This is the mail archive of the
gdb@sourceware.org
mailing list for the GDB project.
Re: GDB paper for GCC summit: New Commands
- From: Eli Zaretskii <eliz at gnu dot org>
- To: pgilliam at us dot ibm dot com
- Cc: gdb at sources dot redhat dot com
- Date: Sat, 01 Apr 2006 16:21:11 +0300
- Subject: Re: GDB paper for GCC summit: New Commands
- References: <1142633147.21920.400.camel@dufur.beaverton.ibm.com> <1143681561.21920.426.camel@dufur.beaverton.ibm.com>
- Reply-to: Eli Zaretskii <eliz at gnu dot org>
> From: PAUL GILLIAM <pgilliam@us.ibm.com>
> Date: Wed, 29 Mar 2006 17:19:21 -0800
>
> 1) How do "set|show detach-on-fork" and "set|show follow-fork" work
> together?
Do you mean "set|show follow-fork-mode"? If so, the answer is in the
manual (in the node "Processes"); if something there is unclear,
please tell what parts need to be clarified.
> I think this is the answer: If 'set detach-on-fork' is
> off, the 'set follow-fork' works as before. If 'set detach-on-fork'
> is on, then the state of 'set follow-fork' determines which for will
> be the 'current' fork. Is this correct?
I think it's the other way around: if detach-on-fork is ON, then "set
follow-fork" behaves as it did before: one of the two processes is
debugged bu GD, while the other runs unimpeded. If it's OFF, then
both parent and the child are under GDB's control.
Again, I think this is stated quite clearly in the manual, so if
anything there sounds confusing, please identify the confusing parts.
> 2) Does the new fork features also work with vfork?
The text suggests that they do, but I don't know if this is correct as
a matter of fact.
> 3) If the name of the main procedure is 'main', is the start command
> equivalent
> to 'tbreak main' followed by 'run', or is there something subtle
> going on?
AFAIK, "start" is equivalent to "tbreak main; run".