scmutil: use the new faster path split Combined with a few other patches in this series, this contributes to improving stream_out performance by 10%.

util: implement a faster os.path.split for posix systems This is not yet used.

scmutil: make join cheaper Combined with a few followup patches, this contributes to improving stream_out performance by 10%.

wireproto: don't format a debug string inside a hot loop This improves stream_out performance by about 5%.

wireproto: bypass filechunkiter for small files when streaming Merely creating and using a generator has a measurable impact, particularly since the common case for stream_out is generators that yield just once. Avoiding generators improves stream_out performance by about 7%.

wireproto: don't audit local paths during stream_out Auditing at this stage is both pointless (paths are already trusted by the local repo) and expensive. Skipping the audits improves stream_out performance by about 15%.

scmutil: delegate mustaudit property to the real opener This will be used by an upcoming patch.

scmutil: turn opener._audit into a property, mustaudit This will be used by an upcoming patch.

perf: add perffncacheencode Examples (all done with somewhat dated clones I found on my disk): Netbeans (~120k entries in fncache): $ hg perffncacheencode ! wall 4.338000 comb 4.336828 user 4.336828 sys 0.000000 (best of 3) Openoffice (~77k entries in fncache)): $ hg perffncacheencode ! wall 1.533000 comb 1.528810 user 1.528810 sys 0.000000 (best of 7) Xen (~10k entries in fncache): $ hg perffncacheencode ! wall 0.198000 comb 0.187201 user 0.187201 sys 0.000000 (best of 51) Done on Windows 7 x64.

Merge with mpm