Hi Chris,
I'm going to crawl out for the woodwork for a few minutes and comment.
On Mon, Apr 21, 2014, at 10:52, Chris Peterson wrote:
> One of Mozilla's 2014 goals is to "Increase active contributors to=20
> Mozilla=E2=80=99s target initiatives by 10x". [1] I'm not sure where the =
10x=20
> number comes from, but the intention is clear. SpiderMonkey is probably=20
> not the ideal component for new contributors to start hacking on Gecko,=20
> but there are exceptions and we can suggest good second bugs to keep=20
> current contributors active.
Actually, I think SpiderMonkey is a rather good place to start hacking
on at least if you know C++. The codebase is self-contained and has been
modernized nicely over the last few years. SM also doesn't take forever
to build, and its tests are fairly easy to write and run. Additionally,
unlike some Gecko modules, it's very active, so reviewers are not hard
to find.
>=20
> Mentored bugs and "good first bugs" are the on-ramp for new=20
> contributors. Many new contributors find bugs through Josh Matthew's=20
> Bugs Ahoy [2] search engine. Unfortunately, Bugs Ahoy only searched one=20
> of SpiderMonkey's five Bugzilla components. I submitted a pull request=20
> [3] to fix this.
>=20
>=20
> * The following bugs have [good first bug] have no [mentor=3D]. We should=
=20
> always have a mentor contact for good first bugs. Any volunteers? :)
>=20
> 689368 - Investigate potential LifoAlloc performance improvements
> 513363 - Improve n-way string concat (when elements are not strings)
>=20
>=20
> * The following bugs are have [mentor=3D] but not [good first bugs]. Are=
=20
> any of them good first bugs? Are these bugs still relevant?
>=20
> 583083 Let the eval "//# sourceURL=3D..." trick work for stack traces
> 715181 Self-host Array.sort
> 852187 Optimize postMessage for large strings
> 894092 IonMonkey: clean-up: Remove VMFunction copy constructor.
> 905210 VMFunctions setup for JIT generates large static initializers
> 920484 Intl prototype initialisation requires empty object
> 950612 Warn about inefficient register-allocation for CallWithABI
> 952650 Kill jsval and all methods like JSVAL_IS_* and JSVAL_TO_*
> 956299 IonMonkey: Clean-up: Abstract trivial callVM codegen.
> 970865 Use GetTempRegForIntArg instead of CallTempReg* for passABIArg
> 974525 IonMonkey: ARM: Try optimizing ma_mod_mask by avoiding branches
> 976110 IonMonkey: Optimize integer divide by constant
> 977966 Optimize String#split('foo').join('bar') pattern
> 989348 BaselineCompiler: Optimize JSOP_ARRAYPUSH
> 995278 JS_NewArrayBufferContents frees user data on error
> 997274 Value.h should use NUNBOX32/PUNBOX64 instead of JS_BITS_PER_WORD
> 998485 Replace EdgeCaseAnalysis with RangeAnalysis
>=20
>=20
> * Do you know of any other bugs (in Bugzilla or in your head) that could=
=20
> be mentored? Are any of those good first or second bugs?
I'm rather out of the loop, so I can't mention anything specifically.
However, if there are any small harmony features to be implemented,
those can be a lot of fun for new contributors. (You can tell your
friends "I implemented that" when they start using a cool new JS
feature.) See the "harmony" meta-bug perhaps.
_______________________________________________
dev-tech-js-engine-internals mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-tech-js-engine-internals
