fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-03-17 03:50:35 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 1
mesa/src/compiler/nir/nir_range_analysis.c
Georg Lehmann a9e75d8ee4 nir: remove nir_analyze_fp_range 
Use fp class analysis instead.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39987>
2026-03-07 05:01:44 +00:00

2534 lines
80 KiB
C
Raw Blame History

/*
* Copyright © 2018 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#include "nir.h"
#include "nir_range_analysis.h"
#include <float.h>
#include <math.h>
#include "util/hash_table.h"
#include "util/u_dynarray.h"
#include "util/u_math.h"
#include "c99_alloca.h"
/**
* Analyzes a sequence of operations to determine some aspects of the range of
* the result.
*/
struct analysis_query {
uint32_t pushed_queries;
uint32_t result_index;
};
struct analysis_state {
nir_shader *shader;
void *range_ht;
struct util_dynarray query_stack;
struct util_dynarray result_stack;
size_t query_size;
uint32_t (*get_key)(struct analysis_query *q);
bool (*lookup)(void *table, uint32_t key, uint32_t *value);
void (*insert)(void *table, uint32_t key, uint32_t value);
void (*process_query)(struct analysis_state *state, struct analysis_query *q,
uint32_t *result, const uint32_t *src);
};
static void *
push_analysis_query(struct analysis_state *state, size_t size)
{
struct analysis_query *q = util_dynarray_grow_bytes(&state->query_stack, 1, size);
q->pushed_queries = 0;
q->result_index = util_dynarray_num_elements(&state->result_stack, uint32_t);
util_dynarray_append_typed(&state->result_stack, uint32_t, 0);
return q;
}
/* Helper for performing range analysis without recursion. */
static uint32_t
perform_analysis(struct analysis_state *state)
{
while (state->query_stack.size) {
struct analysis_query *cur =
(struct analysis_query *)((char *)util_dynarray_end(&state->query_stack) - state->query_size);
uint32_t *result = util_dynarray_element(&state->result_stack, uint32_t, cur->result_index);
uint32_t key = state->get_key(cur);
/* There might be a cycle-resolving entry for loop header phis. Ignore this when finishing
* them by testing pushed_queries.
*/
if (cur->pushed_queries == 0 && key != UINT32_MAX &&
state->lookup(state->range_ht, key, result)) {
state->query_stack.size -= state->query_size;
continue;
}
uint32_t *src = (uint32_t *)util_dynarray_end(&state->result_stack) - cur->pushed_queries;
state->result_stack.size -= sizeof(uint32_t) * cur->pushed_queries;
uint32_t prev_num_queries = state->query_stack.size;
state->process_query(state, cur, result, src);
uint32_t num_queries = state->query_stack.size;
if (num_queries > prev_num_queries) {
cur = (struct analysis_query *)util_dynarray_element(&state->query_stack, char,
prev_num_queries - state->query_size);
cur->pushed_queries = (num_queries - prev_num_queries) / state->query_size;
continue;
}
if (key != UINT32_MAX)
state->insert(state->range_ht, key, *result);
state->query_stack.size -= state->query_size;
}
assert(state->result_stack.size == sizeof(uint32_t));
uint32_t res = util_dynarray_top(&state->result_stack, uint32_t);
util_dynarray_fini(&state->query_stack);
util_dynarray_fini(&state->result_stack);
return res;
}
static fp_class_mask
analyze_fp_constant(const nir_load_const_instr *const load)
{
fp_class_mask result = 0;
for (unsigned i = 0; i < load->def.num_components; ++i) {
const double v = nir_const_value_as_float(load->value[i],
load->def.bit_size);
if (!isnan(v) && floor(v) != v)
result |= FP_CLASS_NON_INTEGRAL;
if (isnan(v))
result |= FP_CLASS_NAN;
else if (v == -INFINITY)
result |= FP_CLASS_NEG_INF;
else if (v < -1.0)
result |= FP_CLASS_LT_NEG_ONE;
else if (v == -1.0)
result |= FP_CLASS_NEG_ONE;
else if (v < 0.0)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE;
else if (dui(v) == 0)
result |= FP_CLASS_POS_ZERO;
else if (v == 0.0)
result |= FP_CLASS_NEG_ZERO;
else if (v < 1.0)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE;
else if (v == 1.0)
result |= FP_CLASS_POS_ONE;
else if (v < INFINITY)
result |= FP_CLASS_GT_POS_ONE;
else
result |= FP_CLASS_POS_INF;
if (v != 0) {
/* handle potential denorm flushing. */
bool is_denorm = false;
switch (load->def.bit_size) {
case 64:
is_denorm = fabs(v) < DBL_MIN;
break;
case 32:
is_denorm = fabs(v) < FLT_MIN;
break;
case 16:
is_denorm = fabs(v) < ldexp(1.0, -14);
break;
default:
UNREACHABLE("unsupported float size");
}
if (is_denorm)
result |= v < 0.0 ? FP_CLASS_NEG_ZERO : FP_CLASS_POS_ZERO;
}
}
return result;
}
struct fp_query {
struct analysis_query head;
const nir_def *def;
};
static void
push_fp_query(struct analysis_state *state, const nir_def *def)
{
struct fp_query *pushed_q = push_analysis_query(state, sizeof(struct fp_query));
pushed_q->def = def;
}
static uint32_t
get_fp_key(struct analysis_query *q)
{
struct fp_query *fp_q = (struct fp_query *)q;
if (!nir_def_is_alu(fp_q->def))
return UINT32_MAX;
return fp_q->def->index;
}
static bool
fp_lookup(void *table, uint32_t key, uint32_t *value)
{
nir_fp_analysis_state *state = table;
if (BITSET_TEST(state->bitset, key)) {
*value = *(uint32_t *)util_sparse_array_get(&state->arr, key);
return true;
} else {
return false;
}
}
static void
fp_insert(void *table, uint32_t key, uint32_t value)
{
nir_fp_analysis_state *state = table;
BITSET_SET(state->bitset, key);
state->max = MAX2(state->max, (int)key);
*(uint32_t *)util_sparse_array_get(&state->arr, key) = value;
}
static fp_class_mask
fneg_fp_class(fp_class_mask src)
{
fp_class_mask result = src & (FP_CLASS_NAN | FP_CLASS_NON_INTEGRAL);
#define NEG_BIT(neg, pos) \
if (src & FP_CLASS_##pos) \
result |= FP_CLASS_##neg; \
if (src & FP_CLASS_##neg) \
result |= FP_CLASS_##pos;
NEG_BIT(NEG_INF, POS_INF);
NEG_BIT(LT_NEG_ONE, GT_POS_ONE);
NEG_BIT(NEG_ONE, POS_ONE);
NEG_BIT(LT_ZERO_GT_NEG_ONE, GT_ZERO_LT_POS_ONE);
NEG_BIT(NEG_ZERO, POS_ZERO);
#undef NEG_BIT
return result;
}
static fp_class_mask
fmul_fp_class(fp_class_mask left, fp_class_mask right, bool mulz, bool src_eq, bool src_neg_eq)
{
/* For runtime performance, shortcut the common completely unknown case. */
if (left == FP_CLASS_UNKNOWN && right == FP_CLASS_UNKNOWN && !src_eq && !src_neg_eq)
return FP_CLASS_UNKNOWN;
fp_class_mask result = 0;
if (left & FP_CLASS_NAN) {
if (right & FP_CLASS_ANY_ZERO)
result |= mulz ? FP_CLASS_POS_ZERO : FP_CLASS_NAN;
if (right & (FP_CLASS_ANY_NEG | FP_CLASS_ANY_POS | FP_CLASS_NAN))
result |= FP_CLASS_NAN;
}
if (left & FP_CLASS_NEG_INF) {
if (right & FP_CLASS_ANY_ZERO)
result |= mulz ? FP_CLASS_POS_ZERO : FP_CLASS_NAN;
if (right & FP_CLASS_ANY_NEG)
result |= FP_CLASS_POS_INF;
if (right & FP_CLASS_ANY_POS)
result |= FP_CLASS_NEG_INF;
if (right & FP_CLASS_NAN)
result |= FP_CLASS_NAN;
}
if (left & FP_CLASS_POS_INF) {
if (right & FP_CLASS_ANY_ZERO)
result |= mulz ? FP_CLASS_POS_ZERO : FP_CLASS_NAN;
if (right & FP_CLASS_ANY_POS)
result |= FP_CLASS_POS_INF;
if (right & FP_CLASS_ANY_NEG)
result |= FP_CLASS_NEG_INF;
if (right & FP_CLASS_NAN)
result |= FP_CLASS_NAN;
}
if (left & FP_CLASS_ANY_NEG_FINITE) {
if (right & FP_CLASS_POS_ZERO)
result |= mulz ? FP_CLASS_POS_ZERO : FP_CLASS_NEG_ZERO;
result |= fneg_fp_class(right & (FP_CLASS_NEG_ZERO | FP_CLASS_POS_INF | FP_CLASS_NEG_INF | FP_CLASS_NAN));
}
if (left & FP_CLASS_ANY_POS_FINITE) {
if (right & FP_CLASS_NEG_ZERO)
result |= mulz ? FP_CLASS_POS_ZERO : FP_CLASS_NEG_ZERO;
result |= right & (FP_CLASS_POS_ZERO | FP_CLASS_POS_INF | FP_CLASS_NEG_INF | FP_CLASS_NAN);
}
if (left & FP_CLASS_LT_NEG_ONE) {
if (right & FP_CLASS_LT_NEG_ONE)
result |= FP_CLASS_POS_INF | FP_CLASS_GT_POS_ONE;
if (right & FP_CLASS_NEG_ONE)
result |= FP_CLASS_GT_POS_ONE;
if (right & FP_CLASS_LT_ZERO_GT_NEG_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_POS_ONE | FP_CLASS_GT_POS_ONE;
if (right & FP_CLASS_GT_POS_ONE)
result |= FP_CLASS_NEG_INF | FP_CLASS_LT_NEG_ONE;
if (right & FP_CLASS_POS_ONE)
result |= FP_CLASS_LT_NEG_ONE;
if (right & FP_CLASS_GT_ZERO_LT_POS_ONE)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_NEG_ONE | FP_CLASS_LT_NEG_ONE;
}
if (left & FP_CLASS_GT_POS_ONE) {
if (right & FP_CLASS_GT_POS_ONE)
result |= FP_CLASS_POS_INF | FP_CLASS_GT_POS_ONE;
if (right & FP_CLASS_POS_ONE)
result |= FP_CLASS_GT_POS_ONE;
if (right & FP_CLASS_GT_ZERO_LT_POS_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_POS_ONE | FP_CLASS_GT_POS_ONE;
if (right & FP_CLASS_LT_NEG_ONE)
result |= FP_CLASS_NEG_INF | FP_CLASS_LT_NEG_ONE;
if (right & FP_CLASS_NEG_ONE)
result |= FP_CLASS_LT_NEG_ONE;
if (right & FP_CLASS_LT_ZERO_GT_NEG_ONE)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_NEG_ONE | FP_CLASS_LT_NEG_ONE;
}
if (left & FP_CLASS_NEG_ONE)
result |= fneg_fp_class(right & ~FP_CLASS_ANY_ZERO);
if (left & FP_CLASS_POS_ONE)
result |= right & ~FP_CLASS_ANY_ZERO;
if (left & FP_CLASS_LT_ZERO_GT_NEG_ONE) {
if (right & FP_CLASS_LT_NEG_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_POS_ONE | FP_CLASS_GT_POS_ONE;
if (right & FP_CLASS_NEG_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE;
if (right & FP_CLASS_LT_ZERO_GT_NEG_ONE)
result |= FP_CLASS_POS_ZERO | FP_CLASS_GT_ZERO_LT_POS_ONE;
if (right & FP_CLASS_GT_POS_ONE)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_NEG_ONE | FP_CLASS_LT_NEG_ONE;
if (right & FP_CLASS_POS_ONE)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE;
if (right & FP_CLASS_GT_ZERO_LT_POS_ONE)
result |= FP_CLASS_NEG_ZERO | FP_CLASS_LT_ZERO_GT_NEG_ONE;
}
if (left & FP_CLASS_GT_ZERO_LT_POS_ONE) {
if (right & FP_CLASS_GT_POS_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_POS_ONE | FP_CLASS_GT_POS_ONE;
if (right & FP_CLASS_POS_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE;
if (right & FP_CLASS_GT_ZERO_LT_POS_ONE)
result |= FP_CLASS_POS_ZERO | FP_CLASS_GT_ZERO_LT_POS_ONE;
if (right & FP_CLASS_LT_NEG_ONE)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_NEG_ONE | FP_CLASS_LT_NEG_ONE;
if (right & FP_CLASS_NEG_ONE)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE;
if (right & FP_CLASS_LT_ZERO_GT_NEG_ONE)
result |= FP_CLASS_NEG_ZERO | FP_CLASS_LT_ZERO_GT_NEG_ONE;
}
if (left & FP_CLASS_NEG_ZERO) {
if (mulz) {
result |= FP_CLASS_POS_ZERO;
} else {
if (right & (FP_CLASS_ANY_INF | FP_CLASS_NAN))
result |= FP_CLASS_NAN;
if (right & (FP_CLASS_ANY_NEG_FINITE | FP_CLASS_NEG_ZERO))
result |= FP_CLASS_POS_ZERO;
if (right & (FP_CLASS_ANY_POS_FINITE | FP_CLASS_POS_ZERO))
result |= FP_CLASS_NEG_ZERO;
}
}
if (left & FP_CLASS_POS_ZERO) {
if (mulz) {
result |= FP_CLASS_POS_ZERO;
} else {
if (right & (FP_CLASS_ANY_INF | FP_CLASS_NAN))
result |= FP_CLASS_NAN;
if (right & (FP_CLASS_ANY_POS_FINITE | FP_CLASS_POS_ZERO))
result |= FP_CLASS_POS_ZERO;
if (right & (FP_CLASS_ANY_NEG_FINITE | FP_CLASS_NEG_ZERO))
result |= FP_CLASS_NEG_ZERO;
}
}
if (src_eq || src_neg_eq) {
/* This case can't create new ones. */
if (!(left & (FP_CLASS_POS_ONE | FP_CLASS_NEG_ONE)))
result &= ~(FP_CLASS_POS_ONE | FP_CLASS_NEG_ONE);
if (src_eq)
result &= ~(FP_CLASS_ANY_NEG | FP_CLASS_NEG_ZERO);
else if (src_neg_eq && mulz)
result &= ~FP_CLASS_ANY_POS;
else if (src_neg_eq)
result &= ~(FP_CLASS_ANY_POS | FP_CLASS_POS_ZERO);
}
if ((left | right) & FP_CLASS_NON_INTEGRAL) {
if (result & (FP_CLASS_LT_NEG_ONE | FP_CLASS_LT_ZERO_GT_NEG_ONE |
FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_GT_POS_ONE))
result |= FP_CLASS_NON_INTEGRAL;
} else {
result &= ~(FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_GT_ZERO_LT_POS_ONE);
}
return result;
}
static fp_class_mask
fadd_fp_class(fp_class_mask left, fp_class_mask right)
{
/* For runtime performance, shortcut the common completely unknown case. */
if (left == FP_CLASS_UNKNOWN && right == FP_CLASS_UNKNOWN)
return FP_CLASS_UNKNOWN;
fp_class_mask result = (left | right) & FP_CLASS_NAN;
/* X + Y is NaN if either operand is NaN or if one operand is +Inf and
* the other is -Inf.
*/
if (left & FP_CLASS_NEG_INF) {
if (right & FP_CLASS_POS_INF)
result |= FP_CLASS_NAN;
if (right & (FP_CLASS_ANY_FINITE | FP_CLASS_NEG_INF))
result |= FP_CLASS_NEG_INF;
}
if (left & FP_CLASS_POS_INF) {
if (right & FP_CLASS_NEG_INF)
result |= FP_CLASS_NAN;
if (right & (FP_CLASS_ANY_FINITE | FP_CLASS_POS_INF))
result |= FP_CLASS_POS_INF;
}
if (left & FP_CLASS_LT_NEG_ONE) {
result |= (right & FP_CLASS_ANY_INF);
if (right & FP_CLASS_LT_NEG_ONE)
result |= FP_CLASS_NEG_INF | FP_CLASS_LT_NEG_ONE;
if (right & (FP_CLASS_NEG_ONE | FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_ANY_ZERO))
result |= FP_CLASS_LT_NEG_ONE;
if (right & (FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_POS_ONE))
result |= FP_CLASS_LT_NEG_ONE | FP_CLASS_NEG_ONE | FP_CLASS_LT_ZERO_GT_NEG_ONE;
if (right & FP_CLASS_GT_POS_ONE)
result |= FP_CLASS_ANY_FINITE;
}
if (left & FP_CLASS_GT_POS_ONE) {
result |= (right & FP_CLASS_ANY_INF);
if (right & FP_CLASS_GT_POS_ONE)
result |= FP_CLASS_POS_INF | FP_CLASS_GT_POS_ONE;
if (right & (FP_CLASS_POS_ONE | FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_ANY_ZERO))
result |= FP_CLASS_GT_POS_ONE;
if (right & (FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_NEG_ONE))
result |= FP_CLASS_GT_POS_ONE | FP_CLASS_POS_ONE | FP_CLASS_GT_ZERO_LT_POS_ONE;
if (right & FP_CLASS_LT_NEG_ONE)
result |= FP_CLASS_ANY_FINITE;
}
if (left & FP_CLASS_NEG_ONE) {
result |= (right & FP_CLASS_ANY_INF);
if (right & (FP_CLASS_LT_NEG_ONE | FP_CLASS_NEG_ONE | FP_CLASS_LT_ZERO_GT_NEG_ONE))
result |= FP_CLASS_LT_NEG_ONE;
if (right & (FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_ANY_ZERO | FP_CLASS_GT_ZERO_LT_POS_ONE))
result |= FP_CLASS_NEG_ONE;
if (right & FP_CLASS_GT_ZERO_LT_POS_ONE)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE;
if (right & FP_CLASS_POS_ONE)
result |= FP_CLASS_POS_ZERO;
if (right & FP_CLASS_GT_POS_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_POS_ONE | FP_CLASS_GT_POS_ONE;
}
if (left & FP_CLASS_POS_ONE) {
result |= (right & FP_CLASS_ANY_INF);
if (right & (FP_CLASS_GT_POS_ONE | FP_CLASS_POS_ONE | FP_CLASS_GT_ZERO_LT_POS_ONE))
result |= FP_CLASS_GT_POS_ONE;
if (right & (FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_ANY_ZERO | FP_CLASS_LT_ZERO_GT_NEG_ONE))
result |= FP_CLASS_POS_ONE;
if (right & FP_CLASS_LT_ZERO_GT_NEG_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE;
if (right & FP_CLASS_NEG_ONE)
result |= FP_CLASS_POS_ZERO;
if (right & FP_CLASS_LT_NEG_ONE)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_NEG_ONE | FP_CLASS_LT_NEG_ONE;
}
if (left & FP_CLASS_LT_ZERO_GT_NEG_ONE) {
result |= (right & FP_CLASS_ANY_INF);
if (right & FP_CLASS_LT_NEG_ONE)
result |= FP_CLASS_LT_NEG_ONE;
if (right & FP_CLASS_NEG_ONE)
result |= FP_CLASS_LT_NEG_ONE | FP_CLASS_NEG_ONE;
if (right & FP_CLASS_LT_ZERO_GT_NEG_ONE)
result |= FP_CLASS_LT_NEG_ONE | FP_CLASS_NEG_ONE | FP_CLASS_LT_ZERO_GT_NEG_ONE;
if (right & FP_CLASS_ANY_ZERO)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE;
if (right & FP_CLASS_GT_ZERO_LT_POS_ONE)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_ANY_ZERO | FP_CLASS_GT_ZERO_LT_POS_ONE;
if (right & FP_CLASS_POS_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_POS_ONE;
if (right & FP_CLASS_GT_POS_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_POS_ONE | FP_CLASS_GT_POS_ONE;
}
if (left & FP_CLASS_GT_ZERO_LT_POS_ONE) {
result |= (right & FP_CLASS_ANY_INF);
if (right & FP_CLASS_GT_POS_ONE)
result |= FP_CLASS_GT_POS_ONE;
if (right & FP_CLASS_POS_ONE)
result |= FP_CLASS_GT_POS_ONE | FP_CLASS_POS_ONE;
if (right & FP_CLASS_GT_ZERO_LT_POS_ONE)
result |= FP_CLASS_GT_POS_ONE | FP_CLASS_POS_ONE | FP_CLASS_GT_ZERO_LT_POS_ONE;
if (right & FP_CLASS_ANY_ZERO)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE;
if (right & FP_CLASS_LT_ZERO_GT_NEG_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_ANY_ZERO | FP_CLASS_LT_ZERO_GT_NEG_ONE;
if (right & FP_CLASS_NEG_ONE)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_NEG_ONE;
if (right & FP_CLASS_LT_NEG_ONE)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_NEG_ONE | FP_CLASS_LT_NEG_ONE;
}
if (left & FP_CLASS_NEG_ZERO) {
result |= right;
}
if (left & FP_CLASS_POS_ZERO) {
result |= right & ~FP_CLASS_NEG_ZERO;
if (right & FP_CLASS_NEG_ZERO)
result |= FP_CLASS_POS_ZERO;
}
if ((left | right) & FP_CLASS_NON_INTEGRAL) {
if (result & (FP_CLASS_LT_NEG_ONE | FP_CLASS_LT_ZERO_GT_NEG_ONE |
FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_GT_POS_ONE))
result |= FP_CLASS_NON_INTEGRAL;
} else {
result &= ~(FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_GT_ZERO_LT_POS_ONE);
}
return result;
}
static fp_class_mask
frcp_fp_class(fp_class_mask src)
{
fp_class_mask result = src & FP_CLASS_NAN;
/* Inf/Zero result in Zero/Inf.*/
if (src & FP_CLASS_NEG_INF)
result |= FP_CLASS_NEG_ZERO;
if (src & FP_CLASS_POS_INF)
result |= FP_CLASS_POS_ZERO;
if (src & FP_CLASS_NEG_ZERO)
result |= FP_CLASS_NEG_INF;
if (src & FP_CLASS_POS_ZERO)
result |= FP_CLASS_POS_INF;
/* One results in one. */
if (src & FP_CLASS_NEG_ONE)
result |= FP_CLASS_NEG_ONE;
if (src & FP_CLASS_POS_ONE)
result |= FP_CLASS_POS_ONE;
if (src & FP_CLASS_LT_NEG_ONE)
result |= FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_NEG_ZERO | FP_CLASS_NON_INTEGRAL;
if (src & FP_CLASS_GT_POS_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_POS_ZERO | FP_CLASS_NON_INTEGRAL;
if (src & FP_CLASS_LT_ZERO_GT_NEG_ONE)
result |= FP_CLASS_LT_NEG_ONE | FP_CLASS_NEG_INF | FP_CLASS_NON_INTEGRAL;
if (src & FP_CLASS_GT_ZERO_LT_POS_ONE)
result |= FP_CLASS_GT_POS_ONE | FP_CLASS_POS_INF | FP_CLASS_NON_INTEGRAL;
return result;
}
static fp_class_mask
fsqrt_fp_class(fp_class_mask src)
{
fp_class_mask result = src & (FP_CLASS_NAN | FP_CLASS_ANY_ZERO);
if (src & FP_CLASS_ANY_NEG)
result |= FP_CLASS_NAN;
if (src & FP_CLASS_GT_ZERO_LT_POS_ONE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_NON_INTEGRAL;
if (src & FP_CLASS_POS_ONE)
result |= FP_CLASS_POS_ONE;
if (src & FP_CLASS_GT_POS_ONE)
result |= FP_CLASS_GT_POS_ONE | FP_CLASS_POS_ONE | FP_CLASS_NON_INTEGRAL;
if (src & FP_CLASS_POS_INF)
result |= FP_CLASS_POS_INF;
return result;
}
static fp_class_mask
fmin_part(fp_class_mask upper_bound, fp_class_mask value)
{
/* Find the highest value in upper_bound, and return all
* smaller or equal values from value.
*/
upper_bound &= FP_CLASS_ANY_NEG | FP_CLASS_ANY_ZERO | FP_CLASS_ANY_POS;
value &= FP_CLASS_ANY_NEG | FP_CLASS_ANY_ZERO | FP_CLASS_ANY_POS;
/* This works even in the case where upper_bound is 0 */
return value & BITFIELD_MASK(util_last_bit(upper_bound));
}
static fp_class_mask
fmin_fp_class(fp_class_mask left, fp_class_mask right)
{
fp_class_mask result = 0;
/* If one source is NaN, we have to include the whole range of the other source. */
if (left & FP_CLASS_NAN)
result |= right;
if (right & FP_CLASS_NAN)
result |= left;
result |= fmin_part(left, right);
result |= fmin_part(right, left);
/* Could probably do better, but meh. */
if ((left | right) & FP_CLASS_NON_INTEGRAL) {
if (result & (FP_CLASS_LT_NEG_ONE | FP_CLASS_LT_ZERO_GT_NEG_ONE |
FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_GT_POS_ONE))
result |= FP_CLASS_NON_INTEGRAL;
}
return result;
}
static fp_class_mask
handle_sz(const nir_alu_instr *alu, fp_class_mask src)
{
if (nir_alu_instr_is_signed_zero_preserve(alu) || !(src & FP_CLASS_ANY_ZERO))
return src;
return src | FP_CLASS_ANY_ZERO;
}
static fp_class_mask
intrinsic_fp_class(const nir_intrinsic_instr *intrin)
{
switch (intrin->intrinsic) {
case nir_intrinsic_load_typed_buffer_amd: {
const enum pipe_format format = nir_intrinsic_format(intrin);
if (format == PIPE_FORMAT_NONE)
return FP_CLASS_UNKNOWN;
const struct util_format_description *desc = util_format_description(format);
int i = util_format_get_first_non_void_channel(format);
if (i == -1)
return FP_CLASS_UNKNOWN;
bool is_signed = desc->channel[i].type == UTIL_FORMAT_TYPE_SIGNED;
bool is_unsigned = desc->channel[i].type == UTIL_FORMAT_TYPE_UNSIGNED;
bool normalized = desc->channel[i].normalized;
if ((!is_signed && !is_unsigned) || desc->channel[i].pure_integer)
return FP_CLASS_UNKNOWN;
fp_class_mask result = FP_CLASS_POS_ZERO | FP_CLASS_POS_ONE;
result |= is_signed ? FP_CLASS_NEG_ONE : 0;
if (normalized) {
result |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_NON_INTEGRAL;
result |= is_signed ? FP_CLASS_LT_ZERO_GT_NEG_ONE : 0;
} else {
result |= FP_CLASS_GT_POS_ONE;
result |= is_signed ? FP_CLASS_LT_NEG_ONE : 0;
}
return result;
}
case nir_intrinsic_load_front_face_fsign:
return FP_CLASS_POS_ONE | FP_CLASS_NEG_ONE;
default:
return FP_CLASS_UNKNOWN;
}
}
static fp_class_mask
tex_fp_class(nir_tex_instr *tex)
{
/* Not much to analyze, except shadow compare. */
if (!tex->is_shadow)
return FP_CLASS_UNKNOWN;
fp_class_mask result = FP_CLASS_POS_ZERO | FP_CLASS_POS_ONE;
/* Gather returns 0 or 1, other ops can interpolate.
* Cube corners are special even for gathers.
*/
if (tex->op != nir_texop_tg4 || tex->sampler_dim == GLSL_SAMPLER_DIM_CUBE)
result |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_NON_INTEGRAL;
return result;
}
/**
* Analyze an expression to determine the possible fp classes of its result
*/
static void
process_fp_query(struct analysis_state *state, struct analysis_query *aq, uint32_t *result,
const uint32_t *src_res)
{
struct fp_query q = *(struct fp_query *)aq;
const nir_def *def = q.def;
if (nir_def_is_const(def)) {
*result = analyze_fp_constant(nir_def_as_load_const(def));
return;
} else if (nir_def_is_intrinsic(def)) {
*result = intrinsic_fp_class(nir_def_as_intrinsic(def));
return;
} else if (nir_def_is_tex(def)) {
*result = tex_fp_class(nir_def_as_tex(def));
return;
} else if (!nir_def_is_alu(def)) {
*result = FP_CLASS_UNKNOWN;
return;
}
const nir_alu_instr *const alu = nir_def_as_alu(def);
if (!aq->pushed_queries) {
switch (alu->op) {
case nir_op_bcsel:
push_fp_query(state, alu->src[1].src.ssa);
push_fp_query(state, alu->src[2].src.ssa);
return;
case nir_op_mov:
case nir_op_fabs:
case nir_op_fexp2:
case nir_op_flog2:
case nir_op_frcp:
case nir_op_fsqrt:
case nir_op_frsq:
case nir_op_fneg:
case nir_op_fsat:
case nir_op_fsign:
case nir_op_ffloor:
case nir_op_fceil:
case nir_op_ftrunc:
case nir_op_fround_even:
case nir_op_ffract:
case nir_op_fsin:
case nir_op_fcos:
case nir_op_fsin_amd:
case nir_op_fcos_amd:
case nir_op_f2f16:
case nir_op_f2f16_rtz:
case nir_op_f2f16_rtne:
case nir_op_f2f32:
case nir_op_f2f64:
case nir_op_fdot2:
case nir_op_fdot3:
case nir_op_fdot4:
case nir_op_fdot8:
case nir_op_fdot16:
case nir_op_fdot2_replicated:
case nir_op_fdot3_replicated:
case nir_op_fdot4_replicated:
case nir_op_fdot8_replicated:
case nir_op_fdot16_replicated:
push_fp_query(state, alu->src[0].src.ssa);
return;
case nir_op_fadd:
case nir_op_fsub:
case nir_op_fmax:
case nir_op_fmin:
case nir_op_fmul:
case nir_op_fmulz:
case nir_op_fpow:
case nir_op_vec2:
push_fp_query(state, alu->src[0].src.ssa);
push_fp_query(state, alu->src[1].src.ssa);
return;
case nir_op_ffma:
case nir_op_ffmaz:
case nir_op_flrp:
push_fp_query(state, alu->src[0].src.ssa);
push_fp_query(state, alu->src[1].src.ssa);
push_fp_query(state, alu->src[2].src.ssa);
return;
default:
break;
}
}
fp_class_mask r = FP_CLASS_UNKNOWN;
switch (alu->op) {
case nir_op_b2i16:
case nir_op_b2i32:
case nir_op_b2i64:
/* b2i32 will generate either 0x00000000 or 0x00000001. When those bit
* patterns are interpreted as floating point, they are 0.0 and
* 1.401298464324817e-45. The latter is subnormal.
*/
r = FP_CLASS_POS_ZERO | FP_CLASS_GT_ZERO_LT_POS_ONE;
break;
case nir_op_b2f16:
case nir_op_b2f32:
case nir_op_b2f64:
r = FP_CLASS_POS_ZERO | FP_CLASS_POS_ONE;
break;
case nir_op_vec2:
case nir_op_bcsel:
r = src_res[0] | src_res[1];
break;
case nir_op_i2f16:
case nir_op_i2f32:
case nir_op_i2f64:
r &= ~FP_CLASS_NAN;
r &= ~FP_CLASS_NON_INTEGRAL;
r &= ~FP_CLASS_GT_ZERO_LT_POS_ONE;
r &= ~FP_CLASS_LT_ZERO_GT_NEG_ONE;
r &= ~FP_CLASS_NEG_ZERO;
if (alu->def.bit_size > 16 || alu->src[0].src.ssa->bit_size <= 16)
r &= ~FP_CLASS_ANY_INF;
break;
case nir_op_u2f16:
case nir_op_u2f32:
case nir_op_u2f64:
r &= ~FP_CLASS_NAN;
r &= ~FP_CLASS_NON_INTEGRAL;
r &= ~FP_CLASS_GT_ZERO_LT_POS_ONE;
r &= ~FP_CLASS_NEG_ZERO;
r &= ~FP_CLASS_ANY_NEG;
if (alu->def.bit_size > 16 || alu->src[0].src.ssa->bit_size < 16)
r &= ~FP_CLASS_ANY_INF;
break;
case nir_op_f2f16:
case nir_op_f2f16_rtz:
case nir_op_f2f16_rtne:
case nir_op_f2f32:
case nir_op_f2f64: {
r = handle_sz(alu, src_res[0]);
if (alu->src[0].src.ssa->bit_size > alu->def.bit_size) {
bool rtz = alu->op == nir_op_f2f16_rtz;
if (alu->op != nir_op_f2f16_rtne && alu->op != nir_op_f2f16_rtz) {
nir_shader *shader = nir_cf_node_get_function(&alu->instr.block->cf_node)->function->shader;
unsigned execution_mode = shader->info.float_controls_execution_mode;
rtz = nir_is_rounding_mode_rtz(execution_mode, alu->def.bit_size);
}
/* Unless we are rounding towards zero, large values can create Inf. */
if (r & FP_CLASS_LT_NEG_ONE) {
if (!rtz)
r |= FP_CLASS_NEG_INF;
r |= FP_CLASS_NEG_ONE;
}
if (r & FP_CLASS_GT_POS_ONE) {
if (!rtz)
r |= FP_CLASS_POS_INF;
r |= FP_CLASS_POS_ONE;
}
/* Underflow can create new zeros. */
if (r & FP_CLASS_LT_ZERO_GT_NEG_ONE) {
if (!rtz)
r |= FP_CLASS_NEG_ONE;
r |= FP_CLASS_NEG_ZERO;
}
if (r & FP_CLASS_GT_ZERO_LT_POS_ONE) {
if (!rtz)
r |= FP_CLASS_POS_ONE;
r |= FP_CLASS_POS_ZERO;
}
}
break;
}
case nir_op_fneg:
r = fneg_fp_class(src_res[0]);
break;
case nir_op_fabs:
r = src_res[0];
r |= fneg_fp_class(r & (FP_CLASS_ANY_NEG | FP_CLASS_NEG_ZERO));
r &= ~(FP_CLASS_ANY_NEG | FP_CLASS_NEG_ZERO);
break;
case nir_op_fadd: {
r = fadd_fp_class(src_res[0], src_res[1]);
break;
}
case nir_op_fsub: {
r = fadd_fp_class(src_res[0], fneg_fp_class(src_res[1]));
break;
}
case nir_op_fexp2: {
fp_class_mask src = src_res[0];
r = 0;
/* If the parameter might be less than zero, the mathematically result
* will be on (0, 1). For sufficiently large magnitude negative
* parameters, the result will flush to zero.
*/
if (src & FP_CLASS_NEG_INF)
r |= FP_CLASS_POS_ZERO;
if (src & FP_CLASS_LT_NEG_ONE)
r |= FP_CLASS_POS_ZERO | FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_NON_INTEGRAL;
if (src & (FP_CLASS_NEG_ONE | FP_CLASS_LT_ZERO_GT_NEG_ONE))
r |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_NON_INTEGRAL;
if (src & (FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_ANY_ZERO | FP_CLASS_GT_ZERO_LT_POS_ONE))
r |= FP_CLASS_POS_ONE;
if (src & (FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_POS_ONE))
r |= FP_CLASS_GT_POS_ONE;
if (src & (FP_CLASS_GT_POS_ONE))
r |= FP_CLASS_GT_POS_ONE | FP_CLASS_POS_INF;
if (src & FP_CLASS_POS_INF)
r |= FP_CLASS_POS_INF;
if (src & FP_CLASS_NON_INTEGRAL)
r |= FP_CLASS_NON_INTEGRAL;
if (src & FP_CLASS_NAN)
r |= FP_CLASS_NAN;
break;
}
case nir_op_flog2: {
r = 0;
if (src_res[0] & (FP_CLASS_ANY_NEG | FP_CLASS_NAN))
r |= FP_CLASS_NAN;
if (src_res[0] & FP_CLASS_ANY_ZERO)
r |= FP_CLASS_NEG_INF;
if (src_res[0] & FP_CLASS_GT_ZERO_LT_POS_ONE)
r |= FP_CLASS_ANY_NEG | FP_CLASS_NON_INTEGRAL;
if (src_res[0] & FP_CLASS_POS_ONE)
r |= FP_CLASS_POS_ZERO;
if (src_res[0] & FP_CLASS_GT_POS_ONE)
r |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_POS_ONE | FP_CLASS_GT_POS_ONE | FP_CLASS_NON_INTEGRAL;
if (src_res[0] & FP_CLASS_POS_INF)
r |= FP_CLASS_POS_INF;
break;
}
case nir_op_fmax: {
fp_class_mask left = fneg_fp_class(src_res[0]);
fp_class_mask right = fneg_fp_class(src_res[1]);
r = fneg_fp_class(fmin_fp_class(left, right));
break;
}
case nir_op_fmin:
r = fmin_fp_class(src_res[0], src_res[1]);
break;
case nir_op_fmul:
case nir_op_fmulz: {
bool mulz = alu->op == nir_op_fmulz;
bool src_eq = nir_alu_srcs_equal(alu, alu, 0, 1);
bool src_neg_eq = !nir_src_is_const(alu->src[0].src) && nir_alu_srcs_negative_equal(alu, alu, 0, 1);
r = fmul_fp_class(src_res[0], src_res[1], mulz, src_eq, src_neg_eq);
break;
}
case nir_op_frcp:
r = frcp_fp_class(handle_sz(alu, src_res[0]));
break;
case nir_op_mov:
r = src_res[0];
break;
case nir_op_fsat: {
r = src_res[0];
/* max(+0.0, x) */
if (r & (FP_CLASS_ANY_NEG | FP_CLASS_NEG_ZERO | FP_CLASS_NAN)) {
r &= ~(FP_CLASS_ANY_NEG | FP_CLASS_NEG_ZERO | FP_CLASS_NAN);
r |= FP_CLASS_POS_ZERO;
}
/* min(+1.0, x) */
if (r & (FP_CLASS_GT_POS_ONE | FP_CLASS_POS_INF)) {
r &= ~(FP_CLASS_GT_POS_ONE | FP_CLASS_POS_INF);
r |= FP_CLASS_POS_ONE;
}
if (!(r & FP_CLASS_GT_ZERO_LT_POS_ONE))
r &= ~FP_CLASS_NON_INTEGRAL;
break;
}
case nir_op_fsign:
r = 0;
if (src_res[0] & FP_CLASS_ANY_NEG)
r |= FP_CLASS_NEG_ONE;
if (src_res[0] & FP_CLASS_ANY_ZERO)
r |= FP_CLASS_ANY_ZERO;
if (src_res[0] & FP_CLASS_ANY_POS)
r |= FP_CLASS_POS_ONE;
/* fsign is -1, 0, or 1, even for NaN */
if (src_res[0] & FP_CLASS_NAN)
r |= FP_CLASS_NEG_ONE | FP_CLASS_ANY_ZERO | FP_CLASS_POS_ONE;
break;
case nir_op_fsqrt:
r = fsqrt_fp_class(src_res[0]);
break;
case nir_op_frsq:
r = frcp_fp_class(fsqrt_fp_class(handle_sz(alu, src_res[0])));
break;
case nir_op_ffloor: {
/* In IEEE 754, floor(NaN) is NaN, and floor(±Inf) is ±Inf. See
* https://pubs.opengroup.org/onlinepubs/9699919799.2016edition/functions/floor.html
*/
r = src_res[0];
if (r & FP_CLASS_NON_INTEGRAL) {
if (r & FP_CLASS_LT_ZERO_GT_NEG_ONE)
r |= FP_CLASS_NEG_ONE;
if (r & FP_CLASS_GT_ZERO_LT_POS_ONE)
r |= FP_CLASS_POS_ZERO;
if (r & FP_CLASS_GT_POS_ONE)
r |= FP_CLASS_POS_ONE;
r &= ~(FP_CLASS_NON_INTEGRAL | FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_GT_ZERO_LT_POS_ONE);
}
break;
}
case nir_op_fceil: {
/* In IEEE 754, ceil(NaN) is NaN, and ceil(±Inf) is ±Inf. See
* https://pubs.opengroup.org/onlinepubs/9699919799.2016edition/functions/ceil.html
*/
r = src_res[0];
if (r & FP_CLASS_NON_INTEGRAL) {
if (r & FP_CLASS_LT_NEG_ONE)
r |= FP_CLASS_NEG_ONE;
if (r & FP_CLASS_LT_ZERO_GT_NEG_ONE)
r |= FP_CLASS_NEG_ZERO;
if (r & FP_CLASS_GT_ZERO_LT_POS_ONE)
r |= FP_CLASS_POS_ONE;
r &= ~(FP_CLASS_NON_INTEGRAL | FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_GT_ZERO_LT_POS_ONE);
}
break;
}
case nir_op_ftrunc: {
/* In IEEE 754, trunc(NaN) is NaN, and trunc(±Inf) is ±Inf. See
* https://pubs.opengroup.org/onlinepubs/9699919799.2016edition/functions/trunc.html
*/
r = src_res[0];
if (r & FP_CLASS_NON_INTEGRAL) {
if (r & FP_CLASS_LT_NEG_ONE)
r |= FP_CLASS_NEG_ONE;
if (r & FP_CLASS_LT_ZERO_GT_NEG_ONE)
r |= FP_CLASS_NEG_ZERO;
if (r & FP_CLASS_GT_ZERO_LT_POS_ONE)
r |= FP_CLASS_POS_ZERO;
if (r & FP_CLASS_GT_POS_ONE)
r |= FP_CLASS_POS_ONE;
r &= ~(FP_CLASS_NON_INTEGRAL | FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_GT_ZERO_LT_POS_ONE);
}
break;
}
case nir_op_fround_even: {
r = src_res[0];
if (r & FP_CLASS_NON_INTEGRAL) {
if (r & FP_CLASS_LT_NEG_ONE)
r |= FP_CLASS_NEG_ONE;
if (r & FP_CLASS_LT_ZERO_GT_NEG_ONE)
r |= FP_CLASS_NEG_ZERO | FP_CLASS_NEG_ONE;
if (r & FP_CLASS_GT_ZERO_LT_POS_ONE)
r |= FP_CLASS_POS_ZERO | FP_CLASS_POS_ONE;
if (r & FP_CLASS_GT_POS_ONE)
r |= FP_CLASS_POS_ONE;
r &= ~(FP_CLASS_NON_INTEGRAL | FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_GT_ZERO_LT_POS_ONE);
}
break;
}
case nir_op_ffract: {
r = 0;
/* fract(±Inf) is NaN. */
if (src_res[0] & (FP_CLASS_ANY_INF | FP_CLASS_NAN))
r |= FP_CLASS_NAN;
/* fract(non_integral) is in (0, 1). */
if (src_res[0] & FP_CLASS_NON_INTEGRAL)
r |= FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_NON_INTEGRAL;
/* fract(small, negative) can be 1.0. */
if (src_res[0] & FP_CLASS_LT_ZERO_GT_NEG_ONE)
r |= FP_CLASS_POS_ONE;
/* fract(integral) is +0.0. */
if (src_res[0] & (FP_CLASS_LT_NEG_ONE | FP_CLASS_NEG_ONE | FP_CLASS_ANY_ZERO | FP_CLASS_POS_ONE | FP_CLASS_GT_POS_ONE))
r |= FP_CLASS_POS_ZERO;
break;
}
case nir_op_fsin:
case nir_op_fcos:
case nir_op_fsin_amd:
case nir_op_fcos_amd: {
/* [-1, +1], and sin/cos(Inf) is NaN */
r = FP_CLASS_NEG_ONE | FP_CLASS_LT_ZERO_GT_NEG_ONE | FP_CLASS_ANY_ZERO |
FP_CLASS_GT_ZERO_LT_POS_ONE | FP_CLASS_POS_ONE | FP_CLASS_NON_INTEGRAL;
if (src_res[0] & (FP_CLASS_NAN | FP_CLASS_ANY_INF))
r |= FP_CLASS_NAN;
break;
}
case nir_op_fdot2:
case nir_op_fdot3:
case nir_op_fdot4:
case nir_op_fdot8:
case nir_op_fdot16:
case nir_op_fdot2_replicated:
case nir_op_fdot3_replicated:
case nir_op_fdot4_replicated:
case nir_op_fdot8_replicated:
case nir_op_fdot16_replicated: {
/* If the two sources are the same SSA value, then the result is either
* NaN or some number >= 0. If one source is the negation of the other,
* the result is either NaN or some number <= 0.
*
* In either of these two cases, if one source is a number, then the
* other must also be a number. Since it should not be possible to get
* Inf-Inf in the dot-product, the result must also be a number.
*/
if (nir_alu_srcs_equal(alu, alu, 0, 1)) {
r = FP_CLASS_ANY_POS | FP_CLASS_POS_ZERO;
} else if (nir_alu_srcs_negative_equal(alu, alu, 0, 1)) {
r = FP_CLASS_ANY_NEG | FP_CLASS_NEG_ZERO;
} else {
r = FP_CLASS_UNKNOWN;
}
if (src_res[0] & FP_CLASS_NAN)
r |= FP_CLASS_NAN;
if (src_res[0] & FP_CLASS_NON_INTEGRAL)
r |= FP_CLASS_NON_INTEGRAL;
break;
}
case nir_op_fpow: {
/* This is a basic port of the old range analysis, the opcode is very
* underdefined. But improvements are likely possible.
* Due to flush-to-zero semanatics of floating-point numbers with very
* small mangnitudes, we can never really be sure a result will be
* non-zero.
*
* NIR uses pow() and powf() to constant evaluate nir_op_fpow. The man
* page for that function says:
*
* If y is 0, the result is 1.0 (even if x is a NaN).
*
* gt_zero: pow(*, eq_zero)
* | pow(eq_zero, lt_zero) # 0^-y = +inf
* | pow(eq_zero, le_zero) # 0^-y = +inf or 0^0 = 1.0
* ;
*
* eq_zero: pow(eq_zero, gt_zero)
* ;
*
* ge_zero: pow(gt_zero, gt_zero)
* | pow(gt_zero, ge_zero)
* | pow(gt_zero, lt_zero)
* | pow(gt_zero, le_zero)
* | pow(gt_zero, ne_zero)
* | pow(gt_zero, unknown)
* | pow(ge_zero, gt_zero)
* | pow(ge_zero, ge_zero)
* | pow(ge_zero, lt_zero)
* | pow(ge_zero, le_zero)
* | pow(ge_zero, ne_zero)
* | pow(ge_zero, unknown)
* | pow(eq_zero, ge_zero) # 0^0 = 1.0 or 0^+y = 0.0
* | pow(eq_zero, ne_zero) # 0^-y = +inf or 0^+y = 0.0
* | pow(eq_zero, unknown) # union of all other y cases
* ;
*
* All other cases are unknown.
*
* We could do better if the right operand is a constant, integral
* value.
*/
fp_class_mask left = src_res[0];
fp_class_mask right = src_res[1];
if (!(right & (FP_CLASS_ANY_NEG | FP_CLASS_ANY_POS))) {
r = FP_CLASS_ANY_POS;
} else if (left & (FP_CLASS_ANY_NEG | FP_CLASS_NEG_ZERO)) {
r = FP_CLASS_UNKNOWN;
} else {
r = FP_CLASS_ANY_POS | FP_CLASS_ANY_ZERO;
if ((right & (FP_CLASS_ANY_NEG | FP_CLASS_NON_INTEGRAL)) || (left & FP_CLASS_NON_INTEGRAL))
r |= FP_CLASS_NON_INTEGRAL;
}
/* Various cases can result in NaN, so assume the worst. */
r |= FP_CLASS_NAN;
break;
}
case nir_op_ffma:
case nir_op_ffmaz: {
bool mulz = alu->op == nir_op_ffmaz;
bool src_eq = nir_alu_srcs_equal(alu, alu, 0, 1);
bool src_neg_eq = !nir_src_is_const(alu->src[0].src) && nir_alu_srcs_negative_equal(alu, alu, 0, 1);
fp_class_mask r_mul = fmul_fp_class(src_res[0], src_res[1], mulz, src_eq, src_neg_eq);
r = fadd_fp_class(r_mul, src_res[2]);
/* fma(a, b, +0.0) can be -0.0 if a * b underflows.
* When fused, the underflow is not flushed before the addition.
*/
bool mul_underflow = (((src_res[0] & FP_CLASS_LT_ZERO_GT_NEG_ONE) && (src_res[1] & FP_CLASS_GT_ZERO_LT_POS_ONE)) ||
((src_res[1] & FP_CLASS_LT_ZERO_GT_NEG_ONE) && (src_res[0] & FP_CLASS_GT_ZERO_LT_POS_ONE)));
if (!src_eq && mul_underflow && (src_res[2] & FP_CLASS_POS_ZERO))
r |= FP_CLASS_NEG_ZERO;
break;
}
case nir_op_flrp: {
/* Decompose the flrp to first + third * (second + -first) */
fp_class_mask inner_fadd_class =
fadd_fp_class(src_res[1], fneg_fp_class(src_res[0]));
fp_class_mask fmul_class =
fmul_fp_class(src_res[2], inner_fadd_class, false, false, false);
r = fadd_fp_class(src_res[0], fmul_class);
/* Various cases can result in NaN, so assume the worst. */
r |= FP_CLASS_NAN;
break;
}
default:
r = FP_CLASS_UNKNOWN;
break;
}
if (nir_alu_type_get_base_type(nir_op_infos[alu->op].output_type) == nir_type_float)
r = handle_sz(alu, r);
assert((r & FP_CLASS_UNKNOWN) == r);
assert((r & ~FP_CLASS_NON_INTEGRAL) != 0);
assert(!(r & FP_CLASS_NON_INTEGRAL) || (r & (FP_CLASS_LT_NEG_ONE | FP_CLASS_LT_ZERO_GT_NEG_ONE |
FP_CLASS_GT_POS_ONE | FP_CLASS_GT_ZERO_LT_POS_ONE)));
*result = r;
}
fp_class_mask
nir_analyze_fp_class(nir_fp_analysis_state *fp_state, const nir_def *def)
{
struct fp_query query_alloc[64];
uint32_t result_alloc[64];
struct analysis_state state;
state.range_ht = fp_state;
util_dynarray_init_from_stack(&state.query_stack, query_alloc, sizeof(query_alloc));
util_dynarray_init_from_stack(&state.result_stack, result_alloc, sizeof(result_alloc));
state.query_size = sizeof(struct fp_query);
state.get_key = &get_fp_key;
state.lookup = &fp_lookup;
state.insert = &fp_insert;
state.process_query = &process_fp_query;
push_fp_query(&state, def);
return perform_analysis(&state);
}
nir_fp_analysis_state
nir_create_fp_analysis_state(nir_function_impl *impl)
{
nir_fp_analysis_state state;
state.impl = impl;
/* Over-allocate the bitset, so that we can keep using the allocated table memory
* even when new SSA values are added. */
state.size = BITSET_BYTES(impl->ssa_alloc + impl->ssa_alloc / 4u);
state.max = -1;
state.bitset = calloc(state.size, 1);
util_sparse_array_init(&state.arr, 4, 256);
return state;
}
void
nir_invalidate_fp_analysis_state(nir_fp_analysis_state *state)
{
if (BITSET_BYTES(state->impl->ssa_alloc) > state->size) {
state->size = BITSET_BYTES(state->impl->ssa_alloc + state->impl->ssa_alloc / 4u);
free(state->bitset);
state->bitset = calloc(state->size, 1);
} else if (state->max >= 0) {
memset(state->bitset, 0, BITSET_BYTES(state->max + 1));
}
state->max = -1;
}
void
nir_free_fp_analysis_state(nir_fp_analysis_state *state)
{
util_sparse_array_finish(&state->arr);
free(state->bitset);
}
static uint32_t
bitmask(uint32_t size)
{
return size >= 32 ? 0xffffffffu : ((uint32_t)1 << size) - 1u;
}
static uint64_t
mul_clamp(uint32_t a, uint32_t b)
{
if (a != 0 && (a * b) / a != b)
return (uint64_t)UINT32_MAX + 1;
else
return a * b;
}
/* recursively gather at most "buf_size" phi/bcsel sources */
static unsigned
search_phi_bcsel(nir_scalar scalar, nir_scalar *buf, unsigned buf_size, struct set *visited)
{
if (_mesa_set_search(visited, scalar.def))
return 0;
_mesa_set_add(visited, scalar.def);
if (nir_def_instr_type(scalar.def) == nir_instr_type_phi) {
nir_phi_instr *phi = nir_def_as_phi(scalar.def);
unsigned num_sources_left = exec_list_length(&phi->srcs);
if (buf_size >= num_sources_left) {
unsigned total_added = 0;
nir_foreach_phi_src(src, phi) {
num_sources_left--;
unsigned added = search_phi_bcsel(nir_get_scalar(src->src.ssa, scalar.comp),
buf + total_added, buf_size - num_sources_left, visited);
assert(added <= buf_size);
buf_size -= added;
total_added += added;
}
return total_added;
}
}
if (nir_scalar_is_alu(scalar)) {
nir_op op = nir_scalar_alu_op(scalar);
if ((op == nir_op_bcsel || op == nir_op_b32csel) && buf_size >= 2) {
nir_scalar src1 = nir_scalar_chase_alu_src(scalar, 1);
nir_scalar src2 = nir_scalar_chase_alu_src(scalar, 2);
unsigned added = search_phi_bcsel(src1, buf, buf_size - 1, visited);
buf_size -= added;
added += search_phi_bcsel(src2, buf + added, buf_size, visited);
return added;
}
}
buf[0] = scalar;
return 1;
}
static uint32_t
get_max_workgroup_invocations(nir_shader *nir)
{
if (!nir->options || !nir->options->max_workgroup_invocations)
return UINT16_MAX;
return nir->options->max_workgroup_invocations;
}
static uint32_t
get_max_workgroup_count(nir_shader *nir, unsigned dim)
{
/* max_workgroup_count represents the maximum compute shader / kernel
* dispatchable work size. On most hardware, this is essentially
* unbounded. On some hardware max_workgroup_count[1] and
* max_workgroup_count[2] may be smaller.
*/
if (!nir->options || !nir->options->max_workgroup_count[dim])
return UINT32_MAX;
return nir->options->max_workgroup_count[dim];
}
struct scalar_query {
struct analysis_query head;
nir_scalar scalar;
};
static void
push_scalar_query(struct analysis_state *state, nir_scalar scalar)
{
struct scalar_query *pushed_q = push_analysis_query(state, sizeof(struct scalar_query));
pushed_q->scalar = scalar;
}
static uint32_t
get_scalar_key(struct analysis_query *q)
{
nir_scalar scalar = ((struct scalar_query *)q)->scalar;
/* keys can't be 0, so we have to add 1 to the index */
unsigned shift_amount = ffs(NIR_MAX_VEC_COMPONENTS) - 1;
return nir_scalar_is_const(scalar)
? UINT32_MAX
: ((scalar.def->index + 1) << shift_amount) | scalar.comp;
}
static bool
scalar_lookup(void *table, uint32_t key, uint32_t *value)
{
struct hash_table *ht = table;
struct hash_entry *he = _mesa_hash_table_search(ht, (void *)(uintptr_t)key);
if (he)
*value = (uintptr_t)he->data;
return he != NULL;
}
static void
scalar_insert(void *table, uint32_t key, uint32_t value)
{
struct hash_table *ht = table;
_mesa_hash_table_insert(ht, (void *)(uintptr_t)key, (void *)(uintptr_t)value);
}
static void
get_intrinsic_uub(struct analysis_state *state, struct scalar_query q, uint32_t *result,
const uint32_t *src)
{
nir_shader *shader = state->shader;
nir_intrinsic_instr *intrin = nir_def_as_intrinsic(q.scalar.def);
switch (intrin->intrinsic) {
case nir_intrinsic_load_local_invocation_index:
/* The local invocation index is used under the hood by RADV for
* some non-compute-like shaders (eg. LS and NGG). These technically
* run in workgroups on the HW, even though this fact is not exposed
* by the API.
* They can safely use the same code path here as variable sized
* compute-like shader stages.
*/
if (!mesa_shader_stage_uses_workgroup(shader->info.stage) ||
shader->info.workgroup_size_variable) {
*result = get_max_workgroup_invocations(shader) - 1;
} else {
*result = (shader->info.workgroup_size[0] *
shader->info.workgroup_size[1] *
shader->info.workgroup_size[2]) -
1u;
}
break;
case nir_intrinsic_load_local_invocation_id:
if (shader->info.workgroup_size_variable)
*result = get_max_workgroup_invocations(shader) - 1u;
else
*result = shader->info.workgroup_size[q.scalar.comp] - 1u;
break;
case nir_intrinsic_load_workgroup_id:
*result = get_max_workgroup_count(shader, q.scalar.comp) - 1u;
break;
case nir_intrinsic_load_num_workgroups:
*result = get_max_workgroup_count(shader, q.scalar.comp);
break;
case nir_intrinsic_load_global_invocation_id:
if (shader->info.workgroup_size_variable) {
*result = mul_clamp(get_max_workgroup_invocations(shader),
get_max_workgroup_count(shader, q.scalar.comp)) -
1u;
} else {
*result = (shader->info.workgroup_size[q.scalar.comp] *
get_max_workgroup_count(shader, q.scalar.comp)) -
1u;
}
break;
case nir_intrinsic_load_invocation_id:
if (shader->info.stage == MESA_SHADER_TESS_CTRL)
*result = shader->info.tess.tcs_vertices_out
? (shader->info.tess.tcs_vertices_out - 1)
: 511; /* Generous maximum output patch size of 512 */
break;
case nir_intrinsic_load_subgroup_invocation:
case nir_intrinsic_first_invocation:
*result = shader->info.max_subgroup_size - 1;
break;
case nir_intrinsic_mbcnt_amd: {
if (!q.head.pushed_queries) {
push_scalar_query(state, nir_get_scalar(intrin->src[1].ssa, 0));
return;
} else {
uint32_t src0 = shader->info.max_subgroup_size - 1;
uint32_t src1 = src[0];
if (src0 + src1 >= src0) /* check overflow */
*result = src0 + src1;
}
break;
}
case nir_intrinsic_load_subgroup_size:
if (shader->info.api_subgroup_size)
*result = shader->info.api_subgroup_size;
else
*result = shader->info.max_subgroup_size;
break;
case nir_intrinsic_load_subgroup_id:
case nir_intrinsic_load_num_subgroups: {
uint32_t workgroup_size = get_max_workgroup_invocations(shader);
if (mesa_shader_stage_uses_workgroup(shader->info.stage) &&
!shader->info.workgroup_size_variable) {
workgroup_size = shader->info.workgroup_size[0] *
shader->info.workgroup_size[1] *
shader->info.workgroup_size[2];
}
*result = DIV_ROUND_UP(workgroup_size, shader->info.min_subgroup_size);
if (intrin->intrinsic == nir_intrinsic_load_subgroup_id)
(*result)--;
break;
}
case nir_intrinsic_reduce:
case nir_intrinsic_inclusive_scan:
case nir_intrinsic_exclusive_scan: {
nir_op op = nir_intrinsic_reduction_op(intrin);
switch (op) {
case nir_op_umin:
case nir_op_umax:
case nir_op_imax:
case nir_op_imin:
case nir_op_iand:
case nir_op_ior:
case nir_op_ixor:
case nir_op_iadd:
if (!q.head.pushed_queries) {
push_scalar_query(state, nir_get_scalar(intrin->src[0].ssa, q.scalar.comp));
return;
}
break;
default:
return;
}
unsigned bit_size = q.scalar.def->bit_size;
bool exclusive = intrin->intrinsic == nir_intrinsic_exclusive_scan;
switch (op) {
case nir_op_umin:
case nir_op_umax:
case nir_op_imax:
case nir_op_imin:
case nir_op_iand:
*result = src[0];
break;
case nir_op_ior:
case nir_op_ixor:
*result = bitmask(util_last_bit64(src[0]));
break;
case nir_op_iadd:
*result = MIN2(*result, (uint64_t)src[0] * (shader->info.max_subgroup_size - exclusive));
break;
default:
UNREACHABLE("unhandled op");
}
if (exclusive) {
uint32_t identity = nir_const_value_as_uint(nir_alu_binop_identity(op, bit_size), bit_size);
*result = MAX2(*result, identity);
}
break;
}
case nir_intrinsic_read_first_invocation:
case nir_intrinsic_read_invocation:
case nir_intrinsic_shuffle:
case nir_intrinsic_shuffle_xor:
case nir_intrinsic_shuffle_up:
case nir_intrinsic_shuffle_down:
case nir_intrinsic_quad_broadcast:
case nir_intrinsic_quad_swap_horizontal:
case nir_intrinsic_quad_swap_vertical:
case nir_intrinsic_quad_swap_diagonal:
case nir_intrinsic_quad_swizzle_amd:
case nir_intrinsic_masked_swizzle_amd:
if (!q.head.pushed_queries) {
push_scalar_query(state, nir_get_scalar(intrin->src[0].ssa, q.scalar.comp));
return;
} else {
*result = src[0];
}
break;
case nir_intrinsic_write_invocation_amd:
if (!q.head.pushed_queries) {
push_scalar_query(state, nir_get_scalar(intrin->src[0].ssa, q.scalar.comp));
push_scalar_query(state, nir_get_scalar(intrin->src[1].ssa, q.scalar.comp));
return;
} else {
*result = MAX2(src[0], src[1]);
}
break;
case nir_intrinsic_load_tess_rel_patch_id_amd:
case nir_intrinsic_load_tcs_num_patches_amd:
/* Very generous maximum: TCS/TES executed by largest possible workgroup */
*result = get_max_workgroup_invocations(shader) / MAX2(shader->info.tess.tcs_vertices_out, 1u);
break;
case nir_intrinsic_load_typed_buffer_amd: {
const enum pipe_format format = nir_intrinsic_format(intrin);
if (format == PIPE_FORMAT_NONE)
break;
const struct util_format_description *desc = util_format_description(format);
if (desc->channel[q.scalar.comp].type != UTIL_FORMAT_TYPE_UNSIGNED)
break;
if (desc->channel[q.scalar.comp].normalized) {
*result = fui(1.0);
break;
}
const uint32_t chan_max = u_uintN_max(desc->channel[q.scalar.comp].size);
*result = desc->channel[q.scalar.comp].pure_integer ? chan_max : fui(chan_max);
break;
}
case nir_intrinsic_load_ttmp_register_amd:
case nir_intrinsic_load_scalar_arg_amd:
case nir_intrinsic_load_vector_arg_amd: {
uint32_t upper_bound = nir_intrinsic_arg_upper_bound_u32_amd(intrin);
if (upper_bound)
*result = upper_bound;
break;
}
case nir_intrinsic_image_samples:
if (state->shader->options->max_samples > 0)
*result = state->shader->options->max_samples;
break;
default:
break;
}
}
static void
get_alu_uub(struct analysis_state *state, struct scalar_query q, uint32_t *result, const uint32_t *src)
{
nir_op op = nir_scalar_alu_op(q.scalar);
/* Early exit for unsupported ALU opcodes. */
switch (op) {
case nir_op_umin:
case nir_op_imin:
case nir_op_imax:
case nir_op_umax:
case nir_op_iand:
case nir_op_ior:
case nir_op_ixor:
case nir_op_ishl:
case nir_op_imul:
case nir_op_ushr:
case nir_op_ishr:
case nir_op_iadd:
case nir_op_umod:
case nir_op_udiv:
case nir_op_bcsel:
case nir_op_b32csel:
case nir_op_ubfe:
case nir_op_bfi:
case nir_op_bfm:
case nir_op_bitfield_select:
case nir_op_extract_u8:
case nir_op_extract_i8:
case nir_op_extract_u16:
case nir_op_extract_i16:
case nir_op_b2i8:
case nir_op_b2i16:
case nir_op_b2i32:
break;
case nir_op_u2u1:
case nir_op_u2u8:
case nir_op_u2u16:
case nir_op_u2u32:
if (nir_scalar_chase_alu_src(q.scalar, 0).def->bit_size > 32) {
/* If src is >32 bits, return max */
return;
}
break;
case nir_op_fsat:
case nir_op_fmul:
case nir_op_fmulz:
case nir_op_f2u32:
case nir_op_f2i32:
if (nir_scalar_chase_alu_src(q.scalar, 0).def->bit_size != 32) {
/* Only 32bit floats support for now, return max */
return;
}
break;
case nir_op_bit_count:
if (nir_scalar_chase_alu_src(q.scalar, 0).def->bit_size > 32) {
*result = nir_scalar_chase_alu_src(q.scalar, 0).def->bit_size;
return;
}
break;
default:
return;
}
if (!q.head.pushed_queries) {
for (unsigned i = 0; i < nir_op_infos[op].num_inputs; i++)
push_scalar_query(state, nir_scalar_chase_alu_src(q.scalar, i));
return;
}
uint32_t max = bitmask(q.scalar.def->bit_size);
switch (op) {
case nir_op_umin:
*result = src[0] < src[1] ? src[0] : src[1];
break;
case nir_op_imin:
case nir_op_imax:
case nir_op_umax:
*result = src[0] > src[1] ? src[0] : src[1];
break;
case nir_op_iand: {
nir_scalar src0_scalar = nir_scalar_chase_alu_src(q.scalar, 0);
nir_scalar src1_scalar = nir_scalar_chase_alu_src(q.scalar, 1);
if (nir_scalar_is_const(src0_scalar))
*result = bitmask(util_last_bit64(src[1])) & nir_scalar_as_uint(src0_scalar);
else if (nir_scalar_is_const(src1_scalar))
*result = bitmask(util_last_bit64(src[0])) & nir_scalar_as_uint(src1_scalar);
else
*result = bitmask(util_last_bit64(src[0])) & bitmask(util_last_bit64(src[1]));
break;
}
case nir_op_ior:
case nir_op_ixor: {
nir_scalar src0_scalar = nir_scalar_chase_alu_src(q.scalar, 0);
nir_scalar src1_scalar = nir_scalar_chase_alu_src(q.scalar, 1);
if (nir_scalar_is_const(src0_scalar))
*result = bitmask(util_last_bit64(src[1])) | nir_scalar_as_uint(src0_scalar);
else if (nir_scalar_is_const(src1_scalar))
*result = bitmask(util_last_bit64(src[0])) | nir_scalar_as_uint(src1_scalar);
else
*result = bitmask(util_last_bit64(src[0])) | bitmask(util_last_bit64(src[1]));
break;
}
case nir_op_ishl: {
uint32_t src1 = MIN2(src[1], q.scalar.def->bit_size - 1u);
if (util_last_bit64(src[0]) + src1 <= q.scalar.def->bit_size) /* check overflow */
*result = src[0] << src1;
*result = MIN2(*result, max);
nir_scalar src1_scalar = nir_scalar_chase_alu_src(q.scalar, 1);
if (nir_scalar_is_const(src1_scalar)) {
uint32_t const_val = 1u << (nir_scalar_as_uint(src1_scalar) & (q.scalar.def->bit_size - 1u));
*result = MIN2(*result, max / const_val * const_val);
}
break;
}
case nir_op_imul: {
if (src[0] == 0 || (src[0] * src[1]) / src[0] == src[1]) /* check overflow */
*result = src[0] * src[1];
*result = MIN2(*result, max);
nir_scalar src0_scalar = nir_scalar_chase_alu_src(q.scalar, 0);
nir_scalar src1_scalar = nir_scalar_chase_alu_src(q.scalar, 1);
if (nir_scalar_is_const(src0_scalar)) {
uint32_t const_val = nir_scalar_as_uint(src0_scalar);
*result = const_val ? MIN2(*result, max / const_val * const_val) : 0;
} else if (nir_scalar_is_const(src1_scalar)) {
uint32_t const_val = nir_scalar_as_uint(src1_scalar);
*result = const_val ? MIN2(*result, max / const_val * const_val) : 0;
}
break;
}
case nir_op_ushr: {
nir_scalar src1_scalar = nir_scalar_chase_alu_src(q.scalar, 1);
uint32_t mask = q.scalar.def->bit_size - 1u;
if (nir_scalar_is_const(src1_scalar))
*result = src[0] >> (nir_scalar_as_uint(src1_scalar) & mask);
else
*result = src[0];
break;
}
case nir_op_ishr: {
nir_scalar src1_scalar = nir_scalar_chase_alu_src(q.scalar, 1);
uint32_t mask = q.scalar.def->bit_size - 1u;
if (src[0] <= 2147483647 && nir_scalar_is_const(src1_scalar))
*result = src[0] >> (nir_scalar_as_uint(src1_scalar) & mask);
else
*result = src[0];
break;
}
case nir_op_iadd:
if (src[0] + src[1] >= src[0]) /* check overflow */
*result = src[0] + src[1];
*result = MIN2(*result, max);
break;
case nir_op_umod:
*result = src[1] ? src[1] - 1 : 0;
break;
case nir_op_udiv: {
nir_scalar src1_scalar = nir_scalar_chase_alu_src(q.scalar, 1);
if (nir_scalar_is_const(src1_scalar))
*result = nir_scalar_as_uint(src1_scalar)
? src[0] / nir_scalar_as_uint(src1_scalar)
: 0;
else
*result = src[0];
break;
}
case nir_op_bcsel:
case nir_op_b32csel:
*result = src[1] > src[2] ? src[1] : src[2];
break;
case nir_op_ubfe:
*result = bitmask(MIN2(src[2], q.scalar.def->bit_size));
break;
case nir_op_bfm: {
nir_scalar src1_scalar = nir_scalar_chase_alu_src(q.scalar, 1);
if (nir_scalar_is_const(src1_scalar)) {
uint32_t src0 = MIN2(src[0], 31);
uint32_t src1 = nir_scalar_as_uint(src1_scalar) & 0x1fu;
*result = bitmask(src0) << src1;
} else {
uint32_t src0 = MIN2(src[0], 31);
uint32_t src1 = MIN2(src[1], 31);
*result = bitmask(MIN2(src0 + src1, 32));
}
break;
}
case nir_op_bfi: {
nir_scalar src0_scalar = nir_scalar_chase_alu_src(q.scalar, 0);
const uint64_t s1 = bitmask(util_last_bit64(src[1]));
const uint64_t s2 = bitmask(util_last_bit64(src[2]));
if (nir_scalar_is_const(src0_scalar)) {
const uint64_t s0 = nir_scalar_as_uint(src0_scalar);
/* This case should be eliminated by opt_algebraic. */
if (s0 == 0) {
*result = s2;
} else {
const int x = ffsll(s0) - 1;
*result = (s0 & (s1 << x)) | (~s0 & s2);
}
} else {
const uint64_t s0 = bitmask(util_last_bit64(src[0]));
/* Due to the unpredictable shift, the true maximum value of (s0 &
* (s1 << x)) cannot be known. However, it cannot be larger than
* s0.
*
* inot doesn't work in get_alu_uub. It is known that (~s0 & s2)
* cannot be larger than s2, so just use s2 as a loose upper bound.
*/
*result = s0 | s2;
}
break;
}
case nir_op_bitfield_select: {
nir_scalar src0_scalar = nir_scalar_chase_alu_src(q.scalar, 0);
const uint64_t s1 = bitmask(util_last_bit64(src[1]));
const uint64_t s2 = bitmask(util_last_bit64(src[2]));
if (nir_scalar_is_const(src0_scalar)) {
const uint64_t s0 = nir_scalar_as_uint(src0_scalar);
*result = (s0 & s1) | (~s0 & s2);
} else {
const uint64_t s0 = bitmask(util_last_bit64(src[0]));
/* inot doesn't work in get_alu_uub. It is known that (~s0 & s2)
* cannot be larger than s2, so just use s2 as a loose upper bound.
*/
*result = (s0 & s1) | s2;
}
break;
}
/* limited floating-point support for f2u32(fmul(load_input(), <constant>)) */
case nir_op_f2i32:
case nir_op_f2u32:
/* infinity/NaN starts at 0x7f800000u, negative numbers at 0x80000000 */
if (src[0] < 0x7f800000u) {
float val;
memcpy(&val, &src[0], 4);
*result = (uint32_t)val;
}
break;
case nir_op_fmul:
case nir_op_fmulz:
/* infinity/NaN starts at 0x7f800000u, negative numbers at 0x80000000 */
if (src[0] < 0x7f800000u && src[1] < 0x7f800000u) {
float src0_f, src1_f;
memcpy(&src0_f, &src[0], 4);
memcpy(&src1_f, &src[1], 4);
/* not a proper rounding-up multiplication, but should be good enough */
float max_f = ceilf(src0_f) * ceilf(src1_f);
memcpy(result, &max_f, 4);
}
break;
case nir_op_fsat:
*result = 0x3f800000u;
break;
case nir_op_u2u1:
case nir_op_u2u8:
case nir_op_u2u16:
case nir_op_u2u32:
*result = MIN2(src[0], max);
break;
case nir_op_b2i8:
case nir_op_b2i16:
case nir_op_b2i32:
*result = 1;
break;
case nir_op_msad_4x8:
*result = MIN2((uint64_t)src[2] + 4 * 255, UINT32_MAX);
break;
case nir_op_extract_u8:
*result = MIN2(src[0], UINT8_MAX);
break;
case nir_op_extract_i8:
*result = (src[0] >= 0x80) ? max : MIN2(src[0], INT8_MAX);
break;
case nir_op_extract_u16:
*result = MIN2(src[0], UINT16_MAX);
break;
case nir_op_extract_i16:
*result = (src[0] >= 0x8000) ? max : MIN2(src[0], INT16_MAX);
break;
case nir_op_bit_count:
*result = util_last_bit64(src[0]);
break;
default:
break;
}
}
static void
get_tex_uub(struct analysis_state *state, struct scalar_query q, uint32_t *result, const uint32_t *src)
{
nir_tex_instr *tex = nir_scalar_as_tex(q.scalar);
if (tex->op == nir_texop_texture_samples && state->shader->options->max_samples > 0)
*result = state->shader->options->max_samples;
}
static void
get_phi_uub(struct analysis_state *state, struct scalar_query q, uint32_t *result, const uint32_t *src)
{
nir_phi_instr *phi = nir_def_as_phi(q.scalar.def);
if (exec_list_is_empty(&phi->srcs))
return;
if (q.head.pushed_queries) {
*result = src[0];
for (unsigned i = 1; i < q.head.pushed_queries; i++)
*result = MAX2(*result, src[i]);
return;
}
nir_cf_node *prev = nir_cf_node_prev(&phi->instr.block->cf_node);
if (!prev || prev->type == nir_cf_node_block) {
/* Resolve cycles by inserting max into range_ht. */
uint32_t max = bitmask(q.scalar.def->bit_size);
scalar_insert(state->range_ht, get_scalar_key(&q.head), max);
struct set *visited = _mesa_pointer_set_create(NULL);
nir_scalar *defs = alloca(sizeof(nir_scalar) * 64);
unsigned def_count = search_phi_bcsel(q.scalar, defs, 64, visited);
_mesa_set_destroy(visited, NULL);
for (unsigned i = 0; i < def_count; i++)
push_scalar_query(state, defs[i]);
} else {
nir_foreach_phi_src(src, phi)
push_scalar_query(state, nir_get_scalar(src->src.ssa, q.scalar.comp));
}
}
static void
process_uub_query(struct analysis_state *state, struct analysis_query *aq, uint32_t *result,
const uint32_t *src)
{
struct scalar_query q = *(struct scalar_query *)aq;
*result = bitmask(q.scalar.def->bit_size);
if (nir_scalar_is_const(q.scalar))
*result = nir_scalar_as_uint(q.scalar);
else if (nir_scalar_is_intrinsic(q.scalar))
get_intrinsic_uub(state, q, result, src);
else if (nir_scalar_is_alu(q.scalar))
get_alu_uub(state, q, result, src);
else if (nir_def_instr_type(q.scalar.def) == nir_instr_type_tex)
get_tex_uub(state, q, result, src);
else if (nir_def_instr_type(q.scalar.def) == nir_instr_type_phi)
get_phi_uub(state, q, result, src);
}
uint32_t
nir_unsigned_upper_bound(nir_shader *shader, struct hash_table *range_ht,
nir_scalar scalar)
{
struct scalar_query query_alloc[16];
uint32_t result_alloc[16];
struct analysis_state state;
state.shader = shader;
state.range_ht = range_ht;
util_dynarray_init_from_stack(&state.query_stack, query_alloc, sizeof(query_alloc));
util_dynarray_init_from_stack(&state.result_stack, result_alloc, sizeof(result_alloc));
state.query_size = sizeof(struct scalar_query);
state.get_key = &get_scalar_key;
state.lookup = &scalar_lookup,
state.insert = &scalar_insert,
state.process_query = &process_uub_query;
push_scalar_query(&state, scalar);
_mesa_hash_table_set_deleted_key(range_ht, (void *)(uintptr_t)UINT32_MAX);
return perform_analysis(&state);
}
bool
nir_addition_might_overflow(nir_shader *shader, struct hash_table *range_ht,
nir_scalar ssa, unsigned const_val)
{
uint32_t ub = nir_unsigned_upper_bound(shader, range_ht, ssa);
return const_val + ub < const_val;
}
static uint64_t
ssa_def_bits_used(const nir_def *def, int recur)
{
uint64_t bits_used = 0;
uint64_t all_bits = BITFIELD64_MASK(def->bit_size);
/* Querying the bits used from a vector is too hard of a question to
* answer. Return the conservative answer that all bits are used. To
* handle this, the function would need to be extended to be a query of a
* single component of the vector. That would also necessary to fully
* handle the 'num_components > 1' inside the loop below.
*
* FINISHME: This restriction will eventually need to be restricted to be
* useful for hardware that uses u16vec2 as the native 16-bit integer type.
*/
if (def->num_components > 1)
return all_bits;
/* Limit recursion */
if (recur-- <= 0)
return all_bits;
nir_foreach_use(src, def) {
switch (nir_src_parent_instr(src)->type) {
case nir_instr_type_alu: {
nir_alu_instr *use_alu = nir_instr_as_alu(nir_src_parent_instr(src));
unsigned src_idx = container_of(src, nir_alu_src, src) - use_alu->src;
/* If a user of the value produces a vector result, return the
* conservative answer that all bits are used. It is possible to
* answer this query by looping over the components used. For example,
*
* vec4 32 ssa_5 = load_const(0x0000f000, 0x00000f00, 0x000000f0, 0x0000000f)
* ...
* vec4 32 ssa_8 = iand ssa_7.xxxx, ssa_5
*
* could conceivably return 0x0000ffff when queyring the bits used of
* ssa_7. This is unlikely to be worth the effort because the
* question can eventually answered after the shader has been
* scalarized.
*/
if (use_alu->def.num_components > 1)
return all_bits;
switch (use_alu->op) {
case nir_op_u2u8:
case nir_op_i2i8:
case nir_op_u2u16:
case nir_op_i2i16:
case nir_op_u2u32:
case nir_op_i2i32:
case nir_op_u2u64:
case nir_op_i2i64: {
uint64_t def_bits_used = ssa_def_bits_used(&use_alu->def, recur);
/* If one of the sign-extended bits is used, set the last src bit
* as used.
*/
if ((use_alu->op == nir_op_i2i8 || use_alu->op == nir_op_i2i16 ||
use_alu->op == nir_op_i2i32 || use_alu->op == nir_op_i2i64) &&
def_bits_used & ~all_bits)
def_bits_used |= BITFIELD64_BIT(def->bit_size - 1);
bits_used |= def_bits_used & all_bits;
break;
}
case nir_op_extract_u8:
case nir_op_extract_i8:
case nir_op_extract_u16:
case nir_op_extract_i16:
if (src_idx == 0 && nir_src_is_const(use_alu->src[1].src)) {
unsigned chunk = nir_alu_src_as_uint(use_alu->src[1]);
uint64_t defs_bits_used = ssa_def_bits_used(&use_alu->def, recur);
unsigned field_bits = use_alu->op == nir_op_extract_u8 ||
use_alu->op == nir_op_extract_i8 ? 8 : 16;
uint64_t field_bitmask = BITFIELD64_MASK(field_bits);
/* If one of the sign-extended bits is used, set the last src bit
* as used.
*/
if ((use_alu->op == nir_op_extract_i8 ||
use_alu->op == nir_op_extract_i16) &&
defs_bits_used & ~field_bitmask)
defs_bits_used |= BITFIELD64_BIT(field_bits - 1);
bits_used |= (field_bitmask & defs_bits_used) <<
(chunk * field_bits);
break;
} else {
return all_bits;
}
case nir_op_ishl:
case nir_op_ishr:
case nir_op_ushr:
if (src_idx == 0 && nir_src_is_const(use_alu->src[1].src)) {
unsigned bit_size = def->bit_size;
unsigned shift = nir_alu_src_as_uint(use_alu->src[1]) & (bit_size - 1);
uint64_t def_bits_used = ssa_def_bits_used(&use_alu->def, recur);
/* If one of the sign-extended bits is used, set the "last src
* bit before shifting" as used.
*/
if (use_alu->op == nir_op_ishr &&
def_bits_used & ~(all_bits >> shift))
def_bits_used |= BITFIELD64_BIT(bit_size - 1 - shift);
/* Reverse the shift to get the bits before shifting. */
if (use_alu->op == nir_op_ushr || use_alu->op == nir_op_ishr)
bits_used |= (def_bits_used << shift) & all_bits;
else
bits_used |= def_bits_used >> shift;
break;
} else if (src_idx == 1) {
bits_used |= use_alu->def.bit_size - 1;
break;
} else {
return all_bits;
}
case nir_op_iand:
case nir_op_ior:
assert(src_idx < 2);
if (nir_src_is_const(use_alu->src[1 - src_idx].src)) {
uint64_t other_src = nir_alu_src_as_uint(use_alu->src[1 - src_idx]);
if (use_alu->op == nir_op_iand)
bits_used |= ssa_def_bits_used(&use_alu->def, recur) & other_src;
else
bits_used |= ssa_def_bits_used(&use_alu->def, recur) & ~other_src;
break;
} else {
return all_bits;
}
case nir_op_ibfe:
case nir_op_ubfe:
if (src_idx == 0 && nir_src_is_const(use_alu->src[1].src)) {
uint64_t def_bits_used = ssa_def_bits_used(&use_alu->def, recur);
unsigned bit_size = use_alu->def.bit_size;
unsigned offset = nir_alu_src_as_uint(use_alu->src[1]) & (bit_size - 1);
unsigned bits = nir_src_is_const(use_alu->src[2].src) ?
nir_alu_src_as_uint(use_alu->src[2]) & (bit_size - 1) :
/* Worst case if bits is not constant. */
(bit_size - offset);
uint64_t field_bitmask = BITFIELD64_MASK(bits);
/* If one of the sign-extended bits is used, set the last src
* bit as used.
* If bits is not constant, all bits can be the last one.
*/
if (use_alu->op == nir_op_ibfe &&
(def_bits_used >> offset) & ~field_bitmask) {
if (nir_alu_src_as_uint(use_alu->src[2]))
def_bits_used |= BITFIELD64_BIT(bits - 1);
else
def_bits_used |= field_bitmask;
}
bits_used |= (field_bitmask & def_bits_used) << offset;
break;
} else if (src_idx == 1 || src_idx == 2) {
bits_used |= use_alu->src[0].src.ssa->bit_size - 1;
break;
} else {
return all_bits;
}
case nir_op_imul24:
case nir_op_umul24:
bits_used |= all_bits & 0xffffff;
break;
case nir_op_mov:
bits_used |= ssa_def_bits_used(&use_alu->def, recur);
break;
case nir_op_bcsel:
if (src_idx == 0)
bits_used |= 0x1;
else
bits_used |= ssa_def_bits_used(&use_alu->def, recur);
break;
default:
/* We don't know what this op does */
return all_bits;
}
break;
}
case nir_instr_type_intrinsic: {
nir_intrinsic_instr *use_intrin =
nir_instr_as_intrinsic(nir_src_parent_instr(src));
unsigned src_idx = src - use_intrin->src;
switch (use_intrin->intrinsic) {
case nir_intrinsic_read_invocation:
case nir_intrinsic_shuffle:
case nir_intrinsic_shuffle_up:
case nir_intrinsic_shuffle_down:
case nir_intrinsic_shuffle_xor:
case nir_intrinsic_quad_broadcast:
case nir_intrinsic_quad_swap_horizontal:
case nir_intrinsic_quad_swap_vertical:
case nir_intrinsic_quad_swap_diagonal:
if (src_idx == 0) {
bits_used |= ssa_def_bits_used(&use_intrin->def, recur);
} else {
if (use_intrin->intrinsic == nir_intrinsic_quad_broadcast) {
bits_used |= 3;
} else {
/* Subgroups larger than 128 are not a thing */
bits_used |= 127;
}
}
break;
case nir_intrinsic_reduce:
case nir_intrinsic_inclusive_scan:
case nir_intrinsic_exclusive_scan:
assert(src_idx == 0);
switch (nir_intrinsic_reduction_op(use_intrin)) {
case nir_op_iadd:
case nir_op_imul:
case nir_op_ior:
case nir_op_iand:
case nir_op_ixor:
bits_used |= ssa_def_bits_used(&use_intrin->def, recur);
break;
default:
return all_bits;
}
break;
default:
/* We don't know what this op does */
return all_bits;
}
break;
}
case nir_instr_type_phi: {
nir_phi_instr *use_phi = nir_instr_as_phi(nir_src_parent_instr(src));
bits_used |= ssa_def_bits_used(&use_phi->def, recur);
break;
}
default:
return all_bits;
}
/* If we've somehow shown that all our bits are used, we're done */
assert((bits_used & ~all_bits) == 0);
if (bits_used == all_bits)
return all_bits;
}
return bits_used;
}
uint64_t
nir_def_bits_used(const nir_def *def)
{
return ssa_def_bits_used(def, 2);
}
static void
get_alu_num_lsb(struct analysis_state *state, struct scalar_query q, uint32_t *result, const uint32_t *src)
{
nir_op op = nir_scalar_alu_op(q.scalar);
switch (op) {
case nir_op_ior:
case nir_op_ixor:
case nir_op_iadd:
case nir_op_iand:
case nir_op_imul:
if (!q.head.pushed_queries) {
push_scalar_query(state, nir_scalar_chase_alu_src(q.scalar, 0));
push_scalar_query(state, nir_scalar_chase_alu_src(q.scalar, 1));
return;
}
break;
case nir_op_ishl:
if (!q.head.pushed_queries) {
push_scalar_query(state, nir_scalar_chase_alu_src(q.scalar, 0));
return;
}
break;
case nir_op_ishr:
case nir_op_ushr:
if (!q.head.pushed_queries) {
if (nir_scalar_is_const(nir_scalar_chase_alu_src(q.scalar, 1)))
push_scalar_query(state, nir_scalar_chase_alu_src(q.scalar, 0));
return;
}
break;
case nir_op_bcsel:
if (!q.head.pushed_queries) {
push_scalar_query(state, nir_scalar_chase_alu_src(q.scalar, 1));
push_scalar_query(state, nir_scalar_chase_alu_src(q.scalar, 2));
return;
}
break;
default:
return;
}
switch (op) {
case nir_op_ior:
case nir_op_ixor:
case nir_op_iadd: {
*result = MIN2(src[0], src[1]);
break;
}
case nir_op_iand: {
*result = MAX2(src[0], src[1]);
break;
}
case nir_op_imul: {
*result = MIN2(src[0] + src[1], q.scalar.def->bit_size);
break;
}
case nir_op_ishl: {
nir_scalar src1 = nir_scalar_chase_alu_src(q.scalar, 1);
uint32_t mask = q.scalar.def->bit_size - 1;
unsigned amount = nir_scalar_is_const(src1) ? nir_scalar_as_uint(src1) & mask : 0;
*result = MIN2(src[0] + amount, q.scalar.def->bit_size);
break;
}
case nir_op_ishr:
case nir_op_ushr: {
nir_scalar src1 = nir_scalar_chase_alu_src(q.scalar, 1);
unsigned amount = nir_scalar_as_uint(src1) & (q.scalar.def->bit_size - 1);
*result = amount > src[0] ? 0 : src[0] - amount;
break;
}
case nir_op_bcsel: {
*result = MIN2(src[0], src[1]);
break;
}
default:
UNREACHABLE("Unknown opcode");
}
}
static void
process_num_lsb_query(struct analysis_state *state, struct analysis_query *aq, uint32_t *result,
const uint32_t *src)
{
struct scalar_query q = *(struct scalar_query *)aq;
*result = 0;
if (nir_scalar_is_const(q.scalar)) {
uint64_t val = nir_scalar_as_uint(q.scalar);
*result = val ? ffsll(val) - 1 : q.scalar.def->bit_size;
} else if (nir_scalar_is_alu(q.scalar)) {
get_alu_num_lsb(state, q, result, src);
}
}
unsigned
nir_def_num_lsb_zero(struct hash_table *numlsb_ht, nir_scalar def)
{
struct scalar_query query_alloc[16];
uint32_t result_alloc[16];
struct analysis_state state;
state.shader = NULL;
state.range_ht = numlsb_ht;
util_dynarray_init_from_stack(&state.query_stack, query_alloc, sizeof(query_alloc));
util_dynarray_init_from_stack(&state.result_stack, result_alloc, sizeof(result_alloc));
state.query_size = sizeof(struct scalar_query);
state.get_key = &get_scalar_key;
state.lookup = &scalar_lookup,
state.insert = &scalar_insert,
state.process_query = &process_num_lsb_query;
push_scalar_query(&state, def);
_mesa_hash_table_set_deleted_key(numlsb_ht, (void *)(uintptr_t)UINT32_MAX);
return perform_analysis(&state);
}
Reference in a new issue View git blame Copy permalink