forked from OSchip/llvm-project
ef72107a49
According to the arm arm specifications, 4 bytes are needed for a shift instead of 8, this was causing the movt instruction to write to a different register sometimes. Patch by Walter Erquinigo! llvm-svn: 280005 |
||
|---|---|---|
| .. | ||
| Interpreter | ||
| MCJIT | ||
| OrcLazy | ||
| OrcMCJIT | ||
| RuntimeDyld | ||
| 2010-01-15-UndefValue.ll | ||
| fma3-jit.ll | ||
| frem.ll | ||
| lit.local.cfg | ||
| mov64zext32.ll | ||
| test-interp-vec-arithm_float.ll | ||
| test-interp-vec-arithm_int.ll | ||
| test-interp-vec-cast.ll | ||
| test-interp-vec-insertelement.ll | ||
| test-interp-vec-insertextractvalue.ll | ||
| test-interp-vec-loadstore.ll | ||
| test-interp-vec-logical.ll | ||
| test-interp-vec-select.ll | ||
| test-interp-vec-setcond-fp.ll | ||
| test-interp-vec-setcond-int.ll | ||
| test-interp-vec-shift.ll | ||
| test-interp-vec-shuffle.ll | ||