From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [PATCH] scsi: Silence unnecessary warnings about ioctl to partition Date: Wed, 02 May 2012 15:59:59 +0200 Message-ID: <4FA13DDF.9010006@redhat.com> References: <1335953452-10460-1-git-send-email-jack@suse.cz> <4FA1092E.9090603@redhat.com> <20120502135123.GF16976@quack.suse.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from mx1.redhat.com ([209.132.183.28]:28159 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754407Ab2EBOA3 (ORCPT ); Wed, 2 May 2012 10:00:29 -0400 In-Reply-To: <20120502135123.GF16976@quack.suse.cz> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Jan Kara Cc: Jens Axboe , LKML , James Bottomley , linux-scsi@vger.kernel.org Il 02/05/2012 15:51, Jan Kara ha scritto: >> > NACK. I would bet that all the warnings you've seen are for ioctl that >> > would have failed anyway with ENOTTY. > Actually, you would loose the bet ;) Doh. :) > The customer was complaining about > warning about SG_IO ioctl. Apparently some Veritas filesystem thread generates > a *lot* of these (I don't know if they happen to do all the filesystem IO > with SG_IO and I'm not sure I want to know ;). Can you at least ask the customer for help finding which command was sent? And perhaps have them try a kernel that blocks SG_IO to see what breaks if anything? > Also I tend to side with Alan that I don't quite see > the point in trying to restrict CAP_SYS_RAWIO threads and thus breaking the > compatibility For example, we have a customer that wants this: * a VM should be able to send vendor-specific commands to a disk via SG_IO (vendor-specific commands require CAP_SYS_RAWIO). * they want to assign logical volumes or partitions to the same VM without letting it read or write outside the logical volume or partition. Of course a better solution for this would be customizable filters for SG_IO commands, where a privileged application would open the block device with CAP_SYS_RAWIO, set the filter and hand the file descriptor to QEMU. Or alternatively some extension of the device cgroup. But either solution would require a large amount of work. Paolo > (if ioctls would be restricted for partitions from the > beginning, then sure it seems like a cleaner choice). But I don't feel that > strongly about it.