Experimental Physics and
Industrial Control System

Eric Norum <[email protected]> · Sun, 16 Sep 2007 07:52:14 -0500

You need to specify the format conversion for a double, not for a float.
Change the format from  %f to %lf and things should work.
Have a look at the scanf man mage for more details.
On Sep 15, 2007, at 9:01 PM, Szalata, Zenon M. wrote:
I have simple softIOC which reads values from a HP34401A
multimeter. The multimeter is connected via GPIB bus to NI PCI-
GPIB interface. The IOC is built with EPICS 3.14.9, ASYN 4-8, and
linux-gpib 3.2.09.
The command which reads the value is:
 {&DSET_LI,GPIBREAD,IB_Q_HIGH,"READ?","%f",0,80,NULL,
0,0,NULL,NULL,NULL},
Caget shows something like: 1091779303, which is nonsence. The
value should be about 9.202. When I change the record to stringi,
then I see correct 9.2 type value. Also in edm I see the same
1091779303 unreal number. When I display it in hex, I see 0x41133AE7.

I also wrote a simple c program where:

int v=0x41133ae7;

float f;

memdpy( &f,&v,sizeof(int));

print( "0x%x %f\n",v,f);
This gives an output:
0x41133ae7 9.201880
All of this is on RHEL4 linux. It seems to me that EPICS is
interpreting the byte order in a way different than the c program.
But then what do I need to do to have the PV floating value be
correct? Thanks in advance,

Zen

--
Eric Norum <[email protected]>
Advanced Photon Source
Argonne National Laboratory
(630) 252-4793

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System