Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751218AbcKEXNf (ORCPT ); Sat, 5 Nov 2016 19:13:35 -0400 Received: from mail-it0-f68.google.com ([209.85.214.68]:33686 "EHLO mail-it0-f68.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750850AbcKEXNd (ORCPT ); Sat, 5 Nov 2016 19:13:33 -0400 MIME-Version: 1.0 In-Reply-To: <20161105193929.GA26349@localhost.localdomain> References: <20161102111031.79519.14741.stgit@ahduyck-blue-test.jf.intel.com> <20161102111252.79519.21950.stgit@ahduyck-blue-test.jf.intel.com> <20161105193929.GA26349@localhost.localdomain> From: Alexander Duyck Date: Sat, 5 Nov 2016 16:13:31 -0700 Message-ID: Subject: Re: [mm PATCH v2 03/26] swiotlb: Add support for DMA_ATTR_SKIP_CPU_SYNC To: Konrad Rzeszutek Wilk Cc: Alexander Duyck , linux-mm , Andrew Morton , Netdev , "linux-kernel@vger.kernel.org" , Konrad Rzeszutek Wilk Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2723 Lines: 62 On Sat, Nov 5, 2016 at 12:39 PM, Konrad Rzeszutek Wilk wrote: > .. snip.. >> @@ -561,6 +565,7 @@ void swiotlb_tbl_unmap_single(struct device *hwdev, phys_addr_t tlb_addr, >> * First, sync the memory before unmapping the entry >> */ >> if (orig_addr != INVALID_PHYS_ADDR && >> + !(attrs & DMA_ATTR_SKIP_CPU_SYNC) && >> ((dir == DMA_FROM_DEVICE) || (dir == DMA_BIDIRECTIONAL))) >> swiotlb_bounce(orig_addr, tlb_addr, size, DMA_FROM_DEVICE); >> >> @@ -654,7 +659,8 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr, >> * GFP_DMA memory; fall back on map_single(), which >> * will grab memory from the lowest available address range. >> */ >> - phys_addr_t paddr = map_single(hwdev, 0, size, DMA_FROM_DEVICE); >> + phys_addr_t paddr = map_single(hwdev, 0, size, >> + DMA_FROM_DEVICE, 0); >> if (paddr == SWIOTLB_MAP_ERROR) >> goto err_warn; >> >> @@ -669,7 +675,8 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr, >> >> /* DMA_TO_DEVICE to avoid memcpy in unmap_single */ >> swiotlb_tbl_unmap_single(hwdev, paddr, >> - size, DMA_TO_DEVICE); >> + size, DMA_TO_DEVICE, >> + DMA_ATTR_SKIP_CPU_SYNC); > > This I believe is redundant. That is swiotlb_tbl_unmap_single only > does an bounce if the dir is DMA_FROM_DEVICE or DMA_BIDIRECTIONAL. > > I added /* optional. */ You are probably right. I don't need to add the DMA_ATTR_SKIP_CPU_SYNC here. >> goto err_warn; >> } >> } >> @@ -699,7 +706,7 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr, >> free_pages((unsigned long)vaddr, get_order(size)); >> else >> /* DMA_TO_DEVICE to avoid memcpy in swiotlb_tbl_unmap_single */ >> - swiotlb_tbl_unmap_single(hwdev, paddr, size, DMA_TO_DEVICE); >> + swiotlb_tbl_unmap_single(hwdev, paddr, size, DMA_TO_DEVICE, 0); > > .. but here you choose to put 0? I changed that to > DMA_ATTR_SKIP_CPU_SYNC and expanded the comment above. > > Time to test the patches. I think I had probably realized the fact that I didn't need it above and so just used 0 here. I can clean this up and resubmit if you want. Do you want me to just split this patch set up so that I submit the swiotlb patches to you and leave the rest of the patches for the mm tree? Thanks. - Alex