Hi, following patches make rv40 decoding on ARM with NEON approximately 20% faster (measured on omap3 adn omap4). The loop filter dominates now the decoding speed with taking more than 50% of decoding time. I'm working on NEON optimisations for the loop filter now.
Janne _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
