This was added during debug/development and
was not deleted by accident.
Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com>
Change-Id: I146748151623a3ea4808c7b5cc34085d30ca81b4
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16388
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
directly
This patch removes hardcoded compressdev code from the
vbdev module and instead uses the accel_fw. The port required
a few changes based on how things are plumbed and accessed,
nothing that isn't be too obscure. CI tests were updated to
run ISAL accel_fw module as well as DPDK compressdev with QAT.
Unit tests for the new module will follow in a separate patch.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: I769cbc888658fb846d89f6f0bfeeb1a2a820767e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/13610
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
ublk backend could support ublk driver with kernel. Specify
configuration parameter to start it up.
Signed-off-by: Yifan Bian <yifan.bian@intel.com>
Co-authored-by: Xiaodong Liu <xiaodong.liu@intel.com>
Change-Id: I55e7d757e04315b25e9bfab5fdcbb6621be3e29e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15680
Reviewed-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
The mlx5 accel module supports crypto operations.
Data buffer is split into `block_size` chunks and each
chunk is enrypted individually.
mlx5 library contains some utility functions that will
later be used by other libraries, this lib will be
exntended later.
Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com>
Change-Id: Iacdd8caaade477277d5a95cfd53e9910e280a73b
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15420
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
All DPDK related code is removed, handling of
RESET command was sligthly updated.
Handling of -ENOMEM was updated for cases when
accel API returns -ENOMEM
Crypto tests in blockdev.sh were extended with more
crypto_bdevs to verify NOMEM cases - that failed
with original vbdev_crypto implementation
Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com>
Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Attempt to send any queued writes even if the socket is in a group. This
will not work if there is a send task already outstanding.
Change-Id: Icc0b5884e3d247042194ad26b30340ceb824886c
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15212
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Add RPC bdev_nvme_get_path_iostat to show I/O statistics for IO paths
of the NVMe bdev.
Change-Id: I22e5ad112d5cfa6d96cf246dcd0e511ae71dc839
Signed-off-by: Richael Zhuang <richael.zhuang@arm.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14745
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Handling for scalar dst buffers were removed in 1866faffe, and all
buffers are now described by iovecs. The dst field was removed from
spdk_accel_task, so we also need to remove any references to that field
from dpdk_compressdev. This patch fixes the build with compressdev
enabled.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I21fbbb45a15e7250944753e445dede7d586bf35b
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16349
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Michal Berger <michal.berger@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
We had it for compress but simply didn't think of a use case for
decompress. During the develpoment of the compressdev accel_fw
module it was discovered that compressdev does indeed provide the
uncompressed length on completion of decompress and the reducelib
uses it. So, add it here.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: I2f6a8bbbe3ef8ebe0b50d6434845f405afa7d37d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16035
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
From an earlier comment on the original patch that introduced the
module. Stop using the thread to associate the device_qp and
simply link it into the chan.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: Ib55979247d9bdacdc868bdb9dea97e1941e1e8ab
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15832
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
This is the port of the vbdev compress logic into the accel
framework. It includes just one enhancement, to only fill each
mbuf in either src or dst array with max "window size" param to
avoid QAT errors. Note that DPDK ISAL PMD was not ported as we
have native ISAL compression in accel now.
Note: ISAL w/DPDK is still built w/this patch, that can't be
removed until the vbdev module moves to accel fw as it still
depends on DPDK ISAL PMD.
Follow-on patches will include addition C API for PMD selection,
this patch just gets equivalent functionality going. Upcoming
patches will also convert the vbdev compress module to use the
accel framework instead of talking directly to compressdev.
More patches will also address comments on vbdev common code
that addressed here would make the review challenging.
This patch also fixes a bug in the ported code that needs to
be fixed here to pass CI. Capability discovery was incorrect
causing all devices to appear to not support chained mbufs,
with the mbuf splitting code this is important to get right.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: I7f526404819b145ef26e40877122ba80a02fcf51
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15178
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
We will be creating a new accel module for DPDK
compressdev. We can heavily leverage all of the
code from the existing vbdev_compress.c file. So
just copy that file exactly to start, and then the
next patch will contain only the deltas for removing
the bdev/reduce plumbing and adding the accel
plumbing.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I4f6ca1c2dad92eb8dd38b351602b65b8cb4ee4e8
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15829
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Community-CI: Mellanox Build Bot
We may need to process multiple events to get all of
the events associated with a newly published service,
so a 1 second poller is actually a bit too long since
we will only get one event per poll iteration. Shorten
it to 100ms instead.
It doesn't appear there is any easy way to use the avahi
API to help with this, maybe we could have some context
variable used to detect if an event was processed or not,
and call avahi_simple_poll_iterate() multiple times if
necessary. This might be something worth investigating
in the future.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ifea7a99dc8b8b501392e29276b20528a873b1ae1
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16327
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Boris Glimcher <Boris.Glimcher@emc.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Make vbdev_crypto_destruct() return 1 to signal that program
execution should wait for spdk_bdev_destruct_done() function,
which is added inside _device_unregister_cb().
This change is related to _vdev_dev_get() not being able
to find the devices, when called from _cryptodev_sym_session_free(),
as it uses device driver name, which might already be freed.
This occurs only during bdev module finish, when crypto bdevs
are being unregistered and vbdev_crypto_finish() proceeds to
call bdev name deletion without waiting for the unregister
callbacks to complete, which ultimately results in reading
freed pointers.
This only happens when code execution takes path for DPDK 22.11+.
Change-Id: Id9a43d07c90aef7a82867383fd77354ac521a3e7
Signed-off-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16290
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Currently delete_crypto_disk() calls spdk_bdev_unregister_bdev_by_name()
and immediately after exiting that function, it proceeds to bdev name
deletion. This causes callbacks triggered by above call to run with
references to bdev names that are partially unavailable due to free()
functions in vbdev_crypto_delete_name().
Add an intermediate callback that will call vbdev_crypto_delete_name()
and then the original callback that was sent as an argument to
delete_crypto_disk(). That way we are sure that bdev name deletion
is performed right after bdev unregister.
Change-Id: I483ce5c3c63596f54d44220ca61c5e97c3807faf
Signed-off-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16253
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
It was found that if system has no crypto devs,
we exit earlier from init function and don't register
io_device. Later in fini function we may try to unregister
invalid io_device
Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com>
Change-Id: I23ac9674c02d4fe97f01c66928a5ec5a2b39ec47
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16321
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This patch is a combination of commits which update vdev_crypto:
110d8411e bdev/crypto: do not create mempool for session private data
495055b05 bdev/crypto: update rte_cryptodev usage for DPDK 22.11
02caed6b5 bdev/crypto: remove mempool usage matching < DPDK 19.02
5887eb321 bdev/crypto: do not track type of crypto session
Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com>
Change-Id: I30c4f76e4e7b4865a7daa638d357888bb5e02071
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16039
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
The io paths' stat will get lost when they are destroyed. Record
the stat in the nvme_ns structure.
Change-Id: I12fc0b04fac0d59e7465fe543ee733f2822a9cdb
Signed-off-by: Richael Zhuang <richael.zhuang@arm.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14744
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Added helper functions nvme_ns_alloc() and nvme_ns_free().
This is a preparation for next patch where we will allocate or free
I/O statistics of nvme_ns when this nvme_ns is allocated or freed.
Change-Id: I54b0dba61bd19923178ba4bed97677cf628ba509
Signed-off-by: Richael Zhuang <richael.zhuang@arm.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16300
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Currently we have stat per bdev I/O channel, but for NVMe bdev
multipath, we don't have stat per I/O path. Especially for
active-active mode, we may want to observe each path's statistics.
This patch support IO stat for nvme_io_path. Record each nvme_io_path
stat using structure spdk_bdev_io_stat.
The following is the comparison of bdevperf test.
Test on Arm server with the following basic configuration.
1 Null bdev: block size: 4K, num_blocks:16k
run bdevperf with io size=4k, qdepth=1/32/128, rw type=randwrite/mixed with 70% read/randread
Each time run 30 seconds, each item run for 16 times and get the average.
The result is as follows.
qdepth type IOPS(default) IOPS(this patch) diff
1 randwrite 7795157.27 7859909.78 0.83%
1 mix(70% r) 7418607.08 7404026.54 -0.20%
1 randread 8053560.83 8046315.44 -0.09%
32 randwrite 15409191.3 15327642.11 -0.53%
32 mix(70% r) 13760145.97 13714666.28 -0.33%
32 randread 16136922.98 16038855.39 -0.61%
128 randwrite 14815647.56 14944902.74 0.87%
128 mix(70% r) 13414858.59 13412317.46 -0.02%
128 randread 15508642.43 15521752.41 0.08%
Change-Id: I4eb5673f49d65d3ff9b930361d2f31ab0ccfa021
Signed-off-by: Richael Zhuang <richael.zhuang@arm.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14743
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Support to specify rr_min_io for multipath round-robin policy,
which makes I/O switches to another io path after rr_min_io I/Os are
rounted to current io path.
Change-Id: I09f0d8d24271c0178ff816fa63ce8576b6e8ae47
Signed-off-by: Richael Zhuang <richael.zhuang@arm.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15445
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Added helper function bdev_nvme_clear_current_io_path for clearing
current_io_path.
The next patch will clear a new variable of the nvme_bdev_channel
when clearing current_io_path. It is convenient to add a helper
function to clear current_io_path.
Change-Id: I7d445638648ea232fedcbf16736e9c4c73cc9c71
Signed-off-by: Richael Zhuang <richael.zhuang@arm.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15681
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Support selecting io path according to number of outstanding io of
each path in a channel. It's optional, and can be set by calling
RPC "bdev_nvme_set_multipath_policy -s queue_depth".
Change-Id: I82cdfbd69b3e105c973844c4f34dc98f0dca2faf
Signed-off-by: Richael Zhuang <richael.zhuang@arm.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14734
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
When we submit more tasks than supported by qp,
extra tasks are queued on io_channel. Later completion
poller tries to resubmit these tasks one by one. That
is not efficient since every enqueu_burst may cause
doorbell updates in HW.
Instead add a check for qpir capacity and submit
appropriate number of requests. If qpair is full,
tasks are queued in dedicated list. This approach
should remove or minimize the need to resubmit
individual crypto operations.
This also handles a case where there are no entries
in global pools (crypto_ops or rte_mbuf)
Fixes issue #2756
Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com>
Change-Id: Iab50e623e7a82a4f5bef7a1e4434e593240ab633
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15769
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Previously vbdev_crypto used DPDK directly and
the restriction on max IO size was propagated to
generic bdev layer which split big IO requests.
Now, when DPDK code is a standalone accel module,
this restriction on max IO size is not visible to
the user and we should get rid of it.
To remove this limitation, allow to submit crypto
operations for part of logical blocks in big IO,
the rest blocks will be processed when all submitted
crypto ops are completed.
To verify this patch, add a functional test which
submits big IO verify mode
Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com>
Change-Id: I0ee89e98195a5c744f3fb2bfc752b578965c3bc5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15768
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
All operations are using iovecs to describe their buffers and only
encrypt/decrypt additionally used nbytes to store the total size of a
src buffer. We don't really need this value in the generic accel code,
so we can let modules calculate it, if necessary. That way, we won't
waste cycles calculating it if a module doesn't use it and it makes the
code a bit easier, as we won't have to deal with the fact that nbytes is
only valid for certain operations.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I29252be34a9af9fd40f4c7fec9d0a0c1139c562d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16306
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Also, since this was the last operation using dst and nbytes, these
fields were removed from spdk_accel_task.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I0d6b090e101c016d1bdcbe7a3bee7d6f691f1c9e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15943
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Also, since this was the last operation using src, remove this field
from spdk_accel_task.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I55fd98697ef4f92a13dd0563b4adf9ccb0af171b
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15942
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Implement the resize function for RAID0. raid0_resize() calculate the
new raid_bdev's block count and if it is different from the old block
count, call spdk_bdev_notify_blockcnt_change() with the new block count.
A raid0 bdev always opens all base bdevs. Hence, if the size of base
bdevs are reduced, resize fails now. This limitation will be removed
later.
Add a simple functional test for this feature. The test is to create
a raid0 bdev with two null bdevs, resize one null bdev, check if the
raid0 bdev is not resize, resize another null bdev, check if the raid0
bdev is resized.
test/iscsi_tgt/resize/resize.sh was used a reference to write the test.
Using jq rather than grep&sed is better and hence replace grep&sed by jq
of test/iscsi_tgt/resize/resize.sh together in this patch.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I07136648c4189b970843fc6da51ff40355423144
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16261
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Also, make it possible to remove copy operations following a fill
operation if they're using the same buffers.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I7da195ce80650a02c5db99d9400ee692f797b1f8
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15940
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Also, replace src2 with an iovec + iovcnt and rename it to s2 to
keep the naming consistent with the source buffer (s).
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I44787128377addd514818ec5aaec084b1a31f0c3
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15939
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Also, replace dst2 with an iovec + iovcnt and rename it to d2 to
keep the naming consistent with the destination buffer (d).
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Ib394c127eeb5890451535ff485f96f7edd2897a4
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15938
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
This patch is first in the series of patches aimed to make all accel
operations describe their buffers with iovecs. The intention is to make
it easier to handle tasks in a generic way.
It doesn't mean that we change the API - all function signatures are
preserved. If a function doesn't use iovecs, we use the aux_iovs array.
However, this does mean that each accel module that provides support for
a given operation will need to be adjusted to use iovecs.
Additionally, update the unit test checking copy elision to verify the
buffers of the copy operation that is left.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I9e6d8d1be3b8b9706cb4a6222dad30e8c373d8fb
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15937
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Factor out block count calculation of raid0 bdev into a helper
function raid0_calculate_blockcnt(). This helper function will be used
to support resize in the following patches.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: Ie99d8686897f1249f48ef8985cb1c8347028fccf
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16283
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The new flag is used in the data processing path.
Change-Id: I93dfc00003d370f8505cc3d334aea161024e2435
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16265
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Non-optimized path was not cached for active/active policy unexpectedly.
Fix the bug in this patch.
This will simplify the code and will be helpful to add more algorithms a
little.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I233d9fa6eca73b1e0fb850c7ea4b3474aff6fc1d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16190
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Richael <richael.zhuang@arm.com>
Unify _bdev_nvme_find_io_path() and bdev_nvme_find_next_io_path()
into _bdev_nvme_find_io_path() by modifying nvme_io_path_get_next().
For active/passive policy, _bdev_nvme_find_io_path() is called only if
nbdev_ch->current_io_path is NULL. Hence, the prev parameter is not
necessary anymore.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: If61b8a24b768a1d571c0033b91d9d9bd487b5cf7
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16189
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Richael <richael.zhuang@arm.com>
nbdev_ch->current_io_path is NULL at initialization.
bdev_nvme_find_io_path() calls _bdev_nvme_find_io_path() if
nbdev_ch->current_io_path is NULL. If there is no optimized path,
nbdev_ch->current_io_path is not filled. Hence, as long as there is no
optimized path, bdev_nvme_find_next_io_path() is never called.
However, bdev_nvme_find_next_io_path() fills nbdev_ch->current_io_path
even if the found path is non optimized.
This is a contradiction and a bug.
To fix this bug, as a preparation, refactor _bdev_nvme_find_io_path() to
cnosolidate bdev_nvme_find_next_io_path() and _bdev_nvme_find_io_path()
into a single function because bdev_nvme_find_next_io_path() and
_bdev_nvme_find_io_path() were almost identical.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I7bc4107ce6ae6f116dee03c9bb67d9e4061ce775
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16188
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Richael <richael.zhuang@arm.com>
When an I/O gets an I/O error, the I/O path to which the I/O was
submitted may be still available. In this case, the I/O should be
retried to the same I/O path. However, a new I/O path was always
selected for an I/O retry.
For the active/passive policy, the same I/O path was selected naturally.
However, for the active/active policy, it was very likely that a
different I/O path was selected.
To use the same I/O path for an I/O retry, add a helper function
bdev_nvme_retry_io() into bdev_nvme_retry_ios() and replace
bdev_nvme_submit_request() by bdev_nvme_retry_io(). bdev_nvme_retry_io()
checks if nbdev_io->io_path is not NULL and is available. Then, call
_bdev_nvme_submit_request() if true, or call bdev_nvme_submit_request()
otherwise. For I/O path error, clear nbdev_io->io_path for
clarification. Add unit test to verify this change.
Linux kernel native NVMe multipath already takes this approach. Hence,
this change will be reasonable.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I7022aafd8b1cdd5830c4f743d64b080aa970cf8d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16015
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Richael <richael.zhuang@arm.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The following patches will clear bio->io_path when the error is a
network path error. As a preparation, cache bio->io_path to a local
variable io_path.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I0a25f4a70dd1064702b2037283bddd6e789dfae1
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16014
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This is in the same effort as gathering all retry functions into the
same place. This will make the following patches easier.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: If2a017ebda45eb1eaf0dc93b15d02b4122dc3be0
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15958
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Relocate abort_retry_ios() in a file to gather all retry functions
in the same place.
I/O retry is very complex and an improvement will be added in the
following patches. This will make the following patches easier.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I84751bd078db4aef753dcc533320cee9aa09c9b2
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15957
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The following patches will change I/O retry to use the same io_path if
it is still available. However, bdev_nvme_submit_request() always calls
bdev_nvme_find_io_path() first. For I/O retry, if possible, we want to
skip calling bdev_nvme_find_io_path() and use nbdev_io->io_path instead.
To reuse the code as much as possible and not to touch the fast code
path, factor out request submit functions from
bdev_nvme_submit_request() into _bdev_nvme_submit_request().
While developing this patch, a bug/mismatch was found such that
bdev_io->internal.ch was different from ch of
bdev_nvme_submit_request(). Fix it together in this patch.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: Id003e033ecde218d1902bca5706c772edef5d5e5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16013
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The new module replaces functionality in vbdev_crypto.
This module is bdev agnostic, so some inernal parts
were reworked.
io_channel: contains a qp of every configured DPDK PMD
crypto key: for mlx5_pci we register a key on each available
device since keys are bound to Protection Domain.
Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com>
Change-Id: If1845cb87eadacbb921c593ba82207a97f2209a3
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14859
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
This patch is just a copy of vbdev_crypto.c and the
corresponding UT file. It makes it easier to review
the next patch which adds accel operations
Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com>
Change-Id: Ib88b45d573b011b1acb35da9bf4dab922d8fb183
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16182
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
