On 02/02/2016 08:06 AM, Damien Le Moal wrote: > > Hello, > > In ll_front_merge_fn, the request size result of the eventual front > merge of a BIO with a request is checked against the maximum > size allowed for the request using blk_rq_get_max_sectors. This > function will also check that the merge result will not span block > chunks for chunked block devices using the function blk_max_size_offset. > > However, blk_rq_get_max_sectors always calls blk_max_size_offset using > the request sector position, which in the case of a front merge is not > the position at which the merged request would be issued: the sector > position to use must be that of the BIO, and not that of the request. > > This problem can trigger a “boundary violation error” for write > requests on ZAC/ZBC host-managed SMR disks as the last write BIO > of a zone (a chunk) can end up being front-merged with the first > request of the following zone (chunk). > > The attached patch against linux-4.5-rc2 fixes this problem by adding > an offset argument to the function blk_rq_get_max_sectors. > Reviewed-by: Hannes Reinecke <hare@xxxxxxxx> I can easily respin this as a proper patch if required. Cheers, Hannes -- Dr. Hannes Reinecke Teamlead Storage & Networking hare@xxxxxxx +49 911 74053 688 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton HRB 21284 (AG Nürnberg) -- To unsubscribe from this list: send the line "unsubscribe linux-block" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
