All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Hawking Zhang <Hawking.Zhang@amd.com>,
	Alex Deucher <alexander.deucher@amd.com>
Subject: [PATCH 5.10 36/48] drm/amdgpu: add support for IP discovery gc_info table v2
Date: Mon,  3 Jan 2022 15:24:13 +0100	[thread overview]
Message-ID: <20220103142054.697936926@linuxfoundation.org> (raw)
In-Reply-To: <20220103142053.466768714@linuxfoundation.org>

From: Alex Deucher <alexander.deucher@amd.com>

commit 5e713c6afa34c0fd6f113bf7bb1c2847172d7b20 upstream.

Used on gfx9 based systems. Fixes incorrect CU counts reported
in the kernel log.

Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1833
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c |   76 ++++++++++++++++++--------
 drivers/gpu/drm/amd/include/discovery.h       |   49 ++++++++++++++++
 2 files changed, 103 insertions(+), 22 deletions(-)

--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -372,10 +372,15 @@ int amdgpu_discovery_get_ip_version(stru
 	return -EINVAL;
 }
 
+union gc_info {
+	struct gc_info_v1_0 v1;
+	struct gc_info_v2_0 v2;
+};
+
 int amdgpu_discovery_get_gfx_info(struct amdgpu_device *adev)
 {
 	struct binary_header *bhdr;
-	struct gc_info_v1_0 *gc_info;
+	union gc_info *gc_info;
 
 	if (!adev->mman.discovery_bin) {
 		DRM_ERROR("ip discovery uninitialized\n");
@@ -383,27 +388,54 @@ int amdgpu_discovery_get_gfx_info(struct
 	}
 
 	bhdr = (struct binary_header *)adev->mman.discovery_bin;
-	gc_info = (struct gc_info_v1_0 *)(adev->mman.discovery_bin +
+	gc_info = (union gc_info *)(adev->mman.discovery_bin +
 			le16_to_cpu(bhdr->table_list[GC].offset));
-
-	adev->gfx.config.max_shader_engines = le32_to_cpu(gc_info->gc_num_se);
-	adev->gfx.config.max_cu_per_sh = 2 * (le32_to_cpu(gc_info->gc_num_wgp0_per_sa) +
-					      le32_to_cpu(gc_info->gc_num_wgp1_per_sa));
-	adev->gfx.config.max_sh_per_se = le32_to_cpu(gc_info->gc_num_sa_per_se);
-	adev->gfx.config.max_backends_per_se = le32_to_cpu(gc_info->gc_num_rb_per_se);
-	adev->gfx.config.max_texture_channel_caches = le32_to_cpu(gc_info->gc_num_gl2c);
-	adev->gfx.config.max_gprs = le32_to_cpu(gc_info->gc_num_gprs);
-	adev->gfx.config.max_gs_threads = le32_to_cpu(gc_info->gc_num_max_gs_thds);
-	adev->gfx.config.gs_vgt_table_depth = le32_to_cpu(gc_info->gc_gs_table_depth);
-	adev->gfx.config.gs_prim_buffer_depth = le32_to_cpu(gc_info->gc_gsprim_buff_depth);
-	adev->gfx.config.double_offchip_lds_buf = le32_to_cpu(gc_info->gc_double_offchip_lds_buffer);
-	adev->gfx.cu_info.wave_front_size = le32_to_cpu(gc_info->gc_wave_size);
-	adev->gfx.cu_info.max_waves_per_simd = le32_to_cpu(gc_info->gc_max_waves_per_simd);
-	adev->gfx.cu_info.max_scratch_slots_per_cu = le32_to_cpu(gc_info->gc_max_scratch_slots_per_cu);
-	adev->gfx.cu_info.lds_size = le32_to_cpu(gc_info->gc_lds_size);
-	adev->gfx.config.num_sc_per_sh = le32_to_cpu(gc_info->gc_num_sc_per_se) /
-					 le32_to_cpu(gc_info->gc_num_sa_per_se);
-	adev->gfx.config.num_packer_per_sc = le32_to_cpu(gc_info->gc_num_packer_per_sc);
-
+	switch (gc_info->v1.header.version_major) {
+	case 1:
+		adev->gfx.config.max_shader_engines = le32_to_cpu(gc_info->v1.gc_num_se);
+		adev->gfx.config.max_cu_per_sh = 2 * (le32_to_cpu(gc_info->v1.gc_num_wgp0_per_sa) +
+						      le32_to_cpu(gc_info->v1.gc_num_wgp1_per_sa));
+		adev->gfx.config.max_sh_per_se = le32_to_cpu(gc_info->v1.gc_num_sa_per_se);
+		adev->gfx.config.max_backends_per_se = le32_to_cpu(gc_info->v1.gc_num_rb_per_se);
+		adev->gfx.config.max_texture_channel_caches = le32_to_cpu(gc_info->v1.gc_num_gl2c);
+		adev->gfx.config.max_gprs = le32_to_cpu(gc_info->v1.gc_num_gprs);
+		adev->gfx.config.max_gs_threads = le32_to_cpu(gc_info->v1.gc_num_max_gs_thds);
+		adev->gfx.config.gs_vgt_table_depth = le32_to_cpu(gc_info->v1.gc_gs_table_depth);
+		adev->gfx.config.gs_prim_buffer_depth = le32_to_cpu(gc_info->v1.gc_gsprim_buff_depth);
+		adev->gfx.config.double_offchip_lds_buf = le32_to_cpu(gc_info->v1.gc_double_offchip_lds_buffer);
+		adev->gfx.cu_info.wave_front_size = le32_to_cpu(gc_info->v1.gc_wave_size);
+		adev->gfx.cu_info.max_waves_per_simd = le32_to_cpu(gc_info->v1.gc_max_waves_per_simd);
+		adev->gfx.cu_info.max_scratch_slots_per_cu = le32_to_cpu(gc_info->v1.gc_max_scratch_slots_per_cu);
+		adev->gfx.cu_info.lds_size = le32_to_cpu(gc_info->v1.gc_lds_size);
+		adev->gfx.config.num_sc_per_sh = le32_to_cpu(gc_info->v1.gc_num_sc_per_se) /
+			le32_to_cpu(gc_info->v1.gc_num_sa_per_se);
+		adev->gfx.config.num_packer_per_sc = le32_to_cpu(gc_info->v1.gc_num_packer_per_sc);
+		break;
+	case 2:
+		adev->gfx.config.max_shader_engines = le32_to_cpu(gc_info->v2.gc_num_se);
+		adev->gfx.config.max_cu_per_sh = le32_to_cpu(gc_info->v2.gc_num_cu_per_sh);
+		adev->gfx.config.max_sh_per_se = le32_to_cpu(gc_info->v2.gc_num_sh_per_se);
+		adev->gfx.config.max_backends_per_se = le32_to_cpu(gc_info->v2.gc_num_rb_per_se);
+		adev->gfx.config.max_texture_channel_caches = le32_to_cpu(gc_info->v2.gc_num_tccs);
+		adev->gfx.config.max_gprs = le32_to_cpu(gc_info->v2.gc_num_gprs);
+		adev->gfx.config.max_gs_threads = le32_to_cpu(gc_info->v2.gc_num_max_gs_thds);
+		adev->gfx.config.gs_vgt_table_depth = le32_to_cpu(gc_info->v2.gc_gs_table_depth);
+		adev->gfx.config.gs_prim_buffer_depth = le32_to_cpu(gc_info->v2.gc_gsprim_buff_depth);
+		adev->gfx.config.double_offchip_lds_buf = le32_to_cpu(gc_info->v2.gc_double_offchip_lds_buffer);
+		adev->gfx.cu_info.wave_front_size = le32_to_cpu(gc_info->v2.gc_wave_size);
+		adev->gfx.cu_info.max_waves_per_simd = le32_to_cpu(gc_info->v2.gc_max_waves_per_simd);
+		adev->gfx.cu_info.max_scratch_slots_per_cu = le32_to_cpu(gc_info->v2.gc_max_scratch_slots_per_cu);
+		adev->gfx.cu_info.lds_size = le32_to_cpu(gc_info->v2.gc_lds_size);
+		adev->gfx.config.num_sc_per_sh = le32_to_cpu(gc_info->v2.gc_num_sc_per_se) /
+			le32_to_cpu(gc_info->v2.gc_num_sh_per_se);
+		adev->gfx.config.num_packer_per_sc = le32_to_cpu(gc_info->v2.gc_num_packer_per_sc);
+		break;
+	default:
+		dev_err(adev->dev,
+			"Unhandled GC info table %d.%d\n",
+			gc_info->v1.header.version_major,
+			gc_info->v1.header.version_minor);
+		return -EINVAL;
+	}
 	return 0;
 }
--- a/drivers/gpu/drm/amd/include/discovery.h
+++ b/drivers/gpu/drm/amd/include/discovery.h
@@ -143,6 +143,55 @@ struct gc_info_v1_0 {
 	uint32_t gc_num_gl2a;
 };
 
+struct gc_info_v1_1 {
+	struct gpu_info_header header;
+
+	uint32_t gc_num_se;
+	uint32_t gc_num_wgp0_per_sa;
+	uint32_t gc_num_wgp1_per_sa;
+	uint32_t gc_num_rb_per_se;
+	uint32_t gc_num_gl2c;
+	uint32_t gc_num_gprs;
+	uint32_t gc_num_max_gs_thds;
+	uint32_t gc_gs_table_depth;
+	uint32_t gc_gsprim_buff_depth;
+	uint32_t gc_parameter_cache_depth;
+	uint32_t gc_double_offchip_lds_buffer;
+	uint32_t gc_wave_size;
+	uint32_t gc_max_waves_per_simd;
+	uint32_t gc_max_scratch_slots_per_cu;
+	uint32_t gc_lds_size;
+	uint32_t gc_num_sc_per_se;
+	uint32_t gc_num_sa_per_se;
+	uint32_t gc_num_packer_per_sc;
+	uint32_t gc_num_gl2a;
+	uint32_t gc_num_tcp_per_sa;
+	uint32_t gc_num_sdp_interface;
+	uint32_t gc_num_tcps;
+};
+
+struct gc_info_v2_0 {
+	struct gpu_info_header header;
+
+	uint32_t gc_num_se;
+	uint32_t gc_num_cu_per_sh;
+	uint32_t gc_num_sh_per_se;
+	uint32_t gc_num_rb_per_se;
+	uint32_t gc_num_tccs;
+	uint32_t gc_num_gprs;
+	uint32_t gc_num_max_gs_thds;
+	uint32_t gc_gs_table_depth;
+	uint32_t gc_gsprim_buff_depth;
+	uint32_t gc_parameter_cache_depth;
+	uint32_t gc_double_offchip_lds_buffer;
+	uint32_t gc_wave_size;
+	uint32_t gc_max_waves_per_simd;
+	uint32_t gc_max_scratch_slots_per_cu;
+	uint32_t gc_lds_size;
+	uint32_t gc_num_sc_per_se;
+	uint32_t gc_num_packer_per_sc;
+};
+
 typedef struct harvest_info_header {
 	uint32_t signature; /* Table Signature */
 	uint32_t version;   /* Table Version */



  parent reply	other threads:[~2022-01-03 14:30 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-03 14:23 [PATCH 5.10 00/48] 5.10.90-rc1 review Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 01/48] Input: i8042 - add deferred probe support Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 02/48] Input: i8042 - enable deferred probe quirk for ASUS UM325UA Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 03/48] tomoyo: Check exceeded quota early in tomoyo_domain_quota_is_ok() Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 04/48] tomoyo: use hwight16() " Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 05/48] parisc: Clear stale IIR value on instruction access rights trap Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 06/48] platform/x86: apple-gmux: use resource_size() with res Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 07/48] memblock: fix memblock_phys_alloc() section mismatch error Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 08/48] recordmcount.pl: fix typo in s390 mcount regex Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 09/48] selinux: initialize proto variable in selinux_ip_postroute_compat() Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 10/48] scsi: lpfc: Terminate string in lpfc_debugfs_nvmeio_trc_write() Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 11/48] net/mlx5: DR, Fix NULL vs IS_ERR checking in dr_domain_init_resources Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 12/48] net/mlx5e: Wrap the tx reporter dump callback to extract the sq Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 13/48] net/mlx5e: Fix ICOSQ recovery flow for XSK Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 14/48] udp: using datalen to cap ipv6 udp max gso segments Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 15/48] selftests: Calculate udpgso segment count without header adjustment Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 16/48] net: phy: fixed_phy: Fix NULL vs IS_ERR() checking in __fixed_phy_register Greg Kroah-Hartman
2022-01-03 19:02   ` Florian Fainelli
2022-01-03 14:23 ` [PATCH 5.10 17/48] sctp: use call_rcu to free endpoint Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 18/48] net/smc: fix using of uninitialized completions Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 19/48] net: usb: pegasus: Do not drop long Ethernet frames Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 20/48] net: ag71xx: Fix a potential double free in error handling paths Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 21/48] net: lantiq_xrx200: fix statistics of received bytes Greg Kroah-Hartman
2022-01-03 14:23 ` [PATCH 5.10 22/48] NFC: st21nfca: Fix memory leak in device probe and remove Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 23/48] net/smc: improved fix wait on already cleared link Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 24/48] net/smc: dont send CDC/LLC message if link not ready Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 25/48] net/smc: fix kernel panic caused by race of smc_sock Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 26/48] igc: Fix TX timestamp support for non-MSI-X platforms Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 27/48] ionic: Initialize the lif->dbid_inuse bitmap Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 28/48] net/mlx5e: Fix wrong features assignment in case of error Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 29/48] selftests/net: udpgso_bench_tx: fix dst ip argument Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 30/48] net/ncsi: check for error return from call to nla_put_u32 Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 31/48] fsl/fman: Fix missing put_device() call in fman_port_probe Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 32/48] i2c: validate user data in compat ioctl Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 33/48] nfc: uapi: use kernel size_t to fix user-space builds Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 34/48] uapi: fix linux/nfc.h userspace compilation errors Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 35/48] drm/amdgpu: When the VCN(1.0) block is suspended, powergating is explicitly enabled Greg Kroah-Hartman
2022-01-03 14:24 ` Greg Kroah-Hartman [this message]
2022-01-03 14:24 ` [PATCH 5.10 37/48] xhci: Fresco FL1100 controller should not have BROKEN_MSI quirk set Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 38/48] usb: gadget: f_fs: Clear ffs_eventfd in ffs_data_clear Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 39/48] usb: mtu3: add memory barrier before set GPDs HWO Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 40/48] usb: mtu3: fix list_head check warning Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 41/48] usb: mtu3: set interval of FS intr and isoc endpoint Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 42/48] binder: fix async_free_space accounting for empty parcels Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 43/48] scsi: vmw_pvscsi: Set residual data length conditionally Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 44/48] Input: appletouch - initialize work before device registration Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 45/48] Input: spaceball - fix parsing of movement data packets Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 46/48] net: fix use-after-free in tw_timer_handler Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 47/48] perf script: Fix CPU filtering of a scripts switch events Greg Kroah-Hartman
2022-01-03 14:24 ` [PATCH 5.10 48/48] bpf: Add kconfig knob for disabling unpriv bpf by default Greg Kroah-Hartman
2022-01-04  1:27 ` [PATCH 5.10 00/48] 5.10.90-rc1 review Guenter Roeck
2022-01-04  5:43 ` Naresh Kamboju
2022-01-04  9:53 ` Jon Hunter

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220103142054.697936926@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=Hawking.Zhang@amd.com \
    --cc=alexander.deucher@amd.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
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.