> Date: Wed, 27 May 2020 19:39:07 +1000 > From: Jonathan Gray <j...@jsg.id.au> > > When testing the row and column increase for efifb on a 1920x1080 > display I noticed the top part of the screen continues to contain part > of a white on blue line from earlier in the dmesg even after the machine > has finished booting. > > RI_CENTER changes the ri_bits offset, doing RI_CLEARMARGINS in cnremap > clears the fragment of a line caused by using RI_CENTER.
ok kettenis@ > Index: efifb.c > =================================================================== > RCS file: /cvs/src/sys/arch/amd64/amd64/efifb.c,v > retrieving revision 1.31 > diff -u -p -r1.31 efifb.c > --- efifb.c 27 May 2020 07:48:02 -0000 1.31 > +++ efifb.c 27 May 2020 09:27:50 -0000 > @@ -219,7 +219,7 @@ efifb_attach(struct device *parent, stru > crow = ri->ri_crow; > > efifb_rasops_preinit(fb); > - ri->ri_flg &= ~RI_CLEAR; > + ri->ri_flg &= ~(RI_CLEAR | RI_CLEARMARGINS); > ri->ri_flg |= RI_VCONS | RI_WRONLY; > > rasops_init(ri, efifb_std_descr.nrows, efifb_std_descr.ncols); > @@ -478,7 +478,7 @@ efifb_cnremap(void) > > efifb_rasops_preinit(fb); > ri->ri_flg &= ~RI_CLEAR; > - ri->ri_flg |= RI_CENTER | RI_WRONLY; > + ri->ri_flg |= RI_CENTER | RI_WRONLY | RI_CLEARMARGINS; > > rasops_init(ri, efifb_std_descr.nrows, efifb_std_descr.ncols); > > >
