[BusyBox] idea regarding space savings wrt usage strings

[Manuel Novoa III]

Erik,

On Tue, 30 Jan 2001, Erik Andersen wrote:
> Hmm.  How ugly is the patch?  I think I am willing to live with
> a slightly larger busybox if the code is more maintainable.  So
> if this makes the code harder to maintain...

Here's the patch.  Most of the size is changes usage.c from individual strings
to a format that the code below it can use to auto-generate usage_gen.[ch].

I built with everything configured and it worked fine.  Sizes below.

[mjn3 at mars embedded]$ ls -l busybox-all busybox-dist-all 
-rwxrwxr-x    1 mjn3     mjn3       263184 Jan 30 17:43 busybox-all
-rwxrwxr-x    1 mjn3     mjn3       264848 Jan 30 16:50 busybox-dist-all
[mjn3 at mars embedded]$ size busybox-all busybox-dist-all 
   text	   data	    bss	    dec	    hex	filename
 232047	  29668	  40540	 302255	  49caf	busybox-all
 233679	  29700	  40540	 303919	  4a32f	busybox-dist-all
[mjn3 at mars embedded]$ ls -l busybox*.gz
-rwxrwxr-x    1 mjn3     mjn3       122112 Jan 30 17:43 busybox-all.gz
-rwxrwxr-x    1 mjn3     mjn3       122258 Jan 30 16:50 busybox-dist-all.gz

I did keep the original usage function, but renamed it usage_from_str.  It was
needed by init.  Also, a currently unused file named usage_gen.data will be
created.  I'll use this in playing with compressed usage messages.

Manuel

-------------- next part --------------
A non-text attachment was scrubbed...
Name: busybox-usage.diff.gz
Type: application/x-gzip
Size: 11398 bytes
Desc: not available
Url : http://lists.busybox.net/pipermail/busybox/attachments/20010130/d9ace90a/attachment.bin