On Thu, Dec 26, 2019 at 04:28:25PM +0100, Håvard Flaget Aasen wrote: > * Package name : clp > Version : 1.17.3+repack1-2
> Changes since the last upload: > > * QA upload > * Add coinor-libclp-doc.links to remove duplicated files > * Set minimum version on runtime-dependencies > * Add patch to check for correct architecture > New release to fix certain issue with the old one, built and tested on six > different architectures this time. Please take a look at the patch to see if > it is acceptable. Hi! I'm afraid that the new check drops optimizations on amd64 -- the arch where 99.9% users are. Your patch is: #ifdef _MSC_VER #include <intrin.h> -#elif defined(__arm__) +#elif defined(__arch64__) #include <arm_neon.h> -#else +#elif defined(__i386__) #include <immintrin.h> //#include <fmaintrin.h> #endif so it checks for __i386__ which is not defined on amd64 (nor x32). There are many defines which would be enough; you can for example check for: #elif defined(__i386__) || defined(__x86_64__) Also, it could be nice to keep the optimized code on armhf as well, which would make the ARM line: #elif defined(__ARM_FEATURE_SIMD32) || defined(__ARM_NEON) Currently, the code _does_ compile but is slow on any arch other than arm64 and i386. And not many people run scientific code on either. Meow! -- ⢀⣴⠾⠻⢶⣦⠀ A MAP07 (Dead Simple) raspberry tincture recipe: 0.5l 95% alcohol, ⣾⠁⢠⠒⠀⣿⡁ 1kg raspberries, 0.4kg sugar; put into a big jar for 1 month. ⢿⡄⠘⠷⠚⠋⠀ Filter out and throw away the fruits (can dump them into a cake, ⠈⠳⣄⠀⠀⠀⠀ etc), let the drink age at least 3-6 months.
