From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756934Ab3BRSHu (ORCPT ); Mon, 18 Feb 2013 13:07:50 -0500 Received: from mail-qe0-f43.google.com ([209.85.128.43]:46879 "EHLO mail-qe0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756814Ab3BRSHq (ORCPT ); Mon, 18 Feb 2013 13:07:46 -0500 MIME-Version: 1.0 In-Reply-To: <51226B46.9080707@linux.vnet.ibm.com> References: <20130218123714.26245.61816.stgit@srivatsabhat.in.ibm.com> <20130218123856.26245.46705.stgit@srivatsabhat.in.ibm.com> <5122551E.1080703@linux.vnet.ibm.com> <51226B46.9080707@linux.vnet.ibm.com> Date: Tue, 19 Feb 2013 02:07:45 +0800 Message-ID: Subject: Re: [PATCH v6 04/46] percpu_rwlock: Implement the core design of Per-CPU Reader-Writer Locks From: Michel Lespinasse To: "Srivatsa S. Bhat" Cc: linux-doc@vger.kernel.org, peterz@infradead.org, fweisbec@gmail.com, linux-kernel@vger.kernel.org, mingo@kernel.org, linux-arch@vger.kernel.org, linux@arm.linux.org.uk, xiaoguangrong@linux.vnet.ibm.com, wangyun@linux.vnet.ibm.com, paulmck@linux.vnet.ibm.com, nikunj@linux.vnet.ibm.com, linux-pm@vger.kernel.org, rusty@rustcorp.com.au, rostedt@goodmis.org, rjw@sisk.pl, namhyung@kernel.org, tglx@linutronix.de, linux-arm-kernel@lists.infradead.org, netdev@vger.kernel.org, oleg@redhat.com, vincent.guittot@linaro.org, sbw@mit.edu, tj@kernel.org, akpm@linux-foundation.org, linuxppc-dev@lists.ozlabs.org Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Feb 19, 2013 at 1:56 AM, Srivatsa S. Bhat wrote: > On 02/18/2013 09:51 PM, Srivatsa S. Bhat wrote: >> On 02/18/2013 09:15 PM, Michel Lespinasse wrote: >>> I don't see anything preventing a race with the corresponding code in >>> percpu_write_unlock() that sets writer_signal back to false. Did I >>> miss something here ? It seems to me we don't have any guarantee that >>> all writer signals will be set to true at the end of the loop... >> >> Ah, thanks for pointing that out! IIRC Oleg had pointed this issue in the last >> version, but back then, I hadn't fully understood what he meant. Your >> explanation made it clear. I'll work on fixing this. > > We can fix this by using the simple patch (untested) shown below. > The alternative would be to acquire the rwlock for write, update the > ->writer_signal values, release the lock, wait for readers to switch, > again acquire the rwlock for write with interrupts disabled etc... which > makes it kinda messy, IMHO. So I prefer the simple version shown below. Looks good. Another alternative would be to make writer_signal an atomic integer instead of a bool. That way writers can increment it before locking and decrement it while unlocking. To reduce the number of atomic ops during writer lock/unlock, the writer_signal could also be a global read_mostly variable (I don't see any downsides to that compared to having it percpu - or is it because you wanted all the fastpath state to be in one single cacheline ?) -- Michel "Walken" Lespinasse A program is never fully debugged until the last user dies. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qe0-f52.google.com (mail-qe0-f52.google.com [209.85.128.52]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority" (not verified)) by ozlabs.org (Postfix) with ESMTPS id 3DF8A2C0292 for ; Tue, 19 Feb 2013 05:07:48 +1100 (EST) Received: by mail-qe0-f52.google.com with SMTP id 6so2623855qeb.11 for ; Mon, 18 Feb 2013 10:07:46 -0800 (PST) MIME-Version: 1.0 In-Reply-To: <51226B46.9080707@linux.vnet.ibm.com> References: <20130218123714.26245.61816.stgit@srivatsabhat.in.ibm.com> <20130218123856.26245.46705.stgit@srivatsabhat.in.ibm.com> <5122551E.1080703@linux.vnet.ibm.com> <51226B46.9080707@linux.vnet.ibm.com> Date: Tue, 19 Feb 2013 02:07:45 +0800 Message-ID: Subject: Re: [PATCH v6 04/46] percpu_rwlock: Implement the core design of Per-CPU Reader-Writer Locks From: Michel Lespinasse To: "Srivatsa S. Bhat" Content-Type: text/plain; charset=ISO-8859-1 Cc: linux-doc@vger.kernel.org, peterz@infradead.org, fweisbec@gmail.com, linux-kernel@vger.kernel.org, namhyung@kernel.org, mingo@kernel.org, linux-arch@vger.kernel.org, linux@arm.linux.org.uk, xiaoguangrong@linux.vnet.ibm.com, wangyun@linux.vnet.ibm.com, paulmck@linux.vnet.ibm.com, nikunj@linux.vnet.ibm.com, linux-pm@vger.kernel.org, rusty@rustcorp.com.au, rostedt@goodmis.org, rjw@sisk.pl, vincent.guittot@linaro.org, tglx@linutronix.de, linux-arm-kernel@lists.infradead.org, netdev@vger.kernel.org, oleg@redhat.com, sbw@mit.edu, tj@kernel.org, akpm@linux-foundation.org, linuxppc-dev@lists.ozlabs.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Tue, Feb 19, 2013 at 1:56 AM, Srivatsa S. Bhat wrote: > On 02/18/2013 09:51 PM, Srivatsa S. Bhat wrote: >> On 02/18/2013 09:15 PM, Michel Lespinasse wrote: >>> I don't see anything preventing a race with the corresponding code in >>> percpu_write_unlock() that sets writer_signal back to false. Did I >>> miss something here ? It seems to me we don't have any guarantee that >>> all writer signals will be set to true at the end of the loop... >> >> Ah, thanks for pointing that out! IIRC Oleg had pointed this issue in the last >> version, but back then, I hadn't fully understood what he meant. Your >> explanation made it clear. I'll work on fixing this. > > We can fix this by using the simple patch (untested) shown below. > The alternative would be to acquire the rwlock for write, update the > ->writer_signal values, release the lock, wait for readers to switch, > again acquire the rwlock for write with interrupts disabled etc... which > makes it kinda messy, IMHO. So I prefer the simple version shown below. Looks good. Another alternative would be to make writer_signal an atomic integer instead of a bool. That way writers can increment it before locking and decrement it while unlocking. To reduce the number of atomic ops during writer lock/unlock, the writer_signal could also be a global read_mostly variable (I don't see any downsides to that compared to having it percpu - or is it because you wanted all the fastpath state to be in one single cacheline ?) -- Michel "Walken" Lespinasse A program is never fully debugged until the last user dies. From mboxrd@z Thu Jan 1 00:00:00 1970 From: walken@google.com (Michel Lespinasse) Date: Tue, 19 Feb 2013 02:07:45 +0800 Subject: [PATCH v6 04/46] percpu_rwlock: Implement the core design of Per-CPU Reader-Writer Locks In-Reply-To: <51226B46.9080707@linux.vnet.ibm.com> References: <20130218123714.26245.61816.stgit@srivatsabhat.in.ibm.com> <20130218123856.26245.46705.stgit@srivatsabhat.in.ibm.com> <5122551E.1080703@linux.vnet.ibm.com> <51226B46.9080707@linux.vnet.ibm.com> Message-ID: To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Tue, Feb 19, 2013 at 1:56 AM, Srivatsa S. Bhat wrote: > On 02/18/2013 09:51 PM, Srivatsa S. Bhat wrote: >> On 02/18/2013 09:15 PM, Michel Lespinasse wrote: >>> I don't see anything preventing a race with the corresponding code in >>> percpu_write_unlock() that sets writer_signal back to false. Did I >>> miss something here ? It seems to me we don't have any guarantee that >>> all writer signals will be set to true at the end of the loop... >> >> Ah, thanks for pointing that out! IIRC Oleg had pointed this issue in the last >> version, but back then, I hadn't fully understood what he meant. Your >> explanation made it clear. I'll work on fixing this. > > We can fix this by using the simple patch (untested) shown below. > The alternative would be to acquire the rwlock for write, update the > ->writer_signal values, release the lock, wait for readers to switch, > again acquire the rwlock for write with interrupts disabled etc... which > makes it kinda messy, IMHO. So I prefer the simple version shown below. Looks good. Another alternative would be to make writer_signal an atomic integer instead of a bool. That way writers can increment it before locking and decrement it while unlocking. To reduce the number of atomic ops during writer lock/unlock, the writer_signal could also be a global read_mostly variable (I don't see any downsides to that compared to having it percpu - or is it because you wanted all the fastpath state to be in one single cacheline ?) -- Michel "Walken" Lespinasse A program is never fully debugged until the last user dies.