[why]
A gfx job may be processed but not finished when reset begin from
compute job timeout. drm_sched_resubmit_jobs_ext in sched_main
assume submitted job unsignaled and always put parent fence.
Resubmission for that job cause underflow. This fix is done in
device reset to avoid changing drm sched
[AMD Official Use Only - General]
Reviewed-by: Tim Huang
-Original Message-
From: Deucher, Alexander
Sent: Friday, May 27, 2022 1:58 AM
To: amd-gfx@lists.freedesktop.org
Cc: Deucher, Alexander ; Huang, Tim
Subject: [PATCH] drm/amdgpu/soc21: add mode2 asic reset for SMU IP v13.0.4
Set
Am 04.06.22 um 00:44 schrieb Felix Kuehling:
[+amd-gfx]
On 2022-06-03 15:37, Felix Kuehling wrote:
On 2022-06-03 06:46, Christian König wrote:
Resources about to be destructed are not tied to BOs any more.
I've been seeing a backtrace in that area with a patch series I'm
working on, but did
Am 06.06.22 um 00:00 schrieb Michal Kubecek:
[SNIP]
This patch should help:
https://patchwork.freedesktop.org/patch/488258/
After ~48 hours with this patch, still no apparent issues.
Tested-by: Michal Kubecek
Thanks, this could be optimized for gfx8 a bit if anybody is interested
in a typi
invalid/prime icahce operation takes effect on both pipes cuconrrently,
therefore CP_MES_IC_BASE_LO/HI and CP_MES_MDBASE_LO/HI have to be both
set before prime icache. Otherwise MES hardware gets garbage data in
above regsters and causes page fault
[ 470.873200] amdgpu :33:00.0: amdgpu: [gfxh
Am 05.06.22 um 18:47 schrieb Daniel Vetter:
On Fri, 27 May 2022 at 01:55, Dmitry Osipenko
wrote:
Introduce a common DRM SHMEM shrinker framework that allows to reduce
code duplication among DRM drivers by replacing theirs custom shrinker
implementations with the generic shrinker.
In order to s
On 6/6/2022 8:41 AM, Evan Quan wrote:
Enable ASPM support for PCIE 7.4.0 and 7.6.0.
Signed-off-by: Evan Quan
Change-Id: Ib3b0e106ff43ad49f0f815e6eeb5c756b6bf4550
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 +
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 1 +
drivers/gpu/
Am 2022-06-03 um 06:46 schrieb Christian König:
Resources about to be destructed are not tied to BOs any more.
Signed-off-by: Christian König
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/ttm/ttm_device.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/driv
On Fri, Jun 03, 2022 at 11:49:31AM -0400, Alex Deucher wrote:
> On Thu, Jun 2, 2022 at 10:22 AM Michal Kubecek wrote:
> >
> > On Thu, Jun 02, 2022 at 09:58:22AM -0400, Alex Deucher wrote:
> > > On Fri, May 27, 2022 at 8:58 AM Michal Kubecek wrote:
> > > > On Fri, May 27, 2022 at 11:00:39AM +0200,
On Sun, Jun 5, 2022 at 10:14 PM Guchun Chen wrote:
>
> This reverts commit da38a66ac46e334f198afcd1b4d4554b4ddca0df.
>
> This causes regression in GPU reset related test.
>
> Cc: Alexander Deucher
> Cc: ricet...@gmail.com
> Signed-off-by: Guchun Chen
Reviewed-by: Alex Deucher
> ---
> drivers
[Public]
Hi all,
This week this patchset was tested on the following systems:
HP Envy 360, with Ryzen 5 4500U
Lenovo Thinkpad T14s Gen2, with AMD Ryzen 5 5650U
Sapphire Pulse RX5700XT
Reference AMD RX6800
Engineering board with Ryzen 9 5900H
These systems were tested on the following displ
The last case label can write two buffers 'mc_reg_address[j]' and
'mc_data[j]' with 'j' offset equal to SMC_NISLANDS_MC_REGISTER_ARRAY_SIZE
since there are no checks for this value in both case labels after the
last 'j++'.
Instead of changing '>' to '>=' there, add the bounds check at the start
of
On 2022-06-03 14:50, Rodrigo Siqueira wrote:
> Signed-off-by: Rodrigo Siqueira
Reviewed-by: Harry Wentland
Harry
> ---
> drivers/gpu/drm/amd/display/include/dal_asic_id.h | 6 --
> 1 file changed, 6 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/display/include/dal_asic_id.h
> b/driv
On 2022-06-03 14:50, Rodrigo Siqueira wrote:
> GCC throw warnings for the function dcn20_update_bounding_box due to its
> frame size that looks like this:
>
> error: the frame size of 1936 bytes is larger than 1024 bytes
> [-Werror=frame-larger-than=]
>
> This commit fixes this issue by elimina
On 2022-06-03 14:50, Rodrigo Siqueira wrote:
> GCC throw warnings for the function dcn301_fpu_update_bw_bounding_box
> due to its frame size that looks like this:
>
> error: the frame size of 1936 bytes is larger than 1024 bytes
> [-Werror=frame-larger-than=]
>
> For fixing this issue I dropped
On 2022-06-03 14:50, Rodrigo Siqueira wrote:
> GCC throw warnings for the function dcn31_update_bw_bounding_box and
> dcn316_update_bw_bounding_box due to its frame size that looks like
> this:
>
> error: the frame size of 1936 bytes is larger than 1024 bytes
> [-Werror=frame-larger-than=]
>
>
On 2022-06-03 14:50, Rodrigo Siqueira wrote:
> GCC throw warnings for the function dcn21_update_bw_bounding_box and
> dcn316_update_bw_bounding_box due to its frame size that looks like
> this:
>
> error: the frame size of 1936 bytes is larger than 1024 bytes
> [-Werror=frame-larger-than=]
>
>
On 2022-06-03 14:50, Rodrigo Siqueira wrote:
> This reverts commit 78baa3c4dfff4375b109bc5e19663a2f7fad1190.
>
> This commit introduced the below compilation error when using
> allmodconfig:
>
> error: implicit declaration of function ‘remove_hpo_dp_link_enc_from_ctx’;
> did you mean ‘add_hpo_dp
Am 2022-06-04 um 06:23 schrieb Errabolu, Ramesh:
+bool amdgpu_device_is_peer_accessible(struct amdgpu_device *adev,
+ struct amdgpu_device *peer_adev)
+{
+#ifdef CONFIG_HSA_AMD_P2P
+ bool p2p_access = false;
+ uint64_t address_mask = peer_adev->de
On 2022-06-03 14:49, Felix Kuehling
wrote:
Removed an unused parameter from two functions and added kernel-doc
comments.
Reviewed-by: Philip Yang
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h| 7 ++-
On Wed, May 25, 2022 at 9:02 AM Daniel Vetter wrote:
> On Mon, May 16, 2022 at 04:56:13PM -0600, Jim Cromie wrote:
> > DRM.debug API is 23 macros, issuing 10 exclusive categories of debug
> > messages. By rough count, they are used 5140 times in the kernel.
> > These all call drm_dbg or drm_devd
Patches 2-6 are:
Reviewed-by: Alex Deucher
On Sun, Jun 5, 2022 at 11:13 PM Evan Quan wrote:
>
> PMFW will handle that properly. Driver involvement may cause some
> unexpected issues.
>
> Signed-off-by: Evan Quan
> Change-Id: I77da7d894485a3ac6a1a956e4d2605d0bc730c25
> ---
> drivers/gpu/drm/amd
Add the xgmi node's physical node id to the kfd topology node properties.
The physical node id is a 0-indexed, monotonically increasing id that is
statically assigned based on the node's frame buffer position within the
hive.
This is useful because the id is also used by the DF for xGMI peer to p
Fixes "no previous prototype" warnings.
Reported-by: kernel test robot
Signed-off-by: Alex Deucher
---
.../gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c | 2 +-
drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c | 8
drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c
Add proper handling for PPC64.
Reported-by: kernel test robot
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/display/dc/dcn32/Makefile | 9 -
drivers/gpu/drm/amd/display/dc/dcn321/Makefile | 9 -
2 files changed, 16 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/d
On 2022-06-06 11:42, Alex Deucher wrote:
> Fixes "no previous prototype" warnings.
>
> Reported-by: kernel test robot
> Signed-off-by: Alex Deucher
Reviewed-by: Harry Wentland
Harry
> ---
> .../gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c | 2 +-
> drivers/gpu/drm/amd/display/dc/dc
On 2022-06-06 11:42, Alex Deucher wrote:
> Add proper handling for PPC64.
>
> Reported-by: kernel test robot
> Signed-off-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/display/dc/dcn32/Makefile | 9 -
> drivers/gpu/drm/amd/display/dc/dcn321/Makefile | 9 -
> 2 files changed, 16
On Mon, Jun 6, 2022 at 11:54 AM Harry Wentland wrote:
>
> On 2022-06-06 11:42, Alex Deucher wrote:
> > Add proper handling for PPC64.
> >
> > Reported-by: kernel test robot
> > Signed-off-by: Alex Deucher
> > ---
> > drivers/gpu/drm/amd/display/dc/dcn32/Makefile | 9 -
> > drivers/gpu/
On Mon, Jun 6, 2022 at 10:16 AM Harry Wentland wrote:
>
> On 2022-06-03 14:50, Rodrigo Siqueira wrote:
> > This reverts commit 78baa3c4dfff4375b109bc5e19663a2f7fad1190.
> >
> > This commit introduced the below compilation error when using
> > allmodconfig:
> >
> > error: implicit declaration of fu
On 2022-06-06 12:00, Alex Deucher wrote:
> On Mon, Jun 6, 2022 at 11:54 AM Harry Wentland wrote:
>>
>> On 2022-06-06 11:42, Alex Deucher wrote:
>>> Add proper handling for PPC64.
>>>
>>> Reported-by: kernel test robot
>>> Signed-off-by: Alex Deucher
>>> ---
>>> drivers/gpu/drm/amd/display/dc
Applied. Thanks!
Alex
On Mon, Jun 6, 2022 at 9:51 AM Alexey Kodanev
wrote:
>
> The last case label can write two buffers 'mc_reg_address[j]' and
> 'mc_data[j]' with 'j' offset equal to SMC_NISLANDS_MC_REGISTER_ARRAY_SIZE
> since there are no checks for this value in both case labels after the
>
On 2022-06-06 03:43, Yiqing Yao wrote:
[why]
A gfx job may be processed but not finished when reset begin from
compute job timeout. drm_sched_resubmit_jobs_ext in sched_main
assume submitted job unsignaled and always put parent fence.
Resubmission for that job cause underflow. This fix is done
On 2022-06-06 12:17, Alex Deucher wrote:
> On Mon, Jun 6, 2022 at 10:16 AM Harry Wentland wrote:
>>
>> On 2022-06-03 14:50, Rodrigo Siqueira wrote:
>>> This reverts commit 78baa3c4dfff4375b109bc5e19663a2f7fad1190.
>>>
>>> This commit introduced the below compilation error when using
>>> allmodc
On Mon, Jun 6, 2022 at 2:02 PM Harry Wentland wrote:
>
>
>
> On 2022-06-06 12:17, Alex Deucher wrote:
> > On Mon, Jun 6, 2022 at 10:16 AM Harry Wentland
> > wrote:
> >>
> >> On 2022-06-03 14:50, Rodrigo Siqueira wrote:
> >>> This reverts commit 78baa3c4dfff4375b109bc5e19663a2f7fad1190.
> >>>
> >
Add support for peer-to-peer communication among AMD GPUs over PCIe
bus. Support REQUIRES enablement of config HSA_AMD_P2P.
Signed-off-by: Ramesh Errabolu
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h| 1 +
.../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 300 ++
drivers/gp
Extend KFD device topology to surface peer-to-peer links among
GPU devices connected over PCIe or xGMI. Enabling HSA_AMD_P2P is
REQUIRED to surface peer-to-peer links.
Prior to this KFD did not expose to user mode any P2P links or
indirect links that go over two or more direct hops. Old versions
o
[AMD Official Use Only - General]
Just posted updated patch addressing the comment
-Original Message-
From: Kuehling, Felix
Sent: Monday, June 6, 2022 7:57 PM
To: Errabolu, Ramesh ; amd-gfx list
Subject: Re: [PATCH 2/3] drm/amdgpu: Add peer-to-peer support among PCIe
connected AMD GP
Ping?
Alex
On Tue, May 24, 2022 at 10:09 PM Alex Deucher wrote:
>
> This is the same as the default case, so drop the extra
> logic.
>
> Signed-off-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c | 20
> 1 file changed, 20 deletions(-)
>
> diff --git a/d
Am 2022-06-06 um 14:07 schrieb Ramesh Errabolu:
Add support for peer-to-peer communication among AMD GPUs over PCIe
bus. Support REQUIRES enablement of config HSA_AMD_P2P.
Signed-off-by: Ramesh Errabolu
Sorry, one more nit-pick inline. With that fixed, the patch is
Reviewed-by: Felix Kuehlin
Am 2022-06-06 um 16:04 schrieb Felix Kuehling:
Am 2022-06-06 um 14:07 schrieb Ramesh Errabolu:
Add support for peer-to-peer communication among AMD GPUs over PCIe
bus. Support REQUIRES enablement of config HSA_AMD_P2P.
Signed-off-by: Ramesh Errabolu
Sorry, one more nit-pick inline. With th
The MODE register contains detailed per-wave information, but UMR
skipped printing it. This patch adds the ability to print each wave's
MODE register as part of the wave scan operation, and prints the MODE
register's sub-fields as part of the deeper print option.
Signed-off-by: Joseph Greathouse
Am 2022-06-06 um 14:08 schrieb Ramesh Errabolu:
Extend KFD device topology to surface peer-to-peer links among
GPU devices connected over PCIe or xGMI. Enabling HSA_AMD_P2P is
REQUIRED to surface peer-to-peer links.
Prior to this KFD did not expose to user mode any P2P links or
indirect links
All other chips, from gfx6-gfx10, now include the MODE register at the
end of the wave debug state. This appears to have been missed in gfx11,
so this patch adds in MODE to the debug state for gfx11.
Signed-off-by: Joseph Greathouse
---
drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 1 +
1 file change
Reviewed-by: Guchun Chen
Regards,
Guchun
-Original Message-
From: amd-gfx On Behalf Of Alex Deucher
Sent: Tuesday, June 7, 2022 2:40 AM
To: Deucher, Alexander
Cc: amd-gfx list
Subject: Re: [PATCH] drm/amdgpu: simplify amdgpu_ucode_get_load_type()
Ping?
Alex
On Tue, May 24, 2022 at
[AMD Official Use Only - General]
It will happed error on loading firmware. we need add firmware version
information.
[ 309.650118] [drm] kiq ring mec 2 pipe 1 q 0
[ 309.652595] [drm] VCN decode and encode initialized successfully(under DPG
Mode).
[ 309.653402] kfd kfd: amdgpu: skipped dev
45 matches
Mail list logo