From: ming.lei@redhat.com (Ming Lei) Date: Thu, 28 Mar 2019 08:55:06 +0800 Subject: [PATCH stable only] nvmet: set loop queue's segment boundary mask as PAGE_SIZE - 1 In-Reply-To: <20190327161658.GL25262@sasha-vm> References: <20190327090230.12671-1-ming.lei@redhat.com> <20190327161658.GL25262@sasha-vm> Message-ID: <20190328005505.GB19708@ming.t460p> On Wed, Mar 27, 2019@12:16:58PM -0400, Sasha Levin wrote: > On Wed, Mar 27, 2019@05:02:30PM +0800, Ming Lei wrote: > > NVMe target only accepts single-page sg list, either file or block > > device backed target code follows this assumption. > > > > However, loop target is one exception, given the sg list is from > > the host queue directly. > > > > This patch sets loop queue's segment boundary mask as PAGE_SIZE - 1 > > for following NVMe target assumption. > > > > Reported-by: Yi Zhang > > Fixes: 3a85a5de29ea ("nvme-loop: add a NVMe loopback host driver") > > Cc: Yi Zhang > > Cc: Sagi Grimberg > > Cc: Chaitanya Kulkarni > > Cc: > > Signed-off-by: Ming Lei > > Could you provide some background as to why this is -stable only in the > commit message? Multi-page bvec has been merged to linus tree already, so the following patch simply can fix the current issue without needing to limit the segment size for nvme-loop: http://lists.infradead.org/pipermail/linux-nvme/2019-March/023105.html Thanks, Ming