• Michael S. Tsirkin's avatar
    locking/x86: Use LOCK ADD for smp_mb() instead of MFENCE · 450cbdd0
    Michael S. Tsirkin authored
    MFENCE appears to be way slower than a locked instruction - let's use
    LOCK ADD unconditionally, as we always did on old 32-bit.
    
    Performance testing results:
    
      perf stat -r 10 -- ./virtio_ring_0_9 --sleep --host-affinity 0 --guest-affinity 0
      Before:
             0.922565990 seconds time elapsed                                          ( +-  1.15% )
      After:
             0.578667024 seconds time elapsed                                          ( +-  1.21% )
    
    i.e. about ~60% faster.
    
    Just poking at SP would be the most natural, but if we then read the
    value from SP, we get a false dependency which will slow us down.
    
    This was noted in this article:
    
      http://shipilev.net/blog/2014/on-the-fence-with-dependencies/
    
    And is easy to reproduce by sticking a barrier in a small non-inline
    function.
    
    So let's use a negative offset - which avoids this problem since we
    build with the red zone disabled.
    
    For userspace, use an address just below the redzone.
    
    The one difference between LOCK ADD and MFENCE is that LOCK ADD does
    not affect CLFLUSH, previous patches converted all uses of CLFLUSH to
    call mb(), such that changes to smp_mb() won't affect it.
    
    Update mb/rmb/wmb() on 32-bit to use the negative offset, too, for
    consistency.
    
    As a follow-up, it might be worth considering switching users
    of CLFLUSH to another API (e.g. clflush_mb()?) - we will
    then be able to convert mb() to smp_mb() again.
    
    Also arguably, GCC should switch to use LOCK ADD for __sync_synchronize().
    This might be worth pursuing separately.
    Suggested-by: 's avatarAndy Lutomirski <luto@amacapital.net>
    Signed-off-by: 's avatarMichael S. Tsirkin <mst@redhat.com>
    Acked-by: 's avatarPeter Zijlstra <peterz@infradead.org>
    Acked-by: 's avatarThomas Gleixner <tglx@linutronix.de>
    Cc: Andrew Morton <akpm@linux-foundation.org>
    Cc: Andy Lutomirski <luto@kernel.org>
    Cc: Borislav Petkov <bp@alien8.de>
    Cc: Brian Gerst <brgerst@gmail.com>
    Cc: Denys Vlasenko <dvlasenk@redhat.com>
    Cc: H. Peter Anvin <hpa@zytor.com>
    Cc: Josh Poimboeuf <jpoimboe@redhat.com>
    Cc: Linus Torvalds <torvalds@linux-foundation.org>
    Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
    Cc: qemu-devel@nongnu.org
    Cc: virtualization@lists.linux-foundation.org
    Link: http://lkml.kernel.org/r/1509118355-4890-1-git-send-email-mst@redhat.comSigned-off-by: 's avatarIngo Molnar <mingo@kernel.org>
    450cbdd0
Name
Last commit
Last update
..
Makefile Loading commit data...
README Loading commit data...
main.c Loading commit data...
main.h Loading commit data...
noring.c Loading commit data...
ptr_ring.c Loading commit data...
ring.c Loading commit data...
run-on-all.sh Loading commit data...
virtio_ring_0_9.c Loading commit data...
virtio_ring_inorder.c Loading commit data...
virtio_ring_poll.c Loading commit data...