Makes sense. Jose
----- Original Message ----- > The assertion added in f09910f3 broke nv50 completely by asserting > that > the number of elements in a dereferenced pointer (i.e. 1) was greater > than i (which ranged up to six), rather than checking the number of > elements in the containing array. > > Signed-off-by: Daniel Stone <dan...@fooishbar.org> > --- > src/gallium/drivers/nv50/nv50_pc_regalloc.c | 2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/src/gallium/drivers/nv50/nv50_pc_regalloc.c > b/src/gallium/drivers/nv50/nv50_pc_regalloc.c > index 72922cb..12a59cb 100644 > --- a/src/gallium/drivers/nv50/nv50_pc_regalloc.c > +++ b/src/gallium/drivers/nv50/nv50_pc_regalloc.c > @@ -421,7 +421,7 @@ phi_opnd_for_bb(struct nv_instruction *phi, > struct nv_basic_block *b, > int i, j; > > for (j = -1, i = 0; i < 6 && phi->src[i]; ++i) { > - assert(i < Elements(phi->src[i])); > + assert(i < Elements(phi->src)); > srci = phi->src[i]; > /* if already replaced, check with original source first */ > if (srci->flags & NV_REF_FLAG_REGALLOC_PRIV) > -- > 1.7.9 > > _______________________________________________ > mesa-dev mailing list > mesa-dev@lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/mesa-dev > _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev