[llvm-dev] [8.0.0 Release] rc1 has been tagged

[Bruce Hoult via llvm-dev]

On Fri, Feb 1, 2019 at 12:56 AM Hans Wennborg <hans at chromium.org> wrote:
> I tried merging r352169 but there merge conflicts in
> lib/Target/RISCV/RISCVInstrInfo.td,
> test/CodeGen/RISCV/atomic-cmpxchg.ll and
> test/CodeGen/RISCV/atomic-rmw.ll.

The RISCVInstrInfo.td should be trivial.

Sadly with this being a commit that changes code generation for a very
common operation (shifts on 32 bit ints on a 64 bit system) there will
be small changes in the machine code for a *lot* of source code
functions. With LLVM tests being as brittle as they are it's likely
that a lot more than just those two tests with merge conflicts will
actually fail, and need adjusting also.

> I didn't actually look at the conflicts, but since the target is still
> in experimental and the merge isn't trivial, maybe it's not worth it?
> I'm still happy to take them if someone wants to help me resolve the
> conflicts though.

I think if we can't also get in the single precision and double
precision float code generation for 64 bit target that landed today,
and change the back end to non-experimental (and probably one or two
other things. TLS? plt@?) there isn't much point.

I think all that stuff will be in trunk by the time LLVM 8.0 actually
releases, but I guess it's missing the release.