linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: Will Deacon <will.deacon@arm.com>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Robin Murphy <robin.murphy@arm.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: linux-arm-kernel@lists.infradead.org,
	iommu@lists.linux-foundation.org, linux-kernel@vger.kernel.org
Subject: [PATCH 09/10] swiotlb: add support for non-coherent DMA
Date: Mon,  8 Oct 2018 10:02:45 +0200	[thread overview]
Message-ID: <20181008080246.20543-10-hch@lst.de> (raw)
In-Reply-To: <20181008080246.20543-1-hch@lst.de>

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 <hch@lst.de>
---
 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 <linux/cache.h>
 #include <linux/dma-direct.h>
+#include <linux/dma-noncoherent.h>
 #include <linux/mm.h>
 #include <linux/export.h>
 #include <linux/spinlock.h>
@@ -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)
+		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);
 }
 
 void
-- 
2.19.0


  parent reply	other threads:[~2018-10-08  8:03 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-10-08  8:02 move swiotlb noncoherent dma support from arm64 to generic code V2 Christoph Hellwig
2018-10-08  8:02 ` [PATCH 01/10] swiotlb: remove a pointless comment Christoph Hellwig
2018-10-11 17:49   ` Robin Murphy
2018-10-19  0:09   ` Konrad Rzeszutek Wilk
2018-10-08  8:02 ` [PATCH 02/10] swiotlb: mark is_swiotlb_buffer static Christoph Hellwig
2018-10-11 17:54   ` Robin Murphy
2018-10-19  0:12   ` Konrad Rzeszutek Wilk
2018-10-08  8:02 ` [PATCH 03/10] swiotlb: do not panic on mapping failures Christoph Hellwig
2018-10-11 18:06   ` Robin Murphy
2018-10-19  0:18     ` Konrad Rzeszutek Wilk
2018-10-19  0:17   ` Konrad Rzeszutek Wilk
2018-10-19  6:04     ` Christoph Hellwig
2018-10-19 13:45       ` Konrad Rzeszutek Wilk
2018-10-08  8:02 ` [PATCH 04/10] swiotlb: remove the overflow buffer Christoph Hellwig
2018-10-11 18:19   ` Robin Murphy
2018-10-12 17:04   ` Catalin Marinas
2018-10-19  0:23   ` Konrad Rzeszutek Wilk
2018-10-08  8:02 ` [PATCH 05/10] swiotlb: merge swiotlb_unmap_page and unmap_single Christoph Hellwig
2018-10-18 17:44   ` Robin Murphy
2018-10-19  0:25   ` Konrad Rzeszutek Wilk
2018-10-08  8:02 ` [PATCH 06/10] swiotlb: use swiotlb_map_page in swiotlb_map_sg_attrs Christoph Hellwig
2018-10-18 17:53   ` Robin Murphy
2018-10-19  0:33   ` Konrad Rzeszutek Wilk
2018-11-07  1:27   ` John Stultz
2018-11-09  7:49     ` Christoph Hellwig
2018-11-09 16:37       ` Robin Murphy
2018-11-19 19:36         ` Robin Murphy
2018-11-20  9:22           ` Christoph Hellwig
2018-11-13  0:07       ` John Stultz
2018-11-13  0:26         ` John Stultz
2018-11-14 14:13         ` Christoph Hellwig
2018-11-14 16:12           ` Christoph Hellwig
2018-11-19 23:22             ` John Stultz
2018-11-20  9:25               ` Christoph Hellwig
2018-11-23 18:27                 ` Will Deacon
2018-11-23 19:34                   ` Robin Murphy
2018-11-26 19:31                     ` Will Deacon
2018-10-08  8:02 ` [PATCH 07/10] swiotlb: refactor swiotlb_map_page Christoph Hellwig
2018-10-18 18:09   ` Robin Murphy
2018-10-19  0:37     ` Konrad Rzeszutek Wilk
2018-10-19  6:52       ` Christoph Hellwig
2018-10-19 13:46         ` Konrad Rzeszutek Wilk
2018-10-08  8:02 ` [PATCH 08/10] swiotlb: don't dip into swiotlb pool for coherent allocations Christoph Hellwig
2018-10-12 17:04   ` Catalin Marinas
2018-10-19  0:40   ` Konrad Rzeszutek Wilk
2018-10-19 16:45   ` Robin Murphy
2018-10-08  8:02 ` Christoph Hellwig [this message]
2018-10-19  0:49   ` [PATCH 09/10] swiotlb: add support for non-coherent DMA Konrad Rzeszutek Wilk
2018-10-22 17:11   ` Robin Murphy
2018-10-26  8:04     ` Christoph Hellwig
2018-10-26  9:59       ` Robin Murphy
2018-10-08  8:02 ` [PATCH 10/10] arm64: use the generic swiotlb_dma_ops Christoph Hellwig
2018-10-12 13:01   ` Robin Murphy
2018-10-12 14:40     ` Christoph Hellwig
2018-10-12 17:05       ` Catalin Marinas
2018-10-22 17:52   ` Robin Murphy
2018-10-26 12:44     ` Christoph Hellwig

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20181008080246.20543-10-hch@lst.de \
    --to=hch@lst.de \
    --cc=catalin.marinas@arm.com \
    --cc=iommu@lists.linux-foundation.org \
    --cc=konrad.wilk@oracle.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=robin.murphy@arm.com \
    --cc=will.deacon@arm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).