KonstantinSeurer
/
mesa
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

nir/lower_double_ops: lower mod() 

There are rounding errors with the division in i965 that affect
the mod(x,y) result when x = N * y. Instead of returning '0' it
was returning 'y'.

This lowering pass fixes those cases.

Signed-off-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
This commit is contained in:
Samuel Iglesias Gonsálvez 2016-04-12 10:55:44 +02:00
parent 9f81434c5f
commit b902377a56
2 changed files with 31 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
src/compiler/nir/nir.h
View File

@ -2425,7 +2425,8 @@ typedef enum {
nir_lower_dfloor = (1 << 4),
nir_lower_dceil = (1 << 5),
nir_lower_dfract = (1 << 6),
nir_lower_dround_even = (1 << 7)
nir_lower_dround_even = (1 << 7),
nir_lower_dmod = (1 << 8)
} nir_lower_doubles_options;
void nir_lower_doubles(nir_shader *shader, nir_lower_doubles_options options);

29
src/compiler/nir/nir_lower_double_ops.c
View File

@ -438,6 +438,24 @@ lower_round_even(nir_builder *b, nir_ssa_def *src)
nir_fsub(b, src, nir_imm_double(b, 0.5)))));
}
static nir_ssa_def *
lower_mod(nir_builder *b, nir_ssa_def *src0, nir_ssa_def *src1)
{
/* mod(x,y) = x - y * floor(x/y)
*
* If the division is lowered, it could add some rounding errors that make
* floor() to return the quotient minus one when x = N * y. If this is the
* case, we return zero because mod(x, y) output value is [0, y).
*/
nir_ssa_def *floor = nir_ffloor(b, nir_fdiv(b, src0, src1));
nir_ssa_def *mod = nir_fsub(b, src0, nir_fmul(b, src1, floor));
return nir_bcsel(b,
nir_fne(b, mod, src1),
mod,
nir_imm_double(b, 0.0));
}
static void
lower_doubles_instr(nir_alu_instr *instr, nir_lower_doubles_options options)
{
@ -486,6 +504,11 @@ lower_doubles_instr(nir_alu_instr *instr, nir_lower_doubles_options options)
return;
break;
case nir_op_fmod:
if (!(options & nir_lower_dmod))
return;
break;
default:
return;
}
@ -525,6 +548,12 @@ lower_doubles_instr(nir_alu_instr *instr, nir_lower_doubles_options options)
result = lower_round_even(&bld, src);
break;
case nir_op_fmod: {
nir_ssa_def *src1 = nir_fmov_alu(&bld, instr->src[1],
instr->dest.dest.ssa.num_components);
result = lower_mod(&bld, src, src1);
}
break;
default:
unreachable("unhandled opcode");
}