Synaptics add new TP firmware ID: 0x2 and 0x3, for now both lower 2 bits
are indicated as TP. Change the constant to bitwise values.
This makes trackpoint to be recognized on Lenovo Carbon X1 Gen5 instead
of it being identified as "PS/2 Generic Mouse".
Signed-off-by: Aaron Ma <aaron.ma@canonical.com>
Cc: stable@vger.kernel.org
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Add ELAN0602 to the list of known ACPI IDs to enable support for ELAN
touchpads found in Lenovo Yoga310.
Signed-off-by: KT Liao <kt.liao@emc.com.tw>
Cc: stable@vger.kernel.org
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Add some additional statistics for tracking VXLAN packets and checksum
errors.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The hardware has the capability to perform checksum offload support
(both Tx and Rx) and TSO support for VXLAN packets. Add the support
required to enable this.
The hardware can only support a single VXLAN port for offload. If more
than one VXLAN port is added then the offload capabilities have to be
disabled and can no longer be advertised.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Convert from using the old u32 supported, advertising, etc. link settings
to the new link mode settings that support bit positions / settings
greater than 32 bits.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add per queue Tx and Rx packet and byte counts.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use the dynamic debug support to output information about the hardware
features reported by the device.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently whenever the driver needs to enable or disable interrupts for
a DMA channel it reads the interrupt enable register (IER), updates the
value and then writes the new value back to the IER. Since the hardware
does not change the IER, software can track this value and elimiate the
need to read it each time.
Add the IER value to the channel related data structure and use that as
the base for enabling and disabling interrupts, thus removing the need
for the MMIO read.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add some additional dynamic debug message to the driver. The new messages
will provide additional information about the PCS window calculation.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Many of the names used by the driver are based upon the name of the device
found during device probe. Move the formatting of the names into the
device open function so that any renaming that occurs before the device is
brought up will be accounted for. This also means moving the creation of
some named workqueues into the device open path.
Add support to register for net events so that if a device is renamed
the corresponding debugfs directory can be renamed.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When transmitting a TSO packet, the driver only increments the TSO packet
statistic by one rather than the number of total packets that were sent.
Update the driver to record the total number of packets that resulted from
TSO transmit.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Sometimes when the driver is being unloaded while the devices are still
up the driver can issue errors. This is based on timing and the double
invocation of some routines. The phy_exit() call needs to be run after
the network device has been closed and unregistered from the system.
Also, the phy_exit() does not need to invoke phy_stop() since that will
be called as part of the device closing, so remove that call.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When running in SGMII mode at speeds below 1000Mbps, the auto-negotition
control register must set the MII control width for the MAC interface
to be 8-bits wide. By default the width is 4-bits.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently the MDIO mode is set to none for the 10000Base-T, which is
incorrect. The MDIO mode should for this configuration should be
clause 45.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, if vport is zero then then an uninialized return status
in err is returned. Since the only return status at the end of the
function esw_add_uc_addr is zero for the current set of return paths
we may as well just return 0 rather than err to fix this issue.
Detected by CoverityScan, CID#1452698 ("Uninitialized scalar variable")
Fixes: eeb66cdb68 ("net/mlx5: Separate between E-Switch and MPFS")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Unsupported events is only useful for developers and does not meaningful
for users. Using dev_dbg makes more sense and reduces noise in kernel
messages.
Signed-off-by: Alex Hung <alex.hung@canonical.com>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>
Unsupported events is only useful for developers and does not meaningful
for users. Using dev_dbg makes more sense and reduces noise in kernel
messages.
Signed-off-by: Alex Hung <alex.hung@canonical.com>
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>
Use errno -ENOSPC ("No space left on device") when the XDP xmit
have no space left on the TX ring buffer, instead of -ENOMEM.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
enable_4k_uar module parameter was added in patch cited below to
address the backward compatibility issue in SRIOV when the VM has
system's PAGE_SIZE uar implementation and the Hypervisor has 4k uar
implementation.
The above compatibility issue does not exist in the non SRIOV case.
In this patch, we always enable 4k uar implementation if SRIOV
is not enabled on mlx4's supported cards.
Fixes: 76e39ccf9c ("net/mlx4_core: Fix backward compatibility on VFs")
Signed-off-by: Huy Nguyen <huyn@mellanox.com>
Reviewed-by: Daniel Jurgens <danielj@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If the pci_find_pcie_root_port() function is called on a root port
itself, return the root port rather than NULL.
This effectively reverts commit 0e40523287 ("PCI: fix oops when
try to find Root Port for a PCI device") which added an extra check
that would now be redundant.
Fixes: a99b646afa ("PCI: Disable PCIe Relaxed Ordering if unsupported")
Fixes: c56d4450eb ("PCI: Turn off Request Attributes to avoid Chelsio T5 Completion erratum")
Signed-off-by: Thierry Reding <treding@nvidia.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Tested-by: Shawn Lin <shawn.lin@rock-chips.com>
Tested-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Rick Farrington <ricardo.farrington@cavium.com>
Signed-off-by: Felix Manlunas <felix.manlunas@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently macvlan devices do not set their hw_enc_features making
encapsulated Tx packets resort to SW fallbacks. Add encapsulation GSO
offloads to ->features as is done for the other GSOs and set
->hw_enc_features.
Signed-off-by: Dimitris Michailidis <dmichail@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use hwmon_device_register_with_groups instead of deprecated
hwmon_device_register and fix a dmesg warning.
This patch however changes the userspace API.
hwmon_device_register_with_groups takes `hwmon' name as an argument and
creates a name file in the `hwmon' device, not in the `platform_device'.
This allows us to remove custom `name' device attribute, but in order to
make lm-sensors happy we also have to move fans and thermal attributes
to the `hwmon' device.
Even though this patch changes userspace API, it's still compatible with
the lm-sensors. Starting with lm-sensors 3.0 (circa 2007), it looks at
both hwmon and the backing device for the name and other attributes.
before:
$ cat /sys/devices/platform/thinkpad_hwmon/{name,fan1_input}
thinkpad
2007
$ cat /sys/devices/platform/thinkpad_hwmon/hwmon/hwmon1/{name,fan1_input}
cat: /sys/devices/platform/thinkpad_hwmon/hwmon/hwmon1/name: No such file or directory
cat: /sys/devices/platform/thinkpad_hwmon/hwmon/hwmon1/fan1_input: No such file or directory
$ cat /sys/class/hwmon/hwmon1/{name,fan1_input}
cat: /sys/class/hwmon/hwmon1/name: No such file or directory
cat: /sys/class/hwmon/hwmon1/fan1_input: No such file or directory
$ sensors
thinkpad-isa-0000
Adapter: ISA adapter
fan1: 3533 RPM
after:
$ cat /sys/devices/platform/thinkpad_hwmon/{name,fan1_input}
cat: /sys/devices/platform/thinkpad_hwmon/name: No such file or directory
cat: /sys/devices/platform/thinkpad_hwmon/fan1_input: No such file or directory
$ cat /sys/devices/platform/thinkpad_hwmon/hwmon/hwmon1/{name,fan1_input}
thinkpad
3478
$ cat /sys/class/hwmon/hwmon1/{name,fan1_input}
thinkpad
3478
$ sensors
thinkpad-isa-0000
Adapter: ISA adapter
fan1: 3489 RPM
$ sensors -v
sensors version 3.4.0 with libsensors version 3.4.0
Signed-off-by: Stanislav Fomichev <kernel@fomichev.me>
[dvhart: cleaned up commit log, bumped version to 4.14 in the doc change]
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>
While in SNK_READY state, the explicit_contract seems to be
set to true irrespective of whether an explicit contract
was established for the current connection. TCPM also seems
to report the pwr_opmode as TYPEC_PWR_MODE_PD always once
the port gets into SNK_READY state. This isn't completely
true as port gets into the SNK_READY state for non-pd
type-c ports as well.
This patch sets the explicit_contract flag only when
the PS_READY message is received and the vbus has been
detected by the port controller.
Signed-off-by: Badhri Jagan Sridharan <Badhri@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Alignment should match with open parenthesis.
This fixes the coding style issue.
Signed-off-by: Shreeya Patel <shreeya.patel23498@gmail.com>
Reviewed-by: Bryan O'Donoghue <pure.logic@nexus-software.ie>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When an external USB synth is unplugged while the module is loaded, we
get a null pointer deref. This is because the tty disappears while
speakup tries to use to to communicate with the synth. This patch fixes
it by checking tty for null before using it. Since tty can become null
between the check and its usage, a mutex is introduced. tty usage is
now surrounded by the mutex, as is the code in speakup_ldisc_close which
sets the tty to null. The mutex also serialises calls to tty from
speakup code.
In case of tty being null, this sets synth->alive to zero and restarts
ttys in case they were stopped by speakup.
Signed-off-by: Okash Khawaja <okash.khawaja@gmail.com>
Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Testing has shown that lp* devices don't work correctly with speakup
just yet. That will require some additional work. Until then, this patch
removes code related to that.
Signed-off-by: Okash Khawaja <okash.khawaja@gmail.com>
Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
platform_get_irq() returns an error code, but the dim2_hdm driver
ignores it and always returns -ENODEV. This is not correct and,
prevents -EPROBE_DEFER from being propagated properly.
Also, notice that platform_get_irq() no longer returns 0 on error:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e330b9a6bb35dc7097a4f02cb1ae7b6f96df92af
Print and propagate the return value of platform_get_irq on failure.
This issue was detected with the help of Coccinelle.
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The 'linkstatus' field from struct 'hfa384x_link_status' represents a
16-bit little-endian integer, so declare it as such to avoid the
following sparse warning when accessing it through the 'le16_to_cpu()'
function:
prism2sta.c:1450:31: warning: cast to restricted __le16
Signed-off-by: Ricardo Silva <rjpdasilva@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The same structure is used in both side, so we dont need to cast.
This will fix the following sparse warnings:
drivers/staging/wlan-ng/prism2sta.c:1139:13: warning: cast to restricted __le16
drivers/staging/wlan-ng/prism2sta.c:1150:24: warning: cast to restricted __le16
drivers/staging/wlan-ng/prism2sta.c:1157:37: warning: cast to restricted __le16
drivers/staging/wlan-ng/prism2sta.c:1158:37: warning: cast to restricted __le16
drivers/staging/wlan-ng/prism2sta.c:1159:40: warning: cast to restricted __le16
drivers/staging/wlan-ng/prism2sta.c:1450:31: warning: cast to restricted __le16
Signed-off-by: AbdAllah MEZITI <abdallah.meziti.pro@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fixed a potential memory leak inside skb_ether_to_p80211()
When the wep_encrypt() fails the code return 2 directly
Which causes the p80211_wep->data dangling
Add a kfree statement to reclaim that memory allocated
Signed-off-by: Lynn Lei <lynnl.yet@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ethernet-rx.c:
fix WARNING: line over 80 characters
The code was restructured a bit, a helper function
was added to cvm_oct_poll.
Signed-off-by: John Smith <catalinnow@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix checkpatch issues:
ERROR: that open brace { should be on the previous line.
WARNING: braces {} are not necessary for single statement blocks.
Signed-off-by: Teo Dacquet <teo.dacquet@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
According to the coding-style documentation, spaces are required
around most binary and ternary operators. Spaces were added to
fix mulitple coding-style errors.
Signed-off-by: Christoffer Kjoelbaek <christoffer@kjoelbaek.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
These vb2_ops structures are only stored in the ops field of a
vb2_queue structure, which is declared as const. Thus the vb2_ops
structures themselves can be const.
Done with the help of Coccinelle.
// <smpl>
@r disable optional_qualifier@
identifier i;
position p;
@@
static struct vb2_ops i@p = { ... };
@ok@
identifier r.i;
struct vb2_queue e;
position p;
@@
e.ops = &i@p;
@bad@
position p != {r.p,ok.p};
identifier r.i;
struct vb2_ops e;
@@
e@i@p
@depends on !bad disable optional_qualifier@
identifier r.i;
@@
static
+const
struct vb2_ops i = { ... };
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
platform_device_id are not supposed to change at runtime. All functions
working with platform_device_id provided by <linux/platform_device.h>
work with const platform_device_id. So mark the non-const structs as
const.
Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Make these const as they are only used during a copy operation.
Done using Coccinelle.
Signed-off-by: Bhumika Goyal <bhumirks@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.
Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.
Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
usb_device_id are not supposed to change at runtime. All functions
working with usb_device_id provided by <linux/usb.h> work with
const usb_device_id. So mark the non-const structs as const.
Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Acked-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
At present, TCPM does not take into account the actual resistor
value presented in the CC line and therefore reports TYPEC_PWR_MODE_USB
irrespective of the power_op_mode it is in.
This patch makes TCPM consider the actual value of Rp.
Signed-off-by: Badhri Jagan Sridharan <Badhri@google.com>
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ 130.893355] state change SNK_DEBOUNCED -> SRC_TRY
[ 130.893363] cc:=3
[ 130.893490] pending state change SRC_TRY -> SNK_TRYWAIT @ 100 ms
[ 130.895602] CC1: 3 -> 0, CC2: 0 -> 0 [state SRC_TRY, polarity 0, disconnected]
[ 130.895613] state change SRC_TRY -> SRC_TRY_DEBOUNCE
[ 130.895621] pending state change SRC_TRY_DEBOUNCE -> SRC_ATTACHED @ 20 ms
[ 130.916843] state change SRC_TRY_DEBOUNCE -> SRC_ATTACHED [delayed 20 ms]
Although the CC state was changing to TYPEC_CC_OPEN, the port entered
SRC_TRY_DEBOUNCE from SRC_TRY. The port must enter SRC_TRY_DEBOUNCE only
if the CC state is TYPEC_CC_RD.
Signed-off-by: Badhri Jagan Sridharan <Badhri@google.com>
[groeck: Wording]
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If the partner is not PD capable, we can not use a power role set request
to swap roles. Use the data role set request instead.
Also, if a partner is not PD capable, it does not really make sense to send
a PD message to trigger a role swap. On top of that, we should really wait
for the attempted role change to complete. Otherwise, it may well be that
user space requests another role change immediately afterwards which will
fail because the port is not yet in ready state.
Trigger the role swap from data role change requests and introduce new
state PORT_RESET and use it to solve the problem. This new state is
mostly identical to ERROR_RECOVERY, only it does not cause a pending
role change to fail. Use this new state also when initializing the driver.
Rename ERROR_RECOVERY_WAIT_OFF to PORT_RESET_WAIT_OFF to better reflect
its new meaning.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Type-C protocol manager state machine could fail, which might result
in role swap requests from user space to hang forever. Add a generous
timeout when waiting for role swaps to complete to avoid this situation.
Originally-from: Badhri Jagan Sridharan <badhri@google.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
tcpm_default_state wasn't considering the port type when determining the
default role. This change makes tcpm_default_state to consider port type
as well.
tcpm_default_state would return the following based on the port type:
TYPEC_PORT_UFP - SNK_UNATTACHED
TYPEC_PORT_DFP - SRC_UNATTACHED
TYPEC_PORT_DRP - based on the preferred_role setting
Signed-off-by: Badhri Jagan Sridharan <Badhri@google.com>
[groeck: Reworded description; minor formatting changes]
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
While exiting ERROR_RECOVERY, choose default state based on the port
type instead of current power role.
Quoting from specification:
4.5.2.2.2 ErrorRecovery State
This state appears in Figure 4-12, Figure 4-13, Figure 4-14, Figure 4-15,
Figure 4-16 and Figure 4-17.
The ErrorRecovery state is where the port removes the terminations from
the CC1 and CC2 pins for tErrorRecovery followed by transitioning to the
appropriate Unattached.SNK or Unattached.SRC state based on port type.
This is the equivalent of forcing a detach event and looking for a new
attach.
Signed-off-by: Badhri Jagan Sridharan <Badhri@google.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Role swap requests fail unless the current role is either SRC_READY or
SNK_READY. This works fine for VCONN and data role swaps, where we
immediately enter READY state after reporting a successful role swap
to user space. However, on power role changes, the role swap is currently
reported as successful while power negotiation is still in process.
User space does not know this, and may request another role swap
immediately after a power role swap is reported to be complete.
This second role swap will fail with -EAGAIN.
To fix the problem, report role swap completion after power negotiation
is complete and the state machine enters SRC_READY or SNK_READY state.
This is better anyway since it captures errors due to failed power
negotiations. It also simplifies the code since the number of calls
needed to report successful role swaps is reduced.
Reported-by: Howard Yen <howard_yen@htc.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>