[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <48F90369174947C98F6AE82888FA60C5@realtek.com.tw>
Date: Tue, 14 Jul 2009 19:39:03 +0800
From: 林建安 <colin@...ltek.com.tw>
To: <linux-kernel@...r.kernel.org>
Subject: jiffies got lost because of printk
Hi there,
We found that system time is not synced with RTC time on our embedded
system.
This is because sometimes we print too many chars to console.
For these lines in the function, release_console_sem, of printk.c,
for ( ; ; ) {
spin_lock_irqsave(&logbuf_lock, flags);
wake_klogd |= log_start - log_end;
if (con_start == log_end)
break; /* Nothing to print */
_con_start = con_start;
_log_end = log_end;
con_start = log_end; /* Flush */
spin_unlock(&logbuf_lock);
call_console_drivers(_con_start, _log_end);
local_irq_restore(flags);
}
Interrupt will be disabled before calling call_console_drivers and enabled
after call_console_drivers finishes.
For the setting of com port baud rate 115200 and HZ 100, if there are more
than 115200/8/100 bytes waiting for output,
interrupt will be disabled for more than 1 HZ and timer interrupt will be
lost.
I know that printing too much in kernel level is not proper. But for 1000HZ,
interrupt may be lost even just output 15 chars.
Any idea?
Regards,
Colin
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists