The KFD VMID assignment was hard-coded in a few places. Consolidate that in
a single variable adev->vm_manager.first_kfd_vmid. The value is still
assigned in gmc-ip-version-specific code.
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 12 +---
When there is no graphics support, KFD can use more of the VMIDs. Graphics
VMIDs are only used for video decoding/encoding and post processing. With
two VCE engines, there is no reason to reserve more than 2 VMIDs for that.
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
Am 2020-06-24 um 8:06 p.m. schrieb Bhardwaj, Rajneesh:
>
> On 6/24/2020 3:05 PM, Alex Deucher wrote:
>> [CAUTION: External Email]
>>
>> The failures with ROCm only happen with noretry=1, so
>> enable runtime pm when noretry=0 (the current default).
>>
>> Signed-off-by: Alex Deucher
>> ---
>>
On 6/24/20 2:11 AM, Greg KH wrote:
On Tue, Jun 23, 2020 at 11:04:30PM -0400, Andrey Grodzovsky wrote:
On 6/23/20 2:05 AM, Greg KH wrote:
On Tue, Jun 23, 2020 at 12:51:00AM -0400, Andrey Grodzovsky wrote:
On 6/22/20 12:45 PM, Greg KH wrote:
On Mon, Jun 22, 2020 at 12:07:25PM -0400, Andrey
Series is
Reviewed-by: Hawking Zhang
Regards,
Hawking
Sent from my iPhone
> On Jun 25, 2020, at 06:52, Sheng, Wenhui wrote:
>
> From: Wenhui Sheng
>
> After we move request full access before set
> ip blocks, we can merge atombios init block
> of SRIOV and baremetal together.
>
>
On 6/24/2020 3:05 PM, Alex Deucher wrote:
[CAUTION: External Email]
Add a switch statement to simplify asic checks. Note
that BACO is not supported on APUs, so there is no
need to check them.
why not base this on smu_context to really query the
SMU_FEATURE_BACO_BIT and then base the below
On 6/24/2020 3:05 PM, Alex Deucher wrote:
[CAUTION: External Email]
The failures with ROCm only happen with noretry=1, so
enable runtime pm when noretry=0 (the current default).
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 6 +-
1 file changed, 5
From: Wenhui Sheng
Move request init data to virt detection func, so we
can insert request full access between request init data
and set ip blocks.
Signed-off-by: Wenhui Sheng
---
drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 28
drivers/gpu/drm/amd/amdgpu/nv.c
From: Wenhui Sheng
>From SIENNA_CICHLID, HW introduce a new protection
feature which can control the FB, doorbell and MMIO
write access for VF, so guest driver should request
full access before ip discovery, or we couldn't access
ip discovery data in FB.
Signed-off-by: Wenhui Sheng
---
From: Wenhui Sheng
Init soc15 reg base early enough so we can touch
mailbox related registers in request full access
for sriov before set_ip_blocks, vi arch doesn't
use reg base in virt ops.
v2: fix reg_base_init missed in bare metal case.
Signed-off-by: Wenhui Sheng
---
From: Wenhui Sheng
After we move request full access before set
ip blocks, we can merge atombios init block
of SRIOV and baremetal together.
Signed-off-by: Wenhui Sheng
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 23 +-
1 file changed, 1 insertion(+), 22 deletions(-)
From: Bhanuprakash Modem
v2:
* Rebase (Manasi)
As both VRR min and max are already part of drm_display_info,
drm can expose this VRR range for each connector.
Hence this logic should move to core DRM.
This reverts commit 727962f030c23422a01e8b22d0f463815fb15ec4.
Signed-off-by: Bhanuprakash
Hi Dave, Daniel,
Fixes for 5.8.
The following changes since commit 48778464bb7d346b47157d21ffde2af6b2d39110:
Linux 5.8-rc2 (2020-06-21 15:45:29 -0700)
are available in the Git repository at:
git://people.freedesktop.org/~agd5f/linux tags/amd-drm-fixes-5.8-2020-06-24
for you to fetch
[AMD Public Use]
Series is:
Reviewed-by: Alex Deucher
From: Jivin, Alex
Sent: Wednesday, June 24, 2020 4:31 PM
To: amd-gfx@lists.freedesktop.org
Cc: Deucher, Alexander
Subject: [PATCH 3/3] drm/amdgpu: SI support for UVD and VCE power managment
Port
Port functionality from the Radeon driver to support
UVD and VCE power management.
Signed-off-by: Alex Jivin
---
drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 67 +++---
drivers/gpu/drm/amd/amdgpu/si_dpm.c| 19
2 files changed, 68 insertions(+), 18 deletions(-)
diff
Port functionality from the Radeon driver to support
UVD clock control.
Signed-off-by: Alex Jivin
---
drivers/gpu/drm/amd/amdgpu/si.c | 234 ++-
drivers/gpu/drm/amd/amdgpu/sid.h | 1 +
2 files changed, 229 insertions(+), 6 deletions(-)
diff --git
Port functionality from the Radeon driver to support
VCE clock control.
Signed-off-by: Alex Jivin
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 9 ++
drivers/gpu/drm/amd/amdgpu/si.c | 126 +++-
drivers/gpu/drm/amd/amdgpu/sid.h| 31 +++
3 files changed, 165
Add a switch statement to simplify asic checks. Note
that BACO is not supported on APUs, so there is no
need to check them.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 34 -
1 file changed, 22 insertions(+), 12 deletions(-)
diff --git
The failures with ROCm only happen with noretry=1, so
enable runtime pm when noretry=0 (the current default).
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
Am 2020-06-24 um 12:06 p.m. schrieb Mukul Joshi:
> Lockdep is spewing circular locking dependency warning when
> reading SDMA usage stats.
>
> 150.887733] ==
> [ 150.893903] WARNING: possible circular locking dependency detected
> [
Lockdep is spewing circular locking dependency warning when
reading SDMA usage stats.
150.887733] ==
[ 150.893903] WARNING: possible circular locking dependency detected
[ 150.905917] --
[
Applied. Thanks!
Alex
On Wed, Jun 24, 2020 at 8:07 AM Colin King wrote:
>
> From: Colin Ian King
>
> There is an off-by-one bounds check on the index into arrays
> table->mc_reg_address and table->mc_reg_table_entry[k].mc_data[j] that
> can lead to reads and writes outside of arrays. Fix the
On Wed, Jun 24, 2020 at 10:54 AM Colin King wrote:
>
> From: Colin Ian King
>
> In the cases where adev->jpeg.num_jpeg_inst is zero or the condition
> adev->jpeg.harvest_config & (1 << i) is always non-zero the variable
> ret is never set to an error condition and the function returns
> an
Applied. Thanks!
Alex
On Wed, Jun 24, 2020 at 10:32 AM Nirmoy wrote:
>
> Acked-by: Nirmoy Das
>
>
> Thanks,
>
> Nirmoy
>
> On 6/24/20 4:14 PM, Colin King wrote:
> > From: Colin Ian King
> >
> > Currently the goto statement is skipping over a lot of setup code
> > because it is outside of an
From: Colin Ian King
In the cases where adev->jpeg.num_jpeg_inst is zero or the condition
adev->jpeg.harvest_config & (1 << i) is always non-zero the variable
ret is never set to an error condition and the function returns
an uninitialized value in ret. Since the only exit condition at
the end
On Wed, Jun 24, 2020 at 7:50 AM Alex Jivin wrote:
>
> Port functionality from the Radeon driver to support
> UVD and VCE power management.
>
> Signed-off-by: Alex Jivin
Please split this into three patches, one to add the set_uvd clocks
interface, one to set the set_vce clocks interface, and
[AMD Public Use]
Got it, will add more info in commit msg.
Brs
Wenhui
-Original Message-
From: Zhang, Hawking
Sent: Wednesday, June 24, 2020 10:40 PM
To: Zhang, Hawking ; Sheng, Wenhui
; amd-gfx@lists.freedesktop.org
Cc: Zhang, Andy ; Liu, Shaoyun
Subject: RE: [PATCH] drm/amdgpu:
[AMD Public Use]
Series is
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: Sheng, Wenhui
Sent: Wednesday, June 24, 2020 22:40
To: amd-gfx@lists.freedesktop.org
Cc: Zhang, Hawking ; Sheng, Wenhui
Subject: [PATCH 4/4] drm/amdgpu: merge atombios init block
From:
From: Wenhui Sheng
Init soc15 reg base early enough so we can touch
mailbox related registers in request full access
for sriov before set_ip_blocks, vi arch doesn't
use reg base in virt ops.
Signed-off-by: Wenhui Sheng
---
drivers/gpu/drm/amd/amdgpu/soc15.c | 25 ++---
1
From: Wenhui Sheng
After we move request full access before set
ip blocks, we can merge atombios init block
of SRIOV and baremetal together.
Signed-off-by: Wenhui Sheng
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 23 +-
1 file changed, 1 insertion(+), 22 deletions(-)
From: Wenhui Sheng
>From SIENNA_CICHLID, HW introduce a new protection
feature which can control the FB, doorbell and MMIO
write access for VF, so guest driver should request
full access before ip discovery, or we couldn't access
ip discovery data in FB.
Signed-off-by: Wenhui Sheng
---
From: Wenhui Sheng
Move request init data to virt detection func, so we
can insert request full access between request init data
and set ip blocks.
Signed-off-by: Wenhui Sheng
---
drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 28
drivers/gpu/drm/amd/amdgpu/nv.c
[AMD Public Use]
To make the description accurate, Let's say only read the first 4KB, starting
from total_vram - 64KB, for ip discovery data.
As discussed, no matter the TMR block is 64KB or 1MB, the ip discovery data
block always located in the same offset and with the same size (4KB)
[AMD Public Use]
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: Sheng, Wenhui
Sent: Wednesday, June 24, 2020 22:35
To: amd-gfx@lists.freedesktop.org
Cc: Zhang, Hawking ; Sheng, Wenhui
Subject: [PATCH] drm/amdgpu: reduce ip discovery data reading size
From:
From: Wenhui Sheng
Only read first 4K data instead of whole TMR block,
so we can reduce the time in full access mode.
Signed-off-by: Wenhui Sheng
---
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h | 3 ++-
2 files changed, 3 insertions(+),
Acked-by: Nirmoy Das
Thanks,
Nirmoy
On 6/24/20 4:14 PM, Colin King wrote:
From: Colin Ian King
Currently the goto statement is skipping over a lot of setup code
because it is outside of an if-block and should be inside it. Fix
this by adding missing if statement braces.
[AMD Public Use]
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: amd-gfx On Behalf Of Alex Deucher
Sent: Wednesday, June 24, 2020 22:20
To: amd-gfx list
Cc: Deucher, Alexander
Subject: Re: [PATCH] drm/amdgpu/uvd3.x: fix register definition warnings
Ping?
On
[AMD Public Use]
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: amd-gfx On Behalf Of Alex Deucher
Sent: Wednesday, June 24, 2020 22:20
To: amd-gfx list
Cc: Deucher, Alexander
Subject: Re: [PATCH] drm/amdgpu: make sure to reserve tmr region on all asics
which
Acked-by: Christian König
Am 24.06.20 um 16:19 schrieb Alex Deucher:
Ping?
On Mon, Jun 22, 2020 at 6:12 PM Alex Deucher wrote:
drop the duplicate register macros from sid.h and use the
standard ones in the oss register headers.
Signed-off-by: Alex Deucher
---
Acked-by: Nirmoy Das
On 6/22/20 11:41 PM, Alex Deucher wrote:
This includes older APUs like renoir.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
Ping?
Alex
On Mon, Jun 22, 2020 at 5:41 PM Alex Deucher wrote:
>
> This includes older APUs like renoir.
>
> Signed-off-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git
Ping?
On Mon, Jun 22, 2020 at 6:12 PM Alex Deucher wrote:
>
> drop the duplicate register macros from sid.h and use the
> standard ones in the oss register headers.
>
> Signed-off-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/si.c| 1 +
> drivers/gpu/drm/amd/amdgpu/si_ih.c | 2 ++
>
From: Colin Ian King
Currently the goto statement is skipping over a lot of setup code
because it is outside of an if-block and should be inside it. Fix
this by adding missing if statement braces.
Addresses-Coverity: ("Structurally dead code")
Fixes: fd151ca5396d ("drm amdgpu: SI UVD v3_1")
[AMD Public Use]
Series is:
Acked-by: Alex Deucher
From: amd-gfx on behalf of Likun Gao
Sent: Wednesday, June 24, 2020 2:35 AM
To: amd-gfx@lists.freedesktop.org
Cc: Gao, Likun ; Feng, Kenneth ;
Zhang, Hawking
Subject: [PATCH 1/2] drm/amd/powerplay: change
From: Colin Ian King
There is an off-by-one bounds check on the index into arrays
table->mc_reg_address and table->mc_reg_table_entry[k].mc_data[j] that
can lead to reads and writes outside of arrays. Fix the bound checking
off-by-one error.
Addresses-Coverity: ("Out-of-bounds read/write")
Port functionality from the Radeon driver to support
UVD and VCE power management.
Signed-off-by: Alex Jivin
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h| 9 +
drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 67 +++--
drivers/gpu/drm/amd/amdgpu/si.c| 361 -
On Tue, Jun 23, 2020 at 11:31:45PM -0400, Andrey Grodzovsky wrote:
>
> On 6/22/20 5:41 AM, Daniel Vetter wrote:
> > On Sun, Jun 21, 2020 at 02:03:02AM -0400, Andrey Grodzovsky wrote:
> > > On device removal reroute all CPU mappings to dummy page per drm_file
> > > instance or imported GEM object.
On Tue, Jun 23, 2020 at 11:04:30PM -0400, Andrey Grodzovsky wrote:
>
> On 6/23/20 2:05 AM, Greg KH wrote:
> > On Tue, Jun 23, 2020 at 12:51:00AM -0400, Andrey Grodzovsky wrote:
> > > On 6/22/20 12:45 PM, Greg KH wrote:
> > > > On Mon, Jun 22, 2020 at 12:07:25PM -0400, Andrey Grodzovsky wrote:
> >
On Sun, Jun 21, 2020 at 01:42:05PM -0400, Qian Cai wrote:
> On Wed, Jun 10, 2020 at 09:41:01PM +0200, Daniel Vetter wrote:
> > fs_reclaim_acquire/release nicely catch recursion issues when
> > allocating GFP_KERNEL memory against shrinkers (which gpu drivers tend
> > to use to keep the excessive
> On Jun 23, 2020, at 6:13 PM, Daniel Vetter wrote:
>
> Ok I tested this. I can't use your script to repro because
> - I don't have a setup with xfs, and the splat points at an issue in xfs
> - reproducing lockdep splats in shrinker callbacks is always a bit tricky
What’s xfs setup are you
From: Likun Gao
Update sienna_cichlid driver if header file to match pptable changes.
Signed-off-by: Likun Gao
Change-Id: Idcc081797f54509ea57503cce1dc915f718f189e
---
.../powerplay/inc/smu11_driver_if_sienna_cichlid.h | 12
drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h
From: Likun Gao
Copy board parameters directly instead of set each parameter for
sienna_cichlid.
Signed-off-by: Likun Gao
---
.../drm/amd/powerplay/sienna_cichlid_ppt.c| 89 +--
1 file changed, 2 insertions(+), 87 deletions(-)
diff --git
52 matches
Mail list logo