I'm setting up Radiance on a computer cluster with lots of nodes. Tweeking gcc flags is not my best quality(understatement), so I bring it here.
Playing a bit with options which I found on Marc's benchmark page, I got rendering times per core over twice as long as on my 2.2 Ghz macbook. Doesn't make me happy
Below I've pasted the cpuinfo from a node on the cluster. Does anyone in the group has an idea what flags I should give to gcc to optimize Radiance?
The flags I've used so far - are: -march=native -m64 -msse -msse2 -funroll-loops -ffast-math -O3 -Dlinux -D_FILE_OFFSET_BITS=64
Cheers,
iebele
cat /proc/cpuinfo:
processor : 7
vendor_id : GenuineIntel
cpu family : 6
model : 23
model name : Intel(R) Xeon(R) CPU L5420 @ 2.50GHz
stepping : 10
cpu MHz : 2493.445
cache size : 6144 KB
physical id : 0
siblings : 4
core id : 2
cpu cores : 4
apicid : 2
initial apicid : 2
fpu : yes
fpu_exception : yes
cpuid level : 13
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 lm constant_tsc arch_perfmon pebs bts rep_good aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm dca sse4_1 xsave lahf_lm tpr_shadow vnmi flexpriority
bogomips : 4987.90
clflush size : 64
cache_alignment : 64
address sizes : 38 bits physical, 48 bits virtual
You've picked what are probably the best compile options for performance. You may or may not see a difference by adding "-fomit-frame-pointer". You may also see a difference by using the gcc 4.6 compilers instead of any earlier versions. 4.6 seems to impart a 5% advantage over even gcc 4.4.
But like Jack mentioned, these compiler options can, at best, give you incremental performance improvements. Whenever I have real work to do, I always use rpiece and a script similar to the "runsmp" that is in the bench4.tar.gz.
If you have the results from some of your tests, I'd like to put them on the Radiance benchmark page.
Mark
路路路
On Sat, 7 Apr 2012, Iebele wrote:
Hi group,
I'm setting up Radiance on a computer cluster with lots of nodes. Tweeking gcc flags is not my best quality(understatement), so I bring it here. Playing a bit with options which I found on Marc's benchmark page, I got rendering times per core over twice as long as on my 2.2 Ghz macbook. Doesn't make me happy Below I've pasted the cpuinfo from a node on the cluster. Does anyone in the group has an idea what flags I should give to gcc to optimize Radiance? The flags I've used so far - are: -march=native -m64 -msse -msse2 -funroll-loops -ffast-math -O3 -Dlinux -D_FILE_OFFSET_BITS=64
Cheers,
iebele
cat /proc/cpuinfo:
processor : 7
vendor_id : GenuineIntel
cpu family : 6
model : 23
model name : Intel(R) Xeon(R) CPU L5420 @ 2.50GHz
stepping : 10
cpu MHz : 2493.445
cache size : 6144 KB
physical id : 0
siblings : 4
core id : 2
cpu cores : 4
apicid : 2
initial apicid : 2
fpu : yes
fpu_exception : yes
cpuid level : 13
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 lm constant_tsc arch_perfmon pebs bts rep_good aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm dca sse4_1 xsave lahf_lm tpr_shadow vnmi flexpriority
bogomips : 4987.90
clflush size : 64
cache_alignment : 64
address sizes : 38 bits physical, 48 bits virtual
_______________________________________________
Radiance-general mailing list [email protected] http://www.radiance-online.org/mailman/listinfo/radiance-general
Thanks a lot for your input, it has been very much supportive.
Marc, I'll run the benchmark one of these weeks and send the results to you,
Iebele
路路路
Op 16 apr. 2012, om 18:58 heeft Mark Stock het volgende geschreven:
Iebele,
You've picked what are probably the best compile options for performance. You may or may not see a difference by adding "-fomit-frame-pointer". You may also see a difference by using the gcc 4.6 compilers instead of any earlier versions. 4.6 seems to impart a 5% advantage over even gcc 4.4.
But like Jack mentioned, these compiler options can, at best, give you incremental performance improvements. Whenever I have real work to do, I always use rpiece and a script similar to the "runsmp" that is in the bench4.tar.gz.
If you have the results from some of your tests, I'd like to put them on the Radiance benchmark page.
Mark
On Sat, 7 Apr 2012, Iebele wrote:
Hi group,
I'm setting up Radiance on a computer cluster with lots of nodes. Tweeking gcc flags is not my best quality(understatement), so I bring it here. Playing a bit with options which I found on Marc's benchmark page, I got rendering times per core over twice as long as on my 2.2 Ghz macbook. Doesn't make me happy Below I've pasted the cpuinfo from a node on the cluster. Does anyone in the group has an idea what flags I should give to gcc to optimize Radiance? The flags I've used so far - are: -march=native -m64 -msse -msse2 -funroll-loops -ffast-math -O3 -Dlinux -D_FILE_OFFSET_BITS=64
Cheers,
iebele
cat /proc/cpuinfo:
processor : 7
vendor_id : GenuineIntel
cpu family : 6
model : 23
model name : Intel(R) Xeon(R) CPU L5420 @ 2.50GHz
stepping : 10
cpu MHz : 2493.445
cache size : 6144 KB
physical id : 0
siblings : 4
core id : 2
cpu cores : 4
apicid : 2
initial apicid : 2
fpu : yes
fpu_exception : yes
cpuid level : 13
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 lm constant_tsc arch_perfmon pebs bts rep_good aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm dca sse4_1 xsave lahf_lm tpr_shadow vnmi flexpriority
bogomips : 4987.90
clflush size : 64
cache_alignment : 64
address sizes : 38 bits physical, 48 bits virtual
_______________________________________________
Radiance-general mailing list [email protected] http://www.radiance-online.org/mailman/listinfo/radiance-general
