diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2013-09-04 08:17:12 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2013-09-04 08:17:12 -0700 |
commit | b854e4de0bf88d094476af82c0d5a80f6f2af916 (patch) | |
tree | 6e9b0a9f28468b043bd395fd816c4b44d4ac8f58 /Documentation/memory-barriers.txt | |
parent | 458c3f60ef12996d5a5b390260a5112bb1da1615 (diff) | |
parent | 7d992feb7694a21ee81f22894b455dadd5d1c110 (diff) | |
download | op-kernel-dev-b854e4de0bf88d094476af82c0d5a80f6f2af916.zip op-kernel-dev-b854e4de0bf88d094476af82c0d5a80f6f2af916.tar.gz |
Merge branch 'core-rcu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull RCU updates from Ingo Molnar:
"Main RCU changes this cycle were:
- Full-system idle detection. This is for use by Frederic
Weisbecker's adaptive-ticks mechanism. Its purpose is to allow the
timekeeping CPU to shut off its tick when all other CPUs are idle.
- Miscellaneous fixes.
- Improved rcutorture test coverage.
- Updated RCU documentation"
* 'core-rcu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (30 commits)
nohz_full: Force RCU's grace-period kthreads onto timekeeping CPU
nohz_full: Add full-system-idle state machine
jiffies: Avoid undefined behavior from signed overflow
rcu: Simplify _rcu_barrier() processing
rcu: Make rcutorture emit online failures if verbose
rcu: Remove unused variable from rcu_torture_writer()
rcu: Sort rcutorture module parameters
rcu: Increase rcutorture test coverage
rcu: Add duplicate-callback tests to rcutorture
doc: Fix memory-barrier control-dependency example
rcu: Update RTFP documentation
nohz_full: Add full-system-idle arguments to API
nohz_full: Add full-system idle states and variables
nohz_full: Add per-CPU idle-state tracking
nohz_full: Add rcu_dyntick data for scalable detection of all-idle state
nohz_full: Add Kconfig parameter for scalable detection of all-idle state
nohz_full: Add testing information to documentation
rcu: Eliminate unused APIs intended for adaptive ticks
rcu: Select IRQ_WORK from TREE_PREEMPT_RCU
rculist: list_first_or_null_rcu() should use list_entry_rcu()
...
Diffstat (limited to 'Documentation/memory-barriers.txt')
-rw-r--r-- | Documentation/memory-barriers.txt | 10 |
1 files changed, 6 insertions, 4 deletions
diff --git a/Documentation/memory-barriers.txt b/Documentation/memory-barriers.txt index fa5d8a9..c8c42e6 100644 --- a/Documentation/memory-barriers.txt +++ b/Documentation/memory-barriers.txt @@ -531,9 +531,10 @@ dependency barrier to make it work correctly. Consider the following bit of code: q = &a; - if (p) + if (p) { + <data dependency barrier> q = &b; - <data dependency barrier> + } x = *q; This will not have the desired effect because there is no actual data @@ -542,9 +543,10 @@ attempting to predict the outcome in advance. In such a case what's actually required is: q = &a; - if (p) + if (p) { + <read barrier> q = &b; - <read barrier> + } x = *q; |