Forgot your password?
typodupeerror
Firefox Mozilla Open Source Software Windows IT News Technology

Mozilla Brings Back Firefox 64-Bit For Windows Nightly Builds 209

Posted by timothy
from the lending-your-closed-source-a-hand dept.
An anonymous reader writes "Last month, Mozilla Engineering Manager Benjamin Smedberg quietly announced that the 64-bit version of Firefox for Windows would never see the light of day. After what he referred to as 'significant negative feedback,' Smedberg has announced he has reviewed that feedback, consulted with his release engineering team, and has decided on a modification to the original plan: Firefox 64-bit for Windows may still never be released, but nightly builds will live another day."
This discussion has been archived. No new comments can be posted.

Mozilla Brings Back Firefox 64-Bit For Windows Nightly Builds

Comments Filter:
  • 64-bit? Bah (Score:5, Funny)

    by Anonymous Coward on Saturday December 22, 2012 @03:17PM (#42370787)

    That means Firefox will still be limited to 4 PiB, which I'm sure it'll be reaching by the release of Firefox 12,458 next year. We need a 128-bit version.

    • by sedmonds (94908) on Saturday December 22, 2012 @04:51PM (#42371317) Homepage
      And by the year after that, they'll need 128-bit just for the firefox version number.
    • Re: (Score:3, Insightful)

      by niftymitch (1625721)

      That means Firefox will still be limited to 4 PiB, which I'm sure it'll be reaching by the release of Firefox 12,458 next year. We need a 128-bit version.

      Naw... it is two things. Cleaner code and the lack of 32bit library cruft in a system. However plugins like flash and even Java32-.vs.-Java64 make me think that well flash is crud and Java not as portable as it should/could be.

    • by lokedhs (672255)
      Actually, that would be 16 EB. You were off by 4 orders of magnitude.
  • by Jafafa Hots (580169) on Saturday December 22, 2012 @03:22PM (#42370823) Homepage Journal

    The 64 bit branch of FireFox and loved it, it was much faster. Some recent update broke compatibility and I have to do a total uninstall and then reinstall while backing up every Firefox thing elsewhere to restore, since it uses the same profile, etc.

    Too lazy to work out the issue, in other words.

    But if you have a nice fast 64bit machine, try Wtaerfox, you'll probably love it. Unless it sucks now.

    • Nonsense, been running WaterFox myself. Works with all the FireFox extensions I can find.

      I do not, for the life of me, understand why FireFox is so hell-bent on 32-bit versions.

      • by Billly Gates (198444) on Saturday December 22, 2012 @03:40PM (#42370939) Journal

        Probably because flash, java, and other plugin makers are so slow to move to 64 bits. Not to mention many out there feel a browser should not use more than 4 gigs of ram and is a light text and graphics reader. Not a minature operating system running complex ajax applications

        • by Anonymous Coward

          Java is 64-bit since a long time. Flash also has a 64-bit version on windows for quite some time. As for Java... it's a PITA to install 32 and 64 bit versions at the same time. Amazing how Oracle messed that up. Going 64-bit only however makes this problem disappear: the 64-bit version is all you need and there are no more clashes. And it gives an immediate speed bonus as well, as 64-bit simply is faster.

          • by murdocj (543661)

            Yikes, Oracle... had a bad flashback to installing the Oracle client. Cold chills go up my spine whenever I think about it. How could installing that damn client be that painful?????

            • by jd2112 (1535857)

              Yikes, Oracle... had a bad flashback to installing the Oracle client. Cold chills go up my spine whenever I think about it. How could installing that damn client be that painful?????

              Removing Oracle is easy compared to Symantec Antivirus.

          • 64-bit isn't just simply faster. It's sometimes/often/usually faster depending on what you do, but it is ALWAYS larger in memory and disk space usage, and plays less well with other applications (Cache size overruns caused by the larger size).

            • by Rockoon (1252108)
              sigh..

              Look, the pointer size is only of concern if you are using a lot of pointers..

              ...but using a lot of pointers negates whatever cache argument that you thought that you had had..
              • Umm... No, it does not.

                • by Rockoon (1252108)

                  Umm... No, it does not.

                  Mr Buzzword doesnt know what he is talking about...

                  The fact that the pointers use an extra 4 bytes is a negligible detail because the L1 caches are huge in relation to those 4 bytes. You would need a very large amount of pointers within the cache to have any sort of measurable effect, so large that following even small percentages of them (why are they in the caches if you arent using them?) will always cause significant L1 thrashing.

                  But an even harder argument for you to try to refute is that Intel is

                  • Mr Buzzword doesnt know what he is talking about...

                    I'm glad you admit you don't know what you are talking about. I wish more people who didn't know would admit it right up front like you did.

                    The fact that the pointers use an extra 4 bytes is a negligible detail because the L1 caches are huge in relation to those 4 bytes. You would need a very large amount of pointers within the cache to have any sort of measurable effect, so large that following even small percentages of them (why are they in the caches if you arent using them?) will always cause significant L1 thrashing.

                    The fact that pointers (and anything dealing with size or offset) uses DOUBLE the size, and isn't negligible. It's effectively halving the size of the caches (total size, line size) in terms of the number of pointers/ints that they can contain.

                    But an even harder argument for you to try to refute is that Intel isnt stupid. The cache parameters (total size, line size, number of sets, for each level of cache) are optimized for 64-bit computing on their 64-bit processors. Intel didnt choose a 64 byte L1 line size willy-nilly. Intel didn't choose 32KB of L1 data willy-nilly. Intel didnt choose 8-way set associativity for their L1 willy-nilly.

                    I'm not going to refute that, nor is that directly related. However, as a side note that I don't want to get tied up in, I wi

                    • *facepalms*

                      But going to 64-bit on an x86 machine also doubles the number of registers, as well as introduces any number of extra instructions that can speed things up. If you aren't getting a significant speedup with a 64-bit build, then something is usually wrong. Like you are casting to a 64-bit integer, but are still treating it like it's 32-bit.

                      Plus your benchmarks mean nothing with FireFox -> the people behind the builds have put that much more effort into optimizing it for 32-bit over 64-bit. It's

                    • Yes, you have twice the number of registers in 64-bit, however, there are some performance penalties for using them (prefix ops), and I'm pretty sure that the register renaming that is done really makes that a rather moot point except for the prefix op penalty still applies if you try it. You are going to find a mixed bag of performance characteristics trying to use them unless you are really really careful, and most compilers aren't, and most apps aren't going to hand code assembler to get it right.

                      Second

        • by dgatwood (11270) on Saturday December 22, 2012 @05:00PM (#42371377) Journal

          Not to mention many out there feel a browser should not use more than 4 gigs of ram and is a light text and graphics reader.

          Having a >4GB footprint is not the only reason to move to a 64-bit address space. As more software becomes 64-bit, those legacy 32-bit apps become more of a problem, both in terms of longer application launch times (because the 32-bit library stack that it uses isn't loaded initially) and in terms of added memory pressure (because of all those unnecessary libraries loaded into RAM).

        • by RabidReindeer (2625839) on Saturday December 22, 2012 @05:32PM (#42371513)

          Probably because flash, java, and other plugin makers are so slow to move to 64 bits. Not to mention many out there feel a browser should not use more than 4 gigs of ram and is a light text and graphics reader. Not a minature operating system running complex ajax applications

          That, after all, is a job for Emacs.

      • by hairyfeet (841228) <bassbeast1968@NOspaM.gmail.com> on Saturday December 22, 2012 @03:52PM (#42371023) Journal

        Pale Moon is also quite good and has a 64bit version although they compile their for newer CPUs (Athlon64, P4 and above) but if you are running 64bits you probably aren't using an ancient CPU anyway.

        But the FF numbers have been dropping and its from nothing but shit like this, being arrogant and not listening to their customers. After all that is EXACTLY what they are, for without them Google wouldn't pay Mozilla for the search rights and the whole company goes down the drain so its time they started actively listening to their customers instead of pissing them off.

        This is one thing we have so much better than we did when FF first came out, it no more "use this or get stuck with IE" nonsense, no more having the entire web coded for IE quirks, now we have this great bounty of choice so if any company refuses to listen to us we can just go somewhere else without having to jump through hoops or deal with a crippled web experience. Just off the top of my head I can name Chrome, Chromium and Comodo Dragon in the Chromium based camp, FF, Pale Moon, Waterfox,Kemelon and Comodo IceDragon in the gecko builds, then you got the more "one offs" like Safari and Opera, that's ten browsers that will ALL render the web just fine, so if the numbers for FireFox go down they have nobody to blame but themselves. maybe next time instead of just pulling a boneheaded stunt how about actually asking your customers first, huh?

        • In this case they are listening.

          I can see where Mozilla is coming from as they have limited resources to double the development efforts for a so called free product. I wish Mozilla would invent Mozilla search to go head to head with Google, but they do not have the revenue for such a risky maneuver.

          • by crutchy (1949900)
            or even better if the Apache Foundation made a search engine.... or maybe a collaborative Mozilla-Apache effort
          • by Tridus (79566)

            If they have limited resources they should direct them in a more appropriate way. 64 bit Firefox has significant benefits to performance and less out of memory related crashing.

            The OS they're working on, OTOH...

          • If they have limited resources, then why don't they abandon the 32-bit branch?

            • Because a lot of otherwise capable hardware is still 32-bit only?

        • by Reziac (43301) *

          I normally use SeaMonkey (as the least obnoxious of the Moz family) but when some website expects yesterday's FF build, I use Palemoon -- the only FF incarnation I can stand. -- Their general memory piggishness seems to be a cache-handling bug, probably a side effect of an hour of use generating 4000 directories and 500 files (not a typo, there are bunches of empty folders) and never cleaning up after either. I found if I clear cache every few minutes, the galloping memory usage doesn't happen.

          Good info abo

      • by icebike (68054)

        I do not, for the life of me, understand why FireFox is so hell-bent on 32-bit versions.

        I do not for the life of me understand this blind push to 64bit when there is no demonstrable speed improvement.
        32bit software on a 64bit platform is not measurably slower for the tasks that a browser needs to do.

        • It may be a blind push to 64-bit, but when AMD and Intel debut some 128-bit CPUs for mainstream computing, which may leave some 32-bit support out just as much 16-bit stuff was left out of 64-bit, obviously there will be a need for at least a 64-bit version of the application. Then you'll see a push for a native 128-bit version. Look at what happened with the switch from Windows 9x/ME to XP. Microsoft put many 16-bit applications on life support after XP came to the desktop to replace the 9x/ME versions
          • by crutchy (1949900)
            i'm pretty sure it would be still be possible to thunk a 16 or 32 bit instruction on a 128 bit processor

            will Microsoft Windows 128-bit support 16 or 32 bit? probably not
            • I don't see why not. Windows XP 64 supported 8 (DOS)/16 (DOS/Win)/32 (DOS/Win)/64 (Win) bit apps. They only recently dropped support for 16-bit Windows Apps.

              • by Dagger2 (1177377)
                XP x64 does not run 16-bit Windows programs. I assume it doesn't run any DOS executables either but I don't have any handy to test with.
                • I stand corrected, you did need to run 16-bit Windows apps/Dos apps inside Virtual PC to get them to run.

            • by Rockoon (1252108)

              i'm pretty sure it would be still be possible to thunk a 16 or 32 bit instruction on a 128 bit processor

              Sure, just like its possible to thunk back to 16-bit once in 64-bit mode on AMD64...

              ...oh wait, it isnt! You dont know what you are talking about.

              • It depends which 16-bit mode you are talking about.

                It IS possible to do 16-bit protected mode (which win16 apps use) while in long mode. Wine has no trouble doing it. MS just couldn't be bothered debugging WOW on top of WOW64 so they left it out.

                OTOH it is not possible to do virtual 8086 mode while in long mode.

        • I do not for the life of me understand this blind push to 64bit when there is no demonstrable speed improvement.

          Part of the problem is all Firefox tabs and windows are part of the same process, unlike IE (and I believe Chrome). So, if a misbehaved AJAX app in one tab uses a gig and a half, every browser window becomes unusable. If FF were 64-bit, then it could use the 8GB or 16GB found in most new machines to mask the problem.

          • by gbjbaanb (229885)

            however, that just masks the problem, doesn't solve it.

            With the current Firefoxes, you can go to the help menu and see what tabs are using what memory from the 'troubleshooting information' item. then you can see a rogue tab has gobbled all your ram, and close it. That's a solution.

            • by ultranova (717540)

              So having a page just work without drama is not a solution, while requiring the user to manually close it (hopefully before FF crashes) and not visit it in the future is?

              • by gbjbaanb (229885)

                if I had a nail stuck in my arm, I'll (manually) pull it out. You, it seems, would leave it stuck in you and say "its ok, I have another arm".

        • What? 64-bit image routines chomp 32-bit image routines! This is a known fact! And that's just the beginning!

          And it's the fact that memory is limited under 32-bit code to 4 GB of RAM that we really care about. FF moves like a pregnant yack when you have 20 tabs open, and has the memory usage of a 700 lb sumo wrestler at an all you can eat buffet.

      • > I do not, for the life of me, understand why FireFox is so hell-bent on 32-bit versions.

        Same here. I've been running Waterfox for some months now on a low-end 64-bit desktop PC. It's very fast, and I've not had any plugin or extension problems. Given what a pig Firefox is for memory (and to be fair Chrome and the other browsers too) if there is one application that should drive 64-bit technology it should be these damned web browsers!
      • by GNious (953874)

        Was using waterfox, but a some plugins refused to work with it, since it isn't firefox by their definition.
        Switched back to 32-bit firefox

    • by CritterNYC (190163) on Saturday December 22, 2012 @03:37PM (#42370909) Homepage
      Waterfox is just Firefox built as 64-bit with some compiler switches and a name change (required by trademark guidelines). It's not a fork and there are no additional bug fixes. It has all the bugs that Firefox does when compiled as a 64-bit binary. You're far better off sticking with Firefox 32-bit which works just fine under 64-bit Windows.
      • by Anonymous Coward

        It has all the bugs that Firefox does when compiled as a 64-bit binary.

        Out of curiosity, which bugs are those? I've seen bugs in FF (such as the well known memory leaks), but they never seemed to be 64 bit specific.

        I've been running the 64 bit firefox for, gosh... I don't even know. 8 or 10 years?

        $ file /usr/lib/firefox/firefox /usr/lib/firefox/firefox: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.24

        • by Anonymous Coward

          It has all the bugs that Firefox does when compiled as a 64-bit binary.

          Out of curiosity, which bugs are those? I've seen bugs in FF (such as the well known memory leaks), but they never seemed to be 64 bit specific.

          I've been running the 64 bit firefox for, gosh... I don't even know. 8 or 10 years?

          $ file /usr/lib/firefox/firefox /usr/lib/firefox/firefox: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.24

          I don't know if this link will work but you can also search to see the 64 bit bugs (some of which also affect x86): https://bugzilla.mozilla.org/buglist.cgi?quicksearch=firefox+64+bit [mozilla.org]

    • I can recommend Palemoon [palemoon.org].instead of Waterfox.
      Its compiled for new (SSE2) cpu's and some "redundant and optional code" is disabled.

      • by NotBorg (829820)
        That's true of any decent compiler targeting x86-64 since all x86-64 processors include SSE2 support. One of the nice things about targeting x86-64 is that you get a nice feature set that you don't have to detect support for at runtime and maintain a multiplicity of code paths to take advantage of features. If you know a feature will be there, it reduces complexity both in source and the compiled output.
    • The 64 bit branch of FireFox and loved it, it was much faster.

      Not really. Look at their own benchmarks [waterfoxproject.org] - at best performance is roughly 15% faster, some cases slower. They also commit the cardinal sin of using benchmark charts that do not start at zero, #1 on the list of "how to lie with numbers."

    • by loufoque (1400831)

      But if you have a nice fast 64bit machine

      Let me fix that for you

      But

      All Intel/AMD CPUs sold in the past 6 years are 64-bit.

      • Only some Atom chips are 64-bit.

  • by Anonymous Coward

    /. Story when turned off: "turned off for now" /. Story now: "...never see the light of day"

    Next story on /. : 64-bit Firefox sent to concentration camp.

  • This will coincide with XP ending support which is the last holdout.

    If Mozilla does not want to double the work then just focus on 64-bit. Besides a few Vista users who went to 7, I do not know anyone who uses the 32-bit version. Usually they tell me some driver or piece of software is not compatible. Most cases running it in XP mode is better nowdays and by 2014 that hardware will very old!

    Maybe release the long term version on that day as the last 32 bit version for 1 year? By 2015 no one should be runnin

    • by Anonymous Coward

      by 2014 that hardware will very old!

      the specs and requirements hasn't changed for MOST applications and tasks since late 2006 (coincides with vista oem release). even the next version of desktop windows is expected to have similar requirements... a dual core athlon from 2006-7 with 2-4gb ram is enough for most people (hard core gamers, 3d graphics artists, and engineers are not 'most people').... and is about the same speed as lower-end desktops today (some "new' ones are actually slower). age of hardware,

    • by clarkn0va (807617)

      By 2015 no one should be running 32 bit XP software or operating systems anymore. I mean enough is enough!

      MS is way ahead of you on that point. [slashdot.org]

  • by Stonefish (210962) on Saturday December 22, 2012 @04:17PM (#42371157)

    It's good to see that someone is being held accountable here. Benjamin Smedberg creates a shitload of negative publicity, pisses off a proportion of dedicated testers and he:
    A. Gets a promotion
    B. Is removed from positions of responsibility because he demonstrates poor judgement
    C. Nothing happens
    D. Gets a pay increase.

    Answer = C
    Come on guys at least make him wear a T shirt for a month that says, I must not override the recommendations of others in relation to 64 bit builds.
    One of the key problems in organisations is that people aren't held accountable for poor judgement, or at least a running sheet is not maintained. Ben will probably continue to be promoted even through he has demonstrated that he has a fundamental lack of connection with what end users want. There is obviously something wrong occurring in the firefox mozilla groupthink and yet nothing is being done.

    • by Tailhook (98486) on Saturday December 22, 2012 @06:08PM (#42371687)

      There is obviously something wrong occurring in the firefox mozilla groupthink and yet nothing is being done.

      That's the feeling I have as well. I don't use 32 bit desktops any longer. Actually haven't consistently used a 32 bit desktop in four years. To somehow not be aware of the behavior of real users is a huge fail.

      And lets not indulge anymore '64 bits isn't necessary' tripe. People that don't understand why key software must adopt the native ISA of a system and avoid backward compatibility kludges need to stop talking about this.

      Anyhow, I just upgraded my main personal desktop hardware and reinstalled my retail Win 7 OS, in addition to Linux. I installed Firefox out of habit but I haven't bothered to reacquire my usual cohort of extensions... I don't use it anymore. Chrome is superior in every way, with the sole exception that noscript is better than scriptno.

      Mozilla is repeating Netscape history. Then as now, leadership is the problem.

      • by Kjella (173770)

        That's the feeling I have as well. I don't use 32 bit desktops any longer. Actually haven't consistently used a 32 bit desktop in four years. To somehow not be aware of the behavior of real users is a huge fail. (...) I don't use [Firefox] anymore. Chrome is superior in every way

        I think they know their users perfectly, even slashdotters seem to be clueless that they're running 32 bit software under their 64 bit OS.

    • by Tridus (79566)

      Mozilla's management has their heads in their asses, and has had them up there for quite a while. They've made a very length list of poor decisions in recent memory.

  • 32-bit is insecure (Score:5, Interesting)

    by r00t (33219) on Saturday December 22, 2012 @04:30PM (#42371213) Journal

    Haven't these people heard of ASLR [wikipedia.org] and heap spraying [wikipedia.org] Do they not understand the concepts?

    Without 64-bit, you have two huge security problems. The first is that there simply isn't enough address space to randomize well. Attackers can guess things. They guess right often enough that the effort is worthwhile. The second huge security problem is that the address space is easy to fill with code-equivalent data for a ROP [wikipedia.org] attack. Actually, with Firefox you could even use real code [wikipedia.org]!

    Using a 32-bit browser in 2012 is kind of insane. It's near-complete security FAIL.

    • IE 10 32-bit has ASLR and heap spraying protection sandboxing right in. I think Chrome does as well but someone can correct me on this if I am wrong. Infact, the only thing Windows 7 64-bit has that the 32-bit does not is signed bootloaders and drivers to prevent rootkits.

      ASLR has been part of 32-bit operating systems for years since Vista.

      If someone knows the ram address of a particular dll they can target it anyway with a poke regardless of the bitness. You can still spray on a 64 bit system as well. It j

      • by r00t (33219)

        IE 10 32-bit has ASLR and heap spraying protection sandboxing right in.

        It barely works. There simply isn't enough address space to make it effective. This is like having a door using a 1-pin lock instead of a 7-pin lock. Picking a 1-pin lock is trivial.

        You can still spray on a 64 bit system as well.

        You can't usefully spray on a 64-bit system. To be useful, your spraying needs to cover a significant portion of the address space. (else you are unlikely to hit it when you trigger the bug) It also can't require more memory than the system has, and it can't need to run for months.

  • Never happened (Score:5, Informative)

    by gfody (514448) on Saturday December 22, 2012 @04:58PM (#42371365)
    I've been using 64bit nightly since the idea of dropping the builds was mentioned in bz. There was a big debate and a bunch of tech news sites picked up the story and now the latest is that they're being "restored". But the 64bit nightly builds never stopped! I'm sure this story is just to get everyone to STFU about it already.
  • As long as it's not the default version, it's not worth mentioning.
  • Have been using Waterfox for some time now, has been great. Features less bloat and fewer annoyances then has been making it's way into the 32 bit releases of firefox.
  • Good news. Even if it doesn't result in 64 bit final products soon it should provide more material for the Waterfox project to develop upon. As long is there is developer support from Firefox for the 64bit the onus is on the developer advocats of 64bit computing to prove that it can show a significant enough performance enhancement to be taken seriously and pushed to mainstream.

  • While I realize that there are theoretically 32 bit versions of Vista, Windows 7, and Windows 8 available, I have never seen one installed on a system. For the Mozilla team to say there will "never" be a 64-bit build for Windows is an asinine and bass-ackwards approach to maintaining compatibility with the OS.

    If it's the plugin-providers causing the problem, then they deserve as much "Boo Hiss" as Mozilla does.

    Let's try to keep up with the times. 64-bit Windows has been stable and standard for years!

    • by BZ (40346)

      > For the Mozilla team to say there will "never" be a
      > 64-bit build for Windows

      Which is something no one at Mozilla ever said. But don't bother reading what they actually said, just read the lies lazy reporters spouted instead.

      What Benjamin said is that there are no plans to ship a final 64-bit product in the next several months.

    • by Luckyo (1726890)

      1. XP is still one of the most popular if not the most popular OS in the world. 64-bit XP on the other hand is non-existent.
      2. 32-bit vista and seven were used quite a lot in older computers that had less then 4GB of RAM. I have a netbook bought a couple of years ago that caps at 2GB that uses 32-bit seven for example. It works fine.

      • by yuhong (1378501)

        AFAIK, I once tried 64-bit Win7 on my system with only 1.5GB of RAM, but decided to move back to 32-bit when I had to reinstall, and is what I am running right now.

  • Either make a portable website that works on every computer or mobile device users have. Or make a native app which, if your wish, can make extensive use of server-based HTML while still incorporating your "plugin" code. But exposing users to overhead, crashes and security issues of your custom code when your application is NOT running is indefensible.

We are Microsoft. Unix is irrelevant. Openness is futile. Prepare to be assimilated.

Working...