Re: [Qemu-devel] [PATCH 1/1] log hmp/qmp command

[Denis V. Lunev]

On 10/12/2015 04:49 PM, Eric Blake wrote:
> On 10/12/2015 02:41 AM, Denis V. Lunev wrote:
> > From: Pavel Butsykin <address@hidden>
> >
> > This log would be very welcome for long-term diagnostics of the system
> > in the production. This log is at least necessary to understand what
> > has been happened on the system and to identify issues at higher-level
> > subsystems (libvirt, etc).
> >
> > Signed-off-by: Pavel Butsykin <address@hidden>
> > Signed-off-by: Denis V. Lunev <address@hidden>
> > CC: Markus Armbruster <address@hidden>
> > CC: Luiz Capitulino <address@hidden>
> > CC: Eric Blake <address@hidden>
> > ---
> > @@ -3822,6 +3824,8 @@ static void handle_qmp_command(JSONMessageParser *parser, 
> > QList *tokens)
> >           error_setg(&local_err, QERR_JSON_PARSING);
> >           goto err_out;
> >       }
> > +    qemu_log_mask(LOG_CMD, "qmp \"%s\" requested\n",
> > +                  qobject_to_json(obj)->string);
> >   
> In addition to the leak already pointed out, qobject_to_json() can be
> expensive, and it looks like we are doing that work unconditionally even
> if the logging is not turned on. Is there a way to optimize so that the
> conversion is only done when logging is enabled?
this seems quite reasonable.

Will it be good if we'll change logging implementation to
use macros, something like

#define qemu_log_mask(int mask, const char *ftp, ...) \
do { \
va_list ap; \
  \
if (!(qemu_loglevel & mask) || !qemu_logfile) {\
      break;\
  }\
  qemu_log(ftp, __VA_ARGS__); \
} while (0)

This IMHO makes sense even out of the scope of this patch.

Den