Commit Graph

51 Commits

Author SHA1 Message Date
est31
2ef8af6619 Adopt let else in more places 2022-02-19 17:27:43 +01:00
Mark Rousskov
0fb2b7a2da Drop json::from_reader
Performing UTF-8 decode outside the JSON module makes more sense in almost all cases.
2022-02-05 15:07:10 -05:00
bjorn3
0b8f3729fb Remove two unnecessary transmutes from opaque Encoder and Decoder 2022-01-31 18:25:05 +01:00
bors
e7825f2b69 Auto merge of #90842 - pierwill:localdefid-indexmap, r=wesleywiser
Use `indexmap` to avoid sorting `LocalDefId`s

See discussion in https://github.com/rust-lang/rust/pull/90408#discussion_r745935459.

Related to work on https://github.com/rust-lang/rust/issues/90317.
2022-01-24 22:04:55 +00:00
pierwill
4f89224f7f Use an indexmap to avoid sorting LocalDefIds
Update `indexmap` to 1.8.0.

Bless test
2022-01-22 22:34:16 -06:00
Nicholas Nethercote
37fbd91eb5 Address review comments. 2022-01-22 10:38:34 +11:00
Nicholas Nethercote
416399dc10 Make Decodable and Decoder infallible.
`Decoder` has two impls:
- opaque: this impl is already partly infallible, i.e. in some places it
  currently panics on failure (e.g. if the input is too short, or on a
  bad `Result` discriminant), and in some places it returns an error
  (e.g. on a bad `Option` discriminant). The number of places where
  either happens is surprisingly small, just because the binary
  representation has very little redundancy and a lot of input reading
  can occur even on malformed data.
- json: this impl is fully fallible, but it's only used (a) for the
  `.rlink` file production, and there's a `FIXME` comment suggesting it
  should change to a binary format, and (b) in a few tests in
  non-fundamental ways. Indeed #85993 is open to remove it entirely.

And the top-level places in the compiler that call into decoding just
abort on error anyway. So the fallibility is providing little value, and
getting rid of it leads to some non-trivial performance improvements.

Much of this commit is pretty boring and mechanical. Some notes about
a few interesting parts:
- The commit removes `Decoder::{Error,error}`.
- `InternIteratorElement::intern_with`: the impl for `T` now has the same
  optimization for small counts that the impl for `Result<T, E>` has,
  because it's now much hotter.
- Decodable impls for SmallVec, LinkedList, VecDeque now all use
  `collect`, which is nice; the one for `Vec` uses unsafe code, because
  that gave better perf on some benchmarks.
2022-01-22 10:38:31 +11:00
Nicholas Nethercote
88600a6d7f Rename Decoder::read_nil and read_unit.
Because `()` is called "unit" and it makes it match
`Encoder::emit_unit`.
2022-01-22 10:22:24 +11:00
bors
38c22af015 Auto merge of #92604 - nnethercote:optimize-impl_read_unsigned_leb128, r=michaelwoerister
Optimize `impl_read_unsigned_leb128`

I see instruction count improvements of up to 3.5% locally with these changes, mostly on the smaller benchmarks.

r? `@michaelwoerister`
2022-01-15 07:27:30 +00:00
Nicholas Nethercote
facba24926 Unpeel the first iteration of the loop in impl_read_unsigned_leb128. 2022-01-07 13:36:26 +11:00
Nicholas Nethercote
5f549d9b49 Modify the buffer position directly when reading leb128 values.
It's a small but clear performance win.
2022-01-07 10:40:51 +11:00
Jakub Beránek
6c9ffe4aec
Do not use LEB128 for encoding u16 and i16 2021-12-28 09:29:08 +01:00
bors
6d6d0899c8 Auto merge of #91626 - klensy:json-less-alloc, r=petrochenkov
rustc_serialize: don't allocate `String`s as key for `BTreeMap`, when `str` is enough
2021-12-13 07:30:38 +00:00
klensy
1b27b69e5a don't allocate strings when str is enought for using as key 2021-12-07 12:54:35 +03:00
The 8472
c640f31c9f avoid string validation in rustc_serialize, check a marker byte instead
since the serialization format isn't self-describing we need a way to detect
when encoder and decoder don't match up. but that doesn't have to
be utf8 validation for strings, which does cost a few % of performance.
Instead we can use a marker byte at the end to be reasonably
sure that we're dealing with a string and it wasn't overwritten in some
way.
2021-12-06 18:43:01 +01:00
Mark Rousskov
3228603cce Avoid generating empty closures for fieldless enums
For many enums, this avoids generating lots of tiny stubs that need to be
codegen'd and then inlined and removed by LLVM.
2021-11-22 21:22:35 -05:00
Matthias Krüger
4457014398 Revert "Auto merge of #89709 - clemenswasser:apply_clippy_suggestions_2, r=petrochenkov"
The PR had some unforseen perf regressions that are not as easy to find.
Revert the PR for now.

This reverts commit 6ae8912a3e, reversing
changes made to 86d6d2b738.
2021-10-15 11:28:23 +02:00
Clemens Wasser
71dd0b928b Apply clippy suggestions 2021-10-10 15:38:19 +02:00
Mark Rousskov
c746be2219 Migrate to 2021 2021-09-20 22:21:42 -04:00
Vadim Petrochenkov
294510e1bb rustc: Remove local variable IDs from Exports
Local variables can never be exported.
2021-09-10 23:41:48 +03:00
Frank Steffahn
be9d2699ca Fix more “a”/“an” typos 2021-08-22 16:35:29 +02:00
Frank Steffahn
bf88b113ea Fix typos “a”→“an” 2021-08-22 15:35:11 +02:00
est31
823e6f161a Remove box syntax from rustc_serialize 2021-08-18 09:25:26 +02:00
Jade
3cf820e17d rfc3052: Remove authors field from Cargo manifests
Since RFC 3052 soft deprecated the authors field anyway, hiding it from
crates.io, docs.rs, and making Cargo not add it by default, and it is
not generally up to date/useful information, we should remove it from
crates in this repo.
2021-07-29 14:56:05 -07:00
Adam Bratschi-Kaye
88b01f1178 Emit warnings for unused fields in custom targets. 2021-06-17 21:48:02 +02:00
LeSeulArtichaut
e3ca81fd5a Use the now available implementation of IntoIterator for arrays 2021-06-14 23:40:09 +02:00
bors
e4a6032706 Auto merge of #85903 - bjorn3:rustc_serialize_cleanup, r=varkor
Remove unused functions and arguments from rustc_serialize
2021-06-07 14:40:26 +00:00
bjorn3
a2c4affe86 Remove unused functions and arguments from rustc_serialize 2021-06-01 19:29:11 +02:00
bjorn3
312f964478 Remove unused feature gates 2021-05-31 13:55:43 +02:00
bors
0ced530534 Auto merge of #83465 - michaelwoerister:safe-read_raw_bytes, r=cjgillot
Allow for reading raw bytes from rustc_serialize::Decoder without unsafe code

The current `read_raw_bytes` method requires using `MaybeUninit` and `unsafe`. I don't think this is necessary. Let's see if a safe interface has any performance drawbacks.

This is a followup to #83273 and will make it easier to rebase #82183.

r? `@cjgillot`
2021-03-26 01:28:59 +00:00
Michael Woerister
517d5ac230 Allow for reading raw bytes from rustc_serialize::Decoder without unsafe code. 2021-03-25 14:05:00 +01:00
Mara Bos
81932be5e7 Revert "Revert stabilizing integer::BITS." 2021-03-24 22:34:36 +01:00
Camille GILLOT
09a638820e Move raw bytes handling to Encoder/Decoder. 2021-03-19 19:35:22 +01:00
Camille GILLOT
e5d09fbbe9 Simplify IntEncodedWithFixedSize. 2021-03-18 20:09:00 +01:00
Camille GILLOT
5003b3dc31 Move IntEncodedWithFixedSize to rustc_serialize. 2021-03-18 20:09:00 +01:00
Joshua Nelson
3733275854 Update the bootstrap compiler
Note this does not change `core::derive` since it was merged after the
beta bump.
2021-02-20 17:19:30 -05:00
klensy
93c8ebe022 bumped smallvec deps 2021-02-14 18:03:11 +03:00
Mara Bos
89882388d9 Revert stabilizing integer::BITS. 2021-02-03 22:23:58 +01:00
Ashley Mannix
8940a2652e stabilize int_bits_const 2021-01-31 21:50:47 +10:00
Tyson Nottingham
f15fae822e rustc_serialize: fix incorrect signed LEB128 decoding
The signed LEB128 decoding function used a hardcoded constant of 64
instead of the number of bits in the type of integer being decoded,
which resulted in incorrect results for some inputs. Fix this, make the
decoding more consistent with the unsigned version, and increase the
LEB128 encoding and decoding test coverage.
2021-01-11 12:13:26 -08:00
Tyson Nottingham
52f21791fb Serialize incr comp structures to file via fixed-size buffer
Reduce a large memory spike that happens during serialization by writing
the incr comp structures to file by way of a fixed-size buffer, rather
than an unbounded vector.

Effort was made to keep the instruction count close to that of the
previous implementation. However, buffered writing to a file inherently
has more overhead than writing to a vector, because each write may
result in a handleable error. To reduce this overhead, arrangements are
made so that each LEB128-encoded integer can be written to the buffer
with only one capacity and error check. Higher-level optimizations in
which entire composite structures can be written with one capacity and
error check are possible, but would require much more work.

The performance is mostly on par with the previous implementation, with
small to moderate instruction count regressions. The memory reduction is
significant, however, so it seems like a worth-while trade-off.
2021-01-11 12:13:22 -08:00
Tyson Nottingham
be79f493fb rustc_serialize: specialize opaque decoding of some u8 sequences 2021-01-01 22:49:16 -08:00
Tyson Nottingham
7c6274d464 rustc_serialize: have read_raw_bytes take MaybeUninit<u8> slice 2021-01-01 22:49:16 -08:00
Tyson Nottingham
a4daa63a90 rustc_serialize: specialize opaque encoding of some u8 sequences 2021-01-01 22:49:14 -08:00
Bastian Kauschke
06cc9c26da stabilize min_const_generics 2020-12-26 18:24:10 +01:00
Joshua Nelson
0ad3dce83a Fix some clippy lints 2020-12-03 17:08:19 -05:00
est31
a0fc455d30 Replace absolute paths with relative ones
Modern compilers allow reaching external crates
like std or core via relative paths in modules
outside of lib.rs and main.rs.
2020-10-13 14:16:45 +02:00
Oliver Scherer
455f284496 Deduplicate and generalize some (de/)serializer impls 2020-09-26 14:55:42 +02:00
Erik Hofmayer
138a2e5eaa /nightly/nightly-rustc 2020-09-23 21:51:56 +02:00
Erik Hofmayer
dd66ea2d3d Updated html_root_url for compiler crates 2020-09-23 21:14:43 +02:00