Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1199392 > unrolled thread

[PATCH] net_dbg_ratelimited: turn into no-op when !DEBUG

Back to article view | Back to linux.kernel

Contents

  [PATCH] net_dbg_ratelimited: turn into no-op when !DEBUG "Jason A. Donenfeld" <Jason@zx2c4.com> - 2015-08-04 05:30 +0200
    Re: [PATCH] net_dbg_ratelimited: turn into no-op when !DEBUG Joe Perches <joe@perches.com> - 2015-08-04 06:00 +0200
      Re: [PATCH] net_dbg_ratelimited: turn into no-op when !DEBUG Joe Perches <joe@perches.com> - 2015-08-04 06:10 +0200
        Re: [PATCH] net_dbg_ratelimited: turn into no-op when !DEBUG David Miller <davem@davemloft.net> - 2015-08-04 07:00 +0200
          Re: [PATCH] net_dbg_ratelimited: turn into no-op when !DEBUG "Jason A. Donenfeld" <Jason@zx2c4.com> - 2015-08-04 17:10 +0200

#1199392 — [PATCH] net_dbg_ratelimited: turn into no-op when !DEBUG

The pr_debug family of functions turns into a no-op when -DDEBUG is not
specified, opting instead to call "no_printk", which gets compiled to a
no-op (but retains gcc's nice warnings about printf-style arguments).

The problem with net_dbg_ratelimited is that it is defined to be a
variant of net_ratelimited_function, which expands to essentially:

    if (net_ratelimit())
        pr_debug(fmt, ...);

When DEBUG is not defined, then this becomes,

    if (net_ratelimit())
        ;

This seems benign, except it isn't. Firstly, there's the obvious
overhead of calling net_ratelimit needlessly, which does quite some book
keeping for the rate limiting. Given that the pr_debug and
net_dbg_ratelimited family of functions are sprinkled liberally through
performance critical code, with developers assuming they'll be compiled
out to a no-op most of the time, we certainly do not want this needless
book keeping. Secondly, and most visibly, even though no debug message
is printed when DEBUG is not defined, if there is a flood of
invocations, dmesg winds up peppered with messages such as
"net_ratelimit: 320 callbacks suppressed". This is because our
aforementioned net_ratelimit() function actually prints this text in
some circumstances. It's especially odd to see this when there isn't any
other accompanying debug message.

So, in sum, it doesn't make sense to have this function's current
behavior, and instead it should match what every other debug family of
functions in the kernel does with !DEBUG -- nothing.

This patch replaces calls to net_dbg_ratelimited when !DEBUG with
no_printk, keeping with the idiom of all the other debug print helpers.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
---
 include/linux/net.h | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/include/linux/net.h b/include/linux/net.h
index 04aa068..500fdfe 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -239,8 +239,13 @@ do {								\
 	net_ratelimited_function(pr_warn, fmt, ##__VA_ARGS__)
 #define net_info_ratelimited(fmt, ...)				\
 	net_ratelimited_function(pr_info, fmt, ##__VA_ARGS__)
+#if defined(DEBUG)
 #define net_dbg_ratelimited(fmt, ...)				\
 	net_ratelimited_function(pr_debug, fmt, ##__VA_ARGS__)
+#else
+#define net_dbg_ratelimited(fmt, ...)				\
+	no_printk(KERN_DEBUG pr_fmt(fmt), ##__VA_ARGS__)
+#endif
 
 bool __net_get_random_once(void *buf, int nbytes, bool *done,
 			   struct static_key *done_key);
-- 
2.4.5

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1199400

On Tue, 2015-08-04 at 05:26 +0200, Jason A. Donenfeld wrote:
> The pr_debug family of functions turns into a no-op when -DDEBUG is not
> specified, opting instead to call "no_printk", which gets compiled to a
> no-op (but retains gcc's nice warnings about printf-style arguments).
> 
> The problem with net_dbg_ratelimited is that it is defined to be a
> variant of net_ratelimited_function, which expands to essentially:
> 
>     if (net_ratelimit())
>         pr_debug(fmt, ...);
> 
> When DEBUG is not defined, then this becomes,
> 
>     if (net_ratelimit())
>         ;
> 
> This seems benign, except it isn't. Firstly, there's the obvious
> overhead of calling net_ratelimit needlessly, which does quite some book
> keeping for the rate limiting. Given that the pr_debug and
> net_dbg_ratelimited family of functions are sprinkled liberally through
> performance critical code, with developers assuming they'll be compiled
> out to a no-op most of the time, we certainly do not want this needless
> book keeping. Secondly, and most visibly, even though no debug message
> is printed when DEBUG is not defined, if there is a flood of
> invocations, dmesg winds up peppered with messages such as
> "net_ratelimit: 320 callbacks suppressed". This is because our
> aforementioned net_ratelimit() function actually prints this text in
> some circumstances. It's especially odd to see this when there isn't any
> other accompanying debug message.
> 
> So, in sum, it doesn't make sense to have this function's current
> behavior, and instead it should match what every other debug family of
> functions in the kernel does with !DEBUG -- nothing.
> 
> This patch replaces calls to net_dbg_ratelimited when !DEBUG with
> no_printk, keeping with the idiom of all the other debug print helpers.

Makes sense, thanks Jason.



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1199402

On Mon, 2015-08-03 at 20:57 -0700, Joe Perches wrote:
> On Tue, 2015-08-04 at 05:26 +0200, Jason A. Donenfeld wrote:
> > This patch replaces calls to net_dbg_ratelimited when !DEBUG with
> > no_printk, keeping with the idiom of all the other debug print helpers.
> 
> Makes sense, thanks Jason.

Perhaps better still would be to use if (0) no_printk so that
the call and whatever argument calls the net_dbg_ratelimited
makes are completely eliminated.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1199410

From: Joe Perches <joe@perches.com>
Date: Mon, 03 Aug 2015 21:02:21 -0700

> On Mon, 2015-08-03 at 20:57 -0700, Joe Perches wrote:
>> On Tue, 2015-08-04 at 05:26 +0200, Jason A. Donenfeld wrote:
>> > This patch replaces calls to net_dbg_ratelimited when !DEBUG with
>> > no_printk, keeping with the idiom of all the other debug print helpers.
>> 
>> Makes sense, thanks Jason.
> 
> Perhaps better still would be to use if (0) no_printk so that
> the call and whatever argument calls the net_dbg_ratelimited
> makes are completely eliminated.

Agreed. Jason please respin your patch to work this way.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1199992

Sorry, I forgot to 'v2' the subject, but here's the respin suggested
by Joe and asked for by David.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web