Date: Mon, 18 Oct 2004 22:45:36 -0700 From: John-Mark Gurney <gurney_j@resnet.uoregon.edu> To: Sean McNeil <sean@mcneil.com> Cc: freebsd-current@freebsd.org Subject: Re: re0 fix that works with polling Message-ID: <20041019054536.GH22681@funkthat.com> In-Reply-To: <1098163629.57566.4.camel@server.mcneil.com> References: <1097896460.1123.2.camel@server.mcneil.com> <20041018011235.GB22681@funkthat.com> <1098135068.2861.6.camel@server.mcneil.com> <20041018230208.GE22681@funkthat.com> <1098141908.3424.6.camel@server.mcneil.com> <20041019030909.GF22681@funkthat.com> <1098163629.57566.4.camel@server.mcneil.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Sean McNeil wrote this message on Mon, Oct 18, 2004 at 22:27 -0700:
> > Have you tried w/o running X? also, tried the bios setting to not
> > assign an irq to the video card? I doubt that the extra interrupts
> > would cause a problem...
>
> I can try this if you think it worthwhile. Remember, though, that all I
> have to do to get it to work for the interrupt case is to decrease the
> count number from 0x800 to 0x200. For polling, it doesn't work without
> the timer interrupt because of apparent starvation. This sort of rules
> out an interrupt sharing issue as polling is interrupt-less doesn't it?
Hmmm... All of my testing was doing in interrupt mode, I haven't tried
polling... I'm going to send you a patch that will print a message to
the console if we run out of tx buffers... do you get a message from
vls about ENOBUFS? tried running truss/ktrace to see if any of the
send requests return ENOBUFS?
--
John-Mark Gurney Voice: +1 415 225 5579
"All that I will do, has been done, All that I have, has not."
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041019054536.GH22681>