[Lldb-commits] [PATCH] D66208: [lldb] Let LLDB depend on intrinsics_gen to fix non-deterministic module build failures

Wed Aug 14 14:35:37 PDT 2019

JDevlieghere added a comment.

In D66208#1630327 <https://reviews.llvm.org/D66208#1630327>, @teemperor wrote:

> In D66208#1629378 <https://reviews.llvm.org/D66208#1629378>, @JDevlieghere wrote:
>
> > LGTM.
> >
> > Out of curiosity, do you know why we don't see this on the CMake bot on GreenDragon, which is a modules build? Is this so infrequent that I might have missed it?
> >
> > http://green.lab.llvm.org/green/view/LLDB/job/lldb-cmake/
>
>
> Do the bots use a fresh build folder? Clang could find the generated headers from a previous build or there are headers with the same name in the possible include paths (even tough they are usually not included as the header that is generated would be before these other headers in the include path). That would be my explanation.

The bot is incremental but the workspace is cleaned once in a while. The combination of both things happening only occasionally makes sense.

Repository:
  rLLDB LLDB

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D66208/new/

https://reviews.llvm.org/D66208