On Wed, Nov 27, 2024 at 2:57 PM Srinivasan Shanmugam
wrote:
>
> This update adds explanations to key functions that manage how the
> Kernel Fusion Driver (KFD) and Kernel Graphics Driver (KGD) share the
> GPU.
>
> amdgpu_gfx_enforce_isolation_wait_for_kfd: Controls the waiting period
> for KFD to
In the function pqm_uninit there is a call-assignment of "pdd =
kfd_get_process_device_data" which could be null, and this value was
later dereferenced without checking.
Fixes: fb91065851cd ("drm/amdkfd: Refactor queue wptr_bo GART mapping")
Signed-off-by: Andrew Martin
---
drivers/gpu/drm/amd/a
If SDMA is not available, warn if we try and copy
a GFX12 DCC buffer with the CPU.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
b/drivers/gpu/drm/amd/amdg
On Sun, 1 Dec 2024 01:55:17 +0200, Dmitry Baryshkov wrote:
> The connector->eld is accessed by the .get_eld() callback. This access
> can collide with the drm_edid_to_eld() updating the data at the same
> time. Add drm_connector.eld_mutex to protect the data from concurrenct
> access.
>
>
> [ ...
Init function is for ras table header read and check function is
responsible for the validation of the header. Call them in different
stages.
Signed-off-by: Tao Zhou
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 15 ++
.../gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c| 20
After the introduction of NPS RAS, one bad page record on eeprom may be
related to 1 or 16 bad pages, so the bad page record and bad page are
two different concepts, define a new variable to store bad page number.
Signed-off-by: Tao Zhou
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 10 +--
On Sun, 01 Dec 2024, Dmitry Baryshkov wrote:
> The connector->eld is accessed by the .get_eld() callback. This access
> can collide with the drm_edid_to_eld() updating the data at the same
> time. Add drm_connector.eld_mutex to protect the data from concurrenct
> access. Individual drivers are not
Add ACA support for jpeg v4.0.3.
Signed-off-by: Yang Wang
Reviewed-by: Hawking Zhang
---
drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 86
1 file changed, 86 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
in
v1:
Add ACA support for vcn v4.0.3.
v2:
- split VCN ACA(v1) to 2 parts: vcn and jpeg.
- move mmSMNAID_AID0_MCA_SMU to amdgpu_aca.h file.
v3:
- split JPEG ACA to another patch.
Signed-off-by: Yang Wang
Reviewed-by: Hawking Zhang
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 85
move common ACA ipid defines into amdgpu_aca.h file.
Signed-off-by: Yang Wang
Reviewed-by: Hawking Zhang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h | 5 +
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 4
drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c | 1 -
3 files changed, 5 insertions(+), 5 de
On Fri, Nov 29, 2024 at 10:40:14AM -0300, André Almeida wrote:
> Hi Raag,
>
> Em 28/11/2024 12:37, Raag Jadav escreveu:
> > Introduce device wedged event, which notifies userspace of 'wedged'
> > (hanged/unusable) state of the DRM device through a uevent. This is
> > useful especially in cases whe
The function atomctrl_get_smc_sclk_range_table() does not check the return
value of smu_atom_get_data_table(). If smu_atom_get_data_table() fails to
retrieve SMU_Info table, it returns NULL which is later dereferenced.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: a23ee
Hi Raag,
Em 28/11/2024 12:37, Raag Jadav escreveu:
Introduce device wedged event, which notifies userspace of 'wedged'
(hanged/unusable) state of the DRM device through a uevent. This is
useful especially in cases where the device is no longer operating as
expected and has become unrecoverable f
On Mon, Dec 02, 2024 at 10:19:41AM +, Maxime Ripard wrote:
> On Sun, 1 Dec 2024 01:55:17 +0200, Dmitry Baryshkov wrote:
> > The connector->eld is accessed by the .get_eld() callback. This access
> > can collide with the drm_edid_to_eld() updating the data at the same
> > time. Add drm_connector
[Public]
From: amd-gfx on behalf of Christian
König
Sent: Monday, December 2, 2024 2:19 PM
To: Deucher, Alexander; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/amdgpu: rework resume handling for display (v2)
>
>Am 02.12.24 um 20:18 schrieb Chri
From: Hawking Zhang
Apply gc v9_5_0 golden settings.
Signed-off-by: Hawking Zhang
Reviewed-by: Asad Kamal
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 16 ++--
1 file changed, 10 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/g
From: Lancelot SIX
Instruction modifiers of the untyped vector memory buffer instructions
(MUBUF encoded) changed in gfx940. The slc, scc and glc modifiers have
been replaced with sc0, sc1 and nt.
The current CWSR trap handler is written using pre-gfx940 modifier
names, making the source incomp
From: Le Ma
add gfx950 basic support
Signed-off-by: Le Ma
Reviewed-by: Hawking Zhang
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c| 1 +
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 5
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 3 +-
drivers/g
From: Le Ma
Update cwsr area size for gfx950 to fit the new user queue buffer validation.
The size of LDS calculation is referred from gfx950 thunk implementation.
Signed-off-by: Le Ma
Acked-by: Hawking Zhang
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdkfd/kfd_queue.c | 10 ++-
From: Lancelot SIX
The gfx-9 trap handler is reading LDS allocation size in 256 bytes
granularity (from SQ_WAVE_LDS_ALLOC), but it using the assumption that
this value is always even (i.e. the LDS allocation is really done in
multiple of 512 bytes). This was true so far, but gfx-950 allocates LD
From: Alex Sierra
For better readability. Also leftover orphaned code.
Signed-off-by: Alex Sierra
Reviewed-by: Felix Kuehling
Reviewed-by: Harish Kasiviswanathan
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 8 +++-
drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 5 +++
From: Alex Sierra
GC 9.5.0 local memory MTYPE default should be set as RW.
Signed-off-by: Alex Sierra
Reviewed-by: Felix Kuehling
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
b/d
From: Alex Sierra
Initial support for GC 9.5.0.
v2: squash in pqm_clean_queue_resource() fix from Lijo
Signed-off-by: Alex Sierra
Reviewed-by: Felix Kuehling
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdkfd/kfd_crat.c | 1 +
drivers/gpu/drm/amd/amdkfd/kfd_debug.h|
From: Le Ma
Add firmware declarations.
Signed-off-by: Le Ma
Reviewed-by: Hawking Zhang
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_
From: Alex Sierra
Update mtype flags to meet gfx 9.5.0 requirements for remote GPU
memory and system memory.
Signed-off-by: Alex Sierra
Reviewed-by: Felix Kuehling
Reviewed-by: Harish Kasiviswanathan
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 8
driver
From: Lancelot SIX
In gfx950, the SQ_WAVE_LDS_ALLOC.LDS_SIZE field is extended to bits 12
to 22. The LDS_SIZE granularity remains unchanged (units of 64 dwords,
or 256 bytes). This patch adjusts the CWSR trap handler to read the
full extent of LDS_SIZE.
Signed-off-by: Lancelot SIX
Reviewed-by
Le 01/12/2024 à 00:55, Dmitry Baryshkov a écrit :
Reading access to connector->eld can happen at the same time the
drm_edid_to_eld() updates the data. Take the newly added eld_mutex in
order to protect connector->eld from concurrent access.
Signed-off-by: Dmitry Baryshkov
Hi Dmitry,
Acked
On Mon, Dec 2, 2024 at 3:27 AM Ivan Stepchenko wrote:
>
> The function atomctrl_get_smc_sclk_range_table() does not check the return
> value of smu_atom_get_data_table(). If smu_atom_get_data_table() fails to
> retrieve SMU_Info table, it returns NULL which is later dereferenced.
>
> Found by Linu
Applied. Thanks!
On Wed, Nov 27, 2024 at 11:17 PM Randy Dunlap wrote:
>
> Make spelling and punctuation changes to ease reading of the comments.
>
> Signed-off-by: Randy Dunlap
> Cc: Alex Deucher
> Cc: Christian König
> Cc: Xinhui Pan
> Cc: amd-gfx@lists.freedesktop.org
> Cc: David Airlie
>
On Wed, Nov 27, 2024 at 1:22 PM David Yat Sin wrote:
>
> Cacheline size is not available in IP discovery for gc943,gc944.
>
> Signed-off-by: David Yat Sin
> Reviewed-by: Harish Kasiviswanathan
Acked-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdkfd/kfd_crat.c | 6 ++
> 1 file changed,
I have the exact same issue with you. I'm using a laptop with 7840hs.
I encountered this warning after upgrading the kerne to 6.11.10.
Maybe some back-port patches break something, I think there is no
warning when I got into kernel 6.11.0.
And the warning also exists in 6.13-rc1.
Maybe you can p
Hi Christian,
Thank you.
Reviewed-by: Arunpravin Paneer Selvam
Regards,
Arun.
On 11/29/2024 8:38 PM, Christian König wrote:
When starting the mpv player, Radeon R9 users are observing
the below error in dmesg.
[drm:amdgpu_uvd_cs_pass2 [amdgpu]]
*ERROR* msg/fb buffer ff00f7c000-ff00f7e000 out
On Mon, Dec 02, 2024 at 01:03:07PM +0200, Dmitry Baryshkov wrote:
> On Mon, Dec 02, 2024 at 10:19:41AM +, Maxime Ripard wrote:
> > On Sun, 1 Dec 2024 01:55:17 +0200, Dmitry Baryshkov wrote:
> > > The connector->eld is accessed by the .get_eld() callback. This access
> > > can collide with the d
This update adds explanations to key functions that manage how the
Kernel Fusion Driver (KFD) and Kernel Graphics Driver (KGD) share the
GPU.
amdgpu_gfx_enforce_isolation_wait_for_kfd: Controls the waiting period
for KFD to ensure it takes turns with KGD in using the GPU. It uses a
mutex to safely
Am 02.12.24 um 17:52 schrieb Alex Deucher:
Split resume into a 3rd step to handle displays when DCC is
enabled on DCN 4.0.1. Move display after the buffer funcs
have been re-enabled so that the GPU will do the move and
properly set the DCC metadata for DCN.
v2: fix fence irq resume ordering
Si
Am 02.12.24 um 20:18 schrieb Christian König:
Am 02.12.24 um 17:52 schrieb Alex Deucher:
Split resume into a 3rd step to handle displays when DCC is
enabled on DCN 4.0.1. Move display after the buffer funcs
have been re-enabled so that the GPU will do the move and
properly set the DCC metadata
[AMD Official Use Only - AMD Internal Distribution Only]
Thanks for the input, as offline syncing will add this remark in the function
amdgpu_ucode_request() description part.
Regards,
Prike
> -Original Message-
> From: Lazar, Lijo
> Sent: Monday, December 2, 2024 3:46 PM
> To:
[AMD Official Use Only - AMD Internal Distribution Only]
The series is:
Reviewed-by: Tao Zhou
> -Original Message-
> From: Wang, Yang(Kevin)
> Sent: Monday, December 2, 2024 4:49 PM
> To: amd-gfx@lists.freedesktop.org
> Cc: Zhang, Hawking ; Zhou1, Tao
> ; Yang, Stanley ; Zhang,
> Hawki
[AMD Official Use Only - AMD Internal Distribution Only]
Series is
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: amd-gfx On Behalf Of Tao Zhou
Sent: Monday, December 2, 2024 18:25
To: amd-gfx@lists.freedesktop.org
Cc: Zhou1, Tao
Subject: [PATCH 2/2] drm/amdgpu:
Split resume into a 3rd step to handle displays when DCC is
enabled on DCN 4.0.1. Move display after the buffer funcs
have been re-enabled so that the GPU will do the move and
properly set the DCC metadata for DCN.
v2: fix fence irq resume ordering
Signed-off-by: Alex Deucher
---
drivers/gpu/d
On 11/30/2024 3:55 PM, Dmitry Baryshkov wrote:
Reading access to connector->eld can happen at the same time the
drm_edid_to_eld() updates the data. Take the newly added eld_mutex in
order to protect connector->eld from concurrent access.
Signed-off-by: Dmitry Baryshkov
---
drivers/gpu/drm/
41 matches
Mail list logo
