Re: [PATCH 3/5] mm: powerpc ptlock comments

[Hugh Dickins <hugh@veritas.com>]

On Tue, 22 Nov 2005, Paul Mackerras wrote:
> 
> > Update comments (only) on page_table_lock and mmap_sem in arch/powerpc.
> 
> [snip]
> 
> >  	/*
> > -	 * No need to use ldarx/stdcx here because all who
> > -	 * might be updating the pte will hold the
> > -	 * page_table_lock
> > +	 * No need to use ldarx/stdcx here
> >  	 */
> 
> If you're going to remove the because clause you might as well remove
> the whole comment.  What you have left is either redundant or
> mystifying (according to the depth of knowledge of the reader).  And
> in fact I don't think I could now say why we don't need an atomic
> update sequence there, so I would appreciate something that updates
> the "because" part.

All I'm doing here is tidying up by removing the no-longer-appropriate
reference to page_table_lock.  As my snipped patch comment said:

> Update comments (only) on page_table_lock and mmap_sem in arch/powerpc.
> Removed the comment on page_table_lock from hash_huge_page: since it's
> no longer taking page_table_lock itself, it's irrelevant whether others
> are; but how it is safe (even against huge file truncation?) I can't say.

I'd appreciate something that updates the "because" part too, but don't
know the answer.  That other patch will need to come from your end - Ben?

Hugh