On 09/02/2016 03:44 AM, Tom St Denis wrote:
> Various whitespace and logical simplifications for gfx6.
>
> Signed-off-by: Tom St Denis
> ---
> drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c | 40
> +--
> 1 file changed, 5 insertions(+), 35 deletions(-)
>
> diff --git a
Reviewed-by: Edward O'Callaghan
On 09/02/2016 04:11 AM, Alex Deucher wrote:
> The mqd is only used on CI and newer.
>
> Signed-off-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c | 23 ---
> 1 file changed, 23 deletions(-)
>
> diff --git a/drivers/gpu/drm/am
On 02/09/16 05:38 AM, Alex Deucher wrote:
> On Wed, Aug 31, 2016 at 9:03 PM, Michel Dänzer wrote:
>> On 01/09/16 07:08 AM, Alex Deucher wrote:
>>> Close was not handled correctly.
>>>
>>> Signed-off-by: Alex Deucher
>>> ---
>>> drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 4
>>> 1 file changed
On 02/09/16 05:44 AM, Alex Deucher wrote:
> On Wed, Aug 31, 2016 at 9:04 PM, Michel Dänzer wrote:
>> On 01/09/16 07:08 AM, Alex Deucher wrote:
>>> Ported from nouveau.
>>>
>>> Signed-off-by: Alex Deucher
>>
>> [...]
>>
>>> +static int
>>> +amdgpufb_open(struct fb_info *info, int user)
>>> +{
>>>
Hi Alex:
Weird, last day, I used the fiji to check the S4 issue, there are some atom
errors, but it did not observed if add atombios_fini/init.
today, I change another fiji and also check the issue with the patch which
comment out the atombios_fini/init. but could not see the atombios error.
I
On Wed, Aug 31, 2016 at 9:04 PM, Michel Dänzer wrote:
> On 01/09/16 07:08 AM, Alex Deucher wrote:
>> Ported from nouveau.
>>
>> Signed-off-by: Alex Deucher
>
> [...]
>
>> +static int
>> +amdgpufb_open(struct fb_info *info, int user)
>> +{
>> + struct amdgpu_fbdev *rfbdev = info->par;
>> +
On Wed, Aug 31, 2016 at 9:03 PM, Michel Dänzer wrote:
> On 01/09/16 07:08 AM, Alex Deucher wrote:
>> Close was not handled correctly.
>>
>> Signed-off-by: Alex Deucher
>> ---
>> drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 4
>> 1 file changed, 4 insertions(+)
>>
>> diff --git a/drivers/gpu/dr
gh
On Thu, Sep 1, 2016 at 7:11 PM, Christian König wrote:
> Thanks for jumping in and explaining what mesa needs on the user space side.
>
> I already had the feeling that this is a bit swampy, but was about to let it
> pass because of lack of time to question it further. But with your notes
> th
The mqd is only used on CI and newer.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c | 23 ---
1 file changed, 23 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index 17a560c..040bc7f 100644
---
Am 01.09.2016 um 19:44 schrieb Tom St Denis:
Various whitespace and logical simplifications for gfx6.
Signed-off-by: Tom St Denis
---
drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c | 40 +--
1 file changed, 5 insertions(+), 35 deletions(-)
diff --git a/drivers/gpu/drm
Am 01.09.2016 um 19:44 schrieb Tom St Denis:
Add missing lock around SE/SH/INSTANCE selections.
Signed-off-by: Tom St Denis
This one and #2 are Reviewed-by: Christian König .
---
drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/gpu
Signed-off-by: Tom St Denis
---
drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c | 8 +---
1 file changed, 1 insertion(+), 7 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index 099539f0fa53..5f508c96496f 100644
--- a/drivers/gpu/drm/amd/amdgpu/
Patch #1 adds missing mutex locking around se/sh/instance writes.
Patches #2/3 perform various cleanups.
___
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
Add missing lock around SE/SH/INSTANCE selections.
Signed-off-by: Tom St Denis
---
drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index 17a560cd6420..099539f0fa53 100
Various whitespace and logical simplifications for gfx6.
Signed-off-by: Tom St Denis
---
drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c | 40 +--
1 file changed, 5 insertions(+), 35 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
b/drivers/gpu/drm/amd/amd
Thanks for jumping in and explaining what mesa needs on the user space side.
I already had the feeling that this is a bit swampy, but was about to
let it pass because of lack of time to question it further. But with
your notes that proposed approach is clearly a NAK.
I would prefer keeping t
On Thu, Sep 1, 2016 at 12:55 PM, Liu, Monk wrote:
>> Why does that makes a difference if it is seen for the first time?
>>
>> [ml] if it is presented for the first time for belonging ctx, means even
>> current CS do not involve context switch, we still need keep the actions in
>> preamble IB.
>>
Well then there is something wrong with the patch. Setting the preamble flag
should result in an IB only being executed when there was a task switch.
We can either implement that as a software solution by skipping the IBs in
question or by using the hardware for this.
Double checking your patch
Which part is no correct ?
When CONTEXT_CONTROL introduced in kernel side, it dynamically skips or keeps
the commands in preamble CEIB (LOAD_CE_RAM), so the Preamble CE IB not needed
to skipped by software method.
But don't forget that CONTEXT_CONTROL also dynamically controls if the
following
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Christian König
> Sent: Thursday, September 01, 2016 10:16 AM
> To: amd-gfx@lists.freedesktop.org
> Subject: [PATCH 2/2] drm/amdgpu: prevent command submission failures
> under memory pressure
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Christian König
> Sent: Thursday, September 01, 2016 10:16 AM
> To: amd-gfx@lists.freedesktop.org
> Subject: [PATCH 1/2] drm/amdgpu: only try again if we actually run into -
> ENOMEM
>
> From
From: Christian König
As last resort try to evict BOs from the current working set into other
memory domains. This effectively prevents command submission failures when
VM page tables have been swapped out.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h| 1 +
drive
From: Christian König
All other errors can't be fixed by using a different memory domain.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 8 +++-
1 file changed, 3 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
b/drivers/gpu/
Am 01.09.2016 um 12:55 schrieb Liu, Monk:
Why does that makes a difference if it is seen for the first time?
[ml] if it is presented for the first time for belonging ctx, means even
current CS do not involve context switch, we still need keep the actions in
preamble IB.
Usually if current CS i
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Michel Dänzer
> Sent: Thursday, September 01, 2016 3:46 AM
> To: amd-gfx@lists.freedesktop.org
> Subject: [PATCH v2 xf86-video-ati 1/2] Only copy from screen pixmap to
> shared pixmap on deman
On Thu, Sep 1, 2016 at 2:22 AM, jimqu wrote:
> reset the asic if adapter is not powerdown when doing freeze()
> thaw() and restore(), in order to get a valid state of adapter.
>
> Change-Id: Id2e4587c06b556eb651ab9277ffc903f780dea56
> Signed-off-by: JimQu
> ---
> drivers/gpu/drm/amd/amdgpu/amdgp
> Why does that makes a difference if it is seen for the first time?
>
> [ml] if it is presented for the first time for belonging ctx, means even
> current CS do not involve context switch, we still need keep the actions in
> preamble IB.
> Usually if current CS is from the same cntx of previous
v1:
for gfx8, use CONTEXT_CONTROL package to dynamically
skip preamble CEIB and other load_xxx command in sequence.
v2:
support GFX7 as well, and bump up version.
remove cntxcntl in compute ring funcs because CPC doesn't
support this packet.
v3: fix reduntant judgement in cntxcntl.
v4: some clean
On Thu, Sep 1, 2016 at 9:37 AM, Liu, Monk wrote:
>
>
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf Of
> Christian K?nig
> Sent: Wednesday, August 31, 2016 7:53 PM
> To: Liu, Monk ; amd-gfx@lists.freedesktop.org
> Subject: Re: [PATCH] drm/amd
v1:
for gfx8, use CONTEXT_CONTROL package to dynamically
skip preamble CEIB and other load_xxx command in sequence.
v2:
support GFX7 as well, and bump up version.
remove cntxcntl in compute ring funcs because CPC doesn't
support this packet.
v3: fix reduntant judgement in cntxcntl.
v4: some clean
From: Michel Dänzer
Only copy once for each time we update the corresponding scanout pixmap.
This can significantly reduce the bandwidth usage when there are
frequent updates to the screen pixmap.
This initial implementation only works when both the master and slave
screens use this driver.
v2:
-Original Message-
From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf Of
Christian K?nig
Sent: Wednesday, August 31, 2016 7:53 PM
To: Liu, Monk ; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/amdgpu:implement CONTEXT_CONTROL (v3)
Looks good to me in general,
32 matches
Mail list logo
