These are useful when you want to catch the signals, like when you're making a kernel, or if you just don't want the overhead. (I don't know if there are any of the second kind of people, I don't think it's a good idea, but hey, choice is good).
Currently, LLVM lowers a cttz8 on x86_64 to these instructions:
```asm
movzbl %dil, %eax
bsfl %eax, %eax
movl $32, %ecx
cmovnel %eax, %ecx
cmpl $32, %ecx
movl $8, %eax
cmovnel %ecx, %eax
```
To improve the codegen, we can zero extend the 8 bit integer, then set
bit 8 and perform a cttz operation on the extended value. That way
there's no conditional operation involved at all.
This was discovered by this benchmark: https://github.com/Kimundi/long_strings_without_repeats
Timings on my box with the current nightly:
```
running 4 tests
test bench_cpp_naive_big ... bench: 5479222 ns/iter (+/- 254222)
test bench_noop_big ... bench: 571405 ns/iter (+/- 111950)
test bench_rust_naive_big ... bench: 7798102 ns/iter (+/- 148841)
test bench_rust_unsafe_big ... bench: 6606488 ns/iter (+/- 67529)
```
Timings with the patch applied:
```
running 4 tests
test bench_cpp_naive_big ... bench: 5470944 ns/iter (+/- 7109)
test bench_noop_big ... bench: 568944 ns/iter (+/- 6895)
test bench_rust_naive_big ... bench: 6795901 ns/iter (+/- 43806)
test bench_rust_unsafe_big ... bench: 5584879 ns/iter (+/- 5291)
```
```asm
movzbl %dil, %eax
bsfl %eax, %eax
movl $32, %ecx
cmovnel %eax, %ecx
cmpl $32, %ecx
movl $8, %eax
cmovnel %ecx, %eax
```
which has some unnecessary overhead, having two conditional moves.
To improve the codegen, we can zero extend the 8 bit integer, then set
bit 8 and perform a cttz operation on the extended value. That way
there's no conditional operation involved at all.
core: Fix size_hint for signed integer `Range<T>` iterators
There was an overflow bug in .size_hint() for signed iterators, which
produced an hilariously incorrect size or an overflow panic.
Incorrect size is a serious bug since the iterators are marked
ExactSizeIterator. (And leads to abort() on (-1i8..127).collect() when
the collection tries to preallocate too much).
> (-1i8..127).size_hint()
(18446744073709551488, Some(18446744073709551488))
Bug found using quickcheck.
Fixes#24851
Instead of using the O(n) defaults, define O(1) shortcuts. I also copied (and slightly modified) the relevant tests from the iter tests into the slice tests just in case someone comes along and changes them in the future.
Partially implements #24214.
There was an overflow bug in .size_hint() for signed iterators, which
produced an hilariously incorrect size or an overflow panic.
Incorrect size is a serious bug since the iterators are marked
ExactSizeIterator. (And leads to abort() on (-1i8..127).collect() when
the collection tries to preallocate too much).
All signed range iterators were affected.
> (-1i8..127).size_hint()
(18446744073709551488, Some(18446744073709551488))
Bug found using quickcheck.
Fixes#24851
These new intrinsics are comparable to `atomic_signal_fence` in C++,
ensuring the compiler will not reorder memory accesses across the
barrier, nor will it emit any machine instructions for it.
Closes#24118, implementing RFC 888.
1. Use next_back for last.
2. Use slices for computing nth. It might be possible to use the same
code for both the mut/const case but I don't know how that will play
with compiler optimizations.
This is an implementation of [RFC 1030][rfc] which adds these traits to the
prelude and additionally removes all inherent `into_iter` methods on collections
in favor of the trait implementation (which is now accessible by default).
[rfc]: https://github.com/rust-lang/rfcs/pull/1030
This is technically a breaking change due to the prelude additions and removal
of inherent methods, but it is expected that essentially no code breaks in
practice.
[breaking-change]
Closes#24538
as accepted in [RFC 526](https://github.com/rust-lang/rfcs/blob/master/text/0526-fmt-text-writer.md).
Note that this brand new method is marked as **stable**. I judged this safe enough: it’s simple enough that it’s very unlikely to change. Still, I can mark it unstable instead if you prefer.
r? @alexcrichton
This commit removes all the old casting/generic traits from `std::num` that are
no longer in use by the standard library. This additionally removes the old
`strconv` module which has not seen much use in quite a long time. All generic
functionality has been supplanted with traits in the `num` crate and the
`strconv` module is supplanted with the [rust-strconv crate][rust-strconv].
[rust-strconv]: https://github.com/lifthrasiir/rust-strconv
This is a breaking change due to the removal of these deprecated crates, and the
alternative crates are listed above.
[breaking-change]
Fixes#24173
These docs could all use examples, so for now, let's just remove the bad one, and when I go over this whole module I'll put in better ones.
This patch
1. renames libunicode to librustc_unicode,
2. deprecates several pieces of libunicode (see below), and
3. removes references to deprecated functions from
librustc_driver and libsyntax. This may change pretty-printed
output from these modules in cases involving wide or combining
characters used in filenames, identifiers, etc.
The following functions are marked deprecated:
1. char.width() and str.width():
--> use unicode-width crate
2. str.graphemes() and str.grapheme_indices():
--> use unicode-segmentation crate
3. str.nfd_chars(), str.nfkd_chars(), str.nfc_chars(), str.nfkc_chars(),
char.compose(), char.decompose_canonical(), char.decompose_compatible(),
char.canonical_combining_class():
--> use unicode-normalization crate
This is an implementation of [RFC 1030][rfc] which adds these traits to the
prelude and additionally removes all inherent `into_iter` methods on collections
in favor of the trait implementation (which is now accessible by default).
[rfc]: https://github.com/rust-lang/rfcs/pull/1030
This is technically a breaking change due to the prelude additions and removal
of inherent methods, but it is expected that essentially no code breaks in
practice.
[breaking-change]
Closes#24538
Add conditional overflow-checking to signed negate operator.
I argue this can land independently of #24420 , because one can write the implementation of `wrapped_neg()` inline if necessary (as illustrated in two cases on this PR).
This needs to go into beta channel.
Alex Crichton