| Age | Commit message (Collapse) | Author |
|---|
|
Now that we have an efficient mechanism to update these two stats,
let's start maintaining them again.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
|
Avoid the overhead of a memory bus lock cycle for counting a value
that is hardly every used.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
|
Receives are frequent events. Avoid the overhead of a memory bus
lock cycle for counting a value that is hardly every used.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
|
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
|
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
|
Clean up: The unit of XDR alignment is defined by RFC 4506,
not as part of the RPC message header. Thus it belongs in
include/linux/sunrpc/xdr.h.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
|
Make the sunrpc trace subsystem trace events easier to use.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
|
The next few patches will employ these strings to help make server-
side trace logs more human-readable. A similar technique is already
in use in kernel RPC client code.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
|
If the port partner is PD2, the PDOs of the local port should follow the
format defined in PD2 Spec. Dynamically modify the pre-defined PD3 PDOs
and transform them into PD2 format before sending them to the PD2 port
partner.
Reviewed-by: Guenter Roeck <linux@roeckus.net>
Signed-off-by: Kyle Tso <kyletso@google.com>
Link: https://lore.kernel.org/r/20210115163311.391332-1-kyletso@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
We need the fixes in here and this resolves a merge issue with
drivers/usb/gadget/udc/bdc/Kconfig.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
We need the fixes in here and this resolves a merge issue in
drivers/tty/tty_io.c
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
We need the IIO/Staging fixes in here as well.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
We can remove 'q' from blk_execute_rq as well after the previous change
in blk_execute_rq_nowait.
And more importantly it never really was needed to start with given
that we can trivial derive it from struct request.
Cc: linux-scsi@vger.kernel.org
Cc: virtualization@lists.linux-foundation.org
Cc: linux-ide@vger.kernel.org
Cc: linux-mmc@vger.kernel.org
Cc: linux-nvme@lists.infradead.org
Cc: linux-nfs@vger.kernel.org
Acked-by: Ulf Hansson <ulf.hansson@linaro.org> # for mmc
Signed-off-by: Guoqing Jiang <guoqing.jiang@cloud.ionos.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
The 'q' is not used since commit a1ce35fa4985 ("block: remove dead
elevator code"), also update the comment of the function.
And more importantly it never really was needed to start with given
that we can trivial derive it from struct request.
Cc: target-devel@vger.kernel.org
Cc: linux-scsi@vger.kernel.org
Cc: virtualization@lists.linux-foundation.org
Cc: linux-ide@vger.kernel.org
Cc: linux-mmc@vger.kernel.org
Cc: linux-nvme@lists.infradead.org
Cc: linux-nfs@vger.kernel.org
Signed-off-by: Guoqing Jiang <guoqing.jiang@cloud.ionos.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux into drm-next
Backmerge v5.11-rc5 into drm-next to clean up a bunch of conflicts we are dragging around.
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
bvec_alloc(), bvec_free() and bvec_nr_vecs() are only used inside block
layer core functions, no need to declare them in public header.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
The inline bvecs won't be used if user needn't bvecs by not passing
BIOSET_NEED_BVECS, so don't allocate bvecs in this situation.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Tested-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Currently when non-mq aware IO scheduler (BFQ, mq-deadline) is used for
a queue with multiple HW queues, the performance it rather bad. The
problem is that these IO schedulers use queue-wide locking and their
dispatch function does not respect the hctx it is passed in and returns
any request it finds appropriate. Thus locality of request access is
broken and dispatch from multiple CPUs just contends on IO scheduler
locks. For these IO schedulers there's little point in dispatching from
multiple CPUs. Instead dispatch always only from a single CPU to limit
contention.
Below is a comparison of dbench runs on XFS filesystem where the storage
is a raid card with 64 HW queues and to it attached a single rotating
disk. BFQ is used as IO scheduler:
clients MQ SQ MQ-Patched
Amean 1 39.12 (0.00%) 43.29 * -10.67%* 36.09 * 7.74%*
Amean 2 128.58 (0.00%) 101.30 * 21.22%* 96.14 * 25.23%*
Amean 4 577.42 (0.00%) 494.47 * 14.37%* 508.49 * 11.94%*
Amean 8 610.95 (0.00%) 363.86 * 40.44%* 362.12 * 40.73%*
Amean 16 391.78 (0.00%) 261.49 * 33.25%* 282.94 * 27.78%*
Amean 32 324.64 (0.00%) 267.71 * 17.54%* 233.00 * 28.23%*
Amean 64 295.04 (0.00%) 253.02 * 14.24%* 242.37 * 17.85%*
Amean 512 10281.61 (0.00%) 10211.16 * 0.69%* 10447.53 * -1.61%*
Numbers are times so lower is better. MQ is stock 5.10-rc6 kernel. SQ is
the same kernel with megaraid_sas.host_tagset_enable=0 so that the card
advertises just a single HW queue. MQ-Patched is a kernel with this
patch applied.
You can see multiple hardware queues heavily hurt performance in
combination with BFQ. The patch restores the performance.
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
This reverts commit b445547ec1bbd3e7bf4b1c142550942f70527d95.
Since both mq-deadline and BFQ completely ignore hctx they are passed to
their dispatch function and dispatch whatever request they deem fit
checking whether any request for a particular hctx is queued is just
pointless since we'll very likely get a request from a different hctx
anyway. In the following commit we'll deal with lock contention in these
IO schedulers in presence of multiple HW queues in a different way.
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Now that no fast path lookups in the partition table are left, there is
no point in micro-optimizing the data structure for it. Just use a bog
standard xarray.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
There is good reason to iterate backwards when deleting all partitions in
del_gendisk, just like we don't in blk_drop_partitions.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Add a helper to call kobject_uevent for the disk and all partitions, and
unexport the disk_part_iter_* helpers that are now only used in the core
block code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Rework the I/O accounting for bio based drivers to use ->bi_bdev. This
means all drivers can now simply use bio_start_io_acct to start
accounting, and it will take partitions into account automatically. To
end I/O account either bio_end_io_acct can be used if the driver never
remaps I/O to a different device, or bio_end_io_acct_remapped if the
driver did remap the I/O.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
There is no good reason to reassign ->bi_bdev when remapping the
partition-relative block number to the device wide one, as all the
information required by the drivers comes from the gendisk anyway.
Keeping the original ->bi_bdev alive will allow to greatly simplify
the partition-away I/O accounting.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Replace the gendisk pointer in struct bio with a pointer to the newly
improved struct block device. From that the gendisk can be trivially
accessed with an extra indirection, but it also allows to directly
look up all information related to partition remapping.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Commit 20bd1d026aac ("scsi: sd: Keep disk read-only when re-reading
partition") addressed a long-standing problem with user read-only
policy being overridden as a result of a device-initiated revalidate.
The commit has since been reverted due to a regression that left some
USB devices read-only indefinitely.
To fix the underlying problems with revalidate we need to keep track
of hardware state and user policy separately.
The gendisk has been updated to reflect the current hardware state set
by the device driver. This is done to allow returning the device to
the hardware state once the user clears the BLKROSET flag.
The resulting semantics are as follows:
- If BLKROSET sets a given partition read-only, that partition will
remain read-only even if the underlying storage stack initiates a
revalidate. However, the BLKRRPART ioctl will cause the partition
table to be dropped and any user policy on partitions will be lost.
- If BLKROSET has not been set, both the whole disk device and any
partitions will reflect the current write-protect state of the
underlying device.
Based on a patch from Martin K. Petersen <martin.petersen@oracle.com>.
Reported-by: Oleksii Kurochko <olkuroch@cisco.com>
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=201221
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Pull block fixes from Jens Axboe:
- NVMe pull request from Christoph:
- fix a status code in nvmet (Chaitanya Kulkarni)
- avoid double completions in nvme-rdma/nvme-tcp (Chao Leng)
- fix the CMB support to cope with NVMe 1.4 controllers (Klaus Jensen)
- fix PRINFO handling in the passthrough ioctl (Revanth Rajashekar)
- fix a double DMA unmap in nvme-pci
- lightnvm error path leak fix (Pan)
- MD pull request from Song:
- Flush request fix (Xiao)
* tag 'block-5.11-2021-01-24' of git://git.kernel.dk/linux-block:
lightnvm: fix memory leak when submit fails
nvme-pci: fix error unwind in nvme_map_data
nvme-pci: refactor nvme_unmap_data
md: Set prev_flush_start and flush_bio in an atomic way
nvmet: set right status on error in id-ns handler
nvme-pci: allow use of cmb on v1.4 controllers
nvme-tcp: avoid request double completion for concurrent nvme_tcp_timeout
nvme-rdma: avoid request double completion for concurrent nvme_rdma_timeout
nvme: check the PRINFO bit before deciding the host buffer length
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
Pull driver core fixes from Greg KH:
"Here are some small driver core fixes for 5.11-rc5 that resolve some
reported problems:
- revert of a -rc1 patch that was causing problems with some machines
- device link device name collision problem fix (busses only have to
name devices unique to their bus, not unique to all busses)
- kernfs splice bugfixes to resolve firmware loading problems for
Qualcomm systems.
- other tiny driver core fixes for minor issues reported.
All of these have been in linux-next with no reported problems"
* tag 'driver-core-5.11-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
driver core: Fix device link device name collision
driver core: Extend device_is_dependent()
kernfs: wire up ->splice_read and ->splice_write
kernfs: implement ->write_iter
kernfs: implement ->read_iter
Revert "driver core: Reorder devices on successful probe"
Driver core: platform: Add extra error check in devm_platform_get_irqs_affinity()
drivers core: Free dma_range_map when driver probe failed
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fixes from Borislav Petkov:
- Correct the marking of kthreads which are supposed to run on a
specific, single CPU vs such which are affine to only one CPU, mark
per-cpu workqueue threads as such and make sure that marking
"survives" CPU hotplug. Fix CPU hotplug issues with such kthreads.
- A fix to not push away tasks on CPUs coming online.
- Have workqueue CPU hotplug code use cpu_possible_mask when breaking
affinity on CPU offlining so that pending workers can finish on newly
arrived onlined CPUs too.
- Dump tasks which haven't vacated a CPU which is currently being
unplugged.
- Register a special scale invariance callback which gets called on
resume from RAM to read out APERF/MPERF after resume and thus make
the schedutil scaling governor more precise.
* tag 'sched_urgent_for_v5.11_rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched: Relax the set_cpus_allowed_ptr() semantics
sched: Fix CPU hotplug / tighten is_per_cpu_kthread()
sched: Prepare to use balance_push in ttwu()
workqueue: Restrict affinity change to rescuer
workqueue: Tag bound workers with KTHREAD_IS_PER_CPU
kthread: Extract KTHREAD_IS_PER_CPU
sched: Don't run cpu-online with balance_push() enabled
workqueue: Use cpu_possible_mask instead of cpu_active_mask to break affinity
sched/core: Print out straggler tasks in sched_cpu_dying()
x86: PM: Register syscore_ops for scale invariance
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull timer fixes from Borislav Petkov:
- Fix an integer overflow in the NTP RTC synchronization which led to
the latter happening every 2 seconds instead of the intended every 11
minutes.
- Get rid of now unused get_seconds().
* tag 'timers_urgent_for_v5.11_rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
ntp: Fix RTC synchronization on 32-bit platforms
timekeeping: Remove unused get_seconds()
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux
Pull misc fixes from Christian Brauner:
- Jann reported sparse complaints because of a missing __user
annotation in a helper we added way back when we added
pidfd_send_signal() to avoid compat syscall handling. Fix it.
- Yanfei replaces a reference in a comment to the _do_fork() helper I
removed a while ago with a reference to the new kernel_clone()
replacement
- Alexander Guril added a simple coding style fix
* tag 'for-linus-2021-01-24' of git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux:
kthread: remove comments about old _do_fork() helper
Kernel: fork.c: Fix coding style: Do not use {} around single-line statements
signal: Add missing __user annotation to copy_siginfo_from_user_any
|
|
Upon receiving a cumulative ACK that changes the congestion state from
Disorder to Open, the TLP timer is not set. If the sender is app-limited,
it can only wait for the RTO timer to expire and retransmit.
The reason for this is that the TLP timer is set before the congestion
state changes in tcp_ack(), so we delay the time point of calling
tcp_set_xmit_timer() until after tcp_fastretrans_alert() returns and
remove the FLAG_SET_XMIT_TIMER from ack_flag when the RACK reorder timer
is set.
This commit has two additional benefits:
1) Make sure to reset RTO according to RFC6298 when receiving ACK, to
avoid spurious RTO caused by RTO timer early expires.
2) Reduce the xmit timer reschedule once per ACK when the RACK reorder
timer is set.
Fixes: df92c8394e6e ("tcp: fix xmit timer to only be reset if data ACKed/SACKed")
Link: https://lore.kernel.org/netdev/1611311242-6675-1-git-send-email-yangpc@wangsu.com
Signed-off-by: Pengcheng Yang <yangpc@wangsu.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Yuchung Cheng <ycheng@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/1611464834-23030-1-git-send-email-yangpc@wangsu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Introduce a new netdev feature, NETIF_F_GRO_UDP_FWD, to allow user
to turn UDP GRO on and off for forwarding.
Defaults to off to not change current datapath.
Suggested-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Alexander Lobakin <alobakin@pm.me>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The TCP_USER_TIMEOUT is checked by the 0-window probe timer. As the
timer has backoff with a max interval of about two minutes, the
actual timeout for TCP_USER_TIMEOUT can be off by up to two minutes.
In this patch the TCP_USER_TIMEOUT is made more accurate by taking it
into account when computing the timer value for the 0-window probes.
This patch is similar to and builds on top of the one that made
TCP_USER_TIMEOUT accurate for RTOs in commit b701a99e431d ("tcp: Add
tcp_clamp_rto_to_user_timeout() helper to improve accuracy").
Fixes: 9721e709fa68 ("tcp: simplify window probe aborting on USER_TIMEOUT")
Signed-off-by: Enke Chen <enchen@paloaltonetworks.com>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20210122191306.GA99540@localhost.localdomain
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
None of these are actually used in the kernel/userspace interface -
there's a userspace component of implementing MRP, and userspace will
need to construct certain frames to put on the wire, but there's no
reason the kernel should provide the relevant definitions in a UAPI
header.
In fact, some of those definitions were broken until previous commit,
so only keep the few that are actually referenced in the kernel code,
and move them to the br_private_mrp.h header.
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Wireshark says that the MRP test packets cannot be decoded - and the
reason for that is that there's a two-byte hole filled with garbage
between the "transitions" and "timestamp" members.
So Wireshark decodes the two garbage bytes and the top two bytes of
the timestamp written by the kernel as the timestamp value (which thus
fluctuates wildly), and interprets the lower two bytes of the
timestamp as a new (type, length) pair, which is of course broken.
Even though this makes the timestamp field in the struct unaligned, it
actually makes it end up on a 32 bit boundary in the frame as mandated
by the standard, since it is preceded by a two byte TLV header.
The struct definitions live under include/uapi/, but they are not
really part of any kernel<->userspace API/ABI, so fixing the
definitions by adding the packed attribute should not cause any
compatibility issues.
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Reviewed-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Add a flag to signal that only pure overwrites are allowed.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
Pass a set of flags to iomap_dio_rw instead of the boolean
wait_for_completion argument. The IOMAP_DIO_FORCE_WAIT flag
replaces the wait_for_completion, but only needs to be passed
when the iocb isn't synchronous to start with to simplify the
callers.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
[djwong: rework xfs_file.c so that we can push iomap changes separately]
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
The fix for a long-standing USB-audio bug required one more dependency
variable to be added to the hw constraints. Unfortunately I didn't
realize at debugging that the new addition may result in the overflow
of the dependency array of each snd_pcm_hw_rule (up to three plus a
sentinel), because USB-audio driver adds one more dependency only for
a certain device and bus, hence it works as is for many devices. But
in a bad case, a simple open always results in -EINVAL (with kernel
WARNING if CONFIG_SND_DEBUG is set) no matter what is passed.
Since the dependencies are real and unavoidable (USB-audio restricts
the hw_params per looping over the format/rate/channels combos), the
only good solution seems to raise the bar for one more dependency for
snd_pcm_hw_rule -- so does this patch: now the hw constraint
dependencies can be up to four.
Fixes: 506c203cc3de ("ALSA: usb-audio: Fix hw constraints dependencies")
Reported-by: Jamie Heilman <jamie@audible.transient.net>
Link: https://lore.kernel.org/r/20210123155730.22576-1-tiwai@suse.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
|
|
The CSR SiRF prima2/atlas platforms are getting removed, so this driver
is no longer needed.
Cc: Barry Song <baohua@kernel.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Barry Song <baohua@kernel.org>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20210120154158.1860736-2-arnd@kernel.org
|
|
The switch ASIC has a limited capacity of physical ('flavour physical'
in devlink terminology) ports that it can support. While each system is
brought up with a different number of ports, this number can be
increased via splitting up to the ASIC's limit.
Expose physical ports as a devlink resource so that user space will have
visibility to the maximum number of ports that can be supported and the
current occupancy.
In addition, add a "Generic Resources" section in devlink-resource
documentation so the different drivers will be aligned by the same resource
name when exposing to user space.
Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
This commit adds support for HTB offload in the mlx5e driver.
Performance:
NIC: Mellanox ConnectX-6 Dx
CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz (24 cores with HT)
100 Gbit/s line rate, 500 UDP streams @ ~200 Mbit/s each
48 traffic classes, flower used for steering
No shaping (rate limits set to 4 Gbit/s per TC) - checking for max
throughput.
Baseline: 98.7 Gbps, 8.25 Mpps
HTB: 6.7 Gbps, 0.56 Mpps
HTB offload: 95.6 Gbps, 8.00 Mpps
Limitations:
1. 256 leaf nodes, 3 levels of depth.
2. Granularity for ceil is 1 Mbit/s. Rates are converted to weights, and
the bandwidth is split among the siblings according to these weights.
Other parameters for classes are not supported.
Ethtool statistics support for QoS SQs are also added. The counters are
called qos_txN_*, where N is the QoS queue number (starting from 0, the
numeration is separate from the normal SQs), and * is the counter name
(the counters are the same as for the normal SQs).
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
HTB doesn't scale well because of contention on a single lock, and it
also consumes CPU. This patch adds support for offloading HTB to
hardware that supports hierarchical rate limiting.
In the offload mode, HTB passes control commands to the driver using
ndo_setup_tc. The driver has to replicate the whole hierarchy of classes
and their settings (rate, ceil) in the NIC. Every modification of the
HTB tree caused by the admin results in ndo_setup_tc being called.
After this setup, the HTB algorithm is done completely in the NIC. An SQ
(send queue) is created for every leaf class and attached to the
hierarchy, so that the NIC can calculate and obey aggregated rate
limits, too. In the future, it can be changed, so that multiple SQs will
back a single leaf class.
ndo_select_queue is responsible for selecting the right queue that
serves the traffic class of each packet.
The data path works as follows: a packet is classified by clsact, the
driver selects a hardware queue according to its class, and the packet
is enqueued into this queue's qdisc.
This solution addresses two main problems of scaling HTB:
1. Contention by flow classification. Currently the filters are attached
to the HTB instance as follows:
# tc filter add dev eth0 parent 1:0 protocol ip flower dst_port 80
classid 1:10
It's possible to move classification to clsact egress hook, which is
thread-safe and lock-free:
# tc filter add dev eth0 egress protocol ip flower dst_port 80
action skbedit priority 1:10
This way classification still happens in software, but the lock
contention is eliminated, and it happens before selecting the TX queue,
allowing the driver to translate the class to the corresponding hardware
queue in ndo_select_queue.
Note that this is already compatible with non-offloaded HTB and doesn't
require changes to the kernel nor iproute2.
2. Contention by handling packets. HTB is not multi-queue, it attaches
to a whole net device, and handling of all packets takes the same lock.
When HTB is offloaded, it registers itself as a multi-queue qdisc,
similarly to mq: HTB is attached to the netdev, and each queue has its
own qdisc.
Some features of HTB may be not supported by some particular hardware,
for example, the maximum number of classes may be limited, the
granularity of rate and ceil parameters may be different, etc. - so, the
offload is not enabled by default, a new parameter is used to enable it:
# tc qdisc replace dev eth0 root handle 1: htb offload
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
In a following commit, sch_htb will start using extack in the delete
class operation to pass hardware errors in offload mode. This commit
prepares for that by adding the extack parameter to this callback and
converting usage of the existing qdiscs.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The existing qdiscs that set TCQ_F_MQROOT don't use sch_tree_lock.
However, hardware-offloaded HTB will start setting this flag while also
using sch_tree_lock.
The current implementation of sch_tree_lock basically locks on
qdisc->dev_queue->qdisc, and it works fine when the tree is attached to
some queue. However, it's not the case for MQROOT qdiscs: such a qdisc
is the root itself, and its dev_queue just points to queue 0, while not
actually being used, because there are real per-queue qdiscs.
This patch changes the logic of sch_tree_lock and sch_tree_unlock to
lock the qdisc itself if it's the MQROOT.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
tcp_recvmsg() uses the CMSG mechanism to receive control information
like packet receive timestamps. This patch adds CMSG fields to
struct tcp_zerocopy_receive, and provides receive timestamps
if available to the user.
Signed-off-by: Arjun Roy <arjunroy@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
These variants were added for bisectability. Remove them, as all call sites
have now been convertd to use the original API.
Link: https://lore.kernel.org/r/20210118100955.1761652-20-a.darwish@linutronix.de
Cc: Jason Yan <yanaijie@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
|
|
All call-sites of below libsas APIs:
- sas_alloc_event()
- sas_notify_port_event()
- sas_notify_phy_event()
have been converted to use the _gfp()-suffixed version. Modify the
original APIs above to take a gfp_t flags parameter by default.
For bisectability, call-sites will be modified again to use the original
libsas APIs (while passing gfp_t). The temporary _gfp()-suffixed versions
can then be removed.
Link: https://lore.kernel.org/r/20210118100955.1761652-13-a.darwish@linutronix.de
Cc: Jason Yan <yanaijie@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
|
|
sas_alloc_event() uses in_interrupt() to decide which allocation should be
used.
The usage of in_interrupt() in drivers is phased out and Linus clearly
requested that code which changes behaviour depending on context should
either be separated or the context be conveyed in an argument passed by the
caller, which usually knows the context.
The in_interrupt() check is also only partially correct, because it fails
to choose the correct code path when just preemption or interrupts are
disabled. For example, as in the following call chain:
mvsas/mv_sas.c: mvs_work_queue() [process context]
spin_lock_irqsave(mvs_info::lock, )
-> libsas/sas_event.c: sas_notify_phy_event()
-> sas_alloc_event()
-> in_interrupt() = false
-> invalid GFP_KERNEL allocation
-> libsas/sas_event.c: sas_notify_port_event()
-> sas_alloc_event()
-> in_interrupt() = false
-> invalid GFP_KERNEL allocation
Introduce sas_alloc_event_gfp(), sas_notify_port_event_gfp(), and
sas_notify_phy_event_gfp(), which all behave like the non _gfp() variants
but use a caller-passed GFP mask for allocations.
For bisectability, all callers will be modified first to pass GFP context,
then the non _gfp() libsas API variants will be modified to take a gfp_t by
default.
Link: https://lore.kernel.org/r/20210118100955.1761652-4-a.darwish@linutronix.de
Fixes: 1c393b970e0f ("scsi: libsas: Use dynamic alloced work to avoid sas event lost")
Cc: Jason Yan <yanaijie@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
|
|
LLDDs report events to libsas with .notify_port_event and .notify_phy_event
callbacks.
These callbacks are fixed and so there is no reason why the functions
cannot be called directly, so do that.
This neatens the code slightly, makes it more obvious, and reduces function
pointer usage, which is generally a good thing. Downside is that there are
2x more symbol exports.
[a.darwish@linutronix.de: Remove the now unused "sas_ha" local variables]
Link: https://lore.kernel.org/r/20210118100955.1761652-3-a.darwish@linutronix.de
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
|
