From owner-freebsd-hackers Fri Oct 25 15:25:17 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id PAA25824 for hackers-outgoing; Fri, 25 Oct 1996 15:25:17 -0700 (PDT) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id PAA25784 for ; Fri, 25 Oct 1996 15:25:09 -0700 (PDT) Received: from clem.systemsix.com (clem.systemsix.com [198.99.86.131]) by who.cdrom.com (8.7.5/8.6.11) with SMTP id NAA17134 for ; Fri, 25 Oct 1996 13:47:54 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by clem.systemsix.com (8.6.12/8.6.12) with SMTP id OAA17045; Fri, 25 Oct 1996 14:47:46 -0600 Message-Id: <199610252047.OAA17045@clem.systemsix.com> X-Authentication-Warning: clem.systemsix.com: Host localhost didn't use HELO protocol X-Mailer: exmh version 1.6.5 12/11/95 From: Steve Passe To: dg@root.com Cc: hackers@freefall.freebsd.org Subject: Re: ed0 timeouts In-reply-to: Your message of "Fri, 25 Oct 1996 13:29:02 PDT." <199610252029.NAA08813@root.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 25 Oct 1996 14:47:45 -0600 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk David, >>I currently am writing code for the FreeBSD SMP kernel that will allow >>us to use symmetric IO handling with all CPUs. This basically >>involves changing the kernel to use a device called the "IO APIC" >>in place of the 8259s. I now have code that does this, with all INT >>devices working properly EXCEPT for my ed NIC card. I have tried different >>slots and different cards, but I continue to get: >> >>[ date/time ] /kernel: ed0: device timeout >> >>I am occasionally loosing the INTs expected to be generated by ed_xmit(). >>The ed_watchdog() routine usually recovers gracefully (network actually locked >>twice since running this code: approx 2/3 weeks). Could someone explain >>why the ed0 cards might loose this INT? Any clues/insight appreciated! > > Not much I can say about this...there isn't anything special I can think of >about the 'ed' cards that would cause them to lose interrupts. What type of >card do you have? I currently am using: ed0 at 0x300-0x31f irq 10 maddr 0xcc000 msize 16384 on isa ed0: address 00:00:c0:e6:59:48, type WD8013EPC (16 bit) I also tried an 8216C Ultra with the same results. I am using the "hard config" jumper, I also tried a soft config @ IRQ5, 0xD8000 my current debug counters show: 0x2cbe calls to ed_xmit() 0x2xb7 tx INTs received 7 watchdog timeouts in an uptime of 1 hour, 37 minutes There is still some possibility that the 8254 clock is being mishandled. The APIC straps it to IRQ2 and ALOT of the kernel code believes it is at IRQ0, I could have missed one of these points. Is there anything possible as reguards 8254 problems??? -- Steve Passe | powered by smp@csn.net | FreeBSD -----BEGIN PGP PUBLIC KEY BLOCK----- Version: 2.6.2 mQCNAzHe7tEAAAEEAM274wAEEdP+grIrV6UtBt54FB5ufifFRA5ujzflrvlF8aoE 04it5BsUPFi3jJLfvOQeydbegexspPXL6kUejYt2OeptHuroIVW5+y2M2naTwqtX WVGeBP6s2q/fPPAS+g+sNZCpVBTbuinKa/C4Q6HJ++M9AyzIq5EuvO0a8Rr9AAUR tBlTdGV2ZSBQYXNzZSA8c21wQGNzbi5uZXQ+ =ds99 -----END PGP PUBLIC KEY BLOCK-----