On Sun, Sep 14, 2014 at 7:54 AM, Markus Trippelsdorf
<mar...@trippelsdorf.de> wrote:
> On 2014.09.14 at 07:09 +0200, Markus Trippelsdorf wrote:
>> Just noticed this today:
>>
>> Sep 14 06:51:57 x4 kernel: [sched_delayed] ^a4CE: hpet increased
>> min_delta_ns to 20115 nsec
>>
>> in hex:
>> 20 01 34 43 45 3A 20
>>
>> Must be a recent regression.
>
> It looks like a combination of commit 504d58745c9ca and commit
> 458df9fd4815b causes the issue.
> 458df9fd4815b changed the loglevel of printk_deferred to a hardcoded
> KERN_WARNING. And 504d58745c9ca changed the printk in
> kernel/time/clockevents.c to printk_deferred.
> But now the KERN_WARNING loglevel of printk_deferred in
> kernel/time/clockevents.c is redundant and responsible for the weird
> 01 34 character combination (KERN_SOH "4").
vprintk_emit() has:
if (in_sched)
text_len = scnprintf(text, sizeof(textbuf),
KERN_WARNING "[sched_delayed] ");
text_len += vscnprintf(text + text_len,
sizeof(textbuf) - text_len, fmt, args);
Hence if in_sched, "[sched_delayed] " is inserted before the printed
message. As most printed messages have a KERN_* prefix, this
prefix now ends up in the middle of the message, causing the issue.
A simple solution is to print a newline after the "[sched_delayed]", but
this will break the line.
Better, but more complicated, would be to remove the KERN_* prefix
(if it exists) from the next message if in_sched is set.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
