Date: Thu, 5 Mar 1998 15:29:05 +1100 From: Bruce Evans <bde@zeta.org.au> To: evanc@synapse.net, shimon@simon-shapiro.org Cc: current@FreeBSD.ORG, Matthew.Thyer@dsto.defence.gov.au, mike@smith.net.au Subject: Re: silo overflows (Was Re: 3.0-RELEASE?) Message-ID: <199803050429.PAA00299@godzilla.zeta.org.au>
next in thread | raw e-mail | index | archive | help
>Assume you are doing something like FTP of packages-current. >You are reading about 16 kilobytes per second, which is just about 1,142 >interrupts per second (assuming a 16 myte FIFO and 14 byte treshold. Every >1KB, you write to disk, so now we have 1,160 interrupts per second. Add >10% ACK on the FTP connection, HZ heartbeat, and you have 1,400 interrupts >per second. Say 2,000 interrupts/Sec. A P6-200 will be safe in this >regard. In a RT O/S this will give us about 500us per interrupt. What do >you think? I think 2000 interrupts/second is not many. A 486/33 can handle about 50000 sio interrupts/second, as it needs to do to handle just 2 16450 UARTs at 115200 bps bidirectional saturated. Running such a setup plus some other sources of interrupts was my standard test until I junked the 486/33 recently. OTOH, a P6/200 can't handle many more sio interrupts/second, since sio interrupt handling consists mainly of PIO, especially on fast systems, especially for buffered UARTs. The 486/33 was replaced by a K6/233 which has significantly (50%) slower ISA PIO than the 486/33. Bruce 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?199803050429.PAA00299>