From: Linus Torvalds <torvalds@linuxfoundation.org>
Date: Fri, 30 May 2008 00:32:30 GMT
Subject: Re: reducing prune sync()s
Message-ID: <alpine.LFD.1.10.0805291727490.3141@woody.linux-foundation.org>
In-Reply-To: <alpine.LFD.1.10.0805291656260.3141@woody.linux-foundation.org>



On Thu, 29 May 2008, Linus Torvalds wrote:
> On Thu, 29 May 2008, Frank Ch. Eigler wrote:
> > 
> >	  Or perhaps having the blanket sync be replaced a
> > list of fsync()s for only the relevant git repository files?
> 
> That would be much better.

Side note: a lot of systems make "fsync()" pretty expensive too. It's one 
of my main disagreements with most log-based filesystems - fsync() can in 
theory be fast, but almost always implies flushing the whole log, even if 
99.9% of that log is totally unrelated to the actual file you want to 
fsync().

So fsync() isn't always all that much better than sync(). It *should* be, 
but reality sometimes bites. So testing should include at least some level 
of "yes, it actually improves things at least on xyz"...

			Linus