From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-0.8 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id F206ECA9ECB for ; Thu, 31 Oct 2019 16:04:39 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id CE6B520873 for ; Thu, 31 Oct 2019 16:04:39 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728534AbfJaQEi (ORCPT ); Thu, 31 Oct 2019 12:04:38 -0400 Received: from mx2.suse.de ([195.135.220.15]:44768 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1726540AbfJaQEi (ORCPT ); Thu, 31 Oct 2019 12:04:38 -0400 X-Virus-Scanned: by amavisd-new at test-mx.suse.de Received: from relay2.suse.de (unknown [195.135.220.254]) by mx1.suse.de (Postfix) with ESMTP id D7552B232; Thu, 31 Oct 2019 16:04:35 +0000 (UTC) Message-ID: <6fd539b82cbbb2ae307a67a76eb4c2ead0bd5d4a.camel@suse.de> Subject: Re: [PATCH v6 3/4] arm64: use both ZONE_DMA and ZONE_DMA32 From: Nicolas Saenz Julienne To: Catalin Marinas Cc: f.fainelli@gmail.com, wahrenst@gmx.net, marc.zyngier@arm.com, will@kernel.org, Linux Kernel Mailing List , Rob Herring , linux-mm@kvack.org, mbrugger@suse.com, Qian Cai , linux-rpi-kernel@lists.infradead.org, phill@raspberrypi.org, Robin Murphy , Christoph Hellwig , linux-arm-kernel@lists.infradead.org, m.szyprowski@samsung.com Date: Thu, 31 Oct 2019 17:04:34 +0100 In-Reply-To: <20191031155145.GF39590@arrakis.emea.arm.com> References: <6703f8dab4a21fe4e1049f8f224502e1733bf72c.camel@suse.de> <9208de061fe2b9ee7b74206b3cd52cc116e43ac0.camel@suse.de> <1956a2c8f4911b2a7e2ba3c53506c0f06efb93f8.camel@suse.de> <20191031155145.GF39590@arrakis.emea.arm.com> Content-Type: multipart/signed; micalg="pgp-sha256"; protocol="application/pgp-signature"; boundary="=-exV1eg95rhXTf/heN52G" User-Agent: Evolution 3.34.1 MIME-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --=-exV1eg95rhXTf/heN52G Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Thu, 2019-10-31 at 15:51 +0000, Catalin Marinas wrote: > (sorry, I've been away last week and only now caught up with emails) >=20 > On Tue, Oct 22, 2019 at 01:23:32PM +0200, Nicolas Saenz Julienne wrote: > > On Mon, 2019-10-21 at 16:36 -0400, Qian Cai wrote: > > > I managed to get more information here, > > >=20 > > > [ 0.000000] cma: dma_contiguous_reserve(limit c0000000) > > > [ 0.000000] cma: dma_contiguous_reserve: reserving 64 MiB for glob= al > > > area > > > [ 0.000000] cma: cma_declare_contiguous(size 0x0000000004000000, b= ase > > > 0x0000000000000000, limit 0x00000000c0000000 alignment 0x000000000000= 0000) > > > [ 0.000000] cma: Failed to reserve 512 MiB > > >=20 > > > Full dmesg: > > >=20 > > > https://cailca.github.io/files/dmesg.txt > >=20 > > OK I got it, reproduced it too. > >=20 > > Here are the relevant logs: > >=20 > > [ 0.000000] DMA [mem 0x00000000802f0000-0x00000000bfffffff] > > [ 0.000000] DMA32 [mem 0x00000000c0000000-0x00000000ffffffff] > > [ 0.000000] Normal [mem 0x0000000100000000-0x00000097fcffffff] > >=20 > > As you can see ZONE_DMA spans from 0x00000000802f0000-0x00000000bffffff= f > > which > > is slightly smaller than 1GB. > >=20 > > [ 0.000000] crashkernel reserved: 0x000000009fe00000 - > > 0x00000000bfe00000 (512 MB) > >=20 > > Here crashkernel reserved 512M in ZONE_DMA. > >=20 > > [ 0.000000] cma: Failed to reserve 512 MiB > >=20 > > CMA tried to allocate 512M in ZONE_DMA which fails as there is no enoug= h > > space. > > Makes sense. > >=20 > > A fix could be moving crashkernel reservations after CMA and then if un= able > > to > > fit in ZONE_DMA try ZONE_DMA32 before bailing out. Maybe it's a little = over > > the > > top, yet although most devices will be fine with ZONE_DMA32, the RPi4 n= eeds > > crashkernel to be reserved in ZONE_DMA. >=20 > Does RPi4 need CMA in ZONE_DMA? If not, I'd rather reserve the CMA from > ZONE_DMA32. Yes, CMA is imperatively to be reserved in ZONE_DMA. > Even if you moved the crash kernel, someone else might complain that > they had 2GB of CMA and it no longer works. I have yet to look into it, but I've been told that on x86/x64 they have a 'high' flag to be set alongside with crashkernel that forces the allocation into ZONE_DMA32. We could mimic this behavior for big servers that don't de= pend on ZONE_DMA but need to reserve big chunks of memory. Regards, Nicolas --=-exV1eg95rhXTf/heN52G Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- iQEzBAABCAAdFiEErOkkGDHCg2EbPcGjlfZmHno8x/4FAl27BhIACgkQlfZmHno8 x/6ytQgAr4/GCUbLvqyD7pRmUlkzRlZj0aJHXFHEsX8ng2CM60HUWp5PPXvp74YZ gIbWndJ96KHhR79C2lG2TcpWJl4nSAtCEJohIE5vFr5//a2uhWy0PiJbYrWnLTwh ZJqmJwSdcBXQ8EQ25bakapXsJdF9Ny2IPe+9Hz/LTJEzC+VYXd8yvuQw3dfHLmGw kgNH0I8mV/fqIxezMhFIOfT5GubP9BNZJL1Kb3iKr1eh9pNpGEDM02xJI02vIElm GzA8KlEBcS5fRvtvInU5vKdBHQy1uVlYHgT5lKArcYCI7wFCkD+sP7NBw1ngHkmT BNmYX1jtM3jEaI2otTOL07aUoQlZsA== =AKNQ -----END PGP SIGNATURE----- --=-exV1eg95rhXTf/heN52G--