Hi Felix,
I think you might want to add a boundary check to limit the ptrace
access inside the BO.
Otherwise there may be a potential security leak allowing others to
access the whole memory by using ptrace system call.
-Alex Bin
On 2017-07-12 01:37 AM, Felix Kuehling wrote:
Any comments
On 2017年07月12日 12:56, Felix Kuehling wrote:
Hi David,
Responses inline ...
Yeah, I see your mean, you mainly want to save some cpu overhead. If
others have no object, feel free add my RB then.
Regards,
David Zhou
On 17-07-12 12:30 AM, zhoucm1 wrote:
nice improvement, just some nitpick in
Any comments?
I believe this is a nice stability improvement. In case of VM faults
they don't take down the whole GPU with an interrupt storm. With KFD we
can recover without a GPU reset in many cases just by unmapping the
offending process' queues.
Regards,
Felix
On 17-07-03 05:11 PM, Felix
Any comments on this one?
This was requested by the HSA runtime team a long time ago as a
debugging feature. It allows gdb to access the content of CPU-mapped
BOs. I imagine this may be useful for user mode driver developers.
Does anyone dare to give me a Reviewed-by?
Regards,
Felix
On 17-07
From: Amber Lin
Set the system bit for foreign BO mappings and use the remote VRAM
BAR address as the VRAM base offset.
Signed-off-by: Amber Lin
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 17 +
1 file changed, 13 insertions(+), 4 deletions(-)
dif
From: Christian König
Signed-off-by: Christian König
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 44ec11d..aef0
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/drm_prime.c | 25 +
include/drm/drmP.h | 2 ++
2 files changed, 27 insertions(+)
diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c
index 25aa455..b1f8445 100644
--- a/drivers/gpu/drm/drm_prim
From: Christian König
We should be able to handle BOs from other instances as well.
v2:
* Add a module option that is off-by-default
* Use new DRM helper function to check the exporting driver
Signed-off-by: Christian König
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu.
From: Christian König
Pinning them in other devices VRAM would obviously not work.
v2: Add checks to DC code paths
Signed-off-by: Christian König
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu_display.c | 6 ++
drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
From: Christian König
This allows us to have multiple GEM objects for one BO.
Signed-off-by: Christian König
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h| 12 +++--
drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c| 41 +++---
drivers/gpu/d
This patch series adds experimental P2P buffer sharing in amdgpu. It's
disabled by default and can be enabled with amdgpu.p2p_sharing=1.
v2:
* Changed drm helper function to cast to GEM object
* Added foreign BO checks to DC code paths
* Updated commit message for amdgpu_cs change
Amber Lin (1):
On 17-07-03 10:55 PM, Alex Deucher wrote:
> On Mon, Jul 3, 2017 at 5:11 PM, Felix Kuehling wrote:
>> From: Christian König
>>
>> They don't support VM mode yet.
>>
>> Signed-off-by: Christian König
>> Reviewed-by: Felix Kuehling
> This could probably be refined since newer asics support VM for
Hi David,
Responses inline ...
On 17-07-12 12:30 AM, zhoucm1 wrote:
> nice improvement, just some nitpick inline, otherwise the patch is
> Reviewed-by: Chunming Zhou .
>
> On 2017年07月12日 11:56, Felix Kuehling wrote:
>> When lots of virtual address spaces is used, there can be thousands
>> of page
Change-Id: I282e1212346ed76470cadf39153109ab135f32f3
Signed-off-by: Chunming Zhou
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 13 +++--
1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_object
nice improvement, just some nitpick inline, otherwise the patch is
Reviewed-by: Chunming Zhou .
On 2017年07月12日 11:56, Felix Kuehling wrote:
When lots of virtual address spaces is used, there can be thousands
of page table BOs. amdgpu_vm_update_level iterates over all of them
recursively. In man
When lots of virtual address spaces is used, there can be thousands
of page table BOs. amdgpu_vm_update_level iterates over all of them
recursively. In many cases only a few or none at all need to be
updated. Minimize unnecessary code execution and memory usage in
those cases.
This speeds up memor
On 2017年07月12日 03:30, Alex Deucher wrote:
On Mon, Jul 10, 2017 at 5:15 PM, Alex Deucher wrote:
The information has moved to different tables, notably
smu_info for core refclk and umc_info for mem refclk.
Signed-off-by: Alex Deucher
ping? this fixes looking up the reference clocks on vega1
Reviewed-by: Chunming Zhou
On 2017年07月12日 07:00, Alex Deucher wrote:
Maybe a leftover from bringup?
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/soc15.c | 7 ---
1 file changed, 7 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c
b/drivers/gpu/drm/amd/amdgpu/
Maybe a leftover from bringup?
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/soc15.c | 7 ---
1 file changed, 7 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c
b/drivers/gpu/drm/amd/amdgpu/soc15.c
index dd58de9..9dff294 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc1
Hello Ben Goz,
The patch bcea30817574: "drm/amdkfd: Add SDMA user-mode queues
support to QCM" from Jan 3, 2015, leads to the following static
checker warning:
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c:191
create_queue_nocpsch()
error: uninitialized symbol 'retval'.
d
This is just future proofing code, not something that can be triggered
in real life. We're testing to make sure we don't shift wrap when we
do "1ull << i" so "i" has to be in the 0-63 range. If it's 64 then we
have gone too far.
Signed-off-by: Dan Carpenter
diff --git a/drivers/gpu/drm/amd/amd
On Tue, Jul 11, 2017 at 12:22 AM, Daniel Vetter wrote:
> On Mon, Jul 10, 2017 at 02:09:42PM -0700, Jason Ekstrand wrote:
> > On Mon, Jul 10, 2017 at 9:15 AM, Christian König <
> deathsim...@vodafone.de>
> > wrote:
> >
> > > Am 10.07.2017 um 17:52 schrieb Jason Ekstrand:
> > >
> > > On Mon, Jul 10
On Mon, Jul 10, 2017 at 5:15 PM, Alex Deucher wrote:
> The information has moved to different tables, notably
> smu_info for core refclk and umc_info for mem refclk.
>
> Signed-off-by: Alex Deucher
ping? this fixes looking up the reference clocks on vega10 and raven.
Alex
> ---
> drivers/gp
On Tue, Jul 11, 2017 at 05:29:38PM +0900, Michel Dänzer wrote:
> On 08/07/17 07:50 AM, John Brooks wrote:
> > On Thu, Jul 06, 2017 at 07:51:25PM +0900, Michel Dänzer wrote:
> >> From: Michel Dänzer
> >>
> >> This is a rearrangement of some patches of John's, with some improvements
> >> of mine, an
On Tue, Jul 11, 2017 at 3:07 PM, kbuild test robot wrote:
> make ARCH=i386
Yeah, either this code shouldn't have been built on 32-bit arch at
all, or be portable.
>arch/x86/pci/fixup.c: In function 'pci_amd_enable_64bit_bar':
>>> arch/x86/pci/fixup.c:674:15: warning: large integer i
On Tue, Jul 11, 2017 at 11:20 AM, Dave Airlie wrote:
> On 11 July 2017 at 18:36, Christian König wrote:
>> Am 11.07.2017 um 08:49 schrieb Dave Airlie:
>>>
>>> On 7 July 2017 at 19:07, Christian König wrote:
Hi Dave,
on first glance that looks rather good to me, but there is o
Hey Alex,
Which apps are having perf problems?
Also, is the issue present when the queue gets mapped to pipe0? (i.e. it
the perf regression only happens on pipe1+)
Regards,
Andres
On 2017-07-11 11:13 AM, Alex Deucher wrote:
Spreading them causes performance regressions using compute
queues.
On Tue, Jul 11, 2017 at 5:17 AM, Michel Dänzer wrote:
> 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
> Signed-off-by: Michel Dänzer
Reviewed-by: Alex Deucher
Alex
> ---
> src/drmmode_
On Tue, Jul 11, 2017 at 12:17 AM, Christian König
wrote:
> Am 11.07.2017 um 04:36 schrieb Michel Dänzer:
>
>> On 11/07/17 06:09 AM, Jason Ekstrand wrote:
>>
>>> On Mon, Jul 10, 2017 at 9:15 AM, Christian König
>>> mailto:deathsim...@vodafone.de>> wrote:
>>>
>>> Am 10.07.2017 um 17:52 schrieb
On Tue, Jul 11, 2017 at 5:17 AM, Michel Dänzer wrote:
> From: Michel Dänzer
>
> Preparation for following change, no functional change intended.
>
> Signed-off-by: Michel Dänzer
Reviewed-by: Alex Deucher
> ---
> src/radeon_dri2.c| 5 +++--
> src/radeon_kms.c | 4 ++--
> src/radeon_p
Spreading them causes performance regressions using compute
queues.
Cc: Jim Qu
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
b/drivers/gpu/drm/amd/amdgpu/amdg
-resource-type-mask-v2/20170711-104904
base: https://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git next
reproduce: make htmldocs
All warnings (new ones prefixed by >>):
WARNING: convert(1) not found, for SVG to PDF conversion install ImageMagick
(https://www.imagemagick.org)
-define-for-the-PCI-resource-type-mask-v2/20170711-104904
base: https://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git next
config: i386-allmodconfig (attached as .config)
compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901
reproduce:
# save the attached .config to linux build tree
That's irrelevant now, Dave Airlied noted in another thread that radv is
already using this.
So we can't remove it anyway because that would break Mesa as well.
Christian.
Am 11.07.2017 um 11:30 schrieb Mao, David:
This is not true.
Can we keep this for a while until we have sync object in pl
From my side, it should be like:
VRAM: 249/4096 vis 50/237 (MiB)
steam:( 3203) : 63304 KiB VRAM, 14088 KiB vis
VRAM, 20072 KiB GTT
compiz:( 2510): 82144 KiB VRAM, 7072 KiB vis
VRAM, 20460 KiB GTT
Xorg:( 1682)
Hi Qiang,
Apologies I booted a 4.9 kernel I was testing and I guess it doesn't
have the ioctl you're using to read visvram.
Booted into 4.11 it's working as expected.
I'll add my Rb and push it momentarily.
Cheers,
Tom
On 11/07/17 07:30 AM, Yu, Qiang wrote:
From my side, it should be lik
On 11 July 2017 at 10:15, Christian König wrote:
> From: Christian König
>
> This reverts commit 6b79c66b841dded6ffa6b56f14e4eb10a90a7c07
> and commit 6afadeaf13279fcdbc48999f522e1dc90a9dfdaf.
>
If/when you're going ahead with this please squash a revert for
25712f1d35f6f64167ede45d3dc72a410f367c
Trying out the patch I see the following in the "vram" list
VRAM: 1096/2048 vis 195/238 (MiB)
xonotic-glx:( 4364) : 965612 KiB VRAM, 0 KiB
vis VRAM, 5064 KiB GTT
umr:( 4315) : 0 KiB VRAM, 0 KiB
vis VRAM, 0 KiB
Hi Christian,
When will sync object support landed in upstream kernel, which version in
specific?
We still rely on legacy semaphore implementation and we have to use it if sync
object still takes time.
Thanks.
Best Regards,
David
> On 11 Jul 2017, at 5:15 PM, Christian König wrote:
>
> From:
On 6 July 2017 at 13:46, Deucher, Alexander wrote:
>> Attach it to analogous primitive?
>
> Radeon libdrm is much different than amdgpu. There is no analog.
>
Upon a closer look, indeed there isn't. Must have gotten confused earlier.
>>
>> > I think the current radeon API is simpler. Maybe a fo
Am 11.07.2017 um 11:20 schrieb Dave Airlie:
On 11 July 2017 at 18:36, Christian König wrote:
Am 11.07.2017 um 08:49 schrieb Dave Airlie:
On 7 July 2017 at 19:07, Christian König wrote:
Hi Dave,
on first glance that looks rather good to me, but there is one things I
don't really like and I s
This is not true.
Can we keep this for a while until we have sync object in place?
Thanks.
Best Regards,
David
> On 11 Jul 2017, at 5:28 PM, Christian König wrote:
>
> I hoped that Dave Airlied will land it together with this patch.
>
> As far as I know the closed source driver already doesn't
I hoped that Dave Airlied will land it together with this patch.
As far as I know the closed source driver already doesn't use that any
more either.
Regards,
Christian.
Am 11.07.2017 um 11:20 schrieb Mao, David:
Hi Christian,
When will sync object support landed in upstream kernel, which ver
On 11 July 2017 at 18:36, Christian König wrote:
> Am 11.07.2017 um 08:49 schrieb Dave Airlie:
>>
>> On 7 July 2017 at 19:07, Christian König wrote:
>>>
>>> Hi Dave,
>>>
>>> on first glance that looks rather good to me, but there is one things I
>>> don't really like and I strongly think Marek wi
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
Signed-off-by: Michel Dänzer
---
src/drmmode_display.c | 12 +++-
src/drmmode_display.h | 2 ++
src/radeon_kms.c | 21 +
From: Michel Dänzer
Preparation for following change, no functional change intended.
Signed-off-by: Michel Dänzer
---
src/radeon_dri2.c| 5 +++--
src/radeon_kms.c | 4 ++--
src/radeon_present.c | 2 +-
3 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/src/radeon_dri2.c b/s
From: Christian König
This reverts commit 6b79c66b841dded6ffa6b56f14e4eb10a90a7c07
and commit 6afadeaf13279fcdbc48999f522e1dc90a9dfdaf.
Semaphore support was never used by any open source project and
not even widely by any closed source driver.
This should be replaced by sync object support.
S
Am 11.07.2017 um 08:49 schrieb Dave Airlie:
On 7 July 2017 at 19:07, Christian König wrote:
Hi Dave,
on first glance that looks rather good to me, but there is one things I
don't really like and I strongly think Marek will absolutely agree on that:
When we add a new CS function then let's get
On 08/07/17 07:50 AM, John Brooks wrote:
> On Thu, Jul 06, 2017 at 07:51:25PM +0900, Michel Dänzer wrote:
>> From: Michel Dänzer
>>
>> This is a rearrangement of some patches of John's, with some improvements
>> of mine, and a patch of mine, rebased on current amd-staging-4.11.
>> Details about ch
On Mon, Jul 10, 2017 at 02:09:42PM -0700, Jason Ekstrand wrote:
> On Mon, Jul 10, 2017 at 9:15 AM, Christian König
> wrote:
>
> > Am 10.07.2017 um 17:52 schrieb Jason Ekstrand:
> >
> > On Mon, Jul 10, 2017 at 8:45 AM, Christian König
> > wrote:
> >
> >> Am 10.07.2017 um 17:28 schrieb Jason Ekstr
Am 11.07.2017 um 04:36 schrieb Michel Dänzer:
On 11/07/17 06:09 AM, Jason Ekstrand wrote:
On Mon, Jul 10, 2017 at 9:15 AM, Christian König
mailto:deathsim...@vodafone.de>> wrote:
Am 10.07.2017 um 17:52 schrieb Jason Ekstrand:
On Mon, Jul 10, 2017 at 8:45 AM, Christian König
mail
51 matches
Mail list logo