See bug 36843: https://bugs.llvm.org/show_bug.cgi?id=36843
I had to correct a lit test data.s because seemingly unused combination of bits
.long 0xbadc0de1
turned out to be a valid gfx9 instruction:
s_call_b64 s[92:93], 0xde1
Isn't it curious?
Paths
| Differential D45268
[AMDGPU][MC][GFX9] Added s_call_b64 ClosedPublic Authored by dp on Apr 4 2018, 10:30 AM.
Details Summary See bug 36843: https://bugs.llvm.org/show_bug.cgi?id=36843 I had to correct a lit test data.s because seemingly unused combination of bits .long 0xbadc0de1 turned out to be a valid gfx9 instruction: s_call_b64 s[92:93], 0xde1 Isn't it curious?
Diff Detail Event TimelineHerald added subscribers: t-tye, tpr, dstuttard and 4 others. · View Herald TranscriptApr 4 2018, 10:30 AM This revision is now accepted and ready to land.Apr 5 2018, 8:53 AM Closed by commit rL329440: [AMDGPU][MC][GFX9] Added s_call_b64 (authored by dpreobra). · Explain WhyApr 6 2018, 11:27 AM This revision was automatically updated to reflect the committed changes.
Revision Contents
Diff 140987 lib/Target/AMDGPU/SOPInstructions.td
test/MC/AMDGPU/data.s
test/MC/AMDGPU/sopk.s
test/MC/Disassembler/AMDGPU/sopk_gfx9.txt
|