17 Oct, 2016
1 commit
-
the mdc value can be quite big (like 65535), so we are in undefined
territory when doing the multiplication with the (also signed)
FCX_MAX_DATA_FACTOR as outlined by UBSAN:UBSAN: Undefined behaviour in drivers/s390/block/dasd_eckd.c:1678:14
signed integer overflow:
65535 * 65536 cannot be represented in type 'int'
CPU: 5 PID: 183 Comm: kworker/u512:1 Not tainted 4.7.0+ #150
Workqueue: events_unbound async_run_entry_fn
000000fb8b59f900 000000fb8b59f990 0000000000000002 0000000000000000
000000fb8b59fa30 000000fb8b59f9a8 000000fb8b59f9a8 000000000011732e
00000000000000a4 0000000000a309e2 0000000000a4c072 000000000000000b
000000fb8b59f9f0 000000fb8b59f990 0000000000000000 0000000000000000
0400000000d83238 000000000011732e 000000fb8b59f990 000000fb8b59f9f0
Call Trace:
([] show_trace+0x98/0xa8)
([] show_stack+0x70/0xf0)
([] dump_stack+0x86/0xb8)
([] ubsan_epilogue+0x28/0x70)
([] handle_overflow+0xde/0xf0)
([] dasd_eckd_check_characteristics+0x50a/0x550)
([] dasd_generic_set_online+0xba/0x380)
([] ccw_device_set_online+0x192/0x550)
([] dasd_generic_auto_online+0x2e/0x70)
([] async_run_entry_fn+0x70/0x270)
([] process_one_work+0x26a/0x638)
([] worker_thread+0x4a/0x658)
([] kthread+0x10c/0x110)
([] kernel_thread_starter+0x6/0xc)
([] kernel_thread_starter+0x0/0xc)As this is a runtime value there is actually no risk of any sane
compiler to detect and (ab)use this undefinedness, but let's make
the multiplication defined by making mdc unsigned.Signed-off-by: Christian Borntraeger
Acked-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
26 Sep, 2016
1 commit
-
If the DASD device gets blocked for any reason, e.g. because it is reserved
somewhere, the host_access_count sysfs entry or the host_access_list
debugfs entry may sleep forever. Make it interruptible so that userspace
can use ^C to abort the operation.Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
11 Aug, 2016
1 commit
-
On LPAR the read message buffer command should be executed on the path
it was received on otherwise there is a chance that the CUIR assignment
might be faulty and the wrong channel path is set online/offline.Fix by setting the path mask accordingly.
On z/VM we might not be able to do I/O on this path but there it does
not matter on which path the read message buffer command is executed.
Therefor implement a retry with an open path mask.Signed-off-by: Stefan Haberland
13 Jun, 2016
1 commit
-
It is possible to specify a user offset for the TOD clock, e.g. +2 hours.
The TOD clock will carry this offset even if the clock is synchronized
with STP. This makes the time stamps acquired with get_sync_clock()
useless as another LPAR migth use a different TOD offset.Use the PTFF instrution to get the TOD epoch difference and subtract
it from the TOD clock value to get a physical timestamp. As the epoch
difference contains the sync check delta as well the LPAR offset value
to the physical clock needs to be refreshed after each clock
synchronization.Signed-off-by: Martin Schwidefsky
16 Apr, 2016
2 commits
-
Implement new DASD IOCTL BIODASDCHECKFMT to check a range of tracks on a
DASD volume for correct formatting. The following characteristics are
checked:
- Block size
- ECKD key length
- ECKD record ID
- Number of records per trackSigned-off-by: Jan Höppner
Signed-off-by: Martin Schwidefsky -
With this feature, applications can query if a DASD volume is online
to another operating system instances by checking the online status of
all attached hosts from the storage server.Reviewed-by: Sebastian Ott
Reviewed-by: Heiko Carstens
Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
17 Mar, 2016
1 commit
-
Reorder lcu and device lock to get rid of the error-prone trylock
mechanism.The locking order is lcu lock -> device lock.
This protects against changes to the lcu device lists and enables us
to iterate over the devices, take the cdev lock and make changes to
the device structures.The complicated part is the summary unit check handler that gets an
interrupt on one device of the lcu that leads to structural changes of
the whole lcu itself. This work needs to be done even if devices on
the lcu disappear. So a device independent worker is used.
The old approach tried to update some lcu structures and set up the
lcu worker in the interrupt context with the device lock held.
But this forced the lock order "cdev lock -> lcu lock" that made it
hard to have the lcu lock held and iterate over all devices and change
them.The new approach is to schedule a device specific worker that gets
out of the interrupt context and rid of the device lock for summary
unit checks. This worker is able to take the lcu lock and schedule the
lcu worker that updates all devices. The time between interrupt and
worker execution is no problem because the devices in the lcu reject
all I/O in this time with an appropriate error. The dasd driver can
deal with this situation and re-drive the I/O later on.Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
07 Mar, 2016
3 commits
-
Convert dasd_device.private to be a void pointer to get
rid of a lot of explicit casts.Signed-off-by: Sebastian Ott
Reviewed-by: Stefan Haberland
Acked-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
Prepare for new format checking function by renaming functions and
moving reusable code to separate functions:- Move sanity checks into a new function and make it reusable.
- Move common format code to a new function called
dasd_eckd_format_process_data.
- Create the generic function dasd_eckd_format_build_ccw_req, which
itself will then decide what ccw request is being built according to
the input data. (with upcoming functionality).Signed-off-by: Jan Höppner
Signed-off-by: Martin Schwidefsky -
- Make sure a calling function can rely on data in fdata by resetting to
its initial values
- Move special treatment for track 0 and 1 to dasd_eckd_build_format
- Replace dangerous backward goto with a loop logic
- Add define for number that specifies the maximum amount of CCWs per
request and is used for format_step calculation
- Remove unused variableSigned-off-by: Jan Höppner
Signed-off-by: Martin Schwidefsky
03 Nov, 2015
3 commits
-
Path verification is either done via dasd_eckd_read_conf() which is
triggered during online processing and resume or via
do_path_verification_work() which is triggered after path events.
The dasd_eckd_read_conf() version added paths unconditionally and did
not check if the path mask was empty. This led to devices having the
disconnected stop flag set but a valid path mask. So they where not
working although they had paths validated successfully. After a resume
this state could even not be solved with additional paths added.Fix by checking for an empty path mask in dasd_eckd_read_conf() and
clearing the device stop bits for a newly added channel path.Reviewed-by: Sebastian Ott
Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky -
For a valid PAV assignment the DASD driver needs to notice possibly
changed configuration data. Thus the failing of read configuration
data should also fail the device restore to prevent invalid PAV
assignment. The failed device may get restored after additional paths
get available later on.
If the restore fails after the device was added to the lcu alias
handling it needs to be removed from the alias handling before exiting
the restore function.Reviewed-by: Sebastian Ott
Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky -
The configuration data is stored per path and also the first valid
configuration data per device. When dasd_eckd_read_conf is called
again after a path got lost the device configuration data is cleared
but possibly not the per path configuration data. This might lead to a
double free when the lost path gets operational again.Fix by clearing all per path configuration data when the first valid
configuration data is received and stored.Reviewed-by: Sebastian Ott
Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
14 Oct, 2015
1 commit
-
We often need to correlate an 8 bit path mask with the position
in a channel path array. Introduce and use pathmask_to_pos for
that task.Reviewed-by: Peter Oberparleiter
Signed-off-by: Sebastian Ott
Signed-off-by: Martin Schwidefsky
19 Aug, 2015
1 commit
-
Remove two more statements which always evaluate to 'false'.
These are more leftovers from the 31 bit era.Signed-off-by: Heiko Carstens
Reviewed-by: David Hildenbrand
Signed-off-by: Martin Schwidefsky
10 Aug, 2015
2 commits
-
This patch adds an enhanced detection for control unit initiated
reconfiguration request scope.
The first approach assumed the scope of the reconfiguration request
to be restricted to the path on which the message was received.
The enhanced approach determines the full scope of the reconfiguration
request by evaluating additional path and device selection information
contained in the reconfiguration message.Reviewed-by: Peter Oberparleiter
Reviewed-by: Sebastian Ott
Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky -
DASD path verification requires the usage of sleep_on_immediatly to
ensure that no other I/O request is blocking the recovery of
disconnected devices. But two concurrent path verification workers for
the same device may kill each others requests due to the usage of the
immediate sleep_on function. This may lead to unsuccessful path
verifications.Prevent that two parallel path verification workers conflict with
each other by implementing a device flag signalling a already running
worker.Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
15 Apr, 2015
1 commit
-
Fix ref counting for DASD devices leading to an inability to set a
DASD device offline.
Before a worker is scheduled the DASD device driver takes a reference
to the device. If the worker was already scheduled this reference was
never freed.
Fix by giving the reference to the DASD device free when
schedule_work() returns false.Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
25 Mar, 2015
1 commit
-
Remove the 31 bit support in order to reduce maintenance cost and
effectively remove dead code. Since a couple of years there is no
distribution left that comes with a 31 bit kernel.The 31 bit kernel also has been broken since more than a year before
anybody noticed. In addition I added a removal warning to the kernel
shown at ipl for 5 minutes: a960062e5826 ("s390: add 31 bit warning
message") which let everybody know about the plan to remove 31 bit
code. We didn't get any response.Given that the last 31 bit only machine was introduced in 1999 let's
remove the code.
Anybody with 31 bit user space code can still use the compat mode.Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky
09 Oct, 2014
1 commit
-
Add support for Control Unit Initiated Reconfiguration (CUIR) to
Linux, a storage server interface to reconcile concurrent hardware
changes between storage and host.Reviewed-by: Stefan Weinhuber
Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
01 Aug, 2014
1 commit
-
Rename enable_PAV to enable_pav.
Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
22 Jul, 2014
2 commits
-
Kernel panic or a hanging device during format if an alias device is
set offline or I/O errors occur.Omit the error recovery procedure for alias devices and do retries on
the base device with full erp.Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky -
Formatting of a previously formatted device is slower than newly
format a device when alias devices are available.
For already formatted devices the alias devices are not used for
formatting.Fix the alias handling for already formatted devices.
Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
01 Jan, 2014
1 commit
-
Needed to bring blk-mq uptodate, since changes have been going in
since for-3.14/core was established.Fixup merge issues related to the immutable biovec changes.
Signed-off-by: Jens Axboe
Conflicts:
block/blk-flush.c
fs/btrfs/check-integrity.c
fs/btrfs/extent_io.c
fs/btrfs/scrub.c
fs/logfs/dev_bdev.c
24 Nov, 2013
1 commit
-
More prep work for immutable biovecs - with immutable bvecs drivers
won't be able to use the biovec directly, they'll need to use helpers
that take into account bio->bi_iter.bi_bvec_done.This updates callers for the new usage without changing the
implementation yet.Signed-off-by: Kent Overstreet
Cc: Jens Axboe
Cc: Geert Uytterhoeven
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: "Ed L. Cashin"
Cc: Nick Piggin
Cc: Lars Ellenberg
Cc: Jiri Kosina
Cc: Paul Clements
Cc: Jim Paris
Cc: Geoff Levand
Cc: Yehuda Sadeh
Cc: Sage Weil
Cc: Alex Elder
Cc: ceph-devel@vger.kernel.org
Cc: Joshua Morris
Cc: Philip Kelleher
Cc: Konrad Rzeszutek Wilk
Cc: Jeremy Fitzhardinge
Cc: Neil Brown
Cc: Martin Schwidefsky
Cc: Heiko Carstens
Cc: linux390@de.ibm.com
Cc: Nagalakshmi Nandigama
Cc: Sreekanth Reddy
Cc: support@lsi.com
Cc: "James E.J. Bottomley"
Cc: Greg Kroah-Hartman
Cc: Alexander Viro
Cc: Steven Whitehouse
Cc: Herton Ronaldo Krzesinski
Cc: Tejun Heo
Cc: Andrew Morton
Cc: Guo Chao
Cc: Asai Thambi S P
Cc: Selvan Mani
Cc: Sam Bradshaw
Cc: Matthew Wilcox
Cc: Keith Busch
Cc: Stephen Hemminger
Cc: Quoc-Son Anh
Cc: Sebastian Ott
Cc: Nitin Gupta
Cc: Minchan Kim
Cc: Jerome Marchand
Cc: Seth Jennings
Cc: "Martin K. Petersen"
Cc: Mike Snitzer
Cc: Vivek Goyal
Cc: "Darrick J. Wong"
Cc: Chris Metcalf
Cc: Jan Kara
Cc: linux-m68k@lists.linux-m68k.org
Cc: linuxppc-dev@lists.ozlabs.org
Cc: drbd-user@lists.linbit.com
Cc: nbd-general@lists.sourceforge.net
Cc: cbe-oss-dev@lists.ozlabs.org
Cc: xen-devel@lists.xensource.com
Cc: virtualization@lists.linux-foundation.org
Cc: linux-raid@vger.kernel.org
Cc: linux-s390@vger.kernel.org
Cc: DL-MPTFusionLinux@lsi.com
Cc: linux-scsi@vger.kernel.org
Cc: devel@driverdev.osuosl.org
Cc: linux-fsdevel@vger.kernel.org
Cc: cluster-devel@redhat.com
Cc: linux-mm@kvack.org
Acked-by: Geoff Levand
20 Nov, 2013
1 commit
-
An I/O request that does not read or write full blocks cannot be
translated into a correct CCW or TCW program and should be rejected
right away. In particular the code that creates TCW requests will not
notice this problem and create broken TCWs that will be rejected by
the hardware.Signed-off-by: Stefan Weinhuber
Reference-ID: RQM1956
15 Oct, 2013
1 commit
-
The prefix command is used instead of a define extent to make use of
PAV alias devices during format. On some older storage servers the
prefix command may not be available and the IO request will fail.
Check for availability of prefix command and use define extent if
not available.Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
22 Aug, 2013
1 commit
-
The ECKD protocol supports reading of tracks with arbitrary format as
raw track images. The DASD device driver supports this in its
raw_track_access mode. In this mode it maps each track to sixteen 4096
byte sectors and rejects all requests that are not properly aligned to
this mapping.An application that wants to use a DASD in raw_track_access mode will
usually use direct I/O to make sure that properly aligned requests are
directly submitted to the driver. However, applications that are not
aware of this mode, e.g. udev, will encounter I/O errors.To make the use without direct I/O possible and avoid this kind of
alignment errors, we now pad unaligned read requests with a dummy
page, so that we can always read full tracks. Please note that
writing is still only possible for full track images that are properly
aligned.Signed-off-by: Stefan Weinhuber
Signed-off-by: Martin Schwidefsky
01 Jul, 2013
2 commits
-
This patch implements generic block layer timeout handling
callbacks for DASDs. When the timeout expires the respective
cqr is aborted.With this timeout handler time-critical request abort
is guaranteed as the abort does not depend on the internal
state of the various DASD driver queues.Signed-off-by: Hannes Reinecke
Acked-by: Stefan Weinhuber
Signed-off-by: Stefan Weinhuber
Signed-off-by: Martin Schwidefsky -
Instead of having the number of retries hard-coded in the various
functions we should be using a default retry value, which can
be modified via sysfs.Signed-off-by: Hannes Reinecke
Signed-off-by: Stefan Weinhuber
Signed-off-by: Martin Schwidefsky
27 Jun, 2013
1 commit
-
Signed-off-by: Thomas Meyer
Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky
17 Apr, 2013
1 commit
-
Reorganize format IO requests and enable usage of PAV.
Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
28 Feb, 2013
1 commit
-
Failfast bit was set incorrectly.
Use set_bit to enable failfast.Reviewed-by: Stefan Weinhuber
Signed-off-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
14 Feb, 2013
1 commit
-
Fix name clash with some common code device drivers and add "tod"
to all tod clock access function names.Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky
08 Jan, 2013
1 commit
-
Now that irq sum accounting for /proc/stat's "intr" line works again we
have the oddity that the sum field (first field) contains only the sum
of the second (external irqs) and third field (I/O interrupts).
The reason for that is that these two fields are already sums of all other
fields. So if we would sum up everything we would count every interrupt
twice.
This is broken since the split interrupt accounting was merged two years
ago: 052ff461c8427629aee887ccc27478fc7373237c "[S390] irq: have detailed
statistics for interrupt types".
To fix this remove the split interrupt fields from /proc/stat's "intr"
line again and only have them in /proc/interrupts.This restores the old behaviour, seems to be the only sane fix and mimics
a behaviour from other architectures where /proc/interrupts also contains
more than /proc/stat's "intr" line does.Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky
30 Nov, 2012
1 commit
-
If a channel path is cabled incorrectly and the device is suspended and
resumed the device may be inaccessible afterwards.
Make the path connection check not interrupt the resume callback there
could be other valid paths available.Signed-off-by: Stefan Haberland
Reference-ID: RQM 1262
Signed-off-by: Martin Schwidefsky
23 Nov, 2012
1 commit
-
Do not use more than one KERN_ per printk.
Signed-off-by: Sebastian Ott
Acked-by: Stefan Haberland
Signed-off-by: Martin Schwidefsky
26 Sep, 2012
3 commits
-
A common way to prepare a z/VM mini disk is to format the real device
with a z/VM tool like CPFMTXA and then define a mini disk that excludes
the first cylinder, i.e. the cylinder 0 of the virtual disk is located
at cylinder 1 of the real device.The DASD device driver will recognize such a mini disk as formatted, as
the uniform record layout on the disk matches that of an LDL formatted
device. However, the cylinder value in the 'count' field of the ECKD
records matches the geometry of the real device, and not that of the
mini disk, so I/O requests will fail with 'record not found' errors.To make the mini disk usable, it needs to be formatted with a tool like
dasdfmt. To enable tools like distribution installation tools to
recognize this situation, the DASD device driver should report such a
mini disk as 'not formatted'.
To this end we need to extend the device recognition code to check not
just for proper record sizes, but also for proper cylinder/head/record
values.Signed-off-by: Stefan Haberland
Reviewed-by: Stefan Weinhuber
Signed-off-by: Martin Schwidefsky -
Change -ENOSYS to -EOPNOTSUPP. Return value is used only internally.
Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
Add the eadm facility bits to the css characteristics and move
them to a new header.Signed-off-by: Sebastian Ott
Signed-off-by: Martin Schwidefsky