Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

SSE/SIMD versions of CRC32 and yEnc-decoder routines #454

Closed
hugbug opened this issue Sep 29, 2017 · 3 comments

Comments

Projects
None yet
2 participants
@hugbug
Copy link
Member

commented Sep 29, 2017

This issue is to test and evaluate SSE/SIMD versions of CRC32 and yEnc-decoder routines.

Details can be found in this interesting but long discussion.

@hugbug hugbug added the feature label Sep 29, 2017

@hugbug hugbug added this to the v20 milestone Sep 29, 2017

@hugbug hugbug referenced this issue Sep 29, 2017

Closed

Speed optimisations in downloader #448

8 of 8 tasks complete

hugbug added a commit that referenced this issue Oct 2, 2017

#454: SSE yEnc and Crc32 routines
for i686 and x86_64 architectures.

hugbug added a commit that referenced this issue Oct 4, 2017

hugbug added a commit that referenced this issue Oct 8, 2017

#454: integrated node-yencode library by Anime Tosho
1) integrated the library; 2) splitted units by CPU architecture; 3)
extended makefile and configure script to detect CPU architecture and
use appropriate compiler flags; 4) runtime CPU features detection for
x86 and ARM with dynamic code  dispatching; 5) temporary (for test
purposes) printing info about SIMD support to stdout on program
startup; 6) new SIMD routines are not yet used in the program

hugbug added a commit that referenced this issue Oct 8, 2017

hugbug added a commit that referenced this issue Oct 8, 2017

hugbug added a commit that referenced this issue Oct 8, 2017

hugbug added a commit that referenced this issue Oct 9, 2017

#454: integrated node-yencode library by Anime Tosho
1) integrated the library; 2) splitted units by CPU architecture; 3)
extended makefile and configure script to detect CPU architecture and
use appropriate compiler flags; 4) runtime CPU features detection for
x86 and ARM with dynamic code  dispatching; 5) temporary (for test
purposes) printing info about SIMD support to stdout on program
startup; 6) new SIMD routines are not yet used in the program

hugbug added a commit that referenced this issue Oct 9, 2017

hugbug added a commit that referenced this issue Oct 9, 2017

#454: integrated node-yencode library by Anime Tosho
1) integrated the library; 2) splitted units by CPU architecture; 3)
extended makefile and configure script to detect CPU architecture and
use appropriate compiler flags; 4) runtime CPU features detection for
x86 and ARM with dynamic code  dispatching; 5) temporary (for test
purposes) printing info about SIMD support to stdout on program
startup; 6) new SIMD routines are not yet used in the program

hugbug added a commit that referenced this issue Oct 9, 2017

hugbug added a commit that referenced this issue Oct 9, 2017

hugbug added a commit that referenced this issue Oct 10, 2017

hugbug added a commit that referenced this issue Oct 11, 2017

#454: integrated node-yencode library by Anime Tosho
1) integrated the library; 2) splitted units by CPU architecture; 3)
extended makefile and configure script to detect CPU architecture and
use appropriate compiler flags; 4) runtime CPU features detection for
x86 and ARM with dynamic code  dispatching; 5) temporary (for test
purposes) printing info about SIMD support to stdout on program
startup; 6) new SIMD routines are not yet used in the program

hugbug added a commit that referenced this issue Oct 12, 2017

hugbug added a commit that referenced this issue Oct 16, 2017

hugbug added a commit that referenced this issue Oct 19, 2017

#454: one-pass simd decoder
updated SIMD decoder, support for end-of-stream detection

hugbug added a commit that referenced this issue Oct 20, 2017

#454: fixed buffer overrun
and compiler warnings on VC++

hugbug added a commit that referenced this issue Oct 20, 2017

hugbug added a commit that referenced this issue Oct 20, 2017

#454: removed option "RateBuffer"
since it’s no longer needed with raw decoder which works on 4KB buffer
already

hugbug added a commit that referenced this issue Oct 20, 2017

#454: removed force-inline
since it’s no longer needed after moving loop into inner functions;
better compatibility with different compilers

hugbug added a commit that referenced this issue Oct 22, 2017

@hugbug hugbug closed this Oct 22, 2017

hugbug added a commit that referenced this issue Oct 25, 2017

@Safihre

This comment has been minimized.

@hugbug

This comment has been minimized.

Copy link
Member Author

commented Nov 1, 2017

That was a bit of exaggeration to explain improved speeds by SIMD only. It was one of many changes contributed to better performance.

I guess that's how advertisement works though 😉

@Safihre

This comment has been minimized.

Copy link

commented Nov 1, 2017

Hahah yes sorry, we know indeed the details, but for them it's the end result that counts 🏎

hugbug added a commit that referenced this issue Nov 1, 2017

#454: fixed: missing data in raw article mode
- option “RawArticle”

hugbug added a commit that referenced this issue Nov 1, 2017

#454, 801bf1a: reactivated simd-decoder
which got accidentally deactivated
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.