Date: Thu, 05 Mar 1998 14:37:03 +1000 From: Stephen McKay <syssgm@dtir.qld.gov.au> To: Mike Smith <mike@smith.net.au> Cc: current@FreeBSD.ORG, syssgm@dtir.qld.gov.au Subject: Re: silo overflows (Was Re: 3.0-RELEASE?) Message-ID: <199803050437.OAA09160@troll.dtir.qld.gov.au> In-Reply-To: <199803050357.TAA24001@dingo.cdrom.com> from Mike Smith at "Wed, 04 Mar 1998 19:57:39 -0800" References: <199803050357.TAA24001@dingo.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wednesday, 4th March 1998, Mike Smith wrote: >> I didn't do enough investigation to even guess at the cause. I looked in >> the XFree86 server code for interrupt disabling but found nothing >> interesting. > >Does the server disable interrupts for extended periods of time? If it does, I couldn't find where. >> I was wondering though if any PCI experts knew if a PCI video card could >> starve the PCI-ISA bridge of cycles. Any experts out there? > >It's quite possible; it would depend on the PCI latency timer settings, >but generally the latency provided for there is only of the order of a >few microseconds. So, how would I print these latency timer settings from FreeBSD? Would I just get a meaningless jumble of numbers anyway? ;-) >I would be more interested in knowing whether the X >server did interrupt-related things specific to the Matrox driver... This was my first thought. I couldn't find any place that it disabled interrupts. There were routines to fiddle processor interrupts but I could find no code that called them. Stephen. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199803050437.OAA09160>