10677 Commits

Author SHA1 Message Date
Shuhei Matsumoto
4c577b1fce lib/iscsi: Change spdk_iscsi_is_deferred_free_pdu() to private and rename it
spdk_iscsi_is_deferred_free_pdu() is called only in a single
place of conn.c.  So change it to private in conn.c.  Additionally,
iscsi_is_free_pdu()_deferred() may be a little more meaningful
and rename to it.

Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Ic1e3d7ff435c454f40e81f9a4f90fe76589ec7b2
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477189
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2019-12-12 14:41:41 +00:00
Shuhei Matsumoto
7ff4d22d4c lib/iscsi: Move declaration of spdk_iscsi_task_mgmt_cpl from iscsi.h to conn.h
spdk_iscsi_task_mgmt_cpl()'s function body is in conn.c, and so move
its declaration from iscsi.h to conn.h.

Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Idd22c360ce1d3d464bf782e21348425ae7debdd8
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477188
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2019-12-12 14:41:41 +00:00
Shuhei Matsumoto
3c78f68363 lib/iscsi: Move declaration of spdk_iscsi_task_cpl() from iscsi.h to conn.h
spdk_iscsi_task_cpl()'s function body is in conn.c, and so move
its declaration from iscsi.h to conn.h.

Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I7bc6feecd5a2ff698c8e30e7aab547ada398c44f
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477187
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2019-12-12 14:41:41 +00:00
Shuhei Matsumoto
62d14e2ca1 ut/iscsi: Re-order operations in abort_queued_datain_task_test()
Re-order operations in abort_queued_datain_task_test() to align it
with comment, and fix a type error.

Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Idde4726bfe370ad742b2de67e0786aebb7237d45
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477409
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-12-12 14:41:41 +00:00
Shuhei Matsumoto
19eb63fcad lib/iscsi: Stop splitting immediately when aborting SCSI read I/O
We changed read I/O submission to stop splitting when detecting LUN
hotplug very recently. We can do the same refinement for read I/O
abortion.

In _iscsi_conn_abort_queued_datain_task(), set all remaining length
to the new task and complete it immediately. We keep the code to
process the case that queued_datain_task completed but is still in
queue, but we can change its if condition to assert.

Simplify the corresponding unit tests accordingly, and set
task->scsi.transfer_len in abort_queued_datain_tasks_test() to
exercise the changed paths.

In iscsi_pdu_payload_po_scsi_read(), if task->scsi.transfer_len is not
larger than SPDK_BDEV_LARGE_BUF_MAX_SIZE, no minimum calculation is
necessary and we can substitute task->scsi.transfer_len to
task->scsi.length simply.  This change is too small to be an independent
patch and is done together.

Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Iea93e51b103eae141a007a0abdaf13cbe6d5287f
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476984
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: SPDK CI Jenkins <sys_sgci@intel.com>
2019-12-12 14:41:41 +00:00
Ben Walker
bed4cdf6c7 nvme: Use sgls, if available, even for contiguous memory
The hardware sgl format can describe large contiguous
buffers using just a single element, so it's more
efficient that a prp list even for a single memory
segment. Always use the sgl format.

Change-Id: I9c62582829f0d64dcd1babdbc48930ddb4d9e626
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/475542
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: 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>
2019-12-11 11:06:57 +00:00
suhua
25baf7148d reactor: Change the context_switch period to microseconds
Signed-off-by: suhua <suhua1@kingsoft.com>
Change-Id: I95a4cc3cb5bef37e31a630171a47e28f14331494
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/475333
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>
2019-12-11 11:06:10 +00:00
Shuhei Matsumoto
b99df8ee33 app/rpc: Add cpumask to output of thread_get_stats RPC
Thread's core affinity had not been visible to users.  This patch
adds thread's core affinity information to the existing
thread_get_stats RPC.  We can create an new RPC to retrieve
thread's core affinity information but using thread_get_stats
RPC will be enough for now.

Besides, the name of SPDK thread 0 is not reactor_0 but app_thread
now.  So fix it together in this patch.

Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I634da46c411d648837538ebf227074d307273c97
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/475187
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Alexey Marchuk <alexeymar@mellanox.com>
2019-12-11 11:05:36 +00:00
Jim Harris
b3d982036a env_dpdk: fix --legacy-mem checks
opts->env_context could have more options specified
than just --legacy-mem.  So strcmp() is not a valid
comparison operator - we need to use strstr instead.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ie4c8cbcbe7c141693a07a11648d6673ec8c012e5

Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477087
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2019-12-11 11:05:10 +00:00
alokkataria
6d6052ac96 env: Use rte_malloc in spdk_mem_register code path when possible
For TCMalloc regions which we register with spdk at runtime in the MMapHook, we
need to ensure that SPDK doesn't do any allocations in that path otherwise we
will hit a livelock situation. MmapHook is invoked when TCMalloc is out of free
memory and needs to get more memory from the system, for the hugepage case it
gets via mmap.

In the current code, we could end up calling malloc in the spdk_mem_register
call via the following call path.

spdk_mem_register -> spdk_mem_map_set_translation -> spdk_mem_map_get_map_1gb

To avoid this livelock situation we call rte_malloc instead which shouldn't
invoke the system allocator. Note that in try_expand_heap_primary() which is
invoked in the rte_malloc code path, we can still call malloc, so we need to
only use this when dynamic memory allocation is disabled via --legacy-mem.
It is possible in the future we could work around even this limitation,
but for now this implementation will be much simpler.

Have verified this change fixes the livelock condition which I was hitting in
my setup without this fix.

Change-Id: I69d0813a70da1f26f8c4d9d8895e406c026be18b
Signed-off-by: Alok N Kataria <alok.kataria@nutanix.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/475943
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: Ben Walker <benjamin.walker@intel.com>
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
2019-12-11 11:05:10 +00:00
Jim Harris
396c445cb1 env_dpdk: tell spdk_mem_map_init whether legacy_mem was specified
We will use this in a future patch to determine whether it's safe
to use DPDK allocated memory when allocating new 1gb page entries.

We could use it in this patch to decide whether or not to register
the memory hotplug handler, but there's really no harm registering
it even when it's not needed.

Ideally DPDK would provide some kind of API to query how DPDK was
configured.  In the normal case we know whether legacy-mem was
specified, but if users initialize DPDK themselves and then call
spdk_env_dpdk_post_init(), we won't know if legacy-mem was specified.
So in that case, we will just assume that it wasn't specified.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ied0e5ff777c8ee651043f46a37ce62e44bfcc5fe

Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477086
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: Ben Walker <benjamin.walker@intel.com>
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
2019-12-11 11:05:10 +00:00
Maciej Wawryk
7561d1b5c2 scripts/vagrant: fix scripts for default emulator
Default emulator does not support OCSSD and Multi-namespace NVMe's.
Add if statement to check if user change emulator.

Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: I30858bfed0c2e571c1aefa6ff7ccbf3bc436a294
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476454
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
2019-12-10 17:33:32 +00:00
jiaqizho
ab04d59c57 ut/tcp: add nvmf incapsule data unit test
This patch is used to test the pending buffer policy
of nvme request which has incapsule data. When
group->pending_buff queue not null , we should make
sure spdk_nvmf_tcp_capsule_cmd_hdr_handle method still
can handle the request.

relate to : issues/995

We try to simulate the following manner: If there
is req waiting for the buffer in the list, the
nvme request which has incapsule data will be
handled firstly even it is added into the
pending_buf_queue. The reason is that currently,
those command will get the buffer from its own queue.

We made these test steps:

 1. submit a tcp request into spdk_nvmf_tcp_req_process
    and make sure it pause in group->pending_buff queue
 2. submit a incapsule tcp request into
    spdk_nvmf_tcp_capsule_cmd_hdr_handle method , and
    this method will call the spdk_nvmf_tcp_req_process
    method to handle this request.
 3. check whether the incapsule tcp request is handled
    correctly. and check the group->pending_buff queue
    remain.

Signed-off-by: jiaqizho <jiaqi.zhou@intel.com>
Change-Id: I85fcbb49e309e1203b4b308115e3bfefc0fcba2a
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/472665
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-12-10 17:33:11 +00:00
Changpeng Liu
4f8c85196f nvmf: remove the unnecessary goto sentences
No funtional code change, removed several unnecessary goto sentences.

Change-Id: I4e802246585573be74d185b6bd6b60270d37296c
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477200
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
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>
2019-12-10 17:32:39 +00:00
Changpeng Liu
2e91b69ccc nvme: remove old spdk_nvme_async_event_config definition
The spdk_nvme_async_event_config structure was very old name definition,
since it already has the new name, so remove the old name support finally.

Change-Id: Ib4f592298bd6f1de2f2fb78814767ed2b2335dea
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477199
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Alexey Marchuk <alexeymar@mellanox.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>
2019-12-10 17:32:39 +00:00
Evgeniy Kochetov
54f81b37ef nvme/rdma: Add 'delay_cmd_submit' option to RDMA transport qpair
Signed-off-by: Evgeniy Kochetov <evgeniik@mellanox.com>
Signed-off-by: Sasha Kotchubievsky <sashakot@mellanox.com>
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I12e12d37baf1f74717a60a4f9d8309a994509e42
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/475308
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>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
2019-12-10 17:32:10 +00:00
Evgeniy Kochetov
d18874592d nvme/conf: Make 'delay_cmd_submit' configurable via config file
Option 'DelayCmdSubmit' has been added to 'Nvme' section of
configuration file.

Signed-off-by: Evgeniy Kochetov <evgeniik@mellanox.com>
Signed-off-by: Sasha Kotchubievsky <sashakot@mellanox.com>
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: Ie367d0012d8274b9bf12c0377d1c4804a700ac94
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/475307
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>
2019-12-10 17:32:10 +00:00
Evgeniy Kochetov
542be8790b nvme/rpc: Make 'delay_cmd_submit' configurable via RPC
Optional parameter 'delay_cmd_submit' has been added to
'bdev_nvme_set_options' RPC method.

Signed-off-by: Evgeniy Kochetov <evgeniik@mellanox.com>
Signed-off-by: Sasha Kotchubievsky <sashakot@mellanox.com>
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: Ia09c57bbac8078ee6ab1e8f0bd279ffeb9e08cb8
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/462584
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>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
2019-12-10 17:32:10 +00:00
Evgeniy Kochetov
38520df476 bdev/nvme: Add 'delay_cmd_submit' to nvme bdev options
Signed-off-by: Evgeniy Kochetov <evgeniik@mellanox.com>
Signed-off-by: Sasha Kotchubievsky <sashakot@mellanox.com>
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I7dada3265f57dda61202797e5a02251b28dd0b30
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/475306
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.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>
2019-12-10 17:32:10 +00:00
Evgeniy Kochetov
ffc7c4a9a9 nvme: Rename 'delay_pcie_doorbell' to 'delay_cmd_submit'
'delay_pcie_doorbel' parameter in 'spdk_nvme_io_qpair_opts' structure
was renamed to 'delay_cmd_submit' to make it suitable for every
transport. Old name is also kept for backward compatibility.

Signed-off-by: Evgeniy Kochetov <evgeniik@mellanox.com>
Signed-off-by: Sasha Kotchubievsky <sashakot@mellanox.com>
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I09ef8028133c4a3d4a5bbc5329ced1f065bcaa46
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/475305
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>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
2019-12-10 17:32:10 +00:00
Seth Howell
5170ac8d83 test/blockdev: move configuration before tests.
This is a significant simplification to the whole test script. Note, it
does force us to run the hello_bdev example against a gpt bdev instead
of an nvme bdev directly, but I see this as an acceptable tradeoff since
it seems like the purpose of running the hello_bdev example is mostly to
ensure that the example itself is not broken.

Change-Id: I8c2a1ede8e7ece2a82222593a7ae46704cc658af
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476935
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
b9ca149ee7 test/nvmf: convert filesystem to use run_test
This test specifically gets a little more complicated looking when we do
the test this way, but the goal is to enable deterministic reporting of
test completions on a per build basis.

Change-Id: Ica36986f821337c22654fb4cc25f9a72a3d73517
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476825
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
67645f4dea test/nvmf/filesystem: explicitly use nvme0n1
This test has always relied on /dev/nvme0n1 being the location of our
test drive which means that it is the only one in the system at the time
the test is run.

So instead of trying to loop multiple non-existant nvme drives, just
call it out directly.

Also, add a todo to remove the dependence on NVMe0n1 since we could run
this test on a machine where the main filesystem is on an NVMe drive
which would break this test.

Change-Id: Ibb2448d9367a5c80d85a1c91b0c6e44e58237751
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476952
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
d29169bd77 test/iscsi: convert data_digests tests to run_test
We had very granular timing reporting on this test, but we really only
had two test cases running for the whole thing.

Change-Id: I11650e124e24e93b1f217c6b8efd2fb006284515
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476824
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
c6767000a4 test/nvmf: convert target_disconnect to run_tests
Change-Id: I8a632fb9c782b50246123f6c2d1cfd2e3f333471
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476823
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
f470cf3d0c nvmf/shutdown.sh: convert to run_test framework
This cleanns up the logs for this test and makes it easier to follow.

Change-Id: Ifc6fcea8672d4d186e3e622eddfc200fdfd076ce
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476822
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
968d4dfd65 test/vmd: convert to run_test framework
This makes the script much more readable and helps delineate test cases
more clearly.

Change-Id: I729ca0380a7173151d074ebd73ed9766c5008480
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476821
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2019-12-10 17:12:03 +00:00
Seth Howell
a0304f54f9 test/nvme: convert tests into run_test
This unifies the script and makes it easy to see what exactly we are
testing.

Change-Id: Icdbc08077ea4397bfde50493c718f99c3c499062
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476809
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
785c3f0fe4 test/ioat: clean up timing calls.
Change-Id: Ife273ac10968ad3fc19f8a46f69e50fa755ba815
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476807
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
3c87c8973c test/rocksdb: convert test cases to run_test
cleanup.

Change-Id: I8e358a1cb327215d0b3b932bd9b1d28d495501a4
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476806
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
fe4d1dce58 test/env: convert program calls to run_test
allows us to remove a lot of extra timing_* calls.

The next several commits are aimed at converting the majority of the
tests to this single framework of using the run_test suite or run_test
case function calls. Hopefully it makes it much easier to understand the
logs over time.

Change-Id: I6d46982108be469a05f70c24b373e60726aa061e
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476805
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
36ac0c4340 test/lvol: use the run_test framework directly
Change-Id: Iabec093a7cd632cf35fb163cfcc0048ed66fc856
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477077
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: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
1fafd71bc3 test: add timing calls to run_test
And clean out the corresponding calls to timing_enter and timing_exit
from the various test scripts.

Change-Id: I0759417b5a529e4c3649ce04cca1799c089da278
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476804
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
2019-12-10 17:12:03 +00:00
Seth Howell
fd17f1aabc test: make run_test more robust
make sure we have enough arguments.

Change-Id: I76ce35635ef14289061323ee401d93d8d081888c
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476801
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
b8042c3850 test: use human readable names in test logs.
Change-Id: I163a7e08bfeff3da2ee627d9fad9f2f1a249c593
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476820
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
2019-12-10 17:12:03 +00:00
Seth Howell
3710048472 test: add a test_name param to run_test
This will allow us to use timing_enter and timing_exit directly
inside the run_test function. That function already lends itself well to
nesting the way we do our timing.

This patch series is aimed at combining the timing_*, run_test, and
report_test_completions calls all into a single place. This will greatly
reduce the number of lines of code in our bash scripts devoted to
tracking timing, formatting, and test completion. It will also enable us
to expand on the reporting of test completions. Further down the line,
this will also allow us to unify test case documentation.

Change-Id: I8e1f4bcea86b2c3b88cc6e42339c57dfce4d58f2
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476799
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
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>
2019-12-10 17:12:03 +00:00
Karol Latecki
e6ee199098 doc: add 19.10 NVMe-oF TCP performance report
Change-Id: I61e5c116921a386d34dd63acf8b88205b442221b
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477006
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Seth Howell <seth.howell@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-12-09 17:51:51 +00:00
Karol Latecki
cbcff1688a doc: break down performance reports into sections
Also rename some of the old NVMe-oF performance reports to
contain "RDMA" in name for easier identification.

Change-Id: I048533dcef12ab3d570fe423adab7f212969f4dd
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476545
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-12-09 17:51:51 +00:00
Karol Latecki
cb507ab820 doc: add 19.10 vhost performance report
Change-Id: I7b5dc4e504ab66bbc7062a799cefb4a13e4758dc
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476544
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Seth Howell <seth.howell@intel.com>
2019-12-09 17:51:51 +00:00
Tomasz Kulasek
716f5e77d0 test/nvme-cli: check if cuse device is destroyed after detach
Change-Id: If1fdafd282c898ccd4e7427d575507c1df4bd778
Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473294
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: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
2019-12-09 17:40:35 +00:00
Tomasz Kulasek
46316bb5db lib/nvme: prevent creating existing cuse device
This patch attempts to solve naming conflict between
CUSE devices created by different SPDK instances.

Each NVMe device is enumerated by SPDK process from 0
up to 127. When process attempts to start cuse device
tries to set exclusive lock on temporary file
"/tmp/spdk_nvme_cuse_lock_<index>" and keep it until
device will be stopped.
If setting lock fails, index is incremented.

It prevents to use the same controller index from
few SPDK instances.

Change-Id: If744ac23f813bd992efb80ae2b61a1acefb5054c
Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/474829
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2019-12-09 17:40:35 +00:00
Tomasz Kulasek
b7b45bc7bc lib/nvme: remove device name parameter from nvme cuse
This patch removes posibility to set cuse device path. Instead
"/dev/spdk/nvme*" path is used.


Change-Id: I7c3087772a3661eebe03fce21356c35cc8204b49
Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/474598
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
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: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
2019-12-09 17:40:35 +00:00
GangCao
5d106e75e0 net: add the check for the add and delete IP operation
To properly return the error case when adding an existing
IP or delete a not existed IP. Proper test case is also
updated.

This is also to fix below issue:
https://github.com/spdk/spdk/issues/992

Change-Id: Ia4d3af8cc86d9bdb66b18a165510cd08f9bfa555
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476543
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-12-09 17:39:50 +00:00
Karol Latecki
a34d7d7b2f test/nvme: wait after reloading nvme module
Make sure that all setting files under /sys/block/nvme*
have time to get created properly.

Change-Id: I9e3bf973fc09d2f5c38be3b29aa1fea4137c0402
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476557
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.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>
2019-12-09 17:39:22 +00:00
Karol Latecki
cfb6aec8ab test/nvme: fix typo in performance script
It's "iostats", not "io_stats".
Remove extraneous "$" from variable substitution.

Change-Id: I287797a8ab44032ce5c32ee48efdec52bba59f53
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476556
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-12-09 17:39:22 +00:00
GangCao
841b7a786e test/vhost/fuzz: remove the unncessary killproces
Change-Id: Ida9b2d6aade8ab12004fccd8830452c93eca30eb
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476702
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Liang Yan <liang.z.yan@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Seth Howell <seth.howell@intel.com>
2019-12-09 17:38:46 +00:00
Jacek Kalwas
baf588a58b nvmf/rdma: remove unsed define
NVMF_REQ_MAX_BUFFERS is defined in nvmf_internal.h and not used in rdma
source code.

Signed-off-by: Jacek Kalwas <jacek.kalwas@intel.com>
Change-Id: I50f24c8fc4ea773378418f7803361d8592f961ae
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/475529
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Alexey Marchuk <alexeymar@mellanox.com>
2019-12-09 14:03:39 +00:00
Jacek Kalwas
f206551388 nvmf: fix status override in case parse_sgl fails
It is valuable to have more detail status instead
SPDK_NVME_SC_INTERNAL_DEVICE_ERROR.

Signed-off-by: Jacek Kalwas <jacek.kalwas@intel.com>
Change-Id: Ifd003b490a7ae9af017645c97636ceaf2f93d4b0
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476634
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
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>
2019-12-09 14:02:37 +00:00
GangCao
b8d0691ff3 nvmf: correct the return code when parsing the subsystem
If succeeded, return 0. Otherwise, return -1

Change-Id: I7c966e09a5efd13a3ddb72ffaaeee4a9d8ca31b0
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476983
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Alexey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Seth Howell <seth.howell@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
2019-12-09 14:01:39 +00:00
Changpeng Liu
f4e3f59e0e nvme: fix potential memory leak when there is controller scan failure
The nvme_transport_ctrlr_scan() may return failure while there are
multiple controllers, so the probe context's init_ctrlrs list may
not null for this case, so when free the probe context, let's ensure
there is no controller in the init_ctrlrs list.  Also added a UT to
cover this case.

Fix issue .

Change-Id: I4d9a10ad73cf00bbe159edd1f5b919797333feb6
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476969
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
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>
2019-12-09 13:58:01 +00:00