6218 Commits

Author SHA1 Message Date
Jim Harris
cd6b3a892d bdev/raid: remove raid_bdev_io_terminate
This has been reduced to just a call to
spdk_bdev_io_complete(), so it is no longer needed.

While here, call spdk_bdev_io_complete() directly
in the strip-crossing error case as well.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia949e3fd55b52e5c873bd591e541ecf6c0c086b1

Reviewed-on: https://review.gerrithub.io/423327
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2018-08-31 19:18:02 +00:00
Jim Harris
29f0c4c76d bdev/raid: remove split counting
Now that I/O never span a strip boundary, we don't
need to count splits.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Iea5df5ca055fe4b338fb5cb0828a45502c239458

Reviewed-on: https://review.gerrithub.io/423325
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2018-08-31 19:18:02 +00:00
Seth Howell
bfcb7f64c1 vhost: Stop memory leak in external foreach loop
This is in response to a Scan-build issue reported on Clang 6.

Change-Id: I2edc853145762998db818cbbe0e9ca0d9b8c123d
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/424139
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.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>
2018-08-31 19:09:07 +00:00
Seth Howell
17e9d38c5b test/unit: Fix Scan-build errors in scsi tests
I had to make small modifications to one test in the dev_ut.c file to
prevent a memory leak error. However, changing the index in this test
from -1 to 0 has no effect on the coverage.

Change-Id: Ia8a475aaa41118f072a5d9006bab9d9c84b4a10a
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/424126
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
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>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-08-31 19:09:07 +00:00
Seth Howell
c0297f9c71 test/unit: Fix scan-build errors in thread and util tests
thread_ut.c: ensure that the channel context is non-null
util: get rid of a couple of uninitialized value errors.

Change-Id: Idbaad4efea14acd7ae8f98b60aa514e45213a134
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/424124
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-08-31 19:09:07 +00:00
Seth Howell
95d97eb965 test/unit: fix scanbuild errors in the nvme tests.
Change-Id: I20366711f5543ac506e3a5c98c21d256b26143f0
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/424121
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-08-31 19:09:07 +00:00
Seth Howell
84009795c5 test/unit: fix Scan-build errors in lvol tests.
Fixes two null pointer dereference errors.

Change-Id: I7d685fa54af60e6efd70f3686d4dc4dba338f555
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/424108
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-08-31 19:09:07 +00:00
Jim Harris
2e6aac525c bdev/raid: use split_on_optimal_io_boundary
Set the bdev->optimal_io_boundary to the strip size, and
set split_on_optimal_io_boundary = true.  This will ensure
that all I/O submitted to the raid module do not cross
a strip boundary, meaning it does not need to be split
across multiple member disks.

This is a step towards removing the iovcnt == 1
limitation.  Further improvements and simplifications
will be made in future patches before removing this
restriction.

Unit tests need to be adjusted here to not span
boundaries either.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I08943805def673288f552a1b7662a4fbe16f25eb

Reviewed-on: https://review.gerrithub.io/423323
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
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>
2018-08-31 17:53:42 +00:00
Jim Harris
4f860d7e40 bdev: only apply split_on_optimal_io_boundary to R/W
Splitting a 1TB unmap into individual 64KB unmap commands
(for a RAID volume with 64KB strip size) would be awful -
the RAID module can be much smarter about this.

So back out the changes for splitting I/O without payload.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I24fe6d911f4e3c9db4b2cb5d66c7236a5596e0d9

Reviewed-on: https://review.gerrithub.io/424103
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2018-08-31 17:53:42 +00:00
zkhatami88
eb6a2cb8f5 nvme:replacing spdk_dma_malloc with spdk_z/malloc
Change-Id: I755d36495e7d8e55c26c4103a9520102223798ef
Signed-off-by: zkhatami88 <z.khatami88@gmail.com>
Reviewed-on: https://review.gerrithub.io/409960
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-08-31 17:42:29 +00:00
wuzhouhui
4c8a350d02 jsonrpc: fix error path of spdk_jsonrpc_parse_request()
Change-Id: Ia1bb2edefe31dc179986b9d4f8ea0d9002af0a40
Signed-off-by: wuzhouhui <wuzhouhui@kingsoft.com>
Reviewed-on: https://review.gerrithub.io/424008
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2018-08-31 17:41:58 +00:00
Ben Walker
28a61c2130 nvmf/rdma: Simplify event acknowledgement in disconnect path
This no longer requires special handling - the event can be
acknowledged like all of the others.

Change-Id: Ib30cf35ec7aff45734ca6fe729e15d8fe41e3838
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/423935
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Seth Howell <seth.howell5141@gmail.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-08-31 14:50:05 +00:00
Ben Walker
81d51948ad nvmf/rdma: Move spdk_nvmf_process_cm_event by event handlers
Keep the code together. This is only code movement.

Change-Id: Ie52f1ab09e197192025f2b664df410ba6e1f06aa
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/423934
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Seth Howell <seth.howell5141@gmail.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-08-31 14:50:05 +00:00
Ben Walker
e6b2caee51 nvmf/rdma: Immediately release resources for requests when killing qpair
Previously, this would release resources for requests if there
was an RDMA error on the qpair. Expand this case to include
scenarios where the qpair is in the process of intentionally
shutting down.

Change-Id: Ib018f190389ee2df20eba3dddcc7dcffdbb4909d
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/423745
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Seth Howell <seth.howell5141@gmail.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-08-31 14:50:05 +00:00
Ben Walker
764346697a nvmf/rdma: Query qp state prior to acknowledging disconnect event
This guarantees that the qpair memory still exists.

Change-Id: I759197b90513f30488aa46bd26535c663e64dae6
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/423744
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Seth Howell <seth.howell5141@gmail.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-08-31 14:50:05 +00:00
Ben Walker
9f6d509bf9 nvmf/rdma: Don't abort commands with pending RDMA ops until quiesced
Don't abort commands in states indicating an RDMA operation
is outstanding until an event indicates that all of the
work items have completed.

Change-Id: Ie2b83604bee142e383ffbcae088f4da0fd0fa658
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/423413
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.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>
Reviewed-by: Seth Howell <seth.howell5141@gmail.com>
2018-08-31 14:50:05 +00:00
Ben Walker
4eb7c48d75 thread: Align members of structs
Change-Id: Iaf339b4a09d3e5d84f64ddda47eaa1dd09d8e268
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/423962
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2018-08-31 05:12:01 +00:00
Jim Harris
051fd012fc ut/blob: fully initialize payload_pattern in split_iov test
Valgrind found this on my system.  The unit test itself
was passing correctly (we wrote uninitialized data and then
successfully read back the same uninitialized data) - but
Valgrind doesn't like code that makes any decisions based
on reading uninitialized memory.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I762f5dd0cb368328bd5eba9cecf7db404b457fad

Reviewed-on: https://review.gerrithub.io/424109
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2018-08-31 04:35:32 +00:00
Seth Howell
93edd7b023 test/unit: fix scan-build errors in iscsi tests
One was an API issue, another was a dereference of a null pointer that
resulted from a possible call to a stubbed function in the unittest
common code, and the last was an erroneous use-after-free due to
scan-build apparently not understanding TAILQ_REMOVE.

Change-Id: I9111817e9e990ff6e388351c08f34c2e945df62d
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/423959
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>
2018-08-31 00:01:09 +00:00
Seth Howell
14ccaed6de test/unit: fix scan-build errors in the event tests
Change-Id: I459c8cdea080af24374c2ad1fa42e291c5021840
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/423950
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>
2018-08-31 00:01:09 +00:00
Seth Howell
438b71d17e test/unit: fix scan-build errors in blobfs tests
fix 3 null pointer dereferences.

Change-Id: I0f11b8e47b05d0e965cea75d7b064cc2c11645e3
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/423942
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>
2018-08-31 00:01:09 +00:00
Seth Howell
11ccf3be2c test/unit: Fix scan-build errors in bdev unittests
Fixes a use after free error in bdev/bdev_ut
Fixes an uninitialized argument value error in mt/bdev/bdev_ut
Adds extra checks in the spdk_bdev_part_construct code
Fixes a dereference of null pointer error in bdev/vbdev_lvol_ut
Change-Id: I1eaa6ebfec0a36a0d910504c70b16ccc651de22a
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/423923
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>
2018-08-31 00:01:09 +00:00
Seth Howell
c8341f12bd bdev/part: use internal buffers for part and product name.
This removes assumptions about hte lifecycle of these variables relative
to the parent bdev and allows us to avoid ambiguity about who
is responsible for freeing these buffers.

Change-Id: Ia996653562d532fa1501faf21d3fdff85033ab33
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/424105
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>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-08-30 22:08:56 +00:00
Jim Harris
9f2035c7cc test/bdev: comment out bdevperf write_zeroes test
This is causing a fair number of failures in both the
Chandler and Jenkins test pools.  This was added fairly
recently, so comment it out for now while it is debugged
further.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I5d227252e27265c2b25ad7e76dd54d6c0501b8c4

Reviewed-on: https://review.gerrithub.io/424132
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-08-30 21:43:33 +00:00
Shuhei Matsumoto
1ac058ca08 bdev/raid: Set RAID bdev's name at its creation to prevent unnamed device
Next patch will keep RAID bdev without removal when adding any base
bdev to the RAID bdev fails for some reason for construct_raid_bdev
RPC.

The half-baked RAID bdev will become unnamed and this should be avoided.

Change-Id: I3ad0d58a04a7638b9b286babe0e56e057e7f5d0d
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/423621
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.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: Kunal Sablok <kunal.sablok@intel.com>
2018-08-30 16:13:35 +00:00
Shuhei Matsumoto
12881ad9bc bdev/raid: Remove duplicated and not enough state check in RPC
When check_and_remove_raid_bdev(raid_cfg) is called, raid_bdev
must be in configuring state. However, offline state will be
probable as well. Besides similar and more comprehensive state
check is done in raid_bdev_cleanup(). Hence remove this check.

Change-Id: I8526da51828f18a8393afbd3d81c0a52f2d33261
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/423620
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Kunal Sablok <kunal.sablok@intel.com>
2018-08-30 16:13:35 +00:00
Shuhei Matsumoto
a295285bf5 bdev/raid: Use descriptive error message for failure in JSON-RPC
Change-Id: I238c2ba43c77f47851aa9d28818d42e8d865f2d4
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/423619
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Kunal Sablok <kunal.sablok@intel.com>
2018-08-30 16:13:35 +00:00
Shuhei Matsumoto
0a46f5c2aa bdev/raid: Extract check claim operation from add base bdev operation
When a raid bdev is constructed by JSON-RPC construct_raid_bdev,
information about config and slot can be passed to
raid_bdev_add_base_device() and raid_bdev_can_claim() doesn't have
to be called.

Hence extract raid_bdev_can_claim_bdev() from raid_bdev_add_base_device()
and put it to raid_bdev_examine().

Change-Id: I92e02bf3661cb97b691246f32198ba946810d96c
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/423618
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Kunal Sablok <kunal.sablok@intel.com>
2018-08-30 16:13:35 +00:00
Shuhei Matsumoto
3333ceb6af bdev/raid: Leave checking bdev duplication with spdk_bdev_register()
When the name of raid bdev is already used by any existing bdev,
spdk_bdev_register() will fail. Hence checking if bdev is duplicated
can be left with spdk_bdev_register().

Change-Id: I22f0c75b0e87c5bd71ece383d146ec61cfaf2627
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/423617
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Kunal Sablok <kunal.sablok@intel.com>
2018-08-30 16:13:35 +00:00
Shuhei Matsumoto
432f8d9160 bdev/raid: Fix uncorrect cleanup call in raid_bdev_alloc_base_bdev_resource()
When raid_bdev_alloc_base_bdev_resource() is called through raid_bdev_examine(),
if raid_bdev_alloc_base_bdev_resource() fails, raid_bdev_cleanup() is called
without freeing base bdev resource.

There is no way to free base bdev resource in this case.

Hence remove cleanup call.

Change-Id: I93494502bfe653293825e11872de5883702e495a
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/423616
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Kunal Sablok <kunal.sablok@intel.com>
2018-08-30 16:13:35 +00:00
Shuhei Matsumoto
2c608684ff bdev/raid: Use descriptive error code for failure of config operation
Change-Id: I2c34e666fbdcfe5640a87398601f0b819c78a36f
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/423615
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Kunal Sablok <kunal.sablok@intel.com>
2018-08-30 16:13:35 +00:00
Shuhei Matsumoto
636b9d597c bdev/raid: Simplify variable name in struct raid_base_bdev_config
raid_base_bdev_config is suffciently descriptive as name and so
"name" can be used as the name of the base bdev.

Other structs have used "name" as the name of bdev.

Change-Id: I090ef541a84fd14d9ec3bfebbdc96ae649709551
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/423614
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.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: Kunal Sablok <kunal.sablok@intel.com>
2018-08-30 16:13:35 +00:00
Tomasz Zawadzki
897eed3179 test/ioat: remove unnecessary config entires
Since IOAT is disabled by default, that option can be removed from
any config file used in tests.

Change-Id: I681bd1eb6a8c81328ba580be75b418994ccd85e3
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/423357
Reviewed-by: GangCao <gang.cao@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2018-08-30 14:45:48 +00:00
Jim Harris
b5d1b4f445 test: don't print backtrace when errexit is off
Some test scripts may temporarily disable errexit -
for example, trying a command several times until
it succeeds.  In these cases we don't want to
print a backtrace every time that command fails.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I481f50b046eb13e23f7048672281ec7d62a573a2

Reviewed-on: https://review.gerrithub.io/423921
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2018-08-30 04:47:39 +00:00
Jim Harris
f4d78f1886 blob: remove unused internal functions
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I67f1bf569a8a21f77729a30e933efd04ee05b86e

Reviewed-on: https://review.gerrithub.io/423957
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>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-08-30 03:41:01 +00:00
Jim Harris
a2794082ca doc/vagrant: add cd to examples/bdev/hello_world
This ensures the bdev.conf file is found.

Reported-by: Amir More (in GitHub issue #420)
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I309d8f211e0a3e0e4530c5d3c225a52b6ea68c2c

Reviewed-on: https://review.gerrithub.io/423931
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
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>
2018-08-30 03:41:01 +00:00
Jim Harris
87d2cd9b69 app: deprecate ReactorMask from ini config file
This is not used anywhere, and is not something we can
set over RPC.  So make this a command-line only option
(as it should be).

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I2df862cdce2f17992d2324312d5c0e98b38a8acd
Reviewed-on: https://review.gerrithub.io/423930
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2018-08-30 03:41:01 +00:00
Jim Harris
42daa14f64 iscsi: accumulate data_transferred properly for large writes
Without this fix, large writes would report underflow.  Linux
initiator seems to ignore the underflow condition, but ESXi
initiator would properly detect it.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Icab28531c44f61a8e5529bcee989d0c2bf41d043

Reviewed-on: https://review.gerrithub.io/423906
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2018-08-30 00:22:42 +00:00
Seth Howell
b6770c73cc pkgdep: minor fixes
CHange double brackets to single brackets for compatibility with
freebsd, also added -y to a pkgdep call to avoid the script hanging when
called programmatically.

Change-Id: I9a851005efae77b9146f14cac8079e2d448143d4
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/423936
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2018-08-29 23:49:12 +00:00
Shuhei Matsumoto
e0fc53f750 doc/jsonrpc: Add Optional column to the param table of set_iscsi_options RPC
Add optional column to the table of parameter of set_iscsi_options RPC.

Change-Id: I7e5d0af2ebe25c84340a12a49b6dc24d19b911e6
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/421466
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-08-29 21:40:31 +00:00
Shuhei Matsumoto
1366b62f70 iscsi: Add CHAP authentication group configuration to JSON config dump
Adding CHAP authentication group configuration to JSON config dump.
This addition is done unconditionally because the path to the CHAP secret file
have to be specified explicitly now.

Change-Id: Ibd4085222a6767dae3a74023f1d6bc270ef4abac
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/421469
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-08-29 21:40:31 +00:00
Shuhei Matsumoto
3a08001dd4 iscsi&scripts/rpc: Add get_iscsi_auth_groups RPC to get current configuration
Add an new RPC to get current authentication group configuration.
This patch is utilized in the next patch to support JSON config dump
for authentication group configuration.

Change-Id: I34be9e196f8e7a484bcd316da54f05d0f6ee0300
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/421468
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-08-29 21:40:31 +00:00
Shuhei Matsumoto
ad323b8064 iscsi&scrpts/rpc: Require to specify CHAP secret file explicitly to load it
Previous patches enabled users to configure CHAP secrets dynamically
by RPCs. Subsequent patches will enable users to load CHAP secrets
from JSON config file.

Loading CHAP secret file is done by default and this will conflict to
JSON config file.

Hence the path to CHAP secret file is required to specify in the config
file or JSON RPC set_iscsi_options explicitly after this patch.

Users who have used CHAP secret file are expected to specify it explicitly
and this will be no harm for them.

Besides, CHAP secret file is not oly for discovery sessions but also for
login to iSCSI targets. However there were wrong description to make user
misunderstand. Hence remove these wrong description in this patch too.

Change-Id: Ic4093cabc0c14b87e26baef4bba6b0d292e40c06
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/421467
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-08-29 21:40:31 +00:00
Seth Howell
aca4ba4dd1 test/vm_setup: fix yum-deprectad for fc28 with vpp
Change-Id: I6608725c031daff6f6d59ced0c327f19b6eea979
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/423755
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-08-29 18:46:02 +00:00
Jim Harris
183f37e8ad bdev: do not reuse bdev_io when splitting write_zeroes
Now that we split on I/O boundaries, that code needs to
be able to use the bdev_io split* members to track
what is left to submit.  This means that the write_zeroes
code cannot submit the parent bdev_io as the child bdev_io,
since the I/O boundary code will overwrite the write_zeroes
split accounting.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I9316b59267508f60799766fc4f1ea05a4b3e5d9e

Reviewed-on: https://review.gerrithub.io/423404
Reviewed-by: Seth Howell <seth.howell5141@gmail.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-08-29 16:42:53 +00:00
Jim Harris
d38b3d28a8 bdev/raid: remove raid_bdev_send_passthru
This function was meant to handle case when a RAID
volume only had one member disk.  This is not a
common case, so just go through the
raid_bdev_submit_children() path in this case.  This
will simplify some upcoming patches to use
split_on_optimal_io_boundary to enable vectored
I/O.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ic7c7f7987d54ffd836deabb98530f7c339094809

Reviewed-on: https://review.gerrithub.io/423322
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
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>
2018-08-29 16:42:53 +00:00
Jim Harris
5443e0aed3 bdev: count down for qos tracking
This will simplify some future patches which will
account for missed timeslice timers by allowing
additional IO/BW in the following timeslice.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I9dd46a768c98ce267c733a9f9719a2d3d2c3c915

Reviewed-on: https://review.gerrithub.io/423579
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: GangCao <gang.cao@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>
2018-08-29 16:42:39 +00:00
Jim Harris
5595a7cad7 test/qos: adjust qos test parameters
Use 512 byte reads (instead of 8KB) to ensure we can
hit the expected IOPs rate.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I204dda9bf5cc168a3f080e6716cacc71fd8952cd

Reviewed-on: https://review.gerrithub.io/423583
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: GangCao <gang.cao@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2018-08-29 16:42:39 +00:00
Jim Harris
b30a59d624 test/qos: use get_bdevs_iostat RPC to get IO count
This eliminates some messy fio output parsing, and has the
side benefit of testing the bdev_iostat RPC code.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia538cd81730704177b137c0c203d557fd640c955

Reviewed-on: https://review.gerrithub.io/423589
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: GangCao <gang.cao@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Chen, LongX O <longx.o.chen@intel.com>
2018-08-29 16:42:39 +00:00
Jim Harris
8a703ca6ef doc: add comment explaining order of INPUT file list
While here, reorder some of the files alphabetically.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I78d7e31e55d94ed4473fa7d3c96fa3fb8481444a

Reviewed-on: https://review.gerrithub.io/423752
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-08-29 16:39:19 +00:00