Re: parsic/sh/sparc tracehook breakage when tracing signals

[Mike Frysinger]

On Thursday 11 February 2010 19:39:50 David Miller wrote:
> From: Mike Frysinger <vapier@xxxxxxxxxx>
> > when i ported the Blackfin code to the tracehook framework, i copied a
> > latent bug from the sparc port.  trying to trace another process while
> > handling signals no longer worked (and subsequently broke some of the gdb
> > tests).
> 
> What you seem to be missing is that on Sparc TIF_SINGLESTEP will never
> be set, because it does not support hardware single step.
> 
> This thread flag is not even defined on that platform.

my point is that these arches never call ptrace_notify(SIGTRAP) from their 
signal handlers.  maybe the arch is unable to due to some port/hardware 
limitation and this is currently the expected behavior.

dont get me wrong ... i dont particularly care if your arch has a bug in it 
here.  my arch did have a bug and on the off chance that others did too, i 
thought i'd drop an e-mail to people.  if your arch is in the "limited" 
category, feel free to ignore this.
-mike

Attachment: signature.asc
Description: This is a digitally signed message part.