On Mar 6, 8:48 am, Cactus <rieman...@googlemail.com> wrote:
> On Mar 6, 7:55 am, Cactus <rieman...@googlemail.com> wrote:
>
>
>
> > On Mar 6, 12:44 am, Jeff Gilchrist <jeff.gilchr...@gmail.com> wrote:
>
> > > Thanks to Jason's recommendation, I was finally able to get a 64bit
> > > Linux distro working on my Core2 machine so I have now been able to
> > > benchmark MPIR in Linux and Vista 64bit on the same machine.
>
> > > I have attached the miprbench and speed test results for both
> > > operating systems. There is definitely a difference between the two.
> > > Vista 64bit = MPIRbench result: 11985.85
> > > Linus 64bit = MPIRbench result: 13172
>
> > > Some of the speed timings are quite different, but the
> > > mpn_mul_basecase more more similar than the rest.
>
> > > Processor info:
>
> > > processor : 0
> > > vendor_id : GenuineIntel
> > > cpu family : 6
> > > model : 23
> > > model name : Intel(R) Core(TM)2 Quad CPU Q9550 @ 2.83GHz
> > > stepping : 7
> > > cpu MHz : 3399.860
> > > cache size : 6144 KB
> > > physical id : 0
> > > siblings : 4
> > > core id : 0
> > > cpu cores : 4
> > > apicid : 0
> > > initial apicid : 0
> > > fpu : yes
> > > fpu_exception : yes
> > > cpuid level : 10
> > > 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 nopl pni monitor ds_cpl
> > > vmx smx est tm2 ssse3 cx16 xtpr sse4_1 lahf_lm
> > > bogomips : 6799.72
> > > clflush size : 64
> > > cache_alignment : 64
> > > address sizes : 36 bits physical, 48 bits virtual
> > > power management:
>
> > > mpirbench_Q9550_3.4GHz_Linux64.txt
> > > 2KViewDownload
>
> > > mpirbench_Q9550_3.4GHz_Vista64.txt
> > > 2KViewDownload
>
> > > speed_Q9550_3.4GHz_Linux64.txt
> > > 11KViewDownload
>
> > > speed_Q9550_3.4GHz_Vista64.txt
> > > 12KViewDownload
>
> > Hi Jeff
>
> > I am really grateful for your efforts on this. It has clearly cost
> > you a lot of time but I assure you it is really appreciated.
>
> > I am now updating the graphs and it seems that mpn_add_n is odd on
> > Windows. As it happens I am worrying about this already so I will
> > hopefully be able to track it down
> es in the Core2 code
> > I will update the mpn_add_n graph now and the others later.
>
> > Brian
>
> All the graphs now updated. The Linux and Windows on the same machine
> are the thicker dashed lines.
My thanks to everyone who submitted results on this thread.
This allowed me to identify several issues and I also discovered that
there were changes in the Linux Core2 code that had not made it into
Windows.
I also found some algorithm and tuning improvements so the trunk SVN
code should now be better on Core2 (although still behind Linux
speed).
Brian
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"mpir-devel" group.
To post to this group, send email to mpir-devel@googlegroups.com
To unsubscribe from this group, send email to
mpir-devel+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/mpir-devel?hl=en
-~----------~----~----~----~------~----~------~--~---