diff options
| author | DanilaZhebryakov <64162836+DanilaZhebryakov@users.noreply.github.com> | 2025-08-20 13:37:14 +0300 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-08-20 12:37:14 +0200 |
| commit | 0a3ee7de9c0a0f779bab035d1cd30f16e333ce76 (patch) | |
| tree | 35717eba58412daeb19c3bf9b113ab89a41df9a7 /mlir/lib/Bytecode/Reader/BytecodeReader.cpp | |
| parent | fea7e6934a2561636213dec0b808a6d4228d58d7 (diff) | |
| download | llvm-0a3ee7de9c0a0f779bab035d1cd30f16e333ce76.zip llvm-0a3ee7de9c0a0f779bab035d1cd30f16e333ce76.tar.gz llvm-0a3ee7de9c0a0f779bab035d1cd30f16e333ce76.tar.bz2 | |
[PowerPC] fix bug affecting float to int32 conversion on LE PowerPC (#150194)
When moving fcti results from float registers to normal registers
through memory, even though MPI was adjusted to account for endianness,
FIPtr was always adjusted for big-endian, which caused loads of wrong
half of a value in little-endian mode.
Diffstat (limited to 'mlir/lib/Bytecode/Reader/BytecodeReader.cpp')
0 files changed, 0 insertions, 0 deletions