openkylin
/
linux
mirror of https://gitee.com/openkylin/linux.git
9
0
Fork 0
Code Issues Projects Releases Wiki Activity

virtio: barrier rework+fixes 

This adds a new kind of barrier, and reworks virtio and xen
 to use it.
 Plus some fixes here and there.
 
 Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1
 
 iQEcBAABAgAGBQJWlU2kAAoJECgfDbjSjVRpZ6IH/Ra19ecG8sCQo9zskr4zo22Z
 DZXC3u0sJDBYjjBAiw3IY1FKh7wx2Fr1RhUOj1bteBgcFCMCV1zInP5ITiCyzd1H
 YYh1w9C2tZaj2T4t9L4hIrAdtIF8fGS+oI2IojXPjOuDLEt6pfFBEjHp/sfl3UJq
 ZmZvw4OXviSNej7jBw8Xni3Uv18yfmLGXvMdkvMSPC1/XL29voGDqTVwhqJwxLVz
 k/ZLcKFOzIs9N7Nja0Jl1EiZtC2Y9cpItqweicNAzszlpkSL44vQxmCSefB+WyQ4
 gt0O3+AxYkLfrxzCBhUA4IpRex3/XPW1b+1e/V1XjfR2n/FlyLe+AIa8uPJElFc=
 =ukaV
 -----END PGP SIGNATURE-----

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
  ...
This commit is contained in:
Linus Torvalds 2016-01-18 16:44:24 -08:00
parent d05d82f711 43e361f23c
commit a200dcb346
44 changed files with 400 additions and 318 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
Documentation/memory-barriers.txt
View File

@ -1655,17 +1655,18 @@ macro is a good place to start looking.
SMP memory barriers are reduced to compiler barriers on uniprocessor compiled 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, systems because it is assumed that a CPU will appear to be self-consistent,
and will order overlapping accesses correctly with respect to itself. 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 [!] 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 references to shared memory on SMP systems, though the use of locking instead
is sufficient. is sufficient.
Mandatory barriers should not be used to control SMP effects, since mandatory Mandatory barriers should not be used to control SMP effects, since mandatory
barriers unnecessarily impose overhead on UP systems. They may, however, be barriers impose unnecessary overhead on both SMP and UP systems. They may,
used to control MMIO effects on accesses through relaxed memory I/O windows. however, be used to control MMIO effects on accesses through relaxed memory I/O
These are required even on non-SMP systems as they affect the order in which windows. These barriers are required even on non-SMP systems as they affect
memory operations appear to a device by prohibiting both the compiler and the the order in which memory operations appear to a device by prohibiting both the
CPU from reordering them. compiler and the CPU from reordering them.
There are some more advanced barrier functions: 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. 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 EXAMPLE USES

35
arch/arm/include/asm/barrier.h
View File

@ -60,38 +60,11 @@ extern void arm_heavy_mb(void);
#define dma_wmb() barrier() #define dma_wmb() barrier()
#endif #endif
#ifndef CONFIG_SMP #define __smp_mb() dmb(ish)
#define smp_mb() barrier() #define __smp_rmb() __smp_mb()
#define smp_rmb() barrier() #define __smp_wmb() dmb(ishst)
#define smp_wmb() barrier()
#else
#define smp_mb() dmb(ish)
#define smp_rmb() smp_mb()
#define smp_wmb() dmb(ishst)
#endif
#define smp_store_release(p, v) \ #include <asm-generic/barrier.h>
do { \
compiletime_assert_atomic_type(*p); \
smp_mb(); \
WRITE_ONCE(*p, v); \
} while (0)
#define smp_load_acquire(p) \
({ \
typeof(*p) ___p1 = READ_ONCE(*p); \
compiletime_assert_atomic_type(*p); \
smp_mb(); \
___p1; \
})
#define read_barrier_depends() do { } while(0)
#define smp_read_barrier_depends() do { } while(0)
#define smp_store_mb(var, value) do { WRITE_ONCE(var, value); smp_mb(); } while (0)
#define smp_mb__before_atomic() smp_mb()
#define smp_mb__after_atomic() smp_mb()
#endif /* !__ASSEMBLY__ */ #endif /* !__ASSEMBLY__ */
#endif /* __ASM_BARRIER_H */ #endif /* __ASM_BARRIER_H */

19
arch/arm64/include/asm/barrier.h
View File

@ -35,11 +35,11 @@
#define dma_rmb() dmb(oshld) #define dma_rmb() dmb(oshld)
#define dma_wmb() dmb(oshst) #define dma_wmb() dmb(oshst)
#define smp_mb() dmb(ish) #define __smp_mb() dmb(ish)
#define smp_rmb() dmb(ishld) #define __smp_rmb() dmb(ishld)
#define smp_wmb() dmb(ishst) #define __smp_wmb() dmb(ishst)
#define smp_store_release(p, v) \ #define __smp_store_release(p, v) \
do { \ do { \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
switch (sizeof(*p)) { \ switch (sizeof(*p)) { \
@ -62,7 +62,7 @@ do { \
} \ } \
} while (0) } while (0)
#define smp_load_acquire(p) \ #define __smp_load_acquire(p) \
({ \ ({ \
union { typeof(*p) __val; char __c[1]; } __u; \ union { typeof(*p) __val; char __c[1]; } __u; \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
@ -91,14 +91,7 @@ do { \
__u.__val; \ __u.__val; \
}) })
#define read_barrier_depends() do { } while(0) #include <asm-generic/barrier.h>
#define smp_read_barrier_depends() do { } while(0)
#define smp_store_mb(var, value) do { WRITE_ONCE(var, value); smp_mb(); } while (0)
#define nop() asm volatile("nop");
#define smp_mb__before_atomic() smp_mb()
#define smp_mb__after_atomic() smp_mb()
#endif /* __ASSEMBLY__ */ #endif /* __ASSEMBLY__ */

4
arch/blackfin/include/asm/barrier.h
View File

@ -78,8 +78,8 @@
#endif /* !CONFIG_SMP */ #endif /* !CONFIG_SMP */
#define smp_mb__before_atomic() barrier() #define __smp_mb__before_atomic() barrier()
#define smp_mb__after_atomic() barrier() #define __smp_mb__after_atomic() barrier()
#include <asm-generic/barrier.h> #include <asm-generic/barrier.h>

24
arch/ia64/include/asm/barrier.h
View File

@ -42,34 +42,24 @@
#define dma_rmb() mb() #define dma_rmb() mb()
#define dma_wmb() mb() #define dma_wmb() mb()
#ifdef CONFIG_SMP # define __smp_mb() mb()
# define smp_mb() mb()
#else
# define smp_mb() barrier()
#endif
#define smp_rmb() smp_mb() #define __smp_mb__before_atomic() barrier()
#define smp_wmb() smp_mb() #define __smp_mb__after_atomic() barrier()
#define read_barrier_depends() do { } while (0)
#define smp_read_barrier_depends() do { } while (0)
#define smp_mb__before_atomic() barrier()
#define smp_mb__after_atomic() barrier()
/* /*
* IA64 GCC turns volatile stores into st.rel and volatile loads into ld.acq no * IA64 GCC turns volatile stores into st.rel and volatile loads into ld.acq no
* need for asm trickery! * need for asm trickery!
*/ */
#define smp_store_release(p, v) \ #define __smp_store_release(p, v) \
do { \ do { \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
barrier(); \ barrier(); \
WRITE_ONCE(*p, v); \ WRITE_ONCE(*p, v); \
} while (0) } while (0)
#define smp_load_acquire(p) \ #define __smp_load_acquire(p) \
({ \ ({ \
typeof(*p) ___p1 = READ_ONCE(*p); \ typeof(*p) ___p1 = READ_ONCE(*p); \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
@ -77,12 +67,12 @@ do { \
___p1; \ ___p1; \
}) })
#define smp_store_mb(var, value) do { WRITE_ONCE(var, value); smp_mb(); } while (0)
/* /*
* The group barrier in front of the rsm & ssm are necessary to ensure * The group barrier in front of the rsm & ssm are necessary to ensure
* that none of the previous instructions in the same group are * that none of the previous instructions in the same group are
* affected by the rsm/ssm. * affected by the rsm/ssm.
*/ */
#include <asm-generic/barrier.h>
#endif /* _ASM_IA64_BARRIER_H */ #endif /* _ASM_IA64_BARRIER_H */

6
arch/ia64/kernel/iosapic.c
View File

@ -256,7 +256,7 @@ set_rte (unsigned int gsi, unsigned int irq, unsigned int dest, int mask)
} }
static void static void
nop (struct irq_data *data) iosapic_nop (struct irq_data *data)
{ {
/* do nothing... */ /* do nothing... */
} }
@ -415,7 +415,7 @@ iosapic_unmask_level_irq (struct irq_data *data)
#define iosapic_shutdown_level_irq mask_irq #define iosapic_shutdown_level_irq mask_irq
#define iosapic_enable_level_irq unmask_irq #define iosapic_enable_level_irq unmask_irq
#define iosapic_disable_level_irq mask_irq #define iosapic_disable_level_irq mask_irq
#define iosapic_ack_level_irq nop #define iosapic_ack_level_irq iosapic_nop
static struct irq_chip irq_type_iosapic_level = { static struct irq_chip irq_type_iosapic_level = {
.name = "IO-SAPIC-level", .name = "IO-SAPIC-level",
@ -453,7 +453,7 @@ iosapic_ack_edge_irq (struct irq_data *data)
} }
#define iosapic_enable_edge_irq unmask_irq #define iosapic_enable_edge_irq unmask_irq
#define iosapic_disable_edge_irq nop #define iosapic_disable_edge_irq iosapic_nop
static struct irq_chip irq_type_iosapic_edge = { static struct irq_chip irq_type_iosapic_edge = {
.name = "IO-SAPIC-edge", .name = "IO-SAPIC-edge",

55
arch/metag/include/asm/barrier.h
View File

@ -44,16 +44,6 @@ static inline void wr_fence(void)
#define rmb() barrier() #define rmb() barrier()
#define wmb() mb() #define wmb() mb()
#define dma_rmb() rmb()
#define dma_wmb() wmb()
#ifndef CONFIG_SMP
#define fence() do { } while (0)
#define smp_mb() barrier()
#define smp_rmb() barrier()
#define smp_wmb() barrier()
#else
#ifdef CONFIG_METAG_SMP_WRITE_REORDERING #ifdef CONFIG_METAG_SMP_WRITE_REORDERING
/* /*
* Write to the atomic memory unlock system event register (command 0). This is * Write to the atomic memory unlock system event register (command 0). This is
@ -63,45 +53,32 @@ static inline void wr_fence(void)
* incoherence). It is therefore ineffective if used after and on the same * incoherence). It is therefore ineffective if used after and on the same
* thread as a write. * thread as a write.
*/ */
static inline void fence(void) static inline void metag_fence(void)
{ {
volatile int *flushptr = (volatile int *) LINSYSEVENT_WR_ATOMIC_UNLOCK; volatile int *flushptr = (volatile int *) LINSYSEVENT_WR_ATOMIC_UNLOCK;
barrier(); barrier();
*flushptr = 0; *flushptr = 0;
barrier(); barrier();
} }
#define smp_mb() fence() #define __smp_mb() metag_fence()
#define smp_rmb() fence() #define __smp_rmb() metag_fence()
#define smp_wmb() barrier() #define __smp_wmb() barrier()
#else
#define metag_fence() do { } while (0)
#define __smp_mb() barrier()
#define __smp_rmb() barrier()
#define __smp_wmb() barrier()
#endif
#ifdef CONFIG_SMP
#define fence() metag_fence()
#else #else
#define fence() do { } while (0) #define fence() do { } while (0)
#define smp_mb() barrier()
#define smp_rmb() barrier()
#define smp_wmb() barrier()
#endif
#endif #endif
#define read_barrier_depends() do { } while (0) #define __smp_mb__before_atomic() barrier()
#define smp_read_barrier_depends() do { } while (0) #define __smp_mb__after_atomic() barrier()
#define smp_store_mb(var, value) do { WRITE_ONCE(var, value); smp_mb(); } while (0) #include <asm-generic/barrier.h>
#define smp_store_release(p, v) \
do { \
compiletime_assert_atomic_type(*p); \
smp_mb(); \
WRITE_ONCE(*p, v); \
} while (0)
#define smp_load_acquire(p) \
({ \
typeof(*p) ___p1 = READ_ONCE(*p); \
compiletime_assert_atomic_type(*p); \
smp_mb(); \
___p1; \
})
#define smp_mb__before_atomic() barrier()
#define smp_mb__after_atomic() barrier()
#endif /* _ASM_METAG_BARRIER_H */ #endif /* _ASM_METAG_BARRIER_H */

49
arch/mips/include/asm/barrier.h
View File

@ -10,9 +10,6 @@
#include <asm/addrspace.h> #include <asm/addrspace.h>
#define read_barrier_depends() do { } while(0)
#define smp_read_barrier_depends() do { } while(0)
#ifdef CONFIG_CPU_HAS_SYNC #ifdef CONFIG_CPU_HAS_SYNC
#define __sync() \ #define __sync() \
__asm__ __volatile__( \ __asm__ __volatile__( \
@ -87,23 +84,21 @@
#define wmb() fast_wmb() #define wmb() fast_wmb()
#define rmb() fast_rmb() #define rmb() fast_rmb()
#define dma_wmb() fast_wmb()
#define dma_rmb() fast_rmb()
#if defined(CONFIG_WEAK_ORDERING) && defined(CONFIG_SMP) #if defined(CONFIG_WEAK_ORDERING)
# ifdef CONFIG_CPU_CAVIUM_OCTEON # ifdef CONFIG_CPU_CAVIUM_OCTEON
# define smp_mb() __sync() # define __smp_mb() __sync()
# define smp_rmb() barrier() # define __smp_rmb() barrier()
# define smp_wmb() __syncw() # define __smp_wmb() __syncw()
# else # else
# define smp_mb() __asm__ __volatile__("sync" : : :"memory") # define __smp_mb() __asm__ __volatile__("sync" : : :"memory")
# define smp_rmb() __asm__ __volatile__("sync" : : :"memory") # define __smp_rmb() __asm__ __volatile__("sync" : : :"memory")
# define smp_wmb() __asm__ __volatile__("sync" : : :"memory") # define __smp_wmb() __asm__ __volatile__("sync" : : :"memory")
# endif # endif
#else #else
#define smp_mb() barrier() #define __smp_mb() barrier()
#define smp_rmb() barrier() #define __smp_rmb() barrier()
#define smp_wmb() barrier() #define __smp_wmb() barrier()
#endif #endif
#if defined(CONFIG_WEAK_REORDERING_BEYOND_LLSC) && defined(CONFIG_SMP) #if defined(CONFIG_WEAK_REORDERING_BEYOND_LLSC) && defined(CONFIG_SMP)
@ -112,13 +107,11 @@
#define __WEAK_LLSC_MB " \n" #define __WEAK_LLSC_MB " \n"
#endif #endif
#define smp_store_mb(var, value) \
do { WRITE_ONCE(var, value); smp_mb(); } while (0)
#define smp_llsc_mb() __asm__ __volatile__(__WEAK_LLSC_MB : : :"memory") #define smp_llsc_mb() __asm__ __volatile__(__WEAK_LLSC_MB : : :"memory")
#ifdef CONFIG_CPU_CAVIUM_OCTEON #ifdef CONFIG_CPU_CAVIUM_OCTEON
#define smp_mb__before_llsc() smp_wmb() #define smp_mb__before_llsc() smp_wmb()
#define __smp_mb__before_llsc() __smp_wmb()
/* Cause previous writes to become visible on all CPUs as soon as possible */ /* Cause previous writes to become visible on all CPUs as soon as possible */
#define nudge_writes() __asm__ __volatile__(".set push\n\t" \ #define nudge_writes() __asm__ __volatile__(".set push\n\t" \
".set arch=octeon\n\t" \ ".set arch=octeon\n\t" \
@ -126,25 +119,13 @@
".set pop" : : : "memory") ".set pop" : : : "memory")
#else #else
#define smp_mb__before_llsc() smp_llsc_mb() #define smp_mb__before_llsc() smp_llsc_mb()
#define __smp_mb__before_llsc() smp_llsc_mb()
#define nudge_writes() mb() #define nudge_writes() mb()
#endif #endif
#define smp_store_release(p, v) \ #define __smp_mb__before_atomic() __smp_mb__before_llsc()
do { \ #define __smp_mb__after_atomic() smp_llsc_mb()
compiletime_assert_atomic_type(*p); \
smp_mb(); \
WRITE_ONCE(*p, v); \
} while (0)
#define smp_load_acquire(p) \ #include <asm-generic/barrier.h>
({ \
typeof(*p) ___p1 = READ_ONCE(*p); \
compiletime_assert_atomic_type(*p); \
smp_mb(); \
___p1; \
})
#define smp_mb__before_atomic() smp_mb__before_llsc()
#define smp_mb__after_atomic() smp_llsc_mb()
#endif /* __ASM_BARRIER_H */ #endif /* __ASM_BARRIER_H */

33
arch/powerpc/include/asm/barrier.h
View File

@ -34,8 +34,6 @@
#define rmb() __asm__ __volatile__ ("sync" : : : "memory") #define rmb() __asm__ __volatile__ ("sync" : : : "memory")
#define wmb() __asm__ __volatile__ ("sync" : : : "memory") #define wmb() __asm__ __volatile__ ("sync" : : : "memory")
#define smp_store_mb(var, value) do { WRITE_ONCE(var, value); smp_mb(); } while (0)
#ifdef __SUBARCH_HAS_LWSYNC #ifdef __SUBARCH_HAS_LWSYNC
# define SMPWMB LWSYNC # define SMPWMB LWSYNC
#else #else
@ -46,22 +44,11 @@
#define dma_rmb() __lwsync() #define dma_rmb() __lwsync()
#define dma_wmb() __asm__ __volatile__ (stringify_in_c(SMPWMB) : : :"memory") #define dma_wmb() __asm__ __volatile__ (stringify_in_c(SMPWMB) : : :"memory")
#ifdef CONFIG_SMP #define __smp_lwsync() __lwsync()
#define smp_lwsync() __lwsync()
#define smp_mb() mb() #define __smp_mb() mb()
#define smp_rmb() __lwsync() #define __smp_rmb() __lwsync()
#define smp_wmb() __asm__ __volatile__ (stringify_in_c(SMPWMB) : : :"memory") #define __smp_wmb() __asm__ __volatile__ (stringify_in_c(SMPWMB) : : :"memory")
#else
#define smp_lwsync() barrier()
#define smp_mb() barrier()
#define smp_rmb() barrier()
#define smp_wmb() barrier()
#endif /* CONFIG_SMP */
#define read_barrier_depends() do { } while (0)
#define smp_read_barrier_depends() do { } while (0)
/* /*
* This is a barrier which prevents following instructions from being * This is a barrier which prevents following instructions from being
@ -72,23 +59,23 @@
#define data_barrier(x) \ #define data_barrier(x) \
asm volatile("twi 0,%0,0; isync" : : "r" (x) : "memory"); asm volatile("twi 0,%0,0; isync" : : "r" (x) : "memory");
#define smp_store_release(p, v) \ #define __smp_store_release(p, v) \
do { \ do { \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
smp_lwsync(); \ __smp_lwsync(); \
WRITE_ONCE(*p, v); \ WRITE_ONCE(*p, v); \
} while (0) } while (0)
#define smp_load_acquire(p) \ #define __smp_load_acquire(p) \
({ \ ({ \
typeof(*p) ___p1 = READ_ONCE(*p); \ typeof(*p) ___p1 = READ_ONCE(*p); \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
smp_lwsync(); \ __smp_lwsync(); \
___p1; \ ___p1; \
}) })
#define smp_mb__before_atomic() smp_mb()
#define smp_mb__after_atomic() smp_mb()
#define smp_mb__before_spinlock() smp_mb() #define smp_mb__before_spinlock() smp_mb()
#include <asm-generic/barrier.h>
#endif /* _ASM_POWERPC_BARRIER_H */ #endif /* _ASM_POWERPC_BARRIER_H */

23
arch/s390/include/asm/barrier.h
View File

@ -26,26 +26,18 @@
#define wmb() barrier() #define wmb() barrier()
#define dma_rmb() mb() #define dma_rmb() mb()
#define dma_wmb() mb() #define dma_wmb() mb()
#define smp_mb() mb() #define __smp_mb() mb()
#define smp_rmb() rmb() #define __smp_rmb() rmb()
#define smp_wmb() wmb() #define __smp_wmb() wmb()
#define read_barrier_depends() do { } while (0) #define __smp_store_release(p, v) \
#define smp_read_barrier_depends() do { } while (0)
#define smp_mb__before_atomic() smp_mb()
#define smp_mb__after_atomic() smp_mb()
#define smp_store_mb(var, value) do { WRITE_ONCE(var, value); smp_mb(); } while (0)
#define smp_store_release(p, v) \
do { \ do { \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
barrier(); \ barrier(); \
WRITE_ONCE(*p, v); \ WRITE_ONCE(*p, v); \
} while (0) } while (0)
#define smp_load_acquire(p) \ #define __smp_load_acquire(p) \
({ \ ({ \
typeof(*p) ___p1 = READ_ONCE(*p); \ typeof(*p) ___p1 = READ_ONCE(*p); \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
@ -53,4 +45,9 @@ do { \
___p1; \ ___p1; \
}) })
#define __smp_mb__before_atomic() barrier()
#define __smp_mb__after_atomic() barrier()
#include <asm-generic/barrier.h>
#endif /* __ASM_BARRIER_H */ #endif /* __ASM_BARRIER_H */

3
arch/sh/include/asm/barrier.h
View File

@ -32,7 +32,8 @@
#define ctrl_barrier() __asm__ __volatile__ ("nop;nop;nop;nop;nop;nop;nop;nop") #define ctrl_barrier() __asm__ __volatile__ ("nop;nop;nop;nop;nop;nop;nop;nop")
#endif #endif
#define smp_store_mb(var, value) do { (void)xchg(&var, value); } while (0) #define __smp_store_mb(var, value) do { (void)xchg(&var, value); } while (0)
#define smp_store_mb(var, value) __smp_store_mb(var, value)
#include <asm-generic/barrier.h> #include <asm-generic/barrier.h>

22
arch/sh/include/asm/cmpxchg-grb.h
View File

@ -23,6 +23,28 @@ static inline unsigned long xchg_u32(volatile u32 *m, unsigned long val)
return retval; return retval;
} }
static inline unsigned long xchg_u16(volatile u16 *m, unsigned long val)
{
unsigned long retval;
__asm__ __volatile__ (
" .align 2 \n\t"
" mova 1f, r0 \n\t" /* r0 = end point */
" mov r15, r1 \n\t" /* r1 = saved sp */
" mov #-6, r15 \n\t" /* LOGIN */
" mov.w @%1, %0 \n\t" /* load old value */
" extu.w %0, %0 \n\t" /* extend as unsigned */
" mov.w %2, @%1 \n\t" /* store new value */
"1: mov r1, r15 \n\t" /* LOGOUT */
: "=&r" (retval),
"+r" (m),
"+r" (val) /* inhibit r15 overloading */
:
: "memory" , "r0", "r1");
return retval;
}
static inline unsigned long xchg_u8(volatile u8 *m, unsigned long val) static inline unsigned long xchg_u8(volatile u8 *m, unsigned long val)
{ {
unsigned long retval; unsigned long retval;

11
arch/sh/include/asm/cmpxchg-irq.h
View File

@ -14,6 +14,17 @@ static inline unsigned long xchg_u32(volatile u32 *m, unsigned long val)
return retval; return retval;
} }
static inline unsigned long xchg_u16(volatile u16 *m, unsigned long val)
{
unsigned long flags, retval;
local_irq_save(flags);
retval = *m;
*m = val;
local_irq_restore(flags);
return retval;
}
static inline unsigned long xchg_u8(volatile u8 *m, unsigned long val) static inline unsigned long xchg_u8(volatile u8 *m, unsigned long val)
{ {
unsigned long flags, retval; unsigned long flags, retval;

25
arch/sh/include/asm/cmpxchg-llsc.h
View File

@ -22,29 +22,8 @@ static inline unsigned long xchg_u32(volatile u32 *m, unsigned long val)
return retval; return retval;
} }
static inline unsigned long xchg_u8(volatile u8 *m, unsigned long val)
{
unsigned long retval;
unsigned long tmp;
__asm__ __volatile__ (
"1: \n\t"
"movli.l @%2, %0 ! xchg_u8 \n\t"
"mov %0, %1 \n\t"
"mov %3, %0 \n\t"
"movco.l %0, @%2 \n\t"
"bf 1b \n\t"
"synco \n\t"
: "=&z"(tmp), "=&r" (retval)
: "r" (m), "r" (val & 0xff)
: "t", "memory"
);
return retval;
}
static inline unsigned long static inline unsigned long
__cmpxchg_u32(volatile int *m, unsigned long old, unsigned long new) __cmpxchg_u32(volatile u32 *m, unsigned long old, unsigned long new)
{ {
unsigned long retval; unsigned long retval;
unsigned long tmp; unsigned long tmp;
@ -68,4 +47,6 @@ __cmpxchg_u32(volatile int *m, unsigned long old, unsigned long new)
return retval; return retval;
} }
#include <asm/cmpxchg-xchg.h>
#endif /* __ASM_SH_CMPXCHG_LLSC_H */ #endif /* __ASM_SH_CMPXCHG_LLSC_H */

51
arch/sh/include/asm/cmpxchg-xchg.h Normal file
View File

@ -0,0 +1,51 @@
#ifndef __ASM_SH_CMPXCHG_XCHG_H
#define __ASM_SH_CMPXCHG_XCHG_H
/*
* Copyright (C) 2016 Red Hat, Inc.
* Author: Michael S. Tsirkin <mst@redhat.com>
*
* This work is licensed under the terms of the GNU GPL, version 2. See the
* file "COPYING" in the main directory of this archive for more details.
*/
#include <linux/bitops.h>
#include <asm/byteorder.h>
/*
* Portable implementations of 1 and 2 byte xchg using a 4 byte cmpxchg.
* Note: this header isn't self-contained: before including it, __cmpxchg_u32
* must be defined first.
*/
static inline u32 __xchg_cmpxchg(volatile void *ptr, u32 x, int size)
{
int off = (unsigned long)ptr % sizeof(u32);
volatile u32 *p = ptr - off;
#ifdef __BIG_ENDIAN
int bitoff = (sizeof(u32) - 1 - off) * BITS_PER_BYTE;
#else
int bitoff = off * BITS_PER_BYTE;
#endif
u32 bitmask = ((0x1 << size * BITS_PER_BYTE) - 1) << bitoff;
u32 oldv, newv;
u32 ret;
do {
oldv = READ_ONCE(*p);
ret = (oldv & bitmask) >> bitoff;
newv = (oldv & ~bitmask) | (x << bitoff);
} while (__cmpxchg_u32(p, oldv, newv) != oldv);
return ret;
}
static inline unsigned long xchg_u16(volatile u16 *m, unsigned long val)
{
return __xchg_cmpxchg(m, val, sizeof *m);
}
static inline unsigned long xchg_u8(volatile u8 *m, unsigned long val)
{
return __xchg_cmpxchg(m, val, sizeof *m);
}
#endif /* __ASM_SH_CMPXCHG_XCHG_H */

3
arch/sh/include/asm/cmpxchg.h
View File

@ -27,6 +27,9 @@ extern void __xchg_called_with_bad_pointer(void);
case 4: \ case 4: \
__xchg__res = xchg_u32(__xchg_ptr, x); \ __xchg__res = xchg_u32(__xchg_ptr, x); \
break; \ break; \
case 2: \
__xchg__res = xchg_u16(__xchg_ptr, x); \
break; \
case 1: \ case 1: \
__xchg__res = xchg_u8(__xchg_ptr, x); \ __xchg__res = xchg_u8(__xchg_ptr, x); \
break; \ break; \

1
arch/sparc/include/asm/barrier_32.h
View File

@ -1,7 +1,6 @@
#ifndef __SPARC_BARRIER_H #ifndef __SPARC_BARRIER_H
#define __SPARC_BARRIER_H #define __SPARC_BARRIER_H
#include <asm/processor.h> /* for nop() */
#include <asm-generic/barrier.h> #include <asm-generic/barrier.h>
#endif /* !(__SPARC_BARRIER_H) */ #endif /* !(__SPARC_BARRIER_H) */

29
arch/sparc/include/asm/barrier_64.h
View File

@ -37,33 +37,14 @@ do { __asm__ __volatile__("ba,pt %%xcc, 1f\n\t" \
#define rmb() __asm__ __volatile__("":::"memory") #define rmb() __asm__ __volatile__("":::"memory")
#define wmb() __asm__ __volatile__("":::"memory") #define wmb() __asm__ __volatile__("":::"memory")
#define dma_rmb() rmb() #define __smp_store_release(p, v) \
#define dma_wmb() wmb()
#define smp_store_mb(__var, __value) \
do { WRITE_ONCE(__var, __value); membar_safe("#StoreLoad"); } while(0)
#ifdef CONFIG_SMP
#define smp_mb() mb()
#define smp_rmb() rmb()
#define smp_wmb() wmb()
#else
#define smp_mb() __asm__ __volatile__("":::"memory")
#define smp_rmb() __asm__ __volatile__("":::"memory")
#define smp_wmb() __asm__ __volatile__("":::"memory")
#endif
#define read_barrier_depends() do { } while (0)
#define smp_read_barrier_depends() do { } while (0)
#define smp_store_release(p, v) \
do { \ do { \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
barrier(); \ barrier(); \
WRITE_ONCE(*p, v); \ WRITE_ONCE(*p, v); \
} while (0) } while (0)
#define smp_load_acquire(p) \ #define __smp_load_acquire(p) \
({ \ ({ \
typeof(*p) ___p1 = READ_ONCE(*p); \ typeof(*p) ___p1 = READ_ONCE(*p); \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
@ -71,7 +52,9 @@ do { \
___p1; \ ___p1; \
}) })
#define smp_mb__before_atomic() barrier() #define __smp_mb__before_atomic() barrier()
#define smp_mb__after_atomic() barrier() #define __smp_mb__after_atomic() barrier()
#include <asm-generic/barrier.h>
#endif /* !(__SPARC64_BARRIER_H) */ #endif /* !(__SPARC64_BARRIER_H) */

3
arch/sparc/include/asm/processor.h
View File

@ -5,7 +5,4 @@
#else #else
#include <asm/processor_32.h> #include <asm/processor_32.h>
#endif #endif
#define nop() __asm__ __volatile__ ("nop")
#endif #endif

9
arch/tile/include/asm/barrier.h
View File

@ -79,11 +79,12 @@ mb_incoherent(void)
* But after the word is updated, the routine issues an "mf" before returning, * But after the word is updated, the routine issues an "mf" before returning,
* and since it's a function call, we don't even need a compiler barrier. * and since it's a function call, we don't even need a compiler barrier.
*/ */
#define smp_mb__before_atomic() smp_mb() #define __smp_mb__before_atomic() __smp_mb()
#define smp_mb__after_atomic() do { } while (0) #define __smp_mb__after_atomic() do { } while (0)
#define smp_mb__after_atomic() __smp_mb__after_atomic()
#else /* 64 bit */ #else /* 64 bit */
#define smp_mb__before_atomic() smp_mb() #define __smp_mb__before_atomic() __smp_mb()
#define smp_mb__after_atomic() smp_mb() #define __smp_mb__after_atomic() __smp_mb()
#endif #endif
#include <asm-generic/barrier.h> #include <asm-generic/barrier.h>

36
arch/x86/include/asm/barrier.h
View File

@ -31,20 +31,10 @@
#endif #endif
#define dma_wmb() barrier() #define dma_wmb() barrier()
#ifdef CONFIG_SMP #define __smp_mb() mb()
#define smp_mb() mb() #define __smp_rmb() dma_rmb()
#define smp_rmb() dma_rmb() #define __smp_wmb() barrier()
#define smp_wmb() barrier() #define __smp_store_mb(var, value) do { (void)xchg(&var, value); } while (0)
#define smp_store_mb(var, value) do { (void)xchg(&var, value); } while (0)
#else /* !SMP */
#define smp_mb() barrier()
#define smp_rmb() barrier()
#define smp_wmb() barrier()
#define smp_store_mb(var, value) do { WRITE_ONCE(var, value); barrier(); } while (0)
#endif /* SMP */
#define read_barrier_depends() do { } while (0)
#define smp_read_barrier_depends() do { } while (0)
#if defined(CONFIG_X86_PPRO_FENCE) #if defined(CONFIG_X86_PPRO_FENCE)
@ -53,31 +43,31 @@
* model and we should fall back to full barriers. * model and we should fall back to full barriers.
*/ */
#define smp_store_release(p, v) \ #define __smp_store_release(p, v) \
do { \ do { \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
smp_mb(); \ __smp_mb(); \
WRITE_ONCE(*p, v); \ WRITE_ONCE(*p, v); \
} while (0) } while (0)
#define smp_load_acquire(p) \ #define __smp_load_acquire(p) \
({ \ ({ \
typeof(*p) ___p1 = READ_ONCE(*p); \ typeof(*p) ___p1 = READ_ONCE(*p); \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
smp_mb(); \ __smp_mb(); \
___p1; \ ___p1; \
}) })
#else /* regular x86 TSO memory ordering */ #else /* regular x86 TSO memory ordering */
#define smp_store_release(p, v) \ #define __smp_store_release(p, v) \
do { \ do { \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
barrier(); \ barrier(); \
WRITE_ONCE(*p, v); \ WRITE_ONCE(*p, v); \
} while (0) } while (0)
#define smp_load_acquire(p) \ #define __smp_load_acquire(p) \
({ \ ({ \
typeof(*p) ___p1 = READ_ONCE(*p); \ typeof(*p) ___p1 = READ_ONCE(*p); \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
@ -88,7 +78,9 @@ do { \
#endif #endif
/* Atomic operations are already serializing on x86 */ /* Atomic operations are already serializing on x86 */
#define smp_mb__before_atomic() barrier() #define __smp_mb__before_atomic() barrier()
#define smp_mb__after_atomic() barrier() #define __smp_mb__after_atomic() barrier()
#include <asm-generic/barrier.h>
#endif /* _ASM_X86_BARRIER_H */ #endif /* _ASM_X86_BARRIER_H */

9
arch/x86/um/asm/barrier.h
View File

@ -36,13 +36,6 @@
#endif /* CONFIG_X86_PPRO_FENCE */ #endif /* CONFIG_X86_PPRO_FENCE */
#define dma_wmb() barrier() #define dma_wmb() barrier()
#define smp_mb() barrier() #include <asm-generic/barrier.h>
#define smp_rmb() barrier()
#define smp_wmb() barrier()
#define smp_store_mb(var, value) do { WRITE_ONCE(var, value); barrier(); } while (0)
#define read_barrier_depends() do { } while (0)
#define smp_read_barrier_depends() do { } while (0)
#endif #endif

4
arch/xtensa/include/asm/barrier.h
View File

@ -13,8 +13,8 @@
#define rmb() barrier() #define rmb() barrier()
#define wmb() mb() #define wmb() mb()
#define smp_mb__before_atomic() barrier() #define __smp_mb__before_atomic() barrier()
#define smp_mb__after_atomic() barrier() #define __smp_mb__after_atomic() barrier()
#include <asm-generic/barrier.h> #include <asm-generic/barrier.h>

2
drivers/gpu/drm/virtio/virtgpu_kms.c
View File

@ -130,7 +130,7 @@ int virtio_gpu_driver_load(struct drm_device *dev, unsigned long flags)
static vq_callback_t *callbacks[] = { static vq_callback_t *callbacks[] = {
virtio_gpu_ctrl_ack, virtio_gpu_cursor_ack virtio_gpu_ctrl_ack, virtio_gpu_cursor_ack
}; };
static const char *names[] = { "control", "cursor" }; static const char * const names[] = { "control", "cursor" };
struct virtio_gpu_device *vgdev; struct virtio_gpu_device *vgdev;
/* this will expand later */ /* this will expand later */

2
drivers/misc/mic/card/mic_virtio.c
View File

@ -311,7 +311,7 @@ static struct virtqueue *mic_find_vq(struct virtio_device *vdev,
static int mic_find_vqs(struct virtio_device *vdev, unsigned nvqs, static int mic_find_vqs(struct virtio_device *vdev, unsigned nvqs,
struct virtqueue *vqs[], struct virtqueue *vqs[],
vq_callback_t *callbacks[], vq_callback_t *callbacks[],
const char *names[]) const char * const names[])
{ {
struct mic_vdev *mvdev = to_micvdev(vdev); struct mic_vdev *mvdev = to_micvdev(vdev);
struct mic_device_ctrl __iomem *dc = mvdev->dc; struct mic_device_ctrl __iomem *dc = mvdev->dc;

2
drivers/remoteproc/remoteproc_virtio.c
View File

@ -147,7 +147,7 @@ static void rproc_virtio_del_vqs(struct virtio_device *vdev)
static int rproc_virtio_find_vqs(struct virtio_device *vdev, unsigned nvqs, static int rproc_virtio_find_vqs(struct virtio_device *vdev, unsigned nvqs,
struct virtqueue *vqs[], struct virtqueue *vqs[],
vq_callback_t *callbacks[], vq_callback_t *callbacks[],
const char *names[]) const char * const names[])
{ {
struct rproc *rproc = vdev_to_rproc(vdev); struct rproc *rproc = vdev_to_rproc(vdev);
int i, ret; int i, ret;

2
drivers/rpmsg/virtio_rpmsg_bus.c
View File

@ -945,7 +945,7 @@ static void rpmsg_ns_cb(struct rpmsg_channel *rpdev, void *data, int len,
static int rpmsg_probe(struct virtio_device *vdev) static int rpmsg_probe(struct virtio_device *vdev)
{ {
vq_callback_t *vq_cbs[] = { rpmsg_recv_done, rpmsg_xmit_done }; vq_callback_t *vq_cbs[] = { rpmsg_recv_done, rpmsg_xmit_done };
const char *names[] = { "input", "output" }; static const char * const names[] = { "input", "output" };
struct virtqueue *vqs[2]; struct virtqueue *vqs[2];
struct virtproc_info *vrp; struct virtproc_info *vrp;
void *bufs_va; void *bufs_va;

2
drivers/s390/virtio/kvm_virtio.c
View File

@ -255,7 +255,7 @@ static void kvm_del_vqs(struct virtio_device *vdev)
static int kvm_find_vqs(struct virtio_device *vdev, unsigned nvqs, static int kvm_find_vqs(struct virtio_device *vdev, unsigned nvqs,
struct virtqueue *vqs[], struct virtqueue *vqs[],
vq_callback_t *callbacks[], vq_callback_t *callbacks[],
const char *names[]) const char * const names[])
{ {
struct kvm_device *kdev = to_kvmdev(vdev); struct kvm_device *kdev = to_kvmdev(vdev);
int i; int i;

2
drivers/s390/virtio/virtio_ccw.c
View File

@ -635,7 +635,7 @@ static int virtio_ccw_register_adapter_ind(struct virtio_ccw_device *vcdev,
static int virtio_ccw_find_vqs(struct virtio_device *vdev, unsigned nvqs, static int virtio_ccw_find_vqs(struct virtio_device *vdev, unsigned nvqs,
struct virtqueue *vqs[], struct virtqueue *vqs[],
vq_callback_t *callbacks[], vq_callback_t *callbacks[],
const char *names[]) const char * const names[])
{ {
struct virtio_ccw_device *vcdev = to_vc_device(vdev); struct virtio_ccw_device *vcdev = to_vc_device(vdev);
unsigned long *indicatorp = NULL; unsigned long *indicatorp = NULL;

4
drivers/virtio/virtio_balloon.c
View File

@ -209,8 +209,8 @@ static unsigned leak_balloon(struct virtio_balloon *vb, size_t num)
*/ */
if (vb->num_pfns != 0) if (vb->num_pfns != 0)
tell_host(vb, vb->deflate_vq); tell_host(vb, vb->deflate_vq);
mutex_unlock(&vb->balloon_lock);
release_pages_balloon(vb); release_pages_balloon(vb);
mutex_unlock(&vb->balloon_lock);
return num_freed_pages; return num_freed_pages;
} }
@ -388,7 +388,7 @@ static int init_vqs(struct virtio_balloon *vb)
{ {
struct virtqueue *vqs[3]; struct virtqueue *vqs[3];
vq_callback_t *callbacks[] = { balloon_ack, balloon_ack, stats_request }; vq_callback_t *callbacks[] = { balloon_ack, balloon_ack, stats_request };
const char *names[] = { "inflate", "deflate", "stats" }; static const char * const names[] = { "inflate", "deflate", "stats" };
int err, nvqs; int err, nvqs;
/* /*

2
drivers/virtio/virtio_input.c
View File

@ -170,7 +170,7 @@ static int virtinput_init_vqs(struct virtio_input *vi)
struct virtqueue *vqs[2]; struct virtqueue *vqs[2];
vq_callback_t *cbs[] = { virtinput_recv_events, vq_callback_t *cbs[] = { virtinput_recv_events,
virtinput_recv_status }; virtinput_recv_status };
static const char *names[] = { "events", "status" }; static const char * const names[] = { "events", "status" };
int err; int err;
err = vi->vdev->config->find_vqs(vi->vdev, 2, vqs, cbs, names); err = vi->vdev->config->find_vqs(vi->vdev, 2, vqs, cbs, names);

2
drivers/virtio/virtio_mmio.c
View File

@ -482,7 +482,7 @@ static struct virtqueue *vm_setup_vq(struct virtio_device *vdev, unsigned index,
static int vm_find_vqs(struct virtio_device *vdev, unsigned nvqs, static int vm_find_vqs(struct virtio_device *vdev, unsigned nvqs,
struct virtqueue *vqs[], struct virtqueue *vqs[],
vq_callback_t *callbacks[], vq_callback_t *callbacks[],
const char *names[]) const char * const names[])
{ {
struct virtio_mmio_device *vm_dev = to_virtio_mmio_device(vdev); struct virtio_mmio_device *vm_dev = to_virtio_mmio_device(vdev);
unsigned int irq = platform_get_irq(vm_dev->pdev, 0); unsigned int irq = platform_get_irq(vm_dev->pdev, 0);

4
drivers/virtio/virtio_pci_common.c
View File

@ -296,7 +296,7 @@ void vp_del_vqs(struct virtio_device *vdev)
static int vp_try_to_find_vqs(struct virtio_device *vdev, unsigned nvqs, static int vp_try_to_find_vqs(struct virtio_device *vdev, unsigned nvqs,
struct virtqueue *vqs[], struct virtqueue *vqs[],
vq_callback_t *callbacks[], vq_callback_t *callbacks[],
const char *names[], const char * const names[],
bool use_msix, bool use_msix,
bool per_vq_vectors) bool per_vq_vectors)
{ {
@ -376,7 +376,7 @@ static int vp_try_to_find_vqs(struct virtio_device *vdev, unsigned nvqs,
int vp_find_vqs(struct virtio_device *vdev, unsigned nvqs, int vp_find_vqs(struct virtio_device *vdev, unsigned nvqs,
struct virtqueue *vqs[], struct virtqueue *vqs[],
vq_callback_t *callbacks[], vq_callback_t *callbacks[],
const char *names[]) const char * const names[])
{ {
int err; int err;

2
drivers/virtio/virtio_pci_common.h
View File

@ -139,7 +139,7 @@ void vp_del_vqs(struct virtio_device *vdev);
int vp_find_vqs(struct virtio_device *vdev, unsigned nvqs, int vp_find_vqs(struct virtio_device *vdev, unsigned nvqs,
struct virtqueue *vqs[], struct virtqueue *vqs[],
vq_callback_t *callbacks[], vq_callback_t *callbacks[],
const char *names[]); const char * const names[]);
const char *vp_bus_name(struct virtio_device *vdev); const char *vp_bus_name(struct virtio_device *vdev);
/* Setup the affinity for a virtqueue: /* Setup the affinity for a virtqueue:

2
drivers/virtio/virtio_pci_modern.c
View File

@ -418,7 +418,7 @@ static struct virtqueue *setup_vq(struct virtio_pci_device *vp_dev,
static int vp_modern_find_vqs(struct virtio_device *vdev, unsigned nvqs, static int vp_modern_find_vqs(struct virtio_device *vdev, unsigned nvqs,
struct virtqueue *vqs[], struct virtqueue *vqs[],
vq_callback_t *callbacks[], vq_callback_t *callbacks[],
const char *names[]) const char * const names[])
{ {
struct virtio_pci_device *vp_dev = to_vp_device(vdev); struct virtio_pci_device *vp_dev = to_vp_device(vdev);
struct virtqueue *vq; struct virtqueue *vq;

15
drivers/virtio/virtio_ring.c
View File

@ -517,10 +517,10 @@ void *virtqueue_get_buf(struct virtqueue *_vq, unsigned int *len)
/* If we expect an interrupt for the next entry, tell host /* If we expect an interrupt for the next entry, tell host
* by writing event index and flush out the write before * by writing event index and flush out the write before
* the read in the next get_buf call. */ * the read in the next get_buf call. */
if (!(vq->avail_flags_shadow & VRING_AVAIL_F_NO_INTERRUPT)) { if (!(vq->avail_flags_shadow & VRING_AVAIL_F_NO_INTERRUPT))
vring_used_event(&vq->vring) = cpu_to_virtio16(_vq->vdev, vq->last_used_idx); virtio_store_mb(vq->weak_barriers,
virtio_mb(vq->weak_barriers); &vring_used_event(&vq->vring),
} cpu_to_virtio16(_vq->vdev, vq->last_used_idx));
#ifdef DEBUG #ifdef DEBUG
vq->last_add_time_valid = false; vq->last_add_time_valid = false;
@ -653,8 +653,11 @@ bool virtqueue_enable_cb_delayed(struct virtqueue *_vq)
} }
/* TODO: tune this threshold */ /* TODO: tune this threshold */
bufs = (u16)(vq->avail_idx_shadow - vq->last_used_idx) * 3 / 4; bufs = (u16)(vq->avail_idx_shadow - vq->last_used_idx) * 3 / 4;
vring_used_event(&vq->vring) = cpu_to_virtio16(_vq->vdev, vq->last_used_idx + bufs);
virtio_mb(vq->weak_barriers); virtio_store_mb(vq->weak_barriers,
&vring_used_event(&vq->vring),
cpu_to_virtio16(_vq->vdev, vq->last_used_idx + bufs));
if (unlikely((u16)(virtio16_to_cpu(_vq->vdev, vq->vring.used->idx) - vq->last_used_idx) > bufs)) { if (unlikely((u16)(virtio16_to_cpu(_vq->vdev, vq->vring.used->idx) - vq->last_used_idx) > bufs)) {
END_USE(vq); END_USE(vq);
return false; return false;

3
drivers/xen/events/events_fifo.c
View File

@ -41,6 +41,7 @@
#include <linux/percpu.h> #include <linux/percpu.h>
#include <linux/cpu.h> #include <linux/cpu.h>
#include <asm/barrier.h>
#include <asm/sync_bitops.h> #include <asm/sync_bitops.h>
#include <asm/xen/hypercall.h> #include <asm/xen/hypercall.h>
#include <asm/xen/hypervisor.h> #include <asm/xen/hypervisor.h>
@ -296,7 +297,7 @@ static void consume_one_event(unsigned cpu,
* control block. * control block.
*/ */
if (head == 0) { if (head == 0) {
rmb(); /* Ensure word is up-to-date before reading head. */ virt_rmb(); /* Ensure word is up-to-date before reading head. */
head = control_block->head[priority]; head = control_block->head[priority];
} }

8
drivers/xen/xenbus/xenbus_comms.c
View File

@ -123,14 +123,14 @@ int xb_write(const void *data, unsigned len)
avail = len; avail = len;
/* Must write data /after/ reading the consumer index. */ /* Must write data /after/ reading the consumer index. */
mb(); virt_mb();
memcpy(dst, data, avail); memcpy(dst, data, avail);
data += avail; data += avail;
len -= avail; len -= avail;
/* Other side must not see new producer until data is there. */ /* Other side must not see new producer until data is there. */
wmb(); virt_wmb();
intf->req_prod += avail; intf->req_prod += avail;
/* Implies mb(): other side will see the updated producer. */ /* Implies mb(): other side will see the updated producer. */
@ -180,14 +180,14 @@ int xb_read(void *data, unsigned len)
avail = len; avail = len;
/* Must read data /after/ reading the producer index. */ /* Must read data /after/ reading the producer index. */
rmb(); virt_rmb();
memcpy(data, src, avail); memcpy(data, src, avail);
data += avail; data += avail;
len -= avail; len -= avail;
/* Other side must not see free space until we've copied out */ /* Other side must not see free space until we've copied out */
mb(); virt_mb();
intf->rsp_cons += avail; intf->rsp_cons += avail;
pr_debug("Finished read of %i bytes (%i to go)\n", avail, len); pr_debug("Finished read of %i bytes (%i to go)\n", avail, len);

106
include/asm-generic/barrier.h
View File

@ -54,22 +54,38 @@
#define read_barrier_depends() do { } while (0) #define read_barrier_depends() do { } while (0)
#endif #endif
#ifndef __smp_mb
#define __smp_mb() mb()
#endif
#ifndef __smp_rmb
#define __smp_rmb() rmb()
#endif
#ifndef __smp_wmb
#define __smp_wmb() wmb()
#endif
#ifndef __smp_read_barrier_depends
#define __smp_read_barrier_depends() read_barrier_depends()
#endif
#ifdef CONFIG_SMP #ifdef CONFIG_SMP
#ifndef smp_mb #ifndef smp_mb
#define smp_mb() mb() #define smp_mb() __smp_mb()
#endif #endif
#ifndef smp_rmb #ifndef smp_rmb
#define smp_rmb() rmb() #define smp_rmb() __smp_rmb()
#endif #endif
#ifndef smp_wmb #ifndef smp_wmb
#define smp_wmb() wmb() #define smp_wmb() __smp_wmb()
#endif #endif
#ifndef smp_read_barrier_depends #ifndef smp_read_barrier_depends
#define smp_read_barrier_depends() read_barrier_depends() #define smp_read_barrier_depends() __smp_read_barrier_depends()
#endif #endif
#else /* !CONFIG_SMP */ #else /* !CONFIG_SMP */
@ -92,32 +108,104 @@
#endif /* CONFIG_SMP */ #endif /* CONFIG_SMP */
#ifndef __smp_store_mb
#define __smp_store_mb(var, value) do { WRITE_ONCE(var, value); __smp_mb(); } while (0)
#endif
#ifndef __smp_mb__before_atomic
#define __smp_mb__before_atomic() __smp_mb()
#endif
#ifndef __smp_mb__after_atomic
#define __smp_mb__after_atomic() __smp_mb()
#endif
#ifndef __smp_store_release
#define __smp_store_release(p, v) \
do { \
compiletime_assert_atomic_type(*p); \
__smp_mb(); \
WRITE_ONCE(*p, v); \
} while (0)
#endif
#ifndef __smp_load_acquire
#define __smp_load_acquire(p) \
({ \
typeof(*p) ___p1 = READ_ONCE(*p); \
compiletime_assert_atomic_type(*p); \
__smp_mb(); \
___p1; \
})
#endif
#ifdef CONFIG_SMP
#ifndef smp_store_mb #ifndef smp_store_mb
#define smp_store_mb(var, value) do { WRITE_ONCE(var, value); smp_mb(); } while (0) #define smp_store_mb(var, value) __smp_store_mb(var, value)
#endif #endif
#ifndef smp_mb__before_atomic #ifndef smp_mb__before_atomic
#define smp_mb__before_atomic() smp_mb() #define smp_mb__before_atomic() __smp_mb__before_atomic()
#endif #endif
#ifndef smp_mb__after_atomic #ifndef smp_mb__after_atomic
#define smp_mb__after_atomic() smp_mb() #define smp_mb__after_atomic() __smp_mb__after_atomic()
#endif #endif
#ifndef smp_store_release
#define smp_store_release(p, v) __smp_store_release(p, v)
#endif
#ifndef smp_load_acquire
#define smp_load_acquire(p) __smp_load_acquire(p)
#endif
#else /* !CONFIG_SMP */
#ifndef smp_store_mb
#define smp_store_mb(var, value) do { WRITE_ONCE(var, value); barrier(); } while (0)
#endif
#ifndef smp_mb__before_atomic
#define smp_mb__before_atomic() barrier()
#endif
#ifndef smp_mb__after_atomic
#define smp_mb__after_atomic() barrier()
#endif
#ifndef smp_store_release
#define smp_store_release(p, v) \ #define smp_store_release(p, v) \
do { \ do { \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
smp_mb(); \ barrier(); \
WRITE_ONCE(*p, v); \ WRITE_ONCE(*p, v); \
} while (0) } while (0)
#endif
#ifndef smp_load_acquire
#define smp_load_acquire(p) \ #define smp_load_acquire(p) \
({ \ ({ \
typeof(*p) ___p1 = READ_ONCE(*p); \ typeof(*p) ___p1 = READ_ONCE(*p); \
compiletime_assert_atomic_type(*p); \ compiletime_assert_atomic_type(*p); \
smp_mb(); \ barrier(); \
___p1; \ ___p1; \
}) })
#endif
#endif
/* Barriers for virtual machine guests when talking to an SMP host */
#define virt_mb() __smp_mb()
#define virt_rmb() __smp_rmb()
#define virt_wmb() __smp_wmb()
#define virt_read_barrier_depends() __smp_read_barrier_depends()
#define virt_store_mb(var, value) __smp_store_mb(var, value)
#define virt_mb__before_atomic() __smp_mb__before_atomic()
#define virt_mb__after_atomic() __smp_mb__after_atomic()
#define virt_store_release(p, v) __smp_store_release(p, v)
#define virt_load_acquire(p) __smp_load_acquire(p)
#endif /* !__ASSEMBLY__ */ #endif /* !__ASSEMBLY__ */
#endif /* __ASM_GENERIC_BARRIER_H */ #endif /* __ASM_GENERIC_BARRIER_H */

2
include/linux/virtio_config.h
View File

@ -70,7 +70,7 @@ struct virtio_config_ops {
int (*find_vqs)(struct virtio_device *, unsigned nvqs, int (*find_vqs)(struct virtio_device *, unsigned nvqs,
struct virtqueue *vqs[], struct virtqueue *vqs[],
vq_callback_t *callbacks[], vq_callback_t *callbacks[],
const char *names[]); const char * const names[]);
void (*del_vqs)(struct virtio_device *); void (*del_vqs)(struct virtio_device *);
u64 (*get_features)(struct virtio_device *vdev); u64 (*get_features)(struct virtio_device *vdev);
int (*finalize_features)(struct virtio_device *vdev); int (*finalize_features)(struct virtio_device *vdev);

21
include/linux/virtio_ring.h
View File

@ -12,7 +12,7 @@
* anyone care? * anyone care?
* *
* For virtio_pci on SMP, we don't need to order with respect to MMIO * For virtio_pci on SMP, we don't need to order with respect to MMIO
* accesses through relaxed memory I/O windows, so smp_mb() et al are * accesses through relaxed memory I/O windows, so virt_mb() et al are
* sufficient. * sufficient.
* *
* For using virtio to talk to real devices (eg. other heterogeneous * For using virtio to talk to real devices (eg. other heterogeneous
@ -23,18 +23,16 @@
static inline void virtio_mb(bool weak_barriers) static inline void virtio_mb(bool weak_barriers)
{ {
#ifdef CONFIG_SMP
if (weak_barriers) if (weak_barriers)
smp_mb(); virt_mb();
else else
#endif
mb(); mb();
} }
static inline void virtio_rmb(bool weak_barriers) static inline void virtio_rmb(bool weak_barriers)
{ {
if (weak_barriers) if (weak_barriers)
dma_rmb(); virt_rmb();
else else
rmb(); rmb();
} }
@ -42,11 +40,22 @@ static inline void virtio_rmb(bool weak_barriers)
static inline void virtio_wmb(bool weak_barriers) static inline void virtio_wmb(bool weak_barriers)
{ {
if (weak_barriers) if (weak_barriers)
dma_wmb(); virt_wmb();
else else
wmb(); wmb();
} }
static inline void virtio_store_mb(bool weak_barriers,
__virtio16 *p, __virtio16 v)
{
if (weak_barriers) {
virt_store_mb(*p, v);
} else {
WRITE_ONCE(*p, v);
mb();
}
}
struct virtio_device; struct virtio_device;
struct virtqueue; struct virtqueue;

16
include/xen/interface/io/ring.h
View File

@ -208,12 +208,12 @@ struct __name##_back_ring { \
#define RING_PUSH_REQUESTS(_r) do { \ #define RING_PUSH_REQUESTS(_r) do { \
wmb(); /* back sees requests /before/ updated producer index */ \ virt_wmb(); /* back sees requests /before/ updated producer index */ \
(_r)->sring->req_prod = (_r)->req_prod_pvt; \ (_r)->sring->req_prod = (_r)->req_prod_pvt; \
} while (0) } while (0)
#define RING_PUSH_RESPONSES(_r) do { \ #define RING_PUSH_RESPONSES(_r) do { \
wmb(); /* front sees responses /before/ updated producer index */ \ virt_wmb(); /* front sees responses /before/ updated producer index */ \
(_r)->sring->rsp_prod = (_r)->rsp_prod_pvt; \ (_r)->sring->rsp_prod = (_r)->rsp_prod_pvt; \
} while (0) } while (0)
@ -250,9 +250,9 @@ struct __name##_back_ring { \
#define RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(_r, _notify) do { \ #define RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(_r, _notify) do { \
RING_IDX __old = (_r)->sring->req_prod; \ RING_IDX __old = (_r)->sring->req_prod; \
RING_IDX __new = (_r)->req_prod_pvt; \ RING_IDX __new = (_r)->req_prod_pvt; \
wmb(); /* back sees requests /before/ updated producer index */ \ virt_wmb(); /* back sees requests /before/ updated producer index */ \
(_r)->sring->req_prod = __new; \ (_r)->sring->req_prod = __new; \
mb(); /* back sees new requests /before/ we check req_event */ \ virt_mb(); /* back sees new requests /before/ we check req_event */ \
(_notify) = ((RING_IDX)(__new - (_r)->sring->req_event) < \ (_notify) = ((RING_IDX)(__new - (_r)->sring->req_event) < \
(RING_IDX)(__new - __old)); \ (RING_IDX)(__new - __old)); \
} while (0) } while (0)
@ -260,9 +260,9 @@ struct __name##_back_ring { \
#define RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(_r, _notify) do { \ #define RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(_r, _notify) do { \
RING_IDX __old = (_r)->sring->rsp_prod; \ RING_IDX __old = (_r)->sring->rsp_prod; \
RING_IDX __new = (_r)->rsp_prod_pvt; \ RING_IDX __new = (_r)->rsp_prod_pvt; \
wmb(); /* front sees responses /before/ updated producer index */ \ virt_wmb(); /* front sees responses /before/ updated producer index */ \
(_r)->sring->rsp_prod = __new; \ (_r)->sring->rsp_prod = __new; \
mb(); /* front sees new responses /before/ we check rsp_event */ \ virt_mb(); /* front sees new responses /before/ we check rsp_event */ \
(_notify) = ((RING_IDX)(__new - (_r)->sring->rsp_event) < \ (_notify) = ((RING_IDX)(__new - (_r)->sring->rsp_event) < \
(RING_IDX)(__new - __old)); \ (RING_IDX)(__new - __old)); \
} while (0) } while (0)
@ -271,7 +271,7 @@ struct __name##_back_ring { \
(_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \ (_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \
if (_work_to_do) break; \ if (_work_to_do) break; \
(_r)->sring->req_event = (_r)->req_cons + 1; \ (_r)->sring->req_event = (_r)->req_cons + 1; \
mb(); \ virt_mb(); \
(_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \ (_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \
} while (0) } while (0)
@ -279,7 +279,7 @@ struct __name##_back_ring { \
(_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \ (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \
if (_work_to_do) break; \ if (_work_to_do) break; \
(_r)->sring->rsp_event = (_r)->rsp_cons + 1; \ (_r)->sring->rsp_event = (_r)->rsp_cons + 1; \
mb(); \ virt_mb(); \
(_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \ (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \
} while (0) } while (0)

4
mm/balloon_compaction.c
View File

@ -61,6 +61,7 @@ struct page *balloon_page_dequeue(struct balloon_dev_info *b_dev_info)
bool dequeued_page; bool dequeued_page;
dequeued_page = false; dequeued_page = false;
spin_lock_irqsave(&b_dev_info->pages_lock, flags);
list_for_each_entry_safe(page, tmp, &b_dev_info->pages, lru) { list_for_each_entry_safe(page, tmp, &b_dev_info->pages, lru) {
/* /*
* Block others from accessing the 'page' while we get around * Block others from accessing the 'page' while we get around
@ -75,15 +76,14 @@ struct page *balloon_page_dequeue(struct balloon_dev_info *b_dev_info)
continue; continue;
} }
#endif #endif
spin_lock_irqsave(&b_dev_info->pages_lock, flags);
balloon_page_delete(page); balloon_page_delete(page);
__count_vm_event(BALLOON_DEFLATE); __count_vm_event(BALLOON_DEFLATE);
spin_unlock_irqrestore(&b_dev_info->pages_lock, flags);
unlock_page(page); unlock_page(page);
dequeued_page = true; dequeued_page = true;
break; break;
} }
} }
spin_unlock_irqrestore(&b_dev_info->pages_lock, flags);
if (!dequeued_page) { if (!dequeued_page) {
/* /*

33
scripts/checkpatch.pl
View File

@ -5116,13 +5116,44 @@ sub process {
} }
} }
# check for memory barriers without a comment. # check for memory barriers without a comment.
if ($line =~ /\b(mb|rmb|wmb|read_barrier_depends|smp_mb|smp_rmb|smp_wmb|smp_read_barrier_depends)\(/) {
my $barriers = qr{
mb|
rmb|
wmb|
read_barrier_depends
}x;
my $barrier_stems = qr{
mb__before_atomic|
mb__after_atomic|
store_release|
load_acquire|
store_mb|
(?:$barriers)
}x;
my $all_barriers = qr{
(?:$barriers)|
smp_(?:$barrier_stems)|
virt_(?:$barrier_stems)
}x;
if ($line =~ /\b(?:$all_barriers)\s*\(/) {
if (!ctx_has_comment($first_line, $linenr)) { if (!ctx_has_comment($first_line, $linenr)) {
WARN("MEMORY_BARRIER", WARN("MEMORY_BARRIER",
"memory barrier without comment\n" . $herecurr); "memory barrier without comment\n" . $herecurr);
} }
} }
my $underscore_smp_barriers = qr{__smp_(?:$barrier_stems)}x;
if ($realfile !~ m@^include/asm-generic/@ &&
$realfile !~ m@/barrier\.h$@ &&
$line =~ m/\b(?:$underscore_smp_barriers)\s*\(/ &&
$line !~ m/^.\s*\#\s*define\s+(?:$underscore_smp_barriers)\s*\(/) {
WARN("MEMORY_BARRIER",
"__smp memory barriers shouldn't be used outside barrier.h and asm-generic\n" . $herecurr);
}
# check for waitqueue_active without a comment. # check for waitqueue_active without a comment.
if ($line =~ /\bwaitqueue_active\s*\(/) { if ($line =~ /\bwaitqueue_active\s*\(/) {
if (!ctx_has_comment($first_line, $linenr)) { if (!ctx_has_comment($first_line, $linenr)) {