On Sun, May 28, 2023 at 04:40:20PM +0000, Alan Huang wrote:
> There seem to be some issues with the description of state transitions.
> This patch fixes that.
>
> Signed-off-by: Alan Huang <mmpgouride@xxxxxxxxx>
> ---
> together/applyrcu.tex | 7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/together/applyrcu.tex b/together/applyrcu.tex
> index 09a265f7..dee027b9 100644
> --- a/together/applyrcu.tex
> +++ b/together/applyrcu.tex
> @@ -676,11 +676,12 @@ The normal path through this state machine traverses the states CLOSED,
> OPEN, CLOSING (with an invocation of \co{call_rcu()}), and back to CLOSED
> once the callback has been invoked.
> If \co{open()} is invoked before the grace period completes, the
> -state machine traverses the cycle OPEN, CLOSING (with an invocation
> -of \co{call_rcu()}), and back to CLOSED once the callback has been invoked.
> +state machine traverses the cycle OPEN, CLOSING (with
> +an invocation of \co{call_rcu()}), REOPENING, and back to OPEN once the
> +callback has been invoked.
> If \co{open()} and then \co{close()} are invoked before the grace period
> completes, the state machine traverses the cycle OPEN, CLOSING (with
> -an invocation of \co{call_rcu()}), REOPENING, and back to OPEN once the
> +an invocation of \co{call_rcu()}), REOPENING, RECLOSING, and back to CLOSING once the
> callback has been invoked.
I took this patch (thank you!), but wonder if we should describe the
additional callback and additional transition to CLOSED, both here
and below. Thoughts?
Thanx, Paul
> Given an indefinite alternating sequence of \co{close()} and \co{open()}
> --
> 2.34.1
>
