Android NDK: GCC is now deprecated, everyone should be switching to Clang

Post by krapp » Mon, Dec 21, 2015, 10:58 PM UTC

simondedalus wrote: post #62

krapp wrote: post #40

Earlier quoted context omitted.

Compiling source code you don't comprehend and aren't capable of rewriting isn't really an exercise of freedom. You could just as well distribute a closed binary with runtime configuration options, for all it matters in that case.

if you're implying that everyone who usefully modifies source code or compiles with options is able to write the code from scratch, i would advise you rethink your position. to put a finer point on it, the concept of "layers of abstraction" is fundamental to computers and software. i can know quite a bit about adjusting the pieces of a program without having the ability to build it as effectively as i modify it.

Post by eddyb » Mon, Dec 21, 2015, 10:59 PM UTC

onedognight wrote: post #68

salgernon wrote: post #34

Earlier quoted context omitted.

Compiler bugs are usually limited to a single compilation unit and are usually submitted by doing a binary search on a pre processed source file. This is certainly true for closed source as well - just redact the (few) symbol names and create something that will compile to an a.out that demonstrates the problem. God saves a puppy anytime someone filed a compiler bug with a trivial test case that demonstrates the prob…

> God saves a puppy anytime someone filed a compiler bug with a trivial test case that demonstrates the problem. The current (r10e) shipping version of clang cannot return 128bit floats from a function. #include int main(int c, char **v) { std::stringstream("1.5"); double x; s >> x; return x == 1.5 ? 0 : 1; } EDIT: You don't need "long double" in the above. double (or float) will show the problem. The standard librar…

Post by krupan » Mon, Dec 21, 2015, 11:04 PM UTC

drzaiusapelord wrote: post #75

jordigh wrote: post #63

Earlier quoted context omitted.

I wonder what kind of world we would have if GNU were completely defeated, if no one released anything under the GPL anymore. Sure, free software would still exist, but nobody would call it "free software" and no one would prevent anyone else from creating non-free derivatives. Nearly all software would be "open core", with some free software here and there but with many or most of the interesting components non-free…

We used to call that "public domain software" and it was big in the 80s from what I recall (I was a kid then). You'd release something to the public domain and people can do with it whatever they wished. Then GNU came around and ate everyone's lunch, for better or worse. I suspect for better. Maybe GNU is just a stop-gap movement to get everyone on board the FOSS train, who knows. BSD licensing hasn't caused any apoc…

Post by sdegutis » Mon, Dec 21, 2015, 11:05 PM UTC

bracewel wrote: post #3

waits for RMS to get mad https://gcc.gnu.org/ml/gcc/2014-01/msg00247.html

Post by Nadya » Mon, Dec 21, 2015, 11:07 PM UTC

eropple wrote: post #69

_delirium wrote: post #21

Earlier quoted context omitted.

> And to preclude modifications which prevent him from effectively monetizing his labour is to infringe on his freedom. This sounds a lot like the people saying that adblocker software is infringing the freedom of people trying to monetize their websites (or maybe even "stealing" from them). And I think the same response applies: the fact that your business model isn't compatible with adblocker software, or with my c…

> Plus it seems extra presumptuous to claim some kind of inherent right to profit off derivative works of my software. Who's claiming that? On the other hand, I've met more than a few pro-FSF people who seem to think it's cromulent to claim that I don't have the right to profit off my work (which is insane; you can pick whatever license you'd like, copyleft or not, and so can I).