[Mesa-dev] [PATCH 1.5/4] [RFC] ac/addrlib: relax an assertion

Marek Olšák maraeo at gmail.com
Tue Sep 12 23:47:32 UTC 2017


Reviewed-by: Marek Olšák <marek.olsak at amd.com>

Marek

On Mon, Sep 11, 2017 at 3:26 PM, Nicolai Hähnle <nhaehnle at gmail.com> wrote:
> From: Nicolai Hähnle <nicolai.haehnle at amd.com>
>
> ---
> We hit this assertion with 3D textures on gfx9.
>
> I'm not aware of any 3D-texture-specific failures, but I'm also not sure
> whether CMASK is supposed to work with 3D textures or whether we've just
> been lucky.
>
> ---
>  src/amd/addrlib/gfx9/gfx9addrlib.cpp | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/src/amd/addrlib/gfx9/gfx9addrlib.cpp b/src/amd/addrlib/gfx9/gfx9addrlib.cpp
> index 57ecb058727..edb4c6e636a 100644
> --- a/src/amd/addrlib/gfx9/gfx9addrlib.cpp
> +++ b/src/amd/addrlib/gfx9/gfx9addrlib.cpp
> @@ -261,21 +261,22 @@ ADDR_E_RETURNCODE Gfx9Lib::HwlComputeHtileInfo(
>  *
>  *   @return
>  *       ADDR_E_RETURNCODE
>  ************************************************************************************************************************
>  */
>  ADDR_E_RETURNCODE Gfx9Lib::HwlComputeCmaskInfo(
>      const ADDR2_COMPUTE_CMASK_INFO_INPUT*    pIn,    ///< [in] input structure
>      ADDR2_COMPUTE_CMASK_INFO_OUTPUT*         pOut    ///< [out] output structure
>      ) const
>  {
> -    ADDR_ASSERT(pIn->resourceType == ADDR_RSRC_TEX_2D);
> +// TODO: Clarify with AddrLib team
> +//     ADDR_ASSERT(pIn->resourceType == ADDR_RSRC_TEX_2D);
>
>      UINT_32 numPipeTotal = GetPipeNumForMetaAddressing(pIn->cMaskFlags.pipeAligned,
>                                                         pIn->swizzleMode);
>
>      UINT_32 numRbTotal = pIn->cMaskFlags.rbAligned ? m_se * m_rbPerSe : 1;
>
>      UINT_32 numCompressBlkPerMetaBlkLog2, numCompressBlkPerMetaBlk;
>
>      if ((numPipeTotal == 1) && (numRbTotal == 1))
>      {
> --
> 2.11.0
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev


More information about the mesa-dev mailing list