Singal completion of every 32 scsi commands and suppress all the rest.
We don't do anything upon getting the completion so no need to "just
consume" it. Cleanup of scsi command is done in cleanup_task callback.
Still keep dataout and control send completions as we may need to
cleanup there. This helps reducing the amount of interrupts/completions
in the IO path.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Poll in batch of 16. Since we don't want it on the stack, keep under
iser completion context (iser_comp).
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Avoid post_send counting (atomic) in the IO path just to keep track of
how many completions we need to consume. Use a beacon post to indicate
that all prior posts completed.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
This will solve a possible condition where we might miss TX completion
(flush error) during session teardown. Since we are using a single
CQ, we don't need to actively drain the TX CQ, instead just wait for
flush_completion (when counters reach zero) and remove iser_poll_for_flush_errors().
This patch might introduce a minor performance regression on its own,
but the next patches will enhance performance using a single CQ for RX
and TX.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Introduce iser_comp which centralizes all iser completion related
items and is referenced by iser_device and each ib_conn.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
iscsi daemon is in user-space, thus we can't rely on it to be invoked
at connection teardown (if not running or does not receive CPU time).
This patch addresses the issue by re-structuring iSER connection
teardown logic and CM events handling.
The CM events will dictate the RDMA resources destruction (ib_conn)
and iser_conn is kept around as long as iscsi_conn is left around
allowing iscsi/iser callbacks to continue after RDMA transport was
destroyed.
This patch introduces a separation in logic when handling CM events:
- DISCONNECTED_HANDLER, ADDR_CHANGED
This events indicate the start of teardown process.
Actions:
1. Terminate the connection: rdma_disconnect (send DREQ/DREP)
2. Notify iSCSI of connection failure
3. Change state to TERMINATING
4. Poll for all flush errors to be consumed
- TIMEWAIT_EXIT, DEVICE_REMOVAL
These events indicate the final stage of termination process and
we can free RDMA related resources.
Actions:
1. Call disconnected handler (we are not guaranteed that DISCONNECTED
event was invoked in the past)
2. Cleanup RDMA related resources
3. For DEVICE_REMOVAL return non-zero rc from cma_handler to
implicitly destroy the cm_id (Can't rely on user-space, make sure
we have forward progress)
We replace flush_completion (indicate all flushes were consumed) with
ib_completion (rdma resources were cleaned up).
The iser_conn_release_work will wait for teardown completions:
- conn_stop was completed (tasks were cleaned-up) - stop_completion
- RDMA resources were destroyed - ib_completion
And then will continue to free iser connection representation (iser_conn).
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Ariel Nahum <arieln@mellanox.com>
Signed-off-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Structure that describes the RDMA relates connection objects. Static
member of iser_conn.
This patch does not change any functionality
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Two reasons why we choose to do this:
1. No point today calling struct iser_conn by another name ib_conn
2. In the next patches we will restructure iser control plane representation
- struct iser_conn: connection logical representation
- struct ib_conn: connection RDMA layout representation
This patch does not change any functionality.
Signed-off-by: Ariel Nahum <arieln@mellanox.com>
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Instead of waiting for events and condition changes of the iser
connection state, we wait for explicit completion of connection
establishment and teardown.
Separate connection establishment wait object from the teardown object
to avoid a situation where racing connection establishment and
teardown may concurrently wakeup each other.
ep_poll will wait for up_completion invoked by
iser_connected_handler() and iser release worker will wait for
flush_completion before releasing the connection.
Bound the completion wait with a 30 seconds timeout for cases where
iscsid (the user space iscsi daemon) is too slow or gone.
Signed-off-by: Ariel Nahum <arieln@mellanox.com>
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
The iser connection state lookups and transitions are not fully protected.
Some transitions are protected with a spinlock, and in some cases the
state is accessed unprotected due to specific assumptions of the flow.
Introduce a new mutex to protect the connection state access. We use a
mutex since we need to also include a scheduling operations executed
under the state lock.
Each state transition/condition and its corresponding action will be
protected with the state mutex.
The rdma_cm events handler acquires the mutex when handling connection
events. Since iser connection state can transition to DOWN
concurrently during connection establishment, we bailout from
addr/route resolution events when the state is not PENDING.
This addresses a scenario where ep_poll retries expire during CMA
connection establishment. In this case ep_disconnect is invoked while
CMA events keep coming (address/route resolution, connected, etc...).
Signed-off-by: Ariel Nahum <arieln@mellanox.com>
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Replace struct sockaddr_in with struct sockaddr which supports both
IPv4 and IPv6, and print using the %pIS format directive.
Signed-off-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
iSER relies on refcounting to manage iser connections establishment
and teardown.
Following commit 39ff05dbbb ("IB/iser: Enhance disconnection logic
for multi-pathing"), iser connection maintain 3 references:
- iscsi_endpoint (at creation stage)
- cma_id (at connection request stage)
- iscsi_conn (at bind stage)
We can avoid taking explicit refcounts by correctly serializing iser
teardown flows (graceful and non-graceful).
Our approach is to trigger a scheduled work to handle ordered teardown
by gracefully waiting for 2 cleanup stages to complete:
1. Cleanup of live pending tasks indicated by iscsi_conn_stop completion
2. Flush errors processing
Each completed stage will notify a waiting worker thread when it is
done to allow teardwon continuation.
Since iSCSI connection establishment may trigger endpoint disconnect
without a successful endpoint connect, we rely on the iscsi <-> iser
binding (.conn_bind) to learn about the teardown policy we should take
wrt cleanup stages.
Since all cleanup worker threads are scheduled (release_wq) in
.ep_disconnect it is safe to assume that when module_exit is called,
all cleanup workers are already scheduled. Thus proper module unload
shall flush all scheduled works before allowing safe exit, to
guarantee no resources got left behind.
Signed-off-by: Ariel Nahum <arieln@mellanox.com>
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Update Mellanox copyrights for 2014 on the iser initiator driver.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
The iscsi stack has existing mechanisms to link back and forth between
the iscsi connection and the iscsi transport (e.g iser/tcp) connection.
This is done through a dd_data pointer field in struct iscsi_conn
which can be set to point to the transport connection, etc.
The iscsi_iser_conn structure was used to get this linking done in
another way, which is uneeded and adds extra complication to the iser
code, so we just remove it.
Signed-off-by: Ariel Nahum <arieln@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Once the iSCSI transaction is completed we must implement
check_protection in order to notify on DIF errors that may have
occured.
The routine boils down to calling ib_check_mr_status to get the
signature status of the transaction.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Alex Tabachnik <alext@mellanox.com>
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Add logic to initialize protection information entities. Upon each
iSCSI task, we keep the scsi_cmnd in order to query the scsi
protection operations and reference to protection buffers.
Modify iser_fast_reg_mr to receive indication whether it is
registering the data or protection buffers.
In addition introduce iser_reg_sig_mr which performs fast registration
work-request for a signature enabled memory region
(IB_WR_REG_SIG_MR). In this routine we set all the protection
relevants for the device to offload protection data-transfer and
verification.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Alex Tabachnik <alext@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
During connection establishment we also initialize T10-PI resources
(QP, PI contexts) in order to support SCSI's protection operations.
Signed-off-by: Alex Tabachnik <alext@mellanox.com>
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Use modparams to activate protection information support.
pi_enable bool: Based on this parameter iSER will know if it should
support T10-PI. We don't want to do this by default as it requires to
allocate and initialize extra resources. In case pi_enable=N, iSER
won't publish to SCSI midlayer any DIF capabilities.
pi_guard int: Based on this parameter iSER will publish DIX guard type
support to SCSI midlayer. 0 means CRC is allowed to be passed in DIX
buffers, 1 (or non-zero) means IP-CSUM is allowed to be passed in DIX
buffers. Note that over the wire, only CRC is allowed.
In the next phase, it is worth considering passing these parameters
from iscsid via nlmsg. This will allow these parameters to be
connection based rather than global.
Signed-off-by: Alex Tabachnik <alext@mellanox.com>
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
This routines operates on data buffers and may also work with
protection infomation buffers. So we generalize them to handle an
iser_data_buf which can be the command data or command protection
information.
This patch does not change any functionality.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
In T10-PI support we will have memory keys for protection buffers and
signature transactions. We prefer to compact indicators rather than
keeping multiple bools.
This commit does not change any functionality.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Alex Tabachnik <alext@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
For T10-PI offload support, we will need to know the device signature
offload capability upon every connection establishment.
This patch does not change any functionality.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Alex Tabachnik <alext@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
FRWR stands for "fast registration work request". We want to avoid
calling the fastreg pool with that name, instead we name it fastreg
which stands for "fast registration".
This pool will include more elements in the future, so it is a good
idea to generalize the name.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Alex Tabachnik <alext@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
In case iSER uses fast registration method, it should not request for
successful completions on fast registration nor local invalidate
requests. We color wr_id with ISER_FRWR_LI_WRID in order to correctly
consume error completions.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Newer HCAs and Virtual functions may not support FMRs but rather a fast
registration model, which we call FRWR - "Fast Registration Work Requests".
This model was introduced in 00f7ec36c ("RDMA/core: Add memory management
extensions support") and works when the IB device supports the
IB_DEVICE_MEM_MGT_EXTENSIONS capability.
Upon creating the iser device iser will test whether the HCA supports
FMRs. If no support for FMRs, check if IB_DEVICE_MEM_MGT_EXTENSIONS
is supported and assign function pointers that handle fast
registration and allocation of appropriate resources (fast_reg
descriptors).
Registration is done using posting IB_WR_FAST_REG_MR to the QP and
invalidations using posting IB_WR_LOCAL_INV.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
This is preparation step for other memory registration methods to be
added. In addition, change reg/unreg routines signature to indicate
they use FMRs.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Currently the driver uses FMRs as the only means to register the
memory pointed by SG provided by the SCSI mid-layer with the RDMA
device.
As preparation step for adding more methods for fast path memory
registration, make the alloc/free and reg/unreg calls function
pointers, which are for now just set to the existing FMR ones.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Use cmds_max passed from user space to be the number of PDUs to be
supported for the session instead of hard-coded ISCSI_DEF_XMIT_CMDS_MAX.
This allow controlling the max number of SCSI commands for the session.
Also don't ignore the qdepth passed from user space.
Derive from session->cmds_max the actual number of RX buffers and FMR
pool size to allocate during the connection bind phase.
Since the iser transport connection is established before the iscsi
session/connection are created and bound, we still use one hard-coded
quantity ISER_DEF_XMIT_CMDS_MAX to compute the maximum number of
work-requests to be supported by the RC QP used for the connection.
The above quantity is made to be a power of two between ISCSI_TOTAL_CMDS_MIN
(16) and ISER_DEF_XMIT_CMDS_MAX (512) inclusive.
Signed-off-by: Shlomo Pongratz <shlomop@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
This is a preparation step to a patch that accepts the number of max
SCSI commands to be supported a session from user space iSCSI tools.
Move the allocation of the login buffer, FMR pool and its associated
page vector from iser_create_ib_conn_res() (which is called prior when
we actually know how many commands should be supported) to
iser_alloc_rx_descriptors() (which is called during the iscsi
connection bind step where this quantity is known).
Also do small refactoring around the deallocation to make that path
similar to the allocation one.
Signed-off-by: Shlomo Pongratz <shlomop@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Commit 4f36388261 ("IB/iser: Move informational messages from error
to info level") set info prints to be emitted at a lower debug level
than warning prints, which is a bit odd. Fix that.
Also move the prints on unaligned SG from warning to debug level.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Add Mellanox copyright to the iser initiator source code which I maintain.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Annex A12 of the IBTA spec defines additional information that needs
to be provided through the CM exchange relating to usage of ZBVA (Zero
Based VAs) and Send With Invalidate over an iSER connection.
Currently, the initiator sets both to not supported, but does provide
the header so that existing iSER targets can be patched to start
looking on the private data carried by the CM.
This is a preparation step to enable iSER with HW drivers for which
FMRs are not supported, such as mlx4 VF instances or new HW devices
which might support only FRWR (Fast Registration Work-Requests) along
the details of the IB_DEVICE_MEM_MGT_EXTENSIONS device capability.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Introduce iser_info() and move informational messages that were
printed as errors to use that macro. Also, cleanup printk leftovers to
use the existing macros.
Signed-off-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
[ Use pr_warn(... instead of printk(KERN_WARNING .... - Roland ]
Signed-off-by: Roland Dreier <roland@purestorage.com>
Add displaying module version, update the version to 1.1,
and remove the DRV_DATE define.
Signed-off-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
ISER_DEF_CMD_PER_LUN was meant to be ISCSI_DEF_XMIT_CMDS_MAX, not plain 128
Signed-off-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
RX/TX CQs will now be selected from a per HCA pool. For the RX flow
this has the effect of using different interrupt vectors when using
low level drivers (such as mlx4) that map the "vector" param provided
by the ULP on CQ creation to a dedicated IRQ/MSI-X vector. This
allows the RX flow processing of IO responses to be distributed across
multiple CPUs.
QPs (--> iSER sessions) are assigned to CQs in round robin order using
the CQ with the minimum number of sessions attached to it.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Alex Tabachnik <alext@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
An iser target may send iscsi NO-OP PDUs as soon as it marks the iSER
iSCSI session as fully operative. This means that there is window
where there are no posted receive buffers on the initiator side, so
it's possible for the iSER RC connection to break because of RNR NAK /
retry errors. To fix this, rely on the flags bits in the login
request to have FFP (0x3) in the lower nibble as a marker for the
final login request, and post an initial chunk of receive buffers
before sending that login request instead of after getting the login
response.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Roland Dreier <roland@purestorage.com>
The current driver never does DMA unmapping on these buffers. Fix that
by adding DMA unmapping to the task cleanup callback, and DMA mapping to
the task init function (drop the headers_initialized micro-optimization).
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
The driver counted on the transactional nature of iSCSI login/text
flows and used the same buffer for both the request and the response.
We also went further and did DMA mapping only once, with
DMA_FROM_DEVICE, which violates the DMA mapping API. Fix that by
using different buffers, one for requests and one for responses, and
use the correct DMA mapping direction for each.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
The code that prepares the SG associated with SCSI command for FMR was
buggy for systems with DMA addresses that don't fit in unsigned long,
e.g under the 32-bit based XenServer dom0 sizeof(dma_addr_t) is 8.
Fix that by casting to unsigned long long a masking constant used by
the code. This resolves a crash in iser_sg_to_page_vec on this system.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il>
Signed-off-by: Roland Dreier <roland@purestorage.com>
* remove interrupt.g inclusion from netdevice.h -- not needed
* fixup fallout, add interrupt.h and hardirq.h back where needed.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The iser connection teardown flow isn't over until the underlying
Connection Manager (e.g the IB CM) delivers a disconnected or timeout
event through the RDMA-CM. When the remote (target) side isn't
reachable, e.g when some HW e.g port/hca/switch isn't functioning or
taken down administratively, the CM timeout flow is used and the event
may be generated only after relatively long time -- on the order of
tens of seconds.
The current iser code exposes this possibly long delay to higher
layers, specifically to the iscsid daemon and iscsi kernel stack. As a
result, the iscsi stack doesn't respond well: this low-level CM delay
is added to the fail-over time under HA schemes such as the one
provided by DM multipath through the multipathd(8) service.
This patch enhances the reference counting scheme on iser's IB
connections so that the disconnect flow initiated by iscsid from user
space (ep_disconnect) doesn't wait for the CM to deliver the
disconnect/timeout event. (The connection teardown isn't done from
iser's view point until the event is delivered)
The iser ib (rdma) connection object is destroyed when its reference
count reaches zero. When this happens on the RDMA-CM callback
context, extra care is taken so that the RDMA-CM does the actual
destroying of the associated ID, since doing it in the callback is
prohibited.
The reference count of iser ib connection normally reaches three,
where the <ref, deref> relations are
1. conn <init, terminate>
2. conn <bind, stop/destroy>
3. cma id <create, disconnect/error/timeout callbacks>
With this patch, multipath fail-over time is about 30 seconds, while
without this patch, multipath fail-over time is about 130 seconds.
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add handler to handle events such as port up and down. This is useful
when testing high-availability schemes such as multi-pathing.
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Remove unnecessary checks for the IB connection state and for QP
overflow, as conn state changes are reported by iSER to libiscsi and
handled there. QP overflow is theoretically possible only when
unsolicited data-outs are used; anyway it's being checked and handled
by HW drivers.
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Simplify and shrink the logic/code used for the send descriptors.
Changes include removing struct iser_dto (an unnecessary abstraction),
using struct iser_regd_buf only for handling SCSI commands, using
dma_sync instead of dma_map/unmap, etc.
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
