From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: michael@ellerman.id.au
Cc: linuxppc-dev@ozlabs.org
Subject: Re: Swiotlb breaks pseries
Date: Tue, 23 Jun 2009 19:32:23 +1000 [thread overview]
Message-ID: <1245749543.4017.69.camel@pasglop> (raw)
In-Reply-To: <1245748428.4550.411.camel@concordia>
On Tue, 2009-06-23 at 19:13 +1000, Michael Ellerman wrote:
> Hi all,
>
> Turning on SWIOTLB selects or enables PPC_NEED_DMA_SYNC_OPS, which means
> we get the non empty versions of dma_sync_* in asm/dma-mapping.h
>
> On my pseries machine the dma_ops have no such routines and we die with
> a null pointer - this patch gets it booting, is there a more elegant way
> to do it?
No that's the best way (checking for NULL in the inlines), avoids
calling through a function pointer when not needed, since this can
be a fairly hot path, especially with networking.
Cheers,
Ben.
> cheers
>
> diff --git a/arch/powerpc/include/asm/dma-mapping.h b/arch/powerpc/include/asm/dma-mapping.h
> index 3d9e887..b44aaab 100644
> --- a/arch/powerpc/include/asm/dma-mapping.h
> +++ b/arch/powerpc/include/asm/dma-mapping.h
> @@ -309,7 +309,9 @@ static inline void dma_sync_single_for_cpu(struct device *dev,
> struct dma_mapping_ops *dma_ops = get_dma_ops(dev);
>
> BUG_ON(!dma_ops);
> - dma_ops->sync_single_range_for_cpu(dev, dma_handle, 0,
> +
> + if (dma_ops->sync_single_range_for_cpu)
> + dma_ops->sync_single_range_for_cpu(dev, dma_handle, 0,
> size, direction);
> }
>
> @@ -320,7 +322,9 @@ static inline void dma_sync_single_for_device(struct device *dev,
> struct dma_mapping_ops *dma_ops = get_dma_ops(dev);
>
> BUG_ON(!dma_ops);
> - dma_ops->sync_single_range_for_device(dev, dma_handle,
> +
> + if (dma_ops->sync_single_range_for_device)
> + dma_ops->sync_single_range_for_device(dev, dma_handle,
> 0, size, direction);
> }
>
> @@ -331,7 +335,9 @@ static inline void dma_sync_sg_for_cpu(struct device *dev,
> struct dma_mapping_ops *dma_ops = get_dma_ops(dev);
>
> BUG_ON(!dma_ops);
> - dma_ops->sync_sg_for_cpu(dev, sgl, nents, direction);
> +
> + if (dma_ops->sync_sg_for_cpu)
> + dma_ops->sync_sg_for_cpu(dev, sgl, nents, direction);
> }
>
> static inline void dma_sync_sg_for_device(struct device *dev,
> @@ -341,7 +347,9 @@ static inline void dma_sync_sg_for_device(struct device *dev,
> struct dma_mapping_ops *dma_ops = get_dma_ops(dev);
>
> BUG_ON(!dma_ops);
> - dma_ops->sync_sg_for_device(dev, sgl, nents, direction);
> +
> + if (dma_ops->sync_sg_for_device)
> + dma_ops->sync_sg_for_device(dev, sgl, nents, direction);
> }
>
> static inline void dma_sync_single_range_for_cpu(struct device *dev,
> @@ -351,7 +359,9 @@ static inline void dma_sync_single_range_for_cpu(struct device *dev,
> struct dma_mapping_ops *dma_ops = get_dma_ops(dev);
>
> BUG_ON(!dma_ops);
> - dma_ops->sync_single_range_for_cpu(dev, dma_handle,
> +
> + if (dma_ops->sync_single_range_for_cpu)
> + dma_ops->sync_single_range_for_cpu(dev, dma_handle,
> offset, size, direction);
> }
>
> @@ -362,7 +372,9 @@ static inline void dma_sync_single_range_for_device(struct device *dev,
> struct dma_mapping_ops *dma_ops = get_dma_ops(dev);
>
> BUG_ON(!dma_ops);
> - dma_ops->sync_single_range_for_device(dev, dma_handle, offset,
> +
> + if (dma_ops->sync_single_range_for_device)
> + dma_ops->sync_single_range_for_device(dev, dma_handle, offset,
> size, direction);
> }
> #else /* CONFIG_PPC_NEED_DMA_SYNC_OPS */
prev parent reply other threads:[~2009-06-23 9:32 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-06-23 9:13 Swiotlb breaks pseries Michael Ellerman
2009-06-23 9:32 ` Benjamin Herrenschmidt [this message]
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=1245749543.4017.69.camel@pasglop \
--to=benh@kernel.crashing.org \
--cc=linuxppc-dev@ozlabs.org \
--cc=michael@ellerman.id.au \
/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).