beremiz

Build by default with optimization level -O2 for GCC

2018-05-25, Andrey Skvortsov
a66062a205ae
Build by default with optimization level -O2 for GCC

before -O0 was used by default, that caused pretty bad performance.

Amd64, i6700k, 4200MHz, GNU/Linux (non-RT kernel), gcc 7.2.0

-------------------------------------
Optimization | EN/ENO |no EN/ENO |
-------------------------------------
default | 11 | 9.5 |
-O3 | 3.9 | 5.2 |
-O2 | 4 | 4.8 |
-Os | 4.1 | 3.5 |
-Ofast | 3.9 | 5.2 |
-------------------------------------

ARM, BBB Cortex-A8, 600Mhz, GNU/Linux, gcc 4.6.3

-------------------------------------
Optimization | EN/ENO |no EN/ENO |
-------------------------------------
default | 273 | 226 |
-O3 | 141.8 | 106.2 |
-O2 | 142 | 107 |
-Os | 152.5 | 112.2 |
-Ofast | 141.7 | 106.2 |
-------------------------------------

For embedded systems with size constaints (like Cortex-Mx, AVR and so
on) I usually use -Os. It gets pretty good results. For
GNU/Linux-based systems -O2 is usually a good choice, as you see the
test results.
<?xml version='1.0' encoding='utf-8'?>
<CFile xmlns:xhtml="http://www.w3.org/1999/xhtml">
<includes>
<xhtml:p><![CDATA[
]]></xhtml:p>
</includes>
<variables>
<variable name="TestInput" type="SINT" initial="0"/>
<variable name="TestOutput" type="SINT"/>
</variables>
<globals>
<xhtml:p><![CDATA[
volatile long Lock=0;
volatile char PtoC=1,CtoP=2;
extern long AtomicCompareExchange(long*,long, long);
extern char *PLC_ID;
int Simple_C_Call(int val){
return val+1;
}
int Python_to_C_Call(char toC, char *fromC){
/* Code called by python should never touch to
variables modified by PLC thread directly
AtomicCompareExchange comes from
beremiz' runtime implementation */
int res = 0;
if(!AtomicCompareExchange((long*)&Lock, 0, 1)){
PtoC=toC;
*fromC=CtoP;
AtomicCompareExchange((long*)&Lock, 1, 0);
res=1;
}
printf("C code called by Python: toC %d fromC %d\n",toC,*fromC);
printf("PLC_ID id %s\n",PLC_ID);
return res;
}
int PLC_C_Call(char fromPLC, char *toPLC){
/* PLC also have to be realy carefull not to
conflict with asynchronous python access */
if(!AtomicCompareExchange((long*)&Lock, 0, 1)){
CtoP = fromPLC;
*toPLC = PtoC;
AtomicCompareExchange((long*)&Lock, 1, 0);
return 1;
}
return 0;
}
]]></xhtml:p>
</globals>
<initFunction>
<xhtml:p><![CDATA[
]]></xhtml:p>
</initFunction>
<cleanUpFunction>
<xhtml:p><![CDATA[
]]></xhtml:p>
</cleanUpFunction>
<retrieveFunction>
<xhtml:p><![CDATA[
]]></xhtml:p>
</retrieveFunction>
<publishFunction>
<xhtml:p><![CDATA[
if(!AtomicCompareExchange((long*)&Lock, 0, 1)){
TestInput = CtoP + PtoC + TestOutput;
AtomicCompareExchange((long*)&Lock, 1, 0);
}
]]></xhtml:p>
</publishFunction>
</CFile>