[PATCH] D105506: [LLD] [COFF] Avoid thread exhaustion on 32-bit Windows host

[Martin Storsjö via Phabricator via llvm-commits]

mstorsjo added a comment.

In D105506#2860673 <https://reviews.llvm.org/D105506#2860673>, @jeremyd2019 wrote:

> I couldn't tell anything useful from that.  Both patched and unpatched lld linked QT6Core.dll (which crashed LLD in a parallel build) in about 0.3 seconds.

Ok, thanks for testing anyway. Based on the numbers from the old commit from @pcc, it'd be a worst case slowdown of 2x, but most actual cases you can link on 32 bit are so small anyway that you probably run into the memory mapping limit fairly soon anyway, so the cases that can be linked are small enough that it's plenty fast anyway.

So with that I think this is fine, but I'd still like an ack from @rnk.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D105506/new/

https://reviews.llvm.org/D105506