LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Ira Weiny <ira.weiny@intel.com>
To: Nicolin Chen <nicoleotsuka@gmail.com>
Cc: hch@lst.de, robin.murphy@arm.com, m.szyprowski@samsung.com,
	vdumpa@nvidia.com, linux@armlinux.org.uk,
	catalin.marinas@arm.com, will.deacon@arm.com, chris@zankel.net,
	jcmvbkbc@gmail.com, joro@8bytes.org, dwmw2@infradead.org,
	tony@atomide.com, akpm@linux-foundation.org,
	sfr@canb.auug.org.au, treding@nvidia.com, keescook@chromium.org,
	iamjoonsoo.kim@lge.com, wsa+renesas@sang-engineering.com,
	linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, linux-xtensa@linux-xtensa.org,
	iommu@lists.linux-foundation.org, dann.frazier@canonical.com
Subject: Re: [PATCH v3 2/2] dma-contiguous: Use fallback alloc_pages for single pages
Date: Fri, 24 May 2019 09:16:19 -0700	[thread overview]
Message-ID: <20190524161618.GB23100@iweiny-DESK2.sc.intel.com> (raw)
In-Reply-To: <20190524040633.16854-3-nicoleotsuka@gmail.com>

On Thu, May 23, 2019 at 09:06:33PM -0700, Nicolin Chen wrote:
> The addresses within a single page are always contiguous, so it's
> not so necessary to always allocate one single page from CMA area.
> Since the CMA area has a limited predefined size of space, it may
> run out of space in heavy use cases, where there might be quite a
> lot CMA pages being allocated for single pages.
> 
> However, there is also a concern that a device might care where a
> page comes from -- it might expect the page from CMA area and act
> differently if the page doesn't.

How does a device know, after this call, if a CMA area was used?  From the
patches I figured a device should not care.

> 
> This patch tries to use the fallback alloc_pages path, instead of
> one-page size allocations from the global CMA area in case that a
> device does not have its own CMA area. This'd save resources from
> the CMA global area for more CMA allocations, and also reduce CMA
> fragmentations resulted from trivial allocations.
> 
> Signed-off-by: Nicolin Chen <nicoleotsuka@gmail.com>
> ---
>  kernel/dma/contiguous.c | 11 ++++++++++-
>  1 file changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/kernel/dma/contiguous.c b/kernel/dma/contiguous.c
> index 21f39a6cb04f..6914b92d5c88 100644
> --- a/kernel/dma/contiguous.c
> +++ b/kernel/dma/contiguous.c
> @@ -223,14 +223,23 @@ bool dma_release_from_contiguous(struct device *dev, struct page *pages,
>   * This function allocates contiguous memory buffer for specified device. It
>   * first tries to use device specific contiguous memory area if available or
>   * the default global one, then tries a fallback allocation of normal pages.
> + *
> + * Note that it byapss one-page size of allocations from the global area as
> + * the addresses within one page are always contiguous, so there is no need
> + * to waste CMA pages for that kind; it also helps reduce fragmentations.
>   */
>  struct page *dma_alloc_contiguous(struct device *dev, size_t size, gfp_t gfp)
>  {
>  	int node = dev ? dev_to_node(dev) : NUMA_NO_NODE;
>  	size_t count = PAGE_ALIGN(size) >> PAGE_SHIFT;
>  	size_t align = get_order(PAGE_ALIGN(size));
> -	struct cma *cma = dev_get_cma_area(dev);
>  	struct page *page = NULL;
> +	struct cma *cma = NULL;
> +
> +	if (dev && dev->cma_area)
> +		cma = dev->cma_area;
> +	else if (count > 1)
> +		cma = dma_contiguous_default_area;

Doesn't dev_get_dma_area() already do this?

Ira

>  
>  	/* CMA can be used only in the context which permits sleeping */
>  	if (cma && gfpflags_allow_blocking(gfp)) {
> -- 
> 2.17.1
> 

  reply	other threads:[~2019-05-24 16:15 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-05-24  4:06 [PATCH v3 0/2] Optimize dma_*_from_contiguous calls Nicolin Chen
2019-05-24  4:06 ` [PATCH v3 1/2] dma-contiguous: Abstract dma_{alloc,free}_contiguous() Nicolin Chen
2019-05-29 18:35   ` Nathan Chancellor
2019-05-29 22:48     ` Nicolin Chen
2019-07-25 16:06   ` Ezequiel Garcia
2019-07-25 16:50     ` Nicolin Chen
2019-07-25 17:31       ` Dafna Hirschfeld
2019-07-25 23:42         ` Nicolin Chen
2019-05-24  4:06 ` [PATCH v3 2/2] dma-contiguous: Use fallback alloc_pages for single pages Nicolin Chen
2019-05-24 16:16   ` Ira Weiny [this message]
2019-05-27 10:55     ` Nicolin Chen
2019-05-24 20:03 ` [PATCH v3 0/2] Optimize dma_*_from_contiguous calls dann frazier
2019-05-28  6:04 ` Christoph Hellwig
2019-05-29 23:06   ` Nicolin Chen

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=20190524161618.GB23100@iweiny-DESK2.sc.intel.com \
    --to=ira.weiny@intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=catalin.marinas@arm.com \
    --cc=chris@zankel.net \
    --cc=dann.frazier@canonical.com \
    --cc=dwmw2@infradead.org \
    --cc=hch@lst.de \
    --cc=iamjoonsoo.kim@lge.com \
    --cc=iommu@lists.linux-foundation.org \
    --cc=jcmvbkbc@gmail.com \
    --cc=joro@8bytes.org \
    --cc=keescook@chromium.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-xtensa@linux-xtensa.org \
    --cc=linux@armlinux.org.uk \
    --cc=m.szyprowski@samsung.com \
    --cc=nicoleotsuka@gmail.com \
    --cc=robin.murphy@arm.com \
    --cc=sfr@canb.auug.org.au \
    --cc=tony@atomide.com \
    --cc=treding@nvidia.com \
    --cc=vdumpa@nvidia.com \
    --cc=will.deacon@arm.com \
    --cc=wsa+renesas@sang-engineering.com \
    --subject='Re: [PATCH v3 2/2] dma-contiguous: Use fallback alloc_pages for single pages' \
    /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

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).