- 14 Nov, 2018 1 commit
-
-
Jens Axboe authored
NVMe always asks for io_queues + 1 worth of IRQ vectors, which means that even when we scale all the way down, we still ask for 2 vectors and get -ENOSPC in return if the system can't support more than 1. Getting just 1 vector is fine, it just means that we'll have 1 IO queue and 1 admin queue, with a shared vector between them. Check for this case and don't add our + 1 if it happens. Fixes: 3b6592f7 ("nvme: utilize two queue maps, one for reads and one for writes") Reported-by:
Guenter Roeck <linux@roeck-us.net> Signed-off-by:
Jens Axboe <axboe@kernel.dk>
-
- 13 Nov, 2018 1 commit
-
-
Jens Axboe authored
We can't use RQF_DONTPREP to see if we should clear ->special, as someone could have set that while inserting the request. Make sure we clear it in our ->initialize_rq_fn() helper instead. Fixes: 22ce0a7c ("ide: don't use req->special") Signed-off-by:
-
- 12 Nov, 2018 1 commit
-
-
Tetsuo Handa authored
Commit 0a42e99b ("loop: Get rid of loop_index_mutex") forgot to remove mutex_unlock(&loop_ctl_mutex) from loop_control_ioctl() when replacing loop_index_mutex with loop_ctl_mutex. Fixes: 0a42e99b ("loop: Get rid of loop_index_mutex") Reported-by:
syzbot <syzbot+c0138741c2290fc5e63f@syzkaller.appspotmail.com> Reviewed-by:
Ming Lei <ming.lei@redhat.com> Reviewed-by:
Jan Kara <jack@suse.cz> Signed-off-by:
Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by:
-
- 10 Nov, 2018 15 commits
-
-
Jens Axboe authored
The compiler rightfully complains: drivers/block/null_blk_main.c: In function ‘null_complete_rq’: drivers/block/null_blk_main.c:647:16: warning: unused variable ‘nullb’ [-Wunused-variable] struct nullb *nullb = rq->q->queuedata; ^~~~~ Fixes: 49f66136 ("nullb: remove leftover legacy request code") Signed-off-by: -
Christoph Hellwig authored
Just replace it with a field of the same name in struct ide_req. Reviewed-by:
Hannes Reinecke <hare@suse.com> Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Christoph Hellwig authored
Reviewed-by:
-
Christoph Hellwig authored
Makes the code a whole lot easier to read. Reviewed-by:
-
Christoph Hellwig authored
Add a retries field to the internal request structure instead, which gets set to zero on the first submission. Reviewed-by:
-
Christoph Hellwig authored
null_softirq_done_fn is only used for the blk-mq path, so remove the other branch. Also rename the function to better match the method name. Reviewed-by:
-
Christoph Hellwig authored
The way these functions abuse ->special to try to store the dummy request looks completely broken, given that it actually stores the original scsi command. Instead switch to ->host_scribble and store the actual dummy command. Reviewed-by:
-
YueHaibing authored
Fixes gcc '-Wunused-but-set-variable' warning: block/blk-ioc.c: In function 'put_io_context_active': block/blk-ioc.c:174:24: warning: variable 'et' set but not used [-Wunused-but-set-variable] It not used any more after commit a1ce35fa ("block: remove dead elevator code") Signed-off-by:
YueHaibing <yuehaibing@huawei.com> Signed-off-by:
-
Christoph Hellwig authored
Unused now. Reviewed-by:
Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by:
Bart Van Assche <bvanassche@acm.org> Signed-off-by:
-
Christoph Hellwig authored
Remove the last use of the old BLKPREP_* values, which get converted to BLK_STS_* later anyway. Reviewed-by:
-
Christoph Hellwig authored
Replace the old BLKPREP_* values with the BLK_STS_ ones that they are converted to later anyway. Reviewed-by:
-
Christoph Hellwig authored
There is no need to call scsi_mq_free_sgtables until we have actually allocated sgtables. Reviewed-by:
-
Christoph Hellwig authored
This just moves the prep_to_mq calls up in preparation of further removal of BLKPREP_* usage. Reviewed-by:
-
Christoph Hellwig authored
Return a blk_status_t directly, and make the code a little more compact by handling the fast path in the caller. Reviewed-by:
-
Christoph Hellwig authored
The return value is just used as a binary yes/no decision, so switch it to a bool instead of the old BLKPREP_* values returned as an int. Also clean up a few related comments. Reviewed-by:
-
- 09 Nov, 2018 12 commits
-
-
Christoph Hellwig authored
Unused now that the legacy request path is gone. Signed-off-by:
-
Christoph Hellwig authored
Use the proper helper instead of manually iterating the scatterlist, which is broken in the presence of chained S/G lists. Signed-off-by:
-
Christoph Hellwig authored
Instead create add to the icmd into struct mtip_cmd which can be unioned with the scatterlist used for the normal I/O path. Signed-off-by:
-
Christoph Hellwig authored
Merging this function into the only callers makes the code flow easier. Signed-off-by:
-
Christoph Hellwig authored
There isn't much need for this helper - we can just calculate the offset for the command header once late in the submission path and fill out the ctba and ctbau fields there. Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Christoph Hellwig authored
There is no good excuse not to use proper __le16/32 types. Signed-off-by:
-
Christoph Hellwig authored
This allows for better error propagation and simpler code. Signed-off-by:
-
Christoph Hellwig authored
Factor out a new is_stopped helper that matches the existing is_se_active helper, and merge the trivial amount of remaining code into the only caller. This also allows better error handling by returning a BLK_STS_* directly instead of explicitly calling blk_mq_end_request, and moving blk_mq_start_request closer to the actual issue to hardware. Signed-off-by:
-
Christoph Hellwig authored
We have all arguments at hand in mtip_hw_submit_io, so keep the rq to sg mapping close to the dma_map_sg call. Signed-off-by:
-
Christoph Hellwig authored
The current sx8 code spends a lot of effort dealing with the fact that tags are per-host, but there might be multiple queues. Now that the driver has been converted to blk-mq it can take care of the blk-mq tag_set concept that has been designed just for that. Signed-off-by:
-
Christoph Hellwig authored
Make the disk/queue alloc and free helpers per-port by moving the trivial loops into the callers. Signed-off-by:
-
- 08 Nov, 2018 10 commits
-
-
Jens Axboe authored
Document the fact that the strategy function passed in can control whether to continue iterating or not. Suggested-by:
-
Jens Axboe authored
Returns true if the queue currently has requests pending, false if not. DM can use this to replace the atomic_inc/dec they do per device to see if a device is busy. Reviewed-by:
Mike Snitzer <snitzer@redhat.com> Reviewed-by:
-
Jens Axboe authored
We have this functionality in sbitmap, but we don't export it in blk-mq for users of the tags busy iteration. This can be useful for stopping the iteration, if the caller doesn't need to find more requests. Reviewed-by:
-
Jan Kara authored
The nested acquisition of loop_ctl_mutex (->lo_ctl_mutex back then) has been introduced by commit f028f3b2 "loop: fix circular locking in loop_clr_fd()" to fix lockdep complains about bd_mutex being acquired after lo_ctl_mutex during partition rereading. Now that these are properly fixed, let's stop fooling lockdep. Signed-off-by:
-
Jan Kara authored
Code in loop_change_fd() drops reference to the old file (and also the new file in a failure case) under loop_ctl_mutex. Similarly to a situation in loop_set_fd() this can create a circular locking dependency if this was the last reference holding the file open. Delay dropping of the file reference until we have released loop_ctl_mutex. Reported-by:
-
Jan Kara authored
Calling blkdev_reread_part() under loop_ctl_mutex causes lockdep to complain about circular lock dependency between bdev->bd_mutex and lo->lo_ctl_mutex. The problem is that on loop device open or close lo_open() and lo_release() get called with bdev->bd_mutex held and they need to acquire loop_ctl_mutex. OTOH when loop_reread_partitions() is called with loop_ctl_mutex held, it will call blkdev_reread_part() which acquires bdev->bd_mutex. See syzbot report for details [1]. Move call to blkdev_reread_part() in __loop_clr_fd() from under loop_ctl_mutex to finish fixing of the lockdep warning and the possible deadlock. [1] https://syzkaller.appspot.com/bug?id=bf154052f0eea4bc7712499e4569505907d1588Reported-by:
syzbot <syzbot+4684a000d5abdade83fac55b1e7d1f935ef1936e@syzkaller.appspotmail.com> Signed-off-by:
-
Jan Kara authored
Calling loop_reread_partitions() under loop_ctl_mutex causes lockdep to complain about circular lock dependency between bdev->bd_mutex and lo->lo_ctl_mutex. The problem is that on loop device open or close lo_open() and lo_release() get called with bdev->bd_mutex held and they need to acquire loop_ctl_mutex. OTOH when loop_reread_partitions() is called with loop_ctl_mutex held, it will call blkdev_reread_part() which acquires bdev->bd_mutex. See syzbot report for details [1]. Move all calls of loop_rescan_partitions() out of loop_ctl_mutex to avoid lockdep warning and fix deadlock possibility. [1] https://syzkaller.appspot.com/bug?id=bf154052f0eea4bc7712499e4569505907d1588Reported-by:
-
Jan Kara authored
The call of __blkdev_reread_part() from loop_reread_partition() happens only when we need to invalidate partitions from loop_release(). Thus move a detection for this into loop_clr_fd() and simplify loop_reread_partition(). This makes loop_reread_partition() safe to use without loop_ctl_mutex because we use only lo->lo_number and lo->lo_file_name in case of error for reporting purposes (thus possibly reporting outdate information is not a big deal) and we are safe from 'lo' going away under us by elevated lo->lo_refcnt. Signed-off-by:
-
Jan Kara authored
Push loop_ctl_mutex down to loop_change_fd(). We will need this to be able to call loop_reread_partitions() without loop_ctl_mutex. Signed-off-by:
-
Jan Kara authored
Push lo_ctl_mutex down to loop_set_fd(). We will need this to be able to call loop_reread_partitions() without lo_ctl_mutex. Signed-off-by:
-
