Gecko-dev Diary
Starting in August 2023 I'll be upgrading the Sailfish OS browser from Gecko version ESR 78 to ESR 91. This page catalogues my progress.
Latest code changes are in the gecko-dev sailfishos-esr91 branch.
There is an index of all posts in case you want to jump to a particular day.
Click the icon for the Gecko-dev Diary RSS feed.
Gecko
5 most recent items
20 Mar 2024 : Day 191 #
This morning I'm taking the first opportunity to test out the new libxul.so binary generated from the build yesterday. Firing up the app there's no render and although initially things seem to be running fine (apart from the complete lack of anything useful on-screen!) after a short time the browser seizes up again.
So sadly this hasn't fixed the underlying issues. But I'm hoping it still fixed something. I can at least push the app through valgrind to see if it had some effect or not.
So this is exactly what I've done.
This is good news: it seems that the change did the trick and, although the render isn't working, the change also doesn't cause the WebView to crash. It feels like this is one step closer to getting our desired result.
For those who are still following along, or just desperate to get your hands on a build to test, I appreciate it's slow progress. But each fix is a fix that's necessary and I remain confident that eventually all of the pieces will fall in to place. This is very definitely a marathon and not a sprint. An Odyssey not a citybreak. We'll get there!
Having exhausted valgrind as a means to find the problem, we now need a new approach. In order to figure out where to focus, I'm going to go back to the SharedSurface_EGLImage::Create() method. You may recall that the seizing up of execution only happened after we persuaded this method to do something sensible and not just return null all the time.
I want to revisit this. My plan is to go through this method and judiciously remove parts of it until the seizing up no longer happens. The plan is to narrow down exactly which part is causing the issue, which should also narrow down where to look for a possible fix.
Unfortunately other commitments mean it's only a short one today. I'm very much hoping to be able to get to this task of cutting out elements to see the effects tomorrow.
If you'd like to read any of my other gecko diary entries, they're all available on my Gecko-dev Diary page.
So sadly this hasn't fixed the underlying issues. But I'm hoping it still fixed something. I can at least push the app through valgrind to see if it had some effect or not.
So this is exactly what I've done.
$ valgrind --log-file=valgrind-harbour-webview-esr91-02.txt harbour-webviewThe resulting output file still — unsurprisingly — contains no shortage of memory mess, but checking the Compositor thread there's no sign of anything related to fGetDisplay(), GetAndInitDisplay() nor CreateOffscreen().
This is good news: it seems that the change did the trick and, although the render isn't working, the change also doesn't cause the WebView to crash. It feels like this is one step closer to getting our desired result.
For those who are still following along, or just desperate to get your hands on a build to test, I appreciate it's slow progress. But each fix is a fix that's necessary and I remain confident that eventually all of the pieces will fall in to place. This is very definitely a marathon and not a sprint. An Odyssey not a citybreak. We'll get there!
Having exhausted valgrind as a means to find the problem, we now need a new approach. In order to figure out where to focus, I'm going to go back to the SharedSurface_EGLImage::Create() method. You may recall that the seizing up of execution only happened after we persuaded this method to do something sensible and not just return null all the time.
I want to revisit this. My plan is to go through this method and judiciously remove parts of it until the seizing up no longer happens. The plan is to narrow down exactly which part is causing the issue, which should also narrow down where to look for a possible fix.
Unfortunately other commitments mean it's only a short one today. I'm very much hoping to be able to get to this task of cutting out elements to see the effects tomorrow.
If you'd like to read any of my other gecko diary entries, they're all available on my Gecko-dev Diary page.
Comments
Uncover Disqus comments