188 lines
5.0 KiB
HTML
188 lines
5.0 KiB
HTML
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">
|
|
<HTML>
|
|
<HEAD>
|
|
<META NAME="GENERATOR" CONTENT="SGML-Tools 1.0.9">
|
|
<TITLE>SPARC-HOWTO.: SMP.</TITLE>
|
|
<LINK HREF="SPARC-HOWTO-22.html" REL=next>
|
|
<LINK HREF="SPARC-HOWTO-20.html" REL=previous>
|
|
<LINK HREF="SPARC-HOWTO.html#toc21" REL=contents>
|
|
</HEAD>
|
|
<BODY>
|
|
<A HREF="SPARC-HOWTO-22.html">Next</A>
|
|
<A HREF="SPARC-HOWTO-20.html">Previous</A>
|
|
<A HREF="SPARC-HOWTO.html#toc21">Contents</A>
|
|
<HR>
|
|
<H2><A NAME="s21">21. SMP.</A></H2>
|
|
|
|
<P>
|
|
<P>Ouaich, SMP is supported, and I am sure that ROSS's dual CPU
|
|
modules are supported as you can see by yourself:
|
|
<P>
|
|
<HR>
|
|
<PRE>
|
|
|
|
piou@ywing:~ > cat /proc/cpuinfo
|
|
cpu : ROSS HyperSparc RT625 or RT626
|
|
fpu : ROSS HyperSparc combined IU/FPU
|
|
promlib : Version 3 Revision 2
|
|
prom : 2.25
|
|
type : sun4m
|
|
ncpus probed : 2
|
|
ncpus active : 2
|
|
Cpu0Bogo : 125.33
|
|
Cpu1Bogo : 125.33
|
|
MMU type : ROSS HyperSparc
|
|
invall : 0
|
|
invmm : 0
|
|
invrnge : 0
|
|
invpg : 0
|
|
contexts : 4096
|
|
CPU0 : online
|
|
CPU1 : online
|
|
piou@ywing:~ >
|
|
</PRE>
|
|
<HR>
|
|
<P>
|
|
<P>The SparcStation 10 and SparcStations 20 are SMP capable computers and
|
|
according to the
|
|
<A HREF="http://faqaboss.sunhelp.org/">FAQABOSS</A>
|
|
the following combinations are known to work :
|
|
<UL>
|
|
<LI>2xSM40 ( model 402 )</LI>
|
|
<LI>2xSM41 ( model 412 )</LI>
|
|
<LI>2xSM51 ( model 512 )</LI>
|
|
<LI>2xSM512 ( model 514 )</LI>
|
|
<LI>2xSM61 ( model 612 )</LI>
|
|
<LI>2xSM71 ( model 712 )</LI>
|
|
<LI>2xSM81 ( model 812 )</LI>
|
|
</UL>
|
|
<P>And, as stated earlier, CPU modules in SparcStations 10 and can run a
|
|
different clock speeds, the following ones __SHOULD__ work:
|
|
<P>
|
|
<UL>
|
|
<LI>2xSM50</LI>
|
|
<LI>SM41, SM51</LI>
|
|
<LI>SM41, SM61</LI>
|
|
<LI>SM51, SM61</LI>
|
|
<LI>SM71, SM81</LI>
|
|
</UL>
|
|
<P>
|
|
<P>
|
|
<P>
|
|
<P>How does it performs? Well, it is fast, really fast. Some of the java
|
|
Demos can run faster on a dual HyperSparc 125Mhz 128MB ( <CODE>ywing</CODE> ) than on a dual
|
|
celeron BP6 433@433Mhz 192MB ( <CODE>calimero</CODE> ). The same applies for the Gimp. When it
|
|
comes to compiling <CODE>calimero</CODE> runs faster than <CODE>ywing</CODE>. Both computers
|
|
running 2.2.16 kernel and <CODE>calimero</CODE>'s hard disk subsystem is full SCSI.
|
|
<P>One important detail when you plan to have different CPU modules in
|
|
your computer is to have the same kind of modules, you cannot mix
|
|
SuperSparc and HyperSparc for example, but you can have an odd number
|
|
of CPUs, for example 3. They are said to be able to run modules
|
|
at different clock speed as written in this article form
|
|
<A HREF="http://www.aceshardware.com/Spades/read.php?article_id=76">AcesHardware</A>
|
|
, but I have not witnessed it.
|
|
<P><CODE>ywing</CODE> has been upgraded to quad-CPU. You can read the
|
|
kernel's message:
|
|
<P>
|
|
<HR>
|
|
<PRE>
|
|
cpu : ROSS HyperSparc RT625 or RT626
|
|
fpu : ROSS HyperSparc combined IU/FPU
|
|
promlib : Version 3 Revision 2
|
|
prom : 2.25
|
|
type : sun4m
|
|
ncpus probed : 4
|
|
ncpus active : 4
|
|
Cpu0Bogo : 125.33
|
|
Cpu1Bogo : 125.33
|
|
Cpu2Bogo : 125.33
|
|
Cpu3Bogo : 125.33
|
|
MMU type : ROSS HyperSparc
|
|
invall : 0
|
|
invmm : 0
|
|
invrnge : 0
|
|
invpg : 0
|
|
contexts : 4096
|
|
CPU0 : online
|
|
CPU1 : online
|
|
CPU2 : online
|
|
CPU3 : online
|
|
</PRE>
|
|
<HR>
|
|
<P>and its main memory is now 256MB:
|
|
<P>
|
|
<HR>
|
|
<PRE>
|
|
total: used: free: shared: buffers: cached:
|
|
Mem: 263028736 29114368 233914368 22958080 1695744 12779520
|
|
Swap: 133849088 0 133849088
|
|
MemTotal: 256864 kB
|
|
MemFree: 228432 kB
|
|
MemShared: 22420 kB
|
|
Buffers: 1656 kB
|
|
Cached: 12480 kB
|
|
BigTotal: 0 kB
|
|
BigFree: 0 kB
|
|
SwapTotal: 130712 kB
|
|
SwapFree: 130712 kB
|
|
</PRE>
|
|
<HR>
|
|
<P>So I have performed an empirical <CODE>proftpd</CODE> compilation test using the
|
|
<CODE>make -JN</CODE>. The results are:
|
|
<HR>
|
|
<PRE>
|
|
with make
|
|
real 3m27.466s
|
|
user 3m15.670s
|
|
sys 0m12.030s
|
|
|
|
with make -j2
|
|
|
|
real 1m52.670s
|
|
user 3m27.210s
|
|
sys 0m15.310s
|
|
|
|
with make -j3
|
|
|
|
real 1m22.560s
|
|
user 3m43.910s
|
|
sys 0m18.070s
|
|
|
|
with make -j4
|
|
real 1m13.582s
|
|
user 4m2.200s
|
|
sys 0m22.830s
|
|
|
|
with make -j5
|
|
real 1m13.445s
|
|
user 4m4.060s
|
|
sys 0m22.640s
|
|
|
|
with make -j8
|
|
|
|
real 1m15.550s
|
|
user 4m1.840s
|
|
sys 0m22.960s
|
|
|
|
with make -j 10
|
|
real 1m20.091s
|
|
user 4m2.440s
|
|
sys 0m22.170s
|
|
</PRE>
|
|
<HR>
|
|
<P>As expected the best results are with make -j5 ( one instance per CPU +
|
|
one ready to enter when a cache miss occurs ); then N increasing the results
|
|
are starting to worsen.
|
|
<P>As a conclusion, those <CODE>sun4m</CODE> SMP systems will be very
|
|
interesting when Gimp 2.0 will be available because of multitreading
|
|
and paralelization of algorithms.
|
|
<P>If you want to learn more about SMP and Linux you can read the
|
|
<CODE>SMP-HOWTO</CODE>
|
|
<A HREF="http://www.tldp.org/HOWTO/SMP-HOWTO.html">http://www.tldp.org/HOWTO/SMP-HOWTO.html</A><P>
|
|
<HR>
|
|
<A HREF="SPARC-HOWTO-22.html">Next</A>
|
|
<A HREF="SPARC-HOWTO-20.html">Previous</A>
|
|
<A HREF="SPARC-HOWTO.html#toc21">Contents</A>
|
|
</BODY>
|
|
</HTML>
|