29 Mar, 2006
2 commits
-
* 'cfq-merge' of git://brick.kernel.dk/data/git/linux-2.6-block:
[BLOCK] cfq-iosched: seek and async performance fixes
[PATCH] ll_rw_blk: fix 80-col offender in put_io_context()
[PATCH] cfq-iosched: small cfq_choose_req() optimization
[PATCH] [BLOCK] cfq-iosched: change cfq io context linking from list to tree -
replaces for_each_cpu with for_each_possible_cpu().
Signed-off-by: KAMEZAWA Hiroyuki
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
28 Mar, 2006
6 commits
-
Detect whether a given process is seeky and if so disable (mostly) the
idle window if it is. We still allow just a little idle time, just enough
to allow that process to submit a new request. That is needed to maintain
fairness across priority groups.In some cases, we could setup several async queues. This is not optimal
from a performance POV, since we want all async io in one queue to perform
good sorting on it. It also impacted sync queues, as async io got too much
slice time.Signed-off-by: Jens Axboe
-
This makes akpm more happy.
Signed-off-by: Jens Axboe
-
this is a small optimization to cfq_choose_req() in the CFQ I/O scheduler
(this function is a semi-often invoked candidate in an oprofile log):
by using a bit mask variable, we can use a simple switch() to check
the various cases instead of having to query two variables for each check.
Benefit: 251 vs. 285 bytes footprint of cfq_choose_req().
Also, common case 0 (no request wrapping) is now checked first in code.Signed-off-by: Andreas Mohr
Signed-off-by: Jens Axboe -
On setups with many disks, we spend a considerable amount of time
looking up the process-disk mapping on each queue of io. Testing with
a NULL based block driver, this costs 40-50% reduction in throughput
for 1000 disks.Signed-off-by: Jens Axboe
-
* 'for-linus' of git://brick.kernel.dk/data/git/linux-2.6-block:
[PATCH] Don't make debugfs depend on DEBUG_KERNEL
[PATCH] Fix blktrace compile with sysfs not defined
[PATCH] unused label in drivers/block/cciss.
[BLOCK] increase size of disk stat counters
[PATCH] blk_execute_rq_nowait-speedup
[PATCH] ide-cd: quiet down GPCMD_READ_CDVD_CAPACITY failure
[BLOCK] ll_rw_blk: kmalloc -> kzalloc conversion
[PATCH] kzalloc() conversion in drivers/block
[PATCH] update max_sectors documentation -
This flag should be set for a virtual device iff it is set for all underlying
devices.Signed-off-by: Neil Brown
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
27 Mar, 2006
6 commits
-
debugfs depends on sysfs, so make blktrace kconfig option depend
on that.Reported by Adrian Bunk.
Signed-off-by: Jens Axboe
-
The kernel's representation of the disk statistics uses the type unsigned
which is 32b on both 32b and 64b platforms. Unfortunately, most system
tools that work with these numbers that are exported in /proc/diskstats
including iostat read these numbers into unsigned longs. This works fine
on 32b platforms and when the number of IO transactions are small on 64b
platforms. However, when the numbers wrap on 64b platforms & you read the
numbers into unsigned longs, and compare the numbers to previous readings,
then you get an unsigned representation of a negative number. This looks
like a very large 64b number & gives you bizarre readouts in iostat:ilc4: Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s rkB/s wkB/s avgrq-sz avgqu-sz await svctm %util
ilc4: sda 5.50 0.00 143.96 0.00 307496983987862656.00 0.00 153748491993931328.00 0.00 2136028725038430.00 7.94 55.12 5.59 80.42Though fixing iostat in user space is possible, and a quick survey
indicates that several other similar tools also use unsigned longs when
processing /proc/diskstats. Therefore, it seems like a better approach
would be to extend the length of the disk_stats structure on 64b
architectures to 64b. The following patch does that. It should not affect
the operation on 32b platforms.Signed-off-by: Ben Woodard
Cc: Rick Lindsley
Signed-off-by: Andrew Morton
Signed-off-by: Jens Axboe -
Both elv_add_request() and generic_unplug_device() grab the queue lock
and disable interrupts, do that locally and use the __ variants.Signed-off-by: Andrew Morton
Signed-off-by: Jens Axboe -
Signed-off-by: Jens Axboe
-
Add blkcnt_t as the type of inode.i_blocks. This enables you to make the size
of blkcnt_t either 4 bytes or 8 bytes on 32 bits architecture with CONFIG_LSF.- CONFIG_LSF
Add new configuration parameter.
- blkcnt_t
On h8300, i386, mips, powerpc, s390 and sh that define sector_t,
blkcnt_t is defined as u64 if CONFIG_LSF is enabled; otherwise it is
defined as unsigned long.
On other architectures, it is defined as unsigned long.
- inode.i_blocks
Change the type from sector_t to blkcnt_t.Signed-off-by: Takashi Sato
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Modify well over a dozen mempool users to call mempool_create_slab_pool()
rather than calling mempool_create() with extra arguments, saving about 30
lines of code and increasing readability.Signed-off-by: Matthew Dobson
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
25 Mar, 2006
1 commit
-
this changes if() BUG(); constructs to BUG_ON() which is
cleaner, contains unlikely() and can better optimized away.Signed-off-by: Eric Sesterhenn
Signed-off-by: Adrian Bunk
24 Mar, 2006
1 commit
-
Signed-off-by: Jens Axboe
23 Mar, 2006
1 commit
-
Semaphore to mutex conversion.
The conversion was generated via scripts, and the result was validated
automatically via a script as well.Signed-off-by: Arjan van de Ven
Signed-off-by: Ingo Molnar
Acked-by: Jens Axboe
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
21 Mar, 2006
1 commit
-
Convert the kobj_map code to use a mutex instead of a semaphore. It
converts the single two users as well, genhd.c and char_dev.c.Signed-off-by: Jes Sorensen
Signed-off-by: Andrew Morton
Signed-off-by: Greg Kroah-Hartman
19 Mar, 2006
17 commits
-
Signed-off-by: Al Viro
-
Signed-off-by: Al Viro
-
Signed-off-by: Al Viro
-
Signed-off-by: Al Viro
-
Signed-off-by: Al Viro
-
Signed-off-by: Al Viro
-
Signed-off-by: Al Viro
-
Signed-off-by: Al Viro
-
Signed-off-by: Al Viro
-
Signed-off-by: Al Viro
-
We don't need to pin ->key down; ->cfqq->cfqd will do that for us.
Incidentally, that stops the leak we had - that reference was never
dropped.Signed-off-by: Al Viro
-
If somebody does a hash lookup for cfq_queue while ioprio of an async queue
is elevated, they shouldn't end up stuck with lowered ioprio when we go back.
Fix is to use ->org_ioprio{,class} in hash lookups.Signed-off-by: Al Viro
-
Signed-off-by: Al Viro
-
Signed-off-by: Al Viro
09 Mar, 2006
1 commit
-
The low level PCI DMA mapping functions should handle it in most cases.
This should fix problems with depleting the DMA zone early. The old
code used precious GFP_DMA memory in many cases where it was not needed.Signed-off-by: Andi Kleen
Cc: Jens Axboe
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
28 Feb, 2006
1 commit
-
During testing of SLES10, we encountered a hang in the CFQ io scheduler.
Turns out the deferred slice expiry logic is buggy, so remove that for
now. We could be left with an idle queue that would never wake up. So
kill that logic, always expire immediately. Also fix a potential timer
race condition.Patch looks bigger than it is, because it moves a function.
Signed-off-by: Jens Axboe
Signed-off-by: Linus Torvalds
08 Feb, 2006
3 commits
-
q->ordcolor must only be flipped on initial queueing of a hardbarrier
request.Constructing ordered sequence and requeueing used to pass through
__elv_add_request() which flips q->ordcolor when it sees a barrier
request.This patch separates out elv_insert() from __elv_add_request() and uses
elv_insert() when constructing ordered sequence and requeueing.
elv_insert() inserts the given request at the specified position and
does nothing else.Signed-off-by: Tejun Heo
Acked-by: Jens Axboe
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
When issuing an SG_IO ioctl through sd that resulted in an unrecoverable
error, a nearly infinite retry loop was discovered. This is due to the
fact that the block layer SG_IO code is not setting up rq->retries. This
patch also fixes up the sg_scsi_ioctl path.Signed-off-by: Brian King
Signed-off-by: Jens Axboe
