linux-stable/block
Martin K. Petersen 8b60e2189f Merge patch series "Add Command Duration Limits support"
Niklas Cassel <nks@flawful.org> says:

This series adds support for Command Duration Limits.
The series is based on linux tag: v6.4-rc1
The series can also be found in git: https://github.com/floatious/linux/commits/cdl-v7

=================
CDL in ATA / SCSI
=================
Command Duration Limits is defined in:
T13 ATA Command Set - 5 (ACS-5) and
T10 SCSI Primary Commands - 6 (SPC-6) respectively
(a simpler version of CDL is defined in T10 SPC-5).

CDL defines Duration Limits Descriptors (DLD).
7 DLDs for read commands and 7 DLDs for write commands.
Simply put, a DLD contains a limit and a policy.

A command can specify that a certain limit should be applied by setting
the DLD index field (3 bits, so 0-7) in the command itself.

The DLD index points to one of the 7 DLDs.
DLD index 0 means no descriptor, so no limit.
DLD index 1-7 means DLD 1-7.

A DLD can have a few different policies, but the two major ones are:
-Policy 0xF (abort), command will be completed with command aborted error
(ATA) or status CHECK CONDITION (SCSI), with sense data indicating that
the command timed out.
-Policy 0xD (complete-unavailable), command will be completed without
error (ATA) or status GOOD (SCSI), with sense data indicating that the
command timed out. Note that the command will not have transferred any
data to/from the device when the command timed out, even though the
command returned success.

Regardless of the CDL policy, in case of a CDL timeout, the I/O will
result in a -ETIME error to user-space.

The DLDs are defined in the CDL log page(s) and are readable and writable.
Reading and writing the CDL DLDs are outside the scope of the kernel.
If a user wants to read or write the descriptors, they can do so using a
user-space application that sends passthrough commands, such as cdl-tools:
https://github.com/westerndigitalcorporation/cdl-tools

================================
The introduction of ioprio hints
================================
What the kernel does provide, is a method to let I/O use one of the CDL DLDs
defined in the device. Note that the kernel will simply forward the DLD index
to the device, so the kernel currently does not know, nor does it need to know,
how the DLDs are defined inside the device.

The way that the CDL DLD index is supplied to the kernel is by introducing a
new 10 bit "ioprio hint" field within the existing 16 bit ioprio definition.

Currently, only 6 out of the 16 ioprio bits are in use, the remaining 10 bits
are unused, and are currently explicitly disallowed to be set by the kernel.

For now, we only add ioprio hints representing CDL DLD index 1-7. Additional
ioprio hints for other QoS features could be defined in the future.

A theoretical future work could be to make an I/O scheduler aware of these
hints. E.g. for CDL, an I/O scheduler could make use of the duration limit
in each descriptor, and take that information into account while scheduling
commands. Right now, the ioprio hints will be ignored by the I/O schedulers.

==============================
How to use CDL from user-space
==============================
Since CDL is mutually exclusive with NCQ priority
(see ncq_prio_enable and sas_ncq_prio_enable in
Documentation/ABI/testing/sysfs-block-device),
CDL has to be explicitly enabled using:
echo 1 > /sys/block/$bdev/device/cdl_enable

Since the ioprio hints are supplied through the existing I/O priority API,
it should be simple for an application to make use of the ioprio hints.

It simply has to reuse one of the new macros defined in
include/uapi/linux/ioprio.h: IOPRIO_PRIO_HINT() or IOPRIO_PRIO_VALUE_HINT(),
and supply one of the new hints defined in include/uapi/linux/ioprio.h:
IOPRIO_HINT_DEV_DURATION_LIMIT_[1-7], which indicates that the I/O should
use the corresponding CDL DLD index 1-7.

By reusing the I/O priority API, the user can both define a DLD to use per
AIO (io_uring sqe->ioprio or libaio iocb->aio_reqprio) or per-thread
(ioprio_set()).

=======
Testing
=======
With the following fio patches:
https://github.com/floatious/fio/commits/cdl

fio adds support for ioprio hints, such that CDL can be tested using e.g.:
fio --ioengine=io_uring --cmdprio_percentage=10 --cmdprio_hint=DLD_index

A simple way to test is to use a DLD with a very short duration limit,
and send large reads. Regardless of the CDL policy, in case of a CDL
timeout, the I/O will result in a -ETIME error to user-space.

We also provide a CDL test suite located in the cdl-tools repo, see:
https://github.com/westerndigitalcorporation/cdl-tools#testing-a-system-command-duration-limits-support

We have tested this patch series using:
-real hardware
-the following QEMU implementation:
https://github.com/floatious/qemu/tree/cdl
(NOTE: the QEMU implementation requires you to define the CDL policy at compile
time, so you currently need to recompile QEMU when switching between policies.)

===================
Further information
===================
For further information about CDL, see Damien's slides:

Presented at SDC 2021:
https://www.snia.org/sites/default/files/SDC/2021/pdfs/SNIA-SDC21-LeMoal-Be-On-Time-command-duration-limits-Feature-Support-in%20Linux.pdf

Presented at Lund Linux Con 2022:
https://drive.google.com/file/d/1I6ChFc0h4JY9qZdO1bY5oCAdYCSZVqWw/view?usp=sharing

================
Changes since V6
================
-Rebased series on v6.4-rc1.
-Picked up Reviewed-by tags from Hannes (Thank you Hannes!)
-Picked up Reviewed-by tag from Christoph (Thank you Christoph!)
-Changed KernelVersion from 6.4 to 6.5 for new sysfs attributes.

For older change logs, see previous patch series versions:
https://lore.kernel.org/linux-scsi/20230406113252.41211-1-nks@flawful.org/
https://lore.kernel.org/linux-scsi/20230404182428.715140-1-nks@flawful.org/
https://lore.kernel.org/linux-scsi/20230309215516.3800571-1-niklas.cassel@wdc.com/
https://lore.kernel.org/linux-scsi/20230124190308.127318-1-niklas.cassel@wdc.com/
https://lore.kernel.org/linux-scsi/20230112140412.667308-1-niklas.cassel@wdc.com/
https://lore.kernel.org/linux-scsi/20221208105947.2399894-1-niklas.cassel@wdc.com/

Link: https://lore.kernel.org/r/20230511011356.227789-1-nks@flawful.org
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2023-05-22 17:09:51 -04:00
..
partitions block: Cleanup set_capacity()/bdev_set_nr_sectors() 2023-04-24 08:48:09 -06:00
badblocks.c
bdev.c block: sync part's ->bd_has_submit_bio with disk's 2023-04-25 07:36:02 -06:00
bfq-cgroup.c blkcg: Restructure blkg_conf_prep() and friends 2023-04-13 06:46:49 -06:00
bfq-iosched.c scsi: block: ioprio: Clean up interface definition 2023-05-22 17:05:18 -04:00
bfq-iosched.h block, bfq: remove BFQ_WEIGHT_LEGACY_DFL 2023-04-06 16:17:32 -06:00
bfq-wf2q.c
bio-integrity.c block: bio-integrity: Copy flags when bio_integrity_payload is cloned 2023-02-16 11:05:41 -07:00
bio.c ext4: Convert ext4_bio_write_page() to use a folio 2023-04-06 13:39:50 -04:00
blk-cgroup-fc-appid.c
blk-cgroup-rwstat.c Revert "blk-cgroup: pin the gendisk in struct blkcg_gq" 2023-02-14 14:24:09 -07:00
blk-cgroup-rwstat.h
blk-cgroup.c for-6.4/block-2023-05-06 2023-05-06 08:28:58 -07:00
blk-cgroup.h for-6.4/block-2023-04-21 2023-04-26 12:52:58 -07:00
blk-core.c Merge patch series "Add Command Duration Limits support" 2023-05-22 17:09:51 -04:00
blk-crypto-fallback.c
blk-crypto-internal.h blk-crypto: remove blk_crypto_insert_cloned_request() 2023-03-16 09:35:09 -06:00
blk-crypto-profile.c blk-crypto: drop the NULL check from blk_crypto_put_keyslot() 2023-03-16 09:35:09 -06:00
blk-crypto-sysfs.c block: make kobj_type structures constant 2023-02-09 09:38:16 -07:00
blk-crypto.c blk-crypto: make blk_crypto_evict_key() more robust 2023-03-16 09:35:09 -06:00
blk-flush.c blk-mq: fix the blk_mq_add_to_requeue_list call in blk_kick_flush 2023-04-16 13:01:43 -06:00
blk-ia-ranges.c block: make kobj_type structures constant 2023-02-09 09:38:16 -07:00
blk-integrity.c blk-integrity: register sysfs attributes on struct device 2023-04-26 18:22:50 -06:00
blk-ioc.c
blk-iocost.c blkcg: Restructure blkg_conf_prep() and friends 2023-04-13 06:46:49 -06:00
blk-iolatency.c blk-iolatency: Make initialization lazy 2023-04-13 06:46:49 -06:00
blk-ioprio.c blk-cgroup: pass a gendisk to pd_alloc_fn 2023-02-03 08:20:05 -07:00
blk-ioprio.h
blk-lib.c
blk-map.c iov_iter: add iter_iovec() helper 2023-03-30 08:12:29 -06:00
blk-merge.c blk-mq: release crypto keyslot before reporting I/O complete 2023-03-16 09:35:09 -06:00
blk-mq-cpumap.c blk-mq: include <linux/blk-mq.h> in block/blk-mq.h 2023-04-13 06:52:29 -06:00
blk-mq-debugfs-zoned.c
blk-mq-debugfs.c blk-mq: include <linux/blk-mq.h> in block/blk-mq.h 2023-04-13 06:52:29 -06:00
blk-mq-debugfs.h
blk-mq-pci.c blk-mq: include <linux/blk-mq.h> in block/blk-mq.h 2023-04-13 06:52:29 -06:00
blk-mq-sched.c blk-mq: cleanup __blk_mq_sched_dispatch_requests 2023-04-13 06:57:18 -06:00
blk-mq-sched.h blk-mq: move blk_mq_sched_insert_request to blk-mq.c 2023-04-13 06:52:29 -06:00
blk-mq-sysfs.c blk-mq: include <linux/blk-mq.h> in block/blk-mq.h 2023-04-13 06:52:29 -06:00
blk-mq-tag.c blk-mq: include <linux/blk-mq.h> in block/blk-mq.h 2023-04-13 06:52:29 -06:00
blk-mq-virtio.c blk-mq: include <linux/blk-mq.h> in block/blk-mq.h 2023-04-13 06:52:29 -06:00
blk-mq.c for-6.4/block-2023-04-21 2023-04-26 12:52:58 -07:00
blk-mq.h for-6.4/block-2023-04-21 2023-04-26 12:52:58 -07:00
blk-pm.c blk-mq: include <linux/blk-mq.h> in block/blk-mq.h 2023-04-13 06:52:29 -06:00
blk-pm.h
blk-rq-qos.c blk-rq-qos: store a gendisk instead of request_queue in struct rq_qos 2023-02-03 08:20:05 -07:00
blk-rq-qos.h blk-iolatency: s/blkcg_rq_qos/iolat_rq_qos/ 2023-04-13 06:46:49 -06:00
blk-settings.c
blk-stat.c blk-mq: include <linux/blk-mq.h> in block/blk-mq.h 2023-04-13 06:52:29 -06:00
blk-stat.h
blk-sysfs.c blk-mq: include <linux/blk-mq.h> in block/blk-mq.h 2023-04-13 06:52:29 -06:00
blk-throttle.c blk-throttle: only enable blk-stat when BLK_DEV_THROTTLING_LOW 2023-04-13 06:48:11 -06:00
blk-throttle.h
blk-timeout.c
blk-wbt.c blk-rq-qos: store a gendisk instead of request_queue in struct rq_qos 2023-02-03 08:20:05 -07:00
blk-wbt.h
blk-zoned.c block: remove more NULL checks after bdev_get_queue() 2023-02-21 09:23:22 -07:00
blk.h blk-integrity: register sysfs attributes on struct device 2023-04-26 18:22:50 -06:00
bounce.c
bsg-lib.c
bsg.c driver core: class: remove module * from class_create() 2023-03-17 15:16:33 +01:00
disk-events.c
elevator.c block: make kobj_type structures constant 2023-02-09 09:38:16 -07:00
elevator.h blk-mq: pass a flags argument to elevator_type->insert_requests 2023-04-13 06:52:30 -06:00
fops.c
genhd.c for-6.4/block-2023-05-06 2023-05-06 08:28:58 -07:00
holder.c
ioctl.c block: fix scan partition for exclusively open device again 2023-02-17 06:15:57 -07:00
ioprio.c scsi: block: ioprio: Clean up interface definition 2023-05-22 17:05:18 -04:00
Kconfig for-6.4/block-2023-04-21 2023-04-26 12:52:58 -07:00
Kconfig.iosched
kyber-iosched.c blk-mq: pass a flags argument to elevator_type->insert_requests 2023-04-13 06:52:30 -06:00
Makefile blk-mq-rdma: remove queue mapping helper for rdma devices 2023-04-13 08:59:05 +02:00
mq-deadline.c blk-mq: pass a flags argument to elevator_type->insert_requests 2023-04-13 06:52:30 -06:00
opal_proto.h sed-opal: allow user authority to get locking range attributes. 2023-04-05 07:46:25 -06:00
sed-opal.c sed-opal: geometry feature reporting command 2023-04-19 14:07:13 -06:00
t10-pi.c