yaxpeax-x86/src/protected_mode/evex.rs, branch 2.1.1 yaxpeax x86 decoder write `apply_disp_scale` in a mode-agnostic way 2021-12-17T19:03:37+00:00 iximeow me@iximeow.net 2021-12-17T19:03:37+00:00 060fb29180f354e04ec8d48f0128433b634fa3c5 `apply_disp_scale` forgot that `wrapping_mul` exists, so we don't need to explicitly write the size of value that `mem_size` should be cast to, in casting to/from a signed integer. taken with `.into()`, we don't need per-architecture stubs to make evex decoding work. 
`apply_disp_scale` forgot that `wrapping_mul` exists, so we don't need
to explicitly write the size of value that `mem_size` should be cast to,
in casting to/from a signed integer. taken with `.into()`, we don't need
per-architecture stubs to make evex decoding work.
displacements are stored as unsigned, but are functionally signed ints 2021-12-17T04:01:26+00:00 iximeow me@iximeow.net 2021-12-17T03:58:15+00:00 f063f747c21548329170fd09fd09d391b2277aed so multiplying to expand EVEX compressed offsets can overflow, and that needs to be okay. 
so multiplying to expand EVEX compressed offsets can overflow, and that
needs to be okay.
reallocate OperandCode, convert disparate registers to array 2021-07-02T06:54:06+00:00 iximeow me@iximeow.net 2021-07-02T06:54:06+00:00 a781322552d9fb52b7b5e51641f49f12678f682f also remove redundant assignments of operand_count and some OperandSpec, bulk-assign all registers and operands on entry to `read_instr`. this all, taken together, shaves off about 7 cycles per decode. 
also remove redundant assignments of operand_count and some OperandSpec,
bulk-assign all registers and operands on entry to `read_instr`. this
all, taken together, shaves off about 7 cycles per decode.
protected-mode avx512 2021-06-27T21:57:03+00:00 iximeow me@iximeow.net 2021-06-27T21:57:03+00:00 c42f84b37c9be599442a44caab289f5fdf971649