public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* Re: [PATCH 9/9] ttm/dma: Implement set_page_caching implementation in the TTM DMA pool code.
@ 2011-09-29 22:16 Kevin Shanahan
  2011-09-29 22:45 ` Konrad Rzeszutek Wilk
  0 siblings, 1 reply; 3+ messages in thread
From: Kevin Shanahan @ 2011-09-29 22:16 UTC (permalink / raw)
  To: Konrad Rzeszutek Wilk; +Cc: linux-kernel

On 2011-09-29 20:33:52, Konrad Rzeszutek Wilk wrote:
> . which is pretty much like the other TTM pool except it
> also handles moving the page to another pool list.
> 
> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
> ---
>  drivers/gpu/drm/ttm/ttm_page_alloc_dma.c |   96 ++++++++++++++++++++++++++++++
>  1 files changed, 96 insertions(+), 0 deletions(-)
> 
> diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c
> index 5909d28..cea031e 100644
> --- a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c
> +++ b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c
> @@ -1307,11 +1307,107 @@ static int ttm_dma_page_alloc_debugfs(struct seq_file *m, void *data)
>  	mutex_unlock(&_manager->lock);
>  	return 0;
>  }
> +#ifdef CONFIG_X86
> +static int ttm_dma_page_set_page_caching(struct page *p,
> +					 int flags,
> +					 enum ttm_caching_state c_old,
> +					 enum ttm_caching_state c_new,
> +					 struct device *dev)
> +{
> +	struct dma_pool *src, *dst;
> +	enum pool_type type;
> +	struct dma_page *dma_p;
> +	bool found = false;
> +	unsigned long irq_flags;
> +	int ret = 0;
> +
> +	if (!p)
> +		return 0;
> +
> +	if (PageHighMem(p))
> +		return 0;
> +
> +	type = ttm_to_type(flags, c_old);
> +	src = ttm_dma_find_pool(dev, type);
> +	if (!src) {
> +		WARN_ON(!src);
> +		return -ENOMEM;
> +	}
> +	type = ttm_to_type(flags, c_new);
> +	dst = ttm_dma_find_pool(dev, type);
> +	if (!dst) {
> +		gfp_t gfp_flags;
> +		if (flags & TTM_PAGE_FLAG_DMA32)
> +			gfp_flags = GFP_USER | GFP_DMA32;
> +		else
> +			gfp_flags = GFP_HIGHUSER;
> +
> +		if (flags & TTM_PAGE_FLAG_ZERO_ALLOC)
> +			gfp_flags |= __GFP_ZERO;
> +
> +		dst = ttm_dma_pool_init(dev, gfp_flags, type);
> +		if (IS_ERR_OR_NULL(dst))
> +			return -ENOMEM;
> +	}
> +
> +	dev_dbg(dev, "(%d) Caching %p (%p) from %x to %x.\n", current->pid,
> +		p, page_address(p), c_old, c_new);
> +
> +	if (c_old != tt_cached) {
> +		/* p isn't in the default caching state, set it to
> +		 * writeback first to free its current memtype. */
> +
> +		ret = set_pages_wb(p, 1);
> +		if (ret)
> +			return ret;
> +	}
>  
> +	if (c_new == tt_wc)
> +		ret = set_memory_wc((unsigned long) page_address(p), 1);
> +	else if (c_new == tt_uncached)
> +		ret = set_pages_uc(p, 1);
> +
> +	if (ret)
> +		return ret;
> +
> +	dev_dbg(src->dev, "(%s:%d) Moving %p (%p) to %s.\n", src->name,
> +		current->pid, p, page_address(p), dst->name);
> +
> +	/* To make it faster we only take the spinlock on list
> +	 * removal, and later on adding the page to the destination pool. */
> +	spin_lock_irqsave(&src->lock, irq_flags);
> +	list_for_each_entry(dma_p, &src->page_list, page_list) {
> +		if (virt_to_page(dma_p->vaddr) != p) {
> +			pr_debug("%s: (%s:%d) Skipping %p (%p) (DMA:0x%lx)\n",
> +				src->dev_name, src->name, current->pid,
> +				dma_p->vaddr,
> +				virt_to_page(dma_p->vaddr),
> +				(unsigned long)dma_p->dma);
> +			continue;
> +		}
> +		list_del(&dma_p->page_list);
> +		src->npages_in_use -= 1;
> +		found = true;
> +		break;
> +	}
> +	spin_lock_irqsave(&src->lock, irq_flags);
        ^^^^^^^^^^^^^^^^^

Was this meant to be spin_unlock_irqrestore?

Cheers,
Kevin.

> +	if (!found)
> +		return -ENODEV;
> +
> +	spin_lock_irqsave(&dst->lock, irq_flags);
> +	list_add(&dma_p->page_list, &dst->page_list);
> +	dst->npages_in_use++;
> +	spin_unlock_irqrestore(&dst->lock, irq_flags);
> +	return 0;
> +};
> +#endif
>  struct ttm_page_alloc_func ttm_page_alloc_dma = {
>  	.get_pages	= ttm_dma_get_pages,
>  	.put_pages	= ttm_dma_put_pages,
>  	.alloc_init	= ttm_dma_page_alloc_init,
>  	.alloc_fini	= ttm_dma_page_alloc_fini,
>  	.debugfs	= ttm_dma_page_alloc_debugfs,
> +#ifdef CONFIG_X86
> +	.set_caching	= ttm_dma_page_set_page_caching,
> +#endif
>  };
> -- 
> 1.7.4.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/etc.

^ permalink raw reply	[flat|nested] 3+ messages in thread
* [PATCH] TTM DMA pool v1.8
@ 2011-09-29 20:33 Konrad Rzeszutek Wilk
  2011-09-29 20:33 ` [PATCH 9/9] ttm/dma: Implement set_page_caching implementation in the TTM DMA pool code Konrad Rzeszutek Wilk
  0 siblings, 1 reply; 3+ messages in thread
From: Konrad Rzeszutek Wilk @ 2011-09-29 20:33 UTC (permalink / raw)
  To: linux-kernel, thellstrom, dri-devel, bskeggs, j.glisse, thomas,
	airlied, airlied, alexdeucher
  Cc: xen-devel

[.. and this is what I said in v1 post]:

Way back in January this patchset:
http://lists.freedesktop.org/archives/dri-devel/2011-January/006905.html
was merged in, but pieces of it had to be reverted b/c they did not
work properly under PowerPC, ARM, and when swapping out pages to disk.

After a bit of discussion on the mailing list
http://marc.info/?i=4D769726.2030307@shipmail.org I started working on it, but
got waylaid by other things .. and finally I am able to post the RFC patches.

There was a lot of discussion about it and I am not sure if I captured
everybody's thoughts - if I did not - that is _not_ intentional - it has just
been quite some time..

Anyhow .. the patches explore what the "lib/dmapool.c" does - which is to have a
DMA pool that the device has associated with. I kind of married that code
along with drivers/gpu/drm/ttm/ttm_page_alloc.c to create a TTM DMA pool code.
The end result is DMA pool with extra features: can do write-combine, uncached,
writeback (and tracks them and sets back to WB when freed); tracks "cached"
pages that don't really need to be returned to a pool; and hooks up to
the shrinker code so that the pools can be shrunk.

If you guys think this set of patches make sense  - my future plans were
 1) Get this in large crowd of testing .. and if it works for a kernel release
 2) to move a bulk of this in the lib/dmapool.c (I spoke with Matthew Wilcox
    about it and he is OK as long as I don't introduce performance regressions).

But before I do any of that a second set of eyes taking a look at these
patches would be most welcome.

In regards to testing, I've been running them non-stop for the last month
(and found some issues which I've fixed up) - and been quite happy with how
they work.

Michel (thanks!) took a spin of the patches on his PowerPC and they did not
cause any regressions (wheew).

The patches are also located in a git tree:

 git://oss.oracle.com/git/kwilk/xen.git devel/ttm.dma_pool.v1.8

 drivers/gpu/drm/nouveau/nouveau_mem.c    |    8 +-
 drivers/gpu/drm/nouveau/nouveau_sgdma.c  |    3 +-
 drivers/gpu/drm/radeon/radeon_device.c   |    6 +
 drivers/gpu/drm/radeon/radeon_gart.c     |    4 +-
 drivers/gpu/drm/radeon/radeon_ttm.c      |    3 +-
 drivers/gpu/drm/ttm/Makefile             |    3 +
 drivers/gpu/drm/ttm/ttm_bo.c             |    4 +-
 drivers/gpu/drm/ttm/ttm_memory.c         |    5 +
 drivers/gpu/drm/ttm/ttm_page_alloc.c     |   63 ++-
 drivers/gpu/drm/ttm/ttm_page_alloc_dma.c | 1317 ++++++++++++++++++++++++++++++
 drivers/gpu/drm/ttm/ttm_tt.c             |    5 +-
 drivers/gpu/drm/vmwgfx/vmwgfx_drv.c      |    4 +-
 drivers/xen/swiotlb-xen.c                |    2 +-
 include/drm/ttm/ttm_bo_driver.h          |    7 +-
 include/drm/ttm/ttm_page_alloc.h         |  100 +++-
 include/linux/swiotlb.h                  |    7 +-
 lib/swiotlb.c                            |    5 +-
 17 files changed, 1516 insertions(+), 30 deletions(-)

^ permalink raw reply	[flat|nested] 3+ messages in thread
end of thread, other threads:[~2011-09-29 22:45 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-09-29 22:16 [PATCH 9/9] ttm/dma: Implement set_page_caching implementation in the TTM DMA pool code Kevin Shanahan
2011-09-29 22:45 ` Konrad Rzeszutek Wilk
  -- strict thread matches above, loose matches on Subject: below --
2011-09-29 20:33 [PATCH] TTM DMA pool v1.8 Konrad Rzeszutek Wilk
2011-09-29 20:33 ` [PATCH 9/9] ttm/dma: Implement set_page_caching implementation in the TTM DMA pool code Konrad Rzeszutek Wilk

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox