crash_kexec gets called before kmsg_dump(KMSG_DUMP_OOPS) if
panic_on_oops is set, so the kernel log buffer is not stored
for this case.
This patch adds a KMSG_DUMP_KEXEC dump type which gets called
when crash_kexec() is invoked. To avoid getting double dumps,
the old KMSG_DUMP_PANIC is moved below crash_kexec(). The
mtdoops driver is modified to handle KMSG_DUMP_KEXEC in the
same way as a panic.
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: Simon Kagstrom <simon.kagstrom@netinsight.net>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The last messages which happens before a crash might contain interesting
information about the crash. This patch reworks mtdoops using the
kmsg_dumper support instead of a console, which simplifies the code and
also includes the messages before the oops started.
On oops callbacks, the MTD device write is scheduled in a work queue (to
be able to use the regular mtd->write call), while panics call
mtd->panic_write directly. Thus, if panic_on_oops is set, the oops will
be written out during the panic.
A parameter to specify which mtd device to use (number or name), as well
as a flag, writable at runtime, to toggle wheter to dump oopses or only
panics (since oopses can often be handled by regular syslog).
The patch was massaged and amended by Artem.
Signed-off-by: Simon Kagstrom <simon.kagstrom@netinsight.net>
Reviewed-by: Anders Grafstrom <anders.grafstrom@netinsight.net>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The main justification for this is to allow catching long messages
during a panic, where the top part might otherwise be lost since moving
to the next block can require a flash erase.
Signed-off-by: Simon Kagstrom <simon.kagstrom@netinsight.net>
Reviewed-by: Anders Grafstrom <anders.grafstrom@netinsight.net>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Make the maximum mtdoops partition size to be 8MiB. Indeed, it does
not make sense to use anything larger than that anyway. This limit
makes it possible to catch stupid mistakes where the user gives e.g.,
a rootfs partition to mtdoops (which will happily erase it).
Signed-off-by: Simon Kagstrom <simon.kagstrom@netinsight.net>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
This patch makes mtdoops keep track of used/unused pages in an array
instead of scanning the flash after a write. The advantage with this
approach is that it avoids calling mtd->read on a panic, which is not
possible for all mtd drivers.
Signed-off-by: Simon Kagstrom <simon.kagstrom@netinsight.net>
Reviewed-by: Anders Grafstrom <anders.grafstrom@netinsight.net>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
While looking into the mtdoops module, I've spotted several minor
imperfections. This patch addresses them. Namely:
1. Remove several trailing white-spaces and tabs
2. Check 'vmalloc()' return code straight away, not several lines
below in the 'mtdoops_console_init()' function.
3. Clean up printks - make them more consistent and use the same
code formatting style for them.
4. Remove silly style of putting brackets around everything in
"if" operators.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Cc: Simon Kagstrom <simon.kagstrom@netinsight.net>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
MTD's have both an index number and a name. Formerly,
the MTD selected for mtdoops was done only by index
number. With this patch, a name can be used instead.
For example, the kernel command line:
console=ttyMTD5
selects MTD 5 for mtdoops. But now this is also possible:
console=ttyMTD,log
which selects the MTD named "log" for mtdoops.
This has the advantage that partitions can be added or
removed that would affect the MTD index number but not the
name, without having to then change the kernel command line.
Signed-off-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Acked-by: Richard Purdie <rpurdie@linux.intel.com>
Acked-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
MTD internal API presently uses 32-bit values to represent
device size. This patch updates them to 64-bits but leaves
the external API unchanged. Extending the external API
is a separate issue for several reasons. First, no one
needs it at the moment. Secondly, whether the implementation
is done with IOCTLs, sysfs or both is still debated. Thirdly
external API changes require the internal API to be accepted
first.
Note that although the MTD API will be able to support 64-bit
device sizes, existing drivers do not and are not required
to do so, although NAND base has been updated.
In general, changing from 32-bit to 64-bit values cause little
or no changes to the majority of the code with the following
exceptions:
- printk message formats
- division and modulus of 64-bit values
- NAND base support
- 32-bit local variables used by mtdpart and mtdconcat
- naughtily assuming one structure maps to another
in MEMERASE ioctl
Signed-off-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
This makes the driver erase a block when it doesn't find any
existing saved log messages which is safer than assuming the
flash was already erased.
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Add a magic number to logged kernel oops messages so that they
can be more accurately detected rather than just having to rely
on the sequence number. This also allows easier detection of
saved crashes by userspace.
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Fix an off by one error in the mtdoops driver
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
struct oops_cxt needlessly became global.
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
drivers/mtd/mtdoops.c: In function ‘mtdoops_console_sync’:
drivers/mtd/mtdoops.c:329: error: implicit declaration of function ‘in_interrupt’
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
When the MTD provides a panic_write function, use it.
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Various minor cleaups to mtdoops:
* Don't support the mtd->erasesize < OOPS_PAGE_SIZE case
* Tweak printks and make the device mtdoops connects to more visible
* CON_PRINTBUFFER flag is uneeded
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Add a spinlock to ensure writes to the mtdoops buffer memory are
sequential and don't race.
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Writing to the flash needs to be done in a workqueue. The console
write functions may be called in any context which can lead to
lockups otherwise.
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Add further error return code checks to the mtdoops driver.
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
drivers/mtd/mtdoops.c: In function 'mtdoops_inc_counter':
drivers/mtd/mtdoops.c:109: warning: format '%d' expects type 'int', but argument 3 has type 'size_t'
drivers/mtd/mtdoops.c: In function 'mtdoops_console_sync':
drivers/mtd/mtdoops.c:277: warning: format '%d' expects type 'int', but argument 3 has type 'size_t'
someone buy Dave an x86_64 box.
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
mtdoops wasn't ensuring data was flushed to flash in crash situations
after recent changes in mainline kernels as tracking the
oops_in_progress variable was no longer enough. We can use the "unblank"
console call as a sync call to tell us to write out the buffer though.
Therefore add a sync function to mtdoops and call this when console
unblank events occur.
Signed-off-by: Richard Purdie <rpurdie@openedhand.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Kernel oops and panic messages are invaluable when debugging crashes.
These messages often don't make it to flash based logging methods (say a
syslog on jffs2) due to the overheads involved in writing to flash.
This patch allows you to turn an MTD partition into a circular log
buffer where kernel oops and panic messages are written to. The messages
are obtained by registering a console driver and checking
oops_in_progress. Erases are performed in advance to maximise the
chances of a saving messages.
To activate it, add console=ttyMTDx to the kernel commandline (where x
is the mtd device number to use).
Signed-off-by: Richard Purdie <rpurdie@openedhand.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
KOSAKI Motohiro