Greg Rose [Wed, 9 Apr 2014 05:59:03 +0000 (05:59 +0000)]
i40e/i40evf: Change type to u32 to avoid sparse error
tx_lpi_status and rx_lpi_status are declared as bool but then used in sizeof
operations in the CORE driver code. Make them u32 to make sparse be quiet.
Change-ID: Iad6daeb1c7149e61ece242acd18c64b320c246a3 Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Jim Young <jamesx.m.young@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
i40e: Report cmd->data in ETHTOOL_GRXCLSRLCNT instead of ETHTOOL_GRXCLSRULE
Based on review feedback from upstream cmd->data is not
defined in ETHTOOL_GRXCLSRULE but needs to be reported in
ETHTOOL_GRXCLSRLCNT. Also use a helper function to calculate the total
filter count.
Change-ID: Iaacbf729527b73290c4fdad837b379b44fd7dd20 Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com> Tested-by: Jim Young <jamesx.m.young@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Kevin Scott [Wed, 9 Apr 2014 05:58:59 +0000 (05:58 +0000)]
i40e/i40evf: Remove reserved PCTYPE defines
Patch to remove PCTYPE definitions which are now reserved.
Change-ID: I66c1c16a45a16f4894b2983101ab2a48ce03f1f4 Signed-off-by: Kevin Scott <kevin.c.scott@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Tested-by: Jim Young <jamesx.m.young@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
This patch changes the declaration of Tx/Rx rings inside several loops. It
eliminates declaring the same rings every time for the duration of the loop,
instead declaring them once before the loop.
Change-ID: I59dea54276f18c47dca522f520c18f65fe42a15d Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Mitch Williams [Wed, 9 Apr 2014 05:58:57 +0000 (05:58 +0000)]
i40evf: tweak Tx rate params and de-magic-ify
In order for the VF to achieve its programmed Tx rate, we need to set
the max credits value to 4. While we're at it, get rid of some magic
numbers.
Change-ID: I4f17b4d3a90f1e069bdd134a543f0aa22feac3a9 Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Mitch Williams [Wed, 9 Apr 2014 05:58:56 +0000 (05:58 +0000)]
i40evf: don't round Tx rate down to 0
Because the hardware configures VF Tx rates in increments of 50 Mbps,
values smaller than that would be rounded down to 0, which was
interpreted as no limit at all. Rather than do this, we round up to 50
Mbps and notify the user.
Change-ID: I5275848233fe7514cf93e11323661c68f4c38737 Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Mitch Williams [Wed, 9 Apr 2014 05:58:55 +0000 (05:58 +0000)]
i40e: enable descriptor prefetch for VFs
As recommended by the hardware guys, enable descriptor prefetch for
rings belonging to VFs. This matches a change already made for ring
belonging to the PF.
Change-ID: Idcc4dc7064bc9144ea81f5701ad07b7ecd11ba49 Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Kevin Scott [Wed, 9 Apr 2014 05:58:54 +0000 (05:58 +0000)]
i40e/i40evf: Update check for AQ aliveness
Update the i40e_check_asq_alive check to ensure that the len register
offset is non-zero, indicating that SW has initialized the AQ.
Change-ID: I9c2e804788b4775bef9c7e80954ab004e6bdb306 Signed-off-by: Kevin Scott <kevin.c.scott@intel.com> Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David S. Miller [Wed, 28 May 2014 16:40:13 +0000 (12:40 -0400)]
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next
Jeff Kirsher says:
====================
Intel Wired LAN Driver Updates
This series contains updates to e1000, e1000e, ixgbe and i40evf.
Emil provides a fix for ixgbe so that non-fiber devices with MNG FW enabled
are able to link at 100Mbps.
Jacob provides several changes for ixgbe, most of which are PTP related.
Renames ixgbe_ptp_enable() to ixgbe_ptp_feature_enable() to better reflect
the functions purpose. Extracts the hardware setup logic for the PTP
hardware bits from the ixgbe_ptp_set_ts_config() to enable future work for
the ixgbe_ptp_reset(). Maintain the hwstamp configuration through a reset
and extracts the creation of the PTP clock device from ptp_init() in order
to properly handle a suspend/resume cycle and only calls it if we don't
already have a ptp_clock pointer.
David provides a patch to expend the e1000e driver to turn on unicast
PROMISC when there is failure to write to a shared receive address register.
The fix update_phy_task() for 82579 is expanded to include newer PHYs as well
so that the dev_spec->eee_lp_ability has the correct value when going into
SX states.
Todd provides a e1000e fix an errata for 82574/82583 where it is possible
bad bits are read from SYSTIMH/L so check to see that the time is
incrementing at a reasonable rate and is a multiple of the time incremental
value. Removes a redundant igb PHY power down register write.
Andi Kleen out of lines two write functions for e1000e to save 30k text size.
Tobias Klauser converts the e1000 and i40evf drivers to use the
is_broadcast_ether_addr() and is_multicast_ether_addr().
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Todd Fujinaka [Thu, 8 May 2014 23:20:24 +0000 (23:20 +0000)]
igb: remove redundant PHY power down register write
One of the registers used to power down the PHY was found to be wrong
(should be bit 2 not bit 1) on further inspection it was also found to
be redundant.
Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Tue, 13 May 2014 00:06:26 +0000 (00:06 +0000)]
e1000e: Fix expand setting EEE link info to all affected parts
Previously, the update_phy_task was only calling e1000_set_eee_pchlan()
for phy.type 82579. This patch is to cause this function to be called
for 82579 and newer phy.types. This causes the dev_spec->eee_lp_ability
to have the correct value when going into SX states.
Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Tue, 13 May 2014 00:02:12 +0000 (00:02 +0000)]
e1000e: Cleanup parenthesis around return value
Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Tue, 6 May 2014 03:50:17 +0000 (03:50 +0000)]
e1000e: Failure to write SHRA turns on PROMISC mode
Previously, the check to turn on promiscuous mode only took into account
the total number of SHared Receive Address (SHRA) registers and if the
request was for a register within that range. It is possible that the
Management Engine might have locked a number of SHRA and not allowed a
new address to be written to the requested register.
Add a function to determine the number of unlocked SHRA registers. Then
determine if the number of registers available is sufficient for our needs,
if not then return -ENOMEM so that UNICAST PROMISC mode is activated.
Since the method by which ME claims SHRA registers is non-deterministic,
also add a return value to the function attempting to write an address
to a SHRA, and return a -E1000_ERR_CONFIG if the write fails. The error
will be passed up the function chain and allow the driver to also set
UNICAST PROMISC when this happens.
Cc: Vlad Yasevich <vyasevic@redhat.com> Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Fri, 16 May 2014 05:12:29 +0000 (05:12 +0000)]
ixgbe: avoid duplicate code in suspend and stop paths
Resume path calls .open but suspend path cannot call .stop because
fdirs should not be freed and control over hardware should not be
released until WoL is configured. To avoid having to duplicate all
changes made in .stop on suspend path split out part of .stop that
is relevant during suspend and call it from .stop and during suspend.
This fix also ensures that ixgbe_ptp_suspend is called during the
suspend path, and helps avoid similar errors. We can't call
ixgbe_ptp_stop, since it will free the PTP clock device, which we
shouldn't be doing during a suspend path.
Signed-off-by: Jakub Kicinski <kubakici@wp.pl> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Fri, 16 May 2014 05:12:28 +0000 (05:12 +0000)]
ixgbe: separate the PTP suspend and stop actions
Since we are adding proper support for suspend of PTP, extract out of
ixgbe_ptp_stop those things relevant to suspend. Then, have
ixgbe_ptp_stop call ixgbe_ptp_suspend. The next patch in the series will
have ixgbe_ptp_suspend called from the ixgbe_suspend path.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Fri, 16 May 2014 05:12:27 +0000 (05:12 +0000)]
ixgbe: extract PTP clock device from ptp_init
In order to properly handle a suspend/resume cycle, we cannot destroy
the PTP clock device. As part of this, we should only re-create the
device on first initialization. After a resume, when ixgbe_ptp_init is
called, we won't create a new clock, and we will use the old clock
device. To that end, this patch extracts the clock creation out of
ptp_init, and only calls it if we don't already have a ptp_clock
pointer.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Fri, 16 May 2014 05:12:26 +0000 (05:12 +0000)]
ixgbe: allow ixgbe_ptp_reset to maintain current hwtstamp config
Rather than clearing the hwtstamp configuration, we should use the known
configuration requested by the user and call the function which has now
been separated from the ioctl. This means that after a reset, the
timestamp mode will be maintained rather than lost. We still can't
maintain the clock value, however.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Fri, 16 May 2014 05:12:25 +0000 (05:12 +0000)]
ixgbe: extract the hardware setup from the ixgbe_ptp_set_ts_config
Currently all of the hardware setup logic for the PTP hardware bits is
buried inside of the ioctl which sets the timestamp configuration. This
makes it hard to use this logic in other places (primarily reset), and
this means we can't restore current timestamp mode upon a MAC reset.
Extracting this logic into a separate function will enable future work
for the ixgbe_ptp_reset function.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Fri, 16 May 2014 05:12:24 +0000 (05:12 +0000)]
ixgbe: rename ixgbe_ptp_enable to ixgbe_ptp_feature_enable
Since the name ixgbe_ptp_enable could be misconstrued as a function
which enables the whole PTP core, rename this function so that it is
clear the function is for enabling of the extra features such as PPS
signal.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Emil Tantilov [Tue, 13 May 2014 08:24:00 +0000 (08:24 +0000)]
ixgbe: fix linking at 100Mbps on copper devices with MNG FW enabled
Driver was calling setup_link to make sure that fiber interfaces with MNG FW
enabled will get link on probe because the laser was most likely turned off.
This prevented non-fiber devices with MNG FW from linking at 100Mbps.
This patch adds a check to only call setup_link for fiber devices.
Reported-and-by: Tony Luck <tony.luck@intel.com> Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> Tested-by: Tony Luck <tony.luck@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Pablo Neira [Sun, 25 May 2014 12:48:33 +0000 (14:48 +0200)]
netfilter: bridge: fix Kconfig unmet dependencies
Before f5efc69 ("netfilter: nf_tables: Add meta expression key for
bridge interface name"), the entire net/bridge/netfilter/ directory
depended on BRIDGE_NF_EBTABLES, ie. on ebtables. However, that
directory already contained the nf_tables bridge extension that
we should allow to compile separately. In f5efc69, we tried to
generalize this by using CONFIG_BRIDGE_NETFILTER which was not a good
idea since this option already existed and it is dedicated to enable
the Netfilter bridge IP/ARP filtering.
Let's try to fix this mess by:
1) making net/bridge/netfilter/ dependent on the toplevel
CONFIG_NETFILTER option, just like we do with the net/netfilter and
net/ipv{4,6}/netfilter/ directories.
2) Changing 'selects' to 'depends on' NETFILTER_XTABLES for
BRIDGE_NF_EBTABLES. I believe this problem was already before f5efc69:
warning: (BRIDGE_NF_EBTABLES) selects NETFILTER_XTABLES which has
unmet direct dependencies (NET && INET && NETFILTER)
3) Fix ebtables/nf_tables bridge dependencies by making NF_TABLES_BRIDGE
and BRIDGE_NF_EBTABLES dependent on BRIDGE and NETFILTER:
warning: (NF_TABLES_BRIDGE && BRIDGE_NF_EBTABLES) selects
BRIDGE_NETFILTER which has unmet direct dependencies (NET && BRIDGE &&
NETFILTER && INET && NETFILTER_ADVANCED)
net/built-in.o: In function `br_parse_ip_options':
br_netfilter.c:(.text+0x4a5ba): undefined reference to `ip_options_compile'
br_netfilter.c:(.text+0x4a5ed): undefined reference to `ip_options_rcv_srr'
net/built-in.o: In function `br_nf_pre_routing_finish':
br_netfilter.c:(.text+0x4a8a4): undefined reference to `ip_route_input_noref'
br_netfilter.c:(.text+0x4a987): undefined reference to `ip_route_output_flow'
make: *** [vmlinux] Error 1
Reported-by: Jim Davis <jim.epost@gmail.com> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: David S. Miller <davem@davemloft.net>
of: mdio: fix compile warning in of_mdiobus_register_phy()
Commit de906af1 (net: phy: make of_set_phy_supported work with genphy driver)
removed the last user of variable 'max_speed' in function
of_mdiobus_register_phy(), leading to compile warning "unused variable
‘max_speed’ [-Wunused-variable]". Thus remove it.
Signed-off-by: Christian Engelmayer <cengelma@gmx.at> Acked-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
tg3: Override clock, link aware and link idle mode during NVRAM dump
When cable is not present the clock speed of some of the devices is
reduced based upon power saving mode setting in NVRAM. Due to this
NVRAM reads take long time to complete as a result CPU soft lockup
message is seen. Fix is to override clock, disable link aware and link
idle modes before NVRAM reads and restore them back after the reads
are complete. During this period also check if the thread needs to be
rescheduled and if there are any signals to handle.
Also decrease the NVRAM command execution timeout value to 1ms.
Signed-off-by: Prashant Sreedharan <prashant@broadcom.com> Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann [Sat, 24 May 2014 19:47:46 +0000 (21:47 +0200)]
team: lb: use sizeof(*fprog) in __fprog_create
sock_fprog and sock_fprog_kern are of equal size, however
it's cleaner to just use sizeof(*fprog) instead to always
have correct type.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
As it may then take longer than what the user specified using
setsockopt(SO_RCVTIMEO).
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
This set of patches properly documents the micrel ks8851 spi ethernet
controller, converts to devm_regulator_get_optional() to make error
paths slightly simpler, and finally adds supports for another
optional regulator and a reset gpio. This allows me to use the ks8851
on my MSM8960 CDP board.
Changes since v1:
* Dropped vendor prefix patch as that should go through DT tree
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Stephen Boyd [Fri, 23 May 2014 19:57:20 +0000 (12:57 -0700)]
net: ks8851: Add of match table
Users are currently just providing "ks8851" as the compatible for
this driver in device tree. Add a compatible string that provides
the vendor name along with the device name to be more explicit.
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: David S. Miller <davem@davemloft.net>
The ks8851 SPI ethernet wasn't documented, but we documented the
optional regulator supply for it under the mll based ethernet
chip. Furthermore, that compatible string needed another 'l'. Fix
all of this and document the optional vdd-io and reset-gpios
properties.
Cc: Nishanth Menon <nm@ti.com> Cc: Rob Herring <robh+dt@kernel.org> Cc: Pawel Moll <pawel.moll@arm.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Ian Campbell <ijc+devicetree@hellion.org.uk> Cc: Kumar Gala <galak@codeaurora.org> Cc: <devicetree@vger.kernel.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Pull networking fixes from David Miller:
"It looks like a sizeble collection but this is nearly 3 weeks of bug
fixing while you were away.
1) Fix crashes over IPSEC tunnels with NAT, the latter can reroute
the packet through a non-IPSEC protected path and the code has to
be able to handle SKBs attached to routes lacking an attached xfrm
state. From Steffen Klassert.
2) Fix OOPSs in ipv4 and ipv6 ipsec layers for unsupported
sub-protocols, also from Steffen Klassert.
3) Set local_df on fragmented netfilter skbs otherwise we won't be
able to forward successfully, from Florian Westphal.
4) cdc_mbim ipv6 neighbour code does __vlan_find_dev_deep without
holding RCU lock, from Bjorn Mork.
5) local_df test in ip_may_fragment is inverted, from Florian
Westphal.
6) jme driver doesn't check for DMA mapping failures, from Neil
Horman.
7) qlogic driver doesn't calculate number of TX queues properly, from
Shahed Shaikh.
8) fib_info_cnt can drift irreversibly positive if we fail to
allocate the fi->fib_metrics array, from Sergey Popovich.
9) Fix use after free in ip6_route_me_harder(), also from Sergey
Popovich.
10) When SYSCTL is disabled, we don't handle local_port_range and
ping_group_range defaults properly at all, from Cong Wang.
11) Unaccelerated VLAN tagged frames improperly handled by cdc_mbim
driver, fix from Bjorn Mork.
12) cassini driver needs nested lock annotations for TX locking, from
Emil Goode.
13) On init error ipv6 VTI driver can unregister pernet ops twice,
oops. Fix from Mahtias Krause.
14) If macvlan device is down, don't propagate IFF_ALLMULTI changes,
from Peter Christensen.
15) Missing NULL pointer check while parsing netlink config options in
ip6_tnl_validate(). From Susant Sahani.
16) Fix handling of neighbour entries during ipv6 router reachability
probing, from Duan Jiong.
17) x86 and s390 JIT address randomization has some address
calculation bugs leading to crashes, from Alexei Starovoitov and
Heiko Carstens.
18) Clear up those uglies with nop patching and net_get_random_once(),
from Hannes Frederic Sowa.
19) Option length miscalculated in ip6_append_data(), fix also from
Hannes Frederic Sowa.
20) A while ago we fixed a race during device unregistry when a
namespace went down, turns out there is a second place that needs
similar protection. From Cong Wang.
21) In the new Altera TSE driver multicast filtering isn't working,
disable it and just use promisc mode until the cause is found.
From Vince Bridgers.
22) When we disable router enabling in ipv6 we have to flush the
cached routes explicitly, from Duan Jiong.
23) NBMA tunnels should not cache routes on the tunnel object because
the key is variable, from Timo Teräs.
24) With stacked devices GRO information in skb->cb[] can be not setup
properly, make sure it is in all code paths. From Eric Dumazet.
25) Really fix stacked vlan locking, multiple levels of nesting with
intervening non-vlan devices are possible. From Vlad Yasevich.
26) Fallback ipip tunnel device's mtu is not setup properly, from
Steffen Klassert.
27) The packet scheduler's tcindex filter can crash because we
structure copy objects with list_head's inside, oops. From Cong
Wang.
28) Fix CHECKSUM_COMPLETE handling for ipv6 GRE tunnels, from Eric
Dumazet.
29) In some configurations 'itag' in __mkroute_input() can end up
being used uninitialized because of how fib_validate_source()
works. Fix it by explitly initializing itag to zero like all the
other fib_validate_source() callers do, from Li RongQing"
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (116 commits)
batman: fix a bogus warning from batadv_is_on_batman_iface()
ipv4: initialise the itag variable in __mkroute_input
bonding: Send ALB learning packets using the right source
bonding: Don't assume 802.1Q when sending alb learning packets.
net: doc: Update references to skb->rxhash
stmmac: Remove unbalanced clk_disable call
ipv6: gro: fix CHECKSUM_COMPLETE support
net_sched: fix an oops in tcindex filter
can: peak_pci: prevent use after free at netdev removal
ip_tunnel: Initialize the fallback device properly
vlan: Fix build error wth vlan_get_encap_level()
can: c_can: remove obsolete STRICT_FRAME_ORDERING Kconfig option
MAINTAINERS: Pravin Shelar is Open vSwitch maintainer.
bnx2x: Convert return 0 to return rc
bonding: Fix alb mode to only use first level vlans.
bonding: Fix stacked device detection in arp monitoring
macvlan: Fix lockdep warnings with stacked macvlan devices
vlan: Fix lockdep warning with stacked vlan devices.
net: Allow for more then a single subclass for netif_addr_lock
net: Find the nesting level of a given device by type.
...
Daniel Borkmann [Fri, 23 May 2014 16:44:01 +0000 (18:44 +0200)]
net: filter: add test case for jump with holes and ret x variants
This patch adds three more test cases:
1) long jumps with holes of unreachable code
2) ret x
3) ldx + ret x
All three tests are for classical BPF and to make sure that
any changes will not break some exotic behaviour that exists
probably since decades. The last two tests are expected to
fail by the BPF checker already, as in classic BPF only K
or A are allowed to be returned. Thus, there are now 52 test
cases for BPF.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Acked-by: Alexei Starovoitov <ast@plumgrid.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann [Fri, 23 May 2014 16:44:00 +0000 (18:44 +0200)]
net: filter: improve test case framework
This patch simplifies and refactors the test case code a
bit and also adds a summary of all test that passed or
failed in the kernel log, so that it's easier to spot if
something has failed.
Future work could further extend the test framework to also
support different input 'stimuli' i.e. related structures
to seccomp.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Acked-by: Alexei Starovoitov <ast@plumgrid.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann [Fri, 23 May 2014 16:43:59 +0000 (18:43 +0200)]
net: filter: doc: add section for BPF test suite
Mention the recently added test suite in the documentation file.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Acked-by: Alexei Starovoitov <ast@plumgrid.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann [Fri, 23 May 2014 16:43:58 +0000 (18:43 +0200)]
net: filter: let unattached filters use sock_fprog_kern
The sk_unattached_filter_create() API is used by BPF filters that
are not directly attached or related to sockets, and are used in
team, ptp, xt_bpf, cls_bpf, etc. As such all users do their own
internal managment of obtaining filter blocks and thus already
have them in kernel memory and set up before calling into
sk_unattached_filter_create(). As a result, due to __user annotation
in sock_fprog, sparse triggers false positives (incorrect type in
assignment [different address space]) when filters are set up before
passing them to sk_unattached_filter_create(). Therefore, let
sk_unattached_filter_create() API use sock_fprog_kern to overcome
this issue.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Acked-by: Alexei Starovoitov <ast@plumgrid.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann [Fri, 23 May 2014 16:43:57 +0000 (18:43 +0200)]
net: filter: remove DL macro
Lets get rid of this macro. After commit 5bcfedf06f7f ("net: filter:
simplify label names from jump-table"), labels have become more
readable due to omission of BPF_ prefix but at the same time more
generic, so that things like `git grep -n` would not find them. As
a middle path, lets get rid of the DL macro as it's not strictly
needed and would otherwise just hide the full name.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Acked-by: Alexei Starovoitov <ast@plumgrid.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 23 May 2014 20:29:04 +0000 (16:29 -0400)]
Merge branch 'inet_csums_part3'
Tom Herbert says:
====================
net: Checksum offload changes - Part III
I am working on overhauling RX checksum offload. Goals of this effort
are:
- Specify what exactly it means when driver returns CHECKSUM_UNNECESSARY
- Preserve CHECKSUM_COMPLETE through encapsulation layers
- Don't do skb_checksum more than once per packet
- Unify GRO and non-GRO csum verification as much as possible
- Unify the checksum functions (checksum_init)
- Simply code
What is in this third patch set:
- Remove sk_no_check from sunrpc (doesn't seem to have any effect)
- Eliminate no_check from protosw. All protocols are using default of
zero for this
- Split sk_no_check into sk_no_check_tx and sk_no_check_rx
- Make enabling of UDP6 more restrictive and explicit
- Support zero UDP6 checksums in l2tp
V2: Took out vxlan changes to set zero csums in IPv6, this will
be in a later patch set.
V3: Fixed bug in restricting UDP6 checksums.
Please review carefully and test if possible, mucking with basic
checksum functions is always a little precarious :-)
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Tom Herbert [Fri, 23 May 2014 15:47:32 +0000 (08:47 -0700)]
net: Make enabling of zero UDP6 csums more restrictive
RFC 6935 permits zero checksums to be used in IPv6 however this is
recommended only for certain tunnel protocols, it does not make
checksums completely optional like they are in IPv4.
This patch restricts the use of IPv6 zero checksums that was previously
intoduced. no_check6_tx and no_check6_rx have been added to control
the use of checksums in UDP6 RX and TX path. The normal
sk_no_check_{rx,tx} settings are not used (this avoids ambiguity when
dealing with a dual stack socket).
A helper function has been added (udp_set_no_check6) which can be
called by tunnel impelmentations to all zero checksums (send on the
socket, and accept them as valid).
Signed-off-by: Tom Herbert <therbert@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Tom Herbert [Fri, 23 May 2014 15:47:19 +0000 (08:47 -0700)]
net: Split sk_no_check into sk_no_check_{rx,tx}
Define separate fields in the sock structure for configuring disabling
checksums in both TX and RX-- sk_no_check_tx and sk_no_check_rx.
The SO_NO_CHECK socket option only affects sk_no_check_tx. Also,
removed UDP_CSUM_* defines since they are no longer necessary.
Signed-off-by: Tom Herbert <therbert@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Tom Herbert [Fri, 23 May 2014 15:47:09 +0000 (08:47 -0700)]
net: Eliminate no_check from protosw
It doesn't seem like an protocols are setting anything other
than the default, and allowing to arbitrarily disable checksums
for a whole protocol seems dangerous. This can be done on a per
socket basis.
Signed-off-by: Tom Herbert <therbert@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 23 May 2014 20:28:18 +0000 (16:28 -0400)]
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next
Jeff Kirsher says:
====================
Intel Wired LAN Driver Updates
This series contains updates to igb, igbvf, ixgbe, i40e and i40evf.
Jacob provides eight patches to cleanup the ixgbe driver to resolve various
checkpatch.pl warnings/errors as well as minor coding style issues.
Stephen Hemminger and I provide simple cleanups of void functions which
had useless return statements at the end of the function which are not
needed.
v2: Dropped Emil's patch "ixgbe: fix the detection of SFP+ capable interfaces"
while I wait for his updated patch to be validated.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
New round for this assorted fixes and clean-up series. There is more room for
clean-ups, and I'll start preparing more patches once these are accepted.
This series consists of cleanups and minor improvements on mvneta, mv643xx_eth
and mvmdio drivers. None of the patches imply any functionality change, except
for the patch six "Change the number of default rx queues to one".
This patch reduces the driver's allocated resources and makes the multiqueue
path in the poll function not get taken. The previous patchset contains more
details:
Ezequiel Garcia [Thu, 22 May 2014 23:07:03 +0000 (20:07 -0300)]
net: mvneta: Remove unneeded 'weigth' field
The 'weight' field is only used to pass the weigth to napi initialization
function. This commit removes the field, and instead uses a fixed value to
initialize the napi context.
Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ezequiel Garcia [Thu, 22 May 2014 23:07:02 +0000 (20:07 -0300)]
net: mvmdio: Use devm_* API to simplify the code
This commit makes use of devm_kmalloc_array() for memory allocation and the
recently introduced devm_mdiobus_alloc() API to simplify driver's code.
While here, remove a redundant out of memory error message.
Acked-by: Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com> Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ezequiel Garcia [Thu, 22 May 2014 23:07:00 +0000 (20:07 -0300)]
net: mvneta: Use prepare/commit API to simplify MAC address setting
Use eth_prepare_mac_addr_change and eth_commit_mac_addr_change, instead
of manually checking and storing the MAC address, which makes the
code slightly more robust. This fixes the lack of valid MAC address check
in the driver's .ndo_set_mac_address hook.
Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ezequiel Garcia [Thu, 22 May 2014 23:06:59 +0000 (20:06 -0300)]
net: mvneta: Clean-up mvneta_init()
This commit cleans-up mvneta_init(), which initializes the hardware
and allocates the rx/qx queues. The queue allocation is simplified
by using devm_kcalloc instead of kzalloc. The unused phy_addr parameter
is removed. While here, the 'hal' references in the comments are removed.
This commit makes no functionality change.
Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ezequiel Garcia [Thu, 22 May 2014 23:06:58 +0000 (20:06 -0300)]
net: mvneta: Check tx queue setup error in mvneta_change_mtu()
This commit checks the return code of mvneta_setup_txq() call
in mvneta_change_mtu(). Also, use the netdevice pointer directly
instead of dereferencing the port structure. While here, let's
fix a tiny comment typo.
Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, mv643xx_eth_adjust_link() is only used to call mv643xx_adjust_pscr().
This commit renames the latter to the former, and therefore removes the extra
and useless function.
Acked-by: Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com> Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andrew Morton [Thu, 22 May 2014 17:16:46 +0000 (10:16 -0700)]
lib/test_bpf.c: don't use gcc union shortcut
Older gcc's (mine is gcc-4.4.4) make a mess of this.
lib/test_bpf.c:74: error: unknown field 'insns' specified in initializer
lib/test_bpf.c:75: warning: missing braces around initializer
lib/test_bpf.c:75: warning: (near initialization for 'tests[0].<anonymous>.insns[0]')
lib/test_bpf.c:76: error: extra brace group at end of initializer
lib/test_bpf.c:76: error: (near initialization for 'tests[0].<anonymous>')
lib/test_bpf.c:76: warning: excess elements in union initializer
lib/test_bpf.c:76: warning: (near initialization for 'tests[0].<anonymous>')
lib/test_bpf.c:77: error: extra brace group at end of initializer
Cc: Alexei Starovoitov <ast@plumgrid.com> Cc: David S. Miller <davem@davemloft.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Alexei Starovoitov <ast@plumgrid.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net-next:v4: Add support to configure SR-IOV VF minimum and maximum Tx rate through ip tool.
o min_tx_rate puts lower limit on the VF bandwidth. VF is guaranteed
to have a bandwidth of at least this value.
max_tx_rate puts cap on the VF bandwidth. VF can have a bandwidth
of up to this value.
o A new handler set_vf_rate for attr IFLA_VF_RATE has been introduced
which takes 4 arguments:
netdev, VF number, min_tx_rate, max_tx_rate
o ndo_set_vf_rate replaces ndo_set_vf_tx_rate handler.
o Drivers that currently implement ndo_set_vf_tx_rate should now call
ndo_set_vf_rate instead and reject attempt to set a minimum bandwidth
greater than 0 for IFLA_VF_TX_RATE when IFLA_VF_RATE is not yet
implemented by driver.
o If user enters only one of either min_tx_rate or max_tx_rate, then,
userland should read back the other value from driver and set both
for IFLA_VF_RATE.
Drivers that have not yet implemented IFLA_VF_RATE should always
return min_tx_rate as 0 when read from ip tool.
o If both IFLA_VF_TX_RATE and IFLA_VF_RATE options are specified, then
IFLA_VF_RATE should override.
o Idea is to have consistent display of rate values to user.
o Usage example: -
./ip link set p4p1 vf 0 rate 900
./ip link show p4p1
32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode
DEFAULT qlen 1000
link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff
vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 900 (Mbps), max_tx_rate 900Mbps
vf 1 MAC f6:c6:7c:3f:3d:6c
vf 2 MAC 56:32:43:98:d7:71
vf 3 MAC d6:be:c3:b5:85:ff
vf 4 MAC ee:a9:9a:1e:19:14
vf 5 MAC 4a:d0:4c:07:52:18
vf 6 MAC 3a:76:44:93:62:f9
vf 7 MAC 82:e9:e7:e3:15:1a
./ip link set p4p1 vf 0 max_tx_rate 300 min_tx_rate 200
./ip link show p4p1
32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode
DEFAULT qlen 1000
link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff
vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 300 (Mbps), max_tx_rate 300Mbps,
min_tx_rate 200Mbps
vf 1 MAC f6:c6:7c:3f:3d:6c
vf 2 MAC 56:32:43:98:d7:71
vf 3 MAC d6:be:c3:b5:85:ff
vf 4 MAC ee:a9:9a:1e:19:14
vf 5 MAC 4a:d0:4c:07:52:18
vf 6 MAC 3a:76:44:93:62:f9
vf 7 MAC 82:e9:e7:e3:15:1a
./ip link set p4p1 vf 0 max_tx_rate 600 rate 300
./ip link show p4p1
32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode
DEFAULT qlen 1000
link/ether 00:0e:1e:08:b0:f brd ff:ff:ff:ff:ff:ff
vf 0 MAC 3e:a0:ca:bd:ae:5, tx rate 600 (Mbps), max_tx_rate 600Mbps,
min_tx_rate 200Mbps
vf 1 MAC f6:c6:7c:3f:3d:6c
vf 2 MAC 56:32:43:98:d7:71
vf 3 MAC d6:be:c3:b5:85:ff
vf 4 MAC ee:a9:9a:1e:19:14
vf 5 MAC 4a:d0:4c:07:52:18
vf 6 MAC 3a:76:44:93:62:f9
vf 7 MAC 82:e9:e7:e3:15:1a
Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Haiyang Zhang [Wed, 21 May 2014 19:55:39 +0000 (12:55 -0700)]
hyperv: Add hash value into RNDIS Per-packet info
It passes the hash value as the RNDIS Per-packet info to the Hyper-V host,
so that the send completion notices can be spread across multiple channels.
MS-TFS: 140273
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Linus Torvalds [Fri, 23 May 2014 17:04:04 +0000 (10:04 -0700)]
Merge branch 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fixes from Ingo Molnar:
"The biggest commit is an irqtime accounting loop latency fix, the rest
are misc fixes all over the place: deadline scheduling, docs, numa,
balancer and a bad to-idle latency fix"
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/numa: Initialize newidle balance stats in sd_numa_init()
sched: Fix updating rq->max_idle_balance_cost and rq->next_balance in idle_balance()
sched: Skip double execution of pick_next_task_fair()
sched: Use CPUPRI_NR_PRIORITIES instead of MAX_RT_PRIO in cpupri check
sched/deadline: Fix memory leak
sched/deadline: Fix sched_yield() behavior
sched: Sanitize irq accounting madness
sched/docbook: Fix 'make htmldocs' warnings caused by missing description
Linus Torvalds [Fri, 23 May 2014 17:02:34 +0000 (10:02 -0700)]
Merge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Ingo Molnar:
"The biggest changes are fixes for races that kept triggering Trinity
crashes, plus liblockdep build fixes and smaller misc fixes.
The liblockdep bits in perf/urgent are a pull mistake - they should
have been in locking/urgent - but by the time I noticed other commits
were added and testing was done :-/ Sorry about that"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf: Fix a race between ring_buffer_detach() and ring_buffer_attach()
perf: Prevent false warning in perf_swevent_add
perf: Limit perf_event_attr::sample_period to 63 bits
tools/liblockdep: Remove all build files when doing make clean
tools/liblockdep: Build liblockdep from tools/Makefile
perf/x86/intel: Fix Silvermont's event constraints
perf: Fix perf_event_init_context()
perf: Fix race in removing an event
Linus Torvalds [Fri, 23 May 2014 16:41:33 +0000 (09:41 -0700)]
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Pull drm radeon and nouveau fixes from Dave Airlie:
"Fixes for the other big two.
The radeon VCE one is large but it fixes some userspace triggerable
issues, otherwise its blackscreens and oopses.
Nouveau fixes a bleeding laptop panel issue when displayport is used
sometimes"
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
drm/radeon/pm: don't allow debugfs/sysfs access when PX card is off (v2)
drm/radeon: avoid segfault on device open when accel is not working.
drm/radeon: fix typo in finding PLL params
drm/radeon: fix register typo on si
drm/radeon: fix buffer placement under memory pressure v2
drm/radeon: fix page directory update size estimation
drm/radeon: handle non-VGA class pci devices with ATRM
drm/radeon: fix DCE83 check for mullins
drm/radeon: check VCE relocation buffer range v3
drm/radeon: also try GART for CPU accessed buffers
drm/gf119-/disp: fix nasty bug which can clobber SOR0's clock setup
drm/nvd9/therm: handle another kind of PWM fan
Linus Torvalds [Fri, 23 May 2014 16:38:07 +0000 (09:38 -0700)]
Merge branch 'akpm' (incoming from Andrew)
Merge misc fixes from Andrew Morton:
"9 fixes"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
MAINTAINERS: add closing angle bracket to Vince Bridgers' email address
Documentation: fix DOCBOOKS=... building
ocfs2: fix double kmem_cache_destroy in dlm_init
mm/memory-failure.c: fix memory leak by race between poison and unpoison
wait: swap EXIT_ZOMBIE(Z) and EXIT_DEAD(X) chars in TASK_STATE_TO_CHAR_STR
memcg: fix swapcache charge from kernel thread context
mm: madvise: fix MADV_WILLNEED on shmem swapouts
mm/filemap.c: avoid always dirtying mapping->flags on O_DIRECT
hwpoison, hugetlb: lock_page/unlock_page does not match for handling a free hugepage
Johannes Berg [Thu, 22 May 2014 18:54:23 +0000 (11:54 -0700)]
Documentation: fix DOCBOOKS=... building
Prior to commit 4266129964b8 ("[media] DocBook: Move all media docbook
stuff into its own directory") it was possible to build only a single
(or more) book(s) by calling, for example
make htmldocs DOCBOOKS=80211.xml
This now fails:
cp: target `.../Documentation/DocBook//media_api' is not a directory
Ignore errors from that copy to make this possible again.
Fixes: 4266129964b8 ("[media] DocBook: Move all media docbook stuff into its own directory") Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Randy Dunlap <rdunlap@xenotime.net> Cc: Mauro Carvalho Chehab <mchehab@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Joseph Qi [Thu, 22 May 2014 18:54:22 +0000 (11:54 -0700)]
ocfs2: fix double kmem_cache_destroy in dlm_init
In dlm_init, if create dlm_lockname_cache failed in
dlm_init_master_caches, it will destroy dlm_lockres_cache which created
before twice. And this will cause system die when loading modules.
Signed-off-by: Joseph Qi <joseph.qi@huawei.com> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Naoya Horiguchi [Thu, 22 May 2014 18:54:21 +0000 (11:54 -0700)]
mm/memory-failure.c: fix memory leak by race between poison and unpoison
When a memory error happens on an in-use page or (free and in-use)
hugepage, the victim page is isolated with its refcount set to one.
When you try to unpoison it later, unpoison_memory() calls put_page()
for it twice in order to bring the page back to free page pool (buddy or
free hugepage list). However, if another memory error occurs on the
page which we are unpoisoning, memory_failure() returns without
releasing the refcount which was incremented in the same call at first,
which results in memory leak and unconsistent num_poisoned_pages
statistics. This patch fixes it.
Masatake YAMATO [Thu, 22 May 2014 18:54:20 +0000 (11:54 -0700)]
wait: swap EXIT_ZOMBIE(Z) and EXIT_DEAD(X) chars in TASK_STATE_TO_CHAR_STR
In commit ad86622b478e ("wait: swap EXIT_ZOMBIE and EXIT_DEAD to hide
EXIT_TRACE from user-space") the order of task state definitions were
changed: EXIT_DEAD and EXIT_ZOMBIE were swapped. Though the charterers
for the states in TASK_STATE_TO_CHAR_STR string were not updated. This
patch synchronizes the string to the order of definitions.
Michal Hocko [Thu, 22 May 2014 18:54:19 +0000 (11:54 -0700)]
memcg: fix swapcache charge from kernel thread context
Commit 284f39afeaa4 ("mm: memcg: push !mm handling out to page cache
charge function") explicitly checks for page cache charges without any
mm context (from kernel thread context[1]).
This seemed to be the only possible case where memory could be charged
without mm context so commit 03583f1a631c ("memcg: remove unnecessary
!mm check from try_get_mem_cgroup_from_mm()") removed the mm check from
get_mem_cgroup_from_mm(). This however caused another NULL ptr
dereference during early boot when loopback kernel thread splices to
tmpfs as reported by Stephan Kulow:
This patch fixes the issue by reintroducing mm check into
get_mem_cgroup_from_mm. We could do the same trick in
__mem_cgroup_try_charge_swapin as we do for the regular page cache path
but it is not worth troubles. The check is not that expensive and it is
better to have get_mem_cgroup_from_mm more robust.
Johannes Weiner [Thu, 22 May 2014 18:54:17 +0000 (11:54 -0700)]
mm: madvise: fix MADV_WILLNEED on shmem swapouts
MADV_WILLNEED currently does not read swapped out shmem pages back in.
Commit 0cd6144aadd2 ("mm + fs: prepare for non-page entries in page
cache radix trees") made find_get_page() filter exceptional radix tree
entries but failed to convert all find_get_page() callers that WANT
exceptional entries over to find_get_entry(). One of them is shmem swap
readahead in madvise, which now skips over any swap-out records.
Convert it to find_get_entry().
Fixes: 0cd6144aadd2 ("mm + fs: prepare for non-page entries in page cache radix trees") Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reported-by: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Jens Axboe [Thu, 22 May 2014 18:54:16 +0000 (11:54 -0700)]
mm/filemap.c: avoid always dirtying mapping->flags on O_DIRECT
In some testing I ran today (some fio jobs that spread over two nodes),
we end up spending 40% of the time in filemap_check_errors(). That
smells fishy. Looking further, this is basically what happens:
blkdev_aio_read()
generic_file_aio_read()
filemap_write_and_wait_range()
if (!mapping->nr_pages)
filemap_check_errors()
and filemap_check_errors() always attempts two test_and_clear_bit() on
the mapping flags, thus dirtying it for every single invocation. The
patch below tests each of these bits before clearing them, avoiding this
issue. In my test case (4-socket box), performance went from 1.7M IOPS
to 4.0M IOPS.
Signed-off-by: Jens Axboe <axboe@fb.com> Acked-by: Jeff Moyer <jmoyer@redhat.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Chen Yucong [Thu, 22 May 2014 18:54:15 +0000 (11:54 -0700)]
hwpoison, hugetlb: lock_page/unlock_page does not match for handling a free hugepage
For handling a free hugepage in memory failure, the race will happen if
another thread hwpoisoned this hugepage concurrently. So we need to
check PageHWPoison instead of !PageHWPoison.
If hwpoison_filter(p) returns true or a race happens, then we need to
unlock_page(hpage).
Linus Torvalds [Fri, 23 May 2014 16:23:51 +0000 (09:23 -0700)]
parisc: 'renameat2()' doesn't need (or have) a separate compat system call
The 'renameat2()' system call was incorrectly added as a ENTRY_COMP() in
the parisc system call table by commit 18e480aa07f78 ("parisc: add
renameat2 syscall"). That causes a link-time error due to there not
being any compat version of that system call:
arch/parisc/kernel/built-in.o: In function `sys_call_table':
(.rodata+0xad0): undefined reference to `compat_sys_renameat2'
make: *** [vmlinux] Error 1
Easily fixed by marking the system call as being the same for compat as
for native by using ENTRY_SAME() instead of ENTRY_COMP().
Remove cases where useless bare return is left at end of function.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jeff Kirsher [Wed, 14 May 2014 08:01:09 +0000 (01:01 -0700)]
igb/ixgbe: remove return statements for void functions
Remove useless return statements for void functions which do not need
it.
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Jacob Keller [Wed, 9 Apr 2014 06:03:17 +0000 (06:03 +0000)]
ixgbe: add /* fallthrough */ comment to case statements
This semicomplex switch-case has various fallthrough portions, that were
not indicated by a /* fallthrough */ comment.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Wed, 9 Apr 2014 06:03:16 +0000 (06:03 +0000)]
ixgbe: add space between operands to &
This patch cleans up a checkpatch.pl style warning in the ixgbe code.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Wed, 9 Apr 2014 06:03:15 +0000 (06:03 +0000)]
ixgbe: don't check NULL for debugfs_remove_recursive
The debugfs_remove_recursive function is NULL-safe, so we don't need to
check here ourselves.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Wed, 9 Apr 2014 06:03:14 +0000 (06:03 +0000)]
ixgbe: add braces around else block
This commit fixes a checkpatch.pl warning for style, by adding braces
around the else block, since the if block requires braces.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Wed, 9 Apr 2014 06:03:13 +0000 (06:03 +0000)]
ixgbe: fix several concatenated strings to single line
This patch fixes various log strings that are split over multiple lines
in the ixgbe driver. This cleans up checkpatch.pl warnings, and makes it
easier to search the code for warning strings displayed to the kernel
log.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Wed, 9 Apr 2014 06:03:12 +0000 (06:03 +0000)]
ixgbe: fix checkpatch style of blank line after declaration
This patch fixes checkpatch warnings in ixgbe, by adding a blank line
between declaration and code blocks.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Wed, 9 Apr 2014 06:03:11 +0000 (06:03 +0000)]
ixgbe: fix function-like macro, remove semicolon
This patch removes the semicolon from the end of the do-while(0)
construct in two function-like macros.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Jacob Keller [Wed, 9 Apr 2014 06:03:10 +0000 (06:03 +0000)]
ixgbe: clean up checkpatch warnings about CODE_INDENT and LEADING_SPACE
The contents of this patch were originally generated by
"scripts/checkpatch.pl --fix-inplace --types CODE_INDENT,LEADING_SPACE
drivers/net/ethernet/ixgbe/*.[ch]", and then hand verified for
consistency.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
ovs_flow_cmd_del() now allocates reply (if needed) after the flow has
already been removed from the flow table. If the reply allocation
fails, a netlink error is signaled with netlink_set_err(), as is
already done in ovs_flow_cmd_new_or_set() in the similar situation.
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com> Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
Jarno Rajahalme [Mon, 5 May 2014 21:28:07 +0000 (14:28 -0700)]
openvswitch: Reduce locking requirements.
Reduce and clarify locking requirements for ovs_flow_cmd_alloc_info(),
ovs_flow_cmd_fill_info() and ovs_flow_cmd_build_info().
A datapath pointer is available only when holding a lock. Change
ovs_flow_cmd_fill_info() and ovs_flow_cmd_build_info() to take a
dp_ifindex directly, rather than a datapath pointer that is then
(only) used to get the dp_ifindex. This is useful, since the
dp_ifindex is available even when the datapath pointer is not, both
before and after taking a lock, which makes further critical section
reduction possible.
Make ovs_flow_cmd_alloc_info() take an 'acts' argument instead a
'flow' pointer. This allows some future patches to do the allocation
before acquiring the flow pointer.
The locking requirements after this patch are:
ovs_flow_cmd_alloc_info(): May be called without locking, must not be
called while holding the RCU read lock (due to memory allocation).
If 'acts' belong to a flow in the flow table, however, then the
caller must hold ovs_mutex.
ovs_flow_cmd_fill_info(): Either ovs_mutex or RCU read lock must be held.
ovs_flow_cmd_build_info(): This calls both of the above, so the caller
must hold ovs_mutex.
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com> Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
projects / linux-beck.git / log