kernel.git/arch/s390/include/asm/atomic.h, branch linux-3.19.y

arch,s390: Convert smp_mb__*()

2014-04-18T12:20:42Z

As per the existing implementation; implement the new one using smp_mb(). AFAICT the s390 compare-and-swap does imply a barrier, however there are some immediate ops that seem to be singly-copy atomic and do not imply a barrier. One such is the "ni" op (which would be and-immediate) which is used for the constant clear_bit implementation. Therefore s390 needs full barriers for the {before,after} atomic ops. Signed-off-by: Peter Zijlstra Acked-by: Paul E. McKenney Link: http://lkml.kernel.org/n/tip-kme5dz5hcobpnufnnkh1ech2@git.kernel.org Cc: Chen Gang Cc: Heiko Carstens Cc: Linus Torvalds Cc: Martin Schwidefsky Cc: Will Deacon Cc: linux390@de.ibm.com Cc: linux-kernel@vger.kernel.org Cc: linux-s390@vger.kernel.org Signed-off-by: Ingo Molnar

s390/bitops,atomic: add missing memory barriers

2014-04-01T07:23:35Z

When reworking the bitops and atomic ops I missed that those instructions that got atomic behaviour only perform a "specific-operand-serialization" instead of a full "serialization". The compare-and-swap instruction used before performs a full serialization before and after the instruction is executed, which means it has full memory barrier semantics. In order to give the new bitops and atomic ops functions also full memory barrier semantics add a "bcr 14,0" before and after each of those new instructions which performs full serialization as well. This restores memory barrier semantics for bitops and atomic ops functions which return values, like e.g. atomic_add_return(), but not for functions which do not return a value, like e.g. atomic_add(). This is consistent to other architectures and what common code requires. Cc: stable@vger.kernel.org # v3.13+ Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky

s390/atomic: use 'unsigned int' instead of 'unsigned long' for atomic_*_mask()

2013-10-24T15:17:02Z

The type of 'v->counter' is always 'int', and related inline assembly code also process 'int', so use 'unsigned int' instead of 'unsigned long' for the 'mask'. Signed-off-by: Chen Gang Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky

s390/atomic: various small cleanups

2013-10-24T15:16:51Z

- add a typecheck to the defines to make sure they operate on an atomic_t - simplify inline assembly constraints - keep variable names common between functions Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky

s390/atomic: optimize atomic_add() for constant values

2013-10-24T15:16:51Z

If the interlocked-access facility 1 is available we can use the asi and agsi instructions for interlocked updates if the to be added value is a contanst and small (in the range of -128..127). asi and agsi do not not return the old or new value, therefore these instructions can only be used for atomic_(add|sub|inc|dec)[64]. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky

s390/atomic: make use of interlocked-access facility 1 instructions

2013-10-24T15:16:46Z

Same as for bitops: make use of the interlocked-access facility 1 instructions which allow to atomically update storage locations without a compare-and-swap loop. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky

s390/atomic: implement atomic_sub_return() with atomic_add_return()

2013-10-24T15:16:46Z

Get rid of the own atomic_sub_return() implementation. Otherwise we can't make use of the interlocked-access facility 1 instructions for atomic_sub_return(), since there is no "load and subtract" instruction available. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky

s390/comments: unify copyright messages and remove file names

2012-07-20T09:15:04Z

Remove the file name from the comment at top of many files. In most cases the file name was wrong anyway, so it's rather pointless. Also unify the IBM copyright statement. We did have a lot of sightly different statements and wanted to change them one after another whenever a file gets touched. However that never happened. Instead people start to take the old/"wrong" statements to use as a template for new files. So unify all of them in one go. Signed-off-by: Heiko Carstens

Disintegrate asm/system.h for S390

2012-03-28T17:30:02Z

Disintegrate asm/system.h for S390. Signed-off-by: David Howells cc: linux-s390@vger.kernel.org

atomic: cleanup asm-generic atomic*.h inclusion

2011-07-26T23:49:47Z

After changing all consumers of atomics to include , we ran into some compile time errors due to this dependency chain: linux/atomic.h -> asm/atomic.h -> asm-generic/atomic-long.h where atomic-long.h could use funcs defined later in linux/atomic.h without a prototype. This patches moves the code that includes asm-generic/atomic*.h to linux/atomic.h. Archs that need need to select CONFIG_GENERIC_ATOMIC64 from now on (some of them used to include it unconditionally). Compile tested on i386 and x86_64 with allnoconfig. Signed-off-by: Arun Sharma Cc: Eric Dumazet Cc: Ingo Molnar Cc: David Miller Acked-by: Mike Frysinger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds