Re: SCHED_ULE should not be the default
On Mon, Dec 12, 2011 at 09:06:04AM -0800, Steve Kargl wrote: Tuning kern.sched.preempt_thresh did not seem to help for my workload. My code is a classic master-slave OpenMPI application where the master runs on one node and all cpu-bound slaves are sent to a second node. If I send send ncpu+1 jobs to the 2nd node with ncpu's, then ncpu-1 jobs are assigned to the 1st ncpu-1 cpus. The last two jobs are assigned to the ncpu'th cpu, and these ping-pong on the this cpu. AFAICT, it is a cpu affinity issue, where ULE is trying to keep each job associated with its initially assigned cpu. While one might suggest that starting ncpu+1 jobs is not prudent, my example is just that. It is an example showing that ULE has performance issues. So, I now can start only ncpu jobs on each node in the cluster and send emails to all other users to not use those node, or use 4BSD and not worry about loading issues. Does it meet your expectations if you start (j modulo ncpu) = 0 jobs on a node? -- Scott LambertKC5MLE Unix SysAdmin lamb...@lambertfam.org ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: ZFS install from -CURRENT snapshot
On Sat, May 28, 2011 at 08:16:37PM -0400, Daniel Staal wrote: Ok, I think I've figured this out, but I want some confirmation before I start playing with actual hardware... The install steps on http://wiki.freebsd.org/RootOnZFS/GPTZFSBoot no longer appear to apply. The /dist/ directory doesn't exist on any of the current snapshots, so the actual install process given in that page can't work. snip If there's a doc someplace on this, feel free to point me to it. I just want a sanity-check before I dive in headfirst... If you want to end up with a mirrored ZFS only setup, have you seen ? http://blogs.freebsdish.org/pjd/2010/08/06/from-sysinstall-to-zfs-only-configuration/ That one was pretty straight-forward for me. You get to use sysinstall and don't have to manually install the OS. -- Scott LambertKC5MLE Unix SysAdmin lamb...@lambertfam.org ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: BTX halts installing 5.x
On Tue, Dec 02, 2003 at 11:11:18AM -0600, Josh Paetzel wrote: I'm trying to install FreeBSD on a Dual Pentium Pro 200mhz 512K cache system and I am getting the following error: CD Loader 1.01 SNIP BTX halted SNIP booting 4.9 cd works 5.anything doesn't Intel PR440FX Mainboard Dual Pentium Pro 200mhz 512K Cache CPUs Intel EtherExpress Pro LAN adapter Adaptec 7880 SCSI controller I have the above running 5.1-CURRENT, but I got there with a source upgrade from 4.8. Works fine. 4.3 gig IBM Wide SCSI-2 Disks I'm not actually using the SCSI controller. I, instead, have a Promise IDE RAID controller. The disks are fast enough for the CPU. I've tried a couple different video cards, Cirrus Logis and SIS PCI 320 Megs of Registered ECC SDRAM ATI Rage video and I'm currently down to 128MB 50ns ECC EDO (SDRAM)? RAM. I was getting random failures and the BIOS had logged some ECC evevents on the 64MB stick. I also have a : bt0: BT-948 FW Rev. 5.06I Ultra Narrow SCSI Host Adapter to run the : sa0: HP C1533A A612 Removable Sequential Access SCSI-2 device My Adaptec controller is completely idle. I've tried googling for a solution for this as well as perused the -CURRENT archives and haven't seen anything relevent to this issue. If you don't get it resolved, I can try booting a 5.x install cd on it's twin brother, 1 CPU, no Promise, no BT-948, which is sitting unpowered at the house, before I scavenge the RAM for this box. -- Scott LambertKC5MLE Unix SysAdmin [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: kernel panic trying to utilize a da(4)/umass(4) device with ohci(4)
On Thu, Nov 20, 2003 at 11:20:18AM +0100, Dag-Erling Smørgrav wrote: Alexander Leidinger [EMAIL PROTECTED] writes: I have no problem using my printer: [...] ulpt0: Hewlett-Packard DeskJet 895C, rev 1.00/1.00, addr 4, iclass 7/1 ulpt0: using uni-directional mode Sure, I can do that: ulpt0: Hewlett-Packard officejet d series, rev 1.10/1.00, addr 3, iclass 7/1 ulpt0: using bi-directional mode but it still won't print. Sometimes it works (albeit very, very slowly) but most of the time the process writing to ulpt0 just hangs. My OfficeJet 6110 works great with FreeBSD 5 as of a month ago. I don't print very often. It is prints quickly. My only problem with installation was because apsfilter' setup script didn't know about the 6110 driver parameter for the hpijs driver. Apsfilter seemed to have missed a couple of newer revs of the hpijs drivers. Unfortunately, the box isn't reachable from here or I'd post configs. -- Scott LambertKC5MLE Unix SysAdmin [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: SCSI cd errors
On Wed, Oct 22, 2003 at 07:59:32PM +0200, Bernd Walter wrote: On Wed, Oct 22, 2003 at 07:47:14PM +0200, Sven Esbjerg wrote: I upgraded my PC yesterday and have experienced hangs in boot. It seems like something has changed in the SCSI or GEOM code. I have an Advansys SCSI controller with two plextor CD drivers attached as well as a HP scanner. When the CD drives are detected the system hangs. Booting with verbose output shows errors and timeouts. adv0: AdvanSys ASC3030/50 SCSI controller port 0x9000-0x90ff mem 0xfa221000-0xfa2210ff irq 2 at device 7.0 on pci0 irq2 is bogus - there is no such thing since 80286 systems. Disabling irq2/9 in BIOS setup may help to work around. My Dual PPro200 seems to like IRQ 2: 14:25:32 Wed Oct 22 $ listirqs.sh IOAPIC #0 intpin 17 - irq 10 IOAPIC #0 intpin 18 - irq 2 IOAPIC #0 intpin 19 - irq 11 atkbd0: irq 1 atapci1: irq 2 atapci2: irq 2 bt0: irq 2 fxp0: irq 2 sio1 irq 3 sio0 irq 4 fdc0: irq 6 ahc0: irq 10 psmcpnp0: irq 12 ata0: irq 14 ata1: irq 15 It's been that way for a long time. -- Scott LambertKC5MLE Unix SysAdmin [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Bad performance
On Wed, Sep 17, 2003 at 08:31:52PM +0200, sebastian ssmoller wrote: On Wed, 2003-09-17 at 20:16, Don Bowman wrote: I wonder how hot your processor is? perhaps ACPI is throttling the clock back, either duty cycle or frequency. In your bios you can set the power mode, perhaps you can set 'full power always'. lmmon might show something. here is my lmmon output. Motherboard Temp Voltages 255C / 491F / 528KVcore1: +3.984V Vcore2: +3.984V Fan Speeds + 3.3V: +3.984V + 5.0V: +6.654V 1:0 rpm+12.0V: +15.938V 2:0 rpm-12.0V: -15.938V 3:0 rpm- 5.0V: -6.654V i'm not sure whether this output is correct : 255 C ?? Obviously, lmmon does not know how to read your environment monitoring data. For teperature on an ACPI enabled machine: http://people.freebsd.org/~hmp/acpi_temp.c -- Scott LambertKC5MLE Unix SysAdmin [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: ATAng problems
I thought I was going to miss out on the ATAng issues... I use a SanDisk CompactFlash PC Card Adapter to read my digital camera pictures. It worked fine before ATAng. Now I get: ata2: spurious interrupt - status=0xff error=0x00 reason=0x00 ata2: spurious interrupt - status=0xff error=0x00 reason=0x00 ata2: spurious interrupt - status=0xff error=0x00 reason=0x00 ata2: spurious interrupt - status=0xff error=0x00 reason=0x00 ata2: spurious interrupt - status=0xff error=0x00 reason=0x00 ... And after transferring some data the machine locks up. I updated last night from my week old -CURRENT but still have the same problem. Starting update: Sat Sep 6 21:27:09 EDT 2003 Finished update: Sat Sep 6 21:35:20 EDT 2003 http://www.lambertfam.org/~lambert/laptop/5/20030907-1331 has the verbose dmesg.boot and dmesg showing the insertion and removal of the card as well as other things. It doesn't lockup until I try to transfer data. When it locks up and I restart, it has typically transferred about 10 - 15 700K to 1.2M photos. -- Scott LambertKC5MLE Unix SysAdmin [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: ACPI battery state and resume not working on Inspiron 5150
On Sat, Aug 09, 2003 at 08:53:54PM -0400, David Gilbert wrote: Kevin == Kevin Oberman [EMAIL PROTECTED] writes: Kevin Sorry, Joe. By the way, are you suspending with acpiconf -s3? Kevin Have you tried creating a hibernation partition (slice) and Kevin using -s4? That appears to work better than suspend on most Kevin platforms that support it at all. Kevin Kevin -- R. Kevin Oberman, Network Engineer What does a hibernation partition look like? My dell has a 31 meg partition that I havn't touched and my FreeBSD partition. And can you determine what the hibernation partition should look like from an acpidump? I have a toshiba that only lists : hw.acpi.supported_sleep_state: S1 S4 S5 And I wiped the disk so fast after I bought it, that I've never seen the hibernation setup that originally came with it. -- Scott LambertKC5MLE Unix SysAdmin [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: DAC960
On Tue, Jul 01, 2003 at 09:25:30PM +0200, Dusan Kozic wrote: Hi, I have DAC960 RAID and SCSI CD-ROM. When I boot FreeBSD 4.8 on this machine, then it doesn't want to mount / partition autimatically. I don't know what this is doing on -current but anyway You don't say what has changed on the machine. Is this a new install? At some point between FreeBSD 2.2 and 4.6 (don't ask) I could install from the CD but not boot off the RAID disk later. Changing a DAC960 BIOS option from 2GB to 8GB made it boot. I have problems with top, vmstat and memcontrol list command too. triton# top top: nlist failed triton# vmstat vmstat: undefined symbols: _kmemstatistics _bucket _zlist triton# memcontrol list memcontrol: can't size range descriptor array: Operation not supported triton# That looks like the mundane kernel/world out of sync. One suggestion based on your dmesg... Have you disabled the bios on your Adaptec SCSI Adapter? If not, it may be loading before the DAC960 bios and only be looking for a bootable disk on that controller. Copyright (c) 1992-2003 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD 4.8-RELEASE #0: Thu Apr 3 10:53:38 GMT 2003 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC Timecounter i8254 frequency 1193182 Hz CPU: Pentium/P54C (133.31-MHz 586-class CPU) mlx0: Mylex version 2 RAID interface port 0x9000-0x907f irq 15 at device 2.0 on pci0 mlx0: DAC960P/PD, 2 channels, firmware 2.73-0-00, 4MB RAM mlxd0: Mylex System Drive on mlx0 mlxd0: 32768MB (67108864 sectors) RAID 5 (online) ahc0: Adaptec aic7870 SCSI adapter port 0xc00-0xcff mem 0x3000-0x3fff irq 11 at device 12.0 on pci0 aic7870: Wide Channel A, SCSI Id=7, 16/253 SCBs Waiting 15 seconds for SCSI devices to settle sa0 at ahc0 bus 0 target 4 lun 0 sa0: ARCHIVE IBM4326NP/RP !D 5500 Removable Sequential Access SCSI-2 device sa0: 7.812MB/s transfers (7.812MHz, offset 15) Manual root filesystem specification: fstype:device Mount device using filesystem fstype eg. ufs:/dev/da0s1a cd0 at ahc0 bus 0 target 6 lun 0 cd0: IBM CDRM00203\\000\\000\\000\\000\\000!K BZ26 Removable CD-ROM SCSI-2 device cd0: 5.000MB/s transfers (5.000MHz, offset 15) cd0: Attempt to query device size failed: NOT READY, Medium not present ? List valid disk boot devices empty line Abort manual input mountroot 8\^H \^Hufs:/dev/mlxd0s1a Mounting root from ufs:/dev/mlxd0s1a -- Scott LambertKC5MLE Unix SysAdmin [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: acpi patch for dell laptop?
On Mon, Jun 23, 2003 at 10:38:26AM -1000, Vincent Poy wrote: Speaking about ACPI, has anyone figured out how to close the lid without suspending? I remember before, when the lid was closed, it would suspend even with sysctl set to not suspend and the screen stays on and when you resume, you can type but the screen freezes and doesn't update. Setting hw.acpi.lid_switch_state=NONE in /etc/sysctl.conf keeps my Toshiba from suspending on lid close. Is that the sysctl you are referring to as not working? -- Scott LambertKC5MLE Unix SysAdmin [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
ACPI testing/debugging guide?
Is there some list of actions to preform and data to collect that would assist with getting the ACPI stuff lined out? I've read the acpiconf man page but don't know that it gives me any way to test for any specific functionality. I've been gradually piecing together the meaning of S1, S2, S3, S4, and S5 and figuring out that the *_(button|switch)_state sysctl oids specify which state to go to on activation of that button rather than being a descriptor of the current state of the buttons. I haven't figured out if the hw.acpi.thermal oids. I think maybe ACPI doesn't recognize the hardware. Is a thermal oid value of 3692 actually 36.92 celcius or some scale from 0x to 0x? Is battery.time supposed to be -1 when on AC? My resume from suspend doesn't work and I don't know how to tell where that is going wrong. I have no serial interface so I can't use a serial console when the LCD doesn't come on. I can live without suspend/resume functionality. You can't miss something you have never had. :-) I would like to be able to tell the fan to stay on anytime it has AC power since it is gradually cooking my lap. How can I help others improve the ACPI support without investing hours I don't have in learning enough reading the source to be useful finding and reporting problems? -- Scott LambertKC5MLE Unix SysAdmin [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: ACPI testing/debugging guide?
On Tue, Jun 17, 2003 at 06:40:26PM -0400, Barney Wolff wrote: On Tue, Jun 17, 2003 at 04:29:59PM -0600, M. Warner Losh wrote: : : ACPI records temperature in tenths of a Kelvin, if you can believe it :) I don't believe that. 369.2K is 96.2C, which is over 200F. That seems to hot to me. My laptop says 2982, which is either about 30C or 15.2C. Given how warm it is on my leg at the moment, I'd guess it is centi-Celcius. Maybe converted internally? Reading the source, it really is tenths Kelvin. Is the 3692 the actual temp, or the CRT, which I assume is the critical temp? In the output of sysctl hw.acpi.thermal.tz0 there are a bunch of values, only one of which is the current temp. The rest are thresholds - AC appears to mean active cooling (aka fan), PSV seems to mean passive. I was shocked at the dK values. However, combined with the fact that this thing is often too hot to touch with bare skin, I am temped to believe in dK. It is definately too hot. Hence, my desire for the ability to keep the fan on all the time. :-) Last night it was: hw.acpi.thermal.tz0.temperature: 3352 (62.05C 143.69F) Current values: hw.acpi.thermal.min_runtime: 0 hw.acpi.thermal.polling_rate: 30 hw.acpi.thermal.tz0.temperature: 3302 (57.05C 134.69F) hw.acpi.thermal.tz0.active: -1 hw.acpi.thermal.tz0.thermal_flags: 0 hw.acpi.thermal.tz0._PSV: 3692 hw.acpi.thermal.tz0._HOT: -1 hw.acpi.thermal.tz0._CRT: 3702 hw.acpi.thermal.tz0._ACx: 3692 3692 -1 -1 -1 -1 -1 -1 -1 -1 -- Scott LambertKC5MLE Unix SysAdmin [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: ACPI testing/debugging guide?
On Tue, Jun 17, 2003 at 03:54:13PM -0700, Hiten Pandya wrote: On Tue, Jun 17, 2003 at 04:29:59PM -0600, M. Warner Losh wrote: In message: [EMAIL PROTECTED] Dan Nelson [EMAIL PROTECTED] writes: : ACPI records temperature in tenths of a Kelvin, if you can believe it :) I don't believe that. 369.2K is 96.2C, which is over 200F. That seems to hot to me. My laptop says 2982, which is either about 30C or 15.2C. Given how warm it is on my leg at the moment, I'd guess it is centi-Celcius. Maybe converted internally? Why not the use http://people.freebsd.org/~hmp/acpi_temp.c 19:07:39 Tue Jun 17 $ ./acpi_temp System temperature = 333.2 K 60.0 C 140.0 F hw.acpi.thermal.tz0.temperature: 3332 but all that program does is read the oid and do the math just like I did on my TI-85. -- Scott LambertKC5MLE Unix SysAdmin [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]