All of lore.kernel.org
 help / color / mirror / Atom feed
* [RFC PATCH] net: sched: fix unsued cpu variable
@ 2014-09-13 18:13 John Fastabend
  2014-09-13 18:25 ` John Fastabend
  0 siblings, 1 reply; 2+ messages in thread
From: John Fastabend @ 2014-09-13 18:13 UTC (permalink / raw)
  To: netdev

kbuild test robot reported an unused variable cpu in cls_u32.c
after the patch below. This happens when PERF and MARK config
variables are disabled

commit 459d5f626da75573e985a7197b0919c3b143146c
Author: John Fastabend <john.fastabend@gmail.com>
Date:   Fri Sep 12 20:08:47 2014 -0700

    net: sched: make cls_u32 per cpu

One way to fix this is to use separate variables for perf
and mark and define the cpu variable inside the ifdef
logic.

I don't really like this though so I'm wondering if I can
clean up the ifdef/endif logic to make this easier to read.
So RFC patch until I convince myself there isn't a better
way. Thought it might be worth sending out in case someone
else has an idea.

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
---
 net/sched/cls_u32.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c
index 5ed5ac4..8cffe5a 100644
--- a/net/sched/cls_u32.c
+++ b/net/sched/cls_u32.c
@@ -788,8 +788,8 @@ static int u32_dump(struct net *net, struct tcf_proto *tp, unsigned long fh,
 	} else {
 #ifdef CONFIG_CLS_U32_PERF
 		struct tc_u32_pcnt *gpf;
-#endif
 		int cpu;
+#endif
 
 		if (nla_put(skb, TCA_U32_SEL,
 			    sizeof(n->sel) + n->sel.nkeys*sizeof(struct tc_u32_key),
@@ -816,9 +816,10 @@ static int u32_dump(struct net *net, struct tcf_proto *tp, unsigned long fh,
 			struct tc_u32_mark mark = {.val = n->val,
 						   .mask = n->mask,
 						   .success = 0};
+			int cpum;
 
-			for_each_possible_cpu(cpu) {
-				__u32 cnt = *per_cpu_ptr(n->pcpu_success, cpu);
+			for_each_possible_cpu(cpum) {
+				__u32 cnt = *per_cpu_ptr(n->pcpu_success, cpum);
 
 				mark.success += cnt;
 			}

^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [RFC PATCH] net: sched: fix unsued cpu variable
  2014-09-13 18:13 [RFC PATCH] net: sched: fix unsued cpu variable John Fastabend
@ 2014-09-13 18:25 ` John Fastabend
  0 siblings, 0 replies; 2+ messages in thread
From: John Fastabend @ 2014-09-13 18:25 UTC (permalink / raw)
  To: John Fastabend, netdev

On 09/13/2014 11:13 AM, John Fastabend wrote:
> kbuild test robot reported an unused variable cpu in cls_u32.c
> after the patch below. This happens when PERF and MARK config
> variables are disabled
> 
> commit 459d5f626da75573e985a7197b0919c3b143146c
> Author: John Fastabend <john.fastabend@gmail.com>
> Date:   Fri Sep 12 20:08:47 2014 -0700
> 
>     net: sched: make cls_u32 per cpu
> 
> One way to fix this is to use separate variables for perf
> and mark and define the cpu variable inside the ifdef
> logic.
> 
> I don't really like this though so I'm wondering if I can
> clean up the ifdef/endif logic to make this easier to read.
> So RFC patch until I convince myself there isn't a better
> way. Thought it might be worth sending out in case someone
> else has an idea.
> 
> Signed-off-by: John Fastabend <john.r.fastabend@intel.com>

kbuild robot kicked me another warning I missed I'll send out
a couple patches tonight or tomorrow morning to resolve them.

Thanks,
John

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2014-09-13 18:25 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-09-13 18:13 [RFC PATCH] net: sched: fix unsued cpu variable John Fastabend
2014-09-13 18:25 ` John Fastabend

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.