kernel/tools/virtio/ringtest, branch linux-4.16.y

Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost

2018-02-08T18:41:00Z

Pull virtio/vhost updates from Michael Tsirkin: "virtio, vhost: fixes, cleanups, features This includes the disk/cache memory stats for for the virtio balloon, as well as multiple fixes and cleanups" * tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost: vhost: don't hold onto file pointer for VHOST_SET_LOG_FD vhost: don't hold onto file pointer for VHOST_SET_VRING_ERR vhost: don't hold onto file pointer for VHOST_SET_VRING_CALL ringtest: ring.c malloc & memset to calloc virtio_vop: don't kfree device on register failure virtio_pci: don't kfree device on register failure virtio: split device_register into device_initialize and device_add vhost: remove unused lock check flag in vhost_dev_cleanup() vhost: Remove the unused variable. virtio_blk: print capacity at probe time virtio: make VIRTIO a menuconfig to ease disabling it all virtio/ringtest: virtio_ring: fix up need_event math virtio/ringtest: fix up need_event math virtio: virtio_mmio: make of_device_ids const. firmware: Use PTR_ERR_OR_ZERO() virtio-mmio: Use PTR_ERR_OR_ZERO() vhost/scsi: Improve a size determination in four functions virtio_balloon: include disk/file caches memory statistics

ringtest: ring.c malloc & memset to calloc

2018-02-01T14:26:46Z

Code cleanup change - moving from malloc & memset to calloc. Signed-off-by: Peter Malone Signed-off-by: Michael S. Tsirkin

virtio/ringtest: virtio_ring: fix up need_event math

2018-01-30T23:47:36Z

last kicked event index must be updated unconditionally: even if we don't need to kick, we do not want to re-check the same entry for events. Reported-by: Cornelia Huck Signed-off-by: Michael S. Tsirkin Reviewed-by: Jens Freimann

virtio/ringtest: fix up need_event math

2018-01-30T23:47:35Z

last kicked event index must be updated unconditionally: even if we don't need to kick, we do not want to re-check the same entry for events. Signed-off-by: Michael S. Tsirkin Acked-by: Cornelia Huck Acked-by: Jason Wang

tools/virtio: fix smp_mb on x86

2018-01-29T17:02:55Z

Offset 128 overlaps the last word of the redzone. Use 132 which is always beyond that. Signed-off-by: Michael S. Tsirkin Signed-off-by: David S. Miller

tools/virtio: copy READ/WRITE_ONCE

2018-01-29T17:02:55Z

This is to make ptr_ring test build again. Signed-off-by: Michael S. Tsirkin Signed-off-by: David S. Miller

tools/virtio: switch to __ptr_ring_empty

2018-01-29T17:02:54Z

We don't rely on lockless guarantees, but it seems cleaner than inverting __ptr_ring_peek. Signed-off-by: Michael S. Tsirkin Signed-off-by: David S. Miller

Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost

2017-12-15T20:56:23Z

Pull virtio regression fixes from Michael Tsirkin: "Fixes two issues in the latest kernel" * tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost: virtio_mmio: fix devm cleanup ptr_ring: fix up after recent ptr_ring changes

ptr_ring: fix up after recent ptr_ring changes

2017-12-08T15:48:38Z

Add more stubs to make it build. Fixes: 81fbfe8a ("ptr_ring: use kmalloc_array()") Signed-off-by: Michael S. Tsirkin

locking/x86: Use LOCK ADD for smp_mb() instead of MFENCE

2017-11-10T12:43:44Z

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: Andy Lutomirski Signed-off-by: Michael S. Tsirkin Acked-by: Peter Zijlstra Acked-by: Thomas Gleixner Cc: Andrew Morton Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Paul E. McKenney 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.com Signed-off-by: Ingo Molnar