tmroeder added a comment.
In https://reviews.llvm.org/D54657#1303320, @joerg wrote:
> I'm sorry, but it still sounds to me like you want to address badly written
> build rules by making the driver more complicated. I don't see that is a
> reasonable goal forward.
My goal is to build a compila
tmroeder updated this revision to Diff 174676.
tmroeder added a comment.
Change CompilationDatabase to not output entries that use -E.
These entries cannot be used by the tooling, so they can be skipped. This
removes the other code and adds tests to make sure that neither -E or -###
entries wri
joerg added a comment.
I'm sorry, but it still sounds to me like you want to address badly written
build rules by making the driver more complicated. I don't see that is a
reasonable goal forward.
Repository:
rC Clang
https://reviews.llvm.org/D54657
__
tmroeder added a comment.
In https://reviews.llvm.org/D54657#1302086, @joerg wrote:
> I don't understand the point here. Why would you want to include
> pre-processing-only commands in the compilation database?
Sorry for not being clear enough.
I don't want to include entries with -E. However
joerg added a comment.
I don't understand the point here. Why would you want to include
pre-processing-only commands in the compilation database?
Repository:
rC Clang
https://reviews.llvm.org/D54657
___
cfe-commits mailing list
cfe-commits@lists
tmroeder created this revision.
tmroeder added a reviewer: klimek.
Herald added a subscriber: cfe-commits.
Some build systems compile files more than once (e.g., Kbuild in the
Linux kernel compiles with -c once and with -E once). In these cases,
the -MJ flag will only record the latest compilation
