Wang Sheng-Hui [Fri, 24 Jun 2016 00:52:11 +0000 (08:52 +0800)]
net/mlx5: use mlx5_buf_alloc_node instead of mlx5_buf_alloc in mlx5_wq_ll_create
Commit 311c7c71c9bb ("net/mlx5e: Allocate DMA coherent memory on
reader NUMA node") introduced mlx5_*_alloc_node() but missed changing
some calling and warn messages. This patch introduces 2 changes:
* Use mlx5_buf_alloc_node() instead of mlx5_buf_alloc() in
mlx5_wq_ll_create()
* Update the failure warn messages with _node postfix for
mlx5_*_alloc function names
Fixes: 311c7c71c9bb ("net/mlx5e: Allocate DMA coherent memory on reader NUMA node") Signed-off-by: Wang Sheng-Hui <[email protected]> Acked-By: Saeed Mahameed <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Philippe Reynes [Thu, 23 Jun 2016 21:48:58 +0000 (23:48 +0200)]
net: ethernet: dnet: use phydev from struct net_device
The private structure contain a pointer to phydev, but the structure
net_device already contain such pointer. So we can remove the pointer
phydev in the private structure, and update the driver to use the
one contained in struct net_device.
Florian Fainelli [Thu, 23 Jun 2016 21:25:33 +0000 (14:25 -0700)]
net: bgmac: Remove superflous netif_carrier_on()
bgmac_open() calls phy_start() to initialize the PHY state machine,
which will set the interface's carrier state accordingly, no need to
force that as this could be conflicting with the PHY state determined by
PHYLIB.
Fixes: dd4544f05469 ("bgmac: driver for GBit MAC core on BCMA bus") Signed-off-by: Florian Fainelli <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Florian Fainelli [Thu, 23 Jun 2016 21:25:32 +0000 (14:25 -0700)]
net: bgmac: Start transmit queue in bgmac_open
The driver does not start the transmit queue in bgmac_open(). If the
queue was stopped prior to closing then re-opening the interface, we
would never be able to wake-up again.
Fixes: dd4544f05469 ("bgmac: driver for GBit MAC core on BCMA bus") Signed-off-by: Florian Fainelli <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Florian Fainelli [Thu, 23 Jun 2016 21:23:12 +0000 (14:23 -0700)]
net: bgmac: Fix SOF bit checking
We are checking for the Start of Frame bit in the ctl1 word, while this
bit is set in the ctl0 word instead. Read the ctl0 word and update the
check to verify that.
Fixes: 9cde94506eac ("bgmac: implement scatter/gather support") Signed-off-by: Florian Fainelli <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Jay Vosburgh [Thu, 23 Jun 2016 21:20:51 +0000 (14:20 -0700)]
bonding: fix 802.3ad aggregator reselection
Since commit 7bb11dc9f59d ("bonding: unify all places where
actor-oper key needs to be updated."), the logic in bonding to handle
selection between multiple aggregators has not functioned.
This affects only configurations wherein the bonding slaves
connect to two discrete aggregators (e.g., two independent switches, each
with LACP enabled), thus creating two separate aggregation groups within a
single bond.
The cause is a change in 7bb11dc9f59d to no longer set
AD_PORT_BEGIN on a port after a link state change, which would cause the
port to be reselected for attachment to an aggregator as if were newly
added to the bond. We cannot restore the prior behavior, as it
contradicts IEEE 802.1AX 5.4.12, which requires ports that "become
inoperable" (lose carrier, setting port_enabled=false as per 802.1AX
5.4.7) to remain selected (i.e., assigned to the aggregator). As the port
now remains selected, the aggregator selection logic is not invoked.
A side effect of this change is that aggregators in bonding will
now contain ports that are link down. The aggregator selection logic
does not currently handle this situation correctly, causing incorrect
aggregator selection.
This patch makes two changes to repair the aggregator selection
logic in bonding to function as documented and within the confines of the
standard:
First, the aggregator selection and related logic now utilizes the
number of active ports per aggregator, not the number of selected ports
(as some selected ports may be down). The ad_select "bandwidth" and
"count" options only consider ports that are link up.
Second, on any carrier state change of any slave, the aggregator
selection logic is explicitly called to insure the correct aggregator is
active.
Reported-by: Veli-Matti Lintu <[email protected]> Fixes: 7bb11dc9f59d ("bonding: unify all places where actor-oper key needs to be updated.") Signed-off-by: Jay Vosburgh <[email protected]> Signed-off-by: David S. Miller <[email protected]>
The test_fp_ctl function is used to test if a given value is a valid
floating-point control. The inline assembly in test_fp_ctl uses an
incorrect constraint for the 'orig_fpc' variable. If the compiler
chooses the same register for 'fpc' and 'orig_fpc' the test_fp_ctl()
function always returns true. This allows user space to trigger
kernel oopses with invalid floating-point control values on the
signal stack.
This problem has been introduced with git commit 4725c86055f5bbdcdf
"s390: fix save and restore of the floating-point-control register"
There are use cases where an intermediate boot kernel (1) uses kexec
to boot the final production kernel (2). For this scenario we should
provide the original boot information to the production kernel (2).
Therefore clearing the boot information during kexec() should not
be done.
Linus Torvalds [Tue, 28 Jun 2016 03:43:00 +0000 (20:43 -0700)]
Merge tag 'for-v4.7-rc' of git://git.kernel.org/pub/scm/linux/kernel/git/sre/linux-power-supply
Pull power supply fixes from Sebastian Reichel.
* tag 'for-v4.7-rc' of git://git.kernel.org/pub/scm/linux/kernel/git/sre/linux-power-supply:
power_supply: tps65217-charger: Fix NULL deref during property export
power_supply: power_supply_read_temp only if use_cnt > 0
Linus Torvalds [Tue, 28 Jun 2016 03:34:43 +0000 (20:34 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Pull input fixes from Dmitry Torokhov.
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
Input: vmmouse - remove port reservation
Input: elantech - add more IC body types to the list
Input: wacom_w8001 - ignore invalid pen data packets
Input: wacom_w8001 - w8001_MAX_LENGTH should be 13
Input: xpad - fix oops when attaching an unknown Xbox One gamepad
MAINTAINERS: add Pali Rohár as reviewer of ALPS PS/2 touchpad driver
Input: add HDMI CEC specific keycodes
Input: add BUS_CEC type
Input: xpad - fix rumble on Xbox One controllers with 2015 firmware
Neerav Parikh [Tue, 7 Jun 2016 16:14:55 +0000 (09:14 -0700)]
i40e: Don't notify client(s) for DCB changes on all VSIs
When LLDP/DCBX change happens the i40e driver code flow tried to
notify the client(s) for each of the PF VSIs. This resulted into
kernel panic on the first VSI that didn't have any netdev
associated to it.
The DCB change notification to the client(s) should be done only
once for the PF/LAN VSI where the client(s) instances have been
added to. Also, move the notification call after the PF driver has
made changes related to the updated DCB configuration.
Tushar Dave [Mon, 16 May 2016 19:40:53 +0000 (12:40 -0700)]
i40e: Fix errors resulted while turning off TSO
On systems with 128 CPUs, turning off TSO results in errors,
i40e 0000:03:00.0: failed to get tracking for 1 vectors for VSI 400, err=-12
i40e 0000:03:00.0: Couldn't create FDir VSI
i40e 0000:03:00.0: i40e_ptp_init: PTP not supported on eth0
i40e 0000:03:00.0: couldn't add VEB, err I40E_ERR_ADMIN_QUEUE_ERROR aq_err I40E_AQ_RC_ENOENT
i40e 0000:03:00.0: rebuild of switch failed: -1, will try to set up simple PF connection
i40e 0000:03:00.0 eth0: adding 00:10:e0:8a:24:b6 vid=0
Enabling FD_SB without checking availability of MSI-X vector is the
root cause. This change adds necessary check.
Shannon Nelson [Mon, 16 May 2016 17:26:44 +0000 (10:26 -0700)]
i40e: add VSI info to macaddr messages
Since the macaddr add and delete happens asynchronously, error
messages don't easily get associated to the actual request. Here
we add a bit of information to the error messages to help
determine the source of the error.
Mitch Williams [Mon, 16 May 2016 17:26:43 +0000 (10:26 -0700)]
i40e: set default VSI without a reset
Remove the need for a reset when the device enters limited promiscuous
mode. This was causing heartburn for people who were using VFs and
bridging, since this would require all of the VFs to undergo a reset
each time the PF changed its promiscuity.
Mitch Williams [Mon, 16 May 2016 17:26:42 +0000 (10:26 -0700)]
i40evf: always activate correct MAC address filter
Always add MAC address at the tail of the MAC filter list. Since the
device's "real" MAC address is added first, it will always be at the
beginning of the list. This prevents an issue where the "real" MAC
filter might not get added if too many other filters are added before
bringing the interface up.
i40e: Fix RSS to not be limited by the number of CPUs
Limiting qcount to pf->num_lan_msix, effectively limits the RSS queues
to only use the number of CPUs, and ignore all other queues. We don't
want to do this. If the user has changed the RSS settings to use more
queues then CPUS, we want to trust they know what they are doing and
let them. More importantly, if we tell them that is what we did, we want
to actually do it and allow traffic into all of the queues we have
allocated. This does not change the default setting to initially
allocate only the number of CPUS of queue pairs.
Serey Kong [Mon, 16 May 2016 17:26:39 +0000 (10:26 -0700)]
i40e: fix missing DA cable check
When a Direct Attach (DA) cable is used, if the i40e_set_settings
function is called it would return an error. Add the DA type so
the function won't fail.
Greg Rose [Mon, 16 May 2016 17:26:38 +0000 (10:26 -0700)]
i40e: Save PCI state before suspend
The i40e_suspend() function was failing to save PCI state
and this would result in a kernel stack trace from a WARN_ONCE in the
pci_legacy_suspend() function.
Add a call to pci_save_state() to fix that problem.
Greg Rose [Mon, 16 May 2016 17:26:37 +0000 (10:26 -0700)]
i40e: Clean up MSIX IRQs before suspend
The i40e_suspend() function calls another function that preps the device
for the power save and resume by freeing all the Tx/Rx resources and
interrupts but that function does not free the "other" causes interrupt
vector and IRQ. It also fails to call synchronize_irq() before freeing
the IRQ vectors. This sometimes may result in some AER errors on those
systems with that PCIe error reporting feature enabled.
Call synchronize_irq() before freeing IRQ vectors and explicitly free
the other causes interrupt resources and shut down that MSIX interrupt.
Mitch Williams [Mon, 16 May 2016 17:26:36 +0000 (10:26 -0700)]
i40evf: don't overflow buffer
If the user adds an obscene amount of MAC addresses, the driver will run
into the situation where it has too many address requests to fit into a
single PF message. The driver checks for this case, and calculates the
maximum number of messages that it can send. Then it completely ignores
this count and overflows the buffer.
Fix this by checking the address count and bailing out of the loop at
the appropriate time.
Mitch Williams [Mon, 16 May 2016 17:26:33 +0000 (10:26 -0700)]
i40e: write HENA for VFs
Now that VF RSS is configured by the PF driver, it needs to set the RSS
Hash Enable registers by default. Without this, no packets will be
hashed and they'll all end up on queue 0.
Mitch Williams [Mon, 16 May 2016 17:26:32 +0000 (10:26 -0700)]
i40e: add hw struct local variable
This function uses the i40e_hw struct all over the place, so why doesn't
it keep a pointer to the struct? Add this pointer as a local variable
and use it consistently throughout the function.
Mitch Williams [Mon, 16 May 2016 17:26:31 +0000 (10:26 -0700)]
i40e: add functions to control default VSI
Add functions to enable and disable default VSI on a VEB. This allows
for configuration of limited promiscuous mode specifically for bridging
purposes.
Linus Torvalds [Mon, 27 Jun 2016 20:38:58 +0000 (13:38 -0700)]
Merge branch 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild
Pull kbuild regression fix from Michal Marek:
"The problem is that commit 9c8fa9bc08f6 ("kbuild: fix if_change and
friends to consider argument order") fixed a potential missed rebuild,
but this results in unnnecessary rebuilds with the packaging targets.
Which is still more correct than the previous logic, but also very
annoying"
* 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild:
kbuild: Initialize exported variables
Linus Torvalds [Mon, 27 Jun 2016 18:23:44 +0000 (11:23 -0700)]
Merge branch 'for-next' of git://git.samba.org/sfrench/cifs-2.6
Pull cifs fixes from Steve French:
"Various small cifs/smb3 fixes, include some for stable, and some from
the recent SMB3 test event"
* 'for-next' of git://git.samba.org/sfrench/cifs-2.6:
File names with trailing period or space need special case conversion
Fix reconnect to not defer smb3 session reconnect long after socket reconnect
cifs: check hash calculating succeeded
cifs: dynamic allocation of ntlmssp blob
cifs: use CIFS_MAX_DOMAINNAME_LEN when converting the domain name
cifs: stuff the fl_owner into "pid" field in the lock request
Linus Torvalds [Mon, 27 Jun 2016 17:59:53 +0000 (10:59 -0700)]
Merge branch 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
Pull crypto fixes from Herbert Xu:
"This fixes the following issues:
- Missing length check for user-space GETALG request
- Bogus memmove length in ux500 driver
- Incorrect priority setting for vmx driver
- Incorrect ABI selection for vmx driver"
* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
crypto: user - re-add size check for CRYPTO_MSG_GETALG
crypto: ux500 - memmove the right size
crypto: vmx - Increase priority of aes-cbc cipher
crypto: vmx - Fix ABI detection
Stefan Hajnoczi [Thu, 23 Jun 2016 15:28:58 +0000 (16:28 +0100)]
vsock: make listener child lock ordering explicit
There are several places where the listener and pending or accept queue
child sockets are accessed at the same time. Lockdep is unhappy that
two locks from the same class are held.
Tell lockdep that it is safe and document the lock ordering.
Originally Claudio Imbrenda <[email protected]> sent a similar
patch asking whether this is safe. I have audited the code and also
covered the vsock_pending_work() function.
Russell King [Thu, 23 Jun 2016 13:50:25 +0000 (14:50 +0100)]
phy: improve safety of fixed-phy MII register reading
There is no prevention of a concurrent call to both fixed_mdio_read()
and fixed_phy_update_state(), which can result in the state being
modified while it's being inspected. Fix this by using a seqcount
to detect modifications, and memcpy()ing the state.
We remain slightly naughty here, calling link_update() and updating
the link status within the read-side loop - which would need rework
of the design to change.
Russell King [Thu, 23 Jun 2016 13:50:20 +0000 (14:50 +0100)]
phy: generate swphy registers on the fly
Generate software phy registers as and when requested, rather than
duplicating the state in fixed_phy. This allows us to eliminate
the duplicate storage of of the same data, which is only different
in format.
As fixed_phy_update_regs() no longer updates register state, rename
it to fixed_phy_update().
Russell King [Thu, 23 Jun 2016 13:50:15 +0000 (14:50 +0100)]
phy: separate swphy state validation from register generation
Separate out the generation of MII registers from the state validation.
This allows us to simplify the error handing in fixed_phy() by allowing
earlier error detection.
Russell King [Thu, 23 Jun 2016 13:50:10 +0000 (14:50 +0100)]
phy: convert swphy register generation to tabular form
Convert the swphy register generation to tabular form which allows us
to eliminate multiple switch() statements. This results in a smaller
object code size, more efficient, and easier to add support for faster
speeds.
Paolo Abeni [Thu, 23 Jun 2016 13:25:09 +0000 (15:25 +0200)]
ipv6: enforce egress device match in per table nexthop lookups
with the commit 8c14586fc320 ("net: ipv6: Use passed in table for
nexthop lookups"), net hop lookup is first performed on route creation
in the passed-in table.
However device match is not enforced in table lookup, so the found
route can be later discarded due to egress device mismatch and no
global lookup will be performed.
This cause the following to fail:
ip link add dummy1 type dummy
ip link add dummy2 type dummy
ip link set dummy1 up
ip link set dummy2 up
ip route add 2001:db8:8086::/48 dev dummy1 metric 20
ip route add 2001:db8:d34d::/64 via 2001:db8:8086::2 dev dummy1 metric 20
ip route add 2001:db8:8086::/48 dev dummy2 metric 21
ip route add 2001:db8:d34d::/64 via 2001:db8:8086::2 dev dummy2 metric 21
RTNETLINK answers: No route to host
This change fixes the issue enforcing device lookup in
ip6_nh_lookup_table()
v1->v2: updated commit message title
Fixes: 8c14586fc320 ("net: ipv6: Use passed in table for nexthop lookups") Reported-and-tested-by: Beniamino Galvani <[email protected]> Signed-off-by: Paolo Abeni <[email protected]> Acked-by: David Ahern <[email protected]> Signed-off-by: David S. Miller <[email protected]>
this is a pull request of 4 patches for net-next/master.
Arnd Bergmann's patch fixes a regresseion in af_can introduced in
linux-can-next-for-4.8-20160617. There are two patches by Ramesh
Shanmugasundaram, which add CAN-2.0 support to the rcar_canfd driver.
And a patch by Ed Spiridonov that adds better error diagnoses messages
to the Ed Spiridonov driver.
====================
David S. Miller [Mon, 27 Jun 2016 14:05:55 +0000 (10:05 -0400)]
Merge tag 'linux-can-fixes-for-4.7-20160623' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can
Marc Kleine-Budde says:
====================
pull-request: can 2016-06-23
this is a pull request of 3 patches for the upcoming linux-4.7 release.
The first two patches are by Oliver Hartkopp fixing oopes in the generic CAN
device netlink handling. Jimmy Assarsson's patch for the kvaser_usb driver adds
support for more devices by adding their USB product ids.
====================
Al Viro [Mon, 20 Jun 2016 17:14:36 +0000 (13:14 -0400)]
make nfs_atomic_open() call d_drop() on all ->open_context() errors.
In "NFSv4: Move dentry instantiation into the NFSv4-specific atomic open code"
unconditional d_drop() after the ->open_context() had been removed. It had
been correct for success cases (there ->open_context() itself had been doing
dcache manipulations), but not for error ones. Only one of those (ENOENT)
got a compensatory d_drop() added in that commit, but in fact it should've
been done for all errors. As it is, the case of O_CREAT non-exclusive open
on a hashed negative dentry racing with e.g. symlink creation from another
client ended up with ->open_context() getting an error and proceeding to
call nfs_lookup(). On a hashed dentry, which would've instantly triggered
BUG_ON() in d_materialise_unique() (or, these days, its equivalent in
d_splice_alias()).
Jeremy Linton [Wed, 22 Jun 2016 17:40:50 +0000 (12:40 -0500)]
net: smsc911x: Fix bug where PHY interrupts are overwritten by 0
By default, mdiobus_alloc() sets the PHYs to polling mode, but a
pointer size memcpy means that a couple IRQs end up being overwritten
with a value of 0. This means that PHY_POLL is disabled and results
in unpredictable behavior depending on the PHY's location on the
MDIO bus. Remove that memcpy and the now unused phy_irq member to
force the SMSC911x PHYs into polling mode 100% of the time.
This series includes multiple features extensions for mlx5 Ethernet netdevice driver.
Namely, TX Rate limiting, RX interrupt moderation, ethtool settings.
TX Rate limiting:
- ConnectX-4 rate limiting infrastructure
- Set max rate NDO support
RX interrupt moderation:
- CQE based coalescing option (controlled via priv flags)
- Adaptive RX coalescing
ethtool settings:
- priv flags callbacks
- Support new ksettings API
- Add 50G missing link mode
- Support auto negotiation on/off
Applied on top: 0e9390ebf1fe ("Merge branch 'mlxsw-next'")
====================
Gal Pressman [Thu, 23 Jun 2016 14:02:46 +0000 (17:02 +0300)]
net/mlx5e: Report correct auto negotiation and allow toggling
Previous to this patch auto negotiation was reported off although it was
on by default in hardware. This patch reports the correct information to
ethtool and allows the user to toggle it on/off.
Added another parameter to set port proto function in order to pass
the auto negotiation field to the hardware.
Gal Pressman [Thu, 23 Jun 2016 14:02:45 +0000 (17:02 +0300)]
net/mlx5e: Use new ethtool get/set link ksettings API
Use new get/set link ksettings and remove get/set settings legacy
callbacks.
This allows us to use bitmasks longer than 32 bit for supported and
advertised link modes and use modes that were previously not supported.
Gal Pressman [Thu, 23 Jun 2016 14:02:42 +0000 (17:02 +0300)]
net/mlx5e: Toggle link only after modifying port parameters
Add a dedicated function to toggle port link. It should be called only
after setting a port register.
Toggle will set port link to down and bring it back up in case that it's
admin status was up.
Tariq Toukan [Thu, 23 Jun 2016 14:02:40 +0000 (17:02 +0300)]
net/mlx5e: CQE based moderation
In this mode the moderation timer will restart upon
new completion (CQE) generation rather than upon interrupt
generation.
The outcome is that for bursty traffic the period timer will never
expire and thus only the moderation frames counter will dictate
interrupt generation, thus the interrupt rate will be relative
to the incoming packets size.
If the burst seizes for "moderation period" time then an interrupt
will be issued immediately.
CQE based moderation is off by default and can be controlled
via ethtool set_priv_flags.
Performance tested on ConnectX4-Lx 50G.
Less packet loss in netperf UDP and TCP tests, with no bw degradation,
for both single and multi streams, with message sizes of
64, 1024, 1472 and 32768 byte.
Yevgeny Petrilin [Thu, 23 Jun 2016 14:02:38 +0000 (17:02 +0300)]
net/mlx5e: Add TXQ set max rate support
Implement set_maxrate ndo.
Use the rate index from the hardware table to attach to channel SQ/TXQ.
In case of failure to configure new rate, the queue remains with
unlimited rate.
We save the configuration on priv structure and apply it each time
Send Queues are being reinitialized (after open/close) operations.
Yevgeny Petrilin [Thu, 23 Jun 2016 14:02:37 +0000 (17:02 +0300)]
net/mlx5: Rate limit tables support
Configuring and managing HW rate limit tables.
The HW holds a table of rate limits, each rate is
associated with an index in that table.
Later a Send Queue uses this index to set the rate limit.
Multiple Send Queues can have the same rate limit, which is
represented by a single entry in this table.
Even though a rate can be shared, each queue is being rate
limited independently of others.
The SW shadow of this table holds the rate itself,
the index in the HW table and the refcount (number of queues)
working with this rate.
The exported functions are mlx5_rl_add_rate and mlx5_rl_remove_rate.
Number of different rates and their values are derived
from HW capabilities.
Sathya Perla [Wed, 22 Jun 2016 12:54:57 +0000 (08:54 -0400)]
be2net: update be2net maintainers list
This patch removes Padmanabh's name from the maintainers list as he's no
longer with the company. It also adds the driver name on the headline to
make it easy to lookup the maintainers list by the driver name.
Somnath Kotur [Wed, 22 Jun 2016 12:54:55 +0000 (08:54 -0400)]
be2net: Fix broadcast echoes from EVB in BE3
On SR-IOV profiles, when the user connects a Linux Bridge or OVS to a BE3
vport, they suffer the "broadcast/multicast echo" problem. BE3 EVB echoes
broadcast and multicast packets back to PF's vport confusing the
Linux bridge. BE3 relies on the src-mac addr being programmed on the
interface to avoid sending back an echo of a broadcast or multicast packet
on a vPort. When a Linux bridge is connected to a BE3, the mac-addr of the
VM behind the bridge doesn't get configured on the vPort and so echo
cancellation doesn't work.
This patch worksaround this problem by disabling the EVB initially
and re-enabling it *only* when SR-IOV is enabled by the user. For the
driver fix to work, the BE3 FW version must be >= 11.1.84.0.
Sathya Perla [Wed, 22 Jun 2016 12:54:54 +0000 (08:54 -0400)]
be2net: support asymmetric rx/tx queue counts
be2net so far supported creation of RX/TX queues only in pairs.
On configs where rx and tx queue counts are different, creation of only
the lesser number of queues has been supported.
This patch now allows a combination of RX/TX-only channels along with
combined channels. N TX-queues and M RX-queues can be created with the
following cmds:
ethtool -L ethX combined N rx M-N (when N < M)
ethtool -L ethX combined M tx N-M (when M < N)
Setting both RX-only and TX-only channels is still not supported.
It is mandatory to create atleast one combined channel.
Sathya Perla [Wed, 22 Jun 2016 12:54:53 +0000 (08:54 -0400)]
be2net: fix definition of be_max_eqs()
The EQs available on a function are shared between NIC and RoCE.
The be_max_eqs() macro was so far being used to refer to the max number of
EQs available for NIC. This has caused some confusion in the code. To fix
this confusion this patch introduces a new macro called be_max_nic_eqs()
to refer to the max number of EQs avialable for NIC only and renames
be_max_eqs() to be_max_func_eqs().
David S. Miller [Mon, 27 Jun 2016 08:02:01 +0000 (04:02 -0400)]
Merge branch 'fec-new-type-device'
Andy Duan says:
====================
net: fec: add new type device
Different i.MX SOC FEC support different features like :
- i.MX6Q/DL FEC does not support AVB and interrupt coalesc
- i.MX6SX/i.MX7D supports AVB and interrupt coalesc
- i.MX6UL/ULL does not support AVB, but support interrupt coalesc
Then, add new quirk flag to judge the supported features, and add new
type device for i.MX6UL.
====================
Fugang Duan [Wed, 22 Jun 2016 10:52:35 +0000 (18:52 +0800)]
net: fec: add interrupt coalesc quirk flag
Different i.MX SOC FEC support different features like :
- i.MX6Q/DL FEC does not support AVB and interrupt coalesc
- i.MX6SX/i.MX7D supports AVB and interrupt coalesc
- i.MX6UL/ULL does not support AVB, but support interrupt coalesc
So, add new quirk flag to judge the supported features.
David S. Miller [Sun, 26 Jun 2016 20:01:54 +0000 (16:01 -0400)]
Merge tag 'rxrpc-rewrite-20160622-2' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs
David Howells says:
====================
rxrpc: Get rid of conn bundle and transport structs
Here's the next part of the AF_RXRPC rewrite. The primary purpose of this
set is to get rid of the rxrpc_conn_bundle and rxrpc_transport structs.
This simplifies things for future development of the connection handling.
To this end, the following significant changes are made:
(1) The rxrpc_connection struct is given pointers to the local and peer
endpoints, inside the rxrpc_conn_parameters struct. Pointers to the
transport's copy of these pointers are then redirected to the
connection struct.
(2) Exclusive connection handling is fixed. Exclusive connections should
do just one call and then be retired. They are used in security
negotiations and, I believe, the idea is to avoid reuse of negotiated
security contexts.
The current code is doing a single connection per socket and doing all
the calls over that. With this change it gets a new connection for
each call made.
(3) A new sendmsg() control message marker is added to make individual
calls operate over exclusive connections. This should be used in
future in preference to the sockopt that marks a socket as "exclusive
connection".
(4) IDs for client connections initiated by a machine are now allocated
from a global pool using the IDR facility and are unique across all
client connections, no matter their destination. The IDR facility is
then used to look up a connection on the connection ID alone. Other
parameters are then verified afterwards.
Note that the IDR facility may use a lot of memory if the IDs it holds
are widely scattered. Given this, in a future commit, client
connections will be retired if they are more than a certain distance
from the last ID allocated.
The client epoch is advanced by 1 each time the client ID counter
wraps. Connections outside the current epoch will also be retired in
a future commit.
(5) The connection bundle concept is removed and the client connection
tree is moved into the local endpoint. The queue for waiting for a
call channel is moved to the rxrpc_connection struct as there can only
be one connection for any particular key going to any particular peer
now.
(6) The rxrpc_transport struct is removed and the service connection tree
is moved into the peer struct.
====================
Xing Zheng [Tue, 21 Jun 2016 12:33:28 +0000 (20:33 +0800)]
net: stmmac: dwmac-rk: add rk3228-specific data
Add constants and callback functions for the dwmac on rk3228/rk3229 socs.
As can be seen, the base structure is the same, only registers and the
bits in them moved slightly.
Linus Torvalds [Sun, 26 Jun 2016 17:08:49 +0000 (10:08 -0700)]
Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI fixes from James Bottomley:
"Two straightforward fixes.
One is a concurrency issue only affecting SAS connected SATA drives,
but which could hang the storage subsystem if it triggers (because the
outstanding command count on error never goes back to zero) and the
other is a NO_TAG fallout from the switch to hostwide tags which
causes the system to crash on module insertion (we've checked
carefully and only the 53c700 family of drivers is vulnerable to this
issue)"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
53c700: fix BUG on untagged commands
scsi: fix race between simultaneous decrements of ->host_failed
David S. Miller [Sat, 25 Jun 2016 16:19:41 +0000 (12:19 -0400)]
Merge branch 'net-sched-bulk-dequeue'
Eric Dumazet says:
====================
net_sched: bulk dequeue and deferred drops
First patch adds an additional parameter to ->enqueue() qdisc method
so that drops can be done outside of critical section
(after locks are released).
Then fq_codel can have a small optimization to reduce number of cache
lines misses during a drop event
(possibly accumulating hundreds of packets to be freed).
A small htb change exports the backlog in class dumps.
Final patch adds bulk dequeue to qdiscs that were lacking this feature.
This series brings a nice qdisc performance increase (more than 80 %
in some cases).
====================
Eric Dumazet [Wed, 22 Jun 2016 06:16:52 +0000 (23:16 -0700)]
net_sched: generalize bulk dequeue
When qdisc bulk dequeue was added in linux-3.18 (commit 5772e9a3463b "qdisc: bulk dequeue support for qdiscs
with TCQ_F_ONETXQUEUE"), it was constrained to some
specific qdiscs.
With some extra care, we can extend this to all qdiscs,
so that typical traffic shaping solutions can benefit from
small batches (8 packets in this patch).
For example, HTB is often used on some multi queue device.
And bonding/team are multi queue devices...
Idea is to bulk-dequeue packets mapping to the same transmit queue.
This brings between 35 and 80 % performance increase in HTB setup
under pressure on a bonding setup :
1) NUMA node contention : 610,000 pps -> 1,110,000 pps
2) No node contention : 1,380,000 pps -> 1,930,000 pps
Now we should work to add batches on the enqueue() side ;)
Eric Dumazet [Wed, 22 Jun 2016 06:16:50 +0000 (23:16 -0700)]
net_sched: fq_codel: cache skb->truesize into skb->cb
Now we defer skb drops, it makes sense to keep a copy
of skb->truesize in struct codel_skb_cb to avoid one
cache line miss per dropped skb in fq_codel_drop(),
to reduce latencies a bit further.
Eric Dumazet [Wed, 22 Jun 2016 06:16:49 +0000 (23:16 -0700)]
net_sched: drop packets after root qdisc lock is released
Qdisc performance suffers when packets are dropped at enqueue()
time because drops (kfree_skb()) are done while qdisc lock is held,
delaying a dequeue() draining the queue.
Nominal throughput can be reduced by 50 % when this happens,
at a time we would like the dequeue() to proceed as fast as possible.
Even FQ is vulnerable to this problem, while one of FQ goals was
to provide some flow isolation.
This patch adds a 'struct sk_buff **to_free' parameter to all
qdisc->enqueue(), and in qdisc_drop() helper.
I measured a performance increase of up to 12 %, but this patch
is a prereq so that future batches in enqueue() can fly.
David S. Miller [Sat, 25 Jun 2016 16:08:41 +0000 (12:08 -0400)]
Merge branch 'liquidio-next'
Raghu Vatsavayi says:
====================
liquidio: updates and bug fixes
Please consider following patch series for liquidio bug fixes
and updates on top of net-next. Following patches should be
applied in the following order as some of them depend on
earlier patches in the series.
====================
Raghu Vatsavayi [Wed, 22 Jun 2016 05:53:09 +0000 (22:53 -0700)]
liquidio: chip reset changes
This patch resolves the order of chip reset while destroying
the resources by postoponing soft reset in destroy resources
function until all queues are removed properly.
Raghu Vatsavayi [Wed, 22 Jun 2016 05:53:06 +0000 (22:53 -0700)]
liquidio: Napi rx/tx traffic
This Patch adds tx buffer handling to Napi along with RX
traffic. Also separate spinlocks are introduced for handling
iq posting and buffer reclaim so that tx path and tx interrupt
do not compete against each other.
Raghu Vatsavayi [Wed, 22 Jun 2016 05:53:04 +0000 (22:53 -0700)]
liquidio: Vlan offloads changes
This patch adds support for vlan offloads for the driver and
receive header structures are also modified appropriately. Also
requestID will not be used in reveive header any more.
Raghu Vatsavayi [Wed, 22 Jun 2016 05:53:03 +0000 (22:53 -0700)]
liquidio: soft command buffer limits
This patch increases the limits of soft command buffer size and
num command buffers. This patch also has changes for queue macros
and limit related changes for new chips.
The issue comes when there are multiple threads attempting to use the
mailbox facility at the same time. The issue is the for the Virtual
Function Driver, the only way to get the Virtual Interface statistics
is to issue mailbox commands to ask the firmware for the VI Stats.
And, because the VI Stats command can only retrieve a smallish number of
stats per mailbox command, we have to issue three mailbox commands in quick
succession. When ethtool or netstat command to get interface stats and
interface up/down is run in a loop for every 0.1 sec, we observed
mailbox collisions. And out of the two commands one would fail with
the present code, since we don't queue the second command.
To overcome the above issue, added a queue to access the mailbox.
Whenever a mailbox command is issued add it to the queue. If its at the
head issue the mailbox command, else wait for the existing command to
complete. Usually command takes less than a milli-second to complete.
Also timeout from the loop, if the command under execution takes
long time to run.
In reality, the number of mailbox access collisions is going to be very
rare since no one runs such abusive script.
Philippe Reynes [Tue, 21 Jun 2016 22:32:35 +0000 (00:32 +0200)]
net: ethernet: macb: use phydev from struct net_device
The private structure contain a pointer to phydev, but the structure
net_device already contain such pointer. So we can remove the pointer
phydev in the private structure, and update the driver to use the
one contained in struct net_device.
Jarno Rajahalme [Tue, 21 Jun 2016 21:59:38 +0000 (14:59 -0700)]
openvswitch: Only set mark and labels with a commit flag.
Only set conntrack mark or labels when the commit flag is specified.
This makes sure we can not set them before the connection has been
persisted, as in that case the mark and labels would be lost in an
event of an userspace upcall.
OVS userspace already requires the commit flag to accept setting
ct_mark and/or ct_labels. Validate for this in the kernel API.
Linus Torvalds [Sat, 25 Jun 2016 15:53:38 +0000 (08:53 -0700)]
Merge branch 'for-linus-4.7-part2' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs
Pull btrfs fixes part 2 from Chris Mason:
"This has one patch from Omar to bring iterate_shared back to btrfs.
We have a tree of work we queue up for directory items and it doesn't
lend itself well to shared access. While we're cleaning it up, Omar
has changed things to use an exclusive lock when there are delayed
items"
* 'for-linus-4.7-part2' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: fix ->iterate_shared() by upgrading i_rwsem for delayed nodes
Linus Torvalds [Sat, 25 Jun 2016 15:42:31 +0000 (08:42 -0700)]
Merge branch 'for-linus-4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs
Pull btrfs fixes from Chris Mason:
"I have a two part pull this time because one of the patches Dave
Sterba collected needed to be against v4.7-rc2 or higher (we used
rc4). I try to make my for-linus-xx branch testable on top of the
last major so we can hand fixes to people on the list more easily, so
I've split this pull in two.
This first part has some fixes and two performance improvements that
we've been testing for some time.
Josef's two performance fixes are most notable. The transid tracking
patch makes a big improvement on pretty much every workload"
* 'for-linus-4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: Force stripesize to the value of sectorsize
btrfs: fix disk_i_size update bug when fallocate() fails
Btrfs: fix error handling in map_private_extent_buffer
Btrfs: fix error return code in btrfs_init_test_fs()
Btrfs: don't do nocow check unless we have to
btrfs: fix deadlock in delayed_ref_async_start
Btrfs: track transid for delayed ref flushing
Vivien Didelot [Tue, 21 Jun 2016 16:28:20 +0000 (12:28 -0400)]
net: dsa: mv88e6xxx: rename single-chip support
With the upcoming support for cross-chip operations, it will be hard to
distinguish portions of code supporting a single-chip or a switch fabric
of interconnected chips.
Make the code clearer now, by renaming the mv88e6xxx_priv_state chip
structure to mv88e6xxx_chip. This patch brings no functional changes.
Vivien Didelot [Tue, 21 Jun 2016 16:28:19 +0000 (12:28 -0400)]
net: dsa: mv88e6xxx: move driver in its own folder
With the upcoming support for cross-chip operations and other mv88e6xxx
enhancements, new files will be added.
Similarly to mlxsw or b53, move mv88e6xxx files into their own folder.
In the meantime, update the MAINTAINERS entry to please checkpatch.pl,
by replacing the invalid 88E6352 entry with 88E6XXX, maintained by
Andrew and myself.
projects / linux.git / log