Re: [tip:x86/asm] x86: Fix and improvepercpu_cmpxchg{8,16}b_double()

[Christoph Lameter]

On Thu, 15 Dec 2011, tip-bot for Jan Beulich wrote:

> x86: Fix and improve percpu_cmpxchg{8,16}b_double()

Great. Thanks. I wish you would have been there when I had to put this
together. Had a difficult time finding relevant docs etc on how to exactly
do this. And the failure messages from the inline asm parser were not that
helpful.

Acked-by: Christoph Lameter <cl@xxxxxxxxx>

> The __dummy variable was pointless (and needlessly initialized
> in the 2x32-bit case), given that local copies of the inputs
> already exist.

Hmm... I had some failures if I did not specify that dummy in the
inline asm. Does this work for all gcc versions?

> The 2x64-bit variant forced the address of the first object into
> %rsi, even though this is needed only for the call to the
> emulation function. The real cmpxchg16b can operate on an
> memory.

Yup. Good idea to code the load into the alternative code path to avoid
the cmpxchg of the primary code path to be restricted to %si register.

You dropped the padding with NOPs. Are the instructions on both paths
always the same length?

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/