[AMDGPU][MC] Corrected parsing of FLAT offset modifier

Summary of changes:

- simplified handling of FLAT offset: offset_s13 and offset_u12 have been replaced with flat_offset;
- provided information about error position for pre-gfx9 targets;
- improved errors handling.

Reviewers: artem.tamazov, arsenm, rampitec

Differential Revision: https://reviews.llvm.org/D64244

llvm-svn: 365321

1 files changed, 1 insertions, 1 deletions

diff --git a/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp b/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp
index 366290d214d..46f1437ab00 100644
--- a/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp
@@ -694,7 +694,7 @@ bool isValidHwregWidth(int64_t Width) {
   return 0 <= (Width - 1) && isUInt<WIDTH_M1_WIDTH_>(Width - 1);
 }
 
-int64_t encodeHwreg(int64_t Id, int64_t Offset, int64_t Width) {
+uint64_t encodeHwreg(uint64_t Id, uint64_t Offset, uint64_t Width) {
   return (Id << ID_SHIFT_) |
          (Offset << OFFSET_SHIFT_) |
          ((Width - 1) << WIDTH_M1_SHIFT_);