[Public]
Reviewed-by: Guchun Chen
Regards,
Guchun
-Original Message-
From: Shi, Leslie
Sent: Friday, December 17, 2021 10:26 AM
To: Grodzovsky, Andrey ; Koenig, Christian
; Pan, Xinhui ; Deucher,
Alexander ; amd-gfx@lists.freedesktop.org
Cc: Chen, Guchun ; Shi, Leslie
Subject:
The return value of dma_alloc_coherent() needs to be checked.
To avoid use of null pointer in memcpy_toio() in case of the failure of
alloc.
Fixes: 57430471e2fa ("drm/amdgpu: Add support for USBC PD FW download")
Signed-off-by: Jiasheng Jiang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 4
Dear Alex:>Is the issue reproducible with the same board in bare metal on x86?Or does it only happen with passthrough on ARM?Unfortunately, my current environment is not convenient to test this GPU board on x86 platform.but I can tell you the problem still occurs on ARM without passthrough to
[Why]
In amdgpu_driver_load_kms, when amdgpu_device_init returns error during driver
modprobe, it
will start the error handle path immediately and call into
amdgpu_device_unmap_mmio as well
to release mapped VRAM. However, in the following release callback, driver
stills visits the
unmapped
[AMD Official Use Only]
OK, I'll rename it before submit.
Regards,
Tao
> -Original Message-
> From: Zhang, Hawking
> Sent: Thursday, December 16, 2021 8:22 PM
> To: Zhou1, Tao ; amd-gfx@lists.freedesktop.org; Yang,
> Stanley ; Chai, Thomas ;
> Kuehling, Felix
> Subject: RE: [PATCH
On Thu, Dec 16, 2021 at 8:43 PM Quan, Evan wrote:
>
> [AMD Official Use Only]
>
> Hi Alex,
>
> Per our checking, vcn_v2 and vcn_v3 already have the dpm disablement(below)
> in their ->suspend routine which should prevent them from the issue here.
> if (adev->pm.dpm_enabled)
>
[AMD Official Use Only]
Hi Alex,
Per our checking, vcn_v2 and vcn_v3 already have the dpm disablement(below) in
their ->suspend routine which should prevent them from the issue here.
if (adev->pm.dpm_enabled)
amdgpu_dpm_enable_uvd(adev, false);
So, maybe it's a different
[AMD Official Use Only]
I will fix the indent issue. There shouldn't be any conflicts between this
patch and the xgmi init changes. We've run with those xgmi init patches applied
on top of mainline-dkms-5.13 (where this change is already present) and found
no issues.
Thanks,
Victor
Hi Dave, Daniel,
More updates for 5.17.
The following changes since commit 3c021931023a30316db415044531b116b85e6ebd:
drm/amdgpu: replace drm_detect_hdmi_monitor() with drm_display_info.is_hdmi
(2021-12-07 13:13:07 -0500)
are available in the Git repository at:
On Thu, Dec 16, 2021 at 2:43 PM Victor Skvortsov
wrote:
>
> Driver needs to call get_xgmi_info() before ip_init
> to determine whether it needs to handle a pending hive reset.
>
> Signed-off-by: Victor Skvortsov
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +++
>
[AMD Official Use Only]
Reviewed by: shaoyun.liu
-Original Message-
From: Skvortsov, Victor
Sent: Thursday, December 16, 2021 2:43 PM
To: amd-gfx@lists.freedesktop.org; Deng, Emily ; Liu, Monk
; Ming, Davis ; Liu, Shaoyun
; Zhou, Peng Ju ; Chen, JingWen
; Chen, Horace ; Nieto,
[AMD Official Use Only]
Reviewed-by: David Nieto
From: Skvortsov, Victor
Sent: Thursday, December 16, 2021 11:42 AM
To: amd-gfx@lists.freedesktop.org ; Deng, Emily
; Liu, Monk ; Ming, Davis
; Liu, Shaoyun ; Zhou, Peng Ju
; Chen, JingWen ; Chen, Horace
;
[AMD Official Use Only]
Reviewed-by: David Nieto
From: Skvortsov, Victor
Sent: Thursday, December 16, 2021 11:42 AM
To: amd-gfx@lists.freedesktop.org ; Deng, Emily
; Liu, Monk ; Ming, Davis
; Liu, Shaoyun ; Zhou, Peng Ju
; Chen, JingWen ; Chen, Horace
;
[AMD Official Use Only]
Reviewed-by: David Nieto
From: Skvortsov, Victor
Sent: Thursday, December 16, 2021 11:42 AM
To: amd-gfx@lists.freedesktop.org ; Deng, Emily
; Liu, Monk ; Ming, Davis
; Liu, Shaoyun ; Zhou, Peng Ju
; Chen, JingWen ; Chen, Horace
;
[AMD Official Use Only]
Reviewed-by: David Nieto
From: Skvortsov, Victor
Sent: Thursday, December 16, 2021 11:42 AM
To: amd-gfx@lists.freedesktop.org ; Deng, Emily
; Liu, Monk ; Ming, Davis
; Liu, Shaoyun ; Zhou, Peng Ju
; Chen, JingWen ; Chen, Horace
;
[AMD Official Use Only]
Reviewed-by: David Nieto
From: Skvortsov, Victor
Sent: Thursday, December 16, 2021 11:42 AM
To: amd-gfx@lists.freedesktop.org ; Deng, Emily
; Liu, Monk ; Ming, Davis
; Liu, Shaoyun ; Zhou, Peng Ju
; Chen, JingWen ; Chen, Horace
;
Expand RLCG interface for new GC read & write commands.
New interface will only be used if the PF enables the flag in pf2vf msg.
v2: Added a description for the scratch registers
Signed-off-by: Victor Skvortsov
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 117 --
1 file
Driver needs to call get_xgmi_info() before ip_init
to determine whether it needs to handle a pending hive reset.
Signed-off-by: Victor Skvortsov
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +++
drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 6 --
Modify GC register access from MMIO to RLCG if the indirect
flag is set
Signed-off-by: Victor Skvortsov
---
.../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 27 +--
1 file changed, 13 insertions(+), 14 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
Modify GC register access from MMIO to RLCG if the
indirect flag is set
v2: Replaced ternary operator with if-else for better
readability
Signed-off-by: Victor Skvortsov
---
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 57 ---
1 file changed, 43 insertions(+), 14
Add helper macros to change register access
from direct to indirect.
Signed-off-by: Victor Skvortsov
---
drivers/gpu/drm/amd/amdgpu/soc15_common.h | 5 +
1 file changed, 5 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h
b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
This patchset introduces an expanded sriov RLCG interface.
This interface will be used by Aldebaran in sriov
mode for indirect GC register access during full access.
v2: Added descriptions to scratch registers, and improved code readability.
v3: Remove the RLC function pointer init change. Move
On Thu, Dec 16, 2021 at 11:14 AM Richard Fontana wrote:
>
> On Thu, Dec 16, 2021 at 4:45 AM Lukas Bulwahn wrote:
> >
> > Commit b5f57384805a ("drm/amdkfd: Add sysfs bitfields and enums to uAPI")
> > adds include/uapi/linux/kfd_sysfs.h with the "GPL-2.0 OR MIT WITH
> > Linux-syscall-note"
The "plane_state" pointer was access before checking if it was NULL.
Avoid a possible NULL pointer dereference by accessing the plane
address after the check.
Addresses-Coverity-ID: 1474582 ("Dereference before null check")
Fixes: 3f68c01be9a22 ("drm/amd/display: add cyan_skillfish display
[Public]
If it needs to be in drm-next, please make sure it lands there.
Alex
From: amd-gfx on behalf of Liu, Shaoyun
Sent: Thursday, December 16, 2021 12:51 PM
To: Skvortsov, Victor ; Alex Deucher
Cc: Ming, Davis ; Chen, JingWen ;
amd-gfx list ; Deng,
[AMD Official Use Only]
Actually I don't know why the change " a35f147621bc drm/amdgpu: get xgmi info
at eary_init " not in drm-next , instead it’s in amd-mainline-dkms-5.13.
That change is necessary for passthrough XGMI hive to a VM and rely on our
driver to do the reset on whole hive
On Thu, Dec 16, 2021 at 4:45 AM Lukas Bulwahn wrote:
>
> Commit b5f57384805a ("drm/amdkfd: Add sysfs bitfields and enums to uAPI")
> adds include/uapi/linux/kfd_sysfs.h with the "GPL-2.0 OR MIT WITH
> Linux-syscall-note" SPDX-License expression.
>
> The command ./scripts/spdxcheck.py warns:
>
>
Maybe we just should use drm_dev_is_unplugged() for this particular case
because, there would be no race since when device is unplugged it's
final. It's the other way around that requires strict drm_dev_enter/exit
scope.
Andrey
On 2021-12-16 3:38 a.m., Christian König wrote:
The
[AMD Official Use Only]
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: amd-gfx On Behalf Of Alex Deucher
Sent: Thursday, December 16, 2021 11:18
To: amd-gfx@lists.freedesktop.org
Cc: Deucher, Alexander
Subject: [PATCH] drm/amdgpu: add support for IP discovery
[AMD Official Use Only]
This one looks better and more logical .
Reviewed By :Shaoyun.liu
-Original Message-
From: Skvortsov, Victor
Sent: Thursday, December 16, 2021 10:39 AM
To: amd-gfx@lists.freedesktop.org; Liu, Shaoyun ; Nieto,
David M
Cc: Skvortsov, Victor
Subject: [PATCH
We want to be able to call virt data exchange conditionally
after gmc sw init to reserve bad pages as early as possible.
Since this is a conditional call, we will need to call
it again unconditionally later in the init sequence.
Refactor the data exchange function so it can be
called multiple
FWIW, it looks like all versions of VCN need the same fix. There have
been reports of suspend failing when VCN is in use on other newer APUs
as well.
Alex
On Tue, Dec 14, 2021 at 12:59 AM Quan, Evan wrote:
>
> [AMD Official Use Only]
>
>
>
>
>
>
>
> From: Zhu, James
> Sent: Monday, December
Is the issue reproducible with the same board in bare metal on x86? Or
does it only happen with passthrough on ARM? Looking through the archives,
the SI patch I made was for an x86 laptop. It would be nice to root cause
this, but there weren't any gfx8 boards with more than 64G of vram, so I
Applied. Thanks!
Alex
On Thu, Dec 16, 2021 at 2:50 AM Jiapeng Chong
wrote:
>
> Fix the following coccicheck warning:
>
> ./drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c:744:35-36:
> WARNING comparing pointer to 0.
>
> Reported-by: Abaci Robot
> Signed-off-by: Jiapeng Chong
> ---
Ping?
On Wed, Dec 15, 2021 at 10:18 PM Alex Deucher wrote:
>
> Used on gfx9 based systems. Fixes incorrect CU counts reported
> in the kernel log.
>
> Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1833
> Signed-off-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c |
On Thu, Dec 16, 2021 at 9:24 AM Christian König
wrote:
>
>
>
> Am 16.12.21 um 15:14 schrieb Alex Deucher:
> > Add a new CTX ioctl operation to set stable pstates for profiling.
> > When creating traces for tools like RGP or using SPM or doing
> > performance profiling, it's required to enable a
On 2021-12-15 3:52 a.m.,
cgel@gmail.com wrote:
From: Changcheng Deng
Use max() and min() in order to make code cleaner.
Reported-by: Zeal Robot
Signed-off-by: Changcheng Deng
Reviewed-by: Philip Yang
Applied, thanks.
[Public]
Hi Victor,
calling get_xgmi_info() in gmc_v9_0_early_init is from below patch, which also
sent out to amd-gfx for review as a upstream patch. Do you know why it's missed
from upstream?
drm/amdgpu: get xgmi info at eary_init
Driver need to get XGMI info
On Tue, Dec 14, 2021 at 12:44 PM Surbhi Kakarya wrote:
>
> On SRIOV environment, if event guard is enabled and VF doesn't
> receive an ack from PF for full access, the guest driver load crashes.
> This is caused due to the call to ttm_device_clear_dma_mappings with
> non-initialized
> mman
[AMD Official Use Only]
Gotcha, I will skip this patch for drm-next
-Original Message-
From: Alex Deucher
Sent: Thursday, December 16, 2021 8:53 AM
To: Skvortsov, Victor
Cc: amd-gfx list ; Deng, Emily
; Liu, Monk ; Ming, Davis
; Liu, Shaoyun ; Zhou, Peng Ju
; Chen, JingWen ; Chen,
Am 16.12.21 um 15:14 schrieb Alex Deucher:
Add a new CTX ioctl operation to set stable pstates for profiling.
When creating traces for tools like RGP or using SPM or doing
performance profiling, it's required to enable a special
stable profiling power state on the GPU. These profiling
states
On 16/12/2021 14:15, Boris Brezillon wrote:
> Hi Steve,
>
> On Thu, 16 Dec 2021 14:02:25 +
> Steven Price wrote:
>
>> + Boris
>>
>> On 16/12/2021 12:08, Dan Carpenter wrote:
>>> Hi DRM Devs,
>>>
>>> In commit 7661809d493b ("mm: don't allow oversized kvmalloc() calls")
>>> from July, Linus
Hi Steve,
On Thu, 16 Dec 2021 14:02:25 +
Steven Price wrote:
> + Boris
>
> On 16/12/2021 12:08, Dan Carpenter wrote:
> > Hi DRM Devs,
> >
> > In commit 7661809d493b ("mm: don't allow oversized kvmalloc() calls")
> > from July, Linus added a WARN_ONCE() for "crazy" allocations over 2GB.
>
So mesa and tools know when this is available.
Mesa MR: https://gitlab.freedesktop.org/mesa/drm/-/merge_requests/207
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git
Add a new CTX ioctl operation to set stable pstates for profiling.
When creating traces for tools like RGP or using SPM or doing
performance profiling, it's required to enable a special
stable profiling power state on the GPU. These profiling
states set fixed clocks and disable certain other
+ Boris
On 16/12/2021 12:08, Dan Carpenter wrote:
> Hi DRM Devs,
>
> In commit 7661809d493b ("mm: don't allow oversized kvmalloc() calls")
> from July, Linus added a WARN_ONCE() for "crazy" allocations over 2GB.
> I have a static checker warning for this and most of the warnings are
> from DRM
[Public]
Reviewed-by: Alex Deucher
From: amd-gfx on behalf of Bokun Zhang
Sent: Wednesday, December 15, 2021 7:52 PM
To: amd-gfx@lists.freedesktop.org
Cc: Zhang, Bokun
Subject: [PATCH] drm/amdgpu: Filter security violation registers
Recently, there is
On Wed, Dec 15, 2021 at 6:58 PM Skvortsov, Victor
wrote:
>
> [AMD Official Use Only]
>
> Hey Alex,
>
> This change was based on the fact that amd-mainline-dkms-5.13 calls
> get_xgmi_info() in gmc_v9_0_early_init(). But I can see that drm-next it's
> instead called in gmc_v9_0_sw_init(). So, I'm
Commit b5f57384805a ("drm/amdkfd: Add sysfs bitfields and enums to uAPI")
adds include/uapi/linux/kfd_sysfs.h with the "GPL-2.0 OR MIT WITH
Linux-syscall-note" SPDX-License expression.
The command ./scripts/spdxcheck.py warns:
include/uapi/linux/kfd_sysfs.h: 1:48 Exception not valid for
Hi Christian,I'm testing for GPU passthrough feature, so I pass through this GPU to virtual machine to use. It based on arm64 system.As far as i know, Alex had dealt with a similar problems on dri/radeon/si.c . Maybe they have a same reason to cause it?the history commit message is
+
+ int (*unmap_queues_cpsch_poison)(struct device_queue_manager *dqm,
+ uint16_t pasid);
};
Might be better call it reset_queue directly (match with update_queue,
create_queue, .etc.,)
Others look good to me
The series (4 patches) is
Reviewed-by:
Hi DRM Devs,
In commit 7661809d493b ("mm: don't allow oversized kvmalloc() calls")
from July, Linus added a WARN_ONCE() for "crazy" allocations over 2GB.
I have a static checker warning for this and most of the warnings are
from DRM ioctls.
drivers/gpu/drm/lima/lima_drv.c:124
The new interface unmaps queues with reset mode for the process consumes
RAS poison, it's only for compute queue.
Signed-off-by: Tao Zhou
---
.../drm/amd/amdkfd/kfd_device_queue_manager.c| 16
.../drm/amd/amdkfd/kfd_device_queue_manager.h| 5 +
2 files changed, 21
CP supports unmap queue with reset mode which only destroys specific queue
without affecting others.
Replacing whole gpu reset with reset queue mode for RAS poison consumption
saves much time, and we can also fallback to gpu reset solution if reset
queue fails.
v2: Return directly if process is
So we can set reset mode for unmap operation, no functional change.
Signed-off-by: Tao Zhou
---
.../gpu/drm/amd/amdkfd/kfd_device_queue_manager.c| 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
Add a reset parameter for umc page retirement, let user decide whether
call gpu reset in umc page retirement.
Signed-off-by: Tao Zhou
---
drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c | 15 ---
drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h | 5 +++--
2 files changed, 15 insertions(+), 5
Hi
Am 01.12.21 um 17:39 schrieb Arunpravin:
- Make drm_buddy_alloc a single function to handle
range allocation and non-range allocation demands
- Implemented a new function alloc_range() which allocates
the requested power-of-two block comply with range limitations
- Moved order
On 15/12/2021 20:46, Arunpravin wrote:
On 14/12/21 12:29 am, Matthew Auld wrote:
On 09/12/2021 15:47, Paneer Selvam, Arunpravin wrote:
[AMD Official Use Only]
Hi Matthew,
Ping on this?
No new comments from me :) I guess just a question of what we should do
with the selftests, and then
Fix the following coccicheck warning:
./drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c:744:35-36:
WARNING comparing pointer to 0.
Reported-by: Abaci Robot
Signed-off-by: Jiapeng Chong
---
drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c | 2 +-
1 file changed, 1
The !drm_dev_enter() is quite unusual and deserves a comment explaining
what's going on here.
Apart from that it looks good with the typos fixed I think.
Christian.
Am 16.12.21 um 08:27 schrieb Chen, Guchun:
[Public]
My BAD to misunderstand this.
There are both spell typos in patch subject
Am 15.12.21 um 22:09 schrieb Ira Weiny:
On Tue, Dec 14, 2021 at 08:09:29AM +0100, Christian König wrote:
Am 14.12.21 um 04:37 schrieb Ira Weiny:
On Mon, Dec 13, 2021 at 09:37:32PM +0100, Christian König wrote:
Am 11.12.21 um 00:24 schrieb ira.we...@intel.com:
From: Ira Weiny
The default
Am 16.12.21 um 05:39 schrieb Huang Rui:
Fix the typo to align with the prototype and function name.
All warnings (new ones prefixed by >>):
drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c:631: warning: expecting
prototype for amdgpu_fence_clear_job_fences(). Prototype was for
Hi Zongmin,
that strongly sounds like the ASIC is not correctly initialized when
trying to read the register.
What board and environment are you using this GPU with? Is that a normal
x86 system?
Regards,
Christian.
Am 16.12.21 um 04:11 schrieb 周宗敏:
1.
the problematic boards that I
63 matches
Mail list logo
