Revert "net: Get rid of consume_skb when tracing is off"
Subsystems (2)
PATCH 1/1 Revert "net: Get rid of consume_skb when tracing is off"
This reverts commit be769db2f95861cc8c7c8fedcc71a8c39b803b10. There is trace_kfree_skb() in kfree_skb(), the trace_kfree_skb() is also a trace function.
--- include/linux/skbuff.h | 9 --------- net/core/skbuff.c | 2 -- 2 files changed, 11 deletions(-)
include/linux/skbuff.h +0 -9
@@ -1072,16 +1072,7 @@ void kfree_skb(struct sk_buff *skb);void kfree_skb_list (struct sk_buff *segs);void skb_dump (const char *level,const struct sk_buff *skb,bool full_pkt);void skb_tx_error (struct sk_buff *skb); - -#ifdef CONFIG_TRACEPOINTS void consume_skb (struct sk_buff *skb); -#else -static inline void consume_skb (struct sk_buff *skb) -{ -return kfree_skb (skb); -} -#endif -void __consume_stateless_skb (struct sk_buff *skb);void __kfree_skb (struct sk_buff *skb);extern struct kmem_cache *skbuff_head_cache;
net/core/skbuff.c +0 -2
@@ -893,7 +893,6 @@ void skb_tx_error(struct sk_buff *skb) }EXPORT_SYMBOL (skb_tx_error); -#ifdef CONFIG_TRACEPOINTS /** * consume_skb - free an skbuff * @skb: buffer to free
@@ -911,7 +910,6 @@ void consume_skb(struct sk_buff *skb)__kfree_skb (skb); }EXPORT_SYMBOL (consume_skb); -#endif /** * __consume_stateless_skb - free an skbuff, assuming it is stateless
Signature · 1 line
2.32.0
Quoted 6 lines · Yajun Deng
This reverts commit
be769db2f95861cc8c7c8fedcc71a8c39b803b10. There istrace_kfree_skb()inkfree_skb(), thetrace_kfree_skb()is also a trace function.Fixes be769db2f958 (net: Get rid of consume_skb when tracing is off)Signed-off-by Yajun Deng <[email protected]>
Please explain in more detail why your patch is needed. As it stands I do not understand the reasoning.
Thanks,
Signature · 3 lines
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Quoted 6 lines · Herbert Xu
Quoted 4 lines · Yajun Deng
This reverts commit
be769db2f95861cc8c7c8fedcc71a8c39b803b10. There istrace_kfree_skb()inkfree_skb(), thetrace_kfree_skb()is also a trace function.
Quoted 6 lines
Quoted 2 lines
Fixes be769db2f958 (net: Get rid of consume_skb when tracing is off)Signed-off-by Yajun Deng <[email protected]>Please explain in more detail why your patch is needed. As it stands I do not understand the reasoning.
if we don't define CONFIG_TRACEPOINTS, consume_skb() wolud called kfree_skb(), there have trace_kfree_skb() in kfree_skb(), the trace_kfree_skb() is also a trace function. So we can trace consume_skb() even if we don't define CONFIG_TRACEPOINTS.
This patch "net: Get rid of consume_skb when tracing is off" does not seem to be effective.
Quoted 5 lines
Thanks,
--
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Quoted 5 lines · On Thu, Jul 29, 2021 at 04:01:28AM +0000, [email protected] wrote:
if we don't define
CONFIG_TRACEPOINTS,consume_skb()wolud calledkfree_skb(), there havetrace_kfree_skb()inkfree_skb(), thetrace_kfree_skb()is also a trace function. So we can traceconsume_skb()even if we don't defineCONFIG_TRACEPOINTS.
This patch "net: Get rid ofconsume_skbwhen tracing is off" does not seem to be effective.
The point of my patch was to get rid of consume_skb because its only purpose is to provide extra information for tracing. If you're not tracing then you don't need that extra information (and overhead).
Cheers,
Signature · 3 lines
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Quoted 11 lines · Herbert Xu
Quoted 4 lines · On Thu, Jul 29, 2021 at 04:01:28AM +0000, [email protected] wrote:
if we don't define
CONFIG_TRACEPOINTS,consume_skb()wolud calledkfree_skb(), there havetrace_kfree_skb()inkfree_skb(), thetrace_kfree_skb()is also a trace function. So we can traceconsume_skb()even if we don't defineCONFIG_TRACEPOINTS.
This patch "net: Get rid ofconsume_skbwhen tracing is off" does not seem to be effective.The point of my patch was to get rid of
consume_skbbecause its only purpose is to provide extra information for tracing. If you're not tracing then you don't need that extra information (and overhead).
Ok, I didn't understand it well.
Quoted 5 lines
Cheers,
--
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt