OSDN Git Service

uclinux-h8/linux.git
5 years agomt76: unify wait_for_mac
Stanislaw Gruszka [Wed, 29 Aug 2018 11:16:32 +0000 (13:16 +0200)]
mt76: unify wait_for_mac

Create new mt76x02_mac.h for common mac functions of mt76x0 and mt76x2
and unity wait_for_mac for both drivers.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agomt76: verify evt type in usb mcu response
Lorenzo Bianconi [Fri, 24 Aug 2018 10:41:44 +0000 (12:41 +0200)]
mt76: verify evt type in usb mcu response

Verify if evt field is set to EVT_CMD_DONE in usb mcu
response messages. This is a preliminary patch for usb_mcu layer
unification between mt76x0u and mt76x2u driver.

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agomt76x2u: run device cleanup routine if resume fails
Lorenzo Bianconi [Fri, 24 Aug 2018 08:17:49 +0000 (10:17 +0200)]
mt76x2u: run device cleanup routine if resume fails

Cleanup {tx,rx} and mcu queues if resume operation fails

Fixes: ee676cd5017c ("mt76: add driver code for MT76x2u based devices")
Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agomt76: Fix comparisons with invalid hardware key index
Geert Uytterhoeven [Thu, 23 Aug 2018 21:27:38 +0000 (23:27 +0200)]
mt76: Fix comparisons with invalid hardware key index

With gcc 4.1.2:

    drivers/net/wireless/mediatek/mt76/mt76x0/tx.c: In function ‘mt76x0_tx’:
    drivers/net/wireless/mediatek/mt76/mt76x0/tx.c:169: warning: comparison is always true due to limited range of data type
    drivers/net/wireless/mediatek/mt76/mt76x2_tx_common.c: In function ‘mt76x2_tx’:
    drivers/net/wireless/mediatek/mt76/mt76x2_tx_common.c:35: warning: comparison is always true due to limited range of data type

While assigning -1 to a u8 works fine, comparing with -1 does not work
as expected.

Fix this by comparing with 0xff, like is already done in some other
places.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agorsi: improve kernel thread handling to fix kernel panic
Siva Rebbagondla [Mon, 27 Aug 2018 11:35:15 +0000 (17:05 +0530)]
rsi: improve kernel thread handling to fix kernel panic

While running regressions, observed below kernel panic when sdio disconnect
called. This is because of, kthread_stop() is taking care of
wait_for_completion() by default. When wait_for_completion triggered
in kthread_stop and as it was done already, giving kernel panic.
Hence, removing redundant wait_for_completion() from rsi_kill_thread().

... skipping ...
BUG: unable to handle kernel NULL pointer dereference at           (null)
IP: [<ffffffff810a63df>] exit_creds+0x1f/0x50
PGD 0
Oops: 0002 [#1] SMP
CPU: 0 PID: 6502 Comm: rmmod Tainted: G  OE   4.15.9-Generic #154-Ubuntu
Hardware name: Dell Inc. Edge Gateway 3003/ , BIOS 01.00.00 04/17/2017
Stack:
ffff88007392e600 ffff880075847dc0 ffffffff8108160a 0000000000000000
ffff88007392e600 ffff880075847de8 ffffffff810a484b ffff880076127000
ffff88003cd3a800 ffff880074f12a00 ffff880075847e28 ffffffffc09bed15
Call Trace:
[<ffffffff8108160a>] __put_task_struct+0x5a/0x140
[<ffffffff810a484b>] kthread_stop+0x10b/0x110
[<ffffffffc09bed15>] rsi_disconnect+0x2f5/0x300 [ven_rsi_sdio]
[<ffffffff81578bcb>] ? __pm_runtime_resume+0x5b/0x80
[<ffffffff816f0918>] sdio_bus_remove+0x38/0x100
[<ffffffff8156cc64>] __device_release_driver+0xa4/0x150
[<ffffffff8156d7a5>] driver_detach+0xb5/0xc0
[<ffffffff8156c6c5>] bus_remove_driver+0x55/0xd0
[<ffffffff8156dfbc>] driver_unregister+0x2c/0x50
[<ffffffff816f0b8a>] sdio_unregister_driver+0x1a/0x20
[<ffffffffc09bf0f5>] rsi_module_exit+0x15/0x30 [ven_rsi_sdio]
[<ffffffff8110cad8>] SyS_delete_module+0x1b8/0x210
[<ffffffff81851dc8>] entry_SYSCALL_64_fastpath+0x1c/0xbb

Signed-off-by: Siva Rebbagondla <siva.rebbagondla@redpinesignals.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agorsi: fix memory alignment issue in ARM32 platforms
Siva Rebbagondla [Mon, 27 Aug 2018 11:35:14 +0000 (17:05 +0530)]
rsi: fix memory alignment issue in ARM32 platforms

During testing in ARM32 platforms, observed below kernel panic, as driver
accessing data beyond the allocated memory while submitting URB to USB.

Fix: Resolved this by specifying correct length by considering 64 bit
alignment. so that, USB bus driver will access only allocated memory.

Unit-test: Tested and confirm that driver bring up and scanning,
connection and data transfer works fine with this fix.

...skipping...
[   25.389450] Unable to handle kernel paging request at virtual
       address 5aa11422
[   25.403078] Internal error: Oops: 5 [#1] SMP ARM
[   25.407703] Modules linked in: rsi_usb
[   25.411473] CPU: 1 PID: 317 Comm: RX-Thread Not tainted 4.18.0-rc7 #1
[   25.419221] Hardware name: Freescale i.MX6 Quad/DualLite (Device Tree)
[   25.425764] PC is at skb_release_data+0x90/0x168
[   25.430393] LR is at skb_release_all+0x28/0x2c
[   25.434842] pc : [<807435b0>] lr : [<80742ba0>] psr: 200e0013 5aa1141e
[   25.464633] Flags: nzCv  IRQs on  FIQs on  Mode SVC_32 ISA ARM Segment none
[   25.477524] Process RX-Thread (pid: 317, stack limit = 0x(ptrval))
[   25.483709] Stack: (0xedf69ed8 to 0xedf6a000)
[   25.569907] Backtrace:
[   25.572368] [<80743520>] (skb_release_data) from [<80742ba0>]
       (skb_release_all+0x28/0x2c)
[   25.580555] r9:7f00258c r8:00000001 r7:ee355000 r6:eddab0d0
       r5:eddab000 r4:eddbb840
[   25.588308] [<80742b78>] (skb_release_all) from [<807432cc>]
       (consume_skb+0x30/0x50)
[   25.596055] r5:eddab000 r4:eddbb840
[   25.599648] [<8074329c>] (consume_skb) from [<7f00117c>]
       (rsi_usb_rx_thread+0x64/0x12c [rsi_usb])
[   25.608524] r5:eddab000 r4:eddbb840
[   25.612116] [<7f001118>] (rsi_usb_rx_thread [rsi_usb]) from
       [<80142750>] (kthread+0x11c/0x15c)
[   25.620735] r10:ee9ff9e0 r9:edcde3b8 r8:ee355000 r7:edf68000
       r6:edd3a780 r5:00000000
[   25.628567] r4:edcde380
[   25.631110] [<80142634>] (kthread) from [<801010e8>]
       (ret_from_fork+0x14/0x2c)
[   25.638336] Exception stack(0xedf69fb0 to 0xedf69ff8)
[   25.682929] ---[ end trace 8236a5496f5b5d3b ]---

Signed-off-by: Siva Rebbagondla <siva.rebbagondla@redpinesignals.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agobrcmfmac: fix wrong strnchr usage
Rasmus Villemoes [Wed, 22 Aug 2018 13:22:15 +0000 (15:22 +0200)]
brcmfmac: fix wrong strnchr usage

strnchr takes arguments in the order of its name: string, max bytes to
read, character to search for. Here we're passing '\n' aka 10 as the
buffer size, and searching for sizeof(buf) aka BRCMF_DCMD_SMLEN aka
256 (aka '\0', since it's implicitly converted to char) within those 10
bytes.

Just interchanging the last two arguments would still leave a bug,
because if we've been successful once, there are not sizeof(buf)
characters left after the new value of p.

Since clmver is immediately afterwards passed as a %s argument, I assume
that it is actually a properly nul-terminated string. For that case, we
have strreplace().

Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agort2x00: use simple_read_from_buffer()
Dan Carpenter [Wed, 22 Aug 2018 10:41:26 +0000 (13:41 +0300)]
rt2x00: use simple_read_from_buffer()

The problem with this copy_to_user() calls is that they don't ensure
that "size" is less than the "length" which the user provided.

Obviously, this is debugfs and "size" is normally going to be very small
so it probably doesn't matter, but this is the correct thing to do.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agomwifex: free rx_cmd skb in suspended state
Ganapathi Bhat [Tue, 21 Aug 2018 07:53:02 +0000 (13:23 +0530)]
mwifex: free rx_cmd skb in suspended state

USB suspend handler will kill the presubmitted rx_cmd URB. This
triggers a call to the corresponding URB complete handler, which
will free the rx_cmd skb, associated with rx_cmd URB. Due to a
possible race betwen suspend handler and main thread, depicted in
'commit bfcacac6c84b ("mwifiex: do no submit URB in suspended
state")', it is possible that the rx_cmd skb will fail to get
freed. This causes a memory leak, since the resume handler will
always allocate a new rx_cmd skb.

To fix this, free the rx_cmd skb in mwifiex_usb_submit_rx_urb, if
the device is in suspended state.

Signed-off-by: Vidya Dharmaraju <vidyad@marvell.com>
Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Ganapathi Bhat <gbhat@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agomwifiex: do no submit URB in suspended state
Ganapathi Bhat [Tue, 21 Aug 2018 07:53:01 +0000 (13:23 +0530)]
mwifiex: do no submit URB in suspended state

There is a possible race between USB suspend and main thread:

1. After processing the command response, main thread will submit
rx_cmd URB back so as to process next command response, by
calling mwifiex_usb_submit_rx_urb.

2. During USB suspend, the suspend handler will check if rx_cmd
URB is pending(submitted) and if true, kill this URB.

There is a possible race between #1 and #2, where rx_cmd URB will
be submitted by main thread(#1) after the suspend handler check
in #2.

To fix this, check if device is already suspended in
mwifiex_usb_submit_rx_urb, in which case do not submit the URB.

Signed-off-by: Vidya Dharmaraju <vidyad@marvell.com>
Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Ganapathi Bhat <gbhat@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agortl8187: Fix warning generated when strncpy() destination length matches the sixe...
Larry Finger [Mon, 20 Aug 2018 18:48:31 +0000 (13:48 -0500)]
rtl8187: Fix warning generated when strncpy() destination length matches the sixe argument

In gcc8, when the 3rd argument (size) of a call to strncpy() matches the
length of the first argument, the compiler warns of the possibility of an
unterminated string. Using strlcpy() forces a null at the end.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agoorinoco: remove unused array encaps_hdr and macro ENCAPS_OVERHEAD
Colin Ian King [Thu, 16 Aug 2018 12:59:44 +0000 (13:59 +0100)]
orinoco: remove unused array encaps_hdr and macro ENCAPS_OVERHEAD

Array encaps_hdr and macro ENCAPS_OVERHEAD are declared but are
not being used, hence they are redundant and can be removed.

Cleans up clang warning:
warning: 'encaps_hdr' defined but not used [-Wunused-const-variable=]

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agobrcmfmac: add 4-way handshake offload detection for FT-802.1X
Chung-Hsien Hsu [Wed, 15 Aug 2018 10:32:40 +0000 (05:32 -0500)]
brcmfmac: add 4-way handshake offload detection for FT-802.1X

Add 4-way handshake offload detection for FT with EAP authentication.

Signed-off-by: Chung-Hsien Hsu <stanley.hsu@cypress.com>
Signed-off-by: Chi-Hsien Lin <chi-hsien.lin@cypress.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agobrcmfmac: add FT-based AKMs in brcmf_set_key_mgmt() for FT support
Chung-Hsien Hsu [Wed, 15 Aug 2018 10:32:39 +0000 (05:32 -0500)]
brcmfmac: add FT-based AKMs in brcmf_set_key_mgmt() for FT support

Add WLAN_AKM_SUITE_FT_8021X and WLAN_AKM_SUITE_FT_PSK in
brcmf_set_key_mgmt() for FT support.

Signed-off-by: Chung-Hsien Hsu <stanley.hsu@cypress.com>
Signed-off-by: Chi-Hsien Lin <chi-hsien.lin@cypress.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agobrcmfmac: add CYW89342 mini-PCIe device
Jia-Shyr Chuang [Wed, 15 Aug 2018 09:23:09 +0000 (04:23 -0500)]
brcmfmac: add CYW89342 mini-PCIe device

CYW89342 is a 2x2 MIMO, 802.11a/b/g/n/ac for WLAN. It is a member of
4355/4359 family.

Signed-off-by: Jia-Shyr Chuang <saint.chuang@cypress.com>
Signed-off-by: Chi-Hsien Lin <chi-hsien.lin@cypress.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agolibertas_tf: prevent underflow in process_cmdrequest()
Dan Carpenter [Tue, 14 Aug 2018 09:07:48 +0000 (12:07 +0300)]
libertas_tf: prevent underflow in process_cmdrequest()

If recvlength is less than MESSAGE_HEADER_LEN (4) we would end up
corrupting memory.

Fixes: c305a19a0d0a ("libertas_tf: usb specific functions")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agomt76x2u: Add support for Alfa AWUS036ACM
Rosen Penev [Sat, 11 Aug 2018 23:33:52 +0000 (16:33 -0700)]
mt76x2u: Add support for Alfa AWUS036ACM

A 2x2 unit with a USB dock for optimal placement.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
Acked-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agocw1200: Remove extra parentheses
Varsha Rao [Wed, 25 Jul 2018 19:00:07 +0000 (21:00 +0200)]
cw1200: Remove extra parentheses

Remove unnecessary parentheses to fix the extraneous parentheses clang
warning.

Signed-off-by: Varsha Rao <rvarsha016@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agobrcmsmac: Remove extra parentheses
Varsha Rao [Wed, 25 Jul 2018 18:56:22 +0000 (20:56 +0200)]
brcmsmac: Remove extra parentheses

Remove the unnecessary parentheses to fix the clang warning of
extraneous parentheses.

Signed-off-by: Varsha Rao <rvarsha016@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agoMerge tag 'iwlwifi-next-for-kalle-2018-08-31' of git://git.kernel.org/pub/scm/linux...
Kalle Valo [Fri, 31 Aug 2018 15:05:12 +0000 (18:05 +0300)]
Merge tag 'iwlwifi-next-for-kalle-2018-08-31' of git://git./linux/kernel/git/iwlwifi/iwlwifi-next

iwlwifi patches for 4.20

* One fix for channel-switch handling;
* Some improvements in the HE code;
* New FW API handling;
* Improvements in the debugging code;
* Some improvements in the PCI recovery mechanism;
* Support for a few FW API changes;
* Enable a new scanning feature;
* Continued work on HE (mostly radiotap);
* Some fixes in the FW rate-scaling implementation;
* Other clean-ups and small fixes and improvements;

5 years agoiwlwifi: fix devices with PCI Device ID 0x34F0 and 11ac RF modules
Luca Coelho [Wed, 15 Aug 2018 19:07:25 +0000 (22:07 +0300)]
iwlwifi: fix devices with PCI Device ID 0x34F0 and 11ac RF modules

The devices with PCI device ID 0x34F0 are part of the SoC and can be
combined with some different external RF modules.  The configuration
for these devices should reflect that, but are currently mixed up.  To
avoid confusion with discrete devices, add part of the firmware to be
used and the official name of the device to the cfg structs.

This is least reorganization possible (without messing things even
more) that could be done as a bugfix for this SoC.  Further
reorganization of this code will be done separately.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoRevert "iwlwifi: allow memory debug TLV to specify the memory type"
Johannes Berg [Thu, 17 May 2018 14:02:36 +0000 (16:02 +0200)]
Revert "iwlwifi: allow memory debug TLV to specify the memory type"

This reverts the addition of memory type in the memory debug TLV
as the firmware never ended up implementing it, and now the PRPH
dump is added as a different TLV.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: activate fragmented EBS in case of fragmented scan
Ayala Beker [Tue, 1 May 2018 11:40:03 +0000 (14:40 +0300)]
iwlwifi: mvm: activate fragmented EBS in case of fragmented scan

In case of fragmented scan on the HB channels, configure EBS to be
fragmented as well.

Signed-off-by: Ayala Beker <ayala.beker@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: refactor txq_alloc for supporting more command type
Golan Ben Ami [Tue, 3 Apr 2018 15:37:51 +0000 (18:37 +0300)]
iwlwifi: refactor txq_alloc for supporting more command type

Support more txq_alloc command types by moving the command declaration
to the gen specific area.  While at it, move some of the code segments
to a common place for re-use.

Signed-off-by: Golan Ben Ami <golan.ben.ami@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: decode HE TB PPDU data
Johannes Berg [Wed, 9 May 2018 09:53:04 +0000 (11:53 +0200)]
iwlwifi: mvm: decode HE TB PPDU data

Decode the HE TB PPDU data that we get in sniffer mode
and use it to populate the HE radiotap information.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: rs-fw: support dcm
Naftali Goldstein [Tue, 1 May 2018 11:40:44 +0000 (14:40 +0300)]
iwlwifi: rs-fw: support dcm

Add flags to the tlc_cfg_cmd for signaling peer support for receiving
Dual Carrier Modulation at BPSK (MCS 0), and set them according to peer
capabilities.

Signed-off-by: Naftali Goldstein <naftali.goldstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: rs-fw: enable STBC in he correctly
Naftali Goldstein [Tue, 1 May 2018 11:36:23 +0000 (14:36 +0300)]
iwlwifi: rs-fw: enable STBC in he correctly

In the HE phy capabilities IE there are 2 bits to signal support for
STBC in bandwidths of 80Mhz or less, and of 160Mhz.
Use these bits to determine STBC support if this IE exists.

Signed-off-by: Naftali Goldstein <naftali.goldstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: avoid code duplication in stopping fw debug data recording
Shahar S Matityahu [Thu, 3 May 2018 06:52:12 +0000 (09:52 +0300)]
iwlwifi: avoid code duplication in stopping fw debug data recording

Make all FW debug data stop recording flows to use
iwl_fw_dbg_stop_recording function instead of writing to FW
registers directly.

Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: change monitor DMA to be coherent
Shahar S Matityahu [Thu, 3 May 2018 06:41:11 +0000 (09:41 +0300)]
iwlwifi: change monitor DMA to be coherent

Allow access to the memory by the host and the device simultaneously.
This will be needed in some future patches.

Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: always init rs_fw with 20MHz bandwidth rates
Naftali Goldstein [Thu, 10 May 2018 17:32:36 +0000 (20:32 +0300)]
iwlwifi: mvm: always init rs_fw with 20MHz bandwidth rates

As with the non-offloaded rs case, during assoc on the ap side the phy
context is set to 20MHz until authorization of a client that supports
wider channel-widths. Support this by sending the initial
tlc_config_cmd with max supported channel width of 20MHz until
authorization succeeds.

Fixes: 6b7a5aea71b3 ("iwlwifi: mvm: always init rs with 20mhz bandwidth rates")
Signed-off-by: Naftali Goldstein <naftali.goldstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: send BCAST management frames to the right station
Emmanuel Grumbach [Mon, 7 May 2018 13:46:35 +0000 (16:46 +0300)]
iwlwifi: mvm: send BCAST management frames to the right station

Probe responses were sent to the multicast station while
they should be routed to the broadcast station.
This has no negative effect since the frame was still
routed to the right queue, but it looked very fishy
to send a frame to a (queue, station) tuple where
'queue' is not mapped to 'station'.

Fixes: 7c305de2b954 ("iwlwifi: mvm: Direct multicast frames to the correct station")
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: Add missing 11n disable module parameter check
Ido Yariv [Tue, 1 May 2018 01:21:19 +0000 (21:21 -0400)]
iwlwifi: Add missing 11n disable module parameter check

In the new method of parsing the NVM the disable_11n module parameter
isn't consulted when setting up HT capabilities. Fix this.

Signed-off-by: Ido Yariv <idox.yariv@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: remove channel 2 from HE radiotap if not applicable
Johannes Berg [Thu, 3 May 2018 11:35:08 +0000 (13:35 +0200)]
iwlwifi: mvm: remove channel 2 from HE radiotap if not applicable

If the bandwidth is only 20 MHz, then the second channel doesn't
exist, but the hardware reports the CRC was OK. Suppress the data
of the second channel in the HE radiotap in this case, by marking
it as not known.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: report # of LTF symbols for extended range SU PPDUs
Johannes Berg [Thu, 3 May 2018 10:52:16 +0000 (12:52 +0200)]
iwlwifi: mvm: report # of LTF symbols for extended range SU PPDUs

This is the same as for SU PPDUs, so it's easy to do.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: properly decode HE GI duration
Johannes Berg [Wed, 2 May 2018 20:53:34 +0000 (22:53 +0200)]
iwlwifi: mvm: properly decode HE GI duration

The GI duration depends on the frame type in some cases,
take that into account when decoding for radiotap.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: put LTF symbol size into HE radiotap
Johannes Berg [Wed, 2 May 2018 09:20:00 +0000 (11:20 +0200)]
iwlwifi: mvm: put LTF symbol size into HE radiotap

I evidently completely confused "number of LTF symbols" and "LTF size".
Radiotap was reporting the former, while I thought it was the latter,
and we really need both.

Add the LTF symbol size into the newly defined field in radiotap.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: implement extended HE-MU sniffer API
Johannes Berg [Thu, 26 Apr 2018 11:14:26 +0000 (13:14 +0200)]
iwlwifi: mvm: implement extended HE-MU sniffer API

Implement the extended HE-MU info type decoding to show the HE-SIG-B
common contents in the HE-MU radiotap field.

The DW4 data is partially overwritten by the hardware in all cases, so
only the higher 16 bits can be used.  To be able to use it for the HE
SIG-B common data anyway, move the bits around in the following way:

  SIG-B common 0: DW 4 -> DW 7
  SIG-B common 1: DW 7 -> DW 8
  SIG-B common 2: DW 8 -> DW 4 (upper half)

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: remove all occurrences of the FSF address paragraph
Luca Coelho [Fri, 24 Aug 2018 07:50:45 +0000 (10:50 +0300)]
iwlwifi: remove all occurrences of the FSF address paragraph

The Free Software Foundation address is superfluous and causes
checkpatch to issue a warning when present.  Remove all paragraphs
with FSF's address to prevent that.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: pcie: set interrupt coalescing also for gen2
Sara Sharon [Mon, 23 Apr 2018 06:38:41 +0000 (09:38 +0300)]
iwlwifi: pcie: set interrupt coalescing also for gen2

We offloaded all the RX configuration of init to firmware. However,
the configuration of interrupt coalescing was left hanging - it wasn't
offloaded nor was it written by host.

This write to the CSR is allowed in gen2, so the host can do it.
Without it we have various issues with RX fullness.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: Send LQ command as async when necessary
Avraham Stern [Thu, 3 May 2018 12:02:16 +0000 (15:02 +0300)]
iwlwifi: mvm: Send LQ command as async when necessary

The parameter that indicated whether the LQ command should be sent
as sync or async was removed, causing the LQ command to be sent as
sync from interrupt context (e.g. from the RX path). This resulted
in a kernel warning: "scheduling while atomic" and failing to send
the LQ command, which ultimately leads to a queue hang.

Fix it by adding back the required parameter to send the command as
sync only when it is allowed.

Fixes: d94c5a820d10 ("iwlwifi: mvm: open BA session only when sta is authorized")
Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: avoid sending too many BARs
Sara Sharon [Tue, 1 May 2018 12:12:08 +0000 (15:12 +0300)]
iwlwifi: mvm: avoid sending too many BARs

When we receive TX response, we may release a few packets
due to a hole that was closed in the transmission window.

However, if that frame failed, we will mark all the released
frames as failed and will send multiple BARs.

This affects statistics badly, and cause unnecessary frames
transmission.

Instead, mark all the following packets as success, with the
desired result of sending a bar for the failed frame only.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: fix BAR seq ctrl reporting
Sara Sharon [Tue, 1 May 2018 11:54:22 +0000 (14:54 +0300)]
iwlwifi: mvm: fix BAR seq ctrl reporting

There is a bug in FW where the sequence control may be
incorrect, and the driver overrides it with the value
of the ieee80211 header.

However, in BAR there is no sequence control in the header,
which result with arbitrary sequence.

This access to an unknown location is bad and it makes the
logs very confusing - so fix it.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: improve the flow when a NIC is disconnected
Emmanuel Grumbach [Tue, 1 May 2018 11:23:56 +0000 (14:23 +0300)]
iwlwifi: improve the flow when a NIC is disconnected

When the NIC is disconnected, we just can't do anything
besides seeking for help from the bus driver.  Dumping the
device's memory is not necessary and just bloats the logs
with unusable data.  Moreover, asking mac80211 to restart
the hardware is also useless.  Bypass all this.

Also, use the STATUS_TRANS_DEAD status bit instead of a
bool inside the transport layer. The advantage of this is
that now, the transport and the op_mode can know what is the
situation and bypass the useless recovery steps mentioned
above.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: support Coex Schema 2
Erel Geron [Mon, 23 Apr 2018 08:43:02 +0000 (11:43 +0300)]
iwlwifi: mvm: support Coex Schema 2

The new coex schema requires moving to SISO only when BT AG is 4.
Adjust the SISO criteria according to the coex schema version reported
by firmware.

Signed-off-by: Erel Geron <erelx.geron@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: pcie: avoid unnecessary work if NIC is disconnected
Matt Chen [Thu, 19 Apr 2018 07:18:00 +0000 (15:18 +0800)]
iwlwifi: pcie: avoid unnecessary work if NIC is disconnected

When the NIC is disconnected from PCI bus, we are not
able to access it anymore. Check the status to avoid
some unnecessary work so can improve the performance.
It will help to make PCI bus rescan to bring back the
device much faster.

The real test is able to improve 7 seconds.

[w/o patch] It takes around 9 seconds
..
2018-04-20T01:22:39.691929-07:00 WARNING kernel:
[   66.335881] Timeout waiting for hardware access (CSR_GP_CNTRL 0xffffffff)
..
2018-04-20T01:22:48.101094-07:00 INFO kernel:
[   74.747364] iwlwifi 0000:01:00.0: loaded firmware version 29.610311.0 op_mode iwlmvm

[w/a patch] It takes about 2 seconds.
..
2018-04-20T01:18:16.454087-07:00 WARNING kernel:
[   75.966860] Timeout waiting for hardware access (CSR_GP_CNTRL 0xffffffff)
..
2018-04-20T01:18:18.602717-07:00 INFO kernel:
[   78.116132] iwlwifi 0000:01:00.0: loaded firmware version 29.610311.0 op_mode iwlmvm
..

Fixes: 49564a806fc5 ("iwlwifi: pcie: remove non-responsive device")
Signed-off-by: Matt Chen <matt.chen@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: don't WARN on trying to dump dead firmware
Johannes Berg [Thu, 19 Apr 2018 08:57:08 +0000 (10:57 +0200)]
iwlwifi: don't WARN on trying to dump dead firmware

There's no point in warning here, the user will just get an
error back to the debugfs file write, and warning just makes
it seem like there's an internal consistency problem when in
reality the user just happened to hit this at a bad time.
Remove the warning.

Fixes: f45f979dc208 ("iwlwifi: mvm: disable dbg data collect when fw isn't alive")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: add 80211 hdr offset to trace data
Mordechay Goodstein [Tue, 10 Apr 2018 15:19:49 +0000 (18:19 +0300)]
iwlwifi: add 80211 hdr offset to trace data

Every rx mpdu cmd is built from cmd_hdr | 80211_hdr.  The problem is
that the size of cmd_hdr changes with API changes and we don't know
where the 80211_hdr starts.

By adding the size of cmd_hdr dynamically, we can ensure that we always
know how to parse mpdu frames, without dependending on the API changes.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: pcie: make gen2 of apm_init non-static
Golan Ben Ami [Thu, 22 Mar 2018 12:57:11 +0000 (14:57 +0200)]
iwlwifi: pcie: make gen2 of apm_init non-static

This will allow using the same init in future generations.

Signed-off-by: Golan Ben Ami <golan.ben.ami@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: pcie: store the default rxq number
Golan Ben Ami [Sun, 25 Mar 2018 08:28:33 +0000 (11:28 +0300)]
iwlwifi: pcie: store the default rxq number

Store the default rxq number in a variable, so we won't need
to use the actual number in the code.

Signed-off-by: Golan Ben Ami <golan.ben.ami@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: support new reduce tx power FW API.
Haim Dreyfuss [Thu, 29 Mar 2018 11:13:54 +0000 (14:13 +0300)]
iwlwifi: mvm: support new reduce tx power FW API.

Update reduce tx power command API to be compatible with new FW API.

Signed-off-by: Haim Dreyfuss <haim.dreyfuss@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: skip EBS in low latency mode while fragmented scan isn't supported
Ayala Beker [Thu, 8 Mar 2018 10:20:46 +0000 (12:20 +0200)]
iwlwifi: mvm: skip EBS in low latency mode while fragmented scan isn't supported

While associated in low latency mode, or when traffic load is high,
don't enable EBS in scan request if fragmented EBS is not supported
by the FW.

Signed-off-by: Ayala Beker <ayala.beker@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: remove FSF's address from the license notice
Golan Ben Ami [Wed, 4 Apr 2018 05:56:05 +0000 (08:56 +0300)]
iwlwifi: remove FSF's address from the license notice

The Free Software Foundation's address shouldn't be in the license
notice anymore, and some of our check scripts complain about it
(via checkpatch.pl).  Remove the address to silence it.

Signed-off-by: Golan Ben Ami <golan.ben.ami@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: remove dump_regs() from transport ops
Johannes Berg [Wed, 11 Apr 2018 14:17:00 +0000 (16:17 +0200)]
iwlwifi: remove dump_regs() from transport ops

This is used only within PCIe, and there's no reason to go through
the transport methods for a function call within PCIe itself.
Remove the dump_regs() method and call the function directly.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: pcie: make non-static hcmd and rx code
Golan Ben Ami [Sun, 25 Mar 2018 06:57:08 +0000 (09:57 +0300)]
iwlwifi: pcie: make non-static hcmd and rx code

Allow other device generations to use the utilities that
are used to send and reclaim host commands and to allocate
rx, by making it non-static.

Signed-off-by: Golan Ben Ami <golan.ben.ami@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: pcie: allow using tx init for other queues but the command queue
Golan Ben Ami [Sun, 25 Mar 2018 13:57:42 +0000 (16:57 +0300)]
iwlwifi: pcie: allow using tx init for other queues but the command queue

We would like to allow using tx init code for other queues but
the command queue - for newer devices.

Signed-off-by: Golan Ben Ami <golan.ben.ami@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: drop packets with bad status in CD
Sara Sharon [Mon, 9 Apr 2018 11:00:27 +0000 (14:00 +0300)]
iwlwifi: drop packets with bad status in CD

We need to drop packets with errors (such as replay,
MIC, ICV, conversion, duplicate and so on).

Drop invalid packets, put the status bits in the metadata and
move the enum definition to the correct place (FW API header).

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: enable reading the value of delay in timestamp_marker cmd
Mordechay Goodstein [Mon, 9 Apr 2018 12:52:16 +0000 (15:52 +0300)]
iwlwifi: enable reading the value of delay in timestamp_marker cmd

The user can validate what was written to the file by reading it,
and check what is the default value before changing it.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: move he RX handling to a separate function
Sara Sharon [Mon, 9 Apr 2018 08:20:09 +0000 (11:20 +0300)]
iwlwifi: mvm: move he RX handling to a separate function

The HE code is bloating the RX handling, and makes it too big.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: move he RX handling to a separate function
Sara Sharon [Mon, 9 Apr 2018 08:20:09 +0000 (11:20 +0300)]
iwlwifi: mvm: move he RX handling to a separate function

The HE code is bloating the RX handling, and makes it too big.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: add support for RX_AMPDU_EOF bit for HE SU
Liad Kaufman [Sun, 1 Apr 2018 14:35:43 +0000 (17:35 +0300)]
iwlwifi: mvm: add support for RX_AMPDU_EOF bit for HE SU

Current implementation turns this bit on only for HE MU.

Signed-off-by: Liad Kaufman <liad.kaufman@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: add bss color to radiotap
Liad Kaufman [Sun, 4 Mar 2018 12:21:49 +0000 (14:21 +0200)]
iwlwifi: mvm: add bss color to radiotap

Add BSS color to the HE radiotap.

Signed-off-by: Liad Kaufman <liad.kaufman@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: add required include to iwl-fh.h
Golan Ben Ami [Tue, 3 Apr 2018 13:39:18 +0000 (16:39 +0300)]
iwlwifi: add required include to iwl-fh.h

Some variables are used in iwl-fh.h, but their type's definition
isn't included. This may break compilation when using the file
without including the necessary dependencies first.

Add the required include to iwl-fh.h.

Signed-off-by: Golan Ben Ami <golan.ben.ami@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: add NOA and CSA to a probe response
Gregory Greenman [Mon, 22 Jan 2018 05:38:12 +0000 (07:38 +0200)]
iwlwifi: mvm: add NOA and CSA to a probe response

A probe response built by a P2P GO should contain:
1. CSA/eCSA IE when relevant
2. If the corresponding probe request had P2P IE, then
need to add P2P IE with NOA attributes.

However, the NOA attributes and the updated channel switch
counter are known only to the FW. The solution is that FW
will send a notification with the relevant probe response
data and the driver will save it and update the probe
response accordingly.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: support new WoWLAN status FW API
Luca Coelho [Mon, 26 Mar 2018 07:24:18 +0000 (10:24 +0300)]
iwlwifi: mvm: support new WoWLAN status FW API

A new FW API has been added for WOWLAN_GET_STATUSES to support
multiple GTK keys and IGTK keys.  Check the respective TLV and use the
new API when it is set.

Let most of the code use the new version (v7) and convert the old
version (v6) to the new one when needed.

Also refactor some functions a bit so that they can be reused more
easily.  Particularly the part that calls WOWLAN_GET_STATUSES which is
reused in D3 and D0i3.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: protect D0i3 code behind CONFIG_PM
Luca Coelho [Tue, 21 Aug 2018 04:24:18 +0000 (07:24 +0300)]
iwlwifi: mvm: protect D0i3 code behind CONFIG_PM

Runtime PM can only be used if CONFIG_PM is set.  Move all the d0i3
code (which is essentially runtime PM) behind the CONFIG_PM flag.

This prevents undefined usage of some functions when CONFIG_PM is not
defined.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: mvm: remove duplicate if in iwl_mvm_setup_connection_keep()
Luca Coelho [Tue, 27 Feb 2018 14:37:56 +0000 (16:37 +0200)]
iwlwifi: mvm: remove duplicate if in iwl_mvm_setup_connection_keep()

We repeated the same if twice in a row.  Remove the second one and
move the code block into the previous one.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: add d3 debug data support
Shahar S Matityahu [Mon, 29 Jan 2018 09:05:37 +0000 (11:05 +0200)]
iwlwifi: add d3 debug data support

During d3, the firmware records debug data into internal buffer
if debug data collection occurs, collect the data that was written to the
buffer

Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: set the tid for non-QOS frames to zero
Avraham Stern [Tue, 20 Mar 2018 15:05:06 +0000 (17:05 +0200)]
iwlwifi: set the tid for non-QOS frames to zero

The tid for Non-QOS frames is set to IWL_MAX_TID_COUNT. This value
is also used for configuring the queue for non-QOS data. However, this
tid is used by the FW for management queues. As a result, the FW
does not encrypt non-QOS data frames.

Fix this by setting the tid for non-QOS data frames to zero, which
is a valid value for data frames in the FW.

This also fixes a bug in sending multicast frames, where the queues
are allocated with tid == 0, but are sent with tid == 8, which may
lead to unexpected behavior.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: runtime: avoid calling debugfs read functions more than once
Naftali Goldstein [Tue, 13 Mar 2018 18:00:35 +0000 (20:00 +0200)]
iwlwifi: runtime: avoid calling debugfs read functions more than once

Upon first calling read() on a debugfs file, invoke
iwl_dbgfs_##name##_read and store the response buffer on the heap, so
subsequent read() calls don't need to invoke said function again.

This is done because cat etc will call read() repeatedly until EOF is
reached (or read() returns 0), which in the current implementation will
cause said function to be invoked multiple times.

With the current implementation this can also cause buggy behavior in
some weird edge cases where the first invocation returns a string of
length n, and the second of length m>n: The last m-n characters of
the second invocation will be printed to screen.

Signed-off-by: Naftali Goldstein <naftali.goldstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: remove unused TLC debugging commands
Luca Coelho [Sat, 18 Aug 2018 07:16:29 +0000 (10:16 +0300)]
iwlwifi: remove unused TLC debugging commands

The iwl_dhc_tlc_cmd struct is not used, so remove it, including the
iwl_tcl_debug_flags enumeration.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoiwlwifi: turn timestamp marker cmd off by default
Mordechay Goodstein [Sun, 11 Feb 2018 18:34:19 +0000 (20:34 +0200)]
iwlwifi: turn timestamp marker cmd off by default

Don't enable debugging timestamps by default, so we save power by not
waking up the FW with timestamp commands.

Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
5 years agoMerge ath-next from git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git
Kalle Valo [Tue, 28 Aug 2018 16:39:22 +0000 (19:39 +0300)]
Merge ath-next from git://git./linux/kernel/git/kvalo/ath.git

ath.git patches for 4.20. Major changes:

ath10k

* limit available channels via DT ieee80211-freq-limit

wil6210

* add 802.11r Fast Roaming support for AP and station modes

* add support for channel 4

5 years agoath10k: limit available channels via DT ieee80211-freq-limit
Sven Eckelmann [Fri, 24 Aug 2018 12:04:59 +0000 (15:04 +0300)]
ath10k: limit available channels via DT ieee80211-freq-limit

Tri-band devices (1x 2.4GHz + 2x 5GHz) often incorporate special filters in
the RX and TX path. These filtered channel can in theory still be used by
the hardware but the signal strength is reduced so much that it makes no
sense.

There is already a DT property to limit the available channels but ath10k
has to manually call this functionality to limit the currrently set wiphy
channels further.

Signed-off-by: Sven Eckelmann <sven.eckelmann@openmesh.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agowil6210: fix invalid memory access for rx_buff_mgmt debugfs
Dedy Lansky [Thu, 23 Aug 2018 11:47:13 +0000 (14:47 +0300)]
wil6210: fix invalid memory access for rx_buff_mgmt debugfs

Check rx_buff_mgmt is allocated before accessing its internal fields.

Signed-off-by: Dedy Lansky <dlansky@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agowil6210: add FT roam support for AP and station
Ahmad Masri [Thu, 23 Aug 2018 11:47:12 +0000 (14:47 +0300)]
wil6210: add FT roam support for AP and station

This feature is needed for enterprise APs and clients to enable
fast roaming as defined in 802.11r between APs in the same ESS.

On AP side, this feature is supported only when disable_ap_sme
is enabled.

Signed-off-by: Ahmad Masri <amasri@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agowil6210: fix RX buffers release and unmap
Maya Erez [Thu, 23 Aug 2018 11:47:11 +0000 (14:47 +0300)]
wil6210: fix RX buffers release and unmap

RX SKBs are released in both wil6210 rmmod and RX handle.
As there is no lock to protect the buffers DMA unmap,
the SKB pointer in buff_arr is used to check if the buffer
memory was already released.
Setting wil->rx_buff_mgmt.buff_arr[buff_id].skb to NULL before the DMA
memory unmap will prevent duplicate unmapping of the same memory.
Move the buffer ID to the free list also in case the SKB is NULL.

Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agowil6210: prevent usage of tx ring 0 for eDMA
Maya Erez [Thu, 23 Aug 2018 11:47:10 +0000 (14:47 +0300)]
wil6210: prevent usage of tx ring 0 for eDMA

In enhanced DMA ring 0 is used for RX ring, hence TX ring 0
is an unused element in ring_tx and ring2cid_tid arrays.
Initialize ring2cid_tid CID to WIL6210_MAX_CID to prevent a false
match of CID 0.
Go over the ring_tx and ring2cid_tid from wil_get_min_tx_ring_id
and on to prevent access to index 0 in eDMA.

Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agowil6210: allocate rx reorder buffer only if rx reorder is enabled
Maya Erez [Thu, 23 Aug 2018 11:47:09 +0000 (14:47 +0300)]
wil6210: allocate rx reorder buffer only if rx reorder is enabled

wil_addba_rx_request allocates the rx reorder buffer regardless
of use_rx_hw_reordering settings.
Fix this by checking wil->use_rx_hw_reordering before allocating
the reorder buffer.

Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agowil6210: set edma variables only for Talyn-MB devices
Maya Erez [Thu, 23 Aug 2018 11:47:08 +0000 (14:47 +0300)]
wil6210: set edma variables only for Talyn-MB devices

use_rx_hw_reordering is already set to true in wil_set_capabilities
for Talyn-MB devices. Remove its setting from wil_priv_init to
prevent its activation for older chips.
Similarly, move the setting of use_compressed_rx_status to
wil_set_capabilities.

Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agowil6210: drop Rx multicast packets that are looped-back to STA
Dedy Lansky [Thu, 23 Aug 2018 11:47:07 +0000 (14:47 +0300)]
wil6210: drop Rx multicast packets that are looped-back to STA

Delivering a looped-back multicast packet to network stack can cause
higher layer protocols to fail like for example IPv6 DAD.
In STA mode, upon receiving Rx multicast packet, check if the source
MAC address is equal to our own MAC address and if so drop the packet.

Signed-off-by: Dedy Lansky <dlansky@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agowil6210: add support for channel 4
Alexei Avshalom Lazar [Thu, 23 Aug 2018 11:47:06 +0000 (14:47 +0300)]
wil6210: add support for channel 4

wil6210 supports channels 1-3 of the 60GHz band.
New FW added support for channel 4. Add support for channel 4 also in
wil6210 driver.

Signed-off-by: Alexei Avshalom Lazar <ailizaro@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agowil6210: remove reset file from debugfs
Karthick Gopalasubramanian [Thu, 23 Aug 2018 11:47:05 +0000 (14:47 +0300)]
wil6210: remove reset file from debugfs

Reset file is not used and may cause race conditions
with operational driver if used.

Signed-off-by: Karthick Gopalasubramanian <kargop@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agoath6kl: Remove unnecessary parentheses
Varsha Rao [Wed, 25 Jul 2018 18:49:48 +0000 (20:49 +0200)]
ath6kl: Remove unnecessary parentheses

Remove extra parentheses to fix the clang warning of extraneous
parentheses.

Signed-off-by: Varsha Rao <rvarsha016@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agoath9k: fix tx99 with monitor mode interface
Felix Fietkau [Mon, 20 Aug 2018 09:37:51 +0000 (11:37 +0200)]
ath9k: fix tx99 with monitor mode interface

Tx99 is typically configured via a monitor mode interface, which does
not get added to the driver as a vif. Since the code currently expects
a configured virtual interface for tx99, enabling tx99 via debugfs fails.
Since the vif is not needed anyway, remove all checks for it.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
[kvalo@codeaurora.org: s/CPTCFG/CONFIG/]
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agoath9k: remove unused array firstep_table
Colin Ian King [Thu, 16 Aug 2018 12:50:30 +0000 (13:50 +0100)]
ath9k: remove unused array firstep_table

Array firstep_table is being assigned but is never used
hence it is redundant and can be removed.

Cleans up clang warning:
warning: 'firstep_table' defined but not used [-Wunused-const-variable=]

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agoath9k: Remove unnecessary parentheses
Varsha Rao [Wed, 25 Jul 2018 18:53:03 +0000 (20:53 +0200)]
ath9k: Remove unnecessary parentheses

Remove extra parentheses to fix the clang warning of extraneous
parentheses.

Signed-off-by: Varsha Rao <rvarsha016@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agoath10k: use struct_size() in kzalloc()
Gustavo A. R. Silva [Fri, 24 Aug 2018 01:12:47 +0000 (20:12 -0500)]
ath10k: use struct_size() in kzalloc()

One of the more common cases of allocation size calculations is finding
the size of a structure that has a zero-sized array at the end, along
with memory for some number of elements for that array. For example:

struct foo {
int stuff;
        void *entry[];
};

instance = kzalloc(sizeof(struct foo) + sizeof(void *) * count, GFP_KERNEL);

Instead of leaving these open-coded and prone to type mistakes, we can
now use the new struct_size() helper:

instance = kzalloc(struct_size(instance, entry, count), GFP_KERNEL);

This issue was detected with the help of Coccinelle.

Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agoath10k: schedule hardware restart if WMI command times out
Martin Willi [Wed, 22 Aug 2018 07:39:52 +0000 (09:39 +0200)]
ath10k: schedule hardware restart if WMI command times out

When running in AP mode, ath10k sometimes suffers from TX credit
starvation. The issue is hard to reproduce and shows up once in a
few days, but has been repeatedly seen with QCA9882 and a large
range of firmwares, including 10.2.4.70.67.

Once the module is in this state, TX credits are never replenished,
which results in "SWBA overrun" errors, as no beacons can be sent.
Even worse, WMI commands run in a timeout while holding the conf
mutex for three seconds each, making any further operations slow
and the whole system unresponsive.

The firmware/driver never recovers from that state automatically,
and triggering TX flush or warm restarts won't work over WMI. So
issue a hardware restart if a WMI command times out due to missing
TX credits. This implies a connectivity outage of about 1.4s in AP
mode, but brings back the interface and the whole system to a usable
state. WMI command timeouts have not been seen in absent of this
specific issue, so taking such drastic actions seems legitimate.

Signed-off-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agoath10k: skip resetting rx filter for WCN3990
Rakesh Pillai [Mon, 6 Aug 2018 14:48:08 +0000 (20:18 +0530)]
ath10k: skip resetting rx filter for WCN3990

WCN3990 has the MAC_PCU_ADDR1 configured properly
and hence it will not send spurious ack frames
during boot up.

Hence the reset_rx_filter workaround is not needed
for WCN3990. Add a hw_param to indicate if hardware rx
filter reset is needed and skip the reset_rx_filter for
WCN3990.

Tested HW: WCN3990
Tested FW: WLAN.HL.2.0-01188-QCAHLSWMTPLZ-1

Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
5 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Mon, 27 Aug 2018 18:59:39 +0000 (11:59 -0700)]
Merge git://git./linux/kernel/git/davem/net

Pull networking fixes from David Miller:

 1) ICE, E1000, IGB, IXGBE, and I40E bug fixes from the Intel folks.

 2) Better fix for AB-BA deadlock in packet scheduler code, from Cong
    Wang.

 3) bpf sockmap fixes (zero sized key handling, etc.) from Daniel
    Borkmann.

 4) Send zero IPID in TCP resets and SYN-RECV state ACKs, to prevent
    attackers using it as a side-channel. From Eric Dumazet.

 5) Memory leak in mediatek bluetooth driver, from Gustavo A. R. Silva.

 6) Hook up rt->dst.input of ipv6 anycast routes properly, from Hangbin
    Liu.

 7) hns and hns3 bug fixes from Huazhong Tan.

 8) Fix RIF leak in mlxsw driver, from Ido Schimmel.

 9) iova range check fix in vhost, from Jason Wang.

10) Fix hang in do_tcp_sendpages() with tls, from John Fastabend.

11) More r8152 chips need to disable RX aggregation, from Kai-Heng Feng.

12) Memory exposure in TCA_U32_SEL handling, from Kees Cook.

13) TCP BBR congestion control fixes from Kevin Yang.

14) hv_netvsc, ignore non-PCI devices, from Stephen Hemminger.

15) qed driver fixes from Tomer Tayar.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (77 commits)
  net: sched: Fix memory exposure from short TCA_U32_SEL
  qed: fix spelling mistake "comparsion" -> "comparison"
  vhost: correctly check the iova range when waking virtqueue
  qlge: Fix netdev features configuration.
  net: macb: do not disable MDIO bus at open/close time
  Revert "net: stmmac: fix build failure due to missing COMMON_CLK dependency"
  net: macb: Fix regression breaking non-MDIO fixed-link PHYs
  mlxsw: spectrum_switchdev: Do not leak RIFs when removing bridge
  i40e: fix condition of WARN_ONCE for stat strings
  i40e: Fix for Tx timeouts when interface is brought up if DCB is enabled
  ixgbe: fix driver behaviour after issuing VFLR
  ixgbe: Prevent unsupported configurations with XDP
  ixgbe: Replace GFP_ATOMIC with GFP_KERNEL
  igb: Replace mdelay() with msleep() in igb_integrated_phy_loopback()
  igb: Replace GFP_ATOMIC with GFP_KERNEL in igb_sw_init()
  igb: Use an advanced ctx descriptor for launchtime
  e1000: ensure to free old tx/rx rings in set_ringparam()
  e1000: check on netif_running() before calling e1000_up()
  ixgb: use dma_zalloc_coherent instead of allocator/memset
  ice: Trivial formatting fixes
  ...

5 years agoFix up libata MAINTAINERS entry
Jens Axboe [Mon, 27 Aug 2018 14:37:46 +0000 (08:37 -0600)]
Fix up libata MAINTAINERS entry

The email was botched in one entry, and I also forgot to update the
location of the git tree. It'll be under the linux-block umbrella, just
with different branches.

Reported-by: Baruch Siach <baruch@tkos.co.il>
Fixes: 7634ccd2da97 ("libata: maintainership update")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
5 years agonet: sched: Fix memory exposure from short TCA_U32_SEL
Kees Cook [Sun, 26 Aug 2018 05:58:01 +0000 (22:58 -0700)]
net: sched: Fix memory exposure from short TCA_U32_SEL

Via u32_change(), TCA_U32_SEL has an unspecified type in the netlink
policy, so max length isn't enforced, only minimum. This means nkeys
(from userspace) was being trusted without checking the actual size of
nla_len(), which could lead to a memory over-read, and ultimately an
exposure via a call to u32_dump(). Reachability is CAP_NET_ADMIN within
a namespace.

Reported-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Jiri Pirko <jiri@resnulli.us>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoLinux 4.19-rc1 v4.19-rc1
Linus Torvalds [Sun, 26 Aug 2018 21:11:59 +0000 (14:11 -0700)]
Linux 4.19-rc1

5 years agoMerge branch 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 20:39:05 +0000 (13:39 -0700)]
Merge branch 'timers-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull timer update from Thomas Gleixner:
 "New defines for the compat time* types so they can be shared between
  32bit and 64bit builds. Not used yet, but merging them now allows the
  actual conversions to be merged through different maintainer trees
  without dependencies

  We still have compat interfaces for 32bit on 64bit even with the new
  2038 safe timespec/val variants because pointer size is different. And
  for the old style timespec/val interfaces we need yet another 'compat'
  interface for both 32bit native and 32bit on 64bit"

* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  y2038: Provide aliases for compat helpers

5 years agoMerge branch 'ida-4.19' of git://git.infradead.org/users/willy/linux-dax
Linus Torvalds [Sun, 26 Aug 2018 18:48:42 +0000 (11:48 -0700)]
Merge branch 'ida-4.19' of git://git.infradead.org/users/willy/linux-dax

Pull IDA updates from Matthew Wilcox:
 "A better IDA API:

      id = ida_alloc(ida, GFP_xxx);
      ida_free(ida, id);

  rather than the cumbersome ida_simple_get(), ida_simple_remove().

  The new IDA API is similar to ida_simple_get() but better named.  The
  internal restructuring of the IDA code removes the bitmap
  preallocation nonsense.

  I hope the net -200 lines of code is convincing"

* 'ida-4.19' of git://git.infradead.org/users/willy/linux-dax: (29 commits)
  ida: Change ida_get_new_above to return the id
  ida: Remove old API
  test_ida: check_ida_destroy and check_ida_alloc
  test_ida: Convert check_ida_conv to new API
  test_ida: Move ida_check_max
  test_ida: Move ida_check_leaf
  idr-test: Convert ida_check_nomem to new API
  ida: Start new test_ida module
  target/iscsi: Allocate session IDs from an IDA
  iscsi target: fix session creation failure handling
  drm/vmwgfx: Convert to new IDA API
  dmaengine: Convert to new IDA API
  ppc: Convert vas ID allocation to new IDA API
  media: Convert entity ID allocation to new IDA API
  ppc: Convert mmu context allocation to new IDA API
  Convert net_namespace to new IDA API
  cb710: Convert to new IDA API
  rsxx: Convert to new IDA API
  osd: Convert to new IDA API
  sd: Convert to new IDA API
  ...

5 years agoMerge tag 'gcc-plugins-v4.19-rc1-fix' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 18:41:08 +0000 (11:41 -0700)]
Merge tag 'gcc-plugins-v4.19-rc1-fix' of git://git./linux/kernel/git/kees/linux

Pull gcc plugin fix from Kees Cook:
 "Lift gcc test into Kconfig. This is for better behavior when the
  kernel is built with Clang, reported by Stefan Agner"

* tag 'gcc-plugins-v4.19-rc1-fix' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
  gcc-plugins: Disable when building under Clang

5 years agoMerge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 18:25:21 +0000 (11:25 -0700)]
Merge branch 'perf-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull perf updates from Thomas Gleixner:
 "Kernel:
   - Improve kallsyms coverage
   - Add x86 entry trampolines to kcore
   - Fix ARM SPE handling
   - Correct PPC event post processing

  Tools:
   - Make the build system more robust
   - Small fixes and enhancements all over the place
   - Update kernel ABI header copies
   - Preparatory work for converting libtraceevnt to a shared library
   - License cleanups"

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (100 commits)
  tools arch: Update arch/x86/lib/memcpy_64.S copy used in 'perf bench mem memcpy'
  tools arch x86: Update tools's copy of cpufeatures.h
  perf python: Fix pyrf_evlist__read_on_cpu() interface
  perf mmap: Store real cpu number in 'struct perf_mmap'
  perf tools: Remove ext from struct kmod_path
  perf tools: Add gzip_is_compressed function
  perf tools: Add lzma_is_compressed function
  perf tools: Add is_compressed callback to compressions array
  perf tools: Move the temp file processing into decompress_kmodule
  perf tools: Use compression id in decompress_kmodule()
  perf tools: Store compression id into struct dso
  perf tools: Add compression id into 'struct kmod_path'
  perf tools: Make is_supported_compression() static
  perf tools: Make decompress_to_file() function static
  perf tools: Get rid of dso__needs_decompress() call in __open_dso()
  perf tools: Get rid of dso__needs_decompress() call in symbol__disassemble()
  perf tools: Get rid of dso__needs_decompress() call in read_object_code()
  tools lib traceevent: Change to SPDX License format
  perf llvm: Allow passing options to llc in addition to clang
  perf parser: Improve error message for PMU address filters
  ...

5 years agoMerge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 17:13:21 +0000 (10:13 -0700)]
Merge branch 'x86-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull x86 fixes from Thomas Gleixner:

 - Correct the L1TF fallout on 32bit and the off by one in the 'too much
   RAM for protection' calculation.

 - Add a helpful kernel message for the 'too much RAM' case

 - Unbreak the VDSO in case that the compiler desides to use indirect
   jumps/calls and emits retpolines which cannot be resolved because the
   kernel uses its own thunks, which does not work for the VDSO. Make it
   use the builtin thunks.

 - Re-export start_thread() which was unexported when the 32/64bit
   implementation was unified. start_thread() is required by modular
   binfmt handlers.

 - Trivial cleanups

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/speculation/l1tf: Suggest what to do on systems with too much RAM
  x86/speculation/l1tf: Fix off-by-one error when warning that system has too much RAM
  x86/kvm/vmx: Remove duplicate l1d flush definitions
  x86/speculation/l1tf: Fix overflow in l1tf_pfn_limit() on 32bit
  x86/process: Re-export start_thread()
  x86/mce: Add notifier_block forward declaration
  x86/vdso: Fix vDSO build if a retpoline is emitted

5 years agoMerge branch 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 16:55:28 +0000 (09:55 -0700)]
Merge branch 'irq-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull irq update from Thomas Gleixner:
 "A small set of updats/fixes for the irq subsystem:

   - Allow GICv3 interrupts to be configured as wake-up sources to
     enable wakeup from suspend

   - Make the error handling of the STM32 irqchip init function work

   - A set of small cleanups and improvements"

* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/gic-v3: Allow interrupt to be configured as wake-up sources
  irqchip/tango: Set irq handler and data in one go
  dt-bindings: irqchip: renesas-irqc: Document r8a774a1 support
  irqchip/s3c24xx: Remove unneeded comparison of unsigned long to 0
  irqchip/stm32: Fix init error handling
  irqchip/bcm7038-l1: Hide cpu offline callback when building for !SMP

5 years agoMerge branch 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 16:47:00 +0000 (09:47 -0700)]
Merge branch 'locking-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull licking update from Thomas Gleixner:
 "Mark the switch cases which fall through to the next case with the
  proper comment so the fallthrough compiler checks can be enabled"

* 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  futex: Mark expected switch fall-throughs