I am using OMAP3515 processor (Generic CortexA8 Device) for my project, I was working on exception handling but this processor dos not generate any exception when any divide by zero instruction is given , it simply goes not next instruction. I want to know whether any configuration available to generate divide by zero exception.
Asked
Active
Viewed 1,197 times
2
-
Are you sure you're not dividing a floating-point operand by zero? – barak manos Dec 31 '14 at 10:10
-
perhaps you need to enable the exceptions? – old_timer Dec 31 '14 at 23:51
-
Hi , I have enabled FPSCR for DZE but even then it is not going to any exception. – vivek Jan 02 '15 at 08:10
1 Answers
2
Cortex A8 does not support divide-by-zero for integer operations. Other than normal interrupts and debug exceptions, the only fault exceptions supported are data abort, prefetch abort, and undefined instruction
The VFP supports divide-by-zero exceptions for floating point operations, which must be explicitly enabled via the FPSCR DZE bit.
Clifford
- 88,407
- 13
- 85
- 165
-
Hi , I have enabled FPSCR for DZE but even then it is not going to any exception. – vivek Jan 02 '15 at 05:18
-
@vivek : Earilier you said you were not using floating-point, so I would not expect an FP exception. If you are using floating point if the hardware enabled ([FPEXC:EN](http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0344f/Ccdhcfga.html))? And is hardware divide supported ([MVFR0:D](http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0344f/Chdebegb.html)) on your target? Have you built your code for hardware floating point (VFP) support? See also [this](http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0344k/Beihdjje.html) i.e. RTFM. – Clifford Jan 02 '15 at 08:59