Under bare metal, there is no more else to take
care of the GPU register access through MMIO.
Under Virtualization, to access GPU register is
implemented through KIQ during run-time due to
world-switch.
Therefore, under SR-IOV user can only access
debugfs to r/w GPU registers when meets all
three
[AMD Official Use Only - Internal Distribution Only]
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: Gao, Likun
Sent: Thursday, April 9, 2020 12:13
To: amd-gfx@lists.freedesktop.org
Cc: Zhang, Hawking ; Gao, Likun
Subject: [PATCH] Revert "drm/amdgpu: change SH MEM
From: Likun Gao
This reverts commit 8d2afe375f6ed9d84ddca234d49a53cf875f9042.
Revert the auto alignment mode set of SH MEM config, as it will result
to OCL Conformance Test fail.
---
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/driver
amdgpu uses lots of pr_* calls for printing error messages.
With this prefix, errors shall be more obvious to the end
use regarding its origin, and may help debugging.
Prefix format:
[xxx.x] amdgpu: ...
Signed-off-by: Aurabindo Pillai
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 6 ++
1 f
Let format prefixes take care of printing the module name
through pr_fmt and dev_fmt definitions.
Signed-off-by: Aurabindo Pillai
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 6 +++---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 ++--
drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
Changes in v2
* Add dev_fmt format prefix
* Removed hardcoded module names in pr_* and dev_* prints
Aurabindo Pillai (3):
drm/amd/amdgpu: add prefix for pr_* prints
drm/amd/amdgpu: add print prefix for dev_* variants
drm/amd/amdgpu: remove hardcoded module name in prints
drivers/gpu/drm/a
Hi Joe,
On Wed, Apr 8, 2020 at 11:37 AM Joe Perches wrote:
>
> All the embedded uses of "amdgpu:" in logging
> messages should also be deleted.
>
> $ git grep -P '(?:dev_|pr_).*"amdgpu:' drivers/gpu/drm/amd/amdgpu/
> drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c:
> pr_err("amdgp
Define dev_fmt macro for informative print messages
Signed-off-by: Aurabindo Pillai
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 67d654a89..7a52d37e6 100644
--- a
The delayed delete list is per device which might be very huge. And in
a heavy workload test, the list might always not be empty. That will
trigger any RCU stall warnings or softlockups in non-preemptible kernels
Lets do schedule out if possible in that case.
Signed-off-by: xinhui pan
---
driver
Am 2020-04-08 um 9:27 a.m. schrieb Deucher, Alexander:
>
> [AMD Public Use]
>
>
> Acked-by: Alex Deucher
>
> *From:* amd-gfx on behalf of
> Jack Zhang
> *Sent:* Wednesday, April 8, 2020 6:33 AM
> *To:* amd-gfx@lists.freede
Hi Dave, Daniel,
Fixes for 5.7.
The following changes since commit 3148a6a0ef3cf93570f30a477292768f7eb5d3c3:
drm/amdkfd: kfree the wrong pointer (2020-04-01 14:44:22 -0400)
are available in the Git repository at:
git://people.freedesktop.org/~agd5f/linux tags/amd-drm-fixes-5.7-2020-04-08
Without CONFIG_DEBUG_FS, this newly added function causes a harmless
warning:
amdgpu_dm/amdgpu_dm.c: In function 'amdgpu_dm_connector_late_register':
amdgpu_dm/amdgpu_dm.c:4723:30: error: unused variable 'amdgpu_dm_connector'
[-Werror=unused-variable]
4723 | struct amdgpu_dm_connector *amdgpu_d
On Wed, Apr 8, 2020 at 3:13 PM Arnd Bergmann wrote:
>
> Without CONFIG_DEBUG_FS, this newly added function causes a harmless
> warning:
>
> amdgpu_dm/amdgpu_dm.c: In function 'amdgpu_dm_connector_late_register':
> amdgpu_dm/amdgpu_dm.c:4723:30: error: unused variable 'amdgpu_dm_connector'
> [-Wer
Add insn_is_fpu(insn) which tells that the insn is
whether touch the FPU/SSE/MMX register or the instruction
of FP coprocessor.
Signed-off-by: Masami Hiramatsu
---
Changes in v2:
- Introduce FPU superscript.
- Fix to add INAT_FPUIFVEX for variant if the first opcode has no
last prefix super
On Wed, 2020-04-08 at 09:37 -0400, Aurabindo Pillai wrote:
> amdgpu uses lots of pr_* calls for printing error messages.
> With this prefix, errors shall be more obvious to the end
> use regarding its origin, and may help debugging.
>
> Prefix format:
>
> [xxx.x] amdgpu: ...
[]
> diff --git a
Am 08.04.20 um 16:00 schrieb Alex Deucher:
On Tue, Apr 7, 2020 at 5:50 AM Christian König
wrote:
Am 07.04.20 um 11:12 schrieb Huang Rui:
ttm_trace_dma_map/ttm_trace_dma_unmap is never used anymore.
v2: remove the file completely
Signed-off-by: Huang Rui
Reviewed-by: Christian König
Can
On Tue, Apr 7, 2020 at 5:50 AM Christian König
wrote:
>
> Am 07.04.20 um 11:12 schrieb Huang Rui:
> > ttm_trace_dma_map/ttm_trace_dma_unmap is never used anymore.
> >
> > v2: remove the file completely
> >
> > Signed-off-by: Huang Rui
>
> Reviewed-by: Christian König
>
Can one of you make sure
On Wed, Apr 8, 2020 at 3:38 AM Thomas Zimmermann wrote:
>
> Hi Alex
>
> Am 07.02.20 um 20:50 schrieb Alex Deucher:
> > These are deprecated and the drm will soon start warning when drivers still
> > use them. It was a long and twisty road, but seems to work.
>
> Are you going to convert radeon as
On Tue, Apr 7, 2020 at 2:30 PM Aurabindo Pillai wrote:
>
> Execution will only reach here if the asserted condition is true.
> Hence there is no need for the additional check.
>
> Signed-off-by: Aurabindo Pillai
Applied. Thanks!
Alex
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 10 -
amdgpu uses lots of pr_* calls for printing error messages.
With this prefix, errors shall be more obvious to the end
use regarding its origin, and may help debugging.
Prefix format:
[xxx.x] amdgpu: ...
Signed-off-by: Aurabindo Pillai
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 6 ++
1 f
Reviewed-by: Mikita Lipski
Thanks,
Mikita
On 2020-04-08 9:31 a.m., Alex Deucher wrote:
fixes unused variable warning.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/d
Am 08.04.20 um 15:31 schrieb Alex Deucher:
fixes unused variable warning.
Signed-off-by: Alex Deucher
Acked-by: Christian König
---
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm
fixes unused variable warning.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 1155fb
[AMD Public Use]
Acked-by: Alex Deucher
From: amd-gfx on behalf of Jack Zhang
Sent: Wednesday, April 8, 2020 6:33 AM
To: amd-gfx@lists.freedesktop.org
Cc: Zhang, Jack (Jian)
Subject: [PATCH] drm/amdgpu/sriov move program of sq_config to PF
Since SQ_CONFIG wi
[AMD Public Use]
Reviewed-by: Alex Deucher
From: Liu, Aaron
Sent: Wednesday, April 8, 2020 2:45 AM
To: amd-gfx@lists.freedesktop.org
Cc: Deucher, Alexander ; Huang, Ray
; Dai, Yuxian (David) ; Liu, Aaron
Subject: [PATCH] drm/amdgpu: unify fw_write_wait for ne
On 4/8/20 12:16 PM, Christian König wrote:
The buffer needs to be two bytes larger when the maximum is 32.
Signed-off-by: Christian König
Acked-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers
-Original Message-
From: Jack Zhang
Sent: Wednesday, April 8, 2020 6:34 PM
To: amd-gfx@lists.freedesktop.org
Cc: Zhang, Jack (Jian)
Subject: [PATCH] drm/amdgpu/sriov move program of sq_config to PF
Since SQ_CONFIG will influence performance behavior, it's unsafe to program in
VF. The
Since SQ_CONFIG will influence performance behavior, it's
unsafe to program in VF. Therefore, program of SQ_CONFIG
need to be moved to PF under siov. Skip it here in guest driver.
Signed-off-by: Jack Zhang
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 3 ++-
1 file changed, 2 insertions(+), 1 dele
The buffer needs to be two bytes larger when the maximum is 32.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c
b/drivers/gpu/drm/amd/amdgpu/amdgp
On Wed, Apr 08, 2020 at 02:45:48PM +0800, Liu, Aaron wrote:
> Make the fw_write_wait default case true since presumably all new
> gfx9 asics will have updated firmware. That is using unique WAIT_REG_MEM
> packet with opration=1.
>
> Signed-off-by: Aaron Liu
> ---
Acked-by: Huang Rui
> drivers
Hi Alex
Am 07.02.20 um 20:50 schrieb Alex Deucher:
> These are deprecated and the drm will soon start warning when drivers still
> use them. It was a long and twisty road, but seems to work.
Are you going to convert radeon as well? It's the only remaining KMS
driver that uses load.
Best regards
Am 07.04.20 um 21:24 schrieb Felix Kuehling:
Am 2020-04-07 um 10:27 a.m. schrieb Christian König:
For HMM support we need the ability to invalidate PTEs from
a MM callback where we can't lock the root PD.
Add a new flag to better support this instead of assuming
that all invalidation updates ar
32 matches
Mail list logo