From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: Bug 205201 - Booting halts if Dawicontrol DC-2976 UW SCSI board installed, unless RAM size limited to 3500M Date: Wed, 27 Nov 2019 09:53:51 +0100 Message-ID: <20191127085351.GA24676@lst.de> References: <6eec5c42-019c-a988-fc2a-cb804194683d@xenosoft.de> <20191121180226.GA3852@lst.de> <2fde79cf-875f-94e6-4a1b-f73ebb2e2c32@xenosoft.de> <20191125073923.GA30168@lst.de> <4681f5fe-c095-15f5-9221-4b55e940bafc@xenosoft.de> <20191126164026.GA8026@lst.de> <20191127065624.GB16913@linux.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20191127065624.GB16913@linux.ibm.com> Sender: linux-kernel-owner@vger.kernel.org To: Mike Rapoport Cc: Christoph Hellwig , Christian Zigotzky , Robin Murphy , linux-arch@vger.kernel.org, darren@stevens-zone.net, mad skateman , Benjamin Herrenschmidt , linux-kernel@vger.kernel.org, linux-mm@kvack.org, iommu@lists.linux-foundation.org, Rob Herring , paulus@samba.org, rtd2@xtra.co.nz, "contact@a-eon.com" , linuxppc-dev , nsaenzjulienne@suse.de List-Id: linux-arch.vger.kernel.org On Wed, Nov 27, 2019 at 08:56:25AM +0200, Mike Rapoport wrote: > Maybe we'll simply force bottom up allocation before calling > swiotlb_init()? Anyway, it's the last memblock allocation. That should work, but I don't think it is the proper fix. The underlying issue here is that ZONE_DMA/DMA32 sizing is something that needs to be propagated to memblock and dma-direct as is based around addressing limitations. But our zone initialization is such a mess that we can't just reuse a variable. Nicolas has started to clean some of this up, but we need to clean that whole zone initialization mess up a lot more. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Wed, 27 Nov 2019 09:53:51 +0100 From: Christoph Hellwig Subject: Re: Bug 205201 - Booting halts if Dawicontrol DC-2976 UW SCSI board installed, unless RAM size limited to 3500M Message-ID: <20191127085351.GA24676@lst.de> References: <6eec5c42-019c-a988-fc2a-cb804194683d@xenosoft.de> <20191121180226.GA3852@lst.de> <2fde79cf-875f-94e6-4a1b-f73ebb2e2c32@xenosoft.de> <20191125073923.GA30168@lst.de> <4681f5fe-c095-15f5-9221-4b55e940bafc@xenosoft.de> <20191126164026.GA8026@lst.de> <20191127065624.GB16913@linux.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20191127065624.GB16913@linux.ibm.com> To: Mike Rapoport Cc: Christoph Hellwig , Christian Zigotzky , Robin Murphy , linux-arch@vger.kernel.org, darren@stevens-zone.net, mad skateman , Benjamin Herrenschmidt , linux-kernel@vger.kernel.org, linux-mm@kvack.org, iommu@lists.linux-foundation.org, Rob Herring , paulus@samba.org, rtd2@xtra.co.nz, "contact@a-eon.com" , linuxppc-dev , nsaenzjulienne@suse.de List-ID: Message-ID: <20191127085351.GfpmJvqcF34IqskwQ8qW5QTb6dFsZKpzVcfrbhDI-Ho@z> On Wed, Nov 27, 2019 at 08:56:25AM +0200, Mike Rapoport wrote: > Maybe we'll simply force bottom up allocation before calling > swiotlb_init()? Anyway, it's the last memblock allocation. That should work, but I don't think it is the proper fix. The underlying issue here is that ZONE_DMA/DMA32 sizing is something that needs to be propagated to memblock and dma-direct as is based around addressing limitations. But our zone initialization is such a mess that we can't just reuse a variable. Nicolas has started to clean some of this up, but we need to clean that whole zone initialization mess up a lot more.