From 094bb20b9fcab3a1652a77741caba6b78097d622 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 Aug 2005 16:25:20 -0400 Subject: [PATCH] RPC: extract socket logic common to both client and server Clean-up: Move some code that is common to both RPC client- and server-side socket transports into its own source file, net/sunrpc/socklib.c. Test-plan: Compile kernel with CONFIG_NFS enabled. Millions of fsx operations over UDP, client and server. Connectathon over UDP. Version: Thu, 11 Aug 2005 16:03:09 -0400 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xdr.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 23448d0fb5bc..d8b7656bca41 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -161,6 +161,7 @@ typedef struct { typedef size_t (*skb_read_actor_t)(skb_reader_t *desc, void *to, size_t len); +extern int csum_partial_copy_to_xdr(struct xdr_buf *, struct sk_buff *); extern ssize_t xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, skb_reader_t *, skb_read_actor_t); -- cgit v1.2.3 From a246b0105bbd9a70a698f69baae2042996f2a0e9 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 Aug 2005 16:25:23 -0400 Subject: [PATCH] RPC: introduce client-side transport switch Move the bulk of client-side socket-specific code into a separate source file, net/sunrpc/xprtsock.c. Test-plan: Millions of fsx operations. Performance characterization such as "sio" or "iozone". Destructive testing (unplugging the network temporarily, server reboots). Connectathon with v2, v3, and v4. Version: Thu, 11 Aug 2005 16:03:38 -0400 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xdr.h | 5 ----- include/linux/sunrpc/xprt.h | 38 +++++++++++++++++++++++++++++++------- 2 files changed, 31 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index d8b7656bca41..5da968729cf8 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -165,11 +165,6 @@ extern int csum_partial_copy_to_xdr(struct xdr_buf *, struct sk_buff *); extern ssize_t xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, skb_reader_t *, skb_read_actor_t); -struct socket; -struct sockaddr; -extern int xdr_sendpages(struct socket *, struct sockaddr *, int, - struct xdr_buf *, unsigned int, int); - extern int xdr_encode_word(struct xdr_buf *, int, u32); extern int xdr_decode_word(struct xdr_buf *, int, u32 *); diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index e618c1649814..d82b47ab73cb 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -59,7 +59,13 @@ extern unsigned int xprt_tcp_slot_table_entries; */ #define RPC_REESTABLISH_TIMEOUT (15*HZ) -/* RPC call and reply header size as number of 32bit words (verifier +/* + * RPC transport idle timeout. + */ +#define RPC_IDLE_DISCONNECT_TIMEOUT (5*60*HZ) + +/* + * RPC call and reply header size as number of 32bit words (verifier * size computed separately) */ #define RPC_CALLHDRSIZE 6 @@ -121,12 +127,19 @@ struct rpc_rqst { #define rq_svec rq_snd_buf.head #define rq_slen rq_snd_buf.len -#define XPRT_LAST_FRAG (1 << 0) -#define XPRT_COPY_RECM (1 << 1) -#define XPRT_COPY_XID (1 << 2) -#define XPRT_COPY_DATA (1 << 3) +struct rpc_task; +struct rpc_xprt; + +struct rpc_xprt_ops { + void (*set_buffer_size)(struct rpc_xprt *xprt); + void (*connect)(struct rpc_task *task); + int (*send_request)(struct rpc_task *task); + void (*close)(struct rpc_xprt *xprt); + void (*destroy)(struct rpc_xprt *xprt); +}; struct rpc_xprt { + struct rpc_xprt_ops * ops; /* transport methods */ struct socket * sock; /* BSD socket layer */ struct sock * inet; /* INET layer */ @@ -199,14 +212,22 @@ struct rpc_xprt { wait_queue_head_t cong_wait; }; +#define XPRT_LAST_FRAG (1 << 0) +#define XPRT_COPY_RECM (1 << 1) +#define XPRT_COPY_XID (1 << 2) +#define XPRT_COPY_DATA (1 << 3) + #ifdef __KERNEL__ struct rpc_xprt * xprt_create_proto(int proto, struct sockaddr_in *addr, struct rpc_timeout *toparms); +void xprt_disconnect(struct rpc_xprt *); int xprt_destroy(struct rpc_xprt *); void xprt_set_timeout(struct rpc_timeout *, unsigned int, unsigned long); - +struct rpc_rqst * xprt_lookup_rqst(struct rpc_xprt *, u32); +void xprt_complete_rqst(struct rpc_xprt *, + struct rpc_rqst *, int); void xprt_reserve(struct rpc_task *); int xprt_prepare_transmit(struct rpc_task *); void xprt_transmit(struct rpc_task *); @@ -214,7 +235,10 @@ void xprt_receive(struct rpc_task *); int xprt_adjust_timeout(struct rpc_rqst *req); void xprt_release(struct rpc_task *); void xprt_connect(struct rpc_task *); -void xprt_sock_setbufsize(struct rpc_xprt *); +int xs_setup_udp(struct rpc_xprt *, + struct rpc_timeout *); +int xs_setup_tcp(struct rpc_xprt *, + struct rpc_timeout *); #define XPRT_LOCKED 0 #define XPRT_CONNECT 1 -- cgit v1.2.3 From 9903cd1c27a1f30e8efea75e125be3b2002f7cb9 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 Aug 2005 16:25:26 -0400 Subject: [PATCH] RPC: transport switch function naming Introduce block header comments and a function naming convention to the socket transport implementation. Provide a debug setting for transports that is separate from RPCDBG_XPRT. Eliminate xprt_default_timeout(). Provide block comments for exposed interfaces in xprt.c, and eliminate the useless obvious comments. Convert printk's to dprintk's. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:04:04 -0400 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/debug.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/debug.h b/include/linux/sunrpc/debug.h index eadb31e3c198..42d299747956 100644 --- a/include/linux/sunrpc/debug.h +++ b/include/linux/sunrpc/debug.h @@ -32,6 +32,7 @@ #define RPCDBG_AUTH 0x0010 #define RPCDBG_PMAP 0x0020 #define RPCDBG_SCHED 0x0040 +#define RPCDBG_TRANS 0x0080 #define RPCDBG_SVCSOCK 0x0100 #define RPCDBG_SVCDSP 0x0200 #define RPCDBG_MISC 0x0400 -- cgit v1.2.3 From 4a0f8c04f2ece949d54a0c4fd7490259cf23a58a Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 Aug 2005 16:25:32 -0400 Subject: [PATCH] RPC: Rename sock_lock Clean-up: replace a name reference to sockets in the generic parts of the RPC client by renaming sock_lock in the rpc_xprt structure. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:05:00 -0400 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index d82b47ab73cb..c4f903f0e17c 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -198,7 +198,7 @@ struct rpc_xprt { /* * Send stuff */ - spinlock_t sock_lock; /* lock socket info */ + spinlock_t transport_lock; /* lock transport info */ spinlock_t xprt_lock; /* lock xprt info */ struct rpc_task * snd_task; /* Task blocked in send */ -- cgit v1.2.3 From 5dc07727f86b25851e95193a0c484ea21b531c47 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 Aug 2005 16:25:35 -0400 Subject: [PATCH] RPC: Rename xprt_lock Clean-up: Replace the xprt_lock with something more aptly named. This lock single-threads the XID and request slot reservation process. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:05:26 -0400 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index c4f903f0e17c..41ce296dded1 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -199,7 +199,7 @@ struct rpc_xprt { * Send stuff */ spinlock_t transport_lock; /* lock transport info */ - spinlock_t xprt_lock; /* lock xprt info */ + spinlock_t reserve_lock; /* lock slot table */ struct rpc_task * snd_task; /* Task blocked in send */ struct list_head recv; -- cgit v1.2.3 From 2226feb6bcd0e5e117a9be3ea3dd3ffc14f3e41e Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 Aug 2005 16:25:38 -0400 Subject: [PATCH] RPC: rename the sockstate field Clean-up: get rid of a name reference to sockets in the generic parts of the RPC client by renaming the sockstate field in the rpc_xprt structure. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:05:53 -0400 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 60 ++++++++++++++++++++++++++++++++++++--------- 1 file changed, 49 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 41ce296dded1..009a3bb4f997 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -163,7 +163,7 @@ struct rpc_xprt { struct list_head free; /* free slots */ struct rpc_rqst * slot; /* slot table storage */ unsigned int max_reqs; /* total slots */ - unsigned long sockstate; /* Socket state */ + unsigned long state; /* transport state */ unsigned char shutdown : 1, /* being shut down */ nocong : 1, /* no congestion control */ resvport : 1, /* use a reserved port */ @@ -240,16 +240,54 @@ int xs_setup_udp(struct rpc_xprt *, int xs_setup_tcp(struct rpc_xprt *, struct rpc_timeout *); -#define XPRT_LOCKED 0 -#define XPRT_CONNECT 1 -#define XPRT_CONNECTING 2 - -#define xprt_connected(xp) (test_bit(XPRT_CONNECT, &(xp)->sockstate)) -#define xprt_set_connected(xp) (set_bit(XPRT_CONNECT, &(xp)->sockstate)) -#define xprt_test_and_set_connected(xp) (test_and_set_bit(XPRT_CONNECT, &(xp)->sockstate)) -#define xprt_test_and_clear_connected(xp) \ - (test_and_clear_bit(XPRT_CONNECT, &(xp)->sockstate)) -#define xprt_clear_connected(xp) (clear_bit(XPRT_CONNECT, &(xp)->sockstate)) +/* + * Reserved bit positions in xprt->state + */ +#define XPRT_LOCKED (0) +#define XPRT_CONNECTED (1) +#define XPRT_CONNECTING (2) + +static inline void xprt_set_connected(struct rpc_xprt *xprt) +{ + set_bit(XPRT_CONNECTED, &xprt->state); +} + +static inline void xprt_clear_connected(struct rpc_xprt *xprt) +{ + clear_bit(XPRT_CONNECTED, &xprt->state); +} + +static inline int xprt_connected(struct rpc_xprt *xprt) +{ + return test_bit(XPRT_CONNECTED, &xprt->state); +} + +static inline int xprt_test_and_set_connected(struct rpc_xprt *xprt) +{ + return test_and_set_bit(XPRT_CONNECTED, &xprt->state); +} + +static inline int xprt_test_and_clear_connected(struct rpc_xprt *xprt) +{ + return test_and_clear_bit(XPRT_CONNECTED, &xprt->state); +} + +static inline void xprt_clear_connecting(struct rpc_xprt *xprt) +{ + smp_mb__before_clear_bit(); + clear_bit(XPRT_CONNECTING, &xprt->state); + smp_mb__after_clear_bit(); +} + +static inline int xprt_connecting(struct rpc_xprt *xprt) +{ + return test_bit(XPRT_CONNECTING, &xprt->state); +} + +static inline int xprt_test_and_set_connecting(struct rpc_xprt *xprt) +{ + return test_and_set_bit(XPRT_CONNECTING, &xprt->state); +} #endif /* __KERNEL__*/ -- cgit v1.2.3 From 44fbac2288dfed6f1963ac00bf922c3bcd779cd1 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 Aug 2005 16:25:44 -0400 Subject: [PATCH] RPC: Add helper for waking tasks pending on a transport Clean-up: remove only reference to xprt->pending from the socket transport implementation. This makes a cleaner interface for other transport implementations as well. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:06:52 -0400 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 009a3bb4f997..d5223993fca9 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -232,6 +232,7 @@ void xprt_reserve(struct rpc_task *); int xprt_prepare_transmit(struct rpc_task *); void xprt_transmit(struct rpc_task *); void xprt_receive(struct rpc_task *); +void xprt_wake_pending_tasks(struct rpc_xprt *, int); int xprt_adjust_timeout(struct rpc_rqst *req); void xprt_release(struct rpc_task *); void xprt_connect(struct rpc_task *); -- cgit v1.2.3 From 55aa4f58aa43dc9a51fb80010630d94b96053a2e Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 Aug 2005 16:25:47 -0400 Subject: [PATCH] RPC: client-side transport switch cleanup Clean-up: change some comments to reflect the realities of the new RPC transport switch mechanism. Get rid of unused xprt_receive() prototype. Also, organize function prototypes in xprt.h by usage and scope. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:07:21 -0400 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 61 ++++++++++++++++++++++++++------------------- 1 file changed, 35 insertions(+), 26 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index d5223993fca9..bfbc492ae36d 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -1,5 +1,5 @@ /* - * linux/include/linux/sunrpc/clnt_xprt.h + * linux/include/linux/sunrpc/xprt.h * * Declarations for the RPC transport interface. * @@ -150,8 +150,8 @@ struct rpc_xprt { unsigned long cong; /* current congestion */ unsigned long cwnd; /* congestion window */ - unsigned int rcvsize, /* socket receive buffer size */ - sndsize; /* socket send buffer size */ + unsigned int rcvsize, /* transport rcv buffer size */ + sndsize; /* transport send buffer size */ size_t max_payload; /* largest RPC payload size, in bytes */ @@ -184,12 +184,12 @@ struct rpc_xprt { unsigned long tcp_copied, /* copied to request */ tcp_flags; /* - * Connection of sockets + * Connection of transports */ - struct work_struct sock_connect; + struct work_struct connect_worker; unsigned short port; /* - * Disconnection of idle sockets + * Disconnection of idle transports */ struct work_struct task_cleanup; struct timer_list timer; @@ -219,27 +219,36 @@ struct rpc_xprt { #ifdef __KERNEL__ -struct rpc_xprt * xprt_create_proto(int proto, struct sockaddr_in *addr, - struct rpc_timeout *toparms); -void xprt_disconnect(struct rpc_xprt *); -int xprt_destroy(struct rpc_xprt *); -void xprt_set_timeout(struct rpc_timeout *, unsigned int, - unsigned long); -struct rpc_rqst * xprt_lookup_rqst(struct rpc_xprt *, u32); -void xprt_complete_rqst(struct rpc_xprt *, - struct rpc_rqst *, int); -void xprt_reserve(struct rpc_task *); -int xprt_prepare_transmit(struct rpc_task *); -void xprt_transmit(struct rpc_task *); -void xprt_receive(struct rpc_task *); -void xprt_wake_pending_tasks(struct rpc_xprt *, int); +/* + * Transport operations used by ULPs + */ +struct rpc_xprt * xprt_create_proto(int proto, struct sockaddr_in *addr, struct rpc_timeout *to); +void xprt_set_timeout(struct rpc_timeout *to, unsigned int retr, unsigned long incr); + +/* + * Generic internal transport functions + */ +void xprt_connect(struct rpc_task *task); +void xprt_reserve(struct rpc_task *task); +int xprt_prepare_transmit(struct rpc_task *task); +void xprt_transmit(struct rpc_task *task); int xprt_adjust_timeout(struct rpc_rqst *req); -void xprt_release(struct rpc_task *); -void xprt_connect(struct rpc_task *); -int xs_setup_udp(struct rpc_xprt *, - struct rpc_timeout *); -int xs_setup_tcp(struct rpc_xprt *, - struct rpc_timeout *); +void xprt_release(struct rpc_task *task); +int xprt_destroy(struct rpc_xprt *xprt); + +/* + * Transport switch helper functions + */ +void xprt_wake_pending_tasks(struct rpc_xprt *xprt, int status); +struct rpc_rqst * xprt_lookup_rqst(struct rpc_xprt *xprt, u32 xid); +void xprt_complete_rqst(struct rpc_xprt *xprt, struct rpc_rqst *req, int copied); +void xprt_disconnect(struct rpc_xprt *xprt); + +/* + * Socket transport setup operations + */ +int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to); +int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to); /* * Reserved bit positions in xprt->state -- cgit v1.2.3 From c7b2cae8a634015b72941ba2fc6c4bc9b8d3a129 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 Aug 2005 16:25:50 -0400 Subject: [PATCH] RPC: separate TCP and UDP write space callbacks Split the socket write space callback function into a TCP version and UDP version, eliminating one dependence on the "xprt->stream" variable. Keep the common pieces of this path in xprt.c so other transports can use it too. Test-plan: Write-intensive workload on a single mount point. Version: Thu, 11 Aug 2005 16:07:51 -0400 Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index bfbc492ae36d..e73174c7e450 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -240,6 +240,8 @@ int xprt_destroy(struct rpc_xprt *xprt); * Transport switch helper functions */ void xprt_wake_pending_tasks(struct rpc_xprt *xprt, int status); +void xprt_wait_for_buffer_space(struct rpc_task *task); +void xprt_write_space(struct rpc_xprt *xprt); struct rpc_rqst * xprt_lookup_rqst(struct rpc_xprt *xprt, u32 xid); void xprt_complete_rqst(struct rpc_xprt *xprt, struct rpc_rqst *req, int copied); void xprt_disconnect(struct rpc_xprt *xprt); -- cgit v1.2.3 From 808012fbb23a52ec59352445d2076d175ad4ab26 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:49 -0700 Subject: [PATCH] RPC: skip over transport-specific heads automatically Add a generic mechanism for skipping over transport-specific headers when constructing an RPC request. This removes another "xprt->stream" dependency. Test-plan: Write-intensive workload on a single mount point (try both UDP and TCP). Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/msg_prot.h | 25 +++++++++++++++++++++++++ include/linux/sunrpc/xprt.h | 7 +++++++ 2 files changed, 32 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/msg_prot.h b/include/linux/sunrpc/msg_prot.h index 15f115332389..f43f237360ae 100644 --- a/include/linux/sunrpc/msg_prot.h +++ b/include/linux/sunrpc/msg_prot.h @@ -76,5 +76,30 @@ enum rpc_auth_stat { #define RPC_MAXNETNAMELEN 256 +/* + * From RFC 1831: + * + * "A record is composed of one or more record fragments. A record + * fragment is a four-byte header followed by 0 to (2**31) - 1 bytes of + * fragment data. The bytes encode an unsigned binary number; as with + * XDR integers, the byte order is from highest to lowest. The number + * encodes two values -- a boolean which indicates whether the fragment + * is the last fragment of the record (bit value 1 implies the fragment + * is the last fragment) and a 31-bit unsigned binary value which is the + * length in bytes of the fragment's data. The boolean value is the + * highest-order bit of the header; the length is the 31 low-order bits. + * (Note that this record specification is NOT in XDR standard form!)" + * + * The Linux RPC client always sends its requests in a single record + * fragment, limiting the maximum payload size for stream transports to + * 2GB. + */ + +typedef u32 rpc_fraghdr; + +#define RPC_LAST_STREAM_FRAGMENT (1U << 31) +#define RPC_FRAGMENT_SIZE_MASK (~RPC_LAST_STREAM_FRAGMENT) +#define RPC_MAX_FRAGMENT_SIZE ((1U << 31) - 1) + #endif /* __KERNEL__ */ #endif /* _LINUX_SUNRPC_MSGPROT_H_ */ diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index e73174c7e450..966c456a0f6d 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -155,6 +155,8 @@ struct rpc_xprt { size_t max_payload; /* largest RPC payload size, in bytes */ + unsigned int tsh_size; /* size of transport specific + header */ struct rpc_wait_queue sending; /* requests waiting to send */ struct rpc_wait_queue resend; /* requests waiting to resend */ @@ -236,6 +238,11 @@ int xprt_adjust_timeout(struct rpc_rqst *req); void xprt_release(struct rpc_task *task); int xprt_destroy(struct rpc_xprt *xprt); +static inline u32 *xprt_skip_transport_header(struct rpc_xprt *xprt, u32 *p) +{ + return p + xprt->tsh_size; +} + /* * Transport switch helper functions */ -- cgit v1.2.3 From 43118c29dea2b23798bd42a147015cceee7fa885 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:49 -0700 Subject: [PATCH] RPC: get rid of xprt->stream Now we can fix up the last few places that use the "xprt->stream" variable, and get rid of it from the rpc_xprt structure. Test-plan: Destructive testing (unplugging the network temporarily). Connectathon with UDP and TCP. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 966c456a0f6d..c9477f022efb 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -168,8 +168,7 @@ struct rpc_xprt { unsigned long state; /* transport state */ unsigned char shutdown : 1, /* being shut down */ nocong : 1, /* no congestion control */ - resvport : 1, /* use a reserved port */ - stream : 1; /* TCP */ + resvport : 1; /* use a reserved port */ /* * XID -- cgit v1.2.3 From fe3aca290f17ae4978bd73d02aa4029f1c9c024c Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:50 -0700 Subject: [PATCH] RPC: add API to set transport-specific timeouts Prepare the way to remove the "xprt->nocong" variable by adding a callout to the RPC client transport switch API to handle setting RPC retransmit timeouts. Add a pair of generic helper functions that provide the ability to set a simple fixed timeout, or to set a timeout based on the state of a round- trip estimator. Test-plan: Use WAN simulation to cause sporadic bursty packet loss. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index c9477f022efb..ac08e99a81cb 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -134,6 +134,7 @@ struct rpc_xprt_ops { void (*set_buffer_size)(struct rpc_xprt *xprt); void (*connect)(struct rpc_task *task); int (*send_request)(struct rpc_task *task); + void (*set_retrans_timeout)(struct rpc_task *task); void (*close)(struct rpc_xprt *xprt); void (*destroy)(struct rpc_xprt *xprt); }; @@ -245,6 +246,8 @@ static inline u32 *xprt_skip_transport_header(struct rpc_xprt *xprt, u32 *p) /* * Transport switch helper functions */ +void xprt_set_retrans_timeout_def(struct rpc_task *task); +void xprt_set_retrans_timeout_rtt(struct rpc_task *task); void xprt_wake_pending_tasks(struct rpc_xprt *xprt, int status); void xprt_wait_for_buffer_space(struct rpc_task *task); void xprt_write_space(struct rpc_xprt *xprt); -- cgit v1.2.3 From 12a804698b29d040b7cdd92e8a44b0e75164dae9 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:51 -0700 Subject: [PATCH] RPC: expose API for serializing access to RPC transports The next several patches introduce an API that allows transports to choose whether the RPC client provides congestion control or whether the transport itself provides it. The first method we abstract is the one that serializes access to the RPC transport to prevent the bytes from different requests from mingling together. This method provides proper request serialization and the opportunity to prevent new requests from being started because the transport is congested. The normal situation is for the transport to handle congestion control itself. Although NFS over UDP was first, it has been recognized after years of experience that having the transport provide congestion control is much better than doing it in the RPC client. Thus TCP, and probably every future transport implementation, will use the default method, xprt_lock_write, provided in xprt.c, which does not provide any kind of congestion control. UDP can continue using the xprt.c-provided Van Jacobson congestion avoidance implementation. Test-plan: Use WAN simulation to cause sporadic bursty packet loss. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index ac08e99a81cb..eee1c6877851 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -132,6 +132,7 @@ struct rpc_xprt; struct rpc_xprt_ops { void (*set_buffer_size)(struct rpc_xprt *xprt); + int (*reserve_xprt)(struct rpc_task *task); void (*connect)(struct rpc_task *task); int (*send_request)(struct rpc_task *task); void (*set_retrans_timeout)(struct rpc_task *task); @@ -232,6 +233,8 @@ void xprt_set_timeout(struct rpc_timeout *to, unsigned int retr, unsigned long */ void xprt_connect(struct rpc_task *task); void xprt_reserve(struct rpc_task *task); +int xprt_reserve_xprt(struct rpc_task *task); +int xprt_reserve_xprt_cong(struct rpc_task *task); int xprt_prepare_transmit(struct rpc_task *task); void xprt_transmit(struct rpc_task *task); int xprt_adjust_timeout(struct rpc_rqst *req); -- cgit v1.2.3 From 49e9a89086b3cae784a4868ca852863e4f4ea3fe Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:51 -0700 Subject: [PATCH] RPC: expose API for serializing access to RPC transports The next method we abstract is the one that releases a transport, allowing another task to have access to the transport. Again, one generic version of this is provided for transports that don't need the RPC client to perform congestion control, and one version is for transports that can use the original Van Jacobson implementation in xprt.c. Test-plan: Use WAN simulation to cause sporadic bursty packet loss. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index eee1c6877851..86833b725bb5 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -133,6 +133,7 @@ struct rpc_xprt; struct rpc_xprt_ops { void (*set_buffer_size)(struct rpc_xprt *xprt); int (*reserve_xprt)(struct rpc_task *task); + void (*release_xprt)(struct rpc_xprt *xprt, struct rpc_task *task); void (*connect)(struct rpc_task *task); int (*send_request)(struct rpc_task *task); void (*set_retrans_timeout)(struct rpc_task *task); @@ -238,6 +239,8 @@ int xprt_reserve_xprt_cong(struct rpc_task *task); int xprt_prepare_transmit(struct rpc_task *task); void xprt_transmit(struct rpc_task *task); int xprt_adjust_timeout(struct rpc_rqst *req); +void xprt_release_xprt(struct rpc_xprt *xprt, struct rpc_task *task); +void xprt_release_xprt_cong(struct rpc_xprt *xprt, struct rpc_task *task); void xprt_release(struct rpc_task *task); int xprt_destroy(struct rpc_xprt *xprt); -- cgit v1.2.3 From 46c0ee8bc4ad3743de05e8b8b20201df44dcb6d3 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:52 -0700 Subject: [PATCH] RPC: separate xprt_timer implementations Allow transports to hook the retransmit timer interrupt. Some transports calculate their congestion window here so that a retransmit timeout has immediate effect on the congestion window. Test-plan: Use WAN simulation to cause sporadic bursty packet loss. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 86833b725bb5..443c3f984cf9 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -137,6 +137,7 @@ struct rpc_xprt_ops { void (*connect)(struct rpc_task *task); int (*send_request)(struct rpc_task *task); void (*set_retrans_timeout)(struct rpc_task *task); + void (*timer)(struct rpc_task *task); void (*close)(struct rpc_xprt *xprt); void (*destroy)(struct rpc_xprt *xprt); }; @@ -257,6 +258,7 @@ void xprt_set_retrans_timeout_rtt(struct rpc_task *task); void xprt_wake_pending_tasks(struct rpc_xprt *xprt, int status); void xprt_wait_for_buffer_space(struct rpc_task *task); void xprt_write_space(struct rpc_xprt *xprt); +void xprt_adjust_cwnd(struct rpc_task *task, int result); struct rpc_rqst * xprt_lookup_rqst(struct rpc_xprt *xprt, u32 xid); void xprt_complete_rqst(struct rpc_xprt *xprt, struct rpc_rqst *req, int copied); void xprt_disconnect(struct rpc_xprt *xprt); -- cgit v1.2.3 From 1570c1e41eabf6b7031f3e4322a2cf1cbe319fee Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:52 -0700 Subject: [PATCH] RPC: add generic interface for adjusting the congestion window A new interface that allows transports to adjust their congestion window using the Van Jacobson implementation in xprt.c is provided. Test-plan: Use WAN simulation to cause sporadic bursty packet loss. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 443c3f984cf9..2e48752d55d9 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -258,9 +258,10 @@ void xprt_set_retrans_timeout_rtt(struct rpc_task *task); void xprt_wake_pending_tasks(struct rpc_xprt *xprt, int status); void xprt_wait_for_buffer_space(struct rpc_task *task); void xprt_write_space(struct rpc_xprt *xprt); +void xprt_update_rtt(struct rpc_task *task); void xprt_adjust_cwnd(struct rpc_task *task, int result); struct rpc_rqst * xprt_lookup_rqst(struct rpc_xprt *xprt, u32 xid); -void xprt_complete_rqst(struct rpc_xprt *xprt, struct rpc_rqst *req, int copied); +void xprt_complete_rqst(struct rpc_task *task, int copied); void xprt_disconnect(struct rpc_xprt *xprt); /* -- cgit v1.2.3 From a58dd398f5db4f73d5c581069fd70a4304cc4f0a Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:53 -0700 Subject: [PATCH] RPC: add a release_rqst callout to the RPC transport switch The final place where congestion control state is adjusted is in xprt_release, where each request is finally released. Add a callout there to allow transports to perform additional processing when a request is about to be released. Test-plan: Use WAN simulation to cause sporadic bursty packet loss. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 2e48752d55d9..64e77658fa30 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -138,6 +138,7 @@ struct rpc_xprt_ops { int (*send_request)(struct rpc_task *task); void (*set_retrans_timeout)(struct rpc_task *task); void (*timer)(struct rpc_task *task); + void (*release_request)(struct rpc_task *task); void (*close)(struct rpc_xprt *xprt); void (*destroy)(struct rpc_xprt *xprt); }; @@ -262,6 +263,7 @@ void xprt_update_rtt(struct rpc_task *task); void xprt_adjust_cwnd(struct rpc_task *task, int result); struct rpc_rqst * xprt_lookup_rqst(struct rpc_xprt *xprt, u32 xid); void xprt_complete_rqst(struct rpc_task *task, int copied); +void xprt_release_rqst_cong(struct rpc_task *task); void xprt_disconnect(struct rpc_xprt *xprt); /* -- cgit v1.2.3 From ed63c003701a314c4893c11eceb9d68f8f46c662 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:53 -0700 Subject: [PATCH] RPC: remove xprt->nocong Get rid of the "xprt->nocong" variable. Test-plan: Use WAN simulation to cause sporadic bursty packet loss with UDP mounts. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 64e77658fa30..559fb471f6f2 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -172,7 +172,6 @@ struct rpc_xprt { unsigned int max_reqs; /* total slots */ unsigned long state; /* transport state */ unsigned char shutdown : 1, /* being shut down */ - nocong : 1, /* no congestion control */ resvport : 1; /* use a reserved port */ /* -- cgit v1.2.3 From 555ee3af161b037865793bd4bebc06b58daafde6 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:54 -0700 Subject: [PATCH] RPC: clean up after nocong was removed Clean-up: Move some macros that are specific to the Van Jacobson implementation into xprt.c. Get rid of the cong_wait field in rpc_xprt, which is no longer used. Get rid of xprt_clear_backlog. Test-plan: Compile with CONFIG_NFS enabled. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 22 ---------------------- 1 file changed, 22 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 559fb471f6f2..dcf0326bda01 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -15,20 +15,6 @@ #include #include -/* - * The transport code maintains an estimate on the maximum number of out- - * standing RPC requests, using a smoothed version of the congestion - * avoidance implemented in 44BSD. This is basically the Van Jacobson - * congestion algorithm: If a retransmit occurs, the congestion window is - * halved; otherwise, it is incremented by 1/cwnd when - * - * - a reply is received and - * - a full number of requests are outstanding and - * - the congestion window hasn't been updated recently. - * - * Upper procedures may check whether a request would block waiting for - * a free RPC slot by using the RPC_CONGESTED() macro. - */ extern unsigned int xprt_udp_slot_table_entries; extern unsigned int xprt_tcp_slot_table_entries; @@ -36,12 +22,6 @@ extern unsigned int xprt_tcp_slot_table_entries; #define RPC_DEF_SLOT_TABLE (16U) #define RPC_MAX_SLOT_TABLE (128U) -#define RPC_CWNDSHIFT (8U) -#define RPC_CWNDSCALE (1U << RPC_CWNDSHIFT) -#define RPC_INITCWND RPC_CWNDSCALE -#define RPC_MAXCWND(xprt) ((xprt)->max_reqs << RPC_CWNDSHIFT) -#define RPCXPRT_CONGESTED(xprt) ((xprt)->cong >= (xprt)->cwnd) - /* Default timeout values */ #define RPC_MAX_UDP_TIMEOUT (60*HZ) #define RPC_MAX_TCP_TIMEOUT (600*HZ) @@ -213,8 +193,6 @@ struct rpc_xprt { void (*old_data_ready)(struct sock *, int); void (*old_state_change)(struct sock *); void (*old_write_space)(struct sock *); - - wait_queue_head_t cong_wait; }; #define XPRT_LAST_FRAG (1 << 0) -- cgit v1.2.3 From 529b33c6db0120126b1381faa51406dc463acdc9 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:54 -0700 Subject: [PATCH] RPC: allow RPC client's port range to be adjustable Select an RPC client source port between 650 and 1023 instead of between 1 and 800. The old range conflicts with a number of network services. Provide sysctls to allow admins to select a different port range. Note that this doesn't affect user-level RPC library behavior, which still uses 1 to 800. Based on a suggestion by Olaf Kirch . Test-plan: Repeated mount and unmount. Destructive testing. Idle timeouts. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/debug.h | 2 ++ include/linux/sunrpc/xprt.h | 17 ++++++++++++++--- 2 files changed, 16 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/debug.h b/include/linux/sunrpc/debug.h index 42d299747956..1a42d902bc11 100644 --- a/include/linux/sunrpc/debug.h +++ b/include/linux/sunrpc/debug.h @@ -95,6 +95,8 @@ enum { CTL_NLMDEBUG, CTL_SLOTTABLE_UDP, CTL_SLOTTABLE_TCP, + CTL_MIN_RESVPORT, + CTL_MAX_RESVPORT, }; #endif /* _LINUX_SUNRPC_DEBUG_H_ */ diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index dcf0326bda01..9d9266cf8a36 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -51,6 +51,17 @@ extern unsigned int xprt_tcp_slot_table_entries; #define RPC_CALLHDRSIZE 6 #define RPC_REPHDRSIZE 4 +/* + * Parameters for choosing a free port + */ +extern unsigned int xprt_min_resvport; +extern unsigned int xprt_max_resvport; + +#define RPC_MIN_RESVPORT (1U) +#define RPC_MAX_RESVPORT (65535U) +#define RPC_DEF_MIN_RESVPORT (650U) +#define RPC_DEF_MAX_RESVPORT (1023U) + /* * This describes a timeout strategy */ @@ -62,6 +73,9 @@ struct rpc_timeout { unsigned char to_exponential; }; +struct rpc_task; +struct rpc_xprt; + /* * This describes a complete RPC request */ @@ -107,9 +121,6 @@ struct rpc_rqst { #define rq_svec rq_snd_buf.head #define rq_slen rq_snd_buf.len -struct rpc_task; -struct rpc_xprt; - struct rpc_xprt_ops { void (*set_buffer_size)(struct rpc_xprt *xprt); int (*reserve_xprt)(struct rpc_task *task); -- cgit v1.2.3 From 03bf4b707eee06706c9db343dd5c905b7ee47ed2 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:55 -0700 Subject: [PATCH] RPC: parametrize various transport connect timeouts Each transport implementation can now set unique bind, connect, reestablishment, and idle timeout values. These are variables, allowing the values to be modified dynamically. This permits exponential backoff of any of these values, for instance. As an example, we implement exponential backoff for the connection reestablishment timeout. Test-plan: Destructive testing (unplugging the network temporarily). Connectathon with UDP and TCP. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 4 ++++ include/linux/sunrpc/xprt.h | 29 ++++++----------------------- 2 files changed, 10 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 9a6047ff1b25..7bac2785c6e4 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -41,6 +41,10 @@ #define NFS_MAX_FILE_IO_BUFFER_SIZE 32768 #define NFS_DEF_FILE_IO_BUFFER_SIZE 4096 +/* Default timeout values */ +#define NFS_MAX_UDP_TIMEOUT (60*HZ) +#define NFS_MAX_TCP_TIMEOUT (600*HZ) + /* * superblock magic number for NFS */ diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 9d9266cf8a36..2543adf18551 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -22,28 +22,6 @@ extern unsigned int xprt_tcp_slot_table_entries; #define RPC_DEF_SLOT_TABLE (16U) #define RPC_MAX_SLOT_TABLE (128U) -/* Default timeout values */ -#define RPC_MAX_UDP_TIMEOUT (60*HZ) -#define RPC_MAX_TCP_TIMEOUT (600*HZ) - -/* - * Wait duration for an RPC TCP connection to be established. Solaris - * NFS over TCP uses 60 seconds, for example, which is in line with how - * long a server takes to reboot. - */ -#define RPC_CONNECT_TIMEOUT (60*HZ) - -/* - * Delay an arbitrary number of seconds before attempting to reconnect - * after an error. - */ -#define RPC_REESTABLISH_TIMEOUT (15*HZ) - -/* - * RPC transport idle timeout. - */ -#define RPC_IDLE_DISCONNECT_TIMEOUT (5*60*HZ) - /* * RPC call and reply header size as number of 32bit words (verifier * size computed separately) @@ -182,14 +160,19 @@ struct rpc_xprt { /* * Connection of transports */ + unsigned long connect_timeout, + bind_timeout, + reestablish_timeout; struct work_struct connect_worker; unsigned short port; + /* * Disconnection of idle transports */ struct work_struct task_cleanup; struct timer_list timer; - unsigned long last_used; + unsigned long last_used, + idle_timeout; /* * Send stuff -- cgit v1.2.3 From 470056c288334eb0b37be26c9ff8aee37ed1cc7a Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 25 Aug 2005 16:25:56 -0700 Subject: [PATCH] RPC: rationalize set_buffer_size In fact, ->set_buffer_size should be completely functionless for non-UDP. Test-plan: Check socket buffer size on UDP sockets over time. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 2543adf18551..99cad3ead81d 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -100,7 +100,7 @@ struct rpc_rqst { #define rq_slen rq_snd_buf.len struct rpc_xprt_ops { - void (*set_buffer_size)(struct rpc_xprt *xprt); + void (*set_buffer_size)(struct rpc_xprt *xprt, size_t sndsize, size_t rcvsize); int (*reserve_xprt)(struct rpc_task *task); void (*release_xprt)(struct rpc_xprt *xprt, struct rpc_task *task); void (*connect)(struct rpc_task *task); @@ -124,7 +124,7 @@ struct rpc_xprt { unsigned long cong; /* current congestion */ unsigned long cwnd; /* congestion window */ - unsigned int rcvsize, /* transport rcv buffer size */ + size_t rcvsize, /* transport rcv buffer size */ sndsize; /* transport send buffer size */ size_t max_payload; /* largest RPC payload size, -- cgit v1.2.3 From 278c995c8a153bb2a9bc427e931cfb9c8034c9d7 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sun, 24 Jul 2005 23:53:01 +0100 Subject: [PATCH] RPC,NFS: new rpc_pipefs patch Currently rpc_mkdir/rpc_rmdir and rpc_mkpipe/mk_unlink have an API that's a little unfortunate. They take a path relative to the rpc_pipefs root and thus need to perform a full lookup. If you look at debugfs or usbfs they always store the dentry for directories they created and thus can pass in a dentry + single pathname component pair into their equivalents of the above functions. And in fact rpc_pipefs actually stores a dentry for all but one component so this change not only simplifies the core rpc_pipe code but also the callers. Unfortuntately this code path is only used by the NFS4 idmapper and AUTH_GSSAPI for which I don't have a test enviroment. Could someone give it a spin? It's the last bit needed before we can rework the lookup_hash API Signed-off-by: Christoph Hellwig Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 2 +- include/linux/sunrpc/rpc_pipe_fs.h | 9 +++++---- 2 files changed, 6 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index ab151bbb66df..b5b51c196690 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -59,7 +59,7 @@ struct rpc_clnt { int cl_nodelen; /* nodename length */ char cl_nodename[UNX_MAXNODENAME]; - char cl_pathname[30];/* Path in rpc_pipe_fs */ + struct dentry * __cl_parent_dentry; struct dentry * cl_dentry; /* inode */ struct rpc_clnt * cl_parent; /* Points to parent of clones */ struct rpc_rtt cl_rtt_default; diff --git a/include/linux/sunrpc/rpc_pipe_fs.h b/include/linux/sunrpc/rpc_pipe_fs.h index 63929349571f..63878d05c9a9 100644 --- a/include/linux/sunrpc/rpc_pipe_fs.h +++ b/include/linux/sunrpc/rpc_pipe_fs.h @@ -41,10 +41,11 @@ RPC_I(struct inode *inode) extern int rpc_queue_upcall(struct inode *, struct rpc_pipe_msg *); -extern struct dentry *rpc_mkdir(char *, struct rpc_clnt *); -extern int rpc_rmdir(char *); -extern struct dentry *rpc_mkpipe(char *, void *, struct rpc_pipe_ops *, int flags); -extern int rpc_unlink(char *); +extern struct dentry *rpc_mkdir(struct dentry *, char *, struct rpc_clnt *); +extern void rpc_rmdir(struct dentry *); +extern struct dentry *rpc_mkpipe(struct dentry *, char *, void *, + struct rpc_pipe_ops *, int flags); +extern void rpc_unlink(struct dentry *); #endif #endif -- cgit v1.2.3 From f134585a7343d71f9be7f0cf97e2145f21dd10c6 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 23 Sep 2005 11:08:25 -0400 Subject: Revert "[PATCH] RPC,NFS: new rpc_pipefs patch" This reverts 17f4e6febca160a9f9dd4bdece9784577a2f4524 commit. --- include/linux/sunrpc/clnt.h | 2 +- include/linux/sunrpc/rpc_pipe_fs.h | 9 ++++----- 2 files changed, 5 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index b5b51c196690..ab151bbb66df 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -59,7 +59,7 @@ struct rpc_clnt { int cl_nodelen; /* nodename length */ char cl_nodename[UNX_MAXNODENAME]; - struct dentry * __cl_parent_dentry; + char cl_pathname[30];/* Path in rpc_pipe_fs */ struct dentry * cl_dentry; /* inode */ struct rpc_clnt * cl_parent; /* Points to parent of clones */ struct rpc_rtt cl_rtt_default; diff --git a/include/linux/sunrpc/rpc_pipe_fs.h b/include/linux/sunrpc/rpc_pipe_fs.h index 63878d05c9a9..63929349571f 100644 --- a/include/linux/sunrpc/rpc_pipe_fs.h +++ b/include/linux/sunrpc/rpc_pipe_fs.h @@ -41,11 +41,10 @@ RPC_I(struct inode *inode) extern int rpc_queue_upcall(struct inode *, struct rpc_pipe_msg *); -extern struct dentry *rpc_mkdir(struct dentry *, char *, struct rpc_clnt *); -extern void rpc_rmdir(struct dentry *); -extern struct dentry *rpc_mkpipe(struct dentry *, char *, void *, - struct rpc_pipe_ops *, int flags); -extern void rpc_unlink(struct dentry *); +extern struct dentry *rpc_mkdir(char *, struct rpc_clnt *); +extern int rpc_rmdir(char *); +extern struct dentry *rpc_mkpipe(char *, void *, struct rpc_pipe_ops *, int flags); +extern int rpc_unlink(char *); #endif #endif -- cgit v1.2.3 From 5e5ce5be6f0161d2a069a4f8a1154fe639c5c02f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 18 Oct 2005 14:20:11 -0700 Subject: RPC: allow call_encode() to delay transmission of an RPC call. Currently, call_encode will cause the entire RPC call to abort if it returns an error. This is unnecessarily rigid, and gets in the way of attempts to allow the NFSv4 layer to order RPC calls that carry sequence ids. Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 99cad3ead81d..068e1fb0868b 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -211,6 +211,7 @@ int xprt_reserve_xprt(struct rpc_task *task); int xprt_reserve_xprt_cong(struct rpc_task *task); int xprt_prepare_transmit(struct rpc_task *task); void xprt_transmit(struct rpc_task *task); +void xprt_abort_transmit(struct rpc_task *task); int xprt_adjust_timeout(struct rpc_rqst *req); void xprt_release_xprt(struct rpc_xprt *xprt, struct rpc_task *task); void xprt_release_xprt_cong(struct rpc_xprt *xprt, struct rpc_task *task); -- cgit v1.2.3 From cee54fc944422c44e476736c045a9e8053cb0644 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 18 Oct 2005 14:20:12 -0700 Subject: NFSv4: Add functions to order RPC calls NFSv4 file state-changing functions such as OPEN, CLOSE, LOCK,... are all labelled with "sequence identifiers" in order to prevent the server from reordering RPC requests, as this could cause its file state to become out of sync with the client. Currently the NFS client code enforces this ordering locally using semaphores to restrict access to structures until the RPC call is done. This, of course, only works with synchronous RPC calls, since the user process must first grab the semaphore. By dropping semaphores, and instead teaching the RPC engine to hold the RPC calls until they are ready to be sent, we can extend this process to work nicely with asynchronous RPC calls too. This patch adds a new list called "rpc_sequence" that defines the order of the RPC calls to be sent. We add one such list for each state_owner. When an RPC call is ready to be sent, it checks if it is top of the rpc_sequence list. If so, it proceeds. If not, it goes back to sleep, and loops until it hits top of the list. Once the RPC call has completed, it can then bump the sequence id counter, and remove itself from the rpc_sequence list, and then wake up the next sleeper. Note that the state_owner sequence ids and lock_owner sequence ids are all indexed to the same rpc_sequence list, so OPEN, LOCK,... requests are all ordered w.r.t. each other. Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index a2bf6914ff1b..d578912bf9a9 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -96,12 +96,13 @@ struct nfs4_change_info { u64 after; }; +struct nfs_seqid; /* * Arguments to the open call. */ struct nfs_openargs { const struct nfs_fh * fh; - __u32 seqid; + struct nfs_seqid * seqid; int open_flags; __u64 clientid; __u32 id; @@ -136,7 +137,7 @@ struct nfs_openres { struct nfs_open_confirmargs { const struct nfs_fh * fh; nfs4_stateid stateid; - __u32 seqid; + struct nfs_seqid * seqid; }; struct nfs_open_confirmres { @@ -149,7 +150,7 @@ struct nfs_open_confirmres { struct nfs_closeargs { struct nfs_fh * fh; nfs4_stateid stateid; - __u32 seqid; + struct nfs_seqid * seqid; int open_flags; }; @@ -165,15 +166,15 @@ struct nfs_lowner { }; struct nfs_open_to_lock { - __u32 open_seqid; + struct nfs_seqid * open_seqid; nfs4_stateid open_stateid; - __u32 lock_seqid; + struct nfs_seqid * lock_seqid; struct nfs_lowner lock_owner; }; struct nfs_exist_lock { nfs4_stateid stateid; - __u32 seqid; + struct nfs_seqid * seqid; }; struct nfs_lock_opargs { @@ -186,7 +187,7 @@ struct nfs_lock_opargs { }; struct nfs_locku_opargs { - __u32 seqid; + struct nfs_seqid * seqid; nfs4_stateid stateid; }; -- cgit v1.2.3 From 9512135df14f8293b9bc5e8fb22d4279dee5ff66 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 18 Oct 2005 14:20:12 -0700 Subject: NFSv4: Fix a potential CLOSE race Once the state_owner and lock_owner semaphores get removed, it will be possible for other OPEN requests to reopen the same file if they have lower sequence ids than our CLOSE call. This patch ensures that we recheck the file state once nfs_wait_on_sequence() has completed waiting. Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index d578912bf9a9..cac0df950c66 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -149,7 +149,7 @@ struct nfs_open_confirmres { */ struct nfs_closeargs { struct nfs_fh * fh; - nfs4_stateid stateid; + nfs4_stateid * stateid; struct nfs_seqid * seqid; int open_flags; }; -- cgit v1.2.3 From faf5f49c2d9c0af2847837c232a432cc146e203b Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 18 Oct 2005 14:20:15 -0700 Subject: NFSv4: Make NFS clean up byte range locks asynchronously Currently we fail to do so if the process was signalled. Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index cac0df950c66..849f95c5fae4 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -188,7 +188,7 @@ struct nfs_lock_opargs { struct nfs_locku_opargs { struct nfs_seqid * seqid; - nfs4_stateid stateid; + nfs4_stateid * stateid; }; struct nfs_lockargs { -- cgit v1.2.3 From 06735b3454824bd561decbde46111f144e905923 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 18 Oct 2005 14:20:15 -0700 Subject: NFSv4: Fix up handling of open_to_lock sequence ids Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 19 ++++--------------- 1 file changed, 4 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 849f95c5fae4..57efcc27f20b 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -165,25 +165,14 @@ struct nfs_lowner { u32 id; }; -struct nfs_open_to_lock { - struct nfs_seqid * open_seqid; - nfs4_stateid open_stateid; +struct nfs_lock_opargs { struct nfs_seqid * lock_seqid; + nfs4_stateid * lock_stateid; + struct nfs_seqid * open_seqid; + nfs4_stateid * open_stateid; struct nfs_lowner lock_owner; -}; - -struct nfs_exist_lock { - nfs4_stateid stateid; - struct nfs_seqid * seqid; -}; - -struct nfs_lock_opargs { __u32 reclaim; __u32 new_lock_owner; - union { - struct nfs_open_to_lock *open_lock; - struct nfs_exist_lock *exist_lock; - } u; }; struct nfs_locku_opargs { -- cgit v1.2.3 From 834f2a4a1554dc5b2598038b3fe8703defcbe467 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 18 Oct 2005 14:20:16 -0700 Subject: VFS: Allow the filesystem to return a full file pointer on open intent This is needed by NFSv4 for atomicity reasons: our open command is in fact a lookup+open, so we need to be able to propagate open context information from lookup() into the resulting struct file's private_data field. Signed-off-by: Trond Myklebust --- include/linux/namei.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include/linux') diff --git a/include/linux/namei.h b/include/linux/namei.h index 7db67b008cac..1c975d0d9e94 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -8,6 +8,7 @@ struct vfsmount; struct open_intent { int flags; int create_mode; + struct file *file; }; enum { MAX_NESTED_LINKS = 5 }; @@ -65,6 +66,13 @@ extern int FASTCALL(link_path_walk(const char *, struct nameidata *)); extern void path_release(struct nameidata *); extern void path_release_on_umount(struct nameidata *); +extern int __user_path_lookup_open(const char __user *, unsigned lookup_flags, struct nameidata *nd, int open_flags); +extern int path_lookup_open(const char *, unsigned lookup_flags, struct nameidata *, int open_flags); +extern struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry *dentry, + int (*open)(struct inode *, struct file *)); +extern struct file *nameidata_to_filp(struct nameidata *nd, int flags); +extern void release_open_intent(struct nameidata *); + extern struct dentry * lookup_one_len(const char *, struct dentry *, int); extern struct dentry * lookup_hash(struct qstr *, struct dentry *); -- cgit v1.2.3 From 02a913a73b52071e93f4b76db3e86138d19efffd Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 18 Oct 2005 14:20:17 -0700 Subject: NFSv4: Eliminate nfsv4 open race... Make NFSv4 return the fully initialized file pointer with the stateid that it created in the lookup w/intent. Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 57efcc27f20b..60086dac11d5 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -712,7 +712,7 @@ struct nfs_rpc_ops { int (*write) (struct nfs_write_data *); int (*commit) (struct nfs_write_data *); int (*create) (struct inode *, struct dentry *, - struct iattr *, int); + struct iattr *, int, struct nameidata *); int (*remove) (struct inode *, struct qstr *); int (*unlink_setup) (struct rpc_message *, struct dentry *, struct qstr *); -- cgit v1.2.3 From 293f1eb551a77fe5c8956a559a3c0baea95cd9bc Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Thu, 13 Oct 2005 16:54:37 -0400 Subject: SUNRPC: Add support for privacy to generic gss-api code. Add support for privacy to generic gss-api code. This is dead code until we have both a mechanism that supports privacy and code in the client or server that uses it. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/gss_api.h | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/gss_api.h b/include/linux/sunrpc/gss_api.h index 689262f63059..e896752ffbf9 100644 --- a/include/linux/sunrpc/gss_api.h +++ b/include/linux/sunrpc/gss_api.h @@ -48,6 +48,17 @@ u32 gss_verify_mic( struct xdr_buf *message, struct xdr_netobj *mic_token, u32 *qstate); +u32 gss_wrap( + struct gss_ctx *ctx_id, + u32 qop, + int offset, + struct xdr_buf *outbuf, + struct page **inpages); +u32 gss_unwrap( + struct gss_ctx *ctx_id, + u32 *qop, + int offset, + struct xdr_buf *inbuf); u32 gss_delete_sec_context( struct gss_ctx **ctx_id); @@ -93,6 +104,17 @@ struct gss_api_ops { struct xdr_buf *message, struct xdr_netobj *mic_token, u32 *qstate); + u32 (*gss_wrap)( + struct gss_ctx *ctx_id, + u32 qop, + int offset, + struct xdr_buf *outbuf, + struct page **inpages); + u32 (*gss_unwrap)( + struct gss_ctx *ctx_id, + u32 *qop, + int offset, + struct xdr_buf *buf); void (*gss_delete_sec_context)( void *internal_ctx_id); }; -- cgit v1.2.3 From ead5e1c26fdcd969cf40c49cb0589d56879d240d Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Thu, 13 Oct 2005 16:54:43 -0400 Subject: SUNRPC: Provide a callback to allow free pages allocated during xdr encoding For privacy, we need to allocate pages to store the encrypted data (passed in pages can't be used without the risk of corrupting data in the page cache). So we need a way to free that memory after the request has been transmitted. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 068e1fb0868b..3b8b6e823c70 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -73,7 +73,10 @@ struct rpc_rqst { int rq_cong; /* has incremented xprt->cong */ int rq_received; /* receive completed */ u32 rq_seqno; /* gss seq no. used on req. */ - + int rq_enc_pages_num; + struct page **rq_enc_pages; /* scratch pages for use by + gss privacy code */ + void (*rq_release_snd_buf)(struct rpc_rqst *); /* release rq_enc_pages */ struct list_head rq_list; struct xdr_buf rq_private_buf; /* The receive buffer -- cgit v1.2.3 From 24b2605becc10ca63c4c30808fa59a8abbf68727 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Thu, 13 Oct 2005 16:54:53 -0400 Subject: RPCSEC_GSS: cleanup au_rslack calculation Various xdr encode routines use au_rslack to guess where the reply argument will end up, so we can set up the xdr_buf to recieve data into the right place for zero copy. Currently we calculate the au_rslack estimate when we check the verifier. Normally this only depends on the verifier size. In the integrity case we add a few bytes to allow for a length and sequence number. It's a bit simpler to calculate only the verifier size when we check the verifier, and delay the full calculation till we unwrap. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/auth.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/auth.h b/include/linux/sunrpc/auth.h index 04ebc24db348..b68c11a2d6dd 100644 --- a/include/linux/sunrpc/auth.h +++ b/include/linux/sunrpc/auth.h @@ -66,7 +66,12 @@ struct rpc_cred_cache { struct rpc_auth { unsigned int au_cslack; /* call cred size estimate */ - unsigned int au_rslack; /* reply verf size guess */ + /* guess at number of u32's auth adds before + * reply data; normally the verifier size: */ + unsigned int au_rslack; + /* for gss, used to calculate au_rslack: */ + unsigned int au_verfsize; + unsigned int au_flags; /* various flags */ struct rpc_authops * au_ops; /* operations */ rpc_authflavor_t au_flavor; /* pseudoflavor (note may -- cgit v1.2.3 From bfa91516b57483fc9c81d8d90325fd2c3c16ac48 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Thu, 13 Oct 2005 16:55:08 -0400 Subject: RPCSEC_GSS: krb5 pre-privacy cleanup The code this was originally derived from processed wrap and mic tokens using the same functions. This required some contortions, and more would be required with the addition of xdr_buf's, so it's better to separate out the two code paths. In preparation for adding privacy support, remove the last vestiges of the old wrap token code. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/gss_krb5.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/gss_krb5.h b/include/linux/sunrpc/gss_krb5.h index ffe31d2eb9ec..cb35833e2ae3 100644 --- a/include/linux/sunrpc/gss_krb5.h +++ b/include/linux/sunrpc/gss_krb5.h @@ -121,13 +121,12 @@ make_checksum(s32 cksumtype, char *header, int hdrlen, struct xdr_buf *body, u32 krb5_make_token(struct krb5_ctx *context_handle, int qop_req, struct xdr_buf *input_message_buffer, - struct xdr_netobj *output_message_buffer, int toktype); + struct xdr_netobj *output_message_buffer); u32 krb5_read_token(struct krb5_ctx *context_handle, struct xdr_netobj *input_token_buffer, - struct xdr_buf *message_buffer, - int *qop_state, int toktype); + struct xdr_buf *message_buffer, int *qop_state); u32 krb5_encrypt(struct crypto_tfm * key, -- cgit v1.2.3 From 14ae162c24d985593d5b19437d7f3d8fd0062b59 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Thu, 13 Oct 2005 16:55:13 -0400 Subject: RPCSEC_GSS: Add support for privacy to krb5 rpcsec_gss mechanism. Add support for privacy to the krb5 rpcsec_gss mechanism. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/gss_krb5.h | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/gss_krb5.h b/include/linux/sunrpc/gss_krb5.h index cb35833e2ae3..7f93c2d5ebdb 100644 --- a/include/linux/sunrpc/gss_krb5.h +++ b/include/linux/sunrpc/gss_krb5.h @@ -116,7 +116,7 @@ enum seal_alg { s32 make_checksum(s32 cksumtype, char *header, int hdrlen, struct xdr_buf *body, - struct xdr_netobj *cksum); + int body_offset, struct xdr_netobj *cksum); u32 krb5_make_token(struct krb5_ctx *context_handle, int qop_req, @@ -128,6 +128,15 @@ krb5_read_token(struct krb5_ctx *context_handle, struct xdr_netobj *input_token_buffer, struct xdr_buf *message_buffer, int *qop_state); +u32 +gss_wrap_kerberos(struct gss_ctx *ctx_id, u32 qop, int offset, + struct xdr_buf *outbuf, struct page **pages); + +u32 +gss_unwrap_kerberos(struct gss_ctx *ctx_id, u32 *qop, int offset, + struct xdr_buf *buf); + + u32 krb5_encrypt(struct crypto_tfm * key, void *iv, void *in, void *out, int length); @@ -136,6 +145,13 @@ u32 krb5_decrypt(struct crypto_tfm * key, void *iv, void *in, void *out, int length); +int +gss_encrypt_xdr_buf(struct crypto_tfm *tfm, struct xdr_buf *outbuf, int offset, + struct page **pages); + +int +gss_decrypt_xdr_buf(struct crypto_tfm *tfm, struct xdr_buf *inbuf, int offset); + s32 krb5_make_seq_num(struct crypto_tfm * key, int direction, -- cgit v1.2.3 From 00fd6e14255fe7a249315746386d640bc4e9e758 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Thu, 13 Oct 2005 16:55:18 -0400 Subject: RPCSEC_GSS remove all qop parameters Not only are the qop parameters that are passed around throughout the gssapi unused by any currently implemented mechanism, but there appears to be some doubt as to whether they will ever be used. Let's just kill them off for now. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/gss_api.h | 13 ++----------- include/linux/sunrpc/gss_err.h | 10 ---------- include/linux/sunrpc/gss_krb5.h | 8 ++++---- include/linux/sunrpc/gss_spkm3.h | 4 ++-- 4 files changed, 8 insertions(+), 27 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/gss_api.h b/include/linux/sunrpc/gss_api.h index e896752ffbf9..9b8bcf125c18 100644 --- a/include/linux/sunrpc/gss_api.h +++ b/include/linux/sunrpc/gss_api.h @@ -40,23 +40,19 @@ int gss_import_sec_context( struct gss_ctx **ctx_id); u32 gss_get_mic( struct gss_ctx *ctx_id, - u32 qop, struct xdr_buf *message, struct xdr_netobj *mic_token); u32 gss_verify_mic( struct gss_ctx *ctx_id, struct xdr_buf *message, - struct xdr_netobj *mic_token, - u32 *qstate); + struct xdr_netobj *mic_token); u32 gss_wrap( struct gss_ctx *ctx_id, - u32 qop, int offset, struct xdr_buf *outbuf, struct page **inpages); u32 gss_unwrap( struct gss_ctx *ctx_id, - u32 *qop, int offset, struct xdr_buf *inbuf); u32 gss_delete_sec_context( @@ -67,7 +63,6 @@ char *gss_service_to_auth_domain_name(struct gss_api_mech *, u32 service); struct pf_desc { u32 pseudoflavor; - u32 qop; u32 service; char *name; char *auth_domain_name; @@ -96,23 +91,19 @@ struct gss_api_ops { struct gss_ctx *ctx_id); u32 (*gss_get_mic)( struct gss_ctx *ctx_id, - u32 qop, struct xdr_buf *message, struct xdr_netobj *mic_token); u32 (*gss_verify_mic)( struct gss_ctx *ctx_id, struct xdr_buf *message, - struct xdr_netobj *mic_token, - u32 *qstate); + struct xdr_netobj *mic_token); u32 (*gss_wrap)( struct gss_ctx *ctx_id, - u32 qop, int offset, struct xdr_buf *outbuf, struct page **inpages); u32 (*gss_unwrap)( struct gss_ctx *ctx_id, - u32 *qop, int offset, struct xdr_buf *buf); void (*gss_delete_sec_context)( diff --git a/include/linux/sunrpc/gss_err.h b/include/linux/sunrpc/gss_err.h index 92608a2e574c..a6807867bd21 100644 --- a/include/linux/sunrpc/gss_err.h +++ b/include/linux/sunrpc/gss_err.h @@ -65,16 +65,6 @@ typedef unsigned int OM_uint32; #define GSS_C_MECH_CODE 2 -/* - * Define the default Quality of Protection for per-message services. Note - * that an implementation that offers multiple levels of QOP may either reserve - * a value (for example zero, as assumed here) to mean "default protection", or - * alternatively may simply equate GSS_C_QOP_DEFAULT to a specific explicit - * QOP value. However a value of 0 should always be interpreted by a GSSAPI - * implementation as a request for the default protection level. - */ -#define GSS_C_QOP_DEFAULT 0 - /* * Expiration time of 2^32-1 seconds means infinite lifetime for a * credential or security context diff --git a/include/linux/sunrpc/gss_krb5.h b/include/linux/sunrpc/gss_krb5.h index 7f93c2d5ebdb..a7bda4edb853 100644 --- a/include/linux/sunrpc/gss_krb5.h +++ b/include/linux/sunrpc/gss_krb5.h @@ -119,21 +119,21 @@ make_checksum(s32 cksumtype, char *header, int hdrlen, struct xdr_buf *body, int body_offset, struct xdr_netobj *cksum); u32 -krb5_make_token(struct krb5_ctx *context_handle, int qop_req, +krb5_make_token(struct krb5_ctx *context_handle, struct xdr_buf *input_message_buffer, struct xdr_netobj *output_message_buffer); u32 krb5_read_token(struct krb5_ctx *context_handle, struct xdr_netobj *input_token_buffer, - struct xdr_buf *message_buffer, int *qop_state); + struct xdr_buf *message_buffer); u32 -gss_wrap_kerberos(struct gss_ctx *ctx_id, u32 qop, int offset, +gss_wrap_kerberos(struct gss_ctx *ctx_id, int offset, struct xdr_buf *outbuf, struct page **pages); u32 -gss_unwrap_kerberos(struct gss_ctx *ctx_id, u32 *qop, int offset, +gss_unwrap_kerberos(struct gss_ctx *ctx_id, int offset, struct xdr_buf *buf); diff --git a/include/linux/sunrpc/gss_spkm3.h b/include/linux/sunrpc/gss_spkm3.h index b5c9968c3c17..0beb2cf00a84 100644 --- a/include/linux/sunrpc/gss_spkm3.h +++ b/include/linux/sunrpc/gss_spkm3.h @@ -41,9 +41,9 @@ struct spkm3_ctx { #define SPKM_WRAP_TOK 5 #define SPKM_DEL_TOK 6 -u32 spkm3_make_token(struct spkm3_ctx *ctx, int qop_req, struct xdr_buf * text, struct xdr_netobj * token, int toktype); +u32 spkm3_make_token(struct spkm3_ctx *ctx, struct xdr_buf * text, struct xdr_netobj * token, int toktype); -u32 spkm3_read_token(struct spkm3_ctx *ctx, struct xdr_netobj *read_token, struct xdr_buf *message_buffer, int *qop_state, int toktype); +u32 spkm3_read_token(struct spkm3_ctx *ctx, struct xdr_netobj *read_token, struct xdr_buf *message_buffer, int toktype); #define CKSUMTYPE_RSA_MD5 0x0007 -- cgit v1.2.3 From a0857d03b21fa54653c9d2fe7a315381176015b4 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Thu, 13 Oct 2005 16:55:23 -0400 Subject: RPCSEC_GSS: krb5 cleanup Remove some senseless wrappers. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/gss_krb5.h | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/gss_krb5.h b/include/linux/sunrpc/gss_krb5.h index a7bda4edb853..2c3601d31045 100644 --- a/include/linux/sunrpc/gss_krb5.h +++ b/include/linux/sunrpc/gss_krb5.h @@ -118,15 +118,11 @@ s32 make_checksum(s32 cksumtype, char *header, int hdrlen, struct xdr_buf *body, int body_offset, struct xdr_netobj *cksum); -u32 -krb5_make_token(struct krb5_ctx *context_handle, - struct xdr_buf *input_message_buffer, - struct xdr_netobj *output_message_buffer); +u32 gss_get_mic_kerberos(struct gss_ctx *, struct xdr_buf *, + struct xdr_netobj *); -u32 -krb5_read_token(struct krb5_ctx *context_handle, - struct xdr_netobj *input_token_buffer, - struct xdr_buf *message_buffer); +u32 gss_verify_mic_kerberos(struct gss_ctx *, struct xdr_buf *, + struct xdr_netobj *); u32 gss_wrap_kerberos(struct gss_ctx *ctx_id, int offset, -- cgit v1.2.3 From 0e574af1be5f569a5d7f2800333b0bfb358a5e34 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 27 Oct 2005 22:12:38 -0400 Subject: NFS: Cleanup initialisation of struct nfs_fattr Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 7bac2785c6e4..8120fd68dee5 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -316,6 +316,11 @@ extern void nfs_file_clear_open_context(struct file *filp); /* linux/net/ipv4/ipconfig.c: trims ip addr off front of name, too. */ extern u32 root_nfs_parse_addr(char *name); /*__init*/ +static inline void nfs_fattr_init(struct nfs_fattr *fattr) +{ + fattr->valid = 0; +} + /* * linux/fs/nfs/file.c */ -- cgit v1.2.3 From 913a70fc170530f7e1ff0693595155457cc6d0ca Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 27 Oct 2005 22:12:38 -0400 Subject: NFS: Convert cache_change_attribute into a jiffy-based value Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 8120fd68dee5..abf890f5fbfb 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -280,7 +280,7 @@ static inline long nfs_save_change_attribute(struct inode *inode) static inline int nfs_verify_change_attribute(struct inode *inode, unsigned long chattr) { return !nfs_caches_unstable(inode) - && chattr == NFS_I(inode)->cache_change_attribute; + && time_after_eq(chattr, NFS_I(inode)->cache_change_attribute); } /* -- cgit v1.2.3 From 33801147a8fda6b04d7e9afe1d42f1c01d3d6837 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 27 Oct 2005 22:12:39 -0400 Subject: NFS: Optimise inode attribute cache updates Allow nfs_refresh_inode() also to update attributes on the inode if the RPC call was sent after the last call to nfs_update_inode(). Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 2 ++ include/linux/nfs_xdr.h | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index abf890f5fbfb..faeaad666ca8 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -141,6 +141,7 @@ struct nfs_inode { unsigned long attrtimeo_timestamp; __u64 change_attr; /* v4 only */ + unsigned long last_updated; /* "Generation counter" for the attribute cache. This is * bumped whenever we update the metadata on the * server. @@ -319,6 +320,7 @@ extern u32 root_nfs_parse_addr(char *name); /*__init*/ static inline void nfs_fattr_init(struct nfs_fattr *fattr) { fattr->valid = 0; + fattr->time_start = jiffies; } /* diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 60086dac11d5..aeaee7e7c51d 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -41,7 +41,7 @@ struct nfs_fattr { __u32 bitmap[2]; /* NFSv4 returned attribute bitmap */ __u64 change_attr; /* NFSv4 change attribute */ __u64 pre_change_attr;/* pre-op NFSv4 change attribute */ - unsigned long timestamp; + unsigned long time_start; }; #define NFS_ATTR_WCC 0x0001 /* pre-op WCC data */ -- cgit v1.2.3 From decf491f3076190262d4c649bed877650623903a Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 27 Oct 2005 22:12:39 -0400 Subject: NFS: Don't let nfs_end_data_update() clobber attribute update information Since we almost always call nfs_end_data_update() after we called nfs_refresh_inode(), we now end up marking the inode metadata as needing revalidation immediately after having updated it. This patch rearranges things so that we mark the inode as needing revalidation _before_ we call nfs_refresh_inode() on those operations that need it. Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index faeaad666ca8..325fe7ae49bb 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -241,13 +241,17 @@ static inline int nfs_caches_unstable(struct inode *inode) return atomic_read(&NFS_I(inode)->data_updates) != 0; } +static inline void nfs_mark_for_revalidate(struct inode *inode) +{ + spin_lock(&inode->i_lock); + NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS; + spin_unlock(&inode->i_lock); +} + static inline void NFS_CACHEINV(struct inode *inode) { - if (!nfs_caches_unstable(inode)) { - spin_lock(&inode->i_lock); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS; - spin_unlock(&inode->i_lock); - } + if (!nfs_caches_unstable(inode)) + nfs_mark_for_revalidate(inode); } static inline int nfs_server_capable(struct inode *inode, int cap) @@ -291,6 +295,7 @@ extern void nfs_zap_caches(struct inode *); extern struct inode *nfs_fhget(struct super_block *, struct nfs_fh *, struct nfs_fattr *); extern int nfs_refresh_inode(struct inode *, struct nfs_fattr *); +extern int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr); extern int nfs_getattr(struct vfsmount *, struct dentry *, struct kstat *); extern int nfs_permission(struct inode *, int, struct nameidata *); extern int nfs_access_get_cached(struct inode *, struct rpc_cred *, struct nfs_access_entry *); -- cgit v1.2.3 From 56ae19f38f10aad4f27f7e12138a29b295dff07a Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 27 Oct 2005 22:12:40 -0400 Subject: NFSv4: Add directory post-op attributes to the CREATE operations. Since the directory attributes change every time we CREATE a file, we might as well pick up the new directory attributes in the same compound. Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index aeaee7e7c51d..6485b8b41b83 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -124,6 +124,7 @@ struct nfs_openres { struct nfs4_change_info cinfo; __u32 rflags; struct nfs_fattr * f_attr; + struct nfs_fattr * dir_attr; const struct nfs_server *server; int delegation_type; nfs4_stateid delegation; @@ -540,6 +541,7 @@ struct nfs4_create_res { struct nfs_fh * fh; struct nfs_fattr * fattr; struct nfs4_change_info dir_cinfo; + struct nfs_fattr * dir_fattr; }; struct nfs4_fsinfo_arg { -- cgit v1.2.3 From 516a6af641bb50c608329a5bd751acd0d65cc4ab Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 27 Oct 2005 22:12:41 -0400 Subject: NFS: Add optional post-op getattr instruction to the NFSv4 file close. "Optional" means that the close call will not fail if the getattr at the end of the compound fails. If it does succeed, try to refresh inode attributes. Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 6485b8b41b83..4f03dc21cf4a 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -153,10 +153,13 @@ struct nfs_closeargs { nfs4_stateid * stateid; struct nfs_seqid * seqid; int open_flags; + const u32 * bitmask; }; struct nfs_closeres { nfs4_stateid stateid; + struct nfs_fattr * fattr; + const struct nfs_server *server; }; /* * * Arguments to the lock,lockt, and locku call. -- cgit v1.2.3 From 91ba2eeec5e8e86e054937eb3bf5aec5b22b1830 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 27 Oct 2005 22:12:42 -0400 Subject: NFSv4: Add post-op attributes to nfs4_proc_link() Optimise attribute revalidation when hardlinking. Add post-op attributes for the directory and the original inode. Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 4f03dc21cf4a..89238b799cfd 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -566,8 +566,17 @@ struct nfs4_link_arg { const struct nfs_fh * fh; const struct nfs_fh * dir_fh; const struct qstr * name; + const u32 * bitmask; +}; + +struct nfs4_link_res { + const struct nfs_server * server; + struct nfs_fattr * fattr; + struct nfs4_change_info cinfo; + struct nfs_fattr * dir_attr; }; + struct nfs4_lookup_arg { const struct nfs_fh * dir_fh; const struct qstr * name; -- cgit v1.2.3 From 6caf2c8276d371679a798058e8fdf49f5ff831a3 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 27 Oct 2005 22:12:43 -0400 Subject: NFSv4: Add post-op attributes to nfs4_proc_rename() Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 89238b799cfd..6f0804280824 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -630,11 +630,15 @@ struct nfs4_rename_arg { const struct nfs_fh * new_dir; const struct qstr * old_name; const struct qstr * new_name; + const u32 * bitmask; }; struct nfs4_rename_res { + const struct nfs_server * server; struct nfs4_change_info old_cinfo; + struct nfs_fattr * old_fattr; struct nfs4_change_info new_cinfo; + struct nfs_fattr * new_fattr; }; struct nfs4_setclientid { -- cgit v1.2.3 From 16e429596dec4d28e16812b3a9be27f18412c567 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 27 Oct 2005 22:12:44 -0400 Subject: NFSv4: Add post-op attributes to nfs4_proc_remove() Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 6f0804280824..deeba7e2c518 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -623,6 +623,13 @@ struct nfs4_readlink { struct nfs4_remove_arg { const struct nfs_fh * fh; const struct qstr * name; + const u32 * bitmask; +}; + +struct nfs4_remove_res { + const struct nfs_server * server; + struct nfs4_change_info cinfo; + struct nfs_fattr * dir_attr; }; struct nfs4_rename_arg { -- cgit v1.2.3 From 4f9838c7ecd14f31f701f64fa65ded132fc0db8a Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 27 Oct 2005 22:12:44 -0400 Subject: NFSv4: Add post-op attributes to NFSv4 write and commit callbacks. Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index deeba7e2c518..40718669b9c8 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -256,6 +256,7 @@ struct nfs_writeargs { enum nfs3_stable_how stable; unsigned int pgbase; struct page ** pages; + const u32 * bitmask; }; struct nfs_writeverf { @@ -267,6 +268,7 @@ struct nfs_writeres { struct nfs_fattr * fattr; struct nfs_writeverf * verf; __u32 count; + const struct nfs_server *server; }; /* -- cgit v1.2.3 From af4ca457eaf2d6682059c18463eb106e2ce58198 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 21 Oct 2005 02:55:38 -0400 Subject: [PATCH] gfp_t: infrastructure Beginning of gfp_t annotations: - -Wbitwise added to CHECKFLAGS - old __bitwise renamed to __bitwise__ - __bitwise defined to either __bitwise__ or nothing, depending on __CHECK_ENDIAN__ being defined - gfp_t switched from __nocast to __bitwise__ - force cast to gfp_t added to __GFP_... constants - new helper - gfp_zone(); extracts zone bits out of gfp_t value and casts the result to int Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/gfp.h | 39 ++++++++++++++++++++------------------- include/linux/types.h | 9 +++++++-- 2 files changed, 27 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gfp.h b/include/linux/gfp.h index 3010e172394d..c3779432a723 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -12,8 +12,8 @@ struct vm_area_struct; * GFP bitmasks.. */ /* Zone modifiers in GFP_ZONEMASK (see linux/mmzone.h - low two bits) */ -#define __GFP_DMA 0x01u -#define __GFP_HIGHMEM 0x02u +#define __GFP_DMA ((__force gfp_t)0x01u) +#define __GFP_HIGHMEM ((__force gfp_t)0x02u) /* * Action modifiers - doesn't change the zoning @@ -26,24 +26,24 @@ struct vm_area_struct; * * __GFP_NORETRY: The VM implementation must not retry indefinitely. */ -#define __GFP_WAIT 0x10u /* Can wait and reschedule? */ -#define __GFP_HIGH 0x20u /* Should access emergency pools? */ -#define __GFP_IO 0x40u /* Can start physical IO? */ -#define __GFP_FS 0x80u /* Can call down to low-level FS? */ -#define __GFP_COLD 0x100u /* Cache-cold page required */ -#define __GFP_NOWARN 0x200u /* Suppress page allocation failure warning */ -#define __GFP_REPEAT 0x400u /* Retry the allocation. Might fail */ -#define __GFP_NOFAIL 0x800u /* Retry for ever. Cannot fail */ -#define __GFP_NORETRY 0x1000u /* Do not retry. Might fail */ -#define __GFP_NO_GROW 0x2000u /* Slab internal usage */ -#define __GFP_COMP 0x4000u /* Add compound page metadata */ -#define __GFP_ZERO 0x8000u /* Return zeroed page on success */ -#define __GFP_NOMEMALLOC 0x10000u /* Don't use emergency reserves */ -#define __GFP_NORECLAIM 0x20000u /* No realy zone reclaim during allocation */ -#define __GFP_HARDWALL 0x40000u /* Enforce hardwall cpuset memory allocs */ +#define __GFP_WAIT ((__force gfp_t)0x10u) /* Can wait and reschedule? */ +#define __GFP_HIGH ((__force gfp_t)0x20u) /* Should access emergency pools? */ +#define __GFP_IO ((__force gfp_t)0x40u) /* Can start physical IO? */ +#define __GFP_FS ((__force gfp_t)0x80u) /* Can call down to low-level FS? */ +#define __GFP_COLD ((__force gfp_t)0x100u) /* Cache-cold page required */ +#define __GFP_NOWARN ((__force gfp_t)0x200u) /* Suppress page allocation failure warning */ +#define __GFP_REPEAT ((__force gfp_t)0x400u) /* Retry the allocation. Might fail */ +#define __GFP_NOFAIL ((__force gfp_t)0x800u) /* Retry for ever. Cannot fail */ +#define __GFP_NORETRY ((__force gfp_t)0x1000u)/* Do not retry. Might fail */ +#define __GFP_NO_GROW ((__force gfp_t)0x2000u)/* Slab internal usage */ +#define __GFP_COMP ((__force gfp_t)0x4000u)/* Add compound page metadata */ +#define __GFP_ZERO ((__force gfp_t)0x8000u)/* Return zeroed page on success */ +#define __GFP_NOMEMALLOC ((__force gfp_t)0x10000u) /* Don't use emergency reserves */ +#define __GFP_NORECLAIM ((__force gfp_t)0x20000u) /* No realy zone reclaim during allocation */ +#define __GFP_HARDWALL ((__force gfp_t)0x40000u) /* Enforce hardwall cpuset memory allocs */ #define __GFP_BITS_SHIFT 20 /* Room for 20 __GFP_FOO bits */ -#define __GFP_BITS_MASK ((1 << __GFP_BITS_SHIFT) - 1) +#define __GFP_BITS_MASK ((__force gfp_t)((1 << __GFP_BITS_SHIFT) - 1)) /* if you forget to add the bitmask here kernel will crash, period */ #define GFP_LEVEL_MASK (__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_FS| \ @@ -64,6 +64,7 @@ struct vm_area_struct; #define GFP_DMA __GFP_DMA +#define gfp_zone(mask) ((__force int)((mask) & (__force gfp_t)GFP_ZONEMASK)) /* * There is only one page-allocator function, and two main namespaces to @@ -94,7 +95,7 @@ static inline struct page *alloc_pages_node(int nid, gfp_t gfp_mask, return NULL; return __alloc_pages(gfp_mask, order, - NODE_DATA(nid)->node_zonelists + (gfp_mask & GFP_ZONEMASK)); + NODE_DATA(nid)->node_zonelists + gfp_zone(gfp_mask)); } #ifdef CONFIG_NUMA diff --git a/include/linux/types.h b/include/linux/types.h index 0aee34f9da9f..21b9ce803644 100644 --- a/include/linux/types.h +++ b/include/linux/types.h @@ -151,7 +151,12 @@ typedef unsigned long sector_t; */ #ifdef __CHECKER__ -#define __bitwise __attribute__((bitwise)) +#define __bitwise__ __attribute__((bitwise)) +#else +#define __bitwise__ +#endif +#ifdef __CHECK_ENDIAN__ +#define __bitwise __bitwise__ #else #define __bitwise #endif @@ -166,7 +171,7 @@ typedef __u64 __bitwise __be64; #endif #ifdef __KERNEL__ -typedef unsigned __nocast gfp_t; +typedef unsigned __bitwise__ gfp_t; #endif struct ustat { -- cgit v1.2.3 From 6daa0e28627abf362138244a620a821a9027d816 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 21 Oct 2005 03:18:50 -0400 Subject: [PATCH] gfp_t: mm/* (easy parts) Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/mm.h | 2 +- include/linux/pagemap.h | 6 +++--- include/linux/slab.h | 2 +- include/linux/swap.h | 4 ++-- 4 files changed, 7 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 097b3a3c693d..e1649578fb0c 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -747,7 +747,7 @@ extern unsigned long do_mremap(unsigned long addr, * The callback will be passed nr_to_scan == 0 when the VM is querying the * cache size, so a fastpath for that case is appropriate. */ -typedef int (*shrinker_t)(int nr_to_scan, unsigned int gfp_mask); +typedef int (*shrinker_t)(int nr_to_scan, gfp_t gfp_mask); /* * Add an aging callback. The int is the number of 'seeks' it takes diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h index acbf31c154f8..efbae53fb078 100644 --- a/include/linux/pagemap.h +++ b/include/linux/pagemap.h @@ -69,7 +69,7 @@ extern struct page * find_lock_page(struct address_space *mapping, extern struct page * find_trylock_page(struct address_space *mapping, unsigned long index); extern struct page * find_or_create_page(struct address_space *mapping, - unsigned long index, unsigned int gfp_mask); + unsigned long index, gfp_t gfp_mask); unsigned find_get_pages(struct address_space *mapping, pgoff_t start, unsigned int nr_pages, struct page **pages); unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index, @@ -92,9 +92,9 @@ extern int read_cache_pages(struct address_space *mapping, struct list_head *pages, filler_t *filler, void *data); int add_to_page_cache(struct page *page, struct address_space *mapping, - unsigned long index, int gfp_mask); + unsigned long index, gfp_t gfp_mask); int add_to_page_cache_lru(struct page *page, struct address_space *mapping, - unsigned long index, int gfp_mask); + unsigned long index, gfp_t gfp_mask); extern void remove_from_page_cache(struct page *page); extern void __remove_from_page_cache(struct page *page); diff --git a/include/linux/slab.h b/include/linux/slab.h index 5fc04a16ecb0..09b9aa60063d 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -121,7 +121,7 @@ extern unsigned int ksize(const void *); extern void *kmem_cache_alloc_node(kmem_cache_t *, gfp_t flags, int node); extern void *kmalloc_node(size_t size, gfp_t flags, int node); #else -static inline void *kmem_cache_alloc_node(kmem_cache_t *cachep, int flags, int node) +static inline void *kmem_cache_alloc_node(kmem_cache_t *cachep, gfp_t flags, int node) { return kmem_cache_alloc(cachep, flags); } diff --git a/include/linux/swap.h b/include/linux/swap.h index a7bf1a3b1496..20c975642cab 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -171,8 +171,8 @@ extern int rotate_reclaimable_page(struct page *page); extern void swap_setup(void); /* linux/mm/vmscan.c */ -extern int try_to_free_pages(struct zone **, unsigned int); -extern int zone_reclaim(struct zone *, unsigned int, unsigned int); +extern int try_to_free_pages(struct zone **, gfp_t); +extern int zone_reclaim(struct zone *, gfp_t, unsigned int); extern int shrink_all_memory(int); extern int vm_swappiness; -- cgit v1.2.3 From fd4f2df24bc23e6b8fc069765b425c7dacf52347 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 21 Oct 2005 03:18:50 -0400 Subject: [PATCH] gfp_t: lib/* Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/idr.h | 2 +- include/linux/kobject.h | 2 +- include/linux/radix-tree.h | 2 +- include/linux/textsearch.h | 4 ++-- 4 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/idr.h b/include/linux/idr.h index 3d5de45f961b..7fb3ff9c7b0e 100644 --- a/include/linux/idr.h +++ b/include/linux/idr.h @@ -71,7 +71,7 @@ struct idr { */ void *idr_find(struct idr *idp, int id); -int idr_pre_get(struct idr *idp, unsigned gfp_mask); +int idr_pre_get(struct idr *idp, gfp_t gfp_mask); int idr_get_new(struct idr *idp, void *ptr, int *id); int idr_get_new_above(struct idr *idp, void *ptr, int starting_id, int *id); void idr_remove(struct idr *idp, int id); diff --git a/include/linux/kobject.h b/include/linux/kobject.h index 3b22304f12fd..7f7403aa4a41 100644 --- a/include/linux/kobject.h +++ b/include/linux/kobject.h @@ -65,7 +65,7 @@ extern void kobject_unregister(struct kobject *); extern struct kobject * kobject_get(struct kobject *); extern void kobject_put(struct kobject *); -extern char * kobject_get_path(struct kobject *, int); +extern char * kobject_get_path(struct kobject *, gfp_t); struct kobj_type { void (*release)(struct kobject *); diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h index 045d4761febc..9f0f9281f42a 100644 --- a/include/linux/radix-tree.h +++ b/include/linux/radix-tree.h @@ -24,7 +24,7 @@ struct radix_tree_root { unsigned int height; - unsigned int gfp_mask; + gfp_t gfp_mask; struct radix_tree_node *rnode; }; diff --git a/include/linux/textsearch.h b/include/linux/textsearch.h index 515046d1b2f4..fc5bb4e91a58 100644 --- a/include/linux/textsearch.h +++ b/include/linux/textsearch.h @@ -40,7 +40,7 @@ struct ts_state struct ts_ops { const char *name; - struct ts_config * (*init)(const void *, unsigned int, int); + struct ts_config * (*init)(const void *, unsigned int, gfp_t); unsigned int (*find)(struct ts_config *, struct ts_state *); void (*destroy)(struct ts_config *); @@ -148,7 +148,7 @@ static inline unsigned int textsearch_get_pattern_len(struct ts_config *conf) extern int textsearch_register(struct ts_ops *); extern int textsearch_unregister(struct ts_ops *); extern struct ts_config *textsearch_prepare(const char *, const void *, - unsigned int, int, int); + unsigned int, gfp_t, int); extern void textsearch_destroy(struct ts_config *conf); extern unsigned int textsearch_find_continuous(struct ts_config *, struct ts_state *, -- cgit v1.2.3 From 7d877f3bda870ab5f001bd92528654471d5966b3 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 21 Oct 2005 03:20:43 -0400 Subject: [PATCH] gfp_t: net/* Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/security.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/security.h b/include/linux/security.h index 627382e74057..dac956ed98f0 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -1210,7 +1210,7 @@ struct security_operations { int (*socket_shutdown) (struct socket * sock, int how); int (*socket_sock_rcv_skb) (struct sock * sk, struct sk_buff * skb); int (*socket_getpeersec) (struct socket *sock, char __user *optval, int __user *optlen, unsigned len); - int (*sk_alloc_security) (struct sock *sk, int family, int priority); + int (*sk_alloc_security) (struct sock *sk, int family, gfp_t priority); void (*sk_free_security) (struct sock *sk); #endif /* CONFIG_SECURITY_NETWORK */ }; -- cgit v1.2.3 From 27496a8c67bef4d789d8e3c8317ca35813a507ae Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 21 Oct 2005 03:20:48 -0400 Subject: [PATCH] gfp_t: fs/* - ->releasepage() annotated (s/int/gfp_t), instances updated - missing gfp_t in fs/* added - fixed misannotation from the original sweep caught by bitwise checks: XFS used __nocast both for gfp_t and for flags used by XFS allocator. The latter left with unsigned int __nocast; we might want to add a different type for those but for now let's leave them alone. That, BTW, is a case when __nocast use had been actively confusing - it had been used in the same code for two different and similar types, with no way to catch misuses. Switch of gfp_t to bitwise had caught that immediately... One tricky bit is left alone to be dealt with later - mapping->flags is a mix of gfp_t and error indications. Left alone for now. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/bio.h | 2 +- include/linux/buffer_head.h | 2 +- include/linux/fs.h | 2 +- include/linux/jbd.h | 4 ++-- include/linux/mbcache.h | 2 +- include/linux/reiserfs_fs.h | 2 +- 6 files changed, 7 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bio.h b/include/linux/bio.h index 3344b4e8e43a..685fd3720df5 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -301,7 +301,7 @@ extern struct bio *bio_map_user_iov(struct request_queue *, struct sg_iovec *, int, int); extern void bio_unmap_user(struct bio *); extern struct bio *bio_map_kern(struct request_queue *, void *, unsigned int, - unsigned int); + gfp_t); extern void bio_set_pages_dirty(struct bio *bio); extern void bio_check_pages_dirty(struct bio *bio); extern struct bio *bio_copy_user(struct request_queue *, unsigned long, unsigned int, int); diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index 6a1d154c0825..88af42f5e04a 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -188,7 +188,7 @@ extern int buffer_heads_over_limit; * Generic address_space_operations implementations for buffer_head-backed * address_spaces. */ -int try_to_release_page(struct page * page, int gfp_mask); +int try_to_release_page(struct page * page, gfp_t gfp_mask); int block_invalidatepage(struct page *page, unsigned long offset); int block_write_full_page(struct page *page, get_block_t *get_block, struct writeback_control *wbc); diff --git a/include/linux/fs.h b/include/linux/fs.h index e0b77c5af9a0..f83d997c5582 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -320,7 +320,7 @@ struct address_space_operations { /* Unfortunately this kludge is needed for FIBMAP. Don't use it */ sector_t (*bmap)(struct address_space *, sector_t); int (*invalidatepage) (struct page *, unsigned long); - int (*releasepage) (struct page *, int); + int (*releasepage) (struct page *, gfp_t); ssize_t (*direct_IO)(int, struct kiocb *, const struct iovec *iov, loff_t offset, unsigned long nr_segs); struct page* (*get_xip_page)(struct address_space *, sector_t, diff --git a/include/linux/jbd.h b/include/linux/jbd.h index ff853b3173c6..be197eb90077 100644 --- a/include/linux/jbd.h +++ b/include/linux/jbd.h @@ -69,7 +69,7 @@ extern int journal_enable_debug; #define jbd_debug(f, a...) /**/ #endif -extern void * __jbd_kmalloc (const char *where, size_t size, int flags, int retry); +extern void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry); #define jbd_kmalloc(size, flags) \ __jbd_kmalloc(__FUNCTION__, (size), (flags), journal_oom_retry) #define jbd_rep_kmalloc(size, flags) \ @@ -890,7 +890,7 @@ extern int journal_forget (handle_t *, struct buffer_head *); extern void journal_sync_buffer (struct buffer_head *); extern int journal_invalidatepage(journal_t *, struct page *, unsigned long); -extern int journal_try_to_free_buffers(journal_t *, struct page *, int); +extern int journal_try_to_free_buffers(journal_t *, struct page *, gfp_t); extern int journal_stop(handle_t *); extern int journal_flush (journal_t *); extern void journal_lock_updates (journal_t *); diff --git a/include/linux/mbcache.h b/include/linux/mbcache.h index 9263d2db2d67..99e044b4efc6 100644 --- a/include/linux/mbcache.h +++ b/include/linux/mbcache.h @@ -22,7 +22,7 @@ struct mb_cache_entry { }; struct mb_cache_op { - int (*free)(struct mb_cache_entry *, int); + int (*free)(struct mb_cache_entry *, gfp_t); }; /* Functions on caches */ diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index af00b10294cd..001ab82df051 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -1972,7 +1972,7 @@ extern struct address_space_operations reiserfs_address_space_operations; /* fix_nodes.c */ #ifdef CONFIG_REISERFS_CHECK -void *reiserfs_kmalloc(size_t size, int flags, struct super_block *s); +void *reiserfs_kmalloc(size_t size, gfp_t flags, struct super_block *s); void reiserfs_kfree(const void *vp, size_t size, struct super_block *s); #else static inline void *reiserfs_kmalloc(size_t size, int flags, -- cgit v1.2.3 From 8267e268e0914ac9371d07f711fcf20cc572993c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 21 Oct 2005 03:20:53 -0400 Subject: [PATCH] gfp_t: block layer core Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/blkdev.h | 14 +++++++------- include/linux/elevator.h | 4 ++-- 2 files changed, 9 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index efdc9b5bc05c..1afbdb2d752c 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -96,8 +96,8 @@ struct io_context { void put_io_context(struct io_context *ioc); void exit_io_context(void); -struct io_context *current_io_context(int gfp_flags); -struct io_context *get_io_context(int gfp_flags); +struct io_context *current_io_context(gfp_t gfp_flags); +struct io_context *get_io_context(gfp_t gfp_flags); void copy_io_context(struct io_context **pdst, struct io_context **psrc); void swap_io_context(struct io_context **ioc1, struct io_context **ioc2); @@ -354,7 +354,7 @@ struct request_queue * queue needs bounce pages for pages above this limit */ unsigned long bounce_pfn; - unsigned int bounce_gfp; + gfp_t bounce_gfp; /* * various queue flags, see QUEUE_* below @@ -550,7 +550,7 @@ extern void generic_make_request(struct bio *bio); extern void blk_put_request(struct request *); extern void blk_end_sync_rq(struct request *rq); extern void blk_attempt_remerge(request_queue_t *, struct request *); -extern struct request *blk_get_request(request_queue_t *, int, int); +extern struct request *blk_get_request(request_queue_t *, int, gfp_t); extern void blk_insert_request(request_queue_t *, struct request *, int, void *); extern void blk_requeue_request(request_queue_t *, struct request *); extern void blk_plug_device(request_queue_t *); @@ -565,7 +565,7 @@ extern void blk_run_queue(request_queue_t *); extern void blk_queue_activity_fn(request_queue_t *, activity_fn *, void *); extern int blk_rq_map_user(request_queue_t *, struct request *, void __user *, unsigned int); extern int blk_rq_unmap_user(struct bio *, unsigned int); -extern int blk_rq_map_kern(request_queue_t *, struct request *, void *, unsigned int, unsigned int); +extern int blk_rq_map_kern(request_queue_t *, struct request *, void *, unsigned int, gfp_t); extern int blk_rq_map_user_iov(request_queue_t *, struct request *, struct sg_iovec *, int); extern int blk_execute_rq(request_queue_t *, struct gendisk *, struct request *, int); @@ -654,8 +654,8 @@ extern void blk_wait_queue_drained(request_queue_t *, int); extern void blk_finish_queue_drain(request_queue_t *); int blk_get_queue(request_queue_t *); -request_queue_t *blk_alloc_queue(int gfp_mask); -request_queue_t *blk_alloc_queue_node(int,int); +request_queue_t *blk_alloc_queue(gfp_t); +request_queue_t *blk_alloc_queue_node(gfp_t, int); #define blk_put_queue(q) blk_cleanup_queue((q)) /* diff --git a/include/linux/elevator.h b/include/linux/elevator.h index ea6bbc2d7407..ed93125c1db5 100644 --- a/include/linux/elevator.h +++ b/include/linux/elevator.h @@ -18,7 +18,7 @@ typedef struct request *(elevator_request_list_fn) (request_queue_t *, struct re typedef void (elevator_completed_req_fn) (request_queue_t *, struct request *); typedef int (elevator_may_queue_fn) (request_queue_t *, int, struct bio *); -typedef int (elevator_set_req_fn) (request_queue_t *, struct request *, struct bio *, int); +typedef int (elevator_set_req_fn) (request_queue_t *, struct request *, struct bio *, gfp_t); typedef void (elevator_put_req_fn) (request_queue_t *, struct request *); typedef void (elevator_deactivate_req_fn) (request_queue_t *, struct request *); @@ -98,7 +98,7 @@ extern int elv_register_queue(request_queue_t *q); extern void elv_unregister_queue(request_queue_t *q); extern int elv_may_queue(request_queue_t *, int, struct bio *); extern void elv_completed_request(request_queue_t *, struct request *); -extern int elv_set_request(request_queue_t *, struct request *, struct bio *, int); +extern int elv_set_request(request_queue_t *, struct request *, struct bio *, gfp_t); extern void elv_put_request(request_queue_t *, struct request *); /* -- cgit v1.2.3 From 55016f10e31bb15b85d8c500f979dfdceb37d548 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 21 Oct 2005 03:21:58 -0400 Subject: [PATCH] gfp_t: drivers/usb Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/usb.h | 8 ++++---- include/linux/usb_gadget.h | 12 ++++++------ 2 files changed, 10 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/usb.h b/include/linux/usb.h index 4dbe580f9335..8f731e8f2821 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -933,17 +933,17 @@ static inline void usb_fill_int_urb (struct urb *urb, } extern void usb_init_urb(struct urb *urb); -extern struct urb *usb_alloc_urb(int iso_packets, unsigned mem_flags); +extern struct urb *usb_alloc_urb(int iso_packets, gfp_t mem_flags); extern void usb_free_urb(struct urb *urb); #define usb_put_urb usb_free_urb extern struct urb *usb_get_urb(struct urb *urb); -extern int usb_submit_urb(struct urb *urb, unsigned mem_flags); +extern int usb_submit_urb(struct urb *urb, gfp_t mem_flags); extern int usb_unlink_urb(struct urb *urb); extern void usb_kill_urb(struct urb *urb); #define HAVE_USB_BUFFERS void *usb_buffer_alloc (struct usb_device *dev, size_t size, - unsigned mem_flags, dma_addr_t *dma); + gfp_t mem_flags, dma_addr_t *dma); void usb_buffer_free (struct usb_device *dev, size_t size, void *addr, dma_addr_t dma); @@ -1050,7 +1050,7 @@ int usb_sg_init ( struct scatterlist *sg, int nents, size_t length, - unsigned mem_flags + gfp_t mem_flags ); void usb_sg_cancel (struct usb_sg_request *io); void usb_sg_wait (struct usb_sg_request *io); diff --git a/include/linux/usb_gadget.h b/include/linux/usb_gadget.h index 71e608607324..ff81117eb733 100644 --- a/include/linux/usb_gadget.h +++ b/include/linux/usb_gadget.h @@ -107,18 +107,18 @@ struct usb_ep_ops { int (*disable) (struct usb_ep *ep); struct usb_request *(*alloc_request) (struct usb_ep *ep, - unsigned gfp_flags); + gfp_t gfp_flags); void (*free_request) (struct usb_ep *ep, struct usb_request *req); void *(*alloc_buffer) (struct usb_ep *ep, unsigned bytes, - dma_addr_t *dma, unsigned gfp_flags); + dma_addr_t *dma, gfp_t gfp_flags); void (*free_buffer) (struct usb_ep *ep, void *buf, dma_addr_t dma, unsigned bytes); // NOTE: on 2.6, drivers may also use dma_map() and // dma_sync_single_*() to directly manage dma overhead. int (*queue) (struct usb_ep *ep, struct usb_request *req, - unsigned gfp_flags); + gfp_t gfp_flags); int (*dequeue) (struct usb_ep *ep, struct usb_request *req); int (*set_halt) (struct usb_ep *ep, int value); @@ -214,7 +214,7 @@ usb_ep_disable (struct usb_ep *ep) * Returns the request, or null if one could not be allocated. */ static inline struct usb_request * -usb_ep_alloc_request (struct usb_ep *ep, unsigned gfp_flags) +usb_ep_alloc_request (struct usb_ep *ep, gfp_t gfp_flags) { return ep->ops->alloc_request (ep, gfp_flags); } @@ -254,7 +254,7 @@ usb_ep_free_request (struct usb_ep *ep, struct usb_request *req) */ static inline void * usb_ep_alloc_buffer (struct usb_ep *ep, unsigned len, dma_addr_t *dma, - unsigned gfp_flags) + gfp_t gfp_flags) { return ep->ops->alloc_buffer (ep, len, dma, gfp_flags); } @@ -330,7 +330,7 @@ usb_ep_free_buffer (struct usb_ep *ep, void *buf, dma_addr_t dma, unsigned len) * reported when the usb peripheral is disconnected. */ static inline int -usb_ep_queue (struct usb_ep *ep, struct usb_request *req, unsigned gfp_flags) +usb_ep_queue (struct usb_ep *ep, struct usb_request *req, gfp_t gfp_flags) { return ep->ops->queue (ep, req, gfp_flags); } -- cgit v1.2.3 From 9796fdd829da626374458e8706daedcc0e432ddd Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 21 Oct 2005 03:22:03 -0400 Subject: [PATCH] gfp_t: kernel/* Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/audit.h | 4 ++-- include/linux/suspend.h | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index b2a2509bd7ea..da3c01955f3d 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -260,11 +260,11 @@ extern int audit_filter_user(struct netlink_skb_parms *cb, int type); #ifdef CONFIG_AUDIT /* These are defined in audit.c */ /* Public API */ -extern void audit_log(struct audit_context *ctx, int gfp_mask, +extern void audit_log(struct audit_context *ctx, gfp_t gfp_mask, int type, const char *fmt, ...) __attribute__((format(printf,4,5))); -extern struct audit_buffer *audit_log_start(struct audit_context *ctx, int gfp_mask, int type); +extern struct audit_buffer *audit_log_start(struct audit_context *ctx, gfp_t gfp_mask, int type); extern void audit_log_format(struct audit_buffer *ab, const char *fmt, ...) __attribute__((format(printf,2,3))); diff --git a/include/linux/suspend.h b/include/linux/suspend.h index ad15a54806d8..ba448c760168 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -71,7 +71,7 @@ void restore_processor_state(void); struct saved_context; void __save_processor_state(struct saved_context *ctxt); void __restore_processor_state(struct saved_context *ctxt); -extern unsigned long get_usable_page(unsigned gfp_mask); +extern unsigned long get_usable_page(gfp_t gfp_mask); extern void free_eaten_memory(void); #endif /* _LINUX_SWSUSP_H */ -- cgit v1.2.3 From b4e3ca1ab1ae9ae86134126dcdc88da1caaa32ca Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 21 Oct 2005 03:22:34 -0400 Subject: [PATCH] gfp_t: remaining bits of drivers/* Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/i2o.h | 4 ++-- include/linux/loop.h | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/i2o.h b/include/linux/i2o.h index bdc286ec947c..b4af45aad25d 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -492,7 +492,7 @@ static inline int i2o_dma_map_sg(struct i2o_controller *c, * Returns 0 on success or -ENOMEM on failure. */ static inline int i2o_dma_alloc(struct device *dev, struct i2o_dma *addr, - size_t len, unsigned int gfp_mask) + size_t len, gfp_t gfp_mask) { struct pci_dev *pdev = to_pci_dev(dev); int dma_64 = 0; @@ -551,7 +551,7 @@ static inline void i2o_dma_free(struct device *dev, struct i2o_dma *addr) * Returns the 0 on success or negative error code on failure. */ static inline int i2o_dma_realloc(struct device *dev, struct i2o_dma *addr, - size_t len, unsigned int gfp_mask) + size_t len, gfp_t gfp_mask) { i2o_dma_free(dev, addr); diff --git a/include/linux/loop.h b/include/linux/loop.h index 53fa51595443..40f63c9879d2 100644 --- a/include/linux/loop.h +++ b/include/linux/loop.h @@ -52,7 +52,7 @@ struct loop_device { unsigned lo_blocksize; void *key_data; - int old_gfp_mask; + gfp_t old_gfp_mask; spinlock_t lo_lock; struct bio *lo_bio; -- cgit v1.2.3 From 260b23674fdb570f3235ce55892246bef1c24c2a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 21 Oct 2005 03:22:44 -0400 Subject: [PATCH] gfp_t: the rest zone handling, mapping->flags handling Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 2 +- include/linux/pagemap.h | 7 ++++--- 2 files changed, 5 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 5ed471b58f4f..7519eb4191e7 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -302,7 +302,7 @@ void get_zone_counts(unsigned long *active, unsigned long *inactive, void build_all_zonelists(void); void wakeup_kswapd(struct zone *zone, int order); int zone_watermark_ok(struct zone *z, int order, unsigned long mark, - int alloc_type, int can_try_harder, int gfp_high); + int alloc_type, int can_try_harder, gfp_t gfp_high); #ifdef CONFIG_HAVE_MEMORY_PRESENT void memory_present(int nid, unsigned long start, unsigned long end); diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h index efbae53fb078..ba6c310a055f 100644 --- a/include/linux/pagemap.h +++ b/include/linux/pagemap.h @@ -21,16 +21,17 @@ static inline gfp_t mapping_gfp_mask(struct address_space * mapping) { - return mapping->flags & __GFP_BITS_MASK; + return (__force gfp_t)mapping->flags & __GFP_BITS_MASK; } /* * This is non-atomic. Only to be used before the mapping is activated. * Probably needs a barrier... */ -static inline void mapping_set_gfp_mask(struct address_space *m, int mask) +static inline void mapping_set_gfp_mask(struct address_space *m, gfp_t mask) { - m->flags = (m->flags & ~__GFP_BITS_MASK) | mask; + m->flags = (m->flags & ~(__force unsigned long)__GFP_BITS_MASK) | + (__force unsigned long)mask; } /* -- cgit v1.2.3