Re: [patch 0/3] kvm tool: Serial emulation overhaul

[Pekka Enberg]

On Mon, Dec 12, 2011 at 7:20 PM, Ingo Molnar <mingo@xxxxxxx> wrote:
>> I'm seeing 1.5 usecs per character for this little benchmark:
>
> Interesting - what do you see with the top -b measurement i
> posted - can you see similar slowdowns?

ssh-4.2# time top -d 0.01 -n 10 -b

[snip]

real	0m2.935
user	0m0.006s
sys	0m0.049s

That's 9.7 usecs per character which is roughly 6x slowdown.

It seems to be related to interrupt handling because if I bump up
TIMER_INTEVAL_NS to 10 msec:

--- a/tools/kvm/kvm.c
+++ b/tools/kvm/kvm.c
@@ -412,7 +412,7 @@ found_kernel:
        return ret;
 }

-#define TIMER_INTERVAL_NS 1000000      /* 1 msec */
+#define TIMER_INTERVAL_NS 10000000     /* 10 msec */

serial console output slows down by the same 10x factor:

real	0m24.631s
user	0m0.007s
sys	0m0.025s

Lowering the interval too 100 usec speeds things up but unfortunately
chokes the serial layer rather quickly:

    8 root      RT   [    6.759222] serial8250: too much work for irq4
0     0    0    0 S  0.0  0.0   0:00.00 migration/1
    9 root      20   0     0    0    0 S  0.0  0.0   0:00.00 kworker/1:0
   10 root      20   0     0    0    0 S  0.0  0.0   0:00.00 ksoftirqd/1
   11 root      20   0     0    0    0 S  0.0  0.0   0:00.00 kworker/0:1
   12 root      RT   0     0    0    0 S  0.0  0.0   0:00.00 watchdog/1
   13 root      RT   0     0    0    0 S  0.0  0.0   0:00.00 migration/2
   14 root      20   0     0    0    0 S  0[    6.759222] serial8250:
too much work for irq4
.0  0.0   0:00.00 kworker/2:0
   15 root      20   0     0    0    0 S  0.0  0.0   0:00.00 ksoftirqd/2

So i'm pretty sure it's some bug in hw/serial.c that's limiting
character output by interrupts.

                                Pekka
--
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/