From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sinan Kaya Subject: Re: [PATCH v7 0/7] netdev: intel: Eliminate duplicate barriers on weakly-ordered archs Date: Tue, 27 Mar 2018 08:42:05 -0400 Message-ID: References: <1521831180-25014-1-git-send-email-okaya@codeaurora.org> <1521849496.15055.16.camel@intel.com> <862cdbeafb9cfd272a426b010943ffc5@codeaurora.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Cc: Alexander Duyck , intel-wired-lan , Netdev , Timur Tabi , sulrich@codeaurora.org, linux-arm-msm@vger.kernel.org, linux-arm-kernel@lists.infradead.org To: Jeff Kirsher Return-path: Received: from smtp.codeaurora.org ([198.145.29.96]:42710 "EHLO smtp.codeaurora.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751732AbeC0MmK (ORCPT ); Tue, 27 Mar 2018 08:42:10 -0400 In-Reply-To: <862cdbeafb9cfd272a426b010943ffc5@codeaurora.org> Content-Language: en-US Sender: netdev-owner@vger.kernel.org List-ID: Jeff, On 3/23/2018 10:34 PM, okaya@codeaurora.org wrote: > On 2018-03-23 19:58, Jeff Kirsher wrote: >> On Fri, 2018-03-23 at 14:53 -0700, Alexander Duyck wrote: >>> On Fri, Mar 23, 2018 at 11:52 AM, Sinan Kaya >>> wrote: >>> > Code includes wmb() followed by writel() in multiple places. writel() >>> > already has a barrier on some architectures like arm64. >>> > >>> > This ends up CPU observing two barriers back to back before executing >>> > the >>> > register write. >>> > >>> > Since code already has an explicit barrier call, changing writel() to >>> > writel_relaxed(). >>> > >>> > I did a regex search for wmb() followed by writel() in each drivers >>> > directory. >>> > I scrubbed the ones I care about in this series. >>> > >>> > I considered "ease of change", "popular usage" and "performance >>> > critical >>> > path" as the determining criteria for my filtering. >>> > >>> > We used relaxed API heavily on ARM for a long time but >>> > it did not exist on other architectures. For this reason, relaxed >>> > architectures have been paying double penalty in order to use the >>> > common >>> > drivers. >>> > >>> > Now that relaxed API is present on all architectures, we can go and >>> > scrub >>> > all drivers to see what needs to change and what can remain. >>> > >>> > We start with mostly used ones and hope to increase the coverage over >>> > time. >>> > It will take a while to cover all drivers. >>> > >>> > Feel free to apply patches individually. >>> >>> I looked over the set and they seem good. >>> >>> Reviewed-by: Alexander Duyck >> >> Grrr, patch 1 does not apply cleanly to my next-queue tree (dev-queue >> branch).  I will deal with this series in a day or two, after I have dealt >> with my driver pull requests. > > Sorry, you will have to replace the ones you took from me. Double sorry now. I don't know if you have been following "RFC on writel and writel_relaxed" thread or not but there are some new developments about wmb() requirement. Basically, wmb() should never be used before writel() as writel() seem to provide coherency and observability guarantee. wmb()+writel_relaxed() is slower on some architectures than plain writel() I'll have to rework these patches to have writel() only. Are you able to drop the applied ones so that I can post V8 or is it too late? Sinan -- Sinan Kaya Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc. Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.