about summary refs log tree commit diff
path: root/src/librustc_data_structures
AgeCommit message (Collapse)AuthorLines
2020-02-19Tune inliningJohn Kåre Alsaker-3/+7
2020-02-15Rollup merge of #68475 - Aaron1011:fix/forest-caching, r=nikomatsakisYuki Okushi-15/+20
Use a `ParamEnvAnd<Predicate>` for caching in `ObligationForest` Previously, we used a plain `Predicate` to cache results (e.g. successes and failures) in ObligationForest. However, fulfillment depends on the precise `ParamEnv` used, so this is unsound in general. This commit changes the impl of `ForestObligation` for `PendingPredicateObligation` to use `ParamEnvAnd<Predicate>` instead of `Predicate` for the associated type. The associated type and method are renamed from 'predicate' to 'cache_key' to reflect the fact that type is no longer just a predicate.
2020-02-14Auto merge of #68693 - Zoxc:query-no-arc, r=michaelwoeristerbors-2/+11
Construct query job latches on-demand r? @michaelwoerister
2020-02-13add selfprofiling for new llvm passmanagerAndreas Jonson-0/+11
2020-02-12Rollup merge of #68914 - nnethercote:speed-up-SipHasher128, r=michaelwoeristerDylan DPC-80/+84
Speed up `SipHasher128`. The current code in `SipHasher128::short_write` is inefficient. It uses `u8to64_le` (which is complex and slow) to extract just the right number of bytes of the input into a u64 and pad the result with zeroes. It then left-shifts that value in order to bitwise-OR it with `self.tail`. For example, imagine we have a u32 input `0xIIHH_GGFF` and only need three bytes to fill up `self.tail`. The current code uses `u8to64_le` to construct `0x0000_0000_00HH_GGFF`, which is just `0xIIHH_GGFF` with the `0xII` removed and zero-extended to a u64. The code then left-shifts that value by five bytes -- discarding the `0x00` byte that replaced the `0xII` byte! -- to give `0xHHGG_FF00_0000_0000`. It then then ORs that value with `self.tail`. There's a much simpler way to do it: zero-extend to u64 first, then left shift. E.g. `0xIIHH_GGFF` is zero-extended to `0x0000_0000_IIHH_GGFF`, and then left-shifted to `0xHHGG_FF00_0000_0000`. We don't have to take time to exclude the unneeded `0xII` byte, because it just gets shifted out anyway! It also avoids multiple occurrences of `unsafe`. There's a similar story with the setting of `self.tail` at the method's end. The current code uses `u8to64_le` to extract the remaining part of the input, but the same effect can be achieved more quickly with a right shift on the zero-extended input. This commit changes `SipHasher128` to use the simpler shift-based approach. The code is also smaller, which means that `short_write` is now inlined where previously it wasn't, which makes things faster again. This gives big speed-ups for all incremental builds, especially "baseline" incremental builds. r? @michaelwoerister
2020-02-12Use a counter instead of pointers to the stackJohn Kåre Alsaker-2/+11
2020-02-12Improve `u8to64_le`.Nicholas Nethercote-41/+12
This makes it faster and also changes it to a safe function. (Thanks to Michael Woerister for the suggestion.) `load_int_le!` is also no longer necessary.
2020-02-11Move macro enum_from_u32 to rustc_data_structures.Camille GILLOT-0/+38
2020-02-11Rollup merge of #66498 - bjorn3:less_feature_flags, r=Dylan-DPCDylan DPC-3/+0
Remove unused feature gates I think many of the remaining unstable things can be easily be replaced with stable things. I have kept the `#![feature(nll)]` even though it is only necessary in `libstd`, to make regressions of it harder.
2020-02-10self-profile: Support arguments for generic_activities.Michael Woerister-34/+73
2020-02-10Speed up `SipHasher128`.Nicholas Nethercote-39/+72
The current code in `SipHasher128::short_write` is inefficient. It uses `u8to64_le` (which is complex and slow) to extract just the right number of bytes of the input into a u64 and pad the result with zeroes. It then left-shifts that value in order to bitwise-OR it with `self.tail`. For example, imagine we have a u32 input 0xIIHH_GGFF and only need three bytes to fill up `self.tail`. The current code uses `u8to64_le` to construct 0x0000_0000_00HH_GGFF, which is just 0xIIHH_GGFF with the 0xII removed and zero-extended to a u64. The code then left-shifts that value by five bytes -- discarding the 0x00 byte that replaced the 0xII byte! -- to give 0xHHGG_FF00_0000_0000. It then then ORs that value with self.tail. There's a much simpler way to do it: zero-extend to u64 first, then left shift. E.g. 0xIIHH_GGFF is zero-extended to 0x0000_0000_IIHH_GGFF, and then left-shifted to 0xHHGG_FF00_0000_0000. We don't have to take time to exclude the unneeded 0xII byte, because it just gets shifted out anyway! It also avoids multiple occurrences of `unsafe`. There's a similar story with the setting of `self.tail` at the method's end. The current code uses `u8to64_le` to extract the remaining part of the input, but the same effect can be achieved more quickly with a right shift on the zero-extended input. All that works on little-endian. It doesn't work for big-endian, but we can just do a `to_le` before calling `short_write` and then it works. This commit changes `SipHasher128` to use the simpler shift-based approach. The code is also smaller, which means that `short_write` is now inlined where previously it wasn't, which makes things faster again. This gives big speed-ups for all incremental builds, especially "baseline" incremental builds.
2020-02-06Rollup merge of #68524 - jonas-schievink:generator-resume-arguments, r=ZoxcDylan DPC-0/+39
Generator Resume Arguments cc https://github.com/rust-lang/rust/issues/43122 and https://github.com/rust-lang/rust/issues/56974 Blockers: * [x] Fix miscompilation when resume argument is live across a yield point (https://github.com/rust-lang/rust/pull/68524#issuecomment-578459069) * [x] Fix 10% compile time regression in `await-call-tree` benchmarks (https://github.com/rust-lang/rust/pull/68524#issuecomment-578487162) * [x] Fix remaining 1-3% regression (https://github.com/rust-lang/rust/pull/68524#issuecomment-579566255) - resolved (https://github.com/rust-lang/rust/pull/68524#issuecomment-581144901) * [x] Make dropck rules account for resume arguments (https://github.com/rust-lang/rust/pull/68524#issuecomment-578541137) Follow-up work: * Change async/await desugaring to make use of this feature * Rewrite [`box_region.rs`](https://github.com/rust-lang/rust/blob/3d8778d767f0dde6fe2bc9459f21ead8e124d8cb/src/librustc_data_structures/box_region.rs) to use resume arguments (this shows up in profiles too)
2020-02-06Remove `RefCell` usage from `ObligationForest`.Nicholas Nethercote-5/+5
It's not needed.
2020-02-04Remove unused feature gates from librustc_data_structuresbjorn3-3/+0
2020-02-02Fix bootstrap rustc buildJonas Schievink-0/+39
2020-02-01Use BufWriterShotaro Yamada-1/+2
2020-01-24[self-profiler] Clean up `EventFilter`Wesley Wiser-7/+3
2020-01-24[self-profiler] Use `ThreadId::as_u64()` instead of transmuteWesley Wiser-7/+3
2020-01-23unused-parens: implement for block return valuesTyler Lanphear-1/+1
2020-01-22Use a `ParamEnvAnd<Predicate>` for caching in `ObligationForest`Aaron Hill-15/+20
Previously, we used a plain `Predicate` to cache results (e.g. successes and failures) in ObligationForest. However, fulfillment depends on the precise `ParamEnv` used, so this is unsound in general. This commit changes the impl of `ForestObligation` for `PendingPredicateObligation` to use `ParamEnvAnd<Predicate>` instead of `Predicate` for the associated type. The associated type and method are renamed from 'predicate' to 'cache_key' to reflect the fact that type is no longer just a predicate.
2020-01-17Rollup merge of #68278 - wesleywiser:doc_query_key_recording, r=michaelwoeristerDylan DPC-0/+2
[self-profiler] Add example to `-Z help` to turn on query key recording Also add the `default` option so that it's easy to add query key recording to the default. r? @michaelwoerister
2020-01-17Rollup merge of #67791 - Zoxc:lift-interning, r=eddybDylan DPC-0/+14
Implement Lift using interners instead of in_arena r? @eddyb cc @cjgillot
2020-01-17[self-profiler] Add example to `-Z help` to turn on query key recordingWesley Wiser-0/+2
Also add the `default` option so that it's easy to add query key recording to the default.
2020-01-15Rollup merge of #68141 - euclio:replace-bindings-with-winapi, r=alexcrichtonYuki Okushi-67/+19
use winapi for non-stdlib Windows bindings
2020-01-12Rollup merge of #67948 - llogiq:gallop, r=Mark-SimulacrumMazdak Farrokhzad-20/+40
Galloping search for binary_search_util This is unlikely to improve perf much unless for synthetic benchmarks, but I figure it likely won't hurt either.
2020-01-11use winapi for non-stdlib Windows bindingsAndy Russell-67/+19
2020-01-12Galloping search for binary_search_utilAndre Bogus-20/+40
2020-01-11Lift using interners instead of in_arenaJohn Kåre Alsaker-0/+14
2020-01-11Rollup merge of #68043 - Zoxc:missing-timers, r=wesleywiserMazdak Farrokhzad-0/+6
Add some missing timers Based on https://github.com/rust-lang/rust/pull/67988 r? @wesleywiser
2020-01-10Fix some rebasing fallout.Michael Woerister-3/+6
2020-01-10Update measureme to 0.7.1 in order to fix compilation error on big-endian ↵Michael Woerister-1/+1
platforms.
2020-01-10Run 'x.py fmt'.Michael Woerister-22/+7
2020-01-10self-profile: Fix issue with handling query blocking.Michael Woerister-0/+1
2020-01-10Initial support for recording query keys in self-profiling data.Michael Woerister-9/+39
2020-01-10self-profile: Switch to new approach for event_id generation that enables ↵Michael Woerister-32/+161
query-invocation-specific event_ids.
2020-01-10Rollup merge of #67922 - Centril:lowering-cleanup, r=petrochenkovMazdak Farrokhzad-0/+11
rustc_ast_lowering: misc cleanup & rustc dep reductions - The first two commits do some code simplification. - The next three do some file splitting (getting `lib.rs` below the 3kloc tidy lint). - The remaining commits reduce the number of `rustc::` imports. This works towards making lowering independent of the `rustc` crate. r? @oli-obk cc @Zoxc
2020-01-09Label unmarked timeJohn Kåre Alsaker-0/+6
2020-01-09{rustc::util -> rustc_data_structures}::capturesMazdak Farrokhzad-0/+11
2020-01-09More commentsJohn Kåre Alsaker-2/+6
2020-01-09Change -Z time event naming scheme and make them generic activitiesJohn Kåre Alsaker-92/+47
2020-01-05Rollup merge of #67882 - euclio:flock, r=rkruppeDylan DPC-106/+15
remove bespoke flock bindings Replaces some `struct flock` definitions with the definition from `libc`.
2020-01-05Use self profile infrastructure for -Z time and -Z time-passesJohn Kåre Alsaker-19/+213
2020-01-04remove bespoke flock bindingsAndy Russell-106/+15
2020-01-04define_id_collections -> rustc_data_structuresMazdak Farrokhzad-0/+8
2019-12-31Revert parts of #66405.Nicholas Nethercote-111/+80
Because it caused major performance regressions in some cases. That PR had five commits, two of which affected performance, and three of which were refactorings. This change undoes the performance-affecting changes, while keeping the refactorings in place. Fixes #67454.
2019-12-29Auto merge of #67614 - Mark-Simulacrum:global-callbacks, r=Zoxcbors-0/+28
Set callbacks globally This sets the callbacks from syntax and rustc_errors just once, utilizing static (rather than thread-local) storage.
2019-12-26Convert collapsed to shortcut reference linksMatthew Kraai-5/+5
2019-12-25Store callbacks in global staticsMark Rousskov-0/+28
The callbacks have precisely two states: the default, and the one present throughout almost all of the rustc run (the filled in value which has access to TyCtxt). We used to store this as a thread local, and reset it on each thread to the non-default value. But this is somewhat wasteful, since there is no reason to set it globally -- while the callbacks themselves access TLS, they do not do so in a manner that fails in when we do not have TLS to work with.
2019-12-22Format the worldMark Rousskov-1286/+1278
2019-12-17Revert "Auto merge of #67362 - Mark-Simulacrum:par-4-default, r=alexcrichton"Mark Rousskov-4/+1
This reverts commit 3ed3b8bb7b100afecf7d5f52eafbb70fec27f537, reversing changes made to 99b89533d4cdf7682ea4054ad0ee36c351d05df1. We will reland a similar patch at a future date but for now we should get a nightly released in a few hours with the parallel patch, so this should be reverted to make sure that the next nightly is not parallel-enabled.
generated by cgit-pink 1.4.1-3-g733a5 (git 2.37.1) at 2026-04-27 21:42:57 +0000