Em Tue, May 19, 2015 at 04:28:15PM +0900, Namhyung Kim escreveu:
> On Mon, May 18, 2015 at 09:52:59AM -0300, Arnaldo Carvalho de Melo wrote:
> > Em Mon, May 18, 2015 at 09:30:48AM +0900, Namhyung Kim escreveu:
> > > With multi-thread report, separate sessions can be passed to each
> > > thread, in this case we should keep a single machine state for all
> > > struct sessions. Separate machines and have a pointer in sessions.
> >
> > I had to look at all the patch to semi-figure this out, i.e. you said it
> > should be separated from 'perf_session', agreed.
> >
> > But who will create it? How will it be passed to the perf_session
> > instances?
> >
> > Most of the patch is making session->machines be turned into a pointer,
> > but the meat, i.e. who creates it, is unclear, I see a malloc in
> > perf_session__new(), where I was kinda expecting that a higer layer,
> > perhaps in struct tool? Would create the list of all machines (struct
> > machines) and then pass it to multiple perf_session__new() calls.
> >
> > But then perf_session__delete() calls 'free(session->machines)', huh?
>
> OK. So, this is what I have in my head:
>
> perf_tool__create_machines(tool) {
> tool->machines = malloc();
> machines__init(tool->machines);
> }
Probably, but then in this case you would call machines__new(), that
does the malloc and init.
>
> perf_session__new(file, repipe, tool) {
> session->machines = tool->machines;
> ...
> }
That could be ok.
>
> perf_tool__delete_machines(tool) {
> /* call machines-related destructors */
> free(tool->machines);
> }
That would be machines__delete(tool->machines), that calls
machine__exit() and then does the free.
- Arnaldo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
