On Mon, Jun 30, 2014 at 10:32 AM, Ian Romanick <i...@freedesktop.org> wrote: > On 06/30/2014 10:11 AM, Matt Turner wrote: >> Setting a couple of bits is the same cost or less as conditionally >> setting a couple of bits. >> --- >> src/mesa/drivers/dri/i965/brw_vec4_generator.cpp | 9 +++------ >> 1 file changed, 3 insertions(+), 6 deletions(-) >> >> diff --git a/src/mesa/drivers/dri/i965/brw_vec4_generator.cpp >> b/src/mesa/drivers/dri/i965/brw_vec4_generator.cpp >> index 8f5b205..509c32c 100644 >> --- a/src/mesa/drivers/dri/i965/brw_vec4_generator.cpp >> +++ b/src/mesa/drivers/dri/i965/brw_vec4_generator.cpp >> @@ -1266,12 +1266,9 @@ vec4_generator::generate_code(exec_list *instructions) >> >> brw_inst *last = &p->store[pre_emit_nr_insn]; >> >> - if (inst->conditional_mod) >> - brw_inst_set_cond_modifier(brw, last, inst->conditional_mod); >> - if (inst->no_dd_clear) >> - brw_inst_set_no_dd_clear(brw, last, true); >> - if (inst->no_dd_check) >> - brw_inst_set_no_dd_check(brw, last, true); >> + brw_inst_set_cond_modifier(brw, last, inst->conditional_mod); >> + brw_inst_set_no_dd_clear(brw, last, inst->no_dd_clear); >> + brw_inst_set_no_dd_check(brw, last, inst->no_dd_check); > > The old code will only set the bits. The new code might also clear the > bits, right? Is that potentially a problem?
Ah, the missing piece of information here is that the bits are already zero when this code is called. _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
