linux/drivers
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
..
accel Driver core changes for 6.4-rc1 2023-04-27 11:53:57 -07:00
accessibility
acpi ACPI: video: Remove acpi_backlight=video quirk for Lenovo ThinkPad W530 2023-05-04 20:23:41 +02:00
amba
android
ata scsi: ata: libata: Handle completion of CDL commands using policy 0xD 2023-05-22 17:05:20 -04:00
atm
auxdisplay
base - Nick Piggin's "shoot lazy tlbs" series, to improve the peformance of 2023-04-27 19:42:02 -07:00
bcma
block for-6.4/io_uring-2023-05-07 2023-05-07 10:00:09 -07:00
bluetooth Bluetooth: btusb: Add WCN6855 devcoredump support 2023-04-23 22:07:27 -07:00
bus modules-6.4-rc1 2023-04-27 16:36:55 -07:00
cdrom
cdx cdx: fix build failure due to sysfs 'bus_type' argument needing to be const 2023-04-27 16:21:32 -07:00
char This PR fixes a critical bug in my first pull request. I fixed the 2023-05-01 10:37:37 -07:00
clk A couple more patches that would be good to get into -rc1. 2023-05-07 10:31:45 -07:00
clocksource Timekeeping and clocksource/event driver updates the second batch: 2023-04-29 10:24:30 -07:00
comedi
connector
counter - New Drivers 2023-05-02 10:41:31 -07:00
cpufreq Driver core changes for 6.4-rc1 2023-04-27 11:53:57 -07:00
cpuidle RISC-V: Align SBI probe implementation with spec 2023-04-29 13:04:50 -07:00
crypto This push fixes the following problems: 2023-05-07 10:57:14 -07:00
cxl cxl for v6.4 2023-04-30 11:51:51 -07:00
dax
dca Mainly singleton patches all over the place. Series of note are: 2023-04-27 19:57:00 -07:00
devfreq Driver core changes for 6.4-rc1 2023-04-27 11:53:57 -07:00
dio
dma dmaengine updates for v6.4 2023-05-03 11:11:56 -07:00
dma-buf - Nick Piggin's "shoot lazy tlbs" series, to improve the peformance of 2023-04-27 19:42:02 -07:00
edac Driver core changes for 6.4-rc1 2023-04-27 11:53:57 -07:00
eisa
extcon
firewire firewire updates for 6.4-rc1 2023-04-27 11:27:48 -07:00
firmware EFI updates for v6.4: 2023-04-29 17:42:33 -07:00
fpga Char/Misc drivers for 6.4-rc1 2023-04-27 12:07:50 -07:00
fsi
gnss
gpio hte: Changes for v6.4-rc1 2023-05-03 11:00:27 -07:00
gpu drm fixes part 2 for 6.4-rc1 2023-05-05 11:49:22 -07:00
greybus
hid Driver core changes for 6.4-rc1 2023-04-27 11:53:57 -07:00
hsi
hte Devicetree updates for v6.4, part 2: 2023-04-27 10:09:05 -07:00
hv hyperv-next for v6.4 2023-04-27 17:17:12 -07:00
hwmon Driver core changes for 6.4-rc1 2023-04-27 11:53:57 -07:00
hwspinlock
hwtracing
i2c i2c: gxp: fix build failure without CONFIG_I2C_SLAVE 2023-05-03 17:27:29 +02:00
i3c i3c: ast2600: set variable ast2600_i3c_ops storage-class-specifier to static 2023-04-30 23:50:26 +02:00
idle intel_idle: mark few variables as __read_mostly 2023-04-27 19:37:36 +02:00
iio Char/Misc drivers for 6.4-rc1 2023-04-27 12:07:50 -07:00
infiniband v6.4 merge window RDMA pull request 2023-04-29 17:21:24 -07:00
input Input updates for 6.4 merge window: 2023-05-01 17:18:56 -07:00
interconnect modules-6.4-rc1 2023-04-27 16:36:55 -07:00
iommu IOMMU Updates for Linux 6.4 2023-04-30 13:00:38 -07:00
ipack
irqchip - Nick Piggin's "shoot lazy tlbs" series, to improve the peformance of 2023-04-27 19:42:02 -07:00
isdn Including fixes from netfilter. 2023-05-05 19:12:01 -07:00
leds - New Drivers 2023-05-02 10:36:02 -07:00
macintosh powerpc updates for 6.4 2023-04-28 16:24:32 -07:00
mailbox - mailbox api: allow direct registration to a channel 2023-05-07 10:17:33 -07:00
mcb mcb-lpc: Reallocate memory region to avoid memory overlapping 2023-04-20 14:24:01 +02:00
md Merge patch series "Use block pr_ops in LIO" 2023-05-22 16:35:02 -04:00
media IOMMU Updates for Linux 6.4 2023-04-30 13:00:38 -07:00
memory ARM: SoC drivers for v6.4 2023-04-25 12:02:16 -07:00
memstick
message Objtool changes for v6.4: 2023-04-28 14:02:54 -07:00
mfd - New Drivers 2023-05-02 10:41:31 -07:00
misc Objtool changes for v6.4: 2023-04-28 14:02:54 -07:00
mmc TTY/Serial changes for 6.4-rc1 2023-04-27 11:46:26 -07:00
most
mtd This pull request contains updates for UBI and UBIFS 2023-05-03 18:58:59 -07:00
mux
net Including fixes from netfilter. 2023-05-05 19:12:01 -07:00
nfc drivers: nfc: nfcsim: remove return value check of dev_dir 2023-04-24 18:12:42 -07:00
ntb
nubus
nvdimm
nvme Merge patch series "Use block pr_ops in LIO" 2023-05-22 16:35:02 -04:00
nvmem modules-6.4-rc1 2023-04-27 16:36:55 -07:00
of Devicetree fixes for 6.4, part 1: 2023-05-05 13:27:59 -07:00
opp Devicetree updates for v6.4, part 2: 2023-04-27 10:09:05 -07:00
parisc parisc: Replace regular spinlock with spin_trylock on panic path 2023-05-03 17:43:26 +02:00
parport
pci cxl for v6.4 2023-04-30 11:51:51 -07:00
pcmcia
peci
perf RISC-V: Align SBI probe implementation with spec 2023-04-29 13:04:50 -07:00
phy phy fixes for 6.4 2023-05-05 11:57:29 -07:00
pinctrl Pin control bulk changes for the v6.4 kernel: 2023-05-02 15:40:41 -07:00
platform modules-6.4-rc1 2023-04-27 16:36:55 -07:00
pnp
power power supply and reset changes for the v6.4 series 2023-04-29 17:37:02 -07:00
powercap
pps
ps3
ptp Driver core changes for 6.4-rc1 2023-04-27 11:53:57 -07:00
pwm pwm: Changes for v6.4-rc1 2023-05-03 11:25:01 -07:00
rapidio Mainly singleton patches all over the place. Series of note are: 2023-04-27 19:57:00 -07:00
ras
regulator modules-6.4-rc1 2023-04-27 16:36:55 -07:00
remoteproc Mainly singleton patches all over the place. Series of note are: 2023-04-27 19:57:00 -07:00
reset Nothing looks out of the ordinary in this batch of clk driver updates. There 2023-04-29 17:29:39 -07:00
rpmsg Driver core changes for 6.4-rc1 2023-04-27 11:53:57 -07:00
rtc - New Drivers 2023-05-02 10:41:31 -07:00
s390 Merge patch series "Use block pr_ops in LIO" 2023-05-22 16:35:02 -04:00
sbus Driver core changes for 6.4-rc1 2023-04-27 11:53:57 -07:00
scsi Merge patch series "Add Command Duration Limits support" 2023-05-22 17:09:51 -04:00
sh
siox
slimbus
soc modules-6.4-rc1 2023-04-27 16:36:55 -07:00
soundwire
spi Char/Misc drivers for 6.4-rc1 2023-04-27 12:07:50 -07:00
spmi spmi: Add a check for remove callback when removing a SPMI driver 2023-04-20 14:16:39 +02:00
ssb
staging modules-6.4-rc1 2023-04-27 16:36:55 -07:00
target Merge patch series "Use block pr_ops in LIO" 2023-05-22 16:35:02 -04:00
tc
tee Driver core changes for 6.4-rc1 2023-04-27 11:53:57 -07:00
thermal thermal: intel: powerclamp: Fix NULL pointer access issue 2023-05-04 20:30:18 +02:00
thunderbolt
tty Char/Misc drivers for 6.4-rc1 2023-04-27 12:07:50 -07:00
ufs scsi: ufs: hwmon: Constify pointers to hwmon_channel_info 2023-05-16 21:29:11 -04:00
uio
usb Char/Misc drivers for 6.4-rc1 2023-04-27 12:07:50 -07:00
vdpa virtio,vhost,vdpa: features, fixes, cleanups 2023-04-27 17:05:34 -07:00
vfio VFIO updates for v6.4-rc1 2023-05-02 11:56:43 -07:00
vhost Scheduler changes for v6.4: 2023-04-28 14:53:30 -07:00
video - Fix-ups 2023-05-02 10:46:11 -07:00
virt Devicetree updates for v6.4, part 2: 2023-04-27 10:09:05 -07:00
virtio - Nick Piggin's "shoot lazy tlbs" series, to improve the peformance of 2023-04-27 19:42:02 -07:00
vlynq
w1 Char/Misc drivers for 6.4-rc1 2023-04-27 12:07:50 -07:00
watchdog linux-watchdog 6.4-rc1 tag 2023-05-04 18:33:56 -07:00
xen xen: branch for v6.4-rc1 2023-04-27 17:27:06 -07:00
zorro
Kconfig
Makefile