Refactor aio submission code:
1. Break up function pqi_raid_bypass_submit_scsi_cmd()
into smaller functions.
2. Add common block (rmd - raid_map_data) to carry around into newly
added functions.
3. Prepare for new AIO functionality.
No functional changes.
Link: https://lore.kernel.org/r/161549371553.25025.8840958689316611074.stgit@brunhilda
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Reviewed-by: Martin Wilck <mwilck@suse.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Add support for newer hardware by adding in a product identifier. This
identifier can then be used to check for the hardware generation.
Link: https://lore.kernel.org/r/161549370966.25025.2968242206975557607.stgit@brunhilda
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Reviewed-by: Martin Wilck <mwilck@suse.com>
Signed-off-by: Kevin Barnett <kevin.barnett@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
While failing queued I/Os in TMF path, there was a request leak and hence
stale entries in request pool with ref count being non-zero. In shutdown
path we have a BUG_ON to catch stuck I/O either in firmware or in the
driver. The stale requests caused a system crash. The I/O request pool
leakage also lead to a significant performance drop.
Link: https://lore.kernel.org/r/161549370379.25025.12793264112620796062.stgit@brunhilda
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Signed-off-by: Murthy Bhat <Murthy.Bhat@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Correct SCSI midlayer sending more requests than exposed host queue depth
causing firmware ASSERT and lockup issues by enabling host-wide tags.
Note: This also results in better performance.
Link: https://lore.kernel.org/r/161549369787.25025.8975999483518581619.stgit@brunhilda
Suggested-by: Ming Lei <ming.lei@redhat.com>
Suggested-by: John Garry <john.garry@huawei.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Pull 5.12/scsi-fixes into the 5.13 SCSI tree to provide a baseline for
some UFS changes that would otherwise cause conflicts during the
merge.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
In __ufshcd_issue_tm_cmd(), it is not correct to use hba->nutrs + req->tag
as the Task Tag in a TMR UPIU. Directly use req->tag as the Task Tag.
Fixes: e293313262 ("scsi: ufs: Fix broken task management command implementation")
Link: https://lore.kernel.org/r/1617262750-4864-3-git-send-email-cang@codeaurora.org
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Can Guo <cang@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
ufshcd_tmc_handler() calls blk_mq_tagset_busy_iter(fn = ufshcd_compl_tm()),
but since blk_mq_tagset_busy_iter() only iterates over all reserved tags
and requests which are not in IDLE state, ufshcd_compl_tm() never gets a
chance to run. Thus, TMR always ends up with completion timeout. Fix it by
calling blk_mq_start_request() in __ufshcd_issue_tm_cmd().
Link: https://lore.kernel.org/r/1617262750-4864-2-git-send-email-cang@codeaurora.org
Fixes: 69a6c269c0 ("scsi: ufs: Use blk_{get,put}_request() to allocate and free TMFs")
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Can Guo <cang@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
There is a regular need in the kernel to provide a way to declare having a
dynamically sized set of trailing elements in a structure. Kernel code
should always use "flexible array members"[1] for these cases. The older
style of one-element or zero-length arrays should no longer be used[2].
Refactor the code according to the use of a flexible-array member in struct
_SGE_TRANSACTION32 instead of one-element array.
Also, this helps with the ongoing efforts to enable -Warray-bounds by
fixing the following warning:
CC [M] drivers/message/fusion/mptlan.o
drivers/message/fusion/mptlan.c: In function ‘mpt_lan_sdu_send’:
drivers/message/fusion/mptlan.c:759:28: warning: array subscript 1 is above array bounds of ‘U32[1]’ {aka ‘unsigned int[1]’} [-Warray-bounds]
759 | pTrans->TransactionDetails[1] = cpu_to_le32((mac[2] << 24) |
| ~~~~~~~~~~~~~~~~~~~~~~~~~~^~~
[1] https://en.wikipedia.org/wiki/Flexible_array_member
[2] https://www.kernel.org/doc/html/v5.10/process/deprecated.html#zero-length-and-one-element-arrays
Link: https://github.com/KSPP/linux/issues/79
Link: https://github.com/KSPP/linux/issues/109
Link: https://lore.kernel.org/r/20210324233344.GA99059@embeddedor
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
There is a regular need in the kernel to provide a way to declare having a
dynamically sized set of trailing elements in a structure. Kernel code
should always use "flexible array members"[1] for these cases. The older
style of one-element or zero-length arrays should no longer be used[2].
Also, this helps with the ongoing efforts to enable -Warray-bounds by
fixing the following warning:
drivers/message/fusion/mptbase.c: In function ‘mptbase_reply’:
drivers/message/fusion/mptbase.c:7747:62: warning: array subscript 1 is above array bounds of ‘U32[1]’ {aka ‘unsigned int[1]’} [-Warray-bounds]
7747 | ioc->events[idx].data[ii] = le32_to_cpu(pEventReply->Data[ii]);
./include/uapi/linux/byteorder/little_endian.h:34:51: note: in definition of macro ‘__le32_to_cpu’
34 | #define __le32_to_cpu(x) ((__force __u32)(__le32)(x))
| ^
drivers/message/fusion/mptbase.c:7747:33: note: in expansion of macro ‘le32_to_cpu’
7747 | ioc->events[idx].data[ii] = le32_to_cpu(pEventReply->Data[ii]);
|
[1] https://en.wikipedia.org/wiki/Flexible_array_member
[2] https://www.kernel.org/doc/html/v5.10/process/deprecated.html#zero-length-and-one-element-arrays
Link: https://github.com/KSPP/linux/issues/79
Link: https://github.com/KSPP/linux/issues/109
Link: https://lore.kernel.org/r/20210324230036.GA67851@embeddedor
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
An old cleanup changed the array size from MAX_ADDR_LEN to unspecified in
the declaration, but now gcc-11 warns about this:
drivers/scsi/fcoe/fcoe_ctlr.c:1972:37: error: argument 1 of type ‘unsigned char[32]’ with mismatched bound [-Werror=array-parameter=]
1972 | u64 fcoe_wwn_from_mac(unsigned char mac[MAX_ADDR_LEN],
| ~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~
In file included from /git/arm-soc/drivers/scsi/fcoe/fcoe_ctlr.c:33:
include/scsi/libfcoe.h:252:37: note: previously declared as ‘unsigned char[]’
252 | u64 fcoe_wwn_from_mac(unsigned char mac[], unsigned int, unsigned int);
| ~~~~~~~~~~~~~~^~~~~
Change the type back to what the function definition uses.
Link: https://lore.kernel.org/r/20210322164702.957810-1-arnd@kernel.org
Fixes: fdd78027fd ("[SCSI] fcoe: cleans up libfcoe.h and adds fcoe.h for fcoe module")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
gcc-11 warns about an strnlen with a length larger than the size of the
passed buffer:
drivers/scsi/lpfc/lpfc_attr.c: In function 'lpfc_nvme_info_show':
drivers/scsi/lpfc/lpfc_attr.c:518:25: error: 'strnlen' specified bound 4095 exceeds source size 24 [-Werror=stringop-overread]
518 | strnlen(LPFC_NVME_INFO_MORE_STR, PAGE_SIZE - 1)
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In this case, the code is entirely valid, as the string is properly
terminated, and the size argument is only there out of extra caution in
case it exceeds a page.
This cannot really happen here, so just simplify it to a sizeof().
Link: https://lore.kernel.org/r/20210322160253.4032422-10-arnd@kernel.org
Fixes: afff0d2321 ("scsi: lpfc: Add Buffer overflow check, when nvme_info larger than PAGE_SIZE")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Building with 'make W=1' shows a few harmless -Wempty-body warning for the
mvsas driver:
drivers/scsi/mvsas/mv_94xx.c: In function 'mvs_94xx_phy_reset':
drivers/scsi/mvsas/mv_94xx.c:278:63: error: suggest braces around empty body in an 'if' statement [-Werror=empty-body]
278 | mv_dprintk("phy hard reset failed.\n");
| ^
drivers/scsi/mvsas/mv_sas.c: In function 'mvs_task_prep':
drivers/scsi/mvsas/mv_sas.c:723:57: error: suggest braces around empty body in an 'else' statement [-Werror=empty-body]
723 | SAS_ADDR(dev->sas_addr));
| ^
Change the empty dprintk() macros to no_printk(), which avoids this warning
and adds format string checking.
Link: https://lore.kernel.org/r/20210322103316.620694-1-arnd@kernel.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
There are a couple of warnings in this driver when building with W=1:
drivers/message/fusion/mptbase.c: In function 'PrimeIocFifos':
drivers/message/fusion/mptbase.c:4608:65: error: suggest braces around empty body in an 'if' statement [-Werror=empty-body]
4608 | "restoring 64 bit addressing\n", ioc->name));
| ^
drivers/message/fusion/mptbase.c:4633:65: error: suggest braces around empty body in an 'if' statement [-Werror=empty-body]
4633 | "restoring 64 bit addressing\n", ioc->name));
The macros are slightly suboptimal since are not proper statements.
Change both versions to the usual "do { ... } while (0)" style to
make them more robust and avoid the warning.
Link: https://lore.kernel.org/r/20210322102549.278661-2-arnd@kernel.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Building with 'make W=1' shows a harmless -Wempty-body warning:
drivers/scsi/aic94xx/aic94xx_init.c: In function 'asd_free_queues':
drivers/scsi/aic94xx/aic94xx_init.c:858:62: error: suggest braces around empty body in an 'if' statement [-Werror=empty-body]
858 | ASD_DPRINTK("Uh-oh! Pending is not empty!\n");
Change the empty ASD_DPRINTK() macro to no_printk(), which avoids this
warning and adds format string checking.
Link: https://lore.kernel.org/r/20210322102549.278661-1-arnd@kernel.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20210330071958.3788214-3-slyfox@gentoo.org
Fixes: f749d8b7a9 ("scsi: hpsa: Correct dev cmds outstanding for retried cmds")
CC: linux-ia64@vger.kernel.org
CC: storagedev@microchip.com
CC: linux-scsi@vger.kernel.org
CC: Joe Szczypek <jszczype@redhat.com>
CC: Scott Benesh <scott.benesh@microchip.com>
CC: Scott Teel <scott.teel@microchip.com>
CC: Tomas Henzl <thenzl@redhat.com>
CC: "Martin K. Petersen" <martin.petersen@oracle.com>
CC: Don Brace <don.brace@microchip.com>
Reported-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Suggested-by: Don Brace <don.brace@microchip.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Boot failure was observed on an HP rx3600 ia64 machine with RAID bus
controller: Hewlett-Packard Company Smart Array P600:
kernel unaligned access to 0xe000000105dd8b95, ip=0xa000000100b87551
kernel unaligned access to 0xe000000105dd8e95, ip=0xa000000100b87551
hpsa 0000:14:01.0: Controller reports max supported commands of 0 Using 16 instead. Ensure that firmware is up to date.
swapper/0[1]: error during unaligned kernel access
The unaligned access comes from 'struct CommandList' that happens to be
packed. Commit f749d8b7a9 ("scsi: hpsa: Correct dev cmds outstanding for
retried cmds") introduced unexpected padding and unaligned atomic_t from
natural alignment to something else.
This change removes packing annotation from a struct not intended to be
sent to controller as is. This restores natural `atomic_t` alignment.
The change was tested on the same rx3600 machine.
Link: https://lore.kernel.org/r/20210330071958.3788214-2-slyfox@gentoo.org
Fixes: f749d8b7a9 ("scsi: hpsa: Correct dev cmds outstanding for retried cmds")
CC: linux-ia64@vger.kernel.org
CC: linux-kernel@vger.kernel.org
CC: storagedev@microchip.com
CC: linux-scsi@vger.kernel.org
CC: Joe Szczypek <jszczype@redhat.com>
CC: Scott Benesh <scott.benesh@microchip.com>
CC: Scott Teel <scott.teel@microchip.com>
CC: Tomas Henzl <thenzl@redhat.com>
CC: "Martin K. Petersen" <martin.petersen@oracle.com>
CC: Don Brace <don.brace@microchip.com>
Reported-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Suggested-by: Don Brace <don.brace@microchip.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The hpsa driver uses data structures which contain a combination of driver
internals and commands sent directly to the hardware. To manage alignment
for the hardware portions the driver used #pragma pack(1).
Commit f749d8b7a9 ("scsi: hpsa: Correct dev cmds outstanding for retried
cmds") switched an existing variable from int to bool. Due to the pragma an
atomic_t in the same data structure ended up being misaligned and broke
boot on ia64.
Add __packed to every struct and union in the header file. Subsequent
commits will address the actual atomic_t misalignment regression.
The commit is a no-op at least on ia64:
$ diff -u <(objdump -d -r old.o) <(objdump -d -r new.o)
Link: https://lore.kernel.org/r/20210330071958.3788214-1-slyfox@gentoo.org
Fixes: f749d8b7a9 ("scsi: hpsa: Correct dev cmds outstanding for retried cmds")
CC: linux-ia64@vger.kernel.org
CC: storagedev@microchip.com
CC: linux-scsi@vger.kernel.org
CC: Joe Szczypek <jszczype@redhat.com>
CC: Scott Benesh <scott.benesh@microchip.com>
CC: Scott Teel <scott.teel@microchip.com>
CC: Tomas Henzl <thenzl@redhat.com>
CC: "Martin K. Petersen" <martin.petersen@oracle.com>
CC: Don Brace <don.brace@microchip.com>
Reported-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Suggested-by: Don Brace <don.brace@microchip.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The variable biosaddr is being assigned a value that is never read, the
variable is redundant and can be safely removed.
Link: https://lore.kernel.org/r/20210325170731.484651-1-colin.king@canonical.com
Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Variable err is assigned -ENOMEM followed by an error return path via label
err_udev that does not access the variable and returns with the -ENOMEM
error return code. The assignment to err is redundant and can be removed.
Link: https://lore.kernel.org/r/20210327230650.25803-1-colin.king@canonical.com
Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
struct request and struct request_queue are declared twice. Remove the
duplicate declarations.
Link: https://lore.kernel.org/r/20210327030850.918018-1-wanjiabing@vivo.com
Signed-off-by: Wan Jiabing <wanjiabing@vivo.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Commit b43abcbbd5 ("scsi: fnic: Ratelimit printks to avoid flooding when
vlan is not set by the switch.i") added printk_ratelimit() in front of a
couple of debug-mode messages to reduce logging overrun when debugging the
driver. The code:
> if (printk_ratelimit())
> FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host,
> "Start VLAN Discovery\n");
ends up calling printk_ratelimit() quite often, triggering many kernel
messages about callbacks being supressed.
The fix is to decompose FNIC_FCS_DBG(), then change the order of checks so
that printk_ratelimit() is only called if driver debugging is enabled.
Link: https://lore.kernel.org/r/20210323172756.5743-1-lduncan@suse.com
Reviewed-by: Laurence Oberman <loberman@redhat.com>
Signed-off-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Perform logout of all remote ports so that all I/Os with driver are
requeued with midlayer for retry.
Link: https://lore.kernel.org/r/20210329085229.4367-12-njavali@marvell.com
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
For the mailbox thread that encounters a PCIe error, pause that thread
until PCIe link reset/recovery has completed to prevent the thread from
possibly unmapping any type of DMA resource that might be in progress.
Link: https://lore.kernel.org/r/20210329085229.4367-10-njavali@marvell.com
Tested-by: Laurence Oberman <loberman@redhat.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
After RISC RESET, the poll time for completion is too short. Fix the
completion polling time.
Link: https://lore.kernel.org/r/20210329085229.4367-8-njavali@marvell.com
Tested-by: Laurence Oberman <loberman@redhat.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
On bsg command completion, bsg_job_done() was called while qla driver
continued to access the bsg_job buffer. bsg_job_done() would free up
resources that ended up being reused by other task while the driver
continued to access the buffers. As a result, driver was reading garbage
data.
localhost kernel: BUG: KASAN: use-after-free in sg_next+0x64/0x80
localhost kernel: Read of size 8 at addr ffff8883228a3330 by task swapper/26/0
localhost kernel:
localhost kernel: CPU: 26 PID: 0 Comm: swapper/26 Kdump:
loaded Tainted: G OE --------- - - 4.18.0-193.el8.x86_64+debug #1
localhost kernel: Hardware name: HP ProLiant DL360
Gen9/ProLiant DL360 Gen9, BIOS P89 08/12/2016
localhost kernel: Call Trace:
localhost kernel: <IRQ>
localhost kernel: dump_stack+0x9a/0xf0
localhost kernel: print_address_description.cold.3+0x9/0x23b
localhost kernel: kasan_report.cold.4+0x65/0x95
localhost kernel: debug_dma_unmap_sg.part.12+0x10d/0x2d0
localhost kernel: qla2x00_bsg_sp_free+0xaf6/0x1010 [qla2xxx]
Link: https://lore.kernel.org/r/20210329085229.4367-6-njavali@marvell.com
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Consolidate zio threshold setting for both FCP & NVMe to prevent one
protocol from clobbering the setting of the other protocol.
Link: https://lore.kernel.org/r/20210329085229.4367-5-njavali@marvell.com
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Session was stuck due to explicit logout to target timing out. The target
was in an unresponsive state. This timeout induced an error to the GNL
command from moving forward.
Link: https://lore.kernel.org/r/20210329085229.4367-4-njavali@marvell.com
Tested-by: Laurence Oberman <loberman@redhat.com>
Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The host:channel:scsi_target_id information is helpful in matching an FC
port with a SCSI device, so add it. For initiator FC ports, a -1 would be
displayed for "target" part.
Link: https://lore.kernel.org/r/20210329085229.4367-3-njavali@marvell.com
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Arun Easi <aeasi@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Removing the response queue processing in the send path is showing IOPS
drop. Add back the process_response_queue() call in the send path.
Link: https://lore.kernel.org/r/20210329085229.4367-2-njavali@marvell.com
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Arun Easi <aeasi@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
A kernel panic was observed due to a timing issue between the sync thread
and the initiator processing a login response from the target. The session
reopen can be invoked both from the session sync thread when iscsid
restarts and from iscsid through the error handler. Before the initiator
receives the response to a login, another reopen request can be sent from
the error handler/sync session. When the initial login response is
subsequently processed, the connection has been closed and the socket has
been released.
To fix this a new connection state, ISCSI_CONN_BOUND, is added:
- Set the connection state value to ISCSI_CONN_DOWN upon
iscsi_if_ep_disconnect() and iscsi_if_stop_conn()
- Set the connection state to the newly created value ISCSI_CONN_BOUND
after bind connection (transport->bind_conn())
- In iscsi_set_param(), return -ENOTCONN if the connection state is not
either ISCSI_CONN_BOUND or ISCSI_CONN_UP
Link: https://lore.kernel.org/r/20210325093248.284678-1-gulam.mohamed@oracle.com
Reviewed-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Gulam Mohamed <gulam.mohamed@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
index 91074fd97f64..f4bf62b007a0 100644
If pscsi_map_sg() fails, make sure to drop references to already allocated
bios.
Link: https://lore.kernel.org/r/20210323212431.15306-2-mwilck@suse.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Martin Wilck <mwilck@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
pscsi_map_sg() uses the variable nr_pages as a hint for bio_kmalloc() how
many vector elements to allocate. If nr_pages is < BIO_MAX_PAGES, it will
be reset to 0 after successful allocation of the bio.
If bio_add_pc_page() fails later for whatever reason, pscsi_map_sg() tries
to allocate another bio, passing nr_vecs = 0. This causes bio_add_pc_page()
to fail immediately in the next call. pci_map_sg() continues to allocate
zero-length bios until memory is exhausted and the kernel crashes with
OOM. This can be easily observed by exporting a SATA DVD drive via pscsi.
The target crashes as soon as the client tries to access the DVD LUN. In
the case I analyzed, bio_add_pc_page() would fail because the DVD device's
max_sectors_kb (128) was exceeded.
Avoid this by simply not resetting nr_pages to 0 after allocating the
bio. This way, the client receives an I/O error when it tries to send
requests exceeding the devices max_sectors_kb, and eventually gets it
right. The client must still limit max_sectors_kb e.g. by an udev rule if
(like in my case) the driver doesn't report valid block limits, otherwise
it encounters I/O errors.
Link: https://lore.kernel.org/r/20210323212431.15306-1-mwilck@suse.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Martin Wilck <mwilck@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
On some configurations, gcc warns about overlapping source and destination
arguments to snprintf:
drivers/scsi/pm8001/pm8001_init.c: In function 'pm8001_request_msix':
drivers/scsi/pm8001/pm8001_init.c:977:3: error: 'snprintf' argument 4 may overlap destination object 'pm8001_ha' [-Werror=restrict]
977 | snprintf(drvname, len, "%s-%d", pm8001_ha->name, i);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/scsi/pm8001/pm8001_init.c:962:56: note: destination object referenced by 'restrict'-qualified argument 1 was declared here
962 | static u32 pm8001_request_msix(struct pm8001_hba_info *pm8001_ha)
| ~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~
I first assumed this was a gcc bug, as that should not happen, but a
reduced test case makes it clear that this happens when the loop counter is
not bounded by the array size.
Help the compiler out by adding an explicit limit here to make the code
slightly more robust and avoid the warning.
Link: https://godbolt.org/z/6T1qPM
Link: https://lore.kernel.org/r/20210323125458.1825564-1-arnd@kernel.org
Acked-by: Jack Wang <jinpu.wang@ionos.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
s/defintions/definitions/ ....two different places.
Link: https://lore.kernel.org/r/20210322205821.1449844-1-unixbhaskar@gmail.com
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Bhaskar Chowdhury <unixbhaskar@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Don Brace
Kevin Barnett
Murthy Bhat
Martin K. Petersen
Can Guo
Gustavo A. R. Silva
Arnd Bergmann
Sergei Trofimovich
Shixin Liu
ganjisheng
Colin Ian King
Wan Jiabing
dudengke
Lee Duncan
Nilesh Javali
Quinn Tran
Arun Easi
Gulam Mohamed
Martin Wilck
Bhaskar Chowdhury
wengjianfeng