David S. Miller [Mon, 28 Jul 2014 05:34:49 +0000 (22:34 -0700)]
Merge branch 'inet_frag_kill_lru_list'
Nikolay Aleksandrov says:
====================
inet: frag: cleanup and update
The end goal of this patchset is to remove the LRU list and to move the
frag eviction to a work queue. It also does a couple of necessary cleanups
and fixes. Brief patch descriptions:
Patches 1 - 3 inclusive: necessary clean ups
Patch 4 moves the eviction from the softirqs to a workqueue.
Patch 5 removes the nqueues counter which was protected by the LRU lock
Patch 6 removes the, by now unused, lru list.
Patch 7 moves the rebuild timer to the workqueue and schedules the rebuilds
only if we've hit the maximum queue length on some of the chains.
Patch 8 migrate the rwlock to a seqlock since the rehash is usually a rare
operation.
Patch 9 introduces an artificial global memory limit based on the value of
init_net's high_thresh which is used to cap the high_thresh of the
other namespaces. Also introduces some sane limits on the other
tunables, and makes it impossible to have low_thresh > high_thresh.
Here are some numbers from running netperf before and after the patchset:
Each test consists of the following setting: -I 95,5 -i 15,10
1. Bound test (-T 4,4)
1.1 Virtio before the patchset -
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.122.177 () port 0 AF_INET : +/-2.500% @ 95% conf. : cpu bind
Socket Message Elapsed Messages CPU Service
Size Size Time Okay Errors Throughput Util Demand
bytes bytes secs # # 10^6bits/sec % SS us/KB
1.2 Virtio after the patchset -
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.122.177 () port 0 AF_INET : +/-2.500% @ 95% conf. : cpu bind
Socket Message Elapsed Messages CPU Service
Size Size Time Okay Errors Throughput Util Demand
bytes bytes secs # # 10^6bits/sec % SS us/KB
2. Virtio unbound test
2.1 Before the patchset
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.122.177 () port 0 AF_INET : +/-2.500% @ 95% conf.
Socket Message Elapsed Messages
Size Size Time Okay Errors Throughput
bytes bytes secs # # 10^6bits/sec
Given the options each netperf ran between 10 and 15 times for 30 seconds
to get the necessary confidence, also the tests themselves ran 3 times and
were consistent.
Another set of tests that I ran were parallel stress tests which consisted
of flooding the machine with fragmented packets from different sources with
frag timeout set to 0 (so there're lots of timeouts) and low_thresh set to
1 byte (so evictions are happening all the time) and on top of that running
a namespace create/destroy endless loop with network interfaces and
addresses that got flooded (for the brief periods they were up) in parallel.
This test ran for an hour without any issues.
====================
inet: frag: set limits and make init_net's high_thresh limit global
This patch makes init_net's high_thresh limit to be the maximum for all
namespaces, thus introducing a global memory limit threshold equal to the
sum of the individual high_thresh limits which are capped.
It also introduces some sane minimums for low_thresh as it shouldn't be
able to drop below 0 (or > high_thresh in the unsigned case), and
overall low_thresh should not ever be above high_thresh, so we make the
following relations for a namespace:
init_net:
high_thresh - max(not capped), min(init_net low_thresh)
low_thresh - max(init_net high_thresh), min (0)
all other namespaces:
high_thresh = max(init_net high_thresh), min(namespace's low_thresh)
low_thresh = max(namespace's high_thresh), min(0)
The major issue with having low_thresh > high_thresh is that we'll
schedule eviction but never evict anything and thus rely only on the
timers.
Instead of rebuilding every n seconds, take advantage of the upper
hash chain length limit.
If we hit it, mark table for rebuild and schedule workqueue.
To prevent frequent rebuilds when we're completely overloaded,
don't rebuild more than once every 5 seconds.
ipfrag_secret_interval sysctl is now obsolete and has been marked as
deprecated, it still can be changed so scripts won't be broken but it
won't have any effect. A comment is left above each unused secret_timer
variable to avoid confusion.
inet: frag: don't account number of fragment queues
The 'nqueues' counter is protected by the lru list lock,
once thats removed this needs to be converted to atomic
counter. Given this isn't used for anything except for
reporting it to userspace via /proc, just remove it.
We still report the memory currently used by fragment
reassembly queues.
When the high_thresh limit is reached we try to toss the 'oldest'
incomplete fragment queues until memory limits are below the low_thresh
value. This happens in softirq/packet processing context.
This has two drawbacks:
1) processors might evict a queue that was about to be completed
by another cpu, because they will compete wrt. resource usage and
resource reclaim.
2) LRU list maintenance is expensive.
But when constantly overloaded, even the 'least recently used' element is
recent, so removing 'lru' queue first is not 'fairer' than removing any
other fragment queue.
This moves eviction out of the fast path:
When the low threshold is reached, a work queue is scheduled
which then iterates over the table and removes the queues that exceed
the memory limits of the namespace. It sets a new flag called
INET_FRAG_EVICTED on the evicted queues so the proper counters will get
incremented when the queue is forcefully expired.
When the high threshold is reached, no more fragment queues are
created until we're below the limit again.
The LRU list is now unused and will be removed in a followup patch.
Li RongQing [Fri, 25 Jul 2014 06:45:11 +0000 (14:45 +0800)]
ipv6: remove obsolete comment in ip6_append_data()
After 11878b40e[net-timestamp: SOCK_RAW and PING timestamping], this comment
becomes obsolete since the codes check not only UDP socket, but also RAW sock;
and the codes are clear, not need the comments
David S. Miller [Fri, 25 Jul 2014 06:40:00 +0000 (23:40 -0700)]
Merge branch 'macb-next'
Cyrille Pitchen says:
====================
net/macb: add HW features to macb driver
this series of patches adds new hardware features to macb driver. These
features can be enabled/disabled at runtime using ethtool. Depending on
hardware and design configuration, some are enabled by default whereas other
are disabled.
For instance, checksum offload features are enabled by default for gem designed
for packet buffer mode but disabled for fifo mode design or for old macb.
Besides, the scatter-gather feature is enabled and tested on macb but disabled
on sama5d3x gem. When testing this feature on sama5d3x gem, TX lockups occured
frequently.
Also, the RX checksum offload feature is enabled at GEM level only when both
IFF_PROMISC bit is clear in dev->flags and NETIF_F_RXCSUM bit is set in
dev->features.
====================
When RX checksum offload is enabled at GEM level (bit 24 set in the Network
Control Register), frames with invalid IP, TCP or UDP checksums are
discarted even if promiscuous mode is enabled (bit 4 set in the Network Control
Register).
This was verified with a simple userspace program, which corrupts UDP checksum
using libnetfilter_queue.
Then both IFF_PROMISC bit must be clear in dev->flags and NETIF_F_RXCSUM bit
must be set in dev->features to enable RX checksum offload at GEM level. This
way tcpdump is still able to capture corrupted frames.
Also skb->ip_summed is set to CHECKSUM_UNNECESSARY only when both TCP/IP or
UDP/IP checksums were verified by the GEM. Indeed the GEM may verify only IP
checksum but not the one for ICMP (or other protocol than TCP or UDP).
The scatter-gather feature will allow to enable the Generic Segmentation Offload.
Generic Segmentation Offload can be enabled/disabled using ethtool -K DEVNAME gso on|off.
e.g:
ethtool -K eth0 gso off
When enabled, the driver may be provided with socket buffers splitted into many fragments.
These fragments need to be queued into the TX ring in reverse order, starting from to the
last one down to the first one, to avoid a race condition with the MAC.
Especially the 'TX_USED' bit in word 1 of the transmit buffer descriptor of the
first fragment should be cleared at the very final step of the queueing algorithm.
This will tell the hardware that fragments are ready to be sent.
Also since the MAC only update the status word of the first buffer descriptor of the
ethernet frame, the queueing algorithm can no longer expect a 'TX_USED' bit to be set by
the MAC into the buffer descriptor following the one for last fragment of the skb.
This is why the driver sets the 'TX_USED' bit before queueing any fragment, so the end of
queue position is well defined for the MAC.
cxgb4: Fixed incorrect check for memory operation in t4_memory_rw
Fix incorrect check introduced in commit fc5ab020 ("cxgb4: Replaced the
backdoor mechanism to access the HW memory with PCIe Window method"). We where
checking for write operation and doing a read, changed it accordingly.
net: filter: rename 'struct sock_filter_int' into 'struct bpf_insn'
eBPF is used by socket filtering, seccomp and soon by tracing and
exposed to userspace, therefore 'sock_filter_int' name is not accurate.
Rename it to 'bpf_insn'
David S. Miller [Fri, 25 Jul 2014 06:18:21 +0000 (23:18 -0700)]
Merge branch 'net_next_ovs' of git://git.kernel.org/pub/scm/linux/kernel/git/pshelar/openvswitch
Pravin B Shelar says:
====================
Open vSwitch
Following patches adds three features to OVS
1. Add fairness to upcall processing.
2. Hash action.
3. Enable Tunnel GSO features.
Rest of patches are bug fixes related to patches from same series.
v2 series changes first patch according to comment from Dave Miller.
v3 series changes first patch according to comment from Nikolay Aleksandrov.
v4 series update recirc patch commit msg.
v5 series resolve conflict with net-next, updated recic action patch.
v6 series sends all patches.
v7 series drop recirc patches.
v8 series checkpatch fix
v9 series drop HASH action patch. update sample action commit msg.
====================
David S. Miller [Fri, 25 Jul 2014 06:08:12 +0000 (23:08 -0700)]
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next
Jeff Kirsher says:
====================
Intel Wired LAN Driver Updates 2014-07-24
This series contains updates to igb, ixgbe, i40e and i40evf.
Mark fixes a possible attempt to dereference a NULL pointer in ixgbe_probe().
Also changes some uses of strncpy to strlcpy when clearing is not needed to
prevent information leakage.
Jacob fixes a bug in the misuse of the list_for_each macro to loop over
every entry in the bus_list. Instead of attempting to loop over the list
from a random entry point, go up to the bus and use the real list_head
entry point. This prevents the possible read or write of unallocated or
incorrectly addressed memory. Then provides a patch to prevent the
display of the minimum link qualification check if we might be in a
virtual machine. This check is incorrect and misleading in this case,
since we actually do not really know what the available bandwidth is.
To do so, we simply check whether each function on the bus matches our
device id.
Carolyn adds a check and prints the error cause register value when the
hardware detects a malformed packet to assist the user.
Toralf Förster fixes a format mismatch in i40e which was found using
cppcheck.
Shannon adds nvmupdate support by implementing a state machine intended
to support the userland tool for updating the device eeprom.
Jesse fixes the extension header checksum logic for IPv6 in i40e and
i40evf.
Mitch reduces a delay in the i40evf driver where we do not need to
delay an entire millisecond to get into our critical section.
Kamil fixes an issue where access to the NVM was being blocked until
a driver reset where a check for NVM related admin queue commands
would not recognize that such a command was received and would not clear
nvm_busy flag.
Catherine fixes a couple of firmware API version errors.
====================
Simon Horman [Mon, 21 Jul 2014 22:12:34 +0000 (15:12 -0700)]
openvswitch: Sample action without side effects
The sample action is rather generic, allowing arbitrary actions to be
executed based on a probability. However its use, within the Open
vSwitch
code-base is limited: only a single user-space action is ever nested.
A consequence of the current implementation of sample actions is that
depending on weather the sample action executed (due to its probability)
any side-effects of nested actions may or may not be present before
executing subsequent actions. This has the potential to complicate
verification of valid actions by the (kernel) datapath. And indeed
adding support for push and pop MPLS actions inside sample actions
is one case where such case.
In order to allow all supported actions to be continue to be nested
inside sample actions without the potential need for complex
verification code this patch changes the implementation of the sample
action in the kernel datapath so that sample actions are more like
a function call and any side effects of nested actions are not
present when executing subsequent actions.
With the above in mind the motivation for this change is twofold:
* To contain side-effects the sample action in the hope of making it
easier to deal with in the future and;
* To avoid some rather complex verification code introduced in the MPLS
datapath patch.
Andy Zhou [Thu, 17 Jul 2014 22:17:44 +0000 (15:17 -0700)]
openvswitch: Avoid memory corruption in queue_userspace_packet()
In queue_userspace_packet(), the ovs_nla_put_flow return value is
not checked. This is fine as long as key_attr_size() returns the
correct value. In case it does not, the current code may corrupt buffer
memory. Add a run time assertion catch this case to avoid silent
failure.
i40e: Give link more time after setting flow control
Give link a little more time to come back up after setting flow control
before resetting. In the new NVMs it is taking longer for link to come back.
This causes the driver to attempt to reset the link, which then errors
because the firmware was already in the middle of a reset. Also, initialize
err to 0.
Reword the error messages. Also add a major version check because
We only want to warn on nvm_minor > expected_minor if
nvm_major == expected_major. Lastly, change an if to an else if
because the two statements will never evaluate to true at the same time.
i40e/i40evf: ARQ copy desc data even for failed commands
Copy desc and buffer data even for ARQ events which return error status.
Previously, a check for NVM related AQ commands which is done later in this
function would not recognize that such a command was received and would
not clear nvm_busy flag. This would block access to NVM until a driver reset.
This will fix that.
The hardware design requires that the driver avoid indicating
checksum offload success on some ipv6 frames with extension
headers.
The code needs to just check for the IPV6EXADD bit and if
it is set punt the checksum to the stack. I don't know why
the code was checking TCP on inner protocol, as that code
doesn't make any sense to me but seems wrong, so remove it.
Shannon Nelson [Wed, 9 Jul 2014 07:46:09 +0000 (07:46 +0000)]
i40e/i40evf: Add nvmupdate support
This implements a state machine intended to support the userland tool for
updating the device eeprom. The state machine implements one-shot reads,
writes, multi-step write sessions, and checksum requests. If we're in the middle
of a multi-step write session, no one should fire off other writes, however, one
shot reads are valid. The userland tool is expected to keep track of its session
status, arrange the placement and ordering of the writes, and deal with the
checksum requirement.
This patch also adds nvmupdate support to ethtool callbacks.
The get_eeprom() and set_eeprom() services in ethtool are used here to
facilitate the userland NVMUpdate tool. The 'magic' value in the get and
set commands is used to pass additional control information for managing
the read and write steps.
The read operation works both as normally expected in the standard ethtool
method, as well as with the extra NVM controls. The write operation
works only for the expanded NVM functions - the normal ethtool method is
not allowed because of the NVM semaphore management needed for multipart
writes, as well as the checksum requirement.
igb: Add message when malformed packets detected by hw
This patch adds a check and prints the error cause register value when
the hardware detects a malformed packet. This is a very unlikely
scenario but has been seen occasionally, so printing the message to
assist the user.
Jacob Keller [Sat, 19 Jul 2014 07:17:17 +0000 (07:17 +0000)]
ixgbe: don't check minimum link when direct assigned to virtual machine
This patch prevents the display of the minimum link qualification check
if we might be in a virtual machine. This check is incorrect and
misleading in this case, since we actually don't really know what the
available bandwidth is. To do so, we simply check whether each function
on the bus matches our device id. If it doesn't the most likely scenario
is that we're directly assigned to a virtual machine.
Jacob Keller [Sat, 19 Jul 2014 07:17:16 +0000 (07:17 +0000)]
ixgbe: fix use of list_for_each in ixgbe_enumerate_functions
Fix a bug in the misuse of the list_for_each macro to loop over every
entry in the bus_list. Instead of attempting to loop over the list from
a random entry point, go up to the bus and use the real list_head entry
point. This prevents the possible read or write of unallocated or
incorrectly addressed memory.
Mark Rustad [Fri, 6 Jun 2014 01:57:06 +0000 (01:57 +0000)]
ixgbe: Change some uses of strncpy to strlcpy
Change some uses of strncpy to use the more appropriate strlcpy
when clearing is not needed to prevent information leakage. Also
change some length arguments to use the preferred sizeof form.
Alex Wang [Thu, 17 Jul 2014 22:14:13 +0000 (15:14 -0700)]
openvswitch: Allow each vport to have an array of 'port_id's.
In order to allow handlers directly read upcalls from datapath,
we need to support per-handler netlink socket for each vport in
datapath. This commit makes this happen. Also, it is guaranteed
to be backward compatible with previous branch.
Mark Rustad [Fri, 6 Jun 2014 01:57:00 +0000 (01:57 +0000)]
ixgbe: Fix possible null-dereference in error path
In ixgbe_probe, the code at label err_dma can dereference adapter
when it has a NULL value. The check is there to avoid disabling a
disabled device. When adapter is NULL, treat it as if the device
is enabled, because it is enabled in that case.
David S. Miller [Thu, 24 Jul 2014 04:06:30 +0000 (21:06 -0700)]
Merge branch 'filter-move'
Alexei Starovoitov says:
====================
I believe my recent set of RFC/patches [1] provided good visibility on where
I would like to take eBPF subsystem. These two trivial patches is a first step
in that direction:
patch 1 - mechanical split of eBPF interpreter out of filter.c
patch 2 - nominate myself as a maintainer for eBPF core pieces
In the foreseeable future eBPF patches will be going through net-next,
so put netdev as a primary mailing list
This patch series contains cleanups for CHECK and WARNINGS reported by
checkpatch.pl. I removed one patch from this series since Joe reported
this was a false positive due to me not using the latest version.
====================
checkpatch.pl flagged a lot of "CHECK: Alignment should match open
parenthesis" checks, fix all of them to make the driver neater. While
at it fix some obvious typos and re-arrange some of the lines to avoid
going over 80 columns.
ipv4: Make IP_MULTICAST_ALL and IP_MSFILTER work on raw sockets
Currently, although IP_MULTICAST_ALL and IP_MSFILTER ioctl calls succeed on
raw sockets, there is no code to implement the functionality on received
packets; it is only implemented for UDP sockets. The raw(7) man page states:
"In addition, all ip(7) IPPROTO_IP socket options valid for datagram sockets
are supported", which implies these ioctls should work on raw sockets.
To fix this, add a call to ip_mc_sf_allow on raw sockets.
This should not break any existing code, since the current position of
not calling ip_mc_sf_filter makes it behave as if neither the IP_MULTICAST_ALL
nor the IP_MSFILTER ioctl had been called. Adding the call to ip_mc_sf_allow
will therefore maintain the current behaviour so long as IP_MULTICAST_ALL and
IP_MSFILTER ioctls are not called. Any code that currently is calling
IP_MULTICAST_ALL or IP_MSFILTER ioctls on raw sockets presumably is wanting
the filter to be applied, although no filtering will currently be occurring.
David S. Miller [Wed, 23 Jul 2014 22:12:32 +0000 (15:12 -0700)]
Merge branch 'virtio-net-busy-poll'
Jason Wang says:
====================
rx busy polling support for virtio-net
This series introduces the support for rx busy polling support. This
was useful for reducing the latency for a kvm guest. Instead of
introducing new states and spinlocks, this series re-uses NAPI state
to synchonrize between NAPI and busy polling. This grealy simplified
the codes and reduce the overheads of spinlocks for normal NAPI fast
path.
Test was done between a kvm guest and an external host. Two hosts were
connected through 40gb mlx4 cards. With both busy_poll and busy_read
are set to 50 in guest, 1 byte netperf tcp_rr shows 127% improvement:
transaction rate was increased from 8353.33 to 18966.87.
Changes from V2:
- Avoid introducing new states and spinlocks by reusuing the NAPI
state
- Fix the budget calculation in virtnet_poll()
- Drop patch 1/3 from V2 since it was useless
Changes from V1:
- split the patch info smaller ones
- add more details about test setup/configuration
====================
Jason Wang [Wed, 23 Jul 2014 08:33:55 +0000 (16:33 +0800)]
virtio-net: rx busy polling support
Add basic support for rx busy polling. Instead of introducing new
states and spinlock to synchronize between NAPI and polling method,
this patch just reuse NAPI state to avoid extra overhead for fast path
and simplified the codes.
Test was done between a kvm guest and an external host. Two hosts were
connected through 40gb mlx4 cards. With both busy_poll and busy_read
are set to 50 in guest, 1 byte netperf tcp_rr shows 127% improvement:
transaction rate was increased from 8353.33 to 18966.87.
This patch use the struct pci_device_id instead of using macro
DEFINE_PCI_DEVICE_TABLE which is deprecated and should not be used.
And also moves these ids after probe and remove functionalities.
This patch converts to use the macro module_pci_driver, which makes
the code smaller and simpler.
Previously in this driver we are having driver version info will be
printed log buffer based on whether the driver selected as module
or statically into image itself. By using the module_pci_driver that
part of the code removed. For the first time of the device init,
we are making the version info to be printed once.
Stefan Sørensen [Tue, 22 Jul 2014 13:20:45 +0000 (15:20 +0200)]
dp83640: Fix receive timestamp race condition
When timestamping received packets, rx_timestamp_work may be scheduled
before the timestamps is received from the hardware resulting in the
packet beeing delivered without the timestamp.
This is fixed by changing the receive timestamp path:
On receiving a packet that need timestamping, the rxts list is
traversed. If a match is found, packet+timestamp are delivered,
otherwise the packet is added to a rx_queue.
When a timestamp arrives rx_queue is traversed and if a matching
packet is found, it is delivered with the timestamp. Otherwise the
timestamp is added to the rxts list for matching with packets arriving
later.
In case the hardware drops a timestamp, a workqueue regularly checks
the queue for old packets and delivers them without a timestamp.
This patch adds the sfc driver code for implementing busy polling.
It adds ndo_busy_poll method and locking between it and napi poll.
It also adds each napi to the napi_hash right after netif_napi_add().
Uses efx_start_eventq and efx_stop_eventq in the self tests.
This patchset contains a fix by Eyal to fix a bug in mlx4_en with bonding
alb/tlb mode.
The second patch, by me, is an ethtool private flag that enable turning off the
BlueFlame hardware feature.
The last two patches, introduce a new low memory profile for mlx4_en and
mlx4_core. This mode is useful when running from a limited resources
environment, like kdump. This will reduce the memory consumed by both drivers
from more than 100M to ~30M.
This patchset was applied and tested on commit 575a193 ("net: mvpp2: Fix error
return code in mvpp2_probe")
====================
Amir Vadai [Tue, 22 Jul 2014 12:44:11 +0000 (15:44 +0300)]
net/mlx4_core: Use low memory profile on kdump kernel
When running in kdump kernel, reduce number of resources allocated for
the hardware. This will enable the NIC to operate in this low memory
environment at the expense of performance and some features not related
to the basic NIC functionality.
Amir Vadai [Tue, 22 Jul 2014 12:44:10 +0000 (15:44 +0300)]
net/mlx4_en: Disable blueflame using ethtool private flags
Enable the user to turn off the hardware feature called BlueFlame.
Since it is something specific to mlx4_en hardware, we control
the feature via ethtool private flags.
Eyal Perry [Tue, 22 Jul 2014 12:44:09 +0000 (15:44 +0300)]
net/mlx4_en: current_mac isn't updated in port up
When port is down dev_addr is changed (e.g. by bonding) but current_mac
is not touched. When port is up again, hash_mac is updated to dev_addr,
but current_mac isn't. This leads to inconsistency between current_mac
and mac_hash. Because of that, mlx4_en_replace_mac() fails to find
current_mac in mac_hash.
Fix is to reset current_mac to dev_addr when port is up - as we do for
mac_hash.
Now that all the users of mvpp2_bm_bufs_free() have been fixed, we can safely
clean the function prototype.
The function is always called to release all the buffers in a BM pool, and
the number of buffers freed is not needed. Therefore, we change the return
to a void, and remove the "num" parameter. This is a cosmetic change, to
make the code slightly cleaner.
After a call to mvpp2_bm_bufs_free(), the caller usually wants to know
if the function successfully freed the requested number. However, this
cannot be done by looking into the BM pool count, because the current
buffer count was updated by mvpp2_bm_bufs_free().
In fact, the current callers of mvpp2_bm_bufs_free() use it to release
all the buffers in the pool, so we can fix this by simply checking
if the pool is not empty.
Marcin Wojtas [Mon, 21 Jul 2014 16:48:12 +0000 (13:48 -0300)]
net: mvpp2: Enable proper PHY polling and fix port functionality
Currently, the network interfaces that are not configured by the bootloader
(using e.g. tftp or ping) can detect the link status but are unable to
transmit data.
The network controller has a functionality that allows the hardware to
continuously poll the PHY and directly update the MAC configuration accordingly
(speed, duplex, etc.). However, this doesn't work well with phylib's
software-based polling and updating MAC configuration in the driver's callback.
This commit fixes this issue by:
1. Setting MVPP2_PHY_AN_STOP_SMI0_MASK in MVPP2_PHY_AN_CFG0_REG in
mvpp2_init(), which disables the harware polling feature.
2. Disabling MVPP2_GMAC_PCS_ENABLE_MASK bit in MVPP2_GMAC_CTRL_2_REG in
mvpp2_port_mii_set() for port types other than SGMII.
David Laight [Tue, 22 Jul 2014 08:59:08 +0000 (08:59 +0000)]
net: sctp: Open out the check for Nagle
The check for Nagle contains 6 separate checks all of which must be true
before a data packet is delayed.
Separate out each into its own 'if (test) return SCTP_XMIT_OK' so that
the reasons can be individually described.
Also return directly with SCTP_XMIT_RWND_FULL.
Delete the now-unused 'retval' variable and 'finish' label from
sctp_packet_can_append_data().
1) Null termination fix in dns_resolver got the pointer dereferncing
wrong, fix from Ben Hutchings.
2) ip_options_compile() has a benign but real buffer overflow when
parsing options. From Eric Dumazet.
3) Table updates can crash in netfilter's nftables if none of the state
flags indicate an actual change, from Pablo Neira Ayuso.
4) Fix race in nf_tables dumping, also from Pablo.
5) GRE-GRO support broke the forwarding path because the segmentation
state was not fully initialized in these paths, from Jerry Chu.
6) sunvnet driver leaks objects and potentially crashes on module
unload, from Sowmini Varadhan.
7) We can accidently generate the same handle for several u32
classifier filters, fix from Cong Wang.
8) Several edge case bug fixes in fragment handling in xen-netback,
from Zoltan Kiss.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (21 commits)
ipv4: fix buffer overflow in ip_options_compile()
batman-adv: fix TT VLAN inconsistency on VLAN re-add
batman-adv: drop QinQ claim frames in bridge loop avoidance
dns_resolver: Null-terminate the right string
xen-netback: Fix pointer incrementation to avoid incorrect logging
xen-netback: Fix releasing header slot on error path
xen-netback: Fix releasing frag_list skbs in error path
xen-netback: Fix handling frag_list on grant op error path
net_sched: avoid generating same handle for u32 filters
net: huawei_cdc_ncm: add "subclass 3" devices
net: qmi_wwan: add two Sierra Wireless/Netgear devices
wan/x25_asy: integer overflow in x25_asy_change_mtu()
net: ppp: fix creating PPP pass and active filters
net/mlx4_en: cq->irq_desc wasn't set in legacy EQ's
sunvnet: clean up objects created in vnet_new() on vnet_exit()
r8169: Enable RX_MULTI_EN for RTL_GIGA_MAC_VER_40
net-gre-gro: Fix a bug that breaks the forwarding path
netfilter: nf_tables: 64bit stats need some extra synchronization
netfilter: nf_tables: set NLM_F_DUMP_INTR if netlink dumping is stale
netfilter: nf_tables: safe RCU iteration on list when dumping
...
Pull IDE fixes from David Miller:
- fix interrupt registry for some Atari IDE chipsets.
- adjust Kconfig dependencies for x86_32 specific chips.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/ide:
ide: Fix SC1200 dependencies
ide: Fix CS5520 and CS5530 dependencies
m68k/atari - ide: do not register interrupt if host->get_lock is set
Merge tag 'trace-fixes-v3.16-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull trace fix from Steven Rostedt:
"Tony Luck found that using the "uptime" trace clock that uses jiffies
as a counter was converted to nanoseconds (silly), and after 1 hour 11
minutes and 34 seconds, this monotonic clock would wrap, causing havoc
with the tracing system and making the clock useless.
He converted that clock to use jiffies_64 and made it into a counter
instead of nanosecond conversions, and displayed the clock with the
straight jiffy count, which works much better than it did in the past"
* tag 'trace-fixes-v3.16-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing: Fix wraparound problems in "uptime" trace clock
David S. Miller [Tue, 22 Jul 2014 03:24:32 +0000 (20:24 -0700)]
Merge branch 'cxgb4-next'
Hariprasad Shenai says:
====================
Misc. fix for cxgb4 and iw_cxgb4
This patch series adds support to enchance error reporting, log detailed
warning for negative advice, support query_qp verb and advertise correct
device max attributes for iwarp.
The patches series is created against 'net-next' tree.
And includes patches on cxgb4 and iw_cxgb4 driver.
Since this patch-series contains changes which are dependent on commit id 4c2c5763 ("cxgb4/iw_cxgb4: use firmware ord/ird resource limits") of net-next
tree we would like to request this patch series to get merged via David Miller's
'net-next' tree.
====================
David S. Miller [Tue, 22 Jul 2014 03:22:52 +0000 (20:22 -0700)]
Merge branch 'enic-next'
Govindarajulu Varadarajan says:
====================
enic: Display classifier filters using ethtool
This series adds ethtool support to show classifier filters added by driver.
v2:
The patch 1/2 removes the $ifdef's around the filter structure. Making it
available always. So that .get_rxnfc() can be implimented without any #ifdefs
====================
enic: remove #ifdef CONFIG_RFS_ACCEL around filter structures
This patch removes the #ifdef CONFIG_RFS_ACCEL around the classifier filter
structures. This makes the filter structures available when CONFIG_RFS_ACCEL = n.
Introduce enic_rfs_timer_start() & enic_rfs_timer_stop() to start/stop the
timer. These two functions are nop when CONFIG_RFS_ACCEL = n.
af_iucv: avoid path quiesce of severed path in shutdown()
An af_iucv stress test showed -EPIPE results for sendmsg()
calls. They are caused by quiescing a path even though it has
been already severed by peer. For IUCV transport shutdown()
consists of 2 steps:
(1) sending the shutdown message to peer
(2) quiescing the iucv path
If the iucv path between these 2 steps is severed due to peer
closing the path, the quiesce step is no longer needed.
Stefan Raspl [Mon, 21 Jul 2014 10:54:43 +0000 (12:54 +0200)]
qeth: Display adjacent switch attributes
Add support to display the adjacent switch port's forwarding
attributes. Currently supports info on forwarding modes '802.1'
and 'rr' (reflective relay).
here you have two fixes that we have been testing for quite some time
(this is why they arrived a bit late in the rc cycle).
Patch 1) ensures that BLA packets get dropped and not forwarded to the
mesh even if they reach batman-adv within QinQ frames. Forwarding them
into the mesh means messing up with the TT database of other nodes which
can generate all kind of unexpected behaviours during route computation.
Patch 2) avoids a couple of race conditions triggered upon fast VLAN
deletion-addition. Such race conditions are pretty dangerous because
they not only create inconsistencies in the TT database of the nodes
in the network, but such scenario is also unrecoverable (unless
nodes are rebooted).
====================
Eric Dumazet [Mon, 21 Jul 2014 05:17:42 +0000 (07:17 +0200)]
ipv4: fix buffer overflow in ip_options_compile()
There is a benign buffer overflow in ip_options_compile spotted by
AddressSanitizer[1] :
Its benign because we always can access one extra byte in skb->head
(because header is followed by struct skb_shared_info), and in this case
this byte is not even used.
projects / linux.git / log