From: Jakub Narebski Date: Sun, 14 Feb 2010 11:07:17 GMT Subject: Re: [PATCH] don't use mmap() to hash files Message-ID: In-Reply-To: <20100214024259.GB9704@dpotapov.dyndns.org> Dmitry Potapov writes: > On Sun, Feb 14, 2010 at 02:53:58AM +0100, Johannes Schindelin wrote: > > On Sun, 14 Feb 2010, Dmitry Potapov wrote: > > > > > + if (strbuf_read(&sbuf, fd, 4096) >= 0) > > > > How certain are you at this point that all of fd's contents fit into your > > memory? > > You can't be sure... In fact, we know mmap() also may fail for huge > files, so can strbuf_read(). Perhaps, mmap() behaves better when you > want to hash a huge file that does not fit in free physical memory, but > I do not think it is an important use case for any VCS, which mostly > stores small text files and a few not so big binary files. Git is not > design to store your video collection. (probably, Git can be improved to > handle big files better but I leave that exercise to those who want to > store their media files in Git). Something like git-bigfiles: http://caca.zoy.org/wiki/git-bigfiles ? (found via http://blog.bitquabit.com/2010/02/10/fightings-been-fun-and-all-its-time-shut-and-get-along/ found via http://tomayko.com/, entry for 10 Feb 2010). -- Jakub Narebski Poland ShadeHawk on #git