Commit graph

678,699 commits

Author SHA1 Message Date
Erik Stromdahl
1c61bedc0a ath10k: fetch firmware images in a loop
To make it easier to handle minimum and maximum firmware API numbers convert
the firmware fetch functionality to a loop. If no firmware image is found print
an error with minimum and maximum API numbers and the name of firmware
directory. This is needed when we switch to using request_firmware_direct()
which doesn't print any errors anymore.

Also add a new function for creating the fw file name dynamically which makes it
easier to add new bus support, for example SDIO and USB, later.

Signed-off-by: Erik Stromdahl <erik.stromdahl@gmail.com>
[kvalo@qca.qualcomm.com: remove sdio/usb part, new error message, clarify commit log]
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2017-02-14 19:52:25 +02:00
David S. Miller
bd08b532f4 Merge branch 'sfc-bogus-interrupt-mode-fallbacks'
Edward Cree says:

====================
sfc: prevent bogus interrupt-mode fallbacks

EF10 VFs only support MSI-X interrupts, not MSI or legacy.  This series
 stops the probe logic from trying to fallback to those if MSI-X interrupt
 probe fails.  It also prevents selecting them with the interrupt_mode
 module parameter.
This avoids producing messages like "failed to hook legacy IRQ 0" and "IRQ
 handler type mismatch for IRQ 0", and ensures that the relevant error
 (from the attempt to enable MSI-X) is reported to the caller.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 12:43:19 -05:00
Andrew Rybchenko
62980cb6dd sfc: only fall back to a lower interrupt mode if it is supported
If we fail to probe interrupts with our minimum mode, return that error.

Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 12:43:18 -05:00
Andrew Rybchenko
6f9f6ec2e0 sfc: MSI-X is the only interrupt mode for EF10 VFs
Add min_interrupt_mode specification per NIC type.
It is a bit confusing because of "highest interrupt mode is less capable".

Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 12:43:18 -05:00
David S. Miller
9b8e1056db Merge branch 'bridge-fdb-minor-cleanup'
Nikolay Aleksandrov says:

====================
bridge: minor fdb cleanup

These patches aim to simplify the bridge fdb API a little by removing some
redundant functions and converting them into wrappers of a single function.
Also add proper lock checking to avoid future mistakes for the search
functions.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 12:41:04 -05:00
Nikolay Aleksandrov
5019ab50f2 bridge: fdb: converge fdb_delete_by functions into one
We can simplify the logic of entries pointing to the bridge by
converging the fdb_delete_by functions, this would allow us to use the
same function for both cases since the fdb's dst is set to NULL if it is
pointing to the bridge thus we can always check for a port match.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 12:41:03 -05:00
Nikolay Aleksandrov
410b3d48f5 bridge: fdb: add proper lock checks in searching functions
In order to avoid new errors add checks to br_fdb_find and fdb_find_rcu
functions. The first requires hash_lock, the second obviously RCU.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 12:41:03 -05:00
Nikolay Aleksandrov
bfd0aeac52 bridge: fdb: converge fdb searching functions into one
Before this patch we had 3 different fdb searching functions which was
confusing. This patch reduces all of them to one - fdb_find_rcu(), and
two flavors: br_fdb_find() which requires hash_lock and br_fdb_find_rcu
which requires RCU. This makes it clear what needs to be used, we also
remove two abusers of __br_fdb_get which called it under hash_lock and
replace them with br_fdb_find().

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 12:41:02 -05:00
Kalle Valo
1427228d58 ath10k: fix napi crash during rmmod when probe firmware fails
This fixes the below crash when ath10k probe firmware fails, NAPI polling tries
to access a rx ring resource which was never allocated. An easy way to
reproduce this is easy to remove all the firmware files, load ath10k modules
and ath10k will crash when calling 'rmmod ath10k_pci'. The fix is to call
napi_enable() from ath10k_pci_hif_start() so that it matches with
napi_disable() being called from ath10k_pci_hif_stop().

Big thanks to Mohammed Shafi Shajakhan who debugged this and provided first
version of the fix. In this patch I just fix the actual problem in pci.c
instead of having a workaround in core.c.

BUG: unable to handle kernel NULL pointer dereference at (null)
IP:  __ath10k_htt_rx_ring_fill_n+0x19/0x230 [ath10k_core]
__ath10k_htt_rx_ring_fill_n+0x19/0x230 [ath10k_core]

Call Trace:

[<ffffffffa113ec62>] ath10k_htt_rx_msdu_buff_replenish+0x42/0x90
[ath10k_core]
[<ffffffffa113f393>] ath10k_htt_txrx_compl_task+0x433/0x17d0
[ath10k_core]
[<ffffffff8114406d>] ? __wake_up_common+0x4d/0x80
[<ffffffff811349ec>] ? cpu_load_update+0xdc/0x150
[<ffffffffa119301d>] ? ath10k_pci_read32+0xd/0x10 [ath10k_pci]
[<ffffffffa1195b17>] ath10k_pci_napi_poll+0x47/0x110 [ath10k_pci]
[<ffffffff817863af>] net_rx_action+0x20f/0x370

Reported-by: Ben Greear <greearb@candelatech.com>
Fixes: 3c97f5de1f ("ath10k: implement NAPI support")
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2017-02-14 19:40:16 +02:00
Amadeusz Sławiński
6d21911356 ath10k: fix comment
I wanted to take a look and it's apparently in other header

Signed-off-by: Amadeusz Sławiński <amadeusz.slawinski@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2017-02-14 19:38:25 +02:00
Amadeusz Sławiński
182f1e5a62 ath10k: use size_t for len variables
cleanup to consolidate type used for len variables

Signed-off-by: Amadeusz Sławiński <amadeusz.slawinski@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2017-02-14 19:38:20 +02:00
Amadeusz Sławiński
56ac13bfc7 ath10k: remove ath10k_vif_to_arvif()
it adds unnecessary level of indirection, while we just access structure
field

Signed-off-by: Amadeusz Sławiński <amadeusz.slawinski@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2017-02-14 19:38:15 +02:00
Richard Leitner
3ec72a2a1e usb: misc: add USB251xB/xBi Hi-Speed Hub Controller Driver
This patch adds a driver for configuration of the Microchip USB251xB/xBi
USB 2.0 hub controller series with USB 2.0 upstream connectivity, SMBus
configuration interface and two to four USB 2.0 downstream ports.

Furthermore add myself as a maintainer for this driver.

The datasheet can be found at the manufacturers website, see [1]. All
device-tree exposed configuration features have been tested on a i.MX6
platform with a USB2512B hub.

[1] http://ww1.microchip.com/downloads/en/DeviceDoc/00001692C.pdf

Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:33:49 -08:00
Colin Ian King
ef5ec7f08d usb: misc: usbtest: remove redundant check on retval < 0
The check for retval being less than zero is always true since
retval equal to -EPIPE at that point.  Replace the existing
conditional with just return retval.

Detected with CoverityScan, CID#114349 ("Logically dead code")

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Peter Chen <peter.chen@nxp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:32:25 -08:00
Arushi Singhal
cf13a747b1 staging:nvec:nvec_ps2.c: Prefering kzalloc(sizeof(*ser_dev)...) over kzalloc(sizeof(struct serio)...)
Prefer kzalloc(sizeof(*ser_dev)...) over kzalloc(sizeof(struct
serio)...) as reported by checkpatch.pl.

Signed-off-by: Arushi Singhal <arushisinghal19971997@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:26:26 -08:00
Julia Lawall
5cec5bacd3 tpm xen: drop unneeded chip variable
The call that used chip was dropped in 1f0f30e404.  Drop the
leftover declaration and initialization.

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
2017-02-14 19:24:33 +02:00
Dmitry Torokhov
39f5712b8f tpm: fix misspelled "facilitate" in module parameter description
I typoed "facilitate" as "faciltate" a few years back...

Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
2017-02-14 19:24:33 +02:00
Wei Yongjun
5939eaf4f9 tpm_tis: fix the error handling of init_tis()
Add the missing platform_driver_unregister() and remove the duplicate
platform_device_unregister(force_pdev) in the error handling case.

Fixes: 00194826e6 ("tpm_tis: Clean up the force=1 module parameter")
Cc: stable@vger.kernel.org
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Reviewed-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
2017-02-14 19:24:33 +02:00
Shishir Kulkarni
842954a8e3 Staging: rtl8188eu: core/rtw_led.c: format line to reduce it below 80 characters
Fixes checkpatch.pl warning: "WARNING: line over 80 characters"

Signed-off-by: Shishir Kulkarni <shishir.kulkarni.2307@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:23:20 -08:00
Rui Sousa
01f8902bcf net: fec: fix multicast filtering hardware setup
Fix hardware setup of multicast address hash:
- Never clear the hardware hash (to avoid packet loss)
- Construct the hash register values in software and then write once
to hardware

Signed-off-by: Rui Sousa <rui.sousa@nxp.com>
Signed-off-by: Fugang Duan <fugang.duan@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 12:15:34 -05:00
Hendrik v. Raven
ebd6228609 staging: bcm2835-audio: remove unused macro
The VC_AUDIO_MAX_MSG_LEN macro is not used by anything in the
bcm2835-audio driver.

Signed-off-by: Hendrik v. Raven <hendrik@consetetur.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:14:29 -08:00
Abhijit Naik
65ab7c2410 staging: bcm2835-audio: Fixed checkpatch warning due to line spacing
bcm2835-vchiq.c:
fixing ERROR: Checkpatch warnings due to extra blank lines or absence of blank
lines where needed.

Signed-off-by: Abhijit Naik <abhijitnaik27@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:14:29 -08:00
Franck Demathieu
4cf64a5f51 staging: nvec: Fix incorrect type of i2c address
The i2c address is unsigned according to the dt-bindings.
Fix sparse issue (-Wtypesign):

  drivers/staging/nvec/nvec.c:781:35: warning: incorrect type in argument 3 (different signedness)
  drivers/staging/nvec/nvec.c:781:35:    expected unsigned int [usertype] *out_value
  drivers/staging/nvec/nvec.c:781:35:    got int *<noident>

Signed-off-by: Franck Demathieu <fdemathieu@gmail.com>
Acked-by: Marc Dietrich <marvin24@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:14:28 -08:00
Bogdan Purcareata
07e9ef1460 staging: fsl-mc: Add missing header
Compiling the fsl-mc bus driver will yield a couple of static analysis
errors:
warning: symbol 'fsl_mc_msi_domain_alloc_irqs' was not declared
warning: symbol 'fsl_mc_msi_domain_free_irqs' was not declared.
warning: symbol 'its_fsl_mc_msi_init' was not declared.
warning: symbol 'its_fsl_mc_msi_cleanup' was not declared.

Since these are properly declared, but the header is not included, add
it in the source files. This way the symbol is properly exported.

Signed-off-by: Bogdan Purcareata <bogdan.purcareata@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:14:28 -08:00
David S. Miller
144adc655f Merge branch 'ipv6-v4mapped'
Jonathan T. Leighton says:

====================
IPv4-mapped on wire, :: dst address issue

Under some circumstances IPv6 datagrams are sent with IPv4-mapped IPv6
addresses as the source. Given an IPv6 socket bound to an IPv4-mapped
IPv6 address, and an IPv6 destination address, both TCP and UDP will
will send packets using the IPv4-mapped IPv6 address as the source. Per
RFC 6890 (Table 20), IPv4-mapped IPv6 source addresses are not allowed
in an IP datagram. The problem can be observed by attempting to
connect() either a TCP or UDP socket, or by using sendmsg() with a UDP
socket. The patch is intended to correct this issue for all socket
types.

linux follows the BSD convention that an IPv6 destination address
specified as in6addr_any is converted to the loopback address.
Currently, neither TCP nor UDP consider the possibility that the source
address is an IPv4-mapped IPv6 address, and assume that the appropriate
loopback address is ::1. The patch adds a check on whether or not the
source address is an IPv4-mapped IPv6 address and then sets the
destination address to either ::ffff:127.0.0.1 or ::1, as appropriate.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 12:13:52 -05:00
Jonathan T. Leighton
052d2369d1 ipv6: Handle IPv4-mapped src to in6addr_any dst.
This patch adds a check on the type of the source address for the case
where the destination address is in6addr_any. If the source is an
IPv4-mapped IPv6 source address, the destination is changed to
::ffff:127.0.0.1, and otherwise the destination is changed to ::1. This
is done in three locations to handle UDP calls to either connect() or
sendmsg() and TCP calls to connect(). Note that udpv6_sendmsg() delays
handling an in6addr_any destination until very late, so the patch only
needs to handle the case where the source is an IPv4-mapped IPv6
address.

Signed-off-by: Jonathan T. Leighton <jtleight@udel.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 12:13:51 -05:00
Jonathan T. Leighton
ec5e3b0a1d ipv6: Inhibit IPv4-mapped src address on the wire.
This patch adds a check for the problematic case of an IPv4-mapped IPv6
source address and a destination address that is neither an IPv4-mapped
IPv6 address nor in6addr_any, and returns an appropriate error. The
check in done before returning from looking up the route.

Signed-off-by: Jonathan T. Leighton <jtleight@udel.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 12:13:51 -05:00
Hauke Mehrtens
17f84b793c spi: lantiq-ssc: add support for Lantiq SSC SPI controller
This driver supports the Lantiq SSC SPI controller in master
mode. This controller is found on Intel (former Lantiq) SoCs like
the Danube, Falcon, xRX200, xRX300.

The hardware uses two hardware FIFOs one for received and one for
transferred bytes. When the driver writes data into the transmit FIFO
the complete word is taken from the FIFO into a shift register. The
data from this shift register is then written to the wire. This driver
uses the interrupts signaling the status of the FIFOs and not the shift
register. It is also possible to use the interrupts for the shift
register, but they will send a signal after every word. When using the
interrupts for the shift register we get a signal when the last word is
written into the shift register and not when it is written to the wire.
After all FIFOs are empty the driver busy waits till the hardware is
not busy any more and returns the transfer status.

Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Mark Brown <broonie@kernel.org>
2017-02-14 17:10:40 +00:00
Karthik Nayak
2c9c5f5a73 staging: comedi: ni_pcimio: remove unused variable 'serial_number'
The struct 'ni_private' holds the variable 'serial_number' which post
assignment is never used. Remove the variable and code pertaining to
obtaining its value.

As a side note, this also fixes the following sparse error:
drivers/staging/comedi/drivers//ni_pcimio.c:1229:32: warning: incorrect type in assignment (different base types)
drivers/staging/comedi/drivers//ni_pcimio.c:1229:32:    expected restricted __be32 [usertype] serial_number
drivers/staging/comedi/drivers//ni_pcimio.c:1229:32:    got unsigned int

Signed-off-by: Karthik Nayak <Karthik.188@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:08:30 -08:00
Saber Rezvani
e1ea433d6e staging: comedi: dyna_pci10xx: usleep_range is preferred over udelay
Fix the checkpatch.pl issue:
CHECK: usleep_range is preferred over udelay

Signed-off-by: Saber Rezvani <irsaber@gmail.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:08:30 -08:00
Saber Rezvani
df6ff8a1ee staging: comedi: s626: usleep_range is preferred over udelay
Fix the checkpatch.pl issue:
CHECK: usleep_range is preferred over udelay

Signed-off-by: Saber Rezvani <irsaber@gmail.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:08:30 -08:00
Saber Rezvani
020883e5e5 staging: comedi: ni_pcidio.c: Spaces preferred around operators
Fix the checkpatch.pl issue:
CHECK: spaces preferred around that '|' (ctx:VxV)

Signed-off-by: Saber Rezvani <irsaber@gmail.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:08:30 -08:00
Saber Rezvani
16cc73893f staging: comedi: ni_pcidio: blank line issues
Fix the checkpatch.pl issue:
CHECK: Please use a blank line after function/struct/union/enum
declarations

Signed-off-by: Saber Rezvani <irsaber@gmail.com>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-02-14 09:08:30 -08:00
Or Gerlitz
fed06ee89b net/mlx5e: Disable preemption when doing TC statistics upcall
When called by HW offloading drivers, the TC action (e.g
net/sched/act_mirred.c) code uses this_cpu logic, e.g

 _bstats_cpu_update(this_cpu_ptr(a->cpu_bstats), bytes, packets)

per the kernel documention, preemption should be disabled, add that.

Before the fix, when running with CONFIG_PREEMPT set, we get a

BUG: using smp_processor_id() in preemptible [00000000] code: tc/3793

asserion from the TC action (mirred) stats_update callback.

Fixes: aad7e08d39 ('net/mlx5e: Hardware offloaded flower filter statistics support')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 11:56:01 -05:00
Volodymyr Bendiuga
91eaa475e2 net:dsa:mv88e6xxx: use watchdog ops for 6097 chip
mv88e6097 chip requires watchdog_ops to be set.

Signed-off-by: Volodymyr Bendiuga <volodymyr.bendiuga@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 11:45:28 -05:00
Talat Batheesh
a748d60df3 IB/mlx4: Take source GID by index from HW GID table
Previously, we used the HW GID index in order to search the source GID
in the software GID cached table. In some cases, for example when
the MAC Address of the network interface is changed, the GID cached table
saves the old-IPv6-link-local GID at the end of the table.

When returning the old MAC address, the software GID cached table tries
to add the new IPv6-link-local GID, and when it identifies that the GID
already exists, the software GID cached does not add it. Thus a mismatch
occurs between the HW and the SW GID tables.

It resulted with sending traffic with the wrong source GID.

This commit fixes the issue by taking both from the HW table.

The problem can be reproduced with the following scenario:
Client:
    # ifconfig ens6 2.2.2.5
    # ifconfig ens6 inet6 add 2001:0db8:0:f101::5/64
    # ifconfig ens6 hw ether f4:52:14:61:a0:71
    # ifconfig ens6 inet6 del 2001:0db8:0:f101::5/64
    # ifconfig ens6 inet6 add 2001:0db8:0:f101::5/64
    # ucmatose -f ipv6 -b 2001:0db8:0:f101::5 -s 2001:0db8:0:f101::6 -p 20156
Server:
    # ucmatose -f ipv6 -b 2001:0db8:0:f101::6 -p 20156

Fixes: 4c3eb3ca13 ('IB/mlx4: Add VLAN support for IBoE')
Signed-off-by: Talat Batheesh <talatb@mellanox.com>
Reviewed-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:44:42 -05:00
Eli Cohen
d8030b0de0 IB/mlx5: Fix blue flame buffer size calculation
A blue flame register is comprised of two buffers of equal size.

Fixes: 5fe9dec0d0 ("IB/mlx5: Use blue flame register allocator in mlx5_ib")
Signed-off-by: Eli Cohen <eli@mellanox.com>
Reviewed-by: Noa Osherovich <noaos@mellanox.com>
Reviewed-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:44:42 -05:00
Leon Romanovsky
850b741514 IB/mlx4: Remove unused variable from function declaration
Remove unused netw_view parameter from eth_link_query_port() function.

Reported-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:44:42 -05:00
Jiri Pirko
d7cf52c249 sched: Fix accidental removal of errout goto
Bring back the goto that was removed by accident.

Reported-by: Colin Ian King <colin.king@canonical.com>
Fixes: 40c81b25b1 ("sched: check negative err value to safe one level of indent")
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-02-14 11:44:14 -05:00
Or Gerlitz
c4550c63b3 IB: Query ports via the core instead of direct into the driver
Change the drivers to call ib_query_port in their get port
immutable handler instead of their own query port handler.

Doing this required to set the core cap flags of this device
before the ib_query_port call is made, since the IB core might
need these caps to serve the port query.

Drivers are ensured by the IB core that the port attributes passed
to the port query verb implementation are zero, and hence we
removed the zeroing from the drivers.

This patch doesn't add any new functionality.

Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Acked-by: Adit Ranadive <aditr@vmware.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:41:22 -05:00
Or Gerlitz
ce1e055fb9 IB: Add protocol for USNIC
Add protocol definition for the proprietary the USNIC driver.

Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Reviewed-by: Christian Benvenuti <benve@cisco.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:41:21 -05:00
Or Gerlitz
bc63f9d558 IB/mlx4: Support raw packet protocol
Mark support for the new raw packet protocol on Eth ports.

Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:41:21 -05:00
Or Gerlitz
72cd57178f IB/mlx5: Support raw packet protocol
Mark support for the new raw packet protocol on Eth ports.

Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:41:20 -05:00
Or Gerlitz
aa773bd495 IB/core: Add raw packet protocol
Define raw packet protocol which comes to denote this port supports
working with raw ethernet frames, e.g as done with RAW_PACKET QPs.

Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:41:20 -05:00
Artemy Kovalyov
81713d3788 IB/mlx5: Add implicit MR support
Add implicit MR, covering entire user address space.
The MR is implemented as an indirect KSM MR consisting of
1GB direct MRs.
Pages and direct MRs are added/removed to MR by ODP.

Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:41:19 -05:00
Artemy Kovalyov
49780d42df IB/mlx5: Expose MR cache for mlx5_ib
Allow other parts of mlx5_ib to use MR cache mechanism.
* Add new functions mlx5_mr_cache_alloc and mlx5_mr_cache_free
* Traditional MTT MKey buckets are limited by MAX_UMR_CACHE_ENTRY
  Additinal buckets may be added above.

Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:41:18 -05:00
Artemy Kovalyov
94990b4989 IB/mlx5: Add null_mkey access
Add mlx5_cmd_null_mkey() function to access null_mkey information
from firmware.

Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:41:18 -05:00
Artemy Kovalyov
d9d0674c0f IB/umem: Indicate that process is being terminated
When process is killed while pagefault operation still in progress -
function will fail. In this specific case we don't want any warnings in
dmesg to avoid log analyzers false alerts. So we need distinct error
code for this case.

Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:41:17 -05:00
Artemy Kovalyov
d07d1d70ce IB/umem: Update on demand page (ODP) support
Currently ODP MR may explicitly register virtual address space area
of limited length.
This change allows MR to cover entire process virtual address space
dynamicaly adding/removing translation entries to device MTT.

Add following changes to support implicit MR:
* Allow umem to be zero size to back-up implicit MR.
* Add new function ib_alloc_odp_umem() to add virtual memory regions
  to implicit MR dynamically on demand.
* Add new function rbt_ib_umem_lookup() to find dynamically added
  virtual memory regions.
* Expose function rbt_ib_umem_for_each_in_range() to other modules and
  make it safe

Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:41:17 -05:00
Artemy Kovalyov
25bf14d6f5 IB/core: Add implicit MR flag
Add flag IB_ODP_SUPPORT_IMPLICIT indicating implicit MR supported.

Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-02-14 11:41:16 -05:00