(Note: this patch was previously applied as b32cfc46 and then reverted
as 63642bef.)
Today the in-guest nvme device shows physical_block_size=512 even though
the backend iSCSI bdev supports physical_block_size=4K
iSCSI targets exposes physical block size using
logical_block_per_physical_block_exponent in READ_CAPACITY_16
NPWG is one of the way to let Linux nvme driver set
physical_block_size of the nvme block device.
This patch adds spdk_bdev.phys_blocklen which is updated if the iSCSI
backend exposes physical_block_size.
Later phys_blocklen is used in nvmf to set NPWG and NAWUPF to report
back during NS identity.
Linux driver uses min(nawupf, npwg) to set physical_block_size.
Similarly in scsi_bdev fill lbppbe in READ_CAP16 response
based on spdk_bdev.phys_blocklen.
Fixes#1884
Signed-off-by: Swapnil Ingle <swapnil.ingle@nutanix.com>
Change-Id: I0b6c81f1937e346d448f49c927eda8c79d2d75c0
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7739
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
This reverts commit b32cfc467b.
This commit fails the ABI checks and only got through because the checks
were disabled until 21.04 hit.
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Change-Id: Id26b8f8ba551193d99b1ccbd31b35378b4095a20
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7731
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Today the in-guest nvme device shows physical_block_size=512 even though
the backend iSCSI bdev supports physical_block_size=4K
iSCSI targets exposes physical block size using
logical_block_per_physical_block_exponent in READ_CAPACITY_16
NPWG is one of the way to let Linux nvme driver set
physical_block_size of the nvme block device.
This patch adds spdk_bdev.phys_blocklen which is updated if the iSCSI
backend exposes physical_block_size.
Later phys_blocklen is used in nvmf to set NPWG and NAWUPF to report
back during NS identity.
Linux driver uses min(nawupf, npwg) to set physical_block_size.
Similarly in scsi_bdev fill lbppbe in READ_CAP16 response
based on spdk_bdev.phys_blocklen.
Fixes#1884
Signed-off-by: Swapnil Ingle <swapnil.ingle@nutanix.com>
Change-Id: I0b6c81f1937e346d448f49c927eda8c79d2d75cf
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7310
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
When zcero copy send is enabled and used by initiator,
it could significantly increase latency in some payloads.
To enable more fine graing configuration of zero copy
send feature, add new parameters enable_zerocopy_send_server
and enable_zerocopy_send_client to spdk_sock_impl_opts to
enable/disable zcopy for specific type of sockets.
Exisiting enable_zerocopy_send parameter affects all types
of sockets.
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I111c75608f8826980a56e210c076ab8ff16ddbdc
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7457
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Each fio thread can have multiple files that it writes to.
Which is why the per thread spdk_fio_setup() fio callback does
for_each_file() {...}.
One of these files can be e.g. a zoned namespace with append support,
another file could be a zoned namespace on another controller without
append support, and a third file could be a conventional namespace
(which never supports the zone append command).
Right now, we will return a fatal error if a thread has e.g. a zoned
namespace (with append support) together with a conventional namespace.
Instead of returning a fatal error, enable zone append only on the
namespaces that support zone append, and allow namespaces that do
not support zone append to continue as usual (using regular writes).
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: Ic6456d408cbe91563acd337a4b70c6e871fe34c6
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7611
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Commit f69367c788 ("fio_nvme: defer qpair allocation to file_open
callback") moved the qpair allocation from spdk_fio_setup() to
spdk_fio_open(). This broke spdk_fio_report_zones(), which needs a
qpair in order to get the initial state of the zones.
setup_files() in FIO calls td->io_ops->setup() (spdk_fio_setup()),
followed by zbd_init_files(), which calls zbd_init_zone_info(),
which calls zbd_create_zone_info(), which calls parse_zone_info(),
which calls zbd_report_zones(), which calls td->io_ops->report_zones()
(spdk_fio_report_zones()).
i.e. spdk_fio_report_zones() will always be called directly after
spdk_fio_setup(). .report_zones() is even called before the per
thread ioengine .init() callback.
Therefore, spdk_fio_report_zones() is called before the ioengine
.open_file() callback.
This is done in order to ensure that all threads will share the same
zbd_info struct, which contains the per zone locks.
Since SPDK nvme ioengine no longer initializes the qpairs in .setup(),
create a temporary qpair in .report_zones().
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: Ic376ac7844e40fceff092900ae7e4714bccf38e6
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7590
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Commit f69367c788 ("fio_nvme: defer qpair allocation to file_open
callback") moved the qpair allocation from spdk_fio_setup() to
spdk_fio_open(). This broke --initial_zone_reset, which needs a qpair
in order to perform the initial zone reset.
While at it, move the initial zone reset from spdk_fio_setup() to
attach_cb(), as this is where all the other fio options are verified.
By placing it in attach_cb(), after the duplicated file check, we
avoid the need to loop through the whole fio_thread->fio_qpair list.
Since SPDK nvme ioengine no longer initializes the qpairs in .setup(),
create a temporary qpair, if the --initial_zone_reset option was used.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I7950304c58aef3ec783f7cd99cfb1e7d7817a197
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7589
Community-CI: Mellanox Build Bot
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>
status code and type is inspected and reported
Fix issue #1893
Signed-off-by: Monica Kenguva <monica.kenguva@intel.com>
Change-Id: I6f181d8c9464182b23c658f4c268b900398fd751
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7567
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
All jobs are created at boot, meaning the setup callback
is invoked for all jobs before any are executed.
But it may be useful to put 'stonewall' parameters in
the job file to execute a bunch of workloads in succession,
starting one workload when the previous one completes.
But since qpairs are created currently during setup, the
total number of workloads that can be expressed is limited
since qpairs for all workloads are allocated up front.
So instead defer allocation of the io qpairs until the
file_open callback. These don't get called until the
job associated with the 'file' (in this case, the
nvme namespace) is ready to execute.
Note that we cannot free the qpairs in the file_close
callback, since fio may 'close' the file before all
I/O have been completed.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I3c60cf27c3660a3c94042c0de719f5bebdb9b417
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7481
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: <dongx.yi@intel.com>
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Perf tool allows to specify g_max_completions
parameter but it is not used.
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: Ib46aaa949ddda1b0ba25c28849a402986a7a3d8e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7373
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
SPDK app framework supports a CLI parameter to
specify IOVA mode, add support for this to perf
Change-Id: Id72755d02c63c443ae3bced1823fc547d9e4b050
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7372
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
When perf tool parses CLI arguments, it saves
each argument to a global value. Later these
values are checked for non-zero and copied to
spdk_env_opts structure. It can be simplified
by saving parsed arguments directly to env_opts
structure. It is applicable to env options only,
options specific for perf are still copied to
global variables.
Change-Id: I298392348df959d8165ea72333764c6505373fc2
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7371
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Some earlier NVMe drives don't implement Deallocate
Logical Block Feature(dlfeat) read value field, but
it can return zeroes in deallocated ranges, that's
the quirk NVME_QUIRK_READ_ZERO_AFTER_DEALLOCATE, so
here we use the API to cover earlier NVMe drives.
Change-Id: I6867618c60dd5488a5d4820d663da7f074af229c
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7525
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
When submit_io cb returns bad status, current
task is dropped and allocated memory is not cleaned.
Change-Id: Ibc33e76e6800644c29eaeb826a3401ad5d5fd582
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7376
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>
Function cleanup_ns_worker_ctx is only called for
ns_ctx which have outstanding IO, so when draining
phase is started and some context doesn't have
outstanding IO, it won't be cleaned properly.
To fix this problem, call cleanup_ns_worker_ctx for
every context when drining is finished.
Fixes issue #1880
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I3ce4716ed6ac1369b6f72b03cbcfd7d407f7de55
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7282
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>
Reviewed-by: <dongx.yi@intel.com>
Check if qpair has a poll group during the connect process,
use poll group's statistics or allocate own structure per
qpair. That is done due to not all applications use poll
groups and we want to avoid "if (qpair->group)"
conditions in data path.
Admin qpair always allocates its own statistics
structure but the statistics are not reported
since this qpair is not attached to a poll group.
Statistics are reported by spdk_nvme_perf tool
if --transport-stats and in bdev_nvme_transport_statistics
RPC method.
Change-Id: I58765be161491fe394968ea65ea22db1478b219a
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6304
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Statistics are dumped in the end of perf tool work,
enabled using long option --transport-stats
Change-Id: Ice3755ba82ebcdacfa72ceb9b3d5a1caee911811
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6302
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: <dongx.yi@intel.com>
I was trying to find a free character to add
statistics reporting and didn't fine a suitable
one. Since perf provides a lot of short options,
let's change it to also use long options.
Change-Id: I2a7fd5619e996a40b2d432017992d5f888abb656
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6301
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: <dongx.yi@intel.com>
This option may have been valid at one point, but is
not any longer. So remove it from the usage text.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I83747968085a77fb52d735026294bf5f82517ffc
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7276
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Optargs are missed about -G and -T
which are used to trace and debug.
Signed-off-by: yidong0635 <dongx.yi@intel.com>
Change-Id: I3624b80f52c1b603acf1bc96126b73d1042f1b34
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7109
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Community-CI: Mellanox Build Bot
spdk_nvme_zns_report_zones() is implemented using
nvme_allocate_request_user_copy(), which under the hood will do
a spdk_zmalloc() with the SPDK_MALLOC_DMA flag set, and will copy
over the result to our buffer.
Therefore, it is redundant for us to use spdk_dma_zmalloc(),
because it will cause us to allocate twice the amount of memory
from the precious DMA pool than needed.
Changing this zone report buffer allocation to a calloc also
has the benefit of making the code uniform with all other
spdk_nvme_zns_report_zones() call sites in the SPDK codebase.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: Ia354fa51c66ae07a38a9a57b07c15d145dd609f0
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7005
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
When support for printing the zone report was added,
the zone report buffer was allocated using calloc().
This was intentionally changed to a malloc in commit
5ef79a17ec ("examples/nvme/identify: add an option to dump
the full zns zone report").
While we shouldn't need to zero the buffer, since the drive
should write the "Number of Zones" field in zone report header,
and we should never read zone descriptors beyond this value,
the ZNS spec also states that reading beyond the last zone
descriptor has undefined results.
Considering that "Number of Zones" field in the zone report
header will only represent the number of zone descriptors
in the buffer when the partial bit was set to true,
always use calloc(), to avoid the chance that someone might
copy this code and call spdk_nvme_zns_report_zones() with the
partial bit set to false.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: Ia39c5235aa5c62a4ec42285f53f4bc80f7ec370f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7004
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
If PRACT is enabled, and metadata size is 8 bytes, for extended
LBA format, the controller will insert/strip the metadata, so
we don't need to pass the metadata buffer, so we should exclude
this metadata buffer from host buffer.
So here add a function to calculate host buffer size.
Change-Id: I42d8d9cbfbf7ba2bc4bf64d65260c6cfe9bd4cb1
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6789
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Niklas Cassel <niklas.cassel@wdc.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Purpose: To setup an accelerated function callback
for created spdk_nvme_poll_group. In this patch,
we just create the interface. The real usage of this
call back will be provided in the other patch.
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: I0d936aa4eba4dbfcc0137942156b9f2919eb5b78
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6758
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
For each active namespace, print the command set identifier.
This will work on namespaces that do not support or report a namespace
type, as spdk_nvme_ns_get_csi() will return SPDK_NVME_CSI_NVM for such
cases.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I05fa7fd6bb3d9ea32dac236c98baef90347094ca
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6905
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
If the test run is aborted early (i.e. Ctrl-C), we
should use the elapsed time as the divisor instead
of the originally specified time.
Fixes issue #1800.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I3312cc83f3422ec46ad584e3ee2b40f3ad9ec6f5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6474
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: <dongx.yi@intel.com>
This is a workaround for issue #1799 that would require
a fix from Google Cloud Platform. GCP NVMe SSDs do
not support overlapped GET_FEATURE commands - the
cdw0 value on completions get mixed up.
On GCP the result is that identify app reports only
1 SQ/CQ supported when in fact it supports 16.
We can easily workaround this in the identify app by
submitting one GET_FEATURE and then polling for its
completion before submitting the next one.
We may consider reverting this in the future should GCP
provide a fix, but there is really no harm in keeping
this patch long term since this isn't an I/O path issue.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I536033677a8364c955d562226e3feba4dbad0e07
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6454
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Community-CI: Mellanox Build Bot
print_qid_mappings=1 will now add logging messages
showing the {filename,qid} tuples associated with
each job.
Note that for the nvme plugin, the filename is
essentially the transport ID. We just print that
filename for simplicity rather than reconstructing
a transport ID string from the ctrlr object.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I9b714ac009fd16b96ed87c2c056be251009815b8
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6396
Reviewed-by: Niklas Cassel <niklas.cassel@wdc.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Now when we have support for spdk_nvme_zns_zone_append() and
spdk_nvme_zns_zone_appendv(), hook them up in the nvme fio plugin.
Note that fio itself does not have support for zone append,
since unlike SPDK, there is no user facing zone append API in
Linux. Therefore, this new option simply replaces writes with
zone appends in the SPDK fio backend.
This is however still useful for the following reasons:
-Provides a way to test zone append in SPDK.
-By using zone append, we can test with iodepth > 1.
With regular writes, the user can only specify iodepth=1.
This is because for zone namespaces, writes have to target
the write pointer. Having more than one write in flight, per
zone, will lead to I/O errors.
In Linux, it is possible to use fio with iodepth > 1
on zoned namespaces, simply because of the mq-deadline
scheduler, which throttles writes such that there is only
one write in flight, per zone, even if user space has
queued up more.
Since a user might not want to use zone append unconditionally,
even on a namespace that supports it, make this an option
rather than enabling it unconditionally.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I028b79f6445bc63b68c97d1370c6f8139779666d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6330
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Use the new function spdk_nvme_zns_ns_get_zone_size_sectors() where
it is appropriate (in comparison to the existing
spdk_nvme_zns_ns_get_zone_size() variant).
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: Ic929ffbc5a1f4a16ba6719a985c05ae625caed46
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6417
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Add an optional limit, -z N, to the zone report dump option -z.
The variable g_zone_report_limit replaces the MAX_ZONE_DESC_ENTRIES such
that the maximum-number of zone-descriptors is overwritable. It also
replaces g_zone_report_full as it is represented by the limit-value 0,
e.g. "no limit" dump the full report.
The print of the section-header now includes the total amount of zones
and the limit. With this information, the header's width varies. A
helper-function, print_uline(), for printing an "underline" using a
given marker, is also added.
Signed-off-by: Simon A. F. Lund <simon.lund@samsung.com>
Change-Id: Ic8abead693ed83bb8612eef1f35605098ccade84
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6036
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Despite spdk_nvme_transport_id_parse() checking optarg for NULL, then
the Jenkins CI doing code-analysis fails with the error message:
"Null pointer passed as an argument to a 'nonnull' parameter"
This adds an assertion to satisfy the code-analysis.
Signed-off-by: Simon A. F. Lund <simon.lund@samsung.com>
Change-Id: I35f7ee659624c1d5a2abda91bccac9fb58393063
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6068
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>
Community-CI: Mellanox Build Bot
Identify application prints the PMR details if it is supported
Signed-off-by: Krishna Kanth Reddy <krish.reddy@samsung.com>
Change-Id: Iaba4c15e18e1402035b11a34b2defe8078855751
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6209
Community-CI: Broadcom CI
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
The original value of bs is 4096, we can rewrite it in another script.
When spdk_nvme_ns_supports_extended_lba(ns) is 1, we need to
test with an integer multiple value of extended_sector_size, such as 4160.
Change-Id: I8257bee4f741f62597bb2bb645ee9c59cf57bf33
Signed-off-by: wanghailiangx <hailiangx.e.wang@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6018
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
NVMe namespace ID (i.e., ns) is a key of Transport ID, which is
supported by perf, but it's missed in the help. In addtion, a
note is added to specify multiple '-r' parameters can be used
to test multiple disks/targets.
Signed-off-by: Allen Zhu <allenz@mellanox.com>
Change-Id: Icf55e6213a63d97396bdc7022e4c0d09c67aab9a
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6141
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
It may be an interesting test case to send I/O to hot
removed namespaces, but for a perf tool it seems better
to stop sending them once we get an INVALID_NAMESPACE
status code.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ie458fa63b12114cf05d5a3f1702e37f91d8a00f6
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6091
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
The nvme perf tool can be useful to generate I/O for
stress conditions. But if we use it for situations
where we expect a high rate of failures, the excessive
spew can significantly clutter the log.
So add a new -Q option (meaning "quiet") which will
rate limit these error messages.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia8ab2e1ea1cfab9f43d87bcabe8f3f7589b77cda
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6077
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Next patch will add rate limiting on some of the
error messages. Separating the read error message
from the write error message will allow us to
rate limit them independently.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Id49b64aa1ee545874d7230399a5127c47f217836
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6076
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: sunshihao <sunshihao@huawei.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
We konw bs should be extended_sector_size(ns) * num_blocks.
In other words, bs should be an integral multiple of extended_sector_size.
num_blocks cannot be got here, so we used integral multiple.
Change-Id: Ie521db194cdad6f2d2247fd2704cab92c36ddb82
Signed-off-by: wanghailiangx <hailiangx.e.wang@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5881
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
When NVMf target linked with vfio-user library, we can use
vfio-user client library to connect to the target.
Here is the three examples that can work with target:
identify -r 'trtype:VFIOUSER traddr:/var/run/muser/domain/muser0/8' -g
perf -r 'trtype:VFIOUSER traddr:/var/run/muser/domain/muser0/8' -g -q 1 -o 4096 -w read -t 10
reconnect -r 'trtype:VFIOUSER traddr:/var/run/muser/domain/muser0/8' -g -q 32 -o 4096 -w randrw \
-M 50 -t 10 -c 0xE
You can run the following test script test/nvmf/target/nvmf_vfio_user.sh to have a quick test,
currently enabled with NVMe Identify,Perf,Reconnect tools.
Change-Id: Ieb9842b2f372184fffbf7f23e4aad26feb47c350
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3839
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
There was some confusion from a user on whether thread=1
means only one thread is supported. So add some extra
documentation around this section to clarify this a bit
more, since the option is a bit confusing.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ibcccd1759180e2e6c47e736d94ffbeafc405f7bb
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5995
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: <dongx.yi@intel.com>
Community-CI: Mellanox Build Bot
Perf tool doesn't have handlers for SIGINT and SIGTERM
signals, so when the tool is killed with e.g. ctrl-c
all SPDK and transport resources are destroyed
ungracefully. In the case of RDMA we may have
IO requests inflight and if the request is processed
by the driver when the corresponding MR is destroyed by
the kernel, it may cause an error on the target side.
Such errors are not harmful but it is better to
have a graceful shutdown procedure.
Fixes issue #1549
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I7818a4705d2b5cf4a5f3ca4745c62392312d22d2
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5869
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Michael Haeuptle <michaelhaeuptle@gmail.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Fixes: 5ef79a17ec (add an option to dump the full zns zone report)
max_zones_per_buf was only used in an assert. Per-patch
testing doesn't do a release build, so this wasn't found
until running nightly tests.
So rework the code a bit to print error message and exit
instead of this unexpected condition occurs.
While here, change another error message to use stderr
instead of stdout.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I2c2893089cc82f76d7dd6b569952dd4a9f907ebc
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5949
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Add a new option, -z, to dump the full zone report for
NVMe Zoned Namepaces.
Rather than allocating a buffer that can exactly hold the number
of zones we request, allocate a fixed size buffer of size MDTS.
That way, the controller will copy as many zones as it can,
while not exceeding the buffer. The nr_zones field in the
received report indicates the number of zones that were
successfully copied to the buffer.
The full zns zone report can be larger than MDTS, therefore
we now need to put the spdk_nvme_zns_report_zones() call
inside a loop.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: If3ff84876289a491d0482e74c58c5a11de7e19df
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5896
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Exit after showing the help text.
If the user specifies -H, it is confusing that
the regular execution of the program continues,
which might mislead the user to think that the
option they specified is not taking effect.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I52d8e0f6c720c7ce1d66bddb99181babb12cef1f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5895
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
A new option -O is added to indicate the IO unit size.
Multiple IOVs are created according to the IO unit size.
We are able to test multiple-SGL SGL requests in NVMEoF
RDMA with this patch.
Signed-off-by: Allen Zhu <allenz@mellanox.com>
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Signed-off-by: Evgeniy Kochetov <evgeniik@mellanox.com>
Change-Id: I7624966b585bf0a9d2bbbb6263fa06fbcdb65820
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4377
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: John Kariuki <John.K.Kariuki@intel.com>
Reviewed-by: sunshihao <sunshihao@huawei.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
