Re: FreeBSD 7 buildworld error
Joshua Coombs wrote: As a follow up: I setup a fresh 7.0 install in a VM and played with it. Using my make.conf, I showed cc1 using 130MB when compiling insn-attrtab.c. I tweaked the VM conf down to 32MB of RAM and redid the compile, and other than taking forever due to swapping, it again churned past insn-attrtab.c using 130MB successfully. My next test is going to be simulating the swap setup I've got on my 386. Rather than one 384MB swap partition, I have three 128MB swap partitions, one per drive. My theory was I'd see some small boost when swapping by spreading the work load. Turns out it doesn't help, I bottleneck on the ISA bus, transferring 1.5MB/sec max no matter how I spread the work across the drives. When I re-partition I'll just pick one drive to hold all swap. That said, given how I butted up against a limit suspiciously close to the size of my swap partitions, rather than the 512MB data size limit reports, thats my next guess as to the cause, swap isn't allowing single processes to page into multiple swap pools? Josh C I can now confirm, using two 128MB swap partitions, plus a third 2GB partition, with 32MB of ram configured results in buildworld's failing, just using a 2GB swap partition works. Josh C ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 7 buildworld error
As a follow up: I setup a fresh 7.0 install in a VM and played with it. Using my make.conf, I showed cc1 using 130MB when compiling insn-attrtab.c. I tweaked the VM conf down to 32MB of RAM and redid the compile, and other than taking forever due to swapping, it again churned past insn-attrtab.c using 130MB successfully. My next test is going to be simulating the swap setup I've got on my 386. Rather than one 384MB swap partition, I have three 128MB swap partitions, one per drive. My theory was I'd see some small boost when swapping by spreading the work load. Turns out it doesn't help, I bottleneck on the ISA bus, transferring 1.5MB/sec max no matter how I spread the work across the drives. When I re-partition I'll just pick one drive to hold all swap. That said, given how I butted up against a limit suspiciously close to the size of my swap partitions, rather than the 512MB data size limit reports, thats my next guess as to the cause, swap isn't allowing single processes to page into multiple swap pools? Josh C ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 7 buildworld error
Derek Taylor wrote: On Fri, Feb 29, 2008 09:50 PM Sean [EMAIL PROTECTED] wrote: Check /etc/make.conf for CFLAGS, and if present remove it. This fixed the problem. Thank you. -Derek. I can confirm a failure in the same spot. What concerns me is in both my failure, and Derek's, the malloc is failing well below what limit says should be allowed. bin/cc/cc_int/../cc_tools -I/usr/src/gnu/usr.bin/cc/cc_int/../../../../contrib/gcc -I/usr/src/gnu/usr.bin/cc/cc_int/../../../../contrib/gcc/config -I/usr/src/gnu/usr.bin/cc/cc_int/../../../../contrib/gcclibs/include -I/usr/src/gnu/usr.bin/cc/cc_int/../../../../contrib/gcclibs/libcpp/include -I/usr/src/gnu/usr.bin/cc/cc_int/../../../../contrib/gcclibs/libdecnumber -I/usr/obj/usr/src/tmp/legacy/usr/include -c ../cc_tools/insn-attrtab.c cc1: out of memory allocating 136475392 bytes *** Error code 1 Stop in /usr/src/gnu/usr.bin/cc/cc_int. *** Error code 1 cyrix-dlc# limit cputime unlimited filesize unlimited datasize 524288 kbytes stacksize65536 kbytes coredumpsize unlimited memoryuseunlimited vmemoryuse unlimited descriptors 957 memorylocked unlimited maxproc 478 sbsize unlimited cc1 was only trying to request 130MB, my datasize is 512MB, why did it fail? Not to mention the additional question of why is cc1 ballooning up so badly on that file, when leading up to it it's using on average 12MB per instance? Prior versions of FreeBSD never took anywhere NEAR this much RAM to build. (I also don't like the 'remove all CFLAGS' fix suggestion, something is broken if a buildworld can't complete on a stock kernel with sane CFLAGS. In my case, I run -Os -pipe, am I now to understand that any CFLAGS setting is inappropriate?) Joshua Coombs ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 7 buildworld error
Jason Evans wrote: cc1 was only trying to request 130MB, my datasize is 512MB, why did it fail? It looks to me like gcc is trying to allocate a single 130MiB object, but you don't say anything about how much memory is already in use. It may well be that there are no remaining places in the memory map to place such a large object. Jason The machine is a pretty lightly loaded, but low physical mem box. 64mb ram, 384mb of swap normally, with an additional 2gb of swap to a file added for my buildworld attempt. Mem: 20M Active, 5604K Inact, 27M Wired, 2876K Cache, 14M Buf, 3064K Free Swap: 2432M Total, 21M Used, 2411M Free 15901 root1 1150 12640K 11628K RUN 2:22 86.96% cc1 Thats while building tree-ssa-ccp.c. How do I test out your theory? (Short of watching top while doing the build.) Josh C ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 7 on old SMP server?
Dominic Fandrey wrote: Why don't you just do the buildworld on another machine? As I don't use the machine as a primary workstation at the moment, it doesn't really matter to me if it takes 1 hour or 3 months, I start the buildworld in a screen'd shell and check on it periodically. When it finishes, I set aside 10 to 20 minutes to do an installworld/kernel mergemaster via serial console, 'just in case.' The closest it gets to annoying is when I start a buildworld based on a new security fix, and 3 days in the fix is revised. : ) And Oliver, you were correct, omitting -pipe didn't allow a buildworld to finish with 448MB ram + swap. Guess I get to repartition after all. My new plan is going to be sourcing a Gigabyte I-RAM, putting 4GB on it, and wiring up an external power supply for it. That plus an SATA to IDE adapter from NewEgg should let me run the I-RAM on my empty ATA controller, giving me 4GB of swap running as fast as the machine can take it. Depending on how it behaves, I may track down a 2MB ISA VGA card again and throw KDE4 on it just to be truly insane. They say it's much better about memory utilization now... :P Joshua Coombs ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 7 on old SMP server?
Oliver Fromme wrote: My hacked up 386 showed gains going from 6.2 to 7, the big win that I've noticed is scp throughput, I can sustain 40 to 45kbps where in the past the box walled at around 30kbps. Apache seems to have less latency responding to gets also. I'm just running a 7b3 kernel at the moment, I'm going to have to repartition with a lot more swap space to be able to build a 7 world (When did the ram use for a buildworld skyrocket?!) but even with this setup, 7 + ULE is a win for me. Are you saying you run FreeBSD 7 on an 80386(SX/DX) machine? How exactly did you hack it? As far as I know, support for 80386 processors was removed from FreeBSD a while ago. For a very short while with 6.0 I was tweaking the kernel to detect 386s as 486s, as well as using CPU_DISABLE_CMPXCHG and having ok luck. I've now got a Cyrix 486DrX-2 66 installed in place of my Am386DX-40, which supports CMPXCHG as well as ID'ing as a 486 so I don't need to do any tweaking to stay running. If I can get another viable 386DX box reassembled I'll see if 7 can be pressed into functioning on it as 6 could. Joshua Coombs ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 7 on old SMP server?
Brian wrote: Would that be a multiday buildworld? Brian 10 days on average. : ) I'm trying a 7 build without -pipe to see if I can squeak it through with 64MB RAM + 384MB swap, I'd much prefer not having to do a dump/restore to reallocate space for more swap. I'm almost thinking a non -pipe build may be quicker given the limited ram I have. Josh C ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 7 on old SMP server?
MaXX wrote: Hi list, I have an old netfinity 7000 (Quad PIII 500, 1Gb RAM) running 6.2 at the moment. I was wondering if it will take benefit of all the SMP improvement of 7 or is it too old? It runs a few postgresql databases, peak loads in the 2 to 4 range. As I do not have another equivalent machine, I prefer to ask before trying myself. I quite happy with the performances of my laptop (centrino duo) under 7, and hope my old server will be as happy as my laptop. thanks in advance, -- MaXX My hacked up 386 showed gains going from 6.2 to 7, the big win that I've noticed is scp throughput, I can sustain 40 to 45kbps where in the past the box walled at around 30kbps. Apache seems to have less latency responding to gets also. I'm just running a 7b3 kernel at the moment, I'm going to have to repartition with a lot more swap space to be able to build a 7 world (When did the ram use for a buildworld skyrocket?!) but even with this setup, 7 + ULE is a win for me. My other test case is a dual P3 1ghz box, running qmail, and moving from 6 to 7 has given me a decent boost in mail throughput capacity. (The box can now sustain approx 250 messages a second where before it would start backlogging before 200 messages a second.) http://www.x386.net/about.html Josh C ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Adaptec SCSI going nuts
Just got a dual Opteron system to play with amd64 builds of FreeBSD. Poking around to see how it behaived, I initially tossed FreeBSD 4.11 on it, and was surprised to see a dump of the scsi controller state at the end of the dmesg. Tried 6.0-Rel, both x86 and amd64, same behavior. Bumped up to 6-stable from yesterday, same behavior. After the dump, the system appears solid, no other errors, so I'm guessing it's just a problem with how the card is initialized, FreeBSD corrects it and moves on. Should I be more concerned, or consider it a quirk of the machine? Joshua Coombs Copyright (c) 1992-2005 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 6.0-RELEASE-p4 #0: Thu Jan 26 15:55:58 EST 2006 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/SMP Timecounter i8254 frequency 1193182 Hz quality 0 CPU: AMD Opteron(tm) Processor 246 (1994.67-MHz K8-class CPU) Origin = AuthenticAMD Id = 0xf5a Stepping = 10 Features=0x78bfbffFPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,MMX,FXSR,SSE,SSE2 AMD Features=0xe0500800SYSCALL,NX,MMX+,LM,3DNow+,3DNow real memory = 1073676288 (1023 MB) avail memory = 1024929792 (977 MB) ACPI APIC Table: A M I OEMAPIC FreeBSD/SMP: Multiprocessor System Detected: 2 CPUs cpu0 (BSP): APIC ID: 0 cpu1 (AP): APIC ID: 1 MADT: Forcing active-low polarity and level trigger for SCI ioapic0 Version 1.1 irqs 0-23 on motherboard ioapic1 Version 1.1 irqs 24-27 on motherboard ioapic2 Version 1.1 irqs 28-31 on motherboard acpi0: A M I OEMXSDT on motherboard acpi0: Power Button (fixed) pci_link0: ACPI PCI Link LNKA irq 9 on acpi0 pci_link1: ACPI PCI Link LNKB irq 10 on acpi0 pci_link2: ACPI PCI Link LNKC irq 11 on acpi0 pci_link3: ACPI PCI Link LNKD irq 15 on acpi0 Timecounter ACPI-fast frequency 3579545 Hz quality 1000 acpi_timer0: 24-bit timer at 3.579545MHz port 0x1008-0x100b on acpi0 cpu0: ACPI CPU on acpi0 acpi_throttle0: ACPI CPU Throttling on cpu0 cpu1: ACPI CPU on acpi0 pcib0: ACPI Host-PCI bridge port 0xcf8-0xcff on acpi0 pci0: ACPI PCI bus on pcib0 pcib1: ACPI PCI-PCI bridge at device 6.0 on pci0 pci3: ACPI PCI bus on pcib1 pci3: display, VGA at device 6.0 (no driver attached) fxp0: Intel 82551 Pro/100 Ethernet port 0xbc00-0xbc3f mem 0xfeafb000-0xfeafbfff,0xfeaa-0xfeab irq 18 at device 8.0 on pci3 miibus0: MII bus on fxp0 inphy0: i82555 10/100 media interface on miibus0 inphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto fxp0: Ethernet address: 00:e0:81:34:64:da isab0: PCI-ISA bridge at device 7.0 on pci0 isa0: ISA bus on isab0 atapci0: AMD 8111 UDMA133 controller port 0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xffa0-0xffaf at device 7.1 on pci0 ata0: ATA channel 0 on atapci0 ata1: ATA channel 1 on atapci0 pci0: serial bus, SMBus at device 7.2 (no driver attached) pci0: bridge at device 7.3 (no driver attached) pcib2: ACPI PCI-PCI bridge at device 10.0 on pci0 pci2: ACPI PCI bus on pcib2 ahd0: Adaptec AIC7902 Ultra320 SCSI adapter port 0x8000-0x80ff,0x7800-0x78ff mem 0xfc8fc000-0xfc8fdfff irq 24 at device 6.0 on pci2 ahd0: [GIANT-LOCKED] aic7902: Ultra320 Wide Channel A, SCSI Id=7, PCI-X 67-100Mhz, 512 SCBs ahd1: Adaptec AIC7902 Ultra320 SCSI adapter port 0x8800-0x88ff,0x8400-0x84ff mem 0xfc8fe000-0xfc8f irq 25 at device 6.1 on pci2 ahd1: [GIANT-LOCKED] aic7902: Ultra320 Wide Channel B, SCSI Id=7, PCI-X 67-100Mhz, 512 SCBs bge0: Broadcom BCM5704C Dual Gigabit Ethernet, ASIC rev. 0x2003 mem 0xfc8b-0xfc8b,0xfc8a-0xfc8a irq 24 at device 9.0 on pci2 miibus1: MII bus on bge0 brgphy0: BCM5704 10/100/1000baseTX PHY on miibus1 brgphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseTX, 1000baseTX-FDX, auto bge0: Ethernet address: 00:e0:81:34:65:28 bge1: Broadcom BCM5704C Dual Gigabit Ethernet, ASIC rev. 0x2003 mem 0xfc8e-0xfc8e,0xfc8d-0xfc8d irq 25 at device 9.1 on pci2 miibus2: MII bus on bge1 brgphy1: BCM5704 10/100/1000baseTX PHY on miibus2 brgphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseTX, 1000baseTX-FDX, auto bge1: Ethernet address: 00:e0:81:34:65:29 pci0: base peripheral, interrupt controller at device 10.1 (no driver attached) pcib3: ACPI PCI-PCI bridge at device 11.0 on pci0 pci1: ACPI PCI bus on pcib3 pci0: base peripheral, interrupt controller at device 11.1 (no driver attached) acpi_button0: Power Button on acpi0 atkbdc0: Keyboard controller (i8042) port 0x60,0x64 irq 1 on acpi0 atkbd0: AT Keyboard flags 0x1 irq 1 on atkbdc0 kbd0 at atkbd0 atkbd0: [GIANT-LOCKED] sio0: 16550A-compatible COM port port 0x3f8-0x3ff irq 4 flags 0x10 on acpi0 sio0: type 16550A fdc0: floppy drive controller (FDE) port 0x3f0-0x3f5,0x3f7 irq 6 drq 2 on acpi0 fdc0: [FAST] fd0: 1440-KB 3.5 drive on fdc0 drive 0 orm0: ISA Option ROM at iomem 0xc-0xc7fff on isa0 ppc0: cannot reserve I/O port range sc0: System console at flags 0x100 on isa0
Re: ZERO_COPY_SOCKETS
Igor Robul [EMAIL PROTECTED] wrote in message news:[EMAIL PROTECTED] On Mon, Dec 05, 2005 at 09:20:28PM -0500, Joshua Coombs wrote: #optionsZERO_COPY_SOCKETS What's the status of this in 6.0-R and 6-stable? The idea of avoiding memory copies when possible seems really appealing for my 386, on which any little boost is significant. : ) http://www.freebsd.org/releases/6.0R/relnotes-i386.html#KERNEL So you can't use FreeBSD-6.0 and greater on 80386 anymore. Ok, lemmie just point out that I'm already running 6.0, just to eliminate that issue. What I'm curious about is weather or not 6.0R's zero copy sockets are considered stable or not. Joshua Coombs ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: ZERO_COPY_SOCKETS
Vivek Khera [EMAIL PROTECTED] wrote in message news:[EMAIL PROTECTED] On Dec 6, 2005, at 7:37 AM, Joshua Coombs wrote: Ok, lemmie just point out that I'm already running 6.0, just to eliminate that issue. So you were not truthful about running a 386... No, I'm running on a 386. It's a bodged, banged up pile, but it's a 386. FreeBSD 6 can infact run on a subset of 386 hardware with the right massaging. What I'm curious about is weather or not 6.0R's zero copy sockets are considered stable or not. Works for me[tm]. Cool, I'll have to give it a go then. Joshua Coombs ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
ZERO_COPY_SOCKETS
#optionsZERO_COPY_SOCKETS What's the status of this in 6.0-R and 6-stable? The idea of avoiding memory copies when possible seems really appealing for my 386, on which any little boost is significant. : ) Joshua Coombs ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: 6.0 release date and stability
Brett Glass [EMAIL PROTECTED] wrote in message news:[EMAIL PROTECTED] The release schedule for FreeBSD 6.0, on the FreeBSD Web site, doesn't show a projected date for the finished product. How close is it? We are (believe it or not) still running and building production servers with 4.11, and would love to move to 6.0 (at least for uniprocessor systems; we may wait for 6.1 for SMP) if it is sufficiently stable and performs adequately. We're running our own tests on RC1, but don't have a lot of spare servers to try it on. So, it's worth asking: How stable is RC1 turning out to be on uniprocessor platforms? On SMP platforms? How is network and disk performance relative to 4.11? (When we tested 5.x, both network and file system performance were worse than that of 4.11.) With what known problems is 6.0 likely to ship, and of these which are likely to impact uniprocessor systems? Are any showstopper bugs merely being worked around for release? And, again, when is the likely release date? --Brett Glass Welp, it's in the RC stage, and I've not seen any reports of massive issues, so I imagine they'll move it through fairly quickly. For what it's worth, on UP, my 386 (stop laughing) is showing twice the inbound and outbound tcp throughput across multiple apps compared to 4.11. Disk throughput is slightly higher, but nothing super impressive. If 6.0 can show gains on a 386, that tells me there is some actual merit to the changes. Joshua Coombs ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
686B Testing
I see it's gone quiet on the 686B front, this due to a lack of testers and if so, what do I need to do to provide usefull info? Joshua Coombs [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-stable in the body of the message