Rdma write with inline flag when sending small packages,
whose length is shorter than the qp's max_inline_data, can
help reducing latency.
In my test environment, which are 2 VMs running on the same
physical host and whose NICs(ConnectX-4Lx) are working on
SR-IOV mode, qperf shows 0.5us-0.7us improvement in latency.
Test command:
server: smc_run taskset -c 1 qperf
client: smc_run taskset -c 1 qperf <server ip> -oo \
msg_size:1:2K:*2 -t 30 -vu tcp_lat
The results shown below:
msgsize before after
1B 11.2 us 10.6 us (-0.6 us)
2B 11.2 us 10.7 us (-0.5 us)
4B 11.3 us 10.7 us (-0.6 us)
8B 11.2 us 10.6 us (-0.6 us)
16B 11.3 us 10.7 us (-0.6 us)
32B 11.3 us 10.6 us (-0.7 us)
64B 11.2 us 11.2 us (0 us)
128B 11.2 us 11.2 us (0 us)
256B 11.2 us 11.2 us (0 us)
512B 11.4 us 11.3 us (-0.1 us)
1KB 11.4 us 11.5 us (0.1 us)
2KB 11.5 us 11.5 us (0 us)
Signed-off-by: Guangguan Wang <guangguan.wang@linux.alibaba.com>
Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
Tested-by: kernel test robot <lkp@intel.com>
Acked-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
As cdc msg's length is 44B, cdc msgs can be sent inline in
most rdma devices, which can help reducing sending latency.
In my test environment, which are 2 VMs running on the same
physical host and whose NICs(ConnectX-4Lx) are working on
SR-IOV mode, qperf shows 0.4us-0.7us improvement in latency.
Test command:
server: smc_run taskset -c 1 qperf
client: smc_run taskset -c 1 qperf <server ip> -oo \
msg_size:1:2K:*2 -t 30 -vu tcp_lat
The results shown below:
msgsize before after
1B 11.9 us 11.2 us (-0.7 us)
2B 11.7 us 11.2 us (-0.5 us)
4B 11.7 us 11.3 us (-0.4 us)
8B 11.6 us 11.2 us (-0.4 us)
16B 11.7 us 11.3 us (-0.4 us)
32B 11.7 us 11.3 us (-0.4 us)
64B 11.7 us 11.2 us (-0.5 us)
128B 11.6 us 11.2 us (-0.4 us)
256B 11.8 us 11.2 us (-0.6 us)
512B 11.8 us 11.4 us (-0.4 us)
1KB 11.9 us 11.4 us (-0.5 us)
2KB 12.1 us 11.5 us (-0.6 us)
Signed-off-by: Guangguan Wang <guangguan.wang@linux.alibaba.com>
Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
Tested-by: kernel test robot <lkp@intel.com>
Acked-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
As per Errata Section 5.1, if EEE is intended to be used, some register
writes must be done once after every hardware reset. This patch now adds
the necessary register writes as listed in the Marvell errata.
Without this fix we experience ethernet problems on some of our boards
equipped with a new version of this ethernet PHY (different supplier).
The fix applies to Marvell Alaska 88E1510/88E1518/88E1512/88E1514
Rev. A0.
Signed-off-by: Leszek Polak <lpolak@arri.de>
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Marek Behún <kabel@kernel.org>
Cc: Andrew Lunn <andrew@lunn.ch>
Cc: Heiner Kallweit <hkallweit1@gmail.com>
Cc: Russell King <linux@armlinux.org.uk>
Cc: David S. Miller <davem@davemloft.net>
Reviewed-by: Marek Behún <kabel@kernel.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20220516070859.549170-1-sr@denx.de
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Calling synchronize_irq() right before free_irq() is quite useless. On one
hand the IRQ can easily fire again before free_irq() is entered, on the
other hand free_irq() itself calls synchronize_irq() internally (in a race
condition free way), before any state associated with the IRQ is freed.
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Link: https://lore.kernel.org/r/20220516082251.1651350-1-chi.minghao@zte.com.cn
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Calling synchronize_irq() right before free_irq() is quite useless. On one
hand the IRQ can easily fire again before free_irq() is entered, on the
other hand free_irq() itself calls synchronize_irq() internally (in a race
condition free way), before any state associated with the IRQ is freed.
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Link: https://lore.kernel.org/r/20220516081914.1651281-1-chi.minghao@zte.com.cn
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Calling synchronize_irq() right before free_irq() is quite useless. On one
hand the IRQ can easily fire again before free_irq() is entered, on the
other hand free_irq() itself calls synchronize_irq() internally (in a race
condition free way), before any state associated with the IRQ is freed.
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Link: https://lore.kernel.org/r/20220516072646.1651109-1-chi.minghao@zte.com.cn
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
devm_kfree check the pointer, there is no need to check before
devm_kfree call.
This change is to cleanup the code a bit.
Signed-off-by: Bernard Zhao <bernard@vivo.com>
Link: https://lore.kernel.org/r/20220516015208.6526-1-bernard@vivo.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
-----BEGIN PGP SIGNATURE-----
iQFHBAABCgAxFiEEBsvAIBsPu6mG7thcrX5LkNig010FAmKCr/0THG1rbEBwZW5n
dXRyb25peC5kZQAKCRCtfkuQ2KDTXbVWCAClh9ywNEPiSxXsB2k9EdVoijz0hv6I
y6YLFPfHxc8w0ewOnBxJOBeLhVjIE2K3n81NaEudheBBImWV8CGlFPHu5sI4S01b
aFcAuLT6SOat52xzkh44WGGjW517+EZM1Of1xzPhKdnR0nLvva52SpJlYwgLrkCN
O6r30cDuU558H4g2BNxavbbNFPzL7HhkBpHmgWQg4q2p7LvLuTE6sN2CvNEgUWYM
nIyK6P/FIyy3l7M7E873WhIEGNGRnAXRBEPXivRmKTlw0wizcvPxlWKm4YxTA0WL
30fw58Hy4x6SZD+7+sP0QhdjiYOofgMtZFZn/LaJYE+rfaYLsFUH3A/2
=tUrE
-----END PGP SIGNATURE-----
Merge tag 'linux-can-next-for-5.19-20220516' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next
Marc Kleine-Budde says:
====================
pull-request: can-next 2022-05-16
the first 2 patches are by me and target the CAN raw protocol. The 1st
removes an unneeded assignment, the other one adds support for
SO_TXTIME/SCM_TXTIME.
Oliver Hartkopp contributes 2 patches for the ISOTP protocol. The 1st
adds support for transmission without flow control, the other let's
bind() return an error on incorrect CAN ID formatting.
Geert Uytterhoeven contributes a patch to clean up ctucanfd's Kconfig
file.
Vincent Mailhol's patch for the slcan driver uses the proper function
to check for invalid CAN frames in the xmit callback.
The next patch is by Geert Uytterhoeven and makes the interrupt-names
of the renesas,rcar-canfd dt bindings mandatory.
A patch by my update the ctucanfd dt bindings to include the common
CAN controller bindings.
The last patch is by Akira Yokosawa and fixes a breakage the
ctucanfd's documentation.
* tag 'linux-can-next-for-5.19-20220516' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next:
docs: ctucanfd: Use 'kernel-figure' directive instead of 'figure'
dt-bindings: can: ctucanfd: include common CAN controller bindings
dt-bindings: can: renesas,rcar-canfd: Make interrupt-names required
can: slcan: slc_xmit(): use can_dropped_invalid_skb() instead of manual check
can: ctucanfd: Let users select instead of depend on CAN_CTUCANFD
can: isotp: isotp_bind(): return -EINVAL on incorrect CAN ID formatting
can: isotp: add support for transmission without flow control
can: raw: add support for SO_TXTIME/SCM_TXTIME
can: raw: raw_sendmsg(): remove not needed setting of skb->sk
====================
Link: https://lore.kernel.org/r/20220516202625.1129281-1-mkl@pengutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
skb_data_area_size() is not needed. As Jakub pointed out [1]:
For Rx, drivers can use the size passed during skb allocation or
use skb_tailroom().
For Tx, drivers should use skb_headlen().
[1] https://lore.kernel.org/netdev/CAHNKnsTmH-rGgWi3jtyC=ktM1DW2W1VJkYoTMJV2Z_Bt498bsg@mail.gmail.com/
Signed-off-by: Ricardo Martinez <ricardo.martinez@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
skb_data_area_size() helper was used to calculate the size of the
DMA mapped buffer passed to the HW. Instead of doing this, use the
size passed to allocate the skbs.
Signed-off-by: Ricardo Martinez <ricardo.martinez@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Mat Martineau says:
====================
mptcp: Updates for net-next
Three independent fixes/features from the MPTCP tree:
Patch 1 is a selftest workaround for older iproute2 packages.
Patch 2 removes superfluous locks that were added with recent MP_FAIL
patches.
Patch 3 adds support for the TCP_DEFER_ACCEPT sockopt.
====================
Link: https://lore.kernel.org/r/20220514002115.725976-1-mathew.j.martineau@linux.intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Support this via passthrough to the underlying tcp listener socket.
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/271
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This reverts commit 4293248c67.
Additional locks are not needed, all the touched sections
are already under mptcp socket lock protection.
Fixes: 4293248c67 ("mptcp: add data lock for sk timers")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Old tc versions (iproute2 5.3) show actions in multiple lines, not a
single line. Then the following unexpected MP_FAIL selftest output
occurs:
file received by server has inverted byte at 169
./mptcp_join.sh: line 1277: [: [{"total acts":1},{"actions":[{"order":0 pedit ,"control_action":{"type":"pipe"}keys 1
index 1 ref 1 bind 1,"installed":0,"last_used":0
key #0 at 148: val ff000000 mask ffffffff
5: integer expression expected
001 Infinite map syn[ ok ] - synack[ ok ] - ack[ ok ]
sum[ ok ] - csum [ ok ]
ftx[ ok ] - failrx[ ok ]
rtx[ ok ] - rstrx [ ok ]
itx[ ok ] - infirx[ ok ]
ftx[ ok ] - failrx[ ok ] invert
This patch adds a 'grep' before 'sed' to fix this.
Fixes: b6e074e171 ("selftests: mptcp: add infinite map testcase")
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Two issues were observed in the ReST doc added by commit c3a0addefb
("docs: ctucanfd: CTU CAN FD open-source IP core documentation.")
with Sphinx versions 2.4.4 and 4.5.0.
The plain "figure" directive broke "make pdfdocs" due to a missing
PDF figure. For conversion of SVG -> PDF to work, the "kernel-figure"
directive, which is an extension for kernel documentation, should
be used instead.
The directive of "code:: raw" causes a warning from both
"make htmldocs" and "make pdfdocs", which reads:
[...]/can/ctu/ctucanfd-driver.rst:75: WARNING: Pygments lexer name
'raw' is not known
A plain literal-block marker should suffice where no syntax
highlighting is intended.
Fix the issues by using suitable directive and marker.
Fixes: c3a0addefb ("docs: ctucanfd: CTU CAN FD open-source IP core documentation.")
Link: https://lore.kernel.org/all/5986752a-1c2a-5d64-f91d-58b1e6decd17@gmail.com
Signed-off-by: Akira Yokosawa <akiyks@gmail.com>
Acked-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>
Cc: Martin Jerabek <martin.jerabek01@gmail.com>
Cc: Ondrej Ille <ondrej.ille@gmail.com>
Cc: Marc Kleine-Budde <mkl@pengutronix.de>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Since commit
| 1f9234401c ("dt-bindings: can: add can-controller.yaml")
there is a common CAN controller binding. Add this to the ctucanfd
binding.
Cc: Ondrej Ille <ondrej.ille@gmail.com>
Acked-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
The Renesas R-Car CAN FD Controller always uses two or more interrupts.
Make the interrupt-names properties a required property, to make it
easier to identify the individual interrupts.
Update the example accordingly.
Link: https://lore.kernel.org/all/a68e65955e0df4db60233d468f348203c2e7b940.1651512451.git.geert+renesas@glider.be
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
slcan does a manual check in slc_xmit() to verify if the skb is valid.
This check is incomplete, use instead can_dropped_invalid_skb().
Link: https://lore.kernel.org/all/20220514141650.1109542-2-mailhol.vincent@wanadoo.fr
Signed-off-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
The CTU CAN-FD IP core is only useful when used with one of the
corresponding PCI/PCIe or platform (FPGA, SoC) drivers, which depend on
PCI resp. OF.
Hence make the users select the core driver code, instead of letting
then depend on it. Keep the core code config option visible when
compile-testing, to maintain compile-coverage.
Link: https://lore.kernel.org/all/887b7440446b6244a20a503cc6e8dc9258846706.1652104941.git.geert+renesas@glider.be
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Commit 3ea566422c ("can: isotp: sanitize CAN ID checks in
isotp_bind()") checks the given CAN ID address information by
sanitizing the input values.
This check (silently) removes obsolete bits by masking the given CAN
IDs.
Derek Will suggested to give a feedback to the application programmer
when the 'sanitizing' was actually needed which means the programmer
provided CAN ID content in a wrong format (e.g. SFF CAN IDs with a CAN
ID > 0x7FF).
Link: https://lore.kernel.org/all/20220515181633.76671-1-socketcan@hartkopp.net
Suggested-by: Derek Will <derekrobertwill@gmail.com>
Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Usually the ISO 15765-2 protocol is a point-to-point protocol to transfer
segmented PDUs to a dedicated receiver. This receiver sends a flow control
message to specify protocol options and timings (e.g. block size / STmin).
The so called functional addressing communication allows a 1:N
communication but is limited to a single frame length.
This new CAN_ISOTP_CF_BROADCAST allows an unconfirmed 1:N communication
with PDU length that would not fit into a single frame. This feature is
not covered by the ISO 15765-2 standard.
Link: https://lore.kernel.org/all/20220507115558.19065-1-socketcan@hartkopp.net
Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
This patch calls into sock_cmsg_send() to parse the user supplied
control information into a struct sockcm_cookie. Then assign the
requested transmit time to the skb.
This makes it possible to use the Earliest TXTIME First (ETF) packet
scheduler with the CAN_RAW protocol. The user can send a CAN_RAW frame
with a TXTIME and the kernel (with the ETF scheduler) will take care
of sending it to the network interface.
Link: https://lore.kernel.org/all/20220502091946.1916211-3-mkl@pengutronix.de
Acked-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
The skb in raw_sendmsg() is allocated with sock_alloc_send_skb(),
which subsequently calls sock_alloc_send_pskb() -> skb_set_owner_w(),
which assigns "skb->sk = sk".
This patch removes the not needed setting of skb->sk.
Link: https://lore.kernel.org/all/20220502091946.1916211-2-mkl@pengutronix.de
Acked-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Now that it is possible to use .probe without having .driver_data, let
KSZ8061 use the kszphy specific hooks for probe,suspend and resume,
which is preferred.
Switch to using the dedicated kszphy probe/suspend/resume functions.
Signed-off-by: Fabio Estevam <festevam@denx.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20220513114613.762810-2-festevam@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently, if the .probe element is present in the phy_driver structure
and the .driver_data is not, a NULL pointer dereference happens.
Allow passing .probe without .driver_data by inserting NULL checks
for priv->type.
Signed-off-by: Fabio Estevam <festevam@denx.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20220513114613.762810-1-festevam@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Calling synchronize_irq() right before free_irq() is quite useless. On one
hand the IRQ can easily fire again before free_irq() is entered, on the
other hand free_irq() itself calls synchronize_irq() internally (in a race
condition free way), before any state associated with the IRQ is freed.
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
t7xx_dl_add_timedout() now return int 'ret', but the return type
is bool. Change the return type to int for furthor errcode upstream.
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
vfree(NULL) is safe. NULL check before vfree() is not needed.
Delete them to simplify the code.
Signed-off-by: Ziyang Xuan <william.xuanziyang@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
octep_init_module misses destroy_workqueue in error path,
this patch fixes that.
Fixes: 862cd659a6 ("octeon_ep: Add driver framework and device initialization")
Signed-off-by: Zheng Bin <zhengbin13@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet says:
====================
net: polish skb defer freeing
While testing this recently added feature on a variety
of platforms/configurations, I found the following issues:
1) A race leading to concurrent calls to smp_call_function_single_async()
2) Missed opportunity to use napi_consume_skb()
3) Need to limit the max length of the per-cpu lists.
4) Process the per-cpu list more frequently, for the
(unusual) case where net_rx_action() has mutiple
napi_poll() to process per round.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
skb_defer_free_flush() can consume cpu cycles,
it seems better to call it in the inner loop:
- Potentially frees page/skb that will be reallocated while hot.
- Account for the cpu cycles in the @time_limit determination.
- Keep softnet_data.defer_count small to reduce chances for
skb_attempt_defer_free() to send an IPI.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit 68822bdf76 ("net: generalize skb freeing
deferral to per-cpu lists") added another per-cpu
cache of skbs. It was expected to be small,
and an IPI was forced whenever the list reached 128
skbs.
We might need to be able to control more precisely
queue capacity and added latency.
An IPI is generated whenever queue reaches half capacity.
Default value of the new limit is 64.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
skb_defer_free_flush() runs from softirq context,
we have the opportunity to refill the napi_alloc_cache,
and/or use kmem_cache_free_bulk() when this cache is full.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A cpu can observe sd->defer_count reaching 128,
and call smp_call_function_single_async()
Problem is that the remote CPU can clear sd->defer_count
before the IPI is run/acknowledged.
Other cpus can queue more packets and also decide
to call smp_call_function_single_async() while the pending
IPI was not yet delivered.
This is a common issue with smp_call_function_single_async().
Callers must ensure correct synchronization and serialization.
I triggered this issue while experimenting smaller threshold.
Performing the call to smp_call_function_single_async()
under sd->defer_lock protection did not solve the problem.
Commit 5a18ceca63 ("smp: Allow smp_call_function_single_async()
to insert locked csd") replaced an informative WARN_ON_ONCE()
with a return of -EBUSY, which is often ignored.
Test of CSD_FLAG_LOCK presence is racy anyway.
Fixes: 68822bdf76 ("net: generalize skb freeing deferral to per-cpu lists")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Works fine on my HP C3600:
[ 274.452394] tulip0: no phy info, aborting mtable build
[ 274.499041] tulip0: MII transceiver #1 config 1000 status 782d advertising 01e1
[ 274.750691] net eth0: Digital DS21142/43 Tulip rev 65 at MMIO 0xf4008000, 00:30:6e:08:7d:21, IRQ 17
[ 283.104520] net eth0: Setting full-duplex based on MII#1 link partner capability of c1e1
Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
hinic_pf_to_mgmt_init misses destroy_workqueue in error path,
this patch fixes that.
Fixes: 6dbb89014d ("hinic: fix sending mailbox timeout in aeq event work")
Signed-off-by: Zheng Bin <zhengbin13@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Menglong Dong says:
====================
net: skb: check the boundrary of skb drop reason
In the commit 1330b6ef33 ("skb: make drop reason booleanable"),
SKB_NOT_DROPPED_YET is added to the enum skb_drop_reason, which makes
the invalid drop reason SKB_NOT_DROPPED_YET can leak to the kfree_skb
tracepoint. Once this happen (it happened, as 4th patch says), it can
cause NULL pointer in drop monitor and result in kernel panic.
Therefore, check the boundrary of drop reason in both kfree_skb_reason
(2th patch) and drop monitor (1th patch) to prevent such case happens
again.
Meanwhile, fix the invalid drop reason passed to kfree_skb_reason() in
tcp_v4_rcv() and tcp_v6_rcv().
Changes since v2:
1/4 - don't reset the reason and print the debug warning only (Jakub
Kicinski)
4/4 - remove new lines between tags
Changes since v1:
- consider tcp_v6_rcv() in the 4th patch
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
The 'drop_reason' that passed to kfree_skb_reason() in tcp_v4_rcv()
and tcp_v6_rcv() can be SKB_NOT_DROPPED_YET(0), as it is used as the
return value of tcp_inbound_md5_hash().
And it can panic the kernel with NULL pointer in
net_dm_packet_report_size() if the reason is 0, as drop_reasons[0]
is NULL.
Fixes: 1330b6ef33 ("skb: make drop reason booleanable")
Reviewed-by: Jiang Biao <benbjiang@tencent.com>
Reviewed-by: Hao Peng <flyingpeng@tencent.com>
Signed-off-by: Menglong Dong <imagedong@tencent.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The SKB_DR_OR() is used to set the drop reason to a value when it is
not set or specified yet. SKB_NOT_DROPPED_YET should also be considered
as not set.
Reviewed-by: Jiang Biao <benbjiang@tencent.com>
Reviewed-by: Hao Peng <flyingpeng@tencent.com>
Signed-off-by: Menglong Dong <imagedong@tencent.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Sometimes, we may forget to reset skb drop reason to NOT_SPECIFIED after
we make it the return value of the functions with return type of enum
skb_drop_reason, such as tcp_inbound_md5_hash. Therefore, its value can
be SKB_NOT_DROPPED_YET(0), which is invalid for kfree_skb_reason().
So we check the range of drop reason in kfree_skb_reason() with
DEBUG_NET_WARN_ON_ONCE().
Reviewed-by: Jiang Biao <benbjiang@tencent.com>
Reviewed-by: Hao Peng <flyingpeng@tencent.com>
Signed-off-by: Menglong Dong <imagedong@tencent.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The 'reason' will be set to 'SKB_DROP_REASON_NOT_SPECIFIED' if it not
small that SKB_DROP_REASON_MAX in net_dm_packet_trace_kfree_skb_hit(),
but it can't avoid it to be 0, which is invalid and can cause NULL
pointer in drop_reasons.
Therefore, reset it to SKB_DROP_REASON_NOT_SPECIFIED when 'reason <= 0'.
Reviewed-by: Jiang Biao <benbjiang@tencent.com>
Reviewed-by: Hao Peng <flyingpeng@tencent.com>
Signed-off-by: Menglong Dong <imagedong@tencent.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Connect with O_NONBLOCK will not be completed immediately
and returns -EINPROGRESS. It is possible to use selector/poll
for completion by selecting the socket for writing. After select
indicates writability, a second connect function call will return
0 to indicate connected successfully as TCP does, but smc returns
-EISCONN. Use socket state for smc to indicate connect state, which
can help smc aligning the connect behaviour with TCP.
Signed-off-by: Guangguan Wang <guangguan.wang@linux.alibaba.com>
Acked-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet says:
====================
net: add annotations for sk->sk_bound_dev_if
While writes on sk->sk_bound_dev_if are protected by socket lock,
we have many lockless reads all over the places.
This is based on syzbot report found in the first patch changelog.
v2: inline ipv6 function only defined if IS_ENABLED(CONFIG_IPV6) (kernel bots)
Change the INET6_MATCH() to inet6_match(), this is no longer a macro.
Change INET_MATCH() to inet_match() (Olivier Hartkopp & Jakub Kicinski)
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
This is no longer a macro, but an inlined function.
INET_MATCH() -> inet_match()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Suggested-by: Olivier Hartkopp <socketcan@hartkopp.net>
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
INET6_MATCH() runs without holding a lock on the socket.
We probably need to annotate most reads.
This patch makes INET6_MATCH() an inline function
to ease our changes.
v2: inline function only defined if IS_ENABLED(CONFIG_IPV6)
Change the name to inet6_match(), this is no longer a macro.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
