All of lore.kernel.org
 help / color / mirror / Atom feed
From: gdavis@mvista.com (George G. Davis)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH] ARM: V6 MPCore v6_dma_inv_range RWFO fix
Date: Wed, 24 Nov 2010 10:10:16 -0500	[thread overview]
Message-ID: <20101124151015.GE31572@mvista.com> (raw)
In-Reply-To: <1290595333.3056.6.camel@e102109-lin.cambridge.arm.com>

On Wed, Nov 24, 2010 at 10:42:13AM +0000, Catalin Marinas wrote:
> On Tue, 2010-11-23 at 22:42 +0000, Russell King - ARM Linux wrote:
> > On Wed, Nov 24, 2010 at 01:28:06AM +0300, Valentine Barshak wrote:
> > > Cache ownership must be acqired by reading/writing data from the
> > > cache line to make cache operation have the desired effect on the
> > > SMP MPCore CPU. However, the ownership is never aquired in the
> > > v6_dma_inv_range function when cleaning the first line and
> > > flushing the last one, in case the address is not aligned
> > > to D_CACHE_LINE_SIZE boundary.
> > > Fix this by reading/writing data if needed, before performing
> > > cache operations.
> > 
> > You should do this on the data _inside_ the requested buffer.  We don't
> > know if the overlapping cache line shares itself with some atomic
> > variable, and doing a read-write on it could undo other updates to it.
> 
> We could just use the boundary addresses to avoid writing beyond the
> buffer. Something like below (pretty much moving the BIC after the RFO,
> untested):
> 
> 
> diff --git a/arch/arm/mm/cache-v6.S b/arch/arm/mm/cache-v6.S
> index 99fa688..d63bb55 100644
> --- a/arch/arm/mm/cache-v6.S
> +++ b/arch/arm/mm/cache-v6.S
> @@ -204,6 +204,10 @@ ENTRY(v6_flush_kern_dcache_area)
>   */
>  v6_dma_inv_range:
>  	tst	r0, #D_CACHE_LINE_SIZE - 1
> +#ifdef CONFIG_DMA_CACHE_RWFO
> +	ldrne	r2, [r0]			@ read for ownership
> +	strne	r2, [r0]			@ write for ownership

My concern doing RWFO prior to aligning "start" is that it could be
non-word-aligned which may result in an alignment fault.

> +#endif
>  	bic	r0, r0, #D_CACHE_LINE_SIZE - 1
>  #ifdef HARVARD_CACHE
>  	mcrne	p15, 0, r0, c7, c10, 1		@ clean D line
> @@ -211,6 +215,10 @@ v6_dma_inv_range:
>  	mcrne	p15, 0, r0, c7, c11, 1		@ clean unified line
>  #endif
>  	tst	r1, #D_CACHE_LINE_SIZE - 1
> +#ifdef CONFIG_DMA_CACHE_RWFO
> +	ldrne	r2, [r1, #-4]			@ read for ownership
> +	strne	r2, [r1, #-4]			@ write for ownership

Same concern here with "end".

Are start and end reasonably expected to be word aligned?

Thanks!

--
Regards,
George
> +#endif
>  	bic	r1, r1, #D_CACHE_LINE_SIZE - 1
>  #ifdef HARVARD_CACHE
>  	mcrne	p15, 0, r1, c7, c14, 1		@ clean & invalidate D line
> 
> 
> 
> -- 
> Catalin
> 
> 

  reply	other threads:[~2010-11-24 15:10 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-11-23 22:28 [PATCH] ARM: V6 MPCore v6_dma_inv_range RWFO fix Valentine Barshak
2010-11-23 22:42 ` Russell King - ARM Linux
2010-11-24  0:24   ` George G. Davis
2010-11-24 10:32     ` Catalin Marinas
2010-11-24 15:14       ` George G. Davis
2010-11-24 10:42   ` Catalin Marinas
2010-11-24 15:10     ` George G. Davis [this message]
2010-11-24 15:35       ` Catalin Marinas
2010-11-24 17:33     ` Russell King - ARM Linux
2010-11-24 17:46       ` Catalin Marinas
2010-11-24 18:01         ` Russell King - ARM Linux
2010-11-24 18:07           ` Catalin Marinas
2010-11-24 18:39             ` [PATCH] ARM: V6 MPCore v6_dma_inv_range and v6_dma_flush_range " Valentine Barshak
2010-12-08 10:25               ` Valentine Barshak
2010-12-09 16:04               ` Catalin Marinas
2010-12-09 17:04                 ` Valentine Barshak
2010-12-09 22:07                 ` Stephen Boyd
2010-12-10  5:44                   ` George G. Davis
2010-12-10  5:26               ` George G. Davis
2010-12-10  9:50                 ` Catalin Marinas
2010-12-10 12:29                   ` Valentine Barshak
2010-12-10 16:49                     ` George G. Davis
2010-11-29 15:28             ` Valentine Barshak
2010-11-24 17:40   ` [PATCH] ARM: V6 MPCore v6_dma_inv_range " Valentine Barshak

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=20101124151015.GE31572@mvista.com \
    --to=gdavis@mvista.com \
    --cc=linux-arm-kernel@lists.infradead.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.