Re: ppc32 debugging ppc64, part 1

[Richard Henderson <rth at redhat dot com>]

On Mon, Sep 12, 2005 at 03:36:37PM -0700, Richard Henderson wrote:
> But for the record, I was seeing odd things and wanted to get the
> libc code out of the way.  I later discovered the typo in my own
> code, but I never got around to reverting the use of syscall.

Oh, yes, one could in fact argue that libc is broken.  It doesn't
handle the PTRACE_PEEK*_3264 commands specially, as it does with
PTRACE_PEEK*.  So you'd have to write 

  if (cmd == *_3264)
    ptrace (cmd, tid, addr, buf)
  else
    *buf = ptrace (cmd, tid, addr, 0);

... and then rely on libc never being fixed.  And in this case I
don't see what going through libc buys us besides frustration.



r~