Re: Problems with AMD64 and 8 GB RAM?

From: Greg 'groggy' Lehey (grog_at_FreeBSD.org)
Date: 03/31/05

  • Next message: Daniel O'Connor: "Re: Problems with AMD64 and 8 GB RAM?" 
    Date: Thu, 31 Mar 2005 11:24:29 +0930
To: Daniel O'Connor <doconnor@gsoft.com.au>

    On Thursday, 31 March 2005 at 10:32:33 +0930, Daniel O'Connor wrote:
    > On Thu, 31 Mar 2005 08:14, Greg 'groggy' Lehey wrote:
    >>> Have you run sysutils/memtest86 with the 8 GB?
    >>
    >> Heh. Difficult when the system doesn't run.
    >
    > You could try http://www.memtest86.com although that doesn't do >4Gb
    :(

    I'm pretty sure it's not the memory. I've tried each pair
    individually, and it's only when they're both in there together that
    it's a problem. And yes, I've tried them in each pair of slots.

    I now have dmesg output for verbose boots with both 4 GB and 8 GB
    memory. The complete dmesg output is at
    http://www.lemis.com/grog/Images/20050331/dmesg.4GB and
    http://www.lemis.com/grog/Images/20050331/dmesg.8GB. The diffs are at
    http://www.lemis.com/grog/Images/20050331/dmesg.diff. Here's a
    truncated summary:

    > --- dmesg.4GB Thu Mar 31 10:47:16 2005
    > +++ dmesg.8GB Thu Mar 31 10:52:32 2005
    > @@ -64,6 +10,7 @@
    > SMAP type=01 base=0000000000100000 len=00000000dfde0000
    > SMAP type=03 base=00000000dfee3000 len=000000000000d000
    > SMAP type=04 base=00000000dfee0000 len=0000000000003000
    > +SMAP type=01 base=0000000100000000 len=0000000100000000
    > Copyright (c) 1992-2005 The FreeBSD Project.
    > @@ -75,7 +22,7 @@
    > Calibrating clock(s) ... i8254 clock: 1193283 Hz
    > CLK_USE_I8254_CALIBRATION not specified - using default frequency
    > Timecounter "i8254" frequency 1193182 Hz quality 0
    > -Calibrating TSC clock ... TSC clock: 1603647337 Hz
    > +Calibrating TSC clock ... TSC clock: 1603647241 Hz
    > CPU: AMD Opteron(tm) Processor 242 (1603.65-MHz K8-class CPU)
    > Origin = "AuthenticAMD" Id = 0xf5a Stepping = 10
    > Features=0x78bfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,MMX,FXSR,SSE,SSE2>
    > @@ -90,11 +37,12 @@
    > L2 4KB data TLB: 512 entries, 4-way associative
    > L2 4KB instruction TLB: 512 entries, 4-way associative
    > L2 unified cache: 1024 kbytes, 64 bytes/line, 1 lines/tag, 16-way associative
    > -real memory = 3756916736 (3582 MB)
    > +real memory = 8589934592 (8192 MB)

    This is interesting in that it has gained 4.5 GB.

    > Physical memory chunk(s):
    > 0x0000000000001000 - 0x000000000009bfff, 634880 bytes (155 pages)
    > -0x0000000000a05000 - 0x00000000d95b7fff, 3636146176 bytes (887731 pages)
    > -avail memory = 3623817216 (3455 MB)
    > +0x0000000000a09000 - 0x00000000dfedffff, 3746394112 bytes (914647 pages)
    > +0x0000000100000000 - 0x00000001f0fcffff, 4043112448 bytes (987088 pages)
    > +avail memory = 7777177600 (7416 MB)
    > ACPI APIC Table: <VIAK8 AWRDACPI>
    > APIC ID: physical 0, logical 0:0
    > APIC ID: physical 1, logical 0:1
    > @@ -138,41 +86,12 @@
    > ioapic0: intpin 9 trigger: level
    > ioapic0: intpin 9 polarity: low
    > lapic0: Routing NMI -> LINT1
    > -A IRQ 3 (edge, high)
    > -ioapic0: intpin 4 -> ISA IRQ 4 (edge, high)
    > -ioapic0: intpin 5 -> ISA IRQ 5 (edge, high)
    > -ioapic0: intpin 6 -> ISA IRQ 6 (edge, high)
    > -ioapic0: intpin 7 -> ISA IRQ 7 (edge, high)
    > -ioapic0: intpin 8 -> ISA IRQ 8 (edge, high)
    > -ioapic0: intpin 9 -> ISA IRQ 9 (edge, high)
    > -ioapic0: intpin 10 -> ISA IRQ 10 (edge, high)
    > -ioapic0: intpin 11 -> ISA IRQ 11 (edge, high)
    > -ioapic0: intpin 12 -> ISA IRQ 12 (edge, high)
    > -ioapic0: intpin 13 -> ISA IRQ 13 (edge, high)
    > -ioapic0: intpin 14 -> ISA IRQ 14 (edge, high)
    > -ioapic0: intpin 15 -> ISA IRQ 15 (edge, high)
    > -ioapic0: intpin 16 -> PCI IRQ 16 (level, low)
    > -ioapic0: intpin 17 -> PCI IRQ 17 (level, low)
    > -ioapic0: intpin 18 -> PCI IRQ 18 (level, low)
    > -ioapic0: intpin 19 -> PCI IRQ 19 (level, low)
    > -ioapic0: intpin 20 -> PCI IRQ 20 (level, low)
    > -ioapic0: intpin 21 -> PCI IRQ 21 (level, low)
    > -ioapic0: intpin 22 -> PCI IRQ 22 (level, low)
    > -ioapic0: intpin 23 -> PCI IRQ 23 (level, low)
    > -MADT: Interrupt override: source 0, irq 2
    > -ioapic0: Routing IRQ 0 -> intpin 2
    > -ioapic0: intpin 2 trigger: edge
    > -ioapic0: intpin 2 polarity: high
    > -MADT: Interrupt override: source 9, irq 9
    > -ioapic0: intpin 9 trigger: level
    > -ioapic0: intpin 9 polarity: low
    > -lapic0: Routing NMI -> LINT1

    This stuff is puzzling. I suppose it could be related. Does anybody
    have any ideas?

    > lapic0: LINT1 trigger: edge
    > lapic0: LINT1 polarity: high
    > lapic1: Routing NMI -> LINT1
    > lapic1: LINT1 trigger: edge
    > lapic1: LINT1 polarity: high
    > -ioapic0 <Version 0.3> irqs 0-23 on motherboard
    > +ioapic0 <Version 0.0> irqs 0-23 on motherboard
    > cpu0 BSP:
    > ID: 0x00000000 VER: 0x00040010 LDR: 0x01000000 DFR: 0x0fffffff
    > lint0: 0x00010700 lint1: 0x00000400 TPR: 0x00000000 SVR: 0x000001ff

    The last lines in the 8 GB dmesg are:

    > bge0: <Broadcom BCM5705 Gigabit Ethernet, ASIC rev. 0x3003> mem 0xfa000000-0xfa00ffff irq 16 at device 11.0 on pci0
    > bge0: Reserved 0x10000 bytes for rid 0x10 type 3 at 0xfa000000

    They're identical in each probe.

    Greg 

    --
See complete headers for address and phone numbers.

  • Next message: Daniel O'Connor: "Re: Problems with AMD64 and 8 GB RAM?"

    Relevant Pages

    • Re: NIC Card Troubles
      ... are you certain that the numbers for card and chip are correct? ... A memory stick is a good way. ... if your machine have usb ports, and you have a memory stick you can try to use that as storage for the dmesg output. ... You need to find out which partition to use on the memory stick. ...
      (comp.unix.bsd.freebsd.misc)
    • Re: very slow X sessions on 5.2-RC2
      ... > I am beginning to wonder if this has to do with the on-board graphics ... > and the amount of memory I have. ... It only seems like the graphics part is sooo slow. ... > any other information is needed, like dmesg output, I can attach them. ...
      (freebsd-current)
    • Re: Great SWT Program
      ... Heh. ... You don't need to guess anything if your memory is even half as good ... ANSI is not rare". ...
      (comp.lang.java.programmer)
    • Re: OT: Report for Snarl (long)
      ... > I wish my memory was better. ... Cuz I ~*KNOW*~ I'm forgetting some choice ...
      (rec.motorcycles.harley)
    • Re: ATTN: Os
      ... Heh, from memory the pre-season before our relegation we didn't win a ... And this was a team that'd finished the previous Premiership season ...
      (uk.sport.football)