Commit Graph

10494 Commits

Author SHA1 Message Date
Artur Paszkiewicz
d3a04643bf module/raid: move error code check out of raid_bdev_queue_io_wait()
Don't pass the error code to the function and instead move the -ENOMEM
check to the callers where it makes more sense.

Signed-off-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Change-Id: Ib2fa92315dc2973d4023fd9509950589de946614
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473452
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>
2019-11-14 16:25:55 +00:00
Artur Paszkiewicz
cb8dbf17dc module/raid: add raid_bdev pointer to struct raid_io
This will make it easier to get to the raid_bdev when we only have the
raid_io.

Signed-off-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Change-Id: I91df14f788a51ada10b0f8356de2162c1f34520c
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/471085
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>
2019-11-14 16:25:55 +00:00
Kozlowski Mateusz
5b88597ac6 bdev: Initialize md_buf for zero copy start request
In emulated zero copy mode on devices with separate metadata support
the start zero copy request could assert after read completion,
during _bdev_io_unset_bounce_buf; the original iovs would be NULL, but
original md_buf would have garbage data.

Signed-off-by: Kozlowski Mateusz <mateusz.kozlowski@intel.com>
Change-Id: Ia52b012c1b7462ebe490c29f5493168775646848
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473455
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: Ben Walker <benjamin.walker@intel.com>
2019-11-14 15:41:40 +00:00
Vitaliy Mysak
fa540f18d8 test/ocf: split tests for io types and get_stats
In bdevperf-iotypes we used to call "bdev_ocf_get_stats" RPC
which is not a part of IO types testing.
As a result, we started to get failures caused by that rpc call which
were not related to ocf bdev IO handling (issue #1025).

This patch isolates the RPC call to separate test.

Change-Id: I3a72ef52452868d49a6ec1ae4d3f9643496c2a28
Signed-off-by: Vitaliy Mysak <vitaliy.mysak@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/474052
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: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2019-11-14 15:40:34 +00:00
Alexey Marchuk
2ad7573393 build/rpc: Add git sha1 to SPDK version macro
Retrieve git sha1 during SPKD build, add sha1 to SPDK_VERSION_STRING
macro and to spdk_get_version RPC method
This might be useful for issues reporting/investigation

Change-Id: I2bfed752839036bf914b546066ed4d23ef4c99b8
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Signed-off-by: Sasha Kotchubievsky <sashakot@mellanox.com>
Signed-off-by: Evgeniy Kochetov <evgeniik@mellanox.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/474004
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2019-11-14 03:46:17 +00:00
Alexey Marchuk
e4aee795c0 configure/ISA-L: Move log file to submodule directory
Running configure script by multiple users on the same machine
leads to access to the same ISA-L log file which is created
with read access for user's group. As the result only the
user who created this file can modify it and other users will
receive "permission denied" error. Move this log file to the
local ISA-L directory to prevent access by multiple users

Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Signed-off-by: Sasha Kotchubievsky <sashakot@mellanox.com>
Signed-off-by: Evgeniy Kochetov <evgeniik@mellanox.com>
Change-Id: I1862738bc0ba242c98816a22dc39bd11ef5d02dd
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473855
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2019-11-14 03:46:17 +00:00
Alexey Marchuk
3cf9ff7495 rdma/conf: Add parsing of no_srq parameter via conf file
Change-Id: I67e8e1f4ce1f3be9a4bb0d470eac4927a6c3fbbd
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Signed-off-by: Sasha Kotchubievsky <sashakot@mellanox.com>
Signed-off-by: Evgeniy Kochetov <evgeniik@mellanox.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473846
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>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2019-11-14 03:46:17 +00:00
Liu Xiaodong
6a382b9860 vhost: delete vhost_put_poll_group declaration
Function vhost_put_poll_group is already removed. There
is no implementation for vhost_put_poll_group, so delete
its declaration.

Change-Id: Ib4766fca93329c4f4faab48857d1ac2eaa268300
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/472513
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2019-11-14 03:44:32 +00:00
Karol Latecki
03c71d3741 test/vhost: update raid bdev entries in config file
Edited config file is loaded in test via rpc.py
load_config. Because of that, each parameter is
sent "as-is" from config file, as opposed to using
rpc.py bdev_create_raid which would cause argparse to
convert all params to desired type.

Commit 445e667 changed raid_level to enum while this
config file still used int which caused nightly tests
to fail.

Change-Id: I729c7263261c73e255ab2e6071f76219a351cb19
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473565
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: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Seth Howell <seth.howell@intel.com>
2019-11-14 03:43:47 +00:00
Yao Liu
ec2ba7aade bdev/uring: add an argument to specify block size
The block size of non-disk files can't be automatically detected,
so add an argument to specify it.

Signed-off-by: Yao Liu <yotta.liu@ucloud.cn>
Change-Id: Ia11ab177b78c66ede4a09a997bba28827d83ec25
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473607
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-11-13 15:06:30 +00:00
Jim Harris
9d9a90faa0 nvme/perf: print periodic stats to stdout every 1s
This is sort of like the periodic stats that fio
prints every second or so.  We can add more to this
later, but this by itself is a nice improvement I
think.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I4ac588c8301bae9924dd6541cbc386ffe8d06b5d

Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473339
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>
2019-11-13 15:03:33 +00:00
Maciej Wawryk
623203b373 test: Shellcheck - include check SC2142
Remove from shellcheck excluded SC2142: Aliases can't use
positional parameters. Use a function.
This warning is not actually present in any of spdk/spdk scripts.

Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: I0af375c062724dc8fcee5a6cd4b8ac99bda260ac
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473292
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2019-11-13 15:03:11 +00:00
Maciej Wawryk
17b26b7690 test: Shellcheck - correct rule: Word is on the form...
Correct shellcheck rule SC2140: Word is on the form "A"B"C"
(B indicated). Did you mean "ABC" or "A\"B\"C"?

Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: I1655b2e76be7a5e3d615c5b34c5af9e4012ac49f
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473291
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2019-11-13 15:03:11 +00:00
Pawel Kaminski
4d0c5091a9 test: Shellcheck - add rule SC2154 to exclude list
Corrected code in some files.
Most of cases cannot be corrected as we use global variables
defined in other files.

SC2154: var is referenced but not assigned.

Signed-off-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Change-Id: Id333587f81b0bc2456b0c97bac5343a8e5a22774
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/472342
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>
2019-11-12 18:14:59 +00:00
Pawel Kaminski
074df1d896 test: Shellcheck - apply rule SC2155
Declare and assign separately to avoid masking return values.

Signed-off-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Change-Id: Ib90598e4268911a3056e8a0baa7d541edaa29b91
Signed-off-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/472287
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>
2019-11-12 18:14:59 +00:00
Pawel Kaminski
c9c576912d test: Shellcheck - move rule SC2162 to exception list
read without -r will mangle backslashes

Change-Id: I9eb59dce40c0bfb59d865319992b419d512dc41b
Signed-off-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/471885
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>
2019-11-12 18:14:59 +00:00
Shuhei Matsumoto
933d990e23 ut/iscsi: Use TAILQ_HEAD_INITIALIZER() for g_write_pdu_list
This clarifies that g_write_pdu_list is global.

Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I4292a2fb455f3a47d79ae6ebb5dc6046f0f7dd52
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/471951
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-11-12 05:31:59 +00:00
Shuhei Matsumoto
22f4e0a12f lib/iscsi: Use cached pdu->data_segment_len throughout in iscsi_pdu_hdr_op_scsi()
Data segment length of the PDU is already cached in pdu->data_segment_len.
Hence additional caching to the local variable data_len is not necessary.

Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I1e596999640229b1b0fa85cbdb342b1636af5076
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/471879
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: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2019-11-12 05:31:59 +00:00
Shuhei Matsumoto
bb61fabf26 lib/iscsi: Clarify only logout with reason "close the session" is acceptable on discovery session
Based on the following description in Chapter 4.3 "iSCSI Session Types in
iSCSI specification

      b) Discovery session - a session only opened for target discovery.
         The target MUST ONLY accept Text Requests with the SendTargets
         key and a Logout Request with reason "close the session".  All
         other requests MUST be rejected.

update the comment slightly, add macro constants for iSCSI logout
reason, and change the ordering of checks to be session type and then
logout reason.

Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Ifc2ecc5b6dde546700662d3cda59d8cc465fd83a
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/472672
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
2019-11-12 05:31:59 +00:00
Shuhei Matsumoto
9c3c488e0c lib/iscsi: Fix NULL pointer access for logout with close connection
conn->sess->session_type must be accessed when conn->sess is not NULL.

Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I3d41443352b65ee5ef4cc1f0d152b9e3221975c9
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/471877
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: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2019-11-12 05:31:59 +00:00
Shuhei Matsumoto
71f63bb7d9 lib/iscsi: Embed _iscsi_conn_free() into the caller and free it
_iscsi_conn_free() is simple enough to embed it into the caller
and remove it.

Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I22bcbbdab15eca647914715754c04b8ec14ad9b2
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/472901
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-11-12 05:31:59 +00:00
Tomasz Zawadzki
9a25fc12bb test: add SPDK_BUILD_PACKAGE flag
Until now the autopackage.sh execution was tied to
RUN_NIGHTLY flag. This was useful to check for
errors on all jobs/systems during a nightly test.
By their nature nightly does not run on per-patch
basis.

This allows for issues to slip in and be noticed only
by the nightly job after merge occurred.

Adding separate SPDK_BUILD_PACKAGE flag, allows to
explicitly test packaging on single job on per-patch.
RUN_NIGHTLY was kept as trigger for autopackage.sh
to still test it on all systems by default.

Change-Id: I29925fd6256b218e24c24ebcc4974c65a9bd986c
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473519
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>
2019-11-08 22:18:24 +00:00
Ziye Yang
542185b7e0 nvme/qpair: merge two if case into one.
Purpose: To remove the duplicated code.

Change-Id: Iab9989f9928698967533e45e7cffad4f09bde16a
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473376
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: Jim Harris <james.r.harris@intel.com>
2019-11-08 22:18:18 +00:00
Ziye Yang
08273e77de tcp: Fix no tcp_req issue while using async writev later.
Purpose: But if we use asynchronous writev
for pdu sending, the call_back of writev may occur
after the new data coming. So it means that the
free tcp request may not be available.
So we use the strategy to check the request status
in TCP_REQUEST_STATE_TRANSFERRING_CONTROLLER_TO_HOST.

So the strategy is checking the state_cntr of all the
reqs in TCP_REQUEST_STATE_TRANSFERRING_CONTROLLER_TO_HOST
state.
1 If the state_cntr > 0, we should queue
the new request.
2 If the statec_cntr == 0, it means that
there is no available slot for the new tcp request
, i.e., the new nvme command comming from the initiator.
If we receive this, it means that the initiator sends more
requests,and we should reject it.

Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: Ifbeb510e669082cb7b80faf2e7987075af31d176
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/472912
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>
2019-11-08 22:17:42 +00:00
Ziye Yang
e19fd311fc nvmf/tcp: Add ttransport variable in spdk_nvmf_tcp_sock_process
To avoid the allocation of ttransport in the sub functions,
and it makes the code much efficient.

Change-Id: Ie4c5a1755ddbecf10dc364ff811f74a7af5f9c3b
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473003
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
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>
2019-11-08 22:17:42 +00:00
Pawel Kaminski
e0c4db2acb rpc: Fix - Use correct deprecated alias name for bdev_nvme_set_hotplug
Change-Id: Ia9869abc73da2d074500f449a943ddb12afc5d76
Signed-off-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473587
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>
2019-11-08 16:02:36 +00:00
Jim Harris
750f2b4b3d thread: fix set-but-unused warning
In release builds, the assert() is compiled out, making
it look like the rc value is never referenced after it's
set.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I59305b0e928f2044146e30b7addc86f81e7a1d3f

Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473472
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2019-11-08 00:07:51 +00:00
Ziye Yang
e9be9df45f nvmf/tcp: Fix the potential issue of connection construction.
When we use async writev (e.g., lib io_uring), we find that
the callback of writev is executed after recving the new
data from the initiator, and this is possible.

For example, if the NVMe-oF TCP target receives the ic_req from the
initiator, and sendout the ic_resp, the state  of tqpair will change from
invalid to running until the callback is executed. And the data of ic_resp
is already sent to the initiator, and we receive the new command later. However,
we may still not get the call back function executed
(i.e, spdk_nvmf_tcp_send_icresp_complete). And it is possible
for using lib io_uring, I faced this issue when using lib uring.

And this patch can fix this issue.

Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Change-Id: I7f4332522866d475e106ac6d36a8ec715133f0dc
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/472770
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>
2019-11-07 23:08:17 +00:00
paul luse
921c378d79 module/crypto: stop touching bdev internal status field
Add a status field to the IO context struct and use it to track
whether an indiivdual crypto op has failed or not so we know
how to complete the bdev IO.

Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: If93bc9938c578184c2db9e015eee50ab758c90f0
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/471693
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>
2019-11-07 23:05:55 +00:00
Konrad Sztyber
933bf632d6 event/vmd: write_config_json support
This patch adds support for saving VMD subsystem's configuration to a
JSON formatted file.  It allows saving and then loading configuration
with VMD enabled without having to send the `enable_vmd` RPC.

Change-Id: I59b380ec1e1f25f60f45d2363724ed7ac78c365c
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473405
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>
2019-11-07 23:05:38 +00:00
Yao Liu
6f3ca0804e bdev/uring: try without O_DIRECT flag if non-disk file first open fails
Signed-off-by: Yao Liu <yotta.liu@ucloud.cn>
Change-Id: I74b5b146c23e5395d26f80ca4f950631bd4721cb
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473373
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-11-07 23:05:06 +00:00
Yao Liu
adf19fe6a3 bdev/uring: open bdev file with O_RDWR flag
Without this flag, write requests will lead to I/O errors.

Signed-off-by: Yao Liu <yotta.liu@ucloud.cn>
Change-Id: I39a453683f7b2cbcd49fb8e0b532bf09f9180a8c
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473372
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: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-11-07 23:05:06 +00:00
Maciej Wawryk
d352c44ad0 test: Shellcheck - correct rule: Consider using grep -c
Correct shellcheck rule SC2126: Consider using
grep -c instead of grep|wc.

Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: Iceeef8b4fae984a95543d8d4b9fcbf90e9c073aa
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473289
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>
2019-11-07 23:04:10 +00:00
Maciej Wawryk
e764b4ca5d test: Shellcheck - correct rule: Assigning an array to a string
Correct shellcheck rule SC2124: Assigning an array to a string!
Assign as array, or use * instead of @ to concatenate.

Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: Id74c55b7e1c41b6469e84ac388cf8a5bd8e562a6
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473276
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>
2019-11-07 23:04:10 +00:00
Maciej Wawryk
98c63be88f test: Shellcheck - include check SC2121
Remove from shellcheck excluded SC2121: To assign a variable,
use just 'var=value', no 'set ..'.
This warning is not actually present in any of spdk/spdk scripts.

Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: I7dae1d73f0e276ef9e5a0cfa78a56227264c0a4d
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473272
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>
2019-11-07 23:04:10 +00:00
Maciej Wawryk
9a4a87b573 test: Shellcheck - correct rule: $/${} is unnecessary on arithmetic variables.
Correct shellcheck rule SC2004: $/${} is unnecessary on arithmetic variables.

Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: Ibf2879360bc50cc058b0f4434a5777c53c0eeffb
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473265
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>
2019-11-07 23:04:10 +00:00
Jim Harris
08a71404a2 test: fix uses of deprecated RPCs
We tried to change all of these when updating the RPC
names, but there were some that were missed apparently.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I1b27d658bb039f201ca003b1d9005e7cfa8c45c2

Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473340
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>
2019-11-07 22:55:38 +00:00
Jim Harris
2be2b6eba5 test: add additional asan suppressions related to fio
We are already suppressing fio (not SPDK fio_plugin)
leaks in a couple of other places, which could likely
be causing the indirect leaks we are now going to
suppress here.

Fixes issue #1003.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ie5283280495e7155cda1a93d2bd3d48ffbb6cba7

Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473495
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: yidong0635 <dongx.yi@intel.com>
Reviewed-by: GangCao <gang.cao@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>
2019-11-07 22:53:21 +00:00
Jim Harris
acea7bbce4 test/rpc_client_test: remove g_rpc_server_th_done sem_t
The pthread_join() is sufficient.  The server thread really
shouldn't take 1 second to run, but when running in a VM
it's possible things are getting scheduled such that the
timer expires.  This should not be a concern of this test
though - it should just test that the rpc server/client
functionality is working as intended.  So remove the
sem_t that was timing out.

Note: I kept the other sem_t's in here for now.  Maybe
they should be removed too, but for now my main intention
was to fix this annoying intermittent failure.

While here, fix some typos and cases where we should be
printing "server" instead of "client".

I think this fixes issue #839.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I647d0b9c5faa9ccac53e6f6387757bb5a7649b5c

Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473490
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: Ben Walker <benjamin.walker@intel.com>
2019-11-07 22:53:21 +00:00
Karol Latecki
8088176127 test/vhost_perf: add option to not run gen_nvme.sh
Do not run gen_nvme.sh and do not automatically create
nvme bdev configuration.
Allows more control over Nvme/bdev configuration when
running performance tests.

Change-Id: I9ef139921efbcc5d92d3a55c05aace929c6284e7
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/464564
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: Maciej Wawryk <maciejx.wawryk@intel.com>
2019-11-07 13:32:41 +00:00
Karol Latecki
7f19b9c66d test/vhost_perf: enable host SAR CPU measurements
Add possibility to measure CPU utilization on host server
while running FIO tests.

Change-Id: Ibc76c2b1469974297350388880a036d8778e4713
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/464563
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2019-11-07 13:32:41 +00:00
jiaqizho
2271c69500 blobfs_bdev_ut : fix memory leak
This memory leak bug is very easy to reproduce.
You can follow this step to reproduce:
   1. clone master spdk , update submoudle
   2. ./configure --with-fuse && make -j24
   3. valgrind --leak-check=full --error-exitcode=2
   test/unit/lib/blobfs/blobfs_bdev.c/blobfs_bdev_ut

It show the memory when you run valgrind.
This is because the ctx in spdk_blobfs_bdev_mount have
not freed when spdk_fs_load(override in blobfs_bdev_ut)
passed.

So i added a unmount operation after mount operation
to fix this problem.

Signed-off-by: jiaqizho <jiaqi.zhou@intel.com>
Change-Id: I770f914123e353dc42d0420c1fb8b34ebdf88f6c
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473171
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-11-07 13:24:15 +00:00
dongx.yi
861010b2d3 test/unit: Add bdev_zone_ut.c to unittest.
Add unittest for bdev_zone.c.

Change-Id: I9883592048471efe7307eb65a46df000fc205f1e
Signed-off-by: yidong0635 <dongx.yi@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/471600
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
2019-11-07 13:23:18 +00:00
Pawel Kaminski
50113a0b79 test/openstack: Add possibility to reinstall devstack on required branch
Change-Id: Ia1a55c32bee9eb60c2d49729915a2b9119b73fb3
Signed-off-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/468991
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.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>
2019-11-07 13:22:50 +00:00
Pawel Kaminski
7b267b4420 test/openstack: Adjust installation file to newest changes in devstack.
Use python3 when installing devstack.
Increase timeouts in order to tests pass in our CI environment.

Change-Id: I6b88e39f9a8f0b8aa255098dc62b881168777ec3
Signed-off-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/472921
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.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>
2019-11-07 13:22:50 +00:00
Xiaodong Liu
6f209f63ae test: fix Blobfs latent failures in test pool
For issue https://github.com/spdk/spdk/issues/990,
It can be solved by setting cache pool a smaller
size in order to fit CI test env.

Change-Id: Ia5478df0f25de523917597eaeedcd858b70ab1e6
Signed-off-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/471957
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-11-07 00:33:25 +00:00
Xiaodong Liu
138e6daec3 rpc/blobfs: add cache size setting rpc
This RPC method can be used to set cache size of
blobfs, in case a smaller or larger memory scale
is requred by cache pool.

Change-Id: I8046c7caeaa15c67825f86d14ae0b79395d51daf
Signed-off-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/471870
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-11-07 00:33:25 +00:00
Xiaodong Liu
2f249ace0c blobfs: add result for set_cache_size
Since g_fs_cache_size only takes effect when creating
g_cache_pool, so spdk_fs_set_cache_size is only
permitted when cache pool is already freed or hasn't
been initialized.
Add a return value to indicate the result
of spdk_fs_set_cache_size.

Change-Id: I3828b136976d6f03f0751b2f20f68cd47c36ec04
Signed-off-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/471869
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-11-07 00:33:25 +00:00
Jim Harris
e93449910a nvme: use -EFAULT for vtophys-related failures
Currently we have a mix of -1 and -EINVAL which
is confusing, especially since these types of failures
also result in the caller's callback routine getting
invoked.

While here, document this new -EFAULT return code for
all of the functions that could return it.

Fixes issue #797.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I8dfbba0ec0b83db0f2ec055b15830981af1965df

Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473054
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: yidong0635 <dongx.yi@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
2019-11-06 15:19:48 +00:00
Jim Harris
0f481877ae nvme: add consistency to I/O command function comments
1) consistently put each possible -errno value on separate
   line
2) fix -ENOMEM misspelling
3) add missing text about submitting from one thread only
   to spdk_nvme_ns_cmd_read_with_md
4) move text about submitting from one thread only for
   spdk_nvme_ns_cmd_readv_with_md
5) remove explanation of DSM from
   spdk_nvme_ns_cmd_dataset_management - we do not generally
   explain the meaning of commands in this header file

It's possible there are some other inconsistencies still, but
I cleaned up all of the ones that I noticed. I was mostly
concerned about item #1 to prepare for a new errno that will
be added in an upcoming patch.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I993c01a2f9f1f8294c1ebf349cc877a663aa43b1

Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/473053
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: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
2019-11-06 15:19:48 +00:00