This is a step toward allowing for multiple maps. Each module may have a
different meaning for placement_id with different uniqueness rules. They
can't all be in the same map.
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Change-Id: I608680a08b947a5d5c0818ff66505ed64e1b891e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7214
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Instead, move it down to the modules. This allows modules
to potentially change the value, if they are able.
Change-Id: I08f5fbadf5d1e96b489ddaaca72aa051ce2cb85c
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7212
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
There are a small, bounded set of placement_ids that the socket layer
will ever encounter, and they remain valid for the lifetime of the
program. The association between a poll group and a placement_id is now
correctly broken when the reference count drops to 0 (in response to
sock_map_release calls), so do not free the entry when the poll group is
destroyed so that it may be reused again.
Change-Id: Iad90e2da7d0860fa8c5cff24f9699bef30cd7bc2
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7210
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Allow the map to have entries with a valid placement_id, but no group.
This will be useful later when the order of placement_id discovery and
group assignment may be reversed.
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Change-Id: Ia39adb3a030135940aeb9eeadf9df78056e59c0d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7209
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: Jim Harris <james.r.harris@intel.com>
We can return error status when processing RELEASE2 without
a reservation, also add a UT to cover this case.
Fix issue #1898.
Change-Id: I56ffa8eabfc0409307500f8740cb627aab9d2f0b
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7379
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
This reverts commit aaac48880d.
This patch was showing issues with SPDK vhost mappings
when handling larger numbers of VMs.
Fixes issue #1901.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I81bd311d26037dcb9340d85abcb4ea45b20a5170
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7424
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>
Change-Id: I2e47b148209ce4c232dbdc5f20c90548be995e1a
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7334
Community-CI: Broadcom CI
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>
This will ensure that we can't exceed the iovcnt when parse
NVMe PRP list to req->iov.
Also comment that the iovcnt in vfio-user transport is used to track
each gpa_to_vva map, for NVMe PRP list command, the PRP2 itself also
will use one entry, so we need add one more entry for this case.
Fix issue #1864.
Change-Id: I06c7137e2c4637c9501f82a9eb1c8e4395d819cd
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7264
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
For NVMe backend device, we should use vtophys to calculate
physical address when doing DMA from/to VM to drives.
Fix#1822.
Change-Id: Ib8fbc371e19e77a20202d408340e7d65644b1eeb
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7261
Community-CI: Broadcom CI
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>
Previously we poll the MMIO callbacks in the context of ADMIN queue's
poll group, here we do some improvement to start a poller to do MMIO
poll, then the group poll will only process NVMe commands while the
MMIO poller will process MMIO access.
This is useful when doing live migration, because the migration region
defined by VFIO is a BAR region, we should stop polling queue pairs
but ack the MMIO accesses during the live migration.
Change-Id: I63bac44889cbe0c31d47599810aab8335dfd4ff5
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7251
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Just code movement for the coming patch.
Change-Id: I7e844bc27a037e086796f9659351f20cdbb517fb
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7333
Community-CI: Broadcom CI
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>
This patch is used to update the field definition related with
work queue in the header file.
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: I68b81d9dfc2497db89e96f0730785be03dcb8add
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7225
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Community-CI: Mellanox Build Bot
NVME_MAX_PRP_LIST_ENTRIES has changed over time, so let's
just remove the reference to the exact value here. Also
explain a bit more why the max size isn't
(NUM_ENTRIES + 1) * page_size.
While here, do a small whitespace cleanup as well.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ib75813788abdd3dbb43192f9fdc27f99b33aeadf
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7328
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>
nvmf_poll_group_remove_subsystem_msg() disconnects all
qpairs associated with controllers in the specified
subsystem. If it finds any controllers that need to
be disconnected, it sends a message to the running
thread to execute the same function again later.
But when it runs again later, the qpair may no longer
be in the poll group, but there could still be
outstanding messages being sent between threads. For
example, _nvmf_qpair_destroy() needs to send a message
to the ctrlr->thread to clear the qpair mask bit.
All of this could result in the nvmf target starting
to destroy poll groups prematurely. Destroy poll
groups results in the nvmf spdk_threads exiting. If
there are still messages being processed from
the STOP_SUBSYSTEMS target state, we can get
use-after-free errors since processing of those
messages could access freed memory associated with
the exited thread.
Fixes issue #1850.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I1e63b9addb2956495a69b5108a41e029f6f9a85d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7275
Community-CI: Broadcom CI
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>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
This include isn't needed in queue_extras.h itself.
There were a few places that were implicitly
depending on this include, so fix those to include
util.h explicitly.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia962ae5a4403ee8ae15f3106d0d5e7d7412a4535
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7172
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@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>
These are interface functions that can be used by
an application e.g. spdk_nvme_perf or bdev_nvme
library. The next patches will add usage of these
functions.
Change-Id: I33b88e0e713c2ea5967f9241885e3257c5070577
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6300
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
The new 2 API function allow to get and free stats
per poll group. New function to get transport name
have been added to report not only transport type but
also the name.
For now only RDMA transport reports statistics,
other transports will be added later.
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I2824cb474fde5fa859cf8196dabac2c48c05709c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6299
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
New statistics include number of poller calls,
number of idle polls and total number of completions.
These statistics allow to estimate % of idle polls
and the number of completions per poll.
Since nvme_rdma_cq_process_completions function
returns number of completed NVMF requests and each
NVMF request consumes 2 RDMA completions (send+recv),
this function was extended to return the number of
RDMA completions.
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: Ifdc1e2e467f645adb5d66d39ff2a379e161fbd77
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6298
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
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>
These statistics allow to estimate WRs batching
efficiency. The number of send WRs equals the total
number of submitted NVME commands.
Change-Id: I96c9836cd6b9070cf5f62e43b4d2738506866e94
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6297
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>
Although currently acking msg_fd inside function
msg_queue_run_batch() will also ack critical_msg's
notification, it is easier to understand the code
if moving acking msg_fd code into
thread_interrupt_msg_process().
Change-Id: I98267c5c28358334a2c1133e3dbc125788de77ab
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7265
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: Changpeng Liu <changpeng.liu@intel.com>
As a start of combining interrupt ability into poller,
it aims to get spdk_thread & spdk_poller runnable between
poll mode and interrupt mode with dynamic switching.
spdk_interrupt_mode_is_enabled() indicate whether interrupt
mode is enabled and dynamic switching is permitted. So
spdk_interrupt_mode_is_enabled==true leads to set up
interrupt mode related resources;
in_interrupt flag indicates whether one spdk_thread now
is running in intr mode.
It is possible that spdk_interrupt_mode_is_enabled==true
but in_interrupt==false. this means spdk_thread & spdk_poller
switched to poll mode from interrupt mode due to heavy
workload coming.
To align with spdk_reactor, use "in_interrupt" to
indicate whether one spdk_thread now runs in intr.
Change-Id: I2cd806bf4dec9969f3df88fac7f6b0c0b716d907
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6540
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: Changpeng Liu <changpeng.liu@intel.com>
Community-CI: Mellanox Build Bot
Restrict spdk_interrupt_mode_enable must be called
once prior to initializing the threading library.
Change-Id: I833ff63fae19882e82154195d03dd7ce56ffb1de
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6707
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: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
This will be helpful to simplify the upcoming change.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I1f170fe48d2ec1b5ea05da6a8aa3589060c5c32d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6438
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Use current_data_offset of task to track the current offset of
large write I/O by following the last patch.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Iec3a371c6050fe11478b6f158259d8f4013f5238
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6424
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Custom transport may not provide the `qpair_abort_request`
callback function, so here for transport API we will just
call it when it's not empty. We will add the callback
support with vfio-user in another patch.
Fix#1883.
Change-Id: Icd82a26bde4ed90068bc85ee04cce9642cb6135d
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7291
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: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
We used the controller ready field to indicate ADMIN queue connection,
but the accept poller and ADMIN poll group may run in different
threads, this may lead vfu_attach_ctx() be called several times, so
change the 'ready' to true when a new socket connection is created.
Fix issue #1854.
Change-Id: Iab6ffd6dffb3fff5cf893e79774bc28fe0b2830c
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7073
Community-CI: Broadcom CI
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>
This will make the code easier to understand.
Change-Id: I7112d3fd5f0d6dce9b66d44375b68ce7d1e8951d
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7072
Community-CI: Broadcom CI
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>
unmap_q is only be called in unmap_qp, so remove this function to make the
code more clear to read.
Change-Id: I627c7a1efdcb85476cb618fced8b0bfc2d8f1f62
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6886
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
When killing QEMU or remote client is terminated normally,
we can release current controller related data structure,
users may restart QEMU to connect the same socket file
again, for the new connection, vfio-user will create
a new controller data structure for it.
Here we add a lock in the endpoint data structure to protect
number of connected queue pairs variable, because controller
data structure is like a session, while endpoint is related
with the socket file, so it's safe here. Moreover, we can
use this lock to protect live migration related data
structures in future.
Change-Id: Ie7060041a253604e7a2242813ec284eae46fe4e8
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6862
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
The following patches will aggregate multiple Data-OUT PDUs into a
single write subtask and we will not be able to use reqh->buffer_offset
to track the current offset of large write I/O to submit write subtasks.
On the other hand, each iscsi_task or iscsi_subtask is only read or write
Hence rename current_datain_offset of iscsi_task by current_data_offset
in this patch.
The next patch will use it to track the current offset of large write I/O
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I922582c5b9474a3c512f81d0f0425158a38a9a8d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6423
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>
The following patches will aggregate multiple Data-OUT PDUs into a
single SCSI write up to 64KB. Any variable to accumulate data length
is necessary.
Hence add data_len to mobj and accumulate read data length into
mobj->data_len, and then refer mobj instead of pdu->data and
pdu->data_segment_len to submit write subtask.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I6354534769e67c0fd995bbc3c2b4a80d21a23915
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6422
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Wrap an operation to get a data buffer from mempool into a helper
function iscsi_datapool_get() and wrap an operation to put a data
buffer to mempool into a helper function iscsi_data_pool_put().
Use inline for both functions.
Besides, as a minor fix, remove duplicated file inclusion between
iscsi.c and iscsi.h.
7
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Ia3005dffaa93a6bca16f19bb467fb5b64ae1aad2
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6366
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: sunshihao <sunshihao@huawei.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Add crc32c to struct spdk_iscsi_pdu and initialize it by SPDK_CRC32C_INITIAL,
and then use it as the initial value of _iscsi_pdu_calc_data_digest().
Separate finalization of crc32c into _iscsi_pdu_finalize_data_digest().
Move the definition of related macro constants from iscsi.c to iscsi.h.
iscsi_pdu_calc_data_digest() is used for read too. So setting
pdu->valid_data_bytes before calling iscsi_pdu_calc_data_digest()
for read.
Data split will be supported only if DIF is disabled, and hence
DIF case is not changed.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I9d24f605fd0d452782e17695b613cd2f63d2e42f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6421
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>
The following patches will want to aggregate multiple Data-OUT PDUs
into the same data buffer, but it will be 64KB at most.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I72eabbeae0b027c2fbff2a5837d180b06b0a1b49
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6418
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The following patches will change the handler for Data-OUT PDU to
submit subtask only when 64KB data is read or F bit is set.
Previously, we had created a subtask when processing header and
before reading data segment. Creating a subtask beforehand is not
convenient for the following changes.
Hence create a subtask after reading data segment.
If LUN is removed while processing the Data-OUT PDU, the corresponding
primary task will be terminated by iscsi_clear_all_transfer_task(),
and any subtask completion is not sent to initiator. Hence we can
reject the received Data-OUT PDU safely.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Ifb6d6988676080b458b31d12fef065f3c1de0cb6
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6415
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>
The hotplug lib can be used for pcie devices
such as nvme, virtio_blk and virtio scsi.
For the sigbus handler, there is only one in a
process and it should handle all the devices.
And align nvme to the hotplug lib
Add the ADD uevent support for allowing the
device hotplug.
Change-Id: I82cd3b4af38ca24cee8b041a215a85c4a69e60f7
Signed-off-by: Jin Yu <jin.yu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5653
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Introduce new utilities NVME_CTRLR_ERRLOG, NVME_CTRLR_WARNLOG and so
on to output the ctrlr's identification at different log levels.
For RDMA and TCP, the subnqn will be output and for PCIe and custom,
the traddr will be ouptput.
Change-Id: I81a112463bf752999aa1fe4e0c867d88e09a2f64
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7057
Reviewed-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Vasuki Manikarnike <vasuki.manikarnike@hpe.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
This can be handled in a cleaner way by having the sock group
create/close operations take an extra reference.
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Change-Id: Id701b6dd9a19b01cd40e0d95eb870aef977eea99
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7208
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
added to a group
The process of adding a socket to a group may, in some scenarios, change
the placement id.
Change-Id: I879d9641099d86978ede5d5e2be1a72eda65a79b
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7207
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: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Also fix the comment. It's never going to make sense to add a socket
to a group twice.
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Change-Id: Id4845b77114aef32bbe4ea0e53d2e1fde8e116f4
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7204
Reviewed-by: Jim Harris <james.r.harris@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>
Community-CI: Mellanox Build Bot
Using nvmf_tcp_destroy() would destroy ttransport->lock, which hasn't
been initialized by that point yet.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: Ie9ced97ef520236dddaa70453b6807e8382ce534
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7235
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: Ben Walker <benjamin.walker@intel.com>
This will be useful as the same purpose as
spdk_io_channel_iter_get_io_device() and will be used in the
following patches.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Id45f5980c65543703b91df2afeb47448232fe503
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7237
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
When a request is submitted, it may have incorrect iov
alignment that doesn't fit PRP requirements. In the
current version an internal function fails such a request
and returns a NULL pointer. This is mapped to -ENOMEM
error which is returned to generic bdev layer where
such a request is queued in a "nomem_io" queue and
later can be resubmitted. That is incorrect and such
a request must be completed immediately. To fail the
request, we need to differentiate between -ENOMEM and
other cases, so we pass a pointer to a result to
local nvme functions
Change-Id: I7120d49114d801497a71fca5a23b172732d088de
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7036
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This function will be used the next patch, current
behaviour remains unchanged
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: Ie74c7395f67a08b0cac018eb5114f358a6b583cb
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7092
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: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
The max_io_size transport option should be a power of 2 and be >= 8KB.
Max data tranfer size is defined in NVMe-oF spec as 2^(mdts cmd field) * 4KB.
Mdts cmd field is calculated as spdk_u32log2(transport->opts.max_io_size / 4096),
so max_io_size < 8KB results in mdts=0, which means no size limit (according to spec).
User can set max_io_size = 0 explicitly to allow no size limit.
Signed-off-by: Maciej Szulik <maciej.szulik@intel.com>
Change-Id: Id88a77efce5f217e1fc7750f61c0bd330aaa3791
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6384
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jacek Kalwas <jacek.kalwas@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
The dbdf format is xxxx:xx:xx.x and with the wrong
format the rte_devargs_parse always fails.
Change-Id: Ia34bc5e68f6401bb25907d5d07c65636b4f491b5
Signed-off-by: Jin Yu <jin.yu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7140
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Community-CI: Mellanox Build Bot
We can send a message to repeat subsystem pause
and free a context that will be used later
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: Ia5e8b0ff43f5e38bd8e659a8a64d42926e1d3c6e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6661
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: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: <dongx.yi@intel.com>
With this change, each polling group will use one
accel_engine channel. This change will be more suitable
to utlize the underlying accelerated device.
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: Ibab183a1f65baff7e58529ee05e96b1b04731285
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7055
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: Aleksey Marchuk <alexeymar@mellanox.com>
Actually we already do this when removeing a memory region, but
the check for it is too strict, we should unmap queue pairs when
the queue pair is in the memory region.
Change-Id: Ia646a0255e32ecdd0a70537a8011ce622eb59195
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6861
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>
The error response should be processed at the beginning of this
function.
Change-Id: Id583951c82981cf58984ab68b23ad6f7ea80cd3f
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6859
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: <dongx.yi@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
When starting VM, there are error logs such as:
vfio_user.c: 510:acq_map: *ERROR*: Map ACQ failed, ACQ 3ffde000, errno -1
vfio_user.c:1043:map_admin_queue: *ERROR*: /var/run/muser/domain/muser1/1: failed to map CQ0: -1
vfio_user.c:1103:memory_region_add_cb: *NOTICE*: Failed to map SQID 1 0x3ffd8000-0x3ffdc000, will try again in next poll
This isn't the error case, because when the Guest memory hot add/remove from QEMU, vfio-user
target will stop and unmap all queue pairs and remap them again, so let's use a more friendly
log instead.
Also use a notice log when adding listener.
Change-Id: Iaa4dc29e02523b5e85ec716d200ec355f8a575ed
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6650
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: <dongx.yi@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Add support in bdev_zone.h for getting the maximum zone append data
transfer size.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I61203e64d51601232c6578a090fa52975364c1f3
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6910
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>
The specification for Maximum Data Transfer Size (MDTS) says this field
should include the length of metadata, if metadata is interleaved with the
logical block data. However, some drives can support MDTS without counting
the interleaved metadata, so for this case SPDK will only use data length
without interleaved metadata length.
Change-Id: I29920a25885699e2689be043b87122367be0e416
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6813
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: Aleksey Marchuk <alexeymar@mellanox.com>
Although the value stored to 'rc' is used in the enclosing
expression, the value is never actually read from 'rc'
Fixes#1860
Change-Id: Id1001552e635968e373cad0fd27d7bda41d887cd
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7082
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
When param len > 0, param data must not be NULL.
So we add a comment to make it clearer.
Change-Id: I053c3e45ddb8fa23fb67ce899d32dadd8e286946
Signed-off-by: wanghailiangx <hailiangx.e.wang@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6618
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Remove the polling group check. Because at this moment,
the qpair is not added into a polling group. If we do
not remove it, we will never enable zcopy feature for
I/O qpair.
And in sock implementmentation, we already fixed the zero copy
handling if a socket is not in a polling group. See
posix_sock_flush function. So we can fix this issue if we directly
remove this check.
Reported by: Aleksey Marchuk <alexeymar@mellanox.com>
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: I969936c4b6c7f13cbfa4d6eb479010c53f3e384a
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/7056
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Community-CI: Mellanox Build Bot
Add print to confirm how groups/queues/engines are being
programmed based on the init RPC used.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: Ic9462c19c6899478a803433f90d9db9249dd5ca1
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6325
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
DPDK has added APIs for registering externally allocated
memory regions. Use them instead of doing our own thing.
We have to postpone spdk_mem_unregister call in
memory_hotplug_cb() because SPDK mutex (g_spdk_mem_map_mutex)
and DPDK mutex (memory_hotplug_lock) may overlap
and cause deadlock when one thread is calling spdk_free()
(locks memory_hotplug_lock first and then tries to lock
g_spdk_mem_map_mutex) and another one is calling
vhost_session_mem_unregister() (locks g_spdk_mem_map_mutex
first and then tries to lock memory_hotplug_lock).
Change-Id: I547b4ffc3987ef088a1b659addba1456ad760a71
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3560
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>
Purpose: The default value of placement_id in spdk_sock
should be set to -1 in spdk_sock_connect_ext. If we still let it to 0 and call
sock_get_placement_id for the spdk socket used in the initiator side,
we will never get the correct placement_id when enable_placement_id configuration
is configured, because we will always get placement_id = 0
instead. And the same comments in spdk_sock_accept function.
And this patch also change the judgement of placement_id in other related places.
PS: Why we need to explictly set default placement_id = -1, because when use
"enable_placement_id=2" for the socket, placment_id=0 is a valid value.
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: I9fcc3a1c6a5007c22d11da5aeed0022577652a76
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6955
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
nvmf_subsystem_remove_listener RPC handler may fail to remove
the listener (e.g. it doesn't exist) but in eror case we
spdk_nvmf_transport_stop_listen_async and send an error
response. In a completion callback passed to
spdk_nvmf_transport_stop_listen_async we try to send a
response again but the response handler had already been
released and we dereference a NULL pointer.
The fix is to skip spdk_nvmf_transport_stop_listen_async
in error case and continue with the subsystem resuming.
Fixes github issue #1821
Change-Id: I8d96b943cca25d9f95d19e8ea600242f019e6b21
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6699
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: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Updating thread stat directly in spdk_thread_poll()
will cover the time spend in msg process in interrupt
mode.
Change-Id: I9b71790281f10fb784ef4fd4059c41438bbaabac
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6722
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
thread_interrupt_msg_process is registered to thread's
fd_group, so it will be called inside spdk_thread_poll.
Since spdk_thread_poll will set/restore tls_thread,
there is no need to set or restore it again here.
Change-Id: Ida10c736ef904ff975eeb42fd0cccad9fd8317cf
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6720
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Move get_rusage directly into reactor_run(), then both
poll mode and interrupt mode can check rusage info.
Change-Id: Id5926752cfb19c13cb969fbfbb35f643e5d49d9a
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6718
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: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
In interrupt mode, reactor spends its valid cpu cycles
to process registered thread interrupt function. So we
can count idle_tsc and busy_tsc in it, and update
reactor's last_tsc in it.
Change-Id: I65f4ae7d3b1e5c7c5c06937d6855f5d1b5c0349f
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6716
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: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
When we iterate qpairs that belong to a subsystem
and try to disconnect them, there is a chance that
some qpair can be disconnected on transport level,
e.g. the initiator may receive a disconnect for
the first qpair and disconnect others. That may lead
to a dead loop when we call spdk_nvmf_qpair_disconnect
with a callback, the callback is called immediatelly
and tries to disconnect the qpair again.
To solve this problem, move part of nvmf_poll_group_remove_subsystem
function to another function nvmf_poll_group_remove_subsystem_msg
which disconnects all qpair at once without any callback
and calls itself via thread_send_msg untill all qpairs are
disconnected.
Fixes github issue #1780
Change-Id: I1000cda73e6164917fc13f7f374366af90571b99
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6597
Community-CI: Broadcom CI
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: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
The NVMe Zoned Namespace Command Set Specification has, in addition to a
Max Open Resources limit, a Max Active Resources limit.
An active resource is defined as zone being in zone state implicit open,
explicit open, or closed.
Create a function spdk_bdev_get_max_active_zones() in the generic SPDK
zone layer, so that this limit can be exposed to the user.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I6f61fc45e1dc38689dc54d5649c35fa9b91dbdfc
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6908
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: Changpeng Liu <changpeng.liu@intel.com>
This change refactors the way nvmf_get_stats RPC works.
The RPC layer passes JSON write context to custom dump function defined within transport ops.
The RPC layer no longer needs to know the structure of transport poll group statictics.
Functions and structures used in the previous flow have been deprecated and will be removed.
JSON returned for RDMA transport should be the same as before this change.
Signed-off-by: Maciej Szulik <maciej.szulik@intel.com>
Change-Id: I03308c45be120793d316bf79814a1295afd9fb95
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6681
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Current update_core_mode is started from the next
core of the master core.
For futher's new scheduler, starting from master
core is required. This change won't impact current
schedulers' behavior.
Change-Id: Ibffd2c93a4288b5e87945ae523ccba88091c4031
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6757
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: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
spdk_nvmf_tgt_listen() is deprecated, so moved
the remaining instance to spdk_nvmf_tgt_listen_ext().
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I32b54e99f83fa10f1074f80aad82bb0608c9ae11
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6630
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Anil Veerabhadrappa <anil.veerabhadrappa@broadcom.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
This API has been deprecated since SPDK 20.07,
see commit (b2947f52).
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: Idb45906c81ea5682c6a67def0265910266d861b5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6629
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Community-CI: Mellanox Build Bot
This patch refactor the pdu sending logic with the async manner,
then if the group contains the accel engine, we can use it.
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: I2d669c0a3255d7a8898441e406906add2f3a3556
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6759
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>
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>
It is invalid to try to delete a NULL qpair, so do
not check for it in nvme_tcp_ctrlr_delete_io_qpair and
return an error when NULL. Just change it to an
assert instead. This makes it consistent with pcie
and rdma.
While here, add an assert in rdma as well.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ic2f76deecb21b78749dac85e33fb1fa0d14a1239
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6917
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: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: <dongx.yi@intel.com>
spdk_bdev_io_complete_nvme_status() had set the bdev_io status to
NVME_ERROR even if it is aborted, i.e, sc is ABORTED_BY_REQUEST.
Fix it to ABORTED, and verify the fix by unit tests.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I6b22547105a6d7986747053f93875854336959b3
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6884
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: Aleksey Marchuk <alexeymar@mellanox.com>
These three goto cases are using device->fd,
so put them in cleanup, it has no impact on
vfio_user_dev_setup failed.
Signed-off-by: yidong0635 <dongx.yi@intel.com>
Change-Id: I28028dda2977cf8158e703afa5b8af38c48f3d85
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6922
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>
Signed-off-by: Jacek Kalwas <jacek.kalwas@intel.com>
Change-Id: I191ad5e3b153fb563256eba1aa695716f66db788
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6377
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
This is an internal API used in several places. The call can fail, so
make sure it can report that correctly.
Change-Id: Iac0ed2c8299c9dd3d2556070278a2224c3807b7b
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6640
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
It is possible that nbd pthread is created but not executed,
then spdk_nbd_stop is call before nbd_pthread's execution,
but nbd pthread starts to execute while nbd is totally stopped.
This patch can get spdk_stop_nbd aligned with nbd pthread.
Change-Id: I57cc92b94d36cd706616c9058134f716f0812892
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6278
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: <dongx.yi@intel.com>
This is better naming to represent their usage.
retry_poller and count can also be used to do
async nbd_stop procedure in the following patch.
Change-Id: Ie5a74e4add3f1a6c7257df00aded8b5d52a09955
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6277
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
spdk_vbdev_register() was deprecated in SPDK 19.04.
config_text field in spdk_bdev_module was deprecated in SPDK 20.10.
spdk_bdev_part_base_construct() was deprecated in SPDK 20.10.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: Ib795ccdf61154c168032ccf8b81ea77e5e663851
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6628
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: Aleksey Marchuk <alexeymar@mellanox.com>
This statistic is incremented when we don't reap
anything from the CQ. Together with the total number
of polls it can be useful to estimate idle percentage.
Change-Id: I61b51d049b0bc506fb8a896e225187e46e75a564
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6295
Community-CI: Broadcom CI
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>
Leverage SO_INCOMING_CPU to get the CPU affinity of connections
(sockets). And allocate the connections to specific poll groups,
which aims to utilize cache locality.
From our test:
6 P4600 NVMe on target,target uses 8 cores, NIC irqs are bound to
these 8 cores, and initiator side uses 24 and 32 cores,
we can get 11%~17% randwrite performance boost for posix, and 8%~12%
for uring.
Change-Id: I011e0a21502c85adcccd4a14fbe9838b43f54976
Signed-off-by: Richael Zhuang <richael.zhuang@arm.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5748
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: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
The shadow registers need to be zero when the qpair is
created. This happens automatically when a given qid
is used for the first time, since the page is allocated
with zmalloc. But if a qid is reused, we need to make
sure its shadow registers are cleared *before* we create
the qpair again with the same qid.
So clear the registers in nvme_pcie_ctrlr_delete_io_qpair,
just after the cq is deleted.
Fixes issue #1795.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I08c30d1ea248559a01b802cd132dd57199b491b5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6752
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
If the iov_len is 0, it is OK for the iov_base to be
NULL.
Reported-by: Yi Ren <yunye.ry@alibaba-inc.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I45c9be68fc2975bf2abd91a9d77935ce516c5210
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6706
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Previously we only process Read/Write/Flush IO commands, we should
not block the DSM command in vfio-user layer if the backend block
device can support it.
Change-Id: Ia6b90397adcc36015f331f011a5bdf3e3d6562d8
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6525
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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>
max_delay_us was deprecated in SPDK 19.04.
config_file was deprecated in SPDK 20.10.
master_core/pci_blacklist/pci_whitelist were deprecated in SPDK 21.01.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: Ie8be4c347de58044a7c3d5b1329d96e47ce084b4
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6594
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Previously the callback parameter for this function is NULL, this will
cause segment fault, so pass the correct parameter here.
Fix#1817
Change-Id: Ie768b7bf4a72862d16a44742ab3032803d0939a2
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6690
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>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Community-CI: Mellanox Build Bot
Can not remove device in the remove event
callback as we can not unregister the remove
callback. So use the alarm_set to fix this issue.
Fixes#1809
Change-Id: Ib86bc4eeecc0fe2bc51538e28684d015405e8835
Signed-off-by: Jin Yu <jin.yu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6553
Reviewed-by: Vasuki Manikarnike <vasuki.manikarnike@hpe.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
nvmf_create_transport rpc parameter to configure the CQ size helps
if the user is aware of CQ size needed as iWARP doesn't support CQ resize.
Fixes issue #1747
Signed-off-by: Monica Kenguva <monica.kenguva@intel.com>
Change-Id: Ia9ba2b5f612993be27ebfa3455fb4fefd80ae738
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6495
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: Aleksey Marchuk <alexeymar@mellanox.com>
iSCSI library had used goto label to consolidate iscsi_reject()
calls but calling iscsi_reject() in return statements will be simpler
and easier to read. This patch series focuses on Data-OUT PDU processing,
and so change goto label to function call in return statements for
Data-OUT PDU first.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I5f30aff764820aab87233ea8cf22263611591a96
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6533
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The spec does not disallow TEXT PDUs with no data. In that
case, just return immediately from iscsi_parse_params.
This avoids a NULL pointer dereference with a TEXT PDU that has
no data, but CONTINUE flag is set.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I2605293daf171633a45132d7b5532fdfc9128aff
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6319
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
instead appending to output file (which occurs on each make execution)
sed is used to modify `Requires` section of the *.pc file
Signed-off-by: Jacek Kalwas <jacek.kalwas@intel.com>
Change-Id: I1a8cb1ec35bf583293c7174a413302191bbbd735
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6460
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: Jim Harris <james.r.harris@intel.com>
These statistics can help to estimate efficiency of
Work Requests batching and show busy/idle polls ratio.
Send: the doorbell updates statistics for verbs
provider are incremented per each ibv_post_send call,
for mlx5_dv per each ibv_wr_complete call.
Recv: the doorbell updates statistics for both
providers are updated when either ibv_post_recv
or ibv_post_srq_recv functions are called.
Each qpair on initialization accepts an optional
pointer to shared statistics (nvmf/nvme poll groups).
If the pointer to statistics is not provided then
qpair allocates its own structure. That is done
to support cases when NVME RDMA initiator doesn't
use poll groups, so we can avoid checks that qpair
has statistics in IO path
Change-Id: I07dea603cb870b85ea23c42e8e2c4520b1c66252
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6293
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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>
This change follows the large read which submits only subtasks, and
simplifies large write cases.
Associate the PDU which sends a SCSI Write PDU with immediate data
with both the primary task and the first secondary task. Then stop
incrementing reference count of the primary task twice.
As same as the last patch, copy the failure status directly among
the primary task and the secondary tasks because the primary task
is not submitted now. Then remove related data from struct
spdk_iscsi_task and related helper functions from conn.c.
Finally simplify unit tests for process_non_read_task_completion().
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I54aa38c9b9fb7d7352da040dcdd8bcc1b1756a83
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6344
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
For the following nvme controller statemachine states:
NVME_CTRLR_STATE_IDENTIFY_NS
NVME_CTRLR_STATE_IDENTIFY_ID_DESCS
NVME_CTRLR_STATE_IDENTIFY_NS_IOCS_SPECIFIC
The statemachine can either:
- Jump to succeeding state
- If active ns list is empty, jump directly to NVME_CTRLR_STATE_CONFIGURE_AER
- In the unlikely case if NVMe completion error, jump to NVME_CTRLR_STATE_ERROR
Simply this such that we either:
- Jump to succeeding state
- In the unlikely case if NVMe completion error, jump to NVME_CTRLR_STATE_ERROR
This will help to reduce the complexity of the nvme controller statemachine,
especially considering that there are new additional states
(NVME_CTRLR_STATE_IDENTIFY_NS_DIRECTIVE and
NVME_CTRLR_STATE_CONFIGURE_NS_STREAMS) currently on review that would continue
with the bad habit of having three possible jump states instead of just two.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I3242052b1108afcd8adbe6d0378b1358fef58ec8
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6521
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: sunshihao <sunshihao@huawei.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
bus_pci depends on pci so it shall be listed before, otherwise it can
result in linking issue e.g.
/usr/bin/ld: /home/jkalwas/spdk/dpdk/build/lib/librte_bus_pci.a(bus_pci_pci_common.c.o): in function `pci_parse':
pci_common.c:(.text+0x6e): undefined reference to `rte_pci_addr_parse'
Signed-off-by: Jacek Kalwas <jacek.kalwas@intel.com>
Change-Id: Idff446df82c37844edc122d5171e8ffa684b296f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6404
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Purpose: To get the optimal group, we need the socket information.
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: I17b048a402fbf002307dd225f64b20a9f876d642
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3324
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>
Community-CI: Mellanox Build Bot
This patch is used to leverage accelerated engine to compute
the data digest in the following case:
1 DIF is not used.
2 The data to compute is aligned with size 4, i.e, %4 = 0.
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: I51fb6e3ab04391062b244cba6e249c8e20d3180f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6014
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>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This patch added the chained crc32 support API for both batched
and non batched mode usage. And also update the accel_perf
program in order to use the revised accelerated crc32 function.
For example, you can use the following command:
./build/examples/accel_perf -C 4 -q 128 -o 4096 -t 5 -w crc32c -y
In this command, "-C 4" means that caculate the chained
crc32 for an iov array.
(even if you do not have the accelerated DSA hardware)
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: Ifede26f9040980b5791da8e5afef41177eede9f6
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6457
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: Changpeng Liu <changpeng.liu@intel.com>
For a SGL using PRPs, there is always an alignment check of the start
address in the beginning of the loop. This is stored in start_valid.
If the start is indeed valid, we might fetch a new SGE,
and then perform a second alignment check on this new SGE.
However, this second alignment check is done unconditionally,
meaning that for the last SGE in a request, we check if the
same start address is aligned twice.
Only perform the second alignment check if we actually fetched
a new SGE.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I9df8038c650b0879f838d1d9d895e8dd7172840d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6493
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: Aleksey Marchuk <alexeymar@mellanox.com>
The statement causes this issue is:
assert(group_impl->num_removed_socks < MAX_EVENTS_PER_POLL);
The call trace is:
The previous solution is:
commitid with: e71e81b631
But with this solution, it will always add the sock
into the removed_socks list even if it is not under polling
context by sock_group_impl_poll_count. So it will exceed the size of
removed_socks array if sock_group_impl_poll_count function will not be
called. And we should not use a large array, because it is just a workaround,
it just hides the bug.
So our current solution is:
1 Remove the code in sock layer, i.e., rollback the commit
e71e81b631. This patch is
not the right fix. The sock->cb_fn's NULL pointer case is
caused by the cb_fn of write operation (if the
spdk_sock_group_remove_sock is inside the cb_fn). And it is not
caused by the epoll related cache issue described in commit
"e7181.." commit, but caused by the following situation:
(1)The socket's cb_fn is set to NULL which is caused by
spdk_sock_group_remove_sock by the socket itself
inside a call back function from a write operation.
(2) And the socket is already in the pending_recv list. It is
not caused by the epoll event issue, e.g., socket A changes Socket B's
cb_fn. By the way, A socket A should never remove a socket B from a polling group.
If it really does it, it should use spdk_thread_sendmsg to make sure
it happens in the next round.
2 Add the code check in each posix, uring implementation module.
If sock->cb_fn is NULL, we will not return the socket to the active socks list.
And this is enough to address the issue.
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: I79187f2f1301c819c46a5c3bdd84372f75534f2f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6472
Reviewed-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Add a function to get the number of max active zones for a zoned
namespace.
The value inside the identify namespace struct is a 0's based value,
where 0xffffffff means unlimited.
If unlimited, the addition will overflow and return 0,
which is the intended value to represent unlimited for this API.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: Ia09e3db157ca0afadbd3ca4032eedd7bcd88248c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6443
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: sunshihao <sunshihao@huawei.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Add a function to get the number of max open zones for a zoned
namespace.
The value inside the identify namespace struct is a 0's based value,
where 0xffffffff means unlimited.
If unlimited, the addition will overflow and return 0,
which is the intended value to represent unlimited for this API.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I4223146bc1ddf90486892a0af5fe5ce006dc5fd3
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6442
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: sunshihao <sunshihao@huawei.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
This patch provides two new accelerated crc32c function interface.
And the next patch will be used to add the real support of chained crc32c feature.
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: I3f8dd55c3da636e29e5fb02fc229b51f05653cd6
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6456
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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>
When read is split, only secondary tasks are submitted. Hence we can
copy the failure status directly among secondary tasks and primary
task now.
Additionally, improve the comment in the source code to make us easier
to understand.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I857711dfaf90515231048f8c31c9273eac854d28
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6343
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
This will make the current code simpler and make the following changes easier.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I5a06f7e876fee03ed05d880525b594f92cadcdca
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6410
Reviewed-by: Ziye Yang <ziye.yang@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>
Community-CI: Mellanox Build Bot
data_buf was duplicated with data and was not necessary. Hence
remove it and use data instead in this patch.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I207047ce73d938f83e39f1454d44a9e4bba6b2f7
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6407
Reviewed-by: Ziye Yang <ziye.yang@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>
Community-CI: Mellanox Build Bot
This patch follows the last cleanup.
Factor out reading PDU payload operation from iscsi_read_pdu() into a
helper function iscsi_pdu_payload_read(). This reduces the nesting
level, improves the readability, and make the following patches easier.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Ie5f51eedefe00f3b43a7b45dcf84be79f8df4e27
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6414
Reviewed-by: Ziye Yang <ziye.yang@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>
Community-CI: Mellanox Build Bot
For the logic in ISCSI_PDU_RECV_STATE_AWAIT_PDU_PAYLOAD case,
this change will make it easier to read.
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: Iccc582dd5c749c60b3d22b2b9b73fb8407e59b0d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6360
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
SPDK block devices can only be resized up when
it is open. So there is no need to pause the
associated namespace itself when resized - just
pausing the subsystem is enough.
Also modify the ns_hotplug_test to do null bdev
resizing - this will help test this resize code path.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I3cb7b9de0892c296f2abf2280bed434d18ebe6b5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6467
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: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Provide a default stub definition for spdk_pci_device_claim/unclaim
for non-linux platforms, rather than just for FreeBSD.
Signed-off-by: Nick Connolly <nick.connolly@mayadata.io>
Change-Id: Ica45d967878582d9a58e37b088eba4bf0d94104e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6464
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Community-CI: Mellanox Build Bot
In _set_thread_name, use pthread_setname_np as the default for
platforms that are not Linux or FreeBSD; it's the most common
'non-portable' pthread extension used to set the thread name.
Signed-off-by: Nick Connolly <nick.connolly@mayadata.io>
Change-Id: Ia841166f0537cd1303eded15bc7ef1a9f03e3b6e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6465
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>
bdev channel is used in nbd fini process, so it should
be released in the latter part of nbd_stop
Change-Id: I87edea63d2d91954cc41cdb71261485ae24c0d9f
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6280
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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: <dongx.yi@intel.com>
Similar issue was fixed in
813869d823
nvmf: Fix possible race condition when adding IO qpair
This patch fixes the same issue which occurs a bit later,
when a message is delivered to another thread. This issue
occurred on CI, callstack is the following:
00:11:46.296 #6 0x00007f2705199f05 in __ubsan_handle_type_mismatch_v1 () from /lib64/libubsan.so.1
00:11:46.296 No symbol table info available.
00:11:46.296 #7 0x00007f27067ace6f in ctrlr_add_qpair_and_update_rsp (qpair=0x221edc0, ctrlr=0x1dc4ea0, rsp=0x2242918) at ctrlr.c:230
00:11:46.296 __PRETTY_FUNCTION__ = "ctrlr_add_qpair_and_update_rsp"
00:11:46.296 __func__ = "ctrlr_add_qpair_and_update_rsp"
00:11:46.296 #8 0x00007f27067b1d0b in nvmf_ctrlr_add_io_qpair (ctx=0x2242540) at ctrlr.c:534
00:11:46.296 req = 0x2242540
00:11:46.296 rsp = 0x2242918
00:11:46.296 qpair = 0x221edc0
00:11:46.296 ctrlr = 0x1dc4ea0
00:11:46.296 __func__ = "nvmf_ctrlr_add_io_qpair"
00:11:46.296 #9 0x00007f27062553ce in msg_queue_run_batch (thread=0x1cff540, max_msgs=8) at thread.c:553
where line 230 in ctrlr.c was
assert(ctrlr->admin_qpair->group->thread == spdk_get_thread());
That means that admin qpair was disconnected from the poll
group and controller is in the process of destruction
Change-Id: I818ba56adda5ed3488a8df78483c0b6839758192
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6364
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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>
We already have support for spdk_nvme_zns_zone_append(),
add support for spdk_nvme_zns_zone_appendv() (zone append with
NVME_PAYLOAD_TYPE_SGL).
_nvme_ns_cmd_rw() currently performs verification of the SGL,
if the parameter check_sgl is set. This parameter is set for all
calls with payload of type NVME_PAYLOAD_TYPE_SGL.
In order to be able to perform the same check_sgl verfication on
zone append vectors, we need to refactor _nvme_ns_cmd_rw() a bit.
Setting check_sgl ensures that _nvme_ns_cmd_split_request_sgl() or
_nvme_ns_cmd_split_request_prp() gets called.
These functions will split an oversized I/O into several different
requests. However, they also iterate the SGE entries, verifies that
the total payload size, total SGE entries is not too many, and that
buffers are properly aligned. A proper request will not get split.
For zone append, splitting a request into several is not allowed,
however, we still want the verification part to be done, such that
(e.g.) a non first/last SGE which is not page aligned, will cause
the whole request to be rejected.
(In the case of spdk_nvme_ns_cmd_write(), a non first/last SGE which
is not page aligned will instead cause the request to be split.)
An alternative would be to try to rip out the verification part from
_nvme_ns_cmd_split_request_sgl() and _nvme_ns_cmd_split_request_prp().
However, that is non-trivial, and would most likely end up with a lot
of duplicated code, which would easily get out of sync.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I2728acdcadeb70b1f0ed628704df19e75d14dcca
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6248
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.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>
Sometimes it is more optimal to get the zone size in number
of sectors, instead of in number of bytes.
Therefore, add a new spdk_nvme_zns_ns_get_zone_size_sectors()
function to get zone size in number of sectors.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I0fe67e00a3d74dd27acfc895ae97448d995b89a3
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6416
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>
In order to make sure we do always response to the kernel module if
there are valid commands in the socket. If we do not see this,
we will see stuck request kernel info in nbd module. And the kernel
will print the timeout message of nbd module again and again.
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: I2ecc3e9c948231a712778f0126e2ecc6220e1d3c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6276
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Community-CI: Mellanox Build Bot
The current implementation treats HPDA/CPDA as the absolute offset
to the beginning of the PDU where the payload data starts. This is
incorrect. The HPDA/CPDA actually specify where the payload data
should start such that the starting location is a multiple of HPDA
(for C2H PDU) or CPDA (for H2C PDU or CapsuleCmd PDU).
The other issue fixed is that the current implementation calculates
padding only when header digest is enabled. This is also incorrect.
Signed-off-by: Wenhua Liu <liuw@vmware.com>
Change-Id: If7a3896a4c1d73f6d062bd3dbe6a912d31771180
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6256
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: Aleksey Marchuk <alexeymar@mellanox.com>
For correct behaviour, pthread_mutex should not be locked after it has
been destroyed.
g_bdev_mgr.mutex is statically initialized. It is destroyed in
bdev_mgr_unregister_cb, but not re-initialized in spdk_bdev_initialize.
Repeated calls to initialize/unregister occur during unit tests.
Remove the destroy from bdev_mgr_unregister_cb, which seems
the simplest way of resolving the issue.
The sequence: spdk_put_io_channel(), spdk_bdev_close(),
spdk_bdev_unregister() occurs during unit tests.
spdk_bdev_unregister() destroys internal.mutex which is then
locked by a call to bdev_channel_destroy() resulting from the
earlier spdk_put_io_channel(). Move the destroy and the free of
internal.qos into bdev_destroy_cb so that they don't occur until
all of the channels have been released. Remove the no longer
required bdev_fini.
Repeat calls to spdk_bdev_unregister that occur after an unregister has
completed will lock internal.mutex which has been destroyed by the
previous unregister. This occurs during unit tests. Defer locking
internal.mutex until after the internal.status has been checked for
SPDK_BDEV_STATUS_REMOVING. This is the only place where
internal.status is set to removing and g_bdev_mgr.mutex alone is
sufficient to ensure atomicity here.
Tested with a pthreads library that contains debugging code to
check the mutex state and a modified version of bdev_io_types_test
to call get_io_channel on a different thread.
Suggested-by: Jim Harris <james.r.harris@intel.com>
Signed-off-by: Nick Connolly <nick.connolly@mayadata.io>
Change-Id: I81cc46a1b8a766700253829b19cc86c7f0eb79f2
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6217
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Vhost is enabled by default, so rte_net was always included.
When disabled, rte_power failed as it depends on rte_ethdev and rte_net.
rte_vhost was only possible to enable on Linux, so there
is no conflict with adding it next to rte_power under this condition.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I2e183004d6457e404471740a0540dcb08aa738d8
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6398
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Jacek Kalwas <jacek.kalwas@intel.com>
Reviewed-by: Nick Connolly <nick.connolly@mayadata.io>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@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>
In dpdk 19.11 version, RTE_VHOST_USER_ASYNC_COPY is not define.
After dpdk 20.08, we can use RTE_VHOST_USER_ASYNC_COPY.
Use version check to avoid this problem.
Signed-off-by: sunshihao <sunshihao@huawei.com>
Change-Id: Iaf9914e8380f3d54cded1e2f16af6a7dc3504f95
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6274
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: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Community-CI: Mellanox Build Bot
There is a special case when using 8-byte metadata + PI + PRACT
where no metadata is transferred to/from controller.
Since _nvme_ns_cmd_rw() already calculates the proper sector size
using _nvme_get_host_buffer_sector_size(), which takes PRACT into
account, change the sectors_per_max_io calculation to also take
PRACT into account.
This will avoid certain requests that don't need splitting getting
split.
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I8d450d37c2458453701189f0e0eca4b8fe71173b
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6247
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
After setting io timeout, host can avoid nbd io
stuck or kernel hang occasionally caused by nbd
stop or underlying bdev removal.
Change-Id: I4ba2a0af7ff7bed369cdaf86121f082136dc1a0b
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6191
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: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Aspects of bit fields are 'implementation defined'. On some platforms
alignment will occur if two adjacent fields are of different types. This
occurs in spdk_nvme_feat_async_event_configutation after the crit_warn
member which is effectively an int8_t, followed by an int16_t. There
isn't a generic way of changing the compiler's behaviour, so the best
options are:
- Change crit_warn to a uint32_t bit field and copy the value to/from
a spdk_nvme_critical_warning_state variable to use it. This requires
changes to code using the field.
- Adjust the structure definition to use smaller types to avoid the
problem. This preserves existing semantics, but the field order will
need to be reviewed if big-endian support is ever added (other places
in nvme_spec.h will need similar attention). A second reserved field
is required.
Use smaller types which seems the most straightforward option. Adjust
the use of the spdk_nvme_feat_async_event_configuration reserved fields
in lib/nvmf/ctrlr.c.
The new structure is binary compatible and the fields behave in the same
way, with the exception of an additional reserved field, so updating
CHANGELOG.md probably isn't necessary.
Signed-off-by: Nick Connolly <nick.connolly@mayadata.io>
Change-Id: I7d8163c84b4f410fc95a5b7064506ad7b4b62c6c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6340
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
__builtin_clzl takes an unsigned long argument which may be smaller
than uint64_t on some platforms. GCC silently ignores the mismatch,
returning the wrong answer at runtime. Use __builtin_clzll instead and
add static assertions to detect the issue.
Attribute 'target_clones' requires 'ifunc' support which only applies to
ELF targets. Add check for defined(__ELF__).
Signed-off-by: Nick Connolly <nick.connolly@mayadata.io>
Change-Id: Iff76640b34223649de531250ad40471d829512c7
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6317
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: Aleksey Marchuk <alexeymar@mellanox.com>
insert_queue() will copy it to internal data structure, so that
before successful map we don't need to consider the error path.
Change-Id: Id7ea2ef73da7914ea430ea568e7981657016d3f7
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6310
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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>
The CQ is created first, so it's more reasonable to connect
the IO queue pair after creating the SQ.
Change-Id: I196c19a54a015310a3777d9bfca7db8735a4d5b2
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6309
Community-CI: Broadcom CI
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>
add_qp() function is only called when creating NVMe SQ/CQ, so unpack
it into the caller to make the code more clear.
Change-Id: Id5cc1152b1684df980909b2f7d73ed2788c0efb2
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6308
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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>
The clients(QEMU and NVMe vfio-user driver) use shared memory with
NVMf vfio-user target for zero-copied IO processing, when memory
region hotplug happens from clients, the backend target can get
notification via callbacks. Here we rename them to reflect the
action. For now NVMe vfio-user example applications use static
memory model and single memory segment.
Change-Id: Icecbe13883668dd8267019b5fe57d0fef1c68b81
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6307
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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>
We can use the NVMf library ABORT implementation directly, so remove
it in vfio-user.
Change-Id: I0f204a869c53c6a6ce67ad900a64d5bb59ac2aab
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6306
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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>
The group poll context is for queue pair state, so we don't need to
check controller state here, and for the disconnect case below, the ADMIN
queue pair will be removed from group poll.
Also add spdk_unlikely in the poll context.
Change-Id: I5ef32ef3cf41ad757a7cb167e1e1fa32c52a84d6
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6227
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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>
Zone append commands cannot be split.
_nvme_ns_cmd_rw() should never cause a NVME_PAYLOAD_TYPE_CONTIG
zone append request to be split.
This is currently true, but add an assert to make sure that
any refactoring to _nvme_ns_cmd_rw() does not break this promise.
Also add error handling, since release builds are built with
asserts disabled.
Follow-up patches will refactor _nvme_ns_cmd_rw().
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Change-Id: I5fd2440c4c9d6bd8d56f30354b208a9047b64729
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6246
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.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>
Community-CI: Mellanox Build Bot
This function allows applications to specify whether
they wish to allow probing a newly attached NVMe
PCIe SSD.
The env layer will only even probe devices that have
been allowed. By default, this is all devices, but
if the user has specified some list of
allowed PCI addresses (via spdk_env_opts pci_allowed)
then newly attached PCIe devices are implicitly not
allowed. This API allows applications to add
device addresses to the allowed list after the
application has started.
This API will be useful for use cases where multiple
SPDK processes are running on one server, and assignment
of PCIe SSDs to those processes are based on some function
of the SSD's PCIe address.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I919bc267f2ad9130ab5c875ff760a301028b047e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6184
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: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: <dongx.yi@intel.com>
The env layer has a pci_allowed list, which specifies
that only a subset of PCI devices may be attached
by the associated process.
But that doesn't cover PCI devices that are hot-inserted
after the application starts, which is common for
storage/NVMe.
So add a new spdk_pci_device_allow() API which allows
an application to add new devices to the allowed list.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I7bd5ff428d84480d46bc236698daadd019b20b8e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6183
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Fixes#1777.
When a qpair cannot be allocated because the transport connection fails,
the qpair was freed without unlinking it from the other structures.
This was leading to a segfault when attempting to create and free other
qpairs.
Also added a unit test to cover this case.
Change-Id: I74b78d1847f90117248b07203b43a11ff5cfa5d6
Signed-off-by: Vasuki Manikarnike <vasuki.manikarnike@hpe.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6272
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Reviewed-by: Michael Haeuptle <michaelhaeuptle@gmail.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
To allow SO_MINOR updates on LTS for the whole year it is supported,
the major version for all components needs to be increased.
This is to prevent scenario where two versions exists with matching
versions, but conflicting ABI.
Ex. Next SPDK release adds an API call increasing the minor version,
then LTS needs just a subset of those additions.
Increasing major so version after LTS, allows the quarterly releases
to update versions as needed. Yet allowing LTS to increase minor
version separately.
Disabled test for increasing SO version without ABI change, as
that is goal of this patch. This check shall be removed with SPDK 21.04
release.
This patch:
- increases SO_VER by 1 for all components
- resets SO_MINOR to 0 for all components
- removes suppressions for ABI tests
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I44d01154430a074103bd21c7084f44932e81fe72
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6167
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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>
In dpdk 19.11, rte_kernel_driver is the old version, add version check before use the members.
Signed-off-by: sunshihao <sunshihao@huawei.com>
Change-Id: Ic1db37cc0760c7d03692fd2cdcbb6ff1e41f872d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6252
Community-CI: Mellanox Build Bot
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>
Developer convenience - make this based on a specific version.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: I228b8aff6e8957cad5e8c1fae5615b113e16cfb5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5950
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Community-CI: Mellanox Build Bot
In the completion handler there's no need to do an MMIO read
unless the completion record indicates there's an error.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: Ic4850d803a23413f9813da30ac6f1b611804f1b5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5847
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: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
For flow control reasons we have to resize the bit arrays we
use to manage flow as channels come and go. However since
channels are assigned to devices, until the channel count
reaches the device count there's no sharing so no resize of
the array is needed. So, when we use a device for the first
time there's no need to run through the rest of the channels
and re-balance.
Same thing is done on destruction. The code to free idxd
specific resources was moved from the rebalance function to
the idxd put channel function which is a much more logical
place for it as well.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: Ib4df163286906f413dd6429dc6833af7b68e208c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5846
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>
Config #1 remains what is shown as an example in the spec. Change
config #0 to just have 1 work group and 1 work queue all backed
by 4 engines. As the majority of initial use cases will not be
implementing separate priorities and/or different back end
targets (mem, pmem, etc) having just 1 group and work queue makes
the most sense as it allows the silicon to decide which engine to
use.
Also, having multiple work queues spreads out the available
entires such that if we're not using all of the work queues then
we're not using all of the resources. As channels are created
they are assigned the next available device. As a channel is
assigned a device that is already in use it will round robin
work queues. If then, for example, we have 16 devices then only
the first work queue will ever be used for the first 16 threads
which seems and if there are even just 2 work queues per device
it would take 32 threads to use all of the resources at the
device.
By haing just one work queue per device we always have the max
number of work queue entries available regardless of how many threads
are being used.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: Ie15ff6bdea12525fe3bfc769613084ddd2de50bf
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5845
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
If rdma_qp_disconnect is not correctly sent out, we will not wait
for the event.
Change-Id: I99701e421dc93909d481ccf35e9bfd8004e60da8
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6163
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Those spdk threads were neither used nor tracked in event framework.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I1becee6390e172a10e08754703e018f7c44e30df
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6213
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
spdk_vhost_init() relies on having a thread on each reactor.
Every vhost controller could be created on the same core and
even passing --cpumask when creating would not affect it.
This has happened before, see patch (7cc83b62).
This patch modifies the g_vhost_core_mask to match the actual
cores in use.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I42a07c5f99690bfa4ecd2a5b9b7b04d1aa7d2800
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6188
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Documentation for vhost target states that CPU mask must be a subset
of application CPU mask. This wasn't enforced right now and allowed
the cpumask on controller creation so long as at least single
CPU core overlapped with application's CPU mask.
This might have been misleading and covered up user configuration errors.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I03f959ec37efd0be9b98cff9c93c5f996b04af35
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6212
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
As of DPDK main commit 682a6454 ("power: add ethdev
power management"), rte_power depends on rte_ethdev.
So add rte_ethdev library, but only on Linux since
that is the only place where we use rte_power.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ibc3473d6ec75c14a8c0c5517616950017cb2ea96
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6210
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Add the real support in nvme tcp transport.
Change-Id: I2aa9b0284d6fe009925e67f602a055e787f77987
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5734
Community-CI: Broadcom CI
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: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
This patch is used to add spdk_nvme_poll_group_get_optimal
public API.
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: Iee34c89e0e1ff1f81167b18e198c144ca28f71de
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3311
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
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>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This reverts commit 72eed604b4.
spdk_vhost_init() relies on having a thread on each reactor.
Without the revert, every vhost controller is created on the same
core and even passing --cpumask when creating does not affect it.
Proper fix would be to change spdk_vhost_init(),
but would require additional testing. For now revert the patch
that spawned always idle threads.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I0dbd6c9fe1d9d23ada260da8fc7b48086223c632
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6186
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Community-CI: Mellanox Build Bot
Recently, we discovered that under the SPDK + UIO mode, it is possible that vm may be wrong when doing io.
such as:
[root@VM_6_151_centos ~]# mkfs.xfs /dev/vdc
meta-data=/dev/vdc isize=512 agcount=4, agsize=234421142 blks
= sectsz=4096 attr=2, projid32bit=1
= crc=1 finobt=0, sparse=0
data = bsize=4096 blocks=937684566, imaxpct=5
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0 ftype=1
log =internal log bsize=4096 blocks=457853, version=2
= sectsz=4096 sunit=1 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
bad magic number
bad magic number
Metadata corruption detected at xfs_sb block 0x0/0x1000
libxfs_writebufr: write verifer failed on xfs_sb bno 0x0/0x1000
the github issue link: https://github.com/spdk/spdk/issues/1738
After investigation, we found as below:
The qemu set to VHOST_USER_SET_MEM_TABLE then call vhost_user_set_mem_table(dpdk version spdk-20.11)->mmap(without MAP_POPULATE)
the mmap will only return the virtual address not map the virtual address to the physical address actually.
after call vtophys_notify->vtophys_get_paddr_pagemap, we will get a wrong physical(the virtual address we not access)
So this patch is to set MAP_POPULATE during we use SPDK + UIO.
Signed-off-by: andypma <andypma@tencent.com>
Reviewed-by: Zhigang Lu <tonnylu@tencent.com>
Reviewed-by: Xun Ni <richardni@tencent.com>
Reviewed-by: YuZhoujian <windyu@tencent.com>
Reviewed-by: LuWeiKang <luweikang@tencent.com>
Change-Id: Ib921b6d7381f46bb5894ed787cef52e02524197a
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5970
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Since iWARP doesn't support CQ resize, we can't
accept new qpairs when CQ size is insufficient.
In other case if we accept new qpairs we can
et CQ overrun and receive IBV_EVENT_CQ_ERR ibv
async event. In that case we will have to close
all qpairs that use this CQ.
Part of the fix for issue #1747
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I30b9163a7c8abb8651f8813c6bbdb80000741c9c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6106
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: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
When SRQ is enabled, all qpairs share common pool of
rdma requests and we should check that rdma_req which
cid matches belongs to the correct qpair.
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I2505ca39c5110e8e67a48abbe34b6df9ff00229f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6110
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Michael Haeuptle <michaelhaeuptle@gmail.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Additionally, the user can specify a namespace to also pause during the
operation.
This allows for the management of hosts, listeners, and the addition of
namespaces all while I/O to other namespaces is occurring. Pausing a
specific namespace also allows for the removal of that namespace without
impacting I/O to other namespaces in the subsystem.
Change-Id: I364336df16df92fe2069114674cb7a68076de6fb
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4997
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
We want to avoid an active for_each_reactor operation
while the reactors are being shut down.
Fixes issue #1766
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I30bc258c5b22545320080d269a1ed8cb0b4e12f2
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6104
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
The data path for PCIe and vfio-user transports are almost
same too, so move the code from nvme_pcie.c to nvme_pcie_common.c,
so that these APIs can be reused by vfio_user.
No logic change for this patch.
Change-Id: I82f480bba3bae0ce35e2a98f29839081095f7d50
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6040
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
dpdk_governor (and therefore scheduler_dynamic) use DPDK
env. It should not be compiled in non-DPDK env.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: I6b7cf15cf7383e6c1e77bfb188b24cabcc1f0fe1
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6102
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Using reactor once insteads of target to fix this warning.
Signed-off-by: yidong0635 <dongx.yi@intel.com>
Change-Id: I30cd6be26669fe1b2c5d694c15bfda223efd3b85
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6086
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>
There is a chance that admin qpair is being destroyed at
the moment when IO qpair is added to a controller due to e.g.
expired keep alive timer. Part of the qpair destruction process
is change of qpair's state to DEACTIVATING and removing it
from poll group. We can check admin qpair's state and poll
group pointer before sending a message to poll group's thread
and fail connect command.
Logs and backtrace from one CI build that hit this problem:
00:10:53.192 [2021-01-22 15:29:46.671869] ctrlr.c: 185:nvmf_ctrlr_keep_alive_poll: *NOTICE*: Disconnecting host from subsystem nqn.2016-06.io.spdk:cnode1 due to keep alive timeout.
00:10:53.374 [2021-01-22 15:29:46.854223] ctrlr.c: 185:nvmf_ctrlr_keep_alive_poll: *NOTICE*: Disconnecting host from subsystem nqn.2016-06.io.spdk:cnode2 due to keep alive timeout.
00:10:53.374 ctrlr.c:587:41: runtime error: member access within null pointer of type 'struct spdk_nvmf_poll_group'
00:10:53.486 #0 0x7f9307d3d3d8 in _nvmf_ctrlr_add_io_qpair /home/vagrant/spdk_repo/spdk/lib/nvmf/ctrlr.c:587
00:10:53.486 #1 0x7f93077ea3cd in msg_queue_run_batch /home/vagrant/spdk_repo/spdk/lib/thread/thread.c:553
00:10:53.486 #2 0x7f93077eb66f in thread_poll /home/vagrant/spdk_repo/spdk/lib/thread/thread.c:631
00:10:53.486 #3 0x7f93077ede54 in spdk_thread_poll /home/vagrant/spdk_repo/spdk/lib/thread/thread.c:740
00:10:53.486 #4 0x7f93078366c3 in _reactor_run /home/vagrant/spdk_repo/spdk/lib/event/reactor.c:677
00:10:53.486 #5 0x7f9307836ec8 in reactor_run /home/vagrant/spdk_repo/spdk/lib/event/reactor.c:721
00:10:53.486 #6 0x7f9307837dfb in spdk_reactors_start /home/vagrant/spdk_repo/spdk/lib/event/reactor.c:838
00:10:53.486 #7 0x7f930782f1c4 in spdk_app_start /home/vagrant/spdk_repo/spdk/lib/event/app.c:580
00:10:53.486 #8 0x4024fa in main /home/vagrant/spdk_repo/spdk/app/nvmf_tgt/nvmf_main.c:75
00:10:53.486 #9 0x7f930716d1a2 in __libc_start_main (/lib64/libc.so.6+0x271a2)
00:10:53.486 #10 0x40228d in _start (/home/vagrant/spdk_repo/spdk/build/bin/nvmf_tgt+0x40228d)
Change-Id: I0968eabd1bcd532b8d69434ad5503204c0a2d92b
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6071
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.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>
Reviewed-by: <dongx.yi@intel.com>
When using interrupt mode we can have a situation
when we create thread (which is always busy) with
a particular core mask, but this thread will be
scheduled for different core, because core
pointed by thread mask is in interrupt mode.
This thread will never be moved by scheduler
to correct core because currently scheduler
do not move busy threads.
This change makes scheduler to move busy threads
if their mask do not match core on which they
are executed currently.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: I35abdc91b197f1b9d40e491f964d31debad72fa5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6073
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
This function will be useful in upcoming patch.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: I2a79305b8cb155a94c83b4baa3f5d7014cb602c9
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/6079
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Save CPU resources by switching reactors w/o any
threads to interrupt mode.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: Ibd1c1d50d1f6c6d7ed226d585e14999c2fd04621
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5456
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
