Eric Dumazet [Wed, 26 Jan 2011 19:28:23 +0000 (19:28 +0000)]
drivers/net: remove some rcu sparse warnings
Add missing __rcu annotations and helpers.
minor : Fix some rcu_dereference() calls in macvtap
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Wed, 26 Jan 2011 18:08:02 +0000 (18:08 +0000)]
net: fix dev_seq_next()
Commit c6d14c84566d (net: Introduce for_each_netdev_rcu() iterator)
added a race in dev_seq_next().
The rcu_dereference() call should be done _before_ testing the end of
list, or we might return a wrong net_device if a concurrent thread
changes net_device list under us.
Note : discovered thanks to a sparse warning :
net/core/dev.c:3919:9: error: incompatible types in comparison expression
(different address spaces)
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> CC: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 27 Jan 2011 22:55:22 +0000 (14:55 -0800)]
ipv6: Remove route peer binding assertions.
They are bogus. The basic idea is that I wanted to make sure
that prefixed routes never bind to peers.
The test I used was whether RTF_CACHE was set.
But first of all, the RTF_CACHE flag is set at different spots
depending upon which ip6_rt_copy() caller you're talking about.
I've validated all of the code paths, and even in the future
where we bind peers more aggressively (for route metric COW'ing)
we never bind to prefix'd routes, only fully specified ones.
This even applies when addrconf or icmp6 routes are allocated.
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Tue, 25 Jan 2011 23:18:38 +0000 (23:18 +0000)]
net: add kmemcheck annotation in __alloc_skb()
pskb_expand_head() triggers a kmemcheck warning when copy of
skb_shared_info is done in pskb_expand_head()
This is because destructor_arg field is not necessarily initialized at
this point. Add kmemcheck_annotate_variable() call in __alloc_skb() to
instruct kmemcheck this is a normal situation.
Resolves bugzilla.kernel.org 27212
Reference: https://bugzilla.kernel.org/show_bug.cgi?id=27212 Reported-by: Christian Casteyde <casteyde.christian@free.fr> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> CC: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Kurt Van Dijck [Wed, 26 Jan 2011 04:55:24 +0000 (04:55 +0000)]
net: fix validate_link_af in rtnetlink core
I'm testing an API that uses IFLA_AF_SPEC attribute.
In the rtnetlink core , the set_link_af() member
of the rtnl_af_ops struct receives the nested attribute
(as I expected), but the validate_link_af() member
receives the parent attribute.
IMO, this patch fixes this.
Signed-off-by: Kurt Van Dijck <kurt.van.dijck@eia.be> Signed-off-by: David S. Miller <davem@davemloft.net>
Patch fixes: "DMA-API: device driver tries to free an invalid DMA
memory address" warning reported here:
https://bugzilla.redhat.com/show_bug.cgi?id=639824
Reported-by: Frantisek Hanzlik <franta@hanzlici.cz> Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ian Campbell [Thu, 27 Jan 2011 04:14:03 +0000 (04:14 +0000)]
xen: netfront: handle incoming GSO SKBs which are not CHECKSUM_PARTIAL
The Linux network stack expects all GSO SKBs to have ip_summed ==
CHECKSUM_PARTIAL (which implies that the frame contains a partial
checksum) and the Xen network ring protocol similarly expects an SKB
which has GSO set to also have NETRX_csum_blank (which also implies a
partial checksum).
However there have been cases of buggy guests which mark a frame as
GSO but do not set csum_blank. If we detect that we a receiving such a
frame (which manifests as ip_summed != PARTIAL && skb_is_gso) then
force the SKB to partial and recalculate the checksum, since we cannot
rely on the peer having done so if they have not set csum_blank.
Add an ethtool stat to track occurances of this event.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: David Miller <davem@davemloft.net> Cc: xen-devel@lists.xensource.com Cc: netdev@vger.kernel.org Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Wed, 26 Jan 2011 00:04:18 +0000 (00:04 +0000)]
econet: remove compiler warnings
net/econet/af_econet.c: In function ‘econet_sendmsg’:
net/econet/af_econet.c:494: warning: label ‘error’ defined but not used
net/econet/af_econet.c:268: warning: unused variable ‘sk’
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Acked-by: Phil Blundell <philb@gnu.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Felix Fietkau [Mon, 24 Jan 2011 18:11:54 +0000 (19:11 +0100)]
ath9k: fix misplaced debug code
The commit 'ath9k: Add more information to debugfs xmit file.' added more
debug counters to ath9k and also added some lines of code to ath9k_hw.
Since ath9k_hw is also used by ath9k_htc, its code must not depend on ath9k
data structures. In this case it was not fatal, but it's still wrong, so
the code needs to be moved back to ath9k.
Signed-off-by: Felix Fietkau <nbd@openwrt.org> Cc: Ben Greear <greearb@candelatech.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
carl9170: utilize fw seq counter for mgmt/non-QoS data frames
"mac80211 will properly assign sequence numbers to QoS-data
frames but cannot do so correctly for non-QoS-data and
management frames because beacons need them from that counter
as well and mac80211 cannot guarantee proper sequencing."
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
David S. Miller [Thu, 27 Jan 2011 04:51:05 +0000 (20:51 -0800)]
net: Implement read-only protection and COW'ing of metrics.
Routing metrics are now copy-on-write.
Initially a route entry points it's metrics at a read-only location.
If a routing table entry exists, it will point there. Else it will
point at the all zero metric place-holder called 'dst_default_metrics'.
The writeability state of the metrics is stored in the low bits of the
metrics pointer, we have two bits left to spare if we want to store
more states.
For the initial implementation, COW is implemented simply via kmalloc.
However future enhancements will change this to place the writable
metrics somewhere else, in order to increase sharing. Very likely
this "somewhere else" will be the inetpeer cache.
Note also that this means that metrics updates may transiently fail
if we cannot COW the metrics successfully.
But even by itself, this patch should decrease memory usage and
increase cache locality especially for routing workloads. In those
cases the read-only metric copies stay in place and never get written
to.
TCP workloads where metrics get updated, and those rare cases where
PMTU triggers occur, will take a very slight performance hit. But
that hit will be alleviated when the long-term writable metrics
move to a more sharable location.
Since the metrics storage went from a u32 array of RTAX_MAX entries to
what is essentially a pointer, some retooling of the dst_entry layout
was necessary.
Most importantly, we need to preserve the alignment of the reference
count so that it doesn't share cache lines with the read-mostly state,
as per Eric Dumazet's alignment assertion checks.
The only non-trivial bit here is the move of the 'flags' member into
the writeable cacheline. This is OK since we are always accessing the
flags around the same moment when we made a modification to the
reference count.
Signed-off-by: David S. Miller <davem@davemloft.net>
Reported-by: Breno Leitao <leitao@linux.vnet.ibm.com> Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Michael Chan [Tue, 25 Jan 2011 22:14:50 +0000 (22:14 +0000)]
cnic: Fix big endian bug
The chip's page tables did not set up properly on big endian machines,
causing EEH errors on PPC machines.
Reported-by: Breno Leitao <leitao@linux.vnet.ibm.com> Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Johannes Berg [Wed, 26 Jan 2011 08:22:15 +0000 (09:22 +0100)]
mac80211: use DECLARE_EVENT_CLASS
For events that include only the local struct as
their parameter, we can use DECLARE_EVENT_CLASS
and save quite some binary size across segments
as well lines of code.
text data bss dec hex filename
375745 19296 916 395957 60ab5 mac80211.ko.before
367473 17888 916 386277 5e4e5 mac80211.ko.after
-8272 -1408 0 -9680 -25d0 delta
Some more tracepoints with identical arguments
could be combined like this but for now this is
the one that benefits most.
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Eric Dumazet [Wed, 26 Jan 2011 07:21:57 +0000 (07:21 +0000)]
net_sched: sch_mqprio: dont leak kernel memory
mqprio_dump() should make sure all fields of struct tc_mqprio_qopt are
initialized.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> CC: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Wed, 26 Jan 2011 21:13:10 +0000 (13:13 -0800)]
tg3: Use new VLAN code
This patch pivots the tg3 driver to the new VLAN infrastructure.
All references to vlgrp have been removed. The driver still attempts to
disable VLAN tag stripping if CONFIG_VLAN_8021Q or
CONFIG_VLAN_8021Q_MODULE is not defined.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Thomas Chou [Tue, 25 Jan 2011 19:22:05 +0000 (19:22 +0000)]
smc91x: add devicetree support
Signed-off-by: Thomas Chou <thomas@wytron.com.tw> Reviewed-by: Grant Likely <grant.likely@secretlab.ca> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Tue, 25 Jan 2011 15:58:57 +0000 (15:58 +0000)]
tg3: Update copyrights and update version to 3.117
This patch updates copyrights and updates the tg3 version to 3.117.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Tue, 25 Jan 2011 15:58:56 +0000 (15:58 +0000)]
tg3: Fix eee preprocessor naming
This patch fixes a preprocessor naming bug for one of the EEE registers.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Tue, 25 Jan 2011 15:58:55 +0000 (15:58 +0000)]
tg3: Fix EEE interoperability issue
This patch fixes a problem where EEE will fail to work in certain
environments.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Tue, 25 Jan 2011 15:58:54 +0000 (15:58 +0000)]
tg3: Disable EEE during loopback tests
EEE interferes with the hardware's ability to loop a packet back to the
host. This patch disables the feature for the duration of the test.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Tue, 25 Jan 2011 15:58:53 +0000 (15:58 +0000)]
tg3: Disable MAC loopback test for CPMU devices
On CPMU devices, the MAC loopback test does not test any important paths
the phy loopback test doesn't also test. The phy loopback test is the
more comprehensive test. This patch disables the MAC loopback test for
these devices.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Tue, 25 Jan 2011 15:58:52 +0000 (15:58 +0000)]
tg3: Fix loopback tests
The half-duplex bit in the MAC MODE register will be set during the
loopback test if the external link is in half-duplex mode. This will
cause the loopback test to fail on newer devices. This patch turns the
half-duplex bit off for the test.
Also, newer devices fail the internal phy loopback test because the phy
link takes a little while to come up. This patch adds code to wait for
the link before proceeding with the test.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Tue, 25 Jan 2011 15:58:51 +0000 (15:58 +0000)]
tg3: Restrict phy ioctl access
If management firmware is present and the device is down, the firmware
will assume control of the phy. If a phy access were allowed from the
host, it will collide with firmware phy accesses, resulting in
unpredictable behavior. This patch fixes the problem by disallowing phy
accesses during the problematic condition.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Tue, 25 Jan 2011 15:58:50 +0000 (15:58 +0000)]
tg3: Disable multivec mode for 1 MSIX vector
For single vector MSI-X allocations, we do not want to enable
multivector modes. This patch makes the necessary corrections.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Tue, 25 Jan 2011 15:58:49 +0000 (15:58 +0000)]
tg3: Use new VLAN code
This patch pivots the tg3 driver to the new VLAN infrastructure.
All references to vlgrp have been removed. The driver still attempts to
disable VLAN tag stripping if CONFIG_VLAN_8021Q or
CONFIG_VLAN_8021Q_MODULE is not defined.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Tue, 25 Jan 2011 15:58:48 +0000 (15:58 +0000)]
tg3: Fix 5719 A0 tx completion bug
The 5719 A0 has a bug that manifests itself as if the chipset were
reordering memory writes. The best known way to solve this problem is
to turn off LSO and jumbo frames.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Tue, 25 Jan 2011 15:58:47 +0000 (15:58 +0000)]
tg3: Revise 5719 internal FIFO overflow solution
Commit cf79003d598b1f82a4caa0564107283b4f560e14, entitled
"tg3: Fix 5719 internal FIFO overflow problem", proposed a way to solve
an internal FIFO overflow problem. We have since discovered a slightly
better way to solve the problem. This patch changes the code so that
the problem is contained closer to the problem source.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jerry Chu [Tue, 25 Jan 2011 21:46:30 +0000 (13:46 -0800)]
TCP: fix a bug that triggers large number of TCP RST by mistake
This patch fixes a bug that causes TCP RST packets to be generated
on otherwise correctly behaved applications, e.g., no unread data
on close,..., etc. To trigger the bug, at least two conditions must
be met:
1. The FIN flag is set on the last data packet, i.e., it's not on a
separate, FIN only packet.
2. The size of the last data chunk on the receive side matches
exactly with the size of buffer posted by the receiver, and the
receiver closes the socket without any further read attempt.
This bug was first noticed on our netperf based testbed for our IW10
proposal to IETF where a large number of RST packets were observed.
netperf's read side code meets the condition 2 above 100%.
Before the fix, tcp_data_queue() will queue the last skb that meets
condition 1 to sk_receive_queue even though it has fully copied out
(skb_copy_datagram_iovec()) the data. Then if condition 2 is also met,
tcp_recvmsg() often returns all the copied out data successfully
without actually consuming the skb, due to a check
"if ((chunk = len - tp->ucopy.len) != 0) {"
and
"len -= chunk;"
after tcp_prequeue_process() that causes "len" to become 0 and an
early exit from the big while loop.
I don't see any reason not to free the skb whose data have been fully
consumed in tcp_data_queue(), regardless of the FIN flag. We won't
get there if MSG_PEEK is on. Am I missing some arcane cases related
to urgent data?
Signed-off-by: H.K. Jerry Chu <hkchu@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Bruno Randolf [Wed, 19 Jan 2011 09:20:52 +0000 (18:20 +0900)]
ath5k: ath5k_setup_channels cleanup and whitespace
Remove useless test_bit - it's not going to happen because of the way this
function is called only when that bit is set.
And fix some whitespace.
Signed-off-by: Bruno Randolf <br1@einfach.org> Acked-by: Bob Copeland <me@bobcopeland.com> Acked-by: Nick Kossifidis <mickflemm@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
ath9k_hw: read and backup AR_WA register value even before chip reset on.
We need to read and backup AR_WA register value permanently and reading
this after the chip is awakened results in this register being zeroed out.
This seems to fix the ASPM with L1 enabled issue that we have observed.
The laptop becomes very slow and hangs mostly with ASPM L1 enabled without
this fix.
Cc: Stable Kernel <stable@kernel.org> Signed-off-by: Senthil Balasubramanian <senthilkumar@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
ath9k_hw: Fix incorrect macversion and macrev checks
There are few places where we are checking for macversion and revsions
before RTC is powered ON. However we are reading the macversion and
revisions only after RTC is powered ON and so both macversion and
revisions are actully zero and this leads to incorrect srev checks
Incorrect srev checks can cause registers to be configured wrongly and can
cause unexpected behavior. Fixing this seems to address the ASPM issue that
we have observed. The laptop becomes very slow and hangs mostly with ASPM L1
enabled without this fix.
fix this by reading the macversion and revisisons even before we start
using them. There is no reason why should we delay reading this info
until RTC is powered on as this is just a register information.
Cc: Stable Kernel <stable@kernel.org> Signed-off-by: Senthil Balasubramanian <senthilkumar@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Reinette Chatre [Tue, 25 Jan 2011 16:38:06 +0000 (08:38 -0800)]
MAINTAINERS: remove Reinette Chatre as iwlwifi maintainer
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Felix Fietkau [Mon, 24 Jan 2011 18:28:49 +0000 (19:28 +0100)]
mac80211: fix a crash in ieee80211_beacon_get_tim on change_interface
Some drivers (e.g. ath9k) do not always disable beacons when they're
supposed to. When an interface is changed using the change_interface op,
the mode specific sdata part is in an undefined state and trying to
get a beacon at this point can produce weird crashes.
To fix this, add a check for ieee80211_sdata_running before using
anything from the sdata.
Signed-off-by: Felix Fietkau <nbd@openwrt.org> Cc: stable@kernel.org Signed-off-by: John W. Linville <linville@tuxdriver.com>
because the core semantic change to ipv6 address handling on ifdown
has broken some things, in particular "disable_ipv6" sysctl handling.
Stephen has made several attempts to get things back in working order,
but nothing has restored disable_ipv6 fully yet.
Reported-by: Eric W. Biederman <ebiederm@xmission.com> Tested-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
textsearch: doc - fix spelling in lib/textsearch.c.
Found the following spelling errors while reading the textsearch code:
"facitilies" -> "facilities"
"continously" -> "continuously"
"arbitary" -> "arbitrary"
"patern" -> "pattern"
"occurences" -> "occurrences"
I'll try to push this patch through DaveM, given the only users
of textsearch is in the net/ tree (nf_conntrack_amanda.c, xt_string.c
and em_text.c)
Signed-off-by: Jesper Sander <sander.contrib@gmail.com> Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk> Signed-off-by: David S. Miller <davem@davemloft.net>
Vlad Dogaru [Mon, 24 Jan 2011 03:37:29 +0000 (03:37 +0000)]
net: add sysfs entry for device group
The group of a network device can be queried or changed from userspace
using sysfs.
For example, considering sysfs mounted in /sys, one can change the group
that interface lo belongs to:
echo 1 > /sys/class/net/lo/group
Signed-off-by: Vlad Dogaru <ddvlad@rosedu.org> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jesper Juhl [Sun, 23 Jan 2011 12:19:55 +0000 (12:19 +0000)]
USB NET KL5KUSB101: Fix mem leak in error path of kaweth_download_firmware()
We will leak the storage allocated by request_firmware() if the size of
the firmware is greater than KAWETH_FIRMWARE_BUF_SIZE.
This removes the leak by calling release_firmware() before we return
-ENOSPC.
Signed-off-by: Jesper Juhl <jj@chaosbits.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Michael Chan [Mon, 24 Jan 2011 12:59:02 +0000 (12:59 +0000)]
bnx2: Always set ETH_FLAG_TXVLAN
TSO does not work if the VLAN tag is in the packet (non-accelerated).
We may be able to remove this restriction in future firmware.
Reported-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
John Fastabend [Fri, 21 Jan 2011 16:35:18 +0000 (16:35 +0000)]
dcbnl: make get_app handling symmetric for IEEE and CEE DCBx
The IEEE get/set app handlers use generic routines and do not
require the net_device to implement the dcbnl_ops routines. This
patch makes it symmetric so user space and drivers do not have
to handle the CEE version and IEEE DCBx versions differently.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Sakari Ailus [Fri, 21 Jan 2011 10:59:30 +0000 (10:59 +0000)]
tlan: Code cleanup: checkpatch.pl is relatively happy now.
- Remove CamelCase.
- Convert hexadecimals to lower case.
- Remove useless comments. Tlan driver contained a name of the function at
the end of it in a comment. Remove those comments.
- Remove local typedefs. Use real types instead of typedefs in code.
- Resolve space issues and reindent.
- One warning remain, it's a case where printing a single line involves a
number of printk()s.
Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi> Signed-off-by: David S. Miller <davem@davemloft.net>
Ben Hutchings [Wed, 19 Jan 2011 11:03:53 +0000 (11:03 +0000)]
net: RPS: Enable hardware acceleration of RFS
Allow drivers for multiqueue hardware with flow filter tables to
accelerate RFS. The driver must:
1. Set net_device::rx_cpu_rmap to a cpu_rmap of the RX completion
IRQs (in queue order). This will provide a mapping from CPUs to the
queues for which completions are handled nearest to them.
2. Implement net_device_ops::ndo_rx_flow_steer. This operation adds
or replaces a filter steering the given flow to the given RX queue, if
possible.
3. Periodically remove filters for which rps_may_expire_flow() returns
true.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ben Hutchings [Wed, 19 Jan 2011 11:03:25 +0000 (11:03 +0000)]
lib: cpu_rmap: CPU affinity reverse-mapping
When initiating I/O on a multiqueue and multi-IRQ device, we may want
to select a queue for which the response will be handled on the same
or a nearby CPU. This requires a reverse-map of IRQ affinity. Add
library functions to support a generic reverse-mapping from CPUs to
objects with affinity and the specific case where the objects are
IRQs.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Mon, 24 Jan 2011 22:41:20 +0000 (14:41 -0800)]
tcp: fix bug in listening_get_next()
commit a8b690f98baf9fb19 (tcp: Fix slowness in read /proc/net/tcp)
introduced a bug in handling of SYN_RECV sockets.
st->offset represents number of sockets found since beginning of
listening_hash[st->bucket].
We should not reset st->offset when iterating through
syn_table[st->sbucket], or else if more than ~25 sockets (if
PAGE_SIZE=4096) are in SYN_RECV state, we exit from listening_get_next()
with a too small st->offset
Next time we enter tcp_seek_last_pos(), we are not able to seek past
already found sockets.
Reported-by: PK <runningdoglackey@yahoo.com> CC: Tom Herbert <therbert@google.com> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Michal Schmidt [Mon, 24 Jan 2011 12:08:48 +0000 (12:08 +0000)]
GRO: fix merging a paged skb after non-paged skbs
Suppose that several linear skbs of the same flow were received by GRO. They
were thus merged into one skb with a frag_list. Then a new skb of the same flow
arrives, but it is a paged skb with data starting in its frags[].
Before adding the skb to the frag_list skb_gro_receive() will of course adjust
the skb to throw away the headers. It correctly modifies the page_offset and
size of the frag, but it leaves incorrect information in the skb:
->data_len is not decreased at all.
->len is decreased only by headlen, as if no change were done to the frag.
Later in a receiving process this causes skb_copy_datagram_iovec() to return
-EFAULT and this is seen in userspace as the result of the recv() syscall.
In practice the bug can be reproduced with the sfc driver. By default the
driver uses an adaptive scheme when it switches between using
napi_gro_receive() (with skbs) and napi_gro_frags() (with pages). The bug is
reproduced when under rx load with enough successful GRO merging the driver
decides to switch from the former to the latter.
Manual control is also possible, so reproducing this is easy with netcat:
- on machine1 (with sfc): nc -l 12345 > /dev/null
- on machine2: nc machine1 12345 < /dev/zero
- on machine1:
echo 1 > /sys/module/sfc/parameters/rx_alloc_method # use skbs
echo 2 > /sys/module/sfc/parameters/rx_alloc_method # use pages
- See that nc has quit suddenly.
[v2: Modified by Eric Dumazet to avoid advancing skb->data past the end
and to use a temporary variable.]
Signed-off-by: Michal Schmidt <mschmidt@redhat.com> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Mon, 24 Jan 2011 21:16:16 +0000 (13:16 -0800)]
net: arp_ioctl() must hold RTNL
Commit 941666c2e3e0 "net: RCU conversion of dev_getbyhwaddr() and
arp_ioctl()" introduced a regression, reported by Jamie Heilman.
"arp -Ds 192.168.2.41 eth0 pub" triggered the ASSERT_RTNL() assert
in pneigh_lookup()
Removing RTNL requirement from arp_ioctl() was a mistake, just revert
that part.
Reported-by: Jamie Heilman <jamie@audible.transient.net> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Shaohua Li [Mon, 24 Jan 2011 08:00:01 +0000 (08:00 +0000)]
fix a shutdown regression in intel_idle
Fix a shutdown regression caused by 2a2d31c8dc6f ("intel_idle: open
broadcast clock event"). The clockevent framework can automatically
shutdown broadcast timers for hotremove CPUs. And we get a shutdown
regression when we shutdown broadcast timer for hot remove CPU, so just
delete some code.
Also fix some section mismatch.
Reported-by: Ari Savolainen <ari.m.savolainen@gmail.com> Signed-off-by: Shaohua Li <shaohua.li@intel.com> Tested-by: Linus Torvalds <torvalds@linux-foundation.org> Cc: stable@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Mon, 24 Jan 2011 19:29:49 +0000 (05:29 +1000)]
Merge branch 'omap-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6
* 'omap-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6:
omap: DMA: clear interrupt status correctly
OMAP3: Devkit8000: Fix tps65930 pullup/pulldown configuration
arm: omap3: cm-t3517: minor comment fix
arm: omap3: cm-t3517: rtc fix
omap1: Fix sched_clock implementation when both MPU timer and 32K timer are used
omap1: Fix booting for 15xx and 730 with omap1_defconfig
omap1: Fix sched_clock for the MPU timer
OMAP: PRCM: remove duplicated headers
OMAP4: clockdomain: bypass unimplemented wake-up dependency functions on OMAP4
OMAP: counter_32k: init clocksource as part of machine timer init
Linus Torvalds [Mon, 24 Jan 2011 19:26:47 +0000 (05:26 +1000)]
Merge branch 'perf-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'perf-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
perf tools: Fix time function double declaration with glibc
perf tools: Fix build by checking if extra warnings are supported
perf tools: Fix build when using gcc 3.4.6
perf tools: Add missing header, fixes build
perf tools: Fix 64 bit integer format strings
perf test: Fix build on older glibcs
perf: perf_event_exit_task_context: s/rcu_dereference/rcu_dereference_raw/
perf test: Use cpu_map->[cpu] when setting affinity
perf symbols: Fix annotation of thumb code
perf: Annotate cpuctx->ctx.mutex to avoid a lockdep splat
powerpc, perf: Fix frequency calculation for overflowing counters (FSL version)
perf: Fix perf_event_init_task()/perf_event_free_task() interaction
perf: Fix find_get_context() vs perf_event_exit_task() race
Linus Torvalds [Mon, 24 Jan 2011 19:25:13 +0000 (05:25 +1000)]
Merge branch 'sched-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'sched-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
sched: Fix poor interactivity on UP systems due to group scheduler nice tune bug
Linus Torvalds [Mon, 24 Jan 2011 19:24:12 +0000 (05:24 +1000)]
Merge branch 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86: Fix jump label with RO/NX module protection crash
x86, hotplug: Fix powersavings with offlined cores on AMD
x86, mcheck, therm_throt.c: Export symbol platform_thermal_notify to allow coretemp to handler intr
x86: Use asm-generic/cacheflush.h
x86: Update CPU cache attributes table descriptors
can: at91_can: make can_id of mailbox 0 configurable
Due to a chip bug (errata 50.2.6.3 & 50.3.5.3 in
"AT91SAM9263 Preliminary 6249H-ATARM-27-Jul-09") the contents of mailbox
0 may be send under certain conditions (even if disabled or in rx mode).
The workaround in the errata suggests not to use the mailbox and load it
with an unused identifier.
This patch implements the second part of the workaround. A sysfs entry
"mb0_id" is introduced. While the interface is down it can be used to
configure the can_id of mailbox 0. The default value id 0x7ff.
In order to use an extended can_id add the CAN_EFF_FLAG (0x80000000U)
to the can_id. Example:
- standard id 0x7ff:
echo 0x7ff > /sys/class/net/can0/mb0_id
- extended id 0x1fffffff:
echo 0x9fffffff > /sys/class/net/can0/mb0_id
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de> Acked-by: Wolfgang Grandegger <wg@grandegger.com> Acked-by: Kurt Van Dijck <kurt.van.dijck@eia.be>
For the Documentation-part: Acked-by: Wolfram Sang <w.sang@pengutronix.de>
Due to a chip bug (errata 50.2.6.3 & 50.3.5.3 in
"AT91SAM9263 Preliminary 6249H-ATARM-27-Jul-09") the contents of mailbox
0 may be send under certain conditions (even if disabled or in rx mode).
The workaround in the errata suggests not to use the mailbox and load it
with a unused identifier.
This patch implements the first part of the workaround, it updates
AT91_MB_RX_NUM and AT91_MB_RX_FIRST (and the inline documentation)
so that mailbox 0 stays unused.
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de> Acked-by: Wolfgang Grandegger <wg@grandegger.com> Acked-by: Kurt Van Dijck <kurt.van.dijck@eia.be>
can: at91_can: clean up usage of AT91_MB_RX_FIRST and AT91_MB_RX_NUM
This patch cleans up the usage of two macros which specify the mailbox
usage. AT91_MB_RX_FIRST and AT91_MB_RX_NUM define the first and the
number of RX mailboxes. The current driver uses these variables in an
unclean way; assuming that AT91_MB_RX_FIRST is 0;
This patch cleans up the usage of these macros, no longer assuming
AT91_MB_RX_FIRST == 0.
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de> Acked-by: Wolfgang Grandegger <wg@grandegger.com>
Yong Zhang [Mon, 24 Jan 2011 07:33:52 +0000 (15:33 +0800)]
sched: Fix poor interactivity on UP systems due to group scheduler nice tune bug
Michael Witten and Christian Kujau reported that the autogroup
scheduling feature hurts interactivity on their UP systems.
It turns out that this is an older bug in the group scheduling code,
and the wider appeal provided by the autogroup feature exposed it
more prominently.
When on UP with FAIR_GROUP_SCHED enabled, tune shares
only affect tg->shares, but is not reflected in
tg->se->load. The reason is that update_cfs_shares()
does nothing on UP.
So introduce update_cfs_shares() for UP && FAIR_GROUP_SCHED.
This issue was found when enable autogroup scheduling was enabled,
but it is an older bug that also exists on cgroup.cpu on UP.
Reported-and-Tested-by: Michael Witten <mfwitten@gmail.com> Reported-and-Tested-by: Christian Kujau <christian@nerdbynature.de> Signed-off-by: Yong Zhang <yong.zhang0@gmail.com> Acked-by: Pekka Enberg <penberg@kernel.org> Acked-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org>
LKML-Reference: <20110124073352.GA24186@windriver.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Linus Torvalds [Mon, 24 Jan 2011 09:58:39 +0000 (19:58 +1000)]
Merge branch 'BUG_ON' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus
* 'BUG_ON' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
Remove MAYBE_BUILD_BUG_ON
BUILD_BUG_ON: make it handle more cases
* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
module: fix missing semicolons in MODULE macro usage
param: add null statement to compiled-in module params
module: fix linker error for MODULE_VERSION when !MODULE and CONFIG_SYSFS=n
module: show version information for built-in modules in sysfs
Bruce Allan [Thu, 6 Jan 2011 07:02:53 +0000 (07:02 +0000)]
e1000e: reduce scope of some variables, remove unnecessary ones
Static analysis of the driver code found some variables for which the scope
can be reduced, or remove the variable altogether.
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Rusty Russell [Mon, 24 Jan 2011 20:45:10 +0000 (14:45 -0600)]
BUILD_BUG_ON: make it handle more cases
BUILD_BUG_ON used to use the optimizer to do code elimination or fail
at link time; it was changed to first the size of a negative array (a
nicer compile time error), then (in 8c87df457cb58fe75b9b893007917cf8095660a0) to a bitfield.
This forced us to change some non-constant cases to MAYBE_BUILD_BUG_ON();
as Jan points out in that commit, it didn't work as intended anyway.
bitfields: needs a literal constant at parse time, and can't be put under
"if (__builtin_constant_p(x))" for example.
negative array: can handle anything, but if the compiler can't tell it's
a constant, silently has no effect.
link time: breaks link if the compiler can't determine the value, but the
linker output is not usually as informative as a compiler error.
If we use the negative-array-size method *and* the link time trick,
we get the ability to use BUILD_BUG_ON() under __builtin_constant_p()
branches, and maximal ability for the compiler to detect errors at
build time.
We also document it thoroughly.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Jan Beulich <JBeulich@novell.com> Acked-by: Hollis Blanchard <hollisb@us.ibm.com>
projects / linux-beck.git / log