[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <46AB631B.6070109@howardsilvan.com>
Date: Sat, 28 Jul 2007 08:39:07 -0700
From: Lee Howard <faxguy@...ardsilvan.com>
To: Alan Cox <alan@...rguk.ukuu.org.uk>
CC: Paul Fulghum <paulkf@...rogate.com>,
Tilman Schmidt <tilman@...p.cc>,
Robert Hancock <hancockr@...w.ca>,
linux-serial@...r.kernel.org, tytso@....edu, rmk@....linux.org.uk,
linux-kernel@...r.kernel.org
Subject: Re: serial flow control appears broken
Alan Cox wrote:
>>Curiously, the session at 38400 bps that skipped 858 bytes... coincided,
>>not just in sequence but also in precice timing within the session, with
>>a small but noticeable disk load that I caused by grepping through a
>>hundred session logs. (I can't reproduce it easily, though, because of
>>disk caching.)
>>
>>
>
>Can you send me a dmesg, there are some cases when high disk load can
>cause high interrupt latency in both 2.2 and 2.6 depending upon what is
>configured.
>
I've attached dmesg output. The os version I used yesterday to run
those tests was Debian 4.0r0 (kernel 2.6.18-4-686). It's still running,
and that's where I give you this dmesg output from.
> I don't think thats related to the main problem but it is
>worth knowing about hdparm -u1
>
# hdparm -u1 /dev/hda
/dev/hda:
setting unmaskirq to 1 (on)
unmaskirq = 1 (on)
#
After doing this I re-ran the 5 test sends at 115200 bps. The number of
lost bytes were: 0, 14, 8, 0, and 3. Compared with yesterday's 63, 5,
44, 48, and 2 this may indicate an improvement. Note also that in the
4th session where no bytes were lost there was still one element of
corrupt data as detected by the image decoder.
Thanks,
Lee.
View attachment "dmesg.out" of type "text/plain" (7309 bytes)
Powered by blists - more mailing lists