| Age | Commit message (Collapse) | Author |
|---|
|
The size of L2TPv2 header with all optional fields is 14 bytes.
l2tp_udp_recv_core only moves 10 bytes to the linear part of a
skb. This may lead to l2tp_recv_common read data outside of a skb.
This patch make sure that there is at least 14 bytes in the linear
part of a skb to meet the maximum need of l2tp_udp_recv_core and
l2tp_recv_common. The minimum size of both PPP HDLC-like frame and
Ethernet frame is larger than 14 bytes, so we are safe to do so.
Also remove L2TP_HDR_SIZE_NOSEQ, it is unused now.
Fixes: fd558d186df2 ("l2tp: Split pppol2tp patch into separate l2tp and ppp parts")
Suggested-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Jacob Wen <jian.w.wen@oracle.com>
Acked-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
WCN3990 is SNOC, not PCI. This prevents probing WCN3990.
Fixes: 367c899f622c ("ath10k: add bus type check in ath10k_init_hw_params")
Signed-off-by: Brian Norris <briannorris@chromium.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
|
|
Report correct eeprom per channel power value.
Fix chan_vs_power map in mt76x0_get_power_info routine
Fixes: f2a2e819d672 ("mt76x0: remove eeprom dependency from mt76x0_get_power_info")
Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
|
|
The are a couple of statments that are one level too deep, fix this by
removing tabs.
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Stefan Schmidt <stefan@datenfreihafen.org>
|
|
The load_microcode_amd() function searches for microcode patches and
attempts to apply a microcode patch if it is of different level than the
currently installed level.
While the processor won't actually load a level that is less than
what is already installed, the logic wrongly returns UCODE_NEW thus
signaling to its caller reload_store() that a late loading should be
attempted.
If the file-system contains an older microcode revision than what is
currently running, such a late microcode reload can result in these
misleading messages:
x86/CPU: CPU features have changed after loading microcode, but might not take effect.
x86/CPU: Please consider either early loading through initrd/built-in or a potential BIOS update.
These messages were issued on a system where SME/SEV are not
enabled by the BIOS (MSR C001_0010[23] = 0b) because during boot,
early_detect_mem_encrypt() is called and cleared the SME and SEV
features in this case.
However, after the wrong late load attempt, get_cpu_cap() is called and
reloads the SME and SEV feature bits, resulting in the messages.
Update the microcode level check to not attempt microcode loading if the
current level is greater than(!) and not only equal to the current patch
level.
[ bp: massage commit message. ]
Fixes: 2613f36ed965 ("x86/microcode: Attempt late loading only when new microcode is present")
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: x86-ml <x86@kernel.org>
Link: https://lkml.kernel.org/r/154894518427.9406.8246222496874202773.stgit@tlendack-t1.amdoffice.net
|
|
There's an issue with how sense requests are handled in IDE. If ide-cd
encounters an error, it queues a sense request. With how IDE request
handling is done, this is the next request we need to handle. But it's
impossible to guarantee this, as another request could come in between
the sense being queued, and ->queue_rq() being run and handling it. If
that request ALSO fails, then we attempt to doubly queue the single
sense request we have.
Since we only support one active request at the time, defer request
processing when a sense request is queued.
Fixes: 600335205b8d "ide: convert to blk-mq"
Reported-by: He Zhe <zhe.he@windriver.com>
Tested-by: He Zhe <zhe.he@windriver.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
After finding a "firmware" dt node stratix10 tries to match it's
compatible string with it. To do so it's calling of_find_matching_node()
which already takes care of decreasing the refcount on the "firmware"
node. We are then incorrectly decreasing the refcount on that node
again.
This patch removes the unwarranted call to of_node_put().
Fixes: e7eef1d7633a ("fpga: add intel stratix10 soc fpga manager driver")
Signed-off-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
[atull: remove unnecessary braces]
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
If debugfs were to return a non-NULL error for a debugfs call, using
that pointer later in debugfs_create_files() would crash.
Fix that by properly checking the pointer before referencing it.
Reported-by: Michal Hocko <mhocko@kernel.org>
Reported-and-tested-by: syzbot+b382ba6a802a3d242790@syzkaller.appspotmail.com
Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
To 2.17
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
The request buffers are freed right before copying the pointers.
Use the func args instead which are identical and still valid.
Simple reproducer (requires KASAN enabled) on a cifs mount:
echo foo > foo ; tail -f foo & rm foo
Cc: <stable@vger.kernel.org> # 4.20
Fixes: 179e44d49c2f ("smb3: add tracepoint for sending lease break responses to server")
Signed-off-by: Aurelien Aptel <aaptel@suse.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Reviewed-by: Paulo Alcantara <palcantara@suse.de>
|
|
If create_buf_file() returns an error, don't try to reference it later
as a valid dentry pointer.
This problem was exposed when debugfs started to return errors instead
of just NULL for some calls when they do not succeed properly.
Also, the check for WARN_ON(dentry) was just wrong :)
Reported-by: Kees Cook <keescook@chromium.org>
Reported-and-tested-by: syzbot+16c3a70e1e9b29346c43@syzkaller.appspotmail.com
Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: David Rientjes <rientjes@google.com>
Fixes: ff9fb72bc077 ("debugfs: return error values, not NULL")
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
According to ARM IHI 0069C (ID070116), we should use GITS_TYPER's
bits [7:4] as ITT_entry_size instead of [8:4]. Although this is
pretty annoying, it only results in a potential over-allocation
of memory, and nothing bad happens.
Fixes: 3dfa576bfb45 ("irqchip/gic-v3-its: Add probing for VLPI properties")
Signed-off-by: Zenghui Yu <yuzenghui@huawei.com>
[maz: massaged subject and commit message]
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
On the System76 Darter Pro (darp5), there is a headset microphone
input attached to 0x1a that does not have a jack detect. In order to
get it working, the pin configuration needs to be set correctly, and
the ALC269_FIXUP_HEADSET_MODE_NO_HP_MIC fixup needs to be applied.
This is similar to the MIC_NO_PRESENCE fixups for some Dell laptops,
except we have a separate microphone jack that is already configured
correctly.
Signed-off-by: Jeremy Soller <jeremy@system76.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
|
|
The wrong size check was performed for output formats like NV24 which
set vfmt->buffers to 1, but vfmt->planes is 2. It was incorrectly
checking the payload size for plane 1, which doesn't exist.
Note: vfmt->buffers refers to the number of per-plane-buffers that
should be allocated. vfmt->planes refers to the number of planes
that make up an image. vfmt->planes may be > vfmt->buffers.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
The buf_out_validate is required for output queues in combination
with requests. Check this.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
Fix kernel-doc warnings in drivers/mtd/nand/raw:
../drivers/mtd/nand/raw/nand_base.c:420: warning: Function parameter or member 'chip' not described in 'nand_fill_oob'
../drivers/mtd/nand/raw/nand_bbt.c:173: warning: Function parameter or member 'this' not described in 'read_bbt'
../drivers/mtd/nand/raw/nand_bbt.c:173: warning: Excess function parameter 'chip' description in 'read_bbt'
Fixes: 0813621ba898a ("mtd: rawnand: Stop passing mtd_info objects to internal functions")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Boris Brezillon <boris.brezillon@bootlin.com>
Cc: Miquel Raynal <miquel.raynal@bootlin.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: linux-mtd@lists.infradead.org
Acked-by: Miquel Raynal <miquel.raynal@bootlin.com>
Signed-off-by: Boris Brezillon <bbrezillon@kernel.org>
|
|
Validate the field for an output buffer. This ensures that the
field is validated when the buffer is queued to a request, and
not when the request itself is queued, which is too late.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
The manufacturer specific initialization has already been done when
block unlocking takes place, and if anything goes wrong during this
procedure we should call spinand_manufacturer_cleanup().
Fixes: 7529df465248 ("mtd: nand: Add core infrastructure to support SPI NANDs")
Cc: <stable@vger.kernel.org>
Signed-off-by: Boris Brezillon <bbrezillon@kernel.org>
Acked-by: Miquel Raynal <miquel.raynal@bootlin.com>
|
|
Looks like PROGRAM LOAD (AKA write cache) does not necessarily reset
the cache content to 0xFF (depends on vendor implementation), so we
must fill the page cache entirely even if we only want to program the
data portion of the page, otherwise we might corrupt the BBM or user
data previously programmed in OOB area.
Fixes: 7529df465248 ("mtd: nand: Add core infrastructure to support SPI NANDs")
Reported-by: Stefan Roese <sr@denx.de>
Cc: <stable@vger.kernel.org>
Signed-off-by: Boris Brezillon <bbrezillon@kernel.org>
Tested-by: Stefan Roese <sr@denx.de>
Reviewed-by: Stefan Roese <sr@denx.de>
Acked-by: Miquel Raynal <miquel.raynal@bootlin.com>
|
|
Validate the field for an output buffer. This ensures that the
field is validated when the buffer is queued to a request, and
not when the request itself is queued, which is too late.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
Validate the field for an output buffer. This ensures that the
field is validated when the buffer is queued to a request, and
not when the request itself is queued, which is too late.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
When queueing a buffer to a request the 'field' value is not validated.
That field is only validated when the _buf_prepare() is called,
which happens when the request is queued.
However, this validation should happen at QBUF time, since you want
to know about this as soon as possible. Also, the spec requires that
the 'field' value is validated at QBUF time.
This patch adds a new buf_out_validate callback to validate the
output buffer at buf_prepare time or when QBUF queues an unprepared
buffer to a request. This callback is mandatory for output queues
that support requests.
This issue was found by v4l2-compliance since it failed to replace
V4L2_FIELD_ANY by a proper field value when testing the vivid video
output in combination with requests.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
Access to reference frames that were imported from dma-buf was taken
care of and is no longer a pending item on the driver's TODO list.
Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
When using vb2 for video decoding, dequeued capture buffers may still
be accessed by the hardware: this is the case when they are used as
reference frames for decoding subsequent frames.
When the buffer is imported with dma-buf, it needs to be mapped before
access. Until now, it was mapped when queuing and unmapped when
dequeuing, which doesn't work for access as a reference frames.
One way to solve this would be to map the buffer again when it is
needed as a reference, but the mapping/unmapping operations can
seriously impact performance. As a result, map the buffer once (when it
is first needed when queued) and keep it mapped until it is freed.
Reviewed-on: https://chromium-review.googlesource.com/334103
[Paul: Updated for mainline and changed commit message]
Signed-off-by: Pawel Osciak <posciak@chromium.org>
Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
This reverts commit cf20ae1535eb690a87c29b9cc7af51881384e967.
The vb2_find_timestamp helper was modified to allow finding buffers
regardless of their current state in the queue. This means that we
no longer have to take particular care of references to the current
capture buffer.
Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
There really is no reason why vb2_find_timestamp can't just find
buffers in any state. Drop that part of the test.
This also means that vb->timestamp should only be set to 0 when
the driver doesn't copy timestamps.
This change allows for more efficient pipelining (i.e. you can use
a buffer for a reference frame even when it is queued).
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Reviewed-by: Tomasz Figa <tfiga@chromium.org>
Reviewed-by: Alexandre Courbot <acourbot@chromium.org>
Reviewed-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
Vdec: Using standard CCF interface to set parent clock and
clock rate in dtsi and using common interface to open/close
video decoder clock.
Venc: Using standard CCF interface to set parent clock in dtsi
and using common interface to open/close video encoder clock.
Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Signed-off-by: Qianqian Yan <qianqian.yan@mediatek.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
Fix MTK binding document for MT8173 dtsi changed in order
to use standard CCF interface.
MT8173 SoC from Mediatek.
Signed-off-by: Yunfei Dong <yunfei.dong@mediatek.com>
Signed-off-by: Qianqian Yan <qianqian.yan@mediatek.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
Upstream must be stopped immediately after receiving the last EOF and
before disabling the IDMA channel. This can be accomplished by moving
upstream stream off to just after receiving the last EOF completion in
prp_stop(). For symmetry also move upstream stream on to end of
prp_start().
This fixes a complete system hard lockup on the SabreAuto when streaming
from the ADV7180, by repeatedly sending a stream off immediately followed
by stream on:
while true; do v4l2-ctl -d1 --stream-mmap --stream-count=3; done
Eventually this either causes the system lockup or EOF timeouts at all
subsequent stream on, until a system reset.
The lockup occurs when disabling the IDMA channel at stream off. Stopping
the video data stream entering the IDMA channel before disabling the
channel itself appears to be a reliable fix for the hard lockup.
Fixes: f0d9c8924e2c3 ("[media] media: imx: Add IC subdev drivers")
Reported-by: Gaël PORTAY <gael.portay@collabora.com>
Tested-by: Gaël PORTAY <gael.portay@collabora.com>
Signed-off-by: Steve Longerbeam <slongerbeam@gmail.com>
Cc: stable@vger.kernel.org # for 4.13 and up
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
Move upstream stream off to just after receiving the last EOF completion
and disabling the CSI (and thus before disabling the IDMA channel) in
csi_stop(). For symmetry also move upstream stream on to beginning of
csi_start().
Doing this makes csi_s_stream() more symmetric with prp_s_stream() which
will require the same change to fix a hard lockup.
Signed-off-by: Steve Longerbeam <slongerbeam@gmail.com>
Cc: stable@vger.kernel.org # for 4.13 and up
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
Disable the CSI immediately after receiving the last EOF before stream
off (and thus before disabling the IDMA channel). Do this by moving the
wait for EOF completion into a new function csi_idmac_wait_last_eof().
This fixes a complete system hard lockup on the SabreAuto when streaming
from the ADV7180, by repeatedly sending a stream off immediately followed
by stream on:
while true; do v4l2-ctl -d4 --stream-mmap --stream-count=3; done
Eventually this either causes the system lockup or EOF timeouts at all
subsequent stream on, until a system reset.
The lockup occurs when disabling the IDMA channel at stream off. Disabling
the CSI before disabling the IDMA channel appears to be a reliable fix for
the hard lockup.
Fixes: 4a34ec8e470cb ("[media] media: imx: Add CSI subdev driver")
Reported-by: Gaël PORTAY <gael.portay@collabora.com>
Signed-off-by: Steve Longerbeam <slongerbeam@gmail.com>
Cc: stable@vger.kernel.org # for 4.13 and up
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
|
|
With support for split pmd lock, we use pmd page pmd_huge_pte pointer
to store the deposited page table. In those config when we move page
tables we need to make sure we move the deposited page table to the
correct pmd page. Otherwise this can result in crash when we withdraw
of deposited page table because we can find the pmd_huge_pte NULL.
eg:
__split_huge_pmd+0x1070/0x1940
__split_huge_pmd+0xe34/0x1940 (unreliable)
vma_adjust_trans_huge+0x110/0x1c0
__vma_adjust+0x2b4/0x9b0
__split_vma+0x1b8/0x280
__do_munmap+0x13c/0x550
sys_mremap+0x220/0x7e0
system_call+0x5c/0x70
Fixes: 675d995297d4 ("powerpc/book3s64: Enable split pmd ptlock.")
Cc: stable@vger.kernel.org # v4.18+
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
|
|
When Sink negotiates PPS, the voltage range of selected PPS APDO might
not cover the previous voltage (out_volt). If the previous out_volt is
lower than the new min_volt, the output voltage in RDO might be set to
an invalid value. For instance, supposed that the previous voltage is
5V, and the new voltage range in the APDO is 7V-12V. Then the output
voltage in the RDO should not be set to 5V which is lower than the
possible min_volt 7V.
Fix this by choosing the maximal value between the previous voltage and
the new min_volt first. And ensure that this value will not exceed the
new max_volt. The new out_volt will fall within the new voltage range
while being the closest value compared to the previous out_volt.
Signed-off-by: Kyle Tso <kyletso@google.com>
Reviewed-by: Adam Thomson <Adam.Thomson.Opensource@diasemi.com>
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Fixes: c710d0bb76ff0 ("usb: typec: tcpm: Extend the matching rules on PPS APDO selection")
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb into usb-linus
Felipe writes:
usb: fixes for v5.0-rc4
Dwc3 got a fix for cases when gadget driver queue an OUT request of
length 0; this is a case that has been overlooked for quite some time
now.
Exynos' dwc3 glue layer got a fix on the error path for those cases
where clk_prepare_enable() fails.
TI's AM335x PHY driver got a fix for a race condition during
probe. This race happened because driver was powering off the PHY only
after adding the PHY handle to the framework. The result is that we
could fall into a situation where user of the PHY (MUSB) could call
phy_init() before phy driver's probe() called phy_poweroff() which
would result in a powered off PHY after phy_init() was called.
The old net2272 driver got a fix for an erroneous use of bitwise
negation.
* tag 'fixes-for-v5.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb:
usb: phy: am335x: fix race condition in _probe
usb: dwc3: exynos: Fix error handling of clk_prepare_enable
usb: phy: fix link errors
usb: gadget: udc: net2272: Fix bitwise and boolean operations
usb: dwc3: gadget: Handle 0 xfer length for OUT EP
|
|
Jose Abreu says:
====================
net: stmmac: Misc fixes
Some misc fixes for stmmac targeting -net.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
In stmmac xmit callback we use a different flow for TSO packets but TSO
xmit callback is not disabling the EEE mode.
Fix this by disabling earlier the EEE mode, i.e. before calling the TSO
xmit callback.
Signed-off-by: Jose Abreu <joabreu@synopsys.com>
Cc: Joao Pinto <jpinto@synopsys.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Cc: Alexandre Torgue <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The number of TSO enabled channels in HW can be different than the
number of total channels. There is no way to determined, at runtime, the
number of TSO capable channels and its safe to assume that if TSO is
enabled then at least channel 0 will be TSO capable.
Lets always send TSO packets from Queue 0.
Signed-off-by: Jose Abreu <joabreu@synopsys.com>
Cc: Joao Pinto <jpinto@synopsys.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Cc: Alexandre Torgue <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
If we don't have DT then stmmac_clk will not be available. Let's add a
new Platform Data field so that we can specify the refclk by this mean.
This way we can still use the coalesce command in PCI based setups.
Signed-off-by: Jose Abreu <joabreu@synopsys.com>
Cc: Joao Pinto <jpinto@synopsys.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Cc: Alexandre Torgue <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
While implementing ipvlan l3 and l3s mode for kubernetes CNI plugin,
I ran into the issue that while l3 mode is working fine, l3s mode
does not have any connectivity to kube-apiserver and hence all pods
end up in Error state as well. The ipvlan master device sits on
top of a bond device and hostns traffic to kube-apiserver (also running
in hostns) is DNATed from 10.152.183.1:443 to 139.178.29.207:37573
where the latter is the address of the bond0. While in l3 mode, a
curl to https://10.152.183.1:443 or to https://139.178.29.207:37573
works fine from hostns, neither of them do in case of l3s. In the
latter only a curl to https://127.0.0.1:37573 appeared to work where
for local addresses of bond0 I saw kernel suddenly starting to emit
ARP requests to query HW address of bond0 which remained unanswered
and neighbor entries in INCOMPLETE state. These ARP requests only
happen while in l3s.
Debugging this further, I found the issue is that l3s mode is piggy-
backing on l3 master device, and in this case local routes are using
l3mdev_master_dev_rcu(dev) instead of net->loopback_dev as per commit
f5a0aab84b74 ("net: ipv4: dst for local input routes should use l3mdev
if relevant") and 5f02ce24c269 ("net: l3mdev: Allow the l3mdev to be
a loopback"). I found that reverting them back into using the
net->loopback_dev fixed ipvlan l3s connectivity and got everything
working for the CNI.
Now judging from 4fbae7d83c98 ("ipvlan: Introduce l3s mode") and the
l3mdev paper in [0] the only sole reason why ipvlan l3s is relying
on l3 master device is to get the l3mdev_ip_rcv() receive hook for
setting the dst entry of the input route without adding its own
ipvlan specific hacks into the receive path, however, any l3 domain
semantics beyond just that are breaking l3s operation. Note that
ipvlan also has the ability to dynamically switch its internal
operation from l3 to l3s for all ports via ipvlan_set_port_mode()
at runtime. In any case, l3 vs l3s soley distinguishes itself by
'de-confusing' netfilter through switching skb->dev to ipvlan slave
device late in NF_INET_LOCAL_IN before handing the skb to L4.
Minimal fix taken here is to add a IFF_L3MDEV_RX_HANDLER flag which,
if set from ipvlan setup, gets us only the wanted l3mdev_l3_rcv() hook
without any additional l3mdev semantics on top. This should also have
minimal impact since dev->priv_flags is already hot in cache. With
this set, l3s mode is working fine and I also get things like
masquerading pod traffic on the ipvlan master properly working.
[0] https://netdevconf.org/1.2/papers/ahern-what-is-l3mdev-paper.pdf
Fixes: f5a0aab84b74 ("net: ipv4: dst for local input routes should use l3mdev if relevant")
Fixes: 5f02ce24c269 ("net: l3mdev: Allow the l3mdev to be a loopback")
Fixes: 4fbae7d83c98 ("ipvlan: Introduce l3s mode")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Mahesh Bandewar <maheshb@google.com>
Cc: David Ahern <dsa@cumulusnetworks.com>
Cc: Florian Westphal <fw@strlen.de>
Cc: Martynas Pumputis <m@lambda.lt>
Acked-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Use pskb_may_pull() to make sure the optional fields are in skb linear
parts, so we can safely read them later.
It's easy to reproduce the issue with a net driver that supports paged
skb data. Just create a L2TPv3 over IP tunnel and then generates some
network traffic.
Once reproduced, rx err in /sys/kernel/debug/l2tp/tunnels will increase.
Changes in v4:
1. s/l2tp_v3_pull_opt/l2tp_v3_ensure_opt_in_linear/
2. s/tunnel->version != L2TP_HDR_VER_2/tunnel->version == L2TP_HDR_VER_3/
3. Add 'Fixes' in commit messages.
Changes in v3:
1. To keep consistency, move the code out of l2tp_recv_common.
2. Use "net" instead of "net-next", since this is a bug fix.
Changes in v2:
1. Only fix L2TPv3 to make code simple.
To fix both L2TPv3 and L2TPv2, we'd better refactor l2tp_recv_common.
It's complicated to do so.
2. Reloading pointers after pskb_may_pull
Fixes: f7faffa3ff8e ("l2tp: Add L2TPv3 protocol support")
Fixes: 0d76751fad77 ("l2tp: Add L2TPv3 IP encapsulation (no UDP) support")
Fixes: a32e0eec7042 ("l2tp: introduce L2TPv3 IP encapsulation support for IPv6")
Signed-off-by: Jacob Wen <jian.w.wen@oracle.com>
Acked-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Call tun_set_real_num_queues() after the increment of tun->numqueues
since the former depends on it. Otherwise, the number of queues is not
correctly accounted for, which results to warnings similar to:
"vnet0 selects TX queue 11, but real number of TX queues is 11".
Fixes: 0b7959b62573 ("tun: publish tfile after it's fully initialized")
Reported-and-tested-by: George Amanakis <gamanakis@gmail.com>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
skb->cb may contain data from previous layers (in an observed case
IPv4 with L3 Master Device). In the observed scenario, the data in
IPCB(skb)->frags was misinterpreted as IP6CB(skb)->frag_max_size,
eventually caused an unexpected IPv6 fragmentation in ip6_fragment()
through ip6_finish_output().
This patch clears IP6CB(skb), which potentially contains garbage data,
on the SRH ip4ip6 encapsulation.
Fixes: 32d99d0b6702 ("ipv6: sr: add support for ip4ip6 encapsulation")
Signed-off-by: Yohei Kanemaru <yohei.kanemaru@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Toshiaki Makita says:
====================
virtio_net: Fix problems around XDP tx and napi_tx
While I'm looking into how to account standard tx counters on XDP tx
processing, I found several bugs around XDP tx and napi_tx.
Patch1: Fix oops on error path. Patch2 depends on this.
Patch2: Fix memory corruption on freeing xdp_frames with napi_tx enabled.
Patch3: Minor fix patch5 depends on.
Patch4: Fix memory corruption on processing xdp_frames when XDP is disabled.
Also patch5 depends on this.
Patch5: Fix memory corruption on processing xdp_frames while XDP is being
disabled.
Patch6: Minor fix patch7 depends on.
Patch7: Fix memory corruption on freeing sk_buff or xdp_frames when a normal
queue is reused for XDP and vise versa.
v2:
- patch5: Make rcu_assign_pointer/synchronize_net conditional instead of
_virtnet_set_queues.
- patch7: Use napi_consume_skb() instead of dev_consume_skb_any()
====================
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
We do not reset or free up unused buffers when enabling/disabling XDP,
so it can happen that xdp_frames are freed after disabling XDP or
sk_buffs are freed after enabling XDP on xdp tx queues.
Thus we need to handle both forms (xdp_frames and sk_buffs) regardless
of XDP setting.
One way to trigger this problem is to disable XDP when napi_tx is
enabled. In that case, virtnet_xdp_set() calls virtnet_napi_enable()
which kicks NAPI. The NAPI handler will call virtnet_poll_cleantx()
which invokes free_old_xmit_skbs() for queues which have been used by
XDP.
Note that even with this change we need to keep skipping
free_old_xmit_skbs() from NAPI handlers when XDP is enabled, because XDP
tx queues do not aquire queue locks.
- v2: Use napi_consume_skb() instead of dev_consume_skb_any()
Fixes: 4941d472bf95 ("virtio-net: do not reset during XDP set")
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
put_page() can work as a fallback for freeing xdp_frames, but the
appropriate way is to use xdp_return_frame().
Fixes: cac320c850ef ("virtio_net: convert to use generic xdp_frame and xdp_return_frame API")
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Commit 8dcc5b0ab0ec ("virtio_net: fix ndo_xdp_xmit crash towards dev not
ready for XDP") tried to avoid access to unexpected sq while XDP is
disabled, but was not complete.
There was a small window which causes out of bounds sq access in
virtnet_xdp_xmit() while disabling XDP.
An example case of
- curr_queue_pairs = 6 (2 for SKB and 4 for XDP)
- online_cpu_num = xdp_queue_paris = 4
when XDP is enabled:
CPU 0 CPU 1
(Disabling XDP) (Processing redirected XDP frames)
virtnet_xdp_xmit()
virtnet_xdp_set()
_virtnet_set_queues()
set curr_queue_pairs (2)
check if rq->xdp_prog is not NULL
virtnet_xdp_sq(vi)
qp = curr_queue_pairs -
xdp_queue_pairs +
smp_processor_id()
= 2 - 4 + 1 = -1
sq = &vi->sq[qp] // out of bounds access
set xdp_queue_pairs (0)
rq->xdp_prog = NULL
Basically we should not change curr_queue_pairs and xdp_queue_pairs
while someone can read the values. Thus, when disabling XDP, assign NULL
to rq->xdp_prog first, and wait for RCU grace period, then change
xxx_queue_pairs.
Note that we need to keep the current order when enabling XDP though.
- v2: Make rcu_assign_pointer/synchronize_net conditional instead of
_virtnet_set_queues.
Fixes: 186b3c998c50 ("virtio-net: support XDP_REDIRECT")
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
When XDP is disabled, curr_queue_pairs + smp_processor_id() can be
larger than max_queue_pairs.
There is no guarantee that we have enough XDP send queues dedicated for
each cpu when XDP is disabled, so do not count drops on sq in that case.
Fixes: 5b8f3c8d30a6 ("virtio_net: Add XDP related stats")
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
When _virtnet_set_queues() failed we did not restore real_num_rx_queues.
Fix this by placing the change of real_num_rx_queues after
_virtnet_set_queues().
This order is also in line with virtnet_set_channels().
Fixes: 4941d472bf95 ("virtio-net: do not reset during XDP set")
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
When napi_tx is enabled, virtnet_poll_cleantx() called
free_old_xmit_skbs() even for xdp send queue.
This is bogus since the queue has xdp_frames, not sk_buffs, thus mangled
device tx bytes counters because skb->len is meaningless value, and even
triggered oops due to general protection fault on freeing them.
Since xdp send queues do not aquire locks, old xdp_frames should be
freed only in virtnet_xdp_xmit(), so just skip free_old_xmit_skbs() for
xdp send queues.
Similarly virtnet_poll_tx() called free_old_xmit_skbs(). This NAPI
handler is called even without calling start_xmit() because cb for tx is
by default enabled. Once the handler is called, it enabled the cb again,
and then the handler would be called again. We don't need this handler
for XDP, so don't enable cb as well as not calling free_old_xmit_skbs().
Also, we need to disable tx NAPI when disabling XDP, so
virtnet_poll_tx() can safely access curr_queue_pairs and
xdp_queue_pairs, which are not atomically updated while disabling XDP.
Fixes: b92f1e6751a6 ("virtio-net: transmit napi")
Fixes: 7b0411ef4aa6 ("virtio-net: clean tx descriptors from rx napi")
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Commit 4e09ff536284 ("virtio-net: disable NAPI only when enabled during
XDP set") tried to fix inappropriate NAPI enabling/disabling when
!netif_running(), but was not complete.
On error path virtio_net could enable NAPI even when !netif_running().
This can cause enabling NAPI twice on virtnet_open(), which would
trigger BUG_ON() in napi_enable().
Fixes: 4941d472bf95b ("virtio-net: do not reset during XDP set")
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
