tcp: Export tcp_write_queue_purge()
Subsystems (3)
PATCH 1/1 tcp: Export tcp_write_queue_purge()
After tcp_write_queue_purge() got uninlined with commit ac3f09ba3e49 ("tcp: uninline tcp_write_queue_purge()"), it became no longer possible to reference this symbol from kernel modules.
--- net/ipv4/tcp.c | 1 + 1 file changed, 1 insertion(+)
net/ipv4/tcp.c +1 -0
@@ -2626,6 +2626,7 @@ void tcp_write_queue_purge(struct sock *sk)tcp_sk (sk)->packets_out =0 ;inet_csk (sk)->icsk_backoff =0 ; } +EXPORT_SYMBOL (tcp_write_queue_purge);int tcp_disconnect (struct sock *sk,int flags) {
Signature · 1 line
2.17.1
Quoted 24 lines · Florian Fainelli
After
tcp_write_queue_purge()got uninlined with commitac3f09ba3e49("tcp: uninlinetcp_write_queue_purge()"), it became no longer possible to reference this symbol from kernel modules.Fixes ac3f09ba3e49 ("tcp: uninline tcp_write_queue_purge()")Signed-off-by Florian Fainelli <[email protected]>--- net/ipv4/tcp.c | 1 + 1 file changed, 1 insertion(+)net/ipv4/tcp.c +1
-1@@ -2626,6 +2626,7 @@ void tcp_write_queue_purge(struct sock *sk)tcp_sk (sk)->packets_out =0 ;inet_csk (sk)->icsk_backoff =0 ; } +EXPORT_SYMBOL (tcp_write_queue_purge);int tcp_disconnect (struct sock *sk,int flags) { --
Quoted 2 lines
2.17.1
Hmmm.... which module would need this exactly ?
How come it took 3 years to discover this issue ?
Quoted 22 lines · Eric Dumazet
Quoted 21 lines · Florian Fainelli
After
tcp_write_queue_purge()got uninlined with commitac3f09ba3e49("tcp: uninlinetcp_write_queue_purge()"), it became no longer possible to reference this symbol from kernel modules.Fixes ac3f09ba3e49 ("tcp: uninline tcp_write_queue_purge()")Signed-off-by Florian Fainelli <[email protected]>--- net/ipv4/tcp.c | 1 + 1 file changed, 1 insertion(+)net/ipv4/tcp.c +1
-0@@ -2626,6 +2626,7 @@ void tcp_write_queue_purge(struct sock *sk)tcp_sk (sk)->packets_out =0 ;inet_csk (sk)->icsk_backoff =0 ; } +EXPORT_SYMBOL (tcp_write_queue_purge);
Quoted 7 lines
Quoted 5 lines
net/ipv4/tcp.c +0
-1int tcp_disconnect (struct sock *sk,int flags) { --2.17.1
Hmmm.... which module would need this exactly ?
None in tree unfortunately, and I doubt it would be published one day. For consistency one could argue that given it used to be accessible, and other symbols within net/ipv4/tcp.c are also exported, so this should one be. Not going to hold that line of argumentation more than in this email, if you object to it, that would be completely fine with me.
Quoted 2 lines
How come it took 3 years to discover this issue ?
We just upgraded our downstream kernel from 4.9 to 5.4 and this is why it took so long.
Signature · 1 line
Florian
David Miller
maintainerFrom: Florian Fainelli <[email protected]>
Date: Thu, 30 Jul 2020 14:23:50 -0700
Quoted 11 lines · Eric Dumazet
Quoted 2 lines · Florian Fainelli
Hmmm.... which module would need this exactly ?
None in tree unfortunately, and I doubt it would be published one day. For consistency one could argue that given it used to be accessible, and other symbols within
net/ipv4/tcp.care also exported, so this should one be. Not going to hold that line of argumentation more than in this email, if you object to it, that would be completely fine with me.Quoted 1 line
Quoted 4 lines
Quoted 1 line
How come it took 3 years to discover this issue ?
We just upgraded our downstream kernel from 4.9 to 5.4 and this is why it took so long.
We really can't do this, sorry.
Quoted 32 lines · Florian Fainelli
Quoted 22 lines · Eric Dumazet
Quoted 21 lines · Florian Fainelli
After
tcp_write_queue_purge()got uninlined with commitac3f09ba3e49("tcp: uninlinetcp_write_queue_purge()"), it became no longer possible to reference this symbol from kernel modules.Fixes ac3f09ba3e49 ("tcp: uninline tcp_write_queue_purge()")Signed-off-by Florian Fainelli <[email protected]>--- net/ipv4/tcp.c | 1 + 1 file changed, 1 insertion(+)net/ipv4/tcp.c +1
-0@@ -2626,6 +2626,7 @@ void tcp_write_queue_purge(struct sock *sk)tcp_sk (sk)->packets_out =0 ;inet_csk (sk)->icsk_backoff =0 ; } +EXPORT_SYMBOL (tcp_write_queue_purge);Quoted 7 lines
Quoted 5 lines
net/ipv4/tcp.c +0
-1int tcp_disconnect (struct sock *sk,int flags) { --2.17.1
Hmmm.... which module would need this exactly ?
Quoted 5 lines
None in tree unfortunately, and I doubt it would be published one day. For consistency one could argue that given it used to be accessible, and other symbols within
net/ipv4/tcp.care also exported, so this should one be. Not going to hold that line of argumentation more than in this email, if you object to it, that would be completely fine with me.
:)
Quoted 6 lines
Quoted 3 lines
How come it took 3 years to discover this issue ?
We just upgraded our downstream kernel from 4.9 to 5.4 and this is why it took so long.
It is not because TCP used an inline function in the past that it means we have to keep
the equivalent function available for all possible out-of-tree modules.
Sorry, we can not accept that out-of-tree modules use TCP stack like that.
You will have to carry this change locally. Or even better get rid of it.
Quoted 48 lines · Eric Dumazet
Quoted 32 lines · Florian Fainelli
Quoted 22 lines · Eric Dumazet
Quoted 21 lines · Florian Fainelli
After
tcp_write_queue_purge()got uninlined with commitac3f09ba3e49("tcp: uninlinetcp_write_queue_purge()"), it became no longer possible to reference this symbol from kernel modules.Fixes ac3f09ba3e49 ("tcp: uninline tcp_write_queue_purge()")Signed-off-by Florian Fainelli <[email protected]>--- net/ipv4/tcp.c | 1 + 1 file changed, 1 insertion(+)net/ipv4/tcp.c +1
-0@@ -2626,6 +2626,7 @@ void tcp_write_queue_purge(struct sock *sk)tcp_sk (sk)->packets_out =0 ;inet_csk (sk)->icsk_backoff =0 ; } +EXPORT_SYMBOL (tcp_write_queue_purge);Quoted 7 lines
Quoted 5 lines
net/ipv4/tcp.c +0
-1int tcp_disconnect (struct sock *sk,int flags) { --2.17.1
Hmmm.... which module would need this exactly ?
Quoted 5 lines
None in tree unfortunately, and I doubt it would be published one day. For consistency one could argue that given it used to be accessible, and other symbols within
net/ipv4/tcp.care also exported, so this should one be. Not going to hold that line of argumentation more than in this email, if you object to it, that would be completely fine with me.:)
Quoted 6 lines
Quoted 3 lines
How come it took 3 years to discover this issue ?
We just upgraded our downstream kernel from 4.9 to 5.4 and this is why it took so long.
Quoted 7 lines
It is not because TCP used an inline function in the past that it means we have to keep
the equivalent function available for all possible out-of-tree modules.Sorry, we can not accept that out-of-tree modules use TCP stack like that.
You will have to carry this change locally. Or even better get rid of it.
Sure, that is completely fair, I had to try though :)
Signature · 1 line
Florian