From owner-freebsd-current Fri Nov 6 10:57:02 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id KAA12878 for freebsd-current-outgoing; Fri, 6 Nov 1998 10:57:02 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from skynet.ctr.columbia.edu (skynet.ctr.columbia.edu [128.59.64.70]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id KAA12856 for ; Fri, 6 Nov 1998 10:56:47 -0800 (PST) (envelope-from wpaul@skynet.ctr.columbia.edu) Received: (from wpaul@localhost) by skynet.ctr.columbia.edu (8.6.12/8.6.9) id NAA15525; Fri, 6 Nov 1998 13:55:12 -0500 From: Bill Paul Message-Id: <199811061855.NAA15525@skynet.ctr.columbia.edu> Subject: Re: Grrr... calcru: negative time blah blah blah To: phk@critter.freebsd.dk (Poul-Henning Kamp) Date: Fri, 6 Nov 1998 13:55:10 -0500 (EST) Cc: current@FreeBSD.ORG In-Reply-To: <2913.910336717@critter.freebsd.dk> from "Poul-Henning Kamp" at Nov 6, 98 08:18:37 am X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Of all the gin joints in all the towns in all the world, Poul-Henning Kamp had to walk into mine and say: > I'm still banging my head against this one. I'm working with a > handfull of people who has this problem, trying to identify what > the heck is happening. I can't reproduce it on any of my machines > here. > > If you have an interrupt storm which could lock out hardclock() > that will certainly screw things up badly. > > I have put some diagnostic patches at: > > http://www.freebsd.org/~phk/tc_diag.diff > > Please notice the patch to sysctl in there too, it is needed to > print out all of the "delta history buffer". > > Please try these patches out, they will not panic the machine if > the tests trigger, merely report some applicable details. > > If any of the tests trigger, please email me: > > /var/run/dmesg.boot > `dmesg` > `sysctl kern.timecounter` > `sysctl debug` I placed a file on ~wpaul on freefall called phk.tar.gz which contains the information from two tests, one with the apm0 device in the kernel (test1) and one without (test2). The behavior was the same in both cases. Basically all I did was boot the system and run xinit. The X server comes partway up, then crashes with a CPU time limit expired error. The whole dmesg is in the file. Here's part of it: Timecounter "i8254" frequency 1193182 Hz CPU: Pentium II (quarter-micron) (686-class CPU) Origin = "GenuineIntel" Id = 0x652 Stepping=2 Features=0x183fbff> real memory = 536870912 (524288K bytes) avail memory = 520142848 (507952K bytes) Programming 24 pins in IOAPIC #0 FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 0, version: 0x00040011, at 0xfee00000 cpu1 (AP): apic id: 1, version: 0x00040011, at 0xfee00000 io0 (APIC): apic id: 2, version: 0x00170011, at 0xfec00000 Probing for devices on PCI bus 0: Devices are: 3Com 3c905B XL integrated 10/100 ethernet Adaptec 7890/91 Ultra2 adapter Adaptec 7880 Ultra SCSI adapter Two QUANTUM VIKING II 9.1WLS 3506 drives on the Ultra2 controller NEC CD-ROM DRIVE:465 1.03 on the 7880 controller ATAPI IOMEGA ZIP on the secondary IDE controller 512MB RAM Here are the sysctl results for test1: sysctl kern.timecounter: kern.timecounter.frequency: 1193182 kern.timecounter.adjustment: 0 sysctl debug: debug.elf_trace: 0 debug.tc_diag_buffer: 11932 11932 11932 11937 11927 11931 11933 11932 11932 11931 11933 11931 11933 11932 11932 11932 11932 11932 89 740 11934 11933 11931 11930 11932 11932 11932 11932 11932 11932 11932 11931 11933 11931 11933 11932 11932 11931 11936 1192 9 11932 11932 11932 11932 11932 11932 11931 11933 11932 11932 11932 11932 11932 11932 11932 11932 11931 11932 11933 11931 11933 11931 11933 11931 11933 11931 11934 11931 11933 11930 11932 11935 11930 11932 11931 11933 11932 11932 11932 49982 50184 50259 1 1934 11930 11932 11936 11928 11932 11932 11931 11932 11933 11931 11933 11931 11933 11932 11932 11934 11930 11932 11932 11932 11932 11931 11933 11932 11928 11936 11932 11932 11932 11932 11933 11931 11932 11932 11937 11932 11927 11939 11926 11933 101 111 11 931 3356 3877 3885 5168 5189 5614 5731 5817 23370 23693 24128 24318 11936 11928 11930 11933 11932 11932 11931 11933 11932 11931 11933 73025 73176 73243 11938 11927 11931 11932 18499 18610 18682 11933 11934 11928 11932 11933 11932 11932 11932 11932 11931 1 1933 11932 11931 11933 11932 11932 11932 11932 11932 11932 11931 11933 11932 11932 11932 11932 11932 11931 11933 11932 11932 11931 11933 11932 11933 11931 11932 29733 31132 31195 11933 11936 11927 11932 11931 11932 11933 11931 11933 11931 11933 11932 1193 2 11932 11932 11932 11931 11933 11932 11932 11932 11932 11932 11932 11932 11934 11930 11933 11931 11932 11932 47338 47978 47986 48807 50401 51051 51072 51518 51730 1556 1658 1863 1950 2198 8500 8578 18966 19741 11920 11930 11931 11933 161 170 11932 11931 debug.tc_diag_index: 19695 debug.tc_diag_stop: 0 debug.tc_diag_maxforward: -30361 debug.fdexpand: 0 debug.debugger_on_panic: 1 debug.ttydebug: 0 debug.nchash: 65535 debug.ncnegfactor: 16 debug.numneg: 19 debug.numcache: 307 debug.vfscache: 1 debug.vnsize: 164 debug.ncsize: 36 debug.numvnodes: 302 debug.wantfreevnodes: 25 debug.freevnodes: 24 debug.disablecwd: 0 debug.bpf_bufsize: 4096 Here are the sysctl results for test2 (with no apm0 device): sysctl kern.timecounter: kern.timecounter.frequency: 1193182 kern.timecounter.adjustment: 0 sysctl debug: debug.elf_trace: 0 debug.tc_diag_buffer: 11932 11932 11932 11932 11931 11933 11933 11933 11930 11931 11933 11932 11932 11932 11933 11931 11932 11931 11934 11933 11932 11930 11932 11931 11934 11931 11933 11931 11932 11932 11932 11933 11931 11931 11933 37 45 11932 11933 11931 11932 11933 11933 11934 11928 11936 11927 11933 11932 11932 11932 11932 11932 11931 11936 11929 11933 11931 11934 11931 11931 11932 11932 11932 11933 11931 11932 11934 11928 11934 11932 11933 11931 11933 11932 11933 11930 11936 11928 11936 11933 11929 11 930 11936 11933 11927 44950 45524 45532 46430 48518 49258 49280 49747 49873 11933 11930 11932 11933 11931 11932 11932 11932 11936 11928 11932 11933 11932 11936 11928 11931 11936 11928 11932 11933 11931 11932 11932 11932 11932 11931 11933 11931 11933 11932 11932 11931 11933 11935 11931 11930 11933 11931 11931 11933 11932 11932 11932 11932 11932 11933 11931 11933 11931 11933 11934 11930 11933 11930 11933 11931 11932 11933 11931 11931 11933 11931 11933 11933 11931 11933 11934 11929 11937 11927 11933 11931 11 932 11933 11935 11929 11931 11936 11932 11928 11932 11932 11933 11932 11931 11932 11933 11931 11934 11931 11933 11931 11932 59 68 77 11932 11931 11932 11932 11932 11932 11934 11930 11932 11935 11929 11934 11930 11932 11932 11932 11932 11932 11932 11932 11 931 11933 11932 11932 11933 11931 11932 11936 11928 11933 11936 11927 11932 11931 11933 11933 11932 11931 11933 11934 11933 11928 11932 11932 11932 11932 11932 11932 11932 11932 11932 11933 11931 11932 11932 11932 11932 11932 11935 11930 11931 11932 11932 11933 11931 debug.tc_diag_index: 26463 debug.tc_diag_stop: 0 debug.tc_diag_maxforward: -41672 debug.fdexpand: 0 debug.debugger_on_panic: 1 debug.ttydebug: 0 debug.nchash: 65535 debug.ncnegfactor: 16 debug.numneg: 18 debug.numcache: 302 debug.vfscache: 1 debug.vnsize: 164 debug.ncsize: 36 debug.numvnodes: 309 debug.wantfreevnodes: 25 debug.freevnodes: 24 debug.disablecwd: 0 debug.bpf_bufsize: 4096 I don't know what any of these numbers mean. If you need me to run more tests, let me know. -Bill -- ============================================================================= -Bill Paul (212) 854-6020 | System Manager, Master of Unix-Fu Work: wpaul@ctr.columbia.edu | Center for Telecommunications Research Home: wpaul@skynet.ctr.columbia.edu | Columbia University, New York City ============================================================================= "It is not I who am crazy; it is I who am mad!" - Ren Hoek, "Space Madness" ============================================================================= To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message