From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Christoph Hellwig <hch@lst.de>, iommu@lists.linux-foundation.org
Cc: Marek Szyprowski <m.szyprowski@samsung.com>,
Robin Murphy <robin.murphy@arm.com>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
linux-kernel@vger.kernel.org, linuxppc-dev@lists.ozlabs.org
Subject: Re: [PATCH 5/5] dma-direct: always allow dma mask <= physiscal memory size
Date: Thu, 27 Sep 2018 11:50:14 +1000 [thread overview]
Message-ID: <985079efc3296ac45de1e6344d7916c42fc7cbdd.camel@kernel.crashing.org> (raw)
In-Reply-To: <20180920185247.20037-6-hch@lst.de>
On Thu, 2018-09-20 at 20:52 +0200, Christoph Hellwig wrote:
> This way an architecture with less than 4G of RAM can support dma_mask
> smaller than 32-bit without a ZONE_DMA. Apparently that is a common
> case on powerpc.
Anything that uses a b43 wifi adapter which has a 31-bit limitation
actually :-)
>
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
> kernel/dma/direct.c | 28 ++++++++++++++++------------
> 1 file changed, 16 insertions(+), 12 deletions(-)
>
> diff --git a/kernel/dma/direct.c b/kernel/dma/direct.c
> index 64466b7ef67b..d1e103c6b107 100644
> --- a/kernel/dma/direct.c
> +++ b/kernel/dma/direct.c
> @@ -4,6 +4,7 @@
> *
> * DMA operations that map physical memory directly without using an IOMMU.
> */
> +#include <linux/bootmem.h> /* for max_pfn */
> #include <linux/export.h>
> #include <linux/mm.h>
> #include <linux/dma-direct.h>
> @@ -283,21 +284,24 @@ int dma_direct_map_sg(struct device *dev, struct scatterlist *sgl, int nents,
> return nents;
> }
>
> +/*
> + * Because 32-bit DMA masks are so common we expect every architecture to be
> + * able to satisfy them - either by not supporting more physical memory, or by
> + * providing a ZONE_DMA32. If neither is the case, the architecture needs to
> + * use an IOMMU instead of the direct mapping.
> + */
> int dma_direct_supported(struct device *dev, u64 mask)
> {
> -#ifdef CONFIG_ZONE_DMA
> - if (mask < phys_to_dma(dev, DMA_BIT_MASK(ARCH_ZONE_DMA_BITS)))
> - return 0;
> -#else
> - /*
> - * Because 32-bit DMA masks are so common we expect every architecture
> - * to be able to satisfy them - either by not supporting more physical
> - * memory, or by providing a ZONE_DMA32. If neither is the case, the
> - * architecture needs to use an IOMMU instead of the direct mapping.
> - */
> - if (mask < phys_to_dma(dev, DMA_BIT_MASK(32)))
> + u64 min_mask;
> +
> + if (IS_ENABLED(CONFIG_ZONE_DMA))
> + min_mask = DMA_BIT_MASK(ARCH_ZONE_DMA_BITS);
> + else
> + min_mask = min_t(u64, DMA_BIT_MASK(32),
> + (max_pfn - 1) << PAGE_SHIFT);
> +
> + if (mask >= phys_to_dma(dev, min_mask))
> return 0;
nitpick ... to be completely "correct", I would have written
if (IS_ENABLED(CONFIG_ZONE_DMA))
min_mask = DMA_BIT_MASK(ARCH_ZONE_DMA_BITS);
else
min_mask = DMA_BIT_MASK(32);
min_mask = min_t(u64, min_mask, (max_pfn - 1) << PAGE_SHIFT);
In "theory" it's also ok to have a mask < ZONE_DMA_BITS as long as it's
big enough to fit all memory :-)
> -#endif
> return 1;
> }
>
next prev parent reply other threads:[~2018-09-27 1:50 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-09-20 18:52 dma mask related fixups (including full bus_dma_mask support) Christoph Hellwig
2018-09-20 18:52 ` [PATCH 1/5] dma-mapping: make the get_required_mask method available unconditionally Christoph Hellwig
2018-09-27 1:28 ` Benjamin Herrenschmidt
2018-09-20 18:52 ` [PATCH 2/5] dma-direct: add an explicit dma_direct_get_required_mask Christoph Hellwig
2018-09-27 1:31 ` Benjamin Herrenschmidt
2018-09-27 14:12 ` Robin Murphy
2018-09-27 15:28 ` Christoph Hellwig
2018-09-27 15:35 ` Robin Murphy
2018-09-20 18:52 ` [PATCH 3/5] dma-direct: refine dma_direct_alloc zone selection Christoph Hellwig
2018-09-27 1:45 ` Benjamin Herrenschmidt
2018-09-27 13:49 ` Christoph Hellwig
2018-09-28 0:06 ` Benjamin Herrenschmidt
2018-09-28 15:46 ` Christoph Hellwig
2018-09-27 14:30 ` Robin Murphy
2018-09-27 15:30 ` Christoph Hellwig
2018-09-27 15:38 ` Robin Murphy
2018-09-27 15:41 ` Christoph Hellwig
2018-09-20 18:52 ` [PATCH 4/5] dma-direct: implement complete bus_dma_mask handling Christoph Hellwig
2018-09-27 14:58 ` Robin Murphy
2018-09-27 15:32 ` Christoph Hellwig
2018-09-27 16:14 ` Robin Murphy
2018-09-27 16:27 ` Christoph Hellwig
2018-09-27 16:41 ` Robin Murphy
2018-09-20 18:52 ` [PATCH 5/5] dma-direct: always allow dma mask <= physiscal memory size Christoph Hellwig
2018-09-27 1:50 ` Benjamin Herrenschmidt [this message]
2018-09-27 13:49 ` Christoph Hellwig
2018-09-27 15:07 ` Robin Murphy
2018-09-27 22:35 dma mask related fixups (including full bus_dma_mask support) v2 Christoph Hellwig
2018-09-27 22:35 ` [PATCH 5/5] dma-direct: always allow dma mask <= physiscal memory size Christoph Hellwig
2018-10-03 23:10 ` Alexander Duyck
2018-10-09 5:01 ` Benjamin Herrenschmidt
2018-11-19 14:18 ` Ramon Fried
2018-11-19 15:50 ` Robin Murphy
2018-11-20 7:38 ` Ramon Fried
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=985079efc3296ac45de1e6344d7916c42fc7cbdd.camel@kernel.crashing.org \
--to=benh@kernel.crashing.org \
--cc=gregkh@linuxfoundation.org \
--cc=hch@lst.de \
--cc=iommu@lists.linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=m.szyprowski@samsung.com \
--cc=robin.murphy@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).