From owner-svn-src-all@freebsd.org Thu Nov 5 00:11:04 2015 Return-Path: Delivered-To: svn-src-all@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id B7CB6A26D6F; Thu, 5 Nov 2015 00:11:04 +0000 (UTC) (envelope-from jason.harmening@gmail.com) Received: from mail-lf0-x22f.google.com (mail-lf0-x22f.google.com [IPv6:2a00:1450:4010:c07::22f]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G2" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 2BD4F1E88; Thu, 5 Nov 2015 00:11:04 +0000 (UTC) (envelope-from jason.harmening@gmail.com) Received: by lfs39 with SMTP id 39so20444025lfs.3; Wed, 04 Nov 2015 16:11:02 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=rXT0IGHUab14ykYIgqGCNe3RPJ277nWS/VeihUtT36s=; b=ZvmJmdA9w7O9WCDmWLvqjG/H1IB6serzX6gU09dWAeM+Jc5P4Ygr4+eQ2YDf9+4fV2 7LTCywS1XRaMh61l67oBBsAMfEc7IhklYosj5bPCeSFvA/rTv/mDJWE3lb7jMA4sMb9U f437tPeAOQnVt9TfalbzWj2MttYhEAo2aTvcXGNow/JzsJmsR56wcFhLEq2ypbILtLAQ CdYDj9T+Z2wNQnTl0cSilKRsGgGQcT2so9zm7g+V2fE6KbMHEXmAqm38x6y4qaVFZn9w avddnHy6r+M5mb3B17U3lYfVfuxRaFOgf/CXgBBRWDcvuS3WsVzfuSnmYupfB3woMSvl 958Q== MIME-Version: 1.0 X-Received: by 10.25.26.79 with SMTP id a76mr1080341lfa.109.1446682261916; Wed, 04 Nov 2015 16:11:01 -0800 (PST) Received: by 10.112.126.104 with HTTP; Wed, 4 Nov 2015 16:11:01 -0800 (PST) In-Reply-To: References: <201510221638.t9MGc1cc053885@repo.freebsd.org> <56348FF8.3010602@gmail.com> <1446394311.91534.236.camel@freebsd.org> Date: Wed, 4 Nov 2015 16:11:01 -0800 Message-ID: Subject: Re: svn commit: r289759 - in head/sys/arm: arm include From: Jason Harmening To: Svatopluk Kraus Cc: Ian Lepore , Ganbold Tsagaankhuu , src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Content-Type: text/plain; charset=UTF-8 X-Content-Filtered-By: Mailman/MimeDel 2.1.20 X-BeenThere: svn-src-all@freebsd.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: "SVN commit messages for the entire src tree \(except for " user" and " projects" \)" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 05 Nov 2015 00:11:04 -0000 On Wed, Nov 4, 2015 at 2:17 PM, Svatopluk Kraus wrote: > On Tue, Nov 3, 2015 at 8:45 AM, Jason Harmening > wrote: > > > > On Sun, Nov 1, 2015 at 8:11 AM, Ian Lepore wrote: > >> > >> > >> It's almost certainly not related to sysinit ordering. This exception > >> is happening during mmc probing after interrupts are enabled. > >> > >> It appears that the problem is the faulting code is running on one of > >> the very early pre-allocated kernel stacks (perhaps in an interrupt > >> handler on an idle thread stack), and these stacks are not in memory > >> represented in the vm page arrays. The mmc code is using a 64-byte > >> buffer on the stack and mapping it for DMA. Normally that causes a > >> bounce for cacheline alignment, but unluckily in this case that buffer > >> on the stack just happened to be aligned to a cacheline boundary and a > >> multiple of the cacheline size, so no bounce. That causes the new sync > >> logic that is based on keeping vm_page_t pointers and offsets to get a > >> NULL pointer back from PHYS_TO_VM_PAGE when mapping, then it dies at > >> sync time trying to dereference that. It used to work because the sync > >> logic used to use the vaddr, not a page pointer. > >> > >> Michal was working on a patch yesterday. > >> > > > > Ah, thanks for pointing that out Ian. I was left scratching my head > > (admittedly on the road and w/o easy access to the code) wondering what > on > > earth would be trying to do DMA during SI_SUB_CPU. > > > > > Using of fictitious pages is not so easy here as in case pointed by > kib@ where they are allocated and freed inside one function. For sync > list sake, they must be allocated when a buffer is loaded and freed > when is unloaded. > > Michal uses pmap_kextract() in case when KVA of buffer is not zero in > his proof-of-concept patch: > https://gist.github.com/strejda/d5ca3ed202427a2e0557 > When KVA of buffer is not zero, temporary mapping is not used at all, > so vm_page_t array is not needed too. > > IMO, buffer's physical address can be saved in sync list to be used > when its KVA is not zero. Thus pmap_kextract() won't be called in > dma_dcache_sync(). > I like the idea of storing off the physical address. If you want to save space in the sync list, I think you can place busaddr and pages in a union, using vaddr == 0 to select which field to use. Some people frown upon use of unions, though. Any reason the panics on PHYS_TO_VM_PAGE in load_buffer() and load_phys() shouldn't be KASSERTs instead?