[RFC 0/7] UVD support for SI in amdgpu

Thu Nov 9 10:09:42 UTC 2017

Am 09.11.2017 um 10:54 schrieb Piotr Redlewski:
> On Wed, Nov 08, 2017 at 06:54:18PM -0500, Alex Deucher wrote:
>> On Wed, Nov 8, 2017 at 5:38 PM, Piotr Redlewski <predlewski at gmail.com> wrote:
>>> Hi,
>>>
>>> Following series implements UVD support for SI in amdgpu driver. Code is based
>>> on CIK's UVD support in amdgpu and SI's UVD support in radeon drivers. To work,
>>> it requires tahiti uvd firmware with added header - I've created simple script
>>> to produce exactly this, so if anyone is interested it can be found here:
>>> https://gist.github.com/anonymous/6d974a970340f7f64b6fcc4f95267e43
>>>
>>> Code is based on amd-staging-drm-next branch in Alex's tree. After applying
>>> these patches, uvd boots up and seems to work ok. I've tested it with vdpauinfo
>>> and mpv.
>>>
>>> Some comments/issues for the patches:
>>> 1. To make uvd work, I had to bring back fb location programming. Using location
>>> programmed by vbios, vram location is not available for uvd mc (at least on my
>>> machine) due to too wide address. Starting address is 40-bit long for fb, but
>>> uvd mc supports only 32-bits (judging by comments in amdgpu code and actual code
>>> in radeon driver)
>> Something else must be going on.  The vram location is irrelevant with
>> respect to the limitations of UVD.  I think the limitations with UVD
>> are more to do with the location of the active buffers relative to
>> each other rather than the absolute location of some aperture in the
>> GPU's address space.  CI has the same limitation as I recall so there
>> is probably a bug somewhere.  Windows has used the fb location as set
>> by the vbios since evergreen, so it definitely should work.
>>
> If this is the case, then there must be something missing in UVD mc controller
> programming. When using vbios, I get following location:
> amdgpu 0000:01:00.0: VRAM: 2048M 0x000000F400000000 - 0x000000F47FFFFFFF (2048M used)
>
> When UVD bo is created, it starts at address 0xf400243000 and this value is used
> for programming UVD mc offsets. Programming is done in the following way:
> addr = (adev->uvd.gpu_addr + AMDGPU_UVD_FIRMWARE_OFFSET) >> 3;
> WREG32(mmUVD_VCPU_CACHE_OFFSET0, addr);
>
> Because address of the bo is wider than 32-bit, this won't work. It would be the
> same if UVD bo would be created at the beginning of the VRAM.
>
> Any ideas how to handle this?

Are you programming UVD_LMI_EXT40_ADDR?

But I'm not sure if we ever handled that correctly in the SI code.

Regards,
Christian.

>
>>> 2. I don't know why, but I couldn't get the uvd to boot without setting uvd mc
>>> offsets before starting other engines. Because of that I set it in .sw_init
>>> function. In my opinion this should be fixed as it generally doesn't follow
>>> amdgpu driver architecture (hardware setup during software setup stage) and
>>> probably will break suspending and resume (I didn't test it). As I mentioned,
>>> I couldn't figure out why this is happening, so I count on help with finding fix
>>> for this.
>>> 3. I found some redefinitions in include/asic_reg/uvd/uvd_4_0_sh_mask.h. I guess
>>> this file is generated, so fix should be made wherever it is generated from. For
>>> now I removed offending lines just to silence the compiler warnings.
>>> 4. I'm not sure whether I choose the right version for the uvd. Existing code in
>>> si.c suggested that it should be 3.1, however I went with the 4.0, because for
>>> this version there are available new style headers.
>> I think the regs are pretty much the same between 3.x and 4.x so it
>> should be fine.
> Great.
>
> Regards,
> Piotr
>> Alex
>>
>>
>>> Regards,
>>> Piotr
>>>
>>> Piotr Redlewski (7):
>>>    drm/amdgpu: remove duplicated definitions of some of the SI registers
>>>    drm/amdgpu/uvd4: fix some register's mask and shift definitions
>>>    drm/amdgpu/gmc6: don't use vram location programmed by the vbios
>>>    drm/amdgpu/uvd4: add early init stage functions for uvd 4.0
>>>    drm/amdgpu/uvd4: add sw init and fini stages' functions for uvd 4.0
>>>    drm/amdgpu/uvd4: add hardware specific functions for uvd 4.0
>>>    drm/amdgpu: enable UVD for SI
>>>
>>>   drivers/gpu/drm/amd/amdgpu/Makefile                |   3 +-
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h           |   6 +
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c            |  14 +
>>>   drivers/gpu/drm/amd/amdgpu/dce_v6_0.c              | 114 ++-
>>>   drivers/gpu/drm/amd/amdgpu/dce_v6_0.h              |   5 +
>>>   drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c              |   7 -
>>>   drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c              |  40 +-
>>>   drivers/gpu/drm/amd/amdgpu/si.c                    | 256 ++++++-
>>>   drivers/gpu/drm/amd/amdgpu/si_ih.c                 |   3 +
>>>   drivers/gpu/drm/amd/amdgpu/sid.h                   |  52 +-
>>>   drivers/gpu/drm/amd/amdgpu/uvd_v4_0.c              | 810 +++++++++++++++++++++
>>>   drivers/gpu/drm/amd/amdgpu/uvd_v4_0.h              |  29 +
>>>   .../drm/amd/include/asic_reg/uvd/uvd_4_0_sh_mask.h |   2 -
>>>   13 files changed, 1273 insertions(+), 68 deletions(-)
>>>   create mode 100644 drivers/gpu/drm/amd/amdgpu/uvd_v4_0.c
>>>   create mode 100644 drivers/gpu/drm/amd/amdgpu/uvd_v4_0.h
>>>
>>> --
>>> 2.15.0
>>>
>>> _______________________________________________
>>> amd-gfx mailing list
>>> amd-gfx at lists.freedesktop.org
>>> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
>> _______________________________________________
>> amd-gfx mailing list
>> amd-gfx at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
> _______________________________________________
> amd-gfx mailing list
> amd-gfx at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/amd-gfx