I have an application that uses the FrontPanel 4.0.8 library running under Fedora Core 15.
This application runs fine when executed on a PC with a native install of Fedora Core 15.
When executed on a PC running Fedora Core 15 in VMware under Windows 7 or XP okFrontPanel_GetDeviceCount returns 0.
The application worked correctly under earlier versions of Fedora under VMware using earlier version of FrontPanel.
I’m thinking about trying a more recent version of FrontPanel, or compiling the application natively under Windows, hopefully with gcc, but I thought I would see if anyone had any other thoughts or suggestions.
There are no silly questions in this business. I will check with our prime contractor, as I have never tried to run it under VMware myself.
They had been running the application with root privileges, but that might have changed, along with what is needed to attached to the underlying USB device.
Another question, if I may. I tried compiling my application, written in “C”, under FC 17 and got the following compile time errors:
…/ok/dist/okFrontPanelDLL.h:278:2: error: unknown type name ?bool?
…/ok/dist/okFrontPanelDLL.h:279:2: error: unknown type name ?bool?
…/ok/dist/okFrontPanelDLL.h:280:2: error: unknown type name ?bool?
If I change “bool” in the following lines to “Bool” then it compiles and links successfully:
On a quick read of …/ok/dist/okFrontPanelDLL.h it seems to use both Bool and bool for different purposes.
I see they are members of a new structure in FrontPanel 4.2.5. Is my change reasonably valid, or will it cause other problems?
The system particulars are:
[LIST]
*] Linux 3.6.7-4.fc17.x86_64 #1 SMP Tue Nov 20 19:40:01 UTC 2012
*] gcc version 4.7.2 20120921 (Red Hat 4.7.2-2) (GCC)
[/LIST]
Bruno-- Thanks for pointing out the issue with bool. It’s not too often that we have customers apply the code in straight-C – most folks build in C++. But the API is designed to apply in both so we’ll chalk this up to an oversight.
We’ll log this as an issue and make an update.
Most gcc compilers (I believe) generate a bool that is sizeof(bool)=1. So you would likely be safe replacing those bool instances with (unsigned char) until we can update the particulars.