 tevador
|
8f2abd6c05
NOP instruction
|
7 лет назад |
|
tevador
|
005c67f64c
Added explicit STORE instructions
|
7 лет назад |
|
tevador
|
d2cb086221
ASM code generator for "small" programs that fit into the uOP cache
|
7 лет назад |
|
tevador
|
bd0dba88a8
4 scratchpad segments
|
7 лет назад |
|
tevador
|
16db607025
Scratchpad size increased to 1 MiB
|
7 лет назад |
|
tevador
|
93c324709b
Related to previous changes
|
7 лет назад |
|
tevador
|
89bc68d093
Memory-bound dataset initialization
|
7 лет назад |
|
tevador
|
4fb168e249
Large page support for cache
|
7 лет назад |
|
tevador
|
8b1102ee05
Interpreter + async mode
|
7 лет назад |
|
tevador
|
a7ffe8c19a
Mix dataset cacheline with registers r0-r7
|
7 лет назад |
|
tevador
|
48d85643de
Dataset intialization algorithm (AES)
|
7 лет назад |
|
tevador
|
67e741ff22
Reduced x86 code size by 512 bytes (and ecx -> and eax)
|
7 лет назад |
|
tevador
|
1426fcbab5
Print average program code size
|
7 лет назад |
|
tevador
|
2756bcdcfe
Added magic division to JIT compiler
|
7 лет назад |
|
tevador
|
451dfc5730
Optimized division by constants
|
7 лет назад |
|
tevador
|
c02ee4291d
FPROUND - variable flag offset
|
7 лет назад |
|
tevador
|
e487092f07
Simplified CALL and RET
|
7 лет назад |
|
tevador
|
557241cd95
JUMP instruction
|
7 лет назад |
|
tevador
|
6941b2cb69
Reworked instruction set documentation
|
7 лет назад |
|
tevador
|
d1a808643d
Random accesses - JIT compiler
|
7 лет назад |
|
tevador
|
b71e0eec65
Optimizations to reduce code size under 32K
|
7 лет назад |
|
tevador
|
b6d654291f
90 address transformations
|
7 лет назад |
|
tevador
|
2f6a599ff6
Inlined calls for memory read
|
7 лет назад |
|
tevador
|
6519fed4d1
Combined prefetch + read into a single step
|
7 лет назад |
|
tevador
|
4189e4ebc6
Original number of VM instructions
|
7 лет назад |
|
tevador
|
619bee5418
Random dataset accesses - asm only
|
7 лет назад |
|
tevador
|
bf8397b08d
Updated specification
|
7 лет назад |
|
tevador
|
3caecc7646
Vector FPU instructions
|
7 лет назад |
|
tevador
|
a09bee8d60
js -> jz to enable macro-op fusion on Intel CPUs (~1% speed-up)
|
7 лет назад |
|
tevador
|
76b6b05cf2
Unconditional RET
|
7 лет назад |
