Re: Making statCounter.syscalls.disk counters more consistent

On Fri, 2003-07-18 at 18:12, Henrik Nordstrom wrote:
> On Friday 18 July 2003 02.13, Duane Wessels wrote:
>
> > Okay. There is already a call to increment disk.unlinks in
> > aiops.c, so if it is unsafe I guess it should be removed.
>
> Indeed.
>
> If you want to have counters in aufs then store_io_aufs or async_io is
> good places to add them. Both are threads safe as opposed to aiops.c.
> Pick the one which makes the counting most unified among the storeio
> implementations.

I wouldn't worry about doing this in HEAD just now. Immediately
post-3.0, I have a large unification patch for ufs, diskd and aufs,
reducing the duplicate code still futher over 3.0's already semi-unified
logic.

Rob

-- 
GPG key available at: <http://members.aardvark.net.au/lifeless/keys.txt>.