* [PATCH 08/14] netvm: Allow skb allocation to use PFMEMALLOC reserves
From: Mel Gorman @ 2011-06-20 13:12 UTC (permalink / raw)
To: Andrew Morton
Cc: Linux-MM, Linux-Netdev, LKML, David Miller, Neil Brown,
Peter Zijlstra, Mel Gorman
In-Reply-To: <1308575540-25219-1-git-send-email-mgorman@suse.de>
Change the skb allocation API to indicate RX usage and use this to fall back
to the PFMEMALLOC reserve when needed. SKBs allocated from the reserve are
tagged in skb->pfmemalloc. If an SKB is allocated from the reserve and
the socket is later found to be unrelated to page reclaim, the packet is
dropped so that the memory remains available for page reclaim. Network
protocols are expected to recover from this packet loss.
[a.p.zijlstra@chello.nl: Ideas taken from various patches]
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
include/linux/gfp.h | 3 ++
include/linux/skbuff.h | 19 ++++++++--
include/net/sock.h | 6 +++
mm/internal.h | 3 --
net/core/filter.c | 8 ++++
net/core/skbuff.c | 95 ++++++++++++++++++++++++++++++++++++++++--------
net/core/sock.c | 4 ++
7 files changed, 116 insertions(+), 22 deletions(-)
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index faefba9..3cc24bb 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -375,6 +375,9 @@ void drain_local_pages(void *dummy);
extern gfp_t gfp_allowed_mask;
+/* Returns true if the gfp_mask allows use of ALLOC_NO_WATERMARK */
+bool gfp_pfmemalloc_allowed(gfp_t gfp_mask);
+
extern void pm_restrict_gfp_mask(void);
extern void pm_restore_gfp_mask(void);
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index c0a4f3a..064d8d4 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -396,6 +396,7 @@ struct sk_buff {
#ifdef CONFIG_IPV6_NDISC_NODETYPE
__u8 ndisc_nodetype:2;
#endif
+ __u8 pfmemalloc:1;
__u8 ooo_okay:1;
kmemcheck_bitfield_end(flags2);
@@ -434,6 +435,15 @@ struct sk_buff {
#include <asm/system.h>
+#define SKB_ALLOC_FCLONE 0x01
+#define SKB_ALLOC_RX 0x02
+
+/* Returns true if the skb was allocated from PFMEMALLOC reserves */
+static inline bool skb_pfmemalloc(struct sk_buff *skb)
+{
+ return unlikely(skb->pfmemalloc);
+}
+
/*
* skb might have a dst pointer attached, refcounted or not.
* _skb_refdst low order bit is set if refcount was _not_ taken
@@ -491,7 +501,7 @@ extern void kfree_skb(struct sk_buff *skb);
extern void consume_skb(struct sk_buff *skb);
extern void __kfree_skb(struct sk_buff *skb);
extern struct sk_buff *__alloc_skb(unsigned int size,
- gfp_t priority, int fclone, int node);
+ gfp_t priority, int flags, int node);
static inline struct sk_buff *alloc_skb(unsigned int size,
gfp_t priority)
{
@@ -501,7 +511,7 @@ static inline struct sk_buff *alloc_skb(unsigned int size,
static inline struct sk_buff *alloc_skb_fclone(unsigned int size,
gfp_t priority)
{
- return __alloc_skb(size, priority, 1, NUMA_NO_NODE);
+ return __alloc_skb(size, priority, SKB_ALLOC_FCLONE, NUMA_NO_NODE);
}
extern bool skb_recycle_check(struct sk_buff *skb, int skb_size);
@@ -1532,7 +1542,8 @@ static inline void __skb_queue_purge(struct sk_buff_head *list)
static inline struct sk_buff *__dev_alloc_skb(unsigned int length,
gfp_t gfp_mask)
{
- struct sk_buff *skb = alloc_skb(length + NET_SKB_PAD, gfp_mask);
+ struct sk_buff *skb = __alloc_skb(length + NET_SKB_PAD, gfp_mask,
+ SKB_ALLOC_RX, NUMA_NO_NODE);
if (likely(skb))
skb_reserve(skb, NET_SKB_PAD);
return skb;
@@ -1583,7 +1594,7 @@ static inline struct sk_buff *netdev_alloc_skb_ip_align(struct net_device *dev,
*/
static inline struct page *__netdev_alloc_page(struct net_device *dev, gfp_t gfp_mask)
{
- return alloc_pages_node(NUMA_NO_NODE, gfp_mask, 0);
+ return alloc_pages_node(NUMA_NO_NODE, gfp_mask | __GFP_MEMALLOC, 0);
}
/**
diff --git a/include/net/sock.h b/include/net/sock.h
index 046bc97..e3aaa88 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -586,6 +586,12 @@ static inline int sock_flag(struct sock *sk, enum sock_flags flag)
return test_bit(flag, &sk->sk_flags);
}
+extern atomic_t memalloc_socks;
+static inline int sk_memalloc_socks(void)
+{
+ return atomic_read(&memalloc_socks);
+}
+
static inline gfp_t sk_allocation(struct sock *sk, gfp_t gfp_mask)
{
return gfp_mask | (sk->sk_allocation & __GFP_MEMALLOC);
diff --git a/mm/internal.h b/mm/internal.h
index a520f3b..d071d380 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -193,9 +193,6 @@ static inline struct page *mem_map_next(struct page *iter,
#define __paginginit __init
#endif
-/* Returns true if the gfp_mask allows use of ALLOC_NO_WATERMARK */
-bool gfp_pfmemalloc_allowed(gfp_t gfp_mask);
-
/* Memory initialisation debug and verification */
enum mminit_level {
MMINIT_WARNING,
diff --git a/net/core/filter.c b/net/core/filter.c
index 36f975f..4ccf6f4 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -80,6 +80,14 @@ int sk_filter(struct sock *sk, struct sk_buff *skb)
int err;
struct sk_filter *filter;
+ /*
+ * If the skb was allocated from pfmemalloc reserves, only
+ * allow SOCK_MEMALLOC sockets to use it as this socket is
+ * helping free memory
+ */
+ if (skb_pfmemalloc(skb) && !sock_flag(sk, SOCK_MEMALLOC))
+ return -ENOMEM;
+
err = security_sock_rcv_skb(sk, skb);
if (err)
return err;
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 46cbd28..d930b92 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -147,6 +147,43 @@ static void skb_under_panic(struct sk_buff *skb, int sz, void *here)
BUG();
}
+
+/*
+ * kmalloc_reserve is a wrapper around kmalloc_node_track_caller that tells
+ * the caller if emergency pfmemalloc reserves are being used. If it is and
+ * the socket is later found to be SOCK_MEMALLOC then PFMEMALLOC reserves
+ * may be used. Otherwise, the packet data may be discarded until enough
+ * memory is free
+ */
+#define kmalloc_reserve(size, gfp, node, pfmemalloc) \
+ __kmalloc_reserve(size, gfp, node, _RET_IP_, pfmemalloc)
+void *__kmalloc_reserve(size_t size, gfp_t flags, int node, unsigned long ip,
+ bool *pfmemalloc)
+{
+ void *obj;
+ bool ret_pfmemalloc = false;
+
+ /*
+ * Try a regular allocation, when that fails and we're not entitled
+ * to the reserves, fail.
+ */
+ obj = kmalloc_node_track_caller(size,
+ flags | __GFP_NOMEMALLOC | __GFP_NOWARN,
+ node);
+ if (obj || !(gfp_pfmemalloc_allowed(flags)))
+ goto out;
+
+ /* Try again but now we are using pfmemalloc reserves */
+ ret_pfmemalloc = true;
+ obj = kmalloc_node_track_caller(size, flags, node);
+
+out:
+ if (pfmemalloc)
+ *pfmemalloc = ret_pfmemalloc;
+
+ return obj;
+}
+
/* Allocate a new skbuff. We do this ourselves so we can fill in a few
* 'private' fields and also do memory statistics to find all the
* [BEEP] leaks.
@@ -157,8 +194,10 @@ static void skb_under_panic(struct sk_buff *skb, int sz, void *here)
* __alloc_skb - allocate a network buffer
* @size: size to allocate
* @gfp_mask: allocation mask
- * @fclone: allocate from fclone cache instead of head cache
- * and allocate a cloned (child) skb
+ * @flags: If SKB_ALLOC_FCLONE is set, allocate from fclone cache
+ * instead of head cache and allocate a cloned (child) skb.
+ * If SKB_ALLOC_RX is set, __GFP_MEMALLOC will be used for
+ * allocations in case the data is required for writeback
* @node: numa node to allocate memory on
*
* Allocate a new &sk_buff. The returned buffer has no headroom and a
@@ -169,14 +208,19 @@ static void skb_under_panic(struct sk_buff *skb, int sz, void *here)
* %GFP_ATOMIC.
*/
struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask,
- int fclone, int node)
+ int flags, int node)
{
struct kmem_cache *cache;
struct skb_shared_info *shinfo;
struct sk_buff *skb;
u8 *data;
+ bool pfmemalloc;
+
+ cache = (flags & SKB_ALLOC_FCLONE)
+ ? skbuff_fclone_cache : skbuff_head_cache;
- cache = fclone ? skbuff_fclone_cache : skbuff_head_cache;
+ if (sk_memalloc_socks() && (flags & SKB_ALLOC_RX))
+ gfp_mask |= __GFP_MEMALLOC;
/* Get the HEAD */
skb = kmem_cache_alloc_node(cache, gfp_mask & ~__GFP_DMA, node);
@@ -185,8 +229,8 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask,
prefetchw(skb);
size = SKB_DATA_ALIGN(size);
- data = kmalloc_node_track_caller(size + sizeof(struct skb_shared_info),
- gfp_mask, node);
+ data = kmalloc_reserve(size + sizeof(struct skb_shared_info),
+ gfp_mask, node, &pfmemalloc);
if (!data)
goto nodata;
prefetchw(data + size);
@@ -197,6 +241,7 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask,
* the tail pointer in struct sk_buff!
*/
memset(skb, 0, offsetof(struct sk_buff, tail));
+ skb->pfmemalloc = pfmemalloc;
skb->truesize = size + sizeof(struct sk_buff);
atomic_set(&skb->users, 1);
skb->head = data;
@@ -213,7 +258,7 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask,
atomic_set(&shinfo->dataref, 1);
kmemcheck_annotate_variable(shinfo->destructor_arg);
- if (fclone) {
+ if (flags & SKB_ALLOC_FCLONE) {
struct sk_buff *child = skb + 1;
atomic_t *fclone_ref = (atomic_t *) (child + 1);
@@ -223,6 +268,7 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask,
atomic_set(fclone_ref, 1);
child->fclone = SKB_FCLONE_UNAVAILABLE;
+ child->pfmemalloc = pfmemalloc;
}
out:
return skb;
@@ -251,7 +297,8 @@ struct sk_buff *__netdev_alloc_skb(struct net_device *dev,
{
struct sk_buff *skb;
- skb = __alloc_skb(length + NET_SKB_PAD, gfp_mask, 0, NUMA_NO_NODE);
+ skb = __alloc_skb(length + NET_SKB_PAD, gfp_mask,
+ SKB_ALLOC_RX, NUMA_NO_NODE);
if (likely(skb)) {
skb_reserve(skb, NET_SKB_PAD);
skb->dev = dev;
@@ -527,6 +574,7 @@ static void __copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
#if defined(CONFIG_IP_VS) || defined(CONFIG_IP_VS_MODULE)
new->ipvs_property = old->ipvs_property;
#endif
+ new->pfmemalloc = old->pfmemalloc;
new->protocol = old->protocol;
new->mark = old->mark;
new->skb_iif = old->skb_iif;
@@ -621,6 +669,9 @@ struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask)
n->fclone = SKB_FCLONE_CLONE;
atomic_inc(fclone_ref);
} else {
+ if (skb_pfmemalloc(skb))
+ gfp_mask |= __GFP_MEMALLOC;
+
n = kmem_cache_alloc(skbuff_head_cache, gfp_mask);
if (!n)
return NULL;
@@ -657,6 +708,13 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
skb_shinfo(new)->gso_type = skb_shinfo(old)->gso_type;
}
+static inline int skb_alloc_rx_flag(const struct sk_buff *skb)
+{
+ if (skb_pfmemalloc((struct sk_buff *)skb))
+ return SKB_ALLOC_RX;
+ return 0;
+}
+
/**
* skb_copy - create private copy of an sk_buff
* @skb: buffer to copy
@@ -678,7 +736,8 @@ struct sk_buff *skb_copy(const struct sk_buff *skb, gfp_t gfp_mask)
{
int headerlen = skb_headroom(skb);
unsigned int size = (skb_end_pointer(skb) - skb->head) + skb->data_len;
- struct sk_buff *n = alloc_skb(size, gfp_mask);
+ struct sk_buff *n = __alloc_skb(size, gfp_mask,
+ skb_alloc_rx_flag(skb), NUMA_NO_NODE);
if (!n)
return NULL;
@@ -712,7 +771,8 @@ EXPORT_SYMBOL(skb_copy);
struct sk_buff *pskb_copy(struct sk_buff *skb, gfp_t gfp_mask)
{
unsigned int size = skb_end_pointer(skb) - skb->head;
- struct sk_buff *n = alloc_skb(size, gfp_mask);
+ struct sk_buff *n = __alloc_skb(size, gfp_mask,
+ skb_alloc_rx_flag(skb), NUMA_NO_NODE);
if (!n)
goto out;
@@ -803,7 +863,10 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail,
goto adjust_others;
}
- data = kmalloc(size + sizeof(struct skb_shared_info), gfp_mask);
+ if (skb_pfmemalloc(skb))
+ gfp_mask |= __GFP_MEMALLOC;
+ data = kmalloc_reserve(size + sizeof(struct skb_shared_info), gfp_mask,
+ NUMA_NO_NODE, NULL);
if (!data)
goto nodata;
@@ -904,8 +967,9 @@ struct sk_buff *skb_copy_expand(const struct sk_buff *skb,
/*
* Allocate the copy buffer
*/
- struct sk_buff *n = alloc_skb(newheadroom + skb->len + newtailroom,
- gfp_mask);
+ struct sk_buff *n = __alloc_skb(newheadroom + skb->len + newtailroom,
+ gfp_mask, skb_alloc_rx_flag(skb),
+ NUMA_NO_NODE);
int oldheadroom = skb_headroom(skb);
int head_copy_len, head_copy_off;
int off;
@@ -2552,8 +2616,9 @@ struct sk_buff *skb_segment(struct sk_buff *skb, u32 features)
skb_release_head_state(nskb);
__skb_push(nskb, doffset);
} else {
- nskb = alloc_skb(hsize + doffset + headroom,
- GFP_ATOMIC);
+ nskb = __alloc_skb(hsize + doffset + headroom,
+ GFP_ATOMIC, skb_alloc_rx_flag(skb),
+ NUMA_NO_NODE);
if (unlikely(!nskb))
goto err;
diff --git a/net/core/sock.c b/net/core/sock.c
index c685eda..8308609 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -219,6 +219,8 @@ __u32 sysctl_rmem_default __read_mostly = SK_RMEM_MAX;
int sysctl_optmem_max __read_mostly = sizeof(unsigned long)*(2*UIO_MAXIOV+512);
EXPORT_SYMBOL(sysctl_optmem_max);
+atomic_t memalloc_socks __read_mostly;
+
/**
* sk_set_memalloc - sets %SOCK_MEMALLOC
* @sk: socket to set it on
@@ -231,6 +233,7 @@ void sk_set_memalloc(struct sock *sk)
{
sock_set_flag(sk, SOCK_MEMALLOC);
sk->sk_allocation |= __GFP_MEMALLOC;
+ atomic_inc(&memalloc_socks);
}
EXPORT_SYMBOL_GPL(sk_set_memalloc);
@@ -238,6 +241,7 @@ void sk_clear_memalloc(struct sock *sk)
{
sock_reset_flag(sk, SOCK_MEMALLOC);
sk->sk_allocation &= ~__GFP_MEMALLOC;
+ atomic_dec(&memalloc_socks);
}
EXPORT_SYMBOL_GPL(sk_clear_memalloc);
--
1.7.3.4
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related
* [PATCH 07/14] netvm: Allow the use of __GFP_MEMALLOC by specific sockets
From: Mel Gorman @ 2011-06-20 13:12 UTC (permalink / raw)
To: Andrew Morton
Cc: Linux-MM, Linux-Netdev, LKML, David Miller, Neil Brown,
Peter Zijlstra, Mel Gorman
In-Reply-To: <1308575540-25219-1-git-send-email-mgorman@suse.de>
Allow specific sockets to be tagged SOCK_MEMALLOC and use __GFP_MEMALLOC
for their allocations. These sockets will be able to go below watermarks
and allocate from the emergency reserve. Such sockets are to be used
to service the VM (iow. to swap over). They must be handled kernel side,
exposing such a socket to user-space is a bug.
There is a risk that the reserves be depleted so for now, the administrator is
responsible for increasing min_free_kbytes as necessary to prevent deadlock
for their workloads.
[a.p.zijlstra@chello.nl: Original patches]
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
include/net/sock.h | 5 ++++-
net/core/sock.c | 22 ++++++++++++++++++++++
2 files changed, 26 insertions(+), 1 deletions(-)
diff --git a/include/net/sock.h b/include/net/sock.h
index e89c38f..046bc97 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -554,6 +554,7 @@ enum sock_flags {
SOCK_RCVTSTAMPNS, /* %SO_TIMESTAMPNS setting */
SOCK_LOCALROUTE, /* route locally only, %SO_DONTROUTE setting */
SOCK_QUEUE_SHRUNK, /* write queue has been shrunk recently */
+ SOCK_MEMALLOC, /* VM depends on this socket for swapping */
SOCK_TIMESTAMPING_TX_HARDWARE, /* %SOF_TIMESTAMPING_TX_HARDWARE */
SOCK_TIMESTAMPING_TX_SOFTWARE, /* %SOF_TIMESTAMPING_TX_SOFTWARE */
SOCK_TIMESTAMPING_RX_HARDWARE, /* %SOF_TIMESTAMPING_RX_HARDWARE */
@@ -587,7 +588,7 @@ static inline int sock_flag(struct sock *sk, enum sock_flags flag)
static inline gfp_t sk_allocation(struct sock *sk, gfp_t gfp_mask)
{
- return gfp_mask;
+ return gfp_mask | (sk->sk_allocation & __GFP_MEMALLOC);
}
static inline void sk_acceptq_removed(struct sock *sk)
@@ -717,6 +718,8 @@ extern int sk_stream_wait_memory(struct sock *sk, long *timeo_p);
extern void sk_stream_wait_close(struct sock *sk, long timeo_p);
extern int sk_stream_error(struct sock *sk, int flags, int err);
extern void sk_stream_kill_queues(struct sock *sk);
+extern void sk_set_memalloc(struct sock *sk);
+extern void sk_clear_memalloc(struct sock *sk);
extern int sk_wait_data(struct sock *sk, long *timeo);
diff --git a/net/core/sock.c b/net/core/sock.c
index 6e81978..c685eda 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -219,6 +219,28 @@ __u32 sysctl_rmem_default __read_mostly = SK_RMEM_MAX;
int sysctl_optmem_max __read_mostly = sizeof(unsigned long)*(2*UIO_MAXIOV+512);
EXPORT_SYMBOL(sysctl_optmem_max);
+/**
+ * sk_set_memalloc - sets %SOCK_MEMALLOC
+ * @sk: socket to set it on
+ *
+ * Set %SOCK_MEMALLOC on a socket for access to emergency reserves.
+ * It's the responsibility of the admin to adjust min_free_kbytes
+ * to meet the requirements
+ */
+void sk_set_memalloc(struct sock *sk)
+{
+ sock_set_flag(sk, SOCK_MEMALLOC);
+ sk->sk_allocation |= __GFP_MEMALLOC;
+}
+EXPORT_SYMBOL_GPL(sk_set_memalloc);
+
+void sk_clear_memalloc(struct sock *sk)
+{
+ sock_reset_flag(sk, SOCK_MEMALLOC);
+ sk->sk_allocation &= ~__GFP_MEMALLOC;
+}
+EXPORT_SYMBOL_GPL(sk_clear_memalloc);
+
#if defined(CONFIG_CGROUPS) && !defined(CONFIG_NET_CLS_CGROUP)
int net_cls_subsys_id = -1;
EXPORT_SYMBOL_GPL(net_cls_subsys_id);
--
1.7.3.4
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related
* [PATCH 06/14] net: Introduce sk_allocation() to allow addition of GFP flags depending on the individual socket
From: Mel Gorman @ 2011-06-20 13:12 UTC (permalink / raw)
To: Andrew Morton
Cc: Linux-MM, Linux-Netdev, LKML, David Miller, Neil Brown,
Peter Zijlstra, Mel Gorman
In-Reply-To: <1308575540-25219-1-git-send-email-mgorman@suse.de>
Introduce sk_allocation(), this function allows to inject sock specific
flags to each sock related allocation. It is only used on allocation
paths that may be required for writing pages back to network storage.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
include/net/sock.h | 5 +++++
net/ipv4/tcp.c | 3 ++-
net/ipv4/tcp_output.c | 13 +++++++------
net/ipv6/tcp_ipv6.c | 12 +++++++++---
4 files changed, 23 insertions(+), 10 deletions(-)
diff --git a/include/net/sock.h b/include/net/sock.h
index f2046e4..e89c38f 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -585,6 +585,11 @@ static inline int sock_flag(struct sock *sk, enum sock_flags flag)
return test_bit(flag, &sk->sk_flags);
}
+static inline gfp_t sk_allocation(struct sock *sk, gfp_t gfp_mask)
+{
+ return gfp_mask;
+}
+
static inline void sk_acceptq_removed(struct sock *sk)
{
sk->sk_ack_backlog--;
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 054a59d..8c1a9d5 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -698,7 +698,8 @@ struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp)
/* The TCP header must be at least 32-bit aligned. */
size = ALIGN(size, 4);
- skb = alloc_skb_fclone(size + sk->sk_prot->max_header, gfp);
+ skb = alloc_skb_fclone(size + sk->sk_prot->max_header,
+ sk_allocation(sk, gfp));
if (skb) {
if (sk_wmem_schedule(sk, skb->truesize)) {
/*
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 882e0b0..87b98f6 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -2324,7 +2324,7 @@ void tcp_send_fin(struct sock *sk)
/* Socket is locked, keep trying until memory is available. */
for (;;) {
skb = alloc_skb_fclone(MAX_TCP_HEADER,
- sk->sk_allocation);
+ sk_allocation(sk, GFP_KERNEL));
if (skb)
break;
yield();
@@ -2350,7 +2350,7 @@ void tcp_send_active_reset(struct sock *sk, gfp_t priority)
struct sk_buff *skb;
/* NOTE: No TCP options attached and we never retransmit this. */
- skb = alloc_skb(MAX_TCP_HEADER, priority);
+ skb = alloc_skb(MAX_TCP_HEADER, sk_allocation(sk, priority));
if (!skb) {
NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTFAILED);
return;
@@ -2423,7 +2423,8 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst,
if (cvp != NULL && cvp->s_data_constant && cvp->s_data_desired)
s_data_desired = cvp->s_data_desired;
- skb = sock_wmalloc(sk, MAX_TCP_HEADER + 15 + s_data_desired, 1, GFP_ATOMIC);
+ skb = sock_wmalloc(sk, MAX_TCP_HEADER + 15 + s_data_desired, 1,
+ sk_allocation(sk, GFP_ATOMIC));
if (skb == NULL)
return NULL;
@@ -2719,7 +2720,7 @@ void tcp_send_ack(struct sock *sk)
* tcp_transmit_skb() will set the ownership to this
* sock.
*/
- buff = alloc_skb(MAX_TCP_HEADER, GFP_ATOMIC);
+ buff = alloc_skb(MAX_TCP_HEADER, sk_allocation(sk, GFP_ATOMIC));
if (buff == NULL) {
inet_csk_schedule_ack(sk);
inet_csk(sk)->icsk_ack.ato = TCP_ATO_MIN;
@@ -2734,7 +2735,7 @@ void tcp_send_ack(struct sock *sk)
/* Send it off, this clears delayed acks for us. */
TCP_SKB_CB(buff)->when = tcp_time_stamp;
- tcp_transmit_skb(sk, buff, 0, GFP_ATOMIC);
+ tcp_transmit_skb(sk, buff, 0, sk_allocation(sk, GFP_ATOMIC));
}
/* This routine sends a packet with an out of date sequence
@@ -2754,7 +2755,7 @@ static int tcp_xmit_probe_skb(struct sock *sk, int urgent)
struct sk_buff *skb;
/* We don't queue it, tcp_transmit_skb() sets ownership. */
- skb = alloc_skb(MAX_TCP_HEADER, GFP_ATOMIC);
+ skb = alloc_skb(MAX_TCP_HEADER, sk_allocation(sk, GFP_ATOMIC));
if (skb == NULL)
return -1;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index d1fd287..62bc424 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -597,7 +597,8 @@ static int tcp_v6_md5_do_add(struct sock *sk, const struct in6_addr *peer,
} else {
/* reallocate new list if current one is full. */
if (!tp->md5sig_info) {
- tp->md5sig_info = kzalloc(sizeof(*tp->md5sig_info), GFP_ATOMIC);
+ tp->md5sig_info = kzalloc(sizeof(*tp->md5sig_info),
+ sk_allocation(sk, GFP_ATOMIC));
if (!tp->md5sig_info) {
kfree(newkey);
return -ENOMEM;
@@ -610,7 +611,8 @@ static int tcp_v6_md5_do_add(struct sock *sk, const struct in6_addr *peer,
}
if (tp->md5sig_info->alloced6 == tp->md5sig_info->entries6) {
keys = kmalloc((sizeof (tp->md5sig_info->keys6[0]) *
- (tp->md5sig_info->entries6 + 1)), GFP_ATOMIC);
+ (tp->md5sig_info->entries6 + 1)),
+ sk_allocation(sk, GFP_ATOMIC));
if (!keys) {
tcp_free_md5sig_pool();
@@ -734,7 +736,8 @@ static int tcp_v6_parse_md5_keys (struct sock *sk, char __user *optval,
struct tcp_sock *tp = tcp_sk(sk);
struct tcp_md5sig_info *p;
- p = kzalloc(sizeof(struct tcp_md5sig_info), GFP_KERNEL);
+ p = kzalloc(sizeof(struct tcp_md5sig_info),
+ sk_allocation(sk, GFP_KERNEL));
if (!p)
return -ENOMEM;
@@ -1084,6 +1087,7 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
struct tcphdr *th = tcp_hdr(skb);
u32 seq = 0, ack_seq = 0;
struct tcp_md5sig_key *key = NULL;
+ gfp_t gfp_mask = GFP_ATOMIC;
if (th->rst)
return;
@@ -1095,6 +1099,8 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
if (sk)
key = tcp_v6_md5_do_lookup(sk, &ipv6_hdr(skb)->daddr);
#endif
+ if (sk)
+ gfp_mask = sk_allocation(sk, gfp_mask);
if (th->ack)
seq = ntohl(th->ack_seq);
--
1.7.3.4
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related
* [PATCH 05/14] mm: Ignore mempolicies when using ALLOC_NO_WATERMARK
From: Mel Gorman @ 2011-06-20 13:12 UTC (permalink / raw)
To: Andrew Morton
Cc: Linux-MM, Linux-Netdev, LKML, David Miller, Neil Brown,
Peter Zijlstra, Mel Gorman
In-Reply-To: <1308575540-25219-1-git-send-email-mgorman@suse.de>
The reserve is proportionally distributed over all !highmem zones in the
system. So we need to allow an emergency allocation access to all zones.
In order to do that we need to break out of any mempolicy boundaries we
might have.
In my opinion that does not break mempolicies as those are user oriented
and not system oriented. That is, system allocations are not guaranteed to
be within mempolicy boundaries. For instance IRQs don't even have a mempolicy.
So breaking out of mempolicy boundaries for 'rare' emergency allocations,
which are always system allocations (as opposed to user) is ok.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
mm/page_alloc.c | 7 +++++++
1 files changed, 7 insertions(+), 0 deletions(-)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 4e19606..ac779f5 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2126,6 +2126,13 @@ rebalance:
/* Allocate without watermarks if the context allows */
if (alloc_flags & ALLOC_NO_WATERMARKS) {
+ /*
+ * Ignore mempolicies if ALLOC_NO_WATERMARKS on the grounds
+ * the allocation is high priority and these type of
+ * allocations are system rather than user orientated
+ */
+ zonelist = node_zonelist(numa_node_id(), gfp_mask);
+
page = __alloc_pages_high_priority(gfp_mask, order,
zonelist, high_zoneidx, nodemask,
preferred_zone, migratetype);
--
1.7.3.4
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related
* [PATCH 04/14] mm: allow PF_MEMALLOC from softirq context
From: Mel Gorman @ 2011-06-20 13:12 UTC (permalink / raw)
To: Andrew Morton
Cc: Linux-MM, Linux-Netdev, LKML, David Miller, Neil Brown,
Peter Zijlstra, Mel Gorman
In-Reply-To: <1308575540-25219-1-git-send-email-mgorman@suse.de>
This is needed to allow network softirq packet processing to make use
of PF_MEMALLOC.
Currently softirq context cannot use PF_MEMALLOC due to it not being
associated with a task, and therefore not having task flags to fiddle with -
thus the gfp to alloc flag mapping ignores the task flags when in interrupts
(hard or soft) context.
Allowing softirqs to make use of PF_MEMALLOC therefore requires some trickery.
We basically borrow the task flags from whatever process happens to be
preempted by the softirq.
So we modify the gfp to alloc flags mapping to not exclude task flags in
softirq context, and modify the softirq code to save, clear and restore
the PF_MEMALLOC flag.
The save and clear, ensures the preempted task's PF_MEMALLOC flag doesn't
leak into the softirq. The restore ensures a softirq's PF_MEMALLOC flag
cannot leak back into the preempted process.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
include/linux/sched.h | 7 +++++++
kernel/softirq.c | 3 +++
mm/page_alloc.c | 5 ++++-
3 files changed, 14 insertions(+), 1 deletions(-)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index a837b20..7089cf1 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1841,6 +1841,13 @@ static inline void rcu_copy_process(struct task_struct *p)
#endif
+static inline void tsk_restore_flags(struct task_struct *p,
+ unsigned long pflags, unsigned long mask)
+{
+ p->flags &= ~mask;
+ p->flags |= pflags & mask;
+}
+
#ifdef CONFIG_SMP
extern void do_set_cpus_allowed(struct task_struct *p,
const struct cpumask *new_mask);
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 1396017..2817c27 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -210,6 +210,8 @@ asmlinkage void __do_softirq(void)
__u32 pending;
int max_restart = MAX_SOFTIRQ_RESTART;
int cpu;
+ unsigned long pflags = current->flags;
+ current->flags &= ~PF_MEMALLOC;
pending = local_softirq_pending();
account_system_vtime(current);
@@ -265,6 +267,7 @@ restart:
account_system_vtime(current);
__local_bh_enable(SOFTIRQ_OFFSET);
+ tsk_restore_flags(current, pflags, PF_MEMALLOC);
}
#ifndef __ARCH_HAS_DO_SOFTIRQ
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 27043e7..4e19606 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2046,7 +2046,10 @@ gfp_to_alloc_flags(gfp_t gfp_mask)
if (likely(!(gfp_mask & __GFP_NOMEMALLOC))) {
if (gfp_mask & __GFP_MEMALLOC)
alloc_flags |= ALLOC_NO_WATERMARKS;
- else if (likely(!(gfp_mask & __GFP_NOMEMALLOC)) && !in_interrupt())
+ else if (!in_irq() && (current->flags & PF_MEMALLOC))
+ alloc_flags |= ALLOC_NO_WATERMARKS;
+ else if (!in_interrupt() &&
+ unlikely(test_thread_flag(TIF_MEMDIE)))
alloc_flags |= ALLOC_NO_WATERMARKS;
}
--
1.7.3.4
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related
* [PATCH 03/14] mm: Introduce __GFP_MEMALLOC to allow access to emergency reserves
From: Mel Gorman @ 2011-06-20 13:12 UTC (permalink / raw)
To: Andrew Morton
Cc: Linux-MM, Linux-Netdev, LKML, David Miller, Neil Brown,
Peter Zijlstra, Mel Gorman
In-Reply-To: <1308575540-25219-1-git-send-email-mgorman@suse.de>
__GFP_MEMALLOC will allow the allocation to disregard the watermarks,
much like PF_MEMALLOC. It allows one to pass along the memalloc state in
object related allocation flags as opposed to task related flags, such
as sk->sk_allocation. This removes the need for ALLOC_PFMEMALLOC as
callers using __GFP_MEMALLOC can get the ALLOC_NO_WATERMARK flag which
is now enough to identify allocations related to page reclaim.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
include/linux/gfp.h | 10 ++++++++--
include/linux/mm_types.h | 2 +-
include/trace/events/gfpflags.h | 1 +
mm/page_alloc.c | 14 ++++++--------
mm/slab.c | 2 +-
5 files changed, 17 insertions(+), 12 deletions(-)
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index cb40892..faefba9 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -23,6 +23,7 @@ struct vm_area_struct;
#define ___GFP_REPEAT 0x400u
#define ___GFP_NOFAIL 0x800u
#define ___GFP_NORETRY 0x1000u
+#define ___GFP_MEMALLOC 0x2000u
#define ___GFP_COMP 0x4000u
#define ___GFP_ZERO 0x8000u
#define ___GFP_NOMEMALLOC 0x10000u
@@ -75,9 +76,14 @@ struct vm_area_struct;
#define __GFP_REPEAT ((__force gfp_t)___GFP_REPEAT) /* See above */
#define __GFP_NOFAIL ((__force gfp_t)___GFP_NOFAIL) /* See above */
#define __GFP_NORETRY ((__force gfp_t)___GFP_NORETRY) /* See above */
+#define __GFP_MEMALLOC ((__force gfp_t)___GFP_MEMALLOC)/* Allow access to emergency reserves */
#define __GFP_COMP ((__force gfp_t)___GFP_COMP) /* Add compound page metadata */
#define __GFP_ZERO ((__force gfp_t)___GFP_ZERO) /* Return zeroed page on success */
-#define __GFP_NOMEMALLOC ((__force gfp_t)___GFP_NOMEMALLOC) /* Don't use emergency reserves */
+#define __GFP_NOMEMALLOC ((__force gfp_t)___GFP_NOMEMALLOC) /* Don't use emergency reserves.
+ * This takes precedence over the
+ * __GFP_MEMALLOC flag if both are
+ * set
+ */
#define __GFP_HARDWALL ((__force gfp_t)___GFP_HARDWALL) /* Enforce hardwall cpuset memory allocs */
#define __GFP_THISNODE ((__force gfp_t)___GFP_THISNODE)/* No fallback, no policies */
#define __GFP_RECLAIMABLE ((__force gfp_t)___GFP_RECLAIMABLE) /* Page is reclaimable */
@@ -127,7 +133,7 @@ struct vm_area_struct;
/* Control page allocator reclaim behavior */
#define GFP_RECLAIM_MASK (__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_FS|\
__GFP_NOWARN|__GFP_REPEAT|__GFP_NOFAIL|\
- __GFP_NORETRY|__GFP_NOMEMALLOC)
+ __GFP_NORETRY|__GFP_MEMALLOC|__GFP_NOMEMALLOC)
/* Control slab gfp mask during early boot */
#define GFP_BOOT_MASK (__GFP_BITS_MASK & ~(__GFP_WAIT|__GFP_IO|__GFP_FS))
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 2986426..319a7cd 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -72,7 +72,7 @@ struct page {
pgoff_t index; /* Our offset within mapping. */
void *freelist; /* SLUB: freelist req. slab lock */
bool pfmemalloc; /* If set by the page allocator,
- * ALLOC_PFMEMALLOC was set and the
+ * ALLOC_NO_WATERMARKS was set and the
* low watermark was not met implying
* that the system is under some
* pressure. The caller should try
diff --git a/include/trace/events/gfpflags.h b/include/trace/events/gfpflags.h
index 9fe3a366..d6fd8e5 100644
--- a/include/trace/events/gfpflags.h
+++ b/include/trace/events/gfpflags.h
@@ -30,6 +30,7 @@
{(unsigned long)__GFP_COMP, "GFP_COMP"}, \
{(unsigned long)__GFP_ZERO, "GFP_ZERO"}, \
{(unsigned long)__GFP_NOMEMALLOC, "GFP_NOMEMALLOC"}, \
+ {(unsigned long)__GFP_MEMALLOC, "GFP_MEMALLOC"}, \
{(unsigned long)__GFP_HARDWALL, "GFP_HARDWALL"}, \
{(unsigned long)__GFP_THISNODE, "GFP_THISNODE"}, \
{(unsigned long)__GFP_RECLAIMABLE, "GFP_RECLAIMABLE"}, \
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 7769a3d..27043e7 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1369,7 +1369,6 @@ failed:
#define ALLOC_HARDER 0x10 /* try to alloc harder */
#define ALLOC_HIGH 0x20 /* __GFP_HIGH set */
#define ALLOC_CPUSET 0x40 /* check for correct cpuset */
-#define ALLOC_PFMEMALLOC 0x80 /* Caller has PF_MEMALLOC set */
#ifdef CONFIG_FAIL_PAGE_ALLOC
@@ -2044,11 +2043,10 @@ gfp_to_alloc_flags(gfp_t gfp_mask)
} else if (unlikely(rt_task(current)) && !in_interrupt())
alloc_flags |= ALLOC_HARDER;
- if ((current->flags & PF_MEMALLOC) ||
- unlikely(test_thread_flag(TIF_MEMDIE))) {
- alloc_flags |= ALLOC_PFMEMALLOC;
-
- if (likely(!(gfp_mask & __GFP_NOMEMALLOC)) && !in_interrupt())
+ if (likely(!(gfp_mask & __GFP_NOMEMALLOC))) {
+ if (gfp_mask & __GFP_MEMALLOC)
+ alloc_flags |= ALLOC_NO_WATERMARKS;
+ else if (likely(!(gfp_mask & __GFP_NOMEMALLOC)) && !in_interrupt())
alloc_flags |= ALLOC_NO_WATERMARKS;
}
@@ -2057,7 +2055,7 @@ gfp_to_alloc_flags(gfp_t gfp_mask)
bool gfp_pfmemalloc_allowed(gfp_t gfp_mask)
{
- return !!(gfp_to_alloc_flags(gfp_mask) & ALLOC_PFMEMALLOC);
+ return !!(gfp_to_alloc_flags(gfp_mask) & ALLOC_NO_WATERMARKS);
}
static inline struct page *
@@ -2239,7 +2237,7 @@ got_pg:
* steps that will free more memory. The caller should avoid the
* page being used for !PFMEMALLOC purposes.
*/
- page->pfmemalloc = !!(alloc_flags & ALLOC_PFMEMALLOC);
+ page->pfmemalloc = !!(alloc_flags & ALLOC_NO_WATERMARKS);
return page;
}
diff --git a/mm/slab.c b/mm/slab.c
index a6480df..fceb656 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2979,7 +2979,7 @@ static int cache_grow(struct kmem_cache *cachep,
if (!slabp)
goto opps1;
- /* Record if ALLOC_PFMEMALLOC was set when allocating the slab */
+ /* Record if ALLOC_NO_WATERMARKS was set when allocating the slab */
if (pfmemalloc) {
struct array_cache *ac = cpu_cache_get(cachep);
slabp->pfmemalloc = true;
--
1.7.3.4
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related
* [PATCH 02/14] mm: sl[au]b: Add knowledge of PFMEMALLOC reserve pages
From: Mel Gorman @ 2011-06-20 13:12 UTC (permalink / raw)
To: Andrew Morton
Cc: Linux-MM, Linux-Netdev, LKML, David Miller, Neil Brown,
Peter Zijlstra, Mel Gorman
In-Reply-To: <1308575540-25219-1-git-send-email-mgorman@suse.de>
Allocations of pages below the min watermark run a risk of the machine
hanging due to lack of memory. To prevent this, only callers who
have PF_MEMALLOC or TIF_MEMDIE set and not processing an interrupt
are allowed to allocate with ALLOC_NO_WATERMARKS. Once they are
allocated to a slab though, nothing prevents other callers consuming
free objects within those slabs. This patch limits access to slab
pages that were alloced from the PFMEMALLOC reserves.
Pages allocated from the reserve are returned with page->pfmemalloc
set and it's up to the caller to determine how the page should be
protected. SLAB restricts access to any page with page->pfmemalloc set
to callers which are known to able to access the PFMEMALLOC reserve. If
one is not available, an attempt is made to allocate a new page rather
than use a reserve. SLUB is a bit more relaxed in that it only records
if the current per-CPU page was allocated from PFMEMALLOC reserve and
uses another partial slab if the caller does not have the necessary
GFP or process flags. This was found to be sufficient in tests to
avoid hangs due to SLUB generally maintaining smaller lists than SLAB.
In low-memory conditions it does mean that !PFMEMALLOC allocators
can fail a slab allocation even though free objects are available
because they are being preserved for callers that are freeing pages.
[a.p.zijlstra@chello.nl: Original implementation]
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
include/linux/mm_types.h | 8 ++
include/linux/slub_def.h | 1 +
mm/internal.h | 3 +
mm/page_alloc.c | 27 +++++-
mm/slab.c | 216 +++++++++++++++++++++++++++++++++++++++-------
mm/slub.c | 33 ++++++--
6 files changed, 244 insertions(+), 44 deletions(-)
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 027935c..2986426 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -71,6 +71,14 @@ struct page {
union {
pgoff_t index; /* Our offset within mapping. */
void *freelist; /* SLUB: freelist req. slab lock */
+ bool pfmemalloc; /* If set by the page allocator,
+ * ALLOC_PFMEMALLOC was set and the
+ * low watermark was not met implying
+ * that the system is under some
+ * pressure. The caller should try
+ * ensure this page is only used to
+ * free other pages.
+ */
};
struct list_head lru; /* Pageout list, eg. active_list
* protected by zone->lru_lock !
diff --git a/include/linux/slub_def.h b/include/linux/slub_def.h
index c8668d1..ff7eb16 100644
--- a/include/linux/slub_def.h
+++ b/include/linux/slub_def.h
@@ -40,6 +40,7 @@ struct kmem_cache_cpu {
unsigned long tid; /* Globally unique transaction id */
struct page *page; /* The slab from which we are allocating */
int node; /* The node of the page (or -1 for debug) */
+ bool pfmemalloc; /* Slab page had pfmemalloc set */
#ifdef CONFIG_SLUB_STATS
unsigned stat[NR_SLUB_STAT_ITEMS];
#endif
diff --git a/mm/internal.h b/mm/internal.h
index d071d380..a520f3b 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -193,6 +193,9 @@ static inline struct page *mem_map_next(struct page *iter,
#define __paginginit __init
#endif
+/* Returns true if the gfp_mask allows use of ALLOC_NO_WATERMARK */
+bool gfp_pfmemalloc_allowed(gfp_t gfp_mask);
+
/* Memory initialisation debug and verification */
enum mminit_level {
MMINIT_WARNING,
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index a327a72..7769a3d 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -656,6 +656,7 @@ static bool free_pages_prepare(struct page *page, unsigned int order)
trace_mm_page_free_direct(page, order);
kmemcheck_free_shadow(page, order);
+ page->pfmemalloc = false;
if (PageAnon(page))
page->mapping = NULL;
for (i = 0; i < (1 << order); i++)
@@ -1174,6 +1175,7 @@ void free_hot_cold_page(struct page *page, int cold)
migratetype = get_pageblock_migratetype(page);
set_page_private(page, migratetype);
+ page->pfmemalloc = false;
local_irq_save(flags);
if (unlikely(wasMlocked))
free_page_mlock(page);
@@ -1367,6 +1369,7 @@ failed:
#define ALLOC_HARDER 0x10 /* try to alloc harder */
#define ALLOC_HIGH 0x20 /* __GFP_HIGH set */
#define ALLOC_CPUSET 0x40 /* check for correct cpuset */
+#define ALLOC_PFMEMALLOC 0x80 /* Caller has PF_MEMALLOC set */
#ifdef CONFIG_FAIL_PAGE_ALLOC
@@ -2041,16 +2044,22 @@ gfp_to_alloc_flags(gfp_t gfp_mask)
} else if (unlikely(rt_task(current)) && !in_interrupt())
alloc_flags |= ALLOC_HARDER;
- if (likely(!(gfp_mask & __GFP_NOMEMALLOC))) {
- if (!in_interrupt() &&
- ((current->flags & PF_MEMALLOC) ||
- unlikely(test_thread_flag(TIF_MEMDIE))))
+ if ((current->flags & PF_MEMALLOC) ||
+ unlikely(test_thread_flag(TIF_MEMDIE))) {
+ alloc_flags |= ALLOC_PFMEMALLOC;
+
+ if (likely(!(gfp_mask & __GFP_NOMEMALLOC)) && !in_interrupt())
alloc_flags |= ALLOC_NO_WATERMARKS;
}
return alloc_flags;
}
+bool gfp_pfmemalloc_allowed(gfp_t gfp_mask)
+{
+ return !!(gfp_to_alloc_flags(gfp_mask) & ALLOC_PFMEMALLOC);
+}
+
static inline struct page *
__alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
struct zonelist *zonelist, enum zone_type high_zoneidx,
@@ -2223,8 +2232,16 @@ nopage:
got_pg:
if (kmemcheck_enabled)
kmemcheck_pagealloc_alloc(page, order, gfp_mask);
- return page;
+ /*
+ * page->pfmemalloc is set when the caller had PFMEMALLOC set or is
+ * been OOM killed. The expectation is that the caller is taking
+ * steps that will free more memory. The caller should avoid the
+ * page being used for !PFMEMALLOC purposes.
+ */
+ page->pfmemalloc = !!(alloc_flags & ALLOC_PFMEMALLOC);
+
+ return page;
}
/*
diff --git a/mm/slab.c b/mm/slab.c
index d96e223..a6480df 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -121,6 +121,8 @@
#include <asm/tlbflush.h>
#include <asm/page.h>
+#include "internal.h"
+
/*
* DEBUG - 1 for kmem_cache_create() to honour; SLAB_RED_ZONE & SLAB_POISON.
* 0 for faster, smaller code (especially in the critical paths).
@@ -227,6 +229,7 @@ struct slab {
unsigned int inuse; /* num of objs active in slab */
kmem_bufctl_t free;
unsigned short nodeid;
+ bool pfmemalloc; /* Slab had pfmemalloc set */
};
struct slab_rcu __slab_cover_slab_rcu;
};
@@ -248,15 +251,37 @@ struct array_cache {
unsigned int avail;
unsigned int limit;
unsigned int batchcount;
- unsigned int touched;
+ bool touched;
+ bool pfmemalloc;
spinlock_t lock;
void *entry[]; /*
* Must have this definition in here for the proper
* alignment of array_cache. Also simplifies accessing
* the entries.
+ *
+ * Entries should not be directly dereferenced as
+ * entries belonging to slabs marked pfmemalloc will
+ * have the lower bits set SLAB_OBJ_PFMEMALLOC
*/
};
+#define SLAB_OBJ_PFMEMALLOC 1
+static inline bool is_obj_pfmemalloc(void *objp)
+{
+ return (unsigned long)objp & SLAB_OBJ_PFMEMALLOC;
+}
+
+static inline void set_obj_pfmemalloc(void **objp)
+{
+ *objp = (void *)((unsigned long)*objp | SLAB_OBJ_PFMEMALLOC);
+ return;
+}
+
+static inline void clear_obj_pfmemalloc(void **objp)
+{
+ *objp = (void *)((unsigned long)*objp & ~SLAB_OBJ_PFMEMALLOC);
+}
+
/*
* bootstrap: The caches do not work without cpuarrays anymore, but the
* cpuarrays are allocated from the generic caches...
@@ -889,12 +914,100 @@ static struct array_cache *alloc_arraycache(int node, int entries,
nc->avail = 0;
nc->limit = entries;
nc->batchcount = batchcount;
- nc->touched = 0;
+ nc->touched = false;
spin_lock_init(&nc->lock);
}
return nc;
}
+/* Clears ac->pfmemalloc if no slabs have pfmalloc set */
+static void check_ac_pfmemalloc(struct kmem_cache *cachep,
+ struct array_cache *ac)
+{
+ struct kmem_list3 *l3 = cachep->nodelists[numa_mem_id()];
+ struct slab *slabp;
+
+ if (!ac->pfmemalloc)
+ return;
+
+ list_for_each_entry(slabp, &l3->slabs_full, list)
+ if (slabp->pfmemalloc)
+ return;
+
+ list_for_each_entry(slabp, &l3->slabs_partial, list)
+ if (slabp->pfmemalloc)
+ return;
+
+ list_for_each_entry(slabp, &l3->slabs_free, list)
+ if (slabp->pfmemalloc)
+ return;
+
+ ac->pfmemalloc = false;
+}
+
+static void *ac_get_obj(struct kmem_cache *cachep, struct array_cache *ac,
+ gfp_t flags, bool force_refill)
+{
+ int i;
+ void *objp = ac->entry[--ac->avail];
+
+ /* Ensure the caller is allowed to use objects from PFMEMALLOC slab */
+ if (unlikely(is_obj_pfmemalloc(objp))) {
+ struct kmem_list3 *l3;
+
+ if (gfp_pfmemalloc_allowed(flags)) {
+ clear_obj_pfmemalloc(&objp);
+ return objp;
+ }
+
+ /* The caller cannot use PFMEMALLOC objects, find another one */
+ for (i = 1; i < ac->avail; i++) {
+ /* If a !PFMEMALLOC object is found, swap them */
+ if (!is_obj_pfmemalloc(ac->entry[i])) {
+ objp = ac->entry[i];
+ ac->entry[i] = ac->entry[ac->avail];
+ ac->entry[ac->avail] = objp;
+ return objp;
+ }
+ }
+
+ /*
+ * If there are empty slabs on the slabs_free list and we are
+ * being forced to refill the cache, mark this one !pfmemalloc.
+ */
+ l3 = cachep->nodelists[numa_mem_id()];
+ if (!list_empty(&l3->slabs_free) && force_refill) {
+ struct slab *slabp = virt_to_slab(objp);
+ slabp->pfmemalloc = false;
+ clear_obj_pfmemalloc(&objp);
+ check_ac_pfmemalloc(cachep, ac);
+ return objp;
+ }
+
+ /* No !PFMEMALLOC objects available */
+ ac->avail++;
+ objp = NULL;
+ }
+
+ return objp;
+}
+
+static void ac_put_obj(struct kmem_cache *cachep, struct array_cache *ac,
+ void *objp)
+{
+ struct slab *slabp;
+
+ /* If there are pfmemalloc slabs, check if the object is part of one */
+ if (unlikely(ac->pfmemalloc)) {
+ slabp = virt_to_slab(objp);
+
+ if (slabp->pfmemalloc)
+ set_obj_pfmemalloc(&objp);
+ }
+
+ ac->entry[ac->avail++] = objp;
+}
+
/*
* Transfer objects in one arraycache to another.
* Locking must be handled by the caller.
@@ -1071,7 +1184,7 @@ static inline int cache_free_alien(struct kmem_cache *cachep, void *objp)
STATS_INC_ACOVERFLOW(cachep);
__drain_alien_cache(cachep, alien, nodeid);
}
- alien->entry[alien->avail++] = objp;
+ ac_put_obj(cachep, alien, objp);
spin_unlock(&alien->lock);
} else {
spin_lock(&(cachep->nodelists[nodeid])->list_lock);
@@ -1678,7 +1791,8 @@ __initcall(cpucache_init);
* did not request dmaable memory, we might get it, but that
* would be relatively rare and ignorable.
*/
-static void *kmem_getpages(struct kmem_cache *cachep, gfp_t flags, int nodeid)
+static void *kmem_getpages(struct kmem_cache *cachep, gfp_t flags, int nodeid,
+ bool *pfmemalloc)
{
struct page *page;
int nr_pages;
@@ -1699,6 +1813,7 @@ static void *kmem_getpages(struct kmem_cache *cachep, gfp_t flags, int nodeid)
page = alloc_pages_exact_node(nodeid, flags | __GFP_NOTRACK, cachep->gfporder);
if (!page)
return NULL;
+ *pfmemalloc = page->pfmemalloc;
nr_pages = (1 << cachep->gfporder);
if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
@@ -2131,7 +2246,7 @@ static int __init_refok setup_cpu_cache(struct kmem_cache *cachep, gfp_t gfp)
cpu_cache_get(cachep)->avail = 0;
cpu_cache_get(cachep)->limit = BOOT_CPUCACHE_ENTRIES;
cpu_cache_get(cachep)->batchcount = 1;
- cpu_cache_get(cachep)->touched = 0;
+ cpu_cache_get(cachep)->touched = false;
cachep->batchcount = 1;
cachep->limit = BOOT_CPUCACHE_ENTRIES;
return 0;
@@ -2678,6 +2793,7 @@ static struct slab *alloc_slabmgmt(struct kmem_cache *cachep, void *objp,
slabp->s_mem = objp + colour_off;
slabp->nodeid = nodeid;
slabp->free = 0;
+ slabp->pfmemalloc = false;
return slabp;
}
@@ -2809,7 +2925,7 @@ static void slab_map_pages(struct kmem_cache *cache, struct slab *slab,
* kmem_cache_alloc() when there are no active objs left in a cache.
*/
static int cache_grow(struct kmem_cache *cachep,
- gfp_t flags, int nodeid, void *objp)
+ gfp_t flags, int nodeid, void *objp, bool pfmemalloc)
{
struct slab *slabp;
size_t offset;
@@ -2853,7 +2969,7 @@ static int cache_grow(struct kmem_cache *cachep,
* 'nodeid'.
*/
if (!objp)
- objp = kmem_getpages(cachep, local_flags, nodeid);
+ objp = kmem_getpages(cachep, local_flags, nodeid, &pfmemalloc);
if (!objp)
goto failed;
@@ -2863,6 +2979,13 @@ static int cache_grow(struct kmem_cache *cachep,
if (!slabp)
goto opps1;
+ /* Record if ALLOC_PFMEMALLOC was set when allocating the slab */
+ if (pfmemalloc) {
+ struct array_cache *ac = cpu_cache_get(cachep);
+ slabp->pfmemalloc = true;
+ ac->pfmemalloc = true;
+ }
+
slab_map_pages(cachep, slabp, objp);
cache_init_objs(cachep, slabp);
@@ -3004,16 +3127,19 @@ bad:
#define check_slabp(x,y) do { } while(0)
#endif
-static void *cache_alloc_refill(struct kmem_cache *cachep, gfp_t flags)
+static void *cache_alloc_refill(struct kmem_cache *cachep, gfp_t flags,
+ bool force_refill)
{
int batchcount;
struct kmem_list3 *l3;
struct array_cache *ac;
int node;
-retry:
check_irq_off();
node = numa_mem_id();
+ if (unlikely(force_refill))
+ goto force_grow;
+retry:
ac = cpu_cache_get(cachep);
batchcount = ac->batchcount;
if (!ac->touched && batchcount > BATCHREFILL_LIMIT) {
@@ -3031,7 +3157,7 @@ retry:
/* See if we can refill from the shared array */
if (l3->shared && transfer_objects(ac, l3->shared, batchcount)) {
- l3->shared->touched = 1;
+ l3->shared->touched = true;
goto alloc_done;
}
@@ -3063,8 +3189,8 @@ retry:
STATS_INC_ACTIVE(cachep);
STATS_SET_HIGH(cachep);
- ac->entry[ac->avail++] = slab_get_obj(cachep, slabp,
- node);
+ ac_put_obj(cachep, ac, slab_get_obj(cachep, slabp,
+ node));
}
check_slabp(cachep, slabp);
@@ -3083,18 +3209,25 @@ alloc_done:
if (unlikely(!ac->avail)) {
int x;
- x = cache_grow(cachep, flags | GFP_THISNODE, node, NULL);
+force_grow:
+ x = cache_grow(cachep, flags | GFP_THISNODE, node, NULL, false);
/* cache_grow can reenable interrupts, then ac could change. */
ac = cpu_cache_get(cachep);
- if (!x && ac->avail == 0) /* no objects in sight? abort */
+
+ /* no objects in sight? abort */
+ if (!x && (ac->avail == 0 || force_refill))
return NULL;
- if (!ac->avail) /* objects refilled by interrupt? */
+ /* objects refilled by interrupt? */
+ if (!ac->avail) {
+ node = numa_node_id();
goto retry;
+ }
}
- ac->touched = 1;
- return ac->entry[--ac->avail];
+ ac->touched = true;
+
+ return ac_get_obj(cachep, ac, flags, force_refill);
}
static inline void cache_alloc_debugcheck_before(struct kmem_cache *cachep,
@@ -3177,23 +3310,35 @@ static inline void *____cache_alloc(struct kmem_cache *cachep, gfp_t flags)
{
void *objp;
struct array_cache *ac;
+ bool force_refill = false;
check_irq_off();
ac = cpu_cache_get(cachep);
if (likely(ac->avail)) {
- STATS_INC_ALLOCHIT(cachep);
- ac->touched = 1;
- objp = ac->entry[--ac->avail];
- } else {
- STATS_INC_ALLOCMISS(cachep);
- objp = cache_alloc_refill(cachep, flags);
+ ac->touched = true;
+ objp = ac_get_obj(cachep, ac, flags, false);
+
/*
- * the 'ac' may be updated by cache_alloc_refill(),
- * and kmemleak_erase() requires its correct value.
+ * Allow for the possibility all avail objects are not allowed
+ * by the current flags
*/
- ac = cpu_cache_get(cachep);
+ if (objp) {
+ STATS_INC_ALLOCHIT(cachep);
+ goto out;
+ }
+ force_refill = true;
}
+
+ STATS_INC_ALLOCMISS(cachep);
+ objp = cache_alloc_refill(cachep, flags, force_refill);
+ /*
+ * the 'ac' may be updated by cache_alloc_refill(),
+ * and kmemleak_erase() requires its correct value.
+ */
+ ac = cpu_cache_get(cachep);
+
+out:
/*
* To avoid a false negative, if an object that is in one of the
* per-CPU caches is leaked, we need to make sure kmemleak doesn't
@@ -3246,6 +3391,7 @@ static void *fallback_alloc(struct kmem_cache *cache, gfp_t flags)
enum zone_type high_zoneidx = gfp_zone(flags);
void *obj = NULL;
int nid;
+ bool pfmemalloc;
if (flags & __GFP_THISNODE)
return NULL;
@@ -3282,7 +3428,8 @@ retry:
if (local_flags & __GFP_WAIT)
local_irq_enable();
kmem_flagcheck(cache, flags);
- obj = kmem_getpages(cache, local_flags, numa_mem_id());
+ obj = kmem_getpages(cache, local_flags, numa_mem_id(),
+ &pfmemalloc);
if (local_flags & __GFP_WAIT)
local_irq_disable();
if (obj) {
@@ -3290,7 +3437,7 @@ retry:
* Insert into the appropriate per node queues
*/
nid = page_to_nid(virt_to_page(obj));
- if (cache_grow(cache, flags, nid, obj)) {
+ if (cache_grow(cache, flags, nid, obj, pfmemalloc)) {
obj = ____cache_alloc_node(cache,
flags | GFP_THISNODE, nid);
if (!obj)
@@ -3362,7 +3509,7 @@ retry:
must_grow:
spin_unlock(&l3->list_lock);
- x = cache_grow(cachep, flags | GFP_THISNODE, nodeid, NULL);
+ x = cache_grow(cachep, flags | GFP_THISNODE, nodeid, NULL, false);
if (x)
goto retry;
@@ -3512,9 +3659,12 @@ static void free_block(struct kmem_cache *cachep, void **objpp, int nr_objects,
struct kmem_list3 *l3;
for (i = 0; i < nr_objects; i++) {
- void *objp = objpp[i];
+ void *objp;
struct slab *slabp;
+ clear_obj_pfmemalloc(&objpp[i]);
+ objp = objpp[i];
+
slabp = virt_to_slab(objp);
l3 = cachep->nodelists[node];
list_del(&slabp->list);
@@ -3627,12 +3777,12 @@ static inline void __cache_free(struct kmem_cache *cachep, void *objp,
if (likely(ac->avail < ac->limit)) {
STATS_INC_FREEHIT(cachep);
- ac->entry[ac->avail++] = objp;
+ ac_put_obj(cachep, ac, objp);
return;
} else {
STATS_INC_FREEMISS(cachep);
cache_flusharray(cachep, ac);
- ac->entry[ac->avail++] = objp;
+ ac_put_obj(cachep, ac, objp);
}
}
@@ -4058,7 +4208,7 @@ static void drain_array(struct kmem_cache *cachep, struct kmem_list3 *l3,
if (!ac || !ac->avail)
return;
if (ac->touched && !force) {
- ac->touched = 0;
+ ac->touched = false;
} else {
spin_lock_irq(&l3->list_lock);
if (ac->avail) {
diff --git a/mm/slub.c b/mm/slub.c
index 35f351f..dd552a5 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -30,6 +30,8 @@
#include <trace/events/kmem.h>
+#include "internal.h"
+
/*
* Lock order:
* 1. slab_lock(page)
@@ -1242,7 +1244,8 @@ static void setup_object(struct kmem_cache *s, struct page *page,
s->ctor(object);
}
-static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node)
+static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node,
+ bool *pfmemalloc)
{
struct page *page;
void *start;
@@ -1257,6 +1260,7 @@ static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node)
goto out;
inc_slabs_node(s, page_to_nid(page), page->objects);
+ *pfmemalloc = page->pfmemalloc;
page->slab = s;
page->flags |= 1 << PG_slab;
@@ -1772,6 +1776,14 @@ slab_out_of_memory(struct kmem_cache *s, gfp_t gfpflags, int nid)
}
}
+static inline bool pfmemalloc_match(struct kmem_cache_cpu *c, gfp_t gfpflags)
+{
+ if (unlikely(c->pfmemalloc))
+ return gfp_pfmemalloc_allowed(gfpflags);
+
+ return true;
+}
+
/*
* Slow path. The lockless freelist is empty or we need to perform
* debugging duties.
@@ -1796,6 +1808,7 @@ static void *__slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
void **object;
struct page *page;
unsigned long flags;
+ bool pfmemalloc = false;
local_irq_save(flags);
#ifdef CONFIG_PREEMPT
@@ -1815,7 +1828,13 @@ static void *__slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
goto new_slab;
slab_lock(page);
- if (unlikely(!node_match(c, node)))
+
+ /*
+ * By rights, we should be searching for a slab page that was
+ * PFMEMALLOC but right now, we are losing the pfmemalloc
+ * information when the page leaves the per-cpu allocator
+ */
+ if (unlikely(!pfmemalloc_match(c, gfpflags) || !node_match(c, node)))
goto another_slab;
stat(s, ALLOC_REFILL);
@@ -1853,7 +1872,7 @@ new_slab:
if (gfpflags & __GFP_WAIT)
local_irq_enable();
- page = new_slab(s, gfpflags, node);
+ page = new_slab(s, gfpflags, node, &pfmemalloc);
if (gfpflags & __GFP_WAIT)
local_irq_disable();
@@ -1868,6 +1887,7 @@ new_slab:
__SetPageSlubFrozen(page);
c->node = page_to_nid(page);
c->page = page;
+ c->pfmemalloc = pfmemalloc;
goto load_freelist;
}
if (!(gfpflags & __GFP_NOWARN) && printk_ratelimit())
@@ -1927,8 +1947,8 @@ redo:
barrier();
object = c->freelist;
- if (unlikely(!object || !node_match(c, node)))
-
+ if (unlikely(!object || !node_match(c, node) ||
+ !pfmemalloc_match(c, gfpflags)))
object = __slab_alloc(s, gfpflags, node, addr, c);
else {
@@ -2351,10 +2371,11 @@ static void early_kmem_cache_node_alloc(int node)
struct page *page;
struct kmem_cache_node *n;
unsigned long flags;
+ bool pfmemalloc; /* Ignore this early in boot */
BUG_ON(kmem_cache_node->size < sizeof(struct kmem_cache_node));
- page = new_slab(kmem_cache_node, GFP_NOWAIT, node);
+ page = new_slab(kmem_cache_node, GFP_NOWAIT, node, &pfmemalloc);
BUG_ON(!page);
if (page_to_nid(page) != node) {
--
1.7.3.4
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related
* [PATCH 01/14] mm: Serialize access to min_free_kbytes
From: Mel Gorman @ 2011-06-20 13:12 UTC (permalink / raw)
To: Andrew Morton
Cc: Linux-MM, Linux-Netdev, LKML, David Miller, Neil Brown,
Peter Zijlstra, Mel Gorman
In-Reply-To: <1308575540-25219-1-git-send-email-mgorman@suse.de>
There is a race between the min_free_kbytes sysctl, memory hotplug
and transparent hugepage support enablement. Memory hotplug uses a
zonelists_mutex to avoid a race when building zonelists. Reuse it to
serialise watermark updates.
[a.p.zijlstra@chello.nl: Older patch fixed the race with spinlock]
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
mm/page_alloc.c | 23 +++++++++++++++--------
1 files changed, 15 insertions(+), 8 deletions(-)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 4e8985a..a327a72 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -5045,14 +5045,7 @@ static void setup_per_zone_lowmem_reserve(void)
calculate_totalreserve_pages();
}
-/**
- * setup_per_zone_wmarks - called when min_free_kbytes changes
- * or when memory is hot-{added|removed}
- *
- * Ensures that the watermark[min,low,high] values for each zone are set
- * correctly with respect to min_free_kbytes.
- */
-void setup_per_zone_wmarks(void)
+static void __setup_per_zone_wmarks(void)
{
unsigned long pages_min = min_free_kbytes >> (PAGE_SHIFT - 10);
unsigned long lowmem_pages = 0;
@@ -5107,6 +5100,20 @@ void setup_per_zone_wmarks(void)
calculate_totalreserve_pages();
}
+/**
+ * setup_per_zone_wmarks - called when min_free_kbytes changes
+ * or when memory is hot-{added|removed}
+ *
+ * Ensures that the watermark[min,low,high] values for each zone are set
+ * correctly with respect to min_free_kbytes.
+ */
+void setup_per_zone_wmarks(void)
+{
+ mutex_lock(&zonelists_mutex);
+ __setup_per_zone_wmarks();
+ mutex_unlock(&zonelists_mutex);
+}
+
/*
* The inactive anon list should be small enough that the VM never has to
* do too much work, but large enough that each inactive page has a chance
--
1.7.3.4
^ permalink raw reply related
* [PATCH 00/14] Swap-over-NBD without deadlocking v5
From: Mel Gorman @ 2011-06-20 13:12 UTC (permalink / raw)
To: Andrew Morton
Cc: Linux-MM, Linux-Netdev, LKML, David Miller, Neil Brown,
Peter Zijlstra, Mel Gorman
Changelog since V4
o Update comment clarifying what protocols can be used (Michal)
o Rebase to 3.0-rc3
Changelog since V3
o Propogate pfmemalloc from packet fragment pages to skb (Neil)
o Rebase to 3.0-rc2
Changelog since V2
o Document that __GFP_NOMEMALLOC overrides __GFP_MEMALLOC (Neil)
o Use wait_event_interruptible (Neil)
o Use !! when casting to bool to avoid any possibilitity of type
truncation (Neil)
o Nicer logic when using skb_pfmemalloc_protocol (Neil)
Changelog since V1
o Rebase on top of mmotm
o Use atomic_t for memalloc_socks (David Miller)
o Remove use of sk_memalloc_socks in vmscan (Neil Brown)
o Check throttle within prepare_to_wait (Neil Brown)
o Add statistics on throttling instead of printk
Swapping over NBD is something that is technically possible but not
often advised. While there are number of guides on the internet
on how to configure it and nbd-client supports a -swap switch to
"prevent deadlocks", the fact of the matter is a machine using NBD
for swap can be locked up within minutes if swap is used intensively.
The problem is that network block devices do not use mempools like
normal block devices do. As the host cannot control where they receive
packets from, they cannot reliably work out in advance how much memory
they might need.
Some years ago, Peter Ziljstra developed a series of patches that
supported swap over an NFS that some distributions are carrying in
their kernels. This patch series borrows very heavily from Peter's work
to support swapping over NBD (the relatively straight-forward case)
and uses throttling instead of dynamically resized memory reserves
so the series is not too unwieldy for review.
Patch 1 serialises access to min_free_kbytes. It's not strictly needed
by this series but as the series cares about watermarks in
general, it's a harmless fix. It could be merged independently.
Patch 2 adds knowledge of the PFMEMALLOC reserves to SLAB and SLUB to
preserve access to pages allocated under low memory situations
to callers that are freeying memory.
Patch 3 introduces __GFP_MEMALLOC to allow access to the PFMEMALLOC
reserves without setting PFMEMALLOC.
Patch 4 opens the possibility for softirqs to use PFMEMALLOC reserves
for later use by network packet processing.
Patch 5 ignores memory policies when ALLOC_NO_WATERMARKS is set.
Patches 6-10 allows network processing to use PFMEMALLOC reserves when
the socket has been marked as being used by the VM to clean
pages. If packets are received and stored in pages that were
allocated under low-memory situations and are unrelated to
the VM, the packets are dropped.
Patch 11 is a micro-optimisation to avoid a function call in the
common case.
Patch 12 tags NBD sockets as being SOCK_MEMALLOC so they can use
PFMEMALLOC if necessary.
Patch 13 notes that it is still possible for the PFMEMALLOC reserve
to be depleted. To prevent this, direct reclaimers get
throttled on a waitqueue if 50% of the PFMEMALLOC reserves are
depleted. It is expected that kswapd and the direct reclaimers
already running will clean enough pages for the low watermark
to be reached and the throttled processes are woken up.
Patch 14 adds a statistic to track how often processes get throttled
Some basic performance testing was run using kernel builds, netperf
on loopback for UDP and TCP, hackbench (pipes and sockets), iozone
and sysbench. Each of them were expected to use the sl*b allocators
reasonably heavily but there did not appear to be significant
performance variances. Here is the results from netperf using
slab as an example
NETPERF UDP
64 237.47 ( 0.00%) 237.34 (-0.05%)
128 472.69 ( 0.00%) 465.96 (-1.44%)
256 926.82 ( 0.00%) 948.40 ( 2.28%)
1024 3260.08 ( 0.00%) 3266.50 ( 0.20%)
2048 5535.11 ( 0.00%) 5453.55 (-1.50%)
3312 7496.60 ( 0.00%)* 7574.44 ( 1.03%)
1.12% 1.00%
4096 8266.35 ( 0.00%)* 8240.06 (-0.32%)*
1.18% 1.49%
8192 11026.01 ( 0.00%) 11010.44 (-0.14%)
16384 14653.98 ( 0.00%) 14666.97 ( 0.09%)
MMTests Statistics: duration
User/Sys Time Running Test (seconds) 2156.64 1873.27
Total Elapsed Time (seconds) 2570.09 2234.10
NETPERF TCP
netperf-tcp tcp-swapnbd
vanilla-slab v4r3-slab
64 1250.76 ( 0.00%) 1256.52 ( 0.46%)
128 2290.70 ( 0.00%) 2336.43 ( 1.96%)
256 3668.42 ( 0.00%) 3751.17 ( 2.21%)
1024 7214.33 ( 0.00%) 7237.23 ( 0.32%)
2048 8230.01 ( 0.00%) 8280.02 ( 0.60%)
3312 8634.95 ( 0.00%) 8758.62 ( 1.41%)
4096 8851.18 ( 0.00%) 9045.88 ( 2.15%)
8192 10067.59 ( 0.00%) 10263.30 ( 1.91%)
16384 11523.26 ( 0.00%) 11654.78 ( 1.13%)
MMTests Statistics: duration
User/Sys Time Running Test (seconds) 1450.23 1389.8
Total Elapsed Time (seconds) 1450.41 1390.35
Here is the equivalent test for SLUB
netperf-udp udp-swapnbd
vanilla-slub v4r3-slub
64 235.33 ( 0.00%) 237.80 ( 1.04%)
128 465.92 ( 0.00%) 469.98 ( 0.86%)
256 907.16 ( 0.00%) 907.58 ( 0.05%)
1024 3240.25 ( 0.00%) 3255.56 ( 0.47%)
2048 5564.87 ( 0.00%) 5446.46 (-2.17%)
3312 7427.65 ( 0.00%)* 7650.00 ( 2.91%)
1.33% 1.00%
4096 8004.51 ( 0.00%)* 8132.79 ( 1.58%)*
1.05% 1.21%
8192 11079.60 ( 0.00%) 10927.09 (-1.40%)
16384 14737.38 ( 0.00%) 15019.50 ( 1.88%)
MMTests Statistics: duration
User/Sys Time Running Test (seconds) 2056.21 2160.38
Total Elapsed Time (seconds) 2426.09 2498.16
NETPERF TCP
netperf-tcp tcp-swapnbd
vanilla-slub v4r3-slub
64 1251.64 ( 0.00%) 1262.89 ( 0.89%)
128 2289.88 ( 0.00%) 2332.94 ( 1.85%)
256 3654.34 ( 0.00%) 3736.48 ( 2.20%)
1024 7192.47 ( 0.00%) 7286.96 ( 1.30%)
2048 8243.55 ( 0.00%) 8291.50 ( 0.58%)
3312 8664.16 ( 0.00%) 8799.88 ( 1.54%)
4096 8869.13 ( 0.00%) 9018.12 ( 1.65%)
8192 10009.53 ( 0.00%) 10214.26 ( 2.00%)
16384 11470.78 ( 0.00%) 11685.20 ( 1.83%)
MMTests Statistics: duration
User/Sys Time Running Test (seconds) 1368.28 1511.81
Total Elapsed Time (seconds) 1370.33 1510.42
Time to completion varied a lot but this can happen with netperf as
it tries to find results within a sufficiently high confidence. There
were some small gains and losses but they are close to the variances
seen between kernel releases.
For testing swap-over-NBD, a machine was booted with 2G of RAM with a
swapfile backed by NBD. 8*NUM_CPU processes were started that create
anonymous memory mappings and read them linearly in a loop. The total
size of the mappings were 4*PHYSICAL_MEMORY to use swap heavily under
memory pressure. Without the patches, the machine locks up within
minutes and runs to completion with them applied.
drivers/block/nbd.c | 7 +-
include/linux/gfp.h | 13 ++-
include/linux/mm_types.h | 8 ++
include/linux/mmzone.h | 1 +
include/linux/sched.h | 7 +
include/linux/skbuff.h | 21 +++-
include/linux/slub_def.h | 1 +
include/linux/vm_event_item.h | 1 +
include/net/sock.h | 19 +++
include/trace/events/gfpflags.h | 1 +
kernel/softirq.c | 3 +
mm/page_alloc.c | 57 +++++++--
mm/slab.c | 240 +++++++++++++++++++++++++++++++++------
mm/slub.c | 33 +++++-
mm/vmscan.c | 55 +++++++++
mm/vmstat.c | 1 +
net/core/dev.c | 48 +++++++-
net/core/filter.c | 8 ++
net/core/skbuff.c | 95 +++++++++++++---
net/core/sock.c | 42 +++++++
net/ipv4/tcp.c | 3 +-
net/ipv4/tcp_output.c | 13 +-
net/ipv6/tcp_ipv6.c | 12 ++-
23 files changed, 602 insertions(+), 87 deletions(-)
--
1.7.3.4
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply
* Re: Linux TCP's Robustness to Multipath Packet Reordering
From: Dominik Kaspar @ 2011-06-20 12:52 UTC (permalink / raw)
To: Ilpo Järvinen
Cc: Alexander Zimmermann, Netdev, Yuchung Cheng, Carsten Wolff,
John Heffner, Eric Dumazet, Lennart Schulte, Arnd Hannemann
In-Reply-To: <alpine.DEB.2.00.1106201336450.17529@wel-95.cs.helsinki.fi>
Hi Ilpo,
> Where did you get this idea of reneging?!?
I observed that my scenario of a retransmitted packet overtaking the
original somehow causes TCP to enter the "Loss" state although no RTO
was caused. And since the Loss state seems to be only entered due to
RTO timeout or SACK reneging, I got the idea that reneging must be
occurring.
> Reneging has nothing to do with DSACKs,
> instead it is only detected if the cumulative ACK stops to such
> boundary where the _next_ segment is SACKed (i.e., some reason
> the receiver "didn't bother" to cumulatively ACK for that too). ...
> That certainly does not happen (ever) for out of window DSACKs.
You are right. If I turn off DSACK, the same thing happens: TCP enters
the Loss state without timeouts occurring. Isn't that a sign of
reneging happening? What else can it be?
Dominik
^ permalink raw reply
* Re: [PATCH 10/14] SIWv2: Transmit path: siw_qp_tx.c
From: Bernard Metzler @ 2011-06-20 12:33 UTC (permalink / raw)
To: Bart Van Assche
Cc: bart.vanassche-Re5JQEeQqe8AvxtiuMwx3w,
linux-rdma-u79uwXL29TY76Z2rM5mHXA, netdev-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <BANLkTimSqwBK_pEGsydCAEvO=S6WVVzDPQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
bart.vanassche-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org wrote on 06/18/2011 07:47:29 PM:
> On Thu, Jun 16, 2011 at 2:42 PM, Bernard Metzler <bmt-OA+xvbQnYDHMbYB6QlFGEg@public.gmane.org>
wrote:
> > ---
> > drivers/infiniband/hw/siw/siw_qp_tx.c | 1332
> +++++++++++++++++++++++++++++++++
> > 1 files changed, 1332 insertions(+), 0 deletions(-)
> > create mode 100644 drivers/infiniband/hw/siw/siw_qp_tx.c
> >
> > diff --git a/drivers/infiniband/hw/siw/siw_qp_tx.c
> b/drivers/infiniband/hw/siw/siw_qp_tx.c
> [ ... ]
> > +/*
> > + * Write out iov referencing hdr, data and trailer of current FPDU.
> > + * Update transmit state dependent on write return status
> > + */
> > +static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
> > +{
> > + struct siw_wqe *wqe = c_tx->wqe;
> > + struct siw_sge *sge = &wqe->wr.sgl.sge[c_tx->sge_idx],
> > + *first_sge = sge;
> > + struct siw_mr *mr = NULL;
> > + struct ib_umem_chunk *chunk = c_tx->umem_chunk;
> > +
> > + struct kvec iov[MAX_ARRAY];
> > + struct page *page_array[MAX_ARRAY];
> > + struct msghdr msg = {.msg_flags = MSG_DONTWAIT};
> > +
> > + int seg = 0, do_crc = c_tx->do_crc,
> is_kva = 0, rv;
> > + unsigned int data_len = c_tx->bytes_unsent,
> > + hdr_len = 0,
> > + trl_len = 0,
> > + sge_off = c_tx->sge_off,
> > + sge_idx = c_tx->sge_idx,
> > + pg_idx = c_tx->pg_idx;
>
> Have you run the siw source code through sparse ? Sparse reports the
> following for the above:
>
yes, I did.
> drivers/infiniband/hw/siw/siw_qp_tx.c:694:1: warning: the frame size
> of 1120 bytes is larger than 1024 bytes
>
> I assume that you know that one should be careful with stack
> allocations in the kernel ?
>
yes. I will have to make the MAX_ARRAY parameter smaller. It
defines the maximum amount of elements in a vector to be pushed to
tcp_sendmsg() in one call.
In an older version of the code we pushed iwarp header, data pages
and iwarp trailer in several consecutive calls. While the code looked
cleaner, it induced a measureable performance penalty. A smaller
MAX_ARRAY will also restrict the maximum allowed scatter-gather-elements
in a work request to a smaller safe value (currently 10 SGEs),
since the code intentionally tries to pack as many SGEs
into one iwarp frame as possible.
Alternatively, the vector could be part of the queue pair
structure, but that would significantly increase per connection
memory footprint, which we wanted to avoid.
> > + if (tx_type == RDMAP_RDMA_READ_REQ) {
>
> This is what the compiler says about the above statement:
>
> drivers/infiniband/hw/siw/siw_qp_tx.c:1145:15: warning: comparison
> between enum siw_wr_opcode and enum rdma_opcode
>
> So how can that statement be correct ?
>
it will not - must be changed to 'SIW_WR_RDMA_READ_REQ'...many thanks!
I obviously used sparse in a wrong way:
building within kernel tree with 'make C=1 CF="-Wsparse-all' does not give
me the indicated warning (neither does setting CF="-Wenum-mismatch").
Can you please advice? Thank you and sorry for the newbie question.
Bernard.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" 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
* Re: [net-next 04/17] ixgbevf: provide 64 bit statistics
From: Jeff Kirsher @ 2011-06-20 12:29 UTC (permalink / raw)
To: Eric Dumazet
Cc: davem@davemloft.net, Stephen Hemminger, netdev@vger.kernel.org,
gospo@redhat.com
In-Reply-To: <1308545992.3539.114.camel@edumazet-laptop>
[-- Attachment #1: Type: text/plain, Size: 1831 bytes --]
On Sun, 2011-06-19 at 21:59 -0700, Eric Dumazet wrote:
> Le dimanche 19 juin 2011 à 17:58 -0700, Jeff Kirsher a écrit :
> > From: Stephen Hemminger <shemminger@vyatta.com>
> >
> > Compute statistics per ring using 64 bits, and provide
> > network device stats in 64 bits.
> >
> > Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
> > Acked-by: Greg Rose <Gregory.v.rose@intel.com>
> > Tested-by: Evan Swanson <evan.swanson@intel.com>
> > Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
> > ---
>
> >
> > +static struct rtnl_link_stats64 *ixgbevf_get_stats(struct net_device *netdev,
> > + struct rtnl_link_stats64 *stats)
> > +{
> > + struct ixgbevf_adapter *adapter = netdev_priv(netdev);
> > + int i;
> > +
> > + ixgbevf_update_stats(adapter);
> > +
> > + stats->multicast = adapter->stats.vfmprc - adapter->stats.base_vfmprc;
> > +
> > + for (i = 0; i < adapter->num_rx_queues; i++) {
> > + stats->rx_bytes += adapter->rx_ring[i].total_bytes;
> > + stats->rx_packets += adapter->rx_ring[i].total_packets;
> > + }
> > +
> > + for (i = 0; i < adapter->num_tx_queues; i++) {
> > + stats->tx_bytes += adapter->tx_ring[i].total_bytes;
> > + stats->tx_packets += adapter->tx_ring[i].total_packets;
> > + }
> > +
> > + return stats;
> > +}
> > +
>
> Unfortunately this patch is racy on 32bit arches.
>
> Jeff, maybe you missed all the discussion we had lately on these
> conversions, needing appropriate synchronization ?
I apologize I did not see the discussion regarding Stephen's patches.
>
> It seems a lot of bugs were added lately with Stephen patches, this is
> rather unfortunate...
>
> I guess I'll have to review them and send fixes.
I will drop Stephen's patch from this series and look forward to your
fixes. Thanks Eric.
[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 490 bytes --]
^ permalink raw reply
* [PATCH] netlink: advertise incomplete dumps
From: Johannes Berg @ 2011-06-20 11:40 UTC (permalink / raw)
To: netdev, linux-wireless
Cc: Samuel Ortiz, aloisio.almeida, John Linville, Thomas Graf
From: Johannes Berg <johannes.berg@intel.com>
Consider the following situation:
* a dump that would show 8 entries, four in the first
round, and four in the second
* between the first and second rounds, 6 entries are
removed
* now the second round will not show any entry, and
even if there is a sequence/generation counter the
application will not know
To solve this problem, add a new flag NLM_F_DUMP_INTR
to the netlink header that indicates the dump wasn't
consistent, this flag can also be set on the MSG_DONE
message that terminates the dump, and as such above
situation can be detected.
To achieve this, add a sequence counter to the netlink
callback struct. Of course, netlink code still needs
to use this new functionality. The correct way to do
that is to always set cb->seq when a dumpit callback
is invoked and call nl_dump_check_consistent() for
each new message. The core code will also call this
function for the final MSG_DONE message.
To make it usable with generic netlink, a new function
genlmsg_nlhdr() is needed to obtain the netlink header
from the genetlink user header.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
Should we merge this through wireless? From wireless-next it'll go into
net-next quickly, but the converse isn't true, so using it in wireless
would be harder if it was merged through net-next I think? Or John can
cherry-pick into wireless-testing?
include/linux/netlink.h | 2 ++
include/net/genetlink.h | 32 ++++++++++++++++++++++++++++++++
include/net/netlink.h | 24 ++++++++++++++++++++++++
net/netlink/af_netlink.c | 2 ++
4 files changed, 60 insertions(+)
--- a/include/linux/netlink.h 2011-06-20 08:42:41.000000000 +0200
+++ b/include/linux/netlink.h 2011-06-20 09:39:19.000000000 +0200
@@ -49,6 +49,7 @@ struct nlmsghdr {
#define NLM_F_MULTI 2 /* Multipart message, terminated by NLMSG_DONE */
#define NLM_F_ACK 4 /* Reply with ack, with zero or error code */
#define NLM_F_ECHO 8 /* Echo this request */
+#define NLM_F_DUMP_INTR 16 /* Dump was inconsistent due to sequence change */
/* Modifiers to GET request */
#define NLM_F_ROOT 0x100 /* specify tree root */
@@ -222,6 +223,7 @@ struct netlink_callback {
struct netlink_callback *cb);
int (*done)(struct netlink_callback *cb);
int family;
+ unsigned int prev_seq, seq;
long args[6];
};
--- a/net/netlink/af_netlink.c 2011-06-20 08:42:41.000000000 +0200
+++ b/net/netlink/af_netlink.c 2011-06-20 09:39:19.000000000 +0200
@@ -1693,6 +1693,8 @@ static int netlink_dump(struct sock *sk)
if (!nlh)
goto errout_skb;
+ nl_dump_check_consistent(cb, nlh);
+
memcpy(nlmsg_data(nlh), &len, sizeof(len));
if (sk_filter(sk, skb))
--- a/include/net/genetlink.h 2011-06-20 08:42:41.000000000 +0200
+++ b/include/net/genetlink.h 2011-06-20 13:36:54.000000000 +0200
@@ -160,6 +160,38 @@ static inline void *genlmsg_put(struct s
}
/**
+ * genlmsg_nlhdr - Obtain netlink header from user specified header
+ * @user_hdr: user header as returned from genlmsg_put()
+ * @family: generic netlink family
+ *
+ * Returns pointer to netlink header.
+ */
+static inline struct nlmsghdr *genlmsg_nlhdr(void *user_hdr,
+ struct genl_family *family)
+{
+ return (struct nlmsghdr *)((char *)user_hdr -
+ family->hdrsize -
+ GENL_HDRLEN -
+ NLMSG_HDRLEN);
+}
+
+/**
+ * genl_dump_check_consistent - check if sequence is consistent and advertise if not
+ * @cb: netlink callback structure that stores the sequence number
+ * @user_hdr: user header as returned from genlmsg_put()
+ * @family: generic netlink family
+ *
+ * Cf. nl_dump_check_consistent(), this just provides a wrapper to make it
+ * simpler to use with generic netlink.
+ */
+static inline void genl_dump_check_consistent(struct netlink_callback *cb,
+ void *user_hdr,
+ struct genl_family *family)
+{
+ nl_dump_check_consistent(cb, genlmsg_nlhdr(user_hdr, family));
+}
+
+/**
* genlmsg_put_reply - Add generic netlink header to a reply message
* @skb: socket buffer holding the message
* @info: receiver info
--- a/include/net/netlink.h 2011-06-20 08:42:41.000000000 +0200
+++ b/include/net/netlink.h 2011-06-20 09:39:19.000000000 +0200
@@ -638,6 +638,30 @@ static inline int nlmsg_unicast(struct s
nlmsg_ok(pos, rem); \
pos = nlmsg_next(pos, &(rem)))
+/**
+ * nl_dump_check_consistent - check if sequence is consistent and advertise if not
+ * @cb: netlink callback structure that stores the sequence number
+ * @nlh: netlink message header to write the flag to
+ *
+ * This function checks if the sequence (generation) number changed during dump
+ * and if it did, advertises it in the netlink message header.
+ *
+ * The correct way to use it is to set cb->seq to the generation counter when
+ * all locks for dumping have been acquired, and then call this function for
+ * each message that is generated.
+ *
+ * Note that due to initialisation concerns, 0 is an invalid sequence number
+ * and must not be used by code that uses this functionality.
+ */
+static inline void
+nl_dump_check_consistent(struct netlink_callback *cb,
+ struct nlmsghdr *nlh)
+{
+ if (cb->prev_seq && cb->seq != cb->prev_seq)
+ nlh->nlmsg_flags |= NLM_F_DUMP_INTR;
+ cb->prev_seq = cb->seq;
+}
+
/**************************************************************************
* Netlink Attributes
**************************************************************************/
^ permalink raw reply
* Re: Linux TCP's Robustness to Multipath Packet Reordering
From: Ilpo Järvinen @ 2011-06-20 10:42 UTC (permalink / raw)
To: Dominik Kaspar
Cc: Alexander Zimmermann, Netdev, Yuchung Cheng, Carsten Wolff,
John Heffner, Eric Dumazet, Lennart Schulte, Arnd Hannemann
In-Reply-To: <BANLkTi=SdhEnyad8keUO30LsNKFNvw6AoA@mail.gmail.com>
On Sun, 19 Jun 2011, Dominik Kaspar wrote:
> Ah... the receiver DSACKs the "original" packet. However, the sender
> already received an ACK for its retransmission and advances SND.UNA.
> When the DSACK finally arrives, it is actually outside of the SND.UNA
> - SND.NXT range, which causes the DSACK to trigger "SACK reneging".
> Did I get that right? :-)
Where did you get this idea of reneging?!? Reneging has nothing to do with
DSACKs, instead it is only detected if the cumulative ACK stops to such
boundary where the _next_ segment is SACKed (i.e., some reason the
receiver "didn't bother" to cumulatively ACK for that too). ...That
certainly does not happen (ever) for out of window DSACKs.
--
i.
^ permalink raw reply
* [PATCH 11/12] batman-adv: throw uevent in userspace on gateway add/change/del event
From: Sven Eckelmann @ 2011-06-20 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, b.a.t.m.a.n, Antonio Quartulli, Marek Lindner,
Sven Eckelmann
In-Reply-To: <1308565025-21293-1-git-send-email-sven@narfation.org>
From: Antonio Quartulli <ordex@autistici.org>
In case of new default gw, changing the default gw or deleting the default gw a
uevent is triggered with type=gw, action=add/change/del and
data={GW_ORIG_ADDRESS} (if any).
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
---
net/batman-adv/gateway_client.c | 6 ++++++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 0350c0c..6381864 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -179,6 +179,7 @@ void gw_election(struct bat_priv *bat_priv)
{
struct gw_node *curr_gw = NULL, *next_gw = NULL;
struct neigh_node *router = NULL;
+ char gw_addr[18] = { '\0' };
/**
* The batman daemon checks here if we already passed a full originator
@@ -200,6 +201,8 @@ void gw_election(struct bat_priv *bat_priv)
goto out;
if (next_gw) {
+ sprintf(gw_addr, "%pM", next_gw->orig_node->orig);
+
router = orig_node_get_router(next_gw->orig_node);
if (!router) {
gw_deselect(bat_priv);
@@ -210,12 +213,14 @@ void gw_election(struct bat_priv *bat_priv)
if ((curr_gw) && (!next_gw)) {
bat_dbg(DBG_BATMAN, bat_priv,
"Removing selected gateway - no gateway in range\n");
+ throw_uevent(bat_priv, UEV_GW, UEV_DEL, NULL);
} else if ((!curr_gw) && (next_gw)) {
bat_dbg(DBG_BATMAN, bat_priv,
"Adding route to gateway %pM (gw_flags: %i, tq: %i)\n",
next_gw->orig_node->orig,
next_gw->orig_node->gw_flags,
router->tq_avg);
+ throw_uevent(bat_priv, UEV_GW, UEV_ADD, gw_addr);
} else {
bat_dbg(DBG_BATMAN, bat_priv,
"Changing route to gateway %pM "
@@ -223,6 +228,7 @@ void gw_election(struct bat_priv *bat_priv)
next_gw->orig_node->orig,
next_gw->orig_node->gw_flags,
router->tq_avg);
+ throw_uevent(bat_priv, UEV_GW, UEV_CHANGE, gw_addr);
}
gw_select(bat_priv, next_gw);
--
1.7.5.3
^ permalink raw reply related
* [PATCH 10/12] batman-adv: gateway election code refactoring
From: Sven Eckelmann @ 2011-06-20 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, b.a.t.m.a.n, Antonio Quartulli, Marek Lindner,
Sven Eckelmann
In-Reply-To: <1308565025-21293-1-git-send-email-sven@narfation.org>
From: Antonio Quartulli <ordex@autistici.org>
The gateway election mechanism has been a little revised. Now the
gw_election is trigered by an atomic_t flag (gw_reselect) which is set
to 1 in case of election needed, avoding to set curr_gw to NULL.
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
---
net/batman-adv/gateway_client.c | 142 +++++++++++++++++++++++----------------
net/batman-adv/main.c | 1 +
net/batman-adv/types.h | 1 +
3 files changed, 85 insertions(+), 59 deletions(-)
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 7248de2..0350c0c 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -20,6 +20,7 @@
*/
#include "main.h"
+#include "bat_sysfs.h"
#include "gateway_client.h"
#include "gateway_common.h"
#include "hard-interface.h"
@@ -97,40 +98,19 @@ static void gw_select(struct bat_priv *bat_priv, struct gw_node *new_gw_node)
void gw_deselect(struct bat_priv *bat_priv)
{
- gw_select(bat_priv, NULL);
+ atomic_set(&bat_priv->gw_reselect, 1);
}
-void gw_election(struct bat_priv *bat_priv)
+static struct gw_node *gw_get_best_gw_node(struct bat_priv *bat_priv)
{
- struct hlist_node *node;
- struct gw_node *gw_node, *curr_gw = NULL, *curr_gw_tmp = NULL;
struct neigh_node *router;
- uint8_t max_tq = 0;
+ struct hlist_node *node;
+ struct gw_node *gw_node, *curr_gw = NULL;
uint32_t max_gw_factor = 0, tmp_gw_factor = 0;
+ uint8_t max_tq = 0;
int down, up;
- /**
- * The batman daemon checks here if we already passed a full originator
- * cycle in order to make sure we don't choose the first gateway we
- * hear about. This check is based on the daemon's uptime which we
- * don't have.
- **/
- if (atomic_read(&bat_priv->gw_mode) != GW_MODE_CLIENT)
- return;
-
- curr_gw = gw_get_selected_gw_node(bat_priv);
- if (curr_gw)
- goto out;
-
rcu_read_lock();
- if (hlist_empty(&bat_priv->gw_list)) {
- bat_dbg(DBG_BATMAN, bat_priv,
- "Removing selected gateway - "
- "no gateway in range\n");
- gw_deselect(bat_priv);
- goto unlock;
- }
-
hlist_for_each_entry_rcu(gw_node, node, &bat_priv->gw_list, list) {
if (gw_node->deleted)
continue;
@@ -139,6 +119,9 @@ void gw_election(struct bat_priv *bat_priv)
if (!router)
continue;
+ if (!atomic_inc_not_zero(&gw_node->refcount))
+ goto next;
+
switch (atomic_read(&bat_priv->gw_sel_class)) {
case 1: /* fast connection */
gw_bandwidth_to_kbit(gw_node->orig_node->gw_flags,
@@ -151,8 +134,12 @@ void gw_election(struct bat_priv *bat_priv)
if ((tmp_gw_factor > max_gw_factor) ||
((tmp_gw_factor == max_gw_factor) &&
- (router->tq_avg > max_tq)))
- curr_gw_tmp = gw_node;
+ (router->tq_avg > max_tq))) {
+ if (curr_gw)
+ gw_node_free_ref(curr_gw);
+ curr_gw = gw_node;
+ atomic_inc(&curr_gw->refcount);
+ }
break;
default: /**
@@ -163,8 +150,12 @@ void gw_election(struct bat_priv *bat_priv)
* soon as a better gateway appears which has
* $routing_class more tq points)
**/
- if (router->tq_avg > max_tq)
- curr_gw_tmp = gw_node;
+ if (router->tq_avg > max_tq) {
+ if (curr_gw)
+ gw_node_free_ref(curr_gw);
+ curr_gw = gw_node;
+ atomic_inc(&curr_gw->refcount);
+ }
break;
}
@@ -174,42 +165,75 @@ void gw_election(struct bat_priv *bat_priv)
if (tmp_gw_factor > max_gw_factor)
max_gw_factor = tmp_gw_factor;
- neigh_node_free_ref(router);
- }
-
- if (curr_gw != curr_gw_tmp) {
- router = orig_node_get_router(curr_gw_tmp->orig_node);
- if (!router)
- goto unlock;
-
- if ((curr_gw) && (!curr_gw_tmp))
- bat_dbg(DBG_BATMAN, bat_priv,
- "Removing selected gateway - "
- "no gateway in range\n");
- else if ((!curr_gw) && (curr_gw_tmp))
- bat_dbg(DBG_BATMAN, bat_priv,
- "Adding route to gateway %pM "
- "(gw_flags: %i, tq: %i)\n",
- curr_gw_tmp->orig_node->orig,
- curr_gw_tmp->orig_node->gw_flags,
- router->tq_avg);
- else
- bat_dbg(DBG_BATMAN, bat_priv,
- "Changing route to gateway %pM "
- "(gw_flags: %i, tq: %i)\n",
- curr_gw_tmp->orig_node->orig,
- curr_gw_tmp->orig_node->gw_flags,
- router->tq_avg);
+ gw_node_free_ref(gw_node);
+next:
neigh_node_free_ref(router);
- gw_select(bat_priv, curr_gw_tmp);
}
-
-unlock:
rcu_read_unlock();
+
+ return curr_gw;
+}
+
+void gw_election(struct bat_priv *bat_priv)
+{
+ struct gw_node *curr_gw = NULL, *next_gw = NULL;
+ struct neigh_node *router = NULL;
+
+ /**
+ * The batman daemon checks here if we already passed a full originator
+ * cycle in order to make sure we don't choose the first gateway we
+ * hear about. This check is based on the daemon's uptime which we
+ * don't have.
+ **/
+ if (atomic_read(&bat_priv->gw_mode) != GW_MODE_CLIENT)
+ goto out;
+
+ if (!atomic_dec_not_zero(&bat_priv->gw_reselect))
+ goto out;
+
+ curr_gw = gw_get_selected_gw_node(bat_priv);
+
+ next_gw = gw_get_best_gw_node(bat_priv);
+
+ if (curr_gw == next_gw)
+ goto out;
+
+ if (next_gw) {
+ router = orig_node_get_router(next_gw->orig_node);
+ if (!router) {
+ gw_deselect(bat_priv);
+ goto out;
+ }
+ }
+
+ if ((curr_gw) && (!next_gw)) {
+ bat_dbg(DBG_BATMAN, bat_priv,
+ "Removing selected gateway - no gateway in range\n");
+ } else if ((!curr_gw) && (next_gw)) {
+ bat_dbg(DBG_BATMAN, bat_priv,
+ "Adding route to gateway %pM (gw_flags: %i, tq: %i)\n",
+ next_gw->orig_node->orig,
+ next_gw->orig_node->gw_flags,
+ router->tq_avg);
+ } else {
+ bat_dbg(DBG_BATMAN, bat_priv,
+ "Changing route to gateway %pM "
+ "(gw_flags: %i, tq: %i)\n",
+ next_gw->orig_node->orig,
+ next_gw->orig_node->gw_flags,
+ router->tq_avg);
+ }
+
+ gw_select(bat_priv, next_gw);
+
out:
if (curr_gw)
gw_node_free_ref(curr_gw);
+ if (next_gw)
+ gw_node_free_ref(next_gw);
+ if (router)
+ neigh_node_free_ref(router);
}
void gw_check_election(struct bat_priv *bat_priv, struct orig_node *orig_node)
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index c2b06b7..e367e69 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -113,6 +113,7 @@ int mesh_init(struct net_device *soft_iface)
if (vis_init(bat_priv) < 1)
goto err;
+ atomic_set(&bat_priv->gw_reselect, 0);
atomic_set(&bat_priv->mesh_state, MESH_ACTIVE);
goto end;
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 11e8569..85cf122 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -201,6 +201,7 @@ struct bat_priv {
struct delayed_work orig_work;
struct delayed_work vis_work;
struct gw_node __rcu *curr_gw; /* rcu protected pointer */
+ atomic_t gw_reselect;
struct hard_iface __rcu *primary_if; /* rcu protected pointer */
struct vis_info *my_vis_info;
};
--
1.7.5.3
^ permalink raw reply related
* [PATCH 09/12] batman-adv: add wrapper function to throw uevent in userspace
From: Sven Eckelmann @ 2011-06-20 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, b.a.t.m.a.n, Antonio Quartulli, Sven Eckelmann
In-Reply-To: <1308565025-21293-1-git-send-email-sven@narfation.org>
From: Antonio Quartulli <ordex@autistici.org>
Using throw_uevent() is now possible to trigger uevent signal that can
be recognised in userspace. Uevents will be triggered through the
/devices/virtual/net/{MESH_IFACE} kobject.
A triggered uevent has three properties:
- type: the event class. Who generates the event (only 'gw' is currently
defined). Corresponds to the BATTYPE uevent variable.
- action: the associated action with the event ('add'/'change'/'del' are
currently defined). Corresponds to the BATACTION uevent variable.
- data: any useful data for the userspace. Corresponds to the BATDATA
uevent variable.
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
---
net/batman-adv/bat_sysfs.c | 71 ++++++++++++++++++++++++++++++++++++++++++++
net/batman-adv/bat_sysfs.h | 2 +
net/batman-adv/main.h | 11 +++++++
3 files changed, 84 insertions(+), 0 deletions(-)
diff --git a/net/batman-adv/bat_sysfs.c b/net/batman-adv/bat_sysfs.c
index 63738ec..cd15deb 100644
--- a/net/batman-adv/bat_sysfs.c
+++ b/net/batman-adv/bat_sysfs.c
@@ -40,6 +40,20 @@ static struct bat_priv *kobj_to_batpriv(struct kobject *obj)
return netdev_priv(net_dev);
}
+#define UEV_TYPE_VAR "BATTYPE="
+#define UEV_ACTION_VAR "BATACTION="
+#define UEV_DATA_VAR "BATDATA="
+
+static char *uev_action_str[] = {
+ "add",
+ "del",
+ "change"
+};
+
+static char *uev_type_str[] = {
+ "gw"
+};
+
/* Use this, if you have customized show and store functions */
#define BAT_ATTR(_name, _mode, _show, _store) \
struct bat_attribute bat_attr_##_name = { \
@@ -601,3 +615,60 @@ void sysfs_del_hardif(struct kobject **hardif_obj)
kobject_put(*hardif_obj);
*hardif_obj = NULL;
}
+
+int throw_uevent(struct bat_priv *bat_priv, enum uev_type type,
+ enum uev_action action, const char *data)
+{
+ int ret = -1;
+ struct hard_iface *primary_if = NULL;
+ struct kobject *bat_kobj;
+ char *uevent_env[4] = { NULL, NULL, NULL, NULL };
+
+ primary_if = primary_if_get_selected(bat_priv);
+ if (!primary_if)
+ goto out;
+
+ bat_kobj = &primary_if->soft_iface->dev.kobj;
+
+ uevent_env[0] = kmalloc(strlen(UEV_TYPE_VAR) +
+ strlen(uev_type_str[type]) + 1,
+ GFP_ATOMIC);
+ if (!uevent_env[0])
+ goto out;
+
+ sprintf(uevent_env[0], "%s%s", UEV_TYPE_VAR, uev_type_str[type]);
+
+ uevent_env[1] = kmalloc(strlen(UEV_ACTION_VAR) +
+ strlen(uev_action_str[action]) + 1,
+ GFP_ATOMIC);
+ if (!uevent_env[1])
+ goto out;
+
+ sprintf(uevent_env[1], "%s%s", UEV_ACTION_VAR, uev_action_str[action]);
+
+ /* If the event is DEL, ignore the data field */
+ if (action != UEV_DEL) {
+ uevent_env[2] = kmalloc(strlen(UEV_DATA_VAR) +
+ strlen(data) + 1, GFP_ATOMIC);
+ if (!uevent_env[2])
+ goto out;
+
+ sprintf(uevent_env[2], "%s%s", UEV_DATA_VAR, data);
+ }
+
+ ret = kobject_uevent_env(bat_kobj, KOBJ_CHANGE, uevent_env);
+out:
+ kfree(uevent_env[0]);
+ kfree(uevent_env[1]);
+ kfree(uevent_env[2]);
+
+ if (primary_if)
+ hardif_free_ref(primary_if);
+
+ if (ret)
+ bat_dbg(DBG_BATMAN, bat_priv, "Impossible to send "
+ "uevent for (%s,%s,%s) event (err: %d)\n",
+ uev_type_str[type], uev_action_str[action],
+ (action == UEV_DEL ? "NULL" : data), ret);
+ return ret;
+}
diff --git a/net/batman-adv/bat_sysfs.h b/net/batman-adv/bat_sysfs.h
index 02f1fa7..a3f75a7 100644
--- a/net/batman-adv/bat_sysfs.h
+++ b/net/batman-adv/bat_sysfs.h
@@ -38,5 +38,7 @@ int sysfs_add_meshif(struct net_device *dev);
void sysfs_del_meshif(struct net_device *dev);
int sysfs_add_hardif(struct kobject **hardif_obj, struct net_device *dev);
void sysfs_del_hardif(struct kobject **hardif_obj);
+int throw_uevent(struct bat_priv *bat_priv, enum uev_type type,
+ enum uev_action action, const char *data);
#endif /* _NET_BATMAN_ADV_SYSFS_H_ */
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 8eae05e..f9e0e17 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -91,6 +91,17 @@ enum mesh_state {
#define BCAST_QUEUE_LEN 256
#define BATMAN_QUEUE_LEN 256
+
+enum uev_action {
+ UEV_ADD = 0,
+ UEV_DEL,
+ UEV_CHANGE
+};
+
+enum uev_type {
+ UEV_GW = 0
+};
+
/*
* Debug Messages
*/
--
1.7.5.3
^ permalink raw reply related
* [PATCH 08/12] batman-adv: protect the local and the global trans-tables with rcu
From: Sven Eckelmann @ 2011-06-20 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, b.a.t.m.a.n, Antonio Quartulli, Sven Eckelmann
In-Reply-To: <1308565025-21293-1-git-send-email-sven@narfation.org>
From: Antonio Quartulli <ordex@autistici.org>
The local and the global translation-tables are now lock free and rcu
protected.
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
---
net/batman-adv/main.c | 2 -
net/batman-adv/routing.c | 2 -
net/batman-adv/translation-table.c | 255 ++++++++++++++++++++----------------
net/batman-adv/types.h | 6 +-
net/batman-adv/vis.c | 13 +-
5 files changed, 150 insertions(+), 128 deletions(-)
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 3318ee2..c2b06b7 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -84,8 +84,6 @@ int mesh_init(struct net_device *soft_iface)
spin_lock_init(&bat_priv->forw_bat_list_lock);
spin_lock_init(&bat_priv->forw_bcast_list_lock);
- spin_lock_init(&bat_priv->tt_lhash_lock);
- spin_lock_init(&bat_priv->tt_ghash_lock);
spin_lock_init(&bat_priv->tt_changes_list_lock);
spin_lock_init(&bat_priv->tt_req_list_lock);
spin_lock_init(&bat_priv->tt_roam_list_lock);
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 05d50ca..0ce090c 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -90,9 +90,7 @@ static void update_transtable(struct bat_priv *bat_priv,
/* Even if we received the crc into the OGM, we prefer
* to recompute it to spot any possible inconsistency
* in the global table */
- spin_lock_bh(&bat_priv->tt_ghash_lock);
orig_node->tt_crc = tt_global_crc(bat_priv, orig_node);
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
/* Roaming phase is over: tables are in sync again. I can
* unset the flag */
orig_node->tt_poss_change = false;
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index d516d85..5f1fcd5 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -80,6 +80,9 @@ static struct tt_local_entry *tt_local_hash_find(struct bat_priv *bat_priv,
if (!compare_eth(tt_local_entry, data))
continue;
+ if (!atomic_inc_not_zero(&tt_local_entry->refcount))
+ continue;
+
tt_local_entry_tmp = tt_local_entry;
break;
}
@@ -109,6 +112,9 @@ static struct tt_global_entry *tt_global_hash_find(struct bat_priv *bat_priv,
if (!compare_eth(tt_global_entry, data))
continue;
+ if (!atomic_inc_not_zero(&tt_global_entry->refcount))
+ continue;
+
tt_global_entry_tmp = tt_global_entry;
break;
}
@@ -125,8 +131,20 @@ static bool is_out_of_time(unsigned long starting_time, unsigned long timeout)
return time_after(jiffies, deadline);
}
+static void tt_local_entry_free_ref(struct tt_local_entry *tt_local_entry)
+{
+ if (atomic_dec_and_test(&tt_local_entry->refcount))
+ kfree_rcu(tt_local_entry, rcu);
+}
+
+static void tt_global_entry_free_ref(struct tt_global_entry *tt_global_entry)
+{
+ if (atomic_dec_and_test(&tt_global_entry->refcount))
+ kfree_rcu(tt_global_entry, rcu);
+}
+
static void tt_local_event(struct bat_priv *bat_priv, uint8_t op,
- const uint8_t *addr, uint8_t roaming)
+ const uint8_t *addr, bool roaming)
{
struct tt_change_node *tt_change_node;
@@ -171,21 +189,19 @@ static int tt_local_init(struct bat_priv *bat_priv)
void tt_local_add(struct net_device *soft_iface, const uint8_t *addr)
{
struct bat_priv *bat_priv = netdev_priv(soft_iface);
- struct tt_local_entry *tt_local_entry;
- struct tt_global_entry *tt_global_entry;
- uint8_t roam_addr[ETH_ALEN];
+ struct tt_local_entry *tt_local_entry = NULL;
+ struct tt_global_entry *tt_global_entry = NULL;
- spin_lock_bh(&bat_priv->tt_lhash_lock);
tt_local_entry = tt_local_hash_find(bat_priv, addr);
if (tt_local_entry) {
tt_local_entry->last_seen = jiffies;
- goto unlock;
+ goto out;
}
tt_local_entry = kmalloc(sizeof(*tt_local_entry), GFP_ATOMIC);
if (!tt_local_entry)
- goto unlock;
+ goto out;
tt_local_event(bat_priv, NO_FLAGS, addr, false);
@@ -195,6 +211,7 @@ void tt_local_add(struct net_device *soft_iface, const uint8_t *addr)
memcpy(tt_local_entry->addr, addr, ETH_ALEN);
tt_local_entry->last_seen = jiffies;
+ atomic_set(&tt_local_entry->refcount, 2);
/* the batman interface mac address should never be purged */
if (compare_eth(addr, soft_iface->dev_addr))
@@ -204,30 +221,26 @@ void tt_local_add(struct net_device *soft_iface, const uint8_t *addr)
hash_add(bat_priv->tt_local_hash, compare_ltt, choose_orig,
tt_local_entry, &tt_local_entry->hash_entry);
+
atomic_inc(&bat_priv->num_local_tt);
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
/* remove address from global hash if present */
- spin_lock_bh(&bat_priv->tt_ghash_lock);
-
tt_global_entry = tt_global_hash_find(bat_priv, addr);
/* Check whether it is a roaming! */
if (tt_global_entry) {
- memcpy(roam_addr, tt_global_entry->addr, ETH_ALEN);
/* This node is probably going to update its tt table */
tt_global_entry->orig_node->tt_poss_change = true;
_tt_global_del(bat_priv, tt_global_entry,
"local tt received");
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
send_roam_adv(bat_priv, tt_global_entry->addr,
- tt_global_entry->orig_node);
- } else
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
-
- return;
-unlock:
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
+ tt_global_entry->orig_node);
+ }
+out:
+ if (tt_local_entry)
+ tt_local_entry_free_ref(tt_local_entry);
+ if (tt_global_entry)
+ tt_global_entry_free_ref(tt_global_entry);
}
int tt_changes_fill_buffer(struct bat_priv *bat_priv,
@@ -309,8 +322,6 @@ int tt_local_seq_print_text(struct seq_file *seq, void *offset)
"announced via TT (TTVN: %u):\n",
net_dev->name, (uint8_t)atomic_read(&bat_priv->ttvn));
- spin_lock_bh(&bat_priv->tt_lhash_lock);
-
buf_size = 1;
/* Estimate length for: " * xx:xx:xx:xx:xx:xx\n" */
for (i = 0; i < hash->size; i++) {
@@ -324,7 +335,6 @@ int tt_local_seq_print_text(struct seq_file *seq, void *offset)
buff = kmalloc(buf_size, GFP_ATOMIC);
if (!buff) {
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
ret = -ENOMEM;
goto out;
}
@@ -344,8 +354,6 @@ int tt_local_seq_print_text(struct seq_file *seq, void *offset)
rcu_read_unlock();
}
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
-
seq_printf(seq, "%s", buff);
kfree(buff);
out:
@@ -354,15 +362,6 @@ out:
return ret;
}
-static void tt_local_entry_free(struct hlist_node *node, void *arg)
-{
- struct bat_priv *bat_priv = arg;
- void *data = container_of(node, struct tt_local_entry, hash_entry);
-
- kfree(data);
- atomic_dec(&bat_priv->num_local_tt);
-}
-
static void tt_local_del(struct bat_priv *bat_priv,
struct tt_local_entry *tt_local_entry,
const char *message)
@@ -375,23 +374,24 @@ static void tt_local_del(struct bat_priv *bat_priv,
hash_remove(bat_priv->tt_local_hash, compare_ltt, choose_orig,
tt_local_entry->addr);
- tt_local_entry_free(&tt_local_entry->hash_entry, bat_priv);
+ tt_local_entry_free_ref(tt_local_entry);
}
void tt_local_remove(struct bat_priv *bat_priv, const uint8_t *addr,
const char *message, bool roaming)
{
- struct tt_local_entry *tt_local_entry;
+ struct tt_local_entry *tt_local_entry = NULL;
- spin_lock_bh(&bat_priv->tt_lhash_lock);
tt_local_entry = tt_local_hash_find(bat_priv, addr);
- if (tt_local_entry) {
- tt_local_event(bat_priv, TT_CHANGE_DEL, tt_local_entry->addr,
- roaming);
- tt_local_del(bat_priv, tt_local_entry, message);
- }
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
+ if (!tt_local_entry)
+ goto out;
+
+ tt_local_event(bat_priv, TT_CHANGE_DEL, tt_local_entry->addr, roaming);
+ tt_local_del(bat_priv, tt_local_entry, message);
+out:
+ if (tt_local_entry)
+ tt_local_entry_free_ref(tt_local_entry);
}
static void tt_local_purge(struct bat_priv *bat_priv)
@@ -400,40 +400,45 @@ static void tt_local_purge(struct bat_priv *bat_priv)
struct tt_local_entry *tt_local_entry;
struct hlist_node *node, *node_tmp;
struct hlist_head *head;
+ spinlock_t *list_lock; /* protects write access to the hash lists */
int i;
- spin_lock_bh(&bat_priv->tt_lhash_lock);
-
for (i = 0; i < hash->size; i++) {
head = &hash->table[i];
+ list_lock = &hash->list_locks[i];
+ spin_lock_bh(list_lock);
hlist_for_each_entry_safe(tt_local_entry, node, node_tmp,
head, hash_entry) {
if (tt_local_entry->never_purge)
continue;
if (!is_out_of_time(tt_local_entry->last_seen,
- TT_LOCAL_TIMEOUT * 1000))
+ TT_LOCAL_TIMEOUT * 1000))
continue;
tt_local_event(bat_priv, TT_CHANGE_DEL,
tt_local_entry->addr, false);
- tt_local_del(bat_priv, tt_local_entry,
- "address timed out");
+ atomic_dec(&bat_priv->num_local_tt);
+ bat_dbg(DBG_TT, bat_priv, "Deleting local "
+ "tt entry (%pM): timed out\n",
+ tt_local_entry->addr);
+ hlist_del_rcu(node);
+ tt_local_entry_free_ref(tt_local_entry);
}
+ spin_unlock_bh(list_lock);
}
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
}
static void tt_local_table_free(struct bat_priv *bat_priv)
{
struct hashtable_t *hash;
- int i;
spinlock_t *list_lock; /* protects write access to the hash lists */
- struct hlist_head *head;
- struct hlist_node *node, *node_tmp;
struct tt_local_entry *tt_local_entry;
+ struct hlist_node *node, *node_tmp;
+ struct hlist_head *head;
+ int i;
if (!bat_priv->tt_local_hash)
return;
@@ -448,7 +453,7 @@ static void tt_local_table_free(struct bat_priv *bat_priv)
hlist_for_each_entry_safe(tt_local_entry, node, node_tmp,
head, hash_entry) {
hlist_del_rcu(node);
- kfree(tt_local_entry);
+ tt_local_entry_free_ref(tt_local_entry);
}
spin_unlock_bh(list_lock);
}
@@ -492,10 +497,9 @@ int tt_global_add(struct bat_priv *bat_priv, struct orig_node *orig_node,
const unsigned char *tt_addr, uint8_t ttvn, bool roaming)
{
struct tt_global_entry *tt_global_entry;
- struct tt_local_entry *tt_local_entry;
struct orig_node *orig_node_tmp;
+ int ret = 0;
- spin_lock_bh(&bat_priv->tt_ghash_lock);
tt_global_entry = tt_global_hash_find(bat_priv, tt_addr);
if (!tt_global_entry) {
@@ -503,7 +507,8 @@ int tt_global_add(struct bat_priv *bat_priv, struct orig_node *orig_node,
kmalloc(sizeof(*tt_global_entry),
GFP_ATOMIC);
if (!tt_global_entry)
- goto unlock;
+ goto out;
+
memcpy(tt_global_entry->addr, tt_addr, ETH_ALEN);
/* Assign the new orig_node */
atomic_inc(&orig_node->refcount);
@@ -511,10 +516,12 @@ int tt_global_add(struct bat_priv *bat_priv, struct orig_node *orig_node,
tt_global_entry->ttvn = ttvn;
tt_global_entry->flags = NO_FLAGS;
tt_global_entry->roam_at = 0;
- atomic_inc(&orig_node->tt_size);
+ atomic_set(&tt_global_entry->refcount, 2);
+
hash_add(bat_priv->tt_global_hash, compare_gtt,
choose_orig, tt_global_entry,
&tt_global_entry->hash_entry);
+ atomic_inc(&orig_node->tt_size);
} else {
if (tt_global_entry->orig_node != orig_node) {
atomic_dec(&tt_global_entry->orig_node->tt_size);
@@ -529,25 +536,18 @@ int tt_global_add(struct bat_priv *bat_priv, struct orig_node *orig_node,
tt_global_entry->roam_at = 0;
}
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
-
bat_dbg(DBG_TT, bat_priv,
"Creating new global tt entry: %pM (via %pM)\n",
tt_global_entry->addr, orig_node->orig);
/* remove address from local hash if present */
- spin_lock_bh(&bat_priv->tt_lhash_lock);
- tt_local_entry = tt_local_hash_find(bat_priv, tt_addr);
-
- if (tt_local_entry)
- tt_local_remove(bat_priv, tt_global_entry->addr,
- "global tt received", roaming);
-
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
- return 1;
-unlock:
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
- return 0;
+ tt_local_remove(bat_priv, tt_global_entry->addr,
+ "global tt received", roaming);
+ ret = 1;
+out:
+ if (tt_global_entry)
+ tt_global_entry_free_ref(tt_global_entry);
+ return ret;
}
int tt_global_seq_print_text(struct seq_file *seq, void *offset)
@@ -584,8 +584,6 @@ int tt_global_seq_print_text(struct seq_file *seq, void *offset)
seq_printf(seq, " %-13s %s %-15s %s\n",
"Client", "(TTVN)", "Originator", "(Curr TTVN)");
- spin_lock_bh(&bat_priv->tt_ghash_lock);
-
buf_size = 1;
/* Estimate length for: " * xx:xx:xx:xx:xx:xx (ttvn) via
* xx:xx:xx:xx:xx:xx (cur_ttvn)\n"*/
@@ -600,10 +598,10 @@ int tt_global_seq_print_text(struct seq_file *seq, void *offset)
buff = kmalloc(buf_size, GFP_ATOMIC);
if (!buff) {
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
ret = -ENOMEM;
goto out;
}
+
buff[0] = '\0';
pos = 0;
@@ -625,8 +623,6 @@ int tt_global_seq_print_text(struct seq_file *seq, void *offset)
rcu_read_unlock();
}
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
-
seq_printf(seq, "%s", buff);
kfree(buff);
out:
@@ -640,7 +636,7 @@ static void _tt_global_del(struct bat_priv *bat_priv,
const char *message)
{
if (!tt_global_entry)
- return;
+ goto out;
bat_dbg(DBG_TT, bat_priv,
"Deleting global tt entry %pM (via %pM): %s\n",
@@ -648,31 +644,35 @@ static void _tt_global_del(struct bat_priv *bat_priv,
message);
atomic_dec(&tt_global_entry->orig_node->tt_size);
+
hash_remove(bat_priv->tt_global_hash, compare_gtt, choose_orig,
tt_global_entry->addr);
- kfree(tt_global_entry);
+out:
+ if (tt_global_entry)
+ tt_global_entry_free_ref(tt_global_entry);
}
void tt_global_del(struct bat_priv *bat_priv,
struct orig_node *orig_node, const unsigned char *addr,
const char *message, bool roaming)
{
- struct tt_global_entry *tt_global_entry;
+ struct tt_global_entry *tt_global_entry = NULL;
- spin_lock_bh(&bat_priv->tt_ghash_lock);
tt_global_entry = tt_global_hash_find(bat_priv, addr);
+ if (!tt_global_entry)
+ goto out;
- if (tt_global_entry && tt_global_entry->orig_node == orig_node) {
+ if (tt_global_entry->orig_node == orig_node) {
if (roaming) {
tt_global_entry->flags |= TT_CLIENT_ROAM;
tt_global_entry->roam_at = jiffies;
goto out;
}
- atomic_dec(&orig_node->tt_size);
_tt_global_del(bat_priv, tt_global_entry, message);
}
out:
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
+ if (tt_global_entry)
+ tt_global_entry_free_ref(tt_global_entry);
}
void tt_global_del_orig(struct bat_priv *bat_priv,
@@ -683,30 +683,28 @@ void tt_global_del_orig(struct bat_priv *bat_priv,
struct hashtable_t *hash = bat_priv->tt_global_hash;
struct hlist_node *node, *safe;
struct hlist_head *head;
+ spinlock_t *list_lock; /* protects write access to the hash lists */
- if (!bat_priv->tt_global_hash)
- return;
-
- spin_lock_bh(&bat_priv->tt_ghash_lock);
for (i = 0; i < hash->size; i++) {
head = &hash->table[i];
+ list_lock = &hash->list_locks[i];
+ spin_lock_bh(list_lock);
hlist_for_each_entry_safe(tt_global_entry, node, safe,
head, hash_entry) {
- if (tt_global_entry->orig_node == orig_node)
- _tt_global_del(bat_priv, tt_global_entry,
- message);
+ if (tt_global_entry->orig_node == orig_node) {
+ bat_dbg(DBG_TT, bat_priv,
+ "Deleting global tt entry %pM "
+ "(via %pM): originator time out\n",
+ tt_global_entry->addr,
+ tt_global_entry->orig_node->orig);
+ hlist_del_rcu(node);
+ tt_global_entry_free_ref(tt_global_entry);
+ }
}
+ spin_unlock_bh(list_lock);
}
atomic_set(&orig_node->tt_size, 0);
-
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
-}
-
-static void tt_global_entry_free(struct hlist_node *node, void *arg)
-{
- void *data = container_of(node, struct tt_global_entry, hash_entry);
- kfree(data);
}
static void tt_global_roam_purge(struct bat_priv *bat_priv)
@@ -715,13 +713,14 @@ static void tt_global_roam_purge(struct bat_priv *bat_priv)
struct tt_global_entry *tt_global_entry;
struct hlist_node *node, *node_tmp;
struct hlist_head *head;
+ spinlock_t *list_lock; /* protects write access to the hash lists */
int i;
- spin_lock_bh(&bat_priv->tt_ghash_lock);
-
for (i = 0; i < hash->size; i++) {
head = &hash->table[i];
+ list_lock = &hash->list_locks[i];
+ spin_lock_bh(list_lock);
hlist_for_each_entry_safe(tt_global_entry, node, node_tmp,
head, hash_entry) {
if (!(tt_global_entry->flags & TT_CLIENT_ROAM))
@@ -730,20 +729,47 @@ static void tt_global_roam_purge(struct bat_priv *bat_priv)
TT_CLIENT_ROAM_TIMEOUT * 1000))
continue;
- _tt_global_del(bat_priv, tt_global_entry,
- "Roaming timeout");
+ bat_dbg(DBG_TT, bat_priv, "Deleting global "
+ "tt entry (%pM): Roaming timeout\n",
+ tt_global_entry->addr);
+ atomic_dec(&tt_global_entry->orig_node->tt_size);
+ hlist_del_rcu(node);
+ tt_global_entry_free_ref(tt_global_entry);
}
+ spin_unlock_bh(list_lock);
}
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
}
static void tt_global_table_free(struct bat_priv *bat_priv)
{
+ struct hashtable_t *hash;
+ spinlock_t *list_lock; /* protects write access to the hash lists */
+ struct tt_global_entry *tt_global_entry;
+ struct hlist_node *node, *node_tmp;
+ struct hlist_head *head;
+ int i;
+
if (!bat_priv->tt_global_hash)
return;
- hash_delete(bat_priv->tt_global_hash, tt_global_entry_free, NULL);
+ hash = bat_priv->tt_global_hash;
+
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+ list_lock = &hash->list_locks[i];
+
+ spin_lock_bh(list_lock);
+ hlist_for_each_entry_safe(tt_global_entry, node, node_tmp,
+ head, hash_entry) {
+ hlist_del_rcu(node);
+ tt_global_entry_free_ref(tt_global_entry);
+ }
+ spin_unlock_bh(list_lock);
+ }
+
+ hash_destroy(hash);
+
bat_priv->tt_global_hash = NULL;
}
@@ -753,19 +779,19 @@ struct orig_node *transtable_search(struct bat_priv *bat_priv,
struct tt_global_entry *tt_global_entry;
struct orig_node *orig_node = NULL;
- spin_lock_bh(&bat_priv->tt_ghash_lock);
tt_global_entry = tt_global_hash_find(bat_priv, addr);
if (!tt_global_entry)
goto out;
if (!atomic_inc_not_zero(&tt_global_entry->orig_node->refcount))
- goto out;
+ goto free_tt;
orig_node = tt_global_entry->orig_node;
+free_tt:
+ tt_global_entry_free_ref(tt_global_entry);
out:
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
return orig_node;
}
@@ -828,7 +854,6 @@ uint16_t tt_local_crc(struct bat_priv *bat_priv)
tt_local_entry->addr[j]);
total ^= total_one;
}
-
rcu_read_unlock();
}
@@ -1371,15 +1396,17 @@ void tt_update_changes(struct bat_priv *bat_priv, struct orig_node *orig_node,
bool is_my_client(struct bat_priv *bat_priv, const uint8_t *addr)
{
- struct tt_local_entry *tt_local_entry;
+ struct tt_local_entry *tt_local_entry = NULL;
+ bool ret = false;
- spin_lock_bh(&bat_priv->tt_lhash_lock);
tt_local_entry = tt_local_hash_find(bat_priv, addr);
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
-
+ if (!tt_local_entry)
+ goto out;
+ ret = true;
+out:
if (tt_local_entry)
- return true;
- return false;
+ tt_local_entry_free_ref(tt_local_entry);
+ return ret;
}
void handle_tt_response(struct bat_priv *bat_priv,
@@ -1416,9 +1443,7 @@ void handle_tt_response(struct bat_priv *bat_priv,
spin_unlock_bh(&bat_priv->tt_req_list_lock);
/* Recalculate the CRC for this orig_node and store it */
- spin_lock_bh(&bat_priv->tt_ghash_lock);
orig_node->tt_crc = tt_global_crc(bat_priv, orig_node);
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
/* Roaming phase is over: tables are in sync again. I can
* unset the flag */
orig_node->tt_poss_change = false;
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 9c84fa9..11e8569 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -184,8 +184,6 @@ struct bat_priv {
spinlock_t forw_bat_list_lock; /* protects forw_bat_list */
spinlock_t forw_bcast_list_lock; /* protects */
spinlock_t tt_changes_list_lock; /* protects tt_changes */
- spinlock_t tt_lhash_lock; /* protects tt_local_hash */
- spinlock_t tt_ghash_lock; /* protects tt_global_hash */
spinlock_t tt_req_list_lock; /* protects tt_req_list */
spinlock_t tt_roam_list_lock; /* protects tt_roam_list */
spinlock_t gw_list_lock; /* protects gw_list and curr_gw */
@@ -226,6 +224,8 @@ struct tt_local_entry {
uint8_t addr[ETH_ALEN];
unsigned long last_seen;
char never_purge;
+ atomic_t refcount;
+ struct rcu_head rcu;
struct hlist_node hash_entry;
};
@@ -235,6 +235,8 @@ struct tt_global_entry {
uint8_t ttvn;
uint8_t flags; /* only TT_GLOBAL_ROAM is used */
unsigned long roam_at; /* time at which TT_GLOBAL_ROAM was set */
+ atomic_t refcount;
+ struct rcu_head rcu;
struct hlist_node hash_entry; /* entry in the global table */
};
diff --git a/net/batman-adv/vis.c b/net/batman-adv/vis.c
index 355c6e5..8a1b985 100644
--- a/net/batman-adv/vis.c
+++ b/net/batman-adv/vis.c
@@ -665,11 +665,12 @@ next:
hash = bat_priv->tt_local_hash;
- spin_lock_bh(&bat_priv->tt_lhash_lock);
for (i = 0; i < hash->size; i++) {
head = &hash->table[i];
- hlist_for_each_entry(tt_local_entry, node, head, hash_entry) {
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(tt_local_entry, node, head,
+ hash_entry) {
entry = (struct vis_info_entry *)
skb_put(info->skb_packet,
sizeof(*entry));
@@ -678,14 +679,12 @@ next:
entry->quality = 0; /* 0 means TT */
packet->entries++;
- if (vis_packet_full(info)) {
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
- return 0;
- }
+ if (vis_packet_full(info))
+ goto unlock;
}
+ rcu_read_unlock();
}
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
return 0;
unlock:
--
1.7.5.3
^ permalink raw reply related
* [PATCH 05/12] batman-adv: Unify the first 3 bytes in each packet
From: Sven Eckelmann @ 2011-06-20 10:16 UTC (permalink / raw)
To: davem; +Cc: netdev, b.a.t.m.a.n, Antonio Quartulli, Sven Eckelmann
In-Reply-To: <1308565025-21293-1-git-send-email-sven@narfation.org>
From: Antonio Quartulli <ordex@autistici.org>
The amount of duplicated code in the receive and routing code can be
reduced when all headers provide the packet type, version and ttl in the
same first bytes.
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
---
net/batman-adv/packet.h | 30 ++++++++++++++++++------------
1 files changed, 18 insertions(+), 12 deletions(-)
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index 9f77086..6ddbfd8 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -34,7 +34,7 @@ enum bat_packettype {
};
/* this file is included by batctl which needs these defines */
-#define COMPAT_VERSION 12
+#define COMPAT_VERSION 14
enum batman_flags {
PRIMARIES_FIRST_HOP = 1 << 4,
@@ -66,15 +66,15 @@ enum unicast_frag_flags {
struct batman_packet {
uint8_t packet_type;
uint8_t version; /* batman version field */
+ uint8_t ttl;
uint8_t flags; /* 0x40: DIRECTLINK flag, 0x20 VIS_SERVER flag... */
- uint8_t tq;
uint32_t seqno;
uint8_t orig[6];
uint8_t prev_sender[6];
- uint8_t ttl;
- uint8_t num_tt;
uint8_t gw_flags; /* flags related to gateway class */
- uint8_t align;
+ uint8_t tq;
+ uint8_t num_tt;
+ uint8_t reserved;
} __packed;
#define BAT_PACKET_LEN sizeof(struct batman_packet)
@@ -82,12 +82,13 @@ struct batman_packet {
struct icmp_packet {
uint8_t packet_type;
uint8_t version; /* batman version field */
- uint8_t msg_type; /* see ICMP message types above */
uint8_t ttl;
+ uint8_t msg_type; /* see ICMP message types above */
uint8_t dst[6];
uint8_t orig[6];
uint16_t seqno;
uint8_t uid;
+ uint8_t reserved;
} __packed;
#define BAT_RR_LEN 16
@@ -97,8 +98,8 @@ struct icmp_packet {
struct icmp_packet_rr {
uint8_t packet_type;
uint8_t version; /* batman version field */
- uint8_t msg_type; /* see ICMP message types above */
uint8_t ttl;
+ uint8_t msg_type; /* see ICMP message types above */
uint8_t dst[6];
uint8_t orig[6];
uint16_t seqno;
@@ -110,16 +111,19 @@ struct icmp_packet_rr {
struct unicast_packet {
uint8_t packet_type;
uint8_t version; /* batman version field */
- uint8_t dest[6];
uint8_t ttl;
+ uint8_t reserved;
+ uint8_t dest[6];
} __packed;
struct unicast_frag_packet {
uint8_t packet_type;
uint8_t version; /* batman version field */
- uint8_t dest[6];
uint8_t ttl;
+ uint8_t reserved;
+ uint8_t dest[6];
uint8_t flags;
+ uint8_t align;
uint8_t orig[6];
uint16_t seqno;
} __packed;
@@ -127,18 +131,20 @@ struct unicast_frag_packet {
struct bcast_packet {
uint8_t packet_type;
uint8_t version; /* batman version field */
- uint8_t orig[6];
uint8_t ttl;
+ uint8_t reserved;
uint32_t seqno;
+ uint8_t orig[6];
} __packed;
struct vis_packet {
uint8_t packet_type;
uint8_t version; /* batman version field */
+ uint8_t ttl; /* TTL */
uint8_t vis_type; /* which type of vis-participant sent this? */
- uint8_t entries; /* number of entries behind this struct */
uint32_t seqno; /* sequence number */
- uint8_t ttl; /* TTL */
+ uint8_t entries; /* number of entries behind this struct */
+ uint8_t reserved;
uint8_t vis_orig[6]; /* originator that announces its neighbors */
uint8_t target_orig[6]; /* who should receive this packet */
uint8_t sender_orig[6]; /* who sent or rebroadcasted this packet */
--
1.7.5.3
^ permalink raw reply related
* [PATCH 01/12] batman-adv: Move compare_orig to originator.c
From: Sven Eckelmann @ 2011-06-20 10:16 UTC (permalink / raw)
To: davem; +Cc: netdev, b.a.t.m.a.n, Sven Eckelmann
In-Reply-To: <1308565025-21293-1-git-send-email-sven@narfation.org>
compare_orig is only used in context of orig_node which is managed
inside originator.c. It is not necessary to keep that function inside
the header originator.h.
Signed-off-by: Sven Eckelmann <sven@narfation.org>
---
net/batman-adv/originator.c | 8 ++++++++
net/batman-adv/originator.h | 8 --------
2 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index a6c35d4..991a6e7 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -37,6 +37,14 @@ static void start_purge_timer(struct bat_priv *bat_priv)
queue_delayed_work(bat_event_workqueue, &bat_priv->orig_work, 1 * HZ);
}
+/* returns 1 if they are the same originator */
+static int compare_orig(const struct hlist_node *node, const void *data2)
+{
+ const void *data1 = container_of(node, struct orig_node, hash_entry);
+
+ return (memcmp(data1, data2, ETH_ALEN) == 0 ? 1 : 0);
+}
+
int originator_init(struct bat_priv *bat_priv)
{
if (bat_priv->orig_hash)
diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h
index 8e307af..cfc1f60 100644
--- a/net/batman-adv/originator.h
+++ b/net/batman-adv/originator.h
@@ -40,14 +40,6 @@ int orig_hash_add_if(struct hard_iface *hard_iface, int max_if_num);
int orig_hash_del_if(struct hard_iface *hard_iface, int max_if_num);
-/* returns 1 if they are the same originator */
-static inline int compare_orig(const struct hlist_node *node, const void *data2)
-{
- const void *data1 = container_of(node, struct orig_node, hash_entry);
-
- return (memcmp(data1, data2, ETH_ALEN) == 0 ? 1 : 0);
-}
-
/* hashfunction to choose an entry in a hash table of given size */
/* hash algorithm from http://en.wikipedia.org/wiki/Hash_table */
static inline int choose_orig(const void *data, int32_t size)
--
1.7.5.3
^ permalink raw reply related
* [PATCH 02/12] batman-adv: Keep interface_tx as local function
From: Sven Eckelmann @ 2011-06-20 10:16 UTC (permalink / raw)
To: davem; +Cc: netdev, b.a.t.m.a.n, Sven Eckelmann
In-Reply-To: <1308565025-21293-1-git-send-email-sven@narfation.org>
interface_tx is not used outside of soft-interface.c and thus doesn't
need to be declared inside soft-interface.h
Signed-off-by: Sven Eckelmann <sven@narfation.org>
---
net/batman-adv/soft-interface.c | 2 +-
net/batman-adv/soft-interface.h | 1 -
2 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index b8d3f248..0fc997e 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -553,7 +553,7 @@ static int interface_change_mtu(struct net_device *dev, int new_mtu)
return 0;
}
-int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
+static int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
{
struct ethhdr *ethhdr = (struct ethhdr *)skb->data;
struct bat_priv *bat_priv = netdev_priv(soft_iface);
diff --git a/net/batman-adv/soft-interface.h b/net/batman-adv/soft-interface.h
index c24906d..001546f 100644
--- a/net/batman-adv/soft-interface.h
+++ b/net/batman-adv/soft-interface.h
@@ -25,7 +25,6 @@
int my_skb_head_push(struct sk_buff *skb, unsigned int len);
int softif_neigh_seq_print_text(struct seq_file *seq, void *offset);
void softif_neigh_purge(struct bat_priv *bat_priv);
-int interface_tx(struct sk_buff *skb, struct net_device *soft_iface);
void interface_rx(struct net_device *soft_iface,
struct sk_buff *skb, struct hard_iface *recv_if,
int hdr_size);
--
1.7.5.3
^ permalink raw reply related
* [PATCH 12/12] batman-adv: improved gateway tq-based selection
From: Sven Eckelmann @ 2011-06-20 10:17 UTC (permalink / raw)
To: davem-fT/PcQaiUtIeIZ0/mPfg9Q
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
b.a.t.m.a.n-ZwoEplunGu2X36UT3dwllkB+6BGkLq7r, Marek Lindner
In-Reply-To: <1308565025-21293-1-git-send-email-sven-KaDOiPu9UxWEi8DpZVb4nw@public.gmane.org>
From: Antonio Quartulli <ordex-GaUfNO9RBHfsrOwW+9ziJQ@public.gmane.org>
If a client issues a DHCPREQUEST for renewal, the packet is dropped
if the old destination (the old gateway for the client) TQ is smaller
than the current best gateway TQ less GW_THRESHOLD
Signed-off-by: Antonio Quartulli <ordex-GaUfNO9RBHfsrOwW+9ziJQ@public.gmane.org>
Signed-off-by: Marek Lindner <lindner_marek-LWAfsSFWpa4@public.gmane.org>
Signed-off-by: Sven Eckelmann <sven-KaDOiPu9UxWEi8DpZVb4nw@public.gmane.org>
---
net/batman-adv/gateway_client.c | 94 ++++++++++++++++++++++++++++++++++++++-
net/batman-adv/gateway_client.h | 3 +-
net/batman-adv/main.h | 3 +-
net/batman-adv/soft-interface.c | 10 +++-
4 files changed, 104 insertions(+), 6 deletions(-)
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 6381864..8b25b52 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -25,11 +25,17 @@
#include "gateway_common.h"
#include "hard-interface.h"
#include "originator.h"
+#include "routing.h"
#include <linux/ip.h>
#include <linux/ipv6.h>
#include <linux/udp.h>
#include <linux/if_vlan.h>
+/* This is the offset of the options field in a dhcp packet starting at
+ * the beginning of the dhcp header */
+#define DHCP_OPTIONS_OFFSET 240
+#define DHCP_REQUEST 3
+
static void gw_node_free_ref(struct gw_node *gw_node)
{
if (atomic_dec_and_test(&gw_node->refcount))
@@ -509,14 +515,75 @@ out:
return ret;
}
-int gw_is_target(struct bat_priv *bat_priv, struct sk_buff *skb)
+static bool is_type_dhcprequest(struct sk_buff *skb, int header_len)
+{
+ int ret = false;
+ unsigned char *p;
+ int pkt_len;
+
+ if (skb_linearize(skb) < 0)
+ goto out;
+
+ pkt_len = skb_headlen(skb);
+
+ if (pkt_len < header_len + DHCP_OPTIONS_OFFSET + 1)
+ goto out;
+
+ p = skb->data + header_len + DHCP_OPTIONS_OFFSET;
+ pkt_len -= header_len + DHCP_OPTIONS_OFFSET + 1;
+
+ /* Access the dhcp option lists. Each entry is made up by:
+ * - octect 1: option type
+ * - octect 2: option data len (only if type != 255 and 0)
+ * - octect 3: option data */
+ while (*p != 255 && !ret) {
+ /* p now points to the first octect: option type */
+ if (*p == 53) {
+ /* type 53 is the message type option.
+ * Jump the len octect and go to the data octect */
+ if (pkt_len < 2)
+ goto out;
+ p += 2;
+
+ /* check if the message type is what we need */
+ if (*p == DHCP_REQUEST)
+ ret = true;
+ break;
+ } else if (*p == 0) {
+ /* option type 0 (padding), just go forward */
+ if (pkt_len < 1)
+ goto out;
+ pkt_len--;
+ p++;
+ } else {
+ /* This is any other option. So we get the length... */
+ if (pkt_len < 1)
+ goto out;
+ pkt_len--;
+ p++;
+
+ /* ...and then we jump over the data */
+ if (pkt_len < *p)
+ goto out;
+ pkt_len -= *p;
+ p += (*p);
+ }
+ }
+out:
+ return ret;
+}
+
+int gw_is_target(struct bat_priv *bat_priv, struct sk_buff *skb,
+ struct orig_node *old_gw)
{
struct ethhdr *ethhdr;
struct iphdr *iphdr;
struct ipv6hdr *ipv6hdr;
struct udphdr *udphdr;
struct gw_node *curr_gw;
+ struct neigh_node *neigh_curr = NULL, *neigh_old = NULL;
unsigned int header_len = 0;
+ int ret = 1;
if (atomic_read(&bat_priv->gw_mode) == GW_MODE_OFF)
return 0;
@@ -584,7 +651,30 @@ int gw_is_target(struct bat_priv *bat_priv, struct sk_buff *skb)
if (!curr_gw)
return 0;
+ /* If old_gw != NULL then this packet is unicast.
+ * So, at this point we have to check the message type: if it is a
+ * DHCPREQUEST we have to decide whether to drop it or not */
+ if (old_gw && curr_gw->orig_node != old_gw) {
+ if (is_type_dhcprequest(skb, header_len)) {
+ /* If the dhcp packet has been sent to a different gw,
+ * we have to evaluate whether the old gw is still
+ * reliable enough */
+ neigh_curr = find_router(bat_priv, curr_gw->orig_node,
+ NULL);
+ neigh_old = find_router(bat_priv, old_gw, NULL);
+ if (!neigh_curr || !neigh_old)
+ goto free_neigh;
+ if (neigh_curr->tq_avg - neigh_old->tq_avg <
+ GW_THRESHOLD)
+ ret = -1;
+ }
+ }
+free_neigh:
+ if (neigh_old)
+ neigh_node_free_ref(neigh_old);
+ if (neigh_curr)
+ neigh_node_free_ref(neigh_curr);
if (curr_gw)
gw_node_free_ref(curr_gw);
- return 1;
+ return ret;
}
diff --git a/net/batman-adv/gateway_client.h b/net/batman-adv/gateway_client.h
index 1ce8c60..b9b983c 100644
--- a/net/batman-adv/gateway_client.h
+++ b/net/batman-adv/gateway_client.h
@@ -31,6 +31,7 @@ void gw_node_update(struct bat_priv *bat_priv,
void gw_node_delete(struct bat_priv *bat_priv, struct orig_node *orig_node);
void gw_node_purge(struct bat_priv *bat_priv);
int gw_client_seq_print_text(struct seq_file *seq, void *offset);
-int gw_is_target(struct bat_priv *bat_priv, struct sk_buff *skb);
+int gw_is_target(struct bat_priv *bat_priv, struct sk_buff *skb,
+ struct orig_node *old_gw);
#endif /* _NET_BATMAN_ADV_GATEWAY_CLIENT_H_ */
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index f9e0e17..4f293b5 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -91,7 +91,6 @@ enum mesh_state {
#define BCAST_QUEUE_LEN 256
#define BATMAN_QUEUE_LEN 256
-
enum uev_action {
UEV_ADD = 0,
UEV_DEL,
@@ -102,6 +101,8 @@ enum uev_type {
UEV_GW = 0
};
+#define GW_THRESHOLD 50
+
/*
* Debug Messages
*/
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 3371ece..2dcdbb7 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -30,6 +30,7 @@
#include "gateway_common.h"
#include "gateway_client.h"
#include "bat_sysfs.h"
+#include "originator.h"
#include <linux/slab.h>
#include <linux/ethtool.h>
#include <linux/etherdevice.h>
@@ -561,6 +562,7 @@ static int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
struct bcast_packet *bcast_packet;
struct vlan_ethhdr *vhdr;
struct softif_neigh *curr_softif_neigh = NULL;
+ struct orig_node *orig_node = NULL;
int data_len = skb->len, ret;
short vid = -1;
bool do_bcast = false;
@@ -595,8 +597,10 @@ static int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
/* Register the client MAC in the transtable */
tt_local_add(soft_iface, ethhdr->h_source);
- if (is_multicast_ether_addr(ethhdr->h_dest)) {
- ret = gw_is_target(bat_priv, skb);
+ orig_node = transtable_search(bat_priv, ethhdr->h_dest);
+ if (is_multicast_ether_addr(ethhdr->h_dest) ||
+ (orig_node && orig_node->gw_flags)) {
+ ret = gw_is_target(bat_priv, skb, orig_node);
if (ret < 0)
goto dropped;
@@ -656,6 +660,8 @@ end:
softif_neigh_free_ref(curr_softif_neigh);
if (primary_if)
hardif_free_ref(primary_if);
+ if (orig_node)
+ orig_node_free_ref(orig_node);
return NETDEV_TX_OK;
}
--
1.7.5.3
^ permalink raw reply related
* [PATCH 07/12] batman-adv: improved roaming mechanism
From: Sven Eckelmann @ 2011-06-20 10:17 UTC (permalink / raw)
To: davem-fT/PcQaiUtIeIZ0/mPfg9Q
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
b.a.t.m.a.n-ZwoEplunGu2X36UT3dwllkB+6BGkLq7r
In-Reply-To: <1308565025-21293-1-git-send-email-sven-KaDOiPu9UxWEi8DpZVb4nw@public.gmane.org>
From: Antonio Quartulli <ordex-GaUfNO9RBHfsrOwW+9ziJQ@public.gmane.org>
With the current client announcement implementation, in case of roaming,
an update is triggered on the new AP serving the client. At that point
the new information is spread around by means of the OGM broadcasting
mechanism. Until this operations is not executed, no node is able to
correctly route traffic towards the client. This obviously causes packet
drops and introduces a delay in the time needed by the client to recover
its connections.
A new packet type called ROAMING_ADVERTISEMENT is added to account this
issue.
This message is sent in case of roaming from the new AP serving the
client to the old one and will contain the client MAC address. In this
way an out-of-OGM update is immediately committed, so that the old node
can update its global translation table. Traffic reaching this node will
then be redirected to the correct destination utilising the fresher
information. Thus reducing the packet drops and the connection recovery
delay.
Signed-off-by: Antonio Quartulli <ordex-GaUfNO9RBHfsrOwW+9ziJQ@public.gmane.org>
Signed-off-by: Sven Eckelmann <sven-KaDOiPu9UxWEi8DpZVb4nw@public.gmane.org>
---
net/batman-adv/hard-interface.c | 4 +
net/batman-adv/main.c | 2 +
net/batman-adv/main.h | 6 +-
net/batman-adv/originator.c | 1 +
net/batman-adv/packet.h | 13 ++-
net/batman-adv/routing.c | 61 +++++++++-
net/batman-adv/routing.h | 1 +
net/batman-adv/send.c | 1 +
net/batman-adv/soft-interface.c | 3 +-
net/batman-adv/translation-table.c | 241 ++++++++++++++++++++++++++++++++---
net/batman-adv/translation-table.h | 8 +-
net/batman-adv/types.h | 26 ++++-
12 files changed, 334 insertions(+), 33 deletions(-)
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index d40426c..55b5def 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -658,6 +658,10 @@ static int batman_skb_recv(struct sk_buff *skb, struct net_device *dev,
case BAT_TT_QUERY:
ret = recv_tt_query(skb, hard_iface);
break;
+ /* Roaming advertisement */
+ case BAT_ROAM_ADV:
+ ret = recv_roam_adv(skb, hard_iface);
+ break;
default:
ret = NET_RX_DROP;
}
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 49a5e64..3318ee2 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -88,6 +88,7 @@ int mesh_init(struct net_device *soft_iface)
spin_lock_init(&bat_priv->tt_ghash_lock);
spin_lock_init(&bat_priv->tt_changes_list_lock);
spin_lock_init(&bat_priv->tt_req_list_lock);
+ spin_lock_init(&bat_priv->tt_roam_list_lock);
spin_lock_init(&bat_priv->tt_buff_lock);
spin_lock_init(&bat_priv->gw_list_lock);
spin_lock_init(&bat_priv->vis_hash_lock);
@@ -101,6 +102,7 @@ int mesh_init(struct net_device *soft_iface)
INIT_HLIST_HEAD(&bat_priv->softif_neigh_vids);
INIT_LIST_HEAD(&bat_priv->tt_changes_list);
INIT_LIST_HEAD(&bat_priv->tt_req_list);
+ INIT_LIST_HEAD(&bat_priv->tt_roam_list);
if (originator_init(bat_priv) < 1)
goto err;
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 6f53a1d..8eae05e 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -42,7 +42,7 @@
* -> TODO: check influence on TQ_LOCAL_WINDOW_SIZE */
#define PURGE_TIMEOUT 200
#define TT_LOCAL_TIMEOUT 3600 /* in seconds */
-
+#define TT_CLIENT_ROAM_TIMEOUT 600
/* sliding packet range of received originator messages in squence numbers
* (should be a multiple of our word size) */
#define TQ_LOCAL_WINDOW_SIZE 64
@@ -55,6 +55,10 @@
#define TT_OGM_APPEND_MAX 3 /* number of OGMs sent with the last tt diff */
+#define ROAMING_MAX_TIME 20 /* Time in which a client can roam at most
+ * ROAMING_MAX_COUNT times */
+#define ROAMING_MAX_COUNT 5
+
#define NO_FLAGS 0
#define NUM_WORDS (TQ_LOCAL_WINDOW_SIZE / WORD_BIT_SIZE)
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 25e7e50..338b3c5 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -219,6 +219,7 @@ struct orig_node *get_orig_node(struct bat_priv *bat_priv, const uint8_t *addr)
/* extra reference for return */
atomic_set(&orig_node->refcount, 2);
+ orig_node->tt_poss_change = false;
orig_node->bat_priv = bat_priv;
memcpy(orig_node->orig, addr, ETH_ALEN);
orig_node->router = NULL;
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index 407dd2e..c5f081d 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -31,7 +31,8 @@ enum bat_packettype {
BAT_BCAST = 0x04,
BAT_VIS = 0x05,
BAT_UNICAST_FRAG = 0x06,
- BAT_TT_QUERY = 0x07
+ BAT_TT_QUERY = 0x07,
+ BAT_ROAM_ADV = 0x08
};
/* this file is included by batctl which needs these defines */
@@ -194,6 +195,16 @@ struct tt_query_packet {
uint16_t tt_data;
} __packed;
+struct roam_adv_packet {
+ uint8_t packet_type;
+ uint8_t version;
+ uint8_t ttl;
+ uint8_t reserved;
+ uint8_t dst[ETH_ALEN];
+ uint8_t src[ETH_ALEN];
+ uint8_t client[ETH_ALEN];
+} __packed;
+
struct tt_change {
uint8_t flags;
uint8_t addr[ETH_ALEN];
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 8b0f833..05d50ca 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -93,6 +93,9 @@ static void update_transtable(struct bat_priv *bat_priv,
spin_lock_bh(&bat_priv->tt_ghash_lock);
orig_node->tt_crc = tt_global_crc(bat_priv, orig_node);
spin_unlock_bh(&bat_priv->tt_ghash_lock);
+ /* Roaming phase is over: tables are in sync again. I can
+ * unset the flag */
+ orig_node->tt_poss_change = false;
} else {
/* if we missed more than one change or our tables are not
* in sync anymore -> request fresh tt data */
@@ -1252,6 +1255,54 @@ out:
return NET_RX_DROP;
}
+int recv_roam_adv(struct sk_buff *skb, struct hard_iface *recv_if)
+{
+ struct bat_priv *bat_priv = netdev_priv(recv_if->soft_iface);
+ struct roam_adv_packet *roam_adv_packet;
+ struct orig_node *orig_node;
+ struct ethhdr *ethhdr;
+
+ /* drop packet if it has not necessary minimum size */
+ if (unlikely(!pskb_may_pull(skb, sizeof(struct roam_adv_packet))))
+ goto out;
+
+ ethhdr = (struct ethhdr *)skb_mac_header(skb);
+
+ /* packet with unicast indication but broadcast recipient */
+ if (is_broadcast_ether_addr(ethhdr->h_dest))
+ goto out;
+
+ /* packet with broadcast sender address */
+ if (is_broadcast_ether_addr(ethhdr->h_source))
+ goto out;
+
+ roam_adv_packet = (struct roam_adv_packet *)skb->data;
+
+ if (!is_my_mac(roam_adv_packet->dst))
+ return route_unicast_packet(skb, recv_if);
+
+ orig_node = orig_hash_find(bat_priv, roam_adv_packet->src);
+ if (!orig_node)
+ goto out;
+
+ bat_dbg(DBG_TT, bat_priv, "Received ROAMING_ADV from %pM "
+ "(client %pM)\n", roam_adv_packet->src,
+ roam_adv_packet->client);
+
+ tt_global_add(bat_priv, orig_node, roam_adv_packet->client,
+ atomic_read(&orig_node->last_ttvn) + 1, true);
+
+ /* Roaming phase starts: I have new information but the ttvn has not
+ * been incremented yet. This flag will make me check all the incoming
+ * packets for the correct destination. */
+ bat_priv->tt_poss_change = true;
+
+ orig_node_free_ref(orig_node);
+out:
+ /* returning NET_RX_DROP will make the caller function kfree the skb */
+ return NET_RX_DROP;
+}
+
/* find a suitable router for this originator, and use
* bonding if possible. increases the found neighbors
* refcount.*/
@@ -1445,6 +1496,7 @@ static int check_unicast_ttvn(struct bat_priv *bat_priv,
struct ethhdr *ethhdr;
struct hard_iface *primary_if;
struct unicast_packet *unicast_packet;
+ bool tt_poss_change;
/* I could need to modify it */
if (skb_cow(skb, sizeof(struct unicast_packet)) < 0)
@@ -1452,27 +1504,28 @@ static int check_unicast_ttvn(struct bat_priv *bat_priv,
unicast_packet = (struct unicast_packet *)skb->data;
- if (is_my_mac(unicast_packet->dest))
+ if (is_my_mac(unicast_packet->dest)) {
+ tt_poss_change = bat_priv->tt_poss_change;
curr_ttvn = (uint8_t)atomic_read(&bat_priv->ttvn);
- else {
+ } else {
orig_node = orig_hash_find(bat_priv, unicast_packet->dest);
if (!orig_node)
return 0;
curr_ttvn = (uint8_t)atomic_read(&orig_node->last_ttvn);
+ tt_poss_change = orig_node->tt_poss_change;
orig_node_free_ref(orig_node);
}
/* Check whether I have to reroute the packet */
- if (seq_before(unicast_packet->ttvn, curr_ttvn)) {
+ if (seq_before(unicast_packet->ttvn, curr_ttvn) || tt_poss_change) {
/* Linearize the skb before accessing it */
if (skb_linearize(skb) < 0)
return 0;
ethhdr = (struct ethhdr *)(skb->data +
sizeof(struct unicast_packet));
-
orig_node = transtable_search(bat_priv, ethhdr->h_dest);
if (!orig_node) {
diff --git a/net/batman-adv/routing.h b/net/batman-adv/routing.h
index e77d464..fb14e95 100644
--- a/net/batman-adv/routing.h
+++ b/net/batman-adv/routing.h
@@ -37,6 +37,7 @@ int recv_bcast_packet(struct sk_buff *skb, struct hard_iface *recv_if);
int recv_vis_packet(struct sk_buff *skb, struct hard_iface *recv_if);
int recv_bat_packet(struct sk_buff *skb, struct hard_iface *recv_if);
int recv_tt_query(struct sk_buff *skb, struct hard_iface *recv_if);
+int recv_roam_adv(struct sk_buff *skb, struct hard_iface *recv_if);
struct neigh_node *find_router(struct bat_priv *bat_priv,
struct orig_node *orig_node,
const struct hard_iface *recv_if);
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index 6b14075..7a2f082 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -303,6 +303,7 @@ void schedule_own_packet(struct hard_iface *hard_iface)
prepare_packet_buffer(bat_priv, hard_iface);
/* Increment the TTVN only once per OGM interval */
atomic_inc(&bat_priv->ttvn);
+ bat_priv->tt_poss_change = false;
}
/* if the changes have been sent enough times */
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index c288d93..3371ece 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -534,7 +534,7 @@ static int interface_set_mac_addr(struct net_device *dev, void *p)
/* only modify transtable if it has been initialised before */
if (atomic_read(&bat_priv->mesh_state) == MESH_ACTIVE) {
tt_local_remove(bat_priv, dev->dev_addr,
- "mac address changed");
+ "mac address changed", false);
tt_local_add(dev, addr->sa_data);
}
@@ -836,6 +836,7 @@ struct net_device *softif_create(const char *name)
bat_priv->tt_buff = NULL;
bat_priv->tt_buff_len = 0;
+ bat_priv->tt_poss_change = false;
bat_priv->primary_if = NULL;
bat_priv->num_ifaces = 0;
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 597cd1a..d516d85 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -126,7 +126,7 @@ static bool is_out_of_time(unsigned long starting_time, unsigned long timeout)
}
static void tt_local_event(struct bat_priv *bat_priv, uint8_t op,
- const uint8_t *addr)
+ const uint8_t *addr, uint8_t roaming)
{
struct tt_change_node *tt_change_node;
@@ -136,6 +136,9 @@ static void tt_local_event(struct bat_priv *bat_priv, uint8_t op,
return;
tt_change_node->change.flags = op;
+ if (roaming)
+ tt_change_node->change.flags |= TT_CLIENT_ROAM;
+
memcpy(tt_change_node->change.addr, addr, ETH_ALEN);
spin_lock_bh(&bat_priv->tt_changes_list_lock);
@@ -170,6 +173,7 @@ void tt_local_add(struct net_device *soft_iface, const uint8_t *addr)
struct bat_priv *bat_priv = netdev_priv(soft_iface);
struct tt_local_entry *tt_local_entry;
struct tt_global_entry *tt_global_entry;
+ uint8_t roam_addr[ETH_ALEN];
spin_lock_bh(&bat_priv->tt_lhash_lock);
tt_local_entry = tt_local_hash_find(bat_priv, addr);
@@ -183,7 +187,7 @@ void tt_local_add(struct net_device *soft_iface, const uint8_t *addr)
if (!tt_local_entry)
goto unlock;
- tt_local_event(bat_priv, NO_FLAGS, addr);
+ tt_local_event(bat_priv, NO_FLAGS, addr, false);
bat_dbg(DBG_TT, bat_priv,
"Creating new local tt entry: %pM (ttvn: %d)\n", addr,
@@ -208,11 +212,19 @@ void tt_local_add(struct net_device *soft_iface, const uint8_t *addr)
tt_global_entry = tt_global_hash_find(bat_priv, addr);
- if (tt_global_entry)
+ /* Check whether it is a roaming! */
+ if (tt_global_entry) {
+ memcpy(roam_addr, tt_global_entry->addr, ETH_ALEN);
+ /* This node is probably going to update its tt table */
+ tt_global_entry->orig_node->tt_poss_change = true;
_tt_global_del(bat_priv, tt_global_entry,
"local tt received");
+ spin_unlock_bh(&bat_priv->tt_ghash_lock);
+ send_roam_adv(bat_priv, tt_global_entry->addr,
+ tt_global_entry->orig_node);
+ } else
+ spin_unlock_bh(&bat_priv->tt_ghash_lock);
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
return;
unlock:
spin_unlock_bh(&bat_priv->tt_lhash_lock);
@@ -367,7 +379,7 @@ static void tt_local_del(struct bat_priv *bat_priv,
}
void tt_local_remove(struct bat_priv *bat_priv, const uint8_t *addr,
- const char *message)
+ const char *message, bool roaming)
{
struct tt_local_entry *tt_local_entry;
@@ -375,7 +387,8 @@ void tt_local_remove(struct bat_priv *bat_priv, const uint8_t *addr,
tt_local_entry = tt_local_hash_find(bat_priv, addr);
if (tt_local_entry) {
- tt_local_event(bat_priv, TT_CHANGE_DEL, tt_local_entry->addr);
+ tt_local_event(bat_priv, TT_CHANGE_DEL, tt_local_entry->addr,
+ roaming);
tt_local_del(bat_priv, tt_local_entry, message);
}
spin_unlock_bh(&bat_priv->tt_lhash_lock);
@@ -404,7 +417,7 @@ static void tt_local_purge(struct bat_priv *bat_priv)
continue;
tt_local_event(bat_priv, TT_CHANGE_DEL,
- tt_local_entry->addr);
+ tt_local_entry->addr, false);
tt_local_del(bat_priv, tt_local_entry,
"address timed out");
}
@@ -476,7 +489,7 @@ static void tt_changes_list_free(struct bat_priv *bat_priv)
/* caller must hold orig_node refcount */
int tt_global_add(struct bat_priv *bat_priv, struct orig_node *orig_node,
- const unsigned char *tt_addr, uint8_t ttvn)
+ const unsigned char *tt_addr, uint8_t ttvn, bool roaming)
{
struct tt_global_entry *tt_global_entry;
struct tt_local_entry *tt_local_entry;
@@ -496,6 +509,8 @@ int tt_global_add(struct bat_priv *bat_priv, struct orig_node *orig_node,
atomic_inc(&orig_node->refcount);
tt_global_entry->orig_node = orig_node;
tt_global_entry->ttvn = ttvn;
+ tt_global_entry->flags = NO_FLAGS;
+ tt_global_entry->roam_at = 0;
atomic_inc(&orig_node->tt_size);
hash_add(bat_priv->tt_global_hash, compare_gtt,
choose_orig, tt_global_entry,
@@ -506,10 +521,12 @@ int tt_global_add(struct bat_priv *bat_priv, struct orig_node *orig_node,
orig_node_tmp = tt_global_entry->orig_node;
atomic_inc(&orig_node->refcount);
tt_global_entry->orig_node = orig_node;
- tt_global_entry->ttvn = ttvn;
orig_node_free_ref(orig_node_tmp);
atomic_inc(&orig_node->tt_size);
}
+ tt_global_entry->ttvn = ttvn;
+ tt_global_entry->flags = NO_FLAGS;
+ tt_global_entry->roam_at = 0;
}
spin_unlock_bh(&bat_priv->tt_ghash_lock);
@@ -523,8 +540,9 @@ int tt_global_add(struct bat_priv *bat_priv, struct orig_node *orig_node,
tt_local_entry = tt_local_hash_find(bat_priv, tt_addr);
if (tt_local_entry)
- tt_local_del(bat_priv, tt_local_entry,
- "global tt received");
+ tt_local_remove(bat_priv, tt_global_entry->addr,
+ "global tt received", roaming);
+
spin_unlock_bh(&bat_priv->tt_lhash_lock);
return 1;
unlock:
@@ -637,7 +655,7 @@ static void _tt_global_del(struct bat_priv *bat_priv,
void tt_global_del(struct bat_priv *bat_priv,
struct orig_node *orig_node, const unsigned char *addr,
- const char *message)
+ const char *message, bool roaming)
{
struct tt_global_entry *tt_global_entry;
@@ -645,9 +663,15 @@ void tt_global_del(struct bat_priv *bat_priv,
tt_global_entry = tt_global_hash_find(bat_priv, addr);
if (tt_global_entry && tt_global_entry->orig_node == orig_node) {
+ if (roaming) {
+ tt_global_entry->flags |= TT_CLIENT_ROAM;
+ tt_global_entry->roam_at = jiffies;
+ goto out;
+ }
atomic_dec(&orig_node->tt_size);
_tt_global_del(bat_priv, tt_global_entry, message);
}
+out:
spin_unlock_bh(&bat_priv->tt_ghash_lock);
}
@@ -685,6 +709,35 @@ static void tt_global_entry_free(struct hlist_node *node, void *arg)
kfree(data);
}
+static void tt_global_roam_purge(struct bat_priv *bat_priv)
+{
+ struct hashtable_t *hash = bat_priv->tt_global_hash;
+ struct tt_global_entry *tt_global_entry;
+ struct hlist_node *node, *node_tmp;
+ struct hlist_head *head;
+ int i;
+
+ spin_lock_bh(&bat_priv->tt_ghash_lock);
+
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+
+ hlist_for_each_entry_safe(tt_global_entry, node, node_tmp,
+ head, hash_entry) {
+ if (!(tt_global_entry->flags & TT_CLIENT_ROAM))
+ continue;
+ if (!is_out_of_time(tt_global_entry->roam_at,
+ TT_CLIENT_ROAM_TIMEOUT * 1000))
+ continue;
+
+ _tt_global_del(bat_priv, tt_global_entry,
+ "Roaming timeout");
+ }
+ }
+
+ spin_unlock_bh(&bat_priv->tt_ghash_lock);
+}
+
static void tt_global_table_free(struct bat_priv *bat_priv)
{
if (!bat_priv->tt_global_hash)
@@ -734,6 +787,12 @@ uint16_t tt_global_crc(struct bat_priv *bat_priv, struct orig_node *orig_node)
head, hash_entry) {
if (compare_eth(tt_global_entry->orig_node,
orig_node)) {
+ /* Roaming clients are in the global table for
+ * consistency only. They don't have to be
+ * taken into account while computing the
+ * global crc */
+ if (tt_global_entry->flags & TT_CLIENT_ROAM)
+ continue;
total_one = 0;
for (j = 0; j < ETH_ALEN; j++)
total_one = crc16_byte(total_one,
@@ -858,6 +917,9 @@ static int tt_global_valid_entry(const void *entry_ptr, const void *data_ptr)
const struct tt_global_entry *tt_global_entry = entry_ptr;
const struct orig_node *orig_node = data_ptr;
+ if (tt_global_entry->flags & TT_CLIENT_ROAM)
+ return 0;
+
return (tt_global_entry->orig_node == orig_node);
}
@@ -1251,10 +1313,11 @@ static void _tt_update_changes(struct bat_priv *bat_priv,
if ((tt_change + i)->flags & TT_CHANGE_DEL)
tt_global_del(bat_priv, orig_node,
(tt_change + i)->addr,
- "tt removed by changes");
+ "tt removed by changes",
+ (tt_change + i)->flags & TT_CLIENT_ROAM);
else
if (!tt_global_add(bat_priv, orig_node,
- (tt_change + i)->addr, ttvn))
+ (tt_change + i)->addr, ttvn, false))
/* In case of problem while storing a
* global_entry, we stop the updating
* procedure without committing the
@@ -1356,6 +1419,9 @@ void handle_tt_response(struct bat_priv *bat_priv,
spin_lock_bh(&bat_priv->tt_ghash_lock);
orig_node->tt_crc = tt_global_crc(bat_priv, orig_node);
spin_unlock_bh(&bat_priv->tt_ghash_lock);
+ /* Roaming phase is over: tables are in sync again. I can
+ * unset the flag */
+ orig_node->tt_poss_change = false;
out:
if (orig_node)
orig_node_free_ref(orig_node);
@@ -1374,16 +1440,134 @@ int tt_init(struct bat_priv *bat_priv)
return 1;
}
-void tt_free(struct bat_priv *bat_priv)
+static void tt_roam_list_free(struct bat_priv *bat_priv)
{
- cancel_delayed_work_sync(&bat_priv->tt_work);
+ struct tt_roam_node *node, *safe;
- tt_local_table_free(bat_priv);
- tt_global_table_free(bat_priv);
- tt_req_list_free(bat_priv);
- tt_changes_list_free(bat_priv);
+ spin_lock_bh(&bat_priv->tt_roam_list_lock);
- kfree(bat_priv->tt_buff);
+ list_for_each_entry_safe(node, safe, &bat_priv->tt_roam_list, list) {
+ list_del(&node->list);
+ kfree(node);
+ }
+
+ spin_unlock_bh(&bat_priv->tt_roam_list_lock);
+}
+
+static void tt_roam_purge(struct bat_priv *bat_priv)
+{
+ struct tt_roam_node *node, *safe;
+
+ spin_lock_bh(&bat_priv->tt_roam_list_lock);
+ list_for_each_entry_safe(node, safe, &bat_priv->tt_roam_list, list) {
+ if (!is_out_of_time(node->first_time,
+ ROAMING_MAX_TIME * 1000))
+ continue;
+
+ list_del(&node->list);
+ kfree(node);
+ }
+ spin_unlock_bh(&bat_priv->tt_roam_list_lock);
+}
+
+/* This function checks whether the client already reached the
+ * maximum number of possible roaming phases. In this case the ROAMING_ADV
+ * will not be sent.
+ *
+ * returns true if the ROAMING_ADV can be sent, false otherwise */
+static bool tt_check_roam_count(struct bat_priv *bat_priv,
+ uint8_t *client)
+{
+ struct tt_roam_node *tt_roam_node;
+ bool ret = false;
+
+ spin_lock_bh(&bat_priv->tt_roam_list_lock);
+ /* The new tt_req will be issued only if I'm not waiting for a
+ * reply from the same orig_node yet */
+ list_for_each_entry(tt_roam_node, &bat_priv->tt_roam_list, list) {
+ if (!compare_eth(tt_roam_node->addr, client))
+ continue;
+
+ if (is_out_of_time(tt_roam_node->first_time,
+ ROAMING_MAX_TIME * 1000))
+ continue;
+
+ if (!atomic_dec_not_zero(&tt_roam_node->counter))
+ /* Sorry, you roamed too many times! */
+ goto unlock;
+ ret = true;
+ break;
+ }
+
+ if (!ret) {
+ tt_roam_node = kmalloc(sizeof(*tt_roam_node), GFP_ATOMIC);
+ if (!tt_roam_node)
+ goto unlock;
+
+ tt_roam_node->first_time = jiffies;
+ atomic_set(&tt_roam_node->counter, ROAMING_MAX_COUNT - 1);
+ memcpy(tt_roam_node->addr, client, ETH_ALEN);
+
+ list_add(&tt_roam_node->list, &bat_priv->tt_roam_list);
+ ret = true;
+ }
+
+unlock:
+ spin_unlock_bh(&bat_priv->tt_roam_list_lock);
+ return ret;
+}
+
+void send_roam_adv(struct bat_priv *bat_priv, uint8_t *client,
+ struct orig_node *orig_node)
+{
+ struct neigh_node *neigh_node = NULL;
+ struct sk_buff *skb = NULL;
+ struct roam_adv_packet *roam_adv_packet;
+ int ret = 1;
+ struct hard_iface *primary_if;
+
+ /* before going on we have to check whether the client has
+ * already roamed to us too many times */
+ if (!tt_check_roam_count(bat_priv, client))
+ goto out;
+
+ skb = dev_alloc_skb(sizeof(struct roam_adv_packet) + ETH_HLEN);
+ if (!skb)
+ goto out;
+
+ skb_reserve(skb, ETH_HLEN);
+
+ roam_adv_packet = (struct roam_adv_packet *)skb_put(skb,
+ sizeof(struct roam_adv_packet));
+
+ roam_adv_packet->packet_type = BAT_ROAM_ADV;
+ roam_adv_packet->version = COMPAT_VERSION;
+ roam_adv_packet->ttl = TTL;
+ primary_if = primary_if_get_selected(bat_priv);
+ if (!primary_if)
+ goto out;
+ memcpy(roam_adv_packet->src, primary_if->net_dev->dev_addr, ETH_ALEN);
+ hardif_free_ref(primary_if);
+ memcpy(roam_adv_packet->dst, orig_node->orig, ETH_ALEN);
+ memcpy(roam_adv_packet->client, client, ETH_ALEN);
+
+ neigh_node = orig_node_get_router(orig_node);
+ if (!neigh_node)
+ goto out;
+
+ bat_dbg(DBG_TT, bat_priv,
+ "Sending ROAMING_ADV to %pM (client %pM) via %pM\n",
+ orig_node->orig, client, neigh_node->addr);
+
+ send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr);
+ ret = 0;
+
+out:
+ if (neigh_node)
+ neigh_node_free_ref(neigh_node);
+ if (ret)
+ kfree_skb(skb);
+ return;
}
static void tt_purge(struct work_struct *work)
@@ -1394,7 +1578,22 @@ static void tt_purge(struct work_struct *work)
container_of(delayed_work, struct bat_priv, tt_work);
tt_local_purge(bat_priv);
+ tt_global_roam_purge(bat_priv);
tt_req_purge(bat_priv);
+ tt_roam_purge(bat_priv);
tt_start_timer(bat_priv);
}
+
+void tt_free(struct bat_priv *bat_priv)
+{
+ cancel_delayed_work_sync(&bat_priv->tt_work);
+
+ tt_local_table_free(bat_priv);
+ tt_global_table_free(bat_priv);
+ tt_req_list_free(bat_priv);
+ tt_changes_list_free(bat_priv);
+ tt_roam_list_free(bat_priv);
+
+ kfree(bat_priv->tt_buff);
+}
diff --git a/net/batman-adv/translation-table.h b/net/batman-adv/translation-table.h
index 51f7e30..1cd2d39 100644
--- a/net/batman-adv/translation-table.h
+++ b/net/batman-adv/translation-table.h
@@ -28,20 +28,20 @@ int tt_changes_fill_buffer(struct bat_priv *bat_priv,
int tt_init(struct bat_priv *bat_priv);
void tt_local_add(struct net_device *soft_iface, const uint8_t *addr);
void tt_local_remove(struct bat_priv *bat_priv,
- const uint8_t *addr, const char *message);
+ const uint8_t *addr, const char *message, bool roaming);
int tt_local_seq_print_text(struct seq_file *seq, void *offset);
void tt_global_add_orig(struct bat_priv *bat_priv,
struct orig_node *orig_node,
const unsigned char *tt_buff, int tt_buff_len);
int tt_global_add(struct bat_priv *bat_priv,
struct orig_node *orig_node, const unsigned char *addr,
- uint8_t ttvn);
+ uint8_t ttvn, bool roaming);
int tt_global_seq_print_text(struct seq_file *seq, void *offset);
void tt_global_del_orig(struct bat_priv *bat_priv,
struct orig_node *orig_node, const char *message);
void tt_global_del(struct bat_priv *bat_priv,
struct orig_node *orig_node, const unsigned char *addr,
- const char *message);
+ const char *message, bool roaming);
struct orig_node *transtable_search(struct bat_priv *bat_priv,
const uint8_t *addr);
void tt_save_orig_buffer(struct bat_priv *bat_priv, struct orig_node *orig_node,
@@ -60,5 +60,7 @@ void tt_update_changes(struct bat_priv *bat_priv, struct orig_node *orig_node,
bool is_my_client(struct bat_priv *bat_priv, const uint8_t *addr);
void handle_tt_response(struct bat_priv *bat_priv,
struct tt_query_packet *tt_response);
+void send_roam_adv(struct bat_priv *bat_priv, uint8_t *client,
+ struct orig_node *orig_node);
#endif /* _NET_BATMAN_ADV_TRANSLATION_TABLE_H_ */
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 3b642a9..9c84fa9 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -81,6 +81,12 @@ struct orig_node {
int16_t tt_buff_len;
spinlock_t tt_buff_lock; /* protects tt_buff */
atomic_t tt_size;
+ /* The tt_poss_change flag is used to detect an ongoing roaming phase.
+ * If true, then I sent a Roaming_adv to this orig_node and I have to
+ * inspect every packet directed to it to check whether it is still
+ * the true destination or not. This flag will be reset to false as
+ * soon as I receive a new TTVN from this orig_node */
+ bool tt_poss_change;
uint32_t last_real_seqno;
uint8_t last_ttl;
unsigned long bcast_bits[NUM_WORDS];
@@ -153,6 +159,12 @@ struct bat_priv {
atomic_t ttvn; /* tranlation table version number */
atomic_t tt_ogm_append_cnt;
atomic_t tt_local_changes; /* changes registered in a OGM interval */
+ /* The tt_poss_change flag is used to detect an ongoing roaming phase.
+ * If true, then I received a Roaming_adv and I have to inspect every
+ * packet directed to me to check whether I am still the true
+ * destination or not. This flag will be reset to false as soon as I
+ * increase my TTVN */
+ bool tt_poss_change;
char num_ifaces;
struct debug_log *debug_log;
struct kobject *mesh_obj;
@@ -167,6 +179,7 @@ struct bat_priv {
struct hashtable_t *tt_local_hash;
struct hashtable_t *tt_global_hash;
struct list_head tt_req_list; /* list of pending tt_requests */
+ struct list_head tt_roam_list;
struct hashtable_t *vis_hash;
spinlock_t forw_bat_list_lock; /* protects forw_bat_list */
spinlock_t forw_bcast_list_lock; /* protects */
@@ -174,6 +187,7 @@ struct bat_priv {
spinlock_t tt_lhash_lock; /* protects tt_local_hash */
spinlock_t tt_ghash_lock; /* protects tt_global_hash */
spinlock_t tt_req_list_lock; /* protects tt_req_list */
+ spinlock_t tt_roam_list_lock; /* protects tt_roam_list */
spinlock_t gw_list_lock; /* protects gw_list and curr_gw */
spinlock_t vis_hash_lock; /* protects vis_hash */
spinlock_t vis_list_lock; /* protects vis_info::recv_list */
@@ -219,8 +233,9 @@ struct tt_global_entry {
uint8_t addr[ETH_ALEN];
struct orig_node *orig_node;
uint8_t ttvn;
- /* entry in the global table */
- struct hlist_node hash_entry;
+ uint8_t flags; /* only TT_GLOBAL_ROAM is used */
+ unsigned long roam_at; /* time at which TT_GLOBAL_ROAM was set */
+ struct hlist_node hash_entry; /* entry in the global table */
};
struct tt_change_node {
@@ -234,6 +249,13 @@ struct tt_req_node {
struct list_head list;
};
+struct tt_roam_node {
+ uint8_t addr[ETH_ALEN];
+ atomic_t counter;
+ unsigned long first_time;
+ struct list_head list;
+};
+
/**
* forw_packet - structure for forw_list maintaining packets to be
* send/forwarded
--
1.7.5.3
^ permalink raw reply related
* [PATCH 06/12] batman-adv: improved client announcement mechanism
From: Sven Eckelmann @ 2011-06-20 10:16 UTC (permalink / raw)
To: davem-fT/PcQaiUtIeIZ0/mPfg9Q
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
b.a.t.m.a.n-ZwoEplunGu2X36UT3dwllkB+6BGkLq7r, Marek Lindner
In-Reply-To: <1308565025-21293-1-git-send-email-sven-KaDOiPu9UxWEi8DpZVb4nw@public.gmane.org>
From: Antonio Quartulli <ordex-GaUfNO9RBHfsrOwW+9ziJQ@public.gmane.org>
The client announcement mechanism informs every mesh node in the network
of any connected non-mesh client, in order to find the path towards that
client from any given point in the mesh.
The old implementation was based on the simple idea of appending a data
buffer to each OGM containing all the client MAC addresses the node is
serving. All other nodes can populate their global translation tables
(table which links client MAC addresses to node addresses) using this
MAC address buffer and linking it to the node's address contained in the
OGM. A node that wants to contact a client has to lookup the node the
client is connected to and its address in the global translation table.
It is easy to understand that this implementation suffers from several
issues:
- big overhead (each and every OGM contains the entire list of
connected clients)
- high latencies for client route updates due to long OGM trip time and
OGM losses
The new implementation addresses these issues by appending client
changes (new client joined or a client left) to the OGM instead of
filling it with all the client addresses each time. In this way nodes
can modify their global tables by means of "updates", thus reducing the
overhead within the OGMs.
To keep the entire network in sync each node maintains a translation
table version number (ttvn) and a translation table checksum. These
values are spread with the OGM to allow all the network participants to
determine whether or not they need to update their translation table
information.
When a translation table lookup is performed in order to send a packet
to a client attached to another node, the destination's ttvn is added to
the payload packet. Forwarding nodes can compare the packet's ttvn with
their destination's ttvn (this node could have a fresher information
than the source) and re-route the packet if necessary. This greatly
reduces the packet loss of clients roaming from one AP to the next.
Signed-off-by: Antonio Quartulli <ordex-GaUfNO9RBHfsrOwW+9ziJQ@public.gmane.org>
Signed-off-by: Marek Lindner <lindner_marek-LWAfsSFWpa4@public.gmane.org>
Signed-off-by: Sven Eckelmann <sven-KaDOiPu9UxWEi8DpZVb4nw@public.gmane.org>
---
net/batman-adv/Kconfig | 1 +
net/batman-adv/aggregation.c | 23 +-
net/batman-adv/aggregation.h | 6 +-
net/batman-adv/bat_sysfs.c | 2 +-
net/batman-adv/hard-interface.c | 13 +-
net/batman-adv/main.c | 13 +-
net/batman-adv/main.h | 7 +-
net/batman-adv/originator.c | 8 +-
net/batman-adv/packet.h | 59 ++-
net/batman-adv/routing.c | 248 +++++++--
net/batman-adv/routing.h | 6 +-
net/batman-adv/send.c | 87 +++-
net/batman-adv/send.h | 2 +-
net/batman-adv/soft-interface.c | 11 +-
net/batman-adv/translation-table.c | 1134 ++++++++++++++++++++++++++++++------
net/batman-adv/translation-table.h | 34 +-
net/batman-adv/types.h | 33 +-
net/batman-adv/unicast.c | 3 +
18 files changed, 1383 insertions(+), 307 deletions(-)
diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index 6c051ad..2b68d06 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -5,6 +5,7 @@
config BATMAN_ADV
tristate "B.A.T.M.A.N. Advanced Meshing Protocol"
depends on NET
+ select CRC16
default n
---help---
diff --git a/net/batman-adv/aggregation.c b/net/batman-adv/aggregation.c
index 29c6740..c583e04 100644
--- a/net/batman-adv/aggregation.c
+++ b/net/batman-adv/aggregation.c
@@ -20,17 +20,12 @@
*/
#include "main.h"
+#include "translation-table.h"
#include "aggregation.h"
#include "send.h"
#include "routing.h"
#include "hard-interface.h"
-/* calculate the size of the tt information for a given packet */
-static int tt_len(const struct batman_packet *batman_packet)
-{
- return batman_packet->num_tt * ETH_ALEN;
-}
-
/* return true if new_packet can be aggregated with forw_packet */
static bool can_aggregate_with(const struct batman_packet *new_batman_packet,
int packet_len,
@@ -264,18 +259,20 @@ void receive_aggr_bat_packet(const struct ethhdr *ethhdr,
batman_packet = (struct batman_packet *)packet_buff;
do {
- /* network to host order for our 32bit seqno, and the
- orig_interval. */
+ /* network to host order for our 32bit seqno and the
+ orig_interval */
batman_packet->seqno = ntohl(batman_packet->seqno);
+ batman_packet->tt_crc = ntohs(batman_packet->tt_crc);
tt_buff = packet_buff + buff_pos + BAT_PACKET_LEN;
- receive_bat_packet(ethhdr, batman_packet,
- tt_buff, tt_len(batman_packet),
- if_incoming);
- buff_pos += BAT_PACKET_LEN + tt_len(batman_packet);
+ receive_bat_packet(ethhdr, batman_packet, tt_buff, if_incoming);
+
+ buff_pos += BAT_PACKET_LEN +
+ tt_len(batman_packet->tt_num_changes);
+
batman_packet = (struct batman_packet *)
(packet_buff + buff_pos);
} while (aggregated_packet(buff_pos, packet_len,
- batman_packet->num_tt));
+ batman_packet->tt_num_changes));
}
diff --git a/net/batman-adv/aggregation.h b/net/batman-adv/aggregation.h
index 0547fd8..216337b 100644
--- a/net/batman-adv/aggregation.h
+++ b/net/batman-adv/aggregation.h
@@ -25,9 +25,11 @@
#include "main.h"
/* is there another aggregated packet here? */
-static inline int aggregated_packet(int buff_pos, int packet_len, int num_tt)
+static inline int aggregated_packet(int buff_pos, int packet_len,
+ int tt_num_changes)
{
- int next_buff_pos = buff_pos + BAT_PACKET_LEN + (num_tt * ETH_ALEN);
+ int next_buff_pos = buff_pos + BAT_PACKET_LEN + (tt_num_changes *
+ sizeof(struct tt_change));
return (next_buff_pos <= packet_len) &&
(next_buff_pos <= MAX_AGGREGATION_BYTES);
diff --git a/net/batman-adv/bat_sysfs.c b/net/batman-adv/bat_sysfs.c
index 924d577..63738ec 100644
--- a/net/batman-adv/bat_sysfs.c
+++ b/net/batman-adv/bat_sysfs.c
@@ -375,7 +375,7 @@ BAT_ATTR_UINT(gw_sel_class, S_IRUGO | S_IWUSR, 1, TQ_MAX_VALUE,
static BAT_ATTR(gw_bandwidth, S_IRUGO | S_IWUSR, show_gw_bwidth,
store_gw_bwidth);
#ifdef CONFIG_BATMAN_ADV_DEBUG
-BAT_ATTR_UINT(log_level, S_IRUGO | S_IWUSR, 0, 3, NULL);
+BAT_ATTR_UINT(log_level, S_IRUGO | S_IWUSR, 0, 7, NULL);
#endif
static struct bat_attribute *mesh_attrs[] = {
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index b55e861..d40426c 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -152,12 +152,6 @@ static void primary_if_select(struct bat_priv *bat_priv,
batman_packet->ttl = TTL;
primary_if_update_addr(bat_priv);
-
- /***
- * hacky trick to make sure that we send the TT information via
- * our new primary interface
- */
- atomic_set(&bat_priv->tt_local_changed, 1);
}
static bool hardif_is_iface_up(const struct hard_iface *hard_iface)
@@ -340,7 +334,8 @@ int hardif_enable_interface(struct hard_iface *hard_iface,
batman_packet->flags = NO_FLAGS;
batman_packet->ttl = 2;
batman_packet->tq = TQ_MAX_VALUE;
- batman_packet->num_tt = 0;
+ batman_packet->tt_num_changes = 0;
+ batman_packet->ttvn = 0;
hard_iface->if_num = bat_priv->num_ifaces;
bat_priv->num_ifaces++;
@@ -659,6 +654,10 @@ static int batman_skb_recv(struct sk_buff *skb, struct net_device *dev,
case BAT_VIS:
ret = recv_vis_packet(skb, hard_iface);
break;
+ /* Translation table query (request or response) */
+ case BAT_TT_QUERY:
+ ret = recv_tt_query(skb, hard_iface);
+ break;
default:
ret = NET_RX_DROP;
}
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 2d6445e..49a5e64 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -86,6 +86,9 @@ int mesh_init(struct net_device *soft_iface)
spin_lock_init(&bat_priv->forw_bcast_list_lock);
spin_lock_init(&bat_priv->tt_lhash_lock);
spin_lock_init(&bat_priv->tt_ghash_lock);
+ spin_lock_init(&bat_priv->tt_changes_list_lock);
+ spin_lock_init(&bat_priv->tt_req_list_lock);
+ spin_lock_init(&bat_priv->tt_buff_lock);
spin_lock_init(&bat_priv->gw_list_lock);
spin_lock_init(&bat_priv->vis_hash_lock);
spin_lock_init(&bat_priv->vis_list_lock);
@@ -96,14 +99,13 @@ int mesh_init(struct net_device *soft_iface)
INIT_HLIST_HEAD(&bat_priv->forw_bcast_list);
INIT_HLIST_HEAD(&bat_priv->gw_list);
INIT_HLIST_HEAD(&bat_priv->softif_neigh_vids);
+ INIT_LIST_HEAD(&bat_priv->tt_changes_list);
+ INIT_LIST_HEAD(&bat_priv->tt_req_list);
if (originator_init(bat_priv) < 1)
goto err;
- if (tt_local_init(bat_priv) < 1)
- goto err;
-
- if (tt_global_init(bat_priv) < 1)
+ if (tt_init(bat_priv) < 1)
goto err;
tt_local_add(soft_iface, soft_iface->dev_addr);
@@ -137,8 +139,7 @@ void mesh_free(struct net_device *soft_iface)
gw_node_purge(bat_priv);
originator_free(bat_priv);
- tt_local_free(bat_priv);
- tt_global_free(bat_priv);
+ tt_free(bat_priv);
softif_neigh_purge(bat_priv);
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 714a241..6f53a1d 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -46,11 +46,15 @@
/* sliding packet range of received originator messages in squence numbers
* (should be a multiple of our word size) */
#define TQ_LOCAL_WINDOW_SIZE 64
+#define TT_REQUEST_TIMEOUT 3 /* seconds we have to keep pending tt_req */
+
#define TQ_GLOBAL_WINDOW_SIZE 5
#define TQ_LOCAL_BIDRECT_SEND_MINIMUM 1
#define TQ_LOCAL_BIDRECT_RECV_MINIMUM 1
#define TQ_TOTAL_BIDRECT_LIMIT 1
+#define TT_OGM_APPEND_MAX 3 /* number of OGMs sent with the last tt diff */
+
#define NO_FLAGS 0
#define NUM_WORDS (TQ_LOCAL_WINDOW_SIZE / WORD_BIT_SIZE)
@@ -96,7 +100,8 @@ enum mesh_state {
enum dbg_level {
DBG_BATMAN = 1 << 0,
DBG_ROUTES = 1 << 1, /* route added / changed / deleted */
- DBG_ALL = 3
+ DBG_TT = 1 << 2, /* translation table operations */
+ DBG_ALL = 7
};
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 991a6e7..25e7e50 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -145,6 +145,7 @@ static void orig_node_free_rcu(struct rcu_head *rcu)
tt_global_del_orig(orig_node->bat_priv, orig_node,
"originator timed out");
+ kfree(orig_node->tt_buff);
kfree(orig_node->bcast_own);
kfree(orig_node->bcast_own_sum);
kfree(orig_node);
@@ -213,6 +214,7 @@ struct orig_node *get_orig_node(struct bat_priv *bat_priv, const uint8_t *addr)
spin_lock_init(&orig_node->ogm_cnt_lock);
spin_lock_init(&orig_node->bcast_seqno_lock);
spin_lock_init(&orig_node->neigh_list_lock);
+ spin_lock_init(&orig_node->tt_buff_lock);
/* extra reference for return */
atomic_set(&orig_node->refcount, 2);
@@ -221,6 +223,8 @@ struct orig_node *get_orig_node(struct bat_priv *bat_priv, const uint8_t *addr)
memcpy(orig_node->orig, addr, ETH_ALEN);
orig_node->router = NULL;
orig_node->tt_buff = NULL;
+ orig_node->tt_buff_len = 0;
+ atomic_set(&orig_node->tt_size, 0);
orig_node->bcast_seqno_reset = jiffies - 1
- msecs_to_jiffies(RESET_PROTECTION_MS);
orig_node->batman_seqno_reset = jiffies - 1
@@ -330,9 +334,7 @@ static bool purge_orig_node(struct bat_priv *bat_priv,
if (purge_orig_neighbors(bat_priv, orig_node,
&best_neigh_node)) {
update_routes(bat_priv, orig_node,
- best_neigh_node,
- orig_node->tt_buff,
- orig_node->tt_buff_len);
+ best_neigh_node);
}
}
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index 6ddbfd8..407dd2e 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -30,7 +30,8 @@ enum bat_packettype {
BAT_UNICAST = 0x03,
BAT_BCAST = 0x04,
BAT_VIS = 0x05,
- BAT_UNICAST_FRAG = 0x06
+ BAT_UNICAST_FRAG = 0x06,
+ BAT_TT_QUERY = 0x07
};
/* this file is included by batctl which needs these defines */
@@ -63,6 +64,25 @@ enum unicast_frag_flags {
UNI_FRAG_LARGETAIL = 1 << 1
};
+/* TT_QUERY subtypes */
+#define TT_QUERY_TYPE_MASK 0x3
+
+enum tt_query_packettype {
+ TT_REQUEST = 0,
+ TT_RESPONSE = 1
+};
+
+/* TT_QUERY flags */
+enum tt_query_flags {
+ TT_FULL_TABLE = 1 << 2
+};
+
+/* TT_CHANGE flags */
+enum tt_change_flags {
+ TT_CHANGE_DEL = 0x01,
+ TT_CLIENT_ROAM = 0x02
+};
+
struct batman_packet {
uint8_t packet_type;
uint8_t version; /* batman version field */
@@ -73,8 +93,9 @@ struct batman_packet {
uint8_t prev_sender[6];
uint8_t gw_flags; /* flags related to gateway class */
uint8_t tq;
- uint8_t num_tt;
- uint8_t reserved;
+ uint8_t tt_num_changes;
+ uint8_t ttvn; /* translation table version number */
+ uint16_t tt_crc;
} __packed;
#define BAT_PACKET_LEN sizeof(struct batman_packet)
@@ -112,7 +133,7 @@ struct unicast_packet {
uint8_t packet_type;
uint8_t version; /* batman version field */
uint8_t ttl;
- uint8_t reserved;
+ uint8_t ttvn; /* destination translation table version number */
uint8_t dest[6];
} __packed;
@@ -120,7 +141,7 @@ struct unicast_frag_packet {
uint8_t packet_type;
uint8_t version; /* batman version field */
uint8_t ttl;
- uint8_t reserved;
+ uint8_t ttvn; /* destination translation table version number */
uint8_t dest[6];
uint8_t flags;
uint8_t align;
@@ -150,4 +171,32 @@ struct vis_packet {
uint8_t sender_orig[6]; /* who sent or rebroadcasted this packet */
} __packed;
+struct tt_query_packet {
+ uint8_t packet_type;
+ uint8_t version; /* batman version field */
+ uint8_t ttl;
+ /* the flag field is a combination of:
+ * - TT_REQUEST or TT_RESPONSE
+ * - TT_FULL_TABLE */
+ uint8_t flags;
+ uint8_t dst[ETH_ALEN];
+ uint8_t src[ETH_ALEN];
+ /* the ttvn field is:
+ * if TT_REQUEST: ttvn that triggered the
+ * request
+ * if TT_RESPONSE: new ttvn for the src
+ * orig_node */
+ uint8_t ttvn;
+ /* tt_data field is:
+ * if TT_REQUEST: crc associated with the
+ * ttvn
+ * if TT_RESPONSE: table_size */
+ uint16_t tt_data;
+} __packed;
+
+struct tt_change {
+ uint8_t flags;
+ uint8_t addr[ETH_ALEN];
+} __packed;
+
#endif /* _NET_BATMAN_ADV_PACKET_H_ */
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index eb6fb7d..8b0f833 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -64,27 +64,56 @@ void slide_own_bcast_window(struct hard_iface *hard_iface)
}
}
-static void update_TT(struct bat_priv *bat_priv, struct orig_node *orig_node,
- const unsigned char *tt_buff, int tt_buff_len)
+static void update_transtable(struct bat_priv *bat_priv,
+ struct orig_node *orig_node,
+ const unsigned char *tt_buff,
+ uint8_t tt_num_changes, uint8_t ttvn,
+ uint16_t tt_crc)
{
- if ((tt_buff_len != orig_node->tt_buff_len) ||
- ((tt_buff_len > 0) &&
- (orig_node->tt_buff_len > 0) &&
- (memcmp(orig_node->tt_buff, tt_buff, tt_buff_len) != 0))) {
+ uint8_t orig_ttvn = (uint8_t)atomic_read(&orig_node->last_ttvn);
+ bool full_table = true;
- if (orig_node->tt_buff_len > 0)
- tt_global_del_orig(bat_priv, orig_node,
- "originator changed tt");
+ /* the ttvn increased by one -> we can apply the attached changes */
+ if (ttvn - orig_ttvn == 1) {
+ /* the OGM could not contain the changes because they were too
+ * many to fit in one frame or because they have already been
+ * sent TT_OGM_APPEND_MAX times. In this case send a tt
+ * request */
+ if (!tt_num_changes) {
+ full_table = false;
+ goto request_table;
+ }
- if ((tt_buff_len > 0) && (tt_buff))
- tt_global_add_orig(bat_priv, orig_node,
- tt_buff, tt_buff_len);
+ tt_update_changes(bat_priv, orig_node, tt_num_changes, ttvn,
+ (struct tt_change *)tt_buff);
+
+ /* Even if we received the crc into the OGM, we prefer
+ * to recompute it to spot any possible inconsistency
+ * in the global table */
+ spin_lock_bh(&bat_priv->tt_ghash_lock);
+ orig_node->tt_crc = tt_global_crc(bat_priv, orig_node);
+ spin_unlock_bh(&bat_priv->tt_ghash_lock);
+ } else {
+ /* if we missed more than one change or our tables are not
+ * in sync anymore -> request fresh tt data */
+ if (ttvn != orig_ttvn || orig_node->tt_crc != tt_crc) {
+request_table:
+ bat_dbg(DBG_TT, bat_priv, "TT inconsistency for %pM. "
+ "Need to retrieve the correct information "
+ "(ttvn: %u last_ttvn: %u crc: %u last_crc: "
+ "%u num_changes: %u)\n", orig_node->orig, ttvn,
+ orig_ttvn, tt_crc, orig_node->tt_crc,
+ tt_num_changes);
+ send_tt_request(bat_priv, orig_node, ttvn, tt_crc,
+ full_table);
+ return;
+ }
}
}
-static void update_route(struct bat_priv *bat_priv, struct orig_node *orig_node,
- struct neigh_node *neigh_node,
- const unsigned char *tt_buff, int tt_buff_len)
+static void update_route(struct bat_priv *bat_priv,
+ struct orig_node *orig_node,
+ struct neigh_node *neigh_node)
{
struct neigh_node *curr_router;
@@ -92,11 +121,10 @@ static void update_route(struct bat_priv *bat_priv, struct orig_node *orig_node,
/* route deleted */
if ((curr_router) && (!neigh_node)) {
-
bat_dbg(DBG_ROUTES, bat_priv, "Deleting route towards: %pM\n",
orig_node->orig);
tt_global_del_orig(bat_priv, orig_node,
- "originator timed out");
+ "Deleted route towards originator");
/* route added */
} else if ((!curr_router) && (neigh_node)) {
@@ -104,9 +132,6 @@ static void update_route(struct bat_priv *bat_priv, struct orig_node *orig_node,
bat_dbg(DBG_ROUTES, bat_priv,
"Adding route towards: %pM (via %pM)\n",
orig_node->orig, neigh_node->addr);
- tt_global_add_orig(bat_priv, orig_node,
- tt_buff, tt_buff_len);
-
/* route changed */
} else if (neigh_node && curr_router) {
bat_dbg(DBG_ROUTES, bat_priv,
@@ -133,8 +158,7 @@ static void update_route(struct bat_priv *bat_priv, struct orig_node *orig_node,
}
void update_routes(struct bat_priv *bat_priv, struct orig_node *orig_node,
- struct neigh_node *neigh_node, const unsigned char *tt_buff,
- int tt_buff_len)
+ struct neigh_node *neigh_node)
{
struct neigh_node *router = NULL;
@@ -144,11 +168,7 @@ void update_routes(struct bat_priv *bat_priv, struct orig_node *orig_node,
router = orig_node_get_router(orig_node);
if (router != neigh_node)
- update_route(bat_priv, orig_node, neigh_node,
- tt_buff, tt_buff_len);
- /* may be just TT changed */
- else
- update_TT(bat_priv, orig_node, tt_buff, tt_buff_len);
+ update_route(bat_priv, orig_node, neigh_node);
out:
if (router)
@@ -360,14 +380,12 @@ static void update_orig(struct bat_priv *bat_priv, struct orig_node *orig_node,
const struct ethhdr *ethhdr,
const struct batman_packet *batman_packet,
struct hard_iface *if_incoming,
- const unsigned char *tt_buff, int tt_buff_len,
- int is_duplicate)
+ const unsigned char *tt_buff, int is_duplicate)
{
struct neigh_node *neigh_node = NULL, *tmp_neigh_node = NULL;
struct neigh_node *router = NULL;
struct orig_node *orig_node_tmp;
struct hlist_node *node;
- int tmp_tt_buff_len;
uint8_t bcast_own_sum_orig, bcast_own_sum_neigh;
bat_dbg(DBG_BATMAN, bat_priv, "update_originator(): "
@@ -432,9 +450,6 @@ static void update_orig(struct bat_priv *bat_priv, struct orig_node *orig_node,
bonding_candidate_add(orig_node, neigh_node);
- tmp_tt_buff_len = (tt_buff_len > batman_packet->num_tt * ETH_ALEN ?
- batman_packet->num_tt * ETH_ALEN : tt_buff_len);
-
/* if this neighbor already is our next hop there is nothing
* to change */
router = orig_node_get_router(orig_node);
@@ -464,15 +479,19 @@ static void update_orig(struct bat_priv *bat_priv, struct orig_node *orig_node,
goto update_tt;
}
- update_routes(bat_priv, orig_node, neigh_node,
- tt_buff, tmp_tt_buff_len);
- goto update_gw;
+ update_routes(bat_priv, orig_node, neigh_node);
update_tt:
- update_routes(bat_priv, orig_node, router,
- tt_buff, tmp_tt_buff_len);
+ /* I have to check for transtable changes only if the OGM has been
+ * sent through a primary interface */
+ if (((batman_packet->orig != ethhdr->h_source) &&
+ (batman_packet->ttl > 2)) ||
+ (batman_packet->flags & PRIMARIES_FIRST_HOP))
+ update_transtable(bat_priv, orig_node, tt_buff,
+ batman_packet->tt_num_changes,
+ batman_packet->ttvn,
+ batman_packet->tt_crc);
-update_gw:
if (orig_node->gw_flags != batman_packet->gw_flags)
gw_node_update(bat_priv, orig_node, batman_packet->gw_flags);
@@ -594,7 +613,7 @@ out:
void receive_bat_packet(const struct ethhdr *ethhdr,
struct batman_packet *batman_packet,
- const unsigned char *tt_buff, int tt_buff_len,
+ const unsigned char *tt_buff,
struct hard_iface *if_incoming)
{
struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
@@ -633,12 +652,14 @@ void receive_bat_packet(const struct ethhdr *ethhdr,
bat_dbg(DBG_BATMAN, bat_priv,
"Received BATMAN packet via NB: %pM, IF: %s [%pM] "
- "(from OG: %pM, via prev OG: %pM, seqno %d, tq %d, "
- "TTL %d, V %d, IDF %d)\n",
+ "(from OG: %pM, via prev OG: %pM, seqno %d, ttvn %u, "
+ "crc %u, changes %u, td %d, TTL %d, V %d, IDF %d)\n",
ethhdr->h_source, if_incoming->net_dev->name,
if_incoming->net_dev->dev_addr, batman_packet->orig,
batman_packet->prev_sender, batman_packet->seqno,
- batman_packet->tq, batman_packet->ttl, batman_packet->version,
+ batman_packet->ttvn, batman_packet->tt_crc,
+ batman_packet->tt_num_changes, batman_packet->tq,
+ batman_packet->ttl, batman_packet->version,
has_directlink_flag);
rcu_read_lock();
@@ -790,14 +811,14 @@ void receive_bat_packet(const struct ethhdr *ethhdr,
((orig_node->last_real_seqno == batman_packet->seqno) &&
(orig_node->last_ttl - 3 <= batman_packet->ttl))))
update_orig(bat_priv, orig_node, ethhdr, batman_packet,
- if_incoming, tt_buff, tt_buff_len, is_duplicate);
+ if_incoming, tt_buff, is_duplicate);
/* is single hop (direct) neighbor */
if (is_single_hop_neigh) {
/* mark direct link on incoming interface */
schedule_forward_packet(orig_node, ethhdr, batman_packet,
- 1, tt_buff_len, if_incoming);
+ 1, if_incoming);
bat_dbg(DBG_BATMAN, bat_priv, "Forwarding packet: "
"rebroadcast neighbor packet with direct link flag\n");
@@ -820,7 +841,7 @@ void receive_bat_packet(const struct ethhdr *ethhdr,
bat_dbg(DBG_BATMAN, bat_priv,
"Forwarding packet: rebroadcast originator packet\n");
schedule_forward_packet(orig_node, ethhdr, batman_packet,
- 0, tt_buff_len, if_incoming);
+ 0, if_incoming);
out_neigh:
if ((orig_neigh_node) && (!is_single_hop_neigh))
@@ -1167,6 +1188,70 @@ static struct neigh_node *find_ifalter_router(struct orig_node *primary_orig,
return router;
}
+int recv_tt_query(struct sk_buff *skb, struct hard_iface *recv_if)
+{
+ struct bat_priv *bat_priv = netdev_priv(recv_if->soft_iface);
+ struct tt_query_packet *tt_query;
+ struct ethhdr *ethhdr;
+
+ /* drop packet if it has not necessary minimum size */
+ if (unlikely(!pskb_may_pull(skb, sizeof(struct tt_query_packet))))
+ goto out;
+
+ /* I could need to modify it */
+ if (skb_cow(skb, sizeof(struct tt_query_packet)) < 0)
+ goto out;
+
+ ethhdr = (struct ethhdr *)skb_mac_header(skb);
+
+ /* packet with unicast indication but broadcast recipient */
+ if (is_broadcast_ether_addr(ethhdr->h_dest))
+ goto out;
+
+ /* packet with broadcast sender address */
+ if (is_broadcast_ether_addr(ethhdr->h_source))
+ goto out;
+
+ tt_query = (struct tt_query_packet *)skb->data;
+
+ tt_query->tt_data = ntohs(tt_query->tt_data);
+
+ switch (tt_query->flags & TT_QUERY_TYPE_MASK) {
+ case TT_REQUEST:
+ /* If we cannot provide an answer the tt_request is
+ * forwarded */
+ if (!send_tt_response(bat_priv, tt_query)) {
+ bat_dbg(DBG_TT, bat_priv,
+ "Routing TT_REQUEST to %pM [%c]\n",
+ tt_query->dst,
+ (tt_query->flags & TT_FULL_TABLE ? 'F' : '.'));
+ tt_query->tt_data = htons(tt_query->tt_data);
+ return route_unicast_packet(skb, recv_if);
+ }
+ break;
+ case TT_RESPONSE:
+ /* packet needs to be linearised to access the TT changes */
+ if (skb_linearize(skb) < 0)
+ goto out;
+
+ if (is_my_mac(tt_query->dst))
+ handle_tt_response(bat_priv, tt_query);
+ else {
+ bat_dbg(DBG_TT, bat_priv,
+ "Routing TT_RESPONSE to %pM [%c]\n",
+ tt_query->dst,
+ (tt_query->flags & TT_FULL_TABLE ? 'F' : '.'));
+ tt_query->tt_data = htons(tt_query->tt_data);
+ return route_unicast_packet(skb, recv_if);
+ }
+ break;
+ }
+
+out:
+ /* returning NET_RX_DROP will make the caller function kfree the skb */
+ return NET_RX_DROP;
+}
+
/* find a suitable router for this originator, and use
* bonding if possible. increases the found neighbors
* refcount.*/
@@ -1353,14 +1438,82 @@ out:
return ret;
}
+static int check_unicast_ttvn(struct bat_priv *bat_priv,
+ struct sk_buff *skb) {
+ uint8_t curr_ttvn;
+ struct orig_node *orig_node;
+ struct ethhdr *ethhdr;
+ struct hard_iface *primary_if;
+ struct unicast_packet *unicast_packet;
+
+ /* I could need to modify it */
+ if (skb_cow(skb, sizeof(struct unicast_packet)) < 0)
+ return 0;
+
+ unicast_packet = (struct unicast_packet *)skb->data;
+
+ if (is_my_mac(unicast_packet->dest))
+ curr_ttvn = (uint8_t)atomic_read(&bat_priv->ttvn);
+ else {
+ orig_node = orig_hash_find(bat_priv, unicast_packet->dest);
+
+ if (!orig_node)
+ return 0;
+
+ curr_ttvn = (uint8_t)atomic_read(&orig_node->last_ttvn);
+ orig_node_free_ref(orig_node);
+ }
+
+ /* Check whether I have to reroute the packet */
+ if (seq_before(unicast_packet->ttvn, curr_ttvn)) {
+ /* Linearize the skb before accessing it */
+ if (skb_linearize(skb) < 0)
+ return 0;
+
+ ethhdr = (struct ethhdr *)(skb->data +
+ sizeof(struct unicast_packet));
+
+ orig_node = transtable_search(bat_priv, ethhdr->h_dest);
+
+ if (!orig_node) {
+ if (!is_my_client(bat_priv, ethhdr->h_dest))
+ return 0;
+ primary_if = primary_if_get_selected(bat_priv);
+ if (!primary_if)
+ return 0;
+ memcpy(unicast_packet->dest,
+ primary_if->net_dev->dev_addr, ETH_ALEN);
+ hardif_free_ref(primary_if);
+ } else {
+ memcpy(unicast_packet->dest, orig_node->orig,
+ ETH_ALEN);
+ curr_ttvn = (uint8_t)
+ atomic_read(&orig_node->last_ttvn);
+ orig_node_free_ref(orig_node);
+ }
+
+ bat_dbg(DBG_ROUTES, bat_priv, "TTVN mismatch (old_ttvn %u "
+ "new_ttvn %u)! Rerouting unicast packet (for %pM) to "
+ "%pM\n", unicast_packet->ttvn, curr_ttvn,
+ ethhdr->h_dest, unicast_packet->dest);
+
+ unicast_packet->ttvn = curr_ttvn;
+ }
+ return 1;
+}
+
int recv_unicast_packet(struct sk_buff *skb, struct hard_iface *recv_if)
{
+ struct bat_priv *bat_priv = netdev_priv(recv_if->soft_iface);
struct unicast_packet *unicast_packet;
int hdr_size = sizeof(*unicast_packet);
if (check_unicast_packet(skb, hdr_size) < 0)
return NET_RX_DROP;
+ if (!check_unicast_ttvn(bat_priv, skb))
+ return NET_RX_DROP;
+
unicast_packet = (struct unicast_packet *)skb->data;
/* packet for me */
@@ -1383,6 +1536,9 @@ int recv_ucast_frag_packet(struct sk_buff *skb, struct hard_iface *recv_if)
if (check_unicast_packet(skb, hdr_size) < 0)
return NET_RX_DROP;
+ if (!check_unicast_ttvn(bat_priv, skb))
+ return NET_RX_DROP;
+
unicast_packet = (struct unicast_frag_packet *)skb->data;
/* packet for me */
diff --git a/net/batman-adv/routing.h b/net/batman-adv/routing.h
index 0ce0392..e77d464 100644
--- a/net/batman-adv/routing.h
+++ b/net/batman-adv/routing.h
@@ -25,11 +25,10 @@
void slide_own_bcast_window(struct hard_iface *hard_iface);
void receive_bat_packet(const struct ethhdr *ethhdr,
struct batman_packet *batman_packet,
- const unsigned char *tt_buff, int tt_buff_len,
+ const unsigned char *tt_buff,
struct hard_iface *if_incoming);
void update_routes(struct bat_priv *bat_priv, struct orig_node *orig_node,
- struct neigh_node *neigh_node, const unsigned char *tt_buff,
- int tt_buff_len);
+ struct neigh_node *neigh_node);
int route_unicast_packet(struct sk_buff *skb, struct hard_iface *recv_if);
int recv_icmp_packet(struct sk_buff *skb, struct hard_iface *recv_if);
int recv_unicast_packet(struct sk_buff *skb, struct hard_iface *recv_if);
@@ -37,6 +36,7 @@ int recv_ucast_frag_packet(struct sk_buff *skb, struct hard_iface *recv_if);
int recv_bcast_packet(struct sk_buff *skb, struct hard_iface *recv_if);
int recv_vis_packet(struct sk_buff *skb, struct hard_iface *recv_if);
int recv_bat_packet(struct sk_buff *skb, struct hard_iface *recv_if);
+int recv_tt_query(struct sk_buff *skb, struct hard_iface *recv_if);
struct neigh_node *find_router(struct bat_priv *bat_priv,
struct orig_node *orig_node,
const struct hard_iface *recv_if);
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index be0d581..6b14075 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -120,7 +120,7 @@ static void send_packet_to_if(struct forw_packet *forw_packet,
/* adjust all flags and log packets */
while (aggregated_packet(buff_pos,
forw_packet->packet_len,
- batman_packet->num_tt)) {
+ batman_packet->tt_num_changes)) {
/* we might have aggregated direct link packets with an
* ordinary base packet */
@@ -135,17 +135,17 @@ static void send_packet_to_if(struct forw_packet *forw_packet,
"Forwarding"));
bat_dbg(DBG_BATMAN, bat_priv,
"%s %spacket (originator %pM, seqno %d, TQ %d, TTL %d,"
- " IDF %s) on interface %s [%pM]\n",
+ " IDF %s, hvn %d) on interface %s [%pM]\n",
fwd_str, (packet_num > 0 ? "aggregated " : ""),
batman_packet->orig, ntohl(batman_packet->seqno),
batman_packet->tq, batman_packet->ttl,
(batman_packet->flags & DIRECTLINK ?
"on" : "off"),
- hard_iface->net_dev->name,
+ batman_packet->ttvn, hard_iface->net_dev->name,
hard_iface->net_dev->dev_addr);
buff_pos += sizeof(*batman_packet) +
- (batman_packet->num_tt * ETH_ALEN);
+ tt_len(batman_packet->tt_num_changes);
packet_num++;
batman_packet = (struct batman_packet *)
(forw_packet->skb->data + buff_pos);
@@ -213,25 +213,18 @@ static void send_packet(struct forw_packet *forw_packet)
rcu_read_unlock();
}
-static void rebuild_batman_packet(struct bat_priv *bat_priv,
- struct hard_iface *hard_iface)
+static void realloc_packet_buffer(struct hard_iface *hard_iface,
+ int new_len)
{
- int new_len;
unsigned char *new_buff;
struct batman_packet *batman_packet;
- new_len = sizeof(*batman_packet) + (bat_priv->num_local_tt * ETH_ALEN);
new_buff = kmalloc(new_len, GFP_ATOMIC);
/* keep old buffer if kmalloc should fail */
if (new_buff) {
memcpy(new_buff, hard_iface->packet_buff,
sizeof(*batman_packet));
- batman_packet = (struct batman_packet *)new_buff;
-
- batman_packet->num_tt = tt_local_fill_buffer(bat_priv,
- new_buff + sizeof(*batman_packet),
- new_len - sizeof(*batman_packet));
kfree(hard_iface->packet_buff);
hard_iface->packet_buff = new_buff;
@@ -239,6 +232,46 @@ static void rebuild_batman_packet(struct bat_priv *bat_priv,
}
}
+/* when calling this function (hard_iface == primary_if) has to be true */
+static void prepare_packet_buffer(struct bat_priv *bat_priv,
+ struct hard_iface *hard_iface)
+{
+ int new_len;
+ struct batman_packet *batman_packet;
+
+ new_len = BAT_PACKET_LEN +
+ tt_len((uint8_t)atomic_read(&bat_priv->tt_local_changes));
+
+ /* if we have too many changes for one packet don't send any
+ * and wait for the tt table request which will be fragmented */
+ if (new_len > hard_iface->soft_iface->mtu)
+ new_len = BAT_PACKET_LEN;
+
+ realloc_packet_buffer(hard_iface, new_len);
+ batman_packet = (struct batman_packet *)hard_iface->packet_buff;
+
+ atomic_set(&bat_priv->tt_crc, tt_local_crc(bat_priv));
+
+ /* reset the sending counter */
+ atomic_set(&bat_priv->tt_ogm_append_cnt, TT_OGM_APPEND_MAX);
+
+ batman_packet->tt_num_changes = tt_changes_fill_buffer(bat_priv,
+ hard_iface->packet_buff + BAT_PACKET_LEN,
+ hard_iface->packet_len - BAT_PACKET_LEN);
+
+}
+
+static void reset_packet_buffer(struct bat_priv *bat_priv,
+ struct hard_iface *hard_iface)
+{
+ struct batman_packet *batman_packet;
+
+ realloc_packet_buffer(hard_iface, BAT_PACKET_LEN);
+
+ batman_packet = (struct batman_packet *)hard_iface->packet_buff;
+ batman_packet->tt_num_changes = 0;
+}
+
void schedule_own_packet(struct hard_iface *hard_iface)
{
struct bat_priv *bat_priv = netdev_priv(hard_iface->soft_iface);
@@ -264,14 +297,22 @@ void schedule_own_packet(struct hard_iface *hard_iface)
if (hard_iface->if_status == IF_TO_BE_ACTIVATED)
hard_iface->if_status = IF_ACTIVE;
- /* if local tt has changed and interface is a primary interface */
- if ((atomic_read(&bat_priv->tt_local_changed)) &&
- (hard_iface == primary_if))
- rebuild_batman_packet(bat_priv, hard_iface);
+ if (hard_iface == primary_if) {
+ /* if at least one change happened */
+ if (atomic_read(&bat_priv->tt_local_changes) > 0) {
+ prepare_packet_buffer(bat_priv, hard_iface);
+ /* Increment the TTVN only once per OGM interval */
+ atomic_inc(&bat_priv->ttvn);
+ }
+
+ /* if the changes have been sent enough times */
+ if (!atomic_dec_not_zero(&bat_priv->tt_ogm_append_cnt))
+ reset_packet_buffer(bat_priv, hard_iface);
+ }
/**
* NOTE: packet_buff might just have been re-allocated in
- * rebuild_batman_packet()
+ * prepare_packet_buffer() or in reset_packet_buffer()
*/
batman_packet = (struct batman_packet *)hard_iface->packet_buff;
@@ -279,6 +320,9 @@ void schedule_own_packet(struct hard_iface *hard_iface)
batman_packet->seqno =
htonl((uint32_t)atomic_read(&hard_iface->seqno));
+ batman_packet->ttvn = atomic_read(&bat_priv->ttvn);
+ batman_packet->tt_crc = htons((uint16_t)atomic_read(&bat_priv->tt_crc));
+
if (vis_server == VIS_TYPE_SERVER_SYNC)
batman_packet->flags |= VIS_SERVER;
else
@@ -307,13 +351,14 @@ void schedule_own_packet(struct hard_iface *hard_iface)
void schedule_forward_packet(struct orig_node *orig_node,
const struct ethhdr *ethhdr,
struct batman_packet *batman_packet,
- int directlink, int tt_buff_len,
+ int directlink,
struct hard_iface *if_incoming)
{
struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
struct neigh_node *router;
uint8_t in_tq, in_ttl, tq_avg = 0;
unsigned long send_time;
+ uint8_t tt_num_changes;
if (batman_packet->ttl <= 1) {
bat_dbg(DBG_BATMAN, bat_priv, "ttl exceeded\n");
@@ -324,6 +369,7 @@ void schedule_forward_packet(struct orig_node *orig_node,
in_tq = batman_packet->tq;
in_ttl = batman_packet->ttl;
+ tt_num_changes = batman_packet->tt_num_changes;
batman_packet->ttl--;
memcpy(batman_packet->prev_sender, ethhdr->h_source, ETH_ALEN);
@@ -356,6 +402,7 @@ void schedule_forward_packet(struct orig_node *orig_node,
batman_packet->ttl);
batman_packet->seqno = htonl(batman_packet->seqno);
+ batman_packet->tt_crc = htons(batman_packet->tt_crc);
/* switch of primaries first hop flag when forwarding */
batman_packet->flags &= ~PRIMARIES_FIRST_HOP;
@@ -367,7 +414,7 @@ void schedule_forward_packet(struct orig_node *orig_node,
send_time = forward_send_time();
add_bat_packet_to_list(bat_priv,
(unsigned char *)batman_packet,
- sizeof(*batman_packet) + tt_buff_len,
+ sizeof(*batman_packet) + tt_len(tt_num_changes),
if_incoming, 0, send_time);
}
diff --git a/net/batman-adv/send.h b/net/batman-adv/send.h
index 6d9c14d..633224a 100644
--- a/net/batman-adv/send.h
+++ b/net/batman-adv/send.h
@@ -28,7 +28,7 @@ void schedule_own_packet(struct hard_iface *hard_iface);
void schedule_forward_packet(struct orig_node *orig_node,
const struct ethhdr *ethhdr,
struct batman_packet *batman_packet,
- int directlink, int tt_buff_len,
+ int directlink,
struct hard_iface *if_outgoing);
int add_bcast_packet_to_list(struct bat_priv *bat_priv,
const struct sk_buff *skb);
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 69c0022..c288d93 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -534,7 +534,7 @@ static int interface_set_mac_addr(struct net_device *dev, void *p)
/* only modify transtable if it has been initialised before */
if (atomic_read(&bat_priv->mesh_state) == MESH_ACTIVE) {
tt_local_remove(bat_priv, dev->dev_addr,
- "mac address changed");
+ "mac address changed");
tt_local_add(dev, addr->sa_data);
}
@@ -592,7 +592,7 @@ static int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
if (curr_softif_neigh)
goto dropped;
- /* TODO: check this for locks */
+ /* Register the client MAC in the transtable */
tt_local_add(soft_iface, ethhdr->h_source);
if (is_multicast_ether_addr(ethhdr->h_dest)) {
@@ -830,7 +830,12 @@ struct net_device *softif_create(const char *name)
atomic_set(&bat_priv->mesh_state, MESH_INACTIVE);
atomic_set(&bat_priv->bcast_seqno, 1);
- atomic_set(&bat_priv->tt_local_changed, 0);
+ atomic_set(&bat_priv->ttvn, 0);
+ atomic_set(&bat_priv->tt_local_changes, 0);
+ atomic_set(&bat_priv->tt_ogm_append_cnt, 0);
+
+ bat_priv->tt_buff = NULL;
+ bat_priv->tt_buff_len = 0;
bat_priv->primary_if = NULL;
bat_priv->num_ifaces = 0;
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 561f769..597cd1a 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -23,13 +23,17 @@
#include "translation-table.h"
#include "soft-interface.h"
#include "hard-interface.h"
+#include "send.h"
#include "hash.h"
#include "originator.h"
+#include "routing.h"
-static void tt_local_purge(struct work_struct *work);
-static void _tt_global_del_orig(struct bat_priv *bat_priv,
- struct tt_global_entry *tt_global_entry,
- const char *message);
+#include <linux/crc16.h>
+
+static void _tt_global_del(struct bat_priv *bat_priv,
+ struct tt_global_entry *tt_global_entry,
+ const char *message);
+static void tt_purge(struct work_struct *work);
/* returns 1 if they are the same mac addr */
static int compare_ltt(const struct hlist_node *node, const void *data2)
@@ -49,10 +53,11 @@ static int compare_gtt(const struct hlist_node *node, const void *data2)
return (memcmp(data1, data2, ETH_ALEN) == 0 ? 1 : 0);
}
-static void tt_local_start_timer(struct bat_priv *bat_priv)
+static void tt_start_timer(struct bat_priv *bat_priv)
{
- INIT_DELAYED_WORK(&bat_priv->tt_work, tt_local_purge);
- queue_delayed_work(bat_event_workqueue, &bat_priv->tt_work, 10 * HZ);
+ INIT_DELAYED_WORK(&bat_priv->tt_work, tt_purge);
+ queue_delayed_work(bat_event_workqueue, &bat_priv->tt_work,
+ msecs_to_jiffies(5000));
}
static struct tt_local_entry *tt_local_hash_find(struct bat_priv *bat_priv,
@@ -112,7 +117,42 @@ static struct tt_global_entry *tt_global_hash_find(struct bat_priv *bat_priv,
return tt_global_entry_tmp;
}
-int tt_local_init(struct bat_priv *bat_priv)
+static bool is_out_of_time(unsigned long starting_time, unsigned long timeout)
+{
+ unsigned long deadline;
+ deadline = starting_time + msecs_to_jiffies(timeout);
+
+ return time_after(jiffies, deadline);
+}
+
+static void tt_local_event(struct bat_priv *bat_priv, uint8_t op,
+ const uint8_t *addr)
+{
+ struct tt_change_node *tt_change_node;
+
+ tt_change_node = kmalloc(sizeof(*tt_change_node), GFP_ATOMIC);
+
+ if (!tt_change_node)
+ return;
+
+ tt_change_node->change.flags = op;
+ memcpy(tt_change_node->change.addr, addr, ETH_ALEN);
+
+ spin_lock_bh(&bat_priv->tt_changes_list_lock);
+ /* track the change in the OGMinterval list */
+ list_add_tail(&tt_change_node->list, &bat_priv->tt_changes_list);
+ atomic_inc(&bat_priv->tt_local_changes);
+ spin_unlock_bh(&bat_priv->tt_changes_list_lock);
+
+ atomic_set(&bat_priv->tt_ogm_append_cnt, 0);
+}
+
+int tt_len(int changes_num)
+{
+ return changes_num * sizeof(struct tt_change);
+}
+
+static int tt_local_init(struct bat_priv *bat_priv)
{
if (bat_priv->tt_local_hash)
return 1;
@@ -122,9 +162,6 @@ int tt_local_init(struct bat_priv *bat_priv)
if (!bat_priv->tt_local_hash)
return 0;
- atomic_set(&bat_priv->tt_local_changed, 0);
- tt_local_start_timer(bat_priv);
-
return 1;
}
@@ -133,40 +170,24 @@ void tt_local_add(struct net_device *soft_iface, const uint8_t *addr)
struct bat_priv *bat_priv = netdev_priv(soft_iface);
struct tt_local_entry *tt_local_entry;
struct tt_global_entry *tt_global_entry;
- int required_bytes;
spin_lock_bh(&bat_priv->tt_lhash_lock);
tt_local_entry = tt_local_hash_find(bat_priv, addr);
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
if (tt_local_entry) {
tt_local_entry->last_seen = jiffies;
- return;
+ goto unlock;
}
- /* only announce as many hosts as possible in the batman-packet and
- space in batman_packet->num_tt That also should give a limit to
- MAC-flooding. */
- required_bytes = (bat_priv->num_local_tt + 1) * ETH_ALEN;
- required_bytes += BAT_PACKET_LEN;
-
- if ((required_bytes > ETH_DATA_LEN) ||
- (atomic_read(&bat_priv->aggregated_ogms) &&
- required_bytes > MAX_AGGREGATION_BYTES) ||
- (bat_priv->num_local_tt + 1 > 255)) {
- bat_dbg(DBG_ROUTES, bat_priv,
- "Can't add new local tt entry (%pM): "
- "number of local tt entries exceeds packet size\n",
- addr);
- return;
- }
-
- bat_dbg(DBG_ROUTES, bat_priv,
- "Creating new local tt entry: %pM\n", addr);
-
tt_local_entry = kmalloc(sizeof(*tt_local_entry), GFP_ATOMIC);
if (!tt_local_entry)
- return;
+ goto unlock;
+
+ tt_local_event(bat_priv, NO_FLAGS, addr);
+
+ bat_dbg(DBG_TT, bat_priv,
+ "Creating new local tt entry: %pM (ttvn: %d)\n", addr,
+ (uint8_t)atomic_read(&bat_priv->ttvn));
memcpy(tt_local_entry->addr, addr, ETH_ALEN);
tt_local_entry->last_seen = jiffies;
@@ -177,13 +198,9 @@ void tt_local_add(struct net_device *soft_iface, const uint8_t *addr)
else
tt_local_entry->never_purge = 0;
- spin_lock_bh(&bat_priv->tt_lhash_lock);
-
hash_add(bat_priv->tt_local_hash, compare_ltt, choose_orig,
tt_local_entry, &tt_local_entry->hash_entry);
- bat_priv->num_local_tt++;
- atomic_set(&bat_priv->tt_local_changed, 1);
-
+ atomic_inc(&bat_priv->num_local_tt);
spin_unlock_bh(&bat_priv->tt_lhash_lock);
/* remove address from global hash if present */
@@ -192,46 +209,60 @@ void tt_local_add(struct net_device *soft_iface, const uint8_t *addr)
tt_global_entry = tt_global_hash_find(bat_priv, addr);
if (tt_global_entry)
- _tt_global_del_orig(bat_priv, tt_global_entry,
- "local tt received");
+ _tt_global_del(bat_priv, tt_global_entry,
+ "local tt received");
spin_unlock_bh(&bat_priv->tt_ghash_lock);
+ return;
+unlock:
+ spin_unlock_bh(&bat_priv->tt_lhash_lock);
}
-int tt_local_fill_buffer(struct bat_priv *bat_priv,
- unsigned char *buff, int buff_len)
+int tt_changes_fill_buffer(struct bat_priv *bat_priv,
+ unsigned char *buff, int buff_len)
{
- struct hashtable_t *hash = bat_priv->tt_local_hash;
- struct tt_local_entry *tt_local_entry;
- struct hlist_node *node;
- struct hlist_head *head;
- int i, count = 0;
+ int count = 0, tot_changes = 0;
+ struct tt_change_node *entry, *safe;
- spin_lock_bh(&bat_priv->tt_lhash_lock);
+ if (buff_len > 0)
+ tot_changes = buff_len / tt_len(1);
- for (i = 0; i < hash->size; i++) {
- head = &hash->table[i];
-
- rcu_read_lock();
- hlist_for_each_entry_rcu(tt_local_entry, node,
- head, hash_entry) {
- if (buff_len < (count + 1) * ETH_ALEN)
- break;
-
- memcpy(buff + (count * ETH_ALEN), tt_local_entry->addr,
- ETH_ALEN);
+ spin_lock_bh(&bat_priv->tt_changes_list_lock);
+ atomic_set(&bat_priv->tt_local_changes, 0);
+ list_for_each_entry_safe(entry, safe, &bat_priv->tt_changes_list,
+ list) {
+ if (count < tot_changes) {
+ memcpy(buff + tt_len(count),
+ &entry->change, sizeof(struct tt_change));
count++;
}
- rcu_read_unlock();
+ list_del(&entry->list);
+ kfree(entry);
}
+ spin_unlock_bh(&bat_priv->tt_changes_list_lock);
- /* if we did not get all new local tts see you next time ;-) */
- if (count == bat_priv->num_local_tt)
- atomic_set(&bat_priv->tt_local_changed, 0);
+ /* Keep the buffer for possible tt_request */
+ spin_lock_bh(&bat_priv->tt_buff_lock);
+ kfree(bat_priv->tt_buff);
+ bat_priv->tt_buff_len = 0;
+ bat_priv->tt_buff = NULL;
+ /* We check whether this new OGM has no changes due to size
+ * problems */
+ if (buff_len > 0) {
+ /**
+ * if kmalloc() fails we will reply with the full table
+ * instead of providing the diff
+ */
+ bat_priv->tt_buff = kmalloc(buff_len, GFP_ATOMIC);
+ if (bat_priv->tt_buff) {
+ memcpy(bat_priv->tt_buff, buff, buff_len);
+ bat_priv->tt_buff_len = buff_len;
+ }
+ }
+ spin_unlock_bh(&bat_priv->tt_buff_lock);
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
- return count;
+ return tot_changes;
}
int tt_local_seq_print_text(struct seq_file *seq, void *offset)
@@ -263,8 +294,8 @@ int tt_local_seq_print_text(struct seq_file *seq, void *offset)
}
seq_printf(seq, "Locally retrieved addresses (from %s) "
- "announced via TT:\n",
- net_dev->name);
+ "announced via TT (TTVN: %u):\n",
+ net_dev->name, (uint8_t)atomic_read(&bat_priv->ttvn));
spin_lock_bh(&bat_priv->tt_lhash_lock);
@@ -311,54 +342,51 @@ out:
return ret;
}
-static void _tt_local_del(struct hlist_node *node, void *arg)
+static void tt_local_entry_free(struct hlist_node *node, void *arg)
{
struct bat_priv *bat_priv = arg;
void *data = container_of(node, struct tt_local_entry, hash_entry);
kfree(data);
- bat_priv->num_local_tt--;
- atomic_set(&bat_priv->tt_local_changed, 1);
+ atomic_dec(&bat_priv->num_local_tt);
}
static void tt_local_del(struct bat_priv *bat_priv,
struct tt_local_entry *tt_local_entry,
const char *message)
{
- bat_dbg(DBG_ROUTES, bat_priv, "Deleting local tt entry (%pM): %s\n",
+ bat_dbg(DBG_TT, bat_priv, "Deleting local tt entry (%pM): %s\n",
tt_local_entry->addr, message);
+ atomic_dec(&bat_priv->num_local_tt);
+
hash_remove(bat_priv->tt_local_hash, compare_ltt, choose_orig,
tt_local_entry->addr);
- _tt_local_del(&tt_local_entry->hash_entry, bat_priv);
+
+ tt_local_entry_free(&tt_local_entry->hash_entry, bat_priv);
}
-void tt_local_remove(struct bat_priv *bat_priv,
- const uint8_t *addr, const char *message)
+void tt_local_remove(struct bat_priv *bat_priv, const uint8_t *addr,
+ const char *message)
{
struct tt_local_entry *tt_local_entry;
spin_lock_bh(&bat_priv->tt_lhash_lock);
-
tt_local_entry = tt_local_hash_find(bat_priv, addr);
- if (tt_local_entry)
+ if (tt_local_entry) {
+ tt_local_event(bat_priv, TT_CHANGE_DEL, tt_local_entry->addr);
tt_local_del(bat_priv, tt_local_entry, message);
-
+ }
spin_unlock_bh(&bat_priv->tt_lhash_lock);
}
-static void tt_local_purge(struct work_struct *work)
+static void tt_local_purge(struct bat_priv *bat_priv)
{
- struct delayed_work *delayed_work =
- container_of(work, struct delayed_work, work);
- struct bat_priv *bat_priv =
- container_of(delayed_work, struct bat_priv, tt_work);
struct hashtable_t *hash = bat_priv->tt_local_hash;
struct tt_local_entry *tt_local_entry;
struct hlist_node *node, *node_tmp;
struct hlist_head *head;
- unsigned long timeout;
int i;
spin_lock_bh(&bat_priv->tt_lhash_lock);
@@ -371,32 +399,53 @@ static void tt_local_purge(struct work_struct *work)
if (tt_local_entry->never_purge)
continue;
- timeout = tt_local_entry->last_seen;
- timeout += TT_LOCAL_TIMEOUT * HZ;
-
- if (time_before(jiffies, timeout))
+ if (!is_out_of_time(tt_local_entry->last_seen,
+ TT_LOCAL_TIMEOUT * 1000))
continue;
+ tt_local_event(bat_priv, TT_CHANGE_DEL,
+ tt_local_entry->addr);
tt_local_del(bat_priv, tt_local_entry,
- "address timed out");
+ "address timed out");
}
}
spin_unlock_bh(&bat_priv->tt_lhash_lock);
- tt_local_start_timer(bat_priv);
}
-void tt_local_free(struct bat_priv *bat_priv)
+static void tt_local_table_free(struct bat_priv *bat_priv)
{
+ struct hashtable_t *hash;
+ int i;
+ spinlock_t *list_lock; /* protects write access to the hash lists */
+ struct hlist_head *head;
+ struct hlist_node *node, *node_tmp;
+ struct tt_local_entry *tt_local_entry;
+
if (!bat_priv->tt_local_hash)
return;
- cancel_delayed_work_sync(&bat_priv->tt_work);
- hash_delete(bat_priv->tt_local_hash, _tt_local_del, bat_priv);
+ hash = bat_priv->tt_local_hash;
+
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+ list_lock = &hash->list_locks[i];
+
+ spin_lock_bh(list_lock);
+ hlist_for_each_entry_safe(tt_local_entry, node, node_tmp,
+ head, hash_entry) {
+ hlist_del_rcu(node);
+ kfree(tt_local_entry);
+ }
+ spin_unlock_bh(list_lock);
+ }
+
+ hash_destroy(hash);
+
bat_priv->tt_local_hash = NULL;
}
-int tt_global_init(struct bat_priv *bat_priv)
+static int tt_global_init(struct bat_priv *bat_priv)
{
if (bat_priv->tt_global_hash)
return 1;
@@ -409,73 +458,78 @@ int tt_global_init(struct bat_priv *bat_priv)
return 1;
}
-void tt_global_add_orig(struct bat_priv *bat_priv,
- struct orig_node *orig_node,
- const unsigned char *tt_buff, int tt_buff_len)
+static void tt_changes_list_free(struct bat_priv *bat_priv)
+{
+ struct tt_change_node *entry, *safe;
+
+ spin_lock_bh(&bat_priv->tt_changes_list_lock);
+
+ list_for_each_entry_safe(entry, safe, &bat_priv->tt_changes_list,
+ list) {
+ list_del(&entry->list);
+ kfree(entry);
+ }
+
+ atomic_set(&bat_priv->tt_local_changes, 0);
+ spin_unlock_bh(&bat_priv->tt_changes_list_lock);
+}
+
+/* caller must hold orig_node refcount */
+int tt_global_add(struct bat_priv *bat_priv, struct orig_node *orig_node,
+ const unsigned char *tt_addr, uint8_t ttvn)
{
struct tt_global_entry *tt_global_entry;
struct tt_local_entry *tt_local_entry;
- int tt_buff_count = 0;
- const unsigned char *tt_ptr;
-
- while ((tt_buff_count + 1) * ETH_ALEN <= tt_buff_len) {
- spin_lock_bh(&bat_priv->tt_ghash_lock);
-
- tt_ptr = tt_buff + (tt_buff_count * ETH_ALEN);
- tt_global_entry = tt_global_hash_find(bat_priv, tt_ptr);
-
- if (!tt_global_entry) {
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
-
- tt_global_entry = kmalloc(sizeof(*tt_global_entry),
- GFP_ATOMIC);
-
- if (!tt_global_entry)
- break;
-
- memcpy(tt_global_entry->addr, tt_ptr, ETH_ALEN);
-
- bat_dbg(DBG_ROUTES, bat_priv,
- "Creating new global tt entry: "
- "%pM (via %pM)\n",
- tt_global_entry->addr, orig_node->orig);
-
- spin_lock_bh(&bat_priv->tt_ghash_lock);
- hash_add(bat_priv->tt_global_hash, compare_gtt,
- choose_orig, tt_global_entry,
- &tt_global_entry->hash_entry);
-
- }
-
+ struct orig_node *orig_node_tmp;
+
+ spin_lock_bh(&bat_priv->tt_ghash_lock);
+ tt_global_entry = tt_global_hash_find(bat_priv, tt_addr);
+
+ if (!tt_global_entry) {
+ tt_global_entry =
+ kmalloc(sizeof(*tt_global_entry),
+ GFP_ATOMIC);
+ if (!tt_global_entry)
+ goto unlock;
+ memcpy(tt_global_entry->addr, tt_addr, ETH_ALEN);
+ /* Assign the new orig_node */
+ atomic_inc(&orig_node->refcount);
tt_global_entry->orig_node = orig_node;
- spin_unlock_bh(&bat_priv->tt_ghash_lock);
-
- /* remove address from local hash if present */
- spin_lock_bh(&bat_priv->tt_lhash_lock);
-
- tt_ptr = tt_buff + (tt_buff_count * ETH_ALEN);
- tt_local_entry = tt_local_hash_find(bat_priv, tt_ptr);
-
- if (tt_local_entry)
- tt_local_del(bat_priv, tt_local_entry,
- "global tt received");
-
- spin_unlock_bh(&bat_priv->tt_lhash_lock);
-
- tt_buff_count++;
- }
-
- /* initialize, and overwrite if malloc succeeds */
- orig_node->tt_buff = NULL;
- orig_node->tt_buff_len = 0;
-
- if (tt_buff_len > 0) {
- orig_node->tt_buff = kmalloc(tt_buff_len, GFP_ATOMIC);
- if (orig_node->tt_buff) {
- memcpy(orig_node->tt_buff, tt_buff, tt_buff_len);
- orig_node->tt_buff_len = tt_buff_len;
+ tt_global_entry->ttvn = ttvn;
+ atomic_inc(&orig_node->tt_size);
+ hash_add(bat_priv->tt_global_hash, compare_gtt,
+ choose_orig, tt_global_entry,
+ &tt_global_entry->hash_entry);
+ } else {
+ if (tt_global_entry->orig_node != orig_node) {
+ atomic_dec(&tt_global_entry->orig_node->tt_size);
+ orig_node_tmp = tt_global_entry->orig_node;
+ atomic_inc(&orig_node->refcount);
+ tt_global_entry->orig_node = orig_node;
+ tt_global_entry->ttvn = ttvn;
+ orig_node_free_ref(orig_node_tmp);
+ atomic_inc(&orig_node->tt_size);
}
}
+
+ spin_unlock_bh(&bat_priv->tt_ghash_lock);
+
+ bat_dbg(DBG_TT, bat_priv,
+ "Creating new global tt entry: %pM (via %pM)\n",
+ tt_global_entry->addr, orig_node->orig);
+
+ /* remove address from local hash if present */
+ spin_lock_bh(&bat_priv->tt_lhash_lock);
+ tt_local_entry = tt_local_hash_find(bat_priv, tt_addr);
+
+ if (tt_local_entry)
+ tt_local_del(bat_priv, tt_local_entry,
+ "global tt received");
+ spin_unlock_bh(&bat_priv->tt_lhash_lock);
+ return 1;
+unlock:
+ spin_unlock_bh(&bat_priv->tt_ghash_lock);
+ return 0;
}
int tt_global_seq_print_text(struct seq_file *seq, void *offset)
@@ -509,17 +563,20 @@ int tt_global_seq_print_text(struct seq_file *seq, void *offset)
seq_printf(seq,
"Globally announced TT entries received via the mesh %s\n",
net_dev->name);
+ seq_printf(seq, " %-13s %s %-15s %s\n",
+ "Client", "(TTVN)", "Originator", "(Curr TTVN)");
spin_lock_bh(&bat_priv->tt_ghash_lock);
buf_size = 1;
- /* Estimate length for: " * xx:xx:xx:xx:xx:xx via xx:xx:xx:xx:xx:xx\n"*/
+ /* Estimate length for: " * xx:xx:xx:xx:xx:xx (ttvn) via
+ * xx:xx:xx:xx:xx:xx (cur_ttvn)\n"*/
for (i = 0; i < hash->size; i++) {
head = &hash->table[i];
rcu_read_lock();
__hlist_for_each_rcu(node, head)
- buf_size += 43;
+ buf_size += 59;
rcu_read_unlock();
}
@@ -538,10 +595,14 @@ int tt_global_seq_print_text(struct seq_file *seq, void *offset)
rcu_read_lock();
hlist_for_each_entry_rcu(tt_global_entry, node,
head, hash_entry) {
- pos += snprintf(buff + pos, 44,
- " * %pM via %pM\n",
+ pos += snprintf(buff + pos, 61,
+ " * %pM (%3u) via %pM (%3u)\n",
tt_global_entry->addr,
- tt_global_entry->orig_node->orig);
+ tt_global_entry->ttvn,
+ tt_global_entry->orig_node->orig,
+ (uint8_t) atomic_read(
+ &tt_global_entry->orig_node->
+ last_ttvn));
}
rcu_read_unlock();
}
@@ -556,64 +617,80 @@ out:
return ret;
}
-static void _tt_global_del_orig(struct bat_priv *bat_priv,
- struct tt_global_entry *tt_global_entry,
- const char *message)
+static void _tt_global_del(struct bat_priv *bat_priv,
+ struct tt_global_entry *tt_global_entry,
+ const char *message)
{
- bat_dbg(DBG_ROUTES, bat_priv,
+ if (!tt_global_entry)
+ return;
+
+ bat_dbg(DBG_TT, bat_priv,
"Deleting global tt entry %pM (via %pM): %s\n",
tt_global_entry->addr, tt_global_entry->orig_node->orig,
message);
+ atomic_dec(&tt_global_entry->orig_node->tt_size);
hash_remove(bat_priv->tt_global_hash, compare_gtt, choose_orig,
tt_global_entry->addr);
kfree(tt_global_entry);
}
+void tt_global_del(struct bat_priv *bat_priv,
+ struct orig_node *orig_node, const unsigned char *addr,
+ const char *message)
+{
+ struct tt_global_entry *tt_global_entry;
+
+ spin_lock_bh(&bat_priv->tt_ghash_lock);
+ tt_global_entry = tt_global_hash_find(bat_priv, addr);
+
+ if (tt_global_entry && tt_global_entry->orig_node == orig_node) {
+ atomic_dec(&orig_node->tt_size);
+ _tt_global_del(bat_priv, tt_global_entry, message);
+ }
+ spin_unlock_bh(&bat_priv->tt_ghash_lock);
+}
+
void tt_global_del_orig(struct bat_priv *bat_priv,
- struct orig_node *orig_node, const char *message)
+ struct orig_node *orig_node, const char *message)
{
struct tt_global_entry *tt_global_entry;
- int tt_buff_count = 0;
- unsigned char *tt_ptr;
+ int i;
+ struct hashtable_t *hash = bat_priv->tt_global_hash;
+ struct hlist_node *node, *safe;
+ struct hlist_head *head;
- if (orig_node->tt_buff_len == 0)
+ if (!bat_priv->tt_global_hash)
return;
spin_lock_bh(&bat_priv->tt_ghash_lock);
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
- while ((tt_buff_count + 1) * ETH_ALEN <= orig_node->tt_buff_len) {
- tt_ptr = orig_node->tt_buff + (tt_buff_count * ETH_ALEN);
- tt_global_entry = tt_global_hash_find(bat_priv, tt_ptr);
-
- if ((tt_global_entry) &&
- (tt_global_entry->orig_node == orig_node))
- _tt_global_del_orig(bat_priv, tt_global_entry,
- message);
-
- tt_buff_count++;
+ hlist_for_each_entry_safe(tt_global_entry, node, safe,
+ head, hash_entry) {
+ if (tt_global_entry->orig_node == orig_node)
+ _tt_global_del(bat_priv, tt_global_entry,
+ message);
+ }
}
+ atomic_set(&orig_node->tt_size, 0);
spin_unlock_bh(&bat_priv->tt_ghash_lock);
-
- orig_node->tt_buff_len = 0;
- kfree(orig_node->tt_buff);
- orig_node->tt_buff = NULL;
}
-static void tt_global_del(struct hlist_node *node, void *arg)
+static void tt_global_entry_free(struct hlist_node *node, void *arg)
{
void *data = container_of(node, struct tt_global_entry, hash_entry);
-
kfree(data);
}
-void tt_global_free(struct bat_priv *bat_priv)
+static void tt_global_table_free(struct bat_priv *bat_priv)
{
if (!bat_priv->tt_global_hash)
return;
- hash_delete(bat_priv->tt_global_hash, tt_global_del, NULL);
+ hash_delete(bat_priv->tt_global_hash, tt_global_entry_free, NULL);
bat_priv->tt_global_hash = NULL;
}
@@ -638,3 +715,686 @@ out:
spin_unlock_bh(&bat_priv->tt_ghash_lock);
return orig_node;
}
+
+/* Calculates the checksum of the local table of a given orig_node */
+uint16_t tt_global_crc(struct bat_priv *bat_priv, struct orig_node *orig_node)
+{
+ uint16_t total = 0, total_one;
+ struct hashtable_t *hash = bat_priv->tt_global_hash;
+ struct tt_global_entry *tt_global_entry;
+ struct hlist_node *node;
+ struct hlist_head *head;
+ int i, j;
+
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(tt_global_entry, node,
+ head, hash_entry) {
+ if (compare_eth(tt_global_entry->orig_node,
+ orig_node)) {
+ total_one = 0;
+ for (j = 0; j < ETH_ALEN; j++)
+ total_one = crc16_byte(total_one,
+ tt_global_entry->addr[j]);
+ total ^= total_one;
+ }
+ }
+ rcu_read_unlock();
+ }
+
+ return total;
+}
+
+/* Calculates the checksum of the local table */
+uint16_t tt_local_crc(struct bat_priv *bat_priv)
+{
+ uint16_t total = 0, total_one;
+ struct hashtable_t *hash = bat_priv->tt_local_hash;
+ struct tt_local_entry *tt_local_entry;
+ struct hlist_node *node;
+ struct hlist_head *head;
+ int i, j;
+
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(tt_local_entry, node,
+ head, hash_entry) {
+ total_one = 0;
+ for (j = 0; j < ETH_ALEN; j++)
+ total_one = crc16_byte(total_one,
+ tt_local_entry->addr[j]);
+ total ^= total_one;
+ }
+
+ rcu_read_unlock();
+ }
+
+ return total;
+}
+
+static void tt_req_list_free(struct bat_priv *bat_priv)
+{
+ struct tt_req_node *node, *safe;
+
+ spin_lock_bh(&bat_priv->tt_req_list_lock);
+
+ list_for_each_entry_safe(node, safe, &bat_priv->tt_req_list, list) {
+ list_del(&node->list);
+ kfree(node);
+ }
+
+ spin_unlock_bh(&bat_priv->tt_req_list_lock);
+}
+
+void tt_save_orig_buffer(struct bat_priv *bat_priv, struct orig_node *orig_node,
+ const unsigned char *tt_buff, uint8_t tt_num_changes)
+{
+ uint16_t tt_buff_len = tt_len(tt_num_changes);
+
+ /* Replace the old buffer only if I received something in the
+ * last OGM (the OGM could carry no changes) */
+ spin_lock_bh(&orig_node->tt_buff_lock);
+ if (tt_buff_len > 0) {
+ kfree(orig_node->tt_buff);
+ orig_node->tt_buff_len = 0;
+ orig_node->tt_buff = kmalloc(tt_buff_len, GFP_ATOMIC);
+ if (orig_node->tt_buff) {
+ memcpy(orig_node->tt_buff, tt_buff, tt_buff_len);
+ orig_node->tt_buff_len = tt_buff_len;
+ }
+ }
+ spin_unlock_bh(&orig_node->tt_buff_lock);
+}
+
+static void tt_req_purge(struct bat_priv *bat_priv)
+{
+ struct tt_req_node *node, *safe;
+
+ spin_lock_bh(&bat_priv->tt_req_list_lock);
+ list_for_each_entry_safe(node, safe, &bat_priv->tt_req_list, list) {
+ if (is_out_of_time(node->issued_at,
+ TT_REQUEST_TIMEOUT * 1000)) {
+ list_del(&node->list);
+ kfree(node);
+ }
+ }
+ spin_unlock_bh(&bat_priv->tt_req_list_lock);
+}
+
+/* returns the pointer to the new tt_req_node struct if no request
+ * has already been issued for this orig_node, NULL otherwise */
+static struct tt_req_node *new_tt_req_node(struct bat_priv *bat_priv,
+ struct orig_node *orig_node)
+{
+ struct tt_req_node *tt_req_node_tmp, *tt_req_node = NULL;
+
+ spin_lock_bh(&bat_priv->tt_req_list_lock);
+ list_for_each_entry(tt_req_node_tmp, &bat_priv->tt_req_list, list) {
+ if (compare_eth(tt_req_node_tmp, orig_node) &&
+ !is_out_of_time(tt_req_node_tmp->issued_at,
+ TT_REQUEST_TIMEOUT * 1000))
+ goto unlock;
+ }
+
+ tt_req_node = kmalloc(sizeof(*tt_req_node), GFP_ATOMIC);
+ if (!tt_req_node)
+ goto unlock;
+
+ memcpy(tt_req_node->addr, orig_node->orig, ETH_ALEN);
+ tt_req_node->issued_at = jiffies;
+
+ list_add(&tt_req_node->list, &bat_priv->tt_req_list);
+unlock:
+ spin_unlock_bh(&bat_priv->tt_req_list_lock);
+ return tt_req_node;
+}
+
+static int tt_global_valid_entry(const void *entry_ptr, const void *data_ptr)
+{
+ const struct tt_global_entry *tt_global_entry = entry_ptr;
+ const struct orig_node *orig_node = data_ptr;
+
+ return (tt_global_entry->orig_node == orig_node);
+}
+
+static struct sk_buff *tt_response_fill_table(uint16_t tt_len, uint8_t ttvn,
+ struct hashtable_t *hash,
+ struct hard_iface *primary_if,
+ int (*valid_cb)(const void *,
+ const void *),
+ void *cb_data)
+{
+ struct tt_local_entry *tt_local_entry;
+ struct tt_query_packet *tt_response;
+ struct tt_change *tt_change;
+ struct hlist_node *node;
+ struct hlist_head *head;
+ struct sk_buff *skb = NULL;
+ uint16_t tt_tot, tt_count;
+ ssize_t tt_query_size = sizeof(struct tt_query_packet);
+ int i;
+
+ if (tt_query_size + tt_len > primary_if->soft_iface->mtu) {
+ tt_len = primary_if->soft_iface->mtu - tt_query_size;
+ tt_len -= tt_len % sizeof(struct tt_change);
+ }
+ tt_tot = tt_len / sizeof(struct tt_change);
+
+ skb = dev_alloc_skb(tt_query_size + tt_len + ETH_HLEN);
+ if (!skb)
+ goto out;
+
+ skb_reserve(skb, ETH_HLEN);
+ tt_response = (struct tt_query_packet *)skb_put(skb,
+ tt_query_size + tt_len);
+ tt_response->ttvn = ttvn;
+ tt_response->tt_data = htons(tt_tot);
+
+ tt_change = (struct tt_change *)(skb->data + tt_query_size);
+ tt_count = 0;
+
+ rcu_read_lock();
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+
+ hlist_for_each_entry_rcu(tt_local_entry, node,
+ head, hash_entry) {
+ if (tt_count == tt_tot)
+ break;
+
+ if ((valid_cb) && (!valid_cb(tt_local_entry, cb_data)))
+ continue;
+
+ memcpy(tt_change->addr, tt_local_entry->addr, ETH_ALEN);
+ tt_change->flags = NO_FLAGS;
+
+ tt_count++;
+ tt_change++;
+ }
+ }
+ rcu_read_unlock();
+
+out:
+ return skb;
+}
+
+int send_tt_request(struct bat_priv *bat_priv, struct orig_node *dst_orig_node,
+ uint8_t ttvn, uint16_t tt_crc, bool full_table)
+{
+ struct sk_buff *skb = NULL;
+ struct tt_query_packet *tt_request;
+ struct neigh_node *neigh_node = NULL;
+ struct hard_iface *primary_if;
+ struct tt_req_node *tt_req_node = NULL;
+ int ret = 1;
+
+ primary_if = primary_if_get_selected(bat_priv);
+ if (!primary_if)
+ goto out;
+
+ /* The new tt_req will be issued only if I'm not waiting for a
+ * reply from the same orig_node yet */
+ tt_req_node = new_tt_req_node(bat_priv, dst_orig_node);
+ if (!tt_req_node)
+ goto out;
+
+ skb = dev_alloc_skb(sizeof(struct tt_query_packet) + ETH_HLEN);
+ if (!skb)
+ goto out;
+
+ skb_reserve(skb, ETH_HLEN);
+
+ tt_request = (struct tt_query_packet *)skb_put(skb,
+ sizeof(struct tt_query_packet));
+
+ tt_request->packet_type = BAT_TT_QUERY;
+ tt_request->version = COMPAT_VERSION;
+ memcpy(tt_request->src, primary_if->net_dev->dev_addr, ETH_ALEN);
+ memcpy(tt_request->dst, dst_orig_node->orig, ETH_ALEN);
+ tt_request->ttl = TTL;
+ tt_request->ttvn = ttvn;
+ tt_request->tt_data = tt_crc;
+ tt_request->flags = TT_REQUEST;
+
+ if (full_table)
+ tt_request->flags |= TT_FULL_TABLE;
+
+ neigh_node = orig_node_get_router(dst_orig_node);
+ if (!neigh_node)
+ goto out;
+
+ bat_dbg(DBG_TT, bat_priv, "Sending TT_REQUEST to %pM via %pM "
+ "[%c]\n", dst_orig_node->orig, neigh_node->addr,
+ (full_table ? 'F' : '.'));
+
+ send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr);
+ ret = 0;
+
+out:
+ if (neigh_node)
+ neigh_node_free_ref(neigh_node);
+ if (primary_if)
+ hardif_free_ref(primary_if);
+ if (ret)
+ kfree_skb(skb);
+ if (ret && tt_req_node) {
+ spin_lock_bh(&bat_priv->tt_req_list_lock);
+ list_del(&tt_req_node->list);
+ spin_unlock_bh(&bat_priv->tt_req_list_lock);
+ kfree(tt_req_node);
+ }
+ return ret;
+}
+
+static bool send_other_tt_response(struct bat_priv *bat_priv,
+ struct tt_query_packet *tt_request)
+{
+ struct orig_node *req_dst_orig_node = NULL, *res_dst_orig_node = NULL;
+ struct neigh_node *neigh_node = NULL;
+ struct hard_iface *primary_if = NULL;
+ uint8_t orig_ttvn, req_ttvn, ttvn;
+ int ret = false;
+ unsigned char *tt_buff;
+ bool full_table;
+ uint16_t tt_len, tt_tot;
+ struct sk_buff *skb = NULL;
+ struct tt_query_packet *tt_response;
+
+ bat_dbg(DBG_TT, bat_priv,
+ "Received TT_REQUEST from %pM for "
+ "ttvn: %u (%pM) [%c]\n", tt_request->src,
+ tt_request->ttvn, tt_request->dst,
+ (tt_request->flags & TT_FULL_TABLE ? 'F' : '.'));
+
+ /* Let's get the orig node of the REAL destination */
+ req_dst_orig_node = get_orig_node(bat_priv, tt_request->dst);
+ if (!req_dst_orig_node)
+ goto out;
+
+ res_dst_orig_node = get_orig_node(bat_priv, tt_request->src);
+ if (!res_dst_orig_node)
+ goto out;
+
+ neigh_node = orig_node_get_router(res_dst_orig_node);
+ if (!neigh_node)
+ goto out;
+
+ primary_if = primary_if_get_selected(bat_priv);
+ if (!primary_if)
+ goto out;
+
+ orig_ttvn = (uint8_t)atomic_read(&req_dst_orig_node->last_ttvn);
+ req_ttvn = tt_request->ttvn;
+
+ /* I have not the requested data */
+ if (orig_ttvn != req_ttvn ||
+ tt_request->tt_data != req_dst_orig_node->tt_crc)
+ goto out;
+
+ /* If it has explicitly been requested the full table */
+ if (tt_request->flags & TT_FULL_TABLE ||
+ !req_dst_orig_node->tt_buff)
+ full_table = true;
+ else
+ full_table = false;
+
+ /* In this version, fragmentation is not implemented, then
+ * I'll send only one packet with as much TT entries as I can */
+ if (!full_table) {
+ spin_lock_bh(&req_dst_orig_node->tt_buff_lock);
+ tt_len = req_dst_orig_node->tt_buff_len;
+ tt_tot = tt_len / sizeof(struct tt_change);
+
+ skb = dev_alloc_skb(sizeof(struct tt_query_packet) +
+ tt_len + ETH_HLEN);
+ if (!skb)
+ goto unlock;
+
+ skb_reserve(skb, ETH_HLEN);
+ tt_response = (struct tt_query_packet *)skb_put(skb,
+ sizeof(struct tt_query_packet) + tt_len);
+ tt_response->ttvn = req_ttvn;
+ tt_response->tt_data = htons(tt_tot);
+
+ tt_buff = skb->data + sizeof(struct tt_query_packet);
+ /* Copy the last orig_node's OGM buffer */
+ memcpy(tt_buff, req_dst_orig_node->tt_buff,
+ req_dst_orig_node->tt_buff_len);
+
+ spin_unlock_bh(&req_dst_orig_node->tt_buff_lock);
+ } else {
+ tt_len = (uint16_t)atomic_read(&req_dst_orig_node->tt_size) *
+ sizeof(struct tt_change);
+ ttvn = (uint8_t)atomic_read(&req_dst_orig_node->last_ttvn);
+
+ skb = tt_response_fill_table(tt_len, ttvn,
+ bat_priv->tt_global_hash,
+ primary_if, tt_global_valid_entry,
+ req_dst_orig_node);
+ if (!skb)
+ goto out;
+
+ tt_response = (struct tt_query_packet *)skb->data;
+ }
+
+ tt_response->packet_type = BAT_TT_QUERY;
+ tt_response->version = COMPAT_VERSION;
+ tt_response->ttl = TTL;
+ memcpy(tt_response->src, req_dst_orig_node->orig, ETH_ALEN);
+ memcpy(tt_response->dst, tt_request->src, ETH_ALEN);
+ tt_response->flags = TT_RESPONSE;
+
+ if (full_table)
+ tt_response->flags |= TT_FULL_TABLE;
+
+ bat_dbg(DBG_TT, bat_priv,
+ "Sending TT_RESPONSE %pM via %pM for %pM (ttvn: %u)\n",
+ res_dst_orig_node->orig, neigh_node->addr,
+ req_dst_orig_node->orig, req_ttvn);
+
+ send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr);
+ ret = true;
+ goto out;
+
+unlock:
+ spin_unlock_bh(&req_dst_orig_node->tt_buff_lock);
+
+out:
+ if (res_dst_orig_node)
+ orig_node_free_ref(res_dst_orig_node);
+ if (req_dst_orig_node)
+ orig_node_free_ref(req_dst_orig_node);
+ if (neigh_node)
+ neigh_node_free_ref(neigh_node);
+ if (primary_if)
+ hardif_free_ref(primary_if);
+ if (!ret)
+ kfree_skb(skb);
+ return ret;
+
+}
+static bool send_my_tt_response(struct bat_priv *bat_priv,
+ struct tt_query_packet *tt_request)
+{
+ struct orig_node *orig_node = NULL;
+ struct neigh_node *neigh_node = NULL;
+ struct hard_iface *primary_if = NULL;
+ uint8_t my_ttvn, req_ttvn, ttvn;
+ int ret = false;
+ unsigned char *tt_buff;
+ bool full_table;
+ uint16_t tt_len, tt_tot;
+ struct sk_buff *skb = NULL;
+ struct tt_query_packet *tt_response;
+
+ bat_dbg(DBG_TT, bat_priv,
+ "Received TT_REQUEST from %pM for "
+ "ttvn: %u (me) [%c]\n", tt_request->src,
+ tt_request->ttvn,
+ (tt_request->flags & TT_FULL_TABLE ? 'F' : '.'));
+
+
+ my_ttvn = (uint8_t)atomic_read(&bat_priv->ttvn);
+ req_ttvn = tt_request->ttvn;
+
+ orig_node = get_orig_node(bat_priv, tt_request->src);
+ if (!orig_node)
+ goto out;
+
+ neigh_node = orig_node_get_router(orig_node);
+ if (!neigh_node)
+ goto out;
+
+ primary_if = primary_if_get_selected(bat_priv);
+ if (!primary_if)
+ goto out;
+
+ /* If the full table has been explicitly requested or the gap
+ * is too big send the whole local translation table */
+ if (tt_request->flags & TT_FULL_TABLE || my_ttvn != req_ttvn ||
+ !bat_priv->tt_buff)
+ full_table = true;
+ else
+ full_table = false;
+
+ /* In this version, fragmentation is not implemented, then
+ * I'll send only one packet with as much TT entries as I can */
+ if (!full_table) {
+ spin_lock_bh(&bat_priv->tt_buff_lock);
+ tt_len = bat_priv->tt_buff_len;
+ tt_tot = tt_len / sizeof(struct tt_change);
+
+ skb = dev_alloc_skb(sizeof(struct tt_query_packet) +
+ tt_len + ETH_HLEN);
+ if (!skb)
+ goto unlock;
+
+ skb_reserve(skb, ETH_HLEN);
+ tt_response = (struct tt_query_packet *)skb_put(skb,
+ sizeof(struct tt_query_packet) + tt_len);
+ tt_response->ttvn = req_ttvn;
+ tt_response->tt_data = htons(tt_tot);
+
+ tt_buff = skb->data + sizeof(struct tt_query_packet);
+ memcpy(tt_buff, bat_priv->tt_buff,
+ bat_priv->tt_buff_len);
+ spin_unlock_bh(&bat_priv->tt_buff_lock);
+ } else {
+ tt_len = (uint16_t)atomic_read(&bat_priv->num_local_tt) *
+ sizeof(struct tt_change);
+ ttvn = (uint8_t)atomic_read(&bat_priv->ttvn);
+
+ skb = tt_response_fill_table(tt_len, ttvn,
+ bat_priv->tt_local_hash,
+ primary_if, NULL, NULL);
+ if (!skb)
+ goto out;
+
+ tt_response = (struct tt_query_packet *)skb->data;
+ }
+
+ tt_response->packet_type = BAT_TT_QUERY;
+ tt_response->version = COMPAT_VERSION;
+ tt_response->ttl = TTL;
+ memcpy(tt_response->src, primary_if->net_dev->dev_addr, ETH_ALEN);
+ memcpy(tt_response->dst, tt_request->src, ETH_ALEN);
+ tt_response->flags = TT_RESPONSE;
+
+ if (full_table)
+ tt_response->flags |= TT_FULL_TABLE;
+
+ bat_dbg(DBG_TT, bat_priv,
+ "Sending TT_RESPONSE to %pM via %pM [%c]\n",
+ orig_node->orig, neigh_node->addr,
+ (tt_response->flags & TT_FULL_TABLE ? 'F' : '.'));
+
+ send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr);
+ ret = true;
+ goto out;
+
+unlock:
+ spin_unlock_bh(&bat_priv->tt_buff_lock);
+out:
+ if (orig_node)
+ orig_node_free_ref(orig_node);
+ if (neigh_node)
+ neigh_node_free_ref(neigh_node);
+ if (primary_if)
+ hardif_free_ref(primary_if);
+ if (!ret)
+ kfree_skb(skb);
+ /* This packet was for me, so it doesn't need to be re-routed */
+ return true;
+}
+
+bool send_tt_response(struct bat_priv *bat_priv,
+ struct tt_query_packet *tt_request)
+{
+ if (is_my_mac(tt_request->dst))
+ return send_my_tt_response(bat_priv, tt_request);
+ else
+ return send_other_tt_response(bat_priv, tt_request);
+}
+
+static void _tt_update_changes(struct bat_priv *bat_priv,
+ struct orig_node *orig_node,
+ struct tt_change *tt_change,
+ uint16_t tt_num_changes, uint8_t ttvn)
+{
+ int i;
+
+ for (i = 0; i < tt_num_changes; i++) {
+ if ((tt_change + i)->flags & TT_CHANGE_DEL)
+ tt_global_del(bat_priv, orig_node,
+ (tt_change + i)->addr,
+ "tt removed by changes");
+ else
+ if (!tt_global_add(bat_priv, orig_node,
+ (tt_change + i)->addr, ttvn))
+ /* In case of problem while storing a
+ * global_entry, we stop the updating
+ * procedure without committing the
+ * ttvn change. This will avoid to send
+ * corrupted data on tt_request
+ */
+ return;
+ }
+}
+
+static void tt_fill_gtable(struct bat_priv *bat_priv,
+ struct tt_query_packet *tt_response)
+{
+ struct orig_node *orig_node = NULL;
+
+ orig_node = orig_hash_find(bat_priv, tt_response->src);
+ if (!orig_node)
+ goto out;
+
+ /* Purge the old table first.. */
+ tt_global_del_orig(bat_priv, orig_node, "Received full table");
+
+ _tt_update_changes(bat_priv, orig_node,
+ (struct tt_change *)(tt_response + 1),
+ tt_response->tt_data, tt_response->ttvn);
+
+ spin_lock_bh(&orig_node->tt_buff_lock);
+ kfree(orig_node->tt_buff);
+ orig_node->tt_buff_len = 0;
+ orig_node->tt_buff = NULL;
+ spin_unlock_bh(&orig_node->tt_buff_lock);
+
+ atomic_set(&orig_node->last_ttvn, tt_response->ttvn);
+
+out:
+ if (orig_node)
+ orig_node_free_ref(orig_node);
+}
+
+void tt_update_changes(struct bat_priv *bat_priv, struct orig_node *orig_node,
+ uint16_t tt_num_changes, uint8_t ttvn,
+ struct tt_change *tt_change)
+{
+ _tt_update_changes(bat_priv, orig_node, tt_change, tt_num_changes,
+ ttvn);
+
+ tt_save_orig_buffer(bat_priv, orig_node, (unsigned char *)tt_change,
+ tt_num_changes);
+ atomic_set(&orig_node->last_ttvn, ttvn);
+}
+
+bool is_my_client(struct bat_priv *bat_priv, const uint8_t *addr)
+{
+ struct tt_local_entry *tt_local_entry;
+
+ spin_lock_bh(&bat_priv->tt_lhash_lock);
+ tt_local_entry = tt_local_hash_find(bat_priv, addr);
+ spin_unlock_bh(&bat_priv->tt_lhash_lock);
+
+ if (tt_local_entry)
+ return true;
+ return false;
+}
+
+void handle_tt_response(struct bat_priv *bat_priv,
+ struct tt_query_packet *tt_response)
+{
+ struct tt_req_node *node, *safe;
+ struct orig_node *orig_node = NULL;
+
+ bat_dbg(DBG_TT, bat_priv, "Received TT_RESPONSE from %pM for "
+ "ttvn %d t_size: %d [%c]\n",
+ tt_response->src, tt_response->ttvn,
+ tt_response->tt_data,
+ (tt_response->flags & TT_FULL_TABLE ? 'F' : '.'));
+
+ orig_node = orig_hash_find(bat_priv, tt_response->src);
+ if (!orig_node)
+ goto out;
+
+ if (tt_response->flags & TT_FULL_TABLE)
+ tt_fill_gtable(bat_priv, tt_response);
+ else
+ tt_update_changes(bat_priv, orig_node, tt_response->tt_data,
+ tt_response->ttvn,
+ (struct tt_change *)(tt_response + 1));
+
+ /* Delete the tt_req_node from pending tt_requests list */
+ spin_lock_bh(&bat_priv->tt_req_list_lock);
+ list_for_each_entry_safe(node, safe, &bat_priv->tt_req_list, list) {
+ if (!compare_eth(node->addr, tt_response->src))
+ continue;
+ list_del(&node->list);
+ kfree(node);
+ }
+ spin_unlock_bh(&bat_priv->tt_req_list_lock);
+
+ /* Recalculate the CRC for this orig_node and store it */
+ spin_lock_bh(&bat_priv->tt_ghash_lock);
+ orig_node->tt_crc = tt_global_crc(bat_priv, orig_node);
+ spin_unlock_bh(&bat_priv->tt_ghash_lock);
+out:
+ if (orig_node)
+ orig_node_free_ref(orig_node);
+}
+
+int tt_init(struct bat_priv *bat_priv)
+{
+ if (!tt_local_init(bat_priv))
+ return 0;
+
+ if (!tt_global_init(bat_priv))
+ return 0;
+
+ tt_start_timer(bat_priv);
+
+ return 1;
+}
+
+void tt_free(struct bat_priv *bat_priv)
+{
+ cancel_delayed_work_sync(&bat_priv->tt_work);
+
+ tt_local_table_free(bat_priv);
+ tt_global_table_free(bat_priv);
+ tt_req_list_free(bat_priv);
+ tt_changes_list_free(bat_priv);
+
+ kfree(bat_priv->tt_buff);
+}
+
+static void tt_purge(struct work_struct *work)
+{
+ struct delayed_work *delayed_work =
+ container_of(work, struct delayed_work, work);
+ struct bat_priv *bat_priv =
+ container_of(delayed_work, struct bat_priv, tt_work);
+
+ tt_local_purge(bat_priv);
+ tt_req_purge(bat_priv);
+
+ tt_start_timer(bat_priv);
+}
diff --git a/net/batman-adv/translation-table.h b/net/batman-adv/translation-table.h
index 0f2b990..51f7e30 100644
--- a/net/batman-adv/translation-table.h
+++ b/net/batman-adv/translation-table.h
@@ -22,23 +22,43 @@
#ifndef _NET_BATMAN_ADV_TRANSLATION_TABLE_H_
#define _NET_BATMAN_ADV_TRANSLATION_TABLE_H_
-int tt_local_init(struct bat_priv *bat_priv);
+int tt_len(int changes_num);
+int tt_changes_fill_buffer(struct bat_priv *bat_priv,
+ unsigned char *buff, int buff_len);
+int tt_init(struct bat_priv *bat_priv);
void tt_local_add(struct net_device *soft_iface, const uint8_t *addr);
void tt_local_remove(struct bat_priv *bat_priv,
const uint8_t *addr, const char *message);
-int tt_local_fill_buffer(struct bat_priv *bat_priv,
- unsigned char *buff, int buff_len);
int tt_local_seq_print_text(struct seq_file *seq, void *offset);
-void tt_local_free(struct bat_priv *bat_priv);
-int tt_global_init(struct bat_priv *bat_priv);
void tt_global_add_orig(struct bat_priv *bat_priv,
struct orig_node *orig_node,
const unsigned char *tt_buff, int tt_buff_len);
+int tt_global_add(struct bat_priv *bat_priv,
+ struct orig_node *orig_node, const unsigned char *addr,
+ uint8_t ttvn);
int tt_global_seq_print_text(struct seq_file *seq, void *offset);
void tt_global_del_orig(struct bat_priv *bat_priv,
- struct orig_node *orig_node, const char *message);
-void tt_global_free(struct bat_priv *bat_priv);
+ struct orig_node *orig_node, const char *message);
+void tt_global_del(struct bat_priv *bat_priv,
+ struct orig_node *orig_node, const unsigned char *addr,
+ const char *message);
struct orig_node *transtable_search(struct bat_priv *bat_priv,
const uint8_t *addr);
+void tt_save_orig_buffer(struct bat_priv *bat_priv, struct orig_node *orig_node,
+ const unsigned char *tt_buff, uint8_t tt_num_changes);
+uint16_t tt_local_crc(struct bat_priv *bat_priv);
+uint16_t tt_global_crc(struct bat_priv *bat_priv, struct orig_node *orig_node);
+void tt_free(struct bat_priv *bat_priv);
+int send_tt_request(struct bat_priv *bat_priv,
+ struct orig_node *dst_orig_node, uint8_t hvn,
+ uint16_t tt_crc, bool full_table);
+bool send_tt_response(struct bat_priv *bat_priv,
+ struct tt_query_packet *tt_request);
+void tt_update_changes(struct bat_priv *bat_priv, struct orig_node *orig_node,
+ uint16_t tt_num_changes, uint8_t ttvn,
+ struct tt_change *tt_change);
+bool is_my_client(struct bat_priv *bat_priv, const uint8_t *addr);
+void handle_tt_response(struct bat_priv *bat_priv,
+ struct tt_query_packet *tt_response);
#endif /* _NET_BATMAN_ADV_TRANSLATION_TABLE_H_ */
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 65b3222..3b642a9 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -75,8 +75,12 @@ struct orig_node {
unsigned long batman_seqno_reset;
uint8_t gw_flags;
uint8_t flags;
+ atomic_t last_ttvn; /* last seen translation table version number */
+ uint16_t tt_crc;
unsigned char *tt_buff;
int16_t tt_buff_len;
+ spinlock_t tt_buff_lock; /* protects tt_buff */
+ atomic_t tt_size;
uint32_t last_real_seqno;
uint8_t last_ttl;
unsigned long bcast_bits[NUM_WORDS];
@@ -94,6 +98,7 @@ struct orig_node {
spinlock_t ogm_cnt_lock;
/* bcast_seqno_lock protects bcast_bits, last_bcast_seqno */
spinlock_t bcast_seqno_lock;
+ spinlock_t tt_list_lock; /* protects tt_list */
atomic_t bond_candidates;
struct list_head bond_list;
};
@@ -145,6 +150,9 @@ struct bat_priv {
atomic_t bcast_seqno;
atomic_t bcast_queue_left;
atomic_t batman_queue_left;
+ atomic_t ttvn; /* tranlation table version number */
+ atomic_t tt_ogm_append_cnt;
+ atomic_t tt_local_changes; /* changes registered in a OGM interval */
char num_ifaces;
struct debug_log *debug_log;
struct kobject *mesh_obj;
@@ -153,22 +161,30 @@ struct bat_priv {
struct hlist_head forw_bcast_list;
struct hlist_head gw_list;
struct hlist_head softif_neigh_vids;
+ struct list_head tt_changes_list; /* tracks changes in a OGM int */
struct list_head vis_send_list;
struct hashtable_t *orig_hash;
struct hashtable_t *tt_local_hash;
struct hashtable_t *tt_global_hash;
+ struct list_head tt_req_list; /* list of pending tt_requests */
struct hashtable_t *vis_hash;
spinlock_t forw_bat_list_lock; /* protects forw_bat_list */
spinlock_t forw_bcast_list_lock; /* protects */
+ spinlock_t tt_changes_list_lock; /* protects tt_changes */
spinlock_t tt_lhash_lock; /* protects tt_local_hash */
spinlock_t tt_ghash_lock; /* protects tt_global_hash */
+ spinlock_t tt_req_list_lock; /* protects tt_req_list */
spinlock_t gw_list_lock; /* protects gw_list and curr_gw */
spinlock_t vis_hash_lock; /* protects vis_hash */
spinlock_t vis_list_lock; /* protects vis_info::recv_list */
spinlock_t softif_neigh_lock; /* protects soft-interface neigh list */
spinlock_t softif_neigh_vid_lock; /* protects soft-interface vid list */
- int16_t num_local_tt;
- atomic_t tt_local_changed;
+ atomic_t num_local_tt;
+ /* Checksum of the local table, recomputed before sending a new OGM */
+ atomic_t tt_crc;
+ unsigned char *tt_buff;
+ int16_t tt_buff_len;
+ spinlock_t tt_buff_lock; /* protects tt_buff */
struct delayed_work tt_work;
struct delayed_work orig_work;
struct delayed_work vis_work;
@@ -202,9 +218,22 @@ struct tt_local_entry {
struct tt_global_entry {
uint8_t addr[ETH_ALEN];
struct orig_node *orig_node;
+ uint8_t ttvn;
+ /* entry in the global table */
struct hlist_node hash_entry;
};
+struct tt_change_node {
+ struct list_head list;
+ struct tt_change change;
+};
+
+struct tt_req_node {
+ uint8_t addr[ETH_ALEN];
+ unsigned long issued_at;
+ struct list_head list;
+};
+
/**
* forw_packet - structure for forw_list maintaining packets to be
* send/forwarded
diff --git a/net/batman-adv/unicast.c b/net/batman-adv/unicast.c
index 6eabf42..32b125f 100644
--- a/net/batman-adv/unicast.c
+++ b/net/batman-adv/unicast.c
@@ -325,6 +325,9 @@ find_router:
unicast_packet->ttl = TTL;
/* copy the destination for faster routing */
memcpy(unicast_packet->dest, orig_node->orig, ETH_ALEN);
+ /* set the destination tt version number */
+ unicast_packet->ttvn =
+ (uint8_t)atomic_read(&orig_node->last_ttvn);
if (atomic_read(&bat_priv->fragmentation) &&
data_len + sizeof(*unicast_packet) >
--
1.7.5.3
^ permalink raw reply related
* [PATCH 04/12] batman-adv: Reduce usage of char
From: Sven Eckelmann @ 2011-06-20 10:16 UTC (permalink / raw)
To: davem-fT/PcQaiUtIeIZ0/mPfg9Q
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
b.a.t.m.a.n-ZwoEplunGu2X36UT3dwllkB+6BGkLq7r
In-Reply-To: <1308565025-21293-1-git-send-email-sven-KaDOiPu9UxWEi8DpZVb4nw@public.gmane.org>
char was used in different places to store information without really
using the characteristics of that data type or by ignoring the fact that
char has not a well defined signedness.
Signed-off-by: Sven Eckelmann <sven-KaDOiPu9UxWEi8DpZVb4nw@public.gmane.org>
---
net/batman-adv/aggregation.c | 2 +-
net/batman-adv/aggregation.h | 2 +-
net/batman-adv/bitarray.c | 4 ++--
net/batman-adv/bitarray.h | 4 ++--
net/batman-adv/gateway_client.c | 2 +-
net/batman-adv/gateway_common.c | 6 +++---
net/batman-adv/main.h | 2 +-
net/batman-adv/routing.c | 8 ++++----
net/batman-adv/send.c | 6 +++---
net/batman-adv/send.h | 2 +-
net/batman-adv/soft-interface.c | 2 +-
11 files changed, 20 insertions(+), 20 deletions(-)
diff --git a/net/batman-adv/aggregation.c b/net/batman-adv/aggregation.c
index 4080970..29c6740 100644
--- a/net/batman-adv/aggregation.c
+++ b/net/batman-adv/aggregation.c
@@ -195,7 +195,7 @@ static void aggregate(struct forw_packet *forw_packet_aggr,
void add_bat_packet_to_list(struct bat_priv *bat_priv,
unsigned char *packet_buff, int packet_len,
- struct hard_iface *if_incoming, char own_packet,
+ struct hard_iface *if_incoming, int own_packet,
unsigned long send_time)
{
/**
diff --git a/net/batman-adv/aggregation.h b/net/batman-adv/aggregation.h
index fedeb8d..0547fd8 100644
--- a/net/batman-adv/aggregation.h
+++ b/net/batman-adv/aggregation.h
@@ -35,7 +35,7 @@ static inline int aggregated_packet(int buff_pos, int packet_len, int num_tt)
void add_bat_packet_to_list(struct bat_priv *bat_priv,
unsigned char *packet_buff, int packet_len,
- struct hard_iface *if_incoming, char own_packet,
+ struct hard_iface *if_incoming, int own_packet,
unsigned long send_time);
void receive_aggr_bat_packet(const struct ethhdr *ethhdr,
unsigned char *packet_buff, int packet_len,
diff --git a/net/batman-adv/bitarray.c b/net/batman-adv/bitarray.c
index 3659a25..c1f4bfc 100644
--- a/net/batman-adv/bitarray.c
+++ b/net/batman-adv/bitarray.c
@@ -127,8 +127,8 @@ static void bit_reset_window(unsigned long *seq_bits)
* 1 if the window was moved (either new or very old)
* 0 if the window was not moved/shifted.
*/
-char bit_get_packet(void *priv, unsigned long *seq_bits,
- int32_t seq_num_diff, int8_t set_mark)
+int bit_get_packet(void *priv, unsigned long *seq_bits,
+ int32_t seq_num_diff, int set_mark)
{
struct bat_priv *bat_priv = priv;
diff --git a/net/batman-adv/bitarray.h b/net/batman-adv/bitarray.h
index 277c037..9c04422 100644
--- a/net/batman-adv/bitarray.h
+++ b/net/batman-adv/bitarray.h
@@ -35,8 +35,8 @@ void bit_mark(unsigned long *seq_bits, int32_t n);
/* receive and process one packet, returns 1 if received seq_num is considered
* new, 0 if old */
-char bit_get_packet(void *priv, unsigned long *seq_bits,
- int32_t seq_num_diff, int8_t set_mark);
+int bit_get_packet(void *priv, unsigned long *seq_bits,
+ int32_t seq_num_diff, int set_mark);
/* count the hamming weight, how many good packets did we receive? */
int bit_packet_count(const unsigned long *seq_bits);
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 24aee56..7248de2 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -360,7 +360,7 @@ void gw_node_purge(struct bat_priv *bat_priv)
struct gw_node *gw_node, *curr_gw;
struct hlist_node *node, *node_tmp;
unsigned long timeout = 2 * PURGE_TIMEOUT * HZ;
- char do_deselect = 0;
+ int do_deselect = 0;
curr_gw = gw_get_selected_gw_node(bat_priv);
diff --git a/net/batman-adv/gateway_common.c b/net/batman-adv/gateway_common.c
index e74307b..18661af 100644
--- a/net/batman-adv/gateway_common.c
+++ b/net/batman-adv/gateway_common.c
@@ -61,9 +61,9 @@ static void kbit_to_gw_bandwidth(int down, int up, long *gw_srv_class)
/* returns the up and downspeeds in kbit, calculated from the class */
void gw_bandwidth_to_kbit(uint8_t gw_srv_class, int *down, int *up)
{
- char sbit = (gw_srv_class & 0x80) >> 7;
- char dpart = (gw_srv_class & 0x78) >> 3;
- char upart = (gw_srv_class & 0x07);
+ int sbit = (gw_srv_class & 0x80) >> 7;
+ int dpart = (gw_srv_class & 0x78) >> 3;
+ int upart = (gw_srv_class & 0x07);
if (!gw_srv_class) {
*down = 0;
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index ed488cb..714a241 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -151,7 +151,7 @@ int debug_log(struct bat_priv *bat_priv, const char *fmt, ...) __printf(2, 3);
while (0)
#else /* !CONFIG_BATMAN_ADV_DEBUG */
__printf(3, 4)
-static inline void bat_dbg(char type __always_unused,
+static inline void bat_dbg(int type __always_unused,
struct bat_priv *bat_priv __always_unused,
const char *fmt __always_unused, ...)
{
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index d5ce644..eb6fb7d 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -163,7 +163,7 @@ static int is_bidirectional_neigh(struct orig_node *orig_node,
struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
struct neigh_node *neigh_node = NULL, *tmp_neigh_node;
struct hlist_node *node;
- unsigned char total_count;
+ uint8_t total_count;
uint8_t orig_eq_count, neigh_rq_count, tq_own;
int tq_asym_penalty, ret = 0;
@@ -602,9 +602,9 @@ void receive_bat_packet(const struct ethhdr *ethhdr,
struct orig_node *orig_neigh_node, *orig_node;
struct neigh_node *router = NULL, *router_router = NULL;
struct neigh_node *orig_neigh_router = NULL;
- char has_directlink_flag;
- char is_my_addr = 0, is_my_orig = 0, is_my_oldorig = 0;
- char is_broadcast = 0, is_bidirectional, is_single_hop_neigh;
+ int has_directlink_flag;
+ int is_my_addr = 0, is_my_orig = 0, is_my_oldorig = 0;
+ int is_broadcast = 0, is_bidirectional, is_single_hop_neigh;
int is_duplicate;
uint32_t if_incoming_seqno;
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index a1b8c31..be0d581 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -165,7 +165,7 @@ static void send_packet(struct forw_packet *forw_packet)
struct bat_priv *bat_priv;
struct batman_packet *batman_packet =
(struct batman_packet *)(forw_packet->skb->data);
- unsigned char directlink = (batman_packet->flags & DIRECTLINK ? 1 : 0);
+ int directlink = (batman_packet->flags & DIRECTLINK ? 1 : 0);
if (!forw_packet->if_incoming) {
pr_err("Error - can't forward packet: incoming iface not "
@@ -307,12 +307,12 @@ void schedule_own_packet(struct hard_iface *hard_iface)
void schedule_forward_packet(struct orig_node *orig_node,
const struct ethhdr *ethhdr,
struct batman_packet *batman_packet,
- uint8_t directlink, int tt_buff_len,
+ int directlink, int tt_buff_len,
struct hard_iface *if_incoming)
{
struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
struct neigh_node *router;
- unsigned char in_tq, in_ttl, tq_avg = 0;
+ uint8_t in_tq, in_ttl, tq_avg = 0;
unsigned long send_time;
if (batman_packet->ttl <= 1) {
diff --git a/net/batman-adv/send.h b/net/batman-adv/send.h
index eceab87..6d9c14d 100644
--- a/net/batman-adv/send.h
+++ b/net/batman-adv/send.h
@@ -28,7 +28,7 @@ void schedule_own_packet(struct hard_iface *hard_iface);
void schedule_forward_packet(struct orig_node *orig_node,
const struct ethhdr *ethhdr,
struct batman_packet *batman_packet,
- uint8_t directlink, int tt_buff_len,
+ int directlink, int tt_buff_len,
struct hard_iface *if_outgoing);
int add_bcast_packet_to_list(struct bat_priv *bat_priv,
const struct sk_buff *skb);
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 0fc997e..69c0022 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -380,7 +380,7 @@ void softif_neigh_purge(struct bat_priv *bat_priv)
struct softif_neigh *softif_neigh, *curr_softif_neigh;
struct softif_neigh_vid *softif_neigh_vid;
struct hlist_node *node, *node_tmp, *node_tmp2;
- char do_deselect;
+ int do_deselect;
rcu_read_lock();
hlist_for_each_entry_rcu(softif_neigh_vid, node,
--
1.7.5.3
^ permalink raw reply related
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