Messages in this thread | | | Date | Tue, 24 Jul 2007 12:08:37 +0530 (IST) | From | Satyam Sharma <> | Subject | Re: [PATCH 6/8] i386: bitops: Don't mark memory as clobbered unnecessarily |
| |
On Tue, 24 Jul 2007, Nick Piggin wrote:
> Satyam Sharma wrote: > > From: Satyam Sharma <ssatyam@cse.iitk.ac.in> > > > > [6/8] i386: bitops: Don't mark memory as clobbered unnecessarily > > > > The goal is to let gcc generate good, beautiful, optimized code. > > > > But test_and_set_bit, test_and_clear_bit, __test_and_change_bit, > > and test_and_change_bit unnecessarily mark all of memory as clobbered, > > thereby preventing gcc from doing perfectly valid optimizations. > > > > The case of __test_and_change_bit() is particularly surprising, given > > that it's a variant where we don't make any guarantees at all. > > __test_and_change_bit is one that you could remove the memory clobber > from.
Yes, for the atomic versions we don't care if we're asking gcc to generate trashy code (even though I'd have wanted to only disallow problematic optimizations -- ones involving the passed bit-string address -- there, and allow other memory references to be optimized as and how the compiler feels like it) because the atomic variants are slow anyway and we probably want to be extra-safe there.
But for the non-atomic variants, it does make sense to remove the memory clobber (and the unneeded __asm__ __volatile__ that another patch did -- for the non-atomic variants, again).
OTOH, as per Linus' review it seems we can drop the "memory" clobber and specify the output operand for the extended asm as "+m". But I must admit I didn't quite understand that at all.
[ I should probably start reading gcc sources, the docs are said to be insufficient/out-of-date, as per the reviews of the patches. ]
Satyam - 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/
| |