https://bugs.llvm.org/show_bug.cgi?id=48638
Bug ID: 48638
Summary: [AMDGPU][MC][DOC] Inconsistent type names for signed
integer
Product: libraries
Version: trunk
Hardware: PC
OS: Windows NT
Status: NEW
Severity: enhancement
Priority: P
Component: Backend: AMDGPU
Assignee: [email protected]
Reporter: [email protected]
CC: [email protected]
AMDGPU assembler documentation uses both 's' and 'i' symbols to refer signed
integer types - see https://llvm.org/docs/AMDGPU/AMDGPUAsmGFX10.html
For example, 32-bit signed integers are referred to as 's32':
global_atomic_smax vdst:opt:s32, vaddr, vdata:s32, saddr offset12s glc slc
On the other hand, 'i16' is used to refer 16-bit signed integers:
v_ldexp_f16_e64 vdst, src0:m, src1:i16 clamp
Documentations should use 'i' for signed integer types to match notation
employed for instruction suffices.
--
You are receiving this mail because:
You are on the CC list for the bug._______________________________________________
llvm-bugs mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
