commit | 645b471d619b654d3bacfa8598f759833e08db4e | [log] [tgz] |
---|---|---|
author | Marek Olšák <marek.olsak@amd.com> | Thu Nov 20 22:16:09 2014 +0100 |
committer | Marek Olšák <marek.olsak@amd.com> | Fri Nov 21 12:15:58 2014 +0100 |
tree | df1ae1582454d5b0627669998f8985cfaecc765d | |
parent | 21577571b37e68edc0422fbf80932588a4614abc [diff] |
radeonsi: use minnum and maxnum LLVM intrinsics for MIN and MAX opcodes So far it has been compiled into pretty ugly code (8 instructions or so for either opcode). Reviewed-by: Tom Stellard <thomas.stellard@amd.com>