problem of i386-stub

[d_luo27 at 263.net (d_luo27 at 263.net)]

> d_luo27 at 263.net wrote:
> > HI:
> > I build the target for pc386(com1 for console, com2 for debug)according to GDB.HOWTO.
> > the program always stop at waitabit( ). the process of program is OK if just go with one step or set breakpoint ,but continue is not.
> >
 erik.ivanenko wrote:
> What is the problem?  What happens when you "continue" from breakpoint?  Since you don't say, I can't guess.

yes, the program can not continue from  waitabit().if the program go by 'next' is right. There is no response for host in gdb, also no response at console. 

> > Another question is timer ISR is not be actived,when I test the sample with  a system call: rtems_task_wake_after(or  rtems_task_wake_when) .the thread can not wake after sleep.
> > this case is also occur when we target for
> > i386ex. some of code of  ROM monitor(we developed on another developing platform) is came from i386-stub.c.
> >
> > Is it a problem in i386-stub.c or i386-stub-glue.c
> >
erik.ivanenko wrote: 
> How can this be a problem?   You are asking the executive to wake a thread after a given number of ticks.  The timer is not ticking.  Result: sleep forever.
> What did you expect to happen????

 I mean when the program go to rtems_task_wake_after(or  rtems_task_wake_when), then I set a breakpoint at timerisr(), the program can not arrive in timerisr. and the thread can not be waked up. 
 But when run the target without the code in GDB.HOWTO , the thread can be waked up. (or without Rom monitor for i386ex).






_____________________________________________
????????--???????????????????? http://www.263.net
???????? ???????? ???????? ???????? ??????????
???????? ???????? ???????? ???????? ????????
???????? ???????? ???????? ???????? ????????