From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Message-ID: <17903.33399.786962.498809@cargo.ozlabs.ibm.com> Date: Thu, 8 Mar 2007 14:26:47 +1100 From: Paul Mackerras To: Bill Huey (hui) Subject: Re: [patch 2/6 -rt] powerpc 2.6.20-rt8: to convert spinlocks to raw ones. In-Reply-To: <20070308004344.GA23200@gnuppy.monkey.org> References: <45EECE5C.6030100@ru.mvista.com> <17902.60669.945986.186995@cargo.ozlabs.ibm.com> <45EEF6A4.9090807@ru.mvista.com> <17903.4302.502364.126701@cargo.ozlabs.ibm.com> <45EF2CD8.3050804@ru.mvista.com> <17903.12035.781782.317694@cargo.ozlabs.ibm.com> <20070308004344.GA23200@gnuppy.monkey.org> Cc: mingo@elte.hu, linux-kernel@vger.kernel.org, linuxppc-dev@ozlabs.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Bill Huey (hui) writes: > The places that need to be reverted to raw spinlocks are generally either > acquired by function calls that allocate the spinlock at a terminal of the > kernel's lock graph or isolated from other callers completely (parts of the > timer for logic for instance). It's all about the collision of various lock > (preemptive and non-preemptive) subtrees and how to avoid scheduling within > atomic violations that lead to deadlocks. The -rt patch gets arbitrary > preemption abilities by shrinking the non-preemptive sub-tree bit to the bare > essentials of what will let a system to run yet still preserve all of > the expected locking semantics of a critical section. Thanks; that's an interesting explanation. It misses the point of what I was saying to Sergei, though, which was *not* "I don't understand your patch", it was "if this patch goes into a git tree, someone coming along in 3 years time won't understand the patch." In other words I was ranting about the need for a decent description to accompany the patch itself, so it would go into the permanent record. Regards, Paul.