Comments on: gcc version comparison, part 1/n: libxul sizes http://blog.mozilla.org/nfroyd/2013/02/01/gcc-version-comparison-part-1n-libxul-sizes/ Improving performance at Mozilla Sat, 20 Apr 2013 10:00:40 +0000 hourly 1 http://wordpress.org/?v=3.5.1 By: Nathan Froyd http://blog.mozilla.org/nfroyd/2013/02/01/gcc-version-comparison-part-1n-libxul-sizes/#comment-896 Nathan Froyd Sun, 03 Feb 2013 02:39:39 +0000 http://blog.mozilla.org/nfroyd/?p=133#comment-896 A clang comparison will hopefully be forthcoming as well.

]]> By: Nathan Froyd http://blog.mozilla.org/nfroyd/2013/02/01/gcc-version-comparison-part-1n-libxul-sizes/#comment-895 Nathan Froyd Sun, 03 Feb 2013 02:39:04 +0000 http://blog.mozilla.org/nfroyd/?p=133#comment-895 Yeah, turns out the linker is responsible for handling this, but–on my machine, at least–Mozilla’s configury determines that those bits of the linker should be disabled so as to not harm the debugging experience. However, we don’t care about the debugging experience in this case…

]]> By: gcc version comparison, part 1.5/n: corrections « Nathan's Blog http://blog.mozilla.org/nfroyd/2013/02/01/gcc-version-comparison-part-1n-libxul-sizes/#comment-893 gcc version comparison, part 1.5/n: corrections « Nathan's Blog Sun, 03 Feb 2013 02:34:12 +0000 http://blog.mozilla.org/nfroyd/?p=133#comment-893 [...] my previous post, I discussed the size of libxul as compiled by various versions of GCC.  Due to some configuration [...]

]]> By: Kyle Huey http://blog.mozilla.org/nfroyd/2013/02/01/gcc-version-comparison-part-1n-libxul-sizes/#comment-892 Kyle Huey Sun, 03 Feb 2013 00:30:02 +0000 http://blog.mozilla.org/nfroyd/?p=133#comment-892 MSVC has declspec novtable for exactly this case.

]]> By: Robert Kaiser http://blog.mozilla.org/nfroyd/2013/02/01/gcc-version-comparison-part-1n-libxul-sizes/#comment-886 Robert Kaiser Sat, 02 Feb 2013 14:50:29 +0000 http://blog.mozilla.org/nfroyd/?p=133#comment-886 Now that’s interesting, esp. as our official builds for Linux right now are using gcc 4.5, which apparently creates the largest size of all. Will be interesting to see how perf compares.

Oh, and if 4.8 improves in perf as well as size, it sounds like it will be awesome!

(Also, how does clang compare?)

]]>