From: Christian Ehrhardt <ehrhardt@xxxxxxxxxxxxxxxxxx> The text might be read a bit misleading and suggest that e.g. test_and_clear_bit does not imply a memory barrier, but the documentation in memory-barriers.txt as well as the function kernel api doc in asm-generic/bitops/atomic.h state that the interface defines an implicit memory barrier. Therefore this patch adds a little statement to defang the potential misleading phrase. Signed-off-by: Christian Ehrhardt <ehrhardt@xxxxxxxxxxxxxxxxxx> --- [diffstat] atomic_ops.txt | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) [diff] Index: linux-2.6/Documentation/atomic_ops.txt =================================================================== --- linux-2.6.orig/Documentation/atomic_ops.txt 2009-06-08 14:46:22.000000000 +0200 +++ linux-2.6/Documentation/atomic_ops.txt 2009-06-08 15:09:54.000000000 +0200 @@ -357,7 +357,8 @@ Like the above, except that these routines return a boolean which indicates whether the changed bit was set _BEFORE_ the atomic bit -operation. +operation. Unlike the above, these extended interfaces imply a +memory barrier. WARNING! It is incredibly important that the value be a boolean, ie. "0" or "1". Do not try to be fancy and save a few instructions by -- To unsubscribe from this list: send the line "unsubscribe linux-doc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html