xfs
[Top] [All Lists]

Re: TAKE 950027 - xfs_icsb_lock_all_counters fails with CONFIG_PREEMPT a

To: dgc@xxxxxxx (David Chinner)
Subject: Re: TAKE 950027 - xfs_icsb_lock_all_counters fails with CONFIG_PREEMPT and >=256p
From: Andi Kleen <ak@xxxxxxx>
Date: 01 Mar 2006 22:36:05 +0100
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <20060301125320.20FDA49F1681@xxxxxxxxxxxxxxxxxxxxxxx>
References: <20060301125320.20FDA49F1681@xxxxxxxxxxxxxxxxxxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3
dgc@xxxxxxx (David Chinner) writes:

> using a spinlock per cpu for superblock counter exclusion results in
> a preempt counter overflow at 256p and above. Change the exclusion mechanism
> to use atomic bit operations and busy wait loops to emulate the spin
> lock exclusion mechanism but without the preempt count issues.

That sounds like the totally wrong place to fix this. 
Wouldn't it be better to fix the spinlocks instead instead of hacking
around this? After all any other code on that big box could run
into the same issue.

-Andi


<Prev in Thread] Current Thread [Next in Thread>