Commit graph

368,847 commits

Author SHA1 Message Date
Davide Caratti
e718fe450e net/mlx4_en: don't set CHECKSUM_COMPLETE on SCTP packets
if the NIC fails to validate the checksum on TCP/UDP, and validation of IP
checksum is successful, the driver subtracts the pseudo-header checksum
from the value obtained by the hardware and sets CHECKSUM_COMPLETE. Don't
do that if protocol is IPPROTO_SCTP, otherwise CRC32c validation fails.

V2: don't test MLX4_CQE_STATUS_IPV6 if MLX4_CQE_STATUS_IPV4 is set

Reported-by: Shuang Li <shuali@redhat.com>
Fixes: f8c6455bb0 ("net/mlx4_en: Extend checksum offloading by CHECKSUM COMPLETE")
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:59:57 -07:00
Greg Edwards
32652c2ac2 igbvf: add lock around mailbox ops
The PF driver assumes the VF will not send another mailbox message until
the PF has written its reply to the previous message.  If the VF does,
that message will be silently dropped by the PF before it writes its
reply to the mailbox.  This results in a VF mailbox timeout for posted
messages waiting for an ACK, and the VF is reset by the
igbvf_watchdog_task in the VM.

Add a lock around the VF mailbox ops to prevent the VF from sending
another message while the PF is still processing the previous one.

Signed-off-by: Greg Edwards <gedwards@ddn.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-08-08 17:59:03 -07:00
Sasha Neftin
48f76b68f9 e1000e: Initial Support for IceLake
i219 (8) and i219 (9) are the next LOM generations that will be available
on the next Intel Client platform (IceLake).
This patch provides the initial support for these devices

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Reviewed-by: Raanan Avargil <raanan.avargil@intel.com>
Reviewed-by: Dima Ruinskiy <dima.ruinskiy@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-08-08 17:54:21 -07:00
Greg Edwards
46b3bb9b47 igb: do not drop PF mailbox lock after read of VF message
When the PF receives a mailbox message from the VF, it grabs the mailbox
lock, reads the VF message from the mailbox, ACKs the message and drops
the lock.

While the PF is performing the action for the VF message, nothing
prevents another VF message from being posted to the mailbox.  The
current code handles this condition by just dropping any new VF messages
without processing them.  This results in a mailbox timeout in the VM
for posted messages waiting for an ACK, and the VF is reset by the
igbvf_watchdog_task in the VM.

Given the right sequence of VF messages and mailbox timeouts, this
condition can go on ad infinitum.

Modify the PF mailbox read method to take an 'unlock' argument that
optionally leaves the mailbox locked by the PF after reading the VF
message.  This ensures another VF message is not posted to the mailbox
until after the PF has completed processing the VF message and written
its reply.

Signed-off-by: Greg Edwards <gedwards@ddn.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-08-08 17:52:45 -07:00
Edward Cree
f1174f77b5 bpf/verifier: rework value tracking
Unifies adjusted and unadjusted register value types (e.g. FRAME_POINTER is
 now just a PTR_TO_STACK with zero offset).
Tracks value alignment by means of tracking known & unknown bits.  This
 also replaces the 'reg->imm' (leading zero bits) calculations for (what
 were) UNKNOWN_VALUEs.
If pointer leaks are allowed, and adjust_ptr_min_max_vals returns -EACCES,
 treat the pointer as an unknown scalar and try again, because we might be
 able to conclude something about the result (e.g. pointer & 0x40 is either
 0 or 0x40).
Verifier hooks in the netronome/nfp driver were changed to match the new
 data structures.

Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:51:34 -07:00
Greg Edwards
1a6c4a3b1e igb: expose mailbox unlock method
Add a mailbox unlock method to e1000_mbx_operations, which will be used
to unlock the PF/VF mailbox by the PF.

Signed-off-by: Greg Edwards <gedwards@ddn.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-08-08 17:50:46 -07:00
Greg Edwards
09fc97ba3e igb: add argument names to mailbox op function declarations
Signed-off-by: Greg Edwards <gedwards@ddn.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-08-08 17:48:10 -07:00
Arvind Yadav
e1cb90f2b8 net: usb: rtl8150: constify usb_device_id
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:47:59 -07:00
Arvind Yadav
9b4355fb03 net: usb: r8152: constify usb_device_id
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:47:59 -07:00
Arvind Yadav
06fa59a018 net: usb: kaweth: constify usb_device_id
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:47:59 -07:00
Arvind Yadav
9befbe13eb net: usb: ipheth: constify usb_device_id
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:47:58 -07:00
Arvind Yadav
7f04c61d91 net: usb: cdc-phonet: constify usb_device_id
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:47:58 -07:00
Arvind Yadav
2f622c4951 net: usb: catc: constify usb_device_id and fix space before '[' error
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.

Fix checkpatch.pl error:
ERROR: space prohibited before open square bracket '['.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:47:58 -07:00
Arvind Yadav
d1bb5aaa29 net: irda: stir4200: constify usb_device_id
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:47:58 -07:00
Arvind Yadav
4730279f10 net: irda: mcs7780: constify usb_device_id
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:47:58 -07:00
Arvind Yadav
7f4e87e971 net: irda: ksdazzle: constify usb_device_id
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:47:58 -07:00
Arvind Yadav
10ffb42270 net: irda: ks959: constify usb_device_id
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:47:58 -07:00
Arvind Yadav
fa00f26ff6 net: irda: kingsun: constify usb_device_id
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:47:58 -07:00
Arvind Yadav
42e6b92de8 net: irda: irda-usb: constify usb_device_id
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-08 17:47:58 -07:00
Corinna Vinschen
2643e6e902 igb: Remove incorrect "unexpected SYS WRAP" log message
TSAUXC.DisableSystime is never set, so SYSTIM runs into a SYS WRAP
every 1100 secs on 80580/i350/i354 (40 bit SYSTIM) and every 35000
secs on 80576 (45 bit SYSTIM).

This wrap event sets the TSICR.SysWrap bit unconditionally.

However, checking TSIM at interrupt time shows that this event does not
actually cause the interrupt.  Rather, it's just bycatch while the
actual interrupt is caused by, for instance, TSICR.TXTS.

The conclusion is that the SYS WRAP is actually expected, so the
"unexpected SYS WRAP" message is entirely bogus and just helps to
confuse users.  Drop it.

Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Acked-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-08-08 17:46:25 -07:00
Gustavo A R Silva
d75372a2da e1000e: add check on e1e_wphy() return value
Check return value from call to e1e_wphy(). This value is being
checked during previous calls to function e1e_wphy() and it seems
a check was missing here.

Addresses-Coverity-ID: 1226905
Signed-off-by: Gustavo A R Silva <garsilva@embeddedor.com>
Reviewed-by: Ethan Zhao <ethan.zhao@oracle.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-08-08 17:44:53 -07:00
Cliff Spradlin
26bd4e2db0 igb: protect TX timestamping from API misuse
HW timestamping can only be requested for a packet if the NIC is first
setup via ioctl(SIOCSHWTSTAMP). If this step was skipped, then the igb
driver still allowed TX packets to request HW timestamping. In this
situation, the _IGB_PTP_TX_IN_PROGRESS flag was set and would never
clear. This prevented any future HW timestamping requests to succeed.

Fix this by checking that the NIC is configured for HW TX timestamping
before accepting a HW TX timestamping request.

Signed-off-by: Cliff Spradlin <cspradlin@google.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-08-08 17:43:18 -07:00
Gangfeng Huang
94221ae75c igb: Fix error of RX network flow classification
After add an ethertype filter, if user change the adapter speed several
times, the error "ethtool -N: etype filters are all used" is reported by
igb driver.

In older patch, function igb_nfc_filter_exit() and igb_nfc_filter_restore()
is not paried. igb_nfc_filter_restore() exist in igb_up(), but function
igb_nfc_filter_exit() is exist in __igb_close(). In the process of speed
changing, only igb_nfc_filter_restore() is called, it will take a position
of ethertype bitmap.

Reproduce steps:
Step 1: Add a etype filter by ethtool
$ethtool -N eth0 flow-type ether proto 0x88F8 action 1
Step 2: Change the adapter speed to 100M/full duplex
$ethtool -s eth0 speed 100 duplex full
Step 3: Change the adapter speed to 1000M/full duplex
ethtool -s eth0 speed 1000 duplex full
Repeat step2 and step3, then dmesg the system log, you can find the error
message, add new ethtype filter is also failed.

This fixing is move igb_nfc_filter_exit() from __igb_close() to igb_down()
to make igb_nfc_filter_restore()/igb_nfc_filter_exit() is paired.

Signed-off-by: Gangfeng Huang <gangfeng.huang@ni.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-08-08 17:28:29 -07:00
Bjorn Andersson
445c2410a4 soc: qcom: mdt_loader: Use request_firmware_into_buf()
By switching to the request_firmware_into_buf() we load the segment data
straight into the preallocated buffers, reducing the need for allocating
scratch buffers for these. In particular the modem firmware consists of
multiple segments in the range 5-15MB, making this worth while.

Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Andy Gross <andy.gross@linaro.org>
2017-08-08 17:41:23 -05:00
Marek Szyprowski
1899bd5757 drm/exynos: forbid creating framebuffers from too small GEM buffers
Add a check if the framebuffer described by the provided drm_mode_fb_cmd2
structure fits into provided GEM buffers. Without this check it is
possible to create a framebuffer object from a small buffer and set it to
the hardware, what results in displaying system memory outside the
allocated GEM buffer.

Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Tobias Jakobi <tjakobi@math.uni-bielefeld.de>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2017-08-09 07:34:23 +09:00
Eric Anholt
3be8eddd9d drm/vc4: Add exec flags to allow forcing a specific X/Y tile walk order.
This is useful to allow GL to provide defined results for overlapping
glBlitFramebuffer, which X11 in turn uses to accelerate uncomposited
window movement without first blitting to a temporary.  x11perf
-copywinwin100 goes from 1850/sec to 4850/sec.

v2: Default to the same behavior as before when the flags aren't
    passed. (suggested by Boris)

Signed-off-by: Eric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20170725162733.28007-2-eric@anholt.net
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
2017-08-08 13:26:44 -07:00
Eric Anholt
fb95992af1 drm/vc4: Demote user-accessible DRM_ERROR paths to DRM_DEBUG.
Userspace shouldn't be able to spam dmesg by passing bad arguments.
This has particularly become an issues since we started using a bad
argument to set_tiling to detect if set_tiling was supported.

Signed-off-by: Eric Anholt <eric@anholt.net>
Fixes: 83753117f1 ("drm/vc4: Add get/set tiling ioctls.")
Link: https://patchwork.freedesktop.org/patch/msgid/20170725162733.28007-1-eric@anholt.net
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
2017-08-08 13:20:11 -07:00
Cihangir Akturk
1d5494e965 drm/vc4: switch to drm_*{get,put} helpers
drm_*_reference() and drm_*_unreference() functions are just
compatibility alias for drm_*_get() and drm_*_put() adn should not be
used by new code. So convert all users of compatibility functions to use
the new APIs.

Signed-off-by: Cihangir Akturk <cakturk@gmail.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/1501761585-11757-26-git-send-email-cakturk@gmail.com
Reviewed-by: Eric Anholt <eric@anholt.net>
2017-08-08 13:20:11 -07:00
Eric Anholt
97ee731da8 drm/vc4: Fix errant drm_bridge_remove() in DSI.
drm_bridge_remove() is for unregistering a bridge driver, not for
detaching a bridge from its consumer.

Fixes: 656fa22f9c ("drm/vc4: Switch DSI to the panel-bridge layer, and support bridges.")
Signed-off-by: Eric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20170802203242.12815-3-eric@anholt.net
Acked-by: Noralf Trønnes <noralf@tronnes.org>
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
2017-08-08 13:20:11 -07:00
Eric Anholt
2cbf59fd90 drm/vc4: Don't disable DSI clocks on component unload.
The clocks are enabled/disabled at encoder enable/disable time, not at
component load.  Fixes a WARN_ON at boot if V3D fails to probe.

Fixes: 4078f57571 ("drm/vc4: Add DSI driver")
Signed-off-by: Eric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20170802203242.12815-2-eric@anholt.net
Acked-by: Noralf Trønnes <noralf@tronnes.org>
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
2017-08-08 13:20:10 -07:00
Eric Anholt
2aee7c5db7 drm/vc4: Fix double destroy of the BO cache on teardown.
It's also destroyed from the top level vc4_drv.c initialization, which
is where the cache was actually initialized from.

This used to just involve duplicate del_timer() and cancel_work_sync()
being called, but it started causing kmalloc issues once we
double-freed the new BO label array.

Fixes: 1908a876f909 ("drm/vc4: Add an ioctl for labeling GEM BOs for summary stats")
Signed-off-by: Eric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20170802203242.12815-1-eric@anholt.net
Tested-by: Noralf Trønnes <noralf@tronnes.org>
Acked-by: Noralf Trønnes <noralf@tronnes.org>
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
2017-08-08 13:20:10 -07:00
Rodrigo Vivi
ea46708f07 drm/i915/cnl: Removing missing DDI_E bits from CNL.
DDI_E is not supported on CNL-U and CNL-Y

When adding the initial support we noticed DDI_E wasn't supported
and removed it on v4 and v5 of that patch.
However for some reason I missed or put back these 2 chunks.

Time to clean it up to avoid later confusion.

Fixes: 8bcd3dd417 ("drm/i915/cnl: Add power wells for CNL")
Cc: Clint Taylor <clinton.a.taylor@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Daniel Vetter <daniel.vetter@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170808193237.17410-1-rodrigo.vivi@intel.com
2017-08-08 13:03:41 -07:00
Srinivas Kandagatla
2a464815d3 soc: qcom: bring all qcom drivers into a submenu
As the number of qcom drivers increase, entries in SOC menu looks
scattered with other SOC drivers. Make a submenu for Qcom drivers
to make it visibly clear while selecting qcom SOC specific drivers.

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Andy Gross <andy.gross@linaro.org>
2017-08-08 14:39:17 -05:00
Rob Clark
b833990901 soc: qcom: wcnss_ctrl: add missing MODULE_DEVICE_TABLE()
This fixes a problem of wifi module not loading on db410c.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Andy Gross <andy.gross@linaro.org>
2017-08-08 14:37:42 -05:00
Rob Clark
3e8b554114 soc: qcom: smsm: fix of_node refcnting problem
of_find_node_with_property() drops the reference to the 'from' node,
which eventually (after enough -EPROBE_DEFERs) drops the last reference
to the node causing all sorts of fun problems, and this nice splat.

  BUG: sleeping function called from invalid context at ../kernel/locking/mutex.c:747
  in_atomic(): 1, irqs_disabled(): 128, pid: 33, name: kworker/0:1
  4 locks held by kworker/0:1/33:
   #0:  ("events"){.+.+.+}, at: [<ffff0000080fa91c>] process_one_work+0x1a4/0x728
   #1:  (deferred_probe_work){+.+.+.}, at: [<ffff0000080fa91c>] process_one_work+0x1a4/0x728
   #2:  (&dev->mutex){......}, at: [<ffff000008676078>] __device_attach+0x30/0x168
   #3:  (devtree_lock){......}, at: [<ffff000008828fd0>] of_find_node_with_property+0x30/0xe0
  irq event stamp: 18976
  hardirqs last  enabled at (18975): [<ffff00000815794c>] __down_trylock_console_sem+0x74/0xb8
  hardirqs last disabled at (18976): [<ffff0000089e26d4>] _raw_spin_lock_irqsave+0x2c/0x78
  softirqs last  enabled at (16880): [<ffff0000080e0f00>] __do_softirq+0x580/0x640
  softirqs last disabled at (16871): [<ffff0000080e13a4>] irq_exit+0xe4/0x138
  CPU: 0 PID: 33 Comm: kworker/0:1 Tainted: G            E   4.12.0-rc5+ #1455
  Hardware name: qualcomm dragonboard410c/dragonboard410c, BIOS 2017.07-rc1-00234-g22fa70a-dirty 06/26/2017
  Workqueue: events deferred_probe_work_func
  Call trace:
  [<ffff000008089ee0>] dump_backtrace+0x0/0x230
  [<ffff00000808a134>] show_stack+0x24/0x30
  [<ffff0000084e1944>] dump_stack+0xac/0xe8
  [<ffff00000810d7e0>] ___might_sleep+0x150/0x230
  [<ffff00000810d918>] __might_sleep+0x58/0x90
  [<ffff0000089dde18>] __mutex_lock+0x50/0x870
  [<ffff0000089de674>] mutex_lock_nested+0x3c/0x50
  [<ffff000008388ae0>] kernfs_remove+0x30/0x50
  [<ffff00000838b720>] sysfs_remove_dir+0x58/0x70
  [<ffff0000084e393c>] kobject_del+0x1c/0x58
  [<ffff0000084e374c>] kobject_put+0xb4/0x208
  [<ffff00000882c364>] of_node_put+0x24/0x30
  [<ffff000008829018>] of_find_node_with_property+0x78/0xe0
  [<ffff000000aff5f4>] qcom_smsm_probe+0x194/0x720 [smsm]
  [<ffff0000086793b4>] platform_drv_probe+0x74/0x110
  [<ffff0000086765bc>] driver_probe_device+0x2b4/0x420
  [<ffff000008676920>] __device_attach_driver+0xd0/0x150
  [<ffff000008673e78>] bus_for_each_drv+0x68/0xa8
  [<ffff00000867611c>] __device_attach+0xd4/0x168
  [<ffff000008676a1c>] device_initial_probe+0x24/0x30
  [<ffff000008675380>] bus_probe_device+0xa0/0xa8
  [<ffff000008675948>] deferred_probe_work_func+0xb8/0xf8
  [<ffff0000080fa9d4>] process_one_work+0x25c/0x728
  [<ffff0000080faef4>] worker_thread+0x54/0x3d8
  [<ffff0000081031d8>] kthread+0x110/0x140
  [<ffff000008082d90>] ret_from_fork+0x10/0x40
  OF: ERROR: Bad of_node_put() on /smsm
  CPU: 0 PID: 33 Comm: kworker/0:1 Tainted: G        W   E   4.12.0-rc5+ #1455
  Hardware name: qualcomm dragonboard410c/dragonboard410c, BIOS 2017.07-rc1-00234-g22fa70a-dirty 06/26/2017
  Workqueue: events deferred_probe_work_func

Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Andy Gross <andy.gross@linaro.org>
2017-08-08 14:37:42 -05:00
Sagi Grimberg
0b36658ca8 nvme-rdma: use intelligent affinity based queue mappings
Use the generic block layer affinity mapping helper. Also,
limit nr_hw_queues to the rdma device number of irq vectors
as we don't really need more.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:58:03 -04:00
Sagi Grimberg
40b24403f3 mlx5: support ->get_vector_affinity
Simply refer to the generic affinity mask helper.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:58:03 -04:00
Sagi Grimberg
a435393aca mlx5: move affinity hints assignments to generic code
generic api takes care of spreading affinity similar to
what mlx5 open coded (and even handles better asymmetric
configurations). Ask the generic API to spread affinity
for us, and feed him pre_vectors that do not participate
in affinity settings (which is an improvement to what we
had before).

The affinity assignments should match what mlx5 tried to
do earlier but now we do not set affinity to async, cmd
and pages dedicated vectors.

Also, remove mlx5e_get_cpu and introduce mlx5e_get_node
(used for allocation purposes) and mlx5_get_vector_affinity
(for indirection table construction) as they provide the needed
information. Luckily, we have generic helpers to get cpumask
and node given a irq vector. mlx5_get_vector_affinity will
be used by mlx5_ib in a subsequent patch.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:55:56 -04:00
Sagi Grimberg
a85e5474f4 mlx5e: don't assume anything on the irq affinity mappings of the device
mlx5e currently assumes that irq affinity is really spread first
irq vectors across device home node cpus, with the new generic affinity
mappings this is no longer the case, hence mlxe should not rely on
this anymore.

Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:53:05 -04:00
Sagi Grimberg
78249c4215 mlx5: convert to generic pci_alloc_irq_vectors
Now that we have a generic code to allocate an array
of irq vectors and even correctly spread their affinity,
correctly handle cpu hotplug events and more, were much
better off using it.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:53:05 -04:00
Dasaratharaman Chandramouli
ac3a949fb2 IB/CM: Set appropriate slid and dlid when handling CM request
If extended LIDs are being used, a connection request contains
OPA GIDs in them. Extract the lids from the OPA gids and populate
slid/dlid fields in the path records that are created when handling
a connection request.

Signed-off-by: Dasaratharaman Chandramouli <dasaratharaman.chandramouli@intel.com>
Reviewed-by: Don Hiatt <don.hiatt@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:50:40 -04:00
Dasaratharaman Chandramouli
6b3c0e6e6d IB/CM: Create appropriate path records when handling CM request
When handling an incoming conection request, ib_cm creates
either an IB or an OPA path record based on the gid field
in the request.

Signed-off-by: Dasaratharaman Chandramouli <dasaratharaman.chandramouli@intel.com>
Reviewed-by: Don Hiatt <don.hiatt@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:50:25 -04:00
Hiatt, Don
e92aa00a51 IB/CM: Add OPA Path record support to CM
Add OPA path record support to the Connection Manager.

Signed-off-by: Don Hiatt <don.hiatt@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:50:25 -04:00
Hiatt, Don
7db20ecd1d IB/core: Change wc.slid from 16 to 32 bits
slid field in struct ib_wc is increased to 32 bits.
This enables core components to use larger LIDs if needed.
The user ABI is unchanged and return 16 bit values when queried.

Signed-off-by: Dasaratharaman Chandramouli <dasaratharaman.chandramouli@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Don Hiatt <don.hiatt@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:50:25 -04:00
Dasaratharaman Chandramouli
db58540b02 IB/core: Change port_attr.sm_lid from 16 to 32 bits
sm_lid field in struct ib_port_attr is increased to 32 bits. This
enables core components to use larger LIDs if needed.
The user ABI is unchanged and return 16 bit values when queried.

Signed-off-by: Dasaratharaman Chandramouli <dasaratharaman.chandramouli@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Don Hiatt <don.hiatt@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:50:25 -04:00
Dasaratharaman Chandramouli
582faf3150 IB/core: Change port_attr.lid size from 16 to 32 bits
lid field in struct ib_port_attr is increased to 32 bits. This enables core
components to use larger LIDs if needed.
The user ABI is unchanged and return 16 bit values when queried.

Signed-off-by: Dasaratharaman Chandramouli <dasaratharaman.chandramouli@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Don Hiatt <don.hiatt@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:50:25 -04:00
Dasaratharaman Chandramouli
1cb2fc0db7 IB/mad: Change slid in RMPP recv from 16 to 32 bits
MAD RMPP contains slid field which is 16 bits in
length, increase it to 32 bits.

Signed-off-by: Dasaratharaman Chandramouli <dasaratharaman.chandramouli@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Don Hiatt <don.hiatt@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:47:18 -04:00
Dasaratharaman Chandramouli
7e93e2cb83 IB/IPoIB: Increase local_lid to 32 bits
IPoIB contains local_lid field which is 16 bits in
length, increase it to 32 bits.

Signed-off-by: Dasaratharaman Chandramouli <dasaratharaman.chandramouli@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Don Hiatt <don.hiatt@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:47:18 -04:00
Dasaratharaman Chandramouli
4c4736905e IB/srpt: Increase lid and sm_lid to 32 bits
srpt contains lid and sm_lid fields which are 16 bits in
length, increase them to 32 bits.

Signed-off-by: Dasaratharaman Chandramouli <dasaratharaman.chandramouli@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Don Hiatt <don.hiatt@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:47:18 -04:00
Dasaratharaman Chandramouli
d541e45500 IB/core: Convert ah_attr from OPA to IB when copying to user
OPA address handle atttibutes that have 32 bit LIDs would have to
be converted to IB address handle attribute with the LID field
programmed in the GID before copying to user space.

Signed-off-by: Dasaratharaman Chandramouli <dasaratharaman.chandramouli@intel.com>
Reviewed-by: Don Hiatt <don.hiatt@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-08-08 14:47:18 -04:00