[issue2273] "hg id" and "hg status" do too much IO

[Joerg Sonnenberger]

New submission from Joerg Sonnenberger <joerg at bec.de>:

I have a medium sized Mercurial repo (for a working copy, 9k files and 33MB
.hg). During testing I am replacing around 30 files (300MB) in the
repository with a data sets that is over all 900MB large. This dramatically
slows down both "hg id" and "hg status".

ktrace gives a 300MB trace, so it seems like it is reading the whole files
instead of just stat(2) them to see that the file size changed.

----------
messages: 13007
nosy: joerg, tonfa
priority: bug
status: unread
title: "hg id" and "hg status" do too much IO
topic: performance

____________________________________________________
Mercurial issue tracker <bugs at mercurial.selenic.com>
<http://mercurial.selenic.com/bts/issue2273>
____________________________________________________