[AMD Official Use Only - Internal Distribution Only]
Christian
what feasible and practice now is:
1) we implement the advanced TDR mode in upstream first (so we can copy the
same scheme in our LTS kernel) -- if you want we can avoid change drm/scheduler
part code, but that one is already reject
[AMD Official Use Only - Internal Distribution Only]
Hi, Christian,
Since this change is a bit critical to our project, we would be grateful that
could get your review.
Are there anything that's is not clear enough I could help to explain?
Again,
Thanks for your huge help to our problem.
Than
[AMD Public Use]
This seems a duplicate of dev_pm_info states. Can't we reuse that?
Thanks,
Lijo
-Original Message-
From: amd-gfx On Behalf Of Alex Deucher
Sent: Tuesday, March 9, 2021 9:40 AM
To: amd-gfx@lists.freedesktop.org
Cc: Deucher, Alexander
Subject: [PATCH 4/7] drm/amdgpu: tra
On Tue, Mar 09, 2021 at 12:45:44PM +0800, Liang, Prike wrote:
>
>
> > -Original Message-
> > From: Alex Deucher
> > Sent: Tuesday, March 9, 2021 12:07 PM
> > To: Liang, Prike
> > Cc: amd-gfx list ; Deucher, Alexander
> > ; Huang, Ray
> > Subject: Re: [PATCH] drm/amdgpu: fix the hiberna
[AMD Public Use]
Thanks Anna. Result is good on SRIOV guest driver as well. Will push with
Reviewed-by: Hawking Zhang
Tested-by Anna Jin < anna@amd.com>
Thanks,
Feifei
-Original Message-
From: Zhang, Hawking
Sent: 2021年3月5日 下午 8:51
To: Xu, Feifei ; amd-gfx@lists.freedesktop.org
> -Original Message-
> From: Alex Deucher
> Sent: Tuesday, March 9, 2021 12:07 PM
> To: Liang, Prike
> Cc: amd-gfx list ; Deucher, Alexander
> ; Huang, Ray
> Subject: Re: [PATCH] drm/amdgpu: fix the hibernation suspend with s0ix
>
> On Mon, Mar 8, 2021 at 10:52 PM Prike Liang wrote:
We reuse the same suspend and resume functions for
all of the pmops states, so flag what state we are
in so that we can alter behavior deeper in the driver
depending on the current flow.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 20 +++-
drivers/gpu/drm/amd/
There's no need to keep vgaswitcheroo around for HG
systems. They don't use muxes and their power control
is handled via ACPI.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h| 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 38 +-
drivers/gpu/
We only need special handling for the S0ix suspend and resume
cases, legacy S3/S4/shutdown/reboot/reset should use the
standard code pathes. This should fix systems with S0ix
plus legacy S4.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h| 6 --
drivers/gpu/drm/
Vram is system memory, so no need to evict.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 8 +++-
1 file changed, 3 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 4b29b8
Move the non-DC specific code into the DCE IP blocks similar
to how we handle DC. This cleans up the common suspend
and resume pathes.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 82 +--
drivers/gpu/drm/amd/amdgpu/amdgpu_display.c | 88 +
Once the device has runtime suspended, we don't need to power it
back up again for system suspend. Likewise for resume, we don't
to power up the device again on resume only to power it back off
again via runtime pm because it's still idle.
v2: add DPM_FLAG_SMART_PREPARE as well
Acked-by: Rajnees
as per:
https://www.kernel.org/doc/html/latest/driver-api/pm/devices.html
The prepare callback is required to support the DPM_FLAG_SMART_SUSPEND
driver flag. This allows runtime pm to auto complete when the
system goes into suspend avoiding a wake up on suspend and on resume.
Apply this for hybri
On Mon, Mar 8, 2021 at 10:52 PM Prike Liang wrote:
>
> During system hibernation suspend still need un-gate gfx CG/PG firstly to
> handle HW
> status check before HW resource destory.
>
> Signed-off-by: Prike Liang
This is fine for stable, but we should work on cleaning this up. I
have a patch
During system hibernation suspend still need un-gate gfx CG/PG firstly to
handle HW
status check before HW resource destory.
Signed-off-by: Prike Liang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdg
On Mon, Mar 8, 2021 at 4:36 PM Mark Yacoub wrote:
>
> From: Mark Yacoub
>
> To initialize the framebuffer, call drm_gem_fb_init_with_funcs which
> verifies that the BO size can fit the FB size by calculating the minimum
> expected size of each plane.
>
> The bug was caught using igt-gpu-tools tes
Am 2021-03-05 um 10:06 a.m. schrieb Nirmoy Das:
> These members are only needed for BOs created by
> amdgpu_gem_object_create(), so we can remove these from the
> base class.
>
> CC: felix.kuehl...@amd.com
> Signed-off-by: Nirmoy Das
Acked-by: Felix Kuehling
> ---
> drivers/gpu/drm/amd/amdgpu
Using 'imply AMD_IOMMU_V2' does not guarantee that the driver can link
against the exported functions. If the GPU driver is built-in but the
IOMMU driver is a loadable module, the kfd_iommu.c file is indeed
built but does not work:
x86_64-linux-ld: drivers/gpu/drm/amd/amdkfd/kfd_iommu.o: in functi
Am 2021-03-08 um 3:45 p.m. schrieb Arnd Bergmann:
> From: Arnd Bergmann
>
> Using 'imply AMD_IOMMU_V2' does not guarantee that the driver can link
> against the exported functions. If the GPU driver is built-in but the
> IOMMU driver is a loadable module, the kfd_iommu.c file is indeed
> built but
update secure display TA header file.
Signed-off-by: Jinzhou Su
---
drivers/gpu/drm/amd/amdgpu/amdgpu_securedisplay.c | 3 +++
drivers/gpu/drm/amd/amdgpu/ta_secureDisplay_if.h | 1 +
2 files changed, 4 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_securedisplay.c
b/drivers/gpu/
[AMD Official Use Only - Internal Distribution Only]
Hi, Christian,
amdgpu_device_skip_hw_access will always assert in reset thread, which
seems not a good idea.
Best Regards
Dennis Li
-Original Message-
From: Christian König
Sent: Tuesday, March 9, 2021 2:07 AM
To: amd-gfx@lists
The nomination period for the 2021 X.Org Foundation Board of Directors
Election closed yesterday and the election is rapidly approaching. We
currently only see membership renewals for 59 people.
If you have not renewed your membership please do so by Thursday, Mar 11
at https://members.x.org.
On 2021-03-08 7:33 a.m., Jack Zhang wrote:
[Why]
Previous tdr design treats the first job in job_timeout as the bad job.
But sometimes a later bad compute job can block a good gfx job and
cause an unexpected gfx job timeout because gfx and compute ring share
internal GC HW mutually.
[How]
Thi
On Sun, Mar 7, 2021 at 10:14 PM Jiapeng Chong
wrote:
>
> Fix the following coccicheck warnings:
>
> ./drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c:1600:40-45: WARNING: conversion
> to bool not needed here.
>
> ./drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c:1598:40-45: WARNING: conversion
> to bool not needed
On Sun, Mar 7, 2021 at 10:00 PM Jiapeng Chong
wrote:
>
> Fix the following coccicheck warnings:
>
> ./drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c:561:34-39: WARNING:
> conversion to bool not needed here.
>
> Reported-by: Abaci Robot
> Signed-off-by: Jiapeng Chong
This patch was already ap
From: Mark Yacoub
To initialize the framebuffer, call drm_gem_fb_init_with_funcs which
verifies that the BO size can fit the FB size by calculating the minimum
expected size of each plane.
The bug was caught using igt-gpu-tools test: kms_addfb_basic.too-high
and kms_addfb_basic.bo-too-small
Tes
Applied. Thanks!
Alex
On Sat, Mar 6, 2021 at 6:05 AM wrote:
>
> From: Zhang Yunkai
>
> 'dce110_resource.h' included in 'dcn21_resource.c' is duplicated.
> 'hw_gpio.h' included in 'hw_factory_dce110.c' is duplicated.
>
> Signed-off-by: Zhang Yunkai
> ---
> drivers/gpu/drm/amd/display/dc/dcn21
Applied. Thanks!
Alex
On Sat, Mar 6, 2021 at 5:48 AM wrote:
>
> From: Zhang Yunkai
>
> 'drm/drm_hdcp.h' included in 'amdgpu_dm.c' is duplicated.
> It is also included in the 79th line.
>
> Signed-off-by: Zhang Yunkai
> ---
> drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 1 -
> 1 file c
On Thu, Mar 4, 2021 at 2:15 PM Mark Yacoub wrote:
>
> From: Mark Yacoub
>
> To initialize the framebuffer, use drm_gem_fb_init_with_funcs which
> verifies that the BO size can fit the FB size by calculating the minimum
> expected size of each plane.
>
> The bug was caught using igt-gpu-tools test
On Fri, Mar 5, 2021 at 1:25 AM Evan Quan wrote:
>
> The "/ 10" should be applied to the right-hand operand instead of
> the left-hand one.
>
> Change-Id: Ie730a1981aa5dee45cd6c3efccc7fb0f088cd679
> Signed-off-by: Evan Quan
> Noticed-by: Georgios Toptsidis
Reviewed-by: Alex Deucher
> ---
> dr
Applied. Thanks!
Alex
On Thu, Mar 4, 2021 at 11:02 PM Quan, Evan wrote:
>
> [AMD Public Use]
>
> Thanks. Reviewed-by: Evan Quan
>
> -Original Message-
> From: Jia-Ju Bai
> Sent: Friday, March 5, 2021 11:54 AM
> To: Deucher, Alexander ; Koenig, Christian
> ; airl...@linux.ie; dan...@f
[AMD Official Use Only - Internal Distribution Only]
> -Original Message-
> From: Koenig, Christian
> Sent: Saturday, March 6, 2021 4:12 AM
> To: Kim, Jonathan ; Christian König
> ; amd-gfx@lists.freedesktop.org
> Cc: Yang, Philip ; Kuehling, Felix
>
> Subject: Re: [PATCH] drm/amdgpu: ad
From: Arnd Bergmann
Using 'imply AMD_IOMMU_V2' does not guarantee that the driver can link
against the exported functions. If the GPU driver is built-in but the
IOMMU driver is a loadable module, the kfd_iommu.c file is indeed
built but does not work:
x86_64-linux-ld: drivers/gpu/drm/amd/amdkfd/
On 2021-03-08 3:18 p.m., Daniel Vetter wrote:
On Fri, Mar 5, 2021 at 10:24 AM Michel Dänzer wrote:
On 2021-03-04 7:26 p.m., Kazlauskas, Nicholas wrote:
On 2021-03-04 10:35 a.m., Michel Dänzer wrote:
On 2021-03-04 4:09 p.m., Kazlauskas, Nicholas wrote:
On 2021-03-04 4:05 a.m., Michel Dänzer
Am 08.03.21 um 21:34 schrieb Alex Deucher:
On Mon, Mar 8, 2021 at 3:20 PM Christian König wrote:
Am 08.03.21 um 16:37 schrieb Nirmoy Das:
FB BO should not be ttm_bo_type_kernel type and
amdgpufb_create_pinned_object() pins the FB BO anyway.
Mhm, why the heck was that a kernel object?
Maybe b
On Mon, Mar 8, 2021 at 9:12 PM Christian König
wrote:
> Am 08.03.21 um 21:02 schrieb Felix Kuehling:
> > Am 2021-03-08 um 2:33 p.m. schrieb Arnd Bergmann:
> > I don't want to create a hard dependency on AMD_IOMMU_V2 if I can avoid
> > it, because it is only really needed for a small number of AMD
On Mon, Mar 8, 2021 at 3:20 PM Christian König wrote:
>
> Am 08.03.21 um 16:37 schrieb Nirmoy Das:
> > FB BO should not be ttm_bo_type_kernel type and
> > amdgpufb_create_pinned_object() pins the FB BO anyway.
>
> Mhm, why the heck was that a kernel object?
Maybe because the fbcon was the main us
Am 08.03.21 um 16:37 schrieb Nirmoy Das:
Implement a new struct amdgpu_bo_user as subclass of
struct amdgpu_bo and a function to created amdgpu_bo_user
bo with a flag to identify the owner.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 28 ++
Am 08.03.21 um 16:37 schrieb Nirmoy Das:
Tiling flag and metadata are only needed for BOs created by
amdgpu_gem_object_create(), so we can remove those from the
base class.
CC: felix.kuehl...@amd.com
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 2 -
drivers/gpu
Am 08.03.21 um 16:37 schrieb Nirmoy Das:
FB BO should not be ttm_bo_type_kernel type and
amdgpufb_create_pinned_object() pins the FB BO anyway.
Mhm, why the heck was that a kernel object?
Signed-off-by: Nirmoy Das
Acked-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c |
On Fri, Mar 5, 2021 at 10:24 AM Michel Dänzer wrote:
>
> On 2021-03-04 7:26 p.m., Kazlauskas, Nicholas wrote:
> > On 2021-03-04 10:35 a.m., Michel Dänzer wrote:
> >> On 2021-03-04 4:09 p.m., Kazlauskas, Nicholas wrote:
> >>> On 2021-03-04 4:05 a.m., Michel Dänzer wrote:
> On 2021-03-03 8:17 p
Am 08.03.21 um 21:02 schrieb Felix Kuehling:
Am 2021-03-08 um 2:33 p.m. schrieb Arnd Bergmann:
On Mon, Mar 8, 2021 at 8:11 PM Felix Kuehling wrote:
Am 2021-03-08 um 2:05 p.m. schrieb Arnd Bergmann:
On Mon, Mar 8, 2021 at 5:24 PM Felix Kuehling wrote:
The driver build should work without IOM
Am 2021-03-08 um 2:33 p.m. schrieb Arnd Bergmann:
> On Mon, Mar 8, 2021 at 8:11 PM Felix Kuehling wrote:
>> Am 2021-03-08 um 2:05 p.m. schrieb Arnd Bergmann:
>>> On Mon, Mar 8, 2021 at 5:24 PM Felix Kuehling
>>> wrote:
The driver build should work without IOMMUv2. In amdkfd/Makefile, we
>>>
On Mon, Mar 8, 2021 at 8:11 PM Felix Kuehling wrote:
>
> Am 2021-03-08 um 2:05 p.m. schrieb Arnd Bergmann:
> > On Mon, Mar 8, 2021 at 5:24 PM Felix Kuehling
> > wrote:
> >> The driver build should work without IOMMUv2. In amdkfd/Makefile, we
> >> have this condition:
> >>
> >> ifneq ($(CONFIG_AM
On Mon, Mar 1, 2021 at 5:25 AM Christian König
wrote:
>
>
>
> Am 01.03.21 um 11:04 schrieb Daniel Vetter:
> > On Mon, Mar 1, 2021 at 10:56 AM Thomas Zimmermann
> > wrote:
> >> (cc'ing amd devs)
> >>
> >> Hi
> >>
> >> Am 28.02.21 um 17:10 schrieb Pavel Turinský:
> >>> The checks were removed in c
Am 2021-03-08 um 2:05 p.m. schrieb Arnd Bergmann:
> On Mon, Mar 8, 2021 at 5:24 PM Felix Kuehling wrote:
>> The driver build should work without IOMMUv2. In amdkfd/Makefile, we
>> have this condition:
>>
>> ifneq ($(CONFIG_AMD_IOMMU_V2),)
>> AMDKFD_FILES += $(AMDKFD_PATH)/kfd_iommu.o
>> endif
>>
>
On Mon, Mar 8, 2021 at 5:24 PM Felix Kuehling wrote:
>
> The driver build should work without IOMMUv2. In amdkfd/Makefile, we
> have this condition:
>
> ifneq ($(CONFIG_AMD_IOMMU_V2),)
> AMDKFD_FILES += $(AMDKFD_PATH)/kfd_iommu.o
> endif
>
> In amdkfd/kfd_iommu.h we define inline stubs of the func
Series is:
Reviewed-by: Alex Deucher
On Mon, Mar 8, 2021 at 1:36 PM Christian König
wrote:
>
> Silence static checker warning.
>
> Signed-off-by: Christian König
> ---
> drivers/gpu/drm/radeon/radeon_ttm.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm
Silence static checker warning.
Signed-off-by: Christian König
---
drivers/gpu/drm/radeon/radeon_ttm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c
b/drivers/gpu/drm/radeon/radeon_ttm.c
index 5ea647f454d3..808941e31d34 100644
--- a/driv
When AGP is compiled as module radeon must be compiled as module as
well.
Signed-off-by: Christian König
---
drivers/gpu/drm/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index e392a90ca687..85b79a7fee63 100644
--- a/drivers/gpu/drm
Otherwise we will run into a NULL ptr deref.
Signed-off-by: Christian König
Bug: https://bugzilla.kernel.org/show_bug.cgi?id=212137
---
drivers/gpu/drm/radeon/radeon.h | 2 ++
drivers/gpu/drm/radeon/radeon_gem.c | 4 ++--
drivers/gpu/drm/radeon/radeon_prime.c | 2 ++
3 files changed, 6 i
Am 08.03.21 um 16:37 schrieb Nirmoy Das:
Allow allocating BO structures with different structure size
than struct amdgpu_bo.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 9 +++--
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | 1 +
2 files changed, 8 insertion
From: Dennis Li
When recovery thread has begun GPU reset, there should be not other
threads to access hardware, otherwise system randomly hang.
v2 (chk): rewritten from scratch, use trylock and lockdep instead of
hand wiring the logic.
Signed-off-by: Dennis Li
Signed-off-by: Christia
Sure, patch 4 Reviewed-by: Andrey Grodzovsky andrey.grodzov...@amd.com
and patch 5 Acked-by: Andrey Grodzovsky andrey.grodzov...@amd.com since
I am not sure about the KFD bits.
Andrey
On 2021-03-08 11:10 a.m., Liu, Shaoyun wrote:
[AMD Official Use Only - Internal Distribution Only]
Hi, Andre
The driver build should work without IOMMUv2. In amdkfd/Makefile, we
have this condition:
ifneq ($(CONFIG_AMD_IOMMU_V2),)
AMDKFD_FILES += $(AMDKFD_PATH)/kfd_iommu.o
endif
In amdkfd/kfd_iommu.h we define inline stubs of the functions that are
causing your link-failures if IOMMU_V2 is not enabled:
[AMD Official Use Only - Internal Distribution Only]
Hi, Andrey.
The first 3 patches in this serial already been acked by Alex. D, can you help
review the rest two ?
Thanks
Shaoyun.liu
-Original Message-
From: Grodzovsky, Andrey
Sent: Monday, March 8, 2021 10:53 AM
To: Liu, Shaoyun
I see, thanks for explaning.
Andrey
On 2021-03-08 10:27 a.m., Liu, Shaoyun wrote:
[AMD Official Use Only - Internal Distribution Only]
Check the function amdgpu_xgmi_add_device, when psp XGMI TA is bot available , the
driver will assign a faked hive ID 0x10 for all GPUs, it means all GPU w
Use amdgpu_bo_create_user() for all the BO allocations for
ttm_bo_type_device type.
CC: felix.kuehl...@amd.com
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 4 +++-
drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c| 4 +++-
2 files changed, 6 insertions(+), 2 deletions(-)
Tiling flag and metadata are only needed for BOs created by
amdgpu_gem_object_create(), so we can remove those from the
base class.
CC: felix.kuehl...@amd.com
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 2 -
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 59
FB BO should not be ttm_bo_type_kernel type and
amdgpufb_create_pinned_object() pins the FB BO anyway.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
b/drivers/gpu/dr
Allow allocating BO structures with different structure size
than struct amdgpu_bo.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 9 +++--
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | 1 +
2 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu
Implement a new struct amdgpu_bo_user as subclass of
struct amdgpu_bo and a function to created amdgpu_bo_user
bo with a flag to identify the owner.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 28 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h |
From: Arnd Bergmann
Using 'imply AMD_IOMMU_V2' does not guarantee that the driver can link
against the exported functions. If the GPU driver is built-in but the
IOMMU driver is a loadable module, the kfd_iommu.c file is indeed
built but does not work:
x86_64-linux-ld: drivers/gpu/drm/amd/amdkfd/
[AMD Official Use Only - Internal Distribution Only]
Check the function amdgpu_xgmi_add_device, when psp XGMI TA is bot available ,
the driver will assign a faked hive ID 0x10 for all GPUs, it means all GPU
will belongs to one same hive . So I can still use hive->tb to sync the reset
on all
The radeon warning is trivial to fix, going to send out a patch in a few
moments.
Regards,
Christian.
Am 08.03.21 um 13:14 schrieb Christophe Leroy:
+Evgeniy for W1 Dallas
+Alex & Christian for RADEON
Le 07/03/2021 à 11:23, kernel test robot a écrit :
Hi Christophe,
I love your patch! Perha
+Evgeniy for W1 Dallas
+Alex & Christian for RADEON
Le 07/03/2021 à 11:23, kernel test robot a écrit :
Hi Christophe,
I love your patch! Perhaps something to improve:
[auto build test WARNING on powerpc/next]
[also build test WARNING on v5.12-rc2 next-20210305]
[If your patch is applied to the
On Fri, 05 Mar 2021, Roland Scheidegger wrote:
> The vmwgfx ones look all good to me, so for
> 23-53: Reviewed-by: Roland Scheidegger
> That said, they were already signed off by Zack, so not sure what
> happened here.
Yes, they were accepted at one point, then dropped without a reason.
Since I
On 3/8/21 2:58 PM, Christian König wrote:
Am 08.03.21 um 14:56 schrieb Nirmoy:
On 3/5/21 4:11 PM, Christian König wrote:
We might need to use this for the KFD as well.
Do you mean for amdgpu_amdkfd_alloc_gws() ?
For example, yes. Basically all places where KFD allocated an BO with
the
Am 08.03.21 um 14:56 schrieb Nirmoy:
On 3/5/21 4:11 PM, Christian König wrote:
We might need to use this for the KFD as well.
Do you mean for amdgpu_amdkfd_alloc_gws() ?
For example, yes. Basically all places where KFD allocated an BO with
the TTM type device/user.
Regards,
Christian.
On 3/5/21 4:11 PM, Christian König wrote:
We might need to use this for the KFD as well.
Do you mean for amdgpu_amdkfd_alloc_gws() ?
Regards,
Nirmoy
Christian.
Am 05.03.21 um 15:35 schrieb Nirmoy Das:
GEM objects encapsulate amdgpu_bo for userspace applications.
Now that we have a ne
Separate the drm_mm_node walking from the actual handling.
Signed-off-by: Christian König
Acked-by: Oak Zeng
Tested-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 67 +++--
1 file changed, 18 insertions(+), 49 deletions(-)
diff --git a/drivers/gpu/drm/amd/amd
Separate the drm_mm_node walking from the actual handling.
Signed-off-by: Christian König
Acked-by: Oak Zeng
Tested-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 54 +-
1 file changed, 18 insertions(+), 36 deletions(-)
diff --git a/drivers/gpu/drm/amd/amd
Separate the drm_mm_node walking from the actual handling.
Signed-off-by: Christian König
Acked-by: Oak Zeng
Tested-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 14 --
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_tt
Separate the drm_mm_node walking from the actual handling.
Signed-off-by: Christian König
Acked-by: Oak Zeng
Tested-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 65 ++---
1 file changed, 15 insertions(+), 50 deletions(-)
diff --git a/drivers/gpu/drm/amd/amd
Separate the drm_mm_node walking from the actual handling.
Signed-off-by: Christian König
Acked-by: Oak Zeng
Tested-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 8 +++-
1 file changed, 3 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
b/
Separate the drm_mm_node walking from the actual handling.
Signed-off-by: Christian König
Acked-by: Oak Zeng
Tested-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 10 ++
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
Allows to walk over the drm_mm nodes in a TTM resource object.
Signed-off-by: Christian König
Acked-by: Oak Zeng
Tested-by: Nirmoy Das
---
.../gpu/drm/amd/amdgpu/amdgpu_res_cursor.h| 105 ++
1 file changed, 105 insertions(+)
create mode 100644 drivers/gpu/drm/amd/amdgpu/am
Separate the drm_mm_node walking from the actual handling.
Signed-off-by: Christian König
Acked-by: Oak Zeng
Tested-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 87 -
1 file changed, 26 insertions(+), 61 deletions(-)
diff --git a/drivers/gpu/drm/amd/amd
[AMD Public Use]
Reviewed-by: Hawking Zhang
Regards,
Hawking
From: Lazar, Lijo
Sent: Monday, March 8, 2021 21:16
To: Lazar, Lijo ; amd-gfx@lists.freedesktop.org
Cc: Deucher, Alexander ; Xu, Feifei
; Zhang, Hawking
Subject: RE: [PATCH] drm/amdgpu: Check if FB BAR is enabled for ROM read
[AMD
[AMD Public Use]
From: amd-gfx On Behalf Of Lazar, Lijo
Sent: Wednesday, March 3, 2021 10:15 AM
To: amd-gfx@lists.freedesktop.org
Cc: Deucher, Alexander ; Xu, Feifei
; Zhang, Hawking
Subject: [PATCH] drm/amdgpu: Check if FB BAR is enabled for ROM read
[AMD Public Use]
Some configurations d
Hi Jack,
yes that comes pretty close. I'm going over the patch right now.
Some things still look a bit complicated to me, but I need to wrap my
head around how and why we are doing it this way once more.
Christian.
Am 08.03.21 um 13:43 schrieb Zhang, Jack (Jian):
[AMD Public Use]
Hi, Chris
[AMD Public Use]
Hi, Christian,
I made some change on V3 patch that insert a dma_fence_wait for the first jobs
after resubmit jobs.
It seems simpler than the V2 patch. Is this what you first thinks of in your
mind?
Thanks,
Jack
-Original Message-
From: Koenig, Christian
Sent: Monday
[Why]
Previous tdr design treats the first job in job_timeout as the bad job.
But sometimes a later bad compute job can block a good gfx job and
cause an unexpected gfx job timeout because gfx and compute ring share
internal GC HW mutually.
[How]
This patch implements an advanced tdr mode.
1.It sy
Fix the following coccicheck warnings:
./drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c:1600:40-45: WARNING: conversion
to bool not needed here.
./drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c:1598:40-45: WARNING: conversion
to bool not needed here.
Reported-by: Abaci Robot
Signed-off-by: Jiapeng Chong
---
Add error return code in error hanlding code of amdgpu_acpi_init().
Reported-by: TOTE Robot
Signed-off-by: Jia-Ju Bai
---
drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
b/drivers/gpu/drm
Fix the following coccicheck warnings:
./drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c:561:34-39: WARNING:
conversion to bool not needed here.
Reported-by: Abaci Robot
Signed-off-by: Jiapeng Chong
---
drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c | 2 +-
1 file changed, 1 insertion(+),
From: Zhang Yunkai
'drm/drm_hdcp.h' included in 'amdgpu_dm.c' is duplicated.
It is also included in the 79th line.
Signed-off-by: Zhang Yunkai
---
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_d
From: Zhang Yunkai
'dce110_resource.h' included in 'dcn21_resource.c' is duplicated.
'hw_gpio.h' included in 'hw_factory_dce110.c' is duplicated.
Signed-off-by: Zhang Yunkai
---
drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c | 1 -
.../gpu/drm/amd/display/dc/gpio/dce110/hw_facto
88 matches
Mail list logo
