Add CUDA & OpenCL support

[crazyks]

Guetzli is an awesome jpeg encoder, however, it works a liitle bit slow. In order to speed it up, we have added CUDA & OpenCL support for Guetzli and we also optimized some procedure and added full jpeg format support.

This work is made by strongtu, ianhuang, tongzhan and me.

We had it tested on our GPU server, and here is the test statistical data for one of the sample pictures.

We hope that our codes can be merged into guetzli/master branch:)

[googlebot]

So there's good news and bad news.

:thumbsup: The good news is that everyone that needs to sign a CLA (the pull request submitter and all commit authors) have done so. Everything is all good there.

:confused: The bad news is that it appears that one or more commits were authored by someone other than the pull request submitter. We need to confirm that they're okay with their commits being contributed to this project. Please have them confirm that here in the pull request.

Note to project maintainer: This is a terminal state, meaning the cla/google commit status will not change from this state. It's up to you to confirm consent of the commit author(s) and merge this pull request when appropriate.

[rogierlommers]

When can we expect these changes to be released?

[kornelski]

@szabadka I don't get what @googlebot is complaining about.

I've rebased/filtered the branch to remove all commits with 3rd party code, and normalized e-mail addresses to match team members' GitHub accounts, but it still complains.

Does it require manual approval by you?

[strongtu]

@pornel May be we need try to fork a new branch and start a new pull request? We can merge code about CUDA support to the new branch manually. But we will lost messages of each commit in the later.

[crazyks]

@robryk @eustas our CLAs and commits seem ok now, but @googlebot is still complaining. Could you please check and approve manually?

[Bukashk0zzz]

Any news on this?

[robryk]

Thanks for this PR; it's quite large, not very trivial, and I don't speak Chinese, so it'll take me some more time to look through it.

@crazyks @pornel Please pay no attention to clabot complaints.

[FireEmerald]

Could someone of you provide a working binary file for windows with cuda support? This would be awesome.

[mikhailnov]

Will this require non-free CUDA (cuda.h) libraries for compilation? If yes, probably a compilation flag is needed to disable them...

[leafjungle]

where is cu_mem defined? cuda or opencl can not find that.

[ianhuang-777]

@leafjungle It's defined in clguetzli.cl.h, not an original CUDA definition.

[joyjoker2017]

which GPU was used in your environment?

[mikhailnov]

Waiting impatiently for this to be merged ))

[rogierlommers]

Yes, we all do!

[ianhuang-777]

@joyjoker2017 Tesla M40

[fvm]

🙏 👍 Fingers crossed 🤞 😄

[fvm]

Bump...

[DanielBiegler]

:sleeping:

[tina-junold]

Any news on this?

[alexblhr]

Is this ever going to be merged ? :)

[jonathas]

Any update on this? Please :)

[magicdoublem]

Did anyone ever succeed in building those binaries and want to share them? ;-)

[EwoutH]

This looks incredible! @crazyks could you resolve the conflicts, update the dependencies and rebase? I will check with someone from Google if we can merge into master. If not, we can create a fork.

[doterax]

Hi there. I have built Guetzli with CUDA for Windows.

You can download binaries from here.

[twitnic]

@crazyks could you resolve the conflict?

[doterax]

@crazyks could you resolve the conflict?

I continue supporting guetzli with CUDA and OpenCl here, where you can also download windows binaries.