• Ma Ling's avatar
    x86, mem: Optimize memmove for small size and unaligned cases · 3b4b682b
    Ma Ling authored
    movs instruction will combine data to accelerate moving data,
    however we need to concern two cases about it.
    
    1. movs instruction need long lantency to startup,
       so here we use general mov instruction to copy data.
    2. movs instruction is not good for unaligned case,
       even if src offset is 0x10, dest offset is 0x0,
       we avoid and handle the case by general mov instruction.
    Signed-off-by: default avatarMa Ling <ling.ma@intel.com>
    LKML-Reference: <1284664360-6138-1-git-send-email-ling.ma@intel.com>
    Signed-off-by: default avatarH. Peter Anvin <hpa@zytor.com>
    3b4b682b
memmove_64.c 3.75 KB