On Fri, Apr 05, 2013 at 06:18:10AM +0400, Max Filippov wrote: > On Fri, Apr 5, 2013 at 5:53 AM, Shaohua Li <s...@kernel.org> wrote: > > On Thu, Apr 04, 2013 at 06:00:18AM +0400, Max Filippov wrote: > > [...] > > >> the commit 0cfbcafcae8b7364b5fa96c2b26ccde7a3a296a9 'block: add plug > >> for blkdev_issue_discard' > >> have added merge opportunity for DISCARD requests. When I do > >> mkfs.f2fs on a 5G partition (0xad8000 sectors) it submits two bios, > >> one for 0x7fe000 sectors (0xffc00000 bytes) and another for > >> 0x2da000 sectors (0x5b400000 bytes). Prior to that commit these > >> bios weren't merged into one request. Now the second bio gets > >> merged with the first, but the request's __data_len field is unsigned int > >> and it gets wrapped to 0x5b000000 bytes instead of 0x15b000000 > >> in the bio_attempt_back_merge. Later this reduced size is passed to > >> the blk_update_request causing KERN_ERR and not completed > >> request. Reverting this commit fixes mkfs.f2fs for me. > > > > A workaround is setting limits.max_discard_sectors to a smaller value. > > I'm not sure: > 1) in my case max_discard_sectors is 0x7fe000 (0xffc00000 bytes, > which still fits into 32 bits) and > 2) this parameter will only change size of individual discard requests for > the discarded range, but as long as these requests are done inside > the plug they will be merged anyway with an overflow if we try > to discard more than 4G at once.
No, merge in plug list checks max_discard_sectors. Please see ll_back_merge_fn() in bio_attempt_back_merge(). Merge in unplug checks max_discard_sectors too. I didn't see any chance a merge doesn't check max_discard_sectors. Can you post a blktrace when the discard ioctl runs and double check if max_discard_sectors is really 0x7fe000? -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/