From: Michel Dänzer
Preparation for following change, no functional change intended.
(Ported from radeon commit aff267ee36cc6a703a532f91f82adc1ba1425ff3)
Signed-off-by: Michel Dänzer
---
src/amdgpu_dri2.c| 5 +++--
src/amdgpu_kms.c | 4 ++--
src/amdgpu_present.c | 2 +-
3 files change
From: Michel Dänzer
In order to avoid possible freeze / log file spam in that case.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99769
(Ported from radeon commit 94dc2b80f3ef0b2c17c20501d824fb0447d52e7a)
Signed-off-by: Michel Dänzer
---
src/amdgpu_kms.c | 21 +--
From: Christian König
The test was relaxed a bit to much.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
b/drivers/gpu/drm/amd/amdgpu/amdgpu_objec
Reviewed-and-Tested-by: Roger He
Thanks
Roger(Hongbo.He)
-Original Message-
From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf Of
Christian K?nig
Sent: Thursday, July 13, 2017 6:23 PM
To: amd-gfx@lists.freedesktop.org
Cc: Huang, Ray
Subject: [PATCH] drm/amdgpu: fix a
Seems to fix the regression I noticed on Carrizo. Can add my Acked-By
if you haven't pushed it already.
Cheers,
Tom
On 13/07/17 08:19 AM, He, Roger wrote:
Reviewed-and-Tested-by: Roger He
Thanks
Roger(Hongbo.He)
-Original Message-
From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesk
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Michel Dänzer
> Sent: Thursday, July 13, 2017 4:46 AM
> To: amd-gfx@lists.freedesktop.org
> Subject: [PATCH xf86-video-amdgpu 2/2] If a TearFree flip fails, fall back to
> non-TearFree operati
From: Christian König
This should save us a bunch of command submission overhead.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 15 +++--
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 56 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h |
From: Christian König
That should reduce our command submission overhead quite a bit.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 7 ++-
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h | 1 +
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/d
From: Christian König
No need to do this on every CS.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 17 -
1 file changed, 12 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.
Signed-off-by: Tom St Denis
---
src/app/ring_read.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/src/app/ring_read.c b/src/app/ring_read.c
index d6b3df58be4d..714a54e4f5ea 100644
--- a/src/app/ring_read.c
+++ b/src/app/ring_read.c
@@ -52,6 +52,9 @@ void umr_read_ring(struct umr_asic *as
Am 12.07.2017 um 22:31 schrieb Felix Kuehling:
I'm wondering if you need a HDP flush in update_directories at all. Are
there cases where the directories are updated but the page tables are not?
It can happen that a page table is swapped out and back in. In this case
on the PDs need an update,
Am 12.07.2017 um 21:38 schrieb Felix Kuehling:
On 17-07-12 03:50 AM, Christian König wrote:
That optimization is unnecessary.
I have a patch in the pipeline that makes the VM BO permanently CPU
mapped and also fixes a couple of bugs in those code path.
You mean just for the CPU-update path?
This patch detects the sequence of PKT0 writes to registers
which configure the VMID/ADDR/SIZE of the IBs in the ring and
then chains them once all 4 pieces of information are found.
Signed-off-by: Tom St Denis
---
src/lib/ring_decode.c | 26 +-
src/umr.h | 1
Makes it more obvious which lines of output are the data
and not headers for PKT0 format packets.
Signed-off-by: Tom St Denis
---
src/lib/ring_decode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/lib/ring_decode.c b/src/lib/ring_decode.c
index d87d489b1e4a..c00300e4ce
Reviewed-by: Felix Kuehling
On 17-07-13 09:25 AM, Christian König wrote:
> From: Christian König
>
> This should save us a bunch of command submission overhead.
>
> Signed-off-by: Christian König
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 15 +++--
> drivers/gpu/drm/amd/amdgpu/amdgp
On 17-07-13 09:25 AM, Christian König wrote:
> From: Christian König
>
> No need to do this on every CS.
>
> Signed-off-by: Christian König
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 17 -
> 1 file changed, 12 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/
On 17-07-13 09:25 AM, Christian König wrote:
> From: Christian König
>
> That should reduce our command submission overhead quite a bit.
> Signed-off-by: Christian König
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 7 ++-
> drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h | 1 +
> 2 files changed
Running the top command on the PF now allows monitoring activity for
a single VF by using [] keys and also to monitor relative assignment
of each VF.
Signed-off-by: Jean-Francois Thibert
---
src/app/top.c | 215 +++---
1 file changed, 205 inser
Applies cleanly but I can't get test it since I don't have a VM setup.
If nobody objects to it in a short while I'll push it through.
Tom
On 13/07/17 12:19 PM, Jean-Francois Thibert wrote:
Running the top command on the PF now allows monitoring activity for
a single VF by using [] keys and als
The redundant fb helpers .load_lut, .gamma_set and .gamma_get are
no longer used. Remove the dead code and hook up the crtc .gamma_set
to use the crtc gamma_store directly instead of duplicating that
info locally.
Acked-by: Daniel Vetter
Signed-off-by: Peter Rosin
---
drivers/gpu/drm/amd/amdgpu
Hi!
While trying to get CLUT support for the atmel_hlcdc driver, and
specifically for the emulated fbdev interface, I received some
push-back that my feeble in-driver attempts should be solved
by the core. This is my attempt to do it right.
I have obviously not tested all of this with more than a
The redundant fb helpers .load_lut, .gamma_set and .gamma_get are
no longer used. Remove the dead code and hook up the crtc .gamma_set
to use the crtc gamma_store directly instead of duplicating that
info locally.
Acked-by: Daniel Vetter
Signed-off-by: Peter Rosin
---
drivers/gpu/drm/radeon/ato
On 2017-07-12 02:26 PM, Jay Cornwall wrote:
The number of compute queues available to the KFD was erroneously
calculated as 64. Only the first MEC can execute compute queues and
it has 32 queue slots.
This caused the oversubscription limit to be calculated incorrectly,
leading to a missing chain
On 2017-07-13 02:36 PM, Andres Rodriguez wrote:
On 2017-07-12 02:26 PM, Jay Cornwall wrote:
The number of compute queues available to the KFD was erroneously
calculated as 64. Only the first MEC can execute compute queues and
it has 32 queue slots.
This caused the oversubscription limit to be
On Thu, Jul 13, 2017, at 13:36, Andres Rodriguez wrote:
> On 2017-07-12 02:26 PM, Jay Cornwall wrote:
> > The number of compute queues available to the KFD was erroneously
> > calculated as 64. Only the first MEC can execute compute queues and
> > it has 32 queue slots.
> >
> > This caused the ove
On 17-07-13 03:15 PM, Jay Cornwall wrote:
> On Thu, Jul 13, 2017, at 13:36, Andres Rodriguez wrote:
>> On 2017-07-12 02:26 PM, Jay Cornwall wrote:
>>> The number of compute queues available to the KFD was erroneously
>>> calculated as 64. Only the first MEC can execute compute queues and
>>> it has
On 2017-07-13 03:35 PM, Felix Kuehling wrote:
On 17-07-13 03:15 PM, Jay Cornwall wrote:
On Thu, Jul 13, 2017, at 13:36, Andres Rodriguez wrote:
On 2017-07-12 02:26 PM, Jay Cornwall wrote:
The number of compute queues available to the KFD was erroneously
calculated as 64. Only the first MEC c
The number of compute queues available to the KFD was erroneously
calculated as 64. Only the first MEC can execute compute queues and
it has 32 queue slots.
This caused the oversubscription limit to be calculated incorrectly,
leading to a missing chained runlist command at the end of an
oversubscr
There is a function get_mec_num use the field , but seems no one call it ,
maybe remove it as well.
Regards
Shaoyun.liu
-Original Message-
From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf Of
Andres Rodriguez
Sent: Thursday, July 13, 2017 3:54 PM
To: Kuehling,
Reviewed-by: Andres Rodriguez
On 2017-07-13 04:23 PM, Jay Cornwall wrote:
The number of compute queues available to the KFD was erroneously
calculated as 64. Only the first MEC can execute compute queues and
it has 32 queue slots.
This caused the oversubscription limit to be calculated incorr
Allows gdb to access contents of user mode mapped BOs. VRAM access
requires the driver to implement a new callback in ttm_bo_driver.
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/ttm/ttm_bo_vm.c | 66 -
include/drm/ttm/ttm_bo_driver.h | 12
2
Allows gdb to access contents of user mode mapped VRAM BOs.
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 59 +
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 2 ++
2 files changed, 61 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgp
People don't seem to like cross-component changes if they can be
avoided. I'd prefer separate commits for KFD, radeon, amdgpu, and
finally the kgd2kfd interface.
With that fixed, Reviewed-by: Felix Kuehling
On 17-07-13 04:36 PM, Andres Rodriguez wrote:
>
> Reviewed-by: Andres Rodriguez
>
> On 2
On 17-07-13 05:08 PM, Felix Kuehling wrote:
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
> @@ -78,6 +78,8 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo,
> struct dma_fence **fence);
>
> int amdgpu_mmap(struct file *filp, s
On Thu, Jul 13, 2017 at 06:22:50PM +0800, Christian König wrote:
> From: Christian König
>
> The test was relaxed a bit to much.
>
> Signed-off-by: Christian König
Nice catch, yes, we need take care the case that no gtt is bound.
Acked-by: Huang Rui
> ---
> drivers/gpu/drm/amd/amdgpu/amdgp
Dead code.
Change-Id: Ic0bb1bcca87e96bc5e8fa9894727b0de152e8818
Signed-off-by: Jay Cornwall
---
drivers/gpu/drm/amd/amdkfd/kfd_device.c | 4
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 7 ---
2 files changed, 11 deletions(-)
diff --git a/drivers/gpu/drm/amd/am
Dead code.
Change-Id: I9575aa73b5741b80dc340f953cc773385c92b2be
Signed-off-by: Jay Cornwall
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 1 -
drivers/gpu/drm/amd/include/kgd_kfd_interface.h | 3 ---
2 files changed, 4 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
Dead code.
Change-Id: I2383e0b541ed55288570b6a0ec8a0d49cdd4df89
Signed-off-by: Jay Cornwall
---
drivers/gpu/drm/radeon/radeon_kfd.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/radeon/radeon_kfd.c
b/drivers/gpu/drm/radeon/radeon_kfd.c
index 719ea51..8f8c7c1 100644
--- a/dr
The number of compute queues available to the KFD was erroneously
calculated as 64. Only the first MEC can execute compute queues and
it has 32 queue slots.
This caused the oversubscription limit to be calculated incorrectly,
leading to a missing chained runlist command at the end of an
oversubscr
'if (parent->entries[pt_idx].addr == pt)' is good enough for condition
of need_update, adding previous Felix's optimization patch, then the
whole picture looks better than this one, IIRC.
Regards,
David Zhou
On 2017年07月13日 21:25, Christian König wrote:
From: Christian König
That should reduc
On 14/07/17 06:23 AM, Felix Kuehling wrote:
> On 17-07-13 05:08 PM, Felix Kuehling wrote:
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
>> @@ -78,6 +78,8 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo,
>> struct dma_fence **fe
On 14/07/17 06:08 AM, Felix Kuehling wrote:
> Allows gdb to access contents of user mode mapped BOs. VRAM access
> requires the driver to implement a new callback in ttm_bo_driver.
Thanks for doing this. Looks mostly good, but I still have some comments.
The shortlog prefix should be "drm/ttm:".
42 matches
Mail list logo
