flypig.co.uk

List items

Items from the current list are shown below.

Gecko

4 Nov 2023 : Day 80 #
I've updated the code in GLContextProviderEGL::CreateWrappingExisting() and GLLibraryEGL::Init() as I described yesterday. I've had to make more change to the upstream code than I was hoping for, but practically speaking I have to make getting something to render on screen my priority over minimising code changes right now.

I'm not quite there yet though. In ESR 78 the DoEnsureInitialized() method — which is the equivalent of ESR 91's Init() method — stored the EGLDisplay in a class variable of GLLibraryEGL. This was the route out of the method for the display value. In ESR 91 that doesn't make so much sense any more. For a start we're not returning an EGLDisplay but rather an EglDisplay that wraps it.

I could update the Init() method so that it returns this as the result of the method, but currently it returns a true/false success/failure condition and changing that is going to get messy.

However, the GLLibraryEGL class does have an mDefaultDisplay field. I wonder if I could use that?

My train is arriving in to London, so it's a question I'll have to try to answer later. That looks like a promising approach though.

[...]

I'm on the train back home and after debugging and debugging and debugging for what feels like weeks, I'm finally building something again. I've made the changes discussed earlier in relation to EglDisplay and EGLDisplay and the updated code is now merrily compiling away.

The code compiles. Once I'm home this evening I'll transfer it over to my phone to see whether it has the positive effect I've been hoping for. If not, I will also hopefully have time to step through it to see whether or not it's doing what I planned.

[...]

I'm home now. The code did compile but the partial build got stuck and refused to produce a library. So I've had to do a full build instead. Which means no testing tonight, but hopefully there'll be a build to test tomorrow morning.

Back on Day 77 I mentioned the fact that from today onwards I'll be taking a pause from these developer diaries while I work on a Turing Way Book Dash. I still plan to have this break but I want to persevere a little longer. I feel like things are close to the point where a render may be possible and I don't want to pause before having tested out these changes. So please don't be surprised if there's another update tomorrow after all.

For all the other entries in my developer diary, check out the Gecko-dev Diary page.

Comments

Uncover Disqus comments