Fulfill the incomplete SMU13 `pp_dpm_dcefclk` implementation.
Reported-by: Guan Yu
Signed-off-by: Evan Quan
---
.../drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c | 27 +++
.../drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c | 27 +++
2 files changed, 54 insertions(+)
The val is defined as unsigned int type, if(val<0) is invalid, hence
modified its type to ssize_t
Fixes the below:
drivers/gpu/drm/amd/pm/amdgpu_pm.c:2800:5-8: WARNING: Unsigned expression
compared with zero: val < 0
drivers/gpu/drm/amd/pm/amdgpu_pm.c:2813:5-8: WARNING: Unsigned expression
[AMD Official Use Only - General]
Ping..
> -Original Message-
> From: Quan, Evan
> Sent: Tuesday, August 15, 2023 4:42 PM
> To: amd-gfx@lists.freedesktop.org
> Cc: Deucher, Alexander ; Quan, Evan
>
> Subject: [PATCH 0/8] A new set of Linux OD interfaces
>
> The existing OD interface
AMDGPU SOCs supports dynamic workload based power profiles, which can
provide fine-tuned performance for a particular type of workload.
This patch series adds an interface to set/reset these power profiles
based on the submitted job. The driver can dynamically switch
the power profiles based on
The'struct amdgpu_smu_workload' initialization/cleanup
functions is added by this patch.
v2:
- Splitting big patch into separate patches.
- Added new fini function.
Cc: Shashank Sharma
Cc: Christian Koenig
Cc: Alex Deucher
Signed-off-by: Arvind Yadav
---
drivers/gpu/drm/amd/amdgpu/Makefile
This patch adds a function which will clear the GPU
power profile after job finished.
This is how it works:
- schedular will set the GPU power profile based on ring_type.
- Schedular will clear the GPU Power profile once job finished.
- Here, the *_workload_profile_set function will set the GPU
This patch adds a function which will change the GPU
power profile based on a submitted job. This can optimize
the power performance when the workload is on.
v2:
- Splitting workload_profile_set and workload_profile_put
into two separate patches.
- Addressed review comment.
Cc: Shashank Sharma
From: Bas Nieuwenhuizen
To prep for allowing different sync modes in a follow-up patch.
Signed-off-by: Bas Nieuwenhuizen
Signed-off-by: Tatsuyuki Ishi
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 +-
From: Bas Nieuwenhuizen
This should be okay because moves themselves use KERNEL usage and
hence still sync with BOOKKEEP usage. Then any later submits still
wait on any pending VM operations.
(i.e. we only made VM ops not wait on BOOKKEEP submits, not the other
way around)
Signed-off-by: Bas
From: Bas Nieuwenhuizen
This changes all BO usages in a submit to BOOKKEEP instead of READ,
which effectively disables implicit sync for these submits.
This is configured at a context level using the existing IOCTL.
Signed-off-by: Bas Nieuwenhuizen
Co-developed-by: Tatsuyuki Ishi
From: Bas Nieuwenhuizen
For detection of the new explicit sync functionality without
having to try the ioctl.
Signed-off-by: Bas Nieuwenhuizen
Signed-off-by: Tatsuyuki Ishi
---
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git
This adds a context option to use DMA_RESV_USAGE_BOOKKEEP for userspace
submissions. This is a respin of [1] but rebased on top of the newly
introduced drm_exec helpers.
Disabling implicit sync is something we've wanted in RADV for a while
for resolving some corner cases. A more immediate thing
The voltage offset setting will be applied to the whole v/f curve line
instead of per anchor point base.
Signed-off-by: Evan Quan
---
drivers/gpu/drm/amd/pm/amdgpu_pm.c| 45 +++
.../drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c | 31 ++---
[AMD Official Use Only - General]
-Original Message-
From: amd-gfx On Behalf Of Srinivasan
Shanmugam
Sent: Monday, August 21, 2023 11:36 AM
To: Koenig, Christian ; Deucher, Alexander
; Chen, Guchun
Cc: Pan, Xinhui ; SHANMUGAM, SRINIVASAN
; amd-gfx@lists.freedesktop.org
Subject:
This patch adds a suspend function that will clear the GPU
power profile before going into suspend state.
v2:
- Add the new suspend function based on review comment.
Cc: Shashank Sharma
Cc: Christian Koenig
Cc: Alex Deucher
Signed-off-by: Arvind Yadav
---
This patch is to switch the GPU workload profile based
on the submitted job. The workload profile is reset to
default when the job is done.
Cc: Shashank Sharma
Cc: Christian Koenig
Cc: Alex Deucher
Signed-off-by: Arvind Yadav
---
drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 5 +
1 file
This patch switches the GPU workload mode to/from
compute mode, while submitting compute workload.
Cc: Christian Koenig
Signed-off-by: Alex Deucher
Signed-off-by: Shashank Sharma
Signed-off-by: Arvind Yadav
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 8 +---
1 file changed, 5
This reverts commit 5ce71f59bb9bd3d8a09b96afdbc92975cb6dc303.
Reason for revert: New amdgpu_workload_profile* api is added
to switch on/off profile mode. These new api will allow to
change the GPU power profile based on a submitted job.
Cc: Shashank Sharma
Cc: Christian Koenig
Cc: Alex
On 8/21/2023 11:59 AM, Kamal, Asad wrote:
[AMD Official Use Only - General]
-Original Message-
From: amd-gfx On Behalf Of Srinivasan
Shanmugam
Sent: Monday, August 21, 2023 11:36 AM
To: Koenig, Christian ; Deucher, Alexander
; Chen, Guchun
Cc: Pan, Xinhui ; SHANMUGAM, SRINIVASAN
Am 20.08.23 um 11:51 schrieb Christophe JAILLET:
This serie simplifies amdgpu_bo_list_create() and usage of the 'struct
amdgpu_bo_list'.
Oh, yes please. That's something I always wanted to cleanup as well.
It is compile tested only.
That bothers me a bit. Arun, Vitaly, Shashank can anybody
Am 21.08.23 um 12:14 schrieb Arunpravin Paneer Selvam:
The way now contiguous requests are implemented such that
the size rounded up to power of 2 and the corresponding order
block picked from the freelist.
In addition to the older method, the new method will rounddown
the size to power of 2
Am 04.08.23 um 07:46 schrieb Srinivasan Shanmugam:
Instead of declaring pointers use READ_ONCE(), when accessing those
values to make sure that the compiler doesn't voilate any cache
coherences
That commit message is a bit confusing and not 100% technically correct.
The compiler is not
[AMD Official Use Only - General]
With the following patch in place
https://patchwork.freedesktop.org/patch/553433/?series=122640=1
Reviewed-by: Asad Kamal
Thanks & Regards
Asad
-Original Message-
From: SHANMUGAM, SRINIVASAN
Sent: Monday, August 21, 2023 12:06 PM
To: Kamal, Asad ;
Am 21.08.23 um 11:14 schrieb Geert Uytterhoeven:
Hi Christian,
On Fri, Jul 7, 2023 at 9:36 PM Geert Uytterhoeven wrote:
On Fri, Jul 7, 2023 at 2:06 PM Christian König
wrote:
Am 06.07.23 um 10:36 schrieb Geert Uytterhoeven:
On 32-bit:
../tests/amdgpu/amdgpu_stress.c: In function
As we have implemented a new method for contiguous allocation
which requires actual size and actual min_block_size, hence we
have moved the roundup and alignment size computation to buddy
allocator. This way gpu drivers pass the required size and
alignment to buddy allocator and rest of the
As we have implemented a new method for contiguous allocation
which requires actual size and actual min_block_size, hence we
have moved the roundup and alignment size computation to buddy
allocator. This way gpu drivers pass the required size and
alignment to buddy allocator and rest of the
The way now contiguous requests are implemented such that
the size rounded up to power of 2 and the corresponding order
block picked from the freelist.
In addition to the older method, the new method will rounddown
the size to power of 2 and the corresponding order block picked
from the freelist.
On Fri, Aug 18, 2023 at 11:26:11AM +0800, Evan Quan wrote:
> drivers/base/Makefile | 1 +
> drivers/base/wbrf.c | 280 ++
Why is a wifi-specific thing going into drivers/base/?
confused,
greg k-h
'struct amdgpu_bo_list' is really used as if it was ended by a flex array.
So make it more explicit and add a 'struct amdgpu_bo_list_entry entries[]'
field at the end of the structure.
This way, struct_size() can be used when it is allocated.
It is less verbose.
Signed-off-by: Christophe JAILLET
Hi,
This soft lockup occurs on amdgpu vanilla torvalds kernel with KCSAN turned on.
The platform is Ubuntu 22.04 LTS.
Please find config and lshw here:
https://domac.alu.unizg.hr/~mtodorov/linux/bugreports/amdgpu/6.5-rc7/
Here is the log excerpt:
[ 69.050693]
Le 21/08/2023 à 08:19, Su Hui a écrit :
smatch error:
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c:1257
amdgpu_discovery_reg_base_init() error:
testing array offset 'adev->vcn.num_vcn_inst' after use.
change the assignment order to avoid buffer overflow.
Fixes: c40bdfb2ffa4 ("drm/amdgpu: fix
Hi,
This soft lockup occurs on amdgpu vanilla torvalds kernel with KCSAN turned on.
The platform is Ubuntu 22.04 LTS.
[ 69.050693]
==
[ 69.198264] watchdog: BUG: soft lockup - CPU#1 stuck for 26s!
[systemd-udevd:635]
[
This serie simplifies amdgpu_bo_list_create() and usage of the 'struct
amdgpu_bo_list'.
It is compile tested only.
Christophe JAILLET (4):
drm/amdgpu: Explicitly add a flexible array at the end of 'struct
amdgpu_bo_list'
drm/amdgpu: Remove a redundant sanity check
drm/amdgpu: Remove
On Fri, Aug 18, 2023 at 05:49:14PM -0500, Limonciello, Mario wrote:
>
>
> On 8/18/2023 4:24 PM, Greg KH wrote:
> > On Fri, Aug 18, 2023 at 11:26:11AM +0800, Evan Quan wrote:
> > > drivers/base/Makefile | 1 +
> > > drivers/base/wbrf.c | 280
The case where 'num_entries' is too big, is already handled by
struct_size(), because kvmalloc() would fail.
It will return -ENOMEM instead of -EINVAL, but it is only related to a
unlikely to happen sanity check.
Signed-off-by: Christophe JAILLET
---
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
kvzalloc() can be used instead of kvmalloc() + memset() + explicit NULL
assignments.
It is less verbose and more future proof.
Signed-off-by: Christophe JAILLET
---
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git
smatch error:
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c:1257
amdgpu_discovery_reg_base_init() error:
testing array offset 'adev->vcn.num_vcn_inst' after use.
change the assignment order to avoid buffer overflow.
Fixes: c40bdfb2ffa4 ("drm/amdgpu: fix incorrect VCN revision in SRIOV")
Now that there is an explicit flexible array at the end of 'struct
amdgpu_bo_list', it can be used to remove amdgpu_bo_list_array_entry() and
simplify some macro.
Signed-off-by: Christophe JAILLET
---
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 2 +-
On 2023/8/21 14:47, Christophe JAILLET wrote:
Le 21/08/2023 à 08:19, Su Hui a écrit :
smatch error:
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c:1257
amdgpu_discovery_reg_base_init() error:
testing array offset 'adev->vcn.num_vcn_inst' after use.
change the assignment order to avoid buffer
smatch error:
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c:1257
amdgpu_discovery_reg_base_init() error:
testing array offset 'adev->vcn.num_vcn_inst' after use.
change the assignment order to avoid buffer overflow.
Fixes: c40bdfb2ffa4 ("drm/amdgpu: fix incorrect VCN revision in SRIOV")
Am 21.08.23 um 09:37 schrieb Su Hui:
smatch error:
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c:1257
amdgpu_discovery_reg_base_init() error:
testing array offset 'adev->vcn.num_vcn_inst' after use.
change the assignment order to avoid buffer overflow.
Fixes: c40bdfb2ffa4 ("drm/amdgpu: fix
Am 21.08.23 um 08:20 schrieb Tatsuyuki Ishi:
From: Bas Nieuwenhuizen
This should be okay because moves themselves use KERNEL usage and
hence still sync with BOOKKEEP usage. Then any later submits still
wait on any pending VM operations.
(i.e. we only made VM ops not wait on BOOKKEEP submits,
Am 21.08.23 um 11:48 schrieb Geert Uytterhoeven:
Hi Christian,
On Mon, Aug 21, 2023 at 11:34 AM Christian König
wrote:
Am 21.08.23 um 11:14 schrieb Geert Uytterhoeven:
On Fri, Jul 7, 2023 at 9:36 PM Geert Uytterhoeven wrote:
On Fri, Jul 7, 2023 at 2:06 PM Christian König
wrote:
Am
It is required for TCP/UTCL1 thrashing
v2: updated rev_id check
Signed-off-by: Mangesh Gadre
Reviewed-by: Hawking Zhang
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
On 8/21/2023 6:36 PM, Shashank Sharma wrote:
Hey Arvind,
On 21/08/2023 08:47, Arvind Yadav wrote:
The'struct amdgpu_smu_workload' initialization/cleanup
functions is added by this patch.
v2:
- Splitting big patch into separate patches.
- Added new fini function.
Cc: Shashank Sharma
Cc:
Someone from MM should also confirm on this, but:
Acked-by: Shashank Sharma
On 21/08/2023 08:47, Arvind Yadav wrote:
This reverts commit 5ce71f59bb9bd3d8a09b96afdbc92975cb6dc303.
Reason for revert: New amdgpu_workload_profile* api is added
to switch on/off profile mode. These new api will
On 17/08/2023 20:20, André Almeida wrote:
Giving that we use codedump just for device resets, move it's functions
and structs to a more semantic file, the amdgpu_reset.{c, h}.
Signed-off-by: André Almeida
---
v5: no change
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h| 9 ---
On 8/21/2023 7:24 PM, Shashank Sharma wrote:
On 21/08/2023 15:35, Yadav, Arvind wrote:
On 8/21/2023 6:36 PM, Shashank Sharma wrote:
Hey Arvind,
On 21/08/2023 08:47, Arvind Yadav wrote:
The'struct amdgpu_smu_workload' initialization/cleanup
functions is added by this patch.
v2:
-
On 21/08/2023 16:12, Yadav, Arvind wrote:
On 8/21/2023 7:24 PM, Shashank Sharma wrote:
On 21/08/2023 15:35, Yadav, Arvind wrote:
On 8/21/2023 6:36 PM, Shashank Sharma wrote:
Hey Arvind,
On 21/08/2023 08:47, Arvind Yadav wrote:
The'struct amdgpu_smu_workload' initialization/cleanup
On 21/08/2023 08:47, Arvind Yadav wrote:
This patch adds a function which will change the GPU
power profile based on a submitted job. This can optimize
the power performance when the workload is on.
v2:
- Splitting workload_profile_set and workload_profile_put
into two separate patches.
-
[Public]
Hi all,
This week this patchset was tested on the following systems:
* Lenovo ThinkBook T13s Gen4 with AMD Ryzen 5 6600U
* MSI Gaming X Trio RX 6800
* Gigabyte Gaming OC RX 7900 XTX
These systems were tested on the following display/connection types:
*
On 21/08/2023 08:47, Arvind Yadav wrote:
This patch adds a suspend function that will clear the GPU
power profile before going into suspend state.
v2:
- Add the new suspend function based on review comment.
Cc: Shashank Sharma
Cc: Christian Koenig
Cc: Alex Deucher
Signed-off-by: Arvind
On 21/08/2023 08:47, Arvind Yadav wrote:
This patch switches the GPU workload mode to/from
compute mode, while submitting compute workload.
Cc: Christian Koenig
Signed-off-by: Alex Deucher
Signed-off-by: Shashank Sharma
Signed-off-by: Arvind Yadav
---
On 21/08/2023 08:47, Arvind Yadav wrote:
This patch is to switch the GPU workload profile based
on the submitted job. The workload profile is reset to
default when the job is done.
Cc: Shashank Sharma
Cc: Christian Koenig
Cc: Alex Deucher
Signed-off-by: Arvind Yadav
---
On 8/21/2023 7:13 PM, Shashank Sharma wrote:
On 21/08/2023 08:47, Arvind Yadav wrote:
This patch adds a suspend function that will clear the GPU
power profile before going into suspend state.
v2:
- Add the new suspend function based on review comment.
Cc: Shashank Sharma
Cc: Christian
On 8/21/2023 7:09 PM, Shashank Sharma wrote:
On 21/08/2023 08:47, Arvind Yadav wrote:
This patch adds a function which will clear the GPU
power profile after job finished.
This is how it works:
- schedular will set the GPU power profile based on ring_type.
- Schedular will clear the GPU
On Fri, 2023-08-18 at 11:26 +0800, Evan Quan wrote:
> To support the WBRF mechanism, Wifi adapters utilized in the system must
> register the frequencies in use(or unregister those frequencies no longer
> used) via the dedicated calls. So that, other drivers responding to the
> frequencies can
+ Amar should be able to help.
Amar,
Can you please check this patch (series if required) with a few IGTs and
probably with Xonotic as well ?
Regards
Shashank
On 21/08/2023 13:03, Christian König wrote:
Am 20.08.23 um 11:51 schrieb Christophe JAILLET:
This serie simplifies
On Fri, 18 Aug 2023, Deucher, Alexander wrote:
> [Public]
>
> > -Original Message-
> > From: Ilpo Järvinen
> > Sent: Monday, July 17, 2023 8:05 AM
> > To: linux-...@vger.kernel.org; Bjorn Helgaas ; Lorenzo
> > Pieralisi ; Rob Herring ;
> > Krzysztof Wilczyński ; Emmanuel Grumbach
> > ;
Hi Christian,
On Mon, Aug 21, 2023 at 11:34 AM Christian König
wrote:
> Am 21.08.23 um 11:14 schrieb Geert Uytterhoeven:
> > On Fri, Jul 7, 2023 at 9:36 PM Geert Uytterhoeven
> > wrote:
> >> On Fri, Jul 7, 2023 at 2:06 PM Christian König
> >> wrote:
> >>> Am 06.07.23 um 10:36 schrieb Geert
Hi Christian,
On Fri, Jul 7, 2023 at 9:36 PM Geert Uytterhoeven wrote:
> On Fri, Jul 7, 2023 at 2:06 PM Christian König
> wrote:
> > Am 06.07.23 um 10:36 schrieb Geert Uytterhoeven:
> > > On 32-bit:
> > >
> > > ../tests/amdgpu/amdgpu_stress.c: In function ‘alloc_bo’:
> > >
Hey Arvind,
On 21/08/2023 08:47, Arvind Yadav wrote:
The'struct amdgpu_smu_workload' initialization/cleanup
functions is added by this patch.
v2:
- Splitting big patch into separate patches.
- Added new fini function.
Cc: Shashank Sharma
Cc: Christian Koenig
Cc: Alex Deucher
Signed-off-by:
On 21/08/2023 08:47, Arvind Yadav wrote:
This patch adds a function which will clear the GPU
power profile after job finished.
This is how it works:
- schedular will set the GPU power profile based on ring_type.
- Schedular will clear the GPU Power profile once job finished.
- Here, the
On 21/08/2023 15:35, Yadav, Arvind wrote:
On 8/21/2023 6:36 PM, Shashank Sharma wrote:
Hey Arvind,
On 21/08/2023 08:47, Arvind Yadav wrote:
The'struct amdgpu_smu_workload' initialization/cleanup
functions is added by this patch.
v2:
- Splitting big patch into separate patches.
- Added new
On 17/08/2023 20:20, André Almeida wrote:
Instead of storing coredump information inside amdgpu_device struct,
move if to a proper separated struct and allocate it dynamically. This
will make it easier to further expand the logged information.
Signed-off-by: André Almeida
---
v5: no change
On 17/08/2023 20:20, André Almeida wrote:
Even if there's nothing currently parsing amdgpu's coredump files, if
we eventually have such tools they will be glad to find a version field
to properly read the file.
Create a version number to be displayed on top of coredump file, to be
incremented
On 8/21/2023 7:16 PM, Shashank Sharma wrote:
On 21/08/2023 08:47, Arvind Yadav wrote:
This patch is to switch the GPU workload profile based
on the submitted job. The workload profile is reset to
default when the job is done.
Cc: Shashank Sharma
Cc: Christian Koenig
Cc: Alex Deucher
On 17/08/2023 20:20, André Almeida wrote:
To better organize struct amdgpu_device, keep all reset information
related fields together in a separated struct.
Signed-off-by: André Almeida
---
v5: new patch, as requested by Shashank Sharma
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 34
Rework the KFD max system memory and ttm limit to allow bigger
system memory allocations upto 63/64 of the available memory which is
controlled by ttm module params pages_limit and page_pool_size. Also for
NPS1 mode, report the max ttm limit as the available VRAM size. For max
system memory limit,
If mGPUs is on same IOMMU group, or is ram direct mapped, then mGPUs
can share the original BO for GTT mapping dma address, without creating
new BO from export/import dmabuf.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 5 +++--
1 file changed, 3
On Mon, Aug 21, 2023 at 2:55 AM Arvind Yadav wrote:
>
> This patch adds a function which will change the GPU
> power profile based on a submitted job. This can optimize
> the power performance when the workload is on.
>
> v2:
> - Splitting workload_profile_set and workload_profile_put
> into
Hi Christian,
I ran GLMark2, unigine heaven and Tomb Raider on steam. I didn't observe
any issues.
Regards,
Arun.
On 8/21/2023 4:33 PM, Christian König wrote:
Am 20.08.23 um 11:51 schrieb Christophe JAILLET:
This serie simplifies amdgpu_bo_list_create() and usage of the 'struct
On Fri, Aug 18, 2023 at 04:12:57PM +, Deucher, Alexander wrote:
> > -Original Message-
> > From: Ilpo Järvinen
> > Sent: Monday, July 17, 2023 8:05 AM
> > To: linux-...@vger.kernel.org; Bjorn Helgaas ; Lorenzo
> > Pieralisi ; Rob Herring ;
> > Krzysztof Wilczyński ; Emmanuel Grumbach
Hi Michel,
Em 17/08/2023 07:37, Michel Dänzer escreveu:
On 8/15/23 20:57, André Almeida wrote:
From: Pekka Paalanen
Specify how the atomic state is maintained between userspace and
kernel, plus the special case for async flips.
Signed-off-by: Pekka Paalanen
Signed-off-by: André Almeida
On Mon, Aug 21, 2023 at 2:55 AM Arvind Yadav wrote:
>
> This patch adds a function which will change the GPU
> power profile based on a submitted job. This can optimize
> the power performance when the workload is on.
>
> v2:
> - Splitting workload_profile_set and workload_profile_put
> into
On 8/21/2023 9:52 PM, Alex Deucher wrote:
On Mon, Aug 21, 2023 at 2:55 AM Arvind Yadav wrote:
This patch adds a function which will change the GPU
power profile based on a submitted job. This can optimize
the power performance when the workload is on.
v2:
- Splitting workload_profile_set
On 8/21/2023 11:36 PM, Alex Deucher wrote:
On Mon, Aug 21, 2023 at 2:55 AM Arvind Yadav wrote:
This patch adds a function which will change the GPU
power profile based on a submitted job. This can optimize
the power performance when the workload is on.
v2:
- Splitting workload_profile_set
Hi,
On 21/08/2023 11:14, Arunpravin Paneer Selvam wrote:
The way now contiguous requests are implemented such that
the size rounded up to power of 2 and the corresponding order
block picked from the freelist.
In addition to the older method, the new method will rounddown
the size to power of 2
On Mon, Aug 21, 2023 at 1:54 PM Yadav, Arvind wrote:
>
>
> On 8/21/2023 9:52 PM, Alex Deucher wrote:
> > On Mon, Aug 21, 2023 at 2:55 AM Arvind Yadav wrote:
> >> This patch adds a function which will change the GPU
> >> power profile based on a submitted job. This can optimize
> >> the power
On 2023-08-21 15:29, Philip Yang wrote:
If mGPUs is on same IOMMU group, or is ram direct mapped, then mGPUs
can share the original BO for GTT mapping dma address, without creating
new BO from export/import dmabuf.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
On Tue, Aug 15, 2023 at 4:50 AM Evan Quan wrote:
>
> The existing OD interface `pp_od_clk_voltage` is unable to meet the growing
> demands for more OD functionalities. Since the buf used within it comes with
> size limit as one page. With more OD functionalities added, we will hit that
> limit
Would it make sense to include a link to a better explanation of the
underlying issue? E.g. https://lwn.net/Articles/624126/?
Regards,
Felix
On 2023-08-21 07:23, Christian König wrote:
Am 04.08.23 um 07:46 schrieb Srinivasan Shanmugam:
Instead of declaring pointers use READ_ONCE(), when
Acked-by: Alex Deucher
On Mon, Aug 21, 2023 at 3:10 AM Evan Quan wrote:
>
> Fulfill the incomplete SMU13 `pp_dpm_dcefclk` implementation.
>
> Reported-by: Guan Yu
> Signed-off-by: Evan Quan
> ---
> .../drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c | 27 +++
>
On 2023-08-21 15:20, Rajneesh Bhardwaj wrote:
Rework the KFD max system memory and ttm limit to allow bigger
system memory allocations upto 63/64 of the available memory which is
controlled by ttm module params pages_limit and page_pool_size. Also for
NPS1 mode, report the max ttm limit as the
Acked-by: Alex Deucher
On Mon, Aug 21, 2023 at 2:50 AM Evan Quan wrote:
>
> The voltage offset setting will be applied to the whole v/f curve line
> instead of per anchor point base.
>
> Signed-off-by: Evan Quan
> ---
> drivers/gpu/drm/amd/pm/amdgpu_pm.c| 45 +++
>
On 8/19/2023 5:50 AM, Greg KH wrote:
On Fri, Aug 18, 2023 at 05:49:14PM -0500, Limonciello, Mario wrote:
On 8/18/2023 4:24 PM, Greg KH wrote:
On Fri, Aug 18, 2023 at 11:26:11AM +0800, Evan Quan wrote:
drivers/base/Makefile | 1 +
drivers/base/wbrf.c
[Public]
-Original Message-
From: amd-gfx On Behalf Of Alex Deucher
Sent: Friday, August 18, 2023 9:55 PM
To: amd-gfx@lists.freedesktop.org
Cc: Deucher, Alexander ; Jiapeng Chong
Subject: [PATCH] drm/amd/pm: fix error flow in sensor fetching
Sensor fetching functions should return an
On 8/21/2023 12:17 PM, Arvind Yadav wrote:
This patch adds a function which will clear the GPU
power profile after job finished.
This is how it works:
- schedular will set the GPU power profile based on ring_type.
- Schedular will clear the GPU Power profile once job finished.
- Here, the
On 8/21/2023 6:17 PM, Mangesh Gadre wrote:
It is required for TCP/UTCL1 thrashing
Description looks a bit odd. You may rephrase as 'Update TCP/UTCL1
thrashing control settings'.
Reviewed-by: Lijo Lazar
Thanks,
Lijo
v2: updated rev_id check
Signed-off-by: Mangesh Gadre
Reviewed-by:
Hi Christian,
On 2023-08-21 07:03, Christian König wrote:
> Am 20.08.23 um 11:51 schrieb Christophe JAILLET:
>> This serie simplifies amdgpu_bo_list_create() and usage of the 'struct
>> amdgpu_bo_list'.
> Oh, yes please. That's something I always wanted to cleanup as well.
>
>> It is compile
90 matches
Mail list logo
