Re: [PATCH 2/2] bdev: Refresh bdev size for disks without partitioning

else >>>> +    rescan_partitions(bdev->bd_disk, bdev); >>> Maybe use the new common helper to replace above. >> What do you mean exactly? Because there's only this place that has the >> code >> pattern here... > > The

Re: [PATCH 1/2 v2] nvme: Cleanup and rename nvme_block_nr()

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE Software Solutions Germany GmbH Maxfeldstr. 5 90409 Nürnberg Germany (HRB 36809, AG Nürnberg

Re: [PATCH 2/2 v2] nvme: Introduce nvme_lba_to_sect()

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE Software Solutions Germany GmbH Maxfeldstr. 5 90409 Nürnberg Germany (HRB 36809, AG Nürnberg

Re: [PATCH v3 5/7] block: Delay default elevator initialization

if (register_queue) > blk_register_queue(disk); That hunk looks unrelated, but anyways: Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE Softwar

Re: [PATCH v3 7/7] sd: Set ELEVATOR_F_ZBD_SEQ_WRITE for ZBC disks

Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE Software Solutions Germany GmbH Maxfeldstr. 5 90409 Nürnberg Germany (HRB 247165, AG München) Key fingerprint = EC38

Re: [PATCH v3 6/7] block: Set ELEVATOR_F_ZBD_SEQ_WRITE for nullblk zoned disks

Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE Software Solutions Germany GmbH Maxfeldstr. 5 90409 Nürnberg Germany (HRB 247165, AG München) Key fingerprint = EC38

Re: [PATCH v3 4/7] block: Improve default elevator selection

Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE Software Solutions Germany GmbH Maxfeldstr. 5 90409 Nürnberg Germany (HRB 247165, AG München) Key fingerprint = EC38

Re: [PATCH v2 3/7] block: Introduce elevator features

My brain autocompleted to 'unsigned long' when I read it. Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE Software Solutions Germany GmbH Maxfeldstr. 5 90409 Nü

Re: [PATCH v2 3/7] block: Introduce elevator features

On 28/08/2019 12:41, Damien Le Moal wrote: > On 2019/08/28 17:16, Johannes Thumshirn wrote: >> What happened to my review comment for v1 of this patch? >> > > I merged the renamed ELEVATOR_F_ZBD_SEQ_WRITE feature into this patch instead > of > following patch and separ

Re: [PATCH v2 3/7] block: Introduce elevator features

What happened to my review comment for v1 of this patch? -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE Software Solutions Germany GmbH Maxfeldstr. 5 90409 Nürnberg Germany (HRB 247165, AG München

Re: [PATCH 5/7] block: Introduce zoned block device elevator feature

ith the feature bits and then one pulling the requirement of ELEVATOR_F_ZONED_BLOCK_DEV in null_blk and sd. -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Fel

Re: [PATCH 4/7] block: Introduce elevator features

nsigned long elevator_features; Wouldn't it make more sens to have a defined number of bits for the features 'unsigned long' means 64 possible features on 64bit and 32 on 32 bit? Better make it explicit (u64 or u32 depending on how many features you think you'll need). Byte,

Re: [PATCH 3/7] block: Remove sysfs lock from elevator_init_rq()

On 23/08/2019 02:15, Damien Le Moal wrote: > Since elevator_init_rq() is called before the device queue is registered elevator_init_mq() ~^ (and in Subject) Otherwise: Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse

Re: [PATCH 2/7] block: Change elevator_init_mq() to always succeed

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH 1/7] block: Cleanup elevator_init_mq() use

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH blktests] Make the NVMe tests more reliable

On Tue, Aug 06, 2019 at 08:11:02AM -0700, Bart Van Assche wrote: > On 8/6/19 1:11 AM, Johannes Thumshirn wrote: > > On 06/08/2019 01:25, Bart Van Assche wrote: > > [...] > > > > > + for ((i=0;i<10;i++)); do > > > +

Re: [PATCH blktests] Make the NVMe tests more reliable

_nvme_loop_dev() does anything with the return value of the function. They expect either a nvme-device or an empty string and fail if the string is empty due to a non-empty diff in the golden output. Thanks, Johannes -- Johannes ThumshirnSUSE Labs Filesystem

Re: [PATCH] nbd: replace kill_bdev() with __invalidate_device() again

On Wed, Jul 31, 2019 at 11:15:46AM +0800, SunKe wrote: > CR: https://code.amazon.com/reviews/CR-7629288 Hi, this link isn't accessible for ordinary people, please remove it from the patch. -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...

Re: [PATCH 3/5] block: Micro-optimize bvec_split_segs()

Looks good, Reviewed-by: Johannes Thumshirn Although I'm always interested in performance numbers when a patch claims to (micro) optimize something. -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689

Re: [PATCH 2/5] block: Document the bio splitting functions

Thanks a lot Bart, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284

Re: [PATCH 1/5] block: Declare several function pointer arguments 'const'

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH blktests v2 11/12] common: Use sysfs instead of modinfo for _have_module_param()

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH blktests v2 12/12] nvme: Ignore errors when removing modules

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH blktests v2 09/12] nvme: Cleanup modprobe lines into helper functions

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH blktests v2 10/12] nvme: Ensure all ports and subsystems are removed on cleanup

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH blktests v2 08/12] check: Add the ability to call a cleanup function after a test ends

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH blktests v2 07/12] nvme/018: Ignore error message generated by nvme read

wise, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG Nürnberg) K

Re: [PATCH blktests v2 06/12] nvme/015: Ensure the namespace is flushed not the char device

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH blktests v2 04/12] nvme/003,004: Add missing calls to nvme disconnect

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH blktests v2 05/12] nvme/005: Don't rely on modprobing to set the multipath paramater

Looks good, eviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH blktests v2 03/12] nvme: Add new test to verify the generation counter

ard-core the transport type in a library function? Maybe sth like this: _discovery_genctr() { local trtype=$1 nvme discover -t ${trtype} | sed -n -e 's/^.*Generation counter \([0-9]\+\).*$/\1/p' } I think Omar can fix it up when applying. Byte,

Re: [PATCH blktests v2 01/12] Add filter function for nvme discover

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH blktests v2 02/12] nvme: More agressively filter the discovery output

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH blktests 11/12] common: Use sysfs instead of modinfo for _have_module_param()

to check if "/sys/module/$1" exists and if it does check for "/sys/module/$1/parameters/$2", if not try modinfo. Does that make sense? Byte, Johannes -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+4

Re: [PATCH blktests 07/12] nvme/018: Ignore error message generated by nvme read]

t accounted for by test 018 and caused it to fail. > > This test does not need to test the error message content, it's > only important that a read past the end of the file fails. Therefore, > pipe stderr of nvme-cli to /dev/null. How about redirecting all of the outpu

Re: [PATCH] block: Disable write plugging for zoned block devices

Not sure if I like the new helper or I'd prefer another 'else' in blk_mq_make_request(). But all variants I can come up with are ugly and disgusting, so let's got the route you proposed. Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnS

Re: [PATCH 1/3] block: improve the gap check in __bio_add_pc_page

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH 2/3] block: create bio_try_merge_pc_page helper __bio_add_pc_page

but admittedly that's pure bikeshedding here. -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 2128

Re: [PATCH 2/3] block: create bio_try_merge_pc_page helper __bio_add_pc_page

queue_max_segment_size(q)) > return false; > - > - return true; > + return __bio_try_merge_page(bio, page, len, offset, same_page); > } That's a lot of spurious whitespace changes here. -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@su

Re: [PATCH 1/3] nvme: 002: fix nvmet pass data with loop

ch makes it impractical. -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG Nürnberg) Key fingerp

Re: [PATCH 8/8] block: never take page references for ITER_BVEC

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH 7/8] direct-io: use bio_release_pages in dio_bio_complete

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH 6/8] block_dev: use bio_release_pages in __blkdev_direct_IO_simple

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH 4/8] iomap: use bio_release_pages in iomap_dio_bio_end_io

On Sat, May 04, 2019 at 02:33:09PM +0200, Johannes Thumshirn wrote: > > Shouldn't this rather be: > > if (!bio_flagged(bio, BIO_NO_PAGE_REF)) > bio_release_pages(bio); OK I apparently can't remember between 3 emails, sorry Review

Re: [PATCH 3/8] block: use bio_release_pages in bio_map_user_iov

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH 5/8] block_dev: use bio_release_pages in blkdev_bio_end_io

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH 4/8] iomap: use bio_release_pages in iomap_dio_bio_end_io

bio_release_pages(bio); Shouldn't this rather be: if (!bio_flagged(bio, BIO_NO_PAGE_REF)) bio_release_pages(bio); -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+

Re: [PATCH 1/8] block: move the BIO_NO_PAGE_REF check into bio_release_pages

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH 1/5] block: rewrite blk_bvec_map_sg to avoid a nth_page call

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH 3/5] block: don't allow multiple bio_iov_iter_get_pages calls per bio

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

Re: [PATCH 2/5] block: refactor __bio_iov_bvec_add_pages

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG

[PATCH] block: bio: ensure newly added bio flags don't override BVEC_POOL_IDX

With the introduction of BIO_NO_PAGE_REF we've used up all available bits in bio::bi_flags. Convert the defines of the flags to an enum and add a BUILD_BUG_ON() call to make sure no-one adds a new one and thus overrides the BVEC_POOL_IDX causing crashes. Signed-off-by: Johannes Thum

Re: [block] 4e366a7962: blktests.nvme.005.fail

m/ I'll have a look, thanks. [...] > Can you trigger this issue on linus tree or 5.2-tmp of block tree? I haven't been able to trigger it on vanilla so far, so it's most likely related to the patch. Will rebase the patchset to 5.2-tmp and retest from there. Thanks, Joh

Re: [block] 4e366a7962: blktests.nvme.005.fail

On 28/03/2019 18:52, Jens Axboe wrote: > On 3/28/19 11:48 AM, Christoph Hellwig wrote: >> On Thu, Mar 28, 2019 at 05:31:29PM +0100, Johannes Thumshirn wrote: >>> In all cases I could reproduce it I had nsegs == 2 and >>> rq->nr_phys_segments == 1 (rq->bio->b

Re: [block] 4e366a7962: blktests.nvme.005.fail

flag and thus we never run out of sync. Adding a call to blk_recount_segments() doesn't do any difference here. It can of be a read herring as well. What makes me suspicious is the fact that I can't reliably trigger it by running nvme/005 but have to run it in a loop ranging bet

Re: [PATCH 03/15] block: add a rq_dma_dir helper

f it :) Fair enough. Byte, Johannes -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG Nürnb

Re: [PATCH v2 1/3] block: bio: kill BIO_SEG_VALID flag

rigger > 100 runs of nvme/005 and I still don't understand what's happening and why. Byte, Johannes -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409

Re: [PATCH v2 1/3] block: bio: kill BIO_SEG_VALID flag

(bio->bi_phys_segments == -1) > bio->bi_phys_segments = 0; > bio->bi_phys_segments++; Yeah, that looks more obvious -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUS

[PATCH v2 0/3] add flag for tracking bio allocation

..@suse.de/ Jens Axboe (1): block: bio: kill BIO_SEG_VALID flag Johannes Thumshirn (2): block: bio: ensure newly added bio flags don't override BVEC_POOL_IDX block: bio: introduce BIO_ALLOCED flag and check it in bio_free block/bio.c | 17 - block/blk-merge

[PATCH v2 1/3] block: bio: kill BIO_SEG_VALID flag

ts ] Signed-off-by: Johannes Thumshirn --- block/bio.c | 15 +-- block/blk-core.c | 1 + block/blk-merge.c | 13 ++--- drivers/md/raid5.c| 2 +- include/linux/blk_types.h | 1 - 5 files changed, 17 insertions(+), 15 deletions(-) diff --g

[PATCH v2 2/3] block: bio: ensure newly added bio flags don't override BVEC_POOL_IDX

With the introduction of BIO_ALLOCED we've used up all available bits in bio::bi_flags. Make sure no-one adds a new one and thus overrides the BVEC_POOL_IDX Signed-off-by: Johannes Thumshirn --- block/bio.c | 3 +++ include/linux/blk_types.h | 25 ++-

[PATCH v2 3/3] block: bio: introduce BIO_ALLOCED flag and check it in bio_free

7;BIO_ALLOCATED' and skip freeing if the flag isn't set. Fixes: 189ce2b9dcc3 ("block: fast-path for small and simple direct I/O requests") Signed-off-by: Johannes Thumshirn --- block/bio.c | 4 include/linux/blk_types.h | 1 + 2 files changed, 5 insert

Re: [PATCH 03/15] block: add a rq_dma_dir helper

sible users to use the new helper in this patch as well? -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HR

Re: [PATCH 1/2] block: bio: ensure newly added bio flags don't override BVEC_POOL_IDX

On 21/03/2019 15:25, Jens Axboe wrote: > On 3/21/19 8:23 AM, Johannes Thumshirn wrote: >> On 21/03/2019 15:21, Jens Axboe wrote: >>> On 3/21/19 8:15 AM, Jens Axboe wrote: >>>> You also haven't solved the issue of now having an extra bit, 2/2 uses >>>

Re: [PATCH 1/2] block: bio: ensure newly added bio flags don't override BVEC_POOL_IDX

se > ->bi_phys_segments to tell if it's valid or not. This patch uses -1 to > signify it's not. > > Totally untested... That sounds like an idea, I'll trow some testing at it and report back. -- Johannes ThumshirnSUSE Labs Filesystems jt

Re: [PATCH 1/2] block: bio: ensure newly added bio flags don't override BVEC_POOL_IDX

ly reason and it is not wrong to do and not the only place where we have it: jthumshirn@glass:linux (master)$ git grep -A 1 "enum {" | grep "= 0" | wc -l 2206 -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de

[PATCH 1/2] block: bio: ensure newly added bio flags don't override BVEC_POOL_IDX

With the introduction of BIO_ALLOCED we've used up all available bits in bio::bi_flags. Make sure no-one adds a new one and thus overrides the BVEC_POOL_IDX Signed-off-by: Johannes Thumshirn --- block/bio.c | 8 include/linux/blk_types.h

[PATCH 2/2] bio: introduce BIO_ALLOCED flag and check it in bio_free

7;BIO_ALLOCATED' and skip freeing if the flag isn't set. Fixes: 189ce2b9dcc3 ("block: fast-path for small and simple direct I/O requests") Signed-off-by: Johannes Thumshirn --- block/bio.c | 4 include/linux/blk_types.h | 1 + 2 files changed, 5 insert

[PATCH 0/2] add flag for tracking bio allocation

C_POOL_IDX(). Note, Jens also staged a patch in his io_uring-next branch taking the last flag. For this reason patch 2/2 might not be applied in this form, but 1/2 is still applicable then. [1] https://lore.kernel.org/linux-block/20190320081253.129688-1-h...@suse.de/ Johannes Thumshirn (2):

Re: [PATCH] block_dev: fix crash on chained bios with O_DIRECT

0 as well. > That said, I do greatly prefer your approach to solving the issue. Any ideas to proceed from here? -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nür

Re: [PATCH] block_dev: fix crash on chained bios with O_DIRECT

On 20/03/2019 12:47, Jan Kara wrote: > On Wed 20-03-19 09:53:10, Johannes Thumshirn wrote: >> On 20/03/2019 09:51, Hannes Reinecke wrote: >>> Yeah, should work, too. >>> But we should be calling bio_uninit() for all bios. >> >> Yup, probably. >

Re: [PATCH] block_dev: fix crash on chained bios with O_DIRECT

On 20/03/2019 09:51, Hannes Reinecke wrote: > Yeah, should work, too. > But we should be calling bio_uninit() for all bios. Yup, probably. > Will you be sending an updated patch? Let's wait what other's thing first. -- Johannes ThumshirnSU

Re: [PATCH] block_dev: fix crash on chained bios with O_DIRECT

tested patch: >From 9c8434e5bf81595e97ea5647437d12bfce0e37b6 Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Wed, 20 Mar 2019 09:40:18 +0100 Subject: [PATCH] bio: Introduce BIO_ALLOCED flag and check it in bio_free When we're submitting a bio from stack and this ends up being split, we call bio_put(). bio_put() w

Re: [PATCH] block: optimize bvec iteration in bvec_iter_advance

this? ^ > because they do classic page-based iteration, or it is larger because > the caller operates on multi-page bvecs. > > This should help optimizing shaving off a few cycles of the I/O hot optimizing || shaving off, both together sounds awkward. Otherwise Reviewed-by: Johannes Thu

Re: [LSF/MM TOPIC] Software RAID Support for NV-DIMM

res already doing data duplication over interleave sets of NV-DIMMs. And if you carve out a bit of your pmem space into an own namespace for the metadata (did I understand you right here?) you still have the problem that all data written to the DIMMs is interleaved in an interlea

[LSF/MM TOPIC] Software RAID Support for NV-DIMM

s, but they are incompatible with DAX. In this session Hannes and I would like to discuss eventual ways how we as an operating system can mitigate these issues for our users. Byte, Johannes -- Johannes ThumshirnSUSE Labs Filesystems jth

Re: [LSF/MM TOPIC] improving storage testing

. Just my two cents, Johannes -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG Nürnberg) Key finge

Re: [LSF/MM TOPIC] improving storage testing

blktests test need filtering. See [1] as an example. [1] https://github.com/osandov/blktests/pull/34 Byte, Johannes -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90

Re: [PATCH] block: avoid setting wbt_lat_usec to current value

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG

Re: [PATCH] blktrace: Show requests without sector

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG

Re: [LSF/MM TOPIC] : blktests: status, an expansion plan for the storage stack test framework

e also lacking tests for things like ioprio, persistent reservation, bcache and so on. Adding support for collecting gcov information after running a test case would also be awesome (this is missing in xfstests as well). So I think a session on blktests can help us get the gap closed. Byte,

Re: [PATCH v2 1/3] blk-iolatency: fix IO hang due to negative inflight counter

> > xfs_io -f -d -c "pwrite 0 4k" /dev/sdg > > echo "`cat /sys/block/sdg/dev` target=100" > $P/io.latency > > xfs_io -f -d -c "pwrite 0 4k" /dev/sdg Can you please turn this into a testcase for blktests as well? Thanks, Johan

Re: [PATCH blktests 1/2] nvme: enable buffered_io for file-backed namespaces

default to the > buffered_io mode. We could also introduce a check if we can use direct IO on the underlying FS, but the only way I can think of from a shell script is doing a 'dd of=$TMPDIR/testfile oflag=direct' and then populate the result to the helper functions. Chaitanya, Omar any id

[PATCH blktests 2/2] nvme: test resize of a namespace with file-backed ns

Test resizing of a NVMe namespace by creating a file backed namespace over nvme-loop with 1G size, connecting to it and then resizing it to 2G. Check if /proc/partitions and blkdev --getsz $DEVICE see the updated size. Signed-off-by: Johannes Thumshirn --- tests/nvme/029 | 67

[PATCH blktests 1/2] nvme: enable buffered_io for file-backed namespaces

arget is file-based enable the use of buffered I/O. Signed-off-by: Johannes Thumshirn --- tests/nvme/rc | 7 +-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/tests/nvme/rc b/tests/nvme/rc index eff1dd992460..ec92e41396be 100644 --- a/tests/nvme/rc +++ b/tests/nvme/rc @@ -48,7

Re: [PATCH blktests 02/14] common: Introduce _test_dev_is_zoned() helper function

ned_file variable out if grep -qe "none" "${TEST_DEV_SYSFS}/queue/zoned"; then [...] works as well and is a bit more descriptive -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 S

Re: [PATCH v3 2/2] trace nvme submit queue status

Is the addition of a trace_pci.h really needed? Why can't you just put it in trace.h? Byte, Johannes -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nür

Re: [PATCH 0/4] nvme multipath: expose slaves/holders

"lsblk: try device/dev to read devno") [2] https://github.com/osandov/blktests/blob/master/common/rc#L185 Byte, Johannes -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfel

Re: [PATCH 2/3] blktests: add python scripts for parsing fio json output

- Repository : openSUSE:Leap:15.0 Name : jq Version: 1.5-lp150.1.10 Arch : x86_64 Vendor : openSUSE Installed Size : 93.4 KiB Installed : Yes [...] -- Johannes ThumshirnSUSE Labs Filesystems jthumsh...@suse.de

Re: [PATCH] block: update documentation

On 28/11/2018 02:25, Damien Le Moal wrote: > Jens Axboe , February 2009 I guess that address of Jens' doesn't work anymore either. -- Johannes ThumshirnSUSE Labs jthumsh...@suse.de+49 911 74053 689 SU

Re: [PATCH 04/20] block: wire up block device iopoll method

Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG Nürnberg

Re: [PATCH 03/20] fs: add an iopoll method to struct file_operations

Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG Nürnberg

Re: [PATCH 02/20] aio: clear IOCB_HIPRI

Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG Nürnberg

Re: [PATCH 01/20] aio: fix failure to put the file pointer

Reivewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG Nürnberg

Re: [PATCH 6/7] block: prevent merging of requests with different priorities

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG

Re: [PATCH 5/7] aio: Fix fallback I/O priority value

Looks good, Reiewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG

Re: [PATCH 4/7] block: Introduce get_current_ioprio()

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG

Re: [PATCH 1/7] aio: Comment use of IOCB_FLAG_IOPRIO aio flag

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG

Re: [PATCH 2/7] block: Remove bio->bi_ioc

Looks good, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard, Graham Norton HRB 21284 (AG

Re: [PATCH 7/7] block: remove the BLKPREP_* values.

So long and thanks for all the fish, Reviewed-by: Johannes Thumshirn -- Johannes ThumshirnSUSE Labs jthumsh...@suse.de+49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Jane Smithard

  1   2   3   4   5   6   >