projects / platform / upstream / mesa.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 69a3559) | patch
intel/fs: QUAD_SWIZZLE requires packed data
authorJason Ekstrand <jason@jlekstrand.net>
Tue, 27 Oct 2020 05:36:53 +0000 (00:36 -0500)
committerMarge Bot <eric+marge@anholt.net>
Fri, 22 Jan 2021 18:38:37 +0000 (18:38 +0000)
commit58bcb5401d85b4a21f6d9ea4eb7bff8e1ed7110f
treefb34bec38f060ef4c257aaf4e842821d76099fb0tree | snapshot
parent69a3559efd6ec036749c3f460f092b7a38c12d2acommit | diff
intel/fs: QUAD_SWIZZLE requires packed data

We could probably support some strides if we tried hard enough but the
whole point of this opcode is to accelerate things with crazy Align16 or
crazy regions.  It's ok if we have to emit an extra MOV to get a packed
source.

Fixes: 8b4a5e641bc3 "intel/fs: Add support for subgroup quad operations"
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7329>
src/intel/compiler/brw_fs_copy_propagation.cpp diff | blob | history
Domain: Graphics System / GL;