memory alignment problems with -current on amd64? [Found Cause]

[Mark Atkinson]

Alan Cox wrote:
> Do you know why NX is disabled on this processor?  If possible, could you
> re-enable it and vm.pmap.pg_ps_enabled, and see if you still have the same
> problem.

I'm not sure why it's off, but I re-enabled it in BIOS with the same result
(still resets or gives a bus error).  

For the processor, I can't seem to find anything that gives the revision.  
Is the Id and stepping good enough?

Ports misc/cpuid seems to give the same information that's in the verbose
boot.

Still can't get the buildworld bus error to dump core for me.  I inserted
a 'ulimit -a' command into some targets in Makefile.inc and it shows
unlimited for the Make environment. 

However, I note that it doesn't say (core dumped) when the error occurs.

example, my latest failure:
peigen.c: In function '_bfd_pei_only_swap_filehdr_out':
peigen.c:2089: internal compiler error: Bus error: 10
Please submit a full bug report,
with preprocessed source if appropriate.

With an deliberately induced bus error, a program will dump core.

> On Wed, Feb 11, 2009 at 5:26 PM, Mark Atkinson <atkin901 at yahoo.com> wrote:
> 
>> Alan Cox wrote:
>>
>> > On Tue, Feb 10, 2009 at 2:20 PM, Mark Atkinson <atkin901 at yahoo.com>
>> wrote:
>> > [snip]
>> >>
>> >>
>> >>
>> >> Well, taking the information I knew -- OCT 15th == good, Mid DEC ==
>> >> BAD,
>> >> I trolled every commit logged between.   Eventually I found this one:
>> >>
>> >> http://svn.freebsd.org/viewvc/base?view=revision&revision=185715
>> >>
>> >> http://docs.freebsd.org/cgi/mid.cgi?200812061937.mB6JbqAI003273
>> >>
>> >> I set vm.pmap.pg_ps_enabled="0" in /boot/loader.conf, and
>> >> was able to complete buildworld and -j16 buildworld and -j8
>> >> buildkernel no problem.
>> >>
>> >> It appears superpage mapping causes alignment problems on this box.
>> >
>> >
>> > Can you please provide more detailed information about this machine, in
>> > particular, the processor including the revision?  It would also be
>> > helpful to see what gdb says about a couple of these crashes,
>> > specifically, the machine registers at the time of the exception.
>>
>> Is there something specifically preventing cores during buildworld?  I
>> can't
>> find one after a bus error.   I'll try to find something to dump the
>> revision for me.  Here's the verbose boot for the processor.
>>
>> CPU: Quad-Core AMD Opteron(tm) Processor 2352 (2100.09-MHz K8-class CPU)
>>  Origin = "AuthenticAMD"  Id = 0x100f23  Stepping = 3
>>
>>
>>
Features=0x178bfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,MMX,FXSR,SSE,SSE2,HTT>
>>  Features2=0x802009<SSE3,MON,CX16,POPCNT>
>>  AMD Features=0xee400800<SYSCALL,MMX+,FFXSR,Page1GB,RDTSCP,LM,3DNow
>> +,3DNow!>
>>  AMD
>> Features2=0x7ff<LAHF,CMP,SVM,ExtAPIC,CR8,ABM,SSE4A,MAS,Prefetch,OSVW,IBS>
>>  TSC: P-state invariant
>>  Cores per package: 4
>> L1 2MB data TLB: 48 entries, fully associative
>> L1 2MB instruction TLB: 16 entries, fully associative
>> L1 4KB data TLB: 48 entries, fully associative
>> L1 4KB instruction TLB: 32 entries, fully associative
>> L1 data cache: 64 kbytes, 64 bytes/line, 1 lines/tag, 2-way associative
>> L1 instruction cache: 64 kbytes, 64 bytes/line, 1 lines/tag, 2-way
>> associative
>> L2 2MB data TLB: 128 entries, 2-way associative
>> L2 2MB instruction TLB: 0 entries, 2-way associative
>> L2 4KB data TLB: 512 entries, 4-way associative
>> L2 4KB instruction TLB: 512 entries, 4-way associative
>> L2 unified cache: 512 kbytes, 64 bytes/line, 1 lines/tag, 16-way
>> associative
>> usable memory = 10720198656 (10223 MB)
>> Physical memory chunk(s):
>> 0x0000000000001000 - 0x000000000009bfff, 634880 bytes (155 pages)
>> 0x0000000000f55000 - 0x00000000cfe4dfff, 3471806464 bytes (847609 pages)
>> 0x00000000cfe56000 - 0x00000000cfe56fff, 4096 bytes (1 pages)
>> 0x0000000100000000 - 0x000000029d212fff, 6931165184 bytes (1692179 pages)
>>
>>
>> --
>> Mark Atkinson
>> atkin901 at yahoo.com
>> (!wired)?(coffee++):(wired);
>>
>> _______________________________________________
>> freebsd-current at freebsd.org mailing list
>> http://lists.freebsd.org/mailman/listinfo/freebsd-current
>> To unsubscribe, send any mail to
>> "freebsd-current-unsubscribe at freebsd.org"
>>
> _______________________________________________
> freebsd-current at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-current
> To unsubscribe, send any mail to "freebsd-current-unsubscribe at freebsd.org"

-- 
Mark Atkinson
atkin901 at yahoo.com
(!wired)?(coffee++):(wired);