Only calling spdk_clear_all_transfer_task cannot solve all
the hotplug issue. The iSCSI task may successfully return
and own the bdev buffer inside the iSCSI task, so we need to
call this flush pdu function.
Change-Id: I255173d0880334e8acccc980a4ce04c380f64435
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-on: https://review.gerrithub.io/428801
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>
Purpose: We need to get the port info in other applications
(e.g., NVMe-oF TCP/IP transport)
Change-Id: I3a4636e764e44425436bb064cb0062c6f3e44035
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/428313
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
The following changes are added in this patch:
1. Remove unused TRACE_ISCSI_CONN_ACTIVE and TRACE_ISCSI_CONN_IDLE.
2. TRACE_ISCSI_TASK_EXECUTED is added. This tracepoint can be used
to record TSC for every PDU, especially non-SCSI PDU.
3. TRACE_ISCSI_PDU_COMPLETED is added. This tracepoint is for SCSI PDU.
It will record the TSC which includes the cycles of flushing writebuf.
Change-Id: I277bb2ce57e3cd2fc6fcb25845abdf2c11ed062b
Signed-off-by: Liang Yan <liang.z.yan@intel.com>
Reviewed-on: https://review.gerrithub.io/425036
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>
Currently all poll group operations are done on the core allocated
to the connection but spdk_env_get_current_core() is called in every
operation and they are not explicitly checked yet.
This patch adds check and uses conn->lcore to operate poll group.
Change-Id: Id3a882a51f4f2d7ed0074f9cd9a8ad5bd830234c
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/424743
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This was added a long time back for tracking an rte_mbuf
whose buffer was a different rte_mbuf - all related to
a userspace TCP stack that is no longer in development.
The concept isn't useful now, so remove it to reduce
the complexity of the tracing code.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I310e492eba7f55df242bb29d82fb19f6daee1f51
Reviewed-on: https://review.gerrithub.io/424565
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>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
To be consistent with nvmf, use underscore.
Change-Id: I5f2ae60518367ab439db9b893ab8ba975e82bed9
Signed-off-by: Liang Yan <liang.z.yan@intel.com>
Reviewed-on: https://review.gerrithub.io/424179
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>
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>
For secure iSCSI targets, dynamic reconfiguration of CHAP secrets is
must to have.
Currently CHAP secrets are loaded for every CHAP authentication
operation. The current implementation will not work correctly
when CHAP secrets in the file are changed dynamically.
If SPDK loads CHAP secrets from the file only at boot and they can be
configured by RPCs, user can change CHAP secrets safely during run time.
Even if there are any users who expect dynamic reconfiguration of
CHAP secrets based on the current implementation, if we provide
this better alternative based on RPCs, they will be able to continue to
satisfy their requirement.
This patch change the current implementation so that SPDK loads
CHAP secrets from the file once at boot and uses them in memory
hereafter.
Besides, use fixed size buffers to hold CHAP secrets.
Previously dynamically allocated buffers by strdup() had been used,
but it required many nomem checks and should be avoided.
Other iSCSI targets/initiators have used fixed size buffers and SPDK
follows others.
Set the size of buffers for both user names and secrets to 256 (the
last byte is for NULL termination). 256 is sufficiently large
compared with others.
CHANGELOG will be updated in the separate patch because new RPCs will
be added instead.
Change-Id: I499e792817c2ed01c3d970bbd3d34a6b1fccf65b
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/421463
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: Jim Harris <james.r.harris@intel.com>
If the connection is already in exiting or exited
status, we should not set it again. And this
could prevent the coredump.
Change-Id: Ia506d13b12c3a6cb5619d65e4b3353b149a85947
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/421668
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>
Reason: If the task is queued in scsi layer to handle, we
should free the task after calling the spdk_iscsi_task_cpl.
Otherwise, if only this task is executed (without subread tasks),
the task will be freed by the loop early, which will cause
the segment fault in spdk_iscsi_task_cpl function.
Change-Id: Ifc42399957b24d976af5fd12f6e33459a3ea86ba
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-on: https://review.gerrithub.io/421706
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>
Due to the network issue, spdk_iscsi_conn_free_pdu in
spdk_iscsi_conn_flush_pdus_internal will not be executed.
So consider the pdu free, we should call spdk_iscsi_conn_free_pdu
in spdk_iscsi_conn_free_tasks.
Actually, for the task which in queued_datain_task, we have
the following case:
1 The task is not sent to the scsi layer: it means that
the task is not freed, so we should call spdk_iscsi_task_put
here.
2 The task is sent to the scsi layer, but no subtasks
are sent to the scsi layer: It means that the call
back function (spdk_iscsi_task_cpl) will be called,
but since it will have the subtask, so spdk_iscsi_task_put
will not be called, thus, we should call spdk_iscsi_task_put
here.
3 The task is sent to the scsi layer, and some subtasks
are also sent to the scsi layer: It also mean that the
spdk_iscsi_task_put will not be called in spdk_iscsi_task_cpl,
and not all the subtasks will be finished, so the father task
will not be freed, so we should still call spdk_iscsi_task_put
here.
4 The task is sent to the scsi layer, and all the subtasks
are also sent to the scsi layer,
thus this task is not in the queued_data_in_task.
So according to 1-4, we should call spdk_iscsi_task_put here,
and also decrease the data_in_cnt;
Change-Id: Icb13df1ae07f6eea0247d45f4a0397edc4aa2500
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/420875
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: Changpeng Liu <changpeng.liu@intel.com>
When hot removal of a LUN is started, callback is called for each
iSCSI connection which accesses the LUN. Callback checks all transfer
tasks complete and then close the LUN.
If the connection clears all transfer tasks before getting all responses
to them from the initiator, the initiator continues to retry data write.
Hence the connection have to wait until all transfer tasks complete.
Change-Id: Iad9063673cfedbd78758890d55a4254512e4fca4
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/417199
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Change-Id: I6babd4cf990bf19b510db88bdfb0ca81e29d9252
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/414700
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Madhu Pai <mpai@netapp.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
This will simplify remove operation a little.
Besides a few unnecessary blank lines are deleted.
Change-Id: Ifd312e5a94ba0668d884dd5baf1ef1f2ebd4417c
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/414562
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Try to make the logic of the function a little clearer.
Change-Id: Ic78bed73feef94d94487f0a4240019789be8d2bc
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/414353
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This is not needed since now we have group iSCSI
poller
Change-Id: I99127a62fc84cee5088b16c908edd78876d0fcfb
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-on: https://review.gerrithub.io/414620
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reason: If we trigger the spdk_iscsi_conn_migration,
we may face the issue that the io channel of lun
is NULL(which means not allocated). So I think
that if we trigger the migration, we need to
stop the executing the further pdu of this conn
by the current polling group.
Also since the connection is triggered migration,
we should stop execution on the current core, and
let this connection be handled by next round.
Change-Id: I0ab89d79c976f3233890ae25cb7eac98de5e30ac
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/409984
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
The initiator may close the connection unexpectly
due to many differnent reasons, which should not
print error log of our SPDK iSCSI target.
Change-Id: I652be75c0762547f27212503e5b52bfc898ecc8f
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/412593
Tested-by: SPDK Automated Test System <sys_sgsw@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>
Store the file descriptor for the shm_open()'d connections array so we
can clean it up at shutdown.
Change-Id: Ia2727f837431cb1de2c077718412e70364135a5f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/408533
Tested-by: 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>
Some places use NULL to check for mmap failure but mmap return
MAP_FAILED in this case.
Change-Id: I4796fa52421da53c94223a9e8cc26ac04968f1d8
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/405648
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
The static variable g_connections_per_lcore can be configured by
.INI config file. However it had not been added in the iSCSI options
because it was not iSCSI global parameter.
This patch is necessary for JSON config file to configure it.
Change-Id: I12aa1d94dd467969a6853b9fb3f8a627d0d70766
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/403623
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The sock abstraction wraps these functions now, so the iSCSI layer
doesn't need to include them directly.
Change-Id: I900ed96a9cfb5d12c7fd190bc590d00846a47034
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/403879
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Trace the initiator and target name for debugging purpose.
Change-Id: I89230484072a7f12a58b4234abd1026a228f302a
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.gerrithub.io/403690
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This will be used to track time used in pollers - each poller can now
indicate if it found any work to do or not.
For cases where it was obvious and the infrastructure was already in
place, existing pollers have been modified to return 0 or a positive
value to indicate whether work was done. Other pollers have been
modified to return -1 by default, indicating that the poller isn't
indicating anything about whether work was performed. This will allow
us to find un-annotated pollers easily in the future and fix them
incrementally.
Change-Id: Ifebfa56604a38434fac5c76ba7263267574ff199
Signed-off-by: Roman Sudarikov <roman.sudarikov@intel.com>
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/391042
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
If we are not in ffp phase, we do not need to update last_nopin
Change-Id: I289beec8437d7117b6261483c333bf5db5fbd11f
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/403166
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Also remove the related login and full feature
function
Change-Id: Ia2d0a6617910134c889d1322b53830fe26b9e956
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/402486
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
This patch will also re-export spdk_iscsi_conn_destruct
Change-Id: I7d6cbba33294c84398a919d888c8b375e901e09e
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/402485
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Put it in the iscsi.c
Change-Id: Ifb2843fb7c78f9ca948eac60704547b8b0635bf0
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/402484
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The better location is to put this function in
the end of process_read_task_completion which can match
the the same function call in spdk_iscsi_op_scsi_read.
Change-Id: I7dbfb23e2c469e0be22e148299643a99b93c8018
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/401985
Tested-by: 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>
Since we have two pollers for each iSCSI polling group,
so it could be possible that the nop_poller to set
iSCSI connection in existed or existing state.
So add a check here, which means the incoming data
is too late.
Change-Id: I94e5090b005c35c9eb5d7eedbd7acdae67327f94
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/402039
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Purpose: To make the logic clear.
Change-Id: I6176f359a23816b316e92b63ee2e5d5175ae7e7b
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/401772
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This patch is used to implement the sock fd event group
polling mechanism if there are incoming data from network
(read event in SPDK iSCSI target side) , thus we can awake
many connections one time, and it can reduce the system call
overhead.
Change-Id: I76c26a89ef9365d7e1ccec616985e7435253896b
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/399796
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The iSCSI connection does not have the poller anymore, so
change the name
Change-Id: I534f72998c6bcc73ad9caf5e8f700751acd95c99
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/401372
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ic2bae1c750d2c9b35bded8ed96c3b84832690ce2
Reviewed-on: https://review.gerrithub.io/401479
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
This patch adds a iSCSI poll group for each core,
thus we can poll a group of connections instead of one
by one on each core.
Change-Id: I9cd82c8c0f7f6e7b3ee34034339638199bfca5da
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/399565
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
In early days of SPDK, there was an experimental userspace
TCP stack called libuns. The SPDK iSCSI target supported
using either the default POSIX/kernel TCP stack or this
libuns TCP stack.
libuns is no longer in use, but work has started on
supporting the FD.io VPP TCP userspace stack. Abstractions
are being put in place to allow *both* the POSIX and VPP
stacks simultaneously.
So remove the concept of a "default" net_framework that is
defined with weak symbols that can be overridden by another
object file. While here, also remove the "clear_socket_association"
concept which was specific to libuns.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I8a0385ca5ae113e34a637f835d8d69f8d510433e
Reviewed-on: https://review.gerrithub.io/400328
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
For now, this provides common abstraction for Linux epoll
and FreeBSD kqueue. It also provides the basis for future
changes where alternate userspace TCP stacks have their own
mechanism for polling a group of descriptors.
While here, remove old epoll/kqueue code in iscsi/conn.c that
was commented out when the iSCSI idle connection code was
recently removed - we now have a real implementation of it
in sock.c so the original code is no longer needed as a
reference.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I664ae32a5ff4d37711b7f534149eb0eb35942335
Reviewed-on: https://review.gerrithub.io/398969
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
This provides an abstraction layer around TCP
sockets. Previously we just used fd integers, but
we don't want to be tied to integers for alternative
userspace TCP stacks.
Future patches will do more work to enable multiple
implementations of this abstraction. For now, just
get the abstraction in place for POSIX sockets and
make all of the iSCSI changes associated with it.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I9a825e9e02eb6927c8702d205665c626f57b3771
Reviewed-on: https://review.gerrithub.io/398861
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
The socket-related code was already broken out into
lib/net/sock.c, so break out the header portions
from include/spdk/net.h into its own sock.h.
This prepares for some upcoming changes in how
TCP sockets are abstracted, to enable alternative
userspace TCP stack implementations to be used with
SPDK.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I40b162e72ea80c235b49f10b17c2085fcfb385d4
Reviewed-on: https://review.gerrithub.io/398851
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
This removes PDU flushing from the main connection poller -
all PDU flushing is now done inline when the PDU is completed,
or from the context of the new flush poller.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I703cb451d4b548e7f75892e31a7927b511fb4f55
Reviewed-on: https://review.gerrithub.io/395543
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I39152be9653316a24135cde497662cfbc8f68e3a
Reviewed-on: https://review.gerrithub.io/395523
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
Fixes github issue #218.
This patch introduces spdk_cpuset object to store and manipulate
the set of individual CPUs. The main objective of this object is
to replace cpumask declared as uint64_t and extend the limitation
of supported CPUs (lcores) above 64 CPUs.
spdk_cpuset is always allocated dynamically and accessed by opaque
pointer, what makes it easier to extend in the future without
breaking API/ABI.
This patch also extends parsing function allowing to set cpumask
using a list of cpus e.g. "[0-4,10,12]" sets mask of 0,1,2,3,4,10,12
as well as hexadecimal string with and without "0x" prefix.
Change-Id: I475c3ba7fab629021a22e03176e57e400dd24a49
Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Reviewed-on: https://review.gerrithub.io/390794
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
To use JSON dump of iSCSI global parameters as configuration file,
at bootup, it must have completeness.
This patch fixes the following:
- DefaultTime2Wait was not contained.
- MinConnectionsPerCore was not contained.
- Converting only 0 to "None" for dump is not efficient.
Change-Id: I740e6938c216c5c62df6ee70b0ceac40fd0cac00
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/396100
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Similar functionality is needed when migrating a connection
to its new core after login. So make spdk_iscsi_conn_stop_poller
reusable for that case, and call it in the login migration path.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ie014a2e50d6ed02165317deab85c943225953197
Reviewed-on: https://review.gerrithub.io/395857
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
All calls to this function now use spdk_env_get_current_core()
as lcore, so remove this parameter.
All calls to this function also use _spdk_iscsi_conn_free() as
the stop_fn, so remove that parameter as well - just call that
function directly from the callsites and eliminate the extra
event.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I639e63d26fbfcc054137d9b453be2927e2c1f625
Reviewed-on: https://review.gerrithub.io/395854
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This function is now only called from one place, and moving
this inline will simplify some future patches which will
consolidate the login and full-feature pollers.
No functional change.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I7d4e6973882c7b579e341283648317e8e06d998a
Reviewed-on: https://review.gerrithub.io/395853
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Some upcoming changes will effectively render this moot
anyways by adding an epoll/kqueue descriptor to poll
on in all cases (not just connections that have been
idle for 5ms).
The epoll/kqueue code was just ifdef'd out instead
of removed - some of this code will be useful
and reusable with minimal changes in the upcoming
patches.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I0c354390537e6369cb3c32e78a59c300dec6d098
Reviewed-on: https://review.gerrithub.io/395553
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
This function was originally named spdk_iscsi_write_pdu()
in lib/iscsi/iscsi.c. Since this is an operation on a
connection, add "conn" to the name and move it to conn.c.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Iaad022907d43788108d5b2660306abcf5e94040d
Reviewed-on: https://review.gerrithub.io/395522
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
This will enable some future enhancements where the main
iSCSI connection poller will not try to flush PDUs, we will
just start a separate poller to periodically flush PDUs in that
case.
This is all part of broader scheme to enable epoll() for
iSCSI. Making these changes allows us to avoid using epoll()
to signal when TCP buffer space is available.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I6fe9314e386673fcc87bb15580ef808838c55a7b
Reviewed-on: https://review.gerrithub.io/395521
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
This concept was added very early in the SPDK iSCSI target
development process when there was a high focus on maximizing
throughput on a single iSCSI connection with 512-byte reads.
Realistically, in multi-connection environments focused on
predominantly 4KB (or more) workloads, this concept loses its
effectiveness - it is relatively rare that PDUs from multiple
I/O would coalesce within the default 8us flush timeout period.
There were no users of flush timeout in the SPDK tree - it was
not even documented in the iSCSI configuration file example and
there was no way to modify it at run time. So leave the change
out of the CHANGELOG.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I08c3e959fb3945dc2c9cb89248305d0c88aa778f
Reviewed-on: https://review.gerrithub.io/395520
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Change-Id: I343d2f1a8f3327ad030a786633bbe9e8c66c14fc
Signed-off-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/394138
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
In the patch 393582, by analyzing the code,
accumulation of write completion to bdev was duplicated by
data_transferred and bytes_completed.
Hence accumulation of data_transferred for write was intended
to be removed.
However by mistake accumulation of data_transferred for read
was removed.
In this patch restore accumulation of data_transferred for
read and remove accumlation of data_transferred for write.
Test code to catch this degradation will be added by another
patch soon.
Change-Id: Iea9883e8ef1bfb0bdf00e291712e6faf2fad281f
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/393713
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Each SCSI task should only report its data_transferred; the
parent/subtask relationship should be handled in iSCSI.
Change-Id: Ibece110fd8cca4e94648942fe6b5e004a4fd8a80
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/393212
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This patch remove need for additional buffer when translating error code
to string.
Change-Id: Iaa60088b5c450581d3cdddbb425119b17d55a44b
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/386114
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I7df72b21f78ab7f0fe0389b478072330bbbd4215
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/393015
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reason: It only handles the queue datain tasks. After the changing,
it would be more accurate for the code reading.
Change-Id: I87999f811810cadd4b58d99be1cdeba0a1a7503f
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/392719
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Also use this function in iscsi/conn.c
Change-Id: I25f6da175eddb12c4ac2624d695c2c43c871d8e8
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/392713
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
For g_num_connections, we should create an proper
array size, we cannot directly create it by the size:
spdk_env_get_core_count(). The reason is that the
core mask can be non-continuous,e.g., 0x1001, thus
for effient access, we create a large array size with
last_core +1, although we will have some space waste,
but this will not be big, but still maintain the fast
array index acccess.
Change-Id: I95e1fc34e0816ac2f8764880c0d0e629f43a5dc4
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/391909
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Load balancing for idle iSCSI connections uses the RTE EAL Launch
state and uses DPDK RTE EAL API.
But all SPDK reactors will exit simultaneously because each SPDK
reactor checks if the global state is RUNNING to exit.
Hence calling rte_eal_get_lcore_state() is not necessary.
When the reactor hot-plug function is supported, this implementation
will be reconsidered together.
Change-Id: I34eaf3e42b5b7deae6473d2bfaf0910aaa9da6de
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/391339
Tested-by: 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>
Currently idle iSCSI connections are managed by the master lcore,
but the master lcore is like BSP of OS and for initialization.
To manage idle iSCSI connections it is important that the core is
consistent.
Hence the first core is better than the master lcore.
In this patch the following are changed together:
- Errors of kqueue() and epoll_create1() are not related with master
lcore. "master lcore" is removed and errno is added into the log.
- In spdk_iscsi_conn_allocate_reactor(), when cpumask is 0, 0 is
selected as core number. 0 is not safe and first_core is used instead.
In spdk_iscsi_conn_allocate_reactor(), when first_core is used instead
of master_lcore, we may observe some contradiction in the following
code. But few changes are done in this patch.
In the current implementation we can assume the first lcore is
equal to the master lcore and the following code will be removed
in the subsequent patch.
/*
* DPDK returns WAIT for the master lcore instead of RUNNING.
* So we always treat the reactor on master core as RUNNING.
*/
if (i == master_lcore) {
state = RUNNING;
} else {
state = rte_eal_get_lcore_state(i);
}
Change-Id: I6cac06c27b289db5ea1f9452e33489286c64d2fa
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/391338
Tested-by: 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>
Before removing the DPDK dependency from the iSCSI connection
load balancing, this should be done.
In spdk_iscsi_conn_allocate_reactor(cpumask)
- if any lcore[i]'s state is FINISHED, the caller calls
rte_eal_wait_lcore(i). But the purpose of rte_eal_wait_lcore()
is to check if the slave is in a WAIT state before calling
rte_eal_remote_launch(). The meaning of this usage is not clear.
- If the state of lcore[i] is WAIT or FINISHED, the reactor does
not run on the lcore[i]. iSCSI connections consist of not reactor
but poller. Hence selecting lcore[i] with the state WAIT or
FINISHED does not look correct.
Change-Id: If8c420f2d16dc44e77f8963f5732faa52e3d829b
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/391332
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
All SPDK libraries should use the spdk/log.h family of functions
for logging.
Change-Id: I2b8ac30f2901b325784552f0016f1058ae2cd577
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/391687
Tested-by: 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>
Require braces around all conditional statements, e.g.:
if (cond)
statement();
becomes:
if (cond) {
statement();
}
This is the style used through most of the SPDK code, but several
exceptions crept in over time. Add the astyle option to make sure we
are consistent.
Change-Id: I5a71980147fe8dfb471ff42e8bc06db2124a1a7f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/390914
Reviewed-by: <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@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>
Disambiguate the log components from the trace functionality
(include/spdk/trace.h).
The internal spdk_trace_flag structure and related functions will be
renamed in a later commit - this is just a find and replace on
SPDK_TRACE_* and SPDK_LOG_REGISTER_TRACE_FLAG().
Change-Id: I617bd5a9fbe35ffb44ae6020b292658c094a0ad6
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/376421
Tested-by: 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>
Make this generic and not directly dependent on
the event framework. That way our libraries can
register pollers without adding a dependency.
Change-Id: I7ad7a7ddc131596ca1791a7b0f43dabfda050f5f
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/387690
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Change-Id: I8ea8c59a0c67176c0c0c39abf807afad61ff3828
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/387689
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Change-Id: I864cd8a6c206dfbe62fcb3f72275c1ae51aa4ed7
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/387688
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
We always unregister pollers from the core they're running on,
so we don't need the async completion event when unregistering
them. To avoid changing the logic, create events instead.
Change-Id: Ib50d93e1578684bea69200d89a5e0d383acc8cfe
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/387685
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Register all pollers to the current core. If necessary, send
an event to the correct core before registering.
Change-Id: Ie34cc8b11143a58c0f621c87c409a3d09d929648
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/387682
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
When an iSCSI connection is closed, spdk_iscsi_conn_stop_poller()
is called in spdk_iscsi_conn_destruct() or
spdk_iscsi_conn_check_shutdown().
However, through both paths, g_num_connections[] is decremented
twice:
(1) in spdk_iscsi_conn_stop_poller()
(2) in spdk_iscsi_conn_free()
This behavior makes g_num_connections[] negative unexpectedly and
for some cases load balancing does not work and the same CPU is
always selected.
How to fix:
spdk_iscsi_conn_stop_poller() is called in the following functions:
- spdk_iscsi_conn_check_shutdown()
- spdk_iscsi_conn_destruct()
- spdk_iscsi_conn_handle_idle().
[Idea 1]
Remove the code to decrement g_num_connections[] from
spdk_iscsi_conn_free().
[Idea 2]
Remove the code to decrement g_num_connections[] from
spdk_iscsi_conn_stop_poller(). Add the code to decrement
g_num_connections[] to __add_idle_conn().
[Idea 1] is simple because just only one line is deleted.
[Idea 2] may be more symmetric than [Idea 1] but to find the right
place to add the code in __add_idle_conn() is not clear and
difficult.
Hence [Idea 1] is proposed.
[Idea 1] may not be the optimal one and more refactoring may be
done.
Change-Id: I81050f4a0e7b3ddd40896f46ab2eb8de14bbcb3a
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/384026
Reviewed-by: Cunyin Chang <cunyin.chang@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
With the new asynchronous subsystem finish framework, we can
drive shutdown of existing connections as part of the subsystem
finish path instead of a separate spdk_iscsi_shutdown function
called as the shutdown function in response to SIGINT.
This is a step towards enabling a single target app that
supports multiple protocols (i.e. iSCSI + vhost + NVMe-oF).
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Id9f596a8091912a72ab7eb93cb45a46fdb130a48
Reviewed-on: https://review.gerrithub.io/386695
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I9218b618544def301a4509eda1b266390a80dba5
Reviewed-on: https://review.gerrithub.io/382830
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
A few foundational change to support safe removal of portal.
- global login poller -> login poller per portal
- Caching portal data for active connection
Change-Id: I62f4d90c9ac11a433ad47421b2b0c69bfc3c70b7
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/379930
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Replace SPDK_TRACE_DEBUG with component-specific flags.
Change-Id: Iee7eafab5e6ac8713f247323a18552b5afb0e86a
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/375834
Tested-by: 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>
This matches the name to the behavior and prepares for addition of a new
log macro for "info" log level.
Change-Id: I94ccd49face4309d3368e399528776ab140748c4
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/375833
Tested-by: 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>
replaces all references to strerror in the spdk lib directory with
references to the thread safe strerror_r
Change-Id: I80d946cce3299007ee10500b93f7e1c8e503ee41
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/374012
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
There is one scenario, when one large read IO is splitted into several
subtasks 1 - N, and the subtask N return successful, it will be stored
in the subtask_list of primary task, but the other tasks could return
error, then when we try to return the subtask N, it will failed, this patch
will make sure the large read IO command return correctly if any of the
subtasks failed.
Change-Id: Ie6bcb79a081acf30f4e4c177c0f7568495ff2b71
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
Reviewed-on: https://review.gerrithub.io/373450
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Change-Id: Ic4221e044eb5af92aa06e468eb989bee285022af
Signed-off-by: Jan Kryl <jan.kryl@nexenta.com>
Reviewed-on: https://review.gerrithub.io/371303
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This patch move the spdk_iscsi_conn_handle_queued_tasks() to the main
loop of connection poller, make the logic more clear, it will also fix
one issue of hotplug:
for large read command, it has the potential risk:
one task will be split into N subtaks, when primary task return error and
try to send data in pdu, it will call spdk_iscsi_conn_handle_queued_tasks(),
the primary task is pending now, and all the subtasks will return error from
lun layer synchronous, this make the primary task return from the function
spdk_iscsi_transfer_in() after all the other subtaks, but when the N - 1 subtask
return from function spdk_iscsi_transfer_in(), it meet the condition:
primary->bytes_completed == primary->scsi.transfer_len
then it will send response pdu, after this, the primary task return from
spdk_iscsi_transfer_in(), it also meet the condition, so it will also try
to send response pdu, this will make the application run into error.
Change-Id: I72206c1ce303f5fb6bd650713742d5819a88a30f
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
Reviewed-on: https://review.gerrithub.io/370339
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
We already dereference task in the line above, and all callers ensure
that task is not NULL already.
Change-Id: Iff23503959116083375082b1ab022a156438cc8b
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/367109
Tested-by: 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>
This is an artifact from a past design. There is no longer
any reason to create an event here - the bdev layer will
correctly queue events and call completions on the correct
thread.
Change-Id: I145dab4046899834c0449ec7380dcbb28215b493
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/364831
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
The SCSI layer was not using the task ID for anything; the iSCSI layer
was using it to store the task tag, so move it there and rename it to
"tag" to make its purpose clear.
Change-Id: Ibda4f4e215056116b9be4a3a0264f98bc4c29535
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The SCSI layer doesn't use subtasks; these are an iSCSI layer concept.
Change-Id: I83871f02362f10fd4ecd4b2a1544eb76bfa53595
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
When destroying a connection, we need to check if we got to
full feature phase before freeing any io channels. This is because
the io channels are only allocated as part of a successful login.
The Calsoft iSCSI test suite has tests which will fail login.
Since the test system was just using a malloc backend with memcpy,
so even though a channel was NULL in some cases, it was never used
since the memcpy engine doesn't need it.
This prepares for some future patches which extend the use
of io channels in the bdev layer.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I2fb7b18a781caa0aadca319aa1e61a6ccf2c55fd
Fix up the existing comment blocks misaligned in the first column.
Also add line numbers to the comment checks.
Change-Id: I9d28c365271df36e7013d74cbb02d0023ab4f581
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This patch fix the issue when large IO failed:
when we handle the read command which need split, we need make
sure all the subtasks to be handled if one of the subtask failed,
this will make sure the command have chance return back to initiator.
Change-Id: I0c01e1a34c6179fce37ab52c8121268b6ee31102
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
By default, all SPDK applications will not share memory.
To share memory, start the applications with the same
shared memory id.
Change-Id: Ib6180369ef0ed12d05983a21d7943e467402b21a
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
