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>
When a connection enters full-feature phase and is assigned to an lcore,
we need to increment the counter for the new lcore, not the connection's
existing lcore.
Change-Id: Idced4090b6e8ac35a767fd223fbd81ba824615d3
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
The 'next' event pointer was never used in the entire code base (always
NULL).
Change-Id: I75f999d3a2e10512d86edec1a5a46ef263e2635b
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This allows the elimination of the spdk_event_get_arg1() and
spdk_event_get_arg2() macros, which accessed the event structure
directly; this was preventing the event structure definition from being
moved out of the public API header.
Change-Id: I74eced799ad7df61ff0b1390c63fb533e3fae8eb
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The spdk_poller_register() function provides a way to pass an event to
call once the poller is registered, but it is always NULL in the current
code base.
Change-Id: I459bf40ae4d050589577d113b7984f1563aaa9cc
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
We cannot quit the process when user did not Logout from the session,
because the active connections always bigger than zero. User cannot
use Ctrl+C to quit SPDK iSCSI target. Add a new state to connection
to avoid destruct connection more than once.
Change-Id: I8efa79aa47534bd6ead965713769f751d9802e47
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
When we enabled the ERL1 configuation, for the DATAIN task release
process, we will queue the task to the SNACK list firstly, and then
remove the list when got ACK from initiator, but for this part of
logic, the reference count of primary task was not released correctly.
Change-Id: Ic5959cf644c74f676be0b84c5650292dc426b2d8
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
This patch make sure the connection in normal state before any further
operation on this connection.
Change-Id: I776740b5b33b1de6707990c09d9131c385adf556
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
The SPDK_TRACELOG macro depends on a CONFIG setting (DEBUG), so it
should not be part of the public API.
Create a new include/spdk_internal directory for headers that should
only be used within SPDK, not exported for public use.
Change-Id: I39b90ce57da3270e735ba32210c4b3a3468c460b
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This removes the 2 bytes of SenseLength from the beginning of the SCSI
sense_data buffer, so now the offsets within sense.data match up to the
expected values from the SCSI spec.
Change-Id: I9188560096a9ec5a8fcf83bec95201521b127494
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
For large writes that require multiple SCSI tasks (one for immediate
data, then one or more for R2T-solicited data), we bump the refcount
for the task associated with the initial immediate data PDU, to
ensure it does not get freed until all of the child tasks are
completed. But in some cases this initial immediate data PDU could
complete after all of the R2T-soliciated data PDUs. The
completion code was not handling this case correctly which would
result in the iSCSI connection thinking it still had outstanding
SCSI tasks when the connection closed.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I9f9c5322755462d1918fde0075c87c84295cb10c
Use standard GCC style atomic operations instead of
the DPDK calls. The DPDK calls end up translating
to the gcc standard inline calls in the generic
case anyway.
Change-Id: I0ea760c4e23c3660b082a803bbc174de7250f365
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
