ok, what's the official take on "endianness"? and hdparm problem

Ken McGuire kenm at desertweyr.com
Wed Jun 7 02:36:49 UTC 2006


On 6 Jun 2006 at 20:28, Rob Landley wrote:

> On Tuesday 06 June 2006 4:11 pm, Ken McGuire wrote:
> >  * This message contains the file 'patch.bb1', which has been
> >  * uuencoded. If you are using Pegasus Mail, then you can use
> >  * the browser's eXtract function to lift the original contents
> >  * out to a file, otherwise you will have to extract the message
> >  * and uudecode it manually.
> 
> Ever heard of mime encoding?
> 

Indeed, my face is red...

> You can also cut and paste the patch inline.  Whitespace mangling's better 
> than that mess...
> 

 Appologies again...

> I uudecoded it by hand, noticed you removed the swab entirely, scratched my 
> head for a bit, went and pondered the code, and realized that both callers of 
> identify() were already adjusting the endianness of the data before calling 
> identify().  (Which is weird, how did this _ever_ work on a big endian 
> system?  It's broken back in 1.1.1, I think...)
> 

I had never used busybox hdparm on ppc before busybox 1.1.2 and it was broken there.
I've been using busybox/buildroot/uClibc since early 2002, mostly on ARM, some x86 and some MIPS.

> The correct thing to do isn't to remove the adjustment from identify, but to 
> remove it from each of the callers so it's done in one place.
> 

Thanks for the patience, I just did a svn up, re-compiled and busybox hdparm 
seems to produce the proper results.. !! 

> http://busybox.net/downloads/patches/svn-15301.patch
> 

I'll read over the patch and see what I can learn, Thanks...

> Rob
> -- 
> Never bet against the cheap plastic solution.
> 

   ...ken...




More information about the busybox mailing list