Messages in this thread | | | Date | Tue, 20 Dec 2005 14:04:07 -0800 (PST) | From | Linus Torvalds <> | Subject | Re: [patch 04/15] Generic Mutex Subsystem, add-atomic-call-func-x86_64.patch |
| |
On Tue, 20 Dec 2005, Nicolas Pitre wrote: > > I mean...... what is it with mutexes that you dislike to the point of > bending backward that far, and even after seeing the numbers, with such > a semaphore implementation that _I_ even wouldn't trust people to use > correctly?
Quite frankly, what has disgusted me about this mutex discussion is the totally specious arguments for the new mutexes that just rubs me entirely the wrong way.
If it had _started_ with a mutex implementation that was faster, simpler, and didn't rename the old and working semaphores, I'd have been perfectly fine with it.
As it is, the discussion has been pretty much everything but that.
And then people who argue about single cycles, end up dismissing the single cycles when I argue that "ld+st" is faster - like you just did.
Be consistent, dammit. If single cycles matter, they matter. If they don't, then the existing code is better, since it's existing and works. You can't have it both ways.
In other words: if people didn't mix up issues that had nothing to do with this into it, I'd be happier. I've already said that a mutex that does _not_ replace semaphore (and doesn't mess with naming) is acceptable.
We've done that before. But do it RIGHT, dammit. And don't mix existing semaphores into it (for example, completions didn't change any old users).
Linus - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |