Re: [PATCH v4 3/4] locking/qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32

From: Peter Zijlstra <peterz@infradead.org>
To: Arnd Bergmann <arnd@arndb.de>
Cc: Stafford Horne <shorne@gmail.com>, Guo Ren <guoren@kernel.org>,
	linux-riscv <linux-riscv@lists.infradead.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	linux-csky@vger.kernel.org,
	linux-arch <linux-arch@vger.kernel.org>,
	Guo Ren <guoren@linux.alibaba.com>, Will Deacon <will@kernel.org>,
	Ingo Molnar <mingo@redhat.com>, Waiman Long <longman@redhat.com>,
	Anup Patel <anup@brainfault.org>
Subject: Re: [PATCH v4 3/4] locking/qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32
Date: Wed, 7 Apr 2021 13:36:45 +0200	[thread overview]
Message-ID: <YG2ZTSFMGrikYWuL@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <CAK8P3a3Pf3TbGoVP7JP7gfPV-WDM8MHV_hdqSwNKKFDr1Sb3zQ@mail.gmail.com>

On Wed, Apr 07, 2021 at 10:42:50AM +0200, Arnd Bergmann wrote:
> Since there are really only a handful of instances in the kernel
> that use the cmpxchg() or xchg() on u8/u16 variables, it would seem
> best to just disallow those completely 

Not going to happen. xchg16 is optimal for qspinlock and if we replace
that with a cmpxchg loop on x86 we're regressing.

> Interestingly, the s390 version using __sync_val_compare_and_swap()
> seems to produce nice output on all architectures that have atomic
> instructions, with any supported compiler, to the point where I think
> we could just use that to replace most of the inline-asm versions except
> for arm64:
> 
> #define cmpxchg(ptr, o, n)                                              \
> ({                                                                      \
>         __typeof__(*(ptr)) __o = (o);                                   \
>         __typeof__(*(ptr)) __n = (n);                                   \
>         (__typeof__(*(ptr))) __sync_val_compare_and_swap((ptr),__o,__n);\
> })

It generates the LL/SC loop, but doesn't do sensible optimizations when
it's again used in a loop itself. That is, it generates a loop of a
loop, just like what you'd expect, which is sub-optimal for LL/SC.

> Not how gcc's acquire/release behavior of __sync_val_compare_and_swap()
> relates to what the kernel wants here.
> 
> The gcc documentation also recommends using the standard
> __atomic_compare_exchange_n() builtin instead, which would allow
> constructing release/acquire/relaxed versions as well, but I could not
> get it to produce equally good output. (possibly I was using it wrong)

I'm scared to death of the C11 crap, the compiler will 'optimize' them
when it feels like it and use the C11 memory model rules for it, which
are not compatible with the kernel rules.

But the same thing applies, it won't do the right thing for composites.

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv