Re: [PATCH 1/5][RFC][CFT] percpu fixes, part 1

[Al Viro <viro@ZenIV.linux.org.uk>]

On Fri, Mar 14, 2014 at 02:47:45PM -0400, Tejun Heo wrote:
> On Fri, Mar 14, 2014 at 06:45:45PM +0000, Al Viro wrote:
> > There's a missing piece in 2/3 (percpu: store offsets instead of lengths in
> > ->map[]) - we need to round size up to multiple of 2 in pcpu_alloc().
> > Updated patch follows; if you want an incremental instead of replacement -
> > please, yell.
> 
> Yeah, I'd prefer an incremental patch.  Also, can you please update
> the comment above alignment / size updates with a bit more detail?
> 
> Thanks!

Done; incremental follows:

diff --git a/mm/percpu.c b/mm/percpu.c
index 5bb658a..65cab3b 100644
--- a/mm/percpu.c
+++ b/mm/percpu.c
@@ -713,11 +713,14 @@ static void __percpu *pcpu_alloc(size_t size, size_t align, bool reserved)
 
 	/*
 	 * We want the lowest bit of offset available for in-use/free
-	 * indicator.
+	 * indicator, so force >= 16bit alignment and make size even.
 	 */
 	if (unlikely(align < 2))
 		align = 2;
 
+	if (unlikely(size & 1))
+		size++;
+
 	if (unlikely(!size || size > PCPU_MIN_UNIT_SIZE || align > PAGE_SIZE)) {
 		WARN(true, "illegal size (%zu) or align (%zu) for "
 		     "percpu allocation\n", size, align);