When the lldd is processing the complete sas task in interrupt and set the
task stat as SAS_TASK_STATE_DONE, the smp timeout timer is able to be
triggered at the same time. And smp_task_timedout() will complete the task
wheter the SAS_TASK_STATE_DONE is set or not. Then the sas task may freed
before lldd end the interrupt process. Thus a use-after-free will happen.
Fix this by calling the complete() only when SAS_TASK_STATE_DONE is not
set. And remove the check of the return value of the del_timer(). Once the
LLDD sets DONE, it must call task->done(), which will call
smp_task_done()->complete() and the task will be completed and freed
correctly.
Reported-by: chenxiang <chenxiang66@hisilicon.com>
Signed-off-by: Jason Yan <yanaijie@huawei.com>
CC: John Garry <john.garry@huawei.com>
CC: Johannes Thumshirn <jthumshirn@suse.de>
CC: Ewan Milne <emilne@redhat.com>
CC: Christoph Hellwig <hch@lst.de>
CC: Tomas Henzl <thenzl@redhat.com>
CC: Dan Williams <dan.j.williams@intel.com>
CC: Hannes Reinecke <hare@suse.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
When ata device IDENTIFY failed, the ata device status is ATA_DEV_UNKNOWN. The
libata reported like:
[113518.620433] ata5.00: qc timeout (cmd 0xec)
[113518.653646] ata5.00: failed to IDENTIFY (I/O error, err_mask=0x4)
But libsas verifies the device status by ata_dev_disabled(), which skipped
ATA_DEV_UNKNOWN. This will make libsas think the ata device probing succeed
the device cannot be actually brought up. And even the new bcast of this
device will be considered as flutter and will not probe this device again.
Change ata_dev_disabled() to !ata_dev_enabled() so that libsas can deal with
this if the ata device probe failed. New bcasts can let us try to probe the
device again and bring it up if it is fine to IDENTIFY.
Tested-by: Zhou Yupeng <zhouyupeng1@huawei.com>
Signed-off-by: Jason Yan <yanaijie@huawei.com>
CC: John Garry <john.garry@huawei.com>
CC: Johannes Thumshirn <jthumshirn@suse.de>
CC: Ewan Milne <emilne@redhat.com>
CC: Christoph Hellwig <hch@lst.de>
CC: Tomas Henzl <thenzl@redhat.com>
CC: Dan Williams <dan.j.williams@intel.com>
CC: Hannes Reinecke <hare@suse.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Commit 2623c7a5f2 ("libata: add refcounting to ata_host") v4.17+ introduced
refcounting to ata_host and will increase or decrease the refcount when
adding or deleting transport ATA port.
Now the ata host for libsas is embedded in domain_device, and the ->kref
member is not initialized. Afer we add ata transport class, ata_host_get()
will be called when adding transport ATA port and a warning will be
triggered as below:
refcount_t: increment on 0; use-after-free.
WARNING: CPU: 2 PID: 103 at
lib/refcount.c:153 refcount_inc+0x40/0x48 ...... Call trace:
refcount_inc+0x40/0x48
ata_host_get+0x10/0x18
ata_tport_add+0x40/0x120
ata_sas_tport_add+0xc/0x14
sas_ata_init+0x7c/0xc8
sas_discover_domain+0x380/0x53c
process_one_work+0x12c/0x288
worker_thread+0x58/0x3f0
kthread+0xfc/0x128
ret_from_fork+0x10/0x18
And also when removing transport ATA port ata_host_put() will be called and
another similar warning will be triggered. If the refcount decreased to
zero, the ata host will be freed. But this ata host is only part of
domain_device, it cannot be freed directly.
So we have to change this embedded static ata host to a dynamically
allocated ata host and initialize the ->kref member. To use ata_host_get()
and ata_host_put() in libsas, we need to move the declaration of these
functions to the public libata.h and export them.
Fixes: b6240a4df0 ("scsi: libsas: add transport class for ATA devices")
Signed-off-by: Jason Yan <yanaijie@huawei.com>
CC: John Garry <john.garry@huawei.com>
CC: Taras Kondratiuk <takondra@cisco.com>
CC: Tejun Heo <tj@kernel.org>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Since commit 312d3e5611 ("[SCSI] libsas: remove ata_port.lock management
duties from lldds") the sas_ata_qc_issue() function unlocks the
ata_port.lock and disables interrupts before doing so. That lock is always
taken with disabled interrupts so at this point, the interrupts are already
disabled. There is no need to disable the interrupts before the unlock
operation because they are already disabled. Restoring the interrupt state
later does not change anything because they were disabled and remain
disabled. Therefore remove the operations which do not change the
behaviour.
Fixes: 312d3e5611 ("[SCSI] libsas: remove ata_port.lock management duties from lldds")
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Acked-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Now ata devices attached with sas controller do not have transport
class, so that we can not see any information of these ata devices in
/sys/class/ata_port(or ata_link or ata_device).
Add transport class for the ata devices attached with sas controller.
The /sys/class directory will show the infomation of the ata devices
as follows:
localhost:/sys/class # ls ata*
ata_device:
dev1.0 dev2.0
ata_link:
link1 link2
ata_port:
ata1 ata2
No functional change of the device scanning and io path. The ata
transport class was deleted when destroying the sas devices.
Signed-off-by: Jason Yan <yanaijie@huawei.com>
CC: Dan Williams <dan.j.williams@intel.com>
CC: Tejun Heo <tj@kernel.org>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Somewhat nasty merge due to conflicts between "33b28357dd00 scsi:
qla2xxx: Fix Async GPN_FT for FCP and FC-NVMe scan" and "2b5b96473efc
scsi: qla2xxx: Fix FC-NVMe LUN discovery"
Merge is non-trivial and has been verified by Qlogic (Cavium)
Signed-off-by: James E.J. Bottomley <jejb@linux.vnet.ibm.com>
When ata device doing EH, some commands still attached with tasks are
not passed to libata when abort failed or recover failed, so libata did
not handle these commands. After these commands done, sas task is freed,
but ata qc is not freed. This will cause ata qc leak and trigger a
warning like below:
WARNING: CPU: 0 PID: 28512 at drivers/ata/libata-eh.c:4037
ata_eh_finish+0xb4/0xcc
CPU: 0 PID: 28512 Comm: kworker/u32:2 Tainted: G W OE 4.14.0#1
......
Call trace:
[<ffff0000088b7bd0>] ata_eh_finish+0xb4/0xcc
[<ffff0000088b8420>] ata_do_eh+0xc4/0xd8
[<ffff0000088b8478>] ata_std_error_handler+0x44/0x8c
[<ffff0000088b8068>] ata_scsi_port_error_handler+0x480/0x694
[<ffff000008875fc4>] async_sas_ata_eh+0x4c/0x80
[<ffff0000080f6be8>] async_run_entry_fn+0x4c/0x170
[<ffff0000080ebd70>] process_one_work+0x144/0x390
[<ffff0000080ec100>] worker_thread+0x144/0x418
[<ffff0000080f2c98>] kthread+0x10c/0x138
[<ffff0000080855dc>] ret_from_fork+0x10/0x18
If ata qc leaked too many, ata tag allocation will fail and io blocked
for ever.
As suggested by Dan Williams, defer ata device commands to libata and
merge sas_eh_finish_cmd() with sas_eh_defer_cmd(). libata will handle
ata qcs correctly after this.
Signed-off-by: Jason Yan <yanaijie@huawei.com>
CC: Xiaofei Tan <tanxiaofei@huawei.com>
CC: John Garry <john.garry@huawei.com>
CC: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Avoid that building with W=1 causes the kernel-doc tool to complain
about function arguments that have not been documented in the libsas
kernel-doc headers. Avoid that the short description starts with a
hyphen by changing "--" into "-" in the first line of the kernel-doc
headers.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Cc: John Garry <john.garry@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Pull SCSI updates from James Bottomley:
"This is mostly updates of the usual driver suspects: arcmsr,
scsi_debug, mpt3sas, lpfc, cxlflash, qla2xxx, aacraid, megaraid_sas,
hisi_sas.
We also have a rework of the libsas hotplug handling to make it more
robust, a slew of 32 bit time conversions and fixes, and a host of the
usual minor updates and style changes. The biggest potential for
regressions is the libsas hotplug changes, but so far they seem stable
under testing"
* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (313 commits)
scsi: qla2xxx: Fix logo flag for qlt_free_session_done()
scsi: arcmsr: avoid do_gettimeofday
scsi: core: Add VENDOR_SPECIFIC sense code definitions
scsi: qedi: Drop cqe response during connection recovery
scsi: fas216: fix sense buffer initialization
scsi: ibmvfc: Remove unneeded semicolons
scsi: hisi_sas: fix a bug in hisi_sas_dev_gone()
scsi: hisi_sas: directly attached disk LED feature for v2 hw
scsi: hisi_sas: devicetree: bindings: add LED feature for v2 hw
scsi: megaraid_sas: NVMe passthrough command support
scsi: megaraid: use ktime_get_real for firmware time
scsi: fnic: use 64-bit timestamps
scsi: qedf: Fix error return code in __qedf_probe()
scsi: devinfo: fix format of the device list
scsi: qla2xxx: Update driver version to 10.00.00.05-k
scsi: qla2xxx: Add XCB counters to debugfs
scsi: qla2xxx: Fix queue ID for async abort with Multiqueue
scsi: qla2xxx: Fix warning for code intentation in __qla24xx_handle_gpdb_event()
scsi: qla2xxx: Fix warning during port_name debug print
scsi: qla2xxx: Fix warning in qla2x00_async_iocb_timeout()
...
Pull SCSI fix from James Bottomley:
"One fix for SAS attached SATA CD-ROMs. It turns out that the libata
handling of CD devices relies on the SCSI error handler, so disable
async aborts (which don't start the error handler) for these devices"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: libsas: Disable asynchronous aborts for SATA devices
There are two places queuing the disco event DISCE_REVALIDATE_DOMAIN.
One is in sas_porte_broadcast_rcvd() and uses sas_chain_event() to queue
the event. The other is in sas_enable_revalidation() and uses
sas_queue_event() to queue the event. We have diffrent work queues for
event and discovery now, so the DISCE_REVALIDATE_DOMAIN event may be
processed in both event queue and discovery queue.
Now since we do synchronous event handling, we cannot do it in discovery
queue, so have to trigger a fake broadcast event to re-trigger the
revalidation from event queue.
Signed-off-by: Jason Yan <yanaijie@huawei.com>
CC: John Garry <john.garry@huawei.com>
CC: Johannes Thumshirn <jthumshirn@suse.de>
CC: Ewan Milne <emilne@redhat.com>
CC: Christoph Hellwig <hch@lst.de>
CC: Tomas Henzl <thenzl@redhat.com>
CC: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
In commit 87c8331fcf ("[SCSI] libsas: prevent domain rediscovery
competing with ata error handling") introduced disco mutex to prevent
rediscovery competing with ata error handling and put the whole
revalidation in the mutex. But the rphy add/remove needs to wait for the
error handling which also grabs the disco mutex. This may leads to dead
lock.So the probe and destruct event were introduce to do the rphy
add/remove asynchronously and out of the lock.
The asynchronously processed workers makes the whole discovery process
not atomic, the other events may interrupt the process. For example,
if a loss of signal event inserted before the probe event, the
sas_deform_port() is called and the port will be deleted.
And sas_port_delete() may run before the destruct event, but the
port-x:x is the top parent of end device or expander. This leads to
a kernel WARNING such as:
[ 82.042979] sysfs group 'power' not found for kobject 'phy-1:0:22'
[ 82.042983] ------------[ cut here ]------------
[ 82.042986] WARNING: CPU: 54 PID: 1714 at fs/sysfs/group.c:237
sysfs_remove_group+0x94/0xa0
[ 82.043059] Call trace:
[ 82.043082] [<ffff0000082e7624>] sysfs_remove_group+0x94/0xa0
[ 82.043085] [<ffff00000864e320>] dpm_sysfs_remove+0x60/0x70
[ 82.043086] [<ffff00000863ee10>] device_del+0x138/0x308
[ 82.043089] [<ffff00000869a2d0>] sas_phy_delete+0x38/0x60
[ 82.043091] [<ffff00000869a86c>] do_sas_phy_delete+0x6c/0x80
[ 82.043093] [<ffff00000863dc20>] device_for_each_child+0x58/0xa0
[ 82.043095] [<ffff000008696f80>] sas_remove_children+0x40/0x50
[ 82.043100] [<ffff00000869d1bc>] sas_destruct_devices+0x64/0xa0
[ 82.043102] [<ffff0000080e93bc>] process_one_work+0x1fc/0x4b0
[ 82.043104] [<ffff0000080e96c0>] worker_thread+0x50/0x490
[ 82.043105] [<ffff0000080f0364>] kthread+0xfc/0x128
[ 82.043107] [<ffff0000080836c0>] ret_from_fork+0x10/0x50
Make probe and destruct a direct call in the disco and revalidate function,
but put them outside the lock. The whole discovery or revalidate won't
be interrupted by other events. And the DISCE_PROBE and DISCE_DESTRUCT
event are deleted as a result of the direct call.
Introduce a new list to destruct the sas_port and put the port delete after
the destruct. This makes sure the right order of destroying the sysfs
kobject and fix the warning above.
In sas_ex_revalidate_domain() have a loop to find all broadcasted
device, and sometimes we have a chance to find the same expander twice.
Because the sas_port will be deleted at the end of the whole revalidate
process, sas_port with the same name cannot be added before this.
Otherwise the sysfs will complain of creating duplicate filename. Since
the LLDD will send broadcast for every device change, we can only
process one expander's revalidation.
[mkp: kbuild test robot warning]
Signed-off-by: Jason Yan <yanaijie@huawei.com>
CC: John Garry <john.garry@huawei.com>
CC: Johannes Thumshirn <jthumshirn@suse.de>
CC: Ewan Milne <emilne@redhat.com>
CC: Christoph Hellwig <hch@lst.de>
CC: Tomas Henzl <thenzl@redhat.com>
CC: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Handling CD-ROM devices from libsas is decidedly odd, as libata relies
on SCSI EH to be started to figure out that no medium is present. So we
cannot do asynchronous aborts for SATA devices.
Fixes: 909657615d ("scsi: libsas: allow async aborts")
Cc: <stable@vger.kernel.org> # 4.12+
Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Tested-by: Yves-Alexis Perez <corsac@debian.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Now we are processing sas event and discover event in different
workqueues. It's safe to wait the discover event done in the sas event
work. Use flush_workqueue() to insure the disco and revalidate events
processed synchronously so that the whole discover and revalidate
process will not be interrupted by other events.
Signed-off-by: Jason Yan <yanaijie@huawei.com>
CC: John Garry <john.garry@huawei.com>
CC: Johannes Thumshirn <jthumshirn@suse.de>
CC: Ewan Milne <emilne@redhat.com>
CC: Christoph Hellwig <hch@lst.de>
CC: Tomas Henzl <thenzl@redhat.com>
CC: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Now all libsas works are queued to scsi host workqueue, include sas
event work post by LLDD and sas discovery work, and a sas hotplug flow
may be divided into several works, e.g libsas receive a
PORTE_BYTES_DMAED event, currently we process it as following steps:
sas_form_port --- run in work in shost workq
sas_discover_domain --- run in another work in shost workq
...
sas_probe_devices --- run in new work in shost workq
We found during hot-add a device, libsas may need run several
works in same workqueue to add device in system, the process is
not atomic, it may interrupt by other sas event works, like
PHYE_LOSS_OF_SIGNAL.
This patch is preparation of execute libsas sas event in sync. We need
to use different workqueue to run sas event and disco event. Otherwise
the work will be blocked for waiting another chained work in the same
workqueue.
Signed-off-by: Yijing Wang <wangyijing@huawei.com>
CC: John Garry <john.garry@huawei.com>
CC: Johannes Thumshirn <jthumshirn@suse.de>
CC: Ewan Milne <emilne@redhat.com>
CC: Christoph Hellwig <hch@lst.de>
CC: Tomas Henzl <thenzl@redhat.com>
CC: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Jason Yan <yanaijie@huawei.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
If the PHY burst too many events, we will alloc a lot of events for the
worker. This may leads to memory exhaustion.
Dan Williams suggested to shut down the PHY if the events reached the
threshold, because in this case the PHY may have gone into some
erroneous state. Users can re-enable the PHY by sysfs if they want.
We cannot use the fixed memory pool because if we run out of events, the
shut down event and loss of signal event will lost too. The events still
need to be allocated and processed in this case.
Suggested-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Jason Yan <yanaijie@huawei.com>
CC: John Garry <john.garry@huawei.com>
CC: Johannes Thumshirn <jthumshirn@suse.de>
CC: Ewan Milne <emilne@redhat.com>
CC: Christoph Hellwig <hch@lst.de>
CC: Tomas Henzl <thenzl@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Now libsas hotplug work is static, every sas event type has its own
static work, LLDD driver queues the hotplug work into shost->work_q. If
LLDD driver burst posts lots hotplug events to libsas, the hotplug
events may pending in the workqueue like
shost->work_q
new work[PORTE_BYTES_DMAED] --> |[PHYE_LOSS_OF_SIGNAL][PORTE_BYTES_DMAED] -> processing
|<-------wait worker to process-------->|
In this case, a new PORTE_BYTES_DMAED event coming, libsas try to queue
it to shost->work_q, but this work is already pending, so it would be
lost. Finally, libsas delete the related sas port and sas devices, but
LLDD driver expect libsas add the sas port and devices(last sas event).
This patch use dynamic allocated work to avoid this issue.
Signed-off-by: Yijing Wang <wangyijing@huawei.com>
CC: John Garry <john.garry@huawei.com>
CC: Johannes Thumshirn <jthumshirn@suse.de>
CC: Ewan Milne <emilne@redhat.com>
CC: Christoph Hellwig <hch@lst.de>
CC: Tomas Henzl <thenzl@redhat.com>
CC: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jason Yan <yanaijie@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The status of SAS PHY is in sas_phy->enabled. There is an issue that the
status of a remote SAS PHY may be initialized incorrectly: if disable
remote SAS PHY through sysfs interface (such as echo 0 >
/sys/class/sas_phy/phy-1:0:0/enable), then reboot the system, and we
will find the status of remote SAS PHY which is disabled before is
1 (cat /sys/class/sas_phy/phy-1:0:0/enable). But actually the status of
remote SAS PHY is disabled and the device attached is not found.
In SAS protocol, NEGOTIATED LOGICAL LINK RATE field of DISCOVER response
is 0x1 when remote SAS PHY is disabled. So initialize sas_phy->enabled
according to the value of NEGOTIATED LOGICAL LINK RATE field.
Signed-off-by: chenxiang <chenxiang66@hisilicon.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Jason Yan <yanaijie@huawei.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The intend purpose here was to goto out if smp_execute_task() returned
error. Obviously something got screwed up. We will never get these link
error statistics below:
~:/sys/class/sas_phy/phy-1:0:12 # cat invalid_dword_count
0
~:/sys/class/sas_phy/phy-1:0:12 # cat running_disparity_error_count
0
~:/sys/class/sas_phy/phy-1:0:12 # cat loss_of_dword_sync_count
0
~:/sys/class/sas_phy/phy-1:0:12 # cat phy_reset_problem_count
0
Obviously we should goto error handler if smp_execute_task() returns
non-zero.
Fixes: 2908d778ab ("[SCSI] aic94xx: new driver")
Signed-off-by: Jason Yan <yanaijie@huawei.com>
CC: John Garry <john.garry@huawei.com>
CC: chenqilin <chenqilin2@huawei.com>
CC: chenxiang <chenxiang66@hisilicon.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Jason Yan
Ming Lei
Sebastian Andrzej Siewior
Kees Cook
James Bottomley
Bart Van Assche
Linus Torvalds
Hannes Reinecke
chenxiang