From: "Paul E. McKenney" <paulmck@xxxxxxxxxx> Protecting the code in a trampoline can also require protecting a number of instructions prior to actually entering the trampoline. For example, these earlier instructions might be computing the address of the trampoline. This commit therefore updates RCU's requirements to record this for posterity. Link: https://lore.kernel.org/lkml/20200511154824.09a18c46@xxxxxxxxxxxxxxxxxx/ Reported-by: Lai Jiangshan <jiangshanlai@xxxxxxxxx> Reported-by: Steven Rostedt <rostedt@xxxxxxxxxxx> Signed-off-by: Paul E. McKenney <paulmck@xxxxxxxxxx> --- Documentation/RCU/Design/Requirements/Requirements.rst | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/Documentation/RCU/Design/Requirements/Requirements.rst b/Documentation/RCU/Design/Requirements/Requirements.rst index 75b8ca0..a69b5c4 100644 --- a/Documentation/RCU/Design/Requirements/Requirements.rst +++ b/Documentation/RCU/Design/Requirements/Requirements.rst @@ -2583,7 +2583,12 @@ not work to have these markers in the trampoline itself, because there would need to be instructions following ``rcu_read_unlock()``. Although ``synchronize_rcu()`` would guarantee that execution reached the ``rcu_read_unlock()``, it would not be able to guarantee that execution -had completely left the trampoline. +had completely left the trampoline. Worse yet, in some situations +the trampoline's protection must extend a few instructions *prior* to +execution reaching the trampoline. For example, these few instructions +might calculate the address of the trampoline, so that entering the +trampoline would be pre-ordained a surprisingly long time before execution +actually reached the trampoline itself. The solution, in the form of `Tasks RCU <https://lwn.net/Articles/607117/>`__, is to have implicit read-side -- 2.9.5
