On 10/3/2023 16:22, Deucher, Alexander wrote:
[Public]
-Original Message-
From: Limonciello, Mario
Sent: Tuesday, October 3, 2023 5:17 PM
To: Deucher, Alexander ; amd-
g...@lists.freedesktop.org
Cc: Wentland, Harry
Subject: Re: [PATCH v3 1/4] drm/amd: Add support for prepare() and
com
WREG32_RLC does not specify the correct XCC so the RLCG interface does
not work.
Define WREG32_RLC_XCC to be like WREG32_RLC but include a parameter to
specify the XCC.
v2: Add new macro WREG32_RLC_XCC instead of modifying exiting WREG32_RLC
macro
Original v1 title: "drm/amdgpu: Add xcc inst
gfx_v9_4_3_xcc_set_compute_eop_interrupt_state should use
*REG32_RLC_XCC so CP_ME1_PIPE0_INT_CNTL registers can be accessed
under SRIOV.
v2: Use RREG32_RLC_XCC
Original v1 title: "drm/amdgpu: Use *REG32_RLC in gfx_v9_4_3.c"
Signed-off-by: Victor Lu
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c |
amdgpu_virt_kiq_reg_write_reg_wait is hardcoded to use MEC engine 0.
Add xcc_inst as a parameter to allow it to use different MEC engines.
v2: rebase
Signed-off-by: Victor Lu
---
drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 5 +++--
drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h | 3 ++-
drivers/gpu/
amdgpu_kiq_wreg/rreg is hardcoded to use MEC engine 0.
Add an xcc_id parameter to amdgpu_kiq_wreg/rreg, define W/RREG32_XCC
and amdgpu_device_xcc_wreg/rreg to to use the new xcc_id parameter.
v2: define amdgpu_device_xcc_wreg/rreg instead of changing parameters
of amdgpu_device_wreg/rreg
Sig
The WREG32/RREG32_SOC15_IP_NO_KIQ call is using XCC0's RLCG interface
when programming other XCCs.
Add xcc instance parameter to them.
v2: rebase
Signed-off-by: Victor Lu
---
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 16
drivers/gpu/drm/amd/amdgpu/soc15_common.h | 6 +++---
The "rlcg_reg_access_supported" flag is missing. Add it back in.
Signed-off-by: Victor Lu
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index fbfe0a1c4b19..e7543bc
From: Xiaogang Chen
This patch implements partial migration in gpu page fault according to migration
granularity(default 2MB) and not split svm range in cpu page fault handling.
A svm range may include pages from both system ram and vram of one gpu now.
These chagnes are expected to improve migra
Prepare for the coming implementation by GCC and Clang of the __counted_by
attribute. Flexible array members annotated with __counted_by can have
their accesses bounds-checked at run-time via CONFIG_UBSAN_BOUNDS (for
array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family
functions).
A
On 10/2/23 10:16, Michel Dänzer wrote:
From: Michel Dänzer
Cleanup, no functional change intended.
Signed-off-by: Michel Dänzer
Seems like this series does indeed fix the issue, so I have applied it
instead of the revert.
---
.../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 23 ++
On 2023-10-03 16:50, Philip Yang wrote:
If there is no VRAM domain, bo_node is NULL and this causes crash.
Refactor the change, and use the module parameter as higher privilege.
Need another patch to support override PTE flag on APU.
Fixes: 55d7e2001c7e ("drm/amdgpu: Add EXT_COHERENT memory a
[Public]
> -Original Message-
> From: Limonciello, Mario
> Sent: Tuesday, October 3, 2023 5:17 PM
> To: Deucher, Alexander ; amd-
> g...@lists.freedesktop.org
> Cc: Wentland, Harry
> Subject: Re: [PATCH v3 1/4] drm/amd: Add support for prepare() and
> complete() callbacks
>
> On 10/3/202
On 10/3/2023 16:11, Deucher, Alexander wrote:
[Public]
-Original Message-
From: amd-gfx On Behalf Of Mario
Limonciello
Sent: Tuesday, October 3, 2023 4:55 PM
To: amd-gfx@lists.freedesktop.org
Cc: Wentland, Harry ; Limonciello, Mario
Subject: [PATCH v3 1/4] drm/amd: Add support for pre
[Public]
> -Original Message-
> From: amd-gfx On Behalf Of Mario
> Limonciello
> Sent: Tuesday, October 3, 2023 4:55 PM
> To: amd-gfx@lists.freedesktop.org
> Cc: Wentland, Harry ; Limonciello, Mario
>
> Subject: [PATCH v3 1/4] drm/amd: Add support for prepare() and complete()
> callbacks
As dc_set_power_state() no longer allocates memory, it's not necessary
to have return types and check return code as it can't fail anymore.
Change it back to `void`.
Signed-off-by: Mario Limonciello
---
.../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 17 +
drivers/gpu/drm/amd/
If there is memory pressure at suspend time then dynamically
allocating a large structure as part of DC suspend code will
fail.
Instead re-use the same structure and clear all members except
those that should be maintained.
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2362
Signed-off-by:
If the system is under high memory pressure, the resources may need
to be evicted into swap instead. If the storage backing for swap
is offlined during the suspend() step then such a call may fail.
So instead move this step into prepare(), while leaving all other
steps that put the GPU into a low
Linux PM core has a prepare() callback run before suspend and complete()
callback ran after resume() for devices to use. Add plumbing to bring
prepare() to amdgpu.
The idea with the new vfuncs for amdgpu is that all IP blocks that
memory allocations during suspend should do the allocation from th
At suspend time if there is memory pressure then dynamically allocating
memory will cause failures that don't clean up properly when trying
suspend a second time.
Move the bigger memory allocations into Linux PM prepare() callback and
drop allocations that aren't really needed in DC code.
v1:
ht
On 10/2/23 13:47, ivlip...@amd.com wrote:
From: Ivan Lipski
This reverts commit a92da5bc33ea99a861f6c422192af6072c145d2d.
V2: Reword commit message
[WHY]
This commit caused regression in which eDP's with PSR support,
but no Replay support (Sink support <= 0x03), failed enabling PSR
and all IG
If there is no VRAM domain, bo_node is NULL and this causes crash.
Refactor the change, and use the module parameter as higher privilege.
Need another patch to support override PTE flag on APU.
Fixes: 55d7e2001c7e ("drm/amdgpu: Add EXT_COHERENT memory allocation flags")
Signed-off-by: Philip Yang
[AMD Official Use Only - General]
You can revert it for now, but that is a bug as replay_feature_enabled should
stop all replay related calls.
From: LIPSKI, IVAN
Sent: October 3, 2023 4:11 PM
To: Lakha, Bhawanpreet ; Li, Sun peng (Leo)
; amd-gfx@lists.freedeskto
[AMD Official Use Only - General]
Setting 'replay_feature_enabled' to false did not fix the issue, and PSR could
not be enabled on an eDP without Replay support (Sink support 0x03 in my case).
Thank you,
Ivan Lipski
From: Lakha, Bhawanpreet
Sent: October 3,
On 2023-10-02 06:58, Sebastian Andrzej Siewior wrote:
> On 2023-09-22 07:33:26 [+0200], Christian König wrote:
>> Am 21.09.23 um 16:15 schrieb Sebastian Andrzej Siewior:
>>> Hi,
>>>
>>> I stumbled uppon the amdgpu driver via a bugzilla report. The actual fix
>>> is #4 + #5 and the rest was made whi
On 2023-09-21 10:15, Sebastian Andrzej Siewior wrote:
> This is a revert of the commit mentioned below while it is not wrong, as
> in the kernel will explode, having migrate_disable() here it is
> complete waste of resources.
>
> Additionally commit message is plain wrong the review tag does not m
When ROCm is active enable additional SMU 13.0.0 optimizations.
This reuses the unused powersave profile on PMFW.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
b/d
[AMD Official Use Only - General]
> If there is no VRAM domain, bo_node is NULL and this causes crash. Move
> the EXT_COHERENT support change to VRAM domain path.
>
> Need another patch to support override PTE flag on APU.
>
> Fixes: 55d7e2001c7e ("drm/amdgpu: Add EXT_COHERENT memory allocation f
On 10/3/2023 2:07 PM, Felix Kuehling wrote:
On 2023-10-02 16:21, Rajneesh Bhardwaj wrote:
On GFXIP9.4.3 APU, allow the memory reporting as per the ttm pages
limit in NPS1 mode.
Signed-off-by: Rajneesh Bhardwaj
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 17 -
drivers
On 2023-10-02 16:21, Rajneesh Bhardwaj wrote:
On GFXIP9.4.3 APU, allow the memory reporting as per the ttm pages
limit in NPS1 mode.
Signed-off-by: Rajneesh Bhardwaj
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 17 -
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 9 +
On 2023-10-03 12:57, Philip Yang wrote:
If there is no VRAM domain, bo_node is NULL and this causes crash. Move
the EXT_COHERENT support change to VRAM domain path.
Need another patch to support override PTE flag on APU.
Fixes: 55d7e2001c7e ("drm/amdgpu: Add EXT_COHERENT memory allocation flags
[AMD Official Use Only - General]
Acked-by: Harish Kasiviswanathan
-Original Message-
From: amd-gfx On Behalf Of Philip Yang
Sent: Tuesday, October 3, 2023 12:57 PM
To: amd-gfx@lists.freedesktop.org
Cc: Yang, Philip ; Kuehling, Felix
Subject: [PATCH] drm/amdkfd: Fix EXT_COHERENT memory
Series is:
Reviewed-by: Alex Deucher
On Tue, Oct 3, 2023 at 1:02 PM Lazar, Lijo wrote:
>
> [AMD Official Use Only - General]
>
> Ping on this series.
>
> Since there is no alternate header to put device property other than
> amdgpu.h, this can be moved also when appropriate split of header file
[AMD Official Use Only - General]
yes, it will disable replay.
From: Li, Sun peng (Leo)
Sent: October 3, 2023 11:40 AM
To: Lakha, Bhawanpreet ; LIPSKI, IVAN
; amd-gfx@lists.freedesktop.org
Cc: Siqueira, Rodrigo ; Mahfooz, Hamza
; Wentland, Harry ; Deucher,
Al
If there is no VRAM domain, bo_node is NULL and this causes crash. Move
the EXT_COHERENT support change to VRAM domain path.
Need another patch to support override PTE flag on APU.
Fixes: 55d7e2001c7e ("drm/amdgpu: Add EXT_COHERENT memory allocation flags")
Signed-off-by: Philip Yang
---
driver
[AMD Official Use Only - General]
Alternatively, might just keep the PKG_TYPE definition in amdgpu_smuio.h
Regards,
Hawking
-Original Message-
From: Lazar, Lijo
Sent: Wednesday, October 4, 2023 00:51
To: Lazar, Lijo ; amd-gfx@lists.freedesktop.org
Cc: Deucher, Alexander ; Zhang, Hawking
[AMD Official Use Only - General]
Ping on this series.
Since there is no alternate header to put device property other than amdgpu.h,
this can be moved also when appropriate split of header files are decided.
Other than that any comments on this?
Thanks,
Lijo
-Original Message-
From:
Hi Alex,
My apology, I was under the impression that RB is a must. I understand
now that ACK is good, checked with Leo after your response. Thank you.
Regards,
Sathish
On 10/3/2023 10:01 PM, Alex Deucher wrote:
On Tue, Oct 3, 2023 at 12:22 PM Sundararaju, Sathishkumar
wrote:
Hi ,
Kind
On Tue, Oct 3, 2023 at 12:22 PM Sundararaju, Sathishkumar
wrote:
>
> Hi ,
>
> Kind request to help review the change. Thank you.
I acked this change back when you sent it out, but if it didn't come
through for some reason:
Acked-by: Alex Deucher
>
> Regards,
>
> Sathish
>
> On 9/21/2023 8:17 PM
On 10/3/2023 07:58, Deucher, Alexander wrote:
[AMD Official Use Only - General]
-Original Message-
From: amd-gfx On Behalf Of Mario
Limonciello
Sent: Monday, October 2, 2023 6:45 PM
To: amd-gfx@lists.freedesktop.org
Cc: Wentland, Harry ; Limonciello, Mario
Subject: [PATCH v2 2/3] drm/
tree/branch:
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
branch HEAD: c9f2baaa18b5ea8f006a2b3a616da9597c71d15e Add linux-next specific
files for 20231003
Error/Warning reports:
https://lore.kernel.org/oe-kbuild-all/202309122047.cri9yjrq-...@intel.com
https
Hi ,
Kind request to help review the change. Thank you.
Regards,
Sathish
On 9/21/2023 8:17 PM, Alex Deucher wrote:
On Thu, Sep 21, 2023 at 9:07 AM Sathishkumar S
wrote:
fix wrong ip count INFO on spatial partitions. update the query
to return the instance count corresponding to the partitio
On 09/27, Harry Wentland wrote:
>
>
> On 2023-09-25 15:49, Melissa Wen wrote:
> > Add 3D LUT property for plane color transformations using a 3D lookup
> > table. 3D LUT allows for highly accurate and complex color
> > transformations and is suitable to adjust the balance between color
> > channe
On 2023-10-03 11:23, Lakha, Bhawanpreet wrote:
[AMD Official Use Only - General]
Why not just set replay_feature_enabled = true; to false?
Would that be the right fix? If so, we can send out a patch
with that instead.
- Leo
---
[AMD Official Use Only - General]
Why not just set replay_feature_enabled = true; to false?
From: Li, Sun peng (Leo)
Sent: October 3, 2023 11:20 AM
To: Lakha, Bhawanpreet ; LIPSKI, IVAN
; amd-gfx@lists.freedesktop.org
Cc: Siqueira, Rodrigo ; Mahfooz, Hamza
; W
On 2023-10-03 11:13, Lakha, Bhawanpreet wrote:
[AMD Official Use Only - General]
Any reason for reverting this instead of looking into
"amdgpu_dm_setup_replay()" and "replay_feature_enabled" to see why reply
is being enabled?
It causes a regression in amd_psr. Unless there's a quick fix,
[AMD Official Use Only - General]
Any reason for reverting this instead of looking into
"amdgpu_dm_setup_replay()" and "replay_feature_enabled" to see why reply is
being enabled?
Bhawan
From: LIPSKI, IVAN
Sent: October 2, 2023 1:47 PM
To: amd-gfx@lists.freedesk
[AMD Official Use Only - General]
> -Original Message-
> From: amd-gfx On Behalf Of Mario
> Limonciello
> Sent: Monday, October 2, 2023 6:45 PM
> To: amd-gfx@lists.freedesktop.org
> Cc: Wentland, Harry ; Limonciello, Mario
>
> Subject: [PATCH v2 2/3] drm/amd: Move evict resources suspend
47 matches
Mail list logo
