This is the mail archive of the java@gcc.gnu.org mailing list for the Java project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: sysdep/i386/backtrace.h infinite loop?


Andrew Haley wrote:
Marco Trudel writes:
> Andrew Haley wrote:
> > Marco Trudel writes:
> > > Maybe a stupid question, but how can this loop in > > > sysdep/i386/backtrace.h ever end:
> > > > > > register _Jv_uintptr_t *_ebp __asm__ ("ebp");
> > > _Jv_uintptr_t rfp;
> > > for (rfp = *_ebp; rfp; rfp = *(_Jv_uintptr_t *)rfp)
> > > {
> > > // only read access for rfp
> > > }
> > > > > > Maybe because "*(_Jv_uintptr_t *)rfp" is changing it, but I can't see how...
> > > > It looks to me like it's walking down a linked list until it hits NULL.
> > Yeah, but where happens the down-walking? In my opinion we do > always-the-same-value-walking...


What, in your opinion, does

rfp = *(_Jv_uintptr_t *)rfp;

do?

Nothing. But as I said already this is probably doing something. But I can't see what. On second thought, what about: Setting rfp to the value stored at the location which was addressed by the value of the old rfp?


Ok, I see. Something is being done and I should work on my C knowledge if I don't understand it. Will do :-)


thanks Marco


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]