From 9d853c3757ef74ded5ae564d68370f22208fb88b Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 15 Jul 2005 07:41:31 -0700 Subject: [PATCH 01/38] [CRYPTO]: Fix zero-extension bug on 64-bit architectures. Noticed by Ken-ichirou MATSUZAWA. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- crypto/cipher.c | 6 +++--- crypto/internal.h | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/crypto/cipher.c b/crypto/cipher.c index 1c92c6bb138b..8da644364cb4 100644 --- a/crypto/cipher.c +++ b/crypto/cipher.c @@ -41,7 +41,7 @@ static unsigned int crypt_slow(const struct cipher_desc *desc, struct scatter_walk *in, struct scatter_walk *out, unsigned int bsize) { - unsigned int alignmask = crypto_tfm_alg_alignmask(desc->tfm); + unsigned long alignmask = crypto_tfm_alg_alignmask(desc->tfm); u8 buffer[bsize * 2 + alignmask]; u8 *src = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1); u8 *dst = src + bsize; @@ -160,7 +160,7 @@ static int crypt_iv_unaligned(struct cipher_desc *desc, unsigned int nbytes) { struct crypto_tfm *tfm = desc->tfm; - unsigned int alignmask = crypto_tfm_alg_alignmask(tfm); + unsigned long alignmask = crypto_tfm_alg_alignmask(tfm); u8 *iv = desc->info; if (unlikely(((unsigned long)iv & alignmask))) { @@ -424,7 +424,7 @@ int crypto_init_cipher_ops(struct crypto_tfm *tfm) } if (ops->cit_mode == CRYPTO_TFM_MODE_CBC) { - unsigned int align; + unsigned long align; unsigned long addr; switch (crypto_tfm_alg_blocksize(tfm)) { diff --git a/crypto/internal.h b/crypto/internal.h index 68612874b5fd..37515beafc8c 100644 --- a/crypto/internal.h +++ b/crypto/internal.h @@ -75,7 +75,7 @@ static inline unsigned int crypto_cipher_ctxsize(struct crypto_alg *alg, switch (flags & CRYPTO_TFM_MODE_MASK) { case CRYPTO_TFM_MODE_CBC: - len = ALIGN(len, alg->cra_alignmask + 1); + len = ALIGN(len, (unsigned long)alg->cra_alignmask + 1); len += alg->cra_blocksize; break; } From a8178345c4f12f4413e5620fc86a6b0381c41bc9 Mon Sep 17 00:00:00 2001 From: Marcelo Feitoza Parisi Date: Fri, 15 Jul 2005 09:59:26 -0700 Subject: [PATCH 02/38] [WAN]: drivers/net/wan/: use of time_after macro From: Marcelo Feitoza Parisi Use of the time_after() macro, defined at linux/jiffies.h, which deal with wrapping correctly and are nicer to read. Signed-off-by: Marcelo Feitoza Parisi Signed-off-by: Domen Puncer Signed-off-by: David S. Miller --- drivers/net/wan/sdla_fr.c | 7 ++++--- drivers/net/wan/sdla_ft1.c | 3 ++- drivers/net/wan/sdla_ppp.c | 3 ++- drivers/net/wan/sdla_x25.c | 3 ++- drivers/net/wan/wanpipe_multppp.c | 9 +++++---- 5 files changed, 15 insertions(+), 10 deletions(-) diff --git a/drivers/net/wan/sdla_fr.c b/drivers/net/wan/sdla_fr.c index 2efccb0554c0..c5f5e62aab8b 100644 --- a/drivers/net/wan/sdla_fr.c +++ b/drivers/net/wan/sdla_fr.c @@ -152,6 +152,7 @@ #include /* for inb(), outb(), etc. */ #include /* for do_gettimeofday */ #include /* sockaddr_in */ +#include /* time_after() macro */ #include #include @@ -773,7 +774,7 @@ static int update(struct wan_device* wandev) for(;;) { if(card->u.f.update_comms_stats == 0) break; - if ((jiffies - timeout) > (1 * HZ)){ + if (time_after(jiffies, timeout + 1 * HZ)){ card->u.f.update_comms_stats = 0; return -EAGAIN; } @@ -4799,7 +4800,7 @@ static void trigger_unconfig_fr(struct net_device *dev) { fr_channel_t *chan = dev->priv; volatile sdla_t *card = chan->card; - u32 timeout; + unsigned long timeout; fr508_flags_t* flags = card->flags; int reset_critical=0; @@ -4821,7 +4822,7 @@ static void trigger_unconfig_fr(struct net_device *dev) if(!(card->u.f.timer_int_enabled & TMR_INT_ENABLED_UNCONFIG)) break; - if ((jiffies - timeout) > (1 * HZ)){ + if (time_after(jiffies, timeout + 1 * HZ)){ card->u.f.timer_int_enabled &= ~TMR_INT_ENABLED_UNCONFIG; printk(KERN_INFO "%s: Failed to delete DLCI %i\n", card->devname,chan->dlci); diff --git a/drivers/net/wan/sdla_ft1.c b/drivers/net/wan/sdla_ft1.c index 5e3124856eb0..9d6528a50f7b 100644 --- a/drivers/net/wan/sdla_ft1.c +++ b/drivers/net/wan/sdla_ft1.c @@ -29,6 +29,7 @@ #include /* WAN router definitions */ #include /* WANPIPE common user API definitions */ #include /* ARPHRD_* defines */ +#include /* time_after() macro */ #include #include @@ -164,7 +165,7 @@ int wpft1_init (sdla_t* card, wandev_conf_t* conf) timeout = jiffies; while (mb->return_code != 'I') /* Wait 1s for board to initialize */ - if ((jiffies - timeout) > 1*HZ) break; + if (time_after(jiffies, timeout + 1*HZ)) break; if (mb->return_code != 'I') { printk(KERN_INFO diff --git a/drivers/net/wan/sdla_ppp.c b/drivers/net/wan/sdla_ppp.c index 1761cb68ab48..a4b489cccbbf 100644 --- a/drivers/net/wan/sdla_ppp.c +++ b/drivers/net/wan/sdla_ppp.c @@ -101,6 +101,7 @@ #include /* ARPHRD_* defines */ #include /* htons(), etc. */ #include /* sockaddr_in */ +#include /* time_after() macro */ #include @@ -482,7 +483,7 @@ static int update(struct wan_device *wandev) if(ppp_priv_area->update_comms_stats == 0){ break; } - if ((jiffies - timeout) > (1 * HZ)){ + if (time_after(jiffies, timeout + 1 * HZ)){ ppp_priv_area->update_comms_stats = 0; ppp_priv_area->timer_int_enabled &= ~TMR_INT_ENABLED_UPDATE; diff --git a/drivers/net/wan/sdla_x25.c b/drivers/net/wan/sdla_x25.c index 3a93d2fd4fbf..8a95d61a2f8f 100644 --- a/drivers/net/wan/sdla_x25.c +++ b/drivers/net/wan/sdla_x25.c @@ -91,6 +91,7 @@ #include /* WAN router definitions */ #include /* WANPIPE common user API definitions */ #include +#include /* time_after() macro */ #include /* htons(), etc. */ #include #include /* Experimental delay */ @@ -867,7 +868,7 @@ static int update(struct wan_device* wandev) if (!(card->u.x.timer_int_enabled & TMR_INT_ENABLED_UPDATE)){ break; } - if ((jiffies-timeout) > 1*HZ){ + if (time_after(jiffies, timeout + 1*HZ)){ card->u.x.timer_int_enabled &= ~TMR_INT_ENABLED_UPDATE; return -EAGAIN; } diff --git a/drivers/net/wan/wanpipe_multppp.c b/drivers/net/wan/wanpipe_multppp.c index 6aa6987d96cb..812a1183c502 100644 --- a/drivers/net/wan/wanpipe_multppp.c +++ b/drivers/net/wan/wanpipe_multppp.c @@ -26,6 +26,7 @@ #include /* WAN router definitions */ #include /* WANPIPE common user API definitions */ #include /* ARPHRD_* defines */ +#include /* time_after() macro */ #include /* sockaddr_in */ #include @@ -270,9 +271,9 @@ int wsppp_init (sdla_t* card, wandev_conf_t* conf) ready to accept commands. We expect this to be completed in less than 1 second. */ - timeout = jiffies; + timeout = jiffies + 1 * HZ; while (mb->return_code != 'I') /* Wait 1s for board to initialize */ - if ((jiffies - timeout) > 1*HZ) break; + if (time_after(jiffies, timeout)) break; if (mb->return_code != 'I') { printk(KERN_INFO @@ -493,11 +494,11 @@ static int update(struct wan_device* wandev) chdlc_priv_area->timer_int_enabled = TMR_INT_ENABLED_UPDATE; /* wait a maximum of 1 second for the statistics to be updated */ - timeout = jiffies; + timeout = jiffies + 1 * HZ; for(;;) { if(chdlc_priv_area->update_comms_stats == 0) break; - if ((jiffies - timeout) > (1 * HZ)){ + if (time_after(jiffies, timeout)){ chdlc_priv_area->update_comms_stats = 0; chdlc_priv_area->timer_int_enabled &= ~TMR_INT_ENABLED_UPDATE; From 4851d3aaa550b7b62553dd19fcadc7b586da4f17 Mon Sep 17 00:00:00 2001 From: Marcelo Feitoza Parisi Date: Fri, 15 Jul 2005 10:00:41 -0700 Subject: [PATCH 03/38] [NET PCMCIA]: drivers/net/pcmcia/smc91c92_cs.c : Use of time_after macro From: Marcelo Feitoza Parisi Use of the time_after() macro, defined at linux/jiffies.h, which deal with wrapping correctly and are nicer to read. Signed-off-by: Marcelo Feitoza Parisi Signed-off-by: Domen Puncer Signed-off-by: David S. Miller --- drivers/net/pcmcia/smc91c92_cs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/pcmcia/smc91c92_cs.c b/drivers/net/pcmcia/smc91c92_cs.c index fbc2f58ff688..0d8bb4cccbb7 100644 --- a/drivers/net/pcmcia/smc91c92_cs.c +++ b/drivers/net/pcmcia/smc91c92_cs.c @@ -41,6 +41,7 @@ #include #include #include +#include #include #include @@ -2092,7 +2093,7 @@ static void media_check(u_long arg) } /* Ignore collisions unless we've had no rx's recently */ - if (jiffies - dev->last_rx > HZ) { + if (time_after(jiffies, dev->last_rx + HZ)) { if (smc->tx_err || (smc->media_status & EPH_16COL)) media |= EPH_16COL; } From d7c7ed4dbc66c5f4dfa5615cdfc5009af3b3b137 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Mon, 18 Jul 2005 13:29:49 -0700 Subject: [PATCH 04/38] [PKT_SCHED]: Remove debugging leftover from textsearch ematch Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/sched/em_text.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/net/sched/em_text.c b/net/sched/em_text.c index 873840d8d072..77beabc91fa3 100644 --- a/net/sched/em_text.c +++ b/net/sched/em_text.c @@ -55,9 +55,6 @@ static int em_text_change(struct tcf_proto *tp, void *data, int len, struct ts_config *ts_conf; int flags = 0; - printk("Configuring text: %s from %d:%d to %d:%d len %d\n", conf->algo, conf->from_offset, - conf->from_layer, conf->to_offset, conf->to_layer, conf->pattern_len); - if (len < sizeof(*conf) || len < (sizeof(*conf) + conf->pattern_len)) return -EINVAL; From 452f299da3253f65020143f743c2e207b752547b Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Mon, 18 Jul 2005 13:30:53 -0700 Subject: [PATCH 05/38] [PKT_SCHED]: Reduce branch mispredictions in pfifo_fast_dequeue The current call to __qdisc_dequeue_head leads to a branch misprediction for every loop iteration, the fact that the most common priority is 2 makes this even worse. This issue has been brought up by Eric Dumazet but unlike his solution which was to manually unroll the loop, this approach preserves the possibility to increase the number of bands at compile time. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/sched/sch_generic.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 73e218e646ac..8edefd5d095d 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -331,11 +331,10 @@ static struct sk_buff *pfifo_fast_dequeue(struct Qdisc* qdisc) int prio; struct sk_buff_head *list = qdisc_priv(qdisc); - for (prio = 0; prio < PFIFO_FAST_BANDS; prio++, list++) { - struct sk_buff *skb = __qdisc_dequeue_head(qdisc, list); - if (skb) { + for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) { + if (!skb_queue_empty(list + prio)) { qdisc->q.qlen--; - return skb; + return __qdisc_dequeue_head(qdisc, list + prio); } } From 23af27eb8fa9ea8614138c4cded7a16cb4197a55 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 18 Jul 2005 13:34:35 -0700 Subject: [PATCH 06/38] [PKT_SCHED]: Kill TCF_META_ID_TCCLASSID. Thomas Graf states: > I used to mark such ids as obsolete in the header but since > skb is on diet anyway and there has been no official > iproute2 release with the ematch bits included it might be > a better idea to remove the ids from the header completely. > Those that have picked up my patch on netdev shouldn't care > about a ABI breakage, actually I doubt that someone is using > it already. So here's the patch to remove it. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/tc_ematch/tc_em_meta.h | 1 - 1 file changed, 1 deletion(-) diff --git a/include/linux/tc_ematch/tc_em_meta.h b/include/linux/tc_ematch/tc_em_meta.h index bcb762d93123..90ab9f95d43d 100644 --- a/include/linux/tc_ematch/tc_em_meta.h +++ b/include/linux/tc_ematch/tc_em_meta.h @@ -53,7 +53,6 @@ enum TCF_META_ID_NFMARK, TCF_META_ID_TCINDEX, TCF_META_ID_TCVERDICT, - TCF_META_ID_TCCLASSID, TCF_META_ID_RTCLASSID, TCF_META_ID_RTIIF, TCF_META_ID_SK_FAMILY, From 37da647d994cdac7e0bc8d2a365fbda403939a2b Mon Sep 17 00:00:00 2001 From: Victor Fusco Date: Mon, 18 Jul 2005 13:35:43 -0700 Subject: [PATCH 07/38] [NETLINK]: Fix "nocast type" warnings From: Victor Fusco Fix the sparse warning "implicit cast to nocast type" Signed-off-by: Victor Fusco Signed-off-by: Domen Puncer Signed-off-by: David S. Miller --- net/netlink/af_netlink.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 3405fdf41b93..ff774a06c89d 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -648,7 +648,8 @@ void netlink_detachskb(struct sock *sk, struct sk_buff *skb) sock_put(sk); } -static inline struct sk_buff *netlink_trim(struct sk_buff *skb, int allocation) +static inline struct sk_buff *netlink_trim(struct sk_buff *skb, + unsigned int __nocast allocation) { int delta; @@ -717,7 +718,7 @@ struct netlink_broadcast_data { int failure; int congested; int delivered; - int allocation; + unsigned int allocation; struct sk_buff *skb, *skb2; }; From e2bf521d9728bfae9b6c3d484614e5962d0b5afd Mon Sep 17 00:00:00 2001 From: Victor Fusco Date: Mon, 18 Jul 2005 13:36:38 -0700 Subject: [PATCH 08/38] [NET]: Fix "nocast type" warnings in skbuff.h From: Victor Fusco Fix the sparse warning "implicit cast to nocast type" Signed-off-by: Victor Fusco Signed-off-by: Domen Puncer Signed-off-by: David S. Miller --- include/linux/skbuff.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 5d4a990d5577..0061c9470482 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -502,7 +502,8 @@ static inline struct sk_buff *skb_share_check(struct sk_buff *skb, * * %NULL is returned on a memory allocation failure. */ -static inline struct sk_buff *skb_unshare(struct sk_buff *skb, int pri) +static inline struct sk_buff *skb_unshare(struct sk_buff *skb, + unsigned int __nocast pri) { might_sleep_if(pri & __GFP_WAIT); if (skb_cloned(skb)) { From ee71a29eb5e341fe977c5ad7a43782c29bd9cb9e Mon Sep 17 00:00:00 2001 From: Christophe Lucas Date: Mon, 18 Jul 2005 13:38:07 -0700 Subject: [PATCH 09/38] [SCTP]: Audit return code of create_proc_* From: Christophe Lucas Audit return of create_proc_* functions. Signed-off-by: Christophe Lucas Signed-off-by: Domen Puncer Signed-off-by: David S. Miller --- net/sctp/objcnt.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/sctp/objcnt.c b/net/sctp/objcnt.c index 0781e5d509fd..8ff588f0d76a 100644 --- a/net/sctp/objcnt.c +++ b/net/sctp/objcnt.c @@ -127,8 +127,12 @@ done: /* Initialize the objcount in the proc filesystem. */ void sctp_dbg_objcnt_init(void) { - create_proc_read_entry("sctp_dbg_objcnt", 0, proc_net_sctp, + struct proc_dir_entry *ent; + ent = create_proc_read_entry("sctp_dbg_objcnt", 0, proc_net_sctp, sctp_dbg_objcnt_read, NULL); + if (!ent) + printk(KERN_WARNING + "sctp_dbg_objcnt: Unable to create /proc entry.\n"); } /* Cleanup the objcount entry in the proc filesystem. */ From d1ad1ff299dd908d07c5e5f27f88bbdb235eb7a5 Mon Sep 17 00:00:00 2001 From: Sridhar Samudrala Date: Mon, 18 Jul 2005 13:44:10 -0700 Subject: [PATCH 10/38] [SCTP]: Fix potential null pointer dereference while handling an icmp error Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- include/net/sctp/sctp.h | 7 ++---- net/sctp/input.c | 47 ++++++++++++----------------------------- net/sctp/ipv6.c | 7 +++--- 3 files changed, 18 insertions(+), 43 deletions(-) diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 4a26adfaed71..e1d5ec1c23c0 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -167,15 +167,12 @@ void sctp_unhash_established(struct sctp_association *); void sctp_hash_endpoint(struct sctp_endpoint *); void sctp_unhash_endpoint(struct sctp_endpoint *); struct sock *sctp_err_lookup(int family, struct sk_buff *, - struct sctphdr *, struct sctp_endpoint **, - struct sctp_association **, + struct sctphdr *, struct sctp_association **, struct sctp_transport **); -void sctp_err_finish(struct sock *, struct sctp_endpoint *, - struct sctp_association *); +void sctp_err_finish(struct sock *, struct sctp_association *); void sctp_icmp_frag_needed(struct sock *, struct sctp_association *, struct sctp_transport *t, __u32 pmtu); void sctp_icmp_proto_unreachable(struct sock *sk, - struct sctp_endpoint *ep, struct sctp_association *asoc, struct sctp_transport *t); diff --git a/net/sctp/input.c b/net/sctp/input.c index 5e085e041a6e..742be9171b7d 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -351,7 +351,6 @@ void sctp_icmp_frag_needed(struct sock *sk, struct sctp_association *asoc, * */ void sctp_icmp_proto_unreachable(struct sock *sk, - struct sctp_endpoint *ep, struct sctp_association *asoc, struct sctp_transport *t) { @@ -367,7 +366,6 @@ void sctp_icmp_proto_unreachable(struct sock *sk, /* Common lookup code for icmp/icmpv6 error handler. */ struct sock *sctp_err_lookup(int family, struct sk_buff *skb, struct sctphdr *sctphdr, - struct sctp_endpoint **epp, struct sctp_association **app, struct sctp_transport **tpp) { @@ -375,11 +373,10 @@ struct sock *sctp_err_lookup(int family, struct sk_buff *skb, union sctp_addr daddr; struct sctp_af *af; struct sock *sk = NULL; - struct sctp_endpoint *ep = NULL; struct sctp_association *asoc = NULL; struct sctp_transport *transport = NULL; - *app = NULL; *epp = NULL; *tpp = NULL; + *app = NULL; *tpp = NULL; af = sctp_get_af_specific(family); if (unlikely(!af)) { @@ -394,27 +391,16 @@ struct sock *sctp_err_lookup(int family, struct sk_buff *skb, * packet. */ asoc = __sctp_lookup_association(&saddr, &daddr, &transport); - if (!asoc) { - /* If there is no matching association, see if it matches any - * endpoint. This may happen for an ICMP error generated in - * response to an INIT_ACK. - */ - ep = __sctp_rcv_lookup_endpoint(&daddr); - if (!ep) { - return NULL; - } + if (!asoc) + return NULL; + + sk = asoc->base.sk; + + if (ntohl(sctphdr->vtag) != asoc->c.peer_vtag) { + ICMP_INC_STATS_BH(ICMP_MIB_INERRORS); + goto out; } - if (asoc) { - sk = asoc->base.sk; - - if (ntohl(sctphdr->vtag) != asoc->c.peer_vtag) { - ICMP_INC_STATS_BH(ICMP_MIB_INERRORS); - goto out; - } - } else - sk = ep->base.sk; - sctp_bh_lock_sock(sk); /* If too many ICMPs get dropped on busy @@ -423,7 +409,6 @@ struct sock *sctp_err_lookup(int family, struct sk_buff *skb, if (sock_owned_by_user(sk)) NET_INC_STATS_BH(LINUX_MIB_LOCKDROPPEDICMPS); - *epp = ep; *app = asoc; *tpp = transport; return sk; @@ -432,21 +417,16 @@ out: sock_put(sk); if (asoc) sctp_association_put(asoc); - if (ep) - sctp_endpoint_put(ep); return NULL; } /* Common cleanup code for icmp/icmpv6 error handler. */ -void sctp_err_finish(struct sock *sk, struct sctp_endpoint *ep, - struct sctp_association *asoc) +void sctp_err_finish(struct sock *sk, struct sctp_association *asoc) { sctp_bh_unlock_sock(sk); sock_put(sk); if (asoc) sctp_association_put(asoc); - if (ep) - sctp_endpoint_put(ep); } /* @@ -471,7 +451,6 @@ void sctp_v4_err(struct sk_buff *skb, __u32 info) int type = skb->h.icmph->type; int code = skb->h.icmph->code; struct sock *sk; - struct sctp_endpoint *ep; struct sctp_association *asoc; struct sctp_transport *transport; struct inet_sock *inet; @@ -488,7 +467,7 @@ void sctp_v4_err(struct sk_buff *skb, __u32 info) savesctp = skb->h.raw; skb->nh.iph = iph; skb->h.raw = (char *)sh; - sk = sctp_err_lookup(AF_INET, skb, sh, &ep, &asoc, &transport); + sk = sctp_err_lookup(AF_INET, skb, sh, &asoc, &transport); /* Put back, the original pointers. */ skb->nh.raw = saveip; skb->h.raw = savesctp; @@ -515,7 +494,7 @@ void sctp_v4_err(struct sk_buff *skb, __u32 info) } else { if (ICMP_PROT_UNREACH == code) { - sctp_icmp_proto_unreachable(sk, ep, asoc, + sctp_icmp_proto_unreachable(sk, asoc, transport); goto out_unlock; } @@ -544,7 +523,7 @@ void sctp_v4_err(struct sk_buff *skb, __u32 info) } out_unlock: - sctp_err_finish(sk, ep, asoc); + sctp_err_finish(sk, asoc); } /* diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index c7e42d125b9c..e9b2fd480d61 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -91,7 +91,6 @@ SCTP_STATIC void sctp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, struct ipv6hdr *iph = (struct ipv6hdr *)skb->data; struct sctphdr *sh = (struct sctphdr *)(skb->data + offset); struct sock *sk; - struct sctp_endpoint *ep; struct sctp_association *asoc; struct sctp_transport *transport; struct ipv6_pinfo *np; @@ -105,7 +104,7 @@ SCTP_STATIC void sctp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, savesctp = skb->h.raw; skb->nh.ipv6h = iph; skb->h.raw = (char *)sh; - sk = sctp_err_lookup(AF_INET6, skb, sh, &ep, &asoc, &transport); + sk = sctp_err_lookup(AF_INET6, skb, sh, &asoc, &transport); /* Put back, the original pointers. */ skb->nh.raw = saveip; skb->h.raw = savesctp; @@ -124,7 +123,7 @@ SCTP_STATIC void sctp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, goto out_unlock; case ICMPV6_PARAMPROB: if (ICMPV6_UNK_NEXTHDR == code) { - sctp_icmp_proto_unreachable(sk, ep, asoc, transport); + sctp_icmp_proto_unreachable(sk, asoc, transport); goto out_unlock; } break; @@ -142,7 +141,7 @@ SCTP_STATIC void sctp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, } out_unlock: - sctp_err_finish(sk, ep, asoc); + sctp_err_finish(sk, asoc); out: if (likely(idev != NULL)) in6_dev_put(idev); From 54208991e15fa00e37a9d172ac0d87191a832165 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 18 Jul 2005 13:45:12 -0700 Subject: [PATCH 11/38] [NET]: Kconfig: NETCONSOLE and NETPOLL together Put NETCONSOLE and NETPOLL options together since they are related. This cuts down on the hassle of flipping back and forth between the Networking menu and the Network drivers menu to change their config settings. Tested with menuconfig, gconfig, and xconfig. gconfig has a small problem with this. I think that it's a bug in gconfig and I will take it up with Romain Lievin. Signed-off-by: Randy Dunlap Signed-off-by: David S. Miller --- drivers/net/Kconfig | 16 ++++++++++++++++ net/Kconfig | 16 ---------------- 2 files changed, 16 insertions(+), 16 deletions(-) diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 9a07ff7a7777..f827f0a1e1a6 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -2549,4 +2549,20 @@ config NETCONSOLE If you want to log kernel messages over the network, enable this. See for details. +config NETPOLL + def_bool NETCONSOLE + +config NETPOLL_RX + bool "Netpoll support for trapping incoming packets" + default n + depends on NETPOLL + +config NETPOLL_TRAP + bool "Netpoll traffic trapping" + default n + depends on NETPOLL + +config NET_POLL_CONTROLLER + def_bool NETPOLL + endmenu diff --git a/net/Kconfig b/net/Kconfig index 2684e809a649..40a31ba86d2c 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -209,22 +209,6 @@ endmenu endmenu -config NETPOLL - def_bool NETCONSOLE - -config NETPOLL_RX - bool "Netpoll support for trapping incoming packets" - default n - depends on NETPOLL - -config NETPOLL_TRAP - bool "Netpoll traffic trapping" - default n - depends on NETPOLL - -config NET_POLL_CONTROLLER - def_bool NETPOLL - source "net/ax25/Kconfig" source "net/irda/Kconfig" source "net/bluetooth/Kconfig" From 6876f95f201ce2d62367d85dd793c1fee351d7a6 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 18 Jul 2005 13:55:19 -0700 Subject: [PATCH 12/38] [IPV4]: fix IP_FIB_HASH kconfig warning This patch fixes the following kconfig warning: net/ipv4/Kconfig:92:warning: defaults for choice values not supported Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- net/ipv4/Kconfig | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index df5386885a90..ecea2bc8229f 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -54,9 +54,9 @@ config IP_ADVANCED_ROUTER choice prompt "Choose IP: FIB lookup algorithm (choose FIB_HASH if unsure)" depends on IP_ADVANCED_ROUTER - default IP_FIB_HASH + default ASK_IP_FIB_HASH -config IP_FIB_HASH +config ASK_IP_FIB_HASH bool "FIB_HASH" ---help--- Current FIB is very proven and good enough for most users. @@ -82,12 +82,8 @@ config IP_FIB_TRIE endchoice -# If the user does not enable advanced routing, he gets the safe -# default of the fib-hash algorithm. config IP_FIB_HASH - bool - depends on !IP_ADVANCED_ROUTER - default y + def_bool ASK_IP_FIB_HASH || !IP_ADVANCED_ROUTER config IP_MULTIPLE_TABLES bool "IP: policy routing" From 3f1c81ff10187396e6d0fd01df82ab0cec256c96 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 18 Jul 2005 17:10:55 -0700 Subject: [PATCH 13/38] [EMATCH]: Kill TCF_META_ID_TCCLASSID reference from meta ematch as well. Signed-off-by: David S. Miller --- net/sched/em_meta.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index 53d98f8d3d80..feefcbfd03d0 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -250,11 +250,6 @@ META_COLLECTOR(int_tcverd) { dst->value = skb->tc_verd; } - -META_COLLECTOR(int_tcclassid) -{ - dst->value = skb->tc_classid; -} #endif /************************************************************************** @@ -529,7 +524,6 @@ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [META_ID(TCINDEX)] = META_FUNC(int_tcindex), #ifdef CONFIG_NET_CLS_ACT [META_ID(TCVERDICT)] = META_FUNC(int_tcverd), - [META_ID(TCCLASSID)] = META_FUNC(int_tcclassid), #endif #ifdef CONFIG_NET_CLS_ROUTE [META_ID(RTCLASSID)] = META_FUNC(int_rtclassid), From 43f51fce0a637e6cdf1285b0fa09b8398ff14834 Mon Sep 17 00:00:00 2001 From: Chas Williams Date: Tue, 19 Jul 2005 13:54:19 -0700 Subject: [PATCH 14/38] [ATM]: [zatm] eliminate kfree warning (from Tobias Hirning ) Signed-off-by: Chas Williams Signed-off-by: David S. Miller --- drivers/atm/zatm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/atm/zatm.c b/drivers/atm/zatm.c index 8d5e65cb9755..afcf1ada5547 100644 --- a/drivers/atm/zatm.c +++ b/drivers/atm/zatm.c @@ -1339,7 +1339,7 @@ static int __init zatm_start(struct atm_dev *dev) return 0; out: for (i = 0; i < NR_MBX; i++) - kfree(zatm_dev->mbx_start[i]); + kfree(&zatm_dev->mbx_start[i]); kfree(zatm_dev->rx_map); kfree(zatm_dev->tx_map); free_irq(zatm_dev->irq, dev); From 322361b3712af97c19ab730b410f9ae5d9aad4c3 Mon Sep 17 00:00:00 2001 From: Chas Williams Date: Tue, 19 Jul 2005 13:54:44 -0700 Subject: [PATCH 15/38] [ATM]: allow bind() on point-to-multpoint svcs (from Martin Whitaker ) Signed-off-by: Chas Williams Signed-off-by: David S. Miller --- net/atm/svc.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/net/atm/svc.c b/net/atm/svc.c index 02f5374a51f2..08e46052a3e4 100644 --- a/net/atm/svc.c +++ b/net/atm/svc.c @@ -118,10 +118,6 @@ static int svc_bind(struct socket *sock,struct sockaddr *sockaddr, goto out; } vcc = ATM_SD(sock); - if (test_bit(ATM_VF_SESSION, &vcc->flags)) { - error = -EINVAL; - goto out; - } addr = (struct sockaddr_atmsvc *) sockaddr; if (addr->sas_family != AF_ATMSVC) { error = -EAFNOSUPPORT; From e1bd232b641a0d0ef184a5154409ef38aebb3155 Mon Sep 17 00:00:00 2001 From: Marcelo Feitoza Parisi Date: Tue, 19 Jul 2005 13:55:14 -0700 Subject: [PATCH 16/38] [ATM]: [idt77252] use time_after() macro Signed-off-by: Marcelo Feitoza Parisi Signed-off-by: Domen Puncer Signed-off-by: Chas Williams Signed-off-by: David S. Miller --- drivers/atm/idt77252.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/atm/idt77252.c b/drivers/atm/idt77252.c index baaf1a3d2242..30b7e990ed0b 100644 --- a/drivers/atm/idt77252.c +++ b/drivers/atm/idt77252.c @@ -46,6 +46,7 @@ static char const rcsid[] = #include #include #include +#include #include #include #include @@ -780,7 +781,7 @@ push_on_scq(struct idt77252_dev *card, struct vc_map *vc, struct sk_buff *skb) return 0; out: - if (jiffies - scq->trans_start > HZ) { + if (time_after(jiffies, scq->trans_start + HZ)) { printk("%s: Error pushing TBD for %d.%d\n", card->name, vc->tx_vcc->vpi, vc->tx_vcc->vci); #ifdef CONFIG_ATM_IDT77252_DEBUG From 5e8676d113a097e19d18492e85215f0f3cc86e43 Mon Sep 17 00:00:00 2001 From: Olaf Hering Date: Tue, 19 Jul 2005 13:55:38 -0700 Subject: [PATCH 17/38] [ATM]: [he] remove linux/version.h include Signed-off-by: Olaf Hering Signed-off-by: Chas Williams Signed-off-by: David S. Miller --- drivers/atm/he.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/atm/he.c b/drivers/atm/he.c index df2c83fd5496..28250c9b32d6 100644 --- a/drivers/atm/he.c +++ b/drivers/atm/he.c @@ -57,7 +57,6 @@ #include #include -#include #include #include #include From c9e42614585dddd544a56907ff5fd5ca55411967 Mon Sep 17 00:00:00 2001 From: Victor Fusco Date: Tue, 19 Jul 2005 13:56:01 -0700 Subject: [PATCH 18/38] [ATM]: [firestream] fix the sparse warning "implicit cast to nocast type" Signed-off-by: Victor Fusco Signed-off-by: Domen Puncer Signed-off-by: Chas Williams Signed-off-by: David S. Miller --- drivers/atm/firestream.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/atm/firestream.c b/drivers/atm/firestream.c index 101f0cc33d10..b078fa548ebf 100644 --- a/drivers/atm/firestream.c +++ b/drivers/atm/firestream.c @@ -1374,7 +1374,8 @@ static void reset_chip (struct fs_dev *dev) } } -static void __devinit *aligned_kmalloc (int size, int flags, int alignment) +static void __devinit *aligned_kmalloc (int size, unsigned int __nocast flags, + int alignment) { void *t; @@ -1464,7 +1465,8 @@ static inline int nr_buffers_in_freepool (struct fs_dev *dev, struct freepool *f does. I've seen "receive abort: no buffers" and things started working again after that... -- REW */ -static void top_off_fp (struct fs_dev *dev, struct freepool *fp, int gfp_flags) +static void top_off_fp (struct fs_dev *dev, struct freepool *fp, + unsigned int __nocast gfp_flags) { struct FS_BPENTRY *qe, *ne; struct sk_buff *skb; From 5938a7b580bb7a738ee349048565ac76e6f3f5d7 Mon Sep 17 00:00:00 2001 From: Victor Fusco Date: Tue, 19 Jul 2005 13:56:29 -0700 Subject: [PATCH 19/38] [ATM]: [ambassador] Fix the sparse warning "implicit cast to nocast type" Signed-off-by: Victor Fusco Signed-off-by: Domen Puncer Signed-off-by: Chas Williams Signed-off-by: David S. Miller --- drivers/atm/ambassador.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/atm/ambassador.c b/drivers/atm/ambassador.c index c46d9520c5a7..73c6b85299c1 100644 --- a/drivers/atm/ambassador.c +++ b/drivers/atm/ambassador.c @@ -794,7 +794,9 @@ static void drain_rx_pools (amb_dev * dev) { drain_rx_pool (dev, pool); } -static inline void fill_rx_pool (amb_dev * dev, unsigned char pool, int priority) { +static inline void fill_rx_pool (amb_dev * dev, unsigned char pool, + unsigned int __nocast priority) +{ rx_in rx; amb_rxq * rxq; From 88e9fa8a54cf4d66ee8fc1d855e82de827233d74 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Tue, 19 Jul 2005 13:56:53 -0700 Subject: [PATCH 20/38] [ATM]: Trivial spelling fix patch for net/Kconfig Signed-off-by: Jesper Juhl Signed-off-by: Adrian Bunk Signed-off-by: Chas Williams Signed-off-by: David S. Miller --- net/atm/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/atm/Kconfig b/net/atm/Kconfig index bea2426229b1..21ff276b2d80 100644 --- a/net/atm/Kconfig +++ b/net/atm/Kconfig @@ -60,7 +60,7 @@ config ATM_BR2684 tristate "RFC1483/2684 Bridged protocols" depends on ATM && INET help - ATM PVCs can carry ethernet PDUs according to rfc2684 (formerly 1483) + ATM PVCs can carry ethernet PDUs according to RFC2684 (formerly 1483) This device will act like an ethernet from the kernels point of view, with the traffic being carried by ATM PVCs (currently 1 PVC/device). This is sometimes used over DSL lines. If in doubt, say N. @@ -69,6 +69,6 @@ config ATM_BR2684_IPFILTER bool "Per-VC IP filter kludge" depends on ATM_BR2684 help - This is an experimental mechanism for users who need to terminating a + This is an experimental mechanism for users who need to terminate a large number of IP-only vcc's. Do not enable this unless you are sure you know what you are doing. From 86cf42e4e029b83110cf98692420239103363dbf Mon Sep 17 00:00:00 2001 From: Duncan Sands Date: Tue, 19 Jul 2005 13:57:17 -0700 Subject: [PATCH 21/38] [ATM]: [speedtch] cure atm_printk() macro gcc-2.95 compile error Signed-off-by: Duncan Sands Signed-off-by: Chas Williams Signed-off-by: David S. Miller --- drivers/usb/atm/speedtch.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/usb/atm/speedtch.c b/drivers/usb/atm/speedtch.c index d0cbbb7f0385..6bd581e69afd 100644 --- a/drivers/usb/atm/speedtch.c +++ b/drivers/usb/atm/speedtch.c @@ -448,19 +448,19 @@ static void speedtch_check_status(struct speedtch_instance_data *instance) case 0: atm_dev->signal = ATM_PHY_SIG_LOST; if (instance->last_status) - atm_info(usbatm, "ADSL line is down\n"); + atm_info(usbatm, "%s\n", "ADSL line is down"); /* It may never resync again unless we ask it to... */ ret = speedtch_start_synchro(instance); break; case 0x08: atm_dev->signal = ATM_PHY_SIG_UNKNOWN; - atm_info(usbatm, "ADSL line is blocked?\n"); + atm_info(usbatm, "%s\n", "ADSL line is blocked?"); break; case 0x10: atm_dev->signal = ATM_PHY_SIG_LOST; - atm_info(usbatm, "ADSL line is synchronising\n"); + atm_info(usbatm, "%s\n", "ADSL line is synchronising"); break; case 0x20: @@ -502,7 +502,7 @@ static void speedtch_status_poll(unsigned long data) if (instance->poll_delay < MAX_POLL_DELAY) mod_timer(&instance->status_checker.timer, jiffies + msecs_to_jiffies(instance->poll_delay)); else - atm_warn(instance->usbatm, "Too many failures - disabling line status polling\n"); + atm_warn(instance->usbatm, "%s\n", "Too many failures - disabling line status polling"); } static void speedtch_resubmit_int(unsigned long data) @@ -545,9 +545,9 @@ static void speedtch_handle_int(struct urb *int_urb, struct pt_regs *regs) if ((count == 6) && !memcmp(up_int, instance->int_data, 6)) { del_timer(&instance->status_checker.timer); - atm_info(usbatm, "DSL line goes up\n"); + atm_info(usbatm, "%s\n", "DSL line goes up"); } else if ((count == 6) && !memcmp(down_int, instance->int_data, 6)) { - atm_info(usbatm, "DSL line goes down\n"); + atm_info(usbatm, "%s\n", "DSL line goes down"); } else { int i; From 6aef4fdfeaec5f2c66415f2cafa98a3ff927501f Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 19 Jul 2005 13:58:40 -0700 Subject: [PATCH 22/38] [NET]: Only build flow.o if CONFIG_XFRM=y Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/core/Makefile | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/core/Makefile b/net/core/Makefile index 5e0c56b7f607..f5f5e58943e8 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -7,9 +7,10 @@ obj-y := sock.o request_sock.o skbuff.o iovec.o datagram.o stream.o scm.o \ obj-$(CONFIG_SYSCTL) += sysctl_net_core.o -obj-y += flow.o dev.o ethtool.o dev_mcast.o dst.o \ +obj-y += dev.o ethtool.o dev_mcast.o dst.o \ neighbour.o rtnetlink.o utils.o link_watch.o filter.o +obj-$(CONFIG_XFRM) += flow.o obj-$(CONFIG_SYSFS) += net-sysfs.o obj-$(CONFIG_NETFILTER) += netfilter.o obj-$(CONFIG_NET_DIVERT) += dv.o From abaacad9bcb3f118cc802f527ab5d7c41b63f83a Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 19 Jul 2005 13:59:17 -0700 Subject: [PATCH 23/38] [IPV4]: Don't select XFRM for ip_gre Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/Kconfig | 1 - 1 file changed, 1 deletion(-) diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index ecea2bc8229f..c7f1f9245bc9 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -252,7 +252,6 @@ config NET_IPIP config NET_IPGRE tristate "IP: GRE tunnels over IP" - select XFRM help Tunneling means encapsulating data of one protocol type within another protocol and sending it over a channel that understands the From eb3f8f5e22cdee8138e654963e371ad337830efb Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 19 Jul 2005 14:00:13 -0700 Subject: [PATCH 24/38] [NET]: BRIDGE_EBT_ARPREPLY must depend on INET BRIDGE_EBT_ARPREPLY=y and INET=n results in the following compile error: net/built-in.o: In function `ebt_target_reply': ebt_arpreply.c:(.text+0x68fb9): undefined reference to `arp_send' make: *** [.tmp_vmlinux1] Error 1 Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- net/bridge/netfilter/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/bridge/netfilter/Kconfig b/net/bridge/netfilter/Kconfig index 68ccef507b49..c70b3be23026 100644 --- a/net/bridge/netfilter/Kconfig +++ b/net/bridge/netfilter/Kconfig @@ -138,7 +138,7 @@ config BRIDGE_EBT_VLAN # config BRIDGE_EBT_ARPREPLY tristate "ebt: arp reply target support" - depends on BRIDGE_NF_EBTABLES + depends on BRIDGE_NF_EBTABLES && INET help This option adds the arp reply target, which allows automatically sending arp replies to arp requests. From 23a534e7b1ad2650002bbc236493791ac23440ee Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 19 Jul 2005 14:00:53 -0700 Subject: [PATCH 25/38] [NET]: NETCONSOLE must depend on INET NETCONSOLE=y and INET=n results in the following compile error: net/built-in.o: In function `netpoll_parse_options': : undefined reference to `in_aton' net/built-in.o: In function `netpoll_parse_options': : undefined reference to `in_aton' Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- drivers/net/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index f827f0a1e1a6..534b598866b3 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -2544,7 +2544,7 @@ config SHAPER config NETCONSOLE tristate "Network console logging support (EXPERIMENTAL)" - depends on NETDEVICES && EXPERIMENTAL + depends on NETDEVICES && INET && EXPERIMENTAL ---help--- If you want to log kernel messages over the network, enable this. See for details. From c877efb207bf4629cfa97ac13412f7392a873485 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 19 Jul 2005 14:01:51 -0700 Subject: [PATCH 26/38] [IPV4]: Fix up lots of little whitespace indentation stuff in fib_trie. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/ipv4/af_inet.c | 2 +- net/ipv4/fib_trie.c | 772 ++++++++++++++++++++++---------------------- 2 files changed, 388 insertions(+), 386 deletions(-) diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index ef7468376ae6..163ae4068b5f 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1157,7 +1157,7 @@ static int __init ipv4_proc_init(void) #ifdef CONFIG_IP_FIB_TRIE if (fib_stat_proc_init()) goto out_fib_stat; - #endif +#endif if (ip_misc_proc_init()) goto out_misc; out: diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 4be234c7d8c3..a701405fab0b 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -90,14 +90,14 @@ typedef unsigned int t_key; #define T_LEAF 1 #define NODE_TYPE_MASK 0x1UL #define NODE_PARENT(_node) \ -((struct tnode *)((_node)->_parent & ~NODE_TYPE_MASK)) + ((struct tnode *)((_node)->_parent & ~NODE_TYPE_MASK)) #define NODE_SET_PARENT(_node, _ptr) \ -((_node)->_parent = (((unsigned long)(_ptr)) | \ + ((_node)->_parent = (((unsigned long)(_ptr)) | \ ((_node)->_parent & NODE_TYPE_MASK))) #define NODE_INIT_PARENT(_node, _type) \ -((_node)->_parent = (_type)) + ((_node)->_parent = (_type)) #define NODE_TYPE(_node) \ -((_node)->_parent & NODE_TYPE_MASK) + ((_node)->_parent & NODE_TYPE_MASK) #define IS_TNODE(n) (!(n->_parent & T_LEAF)) #define IS_LEAF(n) (n->_parent & T_LEAF) @@ -147,7 +147,7 @@ struct trie_stat { unsigned int leaves; unsigned int nullpointers; unsigned int nodesizes[MAX_CHILDS]; -}; +}; struct trie { struct node *trie; @@ -185,9 +185,9 @@ static void trie_bug(char *err) BUG(); } -static inline struct node *tnode_get_child(struct tnode *tn, int i) +static inline struct node *tnode_get_child(struct tnode *tn, int i) { - if (i >= 1<bits) + if (i >= 1<bits) trie_bug("tnode_get_child"); return tn->child[i]; @@ -202,7 +202,7 @@ static inline int tnode_child_length(struct tnode *tn) _________________________________________________________________ | i | i | i | i | i | i | i | N | N | N | S | S | S | S | S | C | ---------------------------------------------------------------- - 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 + 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 _________________________________________________________________ | C | C | C | u | u | u | u | u | u | u | u | u | u | u | u | u | @@ -226,25 +226,25 @@ static inline t_key tkey_extract_bits(t_key a, int offset, int bits) static inline int tkey_equals(t_key a, t_key b) { - return a == b; + return a == b; } static inline int tkey_sub_equals(t_key a, int offset, int bits, t_key b) { - if (bits == 0 || offset >= KEYLENGTH) - return 1; + if (bits == 0 || offset >= KEYLENGTH) + return 1; bits = bits > KEYLENGTH ? KEYLENGTH : bits; return ((a ^ b) << offset) >> (KEYLENGTH - bits) == 0; -} +} static inline int tkey_mismatch(t_key a, int offset, t_key b) { t_key diff = a ^ b; int i = offset; - if(!diff) - return 0; - while((diff << i) >> (KEYLENGTH-1) == 0) + if (!diff) + return 0; + while ((diff << i) >> (KEYLENGTH-1) == 0) i++; return i; } @@ -314,6 +314,7 @@ static void fn_free_alias(struct fib_alias *fa) The bits from (n->pos) to (n->pos + n->bits - 1) - "C" - are the index into n's child array, and will of course be different for each child. + The rest of the bits, from (n->pos + n->bits) onward, are completely unknown at this point. @@ -321,7 +322,7 @@ static void fn_free_alias(struct fib_alias *fa) static void check_tnode(struct tnode *tn) { - if(tn && tn->pos+tn->bits > 32) { + if (tn && tn->pos+tn->bits > 32) { printk("TNODE ERROR tn=%p, pos=%d, bits=%d\n", tn, tn->pos, tn->bits); } } @@ -332,7 +333,7 @@ static int inflate_threshold = 50; static struct leaf *leaf_new(void) { struct leaf *l = kmalloc(sizeof(struct leaf), GFP_KERNEL); - if(l) { + if (l) { NODE_INIT_PARENT(l, T_LEAF); INIT_HLIST_HEAD(&l->list); } @@ -342,7 +343,7 @@ static struct leaf *leaf_new(void) static struct leaf_info *leaf_info_new(int plen) { struct leaf_info *li = kmalloc(sizeof(struct leaf_info), GFP_KERNEL); - if(li) { + if (li) { li->plen = plen; INIT_LIST_HEAD(&li->falh); } @@ -365,7 +366,7 @@ static struct tnode *tnode_alloc(unsigned int size) return kmalloc(size, GFP_KERNEL); } else { return (struct tnode *) - __get_free_pages(GFP_KERNEL, get_order(size)); + __get_free_pages(GFP_KERNEL, get_order(size)); } } @@ -386,7 +387,7 @@ static struct tnode* tnode_new(t_key key, int pos, int bits) int sz = sizeof(struct tnode) + nchildren * sizeof(struct node *); struct tnode *tn = tnode_alloc(sz); - if(tn) { + if (tn) { memset(tn, 0, sz); NODE_INIT_PARENT(tn, T_TNODE); tn->pos = pos; @@ -395,7 +396,8 @@ static struct tnode* tnode_new(t_key key, int pos, int bits) tn->full_children = 0; tn->empty_children = 1< 0) + + if (trie_debug > 0) printk("AT %p s=%u %u\n", tn, (unsigned int) sizeof(struct tnode), (unsigned int) (sizeof(struct node) * 1< 0 ) + if (trie_debug > 0 ) printk("FL %p \n", tn); } - else if(IS_TNODE(tn)) { + else if (IS_TNODE(tn)) { __tnode_free(tn); - if(trie_debug > 0 ) + if (trie_debug > 0 ) printk("FT %p \n", tn); } else { @@ -428,58 +430,58 @@ static void tnode_free(struct tnode *tn) static inline int tnode_full(struct tnode *tn, struct node *n) { - if(n == NULL || IS_LEAF(n)) + if (n == NULL || IS_LEAF(n)) return 0; return ((struct tnode *) n)->pos == tn->pos + tn->bits; } -static inline void put_child(struct trie *t, struct tnode *tn, int i, struct node *n) +static inline void put_child(struct trie *t, struct tnode *tn, int i, struct node *n) { tnode_put_child_reorg(tn, i, n, -1); } - /* + /* * Add a child at position i overwriting the old value. * Update the value of full_children and empty_children. */ -static void tnode_put_child_reorg(struct tnode *tn, int i, struct node *n, int wasfull) +static void tnode_put_child_reorg(struct tnode *tn, int i, struct node *n, int wasfull) { struct node *chi; int isfull; - if(i >= 1<bits) { + if (i >= 1<bits) { printk("bits=%d, i=%d\n", tn->bits, i); trie_bug("tnode_put_child_reorg bits"); } write_lock_bh(&fib_lock); - chi = tn->child[i]; + chi = tn->child[i]; /* update emptyChildren */ if (n == NULL && chi != NULL) tn->empty_children++; else if (n != NULL && chi == NULL) tn->empty_children--; - + /* update fullChildren */ if (wasfull == -1) wasfull = tnode_full(tn, chi); isfull = tnode_full(tn, n); - if (wasfull && !isfull) + if (wasfull && !isfull) tn->full_children--; - - else if (!wasfull && isfull) + + else if (!wasfull && isfull) tn->full_children++; - if(n) - NODE_SET_PARENT(n, tn); + if (n) + NODE_SET_PARENT(n, tn); tn->child[i] = n; write_unlock_bh(&fib_lock); } -static struct node *resize(struct trie *t, struct tnode *tn) +static struct node *resize(struct trie *t, struct tnode *tn) { int i; int err = 0; @@ -487,8 +489,8 @@ static struct node *resize(struct trie *t, struct tnode *tn) if (!tn) return NULL; - if(trie_debug) - printk("In tnode_resize %p inflate_threshold=%d threshold=%d\n", + if (trie_debug) + printk("In tnode_resize %p inflate_threshold=%d threshold=%d\n", tn, inflate_threshold, halve_threshold); /* No children */ @@ -505,7 +507,7 @@ static struct node *resize(struct trie *t, struct tnode *tn) /* compress one level */ struct node *n = tn->child[i]; - if(n) + if (n) NODE_INIT_PARENT(n, NODE_TYPE(n)); write_unlock_bh(&fib_lock); @@ -514,72 +516,72 @@ static struct node *resize(struct trie *t, struct tnode *tn) } write_unlock_bh(&fib_lock); } - /* + /* * Double as long as the resulting node has a number of * nonempty nodes that are above the threshold. */ /* - * From "Implementing a dynamic compressed trie" by Stefan Nilsson of - * the Helsinki University of Technology and Matti Tikkanen of Nokia + * From "Implementing a dynamic compressed trie" by Stefan Nilsson of + * the Helsinki University of Technology and Matti Tikkanen of Nokia * Telecommunications, page 6: - * "A node is doubled if the ratio of non-empty children to all + * "A node is doubled if the ratio of non-empty children to all * children in the *doubled* node is at least 'high'." * - * 'high' in this instance is the variable 'inflate_threshold'. It - * is expressed as a percentage, so we multiply it with - * tnode_child_length() and instead of multiplying by 2 (since the - * child array will be doubled by inflate()) and multiplying - * the left-hand side by 100 (to handle the percentage thing) we + * 'high' in this instance is the variable 'inflate_threshold'. It + * is expressed as a percentage, so we multiply it with + * tnode_child_length() and instead of multiplying by 2 (since the + * child array will be doubled by inflate()) and multiplying + * the left-hand side by 100 (to handle the percentage thing) we * multiply the left-hand side by 50. - * - * The left-hand side may look a bit weird: tnode_child_length(tn) - * - tn->empty_children is of course the number of non-null children - * in the current node. tn->full_children is the number of "full" + * + * The left-hand side may look a bit weird: tnode_child_length(tn) + * - tn->empty_children is of course the number of non-null children + * in the current node. tn->full_children is the number of "full" * children, that is non-null tnodes with a skip value of 0. - * All of those will be doubled in the resulting inflated tnode, so + * All of those will be doubled in the resulting inflated tnode, so * we just count them one extra time here. - * + * * A clearer way to write this would be: - * + * * to_be_doubled = tn->full_children; - * not_to_be_doubled = tnode_child_length(tn) - tn->empty_children - + * not_to_be_doubled = tnode_child_length(tn) - tn->empty_children - * tn->full_children; * * new_child_length = tnode_child_length(tn) * 2; * - * new_fill_factor = 100 * (not_to_be_doubled + 2*to_be_doubled) / + * new_fill_factor = 100 * (not_to_be_doubled + 2*to_be_doubled) / * new_child_length; * if (new_fill_factor >= inflate_threshold) - * - * ...and so on, tho it would mess up the while() loop. - * + * + * ...and so on, tho it would mess up the while () loop. + * * anyway, * 100 * (not_to_be_doubled + 2*to_be_doubled) / new_child_length >= * inflate_threshold - * + * * avoid a division: * 100 * (not_to_be_doubled + 2*to_be_doubled) >= * inflate_threshold * new_child_length - * + * * expand not_to_be_doubled and to_be_doubled, and shorten: - * 100 * (tnode_child_length(tn) - tn->empty_children + + * 100 * (tnode_child_length(tn) - tn->empty_children + * tn->full_children ) >= inflate_threshold * new_child_length - * + * * expand new_child_length: - * 100 * (tnode_child_length(tn) - tn->empty_children + + * 100 * (tnode_child_length(tn) - tn->empty_children + * tn->full_children ) >= * inflate_threshold * tnode_child_length(tn) * 2 - * + * * shorten again: - * 50 * (tn->full_children + tnode_child_length(tn) - - * tn->empty_children ) >= inflate_threshold * + * 50 * (tn->full_children + tnode_child_length(tn) - + * tn->empty_children ) >= inflate_threshold * * tnode_child_length(tn) - * + * */ check_tnode(tn); - + err = 0; while ((tn->full_children > 0 && 50 * (tn->full_children + tnode_child_length(tn) - tn->empty_children) >= @@ -587,7 +589,7 @@ static struct node *resize(struct trie *t, struct tnode *tn) tn = inflate(t, tn, &err); - if(err) { + if (err) { #ifdef CONFIG_IP_FIB_TRIE_STATS t->stats.resize_node_skipped++; #endif @@ -609,7 +611,7 @@ static struct node *resize(struct trie *t, struct tnode *tn) tn = halve(t, tn, &err); - if(err) { + if (err) { #ifdef CONFIG_IP_FIB_TRIE_STATS t->stats.resize_node_skipped++; #endif @@ -617,18 +619,18 @@ static struct node *resize(struct trie *t, struct tnode *tn) } } - + /* Only one child remains */ if (tn->empty_children == tnode_child_length(tn) - 1) for (i = 0; i < tnode_child_length(tn); i++) { - + write_lock_bh(&fib_lock); if (tn->child[i] != NULL) { /* compress one level */ struct node *n = tn->child[i]; - if(n) + if (n) NODE_INIT_PARENT(n, NODE_TYPE(n)); write_unlock_bh(&fib_lock); @@ -648,7 +650,7 @@ static struct tnode *inflate(struct trie *t, struct tnode *tn, int *err) int olen = tnode_child_length(tn); int i; - if(trie_debug) + if (trie_debug) printk("In inflate\n"); tn = tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits + 1); @@ -659,12 +661,12 @@ static struct tnode *inflate(struct trie *t, struct tnode *tn, int *err) } /* - * Preallocate and store tnodes before the actual work so we - * don't get into an inconsistent state if memory allocation - * fails. In case of failure we return the oldnode and inflate + * Preallocate and store tnodes before the actual work so we + * don't get into an inconsistent state if memory allocation + * fails. In case of failure we return the oldnode and inflate * of tnode is ignored. */ - + for(i = 0; i < olen; i++) { struct tnode *inode = (struct tnode *) tnode_get_child(oldtnode, i); @@ -675,20 +677,20 @@ static struct tnode *inflate(struct trie *t, struct tnode *tn, int *err) struct tnode *left, *right; t_key m = TKEY_GET_MASK(inode->pos, 1); - + left = tnode_new(inode->key&(~m), inode->pos + 1, inode->bits - 1); - if(!left) { - *err = -ENOMEM; + if (!left) { + *err = -ENOMEM; break; } - + right = tnode_new(inode->key|m, inode->pos + 1, inode->bits - 1); - if(!right) { - *err = -ENOMEM; + if (!right) { + *err = -ENOMEM; break; } @@ -697,32 +699,32 @@ static struct tnode *inflate(struct trie *t, struct tnode *tn, int *err) } } - if(*err) { + if (*err) { int size = tnode_child_length(tn); int j; - for(j = 0; j < size; j++) - if( tn->child[j]) + for(j = 0; j < size; j++) + if (tn->child[j]) tnode_free((struct tnode *)tn->child[j]); tnode_free(tn); - + *err = -ENOMEM; return oldtnode; } for(i = 0; i < olen; i++) { struct node *node = tnode_get_child(oldtnode, i); - + /* An empty child */ if (node == NULL) continue; /* A leaf or an internal node with skipped bits */ - if(IS_LEAF(node) || ((struct tnode *) node)->pos > + if (IS_LEAF(node) || ((struct tnode *) node)->pos > tn->pos + tn->bits - 1) { - if(tkey_extract_bits(node->key, oldtnode->pos + oldtnode->bits, + if (tkey_extract_bits(node->key, oldtnode->pos + oldtnode->bits, 1) == 0) put_child(t, tn, 2*i, node); else @@ -745,37 +747,37 @@ static struct tnode *inflate(struct trie *t, struct tnode *tn, int *err) struct tnode *left, *right; int size, j; - /* We will replace this node 'inode' with two new - * ones, 'left' and 'right', each with half of the - * original children. The two new nodes will have - * a position one bit further down the key and this - * means that the "significant" part of their keys - * (see the discussion near the top of this file) - * will differ by one bit, which will be "0" in - * left's key and "1" in right's key. Since we are - * moving the key position by one step, the bit that - * we are moving away from - the bit at position - * (inode->pos) - is the one that will differ between + /* We will replace this node 'inode' with two new + * ones, 'left' and 'right', each with half of the + * original children. The two new nodes will have + * a position one bit further down the key and this + * means that the "significant" part of their keys + * (see the discussion near the top of this file) + * will differ by one bit, which will be "0" in + * left's key and "1" in right's key. Since we are + * moving the key position by one step, the bit that + * we are moving away from - the bit at position + * (inode->pos) - is the one that will differ between * left and right. So... we synthesize that bit in the * two new keys. - * The mask 'm' below will be a single "one" bit at + * The mask 'm' below will be a single "one" bit at * the position (inode->pos) */ - /* Use the old key, but set the new significant - * bit to zero. + /* Use the old key, but set the new significant + * bit to zero. */ left = (struct tnode *) tnode_get_child(tn, 2*i); put_child(t, tn, 2*i, NULL); - if(!left) + if (!left) BUG(); right = (struct tnode *) tnode_get_child(tn, 2*i+1); put_child(t, tn, 2*i+1, NULL); - if(!right) + if (!right) BUG(); size = tnode_child_length(left); @@ -800,9 +802,9 @@ static struct tnode *halve(struct trie *t, struct tnode *tn, int *err) int i; int olen = tnode_child_length(tn); - if(trie_debug) printk("In halve\n"); - - tn=tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits - 1); + if (trie_debug) printk("In halve\n"); + + tn = tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits - 1); if (!tn) { *err = -ENOMEM; @@ -810,39 +812,39 @@ static struct tnode *halve(struct trie *t, struct tnode *tn, int *err) } /* - * Preallocate and store tnodes before the actual work so we - * don't get into an inconsistent state if memory allocation - * fails. In case of failure we return the oldnode and halve + * Preallocate and store tnodes before the actual work so we + * don't get into an inconsistent state if memory allocation + * fails. In case of failure we return the oldnode and halve * of tnode is ignored. */ for(i = 0; i < olen; i += 2) { left = tnode_get_child(oldtnode, i); right = tnode_get_child(oldtnode, i+1); - + /* Two nonempty children */ - if( left && right) { + if (left && right) { struct tnode *newBinNode = tnode_new(left->key, tn->pos + tn->bits, 1); - if(!newBinNode) { - *err = -ENOMEM; + if (!newBinNode) { + *err = -ENOMEM; break; } put_child(t, tn, i/2, (struct node *)newBinNode); } } - if(*err) { + if (*err) { int size = tnode_child_length(tn); int j; - for(j = 0; j < size; j++) - if( tn->child[j]) + for(j = 0; j < size; j++) + if (tn->child[j]) tnode_free((struct tnode *)tn->child[j]); tnode_free(tn); - + *err = -ENOMEM; return oldtnode; } @@ -850,7 +852,7 @@ static struct tnode *halve(struct trie *t, struct tnode *tn, int *err) for(i = 0; i < olen; i += 2) { left = tnode_get_child(oldtnode, i); right = tnode_get_child(oldtnode, i+1); - + /* At least one of the children is empty */ if (left == NULL) { if (right == NULL) /* Both are empty */ @@ -858,14 +860,14 @@ static struct tnode *halve(struct trie *t, struct tnode *tn, int *err) put_child(t, tn, i/2, right); } else if (right == NULL) put_child(t, tn, i/2, left); - + /* Two nonempty children */ else { struct tnode *newBinNode = (struct tnode *) tnode_get_child(tn, i/2); put_child(t, tn, i/2, NULL); - if(!newBinNode) + if (!newBinNode) BUG(); put_child(t, newBinNode, 0, left); @@ -879,7 +881,7 @@ static struct tnode *halve(struct trie *t, struct tnode *tn, int *err) static void *trie_init(struct trie *t) { - if(t) { + if (t) { t->size = 0; t->trie = NULL; t->revision = 0; @@ -896,8 +898,7 @@ static struct leaf_info *find_leaf_info(struct hlist_head *head, int plen) struct leaf_info *li; hlist_for_each_entry(li, node, head, hlist) { - - if ( li->plen == plen ) + if (li->plen == plen) return li; } return NULL; @@ -905,35 +906,35 @@ static struct leaf_info *find_leaf_info(struct hlist_head *head, int plen) static inline struct list_head * get_fa_head(struct leaf *l, int plen) { - struct list_head *fa_head=NULL; + struct list_head *fa_head = NULL; struct leaf_info *li = find_leaf_info(&l->list, plen); - - if(li) + + if (li) fa_head = &li->falh; - + return fa_head; } static void insert_leaf_info(struct hlist_head *head, struct leaf_info *new) { - struct leaf_info *li=NULL, *last=NULL; + struct leaf_info *li = NULL, *last = NULL; struct hlist_node *node, *tmp; write_lock_bh(&fib_lock); - - if(hlist_empty(head)) + + if (hlist_empty(head)) hlist_add_head(&new->hlist, head); else { hlist_for_each_entry_safe(li, node, tmp, head, hlist) { - - if (new->plen > li->plen) + + if (new->plen > li->plen) break; - + last = li; } - if(last) + if (last) hlist_add_after(&last->hlist, &new->hlist); - else + else hlist_add_before(&new->hlist, &li->hlist); } write_unlock_bh(&fib_lock); @@ -947,14 +948,14 @@ fib_find_node(struct trie *t, u32 key) struct node *n; pos = 0; - n=t->trie; + n = t->trie; while (n != NULL && NODE_TYPE(n) == T_TNODE) { tn = (struct tnode *) n; - + check_tnode(tn); - - if(tkey_sub_equals(tn->key, pos, tn->pos-pos, key)) { + + if (tkey_sub_equals(tn->key, pos, tn->pos-pos, key)) { pos=tn->pos + tn->bits; n = tnode_get_child(tn, tkey_extract_bits(key, tn->pos, tn->bits)); } @@ -977,23 +978,23 @@ static struct node *trie_rebalance(struct trie *t, struct tnode *tn) t_key cindex, key; struct tnode *tp = NULL; - if(!tn) + if (!tn) BUG(); - + key = tn->key; i = 0; while (tn != NULL && NODE_PARENT(tn) != NULL) { - if( i > 10 ) { + if (i > 10) { printk("Rebalance tn=%p \n", tn); - if(tn) printk("tn->parent=%p \n", NODE_PARENT(tn)); - + if (tn) printk("tn->parent=%p \n", NODE_PARENT(tn)); + printk("Rebalance tp=%p \n", tp); - if(tp) printk("tp->parent=%p \n", NODE_PARENT(tp)); + if (tp) printk("tp->parent=%p \n", NODE_PARENT(tp)); } - if( i > 12 ) BUG(); + if (i > 12) BUG(); i++; tp = NODE_PARENT(tn); @@ -1001,14 +1002,14 @@ static struct node *trie_rebalance(struct trie *t, struct tnode *tn) wasfull = tnode_full(tp, tnode_get_child(tp, cindex)); tn = (struct tnode *) resize (t, (struct tnode *)tn); tnode_put_child_reorg((struct tnode *)tp, cindex,(struct node*)tn, wasfull); - - if(!NODE_PARENT(tn)) + + if (!NODE_PARENT(tn)) break; tn = NODE_PARENT(tn); } /* Handle last (top) tnode */ - if (IS_TNODE(tn)) + if (IS_TNODE(tn)) tn = (struct tnode*) resize(t, (struct tnode *)tn); return (struct node*) tn; @@ -1022,42 +1023,42 @@ fib_insert_node(struct trie *t, int *err, u32 key, int plen) struct node *n; struct leaf *l; int missbit; - struct list_head *fa_head=NULL; + struct list_head *fa_head = NULL; struct leaf_info *li; t_key cindex; pos = 0; - n=t->trie; + n = t->trie; - /* If we point to NULL, stop. Either the tree is empty and we should - * just put a new leaf in if, or we have reached an empty child slot, + /* If we point to NULL, stop. Either the tree is empty and we should + * just put a new leaf in if, or we have reached an empty child slot, * and we should just put our new leaf in that. - * If we point to a T_TNODE, check if it matches our key. Note that - * a T_TNODE might be skipping any number of bits - its 'pos' need + * If we point to a T_TNODE, check if it matches our key. Note that + * a T_TNODE might be skipping any number of bits - its 'pos' need * not be the parent's 'pos'+'bits'! * - * If it does match the current key, get pos/bits from it, extract + * If it does match the current key, get pos/bits from it, extract * the index from our key, push the T_TNODE and walk the tree. * * If it doesn't, we have to replace it with a new T_TNODE. * - * If we point to a T_LEAF, it might or might not have the same key - * as we do. If it does, just change the value, update the T_LEAF's - * value, and return it. + * If we point to a T_LEAF, it might or might not have the same key + * as we do. If it does, just change the value, update the T_LEAF's + * value, and return it. * If it doesn't, we need to replace it with a T_TNODE. */ while (n != NULL && NODE_TYPE(n) == T_TNODE) { tn = (struct tnode *) n; - - check_tnode(tn); - if(tkey_sub_equals(tn->key, pos, tn->pos-pos, key)) { + check_tnode(tn); + + if (tkey_sub_equals(tn->key, pos, tn->pos-pos, key)) { tp = tn; pos=tn->pos + tn->bits; n = tnode_get_child(tn, tkey_extract_bits(key, tn->pos, tn->bits)); - if(n && NODE_PARENT(n) != tn) { + if (n && NODE_PARENT(n) != tn) { printk("BUG tn=%p, n->parent=%p\n", tn, NODE_PARENT(n)); BUG(); } @@ -1069,21 +1070,21 @@ fib_insert_node(struct trie *t, int *err, u32 key, int plen) /* * n ----> NULL, LEAF or TNODE * - * tp is n's (parent) ----> NULL or TNODE + * tp is n's (parent) ----> NULL or TNODE */ - if(tp && IS_LEAF(tp)) + if (tp && IS_LEAF(tp)) BUG(); /* Case 1: n is a leaf. Compare prefixes */ - if (n != NULL && IS_LEAF(n) && tkey_equals(key, n->key)) { + if (n != NULL && IS_LEAF(n) && tkey_equals(key, n->key)) { struct leaf *l = ( struct leaf *) n; - + li = leaf_info_new(plen); - - if(! li) { + + if (!li) { *err = -ENOMEM; goto err; } @@ -1095,7 +1096,7 @@ fib_insert_node(struct trie *t, int *err, u32 key, int plen) t->size++; l = leaf_new(); - if(! l) { + if (!l) { *err = -ENOMEM; goto err; } @@ -1103,7 +1104,7 @@ fib_insert_node(struct trie *t, int *err, u32 key, int plen) l->key = key; li = leaf_info_new(plen); - if(! li) { + if (!li) { tnode_free((struct tnode *) l); *err = -ENOMEM; goto err; @@ -1116,8 +1117,8 @@ fib_insert_node(struct trie *t, int *err, u32 key, int plen) if (t->trie && n == NULL) { NODE_SET_PARENT(l, tp); - - if (!tp) + + if (!tp) BUG(); else { @@ -1127,8 +1128,8 @@ fib_insert_node(struct trie *t, int *err, u32 key, int plen) } /* Case 3: n is a LEAF or a TNODE and the key doesn't match. */ else { - /* - * Add a new tnode here + /* + * Add a new tnode here * first tnode need some special handling */ @@ -1136,39 +1137,39 @@ fib_insert_node(struct trie *t, int *err, u32 key, int plen) pos=tp->pos+tp->bits; else pos=0; - if(n) { + if (n) { newpos = tkey_mismatch(key, pos, n->key); tn = tnode_new(n->key, newpos, 1); } else { newpos = 0; - tn = tnode_new(key, newpos, 1); /* First tnode */ + tn = tnode_new(key, newpos, 1); /* First tnode */ } - if(!tn) { + if (!tn) { free_leaf_info(li); tnode_free((struct tnode *) l); *err = -ENOMEM; goto err; - } - + } + NODE_SET_PARENT(tn, tp); missbit=tkey_extract_bits(key, newpos, 1); put_child(t, tn, missbit, (struct node *)l); put_child(t, tn, 1-missbit, n); - if(tp) { + if (tp) { cindex = tkey_extract_bits(key, tp->pos, tp->bits); put_child(t, (struct tnode *)tp, cindex, (struct node *)tn); } - else { + else { t->trie = (struct node*) tn; /* First tnode */ tp = tn; } } - if(tp && tp->pos+tp->bits > 32) { - printk("ERROR tp=%p pos=%d, bits=%d, key=%0x plen=%d\n", + if (tp && tp->pos+tp->bits > 32) { + printk("ERROR tp=%p pos=%d, bits=%d, key=%0x plen=%d\n", tp, tp->pos, tp->bits, key, plen); } /* Rebalance the trie */ @@ -1185,7 +1186,7 @@ fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta, { struct trie *t = (struct trie *) tb->tb_data; struct fib_alias *fa, *new_fa; - struct list_head *fa_head=NULL; + struct list_head *fa_head = NULL; struct fib_info *fi; int plen = r->rtm_dst_len; int type = r->rtm_type; @@ -1198,17 +1199,17 @@ fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta, return -EINVAL; key = 0; - if (rta->rta_dst) + if (rta->rta_dst) memcpy(&key, rta->rta_dst, 4); key = ntohl(key); - if(trie_debug) + if (trie_debug) printk("Insert table=%d %08x/%d\n", tb->tb_id, key, plen); - mask = ntohl( inet_make_mask(plen) ); + mask = ntohl( inet_make_mask(plen) ); - if(key & ~mask) + if (key & ~mask) return -EINVAL; key = key & mask; @@ -1217,9 +1218,9 @@ fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta, goto err; l = fib_find_node(t, key); - fa = NULL; + fa = NULL; - if(l) { + if (l) { fa_head = get_fa_head(l, plen); fa = fib_find_alias(fa_head, tos, fi->fib_priority); } @@ -1298,16 +1299,16 @@ fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta, new_fa->fa_scope = r->rtm_scope; new_fa->fa_state = 0; #if 0 - new_fa->dst = NULL; + new_fa->dst = NULL; #endif /* * Insert new entry to the list. */ - if(!fa_head) { + if (!fa_head) { fa_head = fib_insert_node(t, &err, key, plen); err = 0; - if(err) + if (err) goto out_free_new_fa; } @@ -1327,11 +1328,11 @@ out_free_new_fa: kmem_cache_free(fn_alias_kmem, new_fa); out: fib_release_info(fi); -err:; +err:; return err; } -static inline int check_leaf(struct trie *t, struct leaf *l, t_key key, int *plen, const struct flowi *flp, +static inline int check_leaf(struct trie *t, struct leaf *l, t_key key, int *plen, const struct flowi *flp, struct fib_result *res, int *err) { int i; @@ -1339,12 +1340,12 @@ static inline int check_leaf(struct trie *t, struct leaf *l, t_key key, int *pl struct leaf_info *li; struct hlist_head *hhead = &l->list; struct hlist_node *node; - + hlist_for_each_entry(li, node, hhead, hlist) { i = li->plen; mask = ntohl(inet_make_mask(i)); - if (l->key != (key & mask)) + if (l->key != (key & mask)) continue; if (((*err) = fib_semantic_match(&li->falh, flp, res, l->key, mask, i)) == 0) { @@ -1376,7 +1377,7 @@ fn_trie_lookup(struct fib_table *tb, const struct flowi *flp, struct fib_result n = t->trie; read_lock(&fib_lock); - if(!n) + if (!n) goto failed; #ifdef CONFIG_IP_FIB_TRIE_STATS @@ -1385,19 +1386,19 @@ fn_trie_lookup(struct fib_table *tb, const struct flowi *flp, struct fib_result /* Just a leaf? */ if (IS_LEAF(n)) { - if( check_leaf(t, (struct leaf *)n, key, &plen, flp, res, &ret) ) + if (check_leaf(t, (struct leaf *)n, key, &plen, flp, res, &ret)) goto found; goto failed; } pn = (struct tnode *) n; chopped_off = 0; - + while (pn) { pos = pn->pos; bits = pn->bits; - if(!chopped_off) + if (!chopped_off) cindex = tkey_extract_bits(MASK_PFX(key, current_prefix_length), pos, bits); n = tnode_get_child(pn, cindex); @@ -1417,33 +1418,33 @@ fn_trie_lookup(struct fib_table *tb, const struct flowi *flp, struct fib_result int mp; /* - * It's a tnode, and we can do some extra checks here if we + * It's a tnode, and we can do some extra checks here if we * like, to avoid descending into a dead-end branch. - * This tnode is in the parent's child array at index - * key[p_pos..p_pos+p_bits] but potentially with some bits - * chopped off, so in reality the index may be just a + * This tnode is in the parent's child array at index + * key[p_pos..p_pos+p_bits] but potentially with some bits + * chopped off, so in reality the index may be just a * subprefix, padded with zero at the end. - * We can also take a look at any skipped bits in this - * tnode - everything up to p_pos is supposed to be ok, + * We can also take a look at any skipped bits in this + * tnode - everything up to p_pos is supposed to be ok, * and the non-chopped bits of the index (se previous - * paragraph) are also guaranteed ok, but the rest is + * paragraph) are also guaranteed ok, but the rest is * considered unknown. * * The skipped bits are key[pos+bits..cn->pos]. */ - - /* If current_prefix_length < pos+bits, we are already doing - * actual prefix matching, which means everything from - * pos+(bits-chopped_off) onward must be zero along some - * branch of this subtree - otherwise there is *no* valid + + /* If current_prefix_length < pos+bits, we are already doing + * actual prefix matching, which means everything from + * pos+(bits-chopped_off) onward must be zero along some + * branch of this subtree - otherwise there is *no* valid * prefix present. Here we can only check the skipped - * bits. Remember, since we have already indexed into the - * parent's child array, we know that the bits we chopped of + * bits. Remember, since we have already indexed into the + * parent's child array, we know that the bits we chopped of * *are* zero. */ /* NOTA BENE: CHECKING ONLY SKIPPED BITS FOR THE NEW NODE HERE */ - + if (current_prefix_length < pos+bits) { if (tkey_extract_bits(cn->key, current_prefix_length, cn->pos - current_prefix_length) != 0 || @@ -1452,13 +1453,13 @@ fn_trie_lookup(struct fib_table *tb, const struct flowi *flp, struct fib_result } /* - * If chopped_off=0, the index is fully validated and we - * only need to look at the skipped bits for this, the new, + * If chopped_off=0, the index is fully validated and we + * only need to look at the skipped bits for this, the new, * tnode. What we actually want to do is to find out if * these skipped bits match our key perfectly, or if we will - * have to count on finding a matching prefix further down, - * because if we do, we would like to have some way of - * verifying the existence of such a prefix at this point. + * have to count on finding a matching prefix further down, + * because if we do, we would like to have some way of + * verifying the existence of such a prefix at this point. */ /* The only thing we can do at this point is to verify that @@ -1470,22 +1471,22 @@ fn_trie_lookup(struct fib_table *tb, const struct flowi *flp, struct fib_result * new tnode's key. */ - /* Note: We aren't very concerned about the piece of the key - * that precede pn->pos+pn->bits, since these have already been - * checked. The bits after cn->pos aren't checked since these are - * by definition "unknown" at this point. Thus, what we want to - * see is if we are about to enter the "prefix matching" state, - * and in that case verify that the skipped bits that will prevail - * throughout this subtree are zero, as they have to be if we are + /* Note: We aren't very concerned about the piece of the key + * that precede pn->pos+pn->bits, since these have already been + * checked. The bits after cn->pos aren't checked since these are + * by definition "unknown" at this point. Thus, what we want to + * see is if we are about to enter the "prefix matching" state, + * and in that case verify that the skipped bits that will prevail + * throughout this subtree are zero, as they have to be if we are * to find a matching prefix. */ node_prefix = MASK_PFX(cn->key, cn->pos); - key_prefix = MASK_PFX(key, cn->pos); + key_prefix = MASK_PFX(key, cn->pos); pref_mismatch = key_prefix^node_prefix; mp = 0; - /* In short: If skipped bits in this node do not match the search + /* In short: If skipped bits in this node do not match the search * key, enter the "prefix matching" state.directly. */ if (pref_mismatch) { @@ -1494,7 +1495,7 @@ fn_trie_lookup(struct fib_table *tb, const struct flowi *flp, struct fib_result pref_mismatch = pref_mismatch <<1; } key_prefix = tkey_extract_bits(cn->key, mp, cn->pos-mp); - + if (key_prefix != 0) goto backtrace; @@ -1505,9 +1506,9 @@ fn_trie_lookup(struct fib_table *tb, const struct flowi *flp, struct fib_result pn = (struct tnode *)n; /* Descend */ chopped_off = 0; continue; - } - if (IS_LEAF(n)) { - if( check_leaf(t, (struct leaf *)n, key, &plen, flp, res, &ret)) + } + if (IS_LEAF(n)) { + if (check_leaf(t, (struct leaf *)n, key, &plen, flp, res, &ret)) goto found; } backtrace: @@ -1521,18 +1522,18 @@ backtrace: /* Decrease current_... with bits chopped off */ if (current_prefix_length > pn->pos + pn->bits - chopped_off) current_prefix_length = pn->pos + pn->bits - chopped_off; - + /* - * Either we do the actual chop off according or if we have + * Either we do the actual chop off according or if we have * chopped off all bits in this tnode walk up to our parent. */ - if(chopped_off <= pn->bits) + if (chopped_off <= pn->bits) cindex &= ~(1 << (chopped_off-1)); else { - if( NODE_PARENT(pn) == NULL) + if (NODE_PARENT(pn) == NULL) goto failed; - + /* Get Child's index */ cindex = tkey_extract_bits(pn->key, NODE_PARENT(pn)->pos, NODE_PARENT(pn)->bits); pn = NODE_PARENT(pn); @@ -1542,10 +1543,10 @@ backtrace: t->stats.backtrack++; #endif goto backtrace; - } + } } failed: - ret = 1; + ret = 1; found: read_unlock(&fib_lock); return ret; @@ -1558,11 +1559,11 @@ static int trie_leaf_remove(struct trie *t, t_key key) struct node *n = t->trie; struct leaf *l; - if(trie_debug) + if (trie_debug) printk("entering trie_leaf_remove(%p)\n", n); /* Note that in the case skipped bits, those bits are *not* checked! - * When we finish this, we will have NULL or a T_LEAF, and the + * When we finish this, we will have NULL or a T_LEAF, and the * T_LEAF may or may not match our key. */ @@ -1571,19 +1572,19 @@ static int trie_leaf_remove(struct trie *t, t_key key) check_tnode(tn); n = tnode_get_child(tn ,tkey_extract_bits(key, tn->pos, tn->bits)); - if(n && NODE_PARENT(n) != tn) { + if (n && NODE_PARENT(n) != tn) { printk("BUG tn=%p, n->parent=%p\n", tn, NODE_PARENT(n)); BUG(); } } l = (struct leaf *) n; - if(!n || !tkey_equals(l->key, key)) + if (!n || !tkey_equals(l->key, key)) return 0; - - /* - * Key found. - * Remove the leaf and rebalance the tree + + /* + * Key found. + * Remove the leaf and rebalance the tree */ t->revision++; @@ -1592,7 +1593,7 @@ static int trie_leaf_remove(struct trie *t, t_key key) tp = NODE_PARENT(n); tnode_free((struct tnode *) n); - if(tp) { + if (tp) { cindex = tkey_extract_bits(key, tp->pos, tp->bits); put_child(t, (struct tnode *)tp, cindex, NULL); t->trie = trie_rebalance(t, tp); @@ -1615,23 +1616,23 @@ fn_trie_delete(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta, struct list_head *fa_head; struct leaf *l; - if (plen > 32) + if (plen > 32) return -EINVAL; key = 0; - if (rta->rta_dst) + if (rta->rta_dst) memcpy(&key, rta->rta_dst, 4); key = ntohl(key); - mask = ntohl( inet_make_mask(plen) ); + mask = ntohl( inet_make_mask(plen) ); - if(key & ~mask) + if (key & ~mask) return -EINVAL; key = key & mask; l = fib_find_node(t, key); - if(!l) + if (!l) return -ESRCH; fa_head = get_fa_head(l, plen); @@ -1677,16 +1678,16 @@ fn_trie_delete(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta, list_del(&fa->fa_list); - if(list_empty(fa_head)) { + if (list_empty(fa_head)) { hlist_del(&li->hlist); kill_li = 1; } write_unlock_bh(&fib_lock); - - if(kill_li) + + if (kill_li) free_leaf_info(li); - if(hlist_empty(&l->list)) + if (hlist_empty(&l->list)) trie_leaf_remove(t, key); if (fa->fa_state & FA_S_ACCESSED) @@ -1705,12 +1706,12 @@ static int trie_flush_list(struct trie *t, struct list_head *head) list_for_each_entry_safe(fa, fa_node, head, fa_list) { struct fib_info *fi = fa->fa_info; - + if (fi && (fi->fib_flags&RTNH_F_DEAD)) { - write_lock_bh(&fib_lock); + write_lock_bh(&fib_lock); list_del(&fa->fa_list); - write_unlock_bh(&fib_lock); + write_unlock_bh(&fib_lock); fn_free_alias(fa); found++; @@ -1727,14 +1728,14 @@ static int trie_flush_leaf(struct trie *t, struct leaf *l) struct leaf_info *li = NULL; hlist_for_each_entry_safe(li, node, tmp, lih, hlist) { - + found += trie_flush_list(t, &li->falh); if (list_empty(&li->falh)) { - write_lock_bh(&fib_lock); + write_lock_bh(&fib_lock); hlist_del(&li->hlist); - write_unlock_bh(&fib_lock); + write_unlock_bh(&fib_lock); free_leaf_info(li); } @@ -1748,8 +1749,8 @@ static struct leaf *nextleaf(struct trie *t, struct leaf *thisleaf) struct tnode *p; int idx; - if(c == NULL) { - if(t->trie == NULL) + if (c == NULL) { + if (t->trie == NULL) return NULL; if (IS_LEAF(t->trie)) /* trie w. just a leaf */ @@ -1757,33 +1758,34 @@ static struct leaf *nextleaf(struct trie *t, struct leaf *thisleaf) p = (struct tnode*) t->trie; /* Start */ } - else + else p = (struct tnode *) NODE_PARENT(c); + while (p) { int pos, last; /* Find the next child of the parent */ - if(c) - pos = 1 + tkey_extract_bits(c->key, p->pos, p->bits); - else + if (c) + pos = 1 + tkey_extract_bits(c->key, p->pos, p->bits); + else pos = 0; last = 1 << p->bits; for(idx = pos; idx < last ; idx++) { - if( p->child[idx]) { + if (p->child[idx]) { /* Decend if tnode */ while (IS_TNODE(p->child[idx])) { p = (struct tnode*) p->child[idx]; idx = 0; - + /* Rightmost non-NULL branch */ - if( p && IS_TNODE(p) ) - while ( p->child[idx] == NULL && idx < (1 << p->bits) ) idx++; + if (p && IS_TNODE(p)) + while (p->child[idx] == NULL && idx < (1 << p->bits)) idx++; /* Done with this tnode? */ - if( idx >= (1 << p->bits) || p->child[idx] == NULL ) + if (idx >= (1 << p->bits) || p->child[idx] == NULL ) goto up; } return (struct leaf*) p->child[idx]; @@ -1816,7 +1818,7 @@ static int fn_trie_flush(struct fib_table *tb) if (ll && hlist_empty(&ll->list)) trie_leaf_remove(t, ll->key); - if(trie_debug) + if (trie_debug) printk("trie_flush found=%d\n", found); return found; } @@ -1839,32 +1841,32 @@ fn_trie_select_default(struct fib_table *tb, const struct flowi *flp, struct fib order = -1; read_lock(&fib_lock); - + l = fib_find_node(t, 0); - if(!l) + if (!l) goto out; fa_head = get_fa_head(l, 0); - if(!fa_head) + if (!fa_head) goto out; - if (list_empty(fa_head)) + if (list_empty(fa_head)) goto out; list_for_each_entry(fa, fa_head, fa_list) { struct fib_info *next_fi = fa->fa_info; - + if (fa->fa_scope != res->scope || fa->fa_type != RTN_UNICAST) continue; - + if (next_fi->fib_priority > res->fi->fib_priority) break; if (!next_fi->fib_nh[0].nh_gw || next_fi->fib_nh[0].nh_scope != RT_SCOPE_LINK) continue; fa->fa_state |= FA_S_ACCESSED; - + if (fi == NULL) { if (next_fi != res->fi) break; @@ -1902,10 +1904,10 @@ fn_trie_select_default(struct fib_table *tb, const struct flowi *flp, struct fib } trie_last_dflt = last_idx; out:; - read_unlock(&fib_lock); + read_unlock(&fib_lock); } -static int fn_trie_dump_fa(t_key key, int plen, struct list_head *fah, struct fib_table *tb, +static int fn_trie_dump_fa(t_key key, int plen, struct list_head *fah, struct fib_table *tb, struct sk_buff *skb, struct netlink_callback *cb) { int i, s_i; @@ -1951,7 +1953,7 @@ static int fn_trie_dump_fa(t_key key, int plen, struct list_head *fah, struct fi return skb->len; } -static int fn_trie_dump_plen(struct trie *t, int plen, struct fib_table *tb, struct sk_buff *skb, +static int fn_trie_dump_plen(struct trie *t, int plen, struct fib_table *tb, struct sk_buff *skb, struct netlink_callback *cb) { int h, s_h; @@ -1968,11 +1970,11 @@ static int fn_trie_dump_plen(struct trie *t, int plen, struct fib_table *tb, str sizeof(cb->args) - 3*sizeof(cb->args[0])); fa_head = get_fa_head(l, plen); - - if(!fa_head) + + if (!fa_head) continue; - if(list_empty(fa_head)) + if (list_empty(fa_head)) continue; if (fn_trie_dump_fa(l->key, plen, fa_head, tb, skb, cb)<0) { @@ -2048,10 +2050,10 @@ struct fib_table * __init fib_hash_init(int id) trie_init(t); - if (id == RT_TABLE_LOCAL) - trie_local=t; - else if (id == RT_TABLE_MAIN) - trie_main=t; + if (id == RT_TABLE_LOCAL) + trie_local = t; + else if (id == RT_TABLE_MAIN) + trie_main = t; if (id == RT_TABLE_LOCAL) printk("IPv4 FIB: Using LC-trie version %s\n", VERSION); @@ -2072,7 +2074,7 @@ static void printbin_seq(struct seq_file *seq, unsigned int v, int bits) seq_printf(seq, "%s", (v & (1<key >> 24, (n->key >> 16) % 256, (n->key >> 8) % 256, n->key % 256); else { - int plen=((struct tnode *)n)->pos; + int plen = ((struct tnode *)n)->pos; t_key prf=MASK_PFX(n->key, plen); - seq_printf(seq, "key=%d.%d.%d.%d/%d\n", + seq_printf(seq, "key=%d.%d.%d.%d/%d\n", prf >> 24, (prf >> 16) % 256, (prf >> 8) % 256, prf % 256, plen); } if (IS_LEAF(n)) { @@ -2103,14 +2105,14 @@ static void printnode_seq(struct seq_file *seq, int indent, struct node *n, struct fib_alias *fa; int i; for (i=32; i>=0; i--) - if(find_leaf_info(&l->list, i)) { - + if (find_leaf_info(&l->list, i)) { + struct list_head *fa_head = get_fa_head(l, i); - - if(!fa_head) + + if (!fa_head) continue; - if(list_empty(fa_head)) + if (list_empty(fa_head)) continue; putspace_seq(seq, indent+2); @@ -2136,7 +2138,7 @@ static void printnode_seq(struct seq_file *seq, int indent, struct node *n, } } else if (IS_TNODE(n)) { - struct tnode *tn=(struct tnode *)n; + struct tnode *tn = (struct tnode *)n; putspace_seq(seq, indent); seq_printf(seq, "| "); seq_printf(seq, "{key prefix=%08x/", tn->key&TKEY_GET_MASK(0, tn->pos)); printbin_seq(seq, tkey_extract_bits(tn->key, 0, tn->pos), tn->pos); @@ -2152,7 +2154,7 @@ static void printnode_seq(struct seq_file *seq, int indent, struct node *n, static void trie_dump_seq(struct seq_file *seq, struct trie *t) { - struct node *n=t->trie; + struct node *n = t->trie; int cindex=0; int indent=1; int pend=0; @@ -2164,7 +2166,7 @@ static void trie_dump_seq(struct seq_file *seq, struct trie *t) if (n) { printnode_seq(seq, indent, n, pend, cindex, 0); if (IS_TNODE(n)) { - struct tnode *tn=(struct tnode *)n; + struct tnode *tn = (struct tnode *)n; pend = tn->pos+tn->bits; putspace_seq(seq, indent); seq_printf(seq, "\\--\n"); indent += 3; @@ -2172,42 +2174,42 @@ static void trie_dump_seq(struct seq_file *seq, struct trie *t) while (tn && cindex < (1 << tn->bits)) { if (tn->child[cindex]) { - + /* Got a child */ - + printnode_seq(seq, indent, tn->child[cindex], pend, cindex, tn->bits); - if (IS_LEAF(tn->child[cindex])) { + if (IS_LEAF(tn->child[cindex])) { cindex++; - + } else { - /* - * New tnode. Decend one level + /* + * New tnode. Decend one level */ - + depth++; - n=tn->child[cindex]; - tn=(struct tnode *)n; - pend=tn->pos+tn->bits; + n = tn->child[cindex]; + tn = (struct tnode *)n; + pend = tn->pos+tn->bits; putspace_seq(seq, indent); seq_printf(seq, "\\--\n"); indent+=3; cindex=0; } } - else + else cindex++; /* - * Test if we are done + * Test if we are done */ - + while (cindex >= (1 << tn->bits)) { /* * Move upwards and test for root * pop off all traversed nodes */ - + if (NODE_PARENT(tn) == NULL) { tn = NULL; n = NULL; @@ -2217,8 +2219,8 @@ static void trie_dump_seq(struct seq_file *seq, struct trie *t) cindex = tkey_extract_bits(tn->key, NODE_PARENT(tn)->pos, NODE_PARENT(tn)->bits); tn = NODE_PARENT(tn); cindex++; - n=(struct node *)tn; - pend=tn->pos+tn->bits; + n = (struct node *)tn; + pend = tn->pos+tn->bits; indent-=3; depth--; } @@ -2236,36 +2238,36 @@ static struct trie_stat *trie_stat_new(void) { struct trie_stat *s = kmalloc(sizeof(struct trie_stat), GFP_KERNEL); int i; - - if(s) { + + if (s) { s->totdepth = 0; s->maxdepth = 0; s->tnodes = 0; s->leaves = 0; s->nullpointers = 0; - + for(i=0; i< MAX_CHILDS; i++) s->nodesizes[i] = 0; } return s; -} +} static struct trie_stat *trie_collect_stats(struct trie *t) { - struct node *n=t->trie; + struct node *n = t->trie; struct trie_stat *s = trie_stat_new(); int cindex = 0; int indent = 1; int pend = 0; int depth = 0; - read_lock(&fib_lock); + read_lock(&fib_lock); if (s) { if (n) { if (IS_TNODE(n)) { struct tnode *tn = (struct tnode *)n; - pend=tn->pos+tn->bits; + pend = tn->pos+tn->bits; indent += 3; s->nodesizes[tn->bits]++; depth++; @@ -2273,26 +2275,26 @@ static struct trie_stat *trie_collect_stats(struct trie *t) while (tn && cindex < (1 << tn->bits)) { if (tn->child[cindex]) { /* Got a child */ - - if (IS_LEAF(tn->child[cindex])) { + + if (IS_LEAF(tn->child[cindex])) { cindex++; - + /* stats */ if (depth > s->maxdepth) s->maxdepth = depth; s->totdepth += depth; s->leaves++; } - + else { - /* - * New tnode. Decend one level + /* + * New tnode. Decend one level */ - + s->tnodes++; s->nodesizes[tn->bits]++; depth++; - + n = tn->child[cindex]; tn = (struct tnode *)n; pend = tn->pos+tn->bits; @@ -2303,13 +2305,13 @@ static struct trie_stat *trie_collect_stats(struct trie *t) } else { cindex++; - s->nullpointers++; + s->nullpointers++; } /* - * Test if we are done + * Test if we are done */ - + while (cindex >= (1 << tn->bits)) { /* @@ -2317,7 +2319,7 @@ static struct trie_stat *trie_collect_stats(struct trie *t) * pop off all traversed nodes */ - + if (NODE_PARENT(tn) == NULL) { tn = NULL; n = NULL; @@ -2326,9 +2328,9 @@ static struct trie_stat *trie_collect_stats(struct trie *t) else { cindex = tkey_extract_bits(tn->key, NODE_PARENT(tn)->pos, NODE_PARENT(tn)->bits); tn = NODE_PARENT(tn); - cindex++; + cindex++; n = (struct node *)tn; - pend=tn->pos+tn->bits; + pend = tn->pos+tn->bits; indent -= 3; depth--; } @@ -2339,7 +2341,7 @@ static struct trie_stat *trie_collect_stats(struct trie *t) } } - read_unlock(&fib_lock); + read_unlock(&fib_lock); return s; } @@ -2375,7 +2377,7 @@ static void fib_triestat_seq_stop(struct seq_file *seq, void *v) } -/* +/* * This outputs /proc/net/fib_triestats * * It always works in backward compatibility mode. @@ -2401,7 +2403,7 @@ static void collect_and_show(struct trie *t, struct seq_file *seq) avdepth=0; seq_printf(seq, "Aver depth: %d.%02d\n", avdepth / 100, avdepth % 100 ); seq_printf(seq, "Max depth: %4d\n", stat->maxdepth); - + seq_printf(seq, "Leaves: %d\n", stat->leaves); bytes += sizeof(struct leaf) * stat->leaves; seq_printf(seq, "Internal nodes: %d\n", stat->tnodes); @@ -2413,7 +2415,7 @@ static void collect_and_show(struct trie *t, struct seq_file *seq) max--; pointers = 0; - for (i = 1; i <= max; i++) + for (i = 1; i <= max; i++) if (stat->nodesizes[i] != 0) { seq_printf(seq, " %d: %d", i, stat->nodesizes[i]); pointers += (1<nodesizes[i]; @@ -2444,30 +2446,30 @@ static void collect_and_show(struct trie *t, struct seq_file *seq) static int fib_triestat_seq_show(struct seq_file *seq, void *v) { char bf[128]; - + if (v == SEQ_START_TOKEN) { - seq_printf(seq, "Basic info: size of leaf: %Zd bytes, size of tnode: %Zd bytes.\n", + seq_printf(seq, "Basic info: size of leaf: %Zd bytes, size of tnode: %Zd bytes.\n", sizeof(struct leaf), sizeof(struct tnode)); - if (trie_local) + if (trie_local) collect_and_show(trie_local, seq); - if (trie_main) + if (trie_main) collect_and_show(trie_main, seq); } else { snprintf(bf, sizeof(bf), "*\t%08X\t%08X", 200, 400); - + seq_printf(seq, "%-127s\n", bf); } return 0; } static struct seq_operations fib_triestat_seq_ops = { - .start = fib_triestat_seq_start, - .next = fib_triestat_seq_next, - .stop = fib_triestat_seq_stop, - .show = fib_triestat_seq_show, + .start = fib_triestat_seq_start, + .next = fib_triestat_seq_next, + .stop = fib_triestat_seq_stop, + .show = fib_triestat_seq_show, }; static int fib_triestat_seq_open(struct inode *inode, struct file *file) @@ -2479,7 +2481,7 @@ static int fib_triestat_seq_open(struct inode *inode, struct file *file) if (rc) goto out_kfree; - seq = file->private_data; + seq = file->private_data; out: return rc; out_kfree: @@ -2487,11 +2489,11 @@ out_kfree: } static struct file_operations fib_triestat_seq_fops = { - .owner = THIS_MODULE, - .open = fib_triestat_seq_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release_private, + .owner = THIS_MODULE, + .open = fib_triestat_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, }; int __init fib_stat_proc_init(void) @@ -2536,7 +2538,7 @@ static void fib_trie_seq_stop(struct seq_file *seq, void *v) } -/* +/* * This outputs /proc/net/fib_trie. * * It always works in backward compatibility mode. @@ -2548,10 +2550,10 @@ static int fib_trie_seq_show(struct seq_file *seq, void *v) char bf[128]; if (v == SEQ_START_TOKEN) { - if (trie_local) + if (trie_local) trie_dump_seq(seq, trie_local); - if (trie_main) + if (trie_main) trie_dump_seq(seq, trie_main); } @@ -2565,10 +2567,10 @@ static int fib_trie_seq_show(struct seq_file *seq, void *v) } static struct seq_operations fib_trie_seq_ops = { - .start = fib_trie_seq_start, - .next = fib_trie_seq_next, - .stop = fib_trie_seq_stop, - .show = fib_trie_seq_show, + .start = fib_trie_seq_start, + .next = fib_trie_seq_next, + .stop = fib_trie_seq_stop, + .show = fib_trie_seq_show, }; static int fib_trie_seq_open(struct inode *inode, struct file *file) @@ -2580,7 +2582,7 @@ static int fib_trie_seq_open(struct inode *inode, struct file *file) if (rc) goto out_kfree; - seq = file->private_data; + seq = file->private_data; out: return rc; out_kfree: @@ -2588,11 +2590,11 @@ out_kfree: } static struct file_operations fib_trie_seq_fops = { - .owner = THIS_MODULE, - .open = fib_trie_seq_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release_private, + .owner = THIS_MODULE, + .open = fib_trie_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release= seq_release_private, }; int __init fib_proc_init(void) From 0303770deb834c15ca664a9d741d40f893c92f4e Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 19 Jul 2005 14:03:34 -0700 Subject: [PATCH 27/38] [NET]: Make ipip/ip6_tunnel independant of XFRM Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/xfrm.h | 2 +- net/ipv4/Kconfig | 1 - net/ipv4/ipip.c | 36 ++++++++++++++++++++++++++++++++---- net/ipv4/xfrm4_tunnel.c | 3 +-- net/ipv6/Kconfig | 1 - net/ipv6/ip6_tunnel.c | 38 +++++++++++++++++++++++++++++++++----- 6 files changed, 67 insertions(+), 14 deletions(-) diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 029522a4ceda..868ef88ef971 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -803,7 +803,7 @@ struct xfrm_algo_desc { /* XFRM tunnel handlers. */ struct xfrm_tunnel { int (*handler)(struct sk_buff *skb); - void (*err_handler)(struct sk_buff *skb, void *info); + void (*err_handler)(struct sk_buff *skb, __u32 info); }; struct xfrm6_tunnel { diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index c7f1f9245bc9..fc561c0ae8e2 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -235,7 +235,6 @@ config IP_PNP_RARP # bool ' IP: ARP support' CONFIG_IP_PNP_ARP config NET_IPIP tristate "IP: tunneling" - select INET_TUNNEL ---help--- Tunneling means encapsulating data of one protocol type within another protocol and sending it over a channel that understands the diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index 68a78731f722..c3947cd566b7 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -273,7 +273,7 @@ static void ipip_tunnel_uninit(struct net_device *dev) dev_put(dev); } -static void ipip_err(struct sk_buff *skb, void *__unused) +static void ipip_err(struct sk_buff *skb, u32 info) { #ifndef I_WISH_WORLD_WERE_PERFECT @@ -852,11 +852,39 @@ static int __init ipip_fb_tunnel_init(struct net_device *dev) return 0; } +#ifdef CONFIG_INET_TUNNEL static struct xfrm_tunnel ipip_handler = { .handler = ipip_rcv, .err_handler = ipip_err, }; +static inline int ipip_register(void) +{ + return xfrm4_tunnel_register(&ipip_handler); +} + +static inline int ipip_unregister(void) +{ + return xfrm4_tunnel_deregister(&ipip_handler); +} +#else +static struct net_protocol ipip_protocol = { + .handler = ipip_rcv, + .err_handler = ipip_err, + .no_policy = 1, +}; + +static inline int ipip_register(void) +{ + return inet_add_protocol(&ipip_protocol, IPPROTO_IPIP); +} + +static inline int ipip_unregister(void) +{ + return inet_del_protocol(&ipip_protocol, IPPROTO_IPIP); +} +#endif + static char banner[] __initdata = KERN_INFO "IPv4 over IPv4 tunneling driver\n"; @@ -866,7 +894,7 @@ static int __init ipip_init(void) printk(banner); - if (xfrm4_tunnel_register(&ipip_handler) < 0) { + if (ipip_register() < 0) { printk(KERN_INFO "ipip init: can't register tunnel\n"); return -EAGAIN; } @@ -888,13 +916,13 @@ static int __init ipip_init(void) err2: free_netdev(ipip_fb_tunnel_dev); err1: - xfrm4_tunnel_deregister(&ipip_handler); + ipip_unregister(); goto out; } static void __exit ipip_fini(void) { - if (xfrm4_tunnel_deregister(&ipip_handler) < 0) + if (ipip_unregister() < 0) printk(KERN_INFO "ipip close: can't deregister tunnel\n"); unregister_netdev(ipip_fb_tunnel_dev); diff --git a/net/ipv4/xfrm4_tunnel.c b/net/ipv4/xfrm4_tunnel.c index e1fe360ed27a..afbb0d4cc305 100644 --- a/net/ipv4/xfrm4_tunnel.c +++ b/net/ipv4/xfrm4_tunnel.c @@ -78,10 +78,9 @@ static int ipip_rcv(struct sk_buff *skb) static void ipip_err(struct sk_buff *skb, u32 info) { struct xfrm_tunnel *handler = ipip_handler; - u32 arg = info; if (handler) - handler->err_handler(skb, &arg); + handler->err_handler(skb, info); } static int ipip_init_state(struct xfrm_state *x) diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig index 95163cd52ae0..ab7a9124f985 100644 --- a/net/ipv6/Kconfig +++ b/net/ipv6/Kconfig @@ -91,7 +91,6 @@ config INET6_TUNNEL config IPV6_TUNNEL tristate "IPv6: IPv6-in-IPv6 tunnel" depends on IPV6 - select INET6_TUNNEL ---help--- Support for IPv6-in-IPv6 tunnels described in RFC 2473. diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index ba3b0c267f75..f39ddeae1eef 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -1110,11 +1110,39 @@ ip6ip6_fb_tnl_dev_init(struct net_device *dev) return 0; } +#ifdef CONFIG_INET6_TUNNEL static struct xfrm6_tunnel ip6ip6_handler = { - .handler = ip6ip6_rcv, - .err_handler = ip6ip6_err, + .handler = ip6ip6_rcv, + .err_handler = ip6ip6_err, }; +static inline int ip6ip6_register(void) +{ + return xfrm6_tunnel_register(&ip6ip6_handler); +} + +static inline int ip6ip6_unregister(void) +{ + return xfrm6_tunnel_unregister(&ip6ip6_handler); +} +#else +static struct inet6_protocol xfrm6_tunnel_protocol = { + .handler = ip6ip6_rcv, + .err_handler = ip6ip6_err, + .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, +}; + +static inline int ip6ip6_register(void) +{ + return inet6_add_protocol(&xfrm6_tunnel_protocol, IPPROTO_IPV6); +} + +static inline int ip6ip6_unregister(void) +{ + return inet6_del_protocol(&xfrm6_tunnel_protocol, IPPROTO_IPV6); +} +#endif + /** * ip6_tunnel_init - register protocol and reserve needed resources * @@ -1125,7 +1153,7 @@ static int __init ip6_tunnel_init(void) { int err; - if (xfrm6_tunnel_register(&ip6ip6_handler) < 0) { + if (ip6ip6_register() < 0) { printk(KERN_ERR "ip6ip6 init: can't register tunnel\n"); return -EAGAIN; } @@ -1144,7 +1172,7 @@ static int __init ip6_tunnel_init(void) } return 0; fail: - xfrm6_tunnel_deregister(&ip6ip6_handler); + ip6ip6_unregister(); return err; } @@ -1154,7 +1182,7 @@ fail: static void __exit ip6_tunnel_cleanup(void) { - if (xfrm6_tunnel_deregister(&ip6ip6_handler) < 0) + if (ip6ip6_unregister() < 0) printk(KERN_INFO "ip6ip6 close: can't deregister tunnel\n"); unregister_netdev(ip6ip6_fb_tnl_dev); From b72f6eccb0fc516070f2de469f73870cb6cb1149 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 19 Jul 2005 14:13:54 -0700 Subject: [PATCH 28/38] [NET]: Fix tc_verd thinko in skb_clone() It was overwriting the computer n->tc_verd value over and over with skb->tc_verd, by mistake. Signed-off-by: David S. Miller --- net/core/skbuff.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index d9f7b06fe886..7eab867ede59 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -377,8 +377,8 @@ struct sk_buff *skb_clone(struct sk_buff *skb, unsigned int __nocast gfp_mask) C(tc_index); #ifdef CONFIG_NET_CLS_ACT n->tc_verd = SET_TC_VERD(skb->tc_verd,0); - n->tc_verd = CLR_TC_OK2MUNGE(skb->tc_verd); - n->tc_verd = CLR_TC_MUNGED(skb->tc_verd); + n->tc_verd = CLR_TC_OK2MUNGE(n->tc_verd); + n->tc_verd = CLR_TC_MUNGED(n->tc_verd); C(input_dev); C(tc_classid); #endif From 4aa49d130df9209707a97786a55a3f584b7345e9 Mon Sep 17 00:00:00 2001 From: Francois Romieu Date: Wed, 20 Jul 2005 12:01:46 -0700 Subject: [PATCH 29/38] [ATM]: zatm: mailbox converted to pci_alloc_consistent() mailbox converted to pci_alloc_consistent() - request_region() is not needed: zatm_init_one() issues pci_request_regions(); - the warning related to kfree(zatm_dev->mbx_start) disappears; Compiled with i386 and sparc64 as target. Signed-off-by: Francois Romieu Signed-off-by: David S. Miller --- drivers/atm/zatm.c | 109 ++++++++++++++++++++++++++------------------- drivers/atm/zatm.h | 1 + 2 files changed, 63 insertions(+), 47 deletions(-) diff --git a/drivers/atm/zatm.c b/drivers/atm/zatm.c index afcf1ada5547..a2b236a966e0 100644 --- a/drivers/atm/zatm.c +++ b/drivers/atm/zatm.c @@ -16,9 +16,9 @@ #include #include #include -#include /* for request_region */ #include #include +#include #include #include #include @@ -1257,22 +1257,22 @@ static int __init zatm_init(struct atm_dev *dev) static int __init zatm_start(struct atm_dev *dev) { - struct zatm_dev *zatm_dev; + struct zatm_dev *zatm_dev = ZATM_DEV(dev); + struct pci_dev *pdev = zatm_dev->pci_dev; unsigned long curr; int pools,vccs,rx; - int error,i,ld; + int error, i, ld; DPRINTK("zatm_start\n"); - zatm_dev = ZATM_DEV(dev); zatm_dev->rx_map = zatm_dev->tx_map = NULL; - for (i = 0; i < NR_MBX; i++) - zatm_dev->mbx_start[i] = 0; - if (request_irq(zatm_dev->irq,&zatm_int,SA_SHIRQ,DEV_LABEL,dev)) { - printk(KERN_ERR DEV_LABEL "(itf %d): IRQ%d is already in use\n", - dev->number,zatm_dev->irq); - return -EAGAIN; + for (i = 0; i < NR_MBX; i++) + zatm_dev->mbx_start[i] = 0; + error = request_irq(zatm_dev->irq, zatm_int, SA_SHIRQ, DEV_LABEL, dev); + if (error < 0) { + printk(KERN_ERR DEV_LABEL "(itf %d): IRQ%d is already in use\n", + dev->number,zatm_dev->irq); + goto done; } - request_region(zatm_dev->base,uPD98401_PORTS,DEV_LABEL); /* define memory regions */ pools = NR_POOLS; if (NR_SHAPERS*SHAPER_SIZE > pools*POOL_SIZE) @@ -1299,51 +1299,66 @@ static int __init zatm_start(struct atm_dev *dev) "%ld VCs\n",dev->number,NR_SHAPERS,pools,rx, (zatm_dev->mem-curr*4)/VC_SIZE); /* create mailboxes */ - for (i = 0; i < NR_MBX; i++) - if (mbx_entries[i]) { - unsigned long here; + for (i = 0; i < NR_MBX; i++) { + void *mbx; + dma_addr_t mbx_dma; - here = (unsigned long) kmalloc(2*MBX_SIZE(i), - GFP_KERNEL); - if (!here) { - error = -ENOMEM; - goto out; - } - if ((here^(here+MBX_SIZE(i))) & ~0xffffUL)/* paranoia */ - here = (here & ~0xffffUL)+0x10000; - zatm_dev->mbx_start[i] = here; - if ((here^virt_to_bus((void *) here)) & 0xffff) { - printk(KERN_ERR DEV_LABEL "(itf %d): system " - "bus incompatible with driver\n", - dev->number); - error = -ENODEV; - goto out; - } - DPRINTK("mbx@0x%08lx-0x%08lx\n",here,here+MBX_SIZE(i)); - zatm_dev->mbx_end[i] = (here+MBX_SIZE(i)) & 0xffff; - zout(virt_to_bus((void *) here) >> 16,MSH(i)); - zout(virt_to_bus((void *) here),MSL(i)); - zout((here+MBX_SIZE(i)) & 0xffff,MBA(i)); - zout(here & 0xffff,MTA(i)); - zout(here & 0xffff,MWA(i)); + if (!mbx_entries[i]) + continue; + mbx = pci_alloc_consistent(pdev, 2*MBX_SIZE(i), &mbx_dma); + if (!mbx) { + error = -ENOMEM; + goto out; } + /* + * Alignment provided by pci_alloc_consistent() isn't enough + * for this device. + */ + if (((unsigned long)mbx ^ mbx_dma) & 0xffff) { + printk(KERN_ERR DEV_LABEL "(itf %d): system " + "bus incompatible with driver\n", dev->number); + pci_free_consistent(pdev, 2*MBX_SIZE(i), mbx, mbx_dma); + error = -ENODEV; + goto out; + } + DPRINTK("mbx@0x%08lx-0x%08lx\n", mbx, mbx + MBX_SIZE(i)); + zatm_dev->mbx_start[i] = (unsigned long)mbx; + zatm_dev->mbx_dma[i] = mbx_dma; + zatm_dev->mbx_end[i] = (zatm_dev->mbx_start[i] + MBX_SIZE(i)) & + 0xffff; + zout(mbx_dma >> 16, MSH(i)); + zout(mbx_dma, MSL(i)); + zout(zatm_dev->mbx_end[i], MBA(i)); + zout((unsigned long)mbx & 0xffff, MTA(i)); + zout((unsigned long)mbx & 0xffff, MWA(i)); + } error = start_tx(dev); - if (error) goto out; + if (error) + goto out; error = start_rx(dev); - if (error) goto out; + if (error) + goto out_tx; error = dev->phy->start(dev); - if (error) goto out; + if (error) + goto out_rx; zout(0xffffffff,IMR); /* enable interrupts */ /* enable TX & RX */ zout(zin(GMR) | uPD98401_GMR_SE | uPD98401_GMR_RE,GMR); - return 0; - out: - for (i = 0; i < NR_MBX; i++) - kfree(&zatm_dev->mbx_start[i]); - kfree(zatm_dev->rx_map); - kfree(zatm_dev->tx_map); - free_irq(zatm_dev->irq, dev); +done: return error; + +out_rx: + kfree(zatm_dev->rx_map); +out_tx: + kfree(zatm_dev->tx_map); +out: + while (i-- > 0) { + pci_free_consistent(pdev, 2*MBX_SIZE(i), + (void *)zatm_dev->mbx_start[i], + zatm_dev->mbx_dma[i]); + } + free_irq(zatm_dev->irq, dev); + goto done; } diff --git a/drivers/atm/zatm.h b/drivers/atm/zatm.h index 34a0480f63d6..416fe0fda60c 100644 --- a/drivers/atm/zatm.h +++ b/drivers/atm/zatm.h @@ -73,6 +73,7 @@ struct zatm_dev { int chans; /* map size, must be 2^n */ /*-------------------------------- mailboxes */ unsigned long mbx_start[NR_MBX];/* start addresses */ + dma_addr_t mbx_dma[NR_MBX]; u16 mbx_end[NR_MBX]; /* end offset (in bytes) */ /*-------------------------------- other pointers */ u32 pool_base; /* Free buffer pool dsc (word addr) */ From 4acdbdbe5089c06d5e0c7e96783fcc4414ded00a Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 21 Jul 2005 13:14:46 -0700 Subject: [PATCH 30/38] [NETFILTER]: ip_conntrack_expect_related must not free expectation If a connection tracking helper tells us to expect a connection, and we're already expecting that connection, we simply free the one they gave us and return success. The problem is that NAT helpers (eg. FTP) have to allocate the expectation first (to see what port is available) then rewrite the packet. If that rewrite fails, they try to remove the expectation, but it was freed in ip_conntrack_expect_related. This is one example of a larger problem: having registered the expectation, the pointer is no longer ours to use. Reference counting is needed for ctnetlink anyway, so introduce it now. To have a single "put" path, we need to grab the reference to the connection on creation, rather than open-coding it in the caller. Signed-off-by: Rusty Russell Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4/ip_conntrack.h | 3 ++ .../netfilter_ipv4/ip_conntrack_helper.h | 7 ++-- net/ipv4/netfilter/ip_conntrack_amanda.c | 8 ++-- net/ipv4/netfilter/ip_conntrack_core.c | 40 +++++++++---------- net/ipv4/netfilter/ip_conntrack_ftp.c | 14 +++---- net/ipv4/netfilter/ip_conntrack_irc.c | 8 ++-- net/ipv4/netfilter/ip_conntrack_standalone.c | 2 +- net/ipv4/netfilter/ip_conntrack_tftp.c | 8 ++-- net/ipv4/netfilter/ip_nat_amanda.c | 4 +- net/ipv4/netfilter/ip_nat_ftp.c | 4 +- net/ipv4/netfilter/ip_nat_irc.c | 4 +- net/ipv4/netfilter/ip_nat_tftp.c | 4 +- 12 files changed, 46 insertions(+), 60 deletions(-) diff --git a/include/linux/netfilter_ipv4/ip_conntrack.h b/include/linux/netfilter_ipv4/ip_conntrack.h index 3781192ce159..f8da7ddeff3a 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack.h +++ b/include/linux/netfilter_ipv4/ip_conntrack.h @@ -197,6 +197,9 @@ struct ip_conntrack_expect /* Timer function; deletes the expectation. */ struct timer_list timeout; + /* Usage count. */ + atomic_t use; + #ifdef CONFIG_IP_NF_NAT_NEEDED /* This is the original per-proto part, used to map the * expected connection the way the recipient expects. */ diff --git a/include/linux/netfilter_ipv4/ip_conntrack_helper.h b/include/linux/netfilter_ipv4/ip_conntrack_helper.h index b1bbba0a12cb..3692daa93dec 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_helper.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_helper.h @@ -30,9 +30,10 @@ extern int ip_conntrack_helper_register(struct ip_conntrack_helper *); extern void ip_conntrack_helper_unregister(struct ip_conntrack_helper *); /* Allocate space for an expectation: this is mandatory before calling - ip_conntrack_expect_related. */ -extern struct ip_conntrack_expect *ip_conntrack_expect_alloc(void); -extern void ip_conntrack_expect_free(struct ip_conntrack_expect *exp); + ip_conntrack_expect_related. You will have to call put afterwards. */ +extern struct ip_conntrack_expect * +ip_conntrack_expect_alloc(struct ip_conntrack *master); +extern void ip_conntrack_expect_put(struct ip_conntrack_expect *exp); /* Add an expected connection: can have more than one per connection */ extern int ip_conntrack_expect_related(struct ip_conntrack_expect *exp); diff --git a/net/ipv4/netfilter/ip_conntrack_amanda.c b/net/ipv4/netfilter/ip_conntrack_amanda.c index a78a320eee08..01e1b58322a9 100644 --- a/net/ipv4/netfilter/ip_conntrack_amanda.c +++ b/net/ipv4/netfilter/ip_conntrack_amanda.c @@ -101,14 +101,13 @@ static int help(struct sk_buff **pskb, if (port == 0 || len > 5) break; - exp = ip_conntrack_expect_alloc(); + exp = ip_conntrack_expect_alloc(ct); if (exp == NULL) { ret = NF_DROP; goto out; } exp->expectfn = NULL; - exp->master = ct; exp->tuple.src.ip = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip; exp->tuple.src.u.tcp.port = 0; @@ -126,10 +125,9 @@ static int help(struct sk_buff **pskb, ret = ip_nat_amanda_hook(pskb, ctinfo, tmp - amanda_buffer, len, exp); - else if (ip_conntrack_expect_related(exp) != 0) { - ip_conntrack_expect_free(exp); + else if (ip_conntrack_expect_related(exp) != 0) ret = NF_DROP; - } + ip_conntrack_expect_put(exp); } out: diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c index 4b78ebeb6635..14af55cad5d6 100644 --- a/net/ipv4/netfilter/ip_conntrack_core.c +++ b/net/ipv4/netfilter/ip_conntrack_core.c @@ -137,19 +137,12 @@ ip_ct_invert_tuple(struct ip_conntrack_tuple *inverse, /* ip_conntrack_expect helper functions */ -static void destroy_expect(struct ip_conntrack_expect *exp) -{ - ip_conntrack_put(exp->master); - IP_NF_ASSERT(!timer_pending(&exp->timeout)); - kmem_cache_free(ip_conntrack_expect_cachep, exp); - CONNTRACK_STAT_INC(expect_delete); -} - static void unlink_expect(struct ip_conntrack_expect *exp) { ASSERT_WRITE_LOCK(&ip_conntrack_lock); + IP_NF_ASSERT(!timer_pending(&exp->timeout)); list_del(&exp->list); - /* Logically in destroy_expect, but we hold the lock here. */ + CONNTRACK_STAT_INC(expect_delete); exp->master->expecting--; } @@ -160,7 +153,7 @@ static void expectation_timed_out(unsigned long ul_expect) write_lock_bh(&ip_conntrack_lock); unlink_expect(exp); write_unlock_bh(&ip_conntrack_lock); - destroy_expect(exp); + ip_conntrack_expect_put(exp); } /* If an expectation for this connection is found, it gets delete from @@ -198,7 +191,7 @@ static void remove_expectations(struct ip_conntrack *ct) list_for_each_entry_safe(i, tmp, &ip_conntrack_expect_list, list) { if (i->master == ct && del_timer(&i->timeout)) { unlink_expect(i); - destroy_expect(i); + ip_conntrack_expect_put(i); } } } @@ -537,7 +530,7 @@ init_conntrack(const struct ip_conntrack_tuple *tuple, if (exp) { if (exp->expectfn) exp->expectfn(conntrack, exp); - destroy_expect(exp); + ip_conntrack_expect_put(exp); } return &conntrack->tuplehash[IP_CT_DIR_ORIGINAL]; @@ -729,14 +722,14 @@ void ip_conntrack_unexpect_related(struct ip_conntrack_expect *exp) if (expect_matches(i, exp) && del_timer(&i->timeout)) { unlink_expect(i); write_unlock_bh(&ip_conntrack_lock); - destroy_expect(i); + ip_conntrack_expect_put(i); return; } } write_unlock_bh(&ip_conntrack_lock); } -struct ip_conntrack_expect *ip_conntrack_expect_alloc(void) +struct ip_conntrack_expect *ip_conntrack_expect_alloc(struct ip_conntrack *me) { struct ip_conntrack_expect *new; @@ -745,18 +738,23 @@ struct ip_conntrack_expect *ip_conntrack_expect_alloc(void) DEBUGP("expect_related: OOM allocating expect\n"); return NULL; } - new->master = NULL; + new->master = me; + atomic_inc(&new->master->ct_general.use); + atomic_set(&new->use, 1); return new; } -void ip_conntrack_expect_free(struct ip_conntrack_expect *expect) +void ip_conntrack_expect_put(struct ip_conntrack_expect *exp) { - kmem_cache_free(ip_conntrack_expect_cachep, expect); + if (atomic_dec_and_test(&exp->use)) { + ip_conntrack_put(exp->master); + kmem_cache_free(ip_conntrack_expect_cachep, exp); + } } static void ip_conntrack_expect_insert(struct ip_conntrack_expect *exp) { - atomic_inc(&exp->master->ct_general.use); + atomic_inc(&exp->use); exp->master->expecting++; list_add(&exp->list, &ip_conntrack_expect_list); @@ -778,7 +776,7 @@ static void evict_oldest_expect(struct ip_conntrack *master) if (i->master == master) { if (del_timer(&i->timeout)) { unlink_expect(i); - destroy_expect(i); + ip_conntrack_expect_put(i); } break; } @@ -810,8 +808,6 @@ int ip_conntrack_expect_related(struct ip_conntrack_expect *expect) /* Refresh timer: if it's dying, ignore.. */ if (refresh_timer(i)) { ret = 0; - /* We don't need the one they've given us. */ - ip_conntrack_expect_free(expect); goto out; } } else if (expect_clash(i, expect)) { @@ -881,7 +877,7 @@ void ip_conntrack_helper_unregister(struct ip_conntrack_helper *me) list_for_each_entry_safe(exp, tmp, &ip_conntrack_expect_list, list) { if (exp->master->helper == me && del_timer(&exp->timeout)) { unlink_expect(exp); - destroy_expect(exp); + ip_conntrack_expect_put(exp); } } /* Get rid of expecteds, set helpers to NULL. */ diff --git a/net/ipv4/netfilter/ip_conntrack_ftp.c b/net/ipv4/netfilter/ip_conntrack_ftp.c index fea6dd2a00b6..7a3b773be3f9 100644 --- a/net/ipv4/netfilter/ip_conntrack_ftp.c +++ b/net/ipv4/netfilter/ip_conntrack_ftp.c @@ -376,7 +376,7 @@ static int help(struct sk_buff **pskb, fb_ptr + matchoff, matchlen, ntohl(th->seq) + matchoff); /* Allocate expectation which will be inserted */ - exp = ip_conntrack_expect_alloc(); + exp = ip_conntrack_expect_alloc(ct); if (exp == NULL) { ret = NF_DROP; goto out; @@ -403,8 +403,7 @@ static int help(struct sk_buff **pskb, networks, or the packet filter itself). */ if (!loose) { ret = NF_ACCEPT; - ip_conntrack_expect_free(exp); - goto out_update_nl; + goto out_put_expect; } exp->tuple.dst.ip = htonl((array[0] << 24) | (array[1] << 16) | (array[2] << 8) | array[3]); @@ -419,7 +418,6 @@ static int help(struct sk_buff **pskb, { 0xFFFFFFFF, { .tcp = { 0xFFFF } }, 0xFF }}); exp->expectfn = NULL; - exp->master = ct; /* Now, NAT might want to mangle the packet, and register the * (possibly changed) expectation itself. */ @@ -428,13 +426,15 @@ static int help(struct sk_buff **pskb, matchoff, matchlen, exp, &seq); else { /* Can't expect this? Best to drop packet now. */ - if (ip_conntrack_expect_related(exp) != 0) { - ip_conntrack_expect_free(exp); + if (ip_conntrack_expect_related(exp) != 0) ret = NF_DROP; - } else + else ret = NF_ACCEPT; } +out_put_expect: + ip_conntrack_expect_put(exp); + out_update_nl: /* Now if this ends in \n, update ftp info. Seq may have been * adjusted by NAT code. */ diff --git a/net/ipv4/netfilter/ip_conntrack_irc.c b/net/ipv4/netfilter/ip_conntrack_irc.c index cd98772cc332..4a28f297d502 100644 --- a/net/ipv4/netfilter/ip_conntrack_irc.c +++ b/net/ipv4/netfilter/ip_conntrack_irc.c @@ -197,7 +197,7 @@ static int help(struct sk_buff **pskb, continue; } - exp = ip_conntrack_expect_alloc(); + exp = ip_conntrack_expect_alloc(ct); if (exp == NULL) { ret = NF_DROP; goto out; @@ -221,16 +221,14 @@ static int help(struct sk_buff **pskb, { { 0, { 0 } }, { 0xFFFFFFFF, { .tcp = { 0xFFFF } }, 0xFF }}); exp->expectfn = NULL; - exp->master = ct; if (ip_nat_irc_hook) ret = ip_nat_irc_hook(pskb, ctinfo, addr_beg_p - ib_ptr, addr_end_p - addr_beg_p, exp); - else if (ip_conntrack_expect_related(exp) != 0) { - ip_conntrack_expect_free(exp); + else if (ip_conntrack_expect_related(exp) != 0) ret = NF_DROP; - } + ip_conntrack_expect_put(exp); goto out; } /* for .. NUM_DCCPROTO */ } /* while data < ... */ diff --git a/net/ipv4/netfilter/ip_conntrack_standalone.c b/net/ipv4/netfilter/ip_conntrack_standalone.c index 1dd824f3cf0a..61798c46e91d 100644 --- a/net/ipv4/netfilter/ip_conntrack_standalone.c +++ b/net/ipv4/netfilter/ip_conntrack_standalone.c @@ -985,7 +985,7 @@ EXPORT_SYMBOL(ip_ct_refresh_acct); EXPORT_SYMBOL(ip_ct_protos); EXPORT_SYMBOL(ip_ct_find_proto); EXPORT_SYMBOL(ip_conntrack_expect_alloc); -EXPORT_SYMBOL(ip_conntrack_expect_free); +EXPORT_SYMBOL(ip_conntrack_expect_put); EXPORT_SYMBOL(ip_conntrack_expect_related); EXPORT_SYMBOL(ip_conntrack_unexpect_related); EXPORT_SYMBOL(ip_conntrack_tuple_taken); diff --git a/net/ipv4/netfilter/ip_conntrack_tftp.c b/net/ipv4/netfilter/ip_conntrack_tftp.c index 992fac3e36ee..f8ff170f390a 100644 --- a/net/ipv4/netfilter/ip_conntrack_tftp.c +++ b/net/ipv4/netfilter/ip_conntrack_tftp.c @@ -65,7 +65,7 @@ static int tftp_help(struct sk_buff **pskb, DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); - exp = ip_conntrack_expect_alloc(); + exp = ip_conntrack_expect_alloc(ct); if (exp == NULL) return NF_DROP; @@ -75,17 +75,15 @@ static int tftp_help(struct sk_buff **pskb, exp->mask.dst.u.udp.port = 0xffff; exp->mask.dst.protonum = 0xff; exp->expectfn = NULL; - exp->master = ct; DEBUGP("expect: "); DUMP_TUPLE(&exp->tuple); DUMP_TUPLE(&exp->mask); if (ip_nat_tftp_hook) ret = ip_nat_tftp_hook(pskb, ctinfo, exp); - else if (ip_conntrack_expect_related(exp) != 0) { - ip_conntrack_expect_free(exp); + else if (ip_conntrack_expect_related(exp) != 0) ret = NF_DROP; - } + ip_conntrack_expect_put(exp); break; case TFTP_OPCODE_DATA: case TFTP_OPCODE_ACK: diff --git a/net/ipv4/netfilter/ip_nat_amanda.c b/net/ipv4/netfilter/ip_nat_amanda.c index da1f412583ed..706c8074f422 100644 --- a/net/ipv4/netfilter/ip_nat_amanda.c +++ b/net/ipv4/netfilter/ip_nat_amanda.c @@ -56,10 +56,8 @@ static unsigned int help(struct sk_buff **pskb, break; } - if (port == 0) { - ip_conntrack_expect_free(exp); + if (port == 0) return NF_DROP; - } sprintf(buffer, "%u", port); ret = ip_nat_mangle_udp_packet(pskb, exp->master, ctinfo, diff --git a/net/ipv4/netfilter/ip_nat_ftp.c b/net/ipv4/netfilter/ip_nat_ftp.c index c6000e794ad6..d83757a70d9f 100644 --- a/net/ipv4/netfilter/ip_nat_ftp.c +++ b/net/ipv4/netfilter/ip_nat_ftp.c @@ -143,10 +143,8 @@ static unsigned int ip_nat_ftp(struct sk_buff **pskb, break; } - if (port == 0) { - ip_conntrack_expect_free(exp); + if (port == 0) return NF_DROP; - } if (!mangle[type](pskb, newip, port, matchoff, matchlen, ct, ctinfo, seq)) { diff --git a/net/ipv4/netfilter/ip_nat_irc.c b/net/ipv4/netfilter/ip_nat_irc.c index 9c1ca3381d56..de31942babe3 100644 --- a/net/ipv4/netfilter/ip_nat_irc.c +++ b/net/ipv4/netfilter/ip_nat_irc.c @@ -65,10 +65,8 @@ static unsigned int help(struct sk_buff **pskb, break; } - if (port == 0) { - ip_conntrack_expect_free(exp); + if (port == 0) return NF_DROP; - } /* strlen("\1DCC CHAT chat AAAAAAAA P\1\n")=27 * strlen("\1DCC SCHAT chat AAAAAAAA P\1\n")=28 diff --git a/net/ipv4/netfilter/ip_nat_tftp.c b/net/ipv4/netfilter/ip_nat_tftp.c index 0343e0d64674..2215317c76b7 100644 --- a/net/ipv4/netfilter/ip_nat_tftp.c +++ b/net/ipv4/netfilter/ip_nat_tftp.c @@ -45,10 +45,8 @@ static unsigned int help(struct sk_buff **pskb, exp->saved_proto.udp.port = exp->tuple.dst.u.tcp.port; exp->dir = IP_CT_DIR_REPLY; exp->expectfn = ip_nat_follow_master; - if (ip_conntrack_expect_related(exp) != 0) { - ip_conntrack_expect_free(exp); + if (ip_conntrack_expect_related(exp) != 0) return NF_DROP; - } return NF_ACCEPT; } From 14a59e1829cd5782ad1a8979e594578f49f3e2c1 Mon Sep 17 00:00:00 2001 From: Loic Le Loarer Date: Thu, 21 Jul 2005 14:16:54 -0700 Subject: [PATCH 31/38] [EQL]: Proper num_slaves decrement Signed-off-by: Loic Le Loarer Signed-off-by: David S. Miller --- drivers/net/eql.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/net/eql.c b/drivers/net/eql.c index dd6865820372..aa1569182fd6 100644 --- a/drivers/net/eql.c +++ b/drivers/net/eql.c @@ -132,7 +132,7 @@ static struct net_device_stats *eql_get_stats(struct net_device *dev); #define eql_is_slave(dev) ((dev->flags & IFF_SLAVE) == IFF_SLAVE) #define eql_is_master(dev) ((dev->flags & IFF_MASTER) == IFF_MASTER) -static void eql_kill_one_slave(slave_t *slave); +static void eql_kill_one_slave(slave_queue_t *queue, slave_t *slave); static void eql_timer(unsigned long param) { @@ -149,7 +149,7 @@ static void eql_timer(unsigned long param) if (slave->bytes_queued < 0) slave->bytes_queued = 0; } else { - eql_kill_one_slave(slave); + eql_kill_one_slave(&eql->queue, slave); } } @@ -214,9 +214,10 @@ static int eql_open(struct net_device *dev) return 0; } -static void eql_kill_one_slave(slave_t *slave) +static void eql_kill_one_slave(slave_queue_t *queue, slave_t *slave) { list_del(&slave->list); + queue->num_slaves--; slave->dev->flags &= ~IFF_SLAVE; dev_put(slave->dev); kfree(slave); @@ -232,8 +233,7 @@ static void eql_kill_slave_queue(slave_queue_t *queue) list_for_each_safe(this, tmp, head) { slave_t *s = list_entry(this, slave_t, list); - eql_kill_one_slave(s); - queue->num_slaves--; + eql_kill_one_slave(queue, s); } spin_unlock_bh(&queue->lock); @@ -318,7 +318,7 @@ static slave_t *__eql_schedule_slaves(slave_queue_t *queue) } } else { /* We found a dead slave, kill it. */ - eql_kill_one_slave(slave); + eql_kill_one_slave(queue, slave); } } return best_slave; @@ -393,7 +393,7 @@ static int __eql_insert_slave(slave_queue_t *queue, slave_t *slave) duplicate_slave = __eql_find_slave_dev(queue, slave->dev); if (duplicate_slave != 0) - eql_kill_one_slave(duplicate_slave); + eql_kill_one_slave(queue, duplicate_slave); list_add(&slave->list, &queue->all_slaves); queue->num_slaves++; @@ -471,7 +471,7 @@ static int eql_emancipate(struct net_device *master_dev, slaving_request_t __use slave_dev); if (slave) { - eql_kill_one_slave(slave); + eql_kill_one_slave(&eql->queue, slave); ret = 0; } } From 28e212fb360ce2568edd60b93d60683d5ad24146 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 22 Jul 2005 11:47:25 -0700 Subject: [PATCH 32/38] [PKT_SCHED]: Kill TCF_META_ID_REALDEV from meta ematch. It won't exist any longer when we shrink the SKB in 2.6.14, and we should kill this off before anyone in userspace starts using it. Signed-off-by: David S. Miller Acked-by: Thomas Graf --- include/linux/tc_ematch/tc_em_meta.h | 1 - net/sched/em_meta.c | 12 ------------ 2 files changed, 13 deletions(-) diff --git a/include/linux/tc_ematch/tc_em_meta.h b/include/linux/tc_ematch/tc_em_meta.h index 90ab9f95d43d..150efe07ff6b 100644 --- a/include/linux/tc_ematch/tc_em_meta.h +++ b/include/linux/tc_ematch/tc_em_meta.h @@ -42,7 +42,6 @@ enum TCF_META_ID_LOADAVG_2, TCF_META_ID_DEV, TCF_META_ID_INDEV, - TCF_META_ID_REALDEV, TCF_META_ID_PRIORITY, TCF_META_ID_PROTOCOL, TCF_META_ID_SECURITY, /* obsolete */ diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index feefcbfd03d0..179efb5bc9b3 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -180,16 +180,6 @@ META_COLLECTOR(var_indev) *err = var_dev(skb->input_dev, dst); } -META_COLLECTOR(int_realdev) -{ - *err = int_dev(skb->real_dev, dst); -} - -META_COLLECTOR(var_realdev) -{ - *err = var_dev(skb->real_dev, dst); -} - /************************************************************************** * skb attributes **************************************************************************/ @@ -501,7 +491,6 @@ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [TCF_META_TYPE_VAR] = { [META_ID(DEV)] = META_FUNC(var_dev), [META_ID(INDEV)] = META_FUNC(var_indev), - [META_ID(REALDEV)] = META_FUNC(var_realdev), [META_ID(SK_BOUND_IF)] = META_FUNC(var_sk_bound_if), }, [TCF_META_TYPE_INT] = { @@ -511,7 +500,6 @@ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [META_ID(LOADAVG_2)] = META_FUNC(int_loadavg_2), [META_ID(DEV)] = META_FUNC(int_dev), [META_ID(INDEV)] = META_FUNC(int_indev), - [META_ID(REALDEV)] = META_FUNC(int_realdev), [META_ID(PRIORITY)] = META_FUNC(int_priority), [META_ID(PROTOCOL)] = META_FUNC(int_protocol), [META_ID(PKTTYPE)] = META_FUNC(int_pkttype), From 4c1217deeb148ff8ab838ba4f1875d0f52dea343 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 22 Jul 2005 12:49:30 -0700 Subject: [PATCH 33/38] [NETFILTER]: Fix deadlock in ip6_queue Already fixed in ip_queue, ip6_queue was missed. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv6/netfilter/ip6_queue.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c index 750943e2d34e..5493180f0d44 100644 --- a/net/ipv6/netfilter/ip6_queue.c +++ b/net/ipv6/netfilter/ip6_queue.c @@ -76,7 +76,9 @@ static DECLARE_MUTEX(ipqnl_sem); static void ipq_issue_verdict(struct ipq_queue_entry *entry, int verdict) { + local_bh_disable(); nf_reinject(entry->skb, entry->info, verdict); + local_bh_enable(); kfree(entry); } From d04b4f8c1c9766e49fad6a141fc61cb30db69a5c Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 22 Jul 2005 12:50:29 -0700 Subject: [PATCH 34/38] [NETFILTER]: Fix potential memory corruption in NAT code (aka memory NAT) The portptr pointing to the port in the conntrack tuple is declared static, which could result in memory corruption when two packets of the same protocol are NATed at the same time and one conntrack goes away. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_nat_proto_tcp.c | 3 ++- net/ipv4/netfilter/ip_nat_proto_udp.c | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/net/ipv4/netfilter/ip_nat_proto_tcp.c b/net/ipv4/netfilter/ip_nat_proto_tcp.c index a91cfceff272..a98e36d2b3c6 100644 --- a/net/ipv4/netfilter/ip_nat_proto_tcp.c +++ b/net/ipv4/netfilter/ip_nat_proto_tcp.c @@ -40,7 +40,8 @@ tcp_unique_tuple(struct ip_conntrack_tuple *tuple, enum ip_nat_manip_type maniptype, const struct ip_conntrack *conntrack) { - static u_int16_t port, *portptr; + static u_int16_t port; + u_int16_t *portptr; unsigned int range_size, min, i; if (maniptype == IP_NAT_MANIP_SRC) diff --git a/net/ipv4/netfilter/ip_nat_proto_udp.c b/net/ipv4/netfilter/ip_nat_proto_udp.c index c669e3b5f5d0..9f66e5625664 100644 --- a/net/ipv4/netfilter/ip_nat_proto_udp.c +++ b/net/ipv4/netfilter/ip_nat_proto_udp.c @@ -41,7 +41,8 @@ udp_unique_tuple(struct ip_conntrack_tuple *tuple, enum ip_nat_manip_type maniptype, const struct ip_conntrack *conntrack) { - static u_int16_t port, *portptr; + static u_int16_t port; + u_int16_t *portptr; unsigned int range_size, min, i; if (maniptype == IP_NAT_MANIP_SRC) From 21f930e4abdcb9649f26e5b959c14dddee4e600b Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 22 Jul 2005 12:51:03 -0700 Subject: [PATCH 35/38] [NETFILTER]: Wait until all references to ip_conntrack_untracked are dropped on unload Fixes a crash when unloading ip_conntrack. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_conntrack_core.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c index 14af55cad5d6..63bf88264980 100644 --- a/net/ipv4/netfilter/ip_conntrack_core.c +++ b/net/ipv4/netfilter/ip_conntrack_core.c @@ -1107,6 +1107,9 @@ void ip_conntrack_cleanup(void) schedule(); goto i_see_dead_people; } + /* wait until all references to ip_conntrack_untracked are dropped */ + while (atomic_read(&ip_conntrack_untracked.ct_general.use) > 1) + schedule(); kmem_cache_destroy(ip_conntrack_cachep); kmem_cache_destroy(ip_conntrack_expect_cachep); From 74bb421da7f39e70ab636ad46ef85ea1178786c5 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 22 Jul 2005 12:51:38 -0700 Subject: [PATCH 36/38] [NETFILTER]: Use correct byteorder in ICMP NAT Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_nat_proto_icmp.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/net/ipv4/netfilter/ip_nat_proto_icmp.c b/net/ipv4/netfilter/ip_nat_proto_icmp.c index a558cf0eee8a..6596c9ee1655 100644 --- a/net/ipv4/netfilter/ip_nat_proto_icmp.c +++ b/net/ipv4/netfilter/ip_nat_proto_icmp.c @@ -35,16 +35,17 @@ icmp_unique_tuple(struct ip_conntrack_tuple *tuple, const struct ip_conntrack *conntrack) { static u_int16_t id; - unsigned int range_size - = (unsigned int)range->max.icmp.id - range->min.icmp.id + 1; + unsigned int range_size; unsigned int i; + range_size = ntohs(range->max.icmp.id) - ntohs(range->min.icmp.id) + 1; /* If no range specified... */ if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) range_size = 0xFFFF; for (i = 0; i < range_size; i++, id++) { - tuple->src.u.icmp.id = range->min.icmp.id + (id % range_size); + tuple->src.u.icmp.id = htons(ntohs(range->min.icmp.id) + + (id % range_size)); if (!ip_nat_used_tuple(tuple, conntrack)) return 1; } From d3984a6b6abac6203868f0e9095c0ed9e33ece03 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 22 Jul 2005 12:52:47 -0700 Subject: [PATCH 37/38] [NETFILTER]: Fix ip6t_LOG MAC format I broke this in the patch that consolidated MAC logging. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv6/netfilter/ip6t_LOG.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c index c44685e391b7..a692e26a4fa3 100644 --- a/net/ipv6/netfilter/ip6t_LOG.c +++ b/net/ipv6/netfilter/ip6t_LOG.c @@ -373,9 +373,10 @@ ip6t_log_packet(unsigned int hooknum, in ? in->name : "", out ? out->name : ""); if (in && !out) { + unsigned int len; /* MAC logging for input chain only. */ printk("MAC="); - if (skb->dev && skb->dev->hard_header_len && + if (skb->dev && (len = skb->dev->hard_header_len) && skb->mac.raw != skb->nh.raw) { unsigned char *p = skb->mac.raw; int i; @@ -384,9 +385,11 @@ ip6t_log_packet(unsigned int hooknum, (p -= ETH_HLEN) < skb->head) p = NULL; - if (p != NULL) - for (i = 0; i < skb->dev->hard_header_len; i++) - printk("%02x", p[i]); + if (p != NULL) { + for (i = 0; i < len; i++) + printk("%02x%s", p[i], + i == len - 1 ? "" : ":"); + } printk(" "); if (skb->dev->type == ARPHRD_SIT) { From 261688d01ec07d3a265b8ace6ec68310fbd96a96 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 22 Jul 2005 14:43:52 -0700 Subject: [PATCH 38/38] [PKT_SCHED]: em_meta: Kill TCF_META_ID_{INDEV,SECURITY,TCVERDICT} More unusable TCF_META_* match types that need to get eliminated before 2.6.13 goes out the door. Signed-off-by: David S. Miller Acked-by: Thomas Graf --- include/linux/tc_ematch/tc_em_meta.h | 3 --- net/sched/em_meta.c | 28 +++------------------------- 2 files changed, 3 insertions(+), 28 deletions(-) diff --git a/include/linux/tc_ematch/tc_em_meta.h b/include/linux/tc_ematch/tc_em_meta.h index 150efe07ff6b..081b1ee8516e 100644 --- a/include/linux/tc_ematch/tc_em_meta.h +++ b/include/linux/tc_ematch/tc_em_meta.h @@ -41,17 +41,14 @@ enum TCF_META_ID_LOADAVG_1, TCF_META_ID_LOADAVG_2, TCF_META_ID_DEV, - TCF_META_ID_INDEV, TCF_META_ID_PRIORITY, TCF_META_ID_PROTOCOL, - TCF_META_ID_SECURITY, /* obsolete */ TCF_META_ID_PKTTYPE, TCF_META_ID_PKTLEN, TCF_META_ID_DATALEN, TCF_META_ID_MACLEN, TCF_META_ID_NFMARK, TCF_META_ID_TCINDEX, - TCF_META_ID_TCVERDICT, TCF_META_ID_RTCLASSID, TCF_META_ID_RTIIF, TCF_META_ID_SK_FAMILY, diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index 179efb5bc9b3..a18b924743d9 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -27,17 +27,17 @@ * lvalue rvalue * +-----------+ +-----------+ * | type: INT | | type: INT | - * def | id: INDEV | | id: VALUE | + * def | id: DEV | | id: VALUE | * | data: | | data: 3 | * +-----------+ +-----------+ * | | - * ---> meta_ops[INT][INDEV](...) | + * ---> meta_ops[INT][DEV](...) | * | | * ----------- | * V V * +-----------+ +-----------+ * | type: INT | | type: INT | - * obj | id: INDEV | | id: VALUE | + * obj | id: DEV | | id: VALUE | * | data: 2 |<--data got filled out | data: 3 | * +-----------+ +-----------+ * | | @@ -170,16 +170,6 @@ META_COLLECTOR(var_dev) *err = var_dev(skb->dev, dst); } -META_COLLECTOR(int_indev) -{ - *err = int_dev(skb->input_dev, dst); -} - -META_COLLECTOR(var_indev) -{ - *err = var_dev(skb->input_dev, dst); -} - /************************************************************************** * skb attributes **************************************************************************/ @@ -235,13 +225,6 @@ META_COLLECTOR(int_tcindex) dst->value = skb->tc_index; } -#ifdef CONFIG_NET_CLS_ACT -META_COLLECTOR(int_tcverd) -{ - dst->value = skb->tc_verd; -} -#endif - /************************************************************************** * Routing **************************************************************************/ @@ -490,7 +473,6 @@ struct meta_ops static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [TCF_META_TYPE_VAR] = { [META_ID(DEV)] = META_FUNC(var_dev), - [META_ID(INDEV)] = META_FUNC(var_indev), [META_ID(SK_BOUND_IF)] = META_FUNC(var_sk_bound_if), }, [TCF_META_TYPE_INT] = { @@ -499,7 +481,6 @@ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [META_ID(LOADAVG_1)] = META_FUNC(int_loadavg_1), [META_ID(LOADAVG_2)] = META_FUNC(int_loadavg_2), [META_ID(DEV)] = META_FUNC(int_dev), - [META_ID(INDEV)] = META_FUNC(int_indev), [META_ID(PRIORITY)] = META_FUNC(int_priority), [META_ID(PROTOCOL)] = META_FUNC(int_protocol), [META_ID(PKTTYPE)] = META_FUNC(int_pkttype), @@ -510,9 +491,6 @@ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [META_ID(NFMARK)] = META_FUNC(int_nfmark), #endif [META_ID(TCINDEX)] = META_FUNC(int_tcindex), -#ifdef CONFIG_NET_CLS_ACT - [META_ID(TCVERDICT)] = META_FUNC(int_tcverd), -#endif #ifdef CONFIG_NET_CLS_ROUTE [META_ID(RTCLASSID)] = META_FUNC(int_rtclassid), #endif