Re: dm: propagate QUEUE_FLAG_NO_SG_MERGE

2014-08-15 Thread Jens Axboe
On 08/11/2014 08:40 AM, Jeff Moyer wrote: > Mike Snitzer writes: > >> Hi Jeff, >> >> Thanks for the patch. But I think you need the following tweak. >> Otherwise if the DM table is reloaded (and the devices in the table >> happen to change) the flag won't get cleared as needed. > > Ah, thanks f

Re: dm: propagate QUEUE_FLAG_NO_SG_MERGE

2014-08-11 Thread Jeff Moyer
Mike Snitzer writes: > Hi Jeff, > > Thanks for the patch. But I think you need the following tweak. > Otherwise if the DM table is reloaded (and the devices in the table > happen to change) the flag won't get cleared as needed. Ah, thanks for the education. Looks good to me. Thanks! Jeff -- T

Re: dm: propagate QUEUE_FLAG_NO_SG_MERGE

2014-08-10 Thread Mike Snitzer
On Fri, Aug 08 2014 at 11:03am -0400, Jeff Moyer wrote: > Hi, > > Commit 05f1dd5 introduced a new queue flag: QUEUE_FLAG_NO_SG_MERGE. > This gets set by default in blk_mq_init_queue for mq-enabled devices. > The effect of the flag is to bypass the SG segment merging. Instead, > the bio->bi_vcnt

Re: [patch] dm: propagate QUEUE_FLAG_NO_SG_MERGE

2014-08-08 Thread Greg KH
On Fri, Aug 08, 2014 at 11:03:41AM -0400, Jeff Moyer wrote: > Hi, > > Commit 05f1dd5 introduced a new queue flag: QUEUE_FLAG_NO_SG_MERGE. > This gets set by default in blk_mq_init_queue for mq-enabled devices. > The effect of the flag is to bypass the SG segment merging. Instead, > the bio->bi_vc

[patch] dm: propagate QUEUE_FLAG_NO_SG_MERGE

2014-08-08 Thread Jeff Moyer
Hi, Commit 05f1dd5 introduced a new queue flag: QUEUE_FLAG_NO_SG_MERGE. This gets set by default in blk_mq_init_queue for mq-enabled devices. The effect of the flag is to bypass the SG segment merging. Instead, the bio->bi_vcnt is used as the number of hardware segments. With a device mapper tar