Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Mon, Jul 06 2015 at 3:44P -0400,
Ming Lin <mlin at kernel.org> wrote:
> From: Kent Overstreet <kent.overstreet at gmail.com>
>
> The way the block layer is currently written, it goes to great lengths
> to avoid having to split bios; upper layer code (such as bio_add_page())
> checks what the underlying device can handle and tries to always create
> bios that don't need to be split.
>
> But this approach becomes unwieldy and eventually breaks down with
> stacked devices and devices with dynamic limits, and it adds a lot of
> complexity. If the block layer could split bios as needed, we could
> eliminate a lot of complexity elsewhere - particularly in stacked
> drivers. Code that creates bios can then create whatever size bios are
> convenient, and more importantly stacked drivers don't have to deal with
> both their own bio size limitations and the limitations of the
> (potentially multiple) devices underneath them. In the future this will
> let us delete merge_bvec_fn and a bunch of other code.
>
> We do this by adding calls to blk_queue_split() to the various
> make_request functions that need it - a few can already handle arbitrary
> size bios. Note that we add the call _after_ any call to
> blk_queue_bounce(); this means that blk_queue_split() and
> blk_recalc_rq_segments() don't need to be concerned with bouncing
> affecting segment merging.
>
> Some make_request_fn() callbacks were simple enough to audit and verify
> they don't need blk_queue_split() calls. The skipped ones are:
>
> * nfhd_make_request (arch/m68k/emu/nfblock.c)
> * axon_ram_make_request (arch/powerpc/sysdev/axonram.c)
> * simdisk_make_request (arch/xtensa/platforms/iss/simdisk.c)
> * brd_make_request (ramdisk - drivers/block/brd.c)
> * mtip_submit_request (drivers/block/mtip32xx/mtip32xx.c)
> * loop_make_request
> * null_queue_bio
> * bcache's make_request fns
>
> Some others are almost certainly safe to remove now, but will be left
> for future patches.
>
> Cc: Jens Axboe <axboe at kernel.dk>
> Cc: Christoph Hellwig <hch at infradead.org>
> Cc: Al Viro <viro at zeniv.linux.org.uk>
> Cc: Ming Lei <ming.lei at canonical.com>
> Cc: Neil Brown <neilb at suse.de>
> Cc: Alasdair Kergon <agk at redhat.com>
> Cc: Mike Snitzer <snitzer at redhat.com>
> Cc: dm-devel at redhat.com
> Cc: Lars Ellenberg <drbd-dev at lists.linbit.com>
> Cc: drbd-user at lists.linbit.com
> Cc: Jiri Kosina <jkosina at suse.cz>
> Cc: Geoff Levand <geoff at infradead.org>
> Cc: Jim Paris <jim at jtan.com>
> Cc: Joshua Morris <josh.h.morris at us.ibm.com>
> Cc: Philip Kelleher <pjk1939 at linux.vnet.ibm.com>
> Cc: Minchan Kim <minchan at kernel.org>
> Cc: Nitin Gupta <ngupta at vflare.org>
> Cc: Oleg Drokin <oleg.drokin at intel.com>
> Cc: Andreas Dilger <andreas.dilger at intel.com>
> Acked-by: NeilBrown <neilb at suse.de> (for the 'md/md.c' bits)
> Signed-off-by: Kent Overstreet <kent.overstreet at gmail.com>
> [dpark: skip more mq-based drivers, resolve merge conflicts, etc.]
> Signed-off-by: Dongsu Park <dpark at posteo.net>
> Signed-off-by: Ming Lin <ming.l at ssi.samsung.com>
...
> diff --git a/block/blk-merge.c b/block/blk-merge.c
> index 30a0d9f..3707f30 100644
> --- a/block/blk-merge.c
> +++ b/block/blk-merge.c
> @@ -9,12 +9,158 @@
>
> #include "blk.h"
>
> +static struct bio *blk_bio_discard_split(struct request_queue *q,
> + struct bio *bio,
> + struct bio_set *bs)
> +{
> + unsigned int max_discard_sectors, granularity;
> + int alignment;
> + sector_t tmp;
> + unsigned split_sectors;
> +
> + /* Zero-sector (unknown) and one-sector granularities are the same. */
> + granularity = max(q->limits.discard_granularity >> 9, 1U);
> +
> + max_discard_sectors = min(q->limits.max_discard_sectors, UINT_MAX >> 9);
> + max_discard_sectors -= max_discard_sectors % granularity;
> +
> + if (unlikely(!max_discard_sectors)) {
> + /* XXX: warn */
> + return NULL;
> + }
> +
> + if (bio_sectors(bio) <= max_discard_sectors)
> + return NULL;
> +
> + split_sectors = max_discard_sectors;
> +
> + /*
> + * If the next starting sector would be misaligned, stop the discard at
> + * the previous aligned sector.
> + */
> + alignment = (q->limits.discard_alignment >> 9) % granularity;
> +
> + tmp = bio->bi_iter.bi_sector + split_sectors - alignment;
> + tmp = sector_div(tmp, granularity);
> +
> + if (split_sectors > tmp)
> + split_sectors -= tmp;
> +
> + return bio_split(bio, split_sectors, GFP_NOIO, bs);
> +}
This code to stop the discard at the previous aligned sector could be
the reason why I have 2 device-mapper-test-suite tests in the
'thin-provisioning' testsuite failing due to this patchset:
4) Failure:
test_discard_a_fragmented_device(DiscardQuickTests)
[/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:49:in `block in assert_no_mappings'
/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:43:in `call'
/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:43:in `block in with_dev_md'
/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:40:in `each'
/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:40:in `with_dev_md'
/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:48:in `assert_no_mappings'
/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:263:in `test_discard_a_fragmented_device']:
<0> expected but was
<131072>.
5) Failure:
test_discard_fully_provisioned_device_benchmark(DiscardQuickTests)
[/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:49:in `block in assert_no_mappings'
/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:43:in `call'
/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:43:in `block in with_dev_md'
/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:40:in `each'
/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:40:in `with_dev_md'
/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:48:in `assert_no_mappings'
/root/snitm/git/device-mapper-test-suite/lib/dmtest/tests/thin-provisioning/discard_tests.rb:225:in `test_discard_fully_provisioned_device_benchmark']:
<0> expected but was
<262144>.
But I haven't yet figured out where the blk_bio_discard_split code, or
elsewhere, is going wrong.
Mike