turn off logs?

[Martin Langhoff]

On Tue, Feb 8, 2011 at 7:58 PM, James Cameron <quozl at laptop.org> wrote:
> On Tue, Feb 08, 2011 at 12:14:56AM -0500, Martin Langhoff wrote:
>> On Tue, Feb 8, 2011 at 12:06 AM, James Cameron <quozl at laptop.org> wrote:
>> > the kernel has the right to convert the write into an
>> > effectively synchronous operation.
>>
>> I am not understanding... can you explain a bit more why it'll turn
>> synchronous?
>
> You isolated this as part of #10045 discussions on devel@ in May last
> year, and my tests confirmed it.  It is documented in kernel
> Documentation/vm.txt ... "process which is generating disk writes will
> itself start writing out dirty data."

Yeah, but before that behaves "effectively synchronous" a lot of other
stuff has to happen.

For example, on XO-1 we had to busy-wait on NAND IO. That was nasty,
and stalled the system while IO was happening. Still, it wasn't
'synchronous'.

So I assume you mean stalls.

On XO-1.5 I don't see whole OS stalls on small IO, nor synchronous behaviour.

> You're asking for the frequency of the behaviour.  It isn't high unless
> there is memory pressure.

Ok.

> On XO-1.5 with 10.1.3 ... using "strace -ff -e write" against olpc-dm,
> followed by use of Terminal, Write, and StopWatch activities, filtering
> by fd 1 and 2, generated 387 writes, of which only three took more than
> 1ms.  The longest was 24ms.
>
> Adding Record activity, and taking four photographs, and a video, the
> total writes is now 5920, and only nine took more than 1ms.  Most writes
> were below 33us.

So still, nothing happens with the small writes.

Big writes stall, and we are having trouble with "streaming" writes --
200% agreed. But I cannot read in your notes nor in the bug you
mention any real situation where small writes.

One easy way these days of generating memory pressure is to write
random data in /var/cache/yum :-)

> Since these logging writes happen in the middle of anything an activity
> does, such as animation or tool tips (I saw warnings about svg's being
> read), they can affect latency.

We can, and should, tighten logging levels --


> May I suggest rsyslogd then, it is well known and well isolated.  We
> already use Python's Logging facility, so we can change handler to
> SysLogHandler, and configure rsyslog.conf to write to some place
> readable by user olpc.

Sure. Sounds reasonable.



m
-- 
 martin.langhoff at gmail.com
 martin at laptop.org -- Software Architect - OLPC
 - ask interesting questions
 - don't get distracted with shiny stuff  - working code first
 - http://wiki.laptop.org/go/User:Martinlanghoff