an interrupt entry is always a 'synchronization point' on Intel. But it's
a nice place for screwup indeed, i think there was a (K6?) errata in this
area, IRQ handlers didnt see a consistent CPU state. But the CPU is
supposed to handle this issue transparently.
-- mingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu