* [PATCH 00/53] Initial 3.12 radeon patches
@ 2013-08-12 16:42 Alex Deucher
2013-08-12 16:42 ` [PATCH 01/53] drm/radeon: switch r6xx+ to using CP DMA for the blit copy callback Alex Deucher
` (53 more replies)
0 siblings, 54 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:42 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
This is the initial set of patches for 3.12 for radeon. The
big changes here are adding ASPM and DPM support to CIK asics.
There are also some cleanups related to removing bo copy support
using the 3D engine in favor of the sDMA engines or CP DMA.
For easier review the patches are also available here:
http://cgit.freedesktop.org/~agd5f/linux/log/?h=drm-next-3.12-wip
Alex Deucher (53):
drm/radeon: switch r6xx+ to using CP DMA for the blit copy callback
drm/radeon/kms: remove r6xx+ blit copy routines
drm/radeon: add UVD->DPM helper function (v5)
drm/radeon/dpm: use multiple UVD power states (v3)
drm/radeon/dpm: rework thermal state handling
drm/radeon: default to 1024M gart size on rv770+
drm/radeon/dpm: use performance state if no UVD state
drm/radeon/kms: fix up dce8 display watermark calc for dpm
drm/radeon/cik: implement some more atom helpers for DPM
drm/radeon: switch CIK to use radeon_ucode.h
drm/radeon/cik: add support for pcie gen1/2/3 switching
drm/radeon: add support for ASPM on CIK asics
drm/radeon/cik: restructure rlc setup
drm/radeon: clean up sumo_rlc_init() for code sharing
drm/radeon: convert SI,CIK to use sumo_rlc functions
drm/radeon: implement clock and power gating for CIK (v3)
drm/radeon: add indirect accessors for dift registers on CIK
drm/radeon/sumo add helper to go from vid7 to vid2
drm/radeon: switch to pptable.h
drm/radeon: add structs to store uvd clock voltage deps
drm/radeon/cik: add rlc helpers for DPM
drm/radeon: add support for thermal controller on KB/KV
drm/radeon: add CI to r600_is_internal_thermal_sensor()
drm/radeon: add KB/KV to r600_is_internal_thermal_sensor
drm/radeon: add get_temperature() callbacks for CIK
drm/radeon: adjust si_dpm function for code sharing
drm/radeon/dpm: update cac leakage table parsing for CI
drm/radeon/dpm: add support for parsing the atom powertune table
drm/radeon/dpm: grab mvdd_dependency_on_mclk info from vbios
drm/radeon: add structs to store vce clock voltage deps
drm/radeon: add clock voltage dep tables for acp, samu
drm/radeon: parse the vce clock voltage deps table
drm/radeon: parse the uvd clock voltage deps table
drm/radeon/dpm: clean up the extended table error pathes
drm/radeon: parse the samu clock voltage deps table
drm/radeon: parse the acp clock voltage deps table
drm/radeon: add r600_get_pcie_lane_support helper
drm/radeon/dpm: add vce clocks to radeon_ps
drm/radeon/dpm: add a helper to encode pcie lane setting
drm/radeon/dpm: add helper to fetch the vrefresh of the current mode
drm/radeon/kms: add dpm support for KB/KV
drm/radeon: add dpm support for CI dGPUs (v2)
drm/radeon/dpm: add debugfs support for CI
drm/radeon/dpm: implement force performance level for CI
drm/radeon/dpm: implement vblank_too_short callback for CI
drm/radeon/dpm: add debugfs support for KB/KV
drm/radeon/dpm: implement force performance level for KB/KV
drm/radeon/dpm: add new callback for powergating UVD (v3)
drm/radeon: restructure UVD code to handle UVD PG (v2)
drm/radeon/dpm: implement UVD powergating for KB/KV
drm/radeon/dpm: implement UVD powergating for CI
drm/radeon/dpm: add reclocking quirk for ASUS K70AF
drm/radeon/dpm: adjust the vblank time checks for eg, ni, si
drivers/gpu/drm/radeon/Makefile | 7 +-
drivers/gpu/drm/radeon/atombios.h | 615 +--
drivers/gpu/drm/radeon/cayman_blit_shaders.c | 54 -
drivers/gpu/drm/radeon/ci_dpm.c | 5179 +++++++++++++++++++++++
drivers/gpu/drm/radeon/ci_dpm.h | 331 ++
drivers/gpu/drm/radeon/ci_smc.c | 262 ++
drivers/gpu/drm/radeon/cik.c | 2115 ++++++++-
drivers/gpu/drm/radeon/cik_reg.h | 3 +
drivers/gpu/drm/radeon/cikd.h | 589 ++-
drivers/gpu/drm/radeon/clearstate_cayman.h | 2 +-
drivers/gpu/drm/radeon/clearstate_ci.h | 944 +++++
drivers/gpu/drm/radeon/clearstate_evergreen.h | 2 +-
drivers/gpu/drm/radeon/cypress_dpm.c | 3 +-
drivers/gpu/drm/radeon/evergreen.c | 296 +-
drivers/gpu/drm/radeon/evergreen_blit_kms.c | 729 ----
drivers/gpu/drm/radeon/evergreen_blit_shaders.c | 54 -
drivers/gpu/drm/radeon/kv_dpm.c | 2610 ++++++++++++
drivers/gpu/drm/radeon/kv_dpm.h | 199 +
drivers/gpu/drm/radeon/kv_smc.c | 207 +
drivers/gpu/drm/radeon/ni.c | 16 +-
drivers/gpu/drm/radeon/ni_dpm.c | 3 +-
drivers/gpu/drm/radeon/ppsmc.h | 57 +
drivers/gpu/drm/radeon/pptable.h | 682 +++
drivers/gpu/drm/radeon/r600.c | 100 +-
drivers/gpu/drm/radeon/r600_blit.c | 31 +
drivers/gpu/drm/radeon/r600_blit_kms.c | 785 ----
drivers/gpu/drm/radeon/r600_blit_shaders.h | 1 -
drivers/gpu/drm/radeon/r600_dpm.c | 258 +-
drivers/gpu/drm/radeon/r600_dpm.h | 6 +
drivers/gpu/drm/radeon/radeon.h | 140 +-
drivers/gpu/drm/radeon/radeon_asic.c | 57 +-
drivers/gpu/drm/radeon/radeon_asic.h | 61 +-
drivers/gpu/drm/radeon/radeon_atombios.c | 122 +-
drivers/gpu/drm/radeon/radeon_blit_common.h | 44 -
drivers/gpu/drm/radeon/radeon_cs.c | 11 +-
drivers/gpu/drm/radeon/radeon_device.c | 22 +-
drivers/gpu/drm/radeon/radeon_drv.c | 4 +-
drivers/gpu/drm/radeon/radeon_pm.c | 88 +-
drivers/gpu/drm/radeon/radeon_ucode.h | 17 +
drivers/gpu/drm/radeon/radeon_uvd.c | 58 +-
drivers/gpu/drm/radeon/rv770.c | 9 +-
drivers/gpu/drm/radeon/rv770_dpm.c | 10 +-
drivers/gpu/drm/radeon/si.c | 180 +-
drivers/gpu/drm/radeon/si_dpm.c | 29 +-
drivers/gpu/drm/radeon/smu7.h | 170 +
drivers/gpu/drm/radeon/smu7_discrete.h | 486 +++
drivers/gpu/drm/radeon/smu7_fusion.h | 300 ++
drivers/gpu/drm/radeon/sumo_dpm.c | 14 +
drivers/gpu/drm/radeon/sumo_dpm.h | 3 +
49 files changed, 15075 insertions(+), 2890 deletions(-)
create mode 100644 drivers/gpu/drm/radeon/ci_dpm.c
create mode 100644 drivers/gpu/drm/radeon/ci_dpm.h
create mode 100644 drivers/gpu/drm/radeon/ci_smc.c
create mode 100644 drivers/gpu/drm/radeon/clearstate_ci.h
delete mode 100644 drivers/gpu/drm/radeon/evergreen_blit_kms.c
create mode 100644 drivers/gpu/drm/radeon/kv_dpm.c
create mode 100644 drivers/gpu/drm/radeon/kv_dpm.h
create mode 100644 drivers/gpu/drm/radeon/kv_smc.c
create mode 100644 drivers/gpu/drm/radeon/pptable.h
delete mode 100644 drivers/gpu/drm/radeon/r600_blit_kms.c
delete mode 100644 drivers/gpu/drm/radeon/radeon_blit_common.h
create mode 100644 drivers/gpu/drm/radeon/smu7.h
create mode 100644 drivers/gpu/drm/radeon/smu7_discrete.h
create mode 100644 drivers/gpu/drm/radeon/smu7_fusion.h
--
1.8.3.1
^ permalink raw reply [flat|nested] 55+ messages in thread
* [PATCH 01/53] drm/radeon: switch r6xx+ to using CP DMA for the blit copy callback
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
@ 2013-08-12 16:42 ` Alex Deucher
2013-08-12 16:42 ` [PATCH 02/53] drm/radeon/kms: remove r6xx+ blit copy routines Alex Deucher
` (52 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:42 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
CP DMA is lighter weight than using the 3D engine.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon_asic.c | 18 +++++++++---------
1 file changed, 9 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index f8f8b31..1926ec0 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -1022,7 +1022,7 @@ static struct radeon_asic r600_asic = {
.hdmi_setmode = &r600_hdmi_setmode,
},
.copy = {
- .blit = &r600_copy_blit,
+ .blit = &r600_copy_cpdma,
.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
.dma = &r600_copy_dma,
.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
@@ -1115,7 +1115,7 @@ static struct radeon_asic rv6xx_asic = {
.get_backlight_level = &atombios_get_backlight_level,
},
.copy = {
- .blit = &r600_copy_blit,
+ .blit = &r600_copy_cpdma,
.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
.dma = &r600_copy_dma,
.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
@@ -1226,7 +1226,7 @@ static struct radeon_asic rs780_asic = {
.hdmi_setmode = &r600_hdmi_setmode,
},
.copy = {
- .blit = &r600_copy_blit,
+ .blit = &r600_copy_cpdma,
.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
.dma = &r600_copy_dma,
.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
@@ -1348,7 +1348,7 @@ static struct radeon_asic rv770_asic = {
.hdmi_setmode = &r600_hdmi_setmode,
},
.copy = {
- .blit = &r600_copy_blit,
+ .blit = &r600_copy_cpdma,
.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
.dma = &rv770_copy_dma,
.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
@@ -1473,7 +1473,7 @@ static struct radeon_asic evergreen_asic = {
.hdmi_setmode = &evergreen_hdmi_setmode,
},
.copy = {
- .blit = &r600_copy_blit,
+ .blit = &r600_copy_cpdma,
.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
.dma = &evergreen_copy_dma,
.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
@@ -1598,7 +1598,7 @@ static struct radeon_asic sumo_asic = {
.hdmi_setmode = &evergreen_hdmi_setmode,
},
.copy = {
- .blit = &r600_copy_blit,
+ .blit = &r600_copy_cpdma,
.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
.dma = &evergreen_copy_dma,
.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
@@ -1722,7 +1722,7 @@ static struct radeon_asic btc_asic = {
.hdmi_setmode = &evergreen_hdmi_setmode,
},
.copy = {
- .blit = &r600_copy_blit,
+ .blit = &r600_copy_cpdma,
.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
.dma = &evergreen_copy_dma,
.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
@@ -1899,7 +1899,7 @@ static struct radeon_asic cayman_asic = {
.hdmi_setmode = &evergreen_hdmi_setmode,
},
.copy = {
- .blit = &r600_copy_blit,
+ .blit = &r600_copy_cpdma,
.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
.dma = &evergreen_copy_dma,
.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
@@ -2074,7 +2074,7 @@ static struct radeon_asic trinity_asic = {
.get_backlight_level = &atombios_get_backlight_level,
},
.copy = {
- .blit = &r600_copy_blit,
+ .blit = &r600_copy_cpdma,
.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
.dma = &evergreen_copy_dma,
.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 02/53] drm/radeon/kms: remove r6xx+ blit copy routines
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
2013-08-12 16:42 ` [PATCH 01/53] drm/radeon: switch r6xx+ to using CP DMA for the blit copy callback Alex Deucher
@ 2013-08-12 16:42 ` Alex Deucher
2013-08-12 16:42 ` [PATCH 03/53] drm/radeon: add UVD->DPM helper function (v5) Alex Deucher
` (51 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:42 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
No longer used now that we use the async dma engines or
CP DMA for bo copies.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/Makefile | 4 +-
drivers/gpu/drm/radeon/cayman_blit_shaders.c | 54 --
drivers/gpu/drm/radeon/evergreen.c | 8 -
drivers/gpu/drm/radeon/evergreen_blit_kms.c | 729 ----------------------
drivers/gpu/drm/radeon/evergreen_blit_shaders.c | 54 --
drivers/gpu/drm/radeon/ni.c | 8 -
drivers/gpu/drm/radeon/r600.c | 26 -
drivers/gpu/drm/radeon/r600_blit.c | 31 +
drivers/gpu/drm/radeon/r600_blit_kms.c | 785 ------------------------
drivers/gpu/drm/radeon/r600_blit_shaders.h | 1 -
drivers/gpu/drm/radeon/radeon.h | 30 -
drivers/gpu/drm/radeon/radeon_asic.h | 16 -
drivers/gpu/drm/radeon/radeon_blit_common.h | 44 --
drivers/gpu/drm/radeon/rv770.c | 7 -
14 files changed, 33 insertions(+), 1764 deletions(-)
delete mode 100644 drivers/gpu/drm/radeon/evergreen_blit_kms.c
delete mode 100644 drivers/gpu/drm/radeon/r600_blit_kms.c
delete mode 100644 drivers/gpu/drm/radeon/radeon_blit_common.h
diff --git a/drivers/gpu/drm/radeon/Makefile b/drivers/gpu/drm/radeon/Makefile
index c3df52c..bfabd69 100644
--- a/drivers/gpu/drm/radeon/Makefile
+++ b/drivers/gpu/drm/radeon/Makefile
@@ -72,8 +72,8 @@ radeon-y += radeon_device.o radeon_asic.o radeon_kms.o \
radeon_cs.o radeon_bios.o radeon_benchmark.o r100.o r300.o r420.o \
rs400.o rs600.o rs690.o rv515.o r520.o r600.o rv770.o radeon_test.o \
r200.o radeon_legacy_tv.o r600_cs.o r600_blit_shaders.o \
- r600_blit_kms.o radeon_pm.o atombios_dp.o r600_audio.o r600_hdmi.o \
- evergreen.o evergreen_cs.o evergreen_blit_shaders.o evergreen_blit_kms.o \
+ radeon_pm.o atombios_dp.o r600_audio.o r600_hdmi.o \
+ evergreen.o evergreen_cs.o evergreen_blit_shaders.o \
evergreen_hdmi.o radeon_trace_points.o ni.o cayman_blit_shaders.o \
atombios_encoders.o radeon_semaphore.o radeon_sa.o atombios_i2c.o si.o \
si_blit_shaders.o radeon_prime.o radeon_uvd.o cik.o cik_blit_shaders.o \
diff --git a/drivers/gpu/drm/radeon/cayman_blit_shaders.c b/drivers/gpu/drm/radeon/cayman_blit_shaders.c
index 19a0114..98d009e 100644
--- a/drivers/gpu/drm/radeon/cayman_blit_shaders.c
+++ b/drivers/gpu/drm/radeon/cayman_blit_shaders.c
@@ -317,58 +317,4 @@ const u32 cayman_default_state[] =
0x00000010, /* */
};
-const u32 cayman_vs[] =
-{
- 0x00000004,
- 0x80400400,
- 0x0000a03c,
- 0x95000688,
- 0x00004000,
- 0x15000688,
- 0x00000000,
- 0x88000000,
- 0x04000000,
- 0x67961001,
-#ifdef __BIG_ENDIAN
- 0x00020000,
-#else
- 0x00000000,
-#endif
- 0x00000000,
- 0x04000000,
- 0x67961000,
-#ifdef __BIG_ENDIAN
- 0x00020008,
-#else
- 0x00000008,
-#endif
- 0x00000000,
-};
-
-const u32 cayman_ps[] =
-{
- 0x00000004,
- 0xa00c0000,
- 0x00000008,
- 0x80400000,
- 0x00000000,
- 0x95000688,
- 0x00000000,
- 0x88000000,
- 0x00380400,
- 0x00146b10,
- 0x00380000,
- 0x20146b10,
- 0x00380400,
- 0x40146b00,
- 0x80380000,
- 0x60146b00,
- 0x00000010,
- 0x000d1000,
- 0xb0800000,
- 0x00000000,
-};
-
-const u32 cayman_ps_size = ARRAY_SIZE(cayman_ps);
-const u32 cayman_vs_size = ARRAY_SIZE(cayman_vs);
const u32 cayman_default_size = ARRAY_SIZE(cayman_default_state);
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index d5b49e3..a5ab569 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -5144,13 +5144,6 @@ static int evergreen_startup(struct radeon_device *rdev)
}
evergreen_gpu_init(rdev);
- r = evergreen_blit_init(rdev);
- if (r) {
- r600_blit_fini(rdev);
- rdev->asic->copy.copy = NULL;
- dev_warn(rdev->dev, "failed blitter (%d) falling back to memcpy\n", r);
- }
-
/* allocate rlc buffers */
if (rdev->flags & RADEON_IS_IGP) {
rdev->rlc.reg_list = sumo_rlc_save_restore_register_list;
@@ -5420,7 +5413,6 @@ int evergreen_init(struct radeon_device *rdev)
void evergreen_fini(struct radeon_device *rdev)
{
r600_audio_fini(rdev);
- r600_blit_fini(rdev);
r700_cp_fini(rdev);
r600_dma_fini(rdev);
r600_irq_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/evergreen_blit_kms.c b/drivers/gpu/drm/radeon/evergreen_blit_kms.c
deleted file mode 100644
index 057c87b..0000000
--- a/drivers/gpu/drm/radeon/evergreen_blit_kms.c
+++ /dev/null
@@ -1,729 +0,0 @@
-/*
- * Copyright 2010 Advanced Micro Devices, Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice (including the next
- * paragraph) shall be included in all copies or substantial portions of the
- * Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- *
- * Authors:
- * Alex Deucher <alexander.deucher@amd.com>
- */
-
-#include <drm/drmP.h>
-#include <drm/radeon_drm.h>
-#include "radeon.h"
-
-#include "evergreend.h"
-#include "evergreen_blit_shaders.h"
-#include "cayman_blit_shaders.h"
-#include "radeon_blit_common.h"
-
-/* emits 17 */
-static void
-set_render_target(struct radeon_device *rdev, int format,
- int w, int h, u64 gpu_addr)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- u32 cb_color_info;
- int pitch, slice;
-
- h = ALIGN(h, 8);
- if (h < 8)
- h = 8;
-
- cb_color_info = CB_FORMAT(format) |
- CB_SOURCE_FORMAT(CB_SF_EXPORT_NORM) |
- CB_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
- pitch = (w / 8) - 1;
- slice = ((w * h) / 64) - 1;
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 15));
- radeon_ring_write(ring, (CB_COLOR0_BASE - PACKET3_SET_CONTEXT_REG_START) >> 2);
- radeon_ring_write(ring, gpu_addr >> 8);
- radeon_ring_write(ring, pitch);
- radeon_ring_write(ring, slice);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, cb_color_info);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, (w - 1) | ((h - 1) << 16));
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 0);
-}
-
-/* emits 5dw */
-static void
-cp_set_surface_sync(struct radeon_device *rdev,
- u32 sync_type, u32 size,
- u64 mc_addr)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- u32 cp_coher_size;
-
- if (size == 0xffffffff)
- cp_coher_size = 0xffffffff;
- else
- cp_coher_size = ((size + 255) >> 8);
-
- if (rdev->family >= CHIP_CAYMAN) {
- /* CP_COHER_CNTL2 has to be set manually when submitting a surface_sync
- * to the RB directly. For IBs, the CP programs this as part of the
- * surface_sync packet.
- */
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
- radeon_ring_write(ring, (0x85e8 - PACKET3_SET_CONFIG_REG_START) >> 2);
- radeon_ring_write(ring, 0); /* CP_COHER_CNTL2 */
- }
- radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3));
- radeon_ring_write(ring, sync_type);
- radeon_ring_write(ring, cp_coher_size);
- radeon_ring_write(ring, mc_addr >> 8);
- radeon_ring_write(ring, 10); /* poll interval */
-}
-
-/* emits 11dw + 1 surface sync = 16dw */
-static void
-set_shaders(struct radeon_device *rdev)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- u64 gpu_addr;
-
- /* VS */
- gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.vs_offset;
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 3));
- radeon_ring_write(ring, (SQ_PGM_START_VS - PACKET3_SET_CONTEXT_REG_START) >> 2);
- radeon_ring_write(ring, gpu_addr >> 8);
- radeon_ring_write(ring, 2);
- radeon_ring_write(ring, 0);
-
- /* PS */
- gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.ps_offset;
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 4));
- radeon_ring_write(ring, (SQ_PGM_START_PS - PACKET3_SET_CONTEXT_REG_START) >> 2);
- radeon_ring_write(ring, gpu_addr >> 8);
- radeon_ring_write(ring, 1);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 2);
-
- gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.vs_offset;
- cp_set_surface_sync(rdev, PACKET3_SH_ACTION_ENA, 512, gpu_addr);
-}
-
-/* emits 10 + 1 sync (5) = 15 */
-static void
-set_vtx_resource(struct radeon_device *rdev, u64 gpu_addr)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- u32 sq_vtx_constant_word2, sq_vtx_constant_word3;
-
- /* high addr, stride */
- sq_vtx_constant_word2 = SQ_VTXC_BASE_ADDR_HI(upper_32_bits(gpu_addr) & 0xff) |
- SQ_VTXC_STRIDE(16);
-#ifdef __BIG_ENDIAN
- sq_vtx_constant_word2 |= SQ_VTXC_ENDIAN_SWAP(SQ_ENDIAN_8IN32);
-#endif
- /* xyzw swizzles */
- sq_vtx_constant_word3 = SQ_VTCX_SEL_X(SQ_SEL_X) |
- SQ_VTCX_SEL_Y(SQ_SEL_Y) |
- SQ_VTCX_SEL_Z(SQ_SEL_Z) |
- SQ_VTCX_SEL_W(SQ_SEL_W);
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_RESOURCE, 8));
- radeon_ring_write(ring, 0x580);
- radeon_ring_write(ring, gpu_addr & 0xffffffff);
- radeon_ring_write(ring, 48 - 1); /* size */
- radeon_ring_write(ring, sq_vtx_constant_word2);
- radeon_ring_write(ring, sq_vtx_constant_word3);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, S__SQ_CONSTANT_TYPE(SQ_TEX_VTX_VALID_BUFFER));
-
- if ((rdev->family == CHIP_CEDAR) ||
- (rdev->family == CHIP_PALM) ||
- (rdev->family == CHIP_SUMO) ||
- (rdev->family == CHIP_SUMO2) ||
- (rdev->family == CHIP_CAICOS))
- cp_set_surface_sync(rdev,
- PACKET3_TC_ACTION_ENA, 48, gpu_addr);
- else
- cp_set_surface_sync(rdev,
- PACKET3_VC_ACTION_ENA, 48, gpu_addr);
-
-}
-
-/* emits 10 */
-static void
-set_tex_resource(struct radeon_device *rdev,
- int format, int w, int h, int pitch,
- u64 gpu_addr, u32 size)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- u32 sq_tex_resource_word0, sq_tex_resource_word1;
- u32 sq_tex_resource_word4, sq_tex_resource_word7;
-
- if (h < 1)
- h = 1;
-
- sq_tex_resource_word0 = TEX_DIM(SQ_TEX_DIM_2D);
- sq_tex_resource_word0 |= ((((pitch >> 3) - 1) << 6) |
- ((w - 1) << 18));
- sq_tex_resource_word1 = ((h - 1) << 0) |
- TEX_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
- /* xyzw swizzles */
- sq_tex_resource_word4 = TEX_DST_SEL_X(SQ_SEL_X) |
- TEX_DST_SEL_Y(SQ_SEL_Y) |
- TEX_DST_SEL_Z(SQ_SEL_Z) |
- TEX_DST_SEL_W(SQ_SEL_W);
-
- sq_tex_resource_word7 = format |
- S__SQ_CONSTANT_TYPE(SQ_TEX_VTX_VALID_TEXTURE);
-
- cp_set_surface_sync(rdev,
- PACKET3_TC_ACTION_ENA, size, gpu_addr);
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_RESOURCE, 8));
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, sq_tex_resource_word0);
- radeon_ring_write(ring, sq_tex_resource_word1);
- radeon_ring_write(ring, gpu_addr >> 8);
- radeon_ring_write(ring, gpu_addr >> 8);
- radeon_ring_write(ring, sq_tex_resource_word4);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, sq_tex_resource_word7);
-}
-
-/* emits 12 */
-static void
-set_scissors(struct radeon_device *rdev, int x1, int y1,
- int x2, int y2)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- /* workaround some hw bugs */
- if (x2 == 0)
- x1 = 1;
- if (y2 == 0)
- y1 = 1;
- if (rdev->family >= CHIP_CAYMAN) {
- if ((x2 == 1) && (y2 == 1))
- x2 = 2;
- }
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
- radeon_ring_write(ring, (PA_SC_SCREEN_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_START) >> 2);
- radeon_ring_write(ring, (x1 << 0) | (y1 << 16));
- radeon_ring_write(ring, (x2 << 0) | (y2 << 16));
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
- radeon_ring_write(ring, (PA_SC_GENERIC_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_START) >> 2);
- radeon_ring_write(ring, (x1 << 0) | (y1 << 16) | (1 << 31));
- radeon_ring_write(ring, (x2 << 0) | (y2 << 16));
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
- radeon_ring_write(ring, (PA_SC_WINDOW_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_START) >> 2);
- radeon_ring_write(ring, (x1 << 0) | (y1 << 16) | (1 << 31));
- radeon_ring_write(ring, (x2 << 0) | (y2 << 16));
-}
-
-/* emits 10 */
-static void
-draw_auto(struct radeon_device *rdev)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
- radeon_ring_write(ring, (VGT_PRIMITIVE_TYPE - PACKET3_SET_CONFIG_REG_START) >> 2);
- radeon_ring_write(ring, DI_PT_RECTLIST);
-
- radeon_ring_write(ring, PACKET3(PACKET3_INDEX_TYPE, 0));
- radeon_ring_write(ring,
-#ifdef __BIG_ENDIAN
- (2 << 2) |
-#endif
- DI_INDEX_SIZE_16_BIT);
-
- radeon_ring_write(ring, PACKET3(PACKET3_NUM_INSTANCES, 0));
- radeon_ring_write(ring, 1);
-
- radeon_ring_write(ring, PACKET3(PACKET3_DRAW_INDEX_AUTO, 1));
- radeon_ring_write(ring, 3);
- radeon_ring_write(ring, DI_SRC_SEL_AUTO_INDEX);
-
-}
-
-/* emits 39 */
-static void
-set_default_state(struct radeon_device *rdev)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- u32 sq_config, sq_gpr_resource_mgmt_1, sq_gpr_resource_mgmt_2, sq_gpr_resource_mgmt_3;
- u32 sq_thread_resource_mgmt, sq_thread_resource_mgmt_2;
- u32 sq_stack_resource_mgmt_1, sq_stack_resource_mgmt_2, sq_stack_resource_mgmt_3;
- int num_ps_gprs, num_vs_gprs, num_temp_gprs;
- int num_gs_gprs, num_es_gprs, num_hs_gprs, num_ls_gprs;
- int num_ps_threads, num_vs_threads, num_gs_threads, num_es_threads;
- int num_hs_threads, num_ls_threads;
- int num_ps_stack_entries, num_vs_stack_entries, num_gs_stack_entries, num_es_stack_entries;
- int num_hs_stack_entries, num_ls_stack_entries;
- u64 gpu_addr;
- int dwords;
-
- /* set clear context state */
- radeon_ring_write(ring, PACKET3(PACKET3_CLEAR_STATE, 0));
- radeon_ring_write(ring, 0);
-
- if (rdev->family < CHIP_CAYMAN) {
- switch (rdev->family) {
- case CHIP_CEDAR:
- default:
- num_ps_gprs = 93;
- num_vs_gprs = 46;
- num_temp_gprs = 4;
- num_gs_gprs = 31;
- num_es_gprs = 31;
- num_hs_gprs = 23;
- num_ls_gprs = 23;
- num_ps_threads = 96;
- num_vs_threads = 16;
- num_gs_threads = 16;
- num_es_threads = 16;
- num_hs_threads = 16;
- num_ls_threads = 16;
- num_ps_stack_entries = 42;
- num_vs_stack_entries = 42;
- num_gs_stack_entries = 42;
- num_es_stack_entries = 42;
- num_hs_stack_entries = 42;
- num_ls_stack_entries = 42;
- break;
- case CHIP_REDWOOD:
- num_ps_gprs = 93;
- num_vs_gprs = 46;
- num_temp_gprs = 4;
- num_gs_gprs = 31;
- num_es_gprs = 31;
- num_hs_gprs = 23;
- num_ls_gprs = 23;
- num_ps_threads = 128;
- num_vs_threads = 20;
- num_gs_threads = 20;
- num_es_threads = 20;
- num_hs_threads = 20;
- num_ls_threads = 20;
- num_ps_stack_entries = 42;
- num_vs_stack_entries = 42;
- num_gs_stack_entries = 42;
- num_es_stack_entries = 42;
- num_hs_stack_entries = 42;
- num_ls_stack_entries = 42;
- break;
- case CHIP_JUNIPER:
- num_ps_gprs = 93;
- num_vs_gprs = 46;
- num_temp_gprs = 4;
- num_gs_gprs = 31;
- num_es_gprs = 31;
- num_hs_gprs = 23;
- num_ls_gprs = 23;
- num_ps_threads = 128;
- num_vs_threads = 20;
- num_gs_threads = 20;
- num_es_threads = 20;
- num_hs_threads = 20;
- num_ls_threads = 20;
- num_ps_stack_entries = 85;
- num_vs_stack_entries = 85;
- num_gs_stack_entries = 85;
- num_es_stack_entries = 85;
- num_hs_stack_entries = 85;
- num_ls_stack_entries = 85;
- break;
- case CHIP_CYPRESS:
- case CHIP_HEMLOCK:
- num_ps_gprs = 93;
- num_vs_gprs = 46;
- num_temp_gprs = 4;
- num_gs_gprs = 31;
- num_es_gprs = 31;
- num_hs_gprs = 23;
- num_ls_gprs = 23;
- num_ps_threads = 128;
- num_vs_threads = 20;
- num_gs_threads = 20;
- num_es_threads = 20;
- num_hs_threads = 20;
- num_ls_threads = 20;
- num_ps_stack_entries = 85;
- num_vs_stack_entries = 85;
- num_gs_stack_entries = 85;
- num_es_stack_entries = 85;
- num_hs_stack_entries = 85;
- num_ls_stack_entries = 85;
- break;
- case CHIP_PALM:
- num_ps_gprs = 93;
- num_vs_gprs = 46;
- num_temp_gprs = 4;
- num_gs_gprs = 31;
- num_es_gprs = 31;
- num_hs_gprs = 23;
- num_ls_gprs = 23;
- num_ps_threads = 96;
- num_vs_threads = 16;
- num_gs_threads = 16;
- num_es_threads = 16;
- num_hs_threads = 16;
- num_ls_threads = 16;
- num_ps_stack_entries = 42;
- num_vs_stack_entries = 42;
- num_gs_stack_entries = 42;
- num_es_stack_entries = 42;
- num_hs_stack_entries = 42;
- num_ls_stack_entries = 42;
- break;
- case CHIP_SUMO:
- num_ps_gprs = 93;
- num_vs_gprs = 46;
- num_temp_gprs = 4;
- num_gs_gprs = 31;
- num_es_gprs = 31;
- num_hs_gprs = 23;
- num_ls_gprs = 23;
- num_ps_threads = 96;
- num_vs_threads = 25;
- num_gs_threads = 25;
- num_es_threads = 25;
- num_hs_threads = 25;
- num_ls_threads = 25;
- num_ps_stack_entries = 42;
- num_vs_stack_entries = 42;
- num_gs_stack_entries = 42;
- num_es_stack_entries = 42;
- num_hs_stack_entries = 42;
- num_ls_stack_entries = 42;
- break;
- case CHIP_SUMO2:
- num_ps_gprs = 93;
- num_vs_gprs = 46;
- num_temp_gprs = 4;
- num_gs_gprs = 31;
- num_es_gprs = 31;
- num_hs_gprs = 23;
- num_ls_gprs = 23;
- num_ps_threads = 96;
- num_vs_threads = 25;
- num_gs_threads = 25;
- num_es_threads = 25;
- num_hs_threads = 25;
- num_ls_threads = 25;
- num_ps_stack_entries = 85;
- num_vs_stack_entries = 85;
- num_gs_stack_entries = 85;
- num_es_stack_entries = 85;
- num_hs_stack_entries = 85;
- num_ls_stack_entries = 85;
- break;
- case CHIP_BARTS:
- num_ps_gprs = 93;
- num_vs_gprs = 46;
- num_temp_gprs = 4;
- num_gs_gprs = 31;
- num_es_gprs = 31;
- num_hs_gprs = 23;
- num_ls_gprs = 23;
- num_ps_threads = 128;
- num_vs_threads = 20;
- num_gs_threads = 20;
- num_es_threads = 20;
- num_hs_threads = 20;
- num_ls_threads = 20;
- num_ps_stack_entries = 85;
- num_vs_stack_entries = 85;
- num_gs_stack_entries = 85;
- num_es_stack_entries = 85;
- num_hs_stack_entries = 85;
- num_ls_stack_entries = 85;
- break;
- case CHIP_TURKS:
- num_ps_gprs = 93;
- num_vs_gprs = 46;
- num_temp_gprs = 4;
- num_gs_gprs = 31;
- num_es_gprs = 31;
- num_hs_gprs = 23;
- num_ls_gprs = 23;
- num_ps_threads = 128;
- num_vs_threads = 20;
- num_gs_threads = 20;
- num_es_threads = 20;
- num_hs_threads = 20;
- num_ls_threads = 20;
- num_ps_stack_entries = 42;
- num_vs_stack_entries = 42;
- num_gs_stack_entries = 42;
- num_es_stack_entries = 42;
- num_hs_stack_entries = 42;
- num_ls_stack_entries = 42;
- break;
- case CHIP_CAICOS:
- num_ps_gprs = 93;
- num_vs_gprs = 46;
- num_temp_gprs = 4;
- num_gs_gprs = 31;
- num_es_gprs = 31;
- num_hs_gprs = 23;
- num_ls_gprs = 23;
- num_ps_threads = 128;
- num_vs_threads = 10;
- num_gs_threads = 10;
- num_es_threads = 10;
- num_hs_threads = 10;
- num_ls_threads = 10;
- num_ps_stack_entries = 42;
- num_vs_stack_entries = 42;
- num_gs_stack_entries = 42;
- num_es_stack_entries = 42;
- num_hs_stack_entries = 42;
- num_ls_stack_entries = 42;
- break;
- }
-
- if ((rdev->family == CHIP_CEDAR) ||
- (rdev->family == CHIP_PALM) ||
- (rdev->family == CHIP_SUMO) ||
- (rdev->family == CHIP_SUMO2) ||
- (rdev->family == CHIP_CAICOS))
- sq_config = 0;
- else
- sq_config = VC_ENABLE;
-
- sq_config |= (EXPORT_SRC_C |
- CS_PRIO(0) |
- LS_PRIO(0) |
- HS_PRIO(0) |
- PS_PRIO(0) |
- VS_PRIO(1) |
- GS_PRIO(2) |
- ES_PRIO(3));
-
- sq_gpr_resource_mgmt_1 = (NUM_PS_GPRS(num_ps_gprs) |
- NUM_VS_GPRS(num_vs_gprs) |
- NUM_CLAUSE_TEMP_GPRS(num_temp_gprs));
- sq_gpr_resource_mgmt_2 = (NUM_GS_GPRS(num_gs_gprs) |
- NUM_ES_GPRS(num_es_gprs));
- sq_gpr_resource_mgmt_3 = (NUM_HS_GPRS(num_hs_gprs) |
- NUM_LS_GPRS(num_ls_gprs));
- sq_thread_resource_mgmt = (NUM_PS_THREADS(num_ps_threads) |
- NUM_VS_THREADS(num_vs_threads) |
- NUM_GS_THREADS(num_gs_threads) |
- NUM_ES_THREADS(num_es_threads));
- sq_thread_resource_mgmt_2 = (NUM_HS_THREADS(num_hs_threads) |
- NUM_LS_THREADS(num_ls_threads));
- sq_stack_resource_mgmt_1 = (NUM_PS_STACK_ENTRIES(num_ps_stack_entries) |
- NUM_VS_STACK_ENTRIES(num_vs_stack_entries));
- sq_stack_resource_mgmt_2 = (NUM_GS_STACK_ENTRIES(num_gs_stack_entries) |
- NUM_ES_STACK_ENTRIES(num_es_stack_entries));
- sq_stack_resource_mgmt_3 = (NUM_HS_STACK_ENTRIES(num_hs_stack_entries) |
- NUM_LS_STACK_ENTRIES(num_ls_stack_entries));
-
- /* disable dyn gprs */
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
- radeon_ring_write(ring, (SQ_DYN_GPR_CNTL_PS_FLUSH_REQ - PACKET3_SET_CONFIG_REG_START) >> 2);
- radeon_ring_write(ring, 0);
-
- /* setup LDS */
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
- radeon_ring_write(ring, (SQ_LDS_RESOURCE_MGMT - PACKET3_SET_CONFIG_REG_START) >> 2);
- radeon_ring_write(ring, 0x10001000);
-
- /* SQ config */
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 11));
- radeon_ring_write(ring, (SQ_CONFIG - PACKET3_SET_CONFIG_REG_START) >> 2);
- radeon_ring_write(ring, sq_config);
- radeon_ring_write(ring, sq_gpr_resource_mgmt_1);
- radeon_ring_write(ring, sq_gpr_resource_mgmt_2);
- radeon_ring_write(ring, sq_gpr_resource_mgmt_3);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, sq_thread_resource_mgmt);
- radeon_ring_write(ring, sq_thread_resource_mgmt_2);
- radeon_ring_write(ring, sq_stack_resource_mgmt_1);
- radeon_ring_write(ring, sq_stack_resource_mgmt_2);
- radeon_ring_write(ring, sq_stack_resource_mgmt_3);
- }
-
- /* CONTEXT_CONTROL */
- radeon_ring_write(ring, 0xc0012800);
- radeon_ring_write(ring, 0x80000000);
- radeon_ring_write(ring, 0x80000000);
-
- /* SQ_VTX_BASE_VTX_LOC */
- radeon_ring_write(ring, 0xc0026f00);
- radeon_ring_write(ring, 0x00000000);
- radeon_ring_write(ring, 0x00000000);
- radeon_ring_write(ring, 0x00000000);
-
- /* SET_SAMPLER */
- radeon_ring_write(ring, 0xc0036e00);
- radeon_ring_write(ring, 0x00000000);
- radeon_ring_write(ring, 0x00000012);
- radeon_ring_write(ring, 0x00000000);
- radeon_ring_write(ring, 0x00000000);
-
- /* set to DX10/11 mode */
- radeon_ring_write(ring, PACKET3(PACKET3_MODE_CONTROL, 0));
- radeon_ring_write(ring, 1);
-
- /* emit an IB pointing at default state */
- dwords = ALIGN(rdev->r600_blit.state_len, 0x10);
- gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.state_offset;
- radeon_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
- radeon_ring_write(ring, gpu_addr & 0xFFFFFFFC);
- radeon_ring_write(ring, upper_32_bits(gpu_addr) & 0xFF);
- radeon_ring_write(ring, dwords);
-
-}
-
-int evergreen_blit_init(struct radeon_device *rdev)
-{
- u32 obj_size;
- int i, r, dwords;
- void *ptr;
- u32 packet2s[16];
- int num_packet2s = 0;
-
- rdev->r600_blit.primitives.set_render_target = set_render_target;
- rdev->r600_blit.primitives.cp_set_surface_sync = cp_set_surface_sync;
- rdev->r600_blit.primitives.set_shaders = set_shaders;
- rdev->r600_blit.primitives.set_vtx_resource = set_vtx_resource;
- rdev->r600_blit.primitives.set_tex_resource = set_tex_resource;
- rdev->r600_blit.primitives.set_scissors = set_scissors;
- rdev->r600_blit.primitives.draw_auto = draw_auto;
- rdev->r600_blit.primitives.set_default_state = set_default_state;
-
- rdev->r600_blit.ring_size_common = 8; /* sync semaphore */
- rdev->r600_blit.ring_size_common += 55; /* shaders + def state */
- rdev->r600_blit.ring_size_common += 16; /* fence emit for VB IB */
- rdev->r600_blit.ring_size_common += 5; /* done copy */
- rdev->r600_blit.ring_size_common += 16; /* fence emit for done copy */
-
- rdev->r600_blit.ring_size_per_loop = 74;
- if (rdev->family >= CHIP_CAYMAN)
- rdev->r600_blit.ring_size_per_loop += 9; /* additional DWs for surface sync */
-
- rdev->r600_blit.max_dim = 16384;
-
- rdev->r600_blit.state_offset = 0;
-
- if (rdev->family < CHIP_CAYMAN)
- rdev->r600_blit.state_len = evergreen_default_size;
- else
- rdev->r600_blit.state_len = cayman_default_size;
-
- dwords = rdev->r600_blit.state_len;
- while (dwords & 0xf) {
- packet2s[num_packet2s++] = cpu_to_le32(PACKET2(0));
- dwords++;
- }
-
- obj_size = dwords * 4;
- obj_size = ALIGN(obj_size, 256);
-
- rdev->r600_blit.vs_offset = obj_size;
- if (rdev->family < CHIP_CAYMAN)
- obj_size += evergreen_vs_size * 4;
- else
- obj_size += cayman_vs_size * 4;
- obj_size = ALIGN(obj_size, 256);
-
- rdev->r600_blit.ps_offset = obj_size;
- if (rdev->family < CHIP_CAYMAN)
- obj_size += evergreen_ps_size * 4;
- else
- obj_size += cayman_ps_size * 4;
- obj_size = ALIGN(obj_size, 256);
-
- /* pin copy shader into vram if not already initialized */
- if (!rdev->r600_blit.shader_obj) {
- r = radeon_bo_create(rdev, obj_size, PAGE_SIZE, true,
- RADEON_GEM_DOMAIN_VRAM,
- NULL, &rdev->r600_blit.shader_obj);
- if (r) {
- DRM_ERROR("evergreen failed to allocate shader\n");
- return r;
- }
-
- r = radeon_bo_reserve(rdev->r600_blit.shader_obj, false);
- if (unlikely(r != 0))
- return r;
- r = radeon_bo_pin(rdev->r600_blit.shader_obj, RADEON_GEM_DOMAIN_VRAM,
- &rdev->r600_blit.shader_gpu_addr);
- radeon_bo_unreserve(rdev->r600_blit.shader_obj);
- if (r) {
- dev_err(rdev->dev, "(%d) pin blit object failed\n", r);
- return r;
- }
- }
-
- DRM_DEBUG("evergreen blit allocated bo %08x vs %08x ps %08x\n",
- obj_size,
- rdev->r600_blit.vs_offset, rdev->r600_blit.ps_offset);
-
- r = radeon_bo_reserve(rdev->r600_blit.shader_obj, false);
- if (unlikely(r != 0))
- return r;
- r = radeon_bo_kmap(rdev->r600_blit.shader_obj, &ptr);
- if (r) {
- DRM_ERROR("failed to map blit object %d\n", r);
- return r;
- }
-
- if (rdev->family < CHIP_CAYMAN) {
- memcpy_toio(ptr + rdev->r600_blit.state_offset,
- evergreen_default_state, rdev->r600_blit.state_len * 4);
-
- if (num_packet2s)
- memcpy_toio(ptr + rdev->r600_blit.state_offset + (rdev->r600_blit.state_len * 4),
- packet2s, num_packet2s * 4);
- for (i = 0; i < evergreen_vs_size; i++)
- *(u32 *)((unsigned long)ptr + rdev->r600_blit.vs_offset + i * 4) = cpu_to_le32(evergreen_vs[i]);
- for (i = 0; i < evergreen_ps_size; i++)
- *(u32 *)((unsigned long)ptr + rdev->r600_blit.ps_offset + i * 4) = cpu_to_le32(evergreen_ps[i]);
- } else {
- memcpy_toio(ptr + rdev->r600_blit.state_offset,
- cayman_default_state, rdev->r600_blit.state_len * 4);
-
- if (num_packet2s)
- memcpy_toio(ptr + rdev->r600_blit.state_offset + (rdev->r600_blit.state_len * 4),
- packet2s, num_packet2s * 4);
- for (i = 0; i < cayman_vs_size; i++)
- *(u32 *)((unsigned long)ptr + rdev->r600_blit.vs_offset + i * 4) = cpu_to_le32(cayman_vs[i]);
- for (i = 0; i < cayman_ps_size; i++)
- *(u32 *)((unsigned long)ptr + rdev->r600_blit.ps_offset + i * 4) = cpu_to_le32(cayman_ps[i]);
- }
- radeon_bo_kunmap(rdev->r600_blit.shader_obj);
- radeon_bo_unreserve(rdev->r600_blit.shader_obj);
-
- radeon_ttm_set_active_vram_size(rdev, rdev->mc.real_vram_size);
- return 0;
-}
diff --git a/drivers/gpu/drm/radeon/evergreen_blit_shaders.c b/drivers/gpu/drm/radeon/evergreen_blit_shaders.c
index f85c0af..d433834 100644
--- a/drivers/gpu/drm/radeon/evergreen_blit_shaders.c
+++ b/drivers/gpu/drm/radeon/evergreen_blit_shaders.c
@@ -300,58 +300,4 @@ const u32 evergreen_default_state[] =
0x00000010, /* */
};
-const u32 evergreen_vs[] =
-{
- 0x00000004,
- 0x80800400,
- 0x0000a03c,
- 0x95000688,
- 0x00004000,
- 0x15200688,
- 0x00000000,
- 0x00000000,
- 0x3c000000,
- 0x67961001,
-#ifdef __BIG_ENDIAN
- 0x000a0000,
-#else
- 0x00080000,
-#endif
- 0x00000000,
- 0x1c000000,
- 0x67961000,
-#ifdef __BIG_ENDIAN
- 0x00020008,
-#else
- 0x00000008,
-#endif
- 0x00000000,
-};
-
-const u32 evergreen_ps[] =
-{
- 0x00000003,
- 0xa00c0000,
- 0x00000008,
- 0x80400000,
- 0x00000000,
- 0x95200688,
- 0x00380400,
- 0x00146b10,
- 0x00380000,
- 0x20146b10,
- 0x00380400,
- 0x40146b00,
- 0x80380000,
- 0x60146b00,
- 0x00000000,
- 0x00000000,
- 0x00000010,
- 0x000d1000,
- 0xb0800000,
- 0x00000000,
-};
-
-const u32 evergreen_ps_size = ARRAY_SIZE(evergreen_ps);
-const u32 evergreen_vs_size = ARRAY_SIZE(evergreen_vs);
const u32 evergreen_default_size = ARRAY_SIZE(evergreen_default_state);
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index ccb4f8b5..acdd603 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -2118,13 +2118,6 @@ static int cayman_startup(struct radeon_device *rdev)
return r;
cayman_gpu_init(rdev);
- r = evergreen_blit_init(rdev);
- if (r) {
- r600_blit_fini(rdev);
- rdev->asic->copy.copy = NULL;
- dev_warn(rdev->dev, "failed blitter (%d) falling back to memcpy\n", r);
- }
-
/* allocate rlc buffers */
if (rdev->flags & RADEON_IS_IGP) {
rdev->rlc.reg_list = tn_rlc_save_restore_register_list;
@@ -2413,7 +2406,6 @@ int cayman_init(struct radeon_device *rdev)
void cayman_fini(struct radeon_device *rdev)
{
- r600_blit_fini(rdev);
cayman_cp_fini(rdev);
cayman_dma_fini(rdev);
r600_irq_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index e66e720..3db2e4d 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -3136,25 +3136,6 @@ void r600_uvd_semaphore_emit(struct radeon_device *rdev,
radeon_ring_write(ring, emit_wait ? 1 : 0);
}
-int r600_copy_blit(struct radeon_device *rdev,
- uint64_t src_offset,
- uint64_t dst_offset,
- unsigned num_gpu_pages,
- struct radeon_fence **fence)
-{
- struct radeon_semaphore *sem = NULL;
- struct radeon_sa_bo *vb = NULL;
- int r;
-
- r = r600_blit_prepare_copy(rdev, num_gpu_pages, fence, &vb, &sem);
- if (r) {
- return r;
- }
- r600_kms_blit_copy(rdev, src_offset, dst_offset, num_gpu_pages, vb);
- r600_blit_done_copy(rdev, fence, vb, sem);
- return 0;
-}
-
/**
* r600_copy_cpdma - copy pages using the CP DMA engine
*
@@ -3356,12 +3337,6 @@ static int r600_startup(struct radeon_device *rdev)
return r;
}
r600_gpu_init(rdev);
- r = r600_blit_init(rdev);
- if (r) {
- r600_blit_fini(rdev);
- rdev->asic->copy.copy = NULL;
- dev_warn(rdev->dev, "failed blitter (%d) falling back to memcpy\n", r);
- }
/* allocate wb buffer */
r = radeon_wb_init(rdev);
@@ -3574,7 +3549,6 @@ int r600_init(struct radeon_device *rdev)
void r600_fini(struct radeon_device *rdev)
{
r600_audio_fini(rdev);
- r600_blit_fini(rdev);
r600_cp_fini(rdev);
r600_dma_fini(rdev);
r600_irq_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/r600_blit.c b/drivers/gpu/drm/radeon/r600_blit.c
index f651881..daf7572 100644
--- a/drivers/gpu/drm/radeon/r600_blit.c
+++ b/drivers/gpu/drm/radeon/r600_blit.c
@@ -31,6 +31,37 @@
#include "r600_blit_shaders.h"
+/* 23 bits of float fractional data */
+#define I2F_FRAC_BITS 23
+#define I2F_MASK ((1 << I2F_FRAC_BITS) - 1)
+
+/*
+ * Converts unsigned integer into 32-bit IEEE floating point representation.
+ * Will be exact from 0 to 2^24. Above that, we round towards zero
+ * as the fractional bits will not fit in a float. (It would be better to
+ * round towards even as the fpu does, but that is slower.)
+ */
+static __pure uint32_t int2float(uint32_t x)
+{
+ uint32_t msb, exponent, fraction;
+
+ /* Zero is special */
+ if (!x) return 0;
+
+ /* Get location of the most significant bit */
+ msb = __fls(x);
+
+ /*
+ * Use a rotate instead of a shift because that works both leftwards
+ * and rightwards due to the mod(32) behaviour. This means we don't
+ * need to check to see if we are above 2^24 or not.
+ */
+ fraction = ror32(x, (msb - I2F_FRAC_BITS) & 0x1f) & I2F_MASK;
+ exponent = (127 + msb) << I2F_FRAC_BITS;
+
+ return fraction + exponent;
+}
+
#define DI_PT_RECTLIST 0x11
#define DI_INDEX_SIZE_16_BIT 0x0
#define DI_SRC_SEL_AUTO_INDEX 0x2
diff --git a/drivers/gpu/drm/radeon/r600_blit_kms.c b/drivers/gpu/drm/radeon/r600_blit_kms.c
deleted file mode 100644
index 9fb5780..0000000
--- a/drivers/gpu/drm/radeon/r600_blit_kms.c
+++ /dev/null
@@ -1,785 +0,0 @@
-/*
- * Copyright 2009 Advanced Micro Devices, Inc.
- * Copyright 2009 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice (including the next
- * paragraph) shall be included in all copies or substantial portions of the
- * Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- *
- */
-
-#include <drm/drmP.h>
-#include <drm/radeon_drm.h>
-#include "radeon.h"
-
-#include "r600d.h"
-#include "r600_blit_shaders.h"
-#include "radeon_blit_common.h"
-
-/* 23 bits of float fractional data */
-#define I2F_FRAC_BITS 23
-#define I2F_MASK ((1 << I2F_FRAC_BITS) - 1)
-
-/*
- * Converts unsigned integer into 32-bit IEEE floating point representation.
- * Will be exact from 0 to 2^24. Above that, we round towards zero
- * as the fractional bits will not fit in a float. (It would be better to
- * round towards even as the fpu does, but that is slower.)
- */
-__pure uint32_t int2float(uint32_t x)
-{
- uint32_t msb, exponent, fraction;
-
- /* Zero is special */
- if (!x) return 0;
-
- /* Get location of the most significant bit */
- msb = __fls(x);
-
- /*
- * Use a rotate instead of a shift because that works both leftwards
- * and rightwards due to the mod(32) behaviour. This means we don't
- * need to check to see if we are above 2^24 or not.
- */
- fraction = ror32(x, (msb - I2F_FRAC_BITS) & 0x1f) & I2F_MASK;
- exponent = (127 + msb) << I2F_FRAC_BITS;
-
- return fraction + exponent;
-}
-
-/* emits 21 on rv770+, 23 on r600 */
-static void
-set_render_target(struct radeon_device *rdev, int format,
- int w, int h, u64 gpu_addr)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- u32 cb_color_info;
- int pitch, slice;
-
- h = ALIGN(h, 8);
- if (h < 8)
- h = 8;
-
- cb_color_info = CB_FORMAT(format) |
- CB_SOURCE_FORMAT(CB_SF_EXPORT_NORM) |
- CB_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
- pitch = (w / 8) - 1;
- slice = ((w * h) / 64) - 1;
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (CB_COLOR0_BASE - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, gpu_addr >> 8);
-
- if (rdev->family > CHIP_R600 && rdev->family < CHIP_RV770) {
- radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_BASE_UPDATE, 0));
- radeon_ring_write(ring, 2 << 0);
- }
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (CB_COLOR0_SIZE - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, (pitch << 0) | (slice << 10));
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (CB_COLOR0_VIEW - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, 0);
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (CB_COLOR0_INFO - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, cb_color_info);
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (CB_COLOR0_TILE - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, 0);
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (CB_COLOR0_FRAG - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, 0);
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (CB_COLOR0_MASK - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, 0);
-}
-
-/* emits 5dw */
-static void
-cp_set_surface_sync(struct radeon_device *rdev,
- u32 sync_type, u32 size,
- u64 mc_addr)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- u32 cp_coher_size;
-
- if (size == 0xffffffff)
- cp_coher_size = 0xffffffff;
- else
- cp_coher_size = ((size + 255) >> 8);
-
- radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3));
- radeon_ring_write(ring, sync_type);
- radeon_ring_write(ring, cp_coher_size);
- radeon_ring_write(ring, mc_addr >> 8);
- radeon_ring_write(ring, 10); /* poll interval */
-}
-
-/* emits 21dw + 1 surface sync = 26dw */
-static void
-set_shaders(struct radeon_device *rdev)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- u64 gpu_addr;
- u32 sq_pgm_resources;
-
- /* setup shader regs */
- sq_pgm_resources = (1 << 0);
-
- /* VS */
- gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.vs_offset;
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (SQ_PGM_START_VS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, gpu_addr >> 8);
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (SQ_PGM_RESOURCES_VS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, sq_pgm_resources);
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (SQ_PGM_CF_OFFSET_VS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, 0);
-
- /* PS */
- gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.ps_offset;
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (SQ_PGM_START_PS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, gpu_addr >> 8);
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (SQ_PGM_RESOURCES_PS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, sq_pgm_resources | (1 << 28));
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (SQ_PGM_EXPORTS_PS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, 2);
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
- radeon_ring_write(ring, (SQ_PGM_CF_OFFSET_PS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, 0);
-
- gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.vs_offset;
- cp_set_surface_sync(rdev, PACKET3_SH_ACTION_ENA, 512, gpu_addr);
-}
-
-/* emits 9 + 1 sync (5) = 14*/
-static void
-set_vtx_resource(struct radeon_device *rdev, u64 gpu_addr)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- u32 sq_vtx_constant_word2;
-
- sq_vtx_constant_word2 = SQ_VTXC_BASE_ADDR_HI(upper_32_bits(gpu_addr) & 0xff) |
- SQ_VTXC_STRIDE(16);
-#ifdef __BIG_ENDIAN
- sq_vtx_constant_word2 |= SQ_VTXC_ENDIAN_SWAP(SQ_ENDIAN_8IN32);
-#endif
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_RESOURCE, 7));
- radeon_ring_write(ring, 0x460);
- radeon_ring_write(ring, gpu_addr & 0xffffffff);
- radeon_ring_write(ring, 48 - 1);
- radeon_ring_write(ring, sq_vtx_constant_word2);
- radeon_ring_write(ring, 1 << 0);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, SQ_TEX_VTX_VALID_BUFFER << 30);
-
- if ((rdev->family == CHIP_RV610) ||
- (rdev->family == CHIP_RV620) ||
- (rdev->family == CHIP_RS780) ||
- (rdev->family == CHIP_RS880) ||
- (rdev->family == CHIP_RV710))
- cp_set_surface_sync(rdev,
- PACKET3_TC_ACTION_ENA, 48, gpu_addr);
- else
- cp_set_surface_sync(rdev,
- PACKET3_VC_ACTION_ENA, 48, gpu_addr);
-}
-
-/* emits 9 */
-static void
-set_tex_resource(struct radeon_device *rdev,
- int format, int w, int h, int pitch,
- u64 gpu_addr, u32 size)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- uint32_t sq_tex_resource_word0, sq_tex_resource_word1, sq_tex_resource_word4;
-
- if (h < 1)
- h = 1;
-
- sq_tex_resource_word0 = S_038000_DIM(V_038000_SQ_TEX_DIM_2D) |
- S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1);
- sq_tex_resource_word0 |= S_038000_PITCH((pitch >> 3) - 1) |
- S_038000_TEX_WIDTH(w - 1);
-
- sq_tex_resource_word1 = S_038004_DATA_FORMAT(format);
- sq_tex_resource_word1 |= S_038004_TEX_HEIGHT(h - 1);
-
- sq_tex_resource_word4 = S_038010_REQUEST_SIZE(1) |
- S_038010_DST_SEL_X(SQ_SEL_X) |
- S_038010_DST_SEL_Y(SQ_SEL_Y) |
- S_038010_DST_SEL_Z(SQ_SEL_Z) |
- S_038010_DST_SEL_W(SQ_SEL_W);
-
- cp_set_surface_sync(rdev,
- PACKET3_TC_ACTION_ENA, size, gpu_addr);
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_RESOURCE, 7));
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, sq_tex_resource_word0);
- radeon_ring_write(ring, sq_tex_resource_word1);
- radeon_ring_write(ring, gpu_addr >> 8);
- radeon_ring_write(ring, gpu_addr >> 8);
- radeon_ring_write(ring, sq_tex_resource_word4);
- radeon_ring_write(ring, 0);
- radeon_ring_write(ring, SQ_TEX_VTX_VALID_TEXTURE << 30);
-}
-
-/* emits 12 */
-static void
-set_scissors(struct radeon_device *rdev, int x1, int y1,
- int x2, int y2)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
- radeon_ring_write(ring, (PA_SC_SCREEN_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, (x1 << 0) | (y1 << 16));
- radeon_ring_write(ring, (x2 << 0) | (y2 << 16));
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
- radeon_ring_write(ring, (PA_SC_GENERIC_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, (x1 << 0) | (y1 << 16) | (1 << 31));
- radeon_ring_write(ring, (x2 << 0) | (y2 << 16));
-
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
- radeon_ring_write(ring, (PA_SC_WINDOW_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
- radeon_ring_write(ring, (x1 << 0) | (y1 << 16) | (1 << 31));
- radeon_ring_write(ring, (x2 << 0) | (y2 << 16));
-}
-
-/* emits 10 */
-static void
-draw_auto(struct radeon_device *rdev)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
- radeon_ring_write(ring, (VGT_PRIMITIVE_TYPE - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
- radeon_ring_write(ring, DI_PT_RECTLIST);
-
- radeon_ring_write(ring, PACKET3(PACKET3_INDEX_TYPE, 0));
- radeon_ring_write(ring,
-#ifdef __BIG_ENDIAN
- (2 << 2) |
-#endif
- DI_INDEX_SIZE_16_BIT);
-
- radeon_ring_write(ring, PACKET3(PACKET3_NUM_INSTANCES, 0));
- radeon_ring_write(ring, 1);
-
- radeon_ring_write(ring, PACKET3(PACKET3_DRAW_INDEX_AUTO, 1));
- radeon_ring_write(ring, 3);
- radeon_ring_write(ring, DI_SRC_SEL_AUTO_INDEX);
-
-}
-
-/* emits 14 */
-static void
-set_default_state(struct radeon_device *rdev)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- u32 sq_config, sq_gpr_resource_mgmt_1, sq_gpr_resource_mgmt_2;
- u32 sq_thread_resource_mgmt, sq_stack_resource_mgmt_1, sq_stack_resource_mgmt_2;
- int num_ps_gprs, num_vs_gprs, num_temp_gprs, num_gs_gprs, num_es_gprs;
- int num_ps_threads, num_vs_threads, num_gs_threads, num_es_threads;
- int num_ps_stack_entries, num_vs_stack_entries, num_gs_stack_entries, num_es_stack_entries;
- u64 gpu_addr;
- int dwords;
-
- switch (rdev->family) {
- case CHIP_R600:
- num_ps_gprs = 192;
- num_vs_gprs = 56;
- num_temp_gprs = 4;
- num_gs_gprs = 0;
- num_es_gprs = 0;
- num_ps_threads = 136;
- num_vs_threads = 48;
- num_gs_threads = 4;
- num_es_threads = 4;
- num_ps_stack_entries = 128;
- num_vs_stack_entries = 128;
- num_gs_stack_entries = 0;
- num_es_stack_entries = 0;
- break;
- case CHIP_RV630:
- case CHIP_RV635:
- num_ps_gprs = 84;
- num_vs_gprs = 36;
- num_temp_gprs = 4;
- num_gs_gprs = 0;
- num_es_gprs = 0;
- num_ps_threads = 144;
- num_vs_threads = 40;
- num_gs_threads = 4;
- num_es_threads = 4;
- num_ps_stack_entries = 40;
- num_vs_stack_entries = 40;
- num_gs_stack_entries = 32;
- num_es_stack_entries = 16;
- break;
- case CHIP_RV610:
- case CHIP_RV620:
- case CHIP_RS780:
- case CHIP_RS880:
- default:
- num_ps_gprs = 84;
- num_vs_gprs = 36;
- num_temp_gprs = 4;
- num_gs_gprs = 0;
- num_es_gprs = 0;
- num_ps_threads = 136;
- num_vs_threads = 48;
- num_gs_threads = 4;
- num_es_threads = 4;
- num_ps_stack_entries = 40;
- num_vs_stack_entries = 40;
- num_gs_stack_entries = 32;
- num_es_stack_entries = 16;
- break;
- case CHIP_RV670:
- num_ps_gprs = 144;
- num_vs_gprs = 40;
- num_temp_gprs = 4;
- num_gs_gprs = 0;
- num_es_gprs = 0;
- num_ps_threads = 136;
- num_vs_threads = 48;
- num_gs_threads = 4;
- num_es_threads = 4;
- num_ps_stack_entries = 40;
- num_vs_stack_entries = 40;
- num_gs_stack_entries = 32;
- num_es_stack_entries = 16;
- break;
- case CHIP_RV770:
- num_ps_gprs = 192;
- num_vs_gprs = 56;
- num_temp_gprs = 4;
- num_gs_gprs = 0;
- num_es_gprs = 0;
- num_ps_threads = 188;
- num_vs_threads = 60;
- num_gs_threads = 0;
- num_es_threads = 0;
- num_ps_stack_entries = 256;
- num_vs_stack_entries = 256;
- num_gs_stack_entries = 0;
- num_es_stack_entries = 0;
- break;
- case CHIP_RV730:
- case CHIP_RV740:
- num_ps_gprs = 84;
- num_vs_gprs = 36;
- num_temp_gprs = 4;
- num_gs_gprs = 0;
- num_es_gprs = 0;
- num_ps_threads = 188;
- num_vs_threads = 60;
- num_gs_threads = 0;
- num_es_threads = 0;
- num_ps_stack_entries = 128;
- num_vs_stack_entries = 128;
- num_gs_stack_entries = 0;
- num_es_stack_entries = 0;
- break;
- case CHIP_RV710:
- num_ps_gprs = 192;
- num_vs_gprs = 56;
- num_temp_gprs = 4;
- num_gs_gprs = 0;
- num_es_gprs = 0;
- num_ps_threads = 144;
- num_vs_threads = 48;
- num_gs_threads = 0;
- num_es_threads = 0;
- num_ps_stack_entries = 128;
- num_vs_stack_entries = 128;
- num_gs_stack_entries = 0;
- num_es_stack_entries = 0;
- break;
- }
-
- if ((rdev->family == CHIP_RV610) ||
- (rdev->family == CHIP_RV620) ||
- (rdev->family == CHIP_RS780) ||
- (rdev->family == CHIP_RS880) ||
- (rdev->family == CHIP_RV710))
- sq_config = 0;
- else
- sq_config = VC_ENABLE;
-
- sq_config |= (DX9_CONSTS |
- ALU_INST_PREFER_VECTOR |
- PS_PRIO(0) |
- VS_PRIO(1) |
- GS_PRIO(2) |
- ES_PRIO(3));
-
- sq_gpr_resource_mgmt_1 = (NUM_PS_GPRS(num_ps_gprs) |
- NUM_VS_GPRS(num_vs_gprs) |
- NUM_CLAUSE_TEMP_GPRS(num_temp_gprs));
- sq_gpr_resource_mgmt_2 = (NUM_GS_GPRS(num_gs_gprs) |
- NUM_ES_GPRS(num_es_gprs));
- sq_thread_resource_mgmt = (NUM_PS_THREADS(num_ps_threads) |
- NUM_VS_THREADS(num_vs_threads) |
- NUM_GS_THREADS(num_gs_threads) |
- NUM_ES_THREADS(num_es_threads));
- sq_stack_resource_mgmt_1 = (NUM_PS_STACK_ENTRIES(num_ps_stack_entries) |
- NUM_VS_STACK_ENTRIES(num_vs_stack_entries));
- sq_stack_resource_mgmt_2 = (NUM_GS_STACK_ENTRIES(num_gs_stack_entries) |
- NUM_ES_STACK_ENTRIES(num_es_stack_entries));
-
- /* emit an IB pointing at default state */
- dwords = ALIGN(rdev->r600_blit.state_len, 0x10);
- gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.state_offset;
- radeon_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
- radeon_ring_write(ring,
-#ifdef __BIG_ENDIAN
- (2 << 0) |
-#endif
- (gpu_addr & 0xFFFFFFFC));
- radeon_ring_write(ring, upper_32_bits(gpu_addr) & 0xFF);
- radeon_ring_write(ring, dwords);
-
- /* SQ config */
- radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 6));
- radeon_ring_write(ring, (SQ_CONFIG - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
- radeon_ring_write(ring, sq_config);
- radeon_ring_write(ring, sq_gpr_resource_mgmt_1);
- radeon_ring_write(ring, sq_gpr_resource_mgmt_2);
- radeon_ring_write(ring, sq_thread_resource_mgmt);
- radeon_ring_write(ring, sq_stack_resource_mgmt_1);
- radeon_ring_write(ring, sq_stack_resource_mgmt_2);
-}
-
-int r600_blit_init(struct radeon_device *rdev)
-{
- u32 obj_size;
- int i, r, dwords;
- void *ptr;
- u32 packet2s[16];
- int num_packet2s = 0;
-
- rdev->r600_blit.primitives.set_render_target = set_render_target;
- rdev->r600_blit.primitives.cp_set_surface_sync = cp_set_surface_sync;
- rdev->r600_blit.primitives.set_shaders = set_shaders;
- rdev->r600_blit.primitives.set_vtx_resource = set_vtx_resource;
- rdev->r600_blit.primitives.set_tex_resource = set_tex_resource;
- rdev->r600_blit.primitives.set_scissors = set_scissors;
- rdev->r600_blit.primitives.draw_auto = draw_auto;
- rdev->r600_blit.primitives.set_default_state = set_default_state;
-
- rdev->r600_blit.ring_size_common = 8; /* sync semaphore */
- rdev->r600_blit.ring_size_common += 40; /* shaders + def state */
- rdev->r600_blit.ring_size_common += 5; /* done copy */
- rdev->r600_blit.ring_size_common += 16; /* fence emit for done copy */
-
- rdev->r600_blit.ring_size_per_loop = 76;
- /* set_render_target emits 2 extra dwords on rv6xx */
- if (rdev->family > CHIP_R600 && rdev->family < CHIP_RV770)
- rdev->r600_blit.ring_size_per_loop += 2;
-
- rdev->r600_blit.max_dim = 8192;
-
- rdev->r600_blit.state_offset = 0;
-
- if (rdev->family >= CHIP_RV770)
- rdev->r600_blit.state_len = r7xx_default_size;
- else
- rdev->r600_blit.state_len = r6xx_default_size;
-
- dwords = rdev->r600_blit.state_len;
- while (dwords & 0xf) {
- packet2s[num_packet2s++] = cpu_to_le32(PACKET2(0));
- dwords++;
- }
-
- obj_size = dwords * 4;
- obj_size = ALIGN(obj_size, 256);
-
- rdev->r600_blit.vs_offset = obj_size;
- obj_size += r6xx_vs_size * 4;
- obj_size = ALIGN(obj_size, 256);
-
- rdev->r600_blit.ps_offset = obj_size;
- obj_size += r6xx_ps_size * 4;
- obj_size = ALIGN(obj_size, 256);
-
- /* pin copy shader into vram if not already initialized */
- if (rdev->r600_blit.shader_obj == NULL) {
- r = radeon_bo_create(rdev, obj_size, PAGE_SIZE, true,
- RADEON_GEM_DOMAIN_VRAM,
- NULL, &rdev->r600_blit.shader_obj);
- if (r) {
- DRM_ERROR("r600 failed to allocate shader\n");
- return r;
- }
-
- r = radeon_bo_reserve(rdev->r600_blit.shader_obj, false);
- if (unlikely(r != 0))
- return r;
- r = radeon_bo_pin(rdev->r600_blit.shader_obj, RADEON_GEM_DOMAIN_VRAM,
- &rdev->r600_blit.shader_gpu_addr);
- radeon_bo_unreserve(rdev->r600_blit.shader_obj);
- if (r) {
- dev_err(rdev->dev, "(%d) pin blit object failed\n", r);
- return r;
- }
- }
-
- DRM_DEBUG("r6xx blit allocated bo %08x vs %08x ps %08x\n",
- obj_size,
- rdev->r600_blit.vs_offset, rdev->r600_blit.ps_offset);
-
- r = radeon_bo_reserve(rdev->r600_blit.shader_obj, false);
- if (unlikely(r != 0))
- return r;
- r = radeon_bo_kmap(rdev->r600_blit.shader_obj, &ptr);
- if (r) {
- DRM_ERROR("failed to map blit object %d\n", r);
- return r;
- }
- if (rdev->family >= CHIP_RV770)
- memcpy_toio(ptr + rdev->r600_blit.state_offset,
- r7xx_default_state, rdev->r600_blit.state_len * 4);
- else
- memcpy_toio(ptr + rdev->r600_blit.state_offset,
- r6xx_default_state, rdev->r600_blit.state_len * 4);
- if (num_packet2s)
- memcpy_toio(ptr + rdev->r600_blit.state_offset + (rdev->r600_blit.state_len * 4),
- packet2s, num_packet2s * 4);
- for (i = 0; i < r6xx_vs_size; i++)
- *(u32 *)((unsigned long)ptr + rdev->r600_blit.vs_offset + i * 4) = cpu_to_le32(r6xx_vs[i]);
- for (i = 0; i < r6xx_ps_size; i++)
- *(u32 *)((unsigned long)ptr + rdev->r600_blit.ps_offset + i * 4) = cpu_to_le32(r6xx_ps[i]);
- radeon_bo_kunmap(rdev->r600_blit.shader_obj);
- radeon_bo_unreserve(rdev->r600_blit.shader_obj);
-
- radeon_ttm_set_active_vram_size(rdev, rdev->mc.real_vram_size);
- return 0;
-}
-
-void r600_blit_fini(struct radeon_device *rdev)
-{
- int r;
-
- radeon_ttm_set_active_vram_size(rdev, rdev->mc.visible_vram_size);
- if (rdev->r600_blit.shader_obj == NULL)
- return;
- /* If we can't reserve the bo, unref should be enough to destroy
- * it when it becomes idle.
- */
- r = radeon_bo_reserve(rdev->r600_blit.shader_obj, false);
- if (!r) {
- radeon_bo_unpin(rdev->r600_blit.shader_obj);
- radeon_bo_unreserve(rdev->r600_blit.shader_obj);
- }
- radeon_bo_unref(&rdev->r600_blit.shader_obj);
-}
-
-static unsigned r600_blit_create_rect(unsigned num_gpu_pages,
- int *width, int *height, int max_dim)
-{
- unsigned max_pages;
- unsigned pages = num_gpu_pages;
- int w, h;
-
- if (num_gpu_pages == 0) {
- /* not supposed to be called with no pages, but just in case */
- h = 0;
- w = 0;
- pages = 0;
- WARN_ON(1);
- } else {
- int rect_order = 2;
- h = RECT_UNIT_H;
- while (num_gpu_pages / rect_order) {
- h *= 2;
- rect_order *= 4;
- if (h >= max_dim) {
- h = max_dim;
- break;
- }
- }
- max_pages = (max_dim * h) / (RECT_UNIT_W * RECT_UNIT_H);
- if (pages > max_pages)
- pages = max_pages;
- w = (pages * RECT_UNIT_W * RECT_UNIT_H) / h;
- w = (w / RECT_UNIT_W) * RECT_UNIT_W;
- pages = (w * h) / (RECT_UNIT_W * RECT_UNIT_H);
- BUG_ON(pages == 0);
- }
-
-
- DRM_DEBUG("blit_rectangle: h=%d, w=%d, pages=%d\n", h, w, pages);
-
- /* return width and height only of the caller wants it */
- if (height)
- *height = h;
- if (width)
- *width = w;
-
- return pages;
-}
-
-
-int r600_blit_prepare_copy(struct radeon_device *rdev, unsigned num_gpu_pages,
- struct radeon_fence **fence, struct radeon_sa_bo **vb,
- struct radeon_semaphore **sem)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- int r;
- int ring_size;
- int num_loops = 0;
- int dwords_per_loop = rdev->r600_blit.ring_size_per_loop;
-
- /* num loops */
- while (num_gpu_pages) {
- num_gpu_pages -=
- r600_blit_create_rect(num_gpu_pages, NULL, NULL,
- rdev->r600_blit.max_dim);
- num_loops++;
- }
-
- /* 48 bytes for vertex per loop */
- r = radeon_sa_bo_new(rdev, &rdev->ring_tmp_bo, vb,
- (num_loops*48)+256, 256, true);
- if (r) {
- return r;
- }
-
- r = radeon_semaphore_create(rdev, sem);
- if (r) {
- radeon_sa_bo_free(rdev, vb, NULL);
- return r;
- }
-
- /* calculate number of loops correctly */
- ring_size = num_loops * dwords_per_loop;
- ring_size += rdev->r600_blit.ring_size_common;
- r = radeon_ring_lock(rdev, ring, ring_size);
- if (r) {
- radeon_sa_bo_free(rdev, vb, NULL);
- radeon_semaphore_free(rdev, sem, NULL);
- return r;
- }
-
- if (radeon_fence_need_sync(*fence, RADEON_RING_TYPE_GFX_INDEX)) {
- radeon_semaphore_sync_rings(rdev, *sem, (*fence)->ring,
- RADEON_RING_TYPE_GFX_INDEX);
- radeon_fence_note_sync(*fence, RADEON_RING_TYPE_GFX_INDEX);
- } else {
- radeon_semaphore_free(rdev, sem, NULL);
- }
-
- rdev->r600_blit.primitives.set_default_state(rdev);
- rdev->r600_blit.primitives.set_shaders(rdev);
- return 0;
-}
-
-void r600_blit_done_copy(struct radeon_device *rdev, struct radeon_fence **fence,
- struct radeon_sa_bo *vb, struct radeon_semaphore *sem)
-{
- struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
- int r;
-
- r = radeon_fence_emit(rdev, fence, RADEON_RING_TYPE_GFX_INDEX);
- if (r) {
- radeon_ring_unlock_undo(rdev, ring);
- return;
- }
-
- radeon_ring_unlock_commit(rdev, ring);
- radeon_sa_bo_free(rdev, &vb, *fence);
- radeon_semaphore_free(rdev, &sem, *fence);
-}
-
-void r600_kms_blit_copy(struct radeon_device *rdev,
- u64 src_gpu_addr, u64 dst_gpu_addr,
- unsigned num_gpu_pages,
- struct radeon_sa_bo *vb)
-{
- u64 vb_gpu_addr;
- u32 *vb_cpu_addr;
-
- DRM_DEBUG("emitting copy %16llx %16llx %d\n",
- src_gpu_addr, dst_gpu_addr, num_gpu_pages);
- vb_cpu_addr = (u32 *)radeon_sa_bo_cpu_addr(vb);
- vb_gpu_addr = radeon_sa_bo_gpu_addr(vb);
-
- while (num_gpu_pages) {
- int w, h;
- unsigned size_in_bytes;
- unsigned pages_per_loop =
- r600_blit_create_rect(num_gpu_pages, &w, &h,
- rdev->r600_blit.max_dim);
-
- size_in_bytes = pages_per_loop * RADEON_GPU_PAGE_SIZE;
- DRM_DEBUG("rectangle w=%d h=%d\n", w, h);
-
- vb_cpu_addr[0] = 0;
- vb_cpu_addr[1] = 0;
- vb_cpu_addr[2] = 0;
- vb_cpu_addr[3] = 0;
-
- vb_cpu_addr[4] = 0;
- vb_cpu_addr[5] = int2float(h);
- vb_cpu_addr[6] = 0;
- vb_cpu_addr[7] = int2float(h);
-
- vb_cpu_addr[8] = int2float(w);
- vb_cpu_addr[9] = int2float(h);
- vb_cpu_addr[10] = int2float(w);
- vb_cpu_addr[11] = int2float(h);
-
- rdev->r600_blit.primitives.set_tex_resource(rdev, FMT_8_8_8_8,
- w, h, w, src_gpu_addr, size_in_bytes);
- rdev->r600_blit.primitives.set_render_target(rdev, COLOR_8_8_8_8,
- w, h, dst_gpu_addr);
- rdev->r600_blit.primitives.set_scissors(rdev, 0, 0, w, h);
- rdev->r600_blit.primitives.set_vtx_resource(rdev, vb_gpu_addr);
- rdev->r600_blit.primitives.draw_auto(rdev);
- rdev->r600_blit.primitives.cp_set_surface_sync(rdev,
- PACKET3_CB_ACTION_ENA | PACKET3_CB0_DEST_BASE_ENA,
- size_in_bytes, dst_gpu_addr);
-
- vb_cpu_addr += 12;
- vb_gpu_addr += 4*12;
- src_gpu_addr += size_in_bytes;
- dst_gpu_addr += size_in_bytes;
- num_gpu_pages -= pages_per_loop;
- }
-}
diff --git a/drivers/gpu/drm/radeon/r600_blit_shaders.h b/drivers/gpu/drm/radeon/r600_blit_shaders.h
index 2f3ce7a..f437d36 100644
--- a/drivers/gpu/drm/radeon/r600_blit_shaders.h
+++ b/drivers/gpu/drm/radeon/r600_blit_shaders.h
@@ -35,5 +35,4 @@ extern const u32 r6xx_default_state[];
extern const u32 r6xx_ps_size, r6xx_vs_size;
extern const u32 r6xx_default_size, r7xx_default_size;
-__pure uint32_t int2float(uint32_t x);
#endif
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 274b8e1..a6da0db 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -844,35 +844,6 @@ struct r600_ih {
bool enabled;
};
-struct r600_blit_cp_primitives {
- void (*set_render_target)(struct radeon_device *rdev, int format,
- int w, int h, u64 gpu_addr);
- void (*cp_set_surface_sync)(struct radeon_device *rdev,
- u32 sync_type, u32 size,
- u64 mc_addr);
- void (*set_shaders)(struct radeon_device *rdev);
- void (*set_vtx_resource)(struct radeon_device *rdev, u64 gpu_addr);
- void (*set_tex_resource)(struct radeon_device *rdev,
- int format, int w, int h, int pitch,
- u64 gpu_addr, u32 size);
- void (*set_scissors)(struct radeon_device *rdev, int x1, int y1,
- int x2, int y2);
- void (*draw_auto)(struct radeon_device *rdev);
- void (*set_default_state)(struct radeon_device *rdev);
-};
-
-struct r600_blit {
- struct radeon_bo *shader_obj;
- struct r600_blit_cp_primitives primitives;
- int max_dim;
- int ring_size_common;
- int ring_size_per_loop;
- u64 shader_gpu_addr;
- u32 vs_offset, ps_offset;
- u32 state_offset;
- u32 state_len;
-};
-
/*
* RLC stuff
*/
@@ -2066,7 +2037,6 @@ struct radeon_device {
const struct firmware *sdma_fw; /* CIK SDMA firmware */
const struct firmware *smc_fw; /* SMC firmware */
const struct firmware *uvd_fw; /* UVD firmware */
- struct r600_blit r600_blit;
struct r600_vram_scratch vram_scratch;
int msi_enabled; /* msi enabled */
struct r600_ih ih; /* r6/700 interrupt ring */
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 3d61d5a..3cf7d89 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -337,9 +337,6 @@ void r600_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib);
int r600_ring_test(struct radeon_device *rdev, struct radeon_ring *cp);
int r600_dma_ring_test(struct radeon_device *rdev, struct radeon_ring *cp);
int r600_uvd_ring_test(struct radeon_device *rdev, struct radeon_ring *ring);
-int r600_copy_blit(struct radeon_device *rdev,
- uint64_t src_offset, uint64_t dst_offset,
- unsigned num_gpu_pages, struct radeon_fence **fence);
int r600_copy_cpdma(struct radeon_device *rdev,
uint64_t src_offset, uint64_t dst_offset,
unsigned num_gpu_pages, struct radeon_fence **fence);
@@ -371,8 +368,6 @@ int r600_count_pipe_bits(uint32_t val);
int r600_mc_wait_for_idle(struct radeon_device *rdev);
int r600_pcie_gart_init(struct radeon_device *rdev);
void r600_scratch_init(struct radeon_device *rdev);
-int r600_blit_init(struct radeon_device *rdev);
-void r600_blit_fini(struct radeon_device *rdev);
int r600_init_microcode(struct radeon_device *rdev);
/* r600 irq */
int r600_irq_process(struct radeon_device *rdev);
@@ -391,16 +386,6 @@ int r600_hdmi_buffer_status_changed(struct drm_encoder *encoder);
void r600_hdmi_update_audio_settings(struct drm_encoder *encoder);
void r600_hdmi_enable(struct drm_encoder *encoder, bool enable);
void r600_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mode);
-/* r600 blit */
-int r600_blit_prepare_copy(struct radeon_device *rdev, unsigned num_gpu_pages,
- struct radeon_fence **fence, struct radeon_sa_bo **vb,
- struct radeon_semaphore **sem);
-void r600_blit_done_copy(struct radeon_device *rdev, struct radeon_fence **fence,
- struct radeon_sa_bo *vb, struct radeon_semaphore *sem);
-void r600_kms_blit_copy(struct radeon_device *rdev,
- u64 src_gpu_addr, u64 dst_gpu_addr,
- unsigned num_gpu_pages,
- struct radeon_sa_bo *vb);
int r600_mc_wait_for_idle(struct radeon_device *rdev);
u32 r600_get_xclk(struct radeon_device *rdev);
uint64_t r600_get_gpu_clock_counter(struct radeon_device *rdev);
@@ -530,7 +515,6 @@ extern u32 evergreen_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_ba
extern void evergreen_post_page_flip(struct radeon_device *rdev, int crtc);
extern void dce4_wait_for_vblank(struct radeon_device *rdev, int crtc);
void evergreen_disable_interrupt_state(struct radeon_device *rdev);
-int evergreen_blit_init(struct radeon_device *rdev);
int evergreen_mc_wait_for_idle(struct radeon_device *rdev);
void evergreen_dma_fence_ring_emit(struct radeon_device *rdev,
struct radeon_fence *fence);
diff --git a/drivers/gpu/drm/radeon/radeon_blit_common.h b/drivers/gpu/drm/radeon/radeon_blit_common.h
deleted file mode 100644
index 4ecbe72..0000000
--- a/drivers/gpu/drm/radeon/radeon_blit_common.h
+++ /dev/null
@@ -1,44 +0,0 @@
-/*
- * Copyright 2009 Advanced Micro Devices, Inc.
- * Copyright 2009 Red Hat Inc.
- * Copyright 2012 Alcatel-Lucent, Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice (including the next
- * paragraph) shall be included in all copies or substantial portions of the
- * Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- *
- */
-
-#ifndef __RADEON_BLIT_COMMON_H__
-
-#define DI_PT_RECTLIST 0x11
-#define DI_INDEX_SIZE_16_BIT 0x0
-#define DI_SRC_SEL_AUTO_INDEX 0x2
-
-#define FMT_8 0x1
-#define FMT_5_6_5 0x8
-#define FMT_8_8_8_8 0x1a
-#define COLOR_8 0x1
-#define COLOR_5_6_5 0x8
-#define COLOR_8_8_8_8 0x1a
-
-#define RECT_UNIT_H 32
-#define RECT_UNIT_W (RADEON_GPU_PAGE_SIZE / 4 / RECT_UNIT_H)
-
-#define __RADEON_BLIT_COMMON_H__
-#endif
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index bcc68ec..8f78d21 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -1852,12 +1852,6 @@ static int rv770_startup(struct radeon_device *rdev)
}
rv770_gpu_init(rdev);
- r = r600_blit_init(rdev);
- if (r) {
- r600_blit_fini(rdev);
- rdev->asic->copy.copy = NULL;
- dev_warn(rdev->dev, "failed blitter (%d) falling back to memcpy\n", r);
- }
/* allocate wb buffer */
r = radeon_wb_init(rdev);
@@ -2092,7 +2086,6 @@ int rv770_init(struct radeon_device *rdev)
void rv770_fini(struct radeon_device *rdev)
{
- r600_blit_fini(rdev);
r700_cp_fini(rdev);
r600_dma_fini(rdev);
r600_irq_fini(rdev);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 03/53] drm/radeon: add UVD->DPM helper function (v5)
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
2013-08-12 16:42 ` [PATCH 01/53] drm/radeon: switch r6xx+ to using CP DMA for the blit copy callback Alex Deucher
2013-08-12 16:42 ` [PATCH 02/53] drm/radeon/kms: remove r6xx+ blit copy routines Alex Deucher
@ 2013-08-12 16:42 ` Alex Deucher
2013-08-12 16:42 ` [PATCH 04/53] drm/radeon/dpm: use multiple UVD power states (v3) Alex Deucher
` (50 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:42 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher, Christian König
Add a helper function for counting the number of open stream handles.
v2: fix copy-pasta in comments and whitespace error
v3: make function static since it's only used in radeon_uvd.c
at the moment
v4: make non-static again for future changes
v5: make static again for new rework of dpm uvd changes
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon.h | 1 +
drivers/gpu/drm/radeon/radeon_uvd.c | 35 +++++++++++++++++++++++++++++++++++
2 files changed, 36 insertions(+)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index a6da0db..23dc0c0 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1441,6 +1441,7 @@ struct radeon_uvd {
void *saved_bo;
atomic_t handles[RADEON_MAX_UVD_HANDLES];
struct drm_file *filp[RADEON_MAX_UVD_HANDLES];
+ unsigned img_size[RADEON_MAX_UVD_HANDLES];
struct delayed_work idle_work;
};
diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c
index b79f4f5..2a4cff1 100644
--- a/drivers/gpu/drm/radeon/radeon_uvd.c
+++ b/drivers/gpu/drm/radeon/radeon_uvd.c
@@ -147,6 +147,7 @@ int radeon_uvd_init(struct radeon_device *rdev)
for (i = 0; i < RADEON_MAX_UVD_HANDLES; ++i) {
atomic_set(&rdev->uvd.handles[i], 0);
rdev->uvd.filp[i] = NULL;
+ rdev->uvd.img_size[i] = 0;
}
return 0;
@@ -347,6 +348,7 @@ static int radeon_uvd_cs_msg(struct radeon_cs_parser *p, struct radeon_bo *bo,
unsigned offset, unsigned buf_sizes[])
{
int32_t *msg, msg_type, handle;
+ unsigned img_size = 0;
void *ptr;
int i, r;
@@ -383,6 +385,8 @@ static int radeon_uvd_cs_msg(struct radeon_cs_parser *p, struct radeon_bo *bo,
if (msg_type == 1) {
/* it's a decode msg, calc buffer sizes */
r = radeon_uvd_cs_msg_decode(msg, buf_sizes);
+ /* calc image size (width * height) */
+ img_size = msg[6] * msg[7];
radeon_bo_kunmap(bo);
if (r)
return r;
@@ -394,6 +398,8 @@ static int radeon_uvd_cs_msg(struct radeon_cs_parser *p, struct radeon_bo *bo,
radeon_bo_kunmap(bo);
return 0;
} else {
+ /* it's a create msg, calc image size (width * height) */
+ img_size = msg[7] * msg[8];
radeon_bo_kunmap(bo);
if (msg_type != 0) {
@@ -414,6 +420,7 @@ static int radeon_uvd_cs_msg(struct radeon_cs_parser *p, struct radeon_bo *bo,
for (i = 0; i < RADEON_MAX_UVD_HANDLES; ++i) {
if (!atomic_cmpxchg(&p->rdev->uvd.handles[i], 0, handle)) {
p->rdev->uvd.filp[i] = p->filp;
+ p->rdev->uvd.img_size[i] = img_size;
return 0;
}
}
@@ -733,6 +740,34 @@ int radeon_uvd_get_destroy_msg(struct radeon_device *rdev, int ring,
return radeon_uvd_send_msg(rdev, ring, bo, fence);
}
+/**
+ * radeon_uvd_count_handles - count number of open streams
+ *
+ * @rdev: radeon_device pointer
+ * @sd: number of SD streams
+ * @hd: number of HD streams
+ *
+ * Count the number of open SD/HD streams as a hint for power mangement
+ */
+static void radeon_uvd_count_handles(struct radeon_device *rdev,
+ unsigned *sd, unsigned *hd)
+{
+ unsigned i;
+
+ *sd = 0;
+ *hd = 0;
+
+ for (i = 0; i < RADEON_MAX_UVD_HANDLES; ++i) {
+ if (!atomic_read(&rdev->uvd.handles[i]))
+ continue;
+
+ if (rdev->uvd.img_size[i] >= 720*576)
+ ++(*hd);
+ else
+ ++(*sd);
+ }
+}
+
static void radeon_uvd_idle_work_handler(struct work_struct *work)
{
struct radeon_device *rdev =
--
1.8.3.1
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 04/53] drm/radeon/dpm: use multiple UVD power states (v3)
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (2 preceding siblings ...)
2013-08-12 16:42 ` [PATCH 03/53] drm/radeon: add UVD->DPM helper function (v5) Alex Deucher
@ 2013-08-12 16:42 ` Alex Deucher
2013-08-12 16:42 ` [PATCH 05/53] drm/radeon/dpm: rework thermal state handling Alex Deucher
` (49 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:42 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Use the UVD handle information to determine which
which power states to select when using UVD. For
example, decoding a single SD stream requires much
lower clocks than multiple HD streams.
v2: switch to a cleaner dpm/uvd interface
v3: change the uvd power state while streams
are active if need be
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon.h | 5 ++++-
drivers/gpu/drm/radeon/radeon_cs.c | 11 +++++++----
drivers/gpu/drm/radeon/radeon_pm.c | 30 ++++++++++++++++++++++++++++++
drivers/gpu/drm/radeon/radeon_uvd.c | 23 ++++++++++++++++-------
4 files changed, 57 insertions(+), 12 deletions(-)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 23dc0c0..7196752 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1360,11 +1360,14 @@ struct radeon_dpm {
struct radeon_dpm_thermal thermal;
/* forced levels */
enum radeon_dpm_forced_level forced_level;
+ /* track UVD streams */
+ unsigned sd;
+ unsigned hd;
};
void radeon_dpm_enable_power_state(struct radeon_device *rdev,
enum radeon_pm_state_type dpm_state);
-
+void radeon_dpm_enable_uvd(struct radeon_device *rdev, bool enable);
struct radeon_pm {
struct mutex mutex;
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 13a130f..5384fa4 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -383,6 +383,10 @@ static int radeon_cs_ib_chunk(struct radeon_device *rdev,
DRM_ERROR("Invalid command stream !\n");
return r;
}
+
+ if (parser->ring == R600_RING_TYPE_UVD_INDEX)
+ radeon_uvd_note_usage(rdev);
+
radeon_cs_sync_rings(parser);
r = radeon_ib_schedule(rdev, &parser->ib, NULL);
if (r) {
@@ -474,6 +478,9 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev,
return r;
}
+ if (parser->ring == R600_RING_TYPE_UVD_INDEX)
+ radeon_uvd_note_usage(rdev);
+
mutex_lock(&rdev->vm_manager.lock);
mutex_lock(&vm->mutex);
r = radeon_vm_alloc_pt(rdev, vm);
@@ -552,10 +559,6 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
return r;
}
- /* XXX pick SD/HD/MVC */
- if (parser.ring == R600_RING_TYPE_UVD_INDEX)
- radeon_uvd_note_usage(rdev);
-
r = radeon_cs_ib_chunk(rdev, &parser);
if (r) {
goto out;
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index c557850..59d7a0c 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -729,6 +729,8 @@ restart_search:
/* use a fallback state if we didn't match */
switch (dpm_state) {
case POWER_STATE_TYPE_INTERNAL_UVD_SD:
+ dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD;
+ goto restart_search;
case POWER_STATE_TYPE_INTERNAL_UVD_HD:
case POWER_STATE_TYPE_INTERNAL_UVD_HD2:
case POWER_STATE_TYPE_INTERNAL_UVD_MVC:
@@ -884,6 +886,34 @@ void radeon_dpm_enable_power_state(struct radeon_device *rdev,
radeon_pm_compute_clocks(rdev);
}
+void radeon_dpm_enable_uvd(struct radeon_device *rdev, bool enable)
+{
+ enum radeon_pm_state_type dpm_state;
+
+ if (enable) {
+ mutex_lock(&rdev->pm.mutex);
+ rdev->pm.dpm.uvd_active = true;
+ if ((rdev->pm.dpm.sd == 1) && (rdev->pm.dpm.hd == 0))
+ dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_SD;
+ else if ((rdev->pm.dpm.sd == 2) && (rdev->pm.dpm.hd == 0))
+ dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD;
+ else if ((rdev->pm.dpm.sd == 0) && (rdev->pm.dpm.hd == 1))
+ dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD;
+ else if ((rdev->pm.dpm.sd == 0) && (rdev->pm.dpm.hd == 2))
+ dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD2;
+ else
+ dpm_state = POWER_STATE_TYPE_INTERNAL_UVD;
+ rdev->pm.dpm.state = dpm_state;
+ mutex_unlock(&rdev->pm.mutex);
+ } else {
+ mutex_lock(&rdev->pm.mutex);
+ rdev->pm.dpm.uvd_active = false;
+ mutex_unlock(&rdev->pm.mutex);
+ }
+
+ radeon_pm_compute_clocks(rdev);
+}
+
static void radeon_pm_suspend_old(struct radeon_device *rdev)
{
mutex_lock(&rdev->pm.mutex);
diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c
index 2a4cff1..1a01bbf 100644
--- a/drivers/gpu/drm/radeon/radeon_uvd.c
+++ b/drivers/gpu/drm/radeon/radeon_uvd.c
@@ -775,10 +775,7 @@ static void radeon_uvd_idle_work_handler(struct work_struct *work)
if (radeon_fence_count_emitted(rdev, R600_RING_TYPE_UVD_INDEX) == 0) {
if ((rdev->pm.pm_method == PM_METHOD_DPM) && rdev->pm.dpm_enabled) {
- mutex_lock(&rdev->pm.mutex);
- rdev->pm.dpm.uvd_active = false;
- mutex_unlock(&rdev->pm.mutex);
- radeon_pm_compute_clocks(rdev);
+ radeon_dpm_enable_uvd(rdev, false);
} else {
radeon_set_uvd_clocks(rdev, 0, 0);
}
@@ -790,13 +787,25 @@ static void radeon_uvd_idle_work_handler(struct work_struct *work)
void radeon_uvd_note_usage(struct radeon_device *rdev)
{
+ bool streams_changed = false;
bool set_clocks = !cancel_delayed_work_sync(&rdev->uvd.idle_work);
set_clocks &= schedule_delayed_work(&rdev->uvd.idle_work,
msecs_to_jiffies(UVD_IDLE_TIMEOUT_MS));
- if (set_clocks) {
+
+ if ((rdev->pm.pm_method == PM_METHOD_DPM) && rdev->pm.dpm_enabled) {
+ unsigned hd = 0, sd = 0;
+ radeon_uvd_count_handles(rdev, &sd, &hd);
+ if ((rdev->pm.dpm.sd != sd) ||
+ (rdev->pm.dpm.hd != hd)) {
+ rdev->pm.dpm.sd = sd;
+ rdev->pm.dpm.hd = hd;
+ streams_changed = true;
+ }
+ }
+
+ if (set_clocks || streams_changed) {
if ((rdev->pm.pm_method == PM_METHOD_DPM) && rdev->pm.dpm_enabled) {
- /* XXX pick SD/HD/MVC */
- radeon_dpm_enable_power_state(rdev, POWER_STATE_TYPE_INTERNAL_UVD);
+ radeon_dpm_enable_uvd(rdev, true);
} else {
radeon_set_uvd_clocks(rdev, 53300, 40000);
}
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 05/53] drm/radeon/dpm: rework thermal state handling
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (3 preceding siblings ...)
2013-08-12 16:42 ` [PATCH 04/53] drm/radeon/dpm: use multiple UVD power states (v3) Alex Deucher
@ 2013-08-12 16:42 ` Alex Deucher
2013-08-12 16:42 ` [PATCH 06/53] drm/radeon: default to 1024M gart size on rv770+ Alex Deucher
` (48 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:42 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
1. Handle the the thermal state directly in the work handler.
Remove the state selection function since nothing else uses it now.
2. On some asics there is no thermal state, so we just use a regular
state and force the low performance state.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon.h | 2 --
drivers/gpu/drm/radeon/radeon_pm.c | 44 +++++++++++++-------------------------
2 files changed, 15 insertions(+), 31 deletions(-)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 7196752..54edeab 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1365,8 +1365,6 @@ struct radeon_dpm {
unsigned hd;
};
-void radeon_dpm_enable_power_state(struct radeon_device *rdev,
- enum radeon_pm_state_type dpm_state);
void radeon_dpm_enable_uvd(struct radeon_device *rdev, bool enable);
struct radeon_pm {
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 59d7a0c..b72508b 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -624,7 +624,15 @@ static void radeon_dpm_thermal_work_handler(struct work_struct *work)
/* switch back the user state */
dpm_state = rdev->pm.dpm.user_state;
}
- radeon_dpm_enable_power_state(rdev, dpm_state);
+ mutex_lock(&rdev->pm.mutex);
+ if (dpm_state == POWER_STATE_TYPE_INTERNAL_THERMAL)
+ rdev->pm.dpm.thermal_active = true;
+ else
+ rdev->pm.dpm.thermal_active = false;
+ rdev->pm.dpm.state = dpm_state;
+ mutex_unlock(&rdev->pm.mutex);
+
+ radeon_pm_compute_clocks(rdev);
}
static struct radeon_ps *radeon_dpm_pick_power_state(struct radeon_device *rdev,
@@ -852,40 +860,18 @@ static void radeon_dpm_change_power_state_locked(struct radeon_device *rdev)
radeon_dpm_post_set_power_state(rdev);
+ /* force low perf level for thermal */
+ if (rdev->pm.dpm.thermal_active &&
+ rdev->asic->dpm.force_performance_level) {
+ radeon_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_LOW);
+ }
+
done:
mutex_unlock(&rdev->ring_lock);
up_write(&rdev->pm.mclk_lock);
mutex_unlock(&rdev->ddev->struct_mutex);
}
-void radeon_dpm_enable_power_state(struct radeon_device *rdev,
- enum radeon_pm_state_type dpm_state)
-{
- if (!rdev->pm.dpm_enabled)
- return;
-
- mutex_lock(&rdev->pm.mutex);
- switch (dpm_state) {
- case POWER_STATE_TYPE_INTERNAL_THERMAL:
- rdev->pm.dpm.thermal_active = true;
- break;
- case POWER_STATE_TYPE_INTERNAL_UVD:
- case POWER_STATE_TYPE_INTERNAL_UVD_SD:
- case POWER_STATE_TYPE_INTERNAL_UVD_HD:
- case POWER_STATE_TYPE_INTERNAL_UVD_HD2:
- case POWER_STATE_TYPE_INTERNAL_UVD_MVC:
- rdev->pm.dpm.uvd_active = true;
- break;
- default:
- rdev->pm.dpm.thermal_active = false;
- rdev->pm.dpm.uvd_active = false;
- break;
- }
- rdev->pm.dpm.state = dpm_state;
- mutex_unlock(&rdev->pm.mutex);
- radeon_pm_compute_clocks(rdev);
-}
-
void radeon_dpm_enable_uvd(struct radeon_device *rdev, bool enable)
{
enum radeon_pm_state_type dpm_state;
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 06/53] drm/radeon: default to 1024M gart size on rv770+
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (4 preceding siblings ...)
2013-08-12 16:42 ` [PATCH 05/53] drm/radeon/dpm: rework thermal state handling Alex Deucher
@ 2013-08-12 16:42 ` Alex Deucher
2013-08-12 16:42 ` [PATCH 07/53] drm/radeon/dpm: use performance state if no UVD state Alex Deucher
` (47 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:42 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Newer asics have a lot of vram so it's less of an
issue to waste a little more space for the gart
page table. This gives us some additional gart space
before having to migrate to non-gart system ram
for games, etc. where we use up most of vram.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon_device.c | 22 +++++++++++++++++-----
drivers/gpu/drm/radeon/radeon_drv.c | 4 ++--
2 files changed, 19 insertions(+), 7 deletions(-)
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 63398ae..e887641 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -1003,16 +1003,28 @@ static void radeon_check_arguments(struct radeon_device *rdev)
radeon_vram_limit = 0;
}
+ if (radeon_gart_size == -1) {
+ /* default to a larger gart size on newer asics */
+ if (rdev->family >= CHIP_RV770)
+ radeon_gart_size = 1024;
+ else
+ radeon_gart_size = 512;
+ }
/* gtt size must be power of two and greater or equal to 32M */
if (radeon_gart_size < 32) {
- dev_warn(rdev->dev, "gart size (%d) too small forcing to 512M\n",
+ dev_warn(rdev->dev, "gart size (%d) too small\n",
radeon_gart_size);
- radeon_gart_size = 512;
-
+ if (rdev->family >= CHIP_RV770)
+ radeon_gart_size = 1024;
+ else
+ radeon_gart_size = 512;
} else if (!radeon_check_pot_argument(radeon_gart_size)) {
dev_warn(rdev->dev, "gart size (%d) must be a power of 2\n",
radeon_gart_size);
- radeon_gart_size = 512;
+ if (rdev->family >= CHIP_RV770)
+ radeon_gart_size = 1024;
+ else
+ radeon_gart_size = 512;
}
rdev->mc.gtt_size = (uint64_t)radeon_gart_size << 20;
@@ -1144,7 +1156,7 @@ int radeon_device_init(struct radeon_device *rdev,
rdev->family = flags & RADEON_FAMILY_MASK;
rdev->is_atom_bios = false;
rdev->usec_timeout = RADEON_MAX_USEC_TIMEOUT;
- rdev->mc.gtt_size = radeon_gart_size * 1024 * 1024;
+ rdev->mc.gtt_size = 512 * 1024 * 1024;
rdev->accel_working = false;
/* set up ring ids */
for (i = 0; i < RADEON_NUM_RINGS; i++) {
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index 29876b1..62bd176 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -154,7 +154,7 @@ int radeon_dynclks = -1;
int radeon_r4xx_atom = 0;
int radeon_agpmode = 0;
int radeon_vram_limit = 0;
-int radeon_gart_size = 512; /* default gart size */
+int radeon_gart_size = -1; /* auto */
int radeon_benchmarking = 0;
int radeon_testing = 0;
int radeon_connector_table = 0;
@@ -187,7 +187,7 @@ module_param_named(vramlimit, radeon_vram_limit, int, 0600);
MODULE_PARM_DESC(agpmode, "AGP Mode (-1 == PCI)");
module_param_named(agpmode, radeon_agpmode, int, 0444);
-MODULE_PARM_DESC(gartsize, "Size of PCIE/IGP gart to setup in megabytes (32, 64, etc)");
+MODULE_PARM_DESC(gartsize, "Size of PCIE/IGP gart to setup in megabytes (32, 64, etc., -1 = auto)");
module_param_named(gartsize, radeon_gart_size, int, 0600);
MODULE_PARM_DESC(benchmark, "Run benchmark");
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 07/53] drm/radeon/dpm: use performance state if no UVD state
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (5 preceding siblings ...)
2013-08-12 16:42 ` [PATCH 06/53] drm/radeon: default to 1024M gart size on rv770+ Alex Deucher
@ 2013-08-12 16:42 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 08/53] drm/radeon/kms: fix up dce8 display watermark calc for dpm Alex Deucher
` (46 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:42 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Newer asics don't have specific UVD states.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon_pm.c | 12 ++++++++++--
1 file changed, 10 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index b72508b..79a03de 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -695,7 +695,10 @@ restart_search:
break;
/* internal states */
case POWER_STATE_TYPE_INTERNAL_UVD:
- return rdev->pm.dpm.uvd_ps;
+ if (rdev->pm.dpm.uvd_ps)
+ return rdev->pm.dpm.uvd_ps;
+ else
+ break;
case POWER_STATE_TYPE_INTERNAL_UVD_SD:
if (ps->class & ATOM_PPLIB_CLASSIFICATION_SDSTATE)
return ps;
@@ -742,7 +745,12 @@ restart_search:
case POWER_STATE_TYPE_INTERNAL_UVD_HD:
case POWER_STATE_TYPE_INTERNAL_UVD_HD2:
case POWER_STATE_TYPE_INTERNAL_UVD_MVC:
- return rdev->pm.dpm.uvd_ps;
+ if (rdev->pm.dpm.uvd_ps) {
+ return rdev->pm.dpm.uvd_ps;
+ } else {
+ dpm_state = POWER_STATE_TYPE_PERFORMANCE;
+ goto restart_search;
+ }
case POWER_STATE_TYPE_INTERNAL_THERMAL:
dpm_state = POWER_STATE_TYPE_INTERNAL_ACPI;
goto restart_search;
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 08/53] drm/radeon/kms: fix up dce8 display watermark calc for dpm
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (6 preceding siblings ...)
2013-08-12 16:42 ` [PATCH 07/53] drm/radeon/dpm: use performance state if no UVD state Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 09/53] drm/radeon/cik: implement some more atom helpers for DPM Alex Deucher
` (45 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Calculate the low and high watermarks based on the low and high
clocks for the current power state. The dynamic pm hw will select
the appropriate watermark based on the internal dpm state.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cik.c | 96 ++++++++++++++++++++++++++++++++++----------
1 file changed, 74 insertions(+), 22 deletions(-)
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 8928bd1..42b143e 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -6826,7 +6826,7 @@ static void dce8_program_watermarks(struct radeon_device *rdev,
u32 lb_size, u32 num_heads)
{
struct drm_display_mode *mode = &radeon_crtc->base.mode;
- struct dce8_wm_params wm;
+ struct dce8_wm_params wm_low, wm_high;
u32 pixel_period;
u32 line_time = 0;
u32 latency_watermark_a = 0, latency_watermark_b = 0;
@@ -6836,35 +6836,82 @@ static void dce8_program_watermarks(struct radeon_device *rdev,
pixel_period = 1000000 / (u32)mode->clock;
line_time = min((u32)mode->crtc_htotal * pixel_period, (u32)65535);
- wm.yclk = rdev->pm.current_mclk * 10;
- wm.sclk = rdev->pm.current_sclk * 10;
- wm.disp_clk = mode->clock;
- wm.src_width = mode->crtc_hdisplay;
- wm.active_time = mode->crtc_hdisplay * pixel_period;
- wm.blank_time = line_time - wm.active_time;
- wm.interlaced = false;
+ /* watermark for high clocks */
+ if ((rdev->pm.pm_method == PM_METHOD_DPM) &&
+ rdev->pm.dpm_enabled) {
+ wm_high.yclk =
+ radeon_dpm_get_mclk(rdev, false) * 10;
+ wm_high.sclk =
+ radeon_dpm_get_sclk(rdev, false) * 10;
+ } else {
+ wm_high.yclk = rdev->pm.current_mclk * 10;
+ wm_high.sclk = rdev->pm.current_sclk * 10;
+ }
+
+ wm_high.disp_clk = mode->clock;
+ wm_high.src_width = mode->crtc_hdisplay;
+ wm_high.active_time = mode->crtc_hdisplay * pixel_period;
+ wm_high.blank_time = line_time - wm_high.active_time;
+ wm_high.interlaced = false;
if (mode->flags & DRM_MODE_FLAG_INTERLACE)
- wm.interlaced = true;
- wm.vsc = radeon_crtc->vsc;
- wm.vtaps = 1;
+ wm_high.interlaced = true;
+ wm_high.vsc = radeon_crtc->vsc;
+ wm_high.vtaps = 1;
if (radeon_crtc->rmx_type != RMX_OFF)
- wm.vtaps = 2;
- wm.bytes_per_pixel = 4; /* XXX: get this from fb config */
- wm.lb_size = lb_size;
- wm.dram_channels = cik_get_number_of_dram_channels(rdev);
- wm.num_heads = num_heads;
+ wm_high.vtaps = 2;
+ wm_high.bytes_per_pixel = 4; /* XXX: get this from fb config */
+ wm_high.lb_size = lb_size;
+ wm_high.dram_channels = cik_get_number_of_dram_channels(rdev);
+ wm_high.num_heads = num_heads;
/* set for high clocks */
- latency_watermark_a = min(dce8_latency_watermark(&wm), (u32)65535);
+ latency_watermark_a = min(dce8_latency_watermark(&wm_high), (u32)65535);
+
+ /* possibly force display priority to high */
+ /* should really do this at mode validation time... */
+ if (!dce8_average_bandwidth_vs_dram_bandwidth_for_display(&wm_high) ||
+ !dce8_average_bandwidth_vs_available_bandwidth(&wm_high) ||
+ !dce8_check_latency_hiding(&wm_high) ||
+ (rdev->disp_priority == 2)) {
+ DRM_DEBUG_KMS("force priority to high\n");
+ }
+
+ /* watermark for low clocks */
+ if ((rdev->pm.pm_method == PM_METHOD_DPM) &&
+ rdev->pm.dpm_enabled) {
+ wm_low.yclk =
+ radeon_dpm_get_mclk(rdev, true) * 10;
+ wm_low.sclk =
+ radeon_dpm_get_sclk(rdev, true) * 10;
+ } else {
+ wm_low.yclk = rdev->pm.current_mclk * 10;
+ wm_low.sclk = rdev->pm.current_sclk * 10;
+ }
+
+ wm_low.disp_clk = mode->clock;
+ wm_low.src_width = mode->crtc_hdisplay;
+ wm_low.active_time = mode->crtc_hdisplay * pixel_period;
+ wm_low.blank_time = line_time - wm_low.active_time;
+ wm_low.interlaced = false;
+ if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+ wm_low.interlaced = true;
+ wm_low.vsc = radeon_crtc->vsc;
+ wm_low.vtaps = 1;
+ if (radeon_crtc->rmx_type != RMX_OFF)
+ wm_low.vtaps = 2;
+ wm_low.bytes_per_pixel = 4; /* XXX: get this from fb config */
+ wm_low.lb_size = lb_size;
+ wm_low.dram_channels = cik_get_number_of_dram_channels(rdev);
+ wm_low.num_heads = num_heads;
+
/* set for low clocks */
- /* wm.yclk = low clk; wm.sclk = low clk */
- latency_watermark_b = min(dce8_latency_watermark(&wm), (u32)65535);
+ latency_watermark_b = min(dce8_latency_watermark(&wm_low), (u32)65535);
/* possibly force display priority to high */
/* should really do this at mode validation time... */
- if (!dce8_average_bandwidth_vs_dram_bandwidth_for_display(&wm) ||
- !dce8_average_bandwidth_vs_available_bandwidth(&wm) ||
- !dce8_check_latency_hiding(&wm) ||
+ if (!dce8_average_bandwidth_vs_dram_bandwidth_for_display(&wm_low) ||
+ !dce8_average_bandwidth_vs_available_bandwidth(&wm_low) ||
+ !dce8_check_latency_hiding(&wm_low) ||
(rdev->disp_priority == 2)) {
DRM_DEBUG_KMS("force priority to high\n");
}
@@ -6889,6 +6936,11 @@ static void dce8_program_watermarks(struct radeon_device *rdev,
LATENCY_HIGH_WATERMARK(line_time)));
/* restore original selection */
WREG32(DPG_WATERMARK_MASK_CONTROL + radeon_crtc->crtc_offset, wm_mask);
+
+ /* save values for DPM */
+ radeon_crtc->line_time = line_time;
+ radeon_crtc->wm_high = latency_watermark_a;
+ radeon_crtc->wm_low = latency_watermark_b;
}
/**
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 09/53] drm/radeon/cik: implement some more atom helpers for DPM
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (7 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 08/53] drm/radeon/kms: fix up dce8 display watermark calc for dpm Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 10/53] drm/radeon: switch CIK to use radeon_ucode.h Alex Deucher
` (44 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Required for DPM on CIK.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon_atombios.c | 115 +++++++++++++++++++++++++++++++
1 file changed, 115 insertions(+)
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index 4ccd61f..27de73c 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -3077,6 +3077,121 @@ int radeon_atom_get_leakage_vddc_based_on_leakage_idx(struct radeon_device *rdev
return radeon_atom_get_max_vddc(rdev, VOLTAGE_TYPE_VDDC, leakage_idx, voltage);
}
+int radeon_atom_get_leakage_id_from_vbios(struct radeon_device *rdev,
+ u16 *leakage_id)
+{
+ union set_voltage args;
+ int index = GetIndexIntoMasterTable(COMMAND, SetVoltage);
+ u8 frev, crev;
+
+ if (!atom_parse_cmd_header(rdev->mode_info.atom_context, index, &frev, &crev))
+ return -EINVAL;
+
+ switch (crev) {
+ case 3:
+ case 4:
+ args.v3.ucVoltageType = 0;
+ args.v3.ucVoltageMode = ATOM_GET_LEAKAGE_ID;
+ args.v3.usVoltageLevel = 0;
+
+ atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
+
+ *leakage_id = le16_to_cpu(args.v3.usVoltageLevel);
+ break;
+ default:
+ DRM_ERROR("Unknown table version %d, %d\n", frev, crev);
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+int radeon_atom_get_leakage_vddc_based_on_leakage_params(struct radeon_device *rdev,
+ u16 *vddc, u16 *vddci,
+ u16 virtual_voltage_id,
+ u16 vbios_voltage_id)
+{
+ int index = GetIndexIntoMasterTable(DATA, ASIC_ProfilingInfo);
+ u8 frev, crev;
+ u16 data_offset, size;
+ int i, j;
+ ATOM_ASIC_PROFILING_INFO_V2_1 *profile;
+ u16 *leakage_bin, *vddc_id_buf, *vddc_buf, *vddci_id_buf, *vddci_buf;
+
+ *vddc = 0;
+ *vddci = 0;
+
+ if (!atom_parse_data_header(rdev->mode_info.atom_context, index, &size,
+ &frev, &crev, &data_offset))
+ return -EINVAL;
+
+ profile = (ATOM_ASIC_PROFILING_INFO_V2_1 *)
+ (rdev->mode_info.atom_context->bios + data_offset);
+
+ switch (frev) {
+ case 1:
+ return -EINVAL;
+ case 2:
+ switch (crev) {
+ case 1:
+ if (size < sizeof(ATOM_ASIC_PROFILING_INFO_V2_1))
+ return -EINVAL;
+ leakage_bin = (u16 *)
+ (rdev->mode_info.atom_context->bios + data_offset +
+ le16_to_cpu(profile->usLeakageBinArrayOffset));
+ vddc_id_buf = (u16 *)
+ (rdev->mode_info.atom_context->bios + data_offset +
+ le16_to_cpu(profile->usElbVDDC_IdArrayOffset));
+ vddc_buf = (u16 *)
+ (rdev->mode_info.atom_context->bios + data_offset +
+ le16_to_cpu(profile->usElbVDDC_LevelArrayOffset));
+ vddci_id_buf = (u16 *)
+ (rdev->mode_info.atom_context->bios + data_offset +
+ le16_to_cpu(profile->usElbVDDCI_IdArrayOffset));
+ vddci_buf = (u16 *)
+ (rdev->mode_info.atom_context->bios + data_offset +
+ le16_to_cpu(profile->usElbVDDCI_LevelArrayOffset));
+
+ if (profile->ucElbVDDC_Num > 0) {
+ for (i = 0; i < profile->ucElbVDDC_Num; i++) {
+ if (vddc_id_buf[i] == virtual_voltage_id) {
+ for (j = 0; j < profile->ucLeakageBinNum; j++) {
+ if (vbios_voltage_id <= leakage_bin[j]) {
+ *vddc = vddc_buf[j * profile->ucElbVDDC_Num + i];
+ break;
+ }
+ }
+ break;
+ }
+ }
+ }
+ if (profile->ucElbVDDCI_Num > 0) {
+ for (i = 0; i < profile->ucElbVDDCI_Num; i++) {
+ if (vddci_id_buf[i] == virtual_voltage_id) {
+ for (j = 0; j < profile->ucLeakageBinNum; j++) {
+ if (vbios_voltage_id <= leakage_bin[j]) {
+ *vddci = vddci_buf[j * profile->ucElbVDDCI_Num + i];
+ break;
+ }
+ }
+ break;
+ }
+ }
+ }
+ break;
+ default:
+ DRM_ERROR("Unknown table version %d, %d\n", frev, crev);
+ return -EINVAL;
+ }
+ break;
+ default:
+ DRM_ERROR("Unknown table version %d, %d\n", frev, crev);
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
int radeon_atom_get_voltage_gpio_settings(struct radeon_device *rdev,
u16 voltage_level, u8 voltage_type,
u32 *gpio_value, u32 *gpio_mask)
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 10/53] drm/radeon: switch CIK to use radeon_ucode.h
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (8 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 09/53] drm/radeon/cik: implement some more atom helpers for DPM Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 11/53] drm/radeon/cik: add support for pcie gen1/2/3 switching Alex Deucher
` (43 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cik.c | 17 +----------------
drivers/gpu/drm/radeon/radeon_ucode.h | 14 ++++++++++++++
2 files changed, 15 insertions(+), 16 deletions(-)
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 42b143e..2ceb900 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -30,22 +30,7 @@
#include "cikd.h"
#include "atom.h"
#include "cik_blit_shaders.h"
-
-/* GFX */
-#define CIK_PFP_UCODE_SIZE 2144
-#define CIK_ME_UCODE_SIZE 2144
-#define CIK_CE_UCODE_SIZE 2144
-/* compute */
-#define CIK_MEC_UCODE_SIZE 4192
-/* interrupts */
-#define BONAIRE_RLC_UCODE_SIZE 2048
-#define KB_RLC_UCODE_SIZE 2560
-#define KV_RLC_UCODE_SIZE 2560
-/* gddr controller */
-#define CIK_MC_UCODE_SIZE 7866
-/* sdma */
-#define CIK_SDMA_UCODE_SIZE 1050
-#define CIK_SDMA_UCODE_VERSION 64
+#include "radeon_ucode.h"
MODULE_FIRMWARE("radeon/BONAIRE_pfp.bin");
MODULE_FIRMWARE("radeon/BONAIRE_me.bin");
diff --git a/drivers/gpu/drm/radeon/radeon_ucode.h b/drivers/gpu/drm/radeon/radeon_ucode.h
index d8b05f7..fad27c0 100644
--- a/drivers/gpu/drm/radeon/radeon_ucode.h
+++ b/drivers/gpu/drm/radeon/radeon_ucode.h
@@ -35,6 +35,12 @@
#define SI_PFP_UCODE_SIZE 2144
#define SI_PM4_UCODE_SIZE 2144
#define SI_CE_UCODE_SIZE 2144
+#define CIK_PFP_UCODE_SIZE 2144
+#define CIK_ME_UCODE_SIZE 2144
+#define CIK_CE_UCODE_SIZE 2144
+
+/* MEC */
+#define CIK_MEC_UCODE_SIZE 4192
/* RLC */
#define R600_RLC_UCODE_SIZE 768
@@ -43,12 +49,20 @@
#define CAYMAN_RLC_UCODE_SIZE 1024
#define ARUBA_RLC_UCODE_SIZE 1536
#define SI_RLC_UCODE_SIZE 2048
+#define BONAIRE_RLC_UCODE_SIZE 2048
+#define KB_RLC_UCODE_SIZE 2560
+#define KV_RLC_UCODE_SIZE 2560
/* MC */
#define BTC_MC_UCODE_SIZE 6024
#define CAYMAN_MC_UCODE_SIZE 6037
#define SI_MC_UCODE_SIZE 7769
#define OLAND_MC_UCODE_SIZE 7863
+#define CIK_MC_UCODE_SIZE 7866
+
+/* SDMA */
+#define CIK_SDMA_UCODE_SIZE 1050
+#define CIK_SDMA_UCODE_VERSION 64
/* SMC */
#define RV770_SMC_UCODE_START 0x0100
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 11/53] drm/radeon/cik: add support for pcie gen1/2/3 switching
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (9 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 10/53] drm/radeon: switch CIK to use radeon_ucode.h Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 12/53] drm/radeon: add support for ASPM on CIK asics Alex Deucher
` (42 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cik.c | 161 ++++++++++++++++++++++++++++++++++++++++++
drivers/gpu/drm/radeon/cikd.h | 57 +++++++++++++++
2 files changed, 218 insertions(+)
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 2ceb900..a30fb32 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -61,6 +61,7 @@ extern void si_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *m
extern void si_rlc_fini(struct radeon_device *rdev);
extern int si_rlc_init(struct radeon_device *rdev);
static void cik_rlc_stop(struct radeon_device *rdev);
+static void cik_pcie_gen3_enable(struct radeon_device *rdev);
/*
* Indirect registers accessor
@@ -5949,6 +5950,9 @@ static int cik_startup(struct radeon_device *rdev)
struct radeon_ring *ring;
int r;
+ /* enable pcie gen2/3 link */
+ cik_pcie_gen3_enable(rdev);
+
cik_mc_program(rdev);
if (rdev->flags & RADEON_IS_IGP) {
@@ -7051,3 +7055,160 @@ int cik_uvd_resume(struct radeon_device *rdev)
return 0;
}
+
+static void cik_pcie_gen3_enable(struct radeon_device *rdev)
+{
+ struct pci_dev *root = rdev->pdev->bus->self;
+ int bridge_pos, gpu_pos;
+ u32 speed_cntl, mask, current_data_rate;
+ int ret, i;
+ u16 tmp16;
+
+ if (radeon_pcie_gen2 == 0)
+ return;
+
+ if (rdev->flags & RADEON_IS_IGP)
+ return;
+
+ if (!(rdev->flags & RADEON_IS_PCIE))
+ return;
+
+ ret = drm_pcie_get_speed_cap_mask(rdev->ddev, &mask);
+ if (ret != 0)
+ return;
+
+ if (!(mask & (DRM_PCIE_SPEED_50 | DRM_PCIE_SPEED_80)))
+ return;
+
+ speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL);
+ current_data_rate = (speed_cntl & LC_CURRENT_DATA_RATE_MASK) >>
+ LC_CURRENT_DATA_RATE_SHIFT;
+ if (mask & DRM_PCIE_SPEED_80) {
+ if (current_data_rate == 2) {
+ DRM_INFO("PCIE gen 3 link speeds already enabled\n");
+ return;
+ }
+ DRM_INFO("enabling PCIE gen 3 link speeds, disable with radeon.pcie_gen2=0\n");
+ } else if (mask & DRM_PCIE_SPEED_50) {
+ if (current_data_rate == 1) {
+ DRM_INFO("PCIE gen 2 link speeds already enabled\n");
+ return;
+ }
+ DRM_INFO("enabling PCIE gen 2 link speeds, disable with radeon.pcie_gen2=0\n");
+ }
+
+ bridge_pos = pci_pcie_cap(root);
+ if (!bridge_pos)
+ return;
+
+ gpu_pos = pci_pcie_cap(rdev->pdev);
+ if (!gpu_pos)
+ return;
+
+ if (mask & DRM_PCIE_SPEED_80) {
+ /* re-try equalization if gen3 is not already enabled */
+ if (current_data_rate != 2) {
+ u16 bridge_cfg, gpu_cfg;
+ u16 bridge_cfg2, gpu_cfg2;
+ u32 max_lw, current_lw, tmp;
+
+ pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &bridge_cfg);
+ pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, &gpu_cfg);
+
+ tmp16 = bridge_cfg | PCI_EXP_LNKCTL_HAWD;
+ pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL, tmp16);
+
+ tmp16 = gpu_cfg | PCI_EXP_LNKCTL_HAWD;
+ pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, tmp16);
+
+ tmp = RREG32_PCIE_PORT(PCIE_LC_STATUS1);
+ max_lw = (tmp & LC_DETECTED_LINK_WIDTH_MASK) >> LC_DETECTED_LINK_WIDTH_SHIFT;
+ current_lw = (tmp & LC_OPERATING_LINK_WIDTH_MASK) >> LC_OPERATING_LINK_WIDTH_SHIFT;
+
+ if (current_lw < max_lw) {
+ tmp = RREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL);
+ if (tmp & LC_RENEGOTIATION_SUPPORT) {
+ tmp &= ~(LC_LINK_WIDTH_MASK | LC_UPCONFIGURE_DIS);
+ tmp |= (max_lw << LC_LINK_WIDTH_SHIFT);
+ tmp |= LC_UPCONFIGURE_SUPPORT | LC_RENEGOTIATE_EN | LC_RECONFIG_NOW;
+ WREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL, tmp);
+ }
+ }
+
+ for (i = 0; i < 10; i++) {
+ /* check status */
+ pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_DEVSTA, &tmp16);
+ if (tmp16 & PCI_EXP_DEVSTA_TRPND)
+ break;
+
+ pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &bridge_cfg);
+ pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, &gpu_cfg);
+
+ pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &bridge_cfg2);
+ pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &gpu_cfg2);
+
+ tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4);
+ tmp |= LC_SET_QUIESCE;
+ WREG32_PCIE_PORT(PCIE_LC_CNTL4, tmp);
+
+ tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4);
+ tmp |= LC_REDO_EQ;
+ WREG32_PCIE_PORT(PCIE_LC_CNTL4, tmp);
+
+ mdelay(100);
+
+ /* linkctl */
+ pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &tmp16);
+ tmp16 &= ~PCI_EXP_LNKCTL_HAWD;
+ tmp16 |= (bridge_cfg & PCI_EXP_LNKCTL_HAWD);
+ pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL, tmp16);
+
+ pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, &tmp16);
+ tmp16 &= ~PCI_EXP_LNKCTL_HAWD;
+ tmp16 |= (gpu_cfg & PCI_EXP_LNKCTL_HAWD);
+ pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, tmp16);
+
+ /* linkctl2 */
+ pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16);
+ tmp16 &= ~((1 << 4) | (7 << 9));
+ tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 9)));
+ pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16);
+
+ pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16);
+ tmp16 &= ~((1 << 4) | (7 << 9));
+ tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 9)));
+ pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16);
+
+ tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4);
+ tmp &= ~LC_SET_QUIESCE;
+ WREG32_PCIE_PORT(PCIE_LC_CNTL4, tmp);
+ }
+ }
+ }
+
+ /* set the link speed */
+ speed_cntl |= LC_FORCE_EN_SW_SPEED_CHANGE | LC_FORCE_DIS_HW_SPEED_CHANGE;
+ speed_cntl &= ~LC_FORCE_DIS_SW_SPEED_CHANGE;
+ WREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL, speed_cntl);
+
+ pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16);
+ tmp16 &= ~0xf;
+ if (mask & DRM_PCIE_SPEED_80)
+ tmp16 |= 3; /* gen3 */
+ else if (mask & DRM_PCIE_SPEED_50)
+ tmp16 |= 2; /* gen2 */
+ else
+ tmp16 |= 1; /* gen1 */
+ pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16);
+
+ speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL);
+ speed_cntl |= LC_INITIATE_LINK_SPEED_CHANGE;
+ WREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL, speed_cntl);
+
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL);
+ if ((speed_cntl & LC_INITIATE_LINK_SPEED_CHANGE) == 0)
+ break;
+ udelay(1);
+ }
+}
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index 7e9275e..a1f376e 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -35,6 +35,63 @@
#define CG_CLKPIN_CNTL 0xC05001A0
# define XTALIN_DIVIDE (1 << 1)
+/* PCIE registers idx/data 0x38/0x3c */
+#define PCIE_LC_STATUS1 0x1400028 /* PCIE */
+# define LC_REVERSE_RCVR (1 << 0)
+# define LC_REVERSE_XMIT (1 << 1)
+# define LC_OPERATING_LINK_WIDTH_MASK (0x7 << 2)
+# define LC_OPERATING_LINK_WIDTH_SHIFT 2
+# define LC_DETECTED_LINK_WIDTH_MASK (0x7 << 5)
+# define LC_DETECTED_LINK_WIDTH_SHIFT 5
+
+#define PCIE_LC_LINK_WIDTH_CNTL 0x100100A2 /* PCIE */
+# define LC_LINK_WIDTH_SHIFT 0
+# define LC_LINK_WIDTH_MASK 0x7
+# define LC_LINK_WIDTH_X0 0
+# define LC_LINK_WIDTH_X1 1
+# define LC_LINK_WIDTH_X2 2
+# define LC_LINK_WIDTH_X4 3
+# define LC_LINK_WIDTH_X8 4
+# define LC_LINK_WIDTH_X16 6
+# define LC_LINK_WIDTH_RD_SHIFT 4
+# define LC_LINK_WIDTH_RD_MASK 0x70
+# define LC_RECONFIG_ARC_MISSING_ESCAPE (1 << 7)
+# define LC_RECONFIG_NOW (1 << 8)
+# define LC_RENEGOTIATION_SUPPORT (1 << 9)
+# define LC_RENEGOTIATE_EN (1 << 10)
+# define LC_SHORT_RECONFIG_EN (1 << 11)
+# define LC_UPCONFIGURE_SUPPORT (1 << 12)
+# define LC_UPCONFIGURE_DIS (1 << 13)
+# define LC_DYN_LANES_PWR_STATE(x) ((x) << 21)
+# define LC_DYN_LANES_PWR_STATE_MASK (0x3 << 21)
+# define LC_DYN_LANES_PWR_STATE_SHIFT 21
+
+#define PCIE_LC_SPEED_CNTL 0x100100A4 /* PCIE */
+# define LC_GEN2_EN_STRAP (1 << 0)
+# define LC_GEN3_EN_STRAP (1 << 1)
+# define LC_TARGET_LINK_SPEED_OVERRIDE_EN (1 << 2)
+# define LC_TARGET_LINK_SPEED_OVERRIDE_MASK (0x3 << 3)
+# define LC_TARGET_LINK_SPEED_OVERRIDE_SHIFT 3
+# define LC_FORCE_EN_SW_SPEED_CHANGE (1 << 5)
+# define LC_FORCE_DIS_SW_SPEED_CHANGE (1 << 6)
+# define LC_FORCE_EN_HW_SPEED_CHANGE (1 << 7)
+# define LC_FORCE_DIS_HW_SPEED_CHANGE (1 << 8)
+# define LC_INITIATE_LINK_SPEED_CHANGE (1 << 9)
+# define LC_SPEED_CHANGE_ATTEMPTS_ALLOWED_MASK (0x3 << 10)
+# define LC_SPEED_CHANGE_ATTEMPTS_ALLOWED_SHIFT 10
+# define LC_CURRENT_DATA_RATE_MASK (0x3 << 13) /* 0/1/2 = gen1/2/3 */
+# define LC_CURRENT_DATA_RATE_SHIFT 13
+# define LC_CLR_FAILED_SPD_CHANGE_CNT (1 << 16)
+# define LC_OTHER_SIDE_EVER_SENT_GEN2 (1 << 18)
+# define LC_OTHER_SIDE_SUPPORTS_GEN2 (1 << 19)
+# define LC_OTHER_SIDE_EVER_SENT_GEN3 (1 << 20)
+# define LC_OTHER_SIDE_SUPPORTS_GEN3 (1 << 21)
+
+#define PCIE_LC_CNTL4 0x100100B6 /* PCIE */
+# define LC_REDO_EQ (1 << 5)
+# define LC_SET_QUIESCE (1 << 13)
+
+/* direct registers */
#define PCIE_INDEX 0x38
#define PCIE_DATA 0x3C
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 12/53] drm/radeon: add support for ASPM on CIK asics
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (10 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 11/53] drm/radeon/cik: add support for pcie gen1/2/3 switching Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 13/53] drm/radeon/cik: restructure rlc setup Alex Deucher
` (41 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Enables PCIE ASPM (Active State Power Management) on
CIK asics.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cik.c | 151 ++++++++++++++++++++++++++++++++++++++++++
drivers/gpu/drm/radeon/cikd.h | 72 +++++++++++++++++++-
2 files changed, 222 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index a30fb32..0da9d67 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -62,6 +62,7 @@ extern void si_rlc_fini(struct radeon_device *rdev);
extern int si_rlc_init(struct radeon_device *rdev);
static void cik_rlc_stop(struct radeon_device *rdev);
static void cik_pcie_gen3_enable(struct radeon_device *rdev);
+static void cik_program_aspm(struct radeon_device *rdev);
/*
* Indirect registers accessor
@@ -5952,6 +5953,8 @@ static int cik_startup(struct radeon_device *rdev)
/* enable pcie gen2/3 link */
cik_pcie_gen3_enable(rdev);
+ /* enable aspm */
+ cik_program_aspm(rdev);
cik_mc_program(rdev);
@@ -7212,3 +7215,151 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev)
udelay(1);
}
}
+
+static void cik_program_aspm(struct radeon_device *rdev)
+{
+ u32 data, orig;
+ bool disable_l0s = false, disable_l1 = false, disable_plloff_in_l1 = false;
+ bool disable_clkreq = false;
+
+ if (radeon_aspm == 0)
+ return;
+
+ /* XXX double check IGPs */
+ if (rdev->flags & RADEON_IS_IGP)
+ return;
+
+ if (!(rdev->flags & RADEON_IS_PCIE))
+ return;
+
+ orig = data = RREG32_PCIE_PORT(PCIE_LC_N_FTS_CNTL);
+ data &= ~LC_XMIT_N_FTS_MASK;
+ data |= LC_XMIT_N_FTS(0x24) | LC_XMIT_N_FTS_OVERRIDE_EN;
+ if (orig != data)
+ WREG32_PCIE_PORT(PCIE_LC_N_FTS_CNTL, data);
+
+ orig = data = RREG32_PCIE_PORT(PCIE_LC_CNTL3);
+ data |= LC_GO_TO_RECOVERY;
+ if (orig != data)
+ WREG32_PCIE_PORT(PCIE_LC_CNTL3, data);
+
+ orig = data = RREG32_PCIE_PORT(PCIE_P_CNTL);
+ data |= P_IGNORE_EDB_ERR;
+ if (orig != data)
+ WREG32_PCIE_PORT(PCIE_P_CNTL, data);
+
+ orig = data = RREG32_PCIE_PORT(PCIE_LC_CNTL);
+ data &= ~(LC_L0S_INACTIVITY_MASK | LC_L1_INACTIVITY_MASK);
+ data |= LC_PMI_TO_L1_DIS;
+ if (!disable_l0s)
+ data |= LC_L0S_INACTIVITY(7);
+
+ if (!disable_l1) {
+ data |= LC_L1_INACTIVITY(7);
+ data &= ~LC_PMI_TO_L1_DIS;
+ if (orig != data)
+ WREG32_PCIE_PORT(PCIE_LC_CNTL, data);
+
+ if (!disable_plloff_in_l1) {
+ bool clk_req_support;
+
+ orig = data = RREG32_PCIE_PORT(PB0_PIF_PWRDOWN_0);
+ data &= ~(PLL_POWER_STATE_IN_OFF_0_MASK | PLL_POWER_STATE_IN_TXS2_0_MASK);
+ data |= PLL_POWER_STATE_IN_OFF_0(7) | PLL_POWER_STATE_IN_TXS2_0(7);
+ if (orig != data)
+ WREG32_PCIE_PORT(PB0_PIF_PWRDOWN_0, data);
+
+ orig = data = RREG32_PCIE_PORT(PB0_PIF_PWRDOWN_1);
+ data &= ~(PLL_POWER_STATE_IN_OFF_1_MASK | PLL_POWER_STATE_IN_TXS2_1_MASK);
+ data |= PLL_POWER_STATE_IN_OFF_1(7) | PLL_POWER_STATE_IN_TXS2_1(7);
+ if (orig != data)
+ WREG32_PCIE_PORT(PB0_PIF_PWRDOWN_1, data);
+
+ orig = data = RREG32_PCIE_PORT(PB1_PIF_PWRDOWN_0);
+ data &= ~(PLL_POWER_STATE_IN_OFF_0_MASK | PLL_POWER_STATE_IN_TXS2_0_MASK);
+ data |= PLL_POWER_STATE_IN_OFF_0(7) | PLL_POWER_STATE_IN_TXS2_0(7);
+ if (orig != data)
+ WREG32_PCIE_PORT(PB1_PIF_PWRDOWN_0, data);
+
+ orig = data = RREG32_PCIE_PORT(PB1_PIF_PWRDOWN_1);
+ data &= ~(PLL_POWER_STATE_IN_OFF_1_MASK | PLL_POWER_STATE_IN_TXS2_1_MASK);
+ data |= PLL_POWER_STATE_IN_OFF_1(7) | PLL_POWER_STATE_IN_TXS2_1(7);
+ if (orig != data)
+ WREG32_PCIE_PORT(PB1_PIF_PWRDOWN_1, data);
+
+ orig = data = RREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL);
+ data &= ~LC_DYN_LANES_PWR_STATE_MASK;
+ data |= LC_DYN_LANES_PWR_STATE(3);
+ if (orig != data)
+ WREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL, data);
+
+ if (!disable_clkreq) {
+ struct pci_dev *root = rdev->pdev->bus->self;
+ u32 lnkcap;
+
+ clk_req_support = false;
+ pcie_capability_read_dword(root, PCI_EXP_LNKCAP, &lnkcap);
+ if (lnkcap & PCI_EXP_LNKCAP_CLKPM)
+ clk_req_support = true;
+ } else {
+ clk_req_support = false;
+ }
+
+ if (clk_req_support) {
+ orig = data = RREG32_PCIE_PORT(PCIE_LC_CNTL2);
+ data |= LC_ALLOW_PDWN_IN_L1 | LC_ALLOW_PDWN_IN_L23;
+ if (orig != data)
+ WREG32_PCIE_PORT(PCIE_LC_CNTL2, data);
+
+ orig = data = RREG32_SMC(THM_CLK_CNTL);
+ data &= ~(CMON_CLK_SEL_MASK | TMON_CLK_SEL_MASK);
+ data |= CMON_CLK_SEL(1) | TMON_CLK_SEL(1);
+ if (orig != data)
+ WREG32_SMC(THM_CLK_CNTL, data);
+
+ orig = data = RREG32_SMC(MISC_CLK_CTRL);
+ data &= ~(DEEP_SLEEP_CLK_SEL_MASK | ZCLK_SEL_MASK);
+ data |= DEEP_SLEEP_CLK_SEL(1) | ZCLK_SEL(1);
+ if (orig != data)
+ WREG32_SMC(MISC_CLK_CTRL, data);
+
+ orig = data = RREG32_SMC(CG_CLKPIN_CNTL);
+ data &= ~BCLK_AS_XCLK;
+ if (orig != data)
+ WREG32_SMC(CG_CLKPIN_CNTL, data);
+
+ orig = data = RREG32_SMC(CG_CLKPIN_CNTL_2);
+ data &= ~FORCE_BIF_REFCLK_EN;
+ if (orig != data)
+ WREG32_SMC(CG_CLKPIN_CNTL_2, data);
+
+ orig = data = RREG32_SMC(MPLL_BYPASSCLK_SEL);
+ data &= ~MPLL_CLKOUT_SEL_MASK;
+ data |= MPLL_CLKOUT_SEL(4);
+ if (orig != data)
+ WREG32_SMC(MPLL_BYPASSCLK_SEL, data);
+ }
+ }
+ } else {
+ if (orig != data)
+ WREG32_PCIE_PORT(PCIE_LC_CNTL, data);
+ }
+
+ orig = data = RREG32_PCIE_PORT(PCIE_CNTL2);
+ data |= SLV_MEM_LS_EN | MST_MEM_LS_EN | REPLAY_MEM_LS_EN;
+ if (orig != data)
+ WREG32_PCIE_PORT(PCIE_CNTL2, data);
+
+ if (!disable_l0s) {
+ data = RREG32_PCIE_PORT(PCIE_LC_N_FTS_CNTL);
+ if((data & LC_N_FTS_MASK) == LC_N_FTS_MASK) {
+ data = RREG32_PCIE_PORT(PCIE_LC_STATUS1);
+ if ((data & LC_REVERSE_XMIT) && (data & LC_REVERSE_RCVR)) {
+ orig = data = RREG32_PCIE_PORT(PCIE_LC_CNTL);
+ data &= ~LC_L0S_INACTIVITY_MASK;
+ if (orig != data)
+ WREG32_PCIE_PORT(PCIE_LC_CNTL, data);
+ }
+ }
+ }
+}
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index a1f376e..d50593f 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -32,10 +32,53 @@
#define GENERAL_PWRMGT 0xC0200000
# define GPU_COUNTER_CLK (1 << 15)
+#define MPLL_BYPASSCLK_SEL 0xC050019C
+# define MPLL_CLKOUT_SEL(x) ((x) << 8)
+# define MPLL_CLKOUT_SEL_MASK 0xFF00
#define CG_CLKPIN_CNTL 0xC05001A0
# define XTALIN_DIVIDE (1 << 1)
+# define BCLK_AS_XCLK (1 << 2)
+#define CG_CLKPIN_CNTL_2 0xC05001A4
+# define FORCE_BIF_REFCLK_EN (1 << 3)
+# define MUX_TCLK_TO_XCLK (1 << 8)
+#define THM_CLK_CNTL 0xC05001A8
+# define CMON_CLK_SEL(x) ((x) << 0)
+# define CMON_CLK_SEL_MASK 0xFF
+# define TMON_CLK_SEL(x) ((x) << 8)
+# define TMON_CLK_SEL_MASK 0xFF00
+#define MISC_CLK_CTRL 0xC05001AC
+# define DEEP_SLEEP_CLK_SEL(x) ((x) << 0)
+# define DEEP_SLEEP_CLK_SEL_MASK 0xFF
+# define ZCLK_SEL(x) ((x) << 8)
+# define ZCLK_SEL_MASK 0xFF00
/* PCIE registers idx/data 0x38/0x3c */
+#define PB0_PIF_PWRDOWN_0 0x1100012 /* PCIE */
+# define PLL_POWER_STATE_IN_TXS2_0(x) ((x) << 7)
+# define PLL_POWER_STATE_IN_TXS2_0_MASK (0x7 << 7)
+# define PLL_POWER_STATE_IN_TXS2_0_SHIFT 7
+# define PLL_POWER_STATE_IN_OFF_0(x) ((x) << 10)
+# define PLL_POWER_STATE_IN_OFF_0_MASK (0x7 << 10)
+# define PLL_POWER_STATE_IN_OFF_0_SHIFT 10
+# define PLL_RAMP_UP_TIME_0(x) ((x) << 24)
+# define PLL_RAMP_UP_TIME_0_MASK (0x7 << 24)
+# define PLL_RAMP_UP_TIME_0_SHIFT 24
+#define PB0_PIF_PWRDOWN_1 0x1100013 /* PCIE */
+# define PLL_POWER_STATE_IN_TXS2_1(x) ((x) << 7)
+# define PLL_POWER_STATE_IN_TXS2_1_MASK (0x7 << 7)
+# define PLL_POWER_STATE_IN_TXS2_1_SHIFT 7
+# define PLL_POWER_STATE_IN_OFF_1(x) ((x) << 10)
+# define PLL_POWER_STATE_IN_OFF_1_MASK (0x7 << 10)
+# define PLL_POWER_STATE_IN_OFF_1_SHIFT 10
+# define PLL_RAMP_UP_TIME_1(x) ((x) << 24)
+# define PLL_RAMP_UP_TIME_1_MASK (0x7 << 24)
+# define PLL_RAMP_UP_TIME_1_SHIFT 24
+
+#define PCIE_CNTL2 0x1001001c /* PCIE */
+# define SLV_MEM_LS_EN (1 << 16)
+# define MST_MEM_LS_EN (1 << 18)
+# define REPLAY_MEM_LS_EN (1 << 19)
+
#define PCIE_LC_STATUS1 0x1400028 /* PCIE */
# define LC_REVERSE_RCVR (1 << 0)
# define LC_REVERSE_XMIT (1 << 1)
@@ -44,6 +87,22 @@
# define LC_DETECTED_LINK_WIDTH_MASK (0x7 << 5)
# define LC_DETECTED_LINK_WIDTH_SHIFT 5
+#define PCIE_P_CNTL 0x1400040 /* PCIE */
+# define P_IGNORE_EDB_ERR (1 << 6)
+
+#define PB1_PIF_PWRDOWN_0 0x2100012 /* PCIE */
+#define PB1_PIF_PWRDOWN_1 0x2100013 /* PCIE */
+
+#define PCIE_LC_CNTL 0x100100A0 /* PCIE */
+# define LC_L0S_INACTIVITY(x) ((x) << 8)
+# define LC_L0S_INACTIVITY_MASK (0xf << 8)
+# define LC_L0S_INACTIVITY_SHIFT 8
+# define LC_L1_INACTIVITY(x) ((x) << 12)
+# define LC_L1_INACTIVITY_MASK (0xf << 12)
+# define LC_L1_INACTIVITY_SHIFT 12
+# define LC_PMI_TO_L1_DIS (1 << 16)
+# define LC_ASPM_TO_L1_DIS (1 << 24)
+
#define PCIE_LC_LINK_WIDTH_CNTL 0x100100A2 /* PCIE */
# define LC_LINK_WIDTH_SHIFT 0
# define LC_LINK_WIDTH_MASK 0x7
@@ -65,7 +124,12 @@
# define LC_DYN_LANES_PWR_STATE(x) ((x) << 21)
# define LC_DYN_LANES_PWR_STATE_MASK (0x3 << 21)
# define LC_DYN_LANES_PWR_STATE_SHIFT 21
-
+#define PCIE_LC_N_FTS_CNTL 0x100100a3 /* PCIE */
+# define LC_XMIT_N_FTS(x) ((x) << 0)
+# define LC_XMIT_N_FTS_MASK (0xff << 0)
+# define LC_XMIT_N_FTS_SHIFT 0
+# define LC_XMIT_N_FTS_OVERRIDE_EN (1 << 8)
+# define LC_N_FTS_MASK (0xff << 24)
#define PCIE_LC_SPEED_CNTL 0x100100A4 /* PCIE */
# define LC_GEN2_EN_STRAP (1 << 0)
# define LC_GEN3_EN_STRAP (1 << 1)
@@ -87,6 +151,12 @@
# define LC_OTHER_SIDE_EVER_SENT_GEN3 (1 << 20)
# define LC_OTHER_SIDE_SUPPORTS_GEN3 (1 << 21)
+#define PCIE_LC_CNTL2 0x100100B1 /* PCIE */
+# define LC_ALLOW_PDWN_IN_L1 (1 << 17)
+# define LC_ALLOW_PDWN_IN_L23 (1 << 18)
+
+#define PCIE_LC_CNTL3 0x100100B5 /* PCIE */
+# define LC_GO_TO_RECOVERY (1 << 30)
#define PCIE_LC_CNTL4 0x100100B6 /* PCIE */
# define LC_REDO_EQ (1 << 5)
# define LC_SET_QUIESCE (1 << 13)
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 13/53] drm/radeon/cik: restructure rlc setup
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (11 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 12/53] drm/radeon: add support for ASPM on CIK asics Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 14/53] drm/radeon: clean up sumo_rlc_init() for code sharing Alex Deucher
` (40 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Restructure rlc setup to handle clock and power
gating.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cik.c | 84 +++++++++++++++++++++++++++----------------
drivers/gpu/drm/radeon/cikd.h | 1 +
drivers/gpu/drm/radeon/si.c | 2 +-
3 files changed, 56 insertions(+), 31 deletions(-)
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 0da9d67..e92a972 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -60,6 +60,7 @@ extern bool evergreen_is_display_hung(struct radeon_device *rdev);
extern void si_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
extern void si_rlc_fini(struct radeon_device *rdev);
extern int si_rlc_init(struct radeon_device *rdev);
+extern void si_rlc_reset(struct radeon_device *rdev);
static void cik_rlc_stop(struct radeon_device *rdev);
static void cik_pcie_gen3_enable(struct radeon_device *rdev);
static void cik_program_aspm(struct radeon_device *rdev);
@@ -4728,31 +4729,34 @@ void cik_dma_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm
* variety of functions, the most important of which is
* the interrupt controller.
*/
-/**
- * cik_rlc_stop - stop the RLC ME
- *
- * @rdev: radeon_device pointer
- *
- * Halt the RLC ME (MicroEngine) (CIK).
- */
-static void cik_rlc_stop(struct radeon_device *rdev)
+static void cik_enable_gui_idle_interrupt(struct radeon_device *rdev,
+ bool enable)
{
- int i, j, k;
- u32 mask, tmp;
+ u32 tmp = RREG32(CP_INT_CNTL_RING0);
- tmp = RREG32(CP_INT_CNTL_RING0);
- tmp &= ~(CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE);
+ if (enable)
+ tmp |= (CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE);
+ else
+ tmp &= ~(CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE);
WREG32(CP_INT_CNTL_RING0, tmp);
+}
- RREG32(CB_CGTT_SCLK_CTRL);
- RREG32(CB_CGTT_SCLK_CTRL);
- RREG32(CB_CGTT_SCLK_CTRL);
- RREG32(CB_CGTT_SCLK_CTRL);
+static void cik_enable_lbpw(struct radeon_device *rdev, bool enable)
+{
+ u32 tmp;
- tmp = RREG32(RLC_CGCG_CGLS_CTRL) & 0xfffffffc;
- WREG32(RLC_CGCG_CGLS_CTRL, tmp);
+ tmp = RREG32(RLC_LB_CNTL);
+ if (enable)
+ tmp |= LOAD_BALANCE_ENABLE;
+ else
+ tmp &= ~LOAD_BALANCE_ENABLE;
+ WREG32(RLC_LB_CNTL, tmp);
+}
- WREG32(RLC_CNTL, 0);
+static void cik_wait_for_rlc_serdes(struct radeon_device *rdev)
+{
+ u32 i, j, k;
+ u32 mask;
for (i = 0; i < rdev->config.cik.max_shader_engines; i++) {
for (j = 0; j < rdev->config.cik.max_sh_per_se; j++) {
@@ -4775,6 +4779,32 @@ static void cik_rlc_stop(struct radeon_device *rdev)
}
/**
+ * cik_rlc_stop - stop the RLC ME
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Halt the RLC ME (MicroEngine) (CIK).
+ */
+static void cik_rlc_stop(struct radeon_device *rdev)
+{
+ u32 tmp;
+
+ cik_enable_gui_idle_interrupt(rdev, false);
+
+ RREG32(CB_CGTT_SCLK_CTRL);
+ RREG32(CB_CGTT_SCLK_CTRL);
+ RREG32(CB_CGTT_SCLK_CTRL);
+ RREG32(CB_CGTT_SCLK_CTRL);
+
+ tmp = RREG32(RLC_CGCG_CGLS_CTRL) & 0xfffffffc;
+ WREG32(RLC_CGCG_CGLS_CTRL, tmp);
+
+ WREG32(RLC_CNTL, 0);
+
+ cik_wait_for_rlc_serdes(rdev);
+}
+
+/**
* cik_rlc_start - start the RLC ME
*
* @rdev: radeon_device pointer
@@ -4783,13 +4813,9 @@ static void cik_rlc_stop(struct radeon_device *rdev)
*/
static void cik_rlc_start(struct radeon_device *rdev)
{
- u32 tmp;
-
WREG32(RLC_CNTL, RLC_ENABLE);
- tmp = RREG32(CP_INT_CNTL_RING0);
- tmp |= (CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE);
- WREG32(CP_INT_CNTL_RING0, tmp);
+ cik_enable_gui_idle_interrupt(rdev, true);
udelay(50);
}
@@ -4827,12 +4853,7 @@ static int cik_rlc_resume(struct radeon_device *rdev)
cik_rlc_stop(rdev);
- WREG32(GRBM_SOFT_RESET, SOFT_RESET_RLC);
- RREG32(GRBM_SOFT_RESET);
- udelay(50);
- WREG32(GRBM_SOFT_RESET, 0);
- RREG32(GRBM_SOFT_RESET);
- udelay(50);
+ si_rlc_reset(rdev);
WREG32(RLC_LB_CNTR_INIT, 0);
WREG32(RLC_LB_CNTR_MAX, 0x00008000);
@@ -4851,6 +4872,9 @@ static int cik_rlc_resume(struct radeon_device *rdev)
WREG32(RLC_GPM_UCODE_DATA, be32_to_cpup(fw_data++));
WREG32(RLC_GPM_UCODE_ADDR, 0);
+ /* XXX - find out what chips support lbpw */
+ cik_enable_lbpw(rdev, false);
+
/* XXX */
clear_state_info[0] = 0;//upper_32_bits(rdev->rlc.save_restore_gpu_addr);
clear_state_info[1] = 0;//rdev->rlc.save_restore_gpu_addr;
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index d50593f..4742f3d 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -905,6 +905,7 @@
#define RLC_LB_CNTR_MAX 0xC348
#define RLC_LB_CNTL 0xC364
+# define LOAD_BALANCE_ENABLE (1 << 0)
#define RLC_LB_CNTR_INIT 0xC36C
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index daa8d2d..610adfc 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -5435,7 +5435,7 @@ int si_rlc_init(struct radeon_device *rdev)
return 0;
}
-static void si_rlc_reset(struct radeon_device *rdev)
+void si_rlc_reset(struct radeon_device *rdev)
{
u32 tmp = RREG32(GRBM_SOFT_RESET);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 14/53] drm/radeon: clean up sumo_rlc_init() for code sharing
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (12 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 13/53] drm/radeon/cik: restructure rlc setup Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 15/53] drm/radeon: convert SI,CIK to use sumo_rlc functions Alex Deucher
` (39 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
This will eventually be shared with newer asics to
reduce code duplication.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/evergreen.c | 221 +++++++++++++++++++------------------
1 file changed, 113 insertions(+), 108 deletions(-)
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index a5ab569..6178408 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -3910,131 +3910,136 @@ int sumo_rlc_init(struct radeon_device *rdev)
dws = rdev->rlc.reg_list_size;
cs_data = rdev->rlc.cs_data;
- /* save restore block */
- if (rdev->rlc.save_restore_obj == NULL) {
- r = radeon_bo_create(rdev, dws * 4, PAGE_SIZE, true,
- RADEON_GEM_DOMAIN_VRAM, NULL, &rdev->rlc.save_restore_obj);
+ if (src_ptr) {
+ /* save restore block */
+ if (rdev->rlc.save_restore_obj == NULL) {
+ r = radeon_bo_create(rdev, dws * 4, PAGE_SIZE, true,
+ RADEON_GEM_DOMAIN_VRAM, NULL, &rdev->rlc.save_restore_obj);
+ if (r) {
+ dev_warn(rdev->dev, "(%d) create RLC sr bo failed\n", r);
+ return r;
+ }
+ }
+
+ r = radeon_bo_reserve(rdev->rlc.save_restore_obj, false);
+ if (unlikely(r != 0)) {
+ sumo_rlc_fini(rdev);
+ return r;
+ }
+ r = radeon_bo_pin(rdev->rlc.save_restore_obj, RADEON_GEM_DOMAIN_VRAM,
+ &rdev->rlc.save_restore_gpu_addr);
if (r) {
- dev_warn(rdev->dev, "(%d) create RLC sr bo failed\n", r);
+ radeon_bo_unreserve(rdev->rlc.save_restore_obj);
+ dev_warn(rdev->dev, "(%d) pin RLC sr bo failed\n", r);
+ sumo_rlc_fini(rdev);
return r;
}
- }
- r = radeon_bo_reserve(rdev->rlc.save_restore_obj, false);
- if (unlikely(r != 0)) {
- sumo_rlc_fini(rdev);
- return r;
- }
- r = radeon_bo_pin(rdev->rlc.save_restore_obj, RADEON_GEM_DOMAIN_VRAM,
- &rdev->rlc.save_restore_gpu_addr);
- if (r) {
+ r = radeon_bo_kmap(rdev->rlc.save_restore_obj, (void **)&rdev->rlc.sr_ptr);
+ if (r) {
+ dev_warn(rdev->dev, "(%d) map RLC sr bo failed\n", r);
+ sumo_rlc_fini(rdev);
+ return r;
+ }
+ /* write the sr buffer */
+ dst_ptr = rdev->rlc.sr_ptr;
+ /* format:
+ * dw0: (reg2 << 16) | reg1
+ * dw1: reg1 save space
+ * dw2: reg2 save space
+ */
+ for (i = 0; i < dws; i++) {
+ data = src_ptr[i] >> 2;
+ i++;
+ if (i < dws)
+ data |= (src_ptr[i] >> 2) << 16;
+ j = (((i - 1) * 3) / 2);
+ dst_ptr[j] = data;
+ }
+ j = ((i * 3) / 2);
+ dst_ptr[j] = RLC_SAVE_RESTORE_LIST_END_MARKER;
+
+ radeon_bo_kunmap(rdev->rlc.save_restore_obj);
radeon_bo_unreserve(rdev->rlc.save_restore_obj);
- dev_warn(rdev->dev, "(%d) pin RLC sr bo failed\n", r);
- sumo_rlc_fini(rdev);
- return r;
- }
- r = radeon_bo_kmap(rdev->rlc.save_restore_obj, (void **)&rdev->rlc.sr_ptr);
- if (r) {
- dev_warn(rdev->dev, "(%d) map RLC sr bo failed\n", r);
- sumo_rlc_fini(rdev);
- return r;
- }
- /* write the sr buffer */
- dst_ptr = rdev->rlc.sr_ptr;
- /* format:
- * dw0: (reg2 << 16) | reg1
- * dw1: reg1 save space
- * dw2: reg2 save space
- */
- for (i = 0; i < dws; i++) {
- data = src_ptr[i] >> 2;
- i++;
- if (i < dws)
- data |= (src_ptr[i] >> 2) << 16;
- j = (((i - 1) * 3) / 2);
- dst_ptr[j] = data;
}
- j = ((i * 3) / 2);
- dst_ptr[j] = RLC_SAVE_RESTORE_LIST_END_MARKER;
-
- radeon_bo_kunmap(rdev->rlc.save_restore_obj);
- radeon_bo_unreserve(rdev->rlc.save_restore_obj);
- /* clear state block */
- reg_list_num = 0;
- dws = 0;
- for (i = 0; cs_data[i].section != NULL; i++) {
- for (j = 0; cs_data[i].section[j].extent != NULL; j++) {
- reg_list_num++;
- dws += cs_data[i].section[j].reg_count;
+ if (cs_data) {
+ /* clear state block */
+ reg_list_num = 0;
+ dws = 0;
+ for (i = 0; cs_data[i].section != NULL; i++) {
+ for (j = 0; cs_data[i].section[j].extent != NULL; j++) {
+ reg_list_num++;
+ dws += cs_data[i].section[j].reg_count;
+ }
}
- }
- reg_list_blk_index = (3 * reg_list_num + 2);
- dws += reg_list_blk_index;
+ reg_list_blk_index = (3 * reg_list_num + 2);
+ dws += reg_list_blk_index;
- if (rdev->rlc.clear_state_obj == NULL) {
- r = radeon_bo_create(rdev, dws * 4, PAGE_SIZE, true,
- RADEON_GEM_DOMAIN_VRAM, NULL, &rdev->rlc.clear_state_obj);
+ if (rdev->rlc.clear_state_obj == NULL) {
+ r = radeon_bo_create(rdev, dws * 4, PAGE_SIZE, true,
+ RADEON_GEM_DOMAIN_VRAM, NULL, &rdev->rlc.clear_state_obj);
+ if (r) {
+ dev_warn(rdev->dev, "(%d) create RLC c bo failed\n", r);
+ sumo_rlc_fini(rdev);
+ return r;
+ }
+ }
+ r = radeon_bo_reserve(rdev->rlc.clear_state_obj, false);
+ if (unlikely(r != 0)) {
+ sumo_rlc_fini(rdev);
+ return r;
+ }
+ r = radeon_bo_pin(rdev->rlc.clear_state_obj, RADEON_GEM_DOMAIN_VRAM,
+ &rdev->rlc.clear_state_gpu_addr);
if (r) {
- dev_warn(rdev->dev, "(%d) create RLC c bo failed\n", r);
+ radeon_bo_unreserve(rdev->rlc.clear_state_obj);
+ dev_warn(rdev->dev, "(%d) pin RLC c bo failed\n", r);
sumo_rlc_fini(rdev);
return r;
}
- }
- r = radeon_bo_reserve(rdev->rlc.clear_state_obj, false);
- if (unlikely(r != 0)) {
- sumo_rlc_fini(rdev);
- return r;
- }
- r = radeon_bo_pin(rdev->rlc.clear_state_obj, RADEON_GEM_DOMAIN_VRAM,
- &rdev->rlc.clear_state_gpu_addr);
- if (r) {
- radeon_bo_unreserve(rdev->rlc.clear_state_obj);
- dev_warn(rdev->dev, "(%d) pin RLC c bo failed\n", r);
- sumo_rlc_fini(rdev);
- return r;
- }
- r = radeon_bo_kmap(rdev->rlc.clear_state_obj, (void **)&rdev->rlc.cs_ptr);
- if (r) {
- dev_warn(rdev->dev, "(%d) map RLC c bo failed\n", r);
- sumo_rlc_fini(rdev);
- return r;
- }
- /* set up the cs buffer */
- dst_ptr = rdev->rlc.cs_ptr;
- reg_list_hdr_blk_index = 0;
- reg_list_mc_addr = rdev->rlc.clear_state_gpu_addr + (reg_list_blk_index * 4);
- data = upper_32_bits(reg_list_mc_addr);
- dst_ptr[reg_list_hdr_blk_index] = data;
- reg_list_hdr_blk_index++;
- for (i = 0; cs_data[i].section != NULL; i++) {
- for (j = 0; cs_data[i].section[j].extent != NULL; j++) {
- reg_num = cs_data[i].section[j].reg_count;
- data = reg_list_mc_addr & 0xffffffff;
- dst_ptr[reg_list_hdr_blk_index] = data;
- reg_list_hdr_blk_index++;
-
- data = (cs_data[i].section[j].reg_index * 4) & 0xffffffff;
- dst_ptr[reg_list_hdr_blk_index] = data;
- reg_list_hdr_blk_index++;
-
- data = 0x08000000 | (reg_num * 4);
- dst_ptr[reg_list_hdr_blk_index] = data;
- reg_list_hdr_blk_index++;
-
- for (k = 0; k < reg_num; k++) {
- data = cs_data[i].section[j].extent[k];
- dst_ptr[reg_list_blk_index + k] = data;
+ r = radeon_bo_kmap(rdev->rlc.clear_state_obj, (void **)&rdev->rlc.cs_ptr);
+ if (r) {
+ dev_warn(rdev->dev, "(%d) map RLC c bo failed\n", r);
+ sumo_rlc_fini(rdev);
+ return r;
+ }
+ /* set up the cs buffer */
+ dst_ptr = rdev->rlc.cs_ptr;
+ reg_list_hdr_blk_index = 0;
+ reg_list_mc_addr = rdev->rlc.clear_state_gpu_addr + (reg_list_blk_index * 4);
+ data = upper_32_bits(reg_list_mc_addr);
+ dst_ptr[reg_list_hdr_blk_index] = data;
+ reg_list_hdr_blk_index++;
+ for (i = 0; cs_data[i].section != NULL; i++) {
+ for (j = 0; cs_data[i].section[j].extent != NULL; j++) {
+ reg_num = cs_data[i].section[j].reg_count;
+ data = reg_list_mc_addr & 0xffffffff;
+ dst_ptr[reg_list_hdr_blk_index] = data;
+ reg_list_hdr_blk_index++;
+
+ data = (cs_data[i].section[j].reg_index * 4) & 0xffffffff;
+ dst_ptr[reg_list_hdr_blk_index] = data;
+ reg_list_hdr_blk_index++;
+
+ data = 0x08000000 | (reg_num * 4);
+ dst_ptr[reg_list_hdr_blk_index] = data;
+ reg_list_hdr_blk_index++;
+
+ for (k = 0; k < reg_num; k++) {
+ data = cs_data[i].section[j].extent[k];
+ dst_ptr[reg_list_blk_index + k] = data;
+ }
+ reg_list_mc_addr += reg_num * 4;
+ reg_list_blk_index += reg_num;
}
- reg_list_mc_addr += reg_num * 4;
- reg_list_blk_index += reg_num;
}
- }
- dst_ptr[reg_list_hdr_blk_index] = RLC_CLEAR_STATE_END_MARKER;
+ dst_ptr[reg_list_hdr_blk_index] = RLC_CLEAR_STATE_END_MARKER;
- radeon_bo_kunmap(rdev->rlc.clear_state_obj);
- radeon_bo_unreserve(rdev->rlc.clear_state_obj);
+ radeon_bo_kunmap(rdev->rlc.clear_state_obj);
+ radeon_bo_unreserve(rdev->rlc.clear_state_obj);
+ }
return 0;
}
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 15/53] drm/radeon: convert SI,CIK to use sumo_rlc functions
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (13 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 14/53] drm/radeon: clean up sumo_rlc_init() for code sharing Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 16/53] drm/radeon: implement clock and power gating for CIK (v3) Alex Deucher
` (38 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
and remove duplicate si_rlc functions.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cik.c | 10 +-
drivers/gpu/drm/radeon/clearstate_cayman.h | 2 +-
drivers/gpu/drm/radeon/clearstate_evergreen.h | 2 +-
drivers/gpu/drm/radeon/evergreen.c | 46 ++++---
drivers/gpu/drm/radeon/ni.c | 6 +-
drivers/gpu/drm/radeon/radeon.h | 4 +-
drivers/gpu/drm/radeon/si.c | 174 ++------------------------
7 files changed, 49 insertions(+), 195 deletions(-)
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index e92a972..8389917 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -57,9 +57,9 @@ extern void r600_ih_ring_fini(struct radeon_device *rdev);
extern void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *save);
extern void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_save *save);
extern bool evergreen_is_display_hung(struct radeon_device *rdev);
+extern void sumo_rlc_fini(struct radeon_device *rdev);
+extern int sumo_rlc_init(struct radeon_device *rdev);
extern void si_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
-extern void si_rlc_fini(struct radeon_device *rdev);
-extern int si_rlc_init(struct radeon_device *rdev);
extern void si_rlc_reset(struct radeon_device *rdev);
static void cik_rlc_stop(struct radeon_device *rdev);
static void cik_pcie_gen3_enable(struct radeon_device *rdev);
@@ -6019,7 +6019,7 @@ static int cik_startup(struct radeon_device *rdev)
cik_gpu_init(rdev);
/* allocate rlc buffers */
- r = si_rlc_init(rdev);
+ r = sumo_rlc_init(rdev);
if (r) {
DRM_ERROR("Failed to init rlc BOs!\n");
return r;
@@ -6343,7 +6343,7 @@ int cik_init(struct radeon_device *rdev)
cik_cp_fini(rdev);
cik_sdma_fini(rdev);
cik_irq_fini(rdev);
- si_rlc_fini(rdev);
+ sumo_rlc_fini(rdev);
cik_mec_fini(rdev);
radeon_wb_fini(rdev);
radeon_ib_pool_fini(rdev);
@@ -6379,7 +6379,7 @@ void cik_fini(struct radeon_device *rdev)
cik_cp_fini(rdev);
cik_sdma_fini(rdev);
cik_irq_fini(rdev);
- si_rlc_fini(rdev);
+ sumo_rlc_fini(rdev);
cik_mec_fini(rdev);
radeon_wb_fini(rdev);
radeon_vm_manager_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/clearstate_cayman.h b/drivers/gpu/drm/radeon/clearstate_cayman.h
index c003394..aa908c5 100644
--- a/drivers/gpu/drm/radeon/clearstate_cayman.h
+++ b/drivers/gpu/drm/radeon/clearstate_cayman.h
@@ -1073,7 +1073,7 @@ static const struct cs_extent_def SECT_CTRLCONST_defs[] =
{SECT_CTRLCONST_def_1, 0x0000f3fc, 2 },
{ 0, 0, 0 }
};
-struct cs_section_def cayman_cs_data[] = {
+static const struct cs_section_def cayman_cs_data[] = {
{ SECT_CONTEXT_defs, SECT_CONTEXT },
{ SECT_CLEAR_defs, SECT_CLEAR },
{ SECT_CTRLCONST_defs, SECT_CTRLCONST },
diff --git a/drivers/gpu/drm/radeon/clearstate_evergreen.h b/drivers/gpu/drm/radeon/clearstate_evergreen.h
index 4791d85..63a1ffb 100644
--- a/drivers/gpu/drm/radeon/clearstate_evergreen.h
+++ b/drivers/gpu/drm/radeon/clearstate_evergreen.h
@@ -1072,7 +1072,7 @@ static const struct cs_extent_def SECT_CTRLCONST_defs[] =
{SECT_CTRLCONST_def_1, 0x0000f3fc, 2 },
{ 0, 0, 0 }
};
-struct cs_section_def evergreen_cs_data[] = {
+static const struct cs_section_def evergreen_cs_data[] = {
{ SECT_CONTEXT_defs, SECT_CONTEXT },
{ SECT_CLEAR_defs, SECT_CLEAR },
{ SECT_CTRLCONST_defs, SECT_CTRLCONST },
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 6178408..6fc876a 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -47,7 +47,7 @@ static const u32 crtc_offsets[6] =
#include "clearstate_evergreen.h"
-static u32 sumo_rlc_save_restore_register_list[] =
+static const u32 sumo_rlc_save_restore_register_list[] =
{
0x98fc,
0x9830,
@@ -131,7 +131,6 @@ static u32 sumo_rlc_save_restore_register_list[] =
0x9150,
0x802c,
};
-static u32 sumo_rlc_save_restore_register_list_size = ARRAY_SIZE(sumo_rlc_save_restore_register_list);
static void evergreen_gpu_init(struct radeon_device *rdev);
void evergreen_fini(struct radeon_device *rdev);
@@ -3898,12 +3897,12 @@ void sumo_rlc_fini(struct radeon_device *rdev)
int sumo_rlc_init(struct radeon_device *rdev)
{
- u32 *src_ptr;
+ const u32 *src_ptr;
volatile u32 *dst_ptr;
u32 dws, data, i, j, k, reg_num;
u32 reg_list_num, reg_list_hdr_blk_index, reg_list_blk_index;
u64 reg_list_mc_addr;
- struct cs_section_def *cs_data;
+ const struct cs_section_def *cs_data;
int r;
src_ptr = rdev->rlc.reg_list;
@@ -3943,22 +3942,28 @@ int sumo_rlc_init(struct radeon_device *rdev)
}
/* write the sr buffer */
dst_ptr = rdev->rlc.sr_ptr;
- /* format:
- * dw0: (reg2 << 16) | reg1
- * dw1: reg1 save space
- * dw2: reg2 save space
- */
- for (i = 0; i < dws; i++) {
- data = src_ptr[i] >> 2;
- i++;
- if (i < dws)
- data |= (src_ptr[i] >> 2) << 16;
- j = (((i - 1) * 3) / 2);
- dst_ptr[j] = data;
+ if (rdev->family >= CHIP_TAHITI) {
+ /* SI */
+ for (i = 0; i < dws; i++)
+ dst_ptr[i] = src_ptr[i];
+ } else {
+ /* ON/LN/TN */
+ /* format:
+ * dw0: (reg2 << 16) | reg1
+ * dw1: reg1 save space
+ * dw2: reg2 save space
+ */
+ for (i = 0; i < dws; i++) {
+ data = src_ptr[i] >> 2;
+ i++;
+ if (i < dws)
+ data |= (src_ptr[i] >> 2) << 16;
+ j = (((i - 1) * 3) / 2);
+ dst_ptr[j] = data;
+ }
+ j = ((i * 3) / 2);
+ dst_ptr[j] = RLC_SAVE_RESTORE_LIST_END_MARKER;
}
- j = ((i * 3) / 2);
- dst_ptr[j] = RLC_SAVE_RESTORE_LIST_END_MARKER;
-
radeon_bo_kunmap(rdev->rlc.save_restore_obj);
radeon_bo_unreserve(rdev->rlc.save_restore_obj);
}
@@ -5152,7 +5157,8 @@ static int evergreen_startup(struct radeon_device *rdev)
/* allocate rlc buffers */
if (rdev->flags & RADEON_IS_IGP) {
rdev->rlc.reg_list = sumo_rlc_save_restore_register_list;
- rdev->rlc.reg_list_size = sumo_rlc_save_restore_register_list_size;
+ rdev->rlc.reg_list_size =
+ (u32)ARRAY_SIZE(sumo_rlc_save_restore_register_list);
rdev->rlc.cs_data = evergreen_cs_data;
r = sumo_rlc_init(rdev);
if (r) {
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index acdd603..21f2ece 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -35,7 +35,7 @@
#include "radeon_ucode.h"
#include "clearstate_cayman.h"
-static u32 tn_rlc_save_restore_register_list[] =
+static const u32 tn_rlc_save_restore_register_list[] =
{
0x98fc,
0x98f0,
@@ -160,7 +160,6 @@ static u32 tn_rlc_save_restore_register_list[] =
0x9830,
0x802c,
};
-static u32 tn_rlc_save_restore_register_list_size = ARRAY_SIZE(tn_rlc_save_restore_register_list);
extern bool evergreen_is_display_hung(struct radeon_device *rdev);
extern void evergreen_print_gpu_status_regs(struct radeon_device *rdev);
@@ -2121,7 +2120,8 @@ static int cayman_startup(struct radeon_device *rdev)
/* allocate rlc buffers */
if (rdev->flags & RADEON_IS_IGP) {
rdev->rlc.reg_list = tn_rlc_save_restore_register_list;
- rdev->rlc.reg_list_size = tn_rlc_save_restore_register_list_size;
+ rdev->rlc.reg_list_size =
+ (u32)ARRAY_SIZE(tn_rlc_save_restore_register_list);
rdev->rlc.cs_data = cayman_cs_data;
r = sumo_rlc_init(rdev);
if (r) {
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 54edeab..0fdce76 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -854,13 +854,13 @@ struct radeon_rlc {
struct radeon_bo *save_restore_obj;
uint64_t save_restore_gpu_addr;
volatile uint32_t *sr_ptr;
- u32 *reg_list;
+ const u32 *reg_list;
u32 reg_list_size;
/* for clear state */
struct radeon_bo *clear_state_obj;
uint64_t clear_state_gpu_addr;
volatile uint32_t *cs_ptr;
- struct cs_section_def *cs_data;
+ const struct cs_section_def *cs_data;
};
int radeon_ib_get(struct radeon_device *rdev, int ring,
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index 610adfc..8b8963d 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -68,6 +68,8 @@ MODULE_FIRMWARE("radeon/HAINAN_smc.bin");
static void si_pcie_gen3_enable(struct radeon_device *rdev);
static void si_program_aspm(struct radeon_device *rdev);
+extern void sumo_rlc_fini(struct radeon_device *rdev);
+extern int sumo_rlc_init(struct radeon_device *rdev);
extern int r600_ih_ring_alloc(struct radeon_device *rdev);
extern void r600_ih_ring_fini(struct radeon_device *rdev);
extern void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev);
@@ -5275,166 +5277,6 @@ static void si_fini_pg(struct radeon_device *rdev)
/*
* RLC
*/
-void si_rlc_fini(struct radeon_device *rdev)
-{
- int r;
-
- /* save restore block */
- if (rdev->rlc.save_restore_obj) {
- r = radeon_bo_reserve(rdev->rlc.save_restore_obj, false);
- if (unlikely(r != 0))
- dev_warn(rdev->dev, "(%d) reserve RLC sr bo failed\n", r);
- radeon_bo_unpin(rdev->rlc.save_restore_obj);
- radeon_bo_unreserve(rdev->rlc.save_restore_obj);
-
- radeon_bo_unref(&rdev->rlc.save_restore_obj);
- rdev->rlc.save_restore_obj = NULL;
- }
-
- /* clear state block */
- if (rdev->rlc.clear_state_obj) {
- r = radeon_bo_reserve(rdev->rlc.clear_state_obj, false);
- if (unlikely(r != 0))
- dev_warn(rdev->dev, "(%d) reserve RLC c bo failed\n", r);
- radeon_bo_unpin(rdev->rlc.clear_state_obj);
- radeon_bo_unreserve(rdev->rlc.clear_state_obj);
-
- radeon_bo_unref(&rdev->rlc.clear_state_obj);
- rdev->rlc.clear_state_obj = NULL;
- }
-}
-
-#define RLC_CLEAR_STATE_END_MARKER 0x00000001
-
-int si_rlc_init(struct radeon_device *rdev)
-{
- volatile u32 *dst_ptr;
- u32 dws, data, i, j, k, reg_num;
- u32 reg_list_num, reg_list_hdr_blk_index, reg_list_blk_index;
- u64 reg_list_mc_addr;
- const struct cs_section_def *cs_data = si_cs_data;
- int r;
-
- /* save restore block */
- if (rdev->rlc.save_restore_obj == NULL) {
- r = radeon_bo_create(rdev, RADEON_GPU_PAGE_SIZE, PAGE_SIZE, true,
- RADEON_GEM_DOMAIN_VRAM, NULL,
- &rdev->rlc.save_restore_obj);
- if (r) {
- dev_warn(rdev->dev, "(%d) create RLC sr bo failed\n", r);
- return r;
- }
- }
-
- r = radeon_bo_reserve(rdev->rlc.save_restore_obj, false);
- if (unlikely(r != 0)) {
- si_rlc_fini(rdev);
- return r;
- }
- r = radeon_bo_pin(rdev->rlc.save_restore_obj, RADEON_GEM_DOMAIN_VRAM,
- &rdev->rlc.save_restore_gpu_addr);
- if (r) {
- radeon_bo_unreserve(rdev->rlc.save_restore_obj);
- dev_warn(rdev->dev, "(%d) pin RLC sr bo failed\n", r);
- si_rlc_fini(rdev);
- return r;
- }
-
- if (rdev->family == CHIP_VERDE) {
- r = radeon_bo_kmap(rdev->rlc.save_restore_obj, (void **)&rdev->rlc.sr_ptr);
- if (r) {
- dev_warn(rdev->dev, "(%d) map RLC sr bo failed\n", r);
- si_rlc_fini(rdev);
- return r;
- }
- /* write the sr buffer */
- dst_ptr = rdev->rlc.sr_ptr;
- for (i = 0; i < ARRAY_SIZE(verde_rlc_save_restore_register_list); i++) {
- dst_ptr[i] = verde_rlc_save_restore_register_list[i];
- }
- radeon_bo_kunmap(rdev->rlc.save_restore_obj);
- }
- radeon_bo_unreserve(rdev->rlc.save_restore_obj);
-
- /* clear state block */
- reg_list_num = 0;
- dws = 0;
- for (i = 0; cs_data[i].section != NULL; i++) {
- for (j = 0; cs_data[i].section[j].extent != NULL; j++) {
- reg_list_num++;
- dws += cs_data[i].section[j].reg_count;
- }
- }
- reg_list_blk_index = (3 * reg_list_num + 2);
- dws += reg_list_blk_index;
-
- if (rdev->rlc.clear_state_obj == NULL) {
- r = radeon_bo_create(rdev, dws * 4, PAGE_SIZE, true,
- RADEON_GEM_DOMAIN_VRAM, NULL, &rdev->rlc.clear_state_obj);
- if (r) {
- dev_warn(rdev->dev, "(%d) create RLC c bo failed\n", r);
- si_rlc_fini(rdev);
- return r;
- }
- }
- r = radeon_bo_reserve(rdev->rlc.clear_state_obj, false);
- if (unlikely(r != 0)) {
- si_rlc_fini(rdev);
- return r;
- }
- r = radeon_bo_pin(rdev->rlc.clear_state_obj, RADEON_GEM_DOMAIN_VRAM,
- &rdev->rlc.clear_state_gpu_addr);
- if (r) {
-
- radeon_bo_unreserve(rdev->rlc.clear_state_obj);
- dev_warn(rdev->dev, "(%d) pin RLC c bo failed\n", r);
- si_rlc_fini(rdev);
- return r;
- }
- r = radeon_bo_kmap(rdev->rlc.clear_state_obj, (void **)&rdev->rlc.cs_ptr);
- if (r) {
- dev_warn(rdev->dev, "(%d) map RLC c bo failed\n", r);
- si_rlc_fini(rdev);
- return r;
- }
- /* set up the cs buffer */
- dst_ptr = rdev->rlc.cs_ptr;
- reg_list_hdr_blk_index = 0;
- reg_list_mc_addr = rdev->rlc.clear_state_gpu_addr + (reg_list_blk_index * 4);
- data = upper_32_bits(reg_list_mc_addr);
- dst_ptr[reg_list_hdr_blk_index] = data;
- reg_list_hdr_blk_index++;
- for (i = 0; cs_data[i].section != NULL; i++) {
- for (j = 0; cs_data[i].section[j].extent != NULL; j++) {
- reg_num = cs_data[i].section[j].reg_count;
- data = reg_list_mc_addr & 0xffffffff;
- dst_ptr[reg_list_hdr_blk_index] = data;
- reg_list_hdr_blk_index++;
-
- data = (cs_data[i].section[j].reg_index * 4) & 0xffffffff;
- dst_ptr[reg_list_hdr_blk_index] = data;
- reg_list_hdr_blk_index++;
-
- data = 0x08000000 | (reg_num * 4);
- dst_ptr[reg_list_hdr_blk_index] = data;
- reg_list_hdr_blk_index++;
-
- for (k = 0; k < reg_num; k++) {
- data = cs_data[i].section[j].extent[k];
- dst_ptr[reg_list_blk_index + k] = data;
- }
- reg_list_mc_addr += reg_num * 4;
- reg_list_blk_index += reg_num;
- }
- }
- dst_ptr[reg_list_hdr_blk_index] = RLC_CLEAR_STATE_END_MARKER;
-
- radeon_bo_kunmap(rdev->rlc.clear_state_obj);
- radeon_bo_unreserve(rdev->rlc.clear_state_obj);
-
- return 0;
-}
-
void si_rlc_reset(struct radeon_device *rdev)
{
u32 tmp = RREG32(GRBM_SOFT_RESET);
@@ -6449,7 +6291,13 @@ static int si_startup(struct radeon_device *rdev)
si_gpu_init(rdev);
/* allocate rlc buffers */
- r = si_rlc_init(rdev);
+ if (rdev->family == CHIP_VERDE) {
+ rdev->rlc.reg_list = verde_rlc_save_restore_register_list;
+ rdev->rlc.reg_list_size =
+ (u32)ARRAY_SIZE(verde_rlc_save_restore_register_list);
+ }
+ rdev->rlc.cs_data = si_cs_data;
+ r = sumo_rlc_init(rdev);
if (r) {
DRM_ERROR("Failed to init rlc BOs!\n");
return r;
@@ -6735,7 +6583,7 @@ int si_init(struct radeon_device *rdev)
si_cp_fini(rdev);
cayman_dma_fini(rdev);
si_irq_fini(rdev);
- si_rlc_fini(rdev);
+ sumo_rlc_fini(rdev);
radeon_wb_fini(rdev);
radeon_ib_pool_fini(rdev);
radeon_vm_manager_fini(rdev);
@@ -6761,7 +6609,7 @@ void si_fini(struct radeon_device *rdev)
si_cp_fini(rdev);
cayman_dma_fini(rdev);
si_irq_fini(rdev);
- si_rlc_fini(rdev);
+ sumo_rlc_fini(rdev);
si_fini_cg(rdev);
si_fini_pg(rdev);
radeon_wb_fini(rdev);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 16/53] drm/radeon: implement clock and power gating for CIK (v3)
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (14 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 15/53] drm/radeon: convert SI,CIK to use sumo_rlc functions Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 17/53] drm/radeon: add indirect accessors for dift registers on CIK Alex Deucher
` (37 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Only the APUs support power gating.
v2: disable cgcg for now
v3: workaround hw issue in mgcg
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cik.c | 1475 +++++++++++++++++++++++++++++++-
drivers/gpu/drm/radeon/cikd.h | 95 +-
drivers/gpu/drm/radeon/clearstate_ci.h | 944 ++++++++++++++++++++
drivers/gpu/drm/radeon/evergreen.c | 57 +-
drivers/gpu/drm/radeon/radeon.h | 14 +
drivers/gpu/drm/radeon/radeon_asic.c | 2 +
drivers/gpu/drm/radeon/si.c | 2 +-
7 files changed, 2565 insertions(+), 24 deletions(-)
create mode 100644 drivers/gpu/drm/radeon/clearstate_ci.h
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 8389917..a36e98c 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -31,6 +31,7 @@
#include "atom.h"
#include "cik_blit_shaders.h"
#include "radeon_ucode.h"
+#include "clearstate_ci.h"
MODULE_FIRMWARE("radeon/BONAIRE_pfp.bin");
MODULE_FIRMWARE("radeon/BONAIRE_me.bin");
@@ -61,9 +62,12 @@ extern void sumo_rlc_fini(struct radeon_device *rdev);
extern int sumo_rlc_init(struct radeon_device *rdev);
extern void si_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
extern void si_rlc_reset(struct radeon_device *rdev);
+extern void si_init_uvd_internal_cg(struct radeon_device *rdev);
static void cik_rlc_stop(struct radeon_device *rdev);
static void cik_pcie_gen3_enable(struct radeon_device *rdev);
static void cik_program_aspm(struct radeon_device *rdev);
+static void cik_init_pg(struct radeon_device *rdev);
+static void cik_init_cg(struct radeon_device *rdev);
/*
* Indirect registers accessor
@@ -86,6 +90,778 @@ void cik_pciep_wreg(struct radeon_device *rdev, u32 reg, u32 v)
(void)RREG32(PCIE_DATA);
}
+static const u32 spectre_rlc_save_restore_register_list[] =
+{
+ (0x0e00 << 16) | (0xc12c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc140 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc150 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc15c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc168 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc170 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc178 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc204 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2b4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2b8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2bc >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2c0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8228 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x829c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x869c >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x98f4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x98f8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9900 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc260 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x90e8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3c000 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3c00c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8c1c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9700 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0x8e00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0x9e00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0xae00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0xbe00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x89bc >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8900 >> 2),
+ 0x00000000,
+ 0x3,
+ (0x0e00 << 16) | (0xc130 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc134 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc1fc >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc208 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc264 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc268 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc26c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc270 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc274 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc278 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc27c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc280 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc284 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc288 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc28c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc290 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc294 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc298 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc29c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2a0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2a4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2a8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2ac >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2b0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x301d0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30238 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30250 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30254 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30258 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3025c >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0xc900 >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0xc900 >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0xc900 >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0xc900 >> 2),
+ 0x00000000,
+ (0x8e00 << 16) | (0xc900 >> 2),
+ 0x00000000,
+ (0x9e00 << 16) | (0xc900 >> 2),
+ 0x00000000,
+ (0xae00 << 16) | (0xc900 >> 2),
+ 0x00000000,
+ (0xbe00 << 16) | (0xc900 >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0xc904 >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0xc904 >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0xc904 >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0xc904 >> 2),
+ 0x00000000,
+ (0x8e00 << 16) | (0xc904 >> 2),
+ 0x00000000,
+ (0x9e00 << 16) | (0xc904 >> 2),
+ 0x00000000,
+ (0xae00 << 16) | (0xc904 >> 2),
+ 0x00000000,
+ (0xbe00 << 16) | (0xc904 >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0xc908 >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0xc908 >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0xc908 >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0xc908 >> 2),
+ 0x00000000,
+ (0x8e00 << 16) | (0xc908 >> 2),
+ 0x00000000,
+ (0x9e00 << 16) | (0xc908 >> 2),
+ 0x00000000,
+ (0xae00 << 16) | (0xc908 >> 2),
+ 0x00000000,
+ (0xbe00 << 16) | (0xc908 >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0xc90c >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0xc90c >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0xc90c >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0xc90c >> 2),
+ 0x00000000,
+ (0x8e00 << 16) | (0xc90c >> 2),
+ 0x00000000,
+ (0x9e00 << 16) | (0xc90c >> 2),
+ 0x00000000,
+ (0xae00 << 16) | (0xc90c >> 2),
+ 0x00000000,
+ (0xbe00 << 16) | (0xc90c >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0xc910 >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0xc910 >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0xc910 >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0xc910 >> 2),
+ 0x00000000,
+ (0x8e00 << 16) | (0xc910 >> 2),
+ 0x00000000,
+ (0x9e00 << 16) | (0xc910 >> 2),
+ 0x00000000,
+ (0xae00 << 16) | (0xc910 >> 2),
+ 0x00000000,
+ (0xbe00 << 16) | (0xc910 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc99c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9834 >> 2),
+ 0x00000000,
+ (0x0000 << 16) | (0x30f00 >> 2),
+ 0x00000000,
+ (0x0001 << 16) | (0x30f00 >> 2),
+ 0x00000000,
+ (0x0000 << 16) | (0x30f04 >> 2),
+ 0x00000000,
+ (0x0001 << 16) | (0x30f04 >> 2),
+ 0x00000000,
+ (0x0000 << 16) | (0x30f08 >> 2),
+ 0x00000000,
+ (0x0001 << 16) | (0x30f08 >> 2),
+ 0x00000000,
+ (0x0000 << 16) | (0x30f0c >> 2),
+ 0x00000000,
+ (0x0001 << 16) | (0x30f0c >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x9b7c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8a14 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8a18 >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x30a00 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8bf0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8bcc >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8b24 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30a04 >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x30a10 >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x30a14 >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x30a18 >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x30a2c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc700 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc704 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc708 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc768 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc770 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc774 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc778 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc77c >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc780 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc784 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc788 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc78c >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc798 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc79c >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc7a0 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc7a4 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc7a8 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc7ac >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc7b0 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc7b4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9100 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3c010 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x92a8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x92ac >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x92b4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x92b8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x92bc >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x92c0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x92c4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x92c8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x92cc >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x92d0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8c00 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8c04 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8c20 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8c38 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8c3c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xae00 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9604 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac08 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac0c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac10 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac14 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac58 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac68 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac6c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac70 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac74 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac78 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac7c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac80 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac84 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac88 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac8c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x970c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9714 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9718 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x971c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0x8e00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0x9e00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0xae00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0xbe00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xcd10 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xcd14 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88b0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88b4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88b8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88bc >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0x89c0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88c4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88c8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88d0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88d4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88d8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8980 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30938 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3093c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30940 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x89a0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30900 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30904 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x89b4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3c210 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3c214 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3c218 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8904 >> 2),
+ 0x00000000,
+ 0x5,
+ (0x0e00 << 16) | (0x8c28 >> 2),
+ (0x0e00 << 16) | (0x8c2c >> 2),
+ (0x0e00 << 16) | (0x8c30 >> 2),
+ (0x0e00 << 16) | (0x8c34 >> 2),
+ (0x0e00 << 16) | (0x9600 >> 2),
+};
+
+static const u32 kalindi_rlc_save_restore_register_list[] =
+{
+ (0x0e00 << 16) | (0xc12c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc140 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc150 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc15c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc168 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc170 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc204 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2b4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2b8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2bc >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2c0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8228 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x829c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x869c >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x98f4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x98f8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9900 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc260 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x90e8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3c000 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3c00c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8c1c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9700 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0xcd20 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x89bc >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8900 >> 2),
+ 0x00000000,
+ 0x3,
+ (0x0e00 << 16) | (0xc130 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc134 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc1fc >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc208 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc264 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc268 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc26c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc270 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc274 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc28c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc290 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc294 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc298 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2a0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2a4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2a8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc2ac >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x301d0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30238 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30250 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30254 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30258 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3025c >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0xc900 >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0xc900 >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0xc900 >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0xc900 >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0xc904 >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0xc904 >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0xc904 >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0xc904 >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0xc908 >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0xc908 >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0xc908 >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0xc908 >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0xc90c >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0xc90c >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0xc90c >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0xc90c >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0xc910 >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0xc910 >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0xc910 >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0xc910 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc99c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9834 >> 2),
+ 0x00000000,
+ (0x0000 << 16) | (0x30f00 >> 2),
+ 0x00000000,
+ (0x0000 << 16) | (0x30f04 >> 2),
+ 0x00000000,
+ (0x0000 << 16) | (0x30f08 >> 2),
+ 0x00000000,
+ (0x0000 << 16) | (0x30f0c >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x9b7c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8a14 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8a18 >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x30a00 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8bf0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8bcc >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8b24 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30a04 >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x30a10 >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x30a14 >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x30a18 >> 2),
+ 0x00000000,
+ (0x0600 << 16) | (0x30a2c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc700 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc704 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc708 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xc768 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc770 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc774 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc798 >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0xc79c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9100 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3c010 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8c00 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8c04 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8c20 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8c38 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8c3c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xae00 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9604 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac08 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac0c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac10 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac14 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac58 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac68 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac6c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac70 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac74 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac78 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac7c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac80 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac84 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac88 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xac8c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x970c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9714 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x9718 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x971c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0x4e00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0x5e00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0x6e00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0x7e00 << 16) | (0x31068 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xcd10 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0xcd14 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88b0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88b4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88b8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88bc >> 2),
+ 0x00000000,
+ (0x0400 << 16) | (0x89c0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88c4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88c8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88d0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88d4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x88d8 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8980 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30938 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3093c >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30940 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x89a0 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30900 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x30904 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x89b4 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3e1fc >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3c210 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3c214 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x3c218 >> 2),
+ 0x00000000,
+ (0x0e00 << 16) | (0x8904 >> 2),
+ 0x00000000,
+ 0x5,
+ (0x0e00 << 16) | (0x8c28 >> 2),
+ (0x0e00 << 16) | (0x8c2c >> 2),
+ (0x0e00 << 16) | (0x8c30 >> 2),
+ (0x0e00 << 16) | (0x8c34 >> 2),
+ (0x0e00 << 16) | (0x9600 >> 2),
+};
+
static const u32 bonaire_golden_spm_registers[] =
{
0x30800, 0xe0ffffff, 0xe0000000
@@ -4778,6 +5554,39 @@ static void cik_wait_for_rlc_serdes(struct radeon_device *rdev)
}
}
+static void cik_update_rlc(struct radeon_device *rdev, u32 rlc)
+{
+ u32 tmp;
+
+ tmp = RREG32(RLC_CNTL);
+ if (tmp != rlc)
+ WREG32(RLC_CNTL, rlc);
+}
+
+static u32 cik_halt_rlc(struct radeon_device *rdev)
+{
+ u32 data, orig;
+
+ orig = data = RREG32(RLC_CNTL);
+
+ if (data & RLC_ENABLE) {
+ u32 i;
+
+ data &= ~RLC_ENABLE;
+ WREG32(RLC_CNTL, data);
+
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ if ((RREG32(RLC_GPM_STAT) & RLC_GPM_BUSY) == 0)
+ break;
+ udelay(1);
+ }
+
+ cik_wait_for_rlc_serdes(rdev);
+ }
+
+ return orig;
+}
+
/**
* cik_rlc_stop - stop the RLC ME
*
@@ -4787,20 +5596,10 @@ static void cik_wait_for_rlc_serdes(struct radeon_device *rdev)
*/
static void cik_rlc_stop(struct radeon_device *rdev)
{
- u32 tmp;
+ WREG32(RLC_CNTL, 0);
cik_enable_gui_idle_interrupt(rdev, false);
- RREG32(CB_CGTT_SCLK_CTRL);
- RREG32(CB_CGTT_SCLK_CTRL);
- RREG32(CB_CGTT_SCLK_CTRL);
- RREG32(CB_CGTT_SCLK_CTRL);
-
- tmp = RREG32(RLC_CGCG_CGLS_CTRL) & 0xfffffffc;
- WREG32(RLC_CGCG_CGLS_CTRL, tmp);
-
- WREG32(RLC_CNTL, 0);
-
cik_wait_for_rlc_serdes(rdev);
}
@@ -4831,8 +5630,7 @@ static void cik_rlc_start(struct radeon_device *rdev)
*/
static int cik_rlc_resume(struct radeon_device *rdev)
{
- u32 i, size;
- u32 clear_state_info[3];
+ u32 i, size, tmp;
const __be32 *fw_data;
if (!rdev->rlc_fw)
@@ -4853,8 +5651,16 @@ static int cik_rlc_resume(struct radeon_device *rdev)
cik_rlc_stop(rdev);
+ /* disable CG */
+ tmp = RREG32(RLC_CGCG_CGLS_CTRL) & 0xfffffffc;
+ WREG32(RLC_CGCG_CGLS_CTRL, tmp);
+
si_rlc_reset(rdev);
+ cik_init_pg(rdev);
+
+ cik_init_cg(rdev);
+
WREG32(RLC_LB_CNTR_INIT, 0);
WREG32(RLC_LB_CNTR_MAX, 0x00008000);
@@ -4875,20 +5681,634 @@ static int cik_rlc_resume(struct radeon_device *rdev)
/* XXX - find out what chips support lbpw */
cik_enable_lbpw(rdev, false);
- /* XXX */
- clear_state_info[0] = 0;//upper_32_bits(rdev->rlc.save_restore_gpu_addr);
- clear_state_info[1] = 0;//rdev->rlc.save_restore_gpu_addr;
- clear_state_info[2] = 0;//cik_default_size;
- WREG32(RLC_GPM_SCRATCH_ADDR, 0x3d);
- for (i = 0; i < 3; i++)
- WREG32(RLC_GPM_SCRATCH_DATA, clear_state_info[i]);
- WREG32(RLC_DRIVER_DMA_STATUS, 0);
+ if (rdev->family == CHIP_BONAIRE)
+ WREG32(RLC_DRIVER_DMA_STATUS, 0);
cik_rlc_start(rdev);
return 0;
}
+static void cik_enable_cgcg(struct radeon_device *rdev, bool enable)
+{
+ u32 data, orig, tmp, tmp2;
+
+ orig = data = RREG32(RLC_CGCG_CGLS_CTRL);
+
+ cik_enable_gui_idle_interrupt(rdev, enable);
+
+ if (enable) {
+ tmp = cik_halt_rlc(rdev);
+
+ cik_select_se_sh(rdev, 0xffffffff, 0xffffffff);
+ WREG32(RLC_SERDES_WR_CU_MASTER_MASK, 0xffffffff);
+ WREG32(RLC_SERDES_WR_NONCU_MASTER_MASK, 0xffffffff);
+ tmp2 = BPM_ADDR_MASK | CGCG_OVERRIDE_0 | CGLS_ENABLE;
+ WREG32(RLC_SERDES_WR_CTRL, tmp2);
+
+ cik_update_rlc(rdev, tmp);
+
+ data |= CGCG_EN | CGLS_EN;
+ } else {
+ RREG32(CB_CGTT_SCLK_CTRL);
+ RREG32(CB_CGTT_SCLK_CTRL);
+ RREG32(CB_CGTT_SCLK_CTRL);
+ RREG32(CB_CGTT_SCLK_CTRL);
+
+ data &= ~(CGCG_EN | CGLS_EN);
+ }
+
+ if (orig != data)
+ WREG32(RLC_CGCG_CGLS_CTRL, data);
+
+}
+
+static void cik_enable_mgcg(struct radeon_device *rdev, bool enable)
+{
+ u32 data, orig, tmp = 0;
+
+ if (enable) {
+ orig = data = RREG32(CP_MEM_SLP_CNTL);
+ data |= CP_MEM_LS_EN;
+ if (orig != data)
+ WREG32(CP_MEM_SLP_CNTL, data);
+
+ orig = data = RREG32(RLC_CGTT_MGCG_OVERRIDE);
+ data &= 0xfffffffd;
+ if (orig != data)
+ WREG32(RLC_CGTT_MGCG_OVERRIDE, data);
+
+ tmp = cik_halt_rlc(rdev);
+
+ cik_select_se_sh(rdev, 0xffffffff, 0xffffffff);
+ WREG32(RLC_SERDES_WR_CU_MASTER_MASK, 0xffffffff);
+ WREG32(RLC_SERDES_WR_NONCU_MASTER_MASK, 0xffffffff);
+ data = BPM_ADDR_MASK | MGCG_OVERRIDE_0;
+ WREG32(RLC_SERDES_WR_CTRL, data);
+
+ cik_update_rlc(rdev, tmp);
+
+ orig = data = RREG32(CGTS_SM_CTRL_REG);
+ data &= ~SM_MODE_MASK;
+ data |= SM_MODE(0x2);
+ data |= SM_MODE_ENABLE;
+ data &= ~CGTS_OVERRIDE;
+ data &= ~CGTS_LS_OVERRIDE;
+ data &= ~ON_MONITOR_ADD_MASK;
+ data |= ON_MONITOR_ADD_EN;
+ data |= ON_MONITOR_ADD(0x96);
+ if (orig != data)
+ WREG32(CGTS_SM_CTRL_REG, data);
+ } else {
+ orig = data = RREG32(RLC_CGTT_MGCG_OVERRIDE);
+ data |= 0x00000002;
+ if (orig != data)
+ WREG32(RLC_CGTT_MGCG_OVERRIDE, data);
+
+ data = RREG32(RLC_MEM_SLP_CNTL);
+ if (data & RLC_MEM_LS_EN) {
+ data &= ~RLC_MEM_LS_EN;
+ WREG32(RLC_MEM_SLP_CNTL, data);
+ }
+
+ data = RREG32(CP_MEM_SLP_CNTL);
+ if (data & CP_MEM_LS_EN) {
+ data &= ~CP_MEM_LS_EN;
+ WREG32(CP_MEM_SLP_CNTL, data);
+ }
+
+ orig = data = RREG32(CGTS_SM_CTRL_REG);
+ data |= CGTS_OVERRIDE | CGTS_LS_OVERRIDE;
+ if (orig != data)
+ WREG32(CGTS_SM_CTRL_REG, data);
+
+ tmp = cik_halt_rlc(rdev);
+
+ cik_select_se_sh(rdev, 0xffffffff, 0xffffffff);
+ WREG32(RLC_SERDES_WR_CU_MASTER_MASK, 0xffffffff);
+ WREG32(RLC_SERDES_WR_NONCU_MASTER_MASK, 0xffffffff);
+ data = BPM_ADDR_MASK | MGCG_OVERRIDE_1;
+ WREG32(RLC_SERDES_WR_CTRL, data);
+
+ cik_update_rlc(rdev, tmp);
+ }
+}
+
+static const u32 mc_cg_registers[] =
+{
+ MC_HUB_MISC_HUB_CG,
+ MC_HUB_MISC_SIP_CG,
+ MC_HUB_MISC_VM_CG,
+ MC_XPB_CLK_GAT,
+ ATC_MISC_CG,
+ MC_CITF_MISC_WR_CG,
+ MC_CITF_MISC_RD_CG,
+ MC_CITF_MISC_VM_CG,
+ VM_L2_CG,
+};
+
+static void cik_enable_mc_ls(struct radeon_device *rdev,
+ bool enable)
+{
+ int i;
+ u32 orig, data;
+
+ for (i = 0; i < ARRAY_SIZE(mc_cg_registers); i++) {
+ orig = data = RREG32(mc_cg_registers[i]);
+ if (enable)
+ data |= MC_LS_ENABLE;
+ else
+ data &= ~MC_LS_ENABLE;
+ if (data != orig)
+ WREG32(mc_cg_registers[i], data);
+ }
+}
+
+static void cik_enable_mc_mgcg(struct radeon_device *rdev,
+ bool enable)
+{
+ int i;
+ u32 orig, data;
+
+ for (i = 0; i < ARRAY_SIZE(mc_cg_registers); i++) {
+ orig = data = RREG32(mc_cg_registers[i]);
+ if (enable)
+ data |= MC_CG_ENABLE;
+ else
+ data &= ~MC_CG_ENABLE;
+ if (data != orig)
+ WREG32(mc_cg_registers[i], data);
+ }
+}
+
+static void cik_enable_sdma_mgcg(struct radeon_device *rdev,
+ bool enable)
+{
+ u32 orig, data;
+
+ if (enable) {
+ WREG32(SDMA0_CLK_CTRL + SDMA0_REGISTER_OFFSET, 0x00000100);
+ WREG32(SDMA0_CLK_CTRL + SDMA1_REGISTER_OFFSET, 0x00000100);
+ } else {
+ orig = data = RREG32(SDMA0_CLK_CTRL + SDMA0_REGISTER_OFFSET);
+ data |= 0xff000000;
+ if (data != orig)
+ WREG32(SDMA0_CLK_CTRL + SDMA0_REGISTER_OFFSET, data);
+
+ orig = data = RREG32(SDMA0_CLK_CTRL + SDMA1_REGISTER_OFFSET);
+ data |= 0xff000000;
+ if (data != orig)
+ WREG32(SDMA0_CLK_CTRL + SDMA1_REGISTER_OFFSET, data);
+ }
+}
+
+static void cik_enable_sdma_mgls(struct radeon_device *rdev,
+ bool enable)
+{
+ u32 orig, data;
+
+ if (enable) {
+ orig = data = RREG32(SDMA0_POWER_CNTL + SDMA0_REGISTER_OFFSET);
+ data |= 0x100;
+ if (orig != data)
+ WREG32(SDMA0_POWER_CNTL + SDMA0_REGISTER_OFFSET, data);
+
+ orig = data = RREG32(SDMA0_POWER_CNTL + SDMA1_REGISTER_OFFSET);
+ data |= 0x100;
+ if (orig != data)
+ WREG32(SDMA0_POWER_CNTL + SDMA1_REGISTER_OFFSET, data);
+ } else {
+ orig = data = RREG32(SDMA0_POWER_CNTL + SDMA0_REGISTER_OFFSET);
+ data &= ~0x100;
+ if (orig != data)
+ WREG32(SDMA0_POWER_CNTL + SDMA0_REGISTER_OFFSET, data);
+
+ orig = data = RREG32(SDMA0_POWER_CNTL + SDMA1_REGISTER_OFFSET);
+ data &= ~0x100;
+ if (orig != data)
+ WREG32(SDMA0_POWER_CNTL + SDMA1_REGISTER_OFFSET, data);
+ }
+}
+
+static void cik_enable_uvd_mgcg(struct radeon_device *rdev,
+ bool enable)
+{
+ u32 orig, data;
+
+ if (enable) {
+ data = RREG32_UVD_CTX(UVD_CGC_MEM_CTRL);
+ data = 0xfff;
+ WREG32_UVD_CTX(UVD_CGC_MEM_CTRL, data);
+
+ orig = data = RREG32(UVD_CGC_CTRL);
+ data |= DCM;
+ if (orig != data)
+ WREG32(UVD_CGC_CTRL, data);
+ } else {
+ data = RREG32_UVD_CTX(UVD_CGC_MEM_CTRL);
+ data &= ~0xfff;
+ WREG32_UVD_CTX(UVD_CGC_MEM_CTRL, data);
+
+ orig = data = RREG32(UVD_CGC_CTRL);
+ data &= ~DCM;
+ if (orig != data)
+ WREG32(UVD_CGC_CTRL, data);
+ }
+}
+
+static void cik_enable_hdp_mgcg(struct radeon_device *rdev,
+ bool enable)
+{
+ u32 orig, data;
+
+ orig = data = RREG32(HDP_HOST_PATH_CNTL);
+
+ if (enable)
+ data &= ~CLOCK_GATING_DIS;
+ else
+ data |= CLOCK_GATING_DIS;
+
+ if (orig != data)
+ WREG32(HDP_HOST_PATH_CNTL, data);
+}
+
+static void cik_enable_hdp_ls(struct radeon_device *rdev,
+ bool enable)
+{
+ u32 orig, data;
+
+ orig = data = RREG32(HDP_MEM_POWER_LS);
+
+ if (enable)
+ data |= HDP_LS_ENABLE;
+ else
+ data &= ~HDP_LS_ENABLE;
+
+ if (orig != data)
+ WREG32(HDP_MEM_POWER_LS, data);
+}
+
+void cik_update_cg(struct radeon_device *rdev,
+ u32 block, bool enable)
+{
+ if (block & RADEON_CG_BLOCK_GFX) {
+ /* order matters! */
+ if (enable) {
+ cik_enable_mgcg(rdev, true);
+ cik_enable_cgcg(rdev, true);
+ } else {
+ cik_enable_cgcg(rdev, false);
+ cik_enable_mgcg(rdev, false);
+ }
+ }
+
+ if (block & RADEON_CG_BLOCK_MC) {
+ if (!(rdev->flags & RADEON_IS_IGP)) {
+ cik_enable_mc_mgcg(rdev, enable);
+ cik_enable_mc_ls(rdev, enable);
+ }
+ }
+
+ if (block & RADEON_CG_BLOCK_SDMA) {
+ cik_enable_sdma_mgcg(rdev, enable);
+ cik_enable_sdma_mgls(rdev, enable);
+ }
+
+ if (block & RADEON_CG_BLOCK_UVD) {
+ if (rdev->has_uvd)
+ cik_enable_uvd_mgcg(rdev, enable);
+ }
+
+ if (block & RADEON_CG_BLOCK_HDP) {
+ cik_enable_hdp_mgcg(rdev, enable);
+ cik_enable_hdp_ls(rdev, enable);
+ }
+}
+
+static void cik_init_cg(struct radeon_device *rdev)
+{
+
+ cik_update_cg(rdev, RADEON_CG_BLOCK_GFX, false); /* XXX true */
+
+ if (rdev->has_uvd)
+ si_init_uvd_internal_cg(rdev);
+
+ cik_update_cg(rdev, (RADEON_CG_BLOCK_MC |
+ RADEON_CG_BLOCK_SDMA |
+ RADEON_CG_BLOCK_UVD |
+ RADEON_CG_BLOCK_HDP), true);
+}
+
+static void cik_enable_sck_slowdown_on_pu(struct radeon_device *rdev,
+ bool enable)
+{
+ u32 data, orig;
+
+ orig = data = RREG32(RLC_PG_CNTL);
+ if (enable)
+ data |= SMU_CLK_SLOWDOWN_ON_PU_ENABLE;
+ else
+ data &= ~SMU_CLK_SLOWDOWN_ON_PU_ENABLE;
+ if (orig != data)
+ WREG32(RLC_PG_CNTL, data);
+}
+
+static void cik_enable_sck_slowdown_on_pd(struct radeon_device *rdev,
+ bool enable)
+{
+ u32 data, orig;
+
+ orig = data = RREG32(RLC_PG_CNTL);
+ if (enable)
+ data |= SMU_CLK_SLOWDOWN_ON_PD_ENABLE;
+ else
+ data &= ~SMU_CLK_SLOWDOWN_ON_PD_ENABLE;
+ if (orig != data)
+ WREG32(RLC_PG_CNTL, data);
+}
+
+static void cik_enable_cp_pg(struct radeon_device *rdev, bool enable)
+{
+ u32 data, orig;
+
+ orig = data = RREG32(RLC_PG_CNTL);
+ if (enable)
+ data &= ~DISABLE_CP_PG;
+ else
+ data |= DISABLE_CP_PG;
+ if (orig != data)
+ WREG32(RLC_PG_CNTL, data);
+}
+
+static void cik_enable_gds_pg(struct radeon_device *rdev, bool enable)
+{
+ u32 data, orig;
+
+ orig = data = RREG32(RLC_PG_CNTL);
+ if (enable)
+ data &= ~DISABLE_GDS_PG;
+ else
+ data |= DISABLE_GDS_PG;
+ if (orig != data)
+ WREG32(RLC_PG_CNTL, data);
+}
+
+#define CP_ME_TABLE_SIZE 96
+#define CP_ME_TABLE_OFFSET 2048
+#define CP_MEC_TABLE_OFFSET 4096
+
+void cik_init_cp_pg_table(struct radeon_device *rdev)
+{
+ const __be32 *fw_data;
+ volatile u32 *dst_ptr;
+ int me, i, max_me = 4;
+ u32 bo_offset = 0;
+ u32 table_offset;
+
+ if (rdev->family == CHIP_KAVERI)
+ max_me = 5;
+
+ if (rdev->rlc.cp_table_ptr == NULL)
+ return;
+
+ /* write the cp table buffer */
+ dst_ptr = rdev->rlc.cp_table_ptr;
+ for (me = 0; me < max_me; me++) {
+ if (me == 0) {
+ fw_data = (const __be32 *)rdev->ce_fw->data;
+ table_offset = CP_ME_TABLE_OFFSET;
+ } else if (me == 1) {
+ fw_data = (const __be32 *)rdev->pfp_fw->data;
+ table_offset = CP_ME_TABLE_OFFSET;
+ } else if (me == 2) {
+ fw_data = (const __be32 *)rdev->me_fw->data;
+ table_offset = CP_ME_TABLE_OFFSET;
+ } else {
+ fw_data = (const __be32 *)rdev->mec_fw->data;
+ table_offset = CP_MEC_TABLE_OFFSET;
+ }
+
+ for (i = 0; i < CP_ME_TABLE_SIZE; i ++) {
+ dst_ptr[bo_offset + i] = be32_to_cpu(fw_data[table_offset + i]);
+ }
+ bo_offset += CP_ME_TABLE_SIZE;
+ }
+}
+
+static void cik_enable_gfx_cgpg(struct radeon_device *rdev,
+ bool enable)
+{
+ u32 data, orig;
+
+ if (enable) {
+ orig = data = RREG32(RLC_PG_CNTL);
+ data |= GFX_PG_ENABLE;
+ if (orig != data)
+ WREG32(RLC_PG_CNTL, data);
+
+ orig = data = RREG32(RLC_AUTO_PG_CTRL);
+ data |= AUTO_PG_EN;
+ if (orig != data)
+ WREG32(RLC_AUTO_PG_CTRL, data);
+ } else {
+ orig = data = RREG32(RLC_PG_CNTL);
+ data &= ~GFX_PG_ENABLE;
+ if (orig != data)
+ WREG32(RLC_PG_CNTL, data);
+
+ orig = data = RREG32(RLC_AUTO_PG_CTRL);
+ data &= ~AUTO_PG_EN;
+ if (orig != data)
+ WREG32(RLC_AUTO_PG_CTRL, data);
+
+ data = RREG32(DB_RENDER_CONTROL);
+ }
+}
+
+static u32 cik_get_cu_active_bitmap(struct radeon_device *rdev, u32 se, u32 sh)
+{
+ u32 mask = 0, tmp, tmp1;
+ int i;
+
+ cik_select_se_sh(rdev, se, sh);
+ tmp = RREG32(CC_GC_SHADER_ARRAY_CONFIG);
+ tmp1 = RREG32(GC_USER_SHADER_ARRAY_CONFIG);
+ cik_select_se_sh(rdev, 0xffffffff, 0xffffffff);
+
+ tmp &= 0xffff0000;
+
+ tmp |= tmp1;
+ tmp >>= 16;
+
+ for (i = 0; i < rdev->config.cik.max_cu_per_sh; i ++) {
+ mask <<= 1;
+ mask |= 1;
+ }
+
+ return (~tmp) & mask;
+}
+
+static void cik_init_ao_cu_mask(struct radeon_device *rdev)
+{
+ u32 i, j, k, active_cu_number = 0;
+ u32 mask, counter, cu_bitmap;
+ u32 tmp = 0;
+
+ for (i = 0; i < rdev->config.cik.max_shader_engines; i++) {
+ for (j = 0; j < rdev->config.cik.max_sh_per_se; j++) {
+ mask = 1;
+ cu_bitmap = 0;
+ counter = 0;
+ for (k = 0; k < rdev->config.cik.max_cu_per_sh; k ++) {
+ if (cik_get_cu_active_bitmap(rdev, i, j) & mask) {
+ if (counter < 2)
+ cu_bitmap |= mask;
+ counter ++;
+ }
+ mask <<= 1;
+ }
+
+ active_cu_number += counter;
+ tmp |= (cu_bitmap << (i * 16 + j * 8));
+ }
+ }
+
+ WREG32(RLC_PG_AO_CU_MASK, tmp);
+
+ tmp = RREG32(RLC_MAX_PG_CU);
+ tmp &= ~MAX_PU_CU_MASK;
+ tmp |= MAX_PU_CU(active_cu_number);
+ WREG32(RLC_MAX_PG_CU, tmp);
+}
+
+static void cik_enable_gfx_static_mgpg(struct radeon_device *rdev,
+ bool enable)
+{
+ u32 data, orig;
+
+ orig = data = RREG32(RLC_PG_CNTL);
+ if (enable)
+ data |= STATIC_PER_CU_PG_ENABLE;
+ else
+ data &= ~STATIC_PER_CU_PG_ENABLE;
+ if (orig != data)
+ WREG32(RLC_PG_CNTL, data);
+}
+
+static void cik_enable_gfx_dynamic_mgpg(struct radeon_device *rdev,
+ bool enable)
+{
+ u32 data, orig;
+
+ orig = data = RREG32(RLC_PG_CNTL);
+ if (enable)
+ data |= DYN_PER_CU_PG_ENABLE;
+ else
+ data &= ~DYN_PER_CU_PG_ENABLE;
+ if (orig != data)
+ WREG32(RLC_PG_CNTL, data);
+}
+
+#define RLC_SAVE_AND_RESTORE_STARTING_OFFSET 0x90
+#define RLC_CLEAR_STATE_DESCRIPTOR_OFFSET 0x3D
+
+static void cik_init_gfx_cgpg(struct radeon_device *rdev)
+{
+ u32 data, orig;
+ u32 i;
+
+ if (rdev->rlc.cs_data) {
+ WREG32(RLC_GPM_SCRATCH_ADDR, RLC_CLEAR_STATE_DESCRIPTOR_OFFSET);
+ WREG32(RLC_GPM_SCRATCH_DATA, upper_32_bits(rdev->rlc.clear_state_gpu_addr));
+ WREG32(RLC_GPM_SCRATCH_DATA, rdev->rlc.clear_state_gpu_addr);
+ WREG32(RLC_GPM_SCRATCH_DATA, rdev->rlc.clear_state_size);
+ } else {
+ WREG32(RLC_GPM_SCRATCH_ADDR, RLC_CLEAR_STATE_DESCRIPTOR_OFFSET);
+ for (i = 0; i < 3; i++)
+ WREG32(RLC_GPM_SCRATCH_DATA, 0);
+ }
+ if (rdev->rlc.reg_list) {
+ WREG32(RLC_GPM_SCRATCH_ADDR, RLC_SAVE_AND_RESTORE_STARTING_OFFSET);
+ for (i = 0; i < rdev->rlc.reg_list_size; i++)
+ WREG32(RLC_GPM_SCRATCH_DATA, rdev->rlc.reg_list[i]);
+ }
+
+ orig = data = RREG32(RLC_PG_CNTL);
+ data |= GFX_PG_SRC;
+ if (orig != data)
+ WREG32(RLC_PG_CNTL, data);
+
+ WREG32(RLC_SAVE_AND_RESTORE_BASE, rdev->rlc.save_restore_gpu_addr >> 8);
+ WREG32(RLC_CP_TABLE_RESTORE, rdev->rlc.cp_table_gpu_addr >> 8);
+
+ data = RREG32(CP_RB_WPTR_POLL_CNTL);
+ data &= ~IDLE_POLL_COUNT_MASK;
+ data |= IDLE_POLL_COUNT(0x60);
+ WREG32(CP_RB_WPTR_POLL_CNTL, data);
+
+ data = 0x10101010;
+ WREG32(RLC_PG_DELAY, data);
+
+ data = RREG32(RLC_PG_DELAY_2);
+ data &= ~0xff;
+ data |= 0x3;
+ WREG32(RLC_PG_DELAY_2, data);
+
+ data = RREG32(RLC_AUTO_PG_CTRL);
+ data &= ~GRBM_REG_SGIT_MASK;
+ data |= GRBM_REG_SGIT(0x700);
+ WREG32(RLC_AUTO_PG_CTRL, data);
+
+}
+
+static void cik_update_gfx_pg(struct radeon_device *rdev, bool enable)
+{
+ bool has_pg = false;
+ bool has_dyn_mgpg = false;
+ bool has_static_mgpg = false;
+
+ /* only APUs have PG */
+ if (rdev->flags & RADEON_IS_IGP) {
+ has_pg = true;
+ has_static_mgpg = true;
+ if (rdev->family == CHIP_KAVERI)
+ has_dyn_mgpg = true;
+ }
+
+ if (has_pg) {
+ cik_enable_gfx_cgpg(rdev, enable);
+ if (enable) {
+ cik_enable_gfx_static_mgpg(rdev, has_static_mgpg);
+ cik_enable_gfx_dynamic_mgpg(rdev, has_dyn_mgpg);
+ } else {
+ cik_enable_gfx_static_mgpg(rdev, false);
+ cik_enable_gfx_dynamic_mgpg(rdev, false);
+ }
+ }
+
+}
+
+void cik_init_pg(struct radeon_device *rdev)
+{
+ bool has_pg = false;
+
+ /* only APUs have PG */
+ if (rdev->flags & RADEON_IS_IGP) {
+ /* XXX disable this for now */
+ /* has_pg = true; */
+ }
+
+ if (has_pg) {
+ cik_enable_sck_slowdown_on_pu(rdev, true);
+ cik_enable_sck_slowdown_on_pd(rdev, true);
+ cik_init_gfx_cgpg(rdev);
+ cik_enable_cp_pg(rdev, true);
+ cik_enable_gds_pg(rdev, true);
+ cik_init_ao_cu_mask(rdev);
+ cik_update_gfx_pg(rdev, true);
+ }
+}
+
/*
* Interrupts
* Starting with r6xx, interrupts are handled via a ring buffer.
@@ -6019,6 +7439,19 @@ static int cik_startup(struct radeon_device *rdev)
cik_gpu_init(rdev);
/* allocate rlc buffers */
+ if (rdev->flags & RADEON_IS_IGP) {
+ if (rdev->family == CHIP_KAVERI) {
+ rdev->rlc.reg_list = spectre_rlc_save_restore_register_list;
+ rdev->rlc.reg_list_size =
+ (u32)ARRAY_SIZE(spectre_rlc_save_restore_register_list);
+ } else {
+ rdev->rlc.reg_list = kalindi_rlc_save_restore_register_list;
+ rdev->rlc.reg_list_size =
+ (u32)ARRAY_SIZE(kalindi_rlc_save_restore_register_list);
+ }
+ }
+ rdev->rlc.cs_data = ci_cs_data;
+ rdev->rlc.cp_table_size = CP_ME_TABLE_SIZE * 5 * 4;
r = sumo_rlc_init(rdev);
if (r) {
DRM_ERROR("Failed to init rlc BOs!\n");
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index 4742f3d..63955ab 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -299,6 +299,10 @@
#define VM_CONTEXT0_PAGE_TABLE_END_ADDR 0x157C
#define VM_CONTEXT1_PAGE_TABLE_END_ADDR 0x1580
+#define VM_L2_CG 0x15c0
+#define MC_CG_ENABLE (1 << 18)
+#define MC_LS_ENABLE (1 << 19)
+
#define MC_SHARED_CHMAP 0x2004
#define NOOFCHAN_SHIFT 12
#define NOOFCHAN_MASK 0x0000f000
@@ -328,6 +332,17 @@
#define MC_SHARED_BLACKOUT_CNTL 0x20ac
+#define MC_HUB_MISC_HUB_CG 0x20b8
+#define MC_HUB_MISC_VM_CG 0x20bc
+
+#define MC_HUB_MISC_SIP_CG 0x20c0
+
+#define MC_XPB_CLK_GAT 0x2478
+
+#define MC_CITF_MISC_RD_CG 0x2648
+#define MC_CITF_MISC_WR_CG 0x264c
+#define MC_CITF_MISC_VM_CG 0x2650
+
#define MC_ARB_RAMCFG 0x2760
#define NOOFBANK_SHIFT 0
#define NOOFBANK_MASK 0x00000003
@@ -357,6 +372,7 @@
#define MC_SEQ_IO_DEBUG_DATA 0x2a48
#define HDP_HOST_PATH_CNTL 0x2C00
+#define CLOCK_GATING_DIS (1 << 23)
#define HDP_NONSURFACE_BASE 0x2C04
#define HDP_NONSURFACE_INFO 0x2C08
#define HDP_NONSURFACE_SIZE 0x2C0C
@@ -364,6 +380,10 @@
#define HDP_ADDR_CONFIG 0x2F48
#define HDP_MISC_CNTL 0x2F4C
#define HDP_FLUSH_INVALIDATE_CACHE (1 << 0)
+#define HDP_MEM_POWER_LS 0x2F50
+#define HDP_LS_ENABLE (1 << 0)
+
+#define ATC_MISC_CG 0x3350
#define IH_RB_CNTL 0x3e00
# define IH_RB_ENABLE (1 << 0)
@@ -631,6 +651,9 @@
#define CP_RB0_RPTR 0x8700
#define CP_RB_WPTR_DELAY 0x8704
+#define CP_RB_WPTR_POLL_CNTL 0x8708
+#define IDLE_POLL_COUNT(x) ((x) << 16)
+#define IDLE_POLL_COUNT_MASK (0xffff << 16)
#define CP_MEQ_THRESHOLDS 0x8764
#define MEQ1_START(x) ((x) << 0)
@@ -857,6 +880,9 @@
# define CP_RINGID1_INT_STAT (1 << 30)
# define CP_RINGID0_INT_STAT (1 << 31)
+#define CP_MEM_SLP_CNTL 0xC1E4
+# define CP_MEM_LS_EN (1 << 0)
+
#define CP_CPF_DEBUG 0xC200
#define CP_PQ_WPTR_POLL_CNTL 0xC20C
@@ -902,6 +928,9 @@
#define RLC_MC_CNTL 0xC30C
+#define RLC_MEM_SLP_CNTL 0xC318
+# define RLC_MEM_LS_EN (1 << 0)
+
#define RLC_LB_CNTR_MAX 0xC348
#define RLC_LB_CNTL 0xC364
@@ -910,7 +939,9 @@
#define RLC_LB_CNTR_INIT 0xC36C
#define RLC_SAVE_AND_RESTORE_BASE 0xC374
-#define RLC_DRIVER_DMA_STATUS 0xC378
+#define RLC_DRIVER_DMA_STATUS 0xC378 /* dGPU */
+#define RLC_CP_TABLE_RESTORE 0xC378 /* APU */
+#define RLC_PG_DELAY_2 0xC37C
#define RLC_GPM_UCODE_ADDR 0xC388
#define RLC_GPM_UCODE_DATA 0xC38C
@@ -919,12 +950,50 @@
#define RLC_CAPTURE_GPU_CLOCK_COUNT 0xC398
#define RLC_UCODE_CNTL 0xC39C
+#define RLC_GPM_STAT 0xC400
+# define RLC_GPM_BUSY (1 << 0)
+
+#define RLC_PG_CNTL 0xC40C
+# define GFX_PG_ENABLE (1 << 0)
+# define GFX_PG_SRC (1 << 1)
+# define DYN_PER_CU_PG_ENABLE (1 << 2)
+# define STATIC_PER_CU_PG_ENABLE (1 << 3)
+# define DISABLE_GDS_PG (1 << 13)
+# define DISABLE_CP_PG (1 << 15)
+# define SMU_CLK_SLOWDOWN_ON_PU_ENABLE (1 << 17)
+# define SMU_CLK_SLOWDOWN_ON_PD_ENABLE (1 << 18)
+
+#define RLC_CGTT_MGCG_OVERRIDE 0xC420
#define RLC_CGCG_CGLS_CTRL 0xC424
+# define CGCG_EN (1 << 0)
+# define CGLS_EN (1 << 1)
+
+#define RLC_PG_DELAY 0xC434
#define RLC_LB_INIT_CU_MASK 0xC43C
#define RLC_LB_PARAMS 0xC444
+#define RLC_PG_AO_CU_MASK 0xC44C
+
+#define RLC_MAX_PG_CU 0xC450
+# define MAX_PU_CU(x) ((x) << 0)
+# define MAX_PU_CU_MASK (0xff << 0)
+#define RLC_AUTO_PG_CTRL 0xC454
+# define AUTO_PG_EN (1 << 0)
+# define GRBM_REG_SGIT(x) ((x) << 3)
+# define GRBM_REG_SGIT_MASK (0xffff << 3)
+
+#define RLC_SERDES_WR_CU_MASTER_MASK 0xC474
+#define RLC_SERDES_WR_NONCU_MASTER_MASK 0xC478
+#define RLC_SERDES_WR_CTRL 0xC47C
+#define BPM_ADDR(x) ((x) << 0)
+#define BPM_ADDR_MASK (0xff << 0)
+#define CGLS_ENABLE (1 << 16)
+#define CGCG_OVERRIDE_0 (1 << 20)
+#define MGCG_OVERRIDE_0 (1 << 22)
+#define MGCG_OVERRIDE_1 (1 << 23)
+
#define RLC_SERDES_CU_MASTER_BUSY 0xC484
#define RLC_SERDES_NONCU_MASTER_BUSY 0xC488
# define SE_MASTER_BUSY_MASK 0x0000ffff
@@ -979,6 +1048,8 @@
#define MQD_VMID(x) ((x) << 0)
#define MQD_VMID_MASK (0xf << 0)
+#define DB_RENDER_CONTROL 0x28000
+
#define PA_SC_RASTER_CONFIG 0x28350
# define RASTER_CONFIG_RB_MAP_0 0
# define RASTER_CONFIG_RB_MAP_1 1
@@ -1072,6 +1143,16 @@
#define CP_PERFMON_CNTL 0x36020
+#define CGTS_SM_CTRL_REG 0x3c000
+#define SM_MODE(x) ((x) << 17)
+#define SM_MODE_MASK (0x7 << 17)
+#define SM_MODE_ENABLE (1 << 20)
+#define CGTS_OVERRIDE (1 << 21)
+#define CGTS_LS_OVERRIDE (1 << 22)
+#define ON_MONITOR_ADD_EN (1 << 23)
+#define ON_MONITOR_ADD(x) ((x) << 24)
+#define ON_MONITOR_ADD_MASK (0xff << 24)
+
#define CGTS_TCC_DISABLE 0x3c00c
#define CGTS_USER_TCC_DISABLE 0x3c010
#define TCC_DISABLE_MASK 0xFFFF0000
@@ -1304,6 +1385,8 @@
#define SDMA0_UCODE_ADDR 0xD000
#define SDMA0_UCODE_DATA 0xD004
+#define SDMA0_POWER_CNTL 0xD008
+#define SDMA0_CLK_CTRL 0xD00C
#define SDMA0_CNTL 0xD010
# define TRAP_ENABLE (1 << 0)
@@ -1428,6 +1511,13 @@
#define UVD_RBC_RB_RPTR 0xf690
#define UVD_RBC_RB_WPTR 0xf694
+#define UVD_CGC_CTRL 0xF4B0
+# define DCM (1 << 0)
+# define CG_DT(x) ((x) << 2)
+# define CG_DT_MASK (0xf << 2)
+# define CLK_OD(x) ((x) << 6)
+# define CLK_OD_MASK (0x1f << 6)
+
/* UVD clocks */
#define CG_DCLK_CNTL 0xC050009C
@@ -1438,4 +1528,7 @@
#define CG_VCLK_CNTL 0xC05000A4
#define CG_VCLK_STATUS 0xC05000A8
+/* UVD CTX indirect */
+#define UVD_CGC_MEM_CTRL 0xC0
+
#endif
diff --git a/drivers/gpu/drm/radeon/clearstate_ci.h b/drivers/gpu/drm/radeon/clearstate_ci.h
new file mode 100644
index 0000000..c3982f9
--- /dev/null
+++ b/drivers/gpu/drm/radeon/clearstate_ci.h
@@ -0,0 +1,944 @@
+/*
+ * Copyright 2013 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+static const unsigned int ci_SECT_CONTEXT_def_1[] =
+{
+ 0x00000000, // DB_RENDER_CONTROL
+ 0x00000000, // DB_COUNT_CONTROL
+ 0x00000000, // DB_DEPTH_VIEW
+ 0x00000000, // DB_RENDER_OVERRIDE
+ 0x00000000, // DB_RENDER_OVERRIDE2
+ 0x00000000, // DB_HTILE_DATA_BASE
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // DB_DEPTH_BOUNDS_MIN
+ 0x00000000, // DB_DEPTH_BOUNDS_MAX
+ 0x00000000, // DB_STENCIL_CLEAR
+ 0x00000000, // DB_DEPTH_CLEAR
+ 0x00000000, // PA_SC_SCREEN_SCISSOR_TL
+ 0x40004000, // PA_SC_SCREEN_SCISSOR_BR
+ 0, // HOLE
+ 0x00000000, // DB_DEPTH_INFO
+ 0x00000000, // DB_Z_INFO
+ 0x00000000, // DB_STENCIL_INFO
+ 0x00000000, // DB_Z_READ_BASE
+ 0x00000000, // DB_STENCIL_READ_BASE
+ 0x00000000, // DB_Z_WRITE_BASE
+ 0x00000000, // DB_STENCIL_WRITE_BASE
+ 0x00000000, // DB_DEPTH_SIZE
+ 0x00000000, // DB_DEPTH_SLICE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // TA_BC_BASE_ADDR
+ 0x00000000, // TA_BC_BASE_ADDR_HI
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // COHER_DEST_BASE_HI_0
+ 0x00000000, // COHER_DEST_BASE_HI_1
+ 0x00000000, // COHER_DEST_BASE_HI_2
+ 0x00000000, // COHER_DEST_BASE_HI_3
+ 0x00000000, // COHER_DEST_BASE_2
+ 0x00000000, // COHER_DEST_BASE_3
+ 0x00000000, // PA_SC_WINDOW_OFFSET
+ 0x80000000, // PA_SC_WINDOW_SCISSOR_TL
+ 0x40004000, // PA_SC_WINDOW_SCISSOR_BR
+ 0x0000ffff, // PA_SC_CLIPRECT_RULE
+ 0x00000000, // PA_SC_CLIPRECT_0_TL
+ 0x40004000, // PA_SC_CLIPRECT_0_BR
+ 0x00000000, // PA_SC_CLIPRECT_1_TL
+ 0x40004000, // PA_SC_CLIPRECT_1_BR
+ 0x00000000, // PA_SC_CLIPRECT_2_TL
+ 0x40004000, // PA_SC_CLIPRECT_2_BR
+ 0x00000000, // PA_SC_CLIPRECT_3_TL
+ 0x40004000, // PA_SC_CLIPRECT_3_BR
+ 0xaa99aaaa, // PA_SC_EDGERULE
+ 0x00000000, // PA_SU_HARDWARE_SCREEN_OFFSET
+ 0xffffffff, // CB_TARGET_MASK
+ 0xffffffff, // CB_SHADER_MASK
+ 0x80000000, // PA_SC_GENERIC_SCISSOR_TL
+ 0x40004000, // PA_SC_GENERIC_SCISSOR_BR
+ 0x00000000, // COHER_DEST_BASE_0
+ 0x00000000, // COHER_DEST_BASE_1
+ 0x80000000, // PA_SC_VPORT_SCISSOR_0_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_0_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_1_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_1_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_2_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_2_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_3_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_3_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_4_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_4_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_5_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_5_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_6_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_6_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_7_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_7_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_8_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_8_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_9_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_9_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_10_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_10_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_11_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_11_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_12_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_12_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_13_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_13_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_14_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_14_BR
+ 0x80000000, // PA_SC_VPORT_SCISSOR_15_TL
+ 0x40004000, // PA_SC_VPORT_SCISSOR_15_BR
+ 0x00000000, // PA_SC_VPORT_ZMIN_0
+ 0x3f800000, // PA_SC_VPORT_ZMAX_0
+ 0x00000000, // PA_SC_VPORT_ZMIN_1
+ 0x3f800000, // PA_SC_VPORT_ZMAX_1
+ 0x00000000, // PA_SC_VPORT_ZMIN_2
+ 0x3f800000, // PA_SC_VPORT_ZMAX_2
+ 0x00000000, // PA_SC_VPORT_ZMIN_3
+ 0x3f800000, // PA_SC_VPORT_ZMAX_3
+ 0x00000000, // PA_SC_VPORT_ZMIN_4
+ 0x3f800000, // PA_SC_VPORT_ZMAX_4
+ 0x00000000, // PA_SC_VPORT_ZMIN_5
+ 0x3f800000, // PA_SC_VPORT_ZMAX_5
+ 0x00000000, // PA_SC_VPORT_ZMIN_6
+ 0x3f800000, // PA_SC_VPORT_ZMAX_6
+ 0x00000000, // PA_SC_VPORT_ZMIN_7
+ 0x3f800000, // PA_SC_VPORT_ZMAX_7
+ 0x00000000, // PA_SC_VPORT_ZMIN_8
+ 0x3f800000, // PA_SC_VPORT_ZMAX_8
+ 0x00000000, // PA_SC_VPORT_ZMIN_9
+ 0x3f800000, // PA_SC_VPORT_ZMAX_9
+ 0x00000000, // PA_SC_VPORT_ZMIN_10
+ 0x3f800000, // PA_SC_VPORT_ZMAX_10
+ 0x00000000, // PA_SC_VPORT_ZMIN_11
+ 0x3f800000, // PA_SC_VPORT_ZMAX_11
+ 0x00000000, // PA_SC_VPORT_ZMIN_12
+ 0x3f800000, // PA_SC_VPORT_ZMAX_12
+ 0x00000000, // PA_SC_VPORT_ZMIN_13
+ 0x3f800000, // PA_SC_VPORT_ZMAX_13
+ 0x00000000, // PA_SC_VPORT_ZMIN_14
+ 0x3f800000, // PA_SC_VPORT_ZMAX_14
+ 0x00000000, // PA_SC_VPORT_ZMIN_15
+ 0x3f800000, // PA_SC_VPORT_ZMAX_15
+};
+static const unsigned int ci_SECT_CONTEXT_def_2[] =
+{
+ 0x00000000, // PA_SC_SCREEN_EXTENT_CONTROL
+ 0, // HOLE
+ 0x00000000, // CP_PERFMON_CNTX_CNTL
+ 0x00000000, // CP_RINGID
+ 0x00000000, // CP_VMID
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0xffffffff, // VGT_MAX_VTX_INDX
+ 0x00000000, // VGT_MIN_VTX_INDX
+ 0x00000000, // VGT_INDX_OFFSET
+ 0x00000000, // VGT_MULTI_PRIM_IB_RESET_INDX
+ 0, // HOLE
+ 0x00000000, // CB_BLEND_RED
+ 0x00000000, // CB_BLEND_GREEN
+ 0x00000000, // CB_BLEND_BLUE
+ 0x00000000, // CB_BLEND_ALPHA
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // DB_STENCIL_CONTROL
+ 0x00000000, // DB_STENCILREFMASK
+ 0x00000000, // DB_STENCILREFMASK_BF
+ 0, // HOLE
+ 0x00000000, // PA_CL_VPORT_XSCALE
+ 0x00000000, // PA_CL_VPORT_XOFFSET
+ 0x00000000, // PA_CL_VPORT_YSCALE
+ 0x00000000, // PA_CL_VPORT_YOFFSET
+ 0x00000000, // PA_CL_VPORT_ZSCALE
+ 0x00000000, // PA_CL_VPORT_ZOFFSET
+ 0x00000000, // PA_CL_VPORT_XSCALE_1
+ 0x00000000, // PA_CL_VPORT_XOFFSET_1
+ 0x00000000, // PA_CL_VPORT_YSCALE_1
+ 0x00000000, // PA_CL_VPORT_YOFFSET_1
+ 0x00000000, // PA_CL_VPORT_ZSCALE_1
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_1
+ 0x00000000, // PA_CL_VPORT_XSCALE_2
+ 0x00000000, // PA_CL_VPORT_XOFFSET_2
+ 0x00000000, // PA_CL_VPORT_YSCALE_2
+ 0x00000000, // PA_CL_VPORT_YOFFSET_2
+ 0x00000000, // PA_CL_VPORT_ZSCALE_2
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_2
+ 0x00000000, // PA_CL_VPORT_XSCALE_3
+ 0x00000000, // PA_CL_VPORT_XOFFSET_3
+ 0x00000000, // PA_CL_VPORT_YSCALE_3
+ 0x00000000, // PA_CL_VPORT_YOFFSET_3
+ 0x00000000, // PA_CL_VPORT_ZSCALE_3
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_3
+ 0x00000000, // PA_CL_VPORT_XSCALE_4
+ 0x00000000, // PA_CL_VPORT_XOFFSET_4
+ 0x00000000, // PA_CL_VPORT_YSCALE_4
+ 0x00000000, // PA_CL_VPORT_YOFFSET_4
+ 0x00000000, // PA_CL_VPORT_ZSCALE_4
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_4
+ 0x00000000, // PA_CL_VPORT_XSCALE_5
+ 0x00000000, // PA_CL_VPORT_XOFFSET_5
+ 0x00000000, // PA_CL_VPORT_YSCALE_5
+ 0x00000000, // PA_CL_VPORT_YOFFSET_5
+ 0x00000000, // PA_CL_VPORT_ZSCALE_5
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_5
+ 0x00000000, // PA_CL_VPORT_XSCALE_6
+ 0x00000000, // PA_CL_VPORT_XOFFSET_6
+ 0x00000000, // PA_CL_VPORT_YSCALE_6
+ 0x00000000, // PA_CL_VPORT_YOFFSET_6
+ 0x00000000, // PA_CL_VPORT_ZSCALE_6
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_6
+ 0x00000000, // PA_CL_VPORT_XSCALE_7
+ 0x00000000, // PA_CL_VPORT_XOFFSET_7
+ 0x00000000, // PA_CL_VPORT_YSCALE_7
+ 0x00000000, // PA_CL_VPORT_YOFFSET_7
+ 0x00000000, // PA_CL_VPORT_ZSCALE_7
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_7
+ 0x00000000, // PA_CL_VPORT_XSCALE_8
+ 0x00000000, // PA_CL_VPORT_XOFFSET_8
+ 0x00000000, // PA_CL_VPORT_YSCALE_8
+ 0x00000000, // PA_CL_VPORT_YOFFSET_8
+ 0x00000000, // PA_CL_VPORT_ZSCALE_8
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_8
+ 0x00000000, // PA_CL_VPORT_XSCALE_9
+ 0x00000000, // PA_CL_VPORT_XOFFSET_9
+ 0x00000000, // PA_CL_VPORT_YSCALE_9
+ 0x00000000, // PA_CL_VPORT_YOFFSET_9
+ 0x00000000, // PA_CL_VPORT_ZSCALE_9
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_9
+ 0x00000000, // PA_CL_VPORT_XSCALE_10
+ 0x00000000, // PA_CL_VPORT_XOFFSET_10
+ 0x00000000, // PA_CL_VPORT_YSCALE_10
+ 0x00000000, // PA_CL_VPORT_YOFFSET_10
+ 0x00000000, // PA_CL_VPORT_ZSCALE_10
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_10
+ 0x00000000, // PA_CL_VPORT_XSCALE_11
+ 0x00000000, // PA_CL_VPORT_XOFFSET_11
+ 0x00000000, // PA_CL_VPORT_YSCALE_11
+ 0x00000000, // PA_CL_VPORT_YOFFSET_11
+ 0x00000000, // PA_CL_VPORT_ZSCALE_11
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_11
+ 0x00000000, // PA_CL_VPORT_XSCALE_12
+ 0x00000000, // PA_CL_VPORT_XOFFSET_12
+ 0x00000000, // PA_CL_VPORT_YSCALE_12
+ 0x00000000, // PA_CL_VPORT_YOFFSET_12
+ 0x00000000, // PA_CL_VPORT_ZSCALE_12
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_12
+ 0x00000000, // PA_CL_VPORT_XSCALE_13
+ 0x00000000, // PA_CL_VPORT_XOFFSET_13
+ 0x00000000, // PA_CL_VPORT_YSCALE_13
+ 0x00000000, // PA_CL_VPORT_YOFFSET_13
+ 0x00000000, // PA_CL_VPORT_ZSCALE_13
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_13
+ 0x00000000, // PA_CL_VPORT_XSCALE_14
+ 0x00000000, // PA_CL_VPORT_XOFFSET_14
+ 0x00000000, // PA_CL_VPORT_YSCALE_14
+ 0x00000000, // PA_CL_VPORT_YOFFSET_14
+ 0x00000000, // PA_CL_VPORT_ZSCALE_14
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_14
+ 0x00000000, // PA_CL_VPORT_XSCALE_15
+ 0x00000000, // PA_CL_VPORT_XOFFSET_15
+ 0x00000000, // PA_CL_VPORT_YSCALE_15
+ 0x00000000, // PA_CL_VPORT_YOFFSET_15
+ 0x00000000, // PA_CL_VPORT_ZSCALE_15
+ 0x00000000, // PA_CL_VPORT_ZOFFSET_15
+ 0x00000000, // PA_CL_UCP_0_X
+ 0x00000000, // PA_CL_UCP_0_Y
+ 0x00000000, // PA_CL_UCP_0_Z
+ 0x00000000, // PA_CL_UCP_0_W
+ 0x00000000, // PA_CL_UCP_1_X
+ 0x00000000, // PA_CL_UCP_1_Y
+ 0x00000000, // PA_CL_UCP_1_Z
+ 0x00000000, // PA_CL_UCP_1_W
+ 0x00000000, // PA_CL_UCP_2_X
+ 0x00000000, // PA_CL_UCP_2_Y
+ 0x00000000, // PA_CL_UCP_2_Z
+ 0x00000000, // PA_CL_UCP_2_W
+ 0x00000000, // PA_CL_UCP_3_X
+ 0x00000000, // PA_CL_UCP_3_Y
+ 0x00000000, // PA_CL_UCP_3_Z
+ 0x00000000, // PA_CL_UCP_3_W
+ 0x00000000, // PA_CL_UCP_4_X
+ 0x00000000, // PA_CL_UCP_4_Y
+ 0x00000000, // PA_CL_UCP_4_Z
+ 0x00000000, // PA_CL_UCP_4_W
+ 0x00000000, // PA_CL_UCP_5_X
+ 0x00000000, // PA_CL_UCP_5_Y
+ 0x00000000, // PA_CL_UCP_5_Z
+ 0x00000000, // PA_CL_UCP_5_W
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // SPI_PS_INPUT_CNTL_0
+ 0x00000000, // SPI_PS_INPUT_CNTL_1
+ 0x00000000, // SPI_PS_INPUT_CNTL_2
+ 0x00000000, // SPI_PS_INPUT_CNTL_3
+ 0x00000000, // SPI_PS_INPUT_CNTL_4
+ 0x00000000, // SPI_PS_INPUT_CNTL_5
+ 0x00000000, // SPI_PS_INPUT_CNTL_6
+ 0x00000000, // SPI_PS_INPUT_CNTL_7
+ 0x00000000, // SPI_PS_INPUT_CNTL_8
+ 0x00000000, // SPI_PS_INPUT_CNTL_9
+ 0x00000000, // SPI_PS_INPUT_CNTL_10
+ 0x00000000, // SPI_PS_INPUT_CNTL_11
+ 0x00000000, // SPI_PS_INPUT_CNTL_12
+ 0x00000000, // SPI_PS_INPUT_CNTL_13
+ 0x00000000, // SPI_PS_INPUT_CNTL_14
+ 0x00000000, // SPI_PS_INPUT_CNTL_15
+ 0x00000000, // SPI_PS_INPUT_CNTL_16
+ 0x00000000, // SPI_PS_INPUT_CNTL_17
+ 0x00000000, // SPI_PS_INPUT_CNTL_18
+ 0x00000000, // SPI_PS_INPUT_CNTL_19
+ 0x00000000, // SPI_PS_INPUT_CNTL_20
+ 0x00000000, // SPI_PS_INPUT_CNTL_21
+ 0x00000000, // SPI_PS_INPUT_CNTL_22
+ 0x00000000, // SPI_PS_INPUT_CNTL_23
+ 0x00000000, // SPI_PS_INPUT_CNTL_24
+ 0x00000000, // SPI_PS_INPUT_CNTL_25
+ 0x00000000, // SPI_PS_INPUT_CNTL_26
+ 0x00000000, // SPI_PS_INPUT_CNTL_27
+ 0x00000000, // SPI_PS_INPUT_CNTL_28
+ 0x00000000, // SPI_PS_INPUT_CNTL_29
+ 0x00000000, // SPI_PS_INPUT_CNTL_30
+ 0x00000000, // SPI_PS_INPUT_CNTL_31
+ 0x00000000, // SPI_VS_OUT_CONFIG
+ 0, // HOLE
+ 0x00000000, // SPI_PS_INPUT_ENA
+ 0x00000000, // SPI_PS_INPUT_ADDR
+ 0x00000000, // SPI_INTERP_CONTROL_0
+ 0x00000002, // SPI_PS_IN_CONTROL
+ 0, // HOLE
+ 0x00000000, // SPI_BARYC_CNTL
+ 0, // HOLE
+ 0x00000000, // SPI_TMPRING_SIZE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // SPI_SHADER_POS_FORMAT
+ 0x00000000, // SPI_SHADER_Z_FORMAT
+ 0x00000000, // SPI_SHADER_COL_FORMAT
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // CB_BLEND0_CONTROL
+ 0x00000000, // CB_BLEND1_CONTROL
+ 0x00000000, // CB_BLEND2_CONTROL
+ 0x00000000, // CB_BLEND3_CONTROL
+ 0x00000000, // CB_BLEND4_CONTROL
+ 0x00000000, // CB_BLEND5_CONTROL
+ 0x00000000, // CB_BLEND6_CONTROL
+ 0x00000000, // CB_BLEND7_CONTROL
+};
+static const unsigned int ci_SECT_CONTEXT_def_3[] =
+{
+ 0x00000000, // PA_CL_POINT_X_RAD
+ 0x00000000, // PA_CL_POINT_Y_RAD
+ 0x00000000, // PA_CL_POINT_SIZE
+ 0x00000000, // PA_CL_POINT_CULL_RAD
+ 0x00000000, // VGT_DMA_BASE_HI
+ 0x00000000, // VGT_DMA_BASE
+};
+static const unsigned int ci_SECT_CONTEXT_def_4[] =
+{
+ 0x00000000, // DB_DEPTH_CONTROL
+ 0x00000000, // DB_EQAA
+ 0x00000000, // CB_COLOR_CONTROL
+ 0x00000000, // DB_SHADER_CONTROL
+ 0x00090000, // PA_CL_CLIP_CNTL
+ 0x00000004, // PA_SU_SC_MODE_CNTL
+ 0x00000000, // PA_CL_VTE_CNTL
+ 0x00000000, // PA_CL_VS_OUT_CNTL
+ 0x00000000, // PA_CL_NANINF_CNTL
+ 0x00000000, // PA_SU_LINE_STIPPLE_CNTL
+ 0x00000000, // PA_SU_LINE_STIPPLE_SCALE
+ 0x00000000, // PA_SU_PRIM_FILTER_CNTL
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // PA_SU_POINT_SIZE
+ 0x00000000, // PA_SU_POINT_MINMAX
+ 0x00000000, // PA_SU_LINE_CNTL
+ 0x00000000, // PA_SC_LINE_STIPPLE
+ 0x00000000, // VGT_OUTPUT_PATH_CNTL
+ 0x00000000, // VGT_HOS_CNTL
+ 0x00000000, // VGT_HOS_MAX_TESS_LEVEL
+ 0x00000000, // VGT_HOS_MIN_TESS_LEVEL
+ 0x00000000, // VGT_HOS_REUSE_DEPTH
+ 0x00000000, // VGT_GROUP_PRIM_TYPE
+ 0x00000000, // VGT_GROUP_FIRST_DECR
+ 0x00000000, // VGT_GROUP_DECR
+ 0x00000000, // VGT_GROUP_VECT_0_CNTL
+ 0x00000000, // VGT_GROUP_VECT_1_CNTL
+ 0x00000000, // VGT_GROUP_VECT_0_FMT_CNTL
+ 0x00000000, // VGT_GROUP_VECT_1_FMT_CNTL
+ 0x00000000, // VGT_GS_MODE
+ 0x00000000, // VGT_GS_ONCHIP_CNTL
+ 0x00000000, // PA_SC_MODE_CNTL_0
+ 0x00000000, // PA_SC_MODE_CNTL_1
+ 0x00000000, // VGT_ENHANCE
+ 0x00000100, // VGT_GS_PER_ES
+ 0x00000080, // VGT_ES_PER_GS
+ 0x00000002, // VGT_GS_PER_VS
+ 0x00000000, // VGT_GSVS_RING_OFFSET_1
+ 0x00000000, // VGT_GSVS_RING_OFFSET_2
+ 0x00000000, // VGT_GSVS_RING_OFFSET_3
+ 0x00000000, // VGT_GS_OUT_PRIM_TYPE
+ 0x00000000, // IA_ENHANCE
+};
+static const unsigned int ci_SECT_CONTEXT_def_5[] =
+{
+ 0x00000000, // WD_ENHANCE
+ 0x00000000, // VGT_PRIMITIVEID_EN
+};
+static const unsigned int ci_SECT_CONTEXT_def_6[] =
+{
+ 0x00000000, // VGT_PRIMITIVEID_RESET
+};
+static const unsigned int ci_SECT_CONTEXT_def_7[] =
+{
+ 0x00000000, // VGT_MULTI_PRIM_IB_RESET_EN
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // VGT_INSTANCE_STEP_RATE_0
+ 0x00000000, // VGT_INSTANCE_STEP_RATE_1
+ 0x000000ff, // IA_MULTI_VGT_PARAM
+ 0x00000000, // VGT_ESGS_RING_ITEMSIZE
+ 0x00000000, // VGT_GSVS_RING_ITEMSIZE
+ 0x00000000, // VGT_REUSE_OFF
+ 0x00000000, // VGT_VTX_CNT_EN
+ 0x00000000, // DB_HTILE_SURFACE
+ 0x00000000, // DB_SRESULTS_COMPARE_STATE0
+ 0x00000000, // DB_SRESULTS_COMPARE_STATE1
+ 0x00000000, // DB_PRELOAD_CONTROL
+ 0, // HOLE
+ 0x00000000, // VGT_STRMOUT_BUFFER_SIZE_0
+ 0x00000000, // VGT_STRMOUT_VTX_STRIDE_0
+ 0, // HOLE
+ 0x00000000, // VGT_STRMOUT_BUFFER_OFFSET_0
+ 0x00000000, // VGT_STRMOUT_BUFFER_SIZE_1
+ 0x00000000, // VGT_STRMOUT_VTX_STRIDE_1
+ 0, // HOLE
+ 0x00000000, // VGT_STRMOUT_BUFFER_OFFSET_1
+ 0x00000000, // VGT_STRMOUT_BUFFER_SIZE_2
+ 0x00000000, // VGT_STRMOUT_VTX_STRIDE_2
+ 0, // HOLE
+ 0x00000000, // VGT_STRMOUT_BUFFER_OFFSET_2
+ 0x00000000, // VGT_STRMOUT_BUFFER_SIZE_3
+ 0x00000000, // VGT_STRMOUT_VTX_STRIDE_3
+ 0, // HOLE
+ 0x00000000, // VGT_STRMOUT_BUFFER_OFFSET_3
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // VGT_STRMOUT_DRAW_OPAQUE_OFFSET
+ 0x00000000, // VGT_STRMOUT_DRAW_OPAQUE_BUFFER_FILLED_SIZE
+ 0x00000000, // VGT_STRMOUT_DRAW_OPAQUE_VERTEX_STRIDE
+ 0, // HOLE
+ 0x00000000, // VGT_GS_MAX_VERT_OUT
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // VGT_SHADER_STAGES_EN
+ 0x00000000, // VGT_LS_HS_CONFIG
+ 0x00000000, // VGT_GS_VERT_ITEMSIZE
+ 0x00000000, // VGT_GS_VERT_ITEMSIZE_1
+ 0x00000000, // VGT_GS_VERT_ITEMSIZE_2
+ 0x00000000, // VGT_GS_VERT_ITEMSIZE_3
+ 0x00000000, // VGT_TF_PARAM
+ 0x00000000, // DB_ALPHA_TO_MASK
+ 0, // HOLE
+ 0x00000000, // PA_SU_POLY_OFFSET_DB_FMT_CNTL
+ 0x00000000, // PA_SU_POLY_OFFSET_CLAMP
+ 0x00000000, // PA_SU_POLY_OFFSET_FRONT_SCALE
+ 0x00000000, // PA_SU_POLY_OFFSET_FRONT_OFFSET
+ 0x00000000, // PA_SU_POLY_OFFSET_BACK_SCALE
+ 0x00000000, // PA_SU_POLY_OFFSET_BACK_OFFSET
+ 0x00000000, // VGT_GS_INSTANCE_CNT
+ 0x00000000, // VGT_STRMOUT_CONFIG
+ 0x00000000, // VGT_STRMOUT_BUFFER_CONFIG
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // PA_SC_CENTROID_PRIORITY_0
+ 0x00000000, // PA_SC_CENTROID_PRIORITY_1
+ 0x00001000, // PA_SC_LINE_CNTL
+ 0x00000000, // PA_SC_AA_CONFIG
+ 0x00000005, // PA_SU_VTX_CNTL
+ 0x3f800000, // PA_CL_GB_VERT_CLIP_ADJ
+ 0x3f800000, // PA_CL_GB_VERT_DISC_ADJ
+ 0x3f800000, // PA_CL_GB_HORZ_CLIP_ADJ
+ 0x3f800000, // PA_CL_GB_HORZ_DISC_ADJ
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2
+ 0x00000000, // PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3
+ 0xffffffff, // PA_SC_AA_MASK_X0Y0_X1Y0
+ 0xffffffff, // PA_SC_AA_MASK_X0Y1_X1Y1
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0, // HOLE
+ 0x0000000e, // VGT_VERTEX_REUSE_BLOCK_CNTL
+ 0x00000010, // VGT_OUT_DEALLOC_CNTL
+ 0x00000000, // CB_COLOR0_BASE
+ 0x00000000, // CB_COLOR0_PITCH
+ 0x00000000, // CB_COLOR0_SLICE
+ 0x00000000, // CB_COLOR0_VIEW
+ 0x00000000, // CB_COLOR0_INFO
+ 0x00000000, // CB_COLOR0_ATTRIB
+ 0, // HOLE
+ 0x00000000, // CB_COLOR0_CMASK
+ 0x00000000, // CB_COLOR0_CMASK_SLICE
+ 0x00000000, // CB_COLOR0_FMASK
+ 0x00000000, // CB_COLOR0_FMASK_SLICE
+ 0x00000000, // CB_COLOR0_CLEAR_WORD0
+ 0x00000000, // CB_COLOR0_CLEAR_WORD1
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // CB_COLOR1_BASE
+ 0x00000000, // CB_COLOR1_PITCH
+ 0x00000000, // CB_COLOR1_SLICE
+ 0x00000000, // CB_COLOR1_VIEW
+ 0x00000000, // CB_COLOR1_INFO
+ 0x00000000, // CB_COLOR1_ATTRIB
+ 0, // HOLE
+ 0x00000000, // CB_COLOR1_CMASK
+ 0x00000000, // CB_COLOR1_CMASK_SLICE
+ 0x00000000, // CB_COLOR1_FMASK
+ 0x00000000, // CB_COLOR1_FMASK_SLICE
+ 0x00000000, // CB_COLOR1_CLEAR_WORD0
+ 0x00000000, // CB_COLOR1_CLEAR_WORD1
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // CB_COLOR2_BASE
+ 0x00000000, // CB_COLOR2_PITCH
+ 0x00000000, // CB_COLOR2_SLICE
+ 0x00000000, // CB_COLOR2_VIEW
+ 0x00000000, // CB_COLOR2_INFO
+ 0x00000000, // CB_COLOR2_ATTRIB
+ 0, // HOLE
+ 0x00000000, // CB_COLOR2_CMASK
+ 0x00000000, // CB_COLOR2_CMASK_SLICE
+ 0x00000000, // CB_COLOR2_FMASK
+ 0x00000000, // CB_COLOR2_FMASK_SLICE
+ 0x00000000, // CB_COLOR2_CLEAR_WORD0
+ 0x00000000, // CB_COLOR2_CLEAR_WORD1
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // CB_COLOR3_BASE
+ 0x00000000, // CB_COLOR3_PITCH
+ 0x00000000, // CB_COLOR3_SLICE
+ 0x00000000, // CB_COLOR3_VIEW
+ 0x00000000, // CB_COLOR3_INFO
+ 0x00000000, // CB_COLOR3_ATTRIB
+ 0, // HOLE
+ 0x00000000, // CB_COLOR3_CMASK
+ 0x00000000, // CB_COLOR3_CMASK_SLICE
+ 0x00000000, // CB_COLOR3_FMASK
+ 0x00000000, // CB_COLOR3_FMASK_SLICE
+ 0x00000000, // CB_COLOR3_CLEAR_WORD0
+ 0x00000000, // CB_COLOR3_CLEAR_WORD1
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // CB_COLOR4_BASE
+ 0x00000000, // CB_COLOR4_PITCH
+ 0x00000000, // CB_COLOR4_SLICE
+ 0x00000000, // CB_COLOR4_VIEW
+ 0x00000000, // CB_COLOR4_INFO
+ 0x00000000, // CB_COLOR4_ATTRIB
+ 0, // HOLE
+ 0x00000000, // CB_COLOR4_CMASK
+ 0x00000000, // CB_COLOR4_CMASK_SLICE
+ 0x00000000, // CB_COLOR4_FMASK
+ 0x00000000, // CB_COLOR4_FMASK_SLICE
+ 0x00000000, // CB_COLOR4_CLEAR_WORD0
+ 0x00000000, // CB_COLOR4_CLEAR_WORD1
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // CB_COLOR5_BASE
+ 0x00000000, // CB_COLOR5_PITCH
+ 0x00000000, // CB_COLOR5_SLICE
+ 0x00000000, // CB_COLOR5_VIEW
+ 0x00000000, // CB_COLOR5_INFO
+ 0x00000000, // CB_COLOR5_ATTRIB
+ 0, // HOLE
+ 0x00000000, // CB_COLOR5_CMASK
+ 0x00000000, // CB_COLOR5_CMASK_SLICE
+ 0x00000000, // CB_COLOR5_FMASK
+ 0x00000000, // CB_COLOR5_FMASK_SLICE
+ 0x00000000, // CB_COLOR5_CLEAR_WORD0
+ 0x00000000, // CB_COLOR5_CLEAR_WORD1
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // CB_COLOR6_BASE
+ 0x00000000, // CB_COLOR6_PITCH
+ 0x00000000, // CB_COLOR6_SLICE
+ 0x00000000, // CB_COLOR6_VIEW
+ 0x00000000, // CB_COLOR6_INFO
+ 0x00000000, // CB_COLOR6_ATTRIB
+ 0, // HOLE
+ 0x00000000, // CB_COLOR6_CMASK
+ 0x00000000, // CB_COLOR6_CMASK_SLICE
+ 0x00000000, // CB_COLOR6_FMASK
+ 0x00000000, // CB_COLOR6_FMASK_SLICE
+ 0x00000000, // CB_COLOR6_CLEAR_WORD0
+ 0x00000000, // CB_COLOR6_CLEAR_WORD1
+ 0, // HOLE
+ 0, // HOLE
+ 0x00000000, // CB_COLOR7_BASE
+ 0x00000000, // CB_COLOR7_PITCH
+ 0x00000000, // CB_COLOR7_SLICE
+ 0x00000000, // CB_COLOR7_VIEW
+ 0x00000000, // CB_COLOR7_INFO
+ 0x00000000, // CB_COLOR7_ATTRIB
+ 0, // HOLE
+ 0x00000000, // CB_COLOR7_CMASK
+ 0x00000000, // CB_COLOR7_CMASK_SLICE
+ 0x00000000, // CB_COLOR7_FMASK
+ 0x00000000, // CB_COLOR7_FMASK_SLICE
+ 0x00000000, // CB_COLOR7_CLEAR_WORD0
+ 0x00000000, // CB_COLOR7_CLEAR_WORD1
+};
+static const struct cs_extent_def ci_SECT_CONTEXT_defs[] =
+{
+ {ci_SECT_CONTEXT_def_1, 0x0000a000, 212 },
+ {ci_SECT_CONTEXT_def_2, 0x0000a0d6, 274 },
+ {ci_SECT_CONTEXT_def_3, 0x0000a1f5, 6 },
+ {ci_SECT_CONTEXT_def_4, 0x0000a200, 157 },
+ {ci_SECT_CONTEXT_def_5, 0x0000a2a0, 2 },
+ {ci_SECT_CONTEXT_def_6, 0x0000a2a3, 1 },
+ {ci_SECT_CONTEXT_def_7, 0x0000a2a5, 233 },
+ { 0, 0, 0 }
+};
+static const struct cs_section_def ci_cs_data[] = {
+ { ci_SECT_CONTEXT_defs, SECT_CONTEXT },
+ { 0, SECT_NONE }
+};
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 6fc876a..2ce12ee 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -140,6 +140,7 @@ extern void cayman_cp_int_cntl_setup(struct radeon_device *rdev,
int ring, u32 cp_int_cntl);
extern void cayman_vm_decode_fault(struct radeon_device *rdev,
u32 status, u32 addr);
+void cik_init_cp_pg_table(struct radeon_device *rdev);
static const u32 evergreen_golden_registers[] =
{
@@ -3893,8 +3894,22 @@ void sumo_rlc_fini(struct radeon_device *rdev)
radeon_bo_unref(&rdev->rlc.clear_state_obj);
rdev->rlc.clear_state_obj = NULL;
}
+
+ /* clear state block */
+ if (rdev->rlc.cp_table_obj) {
+ r = radeon_bo_reserve(rdev->rlc.cp_table_obj, false);
+ if (unlikely(r != 0))
+ dev_warn(rdev->dev, "(%d) reserve RLC cp table bo failed\n", r);
+ radeon_bo_unpin(rdev->rlc.cp_table_obj);
+ radeon_bo_unreserve(rdev->rlc.cp_table_obj);
+
+ radeon_bo_unref(&rdev->rlc.cp_table_obj);
+ rdev->rlc.cp_table_obj = NULL;
+ }
}
+#define CP_ME_TABLE_SIZE 96
+
int sumo_rlc_init(struct radeon_device *rdev)
{
const u32 *src_ptr;
@@ -3980,9 +3995,10 @@ int sumo_rlc_init(struct radeon_device *rdev)
}
reg_list_blk_index = (3 * reg_list_num + 2);
dws += reg_list_blk_index;
+ rdev->rlc.clear_state_size = dws;
if (rdev->rlc.clear_state_obj == NULL) {
- r = radeon_bo_create(rdev, dws * 4, PAGE_SIZE, true,
+ r = radeon_bo_create(rdev, rdev->rlc.clear_state_size * 4, PAGE_SIZE, true,
RADEON_GEM_DOMAIN_VRAM, NULL, &rdev->rlc.clear_state_obj);
if (r) {
dev_warn(rdev->dev, "(%d) create RLC c bo failed\n", r);
@@ -4046,6 +4062,45 @@ int sumo_rlc_init(struct radeon_device *rdev)
radeon_bo_unreserve(rdev->rlc.clear_state_obj);
}
+ if (rdev->rlc.cp_table_size) {
+ if (rdev->rlc.cp_table_obj == NULL) {
+ r = radeon_bo_create(rdev, rdev->rlc.cp_table_size, PAGE_SIZE, true,
+ RADEON_GEM_DOMAIN_VRAM, NULL, &rdev->rlc.cp_table_obj);
+ if (r) {
+ dev_warn(rdev->dev, "(%d) create RLC cp table bo failed\n", r);
+ sumo_rlc_fini(rdev);
+ return r;
+ }
+ }
+
+ r = radeon_bo_reserve(rdev->rlc.cp_table_obj, false);
+ if (unlikely(r != 0)) {
+ dev_warn(rdev->dev, "(%d) reserve RLC cp table bo failed\n", r);
+ sumo_rlc_fini(rdev);
+ return r;
+ }
+ r = radeon_bo_pin(rdev->rlc.cp_table_obj, RADEON_GEM_DOMAIN_VRAM,
+ &rdev->rlc.cp_table_gpu_addr);
+ if (r) {
+ radeon_bo_unreserve(rdev->rlc.cp_table_obj);
+ dev_warn(rdev->dev, "(%d) pin RLC cp_table bo failed\n", r);
+ sumo_rlc_fini(rdev);
+ return r;
+ }
+ r = radeon_bo_kmap(rdev->rlc.cp_table_obj, (void **)&rdev->rlc.cp_table_ptr);
+ if (r) {
+ dev_warn(rdev->dev, "(%d) map RLC cp table bo failed\n", r);
+ sumo_rlc_fini(rdev);
+ return r;
+ }
+
+ cik_init_cp_pg_table(rdev);
+
+ radeon_bo_kunmap(rdev->rlc.cp_table_obj);
+ radeon_bo_unreserve(rdev->rlc.cp_table_obj);
+
+ }
+
return 0;
}
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 0fdce76..ed241d6 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -152,6 +152,14 @@ extern int radeon_aspm;
#define RADEON_RESET_MC (1 << 10)
#define RADEON_RESET_DISPLAY (1 << 11)
+/* CG block flags */
+#define RADEON_CG_BLOCK_GFX (1 << 0)
+#define RADEON_CG_BLOCK_MC (1 << 1)
+#define RADEON_CG_BLOCK_SDMA (1 << 2)
+#define RADEON_CG_BLOCK_UVD (1 << 3)
+#define RADEON_CG_BLOCK_VCE (1 << 4)
+#define RADEON_CG_BLOCK_HDP (1 << 5)
+
/* max cursor sizes (in pixels) */
#define CURSOR_WIDTH 64
#define CURSOR_HEIGHT 64
@@ -861,6 +869,12 @@ struct radeon_rlc {
uint64_t clear_state_gpu_addr;
volatile uint32_t *cs_ptr;
const struct cs_section_def *cs_data;
+ u32 clear_state_size;
+ /* for cp tables */
+ struct radeon_bo *cp_table_obj;
+ uint64_t cp_table_gpu_addr;
+ volatile uint32_t *cp_table_ptr;
+ u32 cp_table_size;
};
int radeon_ib_get(struct radeon_device *rdev, int ring,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 1926ec0..880551b 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2780,6 +2780,7 @@ int radeon_asic_init(struct radeon_device *rdev)
case CHIP_BONAIRE:
rdev->asic = &ci_asic;
rdev->num_crtc = 6;
+ rdev->has_uvd = true;
break;
case CHIP_KAVERI:
case CHIP_KABINI:
@@ -2789,6 +2790,7 @@ int radeon_asic_init(struct radeon_device *rdev)
rdev->num_crtc = 4;
else
rdev->num_crtc = 2;
+ rdev->has_uvd = true;
break;
default:
/* FIXME: not supported yet */
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index 8b8963d..4f91e1f 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -4901,7 +4901,7 @@ static void si_set_uvd_dcm(struct radeon_device *rdev,
WREG32_UVD_CTX(UVD_CGC_CTRL2, tmp2);
}
-static void si_init_uvd_internal_cg(struct radeon_device *rdev)
+void si_init_uvd_internal_cg(struct radeon_device *rdev)
{
bool hw_mode = true;
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 17/53] drm/radeon: add indirect accessors for dift registers on CIK
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (15 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 16/53] drm/radeon: implement clock and power gating for CIK (v3) Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 18/53] drm/radeon/sumo add helper to go from vid7 to vid2 Alex Deucher
` (36 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cik_reg.h | 3 +++
drivers/gpu/drm/radeon/radeon.h | 18 ++++++++++++++++++
2 files changed, 21 insertions(+)
diff --git a/drivers/gpu/drm/radeon/cik_reg.h b/drivers/gpu/drm/radeon/cik_reg.h
index d71e46d..ca1bb61 100644
--- a/drivers/gpu/drm/radeon/cik_reg.h
+++ b/drivers/gpu/drm/radeon/cik_reg.h
@@ -24,6 +24,9 @@
#ifndef __CIK_REG_H__
#define __CIK_REG_H__
+#define CIK_DIDT_IND_INDEX 0xca00
+#define CIK_DIDT_IND_DATA 0xca04
+
#define CIK_DC_GPIO_HPD_MASK 0x65b0
#define CIK_DC_GPIO_HPD_A 0x65b4
#define CIK_DC_GPIO_HPD_EN 0x65b8
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index ed241d6..37b0fc4 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -2141,6 +2141,8 @@ void cik_mm_wdoorbell(struct radeon_device *rdev, u32 offset, u32 v);
#define WREG32_PIF_PHY1(reg, v) eg_pif_phy1_wreg(rdev, (reg), (v))
#define RREG32_UVD_CTX(reg) r600_uvd_ctx_rreg(rdev, (reg))
#define WREG32_UVD_CTX(reg, v) r600_uvd_ctx_wreg(rdev, (reg), (v))
+#define RREG32_DIDT(reg) cik_didt_rreg(rdev, (reg))
+#define WREG32_DIDT(reg, v) cik_didt_wreg(rdev, (reg), (v))
#define WREG32_P(reg, val, mask) \
do { \
uint32_t tmp_ = RREG32(reg); \
@@ -2272,6 +2274,22 @@ static inline void r600_uvd_ctx_wreg(struct radeon_device *rdev, u32 reg, u32 v)
WREG32(R600_UVD_CTX_DATA, (v));
}
+
+static inline u32 cik_didt_rreg(struct radeon_device *rdev, u32 reg)
+{
+ u32 r;
+
+ WREG32(CIK_DIDT_IND_INDEX, (reg));
+ r = RREG32(CIK_DIDT_IND_DATA);
+ return r;
+}
+
+static inline void cik_didt_wreg(struct radeon_device *rdev, u32 reg, u32 v)
+{
+ WREG32(CIK_DIDT_IND_INDEX, (reg));
+ WREG32(CIK_DIDT_IND_DATA, (v));
+}
+
void r100_pll_errata_after_index(struct radeon_device *rdev);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 18/53] drm/radeon/sumo add helper to go from vid7 to vid2
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (16 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 17/53] drm/radeon: add indirect accessors for dift registers on CIK Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 19/53] drm/radeon: switch to pptable.h Alex Deucher
` (35 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Needed for DPM on KB/KV.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/sumo_dpm.c | 14 ++++++++++++++
drivers/gpu/drm/radeon/sumo_dpm.h | 3 +++
2 files changed, 17 insertions(+)
diff --git a/drivers/gpu/drm/radeon/sumo_dpm.c b/drivers/gpu/drm/radeon/sumo_dpm.c
index c0a8503..2cefe59 100644
--- a/drivers/gpu/drm/radeon/sumo_dpm.c
+++ b/drivers/gpu/drm/radeon/sumo_dpm.c
@@ -1530,6 +1530,20 @@ u32 sumo_convert_vid2_to_vid7(struct radeon_device *rdev,
return vid_mapping_table->entries[vid_mapping_table->num_entries - 1].vid_7bit;
}
+u32 sumo_convert_vid7_to_vid2(struct radeon_device *rdev,
+ struct sumo_vid_mapping_table *vid_mapping_table,
+ u32 vid_7bit)
+{
+ u32 i;
+
+ for (i = 0; i < vid_mapping_table->num_entries; i++) {
+ if (vid_mapping_table->entries[i].vid_7bit == vid_7bit)
+ return vid_mapping_table->entries[i].vid_2bit;
+ }
+
+ return vid_mapping_table->entries[vid_mapping_table->num_entries - 1].vid_2bit;
+}
+
static u16 sumo_convert_voltage_index_to_value(struct radeon_device *rdev,
u32 vid_2bit)
{
diff --git a/drivers/gpu/drm/radeon/sumo_dpm.h b/drivers/gpu/drm/radeon/sumo_dpm.h
index 07dda29..db1ea32 100644
--- a/drivers/gpu/drm/radeon/sumo_dpm.h
+++ b/drivers/gpu/drm/radeon/sumo_dpm.h
@@ -202,6 +202,9 @@ void sumo_construct_vid_mapping_table(struct radeon_device *rdev,
u32 sumo_convert_vid2_to_vid7(struct radeon_device *rdev,
struct sumo_vid_mapping_table *vid_mapping_table,
u32 vid_2bit);
+u32 sumo_convert_vid7_to_vid2(struct radeon_device *rdev,
+ struct sumo_vid_mapping_table *vid_mapping_table,
+ u32 vid_7bit);
u32 sumo_get_sleep_divider_from_id(u32 id);
u32 sumo_get_sleep_divider_id_from_clock(struct radeon_device *rdev,
u32 sclk,
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 19/53] drm/radeon: switch to pptable.h
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (17 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 18/53] drm/radeon/sumo add helper to go from vid7 to vid2 Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 20/53] drm/radeon: add structs to store uvd clock voltage deps Alex Deucher
` (34 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Internally we switched to using a separate header for
atombios pplib definitions. Switch over the open source
driver.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/atombios.h | 615 +---------------------------------
drivers/gpu/drm/radeon/pptable.h | 682 ++++++++++++++++++++++++++++++++++++++
2 files changed, 685 insertions(+), 612 deletions(-)
create mode 100644 drivers/gpu/drm/radeon/pptable.h
diff --git a/drivers/gpu/drm/radeon/atombios.h b/drivers/gpu/drm/radeon/atombios.h
index 16b120c..af10f85 100644
--- a/drivers/gpu/drm/radeon/atombios.h
+++ b/drivers/gpu/drm/radeon/atombios.h
@@ -7661,618 +7661,6 @@ typedef struct _ATOM_POWERPLAY_INFO_V3
ATOM_POWERMODE_INFO_V3 asPowerPlayInfo[ATOM_MAX_NUMBEROF_POWER_BLOCK];
}ATOM_POWERPLAY_INFO_V3;
-/* New PPlib */
-/**************************************************************************/
-typedef struct _ATOM_PPLIB_THERMALCONTROLLER
-
-{
- UCHAR ucType; // one of ATOM_PP_THERMALCONTROLLER_*
- UCHAR ucI2cLine; // as interpreted by DAL I2C
- UCHAR ucI2cAddress;
- UCHAR ucFanParameters; // Fan Control Parameters.
- UCHAR ucFanMinRPM; // Fan Minimum RPM (hundreds) -- for display purposes only.
- UCHAR ucFanMaxRPM; // Fan Maximum RPM (hundreds) -- for display purposes only.
- UCHAR ucReserved; // ----
- UCHAR ucFlags; // to be defined
-} ATOM_PPLIB_THERMALCONTROLLER;
-
-#define ATOM_PP_FANPARAMETERS_TACHOMETER_PULSES_PER_REVOLUTION_MASK 0x0f
-#define ATOM_PP_FANPARAMETERS_NOFAN 0x80 // No fan is connected to this controller.
-
-#define ATOM_PP_THERMALCONTROLLER_NONE 0
-#define ATOM_PP_THERMALCONTROLLER_LM63 1 // Not used by PPLib
-#define ATOM_PP_THERMALCONTROLLER_ADM1032 2 // Not used by PPLib
-#define ATOM_PP_THERMALCONTROLLER_ADM1030 3 // Not used by PPLib
-#define ATOM_PP_THERMALCONTROLLER_MUA6649 4 // Not used by PPLib
-#define ATOM_PP_THERMALCONTROLLER_LM64 5
-#define ATOM_PP_THERMALCONTROLLER_F75375 6 // Not used by PPLib
-#define ATOM_PP_THERMALCONTROLLER_RV6xx 7
-#define ATOM_PP_THERMALCONTROLLER_RV770 8
-#define ATOM_PP_THERMALCONTROLLER_ADT7473 9
-#define ATOM_PP_THERMALCONTROLLER_EXTERNAL_GPIO 11
-#define ATOM_PP_THERMALCONTROLLER_EVERGREEN 12
-#define ATOM_PP_THERMALCONTROLLER_EMC2103 13 /* 0x0D */ // Only fan control will be implemented, do NOT show this in PPGen.
-#define ATOM_PP_THERMALCONTROLLER_SUMO 14 /* 0x0E */ // Sumo type, used internally
-#define ATOM_PP_THERMALCONTROLLER_NISLANDS 15
-#define ATOM_PP_THERMALCONTROLLER_SISLANDS 16
-#define ATOM_PP_THERMALCONTROLLER_LM96163 17
-#define ATOM_PP_THERMALCONTROLLER_CISLANDS 18
-
-// Thermal controller 'combo type' to use an external controller for Fan control and an internal controller for thermal.
-// We probably should reserve the bit 0x80 for this use.
-// To keep the number of these types low we should also use the same code for all ASICs (i.e. do not distinguish RV6xx and RV7xx Internal here).
-// The driver can pick the correct internal controller based on the ASIC.
-
-#define ATOM_PP_THERMALCONTROLLER_ADT7473_WITH_INTERNAL 0x89 // ADT7473 Fan Control + Internal Thermal Controller
-#define ATOM_PP_THERMALCONTROLLER_EMC2103_WITH_INTERNAL 0x8D // EMC2103 Fan Control + Internal Thermal Controller
-
-typedef struct _ATOM_PPLIB_STATE
-{
- UCHAR ucNonClockStateIndex;
- UCHAR ucClockStateIndices[1]; // variable-sized
-} ATOM_PPLIB_STATE;
-
-
-typedef struct _ATOM_PPLIB_FANTABLE
-{
- UCHAR ucFanTableFormat; // Change this if the table format changes or version changes so that the other fields are not the same.
- UCHAR ucTHyst; // Temperature hysteresis. Integer.
- USHORT usTMin; // The temperature, in 0.01 centigrades, below which we just run at a minimal PWM.
- USHORT usTMed; // The middle temperature where we change slopes.
- USHORT usTHigh; // The high point above TMed for adjusting the second slope.
- USHORT usPWMMin; // The minimum PWM value in percent (0.01% increments).
- USHORT usPWMMed; // The PWM value (in percent) at TMed.
- USHORT usPWMHigh; // The PWM value at THigh.
-} ATOM_PPLIB_FANTABLE;
-
-typedef struct _ATOM_PPLIB_FANTABLE2
-{
- ATOM_PPLIB_FANTABLE basicTable;
- USHORT usTMax; // The max temperature
-} ATOM_PPLIB_FANTABLE2;
-
-typedef struct _ATOM_PPLIB_EXTENDEDHEADER
-{
- USHORT usSize;
- ULONG ulMaxEngineClock; // For Overdrive.
- ULONG ulMaxMemoryClock; // For Overdrive.
- // Add extra system parameters here, always adjust size to include all fields.
- USHORT usVCETableOffset; //points to ATOM_PPLIB_VCE_Table
- USHORT usUVDTableOffset; //points to ATOM_PPLIB_UVD_Table
- USHORT usSAMUTableOffset; //points to ATOM_PPLIB_SAMU_Table
- USHORT usPPMTableOffset; //points to ATOM_PPLIB_PPM_Table
-} ATOM_PPLIB_EXTENDEDHEADER;
-
-//// ATOM_PPLIB_POWERPLAYTABLE::ulPlatformCaps
-#define ATOM_PP_PLATFORM_CAP_BACKBIAS 1
-#define ATOM_PP_PLATFORM_CAP_POWERPLAY 2
-#define ATOM_PP_PLATFORM_CAP_SBIOSPOWERSOURCE 4
-#define ATOM_PP_PLATFORM_CAP_ASPM_L0s 8
-#define ATOM_PP_PLATFORM_CAP_ASPM_L1 16
-#define ATOM_PP_PLATFORM_CAP_HARDWAREDC 32
-#define ATOM_PP_PLATFORM_CAP_GEMINIPRIMARY 64
-#define ATOM_PP_PLATFORM_CAP_STEPVDDC 128
-#define ATOM_PP_PLATFORM_CAP_VOLTAGECONTROL 256
-#define ATOM_PP_PLATFORM_CAP_SIDEPORTCONTROL 512
-#define ATOM_PP_PLATFORM_CAP_TURNOFFPLL_ASPML1 1024
-#define ATOM_PP_PLATFORM_CAP_HTLINKCONTROL 2048
-#define ATOM_PP_PLATFORM_CAP_MVDDCONTROL 4096
-#define ATOM_PP_PLATFORM_CAP_GOTO_BOOT_ON_ALERT 0x2000 // Go to boot state on alerts, e.g. on an AC->DC transition.
-#define ATOM_PP_PLATFORM_CAP_DONT_WAIT_FOR_VBLANK_ON_ALERT 0x4000 // Do NOT wait for VBLANK during an alert (e.g. AC->DC transition).
-#define ATOM_PP_PLATFORM_CAP_VDDCI_CONTROL 0x8000 // Does the driver control VDDCI independently from VDDC.
-#define ATOM_PP_PLATFORM_CAP_REGULATOR_HOT 0x00010000 // Enable the 'regulator hot' feature.
-#define ATOM_PP_PLATFORM_CAP_BACO 0x00020000 // Does the driver supports BACO state.
-#define ATOM_PP_PLATFORM_CAP_NEW_CAC_VOLTAGE 0x00040000 // Does the driver supports new CAC voltage table.
-#define ATOM_PP_PLATFORM_CAP_REVERT_GPIO5_POLARITY 0x00080000 // Does the driver supports revert GPIO5 polarity.
-#define ATOM_PP_PLATFORM_CAP_OUTPUT_THERMAL2GPIO17 0x00100000 // Does the driver supports thermal2GPIO17.
-#define ATOM_PP_PLATFORM_CAP_VRHOT_GPIO_CONFIGURABLE 0x00200000 // Does the driver supports VR HOT GPIO Configurable.
-
-typedef struct _ATOM_PPLIB_POWERPLAYTABLE
-{
- ATOM_COMMON_TABLE_HEADER sHeader;
-
- UCHAR ucDataRevision;
-
- UCHAR ucNumStates;
- UCHAR ucStateEntrySize;
- UCHAR ucClockInfoSize;
- UCHAR ucNonClockSize;
-
- // offset from start of this table to array of ucNumStates ATOM_PPLIB_STATE structures
- USHORT usStateArrayOffset;
-
- // offset from start of this table to array of ASIC-specific structures,
- // currently ATOM_PPLIB_CLOCK_INFO.
- USHORT usClockInfoArrayOffset;
-
- // offset from start of this table to array of ATOM_PPLIB_NONCLOCK_INFO
- USHORT usNonClockInfoArrayOffset;
-
- USHORT usBackbiasTime; // in microseconds
- USHORT usVoltageTime; // in microseconds
- USHORT usTableSize; //the size of this structure, or the extended structure
-
- ULONG ulPlatformCaps; // See ATOM_PPLIB_CAPS_*
-
- ATOM_PPLIB_THERMALCONTROLLER sThermalController;
-
- USHORT usBootClockInfoOffset;
- USHORT usBootNonClockInfoOffset;
-
-} ATOM_PPLIB_POWERPLAYTABLE;
-
-typedef struct _ATOM_PPLIB_POWERPLAYTABLE2
-{
- ATOM_PPLIB_POWERPLAYTABLE basicTable;
- UCHAR ucNumCustomThermalPolicy;
- USHORT usCustomThermalPolicyArrayOffset;
-}ATOM_PPLIB_POWERPLAYTABLE2, *LPATOM_PPLIB_POWERPLAYTABLE2;
-
-typedef struct _ATOM_PPLIB_POWERPLAYTABLE3
-{
- ATOM_PPLIB_POWERPLAYTABLE2 basicTable2;
- USHORT usFormatID; // To be used ONLY by PPGen.
- USHORT usFanTableOffset;
- USHORT usExtendendedHeaderOffset;
-} ATOM_PPLIB_POWERPLAYTABLE3, *LPATOM_PPLIB_POWERPLAYTABLE3;
-
-typedef struct _ATOM_PPLIB_POWERPLAYTABLE4
-{
- ATOM_PPLIB_POWERPLAYTABLE3 basicTable3;
- ULONG ulGoldenPPID; // PPGen use only
- ULONG ulGoldenRevision; // PPGen use only
- USHORT usVddcDependencyOnSCLKOffset;
- USHORT usVddciDependencyOnMCLKOffset;
- USHORT usVddcDependencyOnMCLKOffset;
- USHORT usMaxClockVoltageOnDCOffset;
- USHORT usVddcPhaseShedLimitsTableOffset; // Points to ATOM_PPLIB_PhaseSheddingLimits_Table
- USHORT usMvddDependencyOnMCLKOffset;
-} ATOM_PPLIB_POWERPLAYTABLE4, *LPATOM_PPLIB_POWERPLAYTABLE4;
-
-typedef struct _ATOM_PPLIB_POWERPLAYTABLE5
-{
- ATOM_PPLIB_POWERPLAYTABLE4 basicTable4;
- ULONG ulTDPLimit;
- ULONG ulNearTDPLimit;
- ULONG ulSQRampingThreshold;
- USHORT usCACLeakageTableOffset; // Points to ATOM_PPLIB_CAC_Leakage_Table
- ULONG ulCACLeakage; // The iLeakage for driver calculated CAC leakage table
- USHORT usTDPODLimit;
- USHORT usLoadLineSlope; // in milliOhms * 100
-} ATOM_PPLIB_POWERPLAYTABLE5, *LPATOM_PPLIB_POWERPLAYTABLE5;
-
-//// ATOM_PPLIB_NONCLOCK_INFO::usClassification
-#define ATOM_PPLIB_CLASSIFICATION_UI_MASK 0x0007
-#define ATOM_PPLIB_CLASSIFICATION_UI_SHIFT 0
-#define ATOM_PPLIB_CLASSIFICATION_UI_NONE 0
-#define ATOM_PPLIB_CLASSIFICATION_UI_BATTERY 1
-#define ATOM_PPLIB_CLASSIFICATION_UI_BALANCED 3
-#define ATOM_PPLIB_CLASSIFICATION_UI_PERFORMANCE 5
-// 2, 4, 6, 7 are reserved
-
-#define ATOM_PPLIB_CLASSIFICATION_BOOT 0x0008
-#define ATOM_PPLIB_CLASSIFICATION_THERMAL 0x0010
-#define ATOM_PPLIB_CLASSIFICATION_LIMITEDPOWERSOURCE 0x0020
-#define ATOM_PPLIB_CLASSIFICATION_REST 0x0040
-#define ATOM_PPLIB_CLASSIFICATION_FORCED 0x0080
-#define ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE 0x0100
-#define ATOM_PPLIB_CLASSIFICATION_OVERDRIVETEMPLATE 0x0200
-#define ATOM_PPLIB_CLASSIFICATION_UVDSTATE 0x0400
-#define ATOM_PPLIB_CLASSIFICATION_3DLOW 0x0800
-#define ATOM_PPLIB_CLASSIFICATION_ACPI 0x1000
-#define ATOM_PPLIB_CLASSIFICATION_HD2STATE 0x2000
-#define ATOM_PPLIB_CLASSIFICATION_HDSTATE 0x4000
-#define ATOM_PPLIB_CLASSIFICATION_SDSTATE 0x8000
-
-//// ATOM_PPLIB_NONCLOCK_INFO::usClassification2
-#define ATOM_PPLIB_CLASSIFICATION2_LIMITEDPOWERSOURCE_2 0x0001
-#define ATOM_PPLIB_CLASSIFICATION2_ULV 0x0002
-#define ATOM_PPLIB_CLASSIFICATION2_MVC 0x0004 //Multi-View Codec (BD-3D)
-
-//// ATOM_PPLIB_NONCLOCK_INFO::ulCapsAndSettings
-#define ATOM_PPLIB_SINGLE_DISPLAY_ONLY 0x00000001
-#define ATOM_PPLIB_SUPPORTS_VIDEO_PLAYBACK 0x00000002
-
-// 0 is 2.5Gb/s, 1 is 5Gb/s
-#define ATOM_PPLIB_PCIE_LINK_SPEED_MASK 0x00000004
-#define ATOM_PPLIB_PCIE_LINK_SPEED_SHIFT 2
-
-// lanes - 1: 1, 2, 4, 8, 12, 16 permitted by PCIE spec
-#define ATOM_PPLIB_PCIE_LINK_WIDTH_MASK 0x000000F8
-#define ATOM_PPLIB_PCIE_LINK_WIDTH_SHIFT 3
-
-// lookup into reduced refresh-rate table
-#define ATOM_PPLIB_LIMITED_REFRESHRATE_VALUE_MASK 0x00000F00
-#define ATOM_PPLIB_LIMITED_REFRESHRATE_VALUE_SHIFT 8
-
-#define ATOM_PPLIB_LIMITED_REFRESHRATE_UNLIMITED 0
-#define ATOM_PPLIB_LIMITED_REFRESHRATE_50HZ 1
-// 2-15 TBD as needed.
-
-#define ATOM_PPLIB_SOFTWARE_DISABLE_LOADBALANCING 0x00001000
-#define ATOM_PPLIB_SOFTWARE_ENABLE_SLEEP_FOR_TIMESTAMPS 0x00002000
-
-#define ATOM_PPLIB_DISALLOW_ON_DC 0x00004000
-
-#define ATOM_PPLIB_ENABLE_VARIBRIGHT 0x00008000
-
-//memory related flags
-#define ATOM_PPLIB_SWSTATE_MEMORY_DLL_OFF 0x000010000
-
-//M3 Arb //2bits, current 3 sets of parameters in total
-#define ATOM_PPLIB_M3ARB_MASK 0x00060000
-#define ATOM_PPLIB_M3ARB_SHIFT 17
-
-#define ATOM_PPLIB_ENABLE_DRR 0x00080000
-
-// remaining 16 bits are reserved
-typedef struct _ATOM_PPLIB_THERMAL_STATE
-{
- UCHAR ucMinTemperature;
- UCHAR ucMaxTemperature;
- UCHAR ucThermalAction;
-}ATOM_PPLIB_THERMAL_STATE, *LPATOM_PPLIB_THERMAL_STATE;
-
-// Contained in an array starting at the offset
-// in ATOM_PPLIB_POWERPLAYTABLE::usNonClockInfoArrayOffset.
-// referenced from ATOM_PPLIB_STATE_INFO::ucNonClockStateIndex
-#define ATOM_PPLIB_NONCLOCKINFO_VER1 12
-#define ATOM_PPLIB_NONCLOCKINFO_VER2 24
-typedef struct _ATOM_PPLIB_NONCLOCK_INFO
-{
- USHORT usClassification;
- UCHAR ucMinTemperature;
- UCHAR ucMaxTemperature;
- ULONG ulCapsAndSettings;
- UCHAR ucRequiredPower;
- USHORT usClassification2;
- ULONG ulVCLK;
- ULONG ulDCLK;
- UCHAR ucUnused[5];
-} ATOM_PPLIB_NONCLOCK_INFO;
-
-// Contained in an array starting at the offset
-// in ATOM_PPLIB_POWERPLAYTABLE::usClockInfoArrayOffset.
-// referenced from ATOM_PPLIB_STATE::ucClockStateIndices
-typedef struct _ATOM_PPLIB_R600_CLOCK_INFO
-{
- USHORT usEngineClockLow;
- UCHAR ucEngineClockHigh;
-
- USHORT usMemoryClockLow;
- UCHAR ucMemoryClockHigh;
-
- USHORT usVDDC;
- USHORT usUnused1;
- USHORT usUnused2;
-
- ULONG ulFlags; // ATOM_PPLIB_R600_FLAGS_*
-
-} ATOM_PPLIB_R600_CLOCK_INFO;
-
-// ulFlags in ATOM_PPLIB_R600_CLOCK_INFO
-#define ATOM_PPLIB_R600_FLAGS_PCIEGEN2 1
-#define ATOM_PPLIB_R600_FLAGS_UVDSAFE 2
-#define ATOM_PPLIB_R600_FLAGS_BACKBIASENABLE 4
-#define ATOM_PPLIB_R600_FLAGS_MEMORY_ODT_OFF 8
-#define ATOM_PPLIB_R600_FLAGS_MEMORY_DLL_OFF 16
-#define ATOM_PPLIB_R600_FLAGS_LOWPOWER 32 // On the RV770 use 'low power' setting (sequencer S0).
-
-typedef struct _ATOM_PPLIB_EVERGREEN_CLOCK_INFO
-{
- USHORT usEngineClockLow;
- UCHAR ucEngineClockHigh;
-
- USHORT usMemoryClockLow;
- UCHAR ucMemoryClockHigh;
-
- USHORT usVDDC;
- USHORT usVDDCI;
- USHORT usUnused;
-
- ULONG ulFlags; // ATOM_PPLIB_R600_FLAGS_*
-
-} ATOM_PPLIB_EVERGREEN_CLOCK_INFO;
-
-typedef struct _ATOM_PPLIB_SI_CLOCK_INFO
-{
- USHORT usEngineClockLow;
- UCHAR ucEngineClockHigh;
-
- USHORT usMemoryClockLow;
- UCHAR ucMemoryClockHigh;
-
- USHORT usVDDC;
- USHORT usVDDCI;
- UCHAR ucPCIEGen;
- UCHAR ucUnused1;
-
- ULONG ulFlags; // ATOM_PPLIB_SI_FLAGS_*, no flag is necessary for now
-
-} ATOM_PPLIB_SI_CLOCK_INFO;
-
-typedef struct _ATOM_PPLIB_CI_CLOCK_INFO
-{
- USHORT usEngineClockLow;
- UCHAR ucEngineClockHigh;
-
- USHORT usMemoryClockLow;
- UCHAR ucMemoryClockHigh;
-
- UCHAR ucPCIEGen;
- USHORT usPCIELane;
-} ATOM_PPLIB_CI_CLOCK_INFO;
-
-typedef struct _ATOM_PPLIB_RS780_CLOCK_INFO
-
-{
- USHORT usLowEngineClockLow; // Low Engine clock in MHz (the same way as on the R600).
- UCHAR ucLowEngineClockHigh;
- USHORT usHighEngineClockLow; // High Engine clock in MHz.
- UCHAR ucHighEngineClockHigh;
- USHORT usMemoryClockLow; // For now one of the ATOM_PPLIB_RS780_SPMCLK_XXXX constants.
- UCHAR ucMemoryClockHigh; // Currentyl unused.
- UCHAR ucPadding; // For proper alignment and size.
- USHORT usVDDC; // For the 780, use: None, Low, High, Variable
- UCHAR ucMaxHTLinkWidth; // From SBIOS - {2, 4, 8, 16}
- UCHAR ucMinHTLinkWidth; // From SBIOS - {2, 4, 8, 16}. Effective only if CDLW enabled. Minimum down stream width could be bigger as display BW requriement.
- USHORT usHTLinkFreq; // See definition ATOM_PPLIB_RS780_HTLINKFREQ_xxx or in MHz(>=200).
- ULONG ulFlags;
-} ATOM_PPLIB_RS780_CLOCK_INFO;
-
-#define ATOM_PPLIB_RS780_VOLTAGE_NONE 0
-#define ATOM_PPLIB_RS780_VOLTAGE_LOW 1
-#define ATOM_PPLIB_RS780_VOLTAGE_HIGH 2
-#define ATOM_PPLIB_RS780_VOLTAGE_VARIABLE 3
-
-#define ATOM_PPLIB_RS780_SPMCLK_NONE 0 // We cannot change the side port memory clock, leave it as it is.
-#define ATOM_PPLIB_RS780_SPMCLK_LOW 1
-#define ATOM_PPLIB_RS780_SPMCLK_HIGH 2
-
-#define ATOM_PPLIB_RS780_HTLINKFREQ_NONE 0
-#define ATOM_PPLIB_RS780_HTLINKFREQ_LOW 1
-#define ATOM_PPLIB_RS780_HTLINKFREQ_HIGH 2
-
-typedef struct _ATOM_PPLIB_SUMO_CLOCK_INFO{
- USHORT usEngineClockLow; //clockfrequency & 0xFFFF. The unit is in 10khz
- UCHAR ucEngineClockHigh; //clockfrequency >> 16.
- UCHAR vddcIndex; //2-bit vddc index;
- USHORT tdpLimit;
- //please initalize to 0
- USHORT rsv1;
- //please initialize to 0s
- ULONG rsv2[2];
-}ATOM_PPLIB_SUMO_CLOCK_INFO;
-
-
-
-typedef struct _ATOM_PPLIB_STATE_V2
-{
- //number of valid dpm levels in this state; Driver uses it to calculate the whole
- //size of the state: sizeof(ATOM_PPLIB_STATE_V2) + (ucNumDPMLevels - 1) * sizeof(UCHAR)
- UCHAR ucNumDPMLevels;
-
- //a index to the array of nonClockInfos
- UCHAR nonClockInfoIndex;
- /**
- * Driver will read the first ucNumDPMLevels in this array
- */
- UCHAR clockInfoIndex[1];
-} ATOM_PPLIB_STATE_V2;
-
-typedef struct _StateArray{
- //how many states we have
- UCHAR ucNumEntries;
-
- ATOM_PPLIB_STATE_V2 states[1];
-}StateArray;
-
-
-typedef struct _ClockInfoArray{
- //how many clock levels we have
- UCHAR ucNumEntries;
-
- //sizeof(ATOM_PPLIB_CLOCK_INFO)
- UCHAR ucEntrySize;
-
- UCHAR clockInfo[1];
-}ClockInfoArray;
-
-typedef struct _NonClockInfoArray{
-
- //how many non-clock levels we have. normally should be same as number of states
- UCHAR ucNumEntries;
- //sizeof(ATOM_PPLIB_NONCLOCK_INFO)
- UCHAR ucEntrySize;
-
- ATOM_PPLIB_NONCLOCK_INFO nonClockInfo[1];
-}NonClockInfoArray;
-
-typedef struct _ATOM_PPLIB_Clock_Voltage_Dependency_Record
-{
- USHORT usClockLow;
- UCHAR ucClockHigh;
- USHORT usVoltage;
-}ATOM_PPLIB_Clock_Voltage_Dependency_Record;
-
-typedef struct _ATOM_PPLIB_Clock_Voltage_Dependency_Table
-{
- UCHAR ucNumEntries; // Number of entries.
- ATOM_PPLIB_Clock_Voltage_Dependency_Record entries[1]; // Dynamically allocate entries.
-}ATOM_PPLIB_Clock_Voltage_Dependency_Table;
-
-typedef struct _ATOM_PPLIB_Clock_Voltage_Limit_Record
-{
- USHORT usSclkLow;
- UCHAR ucSclkHigh;
- USHORT usMclkLow;
- UCHAR ucMclkHigh;
- USHORT usVddc;
- USHORT usVddci;
-}ATOM_PPLIB_Clock_Voltage_Limit_Record;
-
-typedef struct _ATOM_PPLIB_Clock_Voltage_Limit_Table
-{
- UCHAR ucNumEntries; // Number of entries.
- ATOM_PPLIB_Clock_Voltage_Limit_Record entries[1]; // Dynamically allocate entries.
-}ATOM_PPLIB_Clock_Voltage_Limit_Table;
-
-typedef struct _ATOM_PPLIB_CAC_Leakage_Record
-{
- USHORT usVddc; // We use this field for the "fake" standardized VDDC for power calculations; For CI and newer, we use this as the real VDDC value.
- ULONG ulLeakageValue; // For CI and newer we use this as the "fake" standar VDDC value.
-}ATOM_PPLIB_CAC_Leakage_Record;
-
-typedef struct _ATOM_PPLIB_CAC_Leakage_Table
-{
- UCHAR ucNumEntries; // Number of entries.
- ATOM_PPLIB_CAC_Leakage_Record entries[1]; // Dynamically allocate entries.
-}ATOM_PPLIB_CAC_Leakage_Table;
-
-typedef struct _ATOM_PPLIB_PhaseSheddingLimits_Record
-{
- USHORT usVoltage;
- USHORT usSclkLow;
- UCHAR ucSclkHigh;
- USHORT usMclkLow;
- UCHAR ucMclkHigh;
-}ATOM_PPLIB_PhaseSheddingLimits_Record;
-
-typedef struct _ATOM_PPLIB_PhaseSheddingLimits_Table
-{
- UCHAR ucNumEntries; // Number of entries.
- ATOM_PPLIB_PhaseSheddingLimits_Record entries[1]; // Dynamically allocate entries.
-}ATOM_PPLIB_PhaseSheddingLimits_Table;
-
-typedef struct _VCEClockInfo{
- USHORT usEVClkLow;
- UCHAR ucEVClkHigh;
- USHORT usECClkLow;
- UCHAR ucECClkHigh;
-}VCEClockInfo;
-
-typedef struct _VCEClockInfoArray{
- UCHAR ucNumEntries;
- VCEClockInfo entries[1];
-}VCEClockInfoArray;
-
-typedef struct _ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record
-{
- USHORT usVoltage;
- UCHAR ucVCEClockInfoIndex;
-}ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record;
-
-typedef struct _ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table
-{
- UCHAR numEntries;
- ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record entries[1];
-}ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table;
-
-typedef struct _ATOM_PPLIB_VCE_State_Record
-{
- UCHAR ucVCEClockInfoIndex;
- UCHAR ucClockInfoIndex; //highest 2 bits indicates memory p-states, lower 6bits indicates index to ClockInfoArrary
-}ATOM_PPLIB_VCE_State_Record;
-
-typedef struct _ATOM_PPLIB_VCE_State_Table
-{
- UCHAR numEntries;
- ATOM_PPLIB_VCE_State_Record entries[1];
-}ATOM_PPLIB_VCE_State_Table;
-
-
-typedef struct _ATOM_PPLIB_VCE_Table
-{
- UCHAR revid;
-// VCEClockInfoArray array;
-// ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table limits;
-// ATOM_PPLIB_VCE_State_Table states;
-}ATOM_PPLIB_VCE_Table;
-
-
-typedef struct _UVDClockInfo{
- USHORT usVClkLow;
- UCHAR ucVClkHigh;
- USHORT usDClkLow;
- UCHAR ucDClkHigh;
-}UVDClockInfo;
-
-typedef struct _UVDClockInfoArray{
- UCHAR ucNumEntries;
- UVDClockInfo entries[1];
-}UVDClockInfoArray;
-
-typedef struct _ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record
-{
- USHORT usVoltage;
- UCHAR ucUVDClockInfoIndex;
-}ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record;
-
-typedef struct _ATOM_PPLIB_UVD_Clock_Voltage_Limit_Table
-{
- UCHAR numEntries;
- ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record entries[1];
-}ATOM_PPLIB_UVD_Clock_Voltage_Limit_Table;
-
-typedef struct _ATOM_PPLIB_UVD_State_Record
-{
- UCHAR ucUVDClockInfoIndex;
- UCHAR ucClockInfoIndex; //highest 2 bits indicates memory p-states, lower 6bits indicates index to ClockInfoArrary
-}ATOM_PPLIB_UVD_State_Record;
-
-typedef struct _ATOM_PPLIB_UVD_State_Table
-{
- UCHAR numEntries;
- ATOM_PPLIB_UVD_State_Record entries[1];
-}ATOM_PPLIB_UVD_State_Table;
-
-
-typedef struct _ATOM_PPLIB_UVD_Table
-{
- UCHAR revid;
-// UVDClockInfoArray array;
-// ATOM_PPLIB_UVD_Clock_Voltage_Limit_Table limits;
-// ATOM_PPLIB_UVD_State_Table states;
-}ATOM_PPLIB_UVD_Table;
-
-
-typedef struct _ATOM_PPLIB_SAMClk_Voltage_Limit_Record
-{
- USHORT usVoltage;
- USHORT usSAMClockLow;
- UCHAR ucSAMClockHigh;
-}ATOM_PPLIB_SAMClk_Voltage_Limit_Record;
-
-typedef struct _ATOM_PPLIB_SAMClk_Voltage_Limit_Table{
- UCHAR numEntries;
- ATOM_PPLIB_SAMClk_Voltage_Limit_Record entries[1];
-}ATOM_PPLIB_SAMClk_Voltage_Limit_Table;
-
-typedef struct _ATOM_PPLIB_SAMU_Table
-{
- UCHAR revid;
- ATOM_PPLIB_SAMClk_Voltage_Limit_Table limits;
-}ATOM_PPLIB_SAMU_Table;
-
-#define ATOM_PPM_A_A 1
-#define ATOM_PPM_A_I 2
-typedef struct _ATOM_PPLIB_PPM_Table
-{
- UCHAR ucRevId;
- UCHAR ucPpmDesign; //A+I or A+A
- USHORT usCpuCoreNumber;
- ULONG ulPlatformTDP;
- ULONG ulSmallACPlatformTDP;
- ULONG ulPlatformTDC;
- ULONG ulSmallACPlatformTDC;
- ULONG ulApuTDP;
- ULONG ulDGpuTDP;
- ULONG ulDGpuUlvPower;
- ULONG ulTjmax;
-} ATOM_PPLIB_PPM_Table;
-
-/**************************************************************************/
-
// Following definitions are for compatibility issue in different SW components.
#define ATOM_MASTER_DATA_TABLE_REVISION 0x01
@@ -8485,3 +7873,6 @@ typedef struct {
#endif /* _ATOMBIOS_H */
+
+#include "pptable.h"
+
diff --git a/drivers/gpu/drm/radeon/pptable.h b/drivers/gpu/drm/radeon/pptable.h
new file mode 100644
index 0000000..da43ab3
--- /dev/null
+++ b/drivers/gpu/drm/radeon/pptable.h
@@ -0,0 +1,682 @@
+/*
+ * Copyright 2013 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef _PPTABLE_H
+#define _PPTABLE_H
+
+#pragma pack(push, 1)
+
+typedef struct _ATOM_PPLIB_THERMALCONTROLLER
+
+{
+ UCHAR ucType; // one of ATOM_PP_THERMALCONTROLLER_*
+ UCHAR ucI2cLine; // as interpreted by DAL I2C
+ UCHAR ucI2cAddress;
+ UCHAR ucFanParameters; // Fan Control Parameters.
+ UCHAR ucFanMinRPM; // Fan Minimum RPM (hundreds) -- for display purposes only.
+ UCHAR ucFanMaxRPM; // Fan Maximum RPM (hundreds) -- for display purposes only.
+ UCHAR ucReserved; // ----
+ UCHAR ucFlags; // to be defined
+} ATOM_PPLIB_THERMALCONTROLLER;
+
+#define ATOM_PP_FANPARAMETERS_TACHOMETER_PULSES_PER_REVOLUTION_MASK 0x0f
+#define ATOM_PP_FANPARAMETERS_NOFAN 0x80 // No fan is connected to this controller.
+
+#define ATOM_PP_THERMALCONTROLLER_NONE 0
+#define ATOM_PP_THERMALCONTROLLER_LM63 1 // Not used by PPLib
+#define ATOM_PP_THERMALCONTROLLER_ADM1032 2 // Not used by PPLib
+#define ATOM_PP_THERMALCONTROLLER_ADM1030 3 // Not used by PPLib
+#define ATOM_PP_THERMALCONTROLLER_MUA6649 4 // Not used by PPLib
+#define ATOM_PP_THERMALCONTROLLER_LM64 5
+#define ATOM_PP_THERMALCONTROLLER_F75375 6 // Not used by PPLib
+#define ATOM_PP_THERMALCONTROLLER_RV6xx 7
+#define ATOM_PP_THERMALCONTROLLER_RV770 8
+#define ATOM_PP_THERMALCONTROLLER_ADT7473 9
+#define ATOM_PP_THERMALCONTROLLER_KONG 10
+#define ATOM_PP_THERMALCONTROLLER_EXTERNAL_GPIO 11
+#define ATOM_PP_THERMALCONTROLLER_EVERGREEN 12
+#define ATOM_PP_THERMALCONTROLLER_EMC2103 13 /* 0x0D */ // Only fan control will be implemented, do NOT show this in PPGen.
+#define ATOM_PP_THERMALCONTROLLER_SUMO 14 /* 0x0E */ // Sumo type, used internally
+#define ATOM_PP_THERMALCONTROLLER_NISLANDS 15
+#define ATOM_PP_THERMALCONTROLLER_SISLANDS 16
+#define ATOM_PP_THERMALCONTROLLER_LM96163 17
+#define ATOM_PP_THERMALCONTROLLER_CISLANDS 18
+#define ATOM_PP_THERMALCONTROLLER_KAVERI 19
+
+
+// Thermal controller 'combo type' to use an external controller for Fan control and an internal controller for thermal.
+// We probably should reserve the bit 0x80 for this use.
+// To keep the number of these types low we should also use the same code for all ASICs (i.e. do not distinguish RV6xx and RV7xx Internal here).
+// The driver can pick the correct internal controller based on the ASIC.
+
+#define ATOM_PP_THERMALCONTROLLER_ADT7473_WITH_INTERNAL 0x89 // ADT7473 Fan Control + Internal Thermal Controller
+#define ATOM_PP_THERMALCONTROLLER_EMC2103_WITH_INTERNAL 0x8D // EMC2103 Fan Control + Internal Thermal Controller
+
+typedef struct _ATOM_PPLIB_STATE
+{
+ UCHAR ucNonClockStateIndex;
+ UCHAR ucClockStateIndices[1]; // variable-sized
+} ATOM_PPLIB_STATE;
+
+
+typedef struct _ATOM_PPLIB_FANTABLE
+{
+ UCHAR ucFanTableFormat; // Change this if the table format changes or version changes so that the other fields are not the same.
+ UCHAR ucTHyst; // Temperature hysteresis. Integer.
+ USHORT usTMin; // The temperature, in 0.01 centigrades, below which we just run at a minimal PWM.
+ USHORT usTMed; // The middle temperature where we change slopes.
+ USHORT usTHigh; // The high point above TMed for adjusting the second slope.
+ USHORT usPWMMin; // The minimum PWM value in percent (0.01% increments).
+ USHORT usPWMMed; // The PWM value (in percent) at TMed.
+ USHORT usPWMHigh; // The PWM value at THigh.
+} ATOM_PPLIB_FANTABLE;
+
+typedef struct _ATOM_PPLIB_FANTABLE2
+{
+ ATOM_PPLIB_FANTABLE basicTable;
+ USHORT usTMax; // The max temperature
+} ATOM_PPLIB_FANTABLE2;
+
+typedef struct _ATOM_PPLIB_EXTENDEDHEADER
+{
+ USHORT usSize;
+ ULONG ulMaxEngineClock; // For Overdrive.
+ ULONG ulMaxMemoryClock; // For Overdrive.
+ // Add extra system parameters here, always adjust size to include all fields.
+ USHORT usVCETableOffset; //points to ATOM_PPLIB_VCE_Table
+ USHORT usUVDTableOffset; //points to ATOM_PPLIB_UVD_Table
+ USHORT usSAMUTableOffset; //points to ATOM_PPLIB_SAMU_Table
+ USHORT usPPMTableOffset; //points to ATOM_PPLIB_PPM_Table
+ USHORT usACPTableOffset; //points to ATOM_PPLIB_ACP_Table
+ USHORT usPowerTuneTableOffset; //points to ATOM_PPLIB_POWERTUNE_Table
+} ATOM_PPLIB_EXTENDEDHEADER;
+
+//// ATOM_PPLIB_POWERPLAYTABLE::ulPlatformCaps
+#define ATOM_PP_PLATFORM_CAP_BACKBIAS 1
+#define ATOM_PP_PLATFORM_CAP_POWERPLAY 2
+#define ATOM_PP_PLATFORM_CAP_SBIOSPOWERSOURCE 4
+#define ATOM_PP_PLATFORM_CAP_ASPM_L0s 8
+#define ATOM_PP_PLATFORM_CAP_ASPM_L1 16
+#define ATOM_PP_PLATFORM_CAP_HARDWAREDC 32
+#define ATOM_PP_PLATFORM_CAP_GEMINIPRIMARY 64
+#define ATOM_PP_PLATFORM_CAP_STEPVDDC 128
+#define ATOM_PP_PLATFORM_CAP_VOLTAGECONTROL 256
+#define ATOM_PP_PLATFORM_CAP_SIDEPORTCONTROL 512
+#define ATOM_PP_PLATFORM_CAP_TURNOFFPLL_ASPML1 1024
+#define ATOM_PP_PLATFORM_CAP_HTLINKCONTROL 2048
+#define ATOM_PP_PLATFORM_CAP_MVDDCONTROL 4096
+#define ATOM_PP_PLATFORM_CAP_GOTO_BOOT_ON_ALERT 0x2000 // Go to boot state on alerts, e.g. on an AC->DC transition.
+#define ATOM_PP_PLATFORM_CAP_DONT_WAIT_FOR_VBLANK_ON_ALERT 0x4000 // Do NOT wait for VBLANK during an alert (e.g. AC->DC transition).
+#define ATOM_PP_PLATFORM_CAP_VDDCI_CONTROL 0x8000 // Does the driver control VDDCI independently from VDDC.
+#define ATOM_PP_PLATFORM_CAP_REGULATOR_HOT 0x00010000 // Enable the 'regulator hot' feature.
+#define ATOM_PP_PLATFORM_CAP_BACO 0x00020000 // Does the driver supports BACO state.
+#define ATOM_PP_PLATFORM_CAP_NEW_CAC_VOLTAGE 0x00040000 // Does the driver supports new CAC voltage table.
+#define ATOM_PP_PLATFORM_CAP_REVERT_GPIO5_POLARITY 0x00080000 // Does the driver supports revert GPIO5 polarity.
+#define ATOM_PP_PLATFORM_CAP_OUTPUT_THERMAL2GPIO17 0x00100000 // Does the driver supports thermal2GPIO17.
+#define ATOM_PP_PLATFORM_CAP_VRHOT_GPIO_CONFIGURABLE 0x00200000 // Does the driver supports VR HOT GPIO Configurable.
+#define ATOM_PP_PLATFORM_CAP_TEMP_INVERSION 0x00400000 // Does the driver supports Temp Inversion feature.
+#define ATOM_PP_PLATFORM_CAP_EVV 0x00800000
+
+typedef struct _ATOM_PPLIB_POWERPLAYTABLE
+{
+ ATOM_COMMON_TABLE_HEADER sHeader;
+
+ UCHAR ucDataRevision;
+
+ UCHAR ucNumStates;
+ UCHAR ucStateEntrySize;
+ UCHAR ucClockInfoSize;
+ UCHAR ucNonClockSize;
+
+ // offset from start of this table to array of ucNumStates ATOM_PPLIB_STATE structures
+ USHORT usStateArrayOffset;
+
+ // offset from start of this table to array of ASIC-specific structures,
+ // currently ATOM_PPLIB_CLOCK_INFO.
+ USHORT usClockInfoArrayOffset;
+
+ // offset from start of this table to array of ATOM_PPLIB_NONCLOCK_INFO
+ USHORT usNonClockInfoArrayOffset;
+
+ USHORT usBackbiasTime; // in microseconds
+ USHORT usVoltageTime; // in microseconds
+ USHORT usTableSize; //the size of this structure, or the extended structure
+
+ ULONG ulPlatformCaps; // See ATOM_PPLIB_CAPS_*
+
+ ATOM_PPLIB_THERMALCONTROLLER sThermalController;
+
+ USHORT usBootClockInfoOffset;
+ USHORT usBootNonClockInfoOffset;
+
+} ATOM_PPLIB_POWERPLAYTABLE;
+
+typedef struct _ATOM_PPLIB_POWERPLAYTABLE2
+{
+ ATOM_PPLIB_POWERPLAYTABLE basicTable;
+ UCHAR ucNumCustomThermalPolicy;
+ USHORT usCustomThermalPolicyArrayOffset;
+}ATOM_PPLIB_POWERPLAYTABLE2, *LPATOM_PPLIB_POWERPLAYTABLE2;
+
+typedef struct _ATOM_PPLIB_POWERPLAYTABLE3
+{
+ ATOM_PPLIB_POWERPLAYTABLE2 basicTable2;
+ USHORT usFormatID; // To be used ONLY by PPGen.
+ USHORT usFanTableOffset;
+ USHORT usExtendendedHeaderOffset;
+} ATOM_PPLIB_POWERPLAYTABLE3, *LPATOM_PPLIB_POWERPLAYTABLE3;
+
+typedef struct _ATOM_PPLIB_POWERPLAYTABLE4
+{
+ ATOM_PPLIB_POWERPLAYTABLE3 basicTable3;
+ ULONG ulGoldenPPID; // PPGen use only
+ ULONG ulGoldenRevision; // PPGen use only
+ USHORT usVddcDependencyOnSCLKOffset;
+ USHORT usVddciDependencyOnMCLKOffset;
+ USHORT usVddcDependencyOnMCLKOffset;
+ USHORT usMaxClockVoltageOnDCOffset;
+ USHORT usVddcPhaseShedLimitsTableOffset; // Points to ATOM_PPLIB_PhaseSheddingLimits_Table
+ USHORT usMvddDependencyOnMCLKOffset;
+} ATOM_PPLIB_POWERPLAYTABLE4, *LPATOM_PPLIB_POWERPLAYTABLE4;
+
+typedef struct _ATOM_PPLIB_POWERPLAYTABLE5
+{
+ ATOM_PPLIB_POWERPLAYTABLE4 basicTable4;
+ ULONG ulTDPLimit;
+ ULONG ulNearTDPLimit;
+ ULONG ulSQRampingThreshold;
+ USHORT usCACLeakageTableOffset; // Points to ATOM_PPLIB_CAC_Leakage_Table
+ ULONG ulCACLeakage; // The iLeakage for driver calculated CAC leakage table
+ USHORT usTDPODLimit;
+ USHORT usLoadLineSlope; // in milliOhms * 100
+} ATOM_PPLIB_POWERPLAYTABLE5, *LPATOM_PPLIB_POWERPLAYTABLE5;
+
+//// ATOM_PPLIB_NONCLOCK_INFO::usClassification
+#define ATOM_PPLIB_CLASSIFICATION_UI_MASK 0x0007
+#define ATOM_PPLIB_CLASSIFICATION_UI_SHIFT 0
+#define ATOM_PPLIB_CLASSIFICATION_UI_NONE 0
+#define ATOM_PPLIB_CLASSIFICATION_UI_BATTERY 1
+#define ATOM_PPLIB_CLASSIFICATION_UI_BALANCED 3
+#define ATOM_PPLIB_CLASSIFICATION_UI_PERFORMANCE 5
+// 2, 4, 6, 7 are reserved
+
+#define ATOM_PPLIB_CLASSIFICATION_BOOT 0x0008
+#define ATOM_PPLIB_CLASSIFICATION_THERMAL 0x0010
+#define ATOM_PPLIB_CLASSIFICATION_LIMITEDPOWERSOURCE 0x0020
+#define ATOM_PPLIB_CLASSIFICATION_REST 0x0040
+#define ATOM_PPLIB_CLASSIFICATION_FORCED 0x0080
+#define ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE 0x0100
+#define ATOM_PPLIB_CLASSIFICATION_OVERDRIVETEMPLATE 0x0200
+#define ATOM_PPLIB_CLASSIFICATION_UVDSTATE 0x0400
+#define ATOM_PPLIB_CLASSIFICATION_3DLOW 0x0800
+#define ATOM_PPLIB_CLASSIFICATION_ACPI 0x1000
+#define ATOM_PPLIB_CLASSIFICATION_HD2STATE 0x2000
+#define ATOM_PPLIB_CLASSIFICATION_HDSTATE 0x4000
+#define ATOM_PPLIB_CLASSIFICATION_SDSTATE 0x8000
+
+//// ATOM_PPLIB_NONCLOCK_INFO::usClassification2
+#define ATOM_PPLIB_CLASSIFICATION2_LIMITEDPOWERSOURCE_2 0x0001
+#define ATOM_PPLIB_CLASSIFICATION2_ULV 0x0002
+#define ATOM_PPLIB_CLASSIFICATION2_MVC 0x0004 //Multi-View Codec (BD-3D)
+
+//// ATOM_PPLIB_NONCLOCK_INFO::ulCapsAndSettings
+#define ATOM_PPLIB_SINGLE_DISPLAY_ONLY 0x00000001
+#define ATOM_PPLIB_SUPPORTS_VIDEO_PLAYBACK 0x00000002
+
+// 0 is 2.5Gb/s, 1 is 5Gb/s
+#define ATOM_PPLIB_PCIE_LINK_SPEED_MASK 0x00000004
+#define ATOM_PPLIB_PCIE_LINK_SPEED_SHIFT 2
+
+// lanes - 1: 1, 2, 4, 8, 12, 16 permitted by PCIE spec
+#define ATOM_PPLIB_PCIE_LINK_WIDTH_MASK 0x000000F8
+#define ATOM_PPLIB_PCIE_LINK_WIDTH_SHIFT 3
+
+// lookup into reduced refresh-rate table
+#define ATOM_PPLIB_LIMITED_REFRESHRATE_VALUE_MASK 0x00000F00
+#define ATOM_PPLIB_LIMITED_REFRESHRATE_VALUE_SHIFT 8
+
+#define ATOM_PPLIB_LIMITED_REFRESHRATE_UNLIMITED 0
+#define ATOM_PPLIB_LIMITED_REFRESHRATE_50HZ 1
+// 2-15 TBD as needed.
+
+#define ATOM_PPLIB_SOFTWARE_DISABLE_LOADBALANCING 0x00001000
+#define ATOM_PPLIB_SOFTWARE_ENABLE_SLEEP_FOR_TIMESTAMPS 0x00002000
+
+#define ATOM_PPLIB_DISALLOW_ON_DC 0x00004000
+
+#define ATOM_PPLIB_ENABLE_VARIBRIGHT 0x00008000
+
+//memory related flags
+#define ATOM_PPLIB_SWSTATE_MEMORY_DLL_OFF 0x000010000
+
+//M3 Arb //2bits, current 3 sets of parameters in total
+#define ATOM_PPLIB_M3ARB_MASK 0x00060000
+#define ATOM_PPLIB_M3ARB_SHIFT 17
+
+#define ATOM_PPLIB_ENABLE_DRR 0x00080000
+
+// remaining 16 bits are reserved
+typedef struct _ATOM_PPLIB_THERMAL_STATE
+{
+ UCHAR ucMinTemperature;
+ UCHAR ucMaxTemperature;
+ UCHAR ucThermalAction;
+}ATOM_PPLIB_THERMAL_STATE, *LPATOM_PPLIB_THERMAL_STATE;
+
+// Contained in an array starting at the offset
+// in ATOM_PPLIB_POWERPLAYTABLE::usNonClockInfoArrayOffset.
+// referenced from ATOM_PPLIB_STATE_INFO::ucNonClockStateIndex
+#define ATOM_PPLIB_NONCLOCKINFO_VER1 12
+#define ATOM_PPLIB_NONCLOCKINFO_VER2 24
+typedef struct _ATOM_PPLIB_NONCLOCK_INFO
+{
+ USHORT usClassification;
+ UCHAR ucMinTemperature;
+ UCHAR ucMaxTemperature;
+ ULONG ulCapsAndSettings;
+ UCHAR ucRequiredPower;
+ USHORT usClassification2;
+ ULONG ulVCLK;
+ ULONG ulDCLK;
+ UCHAR ucUnused[5];
+} ATOM_PPLIB_NONCLOCK_INFO;
+
+// Contained in an array starting at the offset
+// in ATOM_PPLIB_POWERPLAYTABLE::usClockInfoArrayOffset.
+// referenced from ATOM_PPLIB_STATE::ucClockStateIndices
+typedef struct _ATOM_PPLIB_R600_CLOCK_INFO
+{
+ USHORT usEngineClockLow;
+ UCHAR ucEngineClockHigh;
+
+ USHORT usMemoryClockLow;
+ UCHAR ucMemoryClockHigh;
+
+ USHORT usVDDC;
+ USHORT usUnused1;
+ USHORT usUnused2;
+
+ ULONG ulFlags; // ATOM_PPLIB_R600_FLAGS_*
+
+} ATOM_PPLIB_R600_CLOCK_INFO;
+
+// ulFlags in ATOM_PPLIB_R600_CLOCK_INFO
+#define ATOM_PPLIB_R600_FLAGS_PCIEGEN2 1
+#define ATOM_PPLIB_R600_FLAGS_UVDSAFE 2
+#define ATOM_PPLIB_R600_FLAGS_BACKBIASENABLE 4
+#define ATOM_PPLIB_R600_FLAGS_MEMORY_ODT_OFF 8
+#define ATOM_PPLIB_R600_FLAGS_MEMORY_DLL_OFF 16
+#define ATOM_PPLIB_R600_FLAGS_LOWPOWER 32 // On the RV770 use 'low power' setting (sequencer S0).
+
+typedef struct _ATOM_PPLIB_RS780_CLOCK_INFO
+
+{
+ USHORT usLowEngineClockLow; // Low Engine clock in MHz (the same way as on the R600).
+ UCHAR ucLowEngineClockHigh;
+ USHORT usHighEngineClockLow; // High Engine clock in MHz.
+ UCHAR ucHighEngineClockHigh;
+ USHORT usMemoryClockLow; // For now one of the ATOM_PPLIB_RS780_SPMCLK_XXXX constants.
+ UCHAR ucMemoryClockHigh; // Currentyl unused.
+ UCHAR ucPadding; // For proper alignment and size.
+ USHORT usVDDC; // For the 780, use: None, Low, High, Variable
+ UCHAR ucMaxHTLinkWidth; // From SBIOS - {2, 4, 8, 16}
+ UCHAR ucMinHTLinkWidth; // From SBIOS - {2, 4, 8, 16}. Effective only if CDLW enabled. Minimum down stream width could
+ USHORT usHTLinkFreq; // See definition ATOM_PPLIB_RS780_HTLINKFREQ_xxx or in MHz(>=200).
+ ULONG ulFlags;
+} ATOM_PPLIB_RS780_CLOCK_INFO;
+
+#define ATOM_PPLIB_RS780_VOLTAGE_NONE 0
+#define ATOM_PPLIB_RS780_VOLTAGE_LOW 1
+#define ATOM_PPLIB_RS780_VOLTAGE_HIGH 2
+#define ATOM_PPLIB_RS780_VOLTAGE_VARIABLE 3
+
+#define ATOM_PPLIB_RS780_SPMCLK_NONE 0 // We cannot change the side port memory clock, leave it as it is.
+#define ATOM_PPLIB_RS780_SPMCLK_LOW 1
+#define ATOM_PPLIB_RS780_SPMCLK_HIGH 2
+
+#define ATOM_PPLIB_RS780_HTLINKFREQ_NONE 0
+#define ATOM_PPLIB_RS780_HTLINKFREQ_LOW 1
+#define ATOM_PPLIB_RS780_HTLINKFREQ_HIGH 2
+
+typedef struct _ATOM_PPLIB_EVERGREEN_CLOCK_INFO
+{
+ USHORT usEngineClockLow;
+ UCHAR ucEngineClockHigh;
+
+ USHORT usMemoryClockLow;
+ UCHAR ucMemoryClockHigh;
+
+ USHORT usVDDC;
+ USHORT usVDDCI;
+ USHORT usUnused;
+
+ ULONG ulFlags; // ATOM_PPLIB_R600_FLAGS_*
+
+} ATOM_PPLIB_EVERGREEN_CLOCK_INFO;
+
+typedef struct _ATOM_PPLIB_SI_CLOCK_INFO
+{
+ USHORT usEngineClockLow;
+ UCHAR ucEngineClockHigh;
+
+ USHORT usMemoryClockLow;
+ UCHAR ucMemoryClockHigh;
+
+ USHORT usVDDC;
+ USHORT usVDDCI;
+ UCHAR ucPCIEGen;
+ UCHAR ucUnused1;
+
+ ULONG ulFlags; // ATOM_PPLIB_SI_FLAGS_*, no flag is necessary for now
+
+} ATOM_PPLIB_SI_CLOCK_INFO;
+
+typedef struct _ATOM_PPLIB_CI_CLOCK_INFO
+{
+ USHORT usEngineClockLow;
+ UCHAR ucEngineClockHigh;
+
+ USHORT usMemoryClockLow;
+ UCHAR ucMemoryClockHigh;
+
+ UCHAR ucPCIEGen;
+ USHORT usPCIELane;
+} ATOM_PPLIB_CI_CLOCK_INFO;
+
+typedef struct _ATOM_PPLIB_SUMO_CLOCK_INFO{
+ USHORT usEngineClockLow; //clockfrequency & 0xFFFF. The unit is in 10khz
+ UCHAR ucEngineClockHigh; //clockfrequency >> 16.
+ UCHAR vddcIndex; //2-bit vddc index;
+ USHORT tdpLimit;
+ //please initalize to 0
+ USHORT rsv1;
+ //please initialize to 0s
+ ULONG rsv2[2];
+}ATOM_PPLIB_SUMO_CLOCK_INFO;
+
+typedef struct _ATOM_PPLIB_STATE_V2
+{
+ //number of valid dpm levels in this state; Driver uses it to calculate the whole
+ //size of the state: sizeof(ATOM_PPLIB_STATE_V2) + (ucNumDPMLevels - 1) * sizeof(UCHAR)
+ UCHAR ucNumDPMLevels;
+
+ //a index to the array of nonClockInfos
+ UCHAR nonClockInfoIndex;
+ /**
+ * Driver will read the first ucNumDPMLevels in this array
+ */
+ UCHAR clockInfoIndex[1];
+} ATOM_PPLIB_STATE_V2;
+
+typedef struct _StateArray{
+ //how many states we have
+ UCHAR ucNumEntries;
+
+ ATOM_PPLIB_STATE_V2 states[1];
+}StateArray;
+
+
+typedef struct _ClockInfoArray{
+ //how many clock levels we have
+ UCHAR ucNumEntries;
+
+ //sizeof(ATOM_PPLIB_CLOCK_INFO)
+ UCHAR ucEntrySize;
+
+ UCHAR clockInfo[1];
+}ClockInfoArray;
+
+typedef struct _NonClockInfoArray{
+
+ //how many non-clock levels we have. normally should be same as number of states
+ UCHAR ucNumEntries;
+ //sizeof(ATOM_PPLIB_NONCLOCK_INFO)
+ UCHAR ucEntrySize;
+
+ ATOM_PPLIB_NONCLOCK_INFO nonClockInfo[1];
+}NonClockInfoArray;
+
+typedef struct _ATOM_PPLIB_Clock_Voltage_Dependency_Record
+{
+ USHORT usClockLow;
+ UCHAR ucClockHigh;
+ USHORT usVoltage;
+}ATOM_PPLIB_Clock_Voltage_Dependency_Record;
+
+typedef struct _ATOM_PPLIB_Clock_Voltage_Dependency_Table
+{
+ UCHAR ucNumEntries; // Number of entries.
+ ATOM_PPLIB_Clock_Voltage_Dependency_Record entries[1]; // Dynamically allocate entries.
+}ATOM_PPLIB_Clock_Voltage_Dependency_Table;
+
+typedef struct _ATOM_PPLIB_Clock_Voltage_Limit_Record
+{
+ USHORT usSclkLow;
+ UCHAR ucSclkHigh;
+ USHORT usMclkLow;
+ UCHAR ucMclkHigh;
+ USHORT usVddc;
+ USHORT usVddci;
+}ATOM_PPLIB_Clock_Voltage_Limit_Record;
+
+typedef struct _ATOM_PPLIB_Clock_Voltage_Limit_Table
+{
+ UCHAR ucNumEntries; // Number of entries.
+ ATOM_PPLIB_Clock_Voltage_Limit_Record entries[1]; // Dynamically allocate entries.
+}ATOM_PPLIB_Clock_Voltage_Limit_Table;
+
+union _ATOM_PPLIB_CAC_Leakage_Record
+{
+ struct
+ {
+ USHORT usVddc; // We use this field for the "fake" standardized VDDC for power calculations; For CI and newer, we use this as the real VDDC value. in CI we read it as StdVoltageHiSidd
+ ULONG ulLeakageValue; // For CI and newer we use this as the "fake" standar VDDC value. in CI we read it as StdVoltageLoSidd
+
+ };
+ struct
+ {
+ USHORT usVddc1;
+ USHORT usVddc2;
+ USHORT usVddc3;
+ };
+};
+
+typedef union _ATOM_PPLIB_CAC_Leakage_Record ATOM_PPLIB_CAC_Leakage_Record;
+
+typedef struct _ATOM_PPLIB_CAC_Leakage_Table
+{
+ UCHAR ucNumEntries; // Number of entries.
+ ATOM_PPLIB_CAC_Leakage_Record entries[1]; // Dynamically allocate entries.
+}ATOM_PPLIB_CAC_Leakage_Table;
+
+typedef struct _ATOM_PPLIB_PhaseSheddingLimits_Record
+{
+ USHORT usVoltage;
+ USHORT usSclkLow;
+ UCHAR ucSclkHigh;
+ USHORT usMclkLow;
+ UCHAR ucMclkHigh;
+}ATOM_PPLIB_PhaseSheddingLimits_Record;
+
+typedef struct _ATOM_PPLIB_PhaseSheddingLimits_Table
+{
+ UCHAR ucNumEntries; // Number of entries.
+ ATOM_PPLIB_PhaseSheddingLimits_Record entries[1]; // Dynamically allocate entries.
+}ATOM_PPLIB_PhaseSheddingLimits_Table;
+
+typedef struct _VCEClockInfo{
+ USHORT usEVClkLow;
+ UCHAR ucEVClkHigh;
+ USHORT usECClkLow;
+ UCHAR ucECClkHigh;
+}VCEClockInfo;
+
+typedef struct _VCEClockInfoArray{
+ UCHAR ucNumEntries;
+ VCEClockInfo entries[1];
+}VCEClockInfoArray;
+
+typedef struct _ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record
+{
+ USHORT usVoltage;
+ UCHAR ucVCEClockInfoIndex;
+}ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record;
+
+typedef struct _ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table
+{
+ UCHAR numEntries;
+ ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record entries[1];
+}ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table;
+
+typedef struct _ATOM_PPLIB_VCE_State_Record
+{
+ UCHAR ucVCEClockInfoIndex;
+ UCHAR ucClockInfoIndex; //highest 2 bits indicates memory p-states, lower 6bits indicates index to ClockInfoArrary
+}ATOM_PPLIB_VCE_State_Record;
+
+typedef struct _ATOM_PPLIB_VCE_State_Table
+{
+ UCHAR numEntries;
+ ATOM_PPLIB_VCE_State_Record entries[1];
+}ATOM_PPLIB_VCE_State_Table;
+
+
+typedef struct _ATOM_PPLIB_VCE_Table
+{
+ UCHAR revid;
+// VCEClockInfoArray array;
+// ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table limits;
+// ATOM_PPLIB_VCE_State_Table states;
+}ATOM_PPLIB_VCE_Table;
+
+
+typedef struct _UVDClockInfo{
+ USHORT usVClkLow;
+ UCHAR ucVClkHigh;
+ USHORT usDClkLow;
+ UCHAR ucDClkHigh;
+}UVDClockInfo;
+
+typedef struct _UVDClockInfoArray{
+ UCHAR ucNumEntries;
+ UVDClockInfo entries[1];
+}UVDClockInfoArray;
+
+typedef struct _ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record
+{
+ USHORT usVoltage;
+ UCHAR ucUVDClockInfoIndex;
+}ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record;
+
+typedef struct _ATOM_PPLIB_UVD_Clock_Voltage_Limit_Table
+{
+ UCHAR numEntries;
+ ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record entries[1];
+}ATOM_PPLIB_UVD_Clock_Voltage_Limit_Table;
+
+typedef struct _ATOM_PPLIB_UVD_Table
+{
+ UCHAR revid;
+// UVDClockInfoArray array;
+// ATOM_PPLIB_UVD_Clock_Voltage_Limit_Table limits;
+}ATOM_PPLIB_UVD_Table;
+
+typedef struct _ATOM_PPLIB_SAMClk_Voltage_Limit_Record
+{
+ USHORT usVoltage;
+ USHORT usSAMClockLow;
+ UCHAR ucSAMClockHigh;
+}ATOM_PPLIB_SAMClk_Voltage_Limit_Record;
+
+typedef struct _ATOM_PPLIB_SAMClk_Voltage_Limit_Table{
+ UCHAR numEntries;
+ ATOM_PPLIB_SAMClk_Voltage_Limit_Record entries[1];
+}ATOM_PPLIB_SAMClk_Voltage_Limit_Table;
+
+typedef struct _ATOM_PPLIB_SAMU_Table
+{
+ UCHAR revid;
+ ATOM_PPLIB_SAMClk_Voltage_Limit_Table limits;
+}ATOM_PPLIB_SAMU_Table;
+
+typedef struct _ATOM_PPLIB_ACPClk_Voltage_Limit_Record
+{
+ USHORT usVoltage;
+ USHORT usACPClockLow;
+ UCHAR ucACPClockHigh;
+}ATOM_PPLIB_ACPClk_Voltage_Limit_Record;
+
+typedef struct _ATOM_PPLIB_ACPClk_Voltage_Limit_Table{
+ UCHAR numEntries;
+ ATOM_PPLIB_ACPClk_Voltage_Limit_Record entries[1];
+}ATOM_PPLIB_ACPClk_Voltage_Limit_Table;
+
+typedef struct _ATOM_PPLIB_ACP_Table
+{
+ UCHAR revid;
+ ATOM_PPLIB_ACPClk_Voltage_Limit_Table limits;
+}ATOM_PPLIB_ACP_Table;
+
+typedef struct _ATOM_PowerTune_Table{
+ USHORT usTDP;
+ USHORT usConfigurableTDP;
+ USHORT usTDC;
+ USHORT usBatteryPowerLimit;
+ USHORT usSmallPowerLimit;
+ USHORT usLowCACLeakage;
+ USHORT usHighCACLeakage;
+}ATOM_PowerTune_Table;
+
+typedef struct _ATOM_PPLIB_POWERTUNE_Table
+{
+ UCHAR revid;
+ ATOM_PowerTune_Table power_tune_table;
+}ATOM_PPLIB_POWERTUNE_Table;
+
+typedef struct _ATOM_PPLIB_POWERTUNE_Table_V1
+{
+ UCHAR revid;
+ ATOM_PowerTune_Table power_tune_table;
+ USHORT usMaximumPowerDeliveryLimit;
+ USHORT usReserve[7];
+} ATOM_PPLIB_POWERTUNE_Table_V1;
+
+#define ATOM_PPM_A_A 1
+#define ATOM_PPM_A_I 2
+typedef struct _ATOM_PPLIB_PPM_Table
+{
+ UCHAR ucRevId;
+ UCHAR ucPpmDesign; //A+I or A+A
+ USHORT usCpuCoreNumber;
+ ULONG ulPlatformTDP;
+ ULONG ulSmallACPlatformTDP;
+ ULONG ulPlatformTDC;
+ ULONG ulSmallACPlatformTDC;
+ ULONG ulApuTDP;
+ ULONG ulDGpuTDP;
+ ULONG ulDGpuUlvPower;
+ ULONG ulTjmax;
+} ATOM_PPLIB_PPM_Table;
+
+#pragma pack(pop)
+
+#endif
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 20/53] drm/radeon: add structs to store uvd clock voltage deps
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (18 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 19/53] drm/radeon: switch to pptable.h Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 21/53] drm/radeon/cik: add rlc helpers for DPM Alex Deucher
` (33 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Used for uvd power management.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon.h | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 37b0fc4..00d154d 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1276,6 +1276,17 @@ struct radeon_phase_shedding_limits_table {
struct radeon_phase_shedding_limits_entry *entries;
};
+struct radeon_uvd_clock_voltage_dependency_entry {
+ u32 vclk;
+ u32 dclk;
+ u16 v;
+};
+
+struct radeon_uvd_clock_voltage_dependency_table {
+ u8 count;
+ struct radeon_uvd_clock_voltage_dependency_entry *entries;
+};
+
struct radeon_ppm_table {
u8 ppm_design;
u16 cpu_core_number;
@@ -1294,6 +1305,7 @@ struct radeon_dpm_dynamic_state {
struct radeon_clock_voltage_dependency_table vddci_dependency_on_mclk;
struct radeon_clock_voltage_dependency_table vddc_dependency_on_mclk;
struct radeon_clock_voltage_dependency_table vddc_dependency_on_dispclk;
+ struct radeon_uvd_clock_voltage_dependency_table uvd_clock_voltage_dependency_table;
struct radeon_clock_array valid_sclk_values;
struct radeon_clock_array valid_mclk_values;
struct radeon_clock_and_voltage_limits max_clock_voltage_on_dc;
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 21/53] drm/radeon/cik: add rlc helpers for DPM
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (19 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 20/53] drm/radeon: add structs to store uvd clock voltage deps Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 22/53] drm/radeon: add support for thermal controller on KB/KV Alex Deucher
` (32 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cik.c | 29 +++++++++++++++++++++++++++++
drivers/gpu/drm/radeon/cikd.h | 9 +++++++++
2 files changed, 38 insertions(+)
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index a36e98c..727c296 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -5587,6 +5587,35 @@ static u32 cik_halt_rlc(struct radeon_device *rdev)
return orig;
}
+void cik_enter_rlc_safe_mode(struct radeon_device *rdev)
+{
+ u32 tmp, i, mask;
+
+ tmp = REQ | MESSAGE(MSG_ENTER_RLC_SAFE_MODE);
+ WREG32(RLC_GPR_REG2, tmp);
+
+ mask = GFX_POWER_STATUS | GFX_CLOCK_STATUS;
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ if ((RREG32(RLC_GPM_STAT) & mask) == mask)
+ break;
+ udelay(1);
+ }
+
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ if ((RREG32(RLC_GPR_REG2) & REQ) == 0)
+ break;
+ udelay(1);
+ }
+}
+
+void cik_exit_rlc_safe_mode(struct radeon_device *rdev)
+{
+ u32 tmp;
+
+ tmp = REQ | MESSAGE(MSG_EXIT_RLC_SAFE_MODE);
+ WREG32(RLC_GPR_REG2, tmp);
+}
+
/**
* cik_rlc_stop - stop the RLC ME
*
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index 63955ab..116b313 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -952,6 +952,8 @@
#define RLC_GPM_STAT 0xC400
# define RLC_GPM_BUSY (1 << 0)
+# define GFX_POWER_STATUS (1 << 1)
+# define GFX_CLOCK_STATUS (1 << 2)
#define RLC_PG_CNTL 0xC40C
# define GFX_PG_ENABLE (1 << 0)
@@ -1004,6 +1006,13 @@
#define RLC_GPM_SCRATCH_ADDR 0xC4B0
#define RLC_GPM_SCRATCH_DATA 0xC4B4
+#define RLC_GPR_REG2 0xC4E8
+#define REQ 0x00000001
+#define MESSAGE(x) ((x) << 1)
+#define MESSAGE_MASK 0x0000001e
+#define MSG_ENTER_RLC_SAFE_MODE 1
+#define MSG_EXIT_RLC_SAFE_MODE 0
+
#define CP_HPD_EOP_BASE_ADDR 0xC904
#define CP_HPD_EOP_BASE_ADDR_HI 0xC908
#define CP_HPD_EOP_VMID 0xC90C
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 22/53] drm/radeon: add support for thermal controller on KB/KV
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (20 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 21/53] drm/radeon/cik: add rlc helpers for DPM Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 23/53] drm/radeon: add CI to r600_is_internal_thermal_sensor() Alex Deucher
` (31 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
No support for reading temperature back yet.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon.h | 1 +
drivers/gpu/drm/radeon/radeon_atombios.c | 5 +++++
2 files changed, 6 insertions(+)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 00d154d..876f423 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1132,6 +1132,7 @@ enum radeon_int_thermal_type {
THERMAL_TYPE_SI,
THERMAL_TYPE_EMC2103_WITH_INTERNAL,
THERMAL_TYPE_CI,
+ THERMAL_TYPE_KV,
};
struct radeon_voltage {
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index 27de73c..6247b5e 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -2237,6 +2237,11 @@ static void radeon_atombios_add_pplib_thermal_controller(struct radeon_device *r
(controller->ucFanParameters &
ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with");
rdev->pm.int_thermal_type = THERMAL_TYPE_CI;
+ } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_KAVERI) {
+ DRM_INFO("Internal thermal controller %s fan control\n",
+ (controller->ucFanParameters &
+ ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with");
+ rdev->pm.int_thermal_type = THERMAL_TYPE_KV;
} else if ((controller->ucType ==
ATOM_PP_THERMALCONTROLLER_EXTERNAL_GPIO) ||
(controller->ucType ==
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 23/53] drm/radeon: add CI to r600_is_internal_thermal_sensor()
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (21 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 22/53] drm/radeon: add support for thermal controller on KB/KV Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 24/53] drm/radeon: add KB/KV to r600_is_internal_thermal_sensor Alex Deucher
` (30 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index e5c860f..e246e3a 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -745,6 +745,7 @@ bool r600_is_internal_thermal_sensor(enum radeon_int_thermal_type sensor)
case THERMAL_TYPE_SUMO:
case THERMAL_TYPE_NI:
case THERMAL_TYPE_SI:
+ case THERMAL_TYPE_CI:
return true;
case THERMAL_TYPE_ADT7473_WITH_INTERNAL:
case THERMAL_TYPE_EMC2103_WITH_INTERNAL:
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 24/53] drm/radeon: add KB/KV to r600_is_internal_thermal_sensor
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (22 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 23/53] drm/radeon: add CI to r600_is_internal_thermal_sensor() Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 25/53] drm/radeon: add get_temperature() callbacks for CIK Alex Deucher
` (29 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index e246e3a..d54a838 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -746,6 +746,7 @@ bool r600_is_internal_thermal_sensor(enum radeon_int_thermal_type sensor)
case THERMAL_TYPE_NI:
case THERMAL_TYPE_SI:
case THERMAL_TYPE_CI:
+ case THERMAL_TYPE_KV:
return true;
case THERMAL_TYPE_ADT7473_WITH_INTERNAL:
case THERMAL_TYPE_EMC2103_WITH_INTERNAL:
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 25/53] drm/radeon: add get_temperature() callbacks for CIK
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (23 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 24/53] drm/radeon: add KB/KV to r600_is_internal_thermal_sensor Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 26/53] drm/radeon: adjust si_dpm function for code sharing Alex Deucher
` (28 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cik.c | 37 ++++++++++++++++++++++++++++++++++++
drivers/gpu/drm/radeon/cikd.h | 8 ++++++++
drivers/gpu/drm/radeon/radeon_asic.c | 2 ++
drivers/gpu/drm/radeon/radeon_asic.h | 2 ++
drivers/gpu/drm/radeon/radeon_pm.c | 2 ++
5 files changed, 51 insertions(+)
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 727c296..d0804f7 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -69,6 +69,43 @@ static void cik_program_aspm(struct radeon_device *rdev);
static void cik_init_pg(struct radeon_device *rdev);
static void cik_init_cg(struct radeon_device *rdev);
+/* get temperature in millidegrees */
+int ci_get_temp(struct radeon_device *rdev)
+{
+ u32 temp;
+ int actual_temp = 0;
+
+ temp = (RREG32_SMC(CG_MULT_THERMAL_STATUS) & CTF_TEMP_MASK) >>
+ CTF_TEMP_SHIFT;
+
+ if (temp & 0x200)
+ actual_temp = 255;
+ else
+ actual_temp = temp & 0x1ff;
+
+ actual_temp = actual_temp * 1000;
+
+ return actual_temp;
+}
+
+/* get temperature in millidegrees */
+int kv_get_temp(struct radeon_device *rdev)
+{
+ u32 temp;
+ int actual_temp = 0;
+
+ temp = RREG32_SMC(0xC0300E0C);
+
+ if (temp)
+ actual_temp = (temp / 8) - 49;
+ else
+ actual_temp = 0;
+
+ actual_temp = actual_temp * 1000;
+
+ return actual_temp;
+}
+
/*
* Indirect registers accessor
*/
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index 116b313..b66518a 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -32,6 +32,14 @@
#define GENERAL_PWRMGT 0xC0200000
# define GPU_COUNTER_CLK (1 << 15)
+#define CG_MULT_THERMAL_STATUS 0xC0300010
+#define ASIC_MAX_TEMP(x) ((x) << 0)
+#define ASIC_MAX_TEMP_MASK 0x000001ff
+#define ASIC_MAX_TEMP_SHIFT 0
+#define CTF_TEMP(x) ((x) << 9)
+#define CTF_TEMP_MASK 0x0003fe00
+#define CTF_TEMP_SHIFT 9
+
#define MPLL_BYPASSCLK_SEL 0xC050019C
# define MPLL_CLKOUT_SEL(x) ((x) << 8)
# define MPLL_CLKOUT_SEL_MASK 0xFF00
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 880551b..3a55540 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2452,6 +2452,7 @@ static struct radeon_asic ci_asic = {
.set_pcie_lanes = NULL,
.set_clock_gating = NULL,
.set_uvd_clocks = &cik_set_uvd_clocks,
+ .get_temperature = &ci_get_temp,
},
.pflip = {
.pre_page_flip = &evergreen_pre_page_flip,
@@ -2607,6 +2608,7 @@ static struct radeon_asic kv_asic = {
.set_pcie_lanes = NULL,
.set_clock_gating = NULL,
.set_uvd_clocks = &cik_set_uvd_clocks,
+ .get_temperature = &kv_get_temp,
},
.pflip = {
.pre_page_flip = &evergreen_pre_page_flip,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 3cf7d89..d5c6c5b 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -747,5 +747,7 @@ u32 cik_compute_ring_get_wptr(struct radeon_device *rdev,
struct radeon_ring *ring);
void cik_compute_ring_set_wptr(struct radeon_device *rdev,
struct radeon_ring *ring);
+int ci_get_temp(struct radeon_device *rdev);
+int kv_get_temp(struct radeon_device *rdev);
#endif
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 79a03de..1408014 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -569,6 +569,8 @@ static int radeon_hwmon_init(struct radeon_device *rdev)
case THERMAL_TYPE_NI:
case THERMAL_TYPE_SUMO:
case THERMAL_TYPE_SI:
+ case THERMAL_TYPE_CI:
+ case THERMAL_TYPE_KV:
if (rdev->asic->pm.get_temperature == NULL)
return err;
rdev->pm.int_hwmon_dev = hwmon_device_register(rdev->dev);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 26/53] drm/radeon: adjust si_dpm function for code sharing
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (24 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 25/53] drm/radeon: add get_temperature() callbacks for CIK Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 27/53] drm/radeon/dpm: update cac leakage table parsing for CI Alex Deucher
` (27 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/si_dpm.c | 21 ++++++++++++++-------
1 file changed, 14 insertions(+), 7 deletions(-)
diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
index 88699e3..0f8be48 100644
--- a/drivers/gpu/drm/radeon/si_dpm.c
+++ b/drivers/gpu/drm/radeon/si_dpm.c
@@ -3759,19 +3759,20 @@ static bool si_validate_phase_shedding_tables(struct radeon_device *rdev,
}
static void si_trim_voltage_table_to_fit_state_table(struct radeon_device *rdev,
+ u32 max_voltage_steps,
struct atom_voltage_table *voltage_table)
{
unsigned int i, diff;
- if (voltage_table->count <= SISLANDS_MAX_NO_VREG_STEPS)
+ if (voltage_table->count <= max_voltage_steps)
return;
- diff = voltage_table->count - SISLANDS_MAX_NO_VREG_STEPS;
+ diff = voltage_table->count - max_voltage_steps;
- for (i= 0; i < SISLANDS_MAX_NO_VREG_STEPS; i++)
+ for (i= 0; i < max_voltage_steps; i++)
voltage_table->entries[i] = voltage_table->entries[i + diff];
- voltage_table->count = SISLANDS_MAX_NO_VREG_STEPS;
+ voltage_table->count = max_voltage_steps;
}
static int si_construct_voltage_tables(struct radeon_device *rdev)
@@ -3787,7 +3788,9 @@ static int si_construct_voltage_tables(struct radeon_device *rdev)
return ret;
if (eg_pi->vddc_voltage_table.count > SISLANDS_MAX_NO_VREG_STEPS)
- si_trim_voltage_table_to_fit_state_table(rdev, &eg_pi->vddc_voltage_table);
+ si_trim_voltage_table_to_fit_state_table(rdev,
+ SISLANDS_MAX_NO_VREG_STEPS,
+ &eg_pi->vddc_voltage_table);
if (eg_pi->vddci_control) {
ret = radeon_atom_get_voltage_table(rdev, VOLTAGE_TYPE_VDDCI,
@@ -3796,7 +3799,9 @@ static int si_construct_voltage_tables(struct radeon_device *rdev)
return ret;
if (eg_pi->vddci_voltage_table.count > SISLANDS_MAX_NO_VREG_STEPS)
- si_trim_voltage_table_to_fit_state_table(rdev, &eg_pi->vddci_voltage_table);
+ si_trim_voltage_table_to_fit_state_table(rdev,
+ SISLANDS_MAX_NO_VREG_STEPS,
+ &eg_pi->vddci_voltage_table);
}
if (pi->mvdd_control) {
@@ -3814,7 +3819,9 @@ static int si_construct_voltage_tables(struct radeon_device *rdev)
}
if (si_pi->mvdd_voltage_table.count > SISLANDS_MAX_NO_VREG_STEPS)
- si_trim_voltage_table_to_fit_state_table(rdev, &si_pi->mvdd_voltage_table);
+ si_trim_voltage_table_to_fit_state_table(rdev,
+ SISLANDS_MAX_NO_VREG_STEPS,
+ &si_pi->mvdd_voltage_table);
}
if (si_pi->vddc_phase_shed_control) {
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 27/53] drm/radeon/dpm: update cac leakage table parsing for CI
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (25 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 26/53] drm/radeon: adjust si_dpm function for code sharing Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 28/53] drm/radeon/dpm: add support for parsing the atom powertune table Alex Deucher
` (26 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Uses a different table format if the board supports EVV.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 17 +++++++++++++----
drivers/gpu/drm/radeon/radeon.h | 15 +++++++++++----
2 files changed, 24 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index d54a838..ccdf770 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -956,10 +956,19 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
return -ENOMEM;
}
for (i = 0; i < cac_table->ucNumEntries; i++) {
- rdev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc =
- le16_to_cpu(cac_table->entries[i].usVddc);
- rdev->pm.dpm.dyn_state.cac_leakage_table.entries[i].leakage =
- le32_to_cpu(cac_table->entries[i].ulLeakageValue);
+ if (rdev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_EVV) {
+ rdev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc1 =
+ le16_to_cpu(cac_table->entries[i].usVddc1);
+ rdev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc2 =
+ le16_to_cpu(cac_table->entries[i].usVddc2);
+ rdev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc3 =
+ le16_to_cpu(cac_table->entries[i].usVddc3);
+ } else {
+ rdev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc =
+ le16_to_cpu(cac_table->entries[i].usVddc);
+ rdev->pm.dpm.dyn_state.cac_leakage_table.entries[i].leakage =
+ le32_to_cpu(cac_table->entries[i].ulLeakageValue);
+ }
}
rdev->pm.dpm.dyn_state.cac_leakage_table.count = cac_table->ucNumEntries;
}
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 876f423..e6541e3 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1256,14 +1256,21 @@ struct radeon_clock_voltage_dependency_table {
struct radeon_clock_voltage_dependency_entry *entries;
};
-struct radeon_cac_leakage_entry {
- u16 vddc;
- u32 leakage;
+union radeon_cac_leakage_entry {
+ struct {
+ u16 vddc;
+ u32 leakage;
+ };
+ struct {
+ u16 vddc1;
+ u16 vddc2;
+ u16 vddc3;
+ };
};
struct radeon_cac_leakage_table {
u32 count;
- struct radeon_cac_leakage_entry *entries;
+ union radeon_cac_leakage_entry *entries;
};
struct radeon_phase_shedding_limits_entry {
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 28/53] drm/radeon/dpm: add support for parsing the atom powertune table
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (26 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 27/53] drm/radeon/dpm: update cac leakage table parsing for CI Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 29/53] drm/radeon/dpm: grab mvdd_dependency_on_mclk info from vbios Alex Deucher
` (25 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Needed for DPM on CI.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 44 +++++++++++++++++++++++++++++++++++++++
drivers/gpu/drm/radeon/radeon.h | 12 +++++++++++
2 files changed, 56 insertions(+)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index ccdf770..e6905f0 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -1014,6 +1014,48 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
rdev->pm.dpm.dyn_state.ppm_table->tj_max =
le32_to_cpu(ppm->ulTjmax);
}
+ if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V7) &&
+ ext_hdr->usPowerTuneTableOffset) {
+ u8 rev = *(u8 *)(mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(ext_hdr->usPowerTuneTableOffset));
+ ATOM_PowerTune_Table *pt;
+ rdev->pm.dpm.dyn_state.cac_tdp_table =
+ kzalloc(sizeof(struct radeon_cac_tdp_table), GFP_KERNEL);
+ if (!rdev->pm.dpm.dyn_state.cac_tdp_table) {
+ kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
+ kfree(rdev->pm.dpm.dyn_state.ppm_table);
+ return -ENOMEM;
+ }
+ if (rev > 0) {
+ ATOM_PPLIB_POWERTUNE_Table_V1 *ppt = (ATOM_PPLIB_POWERTUNE_Table_V1 *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(ext_hdr->usPowerTuneTableOffset));
+ rdev->pm.dpm.dyn_state.cac_tdp_table->maximum_power_delivery_limit =
+ ppt->usMaximumPowerDeliveryLimit;
+ pt = &ppt->power_tune_table;
+ } else {
+ ATOM_PPLIB_POWERTUNE_Table *ppt = (ATOM_PPLIB_POWERTUNE_Table *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(ext_hdr->usPowerTuneTableOffset));
+ rdev->pm.dpm.dyn_state.cac_tdp_table->maximum_power_delivery_limit = 255;
+ pt = &ppt->power_tune_table;
+ }
+ rdev->pm.dpm.dyn_state.cac_tdp_table->tdp = le16_to_cpu(pt->usTDP);
+ rdev->pm.dpm.dyn_state.cac_tdp_table->configurable_tdp =
+ le16_to_cpu(pt->usConfigurableTDP);
+ rdev->pm.dpm.dyn_state.cac_tdp_table->tdc = le16_to_cpu(pt->usTDC);
+ rdev->pm.dpm.dyn_state.cac_tdp_table->battery_power_limit =
+ le16_to_cpu(pt->usBatteryPowerLimit);
+ rdev->pm.dpm.dyn_state.cac_tdp_table->small_power_limit =
+ le16_to_cpu(pt->usSmallPowerLimit);
+ rdev->pm.dpm.dyn_state.cac_tdp_table->low_cac_leakage =
+ le16_to_cpu(pt->usLowCACLeakage);
+ rdev->pm.dpm.dyn_state.cac_tdp_table->high_cac_leakage =
+ le16_to_cpu(pt->usHighCACLeakage);
+ }
}
return 0;
@@ -1033,6 +1075,8 @@ void r600_free_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.phase_shedding_limits_table.entries);
if (rdev->pm.dpm.dyn_state.ppm_table)
kfree(rdev->pm.dpm.dyn_state.ppm_table);
+ if (rdev->pm.dpm.dyn_state.cac_tdp_table)
+ kfree(rdev->pm.dpm.dyn_state.cac_tdp_table);
}
enum radeon_pcie_gen r600_get_pcie_gen_support(struct radeon_device *rdev,
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index e6541e3..286e34a 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1308,6 +1308,17 @@ struct radeon_ppm_table {
u32 tj_max;
};
+struct radeon_cac_tdp_table {
+ u16 tdp;
+ u16 configurable_tdp;
+ u16 tdc;
+ u16 battery_power_limit;
+ u16 small_power_limit;
+ u16 low_cac_leakage;
+ u16 high_cac_leakage;
+ u16 maximum_power_delivery_limit;
+};
+
struct radeon_dpm_dynamic_state {
struct radeon_clock_voltage_dependency_table vddc_dependency_on_sclk;
struct radeon_clock_voltage_dependency_table vddci_dependency_on_mclk;
@@ -1325,6 +1336,7 @@ struct radeon_dpm_dynamic_state {
struct radeon_cac_leakage_table cac_leakage_table;
struct radeon_phase_shedding_limits_table phase_shedding_limits_table;
struct radeon_ppm_table *ppm_table;
+ struct radeon_cac_tdp_table *cac_tdp_table;
};
struct radeon_dpm_fan {
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 29/53] drm/radeon/dpm: grab mvdd_dependency_on_mclk info from vbios
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (27 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 28/53] drm/radeon/dpm: add support for parsing the atom powertune table Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 30/53] drm/radeon: add structs to store vce clock voltage deps Alex Deucher
` (24 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Required for dpm on CI.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 19 +++++++++++++++++++
drivers/gpu/drm/radeon/radeon.h | 1 +
2 files changed, 20 insertions(+)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index e6905f0..9dda735 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -877,6 +877,19 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
return ret;
}
}
+ if (power_info->pplib4.usMvddDependencyOnMCLKOffset) {
+ dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(power_info->pplib4.usMvddDependencyOnMCLKOffset));
+ ret = r600_parse_clk_voltage_dep_table(&rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk,
+ dep_table);
+ if (ret) {
+ kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
+ return ret;
+ }
+ }
if (power_info->pplib4.usMaxClockVoltageOnDCOffset) {
ATOM_PPLIB_Clock_Voltage_Limit_Table *clk_v =
(ATOM_PPLIB_Clock_Voltage_Limit_Table *)
@@ -909,6 +922,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
return -ENOMEM;
}
@@ -953,6 +967,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
return -ENOMEM;
}
for (i = 0; i < cac_table->ucNumEntries; i++) {
@@ -991,6 +1006,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
return -ENOMEM;
}
@@ -1025,6 +1041,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
kfree(rdev->pm.dpm.dyn_state.ppm_table);
return -ENOMEM;
@@ -1069,6 +1086,8 @@ void r600_free_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
if (rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries)
kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
+ if (rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries)
+ kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
if (rdev->pm.dpm.dyn_state.cac_leakage_table.entries)
kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
if (rdev->pm.dpm.dyn_state.phase_shedding_limits_table.entries)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 286e34a..e65304a 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1323,6 +1323,7 @@ struct radeon_dpm_dynamic_state {
struct radeon_clock_voltage_dependency_table vddc_dependency_on_sclk;
struct radeon_clock_voltage_dependency_table vddci_dependency_on_mclk;
struct radeon_clock_voltage_dependency_table vddc_dependency_on_mclk;
+ struct radeon_clock_voltage_dependency_table mvdd_dependency_on_mclk;
struct radeon_clock_voltage_dependency_table vddc_dependency_on_dispclk;
struct radeon_uvd_clock_voltage_dependency_table uvd_clock_voltage_dependency_table;
struct radeon_clock_array valid_sclk_values;
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 30/53] drm/radeon: add structs to store vce clock voltage deps
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (28 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 29/53] drm/radeon/dpm: grab mvdd_dependency_on_mclk info from vbios Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 31/53] drm/radeon: add clock voltage dep tables for acp, samu Alex Deucher
` (23 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Used for vce power management.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon.h | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index e65304a..d21fa20 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1295,6 +1295,17 @@ struct radeon_uvd_clock_voltage_dependency_table {
struct radeon_uvd_clock_voltage_dependency_entry *entries;
};
+struct radeon_vce_clock_voltage_dependency_entry {
+ u32 ecclk;
+ u32 evclk;
+ u16 v;
+};
+
+struct radeon_vce_clock_voltage_dependency_table {
+ u8 count;
+ struct radeon_vce_clock_voltage_dependency_entry *entries;
+};
+
struct radeon_ppm_table {
u8 ppm_design;
u16 cpu_core_number;
@@ -1326,6 +1337,7 @@ struct radeon_dpm_dynamic_state {
struct radeon_clock_voltage_dependency_table mvdd_dependency_on_mclk;
struct radeon_clock_voltage_dependency_table vddc_dependency_on_dispclk;
struct radeon_uvd_clock_voltage_dependency_table uvd_clock_voltage_dependency_table;
+ struct radeon_vce_clock_voltage_dependency_table vce_clock_voltage_dependency_table;
struct radeon_clock_array valid_sclk_values;
struct radeon_clock_array valid_mclk_values;
struct radeon_clock_and_voltage_limits max_clock_voltage_on_dc;
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 31/53] drm/radeon: add clock voltage dep tables for acp, samu
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (29 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 30/53] drm/radeon: add structs to store vce clock voltage deps Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 32/53] drm/radeon: parse the vce clock voltage deps table Alex Deucher
` (22 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index d21fa20..661e7c1 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1338,6 +1338,8 @@ struct radeon_dpm_dynamic_state {
struct radeon_clock_voltage_dependency_table vddc_dependency_on_dispclk;
struct radeon_uvd_clock_voltage_dependency_table uvd_clock_voltage_dependency_table;
struct radeon_vce_clock_voltage_dependency_table vce_clock_voltage_dependency_table;
+ struct radeon_clock_voltage_dependency_table samu_clock_voltage_dependency_table;
+ struct radeon_clock_voltage_dependency_table acp_clock_voltage_dependency_table;
struct radeon_clock_array valid_sclk_values;
struct radeon_clock_array valid_mclk_values;
struct radeon_clock_and_voltage_limits max_clock_voltage_on_dc;
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 32/53] drm/radeon: parse the vce clock voltage deps table
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (30 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 31/53] drm/radeon: add clock voltage dep tables for acp, samu Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 33/53] drm/radeon: parse the uvd " Alex Deucher
` (21 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 41 ++++++++++++++++++++++++++++++++++++++-
1 file changed, 40 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index 9dda735..b49b0f0 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -989,12 +989,47 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
}
}
- /* ppm table */
+ /* ext tables */
if (le16_to_cpu(power_info->pplib.usTableSize) >=
sizeof(struct _ATOM_PPLIB_POWERPLAYTABLE3)) {
ATOM_PPLIB_EXTENDEDHEADER *ext_hdr = (ATOM_PPLIB_EXTENDEDHEADER *)
(mode_info->atom_context->bios + data_offset +
le16_to_cpu(power_info->pplib3.usExtendendedHeaderOffset));
+ if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V2) &&
+ ext_hdr->usVCETableOffset) {
+ VCEClockInfoArray *array = (VCEClockInfoArray *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(ext_hdr->usVCETableOffset) + 1);
+ ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table *limits =
+ (ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(ext_hdr->usVCETableOffset) + 1 +
+ 1 + array->ucNumEntries * sizeof(VCEClockInfo));
+ u32 size = limits->numEntries *
+ sizeof(struct radeon_vce_clock_voltage_dependency_entry);
+ rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries =
+ kzalloc(size, GFP_KERNEL);
+ if (!rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries) {
+ kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
+ return -ENOMEM;
+ }
+ rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.count =
+ limits->numEntries;
+ for (i = 0; i < limits->numEntries; i++) {
+ VCEClockInfo *vce_clk =
+ &array->entries[limits->entries[i].ucVCEClockInfoIndex];
+ rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[i].evclk =
+ le16_to_cpu(vce_clk->usEVClkLow) | (vce_clk->ucEVClkHigh << 16);
+ rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[i].ecclk =
+ le16_to_cpu(vce_clk->usECClkLow) | (vce_clk->ucECClkHigh << 16);
+ rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[i].v =
+ le16_to_cpu(limits->entries[i].usVoltage);
+ }
+ }
if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V5) &&
ext_hdr->usPPMTableOffset) {
ATOM_PPLIB_PPM_Table *ppm = (ATOM_PPLIB_PPM_Table *)
@@ -1008,6 +1043,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
+ kfree(rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries);
return -ENOMEM;
}
rdev->pm.dpm.dyn_state.ppm_table->ppm_design = ppm->ucPpmDesign;
@@ -1044,6 +1080,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
kfree(rdev->pm.dpm.dyn_state.ppm_table);
+ kfree(rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries);
return -ENOMEM;
}
if (rev > 0) {
@@ -1096,6 +1133,8 @@ void r600_free_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.ppm_table);
if (rdev->pm.dpm.dyn_state.cac_tdp_table)
kfree(rdev->pm.dpm.dyn_state.cac_tdp_table);
+ if (rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries)
+ kfree(rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries);
}
enum radeon_pcie_gen r600_get_pcie_gen_support(struct radeon_device *rdev,
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 33/53] drm/radeon: parse the uvd clock voltage deps table
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (31 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 32/53] drm/radeon: parse the vce clock voltage deps table Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 34/53] drm/radeon/dpm: clean up the extended table error pathes Alex Deucher
` (20 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 40 +++++++++++++++++++++++++++++++++++++++
1 file changed, 40 insertions(+)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index b49b0f0..c103d3f 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -1030,6 +1030,42 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
le16_to_cpu(limits->entries[i].usVoltage);
}
}
+ if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V3) &&
+ ext_hdr->usUVDTableOffset) {
+ UVDClockInfoArray *array = (UVDClockInfoArray *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(ext_hdr->usUVDTableOffset) + 1);
+ ATOM_PPLIB_UVD_Clock_Voltage_Limit_Table *limits =
+ (ATOM_PPLIB_UVD_Clock_Voltage_Limit_Table *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(ext_hdr->usUVDTableOffset) + 1 +
+ 1 + (array->ucNumEntries * sizeof (UVDClockInfo)));
+ u32 size = limits->numEntries *
+ sizeof(struct radeon_uvd_clock_voltage_dependency_entry);
+ rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries =
+ kzalloc(size, GFP_KERNEL);
+ if (!rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries) {
+ kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
+ kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
+ kfree(rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries);
+ return -ENOMEM;
+ }
+ rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count =
+ limits->numEntries;
+ for (i = 0; i < limits->numEntries; i++) {
+ UVDClockInfo *uvd_clk =
+ &array->entries[limits->entries[i].ucUVDClockInfoIndex];
+ rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].vclk =
+ le16_to_cpu(uvd_clk->usVClkLow) | (uvd_clk->ucVClkHigh << 16);
+ rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].dclk =
+ le16_to_cpu(uvd_clk->usDClkLow) | (uvd_clk->ucDClkHigh << 16);
+ rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].v =
+ le16_to_cpu(limits->entries[i].usVoltage);
+ }
+ }
if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V5) &&
ext_hdr->usPPMTableOffset) {
ATOM_PPLIB_PPM_Table *ppm = (ATOM_PPLIB_PPM_Table *)
@@ -1044,6 +1080,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
kfree(rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries);
+ kfree(rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries);
return -ENOMEM;
}
rdev->pm.dpm.dyn_state.ppm_table->ppm_design = ppm->ucPpmDesign;
@@ -1081,6 +1118,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
kfree(rdev->pm.dpm.dyn_state.ppm_table);
kfree(rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries);
+ kfree(rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries);
return -ENOMEM;
}
if (rev > 0) {
@@ -1135,6 +1173,8 @@ void r600_free_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.cac_tdp_table);
if (rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries)
kfree(rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries);
+ if (rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries)
+ kfree(rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries);
}
enum radeon_pcie_gen r600_get_pcie_gen_support(struct radeon_device *rdev,
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 34/53] drm/radeon/dpm: clean up the extended table error pathes
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (32 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 33/53] drm/radeon: parse the uvd " Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 35/53] drm/radeon: parse the samu clock voltage deps table Alex Deucher
` (19 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 40 ++++++---------------------------------
1 file changed, 6 insertions(+), 34 deletions(-)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index c103d3f..98db6ea 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -919,10 +919,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
sizeof(struct radeon_phase_shedding_limits_entry),
GFP_KERNEL);
if (!rdev->pm.dpm.dyn_state.phase_shedding_limits_table.entries) {
- kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
- kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
+ r600_free_extended_power_table(rdev);
return -ENOMEM;
}
@@ -964,10 +961,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
u32 size = cac_table->ucNumEntries * sizeof(struct radeon_cac_leakage_table);
rdev->pm.dpm.dyn_state.cac_leakage_table.entries = kzalloc(size, GFP_KERNEL);
if (!rdev->pm.dpm.dyn_state.cac_leakage_table.entries) {
- kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
- kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
+ r600_free_extended_power_table(rdev);
return -ENOMEM;
}
for (i = 0; i < cac_table->ucNumEntries; i++) {
@@ -1010,11 +1004,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries =
kzalloc(size, GFP_KERNEL);
if (!rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries) {
- kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
- kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
+ r600_free_extended_power_table(rdev);
return -ENOMEM;
}
rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.count =
@@ -1045,12 +1035,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries =
kzalloc(size, GFP_KERNEL);
if (!rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries) {
- kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
- kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
- kfree(rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries);
+ r600_free_extended_power_table(rdev);
return -ENOMEM;
}
rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count =
@@ -1074,13 +1059,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
rdev->pm.dpm.dyn_state.ppm_table =
kzalloc(sizeof(struct radeon_ppm_table), GFP_KERNEL);
if (!rdev->pm.dpm.dyn_state.ppm_table) {
- kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
- kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
- kfree(rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries);
- kfree(rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries);
+ r600_free_extended_power_table(rdev);
return -ENOMEM;
}
rdev->pm.dpm.dyn_state.ppm_table->ppm_design = ppm->ucPpmDesign;
@@ -1111,14 +1090,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
rdev->pm.dpm.dyn_state.cac_tdp_table =
kzalloc(sizeof(struct radeon_cac_tdp_table), GFP_KERNEL);
if (!rdev->pm.dpm.dyn_state.cac_tdp_table) {
- kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries);
- kfree(rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries);
- kfree(rdev->pm.dpm.dyn_state.cac_leakage_table.entries);
- kfree(rdev->pm.dpm.dyn_state.ppm_table);
- kfree(rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries);
- kfree(rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries);
+ r600_free_extended_power_table(rdev);
return -ENOMEM;
}
if (rev > 0) {
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 35/53] drm/radeon: parse the samu clock voltage deps table
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (33 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 34/53] drm/radeon/dpm: clean up the extended table error pathes Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 36/53] drm/radeon: parse the acp " Alex Deucher
` (18 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 26 ++++++++++++++++++++++++++
1 file changed, 26 insertions(+)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index 98db6ea..314886a 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -1051,6 +1051,30 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
le16_to_cpu(limits->entries[i].usVoltage);
}
}
+ if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V4) &&
+ ext_hdr->usSAMUTableOffset) {
+ ATOM_PPLIB_SAMClk_Voltage_Limit_Table *limits =
+ (ATOM_PPLIB_SAMClk_Voltage_Limit_Table *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(ext_hdr->usSAMUTableOffset) + 1);
+ u32 size = limits->numEntries *
+ sizeof(struct radeon_clock_voltage_dependency_entry);
+ rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries =
+ kzalloc(size, GFP_KERNEL);
+ if (!rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries) {
+ r600_free_extended_power_table(rdev);
+ return -ENOMEM;
+ }
+ rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.count =
+ limits->numEntries;
+ for (i = 0; i < limits->numEntries; i++) {
+ rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries[i].clk =
+ le16_to_cpu(limits->entries[i].usSAMClockLow) |
+ (limits->entries[i].ucSAMClockHigh << 16);
+ rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries[i].v =
+ le16_to_cpu(limits->entries[i].usVoltage);
+ }
+ }
if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V5) &&
ext_hdr->usPPMTableOffset) {
ATOM_PPLIB_PPM_Table *ppm = (ATOM_PPLIB_PPM_Table *)
@@ -1147,6 +1171,8 @@ void r600_free_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries);
if (rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries)
kfree(rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries);
+ if (rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries)
+ kfree(rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries);
}
enum radeon_pcie_gen r600_get_pcie_gen_support(struct radeon_device *rdev,
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 36/53] drm/radeon: parse the acp clock voltage deps table
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (34 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 35/53] drm/radeon: parse the samu clock voltage deps table Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 37/53] drm/radeon: add r600_get_pcie_lane_support helper Alex Deucher
` (17 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 26 ++++++++++++++++++++++++++
1 file changed, 26 insertions(+)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index 314886a..bf851ac 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -1106,6 +1106,30 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
rdev->pm.dpm.dyn_state.ppm_table->tj_max =
le32_to_cpu(ppm->ulTjmax);
}
+ if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V6) &&
+ ext_hdr->usACPTableOffset) {
+ ATOM_PPLIB_ACPClk_Voltage_Limit_Table *limits =
+ (ATOM_PPLIB_ACPClk_Voltage_Limit_Table *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(ext_hdr->usACPTableOffset) + 1);
+ u32 size = limits->numEntries *
+ sizeof(struct radeon_clock_voltage_dependency_entry);
+ rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries =
+ kzalloc(size, GFP_KERNEL);
+ if (!rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries) {
+ r600_free_extended_power_table(rdev);
+ return -ENOMEM;
+ }
+ rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.count =
+ limits->numEntries;
+ for (i = 0; i < limits->numEntries; i++) {
+ rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries[i].clk =
+ le16_to_cpu(limits->entries[i].usACPClockLow) |
+ (limits->entries[i].ucACPClockHigh << 16);
+ rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries[i].v =
+ le16_to_cpu(limits->entries[i].usVoltage);
+ }
+ }
if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V7) &&
ext_hdr->usPowerTuneTableOffset) {
u8 rev = *(u8 *)(mode_info->atom_context->bios + data_offset +
@@ -1173,6 +1197,8 @@ void r600_free_extended_power_table(struct radeon_device *rdev)
kfree(rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries);
if (rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries)
kfree(rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries);
+ if (rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries)
+ kfree(rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries);
}
enum radeon_pcie_gen r600_get_pcie_gen_support(struct radeon_device *rdev,
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 37/53] drm/radeon: add r600_get_pcie_lane_support helper
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (35 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 36/53] drm/radeon: parse the acp " Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 38/53] drm/radeon/dpm: add vce clocks to radeon_ps Alex Deucher
` (16 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 23 +++++++++++++++++++++++
drivers/gpu/drm/radeon/r600_dpm.h | 4 ++++
2 files changed, 27 insertions(+)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index bf851ac..34ea5d6 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -1223,3 +1223,26 @@ enum radeon_pcie_gen r600_get_pcie_gen_support(struct radeon_device *rdev,
}
return RADEON_PCIE_GEN1;
}
+
+u16 r600_get_pcie_lane_support(struct radeon_device *rdev,
+ u16 asic_lanes,
+ u16 default_lanes)
+{
+ switch (asic_lanes) {
+ case 0:
+ default:
+ return default_lanes;
+ case 1:
+ return 1;
+ case 2:
+ return 2;
+ case 4:
+ return 4;
+ case 8:
+ return 8;
+ case 12:
+ return 12;
+ case 16:
+ return 16;
+ }
+}
diff --git a/drivers/gpu/drm/radeon/r600_dpm.h b/drivers/gpu/drm/radeon/r600_dpm.h
index 7c822d9..71d5d93 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.h
+++ b/drivers/gpu/drm/radeon/r600_dpm.h
@@ -224,4 +224,8 @@ enum radeon_pcie_gen r600_get_pcie_gen_support(struct radeon_device *rdev,
enum radeon_pcie_gen asic_gen,
enum radeon_pcie_gen default_gen);
+u16 r600_get_pcie_lane_support(struct radeon_device *rdev,
+ u16 asic_lanes,
+ u16 default_lanes);
+
#endif
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 38/53] drm/radeon/dpm: add vce clocks to radeon_ps
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (36 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 37/53] drm/radeon: add r600_get_pcie_lane_support helper Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 39/53] drm/radeon/dpm: add a helper to encode pcie lane setting Alex Deucher
` (15 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon.h | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 661e7c1..13e402c 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1206,6 +1206,9 @@ struct radeon_ps {
/* UVD clocks */
u32 vclk;
u32 dclk;
+ /* VCE clocks */
+ u32 evclk;
+ u32 ecclk;
/* asic priv */
void *ps_priv;
};
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 39/53] drm/radeon/dpm: add a helper to encode pcie lane setting
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (37 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 38/53] drm/radeon/dpm: add vce clocks to radeon_ps Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 40/53] drm/radeon/dpm: add helper to fetch the vrefresh of the current mode Alex Deucher
` (14 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
convert from number of lanes to register setting.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 10 ++++++++++
drivers/gpu/drm/radeon/r600_dpm.h | 1 +
2 files changed, 11 insertions(+)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index 34ea5d6..89c46c5 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -1246,3 +1246,13 @@ u16 r600_get_pcie_lane_support(struct radeon_device *rdev,
return 16;
}
}
+
+u8 r600_encode_pci_lane_width(u32 lanes)
+{
+ u8 encoded_lanes[] = { 0, 1, 2, 0, 3, 0, 0, 0, 4, 0, 0, 0, 5, 0, 0, 0, 6 };
+
+ if (lanes > 16)
+ return 0;
+
+ return encoded_lanes[lanes];
+}
diff --git a/drivers/gpu/drm/radeon/r600_dpm.h b/drivers/gpu/drm/radeon/r600_dpm.h
index 71d5d93..8dc1fbd 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.h
+++ b/drivers/gpu/drm/radeon/r600_dpm.h
@@ -227,5 +227,6 @@ enum radeon_pcie_gen r600_get_pcie_gen_support(struct radeon_device *rdev,
u16 r600_get_pcie_lane_support(struct radeon_device *rdev,
u16 asic_lanes,
u16 default_lanes);
+u8 r600_encode_pci_lane_width(u32 lanes);
#endif
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 40/53] drm/radeon/dpm: add helper to fetch the vrefresh of the current mode
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (38 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 39/53] drm/radeon/dpm: add a helper to encode pcie lane setting Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 41/53] drm/radeon/kms: add dpm support for KB/KV Alex Deucher
` (13 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Needed for DPM on CI.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/r600_dpm.c | 18 ++++++++++++++++++
drivers/gpu/drm/radeon/r600_dpm.h | 1 +
2 files changed, 19 insertions(+)
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index 89c46c5..26a7878 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -174,6 +174,24 @@ u32 r600_dpm_get_vblank_time(struct radeon_device *rdev)
return vblank_time_us;
}
+u32 r600_dpm_get_vrefresh(struct radeon_device *rdev)
+{
+ struct drm_device *dev = rdev->ddev;
+ struct drm_crtc *crtc;
+ struct radeon_crtc *radeon_crtc;
+ u32 vrefresh = 0;
+
+ list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+ radeon_crtc = to_radeon_crtc(crtc);
+ if (crtc->enabled && radeon_crtc->enabled && radeon_crtc->hw_mode.clock) {
+ vrefresh = radeon_crtc->hw_mode.vrefresh;
+ break;
+ }
+ }
+
+ return vrefresh;
+}
+
void r600_calculate_u_and_p(u32 i, u32 r_c, u32 p_b,
u32 *p, u32 *u)
{
diff --git a/drivers/gpu/drm/radeon/r600_dpm.h b/drivers/gpu/drm/radeon/r600_dpm.h
index 8dc1fbd..1000bf9 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.h
+++ b/drivers/gpu/drm/radeon/r600_dpm.h
@@ -130,6 +130,7 @@ void r600_dpm_print_cap_info(u32 caps);
void r600_dpm_print_ps_status(struct radeon_device *rdev,
struct radeon_ps *rps);
u32 r600_dpm_get_vblank_time(struct radeon_device *rdev);
+u32 r600_dpm_get_vrefresh(struct radeon_device *rdev);
bool r600_is_uvd_state(u32 class, u32 class2);
void r600_calculate_u_and_p(u32 i, u32 r_c, u32 p_b,
u32 *p, u32 *u);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 41/53] drm/radeon/kms: add dpm support for KB/KV
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (39 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 40/53] drm/radeon/dpm: add helper to fetch the vrefresh of the current mode Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 42/53] drm/radeon: add dpm support for CI dGPUs (v2) Alex Deucher
` (12 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
This adds dpm support for KB/KV asics. This includes:
- dynamic engine clock scaling
- dynamic voltage scaling
- power containment
- shader power scaling
Set radeon.dpm=1 to enable.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/Makefile | 2 +-
drivers/gpu/drm/radeon/cik.c | 30 +-
drivers/gpu/drm/radeon/cikd.h | 72 +
drivers/gpu/drm/radeon/kv_dpm.c | 2536 ++++++++++++++++++++++++++++++++++
drivers/gpu/drm/radeon/kv_dpm.h | 199 +++
drivers/gpu/drm/radeon/kv_smc.c | 207 +++
drivers/gpu/drm/radeon/ppsmc.h | 34 +
drivers/gpu/drm/radeon/radeon_asic.c | 14 +
drivers/gpu/drm/radeon/radeon_asic.h | 14 +
drivers/gpu/drm/radeon/radeon_pm.c | 2 +
drivers/gpu/drm/radeon/smu7.h | 170 +++
drivers/gpu/drm/radeon/smu7_fusion.h | 300 ++++
12 files changed, 3576 insertions(+), 4 deletions(-)
create mode 100644 drivers/gpu/drm/radeon/kv_dpm.c
create mode 100644 drivers/gpu/drm/radeon/kv_dpm.h
create mode 100644 drivers/gpu/drm/radeon/kv_smc.c
create mode 100644 drivers/gpu/drm/radeon/smu7.h
create mode 100644 drivers/gpu/drm/radeon/smu7_fusion.h
diff --git a/drivers/gpu/drm/radeon/Makefile b/drivers/gpu/drm/radeon/Makefile
index bfabd69..d3265b5 100644
--- a/drivers/gpu/drm/radeon/Makefile
+++ b/drivers/gpu/drm/radeon/Makefile
@@ -79,7 +79,7 @@ radeon-y += radeon_device.o radeon_asic.o radeon_kms.o \
si_blit_shaders.o radeon_prime.o radeon_uvd.o cik.o cik_blit_shaders.o \
r600_dpm.o rs780_dpm.o rv6xx_dpm.o rv770_dpm.o rv730_dpm.o rv740_dpm.o \
rv770_smc.o cypress_dpm.o btc_dpm.o sumo_dpm.o sumo_smc.o trinity_dpm.o \
- trinity_smc.o ni_dpm.o si_smc.o si_dpm.o
+ trinity_smc.o ni_dpm.o si_smc.o si_dpm.o kv_smc.o kv_dpm.o
radeon-$(CONFIG_COMPAT) += radeon_ioc32.o
radeon-$(CONFIG_VGA_SWITCHEROO) += radeon_atpx_handler.o
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index d0804f7..87e5aee 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -6593,6 +6593,7 @@ int cik_irq_set(struct radeon_device *rdev)
u32 hpd1, hpd2, hpd3, hpd4, hpd5, hpd6;
u32 grbm_int_cntl = 0;
u32 dma_cntl, dma_cntl1;
+ u32 thermal_int;
if (!rdev->irq.installed) {
WARN(1, "Can't enable IRQ/MSI because no handler is installed\n");
@@ -6625,6 +6626,9 @@ int cik_irq_set(struct radeon_device *rdev)
cp_m2p2 = RREG32(CP_ME2_PIPE2_INT_CNTL) & ~TIME_STAMP_INT_ENABLE;
cp_m2p3 = RREG32(CP_ME2_PIPE3_INT_CNTL) & ~TIME_STAMP_INT_ENABLE;
+ thermal_int = RREG32_SMC(CG_THERMAL_INT_CTRL) &
+ ~(THERM_INTH_MASK | THERM_INTL_MASK);
+
/* enable CP interrupts on all rings */
if (atomic_read(&rdev->irq.ring_int[RADEON_RING_TYPE_GFX_INDEX])) {
DRM_DEBUG("cik_irq_set: sw int gfx\n");
@@ -6782,6 +6786,11 @@ int cik_irq_set(struct radeon_device *rdev)
hpd6 |= DC_HPDx_INT_EN;
}
+ if (rdev->irq.dpm_thermal) {
+ DRM_DEBUG("dpm thermal\n");
+ thermal_int |= THERM_INTH_MASK | THERM_INTL_MASK;
+ }
+
WREG32(CP_INT_CNTL_RING0, cp_int_cntl);
WREG32(SDMA0_CNTL + SDMA0_REGISTER_OFFSET, dma_cntl);
@@ -6816,6 +6825,8 @@ int cik_irq_set(struct radeon_device *rdev)
WREG32(DC_HPD5_INT_CONTROL, hpd5);
WREG32(DC_HPD6_INT_CONTROL, hpd6);
+ WREG32_SMC(CG_THERMAL_INT_CTRL, thermal_int);
+
return 0;
}
@@ -7027,6 +7038,7 @@ int cik_irq_process(struct radeon_device *rdev)
bool queue_hotplug = false;
bool queue_reset = false;
u32 addr, status, mc_client;
+ bool queue_thermal = false;
if (!rdev->ih.enabled || rdev->shutdown)
return IRQ_NONE;
@@ -7377,6 +7389,19 @@ restart_ih:
break;
}
break;
+ case 230: /* thermal low to high */
+ DRM_DEBUG("IH: thermal low to high\n");
+ rdev->pm.dpm.thermal.high_to_low = false;
+ queue_thermal = true;
+ break;
+ case 231: /* thermal high to low */
+ DRM_DEBUG("IH: thermal high to low\n");
+ rdev->pm.dpm.thermal.high_to_low = true;
+ queue_thermal = true;
+ break;
+ case 233: /* GUI IDLE */
+ DRM_DEBUG("IH: GUI idle\n");
+ break;
case 241: /* SDMA Privileged inst */
case 247: /* SDMA Privileged inst */
DRM_ERROR("Illegal instruction in SDMA command stream\n");
@@ -7416,9 +7441,6 @@ restart_ih:
break;
}
break;
- case 233: /* GUI IDLE */
- DRM_DEBUG("IH: GUI idle\n");
- break;
default:
DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
break;
@@ -7432,6 +7454,8 @@ restart_ih:
schedule_work(&rdev->hotplug_work);
if (queue_reset)
schedule_work(&rdev->reset_work);
+ if (queue_thermal)
+ schedule_work(&rdev->pm.dpm.thermal.work);
rdev->ih.rptr = rptr;
WREG32(IH_RB_RPTR, rdev->ih.rptr);
atomic_set(&rdev->ih.lock, 0);
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index b66518a..f4c417d 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -28,10 +28,59 @@
#define CIK_RB_BITMAP_WIDTH_PER_SH 2
+/* DIDT IND registers */
+#define DIDT_SQ_CTRL0 0x0
+# define DIDT_CTRL_EN (1 << 0)
+#define DIDT_DB_CTRL0 0x20
+#define DIDT_TD_CTRL0 0x40
+#define DIDT_TCP_CTRL0 0x60
+
/* SMC IND registers */
+#define NB_DPM_CONFIG_1 0x3F9E8
+# define Dpm0PgNbPsLo(x) ((x) << 0)
+# define Dpm0PgNbPsLo_MASK 0x000000ff
+# define Dpm0PgNbPsLo_SHIFT 0
+# define Dpm0PgNbPsHi(x) ((x) << 8)
+# define Dpm0PgNbPsHi_MASK 0x0000ff00
+# define Dpm0PgNbPsHi_SHIFT 8
+# define DpmXNbPsLo(x) ((x) << 16)
+# define DpmXNbPsLo_MASK 0x00ff0000
+# define DpmXNbPsLo_SHIFT 16
+# define DpmXNbPsHi(x) ((x) << 24)
+# define DpmXNbPsHi_MASK 0xff000000
+# define DpmXNbPsHi_SHIFT 24
+
+#define SMC_SYSCON_MSG_ARG_0 0x80000068
+
#define GENERAL_PWRMGT 0xC0200000
+# define GLOBAL_PWRMGT_EN (1 << 0)
# define GPU_COUNTER_CLK (1 << 15)
+#define SCLK_PWRMGT_CNTL 0xC0200008
+# define RESET_BUSY_CNT (1 << 4)
+# define RESET_SCLK_CNT (1 << 5)
+# define DYNAMIC_PM_EN (1 << 21)
+
+#define CG_FTV_0 0xC02001A8
+
+#define LCAC_SX0_OVR_SEL 0xC0400D04
+#define LCAC_SX0_OVR_VAL 0xC0400D08
+
+#define LCAC_MC0_OVR_SEL 0xC0400D34
+#define LCAC_MC0_OVR_VAL 0xC0400D38
+
+#define LCAC_MC1_OVR_SEL 0xC0400D40
+#define LCAC_MC1_OVR_VAL 0xC0400D44
+
+#define LCAC_MC2_OVR_SEL 0xC0400D4C
+#define LCAC_MC2_OVR_VAL 0xC0400D50
+
+#define LCAC_MC3_OVR_SEL 0xC0400D58
+#define LCAC_MC3_OVR_VAL 0xC0400D5C
+
+#define LCAC_CPL_OVR_SEL 0xC0400D84
+#define LCAC_CPL_OVR_VAL 0xC0400D88
+
#define CG_MULT_THERMAL_STATUS 0xC0300010
#define ASIC_MAX_TEMP(x) ((x) << 0)
#define ASIC_MAX_TEMP_MASK 0x000001ff
@@ -60,6 +109,16 @@
# define ZCLK_SEL(x) ((x) << 8)
# define ZCLK_SEL_MASK 0xFF00
+#define CG_THERMAL_INT_CTRL 0xC2100028
+#define DIG_THERM_INTH(x) ((x) << 0)
+#define DIG_THERM_INTH_MASK 0x000000FF
+#define DIG_THERM_INTH_SHIFT 0
+#define DIG_THERM_INTL(x) ((x) << 8)
+#define DIG_THERM_INTL_MASK 0x0000FF00
+#define DIG_THERM_INTL_SHIFT 8
+#define THERM_INTH_MASK (1 << 24)
+#define THERM_INTL_MASK (1 << 25)
+
/* PCIE registers idx/data 0x38/0x3c */
#define PB0_PIF_PWRDOWN_0 0x1100012 /* PCIE */
# define PLL_POWER_STATE_IN_TXS2_0(x) ((x) << 7)
@@ -173,6 +232,19 @@
#define PCIE_INDEX 0x38
#define PCIE_DATA 0x3C
+#define SMC_IND_INDEX_0 0x200
+#define SMC_IND_DATA_0 0x204
+
+#define SMC_IND_ACCESS_CNTL 0x240
+#define AUTO_INCREMENT_IND_0 (1 << 0)
+
+#define SMC_MESSAGE_0 0x250
+#define SMC_MSG_MASK 0xffff
+#define SMC_RESP_0 0x254
+#define SMC_RESP_MASK 0xffff
+
+#define SMC_MSG_ARG_0 0x290
+
#define VGA_HDP_CONTROL 0x328
#define VGA_MEMORY_DISABLE (1 << 4)
diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c
new file mode 100644
index 0000000..2e40163
--- /dev/null
+++ b/drivers/gpu/drm/radeon/kv_dpm.c
@@ -0,0 +1,2536 @@
+/*
+ * Copyright 2013 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include "drmP.h"
+#include "radeon.h"
+#include "cikd.h"
+#include "r600_dpm.h"
+#include "kv_dpm.h"
+
+#define KV_MAX_DEEPSLEEP_DIVIDER_ID 5
+#define KV_MINIMUM_ENGINE_CLOCK 800
+#define SMC_RAM_END 0x40000
+
+static void kv_init_graphics_levels(struct radeon_device *rdev);
+static int kv_calculate_ds_divider(struct radeon_device *rdev);
+static int kv_calculate_nbps_level_settings(struct radeon_device *rdev);
+static int kv_calculate_dpm_settings(struct radeon_device *rdev);
+static void kv_enable_new_levels(struct radeon_device *rdev);
+static void kv_program_nbps_index_settings(struct radeon_device *rdev,
+ struct radeon_ps *new_rps);
+static int kv_set_enabled_levels(struct radeon_device *rdev);
+static int kv_force_dpm_lowest(struct radeon_device *rdev);
+static void kv_apply_state_adjust_rules(struct radeon_device *rdev,
+ struct radeon_ps *new_rps,
+ struct radeon_ps *old_rps);
+static int kv_set_thermal_temperature_range(struct radeon_device *rdev,
+ int min_temp, int max_temp);
+static int kv_init_fps_limits(struct radeon_device *rdev);
+
+static void kv_dpm_powergate_uvd(struct radeon_device *rdev, bool gate);
+static void kv_dpm_powergate_vce(struct radeon_device *rdev, bool gate);
+static void kv_dpm_powergate_samu(struct radeon_device *rdev, bool gate);
+static void kv_dpm_powergate_acp(struct radeon_device *rdev, bool gate);
+
+extern void cik_enter_rlc_safe_mode(struct radeon_device *rdev);
+extern void cik_exit_rlc_safe_mode(struct radeon_device *rdev);
+extern void cik_update_cg(struct radeon_device *rdev,
+ u32 block, bool enable);
+
+static const struct kv_lcac_config_values sx_local_cac_cfg_kv[] =
+{
+ { 0, 4, 1 },
+ { 1, 4, 1 },
+ { 2, 5, 1 },
+ { 3, 4, 2 },
+ { 4, 1, 1 },
+ { 5, 5, 2 },
+ { 6, 6, 1 },
+ { 7, 9, 2 },
+ { 0xffffffff }
+};
+
+static const struct kv_lcac_config_values mc0_local_cac_cfg_kv[] =
+{
+ { 0, 4, 1 },
+ { 0xffffffff }
+};
+
+static const struct kv_lcac_config_values mc1_local_cac_cfg_kv[] =
+{
+ { 0, 4, 1 },
+ { 0xffffffff }
+};
+
+static const struct kv_lcac_config_values mc2_local_cac_cfg_kv[] =
+{
+ { 0, 4, 1 },
+ { 0xffffffff }
+};
+
+static const struct kv_lcac_config_values mc3_local_cac_cfg_kv[] =
+{
+ { 0, 4, 1 },
+ { 0xffffffff }
+};
+
+static const struct kv_lcac_config_values cpl_local_cac_cfg_kv[] =
+{
+ { 0, 4, 1 },
+ { 1, 4, 1 },
+ { 2, 5, 1 },
+ { 3, 4, 1 },
+ { 4, 1, 1 },
+ { 5, 5, 1 },
+ { 6, 6, 1 },
+ { 7, 9, 1 },
+ { 8, 4, 1 },
+ { 9, 2, 1 },
+ { 10, 3, 1 },
+ { 11, 6, 1 },
+ { 12, 8, 2 },
+ { 13, 1, 1 },
+ { 14, 2, 1 },
+ { 15, 3, 1 },
+ { 16, 1, 1 },
+ { 17, 4, 1 },
+ { 18, 3, 1 },
+ { 19, 1, 1 },
+ { 20, 8, 1 },
+ { 21, 5, 1 },
+ { 22, 1, 1 },
+ { 23, 1, 1 },
+ { 24, 4, 1 },
+ { 27, 6, 1 },
+ { 28, 1, 1 },
+ { 0xffffffff }
+};
+
+static const struct kv_lcac_config_reg sx0_cac_config_reg[] =
+{
+ { 0xc0400d00, 0x003e0000, 17, 0x3fc00000, 22, 0x0001fffe, 1, 0x00000001, 0 }
+};
+
+static const struct kv_lcac_config_reg mc0_cac_config_reg[] =
+{
+ { 0xc0400d30, 0x003e0000, 17, 0x3fc00000, 22, 0x0001fffe, 1, 0x00000001, 0 }
+};
+
+static const struct kv_lcac_config_reg mc1_cac_config_reg[] =
+{
+ { 0xc0400d3c, 0x003e0000, 17, 0x3fc00000, 22, 0x0001fffe, 1, 0x00000001, 0 }
+};
+
+static const struct kv_lcac_config_reg mc2_cac_config_reg[] =
+{
+ { 0xc0400d48, 0x003e0000, 17, 0x3fc00000, 22, 0x0001fffe, 1, 0x00000001, 0 }
+};
+
+static const struct kv_lcac_config_reg mc3_cac_config_reg[] =
+{
+ { 0xc0400d54, 0x003e0000, 17, 0x3fc00000, 22, 0x0001fffe, 1, 0x00000001, 0 }
+};
+
+static const struct kv_lcac_config_reg cpl_cac_config_reg[] =
+{
+ { 0xc0400d80, 0x003e0000, 17, 0x3fc00000, 22, 0x0001fffe, 1, 0x00000001, 0 }
+};
+
+static const struct kv_pt_config_reg didt_config_kv[] =
+{
+ { 0x10, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x10, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x10, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x10, 0xff000000, 24, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x11, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x11, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x11, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x11, 0xff000000, 24, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x12, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x12, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x12, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x12, 0xff000000, 24, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x2, 0x00003fff, 0, 0x4, KV_CONFIGREG_DIDT_IND },
+ { 0x2, 0x03ff0000, 16, 0x80, KV_CONFIGREG_DIDT_IND },
+ { 0x2, 0x78000000, 27, 0x3, KV_CONFIGREG_DIDT_IND },
+ { 0x1, 0x0000ffff, 0, 0x3FFF, KV_CONFIGREG_DIDT_IND },
+ { 0x1, 0xffff0000, 16, 0x3FFF, KV_CONFIGREG_DIDT_IND },
+ { 0x0, 0x00000001, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x30, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x30, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x30, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x30, 0xff000000, 24, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x31, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x31, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x31, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x31, 0xff000000, 24, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x32, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x32, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x32, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x32, 0xff000000, 24, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x22, 0x00003fff, 0, 0x4, KV_CONFIGREG_DIDT_IND },
+ { 0x22, 0x03ff0000, 16, 0x80, KV_CONFIGREG_DIDT_IND },
+ { 0x22, 0x78000000, 27, 0x3, KV_CONFIGREG_DIDT_IND },
+ { 0x21, 0x0000ffff, 0, 0x3FFF, KV_CONFIGREG_DIDT_IND },
+ { 0x21, 0xffff0000, 16, 0x3FFF, KV_CONFIGREG_DIDT_IND },
+ { 0x20, 0x00000001, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x50, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x50, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x50, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x50, 0xff000000, 24, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x51, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x51, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x51, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x51, 0xff000000, 24, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x52, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x52, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x52, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x52, 0xff000000, 24, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x42, 0x00003fff, 0, 0x4, KV_CONFIGREG_DIDT_IND },
+ { 0x42, 0x03ff0000, 16, 0x80, KV_CONFIGREG_DIDT_IND },
+ { 0x42, 0x78000000, 27, 0x3, KV_CONFIGREG_DIDT_IND },
+ { 0x41, 0x0000ffff, 0, 0x3FFF, KV_CONFIGREG_DIDT_IND },
+ { 0x41, 0xffff0000, 16, 0x3FFF, KV_CONFIGREG_DIDT_IND },
+ { 0x40, 0x00000001, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x70, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x70, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x70, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x70, 0xff000000, 24, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x71, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x71, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x71, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x71, 0xff000000, 24, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x72, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x72, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x72, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x72, 0xff000000, 24, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0x62, 0x00003fff, 0, 0x4, KV_CONFIGREG_DIDT_IND },
+ { 0x62, 0x03ff0000, 16, 0x80, KV_CONFIGREG_DIDT_IND },
+ { 0x62, 0x78000000, 27, 0x3, KV_CONFIGREG_DIDT_IND },
+ { 0x61, 0x0000ffff, 0, 0x3FFF, KV_CONFIGREG_DIDT_IND },
+ { 0x61, 0xffff0000, 16, 0x3FFF, KV_CONFIGREG_DIDT_IND },
+ { 0x60, 0x00000001, 0, 0x0, KV_CONFIGREG_DIDT_IND },
+ { 0xFFFFFFFF }
+};
+
+static struct kv_ps *kv_get_ps(struct radeon_ps *rps)
+{
+ struct kv_ps *ps = rps->ps_priv;
+
+ return ps;
+}
+
+static struct kv_power_info *kv_get_pi(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = rdev->pm.dpm.priv;
+
+ return pi;
+}
+
+#if 0
+static void kv_program_local_cac_table(struct radeon_device *rdev,
+ const struct kv_lcac_config_values *local_cac_table,
+ const struct kv_lcac_config_reg *local_cac_reg)
+{
+ u32 i, count, data;
+ const struct kv_lcac_config_values *values = local_cac_table;
+
+ while (values->block_id != 0xffffffff) {
+ count = values->signal_id;
+ for (i = 0; i < count; i++) {
+ data = ((values->block_id << local_cac_reg->block_shift) &
+ local_cac_reg->block_mask);
+ data |= ((i << local_cac_reg->signal_shift) &
+ local_cac_reg->signal_mask);
+ data |= ((values->t << local_cac_reg->t_shift) &
+ local_cac_reg->t_mask);
+ data |= ((1 << local_cac_reg->enable_shift) &
+ local_cac_reg->enable_mask);
+ WREG32_SMC(local_cac_reg->cntl, data);
+ }
+ values++;
+ }
+}
+#endif
+
+static int kv_program_pt_config_registers(struct radeon_device *rdev,
+ const struct kv_pt_config_reg *cac_config_regs)
+{
+ const struct kv_pt_config_reg *config_regs = cac_config_regs;
+ u32 data;
+ u32 cache = 0;
+
+ if (config_regs == NULL)
+ return -EINVAL;
+
+ while (config_regs->offset != 0xFFFFFFFF) {
+ if (config_regs->type == KV_CONFIGREG_CACHE) {
+ cache |= ((config_regs->value << config_regs->shift) & config_regs->mask);
+ } else {
+ switch (config_regs->type) {
+ case KV_CONFIGREG_SMC_IND:
+ data = RREG32_SMC(config_regs->offset);
+ break;
+ case KV_CONFIGREG_DIDT_IND:
+ data = RREG32_DIDT(config_regs->offset);
+ break;
+ default:
+ data = RREG32(config_regs->offset << 2);
+ break;
+ }
+
+ data &= ~config_regs->mask;
+ data |= ((config_regs->value << config_regs->shift) & config_regs->mask);
+ data |= cache;
+ cache = 0;
+
+ switch (config_regs->type) {
+ case KV_CONFIGREG_SMC_IND:
+ WREG32_SMC(config_regs->offset, data);
+ break;
+ case KV_CONFIGREG_DIDT_IND:
+ WREG32_DIDT(config_regs->offset, data);
+ break;
+ default:
+ WREG32(config_regs->offset << 2, data);
+ break;
+ }
+ }
+ config_regs++;
+ }
+
+ return 0;
+}
+
+static void kv_do_enable_didt(struct radeon_device *rdev, bool enable)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 data;
+
+ if (pi->caps_sq_ramping) {
+ data = RREG32_DIDT(DIDT_SQ_CTRL0);
+ if (enable)
+ data |= DIDT_CTRL_EN;
+ else
+ data &= ~DIDT_CTRL_EN;
+ WREG32_DIDT(DIDT_SQ_CTRL0, data);
+ }
+
+ if (pi->caps_db_ramping) {
+ data = RREG32_DIDT(DIDT_DB_CTRL0);
+ if (enable)
+ data |= DIDT_CTRL_EN;
+ else
+ data &= ~DIDT_CTRL_EN;
+ WREG32_DIDT(DIDT_DB_CTRL0, data);
+ }
+
+ if (pi->caps_td_ramping) {
+ data = RREG32_DIDT(DIDT_TD_CTRL0);
+ if (enable)
+ data |= DIDT_CTRL_EN;
+ else
+ data &= ~DIDT_CTRL_EN;
+ WREG32_DIDT(DIDT_TD_CTRL0, data);
+ }
+
+ if (pi->caps_tcp_ramping) {
+ data = RREG32_DIDT(DIDT_TCP_CTRL0);
+ if (enable)
+ data |= DIDT_CTRL_EN;
+ else
+ data &= ~DIDT_CTRL_EN;
+ WREG32_DIDT(DIDT_TCP_CTRL0, data);
+ }
+}
+
+static int kv_enable_didt(struct radeon_device *rdev, bool enable)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ int ret;
+
+ if (pi->caps_sq_ramping ||
+ pi->caps_db_ramping ||
+ pi->caps_td_ramping ||
+ pi->caps_tcp_ramping) {
+ cik_enter_rlc_safe_mode(rdev);
+
+ if (enable) {
+ ret = kv_program_pt_config_registers(rdev, didt_config_kv);
+ if (ret) {
+ cik_exit_rlc_safe_mode(rdev);
+ return ret;
+ }
+ }
+
+ kv_do_enable_didt(rdev, enable);
+
+ cik_exit_rlc_safe_mode(rdev);
+ }
+
+ return 0;
+}
+
+#if 0
+static void kv_initialize_hardware_cac_manager(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ if (pi->caps_cac) {
+ WREG32_SMC(LCAC_SX0_OVR_SEL, 0);
+ WREG32_SMC(LCAC_SX0_OVR_VAL, 0);
+ kv_program_local_cac_table(rdev, sx_local_cac_cfg_kv, sx0_cac_config_reg);
+
+ WREG32_SMC(LCAC_MC0_OVR_SEL, 0);
+ WREG32_SMC(LCAC_MC0_OVR_VAL, 0);
+ kv_program_local_cac_table(rdev, mc0_local_cac_cfg_kv, mc0_cac_config_reg);
+
+ WREG32_SMC(LCAC_MC1_OVR_SEL, 0);
+ WREG32_SMC(LCAC_MC1_OVR_VAL, 0);
+ kv_program_local_cac_table(rdev, mc1_local_cac_cfg_kv, mc1_cac_config_reg);
+
+ WREG32_SMC(LCAC_MC2_OVR_SEL, 0);
+ WREG32_SMC(LCAC_MC2_OVR_VAL, 0);
+ kv_program_local_cac_table(rdev, mc2_local_cac_cfg_kv, mc2_cac_config_reg);
+
+ WREG32_SMC(LCAC_MC3_OVR_SEL, 0);
+ WREG32_SMC(LCAC_MC3_OVR_VAL, 0);
+ kv_program_local_cac_table(rdev, mc3_local_cac_cfg_kv, mc3_cac_config_reg);
+
+ WREG32_SMC(LCAC_CPL_OVR_SEL, 0);
+ WREG32_SMC(LCAC_CPL_OVR_VAL, 0);
+ kv_program_local_cac_table(rdev, cpl_local_cac_cfg_kv, cpl_cac_config_reg);
+ }
+}
+#endif
+
+static int kv_enable_smc_cac(struct radeon_device *rdev, bool enable)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ int ret = 0;
+
+ if (pi->caps_cac) {
+ if (enable) {
+ ret = kv_notify_message_to_smu(rdev, PPSMC_MSG_EnableCac);
+ if (ret)
+ pi->cac_enabled = false;
+ else
+ pi->cac_enabled = true;
+ } else if (pi->cac_enabled) {
+ kv_notify_message_to_smu(rdev, PPSMC_MSG_DisableCac);
+ pi->cac_enabled = false;
+ }
+ }
+
+ return ret;
+}
+
+static int kv_process_firmware_header(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 tmp;
+ int ret;
+
+ ret = kv_read_smc_sram_dword(rdev, SMU7_FIRMWARE_HEADER_LOCATION +
+ offsetof(SMU7_Firmware_Header, DpmTable),
+ &tmp, pi->sram_end);
+
+ if (ret == 0)
+ pi->dpm_table_start = tmp;
+
+ ret = kv_read_smc_sram_dword(rdev, SMU7_FIRMWARE_HEADER_LOCATION +
+ offsetof(SMU7_Firmware_Header, SoftRegisters),
+ &tmp, pi->sram_end);
+
+ if (ret == 0)
+ pi->soft_regs_start = tmp;
+
+ return ret;
+}
+
+static int kv_enable_dpm_voltage_scaling(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ int ret;
+
+ pi->graphics_voltage_change_enable = 1;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, GraphicsVoltageChangeEnable),
+ &pi->graphics_voltage_change_enable,
+ sizeof(u8), pi->sram_end);
+
+ return ret;
+}
+
+static int kv_set_dpm_interval(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ int ret;
+
+ pi->graphics_interval = 1;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, GraphicsInterval),
+ &pi->graphics_interval,
+ sizeof(u8), pi->sram_end);
+
+ return ret;
+}
+
+static int kv_set_dpm_boot_state(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ int ret;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, GraphicsBootLevel),
+ &pi->graphics_boot_level,
+ sizeof(u8), pi->sram_end);
+
+ return ret;
+}
+
+static void kv_program_vc(struct radeon_device *rdev)
+{
+ WREG32_SMC(CG_FTV_0, 0x3FFFC000);
+}
+
+static void kv_clear_vc(struct radeon_device *rdev)
+{
+ WREG32_SMC(CG_FTV_0, 0);
+}
+
+static int kv_set_divider_value(struct radeon_device *rdev,
+ u32 index, u32 sclk)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct atom_clock_dividers dividers;
+ int ret;
+
+ ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+ sclk, false, ÷rs);
+ if (ret)
+ return ret;
+
+ pi->graphics_level[index].SclkDid = (u8)dividers.post_div;
+ pi->graphics_level[index].SclkFrequency = cpu_to_be32(sclk);
+
+ return 0;
+}
+
+static u16 kv_convert_8bit_index_to_voltage(struct radeon_device *rdev,
+ u16 voltage)
+{
+ return 6200 - (voltage * 25);
+}
+
+static u16 kv_convert_2bit_index_to_voltage(struct radeon_device *rdev,
+ u32 vid_2bit)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 vid_8bit = sumo_convert_vid2_to_vid7(rdev,
+ &pi->sys_info.vid_mapping_table,
+ vid_2bit);
+
+ return kv_convert_8bit_index_to_voltage(rdev, (u16)vid_8bit);
+}
+
+
+static int kv_set_vid(struct radeon_device *rdev, u32 index, u32 vid)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ pi->graphics_level[index].VoltageDownH = (u8)pi->voltage_drop_t;
+ pi->graphics_level[index].MinVddNb =
+ cpu_to_be32(kv_convert_2bit_index_to_voltage(rdev, vid));
+
+ return 0;
+}
+
+static int kv_set_at(struct radeon_device *rdev, u32 index, u32 at)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ pi->graphics_level[index].AT = cpu_to_be16((u16)at);
+
+ return 0;
+}
+
+static void kv_dpm_power_level_enable(struct radeon_device *rdev,
+ u32 index, bool enable)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ pi->graphics_level[index].EnabledForActivity = enable ? 1 : 0;
+}
+
+static void kv_start_dpm(struct radeon_device *rdev)
+{
+ u32 tmp = RREG32_SMC(GENERAL_PWRMGT);
+
+ tmp |= GLOBAL_PWRMGT_EN;
+ WREG32_SMC(GENERAL_PWRMGT, tmp);
+
+ kv_smc_dpm_enable(rdev, true);
+}
+
+static void kv_stop_dpm(struct radeon_device *rdev)
+{
+ kv_smc_dpm_enable(rdev, false);
+}
+
+static void kv_start_am(struct radeon_device *rdev)
+{
+ u32 sclk_pwrmgt_cntl = RREG32_SMC(SCLK_PWRMGT_CNTL);
+
+ sclk_pwrmgt_cntl &= ~(RESET_SCLK_CNT | RESET_BUSY_CNT);
+ sclk_pwrmgt_cntl |= DYNAMIC_PM_EN;
+
+ WREG32_SMC(SCLK_PWRMGT_CNTL, sclk_pwrmgt_cntl);
+}
+
+static void kv_reset_am(struct radeon_device *rdev)
+{
+ u32 sclk_pwrmgt_cntl = RREG32_SMC(SCLK_PWRMGT_CNTL);
+
+ sclk_pwrmgt_cntl |= (RESET_SCLK_CNT | RESET_BUSY_CNT);
+
+ WREG32_SMC(SCLK_PWRMGT_CNTL, sclk_pwrmgt_cntl);
+}
+
+static int kv_freeze_sclk_dpm(struct radeon_device *rdev, bool freeze)
+{
+ return kv_notify_message_to_smu(rdev, freeze ?
+ PPSMC_MSG_SCLKDPM_FreezeLevel : PPSMC_MSG_SCLKDPM_UnfreezeLevel);
+}
+
+static int kv_force_lowest_valid(struct radeon_device *rdev)
+{
+ return kv_force_dpm_lowest(rdev);
+}
+
+static int kv_unforce_levels(struct radeon_device *rdev)
+{
+ return kv_notify_message_to_smu(rdev, PPSMC_MSG_NoForcedLevel);
+}
+
+static int kv_update_sclk_t(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 low_sclk_interrupt_t = 0;
+ int ret = 0;
+
+ if (pi->caps_sclk_throttle_low_notification) {
+ low_sclk_interrupt_t = cpu_to_be32(pi->low_sclk_interrupt_t);
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, LowSclkInterruptT),
+ (u8 *)&low_sclk_interrupt_t,
+ sizeof(u32), pi->sram_end);
+ }
+ return ret;
+}
+
+static int kv_program_bootup_state(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 i;
+ struct radeon_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
+
+ if (table && table->count) {
+ for (i = pi->graphics_dpm_level_count - 1; i >= 0; i--) {
+ if ((table->entries[i].clk == pi->boot_pl.sclk) ||
+ (i == 0))
+ break;
+ }
+
+ pi->graphics_boot_level = (u8)i;
+ kv_dpm_power_level_enable(rdev, i, true);
+ } else {
+ struct sumo_sclk_voltage_mapping_table *table =
+ &pi->sys_info.sclk_voltage_mapping_table;
+
+ if (table->num_max_dpm_entries == 0)
+ return -EINVAL;
+
+ for (i = pi->graphics_dpm_level_count - 1; i >= 0; i--) {
+ if ((table->entries[i].sclk_frequency == pi->boot_pl.sclk) ||
+ (i == 0))
+ break;
+ }
+
+ pi->graphics_boot_level = (u8)i;
+ kv_dpm_power_level_enable(rdev, i, true);
+ }
+ return 0;
+}
+
+static int kv_enable_auto_thermal_throttling(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ int ret;
+
+ pi->graphics_therm_throttle_enable = 1;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, GraphicsThermThrottleEnable),
+ &pi->graphics_therm_throttle_enable,
+ sizeof(u8), pi->sram_end);
+
+ return ret;
+}
+
+static int kv_upload_dpm_settings(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ int ret;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, GraphicsLevel),
+ (u8 *)&pi->graphics_level,
+ sizeof(SMU7_Fusion_GraphicsLevel) * SMU7_MAX_LEVELS_GRAPHICS,
+ pi->sram_end);
+
+ if (ret)
+ return ret;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, GraphicsDpmLevelCount),
+ &pi->graphics_dpm_level_count,
+ sizeof(u8), pi->sram_end);
+
+ return ret;
+}
+
+static u32 kv_get_clock_difference(u32 a, u32 b)
+{
+ return (a >= b) ? a - b : b - a;
+}
+
+static u32 kv_get_clk_bypass(struct radeon_device *rdev, u32 clk)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 value;
+
+ if (pi->caps_enable_dfs_bypass) {
+ if (kv_get_clock_difference(clk, 40000) < 200)
+ value = 3;
+ else if (kv_get_clock_difference(clk, 30000) < 200)
+ value = 2;
+ else if (kv_get_clock_difference(clk, 20000) < 200)
+ value = 7;
+ else if (kv_get_clock_difference(clk, 15000) < 200)
+ value = 6;
+ else if (kv_get_clock_difference(clk, 10000) < 200)
+ value = 8;
+ else
+ value = 0;
+ } else {
+ value = 0;
+ }
+
+ return value;
+}
+
+static int kv_populate_uvd_table(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct radeon_uvd_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table;
+ struct atom_clock_dividers dividers;
+ int ret;
+ u32 i;
+
+ if (table == NULL || table->count == 0)
+ return 0;
+
+ pi->uvd_level_count = 0;
+ for (i = 0; i < table->count; i++) {
+ if (pi->high_voltage_t &&
+ (pi->high_voltage_t < table->entries[i].v))
+ break;
+
+ pi->uvd_level[i].VclkFrequency = cpu_to_be32(table->entries[i].vclk);
+ pi->uvd_level[i].DclkFrequency = cpu_to_be32(table->entries[i].dclk);
+ pi->uvd_level[i].MinVddNb = cpu_to_be16(table->entries[i].v);
+
+ pi->uvd_level[i].VClkBypassCntl =
+ (u8)kv_get_clk_bypass(rdev, table->entries[i].vclk);
+ pi->uvd_level[i].DClkBypassCntl =
+ (u8)kv_get_clk_bypass(rdev, table->entries[i].dclk);
+
+ ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+ table->entries[i].vclk, false, ÷rs);
+ if (ret)
+ return ret;
+ pi->uvd_level[i].VclkDivider = (u8)dividers.post_div;
+
+ ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+ table->entries[i].dclk, false, ÷rs);
+ if (ret)
+ return ret;
+ pi->uvd_level[i].DclkDivider = (u8)dividers.post_div;
+
+ pi->uvd_level_count++;
+ }
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, UvdLevelCount),
+ (u8 *)&pi->uvd_level_count,
+ sizeof(u8), pi->sram_end);
+ if (ret)
+ return ret;
+
+ pi->uvd_interval = 1;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, UVDInterval),
+ &pi->uvd_interval,
+ sizeof(u8), pi->sram_end);
+ if (ret)
+ return ret;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, UvdLevel),
+ (u8 *)&pi->uvd_level,
+ sizeof(SMU7_Fusion_UvdLevel) * SMU7_MAX_LEVELS_UVD,
+ pi->sram_end);
+
+ return ret;
+
+}
+
+static int kv_populate_vce_table(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ int ret;
+ u32 i;
+ struct radeon_vce_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table;
+ struct atom_clock_dividers dividers;
+
+ if (table == NULL || table->count == 0)
+ return 0;
+
+ pi->vce_level_count = 0;
+ for (i = 0; i < table->count; i++) {
+ if (pi->high_voltage_t &&
+ pi->high_voltage_t < table->entries[i].v)
+ break;
+
+ pi->vce_level[i].Frequency = cpu_to_be32(table->entries[i].evclk);
+ pi->vce_level[i].MinVoltage = cpu_to_be16(table->entries[i].v);
+
+ pi->vce_level[i].ClkBypassCntl =
+ (u8)kv_get_clk_bypass(rdev, table->entries[i].evclk);
+
+ ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+ table->entries[i].evclk, false, ÷rs);
+ if (ret)
+ return ret;
+ pi->vce_level[i].Divider = (u8)dividers.post_div;
+
+ pi->vce_level_count++;
+ }
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, VceLevelCount),
+ (u8 *)&pi->vce_level_count,
+ sizeof(u8),
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ pi->vce_interval = 1;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, VCEInterval),
+ (u8 *)&pi->vce_interval,
+ sizeof(u8),
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, VceLevel),
+ (u8 *)&pi->vce_level,
+ sizeof(SMU7_Fusion_ExtClkLevel) * SMU7_MAX_LEVELS_VCE,
+ pi->sram_end);
+
+ return ret;
+}
+
+static int kv_populate_samu_table(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct radeon_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table;
+ struct atom_clock_dividers dividers;
+ int ret;
+ u32 i;
+
+ if (table == NULL || table->count == 0)
+ return 0;
+
+ pi->samu_level_count = 0;
+ for (i = 0; i < table->count; i++) {
+ if (pi->high_voltage_t &&
+ pi->high_voltage_t < table->entries[i].v)
+ break;
+
+ pi->samu_level[i].Frequency = cpu_to_be32(table->entries[i].clk);
+ pi->samu_level[i].MinVoltage = cpu_to_be16(table->entries[i].v);
+
+ pi->samu_level[i].ClkBypassCntl =
+ (u8)kv_get_clk_bypass(rdev, table->entries[i].clk);
+
+ ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+ table->entries[i].clk, false, ÷rs);
+ if (ret)
+ return ret;
+ pi->samu_level[i].Divider = (u8)dividers.post_div;
+
+ pi->samu_level_count++;
+ }
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, SamuLevelCount),
+ (u8 *)&pi->samu_level_count,
+ sizeof(u8),
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ pi->samu_interval = 1;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, SAMUInterval),
+ (u8 *)&pi->samu_interval,
+ sizeof(u8),
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, SamuLevel),
+ (u8 *)&pi->samu_level,
+ sizeof(SMU7_Fusion_ExtClkLevel) * SMU7_MAX_LEVELS_SAMU,
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ return ret;
+}
+
+
+static int kv_populate_acp_table(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct radeon_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table;
+ struct atom_clock_dividers dividers;
+ int ret;
+ u32 i;
+
+ if (table == NULL || table->count == 0)
+ return 0;
+
+ pi->acp_level_count = 0;
+ for (i = 0; i < table->count; i++) {
+ pi->acp_level[i].Frequency = cpu_to_be32(table->entries[i].clk);
+ pi->acp_level[i].MinVoltage = cpu_to_be16(table->entries[i].v);
+
+ ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+ table->entries[i].clk, false, ÷rs);
+ if (ret)
+ return ret;
+ pi->acp_level[i].Divider = (u8)dividers.post_div;
+
+ pi->acp_level_count++;
+ }
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, AcpLevelCount),
+ (u8 *)&pi->acp_level_count,
+ sizeof(u8),
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ pi->acp_interval = 1;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, ACPInterval),
+ (u8 *)&pi->acp_interval,
+ sizeof(u8),
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, AcpLevel),
+ (u8 *)&pi->acp_level,
+ sizeof(SMU7_Fusion_ExtClkLevel) * SMU7_MAX_LEVELS_ACP,
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ return ret;
+}
+
+static void kv_calculate_dfs_bypass_settings(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 i;
+ struct radeon_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
+
+ if (table && table->count) {
+ for (i = 0; i < pi->graphics_dpm_level_count; i++) {
+ if (pi->caps_enable_dfs_bypass) {
+ if (kv_get_clock_difference(table->entries[i].clk, 40000) < 200)
+ pi->graphics_level[i].ClkBypassCntl = 3;
+ else if (kv_get_clock_difference(table->entries[i].clk, 30000) < 200)
+ pi->graphics_level[i].ClkBypassCntl = 2;
+ else if (kv_get_clock_difference(table->entries[i].clk, 26600) < 200)
+ pi->graphics_level[i].ClkBypassCntl = 7;
+ else if (kv_get_clock_difference(table->entries[i].clk , 20000) < 200)
+ pi->graphics_level[i].ClkBypassCntl = 6;
+ else if (kv_get_clock_difference(table->entries[i].clk , 10000) < 200)
+ pi->graphics_level[i].ClkBypassCntl = 8;
+ else
+ pi->graphics_level[i].ClkBypassCntl = 0;
+ } else {
+ pi->graphics_level[i].ClkBypassCntl = 0;
+ }
+ }
+ } else {
+ struct sumo_sclk_voltage_mapping_table *table =
+ &pi->sys_info.sclk_voltage_mapping_table;
+ for (i = 0; i < pi->graphics_dpm_level_count; i++) {
+ if (pi->caps_enable_dfs_bypass) {
+ if (kv_get_clock_difference(table->entries[i].sclk_frequency, 40000) < 200)
+ pi->graphics_level[i].ClkBypassCntl = 3;
+ else if (kv_get_clock_difference(table->entries[i].sclk_frequency, 30000) < 200)
+ pi->graphics_level[i].ClkBypassCntl = 2;
+ else if (kv_get_clock_difference(table->entries[i].sclk_frequency, 26600) < 200)
+ pi->graphics_level[i].ClkBypassCntl = 7;
+ else if (kv_get_clock_difference(table->entries[i].sclk_frequency, 20000) < 200)
+ pi->graphics_level[i].ClkBypassCntl = 6;
+ else if (kv_get_clock_difference(table->entries[i].sclk_frequency, 10000) < 200)
+ pi->graphics_level[i].ClkBypassCntl = 8;
+ else
+ pi->graphics_level[i].ClkBypassCntl = 0;
+ } else {
+ pi->graphics_level[i].ClkBypassCntl = 0;
+ }
+ }
+ }
+}
+
+static int kv_enable_ulv(struct radeon_device *rdev, bool enable)
+{
+ return kv_notify_message_to_smu(rdev, enable ?
+ PPSMC_MSG_EnableULV : PPSMC_MSG_DisableULV);
+}
+
+static void kv_update_current_ps(struct radeon_device *rdev,
+ struct radeon_ps *rps)
+{
+ struct kv_ps *new_ps = kv_get_ps(rps);
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ pi->current_rps = *rps;
+ pi->current_ps = *new_ps;
+ pi->current_rps.ps_priv = &pi->current_ps;
+}
+
+static void kv_update_requested_ps(struct radeon_device *rdev,
+ struct radeon_ps *rps)
+{
+ struct kv_ps *new_ps = kv_get_ps(rps);
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ pi->requested_rps = *rps;
+ pi->requested_ps = *new_ps;
+ pi->requested_rps.ps_priv = &pi->requested_ps;
+}
+
+int kv_dpm_enable(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ int ret;
+
+ ret = kv_process_firmware_header(rdev);
+ if (ret) {
+ DRM_ERROR("kv_process_firmware_header failed\n");
+ return ret;
+ }
+ kv_init_fps_limits(rdev);
+ kv_init_graphics_levels(rdev);
+ ret = kv_program_bootup_state(rdev);
+ if (ret) {
+ DRM_ERROR("kv_program_bootup_state failed\n");
+ return ret;
+ }
+ kv_calculate_dfs_bypass_settings(rdev);
+ ret = kv_upload_dpm_settings(rdev);
+ if (ret) {
+ DRM_ERROR("kv_upload_dpm_settings failed\n");
+ return ret;
+ }
+ ret = kv_populate_uvd_table(rdev);
+ if (ret) {
+ DRM_ERROR("kv_populate_uvd_table failed\n");
+ return ret;
+ }
+ ret = kv_populate_vce_table(rdev);
+ if (ret) {
+ DRM_ERROR("kv_populate_vce_table failed\n");
+ return ret;
+ }
+ ret = kv_populate_samu_table(rdev);
+ if (ret) {
+ DRM_ERROR("kv_populate_samu_table failed\n");
+ return ret;
+ }
+ ret = kv_populate_acp_table(rdev);
+ if (ret) {
+ DRM_ERROR("kv_populate_acp_table failed\n");
+ return ret;
+ }
+ kv_program_vc(rdev);
+#if 0
+ kv_initialize_hardware_cac_manager(rdev);
+#endif
+ kv_start_am(rdev);
+ if (pi->enable_auto_thermal_throttling) {
+ ret = kv_enable_auto_thermal_throttling(rdev);
+ if (ret) {
+ DRM_ERROR("kv_enable_auto_thermal_throttling failed\n");
+ return ret;
+ }
+ }
+ ret = kv_enable_dpm_voltage_scaling(rdev);
+ if (ret) {
+ DRM_ERROR("kv_enable_dpm_voltage_scaling failed\n");
+ return ret;
+ }
+ ret = kv_set_dpm_interval(rdev);
+ if (ret) {
+ DRM_ERROR("kv_set_dpm_interval failed\n");
+ return ret;
+ }
+ ret = kv_set_dpm_boot_state(rdev);
+ if (ret) {
+ DRM_ERROR("kv_set_dpm_boot_state failed\n");
+ return ret;
+ }
+ ret = kv_enable_ulv(rdev, true);
+ if (ret) {
+ DRM_ERROR("kv_enable_ulv failed\n");
+ return ret;
+ }
+ kv_start_dpm(rdev);
+ ret = kv_enable_didt(rdev, true);
+ if (ret) {
+ DRM_ERROR("kv_enable_didt failed\n");
+ return ret;
+ }
+ ret = kv_enable_smc_cac(rdev, true);
+ if (ret) {
+ DRM_ERROR("kv_enable_smc_cac failed\n");
+ return ret;
+ }
+
+ if (rdev->irq.installed &&
+ r600_is_internal_thermal_sensor(rdev->pm.int_thermal_type)) {
+ ret = kv_set_thermal_temperature_range(rdev, R600_TEMP_RANGE_MIN, R600_TEMP_RANGE_MAX);
+ if (ret) {
+ DRM_ERROR("kv_set_thermal_temperature_range failed\n");
+ return ret;
+ }
+ rdev->irq.dpm_thermal = true;
+ radeon_irq_set(rdev);
+ }
+
+ /* powerdown unused blocks for now */
+ kv_dpm_powergate_acp(rdev, true);
+ kv_dpm_powergate_samu(rdev, true);
+ kv_dpm_powergate_vce(rdev, true);
+
+ kv_update_current_ps(rdev, rdev->pm.dpm.boot_ps);
+
+ return ret;
+}
+
+void kv_dpm_disable(struct radeon_device *rdev)
+{
+ kv_enable_smc_cac(rdev, false);
+ kv_enable_didt(rdev, false);
+ kv_clear_vc(rdev);
+ kv_stop_dpm(rdev);
+ kv_enable_ulv(rdev, false);
+ kv_reset_am(rdev);
+
+ kv_update_current_ps(rdev, rdev->pm.dpm.boot_ps);
+}
+
+#if 0
+static int kv_write_smc_soft_register(struct radeon_device *rdev,
+ u16 reg_offset, u32 value)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ return kv_copy_bytes_to_smc(rdev, pi->soft_regs_start + reg_offset,
+ (u8 *)&value, sizeof(u16), pi->sram_end);
+}
+
+static int kv_read_smc_soft_register(struct radeon_device *rdev,
+ u16 reg_offset, u32 *value)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ return kv_read_smc_sram_dword(rdev, pi->soft_regs_start + reg_offset,
+ value, pi->sram_end);
+}
+#endif
+
+static void kv_init_sclk_t(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ pi->low_sclk_interrupt_t = 0;
+}
+
+static int kv_init_fps_limits(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ int ret = 0;
+
+ if (pi->caps_fps) {
+ u16 tmp;
+
+ tmp = 45;
+ pi->fps_high_t = cpu_to_be16(tmp);
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, FpsHighT),
+ (u8 *)&pi->fps_high_t,
+ sizeof(u16), pi->sram_end);
+
+ tmp = 30;
+ pi->fps_low_t = cpu_to_be16(tmp);
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, FpsLowT),
+ (u8 *)&pi->fps_low_t,
+ sizeof(u16), pi->sram_end);
+
+ }
+ return ret;
+}
+
+static void kv_init_powergate_state(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ pi->uvd_power_gated = false;
+ pi->vce_power_gated = false;
+ pi->samu_power_gated = false;
+ pi->acp_power_gated = false;
+
+}
+
+static int kv_enable_uvd_dpm(struct radeon_device *rdev, bool enable)
+{
+ return kv_notify_message_to_smu(rdev, enable ?
+ PPSMC_MSG_UVDDPM_Enable : PPSMC_MSG_UVDDPM_Disable);
+}
+
+#if 0
+static int kv_enable_vce_dpm(struct radeon_device *rdev, bool enable)
+{
+ return kv_notify_message_to_smu(rdev, enable ?
+ PPSMC_MSG_VCEDPM_Enable : PPSMC_MSG_VCEDPM_Disable);
+}
+#endif
+
+static int kv_enable_samu_dpm(struct radeon_device *rdev, bool enable)
+{
+ return kv_notify_message_to_smu(rdev, enable ?
+ PPSMC_MSG_SAMUDPM_Enable : PPSMC_MSG_SAMUDPM_Disable);
+}
+
+static int kv_enable_acp_dpm(struct radeon_device *rdev, bool enable)
+{
+ return kv_notify_message_to_smu(rdev, enable ?
+ PPSMC_MSG_ACPDPM_Enable : PPSMC_MSG_ACPDPM_Disable);
+}
+
+static int kv_update_uvd_dpm(struct radeon_device *rdev, bool gate)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct radeon_uvd_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table;
+ int ret;
+
+ if (!gate) {
+ if (!pi->caps_uvd_dpm || table->count || pi->caps_stable_p_state)
+ pi->uvd_boot_level = table->count - 1;
+ else
+ pi->uvd_boot_level = 0;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, UvdBootLevel),
+ (uint8_t *)&pi->uvd_boot_level,
+ sizeof(u8), pi->sram_end);
+ if (ret)
+ return ret;
+
+ if (!pi->caps_uvd_dpm ||
+ pi->caps_stable_p_state)
+ kv_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_UVDDPM_SetEnabledMask,
+ (1 << pi->uvd_boot_level));
+ }
+
+ return kv_enable_uvd_dpm(rdev, !gate);
+}
+
+#if 0
+static u8 kv_get_vce_boot_level(struct radeon_device *rdev)
+{
+ u8 i;
+ struct radeon_vce_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table;
+
+ for (i = 0; i < table->count; i++) {
+ if (table->entries[i].evclk >= 0) /* XXX */
+ break;
+ }
+
+ return i;
+}
+
+static int kv_update_vce_dpm(struct radeon_device *rdev,
+ struct radeon_ps *radeon_new_state,
+ struct radeon_ps *radeon_current_state)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct radeon_vce_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table;
+ int ret;
+
+ if (radeon_new_state->evclk > 0 && radeon_current_state->evclk == 0) {
+ if (pi->caps_stable_p_state)
+ pi->vce_boot_level = table->count - 1;
+ else
+ pi->vce_boot_level = kv_get_vce_boot_level(rdev);
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, VceBootLevel),
+ (u8 *)&pi->vce_boot_level,
+ sizeof(u8),
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ if (pi->caps_stable_p_state)
+ kv_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_VCEDPM_SetEnabledMask,
+ (1 << pi->vce_boot_level));
+
+ kv_enable_vce_dpm(rdev, true);
+ } else if (radeon_new_state->evclk == 0 && radeon_current_state->evclk > 0) {
+ kv_enable_vce_dpm(rdev, false);
+ }
+
+ return 0;
+}
+#endif
+
+static int kv_update_samu_dpm(struct radeon_device *rdev, bool gate)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct radeon_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table;
+ int ret;
+
+ if (!gate) {
+ if (pi->caps_stable_p_state)
+ pi->samu_boot_level = table->count - 1;
+ else
+ pi->samu_boot_level = 0;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, SamuBootLevel),
+ (u8 *)&pi->samu_boot_level,
+ sizeof(u8),
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ if (pi->caps_stable_p_state)
+ kv_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_SAMUDPM_SetEnabledMask,
+ (1 << pi->samu_boot_level));
+ }
+
+ return kv_enable_samu_dpm(rdev, !gate);
+}
+
+static int kv_update_acp_dpm(struct radeon_device *rdev, bool gate)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct radeon_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table;
+ int ret;
+
+ if (!gate) {
+ if (pi->caps_stable_p_state)
+ pi->acp_boot_level = table->count - 1;
+ else
+ pi->acp_boot_level = 0;
+
+ ret = kv_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, AcpBootLevel),
+ (u8 *)&pi->acp_boot_level,
+ sizeof(u8),
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ if (pi->caps_stable_p_state)
+ kv_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_ACPDPM_SetEnabledMask,
+ (1 << pi->acp_boot_level));
+ }
+
+ return kv_enable_acp_dpm(rdev, !gate);
+}
+
+static void kv_dpm_powergate_uvd(struct radeon_device *rdev, bool gate)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ if (pi->uvd_power_gated == gate)
+ return;
+
+ pi->uvd_power_gated = gate;
+
+ if (gate) {
+ kv_update_uvd_dpm(rdev, true);
+ if (pi->caps_uvd_pg)
+ kv_notify_message_to_smu(rdev, PPSMC_MSG_UVDPowerOFF);
+ } else {
+ if (pi->caps_uvd_pg)
+ kv_notify_message_to_smu(rdev, PPSMC_MSG_UVDPowerON);
+ kv_update_uvd_dpm(rdev, false);
+ }
+}
+
+static void kv_dpm_powergate_vce(struct radeon_device *rdev, bool gate)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ if (pi->vce_power_gated == gate)
+ return;
+
+ pi->vce_power_gated = gate;
+
+ if (gate) {
+ if (pi->caps_vce_pg)
+ kv_notify_message_to_smu(rdev, PPSMC_MSG_VCEPowerOFF);
+ } else {
+ if (pi->caps_vce_pg)
+ kv_notify_message_to_smu(rdev, PPSMC_MSG_VCEPowerON);
+ }
+}
+
+static void kv_dpm_powergate_samu(struct radeon_device *rdev, bool gate)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ if (pi->samu_power_gated == gate)
+ return;
+
+ pi->samu_power_gated = gate;
+
+ if (gate) {
+ kv_update_samu_dpm(rdev, true);
+ if (pi->caps_samu_pg)
+ kv_notify_message_to_smu(rdev, PPSMC_MSG_SAMPowerOFF);
+ } else {
+ if (pi->caps_samu_pg)
+ kv_notify_message_to_smu(rdev, PPSMC_MSG_SAMPowerON);
+ kv_update_samu_dpm(rdev, false);
+ }
+}
+
+static void kv_dpm_powergate_acp(struct radeon_device *rdev, bool gate)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ if (pi->acp_power_gated == gate)
+ return;
+
+ if (rdev->family == CHIP_KABINI)
+ return;
+
+ pi->acp_power_gated = gate;
+
+ if (gate) {
+ kv_update_acp_dpm(rdev, true);
+ if (pi->caps_acp_pg)
+ kv_notify_message_to_smu(rdev, PPSMC_MSG_ACPPowerOFF);
+ } else {
+ if (pi->caps_acp_pg)
+ kv_notify_message_to_smu(rdev, PPSMC_MSG_ACPPowerON);
+ kv_update_acp_dpm(rdev, false);
+ }
+}
+
+static void kv_set_valid_clock_range(struct radeon_device *rdev,
+ struct radeon_ps *new_rps)
+{
+ struct kv_ps *new_ps = kv_get_ps(new_rps);
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 i;
+ struct radeon_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
+
+ if (table && table->count) {
+ for (i = 0; i < pi->graphics_dpm_level_count; i++) {
+ if ((table->entries[i].clk >= new_ps->levels[0].sclk) ||
+ (i == (pi->graphics_dpm_level_count - 1))) {
+ pi->lowest_valid = i;
+ break;
+ }
+ }
+
+ for (i = pi->graphics_dpm_level_count - 1; i >= 0; i--) {
+ if ((table->entries[i].clk <= new_ps->levels[new_ps->num_levels -1].sclk) ||
+ (i == 0)) {
+ pi->highest_valid = i;
+ break;
+ }
+ }
+
+ if (pi->lowest_valid > pi->highest_valid) {
+ if ((new_ps->levels[0].sclk - table->entries[pi->highest_valid].clk) >
+ (table->entries[pi->lowest_valid].clk - new_ps->levels[new_ps->num_levels - 1].sclk))
+ pi->highest_valid = pi->lowest_valid;
+ else
+ pi->lowest_valid = pi->highest_valid;
+ }
+ } else {
+ struct sumo_sclk_voltage_mapping_table *table =
+ &pi->sys_info.sclk_voltage_mapping_table;
+
+ for (i = 0; i < (int)pi->graphics_dpm_level_count; i++) {
+ if (table->entries[i].sclk_frequency >= new_ps->levels[0].sclk ||
+ i == (int)(pi->graphics_dpm_level_count - 1)) {
+ pi->lowest_valid = i;
+ break;
+ }
+ }
+
+ for (i = pi->graphics_dpm_level_count - 1; i >= 0; i--) {
+ if (table->entries[i].sclk_frequency <=
+ new_ps->levels[new_ps->num_levels - 1].sclk ||
+ i == 0) {
+ pi->highest_valid = i;
+ break;
+ }
+ }
+
+ if (pi->lowest_valid > pi->highest_valid) {
+ if ((new_ps->levels[0].sclk -
+ table->entries[pi->highest_valid].sclk_frequency) >
+ (table->entries[pi->lowest_valid].sclk_frequency -
+ new_ps->levels[new_ps->num_levels -1].sclk))
+ pi->highest_valid = pi->lowest_valid;
+ else
+ pi->lowest_valid = pi->highest_valid;
+ }
+ }
+}
+
+static int kv_update_dfs_bypass_settings(struct radeon_device *rdev,
+ struct radeon_ps *new_rps)
+{
+ struct kv_ps *new_ps = kv_get_ps(new_rps);
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ int ret = 0;
+ u8 clk_bypass_cntl;
+
+ if (pi->caps_enable_dfs_bypass) {
+ clk_bypass_cntl = new_ps->need_dfs_bypass ?
+ pi->graphics_level[pi->graphics_boot_level].ClkBypassCntl : 0;
+ ret = kv_copy_bytes_to_smc(rdev,
+ (pi->dpm_table_start +
+ offsetof(SMU7_Fusion_DpmTable, GraphicsLevel) +
+ (pi->graphics_boot_level * sizeof(SMU7_Fusion_GraphicsLevel)) +
+ offsetof(SMU7_Fusion_GraphicsLevel, ClkBypassCntl)),
+ &clk_bypass_cntl,
+ sizeof(u8), pi->sram_end);
+ }
+
+ return ret;
+}
+
+static int kv_enable_nb_dpm(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ int ret = 0;
+
+ if (pi->enable_nb_dpm && !pi->nb_dpm_enabled) {
+ ret = kv_notify_message_to_smu(rdev, PPSMC_MSG_NBDPM_Enable);
+ if (ret == 0)
+ pi->nb_dpm_enabled = true;
+ }
+
+ return ret;
+}
+
+int kv_dpm_pre_set_power_state(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct radeon_ps requested_ps = *rdev->pm.dpm.requested_ps;
+ struct radeon_ps *new_ps = &requested_ps;
+
+ kv_update_requested_ps(rdev, new_ps);
+
+ kv_apply_state_adjust_rules(rdev,
+ &pi->requested_rps,
+ &pi->current_rps);
+
+ return 0;
+}
+
+int kv_dpm_set_power_state(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct radeon_ps *new_ps = &pi->requested_rps;
+ /*struct radeon_ps *old_ps = &pi->current_rps;*/
+ int ret;
+
+ if (rdev->family == CHIP_KABINI) {
+ if (pi->enable_dpm) {
+ kv_set_valid_clock_range(rdev, new_ps);
+ kv_update_dfs_bypass_settings(rdev, new_ps);
+ ret = kv_calculate_ds_divider(rdev);
+ if (ret) {
+ DRM_ERROR("kv_calculate_ds_divider failed\n");
+ return ret;
+ }
+ kv_calculate_nbps_level_settings(rdev);
+ kv_calculate_dpm_settings(rdev);
+ kv_force_lowest_valid(rdev);
+ kv_enable_new_levels(rdev);
+ kv_upload_dpm_settings(rdev);
+ kv_program_nbps_index_settings(rdev, new_ps);
+ kv_unforce_levels(rdev);
+ kv_set_enabled_levels(rdev);
+ kv_force_lowest_valid(rdev);
+ kv_unforce_levels(rdev);
+#if 0
+ ret = kv_update_vce_dpm(rdev, new_ps, old_ps);
+ if (ret) {
+ DRM_ERROR("kv_update_vce_dpm failed\n");
+ return ret;
+ }
+#endif
+ kv_update_uvd_dpm(rdev, false);
+ kv_update_sclk_t(rdev);
+ }
+ } else {
+ if (pi->enable_dpm) {
+ kv_set_valid_clock_range(rdev, new_ps);
+ kv_update_dfs_bypass_settings(rdev, new_ps);
+ ret = kv_calculate_ds_divider(rdev);
+ if (ret) {
+ DRM_ERROR("kv_calculate_ds_divider failed\n");
+ return ret;
+ }
+ kv_calculate_nbps_level_settings(rdev);
+ kv_calculate_dpm_settings(rdev);
+ kv_freeze_sclk_dpm(rdev, true);
+ kv_upload_dpm_settings(rdev);
+ kv_program_nbps_index_settings(rdev, new_ps);
+ kv_freeze_sclk_dpm(rdev, false);
+ kv_set_enabled_levels(rdev);
+#if 0
+ ret = kv_update_vce_dpm(rdev, new_ps, old_ps);
+ if (ret) {
+ DRM_ERROR("kv_update_vce_dpm failed\n");
+ return ret;
+ }
+#endif
+ kv_update_uvd_dpm(rdev, false);
+ kv_update_sclk_t(rdev);
+ kv_enable_nb_dpm(rdev);
+ }
+ }
+ return 0;
+}
+
+void kv_dpm_post_set_power_state(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct radeon_ps *new_ps = &pi->requested_rps;
+
+ kv_update_current_ps(rdev, new_ps);
+}
+
+void kv_dpm_setup_asic(struct radeon_device *rdev)
+{
+ sumo_take_smu_control(rdev, true);
+ kv_init_powergate_state(rdev);
+ kv_init_sclk_t(rdev);
+}
+
+void kv_dpm_reset_asic(struct radeon_device *rdev)
+{
+ kv_force_lowest_valid(rdev);
+ kv_init_graphics_levels(rdev);
+ kv_program_bootup_state(rdev);
+ kv_upload_dpm_settings(rdev);
+ kv_force_lowest_valid(rdev);
+ kv_unforce_levels(rdev);
+}
+
+//XXX use sumo_dpm_display_configuration_changed
+
+static void kv_construct_max_power_limits_table(struct radeon_device *rdev,
+ struct radeon_clock_and_voltage_limits *table)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ if (pi->sys_info.sclk_voltage_mapping_table.num_max_dpm_entries > 0) {
+ int idx = pi->sys_info.sclk_voltage_mapping_table.num_max_dpm_entries - 1;
+ table->sclk =
+ pi->sys_info.sclk_voltage_mapping_table.entries[idx].sclk_frequency;
+ table->vddc =
+ kv_convert_2bit_index_to_voltage(rdev,
+ pi->sys_info.sclk_voltage_mapping_table.entries[idx].vid_2bit);
+ }
+
+ table->mclk = pi->sys_info.nbp_memory_clock[0];
+}
+
+static void kv_patch_voltage_values(struct radeon_device *rdev)
+{
+ int i;
+ struct radeon_uvd_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table;
+
+ if (table->count) {
+ for (i = 0; i < table->count; i++)
+ table->entries[i].v =
+ kv_convert_8bit_index_to_voltage(rdev,
+ table->entries[i].v);
+ }
+
+}
+
+static void kv_construct_boot_state(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ pi->boot_pl.sclk = pi->sys_info.bootup_sclk;
+ pi->boot_pl.vddc_index = pi->sys_info.bootup_nb_voltage_index;
+ pi->boot_pl.ds_divider_index = 0;
+ pi->boot_pl.ss_divider_index = 0;
+ pi->boot_pl.allow_gnb_slow = 1;
+ pi->boot_pl.force_nbp_state = 0;
+ pi->boot_pl.display_wm = 0;
+ pi->boot_pl.vce_wm = 0;
+}
+
+static int kv_force_dpm_lowest(struct radeon_device *rdev)
+{
+ int ret;
+ u32 enable_mask, i;
+
+ ret = kv_dpm_get_enable_mask(rdev, &enable_mask);
+ if (ret)
+ return ret;
+
+ for (i = 0; i < SMU7_MAX_LEVELS_GRAPHICS; i++) {
+ if (enable_mask & (1 << i))
+ break;
+ }
+
+ return kv_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_DPM_ForceState, i);
+}
+
+static u8 kv_get_sleep_divider_id_from_clock(struct radeon_device *rdev,
+ u32 sclk, u32 min_sclk_in_sr)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 i;
+ u32 temp;
+ u32 min = (min_sclk_in_sr > KV_MINIMUM_ENGINE_CLOCK) ?
+ min_sclk_in_sr : KV_MINIMUM_ENGINE_CLOCK;
+
+ if (sclk < min)
+ return 0;
+
+ if (!pi->caps_sclk_ds)
+ return 0;
+
+ for (i = KV_MAX_DEEPSLEEP_DIVIDER_ID; i <= 0; i--) {
+ temp = sclk / sumo_get_sleep_divider_from_id(i);
+ if ((temp >= min) || (i == 0))
+ break;
+ }
+
+ return (u8)i;
+}
+
+static int kv_get_high_voltage_limit(struct radeon_device *rdev, int *limit)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct radeon_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
+ int i;
+
+ if (table && table->count) {
+ for (i = table->count - 1; i >= 0; i--) {
+ if (pi->high_voltage_t &&
+ (kv_convert_8bit_index_to_voltage(rdev, table->entries[i].v) <=
+ pi->high_voltage_t)) {
+ *limit = i;
+ return 0;
+ }
+ }
+ } else {
+ struct sumo_sclk_voltage_mapping_table *table =
+ &pi->sys_info.sclk_voltage_mapping_table;
+
+ for (i = table->num_max_dpm_entries - 1; i >= 0; i--) {
+ if (pi->high_voltage_t &&
+ (kv_convert_2bit_index_to_voltage(rdev, table->entries[i].vid_2bit) <=
+ pi->high_voltage_t)) {
+ *limit = i;
+ return 0;
+ }
+ }
+ }
+
+ *limit = 0;
+ return 0;
+}
+
+static void kv_apply_state_adjust_rules(struct radeon_device *rdev,
+ struct radeon_ps *new_rps,
+ struct radeon_ps *old_rps)
+{
+ struct kv_ps *ps = kv_get_ps(new_rps);
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 min_sclk = 10000; /* ??? */
+ u32 sclk, mclk = 0;
+ int i, limit;
+ bool force_high;
+ struct radeon_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
+ u32 stable_p_state_sclk = 0;
+ struct radeon_clock_and_voltage_limits *max_limits =
+ &rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
+
+ mclk = max_limits->mclk;
+ sclk = min_sclk;
+
+ if (pi->caps_stable_p_state) {
+ stable_p_state_sclk = (max_limits->sclk * 75) / 100;
+
+ for (i = table->count - 1; i >= 0; i++) {
+ if (stable_p_state_sclk >= table->entries[i].clk) {
+ stable_p_state_sclk = table->entries[i].clk;
+ break;
+ }
+ }
+
+ if (i > 0)
+ stable_p_state_sclk = table->entries[0].clk;
+
+ sclk = stable_p_state_sclk;
+ }
+
+ ps->need_dfs_bypass = true;
+
+ for (i = 0; i < ps->num_levels; i++) {
+ if (ps->levels[i].sclk < sclk)
+ ps->levels[i].sclk = sclk;
+ }
+
+ if (table && table->count) {
+ for (i = 0; i < ps->num_levels; i++) {
+ if (pi->high_voltage_t &&
+ (pi->high_voltage_t <
+ kv_convert_8bit_index_to_voltage(rdev, ps->levels[i].vddc_index))) {
+ kv_get_high_voltage_limit(rdev, &limit);
+ ps->levels[i].sclk = table->entries[limit].clk;
+ }
+ }
+ } else {
+ struct sumo_sclk_voltage_mapping_table *table =
+ &pi->sys_info.sclk_voltage_mapping_table;
+
+ for (i = 0; i < ps->num_levels; i++) {
+ if (pi->high_voltage_t &&
+ (pi->high_voltage_t <
+ kv_convert_8bit_index_to_voltage(rdev, ps->levels[i].vddc_index))) {
+ kv_get_high_voltage_limit(rdev, &limit);
+ ps->levels[i].sclk = table->entries[limit].sclk_frequency;
+ }
+ }
+ }
+
+ if (pi->caps_stable_p_state) {
+ for (i = 0; i < ps->num_levels; i++) {
+ ps->levels[i].sclk = stable_p_state_sclk;
+ }
+ }
+
+ pi->video_start = new_rps->dclk || new_rps->vclk;
+
+ if ((new_rps->class & ATOM_PPLIB_CLASSIFICATION_UI_MASK) ==
+ ATOM_PPLIB_CLASSIFICATION_UI_BATTERY)
+ pi->battery_state = true;
+ else
+ pi->battery_state = false;
+
+ if (rdev->family == CHIP_KABINI) {
+ ps->dpm0_pg_nb_ps_lo = 0x1;
+ ps->dpm0_pg_nb_ps_hi = 0x0;
+ ps->dpmx_nb_ps_lo = 0x1;
+ ps->dpmx_nb_ps_hi = 0x0;
+ } else {
+ ps->dpm0_pg_nb_ps_lo = 0x1;
+ ps->dpm0_pg_nb_ps_hi = 0x0;
+ ps->dpmx_nb_ps_lo = 0x2;
+ ps->dpmx_nb_ps_hi = 0x1;
+
+ if (pi->sys_info.nb_dpm_enable && pi->battery_state) {
+ force_high = (mclk >= pi->sys_info.nbp_memory_clock[3]) ||
+ pi->video_start || (rdev->pm.dpm.new_active_crtc_count >= 3) ||
+ pi->disable_nb_ps3_in_battery;
+ ps->dpm0_pg_nb_ps_lo = force_high ? 0x2 : 0x3;
+ ps->dpm0_pg_nb_ps_hi = 0x2;
+ ps->dpmx_nb_ps_lo = force_high ? 0x2 : 0x3;
+ ps->dpmx_nb_ps_hi = 0x2;
+ }
+ }
+}
+
+static void kv_dpm_power_level_enabled_for_throttle(struct radeon_device *rdev,
+ u32 index, bool enable)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ pi->graphics_level[index].EnabledForThrottle = enable ? 1 : 0;
+}
+
+static int kv_calculate_ds_divider(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 sclk_in_sr = 10000; /* ??? */
+ u32 i;
+
+ if (pi->lowest_valid > pi->highest_valid)
+ return -EINVAL;
+
+ for (i = pi->lowest_valid; i <= pi->highest_valid; i++) {
+ pi->graphics_level[i].DeepSleepDivId =
+ kv_get_sleep_divider_id_from_clock(rdev,
+ be32_to_cpu(pi->graphics_level[i].SclkFrequency),
+ sclk_in_sr);
+ }
+ return 0;
+}
+
+static int kv_calculate_nbps_level_settings(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 i;
+ bool force_high;
+ struct radeon_clock_and_voltage_limits *max_limits =
+ &rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
+ u32 mclk = max_limits->mclk;
+
+ if (pi->lowest_valid > pi->highest_valid)
+ return -EINVAL;
+
+ if (rdev->family == CHIP_KABINI) {
+ for (i = pi->lowest_valid; i <= pi->highest_valid; i++) {
+ pi->graphics_level[i].GnbSlow = 1;
+ pi->graphics_level[i].ForceNbPs1 = 0;
+ pi->graphics_level[i].UpH = 0;
+ }
+
+ if (!pi->sys_info.nb_dpm_enable)
+ return 0;
+
+ force_high = ((mclk >= pi->sys_info.nbp_memory_clock[3]) ||
+ (rdev->pm.dpm.new_active_crtc_count >= 3) || pi->video_start);
+
+ if (force_high) {
+ for (i = pi->lowest_valid; i <= pi->highest_valid; i++)
+ pi->graphics_level[i].GnbSlow = 0;
+ } else {
+ if (pi->battery_state)
+ pi->graphics_level[0].ForceNbPs1 = 1;
+
+ pi->graphics_level[1].GnbSlow = 0;
+ pi->graphics_level[2].GnbSlow = 0;
+ pi->graphics_level[3].GnbSlow = 0;
+ pi->graphics_level[4].GnbSlow = 0;
+ }
+ } else {
+ for (i = pi->lowest_valid; i <= pi->highest_valid; i++) {
+ pi->graphics_level[i].GnbSlow = 1;
+ pi->graphics_level[i].ForceNbPs1 = 0;
+ pi->graphics_level[i].UpH = 0;
+ }
+
+ if (pi->sys_info.nb_dpm_enable && pi->battery_state) {
+ pi->graphics_level[pi->lowest_valid].UpH = 0x28;
+ pi->graphics_level[pi->lowest_valid].GnbSlow = 0;
+ if (pi->lowest_valid != pi->highest_valid)
+ pi->graphics_level[pi->lowest_valid].ForceNbPs1 = 1;
+ }
+ }
+ return 0;
+}
+
+static int kv_calculate_dpm_settings(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 i;
+
+ if (pi->lowest_valid > pi->highest_valid)
+ return -EINVAL;
+
+ for (i = pi->lowest_valid; i <= pi->highest_valid; i++)
+ pi->graphics_level[i].DisplayWatermark = (i == pi->highest_valid) ? 1 : 0;
+
+ return 0;
+}
+
+static void kv_init_graphics_levels(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 i;
+ struct radeon_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
+
+ if (table && table->count) {
+ u32 vid_2bit;
+
+ pi->graphics_dpm_level_count = 0;
+ for (i = 0; i < table->count; i++) {
+ if (pi->high_voltage_t &&
+ (pi->high_voltage_t <
+ kv_convert_8bit_index_to_voltage(rdev, table->entries[i].v)))
+ break;
+
+ kv_set_divider_value(rdev, i, table->entries[i].clk);
+ vid_2bit = sumo_convert_vid7_to_vid2(rdev,
+ &pi->sys_info.vid_mapping_table,
+ table->entries[i].v);
+ kv_set_vid(rdev, i, vid_2bit);
+ kv_set_at(rdev, i, pi->at[i]);
+ kv_dpm_power_level_enabled_for_throttle(rdev, i, true);
+ pi->graphics_dpm_level_count++;
+ }
+ } else {
+ struct sumo_sclk_voltage_mapping_table *table =
+ &pi->sys_info.sclk_voltage_mapping_table;
+
+ pi->graphics_dpm_level_count = 0;
+ for (i = 0; i < table->num_max_dpm_entries; i++) {
+ if (pi->high_voltage_t &&
+ pi->high_voltage_t <
+ kv_convert_2bit_index_to_voltage(rdev, table->entries[i].vid_2bit))
+ break;
+
+ kv_set_divider_value(rdev, i, table->entries[i].sclk_frequency);
+ kv_set_vid(rdev, i, table->entries[i].vid_2bit);
+ kv_set_at(rdev, i, pi->at[i]);
+ kv_dpm_power_level_enabled_for_throttle(rdev, i, true);
+ pi->graphics_dpm_level_count++;
+ }
+ }
+
+ for (i = 0; i < SMU7_MAX_LEVELS_GRAPHICS; i++)
+ kv_dpm_power_level_enable(rdev, i, false);
+}
+
+static void kv_enable_new_levels(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 i;
+
+ for (i = 0; i < SMU7_MAX_LEVELS_GRAPHICS; i++) {
+ if (i >= pi->lowest_valid && i <= pi->highest_valid)
+ kv_dpm_power_level_enable(rdev, i, true);
+ }
+}
+
+static int kv_set_enabled_levels(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 i, new_mask = 0;
+
+ for (i = pi->lowest_valid; i <= pi->highest_valid; i++)
+ new_mask |= (1 << i);
+
+ return kv_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_SCLKDPM_SetEnabledMask,
+ new_mask);
+}
+
+static void kv_program_nbps_index_settings(struct radeon_device *rdev,
+ struct radeon_ps *new_rps)
+{
+ struct kv_ps *new_ps = kv_get_ps(new_rps);
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 nbdpmconfig1;
+
+ if (rdev->family == CHIP_KABINI)
+ return;
+
+ if (pi->sys_info.nb_dpm_enable) {
+ nbdpmconfig1 = RREG32_SMC(NB_DPM_CONFIG_1);
+ nbdpmconfig1 &= ~(Dpm0PgNbPsLo_MASK | Dpm0PgNbPsHi_MASK |
+ DpmXNbPsLo_MASK | DpmXNbPsHi_MASK);
+ nbdpmconfig1 |= (Dpm0PgNbPsLo(new_ps->dpm0_pg_nb_ps_lo) |
+ Dpm0PgNbPsHi(new_ps->dpm0_pg_nb_ps_hi) |
+ DpmXNbPsLo(new_ps->dpmx_nb_ps_lo) |
+ DpmXNbPsHi(new_ps->dpmx_nb_ps_hi));
+ WREG32_SMC(NB_DPM_CONFIG_1, nbdpmconfig1);
+ }
+}
+
+static int kv_set_thermal_temperature_range(struct radeon_device *rdev,
+ int min_temp, int max_temp)
+{
+ int low_temp = 0 * 1000;
+ int high_temp = 255 * 1000;
+ u32 tmp;
+
+ if (low_temp < min_temp)
+ low_temp = min_temp;
+ if (high_temp > max_temp)
+ high_temp = max_temp;
+ if (high_temp < low_temp) {
+ DRM_ERROR("invalid thermal range: %d - %d\n", low_temp, high_temp);
+ return -EINVAL;
+ }
+
+ tmp = RREG32_SMC(CG_THERMAL_INT_CTRL);
+ tmp &= ~(DIG_THERM_INTH_MASK | DIG_THERM_INTL_MASK);
+ tmp |= (DIG_THERM_INTH(49 + (high_temp / 1000)) |
+ DIG_THERM_INTL(49 + (low_temp / 1000)));
+ WREG32_SMC(CG_THERMAL_INT_CTRL, tmp);
+
+ rdev->pm.dpm.thermal.min_temp = low_temp;
+ rdev->pm.dpm.thermal.max_temp = high_temp;
+
+ return 0;
+}
+
+union igp_info {
+ struct _ATOM_INTEGRATED_SYSTEM_INFO info;
+ struct _ATOM_INTEGRATED_SYSTEM_INFO_V2 info_2;
+ struct _ATOM_INTEGRATED_SYSTEM_INFO_V5 info_5;
+ struct _ATOM_INTEGRATED_SYSTEM_INFO_V6 info_6;
+ struct _ATOM_INTEGRATED_SYSTEM_INFO_V1_7 info_7;
+ struct _ATOM_INTEGRATED_SYSTEM_INFO_V1_8 info_8;
+};
+
+static int kv_parse_sys_info_table(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct radeon_mode_info *mode_info = &rdev->mode_info;
+ int index = GetIndexIntoMasterTable(DATA, IntegratedSystemInfo);
+ union igp_info *igp_info;
+ u8 frev, crev;
+ u16 data_offset;
+ int i;
+
+ if (atom_parse_data_header(mode_info->atom_context, index, NULL,
+ &frev, &crev, &data_offset)) {
+ igp_info = (union igp_info *)(mode_info->atom_context->bios +
+ data_offset);
+
+ if (crev != 8) {
+ DRM_ERROR("Unsupported IGP table: %d %d\n", frev, crev);
+ return -EINVAL;
+ }
+ pi->sys_info.bootup_sclk = le32_to_cpu(igp_info->info_8.ulBootUpEngineClock);
+ pi->sys_info.bootup_uma_clk = le32_to_cpu(igp_info->info_8.ulBootUpUMAClock);
+ pi->sys_info.bootup_nb_voltage_index =
+ le16_to_cpu(igp_info->info_8.usBootUpNBVoltage);
+ if (igp_info->info_8.ucHtcTmpLmt == 0)
+ pi->sys_info.htc_tmp_lmt = 203;
+ else
+ pi->sys_info.htc_tmp_lmt = igp_info->info_8.ucHtcTmpLmt;
+ if (igp_info->info_8.ucHtcHystLmt == 0)
+ pi->sys_info.htc_hyst_lmt = 5;
+ else
+ pi->sys_info.htc_hyst_lmt = igp_info->info_8.ucHtcHystLmt;
+ if (pi->sys_info.htc_tmp_lmt <= pi->sys_info.htc_hyst_lmt) {
+ DRM_ERROR("The htcTmpLmt should be larger than htcHystLmt.\n");
+ }
+
+ if (le32_to_cpu(igp_info->info_8.ulSystemConfig) & (1 << 3))
+ pi->sys_info.nb_dpm_enable = true;
+ else
+ pi->sys_info.nb_dpm_enable = false;
+
+ for (i = 0; i < KV_NUM_NBPSTATES; i++) {
+ pi->sys_info.nbp_memory_clock[i] =
+ le32_to_cpu(igp_info->info_8.ulNbpStateMemclkFreq[i]);
+ pi->sys_info.nbp_n_clock[i] =
+ le32_to_cpu(igp_info->info_8.ulNbpStateNClkFreq[i]);
+ }
+ if (le32_to_cpu(igp_info->info_8.ulGPUCapInfo) &
+ SYS_INFO_GPUCAPS__ENABEL_DFS_BYPASS)
+ pi->caps_enable_dfs_bypass = true;
+
+ sumo_construct_sclk_voltage_mapping_table(rdev,
+ &pi->sys_info.sclk_voltage_mapping_table,
+ igp_info->info_8.sAvail_SCLK);
+
+ sumo_construct_vid_mapping_table(rdev,
+ &pi->sys_info.vid_mapping_table,
+ igp_info->info_8.sAvail_SCLK);
+
+ kv_construct_max_power_limits_table(rdev,
+ &rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac);
+ }
+ return 0;
+}
+
+union power_info {
+ struct _ATOM_POWERPLAY_INFO info;
+ struct _ATOM_POWERPLAY_INFO_V2 info_2;
+ struct _ATOM_POWERPLAY_INFO_V3 info_3;
+ struct _ATOM_PPLIB_POWERPLAYTABLE pplib;
+ struct _ATOM_PPLIB_POWERPLAYTABLE2 pplib2;
+ struct _ATOM_PPLIB_POWERPLAYTABLE3 pplib3;
+};
+
+union pplib_clock_info {
+ struct _ATOM_PPLIB_R600_CLOCK_INFO r600;
+ struct _ATOM_PPLIB_RS780_CLOCK_INFO rs780;
+ struct _ATOM_PPLIB_EVERGREEN_CLOCK_INFO evergreen;
+ struct _ATOM_PPLIB_SUMO_CLOCK_INFO sumo;
+};
+
+union pplib_power_state {
+ struct _ATOM_PPLIB_STATE v1;
+ struct _ATOM_PPLIB_STATE_V2 v2;
+};
+
+static void kv_patch_boot_state(struct radeon_device *rdev,
+ struct kv_ps *ps)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ ps->num_levels = 1;
+ ps->levels[0] = pi->boot_pl;
+}
+
+static void kv_parse_pplib_non_clock_info(struct radeon_device *rdev,
+ struct radeon_ps *rps,
+ struct _ATOM_PPLIB_NONCLOCK_INFO *non_clock_info,
+ u8 table_rev)
+{
+ struct kv_ps *ps = kv_get_ps(rps);
+
+ rps->caps = le32_to_cpu(non_clock_info->ulCapsAndSettings);
+ rps->class = le16_to_cpu(non_clock_info->usClassification);
+ rps->class2 = le16_to_cpu(non_clock_info->usClassification2);
+
+ if (ATOM_PPLIB_NONCLOCKINFO_VER1 < table_rev) {
+ rps->vclk = le32_to_cpu(non_clock_info->ulVCLK);
+ rps->dclk = le32_to_cpu(non_clock_info->ulDCLK);
+ } else {
+ rps->vclk = 0;
+ rps->dclk = 0;
+ }
+
+ if (rps->class & ATOM_PPLIB_CLASSIFICATION_BOOT) {
+ rdev->pm.dpm.boot_ps = rps;
+ kv_patch_boot_state(rdev, ps);
+ }
+ if (rps->class & ATOM_PPLIB_CLASSIFICATION_UVDSTATE)
+ rdev->pm.dpm.uvd_ps = rps;
+}
+
+static void kv_parse_pplib_clock_info(struct radeon_device *rdev,
+ struct radeon_ps *rps, int index,
+ union pplib_clock_info *clock_info)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct kv_ps *ps = kv_get_ps(rps);
+ struct kv_pl *pl = &ps->levels[index];
+ u32 sclk;
+
+ sclk = le16_to_cpu(clock_info->sumo.usEngineClockLow);
+ sclk |= clock_info->sumo.ucEngineClockHigh << 16;
+ pl->sclk = sclk;
+ pl->vddc_index = clock_info->sumo.vddcIndex;
+
+ ps->num_levels = index + 1;
+
+ if (pi->caps_sclk_ds) {
+ pl->ds_divider_index = 5;
+ pl->ss_divider_index = 5;
+ }
+}
+
+static int kv_parse_power_table(struct radeon_device *rdev)
+{
+ struct radeon_mode_info *mode_info = &rdev->mode_info;
+ struct _ATOM_PPLIB_NONCLOCK_INFO *non_clock_info;
+ union pplib_power_state *power_state;
+ int i, j, k, non_clock_array_index, clock_array_index;
+ union pplib_clock_info *clock_info;
+ struct _StateArray *state_array;
+ struct _ClockInfoArray *clock_info_array;
+ struct _NonClockInfoArray *non_clock_info_array;
+ union power_info *power_info;
+ int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
+ u16 data_offset;
+ u8 frev, crev;
+ u8 *power_state_offset;
+ struct kv_ps *ps;
+
+ if (!atom_parse_data_header(mode_info->atom_context, index, NULL,
+ &frev, &crev, &data_offset))
+ return -EINVAL;
+ power_info = (union power_info *)(mode_info->atom_context->bios + data_offset);
+
+ state_array = (struct _StateArray *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(power_info->pplib.usStateArrayOffset));
+ clock_info_array = (struct _ClockInfoArray *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(power_info->pplib.usClockInfoArrayOffset));
+ non_clock_info_array = (struct _NonClockInfoArray *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(power_info->pplib.usNonClockInfoArrayOffset));
+
+ rdev->pm.dpm.ps = kzalloc(sizeof(struct radeon_ps) *
+ state_array->ucNumEntries, GFP_KERNEL);
+ if (!rdev->pm.dpm.ps)
+ return -ENOMEM;
+ power_state_offset = (u8 *)state_array->states;
+ rdev->pm.dpm.platform_caps = le32_to_cpu(power_info->pplib.ulPlatformCaps);
+ rdev->pm.dpm.backbias_response_time = le16_to_cpu(power_info->pplib.usBackbiasTime);
+ rdev->pm.dpm.voltage_response_time = le16_to_cpu(power_info->pplib.usVoltageTime);
+ for (i = 0; i < state_array->ucNumEntries; i++) {
+ power_state = (union pplib_power_state *)power_state_offset;
+ non_clock_array_index = power_state->v2.nonClockInfoIndex;
+ non_clock_info = (struct _ATOM_PPLIB_NONCLOCK_INFO *)
+ &non_clock_info_array->nonClockInfo[non_clock_array_index];
+ if (!rdev->pm.power_state[i].clock_info)
+ return -EINVAL;
+ ps = kzalloc(sizeof(struct kv_ps), GFP_KERNEL);
+ if (ps == NULL) {
+ kfree(rdev->pm.dpm.ps);
+ return -ENOMEM;
+ }
+ rdev->pm.dpm.ps[i].ps_priv = ps;
+ k = 0;
+ for (j = 0; j < power_state->v2.ucNumDPMLevels; j++) {
+ clock_array_index = power_state->v2.clockInfoIndex[j];
+ if (clock_array_index >= clock_info_array->ucNumEntries)
+ continue;
+ if (k >= SUMO_MAX_HARDWARE_POWERLEVELS)
+ break;
+ clock_info = (union pplib_clock_info *)
+ &clock_info_array->clockInfo[clock_array_index * clock_info_array->ucEntrySize];
+ kv_parse_pplib_clock_info(rdev,
+ &rdev->pm.dpm.ps[i], k,
+ clock_info);
+ k++;
+ }
+ kv_parse_pplib_non_clock_info(rdev, &rdev->pm.dpm.ps[i],
+ non_clock_info,
+ non_clock_info_array->ucEntrySize);
+ power_state_offset += 2 + power_state->v2.ucNumDPMLevels;
+ }
+ rdev->pm.dpm.num_ps = state_array->ucNumEntries;
+ return 0;
+}
+
+int kv_dpm_init(struct radeon_device *rdev)
+{
+ struct kv_power_info *pi;
+ int ret, i;
+
+ pi = kzalloc(sizeof(struct kv_power_info), GFP_KERNEL);
+ if (pi == NULL)
+ return -ENOMEM;
+ rdev->pm.dpm.priv = pi;
+
+ ret = r600_parse_extended_power_table(rdev);
+ if (ret)
+ return ret;
+
+ for (i = 0; i < SUMO_MAX_HARDWARE_POWERLEVELS; i++)
+ pi->at[i] = TRINITY_AT_DFLT;
+
+ pi->sram_end = SMC_RAM_END;
+
+ if (rdev->family == CHIP_KABINI)
+ pi->high_voltage_t = 4001;
+
+ pi->enable_nb_dpm = true;
+
+ pi->caps_power_containment = true;
+ pi->caps_cac = true;
+ pi->enable_didt = false;
+ if (pi->enable_didt) {
+ pi->caps_sq_ramping = true;
+ pi->caps_db_ramping = true;
+ pi->caps_td_ramping = true;
+ pi->caps_tcp_ramping = true;
+ }
+
+ pi->caps_sclk_ds = true;
+ pi->enable_auto_thermal_throttling = true;
+ pi->disable_nb_ps3_in_battery = false;
+ pi->bapm_enable = true;
+ pi->voltage_drop_t = 0;
+ pi->caps_sclk_throttle_low_notification = false;
+ pi->caps_fps = false; /* true? */
+ pi->caps_uvd_pg = false; /* XXX */
+ pi->caps_uvd_dpm = true;
+ pi->caps_vce_pg = false;
+ pi->caps_samu_pg = false;
+ pi->caps_acp_pg = false;
+ pi->caps_stable_p_state = false;
+
+ ret = kv_parse_sys_info_table(rdev);
+ if (ret)
+ return ret;
+
+ kv_patch_voltage_values(rdev);
+ kv_construct_boot_state(rdev);
+
+ ret = kv_parse_power_table(rdev);
+ if (ret)
+ return ret;
+
+ pi->enable_dpm = true;
+
+ return 0;
+}
+
+void kv_dpm_print_power_state(struct radeon_device *rdev,
+ struct radeon_ps *rps)
+{
+ int i;
+ struct kv_ps *ps = kv_get_ps(rps);
+
+ r600_dpm_print_class_info(rps->class, rps->class2);
+ r600_dpm_print_cap_info(rps->caps);
+ printk("\tuvd vclk: %d dclk: %d\n", rps->vclk, rps->dclk);
+ for (i = 0; i < ps->num_levels; i++) {
+ struct kv_pl *pl = &ps->levels[i];
+ printk("\t\tpower level %d sclk: %u vddc: %u\n",
+ i, pl->sclk,
+ kv_convert_8bit_index_to_voltage(rdev, pl->vddc_index));
+ }
+ r600_dpm_print_ps_status(rdev, rps);
+}
+
+void kv_dpm_fini(struct radeon_device *rdev)
+{
+ int i;
+
+ for (i = 0; i < rdev->pm.dpm.num_ps; i++) {
+ kfree(rdev->pm.dpm.ps[i].ps_priv);
+ }
+ kfree(rdev->pm.dpm.ps);
+ kfree(rdev->pm.dpm.priv);
+ r600_free_extended_power_table(rdev);
+}
+
+void kv_dpm_display_configuration_changed(struct radeon_device *rdev)
+{
+
+}
+
+u32 kv_dpm_get_sclk(struct radeon_device *rdev, bool low)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ struct kv_ps *requested_state = kv_get_ps(&pi->requested_rps);
+
+ if (low)
+ return requested_state->levels[0].sclk;
+ else
+ return requested_state->levels[requested_state->num_levels - 1].sclk;
+}
+
+u32 kv_dpm_get_mclk(struct radeon_device *rdev, bool low)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+
+ return pi->sys_info.bootup_uma_clk;
+}
+
diff --git a/drivers/gpu/drm/radeon/kv_dpm.h b/drivers/gpu/drm/radeon/kv_dpm.h
new file mode 100644
index 0000000..32bb079
--- /dev/null
+++ b/drivers/gpu/drm/radeon/kv_dpm.h
@@ -0,0 +1,199 @@
+/*
+ * Copyright 2013 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef __KV_DPM_H__
+#define __KV_DPM_H__
+
+#define SMU__NUM_SCLK_DPM_STATE 8
+#define SMU__NUM_MCLK_DPM_LEVELS 4
+#define SMU__NUM_LCLK_DPM_LEVELS 8
+#define SMU__NUM_PCIE_DPM_LEVELS 0 /* ??? */
+#include "smu7_fusion.h"
+#include "trinity_dpm.h"
+#include "ppsmc.h"
+
+#define KV_NUM_NBPSTATES 4
+
+enum kv_pt_config_reg_type {
+ KV_CONFIGREG_MMR = 0,
+ KV_CONFIGREG_SMC_IND,
+ KV_CONFIGREG_DIDT_IND,
+ KV_CONFIGREG_CACHE,
+ KV_CONFIGREG_MAX
+};
+
+struct kv_pt_config_reg {
+ u32 offset;
+ u32 mask;
+ u32 shift;
+ u32 value;
+ enum kv_pt_config_reg_type type;
+};
+
+struct kv_lcac_config_values {
+ u32 block_id;
+ u32 signal_id;
+ u32 t;
+};
+
+struct kv_lcac_config_reg {
+ u32 cntl;
+ u32 block_mask;
+ u32 block_shift;
+ u32 signal_mask;
+ u32 signal_shift;
+ u32 t_mask;
+ u32 t_shift;
+ u32 enable_mask;
+ u32 enable_shift;
+};
+
+struct kv_pl {
+ u32 sclk;
+ u8 vddc_index;
+ u8 ds_divider_index;
+ u8 ss_divider_index;
+ u8 allow_gnb_slow;
+ u8 force_nbp_state;
+ u8 display_wm;
+ u8 vce_wm;
+};
+
+struct kv_ps {
+ struct kv_pl levels[SUMO_MAX_HARDWARE_POWERLEVELS];
+ u32 num_levels;
+ bool need_dfs_bypass;
+ u8 dpm0_pg_nb_ps_lo;
+ u8 dpm0_pg_nb_ps_hi;
+ u8 dpmx_nb_ps_lo;
+ u8 dpmx_nb_ps_hi;
+};
+
+struct kv_sys_info {
+ u32 bootup_uma_clk;
+ u32 bootup_sclk;
+ u32 dentist_vco_freq;
+ u32 nb_dpm_enable;
+ u32 nbp_memory_clock[KV_NUM_NBPSTATES];
+ u32 nbp_n_clock[KV_NUM_NBPSTATES];
+ u16 bootup_nb_voltage_index;
+ u8 htc_tmp_lmt;
+ u8 htc_hyst_lmt;
+ struct sumo_sclk_voltage_mapping_table sclk_voltage_mapping_table;
+ struct sumo_vid_mapping_table vid_mapping_table;
+ u32 uma_channel_number;
+};
+
+struct kv_power_info {
+ u32 at[SUMO_MAX_HARDWARE_POWERLEVELS];
+ u32 voltage_drop_t;
+ struct kv_sys_info sys_info;
+ struct kv_pl boot_pl;
+ bool enable_nb_ps_policy;
+ bool disable_nb_ps3_in_battery;
+ bool video_start;
+ bool battery_state;
+ u32 lowest_valid;
+ u32 highest_valid;
+ u16 high_voltage_t;
+ bool cac_enabled;
+ bool bapm_enable;
+ /* smc offsets */
+ u32 sram_end;
+ u32 dpm_table_start;
+ u32 soft_regs_start;
+ /* dpm SMU tables */
+ u8 graphics_dpm_level_count;
+ u8 uvd_level_count;
+ u8 vce_level_count;
+ u8 acp_level_count;
+ u8 samu_level_count;
+ u16 fps_high_t;
+ SMU7_Fusion_GraphicsLevel graphics_level[SMU__NUM_SCLK_DPM_STATE];
+ SMU7_Fusion_ACPILevel acpi_level;
+ SMU7_Fusion_UvdLevel uvd_level[SMU7_MAX_LEVELS_UVD];
+ SMU7_Fusion_ExtClkLevel vce_level[SMU7_MAX_LEVELS_VCE];
+ SMU7_Fusion_ExtClkLevel acp_level[SMU7_MAX_LEVELS_ACP];
+ SMU7_Fusion_ExtClkLevel samu_level[SMU7_MAX_LEVELS_SAMU];
+ u8 uvd_boot_level;
+ u8 vce_boot_level;
+ u8 acp_boot_level;
+ u8 samu_boot_level;
+ u8 uvd_interval;
+ u8 vce_interval;
+ u8 acp_interval;
+ u8 samu_interval;
+ u8 graphics_boot_level;
+ u8 graphics_interval;
+ u8 graphics_therm_throttle_enable;
+ u8 graphics_voltage_change_enable;
+ u8 graphics_clk_slow_enable;
+ u8 graphics_clk_slow_divider;
+ u8 fps_low_t;
+ u32 low_sclk_interrupt_t;
+ bool uvd_power_gated;
+ bool vce_power_gated;
+ bool acp_power_gated;
+ bool samu_power_gated;
+ bool nb_dpm_enabled;
+ /* flags */
+ bool enable_didt;
+ bool enable_dpm;
+ bool enable_auto_thermal_throttling;
+ bool enable_nb_dpm;
+ /* caps */
+ bool caps_cac;
+ bool caps_power_containment;
+ bool caps_sq_ramping;
+ bool caps_db_ramping;
+ bool caps_td_ramping;
+ bool caps_tcp_ramping;
+ bool caps_sclk_throttle_low_notification;
+ bool caps_fps;
+ bool caps_uvd_dpm;
+ bool caps_uvd_pg;
+ bool caps_vce_pg;
+ bool caps_samu_pg;
+ bool caps_acp_pg;
+ bool caps_stable_p_state;
+ bool caps_enable_dfs_bypass;
+ bool caps_sclk_ds;
+ struct radeon_ps current_rps;
+ struct kv_ps current_ps;
+ struct radeon_ps requested_rps;
+ struct kv_ps requested_ps;
+};
+
+
+/* kv_smc.c */
+int kv_notify_message_to_smu(struct radeon_device *rdev, u32 id);
+int kv_dpm_get_enable_mask(struct radeon_device *rdev, u32 *enable_mask);
+int kv_send_msg_to_smc_with_parameter(struct radeon_device *rdev,
+ PPSMC_Msg msg, u32 parameter);
+int kv_read_smc_sram_dword(struct radeon_device *rdev, u32 smc_address,
+ u32 *value, u32 limit);
+int kv_smc_dpm_enable(struct radeon_device *rdev, bool enable);
+int kv_copy_bytes_to_smc(struct radeon_device *rdev,
+ u32 smc_start_address,
+ const u8 *src, u32 byte_count, u32 limit);
+
+#endif
diff --git a/drivers/gpu/drm/radeon/kv_smc.c b/drivers/gpu/drm/radeon/kv_smc.c
new file mode 100644
index 0000000..34a226d
--- /dev/null
+++ b/drivers/gpu/drm/radeon/kv_smc.c
@@ -0,0 +1,207 @@
+/*
+ * Copyright 2013 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Alex Deucher
+ */
+
+#include "drmP.h"
+#include "radeon.h"
+#include "cikd.h"
+#include "kv_dpm.h"
+
+int kv_notify_message_to_smu(struct radeon_device *rdev, u32 id)
+{
+ u32 i;
+ u32 tmp = 0;
+
+ WREG32(SMC_MESSAGE_0, id & SMC_MSG_MASK);
+
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ if ((RREG32(SMC_RESP_0) & SMC_RESP_MASK) != 0)
+ break;
+ udelay(1);
+ }
+ tmp = RREG32(SMC_RESP_0) & SMC_RESP_MASK;
+
+ if (tmp != 1) {
+ if (tmp == 0xFF)
+ return -EINVAL;
+ else if (tmp == 0xFE)
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+int kv_dpm_get_enable_mask(struct radeon_device *rdev, u32 *enable_mask)
+{
+ int ret;
+
+ ret = kv_notify_message_to_smu(rdev, PPSMC_MSG_SCLKDPM_GetEnabledMask);
+
+ if (ret == 0)
+ *enable_mask = RREG32_SMC(SMC_SYSCON_MSG_ARG_0);
+
+ return ret;
+}
+
+int kv_send_msg_to_smc_with_parameter(struct radeon_device *rdev,
+ PPSMC_Msg msg, u32 parameter)
+{
+
+ WREG32(SMC_MSG_ARG_0, parameter);
+
+ return kv_notify_message_to_smu(rdev, msg);
+}
+
+static int kv_set_smc_sram_address(struct radeon_device *rdev,
+ u32 smc_address, u32 limit)
+{
+ if (smc_address & 3)
+ return -EINVAL;
+ if ((smc_address + 3) > limit)
+ return -EINVAL;
+
+ WREG32(SMC_IND_INDEX_0, smc_address);
+ WREG32_P(SMC_IND_ACCESS_CNTL, 0, ~AUTO_INCREMENT_IND_0);
+
+ return 0;
+}
+
+int kv_read_smc_sram_dword(struct radeon_device *rdev, u32 smc_address,
+ u32 *value, u32 limit)
+{
+ int ret;
+
+ ret = kv_set_smc_sram_address(rdev, smc_address, limit);
+ if (ret)
+ return ret;
+
+ *value = RREG32(SMC_IND_DATA_0);
+ return 0;
+}
+
+int kv_smc_dpm_enable(struct radeon_device *rdev, bool enable)
+{
+ if (enable)
+ return kv_notify_message_to_smu(rdev, PPSMC_MSG_DPM_Enable);
+ else
+ return kv_notify_message_to_smu(rdev, PPSMC_MSG_DPM_Disable);
+}
+
+int kv_copy_bytes_to_smc(struct radeon_device *rdev,
+ u32 smc_start_address,
+ const u8 *src, u32 byte_count, u32 limit)
+{
+ int ret;
+ u32 data, original_data, addr, extra_shift, t_byte, count, mask;
+
+ if ((smc_start_address + byte_count) > limit)
+ return -EINVAL;
+
+ addr = smc_start_address;
+ t_byte = addr & 3;
+
+ /* RMW for the initial bytes */
+ if (t_byte != 0) {
+ addr -= t_byte;
+
+ ret = kv_set_smc_sram_address(rdev, addr, limit);
+ if (ret)
+ return ret;
+
+ original_data = RREG32(SMC_IND_DATA_0);
+
+ data = 0;
+ mask = 0;
+ count = 4;
+ while (count > 0) {
+ if (t_byte > 0) {
+ mask = (mask << 8) | 0xff;
+ t_byte--;
+ } else if (byte_count > 0) {
+ data = (data << 8) + *src++;
+ byte_count--;
+ mask <<= 8;
+ } else {
+ data <<= 8;
+ mask = (mask << 8) | 0xff;
+ }
+ count--;
+ }
+
+ data |= original_data & mask;
+
+ ret = kv_set_smc_sram_address(rdev, addr, limit);
+ if (ret)
+ return ret;
+
+ WREG32(SMC_IND_DATA_0, data);
+
+ addr += 4;
+ }
+
+ while (byte_count >= 4) {
+ /* SMC address space is BE */
+ data = (src[0] << 24) + (src[1] << 16) + (src[2] << 8) + src[3];
+
+ ret = kv_set_smc_sram_address(rdev, addr, limit);
+ if (ret)
+ return ret;
+
+ WREG32(SMC_IND_DATA_0, data);
+
+ src += 4;
+ byte_count -= 4;
+ addr += 4;
+ }
+
+ /* RMW for the final bytes */
+ if (byte_count > 0) {
+ data = 0;
+
+ ret = kv_set_smc_sram_address(rdev, addr, limit);
+ if (ret)
+ return ret;
+
+ original_data= RREG32(SMC_IND_DATA_0);
+
+ extra_shift = 8 * (4 - byte_count);
+
+ while (byte_count > 0) {
+ /* SMC address space is BE */
+ data = (data << 8) + *src++;
+ byte_count--;
+ }
+
+ data <<= extra_shift;
+
+ data |= (original_data & ~((~0UL) << extra_shift));
+
+ ret = kv_set_smc_sram_address(rdev, addr, limit);
+ if (ret)
+ return ret;
+
+ WREG32(SMC_IND_DATA_0, data);
+ }
+ return 0;
+}
+
diff --git a/drivers/gpu/drm/radeon/ppsmc.h b/drivers/gpu/drm/radeon/ppsmc.h
index b5564a3..6db6e32 100644
--- a/drivers/gpu/drm/radeon/ppsmc.h
+++ b/drivers/gpu/drm/radeon/ppsmc.h
@@ -99,11 +99,45 @@ typedef uint8_t PPSMC_Result;
#define PPSMC_MSG_ThrottleOVRDSCLKDS ((uint8_t)0x96)
#define PPSMC_MSG_CancelThrottleOVRDSCLKDS ((uint8_t)0x97)
+/* KV/KB */
+#define PPSMC_MSG_UVDDPM_SetEnabledMask ((uint16_t) 0x12D)
+#define PPSMC_MSG_VCEDPM_SetEnabledMask ((uint16_t) 0x12E)
+#define PPSMC_MSG_ACPDPM_SetEnabledMask ((uint16_t) 0x12F)
+#define PPSMC_MSG_SAMUDPM_SetEnabledMask ((uint16_t) 0x130)
+#define PPSMC_MSG_MCLKDPM_ForceState ((uint16_t) 0x131)
+#define PPSMC_MSG_MCLKDPM_NoForcedLevel ((uint16_t) 0x132)
+#define PPSMC_MSG_Voltage_Cntl_Disable ((uint16_t) 0x135)
+#define PPSMC_MSG_PCIeDPM_Enable ((uint16_t) 0x136)
+#define PPSMC_MSG_ACPPowerOFF ((uint16_t) 0x137)
+#define PPSMC_MSG_ACPPowerON ((uint16_t) 0x138)
+#define PPSMC_MSG_SAMPowerOFF ((uint16_t) 0x139)
+#define PPSMC_MSG_SAMPowerON ((uint16_t) 0x13a)
+#define PPSMC_MSG_PCIeDPM_Disable ((uint16_t) 0x13d)
+#define PPSMC_MSG_NBDPM_Enable ((uint16_t) 0x140)
+#define PPSMC_MSG_NBDPM_Disable ((uint16_t) 0x141)
+#define PPSMC_MSG_SCLKDPM_SetEnabledMask ((uint16_t) 0x145)
+#define PPSMC_MSG_DPM_Enable ((uint16_t) 0x14e)
+#define PPSMC_MSG_DPM_Disable ((uint16_t) 0x14f)
+#define PPSMC_MSG_UVDDPM_Enable ((uint16_t) 0x154)
+#define PPSMC_MSG_UVDDPM_Disable ((uint16_t) 0x155)
+#define PPSMC_MSG_SAMUDPM_Enable ((uint16_t) 0x156)
+#define PPSMC_MSG_SAMUDPM_Disable ((uint16_t) 0x157)
+#define PPSMC_MSG_ACPDPM_Enable ((uint16_t) 0x158)
+#define PPSMC_MSG_ACPDPM_Disable ((uint16_t) 0x159)
+#define PPSMC_MSG_VCEDPM_Enable ((uint16_t) 0x15a)
+#define PPSMC_MSG_VCEDPM_Disable ((uint16_t) 0x15b)
+#define PPSMC_MSG_SCLKDPM_GetEnabledMask ((uint16_t) 0x162)
+#define PPSMC_MSG_SCLKDPM_FreezeLevel ((uint16_t) 0x189)
+#define PPSMC_MSG_SCLKDPM_UnfreezeLevel ((uint16_t) 0x18A)
+
/* TN */
#define PPSMC_MSG_DPM_Config ((uint32_t) 0x102)
#define PPSMC_MSG_DPM_ForceState ((uint32_t) 0x104)
#define PPSMC_MSG_PG_SIMD_Config ((uint32_t) 0x108)
#define PPSMC_MSG_DPM_N_LevelsDisabled ((uint32_t) 0x112)
+#define PPSMC_MSG_Voltage_Cntl_Enable ((uint32_t) 0x109)
+#define PPSMC_MSG_VCEPowerOFF ((uint32_t) 0x10e)
+#define PPSMC_MSG_VCEPowerON ((uint32_t) 0x10f)
#define PPSMC_MSG_DCE_RemoveVoltageAdjustment ((uint32_t) 0x11d)
#define PPSMC_MSG_DCE_AllowVoltageAdjustment ((uint32_t) 0x11e)
#define PPSMC_MSG_UVD_DPM_Config ((uint32_t) 0x124)
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 3a55540..9c83ecf 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2610,6 +2610,20 @@ static struct radeon_asic kv_asic = {
.set_uvd_clocks = &cik_set_uvd_clocks,
.get_temperature = &kv_get_temp,
},
+ .dpm = {
+ .init = &kv_dpm_init,
+ .setup_asic = &kv_dpm_setup_asic,
+ .enable = &kv_dpm_enable,
+ .disable = &kv_dpm_disable,
+ .pre_set_power_state = &kv_dpm_pre_set_power_state,
+ .set_power_state = &kv_dpm_set_power_state,
+ .post_set_power_state = &kv_dpm_post_set_power_state,
+ .display_configuration_changed = &kv_dpm_display_configuration_changed,
+ .fini = &kv_dpm_fini,
+ .get_sclk = &kv_dpm_get_sclk,
+ .get_mclk = &kv_dpm_get_mclk,
+ .print_power_state = &kv_dpm_print_power_state,
+ },
.pflip = {
.pre_page_flip = &evergreen_pre_page_flip,
.page_flip = &evergreen_page_flip,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index d5c6c5b..68a1a1f 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -750,4 +750,18 @@ void cik_compute_ring_set_wptr(struct radeon_device *rdev,
int ci_get_temp(struct radeon_device *rdev);
int kv_get_temp(struct radeon_device *rdev);
+int kv_dpm_init(struct radeon_device *rdev);
+int kv_dpm_enable(struct radeon_device *rdev);
+void kv_dpm_disable(struct radeon_device *rdev);
+int kv_dpm_pre_set_power_state(struct radeon_device *rdev);
+int kv_dpm_set_power_state(struct radeon_device *rdev);
+void kv_dpm_post_set_power_state(struct radeon_device *rdev);
+void kv_dpm_setup_asic(struct radeon_device *rdev);
+void kv_dpm_display_configuration_changed(struct radeon_device *rdev);
+void kv_dpm_fini(struct radeon_device *rdev);
+u32 kv_dpm_get_sclk(struct radeon_device *rdev, bool low);
+u32 kv_dpm_get_mclk(struct radeon_device *rdev, bool low);
+void kv_dpm_print_power_state(struct radeon_device *rdev,
+ struct radeon_ps *ps);
+
#endif
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 1408014..37d3d34 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -1202,6 +1202,8 @@ int radeon_pm_init(struct radeon_device *rdev)
case CHIP_VERDE:
case CHIP_OLAND:
case CHIP_HAINAN:
+ case CHIP_KABINI:
+ case CHIP_KAVERI:
/* DPM requires the RLC, RV770+ dGPU requires SMC */
if (!rdev->rlc_fw)
rdev->pm.pm_method = PM_METHOD_PROFILE;
diff --git a/drivers/gpu/drm/radeon/smu7.h b/drivers/gpu/drm/radeon/smu7.h
new file mode 100644
index 0000000..75a380a
--- /dev/null
+++ b/drivers/gpu/drm/radeon/smu7.h
@@ -0,0 +1,170 @@
+/*
+ * Copyright 2013 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef SMU7_H
+#define SMU7_H
+
+#pragma pack(push, 1)
+
+#define SMU7_CONTEXT_ID_SMC 1
+#define SMU7_CONTEXT_ID_VBIOS 2
+
+
+#define SMU7_CONTEXT_ID_SMC 1
+#define SMU7_CONTEXT_ID_VBIOS 2
+
+#define SMU7_MAX_LEVELS_VDDC 8
+#define SMU7_MAX_LEVELS_VDDCI 4
+#define SMU7_MAX_LEVELS_MVDD 4
+#define SMU7_MAX_LEVELS_VDDNB 8
+
+#define SMU7_MAX_LEVELS_GRAPHICS SMU__NUM_SCLK_DPM_STATE // SCLK + SQ DPM + ULV
+#define SMU7_MAX_LEVELS_MEMORY SMU__NUM_MCLK_DPM_LEVELS // MCLK Levels DPM
+#define SMU7_MAX_LEVELS_GIO SMU__NUM_LCLK_DPM_LEVELS // LCLK Levels
+#define SMU7_MAX_LEVELS_LINK SMU__NUM_PCIE_DPM_LEVELS // PCIe speed and number of lanes.
+#define SMU7_MAX_LEVELS_UVD 8 // VCLK/DCLK levels for UVD.
+#define SMU7_MAX_LEVELS_VCE 8 // ECLK levels for VCE.
+#define SMU7_MAX_LEVELS_ACP 8 // ACLK levels for ACP.
+#define SMU7_MAX_LEVELS_SAMU 8 // SAMCLK levels for SAMU.
+#define SMU7_MAX_ENTRIES_SMIO 32 // Number of entries in SMIO table.
+
+#define DPM_NO_LIMIT 0
+#define DPM_NO_UP 1
+#define DPM_GO_DOWN 2
+#define DPM_GO_UP 3
+
+#define SMU7_FIRST_DPM_GRAPHICS_LEVEL 0
+#define SMU7_FIRST_DPM_MEMORY_LEVEL 0
+
+#define GPIO_CLAMP_MODE_VRHOT 1
+#define GPIO_CLAMP_MODE_THERM 2
+#define GPIO_CLAMP_MODE_DC 4
+
+#define SCRATCH_B_TARG_PCIE_INDEX_SHIFT 0
+#define SCRATCH_B_TARG_PCIE_INDEX_MASK (0x7<<SCRATCH_B_TARG_PCIE_INDEX_SHIFT)
+#define SCRATCH_B_CURR_PCIE_INDEX_SHIFT 3
+#define SCRATCH_B_CURR_PCIE_INDEX_MASK (0x7<<SCRATCH_B_CURR_PCIE_INDEX_SHIFT)
+#define SCRATCH_B_TARG_UVD_INDEX_SHIFT 6
+#define SCRATCH_B_TARG_UVD_INDEX_MASK (0x7<<SCRATCH_B_TARG_UVD_INDEX_SHIFT)
+#define SCRATCH_B_CURR_UVD_INDEX_SHIFT 9
+#define SCRATCH_B_CURR_UVD_INDEX_MASK (0x7<<SCRATCH_B_CURR_UVD_INDEX_SHIFT)
+#define SCRATCH_B_TARG_VCE_INDEX_SHIFT 12
+#define SCRATCH_B_TARG_VCE_INDEX_MASK (0x7<<SCRATCH_B_TARG_VCE_INDEX_SHIFT)
+#define SCRATCH_B_CURR_VCE_INDEX_SHIFT 15
+#define SCRATCH_B_CURR_VCE_INDEX_MASK (0x7<<SCRATCH_B_CURR_VCE_INDEX_SHIFT)
+#define SCRATCH_B_TARG_ACP_INDEX_SHIFT 18
+#define SCRATCH_B_TARG_ACP_INDEX_MASK (0x7<<SCRATCH_B_TARG_ACP_INDEX_SHIFT)
+#define SCRATCH_B_CURR_ACP_INDEX_SHIFT 21
+#define SCRATCH_B_CURR_ACP_INDEX_MASK (0x7<<SCRATCH_B_CURR_ACP_INDEX_SHIFT)
+#define SCRATCH_B_TARG_SAMU_INDEX_SHIFT 24
+#define SCRATCH_B_TARG_SAMU_INDEX_MASK (0x7<<SCRATCH_B_TARG_SAMU_INDEX_SHIFT)
+#define SCRATCH_B_CURR_SAMU_INDEX_SHIFT 27
+#define SCRATCH_B_CURR_SAMU_INDEX_MASK (0x7<<SCRATCH_B_CURR_SAMU_INDEX_SHIFT)
+
+
+struct SMU7_PIDController
+{
+ uint32_t Ki;
+ int32_t LFWindupUL;
+ int32_t LFWindupLL;
+ uint32_t StatePrecision;
+ uint32_t LfPrecision;
+ uint32_t LfOffset;
+ uint32_t MaxState;
+ uint32_t MaxLfFraction;
+ uint32_t StateShift;
+};
+
+typedef struct SMU7_PIDController SMU7_PIDController;
+
+// -------------------------------------------------------------------------------------------------------------------------
+#define SMU7_MAX_PCIE_LINK_SPEEDS 3 /* 0:Gen1 1:Gen2 2:Gen3 */
+
+#define SMU7_SCLK_DPM_CONFIG_MASK 0x01
+#define SMU7_VOLTAGE_CONTROLLER_CONFIG_MASK 0x02
+#define SMU7_THERMAL_CONTROLLER_CONFIG_MASK 0x04
+#define SMU7_MCLK_DPM_CONFIG_MASK 0x08
+#define SMU7_UVD_DPM_CONFIG_MASK 0x10
+#define SMU7_VCE_DPM_CONFIG_MASK 0x20
+#define SMU7_ACP_DPM_CONFIG_MASK 0x40
+#define SMU7_SAMU_DPM_CONFIG_MASK 0x80
+#define SMU7_PCIEGEN_DPM_CONFIG_MASK 0x100
+
+#define SMU7_ACP_MCLK_HANDSHAKE_DISABLE 0x00000001
+#define SMU7_ACP_SCLK_HANDSHAKE_DISABLE 0x00000002
+#define SMU7_UVD_MCLK_HANDSHAKE_DISABLE 0x00000100
+#define SMU7_UVD_SCLK_HANDSHAKE_DISABLE 0x00000200
+#define SMU7_VCE_MCLK_HANDSHAKE_DISABLE 0x00010000
+#define SMU7_VCE_SCLK_HANDSHAKE_DISABLE 0x00020000
+
+struct SMU7_Firmware_Header
+{
+ uint32_t Digest[5];
+ uint32_t Version;
+ uint32_t HeaderSize;
+ uint32_t Flags;
+ uint32_t EntryPoint;
+ uint32_t CodeSize;
+ uint32_t ImageSize;
+
+ uint32_t Rtos;
+ uint32_t SoftRegisters;
+ uint32_t DpmTable;
+ uint32_t FanTable;
+ uint32_t CacConfigTable;
+ uint32_t CacStatusTable;
+
+ uint32_t mcRegisterTable;
+
+ uint32_t mcArbDramTimingTable;
+
+ uint32_t PmFuseTable;
+ uint32_t Globals;
+ uint32_t Reserved[42];
+ uint32_t Signature;
+};
+
+typedef struct SMU7_Firmware_Header SMU7_Firmware_Header;
+
+#define SMU7_FIRMWARE_HEADER_LOCATION 0x20000
+
+enum DisplayConfig {
+ PowerDown = 1,
+ DP54x4,
+ DP54x2,
+ DP54x1,
+ DP27x4,
+ DP27x2,
+ DP27x1,
+ HDMI297,
+ HDMI162,
+ LVDS,
+ DP324x4,
+ DP324x2,
+ DP324x1
+};
+
+#pragma pack(pop)
+
+#endif
+
diff --git a/drivers/gpu/drm/radeon/smu7_fusion.h b/drivers/gpu/drm/radeon/smu7_fusion.h
new file mode 100644
index 0000000..78ada9f
--- /dev/null
+++ b/drivers/gpu/drm/radeon/smu7_fusion.h
@@ -0,0 +1,300 @@
+/*
+ * Copyright 2013 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef SMU7_FUSION_H
+#define SMU7_FUSION_H
+
+#include "smu7.h"
+
+#pragma pack(push, 1)
+
+#define SMU7_DTE_ITERATIONS 5
+#define SMU7_DTE_SOURCES 5
+#define SMU7_DTE_SINKS 3
+#define SMU7_NUM_CPU_TES 2
+#define SMU7_NUM_GPU_TES 1
+#define SMU7_NUM_NON_TES 2
+
+// All 'soft registers' should be uint32_t.
+struct SMU7_SoftRegisters
+{
+ uint32_t RefClockFrequency;
+ uint32_t PmTimerP;
+ uint32_t FeatureEnables;
+ uint32_t HandshakeDisables;
+
+ uint8_t DisplayPhy1Config;
+ uint8_t DisplayPhy2Config;
+ uint8_t DisplayPhy3Config;
+ uint8_t DisplayPhy4Config;
+
+ uint8_t DisplayPhy5Config;
+ uint8_t DisplayPhy6Config;
+ uint8_t DisplayPhy7Config;
+ uint8_t DisplayPhy8Config;
+
+ uint32_t AverageGraphicsA;
+ uint32_t AverageMemoryA;
+ uint32_t AverageGioA;
+
+ uint8_t SClkDpmEnabledLevels;
+ uint8_t MClkDpmEnabledLevels;
+ uint8_t LClkDpmEnabledLevels;
+ uint8_t PCIeDpmEnabledLevels;
+
+ uint8_t UVDDpmEnabledLevels;
+ uint8_t SAMUDpmEnabledLevels;
+ uint8_t ACPDpmEnabledLevels;
+ uint8_t VCEDpmEnabledLevels;
+
+ uint32_t DRAM_LOG_ADDR_H;
+ uint32_t DRAM_LOG_ADDR_L;
+ uint32_t DRAM_LOG_PHY_ADDR_H;
+ uint32_t DRAM_LOG_PHY_ADDR_L;
+ uint32_t DRAM_LOG_BUFF_SIZE;
+ uint32_t UlvEnterC;
+ uint32_t UlvTime;
+ uint32_t Reserved[3];
+
+};
+
+typedef struct SMU7_SoftRegisters SMU7_SoftRegisters;
+
+struct SMU7_Fusion_GraphicsLevel
+{
+ uint32_t MinVddNb;
+
+ uint32_t SclkFrequency;
+
+ uint8_t Vid;
+ uint8_t VidOffset;
+ uint16_t AT;
+
+ uint8_t PowerThrottle;
+ uint8_t GnbSlow;
+ uint8_t ForceNbPs1;
+ uint8_t SclkDid;
+
+ uint8_t DisplayWatermark;
+ uint8_t EnabledForActivity;
+ uint8_t EnabledForThrottle;
+ uint8_t UpH;
+
+ uint8_t DownH;
+ uint8_t VoltageDownH;
+ uint8_t DeepSleepDivId;
+
+ uint8_t ClkBypassCntl;
+
+ uint32_t reserved;
+};
+
+typedef struct SMU7_Fusion_GraphicsLevel SMU7_Fusion_GraphicsLevel;
+
+struct SMU7_Fusion_GIOLevel
+{
+ uint8_t EnabledForActivity;
+ uint8_t LclkDid;
+ uint8_t Vid;
+ uint8_t VoltageDownH;
+
+ uint32_t MinVddNb;
+
+ uint16_t ResidencyCounter;
+ uint8_t UpH;
+ uint8_t DownH;
+
+ uint32_t LclkFrequency;
+
+ uint8_t ActivityLevel;
+ uint8_t EnabledForThrottle;
+
+ uint8_t ClkBypassCntl;
+
+ uint8_t padding;
+};
+
+typedef struct SMU7_Fusion_GIOLevel SMU7_Fusion_GIOLevel;
+
+// UVD VCLK/DCLK state (level) definition.
+struct SMU7_Fusion_UvdLevel
+{
+ uint32_t VclkFrequency;
+ uint32_t DclkFrequency;
+ uint16_t MinVddNb;
+ uint8_t VclkDivider;
+ uint8_t DclkDivider;
+
+ uint8_t VClkBypassCntl;
+ uint8_t DClkBypassCntl;
+
+ uint8_t padding[2];
+
+};
+
+typedef struct SMU7_Fusion_UvdLevel SMU7_Fusion_UvdLevel;
+
+// Clocks for other external blocks (VCE, ACP, SAMU).
+struct SMU7_Fusion_ExtClkLevel
+{
+ uint32_t Frequency;
+ uint16_t MinVoltage;
+ uint8_t Divider;
+ uint8_t ClkBypassCntl;
+
+ uint32_t Reserved;
+};
+typedef struct SMU7_Fusion_ExtClkLevel SMU7_Fusion_ExtClkLevel;
+
+struct SMU7_Fusion_ACPILevel
+{
+ uint32_t Flags;
+ uint32_t MinVddNb;
+ uint32_t SclkFrequency;
+ uint8_t SclkDid;
+ uint8_t GnbSlow;
+ uint8_t ForceNbPs1;
+ uint8_t DisplayWatermark;
+ uint8_t DeepSleepDivId;
+ uint8_t padding[3];
+};
+
+typedef struct SMU7_Fusion_ACPILevel SMU7_Fusion_ACPILevel;
+
+struct SMU7_Fusion_NbDpm
+{
+ uint8_t DpmXNbPsHi;
+ uint8_t DpmXNbPsLo;
+ uint8_t Dpm0PgNbPsHi;
+ uint8_t Dpm0PgNbPsLo;
+ uint8_t EnablePsi1;
+ uint8_t SkipDPM0;
+ uint8_t SkipPG;
+ uint8_t Hysteresis;
+ uint8_t EnableDpmPstatePoll;
+ uint8_t padding[3];
+};
+
+typedef struct SMU7_Fusion_NbDpm SMU7_Fusion_NbDpm;
+
+struct SMU7_Fusion_StateInfo
+{
+ uint32_t SclkFrequency;
+ uint32_t LclkFrequency;
+ uint32_t VclkFrequency;
+ uint32_t DclkFrequency;
+ uint32_t SamclkFrequency;
+ uint32_t AclkFrequency;
+ uint32_t EclkFrequency;
+ uint8_t DisplayWatermark;
+ uint8_t McArbIndex;
+ int8_t SclkIndex;
+ int8_t MclkIndex;
+};
+
+typedef struct SMU7_Fusion_StateInfo SMU7_Fusion_StateInfo;
+
+struct SMU7_Fusion_DpmTable
+{
+ uint32_t SystemFlags;
+
+ SMU7_PIDController GraphicsPIDController;
+ SMU7_PIDController GioPIDController;
+
+ uint8_t GraphicsDpmLevelCount;
+ uint8_t GIOLevelCount;
+ uint8_t UvdLevelCount;
+ uint8_t VceLevelCount;
+
+ uint8_t AcpLevelCount;
+ uint8_t SamuLevelCount;
+ uint16_t FpsHighT;
+
+ SMU7_Fusion_GraphicsLevel GraphicsLevel [SMU__NUM_SCLK_DPM_STATE];
+ SMU7_Fusion_ACPILevel ACPILevel;
+ SMU7_Fusion_UvdLevel UvdLevel [SMU7_MAX_LEVELS_UVD];
+ SMU7_Fusion_ExtClkLevel VceLevel [SMU7_MAX_LEVELS_VCE];
+ SMU7_Fusion_ExtClkLevel AcpLevel [SMU7_MAX_LEVELS_ACP];
+ SMU7_Fusion_ExtClkLevel SamuLevel [SMU7_MAX_LEVELS_SAMU];
+
+ uint8_t UvdBootLevel;
+ uint8_t VceBootLevel;
+ uint8_t AcpBootLevel;
+ uint8_t SamuBootLevel;
+ uint8_t UVDInterval;
+ uint8_t VCEInterval;
+ uint8_t ACPInterval;
+ uint8_t SAMUInterval;
+
+ uint8_t GraphicsBootLevel;
+ uint8_t GraphicsInterval;
+ uint8_t GraphicsThermThrottleEnable;
+ uint8_t GraphicsVoltageChangeEnable;
+
+ uint8_t GraphicsClkSlowEnable;
+ uint8_t GraphicsClkSlowDivider;
+ uint16_t FpsLowT;
+
+ uint32_t DisplayCac;
+ uint32_t LowSclkInterruptT;
+
+ uint32_t DRAM_LOG_ADDR_H;
+ uint32_t DRAM_LOG_ADDR_L;
+ uint32_t DRAM_LOG_PHY_ADDR_H;
+ uint32_t DRAM_LOG_PHY_ADDR_L;
+ uint32_t DRAM_LOG_BUFF_SIZE;
+
+};
+
+struct SMU7_Fusion_GIODpmTable
+{
+
+ SMU7_Fusion_GIOLevel GIOLevel [SMU7_MAX_LEVELS_GIO];
+
+ SMU7_PIDController GioPIDController;
+
+ uint32_t GIOLevelCount;
+
+ uint8_t Enable;
+ uint8_t GIOVoltageChangeEnable;
+ uint8_t GIOBootLevel;
+ uint8_t padding;
+ uint8_t padding1[2];
+ uint8_t TargetState;
+ uint8_t CurrenttState;
+ uint8_t ThrottleOnHtc;
+ uint8_t ThermThrottleStatus;
+ uint8_t ThermThrottleTempSelect;
+ uint8_t ThermThrottleEnable;
+ uint16_t TemperatureLimitHigh;
+ uint16_t TemperatureLimitLow;
+
+};
+
+typedef struct SMU7_Fusion_DpmTable SMU7_Fusion_DpmTable;
+typedef struct SMU7_Fusion_GIODpmTable SMU7_Fusion_GIODpmTable;
+
+#pragma pack(pop)
+
+#endif
+
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 42/53] drm/radeon: add dpm support for CI dGPUs (v2)
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (40 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 41/53] drm/radeon/kms: add dpm support for KB/KV Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 43/53] drm/radeon/dpm: add debugfs support for CI Alex Deucher
` (11 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
This adds dpm support for btc asics. This includes:
- dynamic engine clock scaling
- dynamic memory clock scaling
- dynamic voltage scaling
- dynamic pcie gen switching
Set radeon.dpm=1 to enable.
v2: remove unused radeon_atombios.c changes,
make missing smc ucode non-fatal
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/Makefile | 3 +-
drivers/gpu/drm/radeon/ci_dpm.c | 5006 ++++++++++++++++++++++++++++++
drivers/gpu/drm/radeon/ci_dpm.h | 331 ++
drivers/gpu/drm/radeon/ci_smc.c | 262 ++
drivers/gpu/drm/radeon/cik.c | 41 +-
drivers/gpu/drm/radeon/cikd.h | 259 +-
drivers/gpu/drm/radeon/ppsmc.h | 24 +-
drivers/gpu/drm/radeon/radeon.h | 6 +
drivers/gpu/drm/radeon/radeon_asic.c | 14 +
drivers/gpu/drm/radeon/radeon_asic.h | 14 +
drivers/gpu/drm/radeon/radeon_atombios.c | 2 +-
drivers/gpu/drm/radeon/radeon_pm.c | 1 +
drivers/gpu/drm/radeon/radeon_ucode.h | 3 +
drivers/gpu/drm/radeon/si_dpm.c | 10 +-
drivers/gpu/drm/radeon/smu7_discrete.h | 486 +++
15 files changed, 6447 insertions(+), 15 deletions(-)
create mode 100644 drivers/gpu/drm/radeon/ci_dpm.c
create mode 100644 drivers/gpu/drm/radeon/ci_dpm.h
create mode 100644 drivers/gpu/drm/radeon/ci_smc.c
create mode 100644 drivers/gpu/drm/radeon/smu7_discrete.h
diff --git a/drivers/gpu/drm/radeon/Makefile b/drivers/gpu/drm/radeon/Makefile
index d3265b5..ea913cc 100644
--- a/drivers/gpu/drm/radeon/Makefile
+++ b/drivers/gpu/drm/radeon/Makefile
@@ -79,7 +79,8 @@ radeon-y += radeon_device.o radeon_asic.o radeon_kms.o \
si_blit_shaders.o radeon_prime.o radeon_uvd.o cik.o cik_blit_shaders.o \
r600_dpm.o rs780_dpm.o rv6xx_dpm.o rv770_dpm.o rv730_dpm.o rv740_dpm.o \
rv770_smc.o cypress_dpm.o btc_dpm.o sumo_dpm.o sumo_smc.o trinity_dpm.o \
- trinity_smc.o ni_dpm.o si_smc.o si_dpm.o kv_smc.o kv_dpm.o
+ trinity_smc.o ni_dpm.o si_smc.o si_dpm.o kv_smc.o kv_dpm.o ci_smc.o \
+ ci_dpm.o
radeon-$(CONFIG_COMPAT) += radeon_ioc32.o
radeon-$(CONFIG_VGA_SWITCHEROO) += radeon_atpx_handler.o
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
new file mode 100644
index 0000000..72ab92b
--- /dev/null
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -0,0 +1,5006 @@
+/*
+ * Copyright 2013 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include "drmP.h"
+#include "radeon.h"
+#include "cikd.h"
+#include "r600_dpm.h"
+#include "ci_dpm.h"
+#include "atom.h"
+
+#define MC_CG_ARB_FREQ_F0 0x0a
+#define MC_CG_ARB_FREQ_F1 0x0b
+#define MC_CG_ARB_FREQ_F2 0x0c
+#define MC_CG_ARB_FREQ_F3 0x0d
+
+#define SMC_RAM_END 0x40000
+
+#define VOLTAGE_SCALE 4
+#define VOLTAGE_VID_OFFSET_SCALE1 625
+#define VOLTAGE_VID_OFFSET_SCALE2 100
+
+static const struct ci_pt_defaults defaults_bonaire_xt =
+{
+ 1, 0xF, 0xFD, 0x19, 5, 45, 0, 0xB0000,
+ { 0x79, 0x253, 0x25D, 0xAE, 0x72, 0x80, 0x83, 0x86, 0x6F, 0xC8, 0xC9, 0xC9, 0x2F, 0x4D, 0x61 },
+ { 0x17C, 0x172, 0x180, 0x1BC, 0x1B3, 0x1BD, 0x206, 0x200, 0x203, 0x25D, 0x25A, 0x255, 0x2C3, 0x2C5, 0x2B4 }
+};
+
+static const struct ci_pt_defaults defaults_bonaire_pro =
+{
+ 1, 0xF, 0xFD, 0x19, 5, 45, 0, 0x65062,
+ { 0x8C, 0x23F, 0x244, 0xA6, 0x83, 0x85, 0x86, 0x86, 0x83, 0xDB, 0xDB, 0xDA, 0x67, 0x60, 0x5F },
+ { 0x187, 0x193, 0x193, 0x1C7, 0x1D1, 0x1D1, 0x210, 0x219, 0x219, 0x266, 0x26C, 0x26C, 0x2C9, 0x2CB, 0x2CB }
+};
+
+static const struct ci_pt_defaults defaults_saturn_xt =
+{
+ 1, 0xF, 0xFD, 0x19, 5, 55, 0, 0x70000,
+ { 0x8C, 0x247, 0x249, 0xA6, 0x80, 0x81, 0x8B, 0x89, 0x86, 0xC9, 0xCA, 0xC9, 0x4D, 0x4D, 0x4D },
+ { 0x187, 0x187, 0x187, 0x1C7, 0x1C7, 0x1C7, 0x210, 0x210, 0x210, 0x266, 0x266, 0x266, 0x2C9, 0x2C9, 0x2C9 }
+};
+
+static const struct ci_pt_defaults defaults_saturn_pro =
+{
+ 1, 0xF, 0xFD, 0x19, 5, 55, 0, 0x30000,
+ { 0x96, 0x21D, 0x23B, 0xA1, 0x85, 0x87, 0x83, 0x84, 0x81, 0xE6, 0xE6, 0xE6, 0x71, 0x6A, 0x6A },
+ { 0x193, 0x19E, 0x19E, 0x1D2, 0x1DC, 0x1DC, 0x21A, 0x223, 0x223, 0x26E, 0x27E, 0x274, 0x2CF, 0x2D2, 0x2D2 }
+};
+
+static const struct ci_pt_config_reg didt_config_ci[] =
+{
+ { 0x10, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x10, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x10, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x10, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x11, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x11, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x11, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x11, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x12, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x12, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x12, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x12, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x2, 0x00003fff, 0, 0x4, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x2, 0x03ff0000, 16, 0x80, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x2, 0x78000000, 27, 0x3, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x1, 0x0000ffff, 0, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x1, 0xffff0000, 16, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x0, 0x00000001, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x30, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x30, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x30, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x30, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x31, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x31, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x31, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x31, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x32, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x32, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x32, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x32, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x22, 0x00003fff, 0, 0x4, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x22, 0x03ff0000, 16, 0x80, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x22, 0x78000000, 27, 0x3, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x21, 0x0000ffff, 0, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x21, 0xffff0000, 16, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x20, 0x00000001, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x50, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x50, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x50, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x50, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x51, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x51, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x51, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x51, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x52, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x52, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x52, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x52, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x42, 0x00003fff, 0, 0x4, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x42, 0x03ff0000, 16, 0x80, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x42, 0x78000000, 27, 0x3, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x41, 0x0000ffff, 0, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x41, 0xffff0000, 16, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x40, 0x00000001, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x70, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x70, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x70, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x70, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x71, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x71, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x71, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x71, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x72, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x72, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x72, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x72, 0xff000000, 24, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x62, 0x00003fff, 0, 0x4, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x62, 0x03ff0000, 16, 0x80, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x62, 0x78000000, 27, 0x3, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x61, 0x0000ffff, 0, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x61, 0xffff0000, 16, 0x3FFF, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0x60, 0x00000001, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND },
+ { 0xFFFFFFFF }
+};
+
+extern u8 rv770_get_memory_module_index(struct radeon_device *rdev);
+extern int ni_copy_and_switch_arb_sets(struct radeon_device *rdev,
+ u32 arb_freq_src, u32 arb_freq_dest);
+extern u8 si_get_ddr3_mclk_frequency_ratio(u32 memory_clock);
+extern u8 si_get_mclk_frequency_ratio(u32 memory_clock, bool strobe_mode);
+extern void si_trim_voltage_table_to_fit_state_table(struct radeon_device *rdev,
+ u32 max_voltage_steps,
+ struct atom_voltage_table *voltage_table);
+extern void cik_enter_rlc_safe_mode(struct radeon_device *rdev);
+extern void cik_exit_rlc_safe_mode(struct radeon_device *rdev);
+
+static int ci_get_std_voltage_value_sidd(struct radeon_device *rdev,
+ struct atom_voltage_table_entry *voltage_table,
+ u16 *std_voltage_hi_sidd, u16 *std_voltage_lo_sidd);
+static int ci_set_power_limit(struct radeon_device *rdev, u32 n);
+static int ci_set_overdrive_target_tdp(struct radeon_device *rdev,
+ u32 target_tdp);
+static int ci_update_uvd_dpm(struct radeon_device *rdev, bool gate);
+
+static struct ci_power_info *ci_get_pi(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = rdev->pm.dpm.priv;
+
+ return pi;
+}
+
+static struct ci_ps *ci_get_ps(struct radeon_ps *rps)
+{
+ struct ci_ps *ps = rps->ps_priv;
+
+ return ps;
+}
+
+static void ci_initialize_powertune_defaults(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ switch (rdev->pdev->device) {
+ case 0x6650:
+ case 0x6658:
+ case 0x665C:
+ default:
+ pi->powertune_defaults = &defaults_bonaire_xt;
+ break;
+ case 0x6651:
+ case 0x665D:
+ pi->powertune_defaults = &defaults_bonaire_pro;
+ break;
+ case 0x6640:
+ pi->powertune_defaults = &defaults_saturn_xt;
+ break;
+ case 0x6641:
+ pi->powertune_defaults = &defaults_saturn_pro;
+ break;
+ }
+
+ pi->dte_tj_offset = 0;
+
+ pi->caps_power_containment = true;
+ pi->caps_cac = false;
+ pi->caps_sq_ramping = false;
+ pi->caps_db_ramping = false;
+ pi->caps_td_ramping = false;
+ pi->caps_tcp_ramping = false;
+
+ if (pi->caps_power_containment) {
+ pi->caps_cac = true;
+ pi->enable_bapm_feature = true;
+ pi->enable_tdc_limit_feature = true;
+ pi->enable_pkg_pwr_tracking_feature = true;
+ }
+}
+
+static u8 ci_convert_to_vid(u16 vddc)
+{
+ return (6200 - (vddc * VOLTAGE_SCALE)) / 25;
+}
+
+static int ci_populate_bapm_vddc_vid_sidd(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u8 *hi_vid = pi->smc_powertune_table.BapmVddCVidHiSidd;
+ u8 *lo_vid = pi->smc_powertune_table.BapmVddCVidLoSidd;
+ u8 *hi2_vid = pi->smc_powertune_table.BapmVddCVidHiSidd2;
+ u32 i;
+
+ if (rdev->pm.dpm.dyn_state.cac_leakage_table.entries == NULL)
+ return -EINVAL;
+ if (rdev->pm.dpm.dyn_state.cac_leakage_table.count > 8)
+ return -EINVAL;
+ if (rdev->pm.dpm.dyn_state.cac_leakage_table.count !=
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.count)
+ return -EINVAL;
+
+ for (i = 0; i < rdev->pm.dpm.dyn_state.cac_leakage_table.count; i++) {
+ if (rdev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_EVV) {
+ lo_vid[i] = ci_convert_to_vid(rdev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc1);
+ hi_vid[i] = ci_convert_to_vid(rdev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc2);
+ hi2_vid[i] = ci_convert_to_vid(rdev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc3);
+ } else {
+ lo_vid[i] = ci_convert_to_vid(rdev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc);
+ hi_vid[i] = ci_convert_to_vid((u16)rdev->pm.dpm.dyn_state.cac_leakage_table.entries[i].leakage);
+ }
+ }
+ return 0;
+}
+
+static int ci_populate_vddc_vid(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u8 *vid = pi->smc_powertune_table.VddCVid;
+ u32 i;
+
+ if (pi->vddc_voltage_table.count > 8)
+ return -EINVAL;
+
+ for (i = 0; i < pi->vddc_voltage_table.count; i++)
+ vid[i] = ci_convert_to_vid(pi->vddc_voltage_table.entries[i].value);
+
+ return 0;
+}
+
+static int ci_populate_svi_load_line(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ const struct ci_pt_defaults *pt_defaults = pi->powertune_defaults;
+
+ pi->smc_powertune_table.SviLoadLineEn = pt_defaults->svi_load_line_en;
+ pi->smc_powertune_table.SviLoadLineVddC = pt_defaults->svi_load_line_vddc;
+ pi->smc_powertune_table.SviLoadLineTrimVddC = 3;
+ pi->smc_powertune_table.SviLoadLineOffsetVddC = 0;
+
+ return 0;
+}
+
+static int ci_populate_tdc_limit(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ const struct ci_pt_defaults *pt_defaults = pi->powertune_defaults;
+ u16 tdc_limit;
+
+ tdc_limit = rdev->pm.dpm.dyn_state.cac_tdp_table->tdc * 256;
+ pi->smc_powertune_table.TDC_VDDC_PkgLimit = cpu_to_be16(tdc_limit);
+ pi->smc_powertune_table.TDC_VDDC_ThrottleReleaseLimitPerc =
+ pt_defaults->tdc_vddc_throttle_release_limit_perc;
+ pi->smc_powertune_table.TDC_MAWt = pt_defaults->tdc_mawt;
+
+ return 0;
+}
+
+static int ci_populate_dw8(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ const struct ci_pt_defaults *pt_defaults = pi->powertune_defaults;
+ int ret;
+
+ ret = ci_read_smc_sram_dword(rdev,
+ SMU7_FIRMWARE_HEADER_LOCATION +
+ offsetof(SMU7_Firmware_Header, PmFuseTable) +
+ offsetof(SMU7_Discrete_PmFuses, TdcWaterfallCtl),
+ (u32 *)&pi->smc_powertune_table.TdcWaterfallCtl,
+ pi->sram_end);
+ if (ret)
+ return -EINVAL;
+ else
+ pi->smc_powertune_table.TdcWaterfallCtl = pt_defaults->tdc_waterfall_ctl;
+
+ return 0;
+}
+
+static int ci_min_max_v_gnbl_pm_lid_from_bapm_vddc(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u8 *hi_vid = pi->smc_powertune_table.BapmVddCVidHiSidd;
+ u8 *lo_vid = pi->smc_powertune_table.BapmVddCVidLoSidd;
+ int i, min, max;
+
+ min = max = hi_vid[0];
+ for (i = 0; i < 8; i++) {
+ if (0 != hi_vid[i]) {
+ if (min > hi_vid[i])
+ min = hi_vid[i];
+ if (max < hi_vid[i])
+ max = hi_vid[i];
+ }
+
+ if (0 != lo_vid[i]) {
+ if (min > lo_vid[i])
+ min = lo_vid[i];
+ if (max < lo_vid[i])
+ max = lo_vid[i];
+ }
+ }
+
+ if ((min == 0) || (max == 0))
+ return -EINVAL;
+ pi->smc_powertune_table.GnbLPMLMaxVid = (u8)max;
+ pi->smc_powertune_table.GnbLPMLMinVid = (u8)min;
+
+ return 0;
+}
+
+static int ci_populate_bapm_vddc_base_leakage_sidd(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u16 hi_sidd = pi->smc_powertune_table.BapmVddCBaseLeakageHiSidd;
+ u16 lo_sidd = pi->smc_powertune_table.BapmVddCBaseLeakageLoSidd;
+ struct radeon_cac_tdp_table *cac_tdp_table =
+ rdev->pm.dpm.dyn_state.cac_tdp_table;
+
+ hi_sidd = cac_tdp_table->high_cac_leakage / 100 * 256;
+ lo_sidd = cac_tdp_table->low_cac_leakage / 100 * 256;
+
+ pi->smc_powertune_table.BapmVddCBaseLeakageHiSidd = cpu_to_be16(hi_sidd);
+ pi->smc_powertune_table.BapmVddCBaseLeakageLoSidd = cpu_to_be16(lo_sidd);
+
+ return 0;
+}
+
+static int ci_populate_bapm_parameters_in_dpm_table(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ const struct ci_pt_defaults *pt_defaults = pi->powertune_defaults;
+ SMU7_Discrete_DpmTable *dpm_table = &pi->smc_state_table;
+ struct radeon_cac_tdp_table *cac_tdp_table =
+ rdev->pm.dpm.dyn_state.cac_tdp_table;
+ struct radeon_ppm_table *ppm = rdev->pm.dpm.dyn_state.ppm_table;
+ int i, j, k;
+ const u16 *def1;
+ const u16 *def2;
+
+ dpm_table->DefaultTdp = cac_tdp_table->tdp * 256;
+ dpm_table->TargetTdp = cac_tdp_table->configurable_tdp * 256;
+
+ dpm_table->DTETjOffset = (u8)pi->dte_tj_offset;
+ dpm_table->GpuTjMax =
+ (u8)(pi->thermal_temp_setting.temperature_high / 1000);
+ dpm_table->GpuTjHyst = 8;
+
+ dpm_table->DTEAmbientTempBase = pt_defaults->dte_ambient_temp_base;
+
+ if (ppm) {
+ dpm_table->PPM_PkgPwrLimit = cpu_to_be16((u16)ppm->dgpu_tdp * 256 / 1000);
+ dpm_table->PPM_TemperatureLimit = cpu_to_be16((u16)ppm->tj_max * 256);
+ } else {
+ dpm_table->PPM_PkgPwrLimit = cpu_to_be16(0);
+ dpm_table->PPM_TemperatureLimit = cpu_to_be16(0);
+ }
+
+ dpm_table->BAPM_TEMP_GRADIENT = cpu_to_be32(pt_defaults->bapm_temp_gradient);
+ def1 = pt_defaults->bapmti_r;
+ def2 = pt_defaults->bapmti_rc;
+
+ for (i = 0; i < SMU7_DTE_ITERATIONS; i++) {
+ for (j = 0; j < SMU7_DTE_SOURCES; j++) {
+ for (k = 0; k < SMU7_DTE_SINKS; k++) {
+ dpm_table->BAPMTI_R[i][j][k] = cpu_to_be16(*def1);
+ dpm_table->BAPMTI_RC[i][j][k] = cpu_to_be16(*def2);
+ def1++;
+ def2++;
+ }
+ }
+ }
+
+ return 0;
+}
+
+static int ci_populate_pm_base(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 pm_fuse_table_offset;
+ int ret;
+
+ if (pi->caps_power_containment) {
+ ret = ci_read_smc_sram_dword(rdev,
+ SMU7_FIRMWARE_HEADER_LOCATION +
+ offsetof(SMU7_Firmware_Header, PmFuseTable),
+ &pm_fuse_table_offset, pi->sram_end);
+ if (ret)
+ return ret;
+ ret = ci_populate_bapm_vddc_vid_sidd(rdev);
+ if (ret)
+ return ret;
+ ret = ci_populate_vddc_vid(rdev);
+ if (ret)
+ return ret;
+ ret = ci_populate_svi_load_line(rdev);
+ if (ret)
+ return ret;
+ ret = ci_populate_tdc_limit(rdev);
+ if (ret)
+ return ret;
+ ret = ci_populate_dw8(rdev);
+ if (ret)
+ return ret;
+ ret = ci_min_max_v_gnbl_pm_lid_from_bapm_vddc(rdev);
+ if (ret)
+ return ret;
+ ret = ci_populate_bapm_vddc_base_leakage_sidd(rdev);
+ if (ret)
+ return ret;
+ ret = ci_copy_bytes_to_smc(rdev, pm_fuse_table_offset,
+ (u8 *)&pi->smc_powertune_table,
+ sizeof(SMU7_Discrete_PmFuses), pi->sram_end);
+ if (ret)
+ return ret;
+ }
+
+ return 0;
+}
+
+static void ci_do_enable_didt(struct radeon_device *rdev, const bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 data;
+
+ if (pi->caps_sq_ramping) {
+ data = RREG32_DIDT(DIDT_SQ_CTRL0);
+ if (enable)
+ data |= DIDT_CTRL_EN;
+ else
+ data &= ~DIDT_CTRL_EN;
+ WREG32_DIDT(DIDT_SQ_CTRL0, data);
+ }
+
+ if (pi->caps_db_ramping) {
+ data = RREG32_DIDT(DIDT_DB_CTRL0);
+ if (enable)
+ data |= DIDT_CTRL_EN;
+ else
+ data &= ~DIDT_CTRL_EN;
+ WREG32_DIDT(DIDT_DB_CTRL0, data);
+ }
+
+ if (pi->caps_td_ramping) {
+ data = RREG32_DIDT(DIDT_TD_CTRL0);
+ if (enable)
+ data |= DIDT_CTRL_EN;
+ else
+ data &= ~DIDT_CTRL_EN;
+ WREG32_DIDT(DIDT_TD_CTRL0, data);
+ }
+
+ if (pi->caps_tcp_ramping) {
+ data = RREG32_DIDT(DIDT_TCP_CTRL0);
+ if (enable)
+ data |= DIDT_CTRL_EN;
+ else
+ data &= ~DIDT_CTRL_EN;
+ WREG32_DIDT(DIDT_TCP_CTRL0, data);
+ }
+}
+
+static int ci_program_pt_config_registers(struct radeon_device *rdev,
+ const struct ci_pt_config_reg *cac_config_regs)
+{
+ const struct ci_pt_config_reg *config_regs = cac_config_regs;
+ u32 data;
+ u32 cache = 0;
+
+ if (config_regs == NULL)
+ return -EINVAL;
+
+ while (config_regs->offset != 0xFFFFFFFF) {
+ if (config_regs->type == CISLANDS_CONFIGREG_CACHE) {
+ cache |= ((config_regs->value << config_regs->shift) & config_regs->mask);
+ } else {
+ switch (config_regs->type) {
+ case CISLANDS_CONFIGREG_SMC_IND:
+ data = RREG32_SMC(config_regs->offset);
+ break;
+ case CISLANDS_CONFIGREG_DIDT_IND:
+ data = RREG32_DIDT(config_regs->offset);
+ break;
+ default:
+ data = RREG32(config_regs->offset << 2);
+ break;
+ }
+
+ data &= ~config_regs->mask;
+ data |= ((config_regs->value << config_regs->shift) & config_regs->mask);
+ data |= cache;
+
+ switch (config_regs->type) {
+ case CISLANDS_CONFIGREG_SMC_IND:
+ WREG32_SMC(config_regs->offset, data);
+ break;
+ case CISLANDS_CONFIGREG_DIDT_IND:
+ WREG32_DIDT(config_regs->offset, data);
+ break;
+ default:
+ WREG32(config_regs->offset << 2, data);
+ break;
+ }
+ cache = 0;
+ }
+ config_regs++;
+ }
+ return 0;
+}
+
+static int ci_enable_didt(struct radeon_device *rdev, bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ int ret;
+
+ if (pi->caps_sq_ramping || pi->caps_db_ramping ||
+ pi->caps_td_ramping || pi->caps_tcp_ramping) {
+ cik_enter_rlc_safe_mode(rdev);
+
+ if (enable) {
+ ret = ci_program_pt_config_registers(rdev, didt_config_ci);
+ if (ret) {
+ cik_exit_rlc_safe_mode(rdev);
+ return ret;
+ }
+ }
+
+ ci_do_enable_didt(rdev, enable);
+
+ cik_exit_rlc_safe_mode(rdev);
+ }
+
+ return 0;
+}
+
+static int ci_enable_power_containment(struct radeon_device *rdev, bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ PPSMC_Result smc_result;
+ int ret = 0;
+
+ if (enable) {
+ pi->power_containment_features = 0;
+ if (pi->caps_power_containment) {
+ if (pi->enable_bapm_feature) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_EnableDTE);
+ if (smc_result != PPSMC_Result_OK)
+ ret = -EINVAL;
+ else
+ pi->power_containment_features |= POWERCONTAINMENT_FEATURE_BAPM;
+ }
+
+ if (pi->enable_tdc_limit_feature) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_TDCLimitEnable);
+ if (smc_result != PPSMC_Result_OK)
+ ret = -EINVAL;
+ else
+ pi->power_containment_features |= POWERCONTAINMENT_FEATURE_TDCLimit;
+ }
+
+ if (pi->enable_pkg_pwr_tracking_feature) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_PkgPwrLimitEnable);
+ if (smc_result != PPSMC_Result_OK) {
+ ret = -EINVAL;
+ } else {
+ struct radeon_cac_tdp_table *cac_tdp_table =
+ rdev->pm.dpm.dyn_state.cac_tdp_table;
+ u32 default_pwr_limit =
+ (u32)(cac_tdp_table->maximum_power_delivery_limit * 256);
+
+ pi->power_containment_features |= POWERCONTAINMENT_FEATURE_PkgPwrLimit;
+
+ ci_set_power_limit(rdev, default_pwr_limit);
+ }
+ }
+ }
+ } else {
+ if (pi->caps_power_containment && pi->power_containment_features) {
+ if (pi->power_containment_features & POWERCONTAINMENT_FEATURE_TDCLimit)
+ ci_send_msg_to_smc(rdev, PPSMC_MSG_TDCLimitDisable);
+
+ if (pi->power_containment_features & POWERCONTAINMENT_FEATURE_BAPM)
+ ci_send_msg_to_smc(rdev, PPSMC_MSG_DisableDTE);
+
+ if (pi->power_containment_features & POWERCONTAINMENT_FEATURE_PkgPwrLimit)
+ ci_send_msg_to_smc(rdev, PPSMC_MSG_PkgPwrLimitDisable);
+ pi->power_containment_features = 0;
+ }
+ }
+
+ return ret;
+}
+
+static int ci_enable_smc_cac(struct radeon_device *rdev, bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ PPSMC_Result smc_result;
+ int ret = 0;
+
+ if (pi->caps_cac) {
+ if (enable) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_EnableCac);
+ if (smc_result != PPSMC_Result_OK) {
+ ret = -EINVAL;
+ pi->cac_enabled = false;
+ } else {
+ pi->cac_enabled = true;
+ }
+ } else if (pi->cac_enabled) {
+ ci_send_msg_to_smc(rdev, PPSMC_MSG_DisableCac);
+ pi->cac_enabled = false;
+ }
+ }
+
+ return ret;
+}
+
+static int ci_power_control_set_level(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct radeon_cac_tdp_table *cac_tdp_table =
+ rdev->pm.dpm.dyn_state.cac_tdp_table;
+ s32 adjust_percent;
+ s32 target_tdp;
+ int ret = 0;
+ bool adjust_polarity = false; /* ??? */
+
+ if (pi->caps_power_containment &&
+ (pi->power_containment_features & POWERCONTAINMENT_FEATURE_BAPM)) {
+ adjust_percent = adjust_polarity ?
+ rdev->pm.dpm.tdp_adjustment : (-1 * rdev->pm.dpm.tdp_adjustment);
+ target_tdp = ((100 + adjust_percent) *
+ (s32)cac_tdp_table->configurable_tdp) / 100;
+ target_tdp *= 256;
+
+ ret = ci_set_overdrive_target_tdp(rdev, (u32)target_tdp);
+ }
+
+ return ret;
+}
+
+static void ci_dpm_powergate_uvd(struct radeon_device *rdev, bool gate)
+{
+ ci_update_uvd_dpm(rdev, gate);
+}
+
+static void ci_apply_state_adjust_rules(struct radeon_device *rdev,
+ struct radeon_ps *rps)
+{
+ struct ci_ps *ps = ci_get_ps(rps);
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct radeon_clock_and_voltage_limits *max_limits;
+ bool disable_mclk_switching;
+ u32 sclk, mclk;
+ int i;
+
+ if (rdev->pm.dpm.new_active_crtc_count > 1)
+ disable_mclk_switching = true;
+ else
+ disable_mclk_switching = false;
+
+ if ((rps->class & ATOM_PPLIB_CLASSIFICATION_UI_MASK) == ATOM_PPLIB_CLASSIFICATION_UI_BATTERY)
+ pi->battery_state = true;
+ else
+ pi->battery_state = false;
+
+ if (rdev->pm.dpm.ac_power)
+ max_limits = &rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
+ else
+ max_limits = &rdev->pm.dpm.dyn_state.max_clock_voltage_on_dc;
+
+ if (rdev->pm.dpm.ac_power == false) {
+ for (i = 0; i < ps->performance_level_count; i++) {
+ if (ps->performance_levels[i].mclk > max_limits->mclk)
+ ps->performance_levels[i].mclk = max_limits->mclk;
+ if (ps->performance_levels[i].sclk > max_limits->sclk)
+ ps->performance_levels[i].sclk = max_limits->sclk;
+ }
+ }
+
+ /* XXX validate the min clocks required for display */
+
+ if (disable_mclk_switching) {
+ mclk = ps->performance_levels[ps->performance_level_count - 1].mclk;
+ sclk = ps->performance_levels[0].sclk;
+ } else {
+ mclk = ps->performance_levels[0].mclk;
+ sclk = ps->performance_levels[0].sclk;
+ }
+
+ ps->performance_levels[0].sclk = sclk;
+ ps->performance_levels[0].mclk = mclk;
+
+ if (ps->performance_levels[1].sclk < ps->performance_levels[0].sclk)
+ ps->performance_levels[1].sclk = ps->performance_levels[0].sclk;
+
+ if (disable_mclk_switching) {
+ if (ps->performance_levels[0].mclk < ps->performance_levels[1].mclk)
+ ps->performance_levels[0].mclk = ps->performance_levels[1].mclk;
+ } else {
+ if (ps->performance_levels[1].mclk < ps->performance_levels[0].mclk)
+ ps->performance_levels[1].mclk = ps->performance_levels[0].mclk;
+ }
+}
+
+static int ci_set_thermal_temperature_range(struct radeon_device *rdev,
+ int min_temp, int max_temp)
+{
+ int low_temp = 0 * 1000;
+ int high_temp = 255 * 1000;
+ u32 tmp;
+
+ if (low_temp < min_temp)
+ low_temp = min_temp;
+ if (high_temp > max_temp)
+ high_temp = max_temp;
+ if (high_temp < low_temp) {
+ DRM_ERROR("invalid thermal range: %d - %d\n", low_temp, high_temp);
+ return -EINVAL;
+ }
+
+ tmp = RREG32_SMC(CG_THERMAL_INT);
+ tmp &= ~(CI_DIG_THERM_INTH_MASK | CI_DIG_THERM_INTL_MASK);
+ tmp |= CI_DIG_THERM_INTH(high_temp / 1000) |
+ CI_DIG_THERM_INTL(low_temp / 1000);
+ WREG32_SMC(CG_THERMAL_INT, tmp);
+
+#if 0
+ /* XXX: need to figure out how to handle this properly */
+ tmp = RREG32_SMC(CG_THERMAL_CTRL);
+ tmp &= DIG_THERM_DPM_MASK;
+ tmp |= DIG_THERM_DPM(high_temp / 1000);
+ WREG32_SMC(CG_THERMAL_CTRL, tmp);
+#endif
+
+ return 0;
+}
+
+#if 0
+static int ci_read_smc_soft_register(struct radeon_device *rdev,
+ u16 reg_offset, u32 *value)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ return ci_read_smc_sram_dword(rdev,
+ pi->soft_regs_start + reg_offset,
+ value, pi->sram_end);
+}
+#endif
+
+static int ci_write_smc_soft_register(struct radeon_device *rdev,
+ u16 reg_offset, u32 value)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ return ci_write_smc_sram_dword(rdev,
+ pi->soft_regs_start + reg_offset,
+ value, pi->sram_end);
+}
+
+static void ci_init_fps_limits(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ SMU7_Discrete_DpmTable *table = &pi->smc_state_table;
+
+ if (pi->caps_fps) {
+ u16 tmp;
+
+ tmp = 45;
+ table->FpsHighT = cpu_to_be16(tmp);
+
+ tmp = 30;
+ table->FpsLowT = cpu_to_be16(tmp);
+ }
+}
+
+static int ci_update_sclk_t(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ int ret = 0;
+ u32 low_sclk_interrupt_t = 0;
+
+ if (pi->caps_sclk_throttle_low_notification) {
+ low_sclk_interrupt_t = cpu_to_be32(pi->low_sclk_interrupt_t);
+
+ ret = ci_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Discrete_DpmTable, LowSclkInterruptT),
+ (u8 *)&low_sclk_interrupt_t,
+ sizeof(u32), pi->sram_end);
+
+ }
+
+ return ret;
+}
+
+static void ci_get_leakage_voltages(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u16 leakage_id, virtual_voltage_id;
+ u16 vddc, vddci;
+ int i;
+
+ pi->vddc_leakage.count = 0;
+ pi->vddci_leakage.count = 0;
+
+ if (radeon_atom_get_leakage_id_from_vbios(rdev, &leakage_id) == 0) {
+ for (i = 0; i < CISLANDS_MAX_LEAKAGE_COUNT; i++) {
+ virtual_voltage_id = ATOM_VIRTUAL_VOLTAGE_ID0 + i;
+ if (radeon_atom_get_leakage_vddc_based_on_leakage_params(rdev, &vddc, &vddci,
+ virtual_voltage_id,
+ leakage_id) == 0) {
+ if (vddc != 0 && vddc != virtual_voltage_id) {
+ pi->vddc_leakage.actual_voltage[pi->vddc_leakage.count] = vddc;
+ pi->vddc_leakage.leakage_id[pi->vddc_leakage.count] = virtual_voltage_id;
+ pi->vddc_leakage.count++;
+ }
+ if (vddci != 0 && vddci != virtual_voltage_id) {
+ pi->vddci_leakage.actual_voltage[pi->vddci_leakage.count] = vddci;
+ pi->vddci_leakage.leakage_id[pi->vddci_leakage.count] = virtual_voltage_id;
+ pi->vddci_leakage.count++;
+ }
+ }
+ }
+ }
+}
+
+static void ci_set_dpm_event_sources(struct radeon_device *rdev, u32 sources)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ bool want_thermal_protection;
+ enum radeon_dpm_event_src dpm_event_src;
+ u32 tmp;
+
+ switch (sources) {
+ case 0:
+ default:
+ want_thermal_protection = false;
+ break;
+ case (1 << RADEON_DPM_AUTO_THROTTLE_SRC_THERMAL):
+ want_thermal_protection = true;
+ dpm_event_src = RADEON_DPM_EVENT_SRC_DIGITAL;
+ break;
+ case (1 << RADEON_DPM_AUTO_THROTTLE_SRC_EXTERNAL):
+ want_thermal_protection = true;
+ dpm_event_src = RADEON_DPM_EVENT_SRC_EXTERNAL;
+ break;
+ case ((1 << RADEON_DPM_AUTO_THROTTLE_SRC_EXTERNAL) |
+ (1 << RADEON_DPM_AUTO_THROTTLE_SRC_THERMAL)):
+ want_thermal_protection = true;
+ dpm_event_src = RADEON_DPM_EVENT_SRC_DIGIAL_OR_EXTERNAL;
+ break;
+ }
+
+ if (want_thermal_protection) {
+#if 0
+ /* XXX: need to figure out how to handle this properly */
+ tmp = RREG32_SMC(CG_THERMAL_CTRL);
+ tmp &= DPM_EVENT_SRC_MASK;
+ tmp |= DPM_EVENT_SRC(dpm_event_src);
+ WREG32_SMC(CG_THERMAL_CTRL, tmp);
+#endif
+
+ tmp = RREG32_SMC(GENERAL_PWRMGT);
+ if (pi->thermal_protection)
+ tmp &= ~THERMAL_PROTECTION_DIS;
+ else
+ tmp |= THERMAL_PROTECTION_DIS;
+ WREG32_SMC(GENERAL_PWRMGT, tmp);
+ } else {
+ tmp = RREG32_SMC(GENERAL_PWRMGT);
+ tmp |= THERMAL_PROTECTION_DIS;
+ WREG32_SMC(GENERAL_PWRMGT, tmp);
+ }
+}
+
+static void ci_enable_auto_throttle_source(struct radeon_device *rdev,
+ enum radeon_dpm_auto_throttle_src source,
+ bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ if (enable) {
+ if (!(pi->active_auto_throttle_sources & (1 << source))) {
+ pi->active_auto_throttle_sources |= 1 << source;
+ ci_set_dpm_event_sources(rdev, pi->active_auto_throttle_sources);
+ }
+ } else {
+ if (pi->active_auto_throttle_sources & (1 << source)) {
+ pi->active_auto_throttle_sources &= ~(1 << source);
+ ci_set_dpm_event_sources(rdev, pi->active_auto_throttle_sources);
+ }
+ }
+}
+
+static void ci_enable_vr_hot_gpio_interrupt(struct radeon_device *rdev)
+{
+ if (rdev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_REGULATOR_HOT)
+ ci_send_msg_to_smc(rdev, PPSMC_MSG_EnableVRHotGPIOInterrupt);
+}
+
+static int ci_unfreeze_sclk_mclk_dpm(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ PPSMC_Result smc_result;
+
+ if (!pi->need_update_smu7_dpm_table)
+ return 0;
+
+ if ((!pi->sclk_dpm_key_disabled) &&
+ (pi->need_update_smu7_dpm_table & (DPMTABLE_OD_UPDATE_SCLK | DPMTABLE_UPDATE_SCLK))) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_SCLKDPM_UnfreezeLevel);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+
+ if ((!pi->mclk_dpm_key_disabled) &&
+ (pi->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_MCLK)) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_MCLKDPM_UnfreezeLevel);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+
+ pi->need_update_smu7_dpm_table = 0;
+ return 0;
+}
+
+static int ci_enable_sclk_mclk_dpm(struct radeon_device *rdev, bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ PPSMC_Result smc_result;
+
+ if (enable) {
+ if (!pi->sclk_dpm_key_disabled) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_DPM_Enable);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+
+ if (!pi->mclk_dpm_key_disabled) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_MCLKDPM_Enable);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+
+ WREG32_P(MC_SEQ_CNTL_3, CAC_EN, ~CAC_EN);
+
+ WREG32_SMC(LCAC_MC0_CNTL, 0x05);
+ WREG32_SMC(LCAC_MC1_CNTL, 0x05);
+ WREG32_SMC(LCAC_CPL_CNTL, 0x100005);
+
+ udelay(10);
+
+ WREG32_SMC(LCAC_MC0_CNTL, 0x400005);
+ WREG32_SMC(LCAC_MC1_CNTL, 0x400005);
+ WREG32_SMC(LCAC_CPL_CNTL, 0x500005);
+ }
+ } else {
+ if (!pi->sclk_dpm_key_disabled) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_DPM_Disable);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+
+ if (!pi->mclk_dpm_key_disabled) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_MCLKDPM_Disable);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+ }
+
+ return 0;
+}
+
+static int ci_start_dpm(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ PPSMC_Result smc_result;
+ int ret;
+ u32 tmp;
+
+ tmp = RREG32_SMC(GENERAL_PWRMGT);
+ tmp |= GLOBAL_PWRMGT_EN;
+ WREG32_SMC(GENERAL_PWRMGT, tmp);
+
+ tmp = RREG32_SMC(SCLK_PWRMGT_CNTL);
+ tmp |= DYNAMIC_PM_EN;
+ WREG32_SMC(SCLK_PWRMGT_CNTL, tmp);
+
+ ci_write_smc_soft_register(rdev, offsetof(SMU7_SoftRegisters, VoltageChangeTimeout), 0x1000);
+
+ WREG32_P(BIF_LNCNT_RESET, 0, ~RESET_LNCNT_EN);
+
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_Voltage_Cntl_Enable);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+
+ ret = ci_enable_sclk_mclk_dpm(rdev, true);
+ if (ret)
+ return ret;
+
+ if (!pi->pcie_dpm_key_disabled) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_PCIeDPM_Enable);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int ci_freeze_sclk_mclk_dpm(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ PPSMC_Result smc_result;
+
+ if (!pi->need_update_smu7_dpm_table)
+ return 0;
+
+ if ((!pi->sclk_dpm_key_disabled) &&
+ (pi->need_update_smu7_dpm_table & (DPMTABLE_OD_UPDATE_SCLK | DPMTABLE_UPDATE_SCLK))) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_SCLKDPM_FreezeLevel);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+
+ if ((!pi->mclk_dpm_key_disabled) &&
+ (pi->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_MCLK)) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_MCLKDPM_FreezeLevel);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int ci_stop_dpm(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ PPSMC_Result smc_result;
+ int ret;
+ u32 tmp;
+
+ tmp = RREG32_SMC(GENERAL_PWRMGT);
+ tmp &= ~GLOBAL_PWRMGT_EN;
+ WREG32_SMC(GENERAL_PWRMGT, tmp);
+
+ tmp = RREG32(SCLK_PWRMGT_CNTL);
+ tmp &= ~DYNAMIC_PM_EN;
+ WREG32_SMC(SCLK_PWRMGT_CNTL, tmp);
+
+ if (!pi->pcie_dpm_key_disabled) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_PCIeDPM_Disable);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+
+ ret = ci_enable_sclk_mclk_dpm(rdev, false);
+ if (ret)
+ return ret;
+
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_Voltage_Cntl_Disable);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+
+ return 0;
+}
+
+static void ci_enable_sclk_control(struct radeon_device *rdev, bool enable)
+{
+ u32 tmp = RREG32_SMC(SCLK_PWRMGT_CNTL);
+
+ if (enable)
+ tmp &= ~SCLK_PWRMGT_OFF;
+ else
+ tmp |= SCLK_PWRMGT_OFF;
+ WREG32_SMC(SCLK_PWRMGT_CNTL, tmp);
+}
+
+#if 0
+static int ci_notify_hw_of_power_source(struct radeon_device *rdev,
+ bool ac_power)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct radeon_cac_tdp_table *cac_tdp_table =
+ rdev->pm.dpm.dyn_state.cac_tdp_table;
+ u32 power_limit;
+
+ if (ac_power)
+ power_limit = (u32)(cac_tdp_table->maximum_power_delivery_limit * 256);
+ else
+ power_limit = (u32)(cac_tdp_table->battery_power_limit * 256);
+
+ ci_set_power_limit(rdev, power_limit);
+
+ if (pi->caps_automatic_dc_transition) {
+ if (ac_power)
+ ci_send_msg_to_smc(rdev, PPSMC_MSG_RunningOnAC);
+ else
+ ci_send_msg_to_smc(rdev, PPSMC_MSG_Remove_DC_Clamp);
+ }
+
+ return 0;
+}
+#endif
+
+static PPSMC_Result ci_send_msg_to_smc_with_parameter(struct radeon_device *rdev,
+ PPSMC_Msg msg, u32 parameter)
+{
+ WREG32(SMC_MSG_ARG_0, parameter);
+ return ci_send_msg_to_smc(rdev, msg);
+}
+
+static PPSMC_Result ci_send_msg_to_smc_return_parameter(struct radeon_device *rdev,
+ PPSMC_Msg msg, u32 *parameter)
+{
+ PPSMC_Result smc_result;
+
+ smc_result = ci_send_msg_to_smc(rdev, msg);
+
+ if ((smc_result == PPSMC_Result_OK) && parameter)
+ *parameter = RREG32(SMC_MSG_ARG_0);
+
+ return smc_result;
+}
+
+static int ci_dpm_force_state_sclk(struct radeon_device *rdev, u32 n)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ if (!pi->sclk_dpm_key_disabled) {
+ PPSMC_Result smc_result =
+ ci_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_DPM_ForceState, n);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int ci_dpm_force_state_mclk(struct radeon_device *rdev, u32 n)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ if (!pi->mclk_dpm_key_disabled) {
+ PPSMC_Result smc_result =
+ ci_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_MCLKDPM_ForceState, n);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int ci_dpm_force_state_pcie(struct radeon_device *rdev, u32 n)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ if (!pi->pcie_dpm_key_disabled) {
+ PPSMC_Result smc_result =
+ ci_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_PCIeDPM_ForceLevel, n);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int ci_set_power_limit(struct radeon_device *rdev, u32 n)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ if (pi->power_containment_features & POWERCONTAINMENT_FEATURE_PkgPwrLimit) {
+ PPSMC_Result smc_result =
+ ci_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_PkgPwrSetLimit, n);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int ci_set_overdrive_target_tdp(struct radeon_device *rdev,
+ u32 target_tdp)
+{
+ PPSMC_Result smc_result =
+ ci_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_OverDriveSetTargetTdp, target_tdp);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ return 0;
+}
+
+static int ci_set_boot_state(struct radeon_device *rdev)
+{
+ return ci_enable_sclk_mclk_dpm(rdev, false);
+}
+
+static u32 ci_get_average_sclk_freq(struct radeon_device *rdev)
+{
+ u32 sclk_freq;
+ PPSMC_Result smc_result =
+ ci_send_msg_to_smc_return_parameter(rdev,
+ PPSMC_MSG_API_GetSclkFrequency,
+ &sclk_freq);
+ if (smc_result != PPSMC_Result_OK)
+ sclk_freq = 0;
+
+ return sclk_freq;
+}
+
+static u32 ci_get_average_mclk_freq(struct radeon_device *rdev)
+{
+ u32 mclk_freq;
+ PPSMC_Result smc_result =
+ ci_send_msg_to_smc_return_parameter(rdev,
+ PPSMC_MSG_API_GetMclkFrequency,
+ &mclk_freq);
+ if (smc_result != PPSMC_Result_OK)
+ mclk_freq = 0;
+
+ return mclk_freq;
+}
+
+static void ci_dpm_start_smc(struct radeon_device *rdev)
+{
+ int i;
+
+ ci_program_jump_on_start(rdev);
+ ci_start_smc_clock(rdev);
+ ci_start_smc(rdev);
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ if (RREG32_SMC(FIRMWARE_FLAGS) & INTERRUPTS_ENABLED)
+ break;
+ }
+}
+
+static void ci_dpm_stop_smc(struct radeon_device *rdev)
+{
+ ci_reset_smc(rdev);
+ ci_stop_smc_clock(rdev);
+}
+
+static int ci_process_firmware_header(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 tmp;
+ int ret;
+
+ ret = ci_read_smc_sram_dword(rdev,
+ SMU7_FIRMWARE_HEADER_LOCATION +
+ offsetof(SMU7_Firmware_Header, DpmTable),
+ &tmp, pi->sram_end);
+ if (ret)
+ return ret;
+
+ pi->dpm_table_start = tmp;
+
+ ret = ci_read_smc_sram_dword(rdev,
+ SMU7_FIRMWARE_HEADER_LOCATION +
+ offsetof(SMU7_Firmware_Header, SoftRegisters),
+ &tmp, pi->sram_end);
+ if (ret)
+ return ret;
+
+ pi->soft_regs_start = tmp;
+
+ ret = ci_read_smc_sram_dword(rdev,
+ SMU7_FIRMWARE_HEADER_LOCATION +
+ offsetof(SMU7_Firmware_Header, mcRegisterTable),
+ &tmp, pi->sram_end);
+ if (ret)
+ return ret;
+
+ pi->mc_reg_table_start = tmp;
+
+ ret = ci_read_smc_sram_dword(rdev,
+ SMU7_FIRMWARE_HEADER_LOCATION +
+ offsetof(SMU7_Firmware_Header, FanTable),
+ &tmp, pi->sram_end);
+ if (ret)
+ return ret;
+
+ pi->fan_table_start = tmp;
+
+ ret = ci_read_smc_sram_dword(rdev,
+ SMU7_FIRMWARE_HEADER_LOCATION +
+ offsetof(SMU7_Firmware_Header, mcArbDramTimingTable),
+ &tmp, pi->sram_end);
+ if (ret)
+ return ret;
+
+ pi->arb_table_start = tmp;
+
+ return 0;
+}
+
+static void ci_read_clock_registers(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ pi->clock_registers.cg_spll_func_cntl =
+ RREG32_SMC(CG_SPLL_FUNC_CNTL);
+ pi->clock_registers.cg_spll_func_cntl_2 =
+ RREG32_SMC(CG_SPLL_FUNC_CNTL_2);
+ pi->clock_registers.cg_spll_func_cntl_3 =
+ RREG32_SMC(CG_SPLL_FUNC_CNTL_3);
+ pi->clock_registers.cg_spll_func_cntl_4 =
+ RREG32_SMC(CG_SPLL_FUNC_CNTL_4);
+ pi->clock_registers.cg_spll_spread_spectrum =
+ RREG32_SMC(CG_SPLL_SPREAD_SPECTRUM);
+ pi->clock_registers.cg_spll_spread_spectrum_2 =
+ RREG32_SMC(CG_SPLL_SPREAD_SPECTRUM_2);
+ pi->clock_registers.dll_cntl = RREG32(DLL_CNTL);
+ pi->clock_registers.mclk_pwrmgt_cntl = RREG32(MCLK_PWRMGT_CNTL);
+ pi->clock_registers.mpll_ad_func_cntl = RREG32(MPLL_AD_FUNC_CNTL);
+ pi->clock_registers.mpll_dq_func_cntl = RREG32(MPLL_DQ_FUNC_CNTL);
+ pi->clock_registers.mpll_func_cntl = RREG32(MPLL_FUNC_CNTL);
+ pi->clock_registers.mpll_func_cntl_1 = RREG32(MPLL_FUNC_CNTL_1);
+ pi->clock_registers.mpll_func_cntl_2 = RREG32(MPLL_FUNC_CNTL_2);
+ pi->clock_registers.mpll_ss1 = RREG32(MPLL_SS1);
+ pi->clock_registers.mpll_ss2 = RREG32(MPLL_SS2);
+}
+
+static void ci_init_sclk_t(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ pi->low_sclk_interrupt_t = 0;
+}
+
+static void ci_enable_thermal_protection(struct radeon_device *rdev,
+ bool enable)
+{
+ u32 tmp = RREG32_SMC(GENERAL_PWRMGT);
+
+ if (enable)
+ tmp &= ~THERMAL_PROTECTION_DIS;
+ else
+ tmp |= THERMAL_PROTECTION_DIS;
+ WREG32_SMC(GENERAL_PWRMGT, tmp);
+}
+
+static void ci_enable_acpi_power_management(struct radeon_device *rdev)
+{
+ u32 tmp = RREG32_SMC(GENERAL_PWRMGT);
+
+ tmp |= STATIC_PM_EN;
+
+ WREG32_SMC(GENERAL_PWRMGT, tmp);
+}
+
+#if 0
+static int ci_enter_ulp_state(struct radeon_device *rdev)
+{
+
+ WREG32(SMC_MESSAGE_0, PPSMC_MSG_SwitchToMinimumPower);
+
+ udelay(25000);
+
+ return 0;
+}
+
+static int ci_exit_ulp_state(struct radeon_device *rdev)
+{
+ int i;
+
+ WREG32(SMC_MESSAGE_0, PPSMC_MSG_ResumeFromMinimumPower);
+
+ udelay(7000);
+
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ if (RREG32(SMC_RESP_0) == 1)
+ break;
+ udelay(1000);
+ }
+
+ return 0;
+}
+#endif
+
+static int ci_notify_smc_display_change(struct radeon_device *rdev,
+ bool has_display)
+{
+ PPSMC_Msg msg = has_display ? PPSMC_MSG_HasDisplay : PPSMC_MSG_NoDisplay;
+
+ return (ci_send_msg_to_smc(rdev, msg) == PPSMC_Result_OK) ? 0 : -EINVAL;
+}
+
+static int ci_enable_ds_master_switch(struct radeon_device *rdev,
+ bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ if (enable) {
+ if (pi->caps_sclk_ds) {
+ if (ci_send_msg_to_smc(rdev, PPSMC_MSG_MASTER_DeepSleep_ON) != PPSMC_Result_OK)
+ return -EINVAL;
+ } else {
+ if (ci_send_msg_to_smc(rdev, PPSMC_MSG_MASTER_DeepSleep_OFF) != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+ } else {
+ if (pi->caps_sclk_ds) {
+ if (ci_send_msg_to_smc(rdev, PPSMC_MSG_MASTER_DeepSleep_OFF) != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+ }
+
+ return 0;
+}
+
+static void ci_program_display_gap(struct radeon_device *rdev)
+{
+ u32 tmp = RREG32_SMC(CG_DISPLAY_GAP_CNTL);
+ u32 pre_vbi_time_in_us;
+ u32 frame_time_in_us;
+ u32 ref_clock = rdev->clock.spll.reference_freq;
+ u32 refresh_rate = r600_dpm_get_vrefresh(rdev);
+ u32 vblank_time = r600_dpm_get_vblank_time(rdev);
+
+ tmp &= ~DISP_GAP_MASK;
+ if (rdev->pm.dpm.new_active_crtc_count > 0)
+ tmp |= DISP_GAP(R600_PM_DISPLAY_GAP_VBLANK_OR_WM);
+ else
+ tmp |= DISP_GAP(R600_PM_DISPLAY_GAP_IGNORE);
+ WREG32_SMC(CG_DISPLAY_GAP_CNTL, tmp);
+
+ if (refresh_rate == 0)
+ refresh_rate = 60;
+ if (vblank_time == 0xffffffff)
+ vblank_time = 500;
+ frame_time_in_us = 1000000 / refresh_rate;
+ pre_vbi_time_in_us =
+ frame_time_in_us - 200 - vblank_time;
+ tmp = pre_vbi_time_in_us * (ref_clock / 100);
+
+ WREG32_SMC(CG_DISPLAY_GAP_CNTL2, tmp);
+ ci_write_smc_soft_register(rdev, offsetof(SMU7_SoftRegisters, PreVBlankGap), 0x64);
+ ci_write_smc_soft_register(rdev, offsetof(SMU7_SoftRegisters, VBlankTimeout), (frame_time_in_us - pre_vbi_time_in_us));
+
+
+ ci_notify_smc_display_change(rdev, (rdev->pm.dpm.new_active_crtc_count == 1));
+
+}
+
+static void ci_enable_spread_spectrum(struct radeon_device *rdev, bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 tmp;
+
+ if (enable) {
+ if (pi->caps_sclk_ss_support) {
+ tmp = RREG32_SMC(GENERAL_PWRMGT);
+ tmp |= DYN_SPREAD_SPECTRUM_EN;
+ WREG32_SMC(GENERAL_PWRMGT, tmp);
+ }
+ } else {
+ tmp = RREG32_SMC(CG_SPLL_SPREAD_SPECTRUM);
+ tmp &= ~SSEN;
+ WREG32_SMC(CG_SPLL_SPREAD_SPECTRUM, tmp);
+
+ tmp = RREG32_SMC(GENERAL_PWRMGT);
+ tmp &= ~DYN_SPREAD_SPECTRUM_EN;
+ WREG32_SMC(GENERAL_PWRMGT, tmp);
+ }
+}
+
+static void ci_program_sstp(struct radeon_device *rdev)
+{
+ WREG32_SMC(CG_SSP, (SSTU(R600_SSTU_DFLT) | SST(R600_SST_DFLT)));
+}
+
+static void ci_enable_display_gap(struct radeon_device *rdev)
+{
+ u32 tmp = RREG32_SMC(CG_DISPLAY_GAP_CNTL);
+
+ tmp &= ~(DISP_GAP_MASK | DISP_GAP_MCHG_MASK);
+ tmp |= (DISP_GAP(R600_PM_DISPLAY_GAP_IGNORE) |
+ DISP_GAP_MCHG(R600_PM_DISPLAY_GAP_VBLANK));
+
+ WREG32_SMC(CG_DISPLAY_GAP_CNTL, tmp);
+}
+
+static void ci_program_vc(struct radeon_device *rdev)
+{
+ u32 tmp;
+
+ tmp = RREG32_SMC(SCLK_PWRMGT_CNTL);
+ tmp &= ~(RESET_SCLK_CNT | RESET_BUSY_CNT);
+ WREG32_SMC(SCLK_PWRMGT_CNTL, tmp);
+
+ WREG32_SMC(CG_FTV_0, CISLANDS_VRC_DFLT0);
+ WREG32_SMC(CG_FTV_1, CISLANDS_VRC_DFLT1);
+ WREG32_SMC(CG_FTV_2, CISLANDS_VRC_DFLT2);
+ WREG32_SMC(CG_FTV_3, CISLANDS_VRC_DFLT3);
+ WREG32_SMC(CG_FTV_4, CISLANDS_VRC_DFLT4);
+ WREG32_SMC(CG_FTV_5, CISLANDS_VRC_DFLT5);
+ WREG32_SMC(CG_FTV_6, CISLANDS_VRC_DFLT6);
+ WREG32_SMC(CG_FTV_7, CISLANDS_VRC_DFLT7);
+}
+
+static void ci_clear_vc(struct radeon_device *rdev)
+{
+ u32 tmp;
+
+ tmp = RREG32_SMC(SCLK_PWRMGT_CNTL);
+ tmp |= (RESET_SCLK_CNT | RESET_BUSY_CNT);
+ WREG32_SMC(SCLK_PWRMGT_CNTL, tmp);
+
+ WREG32_SMC(CG_FTV_0, 0);
+ WREG32_SMC(CG_FTV_1, 0);
+ WREG32_SMC(CG_FTV_2, 0);
+ WREG32_SMC(CG_FTV_3, 0);
+ WREG32_SMC(CG_FTV_4, 0);
+ WREG32_SMC(CG_FTV_5, 0);
+ WREG32_SMC(CG_FTV_6, 0);
+ WREG32_SMC(CG_FTV_7, 0);
+}
+
+static int ci_upload_firmware(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ int i, ret;
+
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ if (RREG32_SMC(RCU_UC_EVENTS) & BOOT_SEQ_DONE)
+ break;
+ }
+ WREG32_SMC(SMC_SYSCON_MISC_CNTL, 1);
+
+ ci_stop_smc_clock(rdev);
+ ci_reset_smc(rdev);
+
+ ret = ci_load_smc_ucode(rdev, pi->sram_end);
+
+ return ret;
+
+}
+
+static int ci_get_svi2_voltage_table(struct radeon_device *rdev,
+ struct radeon_clock_voltage_dependency_table *voltage_dependency_table,
+ struct atom_voltage_table *voltage_table)
+{
+ u32 i;
+
+ if (voltage_dependency_table == NULL)
+ return -EINVAL;
+
+ voltage_table->mask_low = 0;
+ voltage_table->phase_delay = 0;
+
+ voltage_table->count = voltage_dependency_table->count;
+ for (i = 0; i < voltage_table->count; i++) {
+ voltage_table->entries[i].value = voltage_dependency_table->entries[i].v;
+ voltage_table->entries[i].smio_low = 0;
+ }
+
+ return 0;
+}
+
+static int ci_construct_voltage_tables(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ int ret;
+
+ if (pi->voltage_control == CISLANDS_VOLTAGE_CONTROL_BY_GPIO) {
+ ret = radeon_atom_get_voltage_table(rdev, VOLTAGE_TYPE_VDDC,
+ VOLTAGE_OBJ_GPIO_LUT,
+ &pi->vddc_voltage_table);
+ if (ret)
+ return ret;
+ } else if (pi->voltage_control == CISLANDS_VOLTAGE_CONTROL_BY_SVID2) {
+ ret = ci_get_svi2_voltage_table(rdev,
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk,
+ &pi->vddc_voltage_table);
+ if (ret)
+ return ret;
+ }
+
+ if (pi->vddc_voltage_table.count > SMU7_MAX_LEVELS_VDDC)
+ si_trim_voltage_table_to_fit_state_table(rdev, SMU7_MAX_LEVELS_VDDC,
+ &pi->vddc_voltage_table);
+
+ if (pi->vddci_control == CISLANDS_VOLTAGE_CONTROL_BY_GPIO) {
+ ret = radeon_atom_get_voltage_table(rdev, VOLTAGE_TYPE_VDDCI,
+ VOLTAGE_OBJ_GPIO_LUT,
+ &pi->vddci_voltage_table);
+ if (ret)
+ return ret;
+ } else if (pi->vddci_control == CISLANDS_VOLTAGE_CONTROL_BY_SVID2) {
+ ret = ci_get_svi2_voltage_table(rdev,
+ &rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk,
+ &pi->vddci_voltage_table);
+ if (ret)
+ return ret;
+ }
+
+ if (pi->vddci_voltage_table.count > SMU7_MAX_LEVELS_VDDCI)
+ si_trim_voltage_table_to_fit_state_table(rdev, SMU7_MAX_LEVELS_VDDCI,
+ &pi->vddci_voltage_table);
+
+ if (pi->mvdd_control == CISLANDS_VOLTAGE_CONTROL_BY_GPIO) {
+ ret = radeon_atom_get_voltage_table(rdev, VOLTAGE_TYPE_MVDDC,
+ VOLTAGE_OBJ_GPIO_LUT,
+ &pi->mvdd_voltage_table);
+ if (ret)
+ return ret;
+ } else if (pi->mvdd_control == CISLANDS_VOLTAGE_CONTROL_BY_SVID2) {
+ ret = ci_get_svi2_voltage_table(rdev,
+ &rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk,
+ &pi->mvdd_voltage_table);
+ if (ret)
+ return ret;
+ }
+
+ if (pi->mvdd_voltage_table.count > SMU7_MAX_LEVELS_MVDD)
+ si_trim_voltage_table_to_fit_state_table(rdev, SMU7_MAX_LEVELS_MVDD,
+ &pi->mvdd_voltage_table);
+
+ return 0;
+}
+
+static void ci_populate_smc_voltage_table(struct radeon_device *rdev,
+ struct atom_voltage_table_entry *voltage_table,
+ SMU7_Discrete_VoltageLevel *smc_voltage_table)
+{
+ int ret;
+
+ ret = ci_get_std_voltage_value_sidd(rdev, voltage_table,
+ &smc_voltage_table->StdVoltageHiSidd,
+ &smc_voltage_table->StdVoltageLoSidd);
+
+ if (ret) {
+ smc_voltage_table->StdVoltageHiSidd = voltage_table->value * VOLTAGE_SCALE;
+ smc_voltage_table->StdVoltageLoSidd = voltage_table->value * VOLTAGE_SCALE;
+ }
+
+ smc_voltage_table->Voltage = cpu_to_be16(voltage_table->value * VOLTAGE_SCALE);
+ smc_voltage_table->StdVoltageHiSidd =
+ cpu_to_be16(smc_voltage_table->StdVoltageHiSidd);
+ smc_voltage_table->StdVoltageLoSidd =
+ cpu_to_be16(smc_voltage_table->StdVoltageLoSidd);
+}
+
+static int ci_populate_smc_vddc_table(struct radeon_device *rdev,
+ SMU7_Discrete_DpmTable *table)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ unsigned int count;
+
+ table->VddcLevelCount = pi->vddc_voltage_table.count;
+ for (count = 0; count < table->VddcLevelCount; count++) {
+ ci_populate_smc_voltage_table(rdev,
+ &pi->vddc_voltage_table.entries[count],
+ &table->VddcLevel[count]);
+
+ if (pi->voltage_control == CISLANDS_VOLTAGE_CONTROL_BY_GPIO)
+ table->VddcLevel[count].Smio |=
+ pi->vddc_voltage_table.entries[count].smio_low;
+ else
+ table->VddcLevel[count].Smio = 0;
+ }
+ table->VddcLevelCount = cpu_to_be32(table->VddcLevelCount);
+
+ return 0;
+}
+
+static int ci_populate_smc_vddci_table(struct radeon_device *rdev,
+ SMU7_Discrete_DpmTable *table)
+{
+ unsigned int count;
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ table->VddciLevelCount = pi->vddci_voltage_table.count;
+ for (count = 0; count < table->VddciLevelCount; count++) {
+ ci_populate_smc_voltage_table(rdev,
+ &pi->vddci_voltage_table.entries[count],
+ &table->VddciLevel[count]);
+
+ if (pi->vddci_control == CISLANDS_VOLTAGE_CONTROL_BY_GPIO)
+ table->VddciLevel[count].Smio |=
+ pi->vddci_voltage_table.entries[count].smio_low;
+ else
+ table->VddciLevel[count].Smio = 0;
+ }
+ table->VddciLevelCount = cpu_to_be32(table->VddciLevelCount);
+
+ return 0;
+}
+
+static int ci_populate_smc_mvdd_table(struct radeon_device *rdev,
+ SMU7_Discrete_DpmTable *table)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ unsigned int count;
+
+ table->MvddLevelCount = pi->mvdd_voltage_table.count;
+ for (count = 0; count < table->MvddLevelCount; count++) {
+ ci_populate_smc_voltage_table(rdev,
+ &pi->mvdd_voltage_table.entries[count],
+ &table->MvddLevel[count]);
+
+ if (pi->mvdd_control == CISLANDS_VOLTAGE_CONTROL_BY_GPIO)
+ table->MvddLevel[count].Smio |=
+ pi->mvdd_voltage_table.entries[count].smio_low;
+ else
+ table->MvddLevel[count].Smio = 0;
+ }
+ table->MvddLevelCount = cpu_to_be32(table->MvddLevelCount);
+
+ return 0;
+}
+
+static int ci_populate_smc_voltage_tables(struct radeon_device *rdev,
+ SMU7_Discrete_DpmTable *table)
+{
+ int ret;
+
+ ret = ci_populate_smc_vddc_table(rdev, table);
+ if (ret)
+ return ret;
+
+ ret = ci_populate_smc_vddci_table(rdev, table);
+ if (ret)
+ return ret;
+
+ ret = ci_populate_smc_mvdd_table(rdev, table);
+ if (ret)
+ return ret;
+
+ return 0;
+}
+
+static int ci_populate_mvdd_value(struct radeon_device *rdev, u32 mclk,
+ SMU7_Discrete_VoltageLevel *voltage)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 i = 0;
+
+ if (pi->mvdd_control != CISLANDS_VOLTAGE_CONTROL_NONE) {
+ for (i = 0; i < rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.count; i++) {
+ if (mclk <= rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries[i].clk) {
+ voltage->Voltage = pi->mvdd_voltage_table.entries[i].value;
+ break;
+ }
+ }
+
+ if (i >= rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.count)
+ return -EINVAL;
+ }
+
+ return -EINVAL;
+}
+
+static int ci_get_std_voltage_value_sidd(struct radeon_device *rdev,
+ struct atom_voltage_table_entry *voltage_table,
+ u16 *std_voltage_hi_sidd, u16 *std_voltage_lo_sidd)
+{
+ u16 v_index, idx;
+ bool voltage_found = false;
+ *std_voltage_hi_sidd = voltage_table->value * VOLTAGE_SCALE;
+ *std_voltage_lo_sidd = voltage_table->value * VOLTAGE_SCALE;
+
+ if (rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries == NULL)
+ return -EINVAL;
+
+ if (rdev->pm.dpm.dyn_state.cac_leakage_table.entries) {
+ for (v_index = 0; (u32)v_index < rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.count; v_index++) {
+ if (voltage_table->value ==
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[v_index].v) {
+ voltage_found = true;
+ if ((u32)v_index < rdev->pm.dpm.dyn_state.cac_leakage_table.count)
+ idx = v_index;
+ else
+ idx = rdev->pm.dpm.dyn_state.cac_leakage_table.count - 1;
+ *std_voltage_lo_sidd =
+ rdev->pm.dpm.dyn_state.cac_leakage_table.entries[idx].vddc * VOLTAGE_SCALE;
+ *std_voltage_hi_sidd =
+ rdev->pm.dpm.dyn_state.cac_leakage_table.entries[idx].leakage * VOLTAGE_SCALE;
+ break;
+ }
+ }
+
+ if (!voltage_found) {
+ for (v_index = 0; (u32)v_index < rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.count; v_index++) {
+ if (voltage_table->value <=
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[v_index].v) {
+ voltage_found = true;
+ if ((u32)v_index < rdev->pm.dpm.dyn_state.cac_leakage_table.count)
+ idx = v_index;
+ else
+ idx = rdev->pm.dpm.dyn_state.cac_leakage_table.count - 1;
+ *std_voltage_lo_sidd =
+ rdev->pm.dpm.dyn_state.cac_leakage_table.entries[idx].vddc * VOLTAGE_SCALE;
+ *std_voltage_hi_sidd =
+ rdev->pm.dpm.dyn_state.cac_leakage_table.entries[idx].leakage * VOLTAGE_SCALE;
+ break;
+ }
+ }
+ }
+ }
+
+ return 0;
+}
+
+static void ci_populate_phase_value_based_on_sclk(struct radeon_device *rdev,
+ const struct radeon_phase_shedding_limits_table *limits,
+ u32 sclk,
+ u32 *phase_shedding)
+{
+ unsigned int i;
+
+ *phase_shedding = 1;
+
+ for (i = 0; i < limits->count; i++) {
+ if (sclk < limits->entries[i].sclk) {
+ *phase_shedding = i;
+ break;
+ }
+ }
+}
+
+static void ci_populate_phase_value_based_on_mclk(struct radeon_device *rdev,
+ const struct radeon_phase_shedding_limits_table *limits,
+ u32 mclk,
+ u32 *phase_shedding)
+{
+ unsigned int i;
+
+ *phase_shedding = 1;
+
+ for (i = 0; i < limits->count; i++) {
+ if (mclk < limits->entries[i].mclk) {
+ *phase_shedding = i;
+ break;
+ }
+ }
+}
+
+static int ci_init_arb_table_index(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 tmp;
+ int ret;
+
+ ret = ci_read_smc_sram_dword(rdev, pi->arb_table_start,
+ &tmp, pi->sram_end);
+ if (ret)
+ return ret;
+
+ tmp &= 0x00FFFFFF;
+ tmp |= MC_CG_ARB_FREQ_F1 << 24;
+
+ return ci_write_smc_sram_dword(rdev, pi->arb_table_start,
+ tmp, pi->sram_end);
+}
+
+static int ci_get_dependency_volt_by_clk(struct radeon_device *rdev,
+ struct radeon_clock_voltage_dependency_table *allowed_clock_voltage_table,
+ u32 clock, u32 *voltage)
+{
+ u32 i = 0;
+
+ if (allowed_clock_voltage_table->count == 0)
+ return -EINVAL;
+
+ for (i = 0; i < allowed_clock_voltage_table->count; i++) {
+ if (allowed_clock_voltage_table->entries[i].clk >= clock) {
+ *voltage = allowed_clock_voltage_table->entries[i].v;
+ return 0;
+ }
+ }
+
+ *voltage = allowed_clock_voltage_table->entries[i-1].v;
+
+ return 0;
+}
+
+static u8 ci_get_sleep_divider_id_from_clock(struct radeon_device *rdev,
+ u32 sclk, u32 min_sclk_in_sr)
+{
+ u32 i;
+ u32 tmp;
+ u32 min = (min_sclk_in_sr > CISLAND_MINIMUM_ENGINE_CLOCK) ?
+ min_sclk_in_sr : CISLAND_MINIMUM_ENGINE_CLOCK;
+
+ if (sclk < min)
+ return 0;
+
+ for (i = CISLAND_MAX_DEEPSLEEP_DIVIDER_ID; ; i--) {
+ tmp = sclk / (1 << i);
+ if (tmp >= min || i == 0)
+ break;
+ }
+
+ return (u8)i;
+}
+
+static int ci_initial_switch_from_arb_f0_to_f1(struct radeon_device *rdev)
+{
+ return ni_copy_and_switch_arb_sets(rdev, MC_CG_ARB_FREQ_F0, MC_CG_ARB_FREQ_F1);
+}
+
+static int ci_reset_to_default(struct radeon_device *rdev)
+{
+ return (ci_send_msg_to_smc(rdev, PPSMC_MSG_ResetToDefaults) == PPSMC_Result_OK) ?
+ 0 : -EINVAL;
+}
+
+static int ci_force_switch_to_arb_f0(struct radeon_device *rdev)
+{
+ u32 tmp;
+
+ tmp = (RREG32_SMC(SMC_SCRATCH9) & 0x0000ff00) >> 8;
+
+ if (tmp == MC_CG_ARB_FREQ_F0)
+ return 0;
+
+ return ni_copy_and_switch_arb_sets(rdev, tmp, MC_CG_ARB_FREQ_F0);
+}
+
+static int ci_populate_memory_timing_parameters(struct radeon_device *rdev,
+ u32 sclk,
+ u32 mclk,
+ SMU7_Discrete_MCArbDramTimingTableEntry *arb_regs)
+{
+ u32 dram_timing;
+ u32 dram_timing2;
+ u32 burst_time;
+
+ radeon_atom_set_engine_dram_timings(rdev, sclk, mclk);
+
+ dram_timing = RREG32(MC_ARB_DRAM_TIMING);
+ dram_timing2 = RREG32(MC_ARB_DRAM_TIMING2);
+ burst_time = RREG32(MC_ARB_BURST_TIME) & STATE0_MASK;
+
+ arb_regs->McArbDramTiming = cpu_to_be32(dram_timing);
+ arb_regs->McArbDramTiming2 = cpu_to_be32(dram_timing2);
+ arb_regs->McArbBurstTime = (u8)burst_time;
+
+ return 0;
+}
+
+static int ci_do_program_memory_timing_parameters(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ SMU7_Discrete_MCArbDramTimingTable arb_regs;
+ u32 i, j;
+ int ret = 0;
+
+ memset(&arb_regs, 0, sizeof(SMU7_Discrete_MCArbDramTimingTable));
+
+ for (i = 0; i < pi->dpm_table.sclk_table.count; i++) {
+ for (j = 0; j < pi->dpm_table.mclk_table.count; j++) {
+ ret = ci_populate_memory_timing_parameters(rdev,
+ pi->dpm_table.sclk_table.dpm_levels[i].value,
+ pi->dpm_table.mclk_table.dpm_levels[j].value,
+ &arb_regs.entries[i][j]);
+ if (ret)
+ break;
+ }
+ }
+
+ if (ret == 0)
+ ret = ci_copy_bytes_to_smc(rdev,
+ pi->arb_table_start,
+ (u8 *)&arb_regs,
+ sizeof(SMU7_Discrete_MCArbDramTimingTable),
+ pi->sram_end);
+
+ return ret;
+}
+
+static int ci_program_memory_timing_parameters(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ if (pi->need_update_smu7_dpm_table == 0)
+ return 0;
+
+ return ci_do_program_memory_timing_parameters(rdev);
+}
+
+static void ci_populate_smc_initial_state(struct radeon_device *rdev,
+ struct radeon_ps *radeon_boot_state)
+{
+ struct ci_ps *boot_state = ci_get_ps(radeon_boot_state);
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 level = 0;
+
+ for (level = 0; level < rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.count; level++) {
+ if (rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[level].clk >=
+ boot_state->performance_levels[0].sclk) {
+ pi->smc_state_table.GraphicsBootLevel = level;
+ break;
+ }
+ }
+
+ for (level = 0; level < rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.count; level++) {
+ if (rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries[level].clk >=
+ boot_state->performance_levels[0].mclk) {
+ pi->smc_state_table.MemoryBootLevel = level;
+ break;
+ }
+ }
+}
+
+static u32 ci_get_dpm_level_enable_mask_value(struct ci_single_dpm_table *dpm_table)
+{
+ u32 i;
+ u32 mask_value = 0;
+
+ for (i = dpm_table->count; i > 0; i--) {
+ mask_value = mask_value << 1;
+ if (dpm_table->dpm_levels[i-1].enabled)
+ mask_value |= 0x1;
+ else
+ mask_value &= 0xFFFFFFFE;
+ }
+
+ return mask_value;
+}
+
+static void ci_populate_smc_link_level(struct radeon_device *rdev,
+ SMU7_Discrete_DpmTable *table)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_dpm_table *dpm_table = &pi->dpm_table;
+ u32 i;
+
+ for (i = 0; i < dpm_table->pcie_speed_table.count; i++) {
+ table->LinkLevel[i].PcieGenSpeed =
+ (u8)dpm_table->pcie_speed_table.dpm_levels[i].value;
+ table->LinkLevel[i].PcieLaneCount =
+ r600_encode_pci_lane_width(dpm_table->pcie_speed_table.dpm_levels[i].param1);
+ table->LinkLevel[i].EnabledForActivity = 1;
+ table->LinkLevel[i].DownT = cpu_to_be32(5);
+ table->LinkLevel[i].UpT = cpu_to_be32(30);
+ }
+
+ pi->smc_state_table.LinkLevelCount = (u8)dpm_table->pcie_speed_table.count;
+ pi->dpm_level_enable_mask.pcie_dpm_enable_mask =
+ ci_get_dpm_level_enable_mask_value(&dpm_table->pcie_speed_table);
+}
+
+static int ci_populate_smc_uvd_level(struct radeon_device *rdev,
+ SMU7_Discrete_DpmTable *table)
+{
+ u32 count;
+ struct atom_clock_dividers dividers;
+ int ret = -EINVAL;
+
+ table->UvdLevelCount =
+ rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count;
+
+ for (count = 0; count < table->UvdLevelCount; count++) {
+ table->UvdLevel[count].VclkFrequency =
+ rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[count].vclk;
+ table->UvdLevel[count].DclkFrequency =
+ rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[count].dclk;
+ table->UvdLevel[count].MinVddc =
+ rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[count].v * VOLTAGE_SCALE;
+ table->UvdLevel[count].MinVddcPhases = 1;
+
+ ret = radeon_atom_get_clock_dividers(rdev,
+ COMPUTE_GPUCLK_INPUT_FLAG_DEFAULT_GPUCLK,
+ table->UvdLevel[count].VclkFrequency, false, ÷rs);
+ if (ret)
+ return ret;
+
+ table->UvdLevel[count].VclkDivider = (u8)dividers.post_divider;
+
+ ret = radeon_atom_get_clock_dividers(rdev,
+ COMPUTE_GPUCLK_INPUT_FLAG_DEFAULT_GPUCLK,
+ table->UvdLevel[count].DclkFrequency, false, ÷rs);
+ if (ret)
+ return ret;
+
+ table->UvdLevel[count].DclkDivider = (u8)dividers.post_divider;
+
+ table->UvdLevel[count].VclkFrequency = cpu_to_be32(table->UvdLevel[count].VclkFrequency);
+ table->UvdLevel[count].DclkFrequency = cpu_to_be32(table->UvdLevel[count].DclkFrequency);
+ table->UvdLevel[count].MinVddc = cpu_to_be16(table->UvdLevel[count].MinVddc);
+ }
+
+ return ret;
+}
+
+static int ci_populate_smc_vce_level(struct radeon_device *rdev,
+ SMU7_Discrete_DpmTable *table)
+{
+ u32 count;
+ struct atom_clock_dividers dividers;
+ int ret = -EINVAL;
+
+ table->VceLevelCount =
+ rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.count;
+
+ for (count = 0; count < table->VceLevelCount; count++) {
+ table->VceLevel[count].Frequency =
+ rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[count].evclk;
+ table->VceLevel[count].MinVoltage =
+ (u16)rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[count].v * VOLTAGE_SCALE;
+ table->VceLevel[count].MinPhases = 1;
+
+ ret = radeon_atom_get_clock_dividers(rdev,
+ COMPUTE_GPUCLK_INPUT_FLAG_DEFAULT_GPUCLK,
+ table->VceLevel[count].Frequency, false, ÷rs);
+ if (ret)
+ return ret;
+
+ table->VceLevel[count].Divider = (u8)dividers.post_divider;
+
+ table->VceLevel[count].Frequency = cpu_to_be32(table->VceLevel[count].Frequency);
+ table->VceLevel[count].MinVoltage = cpu_to_be16(table->VceLevel[count].MinVoltage);
+ }
+
+ return ret;
+
+}
+
+static int ci_populate_smc_acp_level(struct radeon_device *rdev,
+ SMU7_Discrete_DpmTable *table)
+{
+ u32 count;
+ struct atom_clock_dividers dividers;
+ int ret = -EINVAL;
+
+ table->AcpLevelCount = (u8)
+ (rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.count);
+
+ for (count = 0; count < table->AcpLevelCount; count++) {
+ table->AcpLevel[count].Frequency =
+ rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries[count].clk;
+ table->AcpLevel[count].MinVoltage =
+ rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries[count].v;
+ table->AcpLevel[count].MinPhases = 1;
+
+ ret = radeon_atom_get_clock_dividers(rdev,
+ COMPUTE_GPUCLK_INPUT_FLAG_DEFAULT_GPUCLK,
+ table->AcpLevel[count].Frequency, false, ÷rs);
+ if (ret)
+ return ret;
+
+ table->AcpLevel[count].Divider = (u8)dividers.post_divider;
+
+ table->AcpLevel[count].Frequency = cpu_to_be32(table->AcpLevel[count].Frequency);
+ table->AcpLevel[count].MinVoltage = cpu_to_be16(table->AcpLevel[count].MinVoltage);
+ }
+
+ return ret;
+}
+
+static int ci_populate_smc_samu_level(struct radeon_device *rdev,
+ SMU7_Discrete_DpmTable *table)
+{
+ u32 count;
+ struct atom_clock_dividers dividers;
+ int ret = -EINVAL;
+
+ table->SamuLevelCount =
+ rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.count;
+
+ for (count = 0; count < table->SamuLevelCount; count++) {
+ table->SamuLevel[count].Frequency =
+ rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries[count].clk;
+ table->SamuLevel[count].MinVoltage =
+ rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries[count].v * VOLTAGE_SCALE;
+ table->SamuLevel[count].MinPhases = 1;
+
+ ret = radeon_atom_get_clock_dividers(rdev,
+ COMPUTE_GPUCLK_INPUT_FLAG_DEFAULT_GPUCLK,
+ table->SamuLevel[count].Frequency, false, ÷rs);
+ if (ret)
+ return ret;
+
+ table->SamuLevel[count].Divider = (u8)dividers.post_divider;
+
+ table->SamuLevel[count].Frequency = cpu_to_be32(table->SamuLevel[count].Frequency);
+ table->SamuLevel[count].MinVoltage = cpu_to_be16(table->SamuLevel[count].MinVoltage);
+ }
+
+ return ret;
+}
+
+static int ci_calculate_mclk_params(struct radeon_device *rdev,
+ u32 memory_clock,
+ SMU7_Discrete_MemoryLevel *mclk,
+ bool strobe_mode,
+ bool dll_state_on)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 dll_cntl = pi->clock_registers.dll_cntl;
+ u32 mclk_pwrmgt_cntl = pi->clock_registers.mclk_pwrmgt_cntl;
+ u32 mpll_ad_func_cntl = pi->clock_registers.mpll_ad_func_cntl;
+ u32 mpll_dq_func_cntl = pi->clock_registers.mpll_dq_func_cntl;
+ u32 mpll_func_cntl = pi->clock_registers.mpll_func_cntl;
+ u32 mpll_func_cntl_1 = pi->clock_registers.mpll_func_cntl_1;
+ u32 mpll_func_cntl_2 = pi->clock_registers.mpll_func_cntl_2;
+ u32 mpll_ss1 = pi->clock_registers.mpll_ss1;
+ u32 mpll_ss2 = pi->clock_registers.mpll_ss2;
+ struct atom_mpll_param mpll_param;
+ int ret;
+
+ ret = radeon_atom_get_memory_pll_dividers(rdev, memory_clock, strobe_mode, &mpll_param);
+ if (ret)
+ return ret;
+
+ mpll_func_cntl &= ~BWCTRL_MASK;
+ mpll_func_cntl |= BWCTRL(mpll_param.bwcntl);
+
+ mpll_func_cntl_1 &= ~(CLKF_MASK | CLKFRAC_MASK | VCO_MODE_MASK);
+ mpll_func_cntl_1 |= CLKF(mpll_param.clkf) |
+ CLKFRAC(mpll_param.clkfrac) | VCO_MODE(mpll_param.vco_mode);
+
+ mpll_ad_func_cntl &= ~YCLK_POST_DIV_MASK;
+ mpll_ad_func_cntl |= YCLK_POST_DIV(mpll_param.post_div);
+
+ if (pi->mem_gddr5) {
+ mpll_dq_func_cntl &= ~(YCLK_SEL_MASK | YCLK_POST_DIV_MASK);
+ mpll_dq_func_cntl |= YCLK_SEL(mpll_param.yclk_sel) |
+ YCLK_POST_DIV(mpll_param.post_div);
+ }
+
+ if (pi->caps_mclk_ss_support) {
+ struct radeon_atom_ss ss;
+ u32 freq_nom;
+ u32 tmp;
+ u32 reference_clock = rdev->clock.mpll.reference_freq;
+
+ if (pi->mem_gddr5)
+ freq_nom = memory_clock * 4;
+ else
+ freq_nom = memory_clock * 2;
+
+ tmp = (freq_nom / reference_clock);
+ tmp = tmp * tmp;
+ if (radeon_atombios_get_asic_ss_info(rdev, &ss,
+ ASIC_INTERNAL_MEMORY_SS, freq_nom)) {
+ u32 clks = reference_clock * 5 / ss.rate;
+ u32 clkv = (u32)((((131 * ss.percentage * ss.rate) / 100) * tmp) / freq_nom);
+
+ mpll_ss1 &= ~CLKV_MASK;
+ mpll_ss1 |= CLKV(clkv);
+
+ mpll_ss2 &= ~CLKS_MASK;
+ mpll_ss2 |= CLKS(clks);
+ }
+ }
+
+ mclk_pwrmgt_cntl &= ~DLL_SPEED_MASK;
+ mclk_pwrmgt_cntl |= DLL_SPEED(mpll_param.dll_speed);
+
+ if (dll_state_on)
+ mclk_pwrmgt_cntl |= MRDCK0_PDNB | MRDCK1_PDNB;
+ else
+ mclk_pwrmgt_cntl &= ~(MRDCK0_PDNB | MRDCK1_PDNB);
+
+ mclk->MclkFrequency = memory_clock;
+ mclk->MpllFuncCntl = mpll_func_cntl;
+ mclk->MpllFuncCntl_1 = mpll_func_cntl_1;
+ mclk->MpllFuncCntl_2 = mpll_func_cntl_2;
+ mclk->MpllAdFuncCntl = mpll_ad_func_cntl;
+ mclk->MpllDqFuncCntl = mpll_dq_func_cntl;
+ mclk->MclkPwrmgtCntl = mclk_pwrmgt_cntl;
+ mclk->DllCntl = dll_cntl;
+ mclk->MpllSs1 = mpll_ss1;
+ mclk->MpllSs2 = mpll_ss2;
+
+ return 0;
+}
+
+static int ci_populate_single_memory_level(struct radeon_device *rdev,
+ u32 memory_clock,
+ SMU7_Discrete_MemoryLevel *memory_level)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ int ret;
+ bool dll_state_on;
+
+ if (rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk.entries) {
+ ret = ci_get_dependency_volt_by_clk(rdev,
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk,
+ memory_clock, &memory_level->MinVddc);
+ if (ret)
+ return ret;
+ }
+
+ if (rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk.entries) {
+ ret = ci_get_dependency_volt_by_clk(rdev,
+ &rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk,
+ memory_clock, &memory_level->MinVddci);
+ if (ret)
+ return ret;
+ }
+
+ if (rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk.entries) {
+ ret = ci_get_dependency_volt_by_clk(rdev,
+ &rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk,
+ memory_clock, &memory_level->MinMvdd);
+ if (ret)
+ return ret;
+ }
+
+ memory_level->MinVddcPhases = 1;
+
+ if (pi->vddc_phase_shed_control)
+ ci_populate_phase_value_based_on_mclk(rdev,
+ &rdev->pm.dpm.dyn_state.phase_shedding_limits_table,
+ memory_clock,
+ &memory_level->MinVddcPhases);
+
+ memory_level->EnabledForThrottle = 1;
+ memory_level->EnabledForActivity = 1;
+ memory_level->UpH = 0;
+ memory_level->DownH = 100;
+ memory_level->VoltageDownH = 0;
+ memory_level->ActivityLevel = (u16)pi->mclk_activity_target;
+
+ memory_level->StutterEnable = false;
+ memory_level->StrobeEnable = false;
+ memory_level->EdcReadEnable = false;
+ memory_level->EdcWriteEnable = false;
+ memory_level->RttEnable = false;
+
+ memory_level->DisplayWatermark = PPSMC_DISPLAY_WATERMARK_LOW;
+
+ if (pi->mclk_stutter_mode_threshold &&
+ (memory_clock <= pi->mclk_stutter_mode_threshold) &&
+ (pi->uvd_enabled == false) &&
+ (RREG32(DPG_PIPE_STUTTER_CONTROL) & STUTTER_ENABLE) &&
+ (rdev->pm.dpm.new_active_crtc_count <= 2))
+ memory_level->StutterEnable = true;
+
+ if (pi->mclk_strobe_mode_threshold &&
+ (memory_clock <= pi->mclk_strobe_mode_threshold))
+ memory_level->StrobeEnable = 1;
+
+ if (pi->mem_gddr5) {
+ memory_level->StrobeRatio =
+ si_get_mclk_frequency_ratio(memory_clock, memory_level->StrobeEnable);
+ if (pi->mclk_edc_enable_threshold &&
+ (memory_clock > pi->mclk_edc_enable_threshold))
+ memory_level->EdcReadEnable = true;
+
+ if (pi->mclk_edc_wr_enable_threshold &&
+ (memory_clock > pi->mclk_edc_wr_enable_threshold))
+ memory_level->EdcWriteEnable = true;
+
+ if (memory_level->StrobeEnable) {
+ if (si_get_mclk_frequency_ratio(memory_clock, true) >=
+ ((RREG32(MC_SEQ_MISC7) >> 16) & 0xf))
+ dll_state_on = ((RREG32(MC_SEQ_MISC5) >> 1) & 0x1) ? true : false;
+ else
+ dll_state_on = ((RREG32(MC_SEQ_MISC6) >> 1) & 0x1) ? true : false;
+ } else {
+ dll_state_on = pi->dll_default_on;
+ }
+ } else {
+ memory_level->StrobeRatio = si_get_ddr3_mclk_frequency_ratio(memory_clock);
+ dll_state_on = ((RREG32(MC_SEQ_MISC5) >> 1) & 0x1) ? true : false;
+ }
+
+ ret = ci_calculate_mclk_params(rdev, memory_clock, memory_level, memory_level->StrobeEnable, dll_state_on);
+ if (ret)
+ return ret;
+
+ memory_level->MinVddc = cpu_to_be32(memory_level->MinVddc * VOLTAGE_SCALE);
+ memory_level->MinVddcPhases = cpu_to_be32(memory_level->MinVddcPhases);
+ memory_level->MinVddci = cpu_to_be32(memory_level->MinVddci * VOLTAGE_SCALE);
+ memory_level->MinMvdd = cpu_to_be32(memory_level->MinMvdd * VOLTAGE_SCALE);
+
+ memory_level->MclkFrequency = cpu_to_be32(memory_level->MclkFrequency);
+ memory_level->ActivityLevel = cpu_to_be16(memory_level->ActivityLevel);
+ memory_level->MpllFuncCntl = cpu_to_be32(memory_level->MpllFuncCntl);
+ memory_level->MpllFuncCntl_1 = cpu_to_be32(memory_level->MpllFuncCntl_1);
+ memory_level->MpllFuncCntl_2 = cpu_to_be32(memory_level->MpllFuncCntl_2);
+ memory_level->MpllAdFuncCntl = cpu_to_be32(memory_level->MpllAdFuncCntl);
+ memory_level->MpllDqFuncCntl = cpu_to_be32(memory_level->MpllDqFuncCntl);
+ memory_level->MclkPwrmgtCntl = cpu_to_be32(memory_level->MclkPwrmgtCntl);
+ memory_level->DllCntl = cpu_to_be32(memory_level->DllCntl);
+ memory_level->MpllSs1 = cpu_to_be32(memory_level->MpllSs1);
+ memory_level->MpllSs2 = cpu_to_be32(memory_level->MpllSs2);
+
+ return 0;
+}
+
+static int ci_populate_smc_acpi_level(struct radeon_device *rdev,
+ SMU7_Discrete_DpmTable *table)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct atom_clock_dividers dividers;
+ SMU7_Discrete_VoltageLevel voltage_level;
+ u32 spll_func_cntl = pi->clock_registers.cg_spll_func_cntl;
+ u32 spll_func_cntl_2 = pi->clock_registers.cg_spll_func_cntl_2;
+ u32 dll_cntl = pi->clock_registers.dll_cntl;
+ u32 mclk_pwrmgt_cntl = pi->clock_registers.mclk_pwrmgt_cntl;
+ int ret;
+
+ table->ACPILevel.Flags &= ~PPSMC_SWSTATE_FLAG_DC;
+
+ if (pi->acpi_vddc)
+ table->ACPILevel.MinVddc = cpu_to_be32(pi->acpi_vddc * VOLTAGE_SCALE);
+ else
+ table->ACPILevel.MinVddc = cpu_to_be32(pi->min_vddc_in_pp_table * VOLTAGE_SCALE);
+
+ table->ACPILevel.MinVddcPhases = pi->vddc_phase_shed_control ? 0 : 1;
+
+ table->ACPILevel.SclkFrequency = rdev->clock.spll.reference_freq;
+
+ ret = radeon_atom_get_clock_dividers(rdev,
+ COMPUTE_GPUCLK_INPUT_FLAG_SCLK,
+ table->ACPILevel.SclkFrequency, false, ÷rs);
+ if (ret)
+ return ret;
+
+ table->ACPILevel.SclkDid = (u8)dividers.post_divider;
+ table->ACPILevel.DisplayWatermark = PPSMC_DISPLAY_WATERMARK_LOW;
+ table->ACPILevel.DeepSleepDivId = 0;
+
+ spll_func_cntl &= ~SPLL_PWRON;
+ spll_func_cntl |= SPLL_RESET;
+
+ spll_func_cntl_2 &= ~SCLK_MUX_SEL_MASK;
+ spll_func_cntl_2 |= SCLK_MUX_SEL(4);
+
+ table->ACPILevel.CgSpllFuncCntl = spll_func_cntl;
+ table->ACPILevel.CgSpllFuncCntl2 = spll_func_cntl_2;
+ table->ACPILevel.CgSpllFuncCntl3 = pi->clock_registers.cg_spll_func_cntl_3;
+ table->ACPILevel.CgSpllFuncCntl4 = pi->clock_registers.cg_spll_func_cntl_4;
+ table->ACPILevel.SpllSpreadSpectrum = pi->clock_registers.cg_spll_spread_spectrum;
+ table->ACPILevel.SpllSpreadSpectrum2 = pi->clock_registers.cg_spll_spread_spectrum_2;
+ table->ACPILevel.CcPwrDynRm = 0;
+ table->ACPILevel.CcPwrDynRm1 = 0;
+
+ table->ACPILevel.Flags = cpu_to_be32(table->ACPILevel.Flags);
+ table->ACPILevel.MinVddcPhases = cpu_to_be32(table->ACPILevel.MinVddcPhases);
+ table->ACPILevel.SclkFrequency = cpu_to_be32(table->ACPILevel.SclkFrequency);
+ table->ACPILevel.CgSpllFuncCntl = cpu_to_be32(table->ACPILevel.CgSpllFuncCntl);
+ table->ACPILevel.CgSpllFuncCntl2 = cpu_to_be32(table->ACPILevel.CgSpllFuncCntl2);
+ table->ACPILevel.CgSpllFuncCntl3 = cpu_to_be32(table->ACPILevel.CgSpllFuncCntl3);
+ table->ACPILevel.CgSpllFuncCntl4 = cpu_to_be32(table->ACPILevel.CgSpllFuncCntl4);
+ table->ACPILevel.SpllSpreadSpectrum = cpu_to_be32(table->ACPILevel.SpllSpreadSpectrum);
+ table->ACPILevel.SpllSpreadSpectrum2 = cpu_to_be32(table->ACPILevel.SpllSpreadSpectrum2);
+ table->ACPILevel.CcPwrDynRm = cpu_to_be32(table->ACPILevel.CcPwrDynRm);
+ table->ACPILevel.CcPwrDynRm1 = cpu_to_be32(table->ACPILevel.CcPwrDynRm1);
+
+ table->MemoryACPILevel.MinVddc = table->ACPILevel.MinVddc;
+ table->MemoryACPILevel.MinVddcPhases = table->ACPILevel.MinVddcPhases;
+
+ if (pi->vddci_control != CISLANDS_VOLTAGE_CONTROL_NONE) {
+ if (pi->acpi_vddci)
+ table->MemoryACPILevel.MinVddci =
+ cpu_to_be32(pi->acpi_vddci * VOLTAGE_SCALE);
+ else
+ table->MemoryACPILevel.MinVddci =
+ cpu_to_be32(pi->min_vddci_in_pp_table * VOLTAGE_SCALE);
+ }
+
+ if (ci_populate_mvdd_value(rdev, 0, &voltage_level))
+ table->MemoryACPILevel.MinMvdd = 0;
+ else
+ table->MemoryACPILevel.MinMvdd =
+ cpu_to_be32(voltage_level.Voltage * VOLTAGE_SCALE);
+
+ mclk_pwrmgt_cntl |= MRDCK0_RESET | MRDCK1_RESET;
+ mclk_pwrmgt_cntl &= ~(MRDCK0_PDNB | MRDCK1_PDNB);
+
+ dll_cntl &= ~(MRDCK0_BYPASS | MRDCK1_BYPASS);
+
+ table->MemoryACPILevel.DllCntl = cpu_to_be32(dll_cntl);
+ table->MemoryACPILevel.MclkPwrmgtCntl = cpu_to_be32(mclk_pwrmgt_cntl);
+ table->MemoryACPILevel.MpllAdFuncCntl =
+ cpu_to_be32(pi->clock_registers.mpll_ad_func_cntl);
+ table->MemoryACPILevel.MpllDqFuncCntl =
+ cpu_to_be32(pi->clock_registers.mpll_dq_func_cntl);
+ table->MemoryACPILevel.MpllFuncCntl =
+ cpu_to_be32(pi->clock_registers.mpll_func_cntl);
+ table->MemoryACPILevel.MpllFuncCntl_1 =
+ cpu_to_be32(pi->clock_registers.mpll_func_cntl_1);
+ table->MemoryACPILevel.MpllFuncCntl_2 =
+ cpu_to_be32(pi->clock_registers.mpll_func_cntl_2);
+ table->MemoryACPILevel.MpllSs1 = cpu_to_be32(pi->clock_registers.mpll_ss1);
+ table->MemoryACPILevel.MpllSs2 = cpu_to_be32(pi->clock_registers.mpll_ss2);
+
+ table->MemoryACPILevel.EnabledForThrottle = 0;
+ table->MemoryACPILevel.EnabledForActivity = 0;
+ table->MemoryACPILevel.UpH = 0;
+ table->MemoryACPILevel.DownH = 100;
+ table->MemoryACPILevel.VoltageDownH = 0;
+ table->MemoryACPILevel.ActivityLevel =
+ cpu_to_be16((u16)pi->mclk_activity_target);
+
+ table->MemoryACPILevel.StutterEnable = false;
+ table->MemoryACPILevel.StrobeEnable = false;
+ table->MemoryACPILevel.EdcReadEnable = false;
+ table->MemoryACPILevel.EdcWriteEnable = false;
+ table->MemoryACPILevel.RttEnable = false;
+
+ return 0;
+}
+
+
+static int ci_enable_ulv(struct radeon_device *rdev, bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_ulv_parm *ulv = &pi->ulv;
+
+ if (ulv->supported) {
+ if (enable)
+ return (ci_send_msg_to_smc(rdev, PPSMC_MSG_EnableULV) == PPSMC_Result_OK) ?
+ 0 : -EINVAL;
+ else
+ return (ci_send_msg_to_smc(rdev, PPSMC_MSG_DisableULV) == PPSMC_Result_OK) ?
+ 0 : -EINVAL;
+ }
+
+ return 0;
+}
+
+static int ci_populate_ulv_level(struct radeon_device *rdev,
+ SMU7_Discrete_Ulv *state)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u16 ulv_voltage = rdev->pm.dpm.backbias_response_time;
+
+ state->CcPwrDynRm = 0;
+ state->CcPwrDynRm1 = 0;
+
+ if (ulv_voltage == 0) {
+ pi->ulv.supported = false;
+ return 0;
+ }
+
+ if (pi->voltage_control != CISLANDS_VOLTAGE_CONTROL_BY_SVID2) {
+ if (ulv_voltage > rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[0].v)
+ state->VddcOffset = 0;
+ else
+ state->VddcOffset =
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[0].v - ulv_voltage;
+ } else {
+ if (ulv_voltage > rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[0].v)
+ state->VddcOffsetVid = 0;
+ else
+ state->VddcOffsetVid = (u8)
+ ((rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[0].v - ulv_voltage) *
+ VOLTAGE_VID_OFFSET_SCALE2 / VOLTAGE_VID_OFFSET_SCALE1);
+ }
+ state->VddcPhase = pi->vddc_phase_shed_control ? 0 : 1;
+
+ state->CcPwrDynRm = cpu_to_be32(state->CcPwrDynRm);
+ state->CcPwrDynRm1 = cpu_to_be32(state->CcPwrDynRm1);
+ state->VddcOffset = cpu_to_be16(state->VddcOffset);
+
+ return 0;
+}
+
+static int ci_calculate_sclk_params(struct radeon_device *rdev,
+ u32 engine_clock,
+ SMU7_Discrete_GraphicsLevel *sclk)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct atom_clock_dividers dividers;
+ u32 spll_func_cntl_3 = pi->clock_registers.cg_spll_func_cntl_3;
+ u32 spll_func_cntl_4 = pi->clock_registers.cg_spll_func_cntl_4;
+ u32 cg_spll_spread_spectrum = pi->clock_registers.cg_spll_spread_spectrum;
+ u32 cg_spll_spread_spectrum_2 = pi->clock_registers.cg_spll_spread_spectrum_2;
+ u32 reference_clock = rdev->clock.spll.reference_freq;
+ u32 reference_divider;
+ u32 fbdiv;
+ int ret;
+
+ ret = radeon_atom_get_clock_dividers(rdev,
+ COMPUTE_GPUCLK_INPUT_FLAG_SCLK,
+ engine_clock, false, ÷rs);
+ if (ret)
+ return ret;
+
+ reference_divider = 1 + dividers.ref_div;
+ fbdiv = dividers.fb_div & 0x3FFFFFF;
+
+ spll_func_cntl_3 &= ~SPLL_FB_DIV_MASK;
+ spll_func_cntl_3 |= SPLL_FB_DIV(fbdiv);
+ spll_func_cntl_3 |= SPLL_DITHEN;
+
+ if (pi->caps_sclk_ss_support) {
+ struct radeon_atom_ss ss;
+ u32 vco_freq = engine_clock * dividers.post_div;
+
+ if (radeon_atombios_get_asic_ss_info(rdev, &ss,
+ ASIC_INTERNAL_ENGINE_SS, vco_freq)) {
+ u32 clk_s = reference_clock * 5 / (reference_divider * ss.rate);
+ u32 clk_v = 4 * ss.percentage * fbdiv / (clk_s * 10000);
+
+ cg_spll_spread_spectrum &= ~CLK_S_MASK;
+ cg_spll_spread_spectrum |= CLK_S(clk_s);
+ cg_spll_spread_spectrum |= SSEN;
+
+ cg_spll_spread_spectrum_2 &= ~CLK_V_MASK;
+ cg_spll_spread_spectrum_2 |= CLK_V(clk_v);
+ }
+ }
+
+ sclk->SclkFrequency = engine_clock;
+ sclk->CgSpllFuncCntl3 = spll_func_cntl_3;
+ sclk->CgSpllFuncCntl4 = spll_func_cntl_4;
+ sclk->SpllSpreadSpectrum = cg_spll_spread_spectrum;
+ sclk->SpllSpreadSpectrum2 = cg_spll_spread_spectrum_2;
+ sclk->SclkDid = (u8)dividers.post_divider;
+
+ return 0;
+}
+
+static int ci_populate_single_graphic_level(struct radeon_device *rdev,
+ u32 engine_clock,
+ u16 sclk_activity_level_t,
+ SMU7_Discrete_GraphicsLevel *graphic_level)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ int ret;
+
+ ret = ci_calculate_sclk_params(rdev, engine_clock, graphic_level);
+ if (ret)
+ return ret;
+
+ ret = ci_get_dependency_volt_by_clk(rdev,
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk,
+ engine_clock, &graphic_level->MinVddc);
+ if (ret)
+ return ret;
+
+ graphic_level->SclkFrequency = engine_clock;
+
+ graphic_level->Flags = 0;
+ graphic_level->MinVddcPhases = 1;
+
+ if (pi->vddc_phase_shed_control)
+ ci_populate_phase_value_based_on_sclk(rdev,
+ &rdev->pm.dpm.dyn_state.phase_shedding_limits_table,
+ engine_clock,
+ &graphic_level->MinVddcPhases);
+
+ graphic_level->ActivityLevel = sclk_activity_level_t;
+
+ graphic_level->CcPwrDynRm = 0;
+ graphic_level->CcPwrDynRm1 = 0;
+ graphic_level->EnabledForActivity = 1;
+ graphic_level->EnabledForThrottle = 1;
+ graphic_level->UpH = 0;
+ graphic_level->DownH = 0;
+ graphic_level->VoltageDownH = 0;
+ graphic_level->PowerThrottle = 0;
+
+ if (pi->caps_sclk_ds)
+ graphic_level->DeepSleepDivId = ci_get_sleep_divider_id_from_clock(rdev,
+ engine_clock,
+ CISLAND_MINIMUM_ENGINE_CLOCK);
+
+ graphic_level->DisplayWatermark = PPSMC_DISPLAY_WATERMARK_LOW;
+
+ graphic_level->Flags = cpu_to_be32(graphic_level->Flags);
+ graphic_level->MinVddc = cpu_to_be32(graphic_level->MinVddc * VOLTAGE_SCALE);
+ graphic_level->MinVddcPhases = cpu_to_be32(graphic_level->MinVddcPhases);
+ graphic_level->SclkFrequency = cpu_to_be32(graphic_level->SclkFrequency);
+ graphic_level->ActivityLevel = cpu_to_be16(graphic_level->ActivityLevel);
+ graphic_level->CgSpllFuncCntl3 = cpu_to_be32(graphic_level->CgSpllFuncCntl3);
+ graphic_level->CgSpllFuncCntl4 = cpu_to_be32(graphic_level->CgSpllFuncCntl4);
+ graphic_level->SpllSpreadSpectrum = cpu_to_be32(graphic_level->SpllSpreadSpectrum);
+ graphic_level->SpllSpreadSpectrum2 = cpu_to_be32(graphic_level->SpllSpreadSpectrum2);
+ graphic_level->CcPwrDynRm = cpu_to_be32(graphic_level->CcPwrDynRm);
+ graphic_level->CcPwrDynRm1 = cpu_to_be32(graphic_level->CcPwrDynRm1);
+
+ return 0;
+}
+
+static int ci_populate_all_graphic_levels(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_dpm_table *dpm_table = &pi->dpm_table;
+ u32 level_array_address = pi->dpm_table_start +
+ offsetof(SMU7_Discrete_DpmTable, GraphicsLevel);
+ u32 level_array_size = sizeof(SMU7_Discrete_GraphicsLevel) *
+ SMU7_MAX_LEVELS_GRAPHICS;
+ SMU7_Discrete_GraphicsLevel *levels = pi->smc_state_table.GraphicsLevel;
+ u32 i, ret;
+
+ memset(levels, 0, level_array_size);
+
+ for (i = 0; i < dpm_table->sclk_table.count; i++) {
+ ret = ci_populate_single_graphic_level(rdev,
+ dpm_table->sclk_table.dpm_levels[i].value,
+ (u16)pi->activity_target[i],
+ &pi->smc_state_table.GraphicsLevel[i]);
+ if (ret)
+ return ret;
+ if (i == (dpm_table->sclk_table.count - 1))
+ pi->smc_state_table.GraphicsLevel[i].DisplayWatermark =
+ PPSMC_DISPLAY_WATERMARK_HIGH;
+ }
+
+ pi->smc_state_table.GraphicsDpmLevelCount = (u8)dpm_table->sclk_table.count;
+ pi->dpm_level_enable_mask.sclk_dpm_enable_mask =
+ ci_get_dpm_level_enable_mask_value(&dpm_table->sclk_table);
+
+ ret = ci_copy_bytes_to_smc(rdev, level_array_address,
+ (u8 *)levels, level_array_size,
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ return 0;
+}
+
+static int ci_populate_ulv_state(struct radeon_device *rdev,
+ SMU7_Discrete_Ulv *ulv_level)
+{
+ return ci_populate_ulv_level(rdev, ulv_level);
+}
+
+static int ci_populate_all_memory_levels(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_dpm_table *dpm_table = &pi->dpm_table;
+ u32 level_array_address = pi->dpm_table_start +
+ offsetof(SMU7_Discrete_DpmTable, MemoryLevel);
+ u32 level_array_size = sizeof(SMU7_Discrete_MemoryLevel) *
+ SMU7_MAX_LEVELS_MEMORY;
+ SMU7_Discrete_MemoryLevel *levels = pi->smc_state_table.MemoryLevel;
+ u32 i, ret;
+
+ memset(levels, 0, level_array_size);
+
+ for (i = 0; i < dpm_table->mclk_table.count; i++) {
+ if (dpm_table->mclk_table.dpm_levels[i].value == 0)
+ return -EINVAL;
+ ret = ci_populate_single_memory_level(rdev,
+ dpm_table->mclk_table.dpm_levels[i].value,
+ &pi->smc_state_table.MemoryLevel[i]);
+ if (ret)
+ return ret;
+ }
+
+ pi->smc_state_table.MemoryLevel[0].ActivityLevel = cpu_to_be16(0x1F);
+
+ pi->smc_state_table.MemoryDpmLevelCount = (u8)dpm_table->mclk_table.count;
+ pi->dpm_level_enable_mask.mclk_dpm_enable_mask =
+ ci_get_dpm_level_enable_mask_value(&dpm_table->mclk_table);
+
+ pi->smc_state_table.MemoryLevel[dpm_table->mclk_table.count - 1].DisplayWatermark =
+ PPSMC_DISPLAY_WATERMARK_HIGH;
+
+ ret = ci_copy_bytes_to_smc(rdev, level_array_address,
+ (u8 *)levels, level_array_size,
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ return 0;
+}
+
+static void ci_reset_single_dpm_table(struct radeon_device *rdev,
+ struct ci_single_dpm_table* dpm_table,
+ u32 count)
+{
+ u32 i;
+
+ dpm_table->count = count;
+ for (i = 0; i < MAX_REGULAR_DPM_NUMBER; i++)
+ dpm_table->dpm_levels[i].enabled = false;
+}
+
+static void ci_setup_pcie_table_entry(struct ci_single_dpm_table* dpm_table,
+ u32 index, u32 pcie_gen, u32 pcie_lanes)
+{
+ dpm_table->dpm_levels[index].value = pcie_gen;
+ dpm_table->dpm_levels[index].param1 = pcie_lanes;
+ dpm_table->dpm_levels[index].enabled = true;
+}
+
+static int ci_setup_default_pcie_tables(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ if (!pi->use_pcie_performance_levels && !pi->use_pcie_powersaving_levels)
+ return -EINVAL;
+
+ if (pi->use_pcie_performance_levels && !pi->use_pcie_powersaving_levels) {
+ pi->pcie_gen_powersaving = pi->pcie_gen_performance;
+ pi->pcie_lane_powersaving = pi->pcie_lane_performance;
+ } else if (!pi->use_pcie_performance_levels && pi->use_pcie_powersaving_levels) {
+ pi->pcie_gen_performance = pi->pcie_gen_powersaving;
+ pi->pcie_lane_performance = pi->pcie_lane_powersaving;
+ }
+
+ ci_reset_single_dpm_table(rdev,
+ &pi->dpm_table.pcie_speed_table,
+ SMU7_MAX_LEVELS_LINK);
+
+ ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 0,
+ pi->pcie_gen_powersaving.min,
+ pi->pcie_lane_powersaving.min);
+ ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 1,
+ pi->pcie_gen_performance.min,
+ pi->pcie_lane_performance.min);
+ ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 2,
+ pi->pcie_gen_powersaving.min,
+ pi->pcie_lane_powersaving.max);
+ ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 3,
+ pi->pcie_gen_performance.min,
+ pi->pcie_lane_performance.max);
+ ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 4,
+ pi->pcie_gen_powersaving.max,
+ pi->pcie_lane_powersaving.max);
+ ci_setup_pcie_table_entry(&pi->dpm_table.pcie_speed_table, 5,
+ pi->pcie_gen_performance.max,
+ pi->pcie_lane_performance.max);
+
+ pi->dpm_table.pcie_speed_table.count = 6;
+
+ return 0;
+}
+
+static int ci_setup_default_dpm_tables(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct radeon_clock_voltage_dependency_table *allowed_sclk_vddc_table =
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
+ struct radeon_clock_voltage_dependency_table *allowed_mclk_table =
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk;
+ struct radeon_cac_leakage_table *std_voltage_table =
+ &rdev->pm.dpm.dyn_state.cac_leakage_table;
+ u32 i;
+
+ if (allowed_sclk_vddc_table == NULL)
+ return -EINVAL;
+ if (allowed_sclk_vddc_table->count < 1)
+ return -EINVAL;
+ if (allowed_mclk_table == NULL)
+ return -EINVAL;
+ if (allowed_mclk_table->count < 1)
+ return -EINVAL;
+
+ memset(&pi->dpm_table, 0, sizeof(struct ci_dpm_table));
+
+ ci_reset_single_dpm_table(rdev,
+ &pi->dpm_table.sclk_table,
+ SMU7_MAX_LEVELS_GRAPHICS);
+ ci_reset_single_dpm_table(rdev,
+ &pi->dpm_table.mclk_table,
+ SMU7_MAX_LEVELS_MEMORY);
+ ci_reset_single_dpm_table(rdev,
+ &pi->dpm_table.vddc_table,
+ SMU7_MAX_LEVELS_VDDC);
+ ci_reset_single_dpm_table(rdev,
+ &pi->dpm_table.vddci_table,
+ SMU7_MAX_LEVELS_VDDCI);
+ ci_reset_single_dpm_table(rdev,
+ &pi->dpm_table.mvdd_table,
+ SMU7_MAX_LEVELS_MVDD);
+
+ pi->dpm_table.sclk_table.count = 0;
+ for (i = 0; i < allowed_sclk_vddc_table->count; i++) {
+ if ((i == 0) ||
+ (pi->dpm_table.sclk_table.dpm_levels[pi->dpm_table.sclk_table.count-1].value !=
+ allowed_sclk_vddc_table->entries[i].clk)) {
+ pi->dpm_table.sclk_table.dpm_levels[pi->dpm_table.sclk_table.count].value =
+ allowed_sclk_vddc_table->entries[i].clk;
+ pi->dpm_table.sclk_table.dpm_levels[pi->dpm_table.sclk_table.count].enabled = true;
+ pi->dpm_table.sclk_table.count++;
+ }
+ }
+
+ pi->dpm_table.mclk_table.count = 0;
+ for (i = 0; i < allowed_mclk_table->count; i++) {
+ if ((i==0) ||
+ (pi->dpm_table.mclk_table.dpm_levels[pi->dpm_table.mclk_table.count-1].value !=
+ allowed_mclk_table->entries[i].clk)) {
+ pi->dpm_table.mclk_table.dpm_levels[pi->dpm_table.mclk_table.count].value =
+ allowed_mclk_table->entries[i].clk;
+ pi->dpm_table.mclk_table.dpm_levels[pi->dpm_table.mclk_table.count].enabled = true;
+ pi->dpm_table.mclk_table.count++;
+ }
+ }
+
+ for (i = 0; i < allowed_sclk_vddc_table->count; i++) {
+ pi->dpm_table.vddc_table.dpm_levels[i].value =
+ allowed_sclk_vddc_table->entries[i].v;
+ pi->dpm_table.vddc_table.dpm_levels[i].param1 =
+ std_voltage_table->entries[i].leakage;
+ pi->dpm_table.vddc_table.dpm_levels[i].enabled = true;
+ }
+ pi->dpm_table.vddc_table.count = allowed_sclk_vddc_table->count;
+
+ allowed_mclk_table = &rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk;
+ if (allowed_mclk_table) {
+ for (i = 0; i < allowed_mclk_table->count; i++) {
+ pi->dpm_table.vddci_table.dpm_levels[i].value =
+ allowed_mclk_table->entries[i].v;
+ pi->dpm_table.vddci_table.dpm_levels[i].enabled = true;
+ }
+ pi->dpm_table.vddci_table.count = allowed_mclk_table->count;
+ }
+
+ allowed_mclk_table = &rdev->pm.dpm.dyn_state.mvdd_dependency_on_mclk;
+ if (allowed_mclk_table) {
+ for (i = 0; i < allowed_mclk_table->count; i++) {
+ pi->dpm_table.mvdd_table.dpm_levels[i].value =
+ allowed_mclk_table->entries[i].v;
+ pi->dpm_table.mvdd_table.dpm_levels[i].enabled = true;
+ }
+ pi->dpm_table.mvdd_table.count = allowed_mclk_table->count;
+ }
+
+ ci_setup_default_pcie_tables(rdev);
+
+ return 0;
+}
+
+static int ci_find_boot_level(struct ci_single_dpm_table *table,
+ u32 value, u32 *boot_level)
+{
+ u32 i;
+ int ret = -EINVAL;
+
+ for(i = 0; i < table->count; i++) {
+ if (value == table->dpm_levels[i].value) {
+ *boot_level = i;
+ ret = 0;
+ }
+ }
+
+ return ret;
+}
+
+static int ci_init_smc_table(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_ulv_parm *ulv = &pi->ulv;
+ struct radeon_ps *radeon_boot_state = rdev->pm.dpm.boot_ps;
+ SMU7_Discrete_DpmTable *table = &pi->smc_state_table;
+ int ret;
+
+ ret = ci_setup_default_dpm_tables(rdev);
+ if (ret)
+ return ret;
+
+ if (pi->voltage_control != CISLANDS_VOLTAGE_CONTROL_NONE)
+ ci_populate_smc_voltage_tables(rdev, table);
+
+ ci_init_fps_limits(rdev);
+
+ if (rdev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_HARDWAREDC)
+ table->SystemFlags |= PPSMC_SYSTEMFLAG_GPIO_DC;
+
+ if (rdev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_STEPVDDC)
+ table->SystemFlags |= PPSMC_SYSTEMFLAG_STEPVDDC;
+
+ if (pi->mem_gddr5)
+ table->SystemFlags |= PPSMC_SYSTEMFLAG_GDDR5;
+
+ if (ulv->supported) {
+ ret = ci_populate_ulv_state(rdev, &pi->smc_state_table.Ulv);
+ if (ret)
+ return ret;
+ WREG32_SMC(CG_ULV_PARAMETER, ulv->cg_ulv_parameter);
+ }
+
+ ret = ci_populate_all_graphic_levels(rdev);
+ if (ret)
+ return ret;
+
+ ret = ci_populate_all_memory_levels(rdev);
+ if (ret)
+ return ret;
+
+ ci_populate_smc_link_level(rdev, table);
+
+ ret = ci_populate_smc_acpi_level(rdev, table);
+ if (ret)
+ return ret;
+
+ ret = ci_populate_smc_vce_level(rdev, table);
+ if (ret)
+ return ret;
+
+ ret = ci_populate_smc_acp_level(rdev, table);
+ if (ret)
+ return ret;
+
+ ret = ci_populate_smc_samu_level(rdev, table);
+ if (ret)
+ return ret;
+
+ ret = ci_do_program_memory_timing_parameters(rdev);
+ if (ret)
+ return ret;
+
+ ret = ci_populate_smc_uvd_level(rdev, table);
+ if (ret)
+ return ret;
+
+ table->UvdBootLevel = 0;
+ table->VceBootLevel = 0;
+ table->AcpBootLevel = 0;
+ table->SamuBootLevel = 0;
+ table->GraphicsBootLevel = 0;
+ table->MemoryBootLevel = 0;
+
+ ret = ci_find_boot_level(&pi->dpm_table.sclk_table,
+ pi->vbios_boot_state.sclk_bootup_value,
+ (u32 *)&pi->smc_state_table.GraphicsBootLevel);
+
+ ret = ci_find_boot_level(&pi->dpm_table.mclk_table,
+ pi->vbios_boot_state.mclk_bootup_value,
+ (u32 *)&pi->smc_state_table.MemoryBootLevel);
+
+ table->BootVddc = pi->vbios_boot_state.vddc_bootup_value;
+ table->BootVddci = pi->vbios_boot_state.vddci_bootup_value;
+ table->BootMVdd = pi->vbios_boot_state.mvdd_bootup_value;
+
+ ci_populate_smc_initial_state(rdev, radeon_boot_state);
+
+ ret = ci_populate_bapm_parameters_in_dpm_table(rdev);
+ if (ret)
+ return ret;
+
+ table->UVDInterval = 1;
+ table->VCEInterval = 1;
+ table->ACPInterval = 1;
+ table->SAMUInterval = 1;
+ table->GraphicsVoltageChangeEnable = 1;
+ table->GraphicsThermThrottleEnable = 1;
+ table->GraphicsInterval = 1;
+ table->VoltageInterval = 1;
+ table->ThermalInterval = 1;
+ table->TemperatureLimitHigh = (u16)((pi->thermal_temp_setting.temperature_high *
+ CISLANDS_Q88_FORMAT_CONVERSION_UNIT) / 1000);
+ table->TemperatureLimitLow = (u16)((pi->thermal_temp_setting.temperature_low *
+ CISLANDS_Q88_FORMAT_CONVERSION_UNIT) / 1000);
+ table->MemoryVoltageChangeEnable = 1;
+ table->MemoryInterval = 1;
+ table->VoltageResponseTime = 0;
+ table->VddcVddciDelta = 4000;
+ table->PhaseResponseTime = 0;
+ table->MemoryThermThrottleEnable = 1;
+ table->PCIeBootLinkLevel = 0;
+ table->PCIeGenInterval = 1;
+ if (pi->voltage_control == CISLANDS_VOLTAGE_CONTROL_BY_SVID2)
+ table->SVI2Enable = 1;
+ else
+ table->SVI2Enable = 0;
+
+ table->ThermGpio = 17;
+ table->SclkStepSize = 0x4000;
+
+ table->SystemFlags = cpu_to_be32(table->SystemFlags);
+ table->SmioMaskVddcVid = cpu_to_be32(table->SmioMaskVddcVid);
+ table->SmioMaskVddcPhase = cpu_to_be32(table->SmioMaskVddcPhase);
+ table->SmioMaskVddciVid = cpu_to_be32(table->SmioMaskVddciVid);
+ table->SmioMaskMvddVid = cpu_to_be32(table->SmioMaskMvddVid);
+ table->SclkStepSize = cpu_to_be32(table->SclkStepSize);
+ table->TemperatureLimitHigh = cpu_to_be16(table->TemperatureLimitHigh);
+ table->TemperatureLimitLow = cpu_to_be16(table->TemperatureLimitLow);
+ table->VddcVddciDelta = cpu_to_be16(table->VddcVddciDelta);
+ table->VoltageResponseTime = cpu_to_be16(table->VoltageResponseTime);
+ table->PhaseResponseTime = cpu_to_be16(table->PhaseResponseTime);
+ table->BootVddc = cpu_to_be16(table->BootVddc * VOLTAGE_SCALE);
+ table->BootVddci = cpu_to_be16(table->BootVddci * VOLTAGE_SCALE);
+ table->BootMVdd = cpu_to_be16(table->BootMVdd * VOLTAGE_SCALE);
+
+ ret = ci_copy_bytes_to_smc(rdev,
+ pi->dpm_table_start +
+ offsetof(SMU7_Discrete_DpmTable, SystemFlags),
+ (u8 *)&table->SystemFlags,
+ sizeof(SMU7_Discrete_DpmTable) - 3 * sizeof(SMU7_PIDController),
+ pi->sram_end);
+ if (ret)
+ return ret;
+
+ return 0;
+}
+
+static void ci_trim_single_dpm_states(struct radeon_device *rdev,
+ struct ci_single_dpm_table *dpm_table,
+ u32 low_limit, u32 high_limit)
+{
+ u32 i;
+
+ for (i = 0; i < dpm_table->count; i++) {
+ if ((dpm_table->dpm_levels[i].value < low_limit) ||
+ (dpm_table->dpm_levels[i].value > high_limit))
+ dpm_table->dpm_levels[i].enabled = false;
+ else
+ dpm_table->dpm_levels[i].enabled = true;
+ }
+}
+
+static void ci_trim_pcie_dpm_states(struct radeon_device *rdev,
+ u32 speed_low, u32 lanes_low,
+ u32 speed_high, u32 lanes_high)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_single_dpm_table *pcie_table = &pi->dpm_table.pcie_speed_table;
+ u32 i, j;
+
+ for (i = 0; i < pcie_table->count; i++) {
+ if ((pcie_table->dpm_levels[i].value < speed_low) ||
+ (pcie_table->dpm_levels[i].param1 < lanes_low) ||
+ (pcie_table->dpm_levels[i].value > speed_high) ||
+ (pcie_table->dpm_levels[i].param1 > lanes_high))
+ pcie_table->dpm_levels[i].enabled = false;
+ else
+ pcie_table->dpm_levels[i].enabled = true;
+ }
+
+ for (i = 0; i < pcie_table->count; i++) {
+ if (pcie_table->dpm_levels[i].enabled) {
+ for (j = i + 1; j < pcie_table->count; j++) {
+ if (pcie_table->dpm_levels[j].enabled) {
+ if ((pcie_table->dpm_levels[i].value == pcie_table->dpm_levels[j].value) &&
+ (pcie_table->dpm_levels[i].param1 == pcie_table->dpm_levels[j].param1))
+ pcie_table->dpm_levels[j].enabled = false;
+ }
+ }
+ }
+ }
+}
+
+static int ci_trim_dpm_states(struct radeon_device *rdev,
+ struct radeon_ps *radeon_state)
+{
+ struct ci_ps *state = ci_get_ps(radeon_state);
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 high_limit_count;
+
+ if (state->performance_level_count < 1)
+ return -EINVAL;
+
+ if (state->performance_level_count == 1)
+ high_limit_count = 0;
+ else
+ high_limit_count = 1;
+
+ ci_trim_single_dpm_states(rdev,
+ &pi->dpm_table.sclk_table,
+ state->performance_levels[0].sclk,
+ state->performance_levels[high_limit_count].sclk);
+
+ ci_trim_single_dpm_states(rdev,
+ &pi->dpm_table.mclk_table,
+ state->performance_levels[0].mclk,
+ state->performance_levels[high_limit_count].mclk);
+
+ ci_trim_pcie_dpm_states(rdev,
+ state->performance_levels[0].pcie_gen,
+ state->performance_levels[0].pcie_lane,
+ state->performance_levels[high_limit_count].pcie_gen,
+ state->performance_levels[high_limit_count].pcie_lane);
+
+ return 0;
+}
+
+static int ci_apply_disp_minimum_voltage_request(struct radeon_device *rdev)
+{
+ struct radeon_clock_voltage_dependency_table *disp_voltage_table =
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk;
+ struct radeon_clock_voltage_dependency_table *vddc_table =
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
+ u32 requested_voltage = 0;
+ u32 i;
+
+ if (disp_voltage_table == NULL)
+ return -EINVAL;
+ if (!disp_voltage_table->count)
+ return -EINVAL;
+
+ for (i = 0; i < disp_voltage_table->count; i++) {
+ if (rdev->clock.current_dispclk == disp_voltage_table->entries[i].clk)
+ requested_voltage = disp_voltage_table->entries[i].v;
+ }
+
+ for (i = 0; i < vddc_table->count; i++) {
+ if (requested_voltage <= vddc_table->entries[i].v) {
+ requested_voltage = vddc_table->entries[i].v;
+ return (ci_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_VddC_Request,
+ requested_voltage * VOLTAGE_SCALE) == PPSMC_Result_OK) ?
+ 0 : -EINVAL;
+ }
+ }
+
+ return -EINVAL;
+}
+
+static int ci_upload_dpm_level_enable_mask(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ PPSMC_Result result;
+
+ if (!pi->sclk_dpm_key_disabled) {
+ if (pi->dpm_level_enable_mask.sclk_dpm_enable_mask) {
+ result = ci_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_SCLKDPM_SetEnabledMask,
+ pi->dpm_level_enable_mask.sclk_dpm_enable_mask);
+ if (result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+ }
+
+ if (!pi->mclk_dpm_key_disabled) {
+ if (pi->dpm_level_enable_mask.mclk_dpm_enable_mask) {
+ result = ci_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_MCLKDPM_SetEnabledMask,
+ pi->dpm_level_enable_mask.mclk_dpm_enable_mask);
+ if (result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+ }
+
+ if (!pi->pcie_dpm_key_disabled) {
+ if (pi->dpm_level_enable_mask.pcie_dpm_enable_mask) {
+ result = ci_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_PCIeDPM_SetEnabledMask,
+ pi->dpm_level_enable_mask.pcie_dpm_enable_mask);
+ if (result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+ }
+
+ ci_apply_disp_minimum_voltage_request(rdev);
+
+ return 0;
+}
+
+static void ci_find_dpm_states_clocks_in_dpm_table(struct radeon_device *rdev,
+ struct radeon_ps *radeon_state)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_ps *state = ci_get_ps(radeon_state);
+ struct ci_single_dpm_table *sclk_table = &pi->dpm_table.sclk_table;
+ u32 sclk = state->performance_levels[state->performance_level_count-1].sclk;
+ struct ci_single_dpm_table *mclk_table = &pi->dpm_table.mclk_table;
+ u32 mclk = state->performance_levels[state->performance_level_count-1].mclk;
+ u32 i;
+
+ pi->need_update_smu7_dpm_table = 0;
+
+ for (i = 0; i < sclk_table->count; i++) {
+ if (sclk == sclk_table->dpm_levels[i].value)
+ break;
+ }
+
+ if (i >= sclk_table->count) {
+ pi->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_SCLK;
+ } else {
+ /* XXX check display min clock requirements */
+ if (0 != CISLAND_MINIMUM_ENGINE_CLOCK)
+ pi->need_update_smu7_dpm_table |= DPMTABLE_UPDATE_SCLK;
+ }
+
+ for (i = 0; i < mclk_table->count; i++) {
+ if (mclk == mclk_table->dpm_levels[i].value)
+ break;
+ }
+
+ if (i >= mclk_table->count)
+ pi->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_MCLK;
+
+ if (rdev->pm.dpm.current_active_crtc_count !=
+ rdev->pm.dpm.new_active_crtc_count)
+ pi->need_update_smu7_dpm_table |= DPMTABLE_UPDATE_MCLK;
+}
+
+static int ci_populate_and_upload_sclk_mclk_dpm_levels(struct radeon_device *rdev,
+ struct radeon_ps *radeon_state)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_ps *state = ci_get_ps(radeon_state);
+ u32 sclk = state->performance_levels[state->performance_level_count-1].sclk;
+ u32 mclk = state->performance_levels[state->performance_level_count-1].mclk;
+ struct ci_dpm_table *dpm_table = &pi->dpm_table;
+ int ret;
+
+ if (!pi->need_update_smu7_dpm_table)
+ return 0;
+
+ if (pi->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_SCLK)
+ dpm_table->sclk_table.dpm_levels[dpm_table->sclk_table.count-1].value = sclk;
+
+ if (pi->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_MCLK)
+ dpm_table->mclk_table.dpm_levels[dpm_table->mclk_table.count-1].value = mclk;
+
+ if (pi->need_update_smu7_dpm_table & (DPMTABLE_OD_UPDATE_SCLK | DPMTABLE_UPDATE_SCLK)) {
+ ret = ci_populate_all_graphic_levels(rdev);
+ if (ret)
+ return ret;
+ }
+
+ if (pi->need_update_smu7_dpm_table & (DPMTABLE_OD_UPDATE_MCLK | DPMTABLE_UPDATE_MCLK)) {
+ ret = ci_populate_all_memory_levels(rdev);
+ if (ret)
+ return ret;
+ }
+
+ return 0;
+}
+
+static int ci_enable_uvd_dpm(struct radeon_device *rdev, bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ const struct radeon_clock_and_voltage_limits *max_limits;
+ int i;
+
+ if (rdev->pm.dpm.ac_power)
+ max_limits = &rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
+ else
+ max_limits = &rdev->pm.dpm.dyn_state.max_clock_voltage_on_dc;
+
+ if (enable) {
+ pi->dpm_level_enable_mask.uvd_dpm_enable_mask = 0;
+
+ for (i = rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count - 1; i >= 0; i--) {
+ if (rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].v <= max_limits->vddc) {
+ pi->dpm_level_enable_mask.uvd_dpm_enable_mask |= 1 << i;
+
+ if (!pi->caps_uvd_dpm)
+ break;
+ }
+ }
+
+ ci_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_UVDDPM_SetEnabledMask,
+ pi->dpm_level_enable_mask.uvd_dpm_enable_mask);
+
+ if (pi->last_mclk_dpm_enable_mask & 0x1) {
+ pi->uvd_enabled = true;
+ pi->dpm_level_enable_mask.mclk_dpm_enable_mask &= 0xFFFFFFFE;
+ ci_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_MCLKDPM_SetEnabledMask,
+ pi->dpm_level_enable_mask.mclk_dpm_enable_mask);
+ }
+ } else {
+ if (pi->last_mclk_dpm_enable_mask & 0x1) {
+ pi->uvd_enabled = false;
+ pi->dpm_level_enable_mask.mclk_dpm_enable_mask |= 1;
+ ci_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_MCLKDPM_SetEnabledMask,
+ pi->dpm_level_enable_mask.mclk_dpm_enable_mask);
+ }
+ }
+
+ return (ci_send_msg_to_smc(rdev, enable ?
+ PPSMC_MSG_UVDDPM_Enable : PPSMC_MSG_UVDDPM_Disable) == PPSMC_Result_OK) ?
+ 0 : -EINVAL;
+}
+
+#if 0
+static int ci_enable_vce_dpm(struct radeon_device *rdev, bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ const struct radeon_clock_and_voltage_limits *max_limits;
+ int i;
+
+ if (rdev->pm.dpm.ac_power)
+ max_limits = &rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
+ else
+ max_limits = &rdev->pm.dpm.dyn_state.max_clock_voltage_on_dc;
+
+ if (enable) {
+ pi->dpm_level_enable_mask.vce_dpm_enable_mask = 0;
+ for (i = rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.count - 1; i >= 0; i--) {
+ if (rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[i].v <= max_limits->vddc) {
+ pi->dpm_level_enable_mask.vce_dpm_enable_mask |= 1 << i;
+
+ if (!pi->caps_vce_dpm)
+ break;
+ }
+ }
+
+ ci_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_VCEDPM_SetEnabledMask,
+ pi->dpm_level_enable_mask.vce_dpm_enable_mask);
+ }
+
+ return (ci_send_msg_to_smc(rdev, enable ?
+ PPSMC_MSG_VCEDPM_Enable : PPSMC_MSG_VCEDPM_Disable) == PPSMC_Result_OK) ?
+ 0 : -EINVAL;
+}
+
+static int ci_enable_samu_dpm(struct radeon_device *rdev, bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ const struct radeon_clock_and_voltage_limits *max_limits;
+ int i;
+
+ if (rdev->pm.dpm.ac_power)
+ max_limits = &rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
+ else
+ max_limits = &rdev->pm.dpm.dyn_state.max_clock_voltage_on_dc;
+
+ if (enable) {
+ pi->dpm_level_enable_mask.samu_dpm_enable_mask = 0;
+ for (i = rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.count - 1; i >= 0; i--) {
+ if (rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries[i].v <= max_limits->vddc) {
+ pi->dpm_level_enable_mask.samu_dpm_enable_mask |= 1 << i;
+
+ if (!pi->caps_samu_dpm)
+ break;
+ }
+ }
+
+ ci_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_SAMUDPM_SetEnabledMask,
+ pi->dpm_level_enable_mask.samu_dpm_enable_mask);
+ }
+ return (ci_send_msg_to_smc(rdev, enable ?
+ PPSMC_MSG_SAMUDPM_Enable : PPSMC_MSG_SAMUDPM_Disable) == PPSMC_Result_OK) ?
+ 0 : -EINVAL;
+}
+
+static int ci_enable_acp_dpm(struct radeon_device *rdev, bool enable)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ const struct radeon_clock_and_voltage_limits *max_limits;
+ int i;
+
+ if (rdev->pm.dpm.ac_power)
+ max_limits = &rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac;
+ else
+ max_limits = &rdev->pm.dpm.dyn_state.max_clock_voltage_on_dc;
+
+ if (enable) {
+ pi->dpm_level_enable_mask.acp_dpm_enable_mask = 0;
+ for (i = rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.count - 1; i >= 0; i--) {
+ if (rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries[i].v <= max_limits->vddc) {
+ pi->dpm_level_enable_mask.acp_dpm_enable_mask |= 1 << i;
+
+ if (!pi->caps_acp_dpm)
+ break;
+ }
+ }
+
+ ci_send_msg_to_smc_with_parameter(rdev,
+ PPSMC_MSG_ACPDPM_SetEnabledMask,
+ pi->dpm_level_enable_mask.acp_dpm_enable_mask);
+ }
+
+ return (ci_send_msg_to_smc(rdev, enable ?
+ PPSMC_MSG_ACPDPM_Enable : PPSMC_MSG_ACPDPM_Disable) == PPSMC_Result_OK) ?
+ 0 : -EINVAL;
+}
+#endif
+
+static int ci_update_uvd_dpm(struct radeon_device *rdev, bool gate)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 tmp;
+
+ if (!gate) {
+ if (pi->caps_uvd_dpm ||
+ (rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count <= 0))
+ pi->smc_state_table.UvdBootLevel = 0;
+ else
+ pi->smc_state_table.UvdBootLevel =
+ rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count - 1;
+
+ tmp = RREG32_SMC(DPM_TABLE_475);
+ tmp &= ~UvdBootLevel_MASK;
+ tmp |= UvdBootLevel(pi->smc_state_table.UvdBootLevel);
+ WREG32_SMC(DPM_TABLE_475, tmp);
+ }
+
+ return ci_enable_uvd_dpm(rdev, !gate);
+}
+
+#if 0
+static u8 ci_get_vce_boot_level(struct radeon_device *rdev)
+{
+ u8 i;
+ u32 min_evclk = 30000; /* ??? */
+ struct radeon_vce_clock_voltage_dependency_table *table =
+ &rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table;
+
+ for (i = 0; i < table->count; i++) {
+ if (table->entries[i].evclk >= min_evclk)
+ return i;
+ }
+
+ return table->count - 1;
+}
+
+static int ci_update_vce_dpm(struct radeon_device *rdev,
+ struct radeon_ps *radeon_new_state,
+ struct radeon_ps *radeon_current_state)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ bool new_vce_clock_non_zero = (radeon_new_state->evclk != 0);
+ bool old_vce_clock_non_zero = (radeon_current_state->evclk != 0);
+ int ret = 0;
+ u32 tmp;
+
+ if (new_vce_clock_non_zero != old_vce_clock_non_zero) {
+ if (new_vce_clock_non_zero) {
+ pi->smc_state_table.VceBootLevel = ci_get_vce_boot_level(rdev);
+
+ tmp = RREG32_SMC(DPM_TABLE_475);
+ tmp &= ~VceBootLevel_MASK;
+ tmp |= VceBootLevel(pi->smc_state_table.VceBootLevel);
+ WREG32_SMC(DPM_TABLE_475, tmp);
+
+ ret = ci_enable_vce_dpm(rdev, true);
+ } else {
+ ret = ci_enable_vce_dpm(rdev, false);
+ }
+ }
+ return ret;
+}
+
+static int ci_update_samu_dpm(struct radeon_device *rdev, bool gate)
+{
+ return ci_enable_samu_dpm(rdev, gate);
+}
+
+static int ci_update_acp_dpm(struct radeon_device *rdev, bool gate)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 tmp;
+
+ if (!gate) {
+ pi->smc_state_table.AcpBootLevel = 0;
+
+ tmp = RREG32_SMC(DPM_TABLE_475);
+ tmp &= ~AcpBootLevel_MASK;
+ tmp |= AcpBootLevel(pi->smc_state_table.AcpBootLevel);
+ WREG32_SMC(DPM_TABLE_475, tmp);
+ }
+
+ return ci_enable_acp_dpm(rdev, !gate);
+}
+#endif
+
+static int ci_generate_dpm_level_enable_mask(struct radeon_device *rdev,
+ struct radeon_ps *radeon_state)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ int ret;
+
+ ret = ci_trim_dpm_states(rdev, radeon_state);
+ if (ret)
+ return ret;
+
+ pi->dpm_level_enable_mask.sclk_dpm_enable_mask =
+ ci_get_dpm_level_enable_mask_value(&pi->dpm_table.sclk_table);
+ pi->dpm_level_enable_mask.mclk_dpm_enable_mask =
+ ci_get_dpm_level_enable_mask_value(&pi->dpm_table.mclk_table);
+ pi->last_mclk_dpm_enable_mask =
+ pi->dpm_level_enable_mask.mclk_dpm_enable_mask;
+ if (pi->uvd_enabled) {
+ if (pi->dpm_level_enable_mask.mclk_dpm_enable_mask & 1)
+ pi->dpm_level_enable_mask.mclk_dpm_enable_mask &= 0xFFFFFFFE;
+ }
+ pi->dpm_level_enable_mask.pcie_dpm_enable_mask =
+ ci_get_dpm_level_enable_mask_value(&pi->dpm_table.pcie_speed_table);
+
+ return 0;
+}
+
+static int ci_set_mc_special_registers(struct radeon_device *rdev,
+ struct ci_mc_reg_table *table)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u8 i, j, k;
+ u32 temp_reg;
+
+ for (i = 0, j = table->last; i < table->last; i++) {
+ if (j >= SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
+ return -EINVAL;
+ switch(table->mc_reg_address[i].s1 << 2) {
+ case MC_SEQ_MISC1:
+ temp_reg = RREG32(MC_PMG_CMD_EMRS);
+ table->mc_reg_address[j].s1 = MC_PMG_CMD_EMRS >> 2;
+ table->mc_reg_address[j].s0 = MC_SEQ_PMG_CMD_EMRS_LP >> 2;
+ for (k = 0; k < table->num_entries; k++) {
+ table->mc_reg_table_entry[k].mc_data[j] =
+ ((temp_reg & 0xffff0000)) | ((table->mc_reg_table_entry[k].mc_data[i] & 0xffff0000) >> 16);
+ }
+ j++;
+ if (j >= SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
+ return -EINVAL;
+
+ temp_reg = RREG32(MC_PMG_CMD_MRS);
+ table->mc_reg_address[j].s1 = MC_PMG_CMD_MRS >> 2;
+ table->mc_reg_address[j].s0 = MC_SEQ_PMG_CMD_MRS_LP >> 2;
+ for (k = 0; k < table->num_entries; k++) {
+ table->mc_reg_table_entry[k].mc_data[j] =
+ (temp_reg & 0xffff0000) | (table->mc_reg_table_entry[k].mc_data[i] & 0x0000ffff);
+ if (!pi->mem_gddr5)
+ table->mc_reg_table_entry[k].mc_data[j] |= 0x100;
+ }
+ j++;
+ if (j > SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
+ return -EINVAL;
+
+ if (!pi->mem_gddr5) {
+ table->mc_reg_address[j].s1 = MC_PMG_AUTO_CMD >> 2;
+ table->mc_reg_address[j].s0 = MC_PMG_AUTO_CMD >> 2;
+ for (k = 0; k < table->num_entries; k++) {
+ table->mc_reg_table_entry[k].mc_data[j] =
+ (table->mc_reg_table_entry[k].mc_data[i] & 0xffff0000) >> 16;
+ }
+ j++;
+ if (j > SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
+ return -EINVAL;
+ }
+ break;
+ case MC_SEQ_RESERVE_M:
+ temp_reg = RREG32(MC_PMG_CMD_MRS1);
+ table->mc_reg_address[j].s1 = MC_PMG_CMD_MRS1 >> 2;
+ table->mc_reg_address[j].s0 = MC_SEQ_PMG_CMD_MRS1_LP >> 2;
+ for (k = 0; k < table->num_entries; k++) {
+ table->mc_reg_table_entry[k].mc_data[j] =
+ (temp_reg & 0xffff0000) | (table->mc_reg_table_entry[k].mc_data[i] & 0x0000ffff);
+ }
+ j++;
+ if (j > SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
+ return -EINVAL;
+ break;
+ default:
+ break;
+ }
+
+ }
+
+ table->last = j;
+
+ return 0;
+}
+
+static bool ci_check_s0_mc_reg_index(u16 in_reg, u16 *out_reg)
+{
+ bool result = true;
+
+ switch(in_reg) {
+ case MC_SEQ_RAS_TIMING >> 2:
+ *out_reg = MC_SEQ_RAS_TIMING_LP >> 2;
+ break;
+ case MC_SEQ_DLL_STBY >> 2:
+ *out_reg = MC_SEQ_DLL_STBY_LP >> 2;
+ break;
+ case MC_SEQ_G5PDX_CMD0 >> 2:
+ *out_reg = MC_SEQ_G5PDX_CMD0_LP >> 2;
+ break;
+ case MC_SEQ_G5PDX_CMD1 >> 2:
+ *out_reg = MC_SEQ_G5PDX_CMD1_LP >> 2;
+ break;
+ case MC_SEQ_G5PDX_CTRL >> 2:
+ *out_reg = MC_SEQ_G5PDX_CTRL_LP >> 2;
+ break;
+ case MC_SEQ_CAS_TIMING >> 2:
+ *out_reg = MC_SEQ_CAS_TIMING_LP >> 2;
+ break;
+ case MC_SEQ_MISC_TIMING >> 2:
+ *out_reg = MC_SEQ_MISC_TIMING_LP >> 2;
+ break;
+ case MC_SEQ_MISC_TIMING2 >> 2:
+ *out_reg = MC_SEQ_MISC_TIMING2_LP >> 2;
+ break;
+ case MC_SEQ_PMG_DVS_CMD >> 2:
+ *out_reg = MC_SEQ_PMG_DVS_CMD_LP >> 2;
+ break;
+ case MC_SEQ_PMG_DVS_CTL >> 2:
+ *out_reg = MC_SEQ_PMG_DVS_CTL_LP >> 2;
+ break;
+ case MC_SEQ_RD_CTL_D0 >> 2:
+ *out_reg = MC_SEQ_RD_CTL_D0_LP >> 2;
+ break;
+ case MC_SEQ_RD_CTL_D1 >> 2:
+ *out_reg = MC_SEQ_RD_CTL_D1_LP >> 2;
+ break;
+ case MC_SEQ_WR_CTL_D0 >> 2:
+ *out_reg = MC_SEQ_WR_CTL_D0_LP >> 2;
+ break;
+ case MC_SEQ_WR_CTL_D1 >> 2:
+ *out_reg = MC_SEQ_WR_CTL_D1_LP >> 2;
+ break;
+ case MC_PMG_CMD_EMRS >> 2:
+ *out_reg = MC_SEQ_PMG_CMD_EMRS_LP >> 2;
+ break;
+ case MC_PMG_CMD_MRS >> 2:
+ *out_reg = MC_SEQ_PMG_CMD_MRS_LP >> 2;
+ break;
+ case MC_PMG_CMD_MRS1 >> 2:
+ *out_reg = MC_SEQ_PMG_CMD_MRS1_LP >> 2;
+ break;
+ case MC_SEQ_PMG_TIMING >> 2:
+ *out_reg = MC_SEQ_PMG_TIMING_LP >> 2;
+ break;
+ case MC_PMG_CMD_MRS2 >> 2:
+ *out_reg = MC_SEQ_PMG_CMD_MRS2_LP >> 2;
+ break;
+ case MC_SEQ_WR_CTL_2 >> 2:
+ *out_reg = MC_SEQ_WR_CTL_2_LP >> 2;
+ break;
+ default:
+ result = false;
+ break;
+ }
+
+ return result;
+}
+
+static void ci_set_valid_flag(struct ci_mc_reg_table *table)
+{
+ u8 i, j;
+
+ for (i = 0; i < table->last; i++) {
+ for (j = 1; j < table->num_entries; j++) {
+ if (table->mc_reg_table_entry[j-1].mc_data[i] !=
+ table->mc_reg_table_entry[j].mc_data[i]) {
+ table->valid_flag |= 1 << i;
+ break;
+ }
+ }
+ }
+}
+
+static void ci_set_s0_mc_reg_index(struct ci_mc_reg_table *table)
+{
+ u32 i;
+ u16 address;
+
+ for (i = 0; i < table->last; i++) {
+ table->mc_reg_address[i].s0 =
+ ci_check_s0_mc_reg_index(table->mc_reg_address[i].s1, &address) ?
+ address : table->mc_reg_address[i].s1;
+ }
+}
+
+static int ci_copy_vbios_mc_reg_table(const struct atom_mc_reg_table *table,
+ struct ci_mc_reg_table *ci_table)
+{
+ u8 i, j;
+
+ if (table->last > SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
+ return -EINVAL;
+ if (table->num_entries > MAX_AC_TIMING_ENTRIES)
+ return -EINVAL;
+
+ for (i = 0; i < table->last; i++)
+ ci_table->mc_reg_address[i].s1 = table->mc_reg_address[i].s1;
+
+ ci_table->last = table->last;
+
+ for (i = 0; i < table->num_entries; i++) {
+ ci_table->mc_reg_table_entry[i].mclk_max =
+ table->mc_reg_table_entry[i].mclk_max;
+ for (j = 0; j < table->last; j++)
+ ci_table->mc_reg_table_entry[i].mc_data[j] =
+ table->mc_reg_table_entry[i].mc_data[j];
+ }
+ ci_table->num_entries = table->num_entries;
+
+ return 0;
+}
+
+static int ci_initialize_mc_reg_table(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct atom_mc_reg_table *table;
+ struct ci_mc_reg_table *ci_table = &pi->mc_reg_table;
+ u8 module_index = rv770_get_memory_module_index(rdev);
+ int ret;
+
+ table = kzalloc(sizeof(struct atom_mc_reg_table), GFP_KERNEL);
+ if (!table)
+ return -ENOMEM;
+
+ WREG32(MC_SEQ_RAS_TIMING_LP, RREG32(MC_SEQ_RAS_TIMING));
+ WREG32(MC_SEQ_CAS_TIMING_LP, RREG32(MC_SEQ_CAS_TIMING));
+ WREG32(MC_SEQ_DLL_STBY_LP, RREG32(MC_SEQ_DLL_STBY));
+ WREG32(MC_SEQ_G5PDX_CMD0_LP, RREG32(MC_SEQ_G5PDX_CMD0));
+ WREG32(MC_SEQ_G5PDX_CMD1_LP, RREG32(MC_SEQ_G5PDX_CMD1));
+ WREG32(MC_SEQ_G5PDX_CTRL_LP, RREG32(MC_SEQ_G5PDX_CTRL));
+ WREG32(MC_SEQ_PMG_DVS_CMD_LP, RREG32(MC_SEQ_PMG_DVS_CMD));
+ WREG32(MC_SEQ_PMG_DVS_CTL_LP, RREG32(MC_SEQ_PMG_DVS_CTL));
+ WREG32(MC_SEQ_MISC_TIMING_LP, RREG32(MC_SEQ_MISC_TIMING));
+ WREG32(MC_SEQ_MISC_TIMING2_LP, RREG32(MC_SEQ_MISC_TIMING2));
+ WREG32(MC_SEQ_PMG_CMD_EMRS_LP, RREG32(MC_PMG_CMD_EMRS));
+ WREG32(MC_SEQ_PMG_CMD_MRS_LP, RREG32(MC_PMG_CMD_MRS));
+ WREG32(MC_SEQ_PMG_CMD_MRS1_LP, RREG32(MC_PMG_CMD_MRS1));
+ WREG32(MC_SEQ_WR_CTL_D0_LP, RREG32(MC_SEQ_WR_CTL_D0));
+ WREG32(MC_SEQ_WR_CTL_D1_LP, RREG32(MC_SEQ_WR_CTL_D1));
+ WREG32(MC_SEQ_RD_CTL_D0_LP, RREG32(MC_SEQ_RD_CTL_D0));
+ WREG32(MC_SEQ_RD_CTL_D1_LP, RREG32(MC_SEQ_RD_CTL_D1));
+ WREG32(MC_SEQ_PMG_TIMING_LP, RREG32(MC_SEQ_PMG_TIMING));
+ WREG32(MC_SEQ_PMG_CMD_MRS2_LP, RREG32(MC_PMG_CMD_MRS2));
+ WREG32(MC_SEQ_WR_CTL_2_LP, RREG32(MC_SEQ_WR_CTL_2));
+
+ ret = radeon_atom_init_mc_reg_table(rdev, module_index, table);
+ if (ret)
+ goto init_mc_done;
+
+ ret = ci_copy_vbios_mc_reg_table(table, ci_table);
+ if (ret)
+ goto init_mc_done;
+
+ ci_set_s0_mc_reg_index(ci_table);
+
+ ret = ci_set_mc_special_registers(rdev, ci_table);
+ if (ret)
+ goto init_mc_done;
+
+ ci_set_valid_flag(ci_table);
+
+init_mc_done:
+ kfree(table);
+
+ return ret;
+}
+
+static int ci_populate_mc_reg_addresses(struct radeon_device *rdev,
+ SMU7_Discrete_MCRegisters *mc_reg_table)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 i, j;
+
+ for (i = 0, j = 0; j < pi->mc_reg_table.last; j++) {
+ if (pi->mc_reg_table.valid_flag & (1 << j)) {
+ if (i >= SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE)
+ return -EINVAL;
+ mc_reg_table->address[i].s0 = cpu_to_be16(pi->mc_reg_table.mc_reg_address[j].s0);
+ mc_reg_table->address[i].s1 = cpu_to_be16(pi->mc_reg_table.mc_reg_address[j].s1);
+ i++;
+ }
+ }
+
+ mc_reg_table->last = (u8)i;
+
+ return 0;
+}
+
+static void ci_convert_mc_registers(const struct ci_mc_reg_entry *entry,
+ SMU7_Discrete_MCRegisterSet *data,
+ u32 num_entries, u32 valid_flag)
+{
+ u32 i, j;
+
+ for (i = 0, j = 0; j < num_entries; j++) {
+ if (valid_flag & (1 << j)) {
+ data->value[i] = cpu_to_be32(entry->mc_data[j]);
+ i++;
+ }
+ }
+}
+
+static void ci_convert_mc_reg_table_entry_to_smc(struct radeon_device *rdev,
+ const u32 memory_clock,
+ SMU7_Discrete_MCRegisterSet *mc_reg_table_data)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 i = 0;
+
+ for(i = 0; i < pi->mc_reg_table.num_entries; i++) {
+ if (memory_clock <= pi->mc_reg_table.mc_reg_table_entry[i].mclk_max)
+ break;
+ }
+
+ if ((i == pi->mc_reg_table.num_entries) && (i > 0))
+ --i;
+
+ ci_convert_mc_registers(&pi->mc_reg_table.mc_reg_table_entry[i],
+ mc_reg_table_data, pi->mc_reg_table.last,
+ pi->mc_reg_table.valid_flag);
+}
+
+static void ci_convert_mc_reg_table_to_smc(struct radeon_device *rdev,
+ SMU7_Discrete_MCRegisters *mc_reg_table)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 i;
+
+ for (i = 0; i < pi->dpm_table.mclk_table.count; i++)
+ ci_convert_mc_reg_table_entry_to_smc(rdev,
+ pi->dpm_table.mclk_table.dpm_levels[i].value,
+ &mc_reg_table->data[i]);
+}
+
+static int ci_populate_initial_mc_reg_table(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ int ret;
+
+ memset(&pi->smc_mc_reg_table, 0, sizeof(SMU7_Discrete_MCRegisters));
+
+ ret = ci_populate_mc_reg_addresses(rdev, &pi->smc_mc_reg_table);
+ if (ret)
+ return ret;
+ ci_convert_mc_reg_table_to_smc(rdev, &pi->smc_mc_reg_table);
+
+ return ci_copy_bytes_to_smc(rdev,
+ pi->mc_reg_table_start,
+ (u8 *)&pi->smc_mc_reg_table,
+ sizeof(SMU7_Discrete_MCRegisters),
+ pi->sram_end);
+}
+
+static int ci_update_and_upload_mc_reg_table(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ if (!(pi->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_MCLK))
+ return 0;
+
+ memset(&pi->smc_mc_reg_table, 0, sizeof(SMU7_Discrete_MCRegisters));
+
+ ci_convert_mc_reg_table_to_smc(rdev, &pi->smc_mc_reg_table);
+
+ return ci_copy_bytes_to_smc(rdev,
+ pi->mc_reg_table_start +
+ offsetof(SMU7_Discrete_MCRegisters, data[0]),
+ (u8 *)&pi->smc_mc_reg_table.data[0],
+ sizeof(SMU7_Discrete_MCRegisterSet) *
+ pi->dpm_table.mclk_table.count,
+ pi->sram_end);
+}
+
+static void ci_enable_voltage_control(struct radeon_device *rdev)
+{
+ u32 tmp = RREG32_SMC(GENERAL_PWRMGT);
+
+ tmp |= VOLT_PWRMGT_EN;
+ WREG32_SMC(GENERAL_PWRMGT, tmp);
+}
+
+static enum radeon_pcie_gen ci_get_maximum_link_speed(struct radeon_device *rdev,
+ struct radeon_ps *radeon_state)
+{
+ struct ci_ps *state = ci_get_ps(radeon_state);
+ int i;
+ u16 pcie_speed, max_speed = 0;
+
+ for (i = 0; i < state->performance_level_count; i++) {
+ pcie_speed = state->performance_levels[i].pcie_gen;
+ if (max_speed < pcie_speed)
+ max_speed = pcie_speed;
+ }
+
+ return max_speed;
+}
+
+static u16 ci_get_current_pcie_speed(struct radeon_device *rdev)
+{
+ u32 speed_cntl = 0;
+
+ speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL) & LC_CURRENT_DATA_RATE_MASK;
+ speed_cntl >>= LC_CURRENT_DATA_RATE_SHIFT;
+
+ return (u16)speed_cntl;
+}
+
+static int ci_get_current_pcie_lane_number(struct radeon_device *rdev)
+{
+ u32 link_width = 0;
+
+ link_width = RREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL) & LC_LINK_WIDTH_RD_MASK;
+ link_width >>= LC_LINK_WIDTH_RD_SHIFT;
+
+ switch (link_width) {
+ case RADEON_PCIE_LC_LINK_WIDTH_X1:
+ return 1;
+ case RADEON_PCIE_LC_LINK_WIDTH_X2:
+ return 2;
+ case RADEON_PCIE_LC_LINK_WIDTH_X4:
+ return 4;
+ case RADEON_PCIE_LC_LINK_WIDTH_X8:
+ return 8;
+ case RADEON_PCIE_LC_LINK_WIDTH_X12:
+ /* not actually supported */
+ return 12;
+ case RADEON_PCIE_LC_LINK_WIDTH_X0:
+ case RADEON_PCIE_LC_LINK_WIDTH_X16:
+ default:
+ return 16;
+ }
+}
+
+static void ci_request_link_speed_change_before_state_change(struct radeon_device *rdev,
+ struct radeon_ps *radeon_new_state,
+ struct radeon_ps *radeon_current_state)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ enum radeon_pcie_gen target_link_speed =
+ ci_get_maximum_link_speed(rdev, radeon_new_state);
+ enum radeon_pcie_gen current_link_speed;
+
+ if (pi->force_pcie_gen == RADEON_PCIE_GEN_INVALID)
+ current_link_speed = ci_get_maximum_link_speed(rdev, radeon_current_state);
+ else
+ current_link_speed = pi->force_pcie_gen;
+
+ pi->force_pcie_gen = RADEON_PCIE_GEN_INVALID;
+ pi->pspp_notify_required = false;
+ if (target_link_speed > current_link_speed) {
+ switch (target_link_speed) {
+ case RADEON_PCIE_GEN3:
+ if (radeon_acpi_pcie_performance_request(rdev, PCIE_PERF_REQ_PECI_GEN3, false) == 0)
+ break;
+ pi->force_pcie_gen = RADEON_PCIE_GEN2;
+ if (current_link_speed == RADEON_PCIE_GEN2)
+ break;
+ case RADEON_PCIE_GEN2:
+ if (radeon_acpi_pcie_performance_request(rdev, PCIE_PERF_REQ_PECI_GEN2, false) == 0)
+ break;
+ default:
+ pi->force_pcie_gen = ci_get_current_pcie_speed(rdev);
+ break;
+ }
+ } else {
+ if (target_link_speed < current_link_speed)
+ pi->pspp_notify_required = true;
+ }
+}
+
+static void ci_notify_link_speed_change_after_state_change(struct radeon_device *rdev,
+ struct radeon_ps *radeon_new_state,
+ struct radeon_ps *radeon_current_state)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ enum radeon_pcie_gen target_link_speed =
+ ci_get_maximum_link_speed(rdev, radeon_new_state);
+ u8 request;
+
+ if (pi->pspp_notify_required) {
+ if (target_link_speed == RADEON_PCIE_GEN3)
+ request = PCIE_PERF_REQ_PECI_GEN3;
+ else if (target_link_speed == RADEON_PCIE_GEN2)
+ request = PCIE_PERF_REQ_PECI_GEN2;
+ else
+ request = PCIE_PERF_REQ_PECI_GEN1;
+
+ if ((request == PCIE_PERF_REQ_PECI_GEN1) &&
+ (ci_get_current_pcie_speed(rdev) > 0))
+ return;
+
+ radeon_acpi_pcie_performance_request(rdev, request, false);
+ }
+}
+
+static int ci_set_private_data_variables_based_on_pptable(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct radeon_clock_voltage_dependency_table *allowed_sclk_vddc_table =
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
+ struct radeon_clock_voltage_dependency_table *allowed_mclk_vddc_table =
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk;
+ struct radeon_clock_voltage_dependency_table *allowed_mclk_vddci_table =
+ &rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk;
+
+ if (allowed_sclk_vddc_table == NULL)
+ return -EINVAL;
+ if (allowed_sclk_vddc_table->count < 1)
+ return -EINVAL;
+ if (allowed_mclk_vddc_table == NULL)
+ return -EINVAL;
+ if (allowed_mclk_vddc_table->count < 1)
+ return -EINVAL;
+ if (allowed_mclk_vddci_table == NULL)
+ return -EINVAL;
+ if (allowed_mclk_vddci_table->count < 1)
+ return -EINVAL;
+
+ pi->min_vddc_in_pp_table = allowed_sclk_vddc_table->entries[0].v;
+ pi->max_vddc_in_pp_table =
+ allowed_sclk_vddc_table->entries[allowed_sclk_vddc_table->count - 1].v;
+
+ pi->min_vddci_in_pp_table = allowed_mclk_vddci_table->entries[0].v;
+ pi->max_vddci_in_pp_table =
+ allowed_mclk_vddci_table->entries[allowed_mclk_vddci_table->count - 1].v;
+
+ rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac.sclk =
+ allowed_sclk_vddc_table->entries[allowed_sclk_vddc_table->count - 1].clk;
+ rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac.mclk =
+ allowed_mclk_vddc_table->entries[allowed_sclk_vddc_table->count - 1].clk;
+ rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac.vddc =
+ allowed_sclk_vddc_table->entries[allowed_sclk_vddc_table->count - 1].v;
+ rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac.vddci =
+ allowed_mclk_vddci_table->entries[allowed_mclk_vddci_table->count - 1].v;
+
+ return 0;
+}
+
+static void ci_patch_with_vddc_leakage(struct radeon_device *rdev, u16 *vddc)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_leakage_voltage *leakage_table = &pi->vddc_leakage;
+ u32 leakage_index;
+
+ for (leakage_index = 0; leakage_index < leakage_table->count; leakage_index++) {
+ if (leakage_table->leakage_id[leakage_index] == *vddc) {
+ *vddc = leakage_table->actual_voltage[leakage_index];
+ break;
+ }
+ }
+}
+
+static void ci_patch_with_vddci_leakage(struct radeon_device *rdev, u16 *vddci)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_leakage_voltage *leakage_table = &pi->vddci_leakage;
+ u32 leakage_index;
+
+ for (leakage_index = 0; leakage_index < leakage_table->count; leakage_index++) {
+ if (leakage_table->leakage_id[leakage_index] == *vddci) {
+ *vddci = leakage_table->actual_voltage[leakage_index];
+ break;
+ }
+ }
+}
+
+static void ci_patch_clock_voltage_dependency_table_with_vddc_leakage(struct radeon_device *rdev,
+ struct radeon_clock_voltage_dependency_table *table)
+{
+ u32 i;
+
+ if (table) {
+ for (i = 0; i < table->count; i++)
+ ci_patch_with_vddc_leakage(rdev, &table->entries[i].v);
+ }
+}
+
+static void ci_patch_clock_voltage_dependency_table_with_vddci_leakage(struct radeon_device *rdev,
+ struct radeon_clock_voltage_dependency_table *table)
+{
+ u32 i;
+
+ if (table) {
+ for (i = 0; i < table->count; i++)
+ ci_patch_with_vddci_leakage(rdev, &table->entries[i].v);
+ }
+}
+
+static void ci_patch_vce_clock_voltage_dependency_table_with_vddc_leakage(struct radeon_device *rdev,
+ struct radeon_vce_clock_voltage_dependency_table *table)
+{
+ u32 i;
+
+ if (table) {
+ for (i = 0; i < table->count; i++)
+ ci_patch_with_vddc_leakage(rdev, &table->entries[i].v);
+ }
+}
+
+static void ci_patch_uvd_clock_voltage_dependency_table_with_vddc_leakage(struct radeon_device *rdev,
+ struct radeon_uvd_clock_voltage_dependency_table *table)
+{
+ u32 i;
+
+ if (table) {
+ for (i = 0; i < table->count; i++)
+ ci_patch_with_vddc_leakage(rdev, &table->entries[i].v);
+ }
+}
+
+static void ci_patch_vddc_phase_shed_limit_table_with_vddc_leakage(struct radeon_device *rdev,
+ struct radeon_phase_shedding_limits_table *table)
+{
+ u32 i;
+
+ if (table) {
+ for (i = 0; i < table->count; i++)
+ ci_patch_with_vddc_leakage(rdev, &table->entries[i].voltage);
+ }
+}
+
+static void ci_patch_clock_voltage_limits_with_vddc_leakage(struct radeon_device *rdev,
+ struct radeon_clock_and_voltage_limits *table)
+{
+ if (table) {
+ ci_patch_with_vddc_leakage(rdev, (u16 *)&table->vddc);
+ ci_patch_with_vddci_leakage(rdev, (u16 *)&table->vddci);
+ }
+}
+
+static void ci_patch_cac_leakage_table_with_vddc_leakage(struct radeon_device *rdev,
+ struct radeon_cac_leakage_table *table)
+{
+ u32 i;
+
+ if (table) {
+ for (i = 0; i < table->count; i++)
+ ci_patch_with_vddc_leakage(rdev, &table->entries[i].vddc);
+ }
+}
+
+static void ci_patch_dependency_tables_with_leakage(struct radeon_device *rdev)
+{
+
+ ci_patch_clock_voltage_dependency_table_with_vddc_leakage(rdev,
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk);
+ ci_patch_clock_voltage_dependency_table_with_vddc_leakage(rdev,
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk);
+ ci_patch_clock_voltage_dependency_table_with_vddc_leakage(rdev,
+ &rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk);
+ ci_patch_clock_voltage_dependency_table_with_vddci_leakage(rdev,
+ &rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk);
+ ci_patch_vce_clock_voltage_dependency_table_with_vddc_leakage(rdev,
+ &rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table);
+ ci_patch_uvd_clock_voltage_dependency_table_with_vddc_leakage(rdev,
+ &rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table);
+ ci_patch_clock_voltage_dependency_table_with_vddc_leakage(rdev,
+ &rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table);
+ ci_patch_clock_voltage_dependency_table_with_vddc_leakage(rdev,
+ &rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table);
+ ci_patch_vddc_phase_shed_limit_table_with_vddc_leakage(rdev,
+ &rdev->pm.dpm.dyn_state.phase_shedding_limits_table);
+ ci_patch_clock_voltage_limits_with_vddc_leakage(rdev,
+ &rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac);
+ ci_patch_clock_voltage_limits_with_vddc_leakage(rdev,
+ &rdev->pm.dpm.dyn_state.max_clock_voltage_on_dc);
+ ci_patch_cac_leakage_table_with_vddc_leakage(rdev,
+ &rdev->pm.dpm.dyn_state.cac_leakage_table);
+
+}
+
+static void ci_get_memory_type(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 tmp;
+
+ tmp = RREG32(MC_SEQ_MISC0);
+
+ if (((tmp & MC_SEQ_MISC0_GDDR5_MASK) >> MC_SEQ_MISC0_GDDR5_SHIFT) ==
+ MC_SEQ_MISC0_GDDR5_VALUE)
+ pi->mem_gddr5 = true;
+ else
+ pi->mem_gddr5 = false;
+
+}
+
+void ci_update_current_ps(struct radeon_device *rdev,
+ struct radeon_ps *rps)
+{
+ struct ci_ps *new_ps = ci_get_ps(rps);
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ pi->current_rps = *rps;
+ pi->current_ps = *new_ps;
+ pi->current_rps.ps_priv = &pi->current_ps;
+}
+
+void ci_update_requested_ps(struct radeon_device *rdev,
+ struct radeon_ps *rps)
+{
+ struct ci_ps *new_ps = ci_get_ps(rps);
+ struct ci_power_info *pi = ci_get_pi(rdev);
+
+ pi->requested_rps = *rps;
+ pi->requested_ps = *new_ps;
+ pi->requested_rps.ps_priv = &pi->requested_ps;
+}
+
+int ci_dpm_pre_set_power_state(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct radeon_ps requested_ps = *rdev->pm.dpm.requested_ps;
+ struct radeon_ps *new_ps = &requested_ps;
+
+ ci_update_requested_ps(rdev, new_ps);
+
+ ci_apply_state_adjust_rules(rdev, &pi->requested_rps);
+
+ return 0;
+}
+
+void ci_dpm_post_set_power_state(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct radeon_ps *new_ps = &pi->requested_rps;
+
+ ci_update_current_ps(rdev, new_ps);
+}
+
+
+void ci_dpm_setup_asic(struct radeon_device *rdev)
+{
+ ci_read_clock_registers(rdev);
+ ci_get_memory_type(rdev);
+ ci_enable_acpi_power_management(rdev);
+ ci_init_sclk_t(rdev);
+}
+
+int ci_dpm_enable(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct radeon_ps *boot_ps = rdev->pm.dpm.boot_ps;
+ int ret;
+
+ if (ci_is_smc_running(rdev))
+ return -EINVAL;
+ if (pi->voltage_control != CISLANDS_VOLTAGE_CONTROL_NONE) {
+ ci_enable_voltage_control(rdev);
+ ret = ci_construct_voltage_tables(rdev);
+ if (ret) {
+ DRM_ERROR("ci_construct_voltage_tables failed\n");
+ return ret;
+ }
+ }
+ if (pi->caps_dynamic_ac_timing) {
+ ret = ci_initialize_mc_reg_table(rdev);
+ if (ret)
+ pi->caps_dynamic_ac_timing = false;
+ }
+ if (pi->dynamic_ss)
+ ci_enable_spread_spectrum(rdev, true);
+ if (pi->thermal_protection)
+ ci_enable_thermal_protection(rdev, true);
+ ci_program_sstp(rdev);
+ ci_enable_display_gap(rdev);
+ ci_program_vc(rdev);
+ ret = ci_upload_firmware(rdev);
+ if (ret) {
+ DRM_ERROR("ci_upload_firmware failed\n");
+ return ret;
+ }
+ ret = ci_process_firmware_header(rdev);
+ if (ret) {
+ DRM_ERROR("ci_process_firmware_header failed\n");
+ return ret;
+ }
+ ret = ci_initial_switch_from_arb_f0_to_f1(rdev);
+ if (ret) {
+ DRM_ERROR("ci_initial_switch_from_arb_f0_to_f1 failed\n");
+ return ret;
+ }
+ ret = ci_init_smc_table(rdev);
+ if (ret) {
+ DRM_ERROR("ci_init_smc_table failed\n");
+ return ret;
+ }
+ ret = ci_init_arb_table_index(rdev);
+ if (ret) {
+ DRM_ERROR("ci_init_arb_table_index failed\n");
+ return ret;
+ }
+ if (pi->caps_dynamic_ac_timing) {
+ ret = ci_populate_initial_mc_reg_table(rdev);
+ if (ret) {
+ DRM_ERROR("ci_populate_initial_mc_reg_table failed\n");
+ return ret;
+ }
+ }
+ ret = ci_populate_pm_base(rdev);
+ if (ret) {
+ DRM_ERROR("ci_populate_pm_base failed\n");
+ return ret;
+ }
+ ci_dpm_start_smc(rdev);
+ ci_enable_vr_hot_gpio_interrupt(rdev);
+ ret = ci_notify_smc_display_change(rdev, false);
+ if (ret) {
+ DRM_ERROR("ci_notify_smc_display_change failed\n");
+ return ret;
+ }
+ ci_enable_sclk_control(rdev, true);
+ ret = ci_enable_ulv(rdev, true);
+ if (ret) {
+ DRM_ERROR("ci_enable_ulv failed\n");
+ return ret;
+ }
+ ret = ci_enable_ds_master_switch(rdev, true);
+ if (ret) {
+ DRM_ERROR("ci_enable_ds_master_switch failed\n");
+ return ret;
+ }
+ ret = ci_start_dpm(rdev);
+ if (ret) {
+ DRM_ERROR("ci_start_dpm failed\n");
+ return ret;
+ }
+ ret = ci_enable_didt(rdev, true);
+ if (ret) {
+ DRM_ERROR("ci_enable_didt failed\n");
+ return ret;
+ }
+ ret = ci_enable_smc_cac(rdev, true);
+ if (ret) {
+ DRM_ERROR("ci_enable_smc_cac failed\n");
+ return ret;
+ }
+ ret = ci_enable_power_containment(rdev, true);
+ if (ret) {
+ DRM_ERROR("ci_enable_power_containment failed\n");
+ return ret;
+ }
+ if (rdev->irq.installed &&
+ r600_is_internal_thermal_sensor(rdev->pm.int_thermal_type)) {
+#if 0
+ PPSMC_Result result;
+#endif
+ ret = ci_set_thermal_temperature_range(rdev, R600_TEMP_RANGE_MIN, R600_TEMP_RANGE_MAX);
+ if (ret) {
+ DRM_ERROR("ci_set_thermal_temperature_range failed\n");
+ return ret;
+ }
+ rdev->irq.dpm_thermal = true;
+ radeon_irq_set(rdev);
+#if 0
+ result = ci_send_msg_to_smc(rdev, PPSMC_MSG_EnableThermalInterrupt);
+
+ if (result != PPSMC_Result_OK)
+ DRM_DEBUG_KMS("Could not enable thermal interrupts.\n");
+#endif
+ }
+
+ ci_enable_auto_throttle_source(rdev, RADEON_DPM_AUTO_THROTTLE_SRC_THERMAL, true);
+
+ ci_update_current_ps(rdev, boot_ps);
+
+ return 0;
+}
+
+void ci_dpm_disable(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct radeon_ps *boot_ps = rdev->pm.dpm.boot_ps;
+
+ if (!ci_is_smc_running(rdev))
+ return;
+
+ if (pi->thermal_protection)
+ ci_enable_thermal_protection(rdev, false);
+ ci_enable_power_containment(rdev, false);
+ ci_enable_smc_cac(rdev, false);
+ ci_enable_didt(rdev, false);
+ ci_enable_spread_spectrum(rdev, false);
+ ci_enable_auto_throttle_source(rdev, RADEON_DPM_AUTO_THROTTLE_SRC_THERMAL, false);
+ ci_stop_dpm(rdev);
+ ci_enable_ds_master_switch(rdev, true);
+ ci_enable_ulv(rdev, false);
+ ci_clear_vc(rdev);
+ ci_reset_to_default(rdev);
+ ci_dpm_stop_smc(rdev);
+ ci_force_switch_to_arb_f0(rdev);
+
+ ci_update_current_ps(rdev, boot_ps);
+}
+
+int ci_dpm_set_power_state(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct radeon_ps *new_ps = &pi->requested_rps;
+ struct radeon_ps *old_ps = &pi->current_rps;
+ int ret;
+
+ ci_find_dpm_states_clocks_in_dpm_table(rdev, new_ps);
+ if (pi->pcie_performance_request)
+ ci_request_link_speed_change_before_state_change(rdev, new_ps, old_ps);
+ ret = ci_freeze_sclk_mclk_dpm(rdev);
+ if (ret) {
+ DRM_ERROR("ci_freeze_sclk_mclk_dpm failed\n");
+ return ret;
+ }
+ ret = ci_populate_and_upload_sclk_mclk_dpm_levels(rdev, new_ps);
+ if (ret) {
+ DRM_ERROR("ci_populate_and_upload_sclk_mclk_dpm_levels failed\n");
+ return ret;
+ }
+ ret = ci_generate_dpm_level_enable_mask(rdev, new_ps);
+ if (ret) {
+ DRM_ERROR("ci_generate_dpm_level_enable_mask failed\n");
+ return ret;
+ }
+#if 0
+ ret = ci_update_vce_dpm(rdev, new_ps, old_ps);
+ if (ret) {
+ DRM_ERROR("ci_update_vce_dpm failed\n");
+ return ret;
+ }
+#endif
+ ret = ci_update_uvd_dpm(rdev, false);
+ if (ret) {
+ DRM_ERROR("ci_update_uvd_dpm failed\n");
+ return ret;
+ }
+ ret = ci_update_sclk_t(rdev);
+ if (ret) {
+ DRM_ERROR("ci_update_sclk_t failed\n");
+ return ret;
+ }
+ if (pi->caps_dynamic_ac_timing) {
+ ret = ci_update_and_upload_mc_reg_table(rdev);
+ if (ret) {
+ DRM_ERROR("ci_update_and_upload_mc_reg_table failed\n");
+ return ret;
+ }
+ }
+ ret = ci_program_memory_timing_parameters(rdev);
+ if (ret) {
+ DRM_ERROR("ci_program_memory_timing_parameters failed\n");
+ return ret;
+ }
+ ret = ci_unfreeze_sclk_mclk_dpm(rdev);
+ if (ret) {
+ DRM_ERROR("ci_unfreeze_sclk_mclk_dpm failed\n");
+ return ret;
+ }
+ ret = ci_upload_dpm_level_enable_mask(rdev);
+ if (ret) {
+ DRM_ERROR("ci_upload_dpm_level_enable_mask failed\n");
+ return ret;
+ }
+ if (pi->pcie_performance_request)
+ ci_notify_link_speed_change_after_state_change(rdev, new_ps, old_ps);
+
+ return 0;
+}
+
+int ci_dpm_power_control_set_level(struct radeon_device *rdev)
+{
+ return ci_power_control_set_level(rdev);
+}
+
+void ci_dpm_reset_asic(struct radeon_device *rdev)
+{
+ ci_set_boot_state(rdev);
+}
+
+void ci_dpm_display_configuration_changed(struct radeon_device *rdev)
+{
+ ci_program_display_gap(rdev);
+}
+
+union power_info {
+ struct _ATOM_POWERPLAY_INFO info;
+ struct _ATOM_POWERPLAY_INFO_V2 info_2;
+ struct _ATOM_POWERPLAY_INFO_V3 info_3;
+ struct _ATOM_PPLIB_POWERPLAYTABLE pplib;
+ struct _ATOM_PPLIB_POWERPLAYTABLE2 pplib2;
+ struct _ATOM_PPLIB_POWERPLAYTABLE3 pplib3;
+};
+
+union pplib_clock_info {
+ struct _ATOM_PPLIB_R600_CLOCK_INFO r600;
+ struct _ATOM_PPLIB_RS780_CLOCK_INFO rs780;
+ struct _ATOM_PPLIB_EVERGREEN_CLOCK_INFO evergreen;
+ struct _ATOM_PPLIB_SUMO_CLOCK_INFO sumo;
+ struct _ATOM_PPLIB_SI_CLOCK_INFO si;
+ struct _ATOM_PPLIB_CI_CLOCK_INFO ci;
+};
+
+union pplib_power_state {
+ struct _ATOM_PPLIB_STATE v1;
+ struct _ATOM_PPLIB_STATE_V2 v2;
+};
+
+static void ci_parse_pplib_non_clock_info(struct radeon_device *rdev,
+ struct radeon_ps *rps,
+ struct _ATOM_PPLIB_NONCLOCK_INFO *non_clock_info,
+ u8 table_rev)
+{
+ rps->caps = le32_to_cpu(non_clock_info->ulCapsAndSettings);
+ rps->class = le16_to_cpu(non_clock_info->usClassification);
+ rps->class2 = le16_to_cpu(non_clock_info->usClassification2);
+
+ if (ATOM_PPLIB_NONCLOCKINFO_VER1 < table_rev) {
+ rps->vclk = le32_to_cpu(non_clock_info->ulVCLK);
+ rps->dclk = le32_to_cpu(non_clock_info->ulDCLK);
+ } else {
+ rps->vclk = 0;
+ rps->dclk = 0;
+ }
+
+ if (rps->class & ATOM_PPLIB_CLASSIFICATION_BOOT)
+ rdev->pm.dpm.boot_ps = rps;
+ if (rps->class & ATOM_PPLIB_CLASSIFICATION_UVDSTATE)
+ rdev->pm.dpm.uvd_ps = rps;
+}
+
+static void ci_parse_pplib_clock_info(struct radeon_device *rdev,
+ struct radeon_ps *rps, int index,
+ union pplib_clock_info *clock_info)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_ps *ps = ci_get_ps(rps);
+ struct ci_pl *pl = &ps->performance_levels[index];
+
+ ps->performance_level_count = index + 1;
+
+ pl->sclk = le16_to_cpu(clock_info->ci.usEngineClockLow);
+ pl->sclk |= clock_info->ci.ucEngineClockHigh << 16;
+ pl->mclk = le16_to_cpu(clock_info->ci.usMemoryClockLow);
+ pl->mclk |= clock_info->ci.ucMemoryClockHigh << 16;
+
+ pl->pcie_gen = r600_get_pcie_gen_support(rdev,
+ pi->sys_pcie_mask,
+ pi->vbios_boot_state.pcie_gen_bootup_value,
+ clock_info->ci.ucPCIEGen);
+ pl->pcie_lane = r600_get_pcie_lane_support(rdev,
+ pi->vbios_boot_state.pcie_lane_bootup_value,
+ le16_to_cpu(clock_info->ci.usPCIELane));
+
+ if (rps->class & ATOM_PPLIB_CLASSIFICATION_ACPI) {
+ pi->acpi_pcie_gen = pl->pcie_gen;
+ }
+
+ if (rps->class2 & ATOM_PPLIB_CLASSIFICATION2_ULV) {
+ pi->ulv.supported = true;
+ pi->ulv.pl = *pl;
+ pi->ulv.cg_ulv_parameter = CISLANDS_CGULVPARAMETER_DFLT;
+ }
+
+ /* patch up boot state */
+ if (rps->class & ATOM_PPLIB_CLASSIFICATION_BOOT) {
+ pl->mclk = pi->vbios_boot_state.mclk_bootup_value;
+ pl->sclk = pi->vbios_boot_state.sclk_bootup_value;
+ pl->pcie_gen = pi->vbios_boot_state.pcie_gen_bootup_value;
+ pl->pcie_lane = pi->vbios_boot_state.pcie_lane_bootup_value;
+ }
+
+ switch (rps->class & ATOM_PPLIB_CLASSIFICATION_UI_MASK) {
+ case ATOM_PPLIB_CLASSIFICATION_UI_BATTERY:
+ pi->use_pcie_powersaving_levels = true;
+ if (pi->pcie_gen_powersaving.max < pl->pcie_gen)
+ pi->pcie_gen_powersaving.max = pl->pcie_gen;
+ if (pi->pcie_gen_powersaving.min > pl->pcie_gen)
+ pi->pcie_gen_powersaving.min = pl->pcie_gen;
+ if (pi->pcie_lane_powersaving.max < pl->pcie_lane)
+ pi->pcie_lane_powersaving.max = pl->pcie_lane;
+ if (pi->pcie_lane_powersaving.min > pl->pcie_lane)
+ pi->pcie_lane_powersaving.min = pl->pcie_lane;
+ break;
+ case ATOM_PPLIB_CLASSIFICATION_UI_PERFORMANCE:
+ pi->use_pcie_performance_levels = true;
+ if (pi->pcie_gen_performance.max < pl->pcie_gen)
+ pi->pcie_gen_performance.max = pl->pcie_gen;
+ if (pi->pcie_gen_performance.min > pl->pcie_gen)
+ pi->pcie_gen_performance.min = pl->pcie_gen;
+ if (pi->pcie_lane_performance.max < pl->pcie_lane)
+ pi->pcie_lane_performance.max = pl->pcie_lane;
+ if (pi->pcie_lane_performance.min > pl->pcie_lane)
+ pi->pcie_lane_performance.min = pl->pcie_lane;
+ break;
+ default:
+ break;
+ }
+}
+
+static int ci_parse_power_table(struct radeon_device *rdev)
+{
+ struct radeon_mode_info *mode_info = &rdev->mode_info;
+ struct _ATOM_PPLIB_NONCLOCK_INFO *non_clock_info;
+ union pplib_power_state *power_state;
+ int i, j, k, non_clock_array_index, clock_array_index;
+ union pplib_clock_info *clock_info;
+ struct _StateArray *state_array;
+ struct _ClockInfoArray *clock_info_array;
+ struct _NonClockInfoArray *non_clock_info_array;
+ union power_info *power_info;
+ int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
+ u16 data_offset;
+ u8 frev, crev;
+ u8 *power_state_offset;
+ struct ci_ps *ps;
+
+ if (!atom_parse_data_header(mode_info->atom_context, index, NULL,
+ &frev, &crev, &data_offset))
+ return -EINVAL;
+ power_info = (union power_info *)(mode_info->atom_context->bios + data_offset);
+
+ state_array = (struct _StateArray *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(power_info->pplib.usStateArrayOffset));
+ clock_info_array = (struct _ClockInfoArray *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(power_info->pplib.usClockInfoArrayOffset));
+ non_clock_info_array = (struct _NonClockInfoArray *)
+ (mode_info->atom_context->bios + data_offset +
+ le16_to_cpu(power_info->pplib.usNonClockInfoArrayOffset));
+
+ rdev->pm.dpm.ps = kzalloc(sizeof(struct radeon_ps) *
+ state_array->ucNumEntries, GFP_KERNEL);
+ if (!rdev->pm.dpm.ps)
+ return -ENOMEM;
+ power_state_offset = (u8 *)state_array->states;
+ rdev->pm.dpm.platform_caps = le32_to_cpu(power_info->pplib.ulPlatformCaps);
+ rdev->pm.dpm.backbias_response_time = le16_to_cpu(power_info->pplib.usBackbiasTime);
+ rdev->pm.dpm.voltage_response_time = le16_to_cpu(power_info->pplib.usVoltageTime);
+ for (i = 0; i < state_array->ucNumEntries; i++) {
+ power_state = (union pplib_power_state *)power_state_offset;
+ non_clock_array_index = power_state->v2.nonClockInfoIndex;
+ non_clock_info = (struct _ATOM_PPLIB_NONCLOCK_INFO *)
+ &non_clock_info_array->nonClockInfo[non_clock_array_index];
+ if (!rdev->pm.power_state[i].clock_info)
+ return -EINVAL;
+ ps = kzalloc(sizeof(struct ci_ps), GFP_KERNEL);
+ if (ps == NULL) {
+ kfree(rdev->pm.dpm.ps);
+ return -ENOMEM;
+ }
+ rdev->pm.dpm.ps[i].ps_priv = ps;
+ ci_parse_pplib_non_clock_info(rdev, &rdev->pm.dpm.ps[i],
+ non_clock_info,
+ non_clock_info_array->ucEntrySize);
+ k = 0;
+ for (j = 0; j < power_state->v2.ucNumDPMLevels; j++) {
+ clock_array_index = power_state->v2.clockInfoIndex[j];
+ if (clock_array_index >= clock_info_array->ucNumEntries)
+ continue;
+ if (k >= CISLANDS_MAX_HARDWARE_POWERLEVELS)
+ break;
+ clock_info = (union pplib_clock_info *)
+ &clock_info_array->clockInfo[clock_array_index * clock_info_array->ucEntrySize];
+ ci_parse_pplib_clock_info(rdev,
+ &rdev->pm.dpm.ps[i], k,
+ clock_info);
+ k++;
+ }
+ power_state_offset += 2 + power_state->v2.ucNumDPMLevels;
+ }
+ rdev->pm.dpm.num_ps = state_array->ucNumEntries;
+ return 0;
+}
+
+int ci_get_vbios_boot_values(struct radeon_device *rdev,
+ struct ci_vbios_boot_state *boot_state)
+{
+ struct radeon_mode_info *mode_info = &rdev->mode_info;
+ int index = GetIndexIntoMasterTable(DATA, FirmwareInfo);
+ ATOM_FIRMWARE_INFO_V2_2 *firmware_info;
+ u8 frev, crev;
+ u16 data_offset;
+
+ if (atom_parse_data_header(mode_info->atom_context, index, NULL,
+ &frev, &crev, &data_offset)) {
+ firmware_info =
+ (ATOM_FIRMWARE_INFO_V2_2 *)(mode_info->atom_context->bios +
+ data_offset);
+ boot_state->mvdd_bootup_value = le16_to_cpu(firmware_info->usBootUpMVDDCVoltage);
+ boot_state->vddc_bootup_value = le16_to_cpu(firmware_info->usBootUpVDDCVoltage);
+ boot_state->vddci_bootup_value = le16_to_cpu(firmware_info->usBootUpVDDCIVoltage);
+ boot_state->pcie_gen_bootup_value = ci_get_current_pcie_speed(rdev);
+ boot_state->pcie_lane_bootup_value = ci_get_current_pcie_lane_number(rdev);
+ boot_state->sclk_bootup_value = le32_to_cpu(firmware_info->ulDefaultEngineClock);
+ boot_state->mclk_bootup_value = le32_to_cpu(firmware_info->ulDefaultMemoryClock);
+
+ return 0;
+ }
+ return -EINVAL;
+}
+
+void ci_dpm_fini(struct radeon_device *rdev)
+{
+ int i;
+
+ for (i = 0; i < rdev->pm.dpm.num_ps; i++) {
+ kfree(rdev->pm.dpm.ps[i].ps_priv);
+ }
+ kfree(rdev->pm.dpm.ps);
+ kfree(rdev->pm.dpm.priv);
+ kfree(rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries);
+ r600_free_extended_power_table(rdev);
+}
+
+int ci_dpm_init(struct radeon_device *rdev)
+{
+ int index = GetIndexIntoMasterTable(DATA, ASIC_InternalSS_Info);
+ u16 data_offset, size;
+ u8 frev, crev;
+ struct ci_power_info *pi;
+ int ret;
+ u32 mask;
+
+ pi = kzalloc(sizeof(struct ci_power_info), GFP_KERNEL);
+ if (pi == NULL)
+ return -ENOMEM;
+ rdev->pm.dpm.priv = pi;
+
+ ret = drm_pcie_get_speed_cap_mask(rdev->ddev, &mask);
+ if (ret)
+ pi->sys_pcie_mask = 0;
+ else
+ pi->sys_pcie_mask = mask;
+ pi->force_pcie_gen = RADEON_PCIE_GEN_INVALID;
+
+ pi->pcie_gen_performance.max = RADEON_PCIE_GEN1;
+ pi->pcie_gen_performance.min = RADEON_PCIE_GEN3;
+ pi->pcie_gen_powersaving.max = RADEON_PCIE_GEN1;
+ pi->pcie_gen_powersaving.min = RADEON_PCIE_GEN3;
+
+ pi->pcie_lane_performance.max = 0;
+ pi->pcie_lane_performance.min = 16;
+ pi->pcie_lane_powersaving.max = 0;
+ pi->pcie_lane_powersaving.min = 16;
+
+ ret = ci_get_vbios_boot_values(rdev, &pi->vbios_boot_state);
+ if (ret) {
+ ci_dpm_fini(rdev);
+ return ret;
+ }
+ ret = ci_parse_power_table(rdev);
+ if (ret) {
+ ci_dpm_fini(rdev);
+ return ret;
+ }
+ ret = r600_parse_extended_power_table(rdev);
+ if (ret) {
+ ci_dpm_fini(rdev);
+ return ret;
+ }
+
+ pi->dll_default_on = false;
+ pi->sram_end = SMC_RAM_END;
+
+ pi->activity_target[0] = CISLAND_TARGETACTIVITY_DFLT;
+ pi->activity_target[1] = CISLAND_TARGETACTIVITY_DFLT;
+ pi->activity_target[2] = CISLAND_TARGETACTIVITY_DFLT;
+ pi->activity_target[3] = CISLAND_TARGETACTIVITY_DFLT;
+ pi->activity_target[4] = CISLAND_TARGETACTIVITY_DFLT;
+ pi->activity_target[5] = CISLAND_TARGETACTIVITY_DFLT;
+ pi->activity_target[6] = CISLAND_TARGETACTIVITY_DFLT;
+ pi->activity_target[7] = CISLAND_TARGETACTIVITY_DFLT;
+
+ pi->mclk_activity_target = CISLAND_MCLK_TARGETACTIVITY_DFLT;
+
+ pi->sclk_dpm_key_disabled = 0;
+ pi->mclk_dpm_key_disabled = 0;
+ pi->pcie_dpm_key_disabled = 0;
+
+ pi->caps_sclk_ds = true;
+
+ pi->mclk_strobe_mode_threshold = 40000;
+ pi->mclk_stutter_mode_threshold = 40000;
+ pi->mclk_edc_enable_threshold = 40000;
+ pi->mclk_edc_wr_enable_threshold = 40000;
+
+ ci_initialize_powertune_defaults(rdev);
+
+ pi->caps_fps = false;
+
+ pi->caps_sclk_throttle_low_notification = false;
+
+ ci_get_leakage_voltages(rdev);
+ ci_patch_dependency_tables_with_leakage(rdev);
+ ci_set_private_data_variables_based_on_pptable(rdev);
+
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries =
+ kzalloc(4 * sizeof(struct radeon_clock_voltage_dependency_entry), GFP_KERNEL);
+ if (!rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries) {
+ ci_dpm_fini(rdev);
+ return -ENOMEM;
+ }
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.count = 4;
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[0].clk = 0;
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[0].v = 0;
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[1].clk = 36000;
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[1].v = 720;
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[2].clk = 54000;
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[2].v = 810;
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[3].clk = 72000;
+ rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[3].v = 900;
+
+ rdev->pm.dpm.dyn_state.mclk_sclk_ratio = 4;
+ rdev->pm.dpm.dyn_state.sclk_mclk_delta = 15000;
+ rdev->pm.dpm.dyn_state.vddc_vddci_delta = 200;
+
+ rdev->pm.dpm.dyn_state.valid_sclk_values.count = 0;
+ rdev->pm.dpm.dyn_state.valid_sclk_values.values = NULL;
+ rdev->pm.dpm.dyn_state.valid_mclk_values.count = 0;
+ rdev->pm.dpm.dyn_state.valid_mclk_values.values = NULL;
+
+ pi->thermal_temp_setting.temperature_low = 99500;
+ pi->thermal_temp_setting.temperature_high = 100000;
+ pi->thermal_temp_setting.temperature_shutdown = 104000;
+
+ pi->uvd_enabled = false;
+
+ pi->voltage_control = CISLANDS_VOLTAGE_CONTROL_NONE;
+ pi->vddci_control = CISLANDS_VOLTAGE_CONTROL_NONE;
+ pi->mvdd_control = CISLANDS_VOLTAGE_CONTROL_NONE;
+ if (radeon_atom_is_voltage_gpio(rdev, VOLTAGE_TYPE_VDDC, VOLTAGE_OBJ_GPIO_LUT))
+ pi->voltage_control = CISLANDS_VOLTAGE_CONTROL_BY_GPIO;
+ else if (radeon_atom_is_voltage_gpio(rdev, VOLTAGE_TYPE_VDDC, VOLTAGE_OBJ_SVID2))
+ pi->voltage_control = CISLANDS_VOLTAGE_CONTROL_BY_SVID2;
+
+ if (rdev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_VDDCI_CONTROL) {
+ if (radeon_atom_is_voltage_gpio(rdev, VOLTAGE_TYPE_VDDCI, VOLTAGE_OBJ_GPIO_LUT))
+ pi->vddci_control = CISLANDS_VOLTAGE_CONTROL_BY_GPIO;
+ else if (radeon_atom_is_voltage_gpio(rdev, VOLTAGE_TYPE_VDDCI, VOLTAGE_OBJ_SVID2))
+ pi->vddci_control = CISLANDS_VOLTAGE_CONTROL_BY_SVID2;
+ else
+ rdev->pm.dpm.platform_caps &= ~ATOM_PP_PLATFORM_CAP_VDDCI_CONTROL;
+ }
+
+ if (rdev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_MVDDCONTROL) {
+ if (radeon_atom_is_voltage_gpio(rdev, VOLTAGE_TYPE_MVDDC, VOLTAGE_OBJ_GPIO_LUT))
+ pi->mvdd_control = CISLANDS_VOLTAGE_CONTROL_BY_GPIO;
+ else if (radeon_atom_is_voltage_gpio(rdev, VOLTAGE_TYPE_MVDDC, VOLTAGE_OBJ_SVID2))
+ pi->mvdd_control = CISLANDS_VOLTAGE_CONTROL_BY_SVID2;
+ else
+ rdev->pm.dpm.platform_caps &= ~ATOM_PP_PLATFORM_CAP_MVDDCONTROL;
+ }
+
+ pi->vddc_phase_shed_control = true;
+
+#if defined(CONFIG_ACPI)
+ pi->pcie_performance_request =
+ radeon_acpi_is_pcie_performance_request_supported(rdev);
+#else
+ pi->pcie_performance_request = false;
+#endif
+
+ if (atom_parse_data_header(rdev->mode_info.atom_context, index, &size,
+ &frev, &crev, &data_offset)) {
+ pi->caps_sclk_ss_support = true;
+ pi->caps_mclk_ss_support = true;
+ pi->dynamic_ss = true;
+ } else {
+ pi->caps_sclk_ss_support = false;
+ pi->caps_mclk_ss_support = false;
+ pi->dynamic_ss = true;
+ }
+
+ if (rdev->pm.int_thermal_type != THERMAL_TYPE_NONE)
+ pi->thermal_protection = true;
+ else
+ pi->thermal_protection = false;
+
+ pi->caps_dynamic_ac_timing = true;
+
+ return 0;
+}
+
+void ci_dpm_print_power_state(struct radeon_device *rdev,
+ struct radeon_ps *rps)
+{
+ struct ci_ps *ps = ci_get_ps(rps);
+ struct ci_pl *pl;
+ int i;
+
+ r600_dpm_print_class_info(rps->class, rps->class2);
+ r600_dpm_print_cap_info(rps->caps);
+ printk("\tuvd vclk: %d dclk: %d\n", rps->vclk, rps->dclk);
+ for (i = 0; i < ps->performance_level_count; i++) {
+ pl = &ps->performance_levels[i];
+ printk("\t\tpower level %d sclk: %u mclk: %u pcie gen: %u pcie lanes: %u\n",
+ i, pl->sclk, pl->mclk, pl->pcie_gen + 1, pl->pcie_lane);
+ }
+ r600_dpm_print_ps_status(rdev, rps);
+}
+
+u32 ci_dpm_get_sclk(struct radeon_device *rdev, bool low)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_ps *requested_state = ci_get_ps(&pi->requested_rps);
+
+ if (low)
+ return requested_state->performance_levels[0].sclk;
+ else
+ return requested_state->performance_levels[requested_state->performance_level_count - 1].sclk;
+}
+
+u32 ci_dpm_get_mclk(struct radeon_device *rdev, bool low)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ struct ci_ps *requested_state = ci_get_ps(&pi->requested_rps);
+
+ if (low)
+ return requested_state->performance_levels[0].mclk;
+ else
+ return requested_state->performance_levels[requested_state->performance_level_count - 1].mclk;
+}
diff --git a/drivers/gpu/drm/radeon/ci_dpm.h b/drivers/gpu/drm/radeon/ci_dpm.h
new file mode 100644
index 0000000..de504b5
--- /dev/null
+++ b/drivers/gpu/drm/radeon/ci_dpm.h
@@ -0,0 +1,331 @@
+/*
+ * Copyright 2013 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef __CI_DPM_H__
+#define __CI_DPM_H__
+
+#include "ppsmc.h"
+
+#define SMU__NUM_SCLK_DPM_STATE 8
+#define SMU__NUM_MCLK_DPM_LEVELS 6
+#define SMU__NUM_LCLK_DPM_LEVELS 8
+#define SMU__NUM_PCIE_DPM_LEVELS 8
+#include "smu7_discrete.h"
+
+#define CISLANDS_MAX_HARDWARE_POWERLEVELS 2
+
+struct ci_pl {
+ u32 mclk;
+ u32 sclk;
+ enum radeon_pcie_gen pcie_gen;
+ u16 pcie_lane;
+};
+
+struct ci_ps {
+ u16 performance_level_count;
+ bool dc_compatible;
+ u32 sclk_t;
+ struct ci_pl performance_levels[CISLANDS_MAX_HARDWARE_POWERLEVELS];
+};
+
+struct ci_dpm_level {
+ bool enabled;
+ u32 value;
+ u32 param1;
+};
+
+#define CISLAND_MAX_DEEPSLEEP_DIVIDER_ID 5
+#define MAX_REGULAR_DPM_NUMBER 8
+#define CISLAND_MINIMUM_ENGINE_CLOCK 800
+
+struct ci_single_dpm_table {
+ u32 count;
+ struct ci_dpm_level dpm_levels[MAX_REGULAR_DPM_NUMBER];
+};
+
+struct ci_dpm_table {
+ struct ci_single_dpm_table sclk_table;
+ struct ci_single_dpm_table mclk_table;
+ struct ci_single_dpm_table pcie_speed_table;
+ struct ci_single_dpm_table vddc_table;
+ struct ci_single_dpm_table vddci_table;
+ struct ci_single_dpm_table mvdd_table;
+};
+
+struct ci_mc_reg_entry {
+ u32 mclk_max;
+ u32 mc_data[SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE];
+};
+
+struct ci_mc_reg_table {
+ u8 last;
+ u8 num_entries;
+ u16 valid_flag;
+ struct ci_mc_reg_entry mc_reg_table_entry[MAX_AC_TIMING_ENTRIES];
+ SMU7_Discrete_MCRegisterAddress mc_reg_address[SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE];
+};
+
+struct ci_ulv_parm
+{
+ bool supported;
+ u32 cg_ulv_parameter;
+ u32 volt_change_delay;
+ struct ci_pl pl;
+};
+
+#define CISLANDS_MAX_LEAKAGE_COUNT 8
+
+struct ci_leakage_voltage {
+ u16 count;
+ u16 leakage_id[CISLANDS_MAX_LEAKAGE_COUNT];
+ u16 actual_voltage[CISLANDS_MAX_LEAKAGE_COUNT];
+};
+
+struct ci_dpm_level_enable_mask {
+ u32 uvd_dpm_enable_mask;
+ u32 vce_dpm_enable_mask;
+ u32 acp_dpm_enable_mask;
+ u32 samu_dpm_enable_mask;
+ u32 sclk_dpm_enable_mask;
+ u32 mclk_dpm_enable_mask;
+ u32 pcie_dpm_enable_mask;
+};
+
+struct ci_vbios_boot_state
+{
+ u16 mvdd_bootup_value;
+ u16 vddc_bootup_value;
+ u16 vddci_bootup_value;
+ u32 sclk_bootup_value;
+ u32 mclk_bootup_value;
+ u16 pcie_gen_bootup_value;
+ u16 pcie_lane_bootup_value;
+};
+
+struct ci_clock_registers {
+ u32 cg_spll_func_cntl;
+ u32 cg_spll_func_cntl_2;
+ u32 cg_spll_func_cntl_3;
+ u32 cg_spll_func_cntl_4;
+ u32 cg_spll_spread_spectrum;
+ u32 cg_spll_spread_spectrum_2;
+ u32 dll_cntl;
+ u32 mclk_pwrmgt_cntl;
+ u32 mpll_ad_func_cntl;
+ u32 mpll_dq_func_cntl;
+ u32 mpll_func_cntl;
+ u32 mpll_func_cntl_1;
+ u32 mpll_func_cntl_2;
+ u32 mpll_ss1;
+ u32 mpll_ss2;
+};
+
+struct ci_thermal_temperature_setting {
+ s32 temperature_low;
+ s32 temperature_high;
+ s32 temperature_shutdown;
+};
+
+struct ci_pcie_perf_range {
+ u16 max;
+ u16 min;
+};
+
+enum ci_pt_config_reg_type {
+ CISLANDS_CONFIGREG_MMR = 0,
+ CISLANDS_CONFIGREG_SMC_IND,
+ CISLANDS_CONFIGREG_DIDT_IND,
+ CISLANDS_CONFIGREG_CACHE,
+ CISLANDS_CONFIGREG_MAX
+};
+
+#define POWERCONTAINMENT_FEATURE_BAPM 0x00000001
+#define POWERCONTAINMENT_FEATURE_TDCLimit 0x00000002
+#define POWERCONTAINMENT_FEATURE_PkgPwrLimit 0x00000004
+
+struct ci_pt_config_reg {
+ u32 offset;
+ u32 mask;
+ u32 shift;
+ u32 value;
+ enum ci_pt_config_reg_type type;
+};
+
+struct ci_pt_defaults {
+ u8 svi_load_line_en;
+ u8 svi_load_line_vddc;
+ u8 tdc_vddc_throttle_release_limit_perc;
+ u8 tdc_mawt;
+ u8 tdc_waterfall_ctl;
+ u8 dte_ambient_temp_base;
+ u32 display_cac;
+ u32 bapm_temp_gradient;
+ u16 bapmti_r[SMU7_DTE_ITERATIONS * SMU7_DTE_SOURCES * SMU7_DTE_SINKS];
+ u16 bapmti_rc[SMU7_DTE_ITERATIONS * SMU7_DTE_SOURCES * SMU7_DTE_SINKS];
+};
+
+#define DPMTABLE_OD_UPDATE_SCLK 0x00000001
+#define DPMTABLE_OD_UPDATE_MCLK 0x00000002
+#define DPMTABLE_UPDATE_SCLK 0x00000004
+#define DPMTABLE_UPDATE_MCLK 0x00000008
+
+struct ci_power_info {
+ struct ci_dpm_table dpm_table;
+ u32 voltage_control;
+ u32 mvdd_control;
+ u32 vddci_control;
+ u32 active_auto_throttle_sources;
+ struct ci_clock_registers clock_registers;
+ u16 acpi_vddc;
+ u16 acpi_vddci;
+ enum radeon_pcie_gen force_pcie_gen;
+ enum radeon_pcie_gen acpi_pcie_gen;
+ struct ci_leakage_voltage vddc_leakage;
+ struct ci_leakage_voltage vddci_leakage;
+ u16 max_vddc_in_pp_table;
+ u16 min_vddc_in_pp_table;
+ u16 max_vddci_in_pp_table;
+ u16 min_vddci_in_pp_table;
+ u32 mclk_strobe_mode_threshold;
+ u32 mclk_stutter_mode_threshold;
+ u32 mclk_edc_enable_threshold;
+ u32 mclk_edc_wr_enable_threshold;
+ struct ci_vbios_boot_state vbios_boot_state;
+ /* smc offsets */
+ u32 sram_end;
+ u32 dpm_table_start;
+ u32 soft_regs_start;
+ u32 mc_reg_table_start;
+ u32 fan_table_start;
+ u32 arb_table_start;
+ /* smc tables */
+ SMU7_Discrete_DpmTable smc_state_table;
+ SMU7_Discrete_MCRegisters smc_mc_reg_table;
+ SMU7_Discrete_PmFuses smc_powertune_table;
+ /* other stuff */
+ struct ci_mc_reg_table mc_reg_table;
+ struct atom_voltage_table vddc_voltage_table;
+ struct atom_voltage_table vddci_voltage_table;
+ struct atom_voltage_table mvdd_voltage_table;
+ struct ci_ulv_parm ulv;
+ u32 power_containment_features;
+ const struct ci_pt_defaults *powertune_defaults;
+ u32 dte_tj_offset;
+ bool vddc_phase_shed_control;
+ struct ci_thermal_temperature_setting thermal_temp_setting;
+ struct ci_dpm_level_enable_mask dpm_level_enable_mask;
+ u32 need_update_smu7_dpm_table;
+ u32 sclk_dpm_key_disabled;
+ u32 mclk_dpm_key_disabled;
+ u32 pcie_dpm_key_disabled;
+ struct ci_pcie_perf_range pcie_gen_performance;
+ struct ci_pcie_perf_range pcie_lane_performance;
+ struct ci_pcie_perf_range pcie_gen_powersaving;
+ struct ci_pcie_perf_range pcie_lane_powersaving;
+ u32 activity_target[SMU7_MAX_LEVELS_GRAPHICS];
+ u32 mclk_activity_target;
+ u32 low_sclk_interrupt_t;
+ u32 last_mclk_dpm_enable_mask;
+ u32 sys_pcie_mask;
+ /* caps */
+ bool caps_power_containment;
+ bool caps_cac;
+ bool caps_sq_ramping;
+ bool caps_db_ramping;
+ bool caps_td_ramping;
+ bool caps_tcp_ramping;
+ bool caps_fps;
+ bool caps_sclk_ds;
+ bool caps_sclk_ss_support;
+ bool caps_mclk_ss_support;
+ bool caps_uvd_dpm;
+ bool caps_vce_dpm;
+ bool caps_samu_dpm;
+ bool caps_acp_dpm;
+ bool caps_automatic_dc_transition;
+ bool caps_sclk_throttle_low_notification;
+ bool caps_dynamic_ac_timing;
+ /* flags */
+ bool thermal_protection;
+ bool pcie_performance_request;
+ bool dynamic_ss;
+ bool dll_default_on;
+ bool cac_enabled;
+ bool uvd_enabled;
+ bool battery_state;
+ bool pspp_notify_required;
+ bool mem_gddr5;
+ bool enable_bapm_feature;
+ bool enable_tdc_limit_feature;
+ bool enable_pkg_pwr_tracking_feature;
+ bool use_pcie_performance_levels;
+ bool use_pcie_powersaving_levels;
+ /* driver states */
+ struct radeon_ps current_rps;
+ struct ci_ps current_ps;
+ struct radeon_ps requested_rps;
+ struct ci_ps requested_ps;
+};
+
+#define CISLANDS_VOLTAGE_CONTROL_NONE 0x0
+#define CISLANDS_VOLTAGE_CONTROL_BY_GPIO 0x1
+#define CISLANDS_VOLTAGE_CONTROL_BY_SVID2 0x2
+
+#define CISLANDS_Q88_FORMAT_CONVERSION_UNIT 256
+
+#define CISLANDS_VRC_DFLT0 0x3FFFC000
+#define CISLANDS_VRC_DFLT1 0x000400
+#define CISLANDS_VRC_DFLT2 0xC00080
+#define CISLANDS_VRC_DFLT3 0xC00200
+#define CISLANDS_VRC_DFLT4 0xC01680
+#define CISLANDS_VRC_DFLT5 0xC00033
+#define CISLANDS_VRC_DFLT6 0xC00033
+#define CISLANDS_VRC_DFLT7 0x3FFFC000
+
+#define CISLANDS_CGULVPARAMETER_DFLT 0x00040035
+#define CISLAND_TARGETACTIVITY_DFLT 30
+#define CISLAND_MCLK_TARGETACTIVITY_DFLT 10
+
+#define PCIE_PERF_REQ_REMOVE_REGISTRY 0
+#define PCIE_PERF_REQ_FORCE_LOWPOWER 1
+#define PCIE_PERF_REQ_PECI_GEN1 2
+#define PCIE_PERF_REQ_PECI_GEN2 3
+#define PCIE_PERF_REQ_PECI_GEN3 4
+
+int ci_copy_bytes_to_smc(struct radeon_device *rdev,
+ u32 smc_start_address,
+ const u8 *src, u32 byte_count, u32 limit);
+void ci_start_smc(struct radeon_device *rdev);
+void ci_reset_smc(struct radeon_device *rdev);
+int ci_program_jump_on_start(struct radeon_device *rdev);
+void ci_stop_smc_clock(struct radeon_device *rdev);
+void ci_start_smc_clock(struct radeon_device *rdev);
+bool ci_is_smc_running(struct radeon_device *rdev);
+PPSMC_Result ci_send_msg_to_smc(struct radeon_device *rdev, PPSMC_Msg msg);
+PPSMC_Result ci_wait_for_smc_inactive(struct radeon_device *rdev);
+int ci_load_smc_ucode(struct radeon_device *rdev, u32 limit);
+int ci_read_smc_sram_dword(struct radeon_device *rdev,
+ u32 smc_address, u32 *value, u32 limit);
+int ci_write_smc_sram_dword(struct radeon_device *rdev,
+ u32 smc_address, u32 value, u32 limit);
+
+#endif
diff --git a/drivers/gpu/drm/radeon/ci_smc.c b/drivers/gpu/drm/radeon/ci_smc.c
new file mode 100644
index 0000000..53b43dd
--- /dev/null
+++ b/drivers/gpu/drm/radeon/ci_smc.c
@@ -0,0 +1,262 @@
+/*
+ * Copyright 2011 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Alex Deucher
+ */
+
+#include <linux/firmware.h>
+#include "drmP.h"
+#include "radeon.h"
+#include "cikd.h"
+#include "ppsmc.h"
+#include "radeon_ucode.h"
+
+static int ci_set_smc_sram_address(struct radeon_device *rdev,
+ u32 smc_address, u32 limit)
+{
+ if (smc_address & 3)
+ return -EINVAL;
+ if ((smc_address + 3) > limit)
+ return -EINVAL;
+
+ WREG32(SMC_IND_INDEX_0, smc_address);
+ WREG32_P(SMC_IND_ACCESS_CNTL, 0, ~AUTO_INCREMENT_IND_0);
+
+ return 0;
+}
+
+int ci_copy_bytes_to_smc(struct radeon_device *rdev,
+ u32 smc_start_address,
+ const u8 *src, u32 byte_count, u32 limit)
+{
+ u32 data, original_data;
+ u32 addr;
+ u32 extra_shift;
+ int ret;
+
+ if (smc_start_address & 3)
+ return -EINVAL;
+ if ((smc_start_address + byte_count) > limit)
+ return -EINVAL;
+
+ addr = smc_start_address;
+
+ while (byte_count >= 4) {
+ /* SMC address space is BE */
+ data = (src[0] << 24) | (src[1] << 16) | (src[2] << 8) | src[3];
+
+ ret = ci_set_smc_sram_address(rdev, addr, limit);
+ if (ret)
+ return ret;
+
+ WREG32(SMC_IND_DATA_0, data);
+
+ src += 4;
+ byte_count -= 4;
+ addr += 4;
+ }
+
+ /* RMW for the final bytes */
+ if (byte_count > 0) {
+ data = 0;
+
+ ret = ci_set_smc_sram_address(rdev, addr, limit);
+ if (ret)
+ return ret;
+
+ original_data = RREG32(SMC_IND_DATA_0);
+
+ extra_shift = 8 * (4 - byte_count);
+
+ while (byte_count > 0) {
+ data = (data << 8) + *src++;
+ byte_count--;
+ }
+
+ data <<= extra_shift;
+
+ data |= (original_data & ~((~0UL) << extra_shift));
+
+ ret = ci_set_smc_sram_address(rdev, addr, limit);
+ if (ret)
+ return ret;
+
+ WREG32(SMC_IND_DATA_0, data);
+ }
+ return 0;
+}
+
+void ci_start_smc(struct radeon_device *rdev)
+{
+ u32 tmp = RREG32_SMC(SMC_SYSCON_RESET_CNTL);
+
+ tmp &= ~RST_REG;
+ WREG32_SMC(SMC_SYSCON_RESET_CNTL, tmp);
+}
+
+void ci_reset_smc(struct radeon_device *rdev)
+{
+ u32 tmp = RREG32_SMC(SMC_SYSCON_RESET_CNTL);
+
+ tmp |= RST_REG;
+ WREG32_SMC(SMC_SYSCON_RESET_CNTL, tmp);
+}
+
+int ci_program_jump_on_start(struct radeon_device *rdev)
+{
+ static u8 data[] = { 0xE0, 0x00, 0x80, 0x40 };
+
+ return ci_copy_bytes_to_smc(rdev, 0x0, data, 4, sizeof(data)+1);
+}
+
+void ci_stop_smc_clock(struct radeon_device *rdev)
+{
+ u32 tmp = RREG32_SMC(SMC_SYSCON_CLOCK_CNTL_0);
+
+ tmp |= CK_DISABLE;
+
+ WREG32_SMC(SMC_SYSCON_CLOCK_CNTL_0, tmp);
+}
+
+void ci_start_smc_clock(struct radeon_device *rdev)
+{
+ u32 tmp = RREG32_SMC(SMC_SYSCON_CLOCK_CNTL_0);
+
+ tmp &= ~CK_DISABLE;
+
+ WREG32_SMC(SMC_SYSCON_CLOCK_CNTL_0, tmp);
+}
+
+bool ci_is_smc_running(struct radeon_device *rdev)
+{
+ u32 clk = RREG32_SMC(SMC_SYSCON_CLOCK_CNTL_0);
+ u32 pc_c = RREG32_SMC(SMC_PC_C);
+
+ if (!(clk & CK_DISABLE) && (0x20100 <= pc_c))
+ return true;
+
+ return false;
+}
+
+PPSMC_Result ci_send_msg_to_smc(struct radeon_device *rdev, PPSMC_Msg msg)
+{
+ u32 tmp;
+ int i;
+
+ if (!ci_is_smc_running(rdev))
+ return PPSMC_Result_Failed;
+
+ WREG32(SMC_MESSAGE_0, msg);
+
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ tmp = RREG32(SMC_RESP_0);
+ if (tmp != 0)
+ break;
+ udelay(1);
+ }
+ tmp = RREG32(SMC_RESP_0);
+
+ return (PPSMC_Result)tmp;
+}
+
+PPSMC_Result ci_wait_for_smc_inactive(struct radeon_device *rdev)
+{
+ u32 tmp;
+ int i;
+
+ if (!ci_is_smc_running(rdev))
+ return PPSMC_Result_OK;
+
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ tmp = RREG32_SMC(SMC_SYSCON_CLOCK_CNTL_0);
+ if ((tmp & CKEN) == 0)
+ break;
+ udelay(1);
+ }
+
+ return PPSMC_Result_OK;
+}
+
+int ci_load_smc_ucode(struct radeon_device *rdev, u32 limit)
+{
+ u32 ucode_start_address;
+ u32 ucode_size;
+ const u8 *src;
+ u32 data;
+
+ if (!rdev->smc_fw)
+ return -EINVAL;
+
+ switch (rdev->family) {
+ case CHIP_BONAIRE:
+ ucode_start_address = BONAIRE_SMC_UCODE_START;
+ ucode_size = BONAIRE_SMC_UCODE_SIZE;
+ break;
+ default:
+ DRM_ERROR("unknown asic in smc ucode loader\n");
+ BUG();
+ }
+
+ if (ucode_size & 3)
+ return -EINVAL;
+
+ src = (const u8 *)rdev->smc_fw->data;
+ WREG32(SMC_IND_INDEX_0, ucode_start_address);
+ WREG32_P(SMC_IND_ACCESS_CNTL, AUTO_INCREMENT_IND_0, ~AUTO_INCREMENT_IND_0);
+ while (ucode_size >= 4) {
+ /* SMC address space is BE */
+ data = (src[0] << 24) | (src[1] << 16) | (src[2] << 8) | src[3];
+
+ WREG32(SMC_IND_DATA_0, data);
+
+ src += 4;
+ ucode_size -= 4;
+ }
+ WREG32_P(SMC_IND_ACCESS_CNTL, 0, ~AUTO_INCREMENT_IND_0);
+
+ return 0;
+}
+
+int ci_read_smc_sram_dword(struct radeon_device *rdev,
+ u32 smc_address, u32 *value, u32 limit)
+{
+ int ret;
+
+ ret = ci_set_smc_sram_address(rdev, smc_address, limit);
+ if (ret)
+ return ret;
+
+ *value = RREG32(SMC_IND_DATA_0);
+ return 0;
+}
+
+int ci_write_smc_sram_dword(struct radeon_device *rdev,
+ u32 smc_address, u32 value, u32 limit)
+{
+ int ret;
+
+ ret = ci_set_smc_sram_address(rdev, smc_address, limit);
+ if (ret)
+ return ret;
+
+ WREG32(SMC_IND_DATA_0, value);
+ return 0;
+}
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 87e5aee..736a416 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -40,6 +40,7 @@ MODULE_FIRMWARE("radeon/BONAIRE_mec.bin");
MODULE_FIRMWARE("radeon/BONAIRE_mc.bin");
MODULE_FIRMWARE("radeon/BONAIRE_rlc.bin");
MODULE_FIRMWARE("radeon/BONAIRE_sdma.bin");
+MODULE_FIRMWARE("radeon/BONAIRE_smc.bin");
MODULE_FIRMWARE("radeon/KAVERI_pfp.bin");
MODULE_FIRMWARE("radeon/KAVERI_me.bin");
MODULE_FIRMWARE("radeon/KAVERI_ce.bin");
@@ -1545,7 +1546,7 @@ static int cik_init_microcode(struct radeon_device *rdev)
const char *chip_name;
size_t pfp_req_size, me_req_size, ce_req_size,
mec_req_size, rlc_req_size, mc_req_size,
- sdma_req_size;
+ sdma_req_size, smc_req_size;
char fw_name[30];
int err;
@@ -1561,6 +1562,7 @@ static int cik_init_microcode(struct radeon_device *rdev)
rlc_req_size = BONAIRE_RLC_UCODE_SIZE * 4;
mc_req_size = CIK_MC_UCODE_SIZE * 4;
sdma_req_size = CIK_SDMA_UCODE_SIZE * 4;
+ smc_req_size = ALIGN(BONAIRE_SMC_UCODE_SIZE, 4);
break;
case CHIP_KAVERI:
chip_name = "KAVERI";
@@ -1652,7 +1654,7 @@ static int cik_init_microcode(struct radeon_device *rdev)
err = -EINVAL;
}
- /* No MC ucode on APUs */
+ /* No SMC, MC ucode on APUs */
if (!(rdev->flags & RADEON_IS_IGP)) {
snprintf(fw_name, sizeof(fw_name), "radeon/%s_mc.bin", chip_name);
err = request_firmware(&rdev->mc_fw, fw_name, rdev->dev);
@@ -1664,6 +1666,21 @@ static int cik_init_microcode(struct radeon_device *rdev)
rdev->mc_fw->size, fw_name);
err = -EINVAL;
}
+
+ snprintf(fw_name, sizeof(fw_name), "radeon/%s_smc.bin", chip_name);
+ err = request_firmware(&rdev->smc_fw, fw_name, rdev->dev);
+ if (err) {
+ printk(KERN_ERR
+ "smc: error loading firmware \"%s\"\n",
+ fw_name);
+ release_firmware(rdev->smc_fw);
+ rdev->smc_fw = NULL;
+ } else if (rdev->smc_fw->size != smc_req_size) {
+ printk(KERN_ERR
+ "cik_smc: Bogus length %zu in firmware \"%s\"\n",
+ rdev->smc_fw->size, fw_name);
+ err = -EINVAL;
+ }
}
out:
@@ -1682,6 +1699,8 @@ out:
rdev->rlc_fw = NULL;
release_firmware(rdev->mc_fw);
rdev->mc_fw = NULL;
+ release_firmware(rdev->smc_fw);
+ rdev->smc_fw = NULL;
}
return err;
}
@@ -6626,8 +6645,12 @@ int cik_irq_set(struct radeon_device *rdev)
cp_m2p2 = RREG32(CP_ME2_PIPE2_INT_CNTL) & ~TIME_STAMP_INT_ENABLE;
cp_m2p3 = RREG32(CP_ME2_PIPE3_INT_CNTL) & ~TIME_STAMP_INT_ENABLE;
- thermal_int = RREG32_SMC(CG_THERMAL_INT_CTRL) &
- ~(THERM_INTH_MASK | THERM_INTL_MASK);
+ if (rdev->flags & RADEON_IS_IGP)
+ thermal_int = RREG32_SMC(CG_THERMAL_INT_CTRL) &
+ ~(THERM_INTH_MASK | THERM_INTL_MASK);
+ else
+ thermal_int = RREG32_SMC(CG_THERMAL_INT) &
+ ~(THERM_INT_MASK_HIGH | THERM_INT_MASK_LOW);
/* enable CP interrupts on all rings */
if (atomic_read(&rdev->irq.ring_int[RADEON_RING_TYPE_GFX_INDEX])) {
@@ -6788,7 +6811,10 @@ int cik_irq_set(struct radeon_device *rdev)
if (rdev->irq.dpm_thermal) {
DRM_DEBUG("dpm thermal\n");
- thermal_int |= THERM_INTH_MASK | THERM_INTL_MASK;
+ if (rdev->flags & RADEON_IS_IGP)
+ thermal_int |= THERM_INTH_MASK | THERM_INTL_MASK;
+ else
+ thermal_int |= THERM_INT_MASK_HIGH | THERM_INT_MASK_LOW;
}
WREG32(CP_INT_CNTL_RING0, cp_int_cntl);
@@ -6825,7 +6851,10 @@ int cik_irq_set(struct radeon_device *rdev)
WREG32(DC_HPD5_INT_CONTROL, hpd5);
WREG32(DC_HPD6_INT_CONTROL, hpd6);
- WREG32_SMC(CG_THERMAL_INT_CTRL, thermal_int);
+ if (rdev->flags & RADEON_IS_IGP)
+ WREG32_SMC(CG_THERMAL_INT_CTRL, thermal_int);
+ else
+ WREG32_SMC(CG_THERMAL_INT, thermal_int);
return 0;
}
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index f4c417d..52cbf6f 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -36,6 +36,23 @@
#define DIDT_TCP_CTRL0 0x60
/* SMC IND registers */
+#define DPM_TABLE_475 0x3F768
+# define SamuBootLevel(x) ((x) << 0)
+# define SamuBootLevel_MASK 0x000000ff
+# define SamuBootLevel_SHIFT 0
+# define AcpBootLevel(x) ((x) << 8)
+# define AcpBootLevel_MASK 0x0000ff00
+# define AcpBootLevel_SHIFT 8
+# define VceBootLevel(x) ((x) << 16)
+# define VceBootLevel_MASK 0x00ff0000
+# define VceBootLevel_SHIFT 16
+# define UvdBootLevel(x) ((x) << 24)
+# define UvdBootLevel_MASK 0xff000000
+# define UvdBootLevel_SHIFT 24
+
+#define FIRMWARE_FLAGS 0x3F800
+# define INTERRUPTS_ENABLED (1 << 0)
+
#define NB_DPM_CONFIG_1 0x3F9E8
# define Dpm0PgNbPsLo(x) ((x) << 0)
# define Dpm0PgNbPsLo_MASK 0x000000ff
@@ -50,25 +67,85 @@
# define DpmXNbPsHi_MASK 0xff000000
# define DpmXNbPsHi_SHIFT 24
+#define SMC_SYSCON_RESET_CNTL 0x80000000
+# define RST_REG (1 << 0)
+#define SMC_SYSCON_CLOCK_CNTL_0 0x80000004
+# define CK_DISABLE (1 << 0)
+# define CKEN (1 << 24)
+
+#define SMC_SYSCON_MISC_CNTL 0x80000010
+
#define SMC_SYSCON_MSG_ARG_0 0x80000068
+#define SMC_PC_C 0x80000370
+
+#define SMC_SCRATCH9 0x80000424
+
+#define RCU_UC_EVENTS 0xC0000004
+# define BOOT_SEQ_DONE (1 << 7)
+
#define GENERAL_PWRMGT 0xC0200000
# define GLOBAL_PWRMGT_EN (1 << 0)
+# define STATIC_PM_EN (1 << 1)
+# define THERMAL_PROTECTION_DIS (1 << 2)
+# define THERMAL_PROTECTION_TYPE (1 << 3)
+# define SW_SMIO_INDEX(x) ((x) << 6)
+# define SW_SMIO_INDEX_MASK (1 << 6)
+# define SW_SMIO_INDEX_SHIFT 6
+# define VOLT_PWRMGT_EN (1 << 10)
# define GPU_COUNTER_CLK (1 << 15)
+# define DYN_SPREAD_SPECTRUM_EN (1 << 23)
+
+#define CNB_PWRMGT_CNTL 0xC0200004
+# define GNB_SLOW_MODE(x) ((x) << 0)
+# define GNB_SLOW_MODE_MASK (3 << 0)
+# define GNB_SLOW_MODE_SHIFT 0
+# define GNB_SLOW (1 << 2)
+# define FORCE_NB_PS1 (1 << 3)
+# define DPM_ENABLED (1 << 4)
#define SCLK_PWRMGT_CNTL 0xC0200008
+# define SCLK_PWRMGT_OFF (1 << 0)
# define RESET_BUSY_CNT (1 << 4)
# define RESET_SCLK_CNT (1 << 5)
# define DYNAMIC_PM_EN (1 << 21)
+#define CG_SSP 0xC0200044
+# define SST(x) ((x) << 0)
+# define SST_MASK (0xffff << 0)
+# define SSTU(x) ((x) << 16)
+# define SSTU_MASK (0xf << 16)
+
+#define CG_DISPLAY_GAP_CNTL 0xC0200060
+# define DISP_GAP(x) ((x) << 0)
+# define DISP_GAP_MASK (3 << 0)
+# define VBI_TIMER_COUNT(x) ((x) << 4)
+# define VBI_TIMER_COUNT_MASK (0x3fff << 4)
+# define VBI_TIMER_UNIT(x) ((x) << 20)
+# define VBI_TIMER_UNIT_MASK (7 << 20)
+# define DISP_GAP_MCHG(x) ((x) << 24)
+# define DISP_GAP_MCHG_MASK (3 << 24)
+
+#define CG_ULV_PARAMETER 0xC0200158
+
#define CG_FTV_0 0xC02001A8
+#define CG_FTV_1 0xC02001AC
+#define CG_FTV_2 0xC02001B0
+#define CG_FTV_3 0xC02001B4
+#define CG_FTV_4 0xC02001B8
+#define CG_FTV_5 0xC02001BC
+#define CG_FTV_6 0xC02001C0
+#define CG_FTV_7 0xC02001C4
+
+#define CG_DISPLAY_GAP_CNTL2 0xC0200230
#define LCAC_SX0_OVR_SEL 0xC0400D04
#define LCAC_SX0_OVR_VAL 0xC0400D08
+#define LCAC_MC0_CNTL 0xC0400D30
#define LCAC_MC0_OVR_SEL 0xC0400D34
#define LCAC_MC0_OVR_VAL 0xC0400D38
-
+#define LCAC_MC1_CNTL 0xC0400D3C
#define LCAC_MC1_OVR_SEL 0xC0400D40
#define LCAC_MC1_OVR_VAL 0xC0400D44
@@ -78,9 +155,28 @@
#define LCAC_MC3_OVR_SEL 0xC0400D58
#define LCAC_MC3_OVR_VAL 0xC0400D5C
+#define LCAC_CPL_CNTL 0xC0400D80
#define LCAC_CPL_OVR_SEL 0xC0400D84
#define LCAC_CPL_OVR_VAL 0xC0400D88
+/* dGPU */
+#define CG_THERMAL_CTRL 0xC0300004
+#define DPM_EVENT_SRC(x) ((x) << 0)
+#define DPM_EVENT_SRC_MASK (7 << 0)
+#define DIG_THERM_DPM(x) ((x) << 14)
+#define DIG_THERM_DPM_MASK 0x003FC000
+#define DIG_THERM_DPM_SHIFT 14
+
+#define CG_THERMAL_INT 0xC030000C
+#define CI_DIG_THERM_INTH(x) ((x) << 8)
+#define CI_DIG_THERM_INTH_MASK 0x0000FF00
+#define CI_DIG_THERM_INTH_SHIFT 8
+#define CI_DIG_THERM_INTL(x) ((x) << 16)
+#define CI_DIG_THERM_INTL_MASK 0x00FF0000
+#define CI_DIG_THERM_INTL_SHIFT 16
+#define THERM_INT_MASK_HIGH (1 << 24)
+#define THERM_INT_MASK_LOW (1 << 25)
+
#define CG_MULT_THERMAL_STATUS 0xC0300010
#define ASIC_MAX_TEMP(x) ((x) << 0)
#define ASIC_MAX_TEMP_MASK 0x000001ff
@@ -89,6 +185,35 @@
#define CTF_TEMP_MASK 0x0003fe00
#define CTF_TEMP_SHIFT 9
+#define CG_SPLL_FUNC_CNTL 0xC0500140
+#define SPLL_RESET (1 << 0)
+#define SPLL_PWRON (1 << 1)
+#define SPLL_BYPASS_EN (1 << 3)
+#define SPLL_REF_DIV(x) ((x) << 5)
+#define SPLL_REF_DIV_MASK (0x3f << 5)
+#define SPLL_PDIV_A(x) ((x) << 20)
+#define SPLL_PDIV_A_MASK (0x7f << 20)
+#define SPLL_PDIV_A_SHIFT 20
+#define CG_SPLL_FUNC_CNTL_2 0xC0500144
+#define SCLK_MUX_SEL(x) ((x) << 0)
+#define SCLK_MUX_SEL_MASK (0x1ff << 0)
+#define CG_SPLL_FUNC_CNTL_3 0xC0500148
+#define SPLL_FB_DIV(x) ((x) << 0)
+#define SPLL_FB_DIV_MASK (0x3ffffff << 0)
+#define SPLL_FB_DIV_SHIFT 0
+#define SPLL_DITHEN (1 << 28)
+#define CG_SPLL_FUNC_CNTL_4 0xC050014C
+
+#define CG_SPLL_SPREAD_SPECTRUM 0xC0500164
+#define SSEN (1 << 0)
+#define CLK_S(x) ((x) << 4)
+#define CLK_S_MASK (0xfff << 4)
+#define CLK_S_SHIFT 4
+#define CG_SPLL_SPREAD_SPECTRUM_2 0xC0500168
+#define CLK_V(x) ((x) << 0)
+#define CLK_V_MASK (0x3ffffff << 0)
+#define CLK_V_SHIFT 0
+
#define MPLL_BYPASSCLK_SEL 0xC050019C
# define MPLL_CLKOUT_SEL(x) ((x) << 8)
# define MPLL_CLKOUT_SEL_MASK 0xFF00
@@ -109,6 +234,7 @@
# define ZCLK_SEL(x) ((x) << 8)
# define ZCLK_SEL_MASK 0xFF00
+/* KV/KB */
#define CG_THERMAL_INT_CTRL 0xC2100028
#define DIG_THERM_INTH(x) ((x) << 0)
#define DIG_THERM_INTH_MASK 0x000000FF
@@ -437,9 +563,37 @@
#define NOOFGROUPS_SHIFT 12
#define NOOFGROUPS_MASK 0x00001000
+#define MC_ARB_DRAM_TIMING 0x2774
+#define MC_ARB_DRAM_TIMING2 0x2778
+
+#define MC_ARB_BURST_TIME 0x2808
+#define STATE0(x) ((x) << 0)
+#define STATE0_MASK (0x1f << 0)
+#define STATE0_SHIFT 0
+#define STATE1(x) ((x) << 5)
+#define STATE1_MASK (0x1f << 5)
+#define STATE1_SHIFT 5
+#define STATE2(x) ((x) << 10)
+#define STATE2_MASK (0x1f << 10)
+#define STATE2_SHIFT 10
+#define STATE3(x) ((x) << 15)
+#define STATE3_MASK (0x1f << 15)
+#define STATE3_SHIFT 15
+
+#define MC_SEQ_RAS_TIMING 0x28a0
+#define MC_SEQ_CAS_TIMING 0x28a4
+#define MC_SEQ_MISC_TIMING 0x28a8
+#define MC_SEQ_MISC_TIMING2 0x28ac
+#define MC_SEQ_PMG_TIMING 0x28b0
+#define MC_SEQ_RD_CTL_D0 0x28b4
+#define MC_SEQ_RD_CTL_D1 0x28b8
+#define MC_SEQ_WR_CTL_D0 0x28bc
+#define MC_SEQ_WR_CTL_D1 0x28c0
+
#define MC_SEQ_SUP_CNTL 0x28c8
#define RUN_MASK (1 << 0)
#define MC_SEQ_SUP_PGM 0x28cc
+#define MC_PMG_AUTO_CMD 0x28d0
#define MC_SEQ_TRAIN_WAKEUP_CNTL 0x28e8
#define TRAIN_DONE_D0 (1 << 30)
@@ -448,9 +602,90 @@
#define MC_IO_PAD_CNTL_D0 0x29d0
#define MEM_FALL_OUT_CMD (1 << 8)
+#define MC_SEQ_MISC0 0x2a00
+#define MC_SEQ_MISC0_VEN_ID_SHIFT 8
+#define MC_SEQ_MISC0_VEN_ID_MASK 0x00000f00
+#define MC_SEQ_MISC0_VEN_ID_VALUE 3
+#define MC_SEQ_MISC0_REV_ID_SHIFT 12
+#define MC_SEQ_MISC0_REV_ID_MASK 0x0000f000
+#define MC_SEQ_MISC0_REV_ID_VALUE 1
+#define MC_SEQ_MISC0_GDDR5_SHIFT 28
+#define MC_SEQ_MISC0_GDDR5_MASK 0xf0000000
+#define MC_SEQ_MISC0_GDDR5_VALUE 5
+#define MC_SEQ_MISC1 0x2a04
+#define MC_SEQ_RESERVE_M 0x2a08
+#define MC_PMG_CMD_EMRS 0x2a0c
+
#define MC_SEQ_IO_DEBUG_INDEX 0x2a44
#define MC_SEQ_IO_DEBUG_DATA 0x2a48
+#define MC_SEQ_MISC5 0x2a54
+#define MC_SEQ_MISC6 0x2a58
+
+#define MC_SEQ_MISC7 0x2a64
+
+#define MC_SEQ_RAS_TIMING_LP 0x2a6c
+#define MC_SEQ_CAS_TIMING_LP 0x2a70
+#define MC_SEQ_MISC_TIMING_LP 0x2a74
+#define MC_SEQ_MISC_TIMING2_LP 0x2a78
+#define MC_SEQ_WR_CTL_D0_LP 0x2a7c
+#define MC_SEQ_WR_CTL_D1_LP 0x2a80
+#define MC_SEQ_PMG_CMD_EMRS_LP 0x2a84
+#define MC_SEQ_PMG_CMD_MRS_LP 0x2a88
+
+#define MC_PMG_CMD_MRS 0x2aac
+
+#define MC_SEQ_RD_CTL_D0_LP 0x2b1c
+#define MC_SEQ_RD_CTL_D1_LP 0x2b20
+
+#define MC_PMG_CMD_MRS1 0x2b44
+#define MC_SEQ_PMG_CMD_MRS1_LP 0x2b48
+#define MC_SEQ_PMG_TIMING_LP 0x2b4c
+
+#define MC_SEQ_WR_CTL_2 0x2b54
+#define MC_SEQ_WR_CTL_2_LP 0x2b58
+#define MC_PMG_CMD_MRS2 0x2b5c
+#define MC_SEQ_PMG_CMD_MRS2_LP 0x2b60
+
+#define MCLK_PWRMGT_CNTL 0x2ba0
+# define DLL_SPEED(x) ((x) << 0)
+# define DLL_SPEED_MASK (0x1f << 0)
+# define DLL_READY (1 << 6)
+# define MC_INT_CNTL (1 << 7)
+# define MRDCK0_PDNB (1 << 8)
+# define MRDCK1_PDNB (1 << 9)
+# define MRDCK0_RESET (1 << 16)
+# define MRDCK1_RESET (1 << 17)
+# define DLL_READY_READ (1 << 24)
+#define DLL_CNTL 0x2ba4
+# define MRDCK0_BYPASS (1 << 24)
+# define MRDCK1_BYPASS (1 << 25)
+
+#define MPLL_FUNC_CNTL 0x2bb4
+#define BWCTRL(x) ((x) << 20)
+#define BWCTRL_MASK (0xff << 20)
+#define MPLL_FUNC_CNTL_1 0x2bb8
+#define VCO_MODE(x) ((x) << 0)
+#define VCO_MODE_MASK (3 << 0)
+#define CLKFRAC(x) ((x) << 4)
+#define CLKFRAC_MASK (0xfff << 4)
+#define CLKF(x) ((x) << 16)
+#define CLKF_MASK (0xfff << 16)
+#define MPLL_FUNC_CNTL_2 0x2bbc
+#define MPLL_AD_FUNC_CNTL 0x2bc0
+#define YCLK_POST_DIV(x) ((x) << 0)
+#define YCLK_POST_DIV_MASK (7 << 0)
+#define MPLL_DQ_FUNC_CNTL 0x2bc4
+#define YCLK_SEL(x) ((x) << 4)
+#define YCLK_SEL_MASK (1 << 4)
+
+#define MPLL_SS1 0x2bcc
+#define CLKV(x) ((x) << 0)
+#define CLKV_MASK (0x3ffffff << 0)
+#define MPLL_SS2 0x2bd0
+#define CLKS(x) ((x) << 0)
+#define CLKS_MASK (0xfff << 0)
+
#define HDP_HOST_PATH_CNTL 0x2C00
#define CLOCK_GATING_DIS (1 << 23)
#define HDP_NONSURFACE_BASE 0x2C04
@@ -465,6 +700,22 @@
#define ATC_MISC_CG 0x3350
+#define MC_SEQ_CNTL_3 0x3600
+# define CAC_EN (1 << 31)
+#define MC_SEQ_G5PDX_CTRL 0x3604
+#define MC_SEQ_G5PDX_CTRL_LP 0x3608
+#define MC_SEQ_G5PDX_CMD0 0x360c
+#define MC_SEQ_G5PDX_CMD0_LP 0x3610
+#define MC_SEQ_G5PDX_CMD1 0x3614
+#define MC_SEQ_G5PDX_CMD1_LP 0x3618
+
+#define MC_SEQ_PMG_DVS_CTL 0x3628
+#define MC_SEQ_PMG_DVS_CTL_LP 0x362c
+#define MC_SEQ_PMG_DVS_CMD 0x3630
+#define MC_SEQ_PMG_DVS_CMD_LP 0x3634
+#define MC_SEQ_DLL_STBY 0x3638
+#define MC_SEQ_DLL_STBY_LP 0x363c
+
#define IH_RB_CNTL 0x3e00
# define IH_RB_ENABLE (1 << 0)
# define IH_RB_SIZE(x) ((x) << 1) /* log2 */
@@ -492,6 +743,9 @@
# define MC_WR_CLEAN_CNT(x) ((x) << 20)
# define MC_VMID(x) ((x) << 25)
+#define BIF_LNCNT_RESET 0x5220
+# define RESET_LNCNT_EN (1 << 0)
+
#define CONFIG_MEMSIZE 0x5428
#define INTERRUPT_CNTL 0x5468
@@ -628,6 +882,9 @@
# define DC_HPDx_RX_INT_TIMER(x) ((x) << 16)
# define DC_HPDx_EN (1 << 28)
+#define DPG_PIPE_STUTTER_CONTROL 0x6cd4
+# define STUTTER_ENABLE (1 << 0)
+
#define GRBM_CNTL 0x8000
#define GRBM_READ_TIMEOUT(x) ((x) << 0)
diff --git a/drivers/gpu/drm/radeon/ppsmc.h b/drivers/gpu/drm/radeon/ppsmc.h
index 6db6e32..4c1ee6d 100644
--- a/drivers/gpu/drm/radeon/ppsmc.h
+++ b/drivers/gpu/drm/radeon/ppsmc.h
@@ -99,7 +99,7 @@ typedef uint8_t PPSMC_Result;
#define PPSMC_MSG_ThrottleOVRDSCLKDS ((uint8_t)0x96)
#define PPSMC_MSG_CancelThrottleOVRDSCLKDS ((uint8_t)0x97)
-/* KV/KB */
+/* CI/KV/KB */
#define PPSMC_MSG_UVDDPM_SetEnabledMask ((uint16_t) 0x12D)
#define PPSMC_MSG_VCEDPM_SetEnabledMask ((uint16_t) 0x12E)
#define PPSMC_MSG_ACPDPM_SetEnabledMask ((uint16_t) 0x12F)
@@ -108,6 +108,7 @@ typedef uint8_t PPSMC_Result;
#define PPSMC_MSG_MCLKDPM_NoForcedLevel ((uint16_t) 0x132)
#define PPSMC_MSG_Voltage_Cntl_Disable ((uint16_t) 0x135)
#define PPSMC_MSG_PCIeDPM_Enable ((uint16_t) 0x136)
+#define PPSMC_MSG_PCIeDPM_Disable ((uint16_t) 0x13d)
#define PPSMC_MSG_ACPPowerOFF ((uint16_t) 0x137)
#define PPSMC_MSG_ACPPowerON ((uint16_t) 0x138)
#define PPSMC_MSG_SAMPowerOFF ((uint16_t) 0x139)
@@ -116,8 +117,13 @@ typedef uint8_t PPSMC_Result;
#define PPSMC_MSG_NBDPM_Enable ((uint16_t) 0x140)
#define PPSMC_MSG_NBDPM_Disable ((uint16_t) 0x141)
#define PPSMC_MSG_SCLKDPM_SetEnabledMask ((uint16_t) 0x145)
+#define PPSMC_MSG_MCLKDPM_SetEnabledMask ((uint16_t) 0x146)
+#define PPSMC_MSG_PCIeDPM_ForceLevel ((uint16_t) 0x147)
+#define PPSMC_MSG_EnableVRHotGPIOInterrupt ((uint16_t) 0x14a)
#define PPSMC_MSG_DPM_Enable ((uint16_t) 0x14e)
#define PPSMC_MSG_DPM_Disable ((uint16_t) 0x14f)
+#define PPSMC_MSG_MCLKDPM_Enable ((uint16_t) 0x150)
+#define PPSMC_MSG_MCLKDPM_Disable ((uint16_t) 0x151)
#define PPSMC_MSG_UVDDPM_Enable ((uint16_t) 0x154)
#define PPSMC_MSG_UVDDPM_Disable ((uint16_t) 0x155)
#define PPSMC_MSG_SAMUDPM_Enable ((uint16_t) 0x156)
@@ -126,9 +132,25 @@ typedef uint8_t PPSMC_Result;
#define PPSMC_MSG_ACPDPM_Disable ((uint16_t) 0x159)
#define PPSMC_MSG_VCEDPM_Enable ((uint16_t) 0x15a)
#define PPSMC_MSG_VCEDPM_Disable ((uint16_t) 0x15b)
+#define PPSMC_MSG_VddC_Request ((uint16_t) 0x15f)
#define PPSMC_MSG_SCLKDPM_GetEnabledMask ((uint16_t) 0x162)
+#define PPSMC_MSG_PCIeDPM_SetEnabledMask ((uint16_t) 0x167)
+#define PPSMC_MSG_TDCLimitEnable ((uint16_t) 0x169)
+#define PPSMC_MSG_TDCLimitDisable ((uint16_t) 0x16a)
+#define PPSMC_MSG_PkgPwrLimitEnable ((uint16_t) 0x185)
+#define PPSMC_MSG_PkgPwrLimitDisable ((uint16_t) 0x186)
+#define PPSMC_MSG_PkgPwrSetLimit ((uint16_t) 0x187)
+#define PPSMC_MSG_OverDriveSetTargetTdp ((uint16_t) 0x188)
#define PPSMC_MSG_SCLKDPM_FreezeLevel ((uint16_t) 0x189)
#define PPSMC_MSG_SCLKDPM_UnfreezeLevel ((uint16_t) 0x18A)
+#define PPSMC_MSG_MCLKDPM_FreezeLevel ((uint16_t) 0x18B)
+#define PPSMC_MSG_MCLKDPM_UnfreezeLevel ((uint16_t) 0x18C)
+#define PPSMC_MSG_MASTER_DeepSleep_ON ((uint16_t) 0x18F)
+#define PPSMC_MSG_MASTER_DeepSleep_OFF ((uint16_t) 0x190)
+#define PPSMC_MSG_Remove_DC_Clamp ((uint16_t) 0x191)
+
+#define PPSMC_MSG_API_GetSclkFrequency ((uint16_t) 0x200)
+#define PPSMC_MSG_API_GetMclkFrequency ((uint16_t) 0x201)
/* TN */
#define PPSMC_MSG_DPM_Config ((uint32_t) 0x102)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 13e402c..297e744 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -246,6 +246,12 @@ int radeon_atom_get_max_vddc(struct radeon_device *rdev, u8 voltage_type,
int radeon_atom_get_leakage_vddc_based_on_leakage_idx(struct radeon_device *rdev,
u16 *voltage,
u16 leakage_idx);
+int radeon_atom_get_leakage_id_from_vbios(struct radeon_device *rdev,
+ u16 *leakage_id);
+int radeon_atom_get_leakage_vddc_based_on_leakage_params(struct radeon_device *rdev,
+ u16 *vddc, u16 *vddci,
+ u16 virtual_voltage_id,
+ u16 vbios_voltage_id);
int radeon_atom_round_to_true_voltage(struct radeon_device *rdev,
u8 voltage_type,
u16 nominal_voltage,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 9c83ecf..c633fa5 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2454,6 +2454,20 @@ static struct radeon_asic ci_asic = {
.set_uvd_clocks = &cik_set_uvd_clocks,
.get_temperature = &ci_get_temp,
},
+ .dpm = {
+ .init = &ci_dpm_init,
+ .setup_asic = &ci_dpm_setup_asic,
+ .enable = &ci_dpm_enable,
+ .disable = &ci_dpm_disable,
+ .pre_set_power_state = &ci_dpm_pre_set_power_state,
+ .set_power_state = &ci_dpm_set_power_state,
+ .post_set_power_state = &ci_dpm_post_set_power_state,
+ .display_configuration_changed = &ci_dpm_display_configuration_changed,
+ .fini = &ci_dpm_fini,
+ .get_sclk = &ci_dpm_get_sclk,
+ .get_mclk = &ci_dpm_get_mclk,
+ .print_power_state = &ci_dpm_print_power_state,
+ },
.pflip = {
.pre_page_flip = &evergreen_pre_page_flip,
.page_flip = &evergreen_page_flip,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 68a1a1f..350da17 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -750,6 +750,20 @@ void cik_compute_ring_set_wptr(struct radeon_device *rdev,
int ci_get_temp(struct radeon_device *rdev);
int kv_get_temp(struct radeon_device *rdev);
+int ci_dpm_init(struct radeon_device *rdev);
+int ci_dpm_enable(struct radeon_device *rdev);
+void ci_dpm_disable(struct radeon_device *rdev);
+int ci_dpm_pre_set_power_state(struct radeon_device *rdev);
+int ci_dpm_set_power_state(struct radeon_device *rdev);
+void ci_dpm_post_set_power_state(struct radeon_device *rdev);
+void ci_dpm_setup_asic(struct radeon_device *rdev);
+void ci_dpm_display_configuration_changed(struct radeon_device *rdev);
+void ci_dpm_fini(struct radeon_device *rdev);
+u32 ci_dpm_get_sclk(struct radeon_device *rdev, bool low);
+u32 ci_dpm_get_mclk(struct radeon_device *rdev, bool low);
+void ci_dpm_print_power_state(struct radeon_device *rdev,
+ struct radeon_ps *ps);
+
int kv_dpm_init(struct radeon_device *rdev);
int kv_dpm_enable(struct radeon_device *rdev);
void kv_dpm_disable(struct radeon_device *rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index 6247b5e..7ba439e 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -211,7 +211,7 @@ void radeon_atombios_i2c_init(struct radeon_device *rdev)
}
static struct radeon_gpio_rec radeon_lookup_gpio(struct radeon_device *rdev,
- u8 id)
+ u8 id)
{
struct atom_context *ctx = rdev->mode_info.atom_context;
struct radeon_gpio_rec gpio;
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 37d3d34..66b04af 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -1202,6 +1202,7 @@ int radeon_pm_init(struct radeon_device *rdev)
case CHIP_VERDE:
case CHIP_OLAND:
case CHIP_HAINAN:
+ case CHIP_BONAIRE:
case CHIP_KABINI:
case CHIP_KAVERI:
/* DPM requires the RLC, RV770+ dGPU requires SMC */
diff --git a/drivers/gpu/drm/radeon/radeon_ucode.h b/drivers/gpu/drm/radeon/radeon_ucode.h
index fad27c0..3385836 100644
--- a/drivers/gpu/drm/radeon/radeon_ucode.h
+++ b/drivers/gpu/drm/radeon/radeon_ucode.h
@@ -140,4 +140,7 @@
#define HAINAN_SMC_UCODE_START 0x10000
#define HAINAN_SMC_UCODE_SIZE 0xe67C
+#define BONAIRE_SMC_UCODE_START 0x20000
+#define BONAIRE_SMC_UCODE_SIZE 0x1FDEC
+
#endif
diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
index 0f8be48..96d96f5 100644
--- a/drivers/gpu/drm/radeon/si_dpm.c
+++ b/drivers/gpu/drm/radeon/si_dpm.c
@@ -3663,7 +3663,7 @@ static void si_clear_vc(struct radeon_device *rdev)
WREG32(CG_FTV, 0);
}
-static u8 si_get_ddr3_mclk_frequency_ratio(u32 memory_clock)
+u8 si_get_ddr3_mclk_frequency_ratio(u32 memory_clock)
{
u8 mc_para_index;
@@ -3676,7 +3676,7 @@ static u8 si_get_ddr3_mclk_frequency_ratio(u32 memory_clock)
return mc_para_index;
}
-static u8 si_get_mclk_frequency_ratio(u32 memory_clock, bool strobe_mode)
+u8 si_get_mclk_frequency_ratio(u32 memory_clock, bool strobe_mode)
{
u8 mc_para_index;
@@ -3758,9 +3758,9 @@ static bool si_validate_phase_shedding_tables(struct radeon_device *rdev,
return true;
}
-static void si_trim_voltage_table_to_fit_state_table(struct radeon_device *rdev,
- u32 max_voltage_steps,
- struct atom_voltage_table *voltage_table)
+void si_trim_voltage_table_to_fit_state_table(struct radeon_device *rdev,
+ u32 max_voltage_steps,
+ struct atom_voltage_table *voltage_table)
{
unsigned int i, diff;
diff --git a/drivers/gpu/drm/radeon/smu7_discrete.h b/drivers/gpu/drm/radeon/smu7_discrete.h
new file mode 100644
index 0000000..82f70c9
--- /dev/null
+++ b/drivers/gpu/drm/radeon/smu7_discrete.h
@@ -0,0 +1,486 @@
+/*
+ * Copyright 2013 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef SMU7_DISCRETE_H
+#define SMU7_DISCRETE_H
+
+#include "smu7.h"
+
+#pragma pack(push, 1)
+
+#define SMU7_DTE_ITERATIONS 5
+#define SMU7_DTE_SOURCES 3
+#define SMU7_DTE_SINKS 1
+#define SMU7_NUM_CPU_TES 0
+#define SMU7_NUM_GPU_TES 1
+#define SMU7_NUM_NON_TES 2
+
+struct SMU7_SoftRegisters
+{
+ uint32_t RefClockFrequency;
+ uint32_t PmTimerP;
+ uint32_t FeatureEnables;
+ uint32_t PreVBlankGap;
+ uint32_t VBlankTimeout;
+ uint32_t TrainTimeGap;
+
+ uint32_t MvddSwitchTime;
+ uint32_t LongestAcpiTrainTime;
+ uint32_t AcpiDelay;
+ uint32_t G5TrainTime;
+ uint32_t DelayMpllPwron;
+ uint32_t VoltageChangeTimeout;
+ uint32_t HandshakeDisables;
+
+ uint8_t DisplayPhy1Config;
+ uint8_t DisplayPhy2Config;
+ uint8_t DisplayPhy3Config;
+ uint8_t DisplayPhy4Config;
+
+ uint8_t DisplayPhy5Config;
+ uint8_t DisplayPhy6Config;
+ uint8_t DisplayPhy7Config;
+ uint8_t DisplayPhy8Config;
+
+ uint32_t AverageGraphicsA;
+ uint32_t AverageMemoryA;
+ uint32_t AverageGioA;
+
+ uint8_t SClkDpmEnabledLevels;
+ uint8_t MClkDpmEnabledLevels;
+ uint8_t LClkDpmEnabledLevels;
+ uint8_t PCIeDpmEnabledLevels;
+
+ uint8_t UVDDpmEnabledLevels;
+ uint8_t SAMUDpmEnabledLevels;
+ uint8_t ACPDpmEnabledLevels;
+ uint8_t VCEDpmEnabledLevels;
+
+ uint32_t DRAM_LOG_ADDR_H;
+ uint32_t DRAM_LOG_ADDR_L;
+ uint32_t DRAM_LOG_PHY_ADDR_H;
+ uint32_t DRAM_LOG_PHY_ADDR_L;
+ uint32_t DRAM_LOG_BUFF_SIZE;
+ uint32_t UlvEnterC;
+ uint32_t UlvTime;
+ uint32_t Reserved[3];
+
+};
+
+typedef struct SMU7_SoftRegisters SMU7_SoftRegisters;
+
+struct SMU7_Discrete_VoltageLevel
+{
+ uint16_t Voltage;
+ uint16_t StdVoltageHiSidd;
+ uint16_t StdVoltageLoSidd;
+ uint8_t Smio;
+ uint8_t padding;
+};
+
+typedef struct SMU7_Discrete_VoltageLevel SMU7_Discrete_VoltageLevel;
+
+struct SMU7_Discrete_GraphicsLevel
+{
+ uint32_t Flags;
+ uint32_t MinVddc;
+ uint32_t MinVddcPhases;
+
+ uint32_t SclkFrequency;
+
+ uint8_t padding1[2];
+ uint16_t ActivityLevel;
+
+ uint32_t CgSpllFuncCntl3;
+ uint32_t CgSpllFuncCntl4;
+ uint32_t SpllSpreadSpectrum;
+ uint32_t SpllSpreadSpectrum2;
+ uint32_t CcPwrDynRm;
+ uint32_t CcPwrDynRm1;
+ uint8_t SclkDid;
+ uint8_t DisplayWatermark;
+ uint8_t EnabledForActivity;
+ uint8_t EnabledForThrottle;
+ uint8_t UpH;
+ uint8_t DownH;
+ uint8_t VoltageDownH;
+ uint8_t PowerThrottle;
+ uint8_t DeepSleepDivId;
+ uint8_t padding[3];
+};
+
+typedef struct SMU7_Discrete_GraphicsLevel SMU7_Discrete_GraphicsLevel;
+
+struct SMU7_Discrete_ACPILevel
+{
+ uint32_t Flags;
+ uint32_t MinVddc;
+ uint32_t MinVddcPhases;
+ uint32_t SclkFrequency;
+ uint8_t SclkDid;
+ uint8_t DisplayWatermark;
+ uint8_t DeepSleepDivId;
+ uint8_t padding;
+ uint32_t CgSpllFuncCntl;
+ uint32_t CgSpllFuncCntl2;
+ uint32_t CgSpllFuncCntl3;
+ uint32_t CgSpllFuncCntl4;
+ uint32_t SpllSpreadSpectrum;
+ uint32_t SpllSpreadSpectrum2;
+ uint32_t CcPwrDynRm;
+ uint32_t CcPwrDynRm1;
+};
+
+typedef struct SMU7_Discrete_ACPILevel SMU7_Discrete_ACPILevel;
+
+struct SMU7_Discrete_Ulv
+{
+ uint32_t CcPwrDynRm;
+ uint32_t CcPwrDynRm1;
+ uint16_t VddcOffset;
+ uint8_t VddcOffsetVid;
+ uint8_t VddcPhase;
+ uint32_t Reserved;
+};
+
+typedef struct SMU7_Discrete_Ulv SMU7_Discrete_Ulv;
+
+struct SMU7_Discrete_MemoryLevel
+{
+ uint32_t MinVddc;
+ uint32_t MinVddcPhases;
+ uint32_t MinVddci;
+ uint32_t MinMvdd;
+
+ uint32_t MclkFrequency;
+
+ uint8_t EdcReadEnable;
+ uint8_t EdcWriteEnable;
+ uint8_t RttEnable;
+ uint8_t StutterEnable;
+
+ uint8_t StrobeEnable;
+ uint8_t StrobeRatio;
+ uint8_t EnabledForThrottle;
+ uint8_t EnabledForActivity;
+
+ uint8_t UpH;
+ uint8_t DownH;
+ uint8_t VoltageDownH;
+ uint8_t padding;
+
+ uint16_t ActivityLevel;
+ uint8_t DisplayWatermark;
+ uint8_t padding1;
+
+ uint32_t MpllFuncCntl;
+ uint32_t MpllFuncCntl_1;
+ uint32_t MpllFuncCntl_2;
+ uint32_t MpllAdFuncCntl;
+ uint32_t MpllDqFuncCntl;
+ uint32_t MclkPwrmgtCntl;
+ uint32_t DllCntl;
+ uint32_t MpllSs1;
+ uint32_t MpllSs2;
+};
+
+typedef struct SMU7_Discrete_MemoryLevel SMU7_Discrete_MemoryLevel;
+
+struct SMU7_Discrete_LinkLevel
+{
+ uint8_t PcieGenSpeed;
+ uint8_t PcieLaneCount;
+ uint8_t EnabledForActivity;
+ uint8_t Padding;
+ uint32_t DownT;
+ uint32_t UpT;
+ uint32_t Reserved;
+};
+
+typedef struct SMU7_Discrete_LinkLevel SMU7_Discrete_LinkLevel;
+
+
+struct SMU7_Discrete_MCArbDramTimingTableEntry
+{
+ uint32_t McArbDramTiming;
+ uint32_t McArbDramTiming2;
+ uint8_t McArbBurstTime;
+ uint8_t padding[3];
+};
+
+typedef struct SMU7_Discrete_MCArbDramTimingTableEntry SMU7_Discrete_MCArbDramTimingTableEntry;
+
+struct SMU7_Discrete_MCArbDramTimingTable
+{
+ SMU7_Discrete_MCArbDramTimingTableEntry entries[SMU__NUM_SCLK_DPM_STATE][SMU__NUM_MCLK_DPM_LEVELS];
+};
+
+typedef struct SMU7_Discrete_MCArbDramTimingTable SMU7_Discrete_MCArbDramTimingTable;
+
+struct SMU7_Discrete_UvdLevel
+{
+ uint32_t VclkFrequency;
+ uint32_t DclkFrequency;
+ uint16_t MinVddc;
+ uint8_t MinVddcPhases;
+ uint8_t VclkDivider;
+ uint8_t DclkDivider;
+ uint8_t padding[3];
+};
+
+typedef struct SMU7_Discrete_UvdLevel SMU7_Discrete_UvdLevel;
+
+struct SMU7_Discrete_ExtClkLevel
+{
+ uint32_t Frequency;
+ uint16_t MinVoltage;
+ uint8_t MinPhases;
+ uint8_t Divider;
+};
+
+typedef struct SMU7_Discrete_ExtClkLevel SMU7_Discrete_ExtClkLevel;
+
+struct SMU7_Discrete_StateInfo
+{
+ uint32_t SclkFrequency;
+ uint32_t MclkFrequency;
+ uint32_t VclkFrequency;
+ uint32_t DclkFrequency;
+ uint32_t SamclkFrequency;
+ uint32_t AclkFrequency;
+ uint32_t EclkFrequency;
+ uint16_t MvddVoltage;
+ uint16_t padding16;
+ uint8_t DisplayWatermark;
+ uint8_t McArbIndex;
+ uint8_t McRegIndex;
+ uint8_t SeqIndex;
+ uint8_t SclkDid;
+ int8_t SclkIndex;
+ int8_t MclkIndex;
+ uint8_t PCIeGen;
+
+};
+
+typedef struct SMU7_Discrete_StateInfo SMU7_Discrete_StateInfo;
+
+
+struct SMU7_Discrete_DpmTable
+{
+ SMU7_PIDController GraphicsPIDController;
+ SMU7_PIDController MemoryPIDController;
+ SMU7_PIDController LinkPIDController;
+
+ uint32_t SystemFlags;
+
+
+ uint32_t SmioMaskVddcVid;
+ uint32_t SmioMaskVddcPhase;
+ uint32_t SmioMaskVddciVid;
+ uint32_t SmioMaskMvddVid;
+
+ uint32_t VddcLevelCount;
+ uint32_t VddciLevelCount;
+ uint32_t MvddLevelCount;
+
+ SMU7_Discrete_VoltageLevel VddcLevel [SMU7_MAX_LEVELS_VDDC];
+// SMU7_Discrete_VoltageLevel VddcStandardReference [SMU7_MAX_LEVELS_VDDC];
+ SMU7_Discrete_VoltageLevel VddciLevel [SMU7_MAX_LEVELS_VDDCI];
+ SMU7_Discrete_VoltageLevel MvddLevel [SMU7_MAX_LEVELS_MVDD];
+
+ uint8_t GraphicsDpmLevelCount;
+ uint8_t MemoryDpmLevelCount;
+ uint8_t LinkLevelCount;
+ uint8_t UvdLevelCount;
+ uint8_t VceLevelCount;
+ uint8_t AcpLevelCount;
+ uint8_t SamuLevelCount;
+ uint8_t MasterDeepSleepControl;
+ uint32_t Reserved[5];
+// uint32_t SamuDefaultLevel;
+
+ SMU7_Discrete_GraphicsLevel GraphicsLevel [SMU7_MAX_LEVELS_GRAPHICS];
+ SMU7_Discrete_MemoryLevel MemoryACPILevel;
+ SMU7_Discrete_MemoryLevel MemoryLevel [SMU7_MAX_LEVELS_MEMORY];
+ SMU7_Discrete_LinkLevel LinkLevel [SMU7_MAX_LEVELS_LINK];
+ SMU7_Discrete_ACPILevel ACPILevel;
+ SMU7_Discrete_UvdLevel UvdLevel [SMU7_MAX_LEVELS_UVD];
+ SMU7_Discrete_ExtClkLevel VceLevel [SMU7_MAX_LEVELS_VCE];
+ SMU7_Discrete_ExtClkLevel AcpLevel [SMU7_MAX_LEVELS_ACP];
+ SMU7_Discrete_ExtClkLevel SamuLevel [SMU7_MAX_LEVELS_SAMU];
+ SMU7_Discrete_Ulv Ulv;
+
+ uint32_t SclkStepSize;
+ uint32_t Smio [SMU7_MAX_ENTRIES_SMIO];
+
+ uint8_t UvdBootLevel;
+ uint8_t VceBootLevel;
+ uint8_t AcpBootLevel;
+ uint8_t SamuBootLevel;
+
+ uint8_t UVDInterval;
+ uint8_t VCEInterval;
+ uint8_t ACPInterval;
+ uint8_t SAMUInterval;
+
+ uint8_t GraphicsBootLevel;
+ uint8_t GraphicsVoltageChangeEnable;
+ uint8_t GraphicsThermThrottleEnable;
+ uint8_t GraphicsInterval;
+
+ uint8_t VoltageInterval;
+ uint8_t ThermalInterval;
+ uint16_t TemperatureLimitHigh;
+
+ uint16_t TemperatureLimitLow;
+ uint8_t MemoryBootLevel;
+ uint8_t MemoryVoltageChangeEnable;
+
+ uint8_t MemoryInterval;
+ uint8_t MemoryThermThrottleEnable;
+ uint16_t VddcVddciDelta;
+
+ uint16_t VoltageResponseTime;
+ uint16_t PhaseResponseTime;
+
+ uint8_t PCIeBootLinkLevel;
+ uint8_t PCIeGenInterval;
+ uint8_t DTEInterval;
+ uint8_t DTEMode;
+
+ uint8_t SVI2Enable;
+ uint8_t VRHotGpio;
+ uint8_t AcDcGpio;
+ uint8_t ThermGpio;
+
+ uint16_t PPM_PkgPwrLimit;
+ uint16_t PPM_TemperatureLimit;
+
+ uint16_t DefaultTdp;
+ uint16_t TargetTdp;
+
+ uint16_t FpsHighT;
+ uint16_t FpsLowT;
+
+ uint16_t BAPMTI_R [SMU7_DTE_ITERATIONS][SMU7_DTE_SOURCES][SMU7_DTE_SINKS];
+ uint16_t BAPMTI_RC [SMU7_DTE_ITERATIONS][SMU7_DTE_SOURCES][SMU7_DTE_SINKS];
+
+ uint8_t DTEAmbientTempBase;
+ uint8_t DTETjOffset;
+ uint8_t GpuTjMax;
+ uint8_t GpuTjHyst;
+
+ uint16_t BootVddc;
+ uint16_t BootVddci;
+
+ uint16_t BootMVdd;
+ uint16_t padding;
+
+ uint32_t BAPM_TEMP_GRADIENT;
+
+ uint32_t LowSclkInterruptT;
+};
+
+typedef struct SMU7_Discrete_DpmTable SMU7_Discrete_DpmTable;
+
+#define SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE 16
+#define SMU7_DISCRETE_MC_REGISTER_ARRAY_SET_COUNT SMU7_MAX_LEVELS_MEMORY
+
+struct SMU7_Discrete_MCRegisterAddress
+{
+ uint16_t s0;
+ uint16_t s1;
+};
+
+typedef struct SMU7_Discrete_MCRegisterAddress SMU7_Discrete_MCRegisterAddress;
+
+struct SMU7_Discrete_MCRegisterSet
+{
+ uint32_t value[SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE];
+};
+
+typedef struct SMU7_Discrete_MCRegisterSet SMU7_Discrete_MCRegisterSet;
+
+struct SMU7_Discrete_MCRegisters
+{
+ uint8_t last;
+ uint8_t reserved[3];
+ SMU7_Discrete_MCRegisterAddress address[SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE];
+ SMU7_Discrete_MCRegisterSet data[SMU7_DISCRETE_MC_REGISTER_ARRAY_SET_COUNT];
+};
+
+typedef struct SMU7_Discrete_MCRegisters SMU7_Discrete_MCRegisters;
+
+struct SMU7_Discrete_PmFuses {
+ // dw0-dw1
+ uint8_t BapmVddCVidHiSidd[8];
+
+ // dw2-dw3
+ uint8_t BapmVddCVidLoSidd[8];
+
+ // dw4-dw5
+ uint8_t VddCVid[8];
+
+ // dw6
+ uint8_t SviLoadLineEn;
+ uint8_t SviLoadLineVddC;
+ uint8_t SviLoadLineTrimVddC;
+ uint8_t SviLoadLineOffsetVddC;
+
+ // dw7
+ uint16_t TDC_VDDC_PkgLimit;
+ uint8_t TDC_VDDC_ThrottleReleaseLimitPerc;
+ uint8_t TDC_MAWt;
+
+ // dw8
+ uint8_t TdcWaterfallCtl;
+ uint8_t LPMLTemperatureMin;
+ uint8_t LPMLTemperatureMax;
+ uint8_t Reserved;
+
+ // dw9-dw10
+ uint8_t BapmVddCVidHiSidd2[8];
+
+ // dw11-dw12
+ uint32_t Reserved6[2];
+
+ // dw13-dw16
+ uint8_t GnbLPML[16];
+
+ // dw17
+ uint8_t GnbLPMLMaxVid;
+ uint8_t GnbLPMLMinVid;
+ uint8_t Reserved1[2];
+
+ // dw18
+ uint16_t BapmVddCBaseLeakageHiSidd;
+ uint16_t BapmVddCBaseLeakageLoSidd;
+};
+
+typedef struct SMU7_Discrete_PmFuses SMU7_Discrete_PmFuses;
+
+
+#pragma pack(pop)
+
+#endif
+
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 43/53] drm/radeon/dpm: add debugfs support for CI
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (41 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 42/53] drm/radeon: add dpm support for CI dGPUs (v2) Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 44/53] drm/radeon/dpm: implement force performance level " Alex Deucher
` (10 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
This allows you to look at the current DPM state via debugfs.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/ci_dpm.c | 11 +++++++++++
drivers/gpu/drm/radeon/cikd.h | 12 ++++++++++++
drivers/gpu/drm/radeon/radeon_asic.c | 1 +
drivers/gpu/drm/radeon/radeon_asic.h | 2 ++
4 files changed, 26 insertions(+)
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index 72ab92b..28b2b36 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -27,6 +27,7 @@
#include "r600_dpm.h"
#include "ci_dpm.h"
#include "atom.h"
+#include <linux/seq_file.h>
#define MC_CG_ARB_FREQ_F0 0x0a
#define MC_CG_ARB_FREQ_F1 0x0b
@@ -4965,6 +4966,16 @@ int ci_dpm_init(struct radeon_device *rdev)
return 0;
}
+void ci_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev,
+ struct seq_file *m)
+{
+ u32 sclk = ci_get_average_sclk_freq(rdev);
+ u32 mclk = ci_get_average_mclk_freq(rdev);
+
+ seq_printf(m, "power level avg sclk: %u mclk: %u\n",
+ sclk, mclk);
+}
+
void ci_dpm_print_power_state(struct radeon_device *rdev,
struct radeon_ps *rps)
{
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index 52cbf6f..baada55 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -110,6 +110,14 @@
# define RESET_SCLK_CNT (1 << 5)
# define DYNAMIC_PM_EN (1 << 21)
+#define TARGET_AND_CURRENT_PROFILE_INDEX 0xC0200014
+# define CURRENT_STATE_MASK (0xf << 4)
+# define CURRENT_STATE_SHIFT 4
+# define CURR_MCLK_INDEX_MASK (0xf << 8)
+# define CURR_MCLK_INDEX_SHIFT 8
+# define CURR_SCLK_INDEX_MASK (0x1f << 16)
+# define CURR_SCLK_INDEX_SHIFT 16
+
#define CG_SSP 0xC0200044
# define SST(x) ((x) << 0)
# define SST_MASK (0xffff << 0)
@@ -126,6 +134,10 @@
# define DISP_GAP_MCHG(x) ((x) << 24)
# define DISP_GAP_MCHG_MASK (3 << 24)
+#define TARGET_AND_CURRENT_PROFILE_INDEX_1 0xC02000F0
+# define CURR_PCIE_INDEX_MASK (0xf << 24)
+# define CURR_PCIE_INDEX_SHIFT 24
+
#define CG_ULV_PARAMETER 0xC0200158
#define CG_FTV_0 0xC02001A8
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index c633fa5..e28f08b 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2467,6 +2467,7 @@ static struct radeon_asic ci_asic = {
.get_sclk = &ci_dpm_get_sclk,
.get_mclk = &ci_dpm_get_mclk,
.print_power_state = &ci_dpm_print_power_state,
+ .debugfs_print_current_performance_level = &ci_dpm_debugfs_print_current_performance_level,
},
.pflip = {
.pre_page_flip = &evergreen_pre_page_flip,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 350da17..371a6f2 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -763,6 +763,8 @@ u32 ci_dpm_get_sclk(struct radeon_device *rdev, bool low);
u32 ci_dpm_get_mclk(struct radeon_device *rdev, bool low);
void ci_dpm_print_power_state(struct radeon_device *rdev,
struct radeon_ps *ps);
+void ci_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev,
+ struct seq_file *m);
int kv_dpm_init(struct radeon_device *rdev);
int kv_dpm_enable(struct radeon_device *rdev);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 44/53] drm/radeon/dpm: implement force performance level for CI
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (42 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 43/53] drm/radeon/dpm: add debugfs support for CI Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 45/53] drm/radeon/dpm: implement vblank_too_short callback " Alex Deucher
` (9 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Allows you to force the selected performance level via sysfs.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/ci_dpm.c | 153 +++++++++++++++++++++++++++++++++++
drivers/gpu/drm/radeon/ppsmc.h | 1 +
drivers/gpu/drm/radeon/radeon_asic.c | 1 +
drivers/gpu/drm/radeon/radeon_asic.h | 2 +
4 files changed, 157 insertions(+)
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index 28b2b36..9eebf1f 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -3601,6 +3601,153 @@ static int ci_generate_dpm_level_enable_mask(struct radeon_device *rdev,
return 0;
}
+static u32 ci_get_lowest_enabled_level(struct radeon_device *rdev,
+ u32 level_mask)
+{
+ u32 level = 0;
+
+ while ((level_mask & (1 << level)) == 0)
+ level++;
+
+ return level;
+}
+
+
+int ci_dpm_force_performance_level(struct radeon_device *rdev,
+ enum radeon_dpm_forced_level level)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ PPSMC_Result smc_result;
+ u32 tmp, levels, i;
+ int ret;
+
+ if (level == RADEON_DPM_FORCED_LEVEL_HIGH) {
+ if ((!pi->sclk_dpm_key_disabled) &&
+ pi->dpm_level_enable_mask.sclk_dpm_enable_mask) {
+ levels = 0;
+ tmp = pi->dpm_level_enable_mask.sclk_dpm_enable_mask;
+ while (tmp >>= 1)
+ levels++;
+ if (levels) {
+ ret = ci_dpm_force_state_sclk(rdev, levels);
+ if (ret)
+ return ret;
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ tmp = (RREG32_SMC(TARGET_AND_CURRENT_PROFILE_INDEX) &
+ CURR_SCLK_INDEX_MASK) >> CURR_SCLK_INDEX_SHIFT;
+ if (tmp == levels)
+ break;
+ udelay(1);
+ }
+ }
+ }
+ if ((!pi->mclk_dpm_key_disabled) &&
+ pi->dpm_level_enable_mask.mclk_dpm_enable_mask) {
+ levels = 0;
+ tmp = pi->dpm_level_enable_mask.mclk_dpm_enable_mask;
+ while (tmp >>= 1)
+ levels++;
+ if (levels) {
+ ret = ci_dpm_force_state_mclk(rdev, levels);
+ if (ret)
+ return ret;
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ tmp = (RREG32_SMC(TARGET_AND_CURRENT_PROFILE_INDEX) &
+ CURR_MCLK_INDEX_MASK) >> CURR_MCLK_INDEX_SHIFT;
+ if (tmp == levels)
+ break;
+ udelay(1);
+ }
+ }
+ }
+ if ((!pi->pcie_dpm_key_disabled) &&
+ pi->dpm_level_enable_mask.pcie_dpm_enable_mask) {
+ levels = 0;
+ tmp = pi->dpm_level_enable_mask.pcie_dpm_enable_mask;
+ while (tmp >>= 1)
+ levels++;
+ if (levels) {
+ ret = ci_dpm_force_state_pcie(rdev, level);
+ if (ret)
+ return ret;
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ tmp = (RREG32_SMC(TARGET_AND_CURRENT_PROFILE_INDEX_1) &
+ CURR_PCIE_INDEX_MASK) >> CURR_PCIE_INDEX_SHIFT;
+ if (tmp == levels)
+ break;
+ udelay(1);
+ }
+ }
+ }
+ } else if (level == RADEON_DPM_FORCED_LEVEL_LOW) {
+ if ((!pi->sclk_dpm_key_disabled) &&
+ pi->dpm_level_enable_mask.sclk_dpm_enable_mask) {
+ levels = ci_get_lowest_enabled_level(rdev,
+ pi->dpm_level_enable_mask.sclk_dpm_enable_mask);
+ ret = ci_dpm_force_state_sclk(rdev, levels);
+ if (ret)
+ return ret;
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ tmp = (RREG32_SMC(TARGET_AND_CURRENT_PROFILE_INDEX) &
+ CURR_SCLK_INDEX_MASK) >> CURR_SCLK_INDEX_SHIFT;
+ if (tmp == levels)
+ break;
+ udelay(1);
+ }
+ }
+ if ((!pi->mclk_dpm_key_disabled) &&
+ pi->dpm_level_enable_mask.mclk_dpm_enable_mask) {
+ levels = ci_get_lowest_enabled_level(rdev,
+ pi->dpm_level_enable_mask.mclk_dpm_enable_mask);
+ ret = ci_dpm_force_state_mclk(rdev, levels);
+ if (ret)
+ return ret;
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ tmp = (RREG32_SMC(TARGET_AND_CURRENT_PROFILE_INDEX) &
+ CURR_MCLK_INDEX_MASK) >> CURR_MCLK_INDEX_SHIFT;
+ if (tmp == levels)
+ break;
+ udelay(1);
+ }
+ }
+ if ((!pi->pcie_dpm_key_disabled) &&
+ pi->dpm_level_enable_mask.pcie_dpm_enable_mask) {
+ levels = ci_get_lowest_enabled_level(rdev,
+ pi->dpm_level_enable_mask.pcie_dpm_enable_mask);
+ ret = ci_dpm_force_state_pcie(rdev, levels);
+ if (ret)
+ return ret;
+ for (i = 0; i < rdev->usec_timeout; i++) {
+ tmp = (RREG32_SMC(TARGET_AND_CURRENT_PROFILE_INDEX_1) &
+ CURR_PCIE_INDEX_MASK) >> CURR_PCIE_INDEX_SHIFT;
+ if (tmp == levels)
+ break;
+ udelay(1);
+ }
+ }
+ } else if (level == RADEON_DPM_FORCED_LEVEL_AUTO) {
+ if (!pi->sclk_dpm_key_disabled) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_NoForcedLevel);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+ if (!pi->mclk_dpm_key_disabled) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_MCLKDPM_NoForcedLevel);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+ if (!pi->pcie_dpm_key_disabled) {
+ smc_result = ci_send_msg_to_smc(rdev, PPSMC_MSG_PCIeDPM_UnForceLevel);
+ if (smc_result != PPSMC_Result_OK)
+ return -EINVAL;
+ }
+ }
+
+ rdev->pm.dpm.forced_level = level;
+
+ return 0;
+}
+
static int ci_set_mc_special_registers(struct radeon_device *rdev,
struct ci_mc_reg_table *table)
{
@@ -4548,6 +4695,12 @@ int ci_dpm_set_power_state(struct radeon_device *rdev)
if (pi->pcie_performance_request)
ci_notify_link_speed_change_after_state_change(rdev, new_ps, old_ps);
+ ret = ci_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_AUTO);
+ if (ret) {
+ DRM_ERROR("ci_dpm_force_performance_level failed\n");
+ return ret;
+ }
+
return 0;
}
diff --git a/drivers/gpu/drm/radeon/ppsmc.h b/drivers/gpu/drm/radeon/ppsmc.h
index 4c1ee6d..6828428 100644
--- a/drivers/gpu/drm/radeon/ppsmc.h
+++ b/drivers/gpu/drm/radeon/ppsmc.h
@@ -119,6 +119,7 @@ typedef uint8_t PPSMC_Result;
#define PPSMC_MSG_SCLKDPM_SetEnabledMask ((uint16_t) 0x145)
#define PPSMC_MSG_MCLKDPM_SetEnabledMask ((uint16_t) 0x146)
#define PPSMC_MSG_PCIeDPM_ForceLevel ((uint16_t) 0x147)
+#define PPSMC_MSG_PCIeDPM_UnForceLevel ((uint16_t) 0x148)
#define PPSMC_MSG_EnableVRHotGPIOInterrupt ((uint16_t) 0x14a)
#define PPSMC_MSG_DPM_Enable ((uint16_t) 0x14e)
#define PPSMC_MSG_DPM_Disable ((uint16_t) 0x14f)
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index e28f08b..705bc74 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2468,6 +2468,7 @@ static struct radeon_asic ci_asic = {
.get_mclk = &ci_dpm_get_mclk,
.print_power_state = &ci_dpm_print_power_state,
.debugfs_print_current_performance_level = &ci_dpm_debugfs_print_current_performance_level,
+ .force_performance_level = &ci_dpm_force_performance_level,
},
.pflip = {
.pre_page_flip = &evergreen_pre_page_flip,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 371a6f2..a887928 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -765,6 +765,8 @@ void ci_dpm_print_power_state(struct radeon_device *rdev,
struct radeon_ps *ps);
void ci_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev,
struct seq_file *m);
+int ci_dpm_force_performance_level(struct radeon_device *rdev,
+ enum radeon_dpm_forced_level level);
int kv_dpm_init(struct radeon_device *rdev);
int kv_dpm_enable(struct radeon_device *rdev);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 45/53] drm/radeon/dpm: implement vblank_too_short callback for CI
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (43 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 44/53] drm/radeon/dpm: implement force performance level " Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 46/53] drm/radeon/dpm: add debugfs support for KB/KV Alex Deucher
` (8 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Check if we can switch the mclk during the vblank time otherwise
we may get artifacts on the screen when the mclk changes.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/ci_dpm.c | 16 +++++++++++++++-
drivers/gpu/drm/radeon/radeon_asic.c | 1 +
drivers/gpu/drm/radeon/radeon_asic.h | 1 +
3 files changed, 17 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index 9eebf1f..6e3d387 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -682,6 +682,19 @@ static void ci_dpm_powergate_uvd(struct radeon_device *rdev, bool gate)
ci_update_uvd_dpm(rdev, gate);
}
+bool ci_dpm_vblank_too_short(struct radeon_device *rdev)
+{
+ struct ci_power_info *pi = ci_get_pi(rdev);
+ u32 vblank_time = r600_dpm_get_vblank_time(rdev);
+ u32 switch_limit = pi->mem_gddr5 ? 450 : 300;
+
+ if (vblank_time < switch_limit)
+ return true;
+ else
+ return false;
+
+}
+
static void ci_apply_state_adjust_rules(struct radeon_device *rdev,
struct radeon_ps *rps)
{
@@ -692,7 +705,8 @@ static void ci_apply_state_adjust_rules(struct radeon_device *rdev,
u32 sclk, mclk;
int i;
- if (rdev->pm.dpm.new_active_crtc_count > 1)
+ if ((rdev->pm.dpm.new_active_crtc_count > 1) ||
+ ci_dpm_vblank_too_short(rdev))
disable_mclk_switching = true;
else
disable_mclk_switching = false;
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 705bc74..2d7bdda 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2469,6 +2469,7 @@ static struct radeon_asic ci_asic = {
.print_power_state = &ci_dpm_print_power_state,
.debugfs_print_current_performance_level = &ci_dpm_debugfs_print_current_performance_level,
.force_performance_level = &ci_dpm_force_performance_level,
+ .vblank_too_short = &ci_dpm_vblank_too_short,
},
.pflip = {
.pre_page_flip = &evergreen_pre_page_flip,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index a887928..b5f4e43 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -767,6 +767,7 @@ void ci_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev,
struct seq_file *m);
int ci_dpm_force_performance_level(struct radeon_device *rdev,
enum radeon_dpm_forced_level level);
+bool ci_dpm_vblank_too_short(struct radeon_device *rdev);
int kv_dpm_init(struct radeon_device *rdev);
int kv_dpm_enable(struct radeon_device *rdev);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 46/53] drm/radeon/dpm: add debugfs support for KB/KV
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (44 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 45/53] drm/radeon/dpm: implement vblank_too_short callback " Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 47/53] drm/radeon/dpm: implement force performance level " Alex Deucher
` (7 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
This allows you to look at the current DPM state via
debugfs.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cikd.h | 4 ++++
drivers/gpu/drm/radeon/kv_dpm.c | 23 +++++++++++++++++++++++
drivers/gpu/drm/radeon/radeon_asic.c | 1 +
drivers/gpu/drm/radeon/radeon_asic.h | 2 ++
4 files changed, 30 insertions(+)
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index baada55..0977cf8 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -134,6 +134,10 @@
# define DISP_GAP_MCHG(x) ((x) << 24)
# define DISP_GAP_MCHG_MASK (3 << 24)
+#define SMU_VOLTAGE_STATUS 0xC0200094
+# define SMU_VOLTAGE_CURRENT_LEVEL_MASK (0xff << 1)
+# define SMU_VOLTAGE_CURRENT_LEVEL_SHIFT 1
+
#define TARGET_AND_CURRENT_PROFILE_INDEX_1 0xC02000F0
# define CURR_PCIE_INDEX_MASK (0xf << 24)
# define CURR_PCIE_INDEX_SHIFT 24
diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c
index 2e40163..d584ee4 100644
--- a/drivers/gpu/drm/radeon/kv_dpm.c
+++ b/drivers/gpu/drm/radeon/kv_dpm.c
@@ -26,6 +26,7 @@
#include "cikd.h"
#include "r600_dpm.h"
#include "kv_dpm.h"
+#include <linux/seq_file.h>
#define KV_MAX_DEEPSLEEP_DIVIDER_ID 5
#define KV_MINIMUM_ENGINE_CLOCK 800
@@ -2481,6 +2482,28 @@ int kv_dpm_init(struct radeon_device *rdev)
return 0;
}
+void kv_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev,
+ struct seq_file *m)
+{
+ struct kv_power_info *pi = kv_get_pi(rdev);
+ u32 current_index =
+ (RREG32_SMC(TARGET_AND_CURRENT_PROFILE_INDEX) & CURR_SCLK_INDEX_MASK) >>
+ CURR_SCLK_INDEX_SHIFT;
+ u32 sclk, tmp;
+ u16 vddc;
+
+ if (current_index >= SMU__NUM_SCLK_DPM_STATE) {
+ seq_printf(m, "invalid dpm profile %d\n", current_index);
+ } else {
+ sclk = be32_to_cpu(pi->graphics_level[current_index].SclkFrequency);
+ tmp = (RREG32_SMC(SMU_VOLTAGE_STATUS) & SMU_VOLTAGE_CURRENT_LEVEL_MASK) >>
+ SMU_VOLTAGE_CURRENT_LEVEL_SHIFT;
+ vddc = kv_convert_8bit_index_to_voltage(rdev, (u16)tmp);
+ seq_printf(m, "power level %d sclk: %u vddc: %u\n",
+ current_index, sclk, vddc);
+ }
+}
+
void kv_dpm_print_power_state(struct radeon_device *rdev,
struct radeon_ps *rps)
{
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 2d7bdda..b25172b 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2640,6 +2640,7 @@ static struct radeon_asic kv_asic = {
.get_sclk = &kv_dpm_get_sclk,
.get_mclk = &kv_dpm_get_mclk,
.print_power_state = &kv_dpm_print_power_state,
+ .debugfs_print_current_performance_level = &kv_dpm_debugfs_print_current_performance_level,
},
.pflip = {
.pre_page_flip = &evergreen_pre_page_flip,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index b5f4e43..80ad5d8 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -782,5 +782,7 @@ u32 kv_dpm_get_sclk(struct radeon_device *rdev, bool low);
u32 kv_dpm_get_mclk(struct radeon_device *rdev, bool low);
void kv_dpm_print_power_state(struct radeon_device *rdev,
struct radeon_ps *ps);
+void kv_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev,
+ struct seq_file *m);
#endif
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 47/53] drm/radeon/dpm: implement force performance level for KB/KV
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (45 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 46/53] drm/radeon/dpm: add debugfs support for KB/KV Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 48/53] drm/radeon/dpm: add new callback for powergating UVD (v3) Alex Deucher
` (6 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Allows you to force the selected performance level via sysfs.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/kv_dpm.c | 43 ++++++++++++++++++++++++++++++++++++
drivers/gpu/drm/radeon/radeon_asic.c | 1 +
drivers/gpu/drm/radeon/radeon_asic.h | 2 ++
3 files changed, 46 insertions(+)
diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c
index d584ee4..c26c4e3 100644
--- a/drivers/gpu/drm/radeon/kv_dpm.c
+++ b/drivers/gpu/drm/radeon/kv_dpm.c
@@ -40,6 +40,7 @@ static void kv_enable_new_levels(struct radeon_device *rdev);
static void kv_program_nbps_index_settings(struct radeon_device *rdev,
struct radeon_ps *new_rps);
static int kv_set_enabled_levels(struct radeon_device *rdev);
+static int kv_force_dpm_highest(struct radeon_device *rdev);
static int kv_force_dpm_lowest(struct radeon_device *rdev);
static void kv_apply_state_adjust_rules(struct radeon_device *rdev,
struct radeon_ps *new_rps,
@@ -1641,6 +1642,30 @@ static int kv_enable_nb_dpm(struct radeon_device *rdev)
return ret;
}
+int kv_dpm_force_performance_level(struct radeon_device *rdev,
+ enum radeon_dpm_forced_level level)
+{
+ int ret;
+
+ if (level == RADEON_DPM_FORCED_LEVEL_HIGH) {
+ ret = kv_force_dpm_highest(rdev);
+ if (ret)
+ return ret;
+ } else if (level == RADEON_DPM_FORCED_LEVEL_LOW) {
+ ret = kv_force_dpm_lowest(rdev);
+ if (ret)
+ return ret;
+ } else if (level == RADEON_DPM_FORCED_LEVEL_AUTO) {
+ ret = kv_unforce_levels(rdev);
+ if (ret)
+ return ret;
+ }
+
+ rdev->pm.dpm.forced_level = level;
+
+ return 0;
+}
+
int kv_dpm_pre_set_power_state(struct radeon_device *rdev)
{
struct kv_power_info *pi = kv_get_pi(rdev);
@@ -1720,6 +1745,7 @@ int kv_dpm_set_power_state(struct radeon_device *rdev)
kv_enable_nb_dpm(rdev);
}
}
+ rdev->pm.dpm.forced_level = RADEON_DPM_FORCED_LEVEL_AUTO;
return 0;
}
@@ -1796,6 +1822,23 @@ static void kv_construct_boot_state(struct radeon_device *rdev)
pi->boot_pl.vce_wm = 0;
}
+static int kv_force_dpm_highest(struct radeon_device *rdev)
+{
+ int ret;
+ u32 enable_mask, i;
+
+ ret = kv_dpm_get_enable_mask(rdev, &enable_mask);
+ if (ret)
+ return ret;
+
+ for (i = SMU7_MAX_LEVELS_GRAPHICS - 1; i >= 0; i--) {
+ if (enable_mask & (1 << i))
+ break;
+ }
+
+ return kv_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_DPM_ForceState, i);
+}
+
static int kv_force_dpm_lowest(struct radeon_device *rdev)
{
int ret;
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index b25172b..1dad80c 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2641,6 +2641,7 @@ static struct radeon_asic kv_asic = {
.get_mclk = &kv_dpm_get_mclk,
.print_power_state = &kv_dpm_print_power_state,
.debugfs_print_current_performance_level = &kv_dpm_debugfs_print_current_performance_level,
+ .force_performance_level = &kv_dpm_force_performance_level,
},
.pflip = {
.pre_page_flip = &evergreen_pre_page_flip,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 80ad5d8..1e386c4 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -784,5 +784,7 @@ void kv_dpm_print_power_state(struct radeon_device *rdev,
struct radeon_ps *ps);
void kv_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev,
struct seq_file *m);
+int kv_dpm_force_performance_level(struct radeon_device *rdev,
+ enum radeon_dpm_forced_level level);
#endif
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 48/53] drm/radeon/dpm: add new callback for powergating UVD (v3)
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (46 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 47/53] drm/radeon/dpm: implement force performance level " Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 49/53] drm/radeon: restructure UVD code to handle UVD PG (v2) Alex Deucher
` (5 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Starting on CIK, multi-media blocks like UVD no longer
have special power state. Rather they have their own
DPM implementation which adjusts their clocks dynamically
when active. When they are not active, the blocks are
powergated to save power.
v2: add missing pm locks
v3: rebase on uvd state selection rework
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/radeon.h | 2 ++
drivers/gpu/drm/radeon/radeon_pm.c | 43 ++++++++++++++++++++++----------------
2 files changed, 27 insertions(+), 18 deletions(-)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 297e744..09a2434 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1730,6 +1730,7 @@ struct radeon_asic {
void (*debugfs_print_current_performance_level)(struct radeon_device *rdev, struct seq_file *m);
int (*force_performance_level)(struct radeon_device *rdev, enum radeon_dpm_forced_level level);
bool (*vblank_too_short)(struct radeon_device *rdev);
+ void (*powergate_uvd)(struct radeon_device *rdev, bool gate);
} dpm;
/* pageflipping */
struct {
@@ -2518,6 +2519,7 @@ void radeon_ring_write(struct radeon_ring *ring, uint32_t v);
#define radeon_dpm_debugfs_print_current_performance_level(rdev, m) rdev->asic->dpm.debugfs_print_current_performance_level((rdev), (m))
#define radeon_dpm_force_performance_level(rdev, l) rdev->asic->dpm.force_performance_level((rdev), (l))
#define radeon_dpm_vblank_too_short(rdev) rdev->asic->dpm.vblank_too_short((rdev))
+#define radeon_dpm_powergate_uvd(rdev, g) rdev->asic->dpm.powergate_uvd((rdev), (g))
/* Common functions */
/* AGP */
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 66b04af..a44ae9a 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -886,28 +886,35 @@ void radeon_dpm_enable_uvd(struct radeon_device *rdev, bool enable)
{
enum radeon_pm_state_type dpm_state;
- if (enable) {
+ if (rdev->asic->dpm.powergate_uvd) {
mutex_lock(&rdev->pm.mutex);
- rdev->pm.dpm.uvd_active = true;
- if ((rdev->pm.dpm.sd == 1) && (rdev->pm.dpm.hd == 0))
- dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_SD;
- else if ((rdev->pm.dpm.sd == 2) && (rdev->pm.dpm.hd == 0))
- dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD;
- else if ((rdev->pm.dpm.sd == 0) && (rdev->pm.dpm.hd == 1))
- dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD;
- else if ((rdev->pm.dpm.sd == 0) && (rdev->pm.dpm.hd == 2))
- dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD2;
- else
- dpm_state = POWER_STATE_TYPE_INTERNAL_UVD;
- rdev->pm.dpm.state = dpm_state;
+ /* enable/disable UVD */
+ radeon_dpm_powergate_uvd(rdev, enable);
mutex_unlock(&rdev->pm.mutex);
} else {
- mutex_lock(&rdev->pm.mutex);
- rdev->pm.dpm.uvd_active = false;
- mutex_unlock(&rdev->pm.mutex);
- }
+ if (enable) {
+ mutex_lock(&rdev->pm.mutex);
+ rdev->pm.dpm.uvd_active = true;
+ if ((rdev->pm.dpm.sd == 1) && (rdev->pm.dpm.hd == 0))
+ dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_SD;
+ else if ((rdev->pm.dpm.sd == 2) && (rdev->pm.dpm.hd == 0))
+ dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD;
+ else if ((rdev->pm.dpm.sd == 0) && (rdev->pm.dpm.hd == 1))
+ dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD;
+ else if ((rdev->pm.dpm.sd == 0) && (rdev->pm.dpm.hd == 2))
+ dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD2;
+ else
+ dpm_state = POWER_STATE_TYPE_INTERNAL_UVD;
+ rdev->pm.dpm.state = dpm_state;
+ mutex_unlock(&rdev->pm.mutex);
+ } else {
+ mutex_lock(&rdev->pm.mutex);
+ rdev->pm.dpm.uvd_active = false;
+ mutex_unlock(&rdev->pm.mutex);
+ }
- radeon_pm_compute_clocks(rdev);
+ radeon_pm_compute_clocks(rdev);
+ }
}
static void radeon_pm_suspend_old(struct radeon_device *rdev)
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 49/53] drm/radeon: restructure UVD code to handle UVD PG (v2)
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (47 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 48/53] drm/radeon/dpm: add new callback for powergating UVD (v3) Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 50/53] drm/radeon/dpm: implement UVD powergating for KB/KV Alex Deucher
` (4 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
When we PG (powergate) UVD, we need to re-initialize it
before we can use it again.
v2: rebase on UVD stop fixes
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cik.c | 14 +++----
drivers/gpu/drm/radeon/evergreen.c | 2 +-
drivers/gpu/drm/radeon/ni.c | 2 +-
drivers/gpu/drm/radeon/r600.c | 74 +++++++++++++++++++-----------------
drivers/gpu/drm/radeon/radeon_asic.h | 4 +-
drivers/gpu/drm/radeon/rv770.c | 2 +-
drivers/gpu/drm/radeon/si.c | 2 +-
7 files changed, 50 insertions(+), 50 deletions(-)
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 736a416..59b866a 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -69,6 +69,7 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev);
static void cik_program_aspm(struct radeon_device *rdev);
static void cik_init_pg(struct radeon_device *rdev);
static void cik_init_cg(struct radeon_device *rdev);
+static void cik_uvd_resume(struct radeon_device *rdev);
/* get temperature in millidegrees */
int ci_get_temp(struct radeon_device *rdev)
@@ -7619,8 +7620,9 @@ static int cik_startup(struct radeon_device *rdev)
return r;
}
- r = cik_uvd_resume(rdev);
+ r = radeon_uvd_resume(rdev);
if (!r) {
+ cik_uvd_resume(rdev);
r = radeon_fence_driver_start_ring(rdev,
R600_RING_TYPE_UVD_INDEX);
if (r)
@@ -7708,7 +7710,7 @@ static int cik_startup(struct radeon_device *rdev)
UVD_RBC_RB_RPTR, UVD_RBC_RB_WPTR,
0, 0xfffff, RADEON_CP_PACKET2);
if (!r)
- r = r600_uvd_init(rdev);
+ r = r600_uvd_init(rdev, true);
if (r)
DRM_ERROR("radeon: failed initializing UVD (%d).\n", r);
}
@@ -8598,15 +8600,10 @@ int cik_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk)
return r;
}
-int cik_uvd_resume(struct radeon_device *rdev)
+static void cik_uvd_resume(struct radeon_device *rdev)
{
uint64_t addr;
uint32_t size;
- int r;
-
- r = radeon_uvd_resume(rdev);
- if (r)
- return r;
/* programm the VCPU memory controller bits 0-27 */
addr = rdev->uvd.gpu_addr >> 3;
@@ -8632,7 +8629,6 @@ int cik_uvd_resume(struct radeon_device *rdev)
addr = (rdev->uvd.gpu_addr >> 32) & 0xFF;
WREG32(UVD_LMI_EXT40_ADDR, addr | (0x9 << 16) | (0x1 << 31));
- return 0;
}
static void cik_pcie_gen3_enable(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 2ce12ee..710c1d4 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -5296,7 +5296,7 @@ static int evergreen_startup(struct radeon_device *rdev)
UVD_RBC_RB_RPTR, UVD_RBC_RB_WPTR,
0, 0xfffff, RADEON_CP_PACKET2);
if (!r)
- r = r600_uvd_init(rdev);
+ r = r600_uvd_init(rdev, true);
if (r)
DRM_ERROR("radeon: error initializing UVD (%d).\n", r);
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index 21f2ece..bc298a35 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -2230,7 +2230,7 @@ static int cayman_startup(struct radeon_device *rdev)
UVD_RBC_RB_RPTR, UVD_RBC_RB_WPTR,
0, 0xfffff, RADEON_CP_PACKET2);
if (!r)
- r = r600_uvd_init(rdev);
+ r = r600_uvd_init(rdev, true);
if (r)
DRM_ERROR("radeon: failed initializing UVD (%d).\n", r);
}
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index 3db2e4d..8a60015 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -2623,7 +2623,7 @@ void r600_dma_fini(struct radeon_device *rdev)
/*
* UVD
*/
-int r600_uvd_rbc_start(struct radeon_device *rdev)
+static int r600_uvd_rbc_start(struct radeon_device *rdev, bool ring_test)
{
struct radeon_ring *ring = &rdev->ring[R600_RING_TYPE_UVD_INDEX];
uint64_t rptr_addr;
@@ -2664,47 +2664,47 @@ int r600_uvd_rbc_start(struct radeon_device *rdev)
rb_bufsz = (0x1 << 8) | rb_bufsz;
WREG32(UVD_RBC_RB_CNTL, rb_bufsz);
- ring->ready = true;
- r = radeon_ring_test(rdev, R600_RING_TYPE_UVD_INDEX, ring);
- if (r) {
- ring->ready = false;
- return r;
- }
+ if (ring_test) {
+ ring->ready = true;
+ r = radeon_ring_test(rdev, R600_RING_TYPE_UVD_INDEX, ring);
+ if (r) {
+ ring->ready = false;
+ return r;
+ }
- r = radeon_ring_lock(rdev, ring, 10);
- if (r) {
- DRM_ERROR("radeon: ring failed to lock UVD ring (%d).\n", r);
- return r;
- }
+ r = radeon_ring_lock(rdev, ring, 10);
+ if (r) {
+ DRM_ERROR("radeon: ring failed to lock UVD ring (%d).\n", r);
+ return r;
+ }
- tmp = PACKET0(UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL, 0);
- radeon_ring_write(ring, tmp);
- radeon_ring_write(ring, 0xFFFFF);
+ tmp = PACKET0(UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL, 0);
+ radeon_ring_write(ring, tmp);
+ radeon_ring_write(ring, 0xFFFFF);
- tmp = PACKET0(UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL, 0);
- radeon_ring_write(ring, tmp);
- radeon_ring_write(ring, 0xFFFFF);
+ tmp = PACKET0(UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL, 0);
+ radeon_ring_write(ring, tmp);
+ radeon_ring_write(ring, 0xFFFFF);
- tmp = PACKET0(UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL, 0);
- radeon_ring_write(ring, tmp);
- radeon_ring_write(ring, 0xFFFFF);
+ tmp = PACKET0(UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL, 0);
+ radeon_ring_write(ring, tmp);
+ radeon_ring_write(ring, 0xFFFFF);
- /* Clear timeout status bits */
- radeon_ring_write(ring, PACKET0(UVD_SEMA_TIMEOUT_STATUS, 0));
- radeon_ring_write(ring, 0x8);
+ /* Clear timeout status bits */
+ radeon_ring_write(ring, PACKET0(UVD_SEMA_TIMEOUT_STATUS, 0));
+ radeon_ring_write(ring, 0x8);
- radeon_ring_write(ring, PACKET0(UVD_SEMA_CNTL, 0));
- radeon_ring_write(ring, 3);
+ radeon_ring_write(ring, PACKET0(UVD_SEMA_CNTL, 0));
+ radeon_ring_write(ring, 3);
- radeon_ring_unlock_commit(rdev, ring);
+ radeon_ring_unlock_commit(rdev, ring);
+ }
return 0;
}
-void r600_uvd_stop(struct radeon_device *rdev)
+void r600_do_uvd_stop(struct radeon_device *rdev)
{
- struct radeon_ring *ring = &rdev->ring[R600_RING_TYPE_UVD_INDEX];
-
/* force RBC into idle state */
WREG32(UVD_RBC_RB_CNTL, 0x11010101);
@@ -2723,11 +2723,17 @@ void r600_uvd_stop(struct radeon_device *rdev)
/* Unstall UMC and register bus */
WREG32_P(UVD_LMI_CTRL2, 0, ~(1 << 8));
WREG32_P(UVD_RB_ARB_CTRL, 0, ~(1 << 3));
+}
+void r600_uvd_stop(struct radeon_device *rdev)
+{
+ struct radeon_ring *ring = &rdev->ring[R600_RING_TYPE_UVD_INDEX];
+
+ r600_do_uvd_stop(rdev);
ring->ready = false;
}
-int r600_uvd_init(struct radeon_device *rdev)
+int r600_uvd_init(struct radeon_device *rdev, bool ring_test)
{
int i, j, r;
/* disable byte swapping */
@@ -2815,17 +2821,17 @@ int r600_uvd_init(struct radeon_device *rdev)
if (r) {
DRM_ERROR("UVD not responding, giving up!!!\n");
- radeon_set_uvd_clocks(rdev, 0, 0);
- return r;
+ goto done;
}
/* enable interupt */
WREG32_P(UVD_MASTINT_EN, 3<<1, ~(3 << 1));
- r = r600_uvd_rbc_start(rdev);
+ r = r600_uvd_rbc_start(rdev, ring_test);
if (!r)
DRM_INFO("UVD initialized successfully.\n");
+done:
/* lower clocks again */
radeon_set_uvd_clocks(rdev, 0, 0);
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 1e386c4..3570817 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -424,8 +424,7 @@ void rs780_dpm_debugfs_print_current_performance_level(struct radeon_device *rde
struct seq_file *m);
/* uvd */
-int r600_uvd_init(struct radeon_device *rdev);
-int r600_uvd_rbc_start(struct radeon_device *rdev);
+int r600_uvd_init(struct radeon_device *rdev, bool ring_test);
void r600_uvd_stop(struct radeon_device *rdev);
int r600_uvd_ib_test(struct radeon_device *rdev, struct radeon_ring *ring);
void r600_uvd_fence_emit(struct radeon_device *rdev,
@@ -696,7 +695,6 @@ u32 cik_get_xclk(struct radeon_device *rdev);
uint32_t cik_pciep_rreg(struct radeon_device *rdev, uint32_t reg);
void cik_pciep_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v);
int cik_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk);
-int cik_uvd_resume(struct radeon_device *rdev);
void cik_sdma_fence_ring_emit(struct radeon_device *rdev,
struct radeon_fence *fence);
void cik_sdma_semaphore_ring_emit(struct radeon_device *rdev,
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index 8f78d21..379b19f 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -1928,7 +1928,7 @@ static int rv770_startup(struct radeon_device *rdev)
UVD_RBC_RB_RPTR, UVD_RBC_RB_WPTR,
0, 0xfffff, RADEON_CP_PACKET2);
if (!r)
- r = r600_uvd_init(rdev);
+ r = r600_uvd_init(rdev, true);
if (r)
DRM_ERROR("radeon: failed initializing UVD (%d).\n", r);
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index 4f91e1f..da23ce8 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -6421,7 +6421,7 @@ static int si_startup(struct radeon_device *rdev)
UVD_RBC_RB_RPTR, UVD_RBC_RB_WPTR,
0, 0xfffff, RADEON_CP_PACKET2);
if (!r)
- r = r600_uvd_init(rdev);
+ r = r600_uvd_init(rdev, true);
if (r)
DRM_ERROR("radeon: failed initializing UVD (%d).\n", r);
}
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 50/53] drm/radeon/dpm: implement UVD powergating for KB/KV
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (48 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 49/53] drm/radeon: restructure UVD code to handle UVD PG (v2) Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 51/53] drm/radeon/dpm: implement UVD powergating for CI Alex Deucher
` (3 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Powergate the UVD block when not in use to save power.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cik.c | 4 ++--
drivers/gpu/drm/radeon/kv_dpm.c | 22 +++++++++++++++-------
drivers/gpu/drm/radeon/radeon_asic.c | 1 +
drivers/gpu/drm/radeon/radeon_asic.h | 1 +
4 files changed, 19 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 59b866a..e661aec 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -69,7 +69,7 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev);
static void cik_program_aspm(struct radeon_device *rdev);
static void cik_init_pg(struct radeon_device *rdev);
static void cik_init_cg(struct radeon_device *rdev);
-static void cik_uvd_resume(struct radeon_device *rdev);
+void cik_uvd_resume(struct radeon_device *rdev);
/* get temperature in millidegrees */
int ci_get_temp(struct radeon_device *rdev)
@@ -8600,7 +8600,7 @@ int cik_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk)
return r;
}
-static void cik_uvd_resume(struct radeon_device *rdev)
+void cik_uvd_resume(struct radeon_device *rdev)
{
uint64_t addr;
uint32_t size;
diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c
index c26c4e3..3e232a4 100644
--- a/drivers/gpu/drm/radeon/kv_dpm.c
+++ b/drivers/gpu/drm/radeon/kv_dpm.c
@@ -49,7 +49,7 @@ static int kv_set_thermal_temperature_range(struct radeon_device *rdev,
int min_temp, int max_temp);
static int kv_init_fps_limits(struct radeon_device *rdev);
-static void kv_dpm_powergate_uvd(struct radeon_device *rdev, bool gate);
+void kv_dpm_powergate_uvd(struct radeon_device *rdev, bool gate);
static void kv_dpm_powergate_vce(struct radeon_device *rdev, bool gate);
static void kv_dpm_powergate_samu(struct radeon_device *rdev, bool gate);
static void kv_dpm_powergate_acp(struct radeon_device *rdev, bool gate);
@@ -59,6 +59,10 @@ extern void cik_exit_rlc_safe_mode(struct radeon_device *rdev);
extern void cik_update_cg(struct radeon_device *rdev,
u32 block, bool enable);
+extern void cik_uvd_resume(struct radeon_device *rdev);
+extern int r600_uvd_init(struct radeon_device *rdev, bool ring_test);
+extern void r600_do_uvd_stop(struct radeon_device *rdev);
+
static const struct kv_lcac_config_values sx_local_cac_cfg_kv[] =
{
{ 0, 4, 1 },
@@ -1201,6 +1205,7 @@ int kv_dpm_enable(struct radeon_device *rdev)
kv_dpm_powergate_acp(rdev, true);
kv_dpm_powergate_samu(rdev, true);
kv_dpm_powergate_vce(rdev, true);
+ kv_dpm_powergate_uvd(rdev, true);
kv_update_current_ps(rdev, rdev->pm.dpm.boot_ps);
@@ -1458,7 +1463,7 @@ static int kv_update_acp_dpm(struct radeon_device *rdev, bool gate)
return kv_enable_acp_dpm(rdev, !gate);
}
-static void kv_dpm_powergate_uvd(struct radeon_device *rdev, bool gate)
+void kv_dpm_powergate_uvd(struct radeon_device *rdev, bool gate)
{
struct kv_power_info *pi = kv_get_pi(rdev);
@@ -1468,13 +1473,18 @@ static void kv_dpm_powergate_uvd(struct radeon_device *rdev, bool gate)
pi->uvd_power_gated = gate;
if (gate) {
- kv_update_uvd_dpm(rdev, true);
+ r600_do_uvd_stop(rdev);
+ cik_update_cg(rdev, RADEON_CG_BLOCK_UVD, false);
+ kv_update_uvd_dpm(rdev, gate);
if (pi->caps_uvd_pg)
kv_notify_message_to_smu(rdev, PPSMC_MSG_UVDPowerOFF);
} else {
if (pi->caps_uvd_pg)
kv_notify_message_to_smu(rdev, PPSMC_MSG_UVDPowerON);
- kv_update_uvd_dpm(rdev, false);
+ cik_uvd_resume(rdev);
+ r600_uvd_init(rdev, false);
+ cik_update_cg(rdev, RADEON_CG_BLOCK_UVD, true);
+ kv_update_uvd_dpm(rdev, gate);
}
}
@@ -1714,7 +1724,6 @@ int kv_dpm_set_power_state(struct radeon_device *rdev)
return ret;
}
#endif
- kv_update_uvd_dpm(rdev, false);
kv_update_sclk_t(rdev);
}
} else {
@@ -1740,7 +1749,6 @@ int kv_dpm_set_power_state(struct radeon_device *rdev)
return ret;
}
#endif
- kv_update_uvd_dpm(rdev, false);
kv_update_sclk_t(rdev);
kv_enable_nb_dpm(rdev);
}
@@ -2502,7 +2510,7 @@ int kv_dpm_init(struct radeon_device *rdev)
pi->voltage_drop_t = 0;
pi->caps_sclk_throttle_low_notification = false;
pi->caps_fps = false; /* true? */
- pi->caps_uvd_pg = false; /* XXX */
+ pi->caps_uvd_pg = true;
pi->caps_uvd_dpm = true;
pi->caps_vce_pg = false;
pi->caps_samu_pg = false;
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 1dad80c..63b6aae 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2642,6 +2642,7 @@ static struct radeon_asic kv_asic = {
.print_power_state = &kv_dpm_print_power_state,
.debugfs_print_current_performance_level = &kv_dpm_debugfs_print_current_performance_level,
.force_performance_level = &kv_dpm_force_performance_level,
+ .powergate_uvd = &kv_dpm_powergate_uvd,
},
.pflip = {
.pre_page_flip = &evergreen_pre_page_flip,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 3570817..9060757 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -784,5 +784,6 @@ void kv_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev,
struct seq_file *m);
int kv_dpm_force_performance_level(struct radeon_device *rdev,
enum radeon_dpm_forced_level level);
+void kv_dpm_powergate_uvd(struct radeon_device *rdev, bool gate);
#endif
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 51/53] drm/radeon/dpm: implement UVD powergating for CI
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (49 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 50/53] drm/radeon/dpm: implement UVD powergating for KB/KV Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 52/53] drm/radeon/dpm: add reclocking quirk for ASUS K70AF Alex Deucher
` (2 subsequent siblings)
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
Disable the UVD block when not in use to save power.
The block is not actually powergated on CI, but we
switch between UVD DPM (where the uvd clocks are
adjusted on demand) and clocks off.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/ci_dpm.c | 7 +------
drivers/gpu/drm/radeon/radeon_asic.c | 1 +
drivers/gpu/drm/radeon/radeon_asic.h | 1 +
3 files changed, 3 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index 6e3d387..e4d9d50 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -677,7 +677,7 @@ static int ci_power_control_set_level(struct radeon_device *rdev)
return ret;
}
-static void ci_dpm_powergate_uvd(struct radeon_device *rdev, bool gate)
+void ci_dpm_powergate_uvd(struct radeon_device *rdev, bool gate)
{
ci_update_uvd_dpm(rdev, gate);
}
@@ -4674,11 +4674,6 @@ int ci_dpm_set_power_state(struct radeon_device *rdev)
return ret;
}
#endif
- ret = ci_update_uvd_dpm(rdev, false);
- if (ret) {
- DRM_ERROR("ci_update_uvd_dpm failed\n");
- return ret;
- }
ret = ci_update_sclk_t(rdev);
if (ret) {
DRM_ERROR("ci_update_sclk_t failed\n");
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 63b6aae..61c0644 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2470,6 +2470,7 @@ static struct radeon_asic ci_asic = {
.debugfs_print_current_performance_level = &ci_dpm_debugfs_print_current_performance_level,
.force_performance_level = &ci_dpm_force_performance_level,
.vblank_too_short = &ci_dpm_vblank_too_short,
+ .powergate_uvd = &ci_dpm_powergate_uvd,
},
.pflip = {
.pre_page_flip = &evergreen_pre_page_flip,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 9060757..5630291 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -766,6 +766,7 @@ void ci_dpm_debugfs_print_current_performance_level(struct radeon_device *rdev,
int ci_dpm_force_performance_level(struct radeon_device *rdev,
enum radeon_dpm_forced_level level);
bool ci_dpm_vblank_too_short(struct radeon_device *rdev);
+void ci_dpm_powergate_uvd(struct radeon_device *rdev, bool gate);
int kv_dpm_init(struct radeon_device *rdev);
int kv_dpm_enable(struct radeon_device *rdev);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 52/53] drm/radeon/dpm: add reclocking quirk for ASUS K70AF
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (50 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 51/53] drm/radeon/dpm: implement UVD powergating for CI Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-12 16:43 ` [PATCH 53/53] drm/radeon/dpm: adjust the vblank time checks for eg, ni, si Alex Deucher
2013-08-13 21:58 ` [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
The LCD has a relatively short vblank time (216us), but
the card is able to reclock memory fine in that time.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reported-by: normalrawr@gmail.com
---
drivers/gpu/drm/radeon/rv770_dpm.c | 10 +++++++++-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/radeon/rv770_dpm.c b/drivers/gpu/drm/radeon/rv770_dpm.c
index 094c67a..44c1e78 100644
--- a/drivers/gpu/drm/radeon/rv770_dpm.c
+++ b/drivers/gpu/drm/radeon/rv770_dpm.c
@@ -2517,8 +2517,16 @@ u32 rv770_dpm_get_mclk(struct radeon_device *rdev, bool low)
bool rv770_dpm_vblank_too_short(struct radeon_device *rdev)
{
u32 vblank_time = r600_dpm_get_vblank_time(rdev);
+ u32 switch_limit = 300;
- if (vblank_time < 300)
+ /* quirks */
+ /* ASUS K70AF */
+ if ((rdev->pdev->device == 0x9553) &&
+ (rdev->pdev->subsystem_vendor == 0x1043) &&
+ (rdev->pdev->subsystem_device == 0x1c42))
+ switch_limit = 200;
+
+ if (vblank_time < switch_limit)
return true;
else
return false;
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* [PATCH 53/53] drm/radeon/dpm: adjust the vblank time checks for eg, ni, si
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (51 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 52/53] drm/radeon/dpm: add reclocking quirk for ASUS K70AF Alex Deucher
@ 2013-08-12 16:43 ` Alex Deucher
2013-08-13 21:58 ` [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-12 16:43 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
According to the internal teams, we never hit the limit for
mclk switching on these asics, so we can disable the check.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/radeon/cypress_dpm.c | 3 ++-
drivers/gpu/drm/radeon/ni_dpm.c | 3 ++-
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/radeon/cypress_dpm.c b/drivers/gpu/drm/radeon/cypress_dpm.c
index 7e5d0b5..95a66db 100644
--- a/drivers/gpu/drm/radeon/cypress_dpm.c
+++ b/drivers/gpu/drm/radeon/cypress_dpm.c
@@ -2166,7 +2166,8 @@ bool cypress_dpm_vblank_too_short(struct radeon_device *rdev)
{
struct rv7xx_power_info *pi = rv770_get_pi(rdev);
u32 vblank_time = r600_dpm_get_vblank_time(rdev);
- u32 switch_limit = pi->mem_gddr5 ? 450 : 300;
+ /* we never hit the non-gddr5 limit so disable it */
+ u32 switch_limit = pi->mem_gddr5 ? 450 : 0;
if (vblank_time < switch_limit)
return true;
diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c
index f0f5f74..238078c 100644
--- a/drivers/gpu/drm/radeon/ni_dpm.c
+++ b/drivers/gpu/drm/radeon/ni_dpm.c
@@ -769,7 +769,8 @@ bool ni_dpm_vblank_too_short(struct radeon_device *rdev)
{
struct rv7xx_power_info *pi = rv770_get_pi(rdev);
u32 vblank_time = r600_dpm_get_vblank_time(rdev);
- u32 switch_limit = pi->mem_gddr5 ? 450 : 300;
+ /* we never hit the non-gddr5 limit so disable it */
+ u32 switch_limit = pi->mem_gddr5 ? 450 : 0;
if (vblank_time < switch_limit)
return true;
--
1.8.3.1
^ permalink raw reply related [flat|nested] 55+ messages in thread
* Re: [PATCH 00/53] Initial 3.12 radeon patches
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
` (52 preceding siblings ...)
2013-08-12 16:43 ` [PATCH 53/53] drm/radeon/dpm: adjust the vblank time checks for eg, ni, si Alex Deucher
@ 2013-08-13 21:58 ` Alex Deucher
53 siblings, 0 replies; 55+ messages in thread
From: Alex Deucher @ 2013-08-13 21:58 UTC (permalink / raw)
To: dri-devel; +Cc: Alex Deucher
On Mon, Aug 12, 2013 at 12:42 PM, Alex Deucher <alexdeucher@gmail.com> wrote:
> This is the initial set of patches for 3.12 for radeon. The
> big changes here are adding ASPM and DPM support to CIK asics.
> There are also some cleanups related to removing bo copy support
> using the 3D engine in favor of the sDMA engines or CP DMA.
>
> For easier review the patches are also available here:
> http://cgit.freedesktop.org/~agd5f/linux/log/?h=drm-next-3.12-wip
Just pushed an updated branch to the same URL. It includes
Christian's ring rework, my recent patches, and a few other odds and
ends.
Alex
>
> Alex Deucher (53):
> drm/radeon: switch r6xx+ to using CP DMA for the blit copy callback
> drm/radeon/kms: remove r6xx+ blit copy routines
> drm/radeon: add UVD->DPM helper function (v5)
> drm/radeon/dpm: use multiple UVD power states (v3)
> drm/radeon/dpm: rework thermal state handling
> drm/radeon: default to 1024M gart size on rv770+
> drm/radeon/dpm: use performance state if no UVD state
> drm/radeon/kms: fix up dce8 display watermark calc for dpm
> drm/radeon/cik: implement some more atom helpers for DPM
> drm/radeon: switch CIK to use radeon_ucode.h
> drm/radeon/cik: add support for pcie gen1/2/3 switching
> drm/radeon: add support for ASPM on CIK asics
> drm/radeon/cik: restructure rlc setup
> drm/radeon: clean up sumo_rlc_init() for code sharing
> drm/radeon: convert SI,CIK to use sumo_rlc functions
> drm/radeon: implement clock and power gating for CIK (v3)
> drm/radeon: add indirect accessors for dift registers on CIK
> drm/radeon/sumo add helper to go from vid7 to vid2
> drm/radeon: switch to pptable.h
> drm/radeon: add structs to store uvd clock voltage deps
> drm/radeon/cik: add rlc helpers for DPM
> drm/radeon: add support for thermal controller on KB/KV
> drm/radeon: add CI to r600_is_internal_thermal_sensor()
> drm/radeon: add KB/KV to r600_is_internal_thermal_sensor
> drm/radeon: add get_temperature() callbacks for CIK
> drm/radeon: adjust si_dpm function for code sharing
> drm/radeon/dpm: update cac leakage table parsing for CI
> drm/radeon/dpm: add support for parsing the atom powertune table
> drm/radeon/dpm: grab mvdd_dependency_on_mclk info from vbios
> drm/radeon: add structs to store vce clock voltage deps
> drm/radeon: add clock voltage dep tables for acp, samu
> drm/radeon: parse the vce clock voltage deps table
> drm/radeon: parse the uvd clock voltage deps table
> drm/radeon/dpm: clean up the extended table error pathes
> drm/radeon: parse the samu clock voltage deps table
> drm/radeon: parse the acp clock voltage deps table
> drm/radeon: add r600_get_pcie_lane_support helper
> drm/radeon/dpm: add vce clocks to radeon_ps
> drm/radeon/dpm: add a helper to encode pcie lane setting
> drm/radeon/dpm: add helper to fetch the vrefresh of the current mode
> drm/radeon/kms: add dpm support for KB/KV
> drm/radeon: add dpm support for CI dGPUs (v2)
> drm/radeon/dpm: add debugfs support for CI
> drm/radeon/dpm: implement force performance level for CI
> drm/radeon/dpm: implement vblank_too_short callback for CI
> drm/radeon/dpm: add debugfs support for KB/KV
> drm/radeon/dpm: implement force performance level for KB/KV
> drm/radeon/dpm: add new callback for powergating UVD (v3)
> drm/radeon: restructure UVD code to handle UVD PG (v2)
> drm/radeon/dpm: implement UVD powergating for KB/KV
> drm/radeon/dpm: implement UVD powergating for CI
> drm/radeon/dpm: add reclocking quirk for ASUS K70AF
> drm/radeon/dpm: adjust the vblank time checks for eg, ni, si
>
> drivers/gpu/drm/radeon/Makefile | 7 +-
> drivers/gpu/drm/radeon/atombios.h | 615 +--
> drivers/gpu/drm/radeon/cayman_blit_shaders.c | 54 -
> drivers/gpu/drm/radeon/ci_dpm.c | 5179 +++++++++++++++++++++++
> drivers/gpu/drm/radeon/ci_dpm.h | 331 ++
> drivers/gpu/drm/radeon/ci_smc.c | 262 ++
> drivers/gpu/drm/radeon/cik.c | 2115 ++++++++-
> drivers/gpu/drm/radeon/cik_reg.h | 3 +
> drivers/gpu/drm/radeon/cikd.h | 589 ++-
> drivers/gpu/drm/radeon/clearstate_cayman.h | 2 +-
> drivers/gpu/drm/radeon/clearstate_ci.h | 944 +++++
> drivers/gpu/drm/radeon/clearstate_evergreen.h | 2 +-
> drivers/gpu/drm/radeon/cypress_dpm.c | 3 +-
> drivers/gpu/drm/radeon/evergreen.c | 296 +-
> drivers/gpu/drm/radeon/evergreen_blit_kms.c | 729 ----
> drivers/gpu/drm/radeon/evergreen_blit_shaders.c | 54 -
> drivers/gpu/drm/radeon/kv_dpm.c | 2610 ++++++++++++
> drivers/gpu/drm/radeon/kv_dpm.h | 199 +
> drivers/gpu/drm/radeon/kv_smc.c | 207 +
> drivers/gpu/drm/radeon/ni.c | 16 +-
> drivers/gpu/drm/radeon/ni_dpm.c | 3 +-
> drivers/gpu/drm/radeon/ppsmc.h | 57 +
> drivers/gpu/drm/radeon/pptable.h | 682 +++
> drivers/gpu/drm/radeon/r600.c | 100 +-
> drivers/gpu/drm/radeon/r600_blit.c | 31 +
> drivers/gpu/drm/radeon/r600_blit_kms.c | 785 ----
> drivers/gpu/drm/radeon/r600_blit_shaders.h | 1 -
> drivers/gpu/drm/radeon/r600_dpm.c | 258 +-
> drivers/gpu/drm/radeon/r600_dpm.h | 6 +
> drivers/gpu/drm/radeon/radeon.h | 140 +-
> drivers/gpu/drm/radeon/radeon_asic.c | 57 +-
> drivers/gpu/drm/radeon/radeon_asic.h | 61 +-
> drivers/gpu/drm/radeon/radeon_atombios.c | 122 +-
> drivers/gpu/drm/radeon/radeon_blit_common.h | 44 -
> drivers/gpu/drm/radeon/radeon_cs.c | 11 +-
> drivers/gpu/drm/radeon/radeon_device.c | 22 +-
> drivers/gpu/drm/radeon/radeon_drv.c | 4 +-
> drivers/gpu/drm/radeon/radeon_pm.c | 88 +-
> drivers/gpu/drm/radeon/radeon_ucode.h | 17 +
> drivers/gpu/drm/radeon/radeon_uvd.c | 58 +-
> drivers/gpu/drm/radeon/rv770.c | 9 +-
> drivers/gpu/drm/radeon/rv770_dpm.c | 10 +-
> drivers/gpu/drm/radeon/si.c | 180 +-
> drivers/gpu/drm/radeon/si_dpm.c | 29 +-
> drivers/gpu/drm/radeon/smu7.h | 170 +
> drivers/gpu/drm/radeon/smu7_discrete.h | 486 +++
> drivers/gpu/drm/radeon/smu7_fusion.h | 300 ++
> drivers/gpu/drm/radeon/sumo_dpm.c | 14 +
> drivers/gpu/drm/radeon/sumo_dpm.h | 3 +
> 49 files changed, 15075 insertions(+), 2890 deletions(-)
> create mode 100644 drivers/gpu/drm/radeon/ci_dpm.c
> create mode 100644 drivers/gpu/drm/radeon/ci_dpm.h
> create mode 100644 drivers/gpu/drm/radeon/ci_smc.c
> create mode 100644 drivers/gpu/drm/radeon/clearstate_ci.h
> delete mode 100644 drivers/gpu/drm/radeon/evergreen_blit_kms.c
> create mode 100644 drivers/gpu/drm/radeon/kv_dpm.c
> create mode 100644 drivers/gpu/drm/radeon/kv_dpm.h
> create mode 100644 drivers/gpu/drm/radeon/kv_smc.c
> create mode 100644 drivers/gpu/drm/radeon/pptable.h
> delete mode 100644 drivers/gpu/drm/radeon/r600_blit_kms.c
> delete mode 100644 drivers/gpu/drm/radeon/radeon_blit_common.h
> create mode 100644 drivers/gpu/drm/radeon/smu7.h
> create mode 100644 drivers/gpu/drm/radeon/smu7_discrete.h
> create mode 100644 drivers/gpu/drm/radeon/smu7_fusion.h
>
> --
> 1.8.3.1
>
^ permalink raw reply [flat|nested] 55+ messages in thread
end of thread, other threads:[~2013-08-13 21:58 UTC | newest]
Thread overview: 55+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-08-12 16:42 [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
2013-08-12 16:42 ` [PATCH 01/53] drm/radeon: switch r6xx+ to using CP DMA for the blit copy callback Alex Deucher
2013-08-12 16:42 ` [PATCH 02/53] drm/radeon/kms: remove r6xx+ blit copy routines Alex Deucher
2013-08-12 16:42 ` [PATCH 03/53] drm/radeon: add UVD->DPM helper function (v5) Alex Deucher
2013-08-12 16:42 ` [PATCH 04/53] drm/radeon/dpm: use multiple UVD power states (v3) Alex Deucher
2013-08-12 16:42 ` [PATCH 05/53] drm/radeon/dpm: rework thermal state handling Alex Deucher
2013-08-12 16:42 ` [PATCH 06/53] drm/radeon: default to 1024M gart size on rv770+ Alex Deucher
2013-08-12 16:42 ` [PATCH 07/53] drm/radeon/dpm: use performance state if no UVD state Alex Deucher
2013-08-12 16:43 ` [PATCH 08/53] drm/radeon/kms: fix up dce8 display watermark calc for dpm Alex Deucher
2013-08-12 16:43 ` [PATCH 09/53] drm/radeon/cik: implement some more atom helpers for DPM Alex Deucher
2013-08-12 16:43 ` [PATCH 10/53] drm/radeon: switch CIK to use radeon_ucode.h Alex Deucher
2013-08-12 16:43 ` [PATCH 11/53] drm/radeon/cik: add support for pcie gen1/2/3 switching Alex Deucher
2013-08-12 16:43 ` [PATCH 12/53] drm/radeon: add support for ASPM on CIK asics Alex Deucher
2013-08-12 16:43 ` [PATCH 13/53] drm/radeon/cik: restructure rlc setup Alex Deucher
2013-08-12 16:43 ` [PATCH 14/53] drm/radeon: clean up sumo_rlc_init() for code sharing Alex Deucher
2013-08-12 16:43 ` [PATCH 15/53] drm/radeon: convert SI,CIK to use sumo_rlc functions Alex Deucher
2013-08-12 16:43 ` [PATCH 16/53] drm/radeon: implement clock and power gating for CIK (v3) Alex Deucher
2013-08-12 16:43 ` [PATCH 17/53] drm/radeon: add indirect accessors for dift registers on CIK Alex Deucher
2013-08-12 16:43 ` [PATCH 18/53] drm/radeon/sumo add helper to go from vid7 to vid2 Alex Deucher
2013-08-12 16:43 ` [PATCH 19/53] drm/radeon: switch to pptable.h Alex Deucher
2013-08-12 16:43 ` [PATCH 20/53] drm/radeon: add structs to store uvd clock voltage deps Alex Deucher
2013-08-12 16:43 ` [PATCH 21/53] drm/radeon/cik: add rlc helpers for DPM Alex Deucher
2013-08-12 16:43 ` [PATCH 22/53] drm/radeon: add support for thermal controller on KB/KV Alex Deucher
2013-08-12 16:43 ` [PATCH 23/53] drm/radeon: add CI to r600_is_internal_thermal_sensor() Alex Deucher
2013-08-12 16:43 ` [PATCH 24/53] drm/radeon: add KB/KV to r600_is_internal_thermal_sensor Alex Deucher
2013-08-12 16:43 ` [PATCH 25/53] drm/radeon: add get_temperature() callbacks for CIK Alex Deucher
2013-08-12 16:43 ` [PATCH 26/53] drm/radeon: adjust si_dpm function for code sharing Alex Deucher
2013-08-12 16:43 ` [PATCH 27/53] drm/radeon/dpm: update cac leakage table parsing for CI Alex Deucher
2013-08-12 16:43 ` [PATCH 28/53] drm/radeon/dpm: add support for parsing the atom powertune table Alex Deucher
2013-08-12 16:43 ` [PATCH 29/53] drm/radeon/dpm: grab mvdd_dependency_on_mclk info from vbios Alex Deucher
2013-08-12 16:43 ` [PATCH 30/53] drm/radeon: add structs to store vce clock voltage deps Alex Deucher
2013-08-12 16:43 ` [PATCH 31/53] drm/radeon: add clock voltage dep tables for acp, samu Alex Deucher
2013-08-12 16:43 ` [PATCH 32/53] drm/radeon: parse the vce clock voltage deps table Alex Deucher
2013-08-12 16:43 ` [PATCH 33/53] drm/radeon: parse the uvd " Alex Deucher
2013-08-12 16:43 ` [PATCH 34/53] drm/radeon/dpm: clean up the extended table error pathes Alex Deucher
2013-08-12 16:43 ` [PATCH 35/53] drm/radeon: parse the samu clock voltage deps table Alex Deucher
2013-08-12 16:43 ` [PATCH 36/53] drm/radeon: parse the acp " Alex Deucher
2013-08-12 16:43 ` [PATCH 37/53] drm/radeon: add r600_get_pcie_lane_support helper Alex Deucher
2013-08-12 16:43 ` [PATCH 38/53] drm/radeon/dpm: add vce clocks to radeon_ps Alex Deucher
2013-08-12 16:43 ` [PATCH 39/53] drm/radeon/dpm: add a helper to encode pcie lane setting Alex Deucher
2013-08-12 16:43 ` [PATCH 40/53] drm/radeon/dpm: add helper to fetch the vrefresh of the current mode Alex Deucher
2013-08-12 16:43 ` [PATCH 41/53] drm/radeon/kms: add dpm support for KB/KV Alex Deucher
2013-08-12 16:43 ` [PATCH 42/53] drm/radeon: add dpm support for CI dGPUs (v2) Alex Deucher
2013-08-12 16:43 ` [PATCH 43/53] drm/radeon/dpm: add debugfs support for CI Alex Deucher
2013-08-12 16:43 ` [PATCH 44/53] drm/radeon/dpm: implement force performance level " Alex Deucher
2013-08-12 16:43 ` [PATCH 45/53] drm/radeon/dpm: implement vblank_too_short callback " Alex Deucher
2013-08-12 16:43 ` [PATCH 46/53] drm/radeon/dpm: add debugfs support for KB/KV Alex Deucher
2013-08-12 16:43 ` [PATCH 47/53] drm/radeon/dpm: implement force performance level " Alex Deucher
2013-08-12 16:43 ` [PATCH 48/53] drm/radeon/dpm: add new callback for powergating UVD (v3) Alex Deucher
2013-08-12 16:43 ` [PATCH 49/53] drm/radeon: restructure UVD code to handle UVD PG (v2) Alex Deucher
2013-08-12 16:43 ` [PATCH 50/53] drm/radeon/dpm: implement UVD powergating for KB/KV Alex Deucher
2013-08-12 16:43 ` [PATCH 51/53] drm/radeon/dpm: implement UVD powergating for CI Alex Deucher
2013-08-12 16:43 ` [PATCH 52/53] drm/radeon/dpm: add reclocking quirk for ASUS K70AF Alex Deucher
2013-08-12 16:43 ` [PATCH 53/53] drm/radeon/dpm: adjust the vblank time checks for eg, ni, si Alex Deucher
2013-08-13 21:58 ` [PATCH 00/53] Initial 3.12 radeon patches Alex Deucher
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.