diff options
| author | Eric Dumazet <edumazet@google.com> | 2021-12-04 07:53:56 +0300 |
|---|---|---|
| committer | Jakub Kicinski <kuba@kernel.org> | 2021-12-07 03:26:46 +0300 |
| commit | 45cac6754529ae17345d8f5b632d9e602a091a20 (patch) | |
| tree | 711a5790a06b73816234f0c860caae947edc5cbe | |
| parent | 4c375272fb0ba19288e323928d1f6937e368851f (diff) | |
| download | linux-45cac6754529ae17345d8f5b632d9e602a091a20.tar.xz | |
net: fix recent csum changes
Vladimir reported csum issues after my recent change in skb_postpull_rcsum()
Issue here is the following:
initial skb->csum is the csum of
[part to be pulled][rest of packet]
Old code:
skb->csum = csum_sub(skb->csum, csum_partial(pull, pull_length, 0));
New code:
skb->csum = ~csum_partial(pull, pull_length, ~skb->csum);
This is broken if the csum of [pulled part]
happens to be equal to skb->csum, because end
result of skb->csum is 0 in new code, instead
of being 0xffffffff
David Laight suggested to use
skb->csum = -csum_partial(pull, pull_length, -skb->csum);
I based my patches on existing code present in include/net/seg6.h,
update_csum_diff4() and update_csum_diff16() which might need
a similar fix.
I guess that my tests, mostly pulling 40 bytes of IPv6 header
were not providing enough entropy to hit this bug.
v2: added wsum_negate() to make sparse happy.
Fixes: 29c3002644bd ("net: optimize skb_postpull_rcsum()")
Fixes: 0bd28476f636 ("gro: optimize skb_gro_postpull_rcsum()")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Suggested-by: David Laight <David.Laight@ACULAB.COM>
Cc: David Lebrun <dlebrun@google.com>
Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://lore.kernel.org/r/20211204045356.3659278-1-eric.dumazet@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
| -rw-r--r-- | include/linux/skbuff.h | 3 | ||||
| -rw-r--r-- | include/net/checksum.h | 4 | ||||
| -rw-r--r-- | include/net/gro.h | 4 |
3 files changed, 8 insertions, 3 deletions
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index eae4bd3237a4..dd262bd8ddbe 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3486,7 +3486,8 @@ static inline void skb_postpull_rcsum(struct sk_buff *skb, const void *start, unsigned int len) { if (skb->ip_summed == CHECKSUM_COMPLETE) - skb->csum = ~csum_partial(start, len, ~skb->csum); + skb->csum = wsum_negate(csum_partial(start, len, + wsum_negate(skb->csum))); else if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_start_offset(skb) < 0) skb->ip_summed = CHECKSUM_NONE; diff --git a/include/net/checksum.h b/include/net/checksum.h index 5b96d5bd6e54..5218041e5c8f 100644 --- a/include/net/checksum.h +++ b/include/net/checksum.h @@ -180,4 +180,8 @@ static inline void remcsum_unadjust(__sum16 *psum, __wsum delta) *psum = csum_fold(csum_sub(delta, (__force __wsum)*psum)); } +static inline __wsum wsum_negate(__wsum val) +{ + return (__force __wsum)-((__force u32)val); +} #endif diff --git a/include/net/gro.h b/include/net/gro.h index b1139fca7c43..8f75802d50fd 100644 --- a/include/net/gro.h +++ b/include/net/gro.h @@ -173,8 +173,8 @@ static inline void skb_gro_postpull_rcsum(struct sk_buff *skb, const void *start, unsigned int len) { if (NAPI_GRO_CB(skb)->csum_valid) - NAPI_GRO_CB(skb)->csum = ~csum_partial(start, len, - ~NAPI_GRO_CB(skb)->csum); + NAPI_GRO_CB(skb)->csum = wsum_negate(csum_partial(start, len, + wsum_negate(NAPI_GRO_CB(skb)->csum))); } /* GRO checksum functions. These are logical equivalents of the normal |