mesa/src/intel/compiler/test_eu_compact.cpp

352 lines
9.7 KiB
C++
Raw Normal View History

/*
* Copyright © 2012 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 <stdlib.h>
#include <stdio.h>
#include <stdbool.h>
#include "util/ralloc.h"
#include "brw_eu.h"
#include <gtest/gtest.h>
struct CompactParams {
unsigned verx10;
unsigned align;
};
std::string
get_compact_params_name(const testing::TestParamInfo<CompactParams> p)
{
CompactParams params = p.param;
std::stringstream ss;
ss << params.verx10 << "_";
switch (params.align) {
case BRW_ALIGN_1:
ss << "Align_1";
break;
case BRW_ALIGN_16:
ss << "Align_16";
break;
default:
unreachable("invalid align");
}
return ss.str();
}
static bool
test_compact_instruction(struct brw_codegen *p, brw_inst src)
{
brw_compact_inst dst;
memset(&dst, 0xd0, sizeof(dst));
if (brw_try_compact_instruction(p->isa, &dst, &src)) {
brw_inst uncompacted;
brw_uncompact_instruction(p->isa, &uncompacted, &dst);
if (memcmp(&uncompacted, &src, sizeof(src))) {
brw_debug_compact_uncompact(p->isa, &src, &uncompacted);
return false;
}
} else {
brw_compact_inst unchanged;
memset(&unchanged, 0xd0, sizeof(unchanged));
/* It's not supposed to change dst unless it compacted. */
if (memcmp(&unchanged, &dst, sizeof(dst))) {
fprintf(stderr, "Failed to compact, but dst changed\n");
fprintf(stderr, " Instruction: ");
brw_disassemble_inst(stderr, p->isa, &src, false, 0, NULL);
return false;
}
}
return true;
}
/**
* When doing fuzz testing, pad bits won't round-trip.
*
* This sort of a superset of skip_bit, which is testing for changing bits that
* aren't worth testing for fuzzing. We also just want to clear bits that
* become meaningless once fuzzing twiddles a related bit.
*/
static void
clear_pad_bits(const struct brw_isa_info *isa, brw_inst *inst)
{
const struct intel_device_info *devinfo = isa->devinfo;
if (brw_inst_opcode(isa, inst) != BRW_OPCODE_SEND &&
brw_inst_opcode(isa, inst) != BRW_OPCODE_SENDC &&
brw_inst_src0_reg_file(devinfo, inst) != BRW_IMMEDIATE_VALUE &&
brw_inst_src1_reg_file(devinfo, inst) != BRW_IMMEDIATE_VALUE) {
brw_inst_set_bits(inst, 127, 111, 0);
}
if (devinfo->ver == 8 && devinfo->platform != INTEL_PLATFORM_CHV &&
is_3src(isa, brw_inst_opcode(isa, inst))) {
brw_inst_set_bits(inst, 105, 105, 0);
brw_inst_set_bits(inst, 84, 84, 0);
brw_inst_set_bits(inst, 36, 35, 0);
}
}
static bool
skip_bit(const struct brw_isa_info *isa, brw_inst *src, int bit)
{
const struct intel_device_info *devinfo = isa->devinfo;
/* pad bit */
if (bit == 7)
return true;
/* The compact bit -- uncompacted can't have it set. */
if (bit == 29)
return true;
if (is_3src(isa, brw_inst_opcode(isa, src))) {
if (devinfo->ver >= 9 || devinfo->platform == INTEL_PLATFORM_CHV) {
if (bit == 127)
return true;
} else {
if (bit >= 126 && bit <= 127)
return true;
if (bit == 105)
return true;
if (bit == 84)
return true;
if (bit >= 35 && bit <= 36)
return true;
}
} else {
if (bit == 47)
return true;
if (devinfo->ver >= 8) {
if (bit == 11)
return true;
if (bit == 95)
return true;
} else {
if (devinfo->ver < 7 && bit == 90)
return true;
if (bit >= 91 && bit <= 95)
return true;
}
}
/* sometimes these are pad bits. */
if (brw_inst_opcode(isa, src) != BRW_OPCODE_SEND &&
brw_inst_opcode(isa, src) != BRW_OPCODE_SENDC &&
brw_inst_src0_reg_file(devinfo, src) != BRW_IMMEDIATE_VALUE &&
brw_inst_src1_reg_file(devinfo, src) != BRW_IMMEDIATE_VALUE &&
bit >= 121) {
return true;
}
return false;
}
static bool
test_fuzz_compact_instruction(struct brw_codegen *p, brw_inst src)
{
for (int bit0 = 0; bit0 < 128; bit0++) {
if (skip_bit(p->isa, &src, bit0))
continue;
for (int bit1 = 0; bit1 < 128; bit1++) {
brw_inst instr = src;
uint64_t *bits = instr.data;
if (skip_bit(p->isa, &src, bit1))
continue;
bits[bit0 / 64] ^= (1ull << (bit0 & 63));
bits[bit1 / 64] ^= (1ull << (bit1 & 63));
clear_pad_bits(p->isa, &instr);
intel/eu: Handle compaction when inserting validation errors When the EU validator encountered an error, it would add an annotation to the disassembly. Unfortunately, the code to insert an error assumed that the next instruction would start at (offset + sizeof(brw_inst)), which is not true if the instruction with an error is compacted. This could lead to cascading disassembly errors, where we started trying to decode the next instruction at the wrong offset, and getting lots of scary looking output: ERROR: Register Regioning patterns where [...] (-f0.1.any16h) illegal(*** invalid execution size value 6 ) { align1 $7.src atomic }; (+f0.1.any16h) illegal.sat(*** invalid execution size value 6 ) { align1 $9.src AccWrEnable }; illegal(*** invalid execution size value 6 ) { align1 $11.src }; (+f0.1) illegal.sat(*** invalid execution size value 6 ) { align1 F@2 AccWrEnable }; (+f0.1) illegal.sat(*** invalid execution size value 6 ) { align1 F@2 AccWrEnable }; (+f0.1) illegal.sat(*** invalid execution size value 6 ) { align1 $15.src AccWrEnable }; illegal(*** invalid execution size value 6 ) { align1 $15.src }; (+f0.1) illegal.sat.g.f0.1(*** invalid execution size value 6 ) { align1 $13.src AccWrEnable }; Only the first instruction was actually wrong - the rest are just a result of starting the disassembler at the wrong offset. Trash ensues! To fix this, just pass the instruction size in a few layers so we can record the next offset properly. Cc: mesa-stable Reviewed-by: Francisco Jerez <currojerez@riseup.net> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17624>
2022-07-19 08:27:29 +01:00
if (!brw_validate_instruction(p->isa, &instr, 0, sizeof(brw_inst), NULL))
continue;
if (!test_compact_instruction(p, instr)) {
printf(" twiddled bits for fuzzing %d, %d\n", bit0, bit1);
return false;
}
}
}
return true;
}
class CompactTestFixture : public testing::TestWithParam<CompactParams> {
protected:
virtual void SetUp() {
CompactParams params = GetParam();
mem_ctx = ralloc_context(NULL);
devinfo = rzalloc(mem_ctx, intel_device_info);
p = rzalloc(mem_ctx, brw_codegen);
devinfo->verx10 = params.verx10;
devinfo->ver = devinfo->verx10 / 10;
brw_init_isa_info(&isa, devinfo);
brw_init_codegen(&isa, p, p);
brw_set_default_predicate_control(p, BRW_PREDICATE_NONE);
brw_set_default_access_mode(p, params.align);
};
virtual void TearDown() {
EXPECT_EQ(p->nr_insn, 1);
EXPECT_TRUE(test_compact_instruction(p, p->store[0]));
EXPECT_TRUE(test_fuzz_compact_instruction(p, p->store[0]));
ralloc_free(mem_ctx);
};
void *mem_ctx;
struct brw_isa_info isa;
intel_device_info *devinfo;
brw_codegen *p;
};
class Instructions : public CompactTestFixture {};
INSTANTIATE_TEST_CASE_P(
CompactTest,
Instructions,
testing::Values(
CompactParams{ 50, BRW_ALIGN_1 }, CompactParams{ 50, BRW_ALIGN_16 },
CompactParams{ 60, BRW_ALIGN_1 }, CompactParams{ 60, BRW_ALIGN_16 },
CompactParams{ 70, BRW_ALIGN_1 }, CompactParams{ 70, BRW_ALIGN_16 },
CompactParams{ 75, BRW_ALIGN_1 }, CompactParams{ 75, BRW_ALIGN_16 },
CompactParams{ 80, BRW_ALIGN_1 }, CompactParams{ 80, BRW_ALIGN_16 },
CompactParams{ 90, BRW_ALIGN_1 }, CompactParams{ 90, BRW_ALIGN_16 },
CompactParams{ 110, BRW_ALIGN_1 },
CompactParams{ 120, BRW_ALIGN_1 },
CompactParams{ 125, BRW_ALIGN_1 }
),
get_compact_params_name);
class InstructionsBeforeIvyBridge : public CompactTestFixture {};
INSTANTIATE_TEST_CASE_P(
CompactTest,
InstructionsBeforeIvyBridge,
testing::Values(
CompactParams{ 50, BRW_ALIGN_1 }, CompactParams{ 50, BRW_ALIGN_16 },
CompactParams{ 60, BRW_ALIGN_1 }, CompactParams{ 60, BRW_ALIGN_16 }
),
get_compact_params_name);
TEST_P(Instructions, ADD_GRF_GRF_GRF)
{
struct brw_reg g0 = brw_vec8_grf(0, 0);
struct brw_reg g2 = brw_vec8_grf(2, 0);
struct brw_reg g4 = brw_vec8_grf(4, 0);
brw_ADD(p, g0, g2, g4);
}
TEST_P(Instructions, ADD_GRF_GRF_IMM)
{
struct brw_reg g0 = brw_vec8_grf(0, 0);
struct brw_reg g2 = brw_vec8_grf(2, 0);
brw_ADD(p, g0, g2, brw_imm_f(1.0));
}
TEST_P(Instructions, ADD_GRF_GRF_IMM_d)
{
struct brw_reg g0 = retype(brw_vec8_grf(0, 0), BRW_REGISTER_TYPE_D);
struct brw_reg g2 = retype(brw_vec8_grf(2, 0), BRW_REGISTER_TYPE_D);
brw_ADD(p, g0, g2, brw_imm_d(1));
}
TEST_P(Instructions, MOV_GRF_GRF)
{
struct brw_reg g0 = brw_vec8_grf(0, 0);
struct brw_reg g2 = brw_vec8_grf(2, 0);
brw_MOV(p, g0, g2);
}
TEST_P(InstructionsBeforeIvyBridge, ADD_MRF_GRF_GRF)
{
struct brw_reg m6 = brw_vec8_reg(BRW_MESSAGE_REGISTER_FILE, 6, 0);
struct brw_reg g2 = brw_vec8_grf(2, 0);
struct brw_reg g4 = brw_vec8_grf(4, 0);
brw_ADD(p, m6, g2, g4);
}
TEST_P(Instructions, ADD_vec1_GRF_GRF_GRF)
{
struct brw_reg g0 = brw_vec1_grf(0, 0);
struct brw_reg g2 = brw_vec1_grf(2, 0);
struct brw_reg g4 = brw_vec1_grf(4, 0);
brw_ADD(p, g0, g2, g4);
}
TEST_P(InstructionsBeforeIvyBridge, PLN_MRF_GRF_GRF)
{
struct brw_reg m6 = brw_vec8_reg(BRW_MESSAGE_REGISTER_FILE, 6, 0);
struct brw_reg interp = brw_vec1_grf(2, 0);
struct brw_reg g4 = brw_vec8_grf(4, 0);
brw_PLN(p, m6, interp, g4);
}
TEST_P(Instructions, f0_0_MOV_GRF_GRF)
{
struct brw_reg g0 = brw_vec8_grf(0, 0);
struct brw_reg g2 = brw_vec8_grf(2, 0);
brw_push_insn_state(p);
brw_set_default_predicate_control(p, BRW_PREDICATE_NORMAL);
brw_MOV(p, g0, g2);
brw_pop_insn_state(p);
}
/* The handling of f0.1 vs f0.0 changes between gfx6 and gfx7. Explicitly test
* it, so that we run the fuzzing can run over all the other bits that might
* interact with it.
*/
TEST_P(Instructions, f0_1_MOV_GRF_GRF)
{
struct brw_reg g0 = brw_vec8_grf(0, 0);
struct brw_reg g2 = brw_vec8_grf(2, 0);
brw_push_insn_state(p);
brw_set_default_predicate_control(p, BRW_PREDICATE_NORMAL);
brw_inst *mov = brw_MOV(p, g0, g2);
brw_inst_set_flag_subreg_nr(p->devinfo, mov, 1);
brw_pop_insn_state(p);
}