FINUFFT_EXECUTE architecture dispatch

[DiamonDinoia]

It is possible to compile the code in finufft_execute for multiple SIMD instructions and select the fastest available at runtime. This might not impact power users (that compile the code themselves) but it will impact python/matlab users are well as developers that cannot use -march=native.

Once this is merged we might consider shipping an linux artifacts on package managers (APT and similars) as these will run with good perfomance everywhere.

[ahbarnett]

These are all great ideas for v 2.4, 2.5, etc. Let's focus on the currently-nearly-done things and get 2.3 stable.

…

On Wed, Jul 17, 2024 at 11:23 AM Marco Barbone ***@***.***> wrote: It is possible to compile the code in finufft_execute for multiple SIMD instructions and select the fastest available at runtime. This might not impact power users (that compile the code themselves) but it will impact python/matlab users are well as developers that cannot use -march=native. Once this is merged we might consider shipping an linux artifacts on package managers (APT and similars) as these will run with good perfomance everywhere. — Reply to this email directly, view it on GitHub <#486>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACNZRSQDVSWNXEGHZ3QWBH3ZM2D5VAVCNFSM6AAAAABLA3YUS6VHI2DSMVQWIX3LMV43ASLTON2WKOZSGQYTGOJRGM3TCNQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

-- *-------------------------------------------------------------------~^`^~._.~' |\ Alex Barnett Center for Computational Mathematics, Flatiron Institute | \ http://users.flatironinstitute.org/~ahb 646-876-5942