- User Since
- Apr 2 2016, 12:44 PM (400 w, 3 d)
Nov 21 2018
The ABI breakage was already there, in the difference between GCC and Clang
- if one compiles against libc++ with gcc, it is not compatible with things
that are compiled with clang. I frankly think that being ABI compatible
with gcc (and correct according to the standard) is far more important than
being ABI compatible with previous versions of clang - if people need the
ABI compatibility, they can and should use -fclang-abi-compat
Oct 25 2018
I don't actually know how to send this upstream, since it's been accepted... How should I proceed?
Oct 24 2018
Add ABI breakage information and reflow
Oct 23 2018
Finish doing the thing! all tests are passing, at least on the revision I'm running on :)
... Yup, that'd do it. Thanks Richard!
Is it possible you have LLVM and Clang checked out at different revisions? That's a common source of weird behavior, particularly in CodeGen tests.
Oct 18 2018
I'm having real difficulty with this - I get really odd errors in seemingly unrelated tests when I change things.
Oct 12 2018
Apr 5 2016
Coalesce the checks.
Add checks for the Ret value as well
Apr 4 2016
Switch to explicit constructor.
So how this would work is; any additional values would be added after LLVMInstructionValueKind, and if any values are removed, they would be removed in this enum, and the values would have integer values assigned.
Fix deadalnix's problems with the revision.
Follow deadalnix's suggestions.
Apr 3 2016
Return a const string instead.
Apr 2 2016
Add tests, fix that switch in LLVMGetValueKind
listening to whitequark :P
I'm sorry, learning the tools :)
This was meant to be four diffs; I don't understand the tools fully yet.