From: Christoph Hellwig <hch@lst.de>
To: Andrew Morton <akpm@linux-foundation.org>,
Haavard Skinnemoen <hskinnemoen@gmail.com>,
Hans-Christian Egtvedt <egtvedt@samfundet.no>,
Steven Miao <realmz6@gmail.com>, Ley Foon Tan <lftan@altera.com>,
David Howells <dhowells@redhat.com>,
Koichi Yasutake <yasutake.koichi@jp.panasonic.com>,
Chris Metcalf <cmetcalf@ezchip.com>,
linux-snps-arc@lists.infraded.org, linux-c6x-dev@linux-c6x.org,
linux-cris-kernel@axis.com, linux-parisc@vger.kernel.org,
linux-m68k@lists.linux-m68k.org, linux-metag@vger.kernel.org,
sparclinux@vger.kernel.org, iommu@lists.linux-foundation.org
Subject: [PATCH 14/16] tile: uninline dma_set_mask
Date: Mon, 9 Nov 2015 19:17:19 +0100 [thread overview]
Message-ID: <1447093041-21832-15-git-send-email-hch@lst.de> (raw)
In-Reply-To: <1447093041-21832-1-git-send-email-hch@lst.de>
We'll soon merge <asm-generic/dma-mapping-common.h> into <linux/dma-mapping.h>
and the reference to dma_capable in the tile dma_set_mask would create a
circular dependency. Fix this by moving the implementation out of line.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/tile/include/asm/dma-mapping.h | 29 +----------------------------
arch/tile/kernel/pci-dma.c | 29 +++++++++++++++++++++++++++++
2 files changed, 30 insertions(+), 28 deletions(-)
diff --git a/arch/tile/include/asm/dma-mapping.h b/arch/tile/include/asm/dma-mapping.h
index 96ac6cc..c342736 100644
--- a/arch/tile/include/asm/dma-mapping.h
+++ b/arch/tile/include/asm/dma-mapping.h
@@ -76,34 +76,7 @@ static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
#include <asm-generic/dma-mapping-common.h>
-static inline int
-dma_set_mask(struct device *dev, u64 mask)
-{
- struct dma_map_ops *dma_ops = get_dma_ops(dev);
-
- /*
- * For PCI devices with 64-bit DMA addressing capability, promote
- * the dma_ops to hybrid, with the consistent memory DMA space limited
- * to 32-bit. For 32-bit capable devices, limit the streaming DMA
- * address range to max_direct_dma_addr.
- */
- if (dma_ops == gx_pci_dma_map_ops ||
- dma_ops == gx_hybrid_pci_dma_map_ops ||
- dma_ops == gx_legacy_pci_dma_map_ops) {
- if (mask == DMA_BIT_MASK(64) &&
- dma_ops == gx_legacy_pci_dma_map_ops)
- set_dma_ops(dev, gx_hybrid_pci_dma_map_ops);
- else if (mask > dev->archdata.max_direct_dma_addr)
- mask = dev->archdata.max_direct_dma_addr;
- }
-
- if (!dev->dma_mask || !dma_supported(dev, mask))
- return -EIO;
-
- *dev->dma_mask = mask;
-
- return 0;
-}
+int dma_set_mask(struct device *dev, u64 mask);
/*
* dma_alloc_noncoherent() is #defined to return coherent memory,
diff --git a/arch/tile/kernel/pci-dma.c b/arch/tile/kernel/pci-dma.c
index 09b5870..b6bc054 100644
--- a/arch/tile/kernel/pci-dma.c
+++ b/arch/tile/kernel/pci-dma.c
@@ -583,6 +583,35 @@ struct dma_map_ops *gx_hybrid_pci_dma_map_ops;
EXPORT_SYMBOL(gx_legacy_pci_dma_map_ops);
EXPORT_SYMBOL(gx_hybrid_pci_dma_map_ops);
+int dma_set_mask(struct device *dev, u64 mask)
+{
+ struct dma_map_ops *dma_ops = get_dma_ops(dev);
+
+ /*
+ * For PCI devices with 64-bit DMA addressing capability, promote
+ * the dma_ops to hybrid, with the consistent memory DMA space limited
+ * to 32-bit. For 32-bit capable devices, limit the streaming DMA
+ * address range to max_direct_dma_addr.
+ */
+ if (dma_ops == gx_pci_dma_map_ops ||
+ dma_ops == gx_hybrid_pci_dma_map_ops ||
+ dma_ops == gx_legacy_pci_dma_map_ops) {
+ if (mask == DMA_BIT_MASK(64) &&
+ dma_ops == gx_legacy_pci_dma_map_ops)
+ set_dma_ops(dev, gx_hybrid_pci_dma_map_ops);
+ else if (mask > dev->archdata.max_direct_dma_addr)
+ mask = dev->archdata.max_direct_dma_addr;
+ }
+
+ if (!dev->dma_mask || !dma_supported(dev, mask))
+ return -EIO;
+
+ *dev->dma_mask = mask;
+
+ return 0;
+}
+EXPORT_SYMBOL(dma_set_mask);
+
#ifdef CONFIG_ARCH_HAS_DMA_SET_COHERENT_MASK
int dma_set_coherent_mask(struct device *dev, u64 mask)
{
--
1.9.1
next prev parent reply other threads:[~2015-11-09 18:17 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-11-09 18:17 use dma_map_ops for all architectures Christoph Hellwig
2015-11-09 18:17 ` [PATCH 03/16] avr32: convert to dma_map_ops Christoph Hellwig
2015-11-09 18:17 ` [PATCH 05/16] c6x: " Christoph Hellwig
2015-11-09 18:17 ` [PATCH 06/16] cris: " Christoph Hellwig
2015-11-11 15:07 ` Jesper Nilsson
2015-11-09 18:17 ` [PATCH 07/16] nios2: " Christoph Hellwig
2015-11-09 18:17 ` [PATCH 08/16] frv: " Christoph Hellwig
2015-11-09 18:17 ` [PATCH 09/16] parisc: " Christoph Hellwig
2015-11-09 19:32 ` Helge Deller
2015-11-09 18:17 ` [PATCH 10/16] mn10300: " Christoph Hellwig
2015-11-09 18:17 ` [PATCH 11/16] m68k: " Christoph Hellwig
2015-11-09 18:17 ` [PATCH 12/16] metag: " Christoph Hellwig
2015-11-09 18:17 ` [PATCH 13/16] sparc: use generic dma_set_mask Christoph Hellwig
2015-11-09 18:17 ` Christoph Hellwig [this message]
[not found] ` <1447093041-21832-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
2015-11-09 18:17 ` [PATCH 01/16] dma-mapping: make the generic coherent dma mmap implementation optional Christoph Hellwig
2015-11-09 18:17 ` [PATCH 02/16] arc: convert to dma_map_ops Christoph Hellwig
2015-11-09 18:17 ` [PATCH 04/16] blackfin: " Christoph Hellwig
2015-11-09 18:17 ` [PATCH 15/16] dma-mapping: always provide the dma_map_ops based implementation Christoph Hellwig
2015-11-09 18:17 ` [PATCH 16/16] dma-mapping: remove <asm-generic/dma-coherent.h> 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=1447093041-21832-15-git-send-email-hch@lst.de \
--to=hch@lst.de \
--cc=akpm@linux-foundation.org \
--cc=cmetcalf@ezchip.com \
--cc=dhowells@redhat.com \
--cc=egtvedt@samfundet.no \
--cc=hskinnemoen@gmail.com \
--cc=iommu@lists.linux-foundation.org \
--cc=lftan@altera.com \
--cc=linux-c6x-dev@linux-c6x.org \
--cc=linux-cris-kernel@axis.com \
--cc=linux-m68k@lists.linux-m68k.org \
--cc=linux-metag@vger.kernel.org \
--cc=linux-parisc@vger.kernel.org \
--cc=linux-snps-arc@lists.infraded.org \
--cc=realmz6@gmail.com \
--cc=sparclinux@vger.kernel.org \
--cc=yasutake.koichi@jp.panasonic.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).