[PATCH] D154364: [LTO] Allow library calls to be internalized in freestanding mode

[Joseph Huber via Phabricator via llvm-commits]

jhuber6 added a comment.

In D154364#4469476 <https://reviews.llvm.org/D154364#4469476>, @efriedma wrote:

>> it sets TLII->disableAllFunctions();. I'm guessing that's not sufficient here?
>
> That just disables TargetLibraryInfo, i.e. recognition of non-intrinsic symbols as having some particular meaning.  Backends generally still assume that compiler-rt.builtins or libgcc, plus memcpy/memmove/memset, are available.

The functions that I'm interested in internalizing are the math functions. Since the GPU resolves everything statically we should be able to internalize all of this and remove the overhead. I'm not sure if there's a good way to express that. Maybe a secondary list?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D154364/new/

https://reviews.llvm.org/D154364