Community
Participate
Working Groups
When an solib event happens, GDB 7.4 outputs: [1,327,436,056,817] *stopped,reason="solib-event",thread-id="1",stopped-threads="all",core="1" while before it output: [1,327,436,104,280] ~"Stopped due to shared library event\n" [1,327,436,104,281] *stopped,thread-id="1",stopped-threads="all",core="0" CDI triggers off the "Stopped due to shared library event" string which is gone. Although CDI was already looking for the reason field, it was expecting "shlib-event" instead of "solib-event"
The result of this bug is that CDI, with GDB 7.4 will always stop at every solib event, even if the user didn't choose that option.
Created attachment 210009 [details] Fix Fix to look for the 'solib-event' reason. I think this should go both into master and cdt_8_0
Mikhail, can you review and let me know if you agree? This brought to my attention by Tom Tromey: http://sourceware.org/ml/gdb-patches/2012-01/msg00828.html
BTW, this is not a problem with DSF because we don't do anything with solib-event yet: Bug 309704
(In reply to comment #3) > Mikhail, can you review and let me know if you agree? > > This brought to my attention by Tom Tromey: > http://sourceware.org/ml/gdb-patches/2012-01/msg00828.html Hi Mikhail, any opinion on this one-line fix for CDI, I would like to put it in the 8.0.2 release which is next Monday. If you don't have time to review, I'll commit this later this week, as I think it is pretty safe. Thanks
(In reply to comment #5) > (In reply to comment #3) > > Mikhail, can you review and let me know if you agree? > > > > This brought to my attention by Tom Tromey: > > http://sourceware.org/ml/gdb-patches/2012-01/msg00828.html > > Hi Mikhail, > > any opinion on this one-line fix for CDI, I would like to put it in the 8.0.2 > release which is next Monday. > > If you don't have time to review, I'll commit this later this week, as I think > it is pretty safe. Marc, Sorry, I've lost the review reminder. I don't remember why we are looking for 'shlib-event' and want to uderstand it first. Will do it later today. Thanks.
(In reply to comment #6) > Marc, > Sorry, I've lost the review reminder. I don't remember why we are looking for > 'shlib-event' and want to uderstand it first. Will do it later today. > Thanks. Actually, you are right. Why not play it safe and do: } else if ("shlib-event".equals(reason) || "solib-event".equals(reason)) { instead of removing the "shlib-event". I'd feel safer with that anyway.
Created attachment 210615 [details] Safer fix Here is the safer fix. Let me know what you think. Thanks
(In reply to comment #8) > Created attachment 210615 [details] > Safer fix > > Here is the safer fix. > Let me know what you think. > > Thanks Yes, this is a better solution. The only link to "shlib-event" (except this discussion) I found was http://permalink.gmane.org/gmane.comp.gdb.devel/6063. It seems that it was and maybe is still used by Apple's GDB.
Thanks Mikhail. I have committed the "safer fix" patch to both master and cdt_8_0.
*** cdt git genie on behalf of Marc Khouzam *** Bug 369594: GDB 7.4 now reports solib events in MI, but no longer in CLI [*] http://git.eclipse.org/c/cdt/org.eclipse.cdt.git/commit/?id=8a73a23e9d27e3b2065394ef76c0e722226cf700 *** cdt git genie on behalf of Marc Khouzam *** Bug 369594: GDB 7.4 now reports solib events in MI, but no longer in CLI [*] http://git.eclipse.org/c/cdt/org.eclipse.cdt.git/commit/?id=8a73a23e9d27e3b2065394ef76c0e722226cf700
*** cdt git genie on behalf of Marc Khouzam *** Bug 369594: GDB 7.4 now reports solib events in MI, but no longer in CLI [*] http://git.eclipse.org/c/cdt/org.eclipse.cdt.git/commit/?id=dbdee9975bbfd8f3608cfe817f2012c8314ef3f3 *** cdt git genie on behalf of Marc Khouzam *** Bug 369594: GDB 7.4 now reports solib events in MI, but no longer in CLI [*] http://git.eclipse.org/c/cdt/org.eclipse.cdt.git/commit/?id=dbdee9975bbfd8f3608cfe817f2012c8314ef3f3
