Module: Mesa
Branch: main
Commit: 34d8bfe65f47df92efe83d902522297614f9a44a
URL:    
http://cgit.freedesktop.org/mesa/mesa/commit/?id=34d8bfe65f47df92efe83d902522297614f9a44a

Author: Lionel Landwerlin <[email protected]>
Date:   Fri Mar 17 09:42:31 2023 +0200

intel/fs: run VGRF compaction just before max live register accounting

There are a number of instances of the dead code elimination pass that
could reduce the count. For some reason this also seems to affect
register allocation itself.

Signed-off-by: Lionel Landwerlin <[email protected]>
Cc: mesa-stable
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Francisco Jerez <[email protected]>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21853>

---

 src/intel/compiler/brw_fs.cpp | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/intel/compiler/brw_fs.cpp b/src/intel/compiler/brw_fs.cpp
index b22b32e5112..77cd0b757b8 100644
--- a/src/intel/compiler/brw_fs.cpp
+++ b/src/intel/compiler/brw_fs.cpp
@@ -6506,6 +6506,8 @@ fs_visitor::allocate_registers(bool allow_spilling)
       "lifo"
    };
 
+   compact_virtual_grfs();
+
    if (needs_register_pressure)
       shader_stats.max_register_pressure = compute_max_register_pressure();
 

Reply via email to