Hi Peter, the problem with the segv in your modified example vanishes if you create the canvas outside the thread, e.g. as a global canvas, or as a canvas belonging to a class as a member. (see examples mhs3 and TMhs on http://www-linux.gsi.de/~go4/HOWTOthreads/, where we use global graphics pads) >From our tests in the course of these examples we always had trouble when the TCanvas was instantiated in the thread itself; maybe this is due to the fact that any Xaction on the canvas is handled by the TThread timer using a thread specific data array pointing to that canvas. However, I did not investigate that in detail; in fact for our purposes we always use canvases which are already existing when the threads start working on them. Peter Christ wrote: > > Hi, > > I'm totally new to ROOT and I plan to use it to develop a program for > data-aquisition. For that the data display should be updated whenever > new data are available (rate: few/s to 1/10s). > The problem: the display should not be locked then i.e. it should be > possible to change ranges etc. > Is that possible - and how? > Since we use threads for a similar purpose, you might have a look at the website of the GSI go4 project http://go4.gsi.de/ where you can find documentation of a display which is asynchronously updated by a remote analysis task. Moreover, we could give you a Threadmanager library based on the root threads which might be useful for development of any own threaded applications in root. (this library might be officially released in the near future, but is already working). Please let me know if there should be any further questions and problems concerning the examples, and your experiences with the root threads. Best regards, Joern -- ///////////////////////////////////////////////////////////////////// // Dr. J"orn Adamczewski (J.Adamczewski@gsi.de) // // GO4 project team Tel: +49-6159-71-2554 // // Datenverarbeitung & ExperimentElektronik FAX: +49-6159-71-2986 // // Ges. f. SchwerIonenforschung, Planckstr.1, D-64291 Darmstadt //
This archive was generated by hypermail 2b29 : Tue Jan 01 2002 - 17:51:02 MET