netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH net-next] net: fix recent csum changes
@ 2021-12-03 18:52 Eric Dumazet
  2021-12-04  3:34 ` kernel test robot
  2021-12-04  3:34 ` kernel test robot
  0 siblings, 2 replies; 8+ messages in thread
From: Eric Dumazet @ 2021-12-03 18:52 UTC (permalink / raw)
  To: David S . Miller, Jakub Kicinski
  Cc: netdev, Eric Dumazet, Eric Dumazet, Vladimir Oltean, David Laight,
	David Lebrun

From: Eric Dumazet <edumazet@google.com>

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.

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>
---
 include/linux/skbuff.h | 2 +-
 include/net/gro.h      | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index eae4bd3237a41cc1b60b44c91fbfe21dfdd8f117..2bbcdaf99ed3df739ddfd2de4be747262226d4b9 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -3486,7 +3486,7 @@ 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 = -csum_partial(start, len, -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/gro.h b/include/net/gro.h
index b1139fca7c435ca0c353c4ed17628dd7f3df4401..4529c4c6f3ca4ac23da569b0bc0e00e3c9dcd765 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 = -csum_partial(start, len,
+						       -NAPI_GRO_CB(skb)->csum);
 }
 
 /* GRO checksum functions. These are logical equivalents of the normal
-- 
2.34.1.400.ga245620fadb-goog


^ permalink raw reply related	[flat|nested] 8+ messages in thread

* Re: [PATCH net-next] net: fix recent csum changes
  2021-12-03 18:52 [PATCH net-next] net: fix recent csum changes Eric Dumazet
@ 2021-12-04  3:34 ` kernel test robot
  2021-12-04  3:34 ` kernel test robot
  1 sibling, 0 replies; 8+ messages in thread
From: kernel test robot @ 2021-12-04  3:34 UTC (permalink / raw)
  To: Eric Dumazet, David S . Miller, Jakub Kicinski
  Cc: kbuild-all, netdev, Eric Dumazet, Vladimir Oltean, David Laight,
	David Lebrun

Hi Eric,

I love your patch! Perhaps something to improve:

[auto build test WARNING on net-next/master]

url:    https://github.com/0day-ci/linux/commits/Eric-Dumazet/net-fix-recent-csum-changes/20211204-025401
base:   https://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git 43332cf97425a3e5508c827c82201ecc5ddd54e0
config: arc-randconfig-s032-20211203 (https://download.01.org/0day-ci/archive/20211204/202112041119.X8sqBWc9-lkp@intel.com/config)
compiler: arc-elf-gcc (GCC) 11.2.0
reproduce:
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # apt-get install sparse
        # sparse version: v0.6.4-dirty
        # https://github.com/0day-ci/linux/commit/c13fbd113358fb59f76f76d25a1fdb57379c4b9c
        git remote add linux-review https://github.com/0day-ci/linux
        git fetch --no-tags linux-review Eric-Dumazet/net-fix-recent-csum-changes/20211204-025401
        git checkout c13fbd113358fb59f76f76d25a1fdb57379c4b9c
        # save the config file to linux build tree
        mkdir build_dir
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-11.2.0 make.cross C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' O=build_dir ARCH=arc SHELL=/bin/bash net/core/ net/ethernet/

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>


sparse warnings: (new ones prefixed by >>)
   net/core/skbuff.c: note: in included file (through include/net/checksum.h, include/linux/skbuff.h, include/net/net_namespace.h, ...):
   arch/arc/include/asm/checksum.h:27:26: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:27:36: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:29:11: sparse: sparse: bad assignment (-=) to restricted __wsum
   arch/arc/include/asm/checksum.h:30:16: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:30:18: sparse: sparse: incorrect type in return expression (different base types) @@     expected restricted __sum16 @@     got unsigned int @@
   arch/arc/include/asm/checksum.h:30:18: sparse:     expected restricted __sum16
   arch/arc/include/asm/checksum.h:30:18: sparse:     got unsigned int
   arch/arc/include/asm/checksum.h:27:26: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:27:36: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:29:11: sparse: sparse: bad assignment (-=) to restricted __wsum
   arch/arc/include/asm/checksum.h:30:16: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:30:18: sparse: sparse: incorrect type in return expression (different base types) @@     expected restricted __sum16 @@     got unsigned int @@
   arch/arc/include/asm/checksum.h:30:18: sparse:     expected restricted __sum16
   arch/arc/include/asm/checksum.h:30:18: sparse:     got unsigned int
   arch/arc/include/asm/checksum.h:27:26: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:27:36: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:29:11: sparse: sparse: bad assignment (-=) to restricted __wsum
   arch/arc/include/asm/checksum.h:30:16: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:30:18: sparse: sparse: incorrect type in return expression (different base types) @@     expected restricted __sum16 @@     got unsigned int @@
   arch/arc/include/asm/checksum.h:30:18: sparse:     expected restricted __sum16
   arch/arc/include/asm/checksum.h:30:18: sparse:     got unsigned int
   net/core/skbuff.c: note: in included file (through include/net/net_namespace.h, include/linux/inet.h):
   include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] sum @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype] sum
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
   include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
   net/core/skbuff.c: note: in included file (through include/net/checksum.h, include/linux/skbuff.h, include/net/net_namespace.h, ...):
   arch/arc/include/asm/checksum.h:27:26: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:27:36: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:29:11: sparse: sparse: bad assignment (-=) to restricted __wsum
   arch/arc/include/asm/checksum.h:30:16: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:30:18: sparse: sparse: incorrect type in return expression (different base types) @@     expected restricted __sum16 @@     got unsigned int @@
   arch/arc/include/asm/checksum.h:30:18: sparse:     expected restricted __sum16
   arch/arc/include/asm/checksum.h:30:18: sparse:     got unsigned int
   net/core/skbuff.c: note: in included file (through include/net/net_namespace.h, include/linux/inet.h):
   include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] sum @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype] sum
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
   include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
   include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] sum @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype] sum
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
   include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
--
   net/core/filter.c:5622:9: sparse: sparse: switch with no cases
   net/core/filter.c:5663:9: sparse: sparse: switch with no cases
   net/core/filter.c:1411:39: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct sock_filter const *filter @@     got struct sock_filter [noderef] __user *filter @@
   net/core/filter.c:1411:39: sparse:     expected struct sock_filter const *filter
   net/core/filter.c:1411:39: sparse:     got struct sock_filter [noderef] __user *filter
   net/core/filter.c:1489:39: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct sock_filter const *filter @@     got struct sock_filter [noderef] __user *filter @@
   net/core/filter.c:1489:39: sparse:     expected struct sock_filter const *filter
   net/core/filter.c:1489:39: sparse:     got struct sock_filter [noderef] __user *filter
   net/core/filter.c:10023:31: sparse: sparse: symbol 'cg_skb_verifier_ops' was not declared. Should it be static?
   net/core/filter.c:10029:27: sparse: sparse: symbol 'cg_skb_prog_ops' was not declared. Should it be static?
   net/core/filter.c:10074:31: sparse: sparse: symbol 'cg_sock_verifier_ops' was not declared. Should it be static?
   net/core/filter.c:10080:27: sparse: sparse: symbol 'cg_sock_prog_ops' was not declared. Should it be static?
   net/core/filter.c:10083:31: sparse: sparse: symbol 'cg_sock_addr_verifier_ops' was not declared. Should it be static?
   net/core/filter.c:10089:27: sparse: sparse: symbol 'cg_sock_addr_prog_ops' was not declared. Should it be static?
   net/core/filter.c:246:32: sparse: sparse: cast to restricted __be16
   net/core/filter.c:246:32: sparse: sparse: cast to restricted __be16
   net/core/filter.c:246:32: sparse: sparse: cast to restricted __be16
   net/core/filter.c:246:32: sparse: sparse: cast to restricted __be16
   net/core/filter.c:273:32: sparse: sparse: cast to restricted __be32
   net/core/filter.c:273:32: sparse: sparse: cast to restricted __be32
   net/core/filter.c:273:32: sparse: sparse: cast to restricted __be32
   net/core/filter.c:273:32: sparse: sparse: cast to restricted __be32
   net/core/filter.c:273:32: sparse: sparse: cast to restricted __be32
   net/core/filter.c:273:32: sparse: sparse: cast to restricted __be32
   net/core/filter.c:1910:43: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected restricted __wsum [usertype] diff @@     got unsigned long long [usertype] to @@
   net/core/filter.c:1910:43: sparse:     expected restricted __wsum [usertype] diff
   net/core/filter.c:1910:43: sparse:     got unsigned long long [usertype] to
   net/core/filter.c:1913:36: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected restricted __be16 [usertype] old @@     got unsigned long long [usertype] from @@
   net/core/filter.c:1913:36: sparse:     expected restricted __be16 [usertype] old
   net/core/filter.c:1913:36: sparse:     got unsigned long long [usertype] from
   net/core/filter.c:1913:42: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __be16 [usertype] new @@     got unsigned long long [usertype] to @@
   net/core/filter.c:1913:42: sparse:     expected restricted __be16 [usertype] new
   net/core/filter.c:1913:42: sparse:     got unsigned long long [usertype] to
   net/core/filter.c:1916:36: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected restricted __be32 [usertype] from @@     got unsigned long long [usertype] from @@
   net/core/filter.c:1916:36: sparse:     expected restricted __be32 [usertype] from
   net/core/filter.c:1916:36: sparse:     got unsigned long long [usertype] from
   net/core/filter.c:1916:42: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __be32 [usertype] to @@     got unsigned long long [usertype] to @@
   net/core/filter.c:1916:42: sparse:     expected restricted __be32 [usertype] to
   net/core/filter.c:1916:42: sparse:     got unsigned long long [usertype] to
   net/core/filter.c:1961:59: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] diff @@     got unsigned long long [usertype] to @@
   net/core/filter.c:1961:59: sparse:     expected restricted __wsum [usertype] diff
   net/core/filter.c:1961:59: sparse:     got unsigned long long [usertype] to
   net/core/filter.c:1964:52: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __be16 [usertype] from @@     got unsigned long long [usertype] from @@
   net/core/filter.c:1964:52: sparse:     expected restricted __be16 [usertype] from
   net/core/filter.c:1964:52: sparse:     got unsigned long long [usertype] from
   net/core/filter.c:1964:58: sparse: sparse: incorrect type in argument 4 (different base types) @@     expected restricted __be16 [usertype] to @@     got unsigned long long [usertype] to @@
   net/core/filter.c:1964:58: sparse:     expected restricted __be16 [usertype] to
   net/core/filter.c:1964:58: sparse:     got unsigned long long [usertype] to
   net/core/filter.c:1967:52: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __be32 [usertype] from @@     got unsigned long long [usertype] from @@
   net/core/filter.c:1967:52: sparse:     expected restricted __be32 [usertype] from
   net/core/filter.c:1967:52: sparse:     got unsigned long long [usertype] from
   net/core/filter.c:1967:58: sparse: sparse: incorrect type in argument 4 (different base types) @@     expected restricted __be32 [usertype] to @@     got unsigned long long [usertype] to @@
   net/core/filter.c:1967:58: sparse:     expected restricted __be32 [usertype] to
   net/core/filter.c:1967:58: sparse:     got unsigned long long [usertype] to
   net/core/filter.c:2013:28: sparse: sparse: incorrect type in return expression (different base types) @@     expected unsigned long long @@     got restricted __wsum @@
   net/core/filter.c:2013:28: sparse:     expected unsigned long long
   net/core/filter.c:2013:28: sparse:     got restricted __wsum
   net/core/filter.c:2035:35: sparse: sparse: incorrect type in return expression (different base types) @@     expected unsigned long long @@     got restricted __wsum [usertype] csum @@
   net/core/filter.c:2035:35: sparse:     expected unsigned long long
   net/core/filter.c:2035:35: sparse:     got restricted __wsum [usertype] csum
   net/core/filter.c: note: in included file (through include/linux/netlink.h, include/linux/sock_diag.h):
   include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] sum @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype] sum
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
   include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
   include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] sum @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype] sum
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
   include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
--
   net/core/dev.c:3207:23: sparse: sparse: incorrect type in argument 4 (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   net/core/dev.c:3207:23: sparse:     expected restricted __wsum [usertype] csum
   net/core/dev.c:3207:23: sparse:     got unsigned int
   net/core/dev.c:3207:23: sparse: sparse: cast from restricted __wsum
   net/core/dev.c: note: in included file (through include/linux/if_ether.h):
   include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] sum @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype] sum
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
   include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
   net/core/dev.c: note: in included file (through include/net/checksum.h, include/linux/skbuff.h, include/linux/if_ether.h):
   arch/arc/include/asm/checksum.h:27:26: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:27:36: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:29:11: sparse: sparse: bad assignment (-=) to restricted __wsum
   arch/arc/include/asm/checksum.h:30:16: sparse: sparse: restricted __wsum degrades to integer
   arch/arc/include/asm/checksum.h:30:18: sparse: sparse: incorrect type in return expression (different base types) @@     expected restricted __sum16 @@     got unsigned int @@
   arch/arc/include/asm/checksum.h:30:18: sparse:     expected restricted __sum16
   arch/arc/include/asm/checksum.h:30:18: sparse:     got unsigned int
   net/core/dev.c:3712:17: sparse: sparse: context imbalance in '__dev_queue_xmit' - different lock contexts for basic block
   net/core/dev.c: note: in included file (through arch/arc/include/asm/irqflags.h, include/linux/irqflags.h, include/linux/rcupdate.h, ...):
   arch/arc/include/asm/irqflags-arcv2.h:80:31: sparse: sparse: undefined identifier '__builtin_arc_lr'
   arch/arc/include/asm/irqflags-arcv2.h:83:17: sparse: sparse: undefined identifier '__builtin_arc_sr'
--
   net/ethernet/eth.c: note: in included file:
   include/net/gro.h:177:56: sparse: sparse: restricted __wsum degrades to integer
>> include/net/gro.h:177:56: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] sum @@     got unsigned int @@
   include/net/gro.h:177:56: sparse:     expected restricted __wsum [usertype] sum
   include/net/gro.h:177:56: sparse:     got unsigned int
   include/net/gro.h:176:42: sparse: sparse: restricted __wsum degrades to integer
   include/net/gro.h:176:40: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/net/gro.h:176:40: sparse:     expected restricted __wsum [usertype] csum
   include/net/gro.h:176:40: sparse:     got unsigned int

vim +3489 include/linux/skbuff.h

  3474	
  3475	/**
  3476	 *	skb_postpull_rcsum - update checksum for received skb after pull
  3477	 *	@skb: buffer to update
  3478	 *	@start: start of data before pull
  3479	 *	@len: length of data pulled
  3480	 *
  3481	 *	After doing a pull on a received packet, you need to call this to
  3482	 *	update the CHECKSUM_COMPLETE checksum, or set ip_summed to
  3483	 *	CHECKSUM_NONE so that it can be recomputed from scratch.
  3484	 */
  3485	static inline void skb_postpull_rcsum(struct sk_buff *skb,
  3486					      const void *start, unsigned int len)
  3487	{
  3488		if (skb->ip_summed == CHECKSUM_COMPLETE)
> 3489			skb->csum = -csum_partial(start, len, -skb->csum);
  3490		else if (skb->ip_summed == CHECKSUM_PARTIAL &&
  3491			 skb_checksum_start_offset(skb) < 0)
  3492			skb->ip_summed = CHECKSUM_NONE;
  3493	}
  3494	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH net-next] net: fix recent csum changes
  2021-12-03 18:52 [PATCH net-next] net: fix recent csum changes Eric Dumazet
  2021-12-04  3:34 ` kernel test robot
@ 2021-12-04  3:34 ` kernel test robot
  2021-12-04  4:40   ` Eric Dumazet
  1 sibling, 1 reply; 8+ messages in thread
From: kernel test robot @ 2021-12-04  3:34 UTC (permalink / raw)
  To: Eric Dumazet, David S . Miller, Jakub Kicinski
  Cc: kbuild-all, netdev, Eric Dumazet, Vladimir Oltean, David Laight,
	David Lebrun

Hi Eric,

I love your patch! Perhaps something to improve:

[auto build test WARNING on net-next/master]

url:    https://github.com/0day-ci/linux/commits/Eric-Dumazet/net-fix-recent-csum-changes/20211204-025401
base:   https://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git 43332cf97425a3e5508c827c82201ecc5ddd54e0
config: parisc-randconfig-s031-20211203 (https://download.01.org/0day-ci/archive/20211204/202112041104.gPgP3Z6U-lkp@intel.com/config)
compiler: hppa64-linux-gcc (GCC) 11.2.0
reproduce:
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # apt-get install sparse
        # sparse version: v0.6.4-dirty
        # https://github.com/0day-ci/linux/commit/c13fbd113358fb59f76f76d25a1fdb57379c4b9c
        git remote add linux-review https://github.com/0day-ci/linux
        git fetch --no-tags linux-review Eric-Dumazet/net-fix-recent-csum-changes/20211204-025401
        git checkout c13fbd113358fb59f76f76d25a1fdb57379c4b9c
        # save the config file to linux build tree
        mkdir build_dir
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-11.2.0 make.cross C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' O=build_dir ARCH=parisc SHELL=/bin/bash net/core/ net/ipv4/

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>


sparse warnings: (new ones prefixed by >>)
   net/core/skbuff.c: note: in included file (through include/net/net_namespace.h, include/linux/inet.h):
>> include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype]
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
>> include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype]
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
>> include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype]
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
--
   net/core/filter.c:1411:39: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct sock_filter const *filter @@     got struct sock_filter [noderef] __user *filter @@
   net/core/filter.c:1411:39: sparse:     expected struct sock_filter const *filter
   net/core/filter.c:1411:39: sparse:     got struct sock_filter [noderef] __user *filter
   net/core/filter.c:1489:39: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct sock_filter const *filter @@     got struct sock_filter [noderef] __user *filter @@
   net/core/filter.c:1489:39: sparse:     expected struct sock_filter const *filter
   net/core/filter.c:1489:39: sparse:     got struct sock_filter [noderef] __user *filter
   net/core/filter.c:2296:45: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected restricted __be32 [usertype] daddr @@     got unsigned int [usertype] ipv4_nh @@
   net/core/filter.c:2296:45: sparse:     expected restricted __be32 [usertype] daddr
   net/core/filter.c:2296:45: sparse:     got unsigned int [usertype] ipv4_nh
   net/core/filter.c:10023:31: sparse: sparse: symbol 'cg_skb_verifier_ops' was not declared. Should it be static?
   net/core/filter.c:10029:27: sparse: sparse: symbol 'cg_skb_prog_ops' was not declared. Should it be static?
   net/core/filter.c:10074:31: sparse: sparse: symbol 'cg_sock_verifier_ops' was not declared. Should it be static?
   net/core/filter.c:10080:27: sparse: sparse: symbol 'cg_sock_prog_ops' was not declared. Should it be static?
   net/core/filter.c:10083:31: sparse: sparse: symbol 'cg_sock_addr_verifier_ops' was not declared. Should it be static?
   net/core/filter.c:10089:27: sparse: sparse: symbol 'cg_sock_addr_prog_ops' was not declared. Should it be static?
   net/core/filter.c:246:32: sparse: sparse: cast to restricted __be16
   net/core/filter.c:273:32: sparse: sparse: cast to restricted __be32
   net/core/filter.c:1910:43: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected restricted __wsum [usertype] diff @@     got unsigned long long [usertype] to @@
   net/core/filter.c:1910:43: sparse:     expected restricted __wsum [usertype] diff
   net/core/filter.c:1910:43: sparse:     got unsigned long long [usertype] to
   net/core/filter.c:1913:36: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected restricted __be16 [usertype] old @@     got unsigned long long [usertype] from @@
   net/core/filter.c:1913:36: sparse:     expected restricted __be16 [usertype] old
   net/core/filter.c:1913:36: sparse:     got unsigned long long [usertype] from
   net/core/filter.c:1913:42: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __be16 [usertype] new @@     got unsigned long long [usertype] to @@
   net/core/filter.c:1913:42: sparse:     expected restricted __be16 [usertype] new
   net/core/filter.c:1913:42: sparse:     got unsigned long long [usertype] to
   net/core/filter.c:1916:36: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected restricted __be32 [usertype] from @@     got unsigned long long [usertype] from @@
   net/core/filter.c:1916:36: sparse:     expected restricted __be32 [usertype] from
   net/core/filter.c:1916:36: sparse:     got unsigned long long [usertype] from
   net/core/filter.c:1916:42: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __be32 [usertype] to @@     got unsigned long long [usertype] to @@
   net/core/filter.c:1916:42: sparse:     expected restricted __be32 [usertype] to
   net/core/filter.c:1916:42: sparse:     got unsigned long long [usertype] to
   net/core/filter.c:1961:59: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] diff @@     got unsigned long long [usertype] to @@
   net/core/filter.c:1961:59: sparse:     expected restricted __wsum [usertype] diff
   net/core/filter.c:1961:59: sparse:     got unsigned long long [usertype] to
   net/core/filter.c:1964:52: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __be16 [usertype] from @@     got unsigned long long [usertype] from @@
   net/core/filter.c:1964:52: sparse:     expected restricted __be16 [usertype] from
   net/core/filter.c:1964:52: sparse:     got unsigned long long [usertype] from
   net/core/filter.c:1964:58: sparse: sparse: incorrect type in argument 4 (different base types) @@     expected restricted __be16 [usertype] to @@     got unsigned long long [usertype] to @@
   net/core/filter.c:1964:58: sparse:     expected restricted __be16 [usertype] to
   net/core/filter.c:1964:58: sparse:     got unsigned long long [usertype] to
   net/core/filter.c:1967:52: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __be32 [usertype] from @@     got unsigned long long [usertype] from @@
   net/core/filter.c:1967:52: sparse:     expected restricted __be32 [usertype] from
   net/core/filter.c:1967:52: sparse:     got unsigned long long [usertype] from
   net/core/filter.c:1967:58: sparse: sparse: incorrect type in argument 4 (different base types) @@     expected restricted __be32 [usertype] to @@     got unsigned long long [usertype] to @@
   net/core/filter.c:1967:58: sparse:     expected restricted __be32 [usertype] to
   net/core/filter.c:1967:58: sparse:     got unsigned long long [usertype] to
   net/core/filter.c:2013:28: sparse: sparse: incorrect type in return expression (different base types) @@     expected unsigned long long @@     got restricted __wsum @@
   net/core/filter.c:2013:28: sparse:     expected unsigned long long
   net/core/filter.c:2013:28: sparse:     got restricted __wsum
   net/core/filter.c:2035:35: sparse: sparse: incorrect type in return expression (different base types) @@     expected unsigned long long @@     got restricted __wsum [usertype] csum @@
   net/core/filter.c:2035:35: sparse:     expected unsigned long long
   net/core/filter.c:2035:35: sparse:     got restricted __wsum [usertype] csum
   net/core/filter.c:5333:17: sparse: sparse: incorrect type in assignment (different base types) @@     expected unsigned int [usertype] spi @@     got restricted __be32 const [usertype] spi @@
   net/core/filter.c:5333:17: sparse:     expected unsigned int [usertype] spi
   net/core/filter.c:5333:17: sparse:     got restricted __be32 const [usertype] spi
   net/core/filter.c:5341:33: sparse: sparse: incorrect type in assignment (different base types) @@     expected unsigned int [usertype] remote_ipv4 @@     got restricted __be32 const [usertype] a4 @@
   net/core/filter.c:5341:33: sparse:     expected unsigned int [usertype] remote_ipv4
   net/core/filter.c:5341:33: sparse:     got restricted __be32 const [usertype] a4
   net/core/filter.c: note: in included file (through include/linux/netlink.h, include/linux/sock_diag.h):
>> include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype]
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
   net/core/filter.c: note: in included file (through include/net/ip.h):
   include/net/route.h:372:48: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected unsigned int [usertype] key @@     got restricted __be32 [usertype] daddr @@
   include/net/route.h:372:48: sparse:     expected unsigned int [usertype] key
   include/net/route.h:372:48: sparse:     got restricted __be32 [usertype] daddr
   include/net/route.h:372:48: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected unsigned int [usertype] key @@     got restricted __be32 [usertype] daddr @@
   include/net/route.h:372:48: sparse:     expected unsigned int [usertype] key
   include/net/route.h:372:48: sparse:     got restricted __be32 [usertype] daddr
   include/net/route.h:372:48: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected unsigned int [usertype] key @@     got restricted __be32 [usertype] daddr @@
   include/net/route.h:372:48: sparse:     expected unsigned int [usertype] key
   include/net/route.h:372:48: sparse:     got restricted __be32 [usertype] daddr
   net/core/filter.c: note: in included file (through include/linux/netlink.h, include/linux/sock_diag.h):
>> include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype]
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
--
   net/core/dev.c:3207:23: sparse: sparse: incorrect type in argument 4 (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   net/core/dev.c:3207:23: sparse:     expected restricted __wsum [usertype] csum
   net/core/dev.c:3207:23: sparse:     got unsigned int
   net/core/dev.c:3207:23: sparse: sparse: cast from restricted __wsum
   net/core/dev.c:3207:23: sparse: sparse: incorrect type in argument 4 (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   net/core/dev.c:3207:23: sparse:     expected restricted __wsum [usertype] csum
   net/core/dev.c:3207:23: sparse:     got unsigned int
   net/core/dev.c:3207:23: sparse: sparse: incorrect type in argument 1 (different base types) @@     expected unsigned int [usertype] val @@     got restricted __wsum @@
   net/core/dev.c:3207:23: sparse:     expected unsigned int [usertype] val
   net/core/dev.c:3207:23: sparse:     got restricted __wsum
   net/core/dev.c:3207:23: sparse: sparse: incorrect type in argument 4 (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   net/core/dev.c:3207:23: sparse:     expected restricted __wsum [usertype] csum
   net/core/dev.c:3207:23: sparse:     got unsigned int
   net/core/dev.c:3207:23: sparse: sparse: cast from restricted __wsum
   net/core/dev.c:3207:23: sparse: sparse: incorrect type in argument 4 (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   net/core/dev.c:3207:23: sparse:     expected restricted __wsum [usertype] csum
   net/core/dev.c:3207:23: sparse:     got unsigned int
   net/core/dev.c:3207:23: sparse: sparse: cast from restricted __wsum
   net/core/dev.c:3207:23: sparse: sparse: incorrect type in argument 4 (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   net/core/dev.c:3207:23: sparse:     expected restricted __wsum [usertype] csum
   net/core/dev.c:3207:23: sparse:     got unsigned int
   net/core/dev.c:3207:23: sparse: sparse: cast from restricted __wsum
   net/core/dev.c:3207:23: sparse: sparse: incorrect type in argument 4 (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   net/core/dev.c:3207:23: sparse:     expected restricted __wsum [usertype] csum
   net/core/dev.c:3207:23: sparse:     got unsigned int
   net/core/dev.c:3207:23: sparse: sparse: cast from restricted __wsum
   net/core/dev.c: note: in included file (through include/linux/if_ether.h):
>> include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype]
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
   net/core/dev.c:3712:17: sparse: sparse: context imbalance in '__dev_queue_xmit' - different lock contexts for basic block
   net/core/dev.c:4918:17: sparse: sparse: context imbalance in 'net_tx_action' - different lock contexts for basic block
--
   net/ipv4/udp_offload.c:139:60: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected restricted __wsum [usertype] res @@     got fouled restricted __sum16 @@
   net/ipv4/udp_offload.c:139:60: sparse:     expected restricted __wsum [usertype] res
   net/ipv4/udp_offload.c:139:60: sparse:     got fouled restricted __sum16
   net/ipv4/udp_offload.c:330:49: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected restricted __wsum [usertype] res @@     got fouled restricted __sum16 @@
   net/ipv4/udp_offload.c:330:49: sparse:     expected restricted __wsum [usertype] res
   net/ipv4/udp_offload.c:330:49: sparse:     got fouled restricted __sum16
   net/ipv4/udp_offload.c:332:60: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected restricted __wsum [usertype] res @@     got fouled restricted __sum16 @@
   net/ipv4/udp_offload.c:332:60: sparse:     expected restricted __wsum [usertype] res
   net/ipv4/udp_offload.c:332:60: sparse:     got fouled restricted __sum16
   net/ipv4/udp_offload.c:348:41: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected restricted __wsum [usertype] res @@     got fouled restricted __sum16 @@
   net/ipv4/udp_offload.c:348:41: sparse:     expected restricted __wsum [usertype] res
   net/ipv4/udp_offload.c:348:41: sparse:     got fouled restricted __sum16
   net/ipv4/udp_offload.c:350:52: sparse: sparse: incorrect type in argument 2 (different base types) @@     expected restricted __wsum [usertype] res @@     got fouled restricted __sum16 @@
   net/ipv4/udp_offload.c:350:52: sparse:     expected restricted __wsum [usertype] res
   net/ipv4/udp_offload.c:350:52: sparse:     got fouled restricted __sum16
   net/ipv4/udp_offload.c: note: in included file:
>> include/net/gro.h:177:56: sparse: sparse: restricted __wsum degrades to integer
>> include/net/gro.h:177:56: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/net/gro.h:177:56: sparse:     expected restricted __wsum [usertype]
   include/net/gro.h:177:56: sparse:     got unsigned int
   include/net/gro.h:176:42: sparse: sparse: restricted __wsum degrades to integer
>> include/net/gro.h:176:40: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/net/gro.h:176:40: sparse:     expected restricted __wsum [usertype] csum
   include/net/gro.h:176:40: sparse:     got unsigned int
>> include/net/gro.h:177:56: sparse: sparse: restricted __wsum degrades to integer
>> include/net/gro.h:177:56: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/net/gro.h:177:56: sparse:     expected restricted __wsum [usertype]
   include/net/gro.h:177:56: sparse:     got unsigned int
   include/net/gro.h:176:42: sparse: sparse: restricted __wsum degrades to integer
>> include/net/gro.h:176:40: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/net/gro.h:176:40: sparse:     expected restricted __wsum [usertype] csum
   include/net/gro.h:176:40: sparse:     got unsigned int
--
   net/ipv4/gre_offload.c: note: in included file:
>> include/net/gro.h:177:56: sparse: sparse: restricted __wsum degrades to integer
>> include/net/gro.h:177:56: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/net/gro.h:177:56: sparse:     expected restricted __wsum [usertype]
   include/net/gro.h:177:56: sparse:     got unsigned int
   include/net/gro.h:176:42: sparse: sparse: restricted __wsum degrades to integer
>> include/net/gro.h:176:40: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/net/gro.h:176:40: sparse:     expected restricted __wsum [usertype] csum
   include/net/gro.h:176:40: sparse:     got unsigned int
--
   net/ipv4/fou.c: note: in included file:
>> include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype]
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
>> include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype]
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
>> include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype]
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int
   net/ipv4/fou.c: note: in included file:
>> include/net/gro.h:177:56: sparse: sparse: restricted __wsum degrades to integer
>> include/net/gro.h:177:56: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/net/gro.h:177:56: sparse:     expected restricted __wsum [usertype]
   include/net/gro.h:177:56: sparse:     got unsigned int
   include/net/gro.h:176:42: sparse: sparse: restricted __wsum degrades to integer
>> include/net/gro.h:176:40: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/net/gro.h:176:40: sparse:     expected restricted __wsum [usertype] csum
   include/net/gro.h:176:40: sparse:     got unsigned int
--
   net/ipv4/ip_tunnel.c: note: in included file:
>> include/linux/skbuff.h:3489:55: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:55: sparse: sparse: incorrect type in argument 3 (different base types) @@     expected restricted __wsum [usertype] @@     got unsigned int @@
   include/linux/skbuff.h:3489:55: sparse:     expected restricted __wsum [usertype]
   include/linux/skbuff.h:3489:55: sparse:     got unsigned int
   include/linux/skbuff.h:3489:29: sparse: sparse: restricted __wsum degrades to integer
>> include/linux/skbuff.h:3489:27: sparse: sparse: incorrect type in assignment (different base types) @@     expected restricted __wsum [usertype] csum @@     got unsigned int @@
   include/linux/skbuff.h:3489:27: sparse:     expected restricted __wsum [usertype] csum
   include/linux/skbuff.h:3489:27: sparse:     got unsigned int

vim +3489 include/linux/skbuff.h

  3474	
  3475	/**
  3476	 *	skb_postpull_rcsum - update checksum for received skb after pull
  3477	 *	@skb: buffer to update
  3478	 *	@start: start of data before pull
  3479	 *	@len: length of data pulled
  3480	 *
  3481	 *	After doing a pull on a received packet, you need to call this to
  3482	 *	update the CHECKSUM_COMPLETE checksum, or set ip_summed to
  3483	 *	CHECKSUM_NONE so that it can be recomputed from scratch.
  3484	 */
  3485	static inline void skb_postpull_rcsum(struct sk_buff *skb,
  3486					      const void *start, unsigned int len)
  3487	{
  3488		if (skb->ip_summed == CHECKSUM_COMPLETE)
> 3489			skb->csum = -csum_partial(start, len, -skb->csum);
  3490		else if (skb->ip_summed == CHECKSUM_PARTIAL &&
  3491			 skb_checksum_start_offset(skb) < 0)
  3492			skb->ip_summed = CHECKSUM_NONE;
  3493	}
  3494	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH net-next] net: fix recent csum changes
  2021-12-04  3:34 ` kernel test robot
@ 2021-12-04  4:40   ` Eric Dumazet
  2021-12-04 14:00     ` David Laight
  0 siblings, 1 reply; 8+ messages in thread
From: Eric Dumazet @ 2021-12-04  4:40 UTC (permalink / raw)
  To: kernel test robot
  Cc: Eric Dumazet, David S . Miller, Jakub Kicinski, kbuild-all,
	netdev, Vladimir Oltean, David Laight, David Lebrun

On Fri, Dec 3, 2021 at 7:34 PM kernel test robot <lkp@intel.com> wrote:
>
> Hi Eric,
>
> I love your patch! Perhaps something to improve:
>
> [auto build test WARNING on net-next/master]
>
> url:    https://github.com/0day-ci/linux/commits/Eric-Dumazet/net-fix-recent-csum-changes/20211204-025401
> base:   https://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git 43332cf97425a3e5508c827c82201ecc5ddd54e0
> config: parisc-randconfig-s031-20211203 (https://download.01.org/0day-ci/archive/20211204/202112041104.gPgP3Z6U-lkp@intel.com/config)
> compiler: hppa64-linux-gcc (GCC) 11.2.0
> reproduce:
>         wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
>         chmod +x ~/bin/make.cross
>         # apt-get install sparse
>         # sparse version: v0.6.4-dirty
>         # https://github.com/0day-ci/linux/commit/c13fbd113358fb59f76f76d25a1fdb57379c4b9c
>         git remote add linux-review https://github.com/0day-ci/linux
>         git fetch --no-tags linux-review Eric-Dumazet/net-fix-recent-csum-changes/20211204-025401
>         git checkout c13fbd113358fb59f76f76d25a1fdb57379c4b9c
>         # save the config file to linux build tree
>         mkdir build_dir
>         COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-11.2.0 make.cross C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' O=build_dir ARCH=parisc SHELL=/bin/bash net/core/ net/ipv4/
>
> If you fix the issue, kindly add following tag as appropriate
> Reported-by: kernel test robot <lkp@intel.com>
>

Yes, keeping sparse happy with these checksum is not easy.

I will add and use this helper, unless someone has a better idea.

diff --git a/include/net/checksum.h b/include/net/checksum.h
index 5b96d5bd6e54532a7a82511ff5d7d4c6f18982d5..5218041e5c8f93cd369a2a3a46add3e6a5e41af7
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

^ permalink raw reply	[flat|nested] 8+ messages in thread

* RE: [PATCH net-next] net: fix recent csum changes
  2021-12-04  4:40   ` Eric Dumazet
@ 2021-12-04 14:00     ` David Laight
  2021-12-04 18:33       ` Eric Dumazet
  0 siblings, 1 reply; 8+ messages in thread
From: David Laight @ 2021-12-04 14:00 UTC (permalink / raw)
  To: 'Eric Dumazet', kernel test robot
  Cc: Eric Dumazet, David S . Miller, Jakub Kicinski,
	kbuild-all@lists.01.org, netdev, Vladimir Oltean, David Lebrun

From: Eric Dumazet
> Sent: 04 December 2021 04:41
> 
> On Fri, Dec 3, 2021 at 7:34 PM kernel test robot <lkp@intel.com> wrote:
> >
> > I love your patch! Perhaps something to improve:
...
> 
> Yes, keeping sparse happy with these checksum is not easy.
> 
> I will add and use this helper, unless someone has a better idea.
> 
> diff --git a/include/net/checksum.h b/include/net/checksum.h
> index 5b96d5bd6e54532a7a82511ff5d7d4c6f18982d5..5218041e5c8f93cd369a2a3a46add3e6a5e41af7
> 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

I was thinking that the expression also requires some comments.
So maybe put a #define / static inline in checksum.h like:

/* Subract the checksum of a buffer.
 * The domain is __wsum is [1..~0u] (ie excludes zero)
 * so ~csum_partial() cannot be used.
 * The two's compliment gives the right answer provided the old 'csum'
 * isn't zero - which it shouldn't be. */
#define csum_partial_sub(buf, len, csum) (-csum_partial(buf, len, -(csum))

and then add the annotations there to keep sparse happy there.

will sparse accept '1 + ~csum' ? The compiler should use negate for it.
It actually makes it slightly more obvious why the code is right.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH net-next] net: fix recent csum changes
  2021-12-04 14:00     ` David Laight
@ 2021-12-04 18:33       ` Eric Dumazet
  2021-12-04 19:03         ` David Laight
  0 siblings, 1 reply; 8+ messages in thread
From: Eric Dumazet @ 2021-12-04 18:33 UTC (permalink / raw)
  To: David Laight
  Cc: kernel test robot, Eric Dumazet, David S . Miller, Jakub Kicinski,
	kbuild-all@lists.01.org, netdev, Vladimir Oltean, David Lebrun

On Sat, Dec 4, 2021 at 6:00 AM David Laight <David.Laight@aculab.com> wrote:
>
> From: Eric Dumazet
> > Sent: 04 December 2021 04:41
> >
> > On Fri, Dec 3, 2021 at 7:34 PM kernel test robot <lkp@intel.com> wrote:
> > >
> > > I love your patch! Perhaps something to improve:
> ...
> >
> > Yes, keeping sparse happy with these checksum is not easy.
> >
> > I will add and use this helper, unless someone has a better idea.
> >
> > diff --git a/include/net/checksum.h b/include/net/checksum.h
> > index 5b96d5bd6e54532a7a82511ff5d7d4c6f18982d5..5218041e5c8f93cd369a2a3a46add3e6a5e41af7
> > 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
>
> I was thinking that the expression also requires some comments.
> So maybe put a #define / static inline in checksum.h like:
>
> /* Subract the checksum of a buffer.
>  * The domain is __wsum is [1..~0u] (ie excludes zero)
>  * so ~csum_partial() cannot be used.
>  * The two's compliment gives the right answer provided the old 'csum'
>  * isn't zero - which it shouldn't be. */
> #define csum_partial_sub(buf, len, csum) (-csum_partial(buf, len, -(csum))
>
> and then add the annotations there to keep sparse happy there.
>
> will sparse accept '1 + ~csum' ? The compiler should use negate for it.
> It actually makes it slightly more obvious why the code is right.

Sparse is not happy with  1 + ~csum,

So we unfortunately would need something ugly like

(__force __wsum)(1 + ~(__force u32)csum)

Which most readers of this code will not find obvious.


>
>         David
>
> -
> Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
> Registration No: 1397386 (Wales)

^ permalink raw reply	[flat|nested] 8+ messages in thread

* RE: [PATCH net-next] net: fix recent csum changes
  2021-12-04 18:33       ` Eric Dumazet
@ 2021-12-04 19:03         ` David Laight
  2021-12-04 19:18           ` David Laight
  0 siblings, 1 reply; 8+ messages in thread
From: David Laight @ 2021-12-04 19:03 UTC (permalink / raw)
  To: 'Eric Dumazet'
  Cc: kernel test robot, Eric Dumazet, David S . Miller, Jakub Kicinski,
	kbuild-all@lists.01.org, netdev, Vladimir Oltean, David Lebrun

From: Eric Dumazet
> Sent: 04 December 2021 18:34
> 
> On Sat, Dec 4, 2021 at 6:00 AM David Laight <David.Laight@aculab.com> wrote:
> >
> > From: Eric Dumazet
> > > Sent: 04 December 2021 04:41
> > >
> > > On Fri, Dec 3, 2021 at 7:34 PM kernel test robot <lkp@intel.com> wrote:
> > > >
> > > > I love your patch! Perhaps something to improve:
> > ...
> > >
> > > Yes, keeping sparse happy with these checksum is not easy.
> > >
> > > I will add and use this helper, unless someone has a better idea.
> > >
> > > diff --git a/include/net/checksum.h b/include/net/checksum.h
> > > index 5b96d5bd6e54532a7a82511ff5d7d4c6f18982d5..5218041e5c8f93cd369a2a3a46add3e6a5e41af7
> > > 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
> >
> > I was thinking that the expression also requires some comments.
> > So maybe put a #define / static inline in checksum.h like:
> >
> > /* Subract the checksum of a buffer.
> >  * The domain is __wsum is [1..~0u] (ie excludes zero)
> >  * so ~csum_partial() cannot be used.
> >  * The two's compliment gives the right answer provided the old 'csum'
> >  * isn't zero - which it shouldn't be. */
> > #define csum_partial_sub(buf, len, csum) (-csum_partial(buf, len, -(csum))
> >
> > and then add the annotations there to keep sparse happy there.
> >
> > will sparse accept '1 + ~csum' ? The compiler should use negate for it.
> > It actually makes it slightly more obvious why the code is right.
> 
> Sparse is not happy with  1 + ~csum,
> 
> So we unfortunately would need something ugly like
> 
> (__force __wsum)(1 + ~(__force u32)csum)
> 
> Which most readers of this code will not find obvious.

Sparse really isn't helping here at all.
Perhaps there should be a way of marking a function so that
sparse just ignores it.

I also rather dislike that the compiler sees the (u32)csum cast.
The sparse annotation should really be either __sparse(u32)csum
or __sparse(u32, csum) to that compiler type checking still applies.

Perhaps adding:
#define WSUM(val) (__force __wsum)(val)
#define U32(csum)  (__force u32)(csum)
before the 'static inlines' in checksum.h and #undeffing them at the end.
Then all the functions could be made a little easier to read.

	return WSUM(1 + ~U32(csum_partial(buf, len, WSUM(1 + ~U32(csum)));

is a bit better than the casts - still not nice.

	David



	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)

^ permalink raw reply	[flat|nested] 8+ messages in thread

* RE: [PATCH net-next] net: fix recent csum changes
  2021-12-04 19:03         ` David Laight
@ 2021-12-04 19:18           ` David Laight
  0 siblings, 0 replies; 8+ messages in thread
From: David Laight @ 2021-12-04 19:18 UTC (permalink / raw)
  To: David Laight, 'Eric Dumazet'
  Cc: kernel test robot, Eric Dumazet, David S . Miller, Jakub Kicinski,
	kbuild-all@lists.01.org, netdev, Vladimir Oltean, David Lebrun

From: David Laight <David.Laight@ACULAB.COM>
> Sent: 04 December 2021 19:03
> 
> From: Eric Dumazet
> > Sent: 04 December 2021 18:34
> >
> > On Sat, Dec 4, 2021 at 6:00 AM David Laight <David.Laight@aculab.com> wrote:
> > >
> > > From: Eric Dumazet
> > > > Sent: 04 December 2021 04:41
> > > >
> > > > On Fri, Dec 3, 2021 at 7:34 PM kernel test robot <lkp@intel.com> wrote:
> > > > >
> > > > > I love your patch! Perhaps something to improve:
> > > ...
> > > >
> > > > Yes, keeping sparse happy with these checksum is not easy.
> > > >
> > > > I will add and use this helper, unless someone has a better idea.
> > > >
> > > > diff --git a/include/net/checksum.h b/include/net/checksum.h
> > > > index 5b96d5bd6e54532a7a82511ff5d7d4c6f18982d5..5218041e5c8f93cd369a2a3a46add3e6a5e41af7
> > > > 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
> > >
> > > I was thinking that the expression also requires some comments.
> > > So maybe put a #define / static inline in checksum.h like:
> > >
> > > /* Subract the checksum of a buffer.
> > >  * The domain is __wsum is [1..~0u] (ie excludes zero)
> > >  * so ~csum_partial() cannot be used.
> > >  * The two's compliment gives the right answer provided the old 'csum'
> > >  * isn't zero - which it shouldn't be. */
> > > #define csum_partial_sub(buf, len, csum) (-csum_partial(buf, len, -(csum))
> > >
> > > and then add the annotations there to keep sparse happy there.
> > >
> > > will sparse accept '1 + ~csum' ? The compiler should use negate for it.
> > > It actually makes it slightly more obvious why the code is right.
> >
> > Sparse is not happy with  1 + ~csum,
> >
> > So we unfortunately would need something ugly like
> >
> > (__force __wsum)(1 + ~(__force u32)csum)
> >
> > Which most readers of this code will not find obvious.
> 
> Sparse really isn't helping here at all.
> Perhaps there should be a way of marking a function so that
> sparse just ignores it.
> 
> I also rather dislike that the compiler sees the (u32)csum cast.
> The sparse annotation should really be either __sparse(u32)csum
> or __sparse(u32, csum) to that compiler type checking still applies.
> 
> Perhaps adding:
> #define WSUM(val) (__force __wsum)(val)
> #define U32(csum)  (__force u32)(csum)
> before the 'static inlines' in checksum.h and #undeffing them at the end.
> Then all the functions could be made a little easier to read.
> 
> 	return WSUM(1 + ~U32(csum_partial(buf, len, WSUM(1 + ~U32(csum)));
> 
> is a bit better than the casts - still not nice.

Thinking again...
#ifdef __sparse
	/* Skip offsetting to save 4 casts. */
	return ~csum_partial(buf, len, ~csum);
#else
	/* Offset by one so that zero is never returned. */
	return 1 + ~csum_partial(bufg, len, 1 + ~csum);
#endif

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)

^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2021-12-04 19:24 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2021-12-03 18:52 [PATCH net-next] net: fix recent csum changes Eric Dumazet
2021-12-04  3:34 ` kernel test robot
2021-12-04  3:34 ` kernel test robot
2021-12-04  4:40   ` Eric Dumazet
2021-12-04 14:00     ` David Laight
2021-12-04 18:33       ` Eric Dumazet
2021-12-04 19:03         ` David Laight
2021-12-04 19:18           ` David Laight

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).