[prev in list] [next in list] [prev in thread] [next in thread] 

List:       squid-dev
Subject:    Re: busy bitmaps and race conditions
From:       Henrik Nordstrom <hno () hem ! passagen ! se>
Date:       2000-05-26 7:14:47
Message-ID: 392E2467.1AADCBFB () hem ! passagen ! se
[Download RAW message or body]

Adrian Chadd wrote:
> 
> While going through diskd, ufs and aufs I've noticed that the async
> unlinks aren't taken into consideration when choosing a new (free)
> bit. I can't see this being too much of a problem in normal use
> but there's a possibility for a race condition if you are running out
> of free bits.
> 
> Am I missing something here?

This is to avoid a race condition where the file is reused while being
removed.

The bitmap grows itself if needed (or am I completely lost what bitmap
you are talking about?), so there shouldn't be any race condition there.

/Henrik

[prev in list] [next in list] [prev in thread] [next in thread]