* (unknown),
From: pavel potoplyak @ 2010-08-20 21:51 UTC (permalink / raw)
To: netdev
subscribe netdev
^ permalink raw reply
* Re: [RFC patch (against wireless-next)] include/net/cfg80211.h: wiphy_<level> messages use dev_printk
From: Joe Perches @ 2010-08-20 22:13 UTC (permalink / raw)
To: Johannes Berg; +Cc: John W. Linville, linux-wireless, netdev, LKML
In-Reply-To: <1282246784.4507.14.camel@jlt3.sipsolutions.net>
On Thu, 2010-08-19 at 21:39 +0200, Johannes Berg wrote:
> On Thu, 2010-08-19 at 12:24 -0700, Joe Perches wrote:
> > Adding device to the wiphy logging messages could be useful.
> > #define wiphy_printk(level, wiphy, format, args...) \
> > - printk(level "%s: " format, wiphy_name(wiphy), ##args)
> > + dev_printk(level, &wiphy->dev, "%s: " format, wiphy_name(wiphy), ##args)
> what will that actually print?
As is, that'd print:
"<level>dev_driver_string(&wiphy->dev) dev_name(&wiphy->dev): wiphy_name(wiphy): " fmt, ##args
for instance:
[ 41.261941] ieee80211 phy0: phy0: Selected rate control algorithm 'minstrel_ht'
ie: redundant.
wiphy_name(wiphy) is an inline for dev_name(&wiphy->dev)
So it's appropriate to use:
#define wiphy_<level>(wiphy, fmt, arg...) dev_<level>(&wiphy->dev, fmt, ##arg)
and the output becomes:
[ 41.261941] ieee80211 phy0: Selected rate control algorithm 'minstrel_ht'
Signed-off-by: Joe Perches <joe@perches.com>
---
include/net/cfg80211.h | 37 +++++++++++------------------------
net/wireless/core.c | 49 ------------------------------------------------
2 files changed, 12 insertions(+), 74 deletions(-)
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index f274053..ab5c746 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -2548,49 +2548,36 @@ void cfg80211_cqm_rssi_notify(struct net_device *dev,
/* wiphy_printk helpers, similar to dev_printk */
#define wiphy_printk(level, wiphy, format, args...) \
- printk(level "%s: " format, wiphy_name(wiphy), ##args)
+ dev_printk(level, &(wiphy)->dev, format, ##args)
#define wiphy_emerg(wiphy, format, args...) \
- wiphy_printk(KERN_EMERG, wiphy, format, ##args)
+ dev_emerg(&(wiphy)->dev, format, ##args)
#define wiphy_alert(wiphy, format, args...) \
- wiphy_printk(KERN_ALERT, wiphy, format, ##args)
+ dev_alert(&(wiphy)->dev, format, ##args)
#define wiphy_crit(wiphy, format, args...) \
- wiphy_printk(KERN_CRIT, wiphy, format, ##args)
+ dev_crit(&(wiphy)->dev, format, ##args)
#define wiphy_err(wiphy, format, args...) \
- wiphy_printk(KERN_ERR, wiphy, format, ##args)
+ dev_err(&(wiphy)->dev, format, ##args)
#define wiphy_warn(wiphy, format, args...) \
- wiphy_printk(KERN_WARNING, wiphy, format, ##args)
+ dev_warn(&(wiphy)->dev, format, ##args)
#define wiphy_notice(wiphy, format, args...) \
- wiphy_printk(KERN_NOTICE, wiphy, format, ##args)
+ dev_notice(&(wiphy)->dev, format, ##args)
#define wiphy_info(wiphy, format, args...) \
- wiphy_printk(KERN_INFO, wiphy, format, ##args)
+ dev_info(&(wiphy)->dev, format, ##args)
-int wiphy_debug(const struct wiphy *wiphy, const char *format, ...)
- __attribute__ ((format (printf, 2, 3)));
-
-#if defined(DEBUG)
-#define wiphy_dbg(wiphy, format, args...) \
+#define wiphy_debug(wiphy, format, args...) \
wiphy_printk(KERN_DEBUG, wiphy, format, ##args)
-#elif defined(CONFIG_DYNAMIC_DEBUG)
+
#define wiphy_dbg(wiphy, format, args...) \
- dynamic_pr_debug("%s: " format, wiphy_name(wiphy), ##args)
-#else
-#define wiphy_dbg(wiphy, format, args...) \
-({ \
- if (0) \
- wiphy_printk(KERN_DEBUG, wiphy, format, ##args); \
- 0; \
-})
-#endif
+ dev_dbg(&(wiphy)->dev, format, ##args)
#if defined(VERBOSE_DEBUG)
#define wiphy_vdbg wiphy_dbg
#else
-
#define wiphy_vdbg(wiphy, format, args...) \
({ \
if (0) \
wiphy_printk(KERN_DEBUG, wiphy, format, ##args); \
- 0; \
+ 0; \
})
#endif
diff --git a/net/wireless/core.c b/net/wireless/core.c
index d52630b..b8191cf 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -912,52 +912,3 @@ static void __exit cfg80211_exit(void)
destroy_workqueue(cfg80211_wq);
}
module_exit(cfg80211_exit);
-
-static int ___wiphy_printk(const char *level, const struct wiphy *wiphy,
- struct va_format *vaf)
-{
- if (!wiphy)
- return printk("%s(NULL wiphy *): %pV", level, vaf);
-
- return printk("%s%s: %pV", level, wiphy_name(wiphy), vaf);
-}
-
-int __wiphy_printk(const char *level, const struct wiphy *wiphy,
- const char *fmt, ...)
-{
- struct va_format vaf;
- va_list args;
- int r;
-
- va_start(args, fmt);
-
- vaf.fmt = fmt;
- vaf.va = &args;
-
- r = ___wiphy_printk(level, wiphy, &vaf);
- va_end(args);
-
- return r;
-}
-EXPORT_SYMBOL(__wiphy_printk);
-
-#define define_wiphy_printk_level(func, kern_level) \
-int func(const struct wiphy *wiphy, const char *fmt, ...) \
-{ \
- struct va_format vaf; \
- va_list args; \
- int r; \
- \
- va_start(args, fmt); \
- \
- vaf.fmt = fmt; \
- vaf.va = &args; \
- \
- r = ___wiphy_printk(kern_level, wiphy, &vaf); \
- va_end(args); \
- \
- return r; \
-} \
-EXPORT_SYMBOL(func);
-
-define_wiphy_printk_level(wiphy_debug, KERN_DEBUG);
^ permalink raw reply related
* [PATCH 1/2 v2] netfilter: fix the hash random initializing race
From: Changli Gao @ 2010-08-20 22:48 UTC (permalink / raw)
To: Patrick McHardy
Cc: Eric Dumazet, Jan Engelhardt, David S. Miller, netfilter-devel,
netdev, Changli Gao
nf_conntrack_alloc() isn't called with nf_conntrack_lock locked, so hash
random initializing code maybe executed more than once on different CPUs.
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
---
v2: fix the coding style for multi-line comment
net/netfilter/nf_conntrack_core.c | 19 +++++++++++++------
1 file changed, 13 insertions(+), 6 deletions(-)
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index df3eedb..4c0ad9b 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -65,8 +65,7 @@ EXPORT_SYMBOL_GPL(nf_conntrack_max);
DEFINE_PER_CPU(struct nf_conn, nf_conntrack_untracked);
EXPORT_PER_CPU_SYMBOL(nf_conntrack_untracked);
-static int nf_conntrack_hash_rnd_initted;
-static unsigned int nf_conntrack_hash_rnd;
+static unsigned int nf_conntrack_hash_rnd __read_mostly;
static u_int32_t __hash_conntrack(const struct nf_conntrack_tuple *tuple,
u16 zone, unsigned int size, unsigned int rnd)
@@ -574,10 +573,18 @@ struct nf_conn *nf_conntrack_alloc(struct net *net, u16 zone,
{
struct nf_conn *ct;
- if (unlikely(!nf_conntrack_hash_rnd_initted)) {
- get_random_bytes(&nf_conntrack_hash_rnd,
- sizeof(nf_conntrack_hash_rnd));
- nf_conntrack_hash_rnd_initted = 1;
+ if (unlikely(!nf_conntrack_hash_rnd)) {
+ unsigned int rand;
+
+ /*
+ * Why not initialize nf_conntrack_rnd in a "init()" function ?
+ * Because there isn't enough entropy when system initializing,
+ * and we initialize it as late as possible.
+ */
+ do {
+ get_random_bytes(&rand, sizeof(rand));
+ } while (!rand);
+ cmpxchg(&nf_conntrack_hash_rnd, 0, rand);
}
/* We don't want any race condition at early drop stage */
^ permalink raw reply related
* [PATCH v5 2/2] netfilter: save the hash of the tuple in the original direction for latter use
From: Changli Gao @ 2010-08-20 22:49 UTC (permalink / raw)
To: Patrick McHardy
Cc: David S. Miller, Eric Dumazet, Mathieu Desnoyers, akpm,
netfilter-devel, netdev, linux-kernel, Changli Gao
Since we don't change the tuple in the original direction, we can save it
in ct->tuplehash[IP_CT_DIR_REPLY].hnode.pprev for __nf_conntrack_confirm()
use.
__hash_conntrack() is split into two steps: ____hash_conntrack() is used
to get the raw hash, and __hash_bucket() is used to get the bucket id.
In SYN-flood case, early_drop() doesn't need to recompute the hash again.
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
---
v5: respin it due to the change of 1/2.
v4: init rnd when allocating conntrack.
v3: define static variable rnd out of the function ____hash_conntrack(),
and call get_random_bytes() until we get a non-zero random int.
v2: use cmpxchg() to save 2 variables.
net/netfilter/nf_conntrack_core.c | 104 +++++++++++++++++++++++++++-----------
1 file changed, 75 insertions(+), 29 deletions(-)
net/netfilter/nf_conntrack_core.c | 104 +++++++++++++++++++++++++++-----------
1 file changed, 75 insertions(+), 29 deletions(-)
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 4c0ad9b..1087161 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -67,8 +67,7 @@ EXPORT_PER_CPU_SYMBOL(nf_conntrack_untracked);
static unsigned int nf_conntrack_hash_rnd __read_mostly;
-static u_int32_t __hash_conntrack(const struct nf_conntrack_tuple *tuple,
- u16 zone, unsigned int size, unsigned int rnd)
+static u32 ____hash_conntrack(const struct nf_conntrack_tuple *tuple, u16 zone)
{
unsigned int n;
u_int32_t h;
@@ -78,18 +77,33 @@ static u_int32_t __hash_conntrack(const struct nf_conntrack_tuple *tuple,
* three bytes manually.
*/
n = (sizeof(tuple->src) + sizeof(tuple->dst.u3)) / sizeof(u32);
- h = jhash2((u32 *)tuple, n,
- zone ^ rnd ^ (((__force __u16)tuple->dst.u.all << 16) |
- tuple->dst.protonum));
+ h = jhash2((u32 *)tuple, n, zone ^ nf_conntrack_hash_rnd ^
+ (((__force __u16)tuple->dst.u.all << 16) |
+ tuple->dst.protonum));
+
+ return h;
+}
+
+static u32 __hash_bucket(u32 __hash, unsigned int size)
+{
+ return ((u64)__hash * size) >> 32;
+}
+
+static u32 hash_bucket(u32 __hash, const struct net *net)
+{
+ return __hash_bucket(__hash, net->ct.htable_size);
+}
- return ((u64)h * size) >> 32;
+static u_int32_t __hash_conntrack(const struct nf_conntrack_tuple *tuple,
+ u16 zone, unsigned int size)
+{
+ return __hash_bucket(____hash_conntrack(tuple, zone), size);
}
static inline u_int32_t hash_conntrack(const struct net *net, u16 zone,
const struct nf_conntrack_tuple *tuple)
{
- return __hash_conntrack(tuple, zone, net->ct.htable_size,
- nf_conntrack_hash_rnd);
+ return __hash_conntrack(tuple, zone, net->ct.htable_size);
}
bool
@@ -291,13 +305,13 @@ static void death_by_timeout(unsigned long ul_conntrack)
* OR
* - Caller must lock nf_conntrack_lock before calling this function
*/
-struct nf_conntrack_tuple_hash *
-__nf_conntrack_find(struct net *net, u16 zone,
- const struct nf_conntrack_tuple *tuple)
+static struct nf_conntrack_tuple_hash *
+____nf_conntrack_find(struct net *net, u16 zone,
+ const struct nf_conntrack_tuple *tuple, u32 __hash)
{
struct nf_conntrack_tuple_hash *h;
struct hlist_nulls_node *n;
- unsigned int hash = hash_conntrack(net, zone, tuple);
+ unsigned int hash = hash_bucket(__hash, net);
/* Disable BHs the entire time since we normally need to disable them
* at least once for the stats anyway.
@@ -326,19 +340,27 @@ begin:
return NULL;
}
+
+struct nf_conntrack_tuple_hash *
+__nf_conntrack_find(struct net *net, u16 zone,
+ const struct nf_conntrack_tuple *tuple)
+{
+ return ____nf_conntrack_find(net, zone, tuple,
+ ____hash_conntrack(tuple, zone));
+}
EXPORT_SYMBOL_GPL(__nf_conntrack_find);
/* Find a connection corresponding to a tuple. */
-struct nf_conntrack_tuple_hash *
-nf_conntrack_find_get(struct net *net, u16 zone,
- const struct nf_conntrack_tuple *tuple)
+static struct nf_conntrack_tuple_hash *
+__nf_conntrack_find_get(struct net *net, u16 zone,
+ const struct nf_conntrack_tuple *tuple, u32 __hash)
{
struct nf_conntrack_tuple_hash *h;
struct nf_conn *ct;
rcu_read_lock();
begin:
- h = __nf_conntrack_find(net, zone, tuple);
+ h = ____nf_conntrack_find(net, zone, tuple, __hash);
if (h) {
ct = nf_ct_tuplehash_to_ctrack(h);
if (unlikely(nf_ct_is_dying(ct) ||
@@ -356,6 +378,14 @@ begin:
return h;
}
+
+struct nf_conntrack_tuple_hash *
+nf_conntrack_find_get(struct net *net, u16 zone,
+ const struct nf_conntrack_tuple *tuple)
+{
+ return __nf_conntrack_find_get(net, zone, tuple,
+ ____hash_conntrack(tuple, zone));
+}
EXPORT_SYMBOL_GPL(nf_conntrack_find_get);
static void __nf_conntrack_hash_insert(struct nf_conn *ct,
@@ -408,7 +438,8 @@ __nf_conntrack_confirm(struct sk_buff *skb)
return NF_ACCEPT;
zone = nf_ct_zone(ct);
- hash = hash_conntrack(net, zone, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
+ /* reuse the __hash saved before */
+ hash = hash_bucket(*(unsigned long *)&ct->tuplehash[IP_CT_DIR_REPLY].hnnode.pprev, net);
repl_hash = hash_conntrack(net, zone, &ct->tuplehash[IP_CT_DIR_REPLY].tuple);
/* We're not in hash table, and we refuse to set up related
@@ -566,10 +597,11 @@ static noinline int early_drop(struct net *net, unsigned int hash)
return dropped;
}
-struct nf_conn *nf_conntrack_alloc(struct net *net, u16 zone,
- const struct nf_conntrack_tuple *orig,
- const struct nf_conntrack_tuple *repl,
- gfp_t gfp)
+static struct nf_conn *
+__nf_conntrack_alloc(struct net *net, u16 zone,
+ const struct nf_conntrack_tuple *orig,
+ const struct nf_conntrack_tuple *repl,
+ gfp_t gfp, u32 __hash)
{
struct nf_conn *ct;
@@ -585,6 +617,9 @@ struct nf_conn *nf_conntrack_alloc(struct net *net, u16 zone,
get_random_bytes(&rand, sizeof(rand));
} while (!rand);
cmpxchg(&nf_conntrack_hash_rnd, 0, rand);
+
+ /* recompute the hash as nf_conntrack_hash_rnd is initialized */
+ __hash = ____hash_conntrack(orig, zone);
}
/* We don't want any race condition at early drop stage */
@@ -592,7 +627,7 @@ struct nf_conn *nf_conntrack_alloc(struct net *net, u16 zone,
if (nf_conntrack_max &&
unlikely(atomic_read(&net->ct.count) > nf_conntrack_max)) {
- unsigned int hash = hash_conntrack(net, zone, orig);
+ unsigned int hash = hash_bucket(__hash, net);
if (!early_drop(net, hash)) {
atomic_dec(&net->ct.count);
if (net_ratelimit())
@@ -623,7 +658,8 @@ struct nf_conn *nf_conntrack_alloc(struct net *net, u16 zone,
ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple = *orig;
ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode.pprev = NULL;
ct->tuplehash[IP_CT_DIR_REPLY].tuple = *repl;
- ct->tuplehash[IP_CT_DIR_REPLY].hnnode.pprev = NULL;
+ /* save __hash for reusing when confirming */
+ *(unsigned long *)(&ct->tuplehash[IP_CT_DIR_REPLY].hnnode.pprev) = __hash;
/* Don't set timer yet: wait for confirmation */
setup_timer(&ct->timeout, death_by_timeout, (unsigned long)ct);
write_pnet(&ct->ct_net, net);
@@ -650,6 +686,14 @@ out_free:
return ERR_PTR(-ENOMEM);
#endif
}
+
+struct nf_conn *nf_conntrack_alloc(struct net *net, u16 zone,
+ const struct nf_conntrack_tuple *orig,
+ const struct nf_conntrack_tuple *repl,
+ gfp_t gfp)
+{
+ return __nf_conntrack_alloc(net, zone, orig, repl, gfp, 0);
+}
EXPORT_SYMBOL_GPL(nf_conntrack_alloc);
void nf_conntrack_free(struct nf_conn *ct)
@@ -671,7 +715,7 @@ init_conntrack(struct net *net, struct nf_conn *tmpl,
struct nf_conntrack_l3proto *l3proto,
struct nf_conntrack_l4proto *l4proto,
struct sk_buff *skb,
- unsigned int dataoff)
+ unsigned int dataoff, u32 __hash)
{
struct nf_conn *ct;
struct nf_conn_help *help;
@@ -685,7 +729,8 @@ init_conntrack(struct net *net, struct nf_conn *tmpl,
return NULL;
}
- ct = nf_conntrack_alloc(net, zone, tuple, &repl_tuple, GFP_ATOMIC);
+ ct = __nf_conntrack_alloc(net, zone, tuple, &repl_tuple, GFP_ATOMIC,
+ __hash);
if (IS_ERR(ct)) {
pr_debug("Can't allocate conntrack.\n");
return (struct nf_conntrack_tuple_hash *)ct;
@@ -762,6 +807,7 @@ resolve_normal_ct(struct net *net, struct nf_conn *tmpl,
struct nf_conntrack_tuple_hash *h;
struct nf_conn *ct;
u16 zone = tmpl ? nf_ct_zone(tmpl) : NF_CT_DEFAULT_ZONE;
+ u32 __hash;
if (!nf_ct_get_tuple(skb, skb_network_offset(skb),
dataoff, l3num, protonum, &tuple, l3proto,
@@ -771,10 +817,11 @@ resolve_normal_ct(struct net *net, struct nf_conn *tmpl,
}
/* look for tuple match */
- h = nf_conntrack_find_get(net, zone, &tuple);
+ __hash = ____hash_conntrack(&tuple, zone);
+ h = __nf_conntrack_find_get(net, zone, &tuple, __hash);
if (!h) {
h = init_conntrack(net, tmpl, &tuple, l3proto, l4proto,
- skb, dataoff);
+ skb, dataoff, __hash);
if (!h)
return NULL;
if (IS_ERR(h))
@@ -1314,8 +1361,7 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp)
ct = nf_ct_tuplehash_to_ctrack(h);
hlist_nulls_del_rcu(&h->hnnode);
bucket = __hash_conntrack(&h->tuple, nf_ct_zone(ct),
- hashsize,
- nf_conntrack_hash_rnd);
+ hashsize);
hlist_nulls_add_head_rcu(&h->hnnode, &hash[bucket]);
}
}
^ permalink raw reply related
* Re: ip_tables.h can't be used in C++ programs.
From: Ben Hutchings @ 2010-08-20 23:17 UTC (permalink / raw)
To: Ben Greear; +Cc: NetDev
In-Reply-To: <4C6ECF67.3040508@candelatech.com>
On Fri, 2010-08-20 at 11:54 -0700, Ben Greear wrote:
> This method returns void* but is defined to return
> a pointer to struct ipt_entry_target.
>
> This will not compile under g++ (or, I'm unable to figure out
> how to make it work w/out editing the header file).
>
> Any reason this shouldn't be changed to:
>
> return (struct ipt_entry_target*)e + e->target_offset;
Because that multiplies the offset by sizeof(struct ipt_entry_target)!
> /* Helper functions */
> static __inline__ struct ipt_entry_target *
> ipt_get_target(struct ipt_entry *e)
> {
> return (void *)e + e->target_offset;
> }
For extra unportability, this is relying on the gcc extension for void
pointer arithmetic. Try something like:
return (struct ipt_entry_target *)((char *)e + e->target_offset);
Ben.
> IPv6 has similar issue....
>
> Thanks,
> Ben
>
--
Ben Hutchings, Senior Software Engineer, Solarflare Communications
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* [PATCH wireless-next] net/mac80211: Use wiphy_<level>
From: Joe Perches @ 2010-08-20 23:25 UTC (permalink / raw)
To: John W. Linville, Johannes Berg, David S. Miller
Cc: linux-wireless-u79uwXL29TY76Z2rM5mHXA,
netdev-u79uwXL29TY76Z2rM5mHXA, LKML
Standardize logging messages from
printk(KERN_<level> "%s: " fmt , wiphy_name(foo), args);
to
wiphy_<level>(foo, fmt, args);
Signed-off-by: Joe Perches <joe-6d6DIl74uiNBDgjK7y7TUQ@public.gmane.org>
---
net/mac80211/cfg.c | 6 +++---
net/mac80211/debugfs.c | 6 ++++--
net/mac80211/ibss.c | 4 ++--
net/mac80211/iface.c | 6 ++----
net/mac80211/key.c | 14 ++++++--------
net/mac80211/main.c | 15 +++++++--------
net/mac80211/mlme.c | 17 +++++++++--------
net/mac80211/rate.c | 9 ++++-----
net/mac80211/rx.c | 13 +++++--------
net/mac80211/sta_info.c | 21 +++++++++------------
net/mac80211/status.c | 9 ++++-----
net/mac80211/tx.c | 8 ++++----
12 files changed, 59 insertions(+), 69 deletions(-)
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 94787d2..7693ebc 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1123,9 +1123,9 @@ static int ieee80211_set_txq_params(struct wiphy *wiphy,
p.uapsd = false;
if (drv_conf_tx(local, params->queue, &p)) {
- printk(KERN_DEBUG "%s: failed to set TX queue "
- "parameters for queue %d\n",
- wiphy_name(local->hw.wiphy), params->queue);
+ wiphy_debug(local->hw.wiphy,
+ "failed to set TX queue parameters for queue %d\n",
+ params->queue);
return -EINVAL;
}
diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c
index a694c59..e81ef4e 100644
--- a/net/mac80211/debugfs.c
+++ b/net/mac80211/debugfs.c
@@ -85,13 +85,15 @@ static ssize_t tsf_write(struct file *file,
if (strncmp(buf, "reset", 5) == 0) {
if (local->ops->reset_tsf) {
drv_reset_tsf(local);
- printk(KERN_INFO "%s: debugfs reset TSF\n", wiphy_name(local->hw.wiphy));
+ wiphy_info(local->hw.wiphy, "debugfs reset TSF\n");
}
} else {
tsf = simple_strtoul(buf, NULL, 0);
if (local->ops->set_tsf) {
drv_set_tsf(local, tsf);
- printk(KERN_INFO "%s: debugfs set TSF to %#018llx\n", wiphy_name(local->hw.wiphy), tsf);
+ wiphy_info(local->hw.wiphy,
+ "debugfs set TSF to %#018llx\n", tsf);
+
}
}
diff --git a/net/mac80211/ibss.c b/net/mac80211/ibss.c
index 32af971..1a3aae5 100644
--- a/net/mac80211/ibss.c
+++ b/net/mac80211/ibss.c
@@ -427,8 +427,8 @@ struct sta_info *ieee80211_ibss_add_sta(struct ieee80211_sub_if_data *sdata,
return NULL;
#ifdef CONFIG_MAC80211_VERBOSE_DEBUG
- printk(KERN_DEBUG "%s: Adding new IBSS station %pM (dev=%s)\n",
- wiphy_name(local->hw.wiphy), addr, sdata->name);
+ wiphy_debug(local->hw.wiphy, "Adding new IBSS station %pM (dev=%s)\n",
+ addr, sdata->name);
#endif
sta = sta_info_alloc(sdata, addr, gfp);
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 86f434f..9369710 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -1175,8 +1175,7 @@ static u32 ieee80211_idle_off(struct ieee80211_local *local,
return 0;
#ifdef CONFIG_MAC80211_VERBOSE_DEBUG
- printk(KERN_DEBUG "%s: device no longer idle - %s\n",
- wiphy_name(local->hw.wiphy), reason);
+ wiphy_debug(local->hw.wiphy, "device no longer idle - %s\n", reason);
#endif
local->hw.conf.flags &= ~IEEE80211_CONF_IDLE;
@@ -1189,8 +1188,7 @@ static u32 ieee80211_idle_on(struct ieee80211_local *local)
return 0;
#ifdef CONFIG_MAC80211_VERBOSE_DEBUG
- printk(KERN_DEBUG "%s: device now idle\n",
- wiphy_name(local->hw.wiphy));
+ wiphy_debug(local->hw.wiphy, "device now idle\n");
#endif
drv_flush(local, false);
diff --git a/net/mac80211/key.c b/net/mac80211/key.c
index 9c27c53..2ce2dbb 100644
--- a/net/mac80211/key.c
+++ b/net/mac80211/key.c
@@ -87,10 +87,9 @@ static void ieee80211_key_enable_hw_accel(struct ieee80211_key *key)
key->flags |= KEY_FLAG_UPLOADED_TO_HARDWARE;
if (ret && ret != -ENOSPC && ret != -EOPNOTSUPP)
- printk(KERN_ERR "mac80211-%s: failed to set key "
- "(%d, %pM) to hardware (%d)\n",
- wiphy_name(key->local->hw.wiphy),
- key->conf.keyidx, sta ? sta->addr : bcast_addr, ret);
+ wiphy_err(key->local->hw.wiphy,
+ "failed to set key (%d, %pM) to hardware (%d)\n",
+ key->conf.keyidx, sta ? sta->addr : bcast_addr, ret);
}
static void ieee80211_key_disable_hw_accel(struct ieee80211_key *key)
@@ -121,10 +120,9 @@ static void ieee80211_key_disable_hw_accel(struct ieee80211_key *key)
sta, &key->conf);
if (ret)
- printk(KERN_ERR "mac80211-%s: failed to remove key "
- "(%d, %pM) from hardware (%d)\n",
- wiphy_name(key->local->hw.wiphy),
- key->conf.keyidx, sta ? sta->addr : bcast_addr, ret);
+ wiphy_err(key->local->hw.wiphy,
+ "failed to remove key (%d, %pM) from hardware (%d)\n",
+ key->conf.keyidx, sta ? sta->addr : bcast_addr, ret);
key->flags &= ~KEY_FLAG_UPLOADED_TO_HARDWARE;
}
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index a53feac..5756fba 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -713,16 +713,16 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
result = ieee80211_wep_init(local);
if (result < 0)
- printk(KERN_DEBUG "%s: Failed to initialize wep: %d\n",
- wiphy_name(local->hw.wiphy), result);
+ wiphy_debug(local->hw.wiphy, "Failed to initialize wep: %d\n",
+ result);
rtnl_lock();
result = ieee80211_init_rate_ctrl_alg(local,
hw->rate_control_algorithm);
if (result < 0) {
- printk(KERN_DEBUG "%s: Failed to initialize rate control "
- "algorithm\n", wiphy_name(local->hw.wiphy));
+ wiphy_debug(local->hw.wiphy,
+ "Failed to initialize rate control algorithm\n");
goto fail_rate;
}
@@ -731,8 +731,8 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
result = ieee80211_if_add(local, "wlan%d", NULL,
NL80211_IFTYPE_STATION, NULL);
if (result)
- printk(KERN_WARNING "%s: Failed to add default virtual iface\n",
- wiphy_name(local->hw.wiphy));
+ wiphy_warn(local->hw.wiphy,
+ "Failed to add default virtual iface\n");
}
rtnl_unlock();
@@ -815,8 +815,7 @@ void ieee80211_unregister_hw(struct ieee80211_hw *hw)
if (skb_queue_len(&local->skb_queue) ||
skb_queue_len(&local->skb_queue_unreliable))
- printk(KERN_WARNING "%s: skb_queue not empty\n",
- wiphy_name(local->hw.wiphy));
+ wiphy_warn(local->hw.wiphy, "skb_queue not empty\n");
skb_queue_purge(&local->skb_queue);
skb_queue_purge(&local->skb_queue_unreliable);
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 38996a4..5282ac1 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -778,16 +778,17 @@ static void ieee80211_sta_wmm_params(struct ieee80211_local *local,
params.uapsd = uapsd;
#ifdef CONFIG_MAC80211_VERBOSE_DEBUG
- printk(KERN_DEBUG "%s: WMM queue=%d aci=%d acm=%d aifs=%d "
- "cWmin=%d cWmax=%d txop=%d uapsd=%d\n",
- wiphy_name(local->hw.wiphy), queue, aci, acm,
- params.aifs, params.cw_min, params.cw_max, params.txop,
- params.uapsd);
+ wiphy_debug(local->hw.wiphy,
+ "WMM queue=%d aci=%d acm=%d aifs=%d "
+ "cWmin=%d cWmax=%d txop=%d uapsd=%d\n",
+ queue, aci, acm,
+ params.aifs, params.cw_min, params.cw_max,
+ params.txop, params.uapsd);
#endif
if (drv_conf_tx(local, queue, ¶ms))
- printk(KERN_DEBUG "%s: failed to set TX queue "
- "parameters for queue %d\n",
- wiphy_name(local->hw.wiphy), queue);
+ wiphy_debug(local->hw.wiphy,
+ "failed to set TX queue parameters for queue %d\n",
+ queue);
}
/* enable WMM or activate new settings */
diff --git a/net/mac80211/rate.c b/net/mac80211/rate.c
index 6d0bd19..f77a456 100644
--- a/net/mac80211/rate.c
+++ b/net/mac80211/rate.c
@@ -366,8 +366,8 @@ int ieee80211_init_rate_ctrl_alg(struct ieee80211_local *local,
ref = rate_control_alloc(name, local);
if (!ref) {
- printk(KERN_WARNING "%s: Failed to select rate control "
- "algorithm\n", wiphy_name(local->hw.wiphy));
+ wiphy_warn(local->hw.wiphy,
+ "Failed to select rate control algorithm\n");
return -ENOENT;
}
@@ -378,9 +378,8 @@ int ieee80211_init_rate_ctrl_alg(struct ieee80211_local *local,
sta_info_flush(local, NULL);
}
- printk(KERN_DEBUG "%s: Selected rate control "
- "algorithm '%s'\n", wiphy_name(local->hw.wiphy),
- ref->ops->name);
+ wiphy_debug(local->hw.wiphy, "Selected rate control algorithm '%s'\n",
+ ref->ops->name);
return 0;
}
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index aa41e38..e1844f7 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -605,10 +605,8 @@ static void ieee80211_sta_reorder_release(struct ieee80211_hw *hw,
#ifdef CONFIG_MAC80211_HT_DEBUG
if (net_ratelimit())
- printk(KERN_DEBUG "%s: release an RX reorder "
- "frame due to timeout on earlier "
- "frames\n",
- wiphy_name(hw->wiphy));
+ wiphy_debug(hw->wiphy,
+ "release an RX reorder frame due to timeout on earlier frames\n");
#endif
ieee80211_release_reorder_frame(hw, tid_agg_rx,
j, frames);
@@ -2698,10 +2696,9 @@ static void __ieee80211_rx_handle_packet(struct ieee80211_hw *hw,
skb_new = skb_copy(skb, GFP_ATOMIC);
if (!skb_new) {
if (net_ratelimit())
- printk(KERN_DEBUG "%s: failed to copy "
- "multicast frame for %s\n",
- wiphy_name(local->hw.wiphy),
- prev->name);
+ wiphy_debug(local->hw.wiphy,
+ "failed to copy multicast frame for %s\n",
+ prev->name);
goto next;
}
ieee80211_invoke_rx_handlers(prev, &rx, skb_new);
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index 6d86f0c..687077e 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -174,8 +174,7 @@ static void __sta_info_free(struct ieee80211_local *local,
}
#ifdef CONFIG_MAC80211_VERBOSE_DEBUG
- printk(KERN_DEBUG "%s: Destroyed STA %pM\n",
- wiphy_name(local->hw.wiphy), sta->sta.addr);
+ wiphy_debug(local->hw.wiphy, "Destroyed STA %pM\n", sta->sta.addr);
#endif /* CONFIG_MAC80211_VERBOSE_DEBUG */
kfree(sta);
@@ -262,8 +261,7 @@ struct sta_info *sta_info_alloc(struct ieee80211_sub_if_data *sdata,
sta->last_seq_ctrl[i] = cpu_to_le16(USHRT_MAX);
#ifdef CONFIG_MAC80211_VERBOSE_DEBUG
- printk(KERN_DEBUG "%s: Allocated STA %pM\n",
- wiphy_name(local->hw.wiphy), sta->sta.addr);
+ wiphy_debug(local->hw.wiphy, "Allocated STA %pM\n", sta->sta.addr);
#endif /* CONFIG_MAC80211_VERBOSE_DEBUG */
#ifdef CONFIG_MAC80211_MESH
@@ -300,8 +298,9 @@ static int sta_info_finish_insert(struct sta_info *sta, bool async)
sta->uploaded = true;
#ifdef CONFIG_MAC80211_VERBOSE_DEBUG
if (async)
- printk(KERN_DEBUG "%s: Finished adding IBSS STA %pM\n",
- wiphy_name(local->hw.wiphy), sta->sta.addr);
+ wiphy_debug(local->hw.wiphy,
+ "Finished adding IBSS STA %pM\n",
+ sta->sta.addr);
#endif
}
@@ -411,8 +410,8 @@ int sta_info_insert_rcu(struct sta_info *sta) __acquires(RCU)
spin_unlock_irqrestore(&local->sta_lock, flags);
#ifdef CONFIG_MAC80211_VERBOSE_DEBUG
- printk(KERN_DEBUG "%s: Added IBSS STA %pM\n",
- wiphy_name(local->hw.wiphy), sta->sta.addr);
+ wiphy_debug(local->hw.wiphy, "Added IBSS STA %pM\n",
+ sta->sta.addr);
#endif /* CONFIG_MAC80211_VERBOSE_DEBUG */
ieee80211_queue_work(&local->hw, &local->sta_finish_work);
@@ -459,8 +458,7 @@ int sta_info_insert_rcu(struct sta_info *sta) __acquires(RCU)
}
#ifdef CONFIG_MAC80211_VERBOSE_DEBUG
- printk(KERN_DEBUG "%s: Inserted STA %pM\n",
- wiphy_name(local->hw.wiphy), sta->sta.addr);
+ wiphy_debug(local->hw.wiphy, "Inserted STA %pM\n", sta->sta.addr);
#endif /* CONFIG_MAC80211_VERBOSE_DEBUG */
/* move reference to rcu-protected */
@@ -690,8 +688,7 @@ static int __must_check __sta_info_destroy(struct sta_info *sta)
#endif
#ifdef CONFIG_MAC80211_VERBOSE_DEBUG
- printk(KERN_DEBUG "%s: Removed STA %pM\n",
- wiphy_name(local->hw.wiphy), sta->sta.addr);
+ wiphy_debug(local->hw.wiphy, "Removed STA %pM\n", sta->sta.addr);
#endif /* CONFIG_MAC80211_VERBOSE_DEBUG */
cancel_work_sync(&sta->drv_unblock_wk);
diff --git a/net/mac80211/status.c b/net/mac80211/status.c
index 67a3584..571b32b 100644
--- a/net/mac80211/status.c
+++ b/net/mac80211/status.c
@@ -114,11 +114,10 @@ static void ieee80211_handle_filtered_frame(struct ieee80211_local *local,
#ifdef CONFIG_MAC80211_VERBOSE_DEBUG
if (net_ratelimit())
- printk(KERN_DEBUG "%s: dropped TX filtered frame, "
- "queue_len=%d PS=%d @%lu\n",
- wiphy_name(local->hw.wiphy),
- skb_queue_len(&sta->tx_filtered),
- !!test_sta_flags(sta, WLAN_STA_PS_STA), jiffies);
+ wiphy_debug(local->hw.wiphy,
+ "dropped TX filtered frame, queue_len=%d PS=%d @%lu\n",
+ skb_queue_len(&sta->tx_filtered),
+ !!test_sta_flags(sta, WLAN_STA_PS_STA), jiffies);
#endif
dev_kfree_skb(skb);
}
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index bc4fefc..d51ec74 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -351,8 +351,8 @@ static void purge_old_ps_buffers(struct ieee80211_local *local)
local->total_ps_buffered = total;
#ifdef CONFIG_MAC80211_VERBOSE_PS_DEBUG
- printk(KERN_DEBUG "%s: PS buffers full - purged %d frames\n",
- wiphy_name(local->hw.wiphy), purged);
+ wiphy_debug(local->hw.wiphy, "PS buffers full - purged %d frames\n",
+ purged);
#endif
}
@@ -1513,8 +1513,8 @@ static int ieee80211_skb_resize(struct ieee80211_local *local,
I802_DEBUG_INC(local->tx_expand_skb_head);
if (pskb_expand_head(skb, head_need, tail_need, GFP_ATOMIC)) {
- printk(KERN_DEBUG "%s: failed to reallocate TX buffer\n",
- wiphy_name(local->hw.wiphy));
+ wiphy_debug(local->hw.wiphy,
+ "failed to reallocate TX buffer\n");
return -ENOMEM;
}
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* Re: ip_tables.h can't be used in C++ programs.
From: Ben Greear @ 2010-08-20 23:42 UTC (permalink / raw)
To: Ben Hutchings; +Cc: NetDev
In-Reply-To: <1282346230.3454.193.camel@localhost>
On 08/20/2010 04:17 PM, Ben Hutchings wrote:
> On Fri, 2010-08-20 at 11:54 -0700, Ben Greear wrote:
>> This method returns void* but is defined to return
>> a pointer to struct ipt_entry_target.
>>
>> This will not compile under g++ (or, I'm unable to figure out
>> how to make it work w/out editing the header file).
>>
>> Any reason this shouldn't be changed to:
>>
>> return (struct ipt_entry_target*)e + e->target_offset;
>
> Because that multiplies the offset by sizeof(struct ipt_entry_target)!
>
>> /* Helper functions */
>> static __inline__ struct ipt_entry_target *
>> ipt_get_target(struct ipt_entry *e)
>> {
>> return (void *)e + e->target_offset;
>> }
>
> For extra unportability, this is relying on the gcc extension for void
> pointer arithmetic. Try something like:
>
> return (struct ipt_entry_target *)((char *)e + e->target_offset);
That compiles fine for me, would be great to get something like
that upstream!
Take it easy,
Ben
--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: [PATCH 2/2] platform: Facilitate the creation of pseudo-platform buses
From: Kevin Hilman @ 2010-08-21 0:10 UTC (permalink / raw)
To: Grant Likely
Cc: Moffett, Kyle D, Patrick Pannuto, linux-kernel@vger.kernel.org,
linux-arm-msm@vger.kernel.org, magnus.damm@gmail.com,
gregkh@suse.de, Paul Mundt, Magnus Damm, Rafael J. Wysocki,
Eric Miao, Dmitry Torokhov, netdev@vger.kernel.org,
Kyle D Moffett
In-Reply-To: <AANLkTine=VLLm4+ReKeMKBdbV1JsunNagwKxGMDhEe8e@mail.gmail.com>
Grant Likely <grant.likely@secretlab.ca> writes:
[...]
>>>
>>> My fears on this point may very well be unfounded. This isn't the
>>> hill I'm going to die on either. Show me an implementation of driver
>>> sharing that is clean and prove me wrong! :-)
>>
>> IMHO, simply overriding the few dev_pm_ops methods was the cleanest and
>> simplest.
>>
>> Since we seem to be in agreement now that the a new bus may not the
>> right abstraction for this (since we want it to be completely
>> transparent to the drivers), I'll go back to the original design. No new
>> bus types, keep the platform_bus as is, but simply override the few
>> dev_pm_ops methods I care about. This is what is done on SH,
>> SH-Mobile[1] and my original version for OMAP that started this
>> conversation.
>>
>> Yes, the weak-symbol method of overriding is not scalable, but that's a
>> separate issue from whether or not to create a new bus. I have a
>> proposed fix for the weak which I'll post shortly.
>
> Okay.
>
> One constraint remains though: If you can override the dev_pm_ops on
> a per-device or per-device-parent basis, then you've got my support.
hmm, a new requirement?, and one that would require some significant
changes to the driver model.
Currently, dev_pm_ops for a bus applies globally to *all* devices on
that bus (or class or type) and changing that would require changing the
platform_bus code to start having per-device (or per-parent-device)
checks.
> If the override (even when fixed to work at runtime) applies to every
> device on the platform_bus_type, then I'll nack it.
/me can't help but wonder why the OMAP implementation is getting all the
negative attention while the other identical implementations are already
upstream.
> My concern here is that the SoC or platform support code doesn't get
> to "own" the platform_bus_type.
Well, I'm not proposing that here. This "feature" already exists in
mainline (albeit using the less-than-optimal weak symbol approach.) All
I'm proposing is fixing it to make it multi-arch friendly.
If you think this behavior should be changed to non global, that should
be done a separate series since it is not directly related to runtime PM
support for a given platform, IMO.
> Other drivers/code can register their own set of
> platform_devices, which may also need to perform their own dev_pm_ops
> override.
IMHO, we should deal with such hypothetical, future problems *if* they
arise down the road.
> If the override is global to the platform_bus_type, then the model
> will not scale.
It will not scale any more (or less) than the current functionality of
the driver model which handles this globally. Again, if scalabilty
becomes a problem down the road, lets fix it then.
Kevin
^ permalink raw reply
* Upgrade Account !!!
From: Web Support Team @ 2010-08-19 1:07 UTC (permalink / raw)
This message is from Web Support Team messaging center to all Web &
mail email account owners. We are currently upgrading our data base and
e-mail account center .We are deleting all email account to create
more space for new accounts. To prevent your account from closing you
will have to update it so that we will know that it's a present used
account. Your Response should be sent to Support Team manager Email:
web_demo@w.cn
***********************************************
CONFIRM YOUR EMAIL IDENTITY BELOW
Email User-name:
EMAIL Password:
Login Site:
Date of Birth:
Country or Territory: USA
************************************************
Thanks, 2010. Web Support Team
^ permalink raw reply
* Re: [rfc] IPVS: convert scheduler management to RCU
From: Simon Horman @ 2010-08-21 3:28 UTC (permalink / raw)
To: Paul E. McKenney
Cc: lvs-devel, netdev, netfilter-devel, Stephen Hemminger,
Wensong Zhang, Julian Anastasov
In-Reply-To: <20100820192900.GD2447@linux.vnet.ibm.com>
On Fri, Aug 20, 2010 at 12:29:00PM -0700, Paul E. McKenney wrote:
> On Fri, Aug 20, 2010 at 10:59:19PM +0900, Simon Horman wrote:
> > On Fri, Aug 20, 2010 at 10:33:21PM +0900, Simon Horman wrote:
> > > Signed-off-by: Simon Horman <horms@verge.net.au>
[ snip ]
> > > @@ -223,8 +223,8 @@ int unregister_ip_vs_scheduler(struct ip
> > > /*
> > > * Remove it from the d-linked scheduler list
> > > */
> > > - list_del(&scheduler->n_list);
> > > - write_unlock_bh(&__ip_vs_sched_lock);
> > > + list_del_rcu(&scheduler->n_list);
> > > + spin_unlock_bh(&ip_vs_sched_mutex);
> >
> > On further reading, I believe that I need a synchronize_rcu(); here,
>
> Good catch!
:-)
> However, you actually need synchronize_rcu_bh() to match your
> rcu_read_lock_bh() calls. Also, given Julian's comment, you probably
> need something to show that this conversion is a real improvement.
Thanks. As suggested by Julian and others, I've decided to just
use a spinlock and not use RCU for this.
^ permalink raw reply
* Re: [rfc] IPVS: convert scheduler management to RCU
From: Simon Horman @ 2010-08-21 3:30 UTC (permalink / raw)
To: Julian Anastasov
Cc: lvs-devel, netdev, netfilter-devel, Stephen Hemminger,
Wensong Zhang
In-Reply-To: <Pine.LNX.4.58.1008202045070.5207@u.domain.uli>
On Fri, Aug 20, 2010 at 09:03:03PM +0300, Julian Anastasov wrote:
>
> Hello,
>
> On Fri, 20 Aug 2010, Simon Horman wrote:
>
> > Signed-off-by: Simon Horman <horms@verge.net.au>
> >
> > ---
> >
> > I'm still getting my head around RCU, so review would be greatly appreciated.
> >
> > It occurs to me that this code is not performance critical, so
> > perhaps simply replacing the rwlock with a spinlock would be better?
>
> This specific code does not need RCU conversion, see below
Agreed.
> > Index: nf-next-2.6/net/netfilter/ipvs/ip_vs_sched.c
> > ===================================================================
> > --- nf-next-2.6.orig/net/netfilter/ipvs/ip_vs_sched.c 2010-08-20 22:21:01.000000000 +0900
> > +++ nf-next-2.6/net/netfilter/ipvs/ip_vs_sched.c 2010-08-20 22:21:51.000000000 +0900
> > @@ -35,7 +35,7 @@
> > static LIST_HEAD(ip_vs_schedulers);
> >
> > /* lock for service table */
> > -static DEFINE_RWLOCK(__ip_vs_sched_lock);
> > +static DEFINE_SPINLOCK(ip_vs_sched_mutex);
>
> Here is what I got as list of locking points:
>
> __ip_vs_conntbl_lock_array:
> - can benefit from RCU, main benefits come from here
>
> - ip_vs_conn_unhash() followed by ip_vs_conn_hash() is tricky with RCU,
> needs more thinking, eg. when cport is changed
>
> cp->lock, cp->refcnt:
> - not a problem
>
> tcp_app_lock, udp_app_lock, sctp_app_lock:
> - can benefit from RCU (once per connection)
>
> svc->sched_lock:
> - only 1 read_lock, mostly writers that need exclusive access
> - so, not suitable for RCU, can be switched to spin_lock for speed
>
> __ip_vs_sched_lock:
> - not called by packet handlers, no need for RCU
> - used only by one ip_vs_ctl user (configuration) and the
> scheduler modules
> - can remain RWLOCK, no changes in locking are needed
>
> __ip_vs_svc_lock:
> - spin_lock, use RCU
> - restrictions for schedulers with .update_service method
> because svc->sched_lock is write locked, see below
>
> __ip_vs_rs_lock:
> - spin_lock, use RCU
>
> Schedulers:
> - every .schedule method has its own locking, two examples:
> - write_lock: to protect the scheduler state (can be
> changed to spin_lock), see WRR. Difficult for RCU.
> - no lock: relies on IP_VS_WAIT_WHILE, no state
> is protected explicitly, fast like RCU, see WLC
>
> Scheduler state, eg. mark->cl:
> - careful RCU assignment, may be all .update_service methods
> should use copy-on-update (WRR). OTOH, ip_vs_wlc_schedule (WLC)
> has no locks at all, thanks to the IP_VS_WAIT_WHILE, so
> it is fast as RCU.
>
> Statistics:
> dest->stats.lock, svc->stats.lock, ip_vs_stats.lock:
> - called for every packet, BAD for SMP, see ip_vs_in_stats(),
> ip_vs_out_stats(), ip_vs_conn_stats()
>
> curr_sb_lock:
> - called for every packet depending on conn state
> - No benefits from RCU, should be spin_lock
>
> To summarize:
>
> - the main problem remains stats:
> dest->stats.lock, svc->stats.lock, ip_vs_stats.lock
>
> - RCU benefits when connection processes many packets per connection, eg.
> for TCP, SCTP, not much for UDP. No gains for the 1st
> packet in connection.
>
> - svc: no benefits from RCU, some schedulers protect state and
> need exclusive access, others have no state (and they do not use
> locks even now)
Thanks for the list. It looks like a good basis for some conversion work.
^ permalink raw reply
* Re: TAHI CN-6-4-1 failed on Linux 2.6.32 kernel
From: Steve Chen @ 2010-08-21 4:16 UTC (permalink / raw)
To: Jesse Gross; +Cc: David Miller, brian.haley, usagi-users-ctl, netdev
In-Reply-To: <AANLkTi=T_=Dy8EhOOyevLd4yj1cy2P8ytXUPS=Y-d6hK@mail.gmail.com>
On Fri, Aug 20, 2010 at 12:35 PM, Jesse Gross <jesse@nicira.com> wrote:
> On Fri, Aug 20, 2010 at 1:16 PM, Steve Chen <schen@mvista.com> wrote:
>>
>> On Thu, Aug 19, 2010 at 7:06 PM, David Miller <davem@davemloft.net> wrote:
>> > From: Steve Chen <schen@mvista.com>
>> > Date: Thu, 19 Aug 2010 13:35:14 -0500
>> >
>> >> I trace through the code. It appears that the network driver (e1000e
>> >> for my setup) always set ip_summed to CHECKSUM_UNNECESSARY. I have
>> >> been unsuccessful to get the driver to take the other branch where
>> >> ip_summed is set to CHECKSUM_COMPLETE. Even when I hard code
>> >> ip_summed to CHECKSUM_COMPLETE, __skb_checksum_complete_head set
>> >> ip_summed to CHECKSUM_UNNECESSARY after recomputing the checksum.
>> >>
>> >> So far the only way I'm able to get ICMP to recompute checksum is
>> >> through the attached hack. Even though I can get all the tests to
>> >> pass, but it just seem wrong.
>> >
>> > If turning off hardware RX checksumming with ethtool has no effect,
>> > and the problem is seen with multiple ethernet cards, the problem
>> > is elsewhere.
>> >
>> > First of all, if you turn RX checksumming off, the checksum field
>> > of the SKB should always be skb->ip_summed = 0. If this is not
>> > happening, find out why.
>>
>> Ahhh, thats my problem. I incorrectly thought the ip_summed should be
>> 2. The ip_summed is set to 1 in
>> __skb_checksum_complete_head. Looking at the code, shouldn't
>>
>> if (likely(!sum))
>>
>> be
>>
>> if (likely(sum))
>>
>> Since sum == 0 would indicate an error?
>
> sum == 0 indicates that the checksum is correct.
> If you compute the checksum of a packet containing the correct checksum the
> result is 0. It's like a slightly more complicated varient of a parity bit.
It appears the issue lies somewhere within CONNTRACK. A co-worker
tested with CONNTRACK disable, and the test passed. Looks like I have
a bit of homework to do. Thank you for all the helpful hints that got
me this far.
Steve
^ permalink raw reply
* [PATCH v8] PPTP: PPP over IPv4 (Point-to-Point Tunneling Protocol)
From: Dmitry Kozlov @ 2010-08-21 6:36 UTC (permalink / raw)
To: netdev
PPP: introduce "pptp" module which implements point-to-point tunneling protocol using pppox framework
NET: introduce the "gre" module for demultiplexing GRE packets on version criteria
(required to pptp and ip_gre may coexists)
NET: ip_gre: update to use the "gre" module
Signed-off-by: Dmitry Kozlov <xeb@mail.ru>
---
This patch introduces then pptp support to the linux kernel which dramatically speeds up pptp vpn connections and decreases cpu usage in comparison of existing user-space implementation (poptop/pptpclient). There is accel-pptp project (https://sourceforge.net/projects/accel-pptp/) to utilize this module, it contains plugin for pppd to use pptp in client-mode and modified pptpd (poptop) to build high-performance pptp NAS.
There was many changes from initial submitted patch, most important are:
1. using rcu instead of read-write locks
2. using static bitmap instead of dynamically allocated
3. using vmalloc for memory allocation instead of BITS_PER_LONG + __get_free_pages
4. fixed many coding style issues
Thanks to Eric Dumazet.
^ permalink raw reply
* [PATCH v8] PPTP: PPP over IPv4 (Point-to-Point Tunneling Protocol)
From: Dmitry Kozlov @ 2010-08-21 6:40 UTC (permalink / raw)
To: netdev
PPP: introduce "pptp" module which implements point-to-point tunneling protocol using pppox framework
NET: introduce the "gre" module for demultiplexing GRE packets on version criteria
(required to pptp and ip_gre may coexists)
NET: ip_gre: update to use the "gre" module
Signed-off-by: Dmitry Kozlov <xeb@mail.ru>
---
This patch introduces then pptp support to the linux kernel which dramatically speeds up pptp vpn connections and decreases cpu usage in comparison of existing user-space implementation (poptop/pptpclient). There is accel-pptp project (https://sourceforge.net/projects/accel-pptp/) to utilize this module, it contains plugin for pppd to use pptp in client-mode and modified pptpd (poptop) to build high-performance pptp NAS.
There was many changes from initial submitted patch, most important are:
1. using rcu instead of read-write locks
2. using static bitmap instead of dynamically allocated
3. using vmalloc for memory allocation instead of BITS_PER_LONG + __get_free_pages
4. fixed many coding style issues
Thanks to Eric Dumazet.
--
MAINTAINERS | 14 +
drivers/net/Kconfig | 11 +
drivers/net/Makefile | 1 +
drivers/net/pptp.c | 726 ++++++++++++++++++++++++++++++++++++++++++++++
include/linux/if_pppox.h | 59 +++--
include/net/gre.h | 18 ++
net/ipv4/Kconfig | 7 +
net/ipv4/Makefile | 1 +
net/ipv4/gre.c | 151 ++++++++++
net/ipv4/ip_gre.c | 14 +-
10 files changed, 975 insertions(+), 27 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 02f75fc..cdae013 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6450,6 +6450,20 @@ M: "Maciej W. Rozycki" <macro@linux-mips.org>
S: Maintained
F: drivers/serial/zs.*
+GRE DEMULTIPLEXER DRIVER
+M: Dmitry Kozlov <xeb@mail.ru>
+L: netdev@vger.kernel.org
+S: Maintained
+F: net/ipv4/gre.c
+F: include/net/gre.h
+
+PPTP DRIVER
+M: Dmitry Kozlov <xeb@mail.ru>
+L: netdev@vger.kernel.org
+S: Maintained
+F: drivers/net/pptp.c
+W: http://sourceforge.net/projects/accel-pptp
+
THE REST
M: Linus Torvalds <torvalds@linux-foundation.org>
L: linux-kernel@vger.kernel.org
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index ce2fcdd..2fa0516 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -3167,6 +3167,17 @@ config PPPOE
which contains instruction on how to use this driver (under
the heading "Kernel mode PPPoE").
+config PPTP
+ tristate "PPP over IPv4 (PPTP) (EXPERIMENTAL)"
+ depends on EXPERIMENTAL && PPP && NET_IPGRE_DEMUX
+ help
+ Support for PPP over IPv4.(Point-to-Point Tunneling Protocol)
+
+ This driver requires pppd plugin to work in client mode or
+ modified pptpd (poptop) to work in server mode.
+ See http://accel-pptp.sourceforge.net/ for information how to
+ utilize this module.
+
config PPPOATM
tristate "PPP over ATM"
depends on ATM && PPP
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index 0a0512a..b33fef1 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -162,6 +162,7 @@ obj-$(CONFIG_PPP_BSDCOMP) += bsd_comp.o
obj-$(CONFIG_PPP_MPPE) += ppp_mppe.o
obj-$(CONFIG_PPPOE) += pppox.o pppoe.o
obj-$(CONFIG_PPPOL2TP) += pppox.o
+obj-$(CONFIG_PPTP) += pppox.o pptp.o
obj-$(CONFIG_SLIP) += slip.o
obj-$(CONFIG_SLHC) += slhc.o
diff --git a/drivers/net/pptp.c b/drivers/net/pptp.c
new file mode 100644
index 0000000..39b1800
--- /dev/null
+++ b/drivers/net/pptp.c
@@ -0,0 +1,726 @@
+/*
+ * Point-to-Point Tunneling Protocol for Linux
+ *
+ * Authors: Dmitry Kozlov <xeb@mail.ru>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ */
+
+#include <linux/string.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/errno.h>
+#include <linux/netdevice.h>
+#include <linux/net.h>
+#include <linux/skbuff.h>
+#include <linux/init.h>
+#include <linux/ppp_channel.h>
+#include <linux/ppp_defs.h>
+#include <linux/if_pppox.h>
+#include <linux/if_ppp.h>
+#include <linux/notifier.h>
+#include <linux/file.h>
+#include <linux/in.h>
+#include <linux/ip.h>
+#include <linux/netfilter.h>
+#include <linux/netfilter_ipv4.h>
+#include <linux/version.h>
+#include <linux/rcupdate.h>
+#include <linux/spinlock.h>
+
+#include <net/sock.h>
+#include <net/protocol.h>
+#include <net/ip.h>
+#include <net/icmp.h>
+#include <net/route.h>
+#include <net/gre.h>
+
+#include <linux/uaccess.h>
+
+#define PPTP_DRIVER_VERSION "0.8.5"
+
+#define MAX_CALLID 65535
+
+static DECLARE_BITMAP(callid_bitmap, MAX_CALLID + 1);
+static struct pppox_sock **callid_sock;
+
+static DEFINE_SPINLOCK(chan_lock);
+
+static struct proto pptp_sk_proto __read_mostly;
+static struct ppp_channel_ops pptp_chan_ops;
+static const struct proto_ops pptp_ops;
+
+#define PPP_LCP_ECHOREQ 0x09
+#define PPP_LCP_ECHOREP 0x0A
+#define SC_RCV_BITS (SC_RCV_B7_1|SC_RCV_B7_0|SC_RCV_ODDP|SC_RCV_EVNP)
+
+#define MISSING_WINDOW 20
+#define WRAPPED(curseq, lastseq)\
+ ((((curseq) & 0xffffff00) == 0) &&\
+ (((lastseq) & 0xffffff00) == 0xffffff00))
+
+#define PPTP_GRE_PROTO 0x880B
+#define PPTP_GRE_VER 0x1
+
+#define PPTP_GRE_FLAG_C 0x80
+#define PPTP_GRE_FLAG_R 0x40
+#define PPTP_GRE_FLAG_K 0x20
+#define PPTP_GRE_FLAG_S 0x10
+#define PPTP_GRE_FLAG_A 0x80
+
+#define PPTP_GRE_IS_C(f) ((f)&PPTP_GRE_FLAG_C)
+#define PPTP_GRE_IS_R(f) ((f)&PPTP_GRE_FLAG_R)
+#define PPTP_GRE_IS_K(f) ((f)&PPTP_GRE_FLAG_K)
+#define PPTP_GRE_IS_S(f) ((f)&PPTP_GRE_FLAG_S)
+#define PPTP_GRE_IS_A(f) ((f)&PPTP_GRE_FLAG_A)
+
+#define PPTP_HEADER_OVERHEAD (2+sizeof(struct pptp_gre_header))
+struct pptp_gre_header {
+ u8 flags;
+ u8 ver;
+ u16 protocol;
+ u16 payload_len;
+ u16 call_id;
+ u32 seq;
+ u32 ack;
+} __packed;
+
+static struct pppox_sock *lookup_chan(u16 call_id, __be32 s_addr)
+{
+ struct pppox_sock *sock;
+ struct pptp_opt *opt;
+
+ rcu_read_lock();
+ sock = rcu_dereference(callid_sock[call_id]);
+ if (sock) {
+ opt = &sock->proto.pptp;
+ if (opt->dst_addr.sin_addr.s_addr != s_addr)
+ sock = NULL;
+ else
+ sock_hold(sk_pppox(sock));
+ }
+ rcu_read_unlock();
+
+ return sock;
+}
+
+static int lookup_chan_dst(u16 call_id, __be32 d_addr)
+{
+ struct pppox_sock *sock;
+ struct pptp_opt *opt;
+ int i;
+
+ rcu_read_lock();
+ for (i = find_next_bit(callid_bitmap, MAX_CALLID, 1); i < MAX_CALLID;
+ i = find_next_bit(callid_bitmap, MAX_CALLID, i + 1)) {
+ sock = rcu_dereference(callid_sock[i]);
+ if (!sock)
+ continue;
+ opt = &sock->proto.pptp;
+ if (opt->dst_addr.call_id == call_id &&
+ opt->dst_addr.sin_addr.s_addr == d_addr)
+ break;
+ }
+ rcu_read_unlock();
+
+ return i < MAX_CALLID;
+}
+
+static int add_chan(struct pppox_sock *sock)
+{
+ static int call_id;
+
+ spin_lock(&chan_lock);
+ if (!sock->proto.pptp.src_addr.call_id) {
+ call_id = find_next_zero_bit(callid_bitmap, MAX_CALLID, call_id + 1);
+ if (call_id == MAX_CALLID) {
+ call_id = find_next_zero_bit(callid_bitmap, MAX_CALLID, 1);
+ if (call_id == MAX_CALLID)
+ goto out_err;
+ }
+ sock->proto.pptp.src_addr.call_id = call_id;
+ } else if (test_bit(sock->proto.pptp.src_addr.call_id, callid_bitmap))
+ goto out_err;
+
+ set_bit(sock->proto.pptp.src_addr.call_id, callid_bitmap);
+ rcu_assign_pointer(callid_sock[sock->proto.pptp.src_addr.call_id], sock);
+ spin_unlock(&chan_lock);
+
+ return 0;
+
+out_err:
+ spin_unlock(&chan_lock);
+ return -1;
+}
+
+static void del_chan(struct pppox_sock *sock)
+{
+ spin_lock(&chan_lock);
+ clear_bit(sock->proto.pptp.src_addr.call_id, callid_bitmap);
+ rcu_assign_pointer(callid_sock[sock->proto.pptp.src_addr.call_id], NULL);
+ spin_unlock(&chan_lock);
+ synchronize_rcu();
+}
+
+static int pptp_xmit(struct ppp_channel *chan, struct sk_buff *skb)
+{
+ struct sock *sk = (struct sock *) chan->private;
+ struct pppox_sock *po = pppox_sk(sk);
+ struct pptp_opt *opt = &po->proto.pptp;
+ struct pptp_gre_header *hdr;
+ unsigned int header_len = sizeof(*hdr);
+ int err = 0;
+ int islcp;
+ int len;
+ unsigned char *data;
+ __u32 seq_recv;
+
+
+ struct rtable *rt;
+ struct net_device *tdev;
+ struct iphdr *iph;
+ int max_headroom;
+
+ if (sk_pppox(po)->sk_state & PPPOX_DEAD)
+ goto tx_error;
+
+ {
+ struct flowi fl = { .oif = 0,
+ .nl_u = {
+ .ip4_u = {
+ .daddr = opt->dst_addr.sin_addr.s_addr,
+ .saddr = opt->src_addr.sin_addr.s_addr,
+ .tos = RT_TOS(0) } },
+ .proto = IPPROTO_GRE };
+ err = ip_route_output_key(&init_net, &rt, &fl);
+ if (err)
+ goto tx_error;
+ }
+ tdev = rt->u.dst.dev;
+
+ max_headroom = LL_RESERVED_SPACE(tdev) + sizeof(*iph) + sizeof(*hdr) + 2;
+
+ if (skb_headroom(skb) < max_headroom || skb_cloned(skb) || skb_shared(skb)) {
+ struct sk_buff *new_skb = skb_realloc_headroom(skb, max_headroom);
+ if (!new_skb) {
+ ip_rt_put(rt);
+ goto tx_error;
+ }
+ if (skb->sk)
+ skb_set_owner_w(new_skb, skb->sk);
+ kfree_skb(skb);
+ skb = new_skb;
+ }
+
+ data = skb->data;
+ islcp = ((data[0] << 8) + data[1]) == PPP_LCP && 1 <= data[2] && data[2] <= 7;
+
+ /* compress protocol field */
+ if ((opt->ppp_flags & SC_COMP_PROT) && data[0] == 0 && !islcp)
+ skb_pull(skb, 1);
+
+ /* Put in the address/control bytes if necessary */
+ if ((opt->ppp_flags & SC_COMP_AC) == 0 || islcp) {
+ data = skb_push(skb, 2);
+ data[0] = PPP_ALLSTATIONS;
+ data[1] = PPP_UI;
+ }
+
+ len = skb->len;
+
+ seq_recv = opt->seq_recv;
+
+ if (opt->ack_sent == seq_recv)
+ header_len -= sizeof(hdr->ack);
+
+ /* Push down and install GRE header */
+ skb_push(skb, header_len);
+ hdr = (struct pptp_gre_header *)(skb->data);
+
+ hdr->flags = PPTP_GRE_FLAG_K;
+ hdr->ver = PPTP_GRE_VER;
+ hdr->protocol = htons(PPTP_GRE_PROTO);
+ hdr->call_id = htons(opt->dst_addr.call_id);
+
+ hdr->flags |= PPTP_GRE_FLAG_S;
+ hdr->seq = htonl(++opt->seq_sent);
+ if (opt->ack_sent != seq_recv) {
+ /* send ack with this message */
+ hdr->ver |= PPTP_GRE_FLAG_A;
+ hdr->ack = htonl(seq_recv);
+ opt->ack_sent = seq_recv;
+ }
+ hdr->payload_len = htons(len);
+
+ /* Push down and install the IP header. */
+
+ skb_reset_transport_header(skb);
+ skb_push(skb, sizeof(*iph));
+ skb_reset_network_header(skb);
+ memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
+ IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED | IPSKB_REROUTED);
+
+ iph = ip_hdr(skb);
+ iph->version = 4;
+ iph->ihl = sizeof(struct iphdr) >> 2;
+ if (ip_dont_fragment(sk, &rt->u.dst))
+ iph->frag_off = htons(IP_DF);
+ else
+ iph->frag_off = 0;
+ iph->protocol = IPPROTO_GRE;
+ iph->tos = 0;
+ iph->daddr = rt->rt_dst;
+ iph->saddr = rt->rt_src;
+ iph->ttl = dst_metric(&rt->u.dst, RTAX_HOPLIMIT);
+ iph->tot_len = htons(skb->len);
+
+ skb_dst_drop(skb);
+ skb_dst_set(skb, &rt->u.dst);
+
+ nf_reset(skb);
+
+ skb->ip_summed = CHECKSUM_NONE;
+ ip_select_ident(iph, &rt->u.dst, NULL);
+ ip_send_check(iph);
+
+ ip_local_out(skb);
+
+tx_error:
+ return 1;
+}
+
+static int pptp_rcv_core(struct sock *sk, struct sk_buff *skb)
+{
+ struct pppox_sock *po = pppox_sk(sk);
+ struct pptp_opt *opt = &po->proto.pptp;
+ int headersize, payload_len, seq;
+ __u8 *payload;
+ struct pptp_gre_header *header;
+
+ if (!(sk->sk_state & PPPOX_CONNECTED)) {
+ if (sock_queue_rcv_skb(sk, skb))
+ goto drop;
+ return NET_RX_SUCCESS;
+ }
+
+ header = (struct pptp_gre_header *)(skb->data);
+
+ /* test if acknowledgement present */
+ if (PPTP_GRE_IS_A(header->ver)) {
+ __u32 ack = (PPTP_GRE_IS_S(header->flags)) ?
+ header->ack : header->seq; /* ack in different place if S = 0 */
+
+ ack = ntohl(ack);
+
+ if (ack > opt->ack_recv)
+ opt->ack_recv = ack;
+ /* also handle sequence number wrap-around */
+ if (WRAPPED(ack, opt->ack_recv))
+ opt->ack_recv = ack;
+ }
+
+ /* test if payload present */
+ if (!PPTP_GRE_IS_S(header->flags))
+ goto drop;
+
+ headersize = sizeof(*header);
+ payload_len = ntohs(header->payload_len);
+ seq = ntohl(header->seq);
+
+ /* no ack present? */
+ if (!PPTP_GRE_IS_A(header->ver))
+ headersize -= sizeof(header->ack);
+ /* check for incomplete packet (length smaller than expected) */
+ if (skb->len - headersize < payload_len)
+ goto drop;
+
+ payload = skb->data + headersize;
+ /* check for expected sequence number */
+ if (seq < opt->seq_recv + 1 || WRAPPED(opt->seq_recv, seq)) {
+ if ((payload[0] == PPP_ALLSTATIONS) && (payload[1] == PPP_UI) &&
+ (PPP_PROTOCOL(payload) == PPP_LCP) &&
+ ((payload[4] == PPP_LCP_ECHOREQ) || (payload[4] == PPP_LCP_ECHOREP)))
+ goto allow_packet;
+ } else {
+ opt->seq_recv = seq;
+allow_packet:
+ skb_pull(skb, headersize);
+
+ if (payload[0] == PPP_ALLSTATIONS && payload[1] == PPP_UI) {
+ /* chop off address/control */
+ if (skb->len < 3)
+ goto drop;
+ skb_pull(skb, 2);
+ }
+
+ if ((*skb->data) & 1) {
+ /* protocol is compressed */
+ skb_push(skb, 1)[0] = 0;
+ }
+
+ skb->ip_summed = CHECKSUM_NONE;
+ skb_set_network_header(skb, skb->head-skb->data);
+ ppp_input(&po->chan, skb);
+
+ return NET_RX_SUCCESS;
+ }
+drop:
+ kfree_skb(skb);
+ return NET_RX_DROP;
+}
+
+static int pptp_rcv(struct sk_buff *skb)
+{
+ struct pppox_sock *po;
+ struct pptp_gre_header *header;
+ struct iphdr *iph;
+
+ if (skb->pkt_type != PACKET_HOST)
+ goto drop;
+
+ if (!pskb_may_pull(skb, 12))
+ goto drop;
+
+ iph = ip_hdr(skb);
+
+ header = (struct pptp_gre_header *)skb->data;
+
+ if (ntohs(header->protocol) != PPTP_GRE_PROTO || /* PPTP-GRE protocol for PPTP */
+ PPTP_GRE_IS_C(header->flags) || /* flag C should be clear */
+ PPTP_GRE_IS_R(header->flags) || /* flag R should be clear */
+ !PPTP_GRE_IS_K(header->flags) || /* flag K should be set */
+ (header->flags&0xF) != 0) /* routing and recursion ctrl = 0 */
+ /* if invalid, discard this packet */
+ goto drop;
+
+ po = lookup_chan(htons(header->call_id), iph->saddr);
+ if (po) {
+ skb_dst_drop(skb);
+ nf_reset(skb);
+ return sk_receive_skb(sk_pppox(po), skb, 0);
+ }
+drop:
+ kfree_skb(skb);
+ return NET_RX_DROP;
+}
+
+static int pptp_bind(struct socket *sock, struct sockaddr *uservaddr,
+ int sockaddr_len)
+{
+ struct sock *sk = sock->sk;
+ struct sockaddr_pppox *sp = (struct sockaddr_pppox *) uservaddr;
+ struct pppox_sock *po = pppox_sk(sk);
+ struct pptp_opt *opt = &po->proto.pptp;
+ int error = 0;
+
+ lock_sock(sk);
+
+ opt->src_addr = sp->sa_addr.pptp;
+ if (add_chan(po)) {
+ release_sock(sk);
+ error = -EBUSY;
+ }
+
+ release_sock(sk);
+ return error;
+}
+
+static int pptp_connect(struct socket *sock, struct sockaddr *uservaddr,
+ int sockaddr_len, int flags)
+{
+ struct sock *sk = sock->sk;
+ struct sockaddr_pppox *sp = (struct sockaddr_pppox *) uservaddr;
+ struct pppox_sock *po = pppox_sk(sk);
+ struct pptp_opt *opt = &po->proto.pptp;
+ struct rtable *rt;
+ int error = 0;
+
+ if (sp->sa_protocol != PX_PROTO_PPTP)
+ return -EINVAL;
+
+ if (lookup_chan_dst(sp->sa_addr.pptp.call_id, sp->sa_addr.pptp.sin_addr.s_addr))
+ return -EALREADY;
+
+ lock_sock(sk);
+ /* Check for already bound sockets */
+ if (sk->sk_state & PPPOX_CONNECTED) {
+ error = -EBUSY;
+ goto end;
+ }
+
+ /* Check for already disconnected sockets, on attempts to disconnect */
+ if (sk->sk_state & PPPOX_DEAD) {
+ error = -EALREADY;
+ goto end;
+ }
+
+ if (!opt->src_addr.sin_addr.s_addr || !sp->sa_addr.pptp.sin_addr.s_addr) {
+ error = -EINVAL;
+ goto end;
+ }
+
+ po->chan.private = sk;
+ po->chan.ops = &pptp_chan_ops;
+
+ {
+ struct flowi fl = {
+ .nl_u = {
+ .ip4_u = {
+ .daddr = opt->dst_addr.sin_addr.s_addr,
+ .saddr = opt->src_addr.sin_addr.s_addr,
+ .tos = RT_CONN_FLAGS(sk) } },
+ .proto = IPPROTO_GRE };
+ security_sk_classify_flow(sk, &fl);
+ if (ip_route_output_key(&init_net, &rt, &fl)) {
+ error = -EHOSTUNREACH;
+ goto end;
+ }
+ sk_setup_caps(sk, &rt->u.dst);
+ }
+ po->chan.mtu = dst_mtu(&rt->u.dst);
+ if (!po->chan.mtu)
+ po->chan.mtu = PPP_MTU;
+ ip_rt_put(rt);
+ po->chan.mtu -= PPTP_HEADER_OVERHEAD;
+
+ po->chan.hdrlen = 2 + sizeof(struct pptp_gre_header);
+ error = ppp_register_channel(&po->chan);
+ if (error) {
+ pr_err("PPTP: failed to register PPP channel (%d)\n", error);
+ goto end;
+ }
+
+ opt->dst_addr = sp->sa_addr.pptp;
+ sk->sk_state = PPPOX_CONNECTED;
+
+ end:
+ release_sock(sk);
+ return error;
+}
+
+static int pptp_getname(struct socket *sock, struct sockaddr *uaddr,
+ int *usockaddr_len, int peer)
+{
+ int len = sizeof(struct sockaddr_pppox);
+ struct sockaddr_pppox sp;
+
+ sp.sa_family = AF_PPPOX;
+ sp.sa_protocol = PX_PROTO_PPTP;
+ sp.sa_addr.pptp = pppox_sk(sock->sk)->proto.pptp.src_addr;
+
+ memcpy(uaddr, &sp, len);
+
+ *usockaddr_len = len;
+
+ return 0;
+}
+
+static int pptp_release(struct socket *sock)
+{
+ struct sock *sk = sock->sk;
+ struct pppox_sock *po;
+ struct pptp_opt *opt;
+ int error = 0;
+
+ if (!sk)
+ return 0;
+
+ lock_sock(sk);
+
+ if (sock_flag(sk, SOCK_DEAD)) {
+ release_sock(sk);
+ return -EBADF;
+ }
+
+ po = pppox_sk(sk);
+ opt = &po->proto.pptp;
+ del_chan(po);
+
+ pppox_unbind_sock(sk);
+ sk->sk_state = PPPOX_DEAD;
+
+ sock_orphan(sk);
+ sock->sk = NULL;
+
+ release_sock(sk);
+ sock_put(sk);
+
+ return error;
+}
+
+static void pptp_sock_destruct(struct sock *sk)
+{
+ if (!(sk->sk_state & PPPOX_DEAD)) {
+ del_chan(pppox_sk(sk));
+ pppox_unbind_sock(sk);
+ }
+ skb_queue_purge(&sk->sk_receive_queue);
+}
+
+static int pptp_create(struct net *net, struct socket *sock)
+{
+ int error = -ENOMEM;
+ struct sock *sk;
+ struct pppox_sock *po;
+ struct pptp_opt *opt;
+
+ sk = sk_alloc(net, PF_PPPOX, GFP_KERNEL, &pptp_sk_proto);
+ if (!sk)
+ goto out;
+
+ sock_init_data(sock, sk);
+
+ sock->state = SS_UNCONNECTED;
+ sock->ops = &pptp_ops;
+
+ sk->sk_backlog_rcv = pptp_rcv_core;
+ sk->sk_state = PPPOX_NONE;
+ sk->sk_type = SOCK_STREAM;
+ sk->sk_family = PF_PPPOX;
+ sk->sk_protocol = PX_PROTO_PPTP;
+ sk->sk_destruct = pptp_sock_destruct;
+
+ po = pppox_sk(sk);
+ opt = &po->proto.pptp;
+
+ opt->seq_sent = 0; opt->seq_recv = 0;
+ opt->ack_recv = 0; opt->ack_sent = 0;
+
+ error = 0;
+out:
+ return error;
+}
+
+static int pptp_ppp_ioctl(struct ppp_channel *chan, unsigned int cmd,
+ unsigned long arg)
+{
+ struct sock *sk = (struct sock *) chan->private;
+ struct pppox_sock *po = pppox_sk(sk);
+ struct pptp_opt *opt = &po->proto.pptp;
+ void __user *argp = (void __user *)arg;
+ int __user *p = argp;
+ int err, val;
+
+ err = -EFAULT;
+ switch (cmd) {
+ case PPPIOCGFLAGS:
+ val = opt->ppp_flags;
+ if (put_user(val, p))
+ break;
+ err = 0;
+ break;
+ case PPPIOCSFLAGS:
+ if (get_user(val, p))
+ break;
+ opt->ppp_flags = val & ~SC_RCV_BITS;
+ err = 0;
+ break;
+ default:
+ err = -ENOTTY;
+ }
+
+ return err;
+}
+
+static struct ppp_channel_ops pptp_chan_ops = {
+ .start_xmit = pptp_xmit,
+ .ioctl = pptp_ppp_ioctl,
+};
+
+static struct proto pptp_sk_proto __read_mostly = {
+ .name = "PPTP",
+ .owner = THIS_MODULE,
+ .obj_size = sizeof(struct pppox_sock),
+};
+
+static const struct proto_ops pptp_ops = {
+ .family = AF_PPPOX,
+ .owner = THIS_MODULE,
+ .release = pptp_release,
+ .bind = pptp_bind,
+ .connect = pptp_connect,
+ .socketpair = sock_no_socketpair,
+ .accept = sock_no_accept,
+ .getname = pptp_getname,
+ .poll = sock_no_poll,
+ .listen = sock_no_listen,
+ .shutdown = sock_no_shutdown,
+ .setsockopt = sock_no_setsockopt,
+ .getsockopt = sock_no_getsockopt,
+ .sendmsg = sock_no_sendmsg,
+ .recvmsg = sock_no_recvmsg,
+ .mmap = sock_no_mmap,
+ .ioctl = pppox_ioctl,
+};
+
+static struct pppox_proto pppox_pptp_proto = {
+ .create = pptp_create,
+ .owner = THIS_MODULE,
+};
+
+static struct gre_protocol gre_pptp_protocol = {
+ .handler = pptp_rcv,
+};
+
+static int __init pptp_init_module(void)
+{
+ int err = 0;
+ pr_info("PPTP driver version " PPTP_DRIVER_VERSION "\n");
+
+ callid_sock = __vmalloc((MAX_CALLID + 1) * sizeof(void *),
+ GFP_KERNEL | __GFP_ZERO, PAGE_KERNEL);
+ if (!callid_sock) {
+ pr_err("PPTP: cann't allocate memory\n");
+ return -ENOMEM;
+ }
+
+ err = gre_add_protocol(&gre_pptp_protocol, GREPROTO_PPTP);
+ if (err) {
+ pr_err("PPTP: can't add gre protocol\n");
+ goto out_mem_free;
+ }
+
+ err = proto_register(&pptp_sk_proto, 0);
+ if (err) {
+ pr_err("PPTP: can't register sk_proto\n");
+ goto out_gre_del_protocol;
+ }
+
+ err = register_pppox_proto(PX_PROTO_PPTP, &pppox_pptp_proto);
+ if (err) {
+ pr_err("PPTP: can't register pppox_proto\n");
+ goto out_unregister_sk_proto;
+ }
+
+ return 0;
+
+out_unregister_sk_proto:
+ proto_unregister(&pptp_sk_proto);
+out_gre_del_protocol:
+ gre_del_protocol(&gre_pptp_protocol, GREPROTO_PPTP);
+out_mem_free:
+ vfree(callid_sock);
+
+ return err;
+}
+
+static void __exit pptp_exit_module(void)
+{
+ unregister_pppox_proto(PX_PROTO_PPTP);
+ proto_unregister(&pptp_sk_proto);
+ gre_del_protocol(&gre_pptp_protocol, GREPROTO_PPTP);
+ vfree(callid_sock);
+}
+
+module_init(pptp_init_module);
+module_exit(pptp_exit_module);
+
+MODULE_DESCRIPTION("Point-to-Point Tunneling Protocol");
+MODULE_AUTHOR("D. Kozlov (xeb@mail.ru)");
+MODULE_LICENSE("GPL");
+
diff --git a/include/linux/if_pppox.h b/include/linux/if_pppox.h
index a6577af..e5c820a 100644
--- a/include/linux/if_pppox.h
+++ b/include/linux/if_pppox.h
@@ -40,26 +40,37 @@
* PPPoE addressing definition
*/
typedef __be16 sid_t;
-struct pppoe_addr{
- sid_t sid; /* Session identifier */
- unsigned char remote[ETH_ALEN]; /* Remote address */
- char dev[IFNAMSIZ]; /* Local device to use */
+struct pppoe_addr {
+ sid_t sid; /* Session identifier */
+ unsigned char remote[ETH_ALEN]; /* Remote address */
+ char dev[IFNAMSIZ]; /* Local device to use */
};
/************************************************************************
- * Protocols supported by AF_PPPOX
- */
+ * PPTP addressing definition
+ */
+struct pptp_addr {
+ u16 call_id;
+ struct in_addr sin_addr;
+};
+
+/************************************************************************
+ * Protocols supported by AF_PPPOX
+ */
#define PX_PROTO_OE 0 /* Currently just PPPoE */
#define PX_PROTO_OL2TP 1 /* Now L2TP also */
-#define PX_MAX_PROTO 2
+#define PX_PROTO_PPTP 2
+#define PX_MAX_PROTO 3
-struct sockaddr_pppox {
- sa_family_t sa_family; /* address family, AF_PPPOX */
- unsigned int sa_protocol; /* protocol identifier */
- union{
- struct pppoe_addr pppoe;
- }sa_addr;
-}__attribute__ ((packed));
+struct sockaddr_pppox {
+ sa_family_t sa_family; /* address family, AF_PPPOX */
+ unsigned int sa_protocol; /* protocol identifier */
+ union {
+ struct pppoe_addr pppoe;
+ struct pptp_addr pptp;
+ } sa_addr;
+} __packed;
+/* hm ... why packed here ? */
/* The use of the above union isn't viable because the size of this
* struct must stay fixed over time -- applications use sizeof(struct
@@ -70,7 +81,7 @@ struct sockaddr_pppol2tp {
sa_family_t sa_family; /* address family, AF_PPPOX */
unsigned int sa_protocol; /* protocol identifier */
struct pppol2tp_addr pppol2tp;
-}__attribute__ ((packed));
+} __packed;
/* The L2TPv3 protocol changes tunnel and session ids from 16 to 32
* bits. So we need a different sockaddr structure.
@@ -79,7 +90,7 @@ struct sockaddr_pppol2tpv3 {
sa_family_t sa_family; /* address family, AF_PPPOX */
unsigned int sa_protocol; /* protocol identifier */
struct pppol2tpv3_addr pppol2tp;
-} __attribute__ ((packed));
+} __packed;
/*********************************************************************
*
@@ -101,7 +112,7 @@ struct pppoe_tag {
__be16 tag_type;
__be16 tag_len;
char tag_data[0];
-} __attribute ((packed));
+} __packed;
/* Tag identifiers */
#define PTT_EOL __cpu_to_be16(0x0000)
@@ -129,7 +140,7 @@ struct pppoe_hdr {
__be16 sid;
__be16 length;
struct pppoe_tag tag[0];
-} __attribute__ ((packed));
+} __packed;
/* Length of entire PPPoE + PPP header */
#define PPPOE_SES_HLEN 8
@@ -150,15 +161,23 @@ struct pppoe_opt {
relayed to (PPPoE relaying) */
};
+struct pptp_opt {
+ struct pptp_addr src_addr;
+ struct pptp_addr dst_addr;
+ u32 ack_sent, ack_recv;
+ u32 seq_sent, seq_recv;
+ int ppp_flags;
+};
#include <net/sock.h>
struct pppox_sock {
/* struct sock must be the first member of pppox_sock */
- struct sock sk;
- struct ppp_channel chan;
+ struct sock sk;
+ struct ppp_channel chan;
struct pppox_sock *next; /* for hash table */
union {
struct pppoe_opt pppoe;
+ struct pptp_opt pptp;
} proto;
__be16 num;
};
diff --git a/include/net/gre.h b/include/net/gre.h
new file mode 100644
index 0000000..8266547
--- /dev/null
+++ b/include/net/gre.h
@@ -0,0 +1,18 @@
+#ifndef __LINUX_GRE_H
+#define __LINUX_GRE_H
+
+#include <linux/skbuff.h>
+
+#define GREPROTO_CISCO 0
+#define GREPROTO_PPTP 1
+#define GREPROTO_MAX 2
+
+struct gre_protocol {
+ int (*handler)(struct sk_buff *skb);
+ void (*err_handler)(struct sk_buff *skb, u32 info);
+};
+
+int gre_add_protocol(const struct gre_protocol *proto, u8 version);
+int gre_del_protocol(const struct gre_protocol *proto, u8 version);
+
+#endif
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 7c3a7d1..7458bda 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -215,8 +215,15 @@ config NET_IPIP
be inserted in and removed from the running kernel whenever you
want). Most people won't need this and can say N.
+config NET_IPGRE_DEMUX
+ tristate "IP: GRE demultiplexer"
+ help
+ This is helper module to demultiplex GRE packets on GRE version field criteria.
+ Required by ip_gre and pptp modules.
+
config NET_IPGRE
tristate "IP: GRE tunnels over IP"
+ depends on NET_IPGRE_DEMUX
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/Makefile b/net/ipv4/Makefile
index 80ff87c..4978d22 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -20,6 +20,7 @@ obj-$(CONFIG_PROC_FS) += proc.o
obj-$(CONFIG_IP_MULTIPLE_TABLES) += fib_rules.o
obj-$(CONFIG_IP_MROUTE) += ipmr.o
obj-$(CONFIG_NET_IPIP) += ipip.o
+obj-$(CONFIG_NET_IPGRE_DEMUX) += gre.o
obj-$(CONFIG_NET_IPGRE) += ip_gre.o
obj-$(CONFIG_SYN_COOKIES) += syncookies.o
obj-$(CONFIG_INET_AH) += ah4.o
diff --git a/net/ipv4/gre.c b/net/ipv4/gre.c
new file mode 100644
index 0000000..b546736
--- /dev/null
+++ b/net/ipv4/gre.c
@@ -0,0 +1,151 @@
+/*
+ * GRE over IPv4 demultiplexer driver
+ *
+ * Authors: Dmitry Kozlov (xeb@mail.ru)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/kmod.h>
+#include <linux/skbuff.h>
+#include <linux/in.h>
+#include <linux/netdevice.h>
+#include <linux/version.h>
+#include <linux/spinlock.h>
+#include <net/protocol.h>
+#include <net/gre.h>
+
+
+const struct gre_protocol *gre_proto[GREPROTO_MAX] __read_mostly;
+static DEFINE_SPINLOCK(gre_proto_lock);
+
+int gre_add_protocol(const struct gre_protocol *proto, u8 version)
+{
+ if (version >= GREPROTO_MAX)
+ goto err_out;
+
+ spin_lock(&gre_proto_lock);
+ if (gre_proto[version])
+ goto err_out_unlock;
+
+ rcu_assign_pointer(gre_proto[version], proto);
+ spin_unlock(&gre_proto_lock);
+ return 0;
+
+err_out_unlock:
+ spin_unlock(&gre_proto_lock);
+err_out:
+ return -1;
+}
+EXPORT_SYMBOL_GPL(gre_add_protocol);
+
+int gre_del_protocol(const struct gre_protocol *proto, u8 version)
+{
+ if (version >= GREPROTO_MAX)
+ goto err_out;
+
+ spin_lock(&gre_proto_lock);
+ if (gre_proto[version] != proto)
+ goto err_out_unlock;
+ rcu_assign_pointer(gre_proto[version], NULL);
+ spin_unlock(&gre_proto_lock);
+ synchronize_rcu();
+ return 0;
+
+err_out_unlock:
+ spin_unlock(&gre_proto_lock);
+err_out:
+ return -1;
+}
+EXPORT_SYMBOL_GPL(gre_del_protocol);
+
+static int gre_rcv(struct sk_buff *skb)
+{
+ const struct gre_protocol *proto;
+ u8 ver;
+ int ret;
+
+ if (!pskb_may_pull(skb, 12))
+ goto drop;
+
+ ver = skb->data[1]&0x7f;
+ if (ver >= GREPROTO_MAX)
+ goto drop;
+
+ rcu_read_lock();
+ proto = rcu_dereference(gre_proto[ver]);
+ if (!proto || !proto->handler)
+ goto drop_unlock;
+ ret = proto->handler(skb);
+ rcu_read_unlock();
+ return ret;
+
+drop_unlock:
+ rcu_read_unlock();
+drop:
+ kfree_skb(skb);
+ return NET_RX_DROP;
+}
+
+static void gre_err(struct sk_buff *skb, u32 info)
+{
+ const struct gre_protocol *proto;
+ u8 ver;
+
+ if (!pskb_may_pull(skb, 12))
+ goto drop;
+
+ ver = skb->data[1]&0x7f;
+ if (ver >= GREPROTO_MAX)
+ goto drop;
+
+ rcu_read_lock();
+ proto = rcu_dereference(gre_proto[ver]);
+ if (!proto || !proto->err_handler)
+ goto drop_unlock;
+ proto->err_handler(skb, info);
+ rcu_read_unlock();
+ return;
+
+drop_unlock:
+ rcu_read_unlock();
+drop:
+ kfree_skb(skb);
+}
+
+static const struct net_protocol net_gre_protocol = {
+ .handler = gre_rcv,
+ .err_handler = gre_err,
+ .netns_ok = 1,
+};
+
+static int __init gre_init(void)
+{
+ pr_info("GRE over IPv4 demultiplexor driver");
+
+ if (inet_add_protocol(&net_gre_protocol, IPPROTO_GRE) < 0) {
+ pr_err("gre: can't add protocol\n");
+ return -EAGAIN;
+ }
+
+ return 0;
+}
+
+static void __exit gre_exit(void)
+{
+ inet_del_protocol(&net_gre_protocol, IPPROTO_GRE);
+}
+
+module_init(gre_init);
+module_exit(gre_exit);
+
+MODULE_DESCRIPTION("GRE over IPv4 demultiplexer driver");
+MODULE_AUTHOR("D. Kozlov (xeb@mail.ru)");
+MODULE_LICENSE("GPL");
+
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 32618e1..5a01d86 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -44,6 +44,7 @@
#include <net/net_namespace.h>
#include <net/netns/generic.h>
#include <net/rtnetlink.h>
+#include <net/gre.h>
#ifdef CONFIG_IPV6
#include <net/ipv6.h>
@@ -1276,10 +1277,9 @@ static void ipgre_fb_tunnel_init(struct net_device *dev)
}
-static const struct net_protocol ipgre_protocol = {
- .handler = ipgre_rcv,
- .err_handler = ipgre_err,
- .netns_ok = 1,
+static const struct gre_protocol ipgre_protocol = {
+ .handler = ipgre_rcv,
+ .err_handler = ipgre_err,
};
static void ipgre_destroy_tunnels(struct ipgre_net *ign, struct list_head *head)
@@ -1661,7 +1661,7 @@ static int __init ipgre_init(void)
if (err < 0)
return err;
- err = inet_add_protocol(&ipgre_protocol, IPPROTO_GRE);
+ err = gre_add_protocol(&ipgre_protocol, GREPROTO_CISCO);
if (err < 0) {
printk(KERN_INFO "ipgre init: can't add protocol\n");
goto add_proto_failed;
@@ -1681,7 +1681,7 @@ out:
tap_ops_failed:
rtnl_link_unregister(&ipgre_link_ops);
rtnl_link_failed:
- inet_del_protocol(&ipgre_protocol, IPPROTO_GRE);
+ gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO);
add_proto_failed:
unregister_pernet_device(&ipgre_net_ops);
goto out;
@@ -1691,7 +1691,7 @@ static void __exit ipgre_fini(void)
{
rtnl_link_unregister(&ipgre_tap_ops);
rtnl_link_unregister(&ipgre_link_ops);
- if (inet_del_protocol(&ipgre_protocol, IPPROTO_GRE) < 0)
+ if (gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO) < 0)
printk(KERN_INFO "ipgre close: can't remove protocol\n");
unregister_pernet_device(&ipgre_net_ops);
}
^ permalink raw reply related
* Re: [PATCH 2/2] platform: Facilitate the creation of pseudo-platform buses
From: Grant Likely @ 2010-08-21 7:10 UTC (permalink / raw)
To: Kevin Hilman
Cc: Moffett, Kyle D, Patrick Pannuto, linux-kernel@vger.kernel.org,
linux-arm-msm@vger.kernel.org, magnus.damm@gmail.com,
gregkh@suse.de, Paul Mundt, Magnus Damm, Rafael J. Wysocki,
Eric Miao, Dmitry Torokhov, netdev@vger.kernel.org,
Kyle D Moffett
In-Reply-To: <87y6c03jxj.fsf@deeprootsystems.com>
On Fri, Aug 20, 2010 at 6:10 PM, Kevin Hilman
<khilman@deeprootsystems.com> wrote:
> Grant Likely <grant.likely@secretlab.ca> writes:
>
> [...]
>
>>>>
>>>> My fears on this point may very well be unfounded. This isn't the
>>>> hill I'm going to die on either. Show me an implementation of driver
>>>> sharing that is clean and prove me wrong! :-)
>>>
>>> IMHO, simply overriding the few dev_pm_ops methods was the cleanest and
>>> simplest.
>>>
>>> Since we seem to be in agreement now that the a new bus may not the
>>> right abstraction for this (since we want it to be completely
>>> transparent to the drivers), I'll go back to the original design. No new
>>> bus types, keep the platform_bus as is, but simply override the few
>>> dev_pm_ops methods I care about. This is what is done on SH,
>>> SH-Mobile[1] and my original version for OMAP that started this
>>> conversation.
>>>
>>> Yes, the weak-symbol method of overriding is not scalable, but that's a
>>> separate issue from whether or not to create a new bus. I have a
>>> proposed fix for the weak which I'll post shortly.
>>
>> Okay.
>>
>> One constraint remains though: If you can override the dev_pm_ops on
>> a per-device or per-device-parent basis, then you've got my support.
>
> hmm, a new requirement?, and one that would require some significant
> changes to the driver model.
Nope! Not a new requirement; this is the issue that this entire
thread has been about. Hijacking the entire platform_bus_type
behaviour for a particular bus is the wrong thing to do because there
are other users in the same system.
> Currently, dev_pm_ops for a bus applies globally to *all* devices on
> that bus (or class or type) and changing that would require changing the
> platform_bus code to start having per-device (or per-parent-device)
> checks.
I'm not opposed to modifying the platform_bus_type to support the use-case.
>> If the override (even when fixed to work at runtime) applies to every
>> device on the platform_bus_type, then I'll nack it.
>
> /me can't help but wonder why the OMAP implementation is getting all the
> negative attention while the other identical implementations are already
> upstream.
OMAP is not getting singled out. This applies to all the implementations.
>> My concern here is that the SoC or platform support code doesn't get
>> to "own" the platform_bus_type.
>
> Well, I'm not proposing that here. This "feature" already exists in
> mainline (albeit using the less-than-optimal weak symbol approach.) All
> I'm proposing is fixing it to make it multi-arch friendly.
Oops, I didn't phrase that very well. What I meant was that SoC or
platform support code must not be allowed to "own" or hijack the
platform_bus_type for it's own purposes. What I wrote sounds like I
was suggesting that as a good thing.
> If you think this behavior should be changed to non global, that should
> be done a separate series since it is not directly related to runtime PM
> support for a given platform, IMO.
>
>> Other drivers/code can register their own set of
>> platform_devices, which may also need to perform their own dev_pm_ops
>> override.
>
> IMHO, we should deal with such hypothetical, future problems *if* they
> arise down the road.
>
>> If the override is global to the platform_bus_type, then the model
>> will not scale.
>
> It will not scale any more (or less) than the current functionality of
> the driver model which handles this globally. Again, if scalabilty
> becomes a problem down the road, lets fix it then.
Alright, I can agree to that. I do agree that the runtime override is
far and away better than the weak symbol approach. However, there
needs to be some very clear rules in place for users of the override,
namely that users must understand that they are stewards of the
platform_bus_type, and must take care to preserve the default
behaviour for "uninteresting" devices.
Also, the expectation should be that it is a temporary measure until a
better abstraction is implemented. It might be that a separate
bus_type is the way to go (if the multiple driver registration problem
can be solved) or it might be a way to differentiate pm_ops either
per-device or per-parent. I'm not sure, but I'm starting on an OMAP
project soon, so I may very well end up working on this.
Cheers,
g.
^ permalink raw reply
* Re: [Bugme-new] [Bug 16626] New: Machine hangs with EIP at skb_copy_and_csum_dev
From: Jarek Poplawski @ 2010-08-21 7:47 UTC (permalink / raw)
To: Plamen Petrov
Cc: Eric Dumazet, Andrew Morton, netdev, bugzilla-daemon,
bugme-daemon
In-Reply-To: <20100820193835.GA6025@del.dom.local>
On Fri, Aug 20, 2010 at 09:38:35PM +0200, Jarek Poplawski wrote:
> Plamen Petrov wrote, On 20.08.2010 12:53:
> > So, I guess its David and Herbert's turn?...
>
> If you're bored in the meantime I'd suggest to do check the realtek
> driver eg:
> - for locking with the patch below,
> - to turn off with ethtool its tx-checksumming and/or scatter-gather,
After rethinking, it's almost impossible this patch could change
anything here, so don't bother, but consider mainly the second
proposal.
Jarek P.
^ permalink raw reply
* Re: [PATCH v8] PPTP: PPP over IPv4 (Point-to-Point Tunneling Protocol)
From: Eric Dumazet @ 2010-08-21 7:49 UTC (permalink / raw)
To: Dmitry Kozlov; +Cc: netdev
In-Reply-To: <E1Omhkb-0005fi-00.xeb-mail-ru@f117.mail.ru>
Le samedi 21 août 2010 à 10:40 +0400, Dmitry Kozlov a écrit :
> PPP: introduce "pptp" module which implements point-to-point tunneling protocol using pppox framework
> NET: introduce the "gre" module for demultiplexing GRE packets on version criteria
> (required to pptp and ip_gre may coexists)
> NET: ip_gre: update to use the "gre" module
>
> Signed-off-by: Dmitry Kozlov <xeb@mail.ru>
> ---
> This patch introduces then pptp support to the linux kernel which
> dramatically speeds up pptp vpn connections and decreases cpu usage in
> comparison of existing user-space implementation (poptop/pptpclient).
> There is accel-pptp project
> (https://sourceforge.net/projects/accel-pptp/) to utilize this module,
> it contains plugin for pppd to use pptp in client-mode and modified
> pptpd (poptop) to build high-performance pptp NAS.
>
> There was many changes from initial submitted patch, most important
> are:
> 1. using rcu instead of read-write locks
> 2. using static bitmap instead of dynamically allocated
> 3. using vmalloc for memory allocation instead of BITS_PER_LONG +
> __get_free_pages
> 4. fixed many coding style issues
> Thanks to Eric Dumazet.
>
> --
> MAINTAINERS | 14 +
> drivers/net/Kconfig | 11 +
> drivers/net/Makefile | 1 +
> drivers/net/pptp.c | 726 ++++++++++++++++++++++++++++++++++++++++++++++
> include/linux/if_pppox.h | 59 +++--
> include/net/gre.h | 18 ++
> net/ipv4/Kconfig | 7 +
> net/ipv4/Makefile | 1 +
> net/ipv4/gre.c | 151 ++++++++++
> net/ipv4/ip_gre.c | 14 +-
> 10 files changed, 975 insertions(+), 27 deletions(-)
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Thanks Dmitry !
^ permalink raw reply
* Re: [Bugme-new] [Bug 16626] New: Machine hangs with EIP at skb_copy_and_csum_dev
From: Eric Dumazet @ 2010-08-21 7:50 UTC (permalink / raw)
To: Jarek Poplawski
Cc: Plamen Petrov, Andrew Morton, netdev, bugzilla-daemon,
bugme-daemon
In-Reply-To: <20100821074742.GA2367@del.dom.local>
Le samedi 21 août 2010 à 09:47 +0200, Jarek Poplawski a écrit :
> On Fri, Aug 20, 2010 at 09:38:35PM +0200, Jarek Poplawski wrote:
> > Plamen Petrov wrote, On 20.08.2010 12:53:
> > > So, I guess its David and Herbert's turn?...
> >
> > If you're bored in the meantime I'd suggest to do check the realtek
> > driver eg:
> > - for locking with the patch below,
> > - to turn off with ethtool its tx-checksumming and/or scatter-gather,
>
> After rethinking, it's almost impossible this patch could change
> anything here, so don't bother, but consider mainly the second
> proposal.
>
> Jarek P.
Indeed ;)
Its true that not many nics use the skb_copy_and_csum_dev() helper,
maybe this one must be updated somehow ?
^ permalink raw reply
* Re: [Bugme-new] [Bug 16626] New: Machine hangs with EIP at skb_copy_and_csum_dev
From: Jarek Poplawski @ 2010-08-21 8:07 UTC (permalink / raw)
To: Eric Dumazet
Cc: Plamen Petrov, Andrew Morton, netdev, bugzilla-daemon,
bugme-daemon
In-Reply-To: <1282377058.2636.12.camel@edumazet-laptop>
On Sat, Aug 21, 2010 at 09:50:58AM +0200, Eric Dumazet wrote:
> Le samedi 21 août 2010 à 09:47 +0200, Jarek Poplawski a écrit :
> > On Fri, Aug 20, 2010 at 09:38:35PM +0200, Jarek Poplawski wrote:
> > > Plamen Petrov wrote, On 20.08.2010 12:53:
> > > > So, I guess its David and Herbert's turn?...
> > >
> > > If you're bored in the meantime I'd suggest to do check the realtek
> > > driver eg:
> > > - for locking with the patch below,
> > > - to turn off with ethtool its tx-checksumming and/or scatter-gather,
> >
> > After rethinking, it's almost impossible this patch could change
> > anything here, so don't bother, but consider mainly the second
> > proposal.
> >
> > Jarek P.
>
> Indeed ;)
>
> Its true that not many nics use the skb_copy_and_csum_dev() helper,
> maybe this one must be updated somehow ?
>
Yes, it seems it should be possible at least to handle the bug with
a warning and error return, considering Plamen's problems with getting
the trace.
Jarek P.
^ permalink raw reply
* Re: [PATCH v8] PPTP: PPP over IPv4 (Point-to-Point Tunneling Protocol)
From: Dmitry Kozlov @ 2010-08-21 9:17 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev
In-Reply-To: <1282376963.2636.10.camel@edumazet-laptop>
> > PPP: introduce "pptp" module which implements point-to-point tunneling protocol using pppox framework
> > NET: introduce the "gre" module for demultiplexing GRE packets on version criteria
> > (required to pptp and ip_gre may coexists)
> > NET: ip_gre: update to use the "gre" module
> >
> > Signed-off-by: Dmitry Kozlov <xeb@mail.ru>
> > ---
> > This patch introduces then pptp support to the linux kernel which
> > dramatically speeds up pptp vpn connections and decreases cpu usage in
> > comparison of existing user-space implementation (poptop/pptpclient).
> > There is accel-pptp project
> > (https://sourceforge.net/projects/accel-pptp/) to utilize this module,
> > it contains plugin for pppd to use pptp in client-mode and modified
> > pptpd (poptop) to build high-performance pptp NAS.
> >
> > There was many changes from initial submitted patch, most important
> > are:
> > 1. using rcu instead of read-write locks
> > 2. using static bitmap instead of dynamically allocated
> > 3. using vmalloc for memory allocation instead of BITS_PER_LONG +
> > __get_free_pages
> > 4. fixed many coding style issues
> > Thanks to Eric Dumazet.
> >
> > --
> > MAINTAINERS | 14 +
> > drivers/net/Kconfig | 11 +
> > drivers/net/Makefile | 1 +
> > drivers/net/pptp.c | 726 ++++++++++++++++++++++++++++++++++++++++++++++
> > include/linux/if_pppox.h | 59 +++--
> > include/net/gre.h | 18 ++
> > net/ipv4/Kconfig | 7 +
> > net/ipv4/Makefile | 1 +
> > net/ipv4/gre.c | 151 ++++++++++
> > net/ipv4/ip_gre.c | 14 +-
> > 10 files changed, 975 insertions(+), 27 deletions(-)
>
> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
>
> Thanks Dmitry !
What is the next step ?
^ permalink raw reply
* Re: [PATCH v8] PPTP: PPP over IPv4 (Point-to-Point Tunneling Protocol)
From: Eric Dumazet @ 2010-08-21 9:31 UTC (permalink / raw)
To: Dmitry Kozlov; +Cc: netdev
In-Reply-To: <E1OmkD3-0004PV-00.xeb-mail-ru@f159.mail.ru>
Le samedi 21 août 2010 à 13:17 +0400, Dmitry Kozlov a écrit :
> What is the next step ?
David S. Miller review the thing, and if no problem remains, adds it to
its net-next-2.6 tree
You can check netdev list of current patches right here :
http://patchwork.ozlabs.org/project/netdev/list/
Your patch is already there ;)
^ permalink raw reply
* [PATCH] mlx4: use bitmap library
From: Akinobu Mita @ 2010-08-21 9:24 UTC (permalink / raw)
To: netdev; +Cc: Akinobu Mita, Roland Dreier, David S. Miller
Use bitmap_set() and bitmap_clear()
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Roland Dreier <rolandd@cisco.com>
Cc: netdev@vger.kernel.org
Cc: "David S. Miller" <davem@davemloft.net>
---
drivers/net/mlx4/alloc.c | 15 ++++-----------
1 files changed, 4 insertions(+), 11 deletions(-)
diff --git a/drivers/net/mlx4/alloc.c b/drivers/net/mlx4/alloc.c
index 8c85156..76595bd 100644
--- a/drivers/net/mlx4/alloc.c
+++ b/drivers/net/mlx4/alloc.c
@@ -74,7 +74,7 @@ void mlx4_bitmap_free(struct mlx4_bitmap *bitmap, u32 obj)
u32 mlx4_bitmap_alloc_range(struct mlx4_bitmap *bitmap, int cnt, int align)
{
- u32 obj, i;
+ u32 obj;
if (likely(cnt == 1 && align == 1))
return mlx4_bitmap_alloc(bitmap);
@@ -91,8 +91,7 @@ u32 mlx4_bitmap_alloc_range(struct mlx4_bitmap *bitmap, int cnt, int align)
}
if (obj < bitmap->max) {
- for (i = 0; i < cnt; i++)
- set_bit(obj + i, bitmap->table);
+ bitmap_set(bitmap->table, obj, cnt);
if (obj == bitmap->last) {
bitmap->last = (obj + cnt);
if (bitmap->last >= bitmap->max)
@@ -109,13 +108,10 @@ u32 mlx4_bitmap_alloc_range(struct mlx4_bitmap *bitmap, int cnt, int align)
void mlx4_bitmap_free_range(struct mlx4_bitmap *bitmap, u32 obj, int cnt)
{
- u32 i;
-
obj &= bitmap->max + bitmap->reserved_top - 1;
spin_lock(&bitmap->lock);
- for (i = 0; i < cnt; i++)
- clear_bit(obj + i, bitmap->table);
+ bitmap_clear(bitmap->table, obj, cnt);
bitmap->last = min(bitmap->last, obj);
bitmap->top = (bitmap->top + bitmap->max + bitmap->reserved_top)
& bitmap->mask;
@@ -125,8 +121,6 @@ void mlx4_bitmap_free_range(struct mlx4_bitmap *bitmap, u32 obj, int cnt)
int mlx4_bitmap_init(struct mlx4_bitmap *bitmap, u32 num, u32 mask,
u32 reserved_bot, u32 reserved_top)
{
- int i;
-
/* num must be a power of 2 */
if (num != roundup_pow_of_two(num))
return -EINVAL;
@@ -142,8 +136,7 @@ int mlx4_bitmap_init(struct mlx4_bitmap *bitmap, u32 num, u32 mask,
if (!bitmap->table)
return -ENOMEM;
- for (i = 0; i < reserved_bot; ++i)
- set_bit(i, bitmap->table);
+ bitmap_set(bitmap->table, 0, reserved_bot);
return 0;
}
--
1.6.0.6
^ permalink raw reply related
* Re: [PATCH v8] PPTP: PPP over IPv4 (Point-to-Point Tunneling Protocol)
From: Dmitry Kozlov @ 2010-08-21 11:24 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev
In-Reply-To: <1282383066.2636.16.camel@edumazet-laptop>
> > What is the next step ?
>
> David S. Miller review the thing, and if no problem remains, adds it to
> its net-next-2.6 tree
>
> You can check netdev list of current patches right here :
>
> http://patchwork.ozlabs.org/project/netdev/list/
>
> Your patch is already there ;)
Ah ok, thanks!
I see some unexpected symbols in mail's subject, maybe i have to resend mail with subject cleaned ?
^ permalink raw reply
* About unix_autobind()
From: Tetsuo Handa @ 2010-08-21 12:01 UTC (permalink / raw)
To: netdev
I was browsing unix_autobind() and wondered what happens if all names in
Unix domain socket's abstract namespace were in use.
Below part is unix_autobind() from linux-2.6.36-rc1/net/unix/af_unix.c
710 retry:
711 addr->len = sprintf(addr->name->sun_path+1, "%05x", ordernum) + 1 + sizeof(short);
712 addr->hash = unix_hash_fold(csum_partial(addr->name, addr->len, 0));
713
714 spin_lock(&unix_table_lock);
715 ordernum = (ordernum+1)&0xFFFFF;
716
717 if (__unix_find_socket_byname(net, addr->name, addr->len, sock->type,
718 addr->hash)) {
719 spin_unlock(&unix_table_lock);
720 /* Sanity yield. It is unusual case, but yet... */
721 if (!(ordernum&0xFF))
722 yield();
723 goto retry;
724 }
We can see that unix_autobind() allows 1048576 names.
A machine with 256MB RAM:
# cat /proc/sys/fs/file-max
24109
# cat /proc/sys/fs/file-nr
608 0 24109
A machine with 1736MB RAM:
# cat /proc/sys/fs/file-max
174347
# cat /proc/sys/fs/file-nr
96 0 174347
/proc/sys/fs/file-max seems to be proportional to the amount of RAM.
I don't have access to a machine with 10GB RAM (where /proc/sys/fs/file-max
becomes larger than 1048576 by default). So, I manually set
# echo 1050000 > /proc/sys/fs/file-max
and executed below program as non-root user.
---------- Test program start ----------
#include <sys/socket.h>
#include <sys/un.h>
#include <unistd.h>
int main(int argc, char *argv[])
{
int i;
for (i = 0; i < 1030; i++) {
switch (fork()) {
case 0:
sleep(5);
close(0);
close(1);
close(2);
for (i = 0; i < 1024; i++) {
struct sockaddr_un addr;
int fd = socket(PF_UNIX, SOCK_DGRAM, 0);
addr.sun_family = AF_UNIX;
bind(fd, (struct sockaddr *) &addr, sizeof(addr.sun_family));
}
while (1)
sleep(1000);
case -1:
write(1, "fork() failed\n", 14);
return 1;
}
}
return 0;
}
---------- Test program end ----------
If there is not enough memory, OOM killer was invoked. OOM killer killed
/usr/sbin/httpd process rather than above test program. (Oops. Non-root user
was able to terminate other user's processes via OOM killer.)
If there is enough memory (I can't test it), OOM killer will not be invoked.
But if all names were occupied, I guess subsequent unix_autobind() by other
users will loop forever because it loops until a name becomes available.
(I had to type "killall a.out" before above program occupies all names, for
the machine became very dull due to unix_autobind().)
Maybe some safeguard is wanted.
^ permalink raw reply
* Re: [PATCH v8] PPTP: PPP over IPv4 (Point-to-Point Tunneling Protocol)
From: xeb @ 2010-08-21 10:09 UTC (permalink / raw)
To: netdev, Eric Dumazet
> > What is the next step ?
>
> David S. Miller review the thing, and if no problem remains, adds it to
> its net-next-2.6 tree
>
> You can check netdev list of current patches right here :
>
> http://patchwork.ozlabs.org/project/netdev/list/
>
> Your patch is already there ;)
Ah ok, thanks!
I see my mail's subject contains some unexpected symbols, maybe i have to
resend patch with cleaned subject ?
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox