Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754508AbdIZRXE (ORCPT ); Tue, 26 Sep 2017 13:23:04 -0400 Received: from mga07.intel.com ([134.134.136.100]:16299 "EHLO mga07.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751485AbdIZRXD (ORCPT ); Tue, 26 Sep 2017 13:23:03 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.42,441,1500966000"; d="scan'208";a="1199250890" Date: Tue, 26 Sep 2017 07:34:41 -0700 From: "Raj, Ashok" To: Robin Murphy Cc: Harsh Jain , Casey Leedom , Herbert Xuy , David Woodhouse , "linux-kernel@vger.kernel.org" , "iommu@lists.linux-foundation.org" , "linux-crypto@vger.kernel.org" , Dan Williams , "dwmw2@infradead.org" , Ashok Raj Subject: Re: DMA error when sg->offset value is greater than PAGE_SIZE in Intel IOMMU Message-ID: <20170926143441.GA136940@otc-nc-03> References: <20170920080151.GA3348@gondor.apana.org.au> <26992a1e-edb3-ed78-ce8e-31e0739d75f4@arm.com> <20170925155430.GB131920@otc-nc-03> <6d2af675-7b97-6eaf-4daa-d7bf80a05923@chelsio.com> <437a9bd8-d4d6-22ca-1a64-1a3e73f1101a@arm.com> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="KsGdsel6WgEHnImy" Content-Disposition: inline In-Reply-To: <437a9bd8-d4d6-22ca-1a64-1a3e73f1101a@arm.com> User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2396 Lines: 73 --KsGdsel6WgEHnImy Content-Type: text/plain; charset=us-ascii Content-Disposition: inline On Tue, Sep 26, 2017 at 03:22:47PM +0100, Robin Murphy wrote: > diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c > index 6784a05dd6b2..d7f7def81613 100644 > --- a/drivers/iommu/intel-iommu.c > +++ b/drivers/iommu/intel-iommu.c > @@ -2254,10 +2254,12 @@ static int __domain_mapping(struct dmar_domain *domain, unsigned long iov_pfn, > uint64_t tmp; > > if (!sg_res) { > + size_t off = sg->offset & ~PAGE_MASK; Should this be VTD_PAGE_MASK? > + > sg_res = aligned_nrpages(sg->offset, sg->length); > - sg->dma_address = ((dma_addr_t)iov_pfn << VTD_PAGE_SHIFT) + sg->offset; > + sg->dma_address = ((dma_addr_t)iov_pfn << VTD_PAGE_SHIFT) + off; > sg->dma_length = sg->length; > - pteval = page_to_phys(sg_page(sg)) | prot; > + pteval = (page_to_phys(sg_page(sg)) + sg->offset - off) | prot; Something seems wrong here.. sg->offset can be > VTD_PAGE_SIZE, think we should add sg->offset and then find the pteval? attached below another cut at fixing the same problem.. if there is something obvious i missed, let me know. again.. untested :-) Cheers, Ashok --KsGdsel6WgEHnImy Content-Type: text/plain; charset=us-ascii Content-Disposition: attachment; filename=fix-vtd-offset-gt4k Sometimes offset can be greater than 4K. vt-d needs to account for that. From: Ashok Raj Signed-off-by: Ashok Raj --- drivers/iommu/intel-iommu.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 6784a05..d43b566 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -2254,10 +2254,13 @@ static int __domain_mapping(struct dmar_domain *domain, unsigned long iov_pfn, uint64_t tmp; if (!sg_res) { + size_t off = sg->offset & ~VTD_PAGE_SHIFT; sg_res = aligned_nrpages(sg->offset, sg->length); - sg->dma_address = ((dma_addr_t)iov_pfn << VTD_PAGE_SHIFT) + sg->offset; + sg->dma_address = ((dma_addr_t) + (iov_pfn + sg->offset) << VTD_PAGE_SHIFT) + off; sg->dma_length = sg->length; - pteval = page_to_phys(sg_page(sg)) | prot; + pteval = (page_to_phys(sg_page(sg)) + + (sg->offset << VTD_PAGE_SHIFT)) | prot; phys_pfn = pteval >> VTD_PAGE_SHIFT; } --KsGdsel6WgEHnImy--