John Hawkinson <jhawk@MIT.EDU> writes:

> Hmm.
> 
> Some comments:
>   Is there a reason not to print what is being skipped,
>   just like the warning for large memory map entries?
> 
>   Shouldn't there be a documentation update indicating
>   what this warning message means if users encounter it?
>   I'm not sure what the best place for that would be, though.

More comments:

	- the code violates some KNF rule like

		+ second level indents are four spaces.
		+ binary operator hang on the previous line when line
		  breaks around it.

	- other cosmetic or minor error like

		+ whitespace at the end of line
		+ warning message prints unnecessary space after newline

	- as far as reading the source, the fix is wrong in some
	  aspect.

		+ seg_end shouldn't be re-calcurated by using
		  bim->entry[x].size after adjusting seg_start.  it
		  overruns.

		+ setting seg_end to 0x9ffff is wrong since it will be
		  truncated to page boundary later.  0xa0000 is fine.

		+ is there no case that the hole is completely
		  included the memory range?  The fix just throws
		  any pages before the hole in such case.  If there is
		  implicit assumption that there is no such case, it
		  should be commented.

enami.