Re: [PATCH] Avoid accessing a slow working copy during diffcore operations.

From: Johannes Schindelin <johannes.schindelin@gmx.de>
Date: Dec 14, 2006, 15:37 UTC
Message-ID: <Pine.LNX.4.63.0612141633340.3635@wbgn013.biozentrum.uni-wuerzburg.de>
In-Reply-To: <81b0412b0612140649i71643aaar847460ca9e4cea48@mail.gmail.com>

Hi,

On Thu, 14 Dec 2006, Alex Riesen wrote:

Show 10 quoted lines

> On 12/14/06, Johannes Schindelin <Johannes.Schindelin@gmx.de> wrote:
> > 
> > It is usually cheaper to just read the file, especially if it is still 
> > cached, because the alternative means unpacking the loose object, or 
> > worse, unpacking the packed object _along_ with the objects in its 
> > delta chain.
> 
> But you have to read less, and even that could be in cache as well and 
> unpacking in userspace could be faster than open/write temporary/ read 
> temporary/close/unlink temporary file on a normal system

You have to unpack anyway, since even the loose objects are packed. But to reconstruct a deltified object, you have to reconstruct possibly many objects.

So yes, if you have it in the working directory (unpacked), it should be faster to just read it, especially if it is still in the filesystem cache.

Show 5 quoted lines

> > Not every OS sucks cache-wise, and you should not make others suffer 
> > for Redmond's shortcomings.
> 
> I'm just do not understand why avoiding temporary file wouldn't help all 
> OSes, even if they do not suck cache-wise.

Ah! But it is not temporary! It is the "working copy", which means the file you have in the working directory. IOW it is an unpacked blob, which happens to be already unpacked anyway.

Ciao, Dscho