From 63038752864c6ab162cc389e5adfd94471a79f41 Mon Sep 17 00:00:00 2001 From: Samy Al Bahra Date: Sat, 13 Oct 2018 22:36:53 -0400 Subject: [PATCH] doc/ck_epoch: update poll and synchronize clarifying expected record state. --- doc/ck_epoch_poll | 7 +++---- doc/ck_epoch_synchronize | 2 ++ 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/doc/ck_epoch_poll b/doc/ck_epoch_poll index 68c4a4e..b1b1f79 100644 --- a/doc/ck_epoch_poll +++ b/doc/ck_epoch_poll @@ -47,10 +47,9 @@ if deemed safe. This function is meant to be used in cases epoch reclamation cost must be amortized over time in a manner that does not affect caller progress. .Sh RETURN VALUES -This function will return true if at least one function was dispatched. -This function will return false if it has determined not all threads have -observed the latest generation of epoch-protected objects. Neither value -indicates an error. +This functions returns false if no memory was reclaimed, the records +are not in a grace period or if no forward progress was made to the +counter. .Sh ERRORS Behavior is undefined if the object pointed to by .Fa record diff --git a/doc/ck_epoch_synchronize b/doc/ck_epoch_synchronize index 6c9a698..6e5f73d 100644 --- a/doc/ck_epoch_synchronize +++ b/doc/ck_epoch_synchronize @@ -40,6 +40,8 @@ The .Fn ck_epoch_synchronize 3 function will block the caller until a grace period has been detected, according to the semantics of epoch reclamation. +It is not safe to call this function on a record that is +in an active section. Any objects requiring safe memory reclamation which are logically deleted are safe for physical deletion following a call to .Fn ck_epoch_synchronize 3 .