Signed-off-by: Kunwu Chan <[email protected]>
---
 defer/seqlock.tex | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/defer/seqlock.tex b/defer/seqlock.tex
index 3ed2f483..6d1d5b78 100644
--- a/defer/seqlock.tex
+++ b/defer/seqlock.tex
@@ -166,7 +166,7 @@ will pass to a later call to \co{read_seqretry()}.
        on the update side.
        This feature is on its way into the C++ standard under the name
        of ``asymmetric fences''~\cite{DavidGoldblatt2022asymmetricFences}.
-       Either way, this trick eliminates the update-side \co{smp_mb()}
+       Either way, this trick eliminates the read-side \co{smp_mb()}
        overhead, which in turn makes eliminating the check more
        attractive, arriving again at the form of the code shown in
        \cref{lst:defer:Sequence-Locked Pre-BSD Routing Table Lookup}.
-- 
2.25.1


Reply via email to