[PATCH] D27858: Default to standalone debug info on Windows

2016-12-16 Thread Reid Kleckner via Phabricator via cfe-commits
rnk added a comment. In https://reviews.llvm.org/D27858#625367, @thakis wrote: > Did you measure what this does to link times? I've measured now, and it's pretty bad. Times to link blink_core increase by 39% with -fstandalone-debug (211s -> 294s). I still think it's what we should do by defau

Re: [PATCH] D27858: Default to standalone debug info on Windows

2016-12-16 Thread Nico Weber via cfe-commits
Won't that make is_win_fastlink builds slow to link? And if we do this, then this change here makes a then-not-very-tested config the default. (I don't have a problem with this patch, go ahead and land if you think it's the way to go. Just wondering aloud.) On Fri, Dec 16, 2016 at 4:39 PM, Reid K

[PATCH] D27858: Default to standalone debug info on Windows

2016-12-16 Thread Reid Kleckner via Phabricator via cfe-commits
rnk added a comment. In https://reviews.llvm.org/D27858#625367, @thakis wrote: > Did you measure what this does to link times? No, but I plan to recover it in Chromium by doing something like: if (is_win && is_clang && !is_win_fastlink) { cflags += ["-flimit-debug-info"] } https://re

[PATCH] D27858: Default to standalone debug info on Windows

2016-12-16 Thread Nico Weber via Phabricator via cfe-commits
thakis added a comment. Did you measure what this does to link times? https://reviews.llvm.org/D27858 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D27858: Default to standalone debug info on Windows

2016-12-16 Thread Reid Kleckner via Phabricator via cfe-commits
rnk created this revision. rnk added reviewers: thakis, hans, dblaikie. rnk added a subscriber: cfe-commits. PDB-based debuggers do not support looking up type names across type stream boundaries. There are two ways where users end up being unable to look into variables with types that were requir