amd64, 2.6.7: several problems
Hi folks,
I've got some questions about amd64.
/proc/cpuinfo returns
processor : 0 vendor_id : AuthenticAMD cpu family : 15 model : 12 model name : AMD Athlon(tm) 64 Processor 3200+ stepping : 0 cpu MHz : 2194.366 cache size : 512 KB fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx mmxext lm 3dnowext 3dnow bogomips : 4341.76 TLB size : 1088 4K pages clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management: ts fid vid ttp
Please check the cache size. It _should_ be 1024 kB, at least that was written on the bill I've got. Is this correct?
Next powernow-k8 refuses to load on my PC:
# modprobe powernow-k8 FATAL: Error inserting powernow_k8 (/lib/modules/2.6.7/kernel/arch/x86_64/kernel/cpufreq/powernow-k8.ko): No such device
This is strange. AFAIK all amd64 CPUs do have Powernow. The ACPI modules (expecially processor), freq_table and cpuid were all loaded. Whats wrong here?
I tried to compile a static powernow-k8, but that gave me the error message
arch/x86_64/kernel/built-in.o(.init.text+0x7d9a): In function `powernowk8_cpu_init':
: undefined reference to acpi_processor_register_performance' arch/x86_64/kernel/built-in.o(.init.text+0x7f39): In function powernowk8_cpu_init':
: undefined reference to acpi_processor_unregister_performance' arch/x86_64/kernel/built-in.o(.exit.text+0x33): In function powernowk8_cpu_exit':
: undefined reference to `acpi_processor_unregister_performance' make[1]: *** [.tmp_vmlinux1] Error 1
make[1]: Leaving directory `/usr/src/kernel-source-2.6.7'
make: *** [stamp-build] Error 2
The ACPI Processor stuff was still set to "build module", so it seems that there is a missing dependency in menuconfig.
Any help would be highly appreciated
Harri
Quoted 10 lines · Harald Dunkel
Next powernow-k8 refuses to load on my PC:
# modprobe powernow-k8 FATAL: Error inserting powernow_k8 (/lib/modules/2.6.7/kernel/arch/x86_64/kernel/cpufreq/powernow-k8.ko): No such deviceThis is strange. AFAIK all amd64 CPUs do have Powernow. The ACPI modules (expecially processor),
freq_tableand cpuid were all loaded. Whats wrong here?
PS: dmesg said
powernow-k8: Found 1 AMD Athlon 64 / Opteron processors (version 1.00.09b)
powernow-k8: BIOS error - no PSB
I found this in powernow-k8.c.
:
if (powernow_k8_cpu_init_acpi(data)) {
/*
* Use the PSB BIOS structure. This is only availabe on
* an UP version, and is deprecated by AMD.
*/
if (pol->cpu != 0) {:
:
i.e. PSB is deprecated. Only if some lookup via ACPI fails, then the kernel tries PSB. If this fails, too, then it prints an error message.
The problem is, why does the kernel print an error message just about the deprecated PSB, if some ACPI stuff doesn't work as expected?
Would it be possible to add a warning message to powernow_k8_cpu_init_acpi(), saying something like
powernow-k8: initialization via ACPI failed,
trying deprecated PSB structure
Regards
Harri
Quoted 27 lines · Harald Dunkel
Hi folks,
I've got some questions about amd64.
/proc/cpuinforeturnsprocessor : 0 vendor_id : AuthenticAMD cpu family : 15 model : 12 model name : AMD Athlon(tm) 64 Processor 3200+ stepping : 0 cpu MHz : 2194.366 cache size : 512 KB fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mcacmov pat pse36 clflush mmx fxsr sse sse2 syscall nx mmxext lm 3dnowext
3dnow bogomips : 4341.76
TLB size : 1088 4K pages
clflush size : 64
cache_alignment: 64
address sizes : 40 bits physical, 48 bits virtual
power management: ts fid vid ttp
Quoted 2 lines
Please check the cache size. It _should_ be 1024 kB, at least that was written on the bill I've got. Is this correct?
Well, not. There are two "versions" of Athlon64 3200+, one of which has 1024 KB of cache and is clocked at 2.0 GHz, and the second has 512 KB of cache and is clocked at 2.2 GHz. Yours is the second one, apparently. More info at:
http://www.amdboard.com/athlon_64_3200.html
Yours,
rjw
Signature · 5 lines
Rafael J. Wysocki ---------------------------- For a successful technology, reality must take precedence over public relations, for nature cannot be fooled. -- Richard P. Feynman