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=-6.8 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_PASS autolearn=ham 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 C58FEC004D3 for ; Mon, 22 Oct 2018 17:11:10 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 4B1FA20663 for ; Mon, 22 Oct 2018 17:11:11 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 4B1FA20663 Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=arm.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728663AbeJWBa2 (ORCPT ); Mon, 22 Oct 2018 21:30:28 -0400 Received: from usa-sjc-mx-foss1.foss.arm.com ([217.140.101.70]:48744 "EHLO foss.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727218AbeJWBa2 (ORCPT ); Mon, 22 Oct 2018 21:30:28 -0400 Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.72.51.249]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id A835F80D; Mon, 22 Oct 2018 10:11:07 -0700 (PDT) Received: from [10.1.196.75] (e110467-lin.cambridge.arm.com [10.1.196.75]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 69C983F5D3; Mon, 22 Oct 2018 10:11:06 -0700 (PDT) Subject: Re: [PATCH 09/10] swiotlb: add support for non-coherent DMA To: Christoph Hellwig , Will Deacon , Catalin Marinas , Konrad Rzeszutek Wilk Cc: linux-arm-kernel@lists.infradead.org, iommu@lists.linux-foundation.org, linux-kernel@vger.kernel.org References: <20181008080246.20543-1-hch@lst.de> <20181008080246.20543-10-hch@lst.de> From: Robin Murphy Message-ID: <9ea28547-dd1a-b4bd-2e91-f71e70417e7d@arm.com> Date: Mon, 22 Oct 2018 18:11:04 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.2.1 MIME-Version: 1.0 In-Reply-To: <20181008080246.20543-10-hch@lst.de> Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-GB Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 08/10/2018 09:02, Christoph Hellwig wrote: > Handle architectures that are not cache coherent directly in the main > swiotlb code by calling arch_sync_dma_for_{device,cpu} in all the right > places from the various dma_map/unmap/sync methods when the device is > non-coherent. > > Because swiotlb now uses dma_direct_alloc for the coherent allocation > that side is already taken care of by the dma-direct code calling into > arch_dma_{alloc,free} for devices that are non-coherent. > > Signed-off-by: Christoph Hellwig > --- > kernel/dma/swiotlb.c | 23 +++++++++++++++++------ > 1 file changed, 17 insertions(+), 6 deletions(-) > > diff --git a/kernel/dma/swiotlb.c b/kernel/dma/swiotlb.c > index 475a41eff3dc..52885b274368 100644 > --- a/kernel/dma/swiotlb.c > +++ b/kernel/dma/swiotlb.c > @@ -21,6 +21,7 @@ > > #include > #include > +#include > #include > #include > #include > @@ -677,6 +678,10 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page, > dma_addr = swiotlb_bounce_page(dev, &phys, size, dir, attrs); > } > > + if (!dev_is_dma_coherent(dev) && > + (attrs & DMA_ATTR_SKIP_CPU_SYNC) == 0) Nit: other parts of the file are already using the "!(...)" style rather than "(...) == 0". > + arch_sync_dma_for_device(dev, phys, size, dir); > + > return dma_addr; > } > > @@ -696,6 +701,10 @@ void swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr, > > BUG_ON(dir == DMA_NONE); > > + if (!dev_is_dma_coherent(hwdev) && > + (attrs & DMA_ATTR_SKIP_CPU_SYNC) == 0) > + arch_sync_dma_for_cpu(hwdev, paddr, size, dir); > + > if (is_swiotlb_buffer(paddr)) { > swiotlb_tbl_unmap_single(hwdev, paddr, size, dir, attrs); > return; > @@ -732,15 +741,17 @@ swiotlb_sync_single(struct device *hwdev, dma_addr_t dev_addr, > > BUG_ON(dir == DMA_NONE); > > - if (is_swiotlb_buffer(paddr)) { > + if (!dev_is_dma_coherent(hwdev) && target == SYNC_FOR_CPU) > + arch_sync_dma_for_cpu(hwdev, paddr, size, dir); > + > + if (is_swiotlb_buffer(paddr)) > swiotlb_tbl_sync_single(hwdev, paddr, size, dir, target); > - return; > - } > > - if (dir != DMA_FROM_DEVICE) > - return; > + if (!dev_is_dma_coherent(hwdev) && target == SYNC_FOR_DEVICE) > + arch_sync_dma_for_device(hwdev, paddr, size, dir); > > - dma_mark_clean(phys_to_virt(paddr), size); > + if (!is_swiotlb_buffer(paddr) && dir == DMA_FROM_DEVICE) > + dma_mark_clean(phys_to_virt(paddr), size); > } All these "if"s end up pretty hard to follow at first glance :( I had a quick play at moving the cache maintenance here out into the callers, which comes out arguably looking perhaps a little cleaner (only +1 source line overall, and actually reduces text size by 32 bytes for my build), but sadly I can't really see any way of doing the equivalent for map/unmap short of duplicating the whole 3-line arch_sync_*() block, which just makes for a different readability problem. As you mentioned on patch #7, I guess this really is just one of those things which has no nice solution, so cosmetics aside, Reviewed-by: Robin Murphy FWIW, below is my "cleanup" attempt (diff on top of the swiotlb-noncoherent.3 branch). Robin. ----->8----- diff --git a/kernel/dma/swiotlb.c b/kernel/dma/swiotlb.c index 52885b274368..43ee29969fdd 100644 --- a/kernel/dma/swiotlb.c +++ b/kernel/dma/swiotlb.c @@ -741,23 +741,24 @@ swiotlb_sync_single(struct device *hwdev, dma_addr_t dev_addr, BUG_ON(dir == DMA_NONE); - if (!dev_is_dma_coherent(hwdev) && target == SYNC_FOR_CPU) - arch_sync_dma_for_cpu(hwdev, paddr, size, dir); - - if (is_swiotlb_buffer(paddr)) + if (is_swiotlb_buffer(paddr)) { swiotlb_tbl_sync_single(hwdev, paddr, size, dir, target); + return; + } - if (!dev_is_dma_coherent(hwdev) && target == SYNC_FOR_DEVICE) - arch_sync_dma_for_device(hwdev, paddr, size, dir); + if (dir != DMA_FROM_DEVICE) + return; - if (!is_swiotlb_buffer(paddr) && dir == DMA_FROM_DEVICE) - dma_mark_clean(phys_to_virt(paddr), size); + dma_mark_clean(phys_to_virt(paddr), size); } void swiotlb_sync_single_for_cpu(struct device *hwdev, dma_addr_t dev_addr, size_t size, enum dma_data_direction dir) { + if (!dev_is_dma_coherent(hwdev)) + arch_sync_dma_for_cpu(hwdev, dma_to_phys(hwdev, dev_addr), + size, dir); swiotlb_sync_single(hwdev, dev_addr, size, dir, SYNC_FOR_CPU); } @@ -766,6 +767,9 @@ swiotlb_sync_single_for_device(struct device *hwdev, dma_addr_t dev_addr, size_t size, enum dma_data_direction dir) { swiotlb_sync_single(hwdev, dev_addr, size, dir, SYNC_FOR_DEVICE); + if (!dev_is_dma_coherent(hwdev)) + arch_sync_dma_for_device(hwdev, dma_to_phys(hwdev, dev_addr), + size, dir); } /* @@ -828,31 +832,28 @@ swiotlb_unmap_sg_attrs(struct device *hwdev, struct scatterlist *sgl, * The same as swiotlb_sync_single_* but for a scatter-gather list, same rules * and usage. */ -static void -swiotlb_sync_sg(struct device *hwdev, struct scatterlist *sgl, - int nelems, enum dma_data_direction dir, - enum dma_sync_target target) +void +swiotlb_sync_sg_for_cpu(struct device *hwdev, struct scatterlist *sgl, + int nelems, enum dma_data_direction dir) { struct scatterlist *sg; int i; for_each_sg(sgl, sg, nelems, i) - swiotlb_sync_single(hwdev, sg->dma_address, - sg_dma_len(sg), dir, target); + swiotlb_sync_single_for_cpu(hwdev, sg->dma_address, + sg_dma_len(sg), dir); } void -swiotlb_sync_sg_for_cpu(struct device *hwdev, struct scatterlist *sg, - int nelems, enum dma_data_direction dir) -{ - swiotlb_sync_sg(hwdev, sg, nelems, dir, SYNC_FOR_CPU); -} - -void -swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg, +swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sgl, int nelems, enum dma_data_direction dir) { - swiotlb_sync_sg(hwdev, sg, nelems, dir, SYNC_FOR_DEVICE); + struct scatterlist *sg; + int i; + + for_each_sg(sgl, sg, nelems, i) + swiotlb_sync_single_for_device(hwdev, sg->dma_address, + sg_dma_len(sg), dir); } /*