 Marcin Ślusarz
|
432e263284
|
intel/compiler: fine-grained control of dispatch widths
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> [v2]
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20854>
|
2023-01-27 11:00:41 +00:00 |
|
|
Nico Cortes
|
29adbb132f
|
Revert "intel/compiler: fine-grained control of dispatch widths"
This reverts commit bed18ab3e2.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8063
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20654>
|
2023-01-12 00:33:25 +00:00 |
|
|
Marcin Ślusarz
|
bed18ab3e2
|
intel/compiler: fine-grained control of dispatch widths
Reviewed-by: Matt Turner <mattst88@gmail.com> [v1]
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20535>
|
2023-01-11 08:17:12 +00:00 |
|
|
Caio Oliveira
|
6c194ddd18
|
intel/compiler: Prepare SIMD selection helpers to handle different prog_datas
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19601>
|
2022-11-15 04:55:18 +00:00 |
|
|
Caio Oliveira
|
6ffa597bcf
|
intel/compiler: Keep track of compiled/spilled in brw_simd_selection_state
We still update the cs_prog_data, but don't rely on it for this state anymore.
This will allow use the SIMD selector with shaders that don't use cs_prog_data.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19601>
|
2022-11-15 04:55:18 +00:00 |
|
|
Caio Oliveira
|
3c52e2d04c
|
intel/compiler: Add a SIMD_COUNT constant
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19601>
|
2022-11-15 04:55:18 +00:00 |
|
|
Caio Oliveira
|
a0580dadfd
|
intel/compiler: Create a struct to hold SIMD selection state
This is a preparation to decouple the storage of what SIMDs
compiled/spilled from the cs_prog_data. This will allow reuse
of SIMD selection code by Bindless Shaders.
And since we have a struct now, move the error array there so
reduce the boilerplate of the users.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19601>
|
2022-11-15 04:55:18 +00:00 |
|
|
Caio Oliveira
|
8cda6cd774
|
intel/compiler: Simplify usage of brw_simd_select_for_workgroup_size()
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19601>
|
2022-11-15 04:55:18 +00:00 |
|
|
Caio Oliveira
|
a943dbf475
|
intel/compiler: Make brw_private.h and simd selector helpers C++
We don't intend to expose neither to drivers, so it is fine to be C++.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19601>
|
2022-11-15 04:55:18 +00:00 |
|