pm24.git/include/linux/bio.h, branch v2.6.30-rc2

block: Add flag for telling the IO schedulers NOT to anticipate more IO

2009-04-06T15:04:54Z

By default, CFQ will anticipate more IO from a given io context if the previously completed IO was sync. This used to be fine, since the only sync IO was reads and O_DIRECT writes. But with more "normal" sync writes being used now, we don't want to anticipate for those. Add a bio/request flag that informs the IO scheduler that this is a sync request that we should not idle for. Introduce WRITE_ODIRECT specifically for O_DIRECT writes, and make sure that the other sync writes set this flag. Signed-off-by: Jens Axboe Signed-off-by: Linus Torvalds

block: add private bio_set for bio integrity allocations

2009-03-24T11:35:17Z

The integrity bio allocation needs its own bio_set to avoid violating the mempool allocation rules and risking deadlocks. Signed-off-by: Martin K. Petersen Signed-off-by: Jens Axboe

block: Add gfp_mask parameter to bio_integrity_clone()

2009-03-14T20:06:51Z

Stricter gfp_mask might be required for clone allocation. For example, request-based dm may clone bio in interrupt context so it has to use GFP_ATOMIC. Signed-off-by: Kiyoshi Ueda Signed-off-by: Jun'ichi Nomura Acked-by: Martin K. Petersen Cc: Alasdair G Kergon Signed-off-by: Jens Axboe

block: fix bad definition of BIO_RW_SYNC

2009-02-18T09:32:00Z

We can't OR shift values, so get rid of BIO_RW_SYNC and use BIO_RW_SYNCIO and BIO_RW_UNPLUG explicitly. This brings back the behaviour from before 213d9417fec62ef4c3675621b9364a667954d4dd. Signed-off-by: Jens Axboe

bio.h: If they MUST be inlined, then use __always_inline

2009-02-02T11:41:07Z

bvec_kmap_irq() and bvec_kunmap_irq() comments say they MUST be inlined, so mark them as __always_inline. Signed-off-by: Alberto Bertogli Signed-off-by: Jens Axboe

Fix misleading comment in bio.h

2009-02-02T11:41:07Z

The comment says "remember to add offset!", but the function already adds it. Signed-off-by: Alberto Bertogli Signed-off-by: Jens Axboe

block: add bio_rw_flagged() for testing bio->bi_rw

2009-01-30T11:34:37Z

The existing functions for checking bio->bi_rw are badly named. So lets mirror what we do for bio->bi_flags testing, use a properly named function so that it's immediately obvious what is being tested. Maintain compatability names for the old macros, eventually we'll get rid of these. Signed-off-by: Jens Axboe

block: seperate bio/request unplug and sync bits

2009-01-30T11:34:37Z

Signed-off-by: Jens Axboe

Fix small typo in bio.h's documentation

2009-01-30T11:34:37Z

Signed-off-by: Alberto Bertogli Signed-off-by: Jens Axboe

block: Don't verify integrity metadata on read error

2009-01-30T11:34:36Z

If we get an I/O error on a read request there is no point in doing a verify pass on the integrity buffer. Adjust the completion path accordingly. Signed-off-by: Martin K. Petersen Signed-off-by: Jens Axboe