[PATCH net-next v2 0/4] rds: Fix semantic annotations

[PATCH net-next v2 0/4] rds: Fix semantic annotations

[Ujwal Kundur]

This patchset addresses all semantic warnings flagged by Sparse for
net/rds.

v1:
 - https://lore.kernel.org/all/20250810171155.3263-1-ujwal.kundur@gmail.com/

Ujwal Kundur (4):
  rds: Replace POLLERR with EPOLLERR
  rds: Fix endianness annotation of jhash wrappers
  rds: Fix endianness annotation for RDS_MPATH_HASH
  rds: Fix endianness annotations for RDS extension headers

 net/rds/af_rds.c     | 2 +-
 net/rds/connection.c | 9 +++++----
 net/rds/message.c    | 4 ++--
 net/rds/rds.h        | 2 +-
 net/rds/recv.c       | 4 ++--
 net/rds/send.c       | 4 ++--
 6 files changed, 13 insertions(+), 12 deletions(-)

-- 
2.30.2

[PATCH net-next v2 1/4] rds: Replace POLLERR with EPOLLERR

[Ujwal Kundur]

Both constants are 1<<3, but EPOLLERR uses the correct annotations.

Flagged by Sparse.

Signed-off-by: Ujwal Kundur <ujwal.kundur@gmail.com>
---
 net/rds/af_rds.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/rds/af_rds.c b/net/rds/af_rds.c
index 086a13170e09..4a7217fbeab6 100644
--- a/net/rds/af_rds.c
+++ b/net/rds/af_rds.c
@@ -242,7 +242,7 @@ static __poll_t rds_poll(struct file *file, struct socket *sock,
 	if (rs->rs_snd_bytes < rds_sk_sndbuf(rs))
 		mask |= (EPOLLOUT | EPOLLWRNORM);
 	if (sk->sk_err || !skb_queue_empty(&sk->sk_error_queue))
-		mask |= POLLERR;
+		mask |= EPOLLERR;
 	read_unlock_irqrestore(&rs->rs_recv_lock, flags);
 
 	/* clear state any time we wake a seen-congested socket */
-- 
2.30.2

Re: [PATCH net-next v2 1/4] rds: Replace POLLERR with EPOLLERR

[Allison Henderson]

On Wed, 2025-08-20 at 23:25 +0530, Ujwal Kundur wrote:
> Both constants are 1<<3, but EPOLLERR uses the correct annotations.
> 
> Flagged by Sparse.
> 
> Signed-off-by: Ujwal Kundur <ujwal.kundur@gmail.com>
> ---
>  net/rds/af_rds.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/net/rds/af_rds.c b/net/rds/af_rds.c
> index 086a13170e09..4a7217fbeab6 100644
> --- a/net/rds/af_rds.c
> +++ b/net/rds/af_rds.c
> @@ -242,7 +242,7 @@ static __poll_t rds_poll(struct file *file, struct socket *sock,
>  	if (rs->rs_snd_bytes < rds_sk_sndbuf(rs))
>  		mask |= (EPOLLOUT | EPOLLWRNORM);
>  	if (sk->sk_err || !skb_queue_empty(&sk->sk_error_queue))
> -		mask |= POLLERR;
> +		mask |= EPOLLERR;
>  	read_unlock_irqrestore(&rs->rs_recv_lock, flags);
>  
>  	/* clear state any time we wake a seen-congested socket */

This looks better.  Thank you for the clean up!
Reviewed-by: Allison Henderson <allison.henderson@oracle.com>

[PATCH net-next v2 2/4] rds: Fix endianness annotation of jhash wrappers

[Ujwal Kundur]

__ipv6_addr_jhash (wrapper around jhash2()) and __inet_ehashfn (wrapper
around jhash_3words()) work with u32 (host endian) values but accept big
endian inputs. Declare the local variables as big endian to avoid
unnecessary casts.

Flagged by Sparse.

Signed-off-by: Ujwal Kundur <ujwal.kundur@gmail.com>
---
 net/rds/connection.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/net/rds/connection.c b/net/rds/connection.c
index d62f486ab29f..ba6fb87647ac 100644
--- a/net/rds/connection.c
+++ b/net/rds/connection.c
@@ -57,16 +57,17 @@ static struct hlist_head *rds_conn_bucket(const struct in6_addr *laddr,
 	static u32 rds6_hash_secret __read_mostly;
 	static u32 rds_hash_secret __read_mostly;
 
-	u32 lhash, fhash, hash;
+	u32 hash;
+	__be32 lhash, fhash;
 
 	net_get_random_once(&rds_hash_secret, sizeof(rds_hash_secret));
 	net_get_random_once(&rds6_hash_secret, sizeof(rds6_hash_secret));
 
-	lhash = (__force u32)laddr->s6_addr32[3];
+	lhash = laddr->s6_addr32[3];
 #if IS_ENABLED(CONFIG_IPV6)
-	fhash = __ipv6_addr_jhash(faddr, rds6_hash_secret);
+	fhash = (__force __be32)__ipv6_addr_jhash(faddr, rds6_hash_secret);
 #else
-	fhash = (__force u32)faddr->s6_addr32[3];
+	fhash = faddr->s6_addr32[3];
 #endif
 	hash = __inet_ehashfn(lhash, 0, fhash, 0, rds_hash_secret);
 
-- 
2.30.2

Re: [PATCH net-next v2 2/4] rds: Fix endianness annotation of jhash wrappers

[Allison Henderson]

On Wed, 2025-08-20 at 23:25 +0530, Ujwal Kundur wrote:
> __ipv6_addr_jhash (wrapper around jhash2()) and __inet_ehashfn (wrapper
> around jhash_3words()) work with u32 (host endian) values but accept big
> endian inputs. Declare the local variables as big endian to avoid
> unnecessary casts.
> 
> Flagged by Sparse.
> 
> Signed-off-by: Ujwal Kundur <ujwal.kundur@gmail.com>
This looks ok now.  Thank you!
Reviewed-by: Allison Henderson <allison.henderson@oracle.com>
> ---
>  net/rds/connection.c | 9 +++++----
>  1 file changed, 5 insertions(+), 4 deletions(-)
> 
> diff --git a/net/rds/connection.c b/net/rds/connection.c
> index d62f486ab29f..ba6fb87647ac 100644
> --- a/net/rds/connection.c
> +++ b/net/rds/connection.c
> @@ -57,16 +57,17 @@ static struct hlist_head *rds_conn_bucket(const struct in6_addr *laddr,
>  	static u32 rds6_hash_secret __read_mostly;
>  	static u32 rds_hash_secret __read_mostly;
>  
> -	u32 lhash, fhash, hash;
> +	u32 hash;
> +	__be32 lhash, fhash;
>  
>  	net_get_random_once(&rds_hash_secret, sizeof(rds_hash_secret));
>  	net_get_random_once(&rds6_hash_secret, sizeof(rds6_hash_secret));
>  
> -	lhash = (__force u32)laddr->s6_addr32[3];
> +	lhash = laddr->s6_addr32[3];
>  #if IS_ENABLED(CONFIG_IPV6)
> -	fhash = __ipv6_addr_jhash(faddr, rds6_hash_secret);
> +	fhash = (__force __be32)__ipv6_addr_jhash(faddr, rds6_hash_secret);
>  #else
> -	fhash = (__force u32)faddr->s6_addr32[3];
> +	fhash = faddr->s6_addr32[3];
>  #endif
>  	hash = __inet_ehashfn(lhash, 0, fhash, 0, rds_hash_secret);
>  

[PATCH net-next v2 3/4] rds: Fix endianness annotation for RDS_MPATH_HASH

[Ujwal Kundur]

jhash_1word accepts host endian inputs while rs_bound_port is a be16
value (sockaddr_in6.sin6_port). Use ntohs() for consistency.

Flagged by Sparse.

Signed-off-by: Ujwal Kundur <ujwal.kundur@gmail.com>
---
 net/rds/rds.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/rds/rds.h b/net/rds/rds.h
index dc360252c515..5b1c072e2e7f 100644
--- a/net/rds/rds.h
+++ b/net/rds/rds.h
@@ -93,7 +93,7 @@ enum {
 
 /* Max number of multipaths per RDS connection. Must be a power of 2 */
 #define	RDS_MPATH_WORKERS	8
-#define	RDS_MPATH_HASH(rs, n) (jhash_1word((rs)->rs_bound_port, \
+#define	RDS_MPATH_HASH(rs, n) (jhash_1word(ntohs((rs)->rs_bound_port), \
 			       (rs)->rs_hash_initval) & ((n) - 1))
 
 #define IS_CANONICAL(laddr, faddr) (htonl(laddr) < htonl(faddr))
-- 
2.30.2

Re: [PATCH net-next v2 3/4] rds: Fix endianness annotation for RDS_MPATH_HASH

[Allison Henderson]

On Wed, 2025-08-20 at 23:25 +0530, Ujwal Kundur wrote:
> jhash_1word accepts host endian inputs while rs_bound_port is a be16
> value (sockaddr_in6.sin6_port). Use ntohs() for consistency.
> 
> Flagged by Sparse.
> 
> Signed-off-by: Ujwal Kundur <ujwal.kundur@gmail.com>
This looks fine.  Thank you!
Reviewed-by: Allison Henderson <allison.henderson@oracle.com>

> ---
>  net/rds/rds.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/net/rds/rds.h b/net/rds/rds.h
> index dc360252c515..5b1c072e2e7f 100644
> --- a/net/rds/rds.h
> +++ b/net/rds/rds.h
> @@ -93,7 +93,7 @@ enum {
>  
>  /* Max number of multipaths per RDS connection. Must be a power of 2 */
>  #define	RDS_MPATH_WORKERS	8
> -#define	RDS_MPATH_HASH(rs, n) (jhash_1word((rs)->rs_bound_port, \
> +#define	RDS_MPATH_HASH(rs, n) (jhash_1word(ntohs((rs)->rs_bound_port), \
>  			       (rs)->rs_hash_initval) & ((n) - 1))
>  
>  #define IS_CANONICAL(laddr, faddr) (htonl(laddr) < htonl(faddr))

[PATCH net-next v2 4/4] rds: Fix endianness annotations for RDS extension headers

[Ujwal Kundur]

Per the RDS 3.1 spec [1], RDS extension headers EXTHDR_NPATHS and
EXTHDR_GEN_NUM are be16 and be32 values respectively, exchanged during
normal operations over-the-wire (RDS Ping/Pong). This contrasts their
declarations as host endian unsigned ints.

Fix the annotations across occurrences. Flagged by Sparse.

[1] https://oss.oracle.com/projects/rds/dist/documentation/rds-3.1-spec.html

Signed-off-by: Ujwal Kundur <ujwal.kundur@gmail.com>
---
 net/rds/message.c | 4 ++--
 net/rds/recv.c    | 4 ++--
 net/rds/send.c    | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/net/rds/message.c b/net/rds/message.c
index 7af59d2443e5..199a899a43e9 100644
--- a/net/rds/message.c
+++ b/net/rds/message.c
@@ -44,8 +44,8 @@ static unsigned int	rds_exthdr_size[__RDS_EXTHDR_MAX] = {
 [RDS_EXTHDR_VERSION]	= sizeof(struct rds_ext_header_version),
 [RDS_EXTHDR_RDMA]	= sizeof(struct rds_ext_header_rdma),
 [RDS_EXTHDR_RDMA_DEST]	= sizeof(struct rds_ext_header_rdma_dest),
-[RDS_EXTHDR_NPATHS]	= sizeof(u16),
-[RDS_EXTHDR_GEN_NUM]	= sizeof(u32),
+[RDS_EXTHDR_NPATHS]	= sizeof(__be16),
+[RDS_EXTHDR_GEN_NUM]	= sizeof(__be32),
 };
 
 void rds_message_addref(struct rds_message *rm)
diff --git a/net/rds/recv.c b/net/rds/recv.c
index 5627f80013f8..66205d6924bf 100644
--- a/net/rds/recv.c
+++ b/net/rds/recv.c
@@ -202,8 +202,8 @@ static void rds_recv_hs_exthdrs(struct rds_header *hdr,
 	unsigned int pos = 0, type, len;
 	union {
 		struct rds_ext_header_version version;
-		u16 rds_npaths;
-		u32 rds_gen_num;
+		__be16 rds_npaths;
+		__be32 rds_gen_num;
 	} buffer;
 	u32 new_peer_gen_num = 0;
 
diff --git a/net/rds/send.c b/net/rds/send.c
index 42d991bc8543..0b3d0ef2f008 100644
--- a/net/rds/send.c
+++ b/net/rds/send.c
@@ -1454,8 +1454,8 @@ rds_send_probe(struct rds_conn_path *cp, __be16 sport,
 
 	if (RDS_HS_PROBE(be16_to_cpu(sport), be16_to_cpu(dport)) &&
 	    cp->cp_conn->c_trans->t_mp_capable) {
-		u16 npaths = cpu_to_be16(RDS_MPATH_WORKERS);
-		u32 my_gen_num = cpu_to_be32(cp->cp_conn->c_my_gen_num);
+		__be16 npaths = cpu_to_be16(RDS_MPATH_WORKERS);
+		__be32 my_gen_num = cpu_to_be32(cp->cp_conn->c_my_gen_num);
 
 		rds_message_add_extension(&rm->m_inc.i_hdr,
 					  RDS_EXTHDR_NPATHS, &npaths,
-- 
2.30.2

Re: [PATCH net-next v2 4/4] rds: Fix endianness annotations for RDS extension headers

[Allison Henderson]

On Wed, 2025-08-20 at 23:25 +0530, Ujwal Kundur wrote:
> Per the RDS 3.1 spec [1], RDS extension headers EXTHDR_NPATHS and
> EXTHDR_GEN_NUM are be16 and be32 values respectively, exchanged during
> normal operations over-the-wire (RDS Ping/Pong). This contrasts their
> declarations as host endian unsigned ints.
> 
> Fix the annotations across occurrences. Flagged by Sparse.
> 
> [1] https://oss.oracle.com/projects/rds/dist/documentation/rds-3.1-spec.html 
> 
> Signed-off-by: Ujwal Kundur <ujwal.kundur@gmail.com>
This looks much better now.  Thank you!
Reviewed-by: Allison Henderson <allison.henderson@oracle.com>
> ---
>  net/rds/message.c | 4 ++--
>  net/rds/recv.c    | 4 ++--
>  net/rds/send.c    | 4 ++--
>  3 files changed, 6 insertions(+), 6 deletions(-)
> 
> diff --git a/net/rds/message.c b/net/rds/message.c
> index 7af59d2443e5..199a899a43e9 100644
> --- a/net/rds/message.c
> +++ b/net/rds/message.c
> @@ -44,8 +44,8 @@ static unsigned int	rds_exthdr_size[__RDS_EXTHDR_MAX] = {
>  [RDS_EXTHDR_VERSION]	= sizeof(struct rds_ext_header_version),
>  [RDS_EXTHDR_RDMA]	= sizeof(struct rds_ext_header_rdma),
>  [RDS_EXTHDR_RDMA_DEST]	= sizeof(struct rds_ext_header_rdma_dest),
> -[RDS_EXTHDR_NPATHS]	= sizeof(u16),
> -[RDS_EXTHDR_GEN_NUM]	= sizeof(u32),
> +[RDS_EXTHDR_NPATHS]	= sizeof(__be16),
> +[RDS_EXTHDR_GEN_NUM]	= sizeof(__be32),
>  };
>  
>  void rds_message_addref(struct rds_message *rm)
> diff --git a/net/rds/recv.c b/net/rds/recv.c
> index 5627f80013f8..66205d6924bf 100644
> --- a/net/rds/recv.c
> +++ b/net/rds/recv.c
> @@ -202,8 +202,8 @@ static void rds_recv_hs_exthdrs(struct rds_header *hdr,
>  	unsigned int pos = 0, type, len;
>  	union {
>  		struct rds_ext_header_version version;
> -		u16 rds_npaths;
> -		u32 rds_gen_num;
> +		__be16 rds_npaths;
> +		__be32 rds_gen_num;
>  	} buffer;
>  	u32 new_peer_gen_num = 0;
>  
> diff --git a/net/rds/send.c b/net/rds/send.c
> index 42d991bc8543..0b3d0ef2f008 100644
> --- a/net/rds/send.c
> +++ b/net/rds/send.c
> @@ -1454,8 +1454,8 @@ rds_send_probe(struct rds_conn_path *cp, __be16 sport,
>  
>  	if (RDS_HS_PROBE(be16_to_cpu(sport), be16_to_cpu(dport)) &&
>  	    cp->cp_conn->c_trans->t_mp_capable) {
> -		u16 npaths = cpu_to_be16(RDS_MPATH_WORKERS);
> -		u32 my_gen_num = cpu_to_be32(cp->cp_conn->c_my_gen_num);
> +		__be16 npaths = cpu_to_be16(RDS_MPATH_WORKERS);
> +		__be32 my_gen_num = cpu_to_be32(cp->cp_conn->c_my_gen_num);
>  
>  		rds_message_add_extension(&rm->m_inc.i_hdr,
>  					  RDS_EXTHDR_NPATHS, &npaths,

Re: [PATCH net-next v2 0/4] rds: Fix semantic annotations

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Wed, 20 Aug 2025 23:25:46 +0530 you wrote:
> This patchset addresses all semantic warnings flagged by Sparse for
> net/rds.
> 
> v1:
>  - https://lore.kernel.org/all/20250810171155.3263-1-ujwal.kundur@gmail.com/
> 
> Ujwal Kundur (4):
>   rds: Replace POLLERR with EPOLLERR
>   rds: Fix endianness annotation of jhash wrappers
>   rds: Fix endianness annotation for RDS_MPATH_HASH
>   rds: Fix endianness annotations for RDS extension headers
> 
> [...]

Here is the summary with links:
  - [net-next,v2,1/4] rds: Replace POLLERR with EPOLLERR
    https://git.kernel.org/netdev/net-next/c/9308987803bb
  - [net-next,v2,2/4] rds: Fix endianness annotation of jhash wrappers
    https://git.kernel.org/netdev/net-next/c/92b925297a2f
  - [net-next,v2,3/4] rds: Fix endianness annotation for RDS_MPATH_HASH
    https://git.kernel.org/netdev/net-next/c/77907a068717
  - [net-next,v2,4/4] rds: Fix endianness annotations for RDS extension headers
    https://git.kernel.org/netdev/net-next/c/bcb28bee987a

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html