From: Peter Zijlstra Subject: Re: Patch "asm-generic/bitops/lock.h: Rewrite using atomic_fetch_" causes kernel crash Date: Thu, 30 Aug 2018 16:17:13 +0200 Message-ID: <20180830141713.GN24082@hirez.programming.kicks-ass.net> References: <1535567633.4465.23.camel@synopsys.com> <20180830094411.GX24124@hirez.programming.kicks-ass.net> <20180830095148.GB5942@arm.com> <1535629996.4465.44.camel@synopsys.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <1535629996.4465.44.camel@synopsys.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=m.gmane.org@lists.infradead.org To: Eugeniy Paltsev Cc: "linux-arch@vger.kernel.org" , "Vineet.Gupta1@synopsys.com" , "Alexey.Brodkin@synopsys.com" , "will.deacon@arm.com" , "linux-kernel@vger.kernel.org" , "yamada.masahiro@socionext.com" , "tglx@linutronix.de" , "linux-snps-arc@lists.infradead.org" , "mingo@kernel.org" , "linux-arm-kernel@lists.infradead.org" List-Id: linux-arch.vger.kernel.org On Thu, Aug 30, 2018 at 11:53:17AM +0000, Eugeniy Paltsev wrote: > I can see crashes with LLSC enabled in both SMP running on 4 cores > and SMP running on 1 core. So you're running on LL/SC enabled hardware; that would make Will's patch irrelevant (although still a good idea for the hardware that does care about that spinlocked atomic crud). Does something like the below cure things? That would confirm the suggestion that the change to __clear_bit_unlock() is the curprit. If that doesn't cure things, then we've been looking in entirely the wrong place. --- diff --git a/include/asm-generic/bitops/lock.h b/include/asm-generic/bitops/lock.h index 3ae021368f48..79c6978152f8 100644 --- a/include/asm-generic/bitops/lock.h +++ b/include/asm-generic/bitops/lock.h @@ -57,12 +57,7 @@ static inline void clear_bit_unlock(unsigned int nr, volatile unsigned long *p) static inline void __clear_bit_unlock(unsigned int nr, volatile unsigned long *p) { - unsigned long old; - - p += BIT_WORD(nr); - old = READ_ONCE(*p); - old &= ~BIT_MASK(nr); - atomic_long_set_release((atomic_long_t *)p, old); + clear_bit_unlock(nr, p); } /**