bnxt_en: Add support for BCM957504

Add support for BCM957504 with device ID 1751

Signed-off-by: Erik Burrows <[email protected]>
Signed-off-by: Michael Chan <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

bnxt_en: Update firmware interface spec. to 1.10.0.47.

Firmware error recover is the major change in this spec.

Signed-off-by: Michael Chan <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

net/tls: Move protocol constants from cipher context to tls context

Each tls context maintains two cipher contexts (one each for tx and rx
directions). For each tls session, the constants such as protocol
version, ciphersuite, iv size, associated data size etc are same for
both the directions and need to be stored only once per tls context.
Hence these are moved from 'struct cipher_context' to 'struct
tls_prot_info' and stored only once in 'struct tls_context'.

Signed-off-by: Vakul Garg <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

rt2800: comment and simplify AGC init for RT6352

We do not need separate lines for calculating register values.
Also add comment that value is different than in vendor driver.

Suggested-by: Daniel Golle <[email protected]>
Signed-off-by: Stanislaw Gruszka <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rt2800: enable TX_PIN_CFG_RFRX_EN only for MT7620

The TX_PIN_CFG_RFRX_EN bit was not set on other devices than MT7620,
restore old behavaviour since setting this bit maight not be
correct for older devices.

Fixes: 41977e86c984 ("rt2x00: add support for MT7620")
Signed-off-by: Stanislaw Gruszka <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rt2800: enable TX_PIN_CFG_LNA_PE_ bits per band

Do not enable TX_PIN_CFG_LNA_PE_A* bits for 2.4GHz band and
vice versa TX_PIN_CFG_LNA_PE_G* bits for 5GHz.

Signed-off-by: Stanislaw Gruszka <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rt2x00: remove confusing AGC register

Register 66 was causing issues on RT6352 if set to the same value as
in MTK driver. With 1c reg value device was working fine in both HT20
and HT40 modes.

Signed-off-by: Tomislav Požega <[email protected]>
Signed-off-by: Stanislaw Gruszka <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rt2x00: remove unneeded check

Remove band check from rf53xx channel config routine since all chips
using it are single band.

Signed-off-by: Tomislav Požega <[email protected]>
Signed-off-by: Stanislaw Gruszka <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

carl9170: clean up a clamp() call

The parameter order for clamp is supposed to be clamp(value, low, high).
When we write it in this order it's equivalent to
min(head->plcp[3] & 0x7f, 75) which works in this context where the min
is zero.  But it's not a correct use of the API.

Signed-off-by: Dan Carpenter <[email protected]>
Acked-by: Christian Lamparter <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtl818x_pci: Remove set but not used variables 'io_addr, mem_addr'

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c: In function 'rtl8180_probe':
drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c:1727:15: warning:
 variable 'io_addr' set but not used [-Wunused-but-set-variable]

drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c:1726:16: warning:
 variable 'mem_addr' set but not used [-Wunused-but-set-variable]

They're never used since introduction.

Signed-off-by: YueHaibing <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

orinoco : Replace function name in string with __func__

Replace hard coded function name with __func__, to
improve robustness and to conform to the Linux kernel coding
style. Issue found using checkpatch.

Signed-off-by: Keyur Patel <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

mwifiex: don't advertise IBSS features without FW support

As it is, doing something like

  # iw phy phy0 interface add foobar type ibss

on a firmware that doesn't have ad-hoc support just yields failures of
HostCmd_CMD_SET_BSS_MODE, which happened to return a '-1' error code
(-EPERM? not really right...) and sometimes may even crash the firmware
along the way.

Let's parse the firmware capability flag while registering the wiphy, so
we don't allow attempting IBSS at all, and we get a proper -EOPNOTSUPP
from nl80211 instead.

Fixes: e267e71e68ae ("mwifiex: Disable adhoc feature based on firmware capability")
Signed-off-by: Brian Norris <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

mwifiex: Fix NL80211_TX_POWER_LIMITED

NL80211_TX_POWER_LIMITED was treated as NL80211_TX_POWER_AUTOMATIC,
which is the opposite of what should happen and can cause nasty
regulatory problems.

if/else converted to a switch without default to make gcc warn
on unhandled enum values.

Signed-off-by: Adrian Bunk <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

libertas_tf: remove set but not used variable 'flags'

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/wireless/marvell/libertas_tf/main.c: In function 'lbtf_rx':
drivers/net/wireless/marvell/libertas_tf/main.c:554:15: warning:
 variable 'flags' set but not used [-Wunused-but-set-variable]

It never used and can be removed.

Signed-off-by: YueHaibing <[email protected]>
Acked-by: Steve deRosier <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

libertas_tf: fix signal reporting

Instead of exposing the signal-to-noise ration, calculate the actual signal
level taking the noise floor into account.

Also, flip the SIGNAL_DBM bit on, so that mac80211 exposes the signal
level along with the station info in scan results. This fills
NetworkManager's "nmcli d wifi output" output with colors, bars and joy.

Signed-off-by: Lubomir Rintel <[email protected]>
Reviewed-by: Steve deRosier <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

libertas_tf: move the banner to a more appropriate place

Also, turn it to a dev_info() to make checkpatch.pl happy.

Signed-off-by: Lubomir Rintel <[email protected]>
Reviewed-by: Steve deRosier <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

libertas_tf: don't set URB_ZERO_PACKET on IN USB transfer

It doesn't make sense and the USB core warns on each submit of such
URB, easily flooding the message buffer with tracebacks.

Analogous issue was fixed in regular libertas driver in commit 6528d8804780
("libertas: don't set URB_ZERO_PACKET on IN USB transfer").

Cc: [email protected]
Signed-off-by: Lubomir Rintel <[email protected]>
Reviewed-by: Steve deRosier <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

libertas_tf: lower the debug level of command trace

Logging each and every command response is way too much for INFO level.
Silence this, unless CONFIG_LIBERTAS_THINFIRM_DEBUG has been enabled.

Signed-off-by: Lubomir Rintel <[email protected]>
Reviewed-by: Steve deRosier <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rsi: remove set but not used variables 'info, vif'

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/wireless/rsi/rsi_91x_main.c: In function 'rsi_prepare_skb':
drivers/net/wireless/rsi/rsi_91x_main.c:127:24: warning:
 variable 'vif' set but not used [-Wunused-but-set-variable]

drivers/net/wireless/rsi/rsi_91x_main.c:124:28: warning:
 variable 'info' set but not used [-Wunused-but-set-variable]

They're not used any more since 160ee2a11ce0 ("rsi: fill rx_params only once.")

Signed-off-by: YueHaibing <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rsi: fix spelling mistakes

Trivial fixes to spelling mistakes in various files in rsi folder.

Signed-off-by: Siva Rebbagondla <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rsi: fix a typo in a debug message

There is a \b (backspace) character in the message that wasn't intended.

Signed-off-by: Dan Carpenter <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

brcmfmac: remove set but not used variable 'old_state'

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c: In function 'brcmf_usb_state_change':
drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c:578:6: warning:
 variable 'old_state' set but not used [-Wunused-but-set-variable]

It's never used and can be removed.

Signed-off-by: YueHaibing <[email protected]>
Acked-by: Arend van Spriel <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

brcmfmac: rework bphy_err() to take struct brcmf_pub argument

This macro will be used in more places not just the cfg80211.c. It makes
sense to pass some common struct to it as "struct wiphy" is mostly
referenced in cfg80211 code only.

A very common one (used above the bus abstraction layer) is struct
brcmf_pub. Many functions already keep reference to it which will make
using bphy_err() simpler. It should also allow extending that macro's
logic if it's ever needed.

This improves code recently added in the commit 3ef005b82e2a ("brcmfmac:
add bphy_err() and use it in the cfg80211.c").

Signed-off-by: Rafał Miłecki <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

brcmfmac: support firmware reporting 160 MHz channels

So far 160 MHz channels were treated as 20 MHz ones which was breaking
support for 40/80 MHz due to the brcmf_construct_chaninfo() logic and
its assumptions.

Signed-off-by: Rafał Miłecki <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

brcmfmac: improve code handling bandwidth of firmware reported channels

1) Use switch to simplify/improve the code & avoid some duplication
2) Add warning for unsupported values

Signed-off-by: Rafał Miłecki <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

brcmfmac: print firmware reported ring status errors

Firmware is capable of reporting ring status. It's used e.g. to signal
some problem with a specific ring setup. This patch adds support for
printing ring & error number which may be useful for debugging setup
issues.

Signed-off-by: Rafał Miłecki <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

brcmfmac: use chipname in brcmf_fw_alloc_request() for error path as well

The local variable chipname is string representation of chip id and revision
which is printed in the good flow of brcmf_fw_alloc_request(). Also use it
for the error path, ie. for unknown/unsupported devices.

Reviewed-by: Hante Meuleman <[email protected]>
Reviewed-by: Pieter-Paul Giesberts <[email protected]>
Reviewed-by: Franky Lin <[email protected]>
Signed-off-by: Arend van Spriel <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

brcmfmac: check and dump trap info during sdio probe

When the firmware crashes during the probe sequence we provide little
information on what really failed. This patch checks the sdpcm shared
location and show the trap information if a firmware trap has happened.

Reviewed-by: Hante Meuleman <[email protected]>
Reviewed-by: Pieter-Paul Giesberts <[email protected]>
Reviewed-by: Franky Lin <[email protected]>
Signed-off-by: Arend van Spriel <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

brcmfmac: disable MBSS feature for bcm4330 device

The MBSS feature was already disabled for bcm43362 as it resulted in a
beacon with BRCM_TEST_SSID regardless user configuration in hostapd. Now
the same has been reported for bcm4330 so disable the feature for this
device as well.

Reported-by: Russell King <[email protected]>
Tested-by: Christopher Martin <[email protected]>
Reviewed-by: Hante Meuleman <[email protected]>
Reviewed-by: Pieter-Paul Giesberts <[email protected]>
Reviewed-by: Franky Lin <[email protected]>
Signed-off-by: Arend van Spriel <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

brcmfmac: create debugfs files for bus-specific layer

Since we moved the drivers debugfs directory under ieee80211 debugfs the
debugfs entries need to be added after wiphy_register() has been called.
For most part that has been done accordingly, but for the debugfs entries
added by SDIO it was not and failed silently. This patch fixes that by
adding a bus-layer callback for it.

Fixes: 856d5a011c86 ("brcmfmac: allocate struct brcmf_pub instance using wiphy_new()")
Reported-by: Russel King <[email protected]>
Reviewed-by: Franky Lin <[email protected]>
Signed-off-by: Arend van Spriel <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

brcmfmac: add subtype check for event handling in data path

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 <[email protected]>
Reviewed-by: Pieter-Paul Giesberts <[email protected]>
Reviewed-by: Franky Lin <[email protected]>
Signed-off-by: Arend van Spriel <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

brcmfmac: assure SSID length from firmware is limited

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 <[email protected]>
Reviewed-by: Pieter-Paul Giesberts <[email protected]>
Reviewed-by: Franky Lin <[email protected]>
Signed-off-by: Arend van Spriel <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8821ae: Remove CamelCase variables

If a macro is in CamelCase, it it converted to upper case. Variables
and routine names are converted to lower case.

The following checkpatch exceptions are also fixed:

WARNING: line over 80 characters
#316: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8821ae/fw.c:1751:
+       SET_8821AE_H2CCMD_AOAC_RSVDPAGE_LOC_GTK_EXT_MEM(u1rsvdpageloc2, GTKEXT_PG);

CHECK: spaces preferred around that '+' (ctx:VxV)
#357: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c:2591:
+                (eeaddr+1), hwinfo[eeaddr+1]);
                        ^

CHECK: spaces preferred around that '+' (ctx:VxV)
#357: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c:2591:
+                (eeaddr+1), hwinfo[eeaddr+1]);
                                          ^

CHECK: spaces preferred around that '+' (ctx:VxV)
#358: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c:2592:
+       if (0xFF == hwinfo[eeaddr+1])  /*YJ,add,120316*/
                                 ^

WARNING: Comparisons should place the constant on the right side of the test
#358: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c:2592:
+       if (0xFF == hwinfo[eeaddr+1])  /*YJ,add,120316*/

CHECK: spaces preferred around that '&' (ctx:VxV)
#501: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c:2666:
+                               pwrinfo24g->bw40_diff[rfpath][txcount] = (hwinfo[eeaddr]&0xf0) >> 4;
                                                                                        ^
Note that not all checkpatch exceptions are addressed. Those will be
handled in later patches.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8723{be,ae}: Remove CamelCase variables

If a macro is in CamelCase, it it converted to upper case. Variables
and routine names are converted to lower case.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8192se: Remove CamelCase variables

If a macro is in CamelCase, it it converted to upper case. Variables
and routine names are converted to lower case.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8188de: Remove CamelCase variables

If a macro is in CamelCase, it it converted to upper case. Variables
and routine names are converted to lower case.

The following checkpatch exceptions are also fixed:

CHECK: No space is necessary after a cast
#211: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c:109:
+                                             (u8 *) (&rfstate));

CHECK: No space is necessary after a cast
#241: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c:277:
+                               ptmp_byte = (u8 *) (&regtoset) + index;

Note that not all checkpatch exceptions are addressed. Those will be
handled in later patches.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8192{c,u}: Remove CamelCase variables

This patch affects modules rtl8192ce, rtl8192cu, and rtl8192com.

The following checkpatch exceptions are also fixed:

CHECK: No space is necessary after a cast
#237: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c:90:
+                                                     (u8 *) (&rfstate));

CHECK: No space is necessary after a cast
#744: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192ce/rf.c:329:
+                       pwr_val[i] = (u8) ((writeval & (0x7f <<

CHECK: No space is necessary after a cast
#784: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192ce/rf.c:365:
+                                              (u8) writeval);

CHECK: spaces preferred around that '/' (ctx:VxV)
#963: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192cu/hw.c:513:
+               txqpageunit = txqpagenum/outepnum;
                                        ^

CHECK: Unnecessary parentheses around 'outepnum > 1'
#975: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192cu/hw.c:521:
+               if ((outepnum > 1) && (txqremaininpage))

CHECK: Alignment should match open parenthesis
#1059: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192cu/hw.c:617:
+static void _rtl92cu_init_chipn_two_out_ep_priority(struct ieee80211_hw *hw,
                                                                bool wmm_enable,

ERROR: "foo * bar" should be "foo *bar"
#1940: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192cu/trx.c:591:
+void rtl92cu_fill_fake_txdesc(struct ieee80211_hw *hw, u8 *pdesc,

Note that not all checkpatch exceptions are addressed. Those will be
handled in later patches.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8188ee: Remove CamelCase variable

This driver contains one instance of a CamelCase variable, namely
IS_81xxC_VENDOR_UMC_A_CUT. This macro is never used, thus it is deleted.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: btcoexist: Remove CamelCase variable

File halbt_precomp.h contains the only CamelCase value, namely
GetDefaultAdapter, but that macro is never used, thus it is deleted.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: Remove CamelCase variables from base code

There are a number of CamelCase variables remaining in the source
files of modules rtl_pci, rtl_usb, and rtlwifi.

The following checkpatch excettions are also fixed:

for_kalle1/0001-rtlwifi-Remove-CamelCase-variables-from-base-code.patch
-----------------------------------------------------------------------
CHECK: Unnecessary parentheses around 'rtlefuse->efuse_map[EFUSE_INIT_MAP][base + i] !=
                             rtlefuse->efuse_map[EFUSE_MODIFY_MAP][base + i]'
#68: FILE: drivers/net/wireless/realtek/rtlwifi/efuse.c:377:
+                       if ((rtlefuse->efuse_map[EFUSE_INIT_MAP][base + i] !=
+                            rtlefuse->efuse_map[EFUSE_MODIFY_MAP][base + i]) ||
+                           (rtlefuse->efuse_map[EFUSE_INIT_MAP][base + i + 1] !=
+                            rtlefuse->efuse_map[EFUSE_MODIFY_MAP][base + i +
                                                                   1])) {

CHECK: Unnecessary parentheses around 'rtlefuse->efuse_map[EFUSE_INIT_MAP][base + i + 1] !=
                             rtlefuse->efuse_map[EFUSE_MODIFY_MAP][base + i +
                                                                   1]'
#68: FILE: drivers/net/wireless/realtek/rtlwifi/efuse.c:377:
+                       if ((rtlefuse->efuse_map[EFUSE_INIT_MAP][base + i] !=
+                            rtlefuse->efuse_map[EFUSE_MODIFY_MAP][base + i]) ||
+                           (rtlefuse->efuse_map[EFUSE_INIT_MAP][base + i + 1] !=
+                            rtlefuse->efuse_map[EFUSE_MODIFY_MAP][base + i +
                                                                   1])) {

WARNING: line over 80 characters
#70: FILE: drivers/net/wireless/realtek/rtlwifi/efuse.c:379:
+                           (rtlefuse->efuse_map[EFUSE_INIT_MAP][base + i + 1] !=

CHECK: No space is necessary after a cast
#186: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c:396:
+       status->rx_is40mhzpacket = (bool) GET_RX_DESC_BW(pdesc);

CHECK: No space is necessary after a cast
#208: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c:335:
+       stats->rx_is40mhzpacket = (bool) GET_RX_DESC_BW(pdesc);

CHECK: No space is necessary after a cast
#243: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192cu/trx.c:301:
+       stats->rx_is40mhzpacket = (bool) GET_RX_DESC_BW(pdesc);

CHECK: No space is necessary after a cast
#252: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192cu/trx.c:368:
+       stats.rx_is40mhzpacket = (bool) GET_RX_DESC_BW(rxdesc);

CHECK: No space is necessary after a cast
#265: FILE: drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c:475:
+       stats->rx_is40mhzpacket = (bool) GET_RX_DESC_BW(pdesc);

WARNING: Unnecessary space before function pointer arguments
#455: FILE: drivers/net/wireless/realtek/rtlwifi/wifi.h:1462:
+       void (*writen_sync) (struct rtl_priv *rtlpriv, u32 addr, void *buf,

WARNING: Unnecessary space before function pointer arguments
#483: FILE: drivers/net/wireless/realtek/rtlwifi/wifi.h:2257:
+       void (*fill_fake_txdesc) (struct ieee80211_hw *hw, u8 *pdesc,

Note that not all checkpatch exceptions are addressed. Those will be
handled in later patches.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8821ae: Replace old-style license information

The traditional license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8723_common: Replace old-style license information

The traditional license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8723be: Replace old-style license information

The traditional license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8723ae: Replace old-style license information

The traditional license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8192se: Replace old-style license information

The traditional license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8192ee: Replace old-style license information

The traditional license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8192de: Replace old-style license information

The traditional license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8192cu: Replace old-style license information

The traditional license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8192ce: Replace old-style license information

The traditional license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8192com: Replace old-style license information

The traditional license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: rtl8188ee: Replace old-style license information

The traditional license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: btcoex: Replace old-style license information

The old-style license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

rtlwifi: Replace old-style license information

The traditional license information is replaced by the SPDX form. There
are no code changes.

Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>

Merge tag 'mt76-for-kvalo-2019-02-18' of https://github.com/nbd168/wireless

mt76 patches for 5.1

* beacon support for USB devices (mesh+ad-hoc only)
* mt76x0 tx power fixes
* fixes for encryption, calibration and ED/CCA
* more code unification
* USB fixes
* fix for using the correct hweight8 function

6lowpan: use rbtree for IP frag queue

This patch aligns IP defragmenation logic in 6lowpan with that
of IPv4 and IPv6: see
commit d4289fcc9b16 ("net: IP6 defrag: use rbtrees for IPv6 defrag")

Modifying ip_defrag selftest seemed like an overkill, as I suspect
most kernel test setups do not have 6lowpan hwsim enabled. So I ran
the following code/script manually:

insmod ./mac802154_hwsim.ko

iwpan dev wpan0 set pan_id 0xbeef
ip link add link wpan0 name lowpan0 type lowpan
ip link set wpan0 up
ip link set lowpan0 up

iwpan dev wpan1 set pan_id 0xbeef
ip netns add foo
iwpan phy1 set netns name foo
ip netns exec foo ip link add link wpan1 name lowpan1 type lowpan
ip netns exec foo ip link set wpan1 up
ip netns exec foo ip link set lowpan1 up

ip -6 addr add "fb01::1/128" nodad dev lowpan0
ip -netns foo -6 addr add "fb02::1/128" nodad dev lowpan1

ip -6 route add "fb02::1/128" dev lowpan0
ip -netns foo -6 route add "fb01::1/128" dev lowpan1

# then in term1:
   ip netns exec foo bash
   ./udp_stream -6

# in term2:
    ./udp_stream -c -6 -H fb02::1

# pr_warn_once showed that the code changed by this patch
# was invoked.

Signed-off-by: Peter Oskolkov <[email protected]>
Acked-by: Alexander Aring <[email protected]>
Signed-off-by: Stefan Schmidt <[email protected]>

Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf

Pablo Neira Ayuso says:

====================
Netfilter/IPVS fixes for net

The following patchset contains Netfilter/IPVS fixes for net:

1) Follow up patch to fix a compilation warning in a recent IPVS fix:
   098e13f5b21d ("ipvs: fix dependency on nf_defrag_ipv6").

2) Bogus ENOENT error on flush after rule deletion in the same batch,
   reported by Phil Sutter.
====================

Signed-off-by: David S. Miller <[email protected]>

net: netcp: Fix ethss driver probe issue

Recent commit below has introduced a bug in netcp driver that causes
the ethss driver probe failure and thus break the networking function
on K2 SoCs such as K2HK, K2L, K2E etc. This patch fixes the issue to
restore networking on the above SoCs.

Fixes: 21c328dcecfc ("net: ethernet: Convert to using %pOFn instead of device_node.name")
Signed-off-by: Murali Karicheri <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

net: hns: Fixes the missing put_device in positive leg for roce reset

This patch fixes the missing device reference release-after-use in
the positive leg of the roce reset API of the HNS DSAF.

Fixes: c969c6e7ab8c ("net: hns: Fix object reference leaks in hns_dsaf_roce_reset()")
Reported-by: John Garry <[email protected]>
Signed-off-by: Salil Mehta <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

Merge tag 'wireless-drivers-for-davem-2019-02-18' of git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers

Kalle Valo says:

====================
wireless-drivers fixes for 5.0

Hopefully the last set of fixes for 5.0, only fix this time.

mt76

* fix regression with resume on mt76x0u USB devices
====================

Signed-off-by: David S. Miller <[email protected]>

net: stmmac: Fix a race in EEE enable callback

We are saving the status of EEE even before we try to enable it. This
leads to a race with XMIT function that tries to arm EEE timer before we
set it up.

Fix this by only saving the EEE parameters after all operations are
performed with success.

Signed-off-by: Jose Abreu <[email protected]>
Fixes: d765955d2ae0 ("stmmac: add the Energy Efficient Ethernet support")
Cc: Joao Pinto <[email protected]>
Cc: David S. Miller <[email protected]>
Cc: Giuseppe Cavallaro <[email protected]>
Cc: Alexandre Torgue <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

Merge branch 'qed-iWARP'

Michal Kalderon says:

====================
qed: iWARP - fix some syn related issues.

This series fixes two bugs related to iWARP syn processing flow.
====================

Signed-off-by: David S. Miller <[email protected]>

qed: Fix iWARP syn packet mac address validation.

The ll2 forwards all syn packets to the driver without validating the mac
address. Add validation check in the driver's iWARP listener flow and drop
the packet if it isn't intended for the device.

Signed-off-by: Ariel Elior <[email protected]>
Signed-off-by: Michal Kalderon <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

qed: Fix iWARP buffer size provided for syn packet processing.

The assumption that the maximum size of a syn packet is 128 bytes
is wrong. Tunneling headers were not accounted for.
Allocate buffers large enough for mtu.

Signed-off-by: Ariel Elior <[email protected]>
Signed-off-by: Michal Kalderon <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

exec: load_script: Do not exec truncated interpreter path

Commit 8099b047ecc4 ("exec: load_script: don't blindly truncate
shebang string") was trying to protect against a confused exec of a
truncated interpreter path. However, it was overeager and also refused
to truncate arguments as well, which broke userspace, and it was
reverted. This attempts the protection again, but allows arguments to
remain truncated. In an effort to improve readability, helper functions
and comments have been added.

Co-developed-by: Linus Torvalds <[email protected]>
Signed-off-by: Kees Cook <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: Samuel Dionne-Riel <[email protected]>
Cc: Richard Weinberger <[email protected]>
Cc: Graham Christensen <[email protected]>
Cc: Michal Hocko <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

bnx2x: Remove set but not used variable 'mfw_vn'

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c: In function 'bnx2x_get_hwinfo':
drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c:11940:10: warning:
 variable 'mfw_vn' set but not used [-Wunused-but-set-variable]

It's never used since introduction.

Signed-off-by: YueHaibing <[email protected]>
Acked-by: Sudarsana Reddy Kalluru <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

Merge branch 'net-phy-add-helpers-for-handling-C45-10GBT-AN-register-values'

Heiner Kallweit says:

====================
net: phy: add helpers for handling C45 10GBT AN register values

Similar to the existing helpers for the Clause 22 registers add helpers
to deal with converting Clause 45 advertisement registers to / from
link mode bitmaps.

Note that these helpers are defined in linux/mdio.h, not like the
Clause 22 helpers in linux/mii.h. Reason is that the Clause 45 register
constants are defined in uapi/linux/mdio.h. And uapi/linux/mdio.h
includes linux/mii.h before defining the C45 register constants.

v2:
- Remove few helpers which aren't used by this series. They will
  follow together with the users.
====================

Signed-off-by: David S. Miller <[email protected]>

net: phy: use mii_10gbt_stat_mod_linkmode_lpa_t in genphy_c45_read_lpa

Use mii_10gbt_stat_mod_linkmode_lpa_t() in genphy_c45_read_lpa() to
simplify the code.

Signed-off-by: Heiner Kallweit <[email protected]>
Reviewed-by: Florian Fainelli <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

net: phy: add helper mii_10gbt_stat_mod_linkmode_lpa_t

Similar to the existing helpers for the Clause 22 registers add helper
mii_10gbt_stat_mod_linkmode_lpa_t.

Note that this helper is defined in linux/mdio.h, not like the
Clause 22 helpers in linux/mii.h. Reason is that the Clause 45 register
constants are defined in uapi/linux/mdio.h. And uapi/linux/mdio.h
includes linux/mii.h before defining the C45 register constants.

v2:
- remove helpers that don't have users in this series

Signed-off-by: Heiner Kallweit <[email protected]>
Reviewed-by: Florian Fainelli <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

liquidio: using NULL instead of plain integer

Fix following warning:

drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c:1453:35: warning: Using plain integer as NULL pointer
drivers/net/ethernet/cavium/liquidio/lio_main.c:2910:23: warning: Using plain integer as NULL pointer

Signed-off-by: YueHaibing <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

r8169: remove unneeded mmiowb barriers

writex() has implicit barriers, that's what makes it different from
writex_relaxed(). Therefore these calls to mmiowb() can be removed.

This patch was recently reverted due to a dependency with another
problematic patch. But because it didn't contribute to the problem
it was rebased and can be resubmitted.

Signed-off-by: Heiner Kallweit <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

r8152: Add support for MAC address pass through on RTL8153-BD

RTL8153-BD is used in Dell DA300 type-C dongle.
It should be added to the whitelist of devices to activate MAC address
pass through.

Per confirming with Realtek all devices containing RTL8153-BD should
activate MAC pass through and there won't use pass through bit on efuse
like in RTL8153-AD.

Signed-off-by: David Chen <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

net: dsa: Implement flow_dissect callback for tag_dsa.

RPS not work for DSA devices since the 'skb_get_hash'
will always get the invalid hash for dsa tagged packets.

"[PATCH] tag_mtk: add flow_dissect callback to the ops struct"
introduced the flow_dissect callback to get the right hash for
MTK tagged packet. Tag_dsa and tag_edsa  also need to implement
the callback.

Signed-off-by: Rundong Ge <[email protected]>
Reviewed-by: Andrew Lunn <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

net: sched: using kfree_rcu() to simplify the code

The callback function of call_rcu() just calls a kfree(), so we
can use kfree_rcu() instead of call_rcu() + callback function.

Signed-off-by: Wei Yongjun <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

mdio_bus: Fix PTR_ERR() usage after initialization to constant

Fix coccinelle warning:

./drivers/net/phy/mdio_bus.c:51:5-12: ERROR: PTR_ERR applied after initialization to constant on line 44
./drivers/net/phy/mdio_bus.c:52:5-12: ERROR: PTR_ERR applied after initialization to constant on line 44

fix this by using IS_ERR before PTR_ERR

Fixes: bafbdd527d56 ("phylib: Add device reset GPIO support")
Signed-off-by: YueHaibing <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

mac80211: mesh: fix missing unlock on error in table_path_del()

spin_lock_bh() is used in table_path_del() but rcu_read_unlock()
is used for unlocking. Fix it by using spin_unlock_bh() instead
of rcu_read_unlock() in the error handling case.

Fixes: b4c3fbe63601 ("mac80211: Use linked list instead of rhashtable walk for mesh tables")
Acked-by: Herbert Xu <[email protected]>
Signed-off-by: Wei Yongjun <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

mlxsw: spectrum: Change IP2ME CPU policer rate and burst size values

The IP2ME packet trap is triggered by packets hitting local routes.
After evaluating current defaults used by the driver it was decided to
reduce the amount of traffic generated by this trap to 1Kpps and
increase the burst size. This is inline with similarly deployed systems.

Signed-off-by: Shalom Toledo <[email protected]>
Signed-off-by: Ido Schimmel <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

net: hamradio: remove unused hweight*() defines

This file does not use hweight*() at all, and the definition is
surrounded by #if 0 ... #endif.

Signed-off-by: Masahiro Yamada <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

net/mlx4_en: fix spelling mistake: "quiting" -> "quitting"

There is a spelling mistake in a en_err error message. Fix it.

Signed-off-by: Colin Ian King <[email protected]>
Reviewed-by: Tariq Toukan <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

net: crypto set sk to NULL when af_alg_release.

KASAN has found use-after-free in sockfs_setattr.
The existed commit 6d8c50dcb029 ("socket: close race condition between sock_close()
and sockfs_setattr()") is to fix this simillar issue, but it seems to ignore
that crypto module forgets to set the sk to NULL after af_alg_release.

KASAN report details as below:
BUG: KASAN: use-after-free in sockfs_setattr+0x120/0x150
Write of size 4 at addr ffff88837b956128 by task syz-executor0/4186

CPU: 2 PID: 4186 Comm: syz-executor0 Not tainted xxx + #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
1.10.2-1ubuntu1 04/01/2014
Call Trace:
 dump_stack+0xca/0x13e
 print_address_description+0x79/0x330
 ? vprintk_func+0x5e/0xf0
 kasan_report+0x18a/0x2e0
 ? sockfs_setattr+0x120/0x150
 sockfs_setattr+0x120/0x150
 ? sock_register+0x2d0/0x2d0
 notify_change+0x90c/0xd40
 ? chown_common+0x2ef/0x510
 chown_common+0x2ef/0x510
 ? chmod_common+0x3b0/0x3b0
 ? __lock_is_held+0xbc/0x160
 ? __sb_start_write+0x13d/0x2b0
 ? __mnt_want_write+0x19a/0x250
 do_fchownat+0x15c/0x190
 ? __ia32_sys_chmod+0x80/0x80
 ? trace_hardirqs_on_thunk+0x1a/0x1c
 __x64_sys_fchownat+0xbf/0x160
 ? lockdep_hardirqs_on+0x39a/0x5e0
 do_syscall_64+0xc8/0x580
 entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x462589
Code: f7 d8 64 89 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 48 89 f8 48 89
f7 48 89 d6 48 89
ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3
48 c7 c1 bc ff ff
ff f7 d8 64 89 01 48
RSP: 002b:00007fb4b2c83c58 EFLAGS: 00000246 ORIG_RAX: 0000000000000104
RAX: ffffffffffffffda RBX: 000000000072bfa0 RCX: 0000000000462589
RDX: 0000000000000000 RSI: 00000000200000c0 RDI: 0000000000000007
RBP: 0000000000000005 R08: 0000000000001000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007fb4b2c846bc
R13: 00000000004bc733 R14: 00000000006f5138 R15: 00000000ffffffff

Allocated by task 4185:
 kasan_kmalloc+0xa0/0xd0
 __kmalloc+0x14a/0x350
 sk_prot_alloc+0xf6/0x290
 sk_alloc+0x3d/0xc00
 af_alg_accept+0x9e/0x670
 hash_accept+0x4a3/0x650
 __sys_accept4+0x306/0x5c0
 __x64_sys_accept4+0x98/0x100
 do_syscall_64+0xc8/0x580
 entry_SYSCALL_64_after_hwframe+0x49/0xbe

Freed by task 4184:
 __kasan_slab_free+0x12e/0x180
 kfree+0xeb/0x2f0
 __sk_destruct+0x4e6/0x6a0
 sk_destruct+0x48/0x70
 __sk_free+0xa9/0x270
 sk_free+0x2a/0x30
 af_alg_release+0x5c/0x70
 __sock_release+0xd3/0x280
 sock_close+0x1a/0x20
 __fput+0x27f/0x7f0
 task_work_run+0x136/0x1b0
 exit_to_usermode_loop+0x1a7/0x1d0
 do_syscall_64+0x461/0x580
 entry_SYSCALL_64_after_hwframe+0x49/0xbe

Syzkaller reproducer:
r0 = perf_event_open(&(0x7f0000000000)={0x0, 0x70, 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, @perf_config_ext}, 0x0, 0x0,
0xffffffffffffffff, 0x0)
r1 = socket$alg(0x26, 0x5, 0x0)
getrusage(0x0, 0x0)
bind(r1, &(0x7f00000001c0)=@alg={0x26, 'hash\x00', 0x0, 0x0,
'sha256-ssse3\x00'}, 0x80)
r2 = accept(r1, 0x0, 0x0)
r3 = accept4$unix(r2, 0x0, 0x0, 0x0)
r4 = dup3(r3, r0, 0x0)
fchownat(r4, &(0x7f00000000c0)='\x00', 0x0, 0x0, 0x1000)

Fixes: 6d8c50dcb029 ("socket: close race condition between sock_close() and sockfs_setattr()")
Signed-off-by: Mao Wenan <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

mt76: mt76x2: simplify per-chain signal strength handling

There is no need to use a for loop here, supported chips can only support
up to 2 chains.

Signed-off-by: Felix Fietkau <[email protected]>

Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf-next

Pablo Neira Ayuso says:

====================
Netfilter/IPVS updates for net-next

The following patchset contains Netfilter/IPVS updates for you net-next
tree:

1) Missing NFTA_RULE_POSITION_ID netlink attribute validation,
   from Phil Sutter.

2) Restrict matching on tunnel metadata to rx/tx path, from wenxu.

3) Avoid indirect calls for IPV6=y, from Florian Westphal.

4) Add two indirections to prepare merger of IPV4 and IPV6 nat
   modules, from Florian Westphal.

5) Broken indentation in ctnetlink, from Colin Ian King.

6) Patches to use struct_size() from netfilter and IPVS,
   from Gustavo A. R. Silva.

7) Display kernel splat only once in case of racing to confirm
   conntrack from bridge plus nfqueue setups, from Chieh-Min Wang.

8) Skip checksum validation for layer 4 protocols that don't need it,
   patch from Alin Nastac.

9) Sparse warning due to symbol that should be static in CLUSTERIP,
   from Wei Yongjun.

10) Add new toggle to disable SDP payload translation when media
    endpoint is reachable though the same interface as the signalling
    peer, from Alin Nastac.
====================

Signed-off-by: David S. Miller <[email protected]>

mt76: fix resetting software IV flag on key delete

It needs to be unset instead of set

Fixes: 23405236460b9 ("mt76: fix transmission of encrypted management frames")
Signed-off-by: Felix Fietkau <[email protected]>

mt76: fix corrupted software generated tx CCMP PN

Since ccmp_pn is u8 *, the second half needs to start at array index 4
instead of 0. Fixes a connection stall after a certain amount of traffic

Fixes: 23405236460b9 ("mt76: fix transmission of encrypted management frames")
Cc: [email protected]
Signed-off-by: Felix Fietkau <[email protected]>

mt76: fix tx status timeout processing

Remove bogus check for MT_PACKET_ID_NO_ACK in mt76_tx_status_skb_get, which
is already handled in callers and turns timeout calls into no-ops
Do not clean up pending status items on reordering of tx status information
if the timeout is not reached yet

Signed-off-by: Felix Fietkau <[email protected]>

mt76: Use the correct hweight8() function

mt76_init_stream_cap() and mt76_get_txpower() call __sw_hweight8()
directly, but that's only defined if CONFIG_GENERIC_HWEIGHT is
enabled.  The function that works on all architectures is hweight8().

Fixes: 551e1ef4d291 ("mt76: add mt76_init_stream_cap routine")
Fixes: 9313faacbb4e ("mt76: move mt76x02_get_txpower to mt76 core")
Signed-off-by: Ben Hutchings <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76x02u: remove bogus check and comment padding

In mt76x02u_skb_dma_info() pad is always non-zero. Patch removes
bogus check and add comments to the function.

Signed-off-by: Stanislaw Gruszka <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76: usb: use dev_err_ratelimited instead of dev_err in mt76u_complete_rx

During device removal the driver can report multiple error messages.
Use dev_err_ratelimited instead of dev_err to display urb errors

Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76: usb: introduce disable_usb_sg parameter

Add disable_usb_sg module parameter to disable scatter-gather on demand

Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76: usb: use a linear buffer for tx/rx datapath if sg is not supported

Use linear fragment and not a single usb scatter-gather buffer in mt76u
{tx,rx} datapath if the usb controller has sg data length constraints

Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76: usb: do not use sg buffers for mcu messages

Do not use scatter-gather buffers for mcu commands.
Introduce mt76u_buf_alloc and mt76u_buf_alloc_sg routines.

Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76: usb: move mt76u_check_sg in usb.c

Move mt76u_check_sg routine in usb.c and introduce sg_en variable
in mt76_usb in order to check if scatter-gather is supported by
mt76u layer

Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76: usb: do not run mt76u_queues_deinit twice

Do not call mt76u_queues_deinit routine in mt76u_alloc_queues error path
since it will be run in mt76x0u_register_device or
mt76x2u_register_device error path. Current implementation triggers the
following kernel warning:

[   67.005516] WARNING: CPU: 2 PID: 761 at lib/refcount.c:187 refcount_sub_and_test_checked+0xa4/0xb8
[   67.019513] refcount_t: underflow; use-after-free.
[   67.099872] Hardware name: BCM2835
[   67.106268] Backtrace:
[   67.111584] [<8010c91c>] (dump_backtrace) from [<8010cc00>] (show_stack+0x20/0x24)
[   67.124974]  r6:60000013 r5:ffffffff r4:00000000 r3:a50bade6
[   67.132226] [<8010cbe0>] (show_stack) from [<807ca5f4>] (dump_stack+0xc8/0x114)
[   67.141225] [<807ca52c>] (dump_stack) from [<8011e65c>] (__warn+0xf4/0x120)
[   67.149849]  r9:000000bb r8:804d0138 r7:00000009 r6:8099dc84 r5:00000000 r4:b66c7b58
[   67.160767] [<8011e568>] (__warn) from [<8011e6d0>] (warn_slowpath_fmt+0x48/0x50)
[   67.171436]  r9:7f65e128 r8:80d1419c r7:80c0bac4 r6:b97b3044 r5:b7368e00 r4:00000000
[   67.182433] [<8011e68c>] (warn_slowpath_fmt) from [<804d0138>] (refcount_sub_and_test_checked+0xa4/0xb8)
[   67.195221]  r3:80c91c25 r2:8099dc94
[   67.200370]  r4:00000000
[   67.204397] [<804d0094>] (refcount_sub_and_test_checked) from [<804d0164>] (refcount_dec_and_test_checked+0x18/0x1c)
[   67.218046]  r4:b7368e00 r3:00000001
[   67.223125] [<804d014c>] (refcount_dec_and_test_checked) from [<805db49c>] (usb_free_urb+0x20/0x4c)
[   67.235358] [<805db47c>] (usb_free_urb) from [<7f639804>] (mt76u_buf_free+0x98/0xac [mt76_usb])
[   67.247302]  r4:00000001 r3:00000001
[   67.252468] [<7f63976c>] (mt76u_buf_free [mt76_usb]) from [<7f639ef8>] (mt76u_queues_deinit+0x44/0x100 [mt76_usb])
[   67.266102]  r8:b8fe8600 r7:b5dac480 r6:b5dace20 r5:00000001 r4:00000000 r3:00000080
[   67.277132] [<7f639eb4>] (mt76u_queues_deinit [mt76_usb]) from [<7f65c040>] (mt76x0u_cleanup+0x40/0x4c [mt76x0u])
[   67.290737]  r7:b5dac480 r6:b8fe8600 r5:ffffffea r4:b5dace20
[   67.298069] [<7f65c000>] (mt76x0u_cleanup [mt76x0u]) from [<7f65c564>] (mt76x0u_probe+0x1f0/0x354 [mt76x0u])
[   67.311174]  r4:b5dace20 r3:00000000
[   67.316312] [<7f65c374>] (mt76x0u_probe [mt76x0u]) from [<805e0b6c>] (usb_probe_interface+0x104/0x240)
[   67.328915]  r7:00000000 r6:7f65e034 r5:b6634800 r4:b8fe8620
[   67.336276] [<805e0a68>] (usb_probe_interface) from [<8056a8bc>] (really_probe+0x224/0x2f8)
[   67.347965]  r10:b65f0a00 r9:00000019 r8:7f65e034 r7:80d3e124 r6:00000000 r5:80d3e120
[   67.359175]  r4:b8fe8620 r3:805e0a68
[   67.364384] [<8056a698>] (really_probe) from [<8056ab60>] (driver_probe_device+0x6c/0x180)
[   67.375974]  r10:b65f0a00 r9:7f65e2c0 r8:b8fe8620 r7:00000000 r6:7f65e034 r5:7f65e034
[   67.387170]  r4:b8fe8620 r3:00000000
[   67.392378] [<8056aaf4>] (driver_probe_device) from [<8056ad54>] (__driver_attach+0xe0/0xe4)
[   67.404097]  r9:7f65e2c0 r8:7f65d22c r7:00000000 r6:b8fe8654 r5:7f65e034 r4:b8fe8620
[   67.415122] [<8056ac74>] (__driver_attach) from [<8056880c>] (bus_for_each_dev+0x68/0xa0)
[   67.426628]  r6:8056ac74 r5:7f65e034 r4:00000000 r3:00000027
[   67.434017] [<805687a4>] (bus_for_each_dev) from [<8056a1cc>] (driver_attach+0x28/0x30)
[   67.445394]  r6:80c6ddc8 r5:b7368f80 r4:7f65e034
[   67.451703] [<8056a1a4>] (driver_attach) from [<80569c24>] (bus_add_driver+0x194/0x21c)
[   67.463081] [<80569a90>] (bus_add_driver) from [<8056b504>] (driver_register+0x8c/0x124)
[   67.474560]  r7:80c6ddc8 r6:7f65e034 r5:00000000 r4:7f65e034
[   67.481964] [<8056b478>] (driver_register) from [<805df510>] (usb_register_driver+0x74/0x140)
[   67.493901]  r5:00000000 r4:7f65e000
[   67.499131] [<805df49c>] (usb_register_driver) from [<7f661024>] (mt76x0_driver_init+0x24/0x1000 [mt76x0u])
[   67.512258]  r9:00000001 r8:7f65e308 r7:00000000 r6:80c08d48 r5:7f661000 r4:7f65e2c0
[   67.523404] [<7f661000>] (mt76x0_driver_init [mt76x0u]) from [<80102f6c>] (do_one_initcall+0x4c/0x210)
[   67.536142] [<80102f20>] (do_one_initcall) from [<801ae63c>] (do_init_module+0x6c/0x21c)
[   67.547639]  r8:7f65e308 r7:80c08d48 r6:b65f0ac0 r5:7f65e2c0 r4:7f65e2c0
[   67.556129] [<801ae5d0>] (do_init_module) from [<801ad68c>] (load_module+0x1d10/0x2304)

Fixes: b40b15e1521f ("mt76: add usb support to mt76 layer")
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76: usb: fix possible memory leak in mt76u_buf_free

Move q->ndesc initialization before the for loop in mt76u_alloc_rx
since otherwise allocated urbs will not be freed in mt76u_buf_free
Double-check scatterlist pointer in mt76u_buf_free

Fixes: b40b15e1521f ("mt76: add usb support to mt76 layer")
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76: usb: fix possible NULL pointer dereference in mt76u_mcu_deinit

Fix possible NULL pointer dereference in mt76u_mcu_deinit routine that
can occur if initialization path fails before calling mt76u_mcu_init_rx

Fixes: ee676cd5017c ("mt76: add driver code for MT76x2u based devices")
Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76x02u: use usb_bulk_msg to upload firmware

We don't need to send firmware data asynchronously, much simpler is just
use synchronous usb_bulk_msg().

Tested-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Stanislaw Gruszka <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76x0: init: introduce mt76x0_init_txpower routine

Add mt76x0_init_txpower in order to initialize max_power per channel
at device bootstrap. Modify mt76x0_get_tx_power_per_rate and
mt76x0_get_power_info signature in order to compute tx power for
non-operating channels

Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76x0: phy: report target_power in debugfs

Initialize target_power variable in mt76x0_phy_set_txpower in order to
report target_power in debugfs

Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76: change the return type of mt76_dma_attach()

There is no need to retun 0 in mt76_dma_attach(), so switch it to void.

Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76x2: remove mt76x2_alloc_device routine

Remove mt76x2_alloc_device since it just runs mt76_alloc_device.
Move mt76_alloc_device call in mt76x2_probe

Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>

mt76x0: remove mt76x0u_alloc_device routine

Remove mt76x0u_alloc_device since it just runs mt76_alloc_device.
Move mt76_alloc_device call in mt76x0u_probe and in mt76x0e_probe

Signed-off-by: Lorenzo Bianconi <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>