Am 02.03.23 um 07:32 schrieb lyndonli:
On GPUs with RAS enabled, below call trace and hang are observed when
shutting down device.
v2: use DRM device unplugged flag instead of shutdown flag as the check to
prevent memory wipe in shutdown stage.
[ +0.00] RIP:
Gpu reset might be needed during driver reloading. To guard that(gpu
reset) work, df cstate needs to be disabled properly.
Signed-off-by: Evan Quan
Change-Id: I5c074c265c0b08a67b6934ae1ad9aa3fed245461
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 9 +
1 file changed, 9 insertions(+)
Am 28.02.23 um 20:13 schrieb Danilo Krummrich:
[SNIP]
+ if (exec->prelocked) {
+ dma_resv_unlock(exec->prelocked->resv);
+ drm_gem_object_put(exec->prelocked);
+ exec->prelocked = NULL;
+ }
Let's say we try to lock 3 objects A, B and C in chronological order
and in
gcc with W=1 reports
drivers/gpu/drm/amd/amdgpu/../pm/swsmu/smu11/vangogh_ppt.c:1600:5:
error: no previous prototype for ‘vangogh_set_apu_thermal_limit’
[-Werror=missing-prototypes]
1600 | int vangogh_set_apu_thermal_limit(struct smu_context *smu, uint32_t
limit)
|
From: Xiaogang Chen
svm_migrate_ram_to_vram migrates a prange from sys ram to vram. The prange may
cross multiple vma. Need remember current dst vram offset in the TTM resource
for
each migration.
Signed-off-by: Xiaogang Chen
---
drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 17
tree/branch:
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
branch HEAD: 1716a175592aff9549a0c07aac8f9cadd03003f5 Add linux-next specific
files for 20230301
Error/Warning reports:
https://lore.kernel.org/oe-kbuild-all/202302111601.jty4lkra-...@intel.com
https
On 3/1/2023 12:54 PM, Felix Kuehling wrote:
Am 2023-03-01 um 11:34 schrieb Xiaogang.Chen:
From: Xiaogang Chen
svm_migrate_ram_to_vram migrates a prange from sys ram to vram. The
prange may
cross multiple vma. Need remember current dst vram offset in the TTM
resource for
each migration.
Am 2023-03-01 um 11:34 schrieb Xiaogang.Chen:
From: Xiaogang Chen
svm_migrate_ram_to_vram migrates a prange from sys ram to vram. The prange may
cross multiple vma. Need remember current dst vram offset in the TTM resource
for
each migration.
Signed-off-by: Xiaogang Chen
The patch looks
On Mon, Jan 23, 2023 at 12:23:19AM +0530, Deepak R Varma wrote:
> On Sun, Jan 15, 2023 at 12:52:10PM -0800, Joe Perches wrote:
> > On Sun, 2023-01-15 at 15:30 +0530, Deepak R Varma wrote:
> > > The if / else block code has same effect irrespective of the logical
> > > evaluation. Hence, simply
From: Le Ma
No mmHDP_READ_CACHE_INVALIDATE register on HDP 4.4.2.
Signed-off-by: Le Ma
Reviewed-by: Hawking Zhang
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c
From: Hawking Zhang
Add HDP IP handling for HDP 4.4.2
Signed-off-by: Hawking Zhang
Reviewed-by: Le Ma
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
This add support for SDMA 4.4.2. The first patch is register
header updates and was too large for the mailing list.
Hawking Zhang (1):
drm/amdgpu: add sdma v4_4_2 ip headers
Le Ma (2):
drm/amdgpu: add sdma v4_4_2 support (v4)
drm/amdgpu: set sdma v4_4_2 ip block
On 2023-03-01 14:24, Chen, Xiaogang wrote:
On 3/1/2023 12:54 PM, Felix Kuehling wrote:
Am 2023-03-01 um 11:34 schrieb Xiaogang.Chen:
From: Xiaogang Chen
svm_migrate_ram_to_vram migrates a prange from sys ram to vram. The
prange may
cross multiple vma. Need remember current dst vram
From: Le Ma
Share same register address with IH 4.4.0.
Signed-off-by: Le Ma
Reviewed-by: Hawking Zhang
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vega20_ih.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
From: Le Ma
No ih2 hardware on IH 4.4.2.
Signed-off-by: Le Ma
Reviewed-by: Hawking Zhang
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vega20_ih.c | 12 +++-
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
From: Hawking Zhang
Add IH IP handling for IH 4.4.2
Signed-off-by: Hawking Zhang
Reviewed-by: Le Ma
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
From: Le Ma
Add support for SDMA 4.4.2.
v1: Create sdma_v4_4_2.[c|h] for initial support (Le)
v2: update amdgpu_ring_init call with atomic score (Hawking)
v3: Squash in sdma_start fixes (Alex)
v4: Comment out currently unused RAS code (Alex)
Signed-off-by: Le Ma
Signed-off-by: Hawking Zhang
From: Le Ma
Use sdma 4.4.2 IP block for chips with sdma 4.4.2 hardware.
Signed-off-by: Le Ma
Reviewed-by: Hawking Zhang
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 4
1 file changed, 4 insertions(+)
diff --git
- adev->in_suspend || adev->ddev.unplugged)
I don't think the code base is correct. Please double check it.
Regards,
Guchun
-Original Message-
From: lyndonli
Sent: Thursday, March 2, 2023 12:57 PM
To: amd-gfx@lists.freedesktop.org
Cc: Koenig, Christian ; Chen, Guchun
;
Reviewed-by: Guchun Chen
Regards,
Guchun
-Original Message-
From: lyndonli
Sent: Thursday, March 2, 2023 2:33 PM
To: amd-gfx@lists.freedesktop.org
Cc: Koenig, Christian ; Chen, Guchun
; Chai, Thomas ; Xu, Feifei
; Li, Lyndon
Subject: [PATCH v2] drm/amdgpu: Fix call trace warning
Reviewed-by: Guchun Chen
Regards,
Guchun
-Original Message-
From: bobzhou
Sent: Thursday, March 2, 2023 2:44 PM
To: amd-gfx@lists.freedesktop.org; Chen, Guchun ; Cui,
Flora ; Shi, Leslie ; Ma, Jun
Cc: Zhou, Bob
Subject: [PATCH] drm/amdgpu: remove unused variable
building with gcc
On GPUs with RAS enabled, below call trace and hang are observed when
shutting down device.
v2: use DRM device unplugged flag instead of shutdown flag as the check to
prevent memory wipe in shutdown stage.
[ +0.00] RIP: 0010:amdgpu_vram_mgr_fini+0x18d/0x1c0 [amdgpu]
[ +0.01] PKRU:
On GPUs with RAS enabled, below call trace and hang are observed when
shutting down device.
v2: use DRM device unplugged flag instead of shutdown flag as the check to
prevent memory wipe in shutdown stage.
[ +0.00] RIP: 0010:amdgpu_vram_mgr_fini+0x18d/0x1c0 [amdgpu]
[ +0.01] PKRU:
Hi Dave, Daniel,
Fixes for 6.3. A bit bigger than usual since it's two weeks worth of fixes.
The following changes since commit a48bba98380cb0b43dcd01d276c7efc282e3c33f:
msm/fbdev: fix unused variable warning with clang. (2023-02-23 09:48:05 +1000)
are available in the Git repository at:
[AMD Official Use Only - General]
After the patch merge, the "tmp" variable will not be used, and a compilation
warning will be generated.
Best Regards,
Kevin
-Original Message-
From: amd-gfx On Behalf Of Yaoyao Lei
Sent: Monday, February 27, 2023 2:46 PM
To:
Sorry, the code base was incorrect.
I just sent out a v2.
Regards,
Lyndon
> -Original Message-
> From: Chen, Guchun
> Sent: Thursday, March 2, 2023 1:28 PM
> To: Li, Lyndon ; amd-gfx@lists.freedesktop.org
> Cc: Koenig, Christian ; Chai, Thomas
> ; Xu, Feifei ; Li, Lyndon
>
> Subject:
building with gcc and W=1 reports
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c:7292:11: error:
unused variable 'tmp' [-Werror=unused-variable]
uint32_t tmp;
^~~
tmp is not used so remove it.
Signed-off-by: bobzhou
---
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 1 -
1 file changed, 1
Am 2023-03-01 um 18:46 schrieb Daniel Phillips:
Our assumptions about how much KFD memory is currently available for
allocation may be violated by various complexities so we define the
reported value as advisory, however we should never report negative
availability.
Signed-off-by: Daniel
Our assumptions about how much KFD memory is currently available for
allocation may be violated by various complexities so we define the
reported value as advisory, however we should never report negative
availability.
Signed-off-by: Daniel Phillips
---
Our assumptions about how much KFD memory is currently available for
allocation may be violated by various complexities so we define the
reported value as advisory, however we should never report negative
availability.
Signed-off-by: Daniel Phillips
---
[AMD Official Use Only - General]
Reviewed-by: Stanley.Yang
Regards,
Stanley
> -Original Message-
> From: amd-gfx On Behalf Of
> Candice Li
> Sent: Wednesday, March 1, 2023 2:10 PM
> To: amd-gfx@lists.freedesktop.org
> Cc: Li, Candice
> Subject: [PATCH] drm/amd/pm: Enable ecc_info
ret = smu_cmn_get_smc_version(smu, _version, _version);
+ if (ret)
+ return -EOPNOTSUPP;
A nit-pick comment, looks querying if_version in this case is redundant,
passing NULL may be more efficient.
Regards,
Guchun
-Original Message-
From: amd-gfx On Behalf Of
On Tue, 28 Feb 2023 10:26:04 -0500
André Almeida wrote:
> Hi Pekka,
>
> Thank you for your feedback,
>
> On 2/28/23 05:02, Pekka Paalanen wrote:
> > On Mon, 27 Feb 2023 15:40:00 -0500
> > André Almeida wrote:
> >
> >> Create a section that specifies how to deal with DRM device resets for
>
33 matches
Mail list logo
