Lines Matching refs:smp_mb
51 smp_mb(), use mb(). See the "Linux Kernel Device Drivers" book or the
60 o The smp_mb() full memory barrier.
67 First, the smp_mb() full memory barrier orders all of the CPU's prior
74 smp_mb(); // Order store to x before load from y.
116 of magnitude greater overhead than smp_mb(), atomic_xchg(), and so on.
133 all architectures is to add a call to smp_mb():
137 smp_mb(); // Inefficient on x86!!!
140 This works, but the added smp_mb() adds needless overhead for
279 ordering, and an smp_mb() would be needed instead:
282 smp_mb();
285 But smp_mb() often incurs much higher overhead than does
289 In contrast, the smp_mb() compiles to an expensive instruction that