On 2/6/24 21:13, Alexander Monakov wrote:
Thank you. I agree we shouldn't penalize non-x86 hosts here, but to be honest
I'd really like to keep this optimization because so many places in Qemu invoke
buffer_is_zero with a constant length, allowing the compiler to optimize out
the length test.
H
On Wed, 7 Feb 2024, Richard Henderson wrote:
> On 2/7/24 06:48, Alexander Monakov wrote:
> > Make buffer_is_zero a 'static inline' function that tests up to three
> > bytes from the buffer before handing off to an unrolled loop. This
> > eliminates call overhead for most non-zero buffers, and al
On 2/7/24 06:48, Alexander Monakov wrote:
Make buffer_is_zero a 'static inline' function that tests up to three
bytes from the buffer before handing off to an unrolled loop. This
eliminates call overhead for most non-zero buffers, and allows to
optimize out length checks when it is known at compi
Make buffer_is_zero a 'static inline' function that tests up to three
bytes from the buffer before handing off to an unrolled loop. This
eliminates call overhead for most non-zero buffers, and allows to
optimize out length checks when it is known at compile time (which is
often the case in Qemu).
