tevador
|
a586751f6b
|
Removed FPNEG instruction
Optimized instruction frequencies
Increased the range for A registers from [1,65536) to [1, 4294967296)
|
5 years ago |
tevador
|
ac4462ad42
|
Renamed floating point instructions
Fixed negative source operand for FMUL_M and FDIV_M
|
5 years ago |
tevador
|
b417fd08ea
|
16 -> 8 chained programs
constant address loads are always from L3
|
5 years ago |
tevador
|
1ee94bef2a
|
Added ISWAP instruction
Scratchpad -> 2 MiB
New scratchpad initialization
New dataset initialization
|
5 years ago |
tevador
|
ab859879a2
|
loop body = 128 instructions
|
5 years ago |
tevador
|
20eb549725
|
Merged load/store of integer and FP registers
|
5 years ago |
tevador
|
8f2abd6c05
|
NOP instruction
register load/store from L3
|
5 years ago |
tevador
|
005c67f64c
|
Added explicit STORE instructions
JIT compiler
|
5 years ago |
tevador
|
7c049cce8d
|
Added store instructions
|
5 years ago |
tevador
|
5b7df0c5e1
|
Test ASM for a new program structure
|
5 years ago |
tevador
|
d2cb086221
|
ASM code generator for "small" programs that fit into the uOP cache
|
5 years ago |
tevador
|
bd0dba88a8
|
4 scratchpad segments
|
5 years ago |
tevador
|
16db607025
|
Scratchpad size increased to 1 MiB
New AES-based scratchpad hashing function
|
5 years ago |
tevador
|
93c324709b
|
Related to previous changes
|
5 years ago |
tevador
|
89bc68d093
|
Memory-bound dataset initialization
|
5 years ago |
tevador
|
4fb168e249
|
Large page support for cache
Bug fixes
|
5 years ago |
tevador
|
8b1102ee05
|
Interpreter + async mode
|
5 years ago |
tevador
|
c2e0122e15
|
Comparison with CryptoNight
|
6 years ago |
tevador
|
a7ffe8c19a
|
Mix dataset cacheline with registers r0-r7
|
6 years ago |
tevador
|
48d85643de
|
Dataset intialization algorithm (AES)
|
6 years ago |
tevador
|
67e741ff22
|
Reduced x86 code size by 512 bytes (and ecx -> and eax)
|
6 years ago |
tevador
|
1426fcbab5
|
Print average program code size
Fixed assembly for MUL_64 and IMUL_32
Division weight 4 -> 8
|
6 years ago |
tevador
|
2756bcdcfe
|
Added magic division to JIT compiler
New B operand selection rules
|
6 years ago |
tevador
|
451dfc5730
|
Optimized division by constants
|
6 years ago |
tevador
|
c02ee4291d
|
FPROUND - variable flag offset
|
6 years ago |
tevador
|
e487092f07
|
Simplified CALL and RET
|
6 years ago |
tevador
|
557241cd95
|
JUMP instruction
|
6 years ago |
tevador
|
6941b2cb69
|
Reworked instruction set documentation
|
6 years ago |
tevador
|
d1a808643d
|
Random accesses - JIT compiler
|
6 years ago |
tevador
|
b71e0eec65
|
Optimizations to reduce code size under 32K
|
6 years ago |
tevador
|
b6d654291f
|
90 address transformations
|
6 years ago |
tevador
|
2f6a599ff6
|
Inlined calls for memory read
|
6 years ago |
tevador
|
6519fed4d1
|
Combined prefetch + read into a single step
|
6 years ago |
tevador
|
4189e4ebc6
|
Original number of VM instructions
|
6 years ago |
tevador
|
619bee5418
|
Random dataset accesses - asm only
Initial support for large pages
|
6 years ago |
tevador
|
bf8397b08d
|
Updated specification
|
6 years ago |
tevador
|
3caecc7646
|
Vector FPU instructions
JitCompilerX86 - static code written in asm
Updated ALU/FPU tests
Updated instruction weights
|
6 years ago |
tevador
|
a09bee8d60
|
js -> jz to enable macro-op fusion on Intel CPUs (~1% speed-up)
|
6 years ago |
tevador
|
76b6b05cf2
|
Unconditional RET
|
6 years ago |
tevador
|
39c569ae44
|
Fixed a potential crash in JitCompilerX86
|
6 years ago |
tevador
|
5bc26348f1
|
Updated readme with performance data
Added --help option
|
6 years ago |
tevador
|
03913d0e81
|
Run a single thread synchronously
|
6 years ago |
tevador
|
c05947db09
|
Bug fixes
|
6 years ago |
tevador
|
ca59925495
|
JitCompilerX86: use mmap to allocate an executable buffer
compile as c++11
|
6 years ago |
tevador
|
740c40b218
|
8 branch conditions for CALL/RET
|
6 years ago |
tevador
|
55afe9646f
|
Debuggable assembly generator
|
6 years ago |
tevador
|
fce6e75689
|
Fixed copyright notice
|
6 years ago |
tevador
|
ffa67295c4
|
Instruction statistics
|
6 years ago |
tevador
|
1db7dd6e8b
|
Renamed immediate constants
|
6 years ago |
tevador
|
b9d2d853aa
|
Support for multiple threads
|
6 years ago |