-
Notifications
You must be signed in to change notification settings - Fork 30
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
OpenPandora port, with separate Commit for GLES, ARM, and Pandora specific #281
base: master
Are you sure you want to change the base?
Conversation
Sorry about the new Pull request, I delete my old branch, so I guess the old Pull closed itself. |
That's fine. I have been really busy with the Emscripten port right now and the work you did with EGL/GLES might help there too.. I have the immediate mode (gl1.c) renderer working (https://twitter.com/danielaquino/status/391786141491281920) but it's wicked slow so I'm working on supporting the GLSL backend but since WebGL is based on OpenGLES then I need to get around not having glMapBuffer and other things. I hope you don't mind using your fork for Pandora builds for a while cause it might take some time to restructure the the various opengl implementations and the changes we're both making so they cleanly implement each backend. It will probably come down to me merging in a new GLES backend that you'll be able to merge into your branch I hope. For things like the commit, "debug: remove of some leftover printf" you can simply use |
Have you had a chance to try out rebase-i ? I think your really going to like it.. It seems that this is targeting GLES 1 ? Do all versions of Pandora support GLES 2 ? |
I haven't tried rebase -i yet. And yes, my port are targeting mainly GLES1 (because it's what I'm 2013/10/28 Daniel Aquino notifications@github.com
|
Yes actually since webgl is ES2 compatible I had to update my Emscripten All I really had to do was use glBufferSubData in place of glMapBuffer and All you have to do is build with GL=2 to if you wanted to try that out. That might give you a performance boost if that's an issue since it really |
Note, for anyone who ends up seeing this later. The fact that this is still here isn't really an indication that the project is dead. The actual proper fix for this requires a bunch of code refactoring that no one has attempted to do yet. A proper fix will apply to this request, Emscripten, and any other platforms that require proper memory alignment. This branch is functional in the meantime and we could even generate builds from it at some point if people wanted. It's just a matter of actually wanting to put in the work... |
Could you add libs back to git ignore? There is a bunch of random white space changes like deleting empty lines that shouldn't really be in there. I'd rename ARM define to something more generic like MEMORY_ALIGNED since it applies to more platforms. I'd like to break this up into separate pull requests to separate RPI, Pandora, memory alignment, opengl changes. Then we can focus on accepting each one quicker. The memory alignment fix could be turned into a new routine to rewrite all the file loading code but at this point it's clear it's too much work and might as well get the changes in as are. |
Ok, I'll see what I can do. I'll see the breakout later. |
I can help to rebase it all into new pull requests and clean things up if you need me too. At first site part of the reason I'd like Pandora changes split out is that adding tons of ifdefs all over the code seems like a headache down the road to maintain. I'd like to instead perhaps expose those settings either as a external config file or grouped up into one set of defines. We can do that after it's all split. |
Also rebasing is your friend here to cleanup commits. I can do it for you but it might change the author of the commit to me which I'd rather not do. |
Please retry analysis of this Pull-Request directly on SonarCloud |
Now pull request, as I delete the original branch from which the previous pull was based.
Probably not the normal way to do it.