David E. Box [Wed, 20 Apr 2022 15:56:21 +0000 (08:56 -0700)]
platform/x86/intel/sdsi: Poll on ready bit for writes
Due to change in firmware flow, update mailbox writes to poll on ready bit
instead of run_busy bit. This change makes the polling method consistent
for both writes and reads, which also uses the ready bit.
David E. Box [Wed, 20 Apr 2022 15:56:20 +0000 (08:56 -0700)]
platform/x86/intel/sdsi: Handle leaky bucket
To prevent an agent from indefinitely holding the mailbox firmware has
implemented a leaky bucket algorithm. Repeated access to the mailbox may
now incur a delay of up to 2.1 seconds. Add a retry loop that tries for
up to 2.5 seconds to acquire the mailbox.
platform/x86: intel-uncore-freq: Prevent driver loading in guests
Loading this driver in guests results in unchecked MSR access error for
MSR 0x620.
There is no use of reading and modifying package/die scope uncore MSRs
in guests. So check for CPU feature X86_FEATURE_HYPERVISOR to prevent
loading of this driver in guests.
platform/x86: dell-laptop: Add quirk entry for Latitude 7520
The Latitude 7520 supports AC timeouts, but it has no KBD_LED_AC_TOKEN
and so changes to stop_timeout appear to have no effect if the laptop
is plugged in.
Hans de Goede [Wed, 27 Apr 2022 11:49:56 +0000 (13:49 +0200)]
platform/x86: asus-wmi: Fix driver not binding when fan curve control probe fails
Before this commit fan_curve_check_present() was trying to not cause
the probe to fail on devices without fan curve control by testing for
known error codes returned by asus_wmi_evaluate_method_buf().
Checking for ENODATA or ENODEV, with the latter being returned by this
function when an ACPI integer with a value of ASUS_WMI_UNSUPPORTED_METHOD
is returned. But for other ACPI integer returns this function just returns
them as is, including the ASUS_WMI_DSTS_UNKNOWN_BIT value of 2.
On the Asus U36SD ASUS_WMI_DSTS_UNKNOWN_BIT gets returned, leading to:
asus-nb-wmi: probe of asus-nb-wmi failed with error 2
Instead of playing whack a mole with error codes here, simply treat all
errors as there not being any fan curves, fixing the driver no longer
loading on the Asus U36SD laptop.
iocost: don't reset the inuse weight of under-weighted debtors
When an iocg is in debt, its inuse weight is owned by debt handling and
should stay at 1. This invariant was broken when determining the amount of
surpluses at the beginning of donation calculation - when an iocg's
hierarchical weight is too low, the iocg is excluded from donation
calculation and its inuse is reset to its active regardless of its
indebtedness, triggering warnings like the following:
As this happens only when an iocg's hierarchical weight is negligible, its
impact likely is limited to triggering the warnings. Fix it by skipping
resetting inuse of under-weighted debtors.
`nf_flowtable_udp_timeout` sysctl option is available only
if CONFIG_NFT_FLOW_OFFLOAD enabled. But infra for this flow
offload UDP timeout was added under CONFIG_NF_FLOW_TABLE
config option. So, if you have CONFIG_NFT_FLOW_OFFLOAD
disabled and CONFIG_NF_FLOW_TABLE enabled, the
`nf_flowtable_udp_timeout` is not present in sysfs.
Please note, that TCP flow offload timeout sysctl option
is present even CONFIG_NFT_FLOW_OFFLOAD is disabled.
I suppose it was a typo in commit that adds UDP flow offload
timeout and CONFIG_NF_FLOW_TABLE should be used instead.
netfilter: nf_conntrack_tcp: re-init for syn packets only
Jaco Kroon reported tcp problems that Eric Dumazet and Neal Cardwell
pinpointed to nf_conntrack tcp_in_window() bug.
tcp trace shows following sequence:
I > R Flags [S], seq 3451342529, win 62580, options [.. tfo [|tcp]>
R > I Flags [S.], seq 2699962254, ack 3451342530, win 65535, options [..]
R > I Flags [P.], seq 1:89, ack 1, [..]
Note 3rd ACK is from responder to initiator so following branch is taken:
} else if (((state->state == TCP_CONNTRACK_SYN_SENT
&& dir == IP_CT_DIR_ORIGINAL)
|| (state->state == TCP_CONNTRACK_SYN_RECV
&& dir == IP_CT_DIR_REPLY))
&& after(end, sender->td_end)) {
... because state == TCP_CONNTRACK_SYN_RECV and dir is REPLY.
This causes the scaling factor to be reset to 0: window scale option
is only present in syn(ack) packets. This in turn makes nf_conntrack
mark valid packets as out-of-window.
This was always broken, it exists even in original commit where
window tracking was added to ip_conntrack (nf_conntrack predecessor)
in 2.6.9-rc1 kernel.
Restrict to 'tcph->syn', just like the 3rd condtional added in
commit 82b72cb94666 ("netfilter: conntrack: re-init state for retransmitted syn-ack").
Upon closer look, those conditionals/branches can be merged:
Because earlier checks prevent syn-ack from showing up in
original direction, the 'dir' checks in the conditional quoted above are
redundant, remove them. Return early for pure syn retransmitted in reply
direction (simultaneous open).
David S. Miller [Wed, 27 Apr 2022 09:58:39 +0000 (10:58 +0100)]
Merge branch '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net
-queue
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2022-04-26
This series contains updates to ice driver only.
Ivan Vecera removes races related to VF message processing by changing
mutex_trylock() call to mutex_lock() and moving additional operations
to occur under mutex.
Petr Oros increases wait time after firmware flash as current time is
not sufficient.
Jake resolves a use-after-free issue for mailbox snapshot.
====================
net: dsa: lantiq_gswip: Don't set GSWIP_MII_CFG_RMII_CLK
Commit 4b5923249b8fa4 ("net: dsa: lantiq_gswip: Configure all remaining
GSWIP_MII_CFG bits") added all known bits in the GSWIP_MII_CFGp
register. It helped bring this register into a well-defined state so the
driver has to rely less on the bootloader to do things right.
Unfortunately it also sets the GSWIP_MII_CFG_RMII_CLK bit without any
possibility to configure it. Upon further testing it turns out that all
boards which are supported by the GSWIP driver in OpenWrt which use an
RMII PHY have a dedicated oscillator on the board which provides the
50MHz RMII reference clock.
Don't set the GSWIP_MII_CFG_RMII_CLK bit (but keep the code which always
clears it) to fix support for the Fritz!Box 7362 SL in OpenWrt. This is
a board with two Atheros AR8030 RMII PHYs. With the "RMII clock" bit set
the MAC also generates the RMII reference clock whose signal then
conflicts with the signal from the oscillator on the board. This results
in a constant cycle of the PHY detecting link up/down (and as a result
of that: the two ports using the AR8030 PHYs are not working).
At the time of writing this patch there's no known board where the MAC
(GSWIP) has to generate the RMII reference clock. If needed this can be
implemented in future by providing a device-tree flag so the
GSWIP_MII_CFG_RMII_CLK bit can be toggled per port.
net: Use this_cpu_inc() to increment net->core_stats
The macro dev_core_stats_##FIELD##_inc() disables preemption and invokes
netdev_core_stats_alloc() to return a per-CPU pointer.
netdev_core_stats_alloc() will allocate memory on its first invocation
which breaks on PREEMPT_RT because it requires non-atomic context for
memory allocation.
This can be avoided by enabling preemption in netdev_core_stats_alloc()
assuming the caller always disables preemption.
It might be better to replace local_inc() with this_cpu_inc() now that
dev_core_stats_##FIELD##_inc() gained a preempt-disable section and does
not rely on already disabled preemption. This results in less
instructions on x86-64:
local_inc:
| incl %gs:__preempt_count(%rip) # __preempt_count
| movq 488(%rdi), %rax # _1->core_stats, _22
| testq %rax, %rax # _22
| je .L585 #,
| add %gs:this_cpu_off(%rip), %rax # this_cpu_off, tcp_ptr__
| .L586:
| testq %rax, %rax # _27
| je .L587 #,
| incq (%rax) # _6->a.counter
| .L587:
| decl %gs:__preempt_count(%rip) # __preempt_count
Merge tag 'for-5.18/fbdev-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/linux-fbdev
Pull fbdev fixes and updates from Helge Deller:
"A bunch of outstanding fbdev patches - all trivial and small"
* tag 'for-5.18/fbdev-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/linux-fbdev:
video: fbdev: clps711x-fb: Use syscon_regmap_lookup_by_phandle
video: fbdev: mmp: replace usage of found with dedicated list iterator variable
video: fbdev: sh_mobile_lcdcfb: Remove sh_mobile_lcdc_check_var() declaration
video: fbdev: i740fb: Error out if 'pixclock' equals zero
video: fbdev: i740fb: use memset_io() to clear screen
video: fbdev: s3fb: Error out if 'pixclock' equals zero
video: fbdev: arkfb: Error out if 'pixclock' equals zero
video: fbdev: tridentfb: Error out if 'pixclock' equals zero
video: fbdev: vt8623fb: Error out if 'pixclock' equals zero
video: fbdev: kyro: Error out if 'lineclock' equals zero
video: fbdev: neofb: Fix the check of 'var->pixclock'
video: fbdev: imxfb: Fix missing of_node_put in imxfb_probe
video: fbdev: omap: Make it CCF clk API compatible
video: fbdev: aty/matrox/...: Prepare cleanup of powerpc's asm/prom.h
video: fbdev: pm2fb: Fix a kernel-doc formatting issue
linux/fb.h: Spelling s/palette/palette/
video: fbdev: sis: fix potential NULL dereference in sisfb_post_sis300()
video: fbdev: pxafb: use if else instead
video: fbdev: udlfb: properly check endpoint type
video: fbdev: of: display_timing: Remove a redundant zeroing of memory
Bluetooth: hci_sync: Cleanup hci_conn if it cannot be aborted
This attempts to cleanup the hci_conn if it cannot be aborted as
otherwise it would likely result in having the controller and host
stack out of sync with respect to connection handle.
Merge tag 'for-5.18-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba:
- direct IO fixes:
- restore passing file offset to correctly calculate checksums
when repairing on read and bio split happens
- use correct bio when sumitting IO on zoned filesystem
- zoned mode fixes:
- fix selection of device to correctly calculate device
capabilities when allocating a new bio
- use a dedicated lock for exclusion during relocation
- fix leaked plug after failure syncing log
- fix assertion during scrub and relocation
* tag 'for-5.18-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: zoned: use dedicated lock for data relocation
btrfs: fix assertion failure during scrub due to block group reallocation
btrfs: fix direct I/O writes for split bios on zoned devices
btrfs: fix direct I/O read repair for split bios
btrfs: fix and document the zoned device choice in alloc_new_bio
btrfs: fix leaked plug after failure syncing log on zoned filesystems
Bluetooth: hci_event: Fix creating hci_conn object on error status
It is useless to create a hci_conn object if on error status as the
result would be it being freed in the process and anyway it is likely
the result of controller and host stack being out of sync.
Bluetooth: hci_event: Fix checking for invalid handle on error status
Commit d5ebaa7c5f6f6 introduces checks for handle range
(e.g HCI_CONN_HANDLE_MAX) but controllers like Intel AX200 don't seem
to respect the valid range int case of error status:
Because of it is impossible to cleanup the connections properly since
the stack would attempt to cancel the connection which is no longer in
progress causing the following trace:
Jacob Keller [Tue, 12 Apr 2022 17:34:22 +0000 (10:34 -0700)]
ice: fix use-after-free when deinitializing mailbox snapshot
During ice_sriov_configure, if num_vfs is 0, we are being asked by the
kernel to remove all VFs.
The driver first de-initializes the snapshot before freeing all the VFs.
This results in a use-after-free BUG detected by KASAN. The bug occurs
because the snapshot can still be accessed until all VFs are removed.
Fix this by freeing all the VFs first before calling
ice_mbx_deinit_snapshot.
[ +0.032591] ==================================================================
[ +0.000021] BUG: KASAN: use-after-free in ice_mbx_vf_state_handler+0x1c3/0x410 [ice]
[ +0.000315] Write of size 28 at addr ffff889908eb6f28 by task kworker/55:2/1530996
[ +0.000025] The buggy address belongs to the object at ffff889908eb6f00
which belongs to the cache kmalloc-96 of size 96
[ +0.000016] The buggy address is located 40 bytes inside of
96-byte region [ffff889908eb6f00, ffff889908eb6f60)
[ +0.000016] Memory state around the buggy address:
[ +0.000012] ffff889908eb6e00: fa fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
[ +0.000014] ffff889908eb6e80: fa fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
[ +0.000014] >ffff889908eb6f00: fa fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
[ +0.000011] ^
[ +0.000013] ffff889908eb6f80: fa fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
[ +0.000013] ffff889908eb7000: fa fb fb fb fb fb fb fb fc fc fc fc fa fb fb fb
[ +0.000012] ==================================================================
Petr Oros [Wed, 13 Apr 2022 15:37:45 +0000 (17:37 +0200)]
ice: wait 5 s for EMP reset after firmware flash
We need to wait 5 s for EMP reset after firmware flash. Code was extracted
from OOT driver (ice v1.8.3 downloaded from sourceforge). Without this
wait, fw_activate let card in inconsistent state and recoverable only
by second flash/activate. Flash was tested on these fw's:
From -> To
3.00 -> 3.10/3.20
3.10 -> 3.00/3.20
3.20 -> 3.00/3.10
Reproducer:
[root@host ~]# devlink dev flash pci/0000:ca:00.0 file E810_XXVDA4_FH_O_SEC_FW_1p6p1p9_NVM_3p10_PLDMoMCTP_0.11_8000AD7B.bin
Preparing to flash
[fw.mgmt] Erasing
[fw.mgmt] Erasing done
[fw.mgmt] Flashing 100%
[fw.mgmt] Flashing done 100%
[fw.undi] Erasing
[fw.undi] Erasing done
[fw.undi] Flashing 100%
[fw.undi] Flashing done 100%
[fw.netlist] Erasing
[fw.netlist] Erasing done
[fw.netlist] Flashing 100%
[fw.netlist] Flashing done 100%
Activate new firmware by devlink reload
[root@host ~]# devlink dev reload pci/0000:ca:00.0 action fw_activate
reload_actions_performed:
fw_activate
[root@host ~]# ip link show ens7f0
71: ens7f0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN mode DEFAULT group default qlen 1000
link/ether b4:96:91:dc:72:e0 brd ff:ff:ff:ff:ff:ff
altname enp202s0f0
dmesg after flash:
[ 55.120788] ice: Copyright (c) 2018, Intel Corporation.
[ 55.274734] ice 0000:ca:00.0: Get PHY capabilities failed status = -5, continuing anyway
[ 55.569797] ice 0000:ca:00.0: The DDP package was successfully loaded: ICE OS Default Package version 1.3.28.0
[ 55.603629] ice 0000:ca:00.0: Get PHY capability failed.
[ 55.608951] ice 0000:ca:00.0: ice_init_nvm_phy_type failed: -5
[ 55.647348] ice 0000:ca:00.0: PTP init successful
[ 55.675536] ice 0000:ca:00.0: DCB is enabled in the hardware, max number of TCs supported on this port are 8
[ 55.685365] ice 0000:ca:00.0: FW LLDP is disabled, DCBx/LLDP in SW mode.
[ 55.692179] ice 0000:ca:00.0: Commit DCB Configuration to the hardware
[ 55.701382] ice 0000:ca:00.0: 126.024 Gb/s available PCIe bandwidth, limited by 16.0 GT/s PCIe x8 link at 0000:c9:02.0 (capable of 252.048 Gb/s with 16.0 GT/s PCIe x16 link)
Reboot doesn’t help, only second flash/activate with OOT or patched
driver put card back in consistent state.
After patch:
[root@host ~]# devlink dev flash pci/0000:ca:00.0 file E810_XXVDA4_FH_O_SEC_FW_1p6p1p9_NVM_3p10_PLDMoMCTP_0.11_8000AD7B.bin
Preparing to flash
[fw.mgmt] Erasing
[fw.mgmt] Erasing done
[fw.mgmt] Flashing 100%
[fw.mgmt] Flashing done 100%
[fw.undi] Erasing
[fw.undi] Erasing done
[fw.undi] Flashing 100%
[fw.undi] Flashing done 100%
[fw.netlist] Erasing
[fw.netlist] Erasing done
[fw.netlist] Flashing 100%
[fw.netlist] Flashing done 100%
Activate new firmware by devlink reload
[root@host ~]# devlink dev reload pci/0000:ca:00.0 action fw_activate
reload_actions_performed:
fw_activate
[root@host ~]# ip link show ens7f0
19: ens7f0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000
link/ether b4:96:91:dc:72:e0 brd ff:ff:ff:ff:ff:ff
altname enp202s0f0
Fixes: 399e27dbbd9e94 ("ice: support immediate firmware activation via devlink reload") Signed-off-by: Petr Oros <[email protected]> Tested-by: Gurucharan <[email protected]> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <[email protected]>
Ivan Vecera [Tue, 19 Apr 2022 14:22:21 +0000 (16:22 +0200)]
ice: Protect vf_state check by cfg_lock in ice_vc_process_vf_msg()
Previous patch labelled "ice: Fix incorrect locking in
ice_vc_process_vf_msg()" fixed an issue with ignored messages
sent by VF driver but a small race window still left.
Recently caught trace during 'ip link set ... vf 0 vlan ...' operation:
[ 7332.995625] ice 0000:3b:00.0: Clearing port VLAN on VF 0
[ 7333.001023] iavf 0000:3b:01.0: Reset indication received from the PF
[ 7333.007391] iavf 0000:3b:01.0: Scheduling reset task
[ 7333.059575] iavf 0000:3b:01.0: PF returned error -5 (IAVF_ERR_PARAM) to our request 3
[ 7333.059626] ice 0000:3b:00.0: Invalid message from VF 0, opcode 3, len 4, error -1
Setting of VLAN for VF causes a reset of the affected VF using
ice_reset_vf() function that runs with cfg_lock taken:
1. ice_notify_vf_reset() informs IAVF driver that reset is needed and
IAVF schedules its own reset procedure
2. Bit ICE_VF_STATE_DIS is set in vf->vf_state
3. Misc initialization steps
4. ice_sriov_post_vsi_rebuild() -> ice_vf_set_initialized() and that
clears ICE_VF_STATE_DIS in vf->vf_state
Step 3 is mentioned race window because IAVF reset procedure runs in
parallel and one of its step is sending of VIRTCHNL_OP_GET_VF_RESOURCES
message (opcode==3). This message is handled in ice_vc_process_vf_msg()
and if it is received during the mentioned race window then it's
marked as invalid and error is returned to VF driver.
Protect vf_state check in ice_vc_process_vf_msg() by cfg_lock to avoid
this race condition.
Ivan Vecera [Fri, 1 Apr 2022 10:40:52 +0000 (12:40 +0200)]
ice: Fix incorrect locking in ice_vc_process_vf_msg()
Usage of mutex_trylock() in ice_vc_process_vf_msg() is incorrect
because message sent from VF is ignored and never processed.
Use mutex_lock() instead to fix the issue. It is safe because this
mutex is used to prevent races between VF related NDOs and
handlers processing request messages from VF and these handlers
are running in ice_service_task() context. Additionally move this
mutex lock prior ice_vc_is_opcode_allowed() call to avoid potential
races during allowlist access.
Fixes: e6ba5273d4ed ("ice: Fix race conditions between virtchnl handling and VF ndo ops") Signed-off-by: Ivan Vecera <[email protected]> Tested-by: Konrad Jankowski <[email protected]> Signed-off-by: Tony Nguyen <[email protected]>
RISC-V: configs: Configs that had RPMSG_CHAR now get RPMSG_CTRL
In the commit 617d32938d1b ("rpmsg: Move the rpmsg control device
from rpmsg_char to rpmsg_ctrl"), we split the rpmsg_char driver in two.
By default give everyone who had the old driver enabled the rpmsg_ctrl
driver too.
xsk: Fix possible crash when multiple sockets are created
Fix a crash that happens if an Rx only socket is created first, then a
second socket is created that is Tx only and bound to the same umem as
the first socket and also the same netdev and queue_id together with the
XDP_SHARED_UMEM flag. In this specific case, the tx_descs array page
pool was not created by the first socket as it was an Rx only socket.
When the second socket is bound it needs this tx_descs array of this
shared page pool as it has a Tx component, but unfortunately it was
never allocated, leading to a crash. Note that this array is only used
for zero-copy drivers using the batched Tx APIs, currently only ice and
i40e.
Detect such case during bind() and allocate this memory region via newly
introduced xp_alloc_tx_descs(). Also, use kvcalloc instead of kcalloc as
for other buffer pool allocations, so that it matches the kvfree() from
xp_destroy().
Adam Zabrocki [Fri, 22 Apr 2022 16:40:27 +0000 (18:40 +0200)]
kprobes: Fix KRETPROBES when CONFIG_KRETPROBE_ON_RETHOOK is set
The recent kernel change in 73f9b911faa7 ("kprobes: Use rethook for kretprobe
if possible"), introduced a potential NULL pointer dereference bug in the
KRETPROBE mechanism. The official Kprobes documentation defines that "Any or
all handlers can be NULL". Unfortunately, there is a missing return handler
verification to fulfill these requirements and can result in a NULL pointer
dereference bug.
This patch adds such verification in kretprobe_rethook_handler() function.
Samuel Holland [Sun, 24 Apr 2022 16:26:21 +0000 (11:26 -0500)]
drm/sun4i: Remove obsolete references to PHYS_OFFSET
commit b4bdc4fbf8d0 ("soc: sunxi: Deal with the MBUS DMA offsets in a
central place") added a platform device notifier that sets the DMA
offset for all of the display engine frontend and backend devices.
The code applying the offset to DMA buffer physical addresses was then
removed from the backend driver in commit 756668ba682e ("drm/sun4i:
backend: Remove the MBUS quirks"), but the code subtracting PHYS_OFFSET
was left in the frontend driver.
As a result, the offset was applied twice in the frontend driver. This
likely went unnoticed because it only affects specific configurations
(scaling or certain pixel formats) where the frontend is used, on boards
with both one of these older SoCs and more than 1 GB of DRAM.
In addition, the references to PHYS_OFFSET prevent compiling the driver
on architectures where PHYS_OFFSET is not defined.
Sean Anderson [Mon, 25 Apr 2022 17:14:09 +0000 (13:14 -0400)]
usb: phy: generic: Get the vbus supply
While support for working with a vbus was added, the regulator was never
actually gotten (despite what was documented). Fix this by actually
getting the supply from the device tree.
If the user sets the usb_request's no_interrupt, then there will be no
completion event for the request. Currently the driver incorrectly uses
the event status of a different request to report the status for a
request with no_interrupt. The dwc3 driver needs to check the TRB status
associated with the request when reporting its status.
Note: this is only applicable to missed_isoc TRB completion status, but
the other status are also listed for completeness/documentation.
virtio_net: fix wrong buf address calculation when using xdp
We received a report[1] of kernel crashes when Cilium is used in XDP
mode with virtio_net after updating to newer kernels. After
investigating the reason it turned out that when using mergeable bufs
with an XDP program which adjusts xdp.data or xdp.data_meta page_to_buf()
calculates the build_skb address wrong because the offset can become less
than the headroom so it gets the address of the previous page (-X bytes
depending on how lower offset is):
page_to_skb: page addr ffff9eb2923e2000 buf ffff9eb2923e1ffc offset 252 headroom 256
This is a pr_err() I added in the beginning of page_to_skb which clearly
shows offset that is less than headroom by adding 4 bytes of metadata
via an xdp prog. The calculations done are:
receive_mergeable():
headroom = VIRTIO_XDP_HEADROOM; // VIRTIO_XDP_HEADROOM == 256 bytes
offset = xdp.data - page_address(xdp_page) -
vi->hdr_len - metasize;
page_to_skb():
p = page_address(page) + offset;
...
buf = p - headroom;
Now buf goes -4 bytes from the page's starting address as can be seen
above which is set as skb->head and skb->data by build_skb later. Depending
on what's done with the skb (when it's freed most often) we get all kinds
of corruptions and BUG_ON() triggers in mm[2]. We have to recalculate
the new headroom after the xdp program has run, similar to how offset
and len are recalculated. Headroom is directly related to
data_hard_start, data and data_meta, so we use them to get the new size.
The result is correct (similar pr_err() in page_to_skb, one case of
xdp_page and one case of virtnet buf):
a) Case with 4 bytes of metadata
[ 115.949641] page_to_skb: page addr ffff8b4dcfad2000 offset 252 headroom 252
[ 121.084105] page_to_skb: page addr ffff8b4dcf018000 offset 20732 headroom 252
b) Case of pushing data +32 bytes
[ 153.181401] page_to_skb: page addr ffff8b4dd0c4d000 offset 288 headroom 288
[ 158.480421] page_to_skb: page addr ffff8b4dd00b0000 offset 24864 headroom 288
c) Case of pushing data -33 bytes
[ 835.906830] page_to_skb: page addr ffff8b4dd3270000 offset 223 headroom 223
[ 840.839910] page_to_skb: page addr ffff8b4dcdd68000 offset 12511 headroom 223
Offset and headroom are equal because offset points to the start of
reserved bytes for the virtio_net header which are at buf start +
headroom, while data points at buf start + vnet hdr size + headroom so
when data or data_meta are adjusted by the xdp prog both the headroom size
and the offset change equally. We can use data_hard_start to compute the
new headroom after the xdp prog (linearized / page start case, the
virtnet buf case is similar just with bigger base offset):
xdp.data_hard_start = page_address + vnet_hdr
xdp.data = page_address + vnet_hdr + headroom
new headroom after xdp prog = xdp.data - xdp.data_hard_start - metasize
net: dsa: mv88e6xxx: Fix port_hidden_wait to account for port_base_addr
The other port_hidden functions rely on the port_read/port_write
functions to access the hidden control port. These functions apply the
offset for port_base_addr where applicable. Update port_hidden_wait to
use the port_wait_bit so that port_base_addr offsets are accounted for
when waiting for the busy bit to change.
Without the offset the port_hidden_wait function would timeout on
devices that have a non-zero port_base_addr (e.g. MV88E6141), however
devices that have a zero port_base_addr would operate correctly (e.g.
MV88E6390).
bug: Have __warn() prototype defined unconditionally
The __warn() prototype is declared in CONFIG_BUG scope but the function
definition in panic.c is unconditional. The IBT enablement started using
it unconditionally but a CONFIG_X86_KERNEL_IBT=y, CONFIG_BUG=n .config
will trigger a
arch/x86/kernel/traps.c: In function ‘__exc_control_protection’:
arch/x86/kernel/traps.c:249:17: error: implicit declaration of function \
‘__warn’; did you mean ‘pr_warn’? [-Werror=implicit-function-declaration]
Pull up the declarations so that they're unconditionally visible too.
Jonathan Lemon [Sun, 24 Apr 2022 16:53:07 +0000 (09:53 -0700)]
net: bcmgenet: hide status block before TX timestamping
The hardware checksum offloading requires use of a transmit
status block inserted before the outgoing frame data, this was
updated in '9a9ba2a4aaaa ("net: bcmgenet: always enable status blocks")'
However, skb_tx_timestamp() assumes that it is passed a raw frame
and PTP parsing chokes on this status block.
Fix this by calling __skb_pull(), which hides the TSB before calling
skb_tx_timestamp(), so an outgoing PTP packet is parsed correctly.
As the data in the skb has already been set up for DMA, and the
dma_unmap_* calls use a separately stored address, there is no
no effective change in the data transmission.
Lin Ma [Fri, 22 Apr 2022 11:43:40 +0000 (19:43 +0800)]
mctp: defer the kfree of object mdev->addrs
The function mctp_unregister() reclaims the device's relevant resource
when a netcard detaches. However, a running routine may be unaware of
this and cause the use-after-free of the mdev->addrs object.
An attacker can adopt the (recent provided) mtcpserial driver with pty
to fake the device detaching and use the userfaultfd to increase the
race success chance (in mctp_sendmsg). The KASan report for such a POC
is shown below:
[ 86.051955] ==================================================================
[ 86.051955] BUG: KASAN: use-after-free in mctp_local_output+0x4e9/0xb7d
[ 86.051955] Read of size 1 at addr ffff888005f298c0 by task poc/295
[ 86.051955]
[ 86.051955] Call Trace:
[ 86.051955] <TASK>
[ 86.051955] dump_stack_lvl+0x33/0x42
[ 86.051955] print_report.cold.13+0xb2/0x6b3
[ 86.051955] ? preempt_schedule_irq+0x57/0x80
[ 86.051955] ? mctp_local_output+0x4e9/0xb7d
[ 86.051955] kasan_report+0xa5/0x120
[ 86.051955] ? mctp_local_output+0x4e9/0xb7d
[ 86.051955] mctp_local_output+0x4e9/0xb7d
[ 86.051955] ? mctp_dev_set_key+0x79/0x79
[ 86.051955] ? copyin+0x38/0x50
[ 86.051955] ? _copy_from_iter+0x1b6/0xf20
[ 86.051955] ? sysvec_apic_timer_interrupt+0x97/0xb0
[ 86.051955] ? asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 86.051955] ? mctp_local_output+0x1/0xb7d
[ 86.051955] mctp_sendmsg+0x64d/0xdb0
[ 86.051955] ? mctp_sk_close+0x20/0x20
[ 86.051955] ? __fget_light+0x2fd/0x4f0
[ 86.051955] ? mctp_sk_close+0x20/0x20
[ 86.051955] sock_sendmsg+0xdd/0x110
[ 86.051955] __sys_sendto+0x1cc/0x2a0
[ 86.051955] ? __ia32_sys_getpeername+0xa0/0xa0
[ 86.051955] ? new_sync_write+0x335/0x550
[ 86.051955] ? alloc_file+0x22f/0x500
[ 86.051955] ? __ip_do_redirect+0x820/0x1820
[ 86.051955] ? vfs_write+0x44d/0x7b0
[ 86.051955] ? vfs_write+0x44d/0x7b0
[ 86.051955] ? fput_many+0x15/0x120
[ 86.051955] ? ksys_write+0x155/0x1b0
[ 86.051955] ? __ia32_sys_read+0xa0/0xa0
[ 86.051955] __x64_sys_sendto+0xd8/0x1b0
[ 86.051955] ? exit_to_user_mode_prepare+0x2f/0x120
[ 86.051955] ? syscall_exit_to_user_mode+0x12/0x20
[ 86.051955] do_syscall_64+0x3a/0x80
[ 86.051955] entry_SYSCALL_64_after_hwframe+0x44/0xae
[ 86.051955] RIP: 0033:0x7f82118a56b3
[ 86.051955] RSP: 002b:00007ffdb154b110 EFLAGS: 00000293 ORIG_RAX: 000000000000002c
[ 86.051955] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f82118a56b3
[ 86.051955] RDX: 0000000000000010 RSI: 00007f8211cd4000 RDI: 0000000000000007
[ 86.051955] RBP: 00007ffdb154c1d0 R08: 00007ffdb154b164 R09: 000000000000000c
[ 86.051955] R10: 0000000000000000 R11: 0000000000000293 R12: 000055d779800db0
[ 86.051955] R13: 00007ffdb154c2b0 R14: 0000000000000000 R15: 0000000000000000
[ 86.051955] </TASK>
[ 86.051955]
[ 86.051955] Allocated by task 295:
[ 86.051955] kasan_save_stack+0x1c/0x40
[ 86.051955] __kasan_kmalloc+0x84/0xa0
[ 86.051955] mctp_rtm_newaddr+0x242/0x610
[ 86.051955] rtnetlink_rcv_msg+0x2fd/0x8b0
[ 86.051955] netlink_rcv_skb+0x11c/0x340
[ 86.051955] netlink_unicast+0x439/0x630
[ 86.051955] netlink_sendmsg+0x752/0xc00
[ 86.051955] sock_sendmsg+0xdd/0x110
[ 86.051955] __sys_sendto+0x1cc/0x2a0
[ 86.051955] __x64_sys_sendto+0xd8/0x1b0
[ 86.051955] do_syscall_64+0x3a/0x80
[ 86.051955] entry_SYSCALL_64_after_hwframe+0x44/0xae
[ 86.051955]
[ 86.051955] Freed by task 301:
[ 86.051955] kasan_save_stack+0x1c/0x40
[ 86.051955] kasan_set_track+0x21/0x30
[ 86.051955] kasan_set_free_info+0x20/0x30
[ 86.051955] __kasan_slab_free+0x104/0x170
[ 86.051955] kfree+0x8c/0x290
[ 86.051955] mctp_dev_notify+0x161/0x2c0
[ 86.051955] raw_notifier_call_chain+0x8b/0xc0
[ 86.051955] unregister_netdevice_many+0x299/0x1180
[ 86.051955] unregister_netdevice_queue+0x210/0x2f0
[ 86.051955] unregister_netdev+0x13/0x20
[ 86.051955] mctp_serial_close+0x6d/0xa0
[ 86.051955] tty_ldisc_kill+0x31/0xa0
[ 86.051955] tty_ldisc_hangup+0x24f/0x560
[ 86.051955] __tty_hangup.part.28+0x2ce/0x6b0
[ 86.051955] tty_release+0x327/0xc70
[ 86.051955] __fput+0x1df/0x8b0
[ 86.051955] task_work_run+0xca/0x150
[ 86.051955] exit_to_user_mode_prepare+0x114/0x120
[ 86.051955] syscall_exit_to_user_mode+0x12/0x20
[ 86.051955] do_syscall_64+0x46/0x80
[ 86.051955] entry_SYSCALL_64_after_hwframe+0x44/0xae
[ 86.051955]
[ 86.051955] The buggy address belongs to the object at ffff888005f298c0
[ 86.051955] which belongs to the cache kmalloc-8 of size 8
[ 86.051955] The buggy address is located 0 bytes inside of
[ 86.051955] 8-byte region [ffff888005f298c0, ffff888005f298c8)
[ 86.051955]
[ 86.051955] The buggy address belongs to the physical page:
[ 86.051955] flags: 0x100000000000200(slab|node=0|zone=1)
[ 86.051955] raw: 0100000000000200dead000000000100dead000000000122ffff888005c42280
[ 86.051955] raw: 0000000000000000000000008066006600000001ffffffff0000000000000000
[ 86.051955] page dumped because: kasan: bad access detected
[ 86.051955]
[ 86.051955] Memory state around the buggy address:
[ 86.051955] ffff888005f29780: 00 fc fc fc fc 00 fc fc fc fc 00 fc fc fc fc 00
[ 86.051955] ffff888005f29800: fc fc fc fc 00 fc fc fc fc 00 fc fc fc fc 00 fc
[ 86.051955] >ffff888005f29880: fc fc fc fb fc fc fc fc fa fc fc fc fc fa fc fc
[ 86.051955] ^
[ 86.051955] ffff888005f29900: fc fc 00 fc fc fc fc 00 fc fc fc fc 00 fc fc fc
[ 86.051955] ffff888005f29980: fc 00 fc fc fc fc 00 fc fc fc fc 00 fc fc fc fc
[ 86.051955] ==================================================================
To this end, just like the commit e04480920d1e ("Bluetooth: defer
cleanup of resources in hci_unregister_dev()") this patch defers the
destructive kfree(mdev->addrs) in mctp_unregister to the mctp_dev_put,
where the refcount of mdev is zero and the entire device is reclaimed.
This prevents the use-after-free because the sendmsg thread holds the
reference of mdev in the mctp_route object.
Ville Syrjälä [Wed, 13 Apr 2022 15:28:51 +0000 (18:28 +0300)]
drm/i915/fbc: Consult hw.crtc instead of uapi.crtc
plane_state->uapi.crtc is not what we want to be looking at.
If bigjoiner is used hw.crtc is what tells us what crtc the plane
is supposedly using.
Not an actual problem on current hardware as the only FBC capable
pipe (A) can't be a bigjoiner slave and thus uapi.crtc==hw.crtc
always here. But when we get more FBC instances this will become
actually important.
It's noted that dcvs interrupts are not self-clearing, thus an interrupt
handler runs constantly, which leads to a severe regression in runtime.
To fix the problem an explicit write to clear interrupt register is
required, note that on OSM platforms the register may not be present.
Daniel Starke [Mon, 25 Apr 2022 10:47:26 +0000 (03:47 -0700)]
tty: n_gsm: fix sometimes uninitialized warning in gsm_dlci_modem_output()
'size' may be used uninitialized in gsm_dlci_modem_output() if called with
an adaption that is neither 1 nor 2. The function is currently only called
by gsm_modem_upd_via_data() and only for adaption 2.
Properly handle every invalid case by returning -EINVAL to silence the
compiler warning and avoid future regressions.
Stephen Boyd [Mon, 25 Apr 2022 23:47:24 +0000 (16:47 -0700)]
Merge tag 'sunxi-clk-fixes-for-5.18-2' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux into clk-fixes
Pull Allwinner clk fixes from Jernej Skrabec:
- Add missing sentinel
- check return value for platform_get_resource()
- mark rtc-32k as critical
* tag 'sunxi-clk-fixes-for-5.18-2' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:
clk: sunxi: sun9i-mmc: check return value after calling platform_get_resource()
clk: sunxi-ng: sun6i-rtc: Mark rtc-32k as critical
clk: sunxi-ng: fix not NULL terminated coccicheck error
Alexander Shiyan [Wed, 20 Apr 2022 07:06:39 +0000 (10:06 +0300)]
video: fbdev: clps711x-fb: Use syscon_regmap_lookup_by_phandle
Since version 5.13, the standard syscon bindings have been added
to all clps711x DT nodes, so we can now use the more general
syscon_regmap_lookup_by_phandle function to get the syscon pointer.
Jakub Kicinski [Mon, 25 Apr 2022 18:03:51 +0000 (11:03 -0700)]
Merge branch 'net-smc-two-fixes-for-smc-fallback'
Wen Gu says:
====================
net/smc: Two fixes for smc fallback
This patch set includes two fixes for smc fallback:
Patch 1/2 introduces some simple helpers to wrap the replacement
and restore of clcsock's callback functions. Make sure that only
the original callbacks will be saved and not overwritten.
Patch 2/2 fixes a syzbot reporting slab-out-of-bound issue where
smc_fback_error_report() accesses the already freed smc sock (see
https://lore.kernel.org/r/00000000000013ca8105d7ae3ada@google.com/).
The patch fixes it by resetting sk_user_data and restoring clcsock
callback functions timely in fallback situation.
But it should be noted that although patch 2/2 can fix the issue
of 'slab-out-of-bounds/use-after-free in smc_fback_error_report',
it can't pass the syzbot reproducer test. Because after applying
these two patches in upstream, syzbot reproducer triggered another
known issue like this:
==================================================================
BUG: KASAN: use-after-free in tcp_retransmit_timer+0x2ef3/0x3360 net/ipv4/tcp_timer.c:511
Read of size 8 at addr ffff888020328380 by task udevd/4158
CPU: 1 PID: 4158 Comm: udevd Not tainted 5.18.0-rc3-syzkaller-00074-gb05a5683eba6-dirty #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
<IRQ>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106
print_address_description.constprop.0.cold+0xeb/0x467 mm/kasan/report.c:313
print_report mm/kasan/report.c:429 [inline]
kasan_report.cold+0xf4/0x1c6 mm/kasan/report.c:491
tcp_retransmit_timer+0x2ef3/0x3360 net/ipv4/tcp_timer.c:511
tcp_write_timer_handler+0x5e6/0xbc0 net/ipv4/tcp_timer.c:622
tcp_write_timer+0xa2/0x2b0 net/ipv4/tcp_timer.c:642
call_timer_fn+0x1a5/0x6b0 kernel/time/timer.c:1421
expire_timers kernel/time/timer.c:1466 [inline]
__run_timers.part.0+0x679/0xa80 kernel/time/timer.c:1737
__run_timers kernel/time/timer.c:1715 [inline]
run_timer_softirq+0xb3/0x1d0 kernel/time/timer.c:1750
__do_softirq+0x29b/0x9c2 kernel/softirq.c:558
invoke_softirq kernel/softirq.c:432 [inline]
__irq_exit_rcu+0x123/0x180 kernel/softirq.c:637
irq_exit_rcu+0x5/0x20 kernel/softirq.c:649
sysvec_apic_timer_interrupt+0x93/0xc0 arch/x86/kernel/apic/apic.c:1097
</IRQ>
...
(detail report can be found in https://syzkaller.appspot.com/text?tag=CrashReport&x=15406b44f00000)
IMHO, the above issue is the same as this known one: https://syzkaller.appspot.com/bug?extid=694120e1002c117747ed,
and it doesn't seem to be related with SMC. The discussion about this known issue is ongoing and can be found in
https://lore.kernel.org/bpf/000000000000f75af905d3ba0716@google.com/T/.
And I added the temporary solution mentioned in the above discussion on
top of my two patches, the syzbot reproducer of 'slab-out-of-bounds/
use-after-free in smc_fback_error_report' no longer triggers any issue.
====================
syzbot reported a slab-out-of-bounds/use-after-free issue,
which was caused by accessing an already freed smc sock in
fallback-specific callback functions of clcsock.
This patch fixes the issue by restoring fallback-specific
callback functions to original ones and resetting clcsock
sk_user_data to NULL before freeing smc sock.
Meanwhile, this patch introduces sk_callback_lock to make
the access and assignment to sk_user_data mutually exclusive.
net/smc: Only save the original clcsock callback functions
Both listen and fallback process will save the current clcsock
callback functions and establish new ones. But if both of them
happen, the saved callback functions will be overwritten.
So this patch introduces some helpers to ensure that only save
the original callback functions of clcsock.
Fixes: 341adeec9ada ("net/smc: Forward wakeup to smc socket waitqueue after fallback") Signed-off-by: Wen Gu <[email protected]> Acked-by: Karsten Graul <[email protected]> Signed-off-by: Jakub Kicinski <[email protected]>
Merge tag 'f2fs-fix-5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs
Pull f2fs fixes from Jaegeuk Kim:
"This includes major bug fixes introduced in 5.18-rc1 and 5.17+:
- Remove obsolete whint_mode (5.18-rc1)
- Fix IO split issue caused by op_flags change in f2fs (5.18-rc1)
- Fix a wrong condition check to detect IO failure loop (5.18-rc1)
- Fix wrong data truncation during roll-forward (5.17+)"
* tag 'f2fs-fix-5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs:
f2fs: should not truncate blocks during roll-forward recovery
f2fs: fix wrong condition check when failing metapage read
f2fs: keep io_flags to avoid IO split due to different op_flags in two fio holders
f2fs: remove obsolete whint_mode
bus: sunxi-rsb: Fix the return value of sunxi_rsb_device_create()
This code is really spurious.
It always returns an ERR_PTR, even when err is known to be 0 and calls
put_device() after a successful device_register() call.
It is likely that the return statement in the normal path is missing.
Add 'return rdev;' to fix it.
no-MMU: expose vmalloc_huge() for alloc_large_system_hash()
It turns out that for the CONFIG_MMU=n builds, vmalloc_huge() was never
defined, since it's defined in mm/vmalloc.c, which doesn't get built for
the no-MMU configurations.
Just implement the trivial wrapper for the no-MMU case too. In fact,
just make it an alias to the existing __vmalloc() function that has the
same signature.
Eric Biggers [Fri, 22 Apr 2022 00:27:31 +0000 (17:27 -0700)]
Documentation: siphash: disambiguate HalfSipHash algorithm from hsiphash functions
Fix the documentation for the hsiphash functions to avoid conflating the
HalfSipHash algorithm with the hsiphash functions, since these functions
actually implement either HalfSipHash or SipHash, and random.c now uses
HalfSipHash (in a very special way) without the hsiphash functions.
This reverts 35a33ff3807d ("random: use memmove instead of memcpy for
remaining 32 bytes"), which was made on a totally bogus basis. The thing
it was worried about overlapping came from the stack, not from one of
its arguments, as Eric pointed out.
But the fact that this confusion even happened draws attention to the
fact that it's a bit non-obvious that the random_data parameter can
alias chacha_state, and in fact should do so when the caller can't rely
on the stack being cleared in a timely manner. So this commit documents
that.
Revert "arm64: dts: tegra: Fix boolean properties with values"
This reverts commit 1a67653de0dd, which caused a boot regression.
The behavior of the "drive-push-pull" in the kernel does not
match what the binding document describes. Revert Rob's patch
to make the DT match the kernel again, rather than the binding.
Fixes: 5b0b9e4c2c89 ("tcp: md5: incorrect tcp_header_len for incoming connections") Signed-off-by: Eric Dumazet <[email protected]> Cc: Francesco Ruggeri <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Eric Dumazet [Mon, 25 Apr 2022 00:34:07 +0000 (17:34 -0700)]
tcp: fix potential xmit stalls caused by TCP_NOTSENT_LOWAT
I had this bug sitting for too long in my pile, it is time to fix it.
Thanks to Doug Porter for reminding me of it!
We had various attempts in the past, including commit 0cbe6a8f089e ("tcp: remove SOCK_QUEUE_SHRUNK"),
but the issue is that TCP stack currently only generates
EPOLLOUT from input path, when tp->snd_una has advanced
and skb(s) cleaned from rtx queue.
If a flow has a big RTT, and/or receives SACKs, it is possible
that the notsent part (tp->write_seq - tp->snd_nxt) reaches 0
and no more data can be sent until tp->snd_una finally advances.
What is needed is to also check if POLLOUT needs to be generated
whenever tp->snd_nxt is advanced, from output path.
This bug triggers more often after an idle period, as
we do not receive ACK for at least one RTT. tcp_notsent_lowat
could be a fraction of what CWND and pacing rate would allow to
send during this RTT.
In a followup patch, I will remove the bogus call
to tcp_chrono_stop(sk, TCP_CHRONO_SNDBUF_LIMITED)
from tcp_check_space(). Fact that we have decided to generate
an EPOLLOUT does not mean the application has immediately
refilled the transmit queue. This optimistic call
might have been the reason the bug seemed not too serious.
Tested:
200 ms rtt, 1% packet loss, 32 MB tcp_rmem[2] and tcp_wmem[2]
$ echo 500000 >/proc/sys/net/ipv4/tcp_notsent_lowat
$ cat bench_rr.sh
SUM=0
for i in {1..10}
do
V=`netperf -H remote_host -l30 -t TCP_RR -- -r 10000000,10000 -o LOCAL_BYTES_SENT | egrep -v "MIGRATED|Bytes"`
echo $V
SUM=$(($SUM + $V))
done
echo SUM=$SUM
Vladimir Oltean [Thu, 21 Apr 2022 23:01:05 +0000 (02:01 +0300)]
net: mscc: ocelot: don't add VID 0 to ocelot->vlans when leaving VLAN-aware bridge
DSA, through dsa_port_bridge_leave(), first notifies the port of the
fact that it left a bridge, then, if that bridge was VLAN-aware, it
notifies the port of the change in VLAN awareness state, towards
VLAN-unaware mode.
So ocelot_port_vlan_filtering() can be called when ocelot_port->bridge
is NULL, and this makes ocelot_add_vlan_unaware_pvid() create a struct
ocelot_bridge_vlan with a vid of 0 and an "untagged" setting of true on
that port.
In a way this structure correctly reflects the reality, but by design,
VID 0 (OCELOT_STANDALONE_PVID) was not meant to be kept in the bridge
VLAN list of the driver, but managed separately.
Having OCELOT_STANDALONE_PVID in ocelot->vlans makes us trip up on
several sanity checks that did not expect to have this VID there.
For example, after we leave a VLAN-aware bridge and we re-join it, we
can no longer program egress-tagged VLANs to hardware:
# ip link add br0 type bridge vlan_filtering 1 && ip link set br0 up
# ip link set swp0 master br0
# ip link set swp0 nomaster
# ip link set swp0 master br0
# bridge vlan add dev swp0 vid 100
Error: mscc_ocelot_switch_lib: Port with more than one egress-untagged VLAN cannot have egress-tagged VLANs.
But this configuration is in fact supported by the hardware, since we
could use OCELOT_PORT_TAG_NATIVE. According to its comment:
/* all VLANs except the native VLAN and VID 0 are egress-tagged */
yet when assessing the eligibility for this mode, we do not check for
VID 0 in ocelot_port_uses_native_vlan(), instead we just ensure that
ocelot_port_num_untagged_vlans() == 1. This is simply because VID 0
doesn't have a bridge VLAN structure.
The way I identify the problem is that ocelot_port_vlan_filtering(false)
only means to call ocelot_add_vlan_unaware_pvid() when we dynamically
turn off VLAN awareness for a bridge we are under, and the PVID changes
from the bridge PVID to a reserved PVID based on the bridge number.
Since OCELOT_STANDALONE_PVID is statically added to the VLAN table
during ocelot_vlan_init() and never removed afterwards, calling
ocelot_add_vlan_unaware_pvid() for it is not intended and does not serve
any purpose.
Fix the issue by avoiding the call to ocelot_add_vlan_unaware_pvid(vid=0)
when we're resetting VLAN awareness after leaving the bridge, to become
a standalone port.
Fixes: 54c319846086 ("net: mscc: ocelot: enforce FDB isolation when VLAN-unaware") Signed-off-by: Vladimir Oltean <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Vladimir Oltean [Thu, 21 Apr 2022 23:01:04 +0000 (02:01 +0300)]
net: mscc: ocelot: ignore VID 0 added by 8021q module
Both the felix DSA driver and ocelot switchdev driver declare
dev->features & NETIF_F_HW_VLAN_CTAG_FILTER under certain circumstances*,
so the 8021q module will add VID 0 to our RX filter when the port goes
up, to ensure 802.1p traffic is not dropped.
We treat VID 0 as a special value (OCELOT_STANDALONE_PVID) which
deliberately does not have a struct ocelot_bridge_vlan associated with
it. Instead, this gets programmed to the VLAN table in ocelot_vlan_init().
If we allow external calls to modify VID 0, we reach the following
situation:
# ip link add br0 type bridge vlan_filtering 1 && ip link set br0 up
# ip link set swp0 master br0
# ip link set swp0 up # this adds VID 0 to ocelot->vlans with untagged=false
bridge vlan
port vlan-id
swp0 1 PVID Egress Untagged # the bridge also adds VID 1
br0 1 PVID Egress Untagged
# bridge vlan add dev swp0 vid 100 untagged
Error: mscc_ocelot_switch_lib: Port with egress-tagged VLANs cannot have more than one egress-untagged (native) VLAN.
This configuration should have been accepted, because
ocelot_port_manage_port_tag() should select OCELOT_PORT_TAG_NATIVE.
Yet it isn't, because we have an entry in ocelot->vlans which says
VID 0 should be egress-tagged, something the hardware can't do.
Fix this by suppressing additions/deletions on VID 0 and managing this
VLAN exclusively using OCELOT_STANDALONE_PVID.
*DSA toggles it when the port becomes VLAN-aware by joining a VLAN-aware
bridge. Ocelot declares it unconditionally for some reason.
Fixes: 54c319846086 ("net: mscc: ocelot: enforce FDB isolation when VLAN-unaware") Signed-off-by: Vladimir Oltean <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Vladimir Oltean [Thu, 21 Apr 2022 22:42:22 +0000 (01:42 +0300)]
net: dsa: flood multicast to CPU when slave has IFF_PROMISC
Certain DSA switches can eliminate flooding to the CPU when none of the
ports have the IFF_ALLMULTI or IFF_PROMISC flags set. This is done by
synthesizing a call to dsa_port_bridge_flags() for the CPU port, a call
which normally comes from the bridge driver via switchdev.
The bridge port flags and IFF_PROMISC|IFF_ALLMULTI have slightly
different semantics, and due to inattention/lack of proper testing, the
IFF_PROMISC flag allows unknown unicast to be flooded to the CPU, but
not unknown multicast.
This must be fixed by setting both BR_FLOOD (unicast) and BR_MCAST_FLOOD
in the synthesized dsa_port_bridge_flags() call, since IFF_PROMISC means
that packets should not be filtered regardless of their MAC DA.
Fixes: 7569459a52c9 ("net: dsa: manage flooding on the CPU ports") Signed-off-by: Vladimir Oltean <[email protected]> Reviewed-by: Florian Fainelli <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Peilin Ye [Thu, 21 Apr 2022 22:09:02 +0000 (15:09 -0700)]
ip_gre, ip6_gre: Fix race condition on o_seqno in collect_md mode
As pointed out by Jakub Kicinski, currently using TUNNEL_SEQ in
collect_md mode is racy for [IP6]GRE[TAP] devices. Consider the
following sequence of events:
1. An [IP6]GRE[TAP] device is created in collect_md mode using "ip link
add ... external". "ip" ignores "[o]seq" if "external" is specified,
so TUNNEL_SEQ is off, and the device is marked as NETIF_F_LLTX (i.e.
it uses lockless TX);
2. Someone sets TUNNEL_SEQ on outgoing skb's, using e.g.
bpf_skb_set_tunnel_key() in an eBPF program attached to this device;
3. gre_fb_xmit() or __gre6_xmit() processes these skb's:
Since we are not using the TX lock (&txq->_xmit_lock), multiple CPUs may
try to do this tunnel->o_seqno++ in parallel, which is racy. Fix it by
making o_seqno atomic_t.
As mentioned by Eric Dumazet in commit b790e01aee74 ("ip_gre: lockless
xmit"), making o_seqno atomic_t increases "chance for packets being out
of order at receiver" when NETIF_F_LLTX is on.
Maybe a better fix would be:
1. Do not ignore "oseq" in external mode. Users MUST specify "oseq" if
they want the kernel to allow sequencing of outgoing packets;
2. Reject all outgoing TUNNEL_SEQ packets if the device was not created
with "oseq".
Unfortunately, that would break userspace.
We could now make [IP6]GRE[TAP] devices always NETIF_F_LLTX, but let us
do it in separate patches to keep this fix minimal.
Peilin Ye [Thu, 21 Apr 2022 22:08:38 +0000 (15:08 -0700)]
ip6_gre: Make o_seqno start from 0 in native mode
For IP6GRE and IP6GRETAP devices, currently o_seqno starts from 1 in
native mode. According to RFC 2890 2.2., "The first datagram is sent
with a sequence number of 0." Fix it.
It is worth mentioning that o_seqno already starts from 0 in collect_md
mode, see the "if (tunnel->parms.collect_md)" clause in __gre6_xmit(),
where tunnel->o_seqno is passed to gre_build_header() before getting
incremented.
Peilin Ye [Thu, 21 Apr 2022 22:07:57 +0000 (15:07 -0700)]
ip_gre: Make o_seqno start from 0 in native mode
For GRE and GRETAP devices, currently o_seqno starts from 1 in native
mode. According to RFC 2890 2.2., "The first datagram is sent with a
sequence number of 0." Fix it.
It is worth mentioning that o_seqno already starts from 0 in collect_md
mode, see gre_fb_xmit(), where tunnel->o_seqno is passed to
gre_build_header() before getting incremented.
Dan Carpenter [Thu, 21 Apr 2022 15:46:13 +0000 (18:46 +0300)]
net: lan966x: fix a couple off by one bugs
The lan966x->ports[] array has lan966x->num_phys_ports elements. These
are assigned in lan966x_probe(). That means the > comparison should be
changed to >=.
The first off by one check is harmless but the second one could lead to
an out of bounds access and a crash.
Fixes: 5ccd66e01cbe ("net: lan966x: add support for interrupts from analyzer") Signed-off-by: Dan Carpenter <[email protected]> Signed-off-by: David S. Miller <[email protected]>
net/smc: sync err code when tcp connection was refused
In the current implementation, when TCP initiates a connection
to an unavailable [ip,port], ECONNREFUSED will be stored in the
TCP socket, but SMC will not. However, some apps (like curl) use
getsockopt(,,SO_ERROR,,) to get the error information, which makes
them miss the error message and behave strangely.
net: hns3: add return value for mailbox handling in PF
Currently, there are some querying mailboxes sent from VF to PF,
and VF will wait the PF's handling result. For mailbox
HCLGE_MBX_GET_QID_IN_PF and HCLGE_MBX_GET_RSS_KEY, it may fail
when the input parameter is invalid, but the prototype of their
handler function is void. In this case, PF always return success
to VF, which may cause the VF get incorrect result.
Fixes it by adding return value for these function.
Fixes: 63b1279d9905 ("net: hns3: check queue id range before using") Fixes: 532cfc0df1e4 ("net: hns3: add a check for index in hclge_get_rss_key()") Signed-off-by: Jian Shen <[email protected]> Signed-off-by: Guangbin Huang <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Jie Wang [Sun, 24 Apr 2022 12:57:23 +0000 (20:57 +0800)]
net: hns3: modify the return code of hclge_get_ring_chain_from_mbx
Currently, function hclge_get_ring_chain_from_mbx will return -ENOMEM if
ring_num is bigger than HCLGE_MBX_MAX_RING_CHAIN_PARAM_NUM. It is better to
return -EINVAL for the invalid parameter case.
So this patch fixes it by return -EINVAL in this abnormal branch.
Fixes: 5d02a58dae60 ("net: hns3: fix for buffer overflow smatch warning") Signed-off-by: Jie Wang <[email protected]> Signed-off-by: Guangbin Huang <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Peng Li [Sun, 24 Apr 2022 12:57:22 +0000 (20:57 +0800)]
net: hns3: fix error log of tx/rx tqps stats
The comments in function hclge_comm_tqps_update_stats is not right,
so fix it.
Fixes: 287db5c40d15 ("net: hns3: create new set of common tqp stats APIs for PF and VF reuse") Signed-off-by: Peng Li <[email protected]> Signed-off-by: Guangbin Huang <[email protected]> Signed-off-by: David S. Miller <[email protected]>
For debugfs node rx/tx_queue_info and rx/tx_bd_info, their output info is
aligned to the right, it's not aligned with output of other debugfs node,
so uniform their output info.
Fixes: 907676b13071 ("net: hns3: use tx bounce buffer for small packets") Fixes: e44c495d95e0 ("net: hns3: refactor queue info of debugfs") Fixes: 77e9184869c9 ("net: hns3: refactor dump bd info of debugfs") Signed-off-by: Hao Chen <[email protected]> Signed-off-by: Guangbin Huang <[email protected]> Signed-off-by: David S. Miller <[email protected]>
net: hns3: clear inited state and stop client after failed to register netdev
If failed to register netdev, it needs to clear INITED state and stop
client in case of cause problem when concurrency with uninitialized
process of driver.
Fixes: a289a7e5c1d4 ("net: hns3: put off calling register_netdev() until client initialize complete") Signed-off-by: Jian Shen <[email protected]> Signed-off-by: Guangbin Huang <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Martin Willi [Tue, 19 Apr 2022 13:47:00 +0000 (15:47 +0200)]
netfilter: Update ip6_route_me_harder to consider L3 domain
The commit referenced below fixed packet re-routing if Netfilter mangles
a routing key property of a packet and the packet is routed in a VRF L3
domain. The fix, however, addressed IPv4 re-routing, only.
This commit applies the same behavior for IPv6. While at it, untangle
the nested ternary operator to make the code more readable.
Xiubo Li [Thu, 14 Apr 2022 01:07:21 +0000 (09:07 +0800)]
ceph: fix possible NULL pointer dereference for req->r_session
The request will be inserted into the ci->i_unsafe_dirops before
assigning the req->r_session, so it's possible that we will hit
NULL pointer dereference bug here.
Xiubo Li [Tue, 22 Mar 2022 03:03:13 +0000 (11:03 +0800)]
ceph: remove incorrect session state check
Once the session is opened the s->s_ttl will be set, and when receiving
a new mdsmap and the MDS map is changed, it will be possibly will close
some sessions and open new ones. And then some sessions will be in
CLOSING state evening without unmounting.
Niels Dossche [Tue, 15 Mar 2022 15:29:47 +0000 (16:29 +0100)]
ceph: get snap_rwsem read lock in handle_cap_export for ceph_add_cap
ceph_add_cap says in its function documentation that the caller should
hold the read lock on the session snap_rwsem. Furthermore, not only
ceph_add_cap needs that lock, when it calls to ceph_lookup_snap_realm it
eventually calls ceph_get_snap_realm which states via lockdep that
snap_rwsem needs to be held. handle_cap_export calls ceph_add_cap
without that mdsc->snap_rwsem held. Thus, since ceph_get_snap_realm
and ceph_add_cap both need the lock, the common place to acquire that
lock is inside handle_cap_export.
if sun50i_cpufreq_get_efuse failed, then opp_tables leak.
Fixes: f328584f7bff ("cpufreq: Add sun50i nvmem based CPU scaling driver") Signed-off-by: Xiaobing Luo <[email protected]> Reviewed-by: Samuel Holland <[email protected]> Signed-off-by: Viresh Kumar <[email protected]>
drm/i915: Check EDID for HDR static metadata when choosing blc
We have now seen panel (XMG Core 15 e21 laptop) advertizing support
for Intel proprietary eDP backlight control via DPCD registers, but
actually working only with legacy pwm control.
This patch adds panel EDID check for possible HDR static metadata and
Intel proprietary eDP backlight control is used only if that exists.
Missing HDR static metadata is ignored if user specifically asks for
Intel proprietary eDP backlight control via enable_dpcd_backlight
parameter.
v2 :
- Ignore missing HDR static metadata if Intel proprietary eDP
backlight control is forced via i915.enable_dpcd_backlight
- Printout info message if panel is missing HDR static metadata and
support for Intel proprietary eDP backlight control is detected
Hans de Goede [Mon, 18 Apr 2022 15:09:36 +0000 (17:09 +0200)]
drm/i915: Fix DISP_POS_Y and DISP_HEIGHT defines
Commit 428cb15d5b00 ("drm/i915: Clean up pre-skl primary plane registers")
introduced DISP_POS_Y and DISP_HEIGHT defines but accidentally set these
their masks to REG_GENMASK(31, 0) instead of REG_GENMASK(31, 16).
This breaks the primary display pane on at least pineview machines, fix
the mask to fix the primary display pane only showing black.
Tested on an Acer One AO532h with an Intel N450 SoC.
Samuel Holland [Mon, 11 Apr 2022 05:00:59 +0000 (00:00 -0500)]
clk: sunxi-ng: sun6i-rtc: Mark rtc-32k as critical
Because some newer hardware variants have multiple possible parents for
the RTC's timekeeping clock, this driver models it as a "rtc-32k" clock.
However, it does not add any consumer for this clock. This causes the
common clock framework to disable it, preventing RTC time access.
Since the RTC's timekeeping clock should always be enabled, regardless
of which drivers are loaded, let's mark this clock as critical instead
of adding a consumer in the RTC driver.
Merge tag 'perf_urgent_for_v5.18_rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Borislav Petkov:
- Add Sapphire Rapids CPU support
- Fix a perf vmalloc-ed buffer mapping error (PERF_USE_VMALLOC in use)
* tag 'perf_urgent_for_v5.18_rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86/cstate: Add SAPPHIRERAPIDS_X CPU support
perf/core: Fix perf_mmap fail when CONFIG_PERF_USE_VMALLOC enabled
Merge tag 'edac_urgent_for_v5.18_rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras
Pull EDAC fix from Borislav Petkov:
- Read the reported error count from the proper register on
synopsys_edac
* tag 'edac_urgent_for_v5.18_rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:
EDAC/synopsys: Read the error count from the correct register
kvmalloc: use vmalloc_huge for vmalloc allocations
Since commit 559089e0a93d ("vmalloc: replace VM_NO_HUGE_VMAP with
VM_ALLOW_HUGE_VMAP"), the use of hugepage mappings for vmalloc is an
opt-in strategy, because it caused a number of problems that weren't
noticed until x86 enabled it too.
One of the issues was fixed by Nick Piggin in commit 3b8000ae185c
("mm/vmalloc: huge vmalloc backing pages should be split rather than
compound"), but I'm still worried about page protection issues, and
VM_FLUSH_RESET_PERMS in particular.
However, like the hash table allocation case (commit f2edd118d02d:
"page_alloc: use vmalloc_huge for large system hash"), the use of
kvmalloc() should be safe from any such games, since the returned
pointer might be a SLUB allocation, and as such no user should
reasonably be using it in any odd ways.
We also know that the allocations are fairly large, since it falls back
to the vmalloc case only when a kmalloc() fails. So using a hugepage
mapping seems both safe and relevant.
This patch does show a weakness in the opt-in strategy: since the opt-in
flag is in the 'vm_flags', not the usual gfp_t allocation flags, very
few of the usual interfaces actually expose it.
That's not much of an issue in this case that already used one of the
fairly specialized low-level vmalloc interfaces for the allocation, but
for a lot of other vmalloc() users that might want to opt in, it's going
to be very inconvenient.
We'll either have to fix any compatibility problems, or expose it in the
gfp flags (__GFP_COMP would have made a lot of sense) to allow normal
vmalloc() users to use hugepage mappings. That said, the cases that
really matter were probably already taken care of by the hash tabel
allocation.
This is due to the AT25 driver using buffers on stack, which is not
possible with CONFIG_VMAP_STACK.
As mentionned in kernel Documentation (Documentation/spi/spi-summary.rst):
- Follow standard kernel rules, and provide DMA-safe buffers in
your messages. That way controller drivers using DMA aren't forced
to make extra copies unless the hardware requires it (e.g. working
around hardware errata that force the use of bounce buffering).
Modify the driver to use a buffer located in the at25 device structure
which is allocated via kmalloc during probe.
Protect writes in this new buffer with the driver's mutex.
Max Krummenacher [Thu, 14 Apr 2022 08:50:54 +0000 (10:50 +0200)]
ARM: dts: imx6ull-colibri: fix vqmmc regulator
The correct spelling for the property is gpios. Otherwise, the regulator
will neither reserve nor control any GPIOs. Thus, any SD/MMC card which
can use UHS-I modes will fail.
Merge tag '5.18-rc3-ksmbd-fixes' of git://git.samba.org/ksmbd
Pull ksmbd server fixes from Steve French:
- cap maximum sector size reported to avoid mount problems
- reference count fix
- fix filename rename race
* tag '5.18-rc3-ksmbd-fixes' of git://git.samba.org/ksmbd:
ksmbd: set fixed sector size to FS_SECTOR_SIZE_INFORMATION
ksmbd: increment reference count of parent fp
ksmbd: remove filename in ksmbd_file
Merge tag 'arc-5.18-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc
Pull ARC fixes from Vineet Gupta:
- Assorted fixes
* tag 'arc-5.18-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc:
ARC: remove redundant READ_ONCE() in cmpxchg loop
ARC: atomic: cleanup atomic-llsc definitions
arc: drop definitions of pgd_index() and pgd_offset{, _k}() entirely
ARC: dts: align SPI NOR node name with dtschema
ARC: Remove a redundant memset()
ARC: fix typos in comments
ARC: entry: fix syscall_trace_exit argument
projects / linux.git / log