Re: [Kgdb-bugreport] [PATCH][3/3] Update CVS KGDB's wrt connect / detach

[Tom Rini <trini@kernel.crashing.org>]

On Thu, Feb 26, 2004 at 01:44:49PM +0530, Amit S. Kale wrote:

> On Thursday 26 Feb 2004 3:23 am, Tom Rini wrote:
> > The following patch fixes a number of little issues here and there, and
> > ends up making things more robust.
> > - We don't need kgdb_might_be_resumed or kgdb_killed_or_detached.
> >   GDB attaching is GDB attaching, we haven't preserved any of the
> >   previous context anyhow.
> 
> If gdb is restarted, kgdb has to remove all breakpoints. Present kgdb does 
> that in the code this patch removes:
> 
> -		if (remcom_in_buffer[0] == 'H' && remcom_in_buffer[1] == 'c') {
> -			remove_all_break();
> -			atomic_set(&kgdb_killed_or_detached, 0);
> -			ok_packet(remcom_out_buffer);
> 
> If we don't remove breakpoints, they stay in kgdb without gdb not knowing it 
> and causes consistency problems.

Er, what do you mean 'restarted' ?  If gdb somehow disconnects 'detach'
or ^D^D, remove_all_break() gets called.  Is there another way for gdb
to somehow disconnect that I don't know of?

> > - Don't try and look for a connection in put_packet, after we've tried
> >   to put a packet.  Instead, when we receive a packet, GDB has
> >   connected.
> 
> We have to check for gdb connection in putpacket or else following problem 
> occurs.
> 
> 1. kgdb console messages are to be put.
> 2. gdb dies

As in doesn't cleanly remove itself?

> 3. putpacket writes the packet and waits for a '+'
> 4. new gdb sends a protocol initialization packet
> 5. putpacket reads characters in that packet hoping for an incoming '+' 
> sending out console message packet on each incoming character
> 6. gdb receives and rejects each console message packet
> 
> > - Remove ok_packet(), excessive, IMHO.
> 
> ok_packet is better than littering "OK" all over the place.

I disagree.  If ok_packet was anything more than
strcpy(remcom_out_buffer, "OK") you'd be right.

-- 
Tom Rini
http://gate.crashing.org/~trini/