Re: CONFIG_PRINTK_TIME woes

[tony . luck]

>I turn off VC logging, and I turn off disk sync'ing, so it goes
>straight to the page cache.
>
>I really do need sub-microsecond timings when I put a lot of
>printk tracing into the stack.

Right now you only have microsecond timing. Although printk() gets
"nanosecond" resolution from sched_clock() it throws away the low
order bits when it outputs the ".%06lu ... nanosec_rem/1000"

>This is a useful feature, please do not labotomize it just because
>it's difficult to implement on ia64.  Just make a
>"printk_get_timestamp_because_ia64_sucks()" interface or something
>like that :-)

I'm just trying to assess how much work is needed at the moment. Both
you and Andrew have come up with scenarios where higher than jiffies
resolution is needed ... so it seems unlikely that I'll be allowed
to do much damage here.  Andrew's proposed:

 __attribute__((weak)) unsigned long long printk_clock(void)

patch would provide the hooks I need.

-Tony
-
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/