linux.git
6 years agonet:ethernet:cadence: Replace dev_kfree_skb_any by dev_consume_skb_any
Huang Zijiang [Thu, 14 Feb 2019 06:41:18 +0000 (14:41 +0800)]
net:ethernet:cadence: Replace dev_kfree_skb_any by dev_consume_skb_any

The skb should be freed by dev_consume_skb_any() in macb_pad_and_fcs()
when *skb is still used. The *skb is be replaced by nskb, so the
original *skb should be consumed(not drop).

Signed-off-by: Huang Zijiang <huang.zijiang@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet:dl2k: Replace dev_kfree_skb_irq by dev_consume_skb_irq
Huang Zijiang [Thu, 14 Feb 2019 06:40:56 +0000 (14:40 +0800)]
net:dl2k: Replace dev_kfree_skb_irq by dev_consume_skb_irq

dev_consume_skb_irq() should be called when skb xmit
done.It makes drop profiles more friendly.

Signed-off-by: Huang Zijiang <huang.zijiang@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet:dl2k: Modify the code style escaping the warning
Huang Zijiang [Thu, 14 Feb 2019 06:40:31 +0000 (14:40 +0800)]
net:dl2k: Modify the code style escaping the warning

modify the code style in order to removing the following warning
when excute the script checkpatch.pl
WARNING: space prohibited between function name and open parenthesis '('

Signed-off-by: Huang Zijiang <huang.zijiang@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoisdn:hisax: Replace dev_kfree_skb_any by dev_consume_skb_any
Huang Zijiang [Thu, 14 Feb 2019 06:39:59 +0000 (14:39 +0800)]
isdn:hisax: Replace dev_kfree_skb_any by dev_consume_skb_any

The skb should be freed by dev_consume_skb_any() in hfcpci_fill_fifo()
when bcs->tx_skb is still used. The bcs->tx_skb is be replaced by
skb_dequeue(&bcs->squeue), so the original bcs->tx_skb should
be consumed(not drop).

Signed-off-by: Huang Zijiang <huang.zijiang@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomm: page_alloc: fix ref bias in page_frag_alloc() for 1-byte allocs
Jann Horn [Wed, 13 Feb 2019 21:45:59 +0000 (22:45 +0100)]
mm: page_alloc: fix ref bias in page_frag_alloc() for 1-byte allocs

The basic idea behind ->pagecnt_bias is: If we pre-allocate the maximum
number of references that we might need to create in the fastpath later,
the bump-allocation fastpath only has to modify the non-atomic bias value
that tracks the number of extra references we hold instead of the atomic
refcount. The maximum number of allocations we can serve (under the
assumption that no allocation is made with size 0) is nc->size, so that's
the bias used.

However, even when all memory in the allocation has been given away, a
reference to the page is still held; and in the `offset < 0` slowpath, the
page may be reused if everyone else has dropped their references.
This means that the necessary number of references is actually
`nc->size+1`.

Luckily, from a quick grep, it looks like the only path that can call
page_frag_alloc(fragsz=1) is TAP with the IFF_NAPI_FRAGS flag, which
requires CAP_NET_ADMIN in the init namespace and is only intended to be
used for kernel testing and fuzzing.

To test for this issue, put a `WARN_ON(page_ref_count(page) == 0)` in the
`offset < 0` path, below the virt_to_page() call, and then repeatedly call
writev() on a TAP device with IFF_TAP|IFF_NO_PI|IFF_NAPI_FRAGS|IFF_NAPI,
with a vector consisting of 15 elements containing 1 byte each.

Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge branch 'net-phy-fix-locking-issue'
David S. Miller [Thu, 14 Feb 2019 17:04:55 +0000 (12:04 -0500)]
Merge branch 'net-phy-fix-locking-issue'

Heiner Kallweit says:

====================
net: phy: fix locking issue

Russell pointed out that the locking used in phy_is_started() isn't
needed and misleading. This locking also contributes to a race fixed
with patch 2.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: phy: fix potential race in the phylib state machine
Heiner Kallweit [Wed, 13 Feb 2019 19:12:54 +0000 (20:12 +0100)]
net: phy: fix potential race in the phylib state machine

Russell reported the following race in the phylib state machine
(quoting from his mail):

if (phy_polling_mode(phydev) && phy_is_started(phydev))
phy_queue_state_machine(phydev, PHY_STATE_TIME);

state = PHY_UP
thread 0 thread 1
phy_disconnect()
+-phy_is_started()
phy_is_started()                |
`-phy_stop()
  +-phydev->state = PHY_HALTED
  `-phy_stop_machine()
    `-cancel_delayed_work_sync()
phy_queue_state_machine()
`-mod_delayed_work()

At this point, the phydev->state_queue() has been added back onto the
system workqueue despite phy_stop_machine() having been called and
cancel_delayed_work_sync() called on it.

Fix this by protecting the complete operation in thread 0.

Fixes: 2b3e88ea6528 ("net: phy: improve phy state checking")
Reported-by: Russell King - ARM Linux admin <linux@armlinux.org.uk>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: phy: don't use locking in phy_is_started
Heiner Kallweit [Wed, 13 Feb 2019 19:11:40 +0000 (20:11 +0100)]
net: phy: don't use locking in phy_is_started

Russell suggested to remove the locking from phy_is_started() because
the read is atomic anyway and actually the locking may be more
misleading.

Fixes: 2b3e88ea6528 ("net: phy: improve phy state checking")
Suggested-by: Russell King - ARM Linux admin <linux@armlinux.org.uk>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoselftests: fix timestamping Makefile
Deepa Dinamani [Wed, 13 Feb 2019 17:09:13 +0000 (09:09 -0800)]
selftests: fix timestamping Makefile

The clean target in the makefile conflicts with the generic
kselftests lib.mk, and fails to properly remove the compiled
test programs.

Remove the redundant rule, the TEST_GEN_FILES will be already
removed by the CLEAN macro in lib.mk.

Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com>
Acked-by: Shuah Khan <shuah@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: ipvlan_l3s: fix kconfig dependency warning
Randy Dunlap [Wed, 13 Feb 2019 16:55:02 +0000 (08:55 -0800)]
net: ipvlan_l3s: fix kconfig dependency warning

Fix the kconfig warning in IPVLAN_L3S when neither INET nor IPV6
is enabled:

WARNING: unmet direct dependencies detected for NET_L3_MASTER_DEV
  Depends on [n]: NET [=y] && (INET [=n] || IPV6 [=n])
  Selected by [y]:
  - IPVLAN_L3S [=y] && NETDEVICES [=y] && NET_CORE [=y] && NETFILTER [=y]

Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Mahesh Bandewar <maheshb@google.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: nuvoton: w90p910_ether: replace dev_kfree_skb_irq by dev_consume_skb_irq for...
Yang Wei [Wed, 13 Feb 2019 15:21:02 +0000 (23:21 +0800)]
net: nuvoton: w90p910_ether: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in w90p910_ether_start_xmit()
when skb xmit done. It makes drop profiles(dropwatch, perf) more
friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: natsemi: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Wed, 13 Feb 2019 15:19:14 +0000 (23:19 +0800)]
net: natsemi: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called when skb xmit done. It makes
drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: micrel: ks8695net: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop...
Yang Wei [Wed, 13 Feb 2019 15:18:09 +0000 (23:18 +0800)]
net: micrel: ks8695net: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in ks8695_tx_irq() when skb
xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: sgi: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Wed, 13 Feb 2019 15:17:06 +0000 (23:17 +0800)]
net: sgi: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called when skb xmit done. It makes
drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: myri10ge: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Wed, 13 Feb 2019 15:15:43 +0000 (23:15 +0800)]
net: myri10ge: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in myri10ge_tx_done() when
skb xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: amd: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Wed, 13 Feb 2019 15:14:54 +0000 (23:14 +0800)]
net: amd: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called when skb xmit done. It makes
drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: dlink: sundance: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Wed, 13 Feb 2019 15:12:02 +0000 (23:12 +0800)]
net: dlink: sundance: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in intr_handler() when skb
xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Remove a redundant blank line in intr_handler().

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge branch 'uapi-Add-a-new-header-for-time-types'
David S. Miller [Thu, 14 Feb 2019 16:51:51 +0000 (11:51 -0500)]
Merge branch 'uapi-Add-a-new-header-for-time-types'

Deepa Dinamani says:

====================
uapi: Add a new header for time types

The series aims at adding a new time header: time_types.h.  This header
is what will eventually hold all the uapi time types that we plan to
leave across the interfaces after the y2038 cleanup.

The series was discussed with Arnd Bergmann.

The second patch fixes the errqueue.h header, which has a dependency on
these types.

Note that there may be a trivial merge conflict with linux-next
c70a772fda11 ("y2038: remove struct definition redirects").
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoerrqueue.h: Include time_types.h
Deepa Dinamani [Wed, 13 Feb 2019 03:26:04 +0000 (19:26 -0800)]
errqueue.h: Include time_types.h

Now that we have a separate header for struct __kernel_timespec,
include it directly without relying on userspace to do it.

Reported-by: Ran Rozenstein <ranro@mellanox.com>
Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agotime: Add time_types.h
Deepa Dinamani [Wed, 13 Feb 2019 03:26:03 +0000 (19:26 -0800)]
time: Add time_types.h

sys/time.h is the mandated include for many time related
defines. However, linux/time.h overlaps sys/time.h
significantly and this makes including both from userspace
or one from the other impossible.

This also means that userspace can get away with including
sys/time.h whenever it needs linux/time.h and this is what's
been happening in the user world usually.

But, we have new data types that we plan to use in the uapi time
interfaces also defined in the linux/time.h. But, we are unable
to use these types when sys/time.h is included.

Hence, move the new types to a new header, time_types.h.
We intend to eventually have all the uapi defines that the kernel
uses defined in this header.
Note that the plan is to replace uapi interfaces with timeval to
use __kernel_old_timeval, timespec to use __kernel_old_timespec etc.

Reported-by: Ran Rozenstein <ranro@mellanox.com>
Fixes: 9718475e6908 ("socket: Add SO_TIMESTAMPING_NEW")
Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge branch 'devlink-region-read-fixes'
David S. Miller [Thu, 14 Feb 2019 16:45:39 +0000 (11:45 -0500)]
Merge branch 'devlink-region-read-fixes'

Parav Pandit says:

====================
devlink: 2 fixes for devlink region read

This 2 patches consist of fixes for devlink region read handling.

v0->v1:
 - Fixed typo from user to use
v1->v2:
 - Rebased
====================

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agodevlink: Fix list access without lock while reading region
Parav Pandit [Tue, 12 Feb 2019 20:24:08 +0000 (14:24 -0600)]
devlink: Fix list access without lock while reading region

While finding the devlink device during region reading,
devlink device list is accessed and devlink device is
returned without holding a lock. This could lead to use-after-free
accesses.

While at it, add lockdep assert to ensure that all future callers hold
the lock when calling devlink_get_from_attrs().

Fixes: 4e54795a27f5 ("devlink: Add support for region snapshot read command")
Signed-off-by: Parav Pandit <parav@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agodevlink: Return right error code in case of errors for region read
Parav Pandit [Tue, 12 Feb 2019 20:23:58 +0000 (14:23 -0600)]
devlink: Return right error code in case of errors for region read

devlink_nl_cmd_region_read_dumpit() misses to return right error code on
most error conditions.
Return the right error code on such errors.

Fixes: 4e54795a27f5 ("devlink: Add support for region snapshot read command")
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agobonding: check slave set command firstly
Tonghao Zhang [Mon, 11 Feb 2019 18:49:48 +0000 (10:49 -0800)]
bonding: check slave set command firstly

This patch is a little improvement. If user use the
command shown as below, we should print the info [1]
instead of [2]. The eth0 exists actually, and it may
confuse user.

$ echo "eth0" > /sys/class/net/bond4/bonding/slaves

[1] "bond4: no command found in slaves file - use +ifname or -ifname"
[2] "write error: No such device"

Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge branch 'mlxsw-hwmon-and-thermal-extensions'
David S. Miller [Thu, 14 Feb 2019 06:33:03 +0000 (22:33 -0800)]
Merge branch 'mlxsw-hwmon-and-thermal-extensions'

Ido Schimmel says:

====================
mlxsw: hwmon and thermal extensions

Vadim says:

This patchset contains various improvements to hwmon and thermal code in
mlxsw. The most significant improvement is the ability to read modules'
temperature attributes (input, fault, critical and emergency thresholds)
as well as fans' fault indication. These new attributes will improve the
ability to monitor the system.

Patches #1-#4 add the necessary device registers and APIs to read
modules' temperature attributes and fans' fault indication.

Patches #5-#8 perform small improvements in hwmon and thermal code such
as using a more indicative name for cooling devices.

Patch #9 exposes fans' fault indication via hwmon.

Patch #10 exposes modules' temperature attributes via hwmon.

Patch #11 adds an hwmon label to modules' temperature sensor. This helps
to parse the output of utilities such as "sensors".

Patch #12 allows to bind an external cooling device ("mlxreg-fan") to
mlxsw thermal zone. This will allow the mlxsw thermal zone to change the
cooling level of cooling devices not programmed via switch registers.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: core: Allow thermal zone binding to an external cooling device
Vadim Pasternak [Wed, 13 Feb 2019 11:28:56 +0000 (11:28 +0000)]
mlxsw: core: Allow thermal zone binding to an external cooling device

Allow thermal zone binding to an external cooling device from the
cooling devices white list.

It provides support for Mellanox next generation systems on which
cooling device logic is not controlled through the switch registers.

Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: core: Add QSFP module temperature label attribute to hwmon
Vadim Pasternak [Wed, 13 Feb 2019 11:28:55 +0000 (11:28 +0000)]
mlxsw: core: Add QSFP module temperature label attribute to hwmon

Add label attribute to hwmon object for exposing QSFP module's
temperature sensor name. Modules are labeled as "front panel xxx". The
label is used by utilities such as "sensors":

front panel 001:   +0.0C  (crit =  +0.0C, emerg =  +0.0C)
..
front panel 020:  +31.0C  (crit = +70.0C, emerg = +80.0C)
..
front panel 056:  +41.0C  (crit = +70.0C, emerg = +80.0C)

Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: core: Extend hwmon interface with QSFP module temperature attributes
Vadim Pasternak [Wed, 13 Feb 2019 11:28:54 +0000 (11:28 +0000)]
mlxsw: core: Extend hwmon interface with QSFP module temperature attributes

Add new attributes to hwmon object for exposing QSFP module temperature
input, fault indication, critical and emergency thresholds. Temperature
input and fault indication are read from Management Temperature Bulk
Register. Temperature thresholds are read from Management Cable Info
Access Register.

Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: core: Extend hwmon interface with fan fault attribute
Vadim Pasternak [Wed, 13 Feb 2019 11:28:53 +0000 (11:28 +0000)]
mlxsw: core: Extend hwmon interface with fan fault attribute

Add new fan hwmon attribute for exposing fan faults (fault indication is
read from Fan Out of Range Event Register).

Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: core: Rename cooling device
Vadim Pasternak [Wed, 13 Feb 2019 11:28:52 +0000 (11:28 +0000)]
mlxsw: core: Rename cooling device

Rename cooling device from "Fan" to "mlxsw_fan".  Name "Fan" is too
common name, and such name is misleading, while it's interpreted by
user. For example name "Fan" could be used by ACPI.

Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: core: Replace thermal temperature trips with defines
Vadim Pasternak [Wed, 13 Feb 2019 11:28:51 +0000 (11:28 +0000)]
mlxsw: core: Replace thermal temperature trips with defines

Replace thermal hardcoded temperature trip values with defines.

Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: core: Modify thermal zone definition
Vadim Pasternak [Wed, 13 Feb 2019 11:28:50 +0000 (11:28 +0000)]
mlxsw: core: Modify thermal zone definition

Modify thermal zone trip points setting for better alignment with system
thermal requirement.

Add hysteresis thresholds for thermal trips in order to avoid throttling
around thermal trip point. If hysteresis temperature is not considered,
PWM can have side effect of flip up/down on thermal trip point boundary.

Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: core: Set different thermal polling time based on bus frequency capability
Vadim Pasternak [Wed, 13 Feb 2019 11:28:48 +0000 (11:28 +0000)]
mlxsw: core: Set different thermal polling time based on bus frequency capability

Add low frequency bus capability in order to allow core functionality
separation based on bus type. Driver could run over PCIe, which is
considered as high frequency bus or I2C, which is considered as low
frequency bus. In the last case time setting, for example, for thermal
polling interval, should be increased.

Use different thermal monitoring based on bus type. For I2C bus time is
set to 20 seconds, while for PCIe 1 second polling interval is used.

Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: core: Add API for QSFP module temperature thresholds reading
Vadim Pasternak [Wed, 13 Feb 2019 11:28:47 +0000 (11:28 +0000)]
mlxsw: core: Add API for QSFP module temperature thresholds reading

Add new API to read QSFP module's temperature thresholds - warning and
critical.

New internal API reads the temperature thresholds from the modules,
which are equipped with the thermal sensor. These thresholds will be
exposed via hwmon subsystem.

Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: reg: Add Fan Out of Range Event Register
Vadim Pasternak [Wed, 13 Feb 2019 11:28:46 +0000 (11:28 +0000)]
mlxsw: reg: Add Fan Out of Range Event Register

Add FORE (Fan Out of Range Event Register), which is used for fan fault
reading.

Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: reg: Add Management Temperature Bulk Register
Vadim Pasternak [Wed, 13 Feb 2019 11:28:45 +0000 (11:28 +0000)]
mlxsw: reg: Add Management Temperature Bulk Register

Add MTBR (Management Temperature Bulk Register), which is used for port
temperature reading in a bulk mode.

Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: spectrum: Move QSFP EEPROM definitions to common location
Vadim Pasternak [Wed, 13 Feb 2019 11:28:44 +0000 (11:28 +0000)]
mlxsw: spectrum: Move QSFP EEPROM definitions to common location

Move QSFP EEPROM definitions to common location from the spectrum driver
in order to make them available for other mlxsw modules. They are common
for all kind of chips and have relation to SFF specifications 8024,
8436, 8472, 8636, rather than to chip type.

Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge tag 'batadv-next-for-davem-20190213' of git://git.open-mesh.org/linux-merge
David S. Miller [Thu, 14 Feb 2019 06:28:11 +0000 (22:28 -0800)]
Merge tag 'batadv-next-for-davem-20190213' of git://git.open-mesh.org/linux-merge

Simon Wunderlich says:

====================
This feature/cleanup patchset includes the following patches:

 - fix memory leak in in batadv_dat_put_dhcp, by Martin Weinelt

 - fix typo, by Sven Eckelmann

 - netlink restructuring patch series (part 2), by Sven Eckelmann
   (19 patches)
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agotest_objagg: Uninitialized variable in error handling
Dan Carpenter [Wed, 13 Feb 2019 08:59:31 +0000 (11:59 +0300)]
test_objagg: Uninitialized variable in error handling

We need to set the error message on this path otherwise some of the
callers, such as test_hints_case(), print from an uninitialized pointer.

We had a similar bug earlier and set "errmsg" to NULL in the caller,
test_delta_action_item().  That code is no longer required so I have
removed it.

Fixes: 9069a3817d82 ("lib: objagg: implement optimization hints assembly and use hints for object creation")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agotest_objagg: Test the correct variable
Dan Carpenter [Wed, 13 Feb 2019 08:58:20 +0000 (11:58 +0300)]
test_objagg: Test the correct variable

There is a typo here.  We intended to check "objagg2" but we instead
test "objagg" which is not an error pointer.

Fixes: 9069a3817d82 ("lib: objagg: implement optimization hints assembly and use hints for object creation")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agolib: objagg: Fix an error code in objagg_hints_get()
Dan Carpenter [Wed, 13 Feb 2019 08:56:50 +0000 (11:56 +0300)]
lib: objagg: Fix an error code in objagg_hints_get()

We need to set the error code on this path otherwise we return
ERR_PTR(0) which would result in a NULL dereference in the caller.

Fixes: 9069a3817d82 ("lib: objagg: implement optimization hints assembly and use hints for object creation")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: dsa: bcm_sf2: potential array overflow in bcm_sf2_sw_suspend()
Dan Carpenter [Wed, 13 Feb 2019 08:23:04 +0000 (11:23 +0300)]
net: dsa: bcm_sf2: potential array overflow in bcm_sf2_sw_suspend()

The value of ->num_ports comes from bcm_sf2_sw_probe() and it is less
than or equal to DSA_MAX_PORTS.  The ds->ports[] array is used inside
the dsa_is_user_port() and dsa_is_cpu_port() functions.  The ds->ports[]
array is allocated in dsa_switch_alloc() and it has ds->num_ports
elements so this leads to a static checker warning about a potential out
of bounds read.

Fixes: 8cfa94984c9c ("net: dsa: bcm_sf2: add suspend/resume callbacks")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Vivien Didelot <vivien.didelot@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agocxgb4vf: Few more link management changes.
Vishal Kulkarni [Wed, 13 Feb 2019 05:18:52 +0000 (10:48 +0530)]
cxgb4vf: Few more link management changes.

CR4_QSFP 10G Speed technology should be 10000baseKR_Full
And also report available FEC modes.

Signed-off-by: Vishal Kulkarni <vishal@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge branch 'pagepool-api-and-dma-address-storage'
David S. Miller [Thu, 14 Feb 2019 06:00:17 +0000 (22:00 -0800)]
Merge branch 'pagepool-api-and-dma-address-storage'

Jesper Dangaard Brouer says:

====================
Fix page_pool API and dma address storage

As pointed out by David Miller in [1] the current page_pool implementation
stores dma_addr_t in page->private. This won't work on 32-bit platforms with
64-bit DMA addresses since the page->private is an unsigned long and the
dma_addr_t a u64.

Since no driver is yet using the DMA mapping capabilities of the API let's
fix this by storing the information in 'struct page' and use that to store
and retrieve DMA addresses from network drivers.

As long as the addresses returned from dma_map_page() are aligned the first
bit, used by the compound pages code should not be set.

Ilias tested the first two patches on Espressobin driver mvneta, for which
we have patches for using the DMA API of page_pool.

[1]: https://lore.kernel.org/netdev/20181207.230655.1261252486319967024.davem@davemloft.net/
====================

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agopage_pool: use DMA_ATTR_SKIP_CPU_SYNC for DMA mappings
Jesper Dangaard Brouer [Wed, 13 Feb 2019 01:55:50 +0000 (02:55 +0100)]
page_pool: use DMA_ATTR_SKIP_CPU_SYNC for DMA mappings

As pointed out by Alexander Duyck, the DMA mapping done in page_pool needs
to use the DMA attribute DMA_ATTR_SKIP_CPU_SYNC.

As the principle behind page_pool keeping the pages mapped is that the
driver takes over the DMA-sync steps.

Reported-by: Alexander Duyck <alexander.duyck@gmail.com>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: page_pool: don't use page->private to store dma_addr_t
Ilias Apalodimas [Wed, 13 Feb 2019 01:55:45 +0000 (02:55 +0100)]
net: page_pool: don't use page->private to store dma_addr_t

As pointed out by David Miller the current page_pool implementation
stores dma_addr_t in page->private.
This won't work on 32-bit platforms with 64-bit DMA addresses since the
page->private is an unsigned long and the dma_addr_t a u64.

A previous patch is adding dma_addr_t on struct page to accommodate this.
This patch adapts the page_pool related functions to use the newly added
struct for storing and retrieving DMA addresses from network drivers.

Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomm: add dma_addr_t to struct page
Jesper Dangaard Brouer [Wed, 13 Feb 2019 01:55:40 +0000 (02:55 +0100)]
mm: add dma_addr_t to struct page

The page_pool API is using page->private to store DMA addresses.
As pointed out by David Miller we can't use that on 32-bit architectures
with 64-bit DMA

This patch adds a new dma_addr_t struct to allow storing DMA addresses

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: sched: remove duplicated include from cls_api.c
YueHaibing [Wed, 13 Feb 2019 01:42:00 +0000 (01:42 +0000)]
net: sched: remove duplicated include from cls_api.c

Remove duplicated include.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoflow_offload: fix block stats
John Hurley [Wed, 13 Feb 2019 00:23:52 +0000 (00:23 +0000)]
flow_offload: fix block stats

With the introduction of flow_stats_update(), drivers now update the stats
fields of the passed tc_cls_flower_offload struct, rather than call
tcf_exts_stats_update() directly to update the stats of offloaded TC
flower rules. However, if multiple qdiscs are registered to a TC shared
block and a flower rule is applied, then, when getting stats for the rule,
multiple callbacks may be made.

Take this into consideration by modifying flow_stats_update to gather the
stats from all callbacks. Currently, the values in tc_cls_flower_offload
only account for the last stats callback in the list.

Fixes: 3b1903ef97c0 ("flow_offload: add statistics retrieval infrastructure and use it")
Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: sched: flower: only return error from hw offload if skip_sw
Vlad Buslov [Tue, 12 Feb 2019 21:39:06 +0000 (23:39 +0200)]
net: sched: flower: only return error from hw offload if skip_sw

Recently introduced tc_setup_flow_action() can fail when parsing tcf_exts
on some unsupported action commands. However, this should not affect the
case when user did not explicitly request hw offload by setting skip_sw
flag. Modify tc_setup_flow_action() callers to only propagate the error if
skip_sw flag is set for filter that is being offloaded, and set extack
error message in that case.

Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Fixes: 3a7b68617de7 ("cls_api: add translator to flow_action representation")
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: fix possible overflow in __sk_mem_raise_allocated()
Eric Dumazet [Tue, 12 Feb 2019 20:26:27 +0000 (12:26 -0800)]
net: fix possible overflow in __sk_mem_raise_allocated()

With many active TCP sockets, fat TCP sockets could fool
__sk_mem_raise_allocated() thanks to an overflow.

They would increase their share of the memory, instead
of decreasing it.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoqlge: fix some indentation issues
Colin Ian King [Tue, 12 Feb 2019 16:08:07 +0000 (16:08 +0000)]
qlge: fix some indentation issues

There are some statements that are indented incorrectly. Fix these.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoqed: fix indentation issue with statements in an if-block
Colin Ian King [Tue, 12 Feb 2019 16:01:53 +0000 (16:01 +0000)]
qed: fix indentation issue with statements in an if-block

There are some statements in an if-block that are not correctly
indented. Fix these.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: ixp4xx_eth: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 12 Feb 2019 16:01:11 +0000 (00:01 +0800)]
net: ixp4xx_eth: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in eth_txdone_irq() when skb
xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: macb: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 12 Feb 2019 16:00:02 +0000 (00:00 +0800)]
net: macb: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in at91ether_interrupt() when
skb xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: sis: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 12 Feb 2019 15:59:04 +0000 (23:59 +0800)]
net: sis: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called when skb xmit done. It makes
drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: fealnx: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 12 Feb 2019 15:56:53 +0000 (23:56 +0800)]
net: fealnx: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in intr_handler() when skb
xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: moxa: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 12 Feb 2019 15:56:00 +0000 (23:56 +0800)]
net: moxa: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in moxart_tx_finished() when
skb xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: apple: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 12 Feb 2019 15:52:53 +0000 (23:52 +0800)]
net: apple: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in mace_interrupt() when skb
xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: atheros: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 12 Feb 2019 15:51:45 +0000 (23:51 +0800)]
net: atheros: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called when skb xmit done. It makes
drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: qualcomm: emac: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 12 Feb 2019 15:49:57 +0000 (23:49 +0800)]
net: qualcomm: emac: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in emac_mac_tx_process() when
skb xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: neterion: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 12 Feb 2019 15:47:31 +0000 (23:47 +0800)]
net: neterion: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called when skb xmit done. It makes
drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agodsa: mv88e6xxx: Ensure all pending interrupts are handled prior to exit
John David Anglin [Mon, 11 Feb 2019 18:40:21 +0000 (13:40 -0500)]
dsa: mv88e6xxx: Ensure all pending interrupts are handled prior to exit

The GPIO interrupt controller on the espressobin board only supports edge interrupts.
If one enables the use of hardware interrupts in the device tree for the 88E6341, it is
possible to miss an edge.  When this happens, the INTn pin on the Marvell switch is
stuck low and no further interrupts occur.

I found after adding debug statements to mv88e6xxx_g1_irq_thread_work() that there is
a race in handling device interrupts (e.g. PHY link interrupts).  Some interrupts are
directly cleared by reading the Global 1 status register.  However, the device interrupt
flag, for example, is not cleared until all the unmasked SERDES and PHY ports are serviced.
This is done by reading the relevant SERDES and PHY status register.

The code only services interrupts whose status bit is set at the time of reading its status
register.  If an interrupt event occurs after its status is read and before all interrupts
are serviced, then this event will not be serviced and the INTn output pin will remain low.

This is not a problem with polling or level interrupts since the handler will be called
again to process the event.  However, it's a big problem when using level interrupts.

The fix presented here is to add a loop around the code servicing switch interrupts.  If
any pending interrupts remain after the current set has been handled, we loop and process
the new set.  If there are no pending interrupts after servicing, we are sure that INTn has
gone high and we will get an edge when a new event occurs.

Tested on espressobin board.

Fixes: dc30c35be720 ("net: dsa: mv88e6xxx: Implement interrupt support.")
Signed-off-by: John David Anglin <dave.anglin@bell.net>
Tested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: phy: fix interrupt handling in non-started states
Heiner Kallweit [Tue, 12 Feb 2019 18:56:15 +0000 (19:56 +0100)]
net: phy: fix interrupt handling in non-started states

phylib enables interrupts before phy_start() has been called, and if
we receive an interrupt in a non-started state, the interrupt handler
returns IRQ_NONE. This causes problems with at least one Marvell chip
as reported by Andrew.
Fix this by handling interrupts the same as in phy_mac_interrupt(),
basically always running the phylib state machine. It knows when it
has to do something and when not.
This change allows to handle interrupts gracefully even if they
occur in a non-started state.

Fixes: 2b3e88ea6528 ("net: phy: improve phy state checking")
Reported-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge branch 'lwt_encap_ip'
Alexei Starovoitov [Thu, 14 Feb 2019 02:27:56 +0000 (18:27 -0800)]
Merge branch 'lwt_encap_ip'

Peter Oskolkov says:

====================
This patchset implements BPF_LWT_ENCAP_IP mode in bpf_lwt_push_encap
BPF helper. It enables BPF programs (specifically, BPF_PROG_TYPE_LWT_IN
and BPF_PROG_TYPE_LWT_XMIT prog types) to add IP encapsulation headers
to packets (e.g. IP/GRE, GUE, IPIP).

This is useful when thousands of different short-lived flows should be
encapped, each with different and dynamically determined destination.
Although lwtunnels can be used in some of these scenarios, the ability
to dynamically generate encap headers adds more flexibility, e.g.
when routing depends on the state of the host (reflected in global bpf
maps).

V2 changes: added flowi-based route lookup, IPv6 encapping, and
   encapping on ingress.

V3 changes: incorporated David Ahern's suggestions:
   - added l3mdev check/oif (patch 2)
   - sync bpf.h from include/uapi into tools/include/uapi
   - selftest tweaks

V4 changes: moved route lookup/dst change from bpf_push_ip_encap
   to when BPF_LWT_REROUTE is handled, as suggested by David Ahern.

V5 changes: added a check in lwt_xmit that skb->protocol stays the
   same if the skb is to be passed back to the stack (ret == BPF_OK).
   Again, suggested by David Ahern.

V6 changes: abandoned.

V7 changes: added handling of GSO packets (patch 3 in the patchset added),
   as suggested by BPF maintainers.

V8 changes:
   - fixed build errors when LWT or IPV6 are not enabled;
   - whitelisted TCP GSO instead of blacklisting SCTP and UDP GSO, as
     suggested by Willem de Bruijn;
   - added validation that pushed length cover needed headers when GRE/UDP
     encap is detected, as suggested by Willem de Bruijn;
   - a couple of minor/stylistic tweaks/fixed typos.

V9 changes:
   - fixed a kbuild test robot compiler warning;
   - added ipv6_route_input to ipv6_stub (patch 4 in the patchset
     added), and IPv6 routing functions are now invoked via ipv6_stub,
     as suggested by David Ahern.

V10 changes:
   - removed unnecessary IS_ENABLED and pr_warn_once from patch 5.

V11 changes: fixed a potential dst leak in patch 5, as suggested by
    David Ahern.
====================

Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
6 years agoselftests: bpf: add test_lwt_ip_encap selftest
Peter Oskolkov [Wed, 13 Feb 2019 19:53:41 +0000 (11:53 -0800)]
selftests: bpf: add test_lwt_ip_encap selftest

This patch adds a bpf self-test to cover BPF_LWT_ENCAP_IP mode
in bpf_lwt_push_encap.

Covered:
- encapping in LWT_IN and LWT_XMIT
- IPv4 and IPv6

A follow-up patch will add GSO and VRF-enabled tests.

Signed-off-by: Peter Oskolkov <posk@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
6 years agobpf: sync <kdir>/include/.../bpf.h with tools/include/.../bpf.h
Peter Oskolkov [Wed, 13 Feb 2019 19:53:40 +0000 (11:53 -0800)]
bpf: sync <kdir>/include/.../bpf.h with tools/include/.../bpf.h

This patch copies changes in bpf.h done by a previous patch
in this patchset from the kernel uapi include dir into tools
uapi include dir.

Signed-off-by: Peter Oskolkov <posk@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
6 years agobpf: add handling of BPF_LWT_REROUTE to lwt_bpf.c
Peter Oskolkov [Wed, 13 Feb 2019 19:53:39 +0000 (11:53 -0800)]
bpf: add handling of BPF_LWT_REROUTE to lwt_bpf.c

This patch builds on top of the previous patch in the patchset,
which added BPF_LWT_ENCAP_IP mode to bpf_lwt_push_encap. As the
encapping can result in the skb needing to go via a different
interface/route/dst, bpf programs can indicate this by returning
BPF_LWT_REROUTE, which triggers a new route lookup for the skb.

v8 changes: fix kbuild errors when LWTUNNEL_BPF is builtin, but
   IPV6 is a module: as LWTUNNEL_BPF can only be either Y or N,
   call IPV6 routing functions only if they are built-in.

v9 changes:
   - fixed a kbuild test robot compiler warning;
   - call IPV6 routing functions via ipv6_stub.

v10 changes: removed unnecessary IS_ENABLED and pr_warn_once.

v11 changes: fixed a potential dst leak.

Signed-off-by: Peter Oskolkov <posk@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
6 years agoipv6_stub: add ipv6_route_input stub/proxy.
Peter Oskolkov [Wed, 13 Feb 2019 19:53:38 +0000 (11:53 -0800)]
ipv6_stub: add ipv6_route_input stub/proxy.

Proxy ip6_route_input via ipv6_stub, for later use by lwt bpf ip encap
(see the next patch in the patchset).

Signed-off-by: Peter Oskolkov <posk@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
6 years agobpf: handle GSO in bpf_lwt_push_encap
Peter Oskolkov [Wed, 13 Feb 2019 19:53:37 +0000 (11:53 -0800)]
bpf: handle GSO in bpf_lwt_push_encap

This patch adds handling of GSO packets in bpf_lwt_push_ip_encap()
(called from bpf_lwt_push_encap):

* IPIP, GRE, and UDP encapsulation types are deduced by looking
  into iphdr->protocol or ipv6hdr->next_header;
* SCTP GSO packets are not supported (as bpf_skb_proto_4_to_6
  and similar do);
* UDP_L4 GSO packets are also not supported (although they are
  not blocked in bpf_skb_proto_4_to_6 and similar), as
  skb_decrease_gso_size() will break it;
* SKB_GSO_DODGY bit is set.

Note: it may be possible to support SCTP and UDP_L4 gso packets;
      but as these cases seem to be not well handled by other
      tunneling/encapping code paths, the solution should
      be generic enough to apply to all tunneling/encapping code.

v8 changes:
   - make sure that if GRE or UDP encap is detected, there is
     enough of pushed bytes to cover both IP[v6] + GRE|UDP headers;
   - do not reject double-encapped packets;
   - whitelist TCP GSO packets rather than block SCTP GSO and
     UDP GSO.

Signed-off-by: Peter Oskolkov <posk@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
6 years agobpf: implement BPF_LWT_ENCAP_IP mode in bpf_lwt_push_encap
Peter Oskolkov [Wed, 13 Feb 2019 19:53:36 +0000 (11:53 -0800)]
bpf: implement BPF_LWT_ENCAP_IP mode in bpf_lwt_push_encap

Implement BPF_LWT_ENCAP_IP mode in bpf_lwt_push_encap BPF helper.
It enables BPF programs (specifically, BPF_PROG_TYPE_LWT_IN and
BPF_PROG_TYPE_LWT_XMIT prog types) to add IP encapsulation headers
to packets (e.g. IP/GRE, GUE, IPIP).

This is useful when thousands of different short-lived flows should be
encapped, each with different and dynamically determined destination.
Although lwtunnels can be used in some of these scenarios, the ability
to dynamically generate encap headers adds more flexibility, e.g.
when routing depends on the state of the host (reflected in global bpf
maps).

v7 changes:
 - added a call skb_clear_hash();
 - removed calls to skb_set_transport_header();
 - refuse to encap GSO-enabled packets.

v8 changes:
 - fix build errors when LWT is not enabled.

Note: the next patch in the patchset with deal with GSO-enabled packets,
which are currently rejected at encapping attempt.

Signed-off-by: Peter Oskolkov <posk@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
6 years agobpf: add plumbing for BPF_LWT_ENCAP_IP in bpf_lwt_push_encap
Peter Oskolkov [Wed, 13 Feb 2019 19:53:35 +0000 (11:53 -0800)]
bpf: add plumbing for BPF_LWT_ENCAP_IP in bpf_lwt_push_encap

This patch adds all needed plumbing in preparation to allowing
bpf programs to do IP encapping via bpf_lwt_push_encap. Actual
implementation is added in the next patch in the patchset.

Of note:
- bpf_lwt_push_encap can now be called from BPF_PROG_TYPE_LWT_XMIT
  prog types in addition to BPF_PROG_TYPE_LWT_IN;
- if the skb being encapped has GSO set, encapsulation is limited
  to IPIP/IP+GRE/IP+GUE (both IPv4 and IPv6);
- as route lookups are different for ingress vs egress, the single
  external bpf_lwt_push_encap BPF helper is routed internally to
  either bpf_lwt_in_push_encap or bpf_lwt_xmit_push_encap BPF_CALLs,
  depending on prog type.

v8 changes: fixed a typo.

Signed-off-by: Peter Oskolkov <posk@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
6 years agosctp: set stream ext to NULL after freeing it in sctp_stream_outq_migrate
Xin Long [Tue, 12 Feb 2019 10:51:01 +0000 (18:51 +0800)]
sctp: set stream ext to NULL after freeing it in sctp_stream_outq_migrate

In sctp_stream_init(), after sctp_stream_outq_migrate() freed the
surplus streams' ext, but sctp_stream_alloc_out() returns -ENOMEM,
stream->outcnt will not be set to 'outcnt'.

With the bigger value on stream->outcnt, when closing the assoc and
freeing its streams, the ext of those surplus streams will be freed
again since those stream exts were not set to NULL after freeing in
sctp_stream_outq_migrate(). Then the invalid-free issue reported by
syzbot would be triggered.

We fix it by simply setting them to NULL after freeing.

Fixes: 5bbbbe32a431 ("sctp: introduce stream scheduler foundations")
Reported-by: syzbot+58e480e7b28f2d890bfd@syzkaller.appspotmail.com
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agosctp: call gso_reset_checksum when computing checksum in sctp_gso_segment
Xin Long [Tue, 12 Feb 2019 10:47:30 +0000 (18:47 +0800)]
sctp: call gso_reset_checksum when computing checksum in sctp_gso_segment

Jianlin reported a panic when running sctp gso over gre over vlan device:

  [   84.772930] RIP: 0010:do_csum+0x6d/0x170
  [   84.790605] Call Trace:
  [   84.791054]  csum_partial+0xd/0x20
  [   84.791657]  gre_gso_segment+0x2c3/0x390
  [   84.792364]  inet_gso_segment+0x161/0x3e0
  [   84.793071]  skb_mac_gso_segment+0xb8/0x120
  [   84.793846]  __skb_gso_segment+0x7e/0x180
  [   84.794581]  validate_xmit_skb+0x141/0x2e0
  [   84.795297]  __dev_queue_xmit+0x258/0x8f0
  [   84.795949]  ? eth_header+0x26/0xc0
  [   84.796581]  ip_finish_output2+0x196/0x430
  [   84.797295]  ? skb_gso_validate_network_len+0x11/0x80
  [   84.798183]  ? ip_finish_output+0x169/0x270
  [   84.798875]  ip_output+0x6c/0xe0
  [   84.799413]  ? ip_append_data.part.50+0xc0/0xc0
  [   84.800145]  iptunnel_xmit+0x144/0x1c0
  [   84.800814]  ip_tunnel_xmit+0x62d/0x930 [ip_tunnel]
  [   84.801699]  gre_tap_xmit+0xac/0xf0 [ip_gre]
  [   84.802395]  dev_hard_start_xmit+0xa5/0x210
  [   84.803086]  sch_direct_xmit+0x14f/0x340
  [   84.803733]  __dev_queue_xmit+0x799/0x8f0
  [   84.804472]  ip_finish_output2+0x2e0/0x430
  [   84.805255]  ? skb_gso_validate_network_len+0x11/0x80
  [   84.806154]  ip_output+0x6c/0xe0
  [   84.806721]  ? ip_append_data.part.50+0xc0/0xc0
  [   84.807516]  sctp_packet_transmit+0x716/0xa10 [sctp]
  [   84.808337]  sctp_outq_flush+0xd7/0x880 [sctp]

It was caused by SKB_GSO_CB(skb)->csum_start not set in sctp_gso_segment.
sctp_gso_segment() calls skb_segment() with 'feature | NETIF_F_HW_CSUM',
which causes SKB_GSO_CB(skb)->csum_start not to be set in skb_segment().

For TCP/UDP, when feature supports HW_CSUM, CHECKSUM_PARTIAL will be set
and gso_reset_checksum will be called to set SKB_GSO_CB(skb)->csum_start.

So SCTP should do the same as TCP/UDP, to call gso_reset_checksum() when
computing checksum in sctp_gso_segment.

Reported-by: Jianlin Shi <jishi@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge branch 'phy-25g'
David S. Miller [Thu, 14 Feb 2019 00:17:53 +0000 (19:17 -0500)]
Merge branch 'phy-25g'

Maxime Chevallier says:

====================
net: phy: Add 2.5G/5GBASET PHYs support

The 802.3bz standard defines 2 modes based on the NBASET alliance work
that allow to use 2.5Gbps and 5Gbps speeds on Cat 5e, 6 and 7 cables.

This series adds the necessary infrastructure to handle these modes with
C45 PHYs. This series was originally part of a bigger one, that has
seen 2 iterations [1] [2] that added support for these modes on Marvell
Alaska PHYs.

Following some discussions with Heiner and Andrew [3], we decided to
split-out the generic parts so that we can work together on the
following steps to get these mode fully working with Aquantia and
Marvell PHYS.

The first 3 patches are reworking some of the internal network phy
infrastructure to handle the new modes in a more generic way.

The 4th patch adds all the C45 register definition and accesses that
follows the 802.3bz standard to support 2.5GBASET and 5GBASET.

[1] : https://lore.kernel.org/netdev/20190118152352.26417-1-maxime.chevallier@bootlin.com/
[2] : https://lore.kernel.org/netdev/20190207094939.27369-1-maxime.chevallier@bootlin.com/
[3] : https://lore.kernel.org/netdev/81c340ea-54b0-1abf-94af-b8dc4ee83e3a@gmail.com/
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: phy: Add generic support for 2.5GBaseT and 5GBaseT
Maxime Chevallier [Mon, 11 Feb 2019 14:25:29 +0000 (15:25 +0100)]
net: phy: Add generic support for 2.5GBaseT and 5GBaseT

The 802.3bz specification, based on previous by the NBASET alliance,
defines the 2.5GBaseT and 5GBaseT link modes for ethernet traffic on
cat5e, cat6 and cat7 cables.

These mode integrate with the already defined C45 MDIO PMA/PMD registers
set that added 10G support, by defining some previously reserved bits,
and adding a new register (2.5G/5G Extended abilities).

This commit adds the required definitions in include/uapi/linux/mdio.h
to support these modes, and detect when a link-partner advertises them.

It also adds support for these mode in the generic C45 PHY
infrastructure.

Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Reviewed-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: phy: Extract genphy_c45_pma_read_abilities from marvell10g
Maxime Chevallier [Mon, 11 Feb 2019 14:25:28 +0000 (15:25 +0100)]
net: phy: Extract genphy_c45_pma_read_abilities from marvell10g

Marvell 10G PHY driver has a generic way of initializing the supported
link modes by reading the PHY's C45 PMA abilities. This can be made
generic, since these registers are part of the 802.3 specifications.

This commit extracts the config_init link_mode initialization code from
marvell10g and uses it to introduce the genphy_c45_pma_read_abilities
function.

Only PMA modes are read, it's still up to the caller to set the Pause
parameters.

Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: phy: Move of_set_phy_eee_broken to phy-core.c
Maxime Chevallier [Mon, 11 Feb 2019 14:25:27 +0000 (15:25 +0100)]
net: phy: Move of_set_phy_eee_broken to phy-core.c

Since of_set_phy_supported was moved to phy-core.c, we can also move
of_set_phy_eee_broken to the same location, so that we have all OF
functions in the same place.

This patch doesn't intend to introduce any change in behaviour.

Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: phy: Mask-out non-compatible modes when setting the max-speed
Maxime Chevallier [Mon, 11 Feb 2019 14:25:26 +0000 (15:25 +0100)]
net: phy: Mask-out non-compatible modes when setting the max-speed

When setting a PHY's max speed using either the max-speed DT property
or ethtool, we should mask-out all non-compatible modes according to the
settings table, instead of just the 10/100BASET modes.

Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Suggested-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf
David S. Miller [Thu, 14 Feb 2019 00:13:47 +0000 (19:13 -0500)]
Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf

Pablo Neira Ayuso says:

====================
Netfilter/IPVS fixes for net

The following patchset contains Netfilter/IPVS fixes for net:

1) Missing structure initialization in ebtables causes splat with
   32-bit user level on a 64-bit kernel, from Francesco Ruggeri.

2) Missing dependency on nf_defrag in IPVS IPv6 codebase, from
   Andrea Claudi.

3) Fix possible use-after-free from release path of target extensions.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge tag 'mlx5-fixes-2019-02-13' of git://git.kernel.org/pub/scm/linux/kernel/git...
David S. Miller [Thu, 14 Feb 2019 00:07:47 +0000 (19:07 -0500)]
Merge tag 'mlx5-fixes-2019-02-13' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux

Saeed Mahameed says:

====================
Mellanox, mlx5 fixes 2019-02-13

This series introduces some fixes to mlx5 driver.
For more information please see tag log below.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/mlx5e: XDP, fix redirect resources availability check
Saeed Mahameed [Tue, 12 Feb 2019 00:27:02 +0000 (16:27 -0800)]
net/mlx5e: XDP, fix redirect resources availability check

Currently mlx5 driver creates xdp redirect hw queues unconditionally on
netdevice open, This is great until someone starts redirecting XDP traffic
via ndo_xdp_xmit on mlx5 device and changes the device configuration at
the same time, this might cause crashes, since the other device's napi
is not aware of the mlx5 state change (resources un-availability).

To fix this we must synchronize with other devices napi's on the system.
Added a new flag under mlx5e_priv to determine XDP TX resources are
available, set/clear it up when necessary and use synchronize_rcu()
when the flag is turned off, so other napi's are in-sync with it, before
we actually cleanup the hw resources.

The flag is tested prior to committing to transmit on mlx5e_xdp_xmit, and
it is sufficient to determine if it safe to transmit or not. The other
two internal flags (MLX5E_STATE_OPENED and MLX5E_SQ_STATE_ENABLED) become
unnecessary. Thus, they are removed from data path.

Fixes: 58b99ee3e3eb ("net/mlx5e: Add support for XDP_REDIRECT in device-out side")
Reported-by: Toke Høiland-Jørgensen <toke@redhat.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5: Fix a compilation warning in events.c
Tariq Toukan [Tue, 22 Jan 2019 12:18:04 +0000 (14:18 +0200)]
net/mlx5: Fix a compilation warning in events.c

Eliminate the following compilation warning:

drivers/net/ethernet/mellanox/mlx5/core/events.c: warning: 'error_str'
may be used uninitialized in this function [-Wuninitialized]:  => 238:3

Fixes: c2fb3db22d35 ("net/mlx5: Rework handling of port module events")
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Mikhael Goikhman <migo@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5: No command allowed when command interface is not ready
Huy Nguyen [Thu, 7 Feb 2019 15:22:56 +0000 (09:22 -0600)]
net/mlx5: No command allowed when command interface is not ready

When EEH is injected and PCI bus stalls, mlx5's pci error detect
function is called to deactivate the command interface and tear down
the device. The issue is that there can be a thread that already
passed MLX5_DEVICE_STATE_INTERNAL_ERROR check, it will send the command
and stuck in the wait_func.

Solution:
Add function mlx5_cmd_flush to disable command interface and clear all
the pending commands. When device state is set to
MLX5_DEVICE_STATE_INTERNAL_ERROR, call mlx5_cmd_flush to ensure all
pending threads waiting for firmware commands completion are terminated.

Fixes: c1d4d2e92ad6 ("net/mlx5: Avoid calling sleeping function by the health poll thread")
Signed-off-by: Huy Nguyen <huyn@mellanox.com>
Reviewed-by: Daniel Jurgens <danielj@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5e: Fix NULL pointer derefernce in set channels error flow
Maria Pasechnik [Sun, 3 Feb 2019 15:55:09 +0000 (17:55 +0200)]
net/mlx5e: Fix NULL pointer derefernce in set channels error flow

New channels are applied to the priv channels only after they
are successfully opened. Then, the indirection table should be built
according to the new number of channels.
Currently, such build is preformed independently of whether the
channels opening is successful, and is not reverted on failure.

The bug is caused due to removal of rss params from channels struct
and moving it to priv struct. That change cause to independency between
channels and rss params.
This causes a crash on a later point, when accessing rqn of a non
existing channel.

This patch fixes it by moving the indirection table build right before
switching the priv channels to new channels struct, after the new set of
channels was successfully opened.

Fixes: bbeb53b8b2c9 ("net/mlx5e: Move RSS params to a dedicated struct")
Signed-off-by: Maria Pasechnik <mariap@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agoMerge tag 'trace-v5.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt...
Linus Torvalds [Wed, 13 Feb 2019 18:28:17 +0000 (10:28 -0800)]
Merge tag 'trace-v5.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace

Pull tracing fix from Steven Rostedt:
 "This fixes kprobes/uprobes dynamic processing of strings, where it
  processes the args but does not update the remaining length of the
  buffer that the string arguments will be placed in. It constantly
  passes in the total size of buffer used instead of passing in the
  remaining size of the buffer used.

  This could cause issues if the strings are larger than the max size of
  an event which could cause the strings to be written beyond what was
  reserved on the buffer"

* tag 'trace-v5.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  tracing: probeevent: Correctly update remaining space in dynamic area

6 years agonet/mlx5: Align ODP capability function with netdev coding style
Leon Romanovsky [Mon, 11 Feb 2019 11:56:07 +0000 (13:56 +0200)]
net/mlx5: Align ODP capability function with netdev coding style

Update newly introduced function to be aligned to netdev coding style.

Fixes: 46861e3e88be ("net/mlx5: Set ODP SRQ support in firmware")
Reported-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
6 years agonetfilter: nft_compat: use-after-free when deleting targets
Pablo Neira Ayuso [Wed, 13 Feb 2019 12:03:53 +0000 (13:03 +0100)]
netfilter: nft_compat: use-after-free when deleting targets

Fetch pointer to module before target object is released.

Fixes: 29e3880109e3 ("netfilter: nf_tables: fix use-after-free when deleting compat expressions")
Fixes: 0ca743a55991 ("netfilter: nf_tables: add compatibility layer for x_tables")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
6 years agodrm/amdgpu/psp11: TA firmware is optional (v3)
Alex Deucher [Tue, 12 Feb 2019 14:54:31 +0000 (09:54 -0500)]
drm/amdgpu/psp11: TA firmware is optional (v3)

Don't warn or fail if it's missing.

v2: handle xgmi case more gracefully.
v3: handle older kernels properly

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Tested-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agoMerge branch 'nvme-5.0' of git://git.infradead.org/nvme into for-linus
Jens Axboe [Wed, 13 Feb 2019 14:32:30 +0000 (07:32 -0700)]
Merge branch 'nvme-5.0' of git://git.infradead.org/nvme into for-linus

Pull single NVMe fix from Christoph

* 'nvme-5.0' of git://git.infradead.org/nvme:
  nvme-pci: add missing unlock for reset error

6 years agosignal: Restore the stop PTRACE_EVENT_EXIT
Eric W. Biederman [Tue, 12 Feb 2019 05:27:42 +0000 (23:27 -0600)]
signal: Restore the stop PTRACE_EVENT_EXIT

In the middle of do_exit() there is there is a call
"ptrace_event(PTRACE_EVENT_EXIT, code);" That call places the process
in TACKED_TRACED aka "(TASK_WAKEKILL | __TASK_TRACED)" and waits for
for the debugger to release the task or SIGKILL to be delivered.

Skipping past dequeue_signal when we know a fatal signal has already
been delivered resulted in SIGKILL remaining pending and
TIF_SIGPENDING remaining set.  This in turn caused the
scheduler to not sleep in PTACE_EVENT_EXIT as it figured
a fatal signal was pending.  This also caused ptrace_freeze_traced
in ptrace_check_attach to fail because it left a per thread
SIGKILL pending which is what fatal_signal_pending tests for.

This difference in signal state caused strace to report
strace: Exit of unknown pid NNNNN ignored

Therefore update the signal handling state like dequeue_signal
would when removing a per thread SIGKILL, by removing SIGKILL
from the per thread signal mask and clearing TIF_SIGPENDING.

Acked-by: Oleg Nesterov <oleg@redhat.com>
Reported-by: Oleg Nesterov <oleg@redhat.com>
Reported-by: Ivan Delalande <colona@arista.com>
Cc: stable@vger.kernel.org
Fixes: 35634ffa1751 ("signal: Always notice exiting tasks")
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
6 years agommc: meson-gx: fix interrupt name
Martin Blumenstingl [Sat, 9 Feb 2019 00:58:50 +0000 (01:58 +0100)]
mmc: meson-gx: fix interrupt name

Commit bb364890323cca ("mmc: meson-gx: Free irq in release() callback")
changed the _probe code to use request_threaded_irq() instead of
devm_request_threaded_irq().
Unfortunately this removes a fallback for the interrupt name:
devm_request_threaded_irq() uses the device name as fallback if the
given IRQ name is NULL. request_threaded_irq() has no such fallback,
thus /proc/interrupts shows "(null)" instead.

Explicitly pass the dev_name() so we get the IRQ name shown in
/proc/interrupts again.
While here, also fix the indentation of the request_threaded_irq()
parameter list.

Fixes: bb364890323cca ("mmc: meson-gx: Free irq in release() callback")
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
6 years agoMerge tag 'imx-drm-fixes-2019-02-12' of git://git.pengutronix.de/pza/linux into drm...
Dave Airlie [Wed, 13 Feb 2019 03:05:03 +0000 (13:05 +1000)]
Merge tag 'imx-drm-fixes-2019-02-12' of git://git.pengutronix.de/pza/linux into drm-fixes

drm/imx: plane, ldb, and ipu-v3 fixes

- Fix CSI register offsets for i.MX51 and i.MX53.
- Fix delayed page flip completion events on i.MX6QP due to unexpected
  behaviour of the PRE when issuing NOP buffer updates to the same
  buffer address.
- Stop throwing errors for plane updates on disabled CRTCs when a
  userspace process is killed while a plane update is pending.
- Add missing of_node_put cleanup in imx_ldb_bind.

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Philipp Zabel <p.zabel@pengutronix.de>
Link: https://patchwork.freedesktop.org/patch/msgid/1549990602.4800.11.camel@pengutronix.de
6 years agoMerge branch 'net-Remove-unused-variables'
David S. Miller [Wed, 13 Feb 2019 01:31:30 +0000 (17:31 -0800)]
Merge branch 'net-Remove-unused-variables'

Florian Fainelli says:

====================
Remove unused variables

This removes unused variables from mlxsw and ethsw after the recent
removal of SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS, build scripts are now
fixed to take care of those warnings :).
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agostaging: fsl-dpaa2: ethsw: Remove unused port_priv variable
Florian Fainelli [Tue, 12 Feb 2019 23:39:59 +0000 (15:39 -0800)]
staging: fsl-dpaa2: ethsw: Remove unused port_priv variable

After 1b8b589d9103 ("staging: fsl-dpaa2: ethsw: Remove getting
PORT_BRIDGE_FLAGS") we are not accessing any driver private data
anymore, remove port_priv which is unused.

Fixes: 1b8b589d9103 ("staging: fsl-dpaa2: ethsw: Remove getting PORT_BRIDGE_FLAGS")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agomlxsw: spectrum_switchdev: Remove unused variables
Florian Fainelli [Tue, 12 Feb 2019 23:39:58 +0000 (15:39 -0800)]
mlxsw: spectrum_switchdev: Remove unused variables

After 1ecb195753a1 ("mlxsw: spectrum_switchdev: Remove getting
PORT_BRIDGE_FLAGS") we are not accessing any driver private data
structure, so the mlxsw_sp_port and mlxsw_sp variables are unused.

Fixes: 1ecb195753a1 ("mlxsw: spectrum_switchdev: Remove getting PORT_BRIDGE_FLAGS")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge branch 'akpm' (patches from Andrew)
Linus Torvalds [Wed, 13 Feb 2019 01:15:33 +0000 (17:15 -0800)]
Merge branch 'akpm' (patches from Andrew)

Merge fixes from Andrew Morton:
 "6 fixes"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  mm: proc: smaps_rollup: fix pss_locked calculation
  Rename include/{uapi => }/asm-generic/shmparam.h really
  Revert "mm: use early_pfn_to_nid in page_ext_init"
  mm/gup: fix gup_pmd_range() for dax
  Revert "mm: slowly shrink slabs with a relatively small number of objects"
  Revert "mm: don't reclaim inodes with many attached pages"

6 years agomm: proc: smaps_rollup: fix pss_locked calculation
Sandeep Patil [Tue, 12 Feb 2019 23:36:11 +0000 (15:36 -0800)]
mm: proc: smaps_rollup: fix pss_locked calculation

The 'pss_locked' field of smaps_rollup was being calculated incorrectly.
It accumulated the current pss everytime a locked VMA was found.  Fix
that by adding to 'pss_locked' the same time as that of 'pss' if the vma
being walked is locked.

Link: http://lkml.kernel.org/r/20190203065425.14650-1-sspatil@android.com
Fixes: 493b0e9d945f ("mm: add /proc/pid/smaps_rollup")
Signed-off-by: Sandeep Patil <sspatil@android.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Daniel Colascione <dancol@google.com>
Cc: <stable@vger.kernel.org> [4.14.x, 4.19.x]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agoRename include/{uapi => }/asm-generic/shmparam.h really
Masahiro Yamada [Tue, 12 Feb 2019 23:36:06 +0000 (15:36 -0800)]
Rename include/{uapi => }/asm-generic/shmparam.h really

Commit 36c0f7f0f899 ("arch: unexport asm/shmparam.h for all
architectures") is different from the patch I submitted.

My patch is this:

  https://lore.kernel.org/lkml/1546904307-11124-1-git-send-email-yamada.masahiro@socionext.com/T/#u

The file renaming part:

  rename include/{uapi => }/asm-generic/shmparam.h (100%)

was lost when it was picked up.

I think it was an accident because Andrew did not say anything.

Link: http://lkml.kernel.org/r/1549158277-24558-1-git-send-email-yamada.masahiro@socionext.com
Fixes: 36c0f7f0f899 ("arch: unexport asm/shmparam.h for all architectures")
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agoRevert "mm: use early_pfn_to_nid in page_ext_init"
Qian Cai [Tue, 12 Feb 2019 23:36:03 +0000 (15:36 -0800)]
Revert "mm: use early_pfn_to_nid in page_ext_init"

This reverts commit fe53ca54270a ("mm: use early_pfn_to_nid in
page_ext_init").

When booting a system with "page_owner=on",

start_kernel
  page_ext_init
    invoke_init_callbacks
      init_section_page_ext
        init_page_owner
          init_early_allocated_pages
            init_zones_in_node
              init_pages_in_zone
                lookup_page_ext
                  page_to_nid

The issue here is that page_to_nid() will not work since some page flags
have no node information until later in page_alloc_init_late() due to
DEFERRED_STRUCT_PAGE_INIT.  Hence, it could trigger an out-of-bounds
access with an invalid nid.

  UBSAN: Undefined behaviour in ./include/linux/mm.h:1104:50
  index 7 is out of range for type 'zone [5]'

Also, kernel will panic since flags were poisoned earlier with,

CONFIG_DEBUG_VM_PGFLAGS=y
CONFIG_NODE_NOT_IN_PAGE_FLAGS=n

start_kernel
  setup_arch
    pagetable_init
      paging_init
        sparse_init
          sparse_init_nid
            memblock_alloc_try_nid_raw

It did not handle it well in init_pages_in_zone() which ends up calling
page_to_nid().

  page:ffffea0004200000 is uninitialized and poisoned
  raw: ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff
  raw: ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff
  page dumped because: VM_BUG_ON_PAGE(PagePoisoned(p))
  page_owner info is not active (free page?)
  kernel BUG at include/linux/mm.h:990!
  RIP: 0010:init_page_owner+0x486/0x520

This means that assumptions behind commit fe53ca54270a ("mm: use
early_pfn_to_nid in page_ext_init") are incomplete.  Therefore, revert
the commit for now.  A proper way to move the page_owner initialization
to sooner is to hook into memmap initialization.

Link: http://lkml.kernel.org/r/20190115202812.75820-1-cai@lca.pw
Signed-off-by: Qian Cai <cai@lca.pw>
Acked-by: Michal Hocko <mhocko@kernel.org>
Cc: Pasha Tatashin <Pavel.Tatashin@microsoft.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Yang Shi <yang.shi@linaro.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This page took 0.118338 seconds and 4 git commands to generate.