4 ms·

Inside JavaScriptCore's low-level interpreter

61 points by p4bl0 14y ago

acqq 14y ago
For those who don't know JavaScriptCore under that name: JavaScriptCore, the WebKit JS implementation http://wingolog.org/archives/2011/10/28/javascriptcore-the-webkit-js-implementation http://wingolog.org/archives/2011/10/28/javascriptcore-the-w... "In 2008, the WebKit folks (...) called the thing SquirrelFish Extreme. Marketers called this Nitro Extreme. (Still, the proper name of the engine is JavaScriptCore; Wikipedia currently gets this one wrong.)" http://en.wikipedia.org/wiki/WebKit#JavaScriptCore http://en.wikipedia.org/wiki/WebKit#JavaScriptCore "JavaScriptCore is a framework that provides a JavaScript engine for WebKit implementations, and provides this type of scripting in other contexts within Mac OS X." Probably should be "within Safari browser." But as the first linked article notes: "JSC folks were doing great, well-factored work. It was so good that SpiderMonkey hackers at Mozilla adopted JSC's regexp JIT compiler and their native-code assembler directly." I believe then it should be "as the whole within Safari browser, and some key parts also within Firefox browser."
- chc 14y ago
  What is it you think should say "within Safari browser"? JavaScriptCore is certainly not inside Safari; it's a part of WebKit. JavaScriptCore is no more "within Safari" than WebKit itself is. But I'm not sure what else you could be referring to.
  - acqq 14y ago
    The Wikipedia article, at the moment I quoted from it related to Max OS X but not mentioning Safari at all, and Safari runs on other platforms: iOS and Windows." Moreover the conclusion that "the JS engine is used wherever WebKit is used" would be false. Wwhen describing the use of the given engine, it's not a good description to say "the engine in WebKit" as there is an example of the very popular use of WebKit without the engine: Google Chrome.
    - chc 14y ago
      Neither I nor Wikipedia said "used wherever WebKit is used". The Wikipedia article said "JavaScript engine for WebKit implementations" and I said "part of WebKit". The fact that Chrome chooses not to use it is immaterial. It is part of WebKit. Check out the WebKit source and you will get the JavaScriptCore source as part of it. Similarly, there's a GIMP-based image editor called Seashore that doesn't use a lot of GIMP's UI, but I still think it's fair to say that the UI code is part of GIMP.
acqq 14y ago
In short: in order to write the fast interpreter for the bytecode of JavaScript in WebKit JS's JavaScriptCore, the author developed a custom variant of an assembly language, wrote the interpreter in it and wrote in Ruby the translator from his custom assembly to the native assembly (allowing him to target different processor architectures like x32 x64 and ARM, at least that's how I understood it from the description). He explains how with C++ he could not get the speed gains he achieved with assembly. One real life example where writing exact assembly code still matters. Coming to a browser near you.
- wingo 14y ago
  No offense intended, but this comment does not add very much while misunderstanding some other things. For example, I did not write the LLInt. It was mostly written by Apple's Filip Pizlo. If this is the sort of topic you are interested in, then you can have the patience to read a few paragraphs. If it's not, then summarizing isn't going to help :)
  - acqq 14y ago
    Thanks, now I know that LLInt (the new interpreter) was written by Filip Pizlo. Note that you mention him in the article only related to DFG JIT. I didn't try to investigate the source files or checkins, I tried to understand only what you write, and that was obviously not clear enough. I hope you see that I'm interested in the topic but read about it the first time. I hope you also can imagine that there are more readers like me. And they would benefit from the correct summary too. Please do write what else I misunderstood, that is exactly the reason why I wrote the short summary, to get the feedback, not to claim that I understand more than you or any insider. It's short, counting the words some 20 times shorter than your article, so I hope it wouldn't be hard for you to point to any other inconsistency. Summarizing helps if the result is something relevant and clear, in a few sentences instead of 1700 words. Without the summary, the most important points can be overlooked/misunderstood by anybody not "close to the sources."
    - wingo 14y ago
      Well, I guess the one thing I would correct is the "why". The LLInt doesn't just produce assembler to be fast, though it is faster than the old interpreter. The real reason it produces assembler is to control the stack representation, so that it works better with tiering, exceptions, and the optimizing compiler (DFG). Otherwise, interpreting was a lose, because tiering up cost too much. As you can see, the situation is a bit complicated. If I could have made the article shorter, I would have :)

61 points by p4bl0 14y ago

acqq 14y ago

For those who don't know JavaScriptCore under that name: JavaScriptCore, the WebKit JS implementation http://wingolog.org/archives/2011/10/28/javascriptcore-the-webkit-js-implementation http://wingolog.org/archives/2011/10/28/javascriptcore-the-w... "In 2008, the WebKit folks (...) called the thing SquirrelFish Extreme. Marketers called this Nitro Extreme. (Still, the proper name of the engine is JavaScriptCore; Wikipedia currently gets this one wrong.)" http://en.wikipedia.org/wiki/WebKit#JavaScriptCore http://en.wikipedia.org/wiki/WebKit#JavaScriptCore "JavaScriptCore is a framework that provides a JavaScript engine for WebKit implementations, and provides this type of scripting in other contexts within Mac OS X." Probably should be "within Safari browser." But as the first linked article notes: "JSC folks were doing great, well-factored work. It was so good that SpiderMonkey hackers at Mozilla adopted JSC's regexp JIT compiler and their native-code assembler directly." I believe then it should be "as the whole within Safari browser, and some key parts also within Firefox browser."

chc 14y ago
What is it you think should say "within Safari browser"? JavaScriptCore is certainly not inside Safari; it's a part of WebKit. JavaScriptCore is no more "within Safari" than WebKit itself is. But I'm not sure what else you could be referring to.
- acqq 14y ago
  The Wikipedia article, at the moment I quoted from it related to Max OS X but not mentioning Safari at all, and Safari runs on other platforms: iOS and Windows." Moreover the conclusion that "the JS engine is used wherever WebKit is used" would be false. Wwhen describing the use of the given engine, it's not a good description to say "the engine in WebKit" as there is an example of the very popular use of WebKit without the engine: Google Chrome.
  - chc 14y ago
    Neither I nor Wikipedia said "used wherever WebKit is used". The Wikipedia article said "JavaScript engine for WebKit implementations" and I said "part of WebKit". The fact that Chrome chooses not to use it is immaterial. It is part of WebKit. Check out the WebKit source and you will get the JavaScriptCore source as part of it. Similarly, there's a GIMP-based image editor called Seashore that doesn't use a lot of GIMP's UI, but I still think it's fair to say that the UI code is part of GIMP.

acqq 14y ago

In short: in order to write the fast interpreter for the bytecode of JavaScript in WebKit JS's JavaScriptCore, the author developed a custom variant of an assembly language, wrote the interpreter in it and wrote in Ruby the translator from his custom assembly to the native assembly (allowing him to target different processor architectures like x32 x64 and ARM, at least that's how I understood it from the description). He explains how with C++ he could not get the speed gains he achieved with assembly. One real life example where writing exact assembly code still matters. Coming to a browser near you.

wingo 14y ago
No offense intended, but this comment does not add very much while misunderstanding some other things. For example, I did not write the LLInt. It was mostly written by Apple's Filip Pizlo. If this is the sort of topic you are interested in, then you can have the patience to read a few paragraphs. If it's not, then summarizing isn't going to help :)
- acqq 14y ago
  Thanks, now I know that LLInt (the new interpreter) was written by Filip Pizlo. Note that you mention him in the article only related to DFG JIT. I didn't try to investigate the source files or checkins, I tried to understand only what you write, and that was obviously not clear enough. I hope you see that I'm interested in the topic but read about it the first time. I hope you also can imagine that there are more readers like me. And they would benefit from the correct summary too. Please do write what else I misunderstood, that is exactly the reason why I wrote the short summary, to get the feedback, not to claim that I understand more than you or any insider. It's short, counting the words some 20 times shorter than your article, so I hope it wouldn't be hard for you to point to any other inconsistency. Summarizing helps if the result is something relevant and clear, in a few sentences instead of 1700 words. Without the summary, the most important points can be overlooked/misunderstood by anybody not "close to the sources."
  - wingo 14y ago
    Well, I guess the one thing I would correct is the "why". The LLInt doesn't just produce assembler to be fast, though it is faster than the old interpreter. The real reason it produces assembler is to control the stack representation, so that it works better with tiering, exceptions, and the optimizing compiler (DFG). Otherwise, interpreting was a lose, because tiering up cost too much. As you can see, the situation is a bit complicated. If I could have made the article shorter, I would have :)