defer/rcuintro: Substitute smp_store_release() for WRITE_ONCE()
PATCH 1/1 defer/rcuintro: Substitute smp_store_release() for WRITE_ONCE()
From 6845d0f978a81ef17d0011bebb500087f39f9a45 Mon Sep 17 00:00:00 2001
From: Akira Yokosawa <[email protected]>
Date: Thu, 24 Oct 2019 00:19:13 +0900
Subject: [PATCH] defer/rcuintro: Substitute smp_store_release() for WRITE_ONCE()
This change corresponds to the update of Figure 9.7 by commit 6bc87460e2ad ("defer/rcuintro: Fix colors, fonts, and primitives in Figures 9.6 and 9.7").
--- defer/rcuintro.tex | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
defer/rcuintro.tex +1 -1
@@ -295,7 +295,7 @@ Again, this same constraint is imposed on reader threads dereferencing they are done using the pointed-to data item. Returning to the second row of Figure~\ref{fig:defer:Deletion With Concurrent Readers}, -where the updater has just completed executing the \co{WRITE_ONCE()}, +where the updater has just completed executing the \co{smp_store_release()}, imagine that CPU~0 executes a context switch. Because readers are not permitted to block while traversing the linked list, we are guaranteed that all prior readers that might have been running on
Signature · 1 line
2.17.1
Paul E. McKenney
AppliedQuoted 10 lines · Akira Yokosawa
Quoted 1 line
From
6845d0f978a81ef17d0011bebb500087f39f9a45Mon Sep 17 00:00:00 2001From: Akira Yokosawa <[email protected]>
Date: Thu, 24 Oct 2019 00:19:13 +0900
Subject: [PATCH] defer/rcuintro: Substitutesmp_store_release()forWRITE_ONCE()This change corresponds to the update of Figure 9.7 by commit
6bc87460e2ad("defer/rcuintro: Fix colors, fonts, and primitives in Figures 9.6 and 9.7").Signed-off-by Akira Yokosawa <[email protected]>
Queued and pushed, thank you!
Thanx, Paul
Quoted 20 lines
--- defer/rcuintro.tex | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)defer/rcuintro.tex +1
-1@@ -295,7 +295,7 @@ Again, this same constraint is imposed on reader threads dereferencing they are done using the pointed-to data item. Returning to the second row of Figure~\ref{fig:defer:Deletion With Concurrent Readers}, -where the updater has just completed executing the \co{WRITE_ONCE()}, +where the updater has just completed executing the \co{smp_store_release()}, imagine that CPU~0 executes a context switch. Because readers are not permitted to block while traversing the linked list, we are guaranteed that all prior readers that might have been running onSignature · 1 line
2.17.1