[Mesa-dev] [PATCH v2 01/11] i965: Introduce new SHADER_OPCODE_URB_WRITE_SIMD8_MASKED/PER_SLOT opcodes.

Kenneth Graunke kenneth at whitecape.org
Wed Oct 21 01:29:21 PDT 2015


On Monday, October 12, 2015 02:49:03 PM Kenneth Graunke wrote:
> In the vec4 backend, we have a vec4_instruction::urb_write_flags field.
> There are many kinds of flags for SIMD4x2 messages.
> 
> However, there are really only two (per-slot offset, use channel masks)
> for SIMD8 messages.  Rather than adding a boolean flag for per-slot
> offsets (polluting all instructions), I decided to just make three new
> opcodes.
> 
> Signed-off-by: Kenneth Graunke <kenneth at whitecape.org>
> ---
>  src/mesa/drivers/dri/i965/brw_defines.h        |  3 +++
>  src/mesa/drivers/dri/i965/brw_fs.cpp           |  9 +++++++++
>  src/mesa/drivers/dri/i965/brw_fs_generator.cpp | 11 +++++++++++
>  src/mesa/drivers/dri/i965/brw_inst.h           |  1 +
>  src/mesa/drivers/dri/i965/brw_shader.cpp       |  9 +++++++++
>  5 files changed, 33 insertions(+)
> 
> Here's the rest of the series that didn't get reviewed last time,
> rebased on Jason's compiler reworks.

Jason landed yet more compiler reworks.  I've pushed a rebased copy
to the 'simd8gs' branch of ~kwg/mesa.  Code got shuffled between
functions or header files, so some of it didn't textually apply, but
the new code isn't significantly different.  I've verified that it
still builds and passes Piglit.

Jason has yet *more* compiler reworks on the mailing list.  I've
preemptively rebased on those and pushed that to my tree as well.
It's the 'simd8gs-rebase-rebase' branch.  That branch doesn't
compile, however - with the roundabout vec4/fs include hell, it's
somehow getting an incomplete type for "struct brw_gs_compile".
I didn't spend the time to figure out why.  Other work to do.

--Ken
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.freedesktop.org/archives/mesa-dev/attachments/20151021/84109dba/attachment.sig>


More information about the mesa-dev mailing list