On Sat, May 05, 2018 at 03:27:51PM +0200, Ingo Molnar wrote: > > * Boqun Feng wrote: > > > > May I suggest the patch below? No change in functionality, but it documents the > > > lack of the cmpxchg_release() APIs and maps them explicitly to the full cmpxchg() > > > version. (Which the generic code does now in a rather roundabout way.) > > > > > > > Hmm.. cmpxchg_release() is actually lwsync() + cmpxchg_relaxed(), but > > you just make it sync() + cmpxchg_relaxed() + sync() with the fallback, > > and sync() is much heavier, so I don't think the fallback is correct. > > Indeed! > > The bit I missed previously is that PowerPC provides its own __atomic_op_release() > method: > > #define __atomic_op_release(op, args...) \ > ({ \ > __asm__ __volatile__(PPC_RELEASE_BARRIER "" : : : "memory"); \ > op##_relaxed(args); \ > }) > > ... which maps to LWSYNC as you say, and my patch made that worse. > > > I think maybe you can move powerpc's __atomic_op_{acqurie,release}() > > from atomic.h to cmpxchg.h (in arch/powerpc/include/asm), and > > > > #define cmpxchg_release __atomic_op_release(cmpxchg, __VA_ARGS__); > > #define cmpxchg64_release __atomic_op_release(cmpxchg64, __VA_ARGS__); > > > > I put a diff below to say what I mean (untested). > > > > > Also, the change to arch/powerpc/include/asm/atomic.h has no functional effect > > > right now either, but should anyone add a _relaxed() variant in the future, with > > > this change atomic_cmpxchg_release() and atomic64_cmpxchg_release() will pick that > > > up automatically. > > > > > > > You mean with your other modification in include/linux/atomic.h, right? > > Because with the unmodified include/linux/atomic.h, we already pick that > > autmatically. If so, I think that's fine. > > > > Here is the diff for the modification for cmpxchg_release(), the idea is > > we generate them in asm/cmpxchg.h other than linux/atomic.h for ppc, so > > we keep the new linux/atomic.h working. Because if I understand > > correctly, the next linux/atomic.h only accepts that > > > > 1) architecture only defines fully ordered primitives > > > > or > > > > 2) architecture only defines _relaxed primitives > > > > or > > > > 3) architecture defines all four (fully, _relaxed, _acquire, > > _release) primitives > > > > So powerpc needs to define all four primitives in its only > > asm/cmpxchg.h. > > Correct, although the new logic is still RFC, PeterZ didn't like the first version > I proposed and might NAK them. > Understood. From my side, I don't have strong feelings for either way. But since powerpc gets affected with the new logic, so I'm glad I could help. > Thanks for the patch - I have created the patch below from it and added your > Signed-off-by. > Thanks ;-) > The only change I made beyond a trivial build fix is that I also added the release > atomics variants explicitly: > > +#define atomic_cmpxchg_release(v, o, n) \ > + cmpxchg_release(&((v)->counter), (o), (n)) > +#define atomic64_cmpxchg_release(v, o, n) \ > + cmpxchg_release(&((v)->counter), (o), (n)) > > It has passed a PowerPC cross-build test here, but no runtime tests. > Do you have the commit at any branch in tip tree? I could pull it and cross-build and check the assembly code of lib/atomic64_test.c, that way I could verify whether we mess something up. > Does this patch look good to you? > Yep! Regards, Boqun > (Still subject to PeterZ's Ack/NAK.) > > Thanks, > > Ingo > > ======================> > From: Boqun Feng > Date: Sat, 5 May 2018 19:28:17 +0800 > Subject: [PATCH] locking/atomics/powerpc: Move cmpxchg helpers to asm/cmpxchg.h and define the full set of cmpxchg APIs > > Move PowerPC's __op_{acqurie,release}() from atomic.h to > cmpxchg.h (in arch/powerpc/include/asm), plus use them to > define these two methods: > > #define cmpxchg_release __op_release(cmpxchg, __VA_ARGS__); > #define cmpxchg64_release __op_release(cmpxchg64, __VA_ARGS__); > > ... the idea is to generate all these methods in cmpxchg.h and to define the full > array of atomic primitives, including the cmpxchg_release() methods which were > defined by the generic code before. > > Also define the atomic[64]_() variants explicitly. > > This ensures that all these low level cmpxchg APIs are defined in > PowerPC headers, with no generic header fallbacks. > > No change in functionality or code generation. > > Signed-off-by: Boqun Feng > Cc: Linus Torvalds > Cc: Mark Rutland > Cc: Peter Zijlstra > Cc: Thomas Gleixner > Cc: aryabinin@virtuozzo.com > Cc: catalin.marinas@arm.com > Cc: dvyukov@google.com > Cc: linux-arm-kernel@lists.infradead.org > Cc: will.deacon@arm.com > Link: http://lkml.kernel.org/r/20180505112817.ihrb726i37bwm4cj@tardis > Signed-off-by: Ingo Molnar > --- > arch/powerpc/include/asm/atomic.h | 22 ++++------------------ > arch/powerpc/include/asm/cmpxchg.h | 24 ++++++++++++++++++++++++ > 2 files changed, 28 insertions(+), 18 deletions(-) > > diff --git a/arch/powerpc/include/asm/atomic.h b/arch/powerpc/include/asm/atomic.h > index 682b3e6a1e21..4e06955ec10f 100644 > --- a/arch/powerpc/include/asm/atomic.h > +++ b/arch/powerpc/include/asm/atomic.h > @@ -13,24 +13,6 @@ > > #define ATOMIC_INIT(i) { (i) } > > -/* > - * Since *_return_relaxed and {cmp}xchg_relaxed are implemented with > - * a "bne-" instruction at the end, so an isync is enough as a acquire barrier > - * on the platform without lwsync. > - */ > -#define __atomic_op_acquire(op, args...) \ > -({ \ > - typeof(op##_relaxed(args)) __ret = op##_relaxed(args); \ > - __asm__ __volatile__(PPC_ACQUIRE_BARRIER "" : : : "memory"); \ > - __ret; \ > -}) > - > -#define __atomic_op_release(op, args...) \ > -({ \ > - __asm__ __volatile__(PPC_RELEASE_BARRIER "" : : : "memory"); \ > - op##_relaxed(args); \ > -}) > - > static __inline__ int atomic_read(const atomic_t *v) > { > int t; > @@ -213,6 +195,8 @@ static __inline__ int atomic_dec_return_relaxed(atomic_t *v) > cmpxchg_relaxed(&((v)->counter), (o), (n)) > #define atomic_cmpxchg_acquire(v, o, n) \ > cmpxchg_acquire(&((v)->counter), (o), (n)) > +#define atomic_cmpxchg_release(v, o, n) \ > + cmpxchg_release(&((v)->counter), (o), (n)) > > #define atomic_xchg(v, new) (xchg(&((v)->counter), new)) > #define atomic_xchg_relaxed(v, new) xchg_relaxed(&((v)->counter), (new)) > @@ -519,6 +503,8 @@ static __inline__ long atomic64_dec_if_positive(atomic64_t *v) > cmpxchg_relaxed(&((v)->counter), (o), (n)) > #define atomic64_cmpxchg_acquire(v, o, n) \ > cmpxchg_acquire(&((v)->counter), (o), (n)) > +#define atomic64_cmpxchg_release(v, o, n) \ > + cmpxchg_release(&((v)->counter), (o), (n)) > > #define atomic64_xchg(v, new) (xchg(&((v)->counter), new)) > #define atomic64_xchg_relaxed(v, new) xchg_relaxed(&((v)->counter), (new)) > diff --git a/arch/powerpc/include/asm/cmpxchg.h b/arch/powerpc/include/asm/cmpxchg.h > index 9b001f1f6b32..e27a612b957f 100644 > --- a/arch/powerpc/include/asm/cmpxchg.h > +++ b/arch/powerpc/include/asm/cmpxchg.h > @@ -8,6 +8,24 @@ > #include > #include > > +/* > + * Since *_return_relaxed and {cmp}xchg_relaxed are implemented with > + * a "bne-" instruction at the end, so an isync is enough as a acquire barrier > + * on the platform without lwsync. > + */ > +#define __atomic_op_acquire(op, args...) \ > +({ \ > + typeof(op##_relaxed(args)) __ret = op##_relaxed(args); \ > + __asm__ __volatile__(PPC_ACQUIRE_BARRIER "" : : : "memory"); \ > + __ret; \ > +}) > + > +#define __atomic_op_release(op, args...) \ > +({ \ > + __asm__ __volatile__(PPC_RELEASE_BARRIER "" : : : "memory"); \ > + op##_relaxed(args); \ > +}) > + > #ifdef __BIG_ENDIAN > #define BITOFF_CAL(size, off) ((sizeof(u32) - size - off) * BITS_PER_BYTE) > #else > @@ -512,6 +530,9 @@ __cmpxchg_acquire(void *ptr, unsigned long old, unsigned long new, > (unsigned long)_o_, (unsigned long)_n_, \ > sizeof(*(ptr))); \ > }) > + > +#define cmpxchg_release(...) __atomic_op_release(cmpxchg, __VA_ARGS__) > + > #ifdef CONFIG_PPC64 > #define cmpxchg64(ptr, o, n) \ > ({ \ > @@ -533,6 +554,9 @@ __cmpxchg_acquire(void *ptr, unsigned long old, unsigned long new, > BUILD_BUG_ON(sizeof(*(ptr)) != 8); \ > cmpxchg_acquire((ptr), (o), (n)); \ > }) > + > +#define cmpxchg64_release(...) __atomic_op_release(cmpxchg64, __VA_ARGS__) > + > #else > #include > #define cmpxchg64_local(ptr, o, n) __cmpxchg64_local_generic((ptr), (o), (n))