Add generic hypercall functions usable for all normal (i.e. not iret)
hypercalls. Depending on the guest type and the processor vendor
different functions need to be used due to the to be used instruction
for entering the hypervisor:
- PV guests need to use syscall
- HVM/PVH guests on Intel need to use vmcall
- HVM/PVH guests on AMD and Hygon need to use vmmcall
As PVH guests need to issue hypercalls very early during boot, there
is a 4th hypercall function needed for HVM/PVH which can be used on
Intel and AMD processors. It will check the vendor type and then set
the Intel or AMD specific function to use via static_call().
This is part of XSA-466 / CVE-2024-53241.
Reported-by: Andrew Cooper <andrew.cooper3@citrix.com> Signed-off-by: Juergen Gross <jgross@suse.com> Co-developed-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The syscall instruction is used in Xen PV mode for doing hypercalls.
Allow syscall to be used in the kernel in case it is tagged with an
unwind hint for objtool.
This is part of XSA-466 / CVE-2024-53241.
Reported-by: Andrew Cooper <andrew.cooper3@citrix.com> Signed-off-by: Juergen Gross <jgross@suse.com> Co-developed-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In order to be able to differentiate between AMD and Intel based
systems for very early hypercalls without having to rely on the Xen
hypercall page, make get_cpu_vendor() non-static.
Refactor early_cpu_init() for the same reason by splitting out the
loop initializing cpu_devs() into an externally callable function.
This is part of XSA-466 / CVE-2024-53241.
Reported-by: Andrew Cooper <andrew.cooper3@citrix.com> Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When removing a netfront device directly after a suspend/resume cycle
it might happen that the queues have not been setup again, causing a
crash during the attempt to stop the queues another time.
Fix that by checking the queues are existing before trying to stop
them.
This is XSA-465 / CVE-2024-53240.
Reported-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com> Fixes: d50b7914fae0 ("xen-netfront: Fix NULL sring after live migration") Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Nikolay Kuratov [Mon, 16 Dec 2024 11:19:23 +0000 (14:19 +0300)]
tracing/kprobes: Skip symbol counting logic for module symbols in create_local_trace_kprobe()
commit b022f0c7e404 ("tracing/kprobes: Return EADDRNOTAVAIL when func matches several symbols")
avoids checking number_of_same_symbols() for module symbol in
__trace_kprobe_create(), but create_local_trace_kprobe() should avoid this
check too. Doing this check leads to ENOENT for module_name:symbol_name
constructions passed over perf_event_open.
No bug in newer kernels as it was fixed more generally by
commit 9d8616034f16 ("tracing/kprobes: Add symbol counting check when module loads")
Range propagation must not affect subreg_def marks, otherwise the
following example is rewritten by verifier incorrectly when
BPF_F_TEST_RND_HI32 flag is set:
(or zero extension of w1 at (2) is missing for architectures that
require zero extension for upper register half).
The following happens w/o this patch:
- r0 is marked as not a subreg at (0);
- w1 is marked as subreg at (2);
- w1 subreg_def is overridden at (3) by copy_register_state();
- w1 is read at (5) but mark_insn_zext() does not mark (2)
for zero extension, because w1 subreg_def is not set;
- because of BPF_F_TEST_RND_HI32 flag verifier inserts random
value for hi32 bits of (2) (marked (r));
- this random value is read at (5).
Fixes: 75748837b7e5 ("bpf: Propagate scalar ranges through register assignments.") Reported-by: Lonial Con <kongln9170@gmail.com> Signed-off-by: Lonial Con <kongln9170@gmail.com> Signed-off-by: Eduard Zingerman <eddyz87@gmail.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Closes: https://lore.kernel.org/bpf/7e2aa30a62d740db182c170fdd8f81c596df280d.camel@gmail.com Link: https://lore.kernel.org/bpf/20240924210844.1758441-1-eddyz87@gmail.com
[ shung-hsi.yu: sync_linked_regs() was called find_equal_scalars() before commit 4bf79f9be434 ("bpf: Track equal scalars history on per-instruction level"), and
modification is done because there is only a single call to
copy_register_state() before commit 98d7ca374ba4 ("bpf: Track delta between
"linked" registers."). ] Signed-off-by: Shung-Hsi Yu <shung-hsi.yu@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
After a recent change to clamp() and its variants [1] that increases the
coverage of the check that high is greater than low because it can be
done through inlining, certain build configurations (such as s390
defconfig) fail to build with clang with:
block/blk-iocost.c:1101:11: error: call to '__compiletime_assert_557' declared with 'error' attribute: clamp() low limit 1 greater than high limit active
1101 | inuse = clamp_t(u32, inuse, 1, active);
| ^
include/linux/minmax.h:218:36: note: expanded from macro 'clamp_t'
218 | #define clamp_t(type, val, lo, hi) __careful_clamp(type, val, lo, hi)
| ^
include/linux/minmax.h:195:2: note: expanded from macro '__careful_clamp'
195 | __clamp_once(type, val, lo, hi, __UNIQUE_ID(v_), __UNIQUE_ID(l_), __UNIQUE_ID(h_))
| ^
include/linux/minmax.h:188:2: note: expanded from macro '__clamp_once'
188 | BUILD_BUG_ON_MSG(statically_true(ulo > uhi), \
| ^
__propagate_weights() is called with an active value of zero in
ioc_check_iocgs(), which results in the high value being less than the
low value, which is undefined because the value returned depends on the
order of the comparisons.
The purpose of this expression is to ensure inuse is not more than
active and at least 1. This could be written more simply with a ternary
expression that uses min(inuse, active) as the condition so that the
value of that condition can be used if it is not zero and one if it is.
Do this conversion to resolve the error and add a comment to deter
people from turning this back into clamp().
This bug was first introduced in c27f3d011b08, where the author of the
patch probably meant to do DeleteMutex instead of ReleaseMutex. The
mutex leak was noticed later on and fixed in e4dfe108371, but the bogus
MutexRelease line was never removed, so do it now.
Similar to bonding driver, add NETIF_F_GSO_ENCAP_ALL to TEAM_VLAN_FEATURES
in order to support slave devices which propagate NETIF_F_GSO_UDP_TUNNEL &
NETIF_F_GSO_UDP_TUNNEL_CSUM as vlan_features.
Fixes: 3625920b62c3 ("teaming: fix vlan_features computing") Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Cc: Nikolay Aleksandrov <razor@blackwall.org> Cc: Ido Schimmel <idosch@idosch.org> Cc: Jiri Pirko <jiri@nvidia.com> Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org> Reviewed-by: Hangbin Liu <liuhangbin@gmail.com> Link: https://patch.msgid.link/20241210141245.327886-5-daniel@iogearbox.net Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Drivers like mlx5 expose NIC's vlan_features such as
NETIF_F_GSO_UDP_TUNNEL & NETIF_F_GSO_UDP_TUNNEL_CSUM which are
later not propagated when the underlying devices are bonded and
a vlan device created on top of the bond.
Right now, the more cumbersome workaround for this is to create
the vlan on top of the mlx5 and then enslave the vlan devices
to a bond.
To fix this, add NETIF_F_GSO_ENCAP_ALL to BOND_VLAN_FEATURES
such that bond_compute_features() can probe and propagate the
vlan_features from the slave devices up to the vlan device.
# ethtool -k enp2s0f0np0 | grep udp
tx-udp_tnl-segmentation: on
tx-udp_tnl-csum-segmentation: on
tx-udp-segmentation: on
rx-udp_tunnel-port-offload: on
rx-udp-gro-forwarding: off
# ethtool -k enp2s0f1np1 | grep udp
tx-udp_tnl-segmentation: on
tx-udp_tnl-csum-segmentation: on
tx-udp-segmentation: on
rx-udp_tunnel-port-offload: on
rx-udp-gro-forwarding: off
# ethtool -k bond0 | grep udp
tx-udp_tnl-segmentation: on
tx-udp_tnl-csum-segmentation: on
tx-udp-segmentation: on
rx-udp_tunnel-port-offload: off [fixed]
rx-udp-gro-forwarding: off
Before:
# ethtool -k bond0.100 | grep udp
tx-udp_tnl-segmentation: off [requested on]
tx-udp_tnl-csum-segmentation: off [requested on]
tx-udp-segmentation: on
rx-udp_tunnel-port-offload: off [fixed]
rx-udp-gro-forwarding: off
After:
# ethtool -k bond0.100 | grep udp
tx-udp_tnl-segmentation: on
tx-udp_tnl-csum-segmentation: on
tx-udp-segmentation: on
rx-udp_tunnel-port-offload: off [fixed]
rx-udp-gro-forwarding: off
Various users have run into this reporting performance issues when
configuring Cilium in vxlan tunneling mode and having the combination
of bond & vlan for the core devices connecting the Kubernetes cluster
to the outside world.
Fixes: a9b3ace44c7d ("bonding: fix vlan_features computing") Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Cc: Nikolay Aleksandrov <razor@blackwall.org> Cc: Ido Schimmel <idosch@idosch.org> Cc: Jiri Pirko <jiri@nvidia.com> Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org> Reviewed-by: Hangbin Liu <liuhangbin@gmail.com> Link: https://patch.msgid.link/20241210141245.327886-3-daniel@iogearbox.net Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Virtual netdevs should use NETIF_F_GSO_SOFTWARE to forward GSO skbs
as-is and let the final drivers deal with them when supported.
Also remove NETIF_F_GSO_UDP_L4 from bonding and team drivers as it's
now included in the "software" list.
Suggested-by: Willem de Bruijn <willemb@google.com> Signed-off-by: Alexander Lobakin <alobakin@pm.me> Acked-by: Willem de Bruijn <willemb@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 77b11c8bf3a2 ("bonding: Fix feature propagation of NETIF_F_GSO_ENCAP_ALL") Signed-off-by: Sasha Levin <sashal@kernel.org>
In general, 'qlen' of any classful qdisc should keep track of the
number of packets that the qdisc itself and all of its children holds.
In case of netem, 'qlen' only accounts for the packets in its internal
tfifo. When netem is used with a child qdisc, the child qdisc can use
'qdisc_tree_reduce_backlog' to inform its parent, netem, about created
or dropped SKBs. This function updates 'qlen' and the backlog statistics
of netem, but netem does not account for changes made by a child qdisc.
'qlen' then indicates the wrong number of packets in the tfifo.
If a child qdisc creates new SKBs during enqueue and informs its parent
about this, netem's 'qlen' value is increased. When netem dequeues the
newly created SKBs from the child, the 'qlen' in netem is not updated.
If 'qlen' reaches the configured sch->limit, the enqueue function stops
working, even though the tfifo is not full.
Reproduce the bug:
Ensure that the sender machine has GSO enabled. Configure netem as root
qdisc and tbf as its child on the outgoing interface of the machine
as follows:
$ tc qdisc add dev <oif> root handle 1: netem delay 100ms limit 100
$ tc qdisc add dev <oif> parent 1:0 tbf rate 50Mbit burst 1542 latency 50ms
Send bulk TCP traffic out via this interface, e.g., by running an iPerf3
client on the machine. Check the qdisc statistics:
$ tc -s qdisc show dev <oif>
Statistics after 10s of iPerf3 TCP test before the fix (note that
netem's backlog > limit, netem stopped accepting packets):
qdisc netem 1: root refcnt 2 limit 1000 delay 100ms
Sent 2767766 bytes 1848 pkt (dropped 652, overlimits 0 requeues 0)
backlog 4294528236b 1155p requeues 0
qdisc tbf 10: parent 1:1 rate 50Mbit burst 1537b lat 50ms
Sent 2767766 bytes 1848 pkt (dropped 327, overlimits 7601 requeues 0)
backlog 0b 0p requeues 0
tbf segments the GSO SKBs (tbf_segment) and updates the netem's 'qlen'.
The interface fully stops transferring packets and "locks". In this case,
the child qdisc and tfifo are empty, but 'qlen' indicates the tfifo is at
its limit and no more packets are accepted.
This patch adds a counter for the entries in the tfifo. Netem's 'qlen' is
only decreased when a packet is returned by its dequeue function, and not
during enqueuing into the child qdisc. External updates to 'qlen' are thus
accounted for and only the behavior of the backlog statistics changes. As
in other qdiscs, 'qlen' then keeps track of how many packets are held in
netem and all of its children. As before, sch->limit remains as the
maximum number of packets in the tfifo. The same applies to netem's
backlog statistics.
Fixes: 50612537e9ab ("netem: fix classful handling") Signed-off-by: Martin Ottens <martin.ottens@fau.de> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Link: https://patch.msgid.link/20241210131412.1837202-1-martin.ottens@fau.de Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The module parameter qcaspi_pluggable controls if QCA7000 signature
should be checked at driver probe (current default) or not. Unfortunately
this could fail in case the chip is temporary in reset, which isn't under
total control by the Linux host. So disable this check per default
in order to avoid unexpected probe failures.
Fixes: 291ab06ecf67 ("net: qualcomm: new Ethernet over SPI driver for QCA7000") Signed-off-by: Stefan Wahren <wahrenst@gmx.net> Link: https://patch.msgid.link/20241206184643.123399-3-wahrenst@gmx.net Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Storing the maximum clock speed in module parameter qcaspi_clkspeed
has the unintended side effect that the first probed instance
defines the value for all other instances. Fix this issue by storing
it in max_speed_hz of the relevant SPI device.
This fix keeps the priority of the speed parameter (module parameter,
device tree property, driver default). Btw this uses the opportunity
to get the rid of the unused member clkspeed.
Fixes: 291ab06ecf67 ("net: qualcomm: new Ethernet over SPI driver for QCA7000") Signed-off-by: Stefan Wahren <wahrenst@gmx.net> Link: https://patch.msgid.link/20241206184643.123399-2-wahrenst@gmx.net Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
t4_set_vf_mac_acl() uses pf to set mac addr, but t4vf_get_vf_mac_acl()
uses port number to get mac addr, this leads to error when an attempt
to set MAC address on VF's of PF2 and PF3.
This patch fixes the issue by using port number to set mac address.
Fixes: e0cdac65ba26 ("cxgb4vf: configure ports accessible by the VF") Signed-off-by: Anumula Murali Mohan Reddy <anumula@chelsio.com> Signed-off-by: Potnuri Bharat Teja <bharat@chelsio.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20241206062014.49414-1-anumula@chelsio.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
In acpi_decode_space() addr->info.mem.caching is checked on main level
for any resource type but addr->info.mem is part of union and thus
valid only if the resource type is memory range.
Move the check inside the preceeding switch/case to only execute it
when the union is of correct type.
Fixes: fcb29bbcd540 ("ACPI: Add prefetch decoding to the address space parser") Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Link: https://patch.msgid.link/20241202100614.20731-1-ilpo.jarvinen@linux.intel.com Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
On both port_tc_ip_test() and port_tc_arp_test(), the max occupancy is
checked on $h2 twice, when only the error message is different and does not
match the check itself.
Remove the two duplicated test cases from the test.
The test is sending only one packet generated with mausezahn from $h1 to
$h2. However, for some reason, it is testing for non-zero maximum occupancy
in both the ingress pool of $h1 and $h2. The former only passes when $h2
happens to send a packet.
Avoid intermittent failures by removing unintentional test case
regarding the ingress pool of $h1.
When TT changes list is too big to fit in packet due to MTU size, an
empty OGM is sent expected other node to send TT request to get the
changes. The issue is that tt.last_changeset was not built thus the
originator was responding with previous changes to those TT requests
(see batadv_send_my_tt_response). Also the changes list was never
cleaned up effectively never ending growing from this point onwards,
repeatedly sending the same TT response changes over and over, and
creating a new empty OGM every OGM interval expecting for the local
changes to be purged.
When there is more TT changes that can fit in packet, drop all changes,
send empty OGM and wait for TT request so we can respond with a full
table instead.
Fixes: e1bf0c14096f ("batman-adv: tvlv - convert tt data sent within OGMs") Signed-off-by: Remi Pommarel <repk@triplefau.lt> Acked-by: Antonio Quartulli <Antonio@mandelbit.com> Signed-off-by: Sven Eckelmann <sven@narfation.org> Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
The number of entries filled by batadv_tt_tvlv_generate() can be less
than initially expected in batadv_tt_prepare_tvlv_{global,local}_data()
(changes can be removed by batadv_tt_local_event() in ADD+DEL sequence
in the meantime as the lock held during the whole tvlv global/local data
generation).
Thus tvlv_len could be bigger than the actual TT entry size that need
to be sent so full table TT_RESPONSE could hold invalid TT entries such
as below.
Remove the extra allocated space to avoid sending uninitialized entries
for full table TT_RESPONSE in both batadv_send_other_tt_response() and
batadv_send_my_tt_response().
Fixes: 7ea7b4a14275 ("batman-adv: make the TT CRC logic VLAN specific") Signed-off-by: Remi Pommarel <repk@triplefau.lt> Signed-off-by: Sven Eckelmann <sven@narfation.org> Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
The number of TT changes can be less than initially expected in
batadv_tt_tvlv_container_update() (changes can be removed by
batadv_tt_local_event() in ADD+DEL sequence between reading
tt_diff_entries_num and actually iterating the change list under lock).
Thus tt_diff_len could be bigger than the actual changes size that need
to be sent. Because batadv_send_my_tt_response sends the whole
packet, uninitialized data can be interpreted as TT changes on other
nodes leading to weird TT global entries on those nodes such as:
BUG: KASAN: vmalloc-out-of-bounds in cmd_to_func drivers/acpi/nfit/
core.c:416 [inline]
BUG: KASAN: vmalloc-out-of-bounds in acpi_nfit_ctl+0x20e8/0x24a0
drivers/acpi/nfit/core.c:459
The issue occurs in cmd_to_func when the call_pkg->nd_reserved2
array is accessed without verifying that call_pkg points to a buffer
that is appropriately sized as a struct nd_cmd_pkg. This can lead
to out-of-bounds access and undefined behavior if the buffer does not
have sufficient space.
To address this, a check was added in acpi_nfit_ctl() to ensure that
buf is not NULL and that buf_len is less than sizeof(*call_pkg)
before accessing it. This ensures safe access to the members of
call_pkg, including the nd_reserved2 array.
When accessing a file with more entries than ES_MAX_ENTRY_NUM, the bh-array
is allocated in __exfat_get_entry_set. The problem is that the bh-array is
allocated with GFP_KERNEL. It does not make sense. In the following cases,
a deadlock for sbi->s_lock between the two processes may occur.
As the final stages of socket destruction may be delayed, it is possible
that virtio_transport_recv_listen() will be called after the accept_queue
has been flushed, but before the SOCK_DONE flag has been set. As a result,
sockets enqueued after the flush would remain unremoved, leading to a
memory leak.
Fixes: 3fe356d58efa ("vsock/virtio: discard packets only when socket is really closed") Reviewed-by: Stefano Garzarella <sgarzare@redhat.com> Signed-off-by: Michal Luczaj <mhal@rbox.co> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
[ Adapted due to missing commit 71dc9ec9ac7d ("virtio/vsock: replace virtio_vsock_pkt with sk_buff") ] Signed-off-by: Tomas Krcka <krckatom@amazon.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
And that includes the new link sock_map_add_link() added just before
the unref.
This results in a sockmap holding a socket, but without the respective
link. This in turn means that close(sock) won't trigger the cleanup,
i.e. a closed socket will not be automatically removed from the sockmap.
Stop tearing the links when a matching link_raw is found.
Fixes: 604326b41a6f ("bpf, sockmap: convert to generic sk_msg interface") Signed-off-by: Michal Luczaj <mhal@rbox.co> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Reviewed-by: John Fastabend <john.fastabend@gmail.com> Link: https://lore.kernel.org/bpf/20241202-sockmap-replace-v1-1-1e88579e7bd5@rbox.co Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix a minor mistakes in the scrub tracepoints that can manifest when
inode-rooted btrees are enabled. The existing code worked fine for bmap
btrees, but we should tighten the code up to be less sloppy.
Cc: <stable@vger.kernel.org> # v5.7 Fixes: 92219c292af8dd ("xfs: convert btree cursor inode-private member names") Signed-off-by: "Darrick J. Wong" <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Way back when we first implemented FICLONE for XFS, life was simple --
either the the entire remapping completed, or something happened and we
had to return an errno explaining what happened. Neither of those
ioctls support returning partial results, so it's all or nothing.
Then things got complicated when copy_file_range came along, because it
actually can return the number of bytes copied, so commit 3f68c1f562f1e4
tried to make it so that we could return a partial result if the
REMAP_FILE_CAN_SHORTEN flag is set. This is also how FIDEDUPERANGE can
indicate that the kernel performed a partial deduplication.
Unfortunately, the logic is wrong if an error stops the remapping and
CAN_SHORTEN is not set. Because those callers cannot return partial
results, it is an error for ->remap_file_range to return a positive
quantity that is less than the @len passed in. Implementations really
should be returning a negative errno in this case, because that's what
btrfs (which introduced FICLONE{,RANGE}) did.
Therefore, ->remap_range implementations cannot silently drop an errno
that they might have when the number of bytes remapped is less than the
number of bytes requested and CAN_SHORTEN is not set.
Found by running generic/562 on a 64k fsblock filesystem and wondering
why it reported corrupt files.
Cc: <stable@vger.kernel.org> # v4.20 Fixes: 3fc9f5e409319e ("xfs: remove xfs_reflink_remap_range")
Really-Fixes: 3f68c1f562f1e4 ("xfs: support returning partial reflink results") Signed-off-by: "Darrick J. Wong" <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Considering that in some extreme cases,
when u_serial driver is accessed by multiple threads,
Thread A is executing the open operation and calling the gs_open,
Thread B is executing the disconnect operation and calling the
gserial_disconnect function,The port->port_usb pointer will be set to NULL.
E.g.
Thread A Thread B
gs_open() gadget_unbind_driver()
gs_start_io() composite_disconnect()
gs_start_rx() gserial_disconnect()
... ...
spin_unlock(&port->port_lock)
status = usb_ep_queue() spin_lock(&port->port_lock)
spin_lock(&port->port_lock) port->port_usb = NULL
gs_free_requests(port->port_usb->in) spin_unlock(&port->port_lock)
Crash
This causes thread A to access a null pointer (port->port_usb is null)
when calling the gs_free_requests function, causing a crash.
If port_usb is NULL, the release request will be skipped as it
will be done by gserial_disconnect.
So add a null pointer check to gs_start_io before attempting
to access the value of the pointer port->port_usb.
On Rasperry Pis without onboard USB hub the power cycle during
power connect init only disable the port but never enabled it again:
usb usb1-port1: attempt power cycle
The port relevant part in dwc2_hcd_hub_control() is skipped in case
port_connect_status = 0 under the assumption the core is or will be soon
in device mode. But this assumption is wrong, because after ClearPortFeature
USB_PORT_FEAT_POWER the port_connect_status will also be 0 and
SetPortFeature (incl. USB_PORT_FEAT_POWER) will be a no-op.
Fix the behavior of dwc2_hcd_hub_control() by replacing the
port_connect_status check with dwc2_is_device_mode().
If the current USB request was aborted, the spi thread would not respond
to any further requests. This is because the "curr_urb" pointer would
not become NULL, so no further requests would be taken off the queue.
The solution here is to set the "urb_done" flag, as this will cause the
correct handling of the URB. Also clear interrupts that should only be
expected if an URB is in progress.
Fixes: 93efb0c65683 ("octeontx2-pf: Fix out-of-bounds read in otx2_get_fecparam()") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fixes: 93efb0c656837 ("octeontx2-pf: Fix out-of-bounds read in otx2_get_fecparam()") Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Per UVC 1.1+ specification 3.7.2, units and terminals must have a non-zero
unique ID.
```
Each Unit and Terminal within the video function is assigned a unique
identification number, the Unit ID (UID) or Terminal ID (TID), contained in
the bUnitID or bTerminalID field of the descriptor. The value 0x00 is
reserved for undefined ID,
```
So, deny allocating an entity with ID 0 or an ID that belongs to a unit
that is already added to the list of entities.
This also prevents some syzkaller reproducers from triggering warnings due
to a chain of entities referring to themselves. In one particular case, an
Output Unit is connected to an Input Unit, both with the same ID of 1. But
when looking up for the source ID of the Output Unit, that same entity is
found instead of the input entity, which leads to such warnings.
In another case, a backward chain was considered finished as the source ID
was 0. Later on, that entity was found, but its pads were not valid.
Here is a sample stack trace for one of those cases.
Cc: stable@vger.kernel.org Reported-by: syzbot+0584f746fde3d52b4675@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=0584f746fde3d52b4675 Reported-by: syzbot+dd320d114deb3f5bb79b@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=dd320d114deb3f5bb79b Fixes: a3fbc2e6bb05 ("media: mc-entity.c: use WARN_ON, validate link pads") Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@igalia.com> Reviewed-by: Ricardo Ribalda <ribalda@chromium.org> Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Link: https://lore.kernel.org/r/20240913180601.1400596-2-cascardo@igalia.com Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
[ ribalda: The context around the changes differs from master. This
version is also missing the gpio unit, so that part is gone from the
patch. ] Signed-off-by: Ricardo Ribalda <ribalda@chromium.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The current USB-audio driver code doesn't check bLength of each
descriptor at traversing for clock descriptors. That is, when a
device provides a bogus descriptor with a shorter bLength, the driver
might hit out-of-bounds reads.
For addressing it, this patch adds sanity checks to the validator
functions for the clock descriptor traversal. When the descriptor
length is shorter than expected, it's skipped in the loop.
For the clock source and clock multiplier descriptors, we can just
check bLength against the sizeof() of each descriptor type.
OTOH, the clock selector descriptor of UAC2 and UAC3 has an array
of bNrInPins elements and two more fields at its tail, hence those
have to be checked in addition to the sizeof() check.
The MODE SELECT(6) command allows handling mode page buffers that are up to
255 bytes, including the 4 byte header needed in front of the page
buffer. For requests larger than this limit, automatically use the MODE
SELECT(10) command.
In both cases, since scsi_mode_select() adds the mode select page header,
checks on the buffer length value must include this header size to avoid
overflows of the command CDB allocation length field.
While at it, use put_unaligned_be16() for setting the header block
descriptor length and CDB allocation length when using MODE SELECT(10).
For devices that explicitly asked for MODE SENSE(10) use, make sure that
scsi_mode_sense() is called with a buffer of at least 8 bytes so that the
sense header fits.
The Rockchip PCIe endpoint controller handles PCIe transfers addresses
by masking the lower bits of the programmed PCI address and using the
same number of lower bits masked from the CPU address space used for the
mapping. For a PCI mapping of <size> bytes starting from <pci_addr>,
the number of bits masked is the number of address bits changing in the
address range [pci_addr..pci_addr + size - 1].
However, rockchip_pcie_prog_ep_ob_atu() calculates num_pass_bits only
using the size of the mapping, resulting in an incorrect number of mask
bits depending on the value of the PCI address to map.
Fix this by introducing the helper function
rockchip_pcie_ep_ob_atu_num_bits() to correctly calculate the number of
mask bits to use to program the address translation unit. The number of
mask bits is calculated depending on both the PCI address and size of
the mapping, and clamped between 8 and 20 using the macros
ROCKCHIP_PCIE_AT_MIN_NUM_BITS and ROCKCHIP_PCIE_AT_MAX_NUM_BITS. As
defined in the Rockchip RK3399 TRM V1.3 Part2, Sections 17.5.5.1.1 and
17.6.8.2.1, this clamping is necessary because:
1) The lower 8 bits of the PCI address to be mapped by the outbound
region are ignored. So a minimum of 8 address bits are needed and
imply that the PCI address must be aligned to 256.
2) The outbound memory regions are 1MB in size. So while we can specify
up to 63-bits for the PCI address (num_bits filed uses bits 0 to 5 of
the outbound address region 0 register), we must limit the number of
valid address bits to 20 to match the memory window maximum size (1
<< 20 = 1MB).
Fixes: cf590b078391 ("PCI: rockchip: Add EP driver for Rockchip PCIe controller") Link: https://lore.kernel.org/r/20241017015849.190271-2-dlemoal@kernel.org Signed-off-by: Damien Le Moal <dlemoal@kernel.org> Signed-off-by: Krzysztof Wilczyński <kwilczynski@kernel.org> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The origin mainline patch fix a buffer overflow issue in
amdgpu_debugfs_gprwave_read(), but it has not been introduced in kernel
6.1 and older kernels. This patch add a check in a wrong function in the
same file.
The compiler can fully inline the actual handler function of an interrupt
entry into the .irqentry.text entry point. If such a function contains an
access which has an exception table entry, modpost complains about a
section mismatch:
WARNING: vmlinux.o(__ex_table+0x447c): Section mismatch in reference ...
The relocation at __ex_table+0x447c references section ".irqentry.text"
which is not in the list of authorized sections.
Add .irqentry.text to OTHER_SECTIONS to cure the issue.
Heming Zhao [Thu, 12 Dec 2024 11:31:05 +0000 (19:31 +0800)]
ocfs2: Revert "ocfs2: fix the la space leak when unmounting an ocfs2 volume"
This reverts commit dfe6c5692fb5 ("ocfs2: fix the la space leak when
unmounting an ocfs2 volume").
In commit dfe6c5692fb5, the commit log "This bug has existed since the
initial OCFS2 code." is wrong. The correct introduction commit is 30dd3478c3cd ("ocfs2: correctly use ocfs2_find_next_zero_bit()").
The influence of commit dfe6c5692fb5 is that it provides a correct
fix for the latest kernel. however, it shouldn't be pushed to stable
branches. Let's use this commit to revert all branches that include dfe6c5692fb5 and use a new fix method to fix commit 30dd3478c3cd.
Fixes: dfe6c5692fb5 ("ocfs2: fix the la space leak when unmounting an ocfs2 volume") Signed-off-by: Heming Zhao <heming.zhao@suse.com> Cc: <stable@vger.kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
BIOS images may fail to load and null checks are added before they are
used.
This fixes 6 NULL_RETURNS issues reported by Coverity.
Reviewed-by: Harry Wentland <harry.wentland@amd.com> Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Keerthana K <keerthana.kalyanasundaram@broadcom.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Don't flush all pending DbC data requests when an endpoint halts.
An endpoint may halt and xHC DbC triggers a STALL error event if there's
an issue with a bulk data transfer. The transfer should restart once xHC
DbC receives a ClearFeature(ENDPOINT_HALT) request from the host.
Once xHC DbC restarts it will start from the TRB pointed to by dequeue
field in the endpoint context, which might be the same TRB we got the
STALL event for. Turn the TRB to a no-op in this case to make sure xHC
DbC doesn't reuse and tries to retransmit this same TRB after we already
handled it, and gave its corresponding data request back.
Other STALL events might be completely bogus.
Lukasz Bartosik discovered that xHC DbC might issue spurious STALL events
if hosts sends a ClearFeature(ENDPOINT_HALT) request to non-halted
endpoints even without any active bulk transfers.
Assume STALL event is spurious if it reports 0 bytes transferred, and
the endpoint stopped on the STALLED TRB.
Don't give back the data request corresponding to the TRB in this case.
The halted status is per endpoint. Track it with a per endpoint flag
instead of the driver invented DbC wide DS_STALLED state.
DbC remains in DbC-Configured state even if endpoints halt. There is no
Stalled state in the DbC Port state Machine (xhci section 7.6.6)
The rtime decompression routine does not fully check bounds during the
entirety of the decompression pass and can corrupt memory outside the
decompression buffer if the compressed data is corrupted. This adds the
required check to prevent this failure mode.
Cc: stable@vger.kernel.org Signed-off-by: Kinsey Moore <kinsey.moore@oarcorp.com> Signed-off-by: Richard Weinberger <richard@nod.at> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When DISCARD frees an ITE, it does not invalidate the
corresponding ITE. In the scenario of continuous saves and
restores, there may be a situation where an ITE is not saved
but is restored. This is unreasonable and may cause restore
to fail. This patch clears the corresponding ITE when DISCARD
frees an ITE.
Cc: stable@vger.kernel.org Fixes: eff484e0298d ("KVM: arm64: vgic-its: ITT save and restore") Signed-off-by: Kunkun Jiang <jiangkunkun@huawei.com>
[Jing: Update with entry write helper] Signed-off-by: Jing Zhang <jingzhangos@google.com> Link: https://lore.kernel.org/r/20241107214137.428439-6-jingzhangos@google.com Signed-off-by: Oliver Upton <oliver.upton@linux.dev> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
vgic_its_save_device_tables will traverse its->device_list to
save DTE for each device. vgic_its_restore_device_tables will
traverse each entry of device table and check if it is valid.
Restore if valid.
But when MAPD unmaps a device, it does not invalidate the
corresponding DTE. In the scenario of continuous saves
and restores, there may be a situation where a device's DTE
is not saved but is restored. This is unreasonable and may
cause restore to fail. This patch clears the corresponding
DTE when MAPD unmaps a device.
In all the vgic_its_save_*() functinos, they do not check whether
the data length is 8 bytes before calling vgic_write_guest_lock.
This patch adds the check. To prevent the kernel from being blown up
when the fault occurs, KVM_BUG_ON() is used. And the other BUG_ON()s
are replaced together.
Cc: stable@vger.kernel.org Signed-off-by: Kunkun Jiang <jiangkunkun@huawei.com>
[Jing: Update with the new entry read/write helpers] Signed-off-by: Jing Zhang <jingzhangos@google.com> Link: https://lore.kernel.org/r/20241107214137.428439-4-jingzhangos@google.com Signed-off-by: Oliver Upton <oliver.upton@linux.dev> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It turns out that we can't do this, because while the old behavior of
ignoring ignorable code points was most definitely wrong, we have
case-folding filesystems with on-disk hash values with that wrong
behavior.
So now you can't look up those names, because they hash to something
different.
Of course, it's also entirely possible that in the meantime people have
created *new* files with the new ("more correct") case folding logic,
and reverting will just make other things break.
The correct solution is to not do case folding in filesystems, but
sadly, people seem to never really understand that. People still see it
as a feature, not a bug.
Reported-by: Qi Han <hanqi@vivo.com> Link: https://bugzilla.kernel.org/show_bug.cgi?id=219586 Cc: Gabriel Krisman Bertazi <krisman@suse.de> Requested-by: Jaegeuk Kim <jaegeuk@kernel.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When activating a swap file we acquire the root's snapshot drew lock and
then check if the root is dead, failing and returning with -EPERM if it's
dead but without unlocking the root's snapshot lock. Fix this by adding
the missing unlock.
Fixes: 60021bd754c6 ("btrfs: prevent subvol with swapfile from being deleted") Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: David Sterba <dsterba@suse.com> Reviewed-by: Qu Wenruo <wqu@suse.com> Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Scheduler raises a SCHED_SOFTIRQ to trigger a load balancing event on
from the IPI handler on the idle CPU. If the SMP function is invoked
from an idle CPU via flush_smp_call_function_queue() then the HARD-IRQ
flag is not set and raise_softirq_irqoff() needlessly wakes ksoftirqd
because soft interrupts are handled before ksoftirqd get on the CPU.
Adding a trace_printk() in nohz_csd_func() at the spot of raising
SCHED_SOFTIRQ and enabling trace events for sched_switch, sched_wakeup,
and softirq_entry (for SCHED_SOFTIRQ vector alone) helps observing the
current behavior:
Use __raise_softirq_irqoff() to raise the softirq. The SMP function call
is always invoked on the requested CPU in an interrupt handler. It is
guaranteed that soft interrupts are handled at the end.
Following are the observations with the changes when enabling the same
set of events:
Commit b2a02fc43a1f ("smp: Optimize send_call_function_single_ipi()")
optimizes IPIs to idle CPUs in TIF_POLLING_NRFLAG mode by setting the
TIF_NEED_RESCHED flag in idle task's thread info and relying on
flush_smp_call_function_queue() in idle exit path to run the
call-function. A softirq raised by the call-function is handled shortly
after in do_softirq_post_smp_call_flush() but the TIF_NEED_RESCHED flag
remains set and is only cleared later when schedule_idle() calls
__schedule().
need_resched() check in _nohz_idle_balance() exists to bail out of load
balancing if another task has woken up on the CPU currently in-charge of
idle load balancing which is being processed in SCHED_SOFTIRQ context.
Since the optimization mentioned above overloads the interpretation of
TIF_NEED_RESCHED, check for idle_cpu() before going with the existing
need_resched() check which can catch a genuine task wakeup on an idle
CPU processing SCHED_SOFTIRQ from do_softirq_post_smp_call_flush(), as
well as the case where ksoftirqd needs to be preempted as a result of
new task wakeup or slice expiry.
In case of PREEMPT_RT or threadirqs, although the idle load balancing
may be inhibited in some cases on the ilb CPU, the fact that ksoftirqd
is the only fair task going back to sleep will trigger a newidle balance
on the CPU which will alleviate some imbalance if it exists if idle
balance fails to do so.
A following patch will trigger NOHZ idle balances as a means to update
nohz.next_balance. Vincent noted that blocked load updates can have
non-negligible overhead, which should be avoided if the intent is to only
update nohz.next_balance.
Add a new NOHZ balance kick flag, NOHZ_NEXT_KICK. Gate NOHZ blocked load
update by the presence of NOHZ_STATS_KICK - currently all NOHZ balance
kicks will have the NOHZ_STATS_KICK flag set, so no change in behaviour is
expected.
Suggested-by: Vincent Guittot <vincent.guittot@linaro.org> Signed-off-by: Valentin Schneider <valentin.schneider@arm.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org> Link: https://lkml.kernel.org/r/20210823111700.2842997-2-valentin.schneider@arm.com
Stable-dep-of: ff47a0acfcce ("sched/fair: Check idle_cpu() before need_resched() to detect ilb CPU turning busy") Signed-off-by: Sasha Levin <sashal@kernel.org>
Instead of waking up a random and already idle CPU, we can take advantage
of this_cpu being about to enter idle to run the ILB and update the
blocked load.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Reviewed-by: Valentin Schneider <valentin.schneider@arm.com> Link: https://lkml.kernel.org/r/20210224133007.28644-7-vincent.guittot@linaro.org
Stable-dep-of: ff47a0acfcce ("sched/fair: Check idle_cpu() before need_resched() to detect ilb CPU turning busy") Signed-off-by: Sasha Levin <sashal@kernel.org>
Remove the specific case for handling this_cpu outside for_each_cpu() loop
when running ILB. Instead we use for_each_cpu_wrap() and start with the
next cpu after this_cpu so we will continue to finish with this_cpu.
update_nohz_stats() is now used for this_cpu too and will prevents
unnecessary update. We don't need a special case for handling the update of
nohz.next_balance for this_cpu anymore because it is now handled by the
loop like others.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Reviewed-by: Valentin Schneider <valentin.schneider@arm.com> Link: https://lkml.kernel.org/r/20210224133007.28644-5-vincent.guittot@linaro.org
Stable-dep-of: ff47a0acfcce ("sched/fair: Check idle_cpu() before need_resched() to detect ilb CPU turning busy") Signed-off-by: Sasha Levin <sashal@kernel.org>
newidle_balance runs with both preempt and irq disabled which prevent
local irq to run during this period. The duration for updating the
blocked load of CPUs varies according to the number of CPU cgroups
with non-decayed load and extends this critical period to an uncontrolled
level.
Remove the update from newidle_balance and trigger a normal ILB that
will take care of the update instead.
This reduces the IRQ latency from O(nr_cgroups * nr_nohz_cpus) to
O(nr_cgroups).
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Reviewed-by: Valentin Schneider <valentin.schneider@arm.com> Link: https://lkml.kernel.org/r/20210224133007.28644-2-vincent.guittot@linaro.org
Stable-dep-of: ff47a0acfcce ("sched/fair: Check idle_cpu() before need_resched() to detect ilb CPU turning busy") Signed-off-by: Sasha Levin <sashal@kernel.org>
The need_resched() check currently in nohz_csd_func() can be tracked
to have been added in scheduler_ipi() back in 2011 via commit ca38062e57e9 ("sched: Use resched IPI to kick off the nohz idle balance")
Since then, it has travelled quite a bit but it seems like an idle_cpu()
check currently is sufficient to detect the need to bail out from an
idle load balancing. To justify this removal, consider all the following
case where an idle load balancing could race with a task wakeup:
o Since commit f3dd3f674555b ("sched: Remove the limitation of WF_ON_CPU
on wakelist if wakee cpu is idle") a target perceived to be idle
(target_rq->nr_running == 0) will return true for
ttwu_queue_cond(target) which will offload the task wakeup to the idle
target via an IPI.
In all such cases target_rq->ttwu_pending will be set to 1 before
queuing the wake function.
If an idle load balance races here, following scenarios are possible:
- The CPU is not in TIF_POLLING_NRFLAG mode in which case an actual
IPI is sent to the CPU to wake it out of idle. If the
nohz_csd_func() queues before sched_ttwu_pending(), the idle load
balance will bail out since idle_cpu(target) returns 0 since
target_rq->ttwu_pending is 1. If the nohz_csd_func() is queued after
sched_ttwu_pending() it should see rq->nr_running to be non-zero and
bail out of idle load balancing.
- The CPU is in TIF_POLLING_NRFLAG mode and instead of an actual IPI,
the sender will simply set TIF_NEED_RESCHED for the target to put it
out of idle and flush_smp_call_function_queue() in do_idle() will
execute the call function. Depending on the ordering of the queuing
of nohz_csd_func() and sched_ttwu_pending(), the idle_cpu() check in
nohz_csd_func() should either see target_rq->ttwu_pending = 1 or
target_rq->nr_running to be non-zero if there is a genuine task
wakeup racing with the idle load balance kick.
o The waker CPU perceives the target CPU to be busy
(targer_rq->nr_running != 0) but the CPU is in fact going idle and due
to a series of unfortunate events, the system reaches a case where the
waker CPU decides to perform the wakeup by itself in ttwu_queue() on
the target CPU but target is concurrently selected for idle load
balance (XXX: Can this happen? I'm not sure, but we'll consider the
mother of all coincidences to estimate the worst case scenario).
ttwu_do_activate() calls enqueue_task() which would increment
"rq->nr_running" post which it calls wakeup_preempt() which is
responsible for setting TIF_NEED_RESCHED (via a resched IPI or by
setting TIF_NEED_RESCHED on a TIF_POLLING_NRFLAG idle CPU) The key
thing to note in this case is that rq->nr_running is already non-zero
in case of a wakeup before TIF_NEED_RESCHED is set which would
lead to idle_cpu() check returning false.
In all cases, it seems that need_resched() check is unnecessary when
checking for idle_cpu() first since an impending wakeup racing with idle
load balancer will either set the "rq->ttwu_pending" or indicate a newly
woken task via "rq->nr_running".
Chasing the reason why this check might have existed in the first place,
I came across Peter's suggestion on the fist iteration of Suresh's
patch from 2011 [1] where the condition to raise the SCHED_SOFTIRQ was:
sched_ttwu_do_pending(list);
if (unlikely((rq->idle == current) &&
rq->nohz_balance_kick &&
!need_resched()))
raise_softirq_irqoff(SCHED_SOFTIRQ);
Since the condition to raise the SCHED_SOFIRQ was preceded by
sched_ttwu_do_pending() (which is equivalent of sched_ttwu_pending()) in
the current upstream kernel, the need_resched() check was necessary to
catch a newly queued task. Peter suggested modifying it to:
if (idle_cpu() && rq->nohz_balance_kick && !need_resched())
raise_softirq_irqoff(SCHED_SOFTIRQ);
where idle_cpu() seems to have replaced "rq->idle == current" check.
Even back then, the idle_cpu() check would have been sufficient to catch
a new task being enqueued. Since commit b2a02fc43a1f ("smp: Optimize
send_call_function_single_ipi()") overloads the interpretation of
TIF_NEED_RESCHED for TIF_POLLING_NRFLAG idling, remove the
need_resched() check in nohz_csd_func() to raise SCHED_SOFTIRQ based
on Peter's suggestion.
Fixes: b2a02fc43a1f ("smp: Optimize send_call_function_single_ipi()") Suggested-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lore.kernel.org/r/20241119054432.6405-3-kprateek.nayak@amd.com Signed-off-by: Sasha Levin <sashal@kernel.org>
Add a quirk similar to eeprom_93xx46 to add an extra clock cycle before
reading data from the EEPROM.
The 93Cx6 family of EEPROMs output a "dummy 0 bit" between the writing
of the op-code/address from the host to the EEPROM and the reading of
the actual data from the EEPROM.
More info can be found on page 6 of the AT93C46 datasheet (linked below).
Similar notes are found in other 93xx6 datasheets.
In summary the read operation for a 93Cx6 EEPROM is:
Write to EEPROM: 110[A5-A0] (9 bits)
Read from EEPROM: 0[D15-D0] (17 bits)
Where:
110 is the start bit and READ OpCode
[A5-A0] is the address to read from
0 is a "dummy bit" preceding the actual data
[D15-D0] is the actual data.
Looking at the READ timing diagrams in the 93Cx6 datasheets the dummy
bit should be clocked out on the last address bit clock cycle meaning it
should be discarded naturally.
However, depending on the hardware configuration sometimes this dummy
bit is not discarded. This is the case with Exar PCI UARTs which require
an extra clock cycle between sending the address and reading the data.
On some powermacs `escc` nodes are missing `#size-cells` properties,
which is deprecated and now triggers a warning at boot since commit 045b14ca5c36 ("of: WARN on deprecated #address-cells/#size-cells
handling").
As powermacs boot via prom_init it's possible to add the missing
properties to the device tree during boot, avoiding the warning. Note
that `escc-legacy` nodes are also missing `#size-cells` properties, but
they are skipped by the macio driver, so leave them alone.
Depends-on: 045b14ca5c36 ("of: WARN on deprecated #address-cells/#size-cells handling") Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Reviewed-by: Rob Herring <robh@kernel.org> Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com> Link: https://patch.msgid.link/20241126025710.591683-1-mpe@ellerman.id.au Signed-off-by: Sasha Levin <sashal@kernel.org>
arch/mips/boot/dts/loongson/ls7a-pch.dtsi:68.16-416.5: Warning (interrupt_provider): /bus@10000000/pci@1a000000: '#interrupt-cells' found, but node is not an interrupt provider
arch/mips/boot/dts/loongson/ls7a-pch.dtsi:68.16-416.5: Warning (interrupt_provider): /bus@10000000/pci@1a000000: '#interrupt-cells' found, but node is not an interrupt provider
arch/mips/boot/dts/loongson/loongson64g_4core_ls7a.dtb: Warning (interrupt_map): Failed prerequisite 'interrupt_provider'
And a runtime warning introduced in commit 045b14ca5c36 ("of: WARN on
deprecated #address-cells/#size-cells handling"):
WARNING: CPU: 0 PID: 1 at drivers/of/base.c:106 of_bus_n_addr_cells+0x9c/0xe0
Missing '#address-cells' in /bus@10000000/pci@1a000000/pci_bridge@9,0
The fix is similar to commit d89a415ff8d5 ("MIPS: Loongson64: DTS: Fix PCIe
port nodes for ls7a"), which has fixed the issue for ls2k (despite its
subject mentions ls7a).
Signed-off-by: Xi Ruoyao <xry111@xry111.site> Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
When completion of a USB transaction results in an error condition, this
bit is set by the Host/Device Controller. This bit is set along with the
USBINT bit, if the TD on which the error interrupt occurred also had its
interrupt on complete (IOC) bit set.
UI is set only when IOC set. Add checking UEI to fix miss call
isr_tr_complete_handler() when IOC have not set and transfer error happen.
Acked-by: Peter Chen <peter.chen@kernel.com> Signed-off-by: Xu Yang <xu.yang_2@nxp.com> Link: https://lore.kernel.org/r/20240926022906.473319-1-xu.yang_2@nxp.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
A deadlock may happen since the i3c_master_register() acquires
&i3cbus->lock twice. See the log below.
Use i3cdev->desc->info instead of calling i3c_device_info() to
avoid acquiring the lock twice.
v2:
- Modified the title and commit message
============================================
WARNING: possible recursive locking detected
6.11.0-mainline
--------------------------------------------
init/1 is trying to acquire lock: f1ffff80a6a40dc0 (&i3cbus->lock){++++}-{3:3}, at: i3c_bus_normaluse_lock
but task is already holding lock: f1ffff80a6a40dc0 (&i3cbus->lock){++++}-{3:3}, at: i3c_master_register
other info that might help us debug this:
Possible unsafe locking scenario:
Wangxun FF5xxx NICs are similar to SFxxx, RP1000 and RP2000 NICs. They may
be multi-function devices, but they do not advertise an ACS capability.
But the hardware does isolate FF5xxx functions as though it had an ACS
capability and PCI_ACS_RR and PCI_ACS_CR were set in the ACS Control
register, i.e., all peer-to-peer traffic is directed upstream instead of
being routed internally.
Add ACS quirk for FF5xxx NICs in pci_quirk_wangxun_nic_acs() so the
functions can be in independent IOMMU groups.
The "bus" and "cxl_bus" reset methods reset a device by asserting Secondary
Bus Reset on the bridge leading to the device. These only work if the
device is the only device below the bridge.
Add a sysfs 'reset_subordinate' attribute on bridges that can assert
Secondary Bus Reset regardless of how many devices are below the bridge.
This resets all the devices below a bridge in a single command, including
the locking and config space save/restore that reset methods normally do.
This may be the only way to reset devices that don't support other reset
methods (ACPI, FLR, PM reset, etc).
creating a large files during checkpoint disable until it runs out of
space and then delete it, then remount to enable checkpoint again, and
then unmount the filesystem triggers the f2fs_bug_on as below:
The root cause is: creating large files during disable checkpoint
period results in not enough free segments, so when writing back root
inode will failed in f2fs_enable_checkpoint. When umount the file
system after enabling checkpoint, the root inode is dirty in
f2fs_evict_inode function, which triggers BUG_ON. The steps to
reproduce are as follows:
dd if=/dev/zero of=f2fs.img bs=1M count=55
mount f2fs.img f2fs_dir -o checkpoint=disable:10%
dd if=/dev/zero of=big bs=1M count=50
sync
rm big
mount -o remount,checkpoint=enable f2fs_dir
umount f2fs_dir
Let's redirty inode when there is not free segments during checkpoint
is disable.
Signed-off-by: Qi Han <hanqi@vivo.com> Reviewed-by: Chao Yu <chao@kernel.org> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
When nd_dax is NULL, nd_pfn is consequently NULL as well. Nevertheless,
it is inadvisable to perform pointer arithmetic or address-taking on a
NULL pointer.
Introduce the nd_dax_devinit() function to enhance the code's logic and
improve its readability.
Most drives rewind the tape when the device is reset. Reading and writing
are not allowed until something is done to make the tape position match the
user's expectation (e.g., rewind the tape). Add MTIOCGET and MTLOAD to
operations allowed after reset. MTIOCGET is modified to not touch the tape
if pos_unknown is non-zero. The tape location is known after MTLOAD.
There is NULL pointer issue observed if from Process A where hid device
being added which results in adding a led_cdev addition and later a
another call to access of led_cdev attribute from Process B can result
in NULL pointer issue.
Use mutex led_cdev->led_access to protect access to led->cdev and its
attribute inside brightness_show() and max_brightness_show() and also
update the comment for mutex that it should be used to protect the led
class device fields.
For no forced preemption model kernel, in the scenario where the
expander is connected to 12 high performance SAS SSDs, the following
call trace may occur:
This is because in the hisi_sas driver, both the hardware interrupt
handler and the interrupt thread are executed on the same CPU. In the
performance test scenario, function irq_wait_for_interrupt() will always
return 0 if lots of interrupts occurs and the CPU will be continuously
consumed. As a result, the CPU cannot run the watchdog thread. When the
watchdog time exceeds the specified time, call trace occurs.
To fix it, add cond_resched() to execute the watchdog thread.
Signed-off-by: Yihang Li <liyihang9@huawei.com> Link: https://lore.kernel.org/r/20241008021822.2617339-8-liyihang9@huawei.com Reviewed-by: Xiang Chen <chenxiang66@hisilicon.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Use atomic64_inc_return(&ref) instead of atomic64_add_return(1, &ref)
to use optimized implementation and ease register pressure around
the primitive for targets that implement optimized variant.
When PINCTRL_IMX_SCU was selected by PINCTRL_IMX8DXL or PINCTRL_IMX8QM
combined with COMPILE_TEST on a non-arm platforms, the IMX_SCU
dependency could not be enabled.
The ndev->npinfo pointer in __netpoll_setup() is RCU-protected but is being
accessed directly for a NULL check. While no RCU read lock is held in this
context, we should still use proper RCU primitives for consistency and
correctness.
Replace the direct NULL check with rcu_access_pointer(), which is the
appropriate primitive when only checking for NULL without dereferencing
the pointer. This function provides the necessary ordering guarantees
without requiring RCU read-side protection.
Commit 51183d233b5a ("net/neighbor: Update neigh_dump_info for strict
data checking") added strict checking. The err variable is not cleared,
so if we find no table to dump we will return the validation error even
if user did not want strict checking.
I think the only way to hit this is to send an buggy request, and ask
for a table which doesn't exist, so there's no point treating this
as a real fix. I only noticed it because a syzbot repro depended on it
to trigger another bug.
Since '1 << rocker_port->pport' may be undefined for port >= 32,
cast the left operand to 'unsigned long long' like it's done in
'rocker_port_set_enable()' above. Compile tested only.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
The ordering in hdmi_codec_get_ch_alloc_table_idx() results in
wrong channel allocation for a number of cases, e.g. when ELD
reports FL|FR|LFE|FC|RL|RR or FL|FR|LFE|FC|RL|RR|RC|RLC|RRC:
ca_id 0x01 with speaker mask FL|FR|LFE is selected instead of
ca_id 0x03 with speaker mask FL|FR|LFE|FC for 4 channels
and
ca_id 0x04 with speaker mask FL|FR|RC gets selected instead of
ca_id 0x0b with speaker mask FL|FR|LFE|FC|RL|RR for 6 channels
Fix this by reordering the channel allocation list with most
specific speaker masks at the top.
This patch fixes a NULL pointer dereference bug in brcmfmac that occurs
when a high 'sd_sgentry_align' value applies (e.g. 512) and a lot of queued SKBs
are sent from the pkt queue.
The problem is the number of entries in the pre-allocated sgtable, it is
nents = max(rxglom_size, txglom_size) + max(rxglom_size, txglom_size) >> 4 + 1.
Given the default [rt]xglom_size=32 it's actually 35 which is too small.
Worst case, the pkt queue can end up with 64 SKBs. This occurs when a new SKB
is added for each original SKB if tailroom isn't enough to hold tail_pad.
At least one sg entry is needed for each SKB. So, eventually the "skb_queue_walk loop"
in brcmf_sdiod_sglist_rw may run out of sg entries. This makes sg_next return
NULL and this causes the oops.
The patch sets nents to max(rxglom_size, txglom_size) * 2 to be able handle
the worst-case.
Btw. this requires only 64-35=29 * 16 (or 20 if CONFIG_NEED_SG_DMA_LENGTH) = 464
additional bytes of memory.
./drivers/net/wireless/intel/ipw2x00/libipw_rx.c:871:2-3: code aligned with following code on line 882.
./drivers/net/wireless/intel/ipw2x00/libipw_rx.c:886:2-3: code aligned with following code on line 900.
Reported-by: Abaci Robot <abaci@linux.alibaba.com> Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=11381 Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Signed-off-by: Kalle Valo <kvalo@kernel.org> Link: https://patch.msgid.link/20241101060725.54640-1-jiapeng.chong@linux.alibaba.com Signed-off-by: Sasha Levin <sashal@kernel.org>
The driver needs to set the correct max_segment_size;
otherwise debug_dma_map_sg() will complain about the
over-mapping of the AMDGPU sg length as following:
Under sriov, host driver will save and restore vf pci cfg space during
reset. And during device init, under sriov, pci_restore_state happens after
fullaccess released, and it can have race condition with mmio protection
enable from host side leading to missing interrupts.
So skip amdgpu_device_cache_pci_state for sriov.
Signed-off-by: Victor Zhao <Victor.Zhao@amd.com> Acked-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
radix_lock() shouldn't be held while holding dma_hash_entry[idx].lock
otherwise, there's a possible deadlock scenario when
dma debug API is called holding rq_lock():
CPU1 situation can happen when it extending radix tree and
it tries to wake up kswapd via wake_all_kswapd().
CPU2 situation can happen while perf_event_task_sched_out()
(i.e. dma sync operation is called while deleting perf_event using
etm and etr tmc which are Arm Coresight hwtracing driver backends).
To remove this possible situation, call dma_entry_free() after
put_hash_bucket() in check_unmap().
Reported-by: Denis Nikitin <denik@chromium.org> Closes: https://lists.linaro.org/archives/list/coresight@lists.linaro.org/thread/2WMS7BBSF5OZYB63VT44U5YWLFP5HL6U/#RWM6MLQX5ANBTEQ2PRM7OXCBGCE6NPWU Signed-off-by: Levi Yun <yeoreum.yun@arm.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
projects / thirdparty / kernel / stable.git / log
