linux-m68k.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Vladimir Murzin <vladimir.murzin@arm.com>
To: Christoph Hellwig <hch@lst.de>, iommu@lists.linux-foundation.org
Cc: linux-mips@linux-mips.org, linux-ia64@vger.kernel.org,
	linux-sh@vger.kernel.org, sparclinux@vger.kernel.org,
	Guan Xuetao <gxt@mprc.pku.edu.cn>,
	linux-arch@vger.kernel.org, linux-s390@vger.kernel.org,
	linux-c6x-dev@linux-c6x.org, linux-hexagon@vger.kernel.org,
	x86@kernel.org, Konrad Rzeszutek Wilk <konrad@darnok.org>,
	linux-snps-arc@lists.infradead.org,
	linux-m68k@lists.linux-m68k.org, patches@groups.riscv.org,
	linux-metag@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	Michal Simek <monstr@monstr.eu>,
	linux-parisc@vger.kernel.org, linux-cris-kernel@axis.com,
	linux-kernel@vger.kernel.org, linux-alpha@vger.kernel.org,
	linuxppc-dev@lists.ozlabs.org
Subject: Re: [PATCH 29/33] dma-direct: retry allocations using GFP_DMA for small masks
Date: Wed, 10 Jan 2018 09:32:16 +0000	[thread overview]
Message-ID: <93966313-cc79-8173-48e6-157caa443322__28874.4759549902$1515576641$gmane$org@arm.com> (raw)
In-Reply-To: <20180110080027.13879-30-hch@lst.de>

On 10/01/18 08:00, Christoph Hellwig wrote:
> If an attempt to allocate memory succeeded, but isn't inside the
> supported DMA mask, retry the allocation with GFP_DMA set as a
> last resort.
> 
> Based on the x86 code, but an off by one error in what is now
> dma_coherent_ok has been fixed vs the x86 code.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
>  lib/dma-direct.c | 25 ++++++++++++++++++++++++-
>  1 file changed, 24 insertions(+), 1 deletion(-)
> 
> diff --git a/lib/dma-direct.c b/lib/dma-direct.c
> index 8f76032ebc3c..4e43c2bb7f5f 100644
> --- a/lib/dma-direct.c
> +++ b/lib/dma-direct.c
> @@ -35,6 +35,11 @@ check_addr(struct device *dev, dma_addr_t dma_addr, size_t size,
>  	return true;
>  }
>  
> +static bool dma_coherent_ok(struct device *dev, phys_addr_t phys, size_t size)
> +{
> +	return phys_to_dma(dev, phys) + size - 1 <= dev->coherent_dma_mask;
> +}
> +
>  static void *dma_direct_alloc(struct device *dev, size_t size,
>  		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
>  {
> @@ -48,11 +53,29 @@ static void *dma_direct_alloc(struct device *dev, size_t size,
>  	if (dev->coherent_dma_mask <= DMA_BIT_MASK(32) && !(gfp & GFP_DMA))
>  		gfp |= GFP_DMA32;
>  
> +again:
>  	/* CMA can be used only in the context which permits sleeping */
> -	if (gfpflags_allow_blocking(gfp))
> +	if (gfpflags_allow_blocking(gfp)) {
>  		page = dma_alloc_from_contiguous(dev, count, page_order, gfp);
> +		if (page && !dma_coherent_ok(dev, page_to_phys(page), size)) {
> +			dma_release_from_contiguous(dev, page, count);
> +			page = NULL;
> +		}
> +	}
>  	if (!page)
>  		page = alloc_pages_node(dev_to_node(dev), gfp, page_order);
> +
> +	if (page && !dma_coherent_ok(dev, page_to_phys(page), size)) {
> +		__free_pages(page, page_order);
> +		page = NULL;
> +
> +		if (dev->coherent_dma_mask < DMA_BIT_MASK(32) &&
> +		    !(gfp & GFP_DMA)) {
> +			gfp = (gfp & ~GFP_DMA32) | GFP_DMA;
> +			goto again;
> +		}
> +	}
> +
>  	if (!page)
>  		return NULL;
>  
> 

Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com>

Cheers
Vladimir

  parent reply	other threads:[~2018-01-10  9:32 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20180110080027.13879-1-hch@lst.de>
2018-01-10  7:59 ` [PATCH 01/33] alpha: mark jensen as broken Christoph Hellwig
2018-01-10  7:59 ` [PATCH 02/33] hexagon: remove unused flush_write_buffers definition Christoph Hellwig
2018-01-10  7:59 ` [PATCH 03/33] m32r: " Christoph Hellwig
2018-01-10  7:59 ` [PATCH 04/33] powerpc: " Christoph Hellwig
2018-01-10  7:59 ` [PATCH 05/33] arc: remove CONFIG_ARC_PLAT_NEEDS_PHYS_TO_DMA Christoph Hellwig
2018-01-10  8:00 ` [PATCH 06/33] m32r: remove the unused dma_capable helper Christoph Hellwig
2018-01-10  8:00 ` [PATCH 07/33] riscv: " Christoph Hellwig
2018-01-10  8:00 ` [PATCH 08/33] s390: " Christoph Hellwig
2018-01-10  8:00 ` [PATCH 09/33] dma-mapping: take dma_pfn_offset into account in dma_max_pfn Christoph Hellwig
2018-01-10  8:00 ` [PATCH 10/33] arm64: don't override dma_max_pfn Christoph Hellwig
2018-01-10  8:00 ` [PATCH 11/33] dma-mapping: move swiotlb arch helpers to a new header Christoph Hellwig
2018-01-10  8:00 ` [PATCH 12/33] dma-mapping: move dma_mark_clean to dma-direct.h Christoph Hellwig
2018-01-10  8:00 ` [PATCH 13/33] hexagon: use the generic dma_capable helper Christoph Hellwig
2018-01-10  8:00 ` [PATCH 14/33] powerpc: rename dma_direct_ to dma_nommu_ Christoph Hellwig
2018-01-10  8:00 ` [PATCH 15/33] microblaze: rename dma_direct to dma_nommu Christoph Hellwig
2018-01-10  8:00 ` [PATCH 16/33] microblaze: remove dma_nommu_dma_supported Christoph Hellwig
2018-01-10  8:00 ` [PATCH 17/33] microblaze: remove the dead !NOT_COHERENT_CACHE dma code Christoph Hellwig
2018-01-10  8:00 ` [PATCH 18/33] s390: move s390_pci_dma_ops to asm/pci_dma.h Christoph Hellwig
2018-01-10  8:00 ` [PATCH 19/33] dma-mapping: warn when there is no coherent_dma_mask Christoph Hellwig
2018-01-10  8:00 ` [PATCH 20/33] dma-mapping: clear harmful GFP_* flags in common code Christoph Hellwig
2018-01-10  8:00 ` [PATCH 21/33] dma-mapping: add an arch_dma_supported hook Christoph Hellwig
2018-01-10  8:00 ` [PATCH 22/33] dma-mapping: provide a generic asm/dma-mapping.h Christoph Hellwig
2018-01-10  8:00 ` [PATCH 23/33] dma-direct: rename dma_noop to dma_direct Christoph Hellwig
2018-01-10  8:00 ` [PATCH 24/33] dma-direct: use phys_to_dma Christoph Hellwig
2018-01-10  8:00 ` [PATCH 25/33] dma-direct: add dma address sanity checks Christoph Hellwig
2018-01-10  8:00 ` [PATCH 26/33] dma-direct: add support for CMA allocation Christoph Hellwig
2018-01-10  8:00 ` [PATCH 27/33] dma-direct: use node local allocations for coherent memory Christoph Hellwig
2018-01-10  8:00 ` [PATCH 28/33] dma-direct: add support for allocation from ZONE_DMA and ZONE_DMA32 Christoph Hellwig
2018-01-10  8:00 ` [PATCH 29/33] dma-direct: retry allocations using GFP_DMA for small masks Christoph Hellwig
2018-01-10  8:00 ` [PATCH 30/33] dma-direct: make dma_direct_{alloc,free} available to other implementations Christoph Hellwig
2018-01-10  8:00 ` [PATCH 31/33] dma-direct: reject too small dma masks Christoph Hellwig
2018-01-10  8:00 ` [PATCH 32/33] cris: use dma-direct Christoph Hellwig
2018-01-10  8:00 ` [PATCH 33/33] h8300: " Christoph Hellwig
     [not found] ` <20180110080027.13879-12-hch@lst.de>
2018-01-10  9:31   ` [PATCH 11/33] dma-mapping: move swiotlb arch helpers to a new header Vladimir Murzin
2018-01-10 14:56   ` Robin Murphy
     [not found]   ` <b2bd6f4b-a932-5251-517b-83bbccfe7c53@arm.com>
2018-01-10 15:22     ` Christoph Hellwig
     [not found]   ` <3721b4ba-0685-255e-06b9-6e60678a1a92@arm.com>
2018-01-10 15:26     ` Christoph Hellwig
2018-01-10 15:31       ` Robin Murphy
     [not found]       ` <b0f90ed0-989f-4dc2-6f86-de8b4e486b55@arm.com>
2018-01-10 15:40         ` Christoph Hellwig
     [not found] ` <20180110080027.13879-30-hch@lst.de>
2018-01-10  9:32   ` Vladimir Murzin [this message]
     [not found] ` <20180110080027.13879-10-hch@lst.de>
2018-01-10 11:07   ` [PATCH 09/33] dma-mapping: take dma_pfn_offset into account in dma_max_pfn Robin Murphy
     [not found] ` <20180110080027.13879-11-hch@lst.de>
2018-01-10 11:08   ` [PATCH 10/33] arm64: don't override dma_max_pfn Robin Murphy
     [not found] ` <20180110080027.13879-32-hch@lst.de>
2018-01-10 11:49   ` [PATCH 31/33] dma-direct: reject too small dma masks Robin Murphy
2018-01-10 15:32     ` Christoph Hellwig
2018-01-10 17:00       ` Robin Murphy
     [not found] ` <20180110080027.13879-21-hch@lst.de>
2018-01-10 11:59   ` [PATCH 20/33] dma-mapping: clear harmful GFP_* flags in common code Robin Murphy
     [not found]   ` <27b90341-f9d0-356f-0194-1c7203a3f93e@arm.com>
2018-01-10 15:29     ` Christoph Hellwig
     [not found] ` <20180110080027.13879-28-hch@lst.de>
2018-01-10 12:06   ` [PATCH 27/33] dma-direct: use node local allocations for coherent memory Robin Murphy
2018-01-10 15:30     ` Christoph Hellwig
     [not found]     ` <20180110153017.GD17790@lst.de>
2018-01-10 16:49       ` Robin Murphy
     [not found] ` <20180110080027.13879-20-hch@lst.de>
2018-01-12 13:16   ` [PATCH 19/33] dma-mapping: warn when there is no coherent_dma_mask Konrad Rzeszutek Wilk
     [not found] ` <20180110080027.13879-22-hch@lst.de>
2018-01-12 13:17   ` [PATCH 21/33] dma-mapping: add an arch_dma_supported hook Konrad Rzeszutek Wilk

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='93966313-cc79-8173-48e6-157caa443322__28874.4759549902$1515576641$gmane$org@arm.com' \
    --to=vladimir.murzin@arm.com \
    --cc=gxt@mprc.pku.edu.cn \
    --cc=hch@lst.de \
    --cc=iommu@lists.linux-foundation.org \
    --cc=konrad@darnok.org \
    --cc=linux-alpha@vger.kernel.org \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-c6x-dev@linux-c6x.org \
    --cc=linux-cris-kernel@axis.com \
    --cc=linux-hexagon@vger.kernel.org \
    --cc=linux-ia64@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-m68k@lists.linux-m68k.org \
    --cc=linux-metag@vger.kernel.org \
    --cc=linux-mips@linux-mips.org \
    --cc=linux-parisc@vger.kernel.org \
    --cc=linux-s390@vger.kernel.org \
    --cc=linux-sh@vger.kernel.org \
    --cc=linux-snps-arc@lists.infradead.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=monstr@monstr.eu \
    --cc=patches@groups.riscv.org \
    --cc=sparclinux@vger.kernel.org \
    --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).