amd64: Fatal Trap 12 in high load situations
Hi I have a problem installing / upgrading FreeBSD 8.0-release on a new machine. The computers specs are: cpu: AMD Phenom II X4 board: Gigabyte MA790GPT-UD3H ram: 4x2GBytes DDR3/1333 hdd: 2xMaxtor STM31000528AS nic: 4x Intel(R) PRO/1000 and I'm running FreeBSD phenom2.localnet 8.0-RELEASE FreeBSD 8.0-RELEASE #0: Sat Nov 21 15:02:08 UTC 2009 r...@mason.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC amd64 During 'make buildworld' the machine regulary crashes with the following panic: Fatal trap 12: page fault while in kernel mode cpuid = 0; apic id = 00 fault virtual adress= 0x8 fault code = supervisor write data, page not present instruction pointer = 0x20:0x80578591 stack pointer = 0x28:0xff80eab94700 frame pointer = 0x28:0xff80eab94720 code segment= base 0x0, limit 0xf, type 0x1b = DPL 0, pres 1, long 1, def32 0, gran 1 processor eflags= interrupt enabled, resume IOPL = 0 current process = 22039 (uudecode) trap number = 12 panic: pagefault cpuid = 0 Uptime: 2h35m4s Physical memory: 8176 MB Dumping 2195 MB: 2180 2164 2148 2132 2116 or this one, its from last night and the machine wrote a minidump before locking up: Fatal trap 12: page fault while in kernel mode cpuid = 0; apic id = 00 fault virtual address = 0x8 fault code = supervisor write data, page not present instruction pointer = 0x20:0x80578591 stack pointer = 0x28:0xff80eab21500 frame pointer = 0x28:0xff80eab21520 code segment= base 0x0, limit 0xf, type 0x1b = DPL 0, pres 1, long 1, def32 0, gran 1 processor eflags= interrupt enabled, resume, IOPL = 0 current process = 5238 (objcopy) trap number = 12 panic: page fault cpuid = 0 Uptime: 1h15m45s Physical memory: 8176 MB Dumping 2148 MB: 2133 2117 2101 2085 2069 2053 2037 2021 2005 1989 1973 1957 1941 1925 1909 1893 1877 1861 1845 1829 1813 1797 1781 1765 1749 1733 1717 1701 1685 1669 1653 1637 1621 1605 1589 1573 1557 1541 1525 1509 1493 1477 1461 1445 1429 1413 1397 1381 1365 1349 1333 1317 1301 1285 1269 1253 1237 1221 1205 1189 1173 1157 1141 1125 1109 1093 1077 1061 1045 1029 1013 997 981 965 949 933 917 901 885 869 853 837 821 805 789 773 757 741 725 709 693 677 661 645 629 613 597 581 565 549 533 517 501 485 469 453 437 421 405 389 373 357 341 325 309 293 277 261 245 229 213 197 181 165 149 133 117 101 85 69 53 37 21 5 While the 'current process' is a different one at any crash, 'Fatal trap 12' and 'supervisor write data, page not present' are always the same, just as the instruction pointer 0x80578591 and virtual address. The most times, the kernel hangs completly so I have to hard reset the machine to get it responding again. About once in every ten crashs it is able to write a dump before rebooting or locking up. I have no knowledge in debugging the kernel (or debugging anything else) so I tried what I found in the handbook. This resulted in the following: kgdb kernel.debug /var/crash/vmcore.0 GNU gdb 6.1.1 [FreeBSD] Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type show copying to see the conditions. There is absolutely no warranty for GDB. Type show warranty for details. This GDB was configured as amd64-marcel-freebsd... Cannot access memory at address 0x400 (kgdb) list *0x80578591 0x80578591 is in lf_advlockasync (/usr/src/sys/kern/kern_lockf.c:604). 599 LIST_INIT(ls-ls_active); 600 LIST_INIT(ls-ls_pending); 601 ls-ls_threads = 1; 602 603 sx_xlock(lf_lock_states_lock); 604 LIST_INSERT_HEAD(lf_lock_states, ls, ls_link); 605 sx_xunlock(lf_lock_states_lock); 606 607 /* 608 * Cope if we lost a race with some other thread while (kgdb) backtrace #0 0x in ?? () Cannot access memory at address 0x0 So here are my questions: 1. Are there any known caveats or quirks regarding my hardware? 2. What can I do to further investigate this issue 3. Not fully on topic but might be related: The buildsystem recognizes my cpu as 686 class cpu wich is wrong. Are there any switches I can set in make.conf to have 'make' use the correct values? Currently I'm using a blank make.conf, meaning it is not present (as it is by default on a fresh installed system). If you need any more Information, I would be happy to provide it best regards, Sven -- 00 -- 00
Re: amd64: Fatal Trap 12 in high load situations
ms80 wrote: Hi I have a problem installing / upgrading FreeBSD 8.0-release on a new machine. The computers specs are: cpu: AMD Phenom II X4 board: Gigabyte MA790GPT-UD3H ram: 4x2GBytes DDR3/1333 hdd: 2xMaxtor STM31000528AS nic: 4x Intel(R) PRO/1000 [snip] So here are my questions: 1. Are there any known caveats or quirks regarding my hardware? 2. What can I do to further investigate this issue 3. Not fully on topic but might be related: The buildsystem recognizes my cpu as 686 class cpu wich is wrong. Are there any switches I can set in make.conf to have 'make' use the correct values? Currently I'm using a blank make.conf, meaning it is not present (as it is by default on a fresh installed system). [snip] I am using this motherboard with an AMD x4 630 Propus cpu and 4G Ram (2x2GB). I have done a basic overclock to 3.36GHz with the ram running at 1600MHz. This is my KDE4 desktop machine running FreeBSD 8 and all ports currently up to date. When selecting the RAM to put on this motherboard you should have consulted the list from Gigabyte for approved memory and chosen very carefully. The memory I actually have was not an exact line item from the list, but it was something extremely close and which was designed and manufactured for use with an AM3 socket motherboard. You will notice that some RAM today is designed for Intel P55 chipsets and Lynnfield processors while other RAM is designed specifically for AM3/AM2 socket use. It is probably not a good idea to disregard this during selection, e.g. memory not specifically meant for AM3 socket mobos may not function correctly. I also seem to recall seeing somewhere that this motherboard acquires limitations in overclocking when all 4 sockets are filled and the best overclocking results when only 2 sockets are in use. I am only using 2 sockets in a 2x2GB arrangement for 4GB RAM total. If you are not overclocking and have all 4 sockets filled you may not be able to go above 1066MHz memory multiplier. With only 2 sockets populated 1333MHz should be attainable. I believe your problem centers around memory. It may not be designed for AM3 socket and/or may not be able to handle a higher memory multiplier. When I first put this motherboard in I attempted to boot from an already installed OS with the memory multiplier set too high and saw numerous examples similar to what you are describing. Since I had bought 1600MHz memory I mistakenly set the multiplier too high. When I set it back to 1333MHz everything was fine. Either the memory multiplier is set too high for your RAM or it is just the wrong RAM to begin with. As far as make.conf goes I use: CPUTYPE?= k8 -Mike ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: amd64: Fatal Trap 12 in high load situations
Am Saturday 06 February 2010 11:38:25 schrob Michael Powell: ms80 wrote: Hi I have a problem installing / upgrading FreeBSD 8.0-release on a new machine. The computers specs are: cpu: AMD Phenom II X4 board: Gigabyte MA790GPT-UD3H ram: 4x2GBytes DDR3/1333 hdd: 2xMaxtor STM31000528AS nic: 4x Intel(R) PRO/1000 [snip] So here are my questions: 1. Are there any known caveats or quirks regarding my hardware? 2. What can I do to further investigate this issue 3. Not fully on topic but might be related: The buildsystem recognizes my cpu as 686 class cpu wich is wrong. Are there any switches I can set in make.conf to have 'make' use the correct values? Currently I'm using a blank make.conf, meaning it is not present (as it is by default on a fresh installed system). [snip] [snip too] I believe your problem centers around memory. It may not be designed for AM3 socket and/or may not be able to handle a higher memory multiplier. When I first put this motherboard in I attempted to boot from an already installed OS with the memory multiplier set too high and saw numerous examples similar to what you are describing. Since I had bought 1600MHz memory I mistakenly set the multiplier too high. When I set it back to 1333MHz everything was fine. Either the memory multiplier is set too high for your RAM or it is just the wrong RAM to begin with. As far as make.conf goes I use: CPUTYPE?= k8 -Mike Hi Thank you for your reply. I'm using two of this: OCZ3P1333LVAM4GK (OCZ DDR3 AMD Edition, rated for 1333MHz at 1.65V). My Board is rated for 1066 - 1600 MHz memory, and neither the website nor the manual say anything about limitations with memory. Anyway: I didn't overclock cpu or memory. I have stability and long life in mind, so I try to keep the hardware cool. During testing I underclocked the memory with 1066 and 800 MHz which didn't help: The machine crashes anyway. The only thing to note is that by default the board tries to set 1.5V DDR3 Voltage which is wrong, you have to set it to 1.65V manually. A faulty piece of hardware was the first thing I suspected and I tested among other things the memory with memtest86+. This runs fine for 4 passes, without any error. As far as I can tell, my memory subsystem is ok. As for make.conf: thanks, I will set this when I try again. with best regards Sven -- 00 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: amd64: Fatal Trap 12 in high load situations
On 6 February 2010 22:18, ms80 m...@dynamik.sytes.net wrote: Am Saturday 06 February 2010 11:38:25 schrob Michael Powell: ms80 wrote: Hi I have a problem installing / upgrading FreeBSD 8.0-release on a new machine. The computers specs are: cpu: AMD Phenom II X4 board: Gigabyte MA790GPT-UD3H ram: 4x2GBytes DDR3/1333 hdd: 2xMaxtor STM31000528AS nic: 4x Intel(R) PRO/1000 [snip] So here are my questions: 1. Are there any known caveats or quirks regarding my hardware? 2. What can I do to further investigate this issue 3. Not fully on topic but might be related: The buildsystem recognizes my cpu as 686 class cpu wich is wrong. Are there any switches I can set in make.conf to have 'make' use the correct values? Currently I'm using a blank make.conf, meaning it is not present (as it is by default on a fresh installed system). [snip] [snip too] I believe your problem centers around memory. It may not be designed for AM3 socket and/or may not be able to handle a higher memory multiplier. When I first put this motherboard in I attempted to boot from an already installed OS with the memory multiplier set too high and saw numerous examples similar to what you are describing. Since I had bought 1600MHz memory I mistakenly set the multiplier too high. When I set it back to 1333MHz everything was fine. Either the memory multiplier is set too high for your RAM or it is just the wrong RAM to begin with. As far as make.conf goes I use: CPUTYPE?= k8 -Mike Hi Thank you for your reply. I'm using two of this: OCZ3P1333LVAM4GK (OCZ DDR3 AMD Edition, rated for 1333MHz at 1.65V). My Board is rated for 1066 - 1600 MHz memory, and neither the website nor the manual say anything about limitations with memory. Anyway: I didn't overclock cpu or memory. I have stability and long life in mind, so I try to keep the hardware cool. During testing I underclocked the memory with 1066 and 800 MHz which didn't help: The machine crashes anyway. The only thing to note is that by default the board tries to set 1.5V DDR3 Voltage which is wrong, you have to set it to 1.65V manually. A faulty piece of hardware was the first thing I suspected and I tested among other things the memory with memtest86+. This runs fine for 4 passes, without any error. As far as I can tell, my memory subsystem is ok. As for make.conf: thanks, I will set this when I try again. with best regards Sven -- 00 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org What power supply do you have? How many watts? brand? If you have insufficient power, it may cause the system to become unstable. Regards David N ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: amd64: Fatal Trap 12 in high load situations
ms80 wrote: [snip] Thank you for your reply. I'm using two of this: OCZ3P1333LVAM4GK (OCZ DDR3 AMD Edition, rated for 1333MHz at 1.65V). My Board is rated for 1066 - 1600 MHz memory, and neither the website nor the manual say anything about limitations with memory. Anyway: I didn't overclock cpu or memory. I have stability and long life in mind, so I try to keep the hardware cool. During testing I underclocked the memory with 1066 and 800 MHz which didn't help: The machine crashes anyway. The only thing to note is that by default the board tries to set 1.5V DDR3 Voltage which is wrong, you have to set it to 1.65V manually. A faulty piece of hardware was the first thing I suspected and I tested among other things the memory with memtest86+. This runs fine for 4 passes, without any error. As far as I can tell, my memory subsystem is ok. Poking around in the OCZ forum for something I thought I recalled seeing somewhere before. I had seen reports that this board might be touchy about 1.65v memory. As far as the consensus goes with the small sampling I looked at, it seemed that 1.63 or 1.64 vdc was the sweet spot. Some claims are that it didn't want to work at anything either above or below this range. My RAM is OCZ3BE1600C8LV4GK (anything with BE or AM in the part number is designed specifically for AM3). I thought it was 1.5v, but since I didn't remember for certain I checked and it shows a spec for 1.65v. However, I rebooted so I could look at the CMOS/BIOS stuff and I have the System Voltage Control section set for AUTO for all. Then I looked in the PC Health Status page and on the DDR3 1.5V line it was only reading 1.600v. There seems to be a general feeling the newer AMD processors don't much care for higher memory voltages. Try lowering your voltages and see if it helps. I am successfully using this board with the CPU clock set at 240MHz, which with the x14 multiplier results in 3.36GHz operation. The Hypertransport and FSB bus speeds are 2400MHz and the memory is running at 1599MHz at the x6.66 multiplier. When I get the RAM up to 1680MHz is where I can get it to freeze. As long as I don't do that it is totally stable. -Mike ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: amd64: Fatal Trap 12 in high load situations
Am Saturday 06 February 2010 14:03:06 schrob David N: [snip] What power supply do you have? How many watts? brand? If you have insufficient power, it may cause the system to become unstable. Regards David N I tested with an Enermax EPR425AWT Pro82+ II, 425W wich was the psu I bought and intended to use with this computer. After stumbling across the instabilities I tested with a HEC 550TE-2WX 550W, but it made no difference, so either both are faulty / insufficient or the problem is something else. regards, Sven -- 00 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: amd64: Fatal Trap 12 in high load situations
Am Saturday 06 February 2010 14:37:16 schrob Michael Powell: ms80 wrote: [snip] There seems to be a general feeling the newer AMD processors don't much care for higher memory voltages. Try lowering your voltages and see if it helps. I am successfully using this board with the CPU clock set at 240MHz, which with the x14 multiplier results in 3.36GHz operation. The Hypertransport and FSB bus speeds are 2400MHz and the memory is running at 1599MHz at the x6.66 multiplier. When I get the RAM up to 1680MHz is where I can get it to freeze. As long as I don't do that it is totally stable. -Mike My CPU is an AMD Phenom II X 4 905e. Its (default) settings are: CPU Clock Ratio (Auto) 2500MHz CPU Northbridge Freq. (Auto) 2000MHz CPU Host Clock Contr. (Auto) HT Link Width (Auto) HT Link Freq. (Auto) 2000MHz Memory Clock(x6.66 ) 1333MHz I set the DDR3 voltage to auto, now it shows about 1.58V. Testing will take a little bit. Thank you for the hint. regards, Sven -- 00 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: amd64: Fatal Trap 12 in high load situations
Am Saturday 06 February 2010 10:17:05 schrob ms80: Hi I have a problem installing / upgrading FreeBSD 8.0-release on a new machine. The computers specs are: cpu: AMD Phenom II X4 board: Gigabyte MA790GPT-UD3H ram: 4x2GBytes DDR3/1333 hdd: 2xMaxtor STM31000528AS nic: 4x Intel(R) PRO/1000 and I'm running FreeBSD phenom2.localnet 8.0-RELEASE FreeBSD 8.0-RELEASE #0: Sat Nov 21 15:02:08 UTC 2009 r...@mason.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC amd64 During 'make buildworld' the machine regulary crashes with the following panic: Fatal trap 12: page fault while in kernel mode cpuid = 0; apic id = 00 fault virtual adress = 0x8 fault code= supervisor write data, page not present instruction pointer = 0x20:0x80578591 stack pointer = 0x28:0xff80eab94700 frame pointer = 0x28:0xff80eab94720 code segment = base 0x0, limit 0xf, type 0x1b = DPL 0, pres 1, long 1, def32 0, gran 1 processor eflags = interrupt enabled, resume IOPL = 0 current process = 22039 (uudecode) trap number = 12 panic: pagefault cpuid = 0 Uptime: 2h35m4s Physical memory: 8176 MB Dumping 2195 MB: 2180 2164 2148 2132 2116 or this one, its from last night and the machine wrote a minidump before locking up: Fatal trap 12: page fault while in kernel mode cpuid = 0; apic id = 00 fault virtual address = 0x8 fault code= supervisor write data, page not present instruction pointer = 0x20:0x80578591 stack pointer = 0x28:0xff80eab21500 frame pointer = 0x28:0xff80eab21520 code segment = base 0x0, limit 0xf, type 0x1b = DPL 0, pres 1, long 1, def32 0, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 5238 (objcopy) trap number = 12 panic: page fault cpuid = 0 Uptime: 1h15m45s Physical memory: 8176 MB Dumping 2148 MB: 2133 2117 2101 2085 2069 2053 2037 2021 2005 1989 1973 1957 1941 1925 1909 1893 1877 1861 1845 1829 1813 1797 1781 1765 1749 1733 1717 1701 1685 1669 1653 1637 1621 1605 1589 1573 1557 1541 1525 1509 1493 1477 1461 1445 1429 1413 1397 1381 1365 1349 1333 1317 1301 1285 1269 1253 1237 1221 1205 1189 1173 1157 1141 1125 1109 1093 1077 1061 1045 1029 1013 997 981 965 949 933 917 901 885 869 853 837 821 805 789 773 757 741 725 709 693 677 661 645 629 613 597 581 565 549 533 517 501 485 469 453 437 421 405 389 373 357 341 325 309 293 277 261 245 229 213 197 181 165 149 133 117 101 85 69 53 37 21 5 [snip] I know, its kind of stupid to reply to my own mails, but for reference: I edited loader.conf to contain ahci_load=YES So far it works: The machine compiled all night and didn't crash. I had the idea because yesterday while testing the proposal to lower the ddr3 voltages, the machine crashed again. Additionally to the panic I'm already used to, I had a second panic in my core.txt.1: This was a fatal trap 1, referencing (current process) to irq 22. I checked what irq22 is and it is my atapci (ATI IXP700/800 SATA300 controller). Googling a bit around I found a tutorial how to activate ahci. I gave it a try and as said above: So far it seems to work. regards Sven -- 00 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
