Files
third_party_mesa3d/src/intel/compiler/brw_ir.h

189 lines
6.1 KiB
C
Raw Normal View History

/* -*- c++ -*- */
/*
* Copyright © 2010-2016 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.
*/
#ifndef BRW_IR_H
#define BRW_IR_H
#include <assert.h>
#include "brw_reg.h"
#include "compiler/glsl/list.h"
#define MAX_SAMPLER_MESSAGE_SIZE 11
#define MAX_VGRF_SIZE 16
#ifdef __cplusplus
struct backend_reg : private brw_reg
{
backend_reg() {}
backend_reg(const struct brw_reg &reg) : brw_reg(reg), offset(0) {}
const brw_reg &as_brw_reg() const
{
assert(file == ARF || file == FIXED_GRF || file == MRF || file == IMM);
assert(offset == 0);
return static_cast<const brw_reg &>(*this);
}
brw_reg &as_brw_reg()
{
assert(file == ARF || file == FIXED_GRF || file == MRF || file == IMM);
assert(offset == 0);
return static_cast<brw_reg &>(*this);
}
bool equals(const backend_reg &r) const;
bool negative_equals(const backend_reg &r) const;
bool is_zero() const;
bool is_one() const;
bool is_negative_one() const;
bool is_null() const;
bool is_accumulator() const;
/** Offset from the start of the (virtual) register in bytes. */
uint16_t offset;
using brw_reg::type;
using brw_reg::file;
using brw_reg::negate;
using brw_reg::abs;
using brw_reg::address_mode;
using brw_reg::subnr;
using brw_reg::nr;
using brw_reg::swizzle;
using brw_reg::writemask;
using brw_reg::indirect_offset;
using brw_reg::vstride;
using brw_reg::width;
using brw_reg::hstride;
using brw_reg::df;
using brw_reg::f;
using brw_reg::d;
using brw_reg::ud;
using brw_reg::d64;
using brw_reg::u64;
};
struct bblock_t;
struct backend_instruction : public exec_node {
bool is_3src(const struct intel_device_info *devinfo) const;
bool is_tex() const;
bool is_math() const;
bool is_control_flow() const;
bool is_commutative() const;
bool can_do_source_mods() const;
bool can_do_saturate() const;
bool can_do_cmod() const;
bool reads_accumulator_implicitly() const;
bool writes_accumulator_implicitly(const struct intel_device_info *devinfo) const;
intel/compiler: Relax some conditions in try_copy_propagate Previously can_do_source_mods was used to determine whether a value with a source modifier or a value from a scalar source (e.g., a uniform) could be copy propagated. The former is a superset of the latter, so this always produces correct results, but it is overly restrictive. For example, a BFI instruction can't have source modifiers, but it can have scalar sources. This was originally authored to prevent a small number of shader-db regressions in a commit that marked SHR has not being able to have source modifiers. That commit has since been dropped in favor of a different method. v2: Refactor register region restriction detection to a helper function. Suggested by Jason. No fossil-db changes on any Intel platform. All Gen7+ platforms had similar results. (Ice Lake shown) total instructions in shared programs: 20039111 -> 20038943 (<.01%) instructions in affected programs: 31736 -> 31568 (-0.53%) helped: 104 HURT: 0 helped stats (abs) min: 1 max: 9 x̄: 1.62 x̃: 1 helped stats (rel) min: 0.30% max: 0.88% x̄: 0.45% x̃: 0.42% 95% mean confidence interval for instructions value: -2.03 -1.20 95% mean confidence interval for instructions %-change: -0.47% -0.42% Instructions are helped. total cycles in shared programs: 980309750 -> 980308897 (<.01%) cycles in affected programs: 591078 -> 590225 (-0.14%) helped: 70 HURT: 26 helped stats (abs) min: 2 max: 622 x̄: 23.94 x̃: 4 helped stats (rel) min: <.01% max: 2.85% x̄: 0.33% x̃: 0.12% HURT stats (abs) min: 2 max: 520 x̄: 31.65 x̃: 6 HURT stats (rel) min: 0.02% max: 2.45% x̄: 0.34% x̃: 0.15% 95% mean confidence interval for cycles value: -26.41 8.64 95% mean confidence interval for cycles %-change: -0.27% -0.03% Inconclusive result (value mean confidence interval includes 0). No shader-db changes on earlier Intel platforms. Reviewed-by: Anuj Phogat anuj.phogat@gmail.com [v1] Reviewed-by: Jason Ekstrand <jason@jlekstrand.net> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9237>
2020-12-06 16:26:04 -08:00
/**
* Instructions that use indirect addressing have additional register
* regioning restrictions.
*/
bool uses_indirect_addressing() const;
void remove(bblock_t *block);
void insert_after(bblock_t *block, backend_instruction *inst);
void insert_before(bblock_t *block, backend_instruction *inst);
void insert_before(bblock_t *block, exec_list *list);
/**
* True if the instruction has side effects other than writing to
* its destination registers. You are expected not to reorder or
* optimize these out unless you know what you are doing.
*/
bool has_side_effects() const;
/**
* True if the instruction might be affected by side effects of other
* instructions.
*/
bool is_volatile() const;
#else
struct backend_instruction {
struct exec_node link;
#endif
/** @{
* Annotation for the generated IR. One of the two can be set.
*/
const void *ir;
const char *annotation;
/** @} */
/**
* Execution size of the instruction. This is used by the generator to
* generate the correct binary for the given instruction. Current valid
* values are 1, 4, 8, 16, 32.
*/
uint8_t exec_size;
/**
* Channel group from the hardware execution and predication mask that
* should be applied to the instruction. The subset of channel enable
* signals (calculated from the EU control flow and predication state)
* given by [group, group + exec_size) will be used to mask GRF writes and
* any other side effects of the instruction.
*/
uint8_t group;
uint32_t offset; /**< spill/unspill offset or texture offset bitfield */
uint8_t mlen; /**< SEND message length */
uint8_t ex_mlen; /**< SENDS extended message length */
int8_t base_mrf; /**< First MRF in the SEND message, if mlen is nonzero. */
uint8_t target; /**< MRT target. */
uint8_t sfid; /**< SFID for SEND instructions */
uint32_t desc; /**< SEND[S] message descriptor immediate */
uint32_t ex_desc; /**< SEND[S] extended message descriptor immediate */
unsigned size_written; /**< Data written to the destination register in bytes. */
enum opcode opcode; /* BRW_OPCODE_* or FS_OPCODE_* */
enum brw_conditional_mod conditional_mod; /**< BRW_CONDITIONAL_* */
enum brw_predicate predicate;
bool predicate_inverse:1;
bool writes_accumulator:1; /**< instruction implicitly writes accumulator */
bool force_writemask_all:1;
bool no_dd_clear:1;
bool no_dd_check:1;
bool saturate:1;
bool shadow_compare:1;
bool check_tdr:1; /**< Only valid for SEND; turns it into a SENDC */
bool send_has_side_effects:1; /**< Only valid for SHADER_OPCODE_SEND */
bool send_is_volatile:1; /**< Only valid for SHADER_OPCODE_SEND */
bool eot:1;
/* Chooses which flag subregister (f0.0 to f1.1) is used for conditional
* mod and predication.
*/
unsigned flag_subreg:2;
/** The number of hardware registers used for a message header. */
uint8_t header_size;
};
#endif