Optimize Integer formatting/parsing #27110

arthurprs · 2015-07-18T15:47:53Z

I wrote a reasonably optimized version for both functions. Further optimizations are possible but I tried to keep the code size small (which I think is important), it's a road of diminished gains.

The repository used for testing/benchmarks is https://github.com/arthurprs/rust-optimized-inttostr

Benchmarks are ran for 3 different distributions, bellow are string length histograms for the u32 type

_h (big numbers skew)
[0, 0, 5, 29, 103, 212, 551, 1138, 1887, 3196, 2879]
_m (slight small number skew):
[0, 2807, 1334, 1057, 905, 821, 772, 707, 627, 605, 365]
_l (small numbers skew):
[0, 8004, 567, 351, 248, 212, 170, 126, 136, 112, 74]

Tested processors are

x64 laptop (i7-2670QM)
x32 server (Digital Ocean E5-2630L-v2)

Display

It uses a small look up table (200 bytes) and decode up to 4 characters at a time. I also took special precautions to reduce 64bit arithmetic on 32bit architectures and the gains are huge in these cases.

Overall, on modern 64bit CPUs it's pretty much the same speed as the stdlib implementation for very small numbers (0..99), but pulls ahead as the length of the decimal increases. On slight older CPUs (w/ worse ALUs) or 32bit architectures it's pretty much always faster.

x64 benchmarks

test bench::display_h_new_u08     ... bench:      71,041 ns/iter (+/- 2,894)
test bench::display_h_new_u16     ... bench:     378,255 ns/iter (+/- 36,547)
test bench::display_h_new_u32     ... bench:   4,232,483 ns/iter (+/- 509,661)
test bench::display_h_new_u64     ... bench:   5,166,740 ns/iter (+/- 421,124)
test bench::display_h_stdlib_u08  ... bench:      73,536 ns/iter (+/- 5,287)
test bench::display_h_stdlib_u16  ... bench:     451,443 ns/iter (+/- 16,879)
test bench::display_h_stdlib_u32  ... bench:   5,551,070 ns/iter (+/- 518,151)
test bench::display_h_stdlib_u64  ... bench:   8,624,374 ns/iter (+/- 643,701)
test bench::display_l_new_u08     ... bench:      71,547 ns/iter (+/- 504)
test bench::display_l_new_u16     ... bench:     399,727 ns/iter (+/- 28,030)
test bench::display_l_new_u32     ... bench:   4,365,303 ns/iter (+/- 414,414)
test bench::display_l_new_u64     ... bench:   5,302,382 ns/iter (+/- 292,324)
test bench::display_l_stdlib_u08  ... bench:      75,445 ns/iter (+/- 2,487)
test bench::display_l_stdlib_u16  ... bench:     444,313 ns/iter (+/- 16,203)
test bench::display_l_stdlib_u32  ... bench:   5,761,801 ns/iter (+/- 387,186)
test bench::display_l_stdlib_u64  ... bench:   8,790,365 ns/iter (+/- 614,846)
test bench::display_m_new_u08     ... bench:      71,820 ns/iter (+/- 2,956)
test bench::display_m_new_u16     ... bench:     399,649 ns/iter (+/- 20,643)
test bench::display_m_new_u32     ... bench:   4,355,561 ns/iter (+/- 179,189)
test bench::display_m_new_u64     ... bench:   5,070,594 ns/iter (+/- 341,950)
test bench::display_m_stdlib_u08  ... bench:      74,900 ns/iter (+/- 1,909)
test bench::display_m_stdlib_u16  ... bench:     448,788 ns/iter (+/- 20,791)
test bench::display_m_stdlib_u32  ... bench:   5,717,939 ns/iter (+/- 316,824)
test bench::display_m_stdlib_u64  ... bench:   8,787,160 ns/iter (+/- 482,864)

x86 benchmarks

test bench::display_h_new_u08     ... bench:      94,246 ns/iter (+/- 34,872)
test bench::display_h_new_u16     ... bench:     533,805 ns/iter (+/- 22,499)
test bench::display_h_new_u32     ... bench:   6,127,747 ns/iter (+/- 2,192,789)
test bench::display_h_new_u64     ... bench:  14,994,203 ns/iter (+/- 1,609,345)
test bench::display_h_stdlib_u08  ... bench:     107,233 ns/iter (+/- 8,571)
test bench::display_h_stdlib_u16  ... bench:     631,186 ns/iter (+/- 11,332)
test bench::display_h_stdlib_u32  ... bench:   7,696,344 ns/iter (+/- 957,917)
test bench::display_h_stdlib_u64  ... bench:  45,677,401 ns/iter (+/- 4,991,344)
test bench::display_l_new_u08     ... bench:      95,855 ns/iter (+/- 27,735)
test bench::display_l_new_u16     ... bench:     532,084 ns/iter (+/- 40,479)
test bench::display_l_new_u32     ... bench:   5,973,953 ns/iter (+/- 211,676)
test bench::display_l_new_u64     ... bench:  14,773,064 ns/iter (+/- 1,276,579)
test bench::display_l_stdlib_u08  ... bench:     106,350 ns/iter (+/- 63,963)
test bench::display_l_stdlib_u16  ... bench:     637,746 ns/iter (+/- 101,005)
test bench::display_l_stdlib_u32  ... bench:   7,740,640 ns/iter (+/- 848,478)
test bench::display_l_stdlib_u64  ... bench:  44,846,932 ns/iter (+/- 4,514,694)
test bench::display_m_new_u08     ... bench:      94,549 ns/iter (+/- 13,029)
test bench::display_m_new_u16     ... bench:     546,030 ns/iter (+/- 35,804)
test bench::display_m_new_u32     ... bench:   5,983,924 ns/iter (+/- 1,180,559)
test bench::display_m_new_u64     ... bench:  14,817,873 ns/iter (+/- 2,271,464)
test bench::display_m_stdlib_u08  ... bench:     107,806 ns/iter (+/- 8,805)
test bench::display_m_stdlib_u16  ... bench:     630,714 ns/iter (+/- 6,586)
test bench::display_m_stdlib_u32  ... bench:   7,784,210 ns/iter (+/- 358,601)
test bench::display_m_stdlib_u64  ... bench:  46,223,927 ns/iter (+/- 6,553,176)

from_str_radix (FromStr)

All valid digits are ascii so I modified the function to use the underlining bytes instead and simplified the match to avoid wasting cycles.

x64 benchmarks

test bench::from_str_h_new_u08    ... bench:      28,153 ns/iter (+/- 624)
test bench::from_str_h_new_u16    ... bench:     223,513 ns/iter (+/- 11,554)
test bench::from_str_h_new_u32    ... bench:   3,098,935 ns/iter (+/- 231,022)
test bench::from_str_h_new_u64    ... bench:   5,009,900 ns/iter (+/- 341,961)
test bench::from_str_h_stdlib_u08 ... bench:      34,033 ns/iter (+/- 2,068)
test bench::from_str_h_stdlib_u16 ... bench:     248,785 ns/iter (+/- 14,208)
test bench::from_str_h_stdlib_u32 ... bench:   4,150,536 ns/iter (+/- 266,070)
test bench::from_str_h_stdlib_u64 ... bench:   6,817,997 ns/iter (+/- 449,838)
test bench::from_str_l_new_u08    ... bench:      27,552 ns/iter (+/- 1,500)
test bench::from_str_l_new_u16    ... bench:     234,360 ns/iter (+/- 13,144)
test bench::from_str_l_new_u32    ... bench:   3,140,261 ns/iter (+/- 248,175)
test bench::from_str_l_new_u64    ... bench:   5,176,583 ns/iter (+/- 350,416)
test bench::from_str_l_stdlib_u08 ... bench:      35,060 ns/iter (+/- 2,154)
test bench::from_str_l_stdlib_u16 ... bench:     252,135 ns/iter (+/- 23,461)
test bench::from_str_l_stdlib_u32 ... bench:   4,154,599 ns/iter (+/- 369,606)
test bench::from_str_l_stdlib_u64 ... bench:   6,892,767 ns/iter (+/- 213,030)
test bench::from_str_m_new_u08    ... bench:      28,252 ns/iter (+/- 1,384)
test bench::from_str_m_new_u16    ... bench:     231,051 ns/iter (+/- 16,540)
test bench::from_str_m_new_u32    ... bench:   3,166,504 ns/iter (+/- 134,418)
test bench::from_str_m_new_u64    ... bench:   5,103,195 ns/iter (+/- 218,912)
test bench::from_str_m_stdlib_u08 ... bench:      35,012 ns/iter (+/- 2,735)
test bench::from_str_m_stdlib_u16 ... bench:     250,967 ns/iter (+/- 14,708)
test bench::from_str_m_stdlib_u32 ... bench:   4,101,845 ns/iter (+/- 205,802)
test bench::from_str_m_stdlib_u64 ... bench:   6,823,001 ns/iter (+/- 267,215)

x86 benchmarks

test bench::from_str_h_new_u08    ... bench:      23,682 ns/iter (+/- 3,590)
test bench::from_str_h_new_u16    ... bench:     190,916 ns/iter (+/- 29,688)
test bench::from_str_h_new_u32    ... bench:   2,649,952 ns/iter (+/- 308,576)
test bench::from_str_h_new_u64    ... bench:  23,458,434 ns/iter (+/- 2,327,427)
test bench::from_str_h_stdlib_u08 ... bench:      45,551 ns/iter (+/- 6,968)
test bench::from_str_h_stdlib_u16 ... bench:     313,739 ns/iter (+/- 17,175)
test bench::from_str_h_stdlib_u32 ... bench:   4,615,669 ns/iter (+/- 470,766)
test bench::from_str_h_stdlib_u64 ... bench:  30,589,482 ns/iter (+/- 2,278,996)
test bench::from_str_l_new_u08    ... bench:      23,763 ns/iter (+/- 5,545)
test bench::from_str_l_new_u16    ... bench:     185,472 ns/iter (+/- 33,097)
test bench::from_str_l_new_u32    ... bench:   2,691,307 ns/iter (+/- 473,886)
test bench::from_str_l_new_u64    ... bench:  22,952,593 ns/iter (+/- 1,963,742)
test bench::from_str_l_stdlib_u08 ... bench:      45,285 ns/iter (+/- 16,337)
test bench::from_str_l_stdlib_u16 ... bench:     313,624 ns/iter (+/- 6,643)
test bench::from_str_l_stdlib_u32 ... bench:   4,595,679 ns/iter (+/- 1,876,361)
test bench::from_str_l_stdlib_u64 ... bench:  30,434,683 ns/iter (+/- 1,901,996)
test bench::from_str_m_new_u08    ... bench:      23,812 ns/iter (+/- 1,505)
test bench::from_str_m_new_u16    ... bench:     185,553 ns/iter (+/- 19,788)
test bench::from_str_m_new_u32    ... bench:   2,614,920 ns/iter (+/- 66,230)
test bench::from_str_m_new_u64    ... bench:  23,241,778 ns/iter (+/- 3,474,077)
test bench::from_str_m_stdlib_u08 ... bench:      45,634 ns/iter (+/- 1,436)
test bench::from_str_m_stdlib_u16 ... bench:     316,479 ns/iter (+/- 21,212)
test bench::from_str_m_stdlib_u32 ... bench:   4,609,147 ns/iter (+/- 487,068)
test bench::from_str_m_stdlib_u64 ... bench:  30,165,173 ns/iter (+/- 1,601,830)

rust-highfive · 2015-07-18T15:48:06Z

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @huonw (or someone else) soon.

If any changes to this PR are deemed necessary, please add them as extra commits. This ensures that the reviewer can see what has changed since they last reviewed the code. The way Github handles out-of-date commits, this should also make it reasonably obvious what issues have or haven't been addressed. Large or tricky changes may require several passes of review and changes.

Please see the contribution instructions for more information.

Gankra · 2015-07-18T16:15:05Z

src/libcore/fmt/num.rs

@@ -233,3 +238,82 @@ integer! { i8, u8 }
 integer! { i16, u16 }
 integer! { i32, u32 }
 integer! { i64, u64 }
+
+const DEC_DIGITS_LUT: [u8; 200] = [
+    '0' as u8,'0' as u8,'0' as u8,'1' as u8,'0' as u8,'2' as u8,'0' as u8,'3' as u8,'0' as u8,'4' as u8,'0' as u8,'5' as u8,'0' as u8,'6' as u8,'0' as u8,'7' as u8,'0' as u8,'8' as u8,'0' as u8,'9' as u8,


Is this something a macro or const fn could make...?

At very least, I feel like you probably want:

static DEC_DIGITS_LUT: &'static [u8; 200] = b"00010203...";

regardless all of these lines fail the style lint (> 100 lines)

Gankra · 2015-07-18T16:17:23Z

So I know we went through a lot of effort to basically get top-of-the-line industrial float formatting. I imagine int formatting is a bit easier, but I'd be remiss to not ask: are there standard "optimal" algorithms that we should be investing in for this?

arthurprs · 2015-07-18T16:34:11Z

@gankro what triggered me to do this work was this repository https://github.com/miloyip/itoa-benchmark which is a good summary of what's out there.

What I have in this PR can be seen as a hybrid of the lut and branchlut versions. My version tries to keep code size small and avoid regressions for small numbers. In addition to that it also reduces 64 arithmetic.

Also note that the biggest increase in speed from lut to branchlut/countlut in that repo are due to avoiding the double buffering, which we can't really avoid in rust due to the fmt machinery.

Gankra · 2015-07-18T16:41:30Z

CC @lifthrasiir

Gankra · 2015-07-18T17:05:01Z

src/libcore/fmt/num.rs

+        #[allow(unused_comparisons)]
+        fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+            let is_positive = *self >= 0;
+            let mut n = if ! is_positive {


style nit, we put the ! right up in the condition's grill

Also presumably you could kill the negation and just reorder the branches.

Gankra · 2015-07-18T17:41:21Z

I'm pretty sure this merits some new tests for interesting corner-cases.

arthurprs · 2015-07-18T22:05:53Z

@gankro thank you very much for your review. I cleaned up the code to account for your suggestions and removed the last bounds checking.

As for new tests I'm not aware of any border cases other than non-digits characters (added) and overflows (already present).

Gankra · 2015-07-18T22:15:32Z

src/libcore/fmt/num.rs

+            }
+
+            let buf_slice = unsafe {
+                str::from_utf8_unchecked(slice::from_raw_parts(buf_ptr.offset(curr)), buf.len() - curr as usize)


Looks like you got your parens wrong here

Also this line is too long (see travis)

arthurprs · 2015-07-18T23:27:46Z

adjusted the nits and parens, travis is green.

Gankra · 2015-07-18T23:32:35Z

This is looking great! The only outstanding issue is @eefriedman's concern about the char conversion, which I don't have strong feelings about.

arthurprs · 2015-07-18T23:51:56Z

I can't find @eefriedman comment here, just in my inbox.

to_digit returns None if the char can't be converted to a number of the specified radix, it sounds simple enough to me. If you guys want I can add the wrapper function with some comments though.

eefriedman · 2015-07-19T00:07:41Z

GitHub collapses comments against old changes; that's probably why you're not finding it.

The problem isn't that to_digit is weird, it's that converting a UTF-8 byte to a Rust char is weird. Might be enough to just expand the "all valid digits are ASCII" comment a bit.

arthurprs · 2015-07-19T01:43:32Z

Updated comment

    // all valid digits are ascii, so we will just iterate over the utf8 bytes
    // and cast them to chars. .to_digit() will safely return None for anything
    // other than a valid ascii digit for a the given radix, including the first-byte
    // of multi-byte sequences.

Gankra · 2015-07-19T02:12:26Z

@bors r+

Awesome stuff!

bors · 2015-07-19T02:12:27Z

📌 Commit 647f87f has been approved by Gankro

bors · 2015-07-19T02:12:30Z

⌛ Testing commit 647f87f with merge b806a4c...

tamird · 2015-07-19T02:29:01Z

src/libcore/fmt/num.rs

+                }
+
+                // if we reach here numbers are <= 9999, so at most 4 chars long
+                let mut n = n as isize; // possibly reduce 64bit math


should this be i64 rather than isize?

nm, i think this is ok

Yeah it's fine specifically because we know n <= 9999

Gankra · 2015-07-19T02:45:10Z

Ehh the build already started. Unless it bounces spuriously (which is not unheard of), just submit a followup pr to remove the annotation.

Such is the danger of weekend PRs... no queue!

bors · 2015-07-19T02:56:07Z

💔 Test failed - auto-mac-64-opt

Gankra · 2015-07-19T02:58:10Z

Non spurious failures even:


rustc: x86_64-apple-darwin/stage2/test/coretesttest-x86_64-apple-darwin
../src/libcoretest/num/mod.rs:121:47: 121:60 error: `ParseIntError` does not name a structure [E0422]
../src/libcoretest/num/mod.rs:121         assert_eq!("--129".parse::<i8>(), Err(ParseIntError{ kind: IntErrorKind::InvalidDigit }));
                                                                                ^~~~~~~~~~~~~
<std macros>:1:1: 9:39 note: in expansion of assert_eq!
../src/libcoretest/num/mod.rs:121:9: 121:99 note: expansion site
../src/libcoretest/num/mod.rs:121:68: 121:94 error: failed to resolve. Use of undeclared type or module `IntErrorKind` [E0433]
../src/libcoretest/num/mod.rs:121         assert_eq!("--129".parse::<i8>(), Err(ParseIntError{ kind: IntErrorKind::InvalidDigit }));
                                                                                                     ^~~~~~~~~~~~~~~~~~~~~~~~~~
<std macros>:1:1: 9:39 note: in expansion of assert_eq!
../src/libcoretest/num/mod.rs:121:9: 121:99 note: expansion site
../src/libcoretest/num/mod.rs:121:68: 121:94 error: unresolved name `IntErrorKind::InvalidDigit` [E0425]
../src/libcoretest/num/mod.rs:121         assert_eq!("--129".parse::<i8>(), Err(ParseIntError{ kind: IntErrorKind::InvalidDigit }));
                                                                                                     ^~~~~~~~~~~~~~~~~~~~~~~~~~
<std macros>:1:1: 9:39 note: in expansion of assert_eq!
../src/libcoretest/num/mod.rs:121:9: 121:99 note: expansion site
../src/libcoretest/num/mod.rs:122:47: 122:60 error: `ParseIntError` does not name a structure [E0422]
../src/libcoretest/num/mod.rs:122         assert_eq!("Съешь".parse::<u8>(), Err(ParseIntError{ kind: IntErrorKind::InvalidDigit }));
                                                                                ^~~~~~~~~~~~~
<std macros>:1:1: 9:39 note: in expansion of assert_eq!
../src/libcoretest/num/mod.rs:122:9: 122:99 note: expansion site
../src/libcoretest/num/mod.rs:122:68: 122:94 error: failed to resolve. Use of undeclared type or module `IntErrorKind` [E0433]
../src/libcoretest/num/mod.rs:122         assert_eq!("Съешь".parse::<u8>(), Err(ParseIntError{ kind: IntErrorKind::InvalidDigit }));
                                                                                                     ^~~~~~~~~~~~~~~~~~~~~~~~~~
<std macros>:1:1: 9:39 note: in expansion of assert_eq!
../src/libcoretest/num/mod.rs:122:9: 122:99 note: expansion site
../src/libcoretest/num/mod.rs:122:68: 122:94 error: unresolved name `IntErrorKind::InvalidDigit` [E0425]
../src/libcoretest/num/mod.rs:122         assert_eq!("Съешь".parse::<u8>(), Err(ParseIntError{ kind: IntErrorKind::InvalidDigit }));
                                                                                                     ^~~~~~~~~~~~~~~~~~~~~~~~~~
<std macros>:1:1: 9:39 note: in expansion of assert_eq!
../src/libcoretest/num/mod.rs:122:9: 122:99 note: expansion site
../src/libcoretest/num/mod.rs:127:43: 127:56 error: `ParseIntError` does not name a structure [E0422]
../src/libcoretest/num/mod.rs:127         assert_eq!("-".parse::<i8>(), Err(ParseIntError{ kind: IntErrorKind::Empty }));
                                                                            ^~~~~~~~~~~~~
<std macros>:1:1: 9:39 note: in expansion of assert_eq!
../src/libcoretest/num/mod.rs:127:9: 127:88 note: expansion site
../src/libcoretest/num/mod.rs:127:64: 127:83 error: failed to resolve. Use of undeclared type or module `IntErrorKind` [E0433]
../src/libcoretest/num/mod.rs:127         assert_eq!("-".parse::<i8>(), Err(ParseIntError{ kind: IntErrorKind::Empty }));
                                                                                                 ^~~~~~~~~~~~~~~~~~~
<std macros>:1:1: 9:39 note: in expansion of assert_eq!
../src/libcoretest/num/mod.rs:127:9: 127:88 note: expansion site
../src/libcoretest/num/mod.rs:127:64: 127:83 error: unresolved name `IntErrorKind::Empty` [E0425]
../src/libcoretest/num/mod.rs:127         assert_eq!("-".parse::<i8>(), Err(ParseIntError{ kind: IntErrorKind::Empty }));
                                                                                                 ^~~~~~~~~~~~~~~~~~~
<std macros>:1:1: 9:39 note: in expansion of assert_eq!
../src/libcoretest/num/mod.rs:127:9: 127:88 note: expansion site
../src/libcoretest/num/mod.rs:128:42: 128:55 error: `ParseIntError` does not name a structure [E0422]
../src/libcoretest/num/mod.rs:128         assert_eq!("".parse::<u8>(), Err(ParseIntError{ kind: IntErrorKind::Empty }));
                                                                           ^~~~~~~~~~~~~
<std macros>:1:1: 9:39 note: in expansion of assert_eq!
../src/libcoretest/num/mod.rs:128:9: 128:87 note: expansion site
../src/libcoretest/num/mod.rs:128:63: 128:82 error: failed to resolve. Use of undeclared type or module `IntErrorKind` [E0433]
../src/libcoretest/num/mod.rs:128         assert_eq!("".parse::<u8>(), Err(ParseIntError{ kind: IntErrorKind::Empty }));
                                                                                                ^~~~~~~~~~~~~~~~~~~
<std macros>:1:1: 9:39 note: in expansion of assert_eq!
../src/libcoretest/num/mod.rs:128:9: 128:87 note: expansion site
../src/libcoretest/num/mod.rs:128:63: 128:82 error: unresolved name `IntErrorKind::Empty` [E0425]
../src/libcoretest/num/mod.rs:128         assert_eq!("".parse::<u8>(), Err(ParseIntError{ kind: IntErrorKind::Empty }));
                                                                                                ^~~~~~~~~~~~~~~~~~~
<std macros>:1:1: 9:39 note: in expansion of assert_eq!
../src/libcoretest/num/mod.rs:128:9: 128:87 note: expansion site
error: aborting due to 12 previous errors
make: *** [x86_64-apple-darwin/stage2/test/coretesttest-x86_64-apple-darwin] Error 101

arthurprs · 2015-07-19T12:57:50Z

Fixed. Sorry, I screwed up the tests, I have to get a better machine so I can compile/test rustc in doable time.

Gankra · 2015-07-19T16:36:09Z

No worries!

We all cut corners testing PRs. :D

@bors r+

bors · 2015-07-19T16:36:10Z

📌 Commit c073f81 has been approved by Gankro

I wrote a reasonably optimized version for both functions. Further optimizations are possible but I tried to keep the code size small (which I think is important), it's a road of diminished gains. The repository used for testing/benchmarks is https://github.com/arthurprs/rust-optimized-inttostr Benchmarks are ran for 3 different distributions, bellow are string length histograms for the u32 type * _h (big numbers skew) [0, 0, 5, 29, 103, 212, 551, 1138, 1887, 3196, 2879] * _m (slight small number skew): [0, 2807, 1334, 1057, 905, 821, 772, 707, 627, 605, 365] * _l (small numbers skew): [0, 8004, 567, 351, 248, 212, 170, 126, 136, 112, 74] Tested processors are * x64 laptop (i7-2670QM) * x32 server (Digital Ocean E5-2630L-v2) ### Display It uses a small look up table (200 bytes) and decode up to 4 characters at a time. I also took special precautions to reduce 64bit arithmetic on 32bit architectures and the gains are huge in these cases. Overall, on modern 64bit CPUs it's pretty much the same speed as the stdlib implementation for very small numbers (0..99), but pulls ahead as the length of the decimal increases. On slight older CPUs (w/ worse ALUs) or 32bit architectures it's pretty much always faster. x64 benchmarks ``` test bench::display_h_new_u08 ... bench: 71,041 ns/iter (+/- 2,894) test bench::display_h_new_u16 ... bench: 378,255 ns/iter (+/- 36,547) test bench::display_h_new_u32 ... bench: 4,232,483 ns/iter (+/- 509,661) test bench::display_h_new_u64 ... bench: 5,166,740 ns/iter (+/- 421,124) test bench::display_h_stdlib_u08 ... bench: 73,536 ns/iter (+/- 5,287) test bench::display_h_stdlib_u16 ... bench: 451,443 ns/iter (+/- 16,879) test bench::display_h_stdlib_u32 ... bench: 5,551,070 ns/iter (+/- 518,151) test bench::display_h_stdlib_u64 ... bench: 8,624,374 ns/iter (+/- 643,701) test bench::display_l_new_u08 ... bench: 71,547 ns/iter (+/- 504) test bench::display_l_new_u16 ... bench: 399,727 ns/iter (+/- 28,030) test bench::display_l_new_u32 ... bench: 4,365,303 ns/iter (+/- 414,414) test bench::display_l_new_u64 ... bench: 5,302,382 ns/iter (+/- 292,324) test bench::display_l_stdlib_u08 ... bench: 75,445 ns/iter (+/- 2,487) test bench::display_l_stdlib_u16 ... bench: 444,313 ns/iter (+/- 16,203) test bench::display_l_stdlib_u32 ... bench: 5,761,801 ns/iter (+/- 387,186) test bench::display_l_stdlib_u64 ... bench: 8,790,365 ns/iter (+/- 614,846) test bench::display_m_new_u08 ... bench: 71,820 ns/iter (+/- 2,956) test bench::display_m_new_u16 ... bench: 399,649 ns/iter (+/- 20,643) test bench::display_m_new_u32 ... bench: 4,355,561 ns/iter (+/- 179,189) test bench::display_m_new_u64 ... bench: 5,070,594 ns/iter (+/- 341,950) test bench::display_m_stdlib_u08 ... bench: 74,900 ns/iter (+/- 1,909) test bench::display_m_stdlib_u16 ... bench: 448,788 ns/iter (+/- 20,791) test bench::display_m_stdlib_u32 ... bench: 5,717,939 ns/iter (+/- 316,824) test bench::display_m_stdlib_u64 ... bench: 8,787,160 ns/iter (+/- 482,864) ``` x86 benchmarks ``` test bench::display_h_new_u08 ... bench: 94,246 ns/iter (+/- 34,872) test bench::display_h_new_u16 ... bench: 533,805 ns/iter (+/- 22,499) test bench::display_h_new_u32 ... bench: 6,127,747 ns/iter (+/- 2,192,789) test bench::display_h_new_u64 ... bench: 14,994,203 ns/iter (+/- 1,609,345) test bench::display_h_stdlib_u08 ... bench: 107,233 ns/iter (+/- 8,571) test bench::display_h_stdlib_u16 ... bench: 631,186 ns/iter (+/- 11,332) test bench::display_h_stdlib_u32 ... bench: 7,696,344 ns/iter (+/- 957,917) test bench::display_h_stdlib_u64 ... bench: 45,677,401 ns/iter (+/- 4,991,344) test bench::display_l_new_u08 ... bench: 95,855 ns/iter (+/- 27,735) test bench::display_l_new_u16 ... bench: 532,084 ns/iter (+/- 40,479) test bench::display_l_new_u32 ... bench: 5,973,953 ns/iter (+/- 211,676) test bench::display_l_new_u64 ... bench: 14,773,064 ns/iter (+/- 1,276,579) test bench::display_l_stdlib_u08 ... bench: 106,350 ns/iter (+/- 63,963) test bench::display_l_stdlib_u16 ... bench: 637,746 ns/iter (+/- 101,005) test bench::display_l_stdlib_u32 ... bench: 7,740,640 ns/iter (+/- 848,478) test bench::display_l_stdlib_u64 ... bench: 44,846,932 ns/iter (+/- 4,514,694) test bench::display_m_new_u08 ... bench: 94,549 ns/iter (+/- 13,029) test bench::display_m_new_u16 ... bench: 546,030 ns/iter (+/- 35,804) test bench::display_m_new_u32 ... bench: 5,983,924 ns/iter (+/- 1,180,559) test bench::display_m_new_u64 ... bench: 14,817,873 ns/iter (+/- 2,271,464) test bench::display_m_stdlib_u08 ... bench: 107,806 ns/iter (+/- 8,805) test bench::display_m_stdlib_u16 ... bench: 630,714 ns/iter (+/- 6,586) test bench::display_m_stdlib_u32 ... bench: 7,784,210 ns/iter (+/- 358,601) test bench::display_m_stdlib_u64 ... bench: 46,223,927 ns/iter (+/- 6,553,176) ``` ### from_str_radix (FromStr) All valid digits are ascii so I modified the function to use the underlining bytes instead and simplified the match to avoid wasting cycles. x64 benchmarks ``` test bench::from_str_h_new_u08 ... bench: 28,153 ns/iter (+/- 624) test bench::from_str_h_new_u16 ... bench: 223,513 ns/iter (+/- 11,554) test bench::from_str_h_new_u32 ... bench: 3,098,935 ns/iter (+/- 231,022) test bench::from_str_h_new_u64 ... bench: 5,009,900 ns/iter (+/- 341,961) test bench::from_str_h_stdlib_u08 ... bench: 34,033 ns/iter (+/- 2,068) test bench::from_str_h_stdlib_u16 ... bench: 248,785 ns/iter (+/- 14,208) test bench::from_str_h_stdlib_u32 ... bench: 4,150,536 ns/iter (+/- 266,070) test bench::from_str_h_stdlib_u64 ... bench: 6,817,997 ns/iter (+/- 449,838) test bench::from_str_l_new_u08 ... bench: 27,552 ns/iter (+/- 1,500) test bench::from_str_l_new_u16 ... bench: 234,360 ns/iter (+/- 13,144) test bench::from_str_l_new_u32 ... bench: 3,140,261 ns/iter (+/- 248,175) test bench::from_str_l_new_u64 ... bench: 5,176,583 ns/iter (+/- 350,416) test bench::from_str_l_stdlib_u08 ... bench: 35,060 ns/iter (+/- 2,154) test bench::from_str_l_stdlib_u16 ... bench: 252,135 ns/iter (+/- 23,461) test bench::from_str_l_stdlib_u32 ... bench: 4,154,599 ns/iter (+/- 369,606) test bench::from_str_l_stdlib_u64 ... bench: 6,892,767 ns/iter (+/- 213,030) test bench::from_str_m_new_u08 ... bench: 28,252 ns/iter (+/- 1,384) test bench::from_str_m_new_u16 ... bench: 231,051 ns/iter (+/- 16,540) test bench::from_str_m_new_u32 ... bench: 3,166,504 ns/iter (+/- 134,418) test bench::from_str_m_new_u64 ... bench: 5,103,195 ns/iter (+/- 218,912) test bench::from_str_m_stdlib_u08 ... bench: 35,012 ns/iter (+/- 2,735) test bench::from_str_m_stdlib_u16 ... bench: 250,967 ns/iter (+/- 14,708) test bench::from_str_m_stdlib_u32 ... bench: 4,101,845 ns/iter (+/- 205,802) test bench::from_str_m_stdlib_u64 ... bench: 6,823,001 ns/iter (+/- 267,215) ``` x86 benchmarks ``` test bench::from_str_h_new_u08 ... bench: 23,682 ns/iter (+/- 3,590) test bench::from_str_h_new_u16 ... bench: 190,916 ns/iter (+/- 29,688) test bench::from_str_h_new_u32 ... bench: 2,649,952 ns/iter (+/- 308,576) test bench::from_str_h_new_u64 ... bench: 23,458,434 ns/iter (+/- 2,327,427) test bench::from_str_h_stdlib_u08 ... bench: 45,551 ns/iter (+/- 6,968) test bench::from_str_h_stdlib_u16 ... bench: 313,739 ns/iter (+/- 17,175) test bench::from_str_h_stdlib_u32 ... bench: 4,615,669 ns/iter (+/- 470,766) test bench::from_str_h_stdlib_u64 ... bench: 30,589,482 ns/iter (+/- 2,278,996) test bench::from_str_l_new_u08 ... bench: 23,763 ns/iter (+/- 5,545) test bench::from_str_l_new_u16 ... bench: 185,472 ns/iter (+/- 33,097) test bench::from_str_l_new_u32 ... bench: 2,691,307 ns/iter (+/- 473,886) test bench::from_str_l_new_u64 ... bench: 22,952,593 ns/iter (+/- 1,963,742) test bench::from_str_l_stdlib_u08 ... bench: 45,285 ns/iter (+/- 16,337) test bench::from_str_l_stdlib_u16 ... bench: 313,624 ns/iter (+/- 6,643) test bench::from_str_l_stdlib_u32 ... bench: 4,595,679 ns/iter (+/- 1,876,361) test bench::from_str_l_stdlib_u64 ... bench: 30,434,683 ns/iter (+/- 1,901,996) test bench::from_str_m_new_u08 ... bench: 23,812 ns/iter (+/- 1,505) test bench::from_str_m_new_u16 ... bench: 185,553 ns/iter (+/- 19,788) test bench::from_str_m_new_u32 ... bench: 2,614,920 ns/iter (+/- 66,230) test bench::from_str_m_new_u64 ... bench: 23,241,778 ns/iter (+/- 3,474,077) test bench::from_str_m_stdlib_u08 ... bench: 45,634 ns/iter (+/- 1,436) test bench::from_str_m_stdlib_u16 ... bench: 316,479 ns/iter (+/- 21,212) test bench::from_str_m_stdlib_u32 ... bench: 4,609,147 ns/iter (+/- 487,068) test bench::from_str_m_stdlib_u64 ... bench: 30,165,173 ns/iter (+/- 1,601,830) ```

bors · 2015-07-19T19:04:16Z

⌛ Testing commit c073f81 with merge a41178e...

bors · 2015-07-19T20:39:37Z

☀️ Test successful - auto-linux-32-nopt-t, auto-linux-32-opt, auto-linux-64-nopt-t, auto-linux-64-opt, auto-linux-64-x-android-t, auto-mac-32-opt, auto-mac-64-nopt-t, auto-mac-64-opt, auto-win-gnu-32-nopt-t, auto-win-gnu-32-opt, auto-win-gnu-64-nopt-t, auto-win-gnu-64-opt

rust-highfive assigned huonw Jul 18, 2015

Gankra reviewed Jul 18, 2015
View reviewed changes

arthurprs force-pushed the optintstr branch from 5c70fbb to 1f3024a Compare July 18, 2015 16:17

Gankra reviewed Jul 18, 2015
View reviewed changes

arthurprs force-pushed the optintstr branch 2 times, most recently from d868815 to bfa1c8d Compare July 18, 2015 22:01

arthurprs force-pushed the optintstr branch from bfa1c8d to ed5d1b2 Compare July 18, 2015 22:07

Gankra reviewed Jul 18, 2015
View reviewed changes

arthurprs force-pushed the optintstr branch 2 times, most recently from 1443a52 to 5843fe4 Compare July 18, 2015 22:52

optimize integer formatting

ebf9e1a

arthurprs force-pushed the optintstr branch from 5843fe4 to 3eaeea8 Compare July 18, 2015 23:07

arthurprs force-pushed the optintstr branch from 3eaeea8 to 55cc0b8 Compare July 19, 2015 01:41

arthurprs force-pushed the optintstr branch 2 times, most recently from a413ab8 to 647f87f Compare July 19, 2015 01:44

tamird reviewed Jul 19, 2015
View reviewed changes

optimize from_str_radix

c073f81

arthurprs force-pushed the optintstr branch from 647f87f to c073f81 Compare July 19, 2015 12:55

bors merged commit c073f81 into rust-lang:master Jul 19, 2015

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Optimize Integer formatting/parsing #27110

Optimize Integer formatting/parsing #27110

arthurprs commented Jul 18, 2015

rust-highfive commented Jul 18, 2015

Gankra Jul 18, 2015

Gankra Jul 18, 2015

Gankra Jul 18, 2015

Gankra commented Jul 18, 2015

arthurprs commented Jul 18, 2015

Gankra commented Jul 18, 2015

Gankra Jul 18, 2015

Gankra Jul 18, 2015

arthurprs Jul 18, 2015

Gankra commented Jul 18, 2015

arthurprs commented Jul 18, 2015

Gankra Jul 18, 2015

Gankra Jul 18, 2015

arthurprs commented Jul 18, 2015

Gankra commented Jul 18, 2015

arthurprs commented Jul 18, 2015

eefriedman commented Jul 19, 2015

arthurprs commented Jul 19, 2015

Gankra commented Jul 19, 2015

bors commented Jul 19, 2015

bors commented Jul 19, 2015

tamird Jul 19, 2015

tamird Jul 19, 2015

Gankra Jul 19, 2015

Gankra commented Jul 19, 2015

bors commented Jul 19, 2015

Gankra commented Jul 19, 2015

arthurprs commented Jul 19, 2015

Gankra commented Jul 19, 2015

bors commented Jul 19, 2015

bors commented Jul 19, 2015

bors commented Jul 19, 2015

Optimize Integer formatting/parsing #27110

Optimize Integer formatting/parsing #27110

Conversation

arthurprs commented Jul 18, 2015

Display

from_str_radix (FromStr)

rust-highfive commented Jul 18, 2015

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Gankra commented Jul 18, 2015

arthurprs commented Jul 18, 2015

Gankra commented Jul 18, 2015

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Gankra commented Jul 18, 2015

arthurprs commented Jul 18, 2015

Choose a reason for hiding this comment

Choose a reason for hiding this comment

arthurprs commented Jul 18, 2015

Gankra commented Jul 18, 2015

arthurprs commented Jul 18, 2015

eefriedman commented Jul 19, 2015

arthurprs commented Jul 19, 2015

Gankra commented Jul 19, 2015

bors commented Jul 19, 2015

bors commented Jul 19, 2015

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Gankra commented Jul 19, 2015

bors commented Jul 19, 2015

Gankra commented Jul 19, 2015

arthurprs commented Jul 19, 2015

Gankra commented Jul 19, 2015

bors commented Jul 19, 2015

bors commented Jul 19, 2015

bors commented Jul 19, 2015