This patch removes unnecessary logspam which resulted from superfluous calls to net_ratelimit(). With the supplied patch, net_ratelimit() is called after the loglevel has been checked.
Signed-off-by: André Gaul gaul@web-yard.de --- main.h | 16 ++++++++++++---- routing.c | 4 ++-- 2 files changed, 14 insertions(+), 6 deletions(-)
diff --git a/main.h b/main.h index 87e7196..df57639 100644 --- a/main.h +++ b/main.h @@ -239,21 +239,29 @@ enum batadv_dbg_level { int batadv_debug_log(struct batadv_priv *bat_priv, const char *fmt, ...) __printf(2, 3);
-#define batadv_dbg(type, bat_priv, fmt, arg...) \ +/* possibly ratelimited debug output */ +#define _batadv_dbg(type, bat_priv, ratelimited, fmt, arg...) \ do { \ - if (atomic_read(&bat_priv->log_level) & type) \ + if (atomic_read(&bat_priv->log_level) & type && \ + (!ratelimited || net_ratelimit())) \ batadv_debug_log(bat_priv, fmt, ## arg);\ } \ while (0) #else /* !CONFIG_BATMAN_ADV_DEBUG */ -__printf(3, 4) -static inline void batadv_dbg(int type __always_unused, +__printf(4, 5) +static inline void _batadv_dbg(int type __always_unused, struct batadv_priv *bat_priv __always_unused, + int ratelimited __always_unused, const char *fmt __always_unused, ...) { } #endif
+#define batadv_dbg(type, bat_priv, arg...) \ + _batadv_dbg(type, bat_priv, 0, ## arg) +#define batadv_dbg_ratelimited(type, bat_priv, arg...) \ + _batadv_dbg(type, bat_priv, 1, ## arg) + #define batadv_info(net_dev, fmt, arg...) \ do { \ struct net_device *_netdev = (net_dev); \ diff --git a/routing.c b/routing.c index 3514153..c679c6f 100644 --- a/routing.c +++ b/routing.c @@ -706,7 +706,7 @@ static int batadv_check_unicast_ttvn(struct batadv_priv *bat_priv, if (batadv_tt_local_client_is_roaming(bat_priv, ethhdr->h_dest, vid)) { if (batadv_reroute_unicast_packet(bat_priv, unicast_packet, ethhdr->h_dest, vid)) - net_ratelimited_function(batadv_dbg, BATADV_DBG_TT, + batadv_dbg_ratelimited(BATADV_DBG_TT, bat_priv, "Rerouting unicast packet to %pM (dst=%pM): Local Roaming\n", unicast_packet->dest, @@ -752,7 +752,7 @@ static int batadv_check_unicast_ttvn(struct batadv_priv *bat_priv, */ if (batadv_reroute_unicast_packet(bat_priv, unicast_packet, ethhdr->h_dest, vid)) { - net_ratelimited_function(batadv_dbg, BATADV_DBG_TT, bat_priv, + batadv_dbg_ratelimited(BATADV_DBG_TT, bat_priv, "Rerouting unicast packet to %pM (dst=%pM): TTVN mismatch old_ttvn=%u new_ttvn=%u\n", unicast_packet->dest, ethhdr->h_dest, old_ttvn, curr_ttvn);
Hi André,
On 04/06/14 21:53, André Gaul wrote:
This patch removes unnecessary logspam which resulted from superfluous calls to net_ratelimit(). With the supplied patch, net_ratelimit() is called after the loglevel has been checked.
good catch!
-#define batadv_dbg(type, bat_priv, fmt, arg...) \ +/* possibly ratelimited debug output */ +#define _batadv_dbg(type, bat_priv, ratelimited, fmt, arg...) \
can you please put the ending "" at the same position of the ones below ? (This way the entire macro definition looks better).
do { \
if (atomic_read(&bat_priv->log_level) & type) \
if (atomic_read(&bat_priv->log_level) & type && \(!ratelimited || net_ratelimit())) \
same here
batadv_debug_log(bat_priv, fmt, ## arg);\} \ while (0) #else /* !CONFIG_BATMAN_ADV_DEBUG */ -__printf(3, 4) -static inline void batadv_dbg(int type __always_unused, +__printf(4, 5) +static inline void _batadv_dbg(int type __always_unused, struct batadv_priv *bat_priv __always_unused,
int ratelimited __always_unused, const char *fmt __always_unused, ...){ } #endif
+#define batadv_dbg(type, bat_priv, arg...) \
- _batadv_dbg(type, bat_priv, 0, ## arg)
+#define batadv_dbg_ratelimited(type, bat_priv, arg...) \
- _batadv_dbg(type, bat_priv, 1, ## arg)
#define batadv_info(net_dev, fmt, arg...) \ do { \ struct net_device *_netdev = (net_dev); \ diff --git a/routing.c b/routing.c index 3514153..c679c6f 100644 --- a/routing.c +++ b/routing.c @@ -706,7 +706,7 @@ static int batadv_check_unicast_ttvn(struct batadv_priv *bat_priv, if (batadv_tt_local_client_is_roaming(bat_priv, ethhdr->h_dest, vid)) { if (batadv_reroute_unicast_packet(bat_priv, unicast_packet, ethhdr->h_dest, vid))
net_ratelimited_function(batadv_dbg, BATADV_DBG_TT,
batadv_dbg_ratelimited(BATADV_DBG_TT, bat_priv, "Rerouting unicast packet to %pM (dst=%pM): Local Roaming\n", unicast_packet->dest,
The arguments on a new line must be lined up to the first column after the opening parenthesis. (I think that checkpatch would have suggested the same)
@@ -752,7 +752,7 @@ static int batadv_check_unicast_ttvn(struct batadv_priv *bat_priv, */ if (batadv_reroute_unicast_packet(bat_priv, unicast_packet, ethhdr->h_dest, vid)) {
net_ratelimited_function(batadv_dbg, BATADV_DBG_TT, bat_priv,
batadv_dbg_ratelimited(BATADV_DBG_TT, bat_priv, "Rerouting unicast packet to %pM (dst=%pM): TTVN mismatch old_ttvn=%u new_ttvn=%u\n", unicast_packet->dest, ethhdr->h_dest, old_ttvn, curr_ttvn);
same here.
The rest looks good!
This patch removes unnecessary logspam which resulted from superfluous calls to net_ratelimit(). With the supplied patch, net_ratelimit() is called after the loglevel has been checked.
Signed-off-by: André Gaul gaul@web-yard.de --- main.h | 20 ++++++++++++++------ routing.c | 18 +++++++++--------- 2 files changed, 23 insertions(+), 15 deletions(-)
diff --git a/main.h b/main.h index 87e7196..4bb500b 100644 --- a/main.h +++ b/main.h @@ -239,21 +239,29 @@ enum batadv_dbg_level { int batadv_debug_log(struct batadv_priv *bat_priv, const char *fmt, ...) __printf(2, 3);
-#define batadv_dbg(type, bat_priv, fmt, arg...) \ +/* possibly ratelimited debug output */ +#define _batadv_dbg(type, bat_priv, ratelimited, fmt, arg...) \ do { \ - if (atomic_read(&bat_priv->log_level) & type) \ + if (atomic_read(&bat_priv->log_level) & type && \ + (!ratelimited || net_ratelimit())) \ batadv_debug_log(bat_priv, fmt, ## arg);\ } \ while (0) #else /* !CONFIG_BATMAN_ADV_DEBUG */ -__printf(3, 4) -static inline void batadv_dbg(int type __always_unused, - struct batadv_priv *bat_priv __always_unused, - const char *fmt __always_unused, ...) +__printf(4, 5) +static inline void _batadv_dbg(int type __always_unused, + struct batadv_priv *bat_priv __always_unused, + int ratelimited __always_unused, + const char *fmt __always_unused, ...) { } #endif
+#define batadv_dbg(type, bat_priv, arg...) \ + _batadv_dbg(type, bat_priv, 0, ## arg) +#define batadv_dbg_ratelimited(type, bat_priv, arg...) \ + _batadv_dbg(type, bat_priv, 1, ## arg) + #define batadv_info(net_dev, fmt, arg...) \ do { \ struct net_device *_netdev = (net_dev); \ diff --git a/routing.c b/routing.c index 3514153..35f76f2 100644 --- a/routing.c +++ b/routing.c @@ -706,11 +706,11 @@ static int batadv_check_unicast_ttvn(struct batadv_priv *bat_priv, if (batadv_tt_local_client_is_roaming(bat_priv, ethhdr->h_dest, vid)) { if (batadv_reroute_unicast_packet(bat_priv, unicast_packet, ethhdr->h_dest, vid)) - net_ratelimited_function(batadv_dbg, BATADV_DBG_TT, - bat_priv, - "Rerouting unicast packet to %pM (dst=%pM): Local Roaming\n", - unicast_packet->dest, - ethhdr->h_dest); + batadv_dbg_ratelimited(BATADV_DBG_TT, + bat_priv, + "Rerouting unicast packet to %pM (dst=%pM): Local Roaming\n", + unicast_packet->dest, + ethhdr->h_dest); /* at this point the mesh destination should have been * substituted with the originator address found in the global * table. If not, let the packet go untouched anyway because @@ -752,10 +752,10 @@ static int batadv_check_unicast_ttvn(struct batadv_priv *bat_priv, */ if (batadv_reroute_unicast_packet(bat_priv, unicast_packet, ethhdr->h_dest, vid)) { - net_ratelimited_function(batadv_dbg, BATADV_DBG_TT, bat_priv, - "Rerouting unicast packet to %pM (dst=%pM): TTVN mismatch old_ttvn=%u new_ttvn=%u\n", - unicast_packet->dest, ethhdr->h_dest, - old_ttvn, curr_ttvn); + batadv_dbg_ratelimited(BATADV_DBG_TT, bat_priv, + "Rerouting unicast packet to %pM (dst=%pM): TTVN mismatch old_ttvn=%u new_ttvn=%u\n", + unicast_packet->dest, ethhdr->h_dest, + old_ttvn, curr_ttvn); return 1; }
On 09/06/14 19:14, André Gaul wrote:
This patch removes unnecessary logspam which resulted from superfluous calls to net_ratelimit(). With the supplied patch, net_ratelimit() is called after the loglevel has been checked.
Signed-off-by: André Gaul gaul@web-yard.de
André,
I'd suggest you to check this patch with checkpatch.pl --strict. Here I get many errors due to non-printable chars at the end of various lines.
Here[1] you find our guidelines (which is a set of complementary rules to the guidelines you find in the kernel tree[2]).
Moreover, the subject must start with "batman-adv: ", because this patch will end up in the kernel and it requires this format to easily distinguish the subsystem that the patch is touching.
A couple of things to keep in mind while sending new versions of the same patch: - don't reply to an existing thread - write [PATCHvN] (with N = version of the patch - 2 in this case) in the subject
Cheers,
[1] http://www.open-mesh.org/projects/open-mesh/wiki/Contribute#Submitting-patch... [2] https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/Documen...
Hey Antonio,
thx for your quick reply!
Am 09.06.2014 23:05, schrieb Antonio Quartulli:
I'd suggest you to check this patch with checkpatch.pl --strict. Here I get many errors due to non-printable chars at the end of various lines.
I did that and I don't get any errors or warnings with checkpatch.pl --strict. It tells me:
------ $ ./scripts/checkpatch.pl --strict /tmp/0001-remove-unnecessary-logspam.patch total: 0 errors, 0 warnings, 65 lines checked
/tmp/0001-remove-unnecessary-logspam.patch has no obvious style problems and is ready for submission. ------
Here[1] you find our guidelines (which is a set of complementary rules to the guidelines you find in the kernel tree[2]).
Moreover, the subject must start with "batman-adv: ", because this patch will end up in the kernel and it requires this format to easily distinguish the subsystem that the patch is touching.
A couple of things to keep in mind while sending new versions of the same patch:
- don't reply to an existing thread
- write [PATCHvN] (with N = version of the patch - 2 in this case) in
the subject
OK, I can resubmit with the corrected subject line [PATCHv2] batman-adv: remove unnecessary logspam but I'd like to know first what this non-printable chars issue is about.
cheers, André
Hi André,
On 09/06/14 23:36, André Gaul wrote:
Hey Antonio,
thx for your quick reply!
Am 09.06.2014 23:05, schrieb Antonio Quartulli:
I'd suggest you to check this patch with checkpatch.pl --strict. Here I get many errors due to non-printable chars at the end of various lines.
This is my output (this is checkpatch from net-next) - you can ignore the warnings about "Macros with complex values should be enclosed in parenthesis":
$ ./exp/linux-merge/scripts/checkpatch.pl --strict [B.A.T.M.A.N.]\ [PATCH]\ remove\ unnecessary\ logspam.eml ERROR: DOS line endings #111: FILE: main.h:242: +/* possibly ratelimited debug output */^M$
ERROR: DOS line endings #112: FILE: main.h:243: +#define _batadv_dbg(type, bat_priv, ratelimited, fmt, arg...)^I^M$
WARNING: Whitespace after \ makes next lines useless #112: FILE: main.h:243: +#define _batadv_dbg(type, bat_priv, ratelimited, fmt, arg...) \
ERROR: Macros with complex values should be enclosed in parenthesis #112: FILE: main.h:243: +#define _batadv_dbg(type, bat_priv, ratelimited, fmt, arg...) \
ERROR: DOS line endings #115: FILE: main.h:245: +^I^Iif (atomic_read(&bat_priv->log_level) & type && ^M$
ERROR: DOS line endings #116: FILE: main.h:246: +^I^I (!ratelimited || net_ratelimit()))^I^I^M$
ERROR: DOS line endings #125: FILE: main.h:251: +__printf(4, 5)^M$
ERROR: DOS line endings #126: FILE: main.h:252: +static inline void _batadv_dbg(int type __always_unused,^M$
ERROR: DOS line endings #127: FILE: main.h:253: +^I^I^I struct batadv_priv *bat_priv __always_unused,^M$
ERROR: DOS line endings #128: FILE: main.h:254: +^I^I^I int ratelimited __always_unused,^M$
ERROR: DOS line endings #129: FILE: main.h:255: +^I^I^I const char *fmt __always_unused, ...)^M$
ERROR: DOS line endings #134: FILE: main.h:260: +#define batadv_dbg(type, bat_priv, arg...) ^M$
WARNING: Whitespace after \ makes next lines useless #134: FILE: main.h:260: +#define batadv_dbg(type, bat_priv, arg...) \
ERROR: Macros with complex values should be enclosed in parenthesis #134: FILE: main.h:260: +#define batadv_dbg(type, bat_priv, arg...) \
ERROR: DOS line endings #135: FILE: main.h:261: +^I_batadv_dbg(type, bat_priv, 0, ## arg)^M$
ERROR: DOS line endings #136: FILE: main.h:262: +#define batadv_dbg_ratelimited(type, bat_priv, arg...) ^M$
WARNING: Whitespace after \ makes next lines useless #136: FILE: main.h:262: +#define batadv_dbg_ratelimited(type, bat_priv, arg...) \
ERROR: Macros with complex values should be enclosed in parenthesis #136: FILE: main.h:262: +#define batadv_dbg_ratelimited(type, bat_priv, arg...) \
ERROR: DOS line endings #137: FILE: main.h:263: +^I_batadv_dbg(type, bat_priv, 1, ## arg)^M$
ERROR: DOS line endings #138: FILE: main.h:264: +^M$
ERROR: DOS line endings #155: FILE: routing.c:709: +^I^I^Ibatadv_dbg_ratelimited(BATADV_DBG_TT,^M$
ERROR: DOS line endings #156: FILE: routing.c:710: +^I^I^I^I^I bat_priv,^M$
ERROR: DOS line endings #157: FILE: routing.c:711: +^I^I^I^I^I "Rerouting unicast packet to %pM (dst=%pM): Local Roaming\n",^M$
ERROR: DOS line endings #158: FILE: routing.c:712: +^I^I^I^I^I unicast_packet->dest,^M$
ERROR: DOS line endings #159: FILE: routing.c:713: +^I^I^I^I^I ethhdr->h_dest);^M$
ERROR: DOS line endings #171: FILE: routing.c:755: +^I^Ibatadv_dbg_ratelimited(BATADV_DBG_TT, bat_priv,^M$
ERROR: DOS line endings #172: FILE: routing.c:756: +^I^I^I^I "Rerouting unicast packet to %pM (dst=%pM): TTVN mismatch old_ttvn=%u new_ttvn=%u\n",^M$
ERROR: DOS line endings #173: FILE: routing.c:757: +^I^I^I^I unicast_packet->dest, ethhdr->h_dest,^M$
ERROR: DOS line endings #174: FILE: routing.c:758: +^I^I^I^I old_ttvn, curr_ttvn);^M$
total: 26 errors, 3 warnings, 0 checks, 65 lines checked
[B.A.T.M.A.N.] [PATCH] remove unnecessary logspam.eml has style problems, please review.
If any of these errors are false positives, please report them to the maintainer, see CHECKPATCH in MAINTAINERS.
maybe it's the mail client that messes the patch up? Are you using git-send-email ?
Cheers,
Hi Antonio!
Am 10.06.2014 08:01, schrieb Antonio Quartulli:
maybe it's the mail client that messes the patch up? Are you using git-send-email ?
Yep, it's git-send-email. Maybe you are running into this mbox/eml issue [1]? Thunderbird seems to save plain text files with DOS line endings when saved as .eml. I can reproduce the errors you got with an .eml file. When saving as .txt it has the correct line endings and the checkpatch script gives no errors or warnings.
If this resolves the issue with the DOS line endings, I'd resubmit with the corrected subject line.
Cheers, André
[1] https://bugzilla.mozilla.org/show_bug.cgi?id=523577
Hi André,
On 10/06/14 17:38, André Gaul wrote:
Hi Antonio!
Am 10.06.2014 08:01, schrieb Antonio Quartulli:
maybe it's the mail client that messes the patch up? Are you using git-send-email ?
Yep, it's git-send-email. Maybe you are running into this mbox/eml issue [1]? Thunderbird seems to save plain text files with DOS line endings when saved as .eml. I can reproduce the errors you got with an .eml file. When saving as .txt it has the correct line endings and the checkpatch script gives no errors or warnings.
If this resolves the issue with the DOS line endings, I'd resubmit with the corrected subject line.
Oh you are right. I should have continued using mutt...
Thanks! :)
Am 10.06.2014 17:40, schrieb Antonio Quartulli:
Oh you are right. I should have continued using mutt...
hehe, I'm glad it works now! :) I'll resubmit in a minute as a new thread.
btw: do you have any pointers to documentation that can help me understand why the kernel developers still use email as the primary channel for submitting patches? Given the fact that git gracefully handles the transmission of code, the handling of all code via email appears archaic and too complicated to me. (no flame war intended, I just want to learn about the reasons!) ;)
Cheers, André
On 10/06/14 17:45, André Gaul wrote:
Am 10.06.2014 17:40, schrieb Antonio Quartulli:
Oh you are right. I should have continued using mutt...
hehe, I'm glad it works now! :) I'll resubmit in a minute as a new thread.
btw: do you have any pointers to documentation that can help me understand why the kernel developers still use email as the primary channel for submitting patches? Given the fact that git gracefully handles the transmission of code, the handling of all code via email appears archaic and too complicated to me. (no flame war intended, I just want to learn about the reasons!) ;)
Honestly I don't know if there is any documentation.
Sending a patch to a mailing list is a way to ask for a review/evaluation, which works well so I don't see a clear reason why it should be changed (just that it appears archaic does not seem to be a good reason).
Only maintainer having their tree on a trusted host can send pull requests without attaching the patches (we do send pull requests along with the patches exactly for this reason..).
And actually git also provides many comfortable commands to do that :) so...! :)
Cheers,
On Tuesday 10 June 2014 17:45:17 André Gaul wrote:
btw: do you have any pointers to documentation that can help me understand why the kernel developers still use email as the primary channel for submitting patches? Given the fact that git gracefully handles the transmission of code, the handling of all code via email appears archaic and too complicated to me. (no flame war intended, I just want to learn about the reasons!) ;)
Traditionally, patches on the mailing list make it easy to review & comment patches by everyone.
What is the alternative you have in mind (no vendor lock-in please) ?
Cheers, Marek
Hi Marek,
Am 14.06.2014 03:44, schrieb Marek Lindner:
Traditionally, patches on the mailing list make it easy to review & comment patches by everyone.
What is the alternative you have in mind (no vendor lock-in please) ?
I got used to simple webuis for reviews+comments à la github or bitbucket. I can fully understand that kernel developers do not want to depend on a company such as github. The only free software alternative that comes to my mind is gitlab but I don't have any experience with gitlab and big projects.
Patches on the mailing list make it easy for people who already do so for some time. E.g., if you're not on the list, then it's not that easy to apply a submitted patch (you'll have to copy+paste the patch). Also the context of a patch is missing in the mail: you don't see the full patched version and every potential reviewer has to apply the patch her/himself, thus collectively wasting time for a task that can be fully automated. Furthermore, we've just seen in this thread how vulnerable the procedure is to misbehaving mail clients. ;) This also wasted some of my and Antonio's time.
As said before: I just wanted to know about the reasons -- refusing vendor lock-ins certainly is a reason and I can agree on trading a bit of comfort for independence. :)
Cheers and keep up the good work on B.A.T.M.A.N., André
b.a.t.m.a.n@lists.open-mesh.org
-
André Gaul -
Antonio Quartulli -
Marek Lindner