sammccall added a comment. In https://reviews.llvm.org/D48071#1171289, @ilya-biryukov wrote:
> A recent change (https://reviews.llvm.org/D49267) is another indication that > caching might be doing more wrong than good. I assume the caching does not > give us much performance-wise, we only request compile commands for file > reparses and reparses tend to be slow and do lots of file system accesses > anyway. > Maybe we should just disable it altogether. @sammccall, @simark, WDYT? Yeah, the caching is bad and we should get rid of it. I thought an option would make this smoother but I no longer think so. AFAIK the only people who actually need this option at present are internal google users due to our suspicious CDB. Those people aren't well-served by having to set an option. Let's find another solution for that and then just remove the caching? Repository: rCTE Clang Tools Extra https://reviews.llvm.org/D48071 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
