This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
| Other format: | [Raw text] | |
On Fri, Jun 21, 2013 at 10:07:08AM +0200, Torvald Riegel wrote: > On Fri, 2013-06-21 at 04:00 +0200, OndÅej BÃlka wrote: > > I choose a O0 as lesser evil than having reference implementation twice > > faster depending what compiler you do use. > > > > One solution is mandate to run benchmarks with fixed version of gcc and > > fixed flags. > > > > Second variant could be have assemblies and regeneration script that would > > be ran with specific gcc. > > Yes you can try to find a niche where you hope you can compare stuff. > But you can as well just get all the measurements you can from people > out there -- with whatever version of gcc is available -- and take this > into account when drawing conclusions from the data. That is, you'd > setup your machine learning in such a way that it looks a data and > checks whether there is high confidence for a certain conclusion (eg, > new version of code faster or not). Confidence will be lower if, for > example, we see performance vary a lot with different versions of gcc, > but remain more or less unchanged when gcc versions don't differ; but if > performance varies independently of the gcc version, that's also useful > to know because it means we draw our conclusion from a wider set of > tests. Likewise for other properties of the test environment such as > the CPU etc. > And what we will do with this data? You typicaly use machine learning to learn trivial facts from data sets that are too vast to browse manualy. It is faster to just browse results and you will train intuition on it. I wrote a simple script (attached with results) to test memcpy with several gcc versions on several architecture. There a O0 is looks most stable. On O2 there are huge regression on nehalem which and variant of regression is present in opteron results. On O3 things get better for core2 but there is regression in nehalem and ivy bridge architectures. The results for nehalem are below. memcpyO0-4.1.s 18.34 18.34 18.34 18.42 18.34 memcpyO0-4.6.s 18.30 18.30 18.30 18.30 18.30 memcpyO0-4.7.s 18.33 18.33 18.32 18.32 18.33 memcpyO0-4.9.s 18.33 18.33 18.33 18.33 18.33 memcpyO2-4.1.s 3.97 3.88 3.88 3.88 3.88 memcpyO2-4.6.s 5.57 5.57 5.57 5.57 5.57 memcpyO2-4.7.s 5.57 5.57 5.57 5.57 5.57 memcpyO2-4.9.s 5.57 5.57 5.57 5.57 5.57 memcpyO3-4.1.s 3.88 3.88 3.88 3.88 3.88 memcpyO3-4.6.s 0.89 0.89 0.88 0.89 0.88 memcpyO3-4.7.s 0.88 0.88 0.88 0.87 0.88 memcpyO3-4.9.s 0.99 0.98 0.98 0.98 0.98 processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 26 model name : Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz stepping : 4 microcode : 0x6 cpu MHz : 2661.000 cache size : 8192 KB physical id : 0 siblings : 8 core id : 0 cpu cores : 4 apicid : 0 initial apicid : 0 fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm sse4_1 sse4_2 popcnt lahf_lm ida dtherm tpr_shadow vnmi flexpriority ept vpid bogomips : 5335.78 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 1 vendor_id : GenuineIntel cpu family : 6 model : 26 model name : Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz stepping : 4 microcode : 0x6 cpu MHz : 1596.000 cache size : 8192 KB physical id : 0 siblings : 8 core id : 1 cpu cores : 4 apicid : 2 initial apicid : 2 fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm sse4_1 sse4_2 popcnt lahf_lm ida dtherm tpr_shadow vnmi flexpriority ept vpid bogomips : 5334.03 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 2 vendor_id : GenuineIntel cpu family : 6 model : 26 model name : Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz stepping : 4 microcode : 0x6 cpu MHz : 1596.000 cache size : 8192 KB physical id : 0 siblings : 8 core id : 2 cpu cores : 4 apicid : 4 initial apicid : 4 fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm sse4_1 sse4_2 popcnt lahf_lm ida dtherm tpr_shadow vnmi flexpriority ept vpid bogomips : 5307.15 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 3 vendor_id : GenuineIntel cpu family : 6 model : 26 model name : Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz stepping : 4 microcode : 0x6 cpu MHz : 1596.000 cache size : 8192 KB physical id : 0 siblings : 8 core id : 3 cpu cores : 4 apicid : 6 initial apicid : 6 fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm sse4_1 sse4_2 popcnt lahf_lm ida dtherm tpr_shadow vnmi flexpriority ept vpid bogomips : 5334.04 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 4 vendor_id : GenuineIntel cpu family : 6 model : 26 model name : Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz stepping : 4 microcode : 0x6 cpu MHz : 1596.000 cache size : 8192 KB physical id : 0 siblings : 8 core id : 0 cpu cores : 4 apicid : 1 initial apicid : 1 fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm sse4_1 sse4_2 popcnt lahf_lm ida dtherm tpr_shadow vnmi flexpriority ept vpid bogomips : 5334.02 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 5 vendor_id : GenuineIntel cpu family : 6 model : 26 model name : Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz stepping : 4 microcode : 0x6 cpu MHz : 1596.000 cache size : 8192 KB physical id : 0 siblings : 8 core id : 1 cpu cores : 4 apicid : 3 initial apicid : 3 fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm sse4_1 sse4_2 popcnt lahf_lm ida dtherm tpr_shadow vnmi flexpriority ept vpid bogomips : 5334.03 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 6 vendor_id : GenuineIntel cpu family : 6 model : 26 model name : Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz stepping : 4 microcode : 0x6 cpu MHz : 1596.000 cache size : 8192 KB physical id : 0 siblings : 8 core id : 2 cpu cores : 4 apicid : 5 initial apicid : 5 fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm sse4_1 sse4_2 popcnt lahf_lm ida dtherm tpr_shadow vnmi flexpriority ept vpid bogomips : 5334.03 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 7 vendor_id : GenuineIntel cpu family : 6 model : 26 model name : Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz stepping : 4 microcode : 0x6 cpu MHz : 1596.000 cache size : 8192 KB physical id : 0 siblings : 8 core id : 3 cpu cores : 4 apicid : 7 initial apicid : 7 fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm sse4_1 sse4_2 popcnt lahf_lm ida dtherm tpr_shadow vnmi flexpriority ept vpid bogomips : 5334.03 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: > Obviously, right now I don't have a patch to do this. But I believe we > need to keep this in mind. This approach isn't rocket science either. > We already make use of it every day.
Attachment:
memcpy_test.tar.bz2
Description: Binary data
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |