On 2019-10-11 1:12 p.m., t...@kernel.org wrote:
Hello, Daniel.
On Wed, Oct 09, 2019 at 06:06:52PM +0200, Daniel Vetter wrote:
That's not the point I was making. For cpu cgroups there's a very well
defined connection between the cpu bitmasks/numbers in cgroups and the cpu
bitmasks you use in
Hi Leo and others,
sorry for the late reply. I just spent some time looking at your patches
and testing them on a Raven DCN-1.
I looked at how the vstartup line is computed in the dc_bandwidth_calcs
etc., and added some DRM_DEBUG statements to the dm_dcn_crtc_high_irq and
dm_pflip_high_irq
Am 29.11.19 um 15:29 schrieb Nirmoy:
Hi Christian,
On 11/26/19 10:45 AM, Christian König wrote:
It looks like a start, but there numerous things which needs to be
fixed.
Question number one is: What's that good for? Entities are not the
problem here. The real issue is the fence ring and the
Luben, This should take care of the warnings that you get when a navi fw
file is missing from initrd.
Regards,
Nirmoy
On 11/29/19 9:26 AM, Yintian Tao wrote:
When load amdgpu failed before create pm_sysfs and ucode_sysfs,
the pm_sysfs and ucode_sysfs should not be removed.
Otherwise, there
For now init_pg() is good enough for bare-metal and SR-IOV, but checkout GFX9,
there are more jobs in init_pg(), so in the future I'd expect there will be
More lines come into init_pg() on gfx10 and by that time I need to modify
init_pg() ,
So IMHO the best way is use init_csb() from the
[AMD Official Use Only - Internal Distribution Only]
It seems to me this is one line code that more than 80 characters. Please
kindly split it to several ones.
Other than that, the Patch is
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: Liu, Monk
Sent:
Well what we do here actually looks like complete overkill to me.
IIRC when the device is removed all subsequent sysfs directories are
removed automatically as well.
So calling sysfs_remove_group() is superflous in the first place.
Regards,
Christian.
Am 29.11.19 um 09:34 schrieb Nirmoy:
still need to init csb even for SRIOV
v2:
drop init_pg() for gfx10 at all since
PG and GFX off feature will be fully controled
by RLC and SMU fw for gfx10
Signed-off-by: Monk Liu
---
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 38 ++
1 file changed, 11
okay
_
Monk Liu|GPU Virtualization Team |AMD
-Original Message-
From: Zhang, Hawking
Sent: Friday, November 29, 2019 5:10 PM
To: Liu, Monk ; amd-gfx@lists.freedesktop.org
Cc: Liu, Monk
Subject: RE: [PATCH] drm/amdgpu: should stop GFX ring in
[AMD Public Use]
BTW, Please also make the change for backdoor loading code path as we need that
for further ASIC bring up. Thanks.
Regards,
Hawking
-Original Message-
From: Zhang, Hawking
Sent: 2019年11月29日 17:08
To: Liu, Monk ; amd-gfx@lists.freedesktop.org
Subject: RE: [PATCH 1/2]
Okay, will do
Thanks
_
Monk Liu|GPU Virtualization Team |AMD
-Original Message-
From: Zhang, Hawking
Sent: Friday, November 29, 2019 4:05 PM
To: Liu, Monk ; amd-gfx@lists.freedesktop.org; Deucher,
Alexander
Subject: RE: [PATCH 3/5] drm/amdgpu:
When load amdgpu failed before create pm_sysfs and ucode_sysfs,
the pm_sysfs and ucode_sysfs should not be removed.
Otherwise, there will be warning call trace just like below.
[ 24.836386] [drm] VCE initialized successfully.
[ 24.841352] amdgpu :00:07.0: amdgpu_device_ip_init failed
[
To align with the scheme from gfx9
disabling GFX ring after VM shutdown could avoid
garbage data be fetched to GFX RB which may lead
to unnecessary screw up on GFX
Signed-off-by: Monk Liu
---
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 1 +
1 file changed, 1 insertion(+)
diff --git
[AMD Official Use Only - Internal Distribution Only]
The gfx10 PG and gfxoff are fully controlled by RLC and SMU fw and don't need
driver to be involved anymore. So the legacy programing for
gfx_idle_threshold.etc will not be applied to gfx10.
More aggressively, let's drop gfx_v10_0_init_pg
Reviewed-by: Nirmoy Das
On 11/29/19 9:26 AM, Yintian Tao wrote:
When load amdgpu failed before create pm_sysfs and ucode_sysfs,
the pm_sysfs and ucode_sysfs should not be removed.
Otherwise, there will be warning call trace just like below.
[ 24.836386] [drm] VCE initialized successfully.
[
[AMD Official Use Only - Internal Distribution Only]
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: Liu, Monk
Sent: 2019年11月29日 15:47
To: Liu, Monk ; amd-gfx@lists.freedesktop.org; Zhang, Hawking
; Deucher, Alexander
Subject: RE: [PATCH 2/5] drm/amdgpu: skip
To align with the scheme from gfx9:
without disabling gfx ring in hw_fini we would
hit GFX hang if a guest VM is destroyed suddenly
when running a game
Signed-off-by: Monk Liu
---
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 1 +
1 file changed, 1 insertion(+)
diff --git
still need to init csb even for SRIOV
Signed-off-by: Monk Liu
---
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 74edfd9..230e8af 100644
---
Sorry, the description of " without disabling gfx ring in hw_fini we would hit
GFX hang if a guest VM is destroyed suddenly when running a game" is not correct
Will send another one.
_
Monk Liu|GPU Virtualization Team |AMD
-Original Message-
From:
[AMD Official Use Only - Internal Distribution Only]
I don't think you need to introduce SRIOV specific path here. The
gfx_v10_0_init_pg should be good enough to cover both bare-metal and
virtualization case.
You can drop the tlb flush in that function if that's your major concern.
Regards,
[AMD Official Use Only - Internal Distribution Only]
The debug message after gfx disablement actually makes me confusing, would you
mind to drop that.
Other than that, the patch is
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: amd-gfx On Behalf Of Monk Liu
Hi Christian
Do you mean we can remove sysfs_remove_group() for pm_sysfs and ucode_sysfs at
amdgpu_device_fini()?
If so , I think the sysfs directories will not be removed automatically.
When I remove sysfs_remove_group() at amdgpu_device_fini() and reload amdgpu,
then it will report the error
Do you mean the code path of " if (adev->firmware.load_type !=
AMDGPU_FW_LOAD_PSP) " , if so already handled
_
Monk Liu|GPU Virtualization Team |AMD
-Original Message-
From: Zhang, Hawking
Sent: Friday, November 29, 2019 5:22 PM
To: Liu, Monk ;
Thanks, done
_
Monk Liu|GPU Virtualization Team |AMD
-Original Message-
From: Zhang, Hawking
Sent: Friday, November 29, 2019 5:29 PM
To: Liu, Monk ; amd-gfx@lists.freedesktop.org
Cc: Liu, Monk
Subject: RE: [PATCH] drm/amdgpu: fix GFX10 missing CSIB
Hi Yintian,
I'm not familiar enough with the firmware handling for this, but feel
free to add an Acked-by: Christian König .
Regards,
Christian.
Am 29.11.19 um 10:32 schrieb Tao, Yintian:
Hi Christian
Thanks a lot. I got it. Can I get your RB?
Best Regards
Yintian Tao
-Original
[AMD Public Use]
Please also drop the tlb flush. It is probably from code used for the emulation.
+ for (i = 0; i < adev->num_vmhubs; i++)
+ amdgpu_gmc_flush_gpu_tlb(adev, 0, i, 0);
With that fixed, the patch is
Reviewed-by: Hawking Zhang
Regards,
Hawking
Am 29.11.19 um 10:25 schrieb Tao, Yintian:
Hi Christian
Do you mean we can remove sysfs_remove_group() for pm_sysfs and ucode_sysfs at
amdgpu_device_fini()?
At least I think so, the question is where this group is added?
If that is for some directory which is removed during driver unload
Hi Christian
Thanks a lot. I got it. Can I get your RB?
Best Regards
Yintian Tao
-Original Message-
From: Christian König
Sent: 2019年11月29日 17:30
To: Tao, Yintian ; Koenig, Christian
; Das, Nirmoy ;
amd-gfx@lists.freedesktop.org
Cc: Tuikov, Luben
Subject: Re: [PATCH] drm/amdgpu:
Yes, this can work using the same way as dqm_lock. This is trivial part,
Felix and Christian is discussing the solution of lock problem.
Regards,
Philip
On 2019-11-28 7:35 p.m., Zeng, Oak wrote:
[AMD Official Use Only - Internal Distribution Only]
Is kmalloc with GFP_NOWAIT an option here?
On Mon, Nov 18, 2019 at 06:18:31PM +0800, Wayne Lin wrote:
> [Why]
> HDMI 2.0 adds aspect ratio attribute to distinguish different
> 4k modes. According to Appendix E of HDMI 2.0 spec, source should
> use VSIF to indicate video mode only when the mode is one defined
> in HDMI 1.4b 4K modes.
Hi Christian,
On 11/26/19 10:45 AM, Christian König wrote:
It looks like a start, but there numerous things which needs to be fixed.
Question number one is: What's that good for? Entities are not the
problem here. The real issue is the fence ring and the rq_list.
The rq_list could actually
On Tue, 29 Oct 2019 14:07:48 -0400
Andrey Grodzovsky wrote:
> Fix a static code checker warning.
>
> v2: Drop PTR_ERR_OR_ZERO.
>
> Signed-off-by: Andrey Grodzovsky
> ---
> drivers/gpu/drm/scheduler/sched_main.c | 7 +--
> 1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git
On 11/28/19 4:00 AM, Ma, Le wrote:
-Original Message-
From: Grodzovsky, Andrey
Sent: Wednesday, November 27, 2019 11:46 PM
To: Ma, Le ; amd-gfx@lists.freedesktop.org
Cc: Chen, Guchun ; Zhou1, Tao
; Deucher, Alexander ;
Li, Dennis ; Zhang, Hawking
Subject: Re: [PATCH 07/10]
33 matches
Mail list logo