From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ingo Molnar Date: Wed, 13 Feb 2019 07:45:44 +0000 Subject: Re: [PATCH v2 2/2] locking/rwsem: Optimize down_read_trylock() Message-Id: <20190213074544.GB62549@gmail.com> List-Id: References: <1549913486-16799-1-git-send-email-longman@redhat.com> <1549913486-16799-3-git-send-email-longman@redhat.com> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Waiman Long Cc: Linus Torvalds , Peter Zijlstra , Ingo Molnar , Will Deacon , Thomas Gleixner , Linux List Kernel Mailing , linux-alpha@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-hexagon@vger.kernel.org, linux-ia64@vger.kernel.org, linuxppc-dev@lists.ozlabs.org, Linux-sh list , sparclinux@vger.kernel.org, linux-xtensa@linux-xtensa.org, linux-arch , the arch/x86 maintainers , Arnd Bergmann , Borislav Petkov , "H. Peter Anvin" , Davidlohr Bueso , Andrew Morton * Waiman Long wrote: > I looked at the assembly code in arch/x86/include/asm/rwsem.h. For both > trylocks (read & write), the count is read first before attempting to > lock it. We did the same for all trylock functions in other locks. > Depending on how the trylock is used and how contended the lock is, it > may help or hurt performance. Changing down_read_trylock to do an > unconditional cmpxchg will change the performance profile of existing > code. So I would prefer keeping the current code. > > I do notice now that the generic down_write_trylock() code is doing an > unconditional compxchg. So I wonder if we should change it to read the > lock first like other trylocks or just leave it as it is. No, I think we should instead move the other trylocks to the try-for-ownership model as well, like Linus suggested. That's the general assumption we make in locking primitives, that we optimize for the common, expected case - which would be that the trylock succeeds, and I don't see why trylock primitives should be different. In fact I can see more ways for read-for-sharing to perform suboptimally on larger systems. Thanks, Ingo