From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Date: Wed, 28 Jan 2004 19:19:22 +0000 Subject: Re: [RFC/PATCH, 2/4] readX_check() performance evaluation Message-Id: <20040128201922.2dc7bec7.ak@suse.de> List-Id: References: <00a301c3e541$c13a6350$2987110a@lsd.css.fujitsu.com> <20040128182003.GL11844@parcelfarce.linux.theplanet.co.uk> In-Reply-To: <20040128182003.GL11844@parcelfarce.linux.theplanet.co.uk> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Matthew Wilcox Cc: torvalds@osdl.org, ishii.hironobu@jp.fujitsu.com, linux-kernel@vger.kernel.org, linux-ia64@vger.kernel.org On Wed, 28 Jan 2004 18:20:03 +0000 Matthew Wilcox wrote: > If there are, Linus' interface is probably the best one. If not, we could > simply have readX_check() / writeX_check() call dev->driver->unregister() > if they notice an error has occurred and then the driver doesn't even > need to call read_pcix_errors(). It just won't really work for platforms with inexact MCEs for IO errors. And even for those with exact MCEs it would probably be a nightmare to implement (writing MCE handlers is extremly hard because you cannot rely on any locking guarantees - even a printk can randomly deadlock) For those the per pci_dev callback is the only realistic way. -Andi