Wed, 11 Nov 2015 17:32:41 -0800 test-merge-prompt.t: rename to test-merge-changedelete.t
Siddharth Agarwal <sid0@fb.com> [Wed, 11 Nov 2015 17:32:41 -0800] rev 26947
test-merge-prompt.t: rename to test-merge-changedelete.t Because: (a) the only prompts we test in this file are change/delete prompts, and (b) we're going to be expanding the scope of this test to cover future improvements to change/delete conflicts, including avoiding prompts altogether.
Thu, 12 Nov 2015 20:48:41 -0600 merge with stable
Matt Mackall <mpm@selenic.com> [Thu, 12 Nov 2015 20:48:41 -0600] rev 26946
merge with stable
Thu, 12 Nov 2015 13:16:04 -0800 tags: create new sortdict for performance reasons stable
Gregory Szorc <gregory.szorc@gmail.com> [Thu, 12 Nov 2015 13:16:04 -0800] rev 26945
tags: create new sortdict for performance reasons sortdict internally maintains a list of keys in insertion order. When a key is replaced via __setitem__, we .remove() from this list. This involves a linear scan and array adjustment. This is an expensive operation. The tags reading code was calling into sortdict.__setitem__ for each tag in a read .hgtags revision. For repositories with thousands of tags or thousands of .hgtags revisions, the overhead from list.remove() noticeable. This patch creates a new sortdict() so __setitem__ calls don't incur a list.remove. This doesn't appear to have any performance impact on my Firefox repository. But that's only because tags reading doesn't show up in profiles to begin with. I'm still waiting to hear from a user with over 10,000 tags and hundreds of heads on the impact of this patch.
(0) -10000 -3000 -1000 -300 -100 -30 -10 -3 +3 +10 +30 +100 +300 +1000 +3000 +10000 tip