SChernykh 3c8c7ee097 Optimized dataset read (#211) ...

* Optimized dataset read

There was a false dependency on readReg2 and readReg3 (caused by `xor rbp, rax` instruction) when reading dataset item (see design.md - 4.6.2 Loop execution, steps 5 and 7). This change uses `ma` register to read dataset item before the whole `rbp` (`ma` and `mx`) is changed, so superscalar and out-of-order CPU can start executing it earlier.

Results: https://i.imgur.com/Bpeq9mx.png

~1% speedup on modern Intel/AMD CPUs.

* ARMv8: optimized dataset read

Break dependency from readReg2 and readReg3.

* Fixed light mode hashing

2021-05-22 13:54:50 +02:00

..

asm

Optimized dataset read (#211)

2021-05-22 13:54:50 +02:00

blake2

Fix symbol collisions with blake2b (#145)

2019-10-30 20:09:27 +01:00

tests

add --noBatch benchmark option

2020-07-04 14:57:56 +02:00

aes_hash.cpp

Combined hash and fill AES loop (#166)

2019-12-01 16:58:38 +01:00

aes_hash.hpp

Combined hash and fill AES loop (#166)

2019-12-01 16:58:38 +01:00

allocator.cpp

Fix inconsistent class/struct usage

2019-11-19 23:17:55 +01:00

allocator.hpp

Relicensed under the 3-clause BSD license

2019-05-18 14:21:47 +02:00

argon2_avx2.c

Optimized Argon2 (SSSE3/AVX2)

2019-10-06 18:07:23 +02:00

argon2_core.c

Fix symbol collisions with blake2b (#145)

2019-10-30 20:09:27 +01:00

argon2_core.h

Optimized Argon2 (SSSE3/AVX2)

2019-10-06 18:07:23 +02:00

argon2_ref.c

Optimized Argon2 (SSSE3/AVX2)

2019-10-06 18:07:23 +02:00

argon2_ssse3.c

use SSSE3 consistently as opposed to SSE3

2019-10-06 23:46:49 +02:00

argon2.h

use SSSE3 consistently as opposed to SSE3

2019-10-06 23:46:49 +02:00

assembly_generator_x86.cpp

Refactoring (#95)

2019-07-03 18:13:20 +02:00

assembly_generator_x86.hpp

Use 'dst' as the CBRANCH condition register

2019-05-21 08:37:36 +02:00

blake2_generator.cpp

Refactoring (#95)

2019-07-03 18:13:20 +02:00

blake2_generator.hpp

Refactoring (#95)

2019-07-03 18:13:20 +02:00

bytecode_machine.cpp

Refactoring (#95)

2019-07-03 18:13:20 +02:00

bytecode_machine.hpp

Regression tests (#73)

2019-06-22 15:54:43 +02:00

common.hpp

JIT compiler for ARMv8 (#125)

2019-09-22 21:06:22 +02:00

configuration.h

Increase the frequency of CBRANCH (#118)

2019-08-30 09:28:18 +02:00

cpu.cpp

Apple silicon: force W^X, enable hardware AES

2020-11-29 20:39:53 +01:00

cpu.hpp

Automatic detection of CPU capabilities

2019-10-08 23:09:35 +02:00

dataset.cpp

Optimized Argon2 (SSSE3/AVX2)

2019-10-06 18:07:23 +02:00

dataset.hpp

Fixes for cmake build with visual studio (#144)

2019-11-22 18:24:16 +01:00

instruction_weights.hpp

Regression tests (#73)

2019-06-22 15:54:43 +02:00

instruction.cpp

Code generator fixups

2019-06-23 23:10:29 +02:00

instruction.hpp

Use strongly typed enums (#55)

2019-06-10 16:02:25 +02:00

instructions_portable.cpp

fix test 92 not failing properly on GCC/amd64

2020-05-06 13:48:53 +02:00

intrin_portable.h

fix test 92 not failing properly on GCC/amd64

2020-05-06 13:48:53 +02:00

jit_compiler_a64_static.hpp

JIT compiler for ARMv8 (#125)

2019-09-22 21:06:22 +02:00

jit_compiler_a64_static.S

Optimized dataset read (#211)

2021-05-22 13:54:50 +02:00

jit_compiler_a64.cpp

Fix illegal instruction crash on some ARM systems

2021-02-01 23:19:14 +01:00

jit_compiler_a64.hpp

Fix inconsistent class/struct usage

2019-11-19 23:17:55 +01:00

jit_compiler_fallback.hpp

Fix inconsistent class/struct usage

2019-11-19 23:17:55 +01:00

jit_compiler_x86_static.asm

remove unnecessary first-load initialization code

2021-01-23 14:56:35 -08:00

jit_compiler_x86_static.hpp

remove unnecessary first-load initialization code

2021-01-23 14:56:35 -08:00

jit_compiler_x86_static.S

remove unnecessary first-load initialization code

2021-01-23 14:56:35 -08:00

jit_compiler_x86.cpp

remove unnecessary first-load initialization code

2021-01-23 14:56:35 -08:00

jit_compiler_x86.hpp

Fix inconsistent class/struct usage

2019-11-19 23:17:55 +01:00

jit_compiler.hpp

Apple silicon: force W^X, enable hardware AES

2020-11-29 20:39:53 +01:00

program.hpp

Relicensed under the 3-clause BSD license

2019-05-18 14:21:47 +02:00

randomx.cpp

Merge pull request #187 from tevador/pr-netbsd

2020-06-28 16:35:19 +02:00

randomx.h

Preserve floating point state when calling randomx_calculate_hash

2020-05-06 12:42:30 +02:00

reciprocal.c

Sanity checks (#88)

2019-06-29 18:53:49 +02:00

reciprocal.h

Regression tests (#73)

2019-06-22 15:54:43 +02:00

soft_aes.cpp

Relicensed under the 3-clause BSD license

2019-05-18 14:21:47 +02:00

soft_aes.h

Relicensed under the 3-clause BSD license

2019-05-18 14:21:47 +02:00

superscalar_program.hpp

Sanity checks (#88)

2019-06-29 18:53:49 +02:00

superscalar.cpp

Fix a possible out-of-bounds access in superscalar generator

2019-10-11 11:31:05 +02:00

superscalar.hpp

Fix header dependency of superscalar_program.hpp

2019-06-24 13:58:41 +02:00

virtual_machine.cpp

Combined hash and fill AES loop (#166)

2019-12-01 16:58:38 +01:00

virtual_machine.hpp

fix potential use-after-free when reallocating cache

2020-06-27 20:21:06 +02:00

virtual_memory.cpp

Faster W^X policy for apple silicon macs

2021-05-20 20:35:18 +01:00

virtual_memory.hpp

Optional W^X policy for JIT pages (#112)

2019-08-25 13:47:40 +02:00

vm_compiled_light.cpp

Optional W^X policy for JIT pages (#112)

2019-08-25 13:47:40 +02:00

vm_compiled_light.hpp

Optional W^X policy for JIT pages (#112)

2019-08-25 13:47:40 +02:00

vm_compiled.cpp

JIT compiler for ARMv8 (#125)

2019-09-22 21:06:22 +02:00

vm_compiled.hpp

Optional W^X policy for JIT pages (#112)

2019-08-25 13:47:40 +02:00

vm_interpreted_light.cpp

Support Dataset size larger than 4 GiB

2019-05-29 17:27:49 +02:00

vm_interpreted_light.hpp

Add Dataset prefetch in interpreted VM (#52)

2019-06-10 16:00:04 +02:00

vm_interpreted.cpp

Regression tests (#73)

2019-06-22 15:54:43 +02:00

vm_interpreted.hpp

Regression tests (#73)

2019-06-22 15:54:43 +02:00