* [patch net-next v4 10/11] act_police: improved accuracy at high rates
From: Jiri Pirko @ 2013-02-10 10:52 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <1360493539-14868-1-git-send-email-jiri@resnulli.us>
Current act_police uses rate table computed by the "tc" userspace program,
which has the following issue:
The rate table has 256 entries to map packet lengths to
token (time units). With TSO sized packets, the 256 entry granularity
leads to loss/gain of rate, making the token bucket inaccurate.
Thus, instead of relying on rate table, this patch explicitly computes
the time and accounts for packet transmission times with nanosecond
granularity.
This is a followup to 56b765b79e9a78dc7d3f8850ba5e5567205a3ecd
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
net/sched/act_police.c | 95 +++++++++++++++++++++++++++-----------------------
1 file changed, 51 insertions(+), 44 deletions(-)
diff --git a/net/sched/act_police.c b/net/sched/act_police.c
index 378a649..71bf4f5 100644
--- a/net/sched/act_police.c
+++ b/net/sched/act_police.c
@@ -26,20 +26,20 @@ struct tcf_police {
struct tcf_common common;
int tcfp_result;
u32 tcfp_ewma_rate;
- u32 tcfp_burst;
+ s64 tcfp_burst;
u32 tcfp_mtu;
- u32 tcfp_toks;
- u32 tcfp_ptoks;
+ s64 tcfp_toks;
+ s64 tcfp_ptoks;
+ s64 tcfp_mtu_ptoks;
psched_time_t tcfp_t_c;
- struct qdisc_rate_table *tcfp_R_tab;
- struct qdisc_rate_table *tcfp_P_tab;
+ struct psched_ratecfg rate;
+ bool rate_present;
+ struct psched_ratecfg peak;
+ bool peak_present;
};
#define to_police(pc) \
container_of(pc, struct tcf_police, common)
-#define L2T(p, L) qdisc_l2t((p)->tcfp_R_tab, L)
-#define L2T_P(p, L) qdisc_l2t((p)->tcfp_P_tab, L)
-
#define POL_TAB_MASK 15
static struct tcf_common *tcf_police_ht[POL_TAB_MASK + 1];
static u32 police_idx_gen;
@@ -123,10 +123,6 @@ static void tcf_police_destroy(struct tcf_police *p)
write_unlock_bh(&police_lock);
gen_kill_estimator(&p->tcf_bstats,
&p->tcf_rate_est);
- if (p->tcfp_R_tab)
- qdisc_put_rtab(p->tcfp_R_tab);
- if (p->tcfp_P_tab)
- qdisc_put_rtab(p->tcfp_P_tab);
/*
* gen_estimator est_timer() might access p->tcf_lock
* or bstats, wait a RCU grace period before freeing p
@@ -227,26 +223,36 @@ override:
}
/* No failure allowed after this point */
- if (R_tab != NULL) {
- qdisc_put_rtab(police->tcfp_R_tab);
- police->tcfp_R_tab = R_tab;
+ police->tcfp_mtu = parm->mtu;
+ if (police->tcfp_mtu == 0) {
+ police->tcfp_mtu = ~0;
+ if (R_tab)
+ police->tcfp_mtu = 255 << R_tab->rate.cell_log;
+ }
+ if (R_tab) {
+ police->rate_present = true;
+ psched_ratecfg_precompute(&police->rate, R_tab->rate.rate);
+ qdisc_put_rtab(R_tab);
+ } else {
+ police->rate_present = false;
}
- if (P_tab != NULL) {
- qdisc_put_rtab(police->tcfp_P_tab);
- police->tcfp_P_tab = P_tab;
+ if (P_tab) {
+ police->peak_present = true;
+ psched_ratecfg_precompute(&police->peak, P_tab->rate.rate);
+ qdisc_put_rtab(P_tab);
+ } else {
+ police->peak_present = false;
}
if (tb[TCA_POLICE_RESULT])
police->tcfp_result = nla_get_u32(tb[TCA_POLICE_RESULT]);
- police->tcfp_toks = police->tcfp_burst = parm->burst;
- police->tcfp_mtu = parm->mtu;
- if (police->tcfp_mtu == 0) {
- police->tcfp_mtu = ~0;
- if (police->tcfp_R_tab)
- police->tcfp_mtu = 255<<police->tcfp_R_tab->rate.cell_log;
+ police->tcfp_burst = PSCHED_TICKS2NS(parm->burst);
+ police->tcfp_toks = police->tcfp_burst;
+ if (police->peak_present) {
+ police->tcfp_mtu_ptoks = (s64) psched_l2t_ns(&police->peak,
+ police->tcfp_mtu);
+ police->tcfp_ptoks = police->tcfp_mtu_ptoks;
}
- if (police->tcfp_P_tab)
- police->tcfp_ptoks = L2T_P(police, police->tcfp_mtu);
police->tcf_action = parm->action;
if (tb[TCA_POLICE_AVRATE])
@@ -256,7 +262,7 @@ override:
if (ret != ACT_P_CREATED)
return ret;
- police->tcfp_t_c = psched_get_time();
+ police->tcfp_t_c = ktime_to_ns(ktime_get());
police->tcf_index = parm->index ? parm->index :
tcf_hash_new_index(&police_idx_gen, &police_hash_info);
h = tcf_hash(police->tcf_index, POL_TAB_MASK);
@@ -303,8 +309,8 @@ static int tcf_act_police(struct sk_buff *skb, const struct tc_action *a,
{
struct tcf_police *police = a->priv;
psched_time_t now;
- long toks;
- long ptoks = 0;
+ s64 toks;
+ s64 ptoks = 0;
spin_lock(&police->tcf_lock);
@@ -320,24 +326,25 @@ static int tcf_act_police(struct sk_buff *skb, const struct tc_action *a,
}
if (qdisc_pkt_len(skb) <= police->tcfp_mtu) {
- if (police->tcfp_R_tab == NULL) {
+ if (!police->rate_present) {
spin_unlock(&police->tcf_lock);
return police->tcfp_result;
}
- now = psched_get_time();
- toks = psched_tdiff_bounded(now, police->tcfp_t_c,
- police->tcfp_burst);
- if (police->tcfp_P_tab) {
+ now = ktime_to_ns(ktime_get());
+ toks = min_t(s64, now - police->tcfp_t_c,
+ police->tcfp_burst);
+ if (police->peak_present) {
ptoks = toks + police->tcfp_ptoks;
- if (ptoks > (long)L2T_P(police, police->tcfp_mtu))
- ptoks = (long)L2T_P(police, police->tcfp_mtu);
- ptoks -= L2T_P(police, qdisc_pkt_len(skb));
+ if (ptoks > police->tcfp_mtu_ptoks)
+ ptoks = police->tcfp_mtu_ptoks;
+ ptoks -= (s64) psched_l2t_ns(&police->peak,
+ qdisc_pkt_len(skb));
}
toks += police->tcfp_toks;
- if (toks > (long)police->tcfp_burst)
+ if (toks > police->tcfp_burst)
toks = police->tcfp_burst;
- toks -= L2T(police, qdisc_pkt_len(skb));
+ toks -= (s64) psched_l2t_ns(&police->rate, qdisc_pkt_len(skb));
if ((toks|ptoks) >= 0) {
police->tcfp_t_c = now;
police->tcfp_toks = toks;
@@ -363,15 +370,15 @@ tcf_act_police_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
.index = police->tcf_index,
.action = police->tcf_action,
.mtu = police->tcfp_mtu,
- .burst = police->tcfp_burst,
+ .burst = PSCHED_NS2TICKS(police->tcfp_burst),
.refcnt = police->tcf_refcnt - ref,
.bindcnt = police->tcf_bindcnt - bind,
};
- if (police->tcfp_R_tab)
- opt.rate = police->tcfp_R_tab->rate;
- if (police->tcfp_P_tab)
- opt.peakrate = police->tcfp_P_tab->rate;
+ if (police->rate_present)
+ opt.rate.rate = psched_ratecfg_getrate(&police->rate);
+ if (police->peak_present)
+ opt.peakrate.rate = psched_ratecfg_getrate(&police->peak);
if (nla_put(skb, TCA_POLICE_TBF, sizeof(opt), &opt))
goto nla_put_failure;
if (police->tcfp_result &&
--
1.8.1.2
^ permalink raw reply related
* [patch net-next v4 09/11] act_police: move struct tcf_police to act_police.c
From: Jiri Pirko @ 2013-02-10 10:52 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <1360493539-14868-1-git-send-email-jiri@resnulli.us>
It's not used anywhere else, so move it.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
include/net/act_api.h | 15 ---------------
net/sched/act_police.c | 15 +++++++++++++++
2 files changed, 15 insertions(+), 15 deletions(-)
diff --git a/include/net/act_api.h b/include/net/act_api.h
index 112c25c..06ef7e9 100644
--- a/include/net/act_api.h
+++ b/include/net/act_api.h
@@ -35,21 +35,6 @@ struct tcf_common {
#define tcf_lock common.tcfc_lock
#define tcf_rcu common.tcfc_rcu
-struct tcf_police {
- struct tcf_common common;
- int tcfp_result;
- u32 tcfp_ewma_rate;
- u32 tcfp_burst;
- u32 tcfp_mtu;
- u32 tcfp_toks;
- u32 tcfp_ptoks;
- psched_time_t tcfp_t_c;
- struct qdisc_rate_table *tcfp_R_tab;
- struct qdisc_rate_table *tcfp_P_tab;
-};
-#define to_police(pc) \
- container_of(pc, struct tcf_police, common)
-
struct tcf_hashinfo {
struct tcf_common **htab;
unsigned int hmask;
diff --git a/net/sched/act_police.c b/net/sched/act_police.c
index 8dbd695..378a649 100644
--- a/net/sched/act_police.c
+++ b/net/sched/act_police.c
@@ -22,6 +22,21 @@
#include <net/act_api.h>
#include <net/netlink.h>
+struct tcf_police {
+ struct tcf_common common;
+ int tcfp_result;
+ u32 tcfp_ewma_rate;
+ u32 tcfp_burst;
+ u32 tcfp_mtu;
+ u32 tcfp_toks;
+ u32 tcfp_ptoks;
+ psched_time_t tcfp_t_c;
+ struct qdisc_rate_table *tcfp_R_tab;
+ struct qdisc_rate_table *tcfp_P_tab;
+};
+#define to_police(pc) \
+ container_of(pc, struct tcf_police, common)
+
#define L2T(p, L) qdisc_l2t((p)->tcfp_R_tab, L)
#define L2T_P(p, L) qdisc_l2t((p)->tcfp_P_tab, L)
--
1.8.1.2
^ permalink raw reply related
* [patch net-next v4 08/11] tbf: fix value set for q->ptokens
From: Jiri Pirko @ 2013-02-10 10:52 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <1360493539-14868-1-git-send-email-jiri@resnulli.us>
q->ptokens is in ns and we are assigning q->mtu directly to it. That is
wrong. psched_l2t_ns() should be used to compute correct value.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
net/sched/sch_tbf.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index dc562a8..0ac6224 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -111,6 +111,7 @@ struct tbf_sched_data {
/* Variables */
s64 tokens; /* Current number of B tokens */
s64 ptokens; /* Current number of P tokens */
+ s64 mtu_ptokens;
psched_time_t t_c; /* Time check-point */
struct Qdisc *qdisc; /* Inner qdisc, default - bfifo queue */
struct qdisc_watchdog watchdog; /* Watchdog timer */
@@ -165,8 +166,8 @@ static struct sk_buff *tbf_dequeue(struct Qdisc *sch)
if (q->peak_present) {
ptoks = toks + q->ptokens;
- if (ptoks > (long)q->mtu)
- ptoks = q->mtu;
+ if (ptoks > q->mtu_ptokens)
+ ptoks = q->mtu_ptokens;
ptoks -= (s64) psched_l2t_ns(&q->peak, len);
}
toks += q->tokens;
@@ -215,7 +216,8 @@ static void tbf_reset(struct Qdisc *sch)
sch->q.qlen = 0;
q->t_c = ktime_to_ns(ktime_get());
q->tokens = q->buffer;
- q->ptokens = q->mtu;
+ if (q->peak_present)
+ q->ptokens = q->mtu_ptokens;
qdisc_watchdog_cancel(&q->watchdog);
}
@@ -296,11 +298,12 @@ static int tbf_change(struct Qdisc *sch, struct nlattr *opt)
q->max_size = max_size;
q->buffer = PSCHED_TICKS2NS(qopt->buffer);
q->tokens = q->buffer;
- q->ptokens = q->mtu;
psched_ratecfg_precompute(&q->rate, rtab->rate.rate);
if (ptab) {
psched_ratecfg_precompute(&q->peak, ptab->rate.rate);
+ q->mtu_ptokens = (s64) psched_l2t_ns(&q->peak, q->mtu);
+ q->ptokens = q->mtu_ptokens;
q->peak_present = true;
} else {
q->peak_present = false;
--
1.8.1.2
^ permalink raw reply related
* [patch net-next v4 07/11] tbf: ignore max_size check for gso skbs
From: Jiri Pirko @ 2013-02-10 10:52 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <1360493539-14868-1-git-send-email-jiri@resnulli.us>
This check made bigger packets incorrectly dropped. Remove this
limitation for gso skbs.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Acked-by: Eric Dumazet <edumazet@google.com>
---
net/sched/sch_tbf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index e05710a..dc562a8 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -121,7 +121,7 @@ static int tbf_enqueue(struct sk_buff *skb, struct Qdisc *sch)
struct tbf_sched_data *q = qdisc_priv(sch);
int ret;
- if (qdisc_pkt_len(skb) > q->max_size)
+ if (qdisc_pkt_len(skb) > q->max_size && !skb_is_gso(skb))
return qdisc_reshape_fail(skb, sch);
ret = qdisc_enqueue(skb, q->qdisc);
--
1.8.1.2
^ permalink raw reply related
* [patch net-next v4 06/11] tbf: improved accuracy at high rates
From: Jiri Pirko @ 2013-02-10 10:52 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <1360493539-14868-1-git-send-email-jiri@resnulli.us>
Current TBF uses rate table computed by the "tc" userspace program,
which has the following issue:
The rate table has 256 entries to map packet lengths to
token (time units). With TSO sized packets, the 256 entry granularity
leads to loss/gain of rate, making the token bucket inaccurate.
Thus, instead of relying on rate table, this patch explicitly computes
the time and accounts for packet transmission times with nanosecond
granularity.
This is a followup to 56b765b79e9a78dc7d3f8850ba5e5567205a3ecd
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Acked-by: Eric Dumazet <edumazet@google.com>
---
net/sched/sch_tbf.c | 60 ++++++++++++++++++++++++++---------------------------
1 file changed, 29 insertions(+), 31 deletions(-)
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 4b056c15..e05710a 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -19,6 +19,7 @@
#include <linux/errno.h>
#include <linux/skbuff.h>
#include <net/netlink.h>
+#include <net/sch_generic.h>
#include <net/pkt_sched.h>
@@ -100,23 +101,21 @@
struct tbf_sched_data {
/* Parameters */
u32 limit; /* Maximal length of backlog: bytes */
- u32 buffer; /* Token bucket depth/rate: MUST BE >= MTU/B */
+ s64 buffer; /* Token bucket depth/rate: MUST BE >= MTU/B */
u32 mtu;
u32 max_size;
- struct qdisc_rate_table *R_tab;
- struct qdisc_rate_table *P_tab;
+ struct psched_ratecfg rate;
+ struct psched_ratecfg peak;
+ bool peak_present;
/* Variables */
- long tokens; /* Current number of B tokens */
- long ptokens; /* Current number of P tokens */
+ s64 tokens; /* Current number of B tokens */
+ s64 ptokens; /* Current number of P tokens */
psched_time_t t_c; /* Time check-point */
struct Qdisc *qdisc; /* Inner qdisc, default - bfifo queue */
struct qdisc_watchdog watchdog; /* Watchdog timer */
};
-#define L2T(q, L) qdisc_l2t((q)->R_tab, L)
-#define L2T_P(q, L) qdisc_l2t((q)->P_tab, L)
-
static int tbf_enqueue(struct sk_buff *skb, struct Qdisc *sch)
{
struct tbf_sched_data *q = qdisc_priv(sch);
@@ -157,23 +156,23 @@ static struct sk_buff *tbf_dequeue(struct Qdisc *sch)
if (skb) {
psched_time_t now;
- long toks;
- long ptoks = 0;
+ s64 toks;
+ s64 ptoks = 0;
unsigned int len = qdisc_pkt_len(skb);
- now = psched_get_time();
- toks = psched_tdiff_bounded(now, q->t_c, q->buffer);
+ now = ktime_to_ns(ktime_get());
+ toks = min_t(s64, now - q->t_c, q->buffer);
- if (q->P_tab) {
+ if (q->peak_present) {
ptoks = toks + q->ptokens;
if (ptoks > (long)q->mtu)
ptoks = q->mtu;
- ptoks -= L2T_P(q, len);
+ ptoks -= (s64) psched_l2t_ns(&q->peak, len);
}
toks += q->tokens;
- if (toks > (long)q->buffer)
+ if (toks > q->buffer)
toks = q->buffer;
- toks -= L2T(q, len);
+ toks -= (s64) psched_l2t_ns(&q->rate, len);
if ((toks|ptoks) >= 0) {
skb = qdisc_dequeue_peeked(q->qdisc);
@@ -214,7 +213,7 @@ static void tbf_reset(struct Qdisc *sch)
qdisc_reset(q->qdisc);
sch->q.qlen = 0;
- q->t_c = psched_get_time();
+ q->t_c = ktime_to_ns(ktime_get());
q->tokens = q->buffer;
q->ptokens = q->mtu;
qdisc_watchdog_cancel(&q->watchdog);
@@ -295,12 +294,17 @@ static int tbf_change(struct Qdisc *sch, struct nlattr *opt)
q->limit = qopt->limit;
q->mtu = qopt->mtu;
q->max_size = max_size;
- q->buffer = qopt->buffer;
+ q->buffer = PSCHED_TICKS2NS(qopt->buffer);
q->tokens = q->buffer;
q->ptokens = q->mtu;
- swap(q->R_tab, rtab);
- swap(q->P_tab, ptab);
+ psched_ratecfg_precompute(&q->rate, rtab->rate.rate);
+ if (ptab) {
+ psched_ratecfg_precompute(&q->peak, ptab->rate.rate);
+ q->peak_present = true;
+ } else {
+ q->peak_present = false;
+ }
sch_tree_unlock(sch);
err = 0;
@@ -319,7 +323,7 @@ static int tbf_init(struct Qdisc *sch, struct nlattr *opt)
if (opt == NULL)
return -EINVAL;
- q->t_c = psched_get_time();
+ q->t_c = ktime_to_ns(ktime_get());
qdisc_watchdog_init(&q->watchdog, sch);
q->qdisc = &noop_qdisc;
@@ -331,12 +335,6 @@ static void tbf_destroy(struct Qdisc *sch)
struct tbf_sched_data *q = qdisc_priv(sch);
qdisc_watchdog_cancel(&q->watchdog);
-
- if (q->P_tab)
- qdisc_put_rtab(q->P_tab);
- if (q->R_tab)
- qdisc_put_rtab(q->R_tab);
-
qdisc_destroy(q->qdisc);
}
@@ -352,13 +350,13 @@ static int tbf_dump(struct Qdisc *sch, struct sk_buff *skb)
goto nla_put_failure;
opt.limit = q->limit;
- opt.rate = q->R_tab->rate;
- if (q->P_tab)
- opt.peakrate = q->P_tab->rate;
+ opt.rate.rate = psched_ratecfg_getrate(&q->rate);
+ if (q->peak_present)
+ opt.peakrate.rate = psched_ratecfg_getrate(&q->peak);
else
memset(&opt.peakrate, 0, sizeof(opt.peakrate));
opt.mtu = q->mtu;
- opt.buffer = q->buffer;
+ opt.buffer = PSCHED_NS2TICKS(q->buffer);
if (nla_put(skb, TCA_TBF_PARMS, sizeof(opt), &opt))
goto nla_put_failure;
--
1.8.1.2
^ permalink raw reply related
* [patch net-next v4 05/11] sch: make htb_rate_cfg and functions around that generic
From: Jiri Pirko @ 2013-02-10 10:52 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <1360493539-14868-1-git-send-email-jiri@resnulli.us>
As it is going to be used in tbf as well, push these to generic code.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Acked-by: Eric Dumazet <edumazet@google.com>
---
include/net/sch_generic.h | 19 ++++++++++++++
net/sched/sch_generic.c | 37 +++++++++++++++++++++++++++
net/sched/sch_htb.c | 65 +++++++----------------------------------------
3 files changed, 65 insertions(+), 56 deletions(-)
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 2d06c2a..2761c90 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -679,4 +679,23 @@ static inline struct sk_buff *skb_act_clone(struct sk_buff *skb, gfp_t gfp_mask,
}
#endif
+struct psched_ratecfg {
+ u64 rate_bps;
+ u32 mult;
+ u32 shift;
+};
+
+static inline u64 psched_l2t_ns(const struct psched_ratecfg *r,
+ unsigned int len)
+{
+ return ((u64)len * r->mult) >> r->shift;
+}
+
+extern void psched_ratecfg_precompute(struct psched_ratecfg *r, u32 rate);
+
+static inline u32 psched_ratecfg_getrate(const struct psched_ratecfg *r)
+{
+ return r->rate_bps >> 3;
+}
+
#endif
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 5d81a44..ffad481 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -25,6 +25,7 @@
#include <linux/rcupdate.h>
#include <linux/list.h>
#include <linux/slab.h>
+#include <net/sch_generic.h>
#include <net/pkt_sched.h>
#include <net/dst.h>
@@ -896,3 +897,39 @@ void dev_shutdown(struct net_device *dev)
WARN_ON(timer_pending(&dev->watchdog_timer));
}
+
+void psched_ratecfg_precompute(struct psched_ratecfg *r, u32 rate)
+{
+ u64 factor;
+ u64 mult;
+ int shift;
+
+ r->rate_bps = rate << 3;
+ r->shift = 0;
+ r->mult = 1;
+ /*
+ * Calibrate mult, shift so that token counting is accurate
+ * for smallest packet size (64 bytes). Token (time in ns) is
+ * computed as (bytes * 8) * NSEC_PER_SEC / rate_bps. It will
+ * work as long as the smallest packet transfer time can be
+ * accurately represented in nanosec.
+ */
+ if (r->rate_bps > 0) {
+ /*
+ * Higher shift gives better accuracy. Find the largest
+ * shift such that mult fits in 32 bits.
+ */
+ for (shift = 0; shift < 16; shift++) {
+ r->shift = shift;
+ factor = 8LLU * NSEC_PER_SEC * (1 << r->shift);
+ mult = div64_u64(factor, r->rate_bps);
+ if (mult > UINT_MAX)
+ break;
+ }
+
+ r->shift = shift - 1;
+ factor = 8LLU * NSEC_PER_SEC * (1 << r->shift);
+ r->mult = div64_u64(factor, r->rate_bps);
+ }
+}
+EXPORT_SYMBOL(psched_ratecfg_precompute);
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 2b22544..03c2692 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -38,6 +38,7 @@
#include <linux/workqueue.h>
#include <linux/slab.h>
#include <net/netlink.h>
+#include <net/sch_generic.h>
#include <net/pkt_sched.h>
/* HTB algorithm.
@@ -71,12 +72,6 @@ enum htb_cmode {
HTB_CAN_SEND /* class can send */
};
-struct htb_rate_cfg {
- u64 rate_bps;
- u32 mult;
- u32 shift;
-};
-
/* interior & leaf nodes; props specific to leaves are marked L: */
struct htb_class {
struct Qdisc_class_common common;
@@ -124,8 +119,8 @@ struct htb_class {
int filter_cnt;
/* token bucket parameters */
- struct htb_rate_cfg rate;
- struct htb_rate_cfg ceil;
+ struct psched_ratecfg rate;
+ struct psched_ratecfg ceil;
s64 buffer, cbuffer; /* token bucket depth/rate */
psched_tdiff_t mbuffer; /* max wait time */
s64 tokens, ctokens; /* current number of tokens */
@@ -168,45 +163,6 @@ struct htb_sched {
struct work_struct work;
};
-static u64 l2t_ns(struct htb_rate_cfg *r, unsigned int len)
-{
- return ((u64)len * r->mult) >> r->shift;
-}
-
-static void htb_precompute_ratedata(struct htb_rate_cfg *r)
-{
- u64 factor;
- u64 mult;
- int shift;
-
- r->shift = 0;
- r->mult = 1;
- /*
- * Calibrate mult, shift so that token counting is accurate
- * for smallest packet size (64 bytes). Token (time in ns) is
- * computed as (bytes * 8) * NSEC_PER_SEC / rate_bps. It will
- * work as long as the smallest packet transfer time can be
- * accurately represented in nanosec.
- */
- if (r->rate_bps > 0) {
- /*
- * Higher shift gives better accuracy. Find the largest
- * shift such that mult fits in 32 bits.
- */
- for (shift = 0; shift < 16; shift++) {
- r->shift = shift;
- factor = 8LLU * NSEC_PER_SEC * (1 << r->shift);
- mult = div64_u64(factor, r->rate_bps);
- if (mult > UINT_MAX)
- break;
- }
-
- r->shift = shift - 1;
- factor = 8LLU * NSEC_PER_SEC * (1 << r->shift);
- r->mult = div64_u64(factor, r->rate_bps);
- }
-}
-
/* find class in global hash table using given handle */
static inline struct htb_class *htb_find(u32 handle, struct Qdisc *sch)
{
@@ -632,7 +588,7 @@ static inline void htb_accnt_tokens(struct htb_class *cl, int bytes, s64 diff)
if (toks > cl->buffer)
toks = cl->buffer;
- toks -= (s64) l2t_ns(&cl->rate, bytes);
+ toks -= (s64) psched_l2t_ns(&cl->rate, bytes);
if (toks <= -cl->mbuffer)
toks = 1 - cl->mbuffer;
@@ -645,7 +601,7 @@ static inline void htb_accnt_ctokens(struct htb_class *cl, int bytes, s64 diff)
if (toks > cl->cbuffer)
toks = cl->cbuffer;
- toks -= (s64) l2t_ns(&cl->ceil, bytes);
+ toks -= (s64) psched_l2t_ns(&cl->ceil, bytes);
if (toks <= -cl->mbuffer)
toks = 1 - cl->mbuffer;
@@ -1134,9 +1090,9 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
memset(&opt, 0, sizeof(opt));
- opt.rate.rate = cl->rate.rate_bps >> 3;
+ opt.rate.rate = psched_ratecfg_getrate(&cl->rate);
opt.buffer = PSCHED_NS2TICKS(cl->buffer);
- opt.ceil.rate = cl->ceil.rate_bps >> 3;
+ opt.ceil.rate = psched_ratecfg_getrate(&cl->ceil);
opt.cbuffer = PSCHED_NS2TICKS(cl->cbuffer);
opt.quantum = cl->quantum;
opt.prio = cl->prio;
@@ -1503,11 +1459,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
cl->prio = TC_HTB_NUMPRIO - 1;
}
- cl->rate.rate_bps = (u64)hopt->rate.rate << 3;
- cl->ceil.rate_bps = (u64)hopt->ceil.rate << 3;
-
- htb_precompute_ratedata(&cl->rate);
- htb_precompute_ratedata(&cl->ceil);
+ psched_ratecfg_precompute(&cl->rate, hopt->rate.rate);
+ psched_ratecfg_precompute(&cl->ceil, hopt->ceil.rate);
cl->buffer = PSCHED_TICKS2NS(hopt->buffer);
cl->cbuffer = PSCHED_TICKS2NS(hopt->buffer);
--
1.8.1.2
^ permalink raw reply related
* [patch net-next v4 04/11] htb: initialize cl->tokens and cl->ctokens correctly
From: Jiri Pirko @ 2013-02-10 10:52 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <1360493539-14868-1-git-send-email-jiri@resnulli.us>
These are in ns so convert from ticks to ns.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Acked-by: Eric Dumazet <edumazet@google.com>
---
net/sched/sch_htb.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 547912e9..2b22544 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1459,8 +1459,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
cl->parent = parent;
/* set class to be in HTB_CAN_SEND state */
- cl->tokens = hopt->buffer;
- cl->ctokens = hopt->cbuffer;
+ cl->tokens = PSCHED_TICKS2NS(hopt->buffer);
+ cl->ctokens = PSCHED_TICKS2NS(hopt->cbuffer);
cl->mbuffer = 60 * PSCHED_TICKS_PER_SEC; /* 1min */
cl->t_c = psched_get_time();
cl->cmode = HTB_CAN_SEND;
--
1.8.1.2
^ permalink raw reply related
* [patch net-next v4 03/11] htb: remove pointless first initialization of buffer and cbuffer
From: Jiri Pirko @ 2013-02-10 10:52 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <1360493539-14868-1-git-send-email-jiri@resnulli.us>
these are initialized correctly couple of lines later in the function.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Acked-by: Eric Dumazet <edumazet@google.com>
---
net/sched/sch_htb.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 14a83dc..547912e9 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1503,9 +1503,6 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
cl->prio = TC_HTB_NUMPRIO - 1;
}
- cl->buffer = hopt->buffer;
- cl->cbuffer = hopt->cbuffer;
-
cl->rate.rate_bps = (u64)hopt->rate.rate << 3;
cl->ceil.rate_bps = (u64)hopt->ceil.rate << 3;
--
1.8.1.2
^ permalink raw reply related
* [patch net-next v4 02/11] htb: fix values in opt dump
From: Jiri Pirko @ 2013-02-10 10:52 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <1360493539-14868-1-git-send-email-jiri@resnulli.us>
in htb_change_class() cl->buffer and cl->buffer are stored in ns.
So in dump, convert them back to psched ticks.
Note this was introduced by:
commit 56b765b79e9a78dc7d3f8850ba5e5567205a3ecd
htb: improved accuracy at high rates
Please consider this for -net/-stable.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Acked-by: Eric Dumazet <edumazet@google.com>
---
net/sched/sch_htb.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 476992c..14a83dc 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1135,9 +1135,9 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
memset(&opt, 0, sizeof(opt));
opt.rate.rate = cl->rate.rate_bps >> 3;
- opt.buffer = cl->buffer;
+ opt.buffer = PSCHED_NS2TICKS(cl->buffer);
opt.ceil.rate = cl->ceil.rate_bps >> 3;
- opt.cbuffer = cl->cbuffer;
+ opt.cbuffer = PSCHED_NS2TICKS(cl->cbuffer);
opt.quantum = cl->quantum;
opt.prio = cl->prio;
opt.level = cl->level;
--
1.8.1.2
^ permalink raw reply related
* [patch net-next v4 01/11] htb: use PSCHED_TICKS2NS()
From: Jiri Pirko @ 2013-02-10 10:52 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <1360493539-14868-1-git-send-email-jiri@resnulli.us>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Acked-by: Eric Dumazet <edumazet@google.com>
---
net/sched/sch_htb.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 51561ea..476992c 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1512,8 +1512,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
htb_precompute_ratedata(&cl->rate);
htb_precompute_ratedata(&cl->ceil);
- cl->buffer = hopt->buffer << PSCHED_SHIFT;
- cl->cbuffer = hopt->buffer << PSCHED_SHIFT;
+ cl->buffer = PSCHED_TICKS2NS(hopt->buffer);
+ cl->cbuffer = PSCHED_TICKS2NS(hopt->buffer);
sch_tree_unlock(sch);
--
1.8.1.2
^ permalink raw reply related
* [patch net-next v4 00/11] couple of net/sched fixes+improvements
From: Jiri Pirko @ 2013-02-10 10:52 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, jhs, kuznet, j.vimal
v3->v4:
- cache mtu_ptokens value instead of compute in by psched_l2t_ns in fast path
(patch "act_police: improved accuracy at high rates" and
patch "tbf: fix value set for q->ptokens")
v2->v3:
- fixed schedule while atomic issue
(patch "act_police: improved accuracy at high rates")
v1->v2:
- made struct psched_ratecfg const in params of couple of inline functions
(patch "sch: make htb_rate_cfg and functions around that generic")
- fixes misspelled "peak"
(patch "tbf: improved accuracy at high rates")
- added last 4 patches to this set
Jiri Pirko (11):
htb: use PSCHED_TICKS2NS()
htb: fix values in opt dump
htb: remove pointless first initialization of buffer and cbuffer
htb: initialize cl->tokens and cl->ctokens correctly
sch: make htb_rate_cfg and functions around that generic
tbf: improved accuracy at high rates
tbf: ignore max_size check for gso skbs
tbf: fix value set for q->ptokens
act_police: move struct tcf_police to act_police.c
act_police: improved accuracy at high rates
act_police: remove <=mtu check for gso skbs
include/net/act_api.h | 15 -------
include/net/sch_generic.h | 19 +++++++++
net/sched/act_police.c | 100 ++++++++++++++++++++++++++++------------------
net/sched/sch_generic.c | 37 +++++++++++++++++
net/sched/sch_htb.c | 80 +++++++------------------------------
net/sched/sch_tbf.c | 73 ++++++++++++++++-----------------
6 files changed, 169 insertions(+), 155 deletions(-)
--
1.8.1.2
^ permalink raw reply
* Re: [net-next (TAKE 2) 0/4] IPv6 over Firewire
From: YOSHIFUJI Hideaki @ 2013-02-10 10:48 UTC (permalink / raw)
To: stephan.gatzka; +Cc: netdev, linux1394-devel, davem, stefanr, YOSHIFUJI Hideaki
In-Reply-To: <51177689.4070904@gmail.com>
Stephan Gatzka wrote:
> Were you able to test your changes against different IP over firewire
> implementations? If not, I will test against the existing Linux
> implementation, MacOSX and Windows.
I have tested against MacOS X 10.5.8.
ARP and ping works but unfortunately NDP does not seem to work.
If I assign static NDP entry on Linux, ping6 ff02::1%fw0 from MacOS works
well.
I checked tcpdump on both side, and I guess this is because MacOS is broken;
- tcpdump on Linux side complains that ICMP checksum is incorrect.
- tcpdump on MacOS side says that length of Link-layer address option is 2 (16)
instead of 3 (24).
I think we should check others like FreeBSD as well.
Thanks.
--yoshfuji
^ permalink raw reply
* Re: [net-next (TAKE 2) 0/4] IPv6 over Firewire
From: Stephan Gatzka @ 2013-02-10 10:29 UTC (permalink / raw)
To: YOSHIFUJI Hideaki; +Cc: netdev, linux1394-devel, davem, stefanr
In-Reply-To: <5117620A.4050005@linux-ipv6.org>
Hi Yoshifuji!
thanks for your submission. The first glance looks good to me but I need
to look over your patch set more thoroughly.
Were you able to test your changes against different IP over firewire
implementations? If not, I will test against the existing Linux
implementation, MacOSX and Windows.
Regards,
Stephan
^ permalink raw reply
* Re: [IPv6] interface-local multicast escapes the local node
From: Hannes Frederic Sowa @ 2013-02-10 9:59 UTC (permalink / raw)
To: YOSHIFUJI Hideaki; +Cc: Erik Hugne, netdev, YOSHIFUJI Hideaki
In-Reply-To: <5116595E.2080601@linux-ipv6.org>
On Sat, Feb 09, 2013 at 11:12:46PM +0900, YOSHIFUJI Hideaki wrote:
> It seems applications will join ff01::/16%eth0 instead of ff01::/16%lo.
> If so, your original patch seems better. My bad, sorry.
>
> Would you update original one, with minor modification that defers
> kfree_skb() after incrementing MIB, please?
I would add another constraint to the if "&& !(dev->flags & IFF_LOOPBACK)", so
it becomes:
if (IPV6_ADDR_MC_SCOPE(&ipv6_hdr(skb)->daddr) <=
IPV6_ADDR_SCOPE_NODELOCAL &&
!(dev->flags & IFF_LOOPBACK))
kfree_skb(skb);
return 0;
}
Otherwise ff01::/16%lo would not work because the multicast mirroring through
dev_loopback_xmit won't be taken and the packet would be dropped after that.
Can you confirm? Thanks.
^ permalink raw reply
* Re: [GIT] Networking
From: Jörg Otte @ 2013-02-10 9:17 UTC (permalink / raw)
To: Francois Romieu
Cc: Linus Torvalds, David Miller, Andrew Morton, Network Development,
Linux Kernel Mailing List, Hayes Wang
In-Reply-To: <20130208225026.GA24672@electric-eye.fr.zoreil.com>
2013/2/8 Francois Romieu <romieu@fr.zoreil.com>:
> Linus Torvalds <torvalds@linux-foundation.org> :
> [...]
>> However, there's still the r8169 regressions (see the emails with the
>> subject "regression: NETDEV WATCHDOG: eth0 (r8169): transmit queue 0
>> timed out"). It's bisected, and a revert is reported to fix things.
>> It's not in this pull request. Comments ?
>
> Jörg asked Hayes. Hayes suggested to revert. Francois reverts, fixes a
> pair of rejects, builds, boot. Anyone should be able to pull from
> branch 'davem.r8169' in repository
>
> git://violet.fr.zoreil.com/romieu/linux davem.r8169
>
> to get the changes below :
>
> Distance from 'davem' (a1c83b054ebe1264ed9ae9d5c286f9eae68e60ea)
> ----------------------------------------------------------------
>
> 4521e1a94279ce610d3f9b7945c17d581f804242
> eef63cc1c6ecf4898a973f870aec95d6e923ea77
>
> Diffstat
> --------
>
> drivers/net/ethernet/realtek/r8169.c | 86 ++++++++----------------------------
> 1 file changed, 19 insertions(+), 67 deletions(-)
>
> Shortlog
> --------
>
> Francois Romieu (2):
> Revert "r8169: enable ALDPS for power saving".
> Revert "r8169: enable internal ASPM and clock request settings".
>
Applied that patch on top of v3.8-rc7. Works well for me.
Thanks, Jörg
^ permalink raw reply
* [net-next (TAKE 2) 4/4] ipv6: IPv6 over Firewire (RFC3146) support.
From: YOSHIFUJI Hideaki @ 2013-02-10 9:02 UTC (permalink / raw)
To: netdev, linux1394-devel; +Cc: stephan.gatzka, davem, stefanr, yoshfuji
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
drivers/firewire/Kconfig | 6 ++--
drivers/firewire/net.c | 85 +++++++++++++++++++++++++++++++++++++++++-----
net/ipv6/addrconf.c | 20 ++++++++++-
3 files changed, 98 insertions(+), 13 deletions(-)
diff --git a/drivers/firewire/Kconfig b/drivers/firewire/Kconfig
index 7224533..ad8d16c 100644
--- a/drivers/firewire/Kconfig
+++ b/drivers/firewire/Kconfig
@@ -47,9 +47,9 @@ config FIREWIRE_NET
tristate "IP networking over 1394"
depends on FIREWIRE && INET
help
- This enables IPv4 over IEEE 1394, providing IP connectivity with
- other implementations of RFC 2734 as found on several operating
- systems. Multicast support is currently limited.
+ This enables IPv4/IPv6 over IEEE 1394, providing IP connectivity
+ with other implementations of RFC 2734/RFC3146 as found on
+ several operating systems. Multicast support is currently limited.
To compile this driver as a module, say M here: The module will be
called firewire-net.
diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c
index 281ef1b..7f87882 100644
--- a/drivers/firewire/net.c
+++ b/drivers/firewire/net.c
@@ -1,5 +1,6 @@
/*
* IPv4 over IEEE 1394, per RFC 2734
+ * IPv6 over IEEE 1394, per RFC 3146
*
* Copyright (C) 2009 Jay Fenlason <fenlason@redhat.com>
*
@@ -46,6 +47,7 @@
#define IANA_SPECIFIER_ID 0x00005eU
#define RFC2734_SW_VERSION 0x000001U
+#define RFC3146_SW_VERSION 0x000002U
#define IEEE1394_GASP_HDR_SIZE 8
@@ -516,6 +518,9 @@ static int fwnet_finish_incoming_packet(struct net_device *net,
switch (ether_type) {
case ETH_P_ARP:
case ETH_P_IP:
+#if IS_ENABLED(CONFIG_IPV6)
+ case ETH_P_IPV6:
+#endif
break;
default:
goto err;
@@ -781,11 +786,18 @@ static void fwnet_receive_broadcast(struct fw_iso_context *context,
ver = be32_to_cpu(buf_ptr[1]) & 0xffffff;
source_node_id = be32_to_cpu(buf_ptr[0]) >> 16;
- if (specifier_id == IANA_SPECIFIER_ID && ver == RFC2734_SW_VERSION) {
- buf_ptr += 2;
- length -= IEEE1394_GASP_HDR_SIZE;
- fwnet_incoming_packet(dev, buf_ptr, length, source_node_id,
- context->card->generation, true);
+ if (specifier_id == IANA_SPECIFIER_ID) {
+ switch (ver) {
+ case RFC2734_SW_VERSION:
+#if IS_ENABLED(CONFIG_IPV6)
+ case RFC3146_SW_VERSION:
+#endif
+ buf_ptr += 2;
+ length -= IEEE1394_GASP_HDR_SIZE;
+ fwnet_incoming_packet(dev, buf_ptr, length,
+ source_node_id,
+ context->card->generation, true);
+ }
}
packet.payload_length = dev->rcv_buffer_size;
@@ -992,8 +1004,18 @@ static int fwnet_send_packet(struct fwnet_packet_task *ptask)
p = skb_push(ptask->skb, IEEE1394_GASP_HDR_SIZE);
put_unaligned_be32(node_id << 16 | IANA_SPECIFIER_ID >> 8, p);
- put_unaligned_be32((IANA_SPECIFIER_ID & 0xff) << 24
- | RFC2734_SW_VERSION, &p[4]);
+
+ switch (ptask->skb->protocol) {
+ default:
+ put_unaligned_be32((IANA_SPECIFIER_ID & 0xff) << 24
+ | RFC2734_SW_VERSION, &p[4]);
+ break;
+#if IS_ENABLED(CONFIG_IPV6)
+ case htons(ETH_P_IPV6):
+ put_unaligned_be32((IANA_SPECIFIER_ID & 0xff) << 24
+ | RFC3146_SW_VERSION, &p[4]);
+#endif
+ }
/* We should not transmit if broadcast_channel.valid == 0. */
fw_send_request(dev->card, &ptask->transaction,
@@ -1215,6 +1237,9 @@ static netdev_tx_t fwnet_tx(struct sk_buff *skb, struct net_device *net)
switch (proto) {
case htons(ETH_P_ARP):
case htons(ETH_P_IP):
+#if IS_ENABLED(CONFIG_IPV6)
+ case htons(ETH_P_IPV6):
+#endif
break;
default:
goto fail;
@@ -1554,6 +1579,14 @@ static const struct ieee1394_device_id fwnet_id_table[] = {
.specifier_id = IANA_SPECIFIER_ID,
.version = RFC2734_SW_VERSION,
},
+#if IS_ENABLED(CONFIG_IPV6)
+ {
+ .match_flags = IEEE1394_MATCH_SPECIFIER_ID |
+ IEEE1394_MATCH_VERSION,
+ .specifier_id = IANA_SPECIFIER_ID,
+ .version = RFC3146_SW_VERSION,
+ },
+#endif
{ }
};
@@ -1591,6 +1624,30 @@ static struct fw_descriptor rfc2374_unit_directory = {
.data = rfc2374_unit_directory_data
};
+#if IS_ENABLED(CONFIG_IPV6)
+static const u32 rfc3146_unit_directory_data[] = {
+ 0x00040000, /* directory_length */
+ 0x1200005e, /* unit_specifier_id: IANA */
+ 0x81000003, /* textual descriptor offset */
+ 0x13000002, /* unit_sw_version: RFC 3146 */
+ 0x81000005, /* textual descriptor offset */
+ 0x00030000, /* descriptor_length */
+ 0x00000000, /* text */
+ 0x00000000, /* minimal ASCII, en */
+ 0x49414e41, /* I A N A */
+ 0x00030000, /* descriptor_length */
+ 0x00000000, /* text */
+ 0x00000000, /* minimal ASCII, en */
+ 0x49507636, /* I P v 6 */
+};
+
+static struct fw_descriptor rfc3146_unit_directory = {
+ .length = ARRAY_SIZE(rfc3146_unit_directory_data),
+ .key = (CSR_DIRECTORY | CSR_UNIT) << 24,
+ .data = rfc3146_unit_directory_data
+};
+#endif
+
static int __init fwnet_init(void)
{
int err;
@@ -1599,11 +1656,17 @@ static int __init fwnet_init(void)
if (err)
return err;
+#if IS_ENABLED(CONFIG_IPV6)
+ err = fw_core_add_descriptor(&rfc3146_unit_directory);
+ if (err)
+ goto out;
+#endif
+
fwnet_packet_task_cache = kmem_cache_create("packet_task",
sizeof(struct fwnet_packet_task), 0, 0, NULL);
if (!fwnet_packet_task_cache) {
err = -ENOMEM;
- goto out;
+ goto out2;
}
err = driver_register(&fwnet_driver.driver);
@@ -1611,7 +1674,11 @@ static int __init fwnet_init(void)
return 0;
kmem_cache_destroy(fwnet_packet_task_cache);
+out2:
+#if IS_ENABLED(CONFIG_IPV6)
+ fw_core_remove_descriptor(&rfc3146_unit_directory);
out:
+#endif
fw_core_remove_descriptor(&rfc2374_unit_directory);
return err;
@@ -1627,6 +1694,6 @@ static void __exit fwnet_cleanup(void)
module_exit(fwnet_cleanup);
MODULE_AUTHOR("Jay Fenlason <fenlason@redhat.com>");
-MODULE_DESCRIPTION("IPv4 over IEEE1394 as per RFC 2734");
+MODULE_DESCRIPTION("IP over IEEE1394 as per RFC 2734/3146");
MODULE_LICENSE("GPL");
MODULE_DEVICE_TABLE(ieee1394, fwnet_id_table);
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index bd9f936..060a08a 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -70,6 +70,7 @@
#include <net/snmp.h>
#include <net/af_ieee802154.h>
+#include <net/firewire.h>
#include <net/ipv6.h>
#include <net/protocol.h>
#include <net/ndisc.h>
@@ -1660,6 +1661,20 @@ static int addrconf_ifid_eui64(u8 *eui, struct net_device *dev)
return 0;
}
+static int addrconf_ifid_ieee1394(u8 *eui, struct net_device *dev)
+{
+ union fwnet_hwaddr *ha;
+
+ if (dev->addr_len != FWNET_ALEN)
+ return -1;
+
+ ha = (union fwnet_hwaddr *)dev->dev_addr;
+
+ memcpy(eui, &ha->uc.uniq_id, sizeof(ha->uc.uniq_id));
+ eui[0] ^= 2;
+ return 0;
+}
+
static int addrconf_ifid_arcnet(u8 *eui, struct net_device *dev)
{
/* XXX: inherit EUI-64 from other interface -- yoshfuji */
@@ -1724,6 +1739,8 @@ static int ipv6_generate_eui64(u8 *eui, struct net_device *dev)
return addrconf_ifid_gre(eui, dev);
case ARPHRD_IEEE802154:
return addrconf_ifid_eui64(eui, dev);
+ case ARPHRD_IEEE1394:
+ return addrconf_ifid_ieee1394(eui, dev);
}
return -1;
}
@@ -2565,7 +2582,8 @@ static void addrconf_dev_config(struct net_device *dev)
(dev->type != ARPHRD_FDDI) &&
(dev->type != ARPHRD_ARCNET) &&
(dev->type != ARPHRD_INFINIBAND) &&
- (dev->type != ARPHRD_IEEE802154)) {
+ (dev->type != ARPHRD_IEEE802154) &&
+ (dev->type != ARPHRD_IEEE1394)) {
/* Alas, we support only Ethernet autoconfiguration. */
return;
}
--
1.7.9.5
^ permalink raw reply related
* [net-next (TAKE 2) 3/4] firewire net, ipv4 arp: Extend hardware address and remove driver-level packet inspection.
From: YOSHIFUJI Hideaki @ 2013-02-10 9:02 UTC (permalink / raw)
To: netdev, linux1394-devel; +Cc: stephan.gatzka, davem, stefanr, yoshfuji
Inspection of upper layer protocol is considered harmful, especially
if it is about ARP or other stateful upper layer protocol; driver
cannot (and should not) have full state of them.
IPv4 over Firewire module used to inspect ARP (both in sending path
and in receiving path), and record peer's GUID, max packet size, max
speed and fifo address. This patch removes such inspection by extending
our "hardware address" definition to include other information as well:
max packet size, max speed anf fifo. By doing this, The neighbour
module in networking subsystem can cache them.
When a packet is being sent, the IP layer fills our virtual header with
the extended "hardware address", including GUID and fifo. The driver
can look-up node-id (the real but rather volatile low-level address)
by GUID, and then the module can send the packet to the wire using
parameters provided in the extendedn hardware address.
This approach is realistic because IP over IEEE1394 (RFC2734) and IPv6
oveer IEEE1394 (RFC3146) share same "hardware address" format
in their address resolution protocols.
Here, extended "hardware address" is defined as follows:
union fwnet_hwaddr {
u8 u[16];
struct {
__be64 uniq_id; /* EUI-64 */
u8 max_rec; /* max packet size */
u8 sspd; /* max speed */
__be16 fifo_hi; /* hi 16bits of FIFO addr */
__be32 fifo_lo; /* lo 32bits of FIFO addr */
} __packed uc;
};
Note that Hardware address is declared as union, so that we can map full
IP address into this, when implementing MCAP (Multicast Cannel Allocation
Protocol) for IPv6, but IP and ARP subsystem do not need to known this
format in detail.
One difference between original ARP (RFC826) and 1394 ARP (RFC2734)
is that 1394 ARP Request/Reply do not contain the target hardware address
field (aka ar$tha). This difference is handled in the ARP subsystem.
CC: Stephan Gatzka <stephan.gatzka@gmail.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
drivers/firewire/net.c | 215 ++++++++++++------------------------------------
include/linux/if_arp.h | 12 ++-
include/net/firewire.h | 25 ++++++
net/ipv4/arp.c | 18 ++--
4 files changed, 100 insertions(+), 170 deletions(-)
create mode 100644 include/net/firewire.h
diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c
index a251379..281ef1b 100644
--- a/drivers/firewire/net.c
+++ b/drivers/firewire/net.c
@@ -28,6 +28,7 @@
#include <asm/unaligned.h>
#include <net/arp.h>
+#include <net/firewire.h>
/* rx limits */
#define FWNET_MAX_FRAGMENTS 30 /* arbitrary, > TX queue depth */
@@ -57,33 +58,6 @@
#define RFC2374_HDR_LASTFRAG 2 /* last fragment */
#define RFC2374_HDR_INTFRAG 3 /* interior fragment */
-#define RFC2734_HW_ADDR_LEN 16
-
-struct rfc2734_arp {
- __be16 hw_type; /* 0x0018 */
- __be16 proto_type; /* 0x0806 */
- u8 hw_addr_len; /* 16 */
- u8 ip_addr_len; /* 4 */
- __be16 opcode; /* ARP Opcode */
- /* Above is exactly the same format as struct arphdr */
-
- __be64 s_uniq_id; /* Sender's 64bit EUI */
- u8 max_rec; /* Sender's max packet size */
- u8 sspd; /* Sender's max speed */
- __be16 fifo_hi; /* hi 16bits of sender's FIFO addr */
- __be32 fifo_lo; /* lo 32bits of sender's FIFO addr */
- __be32 sip; /* Sender's IP Address */
- __be32 tip; /* IP Address of requested hw addr */
-} __packed;
-
-/* This header format is specific to this driver implementation. */
-#define FWNET_ALEN 8
-#define FWNET_HLEN 10
-struct fwnet_header {
- u8 h_dest[FWNET_ALEN]; /* destination address */
- __be16 h_proto; /* packet type ID field */
-} __packed;
-
static bool fwnet_hwaddr_is_multicast(u8 *ha)
{
return !!(*ha & 1);
@@ -196,18 +170,14 @@ struct fwnet_peer {
struct list_head peer_link;
struct fwnet_device *dev;
u64 guid;
- u64 fifo;
- __be32 ip;
/* guarded by dev->lock */
struct list_head pd_list; /* received partial datagrams */
unsigned pdg_size; /* pd_list size */
u16 datagram_label; /* outgoing datagram label */
- u16 max_payload; /* includes RFC2374_FRAG_HDR_SIZE overhead */
int node_id;
int generation;
- unsigned speed;
};
/* This is our task struct. It's used for the packet complete callback. */
@@ -227,6 +197,28 @@ struct fwnet_packet_task {
};
/*
+ * Get fifo address embedded in hwaddr
+ */
+static __u64 fwnet_hwaddr_fifo(union fwnet_hwaddr *ha)
+{
+ return (u64)get_unaligned_be16(&ha->uc.fifo_hi) << 32
+ | get_unaligned_be32(&ha->uc.fifo_lo);
+}
+
+static __u8 fwnet_fixup_speed(__u8 sspd)
+{
+ if (sspd > SCODE_3200)
+ sspd = SCODE_3200;
+ return sspd;
+}
+
+static unsigned fwnet_max_payload(unsigned max_rec, unsigned speed);
+static __u16 fwnet_hwaddr_maxpayload(union fwnet_hwaddr *ha)
+{
+ return fwnet_max_payload(ha->uc.max_rec, ha->uc.sspd);
+}
+
+/*
* saddr == NULL means use device source address.
* daddr == NULL means leave destination address (eg unresolved arp).
*/
@@ -518,7 +510,6 @@ static int fwnet_finish_incoming_packet(struct net_device *net,
bool is_broadcast, u16 ether_type)
{
struct fwnet_device *dev;
- static const __be64 broadcast_hw = cpu_to_be64(~0ULL);
int status;
__be64 guid;
@@ -537,92 +528,13 @@ static int fwnet_finish_incoming_packet(struct net_device *net,
/*
* Parse the encapsulation header. This actually does the job of
- * converting to an ethernet frame header, as well as arp
- * conversion if needed. ARP conversion is easier in this
- * direction, since we are using ethernet as our backend.
+ * converting to an ethernet-like pseudo frame header.
*/
- /*
- * If this is an ARP packet, convert it. First, we want to make
- * use of some of the fields, since they tell us a little bit
- * about the sending machine.
- */
- if (ether_type == ETH_P_ARP) {
- struct rfc2734_arp *arp1394;
- struct arphdr *arp;
- unsigned char *arp_ptr;
- u64 fifo_addr;
- u64 peer_guid;
- unsigned sspd;
- u16 max_payload;
- struct fwnet_peer *peer;
- unsigned long flags;
-
- arp1394 = (struct rfc2734_arp *)skb->data;
- arp = (struct arphdr *)skb->data;
- arp_ptr = (unsigned char *)(arp + 1);
- peer_guid = get_unaligned_be64(&arp1394->s_uniq_id);
- fifo_addr = (u64)get_unaligned_be16(&arp1394->fifo_hi) << 32
- | get_unaligned_be32(&arp1394->fifo_lo);
-
- sspd = arp1394->sspd;
- /* Sanity check. OS X 10.3 PPC reportedly sends 131. */
- if (sspd > SCODE_3200) {
- dev_notice(&net->dev, "sspd %x out of range\n", sspd);
- sspd = SCODE_3200;
- }
- max_payload = fwnet_max_payload(arp1394->max_rec, sspd);
-
- spin_lock_irqsave(&dev->lock, flags);
- peer = fwnet_peer_find_by_guid(dev, peer_guid);
- if (peer) {
- peer->fifo = fifo_addr;
-
- if (peer->speed > sspd)
- peer->speed = sspd;
- if (peer->max_payload > max_payload)
- peer->max_payload = max_payload;
-
- peer->ip = arp1394->sip;
- }
- spin_unlock_irqrestore(&dev->lock, flags);
-
- if (!peer) {
- dev_notice(&net->dev,
- "no peer for ARP packet from %016llx\n",
- (unsigned long long)peer_guid);
- goto no_peer;
- }
-
- /*
- * Now that we're done with the 1394 specific stuff, we'll
- * need to alter some of the data. Believe it or not, all
- * that needs to be done is sender_IP_address needs to be
- * moved, the destination hardware address get stuffed
- * in and the hardware address length set to 8.
- *
- * IMPORTANT: The code below overwrites 1394 specific data
- * needed above so keep the munging of the data for the
- * higher level IP stack last.
- */
-
- arp->ar_hln = 8;
- /* skip over sender unique id */
- arp_ptr += arp->ar_hln;
- /* move sender IP addr */
- put_unaligned(arp1394->sip, (u32 *)arp_ptr);
- /* skip over sender IP addr */
- arp_ptr += arp->ar_pln;
-
- if (arp->ar_op == htons(ARPOP_REQUEST))
- memset(arp_ptr, 0, sizeof(u64));
- else
- memcpy(arp_ptr, net->dev_addr, sizeof(u64));
- }
/* Now add the ethernet header. */
guid = cpu_to_be64(dev->card->guid);
if (dev_hard_header(skb, net, ether_type,
- is_broadcast ? &broadcast_hw : &guid,
+ is_broadcast ? net->broadcast : net->dev_addr,
NULL, skb->len) >= 0) {
struct fwnet_header *eth;
u16 *rawp;
@@ -665,8 +577,7 @@ static int fwnet_finish_incoming_packet(struct net_device *net,
return 0;
- no_peer:
- err:
+err:
net->stats.rx_errors++;
net->stats.rx_dropped++;
@@ -1140,18 +1051,6 @@ static int fwnet_broadcast_start(struct fwnet_device *dev)
unsigned long offset;
unsigned u;
- if (dev->local_fifo == FWNET_NO_FIFO_ADDR) {
- dev->handler.length = 4096;
- dev->handler.address_callback = fwnet_receive_packet;
- dev->handler.callback_data = dev;
-
- retval = fw_core_add_address_handler(&dev->handler,
- &fw_high_memory_region);
- if (retval < 0)
- goto failed_initial;
-
- dev->local_fifo = dev->handler.offset;
- }
max_receive = 1U << (dev->card->max_receive + 1);
num_packets = (FWNET_ISO_PAGE_COUNT * PAGE_SIZE) / max_receive;
@@ -1234,8 +1133,6 @@ static int fwnet_broadcast_start(struct fwnet_device *dev)
dev->broadcast_rcv_context = NULL;
failed_context_create:
fw_core_remove_address_handler(&dev->handler);
- failed_initial:
- dev->local_fifo = FWNET_NO_FIFO_ADDR;
return retval;
}
@@ -1339,42 +1236,23 @@ static netdev_tx_t fwnet_tx(struct sk_buff *skb, struct net_device *net)
ptask->dest_node = IEEE1394_ALL_NODES;
ptask->speed = SCODE_100;
} else {
- __be64 guid = get_unaligned((__be64 *)hdr_buf.h_dest);
+ union fwnet_hwaddr *ha = (union fwnet_hwaddr *)hdr_buf.h_dest;
+ __be64 guid = get_unaligned(&ha->uc.uniq_id);
u8 generation;
peer = fwnet_peer_find_by_guid(dev, be64_to_cpu(guid));
- if (!peer || peer->fifo == FWNET_NO_FIFO_ADDR)
+ if (!peer)
goto fail;
generation = peer->generation;
dest_node = peer->node_id;
- max_payload = peer->max_payload;
+ max_payload = fwnet_hwaddr_maxpayload(ha);
datagram_label_ptr = &peer->datagram_label;
- ptask->fifo_addr = peer->fifo;
+ ptask->fifo_addr = fwnet_hwaddr_fifo(ha);
ptask->generation = generation;
ptask->dest_node = dest_node;
- ptask->speed = peer->speed;
- }
-
- /* If this is an ARP packet, convert it */
- if (proto == htons(ETH_P_ARP)) {
- struct arphdr *arp = (struct arphdr *)skb->data;
- unsigned char *arp_ptr = (unsigned char *)(arp + 1);
- struct rfc2734_arp *arp1394 = (struct rfc2734_arp *)skb->data;
- __be32 ipaddr;
-
- ipaddr = get_unaligned((__be32 *)(arp_ptr + FWNET_ALEN));
-
- arp1394->hw_addr_len = RFC2734_HW_ADDR_LEN;
- arp1394->max_rec = dev->card->max_receive;
- arp1394->sspd = dev->card->link_speed;
-
- put_unaligned_be16(dev->local_fifo >> 32,
- &arp1394->fifo_hi);
- put_unaligned_be32(dev->local_fifo & 0xffffffff,
- &arp1394->fifo_lo);
- put_unaligned(ipaddr, &arp1394->sip);
+ ptask->speed = fwnet_fixup_speed(ha->uc.sspd);
}
ptask->hdr.w0 = 0;
@@ -1491,13 +1369,9 @@ static int fwnet_add_peer(struct fwnet_device *dev,
peer->dev = dev;
peer->guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
- peer->fifo = FWNET_NO_FIFO_ADDR;
- peer->ip = 0;
INIT_LIST_HEAD(&peer->pd_list);
peer->pdg_size = 0;
peer->datagram_label = 0;
- peer->speed = device->max_speed;
- peer->max_payload = fwnet_max_payload(device->max_rec, peer->speed);
peer->generation = device->generation;
smp_rmb();
@@ -1522,6 +1396,7 @@ static int fwnet_probe(struct device *_dev)
struct fwnet_device *dev;
unsigned max_mtu;
int ret;
+ union fwnet_hwaddr *ha;
mutex_lock(&fwnet_device_mutex);
@@ -1546,12 +1421,21 @@ static int fwnet_probe(struct device *_dev)
dev->broadcast_rcv_context = NULL;
dev->broadcast_xmt_max_payload = 0;
dev->broadcast_xmt_datagramlabel = 0;
- dev->local_fifo = FWNET_NO_FIFO_ADDR;
dev->queued_datagrams = 0;
INIT_LIST_HEAD(&dev->peer_list);
dev->card = card;
dev->netdev = net;
+ dev->handler.length = 4096;
+ dev->handler.address_callback = fwnet_receive_packet;
+ dev->handler.callback_data = dev;
+
+ ret = fw_core_add_address_handler(&dev->handler, &fw_high_memory_region);
+ if (ret < 0)
+ goto out;
+
+ dev->local_fifo = dev->handler.offset;
+
/*
* Use the RFC 2734 default 1500 octets or the maximum payload
* as initial MTU
@@ -1561,8 +1445,15 @@ static int fwnet_probe(struct device *_dev)
net->mtu = min(1500U, max_mtu);
/* Set our hardware address while we're at it */
- put_unaligned_be64(card->guid, net->dev_addr);
- put_unaligned_be64(~0ULL, net->broadcast);
+ ha = (union fwnet_hwaddr *)net->dev_addr;
+ put_unaligned_be64(card->guid, &ha->uc.uniq_id);
+ ha->uc.max_rec = dev->card->max_receive;
+ ha->uc.sspd = dev->card->link_speed;
+ put_unaligned_be16(dev->local_fifo >> 32, &ha->uc.fifo_hi);
+ put_unaligned_be32(dev->local_fifo & 0xffffffff, &ha->uc.fifo_lo);
+
+ memset(net->broadcast, -1, net->addr_len);
+
ret = register_netdev(net);
if (ret)
goto out;
@@ -1611,8 +1502,6 @@ static int fwnet_remove(struct device *_dev)
mutex_lock(&fwnet_device_mutex);
net = dev->netdev;
- if (net && peer->ip)
- arp_invalidate(net, peer->ip);
fwnet_remove_peer(peer, dev);
diff --git a/include/linux/if_arp.h b/include/linux/if_arp.h
index 89b4614..f563907 100644
--- a/include/linux/if_arp.h
+++ b/include/linux/if_arp.h
@@ -33,7 +33,15 @@ static inline struct arphdr *arp_hdr(const struct sk_buff *skb)
static inline int arp_hdr_len(struct net_device *dev)
{
- /* ARP header, plus 2 device addresses, plus 2 IP addresses. */
- return sizeof(struct arphdr) + (dev->addr_len + sizeof(u32)) * 2;
+ switch (dev->type) {
+#if IS_ENABLED(CONFIG_FIREWIRE_NET)
+ case ARPHRD_IEEE1394:
+ /* ARP header, device address and 2 IP addresses */
+ return sizeof(struct arphdr) + dev->addr_len + sizeof(u32) * 2;
+#endif
+ default:
+ /* ARP header, plus 2 device addresses, plus 2 IP addresses. */
+ return sizeof(struct arphdr) + (dev->addr_len + sizeof(u32)) * 2;
+ }
}
#endif /* _LINUX_IF_ARP_H */
diff --git a/include/net/firewire.h b/include/net/firewire.h
new file mode 100644
index 0000000..bec4b60
--- /dev/null
+++ b/include/net/firewire.h
@@ -0,0 +1,25 @@
+#ifndef _NET_FIREWIRE_H
+#define _NET_FIREWIRE_H
+
+/* Pseude L2 address */
+#define FWNET_ALEN 16
+union fwnet_hwaddr {
+ u8 u[FWNET_ALEN];
+ /* "Hardware address" defined in RFC2734/RF3146 */
+ struct {
+ __be64 uniq_id; /* EUI-64 */
+ u8 max_rec; /* max packet size */
+ u8 sspd; /* max speed */
+ __be16 fifo_hi; /* hi 16bits of FIFO addr */
+ __be32 fifo_lo; /* lo 32bits of FIFO addr */
+ } __packed uc;
+};
+
+/* Pseudo L2 Header */
+#define FWNET_HLEN 18
+struct fwnet_header {
+ u8 h_dest[FWNET_ALEN]; /* destination address */
+ __be16 h_proto; /* packet type ID field */
+} __packed;
+
+#endif
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 9547a273..2db3af3 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -654,11 +654,19 @@ struct sk_buff *arp_create(int type, int ptype, __be32 dest_ip,
arp_ptr += dev->addr_len;
memcpy(arp_ptr, &src_ip, 4);
arp_ptr += 4;
- if (target_hw != NULL)
- memcpy(arp_ptr, target_hw, dev->addr_len);
- else
- memset(arp_ptr, 0, dev->addr_len);
- arp_ptr += dev->addr_len;
+
+ switch (dev->type) {
+#if IS_ENABLED(CONFIG_FIREWIRE_NET)
+ case ARPHRD_IEEE1394:
+ break;
+#endif
+ default:
+ if (target_hw != NULL)
+ memcpy(arp_ptr, target_hw, dev->addr_len);
+ else
+ memset(arp_ptr, 0, dev->addr_len);
+ arp_ptr += dev->addr_len;
+ }
memcpy(arp_ptr, &dest_ip, 4);
return skb;
--
1.7.9.5
^ permalink raw reply related
* [net-next (TAKE 2) 2/4] firewire net: Send L2 multicast via GASP.
From: YOSHIFUJI Hideaki @ 2013-02-10 9:02 UTC (permalink / raw)
To: netdev, linux1394-devel; +Cc: stephan.gatzka, davem, stefanr, yoshfuji
Send L2 multicast packet via GASP, not by seeing upper-layer
protocol type/address.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
drivers/firewire/net.c | 12 +++++++-----
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c
index a748790..a251379 100644
--- a/drivers/firewire/net.c
+++ b/drivers/firewire/net.c
@@ -84,6 +84,11 @@ struct fwnet_header {
__be16 h_proto; /* packet type ID field */
} __packed;
+static bool fwnet_hwaddr_is_multicast(u8 *ha)
+{
+ return !!(*ha & 1);
+}
+
/* IPv4 and IPv6 encapsulation header */
struct rfc2734_header {
u32 w0;
@@ -626,7 +631,7 @@ static int fwnet_finish_incoming_packet(struct net_device *net,
skb_reset_mac_header(skb);
skb_pull(skb, sizeof(*eth));
eth = (struct fwnet_header *)skb_mac_header(skb);
- if (*eth->h_dest & 1) {
+ if (fwnet_hwaddr_is_multicast(eth->h_dest)) {
if (memcmp(eth->h_dest, net->broadcast,
net->addr_len) == 0)
skb->pkt_type = PACKET_BROADCAST;
@@ -1325,10 +1330,7 @@ static netdev_tx_t fwnet_tx(struct sk_buff *skb, struct net_device *net)
* Set the transmission type for the packet. ARP packets and IP
* broadcast packets are sent via GASP.
*/
- if (memcmp(hdr_buf.h_dest, net->broadcast, FWNET_ALEN) == 0
- || proto == htons(ETH_P_ARP)
- || (proto == htons(ETH_P_IP)
- && IN_MULTICAST(ntohl(ip_hdr(skb)->daddr)))) {
+ if (fwnet_hwaddr_is_multicast(hdr_buf.h_dest)) {
max_payload = dev->broadcast_xmt_max_payload;
datagram_label_ptr = &dev->broadcast_xmt_datagramlabel;
--
1.7.9.5
^ permalink raw reply related
* [net-next (TAKE 2) 1/4] firewire net: Accept IPv4 and ARP only.
From: YOSHIFUJI Hideaki @ 2013-02-10 9:02 UTC (permalink / raw)
To: netdev, linux1394-devel; +Cc: yoshfuji, stefanr, davem
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
drivers/firewire/net.c | 21 +++++++++++++++++++--
1 file changed, 19 insertions(+), 2 deletions(-)
diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c
index 2b27bff..a748790 100644
--- a/drivers/firewire/net.c
+++ b/drivers/firewire/net.c
@@ -517,6 +517,14 @@ static int fwnet_finish_incoming_packet(struct net_device *net,
int status;
__be64 guid;
+ switch (ether_type) {
+ case ETH_P_ARP:
+ case ETH_P_IP:
+ break;
+ default:
+ goto err;
+ }
+
dev = netdev_priv(net);
/* Write metadata, and then pass to the receive level */
skb->dev = net;
@@ -653,6 +661,7 @@ static int fwnet_finish_incoming_packet(struct net_device *net,
return 0;
no_peer:
+ err:
net->stats.rx_errors++;
net->stats.rx_dropped++;
@@ -1299,9 +1308,17 @@ static netdev_tx_t fwnet_tx(struct sk_buff *skb, struct net_device *net)
* We might need to rebuild the header on tx failure.
*/
memcpy(&hdr_buf, skb->data, sizeof(hdr_buf));
- skb_pull(skb, sizeof(hdr_buf));
-
proto = hdr_buf.h_proto;
+
+ switch (proto) {
+ case htons(ETH_P_ARP):
+ case htons(ETH_P_IP):
+ break;
+ default:
+ goto fail;
+ }
+
+ skb_pull(skb, sizeof(hdr_buf));
dg_size = skb->len;
/*
--
1.7.9.5
------------------------------------------------------------------------------
Free Next-Gen Firewall Hardware Offer
Buy your Sophos next-gen firewall before the end March 2013
and get the hardware for free! Learn more.
http://p.sf.net/sfu/sophos-d2d-feb
^ permalink raw reply related
* [net-next (TAKE 2) 0/4] IPv6 over Firewire
From: YOSHIFUJI Hideaki @ 2013-02-10 9:02 UTC (permalink / raw)
To: netdev, linux1394-devel; +Cc: stephan.gatzka, davem, stefanr, yoshfuji
This is TAKE 2 of IPv6 over IEEE1394(Firewire) (RFC3146) support
by extended MAC approach.
YOSHIFUJI Hideaki (4):
firewire net: Accept IPv4 and ARP only.
firewire net: Send L2 multicast via GASP.
firewire net, ipv4 arp: Extend hardware address and remove
driver-level packet inspection.
ipv6: IPv6 over Firewire (RFC3146) support.
drivers/firewire/Kconfig | 6 +-
drivers/firewire/net.c | 329 +++++++++++++++++++++-------------------------
include/linux/if_arp.h | 12 +-
include/net/firewire.h | 25 ++++
net/ipv4/arp.c | 18 ++-
net/ipv6/addrconf.c | 20 ++-
6 files changed, 222 insertions(+), 188 deletions(-)
create mode 100644 include/net/firewire.h
--
1.7.9.5
^ permalink raw reply
* Re: [patch net-next v3 08/11] tbf: fix value set for q->ptokens
From: Jiri Pirko @ 2013-02-10 8:51 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev, davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <1360484498.20362.12.camel@edumazet-glaptop>
Sun, Feb 10, 2013 at 09:21:38AM CET, eric.dumazet@gmail.com wrote:
>On Sun, 2013-02-10 at 09:18 +0100, Jiri Pirko wrote:
>> Sun, Feb 10, 2013 at 02:30:46AM CET, eric.dumazet@gmail.com wrote:
>> >On Sat, 2013-02-09 at 17:45 +0100, Jiri Pirko wrote:
>> >> q->ptokens is in ns and we are assigning q->mtu directly to it. That is
>> >> wrong. psched_l2t_ns() should be used to compute correct value.
>> >>
>> >
>> >
>> >Not clear why its a separate patch, and not folded in the 6th
>>
>> This is independent on 6th. Would be needed even if 6th wouldn't be
>> there.
>
>When was this bug introduced then ?
This has been present since the beginning of git. I'm unable to find bk
repo right now to tell you bk commit :/
>
>
^ permalink raw reply
* [PATCH v2 net-next] skbuff: create skb_panic() to let caller specify panic
From: Jean Sacren @ 2013-02-10 8:23 UTC (permalink / raw)
To: netdev
Combine skb_over_panic() and skb_under_panic() into skb_panic() and let
the caller specify whether it is skb_over_panic or skb_under_panic.
In skb_panic() definition, change 'int sz' and 'here' to 'unsigned int
size' and 'addr' for clarity, and accommodate the output message.
Rewrite kernel doc for skb_panic().
Signed-off-by: Jean Sacren <sakiwit@gmail.com>
---
v2: don't split format over multiple lines as advised by Joe Perches.
net/core/skbuff.c | 49 +++++++++++++++----------------------------------
1 file changed, 15 insertions(+), 34 deletions(-)
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 6114c11..86d915b 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -104,48 +104,27 @@ static const struct pipe_buf_operations sock_pipe_buf_ops = {
.get = sock_pipe_buf_get,
};
-/*
- * Keep out-of-line to prevent kernel bloat.
- * __builtin_return_address is not used because it is not always
- * reliable.
- */
-
/**
- * skb_over_panic - private function
- * @skb: buffer
- * @sz: size
- * @here: address
+ * skb_panic - private function for out-of-line support
+ * @skb: buffer
+ * @size: size
+ * @addr: address
+ * @panic: skb_over_panic or skb_under_panic
*
- * Out of line support code for skb_put(). Not user callable.
- */
-static void skb_over_panic(struct sk_buff *skb, int sz, void *here)
-{
- pr_emerg("%s: text:%p len:%d put:%d head:%p data:%p tail:%#lx end:%#lx dev:%s\n",
- __func__, here, skb->len, sz, skb->head, skb->data,
- (unsigned long)skb->tail, (unsigned long)skb->end,
- skb->dev ? skb->dev->name : "<NULL>");
- BUG();
-}
-
-/**
- * skb_under_panic - private function
- * @skb: buffer
- * @sz: size
- * @here: address
- *
- * Out of line support code for skb_push(). Not user callable.
+ * Out-of-line support for skb_put() and skb_push(). Not user callable.
+ * Keep out-of-line to prevent kernel bloat.
+ * __builtin_return_address is not used because it is not always reliable.
*/
-
-static void skb_under_panic(struct sk_buff *skb, int sz, void *here)
+static void skb_panic(struct sk_buff *skb, unsigned int size, void *addr,
+ const char panic[])
{
pr_emerg("%s: text:%p len:%d put:%d head:%p data:%p tail:%#lx end:%#lx dev:%s\n",
- __func__, here, skb->len, sz, skb->head, skb->data,
+ panic, addr, skb->len, size, skb->head, skb->data,
(unsigned long)skb->tail, (unsigned long)skb->end,
skb->dev ? skb->dev->name : "<NULL>");
BUG();
}
-
/*
* kmalloc_reserve is a wrapper around kmalloc_node_track_caller that tells
* the caller if emergency pfmemalloc reserves are being used. If it is and
@@ -1259,12 +1238,13 @@ EXPORT_SYMBOL(skb_pad);
*/
unsigned char *skb_put(struct sk_buff *skb, unsigned int len)
{
+ const char panic[] = "skb_over_panic";
unsigned char *tmp = skb_tail_pointer(skb);
SKB_LINEAR_ASSERT(skb);
skb->tail += len;
skb->len += len;
if (unlikely(skb->tail > skb->end))
- skb_over_panic(skb, len, __builtin_return_address(0));
+ skb_panic(skb, len, __builtin_return_address(0), panic);
return tmp;
}
EXPORT_SYMBOL(skb_put);
@@ -1280,10 +1260,11 @@ EXPORT_SYMBOL(skb_put);
*/
unsigned char *skb_push(struct sk_buff *skb, unsigned int len)
{
+ const char panic[] = "skb_under_panic";
skb->data -= len;
skb->len += len;
if (unlikely(skb->data<skb->head))
- skb_under_panic(skb, len, __builtin_return_address(0));
+ skb_panic(skb, len, __builtin_return_address(0), panic);
return skb->data;
}
EXPORT_SYMBOL(skb_push);
^ permalink raw reply related
* Re: [patch net-next v3 08/11] tbf: fix value set for q->ptokens
From: Eric Dumazet @ 2013-02-10 8:21 UTC (permalink / raw)
To: Jiri Pirko; +Cc: netdev, davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <20130210081824.GA1570@minipsycho.orion>
On Sun, 2013-02-10 at 09:18 +0100, Jiri Pirko wrote:
> Sun, Feb 10, 2013 at 02:30:46AM CET, eric.dumazet@gmail.com wrote:
> >On Sat, 2013-02-09 at 17:45 +0100, Jiri Pirko wrote:
> >> q->ptokens is in ns and we are assigning q->mtu directly to it. That is
> >> wrong. psched_l2t_ns() should be used to compute correct value.
> >>
> >
> >
> >Not clear why its a separate patch, and not folded in the 6th
>
> This is independent on 6th. Would be needed even if 6th wouldn't be
> there.
When was this bug introduced then ?
^ permalink raw reply
* Re: [patch net-next v3 08/11] tbf: fix value set for q->ptokens
From: Jiri Pirko @ 2013-02-10 8:18 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev, davem, edumazet, jhs, kuznet, j.vimal
In-Reply-To: <1360459846.20362.2.camel@edumazet-glaptop>
Sun, Feb 10, 2013 at 02:30:46AM CET, eric.dumazet@gmail.com wrote:
>On Sat, 2013-02-09 at 17:45 +0100, Jiri Pirko wrote:
>> q->ptokens is in ns and we are assigning q->mtu directly to it. That is
>> wrong. psched_l2t_ns() should be used to compute correct value.
>>
>
>
>Not clear why its a separate patch, and not folded in the 6th
This is independent on 6th. Would be needed even if 6th wouldn't be
there.
>
>
>> Signed-off-by: Jiri Pirko <jiri@resnulli.us>
>> ---
>> net/sched/sch_tbf.c | 9 +++++----
>> 1 file changed, 5 insertions(+), 4 deletions(-)
>>
>> diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
>> index dc562a8..6e8b670 100644
>> --- a/net/sched/sch_tbf.c
>> +++ b/net/sched/sch_tbf.c
>> @@ -165,8 +165,8 @@ static struct sk_buff *tbf_dequeue(struct Qdisc *sch)
>>
>> if (q->peak_present) {
>> ptoks = toks + q->ptokens;
>> - if (ptoks > (long)q->mtu)
>> - ptoks = q->mtu;
>> + if (ptoks > (s64) psched_l2t_ns(&q->peak, q->mtu))
>> + ptoks = (s64) psched_l2t_ns(&q->peak, q->mtu);
>
>It seems a bit expensive to perform this in fast path.
>
>Please add a variable to cache psched_l2t_ns(&q->peak, q->mtu)
Okay, I did not think that this is necessary, but sure, I will do that.
Thanks!
>
>
>> ptoks -= (s64) psched_l2t_ns(&q->peak, len);
>> }
>> toks += q->tokens;
>> @@ -215,7 +215,8 @@ static void tbf_reset(struct Qdisc *sch)
>> sch->q.qlen = 0;
>> q->t_c = ktime_to_ns(ktime_get());
>> q->tokens = q->buffer;
>> - q->ptokens = q->mtu;
>> + if (q->peak_present)
>> + q->ptokens = (s64) psched_l2t_ns(&q->peak, q->mtu);
>> qdisc_watchdog_cancel(&q->watchdog);
>> }
>>
>> @@ -296,11 +297,11 @@ static int tbf_change(struct Qdisc *sch, struct nlattr *opt)
>> q->max_size = max_size;
>> q->buffer = PSCHED_TICKS2NS(qopt->buffer);
>> q->tokens = q->buffer;
>> - q->ptokens = q->mtu;
>>
>> psched_ratecfg_precompute(&q->rate, rtab->rate.rate);
>> if (ptab) {
>> psched_ratecfg_precompute(&q->peak, ptab->rate.rate);
>> + q->ptokens = (s64) psched_l2t_ns(&q->peak, q->mtu);
>
>Here probably.
>
>
>
^ permalink raw reply
* Re: Softirqs without captured packets
From: Eric Dumazet @ 2013-02-10 8:16 UTC (permalink / raw)
To: Javier Domingo; +Cc: David Miller, netdev
In-Reply-To: <CALZVap=_x_nvnH55n7SGMxFGCR9XPKkNzSrt0J1dp41g-HUAqQ@mail.gmail.com>
Please do not top post on netdev
On Sun, 2013-02-10 at 04:02 +0100, Javier Domingo wrote:
> Well, that explains why the 0 :D. I was getting nut. Any way, does
> that happen in e1000e or tg3 drivers? If not, then there is still
> without explanation.
>
The answer is yes.
> Respecting the low numbers, I will do some futher stress tests, but
> this measures where done pinging the computer at 0.001 interval:
>
> ping -i 0.0001 <computer ip>
>
> And received packets where at much, 3 per softirq. Also, I tried
> opening an ftp download session but didn't get much high numbers
> (1Gbps link).
>
Machine is fast enough, thats all.
> When I tried this with the injection done by a dag (hardware injector)
> there were strange results because thought the 0s where explained, the
> were appearing thought the rest of softirqs where about 280+-50.
Have you read this part of net_rx_action() ?
unsigned long time_limit = jiffies + 2;
/* If softirq window is exhuasted then punt.
* Allow this to run for 2 jiffies since which will allow
* an average latency of 1.5/HZ.
*/
if (unlikely(budget <= 0 || time_after(jiffies, time_limit)))
goto softnet_break;
This means your machine handles 280+50 packets per 1.5/HZ
(Plus the hidden TX completions...)
^ 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