Re: [PATCH v2 18/23] sys-rq: Add option to soft dump

From: Randy Dunlap
Date: Mon Nov 08 2010 - 17:33:46 EST


On Mon, 8 Nov 2010 14:27:59 -0800 Mike Waychison wrote:

> On Mon, Nov 8, 2010 at 1:09 PM, Randy Dunlap <randy.dunlap@xxxxxxxxxx> wrote:
> > On Mon, 08 Nov 2010 12:33:23 -0800 Mike Waychison wrote:
> >
> >> It is very useful to provide some means to force the kernel logs to make it out
> >> via the kmsg_oops implementations on the console.  Add a new option 'Y' to
> >> sysrq to allow dumping of logs to kmsg_dumper drivers.
> >>
> >> Signed-off-by: Mike Waychison <mikew@xxxxxxxxxx>
> >
> > Hi,
> >
> > Please add 'y' to Documentation/sysrq.txt .
>
> Ok.
>
> Looking at the docs though, sparc64 has 'y' mapped to "Show global CPU
> Registers". The only slot available is 'a' if the list is up to date.

'a' has a big fat warning not to use it.
'x' is powerpc-specific, just as 'y' is sparc64 specific.

DaveM, any comments/suggestions on these?

Looks like one day soon we'll have to make a way to allow more sysrq keys...


> An alternative may be to overload 'v' "Forcefully restores framebuffer
> console" and "Causes ETM buffer dump [ARM-specific]" as these actions
> seem to be about making crash data visible.
>
> >
> >> ---
> >>  drivers/char/sysrq.c |   14 +++++++++++++-
> >>  1 files changed, 13 insertions(+), 1 deletions(-)
> >>
> >> diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c
> >> index eaa5d3e..058d3c8 100644
> >> --- a/drivers/char/sysrq.c
> >> +++ b/drivers/char/sysrq.c
> >> @@ -41,6 +41,7 @@
> >>  #include <linux/oom.h>
> >>  #include <linux/slab.h>
> >>  #include <linux/input.h>
> >> +#include <linux/kmsg_dump.h>
> >>
> >>  #include <asm/ptrace.h>
> >>  #include <asm/irq_regs.h>
> >> @@ -395,6 +396,17 @@ static struct sysrq_key_op sysrq_unrt_op = {
> >>       .enable_mask    = SYSRQ_ENABLE_RTNICE,
> >>  };
> >>
> >> +static void sysrq_handle_softdump(int key)
> >> +{
> >> +     kmsg_dump(KMSG_DUMP_SOFT, NULL);
> >> +}
> >> +static struct sysrq_key_op sysrq_softdump_op = {
> >> +     .handler        = sysrq_handle_softdump,
> >> +     .help_msg       = "soft-dump(Y)",
> >> +     .action_msg     = "Trigger a soft dump",
> >> +     .enable_mask    = SYSRQ_ENABLE_DUMP,
> >> +};
> >> +
> >>  /* Key Operations table and lock */
> >>  static DEFINE_SPINLOCK(sysrq_key_table_lock);
> >>
> >> @@ -451,7 +463,7 @@ static struct sysrq_key_op *sysrq_key_table[36] = {
> >>       /* x: May be registered on ppc/powerpc for xmon */
> >>       NULL,                           /* x */
> >>       /* y: May be registered on sparc64 for global register dump */
> >> -     NULL,                           /* y */
> >> +     &sysrq_softdump_op,             /* y */
> >>       &sysrq_ftrace_dump_op,          /* z */
> >>  };
> >
> >
> > ---
> > ~Randy
> > *** Remember to use Documentation/SubmitChecklist when testing your code ***
> >
> --
> To unsubscribe from this list: send the line "unsubscribe linux-api" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html


---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/