[Pvfs2-developers] viewdist get-eattr endianness

[Sam Lang]

On Apr 16, 2007, at 12:15 PM, Murali Vilayannur wrote:

> Hi Sam,
> Looks great.
> 2 Questions
> - Should the check in sys-get-eattr.sm be if (ret != 0) instead of  
> if (ret < 0)?
>  If future callouts do not return a -PVFS_* then we could have a  
> problem no?

Hi Murali,

I would want future callouts to return -PVFS_*, but I can change the  
if check.

> - module.mk.in lists a void.c as a new file. It does not seem to be a
> part of this patch? Is it a new file or an existing file?

That's from walt's concurrent state machine merge.  Not sure why its  
showing up in the patch..the void.sm should already have been added.   
I'll check it out!

> Nice cleanups with this new callout framework!
> thanks

Thanks for the comments!

-sam

> Murali
>
> On 4/16/07, Sam Lang <slang at mcs.anl.gov> wrote:
>>
>> Hi All,
>>
>> Attached is a patch that tries to get viewdist working properly when
>> client and server have different endianness.  The basic issue was
>> that viewdist uses get-eattr to get the datafile handle array, which
>> prevents the encoding/decoding of the handles.  The attached patch
>> adds encoding/decoding to known extended attributes (those in the
>> system.pvfs2. namespace).
>>
>> FYI:  I moved the pint-eattr.[hc] files from src/server/ to src/
>> common/misc so that the client can use them as well.  The changes
>> made to those files are essentially in the addition of encoding/
>> decoding callouts for the "dh" eattr.
>>
>> -sam
>>
>>
>> _______________________________________________
>> Pvfs2-developers mailing list
>> Pvfs2-developers at beowulf-underground.org
>> http://www.beowulf-underground.org/mailman/listinfo/pvfs2-developers
>>
>>
>>
>