[issue682] http incoming, outgoing is much slower with many heads

[Daniel Holth]

New submission from Daniel Holth <dholth at fastmail.fm>:

http outgoing is about 5 times slower for me than the same operation against a
local path.

Suggested reproduction steps: create a repository with 50+ heads*; time hg
outgoing against this repository served over "hg serve" vs. as a local path.

* we use a catch-all repository in lieu of always communicating both the path
and revision id

----------
messages: 3801
nosy: dholth
priority: bug
status: unread
title: http incoming, outgoing is much slower with many heads
topic: http_proto, performance

____________________________________________________
Mercurial issue tracker <mercurial-bugs at selenic.com>
<http://www.selenic.com/mercurial/bts/issue682>
____________________________________________________