avr-gcc-list
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Bug in avr-gdb? was:Re: [avr-gcc-list] AVR simulator and interrupts


From: Klaus Rudolph
Subject: Re: Bug in avr-gdb? was:Re: [avr-gcc-list] AVR simulator and interrupts
Date: Tue, 20 Jul 2004 15:54:34 +0200
User-agent: Mozilla/5.0 (X11; U; Linux i686; de-AT; rv:1.6) Gecko/20040114

Theodore A. Roth wrote:

On Sun, 18 Jul 2004, Klaus Rudolph wrote:

Hi Theodore, hi James and the list :-)

Is there a problem with my solution, maybe with other targets as well?
If not it would be nice if this fix could be included to gdb.

Your fix might not be needed. I just tried to duplicate the problem with
the current gdb cvs head and it doesn't show up when running either your
simulator or mine.


Yes, the actual cvs head of gdb handle the problem in a complete different way.
IRQ´s are runnig "behind" the step. So there is no step to the __vector. And
also if you set a breakpoint at address 0x0e the next step is a stepi and the following is then the step to the next source line in irq handler. That is abolutly that what I want to see there! Great, the bug is fixed :-) So lets wait until the release comes.... Maybe we should sometimes
test the head, I have seen bugs come back :-)

Applying the change you suggest to infcmd.c
doesn't change the behaviour in any way I could notice.

That could I not reproduce :-). Maybe you inserted the code at the wrong place? There are two times the same error message in infcmd.c :-) And don´t forget to recompile and maybe you should use the new compiled gdb which is without install under
the gdb src path? :-)))

Please try cvs head and the recently cut 6.2 branch and if either still
exhibits the problem I'll look into it more. I don't want to chase down
a bug that has already been fixed. :-)

I only tested the actual head for now. I need more then 45 minutes to get an "update" of the cvs sources which is much time and so I did not test 6.2 branch. But if you have tested that it
should work because the same software should give same results :-)
Maybe James could do the test for the 6.2 branch. He is the originator of the problem
so he must have the bugfix and so he should do the test :-)


Nice to read you...

   Thanks
      Klaus




reply via email to

[Prev in Thread] Current Thread [Next in Thread]