Commit graph

1,058,063 commits

Author SHA1 Message Date
Kishon Vijay Abraham I
f00bfc6489 dt-bindings: PCI: pci-ep: Add binding to specify virtual function
Add binding to specify virtual function (associated with each physical
function) in endpoint mode.

Link: https://lore.kernel.org/r/20210819123343.1951-2-kishon@ti.com
Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Reviewed-by: Rob Herring <robh@kernel.org>
2021-08-19 14:13:28 +01:00
Steven Rostedt (VMware)
8565a45d08 tracing/probes: Have process_fetch_insn() take a void * instead of pt_regs
In preparation to allow event probes to use the process_fetch_insn()
callback in trace_probe_tmpl.h, change the data passed to it from a
pointer to pt_regs, as the event probe will not be using regs, and make it
a void pointer instead.

Update the process_fetch_insn() callers for kprobe and uprobe events to
have the regs defined in the function and just typecast the void pointer
parameter.

Link: https://lkml.kernel.org/r/20210819041842.291622924@goodmis.org

Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2021-08-19 09:09:03 -04:00
Steven Rostedt (VMware)
007517a019 tracing/probe: Change traceprobe_set_print_fmt() to take a type
Instead of a boolean "is_return" have traceprobe_set_print_fmt() take a
type (currently just PROBE_PRINT_NORMAL and PROBE_PRINT_RETURN). This will
simplify adding different types. For example, the development of the
event_probe, will need its own type as it prints an event, and not an IP.

Link: https://lkml.kernel.org/r/20210819041842.104626301@goodmis.org

Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2021-08-19 09:08:29 -04:00
Dario Binacchi
387da6bc7a can: c_can: cache frames to operate as a true FIFO
As reported by a comment in the c_can_start_xmit() this was not a FIFO.
C/D_CAN controller sends out the buffers prioritized so that the lowest
buffer number wins.

What did c_can_start_xmit() do if head was less tail in the tx ring ? It
waited until all the frames queued in the FIFO was actually transmitted
by the controller before accepting a new CAN frame to transmit, even if
the FIFO was not full, to ensure that the messages were transmitted in
the order in which they were loaded.

By storing the frames in the FIFO without requiring its transmission, we
will be able to use the full size of the FIFO even in cases such as the
one described above. The transmission interrupt will trigger their
transmission only when all the messages previously loaded but stored in
less priority positions of the buffers have been transmitted.

Link: https://lore.kernel.org/r/20210807130800.5246-5-dariobin@libero.it
Suggested-by: Gianluca Falavigna <gianluca.falavigna@inwind.it>
Signed-off-by: Dario Binacchi <dariobin@libero.it>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:06 +02:00
Dario Binacchi
28e86e9ab5 can: c_can: support tx ring algorithm
The algorithm is already used successfully by other CAN drivers
(e.g. mcp251xfd). Its implementation was kindly suggested to me by
Marc Kleine-Budde following a patch I had previously submitted. You can
find every detail at https://lore.kernel.org/patchwork/patch/1422929/.

The idea is that after this patch, it will be easier to patch the driver
to use the message object memory as a true FIFO.

Link: https://lore.kernel.org/r/20210807130800.5246-4-dariobin@libero.it
Suggested-by: Marc Kleine-Budde <mkl@pengutronix.de>
Signed-off-by: Dario Binacchi <dariobin@libero.it>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:05 +02:00
Dario Binacchi
a54cdbba9d can: c_can: exit c_can_do_tx() early if no frames have been sent
The c_can_poll() handles RX/TX events unconditionally. It may therefore
happen that c_can_do_tx() is called unnecessarily because the interrupt
was triggered by the reception of a frame. In these cases, we avoid to
execute unnecessary statements and exit immediately.

Link: https://lore.kernel.org/r/20210807130800.5246-3-dariobin@libero.it
Signed-off-by: Dario Binacchi <dariobin@libero.it>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:05 +02:00
Dario Binacchi
5064e40596 can: c_can: remove struct c_can_priv::priv field
It references the clock but it is never used. So let's remove it.

Link: https://lore.kernel.org/r/20210807130800.5246-2-dariobin@libero.it
Signed-off-by: Dario Binacchi <dariobin@libero.it>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:05 +02:00
Marc Kleine-Budde
05cb2ba4b2 can: c_can: rename IF_RX -> IF_NAPI
The C_CAN/D_CAN cores implement 2 interfaces to manage the message
objects. To avoid concurrency and the need for locking one interface
is used in the TX path (IF_TX). While the other one, named IF_RX is
used from NAPI context only. As this interface is not only used to
manage RX, but also TX message objects, this patch renames IF_RX to
IF_NAPI.

Link: https://lore.kernel.org/r/20210809080608.171545-1-mkl@pengutronix.de
Cc: Dario Binacchi <dariobin@libero.it>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:05 +02:00
Marc Kleine-Budde
236de85f6a can: c_can: c_can_do_tx(): fix typo in comment
This patch fixes a typo in the comment in c_can_do_tx().

Fixes: eddf671150 ("can: c_can: add a comment about IF_RX interface's use")
Link: https://lore.kernel.org/r/20210806105127.103302-1-mkl@pengutronix.de
Cc: Dario Binacchi <dariobin@libero.it>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:05 +02:00
Dario Binacchi
06fc143b2e dt-bindings: net: can: c_can: convert to json-schema
Convert the Bosch C_CAN/D_CAN controller device tree binding
documentation to json-schema.

Document missing properties.
Remove "ti,hwmods" as it is no longer used in TI dts.
Make "clocks" required as it is used in all dts.
Update the examples.

Link: https://lore.kernel.org/r/20210805192750.9051-1-dariobin@libero.it
Signed-off-by: Dario Binacchi <dariobin@libero.it>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:05 +02:00
Matt Kline
812270e544 can: m_can: Batch FIFO writes during CAN transmit
Give FIFO writes the same treatment as reads to avoid fixed costs of
individual transfers on a slow bus (e.g., tcan4x5x).

Link: https://lore.kernel.org/r/20210817050853.14875-4-matt@bitbashing.io
Signed-off-by: Matt Kline <matt@bitbashing.io>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:04 +02:00
Matt Kline
1aa6772f64 can: m_can: Batch FIFO reads during CAN receive
On peripherals communicating over a relatively slow SPI line
(e.g. tcan4x5x), individual transfers have high fixed costs.
This causes the driver to spend most of its time waiting between
transfers and severely limits throughput.

Reduce these overheads by reading more than one word at a time.
Writing could get a similar treatment in follow-on commits.

Link: https://lore.kernel.org/r/20210817050853.14875-3-matt@bitbashing.io
Signed-off-by: Matt Kline <matt@bitbashing.io>
[mkl: remove __packed from struct id_and_dlc]
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:04 +02:00
Matt Kline
e39381770e can: m_can: Disable IRQs on FIFO bus errors
If FIFO reads or writes fail due to the underlying regmap (e.g., SPI)
I/O, propagate that up to the m_can driver, log an error, and disable
interrupts, similar to the mcp251xfd driver.

While reworking the FIFO functions to add this error handling,
add support for bulk reads and writes of multiple registers.

Link: https://lore.kernel.org/r/20210817050853.14875-2-matt@bitbashing.io
Signed-off-by: Matt Kline <matt@bitbashing.io>
[mkl: re-wrap long lines, remove WARN_ON, convert to netdev block comments]
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:04 +02:00
Marc Kleine-Budde
5020ced445 can: m_can: fix block comment style
This patch fixes the commenting style in the m_can driver.

Fixes: 1be37d3b04 ("can: m_can: fix periph RX path: use rx-offload to ensure skbs are sent from softirq context")
Fixes: df06fd6782 ("can: m_can: m_can_chip_config(): enable and configure internal timestamps")
Link: https://lore.kernel.org/r/20210819111703.599686-2-mkl@pengutronix.de
Cc: Chandrasekar Ramakrishnan <rcsekar@samsung.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:04 +02:00
Marc Kleine-Budde
fede1ae2d3 can: tcan4x5x: cdev_to_priv(): remove stray empty line
This patch removes a stray empty line in the cdev_to_priv() function.

Fixes: ac33ffd3e2 ("can: m_can: let m_can_class_allocate_dev() allocate driver specific private data")
Link: https://lore.kernel.org/r/20210819111703.599686-1-mkl@pengutronix.de
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:04 +02:00
Lad Prabhakar
76e9353a80 can: rcar_canfd: Add support for RZ/G2L family
CANFD block on RZ/G2L SoC is almost identical to one found on
R-Car Gen3 SoC's. On RZ/G2L SoC interrupt sources for each channel
are split into different sources and the IP doesn't divide (1/2)
CANFD clock within the IP.

This patch adds compatible string for RZ/G2L family and splits
the irq handlers to accommodate both RZ/G2L and R-Car Gen3 SoC's.

Link: https://lore.kernel.org/r/20210727133022.634-3-prabhakar.mahadev-lad.rj@bp.renesas.com
Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Biju Das <biju.das.jz@bp.renesas.com>
[mkl: fixed typo: recieve -> receive, thanks Geert]
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:04 +02:00
Lad Prabhakar
1aa5a06c0a dt-bindings: net: can: renesas,rcar-canfd: Document RZ/G2L SoC
Add CANFD binding documentation for Renesas RZ/G2L SoC.

Link: https://lore.kernel.org/r/20210727133022.634-2-prabhakar.mahadev-lad.rj@bp.renesas.com
Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:03 +02:00
Marc Kleine-Budde
b2fcc70799 can: mcp251xfd: mark some instances of struct mcp251xfd_priv as const
With the patch 07ff4aed01 ("time/timecounter: Mark 1st argument of
timecounter_cyc2time() as const") some instances of the struct
mcp251xfd_priv can be marked as const. This patch marks these as
const.

Link: https://lore.kernel.org/r/20210813091027.159379-1-mkl@pengutronix.de
Cc: Manivannan Sadhasivam <mani@kernel.org>
Cc: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:03 +02:00
Vincent Mailhol
c734707820 can: etas_es58x: clean-up documentation of struct es58x_fd_tx_conf_msg
The documentation of struct es58x_fd_tx_conf_msg explains in details
the different TDC parameters. However, those description are redundant
with the documentation of struct can_tdc.

Remove most of the description.

Also, fixes a typo in the reference to the datasheet (E701 -> E70).

Link: https://lore.kernel.org/r/20210815033248.98111-8-mailhol.vincent@wanadoo.fr
Signed-off-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:03 +02:00
Vincent Mailhol
7a4573cf3a MAINTAINERS: add Vincent MAILHOL as maintainer for the ETAS ES58X CAN/USB driver
Adding myself (Vincent Mailhol) as a maintainer for the ETAS ES58X
CAN/USB driver.

Link: https://lore.kernel.org/r/20210814093353.74391-1-mailhol.vincent@wanadoo.fr
Signed-off-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:03 +02:00
Vincent Mailhol
e43aaa0fef can: netlink: allow user to turn off unsupported features
The sanity checks on the control modes will reject any request related
to an unsupported features, even turning it off.

Example on an interface which does not support CAN-FD:

$ ip link set can0 type can bitrate 500000 fd off
RTNETLINK answers: Operation not supported

This patch lets such command go through (but requests to turn on an
unsupported feature are, of course, still denied).

Link: https://lore.kernel.org/r/20210815033248.98111-2-mailhol.vincent@wanadoo.fr
Signed-off-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:03 +02:00
Oleksij Rempel
6e86a1543c can: dev: provide optional GPIO based termination support
For CAN buses to work, a termination resistor has to be present at both
ends of the bus. This resistor is usually 120 Ohms, other values may be
required for special bus topologies.

This patch adds support for a generic GPIO based CAN termination. The
resistor value has to be specified via device tree, and it can only be
attached to or detached from the bus. By default the termination is not
active.

Link: https://lore.kernel.org/r/20210818071232.20585-4-o.rempel@pengutronix.de
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:03 +02:00
Oleksij Rempel
fe7edf2482 dt-bindings: can: fsl,flexcan: enable termination-* bindings
Enable termination-* binding and provide validation example for it.

Link: https://lore.kernel.org/r/20210818071232.20585-3-o.rempel@pengutronix.de
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:03 +02:00
Oleksij Rempel
ef82641d68 dt-bindings: can-controller: add support for termination-gpios
Some boards provide GPIO controllable termination resistor. Provide
binding to make use of it.

Link: https://lore.kernel.org/r/20210818071232.20585-2-o.rempel@pengutronix.de
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2021-08-19 15:07:02 +02:00
Gal Pressman
f6018cc460 RDMA/uverbs: Track dmabuf memory regions
The dmabuf memory registrations are missing the restrack handling and
hence do not appear in rdma tool.

Fixes: bfe0cc6eb2 ("RDMA/uverbs: Add uverbs command for dma-buf based MR registration")
Link: https://lore.kernel.org/r/20210812135607.6228-1-galpress@amazon.com
Signed-off-by: Gal Pressman <galpress@amazon.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-08-19 09:59:53 -03:00
Maor Gottlieb
da78fe5fb3 RDMA/mlx5: Fix crash when unbind multiport slave
Fix the below crash when deleting a slave from the unaffiliated list
twice. First time when the slave is bound to the master and the second
when the slave is unloaded.

Fix it by checking if slave is unaffiliated (doesn't have ib device)
before removing from the list.

  RIP: 0010:mlx5r_mp_remove+0x4e/0xa0 [mlx5_ib]
  Call Trace:
   auxiliary_bus_remove+0x18/0x30
   __device_release_driver+0x177/x220
   device_release_driver+0x24/0x30
   bus_remove_device+0xd8/0x140
   device_del+0x18a/0x3e0
   mlx5_rescan_drivers_locked+0xa9/0x210 [mlx5_core]
   mlx5_unregister_device+0x34/0x60 [mlx5_core]
   mlx5_uninit_one+0x32/0x100 [mlx5_core]
   remove_one+0x6e/0xe0 [mlx5_core]
   pci_device_remove+0x36/0xa0
   __device_release_driver+0x177/0x220
   device_driver_detach+0x3c/0xa0
   unbind_store+0x113/0x130
   kernfs_fop_write_iter+0x110/0x1a0
   new_sync_write+0x116/0x1a0
   vfs_write+0x1ba/0x260
   ksys_write+0x5f/0xe0
   do_syscall_64+0x3d/0x90
   entry_SYSCALL_64_after_hwframe+0x44/0xae

Fixes: 93f8244431 ("RDMA/mlx5: Convert mlx5_ib to use auxiliary bus")
Link: https://lore.kernel.org/r/17ec98989b0ba88f7adfbad68eb20bce8d567b44.1628587493.git.leonro@nvidia.com
Signed-off-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-08-19 09:59:20 -03:00
Chuck Lever
729580ddc5 svcrdma: xpt_bc_xprt is already clear in __svc_rdma_free()
svc_xprt_free() already "puts" the bc_xprt before calling the
transport's "free" method. No need to do it twice.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2021-08-19 08:29:32 -04:00
Colin Ian King
5c8a2bb481 net: ethernet: ti: cpsw: make array stpa static const, makes object smaller
Don't populate the array stpa on the stack but instead it
static const. Makes the object code smaller by 81 bytes:

Before:
   text    data   bss    dec    hex filename
  54993   17248     0  72241  11a31 ./drivers/net/ethernet/ti/cpsw_new.o

After:
   text    data   bss    dec    hex filename
  54784   17376     0  72160  119e0 ./drivers/net/ethernet/ti/cpsw_new.o

(gcc version 10.3.0)

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 13:23:20 +01:00
Colin Ian King
0bc277cb82 net: hns3: make array spec_opcode static const, makes object smaller
Don't populate the array spec_opcode on the stack but instead it
static const. Makes the object code smaller by 158 bytes:

Before:
   text   data   bss     dec    hex filename
  12271   3976   128   16375   3ff7 .../hisilicon/hns3/hns3pf/hclge_cmd.o

After:
   text   data   bss     dec    hex filename
  12017   4072   128   16217   3f59 .../hisilicon/hns3/hns3pf/hclge_cmd.o

(gcc version 10.3.0)

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 13:23:20 +01:00
Colin Ian King
36d5825bab hinic: make array speeds static const, makes object smaller
Don't populate the array speeds on the stack but instead it
static const. Makes the object code smaller by 17 bytes:

Before:
   text    data     bss     dec     hex filename
  39987   14200      64   54251    d3eb .../huawei/hinic/hinic_sriov.o

After:
   text    data     bss     dec     hex filename
  39906   14264      64   54234    d3da .../huawei/hinic/hinic_sriov.o

(gcc version 10.3.0)

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 13:23:20 +01:00
David S. Miller
9f3ebe8fb5 Merge branch 'indirect-qdisc-order'
Eli Cohen says:

====================
Indirect dev ingress qdisc creation order

The first patch is just a cleanup of the code.
The second patch is fixing the dependency in ingress qdisc creation
relative to offloading driver registration to filter configurations.

v1 -> v2:
Fix warning - variable set but not used
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 13:19:30 +01:00
Eli Cohen
74fc4f8287 net: Fix offloading indirect devices dependency on qdisc order creation
Currently, when creating an ingress qdisc on an indirect device before
the driver registered for callbacks, the driver will not have a chance
to register its filter configuration callbacks.

To fix that, modify the code such that it keeps track of all the ingress
qdiscs that call flow_indr_dev_setup_offload(). When a driver calls
flow_indr_dev_register(),  go through the list of tracked ingress qdiscs
and call the driver callback entry point so as to give it a chance to
register its callback.

Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Eli Cohen <elic@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 13:19:30 +01:00
Eli Cohen
c1c5cb3aee net/core: Remove unused field from struct flow_indr_dev
rcu field is not used. Remove it.

Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Eli Cohen <elic@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 13:19:30 +01:00
Pavel Skripkin
2274af1d60 net: mii: make mii_ethtool_gset() return void
mii_ethtool_gset() does not return any errors. Since there are no users
of this function that rely on its return value, it can be
made void.

Signed-off-by: Pavel Skripkin <paskripkin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 13:06:53 +01:00
Pavel Skripkin
9fcfd0888c net: pch_gbe: remove mii_ethtool_gset() error handling
mii_ethtool_gset() does not return any errors, so error handling can be
omitted to make code more simple.

Acked-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Pavel Skripkin <paskripkin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 13:06:53 +01:00
David S. Miller
c15128c97b Merge branch 'r8152-bp-settings'
Hayes Wang says:

====================
r8152: fix bp settings

Fix the wrong bp settings of the firmware.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:19:30 +01:00
Hayes Wang
6633fb83f1 r8152: fix the maximum number of PLA bp for RTL8153C
The maximum PLA bp number of RTL8153C is 16, not 8. That is, the
bp 0 ~ 15 are at 0xfc28 ~ 0xfc46, and the bp_en is at 0xfc48.

Fixes: 195aae321c ("r8152: support new chips")
Signed-off-by: Hayes Wang <hayeswang@realtek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:19:30 +01:00
Hayes Wang
a876a33d2a r8152: fix writing USB_BP2_EN
The register of USB_BP2_EN is 16 bits, so we should use
ocp_write_word(), not ocp_write_byte().

Fixes: 9370f2d05a ("support request_firmware for RTL8153")
Signed-off-by: Hayes Wang <hayeswang@realtek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:19:30 +01:00
David S. Miller
d98c821067 Merge branch 'mptcp-fixes'
Mat Martineau says:

====================
mptcp: Bug fixes

Here are two bug fixes for the net tree:

Patch 1 fixes a memory leak that could be encountered when clearing the
list of advertised MPTCP addresses.

Patch 2 fixes a protocol issue early in an MPTCP connection, to ensure
both peers correctly understand that the full MPTCP connection handshake
has completed even when the server side quickly sends an ADD_ADDR
option.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:17:05 +01:00
Matthieu Baerts
67b12f792d mptcp: full fully established support after ADD_ADDR
If directly after an MP_CAPABLE 3WHS, the client receives an ADD_ADDR
with HMAC from the server, it is enough to switch to a "fully
established" mode because it has received more MPTCP options.

It was then OK to enable the "fully_established" flag on the MPTCP
socket. Still, best to check if the ADD_ADDR looks valid by looking if
it contains an HMAC (no 'echo' bit). If an ADD_ADDR echo is received
while we are not in "fully established" mode, it is strange and then
we should not switch to this mode now.

But that is not enough. On one hand, the path-manager has be notified
the state has changed. On the other hand, the "fully_established" flag
on the subflow socket should be turned on as well not to re-send the
MP_CAPABLE 3rd ACK content with the next ACK.

Fixes: 84dfe3677a ("mptcp: send out dedicated ADD_ADDR packet")
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:16:54 +01:00
Paolo Abeni
a0eea5f10e mptcp: fix memory leak on address flush
The endpoint cleanup path is prone to a memory leak, as reported
by syzkaller:

 BUG: memory leak
 unreferenced object 0xffff88810680ea00 (size 64):
   comm "syz-executor.6", pid 6191, jiffies 4295756280 (age 24.138s)
   hex dump (first 32 bytes):
     58 75 7d 3c 80 88 ff ff 22 01 00 00 00 00 ad de  Xu}<....".......
     01 00 02 00 00 00 00 00 ac 1e 00 07 00 00 00 00  ................
   backtrace:
     [<0000000072a9f72a>] kmalloc include/linux/slab.h:591 [inline]
     [<0000000072a9f72a>] mptcp_nl_cmd_add_addr+0x287/0x9f0 net/mptcp/pm_netlink.c:1170
     [<00000000f6e931bf>] genl_family_rcv_msg_doit.isra.0+0x225/0x340 net/netlink/genetlink.c:731
     [<00000000f1504a2c>] genl_family_rcv_msg net/netlink/genetlink.c:775 [inline]
     [<00000000f1504a2c>] genl_rcv_msg+0x341/0x5b0 net/netlink/genetlink.c:792
     [<0000000097e76f6a>] netlink_rcv_skb+0x148/0x430 net/netlink/af_netlink.c:2504
     [<00000000ceefa2b8>] genl_rcv+0x24/0x40 net/netlink/genetlink.c:803
     [<000000008ff91aec>] netlink_unicast_kernel net/netlink/af_netlink.c:1314 [inline]
     [<000000008ff91aec>] netlink_unicast+0x537/0x750 net/netlink/af_netlink.c:1340
     [<0000000041682c35>] netlink_sendmsg+0x846/0xd80 net/netlink/af_netlink.c:1929
     [<00000000df3aa8e7>] sock_sendmsg_nosec net/socket.c:704 [inline]
     [<00000000df3aa8e7>] sock_sendmsg+0x14e/0x190 net/socket.c:724
     [<000000002154c54c>] ____sys_sendmsg+0x709/0x870 net/socket.c:2403
     [<000000001aab01d7>] ___sys_sendmsg+0xff/0x170 net/socket.c:2457
     [<00000000fa3b1446>] __sys_sendmsg+0xe5/0x1b0 net/socket.c:2486
     [<00000000db2ee9c7>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
     [<00000000db2ee9c7>] do_syscall_64+0x38/0x90 arch/x86/entry/common.c:80
     [<000000005873517d>] entry_SYSCALL_64_after_hwframe+0x44/0xae

We should not require an allocation to cleanup stuff.

Rework the code a bit so that the additional RCU work is no more needed.

Fixes: 1729cf186d ("mptcp: create the listening socket for new port")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:16:54 +01:00
David S. Miller
a27919433b Merge branch 'ravb-gbit'
Biju Das says:

====================
ravb: Add Gigabit Ethernet driver support

The DMAC and EMAC blocks of Gigabit Ethernet IP found on RZ/G2L SoC are
similar to the R-Car Ethernet AVB IP.

The Gigabit Ethernet IP consists of Ethernet controller (E-MAC), Internal
TCP/IP Offload Engine (TOE)  and Dedicated Direct memory access controller
(DMAC).

With a few changes in the driver we can support both IPs.

Currently a runtime decision based on the chip type is used to distinguish
the HW differences between the SoC families.

This patch series is in preparation for supporting the RZ/G2L SoC by
replacing driver data chip type with struct ravb_hw_info by moving chip
type to it and also adding gstrings_stats, gstrings_size, net_hw_features,
net_features, aligned_tx, stats_len, max_rx_len variables to
it. This patch also adds the feature bit for {RX, TX} clock internal
delays and TX counters HW features found on R-Car Gen3 to struct
ravb_hw_info.

This patch series is based on net-next.
v2->v3:
 * Removed num_gstat_queue variable from struct ravb_hw_info.
 * started using unsigned int for num_tx_desc variable in struct ravb_private
 * split the patch 'Add struct ravb_hw_info to driver data' into two
 * Renamed skb_sz to max_rx_len and tx_drop_cntrs to tx_counters
   and also updated the comments.
v1->v2:
 * Replaced driver data chip type with struct ravb_hw_info
 * Added gstrings_stats, gstrings_size, net_hw_features, net_features,
   num_gstat_queue, num_tx_desc, stats_len, skb_sz to struct ravb_hw_info
 * Added internal_delay and tx_drop_cntrs hw feature bit to struct ravb_hw_info

RFC->V1
  * Incorporated feedback from Andrew, Sergei, Geert and Prabhakar
  * https://patchwork.kernel.org/project/linux-renesas-soc/list/?series=515525
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:05:29 +01:00
Biju Das
0b81d67311 ravb: Add tx_counters to struct ravb_hw_info
The register for retrieving TX counters is present only on R-Car Gen3
and RZ/G2L; it is not present on R-Car Gen2.

Add the tx_counters hw feature bit to struct ravb_hw_info, to enable this
feature specifically for R-Car Gen3 now and later extend it to RZ/G2L.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:05:16 +01:00
Biju Das
8bc4caa0ab ravb: Add internal delay hw feature to struct ravb_hw_info
R-Car Gen3 supports TX and RX clock internal delay modes, whereas R-Car
Gen2 and RZ/G2L do not support it.
Add an internal_delay hw feature bit to struct ravb_hw_info to enable this
only for R-Car Gen3.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Sergei Shtylyov <sergei.shtylyov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:05:16 +01:00
Biju Das
8912ed25da ravb: Add net_features and net_hw_features to struct ravb_hw_info
On R-Car the checksum calculation on RX frames is done by the E-MAC
module, whereas on RZ/G2L it is done by the TOE.

TOE calculates the checksum of received frames from E-MAC and outputs it to
DMAC. TOE also calculates the checksum of transmission frames from DMAC and
outputs it E-MAC.

Add net_features and net_hw_features to struct ravb_hw_info, to support
subsequent SoCs without any code changes in the ravb_probe function.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Sergei Shtylyov <sergei.shtylyov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:05:16 +01:00
Biju Das
896a818e0e ravb: Add gstrings_stats and gstrings_size to struct ravb_hw_info
The device stats strings for R-Car and RZ/G2L are different.

R-Car provides 30 device stats, whereas RZ/G2L provides only 15. In
addition, RZ/G2L has stats "rx_queue_0_csum_offload_errors" instead of
"rx_queue_0_missed_errors".

Add structure variables gstrings_stats and gstrings_size to struct
ravb_hw_info, so that subsequent SoCs can be added without any code
changes in the ravb_get_strings function.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Sergei Shtylyov <sergei.shtylyov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:05:16 +01:00
Biju Das
25154301fc ravb: Add stats_len to struct ravb_hw_info
R-Car provides 30 device stats, whereas RZ/G2L provides only 15. In
addition, RZ/G2L has stats "rx_queue_0_csum_offload_errors" instead of
"rx_queue_0_missed_errors".

Replace RAVB_STATS_LEN macro with a structure variable stats_len to
struct ravb_hw_info, to support subsequent SoCs without any code changes
to the ravb_get_sset_count function.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Sergei Shtylyov <sergei.shtylyov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:05:15 +01:00
Biju Das
cb01c672c2 ravb: Add max_rx_len to struct ravb_hw_info
The maximum descriptor size that can be specified on the reception side for
R-Car is 2048 bytes, whereas for RZ/G2L it is 8096.

Add the max_rx_len variable to struct ravb_hw_info for allocating different
RX skb buffer sizes for R-Car and RZ/G2L using the netdev_alloc_skb
function.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:05:15 +01:00
Biju Das
68ca3c9232 ravb: Add aligned_tx to struct ravb_hw_info
R-Car Gen2 needs a 4byte aligned address for the transmission buffer,
whereas R-Car Gen3 doesn't have any such restriction.

Add aligned_tx to struct ravb_hw_info to select the driver to choose
between aligned and unaligned tx buffers.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:05:15 +01:00
Biju Das
ebb091461a ravb: Add struct ravb_hw_info to driver data
The DMAC and EMAC blocks of Gigabit Ethernet IP found on RZ/G2L SoC are
similar to the R-Car Ethernet AVB IP. With a few changes in the driver we
can support both IPs.

This patch adds the struct ravb_hw_info to hold hw features, driver data
and function pointers to support both the IPs. It also replaces the driver
data chip type with struct ravb_hw_info by moving chip type to it.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-08-19 12:05:15 +01:00