Skip to content

L
linux
Switch branch/tag
  1. 10 Feb, 2015 2 commits
  3. 04 Feb, 2015 1 commit
    • Peter Zijlstra's avatar
      block: Simplify bsg complete all · 2c561246
      Peter Zijlstra authored 
      It took me a few tries to figure out what this code did; lets rewrite
it into a more regular form.

The thing that makes this one 'special' is the BSG_F_BLOCK flag, if
that is not set we're not supposed/allowed to block and should spin
wait for completion.

The (new) io_wait_event() will never see a false condition in case of
the spinning and we will therefore not block.

Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: default avatarPeter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      2c561246
  5. 03 Feb, 2015 1 commit
  7. 29 Jan, 2015 1 commit
    • Jens Axboe's avatar
      NVMe: avoid kmalloc/kfree for smaller IO · ac3dd5bd
      Jens Axboe authored 
      Currently we allocate an nvme_iod for each IO, which holds the
sg list, prps, and other IO related info. Set a threshold of
2 pages and/or 8KB of data, below which we can just embed this
in the per-command pdu in blk-mq. For any IO at or below
NVME_INT_PAGES and NVME_INT_BYTES, we save a kmalloc and kfree.

For higher IOPS, this saves up to 1% of CPU time.
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
Reviewed-by: default avatarKeith Busch <keith.busch@intel.com>
      ac3dd5bd
  9. 28 Jan, 2015 1 commit
  11. 26 Jan, 2015 1 commit
  13. 25 Jan, 2015 1 commit
  15. 23 Jan, 2015 4 commits
    • Shaohua Li's avatar
      libata: use blk taging · 12cb5ce1
      Shaohua Li authored 
      libata uses its own tag management which is duplication and the
implementation is poor. And if we switch to blk-mq, tag is build-in.
It's time to switch to generic taging.

The SAS driver has its own tag management, and looks we can't directly
map the host controler tag to SATA tag. So I just bypassed the SAS case.

I changed the code/variable name for the tag management of libata to
make it self contained. Only sas will use it. Later if libsas implements
its tag management, the tag management code in libata can be deleted
easily.

Cc: Jens Axboe <axboe@fb.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: default avatarShaohua Li <shli@fb.com>
Acked-by: default avatarTejun Heo <tj@kernel.org>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      12cb5ce1
    • Jens Axboe's avatar
      Merge branch 'for-3.20/core' into for-3.20/drivers · a4a1cc16
      Jens Axboe authored 
      We need the tagging changes for the libata conversion.
      a4a1cc16
    • Shaohua Li's avatar
      blk-mq: add tag allocation policy · 24391c0d
      Shaohua Li authored 
      This is the blk-mq part to support tag allocation policy. The default
allocation policy isn't changed (though it's not a strict FIFO). The new
policy is round-robin for libata. But it's a try-best implementation. If
multiple tasks are competing, the tags returned will be mixed (which is
unavoidable even with !mq, as requests from different tasks can be
mixed in queue)

Cc: Jens Axboe <axboe@fb.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: default avatarShaohua Li <shli@fb.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      24391c0d
    • Shaohua Li's avatar
      block: support different tag allocation policy · ee1b6f7a
      Shaohua Li authored 
      The libata tag allocation is using a round-robin policy. Next patch will
make libata use block generic tag allocation, so let's add a policy to
tag allocation.

Currently two policies: FIFO (default) and round-robin.

Cc: Jens Axboe <axboe@fb.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: default avatarShaohua Li <shli@fb.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      ee1b6f7a
  17. 22 Jan, 2015 2 commits
  19. 21 Jan, 2015 2 commits
    • Martin K. Petersen's avatar
      block: Add discard flag to blkdev_issue_zeroout() function · d93ba7a5
      Martin K. Petersen authored 
      blkdev_issue_discard() will zero a given block range. This is done by
way of explicit writing, thus provisioning or allocating the blocks on
disk.

There are use cases where the desired behavior is to zero the blocks but
unprovision them if possible. The blocks must deterministically contain
zeroes when they are subsequently read back.

This patch adds a flag to blkdev_issue_zeroout() that provides this
variant. If the discard flag is set and a block device guarantees
discard_zeroes_data we will use REQ_DISCARD to clear the block range. If
the device does not support discard_zeroes_data or if the discard
request fails we will fall back to first REQ_WRITE_SAME and then a
regular REQ_WRITE.

Also update the callers of blkdev_issue_zero() to reflect the new flag
and make sb_issue_zeroout() prefer the discard approach.
Signed-off-by: default avatarMartin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: default avatarChristoph Hellwig <hch@lst.de>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      d93ba7a5
    • Jeff Moyer's avatar
      cfq-iosched: fix incorrect filing of rt async cfqq · c6ce1943
      Jeff Moyer authored 
      Hi,

If you can manage to submit an async write as the first async I/O from
the context of a process with realtime scheduling priority, then a
cfq_queue is allocated, but filed into the wrong async_cfqq bucket.  It
ends up in the best effort array, but actually has realtime I/O
scheduling priority set in cfqq->ioprio.

The reason is that cfq_get_queue assumes the default scheduling class and
priority when there is no information present (i.e. when the async cfqq
is created):

static struct cfq_queue *
cfq_get_queue(struct cfq_data *cfqd, bool is_sync, struct cfq_io_cq *cic,
	      struct bio *bio, gfp_t gfp_mask)
{
	const int ioprio_class = IOPRIO_PRIO_CLASS(cic->ioprio);
	const int ioprio = IOPRIO_PRIO_DATA(cic->ioprio);

cic->ioprio starts out as 0, which is "invalid".  So, class of 0
(IOPRIO_CLASS_NONE) is passed to cfq_async_queue_prio like so:

		async_cfqq = cfq_async_queue_prio(cfqd, ioprio_class, ioprio);

static struct cfq_queue **
cfq_async_queue_prio(struct cfq_data *cfqd, int ioprio_class, int ioprio)
{
        switch (ioprio_class) {
        case IOPRIO_CLASS_RT:
                return &cfqd->async_cfqq[0][ioprio];
        case IOPRIO_CLASS_NONE:
                ioprio = IOPRIO_NORM;
                /* fall through */
        case IOPRIO_CLASS_BE:
                return &cfqd->async_cfqq[1][ioprio];
        case IOPRIO_CLASS_IDLE:
                return &cfqd->async_idle_cfqq;
        default:
                BUG();
        }
}

Here, instead of returning a class mapped from the process' scheduling
priority, we get back the bucket associated with IOPRIO_CLASS_BE.

Now, there is no queue allocated there yet, so we create it:

		cfqq = cfq_find_alloc_queue(cfqd, is_sync, cic, bio, gfp_mask);

That function ends up doing this:

			cfq_init_cfqq(cfqd, cfqq, current->pid, is_sync);
			cfq_init_prio_data(cfqq, cic);

cfq_init_cfqq marks the priority as having changed.  Then, cfq_init_prio
data does this:

	ioprio_class = IOPRIO_PRIO_CLASS(cic->ioprio);
	switch (ioprio_class) {
	default:
		printk(KERN_ERR "cfq: bad prio %x\n", ioprio_class);
	case IOPRIO_CLASS_NONE:
		/*
		 * no prio set, inherit CPU scheduling settings
		 */
		cfqq->ioprio = task_nice_ioprio(tsk);
		cfqq->ioprio_class = task_nice_ioclass(tsk);
		break;

So we basically have two code paths that treat IOPRIO_CLASS_NONE
differently, which results in an RT async cfqq filed into a best effort
bucket.

Attached is a patch which fixes the problem.  I'm not sure how to make
it cleaner.  Suggestions would be welcome.
Signed-off-by: default avatarJeff Moyer <jmoyer@redhat.com>
Tested-by: default avatarHidehiro Kawai <hidehiro.kawai.ez@hitachi.com>
Cc: stable@kernel.org
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      c6ce1943
  21. 16 Jan, 2015 1 commit
    • Jens Axboe's avatar
      null_blk: suppress invalid partition info · 227290b4
      Jens Axboe authored 
      null_blk is partitionable, but it doesn't store any of the info. When
it is loaded, you would normally see:

[1226739.343608]  nullb0: unknown partition table
[1226739.343746]  nullb1: unknown partition table

which can confuse some people. Add the appropriate gendisk flag
to suppress this info.
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      227290b4
  23. 14 Jan, 2015 6 commits
    • Jens Axboe's avatar
      blk-mq: fix false negative out-of-tags condition · 0bf36498
      Jens Axboe authored 
      The blk-mq tagging tries to maintain some locality between CPUs and
the tags issued. The tags are split into groups of words, and the
words may not be fully populated. When searching for a new free tag,
blk-mq may look at partial words, hence it passes in an offset/size
to find_next_zero_bit(). However, it does that wrong, the size must
always be the full length of the number of tags in that word,
otherwise we'll potentially miss some near the end.

Another issue is when __bt_get() goes from one word set to the next.
It bumps the index, but not the last_tag associated with the
previous index. Bump that to be in the range of the new word.

Finally, clean up __bt_get() and __bt_get_word() a bit and get
rid of the goto in there, and the unnecessary 'wrap' variable.
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      0bf36498
    • Boaz Harrosh's avatar
      brd: Request from fdisk 4k alignment · c8fa3173
      Boaz Harrosh authored 
      Because of the direct_access() API which returns a PFN. partitions
better start on 4K boundary, else offset ZERO of a partition will
not be aligned and blk_direct_access() will fail the call.

By setting blk_queue_physical_block_size(PAGE_SIZE) we can communicate
this to fdisk and friends.

The call to blk_queue_physical_block_size() is harmless and will
not affect the Kernel behavior in any way. It is only for
communication to user-mode.

before this patch running fdisk on a default size brd of 4M
the first sector offered is 34 (BAD), but after this patch it
will be 40, ie 8 sectors aligned. Also when entering some random
partition sizes the next partition-start sector is offered 8 sectors
aligned after this patch. (Please note that with fdisk the user
can still enter bad values, only the offered default values will
be correct)

Note that with bdev-size > 4M fdisk will try to align on a 1M
boundary (above first-sector will be 2048), in any case.

CC: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: default avatarBoaz Harrosh <boaz@plexistor.com>
Reviewed-by: default avatarMartin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      c8fa3173
    • Boaz Harrosh's avatar
      brd: Fix all partitions BUGs · 937af5ec
      Boaz Harrosh authored 
      This patch fixes up brd's partitions scheme, now enjoying all worlds.

The MAIN fix here is that currently, if one fdisks some partitions,
a BAD bug will make all partitions point to the same start-end sector
ie: 0 - brd_size And an mkfs of any partition would trash the partition
table and the other partition.

Another fix is that "mount -U uuid" will not work if show_part was not
specified, because of the GENHD_FL_SUPPRESS_PARTITION_INFO flag.
We now always load without it and remove the show_part parameter.

[We remove Dmitry's new module-param part_show it is now always
 show]

So NOW the logic goes like this:
* max_part - Just says how many minors to reserve between ramX
  devices. In any way, there can be as many partition as requested.
  If minors between devices ends, then dynamic 259-major ids will
  be allocated on the fly.
  The default is now max_part=1, which means all partitions devt(s)
  will be from the dynamic (259) major-range.
  (If persistent partition minors is needed use max_part=X)
  For example with /dev/sdX max_part is hard coded 16.

* Creation of new devices on the fly still/always work:
  mknod /path/devnod b 1 X
  fdisk -l /path/devnod
  Will create a new device if [X / max_part] was not already
  created before. (Just as before)

  partitions on the dynamically created device will work as well
  Same logic applies with minors as with the pre-created ones.

TODO: dynamic grow of device size. So each device can have it's
      own size.

CC: Dmitry Monakhov <dmonakhov@openvz.org>
Tested-by: default avatarRoss Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: default avatarBoaz Harrosh <boaz@plexistor.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      937af5ec
    • Jens Axboe's avatar
      Merge branch 'for-3.20/core' into for-3.20/drivers · d4119ee0
      Jens Axboe authored
      d4119ee0
    • Matthew Wilcox's avatar
      block: Change direct_access calling convention · dd22f551
      Matthew Wilcox authored 
      In order to support accesses to larger chunks of memory, pass in a
'size' parameter (counted in bytes), and return the amount available at
that address.

Add a new helper function, bdev_direct_access(), to handle common
functionality including partition handling, checking the length requested
is positive, checking for the sector being page-aligned, and checking
the length of the request does not pass the end of the partition.
Signed-off-by: default avatarMatthew Wilcox <matthew.r.wilcox@intel.com>
Reviewed-by: default avatarJan Kara <jack@suse.cz>
Reviewed-by: default avatarBoaz Harrosh <boaz@plexistor.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      dd22f551
    • Matthew Wilcox's avatar
      axonram: Fix bug in direct_access · 91117a20
      Matthew Wilcox authored 
      The 'pfn' returned by axonram was completely bogus, and has been since
2008.
Signed-off-by: default avatarMatthew Wilcox <matthew.r.wilcox@intel.com>
Reviewed-by: default avatarJan Kara <jack@suse.cz>
Reviewed-by: default avatarMathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: stable@vger.kernel.org
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
      91117a20
  25. 02 Jan, 2015 8 commits
  27. 31 Dec, 2014 1 commit
  29. 29 Dec, 2014 1 commit
  31. 28 Dec, 2014 4 commits
  33. 27 Dec, 2014 3 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7