summaryrefslogtreecommitdiff
path: root/crates/hashx
Commit message (Collapse)AuthorAgeFilesLines
* Run cargo update for {hashx,equix}/bench.Gabriela Moldovan2024-06-051-105/+91
| | | | | | | | | Done using ``` (cd crates/hashx/bench && cargo update) (cd crates/equix/bench && cargo update) ```
* Bump the versions of the non-{arti-,tor-} crates.Gabriela Moldovan2024-06-051-1/+1
| | | | | | | | | | | | | | | | | | | | | * crates that have no changes: none * crates that only have non-functional changes (bump the patch version): - test-temp-dir - fslock-guard - hashx - equix - caret - fs-mistrust - safelog - retry-error Done using ``` for c in "${non_functional[@]}"; do maint/bump_nodep $c; done ``` where `non_functional` contains the list above
* Re-run maint/add_warning.Nick Mathewson2024-05-061-2/+2
| | | | This commit is automatically generated.
* Update separate lockfilesIan Jackson2024-04-301-1/+1
|
* Bump versions of crates with nonfunctional changesIan Jackson2024-04-301-1/+1
| | | | | | | | | | for p in `cat ../u`; do maint/bump_nodep $p; git commit -a -m X; done where u contains hashx equix fs-mistrust
* Ship GPL-3 alongside LGPL-3, as requiredIan Jackson2024-04-251-0/+674
|
* Copy LGPL-3 file into hashx and equix cratesIan Jackson2024-04-251-0/+165
| | | | | | | | This is the simplest way to get this included during "cargo publish", which is necessary. Prompted by https://gitlab.torproject.org/tpo/core/arti/-/merge_requests/2094#note_3021187
* Resolve nightly clippy errors in hashx.Nick Mathewson2024-04-221-3/+3
| | | | | | Nightly clippy doesn't like using "expr as T" when the conversion is lossless; it prefers "T::from(expr)" so that if we later change the type of T to something with a lossy conversion, we'll know.
* Run cargo update in {equix,hashx}/benchNick Mathewson2024-04-021-208/+162
|
* Bump patchlevel versions on non-{tor/arti} crates.Nick Mathewson2024-04-021-1/+1
| | | | | | | | | | | | | | | | These have all had backward-compatible changes. Generated with: ``` cargo set-version --bump patch -p fs-mistrust cargo set-version --bump patch -p test-temp-dir cargo set-version --bump patch -p fslock-guard cargo set-version --bump patch -p hashx cargo set-version --bump patch -p equix cargo set-version --bump patch -p caret cargo set-version --bump patch -p safelog cargo set-version --bump patch -p retry-error ```
* deny clippy::unchecked_duration_subtractiontrinity-1686a2024-02-291-0/+1
|
* Fix typosDimitris Apostolou2024-01-081-1/+1
|
* In every crate, change rust-version to 1.70.Nick Mathewson2023-11-281-1/+1
|
* Update other cargo.lock files.Nick Mathewson2023-10-021-1/+1
|
* Patchlevel bumps, with dependents affected.Nick Mathewson2023-10-021-1/+1
|
* Update bench lockfiles for new fixed-capacity-vecNick Mathewson2023-09-201-2/+2
|
* Upgrade to fixed-capacity-vec 1.0Nick Mathewson2023-09-201-1/+1
|
* Switch to FixedCapacityVec 0.1.0 from crates.ioIan Jackson2023-09-065-300/+9
|
* Upgrade bench/ lockfiles too.Nick Mathewson2023-09-051-80/+91
|
* Update patchlevel for crates with nontrivial changes.Nick Mathewson2023-09-051-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | These are: ``` hashx equix tor-async-utils tor-error tor-config tor-rtmock tor-llcrypto tor-bytes tor-hscrypto tor-hspow tor-cert tor-linkspec tor-cell tor-proto tor-netdoc tor-netdir tor-chanmgr tor-guardmgr tor-dirmgr tor-keymgr tor-hsclient tor-hsservice arti-client arti ```
* hashx: Cleanup around Instruction and NUM_INSTRUCTIONSMicah Elizabeth Scott2023-08-256-28/+24
| | | | | | | | | This patch tries to make some of the expressions around NUM_INSTRUCTIONS more convenient. We can import it directly where it's needed, but most uses are replaced by new type aliases for InstructionArray and InstructionVec. No change to any hashx_cachegrind iai benchmarks
* hashx: Avoid memcpy in Assembler::finalize()Micah Elizabeth Scott2023-08-251-1/+7
| | | | | | | | | | | | | | | | | This is a very simple change, just passing 'self' by reference instead of value. The by-value version generates a memcpy of the entire temporary program buffer which doesn't optimize out like I expected it would. The juicy impact here is a much lower cache footprint for compilation, since we avoid having yet another temporary storage location for the program data. generate_compiled_1000x Instructions: 271682605 (-0.627292%) L1 Accesses: 341834751 (-0.813903%) L2 Accesses: 56420 (-39.48365%) RAM Accesses: 618 (-20.25806%) Estimated Cycles: 342138481 (-0.867660%)
* RustfmtIan Jackson2023-08-253-3/+3
|
* RFC: hashx: Make Architecture::compile take an array refIan Jackson2023-08-254-6/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | generate_interp_1000x Instructions: 219216169 (No change) L1 Accesses: 278017243 (-0.000441%) L2 Accesses: 1257 (+4389.286%) RAM Accesses: 415 (-0.479616%) Estimated Cycles: 278038053 (+0.001744%) generate_interp_1000x_c Instructions: 272748034 (No change) L1 Accesses: 349932964 (No change) L2 Accesses: 76 (-1.298701%) RAM Accesses: 411 (+0.243902%) Estimated Cycles: 349947729 (+0.000009%) generate_compiled_1000x Instructions: 256896028 (+0.175731%) L1 Accesses: 342543838 (+0.131802%) L2 Accesses: 149273 (-10.34924%) RAM Accesses: 810 (-0.246305%) Estimated Cycles: 343318553 (+0.106328%) generate_compiled_1000x_c Instructions: 281855218 (No change) L1 Accesses: 362569035 (-0.000001%) L2 Accesses: 88 (+1.149425%) RAM Accesses: 473 (+0.211864%) Estimated Cycles: 362586030 (+0.000010%) interp_u64_hash_1000x Instructions: 13450926 (No change) L1 Accesses: 16622561 (+0.000024%) L2 Accesses: 28 (No change) RAM Accesses: 390 (-1.015228%) Estimated Cycles: 16636351 (-0.000817%) interp_8b_hash_1000x_c Instructions: 8618541 (No change) L1 Accesses: 12316160 (-0.000008%) L2 Accesses: 80 (No change) RAM Accesses: 433 (+0.231481%) Estimated Cycles: 12331715 (+0.000276%) compiled_u64_hash_100000x Instructions: 87311792 (+0.000520%) L1 Accesses: 94396598 (+0.000463%) L2 Accesses: 215 (+2.380952%) RAM Accesses: 774 (-0.641849%) Estimated Cycles: 94424763 (+0.000304%) compiled_8b_hash_100000x_c Instructions: 91547640 (No change) L1 Accesses: 98838166 (-0.000007%) L2 Accesses: 137 (+3.007519%) RAM Accesses: 488 (+0.618557%) Estimated Cycles: 98855931 (+0.000119%)
* Run maint/add-warningIan Jackson2023-08-251-0/+1
| | | | Yet another new module concurrently with a new lint.
* hashx: FixedCapacityVec: Fix clippy lintsIan Jackson2023-08-241-4/+3
|
* RustfmtIan Jackson2023-08-244-6/+12
|
* hashx: FixedCapacityVec: impl Send, Sync, [Ref]UnwindSafeIan Jackson2023-08-241-0/+13
|
* hashx: FixedCapacityVec: TestsIan Jackson2023-08-241-0/+105
| | | | These pass miri too.
* hashx: FixedCapacityVec: introduce push_innerIan Jackson2023-08-241-7/+21
| | | | | | To support testing. No change to iai benchmarks.
* hashx: FixedCapacityVec: use *mut T, and impl DropIan Jackson2023-08-241-24/+68
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | generate_interp_1000x Instructions: 219216169 (-0.408535%) L1 Accesses: 278018470 (-0.322766%) L2 Accesses: 28 (+12.00000%) RAM Accesses: 417 (+0.724638%) Estimated Cycles: 278033205 (-0.322706%) generate_interp_1000x_c Instructions: 272748034 (No change) L1 Accesses: 349932964 (-0.000002%) L2 Accesses: 77 (+1.315789%) RAM Accesses: 410 (+1.234568%) Estimated Cycles: 349947699 (+0.000050%) generate_compiled_1000x Instructions: 256445375 (-0.349434%) L1 Accesses: 342092951 (-0.266541%) L2 Accesses: 166505 (+9.182175%) RAM Accesses: 812 (+0.370828%) Estimated Cycles: 342953896 (-0.245532%) generate_compiled_1000x_c Instructions: 281855218 (No change) L1 Accesses: 362569037 (-0.000002%) L2 Accesses: 87 (No change) RAM Accesses: 472 (+1.287554%) Estimated Cycles: 362585992 (+0.000056%) interp_u64_hash_1000x Instructions: 13450926 (-0.006631%) L1 Accesses: 16622557 (-0.005384%) L2 Accesses: 28 (No change) RAM Accesses: 394 (+0.510204%) Estimated Cycles: 16636487 (-0.004959%) interp_8b_hash_1000x_c Instructions: 8618541 (No change) L1 Accesses: 12316161 (-0.000032%) L2 Accesses: 80 (No change) RAM Accesses: 432 (+0.934579%) Estimated Cycles: 12331681 (+0.001103%) compiled_u64_hash_100000x Instructions: 87311338 (-0.001022%) L1 Accesses: 94396161 (-0.000947%) L2 Accesses: 210 (-0.943396%) RAM Accesses: 779 (+0.386598%) Estimated Cycles: 94424476 (-0.000846%) compiled_8b_hash_100000x_c Instructions: 91547640 (No change) L1 Accesses: 98838173 (-0.000003%) L2 Accesses: 133 (-0.746269%) RAM Accesses: 485 (+0.831601%) Estimated Cycles: 98855813 (+0.000134%)
* hashx: FixedCapacityVec: Note some missing stuffIan Jackson2023-08-241-0/+12
|
* hashx: FixedCapacityVec: DocumentationIan Jackson2023-08-241-0/+27
|
* hashx: FixedCapacityVec: Replace .into_boxed_array method with TryIan Jackson2023-08-242-9/+16
| | | | | This version pushes the panic into the call site, which seems much better. No change to the iai results.
* hashx: FixedCapacityVec: Move into its own moduleIan Jackson2023-08-245-71/+80
|
* RustfmtIan Jackson2023-08-233-9/+9
|
* RFC: hashx: Introduce FixedCapacityVecIan Jackson2023-08-233-6/+81
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is quite shoddy. It shouldn't be merged without some tidying up and unit tests and so on. Also I am confused about the difference between NUM_INSTRUCTIONS and model::REQUIRED_INSTRUCTIONS. However: generate_interp_1000x Instructions: 220115418 (-1.147100%) L1 Accesses: 278918725 (-0.936103%) L2 Accesses: 25 (-98.46248%) RAM Accesses: 414 (-0.956938%) Estimated Cycles: 278933340 (-0.938920%) generate_interp_1000x_c Instructions: 272748034 (No change) L1 Accesses: 349932970 (+0.000001%) L2 Accesses: 76 (-1.298701%) RAM Accesses: 405 (-0.491400%) Estimated Cycles: 349947525 (-0.000021%) generate_compiled_1000x Instructions: 257344624 (-0.982784%) L1 Accesses: 343007206 (-0.753942%) L2 Accesses: 152502 (-17.12839%) RAM Accesses: 809 (-0.369458%) Estimated Cycles: 343798031 (-0.797384%) generate_compiled_1000x_c Instructions: 281855218 (No change) L1 Accesses: 362569043 (+0.000002%) L2 Accesses: 87 (-4.395604%) RAM Accesses: 466 (-0.427350%) Estimated Cycles: 362585788 (-0.000023%) interp_u64_hash_1000x Instructions: 13451818 (-0.100680%) L1 Accesses: 16623452 (-0.105967%) L2 Accesses: 28 (No change) RAM Accesses: 392 (-1.507538%) Estimated Cycles: 16637312 (-0.107138%) interp_8b_hash_1000x_c Instructions: 8618541 (No change) L1 Accesses: 12316165 (+0.000032%) L2 Accesses: 80 (-2.439024%) RAM Accesses: 428 (-0.465116%) Estimated Cycles: 12331545 (-0.000616%) compiled_u64_hash_100000x Instructions: 87312230 (-1.358594%) L1 Accesses: 94397055 (-1.669415%) L2 Accesses: 212 (-0.469484%) RAM Accesses: 776 (-0.767263%) Estimated Cycles: 94425275 (-1.669144%) compiled_8b_hash_100000x_c Instructions: 91547640 (No change) L1 Accesses: 98838176 (+0.000009%) L2 Accesses: 134 (-4.964539%) RAM Accesses: 481 (-0.414079%) Estimated Cycles: 98855681 (-0.000097%)
* RFC: hashx: Make Program a boxed arrayIan Jackson2023-08-231-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | generate_interp_1000x Instructions: 222669662 (-0.146601%) L1 Accesses: 281554367 (+0.069275%) L2 Accesses: 1623 (-90.13494%) RAM Accesses: 418 (No change) Estimated Cycles: 281577112 (+0.042908%) generate_interp_1000x_c Instructions: 272748034 (No change) L1 Accesses: 349932970 (+0.000001%) L2 Accesses: 74 (No change) RAM Accesses: 407 (-0.731707%) Estimated Cycles: 349947585 (-0.000029%) generate_compiled_1000x Instructions: 259898868 (-0.124860%) L1 Accesses: 345603941 (+0.055045%) L2 Accesses: 193008 (-4.001910%) RAM Accesses: 812 (-0.490196%) Estimated Cycles: 346597401 (+0.043228%) generate_compiled_1000x_c Instructions: 281855218 (No change) L1 Accesses: 362569040 (+0.000000%) L2 Accesses: 88 (+2.325581%) RAM Accesses: 468 (-0.636943%) Estimated Cycles: 362585860 (-0.000026%) interp_u64_hash_1000x Instructions: 13465375 (-0.024687%) L1 Accesses: 16641089 (-0.028974%) L2 Accesses: 25 (+8.695652%) RAM Accesses: 398 (+0.505051%) Estimated Cycles: 16655144 (-0.028470%) interp_8b_hash_1000x_c Instructions: 8618541 (No change) L1 Accesses: 12316165 (+0.000016%) L2 Accesses: 78 (No change) RAM Accesses: 430 (-0.462963%) Estimated Cycles: 12331605 (-0.000551%) compiled_u64_hash_100000x Instructions: 88514787 (-0.000365%) L1 Accesses: 95999693 (+0.000196%) L2 Accesses: 208 (-0.952381%) RAM Accesses: 782 (-0.255102%) Estimated Cycles: 96028103 (+0.000112%) compiled_8b_hash_100000x_c Instructions: 91547640 (No change) L1 Accesses: 98838171 (-0.000002%) L2 Accesses: 137 (+3.007519%) RAM Accesses: 483 (-0.412371%) Estimated Cycles: 98855761 (-0.000053%)
* hashx_cachegrind: Fix black_box locationIan Jackson2023-08-231-2/+2
|
* hashx_cachegrind: Make mk_c_equix not shuffle the HashXIan Jackson2023-08-231-8/+8
|
* hashx_cachegrind: Make mk_rust not shuffle the HashXBuilderIan Jackson2023-08-231-10/+9
|
* hashx_cachegrind: Introduce C_HASHX_OK aliasIan Jackson2023-08-231-2/+5
|
* hashx_cachegrind: Introduce bench_loop helper macroIan Jackson2023-08-231-24/+29
| | | | | The macro generates similar but not identical code. There are new bindings.
* hashx_cachegrind: Introduce u32be helper functionIan Jackson2023-08-231-4/+11
| | | | This is going to be more obviously useful in a moment.
* hashx_cachegrind: Introduce mk_c_equix helper macroIan Jackson2023-08-231-4/+9
| | | | The macro generates precisely the existing code.
* hashx_cachegrind: Introduce mk_rust helper macroIan Jackson2023-08-231-8/+14
| | | | The macro generates precisely the existing code.
* hashx: Use a boxed slice for Program storageMicah Elizabeth Scott2023-08-211-3/+3
| | | | | | | | | This is a very small change that converts our Vec cheaply into a boxed slice during program generation. Program generation speed shows no changes, and there's no change when using compiled hashes, but is a surprisingly effective 10% speedup to interpreted hash execution. Signed-off-by: Micah Elizabeth Scott <[email protected]>
* hashx: Assembly buffer sizing and tidyingMicah Elizabeth Scott2023-08-213-34/+104
| | | | | | | | | | | | | | | | | I was looking for ways to optimize out the many redundant capacity checks in the Assembler. I didn't find any promising approaches, but I also saw no evidence that it was an important bottleneck. (A simple unsafe fix didn't improve any important metrics) While I was in there, I tightened up the buffer size definitions for both x86_64 and aarch64, and added assertions to test the limits we set for the size of prologue, epilogue, and single instructions. I kept some of the inlining and data type tweaks, even though benchmarks show no difference. They seem like a step in the right direction, from the disassembly at least. Signed-off-by: Micah Elizabeth Scott <[email protected]>
* hashx: avoid surprising overhead of enum code() methodMicah Elizabeth Scott2023-08-211-3/+3
| | | | | | | | | | | This is a very simple change that avoids a surprising performance pitfall: using the code() method on an enum from another crate caused a non-inlined function call in code where we otherwise expect a high level of compiler optimization. Replacing code() with a cast to u8 avoids this function call and allows more intensive optimization at the call site. Signed-off-by: Micah Elizabeth Scott <[email protected]>
* hashx: Rearrange destination register validator for performanceMicah Elizabeth Scott2023-08-212-12/+49
| | | | | | | | | | | | | This hoists a few decisions out of the innermost portions of choose_dst_reg, by moving what we can out of dst_register_allowed. Wallclock time benchmarks: generate-interp improves, -6.0% Cachegrind benchmarks: generate_interp_1000x, -5.0% instructions, -11.6% L2 access, -6% RAM Signed-off-by: Micah Elizabeth Scott <[email protected]>
generated by cgit v1.2.3 (git 2.46.0) at 2026-08-26 22:58:27 +0000