[remote] Where is S AA p PID currently specified?

[remote] Where is S AA p PID currently specified?

[Pedro Alves]

[Resend - it seems the first time didn't make it]

Hi,

I see this in remote.c:remote_wait, while handling the 'S' and 'T'
stop reply packets:

      switch (buf[0])
       {
        ...

        case 'T':
          ...

	  /* fall through */
	case 'S':		/* Old style status, just signal only.  */
	  if (solibs_changed)
	    status->kind = TARGET_WAITKIND_LOADED;
	  else
	    {
	      status->kind = TARGET_WAITKIND_STOPPED;
	      status->value.sig = (enum target_signal)
		(((fromhex (buf[1])) << 4) + (fromhex (buf[2])));
	    }

	  if (buf[3] == 'p')
	    {
	      thread_num = strtol ((const char *) &buf[4], NULL, 16);
	      record_currthread (thread_num);
	    }
	  goto got_status;


Where's that 'p' after S AA specified?  I don't see it in the docs,
in either S or T stop reply packets description.  Looking through the
file history, it seemed this had to do with some old Cisco
extensions that have since been removed, but this bit got left behind.

Can we remove it ?

-- 
Pedro Alves

Re: [remote] Where is S AA p PID currently specified?

[Michael Snyder]

On Tue, 2008-06-10 at 20:42 +0100, Pedro Alves wrote:

> Where's that 'p' after S AA specified?  I don't see it in the docs,
> in either S or T stop reply packets description.  Looking through the
> file history, it seemed this had to do with some old Cisco
> extensions that have since been removed, but this bit got left behind.

Uh-oh.  If that's true, it probably pre-dates the current 
cvs repo.  Let me look.

What gave you the impression it was related to Cisco?

[remote] Where is S AA p PID currently specified?

[Pedro Alves]

Hi,

I see this in remote.c:remote_wait, while handling the 'S' and 'T'
stop reply packets:

      switch (buf[0])
       {
        ...

        case 'T':
          ...

	  /* fall through */
	case 'S':		/* Old style status, just signal only.  */
	  if (solibs_changed)
	    status->kind = TARGET_WAITKIND_LOADED;
	  else
	    {
	      status->kind = TARGET_WAITKIND_STOPPED;
	      status->value.sig = (enum target_signal)
		(((fromhex (buf[1])) << 4) + (fromhex (buf[2])));
	    }

	  if (buf[3] == 'p')
	    {
	      thread_num = strtol ((const char *) &buf[4], NULL, 16);
	      record_currthread (thread_num);
	    }
	  goto got_status;


Where's that 'p' after S AA specified?  I don't see it in the docs,
in either S or T stop reply packets description.  Looking through the
file history, it seemed this had to do with some old Cisco
extensions that have since been removed, but this bit got left behind.

Can we remove it ?

-- 
Pedro Alves

Re: [remote] Where is S AA p PID currently specified?

[Michael Snyder]

On Tue, 2008-06-10 at 19:44 +0100, Pedro Alves wrote:
> Hi,
> 
> I see this in remote.c:remote_wait, while handling the 'S' and 'T'
> stop reply packets:
> 
[...]
> 	case 'S':		/* Old style status, just signal only.  */
> 	  if (solibs_changed)
> 	    status->kind = TARGET_WAITKIND_LOADED;
> 	  else
> 	    {
> 	      status->kind = TARGET_WAITKIND_STOPPED;
> 	      status->value.sig = (enum target_signal)
> 		(((fromhex (buf[1])) << 4) + (fromhex (buf[2])));
> 	    }
> 
> 	  if (buf[3] == 'p')
> 	    {
> 	      thread_num = strtol ((const char *) &buf[4], NULL, 16);
> 	      record_currthread (thread_num);
> 	    }
> 	  goto got_status;
> 
> 
> Where's that 'p' after S AA specified?  I don't see it in the docs,
> in either S or T stop reply packets description.  Looking through the
> file history, it seemed this had to do with some old Cisco
> extensions that have since been removed, but this bit got left behind.
> 
> Can we remove it ?

OK, you're right -- that 'p' suffix came in between releases 
4.18 and 5.0, and the earliest code has comments mentioning
Cisco kernel threads.

Nuke it!

;-)
Michael

Re: [remote] Where is S AA p PID currently specified?

[Pedro Alves]

A Wednesday 11 June 2008 00:43:27, Michael Snyder wrote:
> On Tue, 2008-06-10 at 19:44 +0100, Pedro Alves wrote:

> > Where's that 'p' after S AA specified?  I don't see it in the docs,
> > in either S or T stop reply packets description.  Looking through the
> > file history, it seemed this had to do with some old Cisco
> > extensions that have since been removed, but this bit got left behind.
> >
> > Can we remove it ?
>
> OK, you're right -- that 'p' suffix came in between releases
> 4.18 and 5.0, and the earliest code has comments mentioning
> Cisco kernel threads.
>
> Nuke it!
>
> ;-)

Thanks for confirming!  I'll post patch in a sec.

-- 
Pedro Alves