Skip to contentSkip to conversation

tcp_cubic: fix the range of delayed_ack

PATCH In review Liu Yu 4 participants 4 messages Apr 30, 2014

Subsystems (3)
Networking Network drivers TCP
lore
Share & tools
lore

Liu Yu

author
#0 lore raw

PATCH 1/1 tcp_cubic: fix the range of delayed_ack

From: Liu Yu <[email protected]>

commit b9f47a3aaeab (tcp_cubic: limit delayed_ack ratio to prevent divide error) try to prevent divide error, but there is still a little chance that delayed_ack can reach zero. In case the param cnt get negative value, then ratio+cnt would overflow and may happen to be zero. As a result, min(ratio, ACK_RATIO_LIMIT) will calculate to be zero.

In some old kernels, such as 2.6.32, there is a bug that would pass negative param, which then ultimately leads to this divide error.

commit 5b35e1e6e9c (tcp: fix tcp_trim_head() to adjust segment count with skb MSS) fixed the negative param issue. However, it's safe that we fix the range of delayed_ack as well, to make sure we do not hit a divide by zero.

CC Stephen Hemminger <[email protected]>
Signed-off-by Liu Yu <[email protected]>
---
 net/ipv4/tcp_cubic.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
net/ipv4/tcp_cubic.c +1 -1
index 8bf2245..b4f1b29 100644
@@ -409,7 +409,7 @@ static void bictcp_acked(struct sock *sk, u32 cnt, s32 rtt_us)
 		ratio -= ca->delayed_ack >> ACK_RATIO_SHIFT;
 		ratio += cnt;
 
-		ca->delayed_ack = min(ratio, ACK_RATIO_LIMIT);
+		ca->delayed_ack = clamp(ratio, 1U, ACK_RATIO_LIMIT);
 	}
 
 	/* Some calls are for duplicates without timetamps */
Signature · 1 line
1.7.1
#1 lore raw
Quoted 16 lines · Liu Yu

From: Liu Yu <[email protected]>

commit b9f47a3aaeab (tcp_cubic: limit delayed_ack ratio to prevent divide error) try to prevent divide error, but there is still a little chance that delayed_ack can reach zero. In case the param cnt get negative value, then ratio+cnt would overflow and may happen to be zero. As a result, min(ratio, ACK_RATIO_LIMIT) will calculate to be zero.

In some old kernels, such as 2.6.32, there is a bug that would pass negative param, which then ultimately leads to this divide error.

commit 5b35e1e6e9c (tcp: fix tcp_trim_head() to adjust segment count with skb MSS) fixed the negative param issue. However, it's safe that we fix the range of delayed_ack as well, to make sure we do not hit a divide by zero.

Quoted 5 lines
CC Stephen Hemminger <[email protected]>
Signed-off-by Liu Yu <[email protected]>
---
 net/ipv4/tcp_cubic.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
Signed-off-by Eric Dumazet <[email protected]>

Thanks !

maintainer Acked-by
#2 lore raw
Quoted 17 lines · Eric Dumazet
Quoted 16 lines · Liu Yu

From: Liu Yu <[email protected]>

commit b9f47a3aaeab (tcp_cubic: limit delayed_ack ratio to prevent divide error) try to prevent divide error, but there is still a little chance that delayed_ack can reach zero. In case the param cnt get negative value, then ratio+cnt would overflow and may happen to be zero. As a result, min(ratio, ACK_RATIO_LIMIT) will calculate to be zero.

In some old kernels, such as 2.6.32, there is a bug that would pass negative param, which then ultimately leads to this divide error.

commit 5b35e1e6e9c (tcp: fix tcp_trim_head() to adjust segment count with skb MSS) fixed the negative param issue. However, it's safe that we fix the range of delayed_ack as well, to make sure we do not hit a divide by zero.

Quoted 7 lines
Quoted 5 lines
CC Stephen Hemminger <[email protected]>
Signed-off-by Liu Yu <[email protected]>
---
 net/ipv4/tcp_cubic.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
Signed-off-by Eric Dumazet <[email protected]>
Acked-by Neal Cardwell <[email protected]>

Indeed, thanks!

neal

maintainer
#3 lore raw
In reply to Liu Yu

From: Liu Yu <[email protected]>
Date: Wed, 30 Apr 2014 17:34:09 +0800

Quoted 12 lines

From: Liu Yu <[email protected]>

commit b9f47a3aaeab (tcp_cubic: limit delayed_ack ratio to prevent divide error) try to prevent divide error, but there is still a little chance that delayed_ack can reach zero. In case the param cnt get negative value, then ratio+cnt would overflow and may happen to be zero. As a result, min(ratio, ACK_RATIO_LIMIT) will calculate to be zero.

In some old kernels, such as 2.6.32, there is a bug that would pass negative param, which then ultimately leads to this divide error.

Quoted 7 lines

commit 5b35e1e6e9c (tcp: fix tcp_trim_head() to adjust segment count with skb MSS) fixed the negative param issue. However, it's safe that we fix the range of delayed_ack as well, to make sure we do not hit a divide by zero.

CC Stephen Hemminger <[email protected]>
Signed-off-by Liu Yu <[email protected]>

Applied and queued up for -stable, thanks.