Google Chrome For Linux Goes 64-bit

Noam.of.Doom writes "The Google Chrome developers announced on August 19th the immediate availability of a new version of the Google Chrome web browser for Linux, Windows and Macintosh operating systems. Google Chrome 4.0.202.2 is here to fix a lot of annoying bugs (see below for details) and it also adds a couple of features only for the Mac platform. However, the good news is that Dean McNamee, one of the Google Chrome engineers, announced yesterday on their mailing list that a working port of the Chrome browser for 64-bit platforms is now available: 'The v8 team did some amazing work this quarter building a working 64-bit port. After a handful of changes on the Chromium side, I've had Chromium Linux building on 64-bit for the last few weeks. I believe mmoss or tony is going to get a buildbot running, and working on packaging.' Until today, Google Chrome was available on both 32- and 64-bit architectures, but it appears that the latter was running based on the 32-bit libraries. Therefore, starting with Google Chrome 4.0.202.2, 64-bit users can enjoy a true x64 version!"

Comments

What a waste of resources

[BadAnalogyGuy]

Come on. This is Linux, not some WIMPy GUI-based OS like Winders or Suckintosh.

I run the results of wget through a custom Perl script and then parse the results and feed image URLs back through wget and into libjpeg.

Why do I need a bloated web browser when I have such an elegant Unix solution?

Re:What a waste of resources

[Yvan256]

This is a really bad analogy, guy.

Re:What a waste of resources

[Hurricane78]

Does it look like the computer of Neo in Matrix 1?

You may kid, but actually, this very *freedom*, to attach everything to everything, and transform the streams of data, while everything is a file in the file system, is *the* killer feature of Linux.

Unfortunately, Linux people somehow act ashamed for it, because the dumb masses, who know jack about actually using you computer and doing it efficiently too, think it would be some cumbersome thing of the past. And the Linux people apparently put the reality of those people above their own much better informed one. (Which is actually pretty pathetic. But understandable, because, well, we aren't the guy that lords the club. Or rules the company party. We aren't the self-confident ones. But hey, there's no reason we can't be. Especially in our expert field.)

Also even more unfortunately, this resulted is the whole GUI desktops imitating Windows and the Mac down to every shitty (or good, the point it that it does not matter) detail, instead of following their own much more powerful and proven philosophies.

I hope that at some time in the future, I will be able to create a team, and hack all those tools into parts and create a new desktop environment, that actually follows that philosophy.
(Anybody interested in sponsoring it? Google? You? :)

Chromium Not Chrome

[thejapanesegeek]

"Google Chrome" has not been released on the Linux platform yet. From the about:linux-splash on my chromium install:

Chromium is an open source browser project. Google Chrome is a browser from Google, based on the Chromium project.

And "Chromium" still doesn't have things like flash and printing, at least not in a stable, usable form.

Re:Chromium Not Chrome

[MrHanky]

Wrong. While still 'un-official', a developer preview of Google Chrome for Linux has been out for a long time, freely available. Link.

Re:Chromium Not Chrome

[Jamamala]

Chromium does have flash support if you start it with --enable-plugins. It works pretty well, although admittedly I haven't tested the latest 4.0 builds or the x64 version.

Re:Chromium Not Chrome

[pablomme]

And "Chromium" still doesn't have things like flash and printing, at least not in a stable, usable form.

Wrong about flash. Add '--enable-plugins' to chromium-browser's command line, and soft link the flash library into chromium's plugins directory (which they fail to tell you to do), e.g. in Ubuntu you would do:

  sudo ln -s /usr/lib/flashplugin-installer/libflashplayer.so /usr/lib/chromium-browser/plugins/

Works well, is stable and is usable, despite the warnings that it may melt your computer etc. Printing is still unavailable.

"Iron" out the privacy bugs...

[Jah-Wren Ryel]

And if we are lucky, there will soon be a privacy-enabled version here:
http://www.srware.net/en/software_srware_iron_download.php

No One Cares What It's Called - It's Fucking Fast

[Anonymous Coward]

Google really hit a home run with Chrome on Linux.

It's as lighting quick on Linux as it is on Windows. And it's just as lighting quick days after heavy use unlike that outdated piece of crap Firefox.

What is puzzling about Chrome/Chromium on Linux is why Google made it look like modern desktop app and not the usual 'designed by a blind person with bad taste using Windows 95 UI widgets' that appears to be the standard.

Re:!x64

[NervousNerd]

From Wikipedia:

The terms x86-64 and x64 are often used as vendor-neutral terms to collectively refer to x86-64 processors from any company.

Serious question

[Thatmushroom]

Can someone explain the particular benefits of having a 64-bit browser? I particularly appreciate the fact that Firefox currently can't hog all of my RAM when something (oftentimes Flash) spirals out of control. Do web developers use memory beyond the 4 gig limit, and is this a godsend for them?

Re:Serious question

[Jah-Wren Ryel]

Can someone explain the particular benefits of having a 64-bit browser?

Not much really. If you frequently browse for porn, I suggest holding out for the TOPS-20 port which will be 69-bit.

Re:Serious question

[dagamer34]

Mainly anything to do with number crunching, 64-bit is a LOT faster. Supposedly also JavaScript in 64-bit will get a boost (as Safari 4 on Snow Leopard seems to show).

Re:Serious question

[harmonise]

Can someone explain the particular benefits of having a 64-bit browser?

For the same reason that text editors might be 32 or 64-bit and not 16-bit. It's what the OS and hardware directly supports. Most computers shipped in the last four or five years are 64-bit and can support 64-bit operating systems. Even Microsoft has said that Windows 7 will be their last operating system to support 32-bit. It makes sense to start developing native 64-bit versions of software instead of sticking with 32-bit and using a compatibility library or layer to further complicate things or possibly screw up or need to debug. There's no direct benefit other than it's the native number of bits for your hardware, but that makes it the right thing to do.

This is why I think leading edge distros like Ubuntu should stop supporting 32-bit and only distribute 64-bit versions of their distros. The only computers that still are 32-bit are either embedded systems or older legacy computers. We need to move on.

Re:Serious question

[FlyingBishop]

Really? I'm posting this from a 32-bit legacy computer that runs 32-bit Ubuntu just fine. I've got a 'legacy' machine on the right with XP and an Nvidia GForce 5600, also 32 bit, that is even more useful, being a Pentium. You are about 4 years too early to even begin to talk about end-of-life for 32 bit.

I couldn't really find numbers, but I supect ditching 32 bit would entail throwing out at least half of the computers currently in use...

I don't have the numbers to back it up, but I'm fairly certain that a sufficiently large portion of computers use 32-bit to make your presumption completely infeasible for the next few years. They were still selling 32-bit machines two years ago, and people can't reasonably be expected to retire those machines until 2011, and many will still be perfectly useful until 2013 or even 2015 with a few repairs.

Meanwhile, you can keep on living in your fantasy world where hardware can magically upgrade itself to run the latest and greatest software.

Re:Serious question

[larry bagina]

x86-64 has more registers. v8 compiles javascript into native code, so more registers means more variables stored in registers (instead of on the stack) and faster execution.

Re:Serious question

[Bert64]

There is already a 64bit version of flash:
http://labs.adobe.com/downloads/flashplayer10.html

Re:Serious question

[bds1986]

This is why I think leading edge distros like Ubuntu should stop supporting 32-bit and only distribute 64-bit versions of their distros. The only computers that still are 32-bit are either embedded systems or older legacy computers. We need to move on.

Ubuntu has nowhere near enough market share to be able to just write off a huge swath of the personal computer userbase without alienating itself. Not to mention that one of the marketing points of Linux is that it will run on the older hardware you have laying around, unlike a certain other OS. Indeed, I'd suspect a sizable portion of Linux installs are run on old "junk hardware". Removing 32-bit for no reason other than "But 64 bit is newer!" wouldn't benefit anyone other than NetBSD uptake stats.

32-bit hardware is perfectly fine, and more than adequate for the needs of many computer users (when was the last time your grandparents needed >4GB RAM?), there's no reason for it not to be supported for the next few years at least. Do you have stock in Intel or something?

Re:Serious question

[SSpade]

64 bit binaries consume more space than 32 bit binaries. That means they make less effective use of system memory and (just as importantly) CPU cache.

All other things being equal, a 64 bit binary will run slower, while using more system resources than a 32 bit binary, so it's something that you'd only want to do if you could actually make use of huge amounts of memory in your application directly.

All other things aren't equal, though, as x86 is rather a "special" architecture. There the 64 bit binary will get access to (amongst other things) more registers. It'll still consume more system resources (ram, disk, power), but may run faster, or slower.

So it's not as simple as "64 bit good, 32 bit bad".

Re:Serious question

[Curtee]

If you are running 64-bit linux already, you will get some performance advantage to running as many apps as possible in 64-bit mode. This is because your shared libraries can actually be shared. Otherwise you end up loading the same shared libraries twice (once for the 64-bit version and once for the 32-bit version). There's noticably longer startup time when you do that.

Re:Serious question

[faragon]

Cons:

- The benefit from passing from 8 to 16 general purpose registers is very little, and often, counterproductive, as total "true registers", the ones used for register renaming in OoOE remain the same, so with twice the general purpose registers, you halve the renaming register pool. That was specially noticeable in firsts AMD64 CPUs, and *very* noticeable on Intel Pentium D CPUs (Pentium 4 with x64 support and other minor changes), acusing of insufficient register pool volume for the OoOE operation in x64 mode. Newer CPUs, having a higher pool of registers, have less impact when executing x64 code.

- Memory and data cache wasting: Pointers take 64 bits, so unles you're doing your own memory management, with 32-bit offsets instead of using the bulk 64-bit space for adresses, you're wasting more memory, and what is worst: higher data cache usage for the same purpose, with unnecessary CPU-RAM bus overload (remember that OoOE implies data fetching! -imagine a contiguous 32 64-bit pointer vector, taking 2048 bits instead of the 1024 bits that it would take with 32-bit pointers-).

Pros:

However, for some things there is true benefit, and is that the number of registers for SSE operations have been also doubled, from 8 to 16. And because of the nature of the SSE code, which is usually less prone to jump misprediction and with less register aliasing, because of the nature of vector processing code.

Corollarius:

In my opinion a 64-bit operating system makes sense, but an application that doesn't need more than 2GB of RAM, and doesn't need to gain an extra 10% of speed up when running optimized SSE vector code, should be compiled in 32-bit mode.

Re:Serious question

[faragon]

64bit also vastly speeds up long and double math. It doesn't really apply to a browser, but if you were using 64bit integers to store currency amounts, you'd notice a huge speedup. Adding/subtracting from longs is one thing that SSE probably won't help. ;)

No speedup for these reasons, at all:

1) In the case of using 64-bit 2's complement integer registers, you're able to speed-up your 64-bit interger code because operating with 64-bit integers without chaining 32-bit results on the 32-bit CPU case. However, you're missing the point that most heavily computing, such as RSA's big numbers, DES, AES, Blowfish, etc. doesn't use general purpose register but vector SIMD opcodes (e.g. SSE*), already available in the 32-bit mode (with 8 instead of 16 registers, yes), which is faster than 64-bit integer operations.

2) Floating point ("double math") remains almost the same, but with also 8 additional SSE registers.

3) Related to "adding/substracting from longs": In 32-bit mode, a SSE3 -or later- functional unit can execute *four* 32-bit instructions per clock (fetching 128-bit data at once), while already being able to execute from 2 to 4 integer + load/store instructions (e.g. Core2Duo or K8), so it would be faster still while chaining 32-bit results.

Re:Serious question

[micheas]

There is *NO FUCKING REASON* for a POS cash register to need 64 bit hardware or software

I can think of one off the top of my head. SSL.

64bit makes encryption much faster. Of course you are probably one of those people that thinks it is just fine that Best Buy has sent credit card numbers with authorization codes over wifi unencrypted at their stores.

Re:Serious question

[atamido]

One huge benefit of adding the ability for x64 compiling is that it forces developers to clean up their code. Even if there is no benefit to running a program in x64, just cleaning up the code to the point that it compiles properly can fix quite a number of bugs.

Re:Serious question

[rolfwind]

They were still selling 32-bit machines two years ago,

Most processors have been 64bit for many years, despite the OS the machine ships with. The one major exception are most Intel Atoms, which are 32 bit. Even that has to change soon, as 1GB is the minimum for netbooks these days, 2GB going to become standard, and after that, they'll only have one more upgrade cycle before hitting the ram limit.

and people can't reasonably be expected to retire those machines until 2011, and many will still be perfectly useful until 2013 or even 2015 with a few repairs.

And that is why Windows 7 will be the last 32 Microsoft OS. It will last that long, at least on a support basis. 32 bit machines can only access 4GB of RAM theoretically, and the max for XP in reality is around 3GB. 32 bit is a deadend. Going into walmart.com, the minimum I see is 1GB for the cheapest of the cheap systems, standard is 3GB, and several 8GB Ram systems was available for under $750. (Although MS would be smart to make incrememntal upgrades much like Ubuntu and Apple is doing, and not just hotfixes with overly ambitious overhauls every 7 years aka longhorn).

Your machines will be indefinitely usable, but, depending on the niche, desktop software will target desktop hardware like a bellcurve -- and figure the systems are around 5 years old on average. In 2015, that 2010 system will be old to a software seller.

Re:chromium?

[Anonymous Coward]

Speed. If you're running it on Ubuntu I'm not sure how noticable it will be, but on Windows it is so absurdly faster than any browser you've ever ran that you'll be amazed at what its capable of. Also true on Macs, but Safari4 is not much slower.

When you can keep the entire google suite open in tabs, with other rich web2.0 sites open at the same time and have none of them be any less responsive than a desktop app would be, it really changes how you view the web.

On top of that the tab seperation is a killer feature in and of itself. If you're playing a flash game in one tab and click a link on irc that crashes in another tab, the first tab is completely unaffected.

The only thing still missing is a good addon engine, but for day to day browsing I just can not go back to Gecko, and thats after using Gecko since the days of Mozilla M11.

Re:chromium?

[rumith]

You have been misinformed. Aside from the Google Update aka Omaha, Google branding and RLZ tracking (http://niichavo.wordpress.com/2009/04/03/construction-complete/">source), it's still the same:

Chromium is the name we have given to the open source project and the browser source code that we released and maintain at www.chromium.org. One can compile this source code to get a fully working browser. Google takes this source code, and adds on the Google name and logo, an auto-updater system called GoogleUpdate, and RLZ (described later in this post), and calls this Google Chrome.

It's like calling Firefox proprietary because you've been shipped an actual binary that uses their TM'd logo, not the one you'd get by default when compiling it from source. And the auto-update mechanism is Windows-only as far as I know (my Debian install of Chrome correctly integrates into the APT system, as it should be) so it shouldn't worry you a lot if you're an Ubuntu user.

Privacy Issues Disclaimer

[Zero__Kelvin]

Some quick research revealed that there are some missing features with regard to privacy that stopped me from checking it out :-( YMMV.

Re:!x64

[macshit]

The "The names x86-64 or x64 are sometimes used as vendor-neutral terms to collectively refer to the two nearly identical implementations." line has been included in the article since at least October 1st, 2006.

What you quote as the "since 2006" phrasing is "sometimes used", which is certainly true (I've certainly heard people use the term "x64"), but the phrasing you originally used was "often used", which is a quite different. Personally I've only ever heard "x64" from microsofties, but it does seem fairly widespread there, at least in MS development.

If looking for a term which is generally understable, however, "x64" doesn't really work, as it seems to be in widespread use only in certain communities. "x86-64" is a bit better since it has an obvious connection with the term "x86", which is much more widespread than either of the others; so it stands a better chance of being understood even by someone who hasn't seen it before.

Re:amazing work indeed

[Bert64]

I believe it's the JIT engine for javascript, which by it's very nature needs to generate native code and is therefore not so simple to port. The rest of it should compile cleanly on 64, at least webkit does anyway.

Remind me why ad companies need installations?

[qubezz]

An advertising company wants to install software on your computer. They profit by data aggregation and accumulation by seeing where on the map you are interested in going, what friends you chat with, what all the files on your computer are and how you search for them, and oh, everything you look at on the internet. Click here to install.

Wake up

[cecil_turtle]

Chrome supports user scripts, and scripts for all of those things already exist. Check out http://www.adsweep.org/ for one. And it is still "just another browser" like Fx, Opera, IE, etc.

Everyone grits teeth, welcomes Chrome/Chromium

[David Gerard]

Google has released its own Web browser, Chrome, with Linux version Chromium. "We absolutely promise that we only want to completely screw over Microsoft with this, and certainly not Mozilla Firefox," said Google's Sundar Pichai. "That we put a pile of our sponsored Mozilla developers on the project is completely irrelevant. We're not evil, remember."

"We are so, so happy with Google Chrome," mumbled Mozilla CEO John Lilly through gritted teeth. "That most of our income is from Google has no bearing on me making this statement."

Microsoft was unfazed. "Browsers don't need to be integrated with online apps," said marketing developer Ian Moulster. "Certainly not like the operating system ... I'll just get back to you."

Google's new browser will give you their web and email services, photo processing, mapping, office applications that will run in said browser and will make you a cup of tea. This is all paid for by personally-directed text ads in your tea leaves, based on analysing a DNA sample taken when you sip the tea and sending your genetic code back to Google for future targeting.

Pichai stressed that Google would maintain complete confidentiality within the marketing department of whatever the browser accessed concerning your confidential business data, bank account details, medical information and personal preferences in pornography. "We're Google. We know where you live. In a completely not evil way. Sponsored link: Get Chrome Browsers on google.com. Or we'll make you use Bing."

Re:amazing work indeed

[kamatsu]

Right, so you're going to make a machine-code emitting JIT emit code that works on two different architectures without porting? wow! i'd love to see that.