This patch bumps the default TCP listen backlog within iscsit_setup_np()
from 5 to 256, in order to reduce the overall latency caused by a small
backlog with 100's of simultaneous login attempts directed to the same
single network portal.
Also add a ISCSIT_TCP_BACKLOG macro in iscsi_target_core.h.
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch bumps the default TA_DEFAULT_CMDSN_DEPTH from 16 -> 64,
which is closer to a sane default for larger pipes @ 10 Gb/sec
with traditional iSCSI, or @ 40/56 Gb/sec Ethernet/Infiniband with
iSCSI Extentions for RDMA.
There is really no downside to increasing this default value for
1 Gb/sec.
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch changes iscsi-target to use transport_alloc_session_tags()
pre-allocation logic for per-cpu session tag pooling with internal
ida_alloc() + ida_free() calls based upon the saved se_cmd->map_tag id.
This includes tag pool setup based upon per NodeACL queue_depth after
locating se_node_acl in iscsi_target_locate_portal().
Also update iscsit_allocate_cmd() and iscsit_release_cmd() to use
percpu_ida_alloc() and percpu_ida_free() respectively.
v5 changes;
- Convert to percpu_ida.h include
v2 changes:
- Fix bug with SessionType=Discovery in iscsi_target_locate_portal()
Cc: Or Gerlitz <ogerlitz@mellanox.com>
Cc: Kent Overstreet <kmo@daterainc.com>
Signed-off-by: Nicholas Bellinger <nab@daterainc.com>
This command converts iscsi/isert-target to use allocations based on
iscsit_transport->priv_size within iscsit_allocate_cmd(), instead of
using an embedded isert_cmd->iscsi_cmd.
This includes removing iscsit_transport->alloc_cmd() usage, along
with updating isert-target code to use iscsit_priv_cmd().
Also, remove left-over iscsit_transport->release_cmd() usage for
direct calls to iscsit_release_cmd(), and drop the now unused
lio_cmd_cache and isert_cmd_cache.
Cc: Or Gerlitz <ogerlitz@mellanox.com>
Cc: Kent Overstreet <kmo@daterainc.com>
Signed-off-by: Nicholas Bellinger <nab@daterainc.com>
This patch adds a sock->sk_state_change() -> iscsi_target_sk_state_change()
callback in order to handle transient TCP failures during the login process,
where sock->sk_data_ready() -> iscsi_target_sk_data_ready() may not be
called to release connection resources, and relinquish tpg->np_login_lock
via iscsit_deaccess_np()
It performs the sk->sk_state check using iscsi_target_sk_state_check() to
look for TCP_CLOSE_WAIT + TCP_CLOSE, and invokes schedule_delayed_work() ->
iscsi_target_do_cleanup() to perform the remaining cleanup from process
context.
It adds an explicit sk_state_check to iscsi_target_do_login() in order
to determine a state failure when iscsi_target_sk_state_change() may
not be able to proceed before LOGIN_FLAGS_READY=1 is set.
Also use sk->sk_sndtimeo -> sk->sk_rcvtimeo settings during login to
iscsi_target_set_sock_callbacks(), and revert back post login to use
MAX_SCHEDULE_TIMEOUT in iscsi_target_restore_sock_callbacks().
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch adds support for login negotiation multi-plexing in
iscsi-target code.
This involves handling the first login request PDU + payload and
login response PDU + payload within __iscsi_target_login_thread()
process context, and then changing struct sock->sk_data_ready()
so that all subsequent exchanges are handled by workqueue process
context, to allow other incoming login requests to be received
in parallel by __iscsi_target_login_thread().
Upon login negotiation completion (or failure), ->sk_data_ready()
is replaced with the original kernel sockets handler saved in
iscsi_conn->orig_data_ready.
v3 changes:
- Convert iscsi_target_sk_data_ready() lock access to
write[lock,unlock]_bh()
- Only clear LOGIN_FLAGS_READ_ACTIVE when iscsi_target_do_login()
returns zero
- Add LOGIN_FLAGS_READY + LOGIN_FLAGS_CLOSED bit checks to
iscsi_target_sk_data_ready()
- Make INIT_DELAYED_WORK() + iscsi_target_set_sock_callbacks() setup
happen earlier by moving from iscsi_target_start_negotiation() into
iscsi_target_locate_portal()
- Set LOGIN_FLAGS_READY bit in iscsi_target_start_negotiation()
after iscsi_target_do_login() returns zero.
v2 changes:
- Add login_timer in iscsi_target_do_login_rx() to avoid
possible endless sleep with MSG_WAITALL for traditional
iscsi-target in certain network configurations.
- Convert lprintk() -> pr_debug()
- Remove forward declarations of iscsi_target_set_sock_callbacks(),
iscsi_target_restore_sock_callbacks() and iscsi_target_sk_data_ready()
- Make iscsi_target_set_sock_callbacks + iscsi_target_restore_sock_callbacks()
static (Fengguang)
- Make iscsi_target_do_login_rx() safe for iser-target w/o conn->sock
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch changes iscsit_add_reject() + iscsit_add_reject_from_cmd()
usage to not sleep on iscsi_cmd->reject_comp to address a free-after-use
usage bug in v3.10 with iser-target code.
It saves ->reject_reason for use within iscsit_build_reject() so the
correct value for both transport cases. It also drops the legacy
fail_conn parameter usage throughput iscsi-target code and adds
two iscsit_add_reject_cmd() and iscsit_reject_cmd helper functions,
along with various small cleanups.
(v2: Re-enable target_put_sess_cmd() to be called from
iscsit_add_reject_from_cmd() for rejects invoked after
target_get_sess_cmd() has been called)
Cc: Or Gerlitz <ogerlitz@mellanox.com>
Cc: Mike Christie <michaelc@cs.wisc.edu>
Cc: stable@vger.kernel.org # 3.10+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
All REJECT response setup of the rejected payload is now done using
on-demand cmd->buf_ptr allocations.
Go ahead and remove dead iscsi_conn->bad_hdr usage rx_opcode path
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch adds a auth configfs group context following existing
explict NodeACL and discovery auth within:
/sys/kernel/config/target/iscsi/$TARGETNAME/$TPGT/auth/
This patch allows these attributes to be used for CHAP authentication
an TPG is configured in demo-mode (generate_node_acl=1).
Note this authentication information takes precedence over NodeACL
authentication when struct se_node_acl->dynamic_node_acl is present.
Cc: Dax Kelson <dkelson@gurulabs.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch changes ISCSI_OP_TEXT handling of SendTargets=[iqn.,eui.]
payloads to return explicit discovery information.
It adds checks to iscsit_process_text_cmd() and adds the special single
$TARGETNAME discovery case in iscsit_build_sendtargets_response() code.
Cc: Or Gerlitz <ogerlitz@mellanox.com>
Cc: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch moves ISCSI_OP_TEXT PDU buffer sanity checks to
iscsit_process_text_cmd() code, so that it can be shared
with iser-target code.
It adds IFC_SENDTARGETS_ALL + iscsi_cmd->text_in_ptr in order
to save text payload for ISCSI_OP_TEXT_RSP, and updates
iscsit_release_cmd() to assigned memory.
Cc: Or Gerlitz <ogerlitz@mellanox.com>
Cc: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch converts struct iscsi_cmd memory allocation + free to use
->iscsit_alloc_cmd() iscsit_transport API caller, and export
iscsit_allocate_cmd() symbols
Also add iscsi_cmd->release_cmd() to be used seperately from
iscsit_transport for connection/session shutdown.
v2 changes:
- Remove unnecessary checks in iscsit_alloc_cmd (asias)
- Drop iscsit_transport->iscsit_free_cmd() usage
- Drop iscsit_transport->iscsit_unmap_cmd() usage
- Add iscsi_cmd->release_cmd()
- Convert lio_release_cmd() to use iscsi_cmd->release_cmd()
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch adds RDMAExtensions, InitiatorRecvDataSegmentLength and
TargetRecvDataSegmentLength parameters keys necessary for iser-target
login to occur.
This includes setting the necessary parameters during login path
code within iscsi_login_zero_tsih_s2(), and currently PAGE_SIZE
aligning the target's advertised MRDSL for immediate data and
unsolicited data-out incoming payloads.
v3 changes:
- Add iscsi_post_login_start_timers FIXME for ISER
v2 changes:
- Fix RDMAExtentions -> RDMAExtensions typo (andy)
- Drop unnecessary '== true' conditional checks for type bool
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch performs the initial conversion of existing traditional iscsi
to use iscsit_transport API callers. This includes:
- iscsi-np cleanups for iscsit_transport_type
- Add iscsi-np transport calls w/ ->iscsit_setup_up() and ->iscsit_free_np()
- Convert login thread process context to use ->iscsit_accept_np() for
connections with pre-allocated struct iscsi_conn
- Convert existing socket accept code to iscsit_accept_np()
- Convert login RX/TX callers to use ->iscsit_get_login_rx() and
->iscsit_put_login_tx() to exchange request/response PDUs
- Convert existing socket login RX/TX calls into iscsit_get_login_rx()
and iscsit_put_login_tx()
- Change iscsit_close_connection() to invoke ->iscsit_free_conn() +
iscsit_put_transport() calls.
- Add iscsit_register_transport() + iscsit_unregister_transport() calls
to module init/exit
v4 changes:
- Add missing iscsit_put_transport() call in iscsi_target_setup_login_socket()
failure case
v2 changes:
- Update module init/exit to use register_transport() + unregister_transport()
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Pass the sense reason as an explicit return value from the I/O submission
path instead of storing it in struct se_cmd and using negative return
values. This cleans up a lot of the code pathes, and with the sparse
annotations for the new sense_reason_t type allows for much better
error checking.
(nab: Convert spc_emulate_modesense + spc_emulate_modeselect to use
sense_reason_t with Roland's MODE SELECT changes)
Signed-off-by: Christoph Hellwig <hch@lst.de>
Cc: Roland Dreier <roland@purestorage.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
The sleeping code in iscsi_target_tx_thread() is susceptible to the classic
missed wakeup race:
- TX thread finishes handle_immediate_queue() and handle_response_queue(),
thinks both queues are empty.
- Another thread adds a queue entry and does wake_up_process(), which does
nothing because the TX thread is still awake.
- TX thread does schedule_timeout() and sleeps forever.
In practice this can kill an iSCSI connection if for example an initiator
does single-threaded writes and the target misses the wakeup window when
queueing an R2T; in this case the connection will be stuck until the
initiator loses patience and does some task management operation (or kills
the connection entirely).
Fix this by converting to wait_event_interruptible(), which does not
suffer from this sort of race.
Signed-off-by: Roland Dreier <roland@purestorage.com>
Cc: Andy Grover <agrover@redhat.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: stable@vger.kernel.org
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch increases the default for nopin_timeout to 15 seconds (wait
between sending a new NopIN ping) and nopin_response_timeout to 30 seconds
(wait for NopOUT response before failing the connection) in order to avoid
false positives by iSCSI Initiators who are not always able (under load) to
respond to NopIN echo PING requests within the current 5 second window.
False positives have been observed recently using Open-iSCSI code on v3.3.x
with heavy large-block READ workloads over small MTU 1 Gb/sec ports, and
increasing these values to more reasonable defaults significantly reduces
the possibility of false positive NopIN response timeout events under
this specific workload.
Historically these have been set low to initiate connection recovery as
soon as possible if we don't hear a ping back, but for modern v3.x code
on 1 -> 10 Gb/sec ports these new defaults make alot more sense.
Cc: Christoph Hellwig <hch@lst.de>
Cc: Andy Grover <agrover@redhat.com>
Cc: Mike Christie <michaelc@cs.wisc.edu>
Cc: Hannes Reinecke <hare@suse.de>
Cc: stable@vger.kernel.org
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Make sure all on the wire types are use as big endian and big endian only so
that sparse can verify all the conversions are done right.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Use the special itt_t type defined by the iscsi headers and the initiator
to make sure it's an opaque value.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
The iSCSI TMR TASK_REASSIGN completion logic in iscsi_tmr_task_reassign()
does an explict check for MRDSL across task reassignment, so go ahead and
add an explict MaxXmitDataSegmentLength check here as well to be safe.
Cc: Mike Christie <michaelc@cs.wisc.edu>
Cc: Andy Grover <agrover@redhat.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Roland Dreier <roland@purestorage.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch introduces a new per connection MaxXmitDataSegmentLength
parameter value used to represent the outgoing MaxRecvDataSegmentLength
that is actually sent over the wire during iSCSI login response back
to the initiator side.
It also adds a new MaxXmitDataSegmentLength configfs attribute to
represent this value within the existing TPG parameter group under
/sys/kernel/config/target/iscsi/$TARGETNAME/$TPGT/param/
Cc: Mike Christie <michaelc@cs.wisc.edu>
Cc: Andy Grover <agrover@redhat.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Roland Dreier <roland@purestorage.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
From Al Viro:
BTW, speaking of struct file treatment related to sockets -
there's this piece of code in iscsi:
/*
* The SCTP stack needs struct socket->file.
*/
if ((np->np_network_transport == ISCSI_SCTP_TCP) ||
(np->np_network_transport == ISCSI_SCTP_UDP)) {
if (!new_sock->file) {
new_sock->file = kzalloc(
sizeof(struct file), GFP_KERNEL);
For one thing, as far as I can see it'not true - sctp does *not* depend on
socket->file being non-NULL; it does, in one place, check socket->file->f_flags
for O_NONBLOCK, but there it treats NULL socket->file as "flag not set".
Which is the case here anyway - the fake struct file created in
__iscsi_target_login_thread() (and in iscsi_target_setup_login_socket(), with
the same excuse) do *not* get that flag set.
Moreover, it's a bloody serious violation of a bunch of asserts in VFS;
all struct file instances should come from filp_cachep, via get_empty_filp()
(or alloc_file(), which is a wrapper for it). FWIW, I'm very tempted to
do this and be done with the entire mess:
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andy Grover <agrover@redhat.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: stable@vger.kernel.org
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Also remove the unused ref_task_lun field in struct se_tmr_req.
(nab: Add missing TASK_REASSIGN ref_lun vs. ref_cmd orig_fe_lun checks
in iscsit_tmr_task_reassign)
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
We originally changed iscsi to allocate its own buffers just as an
intermediate step to clean up some core buffer allocation mechanisms. Now
we can put it back.
Also had to change allocate_iovecs to use data_length instead of
t_data_nents because iovecs are now allocated before the data buffer, thus
t_data_nents is not yet initialized.
Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Redundant, just use iscsi_cmd->se_cmd.data_length once se_cmd is
initialized, or hdr->data_length before then.
Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch includes the handful of squashed patches for target/iscsi from
Andy's original series into lio-core/master code:
*) Make iscsit_add_reject static
*) Remove unused data_offset_end from iscsi_datain_req
*) Remove "#if 0" stubs
*) Rename iscsi_datain_req to cmd_datain_node
*) Cleanups for built_r2ts_for_cmd()
*) Cleanups for Cleanup build_sendtargets_response()
Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
*) Use decoded cmd->immediate_cmd for conditional instead of
re-examining hdr->opcode
*) Make iscist_dataout_post_crc_passed more legible
*) use max() to reduce code in build_r2ts_for_cmd()
*) Remove CONFIG_SMP and if 0 ifdefs
*) Replace if/goto with a while loop
*) Remove unused conn->tx_immediate_queue and tx_response_queue
Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
The name change makes it clear this list_head is so the cmd can be an
item in the connection's conn_cmd_list.
Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Bump core version to v4.1.0-rc2-ml, and for versions from the
following mainline fabric modules:
loopback: v2.1-rc2
tcm_fc: v0.4
iscsi-target: v4.1.0-rc2
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch addresses a bug with sendtargets discovery where INADDR_ANY (0.0.0.0)
+ IN6ADDR_ANY_INIT ([0:0:0:0:0:0:0:0]) network portals where incorrectly being
reported back to initiators instead of the address of the connecting interface.
To address this, save local socket ->getname() output during iscsi login setup,
and makes iscsit_build_sendtargets_response() return these TargetAddress keys
when INADDR_ANY or IN6ADDR_ANY_INIT portals are in use.
Reported-by: Dax Kelson <dkelson@gurulabs.com>
Reported-by: Andy Grover <agrover@redhat.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: <stable@vger.kernel.org>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed bitfields are a problem because instead of being 1 or 0 like
you'd expect they are 0 and -1. It doesn't cause a problem in this case
but sparse complains:
drivers/target/iscsi/iscsi_target_core.h:564:56: error: dubious one-bit
signed bitfield
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Some are never used, some are set but never read, dev_hoq_count is
incremented and decremented, but never read.
Signed-off-by: Joern Engel <joern@logfs.org>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch fixes iscsi-target handling of underflow where residual data is
causing an OOPs by using the incorrect iscsi_cmd_t->data_length initially
assigned in iscsit_allocate_se_cmd(). It resets iscsi_cmd_t->data_length
from se_cmd_t->data_length after transport_generic_allocate_tasks()
has been invoked in iscsit_handle_scsi_cmd() RX context, and converts
iscsi_cmd->residual_count usage to access iscsi_cmd->se_cmd.residual_count
to get the proper residual count set by target-core.
Reported-by: <lists@internyc.net>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Andy Grover <agrover@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch converts target-core to use se_cmd->t_transport_sent instead of
a duplicated se_cmd->transport_sent member in a handful of locations.
It also updates iscsi_target to properly use ->t_transport_sent instead of
it's own iscsi_cmd_t->transport_sent value that was not being assigned.
Reported-by: Christoph Hellwig <hch@lst.de>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This is a squashed version of the following target_core_base.h
cleanup patches:
target: remove the unused SHUTDOWN_SIGS defintion
target: remove unused se_mem leftovers
target: remove the unused map_func_t typedef
target: move TRANSPORT_IOV_DATA_BUFFER to the iscsi-specific code
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
The Linux-iSCSI.org target module is a full featured in-kernel
software implementation of iSCSI target mode (RFC-3720) for the
current WIP mainline target v4.1 infrastructure code for the v3.1
kernel. More information can be found here:
http://linux-iscsi.org/wiki/ISCSI
This includes support for:
* RFC-3720 defined request / response state machines and support for
all defined iSCSI operation codes from Section 10.2.1.2 using libiscsi
include/scsi/iscsi_proto.h PDU definitions
* Target v4.1 compatible control plane using the generic layout in
target_core_fabric_configfs.c and fabric dependent attributes
within /sys/kernel/config/target/iscsi/ subdirectories.
* Target v4.1 compatible iSCSI statistics based on RFC-4544 (iSCSI MIBS)
* Support for IPv6 and IPv4 network portals in M:N mapping to TPGs
* iSCSI Error Recovery Hierarchy support
* Per iSCSI connection RX/TX thread pair scheduling affinity
* crc32c + crc32c_intel SSEv4 instruction offload support using libcrypto
* CHAP Authentication support using libcrypto
* Conversion to use internal SGl allocation with iscsit_alloc_buffs() ->
transport_generic_map_mem_to_cmd()
(nab: Fix iscsi_proto.h struct scsi_lun usage from linux-next in commit:
iscsi: Use struct scsi_lun in iscsi structs instead of u8[8])
(nab: Fix 32-bit compile warnings)
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Andy Grover <agrover@redhat.com>
Acked-by: Roland Dreier <roland@kernel.org>
Signed-off-by: Nicholas A. Bellinger <nab@linux-iscsi.org>
Nicholas Bellinger