Re: [PATCH 4/6] hugetlb: introduce alloc_nodemask_of_node

[Lee Schermerhorn <Lee.Schermerhorn@xxxxxx>]

On Thu, 2009-09-03 at 11:34 -0700, David Rientjes wrote:
> On Tue, 1 Sep 2009, Lee Schermerhorn wrote:
> 
> > > > Index: linux-2.6.31-rc7-mmotm-090827-0057/include/linux/nodemask.h
> > > > ===================================================================
> > > > --- linux-2.6.31-rc7-mmotm-090827-0057.orig/include/linux/nodemask.h	2009-08-28 09:21:19.000000000 -0400
> > > > +++ linux-2.6.31-rc7-mmotm-090827-0057/include/linux/nodemask.h	2009-08-28 09:21:29.000000000 -0400
> > > > @@ -245,18 +245,34 @@ static inline int __next_node(int n, con
> > > >  	return min_t(int,MAX_NUMNODES,find_next_bit(srcp->bits, MAX_NUMNODES, n+1));
> > > >  }
> > > >  
> > > > +#define init_nodemask_of_nodes(mask, node)				\
> > > > +	nodes_clear(*(mask));						\
> > > > +	node_set((node), *(mask));
> > > > +
> > > 
> > > Is the done thing to either make this a static inline or else wrap it in
> > > a do { } while(0) ? The reasoning being that if this is used as part of an
> > > another statement (e.g. a for loop) that it'll actually compile instead of
> > > throw up weird error messages.
> > 
> > Right.  I'll fix this [and signoff/review orders] next time [maybe last
> > time?].  It occurs to me that I can also use this for
> > huge_mpol_nodes_allowed(), so I'll move it up in the series and fix that
> > [which you've already ack'd].  I'll wait a bit to hear from David before
> > I respin.
> > 
> 
> I think it should be an inline function just so there's typechecking on 
> the first argument passed in (and so alloc_nodemask_of_node() below 
> doesn't get a NULL pointer dereference on node_set() if nmp can't be 
> allocated).

OK.  That works.  will be in v6

> 
> I've seen the issue about the signed-off-by/reviewed-by/acked-by order 
> come up before.  I've always put my signed-off-by line last whenever 
> proposing patches because it shows a clear order in who gathered those 
> lines when submitting to -mm, for example.  If I write
> 
> 	Cc: Mel Gorman <mel@xxxxxxxxx>
> 	Signed-off-by: David Rientjes <rientjes@xxxxxxxxxx>
> 
> it is clear that I cc'd Mel on the initial proposal.  If it is the other 
> way around, for example,
> 
> 	Signed-off-by: David Rientjes <rientjes@xxxxxxxxxx>
> 	Cc: Mel Gorman <mel@xxxxxxxxx>
> 	Signed-off-by: Andrew Morton...
> 
> then it indicates Andrew added the cc when merging into -mm.  That's more 
> relevant when such a line is acked-by or reviewed-by since it is now 
> possible to determine who received such acknowledgement from the 
> individual and is responsible for correctly relaying it in the patch 
> submission.
> 
> If it's done this way, it indicates that whoever is signing off the patch 
> is responsible for everything above it.  The type of line (signed-off-by, 
> reviewed-by, acked-by) is enough of an indication about the development 
> history of the patch, I believe, and it doesn't require specific ordering 
> to communicate (and the first line having to be a signed-off-by line isn't 
> really important, it doesn't replace the From: line).
> 
> It also appears to be how both Linus merges his own patches with Cc's.

???

> 
> > > > +/*
> > > > + * returns pointer to kmalloc()'d nodemask initialized to contain the
> > > > + * specified node.  Caller must free with kfree().
> > > > + */
> > > > +#define alloc_nodemask_of_node(node)					\
> > > > +({									\
> > > > +	typeof(_unused_nodemask_arg_) *nmp;				\
> > > > +	nmp = kmalloc(sizeof(*nmp), GFP_KERNEL);			\
> > > > +	if (nmp)							\
> > > > +		init_nodemask_of_nodes(nmp, (node));			\
> > > > +	nmp;								\
> > > > +})
> > > > +
> > > 

--
To unsubscribe from this list: send the line "unsubscribe linux-numa" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html