linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: iommu@lists.linux-foundation.org
Cc: "Konrad Rzeszutek Wilk" <konrad@darnok.org>,
	"Michal Simek" <monstr@monstr.eu>,
	"Guan Xuetao" <gxt@mprc.pku.edu.cn>,
	"Christian König" <ckoenig.leichtzumerken@gmail.com>,
	linux-arm-kernel@lists.infradead.org, linux-ia64@vger.kernel.org,
	linux-mips@linux-mips.org, linuxppc-dev@lists.ozlabs.org,
	x86@kernel.org, linux-arch@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: [PATCH 07/22] swiotlb: add common swiotlb_map_ops
Date: Wed, 10 Jan 2018 09:09:17 +0100	[thread overview]
Message-ID: <20180110080932.14157-8-hch@lst.de> (raw)
In-Reply-To: <20180110080932.14157-1-hch@lst.de>

Currently all architectures that want to use swiotlb have to implement
their own dma_map_ops instances.  Provide a generic one based on the
x86 implementation which first calls into dma_direct to try a full blown
direct mapping implementation (including e.g. CMA) before falling back
allocating from the swiotlb buffer.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 include/linux/swiotlb.h |  8 ++++++++
 lib/swiotlb.c           | 43 +++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 51 insertions(+)

diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h
index 606375e35d87..5b1f2a00491c 100644
--- a/include/linux/swiotlb.h
+++ b/include/linux/swiotlb.h
@@ -66,6 +66,12 @@ extern void swiotlb_tbl_sync_single(struct device *hwdev,
 				    enum dma_sync_target target);
 
 /* Accessory functions. */
+
+void *swiotlb_alloc(struct device *hwdev, size_t size, dma_addr_t *dma_handle,
+		gfp_t flags, unsigned long attrs);
+void swiotlb_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_addr, unsigned long attrs);
+
 extern void
 *swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 			dma_addr_t *dma_handle, gfp_t flags);
@@ -126,4 +132,6 @@ extern void swiotlb_print_info(void);
 extern int is_swiotlb_buffer(phys_addr_t paddr);
 extern void swiotlb_set_max_segment(unsigned int);
 
+extern const struct dma_map_ops swiotlb_dma_ops;
+
 #endif /* __LINUX_SWIOTLB_H */
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index cf5311908fa9..0fae2f45c3c0 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -1087,3 +1087,46 @@ swiotlb_dma_supported(struct device *hwdev, u64 mask)
 	return swiotlb_phys_to_dma(hwdev, io_tlb_end - 1) <= mask;
 }
 EXPORT_SYMBOL(swiotlb_dma_supported);
+
+#ifdef CONFIG_DMA_DIRECT_OPS
+void *swiotlb_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
+		gfp_t gfp, unsigned long attrs)
+{
+	void *vaddr;
+
+	/*
+	 * Don't print a warning when the first allocation attempt fails.
+	 * swiotlb_alloc_coherent() will print a warning when the DMA memory
+	 * allocation ultimately failed.
+	 */
+	gfp |= __GFP_NOWARN;
+
+	vaddr = dma_direct_alloc(dev, size, dma_handle, gfp, attrs);
+	if (!vaddr)
+		vaddr = swiotlb_alloc_coherent(dev, size, dma_handle, gfp);
+	return vaddr;
+}
+
+void swiotlb_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_addr, unsigned long attrs)
+{
+	if (is_swiotlb_buffer(dma_to_phys(dev, dma_addr)))
+		swiotlb_free_coherent(dev, size, vaddr, dma_addr);
+	else
+		dma_direct_free(dev, size, vaddr, dma_addr, attrs);
+}
+
+const struct dma_map_ops swiotlb_dma_ops = {
+	.mapping_error		= swiotlb_dma_mapping_error,
+	.alloc			= swiotlb_alloc,
+	.free			= swiotlb_free,
+	.sync_single_for_cpu	= swiotlb_sync_single_for_cpu,
+	.sync_single_for_device	= swiotlb_sync_single_for_device,
+	.sync_sg_for_cpu	= swiotlb_sync_sg_for_cpu,
+	.sync_sg_for_device	= swiotlb_sync_sg_for_device,
+	.map_sg			= swiotlb_map_sg_attrs,
+	.unmap_sg		= swiotlb_unmap_sg_attrs,
+	.map_page		= swiotlb_map_page,
+	.unmap_page		= swiotlb_unmap_page,
+};
+#endif /* CONFIG_DMA_DIRECT_OPS */
-- 
2.14.2

  parent reply	other threads:[~2018-01-10  8:10 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-01-10  8:09 consolidate swiotlb dma_map implementations Christoph Hellwig
2018-01-10  8:09 ` [PATCH 01/22] swiotlb: suppress warning when __GFP_NOWARN is set Christoph Hellwig
2018-01-10  8:09 ` [PATCH 02/22] arm64: rename swiotlb_dma_ops Christoph Hellwig
2018-01-10 12:13   ` Robin Murphy
2018-01-10  8:09 ` [PATCH 03/22] ia64: " Christoph Hellwig
2018-01-12 13:24   ` Konrad Rzeszutek Wilk
2018-01-10  8:09 ` [PATCH 04/22] powerpc: " Christoph Hellwig
2018-01-12 13:25   ` Konrad Rzeszutek Wilk
2018-01-10  8:09 ` [PATCH 05/22] x86: " Christoph Hellwig
2018-01-12 13:25   ` Konrad Rzeszutek Wilk
2018-01-10  8:09 ` [PATCH 06/22] swiotlb: rename swiotlb_free to swiotlb_exit Christoph Hellwig
2018-01-12 13:39   ` Konrad Rzeszutek Wilk
2018-01-10  8:09 ` Christoph Hellwig [this message]
2018-01-10  8:09 ` [PATCH 08/22] swiotlb: wire up ->dma_supported in swiotlb_dma_ops Christoph Hellwig
2018-01-10 12:16   ` Robin Murphy
2018-01-10 15:35     ` Christoph Hellwig
2018-01-10 17:23       ` Robin Murphy
2018-01-10  8:09 ` [PATCH 09/22] swiotlb: refactor coherent buffer freeing Christoph Hellwig
2018-01-10  8:09 ` [PATCH 10/22] swiotlb: refactor coherent buffer allocation Christoph Hellwig
2018-01-10 12:22   ` Robin Murphy
2018-01-10 15:46     ` Christoph Hellwig
2018-01-10 17:02       ` Robin Murphy
2018-01-15  9:10         ` Christoph Hellwig
2018-01-10  8:09 ` [PATCH 11/22] swiotlb: remove various exports Christoph Hellwig
2018-01-10  8:09 ` [PATCH 12/22] ia64: replace ZONE_DMA with ZONE_DMA32 Christoph Hellwig
2018-01-10  8:09 ` [PATCH 13/22] ia64: use generic swiotlb_ops Christoph Hellwig
2018-01-10  8:09 ` [PATCH 14/22] ia64: clean up swiotlb support Christoph Hellwig
2018-01-10  8:09 ` [PATCH 15/22] ia64: remove an ifdef around the content of pci-dma.c Christoph Hellwig
2018-01-10  8:09 ` [PATCH 16/22] unicore32: use generic swiotlb_ops Christoph Hellwig
2018-01-10  8:09 ` [PATCH 17/22] tile: replace ZONE_DMA with ZONE_DMA32 Christoph Hellwig
2018-01-10  8:09 ` [PATCH 18/22] tile: use generic swiotlb_ops Christoph Hellwig
2018-01-10  8:09 ` [PATCH 19/22] mips/netlogic: remove swiotlb support Christoph Hellwig
2018-01-10  8:09 ` [PATCH 20/22] mips: use swiotlb_{alloc,free} Christoph Hellwig
2018-01-10  8:09 ` [PATCH 21/22] arm64: replace ZONE_DMA with ZONE_DMA32 Christoph Hellwig
2018-01-10 12:58   ` Robin Murphy
2018-01-10 15:55     ` Christoph Hellwig
2018-01-10 15:55       ` Christoph Hellwig
2018-01-10 17:10         ` Robin Murphy
2018-01-10  8:09 ` [PATCH 22/22] arm64: use swiotlb_alloc and swiotlb_free Christoph Hellwig
2018-01-10 13:16   ` Robin Murphy
2018-01-10  8:23 ` consolidate swiotlb dma_map implementations Christian König
2018-01-16  7:53 ` Christoph Hellwig
2018-01-16  8:22   ` Christian König
2018-01-16  8:28     ` Christoph Hellwig
2018-01-16  8:52       ` Christian König

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=20180110080932.14157-8-hch@lst.de \
    --to=hch@lst.de \
    --cc=ckoenig.leichtzumerken@gmail.com \
    --cc=gxt@mprc.pku.edu.cn \
    --cc=iommu@lists.linux-foundation.org \
    --cc=konrad@darnok.org \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-ia64@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mips@linux-mips.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=monstr@monstr.eu \
    --cc=x86@kernel.org \
    /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).