Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

This crashes the kernel when an runtime overlay is applied.

Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

Convert the driver over the the device model component framework, making
use of the drm encoder/connector helpers. This allows adihdmi to be
dynamically selected as an external encoder for drm drivers like tilcdc
that support the DT graph binding which defines ports and remote-endpoints
to attach external encoders.

Also, this driver was modified by another developer to support audio and
tweak some settings.  Along the way it seems to have been reformatted to
4 space tabs which is hard to work with alongside the standard 8 space tabs
in the kernel coding standard. As such, this is reformatted to standard 8
space tabs so it's a bit more readable.

The component and audio support should be merged into the upstream driver
so this adihdmi driver can be removed.

Signed-off-by: Matt Porter <mporter@konsulko.com>

Hacked driver that has audio support. Use this temporarily until
audio support can be added to the upstream adv7511 driver.

Signed-off-by: Jason Kridner <jdk@ti.com>
[Remove slave hacks and use adv75xx compatible strings]
Signed-off-by: Matt Porter <mporter@konsulko.com>

With 4-way handshake offload for 802.1X authentication, the authorized
flag in CONNECT event should be set for a successful connection with
PMK caching. The connection is identified by checking the existence of
PMKID within the Association Request frame with this patch.

Signed-off-by: Chung-Hsien Hsu <stanley.hsu@cypress.com>

Add authorized flag to CONNECT event. It is used for 802.1X 4-way
handshake offload with PMK caching.

Signed-off-by: Chung-Hsien Hsu <stanley.hsu@cypress.com>

commit 1b5e2423 ("brcmfmac: assure SSID length from firmware is
limited") upstream.

The SSID length as received from firmware should not exceed
IEEE80211_MAX_SSID_LEN as that would result in heap overflow.

Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com>
Reviewed-by: Franky Lin <franky.lin@broadcom.com>
Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

commit a4176ec3 ("brcmfmac: add subtype check for event handling in
data path") upstream.

For USB there is no separate channel being used to pass events
from firmware to the host driver and as such are passed over the
data path. In order to detect mock event messages an additional
check is needed on event subtype. This check is added conditionally
using unlikely() keyword.

Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com>
Reviewed-by: Franky Lin <franky.lin@broadcom.com>
Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

Set F2 blocksize to 256 bytes and watermark to 0x40 for 4359. Also
enable and configure F1 MesBusyCtrl. It fixes DMA error while having
UDP bi-directional traffic.

Signed-off-by: Chung-Hsien Hsu <stanley.hsu@cypress.com>

This patch fixes 43455 CRC error while running throughput test with
suspend/resume stress test.

The continuous failure messages before system crash:
brcmfmac: brcmf_sdiod_sglist_rw: CMD53 sg block read failed -84
brcmfmac: brcmf_sdio_rxglom: glom read of 25600 bytes failed: -5
brcmfmac: brcmf_sdio_rxfail: abort command, terminate frame
brcmfmac: brcmf_sdiod_sglist_rw: CMD53 sg block read failed -84
brcmfmac: brcmf_sdio_rxglom: glom read of 24576 bytes failed: -5
brcmfmac: brcmf_sdio_rxfail: abort command, terminate frame

Signed-off-by: Wright Feng <wright.feng@cypress.com>

With asynchronous suspend/resume feature, suspend and resume callbacks to
be executed in parallel with each other. It makes bus changes the state to
BRCMF_BUS_DOWN before all brcmf_cfg80211_suspend IOVAR executions.
The same situation also happens in resume procedure and causes PM mode
keeps in PM_MAX after resume. In order to fix the race condition, We add
one second sleep in bus suspend and cfg80211 resume function.

Signed-off-by: Wright Feng <wright.feng@cypress.com>

The race condition in brcmf_msgbuf_txflow and brcmf_msgbuf_delete_flowring
makes tx_msghdr writing after brcmf_msgbuf_remove_flowring. Host
driver should delete flowring after txflow complete and all txstatus back,
or pend_8021x_cnt will never be zero and cause every connection 950
milliseconds(MAX_WAIT_FOR_8021X_TX) delay.

Signed-off-by: Wright Feng <wright.feng@cypress.com>

With 4-way handshake offload for 802.1X authentication, the authorized
flag in ROAM event should be set for a successful roaming with PMK
caching. The roaming is identified by checking the existence of PMKID
within the (Re)Association Request frame with this patch.

Signed-off-by: Chung-Hsien Hsu <stanley.hsu@cypress.com>

In brcmfmac driver, all ACs were assgined a FW priority based on the
EDCA parameters from AP. In a specific scenario where EDCA parameters
are configured to be same for all ACs, we propose to use the default
FW priority definition to avoid queuing packets of all ACs to the
same priority queue.
Also in case of fcmode=2, throughput of any AC would have depended on
available credits of a single AC, without this fix.

Fixes 11AC CERT 5.2.33 TXOP Limit test

Signed-off-by: Madhan Mohan R <madhanmohan.r@cypress.com>

It was observed that P2P Cert. 5.1.19: DEVUT responds to Service
Discovery request failed due to DUT did not send GAS Comeback Response
after receiving request from test bed P2P peer. To fix this issue,
we need to add P2P Action Frame retry delay to enhance P2P connection
under VSDB and noisy environment, since the peer can be in other
channels under VSDB.

Signed-off-by: Justin Li <justin.li@cypress.com>

Some PCIE firmwares drop txstatus with pktid=0, that makes packet stays in
host side and not release. If the packet is 8021x packet, the
pend_8021x_cnt will be always larger than 0 and see "Timed out waiting
for no pending 802.1x packets" error when sending key to dongle.

Signed-off-by: Wright Feng <wright.feng@cypress.com>

When using PORT_AUTHORIZED event mechanism for offload FT,
wpa_supplicant started a full EAP exchange after a successful roaming.
It was caused by setting portEnabled to FALSE to get EAP state machine
out of the SUCCESS state and eapSuccess cleared when handling ROAM
event in wap_supplicant.

With this patch, the authorized flag in the ROAM event is used to
indicate the connection is authorized. Wpa_supplicant sets portEnabled
according to the flag and no full EAP exchange is performed after the
roaming.

Signed-off-by: Chung-Hsien Hsu <stanley.hsu@cypress.com>

Commit 503c1fb9 ("cfg80211/nl80211: add a port authorized event")
added the NL80211_CMD_PORT_AUTHORIZED event to indicate that a
connection is authorized. It replaced the PORT_AUTHORIZED attribute and
the authorized flag added in commit f45cbe6e
("nl80211: add authorized flag to ROAM event").

However, for offload FT, using PORT_AUTHORIZED event mechanism induces
wpa_supplicant to start a full EAP exchange after a successful roaming.
This patch adds the flag mechanism back to the ROAM event for drivers
to fix the offload FT roaming issue.

Signed-off-by: Chung-Hsien Hsu <stanley.hsu@cypress.com>

P2P group formation fails since either peer is not able to send go-neg
confirm or dut is not able to send go-neg response. To fix this, retry
limit should be increased and dwell time check should be added.

Signed-off-by: Saint Chuang <saint.chuang@cypress.com>

It was observed that TCP Tx/Rx and UDP Tx showed very low throughput
when fcmode=2. This is caused by incorrect default FIFO priority.
BK FIFO queue was used for default 802.1D BE priority traffic.
Hostapd relies on the correct priority setting to deliver good
throughput.

Signed-off-by: Justin Li <justin.li@cypress.com>

commit d2af9b56 ("brcmfmac: validate user provided data for
memdump before copying") upstream.

In patch "brcmfmac: add support for sysfs initiated coredump", a new
scenario of brcmf_debug_create_memdump was added in which the user of
the function might not necessarily provide prefix data. Hence the
function should not assume the data is always valid and should perform a
check before copying.

Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Franky Lin <franky.lin@broadcom.com>
Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

commit 2fd3877b ("brcmfmac: handle FWHALT mailbox indication")
upstream.

The firmware uses a mailbox to communicate to the host what is going
on. In the driver we validate the bit received. Various people seen
the following message:

 brcmfmac: brcmf_sdio_hostmail: Unknown mailbox data content: 0x40012

Bit 4 is cause of this message, but this actually indicates the firmware
has halted. Handle this bit by giving a more meaningful error message.

Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com>
Reviewed-by: Franky Lin <franky.lin@broadcom.com>
Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

Device is left in non-responding state when unloading FMAC driver in
non-DS1 state if it ever enters DS1 at least once. It leaves that state
only after a hard reset or power cycle 43012.

SWWLAN-137253

Signed-off-by: David Weng <david.weng@cypress.com>

The code is duplicate so remove the line from pcie.c

Signed-off-by: Wright Feng <wright.feng@cypress.com>

Changed mutex locking to completions to make code compatible with
Fully Preemptile Kernel (RT) configuration

Signed-off-by: Brian Henriquez <brian.henriquez@cypress.com>

commit 861cb5eb ("brcmfmac: Fix access point mode") upstream.

Since commit 1204aa17 ("brcmfmac: set WIPHY_FLAG_HAVE_AP_SME flag")
the Raspberry Pi 3 A+ (BCM43455) isn't able to operate in AP mode with
hostapd (device_ap_sme=1 use_monitor=0):

brcmfmac: brcmf_cfg80211_stop_ap: setting AP mode failed -52

So add the missing mgmt_stypes for AP mode to fix this.

Fixes: 1204aa17 ("brcmfmac: set WIPHY_FLAG_HAVE_AP_SME flag")
Suggested-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Stefan Wahren <stefan.wahren@i2se.com>
Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

With the addition of skb_orphan in the datapath, though the throughput
increases in TX path, it introduces an issue by removing the flowcontrol
from upper layer and allowing more data to flow for different access
category.

This workaround is to disable using skb_orphan when running multi-stream
data. This change will not be required in linux 4.15 and later versions.

Signed-off-by: Raveendran Somu <raveendran.somu@cypress.com>

This patch is for adding auto suspend feature in FMAC USB bus.

Signed-off-by: Wright Feng <wright.feng@cypress.com>

New code enables PHY_TEMP event in firmware, receives it, processes
into cfg80211 vendor specific event, and forwards to waiting host.

-1- Enable rx PHY_TEMP event from underlying hardware
-2- Process PHY_TEMP event into vendor specific event
-3- Forward vendor specific event to host layer

Signed-off-by: Robert Trask <robert.trask@cypress.com>

4373 has support of 16 WOWL patterns thus increasing the default value

Signed-off-by: Ryohei Kondo <ryohei.kondo@cypress.com>

AOS is a part of the SDIOD core that becomes active when the rest of
SDIOD is sleeping to keep SDIO bus alive responding to reduced set of
commands.

Transaction between AOS and SDIOD is not protected, and if cmd 52 is
received in AOS and in the middle of response state changed from AOS to
SDIOD, response is corrupted and it causes to SDIO Host controller to
hang.

Signed-off-by: Wright Feng <wright.feng@cypress.com>

Bss info flag definition need to be fixed from 0x2 to 0x4
This flag is for rssi info received on channel.
All Firmware branches defined as 0x4 and this is bug in brcmfmac.

Signed-off-by: Prasanna Kerekoppa <prasanna.kerekoppa@cypress.com>

This change is to add support for printing the firmware
console messges of a USB interface chip to the host.
To enable this feature, build option '-msgtrace' should be
enabled in the firmware. And in the host, debug=0x100000
should be provided as a module parameter.

Signed-off-by: Raveendran Somu <raveendran.somu@cypress.com>