From fed0764fafd8e2e629a033c0f7df4106b0dcb7f0 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Mon, 25 Jan 2016 16:33:20 -0500 Subject: [PATCH] locking/atomics: Update comment about READ_ONCE() and structures The comment is out of data. Also point out the performance drawback of the barrier();__builtin_memcpy(); barrier() followed by another copy from stack (__u) to lvalue; Signed-off-by: Konrad Rzeszutek Wilk Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Christian Borntraeger Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Sasha Levin Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1453757600-11441-1-git-send-email-konrad.wilk@oracle.com [ Made it a bit more readable. ] Signed-off-by: Ingo Molnar --- include/linux/compiler.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/include/linux/compiler.h b/include/linux/compiler.h index 00b042c..4291592 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -263,8 +263,9 @@ static __always_inline void __write_once_size(volatile void *p, void *res, int s * In contrast to ACCESS_ONCE these two macros will also work on aggregate * data types like structs or unions. If the size of the accessed data * type exceeds the word size of the machine (e.g., 32 bits or 64 bits) - * READ_ONCE() and WRITE_ONCE() will fall back to memcpy and print a - * compile-time warning. + * READ_ONCE() and WRITE_ONCE() will fall back to memcpy(). There's at + * least two memcpy()s: one for the __builtin_memcpy() and then one for + * the macro doing the copy of variable - '__u' allocated on the stack. * * Their two major use cases are: (1) Mediating communication between * process-level code and irq/NMI handlers, all running on the same CPU, -- 2.7.4