The net_xmit_eval has side effects because it is not making sure that e isn't evaluated twice.
#define net_xmit_eval(e) ((e) == NET_XMIT_CN ? 0 : (e))
The code proposed by David Miller
return net_xmit_eval(dev_queue_xmit(skb));
will therefore get transformed into
return ((dev_queue_xmit(skb)) == NET_XMIT_CN ? 0 : (dev_queue_xmit(skb)))
dev_queue_xmit will therefore be tried again (with an already consumed skb) whenever the return code is not NET_XMIT_CN.
Fixes: 17efab9867c5 ("batman-adv: Simplify handling of NET_XMIT_CN") Signed-off-by: Sven Eckelmann sven@narfation.org --- net/batman-adv/send.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c index d9b28890..1489ec27 100644 --- a/net/batman-adv/send.c +++ b/net/batman-adv/send.c @@ -77,6 +77,7 @@ int batadv_send_skb_packet(struct sk_buff *skb, { struct batadv_priv *bat_priv; struct ethhdr *ethhdr; + int ret;
bat_priv = netdev_priv(hard_iface->soft_iface);
@@ -115,7 +116,8 @@ int batadv_send_skb_packet(struct sk_buff *skb, * congestion and traffic shaping, it drops and returns NET_XMIT_DROP * (which is > 0). This will not be treated as an error. */ - return net_xmit_eval(dev_queue_xmit(skb)); + ret = dev_queue_xmit(skb); + return net_xmit_eval(ret); send_skb_err: kfree_skb(skb); return NET_XMIT_DROP;
IS_ERR/ERR_PTR are not defined in linux/device.h but in linux/err.h. The files using these macros therefore have to include the correct one.
Reported-by: Linus Luessing linus.luessing@web.de Signed-off-by: Sven Eckelmann sven@narfation.org --- net/batman-adv/debugfs.c | 2 +- net/batman-adv/tp_meter.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/batman-adv/debugfs.c b/net/batman-adv/debugfs.c index 5406148b..e32ad47c 100644 --- a/net/batman-adv/debugfs.c +++ b/net/batman-adv/debugfs.c @@ -19,7 +19,7 @@ #include "main.h"
#include <linux/debugfs.h> -#include <linux/device.h> +#include <linux/err.h> #include <linux/errno.h> #include <linux/export.h> #include <linux/fs.h> diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c index 07f64b60..c94ebdec 100644 --- a/net/batman-adv/tp_meter.c +++ b/net/batman-adv/tp_meter.c @@ -23,7 +23,7 @@ #include <linux/byteorder/generic.h> #include <linux/cache.h> #include <linux/compiler.h> -#include <linux/device.h> +#include <linux/err.h> #include <linux/etherdevice.h> #include <linux/fs.h> #include <linux/if_ether.h>
On Sat, Jan 28, 2017 at 10:25:26AM +0100, Sven Eckelmann wrote:
IS_ERR/ERR_PTR are not defined in linux/device.h but in linux/err.h. The files using these macros therefore have to include the correct one.
Reported-by: Linus Luessing linus.luessing@web.de Signed-off-by: Sven Eckelmann sven@narfation.org
Hm, my @web.de mail address is deprecated (read-only, don't ask me why :D).
Would be nice if it could be updated upon committing. (but if someone forgets not a big problem either, as it's documented in .mailmap)
On Sat, Jan 28, 2017 at 10:22:49PM +0100, Linus Lüssing wrote:
On Sat, Jan 28, 2017 at 10:25:26AM +0100, Sven Eckelmann wrote:
IS_ERR/ERR_PTR are not defined in linux/device.h but in linux/err.h. The files using these macros therefore have to include the correct one.
Reported-by: Linus Luessing linus.luessing@web.de Signed-off-by: Sven Eckelmann sven@narfation.org
Hm, my @web.de mail address is deprecated (read-only, don't ask me why :D).
Would be nice if it could be updated upon committing. (but if someone forgets not a big problem either, as it's documented in .mailmap)
Ah, there's a PR already for David - then just forget my last mail :).
On Samstag, 28. Januar 2017 10:25:25 CET Sven Eckelmann wrote:
The net_xmit_eval has side effects because it is not making sure that e isn't evaluated twice.
#define net_xmit_eval(e) ((e) == NET_XMIT_CN ? 0 : (e))The code proposed by David Miller
return net_xmit_eval(dev_queue_xmit(skb));will therefore get transformed into
return ((dev_queue_xmit(skb)) == NET_XMIT_CN ? 0 : (dev_queue_xmit(skb)))dev_queue_xmit will therefore be tried again (with an already consumed skb) whenever the return code is not NET_XMIT_CN.
Fixes: 17efab9867c5 ("batman-adv: Simplify handling of NET_XMIT_CN") Signed-off-by: Sven Eckelmann sven@narfation.org
net/batman-adv/send.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
I was so embarrassed about not catching these two that I've already applied them an prepared the changes for net-next.
Kind regards, Sven
b.a.t.m.a.n@lists.open-mesh.org
-
Linus Lüssing -
Sven Eckelmann