* [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers
@ 2021-10-26 6:38 Vivek Kasireddy
2021-10-26 7:52 ` [Intel-gfx] ✓ Fi.CI.BAT: success for " Patchwork
` (8 more replies)
0 siblings, 9 replies; 22+ messages in thread
From: Vivek Kasireddy @ 2021-10-26 6:38 UTC (permalink / raw)
To: intel-gfx
Cc: Vivek Kasireddy, Maarten Lankhorst, Tvrtko Ursulin, Manasi Navare
On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
more framebuffers/scanout buffers results in only one that is mappable/
fenceable. Therefore, pageflipping between these 2 FBs where only one
is mappable/fenceable creates latencies large enough to miss alternate
vblanks thereby producing less optimal framerate.
This mainly happens because when i915_gem_object_pin_to_display_plane()
is called to pin one of the FB objs, the associated vma is identified
as misplaced and therefore i915_vma_unbind() is called which unbinds and
evicts it. This misplaced vma gets subseqently pinned only when
i915_gem_object_ggtt_pin_ww() is called without the mappable flag. This
results in a latency of ~10ms and happens every other vblank/repaint cycle.
Testcase:
Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
a frame ~7ms before the next vblank, the latencies seen between atomic
commit and flip event is 7, 24 (7 + 16.66), 7, 24..... suggesting that
it misses the vblank every other frame.
Here is the ftrace snippet that shows the source of the ~10ms latency:
i915_gem_object_pin_to_display_plane() {
0.102 us | i915_gem_object_set_cache_level();
i915_gem_object_ggtt_pin_ww() {
0.390 us | i915_vma_instance();
0.178 us | i915_vma_misplaced();
i915_vma_unbind() {
__i915_active_wait() {
0.082 us | i915_active_acquire_if_busy();
0.475 us | }
intel_runtime_pm_get() {
0.087 us | intel_runtime_pm_acquire();
0.259 us | }
__i915_active_wait() {
0.085 us | i915_active_acquire_if_busy();
0.240 us | }
__i915_vma_evict() {
ggtt_unbind_vma() {
gen8_ggtt_clear_range() {
10507.255 us | }
10507.689 us | }
10508.516 us | }
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Manasi Navare <manasi.d.navare@intel.com>
Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
---
drivers/gpu/drm/i915/display/intel_fb_pin.c | 11 +++++++++--
drivers/gpu/drm/i915/display/intel_overlay.c | 11 ++++++++---
drivers/gpu/drm/i915/gem/i915_gem_domain.c | 6 ++++--
drivers/gpu/drm/i915/gem/i915_gem_object.h | 3 ++-
4 files changed, 23 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c b/drivers/gpu/drm/i915/display/intel_fb_pin.c
index 3f77f3013584..53c156d9a9f9 100644
--- a/drivers/gpu/drm/i915/display/intel_fb_pin.c
+++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c
@@ -144,7 +144,7 @@ intel_pin_and_fence_fb_obj(struct drm_framebuffer *fb,
if (!ret) {
vma = i915_gem_object_pin_to_display_plane(obj, &ww, alignment,
- view, pinctl);
+ view, pinctl, uses_fence);
if (IS_ERR(vma)) {
ret = PTR_ERR(vma);
goto err_unpin;
@@ -218,9 +218,16 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state)
INTEL_INFO(dev_priv)->display.cursor_needs_physical;
if (!intel_fb_uses_dpt(fb)) {
+ struct intel_crtc *crtc = to_intel_crtc(plane_state->hw.crtc);
+ struct intel_crtc_state *crtc_state =
+ to_intel_crtc_state(crtc->base.state);
+ bool uses_fence = intel_plane_uses_fence(plane_state);
+ bool is_bigjoiner = crtc_state->bigjoiner ||
+ crtc_state->bigjoiner_slave;
+
vma = intel_pin_and_fence_fb_obj(fb, phys_cursor,
&plane_state->view.gtt,
- intel_plane_uses_fence(plane_state),
+ uses_fence && !is_bigjoiner,
&plane_state->flags);
if (IS_ERR(vma))
return PTR_ERR(vma);
diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c
index 7e3f5c6ca484..e9563b40b911 100644
--- a/drivers/gpu/drm/i915/display/intel_overlay.c
+++ b/drivers/gpu/drm/i915/display/intel_overlay.c
@@ -755,10 +755,14 @@ static u32 overlay_cmd_reg(struct drm_intel_overlay_put_image *params)
return cmd;
}
-static struct i915_vma *intel_overlay_pin_fb(struct drm_i915_gem_object *new_bo)
+static struct i915_vma *intel_overlay_pin_fb(struct drm_i915_gem_object *new_bo,
+ struct intel_overlay *overlay)
{
struct i915_gem_ww_ctx ww;
struct i915_vma *vma;
+ const struct intel_plane_state *plane_state =
+ to_intel_plane_state(overlay->crtc->base.primary->state);
+ bool uses_fence = intel_plane_uses_fence(plane_state);
int ret;
i915_gem_ww_ctx_init(&ww, true);
@@ -766,7 +770,8 @@ static struct i915_vma *intel_overlay_pin_fb(struct drm_i915_gem_object *new_bo)
ret = i915_gem_object_lock(new_bo, &ww);
if (!ret) {
vma = i915_gem_object_pin_to_display_plane(new_bo, &ww, 0,
- NULL, PIN_MAPPABLE);
+ NULL, PIN_MAPPABLE,
+ uses_fence);
ret = PTR_ERR_OR_ZERO(vma);
}
if (ret == -EDEADLK) {
@@ -802,7 +807,7 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay,
atomic_inc(&dev_priv->gpu_error.pending_fb_pin);
- vma = intel_overlay_pin_fb(new_bo);
+ vma = intel_overlay_pin_fb(new_bo, overlay);
if (IS_ERR(vma)) {
ret = PTR_ERR(vma);
goto out_pin_section;
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
index b684a62bf3b0..b9108f13cc49 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
@@ -375,7 +375,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
struct i915_gem_ww_ctx *ww,
u32 alignment,
const struct i915_ggtt_view *view,
- unsigned int flags)
+ unsigned int flags,
+ bool uses_fence)
{
struct drm_i915_private *i915 = to_i915(obj->base.dev);
struct i915_vma *vma;
@@ -411,7 +412,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
*/
vma = ERR_PTR(-ENOSPC);
if ((flags & PIN_MAPPABLE) == 0 &&
- (!view || view->type == I915_GGTT_VIEW_NORMAL))
+ (!view || view->type == I915_GGTT_VIEW_NORMAL) &&
+ uses_fence)
vma = i915_gem_object_ggtt_pin_ww(obj, ww, view, 0, alignment,
flags | PIN_MAPPABLE |
PIN_NONBLOCK);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index a5479ac7a4ad..e83564fa57ca 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -534,7 +534,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
struct i915_gem_ww_ctx *ww,
u32 alignment,
const struct i915_ggtt_view *view,
- unsigned int flags);
+ unsigned int flags,
+ bool uses_fence);
void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj);
void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj);
--
2.31.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers
2021-10-26 6:38 [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Vivek Kasireddy
@ 2021-10-26 7:52 ` Patchwork
2021-10-26 9:07 ` [Intel-gfx] [PATCH] " Ville Syrjälä
` (7 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Patchwork @ 2021-10-26 7:52 UTC (permalink / raw)
To: Vivek Kasireddy; +Cc: intel-gfx
[-- Attachment #1: Type: text/plain, Size: 3819 bytes --]
== Series Details ==
Series: drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers
URL : https://patchwork.freedesktop.org/series/96279/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_10788 -> Patchwork_21442
====================================================
Summary
-------
**SUCCESS**
No regressions found.
External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/index.html
Participating hosts (39 -> 35)
------------------------------
Missing (4): fi-ctg-p8600 fi-bsw-cyan bat-dg1-6 bat-adlp-4
Known issues
------------
Here are the changes found in Patchwork_21442 that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@amdgpu/amd_basic@query-info:
- fi-bsw-kefka: NOTRUN -> [SKIP][1] ([fdo#109271]) +17 similar issues
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/fi-bsw-kefka/igt@amdgpu/amd_basic@query-info.html
* igt@amdgpu/amd_basic@semaphore:
- fi-bdw-5557u: NOTRUN -> [SKIP][2] ([fdo#109271]) +27 similar issues
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/fi-bdw-5557u/igt@amdgpu/amd_basic@semaphore.html
* igt@debugfs_test@read_all_entries:
- fi-apl-guc: [PASS][3] -> [DMESG-WARN][4] ([i915#203])
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/fi-apl-guc/igt@debugfs_test@read_all_entries.html
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/fi-apl-guc/igt@debugfs_test@read_all_entries.html
* igt@gem_exec_suspend@basic-s0:
- fi-apl-guc: [PASS][5] -> [DMESG-WARN][6] ([i915#180] / [i915#203] / [i915#62])
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/fi-apl-guc/igt@gem_exec_suspend@basic-s0.html
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/fi-apl-guc/igt@gem_exec_suspend@basic-s0.html
* igt@kms_chamelium@dp-crc-fast:
- fi-bdw-5557u: NOTRUN -> [SKIP][7] ([fdo#109271] / [fdo#111827]) +8 similar issues
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/fi-bdw-5557u/igt@kms_chamelium@dp-crc-fast.html
#### Possible fixes ####
* igt@gem_exec_suspend@basic-s3:
- fi-bdw-5557u: [INCOMPLETE][8] -> [PASS][9]
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/fi-bdw-5557u/igt@gem_exec_suspend@basic-s3.html
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/fi-bdw-5557u/igt@gem_exec_suspend@basic-s3.html
* igt@i915_selftest@live@execlists:
- fi-bsw-kefka: [INCOMPLETE][10] ([i915#2940]) -> [PASS][11]
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/fi-bsw-kefka/igt@i915_selftest@live@execlists.html
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/fi-bsw-kefka/igt@i915_selftest@live@execlists.html
[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#111827]: https://bugs.freedesktop.org/show_bug.cgi?id=111827
[i915#180]: https://gitlab.freedesktop.org/drm/intel/issues/180
[i915#203]: https://gitlab.freedesktop.org/drm/intel/issues/203
[i915#2940]: https://gitlab.freedesktop.org/drm/intel/issues/2940
[i915#62]: https://gitlab.freedesktop.org/drm/intel/issues/62
Build changes
-------------
* Linux: CI_DRM_10788 -> Patchwork_21442
CI-20190529: 20190529
CI_DRM_10788: e5d0af998bc3f676208222233aa9a6acb06b3e01 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_6261: 0433f0d6063d8450af1e8518047d3679b9e5a6c1 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
Patchwork_21442: aef4429698244c63aaba5323f2bc1e59f08d5338 @ git://anongit.freedesktop.org/gfx-ci/linux
== Linux commits ==
aef442969824 drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/index.html
[-- Attachment #2: Type: text/html, Size: 4795 bytes --]
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers
2021-10-26 6:38 [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Vivek Kasireddy
2021-10-26 7:52 ` [Intel-gfx] ✓ Fi.CI.BAT: success for " Patchwork
@ 2021-10-26 9:07 ` Ville Syrjälä
2021-10-27 6:41 ` Kasireddy, Vivek
2021-10-26 9:33 ` [Intel-gfx] ✗ Fi.CI.IGT: failure for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Patchwork
` (6 subsequent siblings)
8 siblings, 1 reply; 22+ messages in thread
From: Ville Syrjälä @ 2021-10-26 9:07 UTC (permalink / raw)
To: Vivek Kasireddy
Cc: intel-gfx, Maarten Lankhorst, Tvrtko Ursulin, Manasi Navare
On Mon, Oct 25, 2021 at 11:38:11PM -0700, Vivek Kasireddy wrote:
> On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
> more framebuffers/scanout buffers results in only one that is mappable/
> fenceable. Therefore, pageflipping between these 2 FBs where only one
> is mappable/fenceable creates latencies large enough to miss alternate
> vblanks thereby producing less optimal framerate.
>
> This mainly happens because when i915_gem_object_pin_to_display_plane()
> is called to pin one of the FB objs, the associated vma is identified
> as misplaced and therefore i915_vma_unbind() is called which unbinds and
> evicts it. This misplaced vma gets subseqently pinned only when
> i915_gem_object_ggtt_pin_ww() is called without the mappable flag. This
> results in a latency of ~10ms and happens every other vblank/repaint cycle.
>
> Testcase:
> Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
> with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
> a frame ~7ms before the next vblank, the latencies seen between atomic
> commit and flip event is 7, 24 (7 + 16.66), 7, 24..... suggesting that
> it misses the vblank every other frame.
>
> Here is the ftrace snippet that shows the source of the ~10ms latency:
> i915_gem_object_pin_to_display_plane() {
> 0.102 us | i915_gem_object_set_cache_level();
> i915_gem_object_ggtt_pin_ww() {
> 0.390 us | i915_vma_instance();
> 0.178 us | i915_vma_misplaced();
> i915_vma_unbind() {
> __i915_active_wait() {
> 0.082 us | i915_active_acquire_if_busy();
> 0.475 us | }
> intel_runtime_pm_get() {
> 0.087 us | intel_runtime_pm_acquire();
> 0.259 us | }
> __i915_active_wait() {
> 0.085 us | i915_active_acquire_if_busy();
> 0.240 us | }
> __i915_vma_evict() {
> ggtt_unbind_vma() {
> gen8_ggtt_clear_range() {
> 10507.255 us | }
> 10507.689 us | }
> 10508.516 us | }
>
> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> Cc: Manasi Navare <manasi.d.navare@intel.com>
> Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
> ---
> drivers/gpu/drm/i915/display/intel_fb_pin.c | 11 +++++++++--
> drivers/gpu/drm/i915/display/intel_overlay.c | 11 ++++++++---
> drivers/gpu/drm/i915/gem/i915_gem_domain.c | 6 ++++--
> drivers/gpu/drm/i915/gem/i915_gem_object.h | 3 ++-
> 4 files changed, 23 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c b/drivers/gpu/drm/i915/display/intel_fb_pin.c
> index 3f77f3013584..53c156d9a9f9 100644
> --- a/drivers/gpu/drm/i915/display/intel_fb_pin.c
> +++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c
> @@ -144,7 +144,7 @@ intel_pin_and_fence_fb_obj(struct drm_framebuffer *fb,
>
> if (!ret) {
> vma = i915_gem_object_pin_to_display_plane(obj, &ww, alignment,
> - view, pinctl);
> + view, pinctl, uses_fence);
> if (IS_ERR(vma)) {
> ret = PTR_ERR(vma);
> goto err_unpin;
> @@ -218,9 +218,16 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state)
> INTEL_INFO(dev_priv)->display.cursor_needs_physical;
>
> if (!intel_fb_uses_dpt(fb)) {
> + struct intel_crtc *crtc = to_intel_crtc(plane_state->hw.crtc);
> + struct intel_crtc_state *crtc_state =
> + to_intel_crtc_state(crtc->base.state);
> + bool uses_fence = intel_plane_uses_fence(plane_state);
> + bool is_bigjoiner = crtc_state->bigjoiner ||
> + crtc_state->bigjoiner_slave;
Bigjoiner is just an implementation detail. It is not the cause of any
of this.
> +
> vma = intel_pin_and_fence_fb_obj(fb, phys_cursor,
> &plane_state->view.gtt,
> - intel_plane_uses_fence(plane_state),
> + uses_fence && !is_bigjoiner,
> &plane_state->flags);
> if (IS_ERR(vma))
> return PTR_ERR(vma);
> diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c
> index 7e3f5c6ca484..e9563b40b911 100644
> --- a/drivers/gpu/drm/i915/display/intel_overlay.c
> +++ b/drivers/gpu/drm/i915/display/intel_overlay.c
> @@ -755,10 +755,14 @@ static u32 overlay_cmd_reg(struct drm_intel_overlay_put_image *params)
> return cmd;
> }
>
> -static struct i915_vma *intel_overlay_pin_fb(struct drm_i915_gem_object *new_bo)
> +static struct i915_vma *intel_overlay_pin_fb(struct drm_i915_gem_object *new_bo,
> + struct intel_overlay *overlay)
> {
> struct i915_gem_ww_ctx ww;
> struct i915_vma *vma;
> + const struct intel_plane_state *plane_state =
> + to_intel_plane_state(overlay->crtc->base.primary->state);
> + bool uses_fence = intel_plane_uses_fence(plane_state);
> int ret;
>
> i915_gem_ww_ctx_init(&ww, true);
> @@ -766,7 +770,8 @@ static struct i915_vma *intel_overlay_pin_fb(struct drm_i915_gem_object *new_bo)
> ret = i915_gem_object_lock(new_bo, &ww);
> if (!ret) {
> vma = i915_gem_object_pin_to_display_plane(new_bo, &ww, 0,
> - NULL, PIN_MAPPABLE);
> + NULL, PIN_MAPPABLE,
> + uses_fence);
> ret = PTR_ERR_OR_ZERO(vma);
> }
> if (ret == -EDEADLK) {
> @@ -802,7 +807,7 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay,
>
> atomic_inc(&dev_priv->gpu_error.pending_fb_pin);
>
> - vma = intel_overlay_pin_fb(new_bo);
> + vma = intel_overlay_pin_fb(new_bo, overlay);
> if (IS_ERR(vma)) {
> ret = PTR_ERR(vma);
> goto out_pin_section;
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
> index b684a62bf3b0..b9108f13cc49 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
> @@ -375,7 +375,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
> struct i915_gem_ww_ctx *ww,
> u32 alignment,
> const struct i915_ggtt_view *view,
> - unsigned int flags)
> + unsigned int flags,
> + bool uses_fence)
> {
> struct drm_i915_private *i915 = to_i915(obj->base.dev);
> struct i915_vma *vma;
> @@ -411,7 +412,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
> */
> vma = ERR_PTR(-ENOSPC);
> if ((flags & PIN_MAPPABLE) == 0 &&
> - (!view || view->type == I915_GGTT_VIEW_NORMAL))
> + (!view || view->type == I915_GGTT_VIEW_NORMAL) &&
> + uses_fence)
> vma = i915_gem_object_ggtt_pin_ww(obj, ww, view, 0, alignment,
> flags | PIN_MAPPABLE |
> PIN_NONBLOCK);
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
> index a5479ac7a4ad..e83564fa57ca 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
> @@ -534,7 +534,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
> struct i915_gem_ww_ctx *ww,
> u32 alignment,
> const struct i915_ggtt_view *view,
> - unsigned int flags);
> + unsigned int flags,
> + bool uses_fence);
>
> void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj);
> void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj);
> --
> 2.31.1
--
Ville Syrjälä
Intel
^ permalink raw reply [flat|nested] 22+ messages in thread
* [Intel-gfx] ✗ Fi.CI.IGT: failure for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers
2021-10-26 6:38 [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Vivek Kasireddy
2021-10-26 7:52 ` [Intel-gfx] ✓ Fi.CI.BAT: success for " Patchwork
2021-10-26 9:07 ` [Intel-gfx] [PATCH] " Ville Syrjälä
@ 2021-10-26 9:33 ` Patchwork
2021-10-28 14:14 ` [Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev2) Patchwork
` (5 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Patchwork @ 2021-10-26 9:33 UTC (permalink / raw)
To: Vivek Kasireddy; +Cc: intel-gfx
[-- Attachment #1: Type: text/plain, Size: 30292 bytes --]
== Series Details ==
Series: drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers
URL : https://patchwork.freedesktop.org/series/96279/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_10788_full -> Patchwork_21442_full
====================================================
Summary
-------
**FAILURE**
Serious unknown changes coming with Patchwork_21442_full absolutely need to be
verified manually.
If you think the reported changes have nothing to do with the changes
introduced in Patchwork_21442_full, please notify your bug team to allow them
to document this new failure mode, which will reduce false positives in CI.
Participating hosts (10 -> 10)
------------------------------
No changes in participating hosts
Possible new issues
-------------------
Here are the unknown changes that may have been introduced in Patchwork_21442_full:
### IGT changes ###
#### Possible regressions ####
* igt@gem_eio@in-flight-suspend:
- shard-kbl: [PASS][1] -> [INCOMPLETE][2]
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-kbl1/igt@gem_eio@in-flight-suspend.html
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl4/igt@gem_eio@in-flight-suspend.html
* igt@gem_exec_gttfill@basic:
- shard-iclb: [PASS][3] -> [INCOMPLETE][4]
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-iclb6/igt@gem_exec_gttfill@basic.html
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb4/igt@gem_exec_gttfill@basic.html
Known issues
------------
Here are the changes found in Patchwork_21442_full that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@gem_create@create-massive:
- shard-skl: NOTRUN -> [DMESG-WARN][5] ([i915#3002])
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl3/igt@gem_create@create-massive.html
* igt@gem_ctx_persistence@legacy-engines-hang@blt:
- shard-skl: NOTRUN -> [SKIP][6] ([fdo#109271]) +267 similar issues
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl2/igt@gem_ctx_persistence@legacy-engines-hang@blt.html
* igt@gem_exec_fair@basic-none-solo@rcs0:
- shard-kbl: NOTRUN -> [FAIL][7] ([i915#2842])
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl6/igt@gem_exec_fair@basic-none-solo@rcs0.html
* igt@gem_exec_fair@basic-pace-share@rcs0:
- shard-tglb: [PASS][8] -> [FAIL][9] ([i915#2842])
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-tglb1/igt@gem_exec_fair@basic-pace-share@rcs0.html
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb6/igt@gem_exec_fair@basic-pace-share@rcs0.html
* igt@gem_exec_fair@basic-pace@vcs0:
- shard-glk: [PASS][10] -> [FAIL][11] ([i915#2842])
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-glk4/igt@gem_exec_fair@basic-pace@vcs0.html
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-glk5/igt@gem_exec_fair@basic-pace@vcs0.html
* igt@gem_exec_flush@basic-batch-kernel-default-cmd:
- shard-tglb: NOTRUN -> [SKIP][12] ([fdo#109313])
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@gem_exec_flush@basic-batch-kernel-default-cmd.html
* igt@gem_huc_copy@huc-copy:
- shard-apl: NOTRUN -> [SKIP][13] ([fdo#109271] / [i915#2190])
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-apl2/igt@gem_huc_copy@huc-copy.html
* igt@gem_pread@exhaustion:
- shard-apl: NOTRUN -> [WARN][14] ([i915#2658])
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-apl2/igt@gem_pread@exhaustion.html
* igt@gem_pwrite@basic-exhaustion:
- shard-skl: NOTRUN -> [WARN][15] ([i915#2658])
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl2/igt@gem_pwrite@basic-exhaustion.html
* igt@gem_pxp@protected-raw-src-copy-not-readible:
- shard-tglb: NOTRUN -> [SKIP][16] ([i915#4270])
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@gem_pxp@protected-raw-src-copy-not-readible.html
* igt@gem_render_copy@x-tiled-to-vebox-yf-tiled:
- shard-iclb: NOTRUN -> [SKIP][17] ([i915#768])
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb3/igt@gem_render_copy@x-tiled-to-vebox-yf-tiled.html
* igt@gem_userptr_blits@dmabuf-sync:
- shard-skl: NOTRUN -> [SKIP][18] ([fdo#109271] / [i915#3323])
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl10/igt@gem_userptr_blits@dmabuf-sync.html
* igt@gem_userptr_blits@input-checking:
- shard-kbl: NOTRUN -> [DMESG-WARN][19] ([i915#3002])
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl7/igt@gem_userptr_blits@input-checking.html
* igt@gem_userptr_blits@vma-merge:
- shard-skl: NOTRUN -> [FAIL][20] ([i915#3318])
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl4/igt@gem_userptr_blits@vma-merge.html
* igt@gen7_exec_parse@basic-rejected:
- shard-tglb: NOTRUN -> [SKIP][21] ([fdo#109289]) +1 similar issue
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@gen7_exec_parse@basic-rejected.html
* igt@gen9_exec_parse@allowed-single:
- shard-tglb: NOTRUN -> [SKIP][22] ([i915#2856]) +3 similar issues
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb2/igt@gen9_exec_parse@allowed-single.html
- shard-iclb: NOTRUN -> [SKIP][23] ([i915#2856])
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb3/igt@gen9_exec_parse@allowed-single.html
* igt@i915_pm_dc@dc6-psr:
- shard-iclb: [PASS][24] -> [FAIL][25] ([i915#454])
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-iclb7/igt@i915_pm_dc@dc6-psr.html
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb4/igt@i915_pm_dc@dc6-psr.html
- shard-tglb: NOTRUN -> [FAIL][26] ([i915#454])
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@i915_pm_dc@dc6-psr.html
* igt@i915_pm_rpm@dpms-non-lpsp:
- shard-tglb: NOTRUN -> [SKIP][27] ([fdo#111644] / [i915#1397] / [i915#2411]) +1 similar issue
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb8/igt@i915_pm_rpm@dpms-non-lpsp.html
* igt@i915_selftest@live@gt_lrc:
- shard-tglb: NOTRUN -> [DMESG-FAIL][28] ([i915#2373])
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb7/igt@i915_selftest@live@gt_lrc.html
* igt@i915_selftest@live@gt_pm:
- shard-tglb: NOTRUN -> [DMESG-FAIL][29] ([i915#1759])
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb7/igt@i915_selftest@live@gt_pm.html
* igt@kms_addfb_basic@invalid-smem-bo-on-discrete:
- shard-tglb: NOTRUN -> [SKIP][30] ([i915#3826])
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@kms_addfb_basic@invalid-smem-bo-on-discrete.html
* igt@kms_atomic_transition@plane-all-modeset-transition-fencing:
- shard-tglb: NOTRUN -> [SKIP][31] ([i915#1769])
[31]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb7/igt@kms_atomic_transition@plane-all-modeset-transition-fencing.html
* igt@kms_big_fb@x-tiled-8bpp-rotate-90:
- shard-tglb: NOTRUN -> [SKIP][32] ([fdo#111614])
[32]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@kms_big_fb@x-tiled-8bpp-rotate-90.html
* igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-0-hflip:
- shard-skl: NOTRUN -> [SKIP][33] ([fdo#109271] / [i915#3777]) +1 similar issue
[33]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl10/igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-0-hflip.html
* igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-hflip:
- shard-kbl: NOTRUN -> [SKIP][34] ([fdo#109271] / [i915#3777]) +1 similar issue
[34]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl4/igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-hflip.html
* igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-0-async-flip:
- shard-skl: NOTRUN -> [FAIL][35] ([i915#3763])
[35]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl2/igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-0-async-flip.html
* igt@kms_big_fb@yf-tiled-addfb:
- shard-tglb: NOTRUN -> [SKIP][36] ([fdo#111615]) +2 similar issues
[36]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb8/igt@kms_big_fb@yf-tiled-addfb.html
* igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-0-async-flip:
- shard-skl: NOTRUN -> [FAIL][37] ([i915#3743])
[37]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl1/igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-0-async-flip.html
* igt@kms_ccs@pipe-a-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc:
- shard-apl: NOTRUN -> [SKIP][38] ([fdo#109271] / [i915#3886]) +1 similar issue
[38]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-apl2/igt@kms_ccs@pipe-a-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc.html
* igt@kms_ccs@pipe-b-bad-aux-stride-y_tiled_gen12_rc_ccs_cc:
- shard-kbl: NOTRUN -> [SKIP][39] ([fdo#109271] / [i915#3886]) +2 similar issues
[39]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl7/igt@kms_ccs@pipe-b-bad-aux-stride-y_tiled_gen12_rc_ccs_cc.html
* igt@kms_ccs@pipe-b-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc:
- shard-skl: NOTRUN -> [SKIP][40] ([fdo#109271] / [i915#3886]) +13 similar issues
[40]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl10/igt@kms_ccs@pipe-b-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc.html
* igt@kms_ccs@pipe-c-crc-primary-basic-y_tiled_gen12_rc_ccs_cc:
- shard-iclb: NOTRUN -> [SKIP][41] ([fdo#109278] / [i915#3886])
[41]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb3/igt@kms_ccs@pipe-c-crc-primary-basic-y_tiled_gen12_rc_ccs_cc.html
* igt@kms_ccs@pipe-c-missing-ccs-buffer-y_tiled_gen12_mc_ccs:
- shard-tglb: NOTRUN -> [SKIP][42] ([i915#3689] / [i915#3886])
[42]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@kms_ccs@pipe-c-missing-ccs-buffer-y_tiled_gen12_mc_ccs.html
* igt@kms_ccs@pipe-d-missing-ccs-buffer-y_tiled_ccs:
- shard-tglb: NOTRUN -> [SKIP][43] ([i915#3689]) +7 similar issues
[43]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb7/igt@kms_ccs@pipe-d-missing-ccs-buffer-y_tiled_ccs.html
* igt@kms_chamelium@hdmi-aspect-ratio:
- shard-skl: NOTRUN -> [SKIP][44] ([fdo#109271] / [fdo#111827]) +22 similar issues
[44]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl4/igt@kms_chamelium@hdmi-aspect-ratio.html
* igt@kms_chamelium@hdmi-frame-dump:
- shard-iclb: NOTRUN -> [SKIP][45] ([fdo#109284] / [fdo#111827])
[45]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb3/igt@kms_chamelium@hdmi-frame-dump.html
* igt@kms_chamelium@hdmi-mode-timings:
- shard-kbl: NOTRUN -> [SKIP][46] ([fdo#109271] / [fdo#111827]) +5 similar issues
[46]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl4/igt@kms_chamelium@hdmi-mode-timings.html
* igt@kms_chamelium@vga-frame-dump:
- shard-tglb: NOTRUN -> [SKIP][47] ([fdo#109284] / [fdo#111827]) +6 similar issues
[47]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb8/igt@kms_chamelium@vga-frame-dump.html
* igt@kms_color@pipe-d-legacy-gamma:
- shard-apl: NOTRUN -> [SKIP][48] ([fdo#109271]) +37 similar issues
[48]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-apl2/igt@kms_color@pipe-d-legacy-gamma.html
* igt@kms_color_chamelium@pipe-d-ctm-0-75:
- shard-apl: NOTRUN -> [SKIP][49] ([fdo#109271] / [fdo#111827]) +2 similar issues
[49]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-apl2/igt@kms_color_chamelium@pipe-d-ctm-0-75.html
* igt@kms_concurrent@pipe-a:
- shard-skl: [PASS][50] -> [DMESG-WARN][51] ([i915#1982]) +2 similar issues
[50]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-skl10/igt@kms_concurrent@pipe-a.html
[51]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl2/igt@kms_concurrent@pipe-a.html
* igt@kms_content_protection@dp-mst-lic-type-1:
- shard-tglb: NOTRUN -> [SKIP][52] ([i915#3116])
[52]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb8/igt@kms_content_protection@dp-mst-lic-type-1.html
* igt@kms_content_protection@lic:
- shard-tglb: NOTRUN -> [SKIP][53] ([fdo#111828])
[53]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@kms_content_protection@lic.html
* igt@kms_cursor_crc@pipe-a-cursor-suspend:
- shard-tglb: NOTRUN -> [INCOMPLETE][54] ([i915#2411] / [i915#2828] / [i915#456])
[54]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb7/igt@kms_cursor_crc@pipe-a-cursor-suspend.html
* igt@kms_cursor_crc@pipe-b-cursor-32x32-offscreen:
- shard-tglb: NOTRUN -> [SKIP][55] ([i915#3319]) +1 similar issue
[55]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb7/igt@kms_cursor_crc@pipe-b-cursor-32x32-offscreen.html
* igt@kms_cursor_crc@pipe-c-cursor-512x512-random:
- shard-tglb: NOTRUN -> [SKIP][56] ([fdo#109279] / [i915#3359])
[56]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@kms_cursor_crc@pipe-c-cursor-512x512-random.html
* igt@kms_cursor_crc@pipe-d-cursor-256x256-onscreen:
- shard-kbl: NOTRUN -> [SKIP][57] ([fdo#109271]) +55 similar issues
[57]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl7/igt@kms_cursor_crc@pipe-d-cursor-256x256-onscreen.html
* igt@kms_cursor_crc@pipe-d-cursor-32x10-onscreen:
- shard-tglb: NOTRUN -> [SKIP][58] ([i915#3359]) +2 similar issues
[58]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb8/igt@kms_cursor_crc@pipe-d-cursor-32x10-onscreen.html
* igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions:
- shard-skl: NOTRUN -> [FAIL][59] ([i915#2346])
[59]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl10/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions.html
* igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size:
- shard-skl: [PASS][60] -> [FAIL][61] ([i915#2346] / [i915#533])
[60]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-skl8/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
[61]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl1/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
* igt@kms_cursor_legacy@pipe-d-forked-bo:
- shard-iclb: NOTRUN -> [SKIP][62] ([fdo#109278]) +2 similar issues
[62]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb1/igt@kms_cursor_legacy@pipe-d-forked-bo.html
* igt@kms_fbcon_fbt@fbc-suspend:
- shard-apl: [PASS][63] -> [INCOMPLETE][64] ([i915#180] / [i915#1982])
[63]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-apl7/igt@kms_fbcon_fbt@fbc-suspend.html
[64]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-apl4/igt@kms_fbcon_fbt@fbc-suspend.html
* igt@kms_flip@flip-vs-absolute-wf_vblank-interruptible@b-edp1:
- shard-skl: [PASS][65] -> [FAIL][66] ([i915#2122]) +1 similar issue
[65]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-skl10/igt@kms_flip@flip-vs-absolute-wf_vblank-interruptible@b-edp1.html
[66]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl2/igt@kms_flip@flip-vs-absolute-wf_vblank-interruptible@b-edp1.html
* igt@kms_flip@flip-vs-expired-vblank@a-edp1:
- shard-skl: NOTRUN -> [FAIL][67] ([i915#2122])
[67]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl2/igt@kms_flip@flip-vs-expired-vblank@a-edp1.html
* igt@kms_flip@flip-vs-expired-vblank@b-edp1:
- shard-skl: NOTRUN -> [FAIL][68] ([i915#79])
[68]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl2/igt@kms_flip@flip-vs-expired-vblank@b-edp1.html
* igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs:
- shard-skl: NOTRUN -> [SKIP][69] ([fdo#109271] / [i915#2672])
[69]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl1/igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs.html
* igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-draw-pwrite:
- shard-iclb: NOTRUN -> [SKIP][70] ([fdo#109280]) +1 similar issue
[70]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb3/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-draw-pwrite.html
* igt@kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-move:
- shard-tglb: NOTRUN -> [SKIP][71] ([fdo#111825]) +25 similar issues
[71]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-move.html
* igt@kms_hdr@bpc-switch-dpms:
- shard-skl: [PASS][72] -> [FAIL][73] ([i915#1188])
[72]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-skl4/igt@kms_hdr@bpc-switch-dpms.html
[73]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl8/igt@kms_hdr@bpc-switch-dpms.html
* igt@kms_hdr@bpc-switch-suspend:
- shard-kbl: [PASS][74] -> [DMESG-WARN][75] ([i915#180]) +6 similar issues
[74]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-kbl4/igt@kms_hdr@bpc-switch-suspend.html
[75]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl7/igt@kms_hdr@bpc-switch-suspend.html
- shard-apl: [PASS][76] -> [DMESG-WARN][77] ([i915#180]) +4 similar issues
[76]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-apl4/igt@kms_hdr@bpc-switch-suspend.html
[77]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-apl4/igt@kms_hdr@bpc-switch-suspend.html
* igt@kms_plane_alpha_blend@pipe-a-constant-alpha-max:
- shard-skl: NOTRUN -> [FAIL][78] ([fdo#108145] / [i915#265]) +4 similar issues
[78]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl2/igt@kms_plane_alpha_blend@pipe-a-constant-alpha-max.html
* igt@kms_plane_alpha_blend@pipe-c-alpha-opaque-fb:
- shard-kbl: NOTRUN -> [FAIL][79] ([fdo#108145] / [i915#265])
[79]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl7/igt@kms_plane_alpha_blend@pipe-c-alpha-opaque-fb.html
* igt@kms_plane_alpha_blend@pipe-c-alpha-transparent-fb:
- shard-kbl: NOTRUN -> [FAIL][80] ([i915#265])
[80]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl6/igt@kms_plane_alpha_blend@pipe-c-alpha-transparent-fb.html
* igt@kms_plane_lowres@pipe-b-tiling-none:
- shard-tglb: NOTRUN -> [SKIP][81] ([i915#3536])
[81]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb8/igt@kms_plane_lowres@pipe-b-tiling-none.html
* igt@kms_plane_multiple@atomic-pipe-b-tiling-yf:
- shard-tglb: NOTRUN -> [SKIP][82] ([fdo#112054])
[82]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb7/igt@kms_plane_multiple@atomic-pipe-b-tiling-yf.html
* igt@kms_plane_scaling@scaler-with-clipping-clamping@pipe-c-scaler-with-clipping-clamping:
- shard-skl: NOTRUN -> [SKIP][83] ([fdo#109271] / [i915#2733])
[83]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl2/igt@kms_plane_scaling@scaler-with-clipping-clamping@pipe-c-scaler-with-clipping-clamping.html
* igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-2:
- shard-skl: NOTRUN -> [SKIP][84] ([fdo#109271] / [i915#658]) +5 similar issues
[84]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl1/igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-2.html
* igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-4:
- shard-tglb: NOTRUN -> [SKIP][85] ([i915#2920]) +1 similar issue
[85]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb8/igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-4.html
* igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-5:
- shard-kbl: NOTRUN -> [SKIP][86] ([fdo#109271] / [i915#658]) +1 similar issue
[86]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl7/igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-5.html
* igt@kms_psr2_sf@plane-move-sf-dmg-area-2:
- shard-iclb: NOTRUN -> [SKIP][87] ([i915#658])
[87]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb3/igt@kms_psr2_sf@plane-move-sf-dmg-area-2.html
* igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-3:
- shard-apl: NOTRUN -> [SKIP][88] ([fdo#109271] / [i915#658])
[88]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-apl2/igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-3.html
* igt@kms_psr@psr2_primary_page_flip:
- shard-iclb: [PASS][89] -> [SKIP][90] ([fdo#109441]) +2 similar issues
[89]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-iclb2/igt@kms_psr@psr2_primary_page_flip.html
[90]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb7/igt@kms_psr@psr2_primary_page_flip.html
* igt@kms_psr@psr2_sprite_render:
- shard-tglb: NOTRUN -> [FAIL][91] ([i915#132] / [i915#3467])
[91]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb7/igt@kms_psr@psr2_sprite_render.html
* igt@kms_psr@psr2_suspend:
- shard-iclb: NOTRUN -> [SKIP][92] ([fdo#109441])
[92]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb3/igt@kms_psr@psr2_suspend.html
* igt@kms_vblank@pipe-a-ts-continuation-suspend:
- shard-kbl: [PASS][93] -> [DMESG-WARN][94] ([i915#180] / [i915#295])
[93]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-kbl1/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
[94]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl3/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
- shard-apl: [PASS][95] -> [DMESG-WARN][96] ([i915#180] / [i915#295])
[95]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-apl7/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
[96]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-apl6/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
* igt@kms_vblank@pipe-d-wait-idle:
- shard-skl: NOTRUN -> [SKIP][97] ([fdo#109271] / [i915#533]) +3 similar issues
[97]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl1/igt@kms_vblank@pipe-d-wait-idle.html
* igt@kms_vrr@flipline:
- shard-tglb: NOTRUN -> [SKIP][98] ([fdo#109502])
[98]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@kms_vrr@flipline.html
* igt@kms_writeback@writeback-check-output:
- shard-skl: NOTRUN -> [SKIP][99] ([fdo#109271] / [i915#2437])
[99]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl1/igt@kms_writeback@writeback-check-output.html
* igt@kms_writeback@writeback-pixel-formats:
- shard-tglb: NOTRUN -> [SKIP][100] ([i915#2437])
[100]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@kms_writeback@writeback-pixel-formats.html
* igt@nouveau_crc@pipe-a-ctx-flip-detection:
- shard-tglb: NOTRUN -> [SKIP][101] ([i915#2530])
[101]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb8/igt@nouveau_crc@pipe-a-ctx-flip-detection.html
* igt@perf@blocking:
- shard-skl: [PASS][102] -> [FAIL][103] ([i915#1542])
[102]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-skl2/igt@perf@blocking.html
[103]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl9/igt@perf@blocking.html
* igt@perf@non-zero-reason:
- shard-skl: [PASS][104] -> [TIMEOUT][105] ([i915#2836])
[104]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-skl2/igt@perf@non-zero-reason.html
[105]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl3/igt@perf@non-zero-reason.html
* igt@prime_nv_test@nv_write_i915_cpu_mmap_read:
- shard-tglb: NOTRUN -> [SKIP][106] ([fdo#109291]) +2 similar issues
[106]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb1/igt@prime_nv_test@nv_write_i915_cpu_mmap_read.html
* igt@sysfs_clients@fair-7:
- shard-iclb: NOTRUN -> [SKIP][107] ([i915#2994])
[107]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb3/igt@sysfs_clients@fair-7.html
* igt@sysfs_clients@sema-10:
- shard-skl: NOTRUN -> [SKIP][108] ([fdo#109271] / [i915#2994]) +3 similar issues
[108]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl4/igt@sysfs_clients@sema-10.html
* igt@sysfs_clients@split-25:
- shard-kbl: NOTRUN -> [SKIP][109] ([fdo#109271] / [i915#2994])
[109]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl4/igt@sysfs_clients@split-25.html
- shard-tglb: NOTRUN -> [SKIP][110] ([i915#2994])
[110]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb8/igt@sysfs_clients@split-25.html
* igt@sysfs_heartbeat_interval@mixed@bcs0:
- shard-skl: [PASS][111] -> [WARN][112] ([i915#4055])
[111]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-skl10/igt@sysfs_heartbeat_interval@mixed@bcs0.html
[112]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl2/igt@sysfs_heartbeat_interval@mixed@bcs0.html
* igt@sysfs_heartbeat_interval@mixed@rcs0:
- shard-skl: [PASS][113] -> [FAIL][114] ([i915#1731]) +2 similar issues
[113]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-skl10/igt@sysfs_heartbeat_interval@mixed@rcs0.html
[114]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl2/igt@sysfs_heartbeat_interval@mixed@rcs0.html
#### Possible fixes ####
* igt@feature_discovery@psr2:
- shard-iclb: [SKIP][115] ([i915#658]) -> [PASS][116]
[115]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-iclb8/igt@feature_discovery@psr2.html
[116]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-iclb2/igt@feature_discovery@psr2.html
* igt@gem_exec_fair@basic-none@rcs0:
- shard-glk: [FAIL][117] ([i915#2842]) -> [PASS][118] +1 similar issue
[117]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-glk4/igt@gem_exec_fair@basic-none@rcs0.html
[118]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-glk8/igt@gem_exec_fair@basic-none@rcs0.html
* igt@i915_module_load@reload-no-display:
- shard-tglb: [DMESG-WARN][119] ([i915#2867]) -> [PASS][120]
[119]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-tglb7/igt@i915_module_load@reload-no-display.html
[120]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb7/igt@i915_module_load@reload-no-display.html
* igt@kms_async_flips@alternate-sync-async-flip:
- shard-skl: [FAIL][121] ([i915#2521]) -> [PASS][122]
[121]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-skl2/igt@kms_async_flips@alternate-sync-async-flip.html
[122]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-skl9/igt@kms_async_flips@alternate-sync-async-flip.html
* igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size:
- shard-glk: [FAIL][123] ([i915#2346] / [i915#533]) -> [PASS][124]
[123]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-glk4/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
[124]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-glk8/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
- shard-apl: [FAIL][125] ([i915#2346] / [i915#533]) -> [PASS][126]
[125]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-apl7/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
[126]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-apl4/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
* igt@kms_fbcon_fbt@fbc-suspend:
- shard-kbl: [INCOMPLETE][127] ([i915#180] / [i915#636]) -> [PASS][128]
[127]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-kbl1/igt@kms_fbcon_fbt@fbc-suspend.html
[128]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl7/igt@kms_fbcon_fbt@fbc-suspend.html
* igt@kms_frontbuffer_tracking@fbc-suspend:
- shard-kbl: [DMESG-WARN][129] ([i915#180]) -> [PASS][130] +2 similar issues
[129]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-kbl1/igt@kms_frontbuffer_tracking@fbc-suspend.html
[130]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-kbl4/igt@kms_frontbuffer_tracking@fbc-suspend.html
* igt@kms_frontbuffer_tracking@fbcpsr-suspend:
- shard-tglb: [INCOMPLETE][131] ([i915#2411] / [i915#456]) -> [PASS][132]
[131]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-tglb7/igt@kms_frontbuffer_tracking@fbcpsr-suspend.html
[132]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/shard-tglb8/igt@kms_frontbuffer_tracking@fbcpsr-suspend.html
* igt@kms_hdr@bpc-switch-suspend:
- shard-skl: [FAIL][133] ([i915#1188]) -> [PASS][134]
[133]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10788/shard-skl8/igt@kms_hdr@bpc-switch-suspend.html
[
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21442/index.html
[-- Attachment #2: Type: text/html, Size: 33780 bytes --]
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers
2021-10-26 9:07 ` [Intel-gfx] [PATCH] " Ville Syrjälä
@ 2021-10-27 6:41 ` Kasireddy, Vivek
2021-10-27 8:27 ` Ville Syrjälä
0 siblings, 1 reply; 22+ messages in thread
From: Kasireddy, Vivek @ 2021-10-27 6:41 UTC (permalink / raw)
To: Ville Syrjälä
Cc: intel-gfx, Maarten Lankhorst, Tvrtko Ursulin, Navare, Manasi D
Hi Ville,
>
> On Mon, Oct 25, 2021 at 11:38:11PM -0700, Vivek Kasireddy wrote:
> > On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
> > more framebuffers/scanout buffers results in only one that is mappable/
> > fenceable. Therefore, pageflipping between these 2 FBs where only one
> > is mappable/fenceable creates latencies large enough to miss alternate
> > vblanks thereby producing less optimal framerate.
> >
> > This mainly happens because when i915_gem_object_pin_to_display_plane()
> > is called to pin one of the FB objs, the associated vma is identified
> > as misplaced and therefore i915_vma_unbind() is called which unbinds and
> > evicts it. This misplaced vma gets subseqently pinned only when
> > i915_gem_object_ggtt_pin_ww() is called without the mappable flag. This
> > results in a latency of ~10ms and happens every other vblank/repaint cycle.
> >
> > Testcase:
> > Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
> > with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
> > a frame ~7ms before the next vblank, the latencies seen between atomic
> > commit and flip event is 7, 24 (7 + 16.66), 7, 24..... suggesting that
> > it misses the vblank every other frame.
> >
> > Here is the ftrace snippet that shows the source of the ~10ms latency:
> > i915_gem_object_pin_to_display_plane() {
> > 0.102 us | i915_gem_object_set_cache_level();
> > i915_gem_object_ggtt_pin_ww() {
> > 0.390 us | i915_vma_instance();
> > 0.178 us | i915_vma_misplaced();
> > i915_vma_unbind() {
> > __i915_active_wait() {
> > 0.082 us | i915_active_acquire_if_busy();
> > 0.475 us | }
> > intel_runtime_pm_get() {
> > 0.087 us | intel_runtime_pm_acquire();
> > 0.259 us | }
> > __i915_active_wait() {
> > 0.085 us | i915_active_acquire_if_busy();
> > 0.240 us | }
> > __i915_vma_evict() {
> > ggtt_unbind_vma() {
> > gen8_ggtt_clear_range() {
> > 10507.255 us | }
> > 10507.689 us | }
> > 10508.516 us | }
> >
> > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> > Cc: Manasi Navare <manasi.d.navare@intel.com>
> > Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
> > ---
> > drivers/gpu/drm/i915/display/intel_fb_pin.c | 11 +++++++++--
> > drivers/gpu/drm/i915/display/intel_overlay.c | 11 ++++++++---
> > drivers/gpu/drm/i915/gem/i915_gem_domain.c | 6 ++++--
> > drivers/gpu/drm/i915/gem/i915_gem_object.h | 3 ++-
> > 4 files changed, 23 insertions(+), 8 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c
> b/drivers/gpu/drm/i915/display/intel_fb_pin.c
> > index 3f77f3013584..53c156d9a9f9 100644
> > --- a/drivers/gpu/drm/i915/display/intel_fb_pin.c
> > +++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c
> > @@ -144,7 +144,7 @@ intel_pin_and_fence_fb_obj(struct drm_framebuffer *fb,
> >
> > if (!ret) {
> > vma = i915_gem_object_pin_to_display_plane(obj, &ww, alignment,
> > - view, pinctl);
> > + view, pinctl, uses_fence);
> > if (IS_ERR(vma)) {
> > ret = PTR_ERR(vma);
> > goto err_unpin;
> > @@ -218,9 +218,16 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state)
> > INTEL_INFO(dev_priv)->display.cursor_needs_physical;
> >
> > if (!intel_fb_uses_dpt(fb)) {
> > + struct intel_crtc *crtc = to_intel_crtc(plane_state->hw.crtc);
> > + struct intel_crtc_state *crtc_state =
> > + to_intel_crtc_state(crtc->base.state);
> > + bool uses_fence = intel_plane_uses_fence(plane_state);
> > + bool is_bigjoiner = crtc_state->bigjoiner ||
> > + crtc_state->bigjoiner_slave;
>
> Bigjoiner is just an implementation detail. It is not the cause of any
> of this.
[Kasireddy, Vivek] Right, bigjoiner/8K is just exposing the underlying issue; which is
that sometimes, large objects/scanout buffers that fail range overflow checks and thus
are not mappable/fenceable keep getting evicted and reinserted leading to latencies.
I guess I could mark an object/vma as permanently un-mappable/un-fenceable and try
not to map it subsequently but this would result in one scanout buffer that is mappable
but others that are not. Would this be acceptable -- assuming we are pageflipping
between multiple FBs?
Any ideas on solving this issue cleanly?
Thanks,
Vivek
>
> > +
> > vma = intel_pin_and_fence_fb_obj(fb, phys_cursor,
> > &plane_state->view.gtt,
> > - intel_plane_uses_fence(plane_state),
> > + uses_fence && !is_bigjoiner,
> > &plane_state->flags);
> > if (IS_ERR(vma))
> > return PTR_ERR(vma);
> > diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c
> b/drivers/gpu/drm/i915/display/intel_overlay.c
> > index 7e3f5c6ca484..e9563b40b911 100644
> > --- a/drivers/gpu/drm/i915/display/intel_overlay.c
> > +++ b/drivers/gpu/drm/i915/display/intel_overlay.c
> > @@ -755,10 +755,14 @@ static u32 overlay_cmd_reg(struct
> drm_intel_overlay_put_image *params)
> > return cmd;
> > }
> >
> > -static struct i915_vma *intel_overlay_pin_fb(struct drm_i915_gem_object *new_bo)
> > +static struct i915_vma *intel_overlay_pin_fb(struct drm_i915_gem_object *new_bo,
> > + struct intel_overlay *overlay)
> > {
> > struct i915_gem_ww_ctx ww;
> > struct i915_vma *vma;
> > + const struct intel_plane_state *plane_state =
> > + to_intel_plane_state(overlay->crtc->base.primary->state);
> > + bool uses_fence = intel_plane_uses_fence(plane_state);
> > int ret;
> >
> > i915_gem_ww_ctx_init(&ww, true);
> > @@ -766,7 +770,8 @@ static struct i915_vma *intel_overlay_pin_fb(struct
> drm_i915_gem_object *new_bo)
> > ret = i915_gem_object_lock(new_bo, &ww);
> > if (!ret) {
> > vma = i915_gem_object_pin_to_display_plane(new_bo, &ww, 0,
> > - NULL, PIN_MAPPABLE);
> > + NULL, PIN_MAPPABLE,
> > + uses_fence);
> > ret = PTR_ERR_OR_ZERO(vma);
> > }
> > if (ret == -EDEADLK) {
> > @@ -802,7 +807,7 @@ static int intel_overlay_do_put_image(struct intel_overlay
> *overlay,
> >
> > atomic_inc(&dev_priv->gpu_error.pending_fb_pin);
> >
> > - vma = intel_overlay_pin_fb(new_bo);
> > + vma = intel_overlay_pin_fb(new_bo, overlay);
> > if (IS_ERR(vma)) {
> > ret = PTR_ERR(vma);
> > goto out_pin_section;
> > diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c
> b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
> > index b684a62bf3b0..b9108f13cc49 100644
> > --- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c
> > +++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
> > @@ -375,7 +375,8 @@ i915_gem_object_pin_to_display_plane(struct
> drm_i915_gem_object *obj,
> > struct i915_gem_ww_ctx *ww,
> > u32 alignment,
> > const struct i915_ggtt_view *view,
> > - unsigned int flags)
> > + unsigned int flags,
> > + bool uses_fence)
> > {
> > struct drm_i915_private *i915 = to_i915(obj->base.dev);
> > struct i915_vma *vma;
> > @@ -411,7 +412,8 @@ i915_gem_object_pin_to_display_plane(struct
> drm_i915_gem_object *obj,
> > */
> > vma = ERR_PTR(-ENOSPC);
> > if ((flags & PIN_MAPPABLE) == 0 &&
> > - (!view || view->type == I915_GGTT_VIEW_NORMAL))
> > + (!view || view->type == I915_GGTT_VIEW_NORMAL) &&
> > + uses_fence)
> > vma = i915_gem_object_ggtt_pin_ww(obj, ww, view, 0, alignment,
> > flags | PIN_MAPPABLE |
> > PIN_NONBLOCK);
> > diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h
> b/drivers/gpu/drm/i915/gem/i915_gem_object.h
> > index a5479ac7a4ad..e83564fa57ca 100644
> > --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
> > +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
> > @@ -534,7 +534,8 @@ i915_gem_object_pin_to_display_plane(struct
> drm_i915_gem_object *obj,
> > struct i915_gem_ww_ctx *ww,
> > u32 alignment,
> > const struct i915_ggtt_view *view,
> > - unsigned int flags);
> > + unsigned int flags,
> > + bool uses_fence);
> >
> > void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj);
> > void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj);
> > --
> > 2.31.1
>
> --
> Ville Syrjälä
> Intel
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers
2021-10-27 6:41 ` Kasireddy, Vivek
@ 2021-10-27 8:27 ` Ville Syrjälä
2021-10-28 8:04 ` [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large " Vivek Kasireddy
0 siblings, 1 reply; 22+ messages in thread
From: Ville Syrjälä @ 2021-10-27 8:27 UTC (permalink / raw)
To: Kasireddy, Vivek
Cc: intel-gfx, Maarten Lankhorst, Tvrtko Ursulin, Navare, Manasi D
On Wed, Oct 27, 2021 at 06:41:25AM +0000, Kasireddy, Vivek wrote:
> Hi Ville,
>
> >
> > On Mon, Oct 25, 2021 at 11:38:11PM -0700, Vivek Kasireddy wrote:
> > > On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
> > > more framebuffers/scanout buffers results in only one that is mappable/
> > > fenceable. Therefore, pageflipping between these 2 FBs where only one
> > > is mappable/fenceable creates latencies large enough to miss alternate
> > > vblanks thereby producing less optimal framerate.
> > >
> > > This mainly happens because when i915_gem_object_pin_to_display_plane()
> > > is called to pin one of the FB objs, the associated vma is identified
> > > as misplaced and therefore i915_vma_unbind() is called which unbinds and
> > > evicts it. This misplaced vma gets subseqently pinned only when
> > > i915_gem_object_ggtt_pin_ww() is called without the mappable flag. This
> > > results in a latency of ~10ms and happens every other vblank/repaint cycle.
> > >
> > > Testcase:
> > > Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
> > > with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
> > > a frame ~7ms before the next vblank, the latencies seen between atomic
> > > commit and flip event is 7, 24 (7 + 16.66), 7, 24..... suggesting that
> > > it misses the vblank every other frame.
> > >
> > > Here is the ftrace snippet that shows the source of the ~10ms latency:
> > > i915_gem_object_pin_to_display_plane() {
> > > 0.102 us | i915_gem_object_set_cache_level();
> > > i915_gem_object_ggtt_pin_ww() {
> > > 0.390 us | i915_vma_instance();
> > > 0.178 us | i915_vma_misplaced();
> > > i915_vma_unbind() {
> > > __i915_active_wait() {
> > > 0.082 us | i915_active_acquire_if_busy();
> > > 0.475 us | }
> > > intel_runtime_pm_get() {
> > > 0.087 us | intel_runtime_pm_acquire();
> > > 0.259 us | }
> > > __i915_active_wait() {
> > > 0.085 us | i915_active_acquire_if_busy();
> > > 0.240 us | }
> > > __i915_vma_evict() {
> > > ggtt_unbind_vma() {
> > > gen8_ggtt_clear_range() {
> > > 10507.255 us | }
> > > 10507.689 us | }
> > > 10508.516 us | }
> > >
> > > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > > Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> > > Cc: Manasi Navare <manasi.d.navare@intel.com>
> > > Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
> > > ---
> > > drivers/gpu/drm/i915/display/intel_fb_pin.c | 11 +++++++++--
> > > drivers/gpu/drm/i915/display/intel_overlay.c | 11 ++++++++---
> > > drivers/gpu/drm/i915/gem/i915_gem_domain.c | 6 ++++--
> > > drivers/gpu/drm/i915/gem/i915_gem_object.h | 3 ++-
> > > 4 files changed, 23 insertions(+), 8 deletions(-)
> > >
> > > diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c
> > b/drivers/gpu/drm/i915/display/intel_fb_pin.c
> > > index 3f77f3013584..53c156d9a9f9 100644
> > > --- a/drivers/gpu/drm/i915/display/intel_fb_pin.c
> > > +++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c
> > > @@ -144,7 +144,7 @@ intel_pin_and_fence_fb_obj(struct drm_framebuffer *fb,
> > >
> > > if (!ret) {
> > > vma = i915_gem_object_pin_to_display_plane(obj, &ww, alignment,
> > > - view, pinctl);
> > > + view, pinctl, uses_fence);
> > > if (IS_ERR(vma)) {
> > > ret = PTR_ERR(vma);
> > > goto err_unpin;
> > > @@ -218,9 +218,16 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state)
> > > INTEL_INFO(dev_priv)->display.cursor_needs_physical;
> > >
> > > if (!intel_fb_uses_dpt(fb)) {
> > > + struct intel_crtc *crtc = to_intel_crtc(plane_state->hw.crtc);
> > > + struct intel_crtc_state *crtc_state =
> > > + to_intel_crtc_state(crtc->base.state);
> > > + bool uses_fence = intel_plane_uses_fence(plane_state);
> > > + bool is_bigjoiner = crtc_state->bigjoiner ||
> > > + crtc_state->bigjoiner_slave;
> >
> > Bigjoiner is just an implementation detail. It is not the cause of any
> > of this.
> [Kasireddy, Vivek] Right, bigjoiner/8K is just exposing the underlying issue; which is
> that sometimes, large objects/scanout buffers that fail range overflow checks and thus
> are not mappable/fenceable keep getting evicted and reinserted leading to latencies.
> I guess I could mark an object/vma as permanently un-mappable/un-fenceable and try
> not to map it subsequently but this would result in one scanout buffer that is mappable
> but others that are not. Would this be acceptable -- assuming we are pageflipping
> between multiple FBs?
> Any ideas on solving this issue cleanly?
We might just consider skipping PIN_MAPPABLE if the vma is too big.
What the specific defition of "too big" would be I'm not sure.
--
Ville Syrjälä
Intel
^ permalink raw reply [flat|nested] 22+ messages in thread
* [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers
2021-10-27 8:27 ` Ville Syrjälä
@ 2021-10-28 8:04 ` Vivek Kasireddy
2021-10-28 12:53 ` Ville Syrjälä
0 siblings, 1 reply; 22+ messages in thread
From: Vivek Kasireddy @ 2021-10-28 8:04 UTC (permalink / raw)
To: intel-gfx
Cc: Vivek Kasireddy, Ville Syrjälä,
Maarten Lankhorst, Tvrtko Ursulin, Manasi Navare
On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
more framebuffers/scanout buffers results in only one that is mappable/
fenceable. Therefore, pageflipping between these 2 FBs where only one
is mappable/fenceable creates latencies large enough to miss alternate
vblanks thereby producing less optimal framerate.
This mainly happens because when i915_gem_object_pin_to_display_plane()
is called to pin one of the FB objs, the associated vma is identified
as misplaced and therefore i915_vma_unbind() is called which unbinds and
evicts it. This misplaced vma gets subseqently pinned only when
i915_gem_object_ggtt_pin_ww() is called without the mappable flag. This
results in a latency of ~10ms and happens every other vblank/repaint cycle.
Testcase:
Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
a frame ~7ms before the next vblank, the latencies seen between atomic
commit and flip event are 7, 24 (7 + 16.66), 7, 24..... suggesting that
it misses the vblank every other frame.
Here is the ftrace snippet that shows the source of the ~10ms latency:
i915_gem_object_pin_to_display_plane() {
0.102 us | i915_gem_object_set_cache_level();
i915_gem_object_ggtt_pin_ww() {
0.390 us | i915_vma_instance();
0.178 us | i915_vma_misplaced();
i915_vma_unbind() {
__i915_active_wait() {
0.082 us | i915_active_acquire_if_busy();
0.475 us | }
intel_runtime_pm_get() {
0.087 us | intel_runtime_pm_acquire();
0.259 us | }
__i915_active_wait() {
0.085 us | i915_active_acquire_if_busy();
0.240 us | }
__i915_vma_evict() {
ggtt_unbind_vma() {
gen8_ggtt_clear_range() {
10507.255 us | }
10507.689 us | }
10508.516 us | }
v2: Instead of using bigjoiner checks, determine whether a scanout
buffer is too big by checking to see if it is possible to map
two of them into the ggtt.
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Manasi Navare <manasi.d.navare@intel.com>
Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
---
drivers/gpu/drm/i915/i915_gem.c | 48 ++++++++++++++++++++++++++-------
drivers/gpu/drm/i915/i915_vma.c | 2 +-
2 files changed, 40 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 981e383d1a5d..0050c7e4bb51 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -866,6 +866,44 @@ static void discard_ggtt_vma(struct i915_vma *vma)
spin_unlock(&obj->vma.lock);
}
+static bool i915_gem_obj_too_big(struct drm_i915_gem_object *obj)
+{
+ struct drm_i915_private *i915 = to_i915(obj->base.dev);
+ struct i915_ggtt *ggtt = &i915->ggtt;
+ struct drm_mm_node *hole;
+ u64 hole_start, hole_end;
+ u64 fence_size;
+
+ /*
+ * If the required space is larger than the available
+ * aperture, we will not able to find a slot for the
+ * object and unbinding the object now will be in
+ * vain. Worse, doing so may cause us to ping-pong
+ * the object in and out of the Global GTT and
+ * waste a lot of cycles under the mutex.
+ */
+ if (obj->base.size > ggtt->mappable_end)
+ return true;
+
+ fence_size = i915_gem_fence_size(i915, obj->base.size,
+ i915_gem_object_get_tiling(obj),
+ i915_gem_object_get_stride(obj));
+
+ /*
+ * Assuming this object is a large scanout buffer, we try to find
+ * out if there is room to map at-least two of them. There could
+ * be space available to map one but to be consistent, we try to
+ * avoid mapping/fencing any of them.
+ */
+ drm_mm_for_each_hole(hole, &ggtt->vm.mm, hole_start, hole_end) {
+ if (hole_end - hole_start > 2 * fence_size &&
+ hole_start + 2 * fence_size < ggtt->mappable_end)
+ return false;
+ }
+
+ return true;
+}
+
struct i915_vma *
i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
struct i915_gem_ww_ctx *ww,
@@ -879,15 +917,7 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
if (flags & PIN_MAPPABLE &&
(!view || view->type == I915_GGTT_VIEW_NORMAL)) {
- /*
- * If the required space is larger than the available
- * aperture, we will not able to find a slot for the
- * object and unbinding the object now will be in
- * vain. Worse, doing so may cause us to ping-pong
- * the object in and out of the Global GTT and
- * waste a lot of cycles under the mutex.
- */
- if (obj->base.size > ggtt->mappable_end)
+ if (i915_gem_obj_too_big(obj))
return ERR_PTR(-E2BIG);
/*
diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
index 90546fa58fc1..551644dbfa8a 100644
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -977,7 +977,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
if (err)
goto err_active;
- if (i915_is_ggtt(vma->vm))
+ if (i915_is_ggtt(vma->vm) && flags & PIN_MAPPABLE)
__i915_vma_set_map_and_fenceable(vma);
}
--
2.31.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* Re: [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers
2021-10-28 8:04 ` [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large " Vivek Kasireddy
@ 2021-10-28 12:53 ` Ville Syrjälä
2021-10-29 7:43 ` [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v3) Vivek Kasireddy
0 siblings, 1 reply; 22+ messages in thread
From: Ville Syrjälä @ 2021-10-28 12:53 UTC (permalink / raw)
To: Vivek Kasireddy
Cc: intel-gfx, Maarten Lankhorst, Tvrtko Ursulin, Manasi Navare
On Thu, Oct 28, 2021 at 01:04:23AM -0700, Vivek Kasireddy wrote:
> On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
> more framebuffers/scanout buffers results in only one that is mappable/
> fenceable. Therefore, pageflipping between these 2 FBs where only one
> is mappable/fenceable creates latencies large enough to miss alternate
> vblanks thereby producing less optimal framerate.
>
> This mainly happens because when i915_gem_object_pin_to_display_plane()
> is called to pin one of the FB objs, the associated vma is identified
> as misplaced and therefore i915_vma_unbind() is called which unbinds and
> evicts it. This misplaced vma gets subseqently pinned only when
> i915_gem_object_ggtt_pin_ww() is called without the mappable flag. This
> results in a latency of ~10ms and happens every other vblank/repaint cycle.
>
> Testcase:
> Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
> with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
> a frame ~7ms before the next vblank, the latencies seen between atomic
> commit and flip event are 7, 24 (7 + 16.66), 7, 24..... suggesting that
> it misses the vblank every other frame.
>
> Here is the ftrace snippet that shows the source of the ~10ms latency:
> i915_gem_object_pin_to_display_plane() {
> 0.102 us | i915_gem_object_set_cache_level();
> i915_gem_object_ggtt_pin_ww() {
> 0.390 us | i915_vma_instance();
> 0.178 us | i915_vma_misplaced();
> i915_vma_unbind() {
> __i915_active_wait() {
> 0.082 us | i915_active_acquire_if_busy();
> 0.475 us | }
> intel_runtime_pm_get() {
> 0.087 us | intel_runtime_pm_acquire();
> 0.259 us | }
> __i915_active_wait() {
> 0.085 us | i915_active_acquire_if_busy();
> 0.240 us | }
> __i915_vma_evict() {
> ggtt_unbind_vma() {
> gen8_ggtt_clear_range() {
> 10507.255 us | }
> 10507.689 us | }
> 10508.516 us | }
>
> v2: Instead of using bigjoiner checks, determine whether a scanout
> buffer is too big by checking to see if it is possible to map
> two of them into the ggtt.
>
> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> Cc: Manasi Navare <manasi.d.navare@intel.com>
> Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
> ---
> drivers/gpu/drm/i915/i915_gem.c | 48 ++++++++++++++++++++++++++-------
> drivers/gpu/drm/i915/i915_vma.c | 2 +-
> 2 files changed, 40 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> index 981e383d1a5d..0050c7e4bb51 100644
> --- a/drivers/gpu/drm/i915/i915_gem.c
> +++ b/drivers/gpu/drm/i915/i915_gem.c
> @@ -866,6 +866,44 @@ static void discard_ggtt_vma(struct i915_vma *vma)
> spin_unlock(&obj->vma.lock);
> }
>
> +static bool i915_gem_obj_too_big(struct drm_i915_gem_object *obj)
> +{
> + struct drm_i915_private *i915 = to_i915(obj->base.dev);
> + struct i915_ggtt *ggtt = &i915->ggtt;
> + struct drm_mm_node *hole;
> + u64 hole_start, hole_end;
> + u64 fence_size;
> +
> + /*
> + * If the required space is larger than the available
> + * aperture, we will not able to find a slot for the
> + * object and unbinding the object now will be in
> + * vain. Worse, doing so may cause us to ping-pong
> + * the object in and out of the Global GTT and
> + * waste a lot of cycles under the mutex.
> + */
> + if (obj->base.size > ggtt->mappable_end)
> + return true;
> +
> + fence_size = i915_gem_fence_size(i915, obj->base.size,
> + i915_gem_object_get_tiling(obj),
> + i915_gem_object_get_stride(obj));
> +
> + /*
> + * Assuming this object is a large scanout buffer, we try to find
> + * out if there is room to map at-least two of them. There could
> + * be space available to map one but to be consistent, we try to
> + * avoid mapping/fencing any of them.
> + */
> + drm_mm_for_each_hole(hole, &ggtt->vm.mm, hole_start, hole_end) {
> + if (hole_end - hole_start > 2 * fence_size &&
> + hole_start + 2 * fence_size < ggtt->mappable_end)
> + return false;
> + }
Looking for a hole twice the size seems a bit weird. Would make more sense
to check how many of these vmas could we fit in to any of the holes.
This also doesn't seem to handle any alignment constraints.
> +
> + return true;
> +}
> +
> struct i915_vma *
> i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
> struct i915_gem_ww_ctx *ww,
> @@ -879,15 +917,7 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
>
> if (flags & PIN_MAPPABLE &&
> (!view || view->type == I915_GGTT_VIEW_NORMAL)) {
> - /*
> - * If the required space is larger than the available
> - * aperture, we will not able to find a slot for the
> - * object and unbinding the object now will be in
> - * vain. Worse, doing so may cause us to ping-pong
> - * the object in and out of the Global GTT and
> - * waste a lot of cycles under the mutex.
> - */
> - if (obj->base.size > ggtt->mappable_end)
> + if (i915_gem_obj_too_big(obj))
Doing this uncoditionally seems wrong. Some platforms always use
PIN_MAPPABLE.
> return ERR_PTR(-E2BIG);
>
> /*
There is already code around these parts that tries to limit this
ping-pong:
/*
* If NONBLOCK is set the caller is optimistically
* trying to cache the full object within the mappable
* aperture, and *must* have a fallback in place for
* situations where we cannot bind the object. We
* can be a little more lax here and use the fallback
* more often to avoid costly migrations of ourselves
* and other objects within the aperture.
*
* Half-the-aperture is used as a simple heuristic.
* More interesting would to do search for a free
* block prior to making the commitment to unbind.
* That caters for the self-harm case, and with a
* little more heuristics (e.g. NOFAULT, NOEVICT)
* we could try to minimise harm to others.
*/
if (flags & PIN_NONBLOCK &&
obj->base.size > ggtt->mappable_end / 2)
return ERR_PTR(-ENOSPC);
Looks like you're pretty much trying to implement what tha comment says.
> diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
> index 90546fa58fc1..551644dbfa8a 100644
> --- a/drivers/gpu/drm/i915/i915_vma.c
> +++ b/drivers/gpu/drm/i915/i915_vma.c
> @@ -977,7 +977,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
> if (err)
> goto err_active;
>
> - if (i915_is_ggtt(vma->vm))
> + if (i915_is_ggtt(vma->vm) && flags & PIN_MAPPABLE)
> __i915_vma_set_map_and_fenceable(vma);
> }
>
> --
> 2.31.1
--
Ville Syrjälä
Intel
^ permalink raw reply [flat|nested] 22+ messages in thread
* [Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev2)
2021-10-26 6:38 [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Vivek Kasireddy
` (2 preceding siblings ...)
2021-10-26 9:33 ` [Intel-gfx] ✗ Fi.CI.IGT: failure for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Patchwork
@ 2021-10-28 14:14 ` Patchwork
2021-10-29 10:36 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev3) Patchwork
` (4 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Patchwork @ 2021-10-28 14:14 UTC (permalink / raw)
To: Vivek Kasireddy; +Cc: intel-gfx
[-- Attachment #1: Type: text/plain, Size: 9497 bytes --]
== Series Details ==
Series: drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev2)
URL : https://patchwork.freedesktop.org/series/96279/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_10807 -> Patchwork_21473
====================================================
Summary
-------
**FAILURE**
Serious unknown changes coming with Patchwork_21473 absolutely need to be
verified manually.
If you think the reported changes have nothing to do with the changes
introduced in Patchwork_21473, please notify your bug team to allow them
to document this new failure mode, which will reduce false positives in CI.
External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/index.html
Participating hosts (28 -> 31)
------------------------------
Additional (4): fi-kbl-soraka fi-bsw-kefka fi-bsw-nick fi-bsw-n3050
Missing (1): fi-pnv-d510
Possible new issues
-------------------
Here are the unknown changes that may have been introduced in Patchwork_21473:
### IGT changes ###
#### Possible regressions ####
* igt@gem_exec_gttfill@basic:
- fi-apl-guc: [PASS][1] -> [INCOMPLETE][2]
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10807/fi-apl-guc/igt@gem_exec_gttfill@basic.html
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-apl-guc/igt@gem_exec_gttfill@basic.html
- fi-ilk-650: [PASS][3] -> [DMESG-FAIL][4]
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10807/fi-ilk-650/igt@gem_exec_gttfill@basic.html
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-ilk-650/igt@gem_exec_gttfill@basic.html
* igt@gem_exec_parallel@engines@userptr:
- fi-bsw-nick: NOTRUN -> [INCOMPLETE][5]
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-bsw-nick/igt@gem_exec_parallel@engines@userptr.html
* igt@i915_selftest@live@coherency:
- fi-cfl-8109u: [PASS][6] -> [DMESG-FAIL][7]
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10807/fi-cfl-8109u/igt@i915_selftest@live@coherency.html
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-cfl-8109u/igt@i915_selftest@live@coherency.html
- fi-bdw-5557u: NOTRUN -> [DMESG-FAIL][8]
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-bdw-5557u/igt@i915_selftest@live@coherency.html
- fi-snb-2600: [PASS][9] -> [DMESG-FAIL][10]
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10807/fi-snb-2600/igt@i915_selftest@live@coherency.html
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-snb-2600/igt@i915_selftest@live@coherency.html
- fi-bsw-kefka: NOTRUN -> [DMESG-FAIL][11]
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-bsw-kefka/igt@i915_selftest@live@coherency.html
- fi-glk-dsi: [PASS][12] -> [DMESG-FAIL][13]
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10807/fi-glk-dsi/igt@i915_selftest@live@coherency.html
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-glk-dsi/igt@i915_selftest@live@coherency.html
#### Suppressed ####
The following results come from untrusted machines, tests, or statuses.
They do not affect the overall result.
* igt@i915_selftest@live@coherency:
- {fi-tgl-dsi}: [PASS][14] -> [DMESG-FAIL][15]
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10807/fi-tgl-dsi/igt@i915_selftest@live@coherency.html
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-tgl-dsi/igt@i915_selftest@live@coherency.html
- {fi-jsl-1}: [PASS][16] -> [DMESG-FAIL][17]
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10807/fi-jsl-1/igt@i915_selftest@live@coherency.html
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-jsl-1/igt@i915_selftest@live@coherency.html
Known issues
------------
Here are the changes found in Patchwork_21473 that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@amdgpu/amd_basic@query-info:
- fi-bsw-kefka: NOTRUN -> [SKIP][18] ([fdo#109271]) +32 similar issues
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-bsw-kefka/igt@amdgpu/amd_basic@query-info.html
* igt@amdgpu/amd_basic@semaphore:
- fi-bdw-5557u: NOTRUN -> [SKIP][19] ([fdo#109271]) +27 similar issues
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-bdw-5557u/igt@amdgpu/amd_basic@semaphore.html
* igt@fbdev@info:
- fi-bsw-nick: NOTRUN -> [SKIP][20] ([fdo#109271]) +5 similar issues
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-bsw-nick/igt@fbdev@info.html
* igt@gem_exec_fence@basic-busy@bcs0:
- fi-kbl-soraka: NOTRUN -> [SKIP][21] ([fdo#109271]) +8 similar issues
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-kbl-soraka/igt@gem_exec_fence@basic-busy@bcs0.html
* igt@gem_exec_suspend@basic-s3:
- fi-tgl-1115g4: [PASS][22] -> [FAIL][23] ([i915#1888])
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10807/fi-tgl-1115g4/igt@gem_exec_suspend@basic-s3.html
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-tgl-1115g4/igt@gem_exec_suspend@basic-s3.html
* igt@gem_huc_copy@huc-copy:
- fi-kbl-soraka: NOTRUN -> [SKIP][24] ([fdo#109271] / [i915#2190])
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-kbl-soraka/igt@gem_huc_copy@huc-copy.html
* igt@i915_selftest@live@gt_pm:
- fi-kbl-soraka: NOTRUN -> [DMESG-FAIL][25] ([i915#1886] / [i915#2291])
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-kbl-soraka/igt@i915_selftest@live@gt_pm.html
* igt@kms_chamelium@common-hpd-after-suspend:
- fi-kbl-soraka: NOTRUN -> [SKIP][26] ([fdo#109271] / [fdo#111827]) +8 similar issues
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-kbl-soraka/igt@kms_chamelium@common-hpd-after-suspend.html
* igt@kms_chamelium@dp-crc-fast:
- fi-bdw-5557u: NOTRUN -> [SKIP][27] ([fdo#109271] / [fdo#111827]) +8 similar issues
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-bdw-5557u/igt@kms_chamelium@dp-crc-fast.html
* igt@kms_chamelium@hdmi-crc-fast:
- fi-bsw-n3050: NOTRUN -> [SKIP][28] ([fdo#109271] / [fdo#111827]) +8 similar issues
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-bsw-n3050/igt@kms_chamelium@hdmi-crc-fast.html
* igt@kms_chamelium@hdmi-edid-read:
- fi-bsw-kefka: NOTRUN -> [SKIP][29] ([fdo#109271] / [fdo#111827]) +8 similar issues
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-bsw-kefka/igt@kms_chamelium@hdmi-edid-read.html
* igt@kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-d:
- fi-kbl-soraka: NOTRUN -> [SKIP][30] ([fdo#109271] / [i915#533])
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-kbl-soraka/igt@kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-d.html
* igt@kms_pipe_crc_basic@hang-read-crc-pipe-a:
- fi-bsw-n3050: NOTRUN -> [SKIP][31] ([fdo#109271]) +42 similar issues
[31]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-bsw-n3050/igt@kms_pipe_crc_basic@hang-read-crc-pipe-a.html
* igt@runner@aborted:
- fi-ilk-650: NOTRUN -> [FAIL][32] ([i915#4312])
[32]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-ilk-650/igt@runner@aborted.html
- fi-bsw-nick: NOTRUN -> [FAIL][33] ([i915#2722] / [i915#4312])
[33]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-bsw-nick/igt@runner@aborted.html
#### Warnings ####
* igt@runner@aborted:
- fi-apl-guc: [FAIL][34] ([i915#2426] / [i915#3363] / [i915#4312]) -> [FAIL][35] ([i915#2722] / [i915#3363] / [i915#4312])
[34]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10807/fi-apl-guc/igt@runner@aborted.html
[35]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/fi-apl-guc/igt@runner@aborted.html
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#111827]: https://bugs.freedesktop.org/show_bug.cgi?id=111827
[i915#1886]: https://gitlab.freedesktop.org/drm/intel/issues/1886
[i915#1888]: https://gitlab.freedesktop.org/drm/intel/issues/1888
[i915#2190]: https://gitlab.freedesktop.org/drm/intel/issues/2190
[i915#2291]: https://gitlab.freedesktop.org/drm/intel/issues/2291
[i915#2426]: https://gitlab.freedesktop.org/drm/intel/issues/2426
[i915#2722]: https://gitlab.freedesktop.org/drm/intel/issues/2722
[i915#3363]: https://gitlab.freedesktop.org/drm/intel/issues/3363
[i915#4312]: https://gitlab.freedesktop.org/drm/intel/issues/4312
[i915#533]: https://gitlab.freedesktop.org/drm/intel/issues/533
Build changes
-------------
* Linux: CI_DRM_10807 -> Patchwork_21473
CI-20190529: 20190529
CI_DRM_10807: 0cd0ae01802327aedf3ab152e91376b59443a099 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_6262: d1c793b26e31cc6ae3f9fa3239805a9bbcc749fb @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
Patchwork_21473: 9087ca3d310acc9a2f90ff473bfa9058b4f044f4 @ git://anongit.freedesktop.org/gfx-ci/linux
== Linux commits ==
9087ca3d310a drm/i915/gem: Don't try to map and fence large scanout buffers
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21473/index.html
[-- Attachment #2: Type: text/html, Size: 11938 bytes --]
^ permalink raw reply [flat|nested] 22+ messages in thread
* [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v3)
2021-10-28 12:53 ` Ville Syrjälä
@ 2021-10-29 7:43 ` Vivek Kasireddy
2021-11-01 10:13 ` Tvrtko Ursulin
2021-11-01 10:21 ` Ville Syrjälä
0 siblings, 2 replies; 22+ messages in thread
From: Vivek Kasireddy @ 2021-10-29 7:43 UTC (permalink / raw)
To: intel-gfx
Cc: Vivek Kasireddy, Ville Syrjälä,
Maarten Lankhorst, Tvrtko Ursulin, Manasi Navare
On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
more framebuffers/scanout buffers results in only one that is mappable/
fenceable. Therefore, pageflipping between these 2 FBs where only one
is mappable/fenceable creates latencies large enough to miss alternate
vblanks thereby producing less optimal framerate.
This mainly happens because when i915_gem_object_pin_to_display_plane()
is called to pin one of the FB objs, the associated vma is identified
as misplaced and therefore i915_vma_unbind() is called which unbinds and
evicts it. This misplaced vma gets subseqently pinned only when
i915_gem_object_ggtt_pin_ww() is called without the mappable flag. This
results in a latency of ~10ms and happens every other vblank/repaint cycle.
Testcase:
Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
a frame ~7ms before the next vblank, the latencies seen between atomic
commit and flip event are 7, 24 (7 + 16.66), 7, 24..... suggesting that
it misses the vblank every other frame.
Here is the ftrace snippet that shows the source of the ~10ms latency:
i915_gem_object_pin_to_display_plane() {
0.102 us | i915_gem_object_set_cache_level();
i915_gem_object_ggtt_pin_ww() {
0.390 us | i915_vma_instance();
0.178 us | i915_vma_misplaced();
i915_vma_unbind() {
__i915_active_wait() {
0.082 us | i915_active_acquire_if_busy();
0.475 us | }
intel_runtime_pm_get() {
0.087 us | intel_runtime_pm_acquire();
0.259 us | }
__i915_active_wait() {
0.085 us | i915_active_acquire_if_busy();
0.240 us | }
__i915_vma_evict() {
ggtt_unbind_vma() {
gen8_ggtt_clear_range() {
10507.255 us | }
10507.689 us | }
10508.516 us | }
v2: Instead of using bigjoiner checks, determine whether a scanout
buffer is too big by checking to see if it is possible to map
two of them into the ggtt.
v3 (Ville):
- Count how many fb objects can be fit into the available holes
instead of checking for a hole twice the object size.
- Take alignment constraints into account.
- Limit this large scanout buffer check to >= Gen 12 platforms.
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Manasi Navare <manasi.d.navare@intel.com>
Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
---
drivers/gpu/drm/i915/i915_gem.c | 65 ++++++++++++++++++++++++++++-----
drivers/gpu/drm/i915/i915_vma.c | 2 +-
2 files changed, 57 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 981e383d1a5d..761dc385fbfc 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -866,6 +866,61 @@ static void discard_ggtt_vma(struct i915_vma *vma)
spin_unlock(&obj->vma.lock);
}
+static bool i915_gem_obj_too_big(struct drm_i915_gem_object *obj,
+ u64 alignment)
+{
+ struct drm_i915_private *i915 = to_i915(obj->base.dev);
+ struct i915_ggtt *ggtt = &i915->ggtt;
+ struct drm_mm_node *hole;
+ u64 hole_start, hole_end, start, end;
+ u64 fence_size, fence_alignment;
+ unsigned int count = 0;
+
+ /*
+ * If the required space is larger than the available
+ * aperture, we will not able to find a slot for the
+ * object and unbinding the object now will be in
+ * vain. Worse, doing so may cause us to ping-pong
+ * the object in and out of the Global GTT and
+ * waste a lot of cycles under the mutex.
+ */
+ if (obj->base.size > ggtt->mappable_end)
+ return true;
+
+ if (HAS_GMCH(i915) || DISPLAY_VER(i915) < 11 ||
+ !i915_gem_object_is_framebuffer(obj))
+ return false;
+
+ fence_size = i915_gem_fence_size(i915, obj->base.size,
+ i915_gem_object_get_tiling(obj),
+ i915_gem_object_get_stride(obj));
+ fence_alignment = i915_gem_fence_alignment(i915, obj->base.size,
+ i915_gem_object_get_tiling(obj),
+ i915_gem_object_get_stride(obj));
+ alignment = max_t(u64, alignment, fence_alignment);
+
+ /*
+ * Assuming this object is a large scanout buffer, we try to find
+ * out if there is room to map at-least two of them. There could
+ * be space available to map one but to be consistent, we try to
+ * avoid mapping/fencing any of them.
+ */
+ drm_mm_for_each_hole(hole, &ggtt->vm.mm, hole_start, hole_end) {
+ do {
+ start = round_up(hole_start, alignment);
+ end = min_t(u64, hole_end, ggtt->mappable_end);
+
+ if (range_overflows(start, fence_size, end))
+ break;
+
+ count++;
+ hole_start = start + fence_size;
+ } while (1);
+ }
+
+ return count < 2;
+}
+
struct i915_vma *
i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
struct i915_gem_ww_ctx *ww,
@@ -879,15 +934,7 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
if (flags & PIN_MAPPABLE &&
(!view || view->type == I915_GGTT_VIEW_NORMAL)) {
- /*
- * If the required space is larger than the available
- * aperture, we will not able to find a slot for the
- * object and unbinding the object now will be in
- * vain. Worse, doing so may cause us to ping-pong
- * the object in and out of the Global GTT and
- * waste a lot of cycles under the mutex.
- */
- if (obj->base.size > ggtt->mappable_end)
+ if (i915_gem_obj_too_big(obj, alignment))
return ERR_PTR(-E2BIG);
/*
diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
index 90546fa58fc1..551644dbfa8a 100644
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -977,7 +977,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
if (err)
goto err_active;
- if (i915_is_ggtt(vma->vm))
+ if (i915_is_ggtt(vma->vm) && flags & PIN_MAPPABLE)
__i915_vma_set_map_and_fenceable(vma);
}
--
2.31.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev3)
2021-10-26 6:38 [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Vivek Kasireddy
` (3 preceding siblings ...)
2021-10-28 14:14 ` [Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev2) Patchwork
@ 2021-10-29 10:36 ` Patchwork
2021-10-29 11:08 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
` (3 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Patchwork @ 2021-10-29 10:36 UTC (permalink / raw)
To: Vivek Kasireddy; +Cc: intel-gfx
== Series Details ==
Series: drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev3)
URL : https://patchwork.freedesktop.org/series/96279/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
fd0a2abfb9d0 drm/i915/gem: Don't try to map and fence large scanout buffers (v3)
-:106: CHECK:PARENTHESIS_ALIGNMENT: Alignment should match open parenthesis
#106: FILE: drivers/gpu/drm/i915/i915_gem.c:898:
+ fence_alignment = i915_gem_fence_alignment(i915, obj->base.size,
+ i915_gem_object_get_tiling(obj),
total: 0 errors, 0 warnings, 1 checks, 85 lines checked
^ permalink raw reply [flat|nested] 22+ messages in thread
* [Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev3)
2021-10-26 6:38 [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Vivek Kasireddy
` (4 preceding siblings ...)
2021-10-29 10:36 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev3) Patchwork
@ 2021-10-29 11:08 ` Patchwork
2021-10-29 18:21 ` [Intel-gfx] ✗ Fi.CI.IGT: failure " Patchwork
` (2 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Patchwork @ 2021-10-29 11:08 UTC (permalink / raw)
To: Vivek Kasireddy; +Cc: intel-gfx
[-- Attachment #1: Type: text/plain, Size: 7306 bytes --]
== Series Details ==
Series: drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev3)
URL : https://patchwork.freedesktop.org/series/96279/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_10813 -> Patchwork_21487
====================================================
Summary
-------
**SUCCESS**
No regressions found.
External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/index.html
Participating hosts (33 -> 33)
------------------------------
Additional (2): fi-tgl-1115g4 fi-icl-u2
Missing (2): bat-dg1-6 bat-adlp-4
Known issues
------------
Here are the changes found in Patchwork_21487 that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@amdgpu/amd_basic@query-info:
- fi-tgl-1115g4: NOTRUN -> [SKIP][1] ([fdo#109315])
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-tgl-1115g4/igt@amdgpu/amd_basic@query-info.html
* igt@amdgpu/amd_cs_nop@fork-gfx0:
- fi-icl-u2: NOTRUN -> [SKIP][2] ([fdo#109315]) +17 similar issues
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-icl-u2/igt@amdgpu/amd_cs_nop@fork-gfx0.html
* igt@amdgpu/amd_cs_nop@nop-gfx0:
- fi-tgl-1115g4: NOTRUN -> [SKIP][3] ([fdo#109315] / [i915#2575]) +16 similar issues
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-tgl-1115g4/igt@amdgpu/amd_cs_nop@nop-gfx0.html
* igt@gem_huc_copy@huc-copy:
- fi-tgl-1115g4: NOTRUN -> [SKIP][4] ([i915#2190])
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-tgl-1115g4/igt@gem_huc_copy@huc-copy.html
- fi-icl-u2: NOTRUN -> [SKIP][5] ([i915#2190])
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-icl-u2/igt@gem_huc_copy@huc-copy.html
* igt@i915_pm_backlight@basic-brightness:
- fi-tgl-1115g4: NOTRUN -> [SKIP][6] ([i915#1155])
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-tgl-1115g4/igt@i915_pm_backlight@basic-brightness.html
* igt@i915_selftest@live@execlists:
- fi-bsw-nick: [PASS][7] -> [INCOMPLETE][8] ([i915#2940])
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/fi-bsw-nick/igt@i915_selftest@live@execlists.html
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-bsw-nick/igt@i915_selftest@live@execlists.html
* igt@kms_chamelium@common-hpd-after-suspend:
- fi-tgl-1115g4: NOTRUN -> [SKIP][9] ([fdo#111827]) +8 similar issues
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-tgl-1115g4/igt@kms_chamelium@common-hpd-after-suspend.html
* igt@kms_chamelium@hdmi-hpd-fast:
- fi-icl-u2: NOTRUN -> [SKIP][10] ([fdo#111827]) +8 similar issues
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-icl-u2/igt@kms_chamelium@hdmi-hpd-fast.html
* igt@kms_cursor_legacy@basic-busy-flip-before-cursor-atomic:
- fi-tgl-1115g4: NOTRUN -> [SKIP][11] ([i915#4103]) +1 similar issue
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-tgl-1115g4/igt@kms_cursor_legacy@basic-busy-flip-before-cursor-atomic.html
* igt@kms_cursor_legacy@basic-busy-flip-before-cursor-legacy:
- fi-icl-u2: NOTRUN -> [SKIP][12] ([fdo#109278]) +2 similar issues
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-icl-u2/igt@kms_cursor_legacy@basic-busy-flip-before-cursor-legacy.html
* igt@kms_force_connector_basic@force-load-detect:
- fi-tgl-1115g4: NOTRUN -> [SKIP][13] ([fdo#109285])
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-tgl-1115g4/igt@kms_force_connector_basic@force-load-detect.html
- fi-icl-u2: NOTRUN -> [SKIP][14] ([fdo#109285])
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-icl-u2/igt@kms_force_connector_basic@force-load-detect.html
* igt@kms_psr@primary_mmap_gtt:
- fi-tgl-1115g4: NOTRUN -> [SKIP][15] ([i915#1072]) +3 similar issues
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-tgl-1115g4/igt@kms_psr@primary_mmap_gtt.html
* igt@prime_vgem@basic-userptr:
- fi-icl-u2: NOTRUN -> [SKIP][16] ([i915#3301])
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-icl-u2/igt@prime_vgem@basic-userptr.html
- fi-tgl-1115g4: NOTRUN -> [SKIP][17] ([i915#3301])
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-tgl-1115g4/igt@prime_vgem@basic-userptr.html
* igt@runner@aborted:
- fi-bsw-nick: NOTRUN -> [FAIL][18] ([fdo#109271] / [i915#1436] / [i915#3428] / [i915#4312])
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-bsw-nick/igt@runner@aborted.html
#### Possible fixes ####
* igt@kms_frontbuffer_tracking@basic:
- fi-cml-u2: [DMESG-WARN][19] ([i915#4269]) -> [PASS][20]
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/fi-cml-u2/igt@kms_frontbuffer_tracking@basic.html
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-cml-u2/igt@kms_frontbuffer_tracking@basic.html
- {fi-hsw-gt1}: [DMESG-WARN][21] ([i915#4290]) -> [PASS][22]
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/fi-hsw-gt1/igt@kms_frontbuffer_tracking@basic.html
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/fi-hsw-gt1/igt@kms_frontbuffer_tracking@basic.html
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#109278]: https://bugs.freedesktop.org/show_bug.cgi?id=109278
[fdo#109285]: https://bugs.freedesktop.org/show_bug.cgi?id=109285
[fdo#109315]: https://bugs.freedesktop.org/show_bug.cgi?id=109315
[fdo#111827]: https://bugs.freedesktop.org/show_bug.cgi?id=111827
[i915#1072]: https://gitlab.freedesktop.org/drm/intel/issues/1072
[i915#1155]: https://gitlab.freedesktop.org/drm/intel/issues/1155
[i915#1436]: https://gitlab.freedesktop.org/drm/intel/issues/1436
[i915#2190]: https://gitlab.freedesktop.org/drm/intel/issues/2190
[i915#2575]: https://gitlab.freedesktop.org/drm/intel/issues/2575
[i915#2940]: https://gitlab.freedesktop.org/drm/intel/issues/2940
[i915#3301]: https://gitlab.freedesktop.org/drm/intel/issues/3301
[i915#3428]: https://gitlab.freedesktop.org/drm/intel/issues/3428
[i915#4103]: https://gitlab.freedesktop.org/drm/intel/issues/4103
[i915#4269]: https://gitlab.freedesktop.org/drm/intel/issues/4269
[i915#4290]: https://gitlab.freedesktop.org/drm/intel/issues/4290
[i915#4312]: https://gitlab.freedesktop.org/drm/intel/issues/4312
Build changes
-------------
* Linux: CI_DRM_10813 -> Patchwork_21487
CI-20190529: 20190529
CI_DRM_10813: 1025cf52c55966384b9724e0679b37e3115179ab @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_6264: 3458490c14afe3cb8aa873fa9e520e1c815ea068 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
Patchwork_21487: fd0a2abfb9d0c5b13e68751742547c8feaafa0b6 @ git://anongit.freedesktop.org/gfx-ci/linux
== Linux commits ==
fd0a2abfb9d0 drm/i915/gem: Don't try to map and fence large scanout buffers (v3)
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/index.html
[-- Attachment #2: Type: text/html, Size: 8777 bytes --]
^ permalink raw reply [flat|nested] 22+ messages in thread
* [Intel-gfx] ✗ Fi.CI.IGT: failure for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev3)
2021-10-26 6:38 [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Vivek Kasireddy
` (5 preceding siblings ...)
2021-10-29 11:08 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
@ 2021-10-29 18:21 ` Patchwork
2021-11-03 1:12 ` [Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev4) Patchwork
2021-11-03 2:39 ` [Intel-gfx] ✓ Fi.CI.IGT: " Patchwork
8 siblings, 0 replies; 22+ messages in thread
From: Patchwork @ 2021-10-29 18:21 UTC (permalink / raw)
To: Vivek Kasireddy; +Cc: intel-gfx
[-- Attachment #1: Type: text/plain, Size: 30299 bytes --]
== Series Details ==
Series: drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev3)
URL : https://patchwork.freedesktop.org/series/96279/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_10813_full -> Patchwork_21487_full
====================================================
Summary
-------
**FAILURE**
Serious unknown changes coming with Patchwork_21487_full absolutely need to be
verified manually.
If you think the reported changes have nothing to do with the changes
introduced in Patchwork_21487_full, please notify your bug team to allow them
to document this new failure mode, which will reduce false positives in CI.
Participating hosts (9 -> 9)
------------------------------
No changes in participating hosts
Possible new issues
-------------------
Here are the unknown changes that may have been introduced in Patchwork_21487_full:
### IGT changes ###
#### Possible regressions ####
* igt@kms_plane@pixel-format@pipe-a-planes:
- shard-iclb: [PASS][1] -> [INCOMPLETE][2]
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-iclb1/igt@kms_plane@pixel-format@pipe-a-planes.html
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-iclb6/igt@kms_plane@pixel-format@pipe-a-planes.html
Known issues
------------
Here are the changes found in Patchwork_21487_full that come from known issues:
### CI changes ###
#### Possible fixes ####
* boot:
- shard-glk: ([FAIL][3], [PASS][4], [PASS][5], [PASS][6], [PASS][7], [PASS][8], [PASS][9], [PASS][10], [PASS][11], [PASS][12], [PASS][13], [PASS][14], [PASS][15], [PASS][16], [PASS][17], [PASS][18], [PASS][19], [PASS][20], [FAIL][21], [PASS][22], [PASS][23], [PASS][24], [PASS][25], [PASS][26], [PASS][27]) ([i915#4392]) -> ([PASS][28], [PASS][29], [PASS][30], [PASS][31], [PASS][32], [PASS][33], [PASS][34], [PASS][35], [PASS][36], [PASS][37], [PASS][38], [PASS][39], [PASS][40], [PASS][41], [PASS][42], [PASS][43], [PASS][44], [PASS][45], [PASS][46], [PASS][47], [PASS][48], [PASS][49], [PASS][50], [PASS][51], [PASS][52])
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk2/boot.html
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk9/boot.html
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk9/boot.html
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk9/boot.html
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk8/boot.html
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk8/boot.html
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk8/boot.html
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk7/boot.html
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk7/boot.html
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk6/boot.html
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk6/boot.html
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk5/boot.html
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk5/boot.html
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk5/boot.html
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk4/boot.html
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk4/boot.html
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk4/boot.html
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk3/boot.html
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk3/boot.html
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk3/boot.html
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk2/boot.html
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk2/boot.html
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk1/boot.html
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk1/boot.html
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk1/boot.html
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk9/boot.html
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk9/boot.html
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk9/boot.html
[31]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk8/boot.html
[32]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk8/boot.html
[33]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk7/boot.html
[34]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk7/boot.html
[35]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk7/boot.html
[36]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk6/boot.html
[37]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk6/boot.html
[38]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk5/boot.html
[39]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk5/boot.html
[40]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk5/boot.html
[41]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk4/boot.html
[42]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk4/boot.html
[43]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk4/boot.html
[44]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk3/boot.html
[45]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk3/boot.html
[46]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk3/boot.html
[47]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk2/boot.html
[48]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk2/boot.html
[49]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk2/boot.html
[50]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk1/boot.html
[51]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk1/boot.html
[52]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk1/boot.html
### IGT changes ###
#### Issues hit ####
* igt@gem_ctx_param@set-priority-not-supported:
- shard-tglb: NOTRUN -> [SKIP][53] ([fdo#109314])
[53]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb6/igt@gem_ctx_param@set-priority-not-supported.html
* igt@gem_ctx_sseu@mmap-args:
- shard-tglb: NOTRUN -> [SKIP][54] ([i915#280])
[54]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb1/igt@gem_ctx_sseu@mmap-args.html
* igt@gem_exec_fair@basic-deadline:
- shard-glk: [PASS][55] -> [FAIL][56] ([i915#2846])
[55]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk2/igt@gem_exec_fair@basic-deadline.html
[56]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk5/igt@gem_exec_fair@basic-deadline.html
* igt@gem_exec_fair@basic-pace@vcs1:
- shard-iclb: NOTRUN -> [FAIL][57] ([i915#2842])
[57]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-iclb4/igt@gem_exec_fair@basic-pace@vcs1.html
* igt@gem_exec_fair@basic-throttle@rcs0:
- shard-tglb: NOTRUN -> [FAIL][58] ([i915#2842])
[58]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb1/igt@gem_exec_fair@basic-throttle@rcs0.html
- shard-iclb: [PASS][59] -> [FAIL][60] ([i915#2842])
[59]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-iclb8/igt@gem_exec_fair@basic-throttle@rcs0.html
[60]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-iclb6/igt@gem_exec_fair@basic-throttle@rcs0.html
* igt@gem_exec_schedule@submit-early-slice@bcs0:
- shard-tglb: [PASS][61] -> [INCOMPLETE][62] ([i915#3797] / [i915#750])
[61]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-tglb6/igt@gem_exec_schedule@submit-early-slice@bcs0.html
[62]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb5/igt@gem_exec_schedule@submit-early-slice@bcs0.html
* igt@gem_huc_copy@huc-copy:
- shard-glk: NOTRUN -> [SKIP][63] ([fdo#109271] / [i915#2190])
[63]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk6/igt@gem_huc_copy@huc-copy.html
* igt@gem_pread@exhaustion:
- shard-tglb: NOTRUN -> [WARN][64] ([i915#2658])
[64]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb6/igt@gem_pread@exhaustion.html
* igt@gem_pxp@verify-pxp-key-change-after-suspend-resume:
- shard-tglb: NOTRUN -> [SKIP][65] ([i915#4270])
[65]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb6/igt@gem_pxp@verify-pxp-key-change-after-suspend-resume.html
* igt@gem_userptr_blits@dmabuf-sync:
- shard-tglb: NOTRUN -> [SKIP][66] ([i915#3323])
[66]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb1/igt@gem_userptr_blits@dmabuf-sync.html
* igt@gem_userptr_blits@input-checking:
- shard-skl: NOTRUN -> [DMESG-WARN][67] ([i915#3002])
[67]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl9/igt@gem_userptr_blits@input-checking.html
- shard-tglb: NOTRUN -> [DMESG-WARN][68] ([i915#3002])
[68]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb2/igt@gem_userptr_blits@input-checking.html
* igt@gem_userptr_blits@vma-merge:
- shard-glk: NOTRUN -> [FAIL][69] ([i915#3318])
[69]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk5/igt@gem_userptr_blits@vma-merge.html
- shard-skl: NOTRUN -> [FAIL][70] ([i915#3318])
[70]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl1/igt@gem_userptr_blits@vma-merge.html
* igt@gem_workarounds@suspend-resume-context:
- shard-kbl: NOTRUN -> [DMESG-WARN][71] ([i915#180])
[71]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-kbl1/igt@gem_workarounds@suspend-resume-context.html
* igt@gen7_exec_parse@basic-allowed:
- shard-tglb: NOTRUN -> [SKIP][72] ([fdo#109289]) +3 similar issues
[72]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb1/igt@gen7_exec_parse@basic-allowed.html
* igt@gen9_exec_parse@basic-rejected-ctx-param:
- shard-tglb: NOTRUN -> [SKIP][73] ([i915#2856])
[73]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb2/igt@gen9_exec_parse@basic-rejected-ctx-param.html
* igt@i915_pm_rc6_residency@media-rc6-accuracy:
- shard-tglb: NOTRUN -> [SKIP][74] ([fdo#109289] / [fdo#111719])
[74]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb7/igt@i915_pm_rc6_residency@media-rc6-accuracy.html
* igt@i915_pm_rpm@modeset-non-lpsp-stress-no-wait:
- shard-tglb: NOTRUN -> [SKIP][75] ([fdo#111644] / [i915#1397] / [i915#2411])
[75]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb1/igt@i915_pm_rpm@modeset-non-lpsp-stress-no-wait.html
* igt@kms_big_fb@linear-32bpp-rotate-180:
- shard-glk: [PASS][76] -> [DMESG-WARN][77] ([i915#118]) +1 similar issue
[76]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk4/igt@kms_big_fb@linear-32bpp-rotate-180.html
[77]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk7/igt@kms_big_fb@linear-32bpp-rotate-180.html
* igt@kms_big_fb@linear-32bpp-rotate-90:
- shard-tglb: NOTRUN -> [SKIP][78] ([fdo#111614]) +1 similar issue
[78]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb2/igt@kms_big_fb@linear-32bpp-rotate-90.html
* igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip:
- shard-glk: NOTRUN -> [SKIP][79] ([fdo#109271] / [i915#3777])
[79]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk6/igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip.html
* igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-0-hflip-async-flip:
- shard-tglb: NOTRUN -> [SKIP][80] ([fdo#111615]) +1 similar issue
[80]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb2/igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-0-hflip-async-flip.html
* igt@kms_ccs@pipe-a-bad-pixel-format-y_tiled_gen12_mc_ccs:
- shard-apl: NOTRUN -> [SKIP][81] ([fdo#109271] / [i915#3886]) +1 similar issue
[81]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-apl2/igt@kms_ccs@pipe-a-bad-pixel-format-y_tiled_gen12_mc_ccs.html
* igt@kms_ccs@pipe-a-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc:
- shard-skl: NOTRUN -> [SKIP][82] ([fdo#109271] / [i915#3886]) +5 similar issues
[82]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl1/igt@kms_ccs@pipe-a-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc.html
* igt@kms_ccs@pipe-b-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc:
- shard-glk: NOTRUN -> [SKIP][83] ([fdo#109271] / [i915#3886]) +7 similar issues
[83]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk5/igt@kms_ccs@pipe-b-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc.html
* igt@kms_ccs@pipe-c-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc:
- shard-kbl: NOTRUN -> [SKIP][84] ([fdo#109271] / [i915#3886])
[84]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-kbl6/igt@kms_ccs@pipe-c-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc.html
* igt@kms_ccs@pipe-c-random-ccs-data-y_tiled_ccs:
- shard-tglb: NOTRUN -> [SKIP][85] ([i915#3689]) +3 similar issues
[85]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb2/igt@kms_ccs@pipe-c-random-ccs-data-y_tiled_ccs.html
* igt@kms_chamelium@dp-hpd-enable-disable-mode:
- shard-tglb: NOTRUN -> [SKIP][86] ([fdo#109284] / [fdo#111827]) +5 similar issues
[86]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb1/igt@kms_chamelium@dp-hpd-enable-disable-mode.html
* igt@kms_chamelium@vga-hpd:
- shard-skl: NOTRUN -> [SKIP][87] ([fdo#109271] / [fdo#111827]) +5 similar issues
[87]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl1/igt@kms_chamelium@vga-hpd.html
* igt@kms_color@pipe-c-ctm-blue-to-red:
- shard-skl: [PASS][88] -> [DMESG-WARN][89] ([i915#1982]) +5 similar issues
[88]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-skl9/igt@kms_color@pipe-c-ctm-blue-to-red.html
[89]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl4/igt@kms_color@pipe-c-ctm-blue-to-red.html
* igt@kms_color_chamelium@pipe-a-ctm-0-75:
- shard-kbl: NOTRUN -> [SKIP][90] ([fdo#109271] / [fdo#111827]) +2 similar issues
[90]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-kbl6/igt@kms_color_chamelium@pipe-a-ctm-0-75.html
* igt@kms_color_chamelium@pipe-a-gamma:
- shard-iclb: NOTRUN -> [SKIP][91] ([fdo#109284] / [fdo#111827])
[91]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-iclb7/igt@kms_color_chamelium@pipe-a-gamma.html
* igt@kms_color_chamelium@pipe-b-ctm-blue-to-red:
- shard-apl: NOTRUN -> [SKIP][92] ([fdo#109271] / [fdo#111827]) +4 similar issues
[92]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-apl2/igt@kms_color_chamelium@pipe-b-ctm-blue-to-red.html
* igt@kms_color_chamelium@pipe-d-ctm-0-25:
- shard-glk: NOTRUN -> [SKIP][93] ([fdo#109271] / [fdo#111827]) +8 similar issues
[93]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk6/igt@kms_color_chamelium@pipe-d-ctm-0-25.html
* igt@kms_content_protection@atomic-dpms:
- shard-tglb: NOTRUN -> [SKIP][94] ([fdo#111828])
[94]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb1/igt@kms_content_protection@atomic-dpms.html
* igt@kms_cursor_crc@pipe-d-cursor-32x10-sliding:
- shard-tglb: NOTRUN -> [SKIP][95] ([i915#3359]) +3 similar issues
[95]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb1/igt@kms_cursor_crc@pipe-d-cursor-32x10-sliding.html
* igt@kms_cursor_crc@pipe-d-cursor-512x170-onscreen:
- shard-tglb: NOTRUN -> [SKIP][96] ([fdo#109279] / [i915#3359])
[96]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb2/igt@kms_cursor_crc@pipe-d-cursor-512x170-onscreen.html
* igt@kms_cursor_legacy@basic-busy-flip-before-cursor-legacy:
- shard-tglb: NOTRUN -> [SKIP][97] ([i915#4103])
[97]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb1/igt@kms_cursor_legacy@basic-busy-flip-before-cursor-legacy.html
* igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions:
- shard-skl: [PASS][98] -> [FAIL][99] ([i915#2346])
[98]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-skl10/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions.html
[99]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl6/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions.html
* igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size:
- shard-skl: NOTRUN -> [FAIL][100] ([i915#2346] / [i915#533])
[100]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl10/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
* igt@kms_cursor_legacy@flip-vs-cursor-toggle:
- shard-iclb: [PASS][101] -> [FAIL][102] ([i915#2346])
[101]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-iclb5/igt@kms_cursor_legacy@flip-vs-cursor-toggle.html
[102]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-iclb7/igt@kms_cursor_legacy@flip-vs-cursor-toggle.html
* igt@kms_flip@2x-flip-vs-expired-vblank@ab-hdmi-a1-hdmi-a2:
- shard-glk: [PASS][103] -> [FAIL][104] ([i915#79])
[103]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk3/igt@kms_flip@2x-flip-vs-expired-vblank@ab-hdmi-a1-hdmi-a2.html
[104]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk7/igt@kms_flip@2x-flip-vs-expired-vblank@ab-hdmi-a1-hdmi-a2.html
* igt@kms_flip@flip-vs-expired-vblank-interruptible@b-edp1:
- shard-skl: [PASS][105] -> [FAIL][106] ([i915#79])
[105]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-skl1/igt@kms_flip@flip-vs-expired-vblank-interruptible@b-edp1.html
[106]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl9/igt@kms_flip@flip-vs-expired-vblank-interruptible@b-edp1.html
* igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile:
- shard-iclb: [PASS][107] -> [SKIP][108] ([i915#3701])
[107]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-iclb8/igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile.html
[108]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-iclb2/igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile.html
* igt@kms_frontbuffer_tracking@fbc-1p-shrfb-fliptrack-mmap-gtt:
- shard-skl: NOTRUN -> [SKIP][109] ([fdo#109271]) +58 similar issues
[109]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl1/igt@kms_frontbuffer_tracking@fbc-1p-shrfb-fliptrack-mmap-gtt.html
* igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-cur-indfb-draw-pwrite:
- shard-tglb: NOTRUN -> [SKIP][110] ([fdo#111825]) +9 similar issues
[110]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb1/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-cur-indfb-draw-pwrite.html
* igt@kms_frontbuffer_tracking@psr-rgb565-draw-mmap-cpu:
- shard-glk: NOTRUN -> [SKIP][111] ([fdo#109271]) +69 similar issues
[111]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk5/igt@kms_frontbuffer_tracking@psr-rgb565-draw-mmap-cpu.html
* igt@kms_hdr@bpc-switch-suspend:
- shard-kbl: [PASS][112] -> [DMESG-WARN][113] ([i915#180]) +4 similar issues
[112]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-kbl1/igt@kms_hdr@bpc-switch-suspend.html
[113]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-kbl7/igt@kms_hdr@bpc-switch-suspend.html
- shard-skl: [PASS][114] -> [FAIL][115] ([i915#1188]) +1 similar issue
[114]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-skl6/igt@kms_hdr@bpc-switch-suspend.html
[115]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl6/igt@kms_hdr@bpc-switch-suspend.html
* igt@kms_pipe_b_c_ivb@disable-pipe-b-enable-pipe-c:
- shard-apl: NOTRUN -> [SKIP][116] ([fdo#109271]) +52 similar issues
[116]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-apl3/igt@kms_pipe_b_c_ivb@disable-pipe-b-enable-pipe-c.html
* igt@kms_pipe_crc_basic@suspend-read-crc-pipe-b:
- shard-apl: [PASS][117] -> [DMESG-WARN][118] ([i915#180]) +1 similar issue
[117]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-apl8/igt@kms_pipe_crc_basic@suspend-read-crc-pipe-b.html
[118]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-apl8/igt@kms_pipe_crc_basic@suspend-read-crc-pipe-b.html
* igt@kms_plane_alpha_blend@pipe-a-alpha-transparent-fb:
- shard-skl: NOTRUN -> [FAIL][119] ([i915#265])
[119]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl1/igt@kms_plane_alpha_blend@pipe-a-alpha-transparent-fb.html
- shard-glk: NOTRUN -> [FAIL][120] ([i915#265])
[120]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk5/igt@kms_plane_alpha_blend@pipe-a-alpha-transparent-fb.html
* igt@kms_plane_alpha_blend@pipe-b-constant-alpha-max:
- shard-apl: NOTRUN -> [FAIL][121] ([fdo#108145] / [i915#265])
[121]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-apl3/igt@kms_plane_alpha_blend@pipe-b-constant-alpha-max.html
* igt@kms_plane_alpha_blend@pipe-c-coverage-7efc:
- shard-skl: [PASS][122] -> [FAIL][123] ([fdo#108145] / [i915#265]) +1 similar issue
[122]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-skl9/igt@kms_plane_alpha_blend@pipe-c-coverage-7efc.html
[123]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl8/igt@kms_plane_alpha_blend@pipe-c-coverage-7efc.html
* igt@kms_plane_scaling@scaler-with-clipping-clamping@pipe-c-scaler-with-clipping-clamping:
- shard-glk: NOTRUN -> [SKIP][124] ([fdo#109271] / [i915#2733])
[124]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk6/igt@kms_plane_scaling@scaler-with-clipping-clamping@pipe-c-scaler-with-clipping-clamping.html
* igt@kms_psr2_sf@cursor-plane-update-sf:
- shard-skl: NOTRUN -> [SKIP][125] ([fdo#109271] / [i915#658]) +1 similar issue
[125]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl1/igt@kms_psr2_sf@cursor-plane-update-sf.html
- shard-glk: NOTRUN -> [SKIP][126] ([fdo#109271] / [i915#658]) +1 similar issue
[126]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk5/igt@kms_psr2_sf@cursor-plane-update-sf.html
* igt@kms_psr2_sf@overlay-plane-update-sf-dmg-area-4:
- shard-apl: NOTRUN -> [SKIP][127] ([fdo#109271] / [i915#658]) +1 similar issue
[127]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-apl3/igt@kms_psr2_sf@overlay-plane-update-sf-dmg-area-4.html
* igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-1:
- shard-tglb: NOTRUN -> [SKIP][128] ([i915#2920]) +1 similar issue
[128]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb1/igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-1.html
* igt@kms_psr@psr2_sprite_mmap_gtt:
- shard-iclb: [PASS][129] -> [SKIP][130] ([fdo#109441]) +2 similar issues
[129]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-iclb2/igt@kms_psr@psr2_sprite_mmap_gtt.html
[130]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-iclb5/igt@kms_psr@psr2_sprite_mmap_gtt.html
* igt@kms_psr@psr2_sprite_render:
- shard-tglb: NOTRUN -> [FAIL][131] ([i915#132] / [i915#3467])
[131]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb2/igt@kms_psr@psr2_sprite_render.html
* igt@kms_vblank@pipe-a-query-forked:
- shard-snb: [PASS][132] -> [SKIP][133] ([fdo#109271]) +1 similar issue
[132]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-snb2/igt@kms_vblank@pipe-a-query-forked.html
[133]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-snb7/igt@kms_vblank@pipe-a-query-forked.html
* igt@kms_vblank@pipe-d-query-busy-hang:
- shard-iclb: NOTRUN -> [SKIP][134] ([fdo#109278])
[134]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-iclb7/igt@kms_vblank@pipe-d-query-busy-hang.html
* igt@kms_writeback@writeback-invalid-parameters:
- shard-glk: NOTRUN -> [SKIP][135] ([fdo#109271] / [i915#2437])
[135]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk6/igt@kms_writeback@writeback-invalid-parameters.html
* igt@nouveau_crc@pipe-b-source-outp-complete:
- shard-tglb: NOTRUN -> [SKIP][136] ([i915#2530]) +1 similar issue
[136]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb6/igt@nouveau_crc@pipe-b-source-outp-complete.html
* igt@prime_nv_pcopy@test1_micro:
- shard-kbl: NOTRUN -> [SKIP][137] ([fdo#109271]) +28 similar issues
[137]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-kbl6/igt@prime_nv_pcopy@test1_micro.html
* igt@prime_nv_test@nv_write_i915_cpu_mmap_read:
- shard-tglb: NOTRUN -> [SKIP][138] ([fdo#109291]) +1 similar issue
[138]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb2/igt@prime_nv_test@nv_write_i915_cpu_mmap_read.html
* igt@sysfs_clients@busy:
- shard-tglb: NOTRUN -> [SKIP][139] ([i915#2994]) +1 similar issue
[139]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb2/igt@sysfs_clients@busy.html
- shard-skl: NOTRUN -> [SKIP][140] ([fdo#109271] / [i915#2994])
[140]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl9/igt@sysfs_clients@busy.html
* igt@sysfs_clients@split-50:
- shard-glk: NOTRUN -> [SKIP][141] ([fdo#109271] / [i915#2994])
[141]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk6/igt@sysfs_clients@split-50.html
#### Possible fixes ####
* igt@gem_eio@unwedge-stress:
- shard-iclb: [TIMEOUT][142] ([i915#2369] / [i915#2481] / [i915#3070]) -> [PASS][143]
[142]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-iclb6/igt@gem_eio@unwedge-stress.html
[143]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-iclb2/igt@gem_eio@unwedge-stress.html
* igt@gem_exec_fair@basic-none@vcs0:
- shard-apl: [FAIL][144] ([i915#2842]) -> [PASS][145]
[144]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-apl6/igt@gem_exec_fair@basic-none@vcs0.html
[145]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-apl6/igt@gem_exec_fair@basic-none@vcs0.html
* igt@gem_exec_fair@basic-pace-share@rcs0:
- shard-glk: [FAIL][146] ([i915#2842]) -> [PASS][147] +1 similar issue
[146]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-glk4/igt@gem_exec_fair@basic-pace-share@rcs0.html
[147]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-glk7/igt@gem_exec_fair@basic-pace-share@rcs0.html
* igt@i915_pm_backlight@fade_with_suspend:
- shard-tglb: [INCOMPLETE][148] ([i915#456]) -> [PASS][149]
[148]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-tglb7/igt@i915_pm_backlight@fade_with_suspend.html
[149]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb1/igt@i915_pm_backlight@fade_with_suspend.html
* igt@i915_suspend@fence-restore-untiled:
- shard-apl: [DMESG-WARN][150] ([i915#180]) -> [PASS][151] +1 similar issue
[150]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-apl4/igt@i915_suspend@fence-restore-untiled.html
[151]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-apl2/igt@i915_suspend@fence-restore-untiled.html
* igt@kms_async_flips@alternate-sync-async-flip:
- shard-tglb: [FAIL][152] ([i915#2521]) -> [PASS][153]
[152]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-tglb3/igt@kms_async_flips@alternate-sync-async-flip.html
[153]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb3/igt@kms_async_flips@alternate-sync-async-flip.html
* igt@kms_async_flips@crc:
- shard-tglb: [FAIL][154] ([i915#4272]) -> [PASS][155]
[154]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-tglb3/igt@kms_async_flips@crc.html
[155]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb6/igt@kms_async_flips@crc.html
* igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-0-async-flip:
- shard-tglb: [FAIL][156] ([i915#3743]) -> [PASS][157] +1 similar issue
[156]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-tglb3/igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-0-async-flip.html
[157]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-tglb6/igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-0-async-flip.html
* igt@kms_color@pipe-b-ctm-0-5:
- shard-skl: [DMESG-WARN][158] ([i915#1982]) -> [PASS][159]
[158]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/shard-skl6/igt@kms_color@pipe-b-ctm-0-5.html
[159]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/shard-skl6/igt@kms_color@pipe-b-ctm-0-5.html
* igt@kms_cursor_legacy@cursor-vs-flip-varying-size:
- shard-iclb: [DMESG-WARN][160] ([i915#1226]) -> [PASS][161]
[160]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10813/
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21487/index.html
[-- Attachment #2: Type: text/html, Size: 33638 bytes --]
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v3)
2021-10-29 7:43 ` [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v3) Vivek Kasireddy
@ 2021-11-01 10:13 ` Tvrtko Ursulin
2021-11-01 20:52 ` Kasireddy, Vivek
2021-11-01 10:21 ` Ville Syrjälä
1 sibling, 1 reply; 22+ messages in thread
From: Tvrtko Ursulin @ 2021-11-01 10:13 UTC (permalink / raw)
To: Vivek Kasireddy, intel-gfx
On 29/10/2021 08:43, Vivek Kasireddy wrote:
> On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
> more framebuffers/scanout buffers results in only one that is mappable/
> fenceable. Therefore, pageflipping between these 2 FBs where only one
> is mappable/fenceable creates latencies large enough to miss alternate
> vblanks thereby producing less optimal framerate.
>
> This mainly happens because when i915_gem_object_pin_to_display_plane()
> is called to pin one of the FB objs, the associated vma is identified
> as misplaced and therefore i915_vma_unbind() is called which unbinds and
> evicts it. This misplaced vma gets subseqently pinned only when
> i915_gem_object_ggtt_pin_ww() is called without the mappable flag. This
> results in a latency of ~10ms and happens every other vblank/repaint cycle.
This happens because of PIN_MAPPABLE being tried first just in case? Ie.:
vma = ERR_PTR(-ENOSPC);
if ((flags & PIN_MAPPABLE) == 0 &&
(!view || view->type == I915_GGTT_VIEW_NORMAL))
vma = i915_gem_object_ggtt_pin_ww(obj, ww, view, 0, alignment,
flags | PIN_MAPPABLE |
PIN_NONBLOCK);
if (IS_ERR(vma) && vma != ERR_PTR(-EDEADLK))
vma = i915_gem_object_ggtt_pin_ww(obj, ww, view, 0,
alignment, flags);
Comment justifying this behaviour talks about ABI:
/*
* As the user may map the buffer once pinned in the display plane
* (e.g. libkms for the bootup splash), we have to ensure that we
* always use map_and_fenceable for all scanout buffers. However,
* it may simply be too big to fit into mappable, in which case
* put it anyway and hope that userspace can cope (but always first
* try to preserve the existing ABI).
*/
I have no clue how important it is today. So if I ask could we just dropped the "optimistic" PIN_MAPPABLE on modern platforms would that be a stupid question? Because even with this patch the change would effectively already happen for 8K boot splash, no? (Since code can not know if this is single buffered boot splash client or double buffered compositor and the criteria for PIN_MAPPABLE is the same.)
>
> Testcase:
> Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
> with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
> a frame ~7ms before the next vblank, the latencies seen between atomic
> commit and flip event are 7, 24 (7 + 16.66), 7, 24..... suggesting that
> it misses the vblank every other frame.
>
> Here is the ftrace snippet that shows the source of the ~10ms latency:
> i915_gem_object_pin_to_display_plane() {
> 0.102 us | i915_gem_object_set_cache_level();
> i915_gem_object_ggtt_pin_ww() {
> 0.390 us | i915_vma_instance();
> 0.178 us | i915_vma_misplaced();
> i915_vma_unbind() {
> __i915_active_wait() {
> 0.082 us | i915_active_acquire_if_busy();
> 0.475 us | }
> intel_runtime_pm_get() {
> 0.087 us | intel_runtime_pm_acquire();
> 0.259 us | }
> __i915_active_wait() {
> 0.085 us | i915_active_acquire_if_busy();
> 0.240 us | }
> __i915_vma_evict() {
> ggtt_unbind_vma() {
> gen8_ggtt_clear_range() {
> 10507.255 us | }
> 10507.689 us | }
> 10508.516 us | }
>
> v2: Instead of using bigjoiner checks, determine whether a scanout
> buffer is too big by checking to see if it is possible to map
> two of them into the ggtt.
>
> v3 (Ville):
> - Count how many fb objects can be fit into the available holes
> instead of checking for a hole twice the object size.
> - Take alignment constraints into account.
> - Limit this large scanout buffer check to >= Gen 12 platforms.
>
> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> Cc: Manasi Navare <manasi.d.navare@intel.com>
> Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
> ---
> drivers/gpu/drm/i915/i915_gem.c | 65 ++++++++++++++++++++++++++++-----
> drivers/gpu/drm/i915/i915_vma.c | 2 +-
> 2 files changed, 57 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> index 981e383d1a5d..761dc385fbfc 100644
> --- a/drivers/gpu/drm/i915/i915_gem.c
> +++ b/drivers/gpu/drm/i915/i915_gem.c
> @@ -866,6 +866,61 @@ static void discard_ggtt_vma(struct i915_vma *vma)
> spin_unlock(&obj->vma.lock);
> }
>
> +static bool i915_gem_obj_too_big(struct drm_i915_gem_object *obj,
> + u64 alignment)
> +{
> + struct drm_i915_private *i915 = to_i915(obj->base.dev);
> + struct i915_ggtt *ggtt = &i915->ggtt;
> + struct drm_mm_node *hole;
> + u64 hole_start, hole_end, start, end;
> + u64 fence_size, fence_alignment;
> + unsigned int count = 0;
> +
> + /*
> + * If the required space is larger than the available
> + * aperture, we will not able to find a slot for the
> + * object and unbinding the object now will be in
> + * vain. Worse, doing so may cause us to ping-pong
> + * the object in and out of the Global GTT and
> + * waste a lot of cycles under the mutex.
> + */
> + if (obj->base.size > ggtt->mappable_end)
> + return true;
> +
> + if (HAS_GMCH(i915) || DISPLAY_VER(i915) < 11 ||
> + !i915_gem_object_is_framebuffer(obj))
> + return false;
Not sure about these checks - are you sure they are preserving like-for-like?
> +
> + fence_size = i915_gem_fence_size(i915, obj->base.size,
> + i915_gem_object_get_tiling(obj),
> + i915_gem_object_get_stride(obj));
> + fence_alignment = i915_gem_fence_alignment(i915, obj->base.size,
> + i915_gem_object_get_tiling(obj),
> + i915_gem_object_get_stride(obj));
> + alignment = max_t(u64, alignment, fence_alignment);
> +
> + /*
> + * Assuming this object is a large scanout buffer, we try to find
> + * out if there is room to map at-least two of them. There could
> + * be space available to map one but to be consistent, we try to
> + * avoid mapping/fencing any of them.
> + */
> + drm_mm_for_each_hole(hole, &ggtt->vm.mm, hole_start, hole_end) {
> + do {
> + start = round_up(hole_start, alignment);
> + end = min_t(u64, hole_end, ggtt->mappable_end);
> +
> + if (range_overflows(start, fence_size, end))
> + break;
> +
> + count++;
If this solution remains, you could break out when count reaches two.
Regards,
Tvrtko
> + hole_start = start + fence_size;
> + } while (1);
> + }
> +
> + return count < 2;
> +}
> +
> struct i915_vma *
> i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
> struct i915_gem_ww_ctx *ww,
> @@ -879,15 +934,7 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
>
> if (flags & PIN_MAPPABLE &&
> (!view || view->type == I915_GGTT_VIEW_NORMAL)) {
> - /*
> - * If the required space is larger than the available
> - * aperture, we will not able to find a slot for the
> - * object and unbinding the object now will be in
> - * vain. Worse, doing so may cause us to ping-pong
> - * the object in and out of the Global GTT and
> - * waste a lot of cycles under the mutex.
> - */
> - if (obj->base.size > ggtt->mappable_end)
> + if (i915_gem_obj_too_big(obj, alignment))
> return ERR_PTR(-E2BIG);
>
> /*
> diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
> index 90546fa58fc1..551644dbfa8a 100644
> --- a/drivers/gpu/drm/i915/i915_vma.c
> +++ b/drivers/gpu/drm/i915/i915_vma.c
> @@ -977,7 +977,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
> if (err)
> goto err_active;
>
> - if (i915_is_ggtt(vma->vm))
> + if (i915_is_ggtt(vma->vm) && flags & PIN_MAPPABLE)
> __i915_vma_set_map_and_fenceable(vma);
> }
>
>
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v3)
2021-10-29 7:43 ` [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v3) Vivek Kasireddy
2021-11-01 10:13 ` Tvrtko Ursulin
@ 2021-11-01 10:21 ` Ville Syrjälä
2021-11-01 21:27 ` Kasireddy, Vivek
1 sibling, 1 reply; 22+ messages in thread
From: Ville Syrjälä @ 2021-11-01 10:21 UTC (permalink / raw)
To: Vivek Kasireddy; +Cc: intel-gfx
On Fri, Oct 29, 2021 at 12:43:03AM -0700, Vivek Kasireddy wrote:
> On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
> more framebuffers/scanout buffers results in only one that is mappable/
> fenceable. Therefore, pageflipping between these 2 FBs where only one
> is mappable/fenceable creates latencies large enough to miss alternate
> vblanks thereby producing less optimal framerate.
>
> This mainly happens because when i915_gem_object_pin_to_display_plane()
> is called to pin one of the FB objs, the associated vma is identified
> as misplaced and therefore i915_vma_unbind() is called which unbinds and
> evicts it. This misplaced vma gets subseqently pinned only when
> i915_gem_object_ggtt_pin_ww() is called without the mappable flag. This
> results in a latency of ~10ms and happens every other vblank/repaint cycle.
>
> Testcase:
> Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
> with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
> a frame ~7ms before the next vblank, the latencies seen between atomic
> commit and flip event are 7, 24 (7 + 16.66), 7, 24..... suggesting that
> it misses the vblank every other frame.
>
> Here is the ftrace snippet that shows the source of the ~10ms latency:
> i915_gem_object_pin_to_display_plane() {
> 0.102 us | i915_gem_object_set_cache_level();
> i915_gem_object_ggtt_pin_ww() {
> 0.390 us | i915_vma_instance();
> 0.178 us | i915_vma_misplaced();
> i915_vma_unbind() {
> __i915_active_wait() {
> 0.082 us | i915_active_acquire_if_busy();
> 0.475 us | }
> intel_runtime_pm_get() {
> 0.087 us | intel_runtime_pm_acquire();
> 0.259 us | }
> __i915_active_wait() {
> 0.085 us | i915_active_acquire_if_busy();
> 0.240 us | }
> __i915_vma_evict() {
> ggtt_unbind_vma() {
> gen8_ggtt_clear_range() {
> 10507.255 us | }
> 10507.689 us | }
> 10508.516 us | }
>
> v2: Instead of using bigjoiner checks, determine whether a scanout
> buffer is too big by checking to see if it is possible to map
> two of them into the ggtt.
>
> v3 (Ville):
> - Count how many fb objects can be fit into the available holes
> instead of checking for a hole twice the object size.
> - Take alignment constraints into account.
> - Limit this large scanout buffer check to >= Gen 12 platforms.
>
> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> Cc: Manasi Navare <manasi.d.navare@intel.com>
> Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
> ---
> drivers/gpu/drm/i915/i915_gem.c | 65 ++++++++++++++++++++++++++++-----
> drivers/gpu/drm/i915/i915_vma.c | 2 +-
> 2 files changed, 57 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> index 981e383d1a5d..761dc385fbfc 100644
> --- a/drivers/gpu/drm/i915/i915_gem.c
> +++ b/drivers/gpu/drm/i915/i915_gem.c
> @@ -866,6 +866,61 @@ static void discard_ggtt_vma(struct i915_vma *vma)
> spin_unlock(&obj->vma.lock);
> }
>
> +static bool i915_gem_obj_too_big(struct drm_i915_gem_object *obj,
> + u64 alignment)
> +{
> + struct drm_i915_private *i915 = to_i915(obj->base.dev);
> + struct i915_ggtt *ggtt = &i915->ggtt;
> + struct drm_mm_node *hole;
> + u64 hole_start, hole_end, start, end;
> + u64 fence_size, fence_alignment;
> + unsigned int count = 0;
> +
> + /*
> + * If the required space is larger than the available
> + * aperture, we will not able to find a slot for the
> + * object and unbinding the object now will be in
> + * vain. Worse, doing so may cause us to ping-pong
> + * the object in and out of the Global GTT and
> + * waste a lot of cycles under the mutex.
> + */
> + if (obj->base.size > ggtt->mappable_end)
> + return true;
> +
> + if (HAS_GMCH(i915) || DISPLAY_VER(i915) < 11 ||
> + !i915_gem_object_is_framebuffer(obj))
> + return false;
> +
> + fence_size = i915_gem_fence_size(i915, obj->base.size,
> + i915_gem_object_get_tiling(obj),
> + i915_gem_object_get_stride(obj));
> + fence_alignment = i915_gem_fence_alignment(i915, obj->base.size,
> + i915_gem_object_get_tiling(obj),
> + i915_gem_object_get_stride(obj));
> + alignment = max_t(u64, alignment, fence_alignment);
> +
> + /*
> + * Assuming this object is a large scanout buffer, we try to find
> + * out if there is room to map at-least two of them. There could
> + * be space available to map one but to be consistent, we try to
> + * avoid mapping/fencing any of them.
> + */
> + drm_mm_for_each_hole(hole, &ggtt->vm.mm, hole_start, hole_end) {
> + do {
> + start = round_up(hole_start, alignment);
> + end = min_t(u64, hole_end, ggtt->mappable_end);
> +
> + if (range_overflows(start, fence_size, end))
> + break;
> +
> + count++;
> + hole_start = start + fence_size;
> + } while (1);
> + }
> +
> + return count < 2;
> +}
> +
> struct i915_vma *
> i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
> struct i915_gem_ww_ctx *ww,
> @@ -879,15 +934,7 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
>
> if (flags & PIN_MAPPABLE &&
> (!view || view->type == I915_GGTT_VIEW_NORMAL)) {
> - /*
> - * If the required space is larger than the available
> - * aperture, we will not able to find a slot for the
> - * object and unbinding the object now will be in
> - * vain. Worse, doing so may cause us to ping-pong
> - * the object in and out of the Global GTT and
> - * waste a lot of cycles under the mutex.
> - */
> - if (obj->base.size > ggtt->mappable_end)
> + if (i915_gem_obj_too_big(obj, alignment))
> return ERR_PTR(-E2BIG);
>
> /*
As I pointed out we already have the current ping-pong heuristic
right here. You should adjust that instead of adding yet another
heuristic in parallel.
> diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
> index 90546fa58fc1..551644dbfa8a 100644
> --- a/drivers/gpu/drm/i915/i915_vma.c
> +++ b/drivers/gpu/drm/i915/i915_vma.c
> @@ -977,7 +977,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
> if (err)
> goto err_active;
>
> - if (i915_is_ggtt(vma->vm))
> + if (i915_is_ggtt(vma->vm) && flags & PIN_MAPPABLE)
> __i915_vma_set_map_and_fenceable(vma);
> }
>
> --
> 2.31.1
--
Ville Syrjälä
Intel
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v3)
2021-11-01 10:13 ` Tvrtko Ursulin
@ 2021-11-01 20:52 ` Kasireddy, Vivek
0 siblings, 0 replies; 22+ messages in thread
From: Kasireddy, Vivek @ 2021-11-01 20:52 UTC (permalink / raw)
To: Tvrtko Ursulin, intel-gfx
Hi Tvrtko,
> On 29/10/2021 08:43, Vivek Kasireddy wrote:
> > On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
> > more framebuffers/scanout buffers results in only one that is mappable/
> > fenceable. Therefore, pageflipping between these 2 FBs where only one
> > is mappable/fenceable creates latencies large enough to miss alternate
> > vblanks thereby producing less optimal framerate.
> >
> > This mainly happens because when i915_gem_object_pin_to_display_plane()
> > is called to pin one of the FB objs, the associated vma is identified
> > as misplaced and therefore i915_vma_unbind() is called which unbinds and
> > evicts it. This misplaced vma gets subseqently pinned only when
> > i915_gem_object_ggtt_pin_ww() is called without the mappable flag. This
> > results in a latency of ~10ms and happens every other vblank/repaint cycle.
>
> This happens because of PIN_MAPPABLE being tried first just in case? Ie.:
[Kasireddy, Vivek] Yes, that is exactly why this issue happens. I should have
probably described how the patch fixes the problem but the goal with this patch
is to make sure that we return early if we try to pin with PIN_MAPPABLE for
8K FBs.
>
> vma = ERR_PTR(-ENOSPC);
> if ((flags & PIN_MAPPABLE) == 0 &&
> (!view || view->type == I915_GGTT_VIEW_NORMAL))
> vma = i915_gem_object_ggtt_pin_ww(obj, ww, view, 0, alignment,
> flags | PIN_MAPPABLE |
> PIN_NONBLOCK);
> if (IS_ERR(vma) && vma != ERR_PTR(-EDEADLK))
> vma = i915_gem_object_ggtt_pin_ww(obj, ww, view, 0,
> alignment, flags);
>
> Comment justifying this behaviour talks about ABI:
>
> /*
> * As the user may map the buffer once pinned in the display plane
> * (e.g. libkms for the bootup splash), we have to ensure that we
> * always use map_and_fenceable for all scanout buffers. However,
> * it may simply be too big to fit into mappable, in which case
> * put it anyway and hope that userspace can cope (but always first
> * try to preserve the existing ABI).
> */
>
> I have no clue how important it is today. So if I ask could we just dropped the "optimistic"
> PIN_MAPPABLE on modern platforms would that be a stupid question?
[Kasireddy, Vivek] Not sure if we could just drop it as we don't know what is out there that
relies on this. I took a quick look at Plymouth and it does seem to still use single buffered
dirtyfb method (or mmap() depending on the plugin) for updating the FB.
Because even
> with this patch the change would effectively already happen for 8K boot splash, no? (Since
> code can not know if this is single buffered boot splash client or double buffered
> compositor and the criteria for PIN_MAPPABLE is the same.)
[Kasireddy, Vivek] Maybe it would make sense to limit 8K modes only to a compositor
and not expose them to something like bootup splash/Plymouth? Even then, Gnome with X can
still use single buffering/dirtyfb method to update FB. As I mentioned in the patch description,
we are able to map and fence one 8K FB but not more than that. I think we could let this stand
and only prevent the unbind -- which is the cause of the latency. But this might fall apart if the
userspace tries to mmap both FBs -- which could happen if we enable remote desktop streaming
with Weston. Because there are all these special cases that we cannot know beforehand, I
thought marking all the 8K FBs as non-mappable/non-fenceable would automatically
prevent (or fail) such attempts.
>
> >
> > Testcase:
> > Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
> > with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
> > a frame ~7ms before the next vblank, the latencies seen between atomic
> > commit and flip event are 7, 24 (7 + 16.66), 7, 24..... suggesting that
> > it misses the vblank every other frame.
> >
> > Here is the ftrace snippet that shows the source of the ~10ms latency:
> > i915_gem_object_pin_to_display_plane() {
> > 0.102 us | i915_gem_object_set_cache_level();
> > i915_gem_object_ggtt_pin_ww() {
> > 0.390 us | i915_vma_instance();
> > 0.178 us | i915_vma_misplaced();
> > i915_vma_unbind() {
> > __i915_active_wait() {
> > 0.082 us | i915_active_acquire_if_busy();
> > 0.475 us | }
> > intel_runtime_pm_get() {
> > 0.087 us | intel_runtime_pm_acquire();
> > 0.259 us | }
> > __i915_active_wait() {
> > 0.085 us | i915_active_acquire_if_busy();
> > 0.240 us | }
> > __i915_vma_evict() {
> > ggtt_unbind_vma() {
> > gen8_ggtt_clear_range() {
> > 10507.255 us | }
> > 10507.689 us | }
> > 10508.516 us | }
> >
> > v2: Instead of using bigjoiner checks, determine whether a scanout
> > buffer is too big by checking to see if it is possible to map
> > two of them into the ggtt.
> >
> > v3 (Ville):
> > - Count how many fb objects can be fit into the available holes
> > instead of checking for a hole twice the object size.
> > - Take alignment constraints into account.
> > - Limit this large scanout buffer check to >= Gen 12 platforms.
> >
> > Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> > Cc: Manasi Navare <manasi.d.navare@intel.com>
> > Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
> > ---
> > drivers/gpu/drm/i915/i915_gem.c | 65 ++++++++++++++++++++++++++++-----
> > drivers/gpu/drm/i915/i915_vma.c | 2 +-
> > 2 files changed, 57 insertions(+), 10 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> > index 981e383d1a5d..761dc385fbfc 100644
> > --- a/drivers/gpu/drm/i915/i915_gem.c
> > +++ b/drivers/gpu/drm/i915/i915_gem.c
> > @@ -866,6 +866,61 @@ static void discard_ggtt_vma(struct i915_vma *vma)
> > spin_unlock(&obj->vma.lock);
> > }
> >
> > +static bool i915_gem_obj_too_big(struct drm_i915_gem_object *obj,
> > + u64 alignment)
> > +{
> > + struct drm_i915_private *i915 = to_i915(obj->base.dev);
> > + struct i915_ggtt *ggtt = &i915->ggtt;
> > + struct drm_mm_node *hole;
> > + u64 hole_start, hole_end, start, end;
> > + u64 fence_size, fence_alignment;
> > + unsigned int count = 0;
> > +
> > + /*
> > + * If the required space is larger than the available
> > + * aperture, we will not able to find a slot for the
> > + * object and unbinding the object now will be in
> > + * vain. Worse, doing so may cause us to ping-pong
> > + * the object in and out of the Global GTT and
> > + * waste a lot of cycles under the mutex.
> > + */
> > + if (obj->base.size > ggtt->mappable_end)
> > + return true;
> > +
> > + if (HAS_GMCH(i915) || DISPLAY_VER(i915) < 11 ||
> > + !i915_gem_object_is_framebuffer(obj))
> > + return false;
>
> Not sure about these checks - are you sure they are preserving like-for-like?
[Kasireddy, Vivek] I think so; platforms with has_gmch must have PIN_MAPPABLE
and 8K is not available for DISPLAY_VER(i915) < 11. The last check is just
for sanity.
>
> > +
> > + fence_size = i915_gem_fence_size(i915, obj->base.size,
> > + i915_gem_object_get_tiling(obj),
> > + i915_gem_object_get_stride(obj));
> > + fence_alignment = i915_gem_fence_alignment(i915, obj->base.size,
> > + i915_gem_object_get_tiling(obj),
> > + i915_gem_object_get_stride(obj));
> > + alignment = max_t(u64, alignment, fence_alignment);
> > +
> > + /*
> > + * Assuming this object is a large scanout buffer, we try to find
> > + * out if there is room to map at-least two of them. There could
> > + * be space available to map one but to be consistent, we try to
> > + * avoid mapping/fencing any of them.
> > + */
> > + drm_mm_for_each_hole(hole, &ggtt->vm.mm, hole_start, hole_end) {
> > + do {
> > + start = round_up(hole_start, alignment);
> > + end = min_t(u64, hole_end, ggtt->mappable_end);
> > +
> > + if (range_overflows(start, fence_size, end))
> > + break;
> > +
> > + count++;
>
> If this solution remains, you could break out when count reaches two.
[Kasireddy, Vivek] Yes, I was going to do that in v4.
Thanks,
Vivek
>
> Regards,
>
> Tvrtko
>
> > + hole_start = start + fence_size;
> > + } while (1);
> > + }
> > +
> > + return count < 2;
> > +}
> > +
> > struct i915_vma *
> > i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
> > struct i915_gem_ww_ctx *ww,
> > @@ -879,15 +934,7 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object
> *obj,
> >
> > if (flags & PIN_MAPPABLE &&
> > (!view || view->type == I915_GGTT_VIEW_NORMAL)) {
> > - /*
> > - * If the required space is larger than the available
> > - * aperture, we will not able to find a slot for the
> > - * object and unbinding the object now will be in
> > - * vain. Worse, doing so may cause us to ping-pong
> > - * the object in and out of the Global GTT and
> > - * waste a lot of cycles under the mutex.
> > - */
> > - if (obj->base.size > ggtt->mappable_end)
> > + if (i915_gem_obj_too_big(obj, alignment))
> > return ERR_PTR(-E2BIG);
> >
> > /*
> > diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
> > index 90546fa58fc1..551644dbfa8a 100644
> > --- a/drivers/gpu/drm/i915/i915_vma.c
> > +++ b/drivers/gpu/drm/i915/i915_vma.c
> > @@ -977,7 +977,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct
> i915_gem_ww_ctx *ww,
> > if (err)
> > goto err_active;
> >
> > - if (i915_is_ggtt(vma->vm))
> > + if (i915_is_ggtt(vma->vm) && flags & PIN_MAPPABLE)
> > __i915_vma_set_map_and_fenceable(vma);
> > }
> >
> >
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v3)
2021-11-01 10:21 ` Ville Syrjälä
@ 2021-11-01 21:27 ` Kasireddy, Vivek
2021-11-02 14:34 ` Ville Syrjälä
0 siblings, 1 reply; 22+ messages in thread
From: Kasireddy, Vivek @ 2021-11-01 21:27 UTC (permalink / raw)
To: Ville Syrjälä; +Cc: intel-gfx
Hi Ville,
>
> On Fri, Oct 29, 2021 at 12:43:03AM -0700, Vivek Kasireddy wrote:
> > On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
> > more framebuffers/scanout buffers results in only one that is mappable/
> > fenceable. Therefore, pageflipping between these 2 FBs where only one
> > is mappable/fenceable creates latencies large enough to miss alternate
> > vblanks thereby producing less optimal framerate.
> >
> > This mainly happens because when i915_gem_object_pin_to_display_plane()
> > is called to pin one of the FB objs, the associated vma is identified
> > as misplaced and therefore i915_vma_unbind() is called which unbinds and
> > evicts it. This misplaced vma gets subseqently pinned only when
> > i915_gem_object_ggtt_pin_ww() is called without the mappable flag. This
> > results in a latency of ~10ms and happens every other vblank/repaint cycle.
> >
> > Testcase:
> > Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
> > with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
> > a frame ~7ms before the next vblank, the latencies seen between atomic
> > commit and flip event are 7, 24 (7 + 16.66), 7, 24..... suggesting that
> > it misses the vblank every other frame.
> >
> > Here is the ftrace snippet that shows the source of the ~10ms latency:
> > i915_gem_object_pin_to_display_plane() {
> > 0.102 us | i915_gem_object_set_cache_level();
> > i915_gem_object_ggtt_pin_ww() {
> > 0.390 us | i915_vma_instance();
> > 0.178 us | i915_vma_misplaced();
> > i915_vma_unbind() {
> > __i915_active_wait() {
> > 0.082 us | i915_active_acquire_if_busy();
> > 0.475 us | }
> > intel_runtime_pm_get() {
> > 0.087 us | intel_runtime_pm_acquire();
> > 0.259 us | }
> > __i915_active_wait() {
> > 0.085 us | i915_active_acquire_if_busy();
> > 0.240 us | }
> > __i915_vma_evict() {
> > ggtt_unbind_vma() {
> > gen8_ggtt_clear_range() {
> > 10507.255 us | }
> > 10507.689 us | }
> > 10508.516 us | }
> >
> > v2: Instead of using bigjoiner checks, determine whether a scanout
> > buffer is too big by checking to see if it is possible to map
> > two of them into the ggtt.
> >
> > v3 (Ville):
> > - Count how many fb objects can be fit into the available holes
> > instead of checking for a hole twice the object size.
> > - Take alignment constraints into account.
> > - Limit this large scanout buffer check to >= Gen 12 platforms.
> >
> > Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> > Cc: Manasi Navare <manasi.d.navare@intel.com>
> > Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
> > ---
> > drivers/gpu/drm/i915/i915_gem.c | 65 ++++++++++++++++++++++++++++-----
> > drivers/gpu/drm/i915/i915_vma.c | 2 +-
> > 2 files changed, 57 insertions(+), 10 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> > index 981e383d1a5d..761dc385fbfc 100644
> > --- a/drivers/gpu/drm/i915/i915_gem.c
> > +++ b/drivers/gpu/drm/i915/i915_gem.c
> > @@ -866,6 +866,61 @@ static void discard_ggtt_vma(struct i915_vma *vma)
> > spin_unlock(&obj->vma.lock);
> > }
> >
> > +static bool i915_gem_obj_too_big(struct drm_i915_gem_object *obj,
> > + u64 alignment)
> > +{
> > + struct drm_i915_private *i915 = to_i915(obj->base.dev);
> > + struct i915_ggtt *ggtt = &i915->ggtt;
> > + struct drm_mm_node *hole;
> > + u64 hole_start, hole_end, start, end;
> > + u64 fence_size, fence_alignment;
> > + unsigned int count = 0;
> > +
> > + /*
> > + * If the required space is larger than the available
> > + * aperture, we will not able to find a slot for the
> > + * object and unbinding the object now will be in
> > + * vain. Worse, doing so may cause us to ping-pong
> > + * the object in and out of the Global GTT and
> > + * waste a lot of cycles under the mutex.
> > + */
> > + if (obj->base.size > ggtt->mappable_end)
> > + return true;
> > +
> > + if (HAS_GMCH(i915) || DISPLAY_VER(i915) < 11 ||
> > + !i915_gem_object_is_framebuffer(obj))
> > + return false;
> > +
> > + fence_size = i915_gem_fence_size(i915, obj->base.size,
> > + i915_gem_object_get_tiling(obj),
> > + i915_gem_object_get_stride(obj));
> > + fence_alignment = i915_gem_fence_alignment(i915, obj->base.size,
> > + i915_gem_object_get_tiling(obj),
> > + i915_gem_object_get_stride(obj));
> > + alignment = max_t(u64, alignment, fence_alignment);
> > +
> > + /*
> > + * Assuming this object is a large scanout buffer, we try to find
> > + * out if there is room to map at-least two of them. There could
> > + * be space available to map one but to be consistent, we try to
> > + * avoid mapping/fencing any of them.
> > + */
> > + drm_mm_for_each_hole(hole, &ggtt->vm.mm, hole_start, hole_end) {
> > + do {
> > + start = round_up(hole_start, alignment);
> > + end = min_t(u64, hole_end, ggtt->mappable_end);
> > +
> > + if (range_overflows(start, fence_size, end))
> > + break;
> > +
> > + count++;
> > + hole_start = start + fence_size;
> > + } while (1);
> > + }
> > +
> > + return count < 2;
> > +}
> > +
> > struct i915_vma *
> > i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
> > struct i915_gem_ww_ctx *ww,
> > @@ -879,15 +934,7 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object
> *obj,
> >
> > if (flags & PIN_MAPPABLE &&
> > (!view || view->type == I915_GGTT_VIEW_NORMAL)) {
> > - /*
> > - * If the required space is larger than the available
> > - * aperture, we will not able to find a slot for the
> > - * object and unbinding the object now will be in
> > - * vain. Worse, doing so may cause us to ping-pong
> > - * the object in and out of the Global GTT and
> > - * waste a lot of cycles under the mutex.
> > - */
> > - if (obj->base.size > ggtt->mappable_end)
> > + if (i915_gem_obj_too_big(obj, alignment))
> > return ERR_PTR(-E2BIG);
> >
> > /*
>
> As I pointed out we already have the current ping-pong heuristic
> right here. You should adjust that instead of adding yet another
> heuristic in parallel.
[Kasireddy, Vivek] I think the heuristic you are referring to is:
obj->base.size > ggtt->mappable_end / 2
Are you saying that I should do something like
obj->base.size > ggtt->mappable_end / 4
and return early? I guess I could do that and it would work in this case
but I thought it would make more sense to look at the available space
dynamically given that both the 8K FBs fail the
obj->base.size > ggtt->mappable_end / 2 check.
Thanks,
Vivek
>
> > diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
> > index 90546fa58fc1..551644dbfa8a 100644
> > --- a/drivers/gpu/drm/i915/i915_vma.c
> > +++ b/drivers/gpu/drm/i915/i915_vma.c
> > @@ -977,7 +977,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct
> i915_gem_ww_ctx *ww,
> > if (err)
> > goto err_active;
> >
> > - if (i915_is_ggtt(vma->vm))
> > + if (i915_is_ggtt(vma->vm) && flags & PIN_MAPPABLE)
> > __i915_vma_set_map_and_fenceable(vma);
> > }
> >
> > --
> > 2.31.1
>
> --
> Ville Syrjälä
> Intel
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v3)
2021-11-01 21:27 ` Kasireddy, Vivek
@ 2021-11-02 14:34 ` Ville Syrjälä
2021-11-03 0:04 ` [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v4) Vivek Kasireddy
0 siblings, 1 reply; 22+ messages in thread
From: Ville Syrjälä @ 2021-11-02 14:34 UTC (permalink / raw)
To: Kasireddy, Vivek; +Cc: intel-gfx
On Mon, Nov 01, 2021 at 09:27:30PM +0000, Kasireddy, Vivek wrote:
> Hi Ville,
>
> >
> > On Fri, Oct 29, 2021 at 12:43:03AM -0700, Vivek Kasireddy wrote:
> > > On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
> > > more framebuffers/scanout buffers results in only one that is mappable/
> > > fenceable. Therefore, pageflipping between these 2 FBs where only one
> > > is mappable/fenceable creates latencies large enough to miss alternate
> > > vblanks thereby producing less optimal framerate.
> > >
> > > This mainly happens because when i915_gem_object_pin_to_display_plane()
> > > is called to pin one of the FB objs, the associated vma is identified
> > > as misplaced and therefore i915_vma_unbind() is called which unbinds and
> > > evicts it. This misplaced vma gets subseqently pinned only when
> > > i915_gem_object_ggtt_pin_ww() is called without the mappable flag. This
> > > results in a latency of ~10ms and happens every other vblank/repaint cycle.
> > >
> > > Testcase:
> > > Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
> > > with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
> > > a frame ~7ms before the next vblank, the latencies seen between atomic
> > > commit and flip event are 7, 24 (7 + 16.66), 7, 24..... suggesting that
> > > it misses the vblank every other frame.
> > >
> > > Here is the ftrace snippet that shows the source of the ~10ms latency:
> > > i915_gem_object_pin_to_display_plane() {
> > > 0.102 us | i915_gem_object_set_cache_level();
> > > i915_gem_object_ggtt_pin_ww() {
> > > 0.390 us | i915_vma_instance();
> > > 0.178 us | i915_vma_misplaced();
> > > i915_vma_unbind() {
> > > __i915_active_wait() {
> > > 0.082 us | i915_active_acquire_if_busy();
> > > 0.475 us | }
> > > intel_runtime_pm_get() {
> > > 0.087 us | intel_runtime_pm_acquire();
> > > 0.259 us | }
> > > __i915_active_wait() {
> > > 0.085 us | i915_active_acquire_if_busy();
> > > 0.240 us | }
> > > __i915_vma_evict() {
> > > ggtt_unbind_vma() {
> > > gen8_ggtt_clear_range() {
> > > 10507.255 us | }
> > > 10507.689 us | }
> > > 10508.516 us | }
> > >
> > > v2: Instead of using bigjoiner checks, determine whether a scanout
> > > buffer is too big by checking to see if it is possible to map
> > > two of them into the ggtt.
> > >
> > > v3 (Ville):
> > > - Count how many fb objects can be fit into the available holes
> > > instead of checking for a hole twice the object size.
> > > - Take alignment constraints into account.
> > > - Limit this large scanout buffer check to >= Gen 12 platforms.
> > >
> > > Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > > Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> > > Cc: Manasi Navare <manasi.d.navare@intel.com>
> > > Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
> > > ---
> > > drivers/gpu/drm/i915/i915_gem.c | 65 ++++++++++++++++++++++++++++-----
> > > drivers/gpu/drm/i915/i915_vma.c | 2 +-
> > > 2 files changed, 57 insertions(+), 10 deletions(-)
> > >
> > > diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> > > index 981e383d1a5d..761dc385fbfc 100644
> > > --- a/drivers/gpu/drm/i915/i915_gem.c
> > > +++ b/drivers/gpu/drm/i915/i915_gem.c
> > > @@ -866,6 +866,61 @@ static void discard_ggtt_vma(struct i915_vma *vma)
> > > spin_unlock(&obj->vma.lock);
> > > }
> > >
> > > +static bool i915_gem_obj_too_big(struct drm_i915_gem_object *obj,
> > > + u64 alignment)
> > > +{
> > > + struct drm_i915_private *i915 = to_i915(obj->base.dev);
> > > + struct i915_ggtt *ggtt = &i915->ggtt;
> > > + struct drm_mm_node *hole;
> > > + u64 hole_start, hole_end, start, end;
> > > + u64 fence_size, fence_alignment;
> > > + unsigned int count = 0;
> > > +
> > > + /*
> > > + * If the required space is larger than the available
> > > + * aperture, we will not able to find a slot for the
> > > + * object and unbinding the object now will be in
> > > + * vain. Worse, doing so may cause us to ping-pong
> > > + * the object in and out of the Global GTT and
> > > + * waste a lot of cycles under the mutex.
> > > + */
> > > + if (obj->base.size > ggtt->mappable_end)
> > > + return true;
> > > +
> > > + if (HAS_GMCH(i915) || DISPLAY_VER(i915) < 11 ||
> > > + !i915_gem_object_is_framebuffer(obj))
> > > + return false;
> > > +
> > > + fence_size = i915_gem_fence_size(i915, obj->base.size,
> > > + i915_gem_object_get_tiling(obj),
> > > + i915_gem_object_get_stride(obj));
> > > + fence_alignment = i915_gem_fence_alignment(i915, obj->base.size,
> > > + i915_gem_object_get_tiling(obj),
> > > + i915_gem_object_get_stride(obj));
> > > + alignment = max_t(u64, alignment, fence_alignment);
> > > +
> > > + /*
> > > + * Assuming this object is a large scanout buffer, we try to find
> > > + * out if there is room to map at-least two of them. There could
> > > + * be space available to map one but to be consistent, we try to
> > > + * avoid mapping/fencing any of them.
> > > + */
> > > + drm_mm_for_each_hole(hole, &ggtt->vm.mm, hole_start, hole_end) {
> > > + do {
> > > + start = round_up(hole_start, alignment);
> > > + end = min_t(u64, hole_end, ggtt->mappable_end);
> > > +
> > > + if (range_overflows(start, fence_size, end))
> > > + break;
> > > +
> > > + count++;
> > > + hole_start = start + fence_size;
> > > + } while (1);
> > > + }
> > > +
> > > + return count < 2;
> > > +}
> > > +
> > > struct i915_vma *
> > > i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
> > > struct i915_gem_ww_ctx *ww,
> > > @@ -879,15 +934,7 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object
> > *obj,
> > >
> > > if (flags & PIN_MAPPABLE &&
> > > (!view || view->type == I915_GGTT_VIEW_NORMAL)) {
> > > - /*
> > > - * If the required space is larger than the available
> > > - * aperture, we will not able to find a slot for the
> > > - * object and unbinding the object now will be in
> > > - * vain. Worse, doing so may cause us to ping-pong
> > > - * the object in and out of the Global GTT and
> > > - * waste a lot of cycles under the mutex.
> > > - */
> > > - if (obj->base.size > ggtt->mappable_end)
> > > + if (i915_gem_obj_too_big(obj, alignment))
> > > return ERR_PTR(-E2BIG);
> > >
> > > /*
> >
> > As I pointed out we already have the current ping-pong heuristic
> > right here. You should adjust that instead of adding yet another
> > heuristic in parallel.
> [Kasireddy, Vivek] I think the heuristic you are referring to is:
> obj->base.size > ggtt->mappable_end / 2
>
> Are you saying that I should do something like
> obj->base.size > ggtt->mappable_end / 4
> and return early? I guess I could do that and it would work in this case
> but I thought it would make more sense to look at the available space
> dynamically given that both the 8K FBs fail the
> obj->base.size > ggtt->mappable_end / 2 check.
No. I'm saying you do roughly what you do here, but do it in the right
place (ie. replace the /2 heuristic with this fancier one).
--
Ville Syrjälä
Intel
^ permalink raw reply [flat|nested] 22+ messages in thread
* [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v4)
2021-11-02 14:34 ` Ville Syrjälä
@ 2021-11-03 0:04 ` Vivek Kasireddy
2021-11-15 19:32 ` Kasireddy, Vivek
0 siblings, 1 reply; 22+ messages in thread
From: Vivek Kasireddy @ 2021-11-03 0:04 UTC (permalink / raw)
To: intel-gfx
On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
more framebuffers/scanout buffers results in only one that is mappable/
fenceable. Therefore, pageflipping between these 2 FBs where only one
is mappable/fenceable creates latencies large enough to miss alternate
vblanks thereby producing less optimal framerate.
This mainly happens because when i915_gem_object_pin_to_display_plane()
is called to pin one of the FB objs, the associated vma is identified
as misplaced and therefore i915_vma_unbind() is called which unbinds and
evicts it. This misplaced vma gets subseqently pinned only when
i915_gem_object_ggtt_pin_ww() is called without PIN_MAPPABLE. This
results in a latency of ~10ms and happens every other vblank/repaint cycle.
Therefore, to fix this issue, we try to see if there is space to map
at-least two objects of a given size and return early if there isn't. This
would ensure that we do not try with PIN_MAPPABLE for any objects that
are too big to map thereby preventing unncessary unbind.
Testcase:
Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
a frame ~7ms before the next vblank, the latencies seen between atomic
commit and flip event are 7, 24 (7 + 16.66), 7, 24..... suggesting that
it misses the vblank every other frame.
Here is the ftrace snippet that shows the source of the ~10ms latency:
i915_gem_object_pin_to_display_plane() {
0.102 us | i915_gem_object_set_cache_level();
i915_gem_object_ggtt_pin_ww() {
0.390 us | i915_vma_instance();
0.178 us | i915_vma_misplaced();
i915_vma_unbind() {
__i915_active_wait() {
0.082 us | i915_active_acquire_if_busy();
0.475 us | }
intel_runtime_pm_get() {
0.087 us | intel_runtime_pm_acquire();
0.259 us | }
__i915_active_wait() {
0.085 us | i915_active_acquire_if_busy();
0.240 us | }
__i915_vma_evict() {
ggtt_unbind_vma() {
gen8_ggtt_clear_range() {
10507.255 us | }
10507.689 us | }
10508.516 us | }
v2: Instead of using bigjoiner checks, determine whether a scanout
buffer is too big by checking to see if it is possible to map
two of them into the ggtt.
v3 (Ville):
- Count how many fb objects can be fit into the available holes
instead of checking for a hole twice the object size.
- Take alignment constraints into account.
- Limit this large scanout buffer check to >= Gen 11 platforms.
v4:
- Remove existing heuristic that checks just for size. (Ville)
- Return early if we find space to map at-least two objects. (Tvrtko)
- Slightly update the commit message.
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Manasi Navare <manasi.d.navare@intel.com>
Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
---
drivers/gpu/drm/i915/i915_gem.c | 87 ++++++++++++++++++++++-----------
drivers/gpu/drm/i915/i915_vma.c | 2 +-
2 files changed, 59 insertions(+), 30 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index d0e642c82064..287508c37a9a 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -866,6 +866,63 @@ static void discard_ggtt_vma(struct i915_vma *vma)
spin_unlock(&obj->vma.lock);
}
+static bool i915_gem_obj_too_big(struct drm_i915_gem_object *obj,
+ u64 alignment)
+{
+ struct drm_i915_private *i915 = to_i915(obj->base.dev);
+ struct i915_ggtt *ggtt = &i915->ggtt;
+ struct drm_mm_node *hole;
+ u64 hole_start, hole_end, start, end;
+ u64 fence_size, fence_alignment;
+ unsigned int count = 0;
+
+ /*
+ * If the required space is larger than the available
+ * aperture, we will not able to find a slot for the
+ * object and unbinding the object now will be in
+ * vain. Worse, doing so may cause us to ping-pong
+ * the object in and out of the Global GTT and
+ * waste a lot of cycles under the mutex.
+ */
+ if (obj->base.size > ggtt->mappable_end)
+ return true;
+
+ if (HAS_GMCH(i915) || DISPLAY_VER(i915) < 11 ||
+ !i915_gem_object_is_framebuffer(obj))
+ return false;
+
+ fence_size = i915_gem_fence_size(i915, obj->base.size,
+ i915_gem_object_get_tiling(obj),
+ i915_gem_object_get_stride(obj));
+ fence_alignment = i915_gem_fence_alignment(i915, obj->base.size,
+ i915_gem_object_get_tiling(obj),
+ i915_gem_object_get_stride(obj));
+ alignment = max_t(u64, alignment, fence_alignment);
+
+ /*
+ * Assuming this object is a large scanout buffer, we try to find
+ * out if there is room to map at-least two of them. There could
+ * be space available to map one but to be consistent, we try to
+ * avoid mapping/fencing any of them.
+ */
+ drm_mm_for_each_hole(hole, &ggtt->vm.mm, hole_start, hole_end) {
+ do {
+ start = round_up(hole_start, alignment);
+ end = min_t(u64, hole_end, ggtt->mappable_end);
+
+ if (range_overflows(start, fence_size, end))
+ break;
+
+ if (++count >= 2)
+ return false;
+
+ hole_start = start + fence_size;
+ } while (1);
+ }
+
+ return true;
+}
+
struct i915_vma *
i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
struct i915_gem_ww_ctx *ww,
@@ -879,36 +936,8 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
if (flags & PIN_MAPPABLE &&
(!view || view->type == I915_GGTT_VIEW_NORMAL)) {
- /*
- * If the required space is larger than the available
- * aperture, we will not able to find a slot for the
- * object and unbinding the object now will be in
- * vain. Worse, doing so may cause us to ping-pong
- * the object in and out of the Global GTT and
- * waste a lot of cycles under the mutex.
- */
- if (obj->base.size > ggtt->mappable_end)
+ if (i915_gem_obj_too_big(obj, alignment))
return ERR_PTR(-E2BIG);
-
- /*
- * If NONBLOCK is set the caller is optimistically
- * trying to cache the full object within the mappable
- * aperture, and *must* have a fallback in place for
- * situations where we cannot bind the object. We
- * can be a little more lax here and use the fallback
- * more often to avoid costly migrations of ourselves
- * and other objects within the aperture.
- *
- * Half-the-aperture is used as a simple heuristic.
- * More interesting would to do search for a free
- * block prior to making the commitment to unbind.
- * That caters for the self-harm case, and with a
- * little more heuristics (e.g. NOFAULT, NOEVICT)
- * we could try to minimise harm to others.
- */
- if (flags & PIN_NONBLOCK &&
- obj->base.size > ggtt->mappable_end / 2)
- return ERR_PTR(-ENOSPC);
}
new_vma:
diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
index 90546fa58fc1..551644dbfa8a 100644
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -977,7 +977,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
if (err)
goto err_active;
- if (i915_is_ggtt(vma->vm))
+ if (i915_is_ggtt(vma->vm) && flags & PIN_MAPPABLE)
__i915_vma_set_map_and_fenceable(vma);
}
--
2.31.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev4)
2021-10-26 6:38 [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Vivek Kasireddy
` (6 preceding siblings ...)
2021-10-29 18:21 ` [Intel-gfx] ✗ Fi.CI.IGT: failure " Patchwork
@ 2021-11-03 1:12 ` Patchwork
2021-11-03 2:39 ` [Intel-gfx] ✓ Fi.CI.IGT: " Patchwork
8 siblings, 0 replies; 22+ messages in thread
From: Patchwork @ 2021-11-03 1:12 UTC (permalink / raw)
To: Vivek Kasireddy; +Cc: intel-gfx
[-- Attachment #1: Type: text/plain, Size: 6775 bytes --]
== Series Details ==
Series: drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev4)
URL : https://patchwork.freedesktop.org/series/96279/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_10830 -> Patchwork_21510
====================================================
Summary
-------
**SUCCESS**
No regressions found.
External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/index.html
Participating hosts (36 -> 35)
------------------------------
Additional (3): fi-kbl-soraka fi-tgl-1115g4 fi-pnv-d510
Missing (4): fi-kbl-x1275 fi-bsw-cyan bat-dg1-6 bat-adlp-4
Known issues
------------
Here are the changes found in Patchwork_21510 that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@amdgpu/amd_basic@query-info:
- fi-tgl-1115g4: NOTRUN -> [SKIP][1] ([fdo#109315])
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-tgl-1115g4/igt@amdgpu/amd_basic@query-info.html
* igt@amdgpu/amd_cs_nop@nop-gfx0:
- fi-tgl-1115g4: NOTRUN -> [SKIP][2] ([fdo#109315] / [i915#2575]) +16 similar issues
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-tgl-1115g4/igt@amdgpu/amd_cs_nop@nop-gfx0.html
* igt@gem_exec_fence@basic-busy@bcs0:
- fi-kbl-soraka: NOTRUN -> [SKIP][3] ([fdo#109271]) +8 similar issues
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-kbl-soraka/igt@gem_exec_fence@basic-busy@bcs0.html
* igt@gem_exec_suspend@basic-s3:
- fi-bdw-5557u: [PASS][4] -> [INCOMPLETE][5] ([i915#146])
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/fi-bdw-5557u/igt@gem_exec_suspend@basic-s3.html
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-bdw-5557u/igt@gem_exec_suspend@basic-s3.html
* igt@gem_huc_copy@huc-copy:
- fi-kbl-soraka: NOTRUN -> [SKIP][6] ([fdo#109271] / [i915#2190])
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-kbl-soraka/igt@gem_huc_copy@huc-copy.html
- fi-tgl-1115g4: NOTRUN -> [SKIP][7] ([i915#2190])
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-tgl-1115g4/igt@gem_huc_copy@huc-copy.html
* igt@i915_pm_backlight@basic-brightness:
- fi-tgl-1115g4: NOTRUN -> [SKIP][8] ([i915#1155])
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-tgl-1115g4/igt@i915_pm_backlight@basic-brightness.html
* igt@i915_selftest@live@gt_pm:
- fi-kbl-soraka: NOTRUN -> [DMESG-FAIL][9] ([i915#1886] / [i915#2291])
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-kbl-soraka/igt@i915_selftest@live@gt_pm.html
* igt@kms_chamelium@common-hpd-after-suspend:
- fi-tgl-1115g4: NOTRUN -> [SKIP][10] ([fdo#111827]) +8 similar issues
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-tgl-1115g4/igt@kms_chamelium@common-hpd-after-suspend.html
- fi-kbl-soraka: NOTRUN -> [SKIP][11] ([fdo#109271] / [fdo#111827]) +8 similar issues
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-kbl-soraka/igt@kms_chamelium@common-hpd-after-suspend.html
* igt@kms_cursor_legacy@basic-busy-flip-before-cursor-atomic:
- fi-tgl-1115g4: NOTRUN -> [SKIP][12] ([i915#4103]) +1 similar issue
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-tgl-1115g4/igt@kms_cursor_legacy@basic-busy-flip-before-cursor-atomic.html
* igt@kms_force_connector_basic@force-load-detect:
- fi-tgl-1115g4: NOTRUN -> [SKIP][13] ([fdo#109285])
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-tgl-1115g4/igt@kms_force_connector_basic@force-load-detect.html
* igt@kms_frontbuffer_tracking@basic:
- fi-cml-u2: [PASS][14] -> [DMESG-WARN][15] ([i915#4269])
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/fi-cml-u2/igt@kms_frontbuffer_tracking@basic.html
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-cml-u2/igt@kms_frontbuffer_tracking@basic.html
* igt@kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-d:
- fi-kbl-soraka: NOTRUN -> [SKIP][16] ([fdo#109271] / [i915#533])
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-kbl-soraka/igt@kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-d.html
* igt@kms_psr@primary_mmap_gtt:
- fi-tgl-1115g4: NOTRUN -> [SKIP][17] ([i915#1072]) +3 similar issues
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-tgl-1115g4/igt@kms_psr@primary_mmap_gtt.html
* igt@prime_vgem@basic-userptr:
- fi-pnv-d510: NOTRUN -> [SKIP][18] ([fdo#109271]) +53 similar issues
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-pnv-d510/igt@prime_vgem@basic-userptr.html
- fi-tgl-1115g4: NOTRUN -> [SKIP][19] ([i915#3301])
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/fi-tgl-1115g4/igt@prime_vgem@basic-userptr.html
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#109285]: https://bugs.freedesktop.org/show_bug.cgi?id=109285
[fdo#109315]: https://bugs.freedesktop.org/show_bug.cgi?id=109315
[fdo#111827]: https://bugs.freedesktop.org/show_bug.cgi?id=111827
[i915#1072]: https://gitlab.freedesktop.org/drm/intel/issues/1072
[i915#1155]: https://gitlab.freedesktop.org/drm/intel/issues/1155
[i915#146]: https://gitlab.freedesktop.org/drm/intel/issues/146
[i915#1886]: https://gitlab.freedesktop.org/drm/intel/issues/1886
[i915#2190]: https://gitlab.freedesktop.org/drm/intel/issues/2190
[i915#2291]: https://gitlab.freedesktop.org/drm/intel/issues/2291
[i915#2575]: https://gitlab.freedesktop.org/drm/intel/issues/2575
[i915#3301]: https://gitlab.freedesktop.org/drm/intel/issues/3301
[i915#3303]: https://gitlab.freedesktop.org/drm/intel/issues/3303
[i915#4103]: https://gitlab.freedesktop.org/drm/intel/issues/4103
[i915#4269]: https://gitlab.freedesktop.org/drm/intel/issues/4269
[i915#533]: https://gitlab.freedesktop.org/drm/intel/issues/533
Build changes
-------------
* Linux: CI_DRM_10830 -> Patchwork_21510
CI-20190529: 20190529
CI_DRM_10830: 551447131277b251d76fa8db11e5a045bf9d853d @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_6268: c712ecac599add7e877883a7c8e2857d3c19836f @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
Patchwork_21510: 642e66b334c4fc816e552aaf1a77747533f4d300 @ git://anongit.freedesktop.org/gfx-ci/linux
== Linux commits ==
642e66b334c4 drm/i915/gem: Don't try to map and fence large scanout buffers (v4)
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/index.html
[-- Attachment #2: Type: text/html, Size: 8193 bytes --]
^ permalink raw reply [flat|nested] 22+ messages in thread
* [Intel-gfx] ✓ Fi.CI.IGT: success for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev4)
2021-10-26 6:38 [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Vivek Kasireddy
` (7 preceding siblings ...)
2021-11-03 1:12 ` [Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev4) Patchwork
@ 2021-11-03 2:39 ` Patchwork
8 siblings, 0 replies; 22+ messages in thread
From: Patchwork @ 2021-11-03 2:39 UTC (permalink / raw)
To: Vivek Kasireddy; +Cc: intel-gfx
[-- Attachment #1: Type: text/plain, Size: 30299 bytes --]
== Series Details ==
Series: drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev4)
URL : https://patchwork.freedesktop.org/series/96279/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_10830_full -> Patchwork_21510_full
====================================================
Summary
-------
**SUCCESS**
No regressions found.
Participating hosts (10 -> 10)
------------------------------
No changes in participating hosts
Known issues
------------
Here are the changes found in Patchwork_21510_full that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@gem_create@create-massive:
- shard-kbl: NOTRUN -> [DMESG-WARN][1] ([i915#3002])
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@gem_create@create-massive.html
* igt@gem_ctx_isolation@preservation-s3@vcs0:
- shard-kbl: [PASS][2] -> [DMESG-WARN][3] ([i915#180]) +1 similar issue
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl1/igt@gem_ctx_isolation@preservation-s3@vcs0.html
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@gem_ctx_isolation@preservation-s3@vcs0.html
* igt@gem_exec_fair@basic-deadline:
- shard-kbl: NOTRUN -> [FAIL][4] ([i915#2846])
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl1/igt@gem_exec_fair@basic-deadline.html
- shard-skl: NOTRUN -> [FAIL][5] ([i915#2846])
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl8/igt@gem_exec_fair@basic-deadline.html
- shard-glk: [PASS][6] -> [FAIL][7] ([i915#2846])
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-glk6/igt@gem_exec_fair@basic-deadline.html
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-glk1/igt@gem_exec_fair@basic-deadline.html
* igt@gem_exec_fair@basic-none@vcs0:
- shard-apl: [PASS][8] -> [FAIL][9] ([i915#2842])
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-apl7/igt@gem_exec_fair@basic-none@vcs0.html
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl1/igt@gem_exec_fair@basic-none@vcs0.html
* igt@gem_exec_fair@basic-pace-share@rcs0:
- shard-tglb: [PASS][10] -> [FAIL][11] ([i915#2842])
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-tglb7/igt@gem_exec_fair@basic-pace-share@rcs0.html
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb7/igt@gem_exec_fair@basic-pace-share@rcs0.html
* igt@gem_exec_fair@basic-pace-solo@rcs0:
- shard-kbl: [PASS][12] -> [FAIL][13] ([i915#2842]) +2 similar issues
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl7/igt@gem_exec_fair@basic-pace-solo@rcs0.html
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@gem_exec_fair@basic-pace-solo@rcs0.html
* igt@gem_exec_fair@basic-pace@vcs0:
- shard-glk: [PASS][14] -> [FAIL][15] ([i915#2842]) +1 similar issue
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-glk2/igt@gem_exec_fair@basic-pace@vcs0.html
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-glk8/igt@gem_exec_fair@basic-pace@vcs0.html
* igt@gem_exec_params@secure-non-root:
- shard-tglb: NOTRUN -> [SKIP][16] ([fdo#112283])
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@gem_exec_params@secure-non-root.html
* igt@gem_huc_copy@huc-copy:
- shard-kbl: NOTRUN -> [SKIP][17] ([fdo#109271] / [i915#2190])
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl1/igt@gem_huc_copy@huc-copy.html
- shard-skl: NOTRUN -> [SKIP][18] ([fdo#109271] / [i915#2190])
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl8/igt@gem_huc_copy@huc-copy.html
* igt@gem_pxp@create-regular-context-1:
- shard-tglb: NOTRUN -> [SKIP][19] ([i915#4270]) +1 similar issue
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@gem_pxp@create-regular-context-1.html
* igt@gem_softpin@evict-snoop:
- shard-tglb: NOTRUN -> [SKIP][20] ([fdo#109312])
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@gem_softpin@evict-snoop.html
* igt@gem_userptr_blits@unsync-unmap-after-close:
- shard-tglb: NOTRUN -> [SKIP][21] ([i915#3297]) +1 similar issue
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@gem_userptr_blits@unsync-unmap-after-close.html
* igt@gen9_exec_parse@bb-large:
- shard-tglb: NOTRUN -> [SKIP][22] ([i915#2856])
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@gen9_exec_parse@bb-large.html
* igt@i915_pm_dc@dc3co-vpb-simulation:
- shard-tglb: NOTRUN -> [SKIP][23] ([i915#1904])
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@i915_pm_dc@dc3co-vpb-simulation.html
* igt@i915_pm_dc@dc6-dpms:
- shard-skl: NOTRUN -> [FAIL][24] ([i915#454])
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl6/igt@i915_pm_dc@dc6-dpms.html
* igt@i915_pm_rpm@fences-dpms:
- shard-iclb: [PASS][25] -> [INCOMPLETE][26] ([i915#2910])
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-iclb3/igt@i915_pm_rpm@fences-dpms.html
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-iclb7/igt@i915_pm_rpm@fences-dpms.html
* igt@i915_pm_rpm@modeset-lpsp-stress:
- shard-apl: NOTRUN -> [SKIP][27] ([fdo#109271]) +109 similar issues
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl7/igt@i915_pm_rpm@modeset-lpsp-stress.html
* igt@kms_big_fb@x-tiled-32bpp-rotate-180:
- shard-glk: [PASS][28] -> [DMESG-WARN][29] ([i915#118])
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-glk5/igt@kms_big_fb@x-tiled-32bpp-rotate-180.html
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-glk6/igt@kms_big_fb@x-tiled-32bpp-rotate-180.html
* igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
- shard-skl: NOTRUN -> [FAIL][30] ([i915#3743]) +1 similar issue
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl8/igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-async-flip.html
* igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip:
- shard-kbl: NOTRUN -> [SKIP][31] ([fdo#109271] / [i915#3777]) +1 similar issue
[31]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl3/igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip.html
* igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-180-async-flip:
- shard-skl: NOTRUN -> [FAIL][32] ([i915#3763])
[32]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl8/igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-180-async-flip.html
* igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-180-hflip:
- shard-apl: NOTRUN -> [SKIP][33] ([fdo#109271] / [i915#3777])
[33]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl6/igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-180-hflip.html
* igt@kms_big_fb@yf-tiled-8bpp-rotate-0:
- shard-tglb: NOTRUN -> [SKIP][34] ([fdo#111615])
[34]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@kms_big_fb@yf-tiled-8bpp-rotate-0.html
* igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-hflip:
- shard-skl: NOTRUN -> [SKIP][35] ([fdo#109271] / [i915#3777])
[35]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl8/igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-hflip.html
* igt@kms_ccs@pipe-a-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc:
- shard-kbl: NOTRUN -> [SKIP][36] ([fdo#109271] / [i915#3886]) +8 similar issues
[36]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl7/igt@kms_ccs@pipe-a-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc.html
* igt@kms_ccs@pipe-b-bad-aux-stride-y_tiled_gen12_mc_ccs:
- shard-skl: NOTRUN -> [SKIP][37] ([fdo#109271] / [i915#3886]) +8 similar issues
[37]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl8/igt@kms_ccs@pipe-b-bad-aux-stride-y_tiled_gen12_mc_ccs.html
* igt@kms_ccs@pipe-b-random-ccs-data-yf_tiled_ccs:
- shard-tglb: NOTRUN -> [SKIP][38] ([i915#3689]) +1 similar issue
[38]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@kms_ccs@pipe-b-random-ccs-data-yf_tiled_ccs.html
* igt@kms_ccs@pipe-c-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc:
- shard-apl: NOTRUN -> [SKIP][39] ([fdo#109271] / [i915#3886]) +3 similar issues
[39]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl6/igt@kms_ccs@pipe-c-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc.html
* igt@kms_ccs@pipe-c-missing-ccs-buffer-y_tiled_gen12_mc_ccs:
- shard-tglb: NOTRUN -> [SKIP][40] ([i915#3689] / [i915#3886]) +1 similar issue
[40]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@kms_ccs@pipe-c-missing-ccs-buffer-y_tiled_gen12_mc_ccs.html
* igt@kms_ccs@pipe-d-crc-primary-basic-y_tiled_gen12_mc_ccs:
- shard-iclb: NOTRUN -> [SKIP][41] ([fdo#109278])
[41]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-iclb6/igt@kms_ccs@pipe-d-crc-primary-basic-y_tiled_gen12_mc_ccs.html
* igt@kms_chamelium@vga-hpd-for-each-pipe:
- shard-kbl: NOTRUN -> [SKIP][42] ([fdo#109271] / [fdo#111827]) +11 similar issues
[42]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@kms_chamelium@vga-hpd-for-each-pipe.html
* igt@kms_color_chamelium@pipe-b-ctm-max:
- shard-skl: NOTRUN -> [SKIP][43] ([fdo#109271] / [fdo#111827]) +13 similar issues
[43]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl9/igt@kms_color_chamelium@pipe-b-ctm-max.html
* igt@kms_color_chamelium@pipe-c-ctm-0-25:
- shard-apl: NOTRUN -> [SKIP][44] ([fdo#109271] / [fdo#111827]) +4 similar issues
[44]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl3/igt@kms_color_chamelium@pipe-c-ctm-0-25.html
* igt@kms_color_chamelium@pipe-d-ctm-max:
- shard-tglb: NOTRUN -> [SKIP][45] ([fdo#109284] / [fdo#111827]) +4 similar issues
[45]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@kms_color_chamelium@pipe-d-ctm-max.html
* igt@kms_content_protection@atomic-dpms:
- shard-apl: NOTRUN -> [TIMEOUT][46] ([i915#1319]) +1 similar issue
[46]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl7/igt@kms_content_protection@atomic-dpms.html
* igt@kms_content_protection@legacy:
- shard-kbl: NOTRUN -> [TIMEOUT][47] ([i915#1319])
[47]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@kms_content_protection@legacy.html
* igt@kms_cursor_crc@pipe-a-cursor-512x170-sliding:
- shard-kbl: NOTRUN -> [SKIP][48] ([fdo#109271]) +155 similar issues
[48]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@kms_cursor_crc@pipe-a-cursor-512x170-sliding.html
* igt@kms_cursor_crc@pipe-b-cursor-512x512-onscreen:
- shard-tglb: NOTRUN -> [SKIP][49] ([fdo#109279] / [i915#3359]) +1 similar issue
[49]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@kms_cursor_crc@pipe-b-cursor-512x512-onscreen.html
* igt@kms_cursor_crc@pipe-b-cursor-suspend:
- shard-apl: [PASS][50] -> [DMESG-WARN][51] ([i915#180]) +3 similar issues
[50]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-apl2/igt@kms_cursor_crc@pipe-b-cursor-suspend.html
[51]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl8/igt@kms_cursor_crc@pipe-b-cursor-suspend.html
* igt@kms_cursor_crc@pipe-c-cursor-32x32-offscreen:
- shard-tglb: NOTRUN -> [SKIP][52] ([i915#3319])
[52]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@kms_cursor_crc@pipe-c-cursor-32x32-offscreen.html
* igt@kms_cursor_legacy@2x-cursor-vs-flip-atomic:
- shard-iclb: NOTRUN -> [SKIP][53] ([fdo#109274] / [fdo#109278]) +1 similar issue
[53]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-iclb6/igt@kms_cursor_legacy@2x-cursor-vs-flip-atomic.html
* igt@kms_cursor_legacy@pipe-d-torture-bo:
- shard-skl: NOTRUN -> [SKIP][54] ([fdo#109271] / [i915#533])
[54]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl6/igt@kms_cursor_legacy@pipe-d-torture-bo.html
* igt@kms_cursor_legacy@short-busy-flip-before-cursor-atomic-transitions-varying-size:
- shard-tglb: NOTRUN -> [SKIP][55] ([i915#4103])
[55]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@kms_cursor_legacy@short-busy-flip-before-cursor-atomic-transitions-varying-size.html
* igt@kms_flip@2x-wf_vblank-ts-check-interruptible:
- shard-iclb: NOTRUN -> [SKIP][56] ([fdo#109274])
[56]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-iclb6/igt@kms_flip@2x-wf_vblank-ts-check-interruptible.html
* igt@kms_flip@flip-vs-expired-vblank@a-edp1:
- shard-skl: [PASS][57] -> [FAIL][58] ([i915#79])
[57]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-skl1/igt@kms_flip@flip-vs-expired-vblank@a-edp1.html
[58]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl9/igt@kms_flip@flip-vs-expired-vblank@a-edp1.html
* igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs:
- shard-kbl: NOTRUN -> [SKIP][59] ([fdo#109271] / [i915#2672])
[59]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl1/igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs.html
- shard-skl: NOTRUN -> [SKIP][60] ([fdo#109271] / [i915#2672])
[60]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl8/igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs.html
* igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-indfb-draw-blt:
- shard-skl: NOTRUN -> [SKIP][61] ([fdo#109271]) +179 similar issues
[61]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl9/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-indfb-draw-blt.html
* igt@kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-move:
- shard-snb: NOTRUN -> [SKIP][62] ([fdo#109271]) +33 similar issues
[62]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-snb4/igt@kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-move.html
* igt@kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-draw-mmap-gtt:
- shard-tglb: NOTRUN -> [SKIP][63] ([fdo#111825]) +7 similar issues
[63]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-draw-mmap-gtt.html
* igt@kms_frontbuffer_tracking@psr-shrfb-scaledprimary:
- shard-glk: NOTRUN -> [SKIP][64] ([fdo#109271])
[64]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-glk8/igt@kms_frontbuffer_tracking@psr-shrfb-scaledprimary.html
* igt@kms_hdr@bpc-switch-suspend:
- shard-tglb: [PASS][65] -> [INCOMPLETE][66] ([i915#1373] / [i915#2828])
[65]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-tglb2/igt@kms_hdr@bpc-switch-suspend.html
[66]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb7/igt@kms_hdr@bpc-switch-suspend.html
* igt@kms_hdr@static-toggle-suspend:
- shard-tglb: NOTRUN -> [SKIP][67] ([i915#1187])
[67]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@kms_hdr@static-toggle-suspend.html
* igt@kms_pipe_crc_basic@read-crc-pipe-d:
- shard-apl: NOTRUN -> [SKIP][68] ([fdo#109271] / [i915#533]) +1 similar issue
[68]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl6/igt@kms_pipe_crc_basic@read-crc-pipe-d.html
* igt@kms_plane_alpha_blend@pipe-b-alpha-transparent-fb:
- shard-kbl: NOTRUN -> [FAIL][69] ([i915#265]) +1 similar issue
[69]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl1/igt@kms_plane_alpha_blend@pipe-b-alpha-transparent-fb.html
- shard-skl: NOTRUN -> [FAIL][70] ([i915#265])
[70]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl8/igt@kms_plane_alpha_blend@pipe-b-alpha-transparent-fb.html
* igt@kms_plane_alpha_blend@pipe-b-coverage-7efc:
- shard-skl: [PASS][71] -> [FAIL][72] ([fdo#108145] / [i915#265]) +1 similar issue
[71]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-skl2/igt@kms_plane_alpha_blend@pipe-b-coverage-7efc.html
[72]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl1/igt@kms_plane_alpha_blend@pipe-b-coverage-7efc.html
* igt@kms_plane_alpha_blend@pipe-c-alpha-7efc:
- shard-kbl: NOTRUN -> [FAIL][73] ([fdo#108145] / [i915#265]) +4 similar issues
[73]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@kms_plane_alpha_blend@pipe-c-alpha-7efc.html
- shard-apl: NOTRUN -> [FAIL][74] ([fdo#108145] / [i915#265])
[74]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl3/igt@kms_plane_alpha_blend@pipe-c-alpha-7efc.html
* igt@kms_plane_lowres@pipe-a-tiling-x:
- shard-iclb: NOTRUN -> [SKIP][75] ([i915#3536])
[75]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-iclb6/igt@kms_plane_lowres@pipe-a-tiling-x.html
* igt@kms_psr2_sf@overlay-plane-update-sf-dmg-area-4:
- shard-apl: NOTRUN -> [SKIP][76] ([fdo#109271] / [i915#658]) +5 similar issues
[76]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl7/igt@kms_psr2_sf@overlay-plane-update-sf-dmg-area-4.html
* igt@kms_psr2_sf@plane-move-sf-dmg-area-3:
- shard-kbl: NOTRUN -> [SKIP][77] ([fdo#109271] / [i915#658]) +2 similar issues
[77]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@kms_psr2_sf@plane-move-sf-dmg-area-3.html
* igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-3:
- shard-tglb: NOTRUN -> [SKIP][78] ([i915#2920])
[78]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-3.html
* igt@kms_psr2_su@frontbuffer:
- shard-skl: NOTRUN -> [SKIP][79] ([fdo#109271] / [i915#658]) +2 similar issues
[79]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl9/igt@kms_psr2_su@frontbuffer.html
* igt@kms_psr@psr2_suspend:
- shard-iclb: [PASS][80] -> [SKIP][81] ([fdo#109441]) +3 similar issues
[80]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-iclb2/igt@kms_psr@psr2_suspend.html
[81]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-iclb1/igt@kms_psr@psr2_suspend.html
* igt@kms_vblank@pipe-a-ts-continuation-suspend:
- shard-kbl: [PASS][82] -> [DMESG-WARN][83] ([i915#180] / [i915#295])
[82]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl6/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
[83]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl1/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
* igt@kms_writeback@writeback-pixel-formats:
- shard-kbl: NOTRUN -> [SKIP][84] ([fdo#109271] / [i915#2437])
[84]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@kms_writeback@writeback-pixel-formats.html
* igt@nouveau_crc@pipe-a-ctx-flip-detection:
- shard-tglb: NOTRUN -> [SKIP][85] ([i915#2530])
[85]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@nouveau_crc@pipe-a-ctx-flip-detection.html
* igt@perf_pmu@rc6-suspend:
- shard-kbl: NOTRUN -> [DMESG-WARN][86] ([i915#180])
[86]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl6/igt@perf_pmu@rc6-suspend.html
* igt@prime_nv_pcopy@test1_micro:
- shard-tglb: NOTRUN -> [SKIP][87] ([fdo#109291])
[87]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@prime_nv_pcopy@test1_micro.html
* igt@sysfs_clients@fair-0:
- shard-skl: NOTRUN -> [SKIP][88] ([fdo#109271] / [i915#2994]) +2 similar issues
[88]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl9/igt@sysfs_clients@fair-0.html
- shard-apl: NOTRUN -> [SKIP][89] ([fdo#109271] / [i915#2994]) +1 similar issue
[89]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl7/igt@sysfs_clients@fair-0.html
* igt@sysfs_clients@pidname:
- shard-tglb: NOTRUN -> [SKIP][90] ([i915#2994])
[90]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@sysfs_clients@pidname.html
* igt@sysfs_clients@split-25:
- shard-kbl: NOTRUN -> [SKIP][91] ([fdo#109271] / [i915#2994]) +1 similar issue
[91]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl6/igt@sysfs_clients@split-25.html
#### Possible fixes ####
* igt@gem_eio@in-flight-10ms:
- shard-iclb: [TIMEOUT][92] -> [PASS][93]
[92]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-iclb7/igt@gem_eio@in-flight-10ms.html
[93]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-iclb7/igt@gem_eio@in-flight-10ms.html
* igt@gem_eio@unwedge-stress:
- shard-tglb: [TIMEOUT][94] ([i915#2369] / [i915#3063] / [i915#3648]) -> [PASS][95]
[94]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-tglb2/igt@gem_eio@unwedge-stress.html
[95]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb2/igt@gem_eio@unwedge-stress.html
* igt@gem_exec_fair@basic-none@vcs0:
- shard-kbl: [FAIL][96] ([i915#2842]) -> [PASS][97] +2 similar issues
[96]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl3/igt@gem_exec_fair@basic-none@vcs0.html
[97]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@gem_exec_fair@basic-none@vcs0.html
* igt@gem_exec_fair@basic-pace@vecs0:
- shard-kbl: [SKIP][98] ([fdo#109271]) -> [PASS][99] +1 similar issue
[98]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl3/igt@gem_exec_fair@basic-pace@vecs0.html
[99]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl2/igt@gem_exec_fair@basic-pace@vecs0.html
- shard-glk: [FAIL][100] ([i915#2842]) -> [PASS][101]
[100]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-glk2/igt@gem_exec_fair@basic-pace@vecs0.html
[101]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-glk8/igt@gem_exec_fair@basic-pace@vecs0.html
* igt@gem_exec_schedule@u-submit-early-slice@vecs0:
- shard-skl: [INCOMPLETE][102] ([i915#3797]) -> [PASS][103]
[102]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-skl10/igt@gem_exec_schedule@u-submit-early-slice@vecs0.html
[103]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl10/igt@gem_exec_schedule@u-submit-early-slice@vecs0.html
* igt@i915_pm_dc@dc6-psr:
- shard-iclb: [FAIL][104] ([i915#454]) -> [PASS][105]
[104]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-iclb4/igt@i915_pm_dc@dc6-psr.html
[105]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-iclb2/igt@i915_pm_dc@dc6-psr.html
* igt@i915_selftest@live@hangcheck:
- shard-snb: [INCOMPLETE][106] ([i915#3921]) -> [PASS][107]
[106]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-snb7/igt@i915_selftest@live@hangcheck.html
[107]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-snb4/igt@i915_selftest@live@hangcheck.html
* igt@i915_suspend@forcewake:
- shard-tglb: [INCOMPLETE][108] ([i915#2411] / [i915#456]) -> [PASS][109]
[108]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-tglb7/igt@i915_suspend@forcewake.html
[109]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-tglb5/igt@i915_suspend@forcewake.html
* igt@kms_flip@flip-vs-expired-vblank@a-dp1:
- shard-apl: [FAIL][110] ([i915#79]) -> [PASS][111]
[110]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-apl3/igt@kms_flip@flip-vs-expired-vblank@a-dp1.html
[111]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl2/igt@kms_flip@flip-vs-expired-vblank@a-dp1.html
* igt@kms_flip@flip-vs-suspend-interruptible@a-dp1:
- shard-kbl: [DMESG-WARN][112] ([i915#180]) -> [PASS][113] +10 similar issues
[112]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl6/igt@kms_flip@flip-vs-suspend-interruptible@a-dp1.html
[113]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl3/igt@kms_flip@flip-vs-suspend-interruptible@a-dp1.html
* igt@kms_flip@flip-vs-suspend@a-dp1:
- shard-apl: [DMESG-WARN][114] ([i915#180]) -> [PASS][115] +5 similar issues
[114]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-apl6/igt@kms_flip@flip-vs-suspend@a-dp1.html
[115]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl4/igt@kms_flip@flip-vs-suspend@a-dp1.html
* igt@kms_flip@plain-flip-fb-recreate@c-edp1:
- shard-skl: [FAIL][116] ([i915#2122]) -> [PASS][117] +2 similar issues
[116]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-skl2/igt@kms_flip@plain-flip-fb-recreate@c-edp1.html
[117]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-skl7/igt@kms_flip@plain-flip-fb-recreate@c-edp1.html
* igt@kms_psr2_su@frontbuffer:
- shard-iclb: [SKIP][118] ([fdo#109642] / [fdo#111068] / [i915#658]) -> [PASS][119]
[118]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-iclb1/igt@kms_psr2_su@frontbuffer.html
[119]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-iclb2/igt@kms_psr2_su@frontbuffer.html
* igt@kms_psr@psr2_sprite_plane_move:
- shard-iclb: [SKIP][120] ([fdo#109441]) -> [PASS][121] +2 similar issues
[120]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-iclb1/igt@kms_psr@psr2_sprite_plane_move.html
[121]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-iclb2/igt@kms_psr@psr2_sprite_plane_move.html
* igt@sysfs_timeslice_duration@timeout@vecs0:
- shard-apl: [FAIL][122] ([i915#1755]) -> [PASS][123]
[122]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-apl1/igt@sysfs_timeslice_duration@timeout@vecs0.html
[123]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-apl1/igt@sysfs_timeslice_duration@timeout@vecs0.html
#### Warnings ####
* igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-2:
- shard-iclb: [SKIP][124] ([i915#2920]) -> [SKIP][125] ([i915#658])
[124]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-iclb2/igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-2.html
[125]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-iclb5/igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-2.html
* igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-4:
- shard-iclb: [SKIP][126] ([i915#658]) -> [SKIP][127] ([i915#2920]) +1 similar issue
[126]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-iclb4/igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-4.html
[127]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-iclb2/igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-4.html
* igt@runner@aborted:
- shard-kbl: ([FAIL][128], [FAIL][129], [FAIL][130], [FAIL][131], [FAIL][132], [FAIL][133], [FAIL][134], [FAIL][135], [FAIL][136], [FAIL][137]) ([fdo#109271] / [i915#1436] / [i915#180] / [i915#1814] / [i915#3002] / [i915#3363] / [i915#4312]) -> ([FAIL][138], [FAIL][139], [FAIL][140], [FAIL][141], [FAIL][142], [FAIL][143], [FAIL][144]) ([i915#1436] / [i915#180] / [i915#1814] / [i915#3002] / [i915#3363] / [i915#4312] / [i915#602])
[128]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl3/igt@runner@aborted.html
[129]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl1/igt@runner@aborted.html
[130]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl1/igt@runner@aborted.html
[131]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl1/igt@runner@aborted.html
[132]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl7/igt@runner@aborted.html
[133]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl4/igt@runner@aborted.html
[134]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl4/igt@runner@aborted.html
[135]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl4/igt@runner@aborted.html
[136]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl6/igt@runner@aborted.html
[137]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10830/shard-kbl4/igt@runner@aborted.html
[138]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@runner@aborted.html
[139]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@runner@aborted.html
[140]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl6/igt@runner@aborted.html
[141]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl1/igt@runner@aborted.html
[142]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@runner@aborted.html
[143]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl4/igt@runner@aborted.html
[144]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/shard-kbl6/igt@runner@aborted.html
- shard-apl: ([FAIL][145], [FAIL][146], [FAIL][147], [FAIL][148], [FAIL][149], [FAIL][150], [FAIL][151], [FAIL][152]) ([fdo#109271] / [i915#180] / [i915#1814] / [i915#3002] / [i915#3363] /
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21510/index.html
[-- Attachment #2: Type: text/html, Size: 34084 bytes --]
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v4)
2021-11-03 0:04 ` [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v4) Vivek Kasireddy
@ 2021-11-15 19:32 ` Kasireddy, Vivek
0 siblings, 0 replies; 22+ messages in thread
From: Kasireddy, Vivek @ 2021-11-15 19:32 UTC (permalink / raw)
To: intel-gfx
Hi Ville, Tvrtko,
Any further comments on v4 of this patch?
> On platforms capable of allowing 8K (7680 x 4320) modes, pinning 2 or
> more framebuffers/scanout buffers results in only one that is mappable/
> fenceable. Therefore, pageflipping between these 2 FBs where only one
> is mappable/fenceable creates latencies large enough to miss alternate
> vblanks thereby producing less optimal framerate.
>
> This mainly happens because when i915_gem_object_pin_to_display_plane()
> is called to pin one of the FB objs, the associated vma is identified
> as misplaced and therefore i915_vma_unbind() is called which unbinds and
> evicts it. This misplaced vma gets subseqently pinned only when
> i915_gem_object_ggtt_pin_ww() is called without PIN_MAPPABLE. This
> results in a latency of ~10ms and happens every other vblank/repaint cycle.
> Therefore, to fix this issue, we try to see if there is space to map
> at-least two objects of a given size and return early if there isn't. This
> would ensure that we do not try with PIN_MAPPABLE for any objects that
> are too big to map thereby preventing unncessary unbind.
>
> Testcase:
> Running Weston and weston-simple-egl on an Alderlake_S (ADLS) platform
> with a 8K@60 mode results in only ~40 FPS. Since upstream Weston submits
> a frame ~7ms before the next vblank, the latencies seen between atomic
> commit and flip event are 7, 24 (7 + 16.66), 7, 24..... suggesting that
> it misses the vblank every other frame.
>
> Here is the ftrace snippet that shows the source of the ~10ms latency:
> i915_gem_object_pin_to_display_plane() {
> 0.102 us | i915_gem_object_set_cache_level();
> i915_gem_object_ggtt_pin_ww() {
> 0.390 us | i915_vma_instance();
> 0.178 us | i915_vma_misplaced();
> i915_vma_unbind() {
> __i915_active_wait() {
> 0.082 us | i915_active_acquire_if_busy();
> 0.475 us | }
> intel_runtime_pm_get() {
> 0.087 us | intel_runtime_pm_acquire();
> 0.259 us | }
> __i915_active_wait() {
> 0.085 us | i915_active_acquire_if_busy();
> 0.240 us | }
> __i915_vma_evict() {
> ggtt_unbind_vma() {
> gen8_ggtt_clear_range() {
> 10507.255 us | }
> 10507.689 us | }
> 10508.516 us | }
>
> v2: Instead of using bigjoiner checks, determine whether a scanout
> buffer is too big by checking to see if it is possible to map
> two of them into the ggtt.
>
> v3 (Ville):
> - Count how many fb objects can be fit into the available holes
> instead of checking for a hole twice the object size.
> - Take alignment constraints into account.
> - Limit this large scanout buffer check to >= Gen 11 platforms.
>
> v4:
> - Remove existing heuristic that checks just for size. (Ville)
> - Return early if we find space to map at-least two objects. (Tvrtko)
> - Slightly update the commit message.
>
> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> Cc: Manasi Navare <manasi.d.navare@intel.com>
> Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
> ---
> drivers/gpu/drm/i915/i915_gem.c | 87 ++++++++++++++++++++++-----------
> drivers/gpu/drm/i915/i915_vma.c | 2 +-
> 2 files changed, 59 insertions(+), 30 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> index d0e642c82064..287508c37a9a 100644
> --- a/drivers/gpu/drm/i915/i915_gem.c
> +++ b/drivers/gpu/drm/i915/i915_gem.c
> @@ -866,6 +866,63 @@ static void discard_ggtt_vma(struct i915_vma *vma)
> spin_unlock(&obj->vma.lock);
> }
>
> +static bool i915_gem_obj_too_big(struct drm_i915_gem_object *obj,
> + u64 alignment)
> +{
> + struct drm_i915_private *i915 = to_i915(obj->base.dev);
> + struct i915_ggtt *ggtt = &i915->ggtt;
> + struct drm_mm_node *hole;
> + u64 hole_start, hole_end, start, end;
> + u64 fence_size, fence_alignment;
> + unsigned int count = 0;
> +
> + /*
> + * If the required space is larger than the available
> + * aperture, we will not able to find a slot for the
> + * object and unbinding the object now will be in
> + * vain. Worse, doing so may cause us to ping-pong
> + * the object in and out of the Global GTT and
> + * waste a lot of cycles under the mutex.
> + */
> + if (obj->base.size > ggtt->mappable_end)
> + return true;
> +
> + if (HAS_GMCH(i915) || DISPLAY_VER(i915) < 11 ||
> + !i915_gem_object_is_framebuffer(obj))
> + return false;
> +
> + fence_size = i915_gem_fence_size(i915, obj->base.size,
> + i915_gem_object_get_tiling(obj),
> + i915_gem_object_get_stride(obj));
> + fence_alignment = i915_gem_fence_alignment(i915, obj->base.size,
> + i915_gem_object_get_tiling(obj),
> + i915_gem_object_get_stride(obj));
> + alignment = max_t(u64, alignment, fence_alignment);
> +
> + /*
> + * Assuming this object is a large scanout buffer, we try to find
> + * out if there is room to map at-least two of them. There could
> + * be space available to map one but to be consistent, we try to
> + * avoid mapping/fencing any of them.
> + */
> + drm_mm_for_each_hole(hole, &ggtt->vm.mm, hole_start, hole_end) {
> + do {
> + start = round_up(hole_start, alignment);
> + end = min_t(u64, hole_end, ggtt->mappable_end);
> +
> + if (range_overflows(start, fence_size, end))
> + break;
> +
> + if (++count >= 2)
> + return false;
> +
> + hole_start = start + fence_size;
> + } while (1);
> + }
> +
> + return true;
> +}
> +
> struct i915_vma *
> i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj,
> struct i915_gem_ww_ctx *ww,
> @@ -879,36 +936,8 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object
> *obj,
>
> if (flags & PIN_MAPPABLE &&
> (!view || view->type == I915_GGTT_VIEW_NORMAL)) {
> - /*
> - * If the required space is larger than the available
> - * aperture, we will not able to find a slot for the
> - * object and unbinding the object now will be in
> - * vain. Worse, doing so may cause us to ping-pong
> - * the object in and out of the Global GTT and
> - * waste a lot of cycles under the mutex.
> - */
> - if (obj->base.size > ggtt->mappable_end)
> + if (i915_gem_obj_too_big(obj, alignment))
> return ERR_PTR(-E2BIG);
> -
> - /*
> - * If NONBLOCK is set the caller is optimistically
> - * trying to cache the full object within the mappable
> - * aperture, and *must* have a fallback in place for
> - * situations where we cannot bind the object. We
> - * can be a little more lax here and use the fallback
> - * more often to avoid costly migrations of ourselves
> - * and other objects within the aperture.
> - *
> - * Half-the-aperture is used as a simple heuristic.
> - * More interesting would to do search for a free
> - * block prior to making the commitment to unbind.
> - * That caters for the self-harm case, and with a
> - * little more heuristics (e.g. NOFAULT, NOEVICT)
> - * we could try to minimise harm to others.
> - */
> - if (flags & PIN_NONBLOCK &&
> - obj->base.size > ggtt->mappable_end / 2)
> - return ERR_PTR(-ENOSPC);
> }
>
> new_vma:
> diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
> index 90546fa58fc1..551644dbfa8a 100644
> --- a/drivers/gpu/drm/i915/i915_vma.c
> +++ b/drivers/gpu/drm/i915/i915_vma.c
> @@ -977,7 +977,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct
> i915_gem_ww_ctx *ww,
> if (err)
> goto err_active;
>
> - if (i915_is_ggtt(vma->vm))
> + if (i915_is_ggtt(vma->vm) && flags & PIN_MAPPABLE)
> __i915_vma_set_map_and_fenceable(vma);
> }
>
> --
> 2.31.1
^ permalink raw reply [flat|nested] 22+ messages in thread
end of thread, other threads:[~2021-11-15 19:32 UTC | newest]
Thread overview: 22+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-26 6:38 [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Vivek Kasireddy
2021-10-26 7:52 ` [Intel-gfx] ✓ Fi.CI.BAT: success for " Patchwork
2021-10-26 9:07 ` [Intel-gfx] [PATCH] " Ville Syrjälä
2021-10-27 6:41 ` Kasireddy, Vivek
2021-10-27 8:27 ` Ville Syrjälä
2021-10-28 8:04 ` [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large " Vivek Kasireddy
2021-10-28 12:53 ` Ville Syrjälä
2021-10-29 7:43 ` [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v3) Vivek Kasireddy
2021-11-01 10:13 ` Tvrtko Ursulin
2021-11-01 20:52 ` Kasireddy, Vivek
2021-11-01 10:21 ` Ville Syrjälä
2021-11-01 21:27 ` Kasireddy, Vivek
2021-11-02 14:34 ` Ville Syrjälä
2021-11-03 0:04 ` [Intel-gfx] [PATCH] drm/i915/gem: Don't try to map and fence large scanout buffers (v4) Vivek Kasireddy
2021-11-15 19:32 ` Kasireddy, Vivek
2021-10-26 9:33 ` [Intel-gfx] ✗ Fi.CI.IGT: failure for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers Patchwork
2021-10-28 14:14 ` [Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev2) Patchwork
2021-10-29 10:36 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev3) Patchwork
2021-10-29 11:08 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2021-10-29 18:21 ` [Intel-gfx] ✗ Fi.CI.IGT: failure " Patchwork
2021-11-03 1:12 ` [Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915/gem: Don't try to map and fence 8K/bigjoiner scanout buffers (rev4) Patchwork
2021-11-03 2:39 ` [Intel-gfx] ✓ Fi.CI.IGT: " Patchwork
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.