* [PATCH net-next v3] bonding: make global bonding stats more reliable
From: Andy Gospodarek @ 2014-09-26 1:12 UTC (permalink / raw)
To: netdev; +Cc: j.vosburgh, vfalico, nikolay
As the code stands today, bonding stats are based simply on the stats
from the member interfaces. If a member was to be removed from a bond,
the stats would instantly drop. This would be confusing to an admin
would would suddonly see interface stats drop while traffic is still
flowing.
In addition to preventing the stats drops mentioned above, new members
will now be added to the bond and only traffic received after the member
was added to the bond will be counted as part of bonding stats.
v2: Changes suggested by Nik to properly allocate/free stats memory.
v3: Properly destroy workqueue and fix netlink configuration path.
Signed-off-by: Andy Gospodarek <gospo@cumulusnetworks.com>
---
drivers/net/bonding/bond_main.c | 84 +++++++++++++++++++++++++-------------
drivers/net/bonding/bond_netlink.c | 13 +++++-
drivers/net/bonding/bonding.h | 3 ++
3 files changed, 71 insertions(+), 29 deletions(-)
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 5390475..10c2dc3 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1147,17 +1147,26 @@ static struct slave *bond_alloc_slave(struct bonding *bond)
slave = kzalloc(sizeof(struct slave), GFP_KERNEL);
if (!slave)
- return NULL;
+ goto slave_fail;
+
+ slave->slave_stats = kzalloc(sizeof(struct rtnl_link_stats64),
+ GFP_KERNEL);
+ if (!slave->slave_stats)
+ goto slave_stats_fail;
if (BOND_MODE(bond) == BOND_MODE_8023AD) {
SLAVE_AD_INFO(slave) = kzalloc(sizeof(struct ad_slave_info),
GFP_KERNEL);
- if (!SLAVE_AD_INFO(slave)) {
- kfree(slave);
- return NULL;
- }
+ if (!SLAVE_AD_INFO(slave))
+ goto slave_ad_fail;
}
return slave;
+slave_ad_fail:
+ kfree(slave->slave_stats);
+slave_stats_fail:
+ kfree(slave);
+slave_fail:
+ return NULL;
}
static void bond_free_slave(struct slave *slave)
@@ -1167,6 +1176,7 @@ static void bond_free_slave(struct slave *slave)
if (BOND_MODE(bond) == BOND_MODE_8023AD)
kfree(SLAVE_AD_INFO(slave));
+ kfree(slave->slave_stats);
kfree(slave);
}
@@ -1344,6 +1354,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
}
slave_dev->priv_flags |= IFF_BONDING;
+ /* initialize slave stats */
+ dev_get_stats(new_slave->dev, new_slave->slave_stats);
if (bond_is_lb(bond)) {
/* bond_alb_init_slave() must be called before all other stages since
@@ -3085,38 +3097,43 @@ static struct rtnl_link_stats64 *bond_get_stats(struct net_device *bond_dev,
struct list_head *iter;
struct slave *slave;
- memset(stats, 0, sizeof(*stats));
+ memcpy(stats, bond->bond_stats, sizeof(*stats));
bond_for_each_slave(bond, slave, iter) {
const struct rtnl_link_stats64 *sstats =
dev_get_stats(slave->dev, &temp);
+ struct rtnl_link_stats64 *pstats = slave->slave_stats;
+
+ stats->rx_packets += sstats->rx_packets - pstats->rx_packets;
+ stats->rx_bytes += sstats->rx_bytes - pstats->rx_bytes;
+ stats->rx_errors += sstats->rx_errors - pstats->rx_errors;
+ stats->rx_dropped += sstats->rx_dropped - pstats->rx_dropped;
- stats->rx_packets += sstats->rx_packets;
- stats->rx_bytes += sstats->rx_bytes;
- stats->rx_errors += sstats->rx_errors;
- stats->rx_dropped += sstats->rx_dropped;
+ stats->tx_packets += sstats->tx_packets - pstats->tx_packets;;
+ stats->tx_bytes += sstats->tx_bytes - pstats->tx_bytes;
+ stats->tx_errors += sstats->tx_errors - pstats->tx_errors;
+ stats->tx_dropped += sstats->tx_dropped - pstats->tx_dropped;
- stats->tx_packets += sstats->tx_packets;
- stats->tx_bytes += sstats->tx_bytes;
- stats->tx_errors += sstats->tx_errors;
- stats->tx_dropped += sstats->tx_dropped;
+ stats->multicast += sstats->multicast - pstats->multicast;
+ stats->collisions += sstats->collisions - pstats->collisions;
- stats->multicast += sstats->multicast;
- stats->collisions += sstats->collisions;
+ stats->rx_length_errors += sstats->rx_length_errors - pstats->rx_length_errors;
+ stats->rx_over_errors += sstats->rx_over_errors - pstats->rx_over_errors;
+ stats->rx_crc_errors += sstats->rx_crc_errors - pstats->rx_crc_errors;
+ stats->rx_frame_errors += sstats->rx_frame_errors - pstats->rx_frame_errors;
+ stats->rx_fifo_errors += sstats->rx_fifo_errors - pstats->rx_fifo_errors;
+ stats->rx_missed_errors += sstats->rx_missed_errors - pstats->rx_missed_errors;
- stats->rx_length_errors += sstats->rx_length_errors;
- stats->rx_over_errors += sstats->rx_over_errors;
- stats->rx_crc_errors += sstats->rx_crc_errors;
- stats->rx_frame_errors += sstats->rx_frame_errors;
- stats->rx_fifo_errors += sstats->rx_fifo_errors;
- stats->rx_missed_errors += sstats->rx_missed_errors;
+ stats->tx_aborted_errors += sstats->tx_aborted_errors - pstats->tx_aborted_errors;
+ stats->tx_carrier_errors += sstats->tx_carrier_errors - pstats->tx_carrier_errors;
+ stats->tx_fifo_errors += sstats->tx_fifo_errors - pstats->tx_fifo_errors;
+ stats->tx_heartbeat_errors += sstats->tx_heartbeat_errors - pstats->tx_heartbeat_errors;
+ stats->tx_window_errors += sstats->tx_window_errors - pstats->tx_window_errors;
- stats->tx_aborted_errors += sstats->tx_aborted_errors;
- stats->tx_carrier_errors += sstats->tx_carrier_errors;
- stats->tx_fifo_errors += sstats->tx_fifo_errors;
- stats->tx_heartbeat_errors += sstats->tx_heartbeat_errors;
- stats->tx_window_errors += sstats->tx_window_errors;
+ /* save off the slave stats for the next run */
+ memcpy(pstats, sstats, sizeof(*sstats));
}
+ memcpy(bond->bond_stats, stats, sizeof(*stats));
return stats;
}
@@ -3790,6 +3807,9 @@ static void bond_destructor(struct net_device *bond_dev)
struct bonding *bond = netdev_priv(bond_dev);
if (bond->wq)
destroy_workqueue(bond->wq);
+ if (bond->bond_stats)
+ kfree(bond->bond_stats);
+
free_netdev(bond_dev);
}
@@ -4274,7 +4294,8 @@ unsigned int bond_get_num_tx_queues(void)
int bond_create(struct net *net, const char *name)
{
struct net_device *bond_dev;
- int res;
+ struct bonding *bond;
+ int res = -ENOMEM;
rtnl_lock();
@@ -4286,8 +4307,15 @@ int bond_create(struct net *net, const char *name)
rtnl_unlock();
return -ENOMEM;
}
+ /* alloc persistent stats for the bond */
+ bond = netdev_priv(bond_dev);
+ bond->bond_stats = kzalloc(sizeof(struct rtnl_link_stats64),
+ GFP_KERNEL);
+ if (!bond->bond_stats)
+ printk(KERN_CRIT "Stats not available!\n");
dev_net_set(bond_dev, net);
+
bond_dev->rtnl_link_ops = &bond_link_ops;
res = register_netdevice(bond_dev);
diff --git a/drivers/net/bonding/bond_netlink.c b/drivers/net/bonding/bond_netlink.c
index c13d83e..2fdcbe2 100644
--- a/drivers/net/bonding/bond_netlink.c
+++ b/drivers/net/bonding/bond_netlink.c
@@ -381,10 +381,21 @@ static int bond_newlink(struct net *src_net, struct net_device *bond_dev,
struct nlattr *tb[], struct nlattr *data[])
{
int err;
+ struct bonding *bond = netdev_priv(bond_dev);
+
+ /* alloc persistent stats for the bond if not already allocated */
+ if (!bond->bond_stats) {
+ bond->bond_stats = kzalloc(sizeof(struct rtnl_link_stats64),
+ GFP_KERNEL);
+ if (!bond->bond_stats)
+ return -ENOMEM;
+ }
err = bond_changelink(bond_dev, tb, data);
- if (err < 0)
+ if (err < 0) {
+ kfree(bond->bond_stats);
return err;
+ }
return register_netdevice(bond_dev);
}
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index 6140bf0..fe25265 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -24,6 +24,7 @@
#include <linux/inetdevice.h>
#include <linux/etherdevice.h>
#include <linux/reciprocal_div.h>
+#include <linux/if_link.h>
#include "bond_3ad.h"
#include "bond_alb.h"
@@ -175,6 +176,7 @@ struct slave {
struct netpoll *np;
#endif
struct kobject kobj;
+ struct rtnl_link_stats64 *slave_stats;
};
/*
@@ -224,6 +226,7 @@ struct bonding {
/* debugging support via debugfs */
struct dentry *debug_dir;
#endif /* CONFIG_DEBUG_FS */
+ struct rtnl_link_stats64 *bond_stats;
};
#define bond_slave_get_rcu(dev) \
--
1.9.3
^ permalink raw reply related
* Re: FIXMEs in rt2800lib.c
From: nick @ 2014-09-26 17:02 UTC (permalink / raw)
To: Stanislaw Gruszka
Cc: Ivo Van Doorn, helmut.schaa, John Linville, linux-wireless, users,
netdev, julia.lawall@lip6.fr oneukum@suse.de
In-Reply-To: <CAAxn8J=gKVdNt3+Ayx-XdEki7V8sg-M7C=kK7sGADBO3L-PTQw@mail.gmail.com>
On 14-09-26 09:32 AM, Nicholas Krause wrote:
> On Fri, Sep 26, 2014 at 8:22 AM, Stanislaw Gruszka <sgruszka@redhat.com> wrote:
>> On Thu, Sep 25, 2014 at 01:52:43PM -0400, nick wrote:
>>> I seem to be hitting to fix mes in the file I listed in the subject. Would someone tell me if it's Ok to remove
>>> the lines that over write certain registers.
>>
>> No, as long you prove that they are not needed.
>>
>> Stanislaw
> Sure Stanislw,
> I will remove the lines as needed and send it a patch.
> Nick
>
Is that a typo and a yes or not?
Nick
^ permalink raw reply
* [net-next PATCH] net: sched: cls_rcvp, complete rcu conversion
From: John Fastabend @ 2014-09-26 17:02 UTC (permalink / raw)
To: xiyou.wangcong, jhs, eric.dumazet, davem; +Cc: netdev
This completes the cls_rsvp conversion to RCU safe
copy, update semantics.
As a result all cases of tcf_exts_change occur on
empty lists now.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
---
net/sched/cls_rsvp.h | 44 +++++++++++++++++++++++++++++++++++++++++---
1 file changed, 41 insertions(+), 3 deletions(-)
diff --git a/net/sched/cls_rsvp.h b/net/sched/cls_rsvp.h
index b044c20..2512c46 100644
--- a/net/sched/cls_rsvp.h
+++ b/net/sched/cls_rsvp.h
@@ -222,6 +222,33 @@ matched:
return -1;
}
+static void rsvp_replace(struct tcf_proto *tp, struct rsvp_filter *n, u32 h)
+{
+ struct rsvp_head *head = rtnl_dereference(tp->root);
+ struct rsvp_session *s;
+ struct rsvp_filter __rcu **ins;
+ struct rsvp_filter *pins;
+ unsigned int h1 = h & 0xFF;
+ unsigned int h2 = (h >> 8) & 0xFF;
+
+ for (s = rtnl_dereference(head->ht[h1]); s;
+ s = rtnl_dereference(s->next)) {
+ for (ins = &s->ht[h2], pins = rtnl_dereference(*ins); ;
+ ins = &pins->next, pins = rtnl_dereference(*ins)) {
+ if (pins->handle == h) {
+ RCU_INIT_POINTER(n->next, pins->next);
+ rcu_assign_pointer(*ins, n);
+ return;
+ }
+ }
+ }
+
+ /* Something went wrong if we are trying to replace a non-existant
+ * node. Mind as well halt instead of silently failing.
+ */
+ BUG_ON(1);
+}
+
static unsigned long rsvp_get(struct tcf_proto *tp, u32 handle)
{
struct rsvp_head *head = rtnl_dereference(tp->root);
@@ -454,15 +481,26 @@ static int rsvp_change(struct net *net, struct sk_buff *in_skb,
f = (struct rsvp_filter *)*arg;
if (f) {
/* Node exists: adjust only classid */
+ struct rsvp_filter *n;
if (f->handle != handle && handle)
goto errout2;
+
+ n = kmemdup(f, sizeof(*f), GFP_KERNEL);
+ if (!n) {
+ err = -ENOMEM;
+ goto errout2;
+ }
+
+ tcf_exts_init(&n->exts, TCA_RSVP_ACT, TCA_RSVP_POLICE);
+
if (tb[TCA_RSVP_CLASSID]) {
- f->res.classid = nla_get_u32(tb[TCA_RSVP_CLASSID]);
- tcf_bind_filter(tp, &f->res, base);
+ n->res.classid = nla_get_u32(tb[TCA_RSVP_CLASSID]);
+ tcf_bind_filter(tp, &n->res, base);
}
- tcf_exts_change(tp, &f->exts, &e);
+ tcf_exts_change(tp, &n->exts, &e);
+ rsvp_replace(tp, n, handle);
return 0;
}
^ permalink raw reply related
* Re: mmotm 2014-09-25-16-28 uploaded (nf_nat_masquerade_ipv4.c)
From: Arturo Borrero Gonzalez @ 2014-09-26 17:03 UTC (permalink / raw)
To: Randy Dunlap
Cc: akpm, netdev@vger.kernel.org, linux-kernel,
Netfilter Development Mailing list, linux-next, sfr, mhocko
In-Reply-To: <54259A5C.2060102@infradead.org>
On 26 September 2014 18:54, Randy Dunlap <rdunlap@infradead.org> wrote:
> On 09/25/14 16:28, akpm@linux-foundation.org wrote:
>> The mm-of-the-moment snapshot 2014-09-25-16-28 has been uploaded to
>>
>> http://www.ozlabs.org/~akpm/mmotm/
>>
>> mmotm-readme.txt says
>>
>> README for mm-of-the-moment:
>>
>> http://www.ozlabs.org/~akpm/mmotm/
>>
>> This is a snapshot of my -mm patch queue. Uploaded at random hopefully
>> more than once a week.
>
> ../net/ipv4/netfilter/nf_nat_masquerade_ipv4.c: In function 'nf_nat_masquerade_ipv4':
> ../net/ipv4/netfilter/nf_nat_masquerade_ipv4.c:59:5: error: 'struct nf_conn_nat' has no member named 'masq_index'
> nat->masq_index = out->ifindex;
> ^
> ../net/ipv4/netfilter/nf_nat_masquerade_ipv4.c: In function 'device_cmp':
> ../net/ipv4/netfilter/nf_nat_masquerade_ipv4.c:83:12: error: 'const struct nf_conn_nat' has no member named 'masq_index'
> return nat->masq_index == (int)(long)ifindex;
> ^
> ../net/ipv4/netfilter/nf_nat_masquerade_ipv4.c:84:1: warning: control reaches end of non-void function [-Wreturn-type]
>
>
That seems fixed in this commit by Pablo Neira:
http://git.kernel.org/cgit/linux/kernel/git/pablo/nf-next.git/commit/?id=67981fefb20e717cea55b42f9081a833fa46b3be
I guess the patch will hit linux-next soon.
Thanks, regards.
--
Arturo Borrero González
^ permalink raw reply
* Re: [PATCH v5 09/10] ARM: dts: berlin: add the Ethernet node
From: Sergei Shtylyov @ 2014-09-26 17:25 UTC (permalink / raw)
To: Antoine Tenart, sebastian.hesselbarth
Cc: thomas.petazzoni, zmxu, devicetree, netdev, linux-kernel,
alexandre.belloni, jszhang, linux-arm-kernel
In-Reply-To: <1411742036-23520-10-git-send-email-antoine.tenart@free-electrons.com>
Hello.
On 09/26/2014 06:33 PM, Antoine Tenart wrote:
> This patch adds the Ethernet node, enabling the network unit on Berlin
> BG2Q SoCs.
> Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
> Acked-by: Arnd Bergmann <arnd@arndb.de>
> ---
> arch/arm/boot/dts/berlin2q.dtsi | 17 +++++++++++++++++
> 1 file changed, 17 insertions(+)
> diff --git a/arch/arm/boot/dts/berlin2q.dtsi b/arch/arm/boot/dts/berlin2q.dtsi
> index 902eddb19cd8..d442b22fd1ea 100644
> --- a/arch/arm/boot/dts/berlin2q.dtsi
> +++ b/arch/arm/boot/dts/berlin2q.dtsi
> @@ -114,6 +114,23 @@
> #interrupt-cells = <3>;
> };
>
> + eth0: ethernet@b90000 {
> + compatible = "marvell,pxa168-eth";
> + reg = <0xb90000 0x10000>;
> + clocks = <&chip CLKID_GETH0>;
> + interrupts = <GIC_SPI 24 IRQ_TYPE_LEVEL_HIGH>;
> + /* set by bootloader */
> + local-mac-address = [00 00 00 00 00 00];
> + #address-cells = <1>;
> + #size-cells = <0>;
> + phy-handle = <ðphy0>;
> + status = "disabled";
> +
> + ethphy0: ethernet-phy@0 {
> + reg = <0>;
> + };
> + };
> +
Hm, is the PHY internal to the Ethernet controller?
WBR, Sergei
^ permalink raw reply
* tc rsvp filter show broke
From: John Fastabend @ 2014-09-26 17:41 UTC (permalink / raw)
To: Jamal Hadi Salim; +Cc: Cong Wang, netdev
Hi Jamal,
Here is another strange 'tc' result that I think is a
bug. If it was intended I can't see why. Show filter
commands on rsvp classifiers only list the first entry?
# for i in {10..20}; do
tc filter add dev p3p2 protocol ip parent 8001: prio $i \
rsvp session 12.0.0.$i ipproto udp classid 1:$i;
done
# tc filter show dev p3p2
filter parent 8001: protocol ip pref 10 rsvp
# tc filter show dev p3p2 prio 11
filter parent 8001: protocol ip rsvp
# tc filter show dev p3p2 prio 22
Although the entries must still be around because listing
the prio explicitly show the entries. Also the case where
the prio doesn't exist returns without any errors. A better
response might be 'EINVAL' or something.
This is on 3.16 so nothing to do with my changes. I'm
not sure I'll get a chance until Monday to look at it
so figured I'll document it here in case someone else has
a moment.
Thanks,
John
--
John Fastabend Intel Corporation
^ permalink raw reply
* Re: [PATCH] [RESEND] ath5k: Remove AHB bus support
From: John W. Linville @ 2014-09-26 17:45 UTC (permalink / raw)
To: Paul Bolle
Cc: Jiri Slaby, Luis R. Rodriguez, open-5/S+JYg5SzeELgA04lAiVw,
Richard Weinberger, ath5k-devel-juf53994utBLZpfksSYvnA,
linux-wireless-u79uwXL29TY76Z2rM5mHXA,
linux-kernel-u79uwXL29TY76Z2rM5mHXA, Oleksij Rempel,
netdev-u79uwXL29TY76Z2rM5mHXA, Hauke Mehrtens,
list-5/S+JYg5SzeELgA04lAiVw, Jonathan Bither,
openwrt-devel-ZwoEplunGu0+xA9t4nZiwti2O/JbrIOy,
antonynpavlov-Re5JQEeQqe8AvxtiuMwx3w
In-Reply-To: <1410339221.11097.10.camel@x220>
OK, I'm merging this...
On Wed, Sep 10, 2014 at 10:53:41AM +0200, Paul Bolle wrote:
> AHB bus support was added in v2.6.38, through commit a0b907ee2a71
> ("ath5k: Add AHB bus support."). That code can only be build if the
> Kconfig symbol ATHEROS_AR231X is set. But that symbol has never been
> added to the tree. So AHB bus support has always been dead code.
>
> Let's remove all code that depends on ATHEROS_AR231X. If that symbol
> ever gets added to the tree the AHB bus support can be re-added too.
>
> Signed-off-by: Paul Bolle <pebolle-IWqWACnzNjzz+pZb47iToQ@public.gmane.org>
> ---
> First sent in https://lkml.org/lkml/2013/5/13/303 . Updated on top of
> next-20140909.
>
> Build tested only (no access to this hardware).
>
> drivers/net/wireless/ath/ath5k/Kconfig | 14 +-
> drivers/net/wireless/ath/ath5k/Makefile | 1 -
> drivers/net/wireless/ath/ath5k/ahb.c | 234 --------------------------------
> drivers/net/wireless/ath/ath5k/ath5k.h | 28 ----
> drivers/net/wireless/ath/ath5k/base.c | 14 --
> drivers/net/wireless/ath/ath5k/led.c | 6 -
> 6 files changed, 3 insertions(+), 294 deletions(-)
> delete mode 100644 drivers/net/wireless/ath/ath5k/ahb.c
>
> diff --git a/drivers/net/wireless/ath/ath5k/Kconfig b/drivers/net/wireless/ath/ath5k/Kconfig
> index c9f81a388f15..93caf8e68901 100644
> --- a/drivers/net/wireless/ath/ath5k/Kconfig
> +++ b/drivers/net/wireless/ath/ath5k/Kconfig
> @@ -1,13 +1,12 @@
> config ATH5K
> tristate "Atheros 5xxx wireless cards support"
> - depends on (PCI || ATHEROS_AR231X) && MAC80211
> + depends on PCI && MAC80211
> select ATH_COMMON
> select MAC80211_LEDS
> select LEDS_CLASS
> select NEW_LEDS
> select AVERAGE
> - select ATH5K_AHB if (ATHEROS_AR231X && !PCI)
> - select ATH5K_PCI if (!ATHEROS_AR231X && PCI)
> + select ATH5K_PCI
> ---help---
> This module adds support for wireless adapters based on
> Atheros 5xxx chipset.
> @@ -52,16 +51,9 @@ config ATH5K_TRACER
>
> If unsure, say N.
>
> -config ATH5K_AHB
> - bool "Atheros 5xxx AHB bus support"
> - depends on (ATHEROS_AR231X && !PCI)
> - ---help---
> - This adds support for WiSoC type chipsets of the 5xxx Atheros
> - family.
> -
> config ATH5K_PCI
> bool "Atheros 5xxx PCI bus support"
> - depends on (!ATHEROS_AR231X && PCI)
> + depends on PCI
> ---help---
> This adds support for PCI type chipsets of the 5xxx Atheros
> family.
> diff --git a/drivers/net/wireless/ath/ath5k/Makefile b/drivers/net/wireless/ath/ath5k/Makefile
> index 1b3a34f7f224..51e2d8668041 100644
> --- a/drivers/net/wireless/ath/ath5k/Makefile
> +++ b/drivers/net/wireless/ath/ath5k/Makefile
> @@ -17,6 +17,5 @@ ath5k-y += ani.o
> ath5k-y += sysfs.o
> ath5k-y += mac80211-ops.o
> ath5k-$(CONFIG_ATH5K_DEBUG) += debug.o
> -ath5k-$(CONFIG_ATH5K_AHB) += ahb.o
> ath5k-$(CONFIG_ATH5K_PCI) += pci.o
> obj-$(CONFIG_ATH5K) += ath5k.o
> diff --git a/drivers/net/wireless/ath/ath5k/ahb.c b/drivers/net/wireless/ath/ath5k/ahb.c
> deleted file mode 100644
> index 79bffe165cab..000000000000
> --- a/drivers/net/wireless/ath/ath5k/ahb.c
> +++ /dev/null
> @@ -1,234 +0,0 @@
> -/*
> - * Copyright (c) 2008-2009 Atheros Communications Inc.
> - * Copyright (c) 2009 Gabor Juhos <juhosg-p3rKhJxN3npAfugRpC6u6w@public.gmane.org>
> - * Copyright (c) 2009 Imre Kaloz <kaloz-p3rKhJxN3npAfugRpC6u6w@public.gmane.org>
> - *
> - * Permission to use, copy, modify, and/or distribute this software for any
> - * purpose with or without fee is hereby granted, provided that the above
> - * copyright notice and this permission notice appear in all copies.
> - *
> - * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
> - * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
> - * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
> - * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
> - * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
> - * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
> - * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
> - */
> -
> -#include <linux/nl80211.h>
> -#include <linux/platform_device.h>
> -#include <linux/etherdevice.h>
> -#include <linux/export.h>
> -#include <ar231x_platform.h>
> -#include "ath5k.h"
> -#include "debug.h"
> -#include "base.h"
> -#include "reg.h"
> -
> -/* return bus cachesize in 4B word units */
> -static void ath5k_ahb_read_cachesize(struct ath_common *common, int *csz)
> -{
> - *csz = L1_CACHE_BYTES >> 2;
> -}
> -
> -static bool
> -ath5k_ahb_eeprom_read(struct ath_common *common, u32 off, u16 *data)
> -{
> - struct ath5k_hw *ah = common->priv;
> - struct platform_device *pdev = to_platform_device(ah->dev);
> - struct ar231x_board_config *bcfg = dev_get_platdata(&pdev->dev);
> - u16 *eeprom, *eeprom_end;
> -
> - eeprom = (u16 *) bcfg->radio;
> - eeprom_end = ((void *) bcfg->config) + BOARD_CONFIG_BUFSZ;
> -
> - eeprom += off;
> - if (eeprom > eeprom_end)
> - return false;
> -
> - *data = *eeprom;
> - return true;
> -}
> -
> -int ath5k_hw_read_srev(struct ath5k_hw *ah)
> -{
> - struct platform_device *pdev = to_platform_device(ah->dev);
> - struct ar231x_board_config *bcfg = dev_get_platdata(&pdev->dev);
> - ah->ah_mac_srev = bcfg->devid;
> - return 0;
> -}
> -
> -static int ath5k_ahb_eeprom_read_mac(struct ath5k_hw *ah, u8 *mac)
> -{
> - struct platform_device *pdev = to_platform_device(ah->dev);
> - struct ar231x_board_config *bcfg = dev_get_platdata(&pdev->dev);
> - u8 *cfg_mac;
> -
> - if (to_platform_device(ah->dev)->id == 0)
> - cfg_mac = bcfg->config->wlan0_mac;
> - else
> - cfg_mac = bcfg->config->wlan1_mac;
> -
> - memcpy(mac, cfg_mac, ETH_ALEN);
> - return 0;
> -}
> -
> -static const struct ath_bus_ops ath_ahb_bus_ops = {
> - .ath_bus_type = ATH_AHB,
> - .read_cachesize = ath5k_ahb_read_cachesize,
> - .eeprom_read = ath5k_ahb_eeprom_read,
> - .eeprom_read_mac = ath5k_ahb_eeprom_read_mac,
> -};
> -
> -/*Initialization*/
> -static int ath_ahb_probe(struct platform_device *pdev)
> -{
> - struct ar231x_board_config *bcfg = dev_get_platdata(&pdev->dev);
> - struct ath5k_hw *ah;
> - struct ieee80211_hw *hw;
> - struct resource *res;
> - void __iomem *mem;
> - int irq;
> - int ret = 0;
> - u32 reg;
> -
> - if (!dev_get_platdata(&pdev->dev)) {
> - dev_err(&pdev->dev, "no platform data specified\n");
> - ret = -EINVAL;
> - goto err_out;
> - }
> -
> - res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> - if (res == NULL) {
> - dev_err(&pdev->dev, "no memory resource found\n");
> - ret = -ENXIO;
> - goto err_out;
> - }
> -
> - mem = ioremap_nocache(res->start, resource_size(res));
> - if (mem == NULL) {
> - dev_err(&pdev->dev, "ioremap failed\n");
> - ret = -ENOMEM;
> - goto err_out;
> - }
> -
> - res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
> - if (res == NULL) {
> - dev_err(&pdev->dev, "no IRQ resource found\n");
> - ret = -ENXIO;
> - goto err_iounmap;
> - }
> -
> - irq = res->start;
> -
> - hw = ieee80211_alloc_hw(sizeof(struct ath5k_hw), &ath5k_hw_ops);
> - if (hw == NULL) {
> - dev_err(&pdev->dev, "no memory for ieee80211_hw\n");
> - ret = -ENOMEM;
> - goto err_iounmap;
> - }
> -
> - ah = hw->priv;
> - ah->hw = hw;
> - ah->dev = &pdev->dev;
> - ah->iobase = mem;
> - ah->irq = irq;
> - ah->devid = bcfg->devid;
> -
> - if (bcfg->devid >= AR5K_SREV_AR2315_R6) {
> - /* Enable WMAC AHB arbitration */
> - reg = ioread32((void __iomem *) AR5K_AR2315_AHB_ARB_CTL);
> - reg |= AR5K_AR2315_AHB_ARB_CTL_WLAN;
> - iowrite32(reg, (void __iomem *) AR5K_AR2315_AHB_ARB_CTL);
> -
> - /* Enable global WMAC swapping */
> - reg = ioread32((void __iomem *) AR5K_AR2315_BYTESWAP);
> - reg |= AR5K_AR2315_BYTESWAP_WMAC;
> - iowrite32(reg, (void __iomem *) AR5K_AR2315_BYTESWAP);
> - } else {
> - /* Enable WMAC DMA access (assuming 5312 or 231x*/
> - /* TODO: check other platforms */
> - reg = ioread32((void __iomem *) AR5K_AR5312_ENABLE);
> - if (to_platform_device(ah->dev)->id == 0)
> - reg |= AR5K_AR5312_ENABLE_WLAN0;
> - else
> - reg |= AR5K_AR5312_ENABLE_WLAN1;
> - iowrite32(reg, (void __iomem *) AR5K_AR5312_ENABLE);
> -
> - /*
> - * On a dual-band AR5312, the multiband radio is only
> - * used as pass-through. Disable 2 GHz support in the
> - * driver for it
> - */
> - if (to_platform_device(ah->dev)->id == 0 &&
> - (bcfg->config->flags & (BD_WLAN0 | BD_WLAN1)) ==
> - (BD_WLAN1 | BD_WLAN0))
> - ah->ah_capabilities.cap_needs_2GHz_ovr = true;
> - else
> - ah->ah_capabilities.cap_needs_2GHz_ovr = false;
> - }
> -
> - ret = ath5k_init_ah(ah, &ath_ahb_bus_ops);
> - if (ret != 0) {
> - dev_err(&pdev->dev, "failed to attach device, err=%d\n", ret);
> - ret = -ENODEV;
> - goto err_free_hw;
> - }
> -
> - platform_set_drvdata(pdev, hw);
> -
> - return 0;
> -
> - err_free_hw:
> - ieee80211_free_hw(hw);
> - err_iounmap:
> - iounmap(mem);
> - err_out:
> - return ret;
> -}
> -
> -static int ath_ahb_remove(struct platform_device *pdev)
> -{
> - struct ar231x_board_config *bcfg = dev_get_platdata(&pdev->dev);
> - struct ieee80211_hw *hw = platform_get_drvdata(pdev);
> - struct ath5k_hw *ah;
> - u32 reg;
> -
> - if (!hw)
> - return 0;
> -
> - ah = hw->priv;
> -
> - if (bcfg->devid >= AR5K_SREV_AR2315_R6) {
> - /* Disable WMAC AHB arbitration */
> - reg = ioread32((void __iomem *) AR5K_AR2315_AHB_ARB_CTL);
> - reg &= ~AR5K_AR2315_AHB_ARB_CTL_WLAN;
> - iowrite32(reg, (void __iomem *) AR5K_AR2315_AHB_ARB_CTL);
> - } else {
> - /*Stop DMA access */
> - reg = ioread32((void __iomem *) AR5K_AR5312_ENABLE);
> - if (to_platform_device(ah->dev)->id == 0)
> - reg &= ~AR5K_AR5312_ENABLE_WLAN0;
> - else
> - reg &= ~AR5K_AR5312_ENABLE_WLAN1;
> - iowrite32(reg, (void __iomem *) AR5K_AR5312_ENABLE);
> - }
> -
> - ath5k_deinit_ah(ah);
> - iounmap(ah->iobase);
> - ieee80211_free_hw(hw);
> -
> - return 0;
> -}
> -
> -static struct platform_driver ath_ahb_driver = {
> - .probe = ath_ahb_probe,
> - .remove = ath_ahb_remove,
> - .driver = {
> - .name = "ar231x-wmac",
> - .owner = THIS_MODULE,
> - },
> -};
> -
> -module_platform_driver(ath_ahb_driver);
> diff --git a/drivers/net/wireless/ath/ath5k/ath5k.h b/drivers/net/wireless/ath/ath5k/ath5k.h
> index 85316bb3f8c6..ed2468220216 100644
> --- a/drivers/net/wireless/ath/ath5k/ath5k.h
> +++ b/drivers/net/wireless/ath/ath5k/ath5k.h
> @@ -1647,32 +1647,6 @@ static inline struct ath_regulatory *ath5k_hw_regulatory(struct ath5k_hw *ah)
> return &(ath5k_hw_common(ah)->regulatory);
> }
>
> -#ifdef CONFIG_ATHEROS_AR231X
> -#define AR5K_AR2315_PCI_BASE ((void __iomem *)0xb0100000)
> -
> -static inline void __iomem *ath5k_ahb_reg(struct ath5k_hw *ah, u16 reg)
> -{
> - /* On AR2315 and AR2317 the PCI clock domain registers
> - * are outside of the WMAC register space */
> - if (unlikely((reg >= 0x4000) && (reg < 0x5000) &&
> - (ah->ah_mac_srev >= AR5K_SREV_AR2315_R6)))
> - return AR5K_AR2315_PCI_BASE + reg;
> -
> - return ah->iobase + reg;
> -}
> -
> -static inline u32 ath5k_hw_reg_read(struct ath5k_hw *ah, u16 reg)
> -{
> - return ioread32(ath5k_ahb_reg(ah, reg));
> -}
> -
> -static inline void ath5k_hw_reg_write(struct ath5k_hw *ah, u32 val, u16 reg)
> -{
> - iowrite32(val, ath5k_ahb_reg(ah, reg));
> -}
> -
> -#else
> -
> static inline u32 ath5k_hw_reg_read(struct ath5k_hw *ah, u16 reg)
> {
> return ioread32(ah->iobase + reg);
> @@ -1683,8 +1657,6 @@ static inline void ath5k_hw_reg_write(struct ath5k_hw *ah, u32 val, u16 reg)
> iowrite32(val, ah->iobase + reg);
> }
>
> -#endif
> -
> static inline enum ath_bus_type ath5k_get_bus_type(struct ath5k_hw *ah)
> {
> return ath5k_hw_common(ah)->bus_ops->ath_bus_type;
> diff --git a/drivers/net/wireless/ath/ath5k/base.c b/drivers/net/wireless/ath/ath5k/base.c
> index 59a87247aac4..a4a09bb8f2f3 100644
> --- a/drivers/net/wireless/ath/ath5k/base.c
> +++ b/drivers/net/wireless/ath/ath5k/base.c
> @@ -99,15 +99,6 @@ static int ath5k_reset(struct ath5k_hw *ah, struct ieee80211_channel *chan,
>
> /* Known SREVs */
> static const struct ath5k_srev_name srev_names[] = {
> -#ifdef CONFIG_ATHEROS_AR231X
> - { "5312", AR5K_VERSION_MAC, AR5K_SREV_AR5312_R2 },
> - { "5312", AR5K_VERSION_MAC, AR5K_SREV_AR5312_R7 },
> - { "2313", AR5K_VERSION_MAC, AR5K_SREV_AR2313_R8 },
> - { "2315", AR5K_VERSION_MAC, AR5K_SREV_AR2315_R6 },
> - { "2315", AR5K_VERSION_MAC, AR5K_SREV_AR2315_R7 },
> - { "2317", AR5K_VERSION_MAC, AR5K_SREV_AR2317_R1 },
> - { "2317", AR5K_VERSION_MAC, AR5K_SREV_AR2317_R2 },
> -#else
> { "5210", AR5K_VERSION_MAC, AR5K_SREV_AR5210 },
> { "5311", AR5K_VERSION_MAC, AR5K_SREV_AR5311 },
> { "5311A", AR5K_VERSION_MAC, AR5K_SREV_AR5311A },
> @@ -126,7 +117,6 @@ static const struct ath5k_srev_name srev_names[] = {
> { "5418", AR5K_VERSION_MAC, AR5K_SREV_AR5418 },
> { "2425", AR5K_VERSION_MAC, AR5K_SREV_AR2425 },
> { "2417", AR5K_VERSION_MAC, AR5K_SREV_AR2417 },
> -#endif
> { "xxxxx", AR5K_VERSION_MAC, AR5K_SREV_UNKNOWN },
> { "5110", AR5K_VERSION_RAD, AR5K_SREV_RAD_5110 },
> { "5111", AR5K_VERSION_RAD, AR5K_SREV_RAD_5111 },
> @@ -142,10 +132,6 @@ static const struct ath5k_srev_name srev_names[] = {
> { "5413", AR5K_VERSION_RAD, AR5K_SREV_RAD_5413 },
> { "5424", AR5K_VERSION_RAD, AR5K_SREV_RAD_5424 },
> { "5133", AR5K_VERSION_RAD, AR5K_SREV_RAD_5133 },
> -#ifdef CONFIG_ATHEROS_AR231X
> - { "2316", AR5K_VERSION_RAD, AR5K_SREV_RAD_2316 },
> - { "2317", AR5K_VERSION_RAD, AR5K_SREV_RAD_2317 },
> -#endif
> { "xxxxx", AR5K_VERSION_RAD, AR5K_SREV_UNKNOWN },
> };
>
> diff --git a/drivers/net/wireless/ath/ath5k/led.c b/drivers/net/wireless/ath/ath5k/led.c
> index 2062d1190556..0beb7e7d6075 100644
> --- a/drivers/net/wireless/ath/ath5k/led.c
> +++ b/drivers/net/wireless/ath/ath5k/led.c
> @@ -163,20 +163,14 @@ int ath5k_init_leds(struct ath5k_hw *ah)
> {
> int ret = 0;
> struct ieee80211_hw *hw = ah->hw;
> -#ifndef CONFIG_ATHEROS_AR231X
> struct pci_dev *pdev = ah->pdev;
> -#endif
> char name[ATH5K_LED_MAX_NAME_LEN + 1];
> const struct pci_device_id *match;
>
> if (!ah->pdev)
> return 0;
>
> -#ifdef CONFIG_ATHEROS_AR231X
> - match = NULL;
> -#else
> match = pci_match_id(&ath5k_led_devices[0], pdev);
> -#endif
> if (match) {
> __set_bit(ATH_STAT_LEDSOFT, ah->status);
> ah->led_pin = ATH_PIN(match->driver_data);
> --
> 1.9.3
>
>
>
--
John W. Linville Someday the world will need a hero, and you
linville-2XuSBdqkA4R54TAoqtyWWQ@public.gmane.org might be all we have. Be ready.
^ permalink raw reply
* Re: mmotm 2014-09-25-16-28 uploaded (nf_nat_masquerade_ipv4.c)
From: Pablo Neira Ayuso @ 2014-09-26 17:45 UTC (permalink / raw)
To: Randy Dunlap
Cc: akpm, netdev@vger.kernel.org, linux-kernel, netfilter-devel,
linux-next, sfr, mhocko
In-Reply-To: <54259A5C.2060102@infradead.org>
Hi Randy,
On Fri, Sep 26, 2014 at 09:54:52AM -0700, Randy Dunlap wrote:
> On 09/25/14 16:28, akpm@linux-foundation.org wrote:
> > The mm-of-the-moment snapshot 2014-09-25-16-28 has been uploaded to
> >
> > http://www.ozlabs.org/~akpm/mmotm/
> >
> > mmotm-readme.txt says
> >
> > README for mm-of-the-moment:
> >
> > http://www.ozlabs.org/~akpm/mmotm/
> >
> > This is a snapshot of my -mm patch queue. Uploaded at random hopefully
> > more than once a week.
>
> ../net/ipv4/netfilter/nf_nat_masquerade_ipv4.c: In function 'nf_nat_masquerade_ipv4':
> ../net/ipv4/netfilter/nf_nat_masquerade_ipv4.c:59:5: error: 'struct nf_conn_nat' has no member named 'masq_index'
> nat->masq_index = out->ifindex;
> ^
> ../net/ipv4/netfilter/nf_nat_masquerade_ipv4.c: In function 'device_cmp':
> ../net/ipv4/netfilter/nf_nat_masquerade_ipv4.c:83:12: error: 'const struct nf_conn_nat' has no member named 'masq_index'
> return nat->masq_index == (int)(long)ifindex;
> ^
> ../net/ipv4/netfilter/nf_nat_masquerade_ipv4.c:84:1: warning: control reaches end of non-void function [-Wreturn-type]
I'll pass this patch in the upcoming nf-next batch:
http://git.kernel.org/cgit/linux/kernel/git/pablo/nf-next.git/commit/?id=67981fefb20e717cea55b42f9081a833fa46b3be
Thanks.
^ permalink raw reply
* Re: [PATCH V2 1/2] brcm80211: use container_of to resolve brcms_phy from brcms_phy_pub
From: John W. Linville @ 2014-09-26 17:46 UTC (permalink / raw)
To: Fabian Frederick
Cc: brudley-dY08KVG/lbpWk0Htik3J/w, Arend van Spriel,
Franky (Zhenhui) Lin, Hante Meuleman,
linux-wireless-u79uwXL29TY76Z2rM5mHXA,
brcm80211-dev-list-dY08KVG/lbpWk0Htik3J/w,
netdev-u79uwXL29TY76Z2rM5mHXA,
linux-kernel-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1410809448-15982-1-git-send-email-fabf-AgBVmzD5pcezQB+pC5nmwQ@public.gmane.org>
These two patches don't seem to apply on wireless-next, and I don't
have time at the moment to fix them up. Feel free to repost...
On Mon, Sep 15, 2014 at 09:30:46PM +0200, Fabian Frederick wrote:
> Use container_of instead of casting first structure member.
>
> Acked-by: Arend van Spriel <arend-dY08KVG/lbpWk0Htik3J/w@public.gmane.org>
> Signed-off-by: Fabian Frederick <fabf-AgBVmzD5pcezQB+pC5nmwQ@public.gmane.org>
> ---
>
> Compiled but untested.
>
> V2: 2 separate patches for brcm80211 (suggested by Arend van Spriel)
>
> .../net/wireless/brcm80211/brcmsmac/phy/phy_cmn.c | 122 ++++++++++-----------
> .../net/wireless/brcm80211/brcmsmac/phy/phy_lcn.c | 6 +-
> .../net/wireless/brcm80211/brcmsmac/phy/phy_n.c | 8 +-
> 3 files changed, 68 insertions(+), 68 deletions(-)
>
> diff --git a/drivers/net/wireless/brcm80211/brcmsmac/phy/phy_cmn.c b/drivers/net/wireless/brcm80211/brcmsmac/phy/phy_cmn.c
> index 57ecc05..941b1e4 100644
> --- a/drivers/net/wireless/brcm80211/brcmsmac/phy/phy_cmn.c
> +++ b/drivers/net/wireless/brcm80211/brcmsmac/phy/phy_cmn.c
> @@ -128,19 +128,19 @@ static const u8 ofdm_rate_lookup[] = {
>
> void wlc_phyreg_enter(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> wlapi_bmac_ucode_wake_override_phyreg_set(pi->sh->physhim);
> }
>
> void wlc_phyreg_exit(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> wlapi_bmac_ucode_wake_override_phyreg_clear(pi->sh->physhim);
> }
>
> void wlc_radioreg_enter(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> wlapi_bmac_mctrl(pi->sh->physhim, MCTL_LOCK_RADIO, MCTL_LOCK_RADIO);
>
> udelay(10);
> @@ -148,7 +148,7 @@ void wlc_radioreg_enter(struct brcms_phy_pub *pih)
>
> void wlc_radioreg_exit(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> (void)bcma_read16(pi->d11core, D11REGOFFS(phyversion));
> pi->phy_wreg = 0;
> @@ -586,7 +586,7 @@ err:
>
> void wlc_phy_detach(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> if (pih) {
> if (--pi->refcnt)
> @@ -613,7 +613,7 @@ bool
> wlc_phy_get_phyversion(struct brcms_phy_pub *pih, u16 *phytype, u16 *phyrev,
> u16 *radioid, u16 *radiover)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> *phytype = (u16) pi->pubpi.phy_type;
> *phyrev = (u16) pi->pubpi.phy_rev;
> *radioid = pi->pubpi.radioid;
> @@ -624,19 +624,19 @@ wlc_phy_get_phyversion(struct brcms_phy_pub *pih, u16 *phytype, u16 *phyrev,
>
> bool wlc_phy_get_encore(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> return pi->pubpi.abgphy_encore;
> }
>
> u32 wlc_phy_get_coreflags(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> return pi->pubpi.coreflags;
> }
>
> void wlc_phy_anacore(struct brcms_phy_pub *pih, bool on)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> if (ISNPHY(pi)) {
> if (on) {
> @@ -673,7 +673,7 @@ void wlc_phy_anacore(struct brcms_phy_pub *pih, bool on)
>
> u32 wlc_phy_clk_bwbits(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> u32 phy_bw_clkbits = 0;
>
> @@ -698,14 +698,14 @@ u32 wlc_phy_clk_bwbits(struct brcms_phy_pub *pih)
>
> void wlc_phy_por_inform(struct brcms_phy_pub *ppi)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> pi->phy_init_por = true;
> }
>
> void wlc_phy_edcrs_lock(struct brcms_phy_pub *pih, bool lock)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> pi->edcrs_threshold_lock = lock;
>
> @@ -717,14 +717,14 @@ void wlc_phy_edcrs_lock(struct brcms_phy_pub *pih, bool lock)
>
> void wlc_phy_initcal_enable(struct brcms_phy_pub *pih, bool initcal)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> pi->do_initcal = initcal;
> }
>
> void wlc_phy_hw_clk_state_upd(struct brcms_phy_pub *pih, bool newstate)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> if (!pi || !pi->sh)
> return;
> @@ -734,7 +734,7 @@ void wlc_phy_hw_clk_state_upd(struct brcms_phy_pub *pih, bool newstate)
>
> void wlc_phy_hw_state_upd(struct brcms_phy_pub *pih, bool newstate)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> if (!pi || !pi->sh)
> return;
> @@ -746,7 +746,7 @@ void wlc_phy_init(struct brcms_phy_pub *pih, u16 chanspec)
> {
> u32 mc;
> void (*phy_init)(struct brcms_phy *) = NULL;
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> if (pi->init_in_progress)
> return;
> @@ -798,7 +798,7 @@ void wlc_phy_init(struct brcms_phy_pub *pih, u16 chanspec)
>
> void wlc_phy_cal_init(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> void (*cal_init)(struct brcms_phy *) = NULL;
>
> if (WARN((bcma_read32(pi->d11core, D11REGOFFS(maccontrol)) &
> @@ -816,7 +816,7 @@ void wlc_phy_cal_init(struct brcms_phy_pub *pih)
>
> int wlc_phy_down(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> int callbacks = 0;
>
> if (pi->phycal_timer
> @@ -1070,7 +1070,7 @@ void wlc_phy_do_dummy_tx(struct brcms_phy *pi, bool ofdm, bool pa_on)
>
> void wlc_phy_hold_upd(struct brcms_phy_pub *pih, u32 id, bool set)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> if (set)
> mboolset(pi->measure_hold, id);
> @@ -1082,7 +1082,7 @@ void wlc_phy_hold_upd(struct brcms_phy_pub *pih, u32 id, bool set)
>
> void wlc_phy_mute_upd(struct brcms_phy_pub *pih, bool mute, u32 flags)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> if (mute)
> mboolset(pi->measure_hold, PHY_HOLD_FOR_MUTE);
> @@ -1096,7 +1096,7 @@ void wlc_phy_mute_upd(struct brcms_phy_pub *pih, bool mute, u32 flags)
>
> void wlc_phy_clear_tssi(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> if (ISNPHY(pi)) {
> return;
> @@ -1115,7 +1115,7 @@ static bool wlc_phy_cal_txpower_recalc_sw(struct brcms_phy *pi)
>
> void wlc_phy_switch_radio(struct brcms_phy_pub *pih, bool on)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> (void)bcma_read32(pi->d11core, D11REGOFFS(maccontrol));
>
> if (ISNPHY(pi)) {
> @@ -1149,35 +1149,35 @@ void wlc_phy_switch_radio(struct brcms_phy_pub *pih, bool on)
>
> u16 wlc_phy_bw_state_get(struct brcms_phy_pub *ppi)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> return pi->bw;
> }
>
> void wlc_phy_bw_state_set(struct brcms_phy_pub *ppi, u16 bw)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> pi->bw = bw;
> }
>
> void wlc_phy_chanspec_radio_set(struct brcms_phy_pub *ppi, u16 newch)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> pi->radio_chanspec = newch;
>
> }
>
> u16 wlc_phy_chanspec_get(struct brcms_phy_pub *ppi)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> return pi->radio_chanspec;
> }
>
> void wlc_phy_chanspec_set(struct brcms_phy_pub *ppi, u16 chanspec)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> u16 m_cur_channel;
> void (*chanspec_set)(struct brcms_phy *, u16) = NULL;
> m_cur_channel = CHSPEC_CHANNEL(chanspec);
> @@ -1226,7 +1226,7 @@ int wlc_phy_chanspec_bandrange_get(struct brcms_phy *pi, u16 chanspec)
> void wlc_phy_chanspec_ch14_widefilter_set(struct brcms_phy_pub *ppi,
> bool wide_filter)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> pi->channel_14_wide_filter = wide_filter;
>
> @@ -1246,7 +1246,7 @@ void
> wlc_phy_chanspec_band_validch(struct brcms_phy_pub *ppi, uint band,
> struct brcms_chanvec *channels)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> uint i;
> uint channel;
>
> @@ -1267,7 +1267,7 @@ wlc_phy_chanspec_band_validch(struct brcms_phy_pub *ppi, uint band,
>
> u16 wlc_phy_chanspec_band_firstch(struct brcms_phy_pub *ppi, uint band)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> uint i;
> uint channel;
> u16 chspec;
> @@ -1311,7 +1311,7 @@ u16 wlc_phy_chanspec_band_firstch(struct brcms_phy_pub *ppi, uint band)
>
> int wlc_phy_txpower_get(struct brcms_phy_pub *ppi, uint *qdbm, bool *override)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> *qdbm = pi->tx_user_target[0];
> if (override != NULL)
> @@ -1323,7 +1323,7 @@ void wlc_phy_txpower_target_set(struct brcms_phy_pub *ppi,
> struct txpwr_limits *txpwr)
> {
> bool mac_enabled = false;
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> memcpy(&pi->tx_user_target[TXP_FIRST_CCK],
> &txpwr->cck[0], BRCMS_NUM_RATES_CCK);
> @@ -1371,7 +1371,7 @@ void wlc_phy_txpower_target_set(struct brcms_phy_pub *ppi,
>
> int wlc_phy_txpower_set(struct brcms_phy_pub *ppi, uint qdbm, bool override)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> int i;
>
> if (qdbm > 127)
> @@ -1407,7 +1407,7 @@ void
> wlc_phy_txpower_sromlimit(struct brcms_phy_pub *ppi, uint channel, u8 *min_pwr,
> u8 *max_pwr, int txp_rate_idx)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> uint i;
>
> *min_pwr = pi->min_txpower * BRCMS_TXPWR_DB_FACTOR;
> @@ -1456,7 +1456,7 @@ void
> wlc_phy_txpower_sromlimit_max_get(struct brcms_phy_pub *ppi, uint chan,
> u8 *max_txpwr, u8 *min_txpwr)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> u8 tx_pwr_max = 0;
> u8 tx_pwr_min = 255;
> u8 max_num_rate;
> @@ -1493,14 +1493,14 @@ wlc_phy_txpower_boardlimit_band(struct brcms_phy_pub *ppi, uint bandunit,
>
> u8 wlc_phy_txpower_get_target_min(struct brcms_phy_pub *ppi)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> return pi->tx_power_min;
> }
>
> u8 wlc_phy_txpower_get_target_max(struct brcms_phy_pub *ppi)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> return pi->tx_power_max;
> }
> @@ -1812,21 +1812,21 @@ wlc_phy_txpower_reg_limit_calc(struct brcms_phy *pi, struct txpwr_limits *txpwr,
>
> void wlc_phy_txpwr_percent_set(struct brcms_phy_pub *ppi, u8 txpwr_percent)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> pi->txpwr_percent = txpwr_percent;
> }
>
> void wlc_phy_machwcap_set(struct brcms_phy_pub *ppi, u32 machwcap)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> pi->sh->machwcap = machwcap;
> }
>
> void wlc_phy_runbist_config(struct brcms_phy_pub *ppi, bool start_end)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> u16 rxc;
> rxc = 0;
>
> @@ -1857,7 +1857,7 @@ void
> wlc_phy_txpower_limit_set(struct brcms_phy_pub *ppi, struct txpwr_limits *txpwr,
> u16 chanspec)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> wlc_phy_txpower_reg_limit_calc(pi, txpwr, chanspec);
>
> @@ -1881,14 +1881,14 @@ wlc_phy_txpower_limit_set(struct brcms_phy_pub *ppi, struct txpwr_limits *txpwr,
>
> void wlc_phy_ofdm_rateset_war(struct brcms_phy_pub *pih, bool war)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> pi->ofdm_rateset_war = war;
> }
>
> void wlc_phy_bf_preempt_enable(struct brcms_phy_pub *pih, bool bf_preempt)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> pi->bf_preempt_4306 = bf_preempt;
> }
> @@ -1945,7 +1945,7 @@ void wlc_phy_txpower_update_shm(struct brcms_phy *pi)
>
> bool wlc_phy_txpower_hw_ctrl_get(struct brcms_phy_pub *ppi)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> if (ISNPHY(pi))
> return pi->nphy_txpwrctrl;
> @@ -1955,7 +1955,7 @@ bool wlc_phy_txpower_hw_ctrl_get(struct brcms_phy_pub *ppi)
>
> void wlc_phy_txpower_hw_ctrl_set(struct brcms_phy_pub *ppi, bool hwpwrctrl)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> bool suspend;
>
> if (!pi->hwpwrctrl_capable)
> @@ -2038,7 +2038,7 @@ void
> wlc_phy_txpower_get_current(struct brcms_phy_pub *ppi, struct tx_power *power,
> uint channel)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> uint rate, num_rates;
> u8 min_pwr, max_pwr;
>
> @@ -2136,21 +2136,21 @@ wlc_phy_txpower_get_current(struct brcms_phy_pub *ppi, struct tx_power *power,
>
> void wlc_phy_antsel_type_set(struct brcms_phy_pub *ppi, u8 antsel_type)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> pi->antsel_type = antsel_type;
> }
>
> bool wlc_phy_test_ison(struct brcms_phy_pub *ppi)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> return pi->phytest_on;
> }
>
> void wlc_phy_ant_rxdiv_set(struct brcms_phy_pub *ppi, u8 val)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> bool suspend;
>
> pi->sh->rx_antdiv = val;
> @@ -2283,7 +2283,7 @@ static s8 wlc_phy_noise_read_shmem(struct brcms_phy *pi)
>
> void wlc_phy_noise_sample_intr(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> u16 jssi_aux;
> u8 channel = 0;
> s8 noise_dbm = PHY_NOISE_FIXED_VAL_NPHY;
> @@ -2339,7 +2339,7 @@ void wlc_phy_noise_sample_intr(struct brcms_phy_pub *pih)
> static void
> wlc_phy_noise_sample_request(struct brcms_phy_pub *pih, u8 reason, u8 ch)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> s8 noise_dbm = PHY_NOISE_FIXED_VAL_NPHY;
> bool sampling_in_progress = (pi->phynoise_state != 0);
> bool wait_for_intr = true;
> @@ -2531,7 +2531,7 @@ int wlc_phy_rssi_compute(struct brcms_phy_pub *pih,
> {
> int rssi = rxh->PhyRxStatus_1 & PRXS1_JSSI_MASK;
> uint radioid = pih->radioid;
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> if ((pi->sh->corerev >= 11)
> && !(rxh->RxStatus2 & RXS_PHYRXST_VALID)) {
> @@ -2591,7 +2591,7 @@ void wlc_phy_set_deaf(struct brcms_phy_pub *ppi, bool user_flag)
>
> void wlc_phy_watchdog(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> bool delay_phy_cal = false;
> pi->sh->now++;
>
> @@ -2651,7 +2651,7 @@ void wlc_phy_watchdog(struct brcms_phy_pub *pih)
>
> void wlc_phy_BSSinit(struct brcms_phy_pub *pih, bool bonlyap, int rssi)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> uint i;
> uint k;
>
> @@ -2711,7 +2711,7 @@ void wlc_phy_cal_perical(struct brcms_phy_pub *pih, u8 reason)
> s16 nphy_currtemp = 0;
> s16 delta_temp = 0;
> bool do_periodic_cal = true;
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> if (!ISNPHY(pi))
> return;
> @@ -2804,7 +2804,7 @@ u8 wlc_phy_nbits(s32 value)
>
> void wlc_phy_stf_chain_init(struct brcms_phy_pub *pih, u8 txchain, u8 rxchain)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> pi->sh->hw_phytxchain = txchain;
> pi->sh->hw_phyrxchain = rxchain;
> @@ -2815,7 +2815,7 @@ void wlc_phy_stf_chain_init(struct brcms_phy_pub *pih, u8 txchain, u8 rxchain)
>
> void wlc_phy_stf_chain_set(struct brcms_phy_pub *pih, u8 txchain, u8 rxchain)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> pi->sh->phytxchain = txchain;
>
> @@ -2827,7 +2827,7 @@ void wlc_phy_stf_chain_set(struct brcms_phy_pub *pih, u8 txchain, u8 rxchain)
>
> void wlc_phy_stf_chain_get(struct brcms_phy_pub *pih, u8 *txchain, u8 *rxchain)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> *txchain = pi->sh->phytxchain;
> *rxchain = pi->sh->phyrxchain;
> @@ -2837,7 +2837,7 @@ u8 wlc_phy_stf_chain_active_get(struct brcms_phy_pub *pih)
> {
> s16 nphy_currtemp;
> u8 active_bitmap;
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> active_bitmap = (pi->phy_txcore_heatedup) ? 0x31 : 0x33;
>
> @@ -2867,7 +2867,7 @@ u8 wlc_phy_stf_chain_active_get(struct brcms_phy_pub *pih)
>
> s8 wlc_phy_stf_ssmode_get(struct brcms_phy_pub *pih, u16 chanspec)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> u8 siso_mcs_id, cdd_mcs_id;
>
> siso_mcs_id =
> @@ -2944,7 +2944,7 @@ s8 wlc_phy_upd_rssi_offset(struct brcms_phy *pi, s8 rssi, u16 chanspec)
>
> bool wlc_phy_txpower_ipa_ison(struct brcms_phy_pub *ppi)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> if (ISNPHY(pi))
> return wlc_phy_n_txpower_ipa_ison(pi);
> diff --git a/drivers/net/wireless/brcm80211/brcmsmac/phy/phy_lcn.c b/drivers/net/wireless/brcm80211/brcmsmac/phy/phy_lcn.c
> index b2d6d6d..5f13662 100644
> --- a/drivers/net/wireless/brcm80211/brcmsmac/phy/phy_lcn.c
> +++ b/drivers/net/wireless/brcm80211/brcmsmac/phy/phy_lcn.c
> @@ -2865,7 +2865,7 @@ static void wlc_lcnphy_idle_tssi_est(struct brcms_phy_pub *ppi)
> {
> bool suspend, tx_gain_override_old;
> struct lcnphy_txgains old_gains;
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> u16 idleTssi, idleTssi0_2C, idleTssi0_OB, idleTssi0_regvalue_OB,
> idleTssi0_regvalue_2C;
> u16 SAVE_txpwrctrl = wlc_lcnphy_get_tx_pwr_ctrl(pi);
> @@ -3084,7 +3084,7 @@ static void wlc_lcnphy_tx_pwr_ctrl_init(struct brcms_phy_pub *ppi)
> s32 a1, b0, b1;
> s32 tssi, pwr, maxtargetpwr, mintargetpwr;
> bool suspend;
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
>
> suspend = (0 == (bcma_read32(pi->d11core, D11REGOFFS(maccontrol)) &
> MCTL_EN_MAC));
> @@ -4348,7 +4348,7 @@ void wlc_lcnphy_tx_power_adjustment(struct brcms_phy_pub *ppi)
> {
> s8 index;
> u16 index2;
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> struct brcms_phy_lcnphy *pi_lcn = pi->u.pi_lcnphy;
> u16 SAVE_txpwrctrl = wlc_lcnphy_get_tx_pwr_ctrl(pi);
> if (wlc_lcnphy_tempsense_based_pwr_ctrl_enabled(pi) &&
> diff --git a/drivers/net/wireless/brcm80211/brcmsmac/phy/phy_n.c b/drivers/net/wireless/brcm80211/brcmsmac/phy/phy_n.c
> index 93869e8..084f18f 100644
> --- a/drivers/net/wireless/brcm80211/brcmsmac/phy/phy_n.c
> +++ b/drivers/net/wireless/brcm80211/brcmsmac/phy/phy_n.c
> @@ -14121,7 +14121,7 @@ static u8 ant_sw_ctrl_tbl_rev8_2057v7_core1[] = {
>
> bool wlc_phy_bist_check_phy(struct brcms_phy_pub *pih)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> u32 phybist0, phybist1, phybist2, phybist3, phybist4;
>
> if (NREV_GE(pi->pubpi.phy_rev, 16))
> @@ -19734,7 +19734,7 @@ void wlc_phy_rxcore_setstate_nphy(struct brcms_phy_pub *pih, u8 rxcore_bitmask)
> u16 regval;
> u16 tbl_buf[16];
> uint i;
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
> u16 tbl_opcode;
> bool suspend;
>
> @@ -19812,7 +19812,7 @@ void wlc_phy_rxcore_setstate_nphy(struct brcms_phy_pub *pih, u8 rxcore_bitmask)
> u8 wlc_phy_rxcore_getstate_nphy(struct brcms_phy_pub *pih)
> {
> u16 regval, rxen_bits;
> - struct brcms_phy *pi = (struct brcms_phy *) pih;
> + struct brcms_phy *pi = container_of(pih, struct brcms_phy, pubpi_ro);
>
> regval = read_phy_reg(pi, 0xa2);
> rxen_bits = (regval >> 4) & 0xf;
> @@ -21342,7 +21342,7 @@ void wlc_phy_chanspec_set_nphy(struct brcms_phy *pi, u16 chanspec)
>
> void wlc_phy_antsel_init(struct brcms_phy_pub *ppi, bool lut_init)
> {
> - struct brcms_phy *pi = (struct brcms_phy *) ppi;
> + struct brcms_phy *pi = container_of(ppi, struct brcms_phy, pubpi_ro);
> u16 mask = 0xfc00;
> u32 mc = 0;
>
> --
> 1.9.1
>
>
--
John W. Linville Someday the world will need a hero, and you
linville-2XuSBdqkA4R54TAoqtyWWQ@public.gmane.org might be all we have. Be ready.
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [net-next PATCH 1/1 V4] qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE
From: Eric Dumazet @ 2014-09-26 18:06 UTC (permalink / raw)
To: Dave Taht
Cc: Tom Herbert, Jesper Dangaard Brouer, Linux Netdev List,
David S. Miller, Alexander Duyck,
Toke Høiland-Jørgensen, Florian Westphal,
Jamal Hadi Salim, John Fastabend, Daniel Borkmann,
Hannes Frederic Sowa
In-Reply-To: <CAA93jw7Y5-5qwTcoqzRAv-DeRmVUA=hTX6b74hQPM=VmpNWQZA@mail.gmail.com>
On Wed, 2014-09-24 at 19:58 -0700, Dave Taht wrote:
> I have long hoped that the actual BQL limit in play would feed into
> TCP small queues when there are a lot of flows to make each tcp
> "small" queue gradually smaller...
Yep, but what do you do if you have 64 TX queues, each one having
different BQL limits ? This looks like a NP problem when you also have
millions of tcp flows.
Basically BQL limit is more a sign of local host being able to drain TX
completions fast or not.
Here at Google, we have a very tiny queue of at most 2 packets per flow,
and srtt being in usec units gives us a very dynamic signal, based on
end to end measures.
I am experimenting moving TSQ processing from tasklet to workqueue,
because we need faster response to {soft}irq when we so often enter
TCP stack.
^ permalink raw reply
* Re: [Xen-devel] [PATCH] xen/xenbus: Use 'void' instead of 'int' for the return of xenbus_switch_state()
From: David Vrabel @ 2014-09-26 18:07 UTC (permalink / raw)
To: Chen Gang, David Vrabel, konrad.wilk, ian.campbell, wei.liu2,
boris.ostrovsky, bhelgaas, jgross, yongjun_wei, mukesh.rathor
Cc: xen-devel, linux-pci, linux-kernel@vger.kernel.org, linux-scsi,
netdev@vger.kernel.org
In-Reply-To: <5425961A.5000604@gmail.com>
On 26/09/14 17:36, Chen Gang wrote:
> When xenbus_switch_state() fails, it will call xenbus_switch_fatal()
> internally, so need not return any status value, then use 'void' instead
> of 'int' for xenbus_switch_state() and __xenbus_switch_state().
>
> Also need be sure that all callers which check the return value must let
> 'err' be 0.
I've rewritten the commit message as:
xen/xenbus: don't return errors from xenbus_switch_state()
Most users of xenbus_switch_state() weren't handling the failure of
xenbus_switch_state() correctly. They either called
xenbus_dev_fatal() (which xenbus_switch_state() has effectively
already tried to do), or ignored errors.
xenbus_switch_state() may fail because:
a) The device is being unplugged by the toolstack. The device will
shortly be removed and the error does not need to be handled.
b) Xenstore is broken. There isn't much the driver can do in this
case since xenstore is required to signal failure to the toolstack.
So, don't report any errors from xenbus_switch_state() which removes
some unnecessary error handling in some of the drivers.
I'd appreciate a review from some of the other front/backend driver
maintainers on whether this is sensible reasoning.
David
>
> And also need change the related comments for xenbus_switch_state().
>
> Signed-off-by: Chen Gang <gang.chen.5i5j@gmail.com>
> ---
> drivers/block/xen-blkback/xenbus.c | 9 ++-------
> drivers/net/xen-netback/xenbus.c | 5 +----
> drivers/pci/xen-pcifront.c | 15 ++++++---------
> drivers/xen/xen-pciback/xenbus.c | 21 ++++-----------------
> drivers/xen/xen-scsiback.c | 5 +----
> drivers/xen/xenbus/xenbus_client.c | 16 ++++++++--------
> include/xen/xenbus.h | 3 ++-
> 7 files changed, 24 insertions(+), 50 deletions(-)
>
> diff --git a/drivers/block/xen-blkback/xenbus.c b/drivers/block/xen-blkback/xenbus.c
> index 3a8b810..fdcc584 100644
> --- a/drivers/block/xen-blkback/xenbus.c
> +++ b/drivers/block/xen-blkback/xenbus.c
> @@ -587,9 +587,7 @@ static int xen_blkbk_probe(struct xenbus_device *dev,
> if (err)
> goto fail;
>
> - err = xenbus_switch_state(dev, XenbusStateInitWait);
> - if (err)
> - goto fail;
> + xenbus_switch_state(dev, XenbusStateInitWait);
>
> return 0;
>
> @@ -837,10 +835,7 @@ again:
> if (err)
> xenbus_dev_fatal(dev, err, "ending transaction");
>
> - err = xenbus_switch_state(dev, XenbusStateConnected);
> - if (err)
> - xenbus_dev_fatal(dev, err, "%s: switching to Connected state",
> - dev->nodename);
> + xenbus_switch_state(dev, XenbusStateConnected);
>
> return;
> abort:
> diff --git a/drivers/net/xen-netback/xenbus.c b/drivers/net/xen-netback/xenbus.c
> index 9c47b89..b5c3d47 100644
> --- a/drivers/net/xen-netback/xenbus.c
> +++ b/drivers/net/xen-netback/xenbus.c
> @@ -337,10 +337,7 @@ static int netback_probe(struct xenbus_device *dev,
> if (err)
> pr_debug("Error writing multi-queue-max-queues\n");
>
> - err = xenbus_switch_state(dev, XenbusStateInitWait);
> - if (err)
> - goto fail;
> -
> + xenbus_switch_state(dev, XenbusStateInitWait);
> be->state = XenbusStateInitWait;
>
> /* This kicks hotplug scripts, so do it immediately. */
> diff --git a/drivers/pci/xen-pcifront.c b/drivers/pci/xen-pcifront.c
> index 53df39a..c1d73b7 100644
> --- a/drivers/pci/xen-pcifront.c
> +++ b/drivers/pci/xen-pcifront.c
> @@ -901,18 +901,16 @@ static int pcifront_try_connect(struct pcifront_device *pdev)
> }
> }
>
> - err = xenbus_switch_state(pdev->xdev, XenbusStateConnected);
> -
> + xenbus_switch_state(pdev->xdev, XenbusStateConnected);
> + return 0;
> out:
> return err;
> }
>
> static int pcifront_try_disconnect(struct pcifront_device *pdev)
> {
> - int err = 0;
> enum xenbus_state prev_state;
>
> -
> prev_state = xenbus_read_driver_state(pdev->xdev->nodename);
>
> if (prev_state >= XenbusStateClosing)
> @@ -923,11 +921,10 @@ static int pcifront_try_disconnect(struct pcifront_device *pdev)
> pcifront_disconnect(pdev);
> }
>
> - err = xenbus_switch_state(pdev->xdev, XenbusStateClosed);
> + xenbus_switch_state(pdev->xdev, XenbusStateClosed);
>
> out:
> -
> - return err;
> + return 0;
> }
>
> static int pcifront_attach_devices(struct pcifront_device *pdev)
> @@ -1060,8 +1057,8 @@ static int pcifront_detach_devices(struct pcifront_device *pdev)
> domain, bus, slot, func);
> }
>
> - err = xenbus_switch_state(pdev->xdev, XenbusStateReconfiguring);
> -
> + xenbus_switch_state(pdev->xdev, XenbusStateReconfiguring);
> + return 0;
> out:
> return err;
> }
> diff --git a/drivers/xen/xen-pciback/xenbus.c b/drivers/xen/xen-pciback/xenbus.c
> index c214daa..630a215 100644
> --- a/drivers/xen/xen-pciback/xenbus.c
> +++ b/drivers/xen/xen-pciback/xenbus.c
> @@ -184,10 +184,7 @@ static int xen_pcibk_attach(struct xen_pcibk_device *pdev)
>
> dev_dbg(&pdev->xdev->dev, "Connecting...\n");
>
> - err = xenbus_switch_state(pdev->xdev, XenbusStateConnected);
> - if (err)
> - xenbus_dev_fatal(pdev->xdev, err,
> - "Error switching to connected state!");
> + xenbus_switch_state(pdev->xdev, XenbusStateConnected);
>
> dev_dbg(&pdev->xdev->dev, "Connected? %d\n", err);
> out:
> @@ -500,12 +497,7 @@ static int xen_pcibk_reconfigure(struct xen_pcibk_device *pdev)
> }
> }
>
> - err = xenbus_switch_state(pdev->xdev, XenbusStateReconfigured);
> - if (err) {
> - xenbus_dev_fatal(pdev->xdev, err,
> - "Error switching to reconfigured state!");
> - goto out;
> - }
> + xenbus_switch_state(pdev->xdev, XenbusStateReconfigured);
>
> out:
> mutex_unlock(&pdev->dev_lock);
> @@ -640,10 +632,7 @@ static int xen_pcibk_setup_backend(struct xen_pcibk_device *pdev)
> goto out;
> }
>
> - err = xenbus_switch_state(pdev->xdev, XenbusStateInitialised);
> - if (err)
> - xenbus_dev_fatal(pdev->xdev, err,
> - "Error switching to initialised state!");
> + xenbus_switch_state(pdev->xdev, XenbusStateInitialised);
>
> out:
> mutex_unlock(&pdev->dev_lock);
> @@ -683,9 +672,7 @@ static int xen_pcibk_xenbus_probe(struct xenbus_device *dev,
> }
>
> /* wait for xend to configure us */
> - err = xenbus_switch_state(dev, XenbusStateInitWait);
> - if (err)
> - goto out;
> + xenbus_switch_state(dev, XenbusStateInitWait);
>
> /* watch the backend node for backend configuration information */
> err = xenbus_watch_path(dev, dev->nodename, &pdev->be_watch,
> diff --git a/drivers/xen/xen-scsiback.c b/drivers/xen/xen-scsiback.c
> index ad11258..847bc9c 100644
> --- a/drivers/xen/xen-scsiback.c
> +++ b/drivers/xen/xen-scsiback.c
> @@ -1225,10 +1225,7 @@ static int scsiback_probe(struct xenbus_device *dev,
> if (err)
> xenbus_dev_error(dev, err, "writing feature-sg-grant");
>
> - err = xenbus_switch_state(dev, XenbusStateInitWait);
> - if (err)
> - goto fail;
> -
> + xenbus_switch_state(dev, XenbusStateInitWait);
> return 0;
>
> fail:
> diff --git a/drivers/xen/xenbus/xenbus_client.c b/drivers/xen/xenbus/xenbus_client.c
> index ca74410..e2bcd1d 100644
> --- a/drivers/xen/xenbus/xenbus_client.c
> +++ b/drivers/xen/xenbus/xenbus_client.c
> @@ -166,7 +166,7 @@ EXPORT_SYMBOL_GPL(xenbus_watch_pathfmt);
> static void xenbus_switch_fatal(struct xenbus_device *, int, int,
> const char *, ...);
>
> -static int
> +static void
> __xenbus_switch_state(struct xenbus_device *dev,
> enum xenbus_state state, int depth)
> {
> @@ -188,7 +188,7 @@ __xenbus_switch_state(struct xenbus_device *dev,
> int err, abort;
>
> if (state == dev->state)
> - return 0;
> + return;
>
> again:
> abort = 1;
> @@ -196,7 +196,7 @@ again:
> err = xenbus_transaction_start(&xbt);
> if (err) {
> xenbus_switch_fatal(dev, depth, err, "starting transaction");
> - return 0;
> + return;
> }
>
> err = xenbus_scanf(xbt, dev->nodename, "state", "%d", ¤t_state);
> @@ -219,7 +219,7 @@ abort:
> } else
> dev->state = state;
>
> - return 0;
> + return;
> }
>
> /**
> @@ -228,12 +228,12 @@ abort:
> * @state: new state
> *
> * Advertise in the store a change of the given driver to the given new_state.
> - * Return 0 on success, or -errno on error. On error, the device will switch
> - * to XenbusStateClosing, and the error will be saved in the store.
> + * When failure occurs, it will call xenbus_switch_fatal() internally, so need
> + * not return value to notify upper caller.
> */
> -int xenbus_switch_state(struct xenbus_device *dev, enum xenbus_state state)
> +void xenbus_switch_state(struct xenbus_device *dev, enum xenbus_state state)
> {
> - return __xenbus_switch_state(dev, state, 0);
> + __xenbus_switch_state(dev, state, 0);
> }
>
> EXPORT_SYMBOL_GPL(xenbus_switch_state);
> diff --git a/include/xen/xenbus.h b/include/xen/xenbus.h
> index 0324c6d..587c53d 100644
> --- a/include/xen/xenbus.h
> +++ b/include/xen/xenbus.h
> @@ -195,7 +195,8 @@ int xenbus_watch_pathfmt(struct xenbus_device *dev, struct xenbus_watch *watch,
> const char **, unsigned int),
> const char *pathfmt, ...);
>
> -int xenbus_switch_state(struct xenbus_device *dev, enum xenbus_state new_state);
> +void xenbus_switch_state(struct xenbus_device *dev,
> + enum xenbus_state new_state);
> int xenbus_grant_ring(struct xenbus_device *dev, unsigned long ring_mfn);
> int xenbus_map_ring_valloc(struct xenbus_device *dev,
> int gnt_ref, void **vaddr);
>
^ permalink raw reply
* Re: [RFC PATCH net-next v2 0/5] netns: allow to identify peer netns
From: Eric W. Biederman @ 2014-09-26 18:10 UTC (permalink / raw)
To: Nicolas Dichtel
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA,
linux-kernel-u79uwXL29TY76Z2rM5mHXA, luto-kltTT9wpgjJwATOyAt5JVQ,
stephen-OTpzqLSitTUnbdJkjeBofR2eb7JE58TQ, Cong Wang,
linux-api-u79uwXL29TY76Z2rM5mHXA,
akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b,
davem-fT/PcQaiUtIeIZ0/mPfg9Q
In-Reply-To: <1411478430-4989-1-git-send-email-nicolas.dichtel-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>
Nicolas Dichtel <nicolas.dichtel-pdR9zngts4EAvxtiuMwx3w@public.gmane.org> writes:
> The goal of this serie is to be able to multicast netlink messages with an
> attribute that identify a peer netns.
> This is needed by the userland to interpret some informations contained in
> netlink messages (like IFLA_LINK value, but also some other attributes in case
> of x-netns netdevice (see also
> http://thread.gmane.org/gmane.linux.network/315933/focus=316064 and
> http://thread.gmane.org/gmane.linux.kernel.containers/28301/focus=4239)).
I want say that the problem addressed by patch 3/5 of this series is a
fundamentally valid problem. We have network objects spanning network
namespaces and it would be very nice to be able to talk about them in
netlink, and file descriptors are too local and argubably too heavy
weight for netlink quires and especially for netlink broadcast messages.
Furthermore the concept of ineternal concept of peernet2id seems valid.
However what you do not address is a way for CRIU (aka process
migration) to be able to restore these ids after process migration.
Going farther it looks like you are actively breaking process migration
at this time, making this set of patches a no-go.
When adding a new form of namespace id CRIU patches are just about
as necessary as iproute patches.
> Ids are stored in the parent user namespace. These ids are valid only inside
> this user namespace. The user can retrieve these ids via a new netlink messages,
> but only if peer netns are in the same user namespace.
That does not describe what you have actually implemented in the
patches.
I see two ways to go with this.
- A per network namespace table to that you can store ids for ``peer''
network namespaces. The table would need to be populated manually by
the likes of ip netns add.
That flips the order of assignment and makes this idea solid.
Unfortunately in the case of a fully referencing mesh of N network
namespaces such a mesh winds up taking O(N^2) space, which seems
undesirable.
- Add a netlink attribute that says this network element is in a peer
network namespace.
Add a unicast query message that let's you ask if the remote
end of a tunnel is in a network namespace specified by file
descriptor.
I personally lean towards the second version as it is fundamentally
simpler, and generally scales better, and the visibility controls are
the existing visibility controls. The only downside is it requires
a query after receiving a netlink broadcast message for the times that
we care.
Eric
^ permalink raw reply
* [net-next PATCH 1/4] net: sched: make bstats per cpu and estimator RCU safe
From: John Fastabend @ 2014-09-26 18:13 UTC (permalink / raw)
To: xiyou.wangcong, jhs, eric.dumazet, davem; +Cc: netdev
From: John Fastabend <john.fastabend@gmail.com>
In order to run qdisc's without locking statistics and estimators
need to be handled correctly.
To resolve bstats make the statistics per cpu. And because this is
only needed for qdiscs that are running without locks which is not
the case for most qdiscs in the near future only create percpu
stats when qdiscs set the TCQ_F_CPUSTATS flag.
Next because estimators use the bstats to calculate packets per
second and bytes per second the estimator code paths are updated
to use the per cpu statistics.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
---
include/net/gen_stats.h | 11 +++++++++
include/net/sch_generic.h | 22 +++++++++++++++++-
net/core/gen_estimator.c | 29 +++++++++++++++---------
net/core/gen_stats.c | 53 ++++++++++++++++++++++++++++++++++++++++----
net/netfilter/xt_RATEEST.c | 2 +-
net/sched/act_api.c | 5 ++--
net/sched/act_police.c | 2 +-
net/sched/sch_api.c | 29 +++++++++++++++++++-----
net/sched/sch_atm.c | 2 +-
net/sched/sch_cbq.c | 7 +++---
net/sched/sch_drr.c | 7 +++---
net/sched/sch_generic.c | 3 ++
net/sched/sch_hfsc.c | 13 +++++++----
net/sched/sch_htb.c | 12 +++++++---
net/sched/sch_mq.c | 6 ++---
net/sched/sch_mqprio.c | 8 +++----
net/sched/sch_multiq.c | 2 +-
net/sched/sch_prio.c | 2 +-
net/sched/sch_qfq.c | 8 ++++---
19 files changed, 168 insertions(+), 55 deletions(-)
diff --git a/include/net/gen_stats.h b/include/net/gen_stats.h
index ea4271d..ce3c128 100644
--- a/include/net/gen_stats.h
+++ b/include/net/gen_stats.h
@@ -6,6 +6,11 @@
#include <linux/rtnetlink.h>
#include <linux/pkt_sched.h>
+struct gnet_stats_basic_cpu {
+ struct gnet_stats_basic_packed bstats;
+ struct u64_stats_sync syncp;
+};
+
struct gnet_dump {
spinlock_t * lock;
struct sk_buff * skb;
@@ -27,7 +32,11 @@ int gnet_stats_start_copy_compat(struct sk_buff *skb, int type,
spinlock_t *lock, struct gnet_dump *d);
int gnet_stats_copy_basic(struct gnet_dump *d,
+ struct gnet_stats_basic_cpu __percpu *cpu,
struct gnet_stats_basic_packed *b);
+void __gnet_stats_copy_basic(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu,
+ struct gnet_stats_basic_packed *b);
int gnet_stats_copy_rate_est(struct gnet_dump *d,
const struct gnet_stats_basic_packed *b,
struct gnet_stats_rate_est64 *r);
@@ -37,11 +46,13 @@ int gnet_stats_copy_app(struct gnet_dump *d, void *st, int len);
int gnet_stats_finish_copy(struct gnet_dump *d);
int gen_new_estimator(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats,
struct gnet_stats_rate_est64 *rate_est,
spinlock_t *stats_lock, struct nlattr *opt);
void gen_kill_estimator(struct gnet_stats_basic_packed *bstats,
struct gnet_stats_rate_est64 *rate_est);
int gen_replace_estimator(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats,
struct gnet_stats_rate_est64 *rate_est,
spinlock_t *stats_lock, struct nlattr *opt);
bool gen_estimator_active(const struct gnet_stats_basic_packed *bstats,
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index e65b8e0..4b93511 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -6,6 +6,7 @@
#include <linux/rcupdate.h>
#include <linux/pkt_sched.h>
#include <linux/pkt_cls.h>
+#include <linux/percpu.h>
#include <net/gen_stats.h>
#include <net/rtnetlink.h>
@@ -58,6 +59,7 @@ struct Qdisc {
* multiqueue device.
*/
#define TCQ_F_WARN_NONWC (1 << 16)
+#define TCQ_F_CPUSTATS 0x20 /* run using percpu statistics */
u32 limit;
const struct Qdisc_ops *ops;
struct qdisc_size_table __rcu *stab;
@@ -83,7 +85,10 @@ struct Qdisc {
*/
unsigned long state;
struct sk_buff_head q;
- struct gnet_stats_basic_packed bstats;
+ union {
+ struct gnet_stats_basic_packed bstats;
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats;
+ } __packed;
unsigned int __state;
struct gnet_stats_queue qstats;
struct rcu_head rcu_head;
@@ -487,6 +492,10 @@ static inline int qdisc_enqueue_root(struct sk_buff *skb, struct Qdisc *sch)
return qdisc_enqueue(skb, sch) & NET_XMIT_MASK;
}
+static inline bool qdisc_is_percpu_stats(const struct Qdisc *q)
+{
+ return q->flags & TCQ_F_CPUSTATS;
+}
static inline void bstats_update(struct gnet_stats_basic_packed *bstats,
const struct sk_buff *skb)
@@ -495,6 +504,17 @@ static inline void bstats_update(struct gnet_stats_basic_packed *bstats,
bstats->packets += skb_is_gso(skb) ? skb_shinfo(skb)->gso_segs : 1;
}
+static inline void qdisc_bstats_update_cpu(struct Qdisc *sch,
+ const struct sk_buff *skb)
+{
+ struct gnet_stats_basic_cpu *bstats =
+ this_cpu_ptr(sch->cpu_bstats);
+
+ u64_stats_update_begin(&bstats->syncp);
+ bstats_update(&bstats->bstats, skb);
+ u64_stats_update_end(&bstats->syncp);
+}
+
static inline void qdisc_bstats_update(struct Qdisc *sch,
const struct sk_buff *skb)
{
diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c
index 9d33dff..9dfb88a 100644
--- a/net/core/gen_estimator.c
+++ b/net/core/gen_estimator.c
@@ -91,6 +91,8 @@ struct gen_estimator
u32 avpps;
struct rcu_head e_rcu;
struct rb_node node;
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats;
+ struct rcu_head head;
};
struct gen_estimator_head
@@ -115,9 +117,8 @@ static void est_timer(unsigned long arg)
rcu_read_lock();
list_for_each_entry_rcu(e, &elist[idx].list, list) {
- u64 nbytes;
+ struct gnet_stats_basic_packed b = {0};
u64 brate;
- u32 npackets;
u32 rate;
spin_lock(e->stats_lock);
@@ -125,15 +126,15 @@ static void est_timer(unsigned long arg)
if (e->bstats == NULL)
goto skip;
- nbytes = e->bstats->bytes;
- npackets = e->bstats->packets;
- brate = (nbytes - e->last_bytes)<<(7 - idx);
- e->last_bytes = nbytes;
+ __gnet_stats_copy_basic(&b, e->cpu_bstats, e->bstats);
+
+ brate = (b.bytes - e->last_bytes)<<(7 - idx);
+ e->last_bytes = b.bytes;
e->avbps += (brate >> e->ewma_log) - (e->avbps >> e->ewma_log);
e->rate_est->bps = (e->avbps+0xF)>>5;
- rate = (npackets - e->last_packets)<<(12 - idx);
- e->last_packets = npackets;
+ rate = (b.packets - e->last_packets)<<(12 - idx);
+ e->last_packets = b.packets;
e->avpps += (rate >> e->ewma_log) - (e->avpps >> e->ewma_log);
e->rate_est->pps = (e->avpps+0x1FF)>>10;
skip:
@@ -203,12 +204,14 @@ struct gen_estimator *gen_find_node(const struct gnet_stats_basic_packed *bstats
*
*/
int gen_new_estimator(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats,
struct gnet_stats_rate_est64 *rate_est,
spinlock_t *stats_lock,
struct nlattr *opt)
{
struct gen_estimator *est;
struct gnet_estimator *parm = nla_data(opt);
+ struct gnet_stats_basic_packed b = {0};
int idx;
if (nla_len(opt) < sizeof(*parm))
@@ -221,15 +224,18 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats,
if (est == NULL)
return -ENOBUFS;
+ __gnet_stats_copy_basic(&b, cpu_bstats, bstats);
+
idx = parm->interval + 2;
est->bstats = bstats;
est->rate_est = rate_est;
est->stats_lock = stats_lock;
est->ewma_log = parm->ewma_log;
- est->last_bytes = bstats->bytes;
+ est->last_bytes = b.bytes;
est->avbps = rate_est->bps<<5;
- est->last_packets = bstats->packets;
+ est->last_packets = b.packets;
est->avpps = rate_est->pps<<10;
+ est->cpu_bstats = cpu_bstats;
spin_lock_bh(&est_tree_lock);
if (!elist[idx].timer.function) {
@@ -290,11 +296,12 @@ EXPORT_SYMBOL(gen_kill_estimator);
* Returns 0 on success or a negative error code.
*/
int gen_replace_estimator(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats,
struct gnet_stats_rate_est64 *rate_est,
spinlock_t *stats_lock, struct nlattr *opt)
{
gen_kill_estimator(bstats, rate_est);
- return gen_new_estimator(bstats, rate_est, stats_lock, opt);
+ return gen_new_estimator(bstats, cpu_bstats, rate_est, stats_lock, opt);
}
EXPORT_SYMBOL(gen_replace_estimator);
diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c
index 2ddbce4..5ff8e80 100644
--- a/net/core/gen_stats.c
+++ b/net/core/gen_stats.c
@@ -97,6 +97,43 @@ gnet_stats_start_copy(struct sk_buff *skb, int type, spinlock_t *lock,
}
EXPORT_SYMBOL(gnet_stats_start_copy);
+static void
+__gnet_stats_copy_basic_cpu(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu)
+{
+ int i;
+
+ for_each_possible_cpu(i) {
+ struct gnet_stats_basic_cpu *bcpu = per_cpu_ptr(cpu, i);
+ unsigned int start;
+ __u64 bytes;
+ __u32 packets;
+
+ do {
+ start = u64_stats_fetch_begin_irq(&bcpu->syncp);
+ bytes = bcpu->bstats.bytes;
+ packets = bcpu->bstats.packets;
+ } while (u64_stats_fetch_retry_irq(&bcpu->syncp, start));
+
+ bstats->bytes += bcpu->bstats.bytes;
+ bstats->packets += bcpu->bstats.packets;
+ }
+}
+
+void
+__gnet_stats_copy_basic(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu,
+ struct gnet_stats_basic_packed *b)
+{
+ if (cpu) {
+ __gnet_stats_copy_basic_cpu(bstats, cpu);
+ } else {
+ bstats->bytes = b->bytes;
+ bstats->packets = b->packets;
+ }
+}
+EXPORT_SYMBOL(__gnet_stats_copy_basic);
+
/**
* gnet_stats_copy_basic - copy basic statistics into statistic TLV
* @d: dumping handle
@@ -109,19 +146,25 @@ EXPORT_SYMBOL(gnet_stats_start_copy);
* if the room in the socket buffer was not sufficient.
*/
int
-gnet_stats_copy_basic(struct gnet_dump *d, struct gnet_stats_basic_packed *b)
+gnet_stats_copy_basic(struct gnet_dump *d,
+ struct gnet_stats_basic_cpu __percpu *cpu,
+ struct gnet_stats_basic_packed *b)
{
+ struct gnet_stats_basic_packed bstats = {0};
+
+ __gnet_stats_copy_basic(&bstats, cpu, b);
+
if (d->compat_tc_stats) {
- d->tc_stats.bytes = b->bytes;
- d->tc_stats.packets = b->packets;
+ d->tc_stats.bytes = bstats.bytes;
+ d->tc_stats.packets = bstats.packets;
}
if (d->tail) {
struct gnet_stats_basic sb;
memset(&sb, 0, sizeof(sb));
- sb.bytes = b->bytes;
- sb.packets = b->packets;
+ sb.bytes = bstats.bytes;
+ sb.packets = bstats.packets;
return gnet_stats_copy(d, TCA_STATS_BASIC, &sb, sizeof(sb));
}
return 0;
diff --git a/net/netfilter/xt_RATEEST.c b/net/netfilter/xt_RATEEST.c
index 370adf6..604df6f 100644
--- a/net/netfilter/xt_RATEEST.c
+++ b/net/netfilter/xt_RATEEST.c
@@ -136,7 +136,7 @@ static int xt_rateest_tg_checkentry(const struct xt_tgchk_param *par)
cfg.est.interval = info->interval;
cfg.est.ewma_log = info->ewma_log;
- ret = gen_new_estimator(&est->bstats, &est->rstats,
+ ret = gen_new_estimator(&est->bstats, NULL, &est->rstats,
&est->lock, &cfg.opt);
if (ret < 0)
goto err2;
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 648778a..eca4cf9 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -252,7 +252,8 @@ int tcf_hash_create(u32 index, struct nlattr *est, struct tc_action *a,
p->tcfc_tm.install = jiffies;
p->tcfc_tm.lastuse = jiffies;
if (est) {
- int err = gen_new_estimator(&p->tcfc_bstats, &p->tcfc_rate_est,
+ int err = gen_new_estimator(&p->tcfc_bstats, NULL,
+ &p->tcfc_rate_est,
&p->tcfc_lock, est);
if (err) {
kfree(p);
@@ -619,7 +620,7 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a,
if (err < 0)
goto errout;
- if (gnet_stats_copy_basic(&d, &p->tcfc_bstats) < 0 ||
+ if (gnet_stats_copy_basic(&d, NULL, &p->tcfc_bstats) < 0 ||
gnet_stats_copy_rate_est(&d, &p->tcfc_bstats,
&p->tcfc_rate_est) < 0 ||
gnet_stats_copy_queue(&d, &p->tcfc_qstats) < 0)
diff --git a/net/sched/act_police.c b/net/sched/act_police.c
index f32bcb0..69791ca 100644
--- a/net/sched/act_police.c
+++ b/net/sched/act_police.c
@@ -178,7 +178,7 @@ override:
spin_lock_bh(&police->tcf_lock);
if (est) {
- err = gen_replace_estimator(&police->tcf_bstats,
+ err = gen_replace_estimator(&police->tcf_bstats, NULL,
&police->tcf_rate_est,
&police->tcf_lock, est);
if (err)
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 15e7bee..a95e3b4 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -942,6 +942,13 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
sch->handle = handle;
if (!ops->init || (err = ops->init(sch, tca[TCA_OPTIONS])) == 0) {
+ if (qdisc_is_percpu_stats(sch)) {
+ sch->cpu_bstats =
+ alloc_percpu(struct gnet_stats_basic_cpu);
+ if (!sch->cpu_bstats)
+ goto err_out4;
+ }
+
if (tca[TCA_STAB]) {
stab = qdisc_get_stab(tca[TCA_STAB]);
if (IS_ERR(stab)) {
@@ -964,8 +971,11 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
else
root_lock = qdisc_lock(sch);
- err = gen_new_estimator(&sch->bstats, &sch->rate_est,
- root_lock, tca[TCA_RATE]);
+ err = gen_new_estimator(&sch->bstats,
+ sch->cpu_bstats,
+ &sch->rate_est,
+ root_lock,
+ tca[TCA_RATE]);
if (err)
goto err_out4;
}
@@ -984,6 +994,7 @@ err_out:
return NULL;
err_out4:
+ free_percpu(sch->cpu_bstats);
/*
* Any broken qdiscs that would require a ops->reset() here?
* The qdisc was never in action so it shouldn't be necessary.
@@ -1022,9 +1033,11 @@ static int qdisc_change(struct Qdisc *sch, struct nlattr **tca)
because change can't be undone. */
if (sch->flags & TCQ_F_MQROOT)
goto out;
- gen_replace_estimator(&sch->bstats, &sch->rate_est,
- qdisc_root_sleeping_lock(sch),
- tca[TCA_RATE]);
+ gen_replace_estimator(&sch->bstats,
+ sch->cpu_bstats,
+ &sch->rate_est,
+ qdisc_root_sleeping_lock(sch),
+ tca[TCA_RATE]);
}
out:
return 0;
@@ -1299,6 +1312,7 @@ graft:
static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
u32 portid, u32 seq, u16 flags, int event)
{
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats = NULL;
struct tcmsg *tcm;
struct nlmsghdr *nlh;
unsigned char *b = skb_tail_pointer(skb);
@@ -1334,7 +1348,10 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0)
goto nla_put_failure;
- if (gnet_stats_copy_basic(&d, &q->bstats) < 0 ||
+ if (qdisc_is_percpu_stats(q))
+ cpu_bstats = q->cpu_bstats;
+
+ if (gnet_stats_copy_basic(&d, cpu_bstats, &q->bstats) < 0 ||
gnet_stats_copy_rate_est(&d, &q->bstats, &q->rate_est) < 0 ||
gnet_stats_copy_queue(&d, &q->qstats) < 0)
goto nla_put_failure;
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index c398f9c..0101766 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -639,7 +639,7 @@ atm_tc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
flow->qstats.qlen = flow->q->q.qlen;
- if (gnet_stats_copy_basic(d, &flow->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &flow->bstats) < 0 ||
gnet_stats_copy_queue(d, &flow->qstats) < 0)
return -1;
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index d2cd981..22a3a02 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1601,7 +1601,7 @@ cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (cl->undertime != PSCHED_PASTPERFECT)
cl->xstats.undertime = cl->undertime - q->now;
- if (gnet_stats_copy_basic(d, &cl->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
gnet_stats_copy_queue(d, &cl->qstats) < 0)
return -1;
@@ -1759,7 +1759,8 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
}
if (tca[TCA_RATE]) {
- err = gen_replace_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_replace_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err) {
@@ -1852,7 +1853,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
goto failure;
if (tca[TCA_RATE]) {
- err = gen_new_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err) {
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index d8b5ccf..7a6243c 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -88,7 +88,8 @@ static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
if (cl != NULL) {
if (tca[TCA_RATE]) {
- err = gen_replace_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_replace_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err)
@@ -116,7 +117,7 @@ static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
cl->qdisc = &noop_qdisc;
if (tca[TCA_RATE]) {
- err = gen_replace_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_replace_estimator(&cl->bstats, NULL, &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err) {
@@ -282,7 +283,7 @@ static int drr_dump_class_stats(struct Qdisc *sch, unsigned long arg,
cl->qdisc->qstats.qlen = cl->qdisc->q.qlen;
}
- if (gnet_stats_copy_basic(d, &cl->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
gnet_stats_copy_queue(d, &cl->qdisc->qstats) < 0)
return -1;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 11b28f6..7c8e5d7 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -632,6 +632,9 @@ static void qdisc_rcu_free(struct rcu_head *head)
{
struct Qdisc *qdisc = container_of(head, struct Qdisc, rcu_head);
+ if (qdisc_is_percpu_stats(qdisc))
+ free_percpu(qdisc->cpu_bstats);
+
kfree((char *) qdisc - qdisc->padded);
}
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 04b0de4..209b966 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1014,9 +1014,12 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
cur_time = psched_get_time();
if (tca[TCA_RATE]) {
- err = gen_replace_estimator(&cl->bstats, &cl->rate_est,
- qdisc_root_sleeping_lock(sch),
- tca[TCA_RATE]);
+ spinlock_t *lock = qdisc_root_sleeping_lock(sch);
+
+ err = gen_replace_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
+ lock,
+ tca[TCA_RATE]);
if (err)
return err;
}
@@ -1063,7 +1066,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
return -ENOBUFS;
if (tca[TCA_RATE]) {
- err = gen_new_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err) {
@@ -1374,7 +1377,7 @@ hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
xstats.work = cl->cl_total;
xstats.rtwork = cl->cl_cumul;
- if (gnet_stats_copy_basic(d, &cl->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
gnet_stats_copy_queue(d, &cl->qstats) < 0)
return -1;
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 063e953..0256dee 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1144,7 +1144,7 @@ htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d)
cl->xstats.tokens = PSCHED_NS2TICKS(cl->tokens);
cl->xstats.ctokens = PSCHED_NS2TICKS(cl->ctokens);
- if (gnet_stats_copy_basic(d, &cl->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, NULL, &cl->rate_est) < 0 ||
gnet_stats_copy_queue(d, &cl->qstats) < 0)
return -1;
@@ -1402,7 +1402,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
goto failure;
if (htb_rate_est || tca[TCA_RATE]) {
- err = gen_new_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_new_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE] ? : &est.nla);
if (err) {
@@ -1464,8 +1465,11 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
parent->children++;
} else {
if (tca[TCA_RATE]) {
- err = gen_replace_estimator(&cl->bstats, &cl->rate_est,
- qdisc_root_sleeping_lock(sch),
+ spinlock_t *lock = qdisc_root_sleeping_lock(sch);
+
+ err = gen_replace_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
+ lock,
tca[TCA_RATE]);
if (err)
return err;
diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c
index a8b2864..8080eaa 100644
--- a/net/sched/sch_mq.c
+++ b/net/sched/sch_mq.c
@@ -110,8 +110,8 @@ static int mq_dump(struct Qdisc *sch, struct sk_buff *skb)
qdisc = netdev_get_tx_queue(dev, ntx)->qdisc_sleeping;
spin_lock_bh(qdisc_lock(qdisc));
sch->q.qlen += qdisc->q.qlen;
- sch->bstats.bytes += qdisc->bstats.bytes;
- sch->bstats.packets += qdisc->bstats.packets;
+ sch->bstats->bytes += qdisc->bstats.bytes;
+ sch->bstats->packets += qdisc->bstats.packets;
sch->qstats.qlen += qdisc->qstats.qlen;
sch->qstats.backlog += qdisc->qstats.backlog;
sch->qstats.drops += qdisc->qstats.drops;
@@ -201,7 +201,7 @@ static int mq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
sch = dev_queue->qdisc_sleeping;
sch->qstats.qlen = sch->q.qlen;
- if (gnet_stats_copy_basic(d, &sch->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 ||
gnet_stats_copy_queue(d, &sch->qstats) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index 37e7d25..2f0ecd7 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -234,8 +234,8 @@ static int mqprio_dump(struct Qdisc *sch, struct sk_buff *skb)
qdisc = rtnl_dereference(netdev_get_tx_queue(dev, i)->qdisc);
spin_lock_bh(qdisc_lock(qdisc));
sch->q.qlen += qdisc->q.qlen;
- sch->bstats.bytes += qdisc->bstats.bytes;
- sch->bstats.packets += qdisc->bstats.packets;
+ sch->bstats->bytes += qdisc->bstats.bytes;
+ sch->bstats->packets += qdisc->bstats.packets;
sch->qstats.qlen += qdisc->qstats.qlen;
sch->qstats.backlog += qdisc->qstats.backlog;
sch->qstats.drops += qdisc->qstats.drops;
@@ -355,7 +355,7 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
}
/* Reclaim root sleeping lock before completing stats */
spin_lock_bh(d->lock);
- if (gnet_stats_copy_basic(d, &bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &bstats) < 0 ||
gnet_stats_copy_queue(d, &qstats) < 0)
return -1;
} else {
@@ -363,7 +363,7 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
sch = dev_queue->qdisc_sleeping;
sch->qstats.qlen = sch->q.qlen;
- if (gnet_stats_copy_basic(d, &sch->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 ||
gnet_stats_copy_queue(d, &sch->qstats) < 0)
return -1;
}
diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c
index c0466c1..4adbf7f 100644
--- a/net/sched/sch_multiq.c
+++ b/net/sched/sch_multiq.c
@@ -361,7 +361,7 @@ static int multiq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
cl_q = q->queues[cl - 1];
cl_q->qstats.qlen = cl_q->q.qlen;
- if (gnet_stats_copy_basic(d, &cl_q->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 ||
gnet_stats_copy_queue(d, &cl_q->qstats) < 0)
return -1;
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 03ef99e..68a8f25 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -325,7 +325,7 @@ static int prio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
cl_q = q->queues[cl - 1];
cl_q->qstats.qlen = cl_q->q.qlen;
- if (gnet_stats_copy_basic(d, &cl_q->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 ||
gnet_stats_copy_queue(d, &cl_q->qstats) < 0)
return -1;
diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index 602ea01..d59f857 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -459,7 +459,8 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
if (cl != NULL) { /* modify existing class */
if (tca[TCA_RATE]) {
- err = gen_replace_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_replace_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err)
@@ -484,7 +485,8 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
cl->qdisc = &noop_qdisc;
if (tca[TCA_RATE]) {
- err = gen_new_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_new_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err)
@@ -667,7 +669,7 @@ static int qfq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
xstats.weight = cl->agg->class_weight;
xstats.lmax = cl->agg->lmax;
- if (gnet_stats_copy_basic(d, &cl->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
gnet_stats_copy_queue(d, &cl->qdisc->qstats) < 0)
return -1;
^ permalink raw reply related
* [net-next PATCH 2/4] net: sched: implement qstat helper routines
From: John Fastabend @ 2014-09-26 18:14 UTC (permalink / raw)
To: xiyou.wangcong, jhs, eric.dumazet, davem; +Cc: netdev
In-Reply-To: <20140926181346.28430.19380.stgit@nitbit.x32>
This adds helpers to manipulate qstats logic and replaces locations
that touch the counters directly. This simplifies future patches
to push qstats onto per cpu counters.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
---
include/net/sch_generic.h | 43 +++++++++++++++++++++++++++++++++++--------
net/sched/sch_api.c | 2 +-
net/sched/sch_atm.c | 2 +-
net/sched/sch_cbq.c | 10 +++++-----
net/sched/sch_choke.c | 14 +++++++-------
net/sched/sch_codel.c | 2 +-
net/sched/sch_drr.c | 4 ++--
net/sched/sch_dsmark.c | 2 +-
net/sched/sch_fifo.c | 2 +-
net/sched/sch_fq.c | 4 ++--
net/sched/sch_fq_codel.c | 8 ++++----
net/sched/sch_gred.c | 4 ++--
net/sched/sch_hfsc.c | 8 ++++----
net/sched/sch_hhf.c | 8 ++++----
net/sched/sch_htb.c | 6 +++---
net/sched/sch_ingress.c | 2 +-
net/sched/sch_multiq.c | 4 ++--
net/sched/sch_netem.c | 15 +++++++--------
net/sched/sch_pie.c | 2 +-
net/sched/sch_prio.c | 4 ++--
net/sched/sch_qfq.c | 4 ++--
net/sched/sch_red.c | 8 ++++----
net/sched/sch_sfb.c | 10 +++++-----
net/sched/sch_sfq.c | 17 +++++++++--------
net/sched/sch_tbf.c | 8 ++++----
25 files changed, 110 insertions(+), 83 deletions(-)
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 4b93511..5609844 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -521,11 +521,39 @@ static inline void qdisc_bstats_update(struct Qdisc *sch,
bstats_update(&sch->bstats, skb);
}
+static inline void qdisc_qstats_backlog(struct Qdisc *sch,
+ const struct sk_buff *skb)
+{
+ sch->qstats.backlog -= qdisc_pkt_len(skb);
+}
+
+static inline void qdisc_qstats_backlog_inc(struct Qdisc *sch,
+ const struct sk_buff *skb)
+{
+
+ sch->qstats.backlog += qdisc_pkt_len(skb);
+}
+
+static inline void __qdisc_qstats_drop(struct Qdisc *sch, int count)
+{
+ sch->qstats.drops += count;
+}
+
+static inline void qdisc_qstats_drop(struct Qdisc *sch)
+{
+ sch->qstats.drops++;
+}
+
+static inline void qdisc_qstats_overlimit(struct Qdisc *sch)
+{
+ sch->qstats.overlimits++;
+}
+
static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
struct sk_buff_head *list)
{
__skb_queue_tail(list, skb);
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
return NET_XMIT_SUCCESS;
}
@@ -541,7 +569,7 @@ static inline struct sk_buff *__qdisc_dequeue_head(struct Qdisc *sch,
struct sk_buff *skb = __skb_dequeue(list);
if (likely(skb != NULL)) {
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
qdisc_bstats_update(sch, skb);
}
@@ -559,10 +587,9 @@ static inline unsigned int __qdisc_queue_drop_head(struct Qdisc *sch,
struct sk_buff *skb = __skb_dequeue(list);
if (likely(skb != NULL)) {
- unsigned int len = qdisc_pkt_len(skb);
- sch->qstats.backlog -= len;
+ qdisc_qstats_backlog(sch, skb);
kfree_skb(skb);
- return len;
+ return qdisc_pkt_len(skb);
}
return 0;
@@ -579,7 +606,7 @@ static inline struct sk_buff *__qdisc_dequeue_tail(struct Qdisc *sch,
struct sk_buff *skb = __skb_dequeue_tail(list);
if (likely(skb != NULL))
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
return skb;
}
@@ -661,14 +688,14 @@ static inline unsigned int qdisc_queue_drop(struct Qdisc *sch)
static inline int qdisc_drop(struct sk_buff *skb, struct Qdisc *sch)
{
kfree_skb(skb);
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return NET_XMIT_DROP;
}
static inline int qdisc_reshape_fail(struct sk_buff *skb, struct Qdisc *sch)
{
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
#ifdef CONFIG_NET_CLS_ACT
if (sch->reshape_fail == NULL || sch->reshape_fail(skb, sch))
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index a95e3b4..2862bc6 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -763,7 +763,7 @@ void qdisc_tree_decrease_qlen(struct Qdisc *sch, unsigned int n)
cops->put(sch, cl);
}
sch->q.qlen -= n;
- sch->qstats.drops += drops;
+ __qdisc_qstats_drop(sch, drops);
}
}
EXPORT_SYMBOL(qdisc_tree_decrease_qlen);
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index 0101766..040212c 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -417,7 +417,7 @@ done:
if (ret != NET_XMIT_SUCCESS) {
drop: __maybe_unused
if (net_xmit_drop_count(ret)) {
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
if (flow)
flow->qstats.drops++;
}
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 22a3a02..60432c3 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -377,7 +377,7 @@ cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
#endif
if (cl == NULL) {
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
@@ -395,7 +395,7 @@ cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
}
if (net_xmit_drop_count(ret)) {
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
cbq_mark_toplevel(q, cl);
cl->qstats.drops++;
}
@@ -650,11 +650,11 @@ static int cbq_reshape_fail(struct sk_buff *skb, struct Qdisc *child)
return 0;
}
if (net_xmit_drop_count(ret))
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return 0;
}
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return -1;
}
#endif
@@ -995,7 +995,7 @@ cbq_dequeue(struct Qdisc *sch)
*/
if (sch->q.qlen) {
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (q->wd_expires)
qdisc_watchdog_schedule(&q->watchdog,
now + q->wd_expires);
diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c
index 8abc262..a514f42 100644
--- a/net/sched/sch_choke.c
+++ b/net/sched/sch_choke.c
@@ -127,7 +127,7 @@ static void choke_drop_by_idx(struct Qdisc *sch, unsigned int idx)
if (idx == q->tail)
choke_zap_tail_holes(q);
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
qdisc_drop(skb, sch);
qdisc_tree_decrease_qlen(sch, 1);
--sch->q.qlen;
@@ -302,7 +302,7 @@ static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch)
if (q->vars.qavg > p->qth_max) {
q->vars.qcount = -1;
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (use_harddrop(q) || !use_ecn(q) ||
!INET_ECN_set_ce(skb)) {
q->stats.forced_drop++;
@@ -315,7 +315,7 @@ static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch)
q->vars.qcount = 0;
q->vars.qR = red_random(p);
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (!use_ecn(q) || !INET_ECN_set_ce(skb)) {
q->stats.prob_drop++;
goto congestion_drop;
@@ -332,7 +332,7 @@ static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch)
q->tab[q->tail] = skb;
q->tail = (q->tail + 1) & q->tab_mask;
++sch->q.qlen;
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
return NET_XMIT_SUCCESS;
}
@@ -345,7 +345,7 @@ congestion_drop:
other_drop:
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
@@ -365,7 +365,7 @@ static struct sk_buff *choke_dequeue(struct Qdisc *sch)
q->tab[q->head] = NULL;
choke_zap_head_holes(q);
--sch->q.qlen;
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
qdisc_bstats_update(sch, skb);
return skb;
@@ -460,7 +460,7 @@ static int choke_change(struct Qdisc *sch, struct nlattr *opt)
ntab[tail++] = skb;
continue;
}
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
--sch->q.qlen;
qdisc_drop(skb, sch);
}
diff --git a/net/sched/sch_codel.c b/net/sched/sch_codel.c
index 2f9ab17..c56ac44 100644
--- a/net/sched/sch_codel.c
+++ b/net/sched/sch_codel.c
@@ -149,7 +149,7 @@ static int codel_change(struct Qdisc *sch, struct nlattr *opt)
while (sch->q.qlen > sch->limit) {
struct sk_buff *skb = __skb_dequeue(&sch->q);
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
qdisc_drop(skb, sch);
}
qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen);
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index 7a6243c..907b12f 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -360,7 +360,7 @@ static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch)
cl = drr_classify(skb, sch, &err);
if (cl == NULL) {
if (err & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return err;
}
@@ -369,7 +369,7 @@ static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch)
if (unlikely(err != NET_XMIT_SUCCESS)) {
if (net_xmit_drop_count(err)) {
cl->qstats.drops++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
}
return err;
}
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index 485e456..227114f 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -258,7 +258,7 @@ static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch)
err = qdisc_enqueue(skb, p->q);
if (err != NET_XMIT_SUCCESS) {
if (net_xmit_drop_count(err))
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return err;
}
diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c
index e15a9eb..2e2398c 100644
--- a/net/sched/sch_fifo.c
+++ b/net/sched/sch_fifo.c
@@ -42,7 +42,7 @@ static int pfifo_tail_enqueue(struct sk_buff *skb, struct Qdisc *sch)
/* queue full, remove one skb to fulfill the limit */
__qdisc_queue_drop_head(sch, &sch->q);
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
qdisc_enqueue_tail(skb, sch);
return NET_XMIT_CN;
diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c
index e12f997..a5a2e9f 100644
--- a/net/sched/sch_fq.c
+++ b/net/sched/sch_fq.c
@@ -290,7 +290,7 @@ static struct sk_buff *fq_dequeue_head(struct Qdisc *sch, struct fq_flow *flow)
flow->head = skb->next;
skb->next = NULL;
flow->qlen--;
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
sch->q.qlen--;
}
return skb;
@@ -371,7 +371,7 @@ static int fq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
f->qlen++;
if (skb_is_retransmit(skb))
q->stat_tcp_retrans++;
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
if (fq_flow_is_detached(f)) {
fq_flow_add_tail(&q->new_flows, f);
if (time_after(jiffies, f->age + q->flow_refill_delay))
diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c
index 105cf55..3f288ae 100644
--- a/net/sched/sch_fq_codel.c
+++ b/net/sched/sch_fq_codel.c
@@ -164,8 +164,8 @@ static unsigned int fq_codel_drop(struct Qdisc *sch)
q->backlogs[idx] -= len;
kfree_skb(skb);
sch->q.qlen--;
- sch->qstats.drops++;
- sch->qstats.backlog -= len;
+ qdisc_qstats_drop(sch);
+ qdisc_qstats_backlog(sch, skb);
flow->dropped++;
return idx;
}
@@ -180,7 +180,7 @@ static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch)
idx = fq_codel_classify(skb, sch, &ret);
if (idx == 0) {
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
@@ -190,7 +190,7 @@ static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch)
flow = &q->flows[idx];
flow_queue_add(flow, skb);
q->backlogs[idx] += qdisc_pkt_len(skb);
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
if (list_empty(&flow->flowchain)) {
list_add_tail(&flow->flowchain, &q->new_flows);
diff --git a/net/sched/sch_gred.c b/net/sched/sch_gred.c
index 12cbc09..a4ca451 100644
--- a/net/sched/sch_gred.c
+++ b/net/sched/sch_gred.c
@@ -209,7 +209,7 @@ static int gred_enqueue(struct sk_buff *skb, struct Qdisc *sch)
break;
case RED_PROB_MARK:
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (!gred_use_ecn(t) || !INET_ECN_set_ce(skb)) {
q->stats.prob_drop++;
goto congestion_drop;
@@ -219,7 +219,7 @@ static int gred_enqueue(struct sk_buff *skb, struct Qdisc *sch)
break;
case RED_HARD_MARK:
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (gred_use_harddrop(t) || !gred_use_ecn(t) ||
!INET_ECN_set_ce(skb)) {
q->stats.forced_drop++;
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 209b966..ad27825 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1591,7 +1591,7 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
cl = hfsc_classify(skb, sch, &err);
if (cl == NULL) {
if (err & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return err;
}
@@ -1600,7 +1600,7 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
if (unlikely(err != NET_XMIT_SUCCESS)) {
if (net_xmit_drop_count(err)) {
cl->qstats.drops++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
}
return err;
}
@@ -1643,7 +1643,7 @@ hfsc_dequeue(struct Qdisc *sch)
*/
cl = vttree_get_minvt(&q->root, cur_time);
if (cl == NULL) {
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
hfsc_schedule_watchdog(sch);
return NULL;
}
@@ -1698,7 +1698,7 @@ hfsc_drop(struct Qdisc *sch)
list_move_tail(&cl->dlist, &q->droplist);
}
cl->qstats.drops++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
sch->q.qlen--;
return len;
}
diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c
index d85b681..934e9cc 100644
--- a/net/sched/sch_hhf.c
+++ b/net/sched/sch_hhf.c
@@ -376,8 +376,8 @@ static unsigned int hhf_drop(struct Qdisc *sch)
struct sk_buff *skb = dequeue_head(bucket);
sch->q.qlen--;
- sch->qstats.drops++;
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_drop(sch);
+ qdisc_qstats_backlog(sch, skb);
kfree_skb(skb);
}
@@ -395,7 +395,7 @@ static int hhf_enqueue(struct sk_buff *skb, struct Qdisc *sch)
bucket = &q->buckets[idx];
bucket_add(bucket, skb);
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
if (list_empty(&bucket->bucketchain)) {
unsigned int weight;
@@ -457,7 +457,7 @@ begin:
if (bucket->head) {
skb = dequeue_head(bucket);
sch->q.qlen--;
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
}
if (!skb) {
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 0256dee..c40ab7a 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -586,13 +586,13 @@ static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch)
#ifdef CONFIG_NET_CLS_ACT
} else if (!cl) {
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
#endif
} else if ((ret = qdisc_enqueue(skb, cl->un.leaf.q)) != NET_XMIT_SUCCESS) {
if (net_xmit_drop_count(ret)) {
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
cl->qstats.drops++;
}
return ret;
@@ -925,7 +925,7 @@ ok:
goto ok;
}
}
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (likely(next_event > q->now)) {
if (!test_bit(__QDISC_STATE_DEACTIVATED,
&qdisc_root_sleeping(q->watchdog.qdisc)->state)) {
diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c
index b351125..eb5b844 100644
--- a/net/sched/sch_ingress.c
+++ b/net/sched/sch_ingress.c
@@ -69,7 +69,7 @@ static int ingress_enqueue(struct sk_buff *skb, struct Qdisc *sch)
switch (result) {
case TC_ACT_SHOT:
result = TC_ACT_SHOT;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
break;
case TC_ACT_STOLEN:
case TC_ACT_QUEUED:
diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c
index 4adbf7f..7f4e1d8 100644
--- a/net/sched/sch_multiq.c
+++ b/net/sched/sch_multiq.c
@@ -75,7 +75,7 @@ multiq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
if (qdisc == NULL) {
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
@@ -87,7 +87,7 @@ multiq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
return NET_XMIT_SUCCESS;
}
if (net_xmit_drop_count(ret))
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return ret;
}
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 111d70f..ace7ca1 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -429,12 +429,12 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
/* Drop packet? */
if (loss_event(q)) {
if (q->ecn && INET_ECN_set_ce(skb))
- sch->qstats.drops++; /* mark packet */
+ qdisc_qstats_drop(sch); /* mark packet */
else
--count;
}
if (count == 0) {
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return NET_XMIT_SUCCESS | __NET_XMIT_BYPASS;
}
@@ -478,7 +478,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
if (unlikely(skb_queue_len(&sch->q) >= sch->limit))
return qdisc_reshape_fail(skb, sch);
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
cb = netem_skb_cb(skb);
if (q->gap == 0 || /* not doing reordering */
@@ -549,15 +549,14 @@ static unsigned int netem_drop(struct Qdisc *sch)
sch->q.qlen--;
skb->next = NULL;
skb->prev = NULL;
- len = qdisc_pkt_len(skb);
- sch->qstats.backlog -= len;
+ qdisc_qstats_backlog(sch, skb);
kfree_skb(skb);
}
}
if (!len && q->qdisc && q->qdisc->ops->drop)
len = q->qdisc->ops->drop(q->qdisc);
if (len)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return len;
}
@@ -575,7 +574,7 @@ tfifo_dequeue:
skb = __skb_dequeue(&sch->q);
if (skb) {
deliver:
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
qdisc_unthrottled(sch);
qdisc_bstats_update(sch, skb);
return skb;
@@ -610,7 +609,7 @@ deliver:
if (unlikely(err != NET_XMIT_SUCCESS)) {
if (net_xmit_drop_count(err)) {
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
qdisc_tree_decrease_qlen(sch, 1);
}
}
diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c
index fefeeb7..32636c9 100644
--- a/net/sched/sch_pie.c
+++ b/net/sched/sch_pie.c
@@ -232,7 +232,7 @@ static int pie_change(struct Qdisc *sch, struct nlattr *opt)
while (sch->q.qlen > sch->limit) {
struct sk_buff *skb = __skb_dequeue(&sch->q);
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
qdisc_drop(skb, sch);
}
qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen);
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 68a8f25..b411e78 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -77,7 +77,7 @@ prio_enqueue(struct sk_buff *skb, struct Qdisc *sch)
if (qdisc == NULL) {
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
@@ -89,7 +89,7 @@ prio_enqueue(struct sk_buff *skb, struct Qdisc *sch)
return NET_XMIT_SUCCESS;
}
if (net_xmit_drop_count(ret))
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return ret;
}
diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index d59f857..3fb2655 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -1229,7 +1229,7 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
cl = qfq_classify(skb, sch, &err);
if (cl == NULL) {
if (err & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return err;
}
@@ -1249,7 +1249,7 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
pr_debug("qfq_enqueue: enqueue failed %d\n", err);
if (net_xmit_drop_count(err)) {
cl->qstats.drops++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
}
return err;
}
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index 633e32d..6c0534c 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -74,7 +74,7 @@ static int red_enqueue(struct sk_buff *skb, struct Qdisc *sch)
break;
case RED_PROB_MARK:
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (!red_use_ecn(q) || !INET_ECN_set_ce(skb)) {
q->stats.prob_drop++;
goto congestion_drop;
@@ -84,7 +84,7 @@ static int red_enqueue(struct sk_buff *skb, struct Qdisc *sch)
break;
case RED_HARD_MARK:
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (red_use_harddrop(q) || !red_use_ecn(q) ||
!INET_ECN_set_ce(skb)) {
q->stats.forced_drop++;
@@ -100,7 +100,7 @@ static int red_enqueue(struct sk_buff *skb, struct Qdisc *sch)
sch->q.qlen++;
} else if (net_xmit_drop_count(ret)) {
q->stats.pdrop++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
}
return ret;
@@ -142,7 +142,7 @@ static unsigned int red_drop(struct Qdisc *sch)
if (child->ops->drop && (len = child->ops->drop(child)) > 0) {
q->stats.other++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
sch->q.qlen--;
return len;
}
diff --git a/net/sched/sch_sfb.c b/net/sched/sch_sfb.c
index 1562fb2..5819dd8 100644
--- a/net/sched/sch_sfb.c
+++ b/net/sched/sch_sfb.c
@@ -290,7 +290,7 @@ static int sfb_enqueue(struct sk_buff *skb, struct Qdisc *sch)
struct flow_keys keys;
if (unlikely(sch->q.qlen >= q->limit)) {
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
q->stats.queuedrop++;
goto drop;
}
@@ -348,7 +348,7 @@ static int sfb_enqueue(struct sk_buff *skb, struct Qdisc *sch)
sfb_skb_cb(skb)->hashes[slot] = 0;
if (unlikely(minqlen >= q->max)) {
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
q->stats.bucketdrop++;
goto drop;
}
@@ -376,7 +376,7 @@ static int sfb_enqueue(struct sk_buff *skb, struct Qdisc *sch)
}
}
if (sfb_rate_limit(skb, q)) {
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
q->stats.penaltydrop++;
goto drop;
}
@@ -411,7 +411,7 @@ enqueue:
increment_qlen(skb, q);
} else if (net_xmit_drop_count(ret)) {
q->stats.childdrop++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
}
return ret;
@@ -420,7 +420,7 @@ drop:
return NET_XMIT_CN;
other_drop:
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 80c36bd..16fb3e7 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -331,8 +331,8 @@ drop:
sfq_dec(q, x);
kfree_skb(skb);
sch->q.qlen--;
- sch->qstats.drops++;
- sch->qstats.backlog -= len;
+ qdisc_qstats_drop(sch);
+ qdisc_qstats_backlog(sch, skb);
return len;
}
@@ -379,7 +379,7 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
hash = sfq_classify(skb, sch, &ret);
if (hash == 0) {
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
@@ -409,7 +409,7 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
break;
case RED_PROB_MARK:
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (sfq_prob_mark(q)) {
/* We know we have at least one packet in queue */
if (sfq_headdrop(q) &&
@@ -426,7 +426,7 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
goto congestion_drop;
case RED_HARD_MARK:
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (sfq_hard_mark(q)) {
/* We know we have at least one packet in queue */
if (sfq_headdrop(q) &&
@@ -461,7 +461,7 @@ congestion_drop:
}
enqueue:
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
slot->backlog += qdisc_pkt_len(skb);
slot_queue_add(slot, skb);
sfq_inc(q, x);
@@ -520,7 +520,7 @@ next_slot:
sfq_dec(q, a);
qdisc_bstats_update(sch, skb);
sch->q.qlen--;
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
slot->backlog -= qdisc_pkt_len(skb);
/* Is the slot empty? */
if (slot->qlen == 0) {
@@ -586,7 +586,8 @@ static void sfq_rehash(struct Qdisc *sch)
if (x == SFQ_EMPTY_SLOT) {
x = q->dep[0].next; /* get a free slot */
if (x >= SFQ_MAX_FLOWS) {
-drop: sch->qstats.backlog -= qdisc_pkt_len(skb);
+drop:
+ qdisc_qstats_backlog(sch, skb);
kfree_skb(skb);
dropped++;
continue;
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 0c39b75..77edffe 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -175,7 +175,7 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch)
ret = qdisc_enqueue(segs, q->qdisc);
if (ret != NET_XMIT_SUCCESS) {
if (net_xmit_drop_count(ret))
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
} else {
nb++;
}
@@ -201,7 +201,7 @@ static int tbf_enqueue(struct sk_buff *skb, struct Qdisc *sch)
ret = qdisc_enqueue(skb, q->qdisc);
if (ret != NET_XMIT_SUCCESS) {
if (net_xmit_drop_count(ret))
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return ret;
}
@@ -216,7 +216,7 @@ static unsigned int tbf_drop(struct Qdisc *sch)
if (q->qdisc->ops->drop && (len = q->qdisc->ops->drop(q->qdisc)) != 0) {
sch->q.qlen--;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
}
return len;
}
@@ -281,7 +281,7 @@ static struct sk_buff *tbf_dequeue(struct Qdisc *sch)
(cf. CSZ, HPFQ, HFSC)
*/
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
}
return NULL;
}
^ permalink raw reply related
* [net-next PATCH 3/4] net: sched: restrict use of qstats qlen
From: John Fastabend @ 2014-09-26 18:15 UTC (permalink / raw)
To: xiyou.wangcong, jhs, eric.dumazet, davem; +Cc: netdev
In-Reply-To: <20140926181346.28430.19380.stgit@nitbit.x32>
This removes the use of qstats->qlen variable from the classifiers
and makes it an explicit argument to gnet_stats_copy_queue().
The qlen represents the qdisc queue length and is packed into
the qstats at the last moment before passnig to user space. By
handling it explicitely we avoid, in the percpu stats case, having
to figure out which per_cpu variable to put it in.
It would probably be best to remove it from qstats completely
but qstats is a user space ABI and can't be broken. A future
patch could make an internal only qstats structure that would
avoid having to allocate an additional u32 variable on the
Qdisc struct. This would make the qstats struct 128bits instead
of 128+32.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
---
include/net/gen_stats.h | 3 ++-
net/core/gen_stats.c | 6 +++++-
net/sched/act_api.c | 4 +++-
net/sched/sch_api.c | 5 +++--
net/sched/sch_atm.c | 4 +---
net/sched/sch_cbq.c | 3 +--
net/sched/sch_drr.c | 7 +++----
net/sched/sch_fq_codel.c | 2 +-
net/sched/sch_hfsc.c | 3 +--
net/sched/sch_htb.c | 5 +++--
net/sched/sch_mq.c | 8 +++-----
net/sched/sch_mqprio.c | 13 ++++++-------
net/sched/sch_multiq.c | 3 +--
net/sched/sch_prio.c | 3 +--
net/sched/sch_qfq.c | 3 +--
net/sched/sch_sfq.c | 2 +-
16 files changed, 36 insertions(+), 38 deletions(-)
diff --git a/include/net/gen_stats.h b/include/net/gen_stats.h
index ce3c128..de9b3dd 100644
--- a/include/net/gen_stats.h
+++ b/include/net/gen_stats.h
@@ -40,7 +40,8 @@ void __gnet_stats_copy_basic(struct gnet_stats_basic_packed *bstats,
int gnet_stats_copy_rate_est(struct gnet_dump *d,
const struct gnet_stats_basic_packed *b,
struct gnet_stats_rate_est64 *r);
-int gnet_stats_copy_queue(struct gnet_dump *d, struct gnet_stats_queue *q);
+int gnet_stats_copy_queue(struct gnet_dump *d,
+ struct gnet_stats_queue *q, __u32 len);
int gnet_stats_copy_app(struct gnet_dump *d, void *st, int len);
int gnet_stats_finish_copy(struct gnet_dump *d);
diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c
index 5ff8e80..ad3ecb6 100644
--- a/net/core/gen_stats.c
+++ b/net/core/gen_stats.c
@@ -219,6 +219,7 @@ EXPORT_SYMBOL(gnet_stats_copy_rate_est);
* gnet_stats_copy_queue - copy queue statistics into statistics TLV
* @d: dumping handle
* @q: queue statistics
+ * @qlen: queue length statistics
*
* Appends the queue statistics to the top level TLV created by
* gnet_stats_start_copy().
@@ -227,8 +228,11 @@ EXPORT_SYMBOL(gnet_stats_copy_rate_est);
* if the room in the socket buffer was not sufficient.
*/
int
-gnet_stats_copy_queue(struct gnet_dump *d, struct gnet_stats_queue *q)
+gnet_stats_copy_queue(struct gnet_dump *d,
+ struct gnet_stats_queue *q, __u32 qlen)
{
+ q->qlen = qlen;
+
if (d->compat_tc_stats) {
d->tc_stats.drops = q->drops;
d->tc_stats.qlen = q->qlen;
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index eca4cf9..2e13409 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -623,7 +623,9 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a,
if (gnet_stats_copy_basic(&d, NULL, &p->tcfc_bstats) < 0 ||
gnet_stats_copy_rate_est(&d, &p->tcfc_bstats,
&p->tcfc_rate_est) < 0 ||
- gnet_stats_copy_queue(&d, &p->tcfc_qstats) < 0)
+ gnet_stats_copy_queue(&d,
+ &p->tcfc_qstats,
+ p->tcfc_qstats.qlen) < 0)
goto errout;
if (gnet_stats_finish_copy(&d) < 0)
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 2862bc6..ca00ea8 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -1318,6 +1318,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
unsigned char *b = skb_tail_pointer(skb);
struct gnet_dump d;
struct qdisc_size_table *stab;
+ __u32 qlen;
cond_resched();
nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags);
@@ -1335,7 +1336,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
goto nla_put_failure;
if (q->ops->dump && q->ops->dump(q, skb) < 0)
goto nla_put_failure;
- q->qstats.qlen = q->q.qlen;
+ qlen = q->q.qlen;
stab = rtnl_dereference(q->stab);
if (stab && qdisc_dump_stab(skb, stab) < 0)
@@ -1353,7 +1354,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
if (gnet_stats_copy_basic(&d, cpu_bstats, &q->bstats) < 0 ||
gnet_stats_copy_rate_est(&d, &q->bstats, &q->rate_est) < 0 ||
- gnet_stats_copy_queue(&d, &q->qstats) < 0)
+ gnet_stats_copy_queue(&d, &q->qstats, qlen) < 0)
goto nla_put_failure;
if (gnet_stats_finish_copy(&d) < 0)
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index 040212c..c145eb6 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -637,10 +637,8 @@ atm_tc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
{
struct atm_flow_data *flow = (struct atm_flow_data *)arg;
- flow->qstats.qlen = flow->q->q.qlen;
-
if (gnet_stats_copy_basic(d, NULL, &flow->bstats) < 0 ||
- gnet_stats_copy_queue(d, &flow->qstats) < 0)
+ gnet_stats_copy_queue(d, &flow->qstats, flow->q->q.qlen) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 60432c3..c610081 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1594,7 +1594,6 @@ cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
struct cbq_sched_data *q = qdisc_priv(sch);
struct cbq_class *cl = (struct cbq_class *)arg;
- cl->qstats.qlen = cl->q->q.qlen;
cl->xstats.avgidle = cl->avgidle;
cl->xstats.undertime = 0;
@@ -1603,7 +1602,7 @@ cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl->qstats, cl->q->q.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats));
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index 907b12f..5835a93 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -275,17 +275,16 @@ static int drr_dump_class_stats(struct Qdisc *sch, unsigned long arg,
struct gnet_dump *d)
{
struct drr_class *cl = (struct drr_class *)arg;
+ __u32 qlen = cl->qdisc->q.qlen;
struct tc_drr_stats xstats;
memset(&xstats, 0, sizeof(xstats));
- if (cl->qdisc->q.qlen) {
+ if (qlen)
xstats.deficit = cl->deficit;
- cl->qdisc->qstats.qlen = cl->qdisc->q.qlen;
- }
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qdisc->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl->qdisc->qstats, qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c
index 3f288ae..6a42112 100644
--- a/net/sched/sch_fq_codel.c
+++ b/net/sched/sch_fq_codel.c
@@ -550,7 +550,7 @@ static int fq_codel_dump_class_stats(struct Qdisc *sch, unsigned long cl,
qs.backlog = q->backlogs[idx];
qs.drops = flow->dropped;
}
- if (gnet_stats_copy_queue(d, &qs) < 0)
+ if (gnet_stats_copy_queue(d, &qs, 0) < 0)
return -1;
if (idx < q->flows_cnt)
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index ad27825..d364acb 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1370,7 +1370,6 @@ hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
struct hfsc_class *cl = (struct hfsc_class *)arg;
struct tc_hfsc_stats xstats;
- cl->qstats.qlen = cl->qdisc->q.qlen;
cl->qstats.backlog = cl->qdisc->qstats.backlog;
xstats.level = cl->level;
xstats.period = cl->cl_vtperiod;
@@ -1379,7 +1378,7 @@ hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl->qstats, cl->qdisc->q.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index c40ab7a..3a691fd 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1138,15 +1138,16 @@ static int
htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d)
{
struct htb_class *cl = (struct htb_class *)arg;
+ __u32 qlen = 0;
if (!cl->level && cl->un.leaf.q)
- cl->qstats.qlen = cl->un.leaf.q->q.qlen;
+ qlen = cl->un.leaf.q->q.qlen;
cl->xstats.tokens = PSCHED_NS2TICKS(cl->tokens);
cl->xstats.ctokens = PSCHED_NS2TICKS(cl->ctokens);
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, NULL, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl->qstats, qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats));
diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c
index 8080eaa..6416a69 100644
--- a/net/sched/sch_mq.c
+++ b/net/sched/sch_mq.c
@@ -110,9 +110,8 @@ static int mq_dump(struct Qdisc *sch, struct sk_buff *skb)
qdisc = netdev_get_tx_queue(dev, ntx)->qdisc_sleeping;
spin_lock_bh(qdisc_lock(qdisc));
sch->q.qlen += qdisc->q.qlen;
- sch->bstats->bytes += qdisc->bstats.bytes;
- sch->bstats->packets += qdisc->bstats.packets;
- sch->qstats.qlen += qdisc->qstats.qlen;
+ sch->bstats.bytes += qdisc->bstats.bytes;
+ sch->bstats.packets += qdisc->bstats.packets;
sch->qstats.backlog += qdisc->qstats.backlog;
sch->qstats.drops += qdisc->qstats.drops;
sch->qstats.requeues += qdisc->qstats.requeues;
@@ -200,9 +199,8 @@ static int mq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
struct netdev_queue *dev_queue = mq_queue_get(sch, cl);
sch = dev_queue->qdisc_sleeping;
- sch->qstats.qlen = sch->q.qlen;
if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 ||
- gnet_stats_copy_queue(d, &sch->qstats) < 0)
+ gnet_stats_copy_queue(d, &sch->qstats, sch->q.qlen) < 0)
return -1;
return 0;
}
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index 2f0ecd7..03dbeb5 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -234,9 +234,8 @@ static int mqprio_dump(struct Qdisc *sch, struct sk_buff *skb)
qdisc = rtnl_dereference(netdev_get_tx_queue(dev, i)->qdisc);
spin_lock_bh(qdisc_lock(qdisc));
sch->q.qlen += qdisc->q.qlen;
- sch->bstats->bytes += qdisc->bstats.bytes;
- sch->bstats->packets += qdisc->bstats.packets;
- sch->qstats.qlen += qdisc->qstats.qlen;
+ sch->bstats.bytes += qdisc->bstats.bytes;
+ sch->bstats.packets += qdisc->bstats.packets;
sch->qstats.backlog += qdisc->qstats.backlog;
sch->qstats.drops += qdisc->qstats.drops;
sch->qstats.requeues += qdisc->qstats.requeues;
@@ -327,6 +326,7 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
if (cl <= netdev_get_num_tc(dev)) {
int i;
+ __u32 qlen = 0;
struct Qdisc *qdisc;
struct gnet_stats_queue qstats = {0};
struct gnet_stats_basic_packed bstats = {0};
@@ -344,9 +344,9 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
qdisc = rtnl_dereference(q->qdisc);
spin_lock_bh(qdisc_lock(qdisc));
+ qlen += qdisc->q.qlen;
bstats.bytes += qdisc->bstats.bytes;
bstats.packets += qdisc->bstats.packets;
- qstats.qlen += qdisc->qstats.qlen;
qstats.backlog += qdisc->qstats.backlog;
qstats.drops += qdisc->qstats.drops;
qstats.requeues += qdisc->qstats.requeues;
@@ -356,15 +356,14 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
/* Reclaim root sleeping lock before completing stats */
spin_lock_bh(d->lock);
if (gnet_stats_copy_basic(d, NULL, &bstats) < 0 ||
- gnet_stats_copy_queue(d, &qstats) < 0)
+ gnet_stats_copy_queue(d, &qstats, qlen) < 0)
return -1;
} else {
struct netdev_queue *dev_queue = mqprio_queue_get(sch, cl);
sch = dev_queue->qdisc_sleeping;
- sch->qstats.qlen = sch->q.qlen;
if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 ||
- gnet_stats_copy_queue(d, &sch->qstats) < 0)
+ gnet_stats_copy_queue(d, &sch->qstats, sch->q.qlen) < 0)
return -1;
}
return 0;
diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c
index 7f4e1d8..53357b3 100644
--- a/net/sched/sch_multiq.c
+++ b/net/sched/sch_multiq.c
@@ -360,9 +360,8 @@ static int multiq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
struct Qdisc *cl_q;
cl_q = q->queues[cl - 1];
- cl_q->qstats.qlen = cl_q->q.qlen;
if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 ||
- gnet_stats_copy_queue(d, &cl_q->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl_q->qstats, cl_q->q.qlen) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index b411e78..4644f55 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -324,9 +324,8 @@ static int prio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
struct Qdisc *cl_q;
cl_q = q->queues[cl - 1];
- cl_q->qstats.qlen = cl_q->q.qlen;
if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 ||
- gnet_stats_copy_queue(d, &cl_q->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl_q->qstats, cl_q->q.qlen) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index 3fb2655..66df9d9 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -664,14 +664,13 @@ static int qfq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
struct tc_qfq_stats xstats;
memset(&xstats, 0, sizeof(xstats));
- cl->qdisc->qstats.qlen = cl->qdisc->q.qlen;
xstats.weight = cl->agg->class_weight;
xstats.lmax = cl->agg->lmax;
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qdisc->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl->qdisc->qstats, cl->qdisc->q.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 16fb3e7..4805651 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -871,7 +871,7 @@ static int sfq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
qs.qlen = slot->qlen;
qs.backlog = slot->backlog;
}
- if (gnet_stats_copy_queue(d, &qs) < 0)
+ if (gnet_stats_copy_queue(d, &qs, qs.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
}
^ permalink raw reply related
* [net-next PATCH 4/4] net: sched: enable per cpu qstats
From: John Fastabend @ 2014-09-26 18:15 UTC (permalink / raw)
To: xiyou.wangcong, jhs, eric.dumazet, davem; +Cc: netdev
In-Reply-To: <20140926181346.28430.19380.stgit@nitbit.x32>
After previous patches to simplify qstats the qstats can be
made per cpu with a packed union in Qdisc struct.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
---
include/net/gen_stats.h | 3 ++
include/net/sch_generic.h | 12 +++++++++-
net/core/gen_stats.c | 55 +++++++++++++++++++++++++++++++++++++++------
net/sched/act_api.c | 2 +-
net/sched/sch_api.c | 12 ++++++++--
net/sched/sch_atm.c | 2 +-
net/sched/sch_cbq.c | 2 +-
net/sched/sch_drr.c | 2 +-
net/sched/sch_fq_codel.c | 2 +-
net/sched/sch_hfsc.c | 2 +-
net/sched/sch_htb.c | 2 +-
net/sched/sch_mq.c | 2 +-
net/sched/sch_mqprio.c | 5 ++--
net/sched/sch_multiq.c | 2 +-
net/sched/sch_prio.c | 2 +-
net/sched/sch_qfq.c | 3 ++
net/sched/sch_sfq.c | 2 +-
17 files changed, 87 insertions(+), 25 deletions(-)
diff --git a/include/net/gen_stats.h b/include/net/gen_stats.h
index de9b3dd..cbafa37 100644
--- a/include/net/gen_stats.h
+++ b/include/net/gen_stats.h
@@ -41,7 +41,8 @@ int gnet_stats_copy_rate_est(struct gnet_dump *d,
const struct gnet_stats_basic_packed *b,
struct gnet_stats_rate_est64 *r);
int gnet_stats_copy_queue(struct gnet_dump *d,
- struct gnet_stats_queue *q, __u32 len);
+ struct gnet_stats_queue __percpu *cpu_q,
+ struct gnet_stats_queue *q, __u32 qlen);
int gnet_stats_copy_app(struct gnet_dump *d, void *st, int len);
int gnet_stats_finish_copy(struct gnet_dump *d);
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 5609844..4d14c70 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -90,7 +90,10 @@ struct Qdisc {
struct gnet_stats_basic_cpu __percpu *cpu_bstats;
} __packed;
unsigned int __state;
- struct gnet_stats_queue qstats;
+ union {
+ struct gnet_stats_queue qstats;
+ struct gnet_stats_queue __percpu *cpu_qstats;
+ } __packed;
struct rcu_head rcu_head;
int padded;
atomic_t refcnt;
@@ -544,6 +547,13 @@ static inline void qdisc_qstats_drop(struct Qdisc *sch)
sch->qstats.drops++;
}
+static inline void qdisc_qstats_drop_cpu(struct Qdisc *sch)
+{
+ struct gnet_stats_queue *qstats = this_cpu_ptr(sch->cpu_qstats);
+
+ qstats->drops++;
+}
+
static inline void qdisc_qstats_overlimit(struct Qdisc *sch)
{
sch->qstats.overlimits++;
diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c
index ad3ecb6..14681b9 100644
--- a/net/core/gen_stats.c
+++ b/net/core/gen_stats.c
@@ -215,33 +215,74 @@ gnet_stats_copy_rate_est(struct gnet_dump *d,
}
EXPORT_SYMBOL(gnet_stats_copy_rate_est);
+static void
+__gnet_stats_copy_queue_cpu(struct gnet_stats_queue *qstats,
+ const struct gnet_stats_queue __percpu *q)
+{
+ int i;
+
+ for_each_possible_cpu(i) {
+ const struct gnet_stats_queue *qcpu = per_cpu_ptr(q, i);
+
+ qstats->qlen = 0;
+ qstats->backlog += qcpu->backlog;
+ qstats->drops += qcpu->drops;
+ qstats->requeues += qcpu->requeues;
+ qstats->overlimits += qcpu->overlimits;
+ }
+}
+
+static void __gnet_stats_copy_queue(struct gnet_stats_queue *qstats,
+ const struct gnet_stats_queue __percpu *cpu,
+ const struct gnet_stats_queue *q,
+ __u32 qlen)
+{
+ if (cpu) {
+ __gnet_stats_copy_queue_cpu(qstats, cpu);
+ } else {
+ qstats->qlen = q->qlen;
+ qstats->backlog = q->backlog;
+ qstats->drops = q->drops;
+ qstats->requeues = q->requeues;
+ qstats->overlimits = q->overlimits;
+ }
+
+ qstats->qlen = qlen;
+}
+
/**
* gnet_stats_copy_queue - copy queue statistics into statistics TLV
* @d: dumping handle
+ * @cpu_q: per cpu queue statistics
* @q: queue statistics
* @qlen: queue length statistics
*
* Appends the queue statistics to the top level TLV created by
- * gnet_stats_start_copy().
+ * gnet_stats_start_copy(). Using per cpu queue statistics if
+ * they are available.
*
* Returns 0 on success or -1 with the statistic lock released
* if the room in the socket buffer was not sufficient.
*/
int
gnet_stats_copy_queue(struct gnet_dump *d,
+ struct gnet_stats_queue __percpu *cpu_q,
struct gnet_stats_queue *q, __u32 qlen)
{
- q->qlen = qlen;
+ struct gnet_stats_queue qstats = {0};
+
+ __gnet_stats_copy_queue(&qstats, cpu_q, q, qlen);
if (d->compat_tc_stats) {
- d->tc_stats.drops = q->drops;
- d->tc_stats.qlen = q->qlen;
- d->tc_stats.backlog = q->backlog;
- d->tc_stats.overlimits = q->overlimits;
+ d->tc_stats.drops = qstats.drops;
+ d->tc_stats.qlen = qstats.qlen;
+ d->tc_stats.backlog = qstats.backlog;
+ d->tc_stats.overlimits = qstats.overlimits;
}
if (d->tail)
- return gnet_stats_copy(d, TCA_STATS_QUEUE, q, sizeof(*q));
+ return gnet_stats_copy(d, TCA_STATS_QUEUE,
+ &qstats, sizeof(qstats));
return 0;
}
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 2e13409..3d43e49 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -623,7 +623,7 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a,
if (gnet_stats_copy_basic(&d, NULL, &p->tcfc_bstats) < 0 ||
gnet_stats_copy_rate_est(&d, &p->tcfc_bstats,
&p->tcfc_rate_est) < 0 ||
- gnet_stats_copy_queue(&d,
+ gnet_stats_copy_queue(&d, NULL,
&p->tcfc_qstats,
p->tcfc_qstats.qlen) < 0)
goto errout;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index ca00ea8..aa83295 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -947,6 +947,10 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
alloc_percpu(struct gnet_stats_basic_cpu);
if (!sch->cpu_bstats)
goto err_out4;
+
+ sch->cpu_qstats = alloc_percpu(struct gnet_stats_queue);
+ if (!sch->cpu_qstats)
+ goto err_out4;
}
if (tca[TCA_STAB]) {
@@ -995,6 +999,7 @@ err_out:
err_out4:
free_percpu(sch->cpu_bstats);
+ free_percpu(sch->cpu_qstats);
/*
* Any broken qdiscs that would require a ops->reset() here?
* The qdisc was never in action so it shouldn't be necessary.
@@ -1313,6 +1318,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
u32 portid, u32 seq, u16 flags, int event)
{
struct gnet_stats_basic_cpu __percpu *cpu_bstats = NULL;
+ struct gnet_stats_queue __percpu *cpu_qstats = NULL;
struct tcmsg *tcm;
struct nlmsghdr *nlh;
unsigned char *b = skb_tail_pointer(skb);
@@ -1349,12 +1355,14 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0)
goto nla_put_failure;
- if (qdisc_is_percpu_stats(q))
+ if (qdisc_is_percpu_stats(q)) {
cpu_bstats = q->cpu_bstats;
+ cpu_qstats = q->cpu_qstats;
+ }
if (gnet_stats_copy_basic(&d, cpu_bstats, &q->bstats) < 0 ||
gnet_stats_copy_rate_est(&d, &q->bstats, &q->rate_est) < 0 ||
- gnet_stats_copy_queue(&d, &q->qstats, qlen) < 0)
+ gnet_stats_copy_queue(&d, cpu_qstats, &q->qstats, qlen) < 0)
goto nla_put_failure;
if (gnet_stats_finish_copy(&d) < 0)
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index c145eb6..e3e2cc5 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -638,7 +638,7 @@ atm_tc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
struct atm_flow_data *flow = (struct atm_flow_data *)arg;
if (gnet_stats_copy_basic(d, NULL, &flow->bstats) < 0 ||
- gnet_stats_copy_queue(d, &flow->qstats, flow->q->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &flow->qstats, flow->q->q.qlen) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index c610081..beeb75f 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1602,7 +1602,7 @@ cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qstats, cl->q->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &cl->qstats, cl->q->q.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats));
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index 5835a93..3387060 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -284,7 +284,7 @@ static int drr_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qdisc->qstats, qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &cl->qdisc->qstats, qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c
index 6a42112..301c352 100644
--- a/net/sched/sch_fq_codel.c
+++ b/net/sched/sch_fq_codel.c
@@ -550,7 +550,7 @@ static int fq_codel_dump_class_stats(struct Qdisc *sch, unsigned long cl,
qs.backlog = q->backlogs[idx];
qs.drops = flow->dropped;
}
- if (gnet_stats_copy_queue(d, &qs, 0) < 0)
+ if (gnet_stats_copy_queue(d, NULL, &qs, 0) < 0)
return -1;
if (idx < q->flows_cnt)
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index d364acb..e6c7416 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1378,7 +1378,7 @@ hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qstats, cl->qdisc->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &cl->qstats, cl->qdisc->q.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 3a691fd..f1acb0f 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1147,7 +1147,7 @@ htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d)
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, NULL, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qstats, qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &cl->qstats, qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats));
diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c
index 6416a69..f3cbaec 100644
--- a/net/sched/sch_mq.c
+++ b/net/sched/sch_mq.c
@@ -200,7 +200,7 @@ static int mq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
sch = dev_queue->qdisc_sleeping;
if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 ||
- gnet_stats_copy_queue(d, &sch->qstats, sch->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &sch->qstats, sch->q.qlen) < 0)
return -1;
return 0;
}
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index 03dbeb5..3811a74 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -356,14 +356,15 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
/* Reclaim root sleeping lock before completing stats */
spin_lock_bh(d->lock);
if (gnet_stats_copy_basic(d, NULL, &bstats) < 0 ||
- gnet_stats_copy_queue(d, &qstats, qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &qstats, qlen) < 0)
return -1;
} else {
struct netdev_queue *dev_queue = mqprio_queue_get(sch, cl);
sch = dev_queue->qdisc_sleeping;
if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 ||
- gnet_stats_copy_queue(d, &sch->qstats, sch->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL,
+ &sch->qstats, sch->q.qlen) < 0)
return -1;
}
return 0;
diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c
index 53357b3..42dd218 100644
--- a/net/sched/sch_multiq.c
+++ b/net/sched/sch_multiq.c
@@ -361,7 +361,7 @@ static int multiq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
cl_q = q->queues[cl - 1];
if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 ||
- gnet_stats_copy_queue(d, &cl_q->qstats, cl_q->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &cl_q->qstats, cl_q->q.qlen) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 4644f55..8e5cd34 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -325,7 +325,7 @@ static int prio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
cl_q = q->queues[cl - 1];
if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 ||
- gnet_stats_copy_queue(d, &cl_q->qstats, cl_q->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &cl_q->qstats, cl_q->q.qlen) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index 66df9d9..3ec7e88 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -670,7 +670,8 @@ static int qfq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qdisc->qstats, cl->qdisc->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL,
+ &cl->qdisc->qstats, cl->qdisc->q.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 4805651..5814c46 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -871,7 +871,7 @@ static int sfq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
qs.qlen = slot->qlen;
qs.backlog = slot->backlog;
}
- if (gnet_stats_copy_queue(d, &qs, qs.qlen) < 0)
+ if (gnet_stats_copy_queue(d, NULL, &qs, qs.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
}
^ permalink raw reply related
* Re: [net-next PATCH 1/4] net: sched: make bstats per cpu and estimator RCU safe
From: John Fastabend @ 2014-09-26 18:21 UTC (permalink / raw)
To: John Fastabend, xiyou.wangcong, jhs, eric.dumazet, davem; +Cc: netdev
In-Reply-To: <20140926181346.28430.19380.stgit@nitbit.x32>
On 09/26/2014 11:13 AM, John Fastabend wrote:
> From: John Fastabend <john.fastabend@gmail.com>
>
> In order to run qdisc's without locking statistics and estimators
> need to be handled correctly.
>
> To resolve bstats make the statistics per cpu. And because this is
> only needed for qdiscs that are running without locks which is not
> the case for most qdiscs in the near future only create percpu
> stats when qdiscs set the TCQ_F_CPUSTATS flag.
>
> Next because estimators use the bstats to calculate packets per
> second and bytes per second the estimator code paths are updated
> to use the per cpu statistics.
>
> Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
> ---
[...]
> return 0;
> diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
> index 37e7d25..2f0ecd7 100644
> --- a/net/sched/sch_mqprio.c
> +++ b/net/sched/sch_mqprio.c
> @@ -234,8 +234,8 @@ static int mqprio_dump(struct Qdisc *sch, struct sk_buff *skb)
> qdisc = rtnl_dereference(netdev_get_tx_queue(dev, i)->qdisc);
> spin_lock_bh(qdisc_lock(qdisc));
> sch->q.qlen += qdisc->q.qlen;
> - sch->bstats.bytes += qdisc->bstats.bytes;
> - sch->bstats.packets += qdisc->bstats.packets;
> + sch->bstats->bytes += qdisc->bstats.bytes;
> + sch->bstats->packets += qdisc->bstats.packets;
> sch->qstats.qlen += qdisc->qstats.qlen;
> sch->qstats.backlog += qdisc->qstats.backlog;
OK v2 coming this patch got messed up and then fixed in the third
patch.
If you look at the bstats definition you'll see this obviously
wont compile. The end result is good but it wont bisect.
.John
^ permalink raw reply
* Re: [RFC PATCH net-next v2 0/5] netns: allow to identify peer netns
From: Andy Lutomirski @ 2014-09-26 18:26 UTC (permalink / raw)
To: Eric W. Biederman
Cc: Andrew Morton, Network Development, Linux Containers,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Stephen Hemminger, Cong Wang, Linux API, Nicolas Dichtel,
David S. Miller
In-Reply-To: <87ppei45ig.fsf-JOvCrm2gF+uungPnsOpG7nhyD016LWXt@public.gmane.org>
On Fri, Sep 26, 2014 at 11:10 AM, Eric W. Biederman
<ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org> wrote:
> Nicolas Dichtel <nicolas.dichtel-pdR9zngts4EAvxtiuMwx3w@public.gmane.org> writes:
>
>> The goal of this serie is to be able to multicast netlink messages with an
>> attribute that identify a peer netns.
>> This is needed by the userland to interpret some informations contained in
>> netlink messages (like IFLA_LINK value, but also some other attributes in case
>> of x-netns netdevice (see also
>> http://thread.gmane.org/gmane.linux.network/315933/focus=316064 and
>> http://thread.gmane.org/gmane.linux.kernel.containers/28301/focus=4239)).
>
> I want say that the problem addressed by patch 3/5 of this series is a
> fundamentally valid problem. We have network objects spanning network
> namespaces and it would be very nice to be able to talk about them in
> netlink, and file descriptors are too local and argubably too heavy
> weight for netlink quires and especially for netlink broadcast messages.
>
> Furthermore the concept of ineternal concept of peernet2id seems valid.
>
> However what you do not address is a way for CRIU (aka process
> migration) to be able to restore these ids after process migration.
> Going farther it looks like you are actively breaking process migration
> at this time, making this set of patches a no-go.
>
> When adding a new form of namespace id CRIU patches are just about
> as necessary as iproute patches.
>
>> Ids are stored in the parent user namespace. These ids are valid only inside
>> this user namespace. The user can retrieve these ids via a new netlink messages,
>> but only if peer netns are in the same user namespace.
>
> That does not describe what you have actually implemented in the
> patches.
>
> I see two ways to go with this.
>
> - A per network namespace table to that you can store ids for ``peer''
> network namespaces. The table would need to be populated manually by
> the likes of ip netns add.
>
> That flips the order of assignment and makes this idea solid.
>
> Unfortunately in the case of a fully referencing mesh of N network
> namespaces such a mesh winds up taking O(N^2) space, which seems
> undesirable.
>
> - Add a netlink attribute that says this network element is in a peer
> network namespace.
>
> Add a unicast query message that let's you ask if the remote
> end of a tunnel is in a network namespace specified by file
> descriptor.
>
> I personally lean towards the second version as it is fundamentally
> simpler, and generally scales better, and the visibility controls are
> the existing visibility controls. The only downside is it requires
> a query after receiving a netlink broadcast message for the times that
> we care.
The downside of that approach, and all the similar kcmp stuff, is that
it scales poorly for applications using it. This is probably not the
end of the world, but it's not ideal.
--Andy
^ permalink raw reply
* [net-next PATCH v2 1/4] net: sched: make bstats per cpu and estimator RCU safe
From: John Fastabend @ 2014-09-26 18:36 UTC (permalink / raw)
To: xiyou.wangcong, jhs, eric.dumazet, davem; +Cc: netdev
From: John Fastabend <john.fastabend@gmail.com>
In order to run qdisc's without locking statistics and estimators
need to be handled correctly.
To resolve bstats make the statistics per cpu. And because this is
only needed for qdiscs that are running without locks which is not
the case for most qdiscs in the near future only create percpu
stats when qdiscs set the TCQ_F_CPUSTATS flag.
Next because estimators use the bstats to calculate packets per
second and bytes per second the estimator code paths are updated
to use the per cpu statistics.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
---
include/net/gen_stats.h | 11 +++++++++
include/net/sch_generic.h | 22 +++++++++++++++++-
net/core/gen_estimator.c | 29 +++++++++++++++---------
net/core/gen_stats.c | 53 ++++++++++++++++++++++++++++++++++++++++----
net/netfilter/xt_RATEEST.c | 2 +-
net/sched/act_api.c | 5 ++--
net/sched/act_police.c | 2 +-
net/sched/sch_api.c | 29 +++++++++++++++++++-----
net/sched/sch_atm.c | 2 +-
net/sched/sch_cbq.c | 7 +++---
net/sched/sch_drr.c | 7 +++---
net/sched/sch_generic.c | 3 ++
net/sched/sch_hfsc.c | 13 +++++++----
net/sched/sch_htb.c | 12 +++++++---
net/sched/sch_mq.c | 2 +-
net/sched/sch_mqprio.c | 4 ++-
net/sched/sch_multiq.c | 2 +-
net/sched/sch_prio.c | 2 +-
net/sched/sch_qfq.c | 8 ++++---
19 files changed, 164 insertions(+), 51 deletions(-)
diff --git a/include/net/gen_stats.h b/include/net/gen_stats.h
index ea4271d..ce3c128 100644
--- a/include/net/gen_stats.h
+++ b/include/net/gen_stats.h
@@ -6,6 +6,11 @@
#include <linux/rtnetlink.h>
#include <linux/pkt_sched.h>
+struct gnet_stats_basic_cpu {
+ struct gnet_stats_basic_packed bstats;
+ struct u64_stats_sync syncp;
+};
+
struct gnet_dump {
spinlock_t * lock;
struct sk_buff * skb;
@@ -27,7 +32,11 @@ int gnet_stats_start_copy_compat(struct sk_buff *skb, int type,
spinlock_t *lock, struct gnet_dump *d);
int gnet_stats_copy_basic(struct gnet_dump *d,
+ struct gnet_stats_basic_cpu __percpu *cpu,
struct gnet_stats_basic_packed *b);
+void __gnet_stats_copy_basic(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu,
+ struct gnet_stats_basic_packed *b);
int gnet_stats_copy_rate_est(struct gnet_dump *d,
const struct gnet_stats_basic_packed *b,
struct gnet_stats_rate_est64 *r);
@@ -37,11 +46,13 @@ int gnet_stats_copy_app(struct gnet_dump *d, void *st, int len);
int gnet_stats_finish_copy(struct gnet_dump *d);
int gen_new_estimator(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats,
struct gnet_stats_rate_est64 *rate_est,
spinlock_t *stats_lock, struct nlattr *opt);
void gen_kill_estimator(struct gnet_stats_basic_packed *bstats,
struct gnet_stats_rate_est64 *rate_est);
int gen_replace_estimator(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats,
struct gnet_stats_rate_est64 *rate_est,
spinlock_t *stats_lock, struct nlattr *opt);
bool gen_estimator_active(const struct gnet_stats_basic_packed *bstats,
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index e65b8e0..4b93511 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -6,6 +6,7 @@
#include <linux/rcupdate.h>
#include <linux/pkt_sched.h>
#include <linux/pkt_cls.h>
+#include <linux/percpu.h>
#include <net/gen_stats.h>
#include <net/rtnetlink.h>
@@ -58,6 +59,7 @@ struct Qdisc {
* multiqueue device.
*/
#define TCQ_F_WARN_NONWC (1 << 16)
+#define TCQ_F_CPUSTATS 0x20 /* run using percpu statistics */
u32 limit;
const struct Qdisc_ops *ops;
struct qdisc_size_table __rcu *stab;
@@ -83,7 +85,10 @@ struct Qdisc {
*/
unsigned long state;
struct sk_buff_head q;
- struct gnet_stats_basic_packed bstats;
+ union {
+ struct gnet_stats_basic_packed bstats;
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats;
+ } __packed;
unsigned int __state;
struct gnet_stats_queue qstats;
struct rcu_head rcu_head;
@@ -487,6 +492,10 @@ static inline int qdisc_enqueue_root(struct sk_buff *skb, struct Qdisc *sch)
return qdisc_enqueue(skb, sch) & NET_XMIT_MASK;
}
+static inline bool qdisc_is_percpu_stats(const struct Qdisc *q)
+{
+ return q->flags & TCQ_F_CPUSTATS;
+}
static inline void bstats_update(struct gnet_stats_basic_packed *bstats,
const struct sk_buff *skb)
@@ -495,6 +504,17 @@ static inline void bstats_update(struct gnet_stats_basic_packed *bstats,
bstats->packets += skb_is_gso(skb) ? skb_shinfo(skb)->gso_segs : 1;
}
+static inline void qdisc_bstats_update_cpu(struct Qdisc *sch,
+ const struct sk_buff *skb)
+{
+ struct gnet_stats_basic_cpu *bstats =
+ this_cpu_ptr(sch->cpu_bstats);
+
+ u64_stats_update_begin(&bstats->syncp);
+ bstats_update(&bstats->bstats, skb);
+ u64_stats_update_end(&bstats->syncp);
+}
+
static inline void qdisc_bstats_update(struct Qdisc *sch,
const struct sk_buff *skb)
{
diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c
index 9d33dff..9dfb88a 100644
--- a/net/core/gen_estimator.c
+++ b/net/core/gen_estimator.c
@@ -91,6 +91,8 @@ struct gen_estimator
u32 avpps;
struct rcu_head e_rcu;
struct rb_node node;
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats;
+ struct rcu_head head;
};
struct gen_estimator_head
@@ -115,9 +117,8 @@ static void est_timer(unsigned long arg)
rcu_read_lock();
list_for_each_entry_rcu(e, &elist[idx].list, list) {
- u64 nbytes;
+ struct gnet_stats_basic_packed b = {0};
u64 brate;
- u32 npackets;
u32 rate;
spin_lock(e->stats_lock);
@@ -125,15 +126,15 @@ static void est_timer(unsigned long arg)
if (e->bstats == NULL)
goto skip;
- nbytes = e->bstats->bytes;
- npackets = e->bstats->packets;
- brate = (nbytes - e->last_bytes)<<(7 - idx);
- e->last_bytes = nbytes;
+ __gnet_stats_copy_basic(&b, e->cpu_bstats, e->bstats);
+
+ brate = (b.bytes - e->last_bytes)<<(7 - idx);
+ e->last_bytes = b.bytes;
e->avbps += (brate >> e->ewma_log) - (e->avbps >> e->ewma_log);
e->rate_est->bps = (e->avbps+0xF)>>5;
- rate = (npackets - e->last_packets)<<(12 - idx);
- e->last_packets = npackets;
+ rate = (b.packets - e->last_packets)<<(12 - idx);
+ e->last_packets = b.packets;
e->avpps += (rate >> e->ewma_log) - (e->avpps >> e->ewma_log);
e->rate_est->pps = (e->avpps+0x1FF)>>10;
skip:
@@ -203,12 +204,14 @@ struct gen_estimator *gen_find_node(const struct gnet_stats_basic_packed *bstats
*
*/
int gen_new_estimator(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats,
struct gnet_stats_rate_est64 *rate_est,
spinlock_t *stats_lock,
struct nlattr *opt)
{
struct gen_estimator *est;
struct gnet_estimator *parm = nla_data(opt);
+ struct gnet_stats_basic_packed b = {0};
int idx;
if (nla_len(opt) < sizeof(*parm))
@@ -221,15 +224,18 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats,
if (est == NULL)
return -ENOBUFS;
+ __gnet_stats_copy_basic(&b, cpu_bstats, bstats);
+
idx = parm->interval + 2;
est->bstats = bstats;
est->rate_est = rate_est;
est->stats_lock = stats_lock;
est->ewma_log = parm->ewma_log;
- est->last_bytes = bstats->bytes;
+ est->last_bytes = b.bytes;
est->avbps = rate_est->bps<<5;
- est->last_packets = bstats->packets;
+ est->last_packets = b.packets;
est->avpps = rate_est->pps<<10;
+ est->cpu_bstats = cpu_bstats;
spin_lock_bh(&est_tree_lock);
if (!elist[idx].timer.function) {
@@ -290,11 +296,12 @@ EXPORT_SYMBOL(gen_kill_estimator);
* Returns 0 on success or a negative error code.
*/
int gen_replace_estimator(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats,
struct gnet_stats_rate_est64 *rate_est,
spinlock_t *stats_lock, struct nlattr *opt)
{
gen_kill_estimator(bstats, rate_est);
- return gen_new_estimator(bstats, rate_est, stats_lock, opt);
+ return gen_new_estimator(bstats, cpu_bstats, rate_est, stats_lock, opt);
}
EXPORT_SYMBOL(gen_replace_estimator);
diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c
index 2ddbce4..5ff8e80 100644
--- a/net/core/gen_stats.c
+++ b/net/core/gen_stats.c
@@ -97,6 +97,43 @@ gnet_stats_start_copy(struct sk_buff *skb, int type, spinlock_t *lock,
}
EXPORT_SYMBOL(gnet_stats_start_copy);
+static void
+__gnet_stats_copy_basic_cpu(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu)
+{
+ int i;
+
+ for_each_possible_cpu(i) {
+ struct gnet_stats_basic_cpu *bcpu = per_cpu_ptr(cpu, i);
+ unsigned int start;
+ __u64 bytes;
+ __u32 packets;
+
+ do {
+ start = u64_stats_fetch_begin_irq(&bcpu->syncp);
+ bytes = bcpu->bstats.bytes;
+ packets = bcpu->bstats.packets;
+ } while (u64_stats_fetch_retry_irq(&bcpu->syncp, start));
+
+ bstats->bytes += bcpu->bstats.bytes;
+ bstats->packets += bcpu->bstats.packets;
+ }
+}
+
+void
+__gnet_stats_copy_basic(struct gnet_stats_basic_packed *bstats,
+ struct gnet_stats_basic_cpu __percpu *cpu,
+ struct gnet_stats_basic_packed *b)
+{
+ if (cpu) {
+ __gnet_stats_copy_basic_cpu(bstats, cpu);
+ } else {
+ bstats->bytes = b->bytes;
+ bstats->packets = b->packets;
+ }
+}
+EXPORT_SYMBOL(__gnet_stats_copy_basic);
+
/**
* gnet_stats_copy_basic - copy basic statistics into statistic TLV
* @d: dumping handle
@@ -109,19 +146,25 @@ EXPORT_SYMBOL(gnet_stats_start_copy);
* if the room in the socket buffer was not sufficient.
*/
int
-gnet_stats_copy_basic(struct gnet_dump *d, struct gnet_stats_basic_packed *b)
+gnet_stats_copy_basic(struct gnet_dump *d,
+ struct gnet_stats_basic_cpu __percpu *cpu,
+ struct gnet_stats_basic_packed *b)
{
+ struct gnet_stats_basic_packed bstats = {0};
+
+ __gnet_stats_copy_basic(&bstats, cpu, b);
+
if (d->compat_tc_stats) {
- d->tc_stats.bytes = b->bytes;
- d->tc_stats.packets = b->packets;
+ d->tc_stats.bytes = bstats.bytes;
+ d->tc_stats.packets = bstats.packets;
}
if (d->tail) {
struct gnet_stats_basic sb;
memset(&sb, 0, sizeof(sb));
- sb.bytes = b->bytes;
- sb.packets = b->packets;
+ sb.bytes = bstats.bytes;
+ sb.packets = bstats.packets;
return gnet_stats_copy(d, TCA_STATS_BASIC, &sb, sizeof(sb));
}
return 0;
diff --git a/net/netfilter/xt_RATEEST.c b/net/netfilter/xt_RATEEST.c
index 370adf6..604df6f 100644
--- a/net/netfilter/xt_RATEEST.c
+++ b/net/netfilter/xt_RATEEST.c
@@ -136,7 +136,7 @@ static int xt_rateest_tg_checkentry(const struct xt_tgchk_param *par)
cfg.est.interval = info->interval;
cfg.est.ewma_log = info->ewma_log;
- ret = gen_new_estimator(&est->bstats, &est->rstats,
+ ret = gen_new_estimator(&est->bstats, NULL, &est->rstats,
&est->lock, &cfg.opt);
if (ret < 0)
goto err2;
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 648778a..eca4cf9 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -252,7 +252,8 @@ int tcf_hash_create(u32 index, struct nlattr *est, struct tc_action *a,
p->tcfc_tm.install = jiffies;
p->tcfc_tm.lastuse = jiffies;
if (est) {
- int err = gen_new_estimator(&p->tcfc_bstats, &p->tcfc_rate_est,
+ int err = gen_new_estimator(&p->tcfc_bstats, NULL,
+ &p->tcfc_rate_est,
&p->tcfc_lock, est);
if (err) {
kfree(p);
@@ -619,7 +620,7 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a,
if (err < 0)
goto errout;
- if (gnet_stats_copy_basic(&d, &p->tcfc_bstats) < 0 ||
+ if (gnet_stats_copy_basic(&d, NULL, &p->tcfc_bstats) < 0 ||
gnet_stats_copy_rate_est(&d, &p->tcfc_bstats,
&p->tcfc_rate_est) < 0 ||
gnet_stats_copy_queue(&d, &p->tcfc_qstats) < 0)
diff --git a/net/sched/act_police.c b/net/sched/act_police.c
index f32bcb0..69791ca 100644
--- a/net/sched/act_police.c
+++ b/net/sched/act_police.c
@@ -178,7 +178,7 @@ override:
spin_lock_bh(&police->tcf_lock);
if (est) {
- err = gen_replace_estimator(&police->tcf_bstats,
+ err = gen_replace_estimator(&police->tcf_bstats, NULL,
&police->tcf_rate_est,
&police->tcf_lock, est);
if (err)
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 15e7bee..a95e3b4 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -942,6 +942,13 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
sch->handle = handle;
if (!ops->init || (err = ops->init(sch, tca[TCA_OPTIONS])) == 0) {
+ if (qdisc_is_percpu_stats(sch)) {
+ sch->cpu_bstats =
+ alloc_percpu(struct gnet_stats_basic_cpu);
+ if (!sch->cpu_bstats)
+ goto err_out4;
+ }
+
if (tca[TCA_STAB]) {
stab = qdisc_get_stab(tca[TCA_STAB]);
if (IS_ERR(stab)) {
@@ -964,8 +971,11 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
else
root_lock = qdisc_lock(sch);
- err = gen_new_estimator(&sch->bstats, &sch->rate_est,
- root_lock, tca[TCA_RATE]);
+ err = gen_new_estimator(&sch->bstats,
+ sch->cpu_bstats,
+ &sch->rate_est,
+ root_lock,
+ tca[TCA_RATE]);
if (err)
goto err_out4;
}
@@ -984,6 +994,7 @@ err_out:
return NULL;
err_out4:
+ free_percpu(sch->cpu_bstats);
/*
* Any broken qdiscs that would require a ops->reset() here?
* The qdisc was never in action so it shouldn't be necessary.
@@ -1022,9 +1033,11 @@ static int qdisc_change(struct Qdisc *sch, struct nlattr **tca)
because change can't be undone. */
if (sch->flags & TCQ_F_MQROOT)
goto out;
- gen_replace_estimator(&sch->bstats, &sch->rate_est,
- qdisc_root_sleeping_lock(sch),
- tca[TCA_RATE]);
+ gen_replace_estimator(&sch->bstats,
+ sch->cpu_bstats,
+ &sch->rate_est,
+ qdisc_root_sleeping_lock(sch),
+ tca[TCA_RATE]);
}
out:
return 0;
@@ -1299,6 +1312,7 @@ graft:
static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
u32 portid, u32 seq, u16 flags, int event)
{
+ struct gnet_stats_basic_cpu __percpu *cpu_bstats = NULL;
struct tcmsg *tcm;
struct nlmsghdr *nlh;
unsigned char *b = skb_tail_pointer(skb);
@@ -1334,7 +1348,10 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0)
goto nla_put_failure;
- if (gnet_stats_copy_basic(&d, &q->bstats) < 0 ||
+ if (qdisc_is_percpu_stats(q))
+ cpu_bstats = q->cpu_bstats;
+
+ if (gnet_stats_copy_basic(&d, cpu_bstats, &q->bstats) < 0 ||
gnet_stats_copy_rate_est(&d, &q->bstats, &q->rate_est) < 0 ||
gnet_stats_copy_queue(&d, &q->qstats) < 0)
goto nla_put_failure;
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index c398f9c..0101766 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -639,7 +639,7 @@ atm_tc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
flow->qstats.qlen = flow->q->q.qlen;
- if (gnet_stats_copy_basic(d, &flow->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &flow->bstats) < 0 ||
gnet_stats_copy_queue(d, &flow->qstats) < 0)
return -1;
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index d2cd981..22a3a02 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1601,7 +1601,7 @@ cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (cl->undertime != PSCHED_PASTPERFECT)
cl->xstats.undertime = cl->undertime - q->now;
- if (gnet_stats_copy_basic(d, &cl->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
gnet_stats_copy_queue(d, &cl->qstats) < 0)
return -1;
@@ -1759,7 +1759,8 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
}
if (tca[TCA_RATE]) {
- err = gen_replace_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_replace_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err) {
@@ -1852,7 +1853,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
goto failure;
if (tca[TCA_RATE]) {
- err = gen_new_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err) {
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index d8b5ccf..7a6243c 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -88,7 +88,8 @@ static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
if (cl != NULL) {
if (tca[TCA_RATE]) {
- err = gen_replace_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_replace_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err)
@@ -116,7 +117,7 @@ static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
cl->qdisc = &noop_qdisc;
if (tca[TCA_RATE]) {
- err = gen_replace_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_replace_estimator(&cl->bstats, NULL, &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err) {
@@ -282,7 +283,7 @@ static int drr_dump_class_stats(struct Qdisc *sch, unsigned long arg,
cl->qdisc->qstats.qlen = cl->qdisc->q.qlen;
}
- if (gnet_stats_copy_basic(d, &cl->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
gnet_stats_copy_queue(d, &cl->qdisc->qstats) < 0)
return -1;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 11b28f6..7c8e5d7 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -632,6 +632,9 @@ static void qdisc_rcu_free(struct rcu_head *head)
{
struct Qdisc *qdisc = container_of(head, struct Qdisc, rcu_head);
+ if (qdisc_is_percpu_stats(qdisc))
+ free_percpu(qdisc->cpu_bstats);
+
kfree((char *) qdisc - qdisc->padded);
}
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 04b0de4..209b966 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1014,9 +1014,12 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
cur_time = psched_get_time();
if (tca[TCA_RATE]) {
- err = gen_replace_estimator(&cl->bstats, &cl->rate_est,
- qdisc_root_sleeping_lock(sch),
- tca[TCA_RATE]);
+ spinlock_t *lock = qdisc_root_sleeping_lock(sch);
+
+ err = gen_replace_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
+ lock,
+ tca[TCA_RATE]);
if (err)
return err;
}
@@ -1063,7 +1066,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
return -ENOBUFS;
if (tca[TCA_RATE]) {
- err = gen_new_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err) {
@@ -1374,7 +1377,7 @@ hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
xstats.work = cl->cl_total;
xstats.rtwork = cl->cl_cumul;
- if (gnet_stats_copy_basic(d, &cl->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
gnet_stats_copy_queue(d, &cl->qstats) < 0)
return -1;
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 063e953..0256dee 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1144,7 +1144,7 @@ htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d)
cl->xstats.tokens = PSCHED_NS2TICKS(cl->tokens);
cl->xstats.ctokens = PSCHED_NS2TICKS(cl->ctokens);
- if (gnet_stats_copy_basic(d, &cl->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, NULL, &cl->rate_est) < 0 ||
gnet_stats_copy_queue(d, &cl->qstats) < 0)
return -1;
@@ -1402,7 +1402,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
goto failure;
if (htb_rate_est || tca[TCA_RATE]) {
- err = gen_new_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_new_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE] ? : &est.nla);
if (err) {
@@ -1464,8 +1465,11 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
parent->children++;
} else {
if (tca[TCA_RATE]) {
- err = gen_replace_estimator(&cl->bstats, &cl->rate_est,
- qdisc_root_sleeping_lock(sch),
+ spinlock_t *lock = qdisc_root_sleeping_lock(sch);
+
+ err = gen_replace_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
+ lock,
tca[TCA_RATE]);
if (err)
return err;
diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c
index a8b2864..d3a27fb 100644
--- a/net/sched/sch_mq.c
+++ b/net/sched/sch_mq.c
@@ -201,7 +201,7 @@ static int mq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
sch = dev_queue->qdisc_sleeping;
sch->qstats.qlen = sch->q.qlen;
- if (gnet_stats_copy_basic(d, &sch->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 ||
gnet_stats_copy_queue(d, &sch->qstats) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index 37e7d25..8917372 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -355,7 +355,7 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
}
/* Reclaim root sleeping lock before completing stats */
spin_lock_bh(d->lock);
- if (gnet_stats_copy_basic(d, &bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &bstats) < 0 ||
gnet_stats_copy_queue(d, &qstats) < 0)
return -1;
} else {
@@ -363,7 +363,7 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
sch = dev_queue->qdisc_sleeping;
sch->qstats.qlen = sch->q.qlen;
- if (gnet_stats_copy_basic(d, &sch->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 ||
gnet_stats_copy_queue(d, &sch->qstats) < 0)
return -1;
}
diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c
index c0466c1..4adbf7f 100644
--- a/net/sched/sch_multiq.c
+++ b/net/sched/sch_multiq.c
@@ -361,7 +361,7 @@ static int multiq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
cl_q = q->queues[cl - 1];
cl_q->qstats.qlen = cl_q->q.qlen;
- if (gnet_stats_copy_basic(d, &cl_q->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 ||
gnet_stats_copy_queue(d, &cl_q->qstats) < 0)
return -1;
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 03ef99e..68a8f25 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -325,7 +325,7 @@ static int prio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
cl_q = q->queues[cl - 1];
cl_q->qstats.qlen = cl_q->q.qlen;
- if (gnet_stats_copy_basic(d, &cl_q->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 ||
gnet_stats_copy_queue(d, &cl_q->qstats) < 0)
return -1;
diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index 602ea01..d59f857 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -459,7 +459,8 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
if (cl != NULL) { /* modify existing class */
if (tca[TCA_RATE]) {
- err = gen_replace_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_replace_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err)
@@ -484,7 +485,8 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
cl->qdisc = &noop_qdisc;
if (tca[TCA_RATE]) {
- err = gen_new_estimator(&cl->bstats, &cl->rate_est,
+ err = gen_new_estimator(&cl->bstats, NULL,
+ &cl->rate_est,
qdisc_root_sleeping_lock(sch),
tca[TCA_RATE]);
if (err)
@@ -667,7 +669,7 @@ static int qfq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
xstats.weight = cl->agg->class_weight;
xstats.lmax = cl->agg->lmax;
- if (gnet_stats_copy_basic(d, &cl->bstats) < 0 ||
+ if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
gnet_stats_copy_queue(d, &cl->qdisc->qstats) < 0)
return -1;
^ permalink raw reply related
* Re: [net-next PATCH 2/4] net: sched: implement qstat helper routines
From: Eric Dumazet @ 2014-09-26 18:36 UTC (permalink / raw)
To: John Fastabend; +Cc: xiyou.wangcong, jhs, davem, netdev
In-Reply-To: <20140926181440.28430.71445.stgit@nitbit.x32>
On Fri, 2014-09-26 at 11:14 -0700, John Fastabend wrote:
> This adds helpers to manipulate qstats logic and replaces locations
> that touch the counters directly. This simplifies future patches
> to push qstats onto per cpu counters.
>
> Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
> ---
> include/net/sch_generic.h | 43 +++++++++++++++++++++++++++++++++++--------
> net/sched/sch_api.c | 2 +-
> net/sched/sch_atm.c | 2 +-
> net/sched/sch_cbq.c | 10 +++++-----
> net/sched/sch_choke.c | 14 +++++++-------
> net/sched/sch_codel.c | 2 +-
> net/sched/sch_drr.c | 4 ++--
> net/sched/sch_dsmark.c | 2 +-
> net/sched/sch_fifo.c | 2 +-
> net/sched/sch_fq.c | 4 ++--
> net/sched/sch_fq_codel.c | 8 ++++----
> net/sched/sch_gred.c | 4 ++--
> net/sched/sch_hfsc.c | 8 ++++----
> net/sched/sch_hhf.c | 8 ++++----
> net/sched/sch_htb.c | 6 +++---
> net/sched/sch_ingress.c | 2 +-
> net/sched/sch_multiq.c | 4 ++--
> net/sched/sch_netem.c | 15 +++++++--------
> net/sched/sch_pie.c | 2 +-
> net/sched/sch_prio.c | 4 ++--
> net/sched/sch_qfq.c | 4 ++--
> net/sched/sch_red.c | 8 ++++----
> net/sched/sch_sfb.c | 10 +++++-----
> net/sched/sch_sfq.c | 17 +++++++++--------
> net/sched/sch_tbf.c | 8 ++++----
> 25 files changed, 110 insertions(+), 83 deletions(-)
>
> diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
> index 4b93511..5609844 100644
> --- a/include/net/sch_generic.h
> +++ b/include/net/sch_generic.h
> @@ -521,11 +521,39 @@ static inline void qdisc_bstats_update(struct Qdisc *sch,
> bstats_update(&sch->bstats, skb);
> }
>
> +static inline void qdisc_qstats_backlog(struct Qdisc *sch,
> + const struct sk_buff *skb)
qdisc_qstats_backlog_dec() to be consistent with
qdisc_qstats_backlog_dec() ?
> +{
> + sch->qstats.backlog -= qdisc_pkt_len(skb);
> +}
> +
> +static inline void qdisc_qstats_backlog_inc(struct Qdisc *sch,
> + const struct sk_buff *skb)
> +{
> +
> + sch->qstats.backlog += qdisc_pkt_len(skb);
> +}
> +
> +static inline void __qdisc_qstats_drop(struct Qdisc *sch, int count)
> +{
> + sch->qstats.drops += count;
> +}
> +
> +static inline void qdisc_qstats_drop(struct Qdisc *sch)
> +{
> + sch->qstats.drops++;
> +}
> +
> +static inline void qdisc_qstats_overlimit(struct Qdisc *sch)
> +{
> + sch->qstats.overlimits++;
> +}
> +
> static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
> struct sk_buff_head *list)
> {
> __skb_queue_tail(list, skb);
> - sch->qstats.backlog += qdisc_pkt_len(skb);
> + qdisc_qstats_backlog_inc(sch, skb);
>
> return NET_XMIT_SUCCESS;
> }
> @@ -541,7 +569,7 @@ static inline struct sk_buff *__qdisc_dequeue_head(struct Qdisc *sch,
> struct sk_buff *skb = __skb_dequeue(list);
>
> if (likely(skb != NULL)) {
> - sch->qstats.backlog -= qdisc_pkt_len(skb);
> + qdisc_qstats_backlog(sch, skb);
> qdisc_bstats_update(sch, skb);
> }
>
> @@ -559,10 +587,9 @@ static inline unsigned int __qdisc_queue_drop_head(struct Qdisc *sch,
> struct sk_buff *skb = __skb_dequeue(list);
>
> if (likely(skb != NULL)) {
> - unsigned int len = qdisc_pkt_len(skb);
> - sch->qstats.backlog -= len;
> + qdisc_qstats_backlog(sch, skb);
> kfree_skb(skb);
> - return len;
> + return qdisc_pkt_len(skb);
Well, you just added a use after free.
> }
>
> return 0;
> @@ -579,7 +606,7 @@ static inline struct sk_buff *__qdisc_dequeue_tail(struct Qdisc *sch,
> struct sk_buff *skb = __skb_dequeue_tail(list);
>
> if (likely(skb != NULL))
Rest of the patch seems fine, thanks !
^ permalink raw reply
* [net-next PATCH v2 2/4] net: sched: implement qstat helper routines
From: John Fastabend @ 2014-09-26 18:36 UTC (permalink / raw)
To: xiyou.wangcong, jhs, eric.dumazet, davem; +Cc: netdev
In-Reply-To: <20140926183621.2621.72538.stgit@nitbit.x32>
This adds helpers to manipulate qstats logic and replaces locations
that touch the counters directly. This simplifies future patches
to push qstats onto per cpu counters.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
---
include/net/sch_generic.h | 43 +++++++++++++++++++++++++++++++++++--------
net/sched/sch_api.c | 2 +-
net/sched/sch_atm.c | 2 +-
net/sched/sch_cbq.c | 10 +++++-----
net/sched/sch_choke.c | 14 +++++++-------
net/sched/sch_codel.c | 2 +-
net/sched/sch_drr.c | 4 ++--
net/sched/sch_dsmark.c | 2 +-
net/sched/sch_fifo.c | 2 +-
net/sched/sch_fq.c | 4 ++--
net/sched/sch_fq_codel.c | 8 ++++----
net/sched/sch_gred.c | 4 ++--
net/sched/sch_hfsc.c | 8 ++++----
net/sched/sch_hhf.c | 8 ++++----
net/sched/sch_htb.c | 6 +++---
net/sched/sch_ingress.c | 2 +-
net/sched/sch_multiq.c | 4 ++--
net/sched/sch_netem.c | 15 +++++++--------
net/sched/sch_pie.c | 2 +-
net/sched/sch_prio.c | 4 ++--
net/sched/sch_qfq.c | 4 ++--
net/sched/sch_red.c | 8 ++++----
net/sched/sch_sfb.c | 10 +++++-----
net/sched/sch_sfq.c | 17 +++++++++--------
net/sched/sch_tbf.c | 8 ++++----
25 files changed, 110 insertions(+), 83 deletions(-)
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 4b93511..5609844 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -521,11 +521,39 @@ static inline void qdisc_bstats_update(struct Qdisc *sch,
bstats_update(&sch->bstats, skb);
}
+static inline void qdisc_qstats_backlog(struct Qdisc *sch,
+ const struct sk_buff *skb)
+{
+ sch->qstats.backlog -= qdisc_pkt_len(skb);
+}
+
+static inline void qdisc_qstats_backlog_inc(struct Qdisc *sch,
+ const struct sk_buff *skb)
+{
+
+ sch->qstats.backlog += qdisc_pkt_len(skb);
+}
+
+static inline void __qdisc_qstats_drop(struct Qdisc *sch, int count)
+{
+ sch->qstats.drops += count;
+}
+
+static inline void qdisc_qstats_drop(struct Qdisc *sch)
+{
+ sch->qstats.drops++;
+}
+
+static inline void qdisc_qstats_overlimit(struct Qdisc *sch)
+{
+ sch->qstats.overlimits++;
+}
+
static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
struct sk_buff_head *list)
{
__skb_queue_tail(list, skb);
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
return NET_XMIT_SUCCESS;
}
@@ -541,7 +569,7 @@ static inline struct sk_buff *__qdisc_dequeue_head(struct Qdisc *sch,
struct sk_buff *skb = __skb_dequeue(list);
if (likely(skb != NULL)) {
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
qdisc_bstats_update(sch, skb);
}
@@ -559,10 +587,9 @@ static inline unsigned int __qdisc_queue_drop_head(struct Qdisc *sch,
struct sk_buff *skb = __skb_dequeue(list);
if (likely(skb != NULL)) {
- unsigned int len = qdisc_pkt_len(skb);
- sch->qstats.backlog -= len;
+ qdisc_qstats_backlog(sch, skb);
kfree_skb(skb);
- return len;
+ return qdisc_pkt_len(skb);
}
return 0;
@@ -579,7 +606,7 @@ static inline struct sk_buff *__qdisc_dequeue_tail(struct Qdisc *sch,
struct sk_buff *skb = __skb_dequeue_tail(list);
if (likely(skb != NULL))
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
return skb;
}
@@ -661,14 +688,14 @@ static inline unsigned int qdisc_queue_drop(struct Qdisc *sch)
static inline int qdisc_drop(struct sk_buff *skb, struct Qdisc *sch)
{
kfree_skb(skb);
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return NET_XMIT_DROP;
}
static inline int qdisc_reshape_fail(struct sk_buff *skb, struct Qdisc *sch)
{
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
#ifdef CONFIG_NET_CLS_ACT
if (sch->reshape_fail == NULL || sch->reshape_fail(skb, sch))
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index a95e3b4..2862bc6 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -763,7 +763,7 @@ void qdisc_tree_decrease_qlen(struct Qdisc *sch, unsigned int n)
cops->put(sch, cl);
}
sch->q.qlen -= n;
- sch->qstats.drops += drops;
+ __qdisc_qstats_drop(sch, drops);
}
}
EXPORT_SYMBOL(qdisc_tree_decrease_qlen);
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index 0101766..040212c 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -417,7 +417,7 @@ done:
if (ret != NET_XMIT_SUCCESS) {
drop: __maybe_unused
if (net_xmit_drop_count(ret)) {
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
if (flow)
flow->qstats.drops++;
}
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 22a3a02..60432c3 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -377,7 +377,7 @@ cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
#endif
if (cl == NULL) {
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
@@ -395,7 +395,7 @@ cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
}
if (net_xmit_drop_count(ret)) {
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
cbq_mark_toplevel(q, cl);
cl->qstats.drops++;
}
@@ -650,11 +650,11 @@ static int cbq_reshape_fail(struct sk_buff *skb, struct Qdisc *child)
return 0;
}
if (net_xmit_drop_count(ret))
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return 0;
}
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return -1;
}
#endif
@@ -995,7 +995,7 @@ cbq_dequeue(struct Qdisc *sch)
*/
if (sch->q.qlen) {
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (q->wd_expires)
qdisc_watchdog_schedule(&q->watchdog,
now + q->wd_expires);
diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c
index 8abc262..a514f42 100644
--- a/net/sched/sch_choke.c
+++ b/net/sched/sch_choke.c
@@ -127,7 +127,7 @@ static void choke_drop_by_idx(struct Qdisc *sch, unsigned int idx)
if (idx == q->tail)
choke_zap_tail_holes(q);
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
qdisc_drop(skb, sch);
qdisc_tree_decrease_qlen(sch, 1);
--sch->q.qlen;
@@ -302,7 +302,7 @@ static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch)
if (q->vars.qavg > p->qth_max) {
q->vars.qcount = -1;
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (use_harddrop(q) || !use_ecn(q) ||
!INET_ECN_set_ce(skb)) {
q->stats.forced_drop++;
@@ -315,7 +315,7 @@ static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch)
q->vars.qcount = 0;
q->vars.qR = red_random(p);
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (!use_ecn(q) || !INET_ECN_set_ce(skb)) {
q->stats.prob_drop++;
goto congestion_drop;
@@ -332,7 +332,7 @@ static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch)
q->tab[q->tail] = skb;
q->tail = (q->tail + 1) & q->tab_mask;
++sch->q.qlen;
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
return NET_XMIT_SUCCESS;
}
@@ -345,7 +345,7 @@ congestion_drop:
other_drop:
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
@@ -365,7 +365,7 @@ static struct sk_buff *choke_dequeue(struct Qdisc *sch)
q->tab[q->head] = NULL;
choke_zap_head_holes(q);
--sch->q.qlen;
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
qdisc_bstats_update(sch, skb);
return skb;
@@ -460,7 +460,7 @@ static int choke_change(struct Qdisc *sch, struct nlattr *opt)
ntab[tail++] = skb;
continue;
}
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
--sch->q.qlen;
qdisc_drop(skb, sch);
}
diff --git a/net/sched/sch_codel.c b/net/sched/sch_codel.c
index 2f9ab17..c56ac44 100644
--- a/net/sched/sch_codel.c
+++ b/net/sched/sch_codel.c
@@ -149,7 +149,7 @@ static int codel_change(struct Qdisc *sch, struct nlattr *opt)
while (sch->q.qlen > sch->limit) {
struct sk_buff *skb = __skb_dequeue(&sch->q);
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
qdisc_drop(skb, sch);
}
qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen);
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index 7a6243c..907b12f 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -360,7 +360,7 @@ static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch)
cl = drr_classify(skb, sch, &err);
if (cl == NULL) {
if (err & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return err;
}
@@ -369,7 +369,7 @@ static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch)
if (unlikely(err != NET_XMIT_SUCCESS)) {
if (net_xmit_drop_count(err)) {
cl->qstats.drops++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
}
return err;
}
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index 485e456..227114f 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -258,7 +258,7 @@ static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch)
err = qdisc_enqueue(skb, p->q);
if (err != NET_XMIT_SUCCESS) {
if (net_xmit_drop_count(err))
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return err;
}
diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c
index e15a9eb..2e2398c 100644
--- a/net/sched/sch_fifo.c
+++ b/net/sched/sch_fifo.c
@@ -42,7 +42,7 @@ static int pfifo_tail_enqueue(struct sk_buff *skb, struct Qdisc *sch)
/* queue full, remove one skb to fulfill the limit */
__qdisc_queue_drop_head(sch, &sch->q);
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
qdisc_enqueue_tail(skb, sch);
return NET_XMIT_CN;
diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c
index e12f997..a5a2e9f 100644
--- a/net/sched/sch_fq.c
+++ b/net/sched/sch_fq.c
@@ -290,7 +290,7 @@ static struct sk_buff *fq_dequeue_head(struct Qdisc *sch, struct fq_flow *flow)
flow->head = skb->next;
skb->next = NULL;
flow->qlen--;
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
sch->q.qlen--;
}
return skb;
@@ -371,7 +371,7 @@ static int fq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
f->qlen++;
if (skb_is_retransmit(skb))
q->stat_tcp_retrans++;
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
if (fq_flow_is_detached(f)) {
fq_flow_add_tail(&q->new_flows, f);
if (time_after(jiffies, f->age + q->flow_refill_delay))
diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c
index 105cf55..3f288ae 100644
--- a/net/sched/sch_fq_codel.c
+++ b/net/sched/sch_fq_codel.c
@@ -164,8 +164,8 @@ static unsigned int fq_codel_drop(struct Qdisc *sch)
q->backlogs[idx] -= len;
kfree_skb(skb);
sch->q.qlen--;
- sch->qstats.drops++;
- sch->qstats.backlog -= len;
+ qdisc_qstats_drop(sch);
+ qdisc_qstats_backlog(sch, skb);
flow->dropped++;
return idx;
}
@@ -180,7 +180,7 @@ static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch)
idx = fq_codel_classify(skb, sch, &ret);
if (idx == 0) {
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
@@ -190,7 +190,7 @@ static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch)
flow = &q->flows[idx];
flow_queue_add(flow, skb);
q->backlogs[idx] += qdisc_pkt_len(skb);
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
if (list_empty(&flow->flowchain)) {
list_add_tail(&flow->flowchain, &q->new_flows);
diff --git a/net/sched/sch_gred.c b/net/sched/sch_gred.c
index 12cbc09..a4ca451 100644
--- a/net/sched/sch_gred.c
+++ b/net/sched/sch_gred.c
@@ -209,7 +209,7 @@ static int gred_enqueue(struct sk_buff *skb, struct Qdisc *sch)
break;
case RED_PROB_MARK:
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (!gred_use_ecn(t) || !INET_ECN_set_ce(skb)) {
q->stats.prob_drop++;
goto congestion_drop;
@@ -219,7 +219,7 @@ static int gred_enqueue(struct sk_buff *skb, struct Qdisc *sch)
break;
case RED_HARD_MARK:
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (gred_use_harddrop(t) || !gred_use_ecn(t) ||
!INET_ECN_set_ce(skb)) {
q->stats.forced_drop++;
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 209b966..ad27825 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1591,7 +1591,7 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
cl = hfsc_classify(skb, sch, &err);
if (cl == NULL) {
if (err & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return err;
}
@@ -1600,7 +1600,7 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
if (unlikely(err != NET_XMIT_SUCCESS)) {
if (net_xmit_drop_count(err)) {
cl->qstats.drops++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
}
return err;
}
@@ -1643,7 +1643,7 @@ hfsc_dequeue(struct Qdisc *sch)
*/
cl = vttree_get_minvt(&q->root, cur_time);
if (cl == NULL) {
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
hfsc_schedule_watchdog(sch);
return NULL;
}
@@ -1698,7 +1698,7 @@ hfsc_drop(struct Qdisc *sch)
list_move_tail(&cl->dlist, &q->droplist);
}
cl->qstats.drops++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
sch->q.qlen--;
return len;
}
diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c
index d85b681..934e9cc 100644
--- a/net/sched/sch_hhf.c
+++ b/net/sched/sch_hhf.c
@@ -376,8 +376,8 @@ static unsigned int hhf_drop(struct Qdisc *sch)
struct sk_buff *skb = dequeue_head(bucket);
sch->q.qlen--;
- sch->qstats.drops++;
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_drop(sch);
+ qdisc_qstats_backlog(sch, skb);
kfree_skb(skb);
}
@@ -395,7 +395,7 @@ static int hhf_enqueue(struct sk_buff *skb, struct Qdisc *sch)
bucket = &q->buckets[idx];
bucket_add(bucket, skb);
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
if (list_empty(&bucket->bucketchain)) {
unsigned int weight;
@@ -457,7 +457,7 @@ begin:
if (bucket->head) {
skb = dequeue_head(bucket);
sch->q.qlen--;
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
}
if (!skb) {
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 0256dee..c40ab7a 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -586,13 +586,13 @@ static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch)
#ifdef CONFIG_NET_CLS_ACT
} else if (!cl) {
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
#endif
} else if ((ret = qdisc_enqueue(skb, cl->un.leaf.q)) != NET_XMIT_SUCCESS) {
if (net_xmit_drop_count(ret)) {
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
cl->qstats.drops++;
}
return ret;
@@ -925,7 +925,7 @@ ok:
goto ok;
}
}
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (likely(next_event > q->now)) {
if (!test_bit(__QDISC_STATE_DEACTIVATED,
&qdisc_root_sleeping(q->watchdog.qdisc)->state)) {
diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c
index b351125..eb5b844 100644
--- a/net/sched/sch_ingress.c
+++ b/net/sched/sch_ingress.c
@@ -69,7 +69,7 @@ static int ingress_enqueue(struct sk_buff *skb, struct Qdisc *sch)
switch (result) {
case TC_ACT_SHOT:
result = TC_ACT_SHOT;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
break;
case TC_ACT_STOLEN:
case TC_ACT_QUEUED:
diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c
index 4adbf7f..7f4e1d8 100644
--- a/net/sched/sch_multiq.c
+++ b/net/sched/sch_multiq.c
@@ -75,7 +75,7 @@ multiq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
if (qdisc == NULL) {
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
@@ -87,7 +87,7 @@ multiq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
return NET_XMIT_SUCCESS;
}
if (net_xmit_drop_count(ret))
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return ret;
}
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 111d70f..ace7ca1 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -429,12 +429,12 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
/* Drop packet? */
if (loss_event(q)) {
if (q->ecn && INET_ECN_set_ce(skb))
- sch->qstats.drops++; /* mark packet */
+ qdisc_qstats_drop(sch); /* mark packet */
else
--count;
}
if (count == 0) {
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return NET_XMIT_SUCCESS | __NET_XMIT_BYPASS;
}
@@ -478,7 +478,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
if (unlikely(skb_queue_len(&sch->q) >= sch->limit))
return qdisc_reshape_fail(skb, sch);
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
cb = netem_skb_cb(skb);
if (q->gap == 0 || /* not doing reordering */
@@ -549,15 +549,14 @@ static unsigned int netem_drop(struct Qdisc *sch)
sch->q.qlen--;
skb->next = NULL;
skb->prev = NULL;
- len = qdisc_pkt_len(skb);
- sch->qstats.backlog -= len;
+ qdisc_qstats_backlog(sch, skb);
kfree_skb(skb);
}
}
if (!len && q->qdisc && q->qdisc->ops->drop)
len = q->qdisc->ops->drop(q->qdisc);
if (len)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return len;
}
@@ -575,7 +574,7 @@ tfifo_dequeue:
skb = __skb_dequeue(&sch->q);
if (skb) {
deliver:
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
qdisc_unthrottled(sch);
qdisc_bstats_update(sch, skb);
return skb;
@@ -610,7 +609,7 @@ deliver:
if (unlikely(err != NET_XMIT_SUCCESS)) {
if (net_xmit_drop_count(err)) {
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
qdisc_tree_decrease_qlen(sch, 1);
}
}
diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c
index fefeeb7..32636c9 100644
--- a/net/sched/sch_pie.c
+++ b/net/sched/sch_pie.c
@@ -232,7 +232,7 @@ static int pie_change(struct Qdisc *sch, struct nlattr *opt)
while (sch->q.qlen > sch->limit) {
struct sk_buff *skb = __skb_dequeue(&sch->q);
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
qdisc_drop(skb, sch);
}
qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen);
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 68a8f25..b411e78 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -77,7 +77,7 @@ prio_enqueue(struct sk_buff *skb, struct Qdisc *sch)
if (qdisc == NULL) {
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
@@ -89,7 +89,7 @@ prio_enqueue(struct sk_buff *skb, struct Qdisc *sch)
return NET_XMIT_SUCCESS;
}
if (net_xmit_drop_count(ret))
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return ret;
}
diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index d59f857..3fb2655 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -1229,7 +1229,7 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
cl = qfq_classify(skb, sch, &err);
if (cl == NULL) {
if (err & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return err;
}
@@ -1249,7 +1249,7 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
pr_debug("qfq_enqueue: enqueue failed %d\n", err);
if (net_xmit_drop_count(err)) {
cl->qstats.drops++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
}
return err;
}
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index 633e32d..6c0534c 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -74,7 +74,7 @@ static int red_enqueue(struct sk_buff *skb, struct Qdisc *sch)
break;
case RED_PROB_MARK:
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (!red_use_ecn(q) || !INET_ECN_set_ce(skb)) {
q->stats.prob_drop++;
goto congestion_drop;
@@ -84,7 +84,7 @@ static int red_enqueue(struct sk_buff *skb, struct Qdisc *sch)
break;
case RED_HARD_MARK:
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (red_use_harddrop(q) || !red_use_ecn(q) ||
!INET_ECN_set_ce(skb)) {
q->stats.forced_drop++;
@@ -100,7 +100,7 @@ static int red_enqueue(struct sk_buff *skb, struct Qdisc *sch)
sch->q.qlen++;
} else if (net_xmit_drop_count(ret)) {
q->stats.pdrop++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
}
return ret;
@@ -142,7 +142,7 @@ static unsigned int red_drop(struct Qdisc *sch)
if (child->ops->drop && (len = child->ops->drop(child)) > 0) {
q->stats.other++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
sch->q.qlen--;
return len;
}
diff --git a/net/sched/sch_sfb.c b/net/sched/sch_sfb.c
index 1562fb2..5819dd8 100644
--- a/net/sched/sch_sfb.c
+++ b/net/sched/sch_sfb.c
@@ -290,7 +290,7 @@ static int sfb_enqueue(struct sk_buff *skb, struct Qdisc *sch)
struct flow_keys keys;
if (unlikely(sch->q.qlen >= q->limit)) {
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
q->stats.queuedrop++;
goto drop;
}
@@ -348,7 +348,7 @@ static int sfb_enqueue(struct sk_buff *skb, struct Qdisc *sch)
sfb_skb_cb(skb)->hashes[slot] = 0;
if (unlikely(minqlen >= q->max)) {
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
q->stats.bucketdrop++;
goto drop;
}
@@ -376,7 +376,7 @@ static int sfb_enqueue(struct sk_buff *skb, struct Qdisc *sch)
}
}
if (sfb_rate_limit(skb, q)) {
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
q->stats.penaltydrop++;
goto drop;
}
@@ -411,7 +411,7 @@ enqueue:
increment_qlen(skb, q);
} else if (net_xmit_drop_count(ret)) {
q->stats.childdrop++;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
}
return ret;
@@ -420,7 +420,7 @@ drop:
return NET_XMIT_CN;
other_drop:
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 80c36bd..16fb3e7 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -331,8 +331,8 @@ drop:
sfq_dec(q, x);
kfree_skb(skb);
sch->q.qlen--;
- sch->qstats.drops++;
- sch->qstats.backlog -= len;
+ qdisc_qstats_drop(sch);
+ qdisc_qstats_backlog(sch, skb);
return len;
}
@@ -379,7 +379,7 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
hash = sfq_classify(skb, sch, &ret);
if (hash == 0) {
if (ret & __NET_XMIT_BYPASS)
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
kfree_skb(skb);
return ret;
}
@@ -409,7 +409,7 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
break;
case RED_PROB_MARK:
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (sfq_prob_mark(q)) {
/* We know we have at least one packet in queue */
if (sfq_headdrop(q) &&
@@ -426,7 +426,7 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
goto congestion_drop;
case RED_HARD_MARK:
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
if (sfq_hard_mark(q)) {
/* We know we have at least one packet in queue */
if (sfq_headdrop(q) &&
@@ -461,7 +461,7 @@ congestion_drop:
}
enqueue:
- sch->qstats.backlog += qdisc_pkt_len(skb);
+ qdisc_qstats_backlog_inc(sch, skb);
slot->backlog += qdisc_pkt_len(skb);
slot_queue_add(slot, skb);
sfq_inc(q, x);
@@ -520,7 +520,7 @@ next_slot:
sfq_dec(q, a);
qdisc_bstats_update(sch, skb);
sch->q.qlen--;
- sch->qstats.backlog -= qdisc_pkt_len(skb);
+ qdisc_qstats_backlog(sch, skb);
slot->backlog -= qdisc_pkt_len(skb);
/* Is the slot empty? */
if (slot->qlen == 0) {
@@ -586,7 +586,8 @@ static void sfq_rehash(struct Qdisc *sch)
if (x == SFQ_EMPTY_SLOT) {
x = q->dep[0].next; /* get a free slot */
if (x >= SFQ_MAX_FLOWS) {
-drop: sch->qstats.backlog -= qdisc_pkt_len(skb);
+drop:
+ qdisc_qstats_backlog(sch, skb);
kfree_skb(skb);
dropped++;
continue;
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 0c39b75..77edffe 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -175,7 +175,7 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch)
ret = qdisc_enqueue(segs, q->qdisc);
if (ret != NET_XMIT_SUCCESS) {
if (net_xmit_drop_count(ret))
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
} else {
nb++;
}
@@ -201,7 +201,7 @@ static int tbf_enqueue(struct sk_buff *skb, struct Qdisc *sch)
ret = qdisc_enqueue(skb, q->qdisc);
if (ret != NET_XMIT_SUCCESS) {
if (net_xmit_drop_count(ret))
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
return ret;
}
@@ -216,7 +216,7 @@ static unsigned int tbf_drop(struct Qdisc *sch)
if (q->qdisc->ops->drop && (len = q->qdisc->ops->drop(q->qdisc)) != 0) {
sch->q.qlen--;
- sch->qstats.drops++;
+ qdisc_qstats_drop(sch);
}
return len;
}
@@ -281,7 +281,7 @@ static struct sk_buff *tbf_dequeue(struct Qdisc *sch)
(cf. CSZ, HPFQ, HFSC)
*/
- sch->qstats.overlimits++;
+ qdisc_qstats_overlimit(sch);
}
return NULL;
}
^ permalink raw reply related
* [net-next PATCH v2 3/4] net: sched: restrict use of qstats qlen
From: John Fastabend @ 2014-09-26 18:37 UTC (permalink / raw)
To: xiyou.wangcong, jhs, eric.dumazet, davem; +Cc: netdev
In-Reply-To: <20140926183621.2621.72538.stgit@nitbit.x32>
This removes the use of qstats->qlen variable from the classifiers
and makes it an explicit argument to gnet_stats_copy_queue().
The qlen represents the qdisc queue length and is packed into
the qstats at the last moment before passnig to user space. By
handling it explicitely we avoid, in the percpu stats case, having
to figure out which per_cpu variable to put it in.
It would probably be best to remove it from qstats completely
but qstats is a user space ABI and can't be broken. A future
patch could make an internal only qstats structure that would
avoid having to allocate an additional u32 variable on the
Qdisc struct. This would make the qstats struct 128bits instead
of 128+32.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
---
include/net/gen_stats.h | 3 ++-
net/core/gen_stats.c | 6 +++++-
net/sched/act_api.c | 4 +++-
net/sched/sch_api.c | 5 +++--
net/sched/sch_atm.c | 4 +---
net/sched/sch_cbq.c | 3 +--
net/sched/sch_drr.c | 7 +++----
net/sched/sch_fq_codel.c | 2 +-
net/sched/sch_hfsc.c | 3 +--
net/sched/sch_htb.c | 5 +++--
net/sched/sch_mq.c | 4 +---
net/sched/sch_mqprio.c | 9 ++++-----
net/sched/sch_multiq.c | 3 +--
net/sched/sch_prio.c | 3 +--
net/sched/sch_qfq.c | 3 +--
net/sched/sch_sfq.c | 2 +-
16 files changed, 32 insertions(+), 34 deletions(-)
diff --git a/include/net/gen_stats.h b/include/net/gen_stats.h
index ce3c128..de9b3dd 100644
--- a/include/net/gen_stats.h
+++ b/include/net/gen_stats.h
@@ -40,7 +40,8 @@ void __gnet_stats_copy_basic(struct gnet_stats_basic_packed *bstats,
int gnet_stats_copy_rate_est(struct gnet_dump *d,
const struct gnet_stats_basic_packed *b,
struct gnet_stats_rate_est64 *r);
-int gnet_stats_copy_queue(struct gnet_dump *d, struct gnet_stats_queue *q);
+int gnet_stats_copy_queue(struct gnet_dump *d,
+ struct gnet_stats_queue *q, __u32 len);
int gnet_stats_copy_app(struct gnet_dump *d, void *st, int len);
int gnet_stats_finish_copy(struct gnet_dump *d);
diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c
index 5ff8e80..ad3ecb6 100644
--- a/net/core/gen_stats.c
+++ b/net/core/gen_stats.c
@@ -219,6 +219,7 @@ EXPORT_SYMBOL(gnet_stats_copy_rate_est);
* gnet_stats_copy_queue - copy queue statistics into statistics TLV
* @d: dumping handle
* @q: queue statistics
+ * @qlen: queue length statistics
*
* Appends the queue statistics to the top level TLV created by
* gnet_stats_start_copy().
@@ -227,8 +228,11 @@ EXPORT_SYMBOL(gnet_stats_copy_rate_est);
* if the room in the socket buffer was not sufficient.
*/
int
-gnet_stats_copy_queue(struct gnet_dump *d, struct gnet_stats_queue *q)
+gnet_stats_copy_queue(struct gnet_dump *d,
+ struct gnet_stats_queue *q, __u32 qlen)
{
+ q->qlen = qlen;
+
if (d->compat_tc_stats) {
d->tc_stats.drops = q->drops;
d->tc_stats.qlen = q->qlen;
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index eca4cf9..2e13409 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -623,7 +623,9 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a,
if (gnet_stats_copy_basic(&d, NULL, &p->tcfc_bstats) < 0 ||
gnet_stats_copy_rate_est(&d, &p->tcfc_bstats,
&p->tcfc_rate_est) < 0 ||
- gnet_stats_copy_queue(&d, &p->tcfc_qstats) < 0)
+ gnet_stats_copy_queue(&d,
+ &p->tcfc_qstats,
+ p->tcfc_qstats.qlen) < 0)
goto errout;
if (gnet_stats_finish_copy(&d) < 0)
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 2862bc6..ca00ea8 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -1318,6 +1318,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
unsigned char *b = skb_tail_pointer(skb);
struct gnet_dump d;
struct qdisc_size_table *stab;
+ __u32 qlen;
cond_resched();
nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags);
@@ -1335,7 +1336,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
goto nla_put_failure;
if (q->ops->dump && q->ops->dump(q, skb) < 0)
goto nla_put_failure;
- q->qstats.qlen = q->q.qlen;
+ qlen = q->q.qlen;
stab = rtnl_dereference(q->stab);
if (stab && qdisc_dump_stab(skb, stab) < 0)
@@ -1353,7 +1354,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
if (gnet_stats_copy_basic(&d, cpu_bstats, &q->bstats) < 0 ||
gnet_stats_copy_rate_est(&d, &q->bstats, &q->rate_est) < 0 ||
- gnet_stats_copy_queue(&d, &q->qstats) < 0)
+ gnet_stats_copy_queue(&d, &q->qstats, qlen) < 0)
goto nla_put_failure;
if (gnet_stats_finish_copy(&d) < 0)
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index 040212c..c145eb6 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -637,10 +637,8 @@ atm_tc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
{
struct atm_flow_data *flow = (struct atm_flow_data *)arg;
- flow->qstats.qlen = flow->q->q.qlen;
-
if (gnet_stats_copy_basic(d, NULL, &flow->bstats) < 0 ||
- gnet_stats_copy_queue(d, &flow->qstats) < 0)
+ gnet_stats_copy_queue(d, &flow->qstats, flow->q->q.qlen) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 60432c3..c610081 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1594,7 +1594,6 @@ cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
struct cbq_sched_data *q = qdisc_priv(sch);
struct cbq_class *cl = (struct cbq_class *)arg;
- cl->qstats.qlen = cl->q->q.qlen;
cl->xstats.avgidle = cl->avgidle;
cl->xstats.undertime = 0;
@@ -1603,7 +1602,7 @@ cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl->qstats, cl->q->q.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats));
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index 907b12f..5835a93 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -275,17 +275,16 @@ static int drr_dump_class_stats(struct Qdisc *sch, unsigned long arg,
struct gnet_dump *d)
{
struct drr_class *cl = (struct drr_class *)arg;
+ __u32 qlen = cl->qdisc->q.qlen;
struct tc_drr_stats xstats;
memset(&xstats, 0, sizeof(xstats));
- if (cl->qdisc->q.qlen) {
+ if (qlen)
xstats.deficit = cl->deficit;
- cl->qdisc->qstats.qlen = cl->qdisc->q.qlen;
- }
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qdisc->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl->qdisc->qstats, qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c
index 3f288ae..6a42112 100644
--- a/net/sched/sch_fq_codel.c
+++ b/net/sched/sch_fq_codel.c
@@ -550,7 +550,7 @@ static int fq_codel_dump_class_stats(struct Qdisc *sch, unsigned long cl,
qs.backlog = q->backlogs[idx];
qs.drops = flow->dropped;
}
- if (gnet_stats_copy_queue(d, &qs) < 0)
+ if (gnet_stats_copy_queue(d, &qs, 0) < 0)
return -1;
if (idx < q->flows_cnt)
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index ad27825..d364acb 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1370,7 +1370,6 @@ hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
struct hfsc_class *cl = (struct hfsc_class *)arg;
struct tc_hfsc_stats xstats;
- cl->qstats.qlen = cl->qdisc->q.qlen;
cl->qstats.backlog = cl->qdisc->qstats.backlog;
xstats.level = cl->level;
xstats.period = cl->cl_vtperiod;
@@ -1379,7 +1378,7 @@ hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl->qstats, cl->qdisc->q.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index c40ab7a..3a691fd 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1138,15 +1138,16 @@ static int
htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d)
{
struct htb_class *cl = (struct htb_class *)arg;
+ __u32 qlen = 0;
if (!cl->level && cl->un.leaf.q)
- cl->qstats.qlen = cl->un.leaf.q->q.qlen;
+ qlen = cl->un.leaf.q->q.qlen;
cl->xstats.tokens = PSCHED_NS2TICKS(cl->tokens);
cl->xstats.ctokens = PSCHED_NS2TICKS(cl->ctokens);
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, NULL, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl->qstats, qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats));
diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c
index d3a27fb..6416a69 100644
--- a/net/sched/sch_mq.c
+++ b/net/sched/sch_mq.c
@@ -112,7 +112,6 @@ static int mq_dump(struct Qdisc *sch, struct sk_buff *skb)
sch->q.qlen += qdisc->q.qlen;
sch->bstats.bytes += qdisc->bstats.bytes;
sch->bstats.packets += qdisc->bstats.packets;
- sch->qstats.qlen += qdisc->qstats.qlen;
sch->qstats.backlog += qdisc->qstats.backlog;
sch->qstats.drops += qdisc->qstats.drops;
sch->qstats.requeues += qdisc->qstats.requeues;
@@ -200,9 +199,8 @@ static int mq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
struct netdev_queue *dev_queue = mq_queue_get(sch, cl);
sch = dev_queue->qdisc_sleeping;
- sch->qstats.qlen = sch->q.qlen;
if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 ||
- gnet_stats_copy_queue(d, &sch->qstats) < 0)
+ gnet_stats_copy_queue(d, &sch->qstats, sch->q.qlen) < 0)
return -1;
return 0;
}
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index 8917372..03dbeb5 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -236,7 +236,6 @@ static int mqprio_dump(struct Qdisc *sch, struct sk_buff *skb)
sch->q.qlen += qdisc->q.qlen;
sch->bstats.bytes += qdisc->bstats.bytes;
sch->bstats.packets += qdisc->bstats.packets;
- sch->qstats.qlen += qdisc->qstats.qlen;
sch->qstats.backlog += qdisc->qstats.backlog;
sch->qstats.drops += qdisc->qstats.drops;
sch->qstats.requeues += qdisc->qstats.requeues;
@@ -327,6 +326,7 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
if (cl <= netdev_get_num_tc(dev)) {
int i;
+ __u32 qlen = 0;
struct Qdisc *qdisc;
struct gnet_stats_queue qstats = {0};
struct gnet_stats_basic_packed bstats = {0};
@@ -344,9 +344,9 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
qdisc = rtnl_dereference(q->qdisc);
spin_lock_bh(qdisc_lock(qdisc));
+ qlen += qdisc->q.qlen;
bstats.bytes += qdisc->bstats.bytes;
bstats.packets += qdisc->bstats.packets;
- qstats.qlen += qdisc->qstats.qlen;
qstats.backlog += qdisc->qstats.backlog;
qstats.drops += qdisc->qstats.drops;
qstats.requeues += qdisc->qstats.requeues;
@@ -356,15 +356,14 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
/* Reclaim root sleeping lock before completing stats */
spin_lock_bh(d->lock);
if (gnet_stats_copy_basic(d, NULL, &bstats) < 0 ||
- gnet_stats_copy_queue(d, &qstats) < 0)
+ gnet_stats_copy_queue(d, &qstats, qlen) < 0)
return -1;
} else {
struct netdev_queue *dev_queue = mqprio_queue_get(sch, cl);
sch = dev_queue->qdisc_sleeping;
- sch->qstats.qlen = sch->q.qlen;
if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 ||
- gnet_stats_copy_queue(d, &sch->qstats) < 0)
+ gnet_stats_copy_queue(d, &sch->qstats, sch->q.qlen) < 0)
return -1;
}
return 0;
diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c
index 7f4e1d8..53357b3 100644
--- a/net/sched/sch_multiq.c
+++ b/net/sched/sch_multiq.c
@@ -360,9 +360,8 @@ static int multiq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
struct Qdisc *cl_q;
cl_q = q->queues[cl - 1];
- cl_q->qstats.qlen = cl_q->q.qlen;
if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 ||
- gnet_stats_copy_queue(d, &cl_q->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl_q->qstats, cl_q->q.qlen) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index b411e78..4644f55 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -324,9 +324,8 @@ static int prio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
struct Qdisc *cl_q;
cl_q = q->queues[cl - 1];
- cl_q->qstats.qlen = cl_q->q.qlen;
if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 ||
- gnet_stats_copy_queue(d, &cl_q->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl_q->qstats, cl_q->q.qlen) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index 3fb2655..66df9d9 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -664,14 +664,13 @@ static int qfq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
struct tc_qfq_stats xstats;
memset(&xstats, 0, sizeof(xstats));
- cl->qdisc->qstats.qlen = cl->qdisc->q.qlen;
xstats.weight = cl->agg->class_weight;
xstats.lmax = cl->agg->lmax;
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qdisc->qstats) < 0)
+ gnet_stats_copy_queue(d, &cl->qdisc->qstats, cl->qdisc->q.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 16fb3e7..4805651 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -871,7 +871,7 @@ static int sfq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
qs.qlen = slot->qlen;
qs.backlog = slot->backlog;
}
- if (gnet_stats_copy_queue(d, &qs) < 0)
+ if (gnet_stats_copy_queue(d, &qs, qs.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
}
^ permalink raw reply related
* [net-next PATCH v2 4/4] net: sched: enable per cpu qstats
From: John Fastabend @ 2014-09-26 18:37 UTC (permalink / raw)
To: xiyou.wangcong, jhs, eric.dumazet, davem; +Cc: netdev
In-Reply-To: <20140926183621.2621.72538.stgit@nitbit.x32>
After previous patches to simplify qstats the qstats can be
made per cpu with a packed union in Qdisc struct.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
---
include/net/gen_stats.h | 3 ++
include/net/sch_generic.h | 12 +++++++++-
net/core/gen_stats.c | 55 +++++++++++++++++++++++++++++++++++++++------
net/sched/act_api.c | 2 +-
net/sched/sch_api.c | 12 ++++++++--
net/sched/sch_atm.c | 2 +-
net/sched/sch_cbq.c | 2 +-
net/sched/sch_drr.c | 2 +-
net/sched/sch_fq_codel.c | 2 +-
net/sched/sch_hfsc.c | 2 +-
net/sched/sch_htb.c | 2 +-
net/sched/sch_mq.c | 2 +-
net/sched/sch_mqprio.c | 5 ++--
net/sched/sch_multiq.c | 2 +-
net/sched/sch_prio.c | 2 +-
net/sched/sch_qfq.c | 3 ++
net/sched/sch_sfq.c | 2 +-
17 files changed, 87 insertions(+), 25 deletions(-)
diff --git a/include/net/gen_stats.h b/include/net/gen_stats.h
index de9b3dd..cbafa37 100644
--- a/include/net/gen_stats.h
+++ b/include/net/gen_stats.h
@@ -41,7 +41,8 @@ int gnet_stats_copy_rate_est(struct gnet_dump *d,
const struct gnet_stats_basic_packed *b,
struct gnet_stats_rate_est64 *r);
int gnet_stats_copy_queue(struct gnet_dump *d,
- struct gnet_stats_queue *q, __u32 len);
+ struct gnet_stats_queue __percpu *cpu_q,
+ struct gnet_stats_queue *q, __u32 qlen);
int gnet_stats_copy_app(struct gnet_dump *d, void *st, int len);
int gnet_stats_finish_copy(struct gnet_dump *d);
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 5609844..4d14c70 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -90,7 +90,10 @@ struct Qdisc {
struct gnet_stats_basic_cpu __percpu *cpu_bstats;
} __packed;
unsigned int __state;
- struct gnet_stats_queue qstats;
+ union {
+ struct gnet_stats_queue qstats;
+ struct gnet_stats_queue __percpu *cpu_qstats;
+ } __packed;
struct rcu_head rcu_head;
int padded;
atomic_t refcnt;
@@ -544,6 +547,13 @@ static inline void qdisc_qstats_drop(struct Qdisc *sch)
sch->qstats.drops++;
}
+static inline void qdisc_qstats_drop_cpu(struct Qdisc *sch)
+{
+ struct gnet_stats_queue *qstats = this_cpu_ptr(sch->cpu_qstats);
+
+ qstats->drops++;
+}
+
static inline void qdisc_qstats_overlimit(struct Qdisc *sch)
{
sch->qstats.overlimits++;
diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c
index ad3ecb6..14681b9 100644
--- a/net/core/gen_stats.c
+++ b/net/core/gen_stats.c
@@ -215,33 +215,74 @@ gnet_stats_copy_rate_est(struct gnet_dump *d,
}
EXPORT_SYMBOL(gnet_stats_copy_rate_est);
+static void
+__gnet_stats_copy_queue_cpu(struct gnet_stats_queue *qstats,
+ const struct gnet_stats_queue __percpu *q)
+{
+ int i;
+
+ for_each_possible_cpu(i) {
+ const struct gnet_stats_queue *qcpu = per_cpu_ptr(q, i);
+
+ qstats->qlen = 0;
+ qstats->backlog += qcpu->backlog;
+ qstats->drops += qcpu->drops;
+ qstats->requeues += qcpu->requeues;
+ qstats->overlimits += qcpu->overlimits;
+ }
+}
+
+static void __gnet_stats_copy_queue(struct gnet_stats_queue *qstats,
+ const struct gnet_stats_queue __percpu *cpu,
+ const struct gnet_stats_queue *q,
+ __u32 qlen)
+{
+ if (cpu) {
+ __gnet_stats_copy_queue_cpu(qstats, cpu);
+ } else {
+ qstats->qlen = q->qlen;
+ qstats->backlog = q->backlog;
+ qstats->drops = q->drops;
+ qstats->requeues = q->requeues;
+ qstats->overlimits = q->overlimits;
+ }
+
+ qstats->qlen = qlen;
+}
+
/**
* gnet_stats_copy_queue - copy queue statistics into statistics TLV
* @d: dumping handle
+ * @cpu_q: per cpu queue statistics
* @q: queue statistics
* @qlen: queue length statistics
*
* Appends the queue statistics to the top level TLV created by
- * gnet_stats_start_copy().
+ * gnet_stats_start_copy(). Using per cpu queue statistics if
+ * they are available.
*
* Returns 0 on success or -1 with the statistic lock released
* if the room in the socket buffer was not sufficient.
*/
int
gnet_stats_copy_queue(struct gnet_dump *d,
+ struct gnet_stats_queue __percpu *cpu_q,
struct gnet_stats_queue *q, __u32 qlen)
{
- q->qlen = qlen;
+ struct gnet_stats_queue qstats = {0};
+
+ __gnet_stats_copy_queue(&qstats, cpu_q, q, qlen);
if (d->compat_tc_stats) {
- d->tc_stats.drops = q->drops;
- d->tc_stats.qlen = q->qlen;
- d->tc_stats.backlog = q->backlog;
- d->tc_stats.overlimits = q->overlimits;
+ d->tc_stats.drops = qstats.drops;
+ d->tc_stats.qlen = qstats.qlen;
+ d->tc_stats.backlog = qstats.backlog;
+ d->tc_stats.overlimits = qstats.overlimits;
}
if (d->tail)
- return gnet_stats_copy(d, TCA_STATS_QUEUE, q, sizeof(*q));
+ return gnet_stats_copy(d, TCA_STATS_QUEUE,
+ &qstats, sizeof(qstats));
return 0;
}
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 2e13409..3d43e49 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -623,7 +623,7 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a,
if (gnet_stats_copy_basic(&d, NULL, &p->tcfc_bstats) < 0 ||
gnet_stats_copy_rate_est(&d, &p->tcfc_bstats,
&p->tcfc_rate_est) < 0 ||
- gnet_stats_copy_queue(&d,
+ gnet_stats_copy_queue(&d, NULL,
&p->tcfc_qstats,
p->tcfc_qstats.qlen) < 0)
goto errout;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index ca00ea8..aa83295 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -947,6 +947,10 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
alloc_percpu(struct gnet_stats_basic_cpu);
if (!sch->cpu_bstats)
goto err_out4;
+
+ sch->cpu_qstats = alloc_percpu(struct gnet_stats_queue);
+ if (!sch->cpu_qstats)
+ goto err_out4;
}
if (tca[TCA_STAB]) {
@@ -995,6 +999,7 @@ err_out:
err_out4:
free_percpu(sch->cpu_bstats);
+ free_percpu(sch->cpu_qstats);
/*
* Any broken qdiscs that would require a ops->reset() here?
* The qdisc was never in action so it shouldn't be necessary.
@@ -1313,6 +1318,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
u32 portid, u32 seq, u16 flags, int event)
{
struct gnet_stats_basic_cpu __percpu *cpu_bstats = NULL;
+ struct gnet_stats_queue __percpu *cpu_qstats = NULL;
struct tcmsg *tcm;
struct nlmsghdr *nlh;
unsigned char *b = skb_tail_pointer(skb);
@@ -1349,12 +1355,14 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0)
goto nla_put_failure;
- if (qdisc_is_percpu_stats(q))
+ if (qdisc_is_percpu_stats(q)) {
cpu_bstats = q->cpu_bstats;
+ cpu_qstats = q->cpu_qstats;
+ }
if (gnet_stats_copy_basic(&d, cpu_bstats, &q->bstats) < 0 ||
gnet_stats_copy_rate_est(&d, &q->bstats, &q->rate_est) < 0 ||
- gnet_stats_copy_queue(&d, &q->qstats, qlen) < 0)
+ gnet_stats_copy_queue(&d, cpu_qstats, &q->qstats, qlen) < 0)
goto nla_put_failure;
if (gnet_stats_finish_copy(&d) < 0)
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index c145eb6..e3e2cc5 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -638,7 +638,7 @@ atm_tc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
struct atm_flow_data *flow = (struct atm_flow_data *)arg;
if (gnet_stats_copy_basic(d, NULL, &flow->bstats) < 0 ||
- gnet_stats_copy_queue(d, &flow->qstats, flow->q->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &flow->qstats, flow->q->q.qlen) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index c610081..beeb75f 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1602,7 +1602,7 @@ cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qstats, cl->q->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &cl->qstats, cl->q->q.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats));
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index 5835a93..3387060 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -284,7 +284,7 @@ static int drr_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qdisc->qstats, qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &cl->qdisc->qstats, qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c
index 6a42112..301c352 100644
--- a/net/sched/sch_fq_codel.c
+++ b/net/sched/sch_fq_codel.c
@@ -550,7 +550,7 @@ static int fq_codel_dump_class_stats(struct Qdisc *sch, unsigned long cl,
qs.backlog = q->backlogs[idx];
qs.drops = flow->dropped;
}
- if (gnet_stats_copy_queue(d, &qs, 0) < 0)
+ if (gnet_stats_copy_queue(d, NULL, &qs, 0) < 0)
return -1;
if (idx < q->flows_cnt)
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index d364acb..e6c7416 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1378,7 +1378,7 @@ hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qstats, cl->qdisc->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &cl->qstats, cl->qdisc->q.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 3a691fd..f1acb0f 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1147,7 +1147,7 @@ htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d)
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, NULL, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qstats, qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &cl->qstats, qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats));
diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c
index 6416a69..f3cbaec 100644
--- a/net/sched/sch_mq.c
+++ b/net/sched/sch_mq.c
@@ -200,7 +200,7 @@ static int mq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
sch = dev_queue->qdisc_sleeping;
if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 ||
- gnet_stats_copy_queue(d, &sch->qstats, sch->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &sch->qstats, sch->q.qlen) < 0)
return -1;
return 0;
}
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index 03dbeb5..3811a74 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -356,14 +356,15 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
/* Reclaim root sleeping lock before completing stats */
spin_lock_bh(d->lock);
if (gnet_stats_copy_basic(d, NULL, &bstats) < 0 ||
- gnet_stats_copy_queue(d, &qstats, qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &qstats, qlen) < 0)
return -1;
} else {
struct netdev_queue *dev_queue = mqprio_queue_get(sch, cl);
sch = dev_queue->qdisc_sleeping;
if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 ||
- gnet_stats_copy_queue(d, &sch->qstats, sch->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL,
+ &sch->qstats, sch->q.qlen) < 0)
return -1;
}
return 0;
diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c
index 53357b3..42dd218 100644
--- a/net/sched/sch_multiq.c
+++ b/net/sched/sch_multiq.c
@@ -361,7 +361,7 @@ static int multiq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
cl_q = q->queues[cl - 1];
if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 ||
- gnet_stats_copy_queue(d, &cl_q->qstats, cl_q->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &cl_q->qstats, cl_q->q.qlen) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 4644f55..8e5cd34 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -325,7 +325,7 @@ static int prio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
cl_q = q->queues[cl - 1];
if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 ||
- gnet_stats_copy_queue(d, &cl_q->qstats, cl_q->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL, &cl_q->qstats, cl_q->q.qlen) < 0)
return -1;
return 0;
diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index 66df9d9..3ec7e88 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -670,7 +670,8 @@ static int qfq_dump_class_stats(struct Qdisc *sch, unsigned long arg,
if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 ||
gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 ||
- gnet_stats_copy_queue(d, &cl->qdisc->qstats, cl->qdisc->q.qlen) < 0)
+ gnet_stats_copy_queue(d, NULL,
+ &cl->qdisc->qstats, cl->qdisc->q.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 4805651..5814c46 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -871,7 +871,7 @@ static int sfq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
qs.qlen = slot->qlen;
qs.backlog = slot->backlog;
}
- if (gnet_stats_copy_queue(d, &qs, qs.qlen) < 0)
+ if (gnet_stats_copy_queue(d, NULL, &qs, qs.qlen) < 0)
return -1;
return gnet_stats_copy_app(d, &xstats, sizeof(xstats));
}
^ permalink raw reply related
* Re: [net-next PATCH 2/4] net: sched: implement qstat helper routines
From: John Fastabend @ 2014-09-26 18:41 UTC (permalink / raw)
To: Eric Dumazet, John Fastabend; +Cc: xiyou.wangcong, jhs, davem, netdev
In-Reply-To: <1411756601.16953.136.camel@edumazet-glaptop2.roam.corp.google.com>
On 09/26/2014 11:36 AM, Eric Dumazet wrote:
> On Fri, 2014-09-26 at 11:14 -0700, John Fastabend wrote:
>> This adds helpers to manipulate qstats logic and replaces locations
>> that touch the counters directly. This simplifies future patches
>> to push qstats onto per cpu counters.
>>
>> Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
>> ---
>> include/net/sch_generic.h | 43 +++++++++++++++++++++++++++++++++++--------
>> net/sched/sch_api.c | 2 +-
>> net/sched/sch_atm.c | 2 +-
>> net/sched/sch_cbq.c | 10 +++++-----
>> net/sched/sch_choke.c | 14 +++++++-------
>> net/sched/sch_codel.c | 2 +-
>> net/sched/sch_drr.c | 4 ++--
>> net/sched/sch_dsmark.c | 2 +-
>> net/sched/sch_fifo.c | 2 +-
>> net/sched/sch_fq.c | 4 ++--
>> net/sched/sch_fq_codel.c | 8 ++++----
>> net/sched/sch_gred.c | 4 ++--
>> net/sched/sch_hfsc.c | 8 ++++----
>> net/sched/sch_hhf.c | 8 ++++----
>> net/sched/sch_htb.c | 6 +++---
>> net/sched/sch_ingress.c | 2 +-
>> net/sched/sch_multiq.c | 4 ++--
>> net/sched/sch_netem.c | 15 +++++++--------
>> net/sched/sch_pie.c | 2 +-
>> net/sched/sch_prio.c | 4 ++--
>> net/sched/sch_qfq.c | 4 ++--
>> net/sched/sch_red.c | 8 ++++----
>> net/sched/sch_sfb.c | 10 +++++-----
>> net/sched/sch_sfq.c | 17 +++++++++--------
>> net/sched/sch_tbf.c | 8 ++++----
>> 25 files changed, 110 insertions(+), 83 deletions(-)
>>
>> diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
>> index 4b93511..5609844 100644
>> --- a/include/net/sch_generic.h
>> +++ b/include/net/sch_generic.h
>> @@ -521,11 +521,39 @@ static inline void qdisc_bstats_update(struct Qdisc *sch,
>> bstats_update(&sch->bstats, skb);
>> }
>>
>> +static inline void qdisc_qstats_backlog(struct Qdisc *sch,
>> + const struct sk_buff *skb)
>
> qdisc_qstats_backlog_dec() to be consistent with
> qdisc_qstats_backlog_dec() ?
>
>> +{
>> + sch->qstats.backlog -= qdisc_pkt_len(skb);
>> +}
>> +
>> +static inline void qdisc_qstats_backlog_inc(struct Qdisc *sch,
>> + const struct sk_buff *skb)
>> +{
>> +
>> + sch->qstats.backlog += qdisc_pkt_len(skb);
>> +}
>> +
>> +static inline void __qdisc_qstats_drop(struct Qdisc *sch, int count)
>> +{
>> + sch->qstats.drops += count;
>> +}
>> +
>> +static inline void qdisc_qstats_drop(struct Qdisc *sch)
>> +{
>> + sch->qstats.drops++;
>> +}
>> +
>> +static inline void qdisc_qstats_overlimit(struct Qdisc *sch)
>> +{
>> + sch->qstats.overlimits++;
>> +}
>> +
>> static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
>> struct sk_buff_head *list)
>> {
>> __skb_queue_tail(list, skb);
>> - sch->qstats.backlog += qdisc_pkt_len(skb);
>> + qdisc_qstats_backlog_inc(sch, skb);
>>
>> return NET_XMIT_SUCCESS;
>> }
>> @@ -541,7 +569,7 @@ static inline struct sk_buff *__qdisc_dequeue_head(struct Qdisc *sch,
>> struct sk_buff *skb = __skb_dequeue(list);
>>
>> if (likely(skb != NULL)) {
>> - sch->qstats.backlog -= qdisc_pkt_len(skb);
>> + qdisc_qstats_backlog(sch, skb);
>> qdisc_bstats_update(sch, skb);
>> }
>>
>> @@ -559,10 +587,9 @@ static inline unsigned int __qdisc_queue_drop_head(struct Qdisc *sch,
>> struct sk_buff *skb = __skb_dequeue(list);
>>
>> if (likely(skb != NULL)) {
>> - unsigned int len = qdisc_pkt_len(skb);
>> - sch->qstats.backlog -= len;
>> + qdisc_qstats_backlog(sch, skb);
>> kfree_skb(skb);
>> - return len;
>> + return qdisc_pkt_len(skb);
>
> Well, you just added a use after free.
hmm should have held off on the v2 I guess. I'll fix this
and add a test for the only drop_head consumer pfifo_head_drop.
Thanks!
>
>> }
>>
>> return 0;
>> @@ -579,7 +606,7 @@ static inline struct sk_buff *__qdisc_dequeue_tail(struct Qdisc *sch,
>> struct sk_buff *skb = __skb_dequeue_tail(list);
>>
>> if (likely(skb != NULL))
>
>
> Rest of the patch seems fine, thanks !
>
>
> --
> 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
* Re: [RFC PATCH net-next v2 0/5] netns: allow to identify peer netns
From: Eric W. Biederman @ 2014-09-26 18:57 UTC (permalink / raw)
To: Andy Lutomirski
Cc: Andrew Morton, Network Development, Linux Containers,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Stephen Hemminger, Cong Wang, Linux API, Nicolas Dichtel,
David S. Miller
In-Reply-To: <CALCETrX5e0cp4QFCv1eAqR1hjoROU9Rh=cRos9U35DaR-py3Eg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
Andy Lutomirski <luto-kltTT9wpgjJwATOyAt5JVQ@public.gmane.org> writes:
> On Fri, Sep 26, 2014 at 11:10 AM, Eric W. Biederman
> <ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org> wrote:
>> Nicolas Dichtel <nicolas.dichtel-pdR9zngts4EAvxtiuMwx3w@public.gmane.org> writes:
>>
>>> The goal of this serie is to be able to multicast netlink messages with an
>>> attribute that identify a peer netns.
>>> This is needed by the userland to interpret some informations contained in
>>> netlink messages (like IFLA_LINK value, but also some other attributes in case
>>> of x-netns netdevice (see also
>>> http://thread.gmane.org/gmane.linux.network/315933/focus=316064 and
>>> http://thread.gmane.org/gmane.linux.kernel.containers/28301/focus=4239)).
>>
>> I want say that the problem addressed by patch 3/5 of this series is a
>> fundamentally valid problem. We have network objects spanning network
>> namespaces and it would be very nice to be able to talk about them in
>> netlink, and file descriptors are too local and argubably too heavy
>> weight for netlink quires and especially for netlink broadcast messages.
>>
>> Furthermore the concept of ineternal concept of peernet2id seems valid.
>>
>> However what you do not address is a way for CRIU (aka process
>> migration) to be able to restore these ids after process migration.
>> Going farther it looks like you are actively breaking process migration
>> at this time, making this set of patches a no-go.
>>
>> When adding a new form of namespace id CRIU patches are just about
>> as necessary as iproute patches.
>>
>>> Ids are stored in the parent user namespace. These ids are valid only inside
>>> this user namespace. The user can retrieve these ids via a new netlink messages,
>>> but only if peer netns are in the same user namespace.
>>
>> That does not describe what you have actually implemented in the
>> patches.
>>
>> I see two ways to go with this.
>>
>> - A per network namespace table to that you can store ids for ``peer''
>> network namespaces. The table would need to be populated manually by
>> the likes of ip netns add.
>>
>> That flips the order of assignment and makes this idea solid.
>>
>> Unfortunately in the case of a fully referencing mesh of N network
>> namespaces such a mesh winds up taking O(N^2) space, which seems
>> undesirable.
>>
>> - Add a netlink attribute that says this network element is in a peer
>> network namespace.
>>
>> Add a unicast query message that let's you ask if the remote
>> end of a tunnel is in a network namespace specified by file
>> descriptor.
>>
>> I personally lean towards the second version as it is fundamentally
>> simpler, and generally scales better, and the visibility controls are
>> the existing visibility controls. The only downside is it requires
>> a query after receiving a netlink broadcast message for the times that
>> we care.
>
> The downside of that approach, and all the similar kcmp stuff, is that
> it scales poorly for applications using it. This is probably not the
> end of the world, but it's not ideal.
Agreed, the efficiency is not ideal and there is plenty of room for
optimization. We could certainly adopt some of kcmps ordering
infrastructure to make it suck less, or even potentially work out how
to return a file descriptor to the network namespace in question.
The key insight of my second proposal is that we can get out of the
broadcast message business, and only care about the remote namespace for
unicast messages. Putting the work in an infrequently used slow path
instead of a comparitively common path gives us much more freedom in
the implementation.
Eric
^ 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