This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: RFA [PATCH v4] Implement 'catch syscall' for gdbserver (was Re: RFA [PATCH v3] Implement 'catch syscall' for gdbserver)


>>>>> "Pedro" == Pedro Alves <palves@redhat.com> writes:

>> +QCatchSyscalls:1 [;SYSNO]...
>> +QCatchSyscalls:0
>> +  Enable ("QCatchSyscalls:1") or disable ("QCatchSyscalls:0")
>> +  catching syscalls from the inferior process.

Pedro> So, "catch syscall" is per-inferior/process on the GDB side, but
Pedro> this always sets the catchpoints on all processes.  Was that
Pedro> intended?

I wonder whether it is the right thing on the gdb side.

Right now we have the rule that linespecs for breakpoints apply to all
inferiors; but this rule isn't followed for catchpoints.

I tend to think it ought to be, for consistency and simplicity; followed
up by using it{etc}sets for filtering out uninteresting events.


I don't want to derail this patch though.

And arguably it is ok for gdb to present one thing to the user but more
useful for gdbserver to present a different view to gdb.

Tom


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]