From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932583AbdEDSwT (ORCPT ); Thu, 4 May 2017 14:52:19 -0400 Received: from mail-pg0-f42.google.com ([74.125.83.42]:33144 "EHLO mail-pg0-f42.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932073AbdEDSwL (ORCPT ); Thu, 4 May 2017 14:52:11 -0400 MIME-Version: 1.0 In-Reply-To: <0c9a6b09-139c-df49-cf73-6a3935a6aacd@arm.com> References: <1493786795-28153-1-git-send-email-oza.oza@broadcom.com> <1493786795-28153-2-git-send-email-oza.oza@broadcom.com> <0c9a6b09-139c-df49-cf73-6a3935a6aacd@arm.com> From: Oza Oza Date: Fri, 5 May 2017 00:22:10 +0530 Message-ID: Subject: Re: [PATCH 2/3] iommu/pci: reserve iova for PCI masters To: Robin Murphy Cc: Joerg Roedel , Linux IOMMU , linux-pci@vger.kernel.org, linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org, devicetree@vger.kernel.org, BCM Kernel Feedback , Oza Pawandeep Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, May 4, 2017 at 11:50 PM, Robin Murphy wrote: > On 03/05/17 05:46, Oza Pawandeep wrote: >> this patch reserves the iova for PCI masters. >> ARM64 based SOCs may have scattered memory banks. >> such as iproc based SOC has >> >> <0x00000000 0x80000000 0x0 0x80000000>, /* 2G @ 2G */ >> <0x00000008 0x80000000 0x3 0x80000000>, /* 14G @ 34G */ >> <0x00000090 0x00000000 0x4 0x00000000>, /* 16G @ 576G */ >> <0x000000a0 0x00000000 0x4 0x00000000>; /* 16G @ 640G */ >> >> but incoming PCI transcation addressing capability is limited >> by host bridge, for example if max incoming window capability >> is 512 GB, then 0x00000090 and 0x000000a0 will fall beyond it. >> >> to address this problem, iommu has to avoid allocating iovas which >> are reserved. which inturn does not allocate iova if it falls into hole. > > I don't necessarily disagree with doing this, as we could do with facing > up to the issue of discontiguous DMA ranges in particular (I too have a > platform with this problem), but I'm still not overly keen on pulling DT > specifics into this layer. More than that, though, if we are going to do > it, then we should do it for all devices with a restrictive > "dma-ranges", not just PCI ones. > How do you propose to do it ? my thinking is this: iova_reserve_pci_windows is written specific for PCI, and I am adding there. ideally struct pci_host_bridge should have new member: struct list_head inbound_windows; /* resource_entry */ but somehow this resource have to be filled much before iommu_dma_init_domain happens. and use brdge resource directly in iova_reserve_pci_windows as it is already doing it for outbound memory. this will detach the DT specifics from dma-immu layer. let me know how this sounds. >> Bug: SOC-5216 >> Change-Id: Icbfc99a045d730be143fef427098c937b9d46353 >> Signed-off-by: Oza Pawandeep >> Reviewed-on: http://gerrit-ccxsw.broadcom.net/40760 >> Reviewed-by: vpx_checkpatch status >> Reviewed-by: CCXSW >> Tested-by: vpx_autobuild status >> Tested-by: vpx_smoketest status >> Tested-by: CCXSW >> Reviewed-by: Scott Branden >> >> diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c >> index 48d36ce..08764b0 100644 >> --- a/drivers/iommu/dma-iommu.c >> +++ b/drivers/iommu/dma-iommu.c >> @@ -27,6 +27,7 @@ >> #include >> #include >> #include >> +#include >> #include >> #include >> #include >> @@ -171,8 +172,12 @@ static void iova_reserve_pci_windows(struct pci_dev *dev, >> struct iova_domain *iovad) >> { >> struct pci_host_bridge *bridge = pci_find_host_bridge(dev->bus); >> + struct device_node *np = bridge->dev.parent->of_node; >> struct resource_entry *window; >> unsigned long lo, hi; >> + int ret; >> + dma_addr_t tmp_dma_addr = 0, dma_addr; >> + LIST_HEAD(res); >> >> resource_list_for_each_entry(window, &bridge->windows) { >> if (resource_type(window->res) != IORESOURCE_MEM && >> @@ -183,6 +188,36 @@ static void iova_reserve_pci_windows(struct pci_dev *dev, >> hi = iova_pfn(iovad, window->res->end - window->offset); >> reserve_iova(iovad, lo, hi); >> } >> + >> + /* PCI inbound memory reservation. */ >> + ret = of_pci_get_dma_ranges(np, &res); >> + if (!ret) { >> + resource_list_for_each_entry(window, &res) { >> + struct resource *res_dma = window->res; >> + >> + dma_addr = res_dma->start - window->offset; >> + if (tmp_dma_addr > dma_addr) { >> + pr_warn("PCI: failed to reserve iovas; ranges should be sorted\n"); > > I don't see anything in the DT spec about the entries having to be > sorted, and it's not exactly impossible to sort a list if you need it so > (and if I'm being really pedantic, one could still trigger this with a > list that *is* sorted, only by different criteria). > we have to sort it the way we want then. I can make it sort then. thanks for the suggestion. > Robin. > >> + return; >> + } >> + if (tmp_dma_addr != dma_addr) { >> + lo = iova_pfn(iovad, tmp_dma_addr); >> + hi = iova_pfn(iovad, dma_addr - 1); >> + reserve_iova(iovad, lo, hi); >> + } >> + tmp_dma_addr = window->res->end - window->offset; >> + } >> + /* >> + * the last dma-range should honour based on the >> + * 32/64-bit dma addresses. >> + */ >> + if (tmp_dma_addr < DMA_BIT_MASK(sizeof(dma_addr_t) * 8)) { >> + lo = iova_pfn(iovad, tmp_dma_addr); >> + hi = iova_pfn(iovad, >> + DMA_BIT_MASK(sizeof(dma_addr_t) * 8) - 1); >> + reserve_iova(iovad, lo, hi); >> + } >> + } >> } >> >> /** >> >