Am 31.07.2018 um 07:39 schrieb Junwei Zhang:
Signed-off-by: Junwei Zhang
A commit message would be nice to have.
Apart from that the patch is Reviewed-by: Christian König
Christian.
---
tests/amdgpu/bo_tests.c | 32
1 file changed, 32 insertions(+)
d
On 07/30/2018 06:47 PM, Christian König wrote:
Am 30.07.2018 um 12:02 schrieb Junwei Zhang:
From: Chunming Zhou
v2: get original gem handle from gobj
v3: update find bo data structure as union(in, out)
simply some code logic
Do we now have an open source user for this, so that we can up
Am 31.07.2018 um 03:40 schrieb Zhou, David(ChunMing):
Thanks for Jerry still remembers this series.
Hi Christian,
For upstream of this feature, seems we already had agreement long time ago. Two
reasons for upstreaming:
1. this bug was found by an opengl game, so this bug also is in mesa dri
Am 30.07.2018 um 22:14 schrieb Alex Deucher:
On Mon, Jul 30, 2018 at 5:55 AM, Michel Dänzer wrote:
On 2018-07-24 10:53 PM, Alex Deucher wrote:
On Mon, Jul 23, 2018 at 12:32 PM, Gustavo A. R. Silva
wrote:
idx can be indirectly controlled by user-space, hence leading to a
potential exploitatio
Am 31.07.2018 um 08:11 schrieb Thomas Zimmermann:
The function ttm_bo_get acquires a reference on a TTM buffer object. The
function's name is more aligned to the Linux kernel convention of naming
ref-counting function _get and _put.
Signed-off-by: Thomas Zimmermann
amdgpu patches should actua
The function ttm_bo_put releases a reference to a TTM buffer object. The
function's name is more aligned to the Linux kernel convention of naming
ref-counting function _get and _put.
A call to ttm_bo_unref takes the address of the TTM BO object's pointer and
clears the pointer's value to NULL. Thi
The function ttm_bo_get acquires a reference on a TTM buffer object. The
function's name is more aligned to the Linux kernel convention of naming
ref-counting function _get and _put.
Signed-off-by: Thomas Zimmermann
---
drivers/gpu/drm/radeon/radeon_object.c | 2 +-
1 file changed, 1 insertion(+
The function ttm_bo_put releases a reference to a TTM buffer object. The
function's name is more aligned to the Linux kernel convention of naming
ref-counting function _get and _put.
A call to ttm_bo_unref takes the address of the TTM BO object's pointer and
clears the pointer's value to NULL. Thi
The function ttm_bo_get acquires a reference on a TTM buffer object. The
function's name is more aligned to the Linux kernel convention of naming
ref-counting function _get and _put.
Signed-off-by: Thomas Zimmermann
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 2 +-
1 file changed, 1 inserti
On Mon, Jul 30, 2018 at 04:51:57PM +0200, Christian König wrote:
> Add helpers to iterate over all entries in a bo_list.
>
> Signed-off-by: Christian König
Nice wrapper.
Acked-by: Huang Rui
Thanks,
Ray
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 21 ++--
> drivers/gpu/drm/
Signed-off-by: Junwei Zhang
---
tests/amdgpu/bo_tests.c | 32
1 file changed, 32 insertions(+)
diff --git a/tests/amdgpu/bo_tests.c b/tests/amdgpu/bo_tests.c
index 9d4da4a..4d05027 100644
--- a/tests/amdgpu/bo_tests.c
+++ b/tests/amdgpu/bo_tests.c
@@ -27,6 +27,7
On Mon, Jul 30, 2018 at 04:51:56PM +0200, Christian König wrote:
> The RCU grace period is harmless and avoiding it is not worth the effort
> of doubling the implementation.
>
> Signed-off-by: Christian König
Reviewed-by: Huang Rui
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 13 +---
On Mon, Jul 30, 2018 at 04:51:55PM +0200, Christian König wrote:
> When changing a list always completely recreate it. This avoids locking
> in the hot path because the list always stays like it is until it is
> unreferenced.
The fpriv->bo_list_handles is allocated by OP_CREATE, so here we just
re
On Mon, Jul 30, 2018 at 04:51:54PM +0200, Christian König wrote:
> Further demangle amdgpu.h
I also found amdgpu.h is mingled with many things. Moving them to
sub-header is necessary.
Acked-by: Huang Rui
>
> Signed-off-by: Christian König
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu.h |
On Mon, Jul 30, 2018 at 04:51:52PM +0200, Christian König wrote:
> Return -EINVAL when both the BOs as well as a list handle is provided in
> the IOCTL.
>
> Signed-off-by: Christian König
Reviewed-by: Huang Rui
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 9 ++---
> 1 file changed, 6
On 07/31/2018 09:49 AM, Zhou, David(ChunMing) wrote:
Typo, excepted -> expected
-Original Message-
From: amd-gfx On Behalf Of Zhou,
David(ChunMing)
Sent: Tuesday, July 31, 2018 9:41 AM
To: Koenig, Christian ; Zhang, Jerry
; amd-gfx@lists.freedesktop.org
Subject: RE: [PATCH 1/2] drm/am
On Mon, Jul 30, 2018 at 04:51:51PM +0200, Christian König wrote:
> long might only be 32bit in size and we can easily use more than 4GB
> here.
>
> Signed-off-by: Christian König
Acked-by: Huang Rui
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 2 +-
> 1 file changed, 1 insertion(+),
Typo, excepted -> expected
-Original Message-
From: amd-gfx On Behalf Of Zhou,
David(ChunMing)
Sent: Tuesday, July 31, 2018 9:41 AM
To: Koenig, Christian ; Zhang, Jerry
; amd-gfx@lists.freedesktop.org
Subject: RE: [PATCH 1/2] drm/amdgpu: return bo itself if userptr is cpu addr of
bo (v
Thanks for Jerry still remembers this series.
Hi Christian,
For upstream of this feature, seems we already had agreement long time ago. Two
reasons for upstreaming:
1. this bug was found by an opengl game, so this bug also is in mesa driver in
theory.
2. after upstream these patches, we can
Series is Reviewed-by: Chunming Zhou
-Original Message-
From: amd-gfx On Behalf Of Christian
K?nig
Sent: Monday, July 30, 2018 10:52 PM
To: amd-gfx@lists.freedesktop.org
Subject: [PATCH 9/9] drm/amdgpu: create an empty bo_list if no handle is
provided
Instead of having extra handling
On Mon, Jul 30, 2018 at 5:55 AM, Michel Dänzer wrote:
> On 2018-07-24 10:53 PM, Alex Deucher wrote:
>> On Mon, Jul 23, 2018 at 12:32 PM, Gustavo A. R. Silva
>> wrote:
>>> idx can be indirectly controlled by user-space, hence leading to a
>>> potential exploitation of the Spectre variant 1 vulnera
Sorry, I just saw your discussion with Evan. Checking the conditions in
the worker function should work. Feel free to add my Reviewed-by.
Regards,
Felix
On 2018-07-30 03:33 PM, Felix Kuehling wrote:
> NACK.
>
> See inline ...
>
>
> On 2018-07-30 05:23 AM, Rex Zhu wrote:
>> delay to enable gfx
NACK.
See inline ...
On 2018-07-30 05:23 AM, Rex Zhu wrote:
> delay to enable gfx off feature to avoid gfx on/off frequently
> suggested by Alex and Evan.
>
> Signed-off-by: Rex Zhu
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu.h| 2 ++
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 15 ++
[+Oded]
KFD on amd-staging-drm-next doesn't support Raven yet. It's coming soon
via Oded's branch.
To prepare for that, all MMIO access in amdgpu_amdkfd_gfx_v9.c should be
protected by disabling GFXOFF. Since enabling GFXOFF is delayed in your
latest patch series, it shouldn't cause performance i
I'm using that branch myself successfully. For Raven (and other APU)
support, make sure you have AMD_IOMMU_V2 enabled in your kernel config.
Regards,
Felix
On 2018-07-30 09:33 AM, Paul Menzel wrote:
> Dear Felix, dear Yong,
>
>
> On 07/13/18 22:17, Felix Kühling wrote:
>> From: Yong Zhao
>>
>
On Fri, Jul 27, 2018 at 12:04 PM, Michel Dänzer wrote:
> From: Michel Dänzer
>
> We were only storing the FB provided by the client, but on CRTCs with
> TearFree enabled, we use a separate FB. This could cause
> drmmode_flip_handler to fail to clear drmmode_crtc->flip_pending, which
> could resul
Dear Alex,
On 07/30/18 17:32, Alex Deucher wrote:
> On Mon, Jul 30, 2018 at 11:25 AM, Paul Menzel wrote:
>> On 07/30/18 15:53, Oded Gabbay wrote:
>>> Do you have IOMMUv2 enabled on this chipset?
>>
>> No, I did not. With `AMD_IOMMU_V2` selected, the device is found.
>>
>> ```
>> $ grep AMD_IOMMU
On Mon, Jul 30, 2018 at 11:25 AM, Paul Menzel
wrote:
> Dear Oded,
>
>
> On 07/30/18 15:53, Oded Gabbay wrote:
>> Do you have IOMMUv2 enabled on this chipset?
>
> No, I did not. With `AMD_IOMMU_V2` selected, the device is found.
>
> ```
> $ grep AMD_IOMMU /boot/config-4.18.0-rc3-01368-gb5aa3f4aef72
Dear Oded,
On 07/30/18 15:53, Oded Gabbay wrote:
> Do you have IOMMUv2 enabled on this chipset?
No, I did not. With `AMD_IOMMU_V2` selected, the device is found.
```
$ grep AMD_IOMMU /boot/config-4.18.0-rc3-01368-gb5aa3f4aef72
CONFIG_AMD_IOMMU=y
CONFIG_AMD_IOMMU_V2=m
$ sudo dmesg | grep -i kfd
This avoids multiple allocations for the head and the array.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 114 +++-
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h | 17 +++--
2 files changed, 57 insertions(+), 74 deletions(-)
diff --git
Add helpers to iterate over all entries in a bo_list.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 21 ++--
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h | 10 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 51 +
3 files
Return -EINVAL when both the BOs as well as a list handle is provided in
the IOCTL.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 9 ++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
b/drivers/gpu/drm/amd
This allows us to trace all VM ranges which should be valid inside a CS.
v2: dump mappings without BO as well
Signed-off-by: Christian König
Reviewed-by: Chunming Zhou (v1)
Reviewed-and-tested-by: Andrey Grodzovsky (v1)
Reviewed-by: Huang Rui (v1)
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
Instead of having extra handling just create an empty bo_list when no
handle is provided.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 111 ++---
1 file changed, 46 insertions(+), 65 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/
Further demangle amdgpu.h
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 40 +
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h | 70 +
2 files changed, 71 insertions(+), 39 deletions(-)
create mode 100644 drivers/gpu/drm/
When changing a list always completely recreate it. This avoids locking
in the hot path because the list always stays like it is until it is
unreferenced.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 23 ---
drivers/gpu/drm/amd/amdgpu/amdgp
The RCU grace period is harmless and avoiding it is not worth the effort
of doubling the implementation.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 13 +
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h | 1 -
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
long might only be 32bit in size and we can easily use more than 4GB
here.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
b/drivers/gpu/drm/amd/amdgpu/
Yes, that patch won't change any code logic, and just a cleanup.
Thanks,
Ray
-Original Message-
From: Koenig, Christian
Sent: Monday, July 30, 2018 9:31 PM
To: Paul Menzel ; Huang, Ray
; Alex Deucher
Cc: amd-gfx@lists.freedesktop.org; dri-de...@lists.freedesktop.org
Subject: Re: Fix br
Do you have IOMMUv2 enabled on this chipset ?
If not, amdkfd won't include raven support.
Oded
On Mon, Jul 30, 2018 at 4:33 PM Paul Menzel
wrote:
>
> Dear Felix, dear Yong,
>
>
> On 07/13/18 22:17, Felix Kühling wrote:
> > From: Yong Zhao
> >
> > Add DID and kfd_device_info for Raven.
> >
> > Si
Dear Felix, dear Yong,
On 07/13/18 22:17, Felix Kühling wrote:
> From: Yong Zhao
>
> Add DID and kfd_device_info for Raven.
>
> Signed-off-by: Yong Zhao
> Reviewed-by: Felix Kuehling
> Signed-off-by: Felix Kuehling
> Acked-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdkfd/kfd_device.c |
Hi Paul,
well that is rather odd, cause the patch you pointed out is just a code
cleanup without functional change.
Can you double check your bisect result again?
Thanks,
Christian.
Am 30.07.2018 um 14:35 schrieb Paul Menzel:
Dear Linux folks,
On 07/04/18 12:43, Christian König wrote:
Am
A bit large, maybe 100ms is enough.
Apart from that the set now looks good on first glance, but I really
don't know that code that well.
Christian.
Am 30.07.2018 um 12:25 schrieb Zhu, Rex:
One problem,
I delay 500ms to enable gfx off feature.
Any concern about the delay time?
Best Reg
Am 30.07.2018 um 12:02 schrieb Junwei Zhang:
From: Chunming Zhou
v2: get original gem handle from gobj
v3: update find bo data structure as union(in, out)
simply some code logic
Do we now have an open source user for this, so that we can upstream it?
One more point below.
Signed-off
Go ahead with my RB.
-Original Message-
From: amd-gfx On Behalf Of Christian
K?nig
Sent: Monday, July 30, 2018 5:19 PM
To: amd-gfx@lists.freedesktop.org
Subject: [PATCH] drm/amdgpu: add new amdgpu_vm_bo_trace_cs() function v2
This allows us to trace all VM ranges which should be valid i
One problem,
I delay 500ms to enable gfx off feature.
Any concern about the delay time?
Best Regards
Rex
From: amd-gfx on behalf of Rex Zhu
Sent: Monday, July 30, 2018 5:23 PM
To: amd-gfx@lists.freedesktop.org
Cc: Zhu, Rex
Subject: [PATCH v2 2/4] drm/amdg
From: Chunming Zhou
(amdgpu_drm.h will be synced from kernel header as general approach, here
provided for review only)
userspace needs to know if the user memory is from BO or malloc.
v2: modify IOCTL data as union(in, out)
update bo table management(get, set)
simplify code logic
Sig
From: Chunming Zhou
v2: get original gem handle from gobj
v3: update find bo data structure as union(in, out)
simply some code logic
Signed-off-by: Chunming Zhou
Signed-off-by: Junwei Zhang (v3)
Reviewed-by: Christian König
Reviewed-by: Jammy Zhou
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h
bump version for new API involved.
Signed-off-by: Junwei Zhang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 8843a06..6b52551 100644
--- a/
On 2018-07-24 10:53 PM, Alex Deucher wrote:
> On Mon, Jul 23, 2018 at 12:32 PM, Gustavo A. R. Silva
> wrote:
>> idx can be indirectly controlled by user-space, hence leading to a
>> potential exploitation of the Spectre variant 1 vulnerability.
>>
>> This issue was detected with the help of Smatch
use amdgpu_gfx_off_ctrl function so driver can arbitrate
whether the gfx ip can be power off or power on.
Signed-off-by: Rex Zhu
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 6 ++
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 6 ++
2 files changed, 4 insertions(+), 8 deletions(-)
dif
this patch is a workaround for the gpu hang
at video end time if play video with gfx off enabled.
Signed-off-by: Rex Zhu
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_
delay to enable gfx off feature to avoid gfx on/off frequently
suggested by Alex and Evan.
Signed-off-by: Rex Zhu
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h| 2 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 15 +++
drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c| 8 ++--
3 f
v2:
1. drop the special handling for the hw IP
suggested by hawking and Christian.
2. refine the variable name suggested by Flora.
This funciton as the entry of gfx off ctrl feature.
we arbitrat gfx off feature enable/disable in this
function.
Signed-off-by: Rex Zhu
---
drivers/gpu/
This allows us to trace all VM ranges which should be valid inside a CS.
v2: dump mappings without BO as well
Signed-off-by: Christian König
Reviewed-by: Chunming Zhou (v1)
Reviewed-and-tested-by: Andrey Grodzovsky (v1)
Reviewed-by: Huang Rui (v1)
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
On Thu, Jul 26, 2018 at 5:16 PM Christian König <
ckoenig.leichtzumer...@gmail.com> wrote:
> Note which task is using the entity and only kill it if the last user of
> the entity is killed. This should prevent problems when entities are
> leaked to
> child processes.
>
> Signed-off-by: Christian K
I have seen arc architectures use a "arcksyms.c" file.
- sed@ -
[1]
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/plain/arch/arc/kernel/arcksyms.c
[2]
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/arc/kernel/Makefile#n11
[3]
https://git.kerne
Hi,
when compiling with clang-7 and CONFIG_DRM_AMDGPU=m I see the following...
if [ "" = "-pg" ]; then if [ arch/x86/boot/compressed/misc.o !=
"scripts/mod/empty.o" ]; then ./scripts/recordmcount
"arch/x86/boot/compressed/misc.o"; fi; fi;
ERROR: "__addsf3" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko]
On Sun, Jul 29, 2018 at 4:39 PM, Christian König
wrote:
>> Do you need further informations?
>
> No, that is a known issue.
>
Hi Christian,
OK, thanks for the confirmation.
Do you have a discussion thread or a (pending) fix where you can point me?
Thanks.
Regards,
- sed@ -
> Regards,
> Chri
59 matches
Mail list logo