From 2b94895b9281976cf2a0f2f89bf41ee219742235 Mon Sep 17 00:00:00 2001 From: David Howells Date: Sun, 25 Jun 2006 05:48:49 -0700 Subject: [PATCH] [PATCH] Another couple of alterations to the memory barrier doc Make another couple of alterations to the memory barrier document following suggestions by Alan Stern and in co-operation with Paul McKenney: (*) Rework the point of introduction of memory barriers and the description of what they are to reiterate why they're needed. (*) Modify a statement about the use of data dependency barriers to note that other barriers can be used instead (as they imply DD-barriers). Signed-off-by: David Howells Acked-By: Paul E. McKenney Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/memory-barriers.txt | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/Documentation/memory-barriers.txt b/Documentation/memory-barriers.txt index 4710845dbac4..cc53f47a83e8 100644 --- a/Documentation/memory-barriers.txt +++ b/Documentation/memory-barriers.txt @@ -262,9 +262,14 @@ What is required is some way of intervening to instruct the compiler and the CPU to restrict the order. Memory barriers are such interventions. They impose a perceived partial -ordering between the memory operations specified on either side of the barrier. -They request that the sequence of memory events generated appears to other -parts of the system as if the barrier is effective on that CPU. +ordering over the memory operations on either side of the barrier. + +Such enforcement is important because the CPUs and other devices in a system +can use a variety of tricks to improve performance - including reordering, +deferral and combination of memory operations; speculative loads; speculative +branch prediction and various types of caching. Memory barriers are used to +override or suppress these tricks, allowing the code to sanely control the +interaction of multiple CPUs and/or devices. VARIETIES OF MEMORY BARRIER @@ -461,8 +466,8 @@ Whilst this may seem like a failure of coherency or causality maintenance, it isn't, and this behaviour can be observed on certain real CPUs (such as the DEC Alpha). -To deal with this, a data dependency barrier must be inserted between the -address load and the data load: +To deal with this, a data dependency barrier or better must be inserted +between the address load and the data load: CPU 1 CPU 2 =============== =============== -- 2.20.1