If there's a packed epilogue (indicated by the flag E), the EpilogueCount() field actually should be interpreted as EpilogueOffset.
Details
Diff Detail
Event Timeline
| llvm/test/tools/llvm-readobj/COFF/arm64-packed-epilog.s | ||
|---|---|---|
| 2 | If the format is important: # RUN: llvm-mc -filetype=obj -triple aarch64-windows %s -o - |
# RUN: llvm-readobj --unwind - | FileCheck %s --match-full-lines --strict-whitespace
# CHECK:ExceptionData {
# CHECK-NEXT: FunctionLength: 4@grimar may prefer keeping the temporary file as %t | |
| llvm/test/tools/llvm-readobj/COFF/arm64-packed-epilog.s | ||
|---|---|---|
| 2 | Yes, I find it is much more convenient for debugging broken tests to have them. | |
| llvm/test/tools/llvm-readobj/COFF/arm64-packed-epilog.s | ||
|---|---|---|
| 2 | Sure, I can change that. A number of the existing related testcases also use the temp-less form, but I can change to using temp files for the newly added tests at least, and cases where I otherwise touch the RUN lines. | |
| llvm/test/tools/llvm-readobj/COFF/arm64-packed-epilog.s | ||
|---|---|---|
| 2 | Sounds good, thanks! | |
If the format is important:
# RUN: llvm-mc -filetype=obj -triple aarch64-windows %s -o - | # RUN: llvm-readobj --unwind - | FileCheck %s --match-full-lines --strict-whitespace # CHECK:ExceptionData { # CHECK-NEXT: FunctionLength: 4@grimar may prefer keeping the temporary file as %t