Date: Sun, 3 May 1998 10:24:39 -0600 From: Nate Williams <nate@mt.sri.com> To: Matthew Dillon <dillon@backplane.com> Cc: freebsd-bugs@FreeBSD.ORG Subject: Re: i386/5398 Message-ID: <199805031624.KAA20372@mt.sri.com> In-Reply-To: <199805030540.WAA22835@freefall.freebsd.org> References: <199805030540.WAA22835@freefall.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> The following reply was made to PR i386/5398; it has been noted by GNATS. > > From: Matthew Dillon <dillon@backplane.com> > To: Poul-Henning Kamp <phk@critter.freebsd.dk> > Cc: John-Mark Gurney <gurney_j@efn.org>, freebsd-gnats-submit@FreeBSD.ORG > Subject: Re: i386/5398 > Date: Sat, 2 May 1998 22:34:59 -0700 (PDT) > ... > There's only one way to solve this problem and, unfortunately, I don't > have time to do it... and that is to add debugging code to the kernel > to do a histogram of the PC of the pushed context at the beginning of the > core interrupt code to see which diasble/enable pair is causing the > greatest number of latency problems. But, as I said, I don't have the > time to do it myself. I'll pipe in here and state that (at least older versions) XFree86 apparently disable interrupts for long periods of time, because on my box I'll get them for no apparent reason. If I use XIG's server, I do *NOT* see them, but my machine is more unstable since support for my card has got bit rot in it. I *never* see these sorts of problems on any machines which do not have X installed, and since we do not control the X distribution, we can not fix the bug. Based on previous history of this bug and personal experience using FreeBSD on machines with *lots* of serial ports (right now I've got a 486/66 getting beat up by 4 of them full-time), I've got to agree with the boys and say it's an issue with X. The remote end is a P166 and it can't keep up, but the 486/66 on the other end does fine with the same connection *AND* 3 others to faster boxes. Either the Xserver is disabling interrupts, doing something bad with the blitter or some other X interaction. In short, it's only partially FreeBSD's problem. It's FreeBSD problems because ultimately it makes FreeBSD look bad, but in reality this problem is one in XFree86. Note, I've heard that newer versions of XFree86 do things better/differently? Are you using the most recent version of XFree86? Are you using a video card that tends to hog the PCI bus because it's cranked way up? These are the *most* likely scenaries that will cause overflows on your hardware based on my own personal experience. This is not an affront to your intelligence or your time, just observations based on direct experience with FreeBSD and X with similar behavior. Nate To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-bugs" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199805031624.KAA20372>