[Bug breakpoints/17549] New: catch load seems to be inferior-specific

[Bug breakpoints/17549] New: catch load seems to be inferior-specific

[tromey at sourceware dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=17549

            Bug ID: 17549
           Summary: catch load seems to be inferior-specific
           Product: gdb
           Version: 7.7
            Status: NEW
          Severity: normal
          Priority: P2
         Component: breakpoints
          Assignee: unassigned at sourceware dot org
          Reporter: tromey at sourceware dot org

I am debugging firefox in multi-inferior mode.

I did a "catch exec" to catch when the child process
execs.  Then I did "catch load" -- but since I forgot again
about bug #13700, I did this with the parent inferior
selected.

My catchpoint never triggered, even though the child
process loaded many shared libraries.

I think "catch load" should not be inferior-specific.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug breakpoints/17549] catch load seems to be inferior-specific

[palves at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=17549

Pedro Alves <palves at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |palves at redhat dot com

--- Comment #1 from Pedro Alves <palves at redhat dot com> ---
Agreed -- I actually think that no catchpoint should be inferior-specific.

See Bug 13457 too.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug breakpoints/17549] catch load seems to be inferior-specific

[xdje42 at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=17549

Doug Evans <xdje42 at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |xdje42 at gmail dot com

--- Comment #2 from Doug Evans <xdje42 at gmail dot com> ---
Pedro, Can I ask for elaboration on what "I actually think that no catchpoint
should be inferior-specific"?

I can well imagine wanting some catchpoints in only specific inferiors.
[Same for breakpoints, watchpoints, and everything else too.]

For myself, I would even prefer that to be the default.
[which I know it isn't today, but one could add an option to allow it]

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug breakpoints/17549] catch load seems to be inferior-specific

[palves at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=17549

--- Comment #3 from Pedro Alves <palves at redhat dot com> ---
I meant, by default.  I can certainly see how one may want to restrict a
catchpoint to an inferior, but I think that by default, a catchpoint should
trigger in all inferiors, like breakpoints.

I think restricting breakpoints to an inferior by default is more confusing and
surprising to users; users naturally miss/forget this, as evidenced by this PR.
 Bug 13457 shows another example that's easy to miss.

Then if the user wants to restrict it to some inferior, thread, etc., then she
should be able to do that, ideally with itsets (which I've been working on --
all the recent-ish run control fixes are preparatory work for
all-stop-on-non-stop, which is a prerequisite for itsets).

-- 
You are receiving this mail because:
You are on the CC list for the bug.