fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-01-05 02:30:18 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

microsoft/compiler: correct dxil fma opcode

Browse source 
When I originally added the FFMA opcode here, I added the FMAD opcode
instead of the FMA opcode. The reason for this is that it works on
32-bit values as well, so that seemed like a better fit.

But that's not correct, as the FMA opcode isn't a fused operation, so
let's correct the opcode.

This isn't currently in use, because we currently lower away all ffma
opcodes on the NIR level, but that's about to change.

While we're at it, let's also update the opcode name to match the DXIL
documentation.

Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8349>
This commit is contained in:
Erik Faye-Lund 2021-01-06 08:51:18 +01:00 • committed by Marge Bot
parent 317ec8b501
commit c2fa965cc3
1 changed files with 2 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
src/microsoft/compiler/nir_to_dxil.c
View file

@ -207,7 +207,7 @@ enum dxil_intr {
DXIL_INTR_UMAX = 39,
DXIL_INTR_UMIN = 40,
DXIL_INTR_FFMA = 46,
DXIL_INTR_FMA = 47,
DXIL_INTR_CREATE_HANDLE = 57,
DXIL_INTR_CBUFFER_LOAD_LEGACY = 59,
@ -1956,7 +1956,7 @@ emit_alu(struct ntd_context *ctx, nir_alu_instr *alu)
case nir_op_fsqrt: return emit_unary_intin(ctx, alu, DXIL_INTR_SQRT, src[0]);
case nir_op_fmax: return emit_binary_intin(ctx, alu, DXIL_INTR_FMAX, src[0], src[1]);
case nir_op_fmin: return emit_binary_intin(ctx, alu, DXIL_INTR_FMIN, src[0], src[1]);
case nir_op_ffma: return emit_tertiary_intin(ctx, alu, DXIL_INTR_FFMA, src[0], src[1], src[2]);
case nir_op_ffma: return emit_tertiary_intin(ctx, alu, DXIL_INTR_FMA, src[0], src[1], src[2]);
case nir_op_unpack_half_2x16_split_x: return emit_f16tof32(ctx, alu, src[0]);
case nir_op_pack_half_2x16_split: return emit_f32tof16(ctx, alu, src[0]);