On Thu, Jun 21, 2018 at 8:26 AM, Daniel Vetter wrote:
> On Wed, Jun 20, 2018 at 4:21 PM, Christian König
> wrote:
>> Am 20.06.2018 um 16:04 schrieb Christian König:
>>>
>>> Am 20.06.2018 um 14:52 schrieb Daniel Vetter:
On Wed, Jun 20, 2018 at 2:46 PM, Christian König
wrote:
>
On Wed, Jun 20, 2018 at 4:21 PM, Christian König
wrote:
> Am 20.06.2018 um 16:04 schrieb Christian König:
>>
>> Am 20.06.2018 um 14:52 schrieb Daniel Vetter:
>>>
>>> On Wed, Jun 20, 2018 at 2:46 PM, Christian König
>>> wrote:
[SNIP]
>
> Go ahead, that's the point of commit right
On 2018-06-18 11:35 AM, Arnd Bergmann wrote:
> getrawmonotonic64() and get_monotonic_boottime64() are deprecated
> because of the nonstandard naming.
>
> The replacement functions ktime_get_raw_ns() and ktime_get_boot_ns()
> also simplify the callers.
>
> Signed-off-by: Arnd Bergmann
In case this
On Wed, Jun 20, 2018 at 12:48 PM, Tom St Denis wrote:
> This adds what should be a stable interface to read GPU
> load from userspace.
>
> (v2): Fix comments and name of file per recommendations.
>
> Signed-off-by: Tom St Denis
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 47
> +
Acked-by: Slava Abramov
From: amd-gfx on behalf of Tom St Denis
Sent: Wednesday, June 20, 2018 12:48:52 PM
To: amd-gfx@lists.freedesktop.org
Cc: StDenis, Tom
Subject: [PATCH] drm/amd/amdgpu: Add a GPU_LOAD entry to sysfs (v2)
This adds what should be a stable
Hi Dave,
A few fixes for amdgpu for 4.18, nothing major. Most going to stable.
The following changes since commit ce397d215ccd07b8ae3f71db689aedb85d56ab40:
Linux 4.18-rc1 (2018-06-17 08:04:49 +0900)
are available in the git repository at:
git://people.freedesktop.org/~agd5f/linux drm-fixe
In the quest to remove all stack VLA usage from the kernel[1], this
uses the maximum sane buffer size and removes copy/paste code.
[1]
https://lkml.kernel.org/r/CA+55aFzCG-zNmZwX4A2FQpadafLfEzK6CC=qpxydaacu1rq...@mail.gmail.com
Signed-off-by: Kees Cook
---
drivers/gpu/drm/amd/amdgpu/amdgpu_pm.
Series is:
Acked-by: Alex Deucher
From: amd-gfx on behalf of Andrey
Grodzovsky
Sent: Wednesday, June 20, 2018 2:17:22 PM
To: amd-gfx@lists.freedesktop.org; ckoenig.leichtzumer...@gmail.com
Cc: Panariti, David; Haehnle, Nicolai; Grodzovsky, Andrey
Subject: [PAT
Access to SQ_EDC_INFO requires selecting register instance and
hence mutex lock when accessing GRBM_GFX_INDEX for which a work
is schedueled from IH. But SQ interrupt can be raised on many instances
at once which means queuing work will usually succeed for the first one
but fail for the rest since
Switch to using reg fields defines istead of magic values.
Add SH_ID and PRIV fields reading for instr. and err cases.
Signed-off-by: Andrey Grodzovsky
Acked-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 36 +++
1 file changed, 20 insertions(+),
This adds what should be a stable interface to read GPU
load from userspace.
(v2): Fix comments and name of file per recommendations.
Signed-off-by: Tom St Denis
---
drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 47 ++
1 file changed, 47 insertions(+)
diff --git a/dr
OK, according to
https://www.kernel.org/doc/Documentation/memory-barriers.txt, SLEEP AND
WAKE-UP FUNCTIONS sections implies that.
Will respin the patch.
Andrey
On 06/20/2018 11:02 AM, Andrey Grodzovsky wrote:
Are you referring to insert_work->smp_mb(); ?
Andrey
On 06/20/2018 10:50 AM, C
On Wed, Jun 20, 2018 at 4:19 AM, Rex Zhu wrote:
> Missed to set some clock requests and display info which were needed
> by powerplay.
>
> Signed-off-by: Rex Zhu
Acked-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c | 9 +
> 1 file changed, 9 insert
On Wed, Jun 20, 2018 at 3:09 AM, Rex Zhu wrote:
> v2: not change the header file dm_service_types.h, as it was
> shared with other Os. PP need to include this header file and
> remove same data struct defines.
>
> 1. move shared struct dm_pp_wm_sets_with_clock_ranges_soc15 to
>dm_pp_in
On Wed, Jun 20, 2018 at 2:40 AM, Evan Quan wrote:
> On UVD/VCE dpm enabled/disabled, the powergate status will be
> set as false/true. So that we will not try to ungate/gate them(
> enable/disable their dpm) again.
>
> v2: added check for uvd/vce powergate status before gating
> v3: fix typo in de
On Wed, Jun 20, 2018 at 1:58 AM, Rex Zhu wrote:
> use smu_helper function smu_set_watermarks_for_clocks_ranges
> in vega12_set_watermarks_for_clocks_ranges.
>
> Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c | 43
> +---
On Wed, Jun 20, 2018 at 12:56 AM, Rex Zhu wrote:
> change wm_min_memg_clk_in_khz -> wm_min_mem_clk_in_khz
>
> Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c | 8
> drivers/gpu/drm/amd/display/dc/dm_services_types.h
Yes, understood. I just wanted to make sure we didn't merge both and break
things.
Alex
On Wed, Jun 20, 2018 at 10:51 AM, Zhu, Rex wrote:
> Hi Alex,
>
>
> Mikita's patch was still not in drm-next.
>
> I reviewed the interfaces between powerplay and display.
>
> Most of them were still not imp
Are you referring to insert_work->smp_mb(); ?
Andrey
On 06/20/2018 10:50 AM, Christian König wrote:
+ * Try to submit work so SQ_EDC_INFO can be accessed from
+ * BH. If previous work submission hasn't finished yet
+ * just print whatever info is possible directly from the ISR.
+
On Wed, Jun 20, 2018 at 8:31 AM, Tom St Denis wrote:
> This adds what should be a stable interface to read GPU
> load from userspace.
>
> Signed-off-by: Tom St Denis
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 41
> ++
> 1 file changed, 41 insertions(+)
>
>
Hi Alex,
Mikita's patch was still not in drm-next.
I reviewed the interfaces between powerplay and display.
Most of them were still not implemented. so the services powerplay exported to
dc will not be called.
I tried to implement them.
and then we can try to test the strutter mode on rave
On Wed, Jun 20, 2018 at 10:39 AM, Tom St Denis wrote:
>
>
> On 06/20/2018 10:37 AM, Abramov, Slava wrote:
>>
>> I see some functions in amdgpu_pm.c have function level documentation, so
>> that it would be good to have this for newly added functions.
>
>
> Sure I can add some comments/docs.
>
>
>>
Am 20.06.2018 um 16:43 schrieb Andrey Grodzovsky:
Access to SQ_EDC_INFO requires selecting register instance and
hence mutex lock when accessing GRBM_GFX_INDEX for which a work
is schedueled from IH. But SQ interrupt can be raised on many instances
at once which means queuing work will usually su
Hi Slava,
Ah yes, I copied it from another function (because the struct path to
the pp functions is annoying to look up hehehehe).
Thanks!
I'll submit a v2 in a bit.
Cheers,
Tom
On 06/20/2018 10:41 AM, Abramov, Slava wrote:
Should the comment then say 'get the load' instead of 'get the tem
Am 20.06.2018 um 13:25 schrieb Huang Rui:
On Tue, Jun 19, 2018 at 02:57:00PM +0200, Christian König wrote:
Always validating the VM PTs takes to much time. Only always validate
the per VM BOs for now.
Christian, you delete PTE BO instead of moving them into idle list. The
intention is to avoid
Access to SQ_EDC_INFO requires selecting register instance and
hence mutex lock when accessing GRBM_GFX_INDEX for which a work
is schedueled from IH. But SQ interrupt can be raised on many instances
at once which means queuing work will usually succeed for the first one
but fail for the reset since
Switch to using reg fields defines istead of magic values.
Add SH_ID and PRIV fields reading for instr. and err cases.
Signed-off-by: Andrey Grodzovsky
---
drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 36 +++
1 file changed, 20 insertions(+), 16 deletions(-)
diff --gi
Should the comment then say 'get the load' instead of 'get the temperature'?
From: StDenis, Tom
Sent: Wednesday, June 20, 2018 10:39:25 AM
To: Abramov, Slava; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/amd/amdgpu: Add a GPU_LOAD entry to sysfs
On 06/
On 06/20/2018 10:37 AM, Abramov, Slava wrote:
I see some functions in amdgpu_pm.c have function level documentation,
so that it would be good to have this for newly added functions.
Sure I can add some comments/docs.
Another comment is inline.
From: amd-gfx on behalf of Tom St Denis
I see some functions in amdgpu_pm.c have function level documentation, so that
it would be good to have this for newly added functions.
Another comment is inline.
>From: amd-gfx on behalf of Tom St
>Denis
>Sent: Wednesday, June 20, 2018 8:31 AM
>To: amd-gfx@lists.freedesktop.org
>Cc: StDen
Am 20.06.2018 um 16:04 schrieb Christian König:
Am 20.06.2018 um 14:52 schrieb Daniel Vetter:
On Wed, Jun 20, 2018 at 2:46 PM, Christian König
wrote:
[SNIP]
Go ahead, that's the point of commit rights. dim might complain if you
cherry picked them and didn't pick them up using dim apply though
On 2018-06-20 08:05 AM, Dan Carpenter wrote:
> We changed this loop so now it loops 1000 times instead of 10. We need
> to make the counter larger because a uint8_t can't go up to 1000 and
> we need to update the test at the end to test for 1000 instead of 10.
>
> Fixes: 2b6199a1d1b7 ("drm/amd/di
With your proposed changes patch is:
Acked-by: Alex Deucher
From: amd-gfx on behalf of Quan, Evan
Sent: Wednesday, June 20, 2018 2:17:57 AM
To: Alex Deucher
Cc: amd-gfx list
Subject: RE: [PATCH 10/13] drm/amd/powerplay: apply clocks adjust rules on
power stat
Am 20.06.2018 um 14:52 schrieb Daniel Vetter:
On Wed, Jun 20, 2018 at 2:46 PM, Christian König
wrote:
[SNIP]
Go ahead, that's the point of commit rights. dim might complain if you
cherry picked them and didn't pick them up using dim apply though ...
I've fixed up the Link tags, but when I tr
On Wed, Jun 20, 2018 at 2:46 PM, Christian König
wrote:
> Am 19.06.2018 um 17:35 schrieb Daniel Vetter:
>>
>> On Tue, Jun 19, 2018 at 4:47 PM, Christian König
>> wrote:
>>>
>>> Am 18.06.2018 um 10:18 schrieb Daniel Vetter:
On Fri, Jun 01, 2018 at 02:00:17PM +0200, Christian König wrote:
Am 19.06.2018 um 17:35 schrieb Daniel Vetter:
On Tue, Jun 19, 2018 at 4:47 PM, Christian König
wrote:
Am 18.06.2018 um 10:18 schrieb Daniel Vetter:
On Fri, Jun 01, 2018 at 02:00:17PM +0200, Christian König wrote:
Neither used nor correctly implemented anywhere. Just completely remove
the inte
This adds what should be a stable interface to read GPU
load from userspace.
Signed-off-by: Tom St Denis
---
drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 41 ++
1 file changed, 41 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
b/drivers/gpu/drm/am
We changed this loop so now it loops 1000 times instead of 10. We need
to make the counter larger because a uint8_t can't go up to 1000 and
we need to update the test at the end to test for 1000 instead of 10.
Fixes: 2b6199a1d1b7 ("drm/amd/display: replace msleep with udelay in fbc path")
Signed-
On Tue, Jun 19, 2018 at 02:57:00PM +0200, Christian König wrote:
> Always validating the VM PTs takes to much time. Only always validate
> the per VM BOs for now.
Christian, you delete PTE BO instead of moving them into idle list. The
intention is to avoid them do evction back when do vm_validate_
> -Original Message-
> From: Christian König [mailto:ckoenig.leichtzumer...@gmail.com]
> Sent: Wednesday, June 20, 2018 2:37 AM
> To: Grodzovsky, Andrey ; amd-
> g...@lists.freedesktop.org
> Cc: Panariti, David ; Haehnle, Nicolai
>
> Subject: Re: [PATCH 2/2] drm/amdgpu: Add parsing SQ_ED
Missed to set some clock requests and display info which were needed
by powerplay.
Signed-off-by: Rex Zhu
---
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
b/driv
v2: not change the header file dm_service_types.h, as it was
shared with other Os. PP need to include this header file and
remove same data struct defines.
1. move shared struct dm_pp_wm_sets_with_clock_ranges_soc15 to
dm_pp_interface.h.
2. delete the same struct define in powerplay, us
42 matches
Mail list logo