mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
To: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com>,
	<xen-devel@lists.xensource.com>, <linux-kernel@vger.kernel.org>,
	<linux-arm-kernel@lists.infradead.org>, <Ian.Campbell@citrix.com>
Subject: Re: [PATCH v6 12/19] swiotlb: don't assume that io_tlb_start-io_tlb_end is coherent
Date: Wed, 2 Oct 2013 18:31:57 +0100	[thread overview]
Message-ID: <alpine.DEB.2.02.1310021826410.26077@kaball.uk.xensource.com> (raw)
In-Reply-To: <20130930155617.GS3106@phenom.dumpdata.com>

On Mon, 30 Sep 2013, Konrad Rzeszutek Wilk wrote:
> On Fri, Sep 27, 2013 at 05:10:00PM +0100, Stefano Stabellini wrote:
> > The swiotlb code has appropriate calls to dma_mark_clean in place for
> > buffers passed to swiotlb_map_page as an argument. However it assumes
> > that the swiotlb bounce buffer (io_tlb_start-io_tlb_end) is already
> > coherent and doesn't need any calls to dma_mark_clean.
> > 
> > On ARM the swiotlb bounce buffer is not coherent (the memory is
> > writealloc while it should be bufferable) and therefore we need to call
> > dma_mark_clean appropriately on the bounce buffer code paths too.
> > 
> > Note that most architecures have an empty dma_mark_clean implementation
> > anyway.
> 
> The other architecture that uses swiotlb is IA64 and that does have
> an implementation where it touches on page attributes.
> 
> Which means I have to figure out why my HP zx6000 won't boot with 3.11 now :-(
> 

Now this is a very thorny issue.

Honestly I don't like the dma_mark_clean interface very much: it's one
big hammer, when we actually need some finesse to handle coherency.

For example on ARM some devices might not need the dma_mark_clean call,
while others do. Calling it all the times is at the very best
inefficient and incorrect at worst.

I am thinking of calling the original map/unmap_page functions instead
(arm_dma_map_page or arm_coherent_dma_map_page in the arm case).
However in order to do that I would need to add more __get_dma_ops calls in
both lib/swiotlb.c and drivers/xen/swiotlb-xen.c


> > Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
> > ---
> >  lib/swiotlb.c |   13 ++++++++++---
> >  1 files changed, 10 insertions(+), 3 deletions(-)
> > 
> > diff --git a/lib/swiotlb.c b/lib/swiotlb.c
> > index 4e8686c..eb45d17 100644
> > --- a/lib/swiotlb.c
> > +++ b/lib/swiotlb.c
> > @@ -515,6 +515,7 @@ found:
> >  		io_tlb_orig_addr[index+i] = orig_addr + (i << IO_TLB_SHIFT);
> >  	if (dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL)
> >  		swiotlb_bounce(orig_addr, tlb_addr, size, DMA_TO_DEVICE);
> > +	dma_mark_clean(phys_to_virt(tlb_addr), size);
> >  
> >  	return tlb_addr;
> >  }
> > @@ -547,7 +548,10 @@ void swiotlb_tbl_unmap_single(struct device *hwdev, phys_addr_t tlb_addr,
> >  	 * First, sync the memory before unmapping the entry
> >  	 */
> >  	if (orig_addr && ((dir == DMA_FROM_DEVICE) || (dir == DMA_BIDIRECTIONAL)))
> > +	{
> > +		dma_mark_clean(phys_to_virt(tlb_addr), size);
> >  		swiotlb_bounce(orig_addr, tlb_addr, size, DMA_FROM_DEVICE);
> > +	}
> >  
> >  	/*
> >  	 * Return the buffer to the free list by setting the corresponding
> > @@ -587,17 +591,20 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
> >  
> >  	switch (target) {
> >  	case SYNC_FOR_CPU:
> > -		if (likely(dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL))
> > +		if (likely(dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL)) {
> > +			dma_mark_clean(phys_to_virt(tlb_addr), size);
> >  			swiotlb_bounce(orig_addr, tlb_addr,
> >  				       size, DMA_FROM_DEVICE);
> > +		}
> >  		else
> >  			BUG_ON(dir != DMA_TO_DEVICE);
> >  		break;
> >  	case SYNC_FOR_DEVICE:
> > -		if (likely(dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL))
> > +		if (likely(dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL)) {
> >  			swiotlb_bounce(orig_addr, tlb_addr,
> >  				       size, DMA_TO_DEVICE);
> > -		else
> > +			dma_mark_clean(phys_to_virt(tlb_addr), size);
> > +		} else
> >  			BUG_ON(dir != DMA_FROM_DEVICE);
> >  		break;
> >  	default:
> > -- 
> > 1.7.2.5
> > 
> 

  reply	other threads:[~2013-10-02 17:32 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-09-27 16:09 [PATCH v6 0/19] enable swiotlb-xen on arm and arm64 Stefano Stabellini
2013-09-27 16:09 ` [PATCH v6 01/19] arm: make SWIOTLB available Stefano Stabellini
2013-09-27 16:09 ` [PATCH v6 02/19] arm64: define DMA_ERROR_CODE Stefano Stabellini
2013-09-30 10:47   ` Catalin Marinas
2013-09-27 16:09 ` [PATCH v6 03/19] xen: introduce XENMEM_exchange_and_pin and XENMEM_unpin Stefano Stabellini
2013-09-30 14:54   ` Konrad Rzeszutek Wilk
2013-09-27 16:09 ` [PATCH v6 04/19] xen: make xen_create_contiguous_region return the dma address Stefano Stabellini
2013-09-27 16:09 ` [PATCH v6 05/19] swiotlb-xen: support autotranslate guests Stefano Stabellini
2013-09-27 16:09 ` [PATCH v6 06/19] xen/arm,arm64: enable SWIOTLB_XEN Stefano Stabellini
2013-09-30 15:14   ` Konrad Rzeszutek Wilk
2013-09-27 16:09 ` [PATCH v6 07/19] swiotlb-xen: introduce xen_swiotlb_set_dma_mask Stefano Stabellini
2013-09-27 16:09 ` [PATCH v6 08/19] arm/xen: get_dma_ops: return xen_dma_ops if we are running on Xen Stefano Stabellini
2013-09-30 15:17   ` Konrad Rzeszutek Wilk
2013-09-27 16:09 ` [PATCH v6 09/19] arm64/xen: " Stefano Stabellini
2013-09-30 10:48   ` Catalin Marinas
2013-09-27 16:09 ` [PATCH v6 10/19] xen: introduce xen_alloc/free_coherent_pages Stefano Stabellini
2013-09-30 15:31   ` Konrad Rzeszutek Wilk
2013-10-01 13:40     ` Catalin Marinas
2013-10-02 17:03       ` Stefano Stabellini
2013-10-02 17:07         ` Catalin Marinas
2013-10-02 17:14           ` Stefano Stabellini
2013-09-27 16:09 ` [PATCH v6 11/19] swiotlb-xen: use xen_alloc/free_coherent_pages Stefano Stabellini
2013-09-30 15:34   ` Konrad Rzeszutek Wilk
2013-09-27 16:10 ` [PATCH v6 12/19] swiotlb: don't assume that io_tlb_start-io_tlb_end is coherent Stefano Stabellini
2013-09-30 15:56   ` Konrad Rzeszutek Wilk
2013-10-02 17:31     ` Stefano Stabellini [this message]
2013-10-04 13:23       ` Konrad Rzeszutek Wilk
2013-09-27 16:10 ` [PATCH v6 13/19] ASoC: Samsung: Rename dma_ops by samsung_dma_ops Stefano Stabellini
2013-09-27 16:10 ` [PATCH v6 14/19] swiotlb: print a warning when the swiotlb is full Stefano Stabellini
2013-09-30 15:58   ` Konrad Rzeszutek Wilk
2013-10-09 17:31     ` Stefano Stabellini
2013-09-27 16:10 ` [PATCH v6 15/19] swiotlb-xen: call dma_capable only if dev->dma_mask is allocated Stefano Stabellini
2013-09-30 16:02   ` Konrad Rzeszutek Wilk
2013-10-02 17:13     ` Stefano Stabellini
2013-10-02 17:22       ` Konrad Rzeszutek Wilk
2013-10-03 18:35         ` Rob Herring
2013-09-27 16:10 ` [PATCH v6 16/19] arm,arm64: do not always merge biovec if we are running on Xen Stefano Stabellini
2013-09-30 16:06   ` Konrad Rzeszutek Wilk
2013-10-09 17:54     ` Stefano Stabellini
2013-09-27 16:10 ` [PATCH v6 17/19] xen: introduce XENMEM_pin Stefano Stabellini
2013-09-30 17:22   ` Konrad Rzeszutek Wilk
2013-09-27 16:10 ` [PATCH v6 18/19] swiotlb-xen: introduce a rbtree to track phys to bus mappings Stefano Stabellini
2013-09-30 17:27   ` Konrad Rzeszutek Wilk
2013-10-02 17:23     ` Stefano Stabellini
2013-10-09 17:25     ` Stefano Stabellini
2013-09-27 16:10 ` [PATCH v6 19/19] swiotlb-xen: instead of bouncing on the swiotlb, pin single pages Stefano Stabellini
2013-09-30 17:39   ` Konrad Rzeszutek Wilk
2013-10-09 17:27     ` Stefano Stabellini

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=alpine.DEB.2.02.1310021826410.26077@kaball.uk.xensource.com \
    --to=stefano.stabellini@eu.citrix.com \
    --cc=Ian.Campbell@citrix.com \
    --cc=konrad.wilk@oracle.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=xen-devel@lists.xensource.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

all inboxes | Powered by JetHome®