On 27/07/2021 21:27, Thomas Zimmermann wrote:
Drop the DRM IRQ midlayer in favor of Linux IRQ interfaces. DRM's
IRQ helpers are mostly useful for UMS drivers. Modern KMS drivers
don't benefit from using it.
DRM IRQ callbacks are now being called directly or inlined.
Signed-off-by: Thomas Zimmer
If GC has entered CGPG, ringing doorbell > first page doesn't wakeup GC.
Enlarge CP_MEC_DOORBELL_RANGE_UPPER to workaround this issue.
Signed-off-by: Yifan Zhang
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 13 -
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/drivers/gp
Agree. Patch updated. Thanks.
-Original Message-
From: Alex Deucher
Sent: Thursday, July 29, 2021 10:24 AM
To: Zhang, Yifan
Cc: amd-gfx list
Subject: Re: [PATCH v2] drm/amdgpu: fix the doorbell missing when in CGPG issue
for renoir.
On Wed, Jul 28, 2021 at 9:33 PM Zhang, Yifan wrote
On 7/29/2021 1:25 PM, Yifan Zhang wrote:
If GC has entered CGPG, ringing doorbell > first page doesn't wakeup GC.
Enlarge CP_MEC_DOORBELL_RANGE_UPPER to workaround this issue.
Signed-off-by: Yifan Zhang
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 13 -
1 file changed, 12 inser
Hi Guchun,
see below.
Am 29.07.21 um 14:39 schrieb Chen, Guchun:
[Public]
Hi Christian,
Thanks for your feedback.
Originally, drm_sched_fini is part of amdgpu_fence_driver_hw_fini, I did not
move it.
Yeah, not saying that this is your fault, you should just clean that up
more thoughtfull
Exactly that yes.
Thanks,
Christian.
Am 29.07.21 um 14:56 schrieb Chen, Guchun:
[Public]
Got you, so the target is to take this chance to make the code logic more clear
instead of making it just workable on top of mixed functionality code.
I will post a more reasonable patch later on to addr
If GC has entered CGPG, ringing doorbell > first page doesn't wakeup GC.
Enlarge CP_MEC_DOORBELL_RANGE_UPPER to workaround this issue.
Signed-off-by: Yifan Zhang
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 21 -
1 file changed, 20 insertions(+), 1 deletion(-)
diff --git a/dr
Remove the repeated word 'the' from comments
Signed-off-by: Cai Huoqing
---
drivers/gpu/drm/radeon/atombios.h | 4 ++--
drivers/gpu/drm/radeon/r300_reg.h | 2 +-
drivers/gpu/drm/radeon/radeon_device.c | 2 +-
drivers/gpu/drm/radeon/radeon_fence.c | 2 +-
drivers/gpu/drm/radeon/radeon_
This patch fixes a spelling typo in error message.
Signed-off-by: Masanari Iida
---
drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c
b/drivers/gpu/drm/amd/pm/powerplay/hw
Remove the repeated word 'the' from comments
Signed-off-by: Cai Huoqing
---
drivers/gpu/drm/amd/display/dc/core/dc_resource.c | 2 +-
.../gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c | 2 +-
.../drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c | 2 +-
.../gpu/dr
In amdgpu_fence_driver_hw_fini, no need to call drm_sched_fini to stop
scheduler in s3 test, otherwise, fence errors will occur after resume.
So introduce a new parameter to distingiush the case in this API.
Fixes: cd87a6dcf6af drm/amdgpu: adjust fence driver enable sequence
Signed-off-by: Guchun
Am 29.07.21 um 12:49 schrieb Guchun Chen:
In amdgpu_fence_driver_hw_fini, no need to call drm_sched_fini to stop
scheduler in s3 test, otherwise, fence errors will occur after resume.
So introduce a new parameter to distingiush the case in this API.
NAK, the problem is rather that drm_sched_fin
[Public]
Hi Christian,
Thanks for your feedback.
Originally, drm_sched_fini is part of amdgpu_fence_driver_hw_fini, I did not
move it.
Former patch " cd87a6dcf6af drm/amdgpu: adjust fence driver enable sequence "
has dropped amdgpu_fence_driver_suspend, and called amdgpu_fence_driver_hw_fini
[Public]
Got you, so the target is to take this chance to make the code logic more clear
instead of making it just workable on top of mixed functionality code.
I will post a more reasonable patch later on to address this. Thank you.
Regards,
Guchun
-Original Message-
From: Christian Kö
On 7/28/21 8:22 AM, Christoph Hellwig wrote:
> On Tue, Jul 27, 2021 at 05:26:05PM -0500, Tom Lendacky via iommu wrote:
>> Introduce an x86 version of the prot_guest_has() function. This will be
>> used in the more generic x86 code to replace vendor specific calls like
>> sev_active(), etc.
>>
>> Wh
Am 2021-07-28 um 12:10 p.m. schrieb Tim Cahill:
> Hi Felix,
I'm not sure why you're calling me out by name. I'm not working on
anything obviously related to your crashes.
Anyway, I took a quick look at the backtraces. They all point at libgdk.
Two of them are segfaults, one is an abort. It's not
Hi
Am 28.07.21 um 22:11 schrieb Sam Ravnborg:
Hi Dan,
I think I got it - we need to set irq_enabled to true.
The documentation says so:
"
* @irq_enabled:
*
* Indicates that interrupt handling is enabled, specifically vblank
* handling. Drivers which
Hi
Am 29.07.21 um 21:18 schrieb Thomas Zimmermann:
Hi
Am 28.07.21 um 22:11 schrieb Sam Ravnborg:
Hi Dan,
I think I got it - we need to set irq_enabled to true.
The documentation says so:
"
* @irq_enabled:
*
* Indicates that interrupt handling is enabled,
sp
Hi
Am 29.07.21 um 21:24 schrieb dan.sned...@microchip.com:
Hi Thomas,
On 7/29/21 12:18 PM, Thomas Zimmermann wrote:
Hi
Am 28.07.21 um 22:11 schrieb Sam Ravnborg:
Hi Dan,
I think I got it - we need to set irq_enabled to true.
The documentation says so:
"
* @irq_enabled:
Fix the channel_index table layout to fetch the correct
channel_index when calculating physical address from
normalized address during page retirement.
Also, fix the number of UMC instances and number of channels
within each UMC instance for Aldebaran.
Signed-off-by: Mukul Joshi
---
drivers/gpu/
Am 2021-07-19 um 5:18 p.m. schrieb Alex Sierra:
> [Why]
> Avoid conflict with address ranges mapped by SVM
> mechanism that try to be allocated again through
> ioctl_alloc in the same process. And viceversa.
>
> [How]
> For ioctl_alloc_memory_of_gpu allocations
> Check if the address range passed i
Hi Thomas,
>
> Are you sure, you're testing with the latest drm-misc-next or drm-tip?
> Because using irq_enabled is deprecated and the flag was recently replaced
> by commit 1e4cd78ed493 ("drm: Don't test for IRQ support in VBLANK ioctls").
I was looking at drm-misc-fixes which did not have thi
[AMD Official Use Only]
Reviewed-By: John Clements
From: Joshi, Mukul
Sent: Thursday, July 29, 2021 11:37 PM
To: amd-gfx@lists.freedesktop.org
Cc: Clements, John ; Zhang, Hawking
; Joshi, Mukul
Subject: [PATCH] drm/amdgpu: Fix channel_index table layout for A
[Public]
Reviewed-by: Alex Deucher
From: amd-gfx on behalf of Yifan Zhang
Sent: Thursday, July 29, 2021 5:46 AM
To: amd-gfx@lists.freedesktop.org
Cc: Zhang, Yifan
Subject: [PATCH v4] drm/amdgpu: fix the doorbell missing when in CGPG issue for
renoir.
If GC
On 2021-07-28 4:35 p.m., Alex Deucher wrote:
> It's only used and enabled with DCN, so protect it.
>
> Fixes: b6b76b0315ed7b ("drm/amd/display: Fixed EdidUtility build errors")
> Signed-off-by: Alex Deucher
Reviewed-by: Harry Wentland
Harry
> ---
> drivers/gpu/drm/amd/display/dc/core/dc_link
This DC patchset brings improvements in multiple areas. In summary, we have:
* DC version 3.2.147
* DMUB FW release 0.0.77
* LTTPR, MPO improvements
* General bug fixes and stability improvements
--
Anthony Koo (1):
drm/amd/display: [FW Promotion] Release 0.0.77
Aric Cyr (1):
drm/amd/displa
Its a small set of changes this week for DC. In summary, we have:
* DC version 3.2.147
* LTTPR, MPO Improvements
* General stability and bug fixes
--
Anthony Koo (1):
drm/amd/display: [FW Promotion] Release 0.0.77
Aric Cyr (1):
drm/amd/display: 3.2.147
Guo, Bing (2):
drm/amd/display: Fix
Add u32 gfx_version field to kfd_node_properties and kfd_device_info.
Populate _device_info structs accordingly and expose to sysfs.
Signed-off-by: Graham Sider
---
drivers/gpu/drm/amd/amdkfd/kfd_device.c | 29 +++
drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 1 +
drivers/g
Its a small set of changes this week for DC. In summary, we have:
* DC version 3.2.147
* LTTPR, MPO Improvements
* General stability and bug fixes
--
Anthony Koo (1):
drm/amd/display: [FW Promotion] Release 0.0.77
Aric Cyr (1):
drm/amd/display: 3.2.147
Guo, Bing (2):
drm/amd/display: Fix
From: Wesley Chalmers
[WHY]
For DCN31 onward, LTTPR is to be enabled and set to Transparent by
VBIOS. Driver is to assume that VBIOS has done this without needing to
check the VBIOS interop bit.
[HOW]
Add LTTPR enable and interop VBIOS bits into dc->caps, and force-set the
interop bit to true f
From: Bing Guo
Why:
In DCN2x, HW doesn't automatically divide MASTER_UPDATE_LOCK_DB_X
by the number of pipes ODM Combined.
How:
Set MASTER_UPDATE_LOCK_DB_X to the value that is adjusted by the
number of pipes ODM Combined.
Reviewed-by: Martin Leung
Acked-by: Aurabindo Pillai
Signed-off-by: Bi
From: Bing Guo
[Why&How]
Hardware team suggested to use SRExitTime= 35.5us as w/a to prevent
underflow in certain modes.
Reviewed-by: Martin Leung
Acked-by: Aurabindo Pillai
Signed-off-by: Bing Guo
---
drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c | 4 ++--
1 file changed, 2 insert
From: Anthony Koo
Acked-by: Aurabindo Pillai
Signed-off-by: Anthony Koo
---
drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
From: Wesley Chalmers
[WHY]
For DCN31 onward, LTTPR is to be enabled and set to Transparent by
VBIOS. Driver is to assume that VBIOS has done this without needing to
check the VBIOS interop bit.
[HOW]
Add LTTPR enable and interop VBIOS bits into dc->caps, and force-set the
interop bit to true f
Its a small set of changes this week for DC. In summary, we have:
* DC version 3.2.147
* LTTPR, MPO Improvements
* General stability and bug fixes
--
Anthony Koo (1):
drm/amd/display: [FW Promotion] Release 0.0.77
Aric Cyr (1):
drm/amd/display: 3.2.147
Bing Guo (2):
drm/amd/display: Fix
From: Jude Shih
[Why] On S4 resume we also need to fix detection of when to reload DMCUB
firmware because we're currently using the VBIOS version which isn't
compatible with the driver version.
[How] Update the hardware init check for DCN31 since it's the ASIC that
has this issue.
Reviewed-by:
From: Qingqing Zhuo
[Why]
HPD disable and enable sequences are not mutually exclusive
on Linux. For HPDs that spans over 1s (i.e. HPD low = 1s),
part of the disable sequence (specifically, a request to SMU
to lower refclk) could come right before the call to PHY
enable, causing DMUB to access an
From: Jaehyun Chung
[Why]
We do not currently support ODM plus MPO on only one side of the
screen. This unsupported case causes validation calculations to
divide by zero due to invalid viewport values.
[How]
Add stopgap for the validation of ODM plus MPO on one side of
screen case.
Reviewed-by:
From: Roman Li
[Why]
Display Manager initializes array of vblank workqueues, but only 1 is used.
[How]
Use single instance init instead of array.
Reviewed-by: Qingqing Zhou
Acked-by: Aurabindo Pillai
Signed-off-by: Roman Li
---
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 8 ++--
From: Aric Cyr
Acked-by: Aurabindo Pillai
Signed-off-by: Aric Cyr
---
drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h
b/drivers/gpu/drm/amd/display/dc/dc.h
index 22e917714fe2..62c222d0402f 100644
--- a
Remember before enjoying your holiday that the deadline for XDC 2022
proposals is *September 1st, 2021* :-)
Feel free to submit your proposal before, so we can give you early
feedback on it!
Sam
On Thu, 2021-07-01 at 18:14 +0200, Samuel Iglesias Gonsálvez wrote:
> This is a reminder that the cal
I apologize if the name callout is disconcerting. I was trying to
follow instructions for sending bugs and saw your name listed as the
owner of this code area.
FYI, I'd done some more troubleshooting and tinkering regarding the
crashing and Mate seems to be at the center of all the issues. As a
re
On Thu, Jul 29, 2021 at 4:18 AM Cai Huoqing wrote:
>
> Remove the repeated word 'the' from comments
>
> Signed-off-by: Cai Huoqing
Applied. Thanks!
> ---
> drivers/gpu/drm/amd/display/dc/core/dc_resource.c | 2 +-
> .../gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c | 2
Applied. Thanks!
Alex
On Thu, Jul 29, 2021 at 4:21 AM Cai Huoqing wrote:
>
> Remove the repeated word 'the' from comments
>
> Signed-off-by: Cai Huoqing
> ---
> drivers/gpu/drm/radeon/atombios.h | 4 ++--
> drivers/gpu/drm/radeon/r300_reg.h | 2 +-
> drivers/gpu/drm/radeon/radeon_de
Applied. Thanks!
On Thu, Jul 29, 2021 at 9:56 AM Masanari Iida wrote:
>
> This patch fixes a spelling typo in error message.
>
> Signed-off-by: Masanari Iida
> ---
> drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/dr
sts.freedesktop.org
Cc: dri-de...@lists.freedesktop.org
Cc: linux-n...@vger.kernel.org
---
drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
--- linext-20210729.orig/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ linext-20210729/drivers/gpu/drm/amd/amdgpu/amdgpu_acp
[AMD Official Use Only]
> -Original Message-
> From: Ghannam, Yazen
> Sent: Thursday, June 3, 2021 5:13 PM
> To: Joshi, Mukul
> Cc: Borislav Petkov ; Alex Deucher ;
> x86-ml ; Kasiviswanathan, Harish
> ; lkml ;
> amd-gfx@lists.freedesktop.org
> Subject: Re: [PATCH] drm/amdgpu: Register
Get process vm root BO ref in case process is exiting and root BO is
freed, to avoid NULL pointer dereference backtrace:
BUG: unable to handle kernel NULL pointer dereference at
Call Trace:
amdgpu_show_fdinfo+0xfe/0x2a0 [amdgpu]
seq_show+0x12c/0x180
seq_read+0x153/0x410
vfs_read+0
Hi Thomas,
On 7/29/21 12:18 PM, Thomas Zimmermann wrote:
> Hi
>
> Am 28.07.21 um 22:11 schrieb Sam Ravnborg:
>> Hi Dan,
>>
I think I got it - we need to set irq_enabled to true.
The documentation says so:
"
* @irq_enabled:
*
* In
Hi Thomas and Sam,
On 7/29/21 12:48 PM, Sam Ravnborg wrote:
> EXTERNAL EMAIL: Do not click links or open attachments unless you know the
> content is safe
>
> Hi Thomas,
>
>>
>> Are you sure, you're testing with the latest drm-misc-next or drm-tip?
>> Because using irq_enabled is deprecated and
Get process vm root BO ref in case process is exiting and root BO is
freed, to avoid NULL pointer dereference backtrace:
BUG: unable to handle kernel NULL pointer dereference at
Call Trace:
amdgpu_show_fdinfo+0xfe/0x2a0 [amdgpu]
seq_show+0x12c/0x180
seq_read+0x153/0x410
vfs_read+0
This is more useful when talking to the SMU team to have the information
in this format, save one less step to manually do it.
Signed-off-by: Mario Limonciello
---
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 9 +++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm
This adjusts SMU13 to match SMU11 and SMU12 behaviors in not showing
this message all the time. It will still be accessible via
/sys/kernel/debug/dri/0/amdgpu_firmware_info
Signed-off-by: Mario Limonciello
---
drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c | 3 ---
1 file changed, 3 deletions(-
The SMU driver interface and firmware interface messages have comments
indicating that they are backwards compatible. Don't output at WARN
level and scare people there is a problem. Instead bring it down to
INFO level and the details about the mismatch into DEBUG levels.
Signed-off-by: Mario Lim
Correct yellow carp driver-PMFW interface version to v4.
Signed-off-by: Xiaomeng Hou
---
drivers/gpu/drm/amd/pm/inc/smu_v13_0.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v13_0.h
b/drivers/gpu/drm/amd/pm/inc/smu_v13_0.h
index 3fea2430dec0.
[AMD Official Use Only]
Acked-by: Alex Deucher
From: amd-gfx on behalf of Xiaomeng Hou
Sent: Thursday, July 29, 2021 10:40 PM
To: amd-gfx@lists.freedesktop.org
Cc: Hou, Xiaomeng (Matthew) ; Liu, Aaron
Subject: [PATCH] drm/amd/pm: update yellow carp pmfw int
56 matches
Mail list logo