Re: [RFC/PATCH, 2/4] readX_check() performance evaluation

From: Linus Torvalds <>
Date: 2004-01-29 07:30:48
On Wed, 28 Jan 2004, Linus Torvalds wrote:
> If the real error is on the bridge somewhere but you don't even know which
> CPU did the access (and just "somebody" gets an MCE), just set a global
> flag, and have "read_pcix_error()" check the bridge (since it doesn't need
> to look anything up - it already knows the device).
> And in that case then you need to take the proper locks (per-bridge, or
> global, depending on just how much you care) in "clear_pcix_error()" and
> release them in "read_pcix_error()".

Note, in case this wasn't clear already: in this case, the "MCE flag" is 
just a lazy flag saying "you need to check more deeply". It wouldn't cause 
false positives, simply because the _real_ check ends up being 
"read_pcix_error()" actually reading the error status from the bridge or 
the device.

It's just that 99% of the time, you don't want to do even that.

To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to
More majordomo info at
Received on Wed Jan 28 15:39:14 2004

This archive was generated by hypermail 2.1.8 : 2005-08-02 09:20:21 EST