Michael Kelley [Fri, 4 Jun 2021 17:21:02 +0000 (10:21 -0700)]
scsi: storvsc: Update error logging
When an I/O error is reported by the underlying Hyper-V host, current code
provides details only when the logging level is set to WARN, making it more
difficult to diagnose problems in live customer situations. Fix this by
reporting details at ERROR level, which is the default. Also add more
information, including the Hyper-V error code, and the tag # so that the
message can be matched with messages at the SCSI and blk-mq levels.
Also, sense information logging is inconsistent and duplicative. The
existence of sense info is first logged at WARN level, and then full sense
info is logged at ERROR level. Fix this by removing the logging of the
existence of sense info, and change the logging of full sense info to WARN
level in favor of letting the generic SCSI layer handle such logging. With
the change to WARN level, it's no longer necessary to filter out as noise
any NOT READY sense info generated by the virtual DVD device.
Implement CT commands for registering and deregistering the appid for the
application. Also, a small change in decrementing the ndlp ref counter has
been added.
Implement routines to save, retrieve, and remove the VMIDs from the data
structure. A hash table is used to save the VMIDs and the corresponding
UUIDs associated with the application/VMs.
scsi: lpfc: vmid: Add support for VMID in mailbox command
Add supporting datastructures for mailbox command which helps in
determining if the firmware supports appid. Allocate resources for VMID at
initialization time and clean them up on removal.
scsi: lpfc: vmid: Add datastructure for supporting VMID in lpfc
Add the primary datastructures needed to implement VMID in the lpfc
driver. Maintain the capability, current state, and hash table for the
vmid/appid along with other information. This implementation supports the
two versions of vmid implementation (app header and priority tagging).
Muneendra Kumar [Tue, 8 Jun 2021 04:35:45 +0000 (10:05 +0530)]
scsi: blkcg: Add app identifier support for blkcg
Add a unique application identifier (i.e fc_app_id member) in blkcg. This
allows identification of traffic belonging to an specific both on the host
and in the fabric infrastructure. As an example, this allows the storage
stack to uniquely identify traffic belong to particular virtual machine.
Muneendra Kumar [Tue, 8 Jun 2021 04:35:44 +0000 (10:05 +0530)]
scsi: cgroup: Add cgroup_get_from_id()
Add a new function, cgroup_get_from_id(), to retrieve the cgroup associated
with a cgroup id. Also export the function cgroup_get_e_css() as this is
needed in blk-cgroup.h.
Javed Hasan [Thu, 3 Jun 2021 12:16:23 +0000 (05:16 -0700)]
scsi: fc: FDMI enhancement
Added RHBA and RPA attributes type and length.
As per FC_GC_7 document section "Table 400 – Attribute Entry Types and
associated Values" ASCII type attributes length can be vary from "4 to 256
byte". If we keep all RHBA ASCII attributes length 256 then total length
is going upto 2750, which is far more than 2048 (max frame size).
In libfc we do have logic to split FCP commands but not for CT commands.
Practically all version/names get covered with in 64 bytes except OS name,
for that we need 128 bytes. Hence length of all RBHA ASCII attributes
is reduced to 64 bytes and 128 bytes in case of OS name.
Javed Hasan [Thu, 3 Jun 2021 12:16:21 +0000 (05:16 -0700)]
scsi: libfc: Add FDMI-2 attributes
Add all attributes for RHBA and RPA registration.
Fallback mechanism is added between RBHA V2 and RHBA V1 attributes. In case
RHBA get fails for V2 attributes we fall back to V1 attribute registration.
Luo Jiaxing [Mon, 7 Jun 2021 09:29:39 +0000 (17:29 +0800)]
scsi: hisi_sas: Speed up error handling when internal abort timeout occurs
If an internal task abort timeout occurs, the controller has developed a
fault, and needs to be reset to be recovered.
When this occurs during error handling, the current policy is to allow
error handling to continue, and the inevitable nexus ha reset will handle
the required reset.
However various steps of error handling need to taken before this happens.
These also involve some level of HW interaction, which will also fail with
various timeouts.
Speed up this process by recording a HW fault bit for an internal abort
timeout - when this is set, just automatically error any HW interaction,
and essentially go straight to clear nexus ha (to reset the controller).
Luo Jiaxing [Mon, 7 Jun 2021 09:29:38 +0000 (17:29 +0800)]
scsi: hisi_sas: Reset controller for internal abort timeout
If an internal task abort timeout occurs, the controller has developed a
fault, and needs to be reset to be recovered. However if a timeout occurs
during SCSI error handling, issuing a controller reset immediately may
conflict with the error handling.
To handle internal abort in these two scenarios, only queue the reset when
not in an error handling function. In the case of a timeout during error
handling, do nothing and rely on the inevitable ha nexus reset to reset the
controller.
Luo Jiaxing [Mon, 7 Jun 2021 09:29:36 +0000 (17:29 +0800)]
scsi: hisi_sas: Run I_T nexus resets in parallel for clear nexus reset
For a clear nexus reset operation, the I_T nexus resets are executed
serially for each device. For devices attached through an expander, this
may take 2s per device; so, in total, could take a long time.
Reduce the total time by running the I_T nexus resets in parallel through
async operations.
Mike Christie [Wed, 9 Jun 2021 19:27:09 +0000 (14:27 -0500)]
scsi: qedi: Fix host removal with running sessions
qedi_clear_session_ctx() could race with the in-kernel or userspace driven
recovery/removal and we could access a NULL conn or do a double free.
We should be using iscsi_host_remove() to start the removal process from
the driver. It will start the in-kernel recovery and notify userspace that
the driver's scsi_hosts are being removed. iscsid will then drive the
session removal like is done when the logout command is run. When the
sessions are removed, iscsi_host_remove() will return so qedi can finish
knowing there are no running sessions and no new sessions will be allowed.
This also fixes an issue where we check for a NULL conn after already
accessing it introduced in commit 27e986289e73 ("scsi: iscsi: Drop suspend
calls from ep_disconnect") by just removing the function completely.
Dan Carpenter [Wed, 9 Jun 2021 09:27:14 +0000 (12:27 +0300)]
scsi: mpi3mr: Fix error handling in mpi3mr_setup_isr()
The pci_alloc_irq_vectors_affinity() function returns negative error codes
or it returns a number between the minimum vectors (1 in this case) and
max_vectors. It won't return zero. Because "i" is a u16 then the error
handling won't work. And also if it did work the error code was not set.
Really "max_vectors" can be an int as well because we're doing a min_t() on
int type. The other change is that it's better to remove unnecessary
initialization so that static checkers can warn us if there are ever
uninitialized variable bugs introduced in the future.
I changed the error code from -1 (-EPERM) if the kmalloc() failed to
-ENOMEM. And on success path I changed it from "return retval;" to "return
0;" which shouldn't affect the compiled code but makes it more readable.
Dan Carpenter [Wed, 9 Jun 2021 09:26:02 +0000 (12:26 +0300)]
scsi: mpi3mr: Delete unnecessary NULL check
The "mrioc->intr_info" pointer can't be NULL, but if it could then the
second iteration through the loop would Oops. Let's delete the confusing
and impossible NULL check.
Tomas Henzl [Tue, 8 Jun 2021 14:57:12 +0000 (16:57 +0200)]
scsi: mpi3mr: Fix a double free
Fix a double free, scsi_tgt_priv_data will be freed in
mpi3mr_target_destroy() so remove the kfree() from mpi3mr_target_alloc().
I've also removed few unneeded initialisations.
Bean Huo [Mon, 31 May 2021 10:43:08 +0000 (12:43 +0200)]
scsi: ufs: core: Use UPIU query trace in devman_upiu_cmd()
Since devman_upiu_cmd() is not COMMAND UPIU, and doesn't have CDB, it is
better to use UPIU query trace, which provides more helpful information for
issue troubleshooting.
Bean Huo [Mon, 31 May 2021 10:43:07 +0000 (12:43 +0200)]
scsi: ufs: core: Capture command trace only for the cmd != NULL case
For the query request, we already have query_trace, but in
ufshcd_send_command(), there will add two more redundant traces. Since
lrbp->cmd is NULL in the query request, the two trace events below provide
nothing except the tag and DB. Instead of letting them take up the limited
trace ring buffer, it’s better not to print these traces in case of cmd ==
NULL.
The current UPIU completion event trace still prints the COMMAND UPIU
header, rather than the RSP UPIU header. This makes UPIU command trace
useless in problem shooting in case we receive a trace log from the
customer/field.
There are two important fields in RSP UPIU:
1. The response field, which indicates the UFS defined overall success or
failure of the series of Command, Data and RESPONSE UPIU’s that make up
the execution of a task.
2. The Status field, which contains the command set specific status for a
specific command issued by the initiator device.
In preparation to enable -Wimplicit-fallthrough for Clang, fix a
fall-through warning by explicitly adding a break statement instead of just
letting the code fall through to the next case.
In preparation to enable -Wimplicit-fallthrough for Clang, fix a
fall-through warning by replacing a /* fallthrough */ comment with the new
pseudo-keyword macro fallthrough;
Can Guo [Mon, 24 May 2021 08:36:58 +0000 (01:36 -0700)]
scsi: ufs: Utilize Transfer Request List Completion Notification Register
By reading the UTP Transfer Request List Completion Notification Register,
which is added in UFSHCI Ver 3.0, SW can easily get the compeleted transfer
requests. Thus, SW can get rid of host lock, which is used to synchronize
the tr_doorbell and outstanding_reqs, on transfer requests dispatch and
completion paths. This can further benefit random read/write performance.
Can Guo [Mon, 24 May 2021 08:36:57 +0000 (01:36 -0700)]
scsi: ufs: Optimize host lock on transfer requests send/compl paths
Current UFS IRQ handler is completely wrapped by host lock, and because
ufshcd_send_command() is also protected by host lock, when IRQ handler
fires, not only the CPU running the IRQ handler cannot send new requests,
the rest CPUs can neither. Move the host lock wrapping the IRQ handler into
specific branches, i.e., ufshcd_uic_cmd_compl(), ufshcd_check_errors(),
ufshcd_tmc_handler() and ufshcd_transfer_req_compl(). Meanwhile, to further
reduce occpuation of host lock in ufshcd_transfer_req_compl(), host lock is
no longer required to call __ufshcd_transfer_req_compl(). As per test, the
optimization can bring considerable gain to random read/write performance.
Can Guo [Mon, 24 May 2021 08:36:56 +0000 (01:36 -0700)]
scsi: ufs: Remove a redundant command completion logic in error handler
ufshcd_host_reset_and_restore() anyways completes all pending requests
before starts re-probing, so there is no need to complete the command on
the highest bit in tr_doorbell in advance.
Kees Cook [Wed, 2 Jun 2021 18:00:00 +0000 (11:00 -0700)]
scsi: fcoe: Statically initialize flogi_maddr
In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memcpy() avoid using an inline const buffer
argument and instead just statically initialize the destination array
directly.
Saurav Kashyap [Wed, 2 Jun 2021 10:46:53 +0000 (03:46 -0700)]
scsi: qedf: Update the max_id value in host structure
host->max_id defines the maximum target id that the SCSI midlayer will
attempt to manually scan. The default is 8. Update the value to the max
sessions the driver supports.
Bart Van Assche [Mon, 24 May 2021 02:54:55 +0000 (19:54 -0700)]
scsi: libsas: Introduce more SAM status code aliases in enum exec_status
This patch prepares for converting SAM status codes into an enum. Without
this patch converting SAM status codes into an enumeration type would
trigger complaints about enum type mismatches for the SAS code.
Mike Christie [Tue, 25 May 2021 18:18:21 +0000 (13:18 -0500)]
scsi: qedi: Wake up if cmd_cleanup_req is set
If we got a response then we should always wake up the conn. For both the
cmd_cleanup_req == 0 or cmd_cleanup_req > 0, we shouldn't dig into
iscsi_itt_to_task because we don't know what the upper layers are doing.
We can also remove the qedi_clear_task_idx call here because once we signal
success libiscsi will loop over the affected commands and end up calling
the cleanup_task callout which will release it.
Mike Christie [Tue, 25 May 2021 18:18:20 +0000 (13:18 -0500)]
scsi: qedi: Complete TMF works before disconnect
We need to make sure that abort and reset completion work has completed
before ep_disconnect returns. After ep_disconnect we can't manipulate
cmds because libiscsi will call conn_stop and take onwership.
We are trying to make sure abort work and reset completion work has
completed before we do the cmd clean up in ep_disconnect. The problem is
that:
1. the work function sets the QEDI_CONN_FW_CLEANUP bit, so if the work was
still pending we would not see the bit set. We need to do this before
the work is queued.
2. If we had multiple works queued then we could break from the loop in
qedi_ep_disconnect early because when abort work 1 completes it could
clear QEDI_CONN_FW_CLEANUP. qedi_ep_disconnect could then see that
before work 2 has run.
3. A TMF reset completion work could run after ep_disconnect starts
cleaning up cmds via qedi_clearsq. ep_disconnect's call to qedi_clearsq
-> qedi_cleanup_all_io would might think it's done cleaning up cmds,
but the reset completion work could still be running. We then return
from ep_disconnect while still doing cleanup.
This replaces the bit with a counter to track the number of queued TMF
works, and adds a bool to prevent new works from starting from the
completion path once a ep_disconnect starts.
Mike Christie [Tue, 25 May 2021 18:18:18 +0000 (13:18 -0500)]
scsi: qedi: Fix cleanup session block/unblock use
Drivers shouldn't be calling block/unblock session for cmd cleanup because
the functions can change the session state from under libiscsi. This adds
a new a driver level bit so it can block all I/O the host while it drains
the card.
Mike Christie [Tue, 25 May 2021 18:18:17 +0000 (13:18 -0500)]
scsi: qedi: Fix TMF session block/unblock use
Drivers shouldn't be calling block/unblock session for tmf handling because
the functions can change the session state from under libiscsi.
iscsi_queuecommand's call to iscsi_prep_scsi_cmd_pdu->
iscsi_check_tmf_restrictions will prevent new cmds from being sent to qedi
after we've started handling a TMF. So we don't need to try and block it in
the driver, and we can remove these block calls.
Mike Christie [Tue, 25 May 2021 18:18:15 +0000 (13:18 -0500)]
scsi: qedi: Fix TMF tid allocation
qedi_iscsi_abort_work and qedi_tmf_work both allocate a tid then call
qedi_send_iscsi_tmf which also allocates a tid. This removes the tid
allocation from the callers.
Mike Christie [Tue, 25 May 2021 18:18:14 +0000 (13:18 -0500)]
scsi: qedi: Fix use after free during abort cleanup
If qedi_tmf_work's qedi_wait_for_cleanup_request call times out we will
also force the clean up of the qedi_work_map but
qedi_process_cmd_cleanup_resp could still be accessing the qedi_cmd.
To fix this issue we extend where we hold the tmf_work_lock and back_lock
so the qedi_process_cmd_cleanup_resp access is serialized with the cleanup
done in qedi_tmf_work and any completion handling for the iscsi_task.
Mike Christie [Tue, 25 May 2021 18:18:13 +0000 (13:18 -0500)]
scsi: qedi: Fix race during abort timeouts
If the SCSI cmd completes after qedi_tmf_work calls iscsi_itt_to_task then
the qedi qedi_cmd->task_id could be freed and used for another cmd. If we
then call qedi_iscsi_cleanup_task with that task_id we will be cleaning up
the wrong cmd.
Wait to release the task_id until the last put has been done on the
iscsi_task. Because libiscsi grabs a ref to the task when sending the
abort, we know that for the non-abort timeout case that the task_id we are
referencing is for the cmd that was supposed to be aborted.
A latter commit will fix the case where the abort times out while we are
running qedi_tmf_work.
Mike Christie [Tue, 25 May 2021 18:18:12 +0000 (13:18 -0500)]
scsi: qedi: Fix null ref during abort handling
If qedi_process_cmd_cleanup_resp finds the cmd it frees the work and sets
list_tmf_work to NULL, so qedi_tmf_work should check if list_tmf_work is
non-NULL when it wants to force cleanup.
Mike Christie [Tue, 25 May 2021 18:18:11 +0000 (13:18 -0500)]
scsi: iscsi: Move pool freeing
This doesn't fix any bugs, but it makes more sense to free the pool after
we have removed the session. At that time we know nothing is touching any
of the session fields, because all devices have been removed and scans are
stopped.
Mike Christie [Tue, 25 May 2021 18:18:10 +0000 (13:18 -0500)]
scsi: iscsi: Hold task ref during TMF timeout handling
For aborts, qedi needs to cleanup the FW then send the TMF from a worker
thread. While it's doing these the cmd could complete normally and the TMF
could time out. libiscsi would then complete the iscsi_task which will call
into the driver to cleanup the driver level resources while it still might
be accessing them for the cleanup/abort.
This has iscsi_eh_abort keep the iscsi_task ref if the TMF times out, so
qedi does not have to worry about if the task is being freed while in use
and does not need to get its own ref.
Mike Christie [Tue, 25 May 2021 18:18:09 +0000 (13:18 -0500)]
scsi: iscsi: Flush block work before unblock
We set the max_active iSCSI EH works to 1, so all work is going to execute
in order by default. However, userspace can now override this in sysfs. If
max_active > 1, we can end up with the block_work on CPU1 and
iscsi_unblock_session running the unblock_work on CPU2 and the session and
target/device state will end up out of sync with each other.
This adds a flush of the block_work in iscsi_unblock_session.
Mike Christie [Tue, 25 May 2021 18:18:07 +0000 (13:18 -0500)]
scsi: iscsi: Fix shost->max_id use
The iscsi offload drivers are setting the shost->max_id to the max number
of sessions they support. The problem is that max_id is not the max number
of targets but the highest identifier the targets can have. To use it to
limit the number of targets we need to set it to max sessions - 1, or we
can end up with a session we might not have preallocated resources for.
Mike Christie [Tue, 25 May 2021 18:18:06 +0000 (13:18 -0500)]
scsi: iscsi: Fix conn use after free during resets
If we haven't done a unbind target call we can race where
iscsi_conn_teardown wakes up the EH thread and then frees the conn while
those threads are still accessing the conn ehwait.
We can only do one TMF per session so this just moves the TMF fields from
the conn to the session. We can then rely on the
iscsi_session_teardown->iscsi_remove_session->__iscsi_unbind_session call
to remove the target and it's devices, and know after that point there is
no device or scsi-ml callout trying to access the session.
Mike Christie [Tue, 25 May 2021 18:18:05 +0000 (13:18 -0500)]
scsi: iscsi: Get ref to conn during reset handling
The comment in iscsi_eh_session_reset is wrong and we don't wait for the
EH to complete before tearing down the conn. This has us get a ref to the
conn when we are not holding the eh_mutex/frwd_lock so it does not get
freed from under us.
Mike Christie [Tue, 25 May 2021 18:18:04 +0000 (13:18 -0500)]
scsi: iscsi: Have abort handler get ref to conn
If SCSI midlayer is aborting a task when we are tearing down the conn we
could free the conn while the abort thread is accessing the conn. This has
the abort handler get a ref to the conn so it won't be freed from under it.
Note: this is not needed for device/target reset because we are holding the
eh_mutex when accessing the conn.
Mike Christie [Tue, 25 May 2021 18:18:03 +0000 (13:18 -0500)]
scsi: iscsi: Add iscsi_cls_conn refcount helpers
There are a couple places where we could free the iscsi_cls_conn while it's
still in use. This adds some helpers to get/put a refcount on the struct
and converts an exiting user. Subsequent commits will then use the helpers
to fix 2 bugs in the eh code.
Mike Christie [Tue, 25 May 2021 18:18:01 +0000 (13:18 -0500)]
scsi: iscsi: iscsi_tcp: Set no linger
Userspace (open-iscsi based tools at least) sets no linger on the socket to
prevent stale data from being sent. However, with the in-kernel cleanup if
userspace is not up the sockfd_put will release the socket without having
set that sockopt.
iscsid sets that opt at socket close time, but it seems ok to set this at
setup time in the kernel for all tools.
Mike Christie [Tue, 25 May 2021 18:18:00 +0000 (13:18 -0500)]
scsi: iscsi: Fix in-kernel conn failure handling
Commit 0ab710458da1 ("scsi: iscsi: Perform connection failure entirely in
kernel space") has the following regressions/bugs that this patch fixes:
1. It can return cmds to upper layers like dm-multipath where that can
retry them. After they are successful the fs/app can send new I/O to the
same sectors, but we've left the cmds running in FW or in the net layer.
We need to be calling ep_disconnect if userspace is not up.
This patch only fixes the issue for offload drivers. iscsi_tcp will be
fixed in separate commit because it doesn't have a ep_disconnect call.
2. The drivers that implement ep_disconnect expect that it's called before
conn_stop. Besides crashes, if the cleanup_task callout is called before
ep_disconnect it might free up driver/card resources for session1 then they
could be allocated for session2. But because the driver's ep_disconnect is
not called it has not cleaned up the firmware so the card is still using
the resources for the original cmd.
3. The stop_conn_work_fn can run after userspace has done its recovery and
we are happily using the session. We will then end up with various bugs
depending on what is going on at the time.
We may also run stop_conn_work_fn late after userspace has called stop_conn
and ep_disconnect and is now going to call start/bind conn. If
stop_conn_work_fn runs after bind but before start, we would leave the conn
in a unbound but sort of started state where IO might be allowed even
though the drivers have been set in a state where they no longer expect
I/O.
4. Returning -EAGAIN in iscsi_if_destroy_conn if we haven't yet run the in
kernel stop_conn function is breaking userspace. We should have been doing
this for the caller.
Mike Christie [Tue, 25 May 2021 18:17:59 +0000 (13:17 -0500)]
scsi: iscsi: Rel ref after iscsi_lookup_endpoint()
Subsequent commits allow the kernel to do ep_disconnect. In that case we
will have to get a proper refcount on the ep so one thread does not delete
it from under another.
Mike Christie [Tue, 25 May 2021 18:17:57 +0000 (13:17 -0500)]
scsi: iscsi: Force immediate failure during shutdown
If the system is not up, we can just fail immediately since iscsid is not
going to ever answer our netlink events. We are already setting the
recovery_tmo to 0, but by passing stop_conn STOP_CONN_TERM we never will
block the session and start the recovery timer, because for that flag
userspace will do the unbind and destroy events which would remove the
devices and wake up and kill the eh.
Since the conn is dead and the system is going dowm this just has us use
STOP_CONN_RECOVER with recovery_tmo=0 so we fail immediately. However, if
the user has set the recovery_tmo=-1 we let the system hang like they
requested since they might have used that setting for specific reasons
(one known reason is for buggy cluster software).
Mike Christie [Tue, 25 May 2021 18:17:55 +0000 (13:17 -0500)]
scsi: iscsi: Stop queueing during ep_disconnect
During ep_disconnect we have been doing iscsi_suspend_tx/queue to block new
I/O but every driver except cxgbi and iscsi_tcp can still get I/O from
__iscsi_conn_send_pdu() if we haven't called iscsi_conn_failure() before
ep_disconnect. This could happen if we were terminating the session, and
the logout timed out before it was even sent to libiscsi.
Fix the issue by adding a helper which reverses the bind_conn call that
allows new I/O to be queued. Drivers implementing ep_disconnect can use this
to make sure new I/O is not queued to them when handling the disconnect.
scsi: megaraid_sas: Handle missing interrupts while re-enabling IRQs
While reenabling the IRQ after IRQ poll there may be a small window for the
firmware to post the replies with interrupts raised. In that case the
driver will not see the interrupts which leads to I/O timeout.
This issue only happens when there are many I/O completions on a single
reply queue. This forces the driver to switch between the interrupt and IRQ
context.
Make the driver process the reply queue one more time after enabling the
IRQ.
Kashyap Desai [Fri, 28 May 2021 13:13:05 +0000 (18:43 +0530)]
scsi: megaraid_sas: Early detection of VD deletion through RaidMap update
Consider the case where a VD is deleted and the targetID of that VD is
assigned to a newly created VD. If the sequence of deletion/addition of VD
happens very quickly there is a possibility that second event (VD add)
occurs even before the driver processes the first event (VD delete). As
event processing is done in deferred context the device list remains the
same (but targetID is re-used) so driver will not learn the VD
deletion/additon. I/Os meant for the older VD will be directed to new VD
which may lead to data corruption.
Make driver detect the deleted VD as soon as possible based on the RaidMap
update and block further I/O to that device.
scsi: megaraid_sas: Send all non-RW I/Os for TYPE_ENCLOSURE device through firmware
The driver issues all non-ReadWrite I/Os for TYPE_ENCLOSURE devices through
the fast path with invalid dev handle. Fast path in turn directs all the
I/Os to the firmware. As firmware stopped handling those I/Os from SAS3.5
generation of controllers (Ventura generation and onwards) this will lead
to I/O failures.
Switch the driver to issue all the non-ReadWrite I/Os for TYPE_ENCLOSURE
devices directly to firmware for SAS3.5 generation of controllers and
later.
Kashyap Desai [Thu, 20 May 2021 15:25:43 +0000 (20:55 +0530)]
scsi: mpi3mr: Add support for DSN secure firmware check
Read PCI_EXT_CAP_ID_DSN to query security status.
The driver will throw a warning message when a non-secure type controller
is detected. The purpose of this interface is to avoid interacting with any
firmware which is not secured/signed by Broadcom. Any tampering on
firmware component will be detected by hardware and it will be communicated
to the driver to avoid any further interaction with that component.
Kashyap Desai [Thu, 20 May 2021 15:25:39 +0000 (20:55 +0530)]
scsi: mpi3mr: Complete support for soft reset
Unlock the host diagnostic register, write the specific reset type to that
and wait for reset acknowledgment from the controller. If the reset is not
successful retry for the predefined number of times
Kashyap Desai [Thu, 20 May 2021 15:25:38 +0000 (20:55 +0530)]
scsi: mpi3mr: Add support for threaded ISR
Register driver for threaded interrupts.
By default the driver will attempt I/O completion from interrupt context
(primary handler). Since the driver tracks per reply queue outstanding
I/Os, it will schedule threaded ISR if there are any outstanding I/Os
expected on that particular reply queue.
Threaded ISR (secondary handler) will loop for I/O completion as long as
there are outstanding I/Os (speculative method using same per reply queue
outstanding counter) or it has completed some X amount of commands
(something like budget).
Kashyap Desai [Thu, 20 May 2021 15:25:37 +0000 (20:55 +0530)]
scsi: mpi3mr: Hardware workaround for UNMAP commands to NVMe drives
The controller hardware can not handle certain UNMAP commands for NVMe
drives. Add support in the driver for checking those commands and handle
them appropriately.
Kashyap Desai [Thu, 20 May 2021 15:25:36 +0000 (20:55 +0530)]
scsi: mpi3mr: Allow certain commands during pci-remove hook
Instead of driver returning DID_NO_CONNECT during driver unload allow SSU
and Sync Cache commands to be sent to the controller to flush any cached
data from the drive.
Kashyap Desai [Thu, 20 May 2021 15:25:31 +0000 (20:55 +0530)]
scsi: mpi3mr: Add support for timestamp sync with firmware
This operation requests that the IOC update the TimeStamp.
When the I/O Unit is powered on it sets the TimeStamp field value to
0x0000_0000_0000_0000 and increments the current value every millisecond.
A host driver sets the TimeStamp field to the current time by using an
IOCInit request. The TimeStamp field is periodically updated by the host
driver.
Kashyap Desai [Thu, 20 May 2021 15:25:30 +0000 (20:55 +0530)]
scsi: mpi3mr: Add support for recovering controller
Detection of firmware fault or any kind of unresponsiveness in the
controller (any admin command which times out) results in resetting the
controller. The primary reset mechanisms used are either soft reset or diag
fault reset. A reset is performed if the host sets the ResetAction field in
the HostDiagnostic register to either 001b (soft reset) or 007b (diag fault
reset). After successfully resetting the controller the driver
reinitializes the controller by going through start of the day
initialization procedure. Pending I/Os during the reset are returned back
to the SCSI midlayer for retry.
projects / linux.git / log