Lunatics MB opt apps limited to counting 16 CPUs?

[arkayn]

From HAL9000

I was looking over the MB results on my 24 core box & noticed something a bit funny in the stderr information. In that it displayed 16 x 2390 MHz where I would have expected to see 24 x 2390 MHz. As shown in this sample.

<stderr_txt>
Windows optimized S@H Enhanced application by Alex Kan
Version info: SSSE3x (Intel, Core 2-optimized v8-nographics) V5.13 by Alex Kan
SSSE3x Win64 Build 76 , Ported by : Jason G, Raistmer, JDWhale

CPUID: Intel(R) Xeon(R) CPU E5645 @ 2.40GHz
Speed: 16 x 2390 MHz
Cache: L1=64K L2=256K
Features: MMX SSE SSE2 SSE3 SSSE3

Work Unit Info:
...............
Credit multiplier is : 2.85
WU true angle range is : 0.815811

Flopcounter: 19124201267179.797000

Spike count: 2
Pulse count: 0
Triplet count: 0
Gaussian count: 0
called boinc_finish

</stderr_txt>


I don't expect there is anything wrong. It is probably just an aesthetic issue, but figured I would mention it. I would have reported it on the lunatics site, but I don't seem to have an account there again.

[Claggy]

On my i7 2600K, the number of cores is displayed correctly in the AKV8 build 76, but the Clock is wrong and only displays the Stock Clock:

Code: [Select]

Windows optimized S@H Enhanced application by Alex Kan
Version info: SSSE3x (Intel, Core 2-optimized v8-nographics) V5.13 by Alex Kan
SSSE3x Win64 Build 76 , Ported by : Jason G, Raistmer, JDWhale
 
     CPUID:        Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz 
     Speed: 8 x 3400 MHz 
     Cache: L1=64K L2=256K
  Features: MMX SSE SSE2 SSE3 SSSE3 
 
The latest v7 revision gets them both wrong:

Code: [Select]

Windows optimized S@H Enhanced application by Alex Kan
Version info: SSSE3x (Intel, Core 2-optimized v8-nographics) V5.13 by Alex Kan
SSSE3x Win32 Build 374 , Ported by : Jason G, Raistmer, JDWhale


SETI7 update by Raistmer

Build features: SETI7 Non-graphics IPP USE_SSSE3 x86
     CPUID:        Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz
     Speed: 4 x 3394 MHz
     Cache: L1=64K L2=256K

CPU features: FPU TSC PAE CMPXCHG8B APIC SYSENTER MTRR CMOV/CCMP MMX FXSAVE/FXRSTOR SSE SSE2 HT SSE3
CPU type 0x46
CPU-Z output:

Code: [Select]

Processors Information
-------------------------------------------------------------------------

Processor 1 ID = 0
Number of cores 4 (max 8)
Number of threads 8 (max 16)
Name Intel Core i7 2600K
Codename Sandy Bridge
Specification Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz
Package (platform ID) Socket 1155 LGA (0x1)
CPUID 6.A.7
Extended CPUID 6.2A
Core Stepping D2
Technology 32 nm
TDP Limit 95 Watts
Core Speed 4700.4 MHz
Multiplier x FSB 47.0 x 100.0 MHz
Stock frequency 3400 MHz
Instructions sets MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, EM64T, VT-x, AES, AVX
L1 Data cache 4 x 32 KBytes, 8-way set associative, 64-byte line size
L1 Instruction cache 4 x 32 KBytes, 8-way set associative, 64-byte line size
L2 cache 4 x 256 KBytes, 8-way set associative, 64-byte line size
L3 cache 8 MBytes, 16-way set associative, 64-byte line size
FID/VID Control yes

Turbo Mode supported, enabled
Max non-turbo ratio 34x
Max turbo ratio 59x
Max efficiency ratio 16x
Min Power 60 Watts
O/C bins unlimited
Claggy

[Raistmer]

You compared x64 with x32 one
What older x32 build says?

[Claggy]

You compared x64 with x32 one
What older x32 build says?

same as the x64 variant:

Code: [Select]

Windows optimized S@H Enhanced application by Alex Kan
Version info: SSSE3x (Intel, Core 2-optimized v8-nographics) V5.13 by Alex Kan
SSSE3x Win32 Build 76 , Ported by : Jason G, Raistmer, JDWhale

     CPUID:        Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz
     Speed: 8 x 3394 MHz
     Cache: L1=64K L2=256K
  Features: MMX SSE SSE2 SSE3 SSSE3
Claggy

[Raistmer]

No, not the same, look at reported freqs
both new and old builds disagree with x64 one

[Claggy]

My Atom N450 also reports on r246 as only having one core (it should be one core plus one hyper-threaded core):

Code: [Select]

Windows optimized S@H Enhanced application by Alex Kan
Version info: SSSE3x (Intel, Core 2-optimized v8-nographics) V5.13 by Alex Kan
SSSE3x Win32 Build 246 , Ported by : Jason G, Raistmer, JDWhale
 
SETI7 update by Raistmer
 
Build features: SETI7 Non-graphics IPP USE_SSSE3 x86
     CPUID:          Intel(R) Atom(TM) CPU N450   @ 1.66GHz 
     Speed: 1 x 1659 MHz 
     Cache: L1=64K L2=512K
 
CPU features: FPU TSC PAE CMPXCHG8B APIC SYSENTER MTRR CMOV/CCMP MMX FXSAVE/FXRSTOR SSE SSE2 HT SSE3 
CPU type 0x24 
Claggy

[Raistmer]

yes, this feature looks broken and better to remove it in new rebuild.

[Claggy]

No, not the same, look at reported freqs
both new and old builds disagree with x64 one

Most of my v6 results show 8 x 3400 MHz, but one shows Speed: 8 x 3421 MHz and another shows Speed: 8 x 3394 MHz , so what is reported swings around a bit.

Claggy

[Josef W. Segur]

Feature detection in the Lunatics builds is mostly based on Microsoft example code for using the intrinsic __cpuid() function, see e.g. http://msdn.microsoft.com/en-us/library/hskdteyh(v=VS.80).aspx. Speed detection is added using rdtsc, IIRC.

As has been noted, that's basically cosmetic for now. If we ever get around to using some of that information to choose paths within the processing, it will be necessary to try to make it accurate and keep it up to date.
                                                       Joe

[Mike]

V8 only seeing 1 core with correct speed.

Windows optimized S@H Enhanced application by Alex Kan
Version info: SSE3x (AMD/Intel, Core 2-optimized v8-nographics) V5.13 by Alex Kan
SSE3x Win32 Build 76 , Ported by : Jason G, Raistmer, JDWhale

     CPUID: AMD Phenom(tm) II X6 1090T Processor
     Speed: 1 x 3607 MHz
     Cache: L1=64K L2=512K
  Features: MMX SSE SSE2 SSE3

r374 reports correct

Windows optimized S@H Enhanced application by Alex Kan
Version info: SSE2x (AMD/Intel, Core 2-optimized v8-nographics) V5.13 by Alex Kan
SSE2x Win32 Build 374 , Ported by : Jason G, Raistmer, JDWhale


SETI7 update by Raistmer

Build features: SETI7   Non-graphics   IPP   USE_SSE2   x86   
     CPUID: AMD Phenom(tm) II X6 1090T Processor
     Speed: 6 x 3601 MHz
     Cache: L1=64K L2=512K

CPU features: FPU TSC PAE CMPXCHG8B APIC SYSENTER MTRR CMOV/CCMP MMX FXSAVE/FXRSTOR SSE SSE2 HT SSE3
CPU type 0x42

[Fredericx51]

[Also posted on SETI@hone, NC forum.]
I noticed the same info on the I7-2600, HTT turned on, displays 4 cores. These are Bęta builds!

(SETI Bęta : Build features: SETI7 Non-graphics OpenCL USE_OPENCL_HD5xxx IPP AMD specific USE_SSE3 x86
CPUID: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz.  [App. rev365 for ATI GPU]
Speed: 4 x 3386 MHz
Cache: L1=64K L2=256K

CPU features: FPU TSC PAE CMPXCHG8B APIC SYSENTER MTRR CMOV/CCMP MMX FXSAVE/FXRSTOR SSE SSE2 HT SSE3
CPU type 0x46
Number of OpenCL platforms: 1

Bęta rev.365 for CPU, displays :

<stderr_txt>
Windows optimized S@H Enhanced application by Alex Kan
Version info: SSSE3x (Intel, Core 2-optimized v8-nographics) V5.13 by Alex Kan
SSSE3x Win32 Build 365 , Ported by : Jason G, Raistmer, JDWhale


SETI7 update by Raistmer

Build features: SETI7   Non-graphics   IPP   USE_SSSE3   x86   
     CPUID:         Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz
     Speed: 4 x 3349 MHz
     Cache: L1=64K L2=256K

CPU features: FPU TSC PAE CMPXCHG8B APIC SYSENTER MTRR CMOV/CCMP MMX FXSAVE/FXRSTOR SSE SSE2 HT SSE3
CPU type 0x46 .

The x86-64 SSSE3 optimized app., displays correct info on the I7-2600 :
<stderr_txt>
Windows optimized S@H Enhanced application by Alex Kan
Version info: SSSE3x (Intel, Core 2-optimized v8-nographics) V5.13 by Alex Kan
SSSE3x Win64 Build 76 , Ported by : Jason G, Raistmer, JDWhale

     CPUID:         Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz
     Speed: 8 x 3386 MHz
     Cache: L1=64K L2=256K
  Features: MMX SSE SSE2 SSE3 SSSE3

But, as you already stated, it's cosmetic and in this case different from
NON-ATI apps, but it is a QUAD-Core with HTT. (Turned on, it also doesn't
notice the L3 cache, on this CPU, haven't seen any notice of L3 on BOINC either).

The Core i7-2600 processor has 64 KB L1 cache memory per core, 256 KB L2 cache memory per core and an 8 MB L3 Advanced Smart Cache ...

But, IMHO, that's what Bęta-testing/tuning, also is for, spotting inregulaireties and other things, that could be improved ;-)
B.t.w. I'm very glad with both rev.177 (& rev.365 bęta) and rev.521 for AstroPulse, on ATI GPUs.

And shouldn't BOINC display exact CPUID info?