On Sun, May 5, 2013 at 1:01 PM, Maxime Villard <[email protected]> wrote: > I have been making a code scanner for a while, and I wanted to test > a new rule, so I scanned sys/pci/drm. It found an uninitialized > variable in intel_ddi.c. > > Quite simple, if !(val & DDI_BUF_CTL_ENABLE) at l.1420, the variable > 'wait' is not initialized at l.1432. > > I guess I should send it upstream, but, you know, I don't want to > subscribe to the ML only to send one small mail. Is there someone here > who's upstream or who's willing to send that patch? > > Thanks > > > Index: intel_ddi.c > =================================================================== > RCS file: /cvs/src/sys/dev/pci/drm/i915/intel_ddi.c,v > retrieving revision 1.1 > diff -u -r1.1 intel_ddi.c > --- intel_ddi.c 18 Mar 2013 12:36:52 -0000 1.1 > +++ intel_ddi.c 5 May 2013 10:56:40 -0000 > @@ -1412,7 +1412,7 @@ > struct intel_dp *intel_dp = &intel_dig_port->dp; > struct inteldrm_softc *dev_priv = encoder->dev->dev_private; > enum port port = intel_dig_port->port; > - bool wait; > + bool wait = false; > uint32_t val; > > if (I915_READ(DP_TP_CTL(port)) & DP_TP_CTL_ENABLE) { >
Also, a previous declaration of "bool wait" (line 1306 into intel_ddi_post_disable()) is already initialized as "bool wait = false;", so this really sounds like a bug. ok for me to commit this patch. ciao, David
