* [PATCH 1/7] Preparatory refactoring part 1.
From: Corey Hickey @ 2007-07-30 0:21 UTC (permalink / raw)
To: netdev; +Cc: Corey Hickey
In-Reply-To: <11857548771998-git-send-email-bugfood-ml@fatooh.org>
Make a new function sfq_q_enqueue() that operates directly on the
queue data. This will be useful for implementing sfq_change() in
a later patch. A pleasant side-effect is reducing most of the
duplicate code in sfq_enqueue() and sfq_requeue().
Similarly, make a new function sfq_q_dequeue().
Signed-off-by: Corey Hickey <bugfood-ml@fatooh.org>
---
net/sched/sch_sfq.c | 70 ++++++++++++++++++++++++++------------------------
1 files changed, 36 insertions(+), 34 deletions(-)
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 9579573..8ae077f 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -77,6 +77,9 @@
#define SFQ_DEPTH 128
#define SFQ_HASH_DIVISOR 1024
+#define SFQ_HEAD 0
+#define SFQ_TAIL 1
+
/* This type should contain at least SFQ_DEPTH*2 values */
typedef unsigned char sfq_index;
@@ -244,10 +247,8 @@ static unsigned int sfq_drop(struct Qdisc *sch)
return 0;
}
-static int
-sfq_enqueue(struct sk_buff *skb, struct Qdisc* sch)
+static void sfq_q_enqueue(struct sk_buff *skb, struct sfq_sched_data *q, unsigned int end)
{
- struct sfq_sched_data *q = qdisc_priv(sch);
unsigned hash = sfq_hash(q, skb);
sfq_index x;
@@ -256,8 +257,12 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc* sch)
q->ht[hash] = x = q->dep[SFQ_DEPTH].next;
q->hash[x] = hash;
}
- sch->qstats.backlog += skb->len;
- __skb_queue_tail(&q->qs[x], skb);
+
+ if (end == SFQ_TAIL)
+ __skb_queue_tail(&q->qs[x], skb);
+ else
+ __skb_queue_head(&q->qs[x], skb);
+
sfq_inc(q, x);
if (q->qs[x].qlen == 1) { /* The flow is new */
if (q->tail == SFQ_DEPTH) { /* It is the first flow */
@@ -270,12 +275,21 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc* sch)
q->tail = x;
}
}
+}
+
+static int
+sfq_enqueue(struct sk_buff *skb, struct Qdisc* sch)
+{
+ struct sfq_sched_data *q = qdisc_priv(sch);
+ sfq_q_enqueue(skb, q, SFQ_TAIL);
+ sch->qstats.backlog += skb->len;
if (++sch->q.qlen < q->limit-1) {
sch->bstats.bytes += skb->len;
sch->bstats.packets++;
return 0;
}
+ sch->qstats.drops++;
sfq_drop(sch);
return NET_XMIT_CN;
}
@@ -284,28 +298,8 @@ static int
sfq_requeue(struct sk_buff *skb, struct Qdisc* sch)
{
struct sfq_sched_data *q = qdisc_priv(sch);
- unsigned hash = sfq_hash(q, skb);
- sfq_index x;
-
- x = q->ht[hash];
- if (x == SFQ_DEPTH) {
- q->ht[hash] = x = q->dep[SFQ_DEPTH].next;
- q->hash[x] = hash;
- }
+ sfq_q_enqueue(skb, q, SFQ_HEAD);
sch->qstats.backlog += skb->len;
- __skb_queue_head(&q->qs[x], skb);
- sfq_inc(q, x);
- if (q->qs[x].qlen == 1) { /* The flow is new */
- if (q->tail == SFQ_DEPTH) { /* It is the first flow */
- q->tail = x;
- q->next[x] = x;
- q->allot[x] = q->quantum;
- } else {
- q->next[x] = q->next[q->tail];
- q->next[q->tail] = x;
- q->tail = x;
- }
- }
if (++sch->q.qlen < q->limit - 1) {
sch->qstats.requeues++;
return 0;
@@ -316,13 +310,8 @@ sfq_requeue(struct sk_buff *skb, struct Qdisc* sch)
return NET_XMIT_CN;
}
-
-
-
-static struct sk_buff *
-sfq_dequeue(struct Qdisc* sch)
+static struct sk_buff *sfq_q_dequeue(struct sfq_sched_data *q)
{
- struct sfq_sched_data *q = qdisc_priv(sch);
struct sk_buff *skb;
sfq_index a, old_a;
@@ -335,8 +324,6 @@ sfq_dequeue(struct Qdisc* sch)
/* Grab packet */
skb = __skb_dequeue(&q->qs[a]);
sfq_dec(q, a);
- sch->q.qlen--;
- sch->qstats.backlog -= skb->len;
/* Is the slot empty? */
if (q->qs[a].qlen == 0) {
@@ -353,6 +340,21 @@ sfq_dequeue(struct Qdisc* sch)
a = q->next[a];
q->allot[a] += q->quantum;
}
+
+ return skb;
+}
+
+static struct sk_buff
+*sfq_dequeue(struct Qdisc* sch)
+{
+ struct sfq_sched_data *q = qdisc_priv(sch);
+ struct sk_buff *skb;
+
+ skb = sfq_q_dequeue(q);
+ if (skb == NULL)
+ return NULL;
+ sch->q.qlen--;
+ sch->qstats.backlog -= skb->len;
return skb;
}
--
1.5.2.4
^ permalink raw reply related
* [PATCH 5/7] Add divisor.
From: Corey Hickey @ 2007-07-30 0:21 UTC (permalink / raw)
To: netdev; +Cc: Corey Hickey
In-Reply-To: <11857548771998-git-send-email-bugfood-ml@fatooh.org>
Make hash divisor user-configurable.
Signed-off-by: Corey Hickey <bugfood-ml@fatooh.org>
---
net/sched/sch_sfq.c | 17 ++++++++++++-----
1 files changed, 12 insertions(+), 5 deletions(-)
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 70124ac..e6a6a21 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -75,7 +75,7 @@
It is easy to increase these values, but not in flight. */
#define SFQ_DEPTH_DEFAULT 128
-#define SFQ_HASH_DIVISOR 1024
+#define SFQ_DIVISOR_DEFAULT 1024
#define SFQ_HEAD 0
#define SFQ_TAIL 1
@@ -98,6 +98,7 @@ struct sfq_sched_data
unsigned quantum; /* Allotment per round: MUST BE >= MTU */
int limit;
unsigned depth;
+ unsigned hash_divisor;
/* Variables */
struct timer_list perturb_timer;
@@ -105,7 +106,7 @@ struct sfq_sched_data
sfq_index tail; /* Index of current slot in round */
sfq_index max_depth; /* Maximal depth */
- sfq_index ht[SFQ_HASH_DIVISOR]; /* Hash table */
+ sfq_index *ht; /* Hash table */
sfq_index *next; /* Active slots link */
short *allot; /* Current allotment per slot */
unsigned short *hash; /* Hash value indexed by slots */
@@ -120,7 +121,7 @@ static __inline__ unsigned sfq_fold_hash(struct sfq_sched_data *q, u32 h, u32 h1
/* Have we any rotation primitives? If not, WHY? */
h ^= (h1<<pert) ^ (h1>>(0x1F - pert));
h ^= h>>10;
- return h & 0x3FF;
+ return h & (q->hash_divisor-1);
}
static unsigned sfq_hash(struct sfq_sched_data *q, struct sk_buff *skb)
@@ -386,6 +387,7 @@ static void sfq_perturbation(unsigned long arg)
static void sfq_q_destroy(struct sfq_sched_data *q)
{
del_timer(&q->perturb_timer);
+ kfree(q->ht);
kfree(q->dep);
kfree(q->next);
kfree(q->allot);
@@ -411,12 +413,14 @@ static int sfq_q_init(struct sfq_sched_data *q, struct rtattr *opt)
q->max_depth = 0;
if (opt == NULL) {
q->perturb_period = 0;
+ q->hash_divisor = SFQ_DIVISOR_DEFAULT;
q->tail = q->limit = q->depth = SFQ_DEPTH_DEFAULT;
} else {
struct tc_sfq_qopt *ctl = RTA_DATA(opt);
if (ctl->quantum)
q->quantum = ctl->quantum;
q->perturb_period = ctl->perturb_period*HZ;
+ q->hash_divisor = ctl->divisor ? : SFQ_DIVISOR_DEFAULT;
q->tail = q->limit = q->depth = ctl->flows ? : SFQ_DEPTH_DEFAULT;
if (q->depth > SFQ_MAX_DEPTH)
@@ -426,6 +430,9 @@ static int sfq_q_init(struct sfq_sched_data *q, struct rtattr *opt)
q->limit = min_t(u32, ctl->limit, q->depth);
}
+ q->ht = kcalloc(q->hash_divisor, sizeof(sfq_index), GFP_KERNEL);
+ if (!q->ht)
+ goto err_case;
q->dep = kcalloc(1 + q->depth*2, sizeof(struct sfq_head), GFP_KERNEL);
if (!q->dep)
goto err_case;
@@ -442,7 +449,7 @@ static int sfq_q_init(struct sfq_sched_data *q, struct rtattr *opt)
if (!q->qs)
goto err_case;
- for (i=0; i<SFQ_HASH_DIVISOR; i++)
+ for (i=0; i<q->hash_divisor; i++)
q->ht[i] = q->depth;
for (i=0; i < q->depth; i++) {
skb_queue_head_init(&q->qs[i]);
@@ -488,7 +495,7 @@ static int sfq_dump(struct Qdisc *sch, struct sk_buff *skb)
opt.perturb_period = q->perturb_period/HZ;
opt.limit = q->limit;
- opt.divisor = SFQ_HASH_DIVISOR;
+ opt.divisor = q->hash_divisor;
opt.flows = q->depth;
RTA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
--
1.5.2.4
^ permalink raw reply related
* [PATCH 6/7] Make qdisc changeable.
From: Corey Hickey @ 2007-07-30 0:21 UTC (permalink / raw)
To: netdev; +Cc: Corey Hickey
In-Reply-To: <11857548771998-git-send-email-bugfood-ml@fatooh.org>
Re-implement sfq_change() and enable Qdisc_opts.change so "tc qdisc
change" will work.
Signed-off-by: Corey Hickey <bugfood-ml@fatooh.org>
---
net/sched/sch_sfq.c | 51 ++++++++++++++++++++++++++++++++++++++++++++++++++-
1 files changed, 50 insertions(+), 1 deletions(-)
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index e6a6a21..e042cd0 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -485,6 +485,55 @@ static int sfq_init(struct Qdisc *sch, struct rtattr *opt)
return 0;
}
+static int sfq_change(struct Qdisc *sch, struct rtattr *opt)
+{
+ struct sfq_sched_data *q = qdisc_priv(sch);
+ struct sfq_sched_data tmp;
+ struct sk_buff *skb;
+ int err;
+
+ /* set up tmp queue */
+ memset(&tmp, 0, sizeof(struct sfq_sched_data));
+ tmp.quantum = psched_mtu(sch->dev); /* default */
+ if ((err = sfq_q_init(&tmp, opt)))
+ return err;
+
+ /* copy packets from the old queue to the tmp queue */
+ sch_tree_lock(sch);
+ while (sch->q.qlen >= tmp.limit - 1)
+ sfq_drop(sch);
+ while ((skb = sfq_q_dequeue(q)) != NULL)
+ sfq_q_enqueue(skb, &tmp, SFQ_TAIL);
+
+ /* clean up the old queue */
+ sfq_q_destroy(q);
+
+ /* copy elements of the tmp queue into the old queue */
+ q->perturb_period = tmp.perturb_period;
+ q->quantum = tmp.quantum;
+ q->limit = tmp.limit;
+ q->depth = tmp.depth;
+ q->hash_divisor = tmp.hash_divisor;
+ q->tail = tmp.tail;
+ q->max_depth = tmp.max_depth;
+ q->ht = tmp.ht;
+ q->dep = tmp.dep;
+ q->next = tmp.next;
+ q->allot = tmp.allot;
+ q->hash = tmp.hash;
+ q->qs = tmp.qs;
+
+ /* finish up */
+ if (q->perturb_period) {
+ q->perturb_timer.expires = jiffies + q->perturb_period;
+ add_timer(&q->perturb_timer);
+ } else {
+ q->perturbation = 0;
+ }
+ sch_tree_unlock(sch);
+ return 0;
+}
+
static int sfq_dump(struct Qdisc *sch, struct sk_buff *skb)
{
struct sfq_sched_data *q = qdisc_priv(sch);
@@ -519,7 +568,7 @@ static struct Qdisc_ops sfq_qdisc_ops = {
.init = sfq_init,
.reset = sfq_reset,
.destroy = sfq_destroy,
- .change = NULL,
+ .change = sfq_change,
.dump = sfq_dump,
.owner = THIS_MODULE,
};
--
1.5.2.4
^ permalink raw reply related
* [PATCH 4/7] Add "depth".
From: Corey Hickey @ 2007-07-30 0:21 UTC (permalink / raw)
To: netdev; +Cc: Corey Hickey
In-Reply-To: <11857548771998-git-send-email-bugfood-ml@fatooh.org>
Make "depth" (number of queues) user-configurable:
* replace #define with a parameter
* use old hardcoded value as a default
* kcalloc() arrays in sfq_q_init()
* free() arrays in sfq_q_destroy()
Signed-off-by: Corey Hickey <bugfood-ml@fatooh.org>
---
net/sched/sch_sfq.c | 81 +++++++++++++++++++++++++++++++++++----------------
1 files changed, 56 insertions(+), 25 deletions(-)
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 583f925..70124ac 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -74,14 +74,16 @@
It is easy to increase these values, but not in flight. */
-#define SFQ_DEPTH 128
+#define SFQ_DEPTH_DEFAULT 128
#define SFQ_HASH_DIVISOR 1024
#define SFQ_HEAD 0
#define SFQ_TAIL 1
-/* This type should contain at least SFQ_DEPTH*2 values */
-typedef unsigned char sfq_index;
+/* This type must contain greater than depth*2 values, so depth is constrained
+ * accordingly. */
+typedef unsigned int sfq_index;
+#define SFQ_MAX_DEPTH (UINT_MAX / 2 - 1)
struct sfq_head
{
@@ -95,6 +97,7 @@ struct sfq_sched_data
int perturb_period;
unsigned quantum; /* Allotment per round: MUST BE >= MTU */
int limit;
+ unsigned depth;
/* Variables */
struct timer_list perturb_timer;
@@ -103,11 +106,11 @@ struct sfq_sched_data
sfq_index max_depth; /* Maximal depth */
sfq_index ht[SFQ_HASH_DIVISOR]; /* Hash table */
- sfq_index next[SFQ_DEPTH]; /* Active slots link */
- short allot[SFQ_DEPTH]; /* Current allotment per slot */
- unsigned short hash[SFQ_DEPTH]; /* Hash value indexed by slots */
- struct sk_buff_head qs[SFQ_DEPTH]; /* Slot queue */
- struct sfq_head dep[SFQ_DEPTH*2]; /* Linked list of slots, indexed by depth */
+ sfq_index *next; /* Active slots link */
+ short *allot; /* Current allotment per slot */
+ unsigned short *hash; /* Hash value indexed by slots */
+ struct sk_buff_head *qs; /* Slot queue */
+ struct sfq_head *dep; /* Linked list of slots, indexed by depth */
};
static __inline__ unsigned sfq_fold_hash(struct sfq_sched_data *q, u32 h, u32 h1)
@@ -164,7 +167,7 @@ static unsigned sfq_hash(struct sfq_sched_data *q, struct sk_buff *skb)
static inline void sfq_link(struct sfq_sched_data *q, sfq_index x)
{
sfq_index p, n;
- int d = q->qs[x].qlen + SFQ_DEPTH;
+ int d = q->qs[x].qlen + q->depth;
p = d;
n = q->dep[d].next;
@@ -215,7 +218,7 @@ static unsigned int sfq_drop(struct Qdisc *sch)
drop a packet from it */
if (d > 1) {
- sfq_index x = q->dep[d+SFQ_DEPTH].next;
+ sfq_index x = q->dep[d + q->depth].next;
skb = q->qs[x].prev;
len = skb->len;
__skb_unlink(skb, &q->qs[x]);
@@ -238,7 +241,7 @@ static unsigned int sfq_drop(struct Qdisc *sch)
kfree_skb(skb);
sfq_dec(q, d);
sch->q.qlen--;
- q->ht[q->hash[d]] = SFQ_DEPTH;
+ q->ht[q->hash[d]] = q->depth;
sch->qstats.drops++;
sch->qstats.backlog -= len;
return len;
@@ -253,8 +256,8 @@ static void sfq_q_enqueue(struct sk_buff *skb, struct sfq_sched_data *q, unsigne
sfq_index x;
x = q->ht[hash];
- if (x == SFQ_DEPTH) {
- q->ht[hash] = x = q->dep[SFQ_DEPTH].next;
+ if (x == q->depth) {
+ q->ht[hash] = x = q->dep[q->depth].next;
q->hash[x] = hash;
}
@@ -265,7 +268,7 @@ static void sfq_q_enqueue(struct sk_buff *skb, struct sfq_sched_data *q, unsigne
sfq_inc(q, x);
if (q->qs[x].qlen == 1) { /* The flow is new */
- if (q->tail == SFQ_DEPTH) { /* It is the first flow */
+ if (q->tail == q->depth) { /* It is the first flow */
q->tail = x;
q->next[x] = x;
q->allot[x] = q->quantum;
@@ -316,7 +319,7 @@ static struct sk_buff *sfq_q_dequeue(struct sfq_sched_data *q)
sfq_index a, old_a;
/* No active slots */
- if (q->tail == SFQ_DEPTH)
+ if (q->tail == q->depth)
return NULL;
a = old_a = q->next[q->tail];
@@ -327,10 +330,10 @@ static struct sk_buff *sfq_q_dequeue(struct sfq_sched_data *q)
/* Is the slot empty? */
if (q->qs[a].qlen == 0) {
- q->ht[q->hash[a]] = SFQ_DEPTH;
+ q->ht[q->hash[a]] = q->depth;
a = q->next[a];
if (a == old_a) {
- q->tail = SFQ_DEPTH;
+ q->tail = q->depth;
return skb;
}
q->next[q->tail] = a;
@@ -383,6 +386,11 @@ static void sfq_perturbation(unsigned long arg)
static void sfq_q_destroy(struct sfq_sched_data *q)
{
del_timer(&q->perturb_timer);
+ kfree(q->dep);
+ kfree(q->next);
+ kfree(q->allot);
+ kfree(q->hash);
+ kfree(q->qs);
}
static void sfq_destroy(struct Qdisc *sch)
@@ -401,30 +409,53 @@ static int sfq_q_init(struct sfq_sched_data *q, struct rtattr *opt)
q->perturbation = 0;
q->max_depth = 0;
- q->tail = q->limit = SFQ_DEPTH;
if (opt == NULL) {
q->perturb_period = 0;
+ q->tail = q->limit = q->depth = SFQ_DEPTH_DEFAULT;
} else {
struct tc_sfq_qopt *ctl = RTA_DATA(opt);
if (ctl->quantum)
q->quantum = ctl->quantum;
q->perturb_period = ctl->perturb_period*HZ;
+ q->tail = q->limit = q->depth = ctl->flows ? : SFQ_DEPTH_DEFAULT;
+
+ if (q->depth > SFQ_MAX_DEPTH)
+ return -EINVAL;
if (ctl->limit)
- q->limit = min_t(u32, ctl->limit, SFQ_DEPTH);
+ q->limit = min_t(u32, ctl->limit, q->depth);
}
+ q->dep = kcalloc(1 + q->depth*2, sizeof(struct sfq_head), GFP_KERNEL);
+ if (!q->dep)
+ goto err_case;
+ q->next = kcalloc(q->depth, sizeof(sfq_index), GFP_KERNEL);
+ if (!q->next)
+ goto err_case;
+ q->allot = kcalloc(q->depth, sizeof(short), GFP_KERNEL);
+ if (!q->allot)
+ goto err_case;
+ q->hash = kcalloc(q->depth, sizeof(unsigned short), GFP_KERNEL);
+ if (!q->hash)
+ goto err_case;
+ q->qs = kcalloc(q->depth, sizeof(struct sk_buff_head), GFP_KERNEL);
+ if (!q->qs)
+ goto err_case;
+
for (i=0; i<SFQ_HASH_DIVISOR; i++)
- q->ht[i] = SFQ_DEPTH;
- for (i=0; i<SFQ_DEPTH; i++) {
+ q->ht[i] = q->depth;
+ for (i=0; i < q->depth; i++) {
skb_queue_head_init(&q->qs[i]);
- q->dep[i+SFQ_DEPTH].next = i+SFQ_DEPTH;
- q->dep[i+SFQ_DEPTH].prev = i+SFQ_DEPTH;
+ q->dep[i + q->depth].next = i + q->depth;
+ q->dep[i + q->depth].prev = i + q->depth;
}
- for (i=0; i<SFQ_DEPTH; i++)
+ for (i=0; i < q->depth; i++)
sfq_link(q, i);
return 0;
+err_case:
+ sfq_q_destroy(q);
+ return -ENOBUFS;
}
static int sfq_init(struct Qdisc *sch, struct rtattr *opt)
@@ -458,7 +489,7 @@ static int sfq_dump(struct Qdisc *sch, struct sk_buff *skb)
opt.limit = q->limit;
opt.divisor = SFQ_HASH_DIVISOR;
- opt.flows = q->limit;
+ opt.flows = q->depth;
RTA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
--
1.5.2.4
^ permalink raw reply related
* [PATCH 7/7] Remove comments about hardcoded values.
From: Corey Hickey @ 2007-07-30 0:21 UTC (permalink / raw)
To: netdev; +Cc: Corey Hickey
In-Reply-To: <11857548771998-git-send-email-bugfood-ml@fatooh.org>
None of these are true anymore (hooray!).
Signed-off-by: Corey Hickey <bugfood-ml@fatooh.org>
---
include/linux/pkt_sched.h | 8 --------
net/sched/sch_sfq.c | 17 +++--------------
2 files changed, 3 insertions(+), 22 deletions(-)
diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h
index 268c515..58a0ea6 100644
--- a/include/linux/pkt_sched.h
+++ b/include/linux/pkt_sched.h
@@ -148,14 +148,6 @@ struct tc_sfq_qopt
unsigned flows; /* Maximal number of flows */
};
-/*
- * NOTE: limit, divisor and flows are hardwired to code at the moment.
- *
- * limit=flows=128, divisor=1024;
- *
- * The only reason for this is efficiency, it is possible
- * to change these parameters in compile time.
- */
/* RED section */
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index e042cd0..3890452 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -61,18 +61,7 @@
We still need true WFQ for top level CSZ, but using WFQ
for the best effort traffic is absolutely pointless:
- SFQ is superior for this purpose.
-
- IMPLEMENTATION:
- This implementation limits maximal queue length to 128;
- maximal mtu to 2^15-1; number of hash buckets to 1024.
- The only goal of this restrictions was that all data
- fit into one 4K page :-). Struct sfq_sched_data is
- organized in anti-cache manner: all the data for a bucket
- are scattered over different locations. This is not good,
- but it allowed me to put it into 4K.
-
- It is easy to increase these values, but not in flight. */
+ SFQ is superior for this purpose. */
#define SFQ_DEPTH_DEFAULT 128
#define SFQ_DIVISOR_DEFAULT 1024
@@ -491,7 +480,7 @@ static int sfq_change(struct Qdisc *sch, struct rtattr *opt)
struct sfq_sched_data tmp;
struct sk_buff *skb;
int err;
-
+
/* set up tmp queue */
memset(&tmp, 0, sizeof(struct sfq_sched_data));
tmp.quantum = psched_mtu(sch->dev); /* default */
@@ -504,7 +493,7 @@ static int sfq_change(struct Qdisc *sch, struct rtattr *opt)
sfq_drop(sch);
while ((skb = sfq_q_dequeue(q)) != NULL)
sfq_q_enqueue(skb, &tmp, SFQ_TAIL);
-
+
/* clean up the old queue */
sfq_q_destroy(q);
--
1.5.2.4
^ permalink raw reply related
* [PATCH] [iproute2] SFQ: Support changing depth and divisor.
From: Corey Hickey @ 2007-07-30 0:21 UTC (permalink / raw)
To: netdev
In-Reply-To: <11857548771998-git-send-email-bugfood-ml@fatooh.org>
This can safely be applied either before or after the kernel
patches because the tc_sfq_qopt struct is unchanged:
- old kernels will ignore the parameters from new iproute2
- new kernels will use the same default parameters
---
include/linux/pkt_sched.h | 9 ---------
tc/q_sfq.c | 21 ++++++++++++++++++++-
2 files changed, 20 insertions(+), 10 deletions(-)
diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h
index d10f353..37946d4 100644
--- a/include/linux/pkt_sched.h
+++ b/include/linux/pkt_sched.h
@@ -139,15 +139,6 @@ struct tc_sfq_qopt
unsigned flows; /* Maximal number of flows */
};
-/*
- * NOTE: limit, divisor and flows are hardwired to code at the moment.
- *
- * limit=flows=128, divisor=1024;
- *
- * The only reason for this is efficiency, it is possible
- * to change these parameters in compile time.
- */
-
/* RED section */
enum
diff --git a/tc/q_sfq.c b/tc/q_sfq.c
index 05385cf..7754db7 100644
--- a/tc/q_sfq.c
+++ b/tc/q_sfq.c
@@ -25,7 +25,7 @@
static void explain(void)
{
- fprintf(stderr, "Usage: ... sfq [ limit NUMBER ] [ perturb SECS ] [ quantum BYTES ]\n");
+ fprintf(stderr, "Usage: ... sfq [ limit NUMBER ] [ depth FLOWS ] [ divisor HASHBITS ] [ perturb SECS ] [ quantum BYTES ]\n");
}
#define usage() return(-1)
@@ -63,6 +63,25 @@ static int sfq_parse_opt(struct qdisc_util *qu, int argc, char **argv, struct nl
return -1;
}
ok++;
+ } else if (strcmp(*argv, "depth") == 0) {
+ NEXT_ARG();
+ if (get_unsigned(&opt.flows, *argv, 0)) {
+ fprintf(stderr, "Illegal \"depth\"\n");
+ return -1;
+ }
+ ok++;
+ } else if (strcmp(*argv, "divisor") == 0) {
+ NEXT_ARG();
+ if (get_unsigned(&opt.divisor, *argv, 0)) {
+ fprintf(stderr, "Illegal \"divisor\"\n");
+ return -1;
+ }
+ if (opt.divisor >= 15) {
+ fprintf(stderr, "Illegal \"divisor\", must be < 15\n");
+ return -1;
+ }
+ opt.divisor = 1<<opt.divisor;
+ ok++;
} else if (strcmp(*argv, "help") == 0) {
explain();
return -1;
--
1.5.2.4
^ permalink raw reply related
* [PATCH v3 -mm 0/9] netconsole: Multiple targets and dynamic reconfigurability
From: Satyam Sharma @ 2007-07-30 2:47 UTC (permalink / raw)
To: Linux Kernel Mailing List
Cc: Keiichi Kii, Netdev, Joel Becker, Matt Mackall, Andrew Morton,
David Miller
[0/9] netconsole: Multiple targets and dynamic reconfigurability
This is v3 of the patchset, the previous versions are available at:
http://lkml.org/lkml/2007/7/4/107
http://lkml.org/lkml/2007/7/10/78
Diffed against 2.6.23-rc1-git6 (6c8dca5d as of writing), but applies
successfully to 2.6.23-rc1-mm1 as well.
Patches 1-5 are okay-to-apply to current mainline -git, I think.
Patches 6-9 introduce the new functionality and are requested for
inclusion in -mm.
[1/9] netconsole: Cleanups, codingstyle, prettyfication
[2/9] netconsole: Remove bogus check
[3/9] netconsole: Simplify boot/module option setup logic
[4/9] netconsole: Use netif_running() in write_msg()
[5/9] netconsole: Add some useful tips to documentation
[6/9] netconsole: Introduce netconsole_target
[7/9] netconsole: Introduce netconsole_netdev_notifier
[8/9] netconsole: Support multiple logging targets
[9/9] netconsole: Support dynamic reconfiguration using configfs
Changes since v2:
=================
* "enabled" must be defined outside #ifdef NETCONSOLE_DYNAMIC
* Some simple enhancements to documentation
* Drop the use of "unlikely" from a condition where I'd got the
common case wrong
About this patchset:
====================
What?
Support multiple remote logging targets in netconsole. Also, ability to
dynamically add or remove targets or modify parameters (IP, port, remote
MAC address) of targets at runtime, from userspace, using configfs.
Why?
>From Keiichi Kii's original post:
[...] current netconsole is not flexible. For example, if you want to change
ip address for logging agent, in the case of built-in netconsole, you can't
change config except for changing boot parameter and rebooting your system,
or in the case of module netconsole, you need to remove it and reload
with different parameters. [...] and we have been losing serial console
port with PCs and Servers.
(... and especially laptops, I would add.)
Satyam
^ permalink raw reply
* [PATCH v3 -mm 1/9] netconsole: Cleanups, codingstyle, prettyfication
From: Satyam Sharma @ 2007-07-30 2:47 UTC (permalink / raw)
To: Linux Kernel Mailing List
Cc: Keiichi Kii, Netdev, Joel Becker, Matt Mackall, Andrew Morton,
David Miller
In-Reply-To: <20070730024741.10828.48209.sendpatchset@enigma.security.iitk.ac.in>
From: Satyam Sharma <satyam@infradead.org>
[1/9] netconsole: Cleanups, codingstyle, prettyfication
(1) Remove unwanted headers.
(2) Mark __init and __exit as appropriate.
(3) Various trivial codingstyle and prettification stuff.
Signed-off-by: Satyam Sharma <satyam@infradead.org>
Signed-off-by: Keiichi Kii <k-keiichi@bx.jp.nec.com>
Acked-by: Matt Mackall <mpm@selenic.com>
---
drivers/net/netconsole.c | 55 ++++++++++++++++++++++-----------------------
1 files changed, 27 insertions(+), 28 deletions(-)
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index 69233f6..f1c2a2d 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -35,35 +35,32 @@
****************************************************************/
#include <linux/mm.h>
-#include <linux/tty.h>
#include <linux/init.h>
#include <linux/module.h>
#include <linux/console.h>
-#include <linux/tty_driver.h>
#include <linux/moduleparam.h>
#include <linux/string.h>
-#include <linux/sysrq.h>
-#include <linux/smp.h>
#include <linux/netpoll.h>
MODULE_AUTHOR("Maintainer: Matt Mackall <mpm@selenic.com>");
MODULE_DESCRIPTION("Console driver for network interfaces");
MODULE_LICENSE("GPL");
-static char config[256];
-module_param_string(netconsole, config, 256, 0);
+#define MAX_PARAM_LENGTH 256
+#define MAX_PRINT_CHUNK 1000
+
+static char config[MAX_PARAM_LENGTH];
+module_param_string(netconsole, config, MAX_PARAM_LENGTH, 0);
MODULE_PARM_DESC(netconsole, " netconsole=[src-port]@[src-ip]/[dev],[tgt-port]@<tgt-ip>/[tgt-macaddr]\n");
static struct netpoll np = {
- .name = "netconsole",
- .dev_name = "eth0",
- .local_port = 6665,
- .remote_port = 6666,
- .remote_mac = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff},
+ .name = "netconsole",
+ .dev_name = "eth0",
+ .local_port = 6665,
+ .remote_port = 6666,
+ .remote_mac = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff},
};
-static int configured = 0;
-
-#define MAX_PRINT_CHUNK 1000
+static int configured;
static void write_msg(struct console *con, const char *msg, unsigned int len)
{
@@ -75,7 +72,7 @@ static void write_msg(struct console *con, const char *msg, unsigned int len)
local_irq_save(flags);
- for(left = len; left; ) {
+ for (left = len; left;) {
frag = min(left, MAX_PRINT_CHUNK);
netpoll_send_udp(&np, msg, frag);
msg += frag;
@@ -86,12 +83,12 @@ static void write_msg(struct console *con, const char *msg, unsigned int len)
}
static struct console netconsole = {
- .name = "netcon",
- .flags = CON_ENABLED | CON_PRINTBUFFER,
- .write = write_msg
+ .name = "netcon",
+ .flags = CON_ENABLED | CON_PRINTBUFFER,
+ .write = write_msg,
};
-static int option_setup(char *opt)
+static int __init option_setup(char *opt)
{
configured = !netpoll_parse_options(&np, opt);
return 1;
@@ -99,28 +96,30 @@ static int option_setup(char *opt)
__setup("netconsole=", option_setup);
-static int init_netconsole(void)
+static int __init init_netconsole(void)
{
- int err;
+ int err = 0;
- if(strlen(config))
+ if (strnlen(config, MAX_PARAM_LENGTH))
option_setup(config);
- if(!configured) {
- printk("netconsole: not configured, aborting\n");
- return 0;
+ if (!configured) {
+ printk(KERN_INFO "netconsole: not configured, aborting\n");
+ goto out;
}
err = netpoll_setup(&np);
if (err)
- return err;
+ goto out;
register_console(&netconsole);
printk(KERN_INFO "netconsole: network logging started\n");
- return 0;
+
+out:
+ return err;
}
-static void cleanup_netconsole(void)
+static void __exit cleanup_netconsole(void)
{
unregister_console(&netconsole);
netpoll_cleanup(&np);
^ permalink raw reply related
* [PATCH v3 -mm 2/9] netconsole: Remove bogus check
From: Satyam Sharma @ 2007-07-30 2:48 UTC (permalink / raw)
To: Linux Kernel Mailing List
Cc: Keiichi Kii, Netdev, Joel Becker, Matt Mackall, Andrew Morton,
David Miller
In-Reply-To: <20070730024741.10828.48209.sendpatchset@enigma.security.iitk.ac.in>
From: Satyam Sharma <satyam@infradead.org>
[2/9] netconsole: Remove bogus check
The (!np.dev) check in write_msg() is bogus (always false), because:
np.dev is set by netpoll_setup(), which is called by init_netconsole()
before register_console(), so write_msg() cannot be triggered unless
netpoll_setup() successfully set np.dev. Also np.dev cannot go away
from under us, because netpoll_setup() grabs us reference on it. So
let's remove the bogus check.
Signed-off-by: Satyam Sharma <satyam@infradead.org>
Acked-by: Keiichi Kii <k-keiichi@bx.jp.nec.com>
Acked-by: Matt Mackall <mpm@selenic.com>
---
drivers/net/netconsole.c | 3 ---
1 files changed, 0 insertions(+), 3 deletions(-)
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index f1c2a2d..2c2aef1 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -67,9 +67,6 @@ static void write_msg(struct console *con, const char *msg, unsigned int len)
int frag, left;
unsigned long flags;
- if (!np.dev)
- return;
-
local_irq_save(flags);
for (left = len; left;) {
^ permalink raw reply related
* [PATCH v3 -mm 3/9] netconsole: Simplify boot/module option setup logic
From: Satyam Sharma @ 2007-07-30 2:48 UTC (permalink / raw)
To: Linux Kernel Mailing List
Cc: Keiichi Kii, Netdev, Joel Becker, Matt Mackall, Andrew Morton,
David Miller
In-Reply-To: <20070730024741.10828.48209.sendpatchset@enigma.security.iitk.ac.in>
From: Satyam Sharma <satyam@infradead.org>
[3/9] netconsole: Simplify boot/module option setup logic
Presently, boot/module parameters are set up quite differently for the
case of built-in netconsole (__setup() -> obsolete_checksetup() ->
netpoll_parse_options() -> strlen(config) == 0 in init_netconsole())
vs modular netconsole (module_param_string() -> string copied to the
config variable -> strlen(config) != 0 init_netconsole() ->
netpoll_parse_options()).
This patch makes both of them similar by doing exactly the equivalent
of a module_param_string() in option_setup() also -- just copying the
param string passed from the kernel command line into "config" variable.
So, strlen(config) != 0 in both cases, and netpoll_parse_options() is
always called from init_netconsole(), thus making the setup logic for
both cases similar.
Now, option_setup() is only ever called / used for the built-in case,
so we put it inside a #ifndef MODULE, otherwise gcc will complain about
option_setup() being "defined but not used". Also, the "configured"
variable is redundant with this patch and hence removed.
Signed-off-by: Satyam Sharma <satyam@infradead.org>
Signed-off-by: Keiichi Kii <k-keiichi@bx.jp.nec.com>
Acked-by: Matt Mackall <mpm@selenic.com>
---
drivers/net/netconsole.c | 27 ++++++++++++++-------------
1 files changed, 14 insertions(+), 13 deletions(-)
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index 2c2aef1..e56aa6c 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -53,6 +53,15 @@ static char config[MAX_PARAM_LENGTH];
module_param_string(netconsole, config, MAX_PARAM_LENGTH, 0);
MODULE_PARM_DESC(netconsole, " netconsole=[src-port]@[src-ip]/[dev],[tgt-port]@<tgt-ip>/[tgt-macaddr]\n");
+#ifndef MODULE
+static int __init option_setup(char *opt)
+{
+ strlcpy(config, opt, MAX_PARAM_LENGTH);
+ return 1;
+}
+__setup("netconsole=", option_setup);
+#endif /* MODULE */
+
static struct netpoll np = {
.name = "netconsole",
.dev_name = "eth0",
@@ -60,7 +69,6 @@ static struct netpoll np = {
.remote_port = 6666,
.remote_mac = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff},
};
-static int configured;
static void write_msg(struct console *con, const char *msg, unsigned int len)
{
@@ -85,26 +93,19 @@ static struct console netconsole = {
.write = write_msg,
};
-static int __init option_setup(char *opt)
-{
- configured = !netpoll_parse_options(&np, opt);
- return 1;
-}
-
-__setup("netconsole=", option_setup);
-
static int __init init_netconsole(void)
{
int err = 0;
- if (strnlen(config, MAX_PARAM_LENGTH))
- option_setup(config);
-
- if (!configured) {
+ if (!strnlen(config, MAX_PARAM_LENGTH)) {
printk(KERN_INFO "netconsole: not configured, aborting\n");
goto out;
}
+ err = netpoll_parse_options(&np, config);
+ if (err)
+ goto out;
+
err = netpoll_setup(&np);
if (err)
goto out;
^ permalink raw reply related
* [PATCH v3 -mm 4/9] netconsole: Use netif_running() in write_msg()
From: Satyam Sharma @ 2007-07-30 2:48 UTC (permalink / raw)
To: Linux Kernel Mailing List
Cc: Keiichi Kii, Netdev, Joel Becker, Matt Mackall, Andrew Morton,
David Miller
In-Reply-To: <20070730024741.10828.48209.sendpatchset@enigma.security.iitk.ac.in>
From: Satyam Sharma <satyam@infradead.org>
[4/9] netconsole: Use netif_running() in write_msg()
Avoid unnecessarily disabling interrupts and calling netpoll_send_udp()
if the corresponding local interface is not up.
Signed-off-by: Satyam Sharma <satyam@infradead.org>
Acked-by: Keiichi Kii <k-keiichi@bx.jp.nec.com>
Cc: Matt Mackall <mpm@selenic.com>
---
drivers/net/netconsole.c | 18 +++++++++---------
1 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index e56aa6c..75cb761 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -75,16 +75,16 @@ static void write_msg(struct console *con, const char *msg, unsigned int len)
int frag, left;
unsigned long flags;
- local_irq_save(flags);
-
- for (left = len; left;) {
- frag = min(left, MAX_PRINT_CHUNK);
- netpoll_send_udp(&np, msg, frag);
- msg += frag;
- left -= frag;
+ if (netif_running(np.dev)) {
+ local_irq_save(flags);
+ for (left = len; left;) {
+ frag = min(left, MAX_PRINT_CHUNK);
+ netpoll_send_udp(&np, msg, frag);
+ msg += frag;
+ left -= frag;
+ }
+ local_irq_restore(flags);
}
-
- local_irq_restore(flags);
}
static struct console netconsole = {
^ permalink raw reply related
* [PATCH v3 -mm 5/9] netconsole: Add some useful tips to documentation
From: Satyam Sharma @ 2007-07-30 2:48 UTC (permalink / raw)
To: Linux Kernel Mailing List
Cc: Keiichi Kii, Netdev, Joel Becker, Matt Mackall, Andrew Morton,
David Miller
In-Reply-To: <20070730024741.10828.48209.sendpatchset@enigma.security.iitk.ac.in>
From: Satyam Sharma <satyam@infradead.org>
[5/9] netconsole: Add some useful tips to documentation
Add some useful general-purpose tips. Also suggest solution for the frequent
problem of console loglevel set too low numerically (i.e. for high priority
messages only) on the sender.
Signed-off-by: Satyam Sharma <satyam@infradead.org>
Acked-by: Keiichi Kii <k-keiichi@bx.jp.nec.com>
Acked-by: Matt Mackall <mpm@selenic.com>
---
Documentation/networking/netconsole.txt | 25 +++++++++++++++++++++++++
1 files changed, 25 insertions(+), 0 deletions(-)
diff --git a/Documentation/networking/netconsole.txt b/Documentation/networking/netconsole.txt
index 1caa6c7..5962f45 100644
--- a/Documentation/networking/netconsole.txt
+++ b/Documentation/networking/netconsole.txt
@@ -44,11 +44,36 @@ WARNING: the default target ethernet setting uses the broadcast
ethernet address to send packets, which can cause increased load on
other systems on the same ethernet segment.
+TIP: some LAN switches may be configured to suppress ethernet broadcasts
+so it is advised to explicitly specify the remote agents' MAC addresses
+from the config parameters passed to netconsole.
+
+TIP: to find out the MAC address of, say, 10.0.0.2, you may try using:
+
+ ping -c 1 10.0.0.2 ; /sbin/arp -n | grep 10.0.0.2
+
+TIP: in case the remote logging agent is on a separate LAN subnet than
+the sender, it is suggested to try specifying the MAC address of the
+default gateway (you may use /sbin/route -n to find it out) as the
+remote MAC address instead.
+
NOTE: the network device (eth1 in the above case) can run any kind
of other network traffic, netconsole is not intrusive. Netconsole
might cause slight delays in other traffic if the volume of kernel
messages is high, but should have no other impact.
+NOTE: if you find that the remote logging agent is not receiving or
+printing all messages from the sender, it is likely that you have set
+the "console_loglevel" parameter (on the sender) to only send high
+priority messages to the console. You can change this at runtime using:
+
+ dmesg -n 8
+
+or by specifying "debug" on the kernel command line at boot, to send
+all kernel messages to the console. A specific value for this parameter
+can also be set using the "loglevel" kernel boot option. See the
+dmesg(8) man page and Documentation/kernel-parameters.txt for details.
+
Netconsole was designed to be as instantaneous as possible, to
enable the logging of even the most critical kernel bugs. It works
from IRQ contexts as well, and does not enable interrupts while
^ permalink raw reply related
* [PATCH v3 -mm 6/9] netconsole: Introduce netconsole_target
From: Satyam Sharma @ 2007-07-30 2:48 UTC (permalink / raw)
To: Linux Kernel Mailing List
Cc: Keiichi Kii, Netdev, Joel Becker, Matt Mackall, Andrew Morton,
David Miller
In-Reply-To: <20070730024741.10828.48209.sendpatchset@enigma.security.iitk.ac.in>
From: Satyam Sharma <satyam@infradead.org>
[6/9] netconsole: Introduce netconsole_target
Introduce a wrapper structure over netpoll to represent logging targets
configured in netconsole. This will get extended with other members in
further patches.
This is done independent of the (to-be-introduced) NETCONSOLE_DYNAMIC
config option so that we're able to drastically cut down on the #ifdef
complexity of final netconsole.c. Also, struct netconsole_target would be
required for multiple targets support also, and not just dynamic
reconfigurability.
Signed-off-by: Satyam Sharma <satyam@infradead.org>
Signed-off-by: Keiichi Kii <k-keiichi@bx.jp.nec.com>
---
drivers/net/netconsole.c | 36 +++++++++++++++++++++++++-----------
1 files changed, 25 insertions(+), 11 deletions(-)
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index 75cb761..be15ca6 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -62,24 +62,35 @@ static int __init option_setup(char *opt)
__setup("netconsole=", option_setup);
#endif /* MODULE */
-static struct netpoll np = {
- .name = "netconsole",
- .dev_name = "eth0",
- .local_port = 6665,
- .remote_port = 6666,
- .remote_mac = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff},
+/**
+ * struct netconsole_target - Represents a configured netconsole target.
+ * @np: The netpoll structure for this target.
+ */
+struct netconsole_target {
+ struct netpoll np;
+};
+
+static struct netconsole_target default_target = {
+ .np = {
+ .name = "netconsole",
+ .dev_name = "eth0",
+ .local_port = 6665,
+ .remote_port = 6666,
+ .remote_mac = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff},
+ },
};
static void write_msg(struct console *con, const char *msg, unsigned int len)
{
int frag, left;
unsigned long flags;
+ struct netconsole_target *nt = &default_target;
- if (netif_running(np.dev)) {
+ if (netif_running(nt->np.dev)) {
local_irq_save(flags);
for (left = len; left;) {
frag = min(left, MAX_PRINT_CHUNK);
- netpoll_send_udp(&np, msg, frag);
+ netpoll_send_udp(&nt->np, msg, frag);
msg += frag;
left -= frag;
}
@@ -96,17 +107,18 @@ static struct console netconsole = {
static int __init init_netconsole(void)
{
int err = 0;
+ struct netconsole_target *nt = &default_target;
if (!strnlen(config, MAX_PARAM_LENGTH)) {
printk(KERN_INFO "netconsole: not configured, aborting\n");
goto out;
}
- err = netpoll_parse_options(&np, config);
+ err = netpoll_parse_options(&nt->np, config);
if (err)
goto out;
- err = netpoll_setup(&np);
+ err = netpoll_setup(&nt->np);
if (err)
goto out;
@@ -119,8 +131,10 @@ out:
static void __exit cleanup_netconsole(void)
{
+ struct netconsole_target *nt = &default_target;
+
unregister_console(&netconsole);
- netpoll_cleanup(&np);
+ netpoll_cleanup(&nt->np);
}
module_init(init_netconsole);
^ permalink raw reply related
* [PATCH v3 -mm 7/9] netconsole: Introduce netconsole_netdev_notifier
From: Satyam Sharma @ 2007-07-30 2:48 UTC (permalink / raw)
To: Linux Kernel Mailing List
Cc: Keiichi Kii, Netdev, Joel Becker, Matt Mackall, Andrew Morton,
David Miller
In-Reply-To: <20070730024741.10828.48209.sendpatchset@enigma.security.iitk.ac.in>
From: Satyam Sharma <satyam@infradead.org>
[7/9] netconsole: Introduce netconsole_netdev_notifier
To update fields of underlying netpoll structure at runtime on
corresponding NETDEV_CHANGEADDR or NETDEV_CHANGENAME notifications.
ioctl(SIOCSIFHWADDR or SIOCSIFNAME) could be used to change the
hardware/MAC address or name of the local interface that our netpoll
is attached to. Whenever this happens, netdev notifier chain is called
out with the NETDEV_CHANGEADDR or NETDEV_CHANGENAME event message. We
respond to that and update the local_mac or dev_name field of the struct
netpoll. This makes sense anyway, but is especially required for dynamic
netconsole because the netpoll structure's internal members become user
visible files when either sysfs or configfs are used. So this helps us
to keep up with the MAC address/name changes and keep values in struct
netpoll uptodate.
[ Note that ioctl(SIOCSIFADDR) to change IP address of interface at
runtime is not handled (to update local_ip of netpoll) on purpose --
some setups may set the local_ip to a private address, not necessary
the actual IP address of the sender host, as presently allowed. ]
Signed-off-by: Satyam Sharma <satyam@infradead.org>
Signed-off-by: Keiichi Kii <k-keiichi@bx.jp.nec.com>
---
drivers/net/netconsole.c | 32 ++++++++++++++++++++++++++++++++
1 files changed, 32 insertions(+), 0 deletions(-)
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index be15ca6..5557098 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -80,6 +80,33 @@ static struct netconsole_target default_target = {
},
};
+/* Handle network interface device notifications */
+static int netconsole_netdev_event(struct notifier_block *this,
+ unsigned long event,
+ void *ptr)
+{
+ struct net_device *dev = ptr;
+ struct netconsole_target *nt = &default_target;
+
+ if (nt->np.dev == dev) {
+ switch (event) {
+ case NETDEV_CHANGEADDR:
+ memcpy(nt->np.local_mac, dev->dev_addr, ETH_ALEN);
+ break;
+
+ case NETDEV_CHANGENAME:
+ strlcpy(nt->np.dev_name, dev->name, IFNAMSIZ);
+ break;
+ }
+ }
+
+ return NOTIFY_DONE;
+}
+
+static struct notifier_block netconsole_netdev_notifier = {
+ .notifier_call = netconsole_netdev_event,
+};
+
static void write_msg(struct console *con, const char *msg, unsigned int len)
{
int frag, left;
@@ -122,6 +149,10 @@ static int __init init_netconsole(void)
if (err)
goto out;
+ err = register_netdevice_notifier(&netconsole_netdev_notifier);
+ if (err)
+ goto out;
+
register_console(&netconsole);
printk(KERN_INFO "netconsole: network logging started\n");
@@ -134,6 +165,7 @@ static void __exit cleanup_netconsole(void)
struct netconsole_target *nt = &default_target;
unregister_console(&netconsole);
+ unregister_netdevice_notifier(&netconsole_netdev_notifier);
netpoll_cleanup(&nt->np);
}
^ permalink raw reply related
* [PATCH v3 -mm 8/9] netconsole: Support multiple logging targets
From: Satyam Sharma @ 2007-07-30 2:49 UTC (permalink / raw)
To: Linux Kernel Mailing List
Cc: Keiichi Kii, Netdev, Joel Becker, Matt Mackall, Andrew Morton,
David Miller
In-Reply-To: <20070730024741.10828.48209.sendpatchset@enigma.security.iitk.ac.in>
From: Satyam Sharma <satyam@infradead.org>
[8/9] netconsole: Support multiple logging targets
This patch introduces support for multiple targets, independent of
CONFIG_NETCONSOLE_DYNAMIC -- this is useful even in the default case and
(including the infrastructure introduced in previous patches) doesn't
really add too many bytes to module text. All the complexity (and size)
comes with the dynamic reconfigurability / userspace interface patch,
and so it's plausible users may want to keep this enabled but that
disabled (say to avoid a dependency on CONFIG_CONFIGFS_FS too).
Also update documentation to mention the use of ";" separator to specify
multiple logging targets in the boot/module option string.
Brief overview:
We maintain a target_list (and corresponding lock). Get rid of the static
"default_target" and introduce allocation and release functions for our
netconsole_target objects (but keeping sure to preserve previous behaviour
such as default values). During init_netconsole(), ";" is used as the
separator to identify multiple target specifications in the boot/module
option string. The target specifications are parsed and netpolls setup.
During exit, the target_list is torn down and all items released.
Signed-off-by: Satyam Sharma <satyam@infradead.org>
Signed-off-by: Keiichi Kii <k-keiichi@bx.jp.nec.com>
---
Documentation/networking/netconsole.txt | 6 +
drivers/net/netconsole.c | 171 +++++++++++++++++++++++-------
2 files changed, 137 insertions(+), 40 deletions(-)
diff --git a/Documentation/networking/netconsole.txt b/Documentation/networking/netconsole.txt
index 5962f45..1aaa738 100644
--- a/Documentation/networking/netconsole.txt
+++ b/Documentation/networking/netconsole.txt
@@ -34,6 +34,12 @@ Examples:
insmod netconsole netconsole=@/,@10.0.0.2/
+It also supports logging to multiple remote agents by specifying
+parameters for the multiple agents separated by semicolons and the
+complete string enclosed in "quotes", thusly:
+
+ modprobe netconsole netconsole="@/,@10.0.0.2/;@/eth1,6892@10.0.0.3/"
+
Built-in netconsole starts immediately after the TCP stack is
initialized and attempts to bring up the supplied dev at the supplied
address.
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index 5557098..458c4d6 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -62,44 +62,93 @@ static int __init option_setup(char *opt)
__setup("netconsole=", option_setup);
#endif /* MODULE */
+/* Linked list of all configured targets */
+static LIST_HEAD(target_list);
+
+/* This needs to be a spinlock because write_msg() cannot sleep */
+static DEFINE_SPINLOCK(target_list_lock);
+
/**
* struct netconsole_target - Represents a configured netconsole target.
+ * @list: Links this target into the target_list.
* @np: The netpoll structure for this target.
*/
struct netconsole_target {
+ struct list_head list;
struct netpoll np;
};
-static struct netconsole_target default_target = {
- .np = {
- .name = "netconsole",
- .dev_name = "eth0",
- .local_port = 6665,
- .remote_port = 6666,
- .remote_mac = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff},
- },
-};
+/* Allocate new target and setup netpoll for it */
+static struct netconsole_target *alloc_target(char *target_config)
+{
+ int err = -ENOMEM;
+ struct netconsole_target *nt;
+
+ /* Allocate and initialize with defaults */
+ nt = kzalloc(sizeof(*nt), GFP_KERNEL);
+ if (!nt) {
+ printk(KERN_ERR "netconsole: failed to allocate memory\n");
+ goto fail;
+ }
+
+ nt->np.name = "netconsole";
+ strlcpy(nt->np.dev_name, "eth0", IFNAMSIZ);
+ nt->np.local_port = 6665;
+ nt->np.remote_port = 6666;
+ memset(nt->np.remote_mac, 0xff, ETH_ALEN);
+
+ /* Parse parameters and setup netpoll */
+ err = netpoll_parse_options(&nt->np, target_config);
+ if (err)
+ goto fail;
+
+ err = netpoll_setup(&nt->np);
+ if (err)
+ goto fail;
+
+ return nt;
+
+fail:
+ kfree(nt);
+ return ERR_PTR(err);
+}
+
+/* Cleanup netpoll for given target and free it */
+static void free_target(struct netconsole_target *nt)
+{
+ netpoll_cleanup(&nt->np);
+ kfree(nt);
+}
/* Handle network interface device notifications */
static int netconsole_netdev_event(struct notifier_block *this,
unsigned long event,
void *ptr)
{
+ unsigned long flags;
+ struct netconsole_target *nt;
struct net_device *dev = ptr;
- struct netconsole_target *nt = &default_target;
- if (nt->np.dev == dev) {
- switch (event) {
- case NETDEV_CHANGEADDR:
- memcpy(nt->np.local_mac, dev->dev_addr, ETH_ALEN);
- break;
+ if (!(event == NETDEV_CHANGEADDR || event == NETDEV_CHANGENAME))
+ goto done;
+
+ spin_lock_irqsave(&target_list_lock, flags);
+ list_for_each_entry(nt, &target_list, list) {
+ if (nt->np.dev == dev) {
+ switch (event) {
+ case NETDEV_CHANGEADDR:
+ memcpy(nt->np.local_mac, dev->dev_addr, ETH_ALEN);
+ break;
- case NETDEV_CHANGENAME:
- strlcpy(nt->np.dev_name, dev->name, IFNAMSIZ);
- break;
+ case NETDEV_CHANGENAME:
+ strlcpy(nt->np.dev_name, dev->name, IFNAMSIZ);
+ break;
+ }
}
}
+ spin_unlock_irqrestore(&target_list_lock, flags);
+done:
return NOTIFY_DONE;
}
@@ -111,18 +160,32 @@ static void write_msg(struct console *con, const char *msg, unsigned int len)
{
int frag, left;
unsigned long flags;
- struct netconsole_target *nt = &default_target;
-
- if (netif_running(nt->np.dev)) {
- local_irq_save(flags);
- for (left = len; left;) {
- frag = min(left, MAX_PRINT_CHUNK);
- netpoll_send_udp(&nt->np, msg, frag);
- msg += frag;
- left -= frag;
+ struct netconsole_target *nt;
+ const char *tmp;
+
+ /* Avoid taking lock and disabling interrupts unnecessarily */
+ if (list_empty(&target_list))
+ return;
+
+ spin_lock_irqsave(&target_list_lock, flags);
+ list_for_each_entry(nt, &target_list, list) {
+ if (netif_running(nt->np.dev)) {
+ /*
+ * We nest this inside the for-each-target loop above
+ * so that we're able to get as much logging out to
+ * at least one target if we die inside here, instead
+ * of unnecessarily keeping all targets in lock-step.
+ */
+ tmp = msg;
+ for (left = len; left;) {
+ frag = min(left, MAX_PRINT_CHUNK);
+ netpoll_send_udp(&nt->np, tmp, frag);
+ tmp += frag;
+ left -= frag;
+ }
}
- local_irq_restore(flags);
}
+ spin_unlock_irqrestore(&target_list_lock, flags);
}
static struct console netconsole = {
@@ -134,39 +197,67 @@ static struct console netconsole = {
static int __init init_netconsole(void)
{
int err = 0;
- struct netconsole_target *nt = &default_target;
+ struct netconsole_target *nt, *tmp;
+ unsigned long flags;
+ char *target_config;
+ char *input = config;
- if (!strnlen(config, MAX_PARAM_LENGTH)) {
+ if (!strnlen(input, MAX_PARAM_LENGTH)) {
printk(KERN_INFO "netconsole: not configured, aborting\n");
goto out;
}
- err = netpoll_parse_options(&nt->np, config);
- if (err)
- goto out;
-
- err = netpoll_setup(&nt->np);
- if (err)
- goto out;
+ while ((target_config = strsep(&input, ";"))) {
+ nt = alloc_target(target_config);
+ if (IS_ERR(nt)) {
+ err = PTR_ERR(nt);
+ goto fail;
+ }
+ spin_lock_irqsave(&target_list_lock, flags);
+ list_add(&nt->list, &target_list);
+ spin_unlock_irqrestore(&target_list_lock, flags);
+ }
err = register_netdevice_notifier(&netconsole_netdev_notifier);
if (err)
- goto out;
+ goto fail;
register_console(&netconsole);
printk(KERN_INFO "netconsole: network logging started\n");
out:
return err;
+
+fail:
+ printk(KERN_ERR "netconsole: cleaning up\n");
+
+ /*
+ * Remove all targets and destroy them. Skipping the list
+ * lock is safe here, and netpoll_cleanup() will sleep.
+ */
+ list_for_each_entry_safe(nt, tmp, &target_list, list) {
+ list_del(&nt->list);
+ free_target(nt);
+ }
+
+ return err;
}
static void __exit cleanup_netconsole(void)
{
- struct netconsole_target *nt = &default_target;
+ struct netconsole_target *nt, *tmp;
unregister_console(&netconsole);
unregister_netdevice_notifier(&netconsole_netdev_notifier);
- netpoll_cleanup(&nt->np);
+
+ /*
+ * Remove all targets and destroy them. Skipping the list
+ * lock is safe here, and netpoll_cleanup() will sleep.
+ */
+ list_for_each_entry_safe(nt, tmp, &target_list, list) {
+ list_del(&nt->list);
+ free_target(nt);
+ }
}
module_init(init_netconsole);
^ permalink raw reply related
* [PATCH v3 -mm 9/9] netconsole: Support dynamic reconfiguration using configfs
From: Satyam Sharma @ 2007-07-30 2:49 UTC (permalink / raw)
To: Linux Kernel Mailing List
Cc: Keiichi Kii, Netdev, Joel Becker, Matt Mackall, Andrew Morton,
David Miller
In-Reply-To: <20070730024741.10828.48209.sendpatchset@enigma.security.iitk.ac.in>
From: Satyam Sharma <satyam@infradead.org>
[9/9] netconsole: Support dynamic reconfiguration using configfs
This patch introduces support for dynamic reconfiguration (adding, removing
and/or modifying parameters of netconsole targets at runtime) using a
userspace interface exported via configfs. Documentation is also updated
accordingly.
Issues and brief design overview:
(1) Kernel-initiated creation / destruction of kernel objects is not
possible with configfs -- the lifetimes of the "config items" is managed
exclusively from userspace. But netconsole must support boot/module params
too, and these are parsed in kernel and hence netpolls must be setup from
the kernel. Joel Becker suggested to separately manage the lifetimes of
the two kinds of netconsole_target objects -- those created via configfs
mkdir(2) from userspace and those specified from the boot/module option
string. This adds complexity and some redundancy here and also means that
boot/module param-created targets are not exposed through the configfs
namespace (and hence cannot be updated / destroyed dynamically). However,
this saves us from locking / refcounting complexities that would need to
be introduced in configfs to support kernel-initiated item creation /
destroy there.
(2) In configfs, item creation takes place in the call chain of the mkdir(2)
syscall in the driver subsystem. If we used an ioctl(2) to create / destroy
objects from userspace, the special userspace program is able to fill out
the structure to be passed into the ioctl and hence specify attributes such
as local interface that are required at the time we set up the netpoll.
For configfs, this information is not available at the time of mkdir(2).
So, we keep all newly-created targets (via configfs) disabled by default.
The user is expected to set various attributes appropriately (including the
local network interface if required) and then write(2) "1" to the "enabled"
attribute. Thus, netpoll_setup() is then called on the set parameters in the
context of _this_ write(2) on the "enabled" attribute itself. This design
enables the user to reconfigure existing netconsole targets at runtime to
be attached to newly-come-up interfaces that may not have existed when
netconsole was loaded or when the targets were actually created. All this
effectively enables us to get rid of custom ioctls.
(3) Ultra-paranoid configfs attribute show() and store() operations, with
sanity and input range checking, using only safe string primitives, and
compliant with the recommendations in Documentation/filesystems/sysfs.txt.
(4) A new function netpoll_print_options() is created in the netpoll API,
that just prints out the configured parameters for a netpoll structure.
netpoll_parse_options() is modified to use that and it is also exported to
be used from netconsole.
Signed-off-by: Satyam Sharma <satyam@infradead.org>
Acked-by: Keiichi Kii <k-keiichi@bx.jp.nec.com>
---
Documentation/networking/netconsole.txt | 68 ++++
drivers/net/Kconfig | 10 +
drivers/net/netconsole.c | 605 +++++++++++++++++++++++++++++--
include/linux/netpoll.h | 1 +
net/core/netpoll.c | 44 ++-
5 files changed, 683 insertions(+), 45 deletions(-)
diff --git a/Documentation/networking/netconsole.txt b/Documentation/networking/netconsole.txt
index 1aaa738..3c2f2b3 100644
--- a/Documentation/networking/netconsole.txt
+++ b/Documentation/networking/netconsole.txt
@@ -3,6 +3,10 @@ started by Ingo Molnar <mingo@redhat.com>, 2001.09.17
2.6 port and netpoll api by Matt Mackall <mpm@selenic.com>, Sep 9 2003
Please send bug reports to Matt Mackall <mpm@selenic.com>
+and Satyam Sharma <satyam.sharma@gmail.com>
+
+Introduction:
+=============
This module logs kernel printk messages over UDP allowing debugging of
problem where disk logging fails and serial consoles are impractical.
@@ -13,6 +17,9 @@ the specified interface as soon as possible. While this doesn't allow
capture of early kernel panics, it does capture most of the boot
process.
+Sender and receiver configuration:
+==================================
+
It takes a string configuration parameter "netconsole" in the
following format:
@@ -46,6 +53,67 @@ address.
The remote host can run either 'netcat -u -l -p <port>' or syslogd.
+Dynamic reconfiguration:
+========================
+
+Dynamic reconfigurability is a useful addition to netconsole that enables
+remote logging targets to be dynamically added, removed, or have their
+parameters reconfigured at runtime from a configfs-based userspace interface.
+[ Note that the parameters of netconsole targets that were specified/created
+from the boot/module option are not exposed via this interface, and hence
+cannot be modified dynamically. ]
+
+To include this feature, select CONFIG_NETCONSOLE_DYNAMIC when building the
+netconsole module (or kernel, if netconsole is built-in).
+
+Some examples follow (where configfs is mounted at the /sys/kernel/config
+mountpoint).
+
+To add a remote logging target (target names can be arbitrary):
+
+ cd /sys/kernel/config/netconsole/
+ mkdir target1
+
+Note that newly created targets have default parameter values (as mentioned
+above) and are disabled by default -- they must first be enabled by writing
+"1" to the "enabled" attribute (usually after setting parameters accordingly)
+as described below.
+
+To remove a target:
+
+ rmdir /sys/kernel/config/netconsole/othertarget/
+
+The interface exposes these parameters of a netconsole target to userspace:
+
+ enabled Is this target currently enabled? (read-write)
+ dev_name Local network interface name (read-write)
+ local_port Source UDP port to use (read-write)
+ remote_port Remote agent's UDP port (read-write)
+ local_ip Source IP address to use (read-write)
+ remote_ip Remote agent's IP address (read-write)
+ local_mac Local interface's MAC address (read-only)
+ remote_mac Remote agent's MAC address (read-write)
+
+The "enabled" attribute is also used to control whether the parameters of
+a target can be updated or not -- you can modify the parameters of only
+disabled targets (i.e. if "enabled" is 0).
+
+To update a target's parameters:
+
+ cat enabled # check if enabled is 1
+ echo 0 > enabled # disable the target (if required)
+ echo eth2 > dev_name # set local interface
+ echo 10.0.0.4 > remote_ip # update some parameter
+ echo cb:a9:87:65:43:21 > remote_mac # update more parameters
+ echo 1 > enabled # enable target again
+
+You can also update the local interface dynamically. This is especially
+useful if you want to use interfaces that have newly come up (and may not
+have existed when netconsole was loaded / initialized).
+
+Miscellaneous notes:
+====================
+
WARNING: the default target ethernet setting uses the broadcast
ethernet address to send packets, which can cause increased load on
other systems on the same ethernet segment.
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index f8a602c..89c21d7 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2972,6 +2972,16 @@ config NETCONSOLE
If you want to log kernel messages over the network, enable this.
See <file:Documentation/networking/netconsole.txt> for details.
+config NETCONSOLE_DYNAMIC
+ bool "Dynamic reconfiguration of logging targets (EXPERIMENTAL)"
+ depends on NETCONSOLE && SYSFS && EXPERIMENTAL
+ select CONFIGFS_FS
+ help
+ This option enables the ability to dynamically reconfigure target
+ parameters (interface, IP addresses, port numbers, MAC addresses)
+ at runtime through a userspace interface exported using configfs.
+ See <file:Documentation/networking/netconsole.txt> for details.
+
config NETPOLL
def_bool NETCONSOLE
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index 458c4d6..0269323 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -41,6 +41,8 @@
#include <linux/moduleparam.h>
#include <linux/string.h>
#include <linux/netpoll.h>
+#include <linux/inet.h>
+#include <linux/configfs.h>
MODULE_AUTHOR("Maintainer: Matt Mackall <mpm@selenic.com>");
MODULE_DESCRIPTION("Console driver for network interfaces");
@@ -71,20 +73,100 @@ static DEFINE_SPINLOCK(target_list_lock);
/**
* struct netconsole_target - Represents a configured netconsole target.
* @list: Links this target into the target_list.
+ * @item: Links us into the configfs subsystem hierarchy.
+ * @enabled: On / off knob to enable / disable target.
+ * Visible from userspace (read-write).
+ * We maintain a strict 1:1 correspondence between this and
+ * whether the corresponding netpoll is active or inactive.
+ * Also, other parameters of a target may be modified at
+ * runtime only when it is disabled (enabled == 0).
* @np: The netpoll structure for this target.
+ * Contains the other userspace visible parameters:
+ * dev_name (read-write)
+ * local_port (read-write)
+ * remote_port (read-write)
+ * local_ip (read-write)
+ * remote_ip (read-write)
+ * local_mac (read-only)
+ * remote_mac (read-write)
*/
struct netconsole_target {
struct list_head list;
+#ifdef CONFIG_NETCONSOLE_DYNAMIC
+ struct config_item item;
+#endif
+ int enabled;
struct netpoll np;
};
-/* Allocate new target and setup netpoll for it */
-static struct netconsole_target *alloc_target(char *target_config)
+#ifdef CONFIG_NETCONSOLE_DYNAMIC
+
+static struct configfs_subsystem netconsole_subsys;
+
+static int __init dynamic_netconsole_init(void)
+{
+ config_group_init(&netconsole_subsys.su_group);
+ mutex_init(&netconsole_subsys.su_mutex);
+ return configfs_register_subsystem(&netconsole_subsys);
+}
+
+static void __exit dynamic_netconsole_exit(void)
+{
+ configfs_unregister_subsystem(&netconsole_subsys);
+}
+
+/*
+ * Targets that were created by parsing the boot/module option string
+ * do not exist in the configfs hierarchy (and have NULL names) and will
+ * never go away, so make these a no-op for them.
+ */
+static void netconsole_target_get(struct netconsole_target *nt)
+{
+ if (config_item_name(&nt->item))
+ config_item_get(&nt->item);
+}
+
+static void netconsole_target_put(struct netconsole_target *nt)
+{
+ if (config_item_name(&nt->item))
+ config_item_put(&nt->item);
+}
+
+#else /* !CONFIG_NETCONSOLE_DYNAMIC */
+
+static int __init dynamic_netconsole_init(void)
+{
+ return 0;
+}
+
+static void __exit dynamic_netconsole_exit(void)
+{
+}
+
+/*
+ * No danger of targets going away from under us when dynamic
+ * reconfigurability is off.
+ */
+static void netconsole_target_get(struct netconsole_target *nt)
+{
+}
+
+static void netconsole_target_put(struct netconsole_target *nt)
+{
+}
+
+#endif /* CONFIG_NETCONSOLE_DYNAMIC */
+
+/* Allocate new target (from boot/module param) and setup netpoll for it */
+static struct netconsole_target *alloc_param_target(char *target_config)
{
int err = -ENOMEM;
struct netconsole_target *nt;
- /* Allocate and initialize with defaults */
+ /*
+ * Allocate and initialize with defaults.
+ * Note that these targets get their config_item fields zeroed-out.
+ */
nt = kzalloc(sizeof(*nt), GFP_KERNEL);
if (!nt) {
printk(KERN_ERR "netconsole: failed to allocate memory\n");
@@ -106,6 +188,8 @@ static struct netconsole_target *alloc_target(char *target_config)
if (err)
goto fail;
+ nt->enabled = 1;
+
return nt;
fail:
@@ -113,13 +197,469 @@ fail:
return ERR_PTR(err);
}
-/* Cleanup netpoll for given target and free it */
-static void free_target(struct netconsole_target *nt)
+/* Cleanup netpoll for given target (from boot/module param) and free it */
+static void free_param_target(struct netconsole_target *nt)
{
netpoll_cleanup(&nt->np);
kfree(nt);
}
+#ifdef CONFIG_NETCONSOLE_DYNAMIC
+
+/*
+ * Our subsystem hierarchy is:
+ *
+ * /sys/kernel/config/netconsole/
+ * |
+ * <target>/
+ * | enabled
+ * | dev_name
+ * | local_port
+ * | remote_port
+ * | local_ip
+ * | remote_ip
+ * | local_mac
+ * | remote_mac
+ * |
+ * <target>/...
+ */
+
+struct netconsole_target_attr {
+ struct configfs_attribute attr;
+ ssize_t (*show)(struct netconsole_target *nt,
+ char *buf);
+ ssize_t (*store)(struct netconsole_target *nt,
+ const char *buf,
+ size_t count);
+};
+
+static struct netconsole_target *to_target(struct config_item *item)
+{
+ return item ?
+ container_of(item, struct netconsole_target, item) :
+ NULL;
+}
+
+/*
+ * Wrapper over simple_strtol (base 10) with sanity and range checking.
+ * We return (signed) long only because we may want to return errors.
+ * Do not use this to convert numbers that are allowed to be negative.
+ */
+static long strtol10_check_range(const char *cp, long min, long max)
+{
+ long ret;
+ char *p = (char *) cp;
+
+ WARN_ON(min < 0);
+ WARN_ON(max < min);
+
+ ret = simple_strtol(p, &p, 10);
+
+ if (*p && (*p != '\n')) {
+ printk(KERN_ERR "netconsole: invalid input\n");
+ return -EINVAL;
+ }
+ if ((ret < min) || (ret > max)) {
+ printk(KERN_ERR "netconsole: input %ld must be between "
+ "%ld and %ld\n", ret, min, max);
+ return -EINVAL;
+ }
+
+ return ret;
+}
+
+/*
+ * Attribute operations for netconsole_target.
+ */
+
+static ssize_t show_enabled(struct netconsole_target *nt, char *buf)
+{
+ return snprintf(buf, PAGE_SIZE, "%d\n", nt->enabled);
+}
+
+static ssize_t show_dev_name(struct netconsole_target *nt, char *buf)
+{
+ return snprintf(buf, PAGE_SIZE, "%s\n", nt->np.dev_name);
+}
+
+static ssize_t show_local_port(struct netconsole_target *nt, char *buf)
+{
+ return snprintf(buf, PAGE_SIZE, "%d\n", nt->np.local_port);
+}
+
+static ssize_t show_remote_port(struct netconsole_target *nt, char *buf)
+{
+ return snprintf(buf, PAGE_SIZE, "%d\n", nt->np.remote_port);
+}
+
+static ssize_t show_local_ip(struct netconsole_target *nt, char *buf)
+{
+ return snprintf(buf, PAGE_SIZE, "%d.%d.%d.%d\n",
+ HIPQUAD(nt->np.local_ip));
+}
+
+static ssize_t show_remote_ip(struct netconsole_target *nt, char *buf)
+{
+ return snprintf(buf, PAGE_SIZE, "%d.%d.%d.%d\n",
+ HIPQUAD(nt->np.remote_ip));
+}
+
+static ssize_t show_local_mac(struct netconsole_target *nt, char *buf)
+{
+ return snprintf(buf, PAGE_SIZE, "%02x:%02x:%02x:%02x:%02x:%02x\n",
+ nt->np.local_mac[0], nt->np.local_mac[1],
+ nt->np.local_mac[2], nt->np.local_mac[3],
+ nt->np.local_mac[4], nt->np.local_mac[5]);
+}
+
+static ssize_t show_remote_mac(struct netconsole_target *nt, char *buf)
+{
+ return snprintf(buf, PAGE_SIZE, "%02x:%02x:%02x:%02x:%02x:%02x\n",
+ nt->np.remote_mac[0], nt->np.remote_mac[1],
+ nt->np.remote_mac[2], nt->np.remote_mac[3],
+ nt->np.remote_mac[4], nt->np.remote_mac[5]);
+}
+
+/*
+ * This one is special -- targets created through the configfs interface
+ * are not enabled (and the corresponding netpoll activated) by default.
+ * The user is expected to set the desired parameters first (which
+ * would enable him to dynamically add new netpoll targets for new
+ * network interfaces as and when they come up).
+ */
+static ssize_t store_enabled(struct netconsole_target *nt,
+ const char *buf,
+ size_t count)
+{
+ int err;
+ long enabled;
+
+ enabled = strtol10_check_range(buf, 0, 1);
+ if (enabled < 0)
+ return enabled;
+
+ if (enabled) { /* 1 */
+
+ /*
+ * Skip netpoll_parse_options() -- all the attributes are
+ * already configured via configfs. Just print them out.
+ */
+ netpoll_print_options(&nt->np);
+
+ err = netpoll_setup(&nt->np);
+ if (err)
+ return err;
+
+ printk(KERN_INFO "netconsole: network logging started\n");
+
+ } else { /* 0 */
+ netpoll_cleanup(&nt->np);
+ }
+
+ nt->enabled = enabled;
+
+ return strnlen(buf, count);
+}
+
+static ssize_t store_dev_name(struct netconsole_target *nt,
+ const char *buf,
+ size_t count)
+{
+ size_t len;
+
+ if (nt->enabled) {
+ printk(KERN_ERR "netconsole: target (%s) is enabled, "
+ "disable to update parameters\n",
+ config_item_name(&nt->item));
+ return -EINVAL;
+ }
+
+ strlcpy(nt->np.dev_name, buf, IFNAMSIZ);
+
+ /* Get rid of possible trailing newline from echo(1) */
+ len = strnlen(nt->np.dev_name, IFNAMSIZ);
+ if (nt->np.dev_name[len - 1] == '\n')
+ nt->np.dev_name[len - 1] = '\0';
+
+ return strnlen(buf, count);
+}
+
+static ssize_t store_local_port(struct netconsole_target *nt,
+ const char *buf,
+ size_t count)
+{
+ long local_port;
+#define __U16_MAX ((__u16) ~0U)
+
+ if (nt->enabled) {
+ printk(KERN_ERR "netconsole: target (%s) is enabled, "
+ "disable to update parameters\n",
+ config_item_name(&nt->item));
+ return -EINVAL;
+ }
+
+ local_port = strtol10_check_range(buf, 0, __U16_MAX);
+ if (local_port < 0)
+ return local_port;
+
+ nt->np.local_port = local_port;
+
+ return strnlen(buf, count);
+}
+
+static ssize_t store_remote_port(struct netconsole_target *nt,
+ const char *buf,
+ size_t count)
+{
+ long remote_port;
+#define __U16_MAX ((__u16) ~0U)
+
+ if (nt->enabled) {
+ printk(KERN_ERR "netconsole: target (%s) is enabled, "
+ "disable to update parameters\n",
+ config_item_name(&nt->item));
+ return -EINVAL;
+ }
+
+ remote_port = strtol10_check_range(buf, 0, __U16_MAX);
+ if (remote_port < 0)
+ return remote_port;
+
+ nt->np.remote_port = remote_port;
+
+ return strnlen(buf, count);
+}
+
+static ssize_t store_local_ip(struct netconsole_target *nt,
+ const char *buf,
+ size_t count)
+{
+ if (nt->enabled) {
+ printk(KERN_ERR "netconsole: target (%s) is enabled, "
+ "disable to update parameters\n",
+ config_item_name(&nt->item));
+ return -EINVAL;
+ }
+
+ nt->np.local_ip = ntohl(in_aton(buf));
+
+ return strnlen(buf, count);
+}
+
+static ssize_t store_remote_ip(struct netconsole_target *nt,
+ const char *buf,
+ size_t count)
+{
+ if (nt->enabled) {
+ printk(KERN_ERR "netconsole: target (%s) is enabled, "
+ "disable to update parameters\n",
+ config_item_name(&nt->item));
+ return -EINVAL;
+ }
+
+ nt->np.remote_ip = ntohl(in_aton(buf));
+
+ return strnlen(buf, count);
+}
+
+static ssize_t store_remote_mac(struct netconsole_target *nt,
+ const char *buf,
+ size_t count)
+{
+ u8 remote_mac[ETH_ALEN];
+ char *p = (char *) buf;
+ int i;
+
+ if (nt->enabled) {
+ printk(KERN_ERR "netconsole: target (%s) is enabled, "
+ "disable to update parameters\n",
+ config_item_name(&nt->item));
+ return -EINVAL;
+ }
+
+ for (i = 0; i < ETH_ALEN - 1; i++) {
+ remote_mac[i] = simple_strtoul(p, &p, 16);
+ if (*p != ':')
+ goto invalid;
+ p++;
+ }
+ remote_mac[ETH_ALEN - 1] = simple_strtoul(p, &p, 16);
+ if (*p && (*p != '\n'))
+ goto invalid;
+
+ memcpy(nt->np.remote_mac, remote_mac, ETH_ALEN);
+
+ return strnlen(buf, count);
+
+invalid:
+ printk(KERN_ERR "netconsole: invalid input\n");
+ return -EINVAL;
+}
+
+/*
+ * Attribute definitions for netconsole_target.
+ */
+
+#define NETCONSOLE_TARGET_ATTR_RO(_name) \
+static struct netconsole_target_attr netconsole_target_##_name = \
+ __CONFIGFS_ATTR(_name, S_IRUGO, show_##_name, NULL)
+
+#define NETCONSOLE_TARGET_ATTR_RW(_name) \
+static struct netconsole_target_attr netconsole_target_##_name = \
+ __CONFIGFS_ATTR(_name, S_IRUGO | S_IWUSR, show_##_name, store_##_name)
+
+NETCONSOLE_TARGET_ATTR_RW(enabled);
+NETCONSOLE_TARGET_ATTR_RW(dev_name);
+NETCONSOLE_TARGET_ATTR_RW(local_port);
+NETCONSOLE_TARGET_ATTR_RW(remote_port);
+NETCONSOLE_TARGET_ATTR_RW(local_ip);
+NETCONSOLE_TARGET_ATTR_RW(remote_ip);
+NETCONSOLE_TARGET_ATTR_RO(local_mac);
+NETCONSOLE_TARGET_ATTR_RW(remote_mac);
+
+static struct configfs_attribute *netconsole_target_attrs[] = {
+ &netconsole_target_enabled.attr,
+ &netconsole_target_dev_name.attr,
+ &netconsole_target_local_port.attr,
+ &netconsole_target_remote_port.attr,
+ &netconsole_target_local_ip.attr,
+ &netconsole_target_remote_ip.attr,
+ &netconsole_target_local_mac.attr,
+ &netconsole_target_remote_mac.attr,
+ NULL,
+};
+
+/*
+ * Item operations and type for netconsole_target.
+ */
+
+static void netconsole_target_release(struct config_item *item)
+{
+ kfree(to_target(item));
+}
+
+static ssize_t netconsole_target_attr_show(struct config_item *item,
+ struct configfs_attribute *attr,
+ char *buf)
+{
+ ssize_t ret = -EINVAL;
+ struct netconsole_target *nt = to_target(item);
+ struct netconsole_target_attr *na =
+ container_of(attr, struct netconsole_target_attr, attr);
+
+ if (na->show)
+ ret = na->show(nt, buf);
+
+ return ret;
+}
+
+static ssize_t netconsole_target_attr_store(struct config_item *item,
+ struct configfs_attribute *attr,
+ const char *buf,
+ size_t count)
+{
+ ssize_t ret = -EINVAL;
+ struct netconsole_target *nt = to_target(item);
+ struct netconsole_target_attr *na =
+ container_of(attr, struct netconsole_target_attr, attr);
+
+ if (na->store)
+ ret = na->store(nt, buf, count);
+
+ return ret;
+}
+
+static struct configfs_item_operations netconsole_target_item_ops = {
+ .release = netconsole_target_release,
+ .show_attribute = netconsole_target_attr_show,
+ .store_attribute = netconsole_target_attr_store,
+};
+
+static struct config_item_type netconsole_target_type = {
+ .ct_attrs = netconsole_target_attrs,
+ .ct_item_ops = &netconsole_target_item_ops,
+ .ct_owner = THIS_MODULE,
+};
+
+/*
+ * Group operations and type for netconsole_subsys.
+ */
+
+static struct config_item *make_netconsole_target(struct config_group *group,
+ const char *name)
+{
+ unsigned long flags;
+ struct netconsole_target *nt;
+
+ /*
+ * Allocate and initialize with defaults.
+ * Target is disabled at creation (enabled == 0).
+ */
+ nt = kzalloc(sizeof(*nt), GFP_KERNEL);
+ if (!nt) {
+ printk(KERN_ERR "netconsole: failed to allocate memory\n");
+ return NULL;
+ }
+
+ nt->np.name = "netconsole";
+ strlcpy(nt->np.dev_name, "eth0", IFNAMSIZ);
+ nt->np.local_port = 6665;
+ nt->np.remote_port = 6666;
+ memset(nt->np.remote_mac, 0xff, ETH_ALEN);
+
+ /* Initialize the config_item member */
+ config_item_init_type_name(&nt->item, name, &netconsole_target_type);
+
+ /* Adding, but it is disabled */
+ spin_lock_irqsave(&target_list_lock, flags);
+ list_add(&nt->list, &target_list);
+ spin_unlock_irqrestore(&target_list_lock, flags);
+
+ return &nt->item;
+}
+
+static void drop_netconsole_target(struct config_group *group,
+ struct config_item *item)
+{
+ unsigned long flags;
+ struct netconsole_target *nt = to_target(item);
+
+ spin_lock_irqsave(&target_list_lock, flags);
+ list_del(&nt->list);
+ spin_unlock_irqrestore(&target_list_lock, flags);
+
+ /*
+ * The target may have never been enabled, or was manually disabled
+ * before being removed so netpoll may have already been cleaned up.
+ */
+ if (nt->enabled)
+ netpoll_cleanup(&nt->np);
+
+ config_item_put(&nt->item);
+}
+
+static struct configfs_group_operations netconsole_subsys_group_ops = {
+ .make_item = make_netconsole_target,
+ .drop_item = drop_netconsole_target,
+};
+
+static struct config_item_type netconsole_subsys_type = {
+ .ct_group_ops = &netconsole_subsys_group_ops,
+ .ct_owner = THIS_MODULE,
+};
+
+/* The netconsole configfs subsystem */
+static struct configfs_subsystem netconsole_subsys = {
+ .su_group = {
+ .cg_item = {
+ .ci_namebuf = "netconsole",
+ .ci_type = &netconsole_subsys_type,
+ },
+ },
+};
+
+#endif /* CONFIG_NETCONSOLE_DYNAMIC */
+
/* Handle network interface device notifications */
static int netconsole_netdev_event(struct notifier_block *this,
unsigned long event,
@@ -134,6 +674,7 @@ static int netconsole_netdev_event(struct notifier_block *this,
spin_lock_irqsave(&target_list_lock, flags);
list_for_each_entry(nt, &target_list, list) {
+ netconsole_target_get(nt);
if (nt->np.dev == dev) {
switch (event) {
case NETDEV_CHANGEADDR:
@@ -145,6 +686,7 @@ static int netconsole_netdev_event(struct notifier_block *this,
break;
}
}
+ netconsole_target_put(nt);
}
spin_unlock_irqrestore(&target_list_lock, flags);
@@ -169,7 +711,8 @@ static void write_msg(struct console *con, const char *msg, unsigned int len)
spin_lock_irqsave(&target_list_lock, flags);
list_for_each_entry(nt, &target_list, list) {
- if (netif_running(nt->np.dev)) {
+ netconsole_target_get(nt);
+ if (nt->enabled && netif_running(nt->np.dev)) {
/*
* We nest this inside the for-each-target loop above
* so that we're able to get as much logging out to
@@ -184,6 +727,7 @@ static void write_msg(struct console *con, const char *msg, unsigned int len)
left -= frag;
}
}
+ netconsole_target_put(nt);
}
spin_unlock_irqrestore(&target_list_lock, flags);
}
@@ -196,48 +740,52 @@ static struct console netconsole = {
static int __init init_netconsole(void)
{
- int err = 0;
+ int err;
struct netconsole_target *nt, *tmp;
unsigned long flags;
char *target_config;
char *input = config;
- if (!strnlen(input, MAX_PARAM_LENGTH)) {
- printk(KERN_INFO "netconsole: not configured, aborting\n");
- goto out;
- }
-
- while ((target_config = strsep(&input, ";"))) {
- nt = alloc_target(target_config);
- if (IS_ERR(nt)) {
- err = PTR_ERR(nt);
- goto fail;
+ if (strnlen(input, MAX_PARAM_LENGTH)) {
+ while ((target_config = strsep(&input, ";"))) {
+ nt = alloc_param_target(target_config);
+ if (IS_ERR(nt)) {
+ err = PTR_ERR(nt);
+ goto fail;
+ }
+ spin_lock_irqsave(&target_list_lock, flags);
+ list_add(&nt->list, &target_list);
+ spin_unlock_irqrestore(&target_list_lock, flags);
}
- spin_lock_irqsave(&target_list_lock, flags);
- list_add(&nt->list, &target_list);
- spin_unlock_irqrestore(&target_list_lock, flags);
}
err = register_netdevice_notifier(&netconsole_netdev_notifier);
if (err)
goto fail;
+ err = dynamic_netconsole_init();
+ if (err)
+ goto undonotifier;
+
register_console(&netconsole);
printk(KERN_INFO "netconsole: network logging started\n");
-out:
return err;
+undonotifier:
+ unregister_netdevice_notifier(&netconsole_netdev_notifier);
+
fail:
printk(KERN_ERR "netconsole: cleaning up\n");
/*
- * Remove all targets and destroy them. Skipping the list
+ * Remove all targets and destroy them (only targets created
+ * from the boot/module option exist here). Skipping the list
* lock is safe here, and netpoll_cleanup() will sleep.
*/
list_for_each_entry_safe(nt, tmp, &target_list, list) {
list_del(&nt->list);
- free_target(nt);
+ free_param_target(nt);
}
return err;
@@ -248,15 +796,20 @@ static void __exit cleanup_netconsole(void)
struct netconsole_target *nt, *tmp;
unregister_console(&netconsole);
+ dynamic_netconsole_exit();
unregister_netdevice_notifier(&netconsole_netdev_notifier);
/*
- * Remove all targets and destroy them. Skipping the list
- * lock is safe here, and netpoll_cleanup() will sleep.
+ * Targets created via configfs pin references on our module
+ * and would first be rmdir(2)'ed from userspace. We reach
+ * here only when they are already destroyed, and only those
+ * created from the boot/module option are left, so remove and
+ * destroy them. Skipping the list lock is safe here, and
+ * netpoll_cleanup() will sleep.
*/
list_for_each_entry_safe(nt, tmp, &target_list, list) {
list_del(&nt->list);
- free_target(nt);
+ free_param_target(nt);
}
}
diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index 29930b7..e0e0a7b 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -37,6 +37,7 @@ struct netpoll_info {
void netpoll_poll(struct netpoll *np);
void netpoll_send_udp(struct netpoll *np, const char *msg, int len);
+void netpoll_print_options(struct netpoll *np);
int netpoll_parse_options(struct netpoll *np, char *opt);
int netpoll_setup(struct netpoll *np);
int netpoll_trap(void);
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index de1b26a..3f753c6 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -519,6 +519,29 @@ out:
return 0;
}
+void netpoll_print_options(struct netpoll *np)
+{
+ printk(KERN_INFO "%s: local port %d\n",
+ np->name, np->local_port);
+ printk(KERN_INFO "%s: local IP %d.%d.%d.%d\n",
+ np->name, HIPQUAD(np->local_ip));
+ printk(KERN_INFO "%s: interface %s\n",
+ np->name, np->dev_name);
+ printk(KERN_INFO "%s: remote port %d\n",
+ np->name, np->remote_port);
+ printk(KERN_INFO "%s: remote IP %d.%d.%d.%d\n",
+ np->name, HIPQUAD(np->remote_ip));
+ printk(KERN_INFO "%s: remote ethernet address "
+ "%02x:%02x:%02x:%02x:%02x:%02x\n",
+ np->name,
+ np->remote_mac[0],
+ np->remote_mac[1],
+ np->remote_mac[2],
+ np->remote_mac[3],
+ np->remote_mac[4],
+ np->remote_mac[5]);
+}
+
int netpoll_parse_options(struct netpoll *np, char *opt)
{
char *cur=opt, *delim;
@@ -531,7 +554,6 @@ int netpoll_parse_options(struct netpoll *np, char *opt)
cur = delim;
}
cur++;
- printk(KERN_INFO "%s: local port %d\n", np->name, np->local_port);
if (*cur != '/') {
if ((delim = strchr(cur, '/')) == NULL)
@@ -539,9 +561,6 @@ int netpoll_parse_options(struct netpoll *np, char *opt)
*delim = 0;
np->local_ip = ntohl(in_aton(cur));
cur = delim;
-
- printk(KERN_INFO "%s: local IP %d.%d.%d.%d\n",
- np->name, HIPQUAD(np->local_ip));
}
cur++;
@@ -555,8 +574,6 @@ int netpoll_parse_options(struct netpoll *np, char *opt)
}
cur++;
- printk(KERN_INFO "%s: interface %s\n", np->name, np->dev_name);
-
if (*cur != '@') {
/* dst port */
if ((delim = strchr(cur, '@')) == NULL)
@@ -566,7 +583,6 @@ int netpoll_parse_options(struct netpoll *np, char *opt)
cur = delim;
}
cur++;
- printk(KERN_INFO "%s: remote port %d\n", np->name, np->remote_port);
/* dst ip */
if ((delim = strchr(cur, '/')) == NULL)
@@ -575,9 +591,6 @@ int netpoll_parse_options(struct netpoll *np, char *opt)
np->remote_ip = ntohl(in_aton(cur));
cur = delim + 1;
- printk(KERN_INFO "%s: remote IP %d.%d.%d.%d\n",
- np->name, HIPQUAD(np->remote_ip));
-
if (*cur != 0) {
/* MAC address */
if ((delim = strchr(cur, ':')) == NULL)
@@ -608,15 +621,7 @@ int netpoll_parse_options(struct netpoll *np, char *opt)
np->remote_mac[5] = simple_strtol(cur, NULL, 16);
}
- printk(KERN_INFO "%s: remote ethernet address "
- "%02x:%02x:%02x:%02x:%02x:%02x\n",
- np->name,
- np->remote_mac[0],
- np->remote_mac[1],
- np->remote_mac[2],
- np->remote_mac[3],
- np->remote_mac[4],
- np->remote_mac[5]);
+ netpoll_print_options(np);
return 0;
@@ -820,6 +825,7 @@ void netpoll_set_trap(int trap)
EXPORT_SYMBOL(netpoll_set_trap);
EXPORT_SYMBOL(netpoll_trap);
+EXPORT_SYMBOL(netpoll_print_options);
EXPORT_SYMBOL(netpoll_parse_options);
EXPORT_SYMBOL(netpoll_setup);
EXPORT_SYMBOL(netpoll_cleanup);
^ permalink raw reply related
* Re: 2.6.20->2.6.21 - networking dies after random time
From: Marcin Ślusarz @ 2007-07-30 7:29 UTC (permalink / raw)
To: Ingo Molnar, Jarek Poplawski, Thomas Gleixner, Linus Torvalds,
Jean-Baptiste Vignaud, linux-kernel, shemminger, linux-net,
netdev, Andrew Morton, Alan Cox
In-Reply-To: <20070726091254.GA8063@elte.hu>
2007/7/26, Ingo Molnar <mingo@elte.hu>:
> (..)
> yeah - i meant to cover both arches but forgot about x86_64 - updated
> patch attached below.
>
> Ingo
>
> ----------------->
> Subject: x86: activate HARDIRQS_SW_RESEND
> From: Ingo Molnar <mingo@elte.hu>
>
> activate the software-triggered IRQ-resend logic.
>
> it appears some chipsets/cpus do not handle local-APIC driven IRQ
> resends all that well, so always use the soft mechanism to trigger
> the execution of pending interrupts.
>
> Signed-off-by: Ingo Molnar <mingo@elte.hu>
> ---
> arch/i386/Kconfig | 4 ++++
> arch/x86_64/Kconfig | 4 ++++
> kernel/irq/manage.c | 8 ++++++++
> 3 files changed, 16 insertions(+)
>
> Index: linux-rt-rebase.q/arch/i386/Kconfig
> ===================================================================
> --- linux-rt-rebase.q.orig/arch/i386/Kconfig
> +++ linux-rt-rebase.q/arch/i386/Kconfig
> @@ -1284,6 +1284,10 @@ config GENERIC_PENDING_IRQ
> depends on GENERIC_HARDIRQS && SMP
> default y
>
> +config HARDIRQS_SW_RESEND
> + bool
> + default y
> +
> config X86_SMP
> bool
> depends on SMP && !X86_VOYAGER
> Index: linux-rt-rebase.q/arch/x86_64/Kconfig
> ===================================================================
> --- linux-rt-rebase.q.orig/arch/x86_64/Kconfig
> +++ linux-rt-rebase.q/arch/x86_64/Kconfig
> @@ -721,6 +721,10 @@ config GENERIC_PENDING_IRQ
> depends on GENERIC_HARDIRQS && SMP
> default y
>
> +config HARDIRQS_SW_RESEND
> + bool
> + default y
> +
> menu "Power management options"
>
> source kernel/power/Kconfig
> Index: linux-rt-rebase.q/kernel/irq/manage.c
> ===================================================================
> --- linux-rt-rebase.q.orig/kernel/irq/manage.c
> +++ linux-rt-rebase.q/kernel/irq/manage.c
> @@ -175,6 +175,14 @@ void enable_irq(unsigned int irq)
> desc->depth--;
> }
> spin_unlock_irqrestore(&desc->lock, flags);
> +#ifdef CONFIG_HARDIRQS_SW_RESEND
> + /*
> + * Do a bh disable/enable pair to trigger any pending
> + * irq resend logic:
> + */
> + local_bh_disable();
> + local_bh_enable();
> +#endif
> }
> EXPORT_SYMBOL(enable_irq);
This patch didn't help (tested on 2.6.22.1) - ne2k_pci timed out.
ps: I retested all patches posted in this thread on top of 2.6.22.1
and behavior from 2.6.21.3 didn't changed. My next tests will be on
2.6.22.x only.
Regards,
Marcin Slusarz
^ permalink raw reply
* Re: [PATCH] USB Pegasus driver - avoid a potential NULL pointer dereference.
From: Petko Manolov @ 2007-07-30 8:05 UTC (permalink / raw)
To: Oliver Neukum
Cc: linux-usb-devel, Petko Manolov, netdev, Jesper Juhl,
Linux Kernel Mailing List, Greg Kroah-Hartman, Satyam Sharma
In-Reply-To: <200707291049.26619.oliver@neukum.org>
[-- Attachment #1: Type: TEXT/PLAIN, Size: 2206 bytes --]
On Sun, 29 Jul 2007, Oliver Neukum wrote:
> Am Sonntag 29 Juli 2007 schrieb Jesper Juhl:
>> On 29/07/07, Satyam Sharma <satyam.sharma@gmail.com> wrote:
>>> Hi,
>>>
>>> On 7/29/07, Jesper Juhl <jesper.juhl@gmail.com> wrote:
>>>> Hello,
>>>>
>>>> This patch makes sure we don't dereference a NULL pointer in
>>>> drivers/net/usb/pegasus.c::write_bulk_callback() in the initial
>>>> struct net_device *net = pegasus->net; assignment.
>>>> The existing code checks if 'pegasus' is NULL and bails out if
>>>> it is, so we better not touch that pointer until after that check.
>>>> [...]
>>>> diff --git a/drivers/net/usb/pegasus.c b/drivers/net/usb/pegasus.c
>>>> index a05fd97..04cba6b 100644
>>>> --- a/drivers/net/usb/pegasus.c
>>>> +++ b/drivers/net/usb/pegasus.c
>>>> @@ -768,11 +768,13 @@ done:
>>>> static void write_bulk_callback(struct urb *urb)
>>>> {
>>>> pegasus_t *pegasus = urb->context;
>>>> - struct net_device *net = pegasus->net;
>>>> + struct net_device *net;
>>>>
>>>> if (!pegasus)
>>>> return;
>>>>
>>>> + net = pegasus->net;
>>>> +
>>>> if (!netif_device_present(net) || !netif_running(net))
>>>> return;
>>>
>>> Is it really possible that we're called into this function with
>>> urb->context == NULL? If not, I'd suggest let's just get rid of
>>> the check itself, instead.
>>>
>> I'm not sure. I am not very familiar with this code. I just figured
>> that moving the assignment is potentially a little safer and it is
>> certainly no worse than the current code, so that's a safe and
>> potentially benneficial change. Removing the check may be safe but I'm
>> not certain enough to make that call...
>
> pegasus == NULL there would be a kernel bug. Silently ignoring
> it, like the code now wants to do is bad. As the oops has never been
> reported, I figure turning it into an explicit debugging test is overkill,
> so removal seems to be the best option.
In the past urb->context was not guaranteed to be non-null for any
asynchronous calls. If this is not the case anymore then it should be
removed from at least two more locations in the driver.
Attached you'll find the resulting patch.
Petko
[-- Attachment #2: Type: TEXT/x-diff, Size: 1403 bytes --]
--- drivers/net/usb/pegasus.c.old 2007-07-10 11:39:50.000000000 +0300
+++ drivers/net/usb/pegasus.c 2007-07-30 11:02:10.000000000 +0300
@@ -100,9 +100,6 @@ static void ctrl_callback(struct urb *ur
{
pegasus_t *pegasus = urb->context;
- if (!pegasus)
- return;
-
switch (urb->status) {
case 0:
if (pegasus->flags & ETH_REGS_CHANGE) {
@@ -609,15 +606,11 @@ static inline struct sk_buff *pull_skb(p
static void read_bulk_callback(struct urb *urb)
{
pegasus_t *pegasus = urb->context;
- struct net_device *net;
+ struct net_device *net = pegasus->net;
int rx_status, count = urb->actual_length;
u8 *buf = urb->transfer_buffer;
__u16 pkt_len;
- if (!pegasus)
- return;
-
- net = pegasus->net;
if (!netif_device_present(net) || !netif_running(net))
return;
@@ -770,9 +763,6 @@ static void write_bulk_callback(struct u
pegasus_t *pegasus = urb->context;
struct net_device *net = pegasus->net;
- if (!pegasus)
- return;
-
if (!netif_device_present(net) || !netif_running(net))
return;
@@ -805,13 +795,9 @@ static void write_bulk_callback(struct u
static void intr_callback(struct urb *urb)
{
pegasus_t *pegasus = urb->context;
- struct net_device *net;
+ struct net_device *net = pegasus->net;
int status;
- if (!pegasus)
- return;
- net = pegasus->net;
-
switch (urb->status) {
case 0:
break;
[-- Attachment #3: Type: text/plain, Size: 315 bytes --]
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
[-- Attachment #4: Type: text/plain, Size: 191 bytes --]
_______________________________________________
linux-usb-devel@lists.sourceforge.net
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
^ permalink raw reply
* Re: source interface ping bug ?
From: nano bug @ 2007-07-30 8:46 UTC (permalink / raw)
To: netdev
In-Reply-To: <d39c36500707290448p599ff710hf79ba1911c818162@mail.gmail.com>
Can someone have a look a this and tell if it's kernel related or if I
posted this in the wrong place ? Thanks.
On 7/29/07, nano bug <linnewbye@gmail.com> wrote:
> Any news about this ?
>
> On 7/27/07, nano bug <linnewbye@gmail.com> wrote:
> > Hello there,
> >
> > I'm facing the following issue : when I try to ping using source
> > interface instead of a source ip address the ping utility starts to
> > send arp requests instead of icmp requests though the ip address I'm
> > pinging it's not in the subnets directly connected to my linux box.
> > I've noticed this situation since I upgraded from kernel 2.6.20 to
> > 2.6.21. On 2.6.20 and lower I haven't had this problem. Now I upgraded
> > to 2.6.22 but it's the same. I'm using latest iproute and iputils.
> > Here is an output of tcpdump when I try to ping an outside ip address,
> > like for example www.yahoo.com, using source interface :
> >
> > root@darkstar:~# uname -a
> > Linux darkstar 2.6.22 #1 Thu Jul 26 21:22:11 EEST 2007 i686 Pentium II
> > (Deschutes) GenuineIntel GNU/Linux
> >
> > root@darkstar:~# ip -V
> > ip utility, iproute2-ss070710
> > root@darkstar:~#
> >
> >
> > root@darkstar:~# ip address show dev eth2
> > 3: eth2: <BROADCAST,MULTICAST,NOTRAILERS,UP,LOWER_UP> mtu 1500 qdisc
> > pfifo_fast qlen 1000
> > link/ether 00:90:27:0f:79:f3 brd ff:ff:ff:ff:ff:ff
> > inet 86.106.19.75/23 brd 86.106.19.255 scope global eth2
> > root@darkstar:~# ip route get 87.248.113.14 from 86.106.19.75 oif eth2
> > 87.248.113.14 from 86.106.19.75 via 86.106.18.1 dev eth2
> > cache mtu 1500 advmss 1460 hoplimit 64
> > root@darkstar:~#
> >
> >
> > root@darkstar:~/iputils# ./ping -V
> > ping utility, iputils-sss20070202
> > root@darkstar:~/iputils# ./ping -I 86.106.19.75 87.248.113.14 -c 2
> > PING 87.248.113.14 (87.248.113.14) from 86.106.19.75 : 56(84) bytes of data.
> > 64 bytes from 87.248.113.14: icmp_seq=1 ttl=51 time=60.5 ms
> > 64 bytes from 87.248.113.14: icmp_seq=2 ttl=51 time=63.2 ms
> >
> > --- 87.248.113.14 ping statistics ---
> > 2 packets transmitted, 2 received, 0% packet loss, time 999ms
> > rtt min/avg/max/mdev = 60.574/61.924/63.274/1.350 ms
> > root@darkstar:~/iputils#
> >
> >
> > root@darkstar:~# tcpdump -i eth2 -vvv -n host 87.248.113.14 and host
> > 86.106.19.75
> > tcpdump: listening on eth2, link-type EN10MB (Ethernet), capture size 96 bytes
> > 01:18:09.572603 IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF],
> > proto: ICMP (1), length: 84) 86.106.19.75 > 87.248.113.14: ICMP echo
> > request, id 27166, seq 1, length 64
> > 01:18:09.632861 IP (tos 0x0, ttl 51, id 6100, offset 0, flags [none],
> > proto: ICMP (1), length: 84) 87.248.113.14 > 86.106.19.75: ICMP echo
> > reply, id 27166, seq 1, length 64
> > 01:18:10.572746 IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF],
> > proto: ICMP (1), length: 84) 86.106.19.75 > 87.248.113.14: ICMP echo
> > request, id 27166, seq 2, length 64
> > 01:18:10.634951 IP (tos 0x0, ttl 51, id 8790, offset 0, flags [none],
> > proto: ICMP (1), length: 84) 87.248.113.14 > 86.106.19.75: ICMP echo
> > reply, id 27166, seq 2, length 64
> >
> >
> >
> > using source interface :
> >
> > root@darkstar:~/iputils# ./ping -I eth2 87.248.113.14
> > PING 87.248.113.14 (87.248.113.14) from 86.106.19.75 eth2: 56(84) bytes of data.
> > From 86.106.19.75 icmp_seq=1 Destination Host Unreachable
> > From 86.106.19.75 icmp_seq=2 Destination Host Unreachable
> > From 86.106.19.75 icmp_seq=3 Destination Host Unreachable
> > From 86.106.19.75 icmp_seq=5 Destination Host Unreachable
> > From 86.106.19.75 icmp_seq=6 Destination Host Unreachable
> > From 86.106.19.75 icmp_seq=7 Destination Host Unreachable
> > From 86.106.19.75 icmp_seq=9 Destination Host Unreachable
> > From 86.106.19.75 icmp_seq=10 Destination Host Unreachable
> > From 86.106.19.75 icmp_seq=11 Destination Host Unreachable
> >
> > --- 87.248.113.14 ping statistics ---
> > 13 packets transmitted, 0 received, +9 errors, 100% packet loss, time 12006ms
> > , pipe 3
> > root@darkstar:~/iputils#
> >
> >
> > root@darkstar:~# tcpdump -i eth2 -vvv -n host 87.248.113.14 and host
> > 86.106.19.75
> > tcpdump: listening on eth2, link-type EN10MB (Ethernet), capture size 96 bytes
> > 01:19:24.292911 arp who-has 87.248.113.14 tell 86.106.19.75
> > 01:19:25.292897 arp who-has 87.248.113.14 tell 86.106.19.75
> > 01:19:26.292901 arp who-has 87.248.113.14 tell 86.106.19.75
> > 01:19:27.302906 arp who-has 87.248.113.14 tell 86.106.19.75
> > 01:19:28.302911 arp who-has 87.248.113.14 tell 86.106.19.75
> > 01:19:29.302912 arp who-has 87.248.113.14 tell 86.106.19.75
> > 01:19:31.302917 arp who-has 87.248.113.14 tell 86.106.19.75
> > 01:19:32.302921 arp who-has 87.248.113.14 tell 86.106.19.75
> > 01:19:33.302923 arp who-has 87.248.113.14 tell 86.106.19.75
> > 01:19:35.302932 arp who-has 87.248.113.14 tell 86.106.19.75
> > 01:19:36.302932 arp who-has 87.248.113.14 tell 86.106.19.75
> > 01:19:37.302939 arp who-has 87.248.113.14 tell 86.106.19.75
> >
> > 12 packets captured
> > 12 packets received by filter
> > 0 packets dropped by kernel
> > root@darkstar:~#
> >
> >
> > There is one exception though, it works when using eth0. I'm
> > administrating multiple linux boxes with 2 or 3 ethernet cards and if
> > I try pinging with eth0 it does send icmp like it should but when
> > specifying eth1 or eth2 is sends arp requests. The distro I'm using is
> > Slackware 12.0. I already reported this to the iputils maintainer and
> > he recommended to report it here. Thanks in advance.
> >
> > Regards.
> >
>
^ permalink raw reply
* Re: 2.6.20->2.6.21 - networking dies after random time
From: Ingo Molnar @ 2007-07-30 8:46 UTC (permalink / raw)
To: Alan Cox
Cc: Jarek Poplawski, Thomas Gleixner, Linus Torvalds, Marcin ??lusarz,
Jean-Baptiste Vignaud, linux-kernel, shemminger, linux-net,
netdev, Andrew Morton
In-Reply-To: <20070725154656.54fd6f13@the-village.bc.nu>
* Alan Cox <alan@lxorguk.ukuu.org.uk> wrote:
> Ok the logic behind the 8390 is very simple:
thanks for the explanation Alan! A few comments and a question:
> Things to know
> - IRQ delivery is asynchronous to the PCI bus
> - Blocking the local CPU IRQ via spin locks was too slow
> - The chip has register windows needing locking work
>
> So the path was once (I say once as people appear to have changed it
> in the mean time and it now looks rather bogus if the changes to use
> disable_irq_nosync_irqsave are disabling the local IRQ)
>
>
> Take the page lock
> Mask the IRQ on chip
> Disable the IRQ (but not mask locally- someone seems to have
> broken this with the lock validator stuff)
> [This must be _nosync as the page lock may otherwise
> deadlock us]
( side-note: you can ignore the lock validator stuff here, the validator
changes are supposed to a NOP on the !lockdep case. Local irqs will
only be disabled if the validator is running. This could cause dropped
serial irqs on very old boxes but i doubt anyone will want to run the
validator on those. )
> Drop the page lock and turn IRQs back on
>
> At this point an existing IRQ may still be running but we can't
> get a new one
>
> Take the lock (so we know the IRQ has terminated) but don't mask
> the IRQs on the processor
> Set irqlock [for debug]
>
> Transmit (slow as ****)
>
> re-enable the IRQ
>
>
> We have to use disable_irq because otherwise you will get delayed
> interrupts on the APIC bus deadlocking the transmit path.
>
> Quite hairy but the chip simply wasn't designed for SMP and you can't
> even ACK an interrupt without risking corrupting other parallel
> activities on the chip.
So the whole locking is to be able to keep irqs enabled for a long time,
without risking entry of the same IRQ handler on this same CPU, correct?
Marcin's test results suggest that if an IRQ is resent right at the
enable_irq() point [be that via the hw irq-resend mechanism or the sw
irq-resend mechanism], the hang happens.
In the previous 2.6.20 logic we'd not normally generate an IRQ at that
point (because we masked the irq and the card itself deasserts the line
so any level-triggered irq is now moot).
Once Thomas hacked off this resend mechanism for level-triggered irqs,
Marcin saw the hangs go away.
So it seems to me that maybe the driver could be surprised via these
spurious interrupts that happen right after the irq_enable(). Does the
patch below make any sense in your opinion?
Ingo
Index: linux/drivers/net/lib8390.c
===================================================================
--- linux.orig/drivers/net/lib8390.c
+++ linux/drivers/net/lib8390.c
@@ -375,6 +375,8 @@ static int ei_start_xmit(struct sk_buff
/* Turn 8390 interrupts back on. */
ei_local->irqlock = 0;
ei_outb_p(ENISR_ALL, e8390_base + EN0_IMR);
+ /* force POST: */
+ ei_inb_p(e8390_base + EN0_IMR);
spin_unlock(&ei_local->page_lock);
enable_irq_lockdep_irqrestore(dev->irq, &flags);
^ permalink raw reply
* Re: 2.6.20->2.6.21 - networking dies after random time
From: Ingo Molnar @ 2007-07-30 8:49 UTC (permalink / raw)
To: Marcin Ślusarz
Cc: Jarek Poplawski, Thomas Gleixner, Linus Torvalds,
Jean-Baptiste Vignaud, linux-kernel, shemminger, linux-net,
netdev, Andrew Morton, Alan Cox
In-Reply-To: <4bacf17f0707300029g5116e70bq4808059dc8b069f1@mail.gmail.com>
* Marcin Ślusarz <marcin.slusarz@gmail.com> wrote:
> > Subject: x86: activate HARDIRQS_SW_RESEND
> > From: Ingo Molnar <mingo@elte.hu>
> >
> > activate the software-triggered IRQ-resend logic.
> This patch didn't help (tested on 2.6.22.1) - ne2k_pci timed out.
ok. This makes it more likely that the driver itself (or the card) gets
confused by the resend.
does the patch below fix those timeouts? It tests the theory whether any
POST latency could expose this problem.
Ingo
Index: linux/drivers/net/lib8390.c
===================================================================
--- linux.orig/drivers/net/lib8390.c
+++ linux/drivers/net/lib8390.c
@@ -375,6 +375,8 @@ static int ei_start_xmit(struct sk_buff
/* Turn 8390 interrupts back on. */
ei_local->irqlock = 0;
ei_outb_p(ENISR_ALL, e8390_base + EN0_IMR);
+ /* force POST: */
+ ei_inb_p(e8390_base + EN0_IMR);
spin_unlock(&ei_local->page_lock);
enable_irq_lockdep_irqrestore(dev->irq, &flags);
^ permalink raw reply
* Re: [PATCH] USB Pegasus driver - avoid a potential NULL pointer dereference.
From: Satyam Sharma @ 2007-07-30 8:54 UTC (permalink / raw)
To: Petko Manolov
Cc: linux-usb-devel, Petko Manolov, Greg Kroah-Hartman, Jesper Juhl,
Oliver Neukum, Linux Kernel Mailing List, netdev
In-Reply-To: <Pine.LNX.4.64.0707301053020.3113@bender.nucleusys.com>
On Mon, 30 Jul 2007, Petko Manolov wrote:
> On Sun, 29 Jul 2007, Oliver Neukum wrote:
>
> > [...]
> > pegasus == NULL there would be a kernel bug. Silently ignoring
> > it, like the code now wants to do is bad. As the oops has never been
> > reported, I figure turning it into an explicit debugging test is overkill,
> > so removal seems to be the best option.
>
> In the past urb->context was not guaranteed to be non-null for any
> asynchronous calls. If this is not the case anymore then it should be removed
> from at least two more locations in the driver.
>
> Attached you'll find the resulting patch.
Given Oliver's earlier comment, it looks okay to me. Thanks.
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
linux-usb-devel@lists.sourceforge.net
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
^ permalink raw reply
* Re: Linksys Gigabit USB2.0 adapter (asix) regression
From: Erik Slagter @ 2007-07-30 9:27 UTC (permalink / raw)
To: David Hollis; +Cc: netdev
In-Reply-To: <1182923692.6644.4.camel@dhollis-lnx.sunera.com>
David Hollis wrote:
> On Mon, 2007-06-25 at 19:05 +0200, Erik Slagter wrote:
>> drivers/net/usb/asix.c: PHYID=0x01410cc2
>
> Ok, it is using a Marvell PHY so that part should be fine. You
> mentioned that it looks like the packets are being transmitted, but are
> garbled in some way. The device does prepend a 'header' to ethernet
> packets as they are transmitted down the USB pipe. The device strips
> this off and puts the packets on the wire. This could be where the
> issue lies. Are you on x86 by chance or something else?
They are either garbled are they are not passed on the wire. The
transmitted packets are shown by tshark, but a tshark run on "the other
end of the line" does not show them.
Platform is indeed x86, to be precise: fedora 7, kernel 2.6.22-rc6, cpu
pentium M, dell laptop inspiron 9300, ICH6.
If you want me to test something please yell, it's no trouble at all to
change a few lines in the driver's source and recompile the module.
Please note I cannot send mail to you: "(conversation with
dhollis.dyndns.org[71.251.104.159] timed out while sending MAIL FROM)"
^ permalink raw reply
* [PATCH 2.6.23-rc1] PPPOL2TP: Add CONFIG_INET Kconfig dependency
From: James Chapman @ 2007-07-30 12:05 UTC (permalink / raw)
To: netdev; +Cc: toralf.foerster
[PPPOL2TP]: Add CONFIG_INET Kconfig dependency.
PPPOL2TP uses UDP so it obviously depends on CONFIG_INET.
Signed-off-by: James Chapman <jchapman@katalix.com>
---
Toralf Foerster reported that make rndconfig failed in 2.6.23-rc1
when selecting CONFIG_PPPOL2TP without CONFIG_INET.
Index: linux-2.6.23-rc1/drivers/net/Kconfig
===================================================================
--- linux-2.6.23-rc1.orig/drivers/net/Kconfig
+++ linux-2.6.23-rc1/drivers/net/Kconfig
@@ -2851,7 +2851,7 @@ config PPPOATM
config PPPOL2TP
tristate "PPP over L2TP (EXPERIMENTAL)"
- depends on EXPERIMENTAL && PPP
+ depends on EXPERIMENTAL && PPP && INET
help
Support for PPP-over-L2TP socket family. L2TP is a protocol
used by ISPs and enterprises to tunnel PPP traffic over UDP
^ permalink raw reply
* Re: [2.6 patch] make nf_ct_ipv6_skip_exthdr() static
From: Patrick McHardy @ 2007-07-30 12:30 UTC (permalink / raw)
To: Adrian Bunk; +Cc: Yasuyuki Kozakai, netfilter-devel, netdev, linux-kernel
In-Reply-To: <20070729145846.GZ16817@stusta.de>
Adrian Bunk wrote:
> nf_ct_ipv6_skip_exthdr() can now become static.
Applied, thanks.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox