We do not support Soft RoCE anymore. Remove a workaround for Soft RoCE's
bug that we amy receive a completion without error status after qpair is
disconnected/destroyed. Then add a assert to check if rdma_req->req is
not NULL. This will simplify the code and the following patches.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I80c349053adc0f79679eaf8a5d7265d555d3c2b0
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14909
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
The following patches will support SRQ and SRQ will be per poller.
We will need SRQ in nvme_rdma_cq_process_completions().
It is not possible to identify poller if poll_group is passed to
nvme_rdma_cq_process_completions().
Based on these thoughts, add poll_group pointer to poller and pass
poller to nvme_rdma_cq_process_completions() instead of poll_group.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Signed-off-by: Denis Nagorny <denisn@nvidia.com>
Signed-off-by: Evgeniy Kochetov <evgeniik@nvidia.com>
Change-Id: I322a7a0cc08bdcc8e87e720ad65dd8f0b6ae9112
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14282
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
NVMe-RDMA target has a helper function get_rdma_qpair_from_wc() and
uses it to identify a qpair from a WC.
NVMe-RDMA initiator has a similar function
nvme_rdma_poll_group_get_qpair_by_id().
NVMe-RDMA initiator will support SRQ in the following patches, and
it will want to identify a qpair from a WC.
get_rdma_qpair_from_wc() of NVMe-RDMA target uses wc->qp_num internally
anyway.
However, the upcoming custom transport for RDMA will have to use other
variables of WC.
Hence, it will be convenient to pass WC instead of qp_num if we consider
future enhancements.
Based on these thoughts, for NVMe-RDMA initiator rename
nvme_rdma_poll_group_get_qpair_by_id() by get_rdma_qpair_from_wc().
remove unnecessary declaration, and pass WC instead of qp_num.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Signed-off-by: Denis Nagorny <denisn@nvidia.com>
Signed-off-by: Evgeniy Kochetov <evgeniik@nvidia.com>
Change-Id: I01ead4730207e2c6ac53b83f151bd5f977a11465
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14279
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Poller will have more shared resources when SRQ is supported.
This is a preparation.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Signed-off-by: Denis Nagorny <denisn@nvidia.com>
Signed-off-by: Evgeniy Kochetov <evgeniik@nvidia.com>
Change-Id: Ic3d1cb93dde3f53653a9536a103e5518cebd58e1
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14173
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
nvme_rdma_ctrlr_disconnect_qpair() does not poll the qpair until it is
actually disconnected if it is in a poll group even if its async mode
is disabled. Hence, spdk_nvme_ctrlr_free_io_qpair() removes the qpair
from a poll group when it is being disconnected.
On the other hand, I/O qpair is destroyed after it is actually
disconnected.
When SRQ is enabled and used, a SRQ is destroyed if the corresponding
poller does not have any I/O qpair after an I/O qpair is removed from
the poller.
In particular, if we use spdk_nvme_ctrlr_free_io_qpair(), a SRQ is
destroyed before the corresponding I/O qpairs are destroyed.
Destroying a SRQ failed because it is still referenced by I/O qpairs.
This bug was found when running the SPDK NVMe perf tool with SRQ.
The reason was we had nvme_rdma_poll_group_process_completions() to call
disconnected_qpair_cb after the qpair is actually disconnected.
However, it is ensured that nvme_rdma_poll_group_process_completions()
calls disconnected_qpair_cb for any disconnected qpair.
Hence, remove a check if qpair->poll_group is not NULL from
nvme_rdma_ctrlr_disconnect_qpair() and update the comment.
Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I0fde0d827eec3280e1cc5a0fce34d163a6069bc4
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14908
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
With RDMA, the admin poller can experience a remote disconnect when
processing completions. The admin qpair will be disconnected to handle
this. The disconnect code path will manually complete queued aborts.
However, the completion callback for the abort will attempt to resubmit
other queued aborts from the queue, which will result in a very large
stack and can eventually cause a segfault.
The fix is to not resubmit queued aborts if the admin qpair is in any
kind of failed state.
Change-Id: I4a6f959232c8a1bd30c87ca50459014e556cbaa0
Signed-off-by: Vasuki Manikarnike <vasuki.manikarnike@hpe.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15114
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Michael Haeuptle <michaelhaeuptle@gmail.com>
A loop inside 'nvme_tcp_qpair_process_completions' makes
'max_completions' actually behaving like a minimum:
do {
rc = nvme_tcp_read_pdu(tqpair, &reaped);
[...]
} while (reaped < max_completions);
Before this change 'max_completion' constraint, in its true sense,
was actually not respected and a loop inside 'nvme_tcp_read_pdu'
could be executed indefinitely as long as a recv state changed.
To prevent this behavior, max_completion must be passed to
'nvme_tcp_read_pdu' and used as an additional exit condition.
Signed-off-by: Szulik, Maciej <maciej.szulik@intel.com>
Change-Id: I28da962f4a62f08ddb51915b5d0dae9611a82dee
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15136
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Some reset/disable paths are freeing the shadow doorbells without
switching the SQs back to BAR0. Fix this up, and add a small cleanup
when initializing the shadow doorbells.
Signed-off-by: John Levon <john.levon@nutanix.com>
Change-Id: Ia5e5b91b7dc696a558eb0ad59cc554abced47cca
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14901
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>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
To support SQs allocated to a poll group other than the controller's
main poll group, we need to make sure to poll those SQs when we wake up
and handle the controller interrupt. As they will be running in a
separate SPDK thread, we will arrange for all poll groups to wake up
when we receive an interrupt corresponding to a vfio-user message
arriving.
This can mean needless wakeups: we don't (yet) have a mechanism to only
wake up the poll groups that correspond to a particular SQ write.
Additionally, as we don't have any notion of a poll group per
controller, this ends up polling all SQs in the entire poll group, not
just the ones corresponding to the controller we were handling.
As this has potential performance issues in many cases, it defaults to
disabled.
Signed-off-by: John Levon <john.levon@nutanix.com>
Change-Id: I3d9f32625529455f8d55578ae9cd7b84265f67ab
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14120
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>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
When crc32c is invoked with a multiple entry input iov,
only the last op has crc_dst set in order to write the final
crc value into the user supplied location.
spdk_idxd_process_events() for every successfully completed
CRC op writes the value into *op->crc_dst
UNLESS it is NULL.
The problem is that _idxd_prep_batch_cmd() that allocates
new ops left op->crc_dst uninitialized.
This results in a memory corruption (use after free)
in the following scenario:
1) op A is allocated an crc_dst is set to point to user memory X.
2) Op A is compeleted
3) User memory X is freed.
4) Ops B and C are allocated (chained), C has crc_dst set.
=> B reused op A memory and crc_dst still points to the
now stale user location (1)
5) B is complered, spdk_idxd_process_events() writes into X
as B->crc_dst = X.
Fix: _idxd_prep_batch_cmd() should initialize crc_dst to NULL.
Signed-off-by: Anton Eidelman <anton@lightbitslabs.com>
Change-Id: I9e7d57ec43a8fbcb3750906015a5cb7291278c35
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15115
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
We were missing a check when ISAL uses the complete output buffer
on compression to determine whether it was s perfect fit or if
simply not enough buffer was provided.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: I73532666f50cb9fbef3c42f6bfb25fc5c7de01c6
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14874
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: Aleksey Marchuk <alexeymar@nvidia.com>
Prevent user from switching back to static scheduler after
different scheduler has been selected. Currently we
do not have a way to save initial thread distribution
configuration, so each time user switches from dynamic
scheduler back to static, the SPDK threads may end up on
different reactors. This would cause discrepancy in
performance statistics of SPDK managed by static scheduler.
Change-Id: Ic17a6be55eaea0e1a748f92e01f7075540403637
Signed-off-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15055
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: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
This helps generate slightly better code in this function,
which can have a noticeable impact for high trace
event workloads.
Tested with bdevperf, single malloc or null bdev,
qd=32, 512B randreads on a single Xeon core.
Specify "-e bdev" to enable bdev trace events.
Null:
Before: 8.09M/s (123ns per IO)
After: 8.68M/s (115ns per IO)
Malloc:
Before: 4.21M/s (237ns per IO)
After: 4.34M/s (230ns per IO)
Note that each bdev I/O generates two trace events (START
and END) - meaning this change removes 7-8ns of overhead
for every 2 trace events, at least on my system.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I7021b7f9e28b4a7cb16f8a97b4d4004ae165efd2
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15096
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
When spdk_idxd_submit_crc32c() handles input
with multiple iovs (or multiple ops are generated
due to physically discontinuous buffers),
the first op has the original seed, while the
subsequent ops instruct the hardware to
to fetch the seed from the output of the previous op
(op->hw.crc32c_val):
void *prev_crc;
...
desc->flags |= IDXD_FLAG_FENCE | IDXD_FLAG_CRC_READ_CRC_SEED;
desc->crc32c.addr = (uint64_t)prev_crc; <<< virtual addr
The problem is the prev_crc is a virtual address,
so the hardware (at least with no IOMMU configured)
reports: DSA_COMP_HW_ERR1
spdk_idxd_process_events: Completion status 0x20
Solution:
Set crc32c.addr to the physical address of
the crc32c_val field in the previous desc.
Since desc->completion_addr already holds the physical address
of the dsa_hw_comp_record, we use this with the crc32c_val offset.
Signed-off-by: Anton Eidelman <anton@lightbitslabs.com>
Change-Id: I330e98c2f3fd6da5cb4fc03d0745df09a9ff0e0c
Signed-off-by: Anton Eidelman <anton@lightbitslabs.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14954
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: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: John Kariuki <John.K.Kariuki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
It allows the users to specify the path to the RPC socket on a NFS
mounted filesystem. This is necessary, because flock(2) on NFS requires
write access to place an exclusive lock.
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Change-Id: If197498ed5bdcb4e02c5f2f2b2c1ef388872c457
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14993
Reviewed-by: Jacek Kalwas <jacek.kalwas@intel.com>
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>
Community-CI: Mellanox Build Bot
When Link Time Optimization is enabled, compiler can sometimes produce
additional warnings saying that some variables may be uninitialized.
To supress the warning it is enough to add explicit initialization
of the variable causing the issue, in this case '*module_name = NULL'
and "*writer = NULL".
Signed-off-by: Szulik, Maciej <maciej.szulik@intel.com>
Change-Id: I30492115b28a18554b08a6f575cbcc9538f3b848
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14849
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jacek Kalwas <jacek.kalwas@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Fixes#2693
spdk threads should not be placed in interrupt mode
if the application does not have interrupt mode enabled.
This resulted in race condition, while reactor was placed
in interrupt mode, thread was scheduled on it.
Such operation is a valid one, but never should be attempt
to change the threads mode in this case.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I10b0bbacac1df812badb91b37064528f66743e51
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14815
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Patch below added copies of pci realted headers to keep
compatiblity with <= DPDK 22.07.
(1eb35ac) env_dpdk: add copies of 22.07 pci-related header files
Unfortunetly the rte_bus/bus_pci/dev headers from DPDK 22.07 are
not compatibile going back to DPDK 20.11.
The issues are:
- lack of RTE_TAILQ_ENTRY defined in rte_os.h
- rte_intr_handle being part of rte_pci_device rather than pointer
pci_dpdk_2207.c even before this patch is not binary compatible with
DPDK 20.11 - see pci_device_*_interrupt_2207() functions.
There would need to be another copy of headers matching that version
of DPDK to resolve this issue.
SPDK supports up to two latest LTS releases. Which right now includes
DPDK 20.11, but soon will be dropped due to DPDK 22.11 release.
Having compile time defines here, keeps the older DPDK working.
Meanwhile backwards compatiblity in SPDK is no worse than before.
The recent changes to env_dpdk, are aiming to improve support
with newer versions of DPDK.
Change-Id: If4dc601cb03e18c2cad61f3a93080e8265ca5fcc
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14795
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@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Add new bdev property split_on_write_unit which, if set to true, causes
writes to be split to match write_unit_size and fail if not aligned to
or not multiple of write_unit_size.
Signed-off-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Change-Id: Id49f58a3288ddf5cfe4921ce4020ae4bcdd67298
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11390
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The client vfio_user library doesn't require this flag as
it is totally owned in SPDK, so remove it.
Change-Id: I8f7b1df18017ceac24dbb8a0417871f25f6bee0d
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/13895
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Previously SPDK use libvfio-user library to provide emulated NVMe
devices to VM, but it's limited to NVMe device type only. Here we
add SPDK vfu_target library abstraction based on libvfio-user which
supports more PCI device types.
We will add virtio-blk and virtio-scsi devices emulation based on
vfu_tgt library in following patches, actually this library can
support NVMe emulation too, due to the fact that the NVMe emulation
is already exist, so we will keep the NVMe emulation which based on
libvfio-user directly as it is.
Change-Id: Ib0ead6c6118fa62308355fe432003dd928a2fae9
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12597
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This allows eliminating dpdk_pci_device_vtophys and
dpdk_pci_device_map_bar, reducing the amount of
code we need to maintain in the per-DPDK version
implementations.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I73d15eb75bf7fe8340d85494425e15651fec5425
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14722
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Break this function up into three APIs instead:
* dpdk_pci_device_get_addr
* dpdk_pci_device_get_id
* dpdk_pci_device_get_numa_node
This more clearly delineates the requirements we
have from the DPDK PCI device/driver APIs.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ie585c8252d63c15c6e6884d60f8a064c3f0ab94f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14684
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Moving forward, we want to still be able to run against
<= 22.07 versions of DPDK, which exposed the necessary
data structures in public header files. But since we
will be building against newer versions of DPDK which
don't expose them publicly, we need a copy of the 22.07
header files in our tree.
Exclude these header files from astyle and POSIX include
file checks in check_format.sh
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Icd8a067af41a2ba031ce8f875a8a2b63f722ab69
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14683
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
This was a remnant from ages ago when we had rte_vhost
DPDK code copied into our repo. We actually have a file
named rte_vhost_user.c which is not DPDK code that was
getting excluded from astyle checking.
So this also includes the astyle violations that had
crept into this file. In a couple of places, change
the enum return type to int, this reduces astyle
confusion on function and if brace style.
Same applies to POSIX include checking - we don't need
to exclude rte_vhost_user.c from this either.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: If3a25011ad54c694c15a91f7be66d862c765c5db
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14688
Community-CI: Mellanox Build Bot
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
For example, in the calling from spdk_bdev_get_current_qd(), if
spdk_for_each_channel() failed to allocate struct spdk_io_channel_iter,
it will just return and the ctx allocated in spdk_bdev_get_current_qd()
is not released.
Instead to change the public API of spdk_for_each_channel() to return
the failed status to let the caller properly handle the NOMEM case and
release the allocation, it just adds the assert here.
Change-Id: I6a95207dd390586bdae4e86e5d550cdac709e10a
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14657
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
max_aq_depth should be not smaller than 2 or greater
than 4096
Signed-off-by: MengjinWu <mengjin.wu@intel.com>
Change-Id: I205fbb4345cfdc41ebaf30c953da263fe9f0e9a8
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14691
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: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Mellanox Build Bot
max_queue_depth should be not smaller than 2 or greater
than 65536
Signed-off-by: MengjinWu <mengjin.wu@intel.com>
Change-Id: I0f2a4b8df6eb1b140a11936fc6929f1285a7d717
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14619
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: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Mellanox Build Bot
Refine the macro definition name about queue depth and
prepare for next patch.
Signed-off-by: MengjinWu <mengjin.wu@intel.com>
Change-Id: I85bee2528ae4ab70292fc11aa62d05bae0c28a77
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14664
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Delete bit masks from trace help (found inside
build/bin/spdk_tgt -h help text), as they do not
provide useful information, are much harder to
remember and use, and migh leave user confused.
Since we provide trace group names anyway, bit masks
are excessive.
Change --tpoint-group-mask parameter name to
--tpoint-group, because we do not provide
bit masks anymore.
Drop "default" tpoint group mask from help text,
since it does not enable any tracepoints and
may confuse the user.
Change-Id: I2ca780883dfa7822e76523e9ba1fc65a7bfe5a99
Signed-off-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14656
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>
When Link Time Optimization is enabled, compiler can sometimes produce
additional warnings saying that some variables may be uninitialized.
To supress the warning it is enough to add explicit initialization
of the variable causing the issue, in this case 'iovcnt = 0'.
Signed-off-by: Szulik, Maciej <maciej.szulik@intel.com>
Change-Id: I080b20a6008643ae78c8e3a6c2d183193ef6c1bf
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14674
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Jacek Kalwas <jacek.kalwas@intel.com>
Community-CI: Mellanox Build Bot
When data_local.num_async_events >
SPDK_NVMF_MIGR_MAX_PENDING_AERS, data_local.async_events
was already indexed by 256, and it was out of bounds.
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Change-Id: I15cfdeb9bc165de0c73fbc9171b0ce6d8689c0aa
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14666
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
If the kernel is booted with the IOMMU enabled and Shared Memory mode
enabled (which are the expected boot parameters for production servers),
then the kernel idxd driver will automatically register a dedicated work
queue with the PASID for the process that opens it. This means that the
descriptors written into the portal for that work queue should be
*virtual* addresses.
If the IOMMU is enabled but Shared Memory mode is disabled, then the
kernel has registered the device with the IOMMU and assigned it I/O
virtual addresses. We have no way to get those addresses from user
space, so we cannot use the kernel driver in this mode. Add a check to
catch that.
If the IOMMU is disabled, then physical addresses are used everywherre.
Change-Id: I0bf079835ad4df1128ef9db54f5564050327e9f7
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14019
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: Konrad Sztyber <konrad.sztyber@intel.com>
The DSA specification calls out that software must use a memory barrier
such as sfence prior to writing a descriptor or incorrect data may be
transferred during the operation.
Change-Id: I12f20e5a748e41616c7a542ccdb158c6b548eea4
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14018
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
By doing the registration immediately upon mapping the BAR instead of
when the memory is inserted into the spdk_mem_map, we're able to
register BARs that are not 2MB multiples in size and alignment. The SPDK
API for registering a BAR already returns the physical/io address in the
map call, and it can be used directly without a call to
spdk_mem_register().
If the user does elect to later register the BAR using
spdk_mem_register(), we attempt to insert the 2MB aligned segments we
can into the spdk_mem_map. Users may still need to register memory for a
few reasons, such as making spdk_vtophys() work, or for setting up the
BAR as a target for RDMA. These cases still require 2MB aligned and
sized segments.
Change-Id: I395ae8803ec4bf22703f6f76db54200949e82532
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14017
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
These variants did not exist in DPDK 20.11 which is
still supported by SPDK.
So we will instead need to scan the rte_version()
string to get these values.
Fixes issue #2715.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I79657002a7a605a38a0d98b944ac53c02fa6d78c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14661
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Pawel Piatek <pawelx.piatek@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
In-capsule data length should be the same with the SGL data length.
Signed-off-by: MengjinWu <mengjin.wu@intel.com>
Change-Id: I7eefecb8baebb76850a48689907aff27a8946f98
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14602
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>
Fixed error handles which are violated with spec:
1. 'data length > MAXH2CDATA' is a fatal error.
2. 'ICDOFF != 0' should abort the IO.
Other errors which are not defined in spec:
1. invalid sgl type
2. In-capsule Data length > In-capsule Data size
Because this function runs before data part receiving, it is hard
to skip the following data segment if we want to handle some error
as non-fatal.
Currently, we have to handle all undefined errors as fatal errors.
I think after this release, we can change receving process. This will
be helpful for error handling. But this work is not small.
Signed-off-by: MengjinWu <mengjin.wu@intel.com>
Change-Id: I8fc0d2d743505e49a93be19fd217e7ad6ca06622
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14580
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>
Fuzzing vfio-user require access to send request api
Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: I6c58b8ab4fd3394150bbb3e64b4f95bff93dae6e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/13881
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
During fuzzing vfio-user client and server are started from same
process causing deadlock. SO_PEERCRED return pid of process
connected to vfio endpoint.
Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Change-Id: I6fc2db5d58a459a30fec116a9de3c69d48acf75e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14559
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This checks the current version to make sure we have
a dpdk_fn_table that supports it.
This is easy for now, since the DPDK PCI API is
public. Moving forward, DPDK 22.11 will likely make
these APIs private, requiring us to carry header file
copies for different DPDK versions so that we can
not only build against DPDK but also use the correct
data strucures and APIs to interact with those private
DPDK interfaces. We will also need to consider
minor (i.e. stable or point) releases since they
could technically change PCI ABI as well - the current
year + month checks won't be sufficient.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ic9f41d9d13778f3d078b20b08da48d8d16362b11
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14637
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
This allows it to return error codes. Have the
init code check the return value and fail the init
process when pci_env_init() returns error.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I7c8a4f9a6da6b3438ed09a881153b7a4ceef3a83
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14635
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Get ready to have multiple implementations of the
dpdk_fn_table. We could do some fancy self-registering
constructor functions, but let's just keep it simple
for now and extern declare each implementation in
the pci_dpdk.h header file.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I8f5621412d1c8bd22c95ab74ef66c5bcc41d1380
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14636
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
