Re: [PATCH 5/8] drm/amdgpu: Create context for usermode queue

Am 07.02.23 um 08:51 schrieb Shashank Sharma: On 07/02/2023 08:14, Christian König wrote: Am 03.02.23 um 22:54 schrieb Shashank Sharma: The FW expects us to allocate atleast one page as context space to process gang, process, shadow, GDS and FW_space related work. This patch creates some

Re: [PATCH 6/8] drm/amdgpu: Map userqueue into HW

On 07/02/2023 08:20, Christian König wrote: Am 03.02.23 um 22:54 schrieb Shashank Sharma: From: Shashank Sharma This patch adds new fptrs to prepare the usermode queue to be mapped or unmapped into the HW. After this mapping, the queue will be ready to accept the workload. V1: Addressed

Re: [PATCH 5/8] drm/amdgpu: Create context for usermode queue

On 07/02/2023 08:14, Christian König wrote: Am 03.02.23 um 22:54 schrieb Shashank Sharma: The FW expects us to allocate atleast one page as context space to process gang, process, shadow, GDS and FW_space related work. This patch creates some object for the same, and adds an IP specific

Re: [PATCH 3/8] drm/amdgpu: introduce userqueue MQD handlers

On 07/02/2023 08:11, Christian König wrote: Am 03.02.23 um 22:54 schrieb Shashank Sharma: From: Shashank Sharma A Memory queue descriptor (MQD) of a userqueue defines it in the harware's context. As the method of formation of a MQD, and its format can vary between different graphics IPs,

Re: [PATCH 2/8] drm/amdgpu: add usermode queues

On 07/02/2023 08:08, Christian König wrote: Am 03.02.23 um 22:54 schrieb Shashank Sharma: From: Shashank Sharma This patch adds skeleton code for usermode queue creation. It typically contains: - A new structure to keep all the user queue data in one place. - An IOCTL function to

Re: [PATCH 1/8] drm/amdgpu: UAPI for user queue management

On 07/02/2023 08:03, Christian König wrote: Am 06.02.23 um 22:03 schrieb Alex Deucher: On Mon, Feb 6, 2023 at 12:01 PM Christian König wrote: Am 06.02.23 um 17:56 schrieb Alex Deucher: On Fri, Feb 3, 2023 at 5:26 PM Shashank Sharma wrote: Hey Alex, On 03/02/2023 23:07, Alex Deucher

Re: [PATCH] drm/amdgpu: Use the TGID for trace_amdgpu_vm_update_ptes

That sounds like a good idea to me as well. If you think that a patch should be backported please add a "CC: sta...@vger.kernel.org" tag to it before sending it out. We can always remove it if we don't think a backport is appropriated, but maintainers seldom add it by themself. Thanks,

Re: [PATCH 6/8] drm/amdgpu: Map userqueue into HW

Am 03.02.23 um 22:54 schrieb Shashank Sharma: From: Shashank Sharma This patch adds new fptrs to prepare the usermode queue to be mapped or unmapped into the HW. After this mapping, the queue will be ready to accept the workload. V1: Addressed review comments from Alex on the RFC patch

Re: [PATCH 5/8] drm/amdgpu: Create context for usermode queue

Am 03.02.23 um 22:54 schrieb Shashank Sharma: The FW expects us to allocate atleast one page as context space to process gang, process, shadow, GDS and FW_space related work. This patch creates some object for the same, and adds an IP specific functions to do this. Cc: Alex Deucher Cc:

Re: [PATCH 3/8] drm/amdgpu: introduce userqueue MQD handlers

Am 03.02.23 um 22:54 schrieb Shashank Sharma: From: Shashank Sharma A Memory queue descriptor (MQD) of a userqueue defines it in the harware's context. As the method of formation of a MQD, and its format can vary between different graphics IPs, we need gfx GEN specific handlers to create MQDs.

Re: [PATCH 2/8] drm/amdgpu: add usermode queues

Am 03.02.23 um 22:54 schrieb Shashank Sharma: From: Shashank Sharma This patch adds skeleton code for usermode queue creation. It typically contains: - A new structure to keep all the user queue data in one place. - An IOCTL function to create/free a usermode queue. - A function to generate

Re: [PATCH 1/8] drm/amdgpu: UAPI for user queue management

Am 06.02.23 um 22:03 schrieb Alex Deucher: On Mon, Feb 6, 2023 at 12:01 PM Christian König wrote: Am 06.02.23 um 17:56 schrieb Alex Deucher: On Fri, Feb 3, 2023 at 5:26 PM Shashank Sharma wrote: Hey Alex, On 03/02/2023 23:07, Alex Deucher wrote: On Fri, Feb 3, 2023 at 4:54 PM Shashank

Re: Indexing of FeatureCtrlMask for SMU13 OverDrive

On Mon, Feb 6, 2023 at 8:17 PM Matt Coffin wrote: > > Hello again, > > I've been working on OverDrive support for smu13, as you probably > already know. In that endeavor, it also contains the following: > > 1. I've come up with a few patterns that I think will reduce the > amount of boilerplate

Re: gpu_metrics does not provide 'current_gfxclk', 'current_uclk', 'average_cpu_power' & 'temperature_core' on AMD Ryzen 7000 CPU

On Mon, Feb 6, 2023 at 5:48 PM sfrcorne wrote: > > Dear Alex, > > First of all, thank you for your response. Personally, I use a Ryzen 5 7600X > however people with a Ryzen 9 7900X are also reporting this issue. The > relevant bug report in Mangohud can be found here: >

[PATCH] drm/amd/display: Align num_crtc to max_streams

From: tiancyin [Why] Display pipe might be harvested on some SKUs, that cause the adev->mode_info.num_crtc mismatch with the usable crtc number, then below error dmesgs observed after GPU recover. *ERROR* amdgpu_dm_set_crtc_irq_state: crtc is NULL at id :3 *ERROR*

RE: [PATCH] drm/amd/pm/smu7: move variables to where they are used

[AMD Official Use Only - General] Reviewed-by: Evan Quan > -Original Message- > From: Deucher, Alexander > Sent: Tuesday, February 7, 2023 1:05 AM > To: amd-gfx@lists.freedesktop.org > Cc: Deucher, Alexander ; Quan, Evan > > Subject: [PATCH] drm/amd/pm/smu7: move variables to where

RE: [PATCH 1/2] drm/amd/pm: bump SMU 13.0.0 driver_if header version

Acked-by: Guchun Chen Regards, Guchun -Original Message- From: amd-gfx On Behalf Of Evan Quan Sent: Tuesday, February 7, 2023 10:49 AM To: amd-gfx@lists.freedesktop.org Cc: Deucher, Alexander ; Quan, Evan Subject: [PATCH 1/2] drm/amd/pm: bump SMU 13.0.0 driver_if header version This

Re: [PATCH 2/2] drm/amd/pm: bump SMU 13.0.7 driver_if header version

Series is: Acked-by: Alex Deucher On Mon, Feb 6, 2023 at 9:49 PM Evan Quan wrote: > > This can suppress the warning caused by version mismatch. > > Signed-off-by: Evan Quan > Change-Id: Id3331a329ea9b1dbc45d8a4e773af1cbe8e21a27 > --- >

Re: gpu_metrics does not provide 'current_gfxclk', 'current_uclk', 'average_cpu_power' & 'temperature_core' on AMD Ryzen 7000 CPU

Dear Alex, First of all, thank you for your response. Personally, I use a Ryzen 5 7600X however people with a Ryzen 9 7900X are also reporting this issue. The relevant bug report in Mangohud can be found here: "https://github.com/flightlessmango/MangoHud/issues/868;. I looked around a bit in

[PATCH 1/2] drm/amd/pm: bump SMU 13.0.0 driver_if header version

This can suppress the warning caused by version mismatch. Signed-off-by: Evan Quan Change-Id: I5e62de359015ac93b2dfd6a257584a5e6d38e1f8 --- .../drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h | 5 +++-- drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h | 2 +- 2 files

[PATCH 2/2] drm/amd/pm: bump SMU 13.0.7 driver_if header version

This can suppress the warning caused by version mismatch. Signed-off-by: Evan Quan Change-Id: Id3331a329ea9b1dbc45d8a4e773af1cbe8e21a27 --- .../gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h | 3 ++- drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h | 2 +- 2 files

RE: [PATCH] drm/amd/pm: add SMU 13.0.7 missing GetPptLimit message mapping

[AMD Official Use Only - General] Reviewed-by: Feifei Xu -Original Message- From: amd-gfx On Behalf Of Evan Quan Sent: Friday, February 3, 2023 5:39 PM To: amd-gfx@lists.freedesktop.org Cc: Deucher, Alexander ; Quan, Evan Subject: [PATCH] drm/amd/pm: add SMU 13.0.7 missing

Indexing of FeatureCtrlMask for SMU13 OverDrive

Hello again, I've been working on OverDrive support for smu13, as you probably already know. In that endeavor, it also contains the following: 1. I've come up with a few patterns that I think will reduce the amount of boilerplate and SMU-specific code required to do implement these interfaces in

Re: [PATCH] drm/amdgpu: Use the TGID for trace_amdgpu_vm_update_ptes

Hi, thanks for applying the patch! Do you think it'd also be possible to backport it to previous kernel versions or do you already plan to do that? Since it is a one-liner bugfix it shouldn't be too hard to backport. Thank you, Friedrich Vock On 06.02.23 21:26, Alex Deucher wrote: Applied.

Re: gpu_metrics does not provide 'current_gfxclk', 'current_uclk', 'average_cpu_power' & 'temperature_core' on AMD Ryzen 7000 CPU

On Mon, Feb 6, 2023 at 9:22 AM sfrcorne wrote: > > Hello, > > I hope this is the correct place to ask my question. I was not sure if I > should have opened a new issue on Gitlab or send an email here, since I don't > know know whether this is a bug or intended behaviour. > > The question is

Re: [PATCH 1/8] drm/amdgpu: UAPI for user queue management

On Mon, Feb 6, 2023 at 12:01 PM Christian König wrote: > > Am 06.02.23 um 17:56 schrieb Alex Deucher: > > On Fri, Feb 3, 2023 at 5:26 PM Shashank Sharma > > wrote: > >> Hey Alex, > >> > >> On 03/02/2023 23:07, Alex Deucher wrote: > >>> On Fri, Feb 3, 2023 at 4:54 PM Shashank Sharma > >>>

Re: [PATCH] drm/amdkfd: To fix sdma page fault issue for GC 11.x

On 2023-02-06 07:58, Ji, Ruili wrote: From: Ruili Ji For the MQD memory, KMD would always allocate 4K memory, and mes scheduler would write to the end of MQD for unmap flag. Signed-off-by: Ruili Ji --- .../drm/amd/amdkfd/kfd_device_queue_manager.c | 20 +++

Re: [PATCH] drm/amdgpu: Use the TGID for trace_amdgpu_vm_update_ptes

Applied. Thanks! Alex On Mon, Feb 6, 2023 at 3:35 AM Christian König wrote: > > > > Am 02.02.23 um 17:21 schrieb Friedrich Vock: > > The pid field corresponds to the result of gettid() in userspace. > > However, userspace cannot reliably attribute PTE events to processes > > with just the

Re: [PATCH] [SUBMITTED 20210927] [RESEND^2] drm/amdgpu: fix enum odm_combine_mode mismatch

On Mon, Feb 6, 2023 at 2:36 PM Arnd Bergmann wrote: > > From: Arnd Bergmann > > A conversion from 'bool' to 'enum odm_combine_mode' was incomplete, > and gcc warns about this with many instances of > > display/dc/dml/dcn20/display_mode_vba_20.c:3899:44: warning: implicit > conversion from 'enum

Re: [PATCH] drm/amd/amdgpu: add complete header search path

Applied. Thanks! On Fri, Feb 3, 2023 at 10:27 PM Randy Dunlap wrote: > > The path for the "mod_info_packet.h" header file is > incomplete, so add its location to the header search path > in the amdgpu Makefile. > > See on ARCH=alpha (275 times in one build). > > In file included from

RE: [PATCH 1/2] drm/amdgpu: Fix incorrect filenames in sysfs comments

[AMD Official Use Only - General] This series Reviewed-by: Harish Kasiviswanathan -Original Message- From: amd-gfx On Behalf Of kent.russ...@amd.com Sent: Monday, February 6, 2023 12:26 PM To: amd-gfx@lists.freedesktop.org Cc: Russell, Kent Subject: [PATCH 1/2] drm/amdgpu: Fix

[PATCH] drm: Rename headers to match DP2.1 spec

This patch changes the headers defined in drm_dp.h to match the DP 2.1 spec. Signed-off-by: Jasdeep Dhillon --- drivers/gpu/drm/tegra/dp.c | 2 +- include/drm/display/drm_dp.h | 13 +++-- 2 files changed, 8 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/tegra/dp.c

[PATCH] [SUBMITTED 20210927] [RESEND^2] drm/amdgpu: fix enum odm_combine_mode mismatch

From: Arnd Bergmann A conversion from 'bool' to 'enum odm_combine_mode' was incomplete, and gcc warns about this with many instances of display/dc/dml/dcn20/display_mode_vba_20.c:3899:44: warning: implicit conversion from 'enum ' to 'enum odm_combine_mode' [-Wenum-conversion] 3899 |

Re: [PATCH] drm/amdgpu: Fix potential race processing vm->freed

Am 06.02.23 um 19:21 schrieb Rob Clark: On Mon, Feb 6, 2023 at 8:05 AM Christian König wrote: Am 06.02.23 um 16:52 schrieb Rob Clark: On Mon, Feb 6, 2023 at 2:15 AM Christian König wrote: Am 03.02.23 um 19:10 schrieb Rob Clark: From: Rob Clark If userspace calls the AMDGPU_CS ioctl from

Re: [PATCH] drm/amdgpu: Fix potential race processing vm->freed

On Mon, Feb 6, 2023 at 8:05 AM Christian König wrote: > > Am 06.02.23 um 16:52 schrieb Rob Clark: > > On Mon, Feb 6, 2023 at 2:15 AM Christian König > > wrote: > >> Am 03.02.23 um 19:10 schrieb Rob Clark: > >>> From: Rob Clark > >>> > >>> If userspace calls the AMDGPU_CS ioctl from multiple

[linux-next:master] BUILD REGRESSION 129af770823407ee115a56c69a04b440fd2fbe61

tree/branch: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master branch HEAD: 129af770823407ee115a56c69a04b440fd2fbe61 Add linux-next specific files for 20230206 Error/Warning reports: https://lore.kernel.org/oe-kbuild-all/202301230743.xnut0zvc-...@intel.com https

Re: [PATCH] drm/amd/display: fix cursor offset on rotation 180

On 1/31/23 11:05, Melissa Wen wrote: Cursor gets clipped off in the middle of the screen with hw rotation 180. Fix a miscalculation of cursor offset when it's placed near the edges in the pipe split case. Cursor bugs with hw rotation were reported on AMD issue tracker:

[PATCH 2/2] drm/amdgpu: Add unique_id support for GC 11.0.1/2

These can support unique_id, so create the sysfs file for them Signed-off-by: kent.russ...@amd.com --- drivers/gpu/drm/amd/pm/amdgpu_pm.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c index

[PATCH 1/2] drm/amdgpu: Fix incorrect filenames in sysfs comments

This looks like a standard copy/paste mistake. Replace the incorrect serial_number references with product_name and product_model Signed-off-by: kent.russ...@amd.com --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git

Re: [PATCH] drm/amd/display: fix cursor offset on rotation 180

On 1/31/23 11:05, Melissa Wen wrote: > Cursor gets clipped off in the middle of the screen with hw rotation > 180. Fix a miscalculation of cursor offset when it's placed near the > edges in the pipe split case. > > Cursor bugs with hw rotation were reported on AMD issue tracker: >

Re: [PATCH 14/14] drm/amdgpu: introduce doorbell bo in kernel

On 06/02/2023 17:57, Christian König wrote: Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher This patch does the following: - Removes doorbell ptr from adev. - Moves doorbell.num_doorbells into mman.num_doorbells. - Introduces a kernel bo for doorbell management in form of

Re: [PATCH 3/3] drm/connector: Deprecate split for BT.2020 in drm_colorspace enum

On 2/6/23 04:47, Ville Syrjälä wrote: > On Sat, Feb 04, 2023 at 06:09:45AM +, Joshua Ashton wrote: >> >> >> On 2/3/23 19:34, Ville Syrjälä wrote: >>> On Fri, Feb 03, 2023 at 09:25:38PM +0200, Ville Syrjälä wrote: On Fri, Feb 03, 2023 at 08:56:55PM +0200, Ville Syrjälä wrote: > On

[PATCH] drm/amd/pm/smu7: move variables to where they are used

Move variable declarations to where they are used. Fixes a segfault on smu7 V0 structures where some tables don't exist. Cc: Evan Quan Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2388 Fixes: 711d3c39503b ("drm/amd/pm: fulfill powerplay peak profiling mode shader/memory clock

Re: [PATCH 12/13] drm/amdgpu: add domain info in bo_create_kernel_at

Am 06.02.23 um 18:01 schrieb Alex Deucher: On Mon, Feb 6, 2023 at 11:51 AM Christian König wrote: Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Shashank Sharma This patch adds a domain input variable for amdgpu_bo_create_kernel_at, so that it could be used for both VRAM and DOORBELL

Re: [PATCH 1/8] drm/amdgpu: UAPI for user queue management

Am 06.02.23 um 17:56 schrieb Alex Deucher: On Fri, Feb 3, 2023 at 5:26 PM Shashank Sharma wrote: Hey Alex, On 03/02/2023 23:07, Alex Deucher wrote: On Fri, Feb 3, 2023 at 4:54 PM Shashank Sharma wrote: From: Alex Deucher This patch intorduces new UAPI/IOCTL for usermode graphics queue.

Re: [PATCH 12/13] drm/amdgpu: add domain info in bo_create_kernel_at

On Mon, Feb 6, 2023 at 11:51 AM Christian König wrote: > > Am 03.02.23 um 20:08 schrieb Shashank Sharma: > > From: Shashank Sharma > > > > This patch adds a domain input variable for amdgpu_bo_create_kernel_at, > > so that it could be used for both VRAM and DOORBELL domains objects. It > > also

Re: [PATCH 14/14] drm/amdgpu: introduce doorbell bo in kernel

Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher This patch does the following: - Removes doorbell ptr from adev. - Moves doorbell.num_doorbells into mman.num_doorbells. - Introduces a kernel bo for doorbell management in form of mman.doorbell_kernel_bo. This bo holds the

Re: [PATCH 1/8] drm/amdgpu: UAPI for user queue management

On Fri, Feb 3, 2023 at 5:26 PM Shashank Sharma wrote: > > Hey Alex, > > On 03/02/2023 23:07, Alex Deucher wrote: > > On Fri, Feb 3, 2023 at 4:54 PM Shashank Sharma > > wrote: > >> From: Alex Deucher > >> > >> This patch intorduces new UAPI/IOCTL for usermode graphics > >> queue. The userspace

Re: [PATCH 12/14] drm/amdgpu: initialize doorbell memory pool

Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher This patch initializes doorbell pool with bar manager, which will divide all the doorbell memory into pages. Signed-off-by: Alex Deucher Signed-off-by: Shashank Sharma --- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 11

Re: [PATCH 12/13] drm/amdgpu: add domain info in bo_create_kernel_at

Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Shashank Sharma This patch adds a domain input variable for amdgpu_bo_create_kernel_at, so that it could be used for both VRAM and DOORBELL domains objects. It also adds supporting code for existing callers. We should probably drop this one

Re: [PATCH 10/13] drm/amdgpu: doorbell support in get_memory functions

Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher This patch adds section for doorbell memory in memory status reporting functions like vm/bo_get_memory. Marek reworked this just recently to pass around a structure. You should probably rebase the code when that patch lands.

Re: [PATCH 09/13] drm/amdgpu: accommodate DOMAIN/PL_DOORBELL

On Mon, Feb 6, 2023 at 6:30 AM Christian König wrote: > > Am 03.02.23 um 20:08 schrieb Shashank Sharma: > > From: Alex Deucher > > > > This patch adds changes to accommodate the new GEM/TTM domain > > for doorbell memory. > > > > Signed-off-by: Alex Deucher > > Signed-off-by: Shashank Sharma >

Re: [PATCH 09/14] drm/amdgpu: move doorbell aperture handling into ttm_init

Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher To consolidate it with vram handling. Signed-off-by: Alex Deucher Signed-off-by: Alex Deucher I can't check all the implementation details, but from a mile high view that looks correct. Acked-by: Christian König

Re: [PATCH 02/13] drm/amdgpu: rename vram_mgr functions to bar_mgr

On Mon, Feb 6, 2023 at 6:21 AM Christian König wrote: > > Am 03.02.23 um 20:08 schrieb Shashank Sharma: > > From: Alex Deucher > > > > Rename the VRAM manager functions so they can be resused to manage > > doorbell BAR as well. > > Yeah, as said before Alex did this before we had the buddy

Re: [PATCH 0/8] AMDGPU usermode queues

On Mon, Feb 6, 2023 at 10:39 AM Michel Dänzer wrote: > > On 2/3/23 22:54, Shashank Sharma wrote: > > From: Shashank Sharma > > > > This patch series introduces AMDGPU usermode graphics queues. > > User queues is a method of GPU workload submission into the graphics > > hardware without any

Re: [PATCH] drm/amdgpu: Fix potential race processing vm->freed

Am 06.02.23 um 16:52 schrieb Rob Clark: On Mon, Feb 6, 2023 at 2:15 AM Christian König wrote: Am 03.02.23 um 19:10 schrieb Rob Clark: From: Rob Clark If userspace calls the AMDGPU_CS ioctl from multiple threads, because the vm is global to the drm_file, you can end up with multiple threads

Re: [PATCH 02/13] drm/amdgpu: rename vram_mgr functions to bar_mgr

Am 06.02.23 um 16:34 schrieb Shashank Sharma: On 06/02/2023 12:20, Christian König wrote: Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher Rename the VRAM manager functions so they can be resused to manage doorbell BAR as well. Yeah, as said before Alex did this before we

Re: [PATCH] drm/amdgpu: Fix potential race processing vm->freed

On Mon, Feb 6, 2023 at 2:15 AM Christian König wrote: > > Am 03.02.23 um 19:10 schrieb Rob Clark: > > From: Rob Clark > > > > If userspace calls the AMDGPU_CS ioctl from multiple threads, because > > the vm is global to the drm_file, you can end up with multiple threads > > racing in

Re: [PATCH 0/8] AMDGPU usermode queues

On 2/3/23 22:54, Shashank Sharma wrote: > From: Shashank Sharma > > This patch series introduces AMDGPU usermode graphics queues. > User queues is a method of GPU workload submission into the graphics > hardware without any interaction with kernel/DRM schedulers. In this > method, a userspace

Re: [PATCH 02/13] drm/amdgpu: rename vram_mgr functions to bar_mgr

On 06/02/2023 12:20, Christian König wrote: Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher Rename the VRAM manager functions so they can be resused to manage doorbell BAR as well. Yeah, as said before Alex did this before we had the buddy allocator. This doesn't makes

Re: [PATCH 01/13] drm/amdgpu: add UAPI for allocating doorbell memory

Hey Christian, On 06/02/2023 12:19, Christian König wrote: Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher Signed-off-by: Alex Deucher ---   include/uapi/drm/amdgpu_drm.h | 7 ++-   1 file changed, 6 insertions(+), 1 deletion(-) diff --git

[PATCH] drm/amd/amdgpu: add complete header search path

The path for the "mod_info_packet.h" header file is incomplete, so add its location to the header search path in the amdgpu Makefile. See on ARCH=alpha (275 times in one build). In file included from ../drivers/gpu/drm/amd/amdgpu/amdgpu.h:90, from

[bug report] drm/amd/display: Disable MALL SS and messages for PSR supported configs

Hello Dillon Varone, This is a semi-automatic email about new static checker warnings. The patch 0bed85e48af1: "drm/amd/display: Disable MALL SS and messages for PSR supported configs" from Jan 4, 2023, leads to the following Smatch complaint:

gpu_metrics does not provide 'current_gfxclk', 'current_uclk', 'average_cpu_power' & 'temperature_core' on AMD Ryzen 7000 CPU

Hello, I hope this is the correct place to ask my question. I was not sure if I should have opened a new issue on Gitlab or send an email here, since I don't know know whether this is a bug or intended behaviour. The question is about the new AMD Ryzen 7000 CPU's. These new CPU's have an iGPU

Re: Patched macOS kexts start Raven iGPU, but GPUVM page fault occurs on the first GFX and SDMA IB submitted by WindowServer. Help?

While simply waiting for a reply to the email was an attractive option, we chose to investigate other parts of the code during the last three days. More precisely, we investigated _mmhub_1_0_update_medium_grain_clock_gating and seemed to have discovered a register offset mismatch. However, we

RE: [PATCH] drm/amdkfd: To fix sdma page fault issue for GC 11.x

[AMD Official Use Only - General] Reviewed-by: Aaron Liu > -Original Message- > From: Ji, Ruili > Sent: Monday, February 6, 2023 8:58 PM > To: amd-gfx@lists.freedesktop.org > Cc: Deucher, Alexander ; Kuehling, Felix > ; Liu, Aaron ; Zhang, Yifan > ; Ji, Ruili > Subject: [PATCH]

[PATCH] drm/amdkfd: To fix sdma page fault issue for GC 11.x

From: Ruili Ji For the MQD memory, KMD would always allocate 4K memory, and mes scheduler would write to the end of MQD for unmap flag. Signed-off-by: Ruili Ji --- .../drm/amd/amdkfd/kfd_device_queue_manager.c | 20 +++ drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c | 12

Re: [PATCH 09/13] drm/amdgpu: accommodate DOMAIN/PL_DOORBELL

Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher This patch adds changes to accommodate the new GEM/TTM domain for doorbell memory. Signed-off-by: Alex Deucher Signed-off-by: Shashank Sharma --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 +

Re: [PATCH 08/13] drm/amdgpu: move doorbell ptr into mman structure

Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher This patch: - moves the doorbell.ptr variable to mman structure - renames it to doorbell_aper_base_kaddr for better readability; This change is to make doorbell's ttm management similar to vram's. Signed-off-by: Alex Deucher

Re: [PATCH 07/13] drm/amdgpu: store doorbell info in gmc structure

Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher This patch moves doorbell info into adev->gmc structure, to align with vram info. This will prepare structures for proper ttm management of the doorbell BAR. Mhm, this is most likely not a good idea either. The doorbell isn't

Re: [PATCH 06/13] drm/amdgpu: rename gmc.aper_base/size

Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher This patch renames aper_base and aper_size parameters (in adev->gmc), to vram_aper_base and vram_aper_size, to differentiate it from the doorbell BAR. Signed-off-by: Alex Deucher Signed-off-by: Shashank Sharma Acked-by:

Re: [PATCH 04/13] drm/amdgpu: replace aper_base_kaddr with vram_aper_base_kaddr

Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher To differentiate it from the doorbell BAR. Signed-off-by: Alex Deucher Acked-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 10 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c| 14

Re: [PATCH 02/13] drm/amdgpu: rename vram_mgr functions to bar_mgr

Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher Rename the VRAM manager functions so they can be resused to manage doorbell BAR as well. Yeah, as said before Alex did this before we had the buddy allocator. This doesn't makes sense any more and should probably be dropped

Re: [PATCH 01/13] drm/amdgpu: add UAPI for allocating doorbell memory

Am 03.02.23 um 20:08 schrieb Shashank Sharma: From: Alex Deucher Signed-off-by: Alex Deucher --- include/uapi/drm/amdgpu_drm.h | 7 ++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index

Re: [PATCH] drm/amdgpu: Fix potential race processing vm->freed

Am 03.02.23 um 19:10 schrieb Rob Clark: From: Rob Clark If userspace calls the AMDGPU_CS ioctl from multiple threads, because the vm is global to the drm_file, you can end up with multiple threads racing in amdgpu_vm_clear_freed(). So the freed list should be protected with the status_lock,

Re: [PATCH 3/3] drm/connector: Deprecate split for BT.2020 in drm_colorspace enum

On Sat, Feb 04, 2023 at 06:09:45AM +, Joshua Ashton wrote: > > > On 2/3/23 19:34, Ville Syrjälä wrote: > > On Fri, Feb 03, 2023 at 09:25:38PM +0200, Ville Syrjälä wrote: > >> On Fri, Feb 03, 2023 at 08:56:55PM +0200, Ville Syrjälä wrote: > >>> On Fri, Feb 03, 2023 at 01:28:20PM -0500, Harry

Re: [PATCH] drm/amdgpu: expose more memory stats in fdinfo

Just two nit picks: +    seq_printf(m, "drm-evicted-visible-vram:\t%llu KiB\n", +           stats.evicted_visible_vram/1024UL); For the values not standardized for all DRM drivers we might want to use amd as prefix here instead of drm. +    uint64_t requested_gtt;        /* how much

Re: [PATCH] drm/amdgpu: only WARN freeing buffers when DMA is unavailable

Am 06.02.23 um 09:28 schrieb Xiao, Jack: [AMD Official Use Only - General]    >> >> It's simply not allowed to free up resources during suspend since those can't be acquired again during resume.       >> The in_suspend flag is set at the beginning of

Re: [PATCH 0/8] AMDGPU usermode queues

Am 06.02.23 um 01:52 schrieb Dave Airlie: On Sat, 4 Feb 2023 at 07:54, Shashank Sharma wrote: From: Shashank Sharma This patch series introduces AMDGPU usermode graphics queues. User queues is a method of GPU workload submission into the graphics hardware without any interaction with

Re: [PATCH] drm/amdgpu: Use the TGID for trace_amdgpu_vm_update_ptes

Am 02.02.23 um 17:21 schrieb Friedrich Vock: The pid field corresponds to the result of gettid() in userspace. However, userspace cannot reliably attribute PTE events to processes with just the thread id. This patch allows userspace to easily attribute PTE update events to specific processes

RE: [PATCH] drm/amdgpu: only WARN freeing buffers when DMA is unavailable

[AMD Official Use Only - General] >> >> It's simply not allowed to free up resources during suspend since those can't be acquired again during resume. >> The in_suspend flag is set at the beginning of suspend and unset at the end of

RE: [PATCH] drm/amd/display: Disable migration to ensure consistency of per-CPU variable

[AMD Official Use Only - General] Loop Charlie. Thanks, Rico Sent from Mail for Windows From: Tianci Yin Sent: Monday, February 6, 2023 3:59 PM To: amd-gfx@lists.freedesktop.org

Re: [PATCH] drm/amdgpu: only WARN freeing buffers when DMA is unavailable

Am 06.02.23 um 08:23 schrieb Xiao, Jack: [AMD Official Use Only - General] >> Nope, that is not related to any hw state. can use other flag. >> It's simply not allowed to free up resources during suspend since those can't be acquired again during resume. The in_suspend flag is set at the