repeated histogram_start for adAcquire?

[newville]

@TomQD-94 I'm seeing occasional segfaults of the xspress3App that I am not sure I fully understand. It seems to happen most often when I start a "second run", as for a second row of a map. That is, the first row of a map completes, including writing the HDF5 file. My scanning code then resets some parameters and starts the next row. When starting the xspress3 may cause the segfault.

Starting at line 1215 of xspress3Epics.cpp (

xspress3/xspress3App/src/xspress3Epics.cpp

Line 1215 in 36e9c41

else if (function == ADAcquire) {

)

Curiously, at line 1233 (

xspress3/xspress3App/src/xspress3Epics.cpp

Line 1233 in 36e9c41

setupITFG();

)

the code has

	  setupITFG();
	  xsp3_status = xsp3->histogram_start(xsp3_handle_, -1 );
	  if (xsp3_status != XSP3_OK) {
	    checkStatus(xsp3_status, "xsp3_histogram_start", functionName);
	    status = asynError;

	  } else {
	    setupITFG(); 
	    xsp3_status = xsp3->histogram_start(xsp3_handle_, -1 );
		
	    if (xsp3_status != XSP3_OK) {
	      checkStatus(xsp3_status, "xsp3_histogram_start", functionName);
	      status = asynError;
	    }

I think that repeated call to setupITF() and xsp3->histogram_start() must be a mistake, no?

I'm not convinced that this is the only problem I am seeing, but commenting out the second set seems to help and not cause any different problems. Do you think this should be removed?

[TomQD-94]

Hi Matt,

I think you are right, I can't see the logic in calling it twice.

Additionally, this call is not on some of the other branches in this repository, but it is on the fork I made when doing some Mk2 changes so it is hard to see when or why this was added.

[BenBradnick]

@newville I think this can be closed now