On Wed, Jul 03, 2019 at 06:25:01PM +0100, Daniel P. Berrangé wrote:
> On Wed, Jul 03, 2019 at 07:10:05PM +0200, Philippe Mathieu-Daudé wrote:
> > @@ -306,6 +308,13 @@ bool trace_init_backends(void)
> > openlog(NULL, LOG_PID, LOG_DAEMON);
> > #endif
> >
> > + trace_init_file(getenv("QEMU_TRACE_LOGFILE"));
> > + trace_init_events(getenv("QEMU_TRACE_EVENTFILE"));
> > + trace_env = getenv("QEMU_TRACE_EVENTS");
> > + if (trace_env) {
> > + trace_enable_events(trace_env);
> > + }
> > +
>
> I don't think it is a nice idea to add this via environment variables
> to QEMU itself. Why not modify libqtest qtest_init_without_qmp_handshake
> to read the env vars and then pass a suitable -trace arg when spawning
> QEMU ?What is the concern about adding these environment variables to QEMU? It is convenient to be able to use tracing even if QEMU is invoked by something you cannot modify/control. The main issues I see with environment variables are: 1. Security. Is there a scenario where an attacker can use environment variables to influence the behavior of a QEMU process running at a different trust level? 2. Name collision. What is the chance that existing users already define environment variables with these names and that unexpected behavior could result? Stefan
signature.asc
Description: PGP signature
