mesa/src/panfrost/util
Alyssa Rosenzweig 5f93feed61 panfrost: Don't merge workgroups with variable shared mem
If nir->info.shared_size = 0 but grid->variable_shared_mem > 0, the shader uses
shared memory but the compiler may not realize that. We need to disable
workgroup merging even in this case. The alternate approach is to statically
check for shared intrinsics in the compiler, but this is a bit easier all things
considered.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18581>
2022-11-02 23:36:56 +00:00
..
lcra.c
lcra.h
meson.build panfrost: Don't use nir_variable to link varyings 2022-11-02 16:52:11 +00:00
nir_mod_helpers.c
pan_collect_varyings.c panfrost: Don't use nir_variable to link varyings 2022-11-02 16:52:11 +00:00
pan_ir.c
pan_ir.h panfrost: Don't merge workgroups with variable shared mem 2022-11-02 23:36:56 +00:00
pan_liveness.c
pan_lower_64bit_intrin.c
pan_lower_framebuffer.c panfrost: Don't use nir_variable in the compilers 2022-11-02 04:22:06 +00:00
pan_lower_framebuffer.h
pan_lower_helper_invocation.c
pan_lower_sample_position.c
pan_lower_store_component.c pan/mdg: Use bifrost_nir_lower_store_component 2022-11-02 16:52:11 +00:00
pan_lower_writeout.c panfrost: Don't use nir_variable in the compilers 2022-11-02 04:22:06 +00:00
pan_lower_xfb.c panfrost: Use compute-based XFB on Midgard 2022-10-27 20:13:11 +00:00
pan_sysval.c pan/bi: Handle transform feedback intrinsics 2022-06-04 14:35:56 +00:00