Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 27 Sep 2002 21:36:27 +0200 (CEST)
From:      Dirk-Willem van Gulik <dirkx@webweaving.org>
To:        freebsd-hackers@freebsd.org
Subject:   4.6.2/RELENG_4: timeout in wi_seek and wi_cmd
Message-ID:  <20020927164601.P49453-100000@foem.leiden.webweaving.org>

next in thread | raw e-mail | index | archive | help

On a machine with either 4.6.2 of RELENG_4 as of today with:

3x	wiX: <Intersil Prism2.5> mem 0xfedf8000-0xfedf8fff irq 11 at device 19.X on pci0
	wiX: 802.11 address: 00:06:25:a7:a7:2a
	wiX: using RF:PRISM2.5 MAC:ISL3874A(Mini-PCI)
	wiX: Intersil Firmware: Primary 1.00.05, Station 1.03.04

I am getting regular errors like:
	wi1: timeout in wi_cmd 0x010b; event status 0x8000
and at timesL
	wi0: timeout in wi_seek to 146/3c; last status 403c

when doing repeated/contineous HTTP fetches of large file for a few hours
(at around 1-2Mbit, 1Gbyte or bigger) across just one interface (i.e.
we're not bridging or routing - but sending internal to /dev/null). Lots
of short/quick small files does not seem to trip the above.

And in particular once the latter starts happenig - it seems to get
gradually worse / interspeced with 0.5 -2 seconds hangs of all traffic on
all interfaces.

RELENG_4 seems to be slightly less of an issue than 4.6.2 - but no
fundamental changes/differences seen.

Is this firmware related ? is WI_TIMEOUT timeout soo long that it causes
other things to fail ? Any way to stop the 'hangs' from hitting other
interfaces. Or any other hints/insights ?

Note that we did have to increase mbuff sizes to 4096 to quell mbuff
errors/warnings.

Ta!

Dw




To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020927164601.P49453-100000>