summaryrefslogtreecommitdiff
path: root/Documentation/memory-barriers.txt
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2016-01-19 03:44:24 +0300
committerLinus Torvalds <torvalds@linux-foundation.org>2016-01-19 03:44:24 +0300
commita200dcb34693084e56496960d855afdeaaf9578f (patch)
treebf65e4350460b7f98247278469f7600d1808c3fc /Documentation/memory-barriers.txt
parentd05d82f7110b08fd36178a641b69a1f206e1142b (diff)
parent43e361f23c49dbddf74f56ddf6cdd85c5dbff6da (diff)
downloadlinux-a200dcb34693084e56496960d855afdeaaf9578f.tar.xz
Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost
Pull virtio barrier rework+fixes from Michael Tsirkin: "This adds a new kind of barrier, and reworks virtio and xen to use it. Plus some fixes here and there" * tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost: (44 commits) checkpatch: add virt barriers checkpatch: check for __smp outside barrier.h checkpatch.pl: add missing memory barriers virtio: make find_vqs() checkpatch.pl-friendly virtio_balloon: fix race between migration and ballooning virtio_balloon: fix race by fill and leak s390: more efficient smp barriers s390: use generic memory barriers xen/events: use virt_xxx barriers xen/io: use virt_xxx barriers xenbus: use virt_xxx barriers virtio_ring: use virt_store_mb sh: move xchg_cmpxchg to a header by itself sh: support 1 and 2 byte xchg virtio_ring: update weak barriers to use virt_xxx Revert "virtio_ring: Update weak barriers to use dma_wmb/rmb" asm-generic: implement virt_xxx memory barriers x86: define __smp_xxx xtensa: define __smp_xxx tile: define __smp_xxx ...
Diffstat (limited to 'Documentation/memory-barriers.txt')
-rw-r--r--Documentation/memory-barriers.txt28
1 files changed, 23 insertions, 5 deletions
diff --git a/Documentation/memory-barriers.txt b/Documentation/memory-barriers.txt
index a61be39c7b51..904ee42d078e 100644
--- a/Documentation/memory-barriers.txt
+++ b/Documentation/memory-barriers.txt
@@ -1655,17 +1655,18 @@ macro is a good place to start looking.
SMP memory barriers are reduced to compiler barriers on uniprocessor compiled
systems because it is assumed that a CPU will appear to be self-consistent,
and will order overlapping accesses correctly with respect to itself.
+However, see the subsection on "Virtual Machine Guests" below.
[!] Note that SMP memory barriers _must_ be used to control the ordering of
references to shared memory on SMP systems, though the use of locking instead
is sufficient.
Mandatory barriers should not be used to control SMP effects, since mandatory
-barriers unnecessarily impose overhead on UP systems. They may, however, be
-used to control MMIO effects on accesses through relaxed memory I/O windows.
-These are required even on non-SMP systems as they affect the order in which
-memory operations appear to a device by prohibiting both the compiler and the
-CPU from reordering them.
+barriers impose unnecessary overhead on both SMP and UP systems. They may,
+however, be used to control MMIO effects on accesses through relaxed memory I/O
+windows. These barriers are required even on non-SMP systems as they affect
+the order in which memory operations appear to a device by prohibiting both the
+compiler and the CPU from reordering them.
There are some more advanced barrier functions:
@@ -2948,6 +2949,23 @@ The Alpha defines the Linux kernel's memory barrier model.
See the subsection on "Cache Coherency" above.
+VIRTUAL MACHINE GUESTS
+-------------------
+
+Guests running within virtual machines might be affected by SMP effects even if
+the guest itself is compiled without SMP support. This is an artifact of
+interfacing with an SMP host while running an UP kernel. Using mandatory
+barriers for this use-case would be possible but is often suboptimal.
+
+To handle this case optimally, low-level virt_mb() etc macros are available.
+These have the same effect as smp_mb() etc when SMP is enabled, but generate
+identical code for SMP and non-SMP systems. For example, virtual machine guests
+should use virt_mb() rather than smp_mb() when synchronizing against a
+(possibly SMP) host.
+
+These are equivalent to smp_mb() etc counterparts in all other respects,
+in particular, they do not control MMIO effects: to control
+MMIO effects, use mandatory barriers.
============
EXAMPLE USES