transform the above code into the following:
q = READ_ONCE(a);
- WRITE_ONCE(b, 1);
+ WRITE_ONCE(b, 2);
do_something_else();
Given this transformation, the CPU is not required to respect the ordering
spin_unlock(Q);
-See Documentation/DocBook/deviceiobook.tmpl for more information.
+See Documentation/driver-api/device-io.rst for more information.
=================================
relaxed memory access properties, then _mandatory_ memory barriers are
required to enforce ordering.
-See Documentation/DocBook/deviceiobook.tmpl for more information.
+See Documentation/driver-api/device-io.rst for more information.
INTERRUPTS