bdev/nvme: add option to control accel sequence support

There are still some limitiations to the way IOs with an accel sequence
are handled: mainly, they'll never be retried by bdev_nvme, because it

bdev/nvme: add option to control accel sequence support

There are still some limitiations to the way IOs with an accel sequence
are handled: mainly, they'll never be retried by bdev_nvme, because it
cannot re-execute an accel sequence (and it doesn't even know whether
that sequence was executed or not).

We plan to address this issue in the future, but, in the interim, a new
option, allow_accel_sequence, was added to allow users to control
whether NVMe bdevs report support for accel sequences (of course the
underlying controller must also support them).

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I560419c605b1f6e2f825a33411e06d120c4d5629
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20014
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

test/bdev: several small chaining.sh improvements

1) `rm -f` ignores non-existing files and arguments, so any checks
before calling it are unnecessary.
2) Changed the test before killing bdevperf

test/bdev: several small chaining.sh improvements

1) `rm -f` ignores non-existing files and arguments, so any checks
before calling it are unnecessary.
2) Changed the test before killing bdevperf, as -v will only ensure that
a variable was declared, but it could still be empty.
3) Quoted all keys referencing the stats array to emphasize that it is
an associative array.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Id230b133f79475ebaa382cb30d37e95b1226b14d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/18870
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jim Harris <jim.harris@gmail.com>
Reviewed-by: Pawel Piatek <pawelx.piatek@intel.com>

show more ...

nvme/tcp: report accel sequence support

That allows users to submit requests with an associated accel sequence
that will be executed prior/after the data is sent/received (depending
on the IO type).

nvme/tcp: report accel sequence support

That allows users to submit requests with an associated accel sequence
that will be executed prior/after the data is sent/received (depending
on the IO type). The support is only advertised if data digest is also
enabled, otherwise there's no benefit in doing so.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: If83808c60b78cddfbe591b23e0239c6b6fd6c6ce
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/18799
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <jim.harris@gmail.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

test/bdev: verify chaining integration with NVMe/TCP

Add an extra case to the chaining test that runs rw traffic with
data digest enabled and verifies that the total number of executed
operations/se

test/bdev: verify chaining integration with NVMe/TCP

Add an extra case to the chaining test that runs rw traffic with
data digest enabled and verifies that the total number of executed
operations/sequences is as expected.

Currently, the TCP transport doesn't report support for accel sequences,
so the operations cannot be chained and the expected number of executed
accel sequences is equal to the total number of operations.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Ie4bad7ffc9b221dc1a008c4664fb3f652c480767
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/18798
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Jim Harris <jim.harris@gmail.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

test/bdev: extend chaining test with bdev layer ENOMEM case

The test already checked ENOMEM handling, but it only used bdevs that
support chaining (crypto, malloc), so bdev layer didn't need to exec

test/bdev: extend chaining test with bdev layer ENOMEM case

The test already checked ENOMEM handling, but it only used bdevs that
support chaining (crypto, malloc), so bdev layer didn't need to execute
any accel operations. So, to force bdev layer to do that, a passthru
bdev was added, as it doesn't support chaining.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I322a65ccebb0f144c759692fff285cfd44bbab4b
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/17766
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>

show more ...

bdev/crypto: complete IOs on ENOMEM from accel

spdk_bdev_queue_io_wait() can only be used when one of bdev submission
functions returns ENOMEM (i.e. there are no more spdk_bdev_ios on that
IO channe

bdev/crypto: complete IOs on ENOMEM from accel

spdk_bdev_queue_io_wait() can only be used when one of bdev submission
functions returns ENOMEM (i.e. there are no more spdk_bdev_ios on that
IO channel). Using it in any other case, e.g. on spdk_accel_append_*()
returning ENOMEM, will most likely result in failure. Therefore, to
avoid that, the IOs are completed with NOMEM status relying on the bdev
layer to retry them.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Ie0f03496e5d3180c481815b3f1b021e74ae2f46d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/17319
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>

show more ...

bdev/malloc: report accel sequence support

This actually allows malloc bdev to chain multiple accel operations
together. And, since the last operation will always be a copy, accel
should remove tha

bdev/malloc: report accel sequence support

This actually allows malloc bdev to chain multiple accel operations
together. And, since the last operation will always be a copy, accel
should remove that copy by modifying previous operation's dst/src.

On my system, it improved bdevperf performance (single core, qd=4,
bs=128k, bdev_crypto on top of bdev_malloc, crypto_sw):

randread: 5668M/s -> 8201M/s
randwrite: 5148M/s -> 7856M/s

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I5b9173fa70a42ee56f56c496a34037d46d2f420f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/17202
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

bdev/malloc: use appends for write requests

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Ief6c873a5f65274a25b67bc3f2811d8f3e4a33b3
Reviewed-on: https://review.spdk.io/gerrit/c

bdev/malloc: use appends for write requests

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Ief6c873a5f65274a25b67bc3f2811d8f3e4a33b3
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/17200
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>

show more ...

bdev/malloc: use appends for read requests

This only changes the interface bdev_malloc uses for scheduling the copy
to appends, but it won't chain those copies to an existing sequence, as
bdev_mallo

bdev/malloc: use appends for read requests

This only changes the interface bdev_malloc uses for scheduling the copy
to appends, but it won't chain those copies to an existing sequence, as
bdev_malloc doesn't report support for accel sequences yet. That will
be changed in one of the subsequent patches.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: I6db2c79b15cb96a1b07c6cf5514004c76b9d2e92
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/17198
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
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>

show more ...

test/bdev: accel chaining test

This test sends several read/write requests and verifies the expected
number of accel operations have been executed.

Signed-off-by: Konrad Sztyber <konrad.sztyber@int

test/bdev: accel chaining test

This test sends several read/write requests and verifies the expected
number of accel operations have been executed.

Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Idda46ef00dc5bcc0a176d3dfb39f3f3861964741
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/17193
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>

show more ...