[PATCH] D65582: IR: accept and print numbered %N names for function args

2019-08-04 Thread Tim Northover via Phabricator via cfe-commits
t.p.northover accepted this revision. t.p.northover marked an inline comment as done. t.p.northover added a comment. Thanks Reid, committed as r367755. Comment at: llvm/utils/add_argument_names.py:5 +def fix_string(s): +TYPE = re.compile('\s*(i[0-9]+|float|double|x86_fp80|

[PATCH] D65582: IR: accept and print numbered %N names for function args

2019-08-02 Thread Reid Kleckner via Phabricator via cfe-commits
rnk accepted this revision. rnk added a comment. This revision is now accepted and ready to land. lgtm Comment at: llvm/utils/add_argument_names.py:5 +def fix_string(s): +TYPE = re.compile('\s*(i[0-9]+|float|double|x86_fp80|fp128|ppc_fp128|\[\[.*?\]\]|\[2 x \[\[[A-Z_0-9]+

[PATCH] D65582: IR: accept and print numbered %N names for function args

2019-08-01 Thread Reid Kleckner via Phabricator via cfe-commits
rnk added a comment. This is awesome! Thanks for doing the test updates. As I read it, LLVM will still read old-style IR just fine, which seems nice. I think it would be worth noting this in the 10.0 release notes, since it will affect someone upgrading a frontend as you mention. Otherwise, I

[PATCH] D65582: IR: accept and print numbered %N names for function args

2019-08-01 Thread Johannes Doerfert via Phabricator via cfe-commits
jdoerfert resigned from this revision. jdoerfert added a comment. I like the idea but I am not the right person to review. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65582/new/ https://reviews.llvm.org/D65582 _

[PATCH] D65582: IR: accept and print numbered %N names for function args

2019-08-01 Thread James Y Knight via Phabricator via cfe-commits
jyknight added a comment. Herald added a subscriber: wuzish. +1 for doing this. I started looking at fixing this when I modified the printer to print proper labels for numbered basic-blocks (instead of comments), but I didn't do so because of the amount of test churn was off-putting. I think th