* [Patch net-next v3 4/5] net_sched: act: refuse to remove bound action outside
From: Cong Wang @ 2014-02-12 1:07 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang, Jamal Hadi Salim, David S. Miller
In-Reply-To: <1392167255-21744-1-git-send-email-xiyou.wangcong@gmail.com>
When an action is bonnd to a filter, there is no point to
remove it outside. Currently we just silently decrease the refcnt,
we should reject this explicitly with EPERM.
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
---
include/net/act_api.h | 2 +-
net/sched/act_api.c | 26 ++++++++++++++++++++------
2 files changed, 21 insertions(+), 7 deletions(-)
diff --git a/include/net/act_api.h b/include/net/act_api.h
index 969cac6..3ee4c92 100644
--- a/include/net/act_api.h
+++ b/include/net/act_api.h
@@ -109,7 +109,7 @@ void tcf_hash_insert(struct tc_action *a);
int tcf_register_action(struct tc_action_ops *a, unsigned int mask);
int tcf_unregister_action(struct tc_action_ops *a);
-void tcf_action_destroy(struct list_head *actions, int bind);
+int tcf_action_destroy(struct list_head *actions, int bind);
int tcf_action_exec(struct sk_buff *skb, const struct list_head *actions,
struct tcf_result *res);
int tcf_action_init(struct net *net, struct nlattr *nla,
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index c88d382..27e4c53 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -53,6 +53,8 @@ int tcf_hash_release(struct tc_action *a, int bind)
if (p) {
if (bind)
p->tcfc_bindcnt--;
+ else if (p->tcfc_bindcnt > 0)
+ return -EPERM;
p->tcfc_refcnt--;
if (p->tcfc_bindcnt <= 0 && p->tcfc_refcnt <= 0) {
@@ -123,6 +125,7 @@ static int tcf_del_walker(struct sk_buff *skb, struct tc_action *a)
struct tcf_common *p;
struct nlattr *nest;
int i = 0, n_i = 0;
+ int ret = -EINVAL;
nest = nla_nest_start(skb, a->order);
if (nest == NULL)
@@ -133,10 +136,12 @@ static int tcf_del_walker(struct sk_buff *skb, struct tc_action *a)
head = &hinfo->htab[tcf_hash(i, hinfo->hmask)];
hlist_for_each_entry_safe(p, n, head, tcfc_head) {
a->priv = p;
- if (ACT_P_DELETED == tcf_hash_release(a, 0)) {
+ ret = tcf_hash_release(a, 0);
+ if (ret == ACT_P_DELETED) {
module_put(a->ops->owner);
n_i++;
- }
+ } else if (ret < 0)
+ goto nla_put_failure;
}
}
if (nla_put_u32(skb, TCA_FCNT, n_i))
@@ -146,7 +151,7 @@ static int tcf_del_walker(struct sk_buff *skb, struct tc_action *a)
return n_i;
nla_put_failure:
nla_nest_cancel(skb, nest);
- return -EINVAL;
+ return ret;
}
static int tcf_generic_walker(struct sk_buff *skb, struct netlink_callback *cb,
@@ -401,16 +406,21 @@ exec_done:
}
EXPORT_SYMBOL(tcf_action_exec);
-void tcf_action_destroy(struct list_head *actions, int bind)
+int tcf_action_destroy(struct list_head *actions, int bind)
{
struct tc_action *a, *tmp;
+ int ret = 0;
list_for_each_entry_safe(a, tmp, actions, list) {
- if (tcf_hash_release(a, bind) == ACT_P_DELETED)
+ ret = tcf_hash_release(a, bind);
+ if (ret == ACT_P_DELETED)
module_put(a->ops->owner);
+ else if (ret < 0)
+ return ret;
list_del(&a->list);
kfree(a);
}
+ return ret;
}
int
@@ -838,7 +848,11 @@ tcf_del_notify(struct net *net, struct nlmsghdr *n, struct list_head *actions,
}
/* now do the delete */
- tcf_action_destroy(actions, 0);
+ ret = tcf_action_destroy(actions, 0);
+ if (ret < 0) {
+ kfree_skb(skb);
+ return ret;
+ }
ret = rtnetlink_send(skb, net, portid, RTNLGRP_TC,
n->nlmsg_flags & NLM_F_ECHO);
--
1.8.3.1
^ permalink raw reply related
* [Patch net-next v3 5/5] net_sched: act: clean up tca_action_flush()
From: Cong Wang @ 2014-02-12 1:07 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang, Jamal Hadi Salim, David S. Miller
In-Reply-To: <1392167255-21744-1-git-send-email-xiyou.wangcong@gmail.com>
We could allocate tc_action on stack in tca_action_flush(),
since it is not large.
Also, we could use create_a() in tcf_action_get_1().
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
---
net/sched/act_api.c | 53 +++++++++++++++++++++++------------------------------
1 file changed, 23 insertions(+), 30 deletions(-)
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 27e4c53..8a5ba5a 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -685,6 +685,20 @@ act_get_notify(struct net *net, u32 portid, struct nlmsghdr *n,
return rtnl_unicast(skb, net, portid);
}
+static struct tc_action *create_a(int i)
+{
+ struct tc_action *act;
+
+ act = kzalloc(sizeof(*act), GFP_KERNEL);
+ if (act == NULL) {
+ pr_debug("create_a: failed to alloc!\n");
+ return NULL;
+ }
+ act->order = i;
+ INIT_LIST_HEAD(&act->list);
+ return act;
+}
+
static struct tc_action *
tcf_action_get_1(struct nlattr *nla, struct nlmsghdr *n, u32 portid)
{
@@ -704,11 +718,10 @@ tcf_action_get_1(struct nlattr *nla, struct nlmsghdr *n, u32 portid)
index = nla_get_u32(tb[TCA_ACT_INDEX]);
err = -ENOMEM;
- a = kzalloc(sizeof(struct tc_action), GFP_KERNEL);
+ a = create_a(0);
if (a == NULL)
goto err_out;
- INIT_LIST_HEAD(&a->list);
err = -EINVAL;
a->ops = tc_lookup_action(tb[TCA_ACT_KIND]);
if (a->ops == NULL) /* could happen in batch of actions */
@@ -738,20 +751,6 @@ static void cleanup_a(struct list_head *actions)
}
}
-static struct tc_action *create_a(int i)
-{
- struct tc_action *act;
-
- act = kzalloc(sizeof(*act), GFP_KERNEL);
- if (act == NULL) {
- pr_debug("create_a: failed to alloc!\n");
- return NULL;
- }
- act->order = i;
- INIT_LIST_HEAD(&act->list);
- return act;
-}
-
static int tca_action_flush(struct net *net, struct nlattr *nla,
struct nlmsghdr *n, u32 portid)
{
@@ -763,18 +762,12 @@ static int tca_action_flush(struct net *net, struct nlattr *nla,
struct nlattr *nest;
struct nlattr *tb[TCA_ACT_MAX + 1];
struct nlattr *kind;
- struct tc_action *a = create_a(0);
+ struct tc_action a;
int err = -ENOMEM;
- if (a == NULL) {
- pr_debug("tca_action_flush: couldnt create tc_action\n");
- return err;
- }
-
skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
if (!skb) {
pr_debug("tca_action_flush: failed skb alloc\n");
- kfree(a);
return err;
}
@@ -786,8 +779,10 @@ static int tca_action_flush(struct net *net, struct nlattr *nla,
err = -EINVAL;
kind = tb[TCA_ACT_KIND];
- a->ops = tc_lookup_action(kind);
- if (a->ops == NULL) /*some idjot trying to flush unknown action */
+ memset(&a, 0, sizeof(struct tc_action));
+ INIT_LIST_HEAD(&a.list);
+ a.ops = tc_lookup_action(kind);
+ if (a.ops == NULL) /*some idjot trying to flush unknown action */
goto err_out;
nlh = nlmsg_put(skb, portid, n->nlmsg_seq, RTM_DELACTION, sizeof(*t), 0);
@@ -802,7 +797,7 @@ static int tca_action_flush(struct net *net, struct nlattr *nla,
if (nest == NULL)
goto out_module_put;
- err = a->ops->walk(skb, &dcb, RTM_DELACTION, a);
+ err = a.ops->walk(skb, &dcb, RTM_DELACTION, &a);
if (err < 0)
goto out_module_put;
if (err == 0)
@@ -812,8 +807,7 @@ static int tca_action_flush(struct net *net, struct nlattr *nla,
nlh->nlmsg_len = skb_tail_pointer(skb) - b;
nlh->nlmsg_flags |= NLM_F_ROOT;
- module_put(a->ops->owner);
- kfree(a);
+ module_put(a.ops->owner);
err = rtnetlink_send(skb, net, portid, RTNLGRP_TC,
n->nlmsg_flags & NLM_F_ECHO);
if (err > 0)
@@ -822,11 +816,10 @@ static int tca_action_flush(struct net *net, struct nlattr *nla,
return err;
out_module_put:
- module_put(a->ops->owner);
+ module_put(a.ops->owner);
err_out:
noflush_out:
kfree_skb(skb);
- kfree(a);
return err;
}
--
1.8.3.1
^ permalink raw reply related
* [Patch net-next v2] net: allow setting mac address of loopback device
From: Cong Wang @ 2014-02-12 1:21 UTC (permalink / raw)
To: netdev
Cc: Hannes Frederic Sowa, Neil Horman, Stephen Hemminger,
Eric Dumazet, David S. Miller, Cong Wang
We are trying to mirror the local traffic from lo to eth0,
allowing setting mac address of lo to eth0 would make
the ether addresses in these packets correct, so that
we don't have to modify the ether header again.
Since usually no one cares about its mac address (all-zero),
it is safe to allow those who care to set its mac address.
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: Stephen Hemminger <stephen@networkplumber.org>
Cc: Eric Dumazet <edumazet@google.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
---
diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c
index c5011e0..e7c1d5f 100644
--- a/drivers/net/loopback.c
+++ b/drivers/net/loopback.c
@@ -160,6 +160,7 @@ static const struct net_device_ops loopback_ops = {
.ndo_init = loopback_dev_init,
.ndo_start_xmit= loopback_xmit,
.ndo_get_stats64 = loopback_get_stats64,
+ .ndo_set_mac_address = eth_mac_addr,
};
/*
@@ -174,6 +175,7 @@ static void loopback_setup(struct net_device *dev)
dev->tx_queue_len = 0;
dev->type = ARPHRD_LOOPBACK; /* 0x0001*/
dev->flags = IFF_LOOPBACK;
+ dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
dev->hw_features = NETIF_F_ALL_TSO | NETIF_F_UFO;
dev->features = NETIF_F_SG | NETIF_F_FRAGLIST
^ permalink raw reply related
* Re: [PATCH v2 2/2] sctp: optimize the sctp_sysctl_net_register
From: Wang Weidong @ 2014-02-12 1:21 UTC (permalink / raw)
To: Sergei Shtylyov, nhorman, davem, vyasevich; +Cc: dborkman, netdev
In-Reply-To: <52FA5418.6090007@cogentembedded.com>
On 2014/2/12 0:47, Sergei Shtylyov wrote:
> Hello.
>
> On 02/11/2014 04:49 AM, Wang Weidong wrote:
>
>> Here, when the net is init_net, we needn't to kmemdup the ctl_table
>> again. So add a check for net. Also we can save some memory.
>
>> Signed-off-by: Wang Weidong <wangweidong1@huawei.com>
>> ---
>> net/sctp/sysctl.c | 16 +++++++++-------
>> 1 file changed, 9 insertions(+), 7 deletions(-)
>
>> diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c
>> index 2ddb401..b65396b 100644
>> --- a/net/sctp/sysctl.c
>> +++ b/net/sctp/sysctl.c
>> @@ -403,15 +403,17 @@ static int proc_sctp_do_rto_max(struct ctl_table *ctl, int write,
>>
>> int sctp_sysctl_net_register(struct net *net)
>> {
>> - struct ctl_table *table;
>> - int i;
>> + struct ctl_table *table = sctp_net_table;
>>
>> - table = kmemdup(sctp_net_table, sizeof(sctp_net_table), GFP_KERNEL);
>> - if (!table)
>> - return -ENOMEM;
>> + if (!net_eq(net, &init_net)) {
>> + int i;
>
> Empty line after declaration wouldn't hurt, just like above.
>
Yeah. I will change it soon.
Thanks
Wang
>> + table = kmemdup(sctp_net_table, sizeof(sctp_net_table), GFP_KERNEL);
>> + if (!table)
>> + return -ENOMEM;
>
> WBR, Sergei
>
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
>
^ permalink raw reply
* [PATCH net-next] intel: Remove unnecessary OOM messages
From: Joe Perches @ 2014-02-12 1:26 UTC (permalink / raw)
To: Jeff Kirsher; +Cc: e1000-devel, netdev
Don't emit these as there's a generic OOM with a dump_stack()
on allocation failures.
Signed-off-by: Joe Perches <joe@perches.com>
---
^ permalink raw reply
* [PATCH 01/10] net: phy: use network device in phy_print_status
From: Florian Fainelli @ 2014-02-12 1:27 UTC (permalink / raw)
To: netdev; +Cc: davem, Florian Fainelli
In-Reply-To: <1392168462-18888-1-git-send-email-f.fainelli@gmail.com>
phy_print_status() currently uses dev_name(&phydev->dev) which will
usually result in printing something along those lines for Device Tree
aware drivers:
libphy: f0b60000.etherne:0a - Link is Down
libphy: f0ba0000.etherne:00 - Link is Up - 1000/Full
This is not terribly useful for network administrators or users since we
expect a network interface name to be able to correlate link events with
interfaces. Update phy_print_status() to use netdev_info() with
phydev->attached_dev which is the backing network device for our PHY
device. The leading dash is removed since netdev_info() prefixes the
messages with "<interface>: " already.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/phy/phy.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index 19c9eca..c35b2e7 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -45,12 +45,11 @@
void phy_print_status(struct phy_device *phydev)
{
if (phydev->link) {
- pr_info("%s - Link is Up - %d/%s\n",
- dev_name(&phydev->dev),
+ netdev_info(phydev->attached_dev, "Link is Up - %d/%s\n",
phydev->speed,
DUPLEX_FULL == phydev->duplex ? "Full" : "Half");
} else {
- pr_info("%s - Link is Down\n", dev_name(&phydev->dev));
+ netdev_info(phydev->attached_dev, "Link is Down\n");
}
}
EXPORT_SYMBOL(phy_print_status);
--
1.8.3.2
^ permalink raw reply related
* [PATCH 03/10] net: phy: display human readable PHY speed settings
From: Florian Fainelli @ 2014-02-12 1:27 UTC (permalink / raw)
To: netdev; +Cc: davem, Florian Fainelli
In-Reply-To: <1392168462-18888-1-git-send-email-f.fainelli@gmail.com>
Use a convenience function: phy_speed_to_str() which will display human
readable speeds.
Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/phy/phy.c | 24 ++++++++++++++++++++++--
1 file changed, 22 insertions(+), 2 deletions(-)
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index 8ae2260..36fc6e1 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -38,6 +38,26 @@
#include <asm/irq.h>
+static const char *phy_speed_to_str(int speed)
+{
+ switch (speed) {
+ case SPEED_10:
+ return "10Mbps";
+ case SPEED_100:
+ return "100Mbps";
+ case SPEED_1000:
+ return "1Gbps";
+ case SPEED_2500:
+ return "2.5Gbps";
+ case SPEED_10000:
+ return "10Gbps";
+ case SPEED_UNKNOWN:
+ return "Unknown";
+ default:
+ return "Unsupported (update phy.c)";
+ }
+}
+
/**
* phy_print_status - Convenience function to print out the current phy status
* @phydev: the phy_device struct
@@ -46,8 +66,8 @@ void phy_print_status(struct phy_device *phydev)
{
if (phydev->link) {
netdev_info(phydev->attached_dev,
- "Link is Up - %d/%s - flow control %s\n",
- phydev->speed,
+ "Link is Up - %s/%s - flow control %s\n",
+ phy_speed_to_str(phydev->speed),
DUPLEX_FULL == phydev->duplex ? "Full" : "Half",
phydev->pause ? "rx/tx" : "off");
} else {
--
1.8.3.2
^ permalink raw reply related
* [PATCH 02/10] net: phy: update phy_print_status to show pause settings
From: Florian Fainelli @ 2014-02-12 1:27 UTC (permalink / raw)
To: netdev; +Cc: davem, Florian Fainelli
In-Reply-To: <1392168462-18888-1-git-send-email-f.fainelli@gmail.com>
Update phy_print_status() to also display the PHY device pause settings
(rx/tx or off).
Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/phy/phy.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index c35b2e7..8ae2260 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -45,9 +45,11 @@
void phy_print_status(struct phy_device *phydev)
{
if (phydev->link) {
- netdev_info(phydev->attached_dev, "Link is Up - %d/%s\n",
+ netdev_info(phydev->attached_dev,
+ "Link is Up - %d/%s - flow control %s\n",
phydev->speed,
- DUPLEX_FULL == phydev->duplex ? "Full" : "Half");
+ DUPLEX_FULL == phydev->duplex ? "Full" : "Half",
+ phydev->pause ? "rx/tx" : "off");
} else {
netdev_info(phydev->attached_dev, "Link is Down\n");
}
--
1.8.3.2
^ permalink raw reply related
* [PATCH 07/10] net: phy: re-design phy_modes to be self-contained
From: Florian Fainelli @ 2014-02-12 1:27 UTC (permalink / raw)
To: netdev; +Cc: davem, Florian Fainelli
In-Reply-To: <1392168462-18888-1-git-send-email-f.fainelli@gmail.com>
of_get_phy_mode() uses a local array to map phy_interface_t values from
include/linux/net/phy.h to a string which is read from the 'phy-mode' or
'phy-connection-type' property. In preparation for exposing the PHY
interface mode through sysfs, perform the following:
- mode phy_modes from drivers/of/of_net.c to include/linux/phy.h such
that it is right below the phy_interface_t enum
- make it a static inline function returning the string such that we can
use it by just including include/linux/net/phy.h
- add a PHY_INTERFACE_MODE_MAX enum value to guard the iteration in
of_get_phy_mode()
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/of/of_net.c | 26 ++------------------------
include/linux/phy.h | 42 ++++++++++++++++++++++++++++++++++++++++++
2 files changed, 44 insertions(+), 24 deletions(-)
diff --git a/drivers/of/of_net.c b/drivers/of/of_net.c
index a208a45..84215c1 100644
--- a/drivers/of/of_net.c
+++ b/drivers/of/of_net.c
@@ -12,28 +12,6 @@
#include <linux/export.h>
/**
- * It maps 'enum phy_interface_t' found in include/linux/phy.h
- * into the device tree binding of 'phy-mode', so that Ethernet
- * device driver can get phy interface from device tree.
- */
-static const char *phy_modes[] = {
- [PHY_INTERFACE_MODE_NA] = "",
- [PHY_INTERFACE_MODE_MII] = "mii",
- [PHY_INTERFACE_MODE_GMII] = "gmii",
- [PHY_INTERFACE_MODE_SGMII] = "sgmii",
- [PHY_INTERFACE_MODE_TBI] = "tbi",
- [PHY_INTERFACE_MODE_REVMII] = "rev-mii",
- [PHY_INTERFACE_MODE_RMII] = "rmii",
- [PHY_INTERFACE_MODE_RGMII] = "rgmii",
- [PHY_INTERFACE_MODE_RGMII_ID] = "rgmii-id",
- [PHY_INTERFACE_MODE_RGMII_RXID] = "rgmii-rxid",
- [PHY_INTERFACE_MODE_RGMII_TXID] = "rgmii-txid",
- [PHY_INTERFACE_MODE_RTBI] = "rtbi",
- [PHY_INTERFACE_MODE_SMII] = "smii",
- [PHY_INTERFACE_MODE_XGMII] = "xgmii",
-};
-
-/**
* of_get_phy_mode - Get phy mode for given device_node
* @np: Pointer to the given device_node
*
@@ -49,8 +27,8 @@ int of_get_phy_mode(struct device_node *np)
if (err < 0)
return err;
- for (i = 0; i < ARRAY_SIZE(phy_modes); i++)
- if (!strcasecmp(pm, phy_modes[i]))
+ for (i = 0; i < PHY_INTERFACE_MODE_MAX; i++)
+ if (!strcasecmp(pm, phy_modes(i)))
return i;
return -ENODEV;
diff --git a/include/linux/phy.h b/include/linux/phy.h
index eede657..ef7fa11 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -74,8 +74,50 @@ typedef enum {
PHY_INTERFACE_MODE_RTBI,
PHY_INTERFACE_MODE_SMII,
PHY_INTERFACE_MODE_XGMII,
+ PHY_INTERFACE_MODE_MAX,
} phy_interface_t;
+/**
+ * It maps 'enum phy_interface_t' found in include/linux/phy.h
+ * into the device tree binding of 'phy-mode', so that Ethernet
+ * device driver can get phy interface from device tree.
+ */
+static inline const char *phy_modes(phy_interface_t interface)
+{
+ switch (interface) {
+ case PHY_INTERFACE_MODE_NA:
+ return "";
+ case PHY_INTERFACE_MODE_MII:
+ return "mii";
+ case PHY_INTERFACE_MODE_GMII:
+ return "gmii";
+ case PHY_INTERFACE_MODE_SGMII:
+ return "sgmii";
+ case PHY_INTERFACE_MODE_TBI:
+ return "tbi";
+ case PHY_INTERFACE_MODE_REVMII:
+ return "rev-mii";
+ case PHY_INTERFACE_MODE_RMII:
+ return "rmii";
+ case PHY_INTERFACE_MODE_RGMII:
+ return "rgmii";
+ case PHY_INTERFACE_MODE_RGMII_ID:
+ return "rgmii-id";
+ case PHY_INTERFACE_MODE_RGMII_RXID:
+ return "rgmii-rxid";
+ case PHY_INTERFACE_MODE_RGMII_TXID:
+ return "rgmii-txid";
+ case PHY_INTERFACE_MODE_RTBI:
+ return "rtbi";
+ case PHY_INTERFACE_MODE_SMII:
+ return "smii";
+ case PHY_INTERFACE_MODE_XGMII:
+ return "xgmii";
+ default:
+ return "unknown";
+ }
+}
+
#define PHY_INIT_TIMEOUT 100000
#define PHY_STATE_TIME 1
--
1.8.3.2
^ permalink raw reply related
* [PATCH 09/10] net: phy: add "has_fixups" boolean property
From: Florian Fainelli @ 2014-02-12 1:27 UTC (permalink / raw)
To: netdev; +Cc: davem, Florian Fainelli
In-Reply-To: <1392168462-18888-1-git-send-email-f.fainelli@gmail.com>
Add a boolean property which indicates if the PHY has had any fixup
routine ran on it. We are later going to use that boolean to expose it
as a sysfs property to help troubleshooting.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/phy/phy_device.c | 1 +
include/linux/phy.h | 1 +
2 files changed, 2 insertions(+)
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 7c18420..c2d778d 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -139,6 +139,7 @@ static int phy_scan_fixups(struct phy_device *phydev)
mutex_unlock(&phy_fixup_lock);
return err;
}
+ phydev->has_fixups = true;
}
}
mutex_unlock(&phy_fixup_lock);
diff --git a/include/linux/phy.h b/include/linux/phy.h
index ef7fa11..42f1bc7 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -350,6 +350,7 @@ struct phy_device {
struct phy_c45_device_ids c45_ids;
bool is_c45;
bool is_internal;
+ bool has_fixups;
enum phy_state state;
--
1.8.3.2
^ permalink raw reply related
* [PATCH 05/10] net: phy: allow driver to implement their own aneg_done
From: Florian Fainelli @ 2014-02-12 1:27 UTC (permalink / raw)
To: netdev; +Cc: davem, Florian Fainelli
In-Reply-To: <1392168462-18888-1-git-send-email-f.fainelli@gmail.com>
Some PHYs out there can be very quirky with respect to how they would
report the auto-negotiation is completed. Allow drivers to override the
generic aneg_done() implementation by providing their own.
Since not all drivers have been updated yet to use genphy_aneg_done() as
aneg_done() callback, we explicitely check that this callback is valid
before calling into it.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/phy/phy.c | 9 ++++++---
drivers/net/phy/phy_device.c | 1 +
include/linux/phy.h | 3 +++
3 files changed, 10 insertions(+), 3 deletions(-)
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index db9c543..2fa4611 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -114,12 +114,15 @@ static int phy_config_interrupt(struct phy_device *phydev, u32 interrupts)
* phy_aneg_done - return auto-negotiation status
* @phydev: target phy_device struct
*
- * Description: Reads the status register and returns 0 either if
- * auto-negotiation is incomplete, or if there was an error.
- * Returns BMSR_ANEGCOMPLETE if auto-negotiation is done.
+ * Description: Return the auto-negotiation status from this @phydev
+ * Returns > 0 on success or < 0 on error. 0 means that auto-negotiation
+ * is still pending.
*/
static inline int phy_aneg_done(struct phy_device *phydev)
{
+ if (phydev->drv->aneg_done)
+ return phydev->drv->aneg_done(phydev);
+
return genphy_aneg_done(phydev);
}
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 4e7db72..7c18420 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -1267,6 +1267,7 @@ static struct phy_driver genphy_driver[] = {
.config_init = genphy_config_init,
.features = 0,
.config_aneg = genphy_config_aneg,
+ .aneg_done = genphy_aneg_done,
.read_status = genphy_read_status,
.suspend = genphy_suspend,
.resume = genphy_resume,
diff --git a/include/linux/phy.h b/include/linux/phy.h
index c572842..eede657 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -417,6 +417,9 @@ struct phy_driver {
*/
int (*config_aneg)(struct phy_device *phydev);
+ /* Determines the auto negotiation result */
+ int (*aneg_done)(struct phy_device *phydev);
+
/* Determines the negotiated speed and duplex */
int (*read_status)(struct phy_device *phydev);
--
1.8.3.2
^ permalink raw reply related
* [PATCH 06/10] net: phy: fix phy_{clear,config}_interrupt comment typos
From: Florian Fainelli @ 2014-02-12 1:27 UTC (permalink / raw)
To: netdev; +Cc: davem, Florian Fainelli
In-Reply-To: <1392168462-18888-1-git-send-email-f.fainelli@gmail.com>
The comments above phy_{clear,config}_interrupt used the word "on"
instead of "or", when talking about the return values of the functions,
fix these two typos.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/phy/phy.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index 2fa4611..fc918b6 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -83,7 +83,7 @@ EXPORT_SYMBOL(phy_print_status);
* If the @phydev driver has an ack_interrupt function, call it to
* ack and clear the phy device's interrupt.
*
- * Returns 0 on success on < 0 on error.
+ * Returns 0 on success or < 0 on error.
*/
static int phy_clear_interrupt(struct phy_device *phydev)
{
@@ -98,7 +98,7 @@ static int phy_clear_interrupt(struct phy_device *phydev)
* @phydev: the phy_device struct
* @interrupts: interrupt flags to configure for this @phydev
*
- * Returns 0 on success on < 0 on error.
+ * Returns 0 on success or < 0 on error.
*/
static int phy_config_interrupt(struct phy_device *phydev, u32 interrupts)
{
--
1.8.3.2
^ permalink raw reply related
* [PATCH 10/10] net: phy: expose phydev->has_fixups to sysfs
From: Florian Fainelli @ 2014-02-12 1:27 UTC (permalink / raw)
To: netdev; +Cc: davem, Florian Fainelli
In-Reply-To: <1392168462-18888-1-git-send-email-f.fainelli@gmail.com>
Expose the PHY device has_fixups boolean as a sysfs property to help
troubleshooting PHY configurations.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
Documentation/ABI/testing/sysfs-bus-mdio | 11 +++++++++++
drivers/net/phy/mdio_bus.c | 10 ++++++++++
2 files changed, 21 insertions(+)
diff --git a/Documentation/ABI/testing/sysfs-bus-mdio b/Documentation/ABI/testing/sysfs-bus-mdio
index 2133afd..fcab995 100644
--- a/Documentation/ABI/testing/sysfs-bus-mdio
+++ b/Documentation/ABI/testing/sysfs-bus-mdio
@@ -17,3 +17,14 @@ Description:
Ethernet driver during bus enumeration, encoded in string.
This interface mode is used to configure the Ethernet MAC with the
appropriate mode for its data lines to the PHY hardware.
+
+What: /sys/bus/mdio_bus/devices/.../phy_has_fixups
+Date: February 2014
+KernelVersion: 3.15
+Contact: netdev@vger.kernel.org
+Description:
+ This attribute contains the boolean value whether a given PHY
+ device has had any "fixup" workaround running on it, encoded as
+ a boolean. This information is provided to help troubleshooting
+ PHY configurations.
+
diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c
index 7c66ea0..76f54b3 100644
--- a/drivers/net/phy/mdio_bus.c
+++ b/drivers/net/phy/mdio_bus.c
@@ -441,9 +441,19 @@ phy_interface_show(struct device *dev, struct device_attribute *attr, char *buf)
}
static DEVICE_ATTR_RO(phy_interface);
+static ssize_t
+phy_has_fixups_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+ struct phy_device *phydev = to_phy_device(dev);
+
+ return sprintf(buf, "%d\n", phydev->has_fixups);
+}
+static DEVICE_ATTR_RO(phy_has_fixups);
+
static struct attribute *mdio_dev_attrs[] = {
&dev_attr_phy_id.attr,
&dev_attr_phy_interface.attr,
+ &dev_attr_phy_has_fixups.attr,
NULL,
};
ATTRIBUTE_GROUPS(mdio_dev);
--
1.8.3.2
^ permalink raw reply related
* [PATCH 00/10] net: phy: small cleanups and improvements
From: Florian Fainelli @ 2014-02-12 1:27 UTC (permalink / raw)
To: netdev; +Cc: davem, Florian Fainelli
Hi David,
This patchset contains some small cleanups which have been sent on the
mailing-list before to improve phy_print_status() and make it useful.
The second part of the patch allows PHY drivers to implement their own
aneg_done() callback.
Finally, two new sysfs properties are exposed to help troubleshoot libphy
aware Linux systems.
Thanks!
Florian Fainelli (10):
net: phy: use network device in phy_print_status
net: phy: update phy_print_status to show pause settings
net: phy: display human readable PHY speed settings
net: phy: add genphy_aneg_done()
net: phy: allow driver to implement their own aneg_done
net: phy: fix phy_{clear,config}_interrupt comment typos
net: phy: re-design phy_modes to be self-contained
net: phy: expose PHY device interface mode
net: phy: add "has_fixups" boolean property
net: phy: expose phydev->has_fixups to sysfs
Documentation/ABI/testing/sysfs-bus-mdio | 21 ++++++++++++++
drivers/net/phy/mdio_bus.c | 20 ++++++++++++++
drivers/net/phy/phy.c | 46 +++++++++++++++++++++++--------
drivers/net/phy/phy_device.c | 18 ++++++++++++
drivers/of/of_net.c | 26 ++----------------
include/linux/phy.h | 47 ++++++++++++++++++++++++++++++++
6 files changed, 142 insertions(+), 36 deletions(-)
--
1.8.3.2
^ permalink raw reply
* [PATCH 08/10] net: phy: expose PHY device interface mode
From: Florian Fainelli @ 2014-02-12 1:27 UTC (permalink / raw)
To: netdev; +Cc: davem, Florian Fainelli
In-Reply-To: <1392168462-18888-1-git-send-email-f.fainelli@gmail.com>
Expose the PHY device interface mode through sysfs since this is an
useful piece of information for knowing how the attached networking
device will have configured its transmit/receive path.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
Documentation/ABI/testing/sysfs-bus-mdio | 10 ++++++++++
drivers/net/phy/mdio_bus.c | 10 ++++++++++
2 files changed, 20 insertions(+)
diff --git a/Documentation/ABI/testing/sysfs-bus-mdio b/Documentation/ABI/testing/sysfs-bus-mdio
index 6349749..2133afd 100644
--- a/Documentation/ABI/testing/sysfs-bus-mdio
+++ b/Documentation/ABI/testing/sysfs-bus-mdio
@@ -7,3 +7,13 @@ Description:
by the device during bus enumeration, encoded in hexadecimal.
This ID is used to match the device with the appropriate
driver.
+
+What: /sys/bus/mdio_bus/devices/.../phy_interface
+Date: February 2014
+KernelVersion: 3.15
+Contact: netdev@vger.kernel.org
+Description:
+ This attribute contains the PHY interface as configured by the
+ Ethernet driver during bus enumeration, encoded in string.
+ This interface mode is used to configure the Ethernet MAC with the
+ appropriate mode for its data lines to the PHY hardware.
diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c
index 71e4900..7c66ea0 100644
--- a/drivers/net/phy/mdio_bus.c
+++ b/drivers/net/phy/mdio_bus.c
@@ -432,8 +432,18 @@ phy_id_show(struct device *dev, struct device_attribute *attr, char *buf)
}
static DEVICE_ATTR_RO(phy_id);
+static ssize_t
+phy_interface_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+ struct phy_device *phydev = to_phy_device(dev);
+
+ return sprintf(buf, "%s\n", phy_modes(phydev->interface));
+}
+static DEVICE_ATTR_RO(phy_interface);
+
static struct attribute *mdio_dev_attrs[] = {
&dev_attr_phy_id.attr,
+ &dev_attr_phy_interface.attr,
NULL,
};
ATTRIBUTE_GROUPS(mdio_dev);
--
1.8.3.2
^ permalink raw reply related
* [PATCH 04/10] net: phy: add genphy_aneg_done()
From: Florian Fainelli @ 2014-02-12 1:27 UTC (permalink / raw)
To: netdev; +Cc: davem, Florian Fainelli
In-Reply-To: <1392168462-18888-1-git-send-email-f.fainelli@gmail.com>
In preparation for allowing PHY drivers to potentially override their
auto-negotiation done callback, move the contents of phy_aneg_done() to
genphy_aneg_done() since that function really is the generic
implementation based on the BMSR_ANEGCOMPLETE status.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/phy/phy.c | 4 +---
drivers/net/phy/phy_device.c | 16 ++++++++++++++++
include/linux/phy.h | 1 +
3 files changed, 18 insertions(+), 3 deletions(-)
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index 36fc6e1..db9c543 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -120,9 +120,7 @@ static int phy_config_interrupt(struct phy_device *phydev, u32 interrupts)
*/
static inline int phy_aneg_done(struct phy_device *phydev)
{
- int retval = phy_read(phydev, MII_BMSR);
-
- return (retval < 0) ? retval : (retval & BMSR_ANEGCOMPLETE);
+ return genphy_aneg_done(phydev);
}
/* A structure for mapping a particular speed and duplex
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 82514e7..4e7db72 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -865,6 +865,22 @@ int genphy_config_aneg(struct phy_device *phydev)
}
EXPORT_SYMBOL(genphy_config_aneg);
+/**
+ * genphy_aneg_done - return auto-negotiation status
+ * @phydev: target phy_device struct
+ *
+ * Description: Reads the status register and returns 0 either if
+ * auto-negotiation is incomplete, or if there was an error.
+ * Returns BMSR_ANEGCOMPLETE if auto-negotiation is done.
+ */
+int genphy_aneg_done(struct phy_device *phydev)
+{
+ int retval = phy_read(phydev, MII_BMSR);
+
+ return (retval < 0) ? retval : (retval & BMSR_ANEGCOMPLETE);
+}
+EXPORT_SYMBOL(genphy_aneg_done);
+
static int gen10g_config_aneg(struct phy_device *phydev)
{
return 0;
diff --git a/include/linux/phy.h b/include/linux/phy.h
index 565188c..c572842 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -612,6 +612,7 @@ static inline int phy_read_status(struct phy_device *phydev)
int genphy_setup_forced(struct phy_device *phydev);
int genphy_restart_aneg(struct phy_device *phydev);
int genphy_config_aneg(struct phy_device *phydev);
+int genphy_aneg_done(struct phy_device *phydev);
int genphy_update_link(struct phy_device *phydev);
int genphy_read_status(struct phy_device *phydev);
int genphy_suspend(struct phy_device *phydev);
--
1.8.3.2
^ permalink raw reply related
* [PATCH net-next V2] intel: Remove unnecessary OOM messages
From: Joe Perches @ 2014-02-12 1:30 UTC (permalink / raw)
To: Jeff Kirsher; +Cc: e1000-devel, netdev
In-Reply-To: <1392168397.23721.2.camel@joe-AO722>
Don't emit these as there's a generic OOM with a dump_stack()
on allocation failures.
Signed-off-by: Joe Perches <joe@perches.com>
---
With patch attached this time...
drivers/net/ethernet/intel/i40e/i40e_ethtool.c | 5 +---
drivers/net/ethernet/intel/i40evf/i40evf_main.c | 20 ++++---------
.../net/ethernet/intel/i40evf/i40evf_virtchnl.c | 35 +++++++---------------
drivers/net/ethernet/intel/igb/igb_main.c | 2 --
4 files changed, 17 insertions(+), 45 deletions(-)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index b1d7d8c..d4d0761 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -1506,11 +1506,8 @@ static int i40e_add_del_fdir_ethtool(struct i40e_vsi *vsi,
*/
fd_data.raw_packet = kzalloc(I40E_FDIR_MAX_RAW_PACKET_LOOKUP,
GFP_KERNEL);
-
- if (!fd_data.raw_packet) {
- dev_info(&pf->pdev->dev, "Could not allocate memory\n");
+ if (!fd_data.raw_packet)
return -ENOMEM;
- }
fd_data.q_index = fsp->ring_cookie;
fd_data.flex_off = 0;
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index f5caf44..5331fcb 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -650,12 +650,9 @@ i40evf_vlan_filter *i40evf_add_vlan(struct i40evf_adapter *adapter, u16 vlan)
f = i40evf_find_vlan(adapter, vlan);
if (NULL == f) {
f = kzalloc(sizeof(*f), GFP_ATOMIC);
- if (NULL == f) {
- dev_info(&adapter->pdev->dev,
- "%s: no memory for new VLAN filter\n",
- __func__);
+ if (NULL == f)
return NULL;
- }
+
f->vlan = vlan;
INIT_LIST_HEAD(&f->list);
@@ -760,8 +757,6 @@ i40evf_mac_filter *i40evf_add_filter(struct i40evf_adapter *adapter,
if (NULL == f) {
f = kzalloc(sizeof(*f), GFP_ATOMIC);
if (NULL == f) {
- dev_info(&adapter->pdev->dev,
- "%s: no memory for new filter\n", __func__);
clear_bit(__I40EVF_IN_CRITICAL_TASK,
&adapter->crit_section);
return NULL;
@@ -1507,11 +1502,9 @@ static void i40evf_adminq_task(struct work_struct *work)
event.msg_size = I40EVF_MAX_AQ_BUF_SIZE;
event.msg_buf = kzalloc(event.msg_size, GFP_KERNEL);
- if (!event.msg_buf) {
- dev_info(&adapter->pdev->dev, "%s: no memory for ARQ clean\n",
- __func__);
+ if (!event.msg_buf)
return;
- }
+
v_msg = (struct i40e_virtchnl_msg *)&event.desc;
do {
ret = i40evf_clean_arq_element(hw, &event, &pending);
@@ -1902,11 +1895,8 @@ static void i40evf_init_task(struct work_struct *work)
(I40E_MAX_VF_VSI *
sizeof(struct i40e_virtchnl_vsi_resource));
adapter->vf_res = kzalloc(bufsz, GFP_KERNEL);
- if (!adapter->vf_res) {
- dev_err(&pdev->dev, "%s: unable to allocate memory\n",
- __func__);
+ if (!adapter->vf_res)
goto err;
- }
}
err = i40evf_get_vf_config(adapter);
if (err == I40E_ERR_ADMIN_QUEUE_NO_WORK)
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
index e6978d7..6412485 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
@@ -213,11 +213,9 @@ void i40evf_configure_queues(struct i40evf_adapter *adapter)
len = sizeof(struct i40e_virtchnl_vsi_queue_config_info) +
(sizeof(struct i40e_virtchnl_queue_pair_info) * pairs);
vqci = kzalloc(len, GFP_ATOMIC);
- if (!vqci) {
- dev_err(&adapter->pdev->dev, "%s: unable to allocate memory\n",
- __func__);
+ if (!vqci)
return;
- }
+
vqci->vsi_id = adapter->vsi_res->vsi_id;
vqci->num_queue_pairs = pairs;
vqpi = vqci->qpair;
@@ -326,11 +324,8 @@ void i40evf_map_queues(struct i40evf_adapter *adapter)
(adapter->num_msix_vectors *
sizeof(struct i40e_virtchnl_vector_map));
vimi = kzalloc(len, GFP_ATOMIC);
- if (!vimi) {
- dev_err(&adapter->pdev->dev, "%s: unable to allocate memory\n",
- __func__);
+ if (!vimi)
return;
- }
vimi->num_vectors = adapter->num_msix_vectors;
/* Queue vectors first */
@@ -396,11 +391,9 @@ void i40evf_add_ether_addrs(struct i40evf_adapter *adapter)
}
veal = kzalloc(len, GFP_ATOMIC);
- if (!veal) {
- dev_err(&adapter->pdev->dev, "%s: unable to allocate memory\n",
- __func__);
+ if (!veal)
return;
- }
+
veal->vsi_id = adapter->vsi_res->vsi_id;
veal->num_elements = count;
list_for_each_entry(f, &adapter->mac_filter_list, list) {
@@ -459,11 +452,9 @@ void i40evf_del_ether_addrs(struct i40evf_adapter *adapter)
len = I40EVF_MAX_AQ_BUF_SIZE;
}
veal = kzalloc(len, GFP_ATOMIC);
- if (!veal) {
- dev_err(&adapter->pdev->dev, "%s: unable to allocate memory\n",
- __func__);
+ if (!veal)
return;
- }
+
veal->vsi_id = adapter->vsi_res->vsi_id;
veal->num_elements = count;
list_for_each_entry_safe(f, ftmp, &adapter->mac_filter_list, list) {
@@ -523,11 +514,9 @@ void i40evf_add_vlans(struct i40evf_adapter *adapter)
len = I40EVF_MAX_AQ_BUF_SIZE;
}
vvfl = kzalloc(len, GFP_ATOMIC);
- if (!vvfl) {
- dev_err(&adapter->pdev->dev, "%s: unable to allocate memory\n",
- __func__);
+ if (!vvfl)
return;
- }
+
vvfl->vsi_id = adapter->vsi_res->vsi_id;
vvfl->num_elements = count;
list_for_each_entry(f, &adapter->vlan_filter_list, list) {
@@ -585,11 +574,9 @@ void i40evf_del_vlans(struct i40evf_adapter *adapter)
len = I40EVF_MAX_AQ_BUF_SIZE;
}
vvfl = kzalloc(len, GFP_ATOMIC);
- if (!vvfl) {
- dev_err(&adapter->pdev->dev, "%s: unable to allocate memory\n",
- __func__);
+ if (!vvfl)
return;
- }
+
vvfl->vsi_id = adapter->vsi_res->vsi_id;
vvfl->num_elements = count;
list_for_each_entry_safe(f, ftmp, &adapter->vlan_filter_list, list) {
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 46d31a4..235ffe4 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -2665,8 +2665,6 @@ static int igb_enable_sriov(struct pci_dev *pdev, int num_vfs)
/* if allocation failed then we do not support SR-IOV */
if (!adapter->vf_data) {
adapter->vfs_allocated_count = 0;
- dev_err(&pdev->dev,
- "Unable to allocate memory for VF Data Storage\n");
err = -ENOMEM;
goto out;
}
^ permalink raw reply related
* [PATCH v3 0/2] sctp: fix a problem with net_namespace
From: Wang Weidong @ 2014-02-12 1:44 UTC (permalink / raw)
To: nhorman, davem, vyasevich; +Cc: dborkman, sergei.shtylyov, netdev
fix a problem with net_namespace, and optimize
the sctp_sysctl_net_register.
v2 -> v3:
-patch2: add empty line after declaration as potined out by Sergei.
v1 -> v2:
-patch1: add Neil's ACK.
Wang Weidong (2):
sctp: fix a missed .data initialization
sctp: optimize the sctp_sysctl_net_register
net/sctp/sysctl.c | 18 +++++++++++-------
1 file changed, 11 insertions(+), 7 deletions(-)
--
1.7.12
^ permalink raw reply
* [PATCH v3 2/2] sctp: optimize the sctp_sysctl_net_register
From: Wang Weidong @ 2014-02-12 1:44 UTC (permalink / raw)
To: nhorman, davem, vyasevich; +Cc: dborkman, sergei.shtylyov, netdev
In-Reply-To: <1392169484-8256-1-git-send-email-wangweidong1@huawei.com>
Here, when the net is init_net, we needn't to kmemdup the ctl_table
again. So add a check for net. Also we can save some memory.
Signed-off-by: Wang Weidong <wangweidong1@huawei.com>
---
net/sctp/sysctl.c | 17 ++++++++++-------
1 file changed, 10 insertions(+), 7 deletions(-)
diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c
index d354de5..35c8923 100644
--- a/net/sctp/sysctl.c
+++ b/net/sctp/sysctl.c
@@ -402,15 +402,18 @@ static int proc_sctp_do_rto_max(struct ctl_table *ctl, int write,
int sctp_sysctl_net_register(struct net *net)
{
- struct ctl_table *table;
- int i;
+ struct ctl_table *table = sctp_net_table;
+
+ if (!net_eq(net, &init_net)) {
+ int i;
- table = kmemdup(sctp_net_table, sizeof(sctp_net_table), GFP_KERNEL);
- if (!table)
- return -ENOMEM;
+ table = kmemdup(sctp_net_table, sizeof(sctp_net_table), GFP_KERNEL);
+ if (!table)
+ return -ENOMEM;
- for (i = 0; table[i].data; i++)
- table[i].data += (char *)(&net->sctp) - (char *)&init_net.sctp;
+ for (i = 0; table[i].data; i++)
+ table[i].data += (char *)(&net->sctp) - (char *)&init_net.sctp;
+ }
net->sctp.sysctl_header = register_net_sysctl(net, "net/sctp", table);
return 0;
--
1.7.12
^ permalink raw reply related
* [PATCH v3 1/2] sctp: fix a missed .data initialization
From: Wang Weidong @ 2014-02-12 1:44 UTC (permalink / raw)
To: nhorman, davem, vyasevich; +Cc: dborkman, sergei.shtylyov, netdev
In-Reply-To: <1392169484-8256-1-git-send-email-wangweidong1@huawei.com>
As commit 3c68198e75111a90("sctp: Make hmac algorithm selection for
cookie generation dynamic"), we miss the .data initialization.
If we don't use the net_namespace, the problem that parts of the
sysctl configuration won't be isolation and won't occur.
In sctp_sysctl_net_register(), we register the sysctl for each
net, in the for(), we use the 'table[i].data' as check condition, so
when the 'i' is the index of sctp_hmac_alg, the data is NULL, then
break. So add the .data initialization.
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Wang Weidong <wangweidong1@huawei.com>
---
net/sctp/sysctl.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c
index 7135e61..d354de5 100644
--- a/net/sctp/sysctl.c
+++ b/net/sctp/sysctl.c
@@ -151,6 +151,7 @@ static struct ctl_table sctp_net_table[] = {
},
{
.procname = "cookie_hmac_alg",
+ .data = &init_net.sctp.sctp_hmac_alg,
.maxlen = 8,
.mode = 0644,
.proc_handler = proc_sctp_do_hmac_alg,
--
1.7.12
^ permalink raw reply related
* [PATCH] Documentation/networking: delete orphaned 3c505.txt file.
From: Paul Gortmaker @ 2014-02-12 1:52 UTC (permalink / raw)
To: netdev; +Cc: linux-doc, Paul Gortmaker
In the commit 0e245dbaac9fa1c2fd0f4e2af7b9f6d874083a8b
("drivers/net: delete the 3Com 3c505/3c507 intel i825xx support")
we clobbered the 3c505 driver (over a year ago) along with other
abandoned ISA drivers.
However, this orphaned README file escaped detection at that
time, and has lived on until today. Get rid of it now.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
---
[I did not mark this net vs. net-next; as it is a fix against
old work (in net) but at the same time harmless & OK delayed
to new work (net-next) and those new commits -- so in the end,
please feel free to put it where you think it fits easiest.]
Documentation/networking/3c505.txt | 45 --------------------------------------
1 file changed, 45 deletions(-)
delete mode 100644 Documentation/networking/3c505.txt
diff --git a/Documentation/networking/3c505.txt b/Documentation/networking/3c505.txt
deleted file mode 100644
index 72f38b13101d..000000000000
--- a/Documentation/networking/3c505.txt
+++ /dev/null
@@ -1,45 +0,0 @@
-The 3Com Etherlink Plus (3c505) driver.
-
-This driver now uses DMA. There is currently no support for PIO operation.
-The default DMA channel is 6; this is _not_ autoprobed, so you must
-make sure you configure it correctly. If loading the driver as a
-module, you can do this with "modprobe 3c505 dma=n". If the driver is
-linked statically into the kernel, you must either use an "ether="
-statement on the command line, or change the definition of ELP_DMA in 3c505.h.
-
-The driver will warn you if it has to fall back on the compiled in
-default DMA channel.
-
-If no base address is given at boot time, the driver will autoprobe
-ports 0x300, 0x280 and 0x310 (in that order). If no IRQ is given, the driver
-will try to probe for it.
-
-The driver can be used as a loadable module.
-
-Theoretically, one instance of the driver can now run multiple cards,
-in the standard way (when loading a module, say "modprobe 3c505
-io=0x300,0x340 irq=10,11 dma=6,7" or whatever). I have not tested
-this, though.
-
-The driver may now support revision 2 hardware; the dependency on
-being able to read the host control register has been removed. This
-is also untested, since I don't have a suitable card.
-
-Known problems:
- I still see "DMA upload timed out" messages from time to time. These
-seem to be fairly non-fatal though.
- The card is old and slow.
-
-To do:
- Improve probe/setup code
- Test multicast and promiscuous operation
-
-Authors:
- The driver is mainly written by Craig Southeren, email
- <craigs@ineluki.apana.org.au>.
- Parts of the driver (adapting the driver to 1.1.4+ kernels,
- IRQ/address detection, some changes) and this README by
- Juha Laiho <jlaiho@ichaos.nullnet.fi>.
- DMA mode, more fixes, etc, by Philip Blundell <pjb27@cam.ac.uk>
- Multicard support, Software configurable DMA, etc., by
- Christopher Collins <ccollins@pcug.org.au>
--
1.8.5.2
^ permalink raw reply related
* Re: 3.14-mw regression: rtl8169 WARNING: DMA-API: exceeded 7 overlapping mappings of pfn 55ebe
From: Dan Williams @ 2014-02-12 2:07 UTC (permalink / raw)
To: Sander Eikelenboom
Cc: Konrad Rzeszutek Wilk, Wei Liu, Francois Romieu,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org, Dave Jones
In-Reply-To: <1434499348.20140211205617@eikelenboom.it>
[-- Attachment #1: Type: text/plain, Size: 9796 bytes --]
On Tue, Feb 11, 2014 at 11:56 AM, Sander Eikelenboom
<linux@eikelenboom.it> wrote:
> Hi Dan,
>
> FYI just tested and put Xen out of the equation (booting baremetal) and it still persists.
>
> I tried something else .. don't know if it gives you anymore insights, but it's worth the try:
This is great! See below:
>
> diff --git a/lib/dma-debug.c b/lib/dma-debug.c
> index 2defd13..0fe5b75 100644
> --- a/lib/dma-debug.c
> +++ b/lib/dma-debug.c
> @@ -474,11 +474,11 @@ static int active_pfn_set_overlap(unsigned long pfn, int overlap)
> return overlap;
> }
>
> -static void active_pfn_inc_overlap(unsigned long pfn)
> +static void active_pfn_inc_overlap(struct dma_debug_entry *ent)
> {
> - int overlap = active_pfn_read_overlap(pfn);
> + int overlap = active_pfn_read_overlap(ent->pfn);
>
> - overlap = active_pfn_set_overlap(pfn, ++overlap);
> + overlap = active_pfn_set_overlap(ent->pfn, ++overlap);
>
> /* If we overflowed the overlap counter then we're potentially
> * leaking dma-mappings. Otherwise, if maps and unmaps are
> @@ -486,15 +486,43 @@ static void active_pfn_inc_overlap(unsigned long pfn)
> * debug_dma_assert_idle() as the pfn may be marked idle
> * prematurely.
> */
> +
> WARN_ONCE(overlap > ACTIVE_PFN_MAX_OVERLAP,
> "DMA-API: exceeded %d overlapping mappings of pfn %lx\n",
> - ACTIVE_PFN_MAX_OVERLAP, pfn);
> + ACTIVE_PFN_MAX_OVERLAP, ent->pfn);
> +
> + if(overlap > ACTIVE_PFN_MAX_OVERLAP){
> +
> + dev_info(ent->dev, "DMA-API: exceeded %d overlapping mappings of pfn %lx .. start dump\n", ACTIVE_PFN_MAX_OVERLAP, ent->pfn);
> + int idx;
> +
> + for (idx = 0; idx < HASH_SIZE; idx++) {
> + struct hash_bucket *bucket = &dma_entry_hash[idx];
> + struct dma_debug_entry *entry;
> + unsigned long flags;
> +
> + list_for_each_entry(entry, &bucket->list, list) {
> + if (entry->pfn == ent->pfn) {
> + dev_info(entry->dev, "%s idx %d P=%Lx N=%lx D=%Lx L=%Lx %s %s\n",
> + type2name[entry->type], idx,
> + phys_addr(entry), entry->pfn,
> + entry->dev_addr, entry->size,
> + dir2name[entry->direction],
> + maperr2str[entry->map_err_type]);
> + }
> + }
> + }
> + dev_info(ent->dev, "DMA-API: exceeded %d overlapping mappings of pfn %lx .. end of dump\n", ACTIVE_PFN_MAX_OVERLAP, ent->pfn);
> + }
> }
>
>
> @@ -505,10 +533,10 @@ static int active_pfn_insert(struct dma_debug_entry *entry)
>
> spin_lock_irqsave(&radix_lock, flags);
> rc = radix_tree_insert(&dma_active_pfn, entry->pfn, entry);
> - if (rc == -EEXIST)
> - active_pfn_inc_overlap(entry->pfn);
> + if (rc == -EEXIST){
> + active_pfn_inc_overlap(entry);
> + }
> spin_unlock_irqrestore(&radix_lock, flags);
> -
> return rc;
> }
>
>
> This results in:
> [ 27.708678] r8169 0000:0a:00.0 eth1: link down
> [ 27.712102] r8169 0000:0a:00.0 eth1: link down
> [ 28.015340] r8169 0000:0b:00.0 eth0: link down
> [ 28.015368] r8169 0000:0b:00.0 eth0: link down
> [ 29.654844] r8169 0000:0b:00.0 eth0: link up
> [ 30.278542] r8169 0000:0a:00.0 eth1: link up
> [ 60.829503] EXT4-fs (dm-2): mounted filesystem with ordered data mode. Opts: barrier=1,errors=remount-ro
> [ 69.708979] EXT4-fs (dm-42): mounted filesystem with ordered data mode. Opts: barrier=1,errors=remount-ro
> [ 76.128678] EXT4-fs (dm-43): mounted filesystem with ordered data mode. Opts: barrier=1,errors=remount-ro
> [ 82.922836] EXT4-fs (dm-44): mounted filesystem with ordered data mode. Opts: barrier=1,errors=remount-ro
> [ 89.232889] EXT4-fs (dm-45): mounted filesystem with ordered data mode. Opts: barrier=1,errors=remount-ro
> [ 95.359859] EXT4-fs (dm-46): mounted filesystem with ordered data mode. Opts: barrier=1,errors=remount-ro
> [ 101.638559] EXT4-fs (sdb1): mounted filesystem with ordered data mode. Opts: barrier=1,errors=remount-ro
> [ 218.073407] ------------[ cut here ]------------
> [ 218.080983] WARNING: CPU: 5 PID: 0 at lib/dma-debug.c:492 add_dma_entry+0xf1/0x210()
> [ 218.088550] DMA-API: exceeded 7 overlapping mappings of pfn 3c421
> [ 218.095988] Modules linked in:
> [ 218.103270] CPU: 5 PID: 0 Comm: swapper/5 Tainted: G W 3.14.0-rc2-20140211-pcireset-net-btrevert-xenblock-dmadebug5+ #1
> [ 218.110712] Hardware name: MSI MS-7640/890FXA-GD70 (MS-7640) , BIOS V1.8B1 09/13/2010
> [ 218.118134] 0000000000000009 ffff88003fd437b8 ffffffff81b809c4 ffff88003e308000
> [ 218.125556] ffff88003fd43808 ffff88003fd437f8 ffffffff810c985c 0000000000000000
> [ 218.132917] 00000000ffffffef 0000000000000036 ffff88003d9d3c00 0000000000000282
> [ 218.140154] Call Trace:
> [ 218.147193] <IRQ> [<ffffffff81b809c4>] dump_stack+0x46/0x58
> [ 218.154271] [<ffffffff810c985c>] warn_slowpath_common+0x8c/0xc0
> [ 218.161293] [<ffffffff810c9946>] warn_slowpath_fmt+0x46/0x50
> [ 218.168227] [<ffffffff814f2cfa>] ? active_pfn_read_overlap+0x3a/0x70
> [ 218.175116] [<ffffffff814f41d1>] add_dma_entry+0xf1/0x210
> [ 218.181865] [<ffffffff814f4646>] debug_dma_map_page+0x126/0x150
> [ 218.188484] [<ffffffff817aabeb>] rtl8169_start_xmit+0x21b/0xa20
> [ 218.195042] [<ffffffff81a01877>] ? dev_queue_xmit_nit+0x1d7/0x260
> [ 218.201553] [<ffffffff81a0188f>] ? dev_queue_xmit_nit+0x1ef/0x260
> [ 218.207965] [<ffffffff81a016a5>] ? dev_queue_xmit_nit+0x5/0x260
> [ 218.214290] [<ffffffff81a0661f>] dev_hard_start_xmit+0x37f/0x590
> [ 218.220481] [<ffffffff81a26cae>] sch_direct_xmit+0xfe/0x280
> [ 218.226529] [<ffffffff81a06a7f>] __dev_queue_xmit+0x24f/0x660
> [ 218.232521] [<ffffffff81a06835>] ? __dev_queue_xmit+0x5/0x660
> [ 218.238439] [<ffffffff81ab21b9>] ? ip_output+0x59/0xf0
> [ 218.244272] [<ffffffff81a06eb0>] dev_queue_xmit+0x10/0x20
> [ 218.250043] [<ffffffff81ab076b>] ip_finish_output+0x2cb/0x670
> [ 218.255682] [<ffffffff81ab21b9>] ? ip_output+0x59/0xf0
> [ 218.261168] [<ffffffff81ab21b9>] ip_output+0x59/0xf0
> [ 218.266559] [<ffffffff81aad596>] ip_forward_finish+0x76/0x1a0
> [ 218.271883] [<ffffffff81aad86b>] ip_forward+0x1ab/0x440
> [ 218.277148] [<ffffffff81aab380>] ip_rcv_finish+0x150/0x660
> [ 218.282373] [<ffffffff81aabe3b>] ip_rcv+0x22b/0x370
> [ 218.287436] [<ffffffff81b09bc7>] ? packet_rcv_spkt+0x47/0x190
> [ 218.292372] [<ffffffff81a03272>] __netif_receive_skb_core+0x722/0x8f0
> [ 218.297328] [<ffffffff81a02c75>] ? __netif_receive_skb_core+0x125/0x8f0
> [ 218.302304] [<ffffffff8112ce6e>] ? getnstimeofday+0xe/0x30
> [ 218.307296] [<ffffffff819f42c5>] ? __netdev_alloc_frag+0x175/0x1b0
> [ 218.312166] [<ffffffff81a03461>] __netif_receive_skb+0x21/0x70
> [ 218.316904] [<ffffffff81a034d3>] netif_receive_skb_internal+0x23/0xf0
> [ 218.321596] [<ffffffff81a04d2d>] napi_gro_receive+0x8d/0x100
> [ 218.326219] [<ffffffff817a7bc3>] rtl8169_poll+0x2d3/0x680
> [ 218.330754] [<ffffffff8112e366>] ? update_wall_time+0x356/0x690
> [ 218.335208] [<ffffffff81a03a0a>] net_rx_action+0x18a/0x2c0
> [ 218.339595] [<ffffffff810ce6f1>] ? __do_softirq+0xc1/0x300
> [ 218.343890] [<ffffffff810ce767>] __do_softirq+0x137/0x300
> [ 218.348085] [<ffffffff810cec9a>] irq_exit+0xaa/0xd0
> [ 218.352203] [<ffffffff81b8e5a7>] do_IRQ+0x67/0x110
> [ 218.356225] [<ffffffff81b8b772>] common_interrupt+0x72/0x72
> [ 218.360156] <EOI> [<ffffffff810536e6>] ? native_safe_halt+0x6/0x10
> [ 218.364087] [<ffffffff81113a7d>] ? trace_hardirqs_on+0xd/0x10
> [ 218.367935] [<ffffffff81020632>] default_idle+0x32/0xd0
> [ 218.371691] [<ffffffff8102071e>] amd_e400_idle+0x4e/0x140
> [ 218.375360] [<ffffffff81020f86>] arch_cpu_idle+0x36/0x40
> [ 218.378921] [<ffffffff81120a01>] cpu_startup_entry+0xa1/0x2a0
> [ 218.382508] [<ffffffff810473cf>] start_secondary+0x1af/0x210
> [ 218.386133] ---[ end trace 0e12f271209e2c18 ]---
> [ 218.389769] r8169 0000:0b:00.0: DMA-API: exceeded 7 overlapping mappings of pfn 3c421 .. start dump
> [ 218.393566] r8169 0000:0b:00.0: single idx 563 P=3c421100 N=3c421 D=c66100 L=36 DMA_TO_DEVICE dma map error checked
> [ 218.397379] r8169 0000:0b:00.0: single idx 563 P=3c4212c0 N=3c421 D=c672c0 L=36 DMA_TO_DEVICE dma map error checked
> [ 218.401094] r8169 0000:0b:00.0: single idx 564 P=3c421480 N=3c421 D=c68480 L=36 DMA_TO_DEVICE dma map error checked
> [ 218.404730] r8169 0000:0b:00.0: single idx 564 P=3c421640 N=3c421 D=c69640 L=36 DMA_TO_DEVICE dma map error checked
> [ 218.408310] r8169 0000:0b:00.0: single idx 565 P=3c421800 N=3c421 D=c6a800 L=36 DMA_TO_DEVICE dma map error checked
> [ 218.411762] r8169 0000:0b:00.0: single idx 565 P=3c4219c0 N=3c421 D=c6b9c0 L=36 DMA_TO_DEVICE dma map error checked
> [ 218.415075] r8169 0000:0b:00.0: single idx 566 P=3c421b80 N=3c421 D=c6cb80 L=9b DMA_TO_DEVICE dma map error checked
> [ 218.418305] r8169 0000:0b:00.0: single idx 566 P=3c421dc0 N=3c421 D=c6ddc0 L=36 DMA_TO_DEVICE dma map error checked
> [ 218.421502] r8169 0000:0b:00.0: single idx 567 P=3c421f80 N=3c421 D=c6ef80 L=36 DMA_TO_DEVICE dma map error not checked
The overlap granularity is too large. Multiple dma_map_single
mappings are allowed to a given page as long as they don't collide on
the same cache line.
Please try the attached patch to see if it fixes this issue. Works ok for me.
[-- Attachment #2: fix-dma-debug-overlap.patch --]
[-- Type: text/x-patch, Size: 9843 bytes --]
dma debug: account for cachelines in overlap tracking
From: Dan Williams <dan.j.williams@intel.com>
While debug_dma_assert_idle() checks if a given *page* is actively
undergoing dma the valid granularity of a dma mapping is a *cacheline*.
Sander's testing shows that the warning message "DMA-API: exceeded 7
overlapping mappings of pfn..." is falsely triggering. The test is
simply mapping multiple cachelines in a given page.
Ultimately we want overlap tracking to be valid as it is a real api
violation, to that end we need to track active mappings by cachelines.
To this end, update the active dma tracking to use the
page-frame-relative cacheline of the mapping as the key, and update
debug_dma_assert_idle() to check for all possible mapped cachelines for
a given page.
Note, the radix gang lookup is suboptimal. It would be best if it
stopped fetching entries once the search passed a page boundary.
Nevertheless, this implementation does not perturb the original net_dma
failing case. That is to say the extra overhead does not show up in
terms of making the failing case pass due to a timing change.
Reported-by: Sander Eikelenboom <linux@eikelenboom.it>
Reported-by: Dave Jones <davej@redhat.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Francois Romieu <romieu@fr.zoreil.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Wei Liu <wei.liu2@citrix.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
lib/dma-debug.c | 112 +++++++++++++++++++++++++++++++++----------------------
1 files changed, 67 insertions(+), 45 deletions(-)
diff --git a/lib/dma-debug.c b/lib/dma-debug.c
index 2defd1308b04..42b12740940b 100644
--- a/lib/dma-debug.c
+++ b/lib/dma-debug.c
@@ -424,111 +424,121 @@ void debug_dma_dump_mappings(struct device *dev)
EXPORT_SYMBOL(debug_dma_dump_mappings);
/*
- * For each page mapped (initial page in the case of
- * dma_alloc_coherent/dma_map_{single|page}, or each page in a
- * scatterlist) insert into this tree using the pfn as the key. At
+ * For each mapping (initial cacheline in the case of
+ * dma_alloc_coherent/dma_map_page, initial cacheline in each page of a
+ * scatterlist, or the cacheline specified in dma_map_single) insert
+ * into this tree using the cacheline as the key. At
* dma_unmap_{single|sg|page} or dma_free_coherent delete the entry. If
- * the pfn already exists at insertion time add a tag as a reference
+ * the entry already exists at insertion time add a tag as a reference
* count for the overlapping mappings. For now, the overlap tracking
- * just ensures that 'unmaps' balance 'maps' before marking the pfn
- * idle, but we should also be flagging overlaps as an API violation.
+ * just ensures that 'unmaps' balance 'maps' before marking the
+ * cacheline idle, but we should also be flagging overlaps as an API
+ * violation.
*
* Memory usage is mostly constrained by the maximum number of available
* dma-debug entries in that we need a free dma_debug_entry before
- * inserting into the tree. In the case of dma_map_{single|page} and
- * dma_alloc_coherent there is only one dma_debug_entry and one pfn to
- * track per event. dma_map_sg(), on the other hand,
- * consumes a single dma_debug_entry, but inserts 'nents' entries into
- * the tree.
+ * inserting into the tree. In the case of dma_map_page and
+ * dma_alloc_coherent there is only one dma_debug_entry and one
+ * dma_active_cacheline entry to track per event. dma_map_sg(), on the
+ * other hand, consumes a single dma_debug_entry, but inserts 'nents'
+ * entries into the tree.
*
* At any time debug_dma_assert_idle() can be called to trigger a
- * warning if the given page is in the active set.
+ * warning if any cachelines in the given page are in the active set.
*/
-static RADIX_TREE(dma_active_pfn, GFP_NOWAIT);
+static RADIX_TREE(dma_active_cacheline, GFP_NOWAIT);
static DEFINE_SPINLOCK(radix_lock);
-#define ACTIVE_PFN_MAX_OVERLAP ((1 << RADIX_TREE_MAX_TAGS) - 1)
+#define ACTIVE_CLN_MAX_OVERLAP ((1 << RADIX_TREE_MAX_TAGS) - 1)
+#define CACHELINE_PER_PAGE_SHIFT (PAGE_SHIFT - L1_CACHE_SHIFT)
+#define CACHELINES_PER_PAGE (1 << CACHELINE_PER_PAGE_SHIFT)
-static int active_pfn_read_overlap(unsigned long pfn)
+unsigned long to_cln(struct dma_debug_entry *entry)
+{
+ return (entry->pfn << CACHELINE_PER_PAGE_SHIFT) +
+ (entry->offset >> L1_CACHE_SHIFT);
+}
+
+static int active_cln_read_overlap(unsigned long cln)
{
int overlap = 0, i;
for (i = RADIX_TREE_MAX_TAGS - 1; i >= 0; i--)
- if (radix_tree_tag_get(&dma_active_pfn, pfn, i))
+ if (radix_tree_tag_get(&dma_active_cacheline, cln, i))
overlap |= 1 << i;
return overlap;
}
-static int active_pfn_set_overlap(unsigned long pfn, int overlap)
+static int active_cln_set_overlap(unsigned long cln, int overlap)
{
int i;
- if (overlap > ACTIVE_PFN_MAX_OVERLAP || overlap < 0)
+ if (overlap > ACTIVE_CLN_MAX_OVERLAP || overlap < 0)
return overlap;
for (i = RADIX_TREE_MAX_TAGS - 1; i >= 0; i--)
if (overlap & 1 << i)
- radix_tree_tag_set(&dma_active_pfn, pfn, i);
+ radix_tree_tag_set(&dma_active_cacheline, cln, i);
else
- radix_tree_tag_clear(&dma_active_pfn, pfn, i);
+ radix_tree_tag_clear(&dma_active_cacheline, cln, i);
return overlap;
}
-static void active_pfn_inc_overlap(unsigned long pfn)
+static void active_cln_inc_overlap(unsigned long cln)
{
- int overlap = active_pfn_read_overlap(pfn);
+ int overlap = active_cln_read_overlap(cln);
- overlap = active_pfn_set_overlap(pfn, ++overlap);
+ overlap = active_cln_set_overlap(cln, ++overlap);
/* If we overflowed the overlap counter then we're potentially
* leaking dma-mappings. Otherwise, if maps and unmaps are
* balanced then this overflow may cause false negatives in
- * debug_dma_assert_idle() as the pfn may be marked idle
+ * debug_dma_assert_idle() as the cln may be marked idle
* prematurely.
*/
- WARN_ONCE(overlap > ACTIVE_PFN_MAX_OVERLAP,
- "DMA-API: exceeded %d overlapping mappings of pfn %lx\n",
- ACTIVE_PFN_MAX_OVERLAP, pfn);
+ WARN_ONCE(overlap > ACTIVE_CLN_MAX_OVERLAP,
+ "DMA-API: exceeded %d overlapping mappings of cln %lx\n",
+ ACTIVE_CLN_MAX_OVERLAP, cln);
}
-static int active_pfn_dec_overlap(unsigned long pfn)
+static int active_cln_dec_overlap(unsigned long cln)
{
- int overlap = active_pfn_read_overlap(pfn);
+ int overlap = active_cln_read_overlap(cln);
- return active_pfn_set_overlap(pfn, --overlap);
+ return active_cln_set_overlap(cln, --overlap);
}
-static int active_pfn_insert(struct dma_debug_entry *entry)
+static int active_cln_insert(struct dma_debug_entry *entry)
{
unsigned long flags;
int rc;
spin_lock_irqsave(&radix_lock, flags);
- rc = radix_tree_insert(&dma_active_pfn, entry->pfn, entry);
+ rc = radix_tree_insert(&dma_active_cacheline, to_cln(entry), entry);
if (rc == -EEXIST)
- active_pfn_inc_overlap(entry->pfn);
+ active_cln_inc_overlap(to_cln(entry));
spin_unlock_irqrestore(&radix_lock, flags);
return rc;
}
-static void active_pfn_remove(struct dma_debug_entry *entry)
+static void active_cln_remove(struct dma_debug_entry *entry)
{
unsigned long flags;
spin_lock_irqsave(&radix_lock, flags);
/* since we are counting overlaps the final put of the
- * entry->pfn will occur when the overlap count is 0.
- * active_pfn_dec_overlap() returns -1 in that case
+ * cacheline will occur when the overlap count is 0.
+ * active_cln_dec_overlap() returns -1 in that case
*/
- if (active_pfn_dec_overlap(entry->pfn) < 0)
- radix_tree_delete(&dma_active_pfn, entry->pfn);
+ if (active_cln_dec_overlap(to_cln(entry)) < 0)
+ radix_tree_delete(&dma_active_cacheline, to_cln(entry));
spin_unlock_irqrestore(&radix_lock, flags);
}
/**
* debug_dma_assert_idle() - assert that a page is not undergoing dma
- * @page: page to lookup in the dma_active_pfn tree
+ * @page: page to lookup in the dma_active_cacheline tree
*
* Place a call to this routine in cases where the cpu touching the page
* before the dma completes (page is dma_unmapped) will lead to data
@@ -536,14 +546,26 @@ static void active_pfn_remove(struct dma_debug_entry *entry)
*/
void debug_dma_assert_idle(struct page *page)
{
+ unsigned long cln = page_to_pfn(page) << CACHELINE_PER_PAGE_SHIFT;
+ static struct dma_debug_entry *ents[CACHELINES_PER_PAGE];
+ struct dma_debug_entry *entry = NULL;
+ void **results = (void **) &ents;
+ unsigned int nents, i;
unsigned long flags;
- struct dma_debug_entry *entry;
if (!page)
return;
spin_lock_irqsave(&radix_lock, flags);
- entry = radix_tree_lookup(&dma_active_pfn, page_to_pfn(page));
+ nents = radix_tree_gang_lookup(&dma_active_cacheline, results, cln,
+ CACHELINES_PER_PAGE);
+ for (i = 0; i < nents; i++) {
+ if (to_cln(ents[i]) == cln) {
+ entry = ents[i];
+ break;
+ } else if (to_cln(ents[i]) >= cln + CACHELINES_PER_PAGE)
+ break;
+ }
spin_unlock_irqrestore(&radix_lock, flags);
if (!entry)
@@ -551,7 +573,7 @@ void debug_dma_assert_idle(struct page *page)
err_printk(entry->dev, entry,
"DMA-API: cpu touching an active dma mapped page "
- "[pfn=0x%lx]\n", entry->pfn);
+ "[cln=0x%lx]\n", to_cln(entry));
}
/*
@@ -568,9 +590,9 @@ static void add_dma_entry(struct dma_debug_entry *entry)
hash_bucket_add(bucket, entry);
put_hash_bucket(bucket, &flags);
- rc = active_pfn_insert(entry);
+ rc = active_cln_insert(entry);
if (rc == -ENOMEM) {
- pr_err("DMA-API: pfn tracking ENOMEM, dma-debug disabled\n");
+ pr_err("DMA-API: cacheline tracking ENOMEM, dma-debug disabled\n");
global_disable = true;
}
@@ -631,7 +653,7 @@ static void dma_entry_free(struct dma_debug_entry *entry)
{
unsigned long flags;
- active_pfn_remove(entry);
+ active_cln_remove(entry);
/*
* add to beginning of the list - this way the entries are
^ permalink raw reply related
* [PATCH net-next 1/4] sch_netem: return errcode before setting params
From: Yang Yingliang @ 2014-02-12 2:58 UTC (permalink / raw)
To: netdev; +Cc: davem, stephen
In-Reply-To: <1392173895-5012-1-git-send-email-yangyingliang@huawei.com>
get_dist_table() and get_loss_clg() may be failed. These
two functions should be called after setting the members
of qdisc_priv(sch), or it will break the old settings while
either of them is failed.
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
---
net/sched/sch_netem.c | 39 +++++++++++++++++++++++++++++----------
1 file changed, 29 insertions(+), 10 deletions(-)
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index de1059a..b341943 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -821,6 +821,8 @@ static int netem_change(struct Qdisc *sch, struct nlattr *opt)
struct netem_sched_data *q = qdisc_priv(sch);
struct nlattr *tb[TCA_NETEM_MAX + 1];
struct tc_netem_qopt *qopt;
+ struct clgstate old_clg;
+ int old_loss_model = CLG_RANDOM;
int ret;
if (opt == NULL)
@@ -831,6 +833,33 @@ static int netem_change(struct Qdisc *sch, struct nlattr *opt)
if (ret < 0)
return ret;
+ /* backup q->clg and q->loss_model */
+ old_clg = q->clg;
+ old_loss_model = q->loss_model;
+
+ if (tb[TCA_NETEM_LOSS]) {
+ ret = get_loss_clg(sch, tb[TCA_NETEM_LOSS]);
+ if (ret) {
+ q->loss_model = old_loss_model;
+ return ret;
+ }
+ } else {
+ q->loss_model = CLG_RANDOM;
+ }
+
+ if (tb[TCA_NETEM_DELAY_DIST]) {
+ ret = get_dist_table(sch, tb[TCA_NETEM_DELAY_DIST]);
+ if (ret) {
+ /* recover clg and loss_model, in case of
+ * q->clg and q->loss_model were modified
+ * in get_loss_clg()
+ */
+ q->clg = old_clg;
+ q->loss_model = old_loss_model;
+ return ret;
+ }
+ }
+
sch->limit = qopt->limit;
q->latency = qopt->latency;
@@ -850,12 +879,6 @@ static int netem_change(struct Qdisc *sch, struct nlattr *opt)
if (tb[TCA_NETEM_CORR])
get_correlation(sch, tb[TCA_NETEM_CORR]);
- if (tb[TCA_NETEM_DELAY_DIST]) {
- ret = get_dist_table(sch, tb[TCA_NETEM_DELAY_DIST]);
- if (ret)
- return ret;
- }
-
if (tb[TCA_NETEM_REORDER])
get_reorder(sch, tb[TCA_NETEM_REORDER]);
@@ -872,10 +895,6 @@ static int netem_change(struct Qdisc *sch, struct nlattr *opt)
if (tb[TCA_NETEM_ECN])
q->ecn = nla_get_u32(tb[TCA_NETEM_ECN]);
- q->loss_model = CLG_RANDOM;
- if (tb[TCA_NETEM_LOSS])
- ret = get_loss_clg(sch, tb[TCA_NETEM_LOSS]);
-
return ret;
}
--
1.8.0
^ permalink raw reply related
* [PATCH net-next 3/4] sch_netem: replace magic numbers with enumerate in GE model
From: Yang Yingliang @ 2014-02-12 2:58 UTC (permalink / raw)
To: netdev; +Cc: davem, stephen
In-Reply-To: <1392173895-5012-1-git-send-email-yangyingliang@huawei.com>
Replace some magic numbers which describe states of GE model
loss generator with enumerate.
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
---
net/sched/sch_netem.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 4a5eb28..4fced67 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -117,6 +117,11 @@ struct netem_sched_data {
LOST_IN_BURST_PERIOD,
} _4_state_model;
+ enum {
+ GOOD_STATE = 1,
+ BAD_STATE,
+ } GE_state_model;
+
/* Correlated Loss Generation models */
struct clgstate {
/* state of the Markov chain */
@@ -272,15 +277,15 @@ static bool loss_gilb_ell(struct netem_sched_data *q)
struct clgstate *clg = &q->clg;
switch (clg->state) {
- case 1:
+ case GOOD_STATE:
if (prandom_u32() < clg->a1)
- clg->state = 2;
+ clg->state = BAD_STATE;
if (prandom_u32() < clg->a4)
return true;
break;
- case 2:
+ case BAD_STATE:
if (prandom_u32() < clg->a2)
- clg->state = 1;
+ clg->state = GOOD_STATE;
if (prandom_u32() > clg->a3)
return true;
}
--
1.8.0
^ permalink raw reply related
* [PATCH net-next 4/4] sch_netem: replace spin_(un)lock_bh with sch_tree_(un)lock
From: Yang Yingliang @ 2014-02-12 2:58 UTC (permalink / raw)
To: netdev; +Cc: davem, stephen
In-Reply-To: <1392173895-5012-1-git-send-email-yangyingliang@huawei.com>
spin_(un)lock_bh(root_lock) is same as sch_tree_(un)lock.
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
---
net/sched/sch_netem.c | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 4fced67..62811d5 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -665,7 +665,6 @@ static int get_dist_table(struct Qdisc *sch, const struct nlattr *attr)
struct netem_sched_data *q = qdisc_priv(sch);
size_t n = nla_len(attr)/sizeof(__s16);
const __s16 *data = nla_data(attr);
- spinlock_t *root_lock;
struct disttable *d;
int i;
size_t s;
@@ -684,11 +683,9 @@ static int get_dist_table(struct Qdisc *sch, const struct nlattr *attr)
for (i = 0; i < n; i++)
d->table[i] = data[i];
- root_lock = qdisc_root_sleeping_lock(sch);
-
- spin_lock_bh(root_lock);
+ sch_tree_lock(sch);
swap(q->delay_dist, d);
- spin_unlock_bh(root_lock);
+ sch_tree_unlock(sch);
dist_free(d);
return 0;
--
1.8.0
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox