From owner-freebsd-current Wed Mar 4 20:34:14 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id UAA25446 for freebsd-current-outgoing; Wed, 4 Mar 1998 20:34:14 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from sendero.simon-shapiro.org (sendero-fxp0.Simon-Shapiro.ORG [206.190.148.34]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id UAA25410 for ; Wed, 4 Mar 1998 20:34:03 -0800 (PST) (envelope-from shimon@sendero-fxp0.simon-shapiro.org) Received: (qmail 19524 invoked by uid 1000); 5 Mar 1998 04:40:42 -0000 Message-ID: X-Mailer: XFMail 1.3-alpha-021598 [p0] on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <199803050323.TAA23786@dingo.cdrom.com> Date: Wed, 04 Mar 1998 20:40:42 -0800 (PST) Reply-To: shimon@simon-shapiro.org Organization: The Simon Shapiro Foundation From: Simon Shapiro To: Mike Smith Subject: Re: silo overflows (Was Re: 3.0-RELEASE?) Cc: Evan Champion , current@FreeBSD.ORG, Matthew Thyer Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On 05-Mar-98 Mike Smith wrote: ... > Interrupt rate is less significant in this issue than interrupt > latency. At 230kbps you have about 43us/character. Without a 16650 > datasheet I can't confirm whether the standard FIFO trigger levels are > doubled or not; if they are, you have 4 bytes or about 170us between > when the interrupt is generated and "too late". That means that the built-in fifo should go down from 14 to 12 bytes, or probably 10. Does the IOCTL that changes baud rate dividers do that too? > Because of the way the sio driver handles interrupts, you want to look > for code sections bracketed with disable_intr/enable_intr calls to find > possible culprits. If the driver disables interrupts then loses them, is this like shooting yourself in the foot and complain about the pain? :-) >> Every >> 1KB, you write to disk, so now we have 1,160 interrupts per second. > > There's no guarantee of this; particularly not if I/O clustering, async > or soft updates are enabled. Filesystems are mounted async during an > install. Of course. The purpuse of this napkin computation was to see, crudely, if we have congestion, or not. I think we can easily see that we do, and some overrun control is necessary. I simply do not know enough of the internals of FreeBSD to speculate any further. It has to be a ssoluble problem. We have more MIPS, more bandwidth and more control than a modem has. No? ---------- Sincerely Yours, Simon Shapiro Shimon@Simon-Shapiro.ORG Voice: 503.799.2313 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message