Re: [patch] smarter atime updates

Andreas Dilger (adilger@turbolabs.com)
Fri, 30 Nov 2001 14:52:23 -0700


On Nov 30, 2001 13:44 -0200, Marcelo Tosatti wrote:
> Now are you sure this can't break anything ?

What fun would that be, if you couldn't follow in Linus' footsteps?
People would get complacent if things didn't break now and again ;-).

Anyways, you never want to put a change that is not a specific bug
fix in a -rc patch anyways. Save it for a -pre patch, where you expect
at least some testing before it is released.

> On Thu, 29 Nov 2001, Andrew Morton wrote:
> > mark_inode_dirty() is quite expensive for journalling filesystems,
> > and we're calling it a lot more than we need to.
> >
> > --- linux-2.4.17-pre1/fs/inode.c Mon Nov 26 11:52:07 2001
> > +++ linux-akpm/fs/inode.c Thu Nov 29 21:53:02 2001
> > @@ -1187,6 +1187,8 @@ void __init inode_init(unsigned long mem
> >
> > void update_atime (struct inode *inode)
> > {
> > + if (inode->i_atime == CURRENT_TIME)
> > + return;
> > if ( IS_NOATIME (inode) ) return;
> > if ( IS_NODIRATIME (inode) && S_ISDIR (inode->i_mode) ) return;
> > if ( IS_RDONLY (inode) ) return;
> >
> >
> > with this patch, the time to read a 10 meg file with 10 million
> > read()s falls from 38 seconds (ext3), 39 seconds (reiserfs) and
> > 11.6 seconds (ext2) down to 10.5 seconds.

Well, just doing a code check of the update_atime() and UPDATE_ATIME()
users, and they are all in readlink(), follow_link(), open_namei(),
and various fs _readdir() codes. None of them (AFAICS) depend on the
mark_inode_dirty() as a side-effect. This means it should be safe.

Cheers, Andreas

--
Andreas Dilger
http://sourceforge.net/projects/ext2resize/
http://www-mddsp.enel.ucalgary.ca/People/adilger/

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/