3a37fd4010
This CL replaces the original subVW implementation with a implementation that uses a similar idea as CL 164968. When we know the borrow bit is zero, we can copy the rest of words as they will not be updated. Also, since we are copying vector of a words, a faster implementation of copy is written in this CL to copy a word or multiple words at a time. Benchmarks: name old time/op new time/op delta SubVW/1-18 4.43ns ± 0% 3.82ns ± 0% -13.85% (p=0.000 n=20+20) SubVW/2-18 5.39ns ± 0% 4.25ns ± 0% -21.23% (p=0.000 n=20+20) SubVW/3-18 6.29ns ± 0% 4.65ns ± 0% -26.07% (p=0.000 n=16+19) SubVW/4-18 6.08ns ± 2% 4.84ns ± 0% -20.43% (p=0.000 n=20+20) SubVW/5-18 7.06ns ± 1% 4.93ns ± 0% -30.18% (p=0.000 n=20+20) SubVW/10-18 10.3ns ± 2% 7.2ns ± 0% -30.35% (p=0.000 n=20+19) SubVW/100-18 48.0ns ± 4% 17.6ns ± 0% -63.32% (p=0.000 n=18+19) SubVW/1000-18 448ns ±10% 236ns ± 1% -47.24% (p=0.000 n=20+20) SubVW/10000-18 4.83µs ± 5% 2.96µs ± 0% -38.73% (p=0.000 n=20+19) SubVW/100000-18 46.6µs ± 3% 30.6µs ± 1% -34.30% (p=0.000 n=20+20) [Geo mean] 56.3ns 37.0ns -34.24% name old speed new speed delta SubVW/1-18 1.80GB/s ± 0% 2.10GB/s ± 0% +16.16% (p=0.000 n=20+20) SubVW/2-18 2.97GB/s ± 0% 3.77GB/s ± 0% +26.95% (p=0.000 n=20+20) SubVW/3-18 3.82GB/s ± 0% 5.16GB/s ± 0% +35.26% (p=0.000 n=20+19) SubVW/4-18 5.26GB/s ± 1% 6.61GB/s ± 0% +25.59% (p=0.000 n=20+20) SubVW/5-18 5.67GB/s ± 1% 8.11GB/s ± 0% +43.12% (p=0.000 n=20+20) SubVW/10-18 7.79GB/s ± 2% 11.17GB/s ± 0% +43.52% (p=0.000 n=20+19) SubVW/100-18 16.7GB/s ± 4% 45.5GB/s ± 0% +172.61% (p=0.000 n=18+20) SubVW/1000-18 17.9GB/s ± 9% 33.9GB/s ± 1% +89.25% (p=0.000 n=20+20) SubVW/10000-18 16.6GB/s ± 5% 27.0GB/s ± 0% +63.08% (p=0.000 n=20+19) SubVW/100000-18 17.2GB/s ± 2% 26.1GB/s ± 1% +52.18% (p=0.000 n=20+20) [Geo mean] 7.25GB/s 11.03GB/s +52.01% Change-Id: I32e99cbab3260054a96231d02b87049c833ab77e Reviewed-on: https://go-review.googlesource.com/c/go/+/227297 Reviewed-by: Michael Munday <mike.munday@ibm.com> Run-TryBot: Michael Munday <mike.munday@ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> |
||
|---|---|---|
| .github | ||
| api | ||
| doc | ||
| lib/time | ||
| misc | ||
| src | ||
| test | ||
| .gitattributes | ||
| .gitignore | ||
| AUTHORS | ||
| CONTRIBUTING.md | ||
| CONTRIBUTORS | ||
| favicon.ico | ||
| LICENSE | ||
| PATENTS | ||
| README.md | ||
| robots.txt | ||
| SECURITY.md | ||
The Go Programming Language
Go is an open source programming language that makes it easy to build simple, reliable, and efficient software.
Gopher image by Renee French, licensed under Creative Commons 3.0 Attributions license.
Our canonical Git repository is located at https://go.googlesource.com/go. There is a mirror of the repository at https://github.com/golang/go.
Unless otherwise noted, the Go source files are distributed under the BSD-style license found in the LICENSE file.
Download and Install
Binary Distributions
Official binary distributions are available at https://golang.org/dl/.
After downloading a binary release, visit https://golang.org/doc/install or load doc/install.html in your web browser for installation instructions.
Install From Source
If a binary distribution is not available for your combination of operating system and architecture, visit https://golang.org/doc/install/source or load doc/install-source.html in your web browser for source installation instructions.
Contributing
Go is the work of thousands of contributors. We appreciate your help!
To contribute, please read the contribution guidelines: https://golang.org/doc/contribute.html
Note that the Go project uses the issue tracker for bug reports and proposals only. See https://golang.org/wiki/Questions for a list of places to ask questions about the Go language.