[Lldb-commits] [PATCH] D70840: [LLDB] [DWARF] Strip out the thumb bit from addresses on ARM

Fri Nov 29 03:30:36 PST 2019

mstorsjo added a comment.

In D70840#1763639 <https://reviews.llvm.org/D70840#1763639>, @labath wrote:

> Yeah, this is going to be tricky... I don't really know what's the right way to do this, but here are the thoughts I have on this so far:
>
> - The new DWARFDebugInfoEntry member is going to substantially increase our memory footprint -- that's a non-starter

Ok, I feared that class was one where the size is critical yeah...

> - I don't like the inconsistency where all addresses are demangled in the DWARF code, except the line tables, which are in the "generic" code

Yup. The reason for doing it like this was as I tried to find the most narrow interface where I could catch all of them.

> - I think we ought to have some kind of a utility function to hold the logic for the `&~1` stuff. there is something like that in Architecture::GetOpcodeLoadAddress. The Architecture class is mostly independent of a target, so we may be able create one instance for each module or symbol file, but that feels quite heavy. I might even go for putting something like that into the ArchSpec class. The raison d'être of the Architecture class was to evict some heavy code out of ArchSpec -- this was ArchitectureArm::OverrideStopInfo. That one is definitely too heavy, so still don't thing it belongs in ArchSpec, but the `&~1` thingy seems like it could find a place there.)
> - is this behavior reproducible with lld ? If so, I think it'd make the test more readable to run llvm-mc+lld as a part of the test

Yes it is. Ok, that's fine with me.

> - the address byte size would best be handled separately

On second thought, yes, this should be trivial to split out and make a similar test with inspecting line tables of an i386 binary.

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D70840/new/

https://reviews.llvm.org/D70840