El 30/10/17 a las 23:40, Jason Ekstrand escribió: > On Thu, Oct 12, 2017 at 11:38 AM, Jose Maria Casanova Crespo > <jmcasan...@igalia.com <mailto:jmcasan...@igalia.com>> wrote: > > From: Alejandro Piñeiro <apinhe...@igalia.com > <mailto:apinhe...@igalia.com>> > > By default we don't set the rounding mode. We only set > round-to-near-even or round-to-zero mode if explicitly set from nir. > > v2: Use a single SHADER_OPCODE_RND_MODE opcode taking an immediate > with the rounding mode (Curro) > > Signed-off-by: Jose Maria Casanova Crespo <jmcasan...@igalia.com > <mailto:jmcasan...@igalia.com>> > Signed-off-by: Alejandro Piñeiro <apinhe...@igalia.com > <mailto:apinhe...@igalia.com>> > --- > src/intel/compiler/brw_fs_nir.cpp | 8 ++++++++ > 1 file changed, 8 insertions(+) > > diff --git a/src/intel/compiler/brw_fs_nir.cpp > b/src/intel/compiler/brw_fs_nir.cpp > index 6908c7ea02..b356836e80 100644 > --- a/src/intel/compiler/brw_fs_nir.cpp > +++ b/src/intel/compiler/brw_fs_nir.cpp > @@ -693,6 +693,14 @@ fs_visitor::nir_emit_alu(const fs_builder > &bld, nir_alu_instr *instr) > inst->saturate = instr->dest.saturate; > break; > > + case nir_op_f2f16_rtne: > + case nir_op_f2f16_rtz: > + if (instr->op == nir_op_f2f16_rtz) > + bld.emit(SHADER_OPCODE_RND_MODE, bld.null_reg_ud(), > brw_imm_d(BRW_RND_MODE_RTZ)); > + else if (instr->op == nir_op_f2f16_rtne) > + bld.emit(SHADER_OPCODE_RND_MODE, bld.null_reg_ud(), > brw_imm_d(BRW_RND_MODE_RTNE)); > + /* fallthrough */ > > > It might look a little nicer (though it's more lines of code) to have > a little brw_from_nir_rounding_mode helper and then we could have just > the one emit call. I don't care too much though.
I agree, and it could simplify if we want to enable other rounding modes supported by the hardware in the future. > > > + > /* In theory, it would be better to use BRW_OPCODE_F32TO16. > Depending > * on the HW gen, it is a special hw opcode or just a MOV, and > * brw_F32TO16 (at brw_eu_emit) would do the work to chose. > -- > 2.13.6 > > _______________________________________________ > mesa-dev mailing list > mesa-dev@lists.freedesktop.org <mailto:mesa-dev@lists.freedesktop.org> > https://lists.freedesktop.org/mailman/listinfo/mesa-dev > <https://lists.freedesktop.org/mailman/listinfo/mesa-dev> > > > > > _______________________________________________ > mesa-dev mailing list > mesa-dev@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/mesa-dev _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
