7 years agonet/mlx4_core: Unmap UAR also in the case of error flow
Dotan Barak [Thu, 25 Oct 2012 01:12:49 +0000]
net/mlx4_core: Unmap UAR also in the case of error flow

If a failure takes place during the EQ creation, we need to unmap the
UAR memory block too.

Signed-off-by: Dotan Barak <dotanb@dev.mellanox.co.il>
Signed-off-by: Uri Habusha <urih@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agonet/mlx4_en: Don't use vlan tag value as an indication for vlan presence
Moni Shoua [Thu, 25 Oct 2012 01:12:48 +0000]
net/mlx4_en: Don't use vlan tag value as an indication for vlan presence

The vlan tag can be zero. This is why it can't serve as an indication
that packet requires VLAN header in the TX flow.

Signed-off-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agonet/mlx4_en: Fix double-release-range in tx-rings
Jack Morgenstein [Thu, 25 Oct 2012 01:12:47 +0000]
net/mlx4_en: Fix double-release-range in tx-rings

The QP range is reserved as a single block. However, when freeing the
en resources, the tx-ring QPs are released both in mlx4_en_destroy_tx_ring
(one at a time) and in mlx4_en_free_resources (as a block release).

Fix by eliminating the one-at-a-time release in mlx4_en_destroy_tx_ring.

Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agobas_gigaset: fix pre_reset handling
Tilman Schmidt [Wed, 24 Oct 2012 08:44:32 +0000]
bas_gigaset: fix pre_reset handling

The delayed work function int_in_work() may call usb_reset_device()
and thus, indirectly, the driver's pre_reset method. Trying to
cancel the work synchronously in that situation would deadlock.
Fix by avoiding cancel_work_sync() in the pre_reset method.

If the reset was NOT initiated by int_in_work() this might cause
int_in_work() to run after the post_reset method, with urb_int_in
already resubmitted, so handle that case gracefully.

Signed-off-by: Tilman Schmidt <tilman@imap.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agovhost: fix mergeable bufs on BE hosts
Michael S. Tsirkin [Wed, 24 Oct 2012 18:37:51 +0000]
vhost: fix mergeable bufs on BE hosts

We copy head count to a 16 bit field, this works by chance on LE but on
BE guest gets 0. Fix it up.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Alexander Graf <agraf@suse.de>
Cc: stable@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agogianfar_ptp: use iomem, not ioports resource tree in probe
Wei Yang [Wed, 24 Oct 2012 05:21:36 +0000]
gianfar_ptp: use iomem, not ioports resource tree in probe

When using a 36 bit dtb file, the driver complains "resource busy".

Investigating the source of the message leads one to the
gianfar_ptp_probe function.

Since the type of the device resource requested in this function
is IORESOURCE_MEM, it should use "iomem_resource" instead of

Signed-off-by: Wei Yang <Wei.Yang@windriver.com>
Cc: Claudiu Manoil <claudiu.manoil@freescale.com>
Cc: Timur Tabi <timur@freescale.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoMerge branch 'fixes-for-3.7' of git://gitorious.org/linux-can/linux-can
David S. Miller [Thu, 25 Oct 2012 03:16:49 +0000]
Merge branch 'fixes-for-3.7' of git://gitorious.org/linux-can/linux-can

Marc Kleine-Budde says:

here are two patches for the v3.7 release cycle. A patch by Wolfgang Grandegger
for the flexcan driver, which switches off a workaround on the imx6q that is
not needed, because the hardware is not affected by that bug. And a patch by
Stephane Grosjean which updates the pci device table for the peak pci sja1000

Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoipv6: Set default hoplimit as zero.
Li RongQing [Wed, 24 Oct 2012 06:01:18 +0000]
ipv6: Set default hoplimit as zero.

Commit a02e4b7dae4551(Demark default hoplimit as zero) only changes the
hoplimit checking condition and default value in ip6_dst_hoplimit, not
zeros all hoplimit default value.

Keep the zeroing ip6_template_metrics[RTAX_HOPLIMIT - 1] to force it as
const, cause as a37e6e344910(net: force dst_default_metrics to const

Signed-off-by: Li RongQing <roy.qing.li@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoNET_VENDOR_TI: make available for am33xx as well
Peter Korsgaard [Tue, 23 Oct 2012 12:11:55 +0000]
NET_VENDOR_TI: make available for am33xx as well

The cpsw/davinci mdio ip cores are present on am33xx, so make NET_VENDOR_TI
visible for it as well.

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agopch_gbe: fix error handling in pch_gbe_up()
Veaceslav Falico [Tue, 23 Oct 2012 04:54:34 +0000]
pch_gbe: fix error handling in pch_gbe_up()

If we fail to allocate rx buffers pool by any reason, we'll just return
with an error, however we've previously successfully requested an irq. Fix
this by releasing the irq before returning the error.

Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agocan: sja1000: fix/add miniPCIe/cPCI PC/104Plus PCI/104e PEAK-System boards
Stephane Grosjean [Tue, 23 Oct 2012 10:37:17 +0000]
can: sja1000: fix/add miniPCIe/cPCI PC/104Plus PCI/104e PEAK-System boards

This patch adds the support for 4 new PCI boards based on the SJA1000 CAN
controller, from PEAK-System Technik:

. PCAN-miniPCIe (PCI-Express Mini slots, available as single or dual-channel)
. PCAN-cPCI (CompactPCI format, available as single or dual-channel)
. PCAN-PC/104-Plus (PC/104-Plus system, av. as 1, 2 or 4 channels)
. PCAN-PCI/104-Express (PCI/104-Express system, av. as 1 or 2 channels)

This patch also fixes a typo in existing "PEAK_MPCI_DEVICE_ID" identifier
(missing "e" for Express). Finally, it also changes the author as well as it
updates the module supported devices list.

Signed-off-by: Stephane Grosjean <s.grosjean@peak-system.com>
Acked-by: Oliver Hartkopp <socketcan@hartkopp.net>
Acked-by: Wolfgang Grandegger <wg@grandegger.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

7 years agocan: flexcan: disable bus error interrupts for the i.MX6q
Wolfgang Grandegger [Wed, 10 Oct 2012 19:10:42 +0000]
can: flexcan: disable bus error interrupts for the i.MX6q

This patch adds some Flexcan version info and removes the feature flag
FLEXCAN_HAS_BROKEN_ERR_STATE for the i.MX6Q. It also has the line [TR]WRN_INT
properly connected.

Cc: Hui Wang <jason77.wang@gmail.com>
Cc: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
[mkl: convert to incremental patch]
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

7 years agotcp: Reject invalid ack_seq to Fast Open sockets
Jerry Chu [Mon, 22 Oct 2012 11:26:36 +0000]
tcp: Reject invalid ack_seq to Fast Open sockets

A packet with an invalid ack_seq may cause a TCP Fast Open socket to switch
to the unexpected TCP_CLOSING state, triggering a BUG_ON kernel panic.

When a FIN packet with an invalid ack_seq# arrives at a socket in
the TCP_FIN_WAIT1 state, rather than discarding the packet, the current
code will accept the FIN, causing state transition to TCP_CLOSING.

This may be a small deviation from RFC793, which seems to say that the
packet should be dropped. Unfortunately I did not expect this case for
Fast Open hence it will trigger a BUG_ON panic.

It turns out there is really nothing bad about a TFO socket going into
TCP_CLOSING state so I could just remove the BUG_ON statements. But after
some thought I think it's better to treat this case like TCP_SYN_RECV
and return a RST to the confused peer who caused the unacceptable ack_seq
to be generated in the first place.

Signed-off-by: H.K. Jerry Chu <hkchu@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Acked-by: Yuchung Cheng <ycheng@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoMAINTAINERS: NETWORKING DRIVERS matches too much
Jean Delvare [Thu, 18 Oct 2012 20:11:38 +0000]

File pattern include/linux/*device.h matches too much, including
completely unrelated files. Replace it with an explicit list of
network device-related header files.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Joe Perches <joe@perches.com>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agopch_gbe: don't reset MAC_RX on FIFO overflow
Veaceslav Falico [Mon, 22 Oct 2012 04:43:25 +0000]
pch_gbe: don't reset MAC_RX on FIFO overflow

Currently, when FIFO_ERR happens, we stop the dma, wait for it to become
idle and then reset the whole MAC_RX logic (and after that we must re-set
multicast addresses and also re-enable MAC_RX when we're finally ready to
accept new packets). This leads to CRC errors on high number of incoming
packets and is not needed according to the datasheet.

This patch fixes it by the following steps:

1) remove this reset in pch_gbe_stop_receive(), which causes some functions
to not be used anywhere
2) remove already unused functions pch_gbe_wait_clr_bit_irq() and
pch_gbe_mac_reset_rx() to correctly build
3) move pch_gbe_enable_mac_rx() out of pch_gbe_start_receive() to
pch_gbe_up() where it's only needed after we've removed the MAC_RX reset
4) rename pch_gbe_start/stop_receive() to pch_gbe_enable/disable_dma_rx()
to more precisely reflect what the functions are now doing.

After these changes we already don't see the CRC errors and gain some
increase in RX processing speed.

Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agopch_gbe: don't re-set RX_FIFO_ERR flag in napi_poll
Veaceslav Falico [Mon, 22 Oct 2012 04:43:24 +0000]
pch_gbe: don't re-set RX_FIFO_ERR flag in napi_poll

If we were in RX_FIFO_ERR state and entered pch_gbe_napi_poll(), we'll
anyway clean some rx space and thus can continue to receive more packets.
Currently, we re-set the RX_FIFO_ERR in situations when we've exhausted our
budget for RX cleaning or cleaned some TX packets. Removing it gives us
+20%-40% speed increase and a lot less of RX_FIFO_ERRors reported.

Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agopch_gbe: create functions for MAC_RX {en,dis}able
Veaceslav Falico [Mon, 22 Oct 2012 04:43:23 +0000]
pch_gbe: create functions for MAC_RX {en,dis}able

Move MAC_RX-related bits into separate functions.

Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agonet: fix secpath kmemleak
Eric Dumazet [Mon, 22 Oct 2012 09:03:40 +0000]
net: fix secpath kmemleak

Mike Kazantsev found 3.5 kernels and beyond were leaking memory,
and tracked the faulty commit to a1c7fff7e18f59e ("net:
netdev_alloc_skb() use build_skb()")

While this commit seems fine, it uncovered a bug introduced
in commit bad43ca8325 ("net: introduce skb_try_coalesce()), in function

If head is stolen, we free the sk_buff,
without removing references on secpath (skb->sp).

So IPsec + IP defrag/reassembly (using skb coalescing), or
TCP coalescing could leak secpath objects.

Fix this bug by calling skb_release_head_state(skb) to properly
release all possible references to linked objects.

Reported-by: Mike Kazantsev <mk.fraggod@gmail.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Bisected-by: Mike Kazantsev <mk.fraggod@gmail.com>
Tested-by: Mike Kazantsev <mk.fraggod@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoqla3xxx: Ensure request/response queue addr writes to the registers
Joe Jin [Sun, 21 Oct 2012 14:40:36 +0000]
qla3xxx: Ensure request/response queue addr writes to the registers

Before use the request and response queue addr, make sure it has wrote
to the registers.

Signed-off-by: Joe Jin <joe.jin@oracle.com>
Cc: Jitendra Kalsaria <jitendra.kalsaria@qlogic.com>
Cc: Ron Mercer <ron.mercer@qlogic.com>
Acked-by: Jitendra Kalsaria <jitendra.kalsaria@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agotcp: add SYN/data info to TCP_INFO
Yuchung Cheng [Fri, 19 Oct 2012 15:14:44 +0000]
tcp: add SYN/data info to TCP_INFO

Add a bit TCPI_OPT_SYN_DATA (32) to the socket option TCP_INFO:tcpi_options.
It's set if the data in SYN (sent or received) is acked by SYN-ACK. Server or
client application can use this information to check Fast Open success rate.

Signed-off-by: Yuchung Cheng <ycheng@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoMerge branch 'for-davem' of git://git.kernel.org/pub/scm/linux/kernel/git/linville...
David S. Miller [Mon, 22 Oct 2012 00:57:37 +0000]
Merge branch 'for-davem' of git://git./linux/kernel/git/linville/wireless

John W. Linville says:

This is a batch of fixes intended for the 3.7 stream.

Dan Carpenter brings a fix for a simple signedness bug that could
prevent the proper termination of a loop.

Felix Fietkau found a few more places that need to use
ieee80211_free_txskb for properly releasing SKBs used by mac80211.

Franky Lin offers a pair of brcmfmac fixes, both fixing simple state
reporting errors.

Hante Meuleman corrects an error reporting case that wasn't handling
all types of errors properly.

Johan Hedberg offers a fix for an issue discovered at the Bluetooth
UnPlug Fest.  Gustavo says "the patch fixes a failure to pair with
devices that support the LE Secure Connections feature."

Johannes Berg sends an iwlwifi fix to handle a message type that
is too large for the normal command mechanism.  He also provides a
mac80211 fix to use HT20 channels when HT40 channels are not permitted.

Jouni Malinen offers a mac80211 fix for a masking error that was
incorrectly marking some frames.

Piotr Haber provides a fix to make sure bcma devices are unregistered

Stanislav Yakovlev gives us a fix for a panic in the ipw2200 driver.

Stanislaw Gruszka sends a pair of fixes: one prevents a mismatch on
connection states between cfg80211 and mac80211; the other prevents
some frame corruption related to handling encryption.

Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agocxgb4: Remove unnecessary #ifdef condition
Vipul Pandya [Fri, 19 Oct 2012 02:09:53 +0000]
cxgb4: Remove unnecessary #ifdef condition

This patch also fixes the build failure caused due to removal of #ifdef

Signed-off-by: Vipul Pandya <vipul@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wirel...
John W. Linville [Fri, 19 Oct 2012 17:55:42 +0000]
Merge branch 'master' of git://git./linux/kernel/git/linville/wireless into for-davem

7 years agoMerge tag 'batman-adv-fix-for-davem' of git://git.open-mesh.org/linux-merge
David S. Miller [Thu, 18 Oct 2012 19:36:59 +0000]
Merge tag 'batman-adv-fix-for-davem' of git://git.open-mesh.org/linux-merge

Included fixes:
- Fix broadcast packet CRC calculation which can lead to ~80% broadcast packet
- Fix a race condition in duplicate broadcast packet check

Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agotcp: fix FIONREAD/SIOCINQ
Eric Dumazet [Thu, 18 Oct 2012 09:14:12 +0000]

tcp_ioctl() tries to take into account if tcp socket received a FIN
to report correct number bytes in receive queue.

But its flaky because if the application ate the last skb,
we return 1 instead of 0.

Correct way to detect that FIN was received is to test SOCK_DONE.

Reported-by: Elliot Hughes <enh@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agonet: qmi_wwan: adding more ZTE devices
Bjørn Mork [Thu, 18 Oct 2012 05:11:29 +0000]
net: qmi_wwan: adding more ZTE devices

Analyzed a few Windows driver description files, supporting
this long list of devices:

%ztewwan.DeviceDesc0002%    = ztewwan.ndi, USB\VID_19D2&PID_0002&MI_01
%ztewwan.DeviceDesc0012%    = ztewwan.ndi, USB\VID_19D2&PID_0012&MI_01
%ztewwan.DeviceDesc0017%    = ztewwan.ndi, USB\VID_19D2&PID_0017&MI_03
%ztewwan.DeviceDesc0021%    = ztewwan.ndi, USB\VID_19D2&PID_0021&MI_04
%ztewwan.DeviceDesc0025%    = ztewwan.ndi, USB\VID_19D2&PID_0025&MI_01
%ztewwan.DeviceDesc0031%    = ztewwan.ndi, USB\VID_19D2&PID_0031&MI_04
%ztewwan.DeviceDesc0042%    = ztewwan.ndi, USB\VID_19D2&PID_0042&MI_04
%ztewwan.DeviceDesc0049%    = ztewwan.ndi, USB\VID_19D2&PID_0049&MI_05
%ztewwan.DeviceDesc0052%    = ztewwan.ndi, USB\VID_19D2&PID_0052&MI_04
%ztewwan.DeviceDesc0055%    = ztewwan.ndi, USB\VID_19D2&PID_0055&MI_01
%ztewwan.DeviceDesc0058%    = ztewwan.ndi, USB\VID_19D2&PID_0058&MI_04
%ztewwan.DeviceDesc0063%    = ztewwan.ndi, USB\VID_19D2&PID_0063&MI_04
%ztewwan.DeviceDesc2002%    = ztewwan.ndi, USB\VID_19D2&PID_2002&MI_04
%ztewwan.DeviceDesc0104%    = ztewwan.ndi, USB\VID_19D2&PID_0104&MI_04
%ztewwan.DeviceDesc0113%    = ztewwan.ndi, USB\VID_19D2&PID_0113&MI_05
%ztewwan.DeviceDesc0118%    = ztewwan.ndi, USB\VID_19D2&PID_0118&MI_05
%ztewwan.DeviceDesc0121%    = ztewwan.ndi, USB\VID_19D2&PID_0121&MI_05
%ztewwan.DeviceDesc0123%    = ztewwan.ndi, USB\VID_19D2&PID_0123&MI_04
%ztewwan.DeviceDesc0124%    = ztewwan.ndi, USB\VID_19D2&PID_0124&MI_05
%ztewwan.DeviceDesc0125%    = ztewwan.ndi, USB\VID_19D2&PID_0125&MI_06
%ztewwan.DeviceDesc0126%    = ztewwan.ndi, USB\VID_19D2&PID_0126&MI_05
%ztewwan.DeviceDesc1008%    = ztewwan.ndi, USB\VID_19D2&PID_1008&MI_04
%ztewwan.DeviceDesc1010%    = ztewwan.ndi, USB\VID_19D2&PID_1010&MI_04
%ztewwan.DeviceDesc1012%    = ztewwan.ndi, USB\VID_19D2&PID_1012&MI_04
%ztewwan.DeviceDesc1402%    = ztewwan.ndi, USB\VID_19D2&PID_1402&MI_02
%ztewwan.DeviceDesc0157%    = ztewwan.ndi, USB\VID_19D2&PID_0157&MI_05
%ztewwan.DeviceDesc0158%    = ztewwan.ndi, USB\VID_19D2&PID_0158&MI_03
%ztewwan.DeviceDesc1401%    = ztewwan.ndi, USB\VID_19D2&PID_1401&MI_02
%ztewwan.DeviceDesc0130%    = ztewwan.ndi, USB\VID_19D2&PID_0130&MI_01
%ztewwan.DeviceDesc0133%    = ztewwan.ndi, USB\VID_19D2&PID_0133&MI_03
%ztewwan.DeviceDesc0176%    = ztewwan.ndi, USB\VID_19D2&PID_0176&MI_03
%ztewwan.DeviceDesc0178%    = ztewwan.ndi, USB\VID_19D2&PID_0178&MI_03
%ztewwan.DeviceDesc0168%    = ztewwan.ndi, USB\VID_19D2&PID_0168&MI_04
%ztewwan.DeviceDesc0191%    = ztewwan.ndi, USB\VID_19D2&PID_0191&MI_04
%ztewwan.DeviceDesc0167%    = ztewwan.ndi, USB\VID_19D2&PID_0167&MI_04
%ztewwan.DeviceDesc0199%    = ztewwan.ndi, USB\VID_19D2&PID_0199&MI_01
%ztewwan.DeviceDesc0200%    = ztewwan.ndi, USB\VID_19D2&PID_0200&MI_01
%ztewwan.DeviceDesc0257%    = ztewwan.ndi, USB\VID_19D2&PID_0257&MI_03
%ztewwan.DeviceDesc1018%    = ztewwan.ndi, USB\VID_19D2&PID_1018&MI_03
%ztewwan.DeviceDesc1426%    = ztewwan.ndi, USB\VID_19D2&PID_1426&MI_02
%ztewwan.DeviceDesc1247%    = ztewwan.ndi, USB\VID_19D2&PID_1247&MI_04
%ztewwan.DeviceDesc1425%    = ztewwan.ndi, USB\VID_19D2&PID_1425&MI_02
%ztewwan.DeviceDesc1424%    = ztewwan.ndi, USB\VID_19D2&PID_1424&MI_02
%ztewwan.DeviceDesc1252%    = ztewwan.ndi, USB\VID_19D2&PID_1252&MI_04
%ztewwan.DeviceDesc1254%    = ztewwan.ndi, USB\VID_19D2&PID_1254&MI_04
%ztewwan.DeviceDesc1255A%   = ztewwan.ndi, USB\VID_19D2&PID_1255&MI_03
%ztewwan.DeviceDesc1255B%   = ztewwan.ndi, USB\VID_19D2&PID_1255&MI_04
%ztewwan.DeviceDesc1256%    = ztewwan.ndi, USB\VID_19D2&PID_1256&MI_04
%ztewwan.DeviceDesc1245%    = ztewwanCombB.ndi, USB\VID_19D2&PID_1245&MI_04
%ztewwan.DeviceDesc1021%    = ztewwan.ndi, USB\VID_19D2&PID_1021&MI_02

Adding the ones we were missing.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agonetlink: use kfree_rcu() in netlink_release()
Eric Dumazet [Thu, 18 Oct 2012 03:21:55 +0000]
netlink: use kfree_rcu() in netlink_release()

On some suspend/resume operations involving wimax device, we have
noticed some intermittent memory corruptions in netlink code.

Stéphane Marchesin tracked this corruption in netlink_update_listeners()
and suggested a patch.

It appears netlink_release() should use kfree_rcu() instead of kfree()
for the listeners structure as it may be used by other cpus using RCU

netlink_release() must set to NULL the listeners pointer when
it is about to be freed.

Also have to protect netlink_update_listeners() and
netlink_has_listeners() if listeners is NULL.

Add a nl_deref_protected() lockdep helper to properly document which
locks protects us.

Reported-by: Jonathan Kliegman <kliegs@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Stéphane Marchesin <marcheu@google.com>
Cc: Sam Leffler <sleffler@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoipv4: Fix flushing of cached routing informations
Steffen Klassert [Wed, 17 Oct 2012 21:17:44 +0000]
ipv4: Fix flushing of cached routing informations

Currently we can not flush cached pmtu/redirect informations via
the ipv4_sysctl_rtcache_flush sysctl. We need to check the rt_genid
of the old route and reset the nh exeption if the old route is
expired when we bind a new route to a nh exeption.

Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agovlan: allow to change type when no vlan device is hooked on netdev
Jiri Pirko [Wed, 17 Oct 2012 01:37:36 +0000]
vlan: allow to change type when no vlan device is hooked on netdev

vlan_info might be present but still no vlan devices might be there.
That is in case of vlan0 automatically added.

So in that case, allow to change netdev type.

Reported-by: Jon Stanley <jstanley@rmrf.net>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agophy: add AT803x driver
Matus Ujhelyi [Sun, 14 Oct 2012 19:07:16 +0000]
phy: add AT803x driver

This driver add support for wake over lan on AT803x phys.

Signed-off-by: Matus Ujhelyi <ujhelyi.m@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agousb/ipheth: Add iPhone 5 support
Jay Purohit [Sun, 14 Oct 2012 07:07:21 +0000]
usb/ipheth: Add iPhone 5 support

I noticed that the iPhone ethernet driver did not support
iPhone 5. I quickly added support to it in my kernel, here's
a patch.

Signed-off-by: Jay Purohit <jspurohit@velocitylimitless.com>
Acked-by: Valdis Kletnieks <valdis.kletnieks@vt.edu>
Signed-off-by: Jan Ceuleers <jan.ceuleers@computer.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoMerge branch 'master' of git://1984.lsi.us.es/nf
David S. Miller [Thu, 18 Oct 2012 19:17:09 +0000]
Merge branch 'master' of git://1984.lsi.us.es/nf

Pablo Neira Ayuso Says:

The following patchset contains Netfilter/IPVS updates for your
net tree, they are:

* Fix incorrect hooks for SNAT and DNAT (bug introduced in recent IPv6
  NAT changes), from Elison Niven.

* Fix xt_TEE (got broken with recent rt_gateway semantic change),
  from Eric Dumazet.

* Fix custom conntrack timeout policy attachment for IPv6, from myself.

* Always initialize ip_vs_timeout_user in case that TCP or UDP protocols
  is disabled, from Arnd Bergmann.

Note that I had to pull from your tree to obtain:

(c92b96553a80c1 ipv4: Add FLOWI_FLAG_KNOWN_NH)

which was required for the xt_TEE fix.

Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agobatman-adv: Fix potential broadcast BLA-duplicate-check race condition
Linus Lüssing [Wed, 17 Oct 2012 12:53:05 +0000]
batman-adv: Fix potential broadcast BLA-duplicate-check race condition

Threads in the bottom half of batadv_bla_check_bcast_duplist() might
otherwise for instance overwrite variables which other threads might
be using/reading at the same time in the top half, potentially
leading to messing up the bcast_duplist, possibly resulting in false
bridge loop avoidance duplicate check decisions.

Signed-off-by: Linus Lüssing <linus.luessing@web.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>

7 years agobatman-adv: Fix broadcast packet CRC calculation
Linus Lüssing [Wed, 17 Oct 2012 12:53:04 +0000]
batman-adv: Fix broadcast packet CRC calculation

So far the crc16 checksum for a batman-adv broadcast data packet, received
on a batman-adv hard interface, was calculated over zero bytes of its
content leading to many incoming broadcast data packets wrongly being
dropped (60-80% packet loss).

This patch fixes this issue by calculating the crc16 over the actual,
complete broadcast payload.

The issue is a regression introduced by
("batman-adv: add broadcast duplicate check").

Signed-off-by: Linus Lüssing <linus.luessing@web.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>

7 years agoapparmor: fix apparmor OOPS in audit_log_untrustedstring+0x1c/0x40
John Johansen [Wed, 17 Oct 2012 20:29:33 +0000]
apparmor: fix apparmor OOPS in audit_log_untrustedstring+0x1c/0x40

The capability defines have moved causing the auto generated names
of capabilities that apparmor uses in logging to be incorrect.

Fix the autogenerated table source to uapi/linux/capability.h

Reported-by: YanHong <clouds.yan@gmail.com>
Reported-by: Krzysztof Kolasa <kkolasa@winsoft.pl>
Analyzed-by: Al Viro <viro@ZenIV.linux.org.uk>
Signed-off-by: John Johansen <john.johansen@canonical.com>
Acked-by: David Howells <dhowells@redhat.com>
Acked-by: James Morris <james.l.morris@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agoMerge tag 'uapi-fixes-20121017' of git://git.infradead.org/users/dhowells/linux-headers
Linus Torvalds [Wed, 17 Oct 2012 20:40:22 +0000]
Merge tag 'uapi-fixes-20121017' of git://git.infradead.org/users/dhowells/linux-headers

Pull misc UAPI fixes from David Howells:
 "They do a number of things:

  (1) Import a patch from Catalin Marinas to extend the generic-y in
      Kbuild facility to uapi directories.

  (2) Make arch/tile's ucontext.h file use (1) and remove the header-y
      line from the kernel internal side of things.

  (3) Remove some now-empty conditional bits from include/linux/Kbuild.
      The contents got moved to the UAPI side of things along with new

  (4) Deal with now-empty files:

     (a) Empty Kbuild files under include/ get removed.

     (b) Empty Kbuild files under arch/ get comments to hold them as
         they are likely to end up with generic-y or genhdr-y lines.
         Deleting them appears to work if we want to go that route.

     (c) Put a comment into uapi/asm-generic/kvm_para.h to prevent the
         patch program from deleting that, and made the arches with
         empty kvm_para.h uapi files use that instead of having their
         own files.

     (d) Put comments into four other empty uapi/ headers to prevent the
         patch program from deleting them.

  A question: Is this the right way to deal with the now-empty Kbuild

  The ones under include/ are unlikely to be used - even for generated
  files, I think - so getting rid of them is probably okay.  Once all
  the bits are in, we can probably remove all the Kbuild files under
  include/ that aren't also under include/uapi/.

  The ones under arch/ are more of an issue because of the potential for
  generic-y and genhdr-y."

* tag 'uapi-fixes-20121017' of git://git.infradead.org/users/dhowells/linux-headers:
  UAPI: Make arch/sparc/include/uapi/asm/sigcontext.h non-empty
  UAPI: Make arch/sh/include/uapi/asm/hw_breakpoint.h non-empty
  UAPI: Make arch/mn10300/include/uapi/asm/setup.h non-empty
  UAPI: Put a comment into uapi/asm-generic/kvm_para.h and use it from arches
  UAPI: The tile arch uses the generic ucontext.h file
  UAPI: Place comments in empty arch Kbuilds to make them non-empty
  UAPI: Remove empty non-UAPI Kbuild files
  UAPI: Remove empty conditionals from include/linux/Kbuild
  UAPI: Make uapi/linux/irqnr.h non-empty
  uapi: Allow automatic generation of uapi/asm/ header files

7 years agoMerge branch 'for-john' of git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211
John W. Linville [Wed, 17 Oct 2012 20:23:33 +0000]
Merge branch 'for-john' of git://git./linux/kernel/git/jberg/mac80211

7 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Wed, 17 Oct 2012 19:41:18 +0000]
Merge git://git./linux/kernel/git/davem/net

Pull networking fixes from David Miller:

 1) Fix regression in /proc/net/if_inet6, sometimes devices do not get
    listed.  From Eric Dumazet.

 2) Add IPSEC networking sub-section to MAINTAINERS.

 3) S390 networking fixes from Hendrik Brueckner and Stefan Raspl.

 4) Fix enslavement of devices that can't do VLAN properly, from Jiri

 5) SCTP sack handling fix from Zijie Pan.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
  ipv6: addrconf: fix /proc/net/if_inet6
  bnx2x: fix handling mf storage modes
  qeth: fix deadlock between recovery and bonding driver
  smsgiucv: reestablish IUCV path after resume
  sctp: fix call to SCTP_CMD_PROCESS_SACK in sctp_cmd_interpreter()
  vlan: fix bond/team enslave of vlan challenged slave/port
  MAINTAINERS: Add explicit section for IPSEC networking.

7 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc
Linus Torvalds [Wed, 17 Oct 2012 19:40:44 +0000]
Merge git://git./linux/kernel/git/davem/sparc

Pull sparc fixes from David Miller:
 "Two sparc64 perf bug fixes and add a sysrq facility so I can diagnose
  these kinds of problems more quickly in the future."

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc:
  sparc64: Fix bit twiddling in sparc_pmu_enable_event().
  sparc64: Add global PMU register dumping via sysrq.
  sparc64: Like x86 we should check current->mm during perf backtrace generation.

7 years agomac80211: connect with HT20 if HT40 is not permitted
Johannes Berg [Mon, 1 Oct 2012 13:52:00 +0000]
mac80211: connect with HT20 if HT40 is not permitted

Some changes to fix issues with HT40 APs in Korea
and follow-up changes to allow using HT40 even if
the local regulatory database disallows it caused
issues with iwlwifi (and could cause issues with
other devices); iwlwifi firmware would assert if
you tried to connect to an AP that has an invalid
configuration (e.g. using HT40- on channel 140.)

Fix this, while avoiding the "Korean AP" issue by
disabling HT40 and advertising HT20 to the AP
when connecting.

Cc: stable@vger.kernel.org [3.6]
Reported-by: Florian Reitmeir <florian@reitmeir.org>
Tested-by: Florian Reitmeir <florian@reitmeir.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

7 years agoUAPI: Make arch/sparc/include/uapi/asm/sigcontext.h non-empty
David Howells [Wed, 17 Oct 2012 11:32:35 +0000]
UAPI: Make arch/sparc/include/uapi/asm/sigcontext.h non-empty

arch/sparc/include/uapi/asm/sigcontext.h was emitted by the UAPI disintegration
script as an empty file because the parent file had no UAPI stuff in it,
despite being marked with "header-y".

Unfortunately, the patch program deletes resultant empty files when applying a
kernel patch.

So just stick a comment in there as a placeholder.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: David S. Miller <davem@davemloft.net>
cc: sparclinux@vger.kernel.org

7 years agoUAPI: Make arch/sh/include/uapi/asm/hw_breakpoint.h non-empty
David Howells [Wed, 17 Oct 2012 11:32:35 +0000]
UAPI: Make arch/sh/include/uapi/asm/hw_breakpoint.h non-empty

arch/sh/include/uapi/asm/hw_breakpoint.h was emitted by the UAPI disintegration
script as an empty file because the parent file had no UAPI stuff in it,
despite being marked with "header-y".

Unfortunately, the patch program deletes resultant empty files when applying a
kernel patch.

So just stick a comment in there as a placeholder.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Paul Mundt <lethal@linux-sh.org>
cc: linux-sh@vger.kernel.org

7 years agoUAPI: Make arch/mn10300/include/uapi/asm/setup.h non-empty
David Howells [Wed, 17 Oct 2012 11:32:35 +0000]
UAPI: Make arch/mn10300/include/uapi/asm/setup.h non-empty

arch/mn10300/include/uapi/asm/setup.h was emitted by the UAPI disintegration
script as an empty file because the parent file had no UAPI stuff in it,
despite being marked with "header-y".

Unfortunately, the patch program deletes resultant empty files when applying a
kernel patch.

So just stick a comment in there as a placeholder.

Signed-off-by: David Howells <dhowells@redhat.com>

7 years agoUAPI: Put a comment into uapi/asm-generic/kvm_para.h and use it from arches
David Howells [Wed, 17 Oct 2012 11:32:07 +0000]
UAPI: Put a comment into uapi/asm-generic/kvm_para.h and use it from arches

Make uapi/asm-generic/kvm_para.h non-empty by addition of a comment to stop
the patch program from deleting it when it creates it.

Then delete empty arch-specific uapi/asm/kvm_para.h files and tell the Kbuild
files to use the generic instead.

Should this perhaps instead be a #warning or #error that the facility is
unsupported on this arch?

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Arnd Bergmann <arnd@arndb.de>
cc: Avi Kivity <avi@redhat.com>
cc: Marcelo Tosatti <mtosatti@redhat.com>
cc: kvm@vger.kernel.org

7 years agoUAPI: The tile arch uses the generic ucontext.h file
David Howells [Wed, 17 Oct 2012 11:31:16 +0000]
UAPI: The tile arch uses the generic ucontext.h file

Move the header-y and generic-y lines for ucontext.h from
arch/tile/include/asm/Kbuild to the uapi/ Kbuild as the asm-generic variant is

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Chris Metcalf <cmetcalf@tilera.com>

7 years agoUAPI: Place comments in empty arch Kbuilds to make them non-empty
David Howells [Wed, 17 Oct 2012 11:31:16 +0000]
UAPI: Place comments in empty arch Kbuilds to make them non-empty

Place comments in:


to make them non-empty so that the patch program doesn't remove them when it
reduces them to nothing.

Possibly they should be just deleted, but it's possible that they'll acquire
generic-y or genhdr-y lines in future, so I'm keeping them around for the

Note that MIPS will compile happily if the file is deleted instead.  I haven't
tested TILE, but I suspect it will be the same there.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Ralf Baechle <ralf@linux-mips.org>
cc: Chris Metcalf <cmetcalf@tilera.com>

7 years agoUAPI: Remove empty non-UAPI Kbuild files
David Howells [Wed, 17 Oct 2012 11:31:15 +0000]
UAPI: Remove empty non-UAPI Kbuild files

Remove non-UAPI Kbuild files that have become empty as a result of UAPI
disintegration.  They used to have only header-y lines in them and those have
now moved to the Kbuild files in the corresponding uapi/ directories.

Possibly these should not be removed but rather have a comment inserted to say
they are intentionally left blank.  This would make it easier to add generated
header lines in future without having to restore the infrastructure.

Note that at this point not all the UAPI disintegration parts have been merged,
so it is likely that more empty Kbuild files will turn up.

It is probably necessary to make the files non-empty to prevent the patch
program from automatically deleting them when it reduces them to nothing.

Signed-off-by: David Howells <dhowells@redhat.com>

7 years agoUAPI: Remove empty conditionals from include/linux/Kbuild
David Howells [Wed, 17 Oct 2012 11:31:15 +0000]
UAPI: Remove empty conditionals from include/linux/Kbuild

Remove empty conditionals from include/linux/Kbuild as the contents, with new
conditionals, have moved to include/uapi/linux/Kbuild.

Signed-off-by: David Howells <dhowells@redhat.com>

7 years agoUAPI: Make uapi/linux/irqnr.h non-empty
David Howells [Wed, 17 Oct 2012 11:31:15 +0000]
UAPI: Make uapi/linux/irqnr.h non-empty

uapi/linux/irqnr.h was emitted by the UAPI disintegration script as an empty
file because the parent linux/irqnr.h had no UAPI stuff in it, despite being
marked with "header-y".

Unfortunately, the patch program deletes the empty file when applying a kernel

It's not clear why this file is part of the UAPI at all.  Looking in:


there's nothing there but a header reinclusion guard and a comment.

So just stick a comment in there as a placeholder.

Without this, if the kernel is fabricated from, say, a tarball and a patch, you
can get this error when building x86_64 or usermode Linux (and probably

include/linux/irqnr.h:4:30: fatal error: uapi/linux/irqnr.h: No such file or directory

Reported-by: Randy Dunlap <rdunlap@xenotime.net>
Reported-by: Alessandro Suardi <alessandro.suardi@gmail.com>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Randy Dunlap <rdunlap@xenotime.net>
cc: Alessandro Suardi <alessandro.suardi@gmail.com>

7 years agouapi: Allow automatic generation of uapi/asm/ header files
Catalin Marinas [Wed, 17 Oct 2012 11:25:44 +0000]
uapi: Allow automatic generation of uapi/asm/ header files

Several arch/*/include/uapi/asm/* header simply include the
corresponding <asm-generic/*> file. This patch allows such files to be
specified in uapi/asm/Kbuild via "generic-y += ..." to be automatically
generated (similar to asm/Kbuild).

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Michal Marek <mmarek@suse.cz>
Cc: Arnd Bergmann <arnd@arndb.de>

7 years agonetfilter: xt_TEE: don't use destination address found in header
Eric Dumazet [Tue, 16 Oct 2012 22:33:29 +0000]
netfilter: xt_TEE: don't use destination address found in header

Torsten Luettgert bisected TEE regression starting with commit
f8126f1d5136be1 (ipv4: Adjust semantics of rt->rt_gateway.)

The problem is that it tries to ARP-lookup the original destination
address of the forwarded packet, not the address of the gateway.

Fix this using FLOWI_FLAG_KNOWN_NH Julian added in commit
c92b96553a80c1 (ipv4: Add FLOWI_FLAG_KNOWN_NH), so that known
nexthop (info->gw.ip) has preference on resolving.

Reported-by: Torsten Luettgert <ml-netfilter@enda.eu>
Bisected-by: Torsten Luettgert <ml-netfilter@enda.eu>
Tested-by: Torsten Luettgert <ml-netfilter@enda.eu>
Cc: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

7 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Pablo Neira Ayuso [Wed, 17 Oct 2012 08:59:20 +0000]
Merge branch 'master' of git://git./linux/kernel/git/davem/net

To obtain new flag FLOWI_FLAG_KNOWN_NH to fix netfilter's xt_TEE target.

7 years agoMerge tag 'sh-for-linus' of git://github.com/pmundt/linux-sh
Linus Torvalds [Wed, 17 Oct 2012 02:24:00 +0000]
Merge tag 'sh-for-linus' of git://github.com/pmundt/linux-sh

Pull SuperH updates from Paul Mundt.

The bulk of this is the UAPI disintegration for SH.

* tag 'sh-for-linus' of git://github.com/pmundt/linux-sh:
  sh: Fix up more fallout from pointless ARM __iomem churn.
  sh: Wire up kcmp syscall.
  UAPI: (Scripted) Disintegrate arch/sh/include/asm

7 years agoMerge branch 'frv' (FRV patches from David Howells)
Linus Torvalds [Wed, 17 Oct 2012 01:49:22 +0000]
Merge branch 'frv' (FRV patches from David Howells)

Merge emailed FRV fixes from David Howells.

* frv:
  FRV: Fix linux/elf-fdpic.h
  FRV: Fix const sections change
  FRV: Fix incorrect symbol in copy_thread()
  FRV: Fix VLIW packing constraint violation in entry.S

7 years agoFRV: Fix linux/elf-fdpic.h
David Howells [Mon, 15 Oct 2012 23:10:35 +0000]
FRV: Fix linux/elf-fdpic.h

It seems I accidentally switched the guard on linux/elf-fdpic.h from #ifdef
__KERNEL__ to #ifndef __KERNEL__ when attempting to expand the guarded region
to cover the elf_fdpic_params struct when doing the UAPI split - with the
result that the struct became unavailable to kernel code.

Move incorrectly guarded bits back to the kernelspace header.

Whilst we're at it, the __KERNEL__ guards can be deleted as they're no longer

Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Reported-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Fengguang Wu <fengguang.wu@intel.com>
cc: Lars-Peter Clausen <lars@metafoo.de>
cc: uclinux-dev@uclinux.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agoFRV: Fix const sections change
Andi Kleen [Mon, 15 Oct 2012 23:10:28 +0000]
FRV: Fix const sections change

Add __pminitconst to fix the build, otherwise the following error can occur:

  arch/frv/kernel/setup.c:187:47: error: expected '=', ',', ';', 'asm' or '__attribute__' before '*' token
  arch/frv/kernel/setup.c:386:2: error: 'clock_cmodes' undeclared (first use in this function)
  arch/frv/kernel/setup.c:571:6: error: 'clock_cmodes' undeclared (first use in this function)
  make[2]: *** [arch/frv/kernel/setup.o] Error 1


Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agoFRV: Fix incorrect symbol in copy_thread()
David Howells [Mon, 15 Oct 2012 23:10:21 +0000]
FRV: Fix incorrect symbol in copy_thread()

Fix an incorrect symbol in copy_thread():

  arch/frv/kernel/process.c: In function 'copy_thread':
  arch/frv/kernel/process.c:197: error: 'chilregs' undeclared (first use in this function)
  arch/frv/kernel/process.c:197: error: (Each undeclared identifier is reported only once
  arch/frv/kernel/process.c:197: error: for each function it appears in.)

Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Al Viro <viro@ZenIV.linux.org.uk>
cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agoFRV: Fix VLIW packing constraint violation in entry.S
David Howells [Mon, 15 Oct 2012 23:10:13 +0000]
FRV: Fix VLIW packing constraint violation in entry.S

Fix VLIW packing constraint violation in entry.S:

  arch/frv/kernel/entry.S: Assembler messages:
  arch/frv/kernel/entry.S:871: Error: VLIW packing constraint violation

When packing CALLL with OR, CALLL must go in the first slot.  The
instructions are executed simultaneously, so it doesn't matter which way
round they're packed from that point of view.

Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Al Viro <viro@ZenIV.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agoprintk: Fix scheduling-while-atomic problem in console_cpu_notify()
Paul E. McKenney [Tue, 16 Oct 2012 04:35:59 +0000]
printk: Fix scheduling-while-atomic problem in console_cpu_notify()

The console_cpu_notify() function runs with interrupts disabled in the
CPU_DYING case.  It therefore cannot block, for example, as will happen
when it calls console_lock().  Therefore, remove the CPU_DYING leg of
the switch statement to avoid this problem.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agoMerge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Linus Torvalds [Wed, 17 Oct 2012 01:14:31 +0000]
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux

Pull drm fixes from Dave Airlie:
 "Fixes for i915, nouveau and radeon:

   - i915: haswell stability, modeset rework fallout, ums fix
   - nouveau: misc fixes from code rework
   - radeon: pll rework fixes, more 2 level PTE cleanups.
   - core: warning fixes on 32-bit."

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux: (31 commits)
  nouveau: fix warning on 32-bit build.
  drm/nouveau/bios: fix typo in error message
  drm/nouveau: only call ttm_agp_tt_create when __OS_HAS_AGP
  drm/nv50/fb: fix double free of vram mm
  drm/nouveau/pm: do not stop reclocking if failing to set the fan speed
  drm/nouveau/pm: fix a typo related to the move to the therm subdev
  drm/nouveau/hwmon: fix the initialization condition
  drm: fix warning on 32-bit.
  drm: radeon: fix printk format warning
  drm/radeon: fix spelling typos in debugging output
  drm/radeon: Don't destroy I2C Bus Rec in radeon_ext_tmds_enc_destroy().
  drm/radeon: check if pcie gen 2 is already enabled (v2)
  drm/radeon/cayman: set VM max pfn at MC init
  drm/radeon: separate pt alloc from lru add
  drm/radeon: don't add the IB pool to all VMs v2
  drm/radeon: allocate page tables on demand v4
  drm/radeon: update comments to clarify VM setup (v2)
  drm/radeon: allocate PPLLs from low to high
  drm/radeon: fix compilation with backlight disabled
  drm/radeon: use %zu for formatting size_t

7 years agoMerge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs
Linus Torvalds [Wed, 17 Oct 2012 01:12:38 +0000]
Merge branch 'for_linus' of git://git./linux/kernel/git/jack/linux-fs

Pull ext2, ext3, quota fixes from Jan Kara:
 "Fix three regressions caused by user namespace conversions (ext2,
  ext3, quota) and minor ext3 fix and cleanup."

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
  quota: Silence warning about PRJQUOTA not being handled in need_print_warning()
  ext3: fix return values on parse_options() failure
  ext2: fix return values on parse_options() failure
  ext3: ext3_bread usage audit
  ext3: fix possible non-initialized variable on htree_dirblock_to_tree()

7 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Linus Torvalds [Wed, 17 Oct 2012 01:11:48 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/viro/vfs

Pull vfs fixes from Al Viro:
 "Fix for my braino in replace_fd(), dhowell's fix for the fallout from
  over-enthusiastic bo^Wdeclaration movements plus crapectomy that
  should've happened a long time ago (SEL_...  definitions)."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  bury SEL_{IN,OUT,EX}
  Unexport some bits of linux/fs.h
  fix a leak in replace_fd() users

7 years agoMerge tag 'pinctrl-fixes-v3.7-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Wed, 17 Oct 2012 01:10:26 +0000]
Merge tag 'pinctrl-fixes-v3.7-rc2' of git://git./linux/kernel/git/linusw/linux-pinctrl

Pull pinctrl fixes from Linus Walleij:
 "A number of pinctrl fixes for the v3.7 series:
   - duplicate includes, section markup, code mishaps
   - erroneous return value in errorpath on the bcm2835 driver
   - remove an unused sirf function that was causing build errors
   - multiple-platform compilation stubs and a missed code review
     comment fixup on the nomadik pin controller"

* tag 'pinctrl-fixes-v3.7-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:
  pinctrl/nomadik: always use the simple irqdomain
  pinctrl/nomadik: provide stubs for legacy Nomadik
  pinctrl: remove duplicated include from pinctrl-xway.c
  pinctrl: sirf: remove sirfsoc_gpio_set_pull function
  pinctrl: fix return value in bcm2835_pinctrl_probe()
  pinctrl: remove duplicated include from pinctrl-bcm2835.c
  pinctrl: bcm2835: Use existing pointer to struct device
  pinctrl: samsung: use __devinit section for init code

7 years agoMerge branch 'ipmi' (IPMI patches from Corey Minyard)
Linus Torvalds [Wed, 17 Oct 2012 01:08:21 +0000]
Merge branch 'ipmi' (IPMI patches from Corey Minyard)

Merge emailed patches from Corey Minyard:
 "Remove some bogus docs, Fix ACPI/IPMI interactions, fix some warnings,
  and add register spacing detection for PCI interfaces."

* ipmi:
  IPMI: Detect register spacing on PCI interfaces
  IPMI: Fix some uninitialized warning
  IPMI: Change link order
  ACPI: Reorder IPMI driver before any other ACPI drivers
  IPMI: Remove SMBus driver info from the docs

7 years agoIPMI: Detect register spacing on PCI interfaces
Corey Minyard [Tue, 16 Oct 2012 20:53:40 +0000]
IPMI: Detect register spacing on PCI interfaces

The IPMI spec defines a way to detect register spacing for PCI interfaces,
so implement it.

Signed-off-by: Steven Hsieh <sshsieh@broadcom.com>
Signed-off-by: Corey Minyard <cminyard@mvista.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agoIPMI: Fix some uninitialized warning
Corey Minyard [Tue, 16 Oct 2012 20:53:39 +0000]
IPMI: Fix some uninitialized warning

There was a spot where the compiler couldn't tell some variables
would be set.  So initialize them to make the warning go away.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agoIPMI: Change link order
Matthew Garrett [Tue, 16 Oct 2012 20:53:38 +0000]
IPMI: Change link order

IPMI must be initialised before ACPI in order to ensure that any IPMI
services are available before ACPI driver initialisation attempts to use
any IPMI operation regions.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Corey Minyard <cminyard@mvista.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agoACPI: Reorder IPMI driver before any other ACPI drivers
Matthew Garrett [Tue, 16 Oct 2012 20:53:37 +0000]
ACPI: Reorder IPMI driver before any other ACPI drivers

Drivers may make calls that require the ACPI IPMI driver to have been
initialised already, so make sure that it appears earlier in the build

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Corey Minyard <cminyard@mvista.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agoIPMI: Remove SMBus driver info from the docs
Corey Minyard [Tue, 16 Oct 2012 20:53:36 +0000]
IPMI: Remove SMBus driver info from the docs

Some documentation for the SMBus driver is in the IPMI docs, but that
code is not in the kernel tree at this point.  So remove the docs to
avoid confusion.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agoMAINTAINERS: change email after moving for LED subsystem maintaining
Bryan Wu [Tue, 16 Oct 2012 19:55:19 +0000]
MAINTAINERS: change email after moving for LED subsystem maintaining

Signed-off-by: Bryan Wu <cooloney@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agomm, mempolicy: fix printing stack contents in numa_maps
David Rientjes [Wed, 17 Oct 2012 00:31:23 +0000]
mm, mempolicy: fix printing stack contents in numa_maps

When reading /proc/pid/numa_maps, it's possible to return the contents of
the stack where the mempolicy string should be printed if the policy gets
freed from beneath us.

This happens because mpol_to_str() may return an error the
stack-allocated buffer is then printed without ever being stored.

There are two possible error conditions in mpol_to_str():

 - if the buffer allocated is insufficient for the string to be stored,

 - if the mempolicy has an invalid mode.

The first error condition is not triggered in any of the callers to
mpol_to_str(): at least 50 bytes is always allocated on the stack and this
is sufficient for the string to be written.  A future patch should convert
this into BUILD_BUG_ON() since we know the maximum strlen possible, but
that's not -rc material.

The second error condition is possible if a race occurs in dropping a
reference to a task's mempolicy causing it to be freed during the read().
The slab poison value is then used for the mode and mpol_to_str() returns

This race is only possible because get_vma_policy() believes that
mm->mmap_sem protects task->mempolicy, which isn't true.  The exit path
does not hold mm->mmap_sem when dropping the reference or setting
task->mempolicy to NULL: it uses task_lock(task) instead.

Thus, it's required for the caller of a task mempolicy to hold
task_lock(task) while grabbing the mempolicy and reading it.  Callers with
a vma policy store their mempolicy earlier and can simply increment the
reference count so it's guaranteed not to be freed.

Reported-by: Dave Jones <davej@redhat.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agosparc64: Fix bit twiddling in sparc_pmu_enable_event().
David S. Miller [Tue, 16 Oct 2012 20:05:25 +0000]
sparc64: Fix bit twiddling in sparc_pmu_enable_event().

There was a serious disconnect in the logic happening in
sparc_pmu_disable_event() vs. sparc_pmu_enable_event().

Event disable is implemented by programming a NOP event into the PCR.

However, event enable was not reversing this operation.  Instead, it
was setting the User/Priv/Hypervisor trace enable bits.

That's not sparc_pmu_enable_event()'s job, that's what
sparc_pmu_enable() and sparc_pmu_disable() do .

The intent of sparc_pmu_enable_event() is clear, since it first clear
out the event type encoding field.  So fix this by OR'ing in the event
encoding rather than the trace enable bits.

Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoipv6: addrconf: fix /proc/net/if_inet6
Eric Dumazet [Tue, 16 Oct 2012 07:37:27 +0000]
ipv6: addrconf: fix /proc/net/if_inet6

Commit 1d5783030a1 (ipv6/addrconf: speedup /proc/net/if_inet6 filling)
added bugs hiding some devices from if_inet6 and breaking applications.

"ip -6 addr" could still display all IPv6 addresses, while "ifconfig -a"

One way to reproduce the bug is by starting in a shell :

unshare -n /bin/bash
ifconfig lo up

And in original net namespace, lo device disappeared from if_inet6

Reported-by: Jan Hinnerk Stosch <janhinnerk.stosch@gmail.com>
Tested-by: Jan Hinnerk Stosch <janhinnerk.stosch@gmail.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Mihai Maruseac <mihai.maruseac@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agobnx2x: fix handling mf storage modes
Dmitry Kravkov [Tue, 16 Oct 2012 01:28:27 +0000]
bnx2x: fix handling mf storage modes

Since commit a3348722 AFEX FCoE function is continuously reset.
The patch prevents the resetting and removes debug print
to stop garbaging syslog.

Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoqeth: fix deadlock between recovery and bonding driver
Stefan Raspl [Mon, 15 Oct 2012 19:21:18 +0000]
qeth: fix deadlock between recovery and bonding driver

The recovery thread, when failing, tears down the respective interface. To do
so, it needs to obtain the rtnl lock first, as the interface configuration is
If another process tries to modify an interface setting at the same time, that
process can obtain the rtnl lock first, but the respective callback in the qeth
driver will block until recovery has completed - which cannot happen since the
calling process already obtained it.
In one particular case, the bonding driver acquired the rtnl lock to modify the
card's MAC address, while the recovery failed at the same time due to the card
being removed. Hence qeth_l2_set_mac_address (implicitly holding the rtnl lock)
was waiting on qeth_l2_recover, which deadlocked when waiting on the rtnl lock.
This patch uses rtnl_trylock instead of rtnl_lock in the recovery thread. If the
lock cannot be obtained, the interface will be left up, but the card state
remains in CARD_STATE_RECOVER, which will prevent any further activities on the

Signed-off-by: Stefan Raspl <raspl@linux.vnet.ibm.com>
Signed-off-by: Frank Blaschka <frank.blaschka@de.ibm.com>
Reviewed-by: Ursula Braun <ursula.braun@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agosmsgiucv: reestablish IUCV path after resume
Hendrik Brueckner [Mon, 15 Oct 2012 19:21:17 +0000]
smsgiucv: reestablish IUCV path after resume

smsg_pm_restore_thaw() uses wrong checking before reconnecting
the IUCV path to *MSG. It is corrected with this patch.

Signed-off-by: Ursula Braun <ursula.braun@de.ibm.com>
Signed-off-by: Frank Blaschka <frank.blaschka@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agosctp: fix call to SCTP_CMD_PROCESS_SACK in sctp_cmd_interpreter()
Zijie Pan [Mon, 15 Oct 2012 03:56:39 +0000]
sctp: fix call to SCTP_CMD_PROCESS_SACK in sctp_cmd_interpreter()

Bug introduced by commit edfee0339e681a784ebacec7e8c2dc97dc6d2839
(sctp: check src addr when processing SACK to update transport state)

Signed-off-by: Zijie Pan <zijie.pan@6wind.com>
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agovlan: fix bond/team enslave of vlan challenged slave/port
Jiri Pirko [Sun, 14 Oct 2012 04:30:56 +0000]
vlan: fix bond/team enslave of vlan challenged slave/port

In vlan_uses_dev() check for number of vlan devs rather than existence
of vlan_info. The reason is that vlan id 0 is there without appropriate
vlan dev on it by default which prevented from enslaving vlan challenged

Reported-by: Jon Stanley <jstanley@rmrf.net>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoMAINTAINERS: Add explicit section for IPSEC networking.
David S. Miller [Tue, 16 Oct 2012 18:08:40 +0000]
MAINTAINERS: Add explicit section for IPSEC networking.

Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agobury SEL_{IN,OUT,EX}
Al Viro [Tue, 16 Oct 2012 17:37:17 +0000]
bury SEL_{IN,OUT,EX}

Had not been used for more than a decade and half; it used
to be a part of (in-kernel) ->select() API and it has been pining
for fjords since 2.1.23pre1.  This is an ex-parrot...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

7 years agoUnexport some bits of linux/fs.h
David Howells [Mon, 15 Oct 2012 15:40:35 +0000]
Unexport some bits of linux/fs.h

There are some bits of linux/fs.h which are only used within the kernel and
shouldn't be in the UAPI.  Move these from uapi/linux/fs.h into linux/fs.h.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

7 years agofix a leak in replace_fd() users
Al Viro [Tue, 16 Oct 2012 17:30:07 +0000]
fix a leak in replace_fd() users

replace_fd() began with "eats a reference, tries to insert into
descriptor table" semantics; at some point I'd switched it to
much saner current behaviour ("try to insert into descriptor
table, grabbing a new reference if inserted; caller should do
fput() in any case"), but forgot to update the callers.
Mea culpa...

[Spotted by Pavel Roskin, who has really weird system with pipe-fed
coredumps as part of what he considers a normal boot ;-)]

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

7 years agosparc64: Add global PMU register dumping via sysrq.
David S. Miller [Tue, 16 Oct 2012 16:34:01 +0000]
sparc64: Add global PMU register dumping via sysrq.

Signed-off-by: David S. Miller <davem@davemloft.net>

7 years agoiwlwifi: fix 6000 series channel switch command
Johannes Berg [Tue, 25 Sep 2012 14:40:12 +0000]
iwlwifi: fix 6000 series channel switch command

The channel switch command for 6000 series devices
is larger than the maximum inline command size of
320 bytes. The command is therefore refused with a
warning. Fix this by allocating the command and
using the NOCOPY mechanism.

Cc: stable@kernel.org
Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

7 years agonouveau: fix warning on 32-bit build.
Dave Airlie [Tue, 16 Oct 2012 06:40:53 +0000]
nouveau: fix warning on 32-bit build.

Signed-off-by: Dave Airlie <airlied@redhat.com>

7 years agoMerge branch 'drm-nouveau-fixes' of git://anongit.freedesktop.org/git/nouveau/linux...
Dave Airlie [Tue, 16 Oct 2012 06:33:48 +0000]
Merge branch 'drm-nouveau-fixes' of git://anongit.freedesktop.org/git/nouveau/linux-2.6 into drm-fixes

minor set of nouveau fixes.

* 'drm-nouveau-fixes' of git://anongit.freedesktop.org/git/nouveau/linux-2.6:
  drm/nouveau/bios: fix typo in error message
  drm/nouveau: only call ttm_agp_tt_create when __OS_HAS_AGP
  drm/nv50/fb: fix double free of vram mm
  drm/nouveau/pm: do not stop reclocking if failing to set the fan speed
  drm/nouveau/pm: fix a typo related to the move to the therm subdev
  drm/nouveau/hwmon: fix the initialization condition

7 years agodrm/nouveau/bios: fix typo in error message
Ben Skeggs [Tue, 16 Oct 2012 06:25:08 +0000]
drm/nouveau/bios: fix typo in error message

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

7 years agodrm/nouveau: only call ttm_agp_tt_create when __OS_HAS_AGP
Max Filippov [Sat, 13 Oct 2012 21:58:26 +0000]
drm/nouveau: only call ttm_agp_tt_create when __OS_HAS_AGP

ttm_agp_tt_create is itself defined under CONFIG_AGP, so there's no
point calling it otherwise.

Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

7 years agodrm/nv50/fb: fix double free of vram mm
Marcin Slusarz [Thu, 11 Oct 2012 21:53:48 +0000]
drm/nv50/fb: fix double free of vram mm

nouveau_fb_destroy already calls nouveau_mm_fini on vram mm.

Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

7 years agodrm/nouveau/pm: do not stop reclocking if failing to set the fan speed
Martin Peres [Wed, 3 Oct 2012 23:00:13 +0000]
drm/nouveau/pm: do not stop reclocking if failing to set the fan speed

With the introduction of fan management modes, fan may not be drivable.
We should allow reclocking nonetheless.

This return was stupid to begin with since it may have left the card
in an intermediate state (clocks corresponding to a perflvl and voltage
corresponding to another one). The reclocking code will need to be
rewritten in a near-future in order to provide a better error handling.

Reported-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: Martin Peres <martin.peres@labri.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

7 years agodrm/nouveau/pm: fix a typo related to the move to the therm subdev
Martin Peres [Wed, 3 Oct 2012 22:44:19 +0000]
drm/nouveau/pm: fix a typo related to the move to the therm subdev

Reported-by: Vekin on IRC
Reported-by: Emil Velikov <emil.l.velikov@gmail.com>
Signed-off-by: Martin Peres <martin.peres@labri.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

7 years agodrm/nouveau/hwmon: fix the initialization condition
Martin Peres [Wed, 3 Oct 2012 22:28:21 +0000]
drm/nouveau/hwmon: fix the initialization condition

Signed-off-by: Martin Peres <martin.peres@labri.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

7 years agodrm: fix warning on 32-bit.
Dave Airlie [Tue, 16 Oct 2012 00:28:21 +0000]
drm: fix warning on 32-bit.

This cast was causing a warning on 32-bit builds.

Signed-off-by: Dave Airlie <airlied@redhat.com>

7 years agodrm: radeon: fix printk format warning
Randy Dunlap [Tue, 16 Oct 2012 00:15:45 +0000]
drm: radeon: fix printk format warning

drivers/gpu/drm/radeon/radeon_atpx_handler.c:151:3: warning: format '%lu' expects type 'long unsigned int', but argument 2 has type 'size_t'

[airlied: Alex had others fixed already, except for atpx one]

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Dave Airlie <airlied@redhat.com>

7 years agoMerge branch 'drm-fixes-3.7' of git://people.freedesktop.org/~agd5f/linux into drm...
Dave Airlie [Tue, 16 Oct 2012 00:13:24 +0000]
Merge branch 'drm-fixes-3.7' of git://people.freedesktop.org/~agd5f/linux into drm-fixes

Alex writes:
"This is the first -fixes pull for 3.7.  I would have preferred
to have gotten it out a bit sooner, but I was on holiday last week.
- Cleanup of the new 2 level page table code it get it in
  better shape and using less memory.
- Fix some display issues related to the PLL rework.
- Fix some cmpiler warnings and errors with certain config
- Other misc bug fixes."

* 'drm-fixes-3.7' of git://people.freedesktop.org/~agd5f/linux:
  drm/radeon: fix spelling typos in debugging output
  drm/radeon: Don't destroy I2C Bus Rec in radeon_ext_tmds_enc_destroy().
  drm/radeon: check if pcie gen 2 is already enabled (v2)
  drm/radeon/cayman: set VM max pfn at MC init
  drm/radeon: separate pt alloc from lru add
  drm/radeon: don't add the IB pool to all VMs v2
  drm/radeon: allocate page tables on demand v4
  drm/radeon: update comments to clarify VM setup (v2)
  drm/radeon: allocate PPLLs from low to high
  drm/radeon: fix compilation with backlight disabled
  drm/radeon: use %zu for formatting size_t

7 years agoMerge branch 'drm-intel-fixes' of git://people.freedesktop.org/~danvet/drm-intel...
Dave Airlie [Tue, 16 Oct 2012 00:11:29 +0000]
Merge branch 'drm-intel-fixes' of git://people.freedesktop.org/~danvet/drm-intel into drm-fixes

Daniel writes:
"- some register magic to fix hsw crw (Paulo&Ben)
- fix backlight destruction for cpu edp (Jani)
- fix gen ch7xxx dvo ->get_hw_state
- fixup the plane->pipe fixup code, the broken version massively angers
  the modeset sanity checks
- kill pipe A quirk for i855gm, otherwise I get a black screen with the
  above patch
- fixup for gem_get_page helper (Chris)
- fixup guardband clipping w/a (Ken), without this mesa master can erronously
  drop vertices on snb, mesa 9.0 has the optimization reverted
- another pageflip vs. modeset fix
- kill bogus BUG_ON which broke ums+gem from Willy Tarreau (gasp, people
  are still using this!)"

* 'drm-intel-fixes' of git://people.freedesktop.org/~danvet/drm-intel:
  drm/i915: fix non-DP-D eDP backlight cleanup and module reload
  drm/i915: HSW CRW stability magic
  drm/i915/dvo-ch7xxx: fix get_hw_state
  drm/i915: fixup the plane->pipe fixup code
  drm/i915: rip out the pipe A quirk for i855gm
  drm/i915: disable wc gtt pte mappings on gen2
  drm/i915: fixup i915_gem_object_get_page inline helper
  drm/i915: Disallow preallocation of requests
  drm/i915: Set guardband clipping workaround bit in the right register.
  drm/i915: paper over a pipe-enable vs pageflip race
  drm/i915: remove useless BUG_ON which caused a regression in 3.5.

7 years agothermal, cpufreq: Fix build when CPU_FREQ_TABLE isn't configured
David Rientjes [Mon, 15 Oct 2012 20:40:15 +0000]
thermal, cpufreq: Fix build when CPU_FREQ_TABLE isn't configured

Commit 023614183768 ("thermal: add generic cpufreq cooling
implementation") requires cpufreq_frequency_get_table(), but that
function is only defined for CONFIG_CPU_FREQ_TABLE resulting in the
following build error:

  drivers/built-in.o: In function `cpufreq_get_max_state':
  drivers/thermal/cpu_cooling.c:259: undefined reference to `cpufreq_frequency_get_table'
  drivers/built-in.o: In function `get_cpu_frequency':
  drivers/thermal/cpu_cooling.c:129: undefined reference to `cpufreq_frequency_get_table'

Fix it by selecting CONFIG_CPU_FREQ_TABLE for such a configuration.

select it there as well.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agokbuild: Fix accidental revert in commit fe04ddf
Michal Marek [Mon, 15 Oct 2012 19:16:56 +0000]
kbuild: Fix accidental revert in commit fe04ddf

Commit fe04ddf7c291 ("kbuild: Do not package /boot and /lib in make
tar-pkg") accidentally reverted two previous kbuild commits.  I don't
know what I was thinking.

This brings back changes made by commits 24cc7fb69a5b ("x86/kbuild:
archscripts depends on scripts_basic") and c1c1a59e37da ("firmware: fix
directory creation rule matching with make 3.80")

Reported-by: Jan Beulich <JBeulich@suse.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Michal Marek <mmarek@suse.cz>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

7 years agonet/wireless: ipw2200: Fix panic occurring in ipw_handle_promiscuous_tx()
Stanislav Yakovlev [Mon, 15 Oct 2012 14:14:32 +0000]
net/wireless: ipw2200: Fix panic occurring in ipw_handle_promiscuous_tx()

The driver does not count space of radiotap fields when allocating skb for
radiotap packet. This leads to kernel panic with the following call trace:

[67607.676067] [<c152f90f>] error_code+0x67/0x6c
[67607.676067] [<c142f831>] ? skb_put+0x91/0xa0
[67607.676067] [<f8cf5e5b>] ? ipw_handle_promiscuous_tx+0x16b/0x2d0 [ipw2200]
[67607.676067] [<f8cf5e5b>] ipw_handle_promiscuous_tx+0x16b/0x2d0 [ipw2200]
[67607.676067] [<f8cf899b>] ipw_net_hard_start_xmit+0x8b/0x90 [ipw2200]
[67607.676067] [<f8741c5a>] libipw_xmit+0x55a/0x980 [libipw]
[67607.676067] [<c143d3e8>] dev_hard_start_xmit+0x218/0x4d0

This bug was found by VittGam.

Cc: stable@kernel.org
Signed-off-by: Stanislav Yakovlev <stas.yakovlev@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

7 years agobcma: fix unregistration of cores
Piotr Haber [Thu, 11 Oct 2012 12:05:15 +0000]
bcma: fix unregistration of cores

When cores are unregistered, entries
need to be removed from cores list in a safe manner.

Reported-by: Stanislaw Gruszka <sgruszka@redhat.com>
Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Signed-off-by: Piotr Haber <phaber@broadcom.com>
Cc: stable@vger.kernel.org
Signed-off-by: John W. Linville <linville@tuxdriver.com>