Date: Wed, 04 Mar 1998 19:37:59 -0800 From: Mike Smith <mike@smith.net.au> To: John Saunders <john.saunders@scitec.com.au> Cc: shimon@simon-shapiro.org, freebsd-current@FreeBSD.ORG Subject: Re: silo overflows (Was Re: 3.0-RELEASE?) Message-ID: <199803050338.TAA23878@dingo.cdrom.com> In-Reply-To: Your message of "Thu, 05 Mar 1998 14:16:45 %2B1100." <34FE191D.1D676AC3@scitec.com.au>
next in thread | previous in thread | raw e-mail | index | archive | help
> Simon Shapiro wrote: > > [snip] Every 1KB, you write to disk [snip] > > If you use an IDE disk then 10 to 1 that's the problem. Particularly > if the IDE driver is doing multi-sector transfers. Interrupts must be > disabled during IDE PIO transfers (or you can risk data corruption) > and the time taken to transfer, say 16 sectors, is significant. When > Linux supported Intel Triton busmaster IDE the occurance of FIFO > overflows disappeared on Triton boards. For what it's worth, interrupts don't actually appear to be disabled for PIO IDE transfers. (Feel free to check the source, FWIW.) I don't believe that interrupts for the serial ports are blocked by the wdintr() handler; they use "fast" handlers which are immune to blocking by the normal spl() mechanism. > The solution is to use either a SCSI disk system, or support > busmastering IDE. The real problem is the crazy IRQ priorities of > the PC architeture. This last hits the nail square on the head. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com 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?199803050338.TAA23878>