If a command status of CMD_PROTOCOL_ERR is received, this
information should be conveyed to the SCSI mid layer, not
dropped on the floor. CMD_PROTOCOL_ERR may be received
from the Smart Array for any commands destined for an external
RAID controller such as a P2000, or commands destined for tape
drives or CD/DVD-ROM drives, if for instance a cable is
disconnected. This mostly affects multipath configurations, as
disconnecting a cable on a non-multipath configuration is not
going to do anything good regardless of whether CMD_PROTOCOL_ERR
is handled correctly or not. Not handling CMD_PROTOCOL_ERR
correctly in a multipath configaration involving external RAID
controllers may cause data corruption, so this is quite a serious
bug. This bug should not normally cause a problem for direct
attached disk storage.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Cc: stable@vger.kernel.org
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
I think ioremap() ends up being equivalent to ioremap_nocache
by default, but we should signal our intent that these mappings
should be non-cacheable.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
In the abort handler, when asked to abort a command which
is not known to the driver, SUCCESS is returned, but the
diagnostic message incorrectly indicates the abort failed.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
It turns out Smart Array logical drives do not support target
reset and when the target reset fails, the logical drive will
be taken off line. Symptoms look like this:
hpsa 0000:03:00.0: Abort request on C1:B0:T0:L0
hpsa 0000:03:00.0: resetting device 1:0:0:0
hpsa 0000:03:00.0: cp ffff880037c56000 is reported invalid (probably means target device no longer present)
hpsa 0000:03:00.0: resetting device failed.
sd 1:0:0:0: Device offlined - not ready after error recovery
sd 1:0:0:0: rejecting I/O to offline device
EXT3-fs error (device sdb1): read_block_bitmap:
LUN reset is supported though, and is what we should be using.
Target reset is also disruptive in shared SAS situations,
for example, an external MSA1210m which does support target
reset attached to Smart Arrays in multiple hosts -- a target
reset from one host is disruptive to other hosts as all LUNs
on the target will be reset and will abort all outstanding i/os
back to all the attached hosts. So we should use LUN reset,
not target reset.
Tested this with Smart Array logical drives and with tape drives.
Not sure how this bug survived since 2009, except it must be very
rare for a Smart Array to require more than 30s to complete a request.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Cc: stable@vger.kernel.org
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Dial back the aggressiveness of the controller lockup detection thread.
Currently it will declare the controller to be locked up if it goes
for 10 seconds with no interrupts and no change in the heartbeat
register. Dial back this to 30 seconds with no heartbeat change, and
also snoop the ioctl path and if a firmware flash command is detected,
dial it back further to 4 minutes until the firmware flash command
completes. The reason for this is that during the firmware flash
operation, the controller apparently doesn't update the heartbeat
register as frequently as it is supposed to, and we can get a false
positive.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Signed-off-by: Mike Miller <mikem@beardog.cce.hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Use spinlocks with finer granularity in the submission and
completion paths to allow concurrent execution for multiple
reply queues. In particular, do not hold a spin lock while
submitting a request to the device, nor during most of the
interrupt handler.
Signed-off-by: Matt Gates <matthew.gates@hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Smart Arrays can support multiple reply queues onto which command
completions may be deposited. It can help performance quite a bit
to arrange for command completions to be processed on the same CPU
from which they were submitted to increase the likelihood of cache
hits.
Signed-off-by: Matt Gates <matthew.gates@hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
This is in order to smooth the way for upcoming changes to allow use of
multiple reply queues for command completions.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: Matt Gates <matthew.gates@hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
When aborting a command, the tag is supposed to be
specified as 64-bit little endian. However, some smart
arrays expect the tag of the command to be aborted to be
specified in a strange byte order. How to tell which sort
of Smart Array firmware we're dealing with is not obvious.
However, because of the way we construct our tags, the values
of any outstanding tag when specified with the "strange" byte
order will not collide with the value specified in the correct
order. That means we can safely attempt the abort both ways.
Signed-off-by: Stephen M. Cameron <stephenmcameron@gmail.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Instead of giving up after 3 immediate retries of driver initiated
commands, back off the rate of retries and retry a bunch more times.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Reviewed-by: Andi Shyti <andi.shyti@gmail.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
In shared SAS configurations we might get a busy status
during driver initiated commands (e.g. during rescan for
devices). We should retry the command in such cases rather
than giving up.
Signed-off-by: Matt Bondurant <Matthew.dav.bondurant@hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Default behavior for any CHECK CONDITION excepting a few special cases is to
print out certain parts of the sense buffer and the CDB. Default behavior
should be to print nothing and let the upper layers or applications decide what
to do about these. The same information is already available by setting the
appropriate bits of the scsi_logging_level kernel parameter or via
/proc/sys/dev/scsi/logging_level.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
pci_disable_device() disables the bus master bit and pci_enable_device does
not re-enable it. It needs to be enabled.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
There was code to skip "disabled" devices which was intended to
skip devices disabled in the BIOS, but it really just checks to
see if the device can write to host memory, which this is disabled
by pci_disable_device on driver unload, so this check has the effect
of preventing subsequent load of the driver. And devices disabled in
the BIOS don't show up at all anyway, so this check never made any
sense to begin with, and should be removed.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
As Jenx Axboe explained to me: "In earlier times (2.6.18 and pre, iirc), Linux
disabled IO and mem bars on pci_disable_device(). Now in newer kernel it does
not. And in the newer kernels you run into problems if you DON'T disable the
device on exit, since when it later loads the device is already in the enabled
state - and pci_enable_device() then does nothing. This typically screws
MSI/MSI-X." This is what the big scary comment that says pci_disable_device
does "something nasty" to smart arrays was evidently referring to.
If pci_disable_device is not called on driver rmmod, subsequently insmod'ing
the driver may in result in some cases fail to be able to receive interrupts,
esp. if other drivers are loaded between unloading and loading hpsa.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Use check_signature to find a signature in the mmio address.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Acked-by: Mike Miller <mike.miller@hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Some other older controllers also do have problems to perform a kdump.
Adding controllers to this list means that the driver will signal
this non-ability via a resettable flag correctly.
The unsupported list was created after a consultation with HP.
Signed-off-by: Tomas Henzl <thenzl@redhat.com>
Acked-by: Mike Miller <mike.miller@hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Use find_first_zero_bit to find the first cleared bit in a memory region.
This also includes the following minor changes.
- Use bitmap_zero
- Reduce unnecessary atomic bitops usage
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Acked-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Certain types of changes to devices should not be interpreted as a device
change that would cause the device to be removed and re-added. These include
RAID level and Firmware revision changes. However, these attribute changes DO
need to be reflected in the controller info structure's dev structure list, so
that sysfs and /proc info files for the devices will reflect the new values.
Signed-off-by: Scott Teel <scott.stacy.teel@hp.com>
Acked-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Reduce confusion and inaccuracy caused by dated naming of vars and functions
referring to external target devices.
CURRENT NAMING: PROPOSED NAMING:
"MSA2xxx devices" "external target devices"
msa2xxx_model ext_target_model
is_msa2xxx is_ext_target
add_msa2xxx_enclosure add_ext_target_dev
nmsa2xxx_enclosures n_ext_target_devs
Signed-off-by: Scott Teel <scott.teel@hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Driver limits SAS external target IDs to range 1-8.
Need to increase limit and clean up overlapping concepts of targets and paths
in the code.
There are several defined constants that control this:
HPSA_MAX_TARGETS_PER_CTLR 16
MAX_MSA2XXX_ENCLOSURES 32
HPSA_MAX_PATHS 8
We can condense this to one constant:
MAX_EXT_TARGETS 32
SAS switches allow for 8 connections, and there is capacity for 4 switches per
enclosure in largest blade enclosure type.
Signed-off-by: Scott Teel <scott.teel@hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
It should call hpsa_set_bus_target_lun rather
than individually setting bus, target and lun.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Some distros have a "rescan-scsi-bus.sh" script which depends on
SCSI REPORT LUNs not reporting something different than what the
driver tells the kernel, even if the driver uses scan_start and
scan_finished methods of the SCSI host template to override the
usual SCSI midlayer discovery code. Previously, 1 was added to
the LUN to make room to insert the RAID controller device at
LUN 0. Now, the RAID controller is moved to bus 3, and 1 is no
longer added to the LUN. However, SCSI REPORT LUNS on Smart Array
doesn't report physical devices like tape drives or auto-loaders
as it turns out, so those particular device types still won't match.
Generally the logical drives are reported first however, so at
least those should match.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Upgraded firmware on Smart Array P7xx (and some others) made them show up as
SCSI revision 5 devices and this caused the driver to fail to map MSA2xxx
logical drives to the correct bus/target/lun. A symptom of this would be that
the target ID of the logical drives as presented by the external storage array
is ignored, and all such logical drives are assigned to target zero,
differentiated only by LUN. Some multipath software reportedly does not deal
well with this behavior, failing to recognize different paths to the same
device as such.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: Scott Teel <scott.teel@hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Sometimes, for testing purposes (e.g. testing rmmod on a system
that normally boots using hpsa) it's nice to rename the driver
and split it into two drivers and restrict it to certain
controllers. This makes that easier.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
hpsa_register_scsi just calls hpsa_scsi_detect. Move
the guts of hpsa_scsi_detect into hpsa_register_scsi and
get rid of hpsa_scsi_detect.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
We had both h->max_sg_entries and h->maxsgentries in the per controller
structure which is terribly confusing. max_sg_entries was really
just a constant, 32, which defines how big the "block fetch table"
is, which is as large as the max number of SG elements embedded
within a command (excluding SG elements in chain blocks).
MAXSGENTRIES was the constant used to denote the max number of SG
elements embedded within a command, also a poor name.
So renamed MAXSGENTREIS to SG_ENTRIES_IN_CMD, and removed
h->max_sg_entries and replaced it with SG_ENTRIES_IN_CMD.
h->maxsgentries is unchanged, and is the maximum number of sg
elements the controller will support in a command, including
those in chain blocks, minus 1 for the chain block pointer..
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Don't call kthread_stop with a spin lock held and interrupts
disabled because kthread_stop will sleep waiting for the thread
to stop.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
The '5i' controller freezes when a kdump is attemted.
This patch admits it and adds the controller
to the unresetable list.
Signed-off-by: Tomas Henzl <thenzl@redhat.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
IRQF_SHARED is required for older controllers that don't support MSI(X)
and which may end up sharing an interrupt. All the controllers hpsa
normally supports have MSI(X) capability, but older controllers may be
encountered via the hpsa_allow_any=1 module parameter.
Also remove deprecated IRQF_DISABLED.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
The Windows driver .inf disables ASPM on hpsa devices. Do the same because the
selection of a non default ASPM policy can cause the device to hang.
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Cc: stable@kernel.org
Acked-by: Mike Miller <mike.miller@hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6: (45 commits)
[SCSI] Fix block queue and elevator memory leak in scsi_alloc_sdev
[SCSI] scsi_dh_alua: Fix the time inteval for alua rtpg commands
[SCSI] scsi_transport_iscsi: Fix documentation os parameter
[SCSI] mv_sas: OCZ RevoDrive3 & zDrive R4 support
[SCSI] libfc: improve flogi retries to avoid lport stuck
[SCSI] libfc: avoid exchanges collision during lport reset
[SCSI] libfc: fix checking FC_TYPE_BLS
[SCSI] edd: Treat "XPRS" host bus type the same as "PCI"
[SCSI] isci: overriding max_concurr_spinup oem parameter by max(oem, user)
[SCSI] isci: revert bcn filtering
[SCSI] isci: Fix hard reset timeout conditions.
[SCSI] isci: No need to manage the pending reset bit on pending requests.
[SCSI] isci: Remove redundant isci_request.ttype field.
[SCSI] isci: Fix task management for SMP, SATA and on dev remove.
[SCSI] isci: No task_done callbacks in error handler paths.
[SCSI] isci: Handle task request timeouts correctly.
[SCSI] isci: Fix tag leak in tasks and terminated requests.
[SCSI] isci: Immediately fail I/O to removed devices.
[SCSI] isci: Lookup device references through requests in completions.
[SCSI] ipr: add definitions for additional adapter
...
* 'for-3.2/drivers' of git://git.kernel.dk/linux-block: (30 commits)
virtio-blk: use ida to allocate disk index
hpsa: add small delay when using PCI Power Management to reset for kump
cciss: add small delay when using PCI Power Management to reset for kump
xen/blkback: Fix two races in the handling of barrier requests.
xen/blkback: Check for proper operation.
xen/blkback: Fix the inhibition to map pages when discarding sector ranges.
xen/blkback: Report VBD_WSECT (wr_sect) properly.
xen/blkback: Support 'feature-barrier' aka old-style BARRIER requests.
xen-blkfront: plug device number leak in xlblk_init() error path
xen-blkfront: If no barrier or flush is supported, use invalid operation.
xen-blkback: use kzalloc() in favor of kmalloc()+memset()
xen-blkback: fixed indentation and comments
xen-blkfront: fix a deadlock while handling discard response
xen-blkfront: Handle discard requests.
xen-blkback: Implement discard requests ('feature-discard')
xen-blkfront: add BLKIF_OP_DISCARD and discard request struct
drivers/block/loop.c: remove unnecessary bdev argument from loop_clr_fd()
drivers/block/loop.c: emit uevent on auto release
drivers/block/cpqarray.c: use pci_dev->revision
loop: always allow userspace partitions and optionally support automatic scanning
...
Fic up trivial header file includsion conflict in drivers/block/loop.c
When controller lockup condition is detected,
we should fail all outstanding commands and disable
the controller. This will enable multipath solutions
to recover gracefully.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
We weren't filling in the transfer length of the
flush cache command (it transfers 4 bytes of zeroes).
Firmware didn't seem to be bothered by this, but it
should be fixed.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
The currentsd[] array in hpsa_update_scsi_devices had room for
256 devices. The code was iterating over however many physical
and logical devices plus an additional number of possible external
MSA2XXX controllers, which together could potentially exceed 256.
We increased the size of the currentsd array to 1024 + 1024 + 32 + 1
elements to reflect a reasonable maximum possible number of devices
which might be encountered. We also don't just walk off the end
of the array if the array controller reports more devices than we
are prepared to handle, we just ignore the excessive devices.
Signed-off-by: Scott Teel <scott.teel@hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Rename HPSA_MAX_SCSI_DEVS_PER_HBA to HPSA_MAX_DEVICES
Signed-off-by: Scott Teel <scott.teel@hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Set the max hardware sectors in the SCSI host template to 8192
to allow for larger i/o's (8192 is the same limit the cciss
driver currently has.)
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
The P600 requires a small delay when changing states. Otherwise we may think
the board did not reset and we bail. This for kdump only and is particular
to the P600.
Signed-off-by: Mike Miller <mike.miller@hp.com>
Cc: stable@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The following warning message may be confusing to some users:
dev_warn(&pdev->dev, "Controller claims that "
"'Bit 2 doorbell reset' is "
"supported, but not 'bit 5 doorbell reset'. "
"Firmware update is recommended.\n");
Most users don't know or care what bit we may be hitting. Also change
"recommended" to "required."
Signed-off-by: Mike Miller <mike.miller@hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
If a physical device exposed to the OS by hpsa
is replaced (e.g. one hot plug tape drive is replaced
by another, or a tape drive is placed into "OBDR" mode
in which it acts like a CD-ROM device) and a rescan is
initiated, the replaced device will be added to the
SCSI midlayer with target and lun numbers set to -1.
After that, a panic is likely to ensue. When a physical
device is replaced, the lun and target number should be
preserved.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Cc: stable@kernel.org
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
The test to detect OBDR ("One Button Disaster Recovery")
cd-rom devices was comparing against uninitialized data.
Fixed by moving the test for the device to where the
inquiry data is collected, and uninitialized variable
altogether as it wasn't really being used.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Cc: stable@kernel.org
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6: (71 commits)
[SCSI] fcoe: cleanup cpu selection for incoming requests
[SCSI] fcoe: add fip retry to avoid missing critical keep alive
[SCSI] libfc: fix warn on in lport retry
[SCSI] libfc: Remove the reference to FCP packet from scsi_cmnd in case of error
[SCSI] libfc: cleanup sending SRR request
[SCSI] libfc: two minor changes in comments
[SCSI] libfc, fcoe: ignore rx frame with wrong xid info
[SCSI] libfc: release exchg cache
[SCSI] libfc: use FC_MAX_ERROR_CNT
[SCSI] fcoe: remove unused ptype field in fcoe_rcv_info
[SCSI] bnx2fc: Update copyright and bump version to 1.0.4
[SCSI] bnx2fc: Tx BDs cache in write tasks
[SCSI] bnx2fc: Do not arm CQ when there are no CQEs
[SCSI] bnx2fc: hold tgt lock when calling cmd_release
[SCSI] bnx2fc: Enable support for sequence level error recovery
[SCSI] bnx2fc: HSI changes for tape
[SCSI] bnx2fc: Handle REC_TOV error code from firmware
[SCSI] bnx2fc: REC/SRR link service request and response handling
[SCSI] bnx2fc: Support 'sequence cleanup' task
[SCSI] dh_rdac: Associate HBA and storage in rdac_controller to support partitions in storage
...
In a shared SAS setup, target devices may be reset by one of
several hosts, and outstanding commands on that device will be
completed to corresponding hosts with status of UNSOLICITED_ABORT.
Such commands should be retried instead of being treated as i/o
errors. Also fixed a nearby spelling error.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
This allows us to move duplicated code in <asm/atomic.h>
(atomic_inc_not_zero() for now) to <linux/atomic.h>
Signed-off-by: Arun Sharma <asharma@fb.com>
Reviewed-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This memcpy:
memcpy(cmd->sense_buffer, ei->SenseInfo,
ei->SenseLen > SCSI_SENSE_BUFFERSIZE ?
SCSI_SENSE_BUFFERSIZE :
ei->SenseLen);
The ei->SenseLen field is filled in by the Smart Array. For requests to
logical drives, it will not exceed 32 bytes, so should be ok, but for physical
requests it depends on the target device, not the Smart Array. It's conceivable
that this could exceed the 32 byte size of ei->SenseInfo. In that case, the memcpy
would read past the end of ei->SenseInfo, copying data from the next command,
as if it were sense data, or, if it happened to be the very last command in the
block of allocated commands, could fall off the end of the allocated area and
crash. I'm not aware of anyone ever encountering this behavior, but it could
conceivably happen. This bug was found by Coverity.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Not at all sure this is correct or appropriate to change,
but this seems odd.
Found via coccinelle script
@@
type T;
T* ptr;
expression E1;
@@
* memset(E1, 0, sizeof(ptr));
Signed-off-by: Joe Perches <joe@perches.com>
Acked-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <jbottomley@parallels.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Scott Teel <scott.stacy.teel@hp.com>
Signed-off-by: James Bottomley <jbottomley@parallels.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Just go straight to the soft-reset method instead.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <jbottomley@parallels.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <jbottomley@parallels.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
on driver load, if reset_devices is set, and the hard reset
attempts fail, try to bring up the controller to the point that
a command can be sent, and send it a soft reset command, then
after the reset undo whatever driver initialization was done to get
it to the point to take a command, and re-do it after the reset.
This is to get kdump to work on all the "non-resettable" controllers
(except 64xx controllers which can't be reset due to the potentially
shared cache module.)
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <jbottomley@parallels.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The bit-2-doorbell reset method seemed to cause (survivable) NMIs
on some systems and (unsurvivable) IOCK NMIs on some G7 servers.
Firmware guys implemented a new doorbell method to alleviate these
problems triggered by bit 5 of the doorbell register. We want to
use it if it's available.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <jbottomley@parallels.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
hpsa_scsi_setup at one time contained enough code to justify
its existence, but that time has passed.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <jbottomley@parallels.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
When waiting for the board to become "not ready"
don't print a message saying "waiting for board to
become ready" (possibly followed by a message saying
"failed waiting for board to become not ready". Instead,
it should be "waiting for board to reset" and "failed
waiting for board to reset."
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <jbottomley@parallels.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <jbottomley@parallels.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <jbottomley@parallels.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Detect failure of controller reset by noticing if the 32 bytes of
"driver version" we store on the hardware in the config table
fail to get zeroed out. Previously we noticed if the controller
did not transition to "simple mode", but this did not detect reset
failure if the controller was already in simple mode prior to
the reset attempt (e.g. due to module parameter hpsa_simple_mode=1).
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <jbottomley@parallels.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <jbottomley@parallels.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
This attribute, requested by Redhat, allows kexec-tools to know
whether the controller can honor the reset_devices kernel parameter
and actually reset the controller. For kdump to work properly it
is necessary that the reset_devices parameter be honored. This
attribute enables kexec-tools to warn the user if they attempt to
designate a non-resettable controller as the dump device.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
My first attempt was botched, got the wrong PCI Device ID
(used PCI_DEVICE_ID_HP_CISSE, should have been PCI_DEVICE_ID_HP_CISSF)
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
'!' has higher precedence than '&'. CFGTBL_ChangeReq is 0x1 so the
original code is equivelent to if (!doorbell_value) {...
Signed-off-by: Dan Carpenter <error27@gmail.com>
Acked-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
We can get completions left over from before the attempted reset which
will interfere with the kdump. Better to just not make the attempt in
that case.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Controller will transfer only 32-bits on completion if it
knows we are only using 32-bit tags. Also, some newer controllers
apparently (and erroneously) require that we only use 32-bit tags,
and that we inform the controller of this.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
It's not enough to simple avoid putting the board into performant
mode, as we have to set up the interrupts differently, etc. When
I originally tested this module parameter, I tested it incorrectly
without realizing it, and the driver was running in performant mode
the whole time unbeknownst to me.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Driver's internal queues should be FIFO, not LIFO.
This is a port of an almost identical patch from cciss by Jens Axboe.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
memset arg64 to zero in the passthrough ioctls to avoid leaking contents
of kernel stack memory to userland via uninitialized padding fields
inserted by the compiler for alignment reasons.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Thanks to Scott Teel for noticing this.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Need to take the lock while accessing the register to check to
see if config table changes have taken effect.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
This is to prevent hpsa from resetting older boards
which the cciss driver may be controlling.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
This is to conserve memory in a memory-limited kdump scenario
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
and use the doorbell reset method if available (which doesn't
lock up the controller if you properly save and restore all
the PCI registers that you're supposed to.)
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
After a reset, we should first wait for the board to become "not ready",
and then wait for it to become "ready", instead of immediately
waiting for it to become "ready", and do this waiting *after*
restoring PCI config space registers. Also, only wait 10 secs
for board to become "not ready" after a reset (it should quickly
become not ready.)
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
They are defined in hpsa_cmd.h
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Some low bits might have been set by the driver, causing
a message like this to come out:
[ 13.288062] ------------[ cut here ]------------
[ 13.293211] WARNING: at lib/dma-debug.c:803 check_unmap+0x1a1/0x654()
[ 13.300387] Hardware name: ProLiant DL180 G6
[ 13.305335] hpsa 0000:06:00.0: DMA-API: device driver tries to free
DMA memory it has not allocated [device address=0x000000007f81e001]
[size=640 bytes]
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Otherwise, after doing a RAID level migration, the disk will be
disruptively removed and re-added as a different disk on rescan.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The firmware may have been updated, in which case, it's the same device,
and in that case, we do not want to remove and add the device, we want to
let it continue as is.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
PCI_DEVICE_ID_CISSF is defined as 323b in pci_ids.h but redefined as 3fff in
hpsa.c. The ID of 3fff will _never_ ship as a standalone controller. It is
intended only as part a complete storage solution. As such, this patch
removes the redefinition and the StorageWorks P1210m from the product table.
It also removes a duplicate line for the "unknown" controller support.
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The big kernel lock has been removed from all these files at some point,
leaving only the #include.
Remove this too as a cleanup.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Move the mid-layer's ->queuecommand() invocation from being locked
with the host lock to being unlocked to facilitate speeding up the
critical path for drivers who don't need this lock taken anyway.
The patch below presents a simple SCSI host lock push-down as an
equivalent transformation. No locking or other behavior should change
with this patch. All existing bugs and locking orders are preserved.
Additionally, add one parameter to queuecommand,
struct Scsi_Host *
and remove one parameter from queuecommand,
void (*done)(struct scsi_cmnd *)
Scsi_Host* is a convenient pointer that most host drivers need anyway,
and 'done' is redundant to struct scsi_cmnd->scsi_done.
Minimal code disturbance was attempted with this change. Most drivers
needed only two one-line modifications for their host lock push-down.
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Acked-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The doorbell reset initially appears to work correctly,
the controller resets, comes up, some i/o can even be
done, but on at least some Smart Arrays in some servers,
it eventually causes a subsequent controller lockup due
to some kind of PCIe error, and kdump can end up leaving
the root filesystem in an unbootable state. For this
reason, until the problem is fixed, or at least isolated
to certain hardware enough to be avoided, the doorbell
reset should not be used at all.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Some controllers might try to tell us they support 0 commands
in performant mode. This is a lie told by buggy firmware.
We have to be wary of this lest we try to allocate a negative
number of command blocks, which will be treated as unsigned,
and get an out of memory condition.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
There are things which need to be done in the intx
interrupt handler which do not need to be done in
the msi/msix interrupt handler, like checking that
the interrupt is actually for us, and checking that the
interrupt pending bit on the hardware is set (which we
weren't previously doing at all, which means old controllers
wouldn't work), so it makes sense to separate these into
two functions.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The 6402/6404 are two PCI devices -- two Smart Array controllers
-- that fit into one slot. It is possible to reset them independently,
however, they share a battery backed cache module. One of the pair
controls the cache and the 2nd one access the cache through the first
one. If you reset the one controlling the cache, the other one will
not be a happy camper. So we just forbid resetting this conjoined
mess.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Smart Array controllers newer than the P600 do not honor the
PCI power state method of resetting the controllers. Instead,
in these cases we can get them to reset via the "doorbell" register.
This escaped notice until we began using "performant" mode because
the fact that the controllers did not reset did not normally
impede subsequent operation, and so things generally appeared to
"work". Once the performant mode code was added, if the controller
does not reset, it remains in performant mode. The code immediately
after the reset presumes the controller is in "simple" mode
(which previously, it had remained in simple mode the whole time).
If the controller remains in performant mode any code which presumes
it is in simple mode will not work. So the reset needs to be fixed.
Unfortunately there are some controllers which cannot be reset by
either method. (eg. p800). We detect these cases by noticing that
the controller seems to remain in performant mode even after a
reset has been attempted. In those case, we proceed anyway,
as if the reset has happened (and skip the step of waiting for
the controller to become ready -- which is expecting it to be in
"simple" mode.) To sum up, we try to do a better job of resetting
the controller if "reset_devices" is set, and if it doesn't work,
we print a message and try to continue anyway.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Rationale for this is that I will also need to use this code
in fixing kdump host reset code prior to having the hba structure.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Rationale for this is that in order to fix the hard reset code used
by kdump, we need to use this function before we even have the per
HBA structure.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
We were previously only accepting HP boards.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Add 5 CCISSE smart array controllers
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
We have 32 (MAXSGENTRIES) scatter gather elements embedded
in the command. With all these, the total command size is
about 576 bytes. However, the last entry in the block fetch table
is 35. (the block fetch table contains the number of 16-byte chunks
the firmware needs to fetch for a given number of scatter gather
elements.) 35 * 16 = 560 bytes, which isn't enough. It needs to be
36. (36 * 16 == 576) or, MAXSGENTRIES + 4. (plus 4 because there's a
bunch of stuff at the front of the command before the first scatter
gather element that takes up 4 * 16 bytes.) Without this fix, the
controller may have to perform two DMA operations to fetch the
command since the first one may not get the whole thing.
Signed-off-by: Don Brace <brace@beardog.cce.hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
before trying to enter simple mode transport method.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
This patch removes unnecessary #define's from hpsa. The SCSI midlayer
handles all this for us.
Signed-off-by: Mike Miller <mike.miller@hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
This uses the scatter-gather chaining feature of Smart Array
controllers. 32 scatter-gather elements are embedded in the
"command list", and the last element in the list may be marked
as a "chain pointer", and point to an additional block of
scatter gather elements. The precise number of scatter gather
elements supported is dependent on the particular kind of
Smart Array, and is determined at runtime by querying the
hardware.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The intent of the scan thread was to allow a UNIT ATTENTION/LUN
DATA CHANGED condition encountered in the interrupt handler
to trigger a rescan of devices, which can't be done in interrupt
context. However, we weren't able to get this to work, due to
multiple such UNIT ATTENTION conditions arriving during the rescan,
during updating of the SCSI mid layer, etc. There's no way to tell
the devices, "stand still while I scan you!" Since it doesn't work,
there's no point in having the thread, as the rescan triggered via
ioctl or sysfs can be done without such a thread.
Signed-off-by: Mike Miller <mikem@beardog.cce.hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The SCSI status does not need to be shifted.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The use of the big kernel lock here appears
to be ancient cruft that is no longer needed.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
fix bug in adjust_hpsa_scsi_table which caused devices which have
changed size, etc. to do the wrong thing.
The problem was as follows:
The driver maintains its current idea of what devices are present
in the h->dev[] array. When it updates this array, it scans the
hardware, and produces a new list of devices, call it sd[], for
scsi devices.
Then, it compares each item in h->dev[] vs. sd[], and any items which
are not present sd it removes from h->dev[], and any items present
in sd[], but different, it modifies in h->dev[].
Then, it looks for items in sd[] which are not present in h->dev[],
and adds those items into h->dev[]. All the while, it keeps track
of what items were added and removed to/from h->dev[].
Finally, it updates the SCSI mid-layer by removing and adding
the same devices it removed and added to/from h->dev[]. (modified
devices count as a remove then add.)
originally, when a "changed" device was discovered, it was
removed then added to h->dev[]. The item was added to the *end*
of h->dev[]. And, the item was removed from sd[] as well
(nulled out). As it processed h->dev[], these newly added items
at the end of the list were encountered, and sd[] was searched,
but those items were nulled out. So they ended up getting removed
immediately after they were added.
The solution is to have a way to replace items in the h->dev[]
array instead of doing a remove + add. Then the "changed" items.
are not encountered a second time, and removed.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Mike Miller <mikem@beardog.cce.hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
use scan_start and scan_finished entry points for scanning and route
the CCISS_REGNEWD ioctl and sysfs triggering of same functionality
through hpsa_scan_start.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Matt Gates <matthew.gates@hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Matt Gates <matthew.gates@hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The commands should be retried, and this will make that happen,
instead of resulting in an i/o error.
Signed-off-by: Matt Gates <matthew.gates@hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The p1210m responsds to SCSI report LUNs, unlike traditional Smart
Array controllers. This means that the bus, target, and lun
assignments done by the driver cannot be arbitrary, but must match
what SCSI REPORT LUNS returns.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
and update pci_ids.h to include new PCI ID for StorageWorks 1210m variant.
Signed-off-by: Mike Miller <mikem@beardog.cce.hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
This is done by adding support for the so-called "performant mode"
(that's really what they called it). Smart Array controllers
have a mode which enables multiple command completions to be
delivered with a single interrupt, "performant" mode. We want to use
that mode, as some newer controllers will be requiring this mode.
Signed-off-by: Don Brace <brace@beardog.cce.hp.com>
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: Mike Miller <mikem@beardog.cce.hp.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
