And after it is ready, we can go ahead with moving
> > > it out of staging to drm.
> >
> > Please keep the todo item that it needs to be converted to atomic. And
> > tbh, it's probably faster if you just submit it to dri-devel, assuming you
> > have time to work on i
Was missing license text.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/acp/Makefile | 21
drivers/gpu/drm/amd/amdgpu/Makefile| 21
drivers/gpu/drm/amd/amdkfd/Makefile| 21
driv
These files were missing it before.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 2 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h | 23 ++
drivers/gpu/drm/amd/display/dc/core/dc_debug.c | 22 +
drivers/g
; have time to work on it. For small drivers we tend to be fairly quick in
> getting them into good enough shape.
I have received the driver from Teddy and pushed it to
https://github.com/sudipm-mukherjee/parport/tree/drm_smi for your first
look into it. It is not even building with next-20171
Hi Shawn,
Finally got a chance to take another look at this and realized that my patch is
somewhat wrong and also didn't really give me the info I was hoping for.
Other than the warning do you see any functional problems?
Which laptop is this? This seems very much system-specific.
Harry
On 2
From: Christian König
Otherwise somebody could try to evict it at the same time and try to use
half torn down structures.
commit 2642cf110d08a403f585a051e4cbf45a90b3adea upstream
Signed-off-by: Christian König
Reviewed-and-Tested-by: Michel Dänzer
Reviewed-by: Alex Deucher
Signed-off-by: Ale
I haven't gone to see where it started, but as of late a good number of
pretty nasty deadlock issues have appeared with the kernel. Easy
reproduction recipe on a laptop with i915/amdgpu prime with lockdep enabled:
DRI_PRIME=1 glxinfo
Additionally, some more race conditions exist that I've managed
On Wed, 2017-11-29 at 16:58 -0500, Felix Kuehling wrote:
> You can see the state of the queues in debugfs:
> /sys/kernel/debug/kfd/... You can look at MQDs and HQDs.
thanks. how do I decode the information?
The rptr always stops at pos 60 which looks like this in mqds:
DIQ on device 45a2
000
Change-Id: I59828a9a10652988e22b50d87dd1ec9df8ae7a1d
Signed-off-by: Shaoyun Liu
---
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 228 -
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 20 +--
drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c | 22 ++--
drivers/gpu/drm/amd/amdgpu/nbio_
Change-Id: I29f33ee3b4bbd6737f3426385a9e8452fb528a67
Signed-off-by: Shaoyun Liu
---
drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c| 126 ++
drivers/gpu/drm/amd/amdgpu/soc15_common.h | 34 ++--
2 files changed, 65 insertions(+), 95 deletions(-)
diff --git a/drivers/gp
Ok , I think can make the change . It involved logic change , please
carefully review it .
How to change the SOC15_REG_OFFSET used in the following case :
const struct nbio_hdp_flush_reg nbio_v7_0_hdp_flush_reg = {
.hdp_flush_req_offset = SOC15_REG_OFFSET(NBIO, 0, mmGPU_HDP_FLUSH_RE
Like static const struct amdgpu_gds_reg_offset amdgpu_gds_reg_offset[]
Actually those could be completely nuked instead.
We added those because we feared that the GDS registers similar to the
VMID base registers aren't continuously for older hardware generations.
But somebody just copy & past
Hi Kent,
I just realized when looking at this again, that you could remove the 0x
before %p. I think %p will print 0x itself.
Regards,
Felix
On 2017-11-30 11:49 AM, Kent Russell wrote:
> Just print a pointer instead of casting
>
> Signed-off-by: Kent Russell
> ---
> drivers/gpu/drm/amd/amdk
I think what you are talking is the golden setting registers as such as
golden_settings_gc_9_0[] etc . I can change them as suggested (Use
SOC15_REG_GOLDEN_VALUE and define a structure for the golden register values).
There are some places that requires directly use the register offset
Just print a pointer instead of casting
Signed-off-by: Kent Russell
---
drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c
b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c
index feb76c2..8abaa
On Tue, Nov 28, 2017 at 1:29 AM, Felix Kuehling wrote:
> From: Philip Yang
>
> After debugger is registered, the pqm_destroy_queue fails because is_debug
> is true, the queue should not be removed from process_queue_list since
> the count is not reduced.
>
> Test application calls debugger unregi
On Tue, Nov 28, 2017 at 1:29 AM, Felix Kuehling wrote:
> Use proper powerplay function. This fixes OpenCL initialization
> problems.
>
> Signed-off-by: Felix Kuehling
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 7 +--
> 1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git
On Wed, Sep 27, 2017 at 7:09 AM, Felix Kuehling wrote:
> From: shaoyunl
>
> HWS does not support over-subscription and the scheduler can not internally
> modify the engine. Driver needs to program the correct engine ID.
>
> Fix the queue and engine selection to create queues on alternating SDMA
>
On Fri, Nov 24, 2017 at 11:44 PM, Felix Kuehling wrote:
>
> Hi Oded,
>
> Yong made this patch against our internal KFD branch. I asked him to
> send it to upstream as well, since it applies to code I have already
> upstreamed. I'm going to do this for more patches in the future as more
> KFD code
Am 28.11.2017 um 16:32 schrieb Andrey Grodzovsky:
The test is as following:
1) Create context A & B
2) Send a command submission using context A which fires up a compute shader.
3) The shader wait a bit and then write a value to a memory location.
4) Send a command submission using context B whi
Signed-off-by: Tom St Denis
---
src/lib/ring_decode.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/lib/ring_decode.c b/src/lib/ring_decode.c
index dee1502cc788..c904df0e281c 100644
--- a/src/lib/ring_decode.c
+++ b/src/lib/ring_decode.c
@@ -395,7 +395,7 @@ static void
I would rather go with something like this:
#define SOC15_REG_GOLDEN_VALUE(ip, inst, reg, value, mask) { ip##_HWIP,
inst, reg##_BASE_IDX, reg, value, mask }
And please define a structure for the golden register values instead of
just using an u32 array.
Regards,
Christian.
Am 29.11.2017 um
The reason to use define here is the original function didn't take the adev as
parameter.
Yeah, I've seen that and this is exactly the reason why I want to avoid
using a macro here :)
If you don't want to change every caller then at least make the macro
name capital, so that we can see in the
23 matches
Mail list logo
