aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2022-12-03roll up decoding loop changes for 16-bit and 32-bit decodersiximeow
this applies * f338c74656f6eef8b3080fa9f249b1cb733fd1a9 * bece19e6a69b158893abbf56a6cac25eb25d9a32 * 6353f58170d28a142e3b012c2c86f684d50dea45 * 67be1c0983244645a3c762b7aa0601f0d0ba4bb3 * 091f1d66ef853d6339a96e43d71c137ee7d3907a as one unit to both the 16-bit and 32-bit decoders.
2022-12-03apply e7f49509 to 16-bit and 32-bit decodersiximeow
2022-12-03apply 2444de11 to 16-bit and 32-bit decodersiximeow
these don't need the extra `rex`-supporting index space, so they don't have it.
2022-12-03fix incorrect rex selection and field description offsetsiximeow
2022-12-0366 prefixes are common, 0f opcodes are commoniximeow
2022-12-03support a fast path through the decoder for [rex-prefixed]opcode instsiximeow
the overwhelming majority of x86 instructions are either a single-byte opcode or a single-byte opcode with a rex prefix. supporting these specially means that we don't have to length-check on every byte or go through the full decode loop while reading the most likely instructions. this is a significant improvement on typical x86 streams, but comes at a moderate penalty for crafted x86 instructions. the penalty is still not very bad, as the fast path is exited in favor of the full decode loop as soon as we see a non-rex prefix byte; this adds maybe a dozen instructions to the slow path.
2022-12-03just a bit more code motion that seemed to help things sometimesiximeow
2022-12-03reorder prefix checks, extract vex/evex prefix handlingiximeow
sharing vex/evex invalid prefix checks improves codegen a bit, but ordering prefix checks by likeliest prefix first reduces time falling through prefix handling arms. both together are a notable improvement in throughput on typical x86 code. bundled in here is some code motion to where `mem_size = 0` and `operand_count = 2` are executed; this is because, at least on zen2 and cascade lake parts, bunching all stores to the instruction together caused small stalls getting into the decoder. spreading out stores seems to mix these assignments with parts of code that was not using memory anyway, and pipelines better.
2022-12-03move opcode lookup tables into const arraysiximeow
cleanliness, but also slightly better codegen somehow?
2022-12-03replace size lookup logic with a LUTiximeow
the match compiled into some indirect branch awfulness!! no thank you
2022-09-23Fix some typos.Bruce Mitchener
2022-05-30pshufb annotations use incorrect register banks (for now?)iximeow
the correct bank is applied far after register numbers are read. a correct annotation would need to know to defer emission until setting register banks, but also would need to work backwards for the number of bits between the current byte and modrm. not impossible, but substantial refactoring.
2022-05-07more annotation fixes?iximeow
2022-05-01add testing setup for field descriptionsiximeow
2022-04-30support 0x9a callf in 16/32-bit modesiximeow
2022-04-24fix a few issues preventing no-std builds from ... buildingiximeow
this includes a `Makefile` that exercises the various crate configs. most annoyingly, several doc comments needed to grow `#[cfg(feature="fmt")]` blocks so docs continue to build with that feature enabled or disabled. carved out a way to run exhaustive tests; they should be written as `#[ignore]`, and then the makefile will run even ignored tests on the expectation that this will run the exhaustive (but slower) suite. exhaustive tests are not yet written. they'll probably involve spanning 4 byte sequences from 0 to 2^32-1.
2022-01-12fuzz DisplayStyle::C and fix corresponding issues1.1.4iximeow
2022-01-02update changelogiximeow
2022-01-02fix incorrect decoder used in docs testiximeow
2022-01-02actually include a linkiximeow
2022-01-02explicit inline annotations for kinda_uncheckedsiximeow
unfortunately something about the wrapper functions adjusts codegen even when the wrapper functions themselves are just calls to inner functions. the in-tree benchmark (known to not be comprehensive, but enough to spot a difference), showed a ~3.5% regression in throughput with the prior commit, even though it doesn't change behavior at all. explicit #[inline(always)] gets things to a state where the wrapper functions do not penalize performance. for an example of the differences in codegen, see below. before: ``` < 141d4: 48 39 fa cmp %rdi,%rdx < 141d7: 0f 84 0b 4d 00 00 je 18ee8 <_ZN5bench16do_decode_swathe17h694154735739ce4cE+0x4e58> < 141dd: 0f b6 0f movzbl (%rdi),%ecx < 141e0: 48 83 c7 01 add $0x1,%rdi < 141e4: 48 89 7c 24 38 mov %rdi,0x38(%rsp) ... snip ... ``` after: ``` > 141d4: 48 39 ea cmp %rbp,%rdx > 141d7: 0f 84 97 4c 00 00 je 18e74 <_ZN5bench16do_decode_swathe17h694154735739ce4cE+0x4de4> > 141dd: 0f b6 4d 00 movzbl 0x0(%rbp),%ecx > 141e1: 48 83 c5 01 add $0x1,%rbp > 141e5: 48 89 6c 24 38 mov %rbp,0x38(%rsp) ... snip ... ``` there are several spans of code with this kind of change involved; there are no explicit calls to `get_kinda_unchecked` or `unreachable_kinda_unchecked` but clearly a difference did make it through to the benchmark's code. while the choice of `rbp` instead of `rdi` wouldn't seem very interesting, the instructions themselves are more substantially different. `0fb60f` vs `0fb64d00`; to encode `[rbp + 0]`, the instruction requires a displacement, and is one byte longer as a result. there are several instructions so-impacted, and i suspect the increased code size is what ended up changing benchmark behavior. after adding these `#[inline(always)]` annotations, there is no difference in generated code with or without the `kinda_unchecked` helpers!
2022-01-02Wrap unsafe functions to catch errors in debug5225225
Closes https://github.com/iximeow/yaxpeax-x86/issues/16
2021-12-19prep for 1.1.3 releaseiximeow
actual release is being held until cargo fuzz runs a while without a panic
2021-12-19add in-tree cargo fuzz targets for decode and display implsiximeow
2021-12-19fix incorrect memory size for f30f1e-style nopiximeow
not only did the instruction have wrong data, but if displayed, the formatter would panic.
2021-12-19test that invalid RegSpec constructions panic as expectediximeow
in the process, fix 64-bit rex-byte limit, 32/16-bit mode mask reg limit
2021-12-17write `apply_disp_scale` in a mode-agnostic wayiximeow
`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.
2021-12-17do not panic on negative compressed displacements, i mean it!!iximeow
2021-12-16bump version to 1.1.21.1.2iximeow
2021-12-16displacements are stored as unsigned, but are functionally signed intsiximeow
so multiplying to expand EVEX compressed offsets can overflow, and that needs to be okay.
2021-10-10bump version to 1.1.11.1.1iximeow
2021-10-10talk about contribution policy a littleiximeow
2021-10-10downgrade "most hardware" to "some hardware"iximeow
alas
2021-10-10add `InstructionDisplayer` export to changelogiximeow
2021-10-10support endbr{32,64}iximeow
2021-10-10consistentify doc styleiximeow
2021-10-10export `InstructionDisplayer` (#9)i509VCB
This makes generated docs refer to a type and show said type in the list of all structs rather than rustdoc showing gray text in return types. quote doc references
2021-08-22bump to yaxpeax-arch 0.2.7 and proper field description support1.1.0iximeow
2021-08-22add 16/32-bit opcode/operand boundary desc, consistentify memory descriptionsiximeow
2021-08-21add `AnnotatingDecoder` note to CHANGELOG and publicize descriptionsiximeow
2021-08-21improve relative branch offset formatting for DisplayStyle::Ciximeow
2021-08-21maintain pre-annotation inlining propertiesiximeow
this gets yaxpeax-x86 in no-inline configurations back to building as it did before, but is quite a blunt hammer. it seems that extra calls to `sink.record` trips the inlining thresholds for `read_with_annotation`, and then its caller, and its caller, even when one of them is just a delegation to its inner call. this is particularly unfortunate because yaxpeax-x86 is now making a decision about the inlining of a rather large function at the public edge of its API, but these attributes match the inlining decisions that LLVM was making before adding `DescriptionSink`. hopefully not too bad. not sure how to handle this in the future.
2021-08-21add descriptions for other prefixes, 16-bit addressingiximeow
2021-08-21add description reporting for segment prefixes and opcodes for 32-bit and 16-bitiximeow
2021-08-21provide decoder annotation for evex prefixiximeow
2021-08-21report barebones decoder annotation for vex-coded instructionsiximeow
2021-08-21extend decoder annotation through all of 64-, 32-, and 16-bit modesiximeow
2021-08-21force read_sib inlining in 64-bit modeiximeow
even though NullSink is no-ops, it causes llvm to not inline this function, for a net perf reduction
2021-08-21extend annotation reporting to 32- and 16-bit modes, kindaiximeow
2021-08-21wipiximeow