iommu-helpers.h patch

iommu-helpers.h patch

[rubisher]

Hello *,

This seems to be missed a while ago ;-)

--- drivers-parisc-iommu-helpers.h-git20071218	2007-12-21 19:02:38.000000000 +0000
+++ drivers-parisc-iommu-helpers.h-git20071218.new	2007-12-22 13:39:04.000000000 +0000
@@ -29,7 +29,7 @@
  		long size;

  		DBG_RUN_SG(" %d : %08lx/%05x %08lx/%05x\n", nents,
-			   (unsigned long)sg_dma_address(startsg), cnt,
+			   (unsigned long)sg_dma_address(startsg), size,
  			   sg_virt_addr(startsg), startsg->length
  		);
=== <> ===

Cheers,
	r.

Re: iommu-helpers.h patch

[Grant Grundler]

On Sat, Dec 22, 2007 at 01:45:45PM +0000, rubisher wrote:
> Hello *,
>
> This seems to be missed a while ago ;-)
>
> --- drivers-parisc-iommu-helpers.h-git20071218	2007-12-21 
> 19:02:38.000000000 +0000
> +++ drivers-parisc-iommu-helpers.h-git20071218.new	2007-12-22 
> 13:39:04.000000000 +0000
> @@ -29,7 +29,7 @@
>  		long size;
>
>  		DBG_RUN_SG(" %d : %08lx/%05x %08lx/%05x\n", nents,
> -			   (unsigned long)sg_dma_address(startsg), cnt,
> +			   (unsigned long)sg_dma_address(startsg), size,

Sorry - but this is wrong. "size" isn't initialized until later.
"cnt" was the thing I needed/wanted dumped.

cheers,
grant

>  			   sg_virt_addr(startsg), startsg->length
>  		);
> === <> ===
>
> Cheers,
> 	r.
> -
> To unsubscribe from this list: send the line "unsubscribe linux-parisc" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: iommu-helpers.h patch

[rubisher]

Grant Grundler wrote:
> On Sat, Dec 22, 2007 at 01:45:45PM +0000, rubisher wrote:
>> Hello *,
>>
>> This seems to be missed a while ago ;-)
>>
>> --- drivers-parisc-iommu-helpers.h-git20071218	2007-12-21 
>> 19:02:38.000000000 +0000
>> +++ drivers-parisc-iommu-helpers.h-git20071218.new	2007-12-22 
>> 13:39:04.000000000 +0000
>> @@ -29,7 +29,7 @@
>>  		long size;
>>
>>  		DBG_RUN_SG(" %d : %08lx/%05x %08lx/%05x\n", nents,
>> -			   (unsigned long)sg_dma_address(startsg), cnt,
>> +			   (unsigned long)sg_dma_address(startsg), size,
> 
> Sorry - but this is wrong. "size" isn't initialized until later.
Agree with you: size is initialized later.

Anyway cnt variable was gone:
# grep cnt drivers/parisc/iommu-helpers.h
                            (unsigned long)sg_dma_address(startsg), cnt,

and imo it was with this patch:
<http://cvs.parisc-linux.org/linux-2.6/drivers/parisc/ccio-dma.c?r1=1.11&r2=1.12&makepatch=1&diff_format=u>

the interesting hunk's part was:
-		int cnt = sg_dma_len(startsg);
-		sg_dma_len(startsg) = 0;
+		unsigned long vaddr;
+		long size;

  		DBG_RUN_SG(" %d : %08lx/%05x %08lx/%05x\n", nents,
  			   (unsigned long)sg_dma_address(startsg), cnt,
  			   sg_virt_addr(startsg), startsg->length
  		);

+

(and make a kernel with DEBUG_CCIO_RUN_SG failed with an error at this place, so we need to patch ;-)

> "cnt" was the thing I needed/wanted dumped.
> 
So may be something like:

                 DBG_RUN_SG(" %d : %08lx/%05x %08lx/%05x\n", nents,
-                          (unsigned long)sg_dma_address(startsg), cnt,
+                          (unsigned long)sg_dma_address(startsg), sg_dma_len(startsg),
                            sg_virt_addr(startsg), startsg->length
                 );

Tx,
	r.

> cheers,
> grant
> 
>>  			   sg_virt_addr(startsg), startsg->length
>>  		);
>> === <> ===
>>
>> Cheers,
>> 	r.
>> -
>> To unsubscribe from this list: send the line "unsubscribe linux-parisc" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
> 

Re: iommu-helpers.h patch

[Grant Grundler]

On Fri, Dec 28, 2007 at 03:27:26PM +0000, rubisher wrote:
> Anyway cnt variable was gone:
> # grep cnt drivers/parisc/iommu-helpers.h
>                            (unsigned long)sg_dma_address(startsg), cnt,

oh! that's definitely wrong too then...

> the interesting hunk's part was:
> -		int cnt = sg_dma_len(startsg);
> -		sg_dma_len(startsg) = 0;
> +		unsigned long vaddr;
> +		long size;

Yup...

> (and make a kernel with DEBUG_CCIO_RUN_SG failed with an error at this 
> place, so we need to patch ;-)
>
>> "cnt" was the thing I needed/wanted dumped.
> So may be something like:
>
>                 DBG_RUN_SG(" %d : %08lx/%05x %08lx/%05x\n", nents,
> -                          (unsigned long)sg_dma_address(startsg), cnt,
> +                          (unsigned long)sg_dma_address(startsg), 
> sg_dma_len(startsg),
>                            sg_virt_addr(startsg), startsg->length
>                 );

Yeah, that would be better I think.

grant