[PATCH] D25073: [DebugInfo]: preparation to implement DW_AT_alignment

Victor Leschuk via llvm-commits llvm-commits at lists.llvm.org
Mon Oct 10 02:51:03 PDT 2016


vleschuk marked an inline comment as done.
vleschuk added a comment.

In https://reviews.llvm.org/D25073#563561, @aprantl wrote:

> An alternative for DILocalVariable would be to stick a HasAlignment bit into Record[0] & 2, similar to how DISubprogram works. This may be simpler / more readable / more space efficient.


Done, thanks for the tip!

> Do any of the bitfield tests need upgrading to remove the alignment and verify they still work?

Yep, thanks for reminding. Removed alignment debug info from bitfield-related tests and checked that they pass.



================
Comment at: include/llvm/IR/DIBuilder.h:209
         DIScope *Scope, StringRef Name, DIFile *File, unsigned LineNo,
-        uint64_t SizeInBits, uint64_t AlignInBits, uint64_t OffsetInBits,
+        uint64_t SizeInBits, uint64_t OffsetInBits,
         uint64_t StorageOffsetInBits, DINode::DIFlags Flags, DIType *Ty);
----------------
aprantl wrote:
> Have you verified that this doesn't break bitfields?
Yes, I ran some manual tests, and also double checked bitfield-related tests (both for DWARF and COFF).


https://reviews.llvm.org/D25073





More information about the llvm-commits mailing list