From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Subject: Re: [PATCH v2] block: bio_check_eod() needs to consider partition To: Bart Van Assche , "hch@lst.de" Cc: "jiufei.xue@linux.alibaba.com" , "linux-block@vger.kernel.org" References: <20180313091819.26881-1-hch@lst.de> <7c4181b7-788f-bf90-8397-c09beceedcb7@kernel.dk> <1520958339.2638.8.camel@wdc.com> From: Jens Axboe Message-ID: <315d98a1-7ade-4d8c-a71d-7716d27413ac@kernel.dk> Date: Tue, 13 Mar 2018 10:37:05 -0700 MIME-Version: 1.0 In-Reply-To: <1520958339.2638.8.camel@wdc.com> Content-Type: text/plain; charset=utf-8 List-ID: On 3/13/18 9:25 AM, Bart Van Assche wrote: > On Tue, 2018-03-13 at 07:49 -0700, Jens Axboe wrote: >> On 3/13/18 2:18 AM, Christoph Hellwig wrote: >>> bio_check_eod() should check partiton size not the whole disk if >>> bio->bi_partno is non-zero. Does this by taking the call to bio_check_eod >>> into blk_partition_remap. >> >> Applied, thanks. > > Hello Christoph and Jens, > > I think that this patch introduces a severe regression: with this patch applied > a VM that I use for kernel testing doesn't boot anymore. If I revert this patch > that VM boots fine again. This is what I see on the serial console with this > patch applied: > > virtio_blk virtio4: [vda] 41943040 512-byte logical blocks (21.5 GB/20.0 GiB) > vda: vda1 > attempt to access beyond end of device > vda: rw=0, want=41940872, limit=41938944 > attempt to access beyond end of device > vda: rw=0, want=41940872, limit=41938944 I've killed the patch for now, thanks for the quick test, Bart. -- Jens Axboe