Saturday, November 24, 2012

"17.0.1pre" available

I put the "pre" in quotation marks because it still appears as 17.0, but this contains the new atomics as discussed in our last blog post. However, because JavaScript does not require (at least currently) the same memory guarantees as the rest of the browser does (might), I implemented an atomic set routine just for JavaScript that has no read or write barriers, which is a lot faster than the other approach (which is faster still than the old mutex method). Get it from the Downloads tab. The only difference you should note is that the browser is faster; there are no other changes.

Based on Tobias' initial analysis and work in 19, there appears to be enough stuff that Mozilla broke that we should just skip 18 and go to that to get started on undoing the bustage right away. Thus, Aurora 19 will be the inaugural release of the next unstable series. I'll probably start work on that this week once I've sufficiently recovered from tryptophan-induced Thanksgiving turkey coma (though actually this year we had roast duck).

8 comments:

  1. Gotta believe there's tryptophan in duck as well, 'cause their cousins......bird cousins all the way!

    ReplyDelete
  2. This comment has been removed by the author.

    ReplyDelete
  3. Thank you for TenFourFox 17.0. It runs very well on Mac OS X 10.4.11 Tiger and 10.5.8 Leopard (Power Mac G5 2003).

    Unfortunately there isn't Firefox 17.0 available for my Intel iMac with 10.5.8 Leopard. What do you think of TenFourFox for Intel Macs with Tiger and Leopard?

    ReplyDelete
  4. I think it's a great idea, but it's not one I myself would work on -- the only Intel Mac I own is for a specific purpose and it runs 10.6.8 anyway (my daily drivers are this quad G5 and an iMac G4).

    I should write a blog post on what would be required to get it to work on Intel. It would be non-trivial, but it would not be difficult, and if someone(tm) decided they wanted to do it, I would be willing to direct people to it and help host builds. However, they would be the primary responsibility. Do you feel up to it? A few people looked at it but nothing ever happened -- but now that 10.4 and 10.5 support is gone, I bet there will be some additional interest.

    ReplyDelete
  5. AuroraFox should be buildable for Intel 10.5 (32 bit only), probably with some little adjustments. A universal version should be doable as well.

    ReplyDelete
  6. -- I should write a blog post on what would -- be required to get it to work on Intel.

    That's a good idea :)

    ReplyDelete
  7. On skipping 18 to go to 19:

    Wait, I'm confused. What if Mozilla manages to break as much in their 19 as well? Will your work in 19 then be wasted, so you'll have to skip onward to 20?

    ReplyDelete
  8. That's a fair question and ultimately it's a judgment call. The good news is that most of the breaking changes we expected (dropping ATSFont support, which is needed for 10.4/5; dropping some of the low-level library plumbing) did indeed land in 19. There's always a risk of them doing something else or making a change we didn't expect, but these were the big shoes we were waiting to drop, so we'll just jump to where they need to be fixed since 18 doesn't have those issues. If 19 has some other big problems I might revert to 18 to see what the difference was.

    One silver lining to the rapid release system, though it's a lot more work over time, is that the changes between versions are easier to track. And that makes it easier for us to repair our system dependencies since their modifications are more incremental. But the concern is certainly well-taken.

    ReplyDelete

Due to an increased frequency of spam, comments are now subject to moderation.