[$] Expediting membarrier()

The membarrier()system call is arguably one of the strangest offered by the Linux kernel. It expensively emulates an operation that can beperformed by a single unprivileged barrier instruction, using an invocationof the kernel's read-copy-update (RCU) machinery - all in the name ofperformance. But, it would seem, membarrier() is not fast enough,causing users to fall back to complex and brittle tricks. An attempt tofix the problem is now under discussion, but not everybody is convincedthat the cure is better than the disease.