4 Commits

Author SHA1 Message Date
Stanislav Mekhanoshin
8529bd7b96
[AMDGPU] Respect MBB alignment in the getFunctionCodeSize() (#127142) 2025-02-18 13:19:33 -08:00
Stanislav Mekhanoshin
bc4f05d8a8
[AMDGPU] Early bail in getFunctionCodeSize for meta inst. NFC. (#127129)
It does not change the estimate because getInstSizeInBytes() already
returns 0 for meta instructions, but added a test and early bail.
2025-02-18 02:08:28 -08:00
pvanhout
69036eb735 [AMDGPU] Fix code-size-estimate.mir test
Expensive-checks was failing on it.
2023-09-07 14:04:12 +02:00
Pierre van Houtryve
30955c9d22
[AMDGPU] Fix V_MOV_B32_indirect inst size (#65584)
This inst lowers to a normal v_mov_b32 so it's not zero-sized, but has a
size of 4.

Solves SWDEV-416337
2023-09-07 13:12:58 +02:00