Calling spdk_nvmf_tgt_accept() now automatically assigns new qpairs
to the best available poll group.
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Change-Id: I3df2a2c5a28dba45c5ba0cbd1e8c28dd7e56cf9e
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2813
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
As we have done for SPDK reactor, SPDK thread termination needs communication
among SPDK threads. So At application shutdown, create the while loop
until all threads are destroyed and in the loop, check each thread
is exited, and destroy the thread if exited or mark it as exiting
and retry checking after traversing other threads.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I7a1008b95f3303de17449d574e917eb47cbe54b5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2660
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
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>
Add simple periodic round robin poll group migration to the example
application to exercise SPDK thread dynamic scheduling.
Add CLI option to enable the feature manually. Default setting is
disabled, and the period is specified as micro second.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I4639e94ce00611a600ff068e754ee6f31905565c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2556
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Add the flag, resched, to struct nvmf_lw_thread.
Add an new function nvmf_request_spdk_thread_reschedule() to set
the resched flag of the current thread to true.
nvmf_reactor_run() checks the resched flag and call
nvmf_schedule_spdk_thread() and clear it if it is true.
Enable the thread operation SPDK_THREAD_OP_RESCHED and add
nvmf_request_spdk_thread_reschedule() to the case of
SPDK_THREAD_OP_RESCHED in nvmf_reactor_thread_op().
Following patch will add repeated calls to spdk_thread_set_cpumask()
to exercise the feature.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I60a4213c139b3e6f16ccb7065dea34ffe760de13
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2541
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Replace mutex and linked list by lockless ring for nvmf_reactor->threads.
If we create event call, we will be able to keep linked list and
remove only mutex. However, simple replacement of locked linked list
by lockless ring will be enough for now.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I590a385ff018ad20c7d6752b0a3a5562e3b68510
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2579
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Stop creating the default lightweight thread per POSIX thread. Additionally,
substitute the created master thread into g_init_thread directly as a
preparation to the next patch.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I09643491e6b5aaf91d9958481f05e264a2ac5f7d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2542
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Replace spdk_thread_lib_init() by spdk_thread_lib_init_ext() to
support dynamic SPDK thread scheduling in this sample application.
Add nvmf_reactor_thread_op() and nvmf_reactor_thread_op_supported(),
and then pass them by spdk_thread_lib_init_ext().
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I9498d45d74757d8893b2b415fe28bf65f8fd02d6
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2540
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
By following the SPDK NVMe-oF target application, create and destroy
a lightweight thread per poll group instead of assuming a pool of
lightweight threads already exist at start up time. Each poll group
is only accessed from the associated lightweight thread.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I0a221594d51373237f4ead07a3f5592dbbffa7a2
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2539
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
if spdk_thread_poll() calls spdk_thread_create() to create SPDK
thread dynamically, it will acquire nvmf_reactor->mutex and will
cause deadlock because the queue nvmf_reactor->threads is accessed
by any POSIX thread by holding nvmf_reactor->mutex.
If any non-owner POSIX thread manipulate nvmf_reactor->threads by
holding nvmf_reactor->threads, like general priority based scheduler,
running SPDK threads by FIFO will be better.
Change nvmf_reactor_run() to dequeue a SPDK thread from the head, run
it, and then enqueue it to the tail. Release nvmf_reactor->mutex
while spdk_thread_poll() is executed.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I9b4040cc4d995beaf4a1f9ed8260bb1254d7d7fb
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2537
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
nvmf_schedule_spdk_thread() is not so performance critical but
should work as designed. Guard the whole operation to decide the target
core by g_mutex. This matches the reactor's _reactor_schedule_thread().
If _reactor_schedule_thread() is refined, change this accordingly too.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I9382230577442897d4d2d22b85b1ae4edd77aa98
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2536
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
We will be create fine name for each poller but it will need large
effort. Replacing spdk_poller_register by the macro SPDK_POLLER_REGISTER
will provide better name than function address with minimum effort.
Following patches may improve function name for clarification.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: If862a274c5879065c3f7cb04dcb5ca7844523e68
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/1781
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: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Community-CI: Broadcom CI
Following patches will require thread termination so that thread_exit()
will move thread to exiting, thread_poll() will move thread from
exiting to exited, thread_is_exited() will detect the threadis exited,
and then call thread_destroy().
So change all places as a preparation.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I6b2e8aee5ed7cd160a88b4c9aaed7d90bd9dac07
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/1640
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The next patch will create poll group threads dynamically for
NVMe-oF target, and will need to wait for completion of poll group and
I/O channel destroy. This is a preparation for the next patch.
Add callback function and its argument to spdk_nvmf_poll_group_destroy(),
and to struct spdk_nvmf_poll_group, respectively.
The callback has not only cb_arg but also status as its parameters even
if the next patch always sets the status to zero. The reason is to follow
spdk_nvmf_tgt_destroy's callback and to process any case that the status
is nonzero in future.
spdk_nvmf_poll_group_destroy() sets the passed callback to the passed
poll group.
Then spdk_nvmf_tgt_destroy_poll_group() calls the held callback in the
end.
This change will ensure all pollers are being unregistered and
all I/O channels are being released.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Ifb854066a5259a6029d55b88de358e3346c63f18
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/495
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
The copy engine library, modules and public APIs have been renamed.
Use of the word `copy` has been replaced with the word `accel`
short for accelerator in preparation for adding new capabilities
in the future. Additionally, APIs for what was previously called
the `memcpy` engine have been renamed to identify the engine as a
software accelerator.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: Ia607aa718416146fbba1e6792b8de0f66bd8a5de
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/576
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>
MD026 Trailing punctuation in header
This rule is triggered on any header that has a
punctuation character as the last character in the line
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: I3ab4894092ef6b5f920d89b74e43c2e46e9581c0
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/657
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>
MD003 - Header style
Triggered when inconsistent header styles are used accros file.
Also adjust other headers in file to fix
MD022 - Headers should be surrounded by blank lines
MD025 - Multiple top level headers in the same document
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: I1937abc0d29dc8a21a8eac54d0af2427ebf95bdc
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/653
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>
When the call to spdk_subsystem_init returns synchronously, calling the
callback results in double iterating over states in the state machine.
Returning after calling spdk_subsystem_init allows us to reliably begin
executing after the callback function is executed with no double
execution.
Change-Id: Ieb9003acece7483c956f19be4f131c8d4f64fbaa
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/480007
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>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Add the same handler for SIGINT and SIGTERM so that
the working thread can exit.
Change-Id: If1ce617ddd778e92bed6f95089c48caf841e256f
Signed-off-by: JinYu <jin.yu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/468661
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: GangCao <gang.cao@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>
Add fc library path if FC transport is configured.
Signed-off-by: Anil Veerabhadrappa <anil.veerabhadrappa@broadcom.com>
Change-Id: Ifca51aed9cfde6fda347bbe9e331026bd7246a97
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478771
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>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
use the nvmf lib to add an nvmf example which
doesn't use the spdk app framework. It can show
how to use the nvmf lib to build your own nvmf target.
We encouage you to use the RPC cmds so this example will
not supply the configuration file. You can read the
README.md to get the steps how to build your nvmf target
and test the IO status.
Change-Id: I8d175934f7364d40fb44d36e85c31d8ea5063654
Signed-off-by: JinYu <jin.yu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/468457
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: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: GangCao <gang.cao@intel.com>
