The commit mentioned below has split the parameters of ICOSQ and async
ICOSQ, but it contained a typo: the CQ parameters were swapped for ICOSQ
and async ICOSQ. Async ICOSQ is longer than the normal ICOSQ, and the CQ
size must be the same as the size of the corresponding SQ, but due to
this bug, the CQ of async ICOSQ was much shorter than async ICOSQ
itself. It led to overflows of the CQ with such messages in dmesg, in
particular, when running multiple kTLS-offloaded streams:
net/mlx5e: Replace synchronize_rcu with synchronize_net
The commit cited below switched from using napi_synchronize to
synchronize_rcu to have a guarantee that it will finish in finite time.
However, on average, synchronize_rcu takes more time than
napi_synchronize. Given that it's called multiple times per channel on
deactivation, it accumulates to a significant amount, which causes
timeouts in some applications (for example, when using bonding with
NetworkManager).
This commit replaces synchronize_rcu with synchronize_net, which is
faster when called under rtnl_lock, allowing to speed up the described
flow.
Fixes: 9c25a22dfb00 ("net/mlx5e: Use synchronize_rcu to sync with NAPI") Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com> Reviewed-by: Tariq Toukan <tariqt@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Shay Drory [Mon, 23 Nov 2020 06:39:10 +0000 (08:39 +0200)]
net/mlx5: Fix health error state handling
Currently, when we discover a fatal error, we are queueing a work that
will wait for a lock in order to enter the device to error state.
Meanwhile, FW commands are still being processed, and gets timeouts.
This can block the driver for few minutes before the work will manage
to get the lock and enter to error state.
Setting the device to error state before queueing health work, in order
to avoid FW commands being processed while the work is waiting for the
lock.
Fixes: c1d4d2e92ad6 ("net/mlx5: Avoid calling sleeping function by the health poll thread") Signed-off-by: Shay Drory <shayd@nvidia.com> Reviewed-by: Moshe Shemesh <moshe@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
net/mlx5e: Change interrupt moderation channel params also when channels are closed
struct mlx5e_params contains fields ({rx,tx}_cq_moderation) that depend
on two things: whether DIM is enabled and the state of a private flag
(MLX5E_PFLAG_{RX,TX}_CQE_BASED_MODER). Whenever the DIM state changes,
mlx5e_reset_{rx,tx}_moderation is called to update the fields, however,
only if the channels are open. The flow where the channels are closed
misses the required update of the fields. This commit moves the calls of
mlx5e_reset_{rx,tx}_moderation, so that they run in both flows.
net/mlx5e: Don't change interrupt moderation params when DIM is enabled
When mlx5e_ethtool_set_coalesce doesn't change DIM state
(enabled/disabled), it calls mlx5e_set_priv_channels_coalesce
unconditionally, which in turn invokes a firmware command to set
interrupt moderation parameters. It shouldn't happen while DIM manages
those parameters dynamically (it might even be happening at the same
time).
This patch fixes it by splitting mlx5e_set_priv_channels_coalesce into
two functions (for RX and TX) and calling them only when DIM is disabled
(for RX and TX respectively).
Raed Salem [Mon, 25 Jan 2021 07:58:22 +0000 (09:58 +0200)]
net/mlx5e: Enable XDP for Connect-X IPsec capable devices
This limitation was inherited by previous Innova (FPGA) IPsec
implementation, it uses its private set of RQ handlers which
does not support XDP, for Connect-X this is no longer true.
Fix by keeping this limitation only for Innova IPsec supporting devices,
as otherwise this limitation effectively wrongly blocks XDP for all
future Connect-X devices for all flows even if IPsec offload is not
used.
Raed Salem [Sun, 24 Jan 2021 20:40:23 +0000 (22:40 +0200)]
net/mlx5e: Enable striding RQ for Connect-X IPsec capable devices
This limitation was inherited by previous Innova (FPGA) IPsec
implementation, it uses its private set of RQ handlers which does
not support striding rq, for Connect-X this is no longer true.
Fix by keeping this limitation only for Innova IPsec supporting devices,
as otherwise this limitation effectively wrongly blocks striding RQs for
all future Connect-X devices for all flows even if IPsec offload is not
used.
Parav Pandit [Tue, 12 Jan 2021 14:13:22 +0000 (16:13 +0200)]
net/mlx5e: E-switch, Fix rate calculation for overflow
rate_bytes_ps is a 64-bit field. It passed as 32-bit field to
apply_police_params(). Due to this when police rate is higher
than 4Gbps, 32-bit calculation ignores the carry. This results
in incorrect rate configurationn the device.
Paolo Abeni [Thu, 11 Feb 2021 23:30:42 +0000 (15:30 -0800)]
mptcp: add a missing retransmission timer scheduling
Currently we do not schedule the MPTCP retransmission
timer after pushing the data when such action happens
in the subflow context.
This may cause hang-up on active-backup scenarios, or
even when only single subflow msks are involved, if we lost
some peer's ack.
Fixes: 6e628cd3a8f7 ("mptcp: use mptcp release_cb for delayed tasks") Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Thu, 11 Feb 2021 23:30:41 +0000 (15:30 -0800)]
mptcp: better msk receive window updates
Move mptcp_cleanup_rbuf() related checks inside the mentioned
helper and extend them to mirror TCP checks more closely.
Additionally drop the 'rmem_pending' hack, since commit 879526030c8b
("mptcp: protect the rx path with the msk socket spinlock") we
can use instead 'rmem_released'.
Fixes: ea4ca586b16f ("mptcp: refine MPTCP-level ack scheduling") Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Thu, 11 Feb 2021 23:30:40 +0000 (15:30 -0800)]
mptcp: init mptcp request socket earlier
The mptcp subflow route_req() callback performs the subflow
req initialization after the route_req() check. If the latter
fails, mptcp-specific bits of the current request sockets
are left uninitialized.
The above causes bad things at req socket disposal time, when
the mptcp resources are cleared.
This change addresses the issue by splitting subflow_init_req()
into the actual initialization and the mptcp-specific checks.
The initialization is moved before any possibly failing check.
Reported-by: Christoph Paasch <cpaasch@apple.com> Fixes: 7ea851d19b23 ("tcp: merge 'init_req' and 'route_req' functions") Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
The mptcp_worker tries to update the MPTCP retransmission timer
even if such timer is not currently scheduled.
The mptcp_rtx_head() return value is bogus: we can have enqueued
data not yet transmitted. The above may additionally cause spurious,
unneeded MPTCP-level retransmissions.
Fix the issue adding an explicit clearing of the rtx queue before
trying to retransmit and checking for unacked data.
Additionally drop an unneeded timer stop call and the unused
mptcp_rtx_tail() helper.
Reported-by: Christoph Paasch <cpaasch@apple.com> Fixes: 6e628cd3a8f7 ("mptcp: use mptcp release_cb for delayed tasks") Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Thu, 11 Feb 2021 23:30:38 +0000 (15:30 -0800)]
mptcp: fix poll after shutdown
The current mptcp_poll() implementation gives unexpected
results after shutdown(SEND_SHUTDOWN) and when the msk
status is TCP_CLOSE.
Set the correct mask.
Fixes: 8edf08649eed ("mptcp: rework poll+nospace handling") Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Thu, 11 Feb 2021 23:30:37 +0000 (15:30 -0800)]
mptcp: deliver ssk errors to msk
Currently all errors received on msk subflows are ignored.
We need to catch at least the errors on connect() and
on fallback sockets.
Use a custom sk_error_report callback at subflow level,
and do the real action under the msk socket lock - via
the usual sock_owned_by_user()/release_callback() schema.
Fixes: 6e628cd3a8f7 ("mptcp: use mptcp release_cb for delayed tasks") Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit [Thu, 11 Feb 2021 21:32:52 +0000 (22:32 +0100)]
net: phy: consider that suspend2ram may cut off PHY power
Claudiu reported that on his system S2R cuts off power to the PHY and
after resuming certain PHY settings are lost. The PM folks confirmed
that cutting off power to selected components in S2R is a valid case.
Therefore resuming from S2R, same as from hibernation, has to assume
that the PHY has power-on defaults. As a consequence use the restore
callback also as resume callback.
In addition make sure that the interrupt configuration is restored.
Let's do this in phy_init_hw() and ensure that after this call
actual interrupt configuration is in sync with phydev->interrupts.
Currently, if interrupt was enabled before hibernation, we would
resume with interrupt disabled because that's the power-on default.
This fix applies cleanly only after the commit marked as fixed.
I don't have an affected system, therefore change is compile-tested
only.
Ioana Ciornei [Thu, 11 Feb 2021 19:51:22 +0000 (21:51 +0200)]
dpaa2-eth: fix memory leak in XDP_REDIRECT
If xdp_do_redirect() fails, the calling driver should handle recycling
or freeing of the page associated with the frame. The dpaa2-eth driver
didn't do either of them and just incremented a counter.
Fix this by trying to DMA map back the page and recycle it or, if the
mapping fails, just free it.
Fixes: d678be1dc1ec ("dpaa2-eth: add XDP_REDIRECT support") Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Nathan Rossi [Thu, 11 Feb 2021 05:17:57 +0000 (05:17 +0000)]
net: ethernet: aquantia: Handle error cleanup of start on open
The aq_nic_start function can fail in a variety of cases which leaves
the device in broken state.
An example case where the start function fails is the
request_threaded_irq which can be interrupted, resulting in a EINTR
result. This can be manually triggered by bringing the link up (e.g. ip
link set up) and triggering a SIGINT on the initiating process (e.g.
Ctrl+C). This would put the device into a half configured state.
Subsequently bringing the link up again would cause the napi_enable to
BUG.
In order to correctly clean up the failed attempt to start a device call
aq_nic_stop.
Signed-off-by: Nathan Rossi <nathan.rossi@digi.com> Reviewed-by: Igor Russkikh <irusskikh@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 11 Feb 2021 22:36:22 +0000 (14:36 -0800)]
Merge branch 'bnxt_en-fixes'
Michael Chan says:
====================
bnxt_en: 2 bug fixes.
Two unrelated fixes. The first one fixes intermittent false TX timeouts
during ring reconfigurations. The second one fixes a formatting
discrepancy between the stored and the running FW versions.
Please also queue these for -stable. Thanks.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Vasundhara Volam [Thu, 11 Feb 2021 07:24:24 +0000 (02:24 -0500)]
bnxt_en: Fix devlink info's stored fw.psid version format.
The running fw.psid version is in decimal format but the stored
fw.psid is in hex format. This can mislead the user to reset the
NIC to activate the stored version to become the running version.
Fix it to display the stored fw.psid in decimal format.
Fixes: 1388875b3916 ("bnxt_en: Add stored FW version info to devlink info_get cb.") Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com> Signed-off-by: Michael Chan <michael.chan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Edwin Peer [Thu, 11 Feb 2021 07:24:23 +0000 (02:24 -0500)]
bnxt_en: reverse order of TX disable and carrier off
A TX queue can potentially immediately timeout after it is stopped
and the last TX timestamp on that queue was more than 5 seconds ago with
carrier still up. Prevent these intermittent false TX timeouts
by bringing down carrier first before calling netif_tx_disable().
Fixes: c0c050c58d84 ("bnxt_en: New Broadcom ethernet driver.") Signed-off-by: Edwin Peer <edwin.peer@broadcom.com> Signed-off-by: Michael Chan <michael.chan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
If set_link_state() fails for any reason, we still cleanup the adapter
state and cannot recover from a partial close anyway. So set the adapter
to CLOSED state. That way if a new soft/hard reset is processed, the
adapter will remain in the CLOSED state until the next ibmvnic_open().
Fixes: 01d9bd792d16 ("ibmvnic: Reorganize device close") Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.ibm.com> Reported-by: Abdul Haleem <abdhalee@in.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Wed, 10 Feb 2021 17:13:33 +0000 (09:13 -0800)]
tcp: fix tcp_rmem documentation
tcp_rmem[1] has been changed to 131072, we should update the documentation
to reflect this.
Fixes: a337531b942b ("tcp: up initial rmem to 128KB and SYN rwin to around 64KB") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Zhibin Liu <zhibinliu@google.com> Cc: Yuchung Cheng <ycheng@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Linus Torvalds [Wed, 10 Feb 2021 20:03:35 +0000 (12:03 -0800)]
Merge tag 'pm-5.11-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull power management fixes from Rafael Wysocki:
"Address a performance regression related to scale-invariance on x86
that may prevent turbo CPU frequencies from being used in certain
workloads on systems using acpi-cpufreq as the CPU performance scaling
driver and schedutil as the scaling governor"
* tag 'pm-5.11-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
cpufreq: ACPI: Update arch scale-invariance max perf ratio if CPPC is not there
cpufreq: ACPI: Extend frequency tables to cover boost frequencies
Linus Torvalds [Wed, 10 Feb 2021 19:58:21 +0000 (11:58 -0800)]
Merge tag 'acpi-5.11-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull ACPI fix from Rafael Wysocki:
"Revert a problematic ACPICA commit that changed the code to attempt to
update memory regions which may be read-only on some systems (Ard
Biesheuvel)"
* tag 'acpi-5.11-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
Revert "ACPICA: Interpreter: fix memory leak by using existing buffer"
Pull networking fixes from David Miller:
"Another pile of networing fixes:
1) ath9k build error fix from Arnd Bergmann
2) dma memory leak fix in mediatec driver from Lorenzo Bianconi.
3) bpf int3 kprobe fix from Alexei Starovoitov.
4) bpf stackmap integer overflow fix from Bui Quang Minh.
5) Add usb device ids for Cinterion MV31 to qmi_qwwan driver, from
Christoph Schemmel.
6) Don't update deleted entry in xt_recent netfilter module, from
Jazsef Kadlecsik.
7) Use after free in nftables, fix from Pablo Neira Ayuso.
8) Header checksum fix in flowtable from Sven Auhagen.
9) Validate user controlled length in qrtr code, from Sabyrzhan
Tasbolatov.
10) Fix race in xen/netback, from Juergen Gross,
11) New device ID in cxgb4, from Raju Rangoju.
12) Fix ring locking in rxrpc release call, from David Howells.
13) Don't return LAPB error codes from x25_open(), from Xie He.
14) Missing error returns in gsi_channel_setup() from Alex Elder.
15) Get skb_copy_and_csum_datagram working properly with odd segment
sizes, from Willem de Bruijn.
16) Missing RFS/RSS table init in enetc driver, from Vladimir Oltean.
17) Do teardown on probe failure in DSA, from Vladimir Oltean.
18) Fix compilation failures of txtimestamp selftest, from Vadim
Fedorenko.
19) Limit rx per-napi gro queue size to fix latency regression, from
Eric Dumazet.
20) dpaa_eth xdp fixes from Camelia Groza.
21) Missing txq mode update when switching CBS off, in stmmac driver,
from Mohammad Athari Bin Ismail.
22) Failover pending logic fix in ibmvnic driver, from Sukadev
Bhattiprolu.
23) Null deref fix in vmw_vsock, from Norbert Slusarek.
24) Missing verdict update in xdp paths of ena driver, from Shay
Agroskin.
25) seq_file iteration fix in sctp from Neil Brown.
26) bpf 32-bit src register truncation fix on div/mod, from Daniel
Borkmann.
27) Fix jmp32 pruning in bpf verifier, from Daniel Borkmann.
28) Fix locking in vsock_shutdown(), from Stefano Garzarella.
29) Various missing index bound checks in hns3 driver, from Yufeng Mo.
30) Flush ports on .phylink_mac_link_down() in dsa felix driver, from
Vladimir Oltean.
31) Don't mix up stp and mrp port states in bridge layer, from Horatiu
Vultur.
32) Fix locking during netif_tx_disable(), from Edwin Peer"
* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (45 commits)
bpf: Fix 32 bit src register truncation on div/mod
bpf: Fix verifier jmp32 pruning decision logic
bpf: Fix verifier jsgt branch analysis on max bound
vsock: fix locking in vsock_shutdown()
net: hns3: add a check for index in hclge_get_rss_key()
net: hns3: add a check for tqp_index in hclge_get_ring_chain_from_mbx()
net: hns3: add a check for queue_id in hclge_reset_vf_queue()
net: dsa: felix: implement port flushing on .phylink_mac_link_down
switchdev: mrp: Remove SWITCHDEV_ATTR_ID_MRP_PORT_STAT
bridge: mrp: Fix the usage of br_mrp_port_switchdev_set_state
net: watchdog: hold device global xmit lock during tx disable
netfilter: nftables: relax check for stateful expressions in set definition
netfilter: conntrack: skip identical origin tuple in same zone only
vsock/virtio: update credit only if socket is not closed
net: fix iteration for sctp transport seq_files
net: ena: Update XDP verdict upon failure
net/vmw_vsock: improve locking in vsock_connect_timeout()
net/vmw_vsock: fix NULL pointer dereference
ibmvnic: Clear failover_pending if unable to schedule
net: stmmac: set TxQ mode back to DCB after disabling CBS
...
Linus Torvalds [Wed, 10 Feb 2021 19:22:41 +0000 (11:22 -0800)]
Merge branch 'akpm' (patches from Andrew)
Merge misc fixes from Andrew Morton:
"14 patches.
Subsystems affected by this patch series: mm (kasan, mremap, tmpfs,
selftests, memcg, and slub), MAINTAINERS, squashfs, nilfs2, and
firmware"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
nilfs2: make splice write available again
mm, slub: better heuristic for number of cpus when calculating slab order
Revert "mm: memcontrol: avoid workload stalls when lowering memory.high"
MAINTAINERS: update Andrey Ryabinin's email address
selftests/vm: rename file run_vmtests to run_vmtests.sh
tmpfs: disallow CONFIG_TMPFS_INODE64 on alpha
tmpfs: disallow CONFIG_TMPFS_INODE64 on s390
mm/mremap: fix BUILD_BUG_ON() error in get_extent
firmware_loader: align .builtin_fw to 8
kasan: fix stack traces dependency for HW_TAGS
squashfs: add more sanity checks in xattr id lookup
squashfs: add more sanity checks in inode lookup
squashfs: add more sanity checks in id lookup
squashfs: avoid out of bounds writes in decompressors
Joachim Henke [Tue, 9 Feb 2021 21:42:36 +0000 (13:42 -0800)]
nilfs2: make splice write available again
Since 5.10, splice() or sendfile() to NILFS2 return EINVAL. This was
caused by commit 36e2c7421f02 ("fs: don't allow splice read/write
without explicit ops").
This patch initializes the splice_write field in file_operations, like
most file systems do, to restore the functionality.
Vlastimil Babka [Tue, 9 Feb 2021 21:42:32 +0000 (13:42 -0800)]
mm, slub: better heuristic for number of cpus when calculating slab order
When creating a new kmem cache, SLUB determines how large the slab pages
will based on number of inputs, including the number of CPUs in the
system. Larger slab pages mean that more objects can be allocated/free
from per-cpu slabs before accessing shared structures, but also
potentially more memory can be wasted due to low slab usage and
fragmentation. The rough idea of using number of CPUs is that larger
systems will be more likely to benefit from reduced contention, and also
should have enough memory to spare.
Number of CPUs used to be determined as nr_cpu_ids, which is number of
possible cpus, but on some systems many will never be onlined, thus
commit 045ab8c9487b ("mm/slub: let number of online CPUs determine the
slub page order") changed it to nr_online_cpus(). However, for kmem
caches created early before CPUs are onlined, this may lead to
permamently low slab page sizes.
Vincent reports a regression [1] of hackbench on arm64 systems:
"I'm facing significant performances regression on a large arm64
server system (224 CPUs). Regressions is also present on small arm64
system (8 CPUs) but in a far smaller order of magnitude
On 224 CPUs system : 9 iterations of hackbench -l 16000 -g 16
v5.11-rc4 : 9.135sec (+/- 0.45%)
v5.11-rc4 + revert this patch: 3.173sec (+/- 0.48%)
v5.10: 3.136sec (+/- 0.40%)"
Mel reports a regression [2] of hackbench on x86_64, with lockstat suggesting
page allocator contention:
"i.e. the patch incurs a 7% to 32% performance penalty. This bisected
cleanly yesterday when I was looking for the regression and then
found the thread.
Numerous caches change size. For example, kmalloc-512 goes from
order-0 (vanilla) to order-2 with the revert.
So mostly this is down to the number of times SLUB calls into the
page allocator which only caches order-0 pages on a per-cpu basis"
Clearly num_online_cpus() doesn't work too early in bootup. We could
change the order dynamically in a memory hotplug callback, but runtime
order changing for existing kmem caches has been already shown as
dangerous, and removed in 32a6f409b693 ("mm, slub: remove runtime
allocation order changes").
It could be resurrected in a safe manner with some effort, but to fix
the regression we need something simpler.
We could use num_present_cpus() that should be the number of physically
present CPUs even before they are onlined. That would work for PowerPC
[3], which triggered the original commit, but that still doesn't work on
arm64 [4] as explained in [5].
So this patch tries to determine the best available value without
specific arch knowledge.
- num_present_cpus() if the number is larger than 1, as that means the
arch is likely setting it properly
- nr_cpu_ids otherwise
This should fix the reported regressions while also keeping the effect
of 045ab8c9487b for PowerPC systems. It's possible there are
configurations where num_present_cpus() is 1 during boot while
nr_cpu_ids is at the same time bloated, so these (if they exist) would
keep the large orders based on nr_cpu_ids as was before 045ab8c9487b.
Johannes Weiner [Tue, 9 Feb 2021 21:42:28 +0000 (13:42 -0800)]
Revert "mm: memcontrol: avoid workload stalls when lowering memory.high"
This reverts commit 536d3bf261a2fc3b05b3e91e7eef7383443015cf, as it can
cause writers to memory.high to get stuck in the kernel forever,
performing page reclaim and consuming excessive amounts of CPU cycles.
Before the patch, a write to memory.high would first put the new limit
in place for the workload, and then reclaim the requested delta. After
the patch, the kernel tries to reclaim the delta before putting the new
limit into place, in order to not overwhelm the workload with a sudden,
large excess over the limit. However, if reclaim is actively racing
with new allocations from the uncurbed workload, it can keep the write()
working inside the kernel indefinitely.
This is causing problems in Facebook production. A privileged
system-level daemon that adjusts memory.high for various workloads
running on a host can get unexpectedly stuck in the kernel and
essentially turn into a sort of involuntary kswapd for one of the
workloads. We've observed that daemon busy-spin in a write() for
minutes at a time, neglecting its other duties on the system, and
expending privileged system resources on behalf of a workload.
To remedy this, we have first considered changing the reclaim logic to
break out after a couple of loops - whether the workload has converged
to the new limit or not - and bound the write() call this way. However,
the root cause that inspired the sequence change in the first place has
been fixed through other means, and so a revert back to the proven
limit-setting sequence, also used by memory.max, is preferable.
The sequence was changed to avoid extreme latencies in the workload when
the limit was lowered: the sudden, large excess created by the limit
lowering would erroneously trigger the penalty sleeping code that is
meant to throttle excessive growth from below. Allocating threads could
end up sleeping long after the write() had already reclaimed the delta
for which they were being punished.
However, erroneous throttling also caused problems in other scenarios at
around the same time. This resulted in commit b3ff92916af3 ("mm, memcg:
reclaim more aggressively before high allocator throttling"), included
in the same release as the offending commit. When allocating threads
now encounter large excess caused by a racing write() to memory.high,
instead of entering punitive sleeps, they will simply be tasked with
helping reclaim down the excess, and will be held no longer than it
takes to accomplish that. This is in line with regular limit
enforcement - i.e. if the workload allocates up against or over an
otherwise unchanged limit from below.
With the patch breaking userspace, and the root cause addressed by other
means already, revert it again.
Link: https://lkml.kernel.org/r/20210122184341.292461-1-hannes@cmpxchg.org Fixes: 536d3bf261a2 ("mm: memcontrol: avoid workload stalls when lowering memory.high") Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reported-by: Tejun Heo <tj@kernel.org> Acked-by: Chris Down <chris@chrisdown.name> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Roman Gushchin <guro@fb.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: Michal Koutný <mkoutny@suse.com> Cc: <stable@vger.kernel.org> [5.8+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Seth Forshee [Tue, 9 Feb 2021 21:42:17 +0000 (13:42 -0800)]
tmpfs: disallow CONFIG_TMPFS_INODE64 on alpha
As with s390, alpha is a 64-bit architecture with a 32-bit ino_t. With
CONFIG_TMPFS_INODE64=y tmpfs mounts will get 64-bit inode numbers and
display "inode64" in the mount options, whereas passing "inode64" in the
mount options will fail. This leads to erroneous behaviours such as
this:
# mkdir mnt
# mount -t tmpfs nodev mnt
# mount -o remount,rw mnt
mount: /home/ubuntu/mnt: mount point not mounted or bad option.
Prevent CONFIG_TMPFS_INODE64 from being selected on alpha.
Link: https://lkml.kernel.org/r/20210208215726.608197-1-seth.forshee@canonical.com Fixes: ea3271f7196c ("tmpfs: support 64-bit inums per-sb") Signed-off-by: Seth Forshee <seth.forshee@canonical.com> Acked-by: Hugh Dickins <hughd@google.com> Cc: Chris Down <chris@chrisdown.name> Cc: Amir Goldstein <amir73il@gmail.com> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Cc: <stable@vger.kernel.org> [5.9+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Seth Forshee [Tue, 9 Feb 2021 21:42:14 +0000 (13:42 -0800)]
tmpfs: disallow CONFIG_TMPFS_INODE64 on s390
Currently there is an assumption in tmpfs that 64-bit architectures also
have a 64-bit ino_t. This is not true on s390 which has a 32-bit ino_t.
With CONFIG_TMPFS_INODE64=y tmpfs mounts will get 64-bit inode numbers
and display "inode64" in the mount options, but passing the "inode64"
mount option will fail. This leads to the following behavior:
# mkdir mnt
# mount -t tmpfs nodev mnt
# mount -o remount,rw mnt
mount: /home/ubuntu/mnt: mount point not mounted or bad option.
As mount sees "inode64" in the mount options and thus passes it in the
options for the remount.
So prevent CONFIG_TMPFS_INODE64 from being selected on s390.
Link: https://lkml.kernel.org/r/20210205230620.518245-1-seth.forshee@canonical.com Fixes: ea3271f7196c ("tmpfs: support 64-bit inums per-sb") Signed-off-by: Seth Forshee <seth.forshee@canonical.com> Acked-by: Hugh Dickins <hughd@google.com> Cc: Chris Down <chris@chrisdown.name> Cc: Hugh Dickins <hughd@google.com> Cc: Amir Goldstein <amir73il@gmail.com> Cc: Heiko Carstens <hca@linux.ibm.com> Cc: Vasily Gorbik <gor@linux.ibm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: <stable@vger.kernel.org> [5.9+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fangrui Song [Tue, 9 Feb 2021 21:42:07 +0000 (13:42 -0800)]
firmware_loader: align .builtin_fw to 8
arm64 references the start address of .builtin_fw (__start_builtin_fw)
with a pair of R_AARCH64_ADR_PREL_PG_HI21/R_AARCH64_LDST64_ABS_LO12_NC
relocations. The compiler is allowed to emit the
R_AARCH64_LDST64_ABS_LO12_NC relocation because struct builtin_fw in
include/linux/firmware.h is 8-byte aligned.
The R_AARCH64_LDST64_ABS_LO12_NC relocation requires the address to be a
multiple of 8, which may not be the case if .builtin_fw is empty.
Unconditionally align .builtin_fw to fix the linker error. 32-bit
architectures could use ALIGN(4) but that would add unnecessary
complexity, so just use ALIGN(8).
Link: https://lkml.kernel.org/r/20201208054646.2913063-1-maskray@google.com Link: https://github.com/ClangBuiltLinux/linux/issues/1204 Fixes: 5658c76 ("firmware: allow firmware files to be built into kernel image") Signed-off-by: Fangrui Song <maskray@google.com> Reported-by: kernel test robot <lkp@intel.com> Acked-by: Arnd Bergmann <arnd@arndb.de> Reviewed-by: Nick Desaulniers <ndesaulniers@google.com> Tested-by: Nick Desaulniers <ndesaulniers@google.com> Tested-by: Douglas Anderson <dianders@chromium.org> Acked-by: Nathan Chancellor <nathan@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently, whether the alloc/free stack traces collection is enabled by
default for hardware tag-based KASAN depends on CONFIG_DEBUG_KERNEL.
The intention for this dependency was to only enable collection on slow
debug kernels due to a significant perf and memory impact.
As it turns out, CONFIG_DEBUG_KERNEL is not considered a debug option
and is enabled on many productions kernels including Android and Ubuntu.
As the result, this dependency is pointless and only complicates the
code and documentation.
Having stack traces collection disabled by default would make the
hardware mode work differently to to the software ones, which is
confusing.
This change removes the dependency and enables stack traces collection
by default.
Looking into the future, this default might makes sense for production
kernels, assuming we implement a fast stack trace collection approach.
Link: https://lkml.kernel.org/r/6678d77ceffb71f1cff2cf61560e2ffe7bb6bfe9.1612808820.git.andreyknvl@google.com Signed-off-by: Andrey Konovalov <andreyknvl@google.com> Reviewed-by: Marco Elver <elver@google.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Vincenzo Frascino <vincenzo.frascino@arm.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Alexander Potapenko <glider@google.com> Cc: Will Deacon <will.deacon@arm.com> Cc: Andrey Ryabinin <aryabinin@virtuozzo.com> Cc: Peter Collingbourne <pcc@google.com> Cc: Evgenii Stepanov <eugenis@google.com> Cc: Branislav Rankov <Branislav.Rankov@arm.com> Cc: Kevin Brodsky <kevin.brodsky@arm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Phillip Lougher [Tue, 9 Feb 2021 21:42:00 +0000 (13:42 -0800)]
squashfs: add more sanity checks in xattr id lookup
Sysbot has reported a warning where a kmalloc() attempt exceeds the
maximum limit. This has been identified as corruption of the xattr_ids
count when reading the xattr id lookup table.
This patch adds a number of additional sanity checks to detect this
corruption and others.
1. It checks for a corrupted xattr index read from the inode. This could
be because the metadata block is uncompressed, or because the
"compression" bit has been corrupted (turning a compressed block
into an uncompressed block). This would cause an out of bounds read.
2. It checks against corruption of the xattr_ids count. This can either
lead to the above kmalloc failure, or a smaller than expected
table to be read.
3. It checks the contents of the index table for corruption.
Phillip Lougher [Tue, 9 Feb 2021 21:41:56 +0000 (13:41 -0800)]
squashfs: add more sanity checks in inode lookup
Sysbot has reported an "slab-out-of-bounds read" error which has been
identified as being caused by a corrupted "ino_num" value read from the
inode. This could be because the metadata block is uncompressed, or
because the "compression" bit has been corrupted (turning a compressed
block into an uncompressed block).
This patch adds additional sanity checks to detect this, and the
following corruption.
1. It checks against corruption of the inodes count. This can either
lead to a larger table to be read, or a smaller than expected
table to be read.
In the case of a too large inodes count, this would often have been
trapped by the existing sanity checks, but this patch introduces
a more exact check, which can identify too small values.
2. It checks the contents of the index table for corruption.
Phillip Lougher [Tue, 9 Feb 2021 21:41:53 +0000 (13:41 -0800)]
squashfs: add more sanity checks in id lookup
Sysbot has reported a number of "slab-out-of-bounds reads" and
"use-after-free read" errors which has been identified as being caused
by a corrupted index value read from the inode. This could be because
the metadata block is uncompressed, or because the "compression" bit has
been corrupted (turning a compressed block into an uncompressed block).
This patch adds additional sanity checks to detect this, and the
following corruption.
1. It checks against corruption of the ids count. This can either
lead to a larger table to be read, or a smaller than expected
table to be read.
In the case of a too large ids count, this would often have been
trapped by the existing sanity checks, but this patch introduces
a more exact check, which can identify too small values.
2. It checks the contents of the index table for corruption.
Phillip Lougher [Tue, 9 Feb 2021 21:41:50 +0000 (13:41 -0800)]
squashfs: avoid out of bounds writes in decompressors
Patch series "Squashfs: fix BIO migration regression and add sanity checks".
Patch [1/4] fixes a regression introduced by the "migrate from
ll_rw_block usage to BIO" patch, which has produced a number of
Sysbot/Syzkaller reports.
Patches [2/4], [3/4], and [4/4] fix a number of filesystem corruption
issues which have produced Sysbot reports in the id, inode and xattr
lookup code.
Each patch has been tested against the Sysbot reproducers using the
given kernel configuration. They have the appropriate "Reported-by:"
lines added.
Additionally, all of the reproducer filesystems are indirectly fixed by
patch [4/4] due to the fact they all have xattr corruption which is now
detected there.
Additional testing with other configurations and architectures (32bit,
big endian), and normal filesystems has also been done to trap any
inadvertent regressions caused by the additional sanity checks.
This patch (of 4):
This is a regression introduced by the patch "migrate from ll_rw_block
usage to BIO".
Sysbot/Syskaller has reported a number of "out of bounds writes" and
"unable to handle kernel paging request in squashfs_decompress" errors
which have been identified as a regression introduced by the above
patch.
Specifically, the patch removed the following sanity check
1. It ensured any reads were not beyond the end of the filesystem
2. It ensured that the "length" field read from the filesystem
was within the expected maximum length. Without this any
corrupted values can over-run allocated buffers.
Linus Torvalds [Wed, 10 Feb 2021 01:19:56 +0000 (17:19 -0800)]
Merge tag 'i3c/fixes-for-5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux
Pull i3c fix from Alexandre Belloni:
"A single build warning fix"
* tag 'i3c/fixes-for-5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux:
i3c/master/mipi-i3c-hci: Fix position of __maybe_unused in i3c_hci_of_match
Daniel Borkmann [Tue, 9 Feb 2021 18:46:10 +0000 (18:46 +0000)]
bpf: Fix 32 bit src register truncation on div/mod
While reviewing a different fix, John and I noticed an oddity in one of the
BPF program dumps that stood out, for example:
# bpftool p d x i 13
0: (b7) r0 = 808464450
1: (b4) w4 = 808464432
2: (bc) w0 = w0
3: (15) if r0 == 0x0 goto pc+1
4: (9c) w4 %= w0
[...]
In line 2 we noticed that the mov32 would 32 bit truncate the original src
register for the div/mod operation. While for the two operations the dst
register is typically marked unknown e.g. from adjust_scalar_min_max_vals()
the src register is not, and thus verifier keeps tracking original bounds,
simplified:
Runtime result of r0 at exit is 0 instead of expected -1. Remove the
verifier mov32 src rewrite in div/mod and replace it with a jmp32 test
instead. After the fix, we result in the following code generation when
having dividend r1 and divisor r6:
x86 in particular can throw a 'divide error' exception for div
instruction not only for divisor being zero, but also for the case
when the quotient is too large for the designated register. For the
edx:eax and rdx:rax dividend pair it is not an issue in x86 BPF JIT
since we always zero edx (rdx). Hence really the only protection
needed is against divisor being zero.
Fixes: 68fda450a7df ("bpf: fix 32-bit divide by zero") Co-developed-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: John Fastabend <john.fastabend@gmail.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Alexei Starovoitov <ast@kernel.org>
from 6 to 7: safe
processed 15 insns (limit 1000000) max_states_per_insn 0 total_states 1 peak_states 1 mark_read 1
The underlying program was xlated as follows:
# bpftool p d x i 10
0: (b7) r0 = 808464450
1: (b4) w4 = 808464432
2: (bc) w0 = w0
3: (15) if r0 == 0x0 goto pc+1
4: (9c) w4 %= w0
5: (66) if w4 s> 0x30303030 goto pc+0
6: (7f) r0 >>= r0
7: (bc) w0 = w0
8: (15) if r0 == 0x0 goto pc+1
9: (9c) w4 %= w0
10: (66) if w0 s> 0x3030 goto pc+0
11: (d6) if w0 s<= 0x303030 goto pc+1
12: (05) goto pc-1
13: (95) exit
The verifier rewrote original instructions it recognized as dead code with
'goto pc-1', but reality differs from verifier simulation in that we are
actually able to trigger a hang due to hitting the 'goto pc-1' instructions.
Taking a closer look at the verifier analysis, the reason is that it misjudges
its pruning decision at the first 'from 6 to 7: safe' occasion. What happens
is that while both old/cur registers are marked as precise, they get misjudged
for the jmp32 case as range_within() yields true, meaning that the prior
verification path with a wider register bound could be verified successfully
and therefore the current path with a narrower register bound is deemed safe
as well whereas in reality it's not. R0 old/cur path's bounds compare as
follows:
The 64 bit bounds generally look okay and while the information that got
propagated from 32 to 64 bit looks correct as well, it's not precise enough
for judging a conditional jmp32. Given the latter only operates on subregisters
we also need to take these into account as well for a range_within() probe
in order to be able to prune paths. Extending the range_within() constraint
to both bounds will be able to tell us that the old signed 32 bit bounds are
not wider than the cur signed 32 bit bounds.
With the fix in place, the program will now verify the 'goto' branch case as
it should have been:
The bug is quite subtle in the sense that when verifier would determine that
a given branch is dead code, it would (here: wrongly) remove these instructions
from the program and hard-wire the taken branch for privileged programs instead
of the 'goto pc-1' rewrites which will cause hard to debug problems.
Fixes: 3f50f132d840 ("bpf: Verifier, do explicit ALU32 bounds tracking") Reported-by: Anatoly Trosinenko <anatoly.trosinenko@gmail.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Reviewed-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Alexei Starovoitov <ast@kernel.org>
Daniel Borkmann [Fri, 5 Feb 2021 16:20:14 +0000 (17:20 +0100)]
bpf: Fix verifier jsgt branch analysis on max bound
Fix incorrect is_branch{32,64}_taken() analysis for the jsgt case. The return
code for both will tell the caller whether a given conditional jump is taken
or not, e.g. 1 means branch will be taken [for the involved registers] and the
goto target will be executed, 0 means branch will not be taken and instead we
fall-through to the next insn, and last but not least a -1 denotes that it is
not known at verification time whether a branch will be taken or not. Now while
the jsgt has the branch-taken case correct with reg->s32_min_value > sval, the
branch-not-taken case is off-by-one when testing for reg->s32_max_value < sval
since the branch will also be taken for reg->s32_max_value == sval. The jgt
branch analysis, for example, gets this right.
Fixes: 3f50f132d840 ("bpf: Verifier, do explicit ALU32 bounds tracking") Fixes: 4f7b3e82589e ("bpf: improve verifier branch analysis") Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Reviewed-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Alexei Starovoitov <ast@kernel.org>
In vsock_shutdown() we touched some socket fields without holding the
socket lock, such as 'state' and 'sk_flags'.
Also, after the introduction of multi-transport, we are accessing
'vsk->transport' in vsock_send_shutdown() without holding the lock
and this call can be made while the connection is in progress, so
the transport can change in the meantime.
To avoid issues, we hold the socket lock when we enter in
vsock_shutdown() and release it when we leave.
Among the transports that implement the 'shutdown' callback, only
hyperv_transport acquired the lock. Since the caller now holds it,
we no longer take it.
Fixes: d021c344051a ("VSOCK: Introduce VM Sockets") Signed-off-by: Stefano Garzarella <sgarzare@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yufeng Mo [Tue, 9 Feb 2021 09:03:07 +0000 (17:03 +0800)]
net: hns3: add a check for index in hclge_get_rss_key()
The index is received from vf, if use it directly,
an out-of-bound issue may be caused, so add a check for
this index before using it in hclge_get_rss_key().
Fixes: a638b1d8cc87 ("net: hns3: fix get VF RSS issue") Signed-off-by: Yufeng Mo <moyufeng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yufeng Mo [Tue, 9 Feb 2021 09:03:06 +0000 (17:03 +0800)]
net: hns3: add a check for tqp_index in hclge_get_ring_chain_from_mbx()
The tqp_index is received from vf, if use it directly,
an out-of-bound issue may be caused, so add a check for
this tqp_index before using it in hclge_get_ring_chain_from_mbx().
Fixes: 84e095d64ed9 ("net: hns3: Change PF to add ring-vect binding & resetQ to mailbox") Signed-off-by: Yufeng Mo <moyufeng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yufeng Mo [Tue, 9 Feb 2021 09:03:05 +0000 (17:03 +0800)]
net: hns3: add a check for queue_id in hclge_reset_vf_queue()
The queue_id is received from vf, if use it directly,
an out-of-bound issue may be caused, so add a check for
this queue_id before using it in hclge_reset_vf_queue().
Fixes: 1a426f8b40fc ("net: hns3: fix the VF queue reset flow error") Signed-off-by: Yufeng Mo <moyufeng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Vladimir Oltean [Mon, 8 Feb 2021 17:36:27 +0000 (19:36 +0200)]
net: dsa: felix: implement port flushing on .phylink_mac_link_down
There are several issues which may be seen when the link goes down while
forwarding traffic, all of which can be attributed to the fact that the
port flushing procedure from the reference manual was not closely
followed.
With flow control enabled on both the ingress port and the egress port,
it may happen when a link goes down that Ethernet packets are in flight.
In flow control mode, frames are held back and not dropped. When there
is enough traffic in flight (example: iperf3 TCP), then the ingress port
might enter congestion and never exit that state. This is a problem,
because it is the egress port's link that went down, and that has caused
the inability of the ingress port to send packets to any other port.
This is solved by flushing the egress port's queues when it goes down.
There is also a problem when performing stream splitting for
IEEE 802.1CB traffic (not yet upstream, but a sort of multicast,
basically). There, if one port from the destination ports mask goes
down, splitting the stream towards the other destinations will no longer
be performed. This can be traced down to this line:
Basically only DEV_MAC_ENA_CFG_RX_ENA should be disabled, but not
DEV_MAC_ENA_CFG_TX_ENA - I don't have further insight into why that is
the case, but apparently multicasting to several ports will cause issues
if at least one of them doesn't have DEV_MAC_ENA_CFG_TX_ENA set.
I am not sure what the state of the Ocelot VSC7514 driver is, but
probably not as bad as Felix/Seville, since VSC7514 uses phylib and has
the following in ocelot_adjust_link:
if (!phydev->link)
return;
therefore the port is not really put down when the link is lost, unlike
the DSA drivers which use .phylink_mac_link_down for that.
Nonetheless, I put ocelot_port_flush() in the common ocelot.c because it
needs to access some registers from drivers/net/ethernet/mscc/ocelot_rew.h
which are not exported in include/soc/mscc/ and a bugfix patch should
probably not move headers around.
Fixes: bdeced75b13f ("net: dsa: felix: Add PCS operations for PHYLINK") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Based on the discussion here[1], there was a problem with the function
br_mrp_port_switchdev_set_state. The problem was that it was called
both with BR_STATE* and BR_MRP_PORT_STATE* types. This patch series
fixes this issue and removes SWITCHDEV_ATTR_ID_MRP_PORT_STAT because
is not used anymore.
Now that MRP started to use also SWITCHDEV_ATTR_ID_PORT_STP_STATE to
notify HW, then SWITCHDEV_ATTR_ID_MRP_PORT_STAT is not used anywhere
else, therefore we can remove it.
Fixes: c284b545900830 ("switchdev: mrp: Extend switchdev API to offload MRP") Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Horatiu Vultur [Sat, 6 Feb 2021 21:47:33 +0000 (22:47 +0100)]
bridge: mrp: Fix the usage of br_mrp_port_switchdev_set_state
The function br_mrp_port_switchdev_set_state was called both with MRP
port state and STP port state, which is an issue because they don't
match exactly.
Therefore, update the function to be used only with STP port state and
use the id SWITCHDEV_ATTR_ID_PORT_STP_STATE.
The choice of using STP over MRP is that the drivers already implement
SWITCHDEV_ATTR_ID_PORT_STP_STATE and already in SW we update the port
STP state.
Fixes: 9a9f26e8f7ea30 ("bridge: mrp: Connect MRP API with the switchdev API") Fixes: fadd409136f0f2 ("bridge: switchdev: mrp: Implement MRP API for switchdev") Fixes: 2f1a11ae11d222 ("bridge: mrp: Add MRP interface.") Reported-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk> Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Edwin Peer [Sat, 6 Feb 2021 01:37:32 +0000 (17:37 -0800)]
net: watchdog: hold device global xmit lock during tx disable
Prevent netif_tx_disable() running concurrently with dev_watchdog() by
taking the device global xmit lock. Otherwise, the recommended:
netif_carrier_off(dev);
netif_tx_disable(dev);
driver shutdown sequence can happen after the watchdog has already
checked carrier, resulting in possible false alarms. This is because
netif_tx_lock() only sets the frozen bit without maintaining the locks
on the individual queues.
Fixes: c3f26a269c24 ("netdev: Fix lockdep warnings in multiqueue configurations.") Signed-off-by: Edwin Peer <edwin.peer@broadcom.com> Reviewed-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
netfilter: nftables: relax check for stateful expressions in set definition
Restore the original behaviour where users are allowed to add an element
with any stateful expression if the set definition specifies no stateful
expressions. Make sure upper maximum number of stateful expressions of
NFT_SET_EXPR_MAX is not reached.
Fixes: 8cfd9b0f8515 ("netfilter: nftables: generalize set expressions support") Fixes: 48b0ae046ee9 ("netfilter: nftables: netlink support for several set element expressions") Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Linus Torvalds [Mon, 8 Feb 2021 19:32:39 +0000 (11:32 -0800)]
Merge tag 'trace-v5.11-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull tracing fix from Steven Rostedt:
"Fix output of top level event tracing 'enable' file.
When writing a tool for enabling events in the tracing system, an
anomaly was discovered. The top level event 'enable' file would never
show '1' when all events were enabled.
The system and event 'enable' files worked as expected.
The reason was because the top level event 'enable' file included the
'ftrace' tracer events, which are not controlled by the 'enable' file
and would cause the output to be wrong. This appears to have been a
bug since it was created"
* tag 'trace-v5.11-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing: Do not count ftrace events in top level enable output
NeilBrown [Fri, 5 Feb 2021 00:36:30 +0000 (11:36 +1100)]
net: fix iteration for sctp transport seq_files
The sctp transport seq_file iterators take a reference to the transport
in the ->start and ->next functions and releases the reference in the
->show function. The preferred handling for such resources is to
release them in the subsequent ->next or ->stop function call.
Since Commit 1f4aace60b0e ("fs/seq_file.c: simplify seq_file iteration
code and interface") there is no guarantee that ->show will be called
after ->next, so this function can now leak references.
So move the sctp_transport_put() call to ->next and ->stop.
Fixes: 1f4aace60b0e ("fs/seq_file.c: simplify seq_file iteration code and interface") Reported-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: NeilBrown <neilb@suse.de> Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The 'exisitng buffer' in this case is the firmware provided table, and
we should not modify that in place. This fixes a crash on arm64 with
initrd table overrides, in which case the DSDT is not mapped with
read/write permissions.
Reported-by: Shawn Guo <shawn.guo@linaro.org> Signed-off-by: Ard Biesheuvel <ardb@kernel.org> Tested-by: Shawn Guo <shawn.guo@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
cpufreq: ACPI: Update arch scale-invariance max perf ratio if CPPC is not there
If the maximum performance level taken for computing the
arch_max_freq_ratio value used in the x86 scale-invariance code is
higher than the one corresponding to the cpuinfo.max_freq value
coming from the acpi_cpufreq driver, the scale-invariant utilization
falls below 100% even if the CPU runs at cpuinfo.max_freq or slightly
faster, which causes the schedutil governor to select a frequency
below cpuinfo.max_freq. That frequency corresponds to a frequency
table entry below the maximum performance level necessary to get to
the "boost" range of CPU frequencies which prevents "boost"
frequencies from being used in some workloads.
While this issue is related to scale-invariance, it may be amplified
by commit db865272d9c4 ("cpufreq: Avoid configuring old governors as
default with intel_pstate") from the 5.10 development cycle which
made it extremely easy to default to schedutil even if the preferred
driver is acpi_cpufreq as long as intel_pstate is built too, because
the mere presence of the latter effectively removes the ondemand
governor from the defaults. Distro kernels are likely to include
both intel_pstate and acpi_cpufreq on x86, so their users who cannot
use intel_pstate or choose to use acpi_cpufreq may easily be
affectecd by this issue.
If CPPC is available, it can be used to address this issue by
extending the frequency tables created by acpi_cpufreq to cover the
entire available frequency range (including "boost" frequencies) for
each CPU, but if CPPC is not there, acpi_cpufreq has no idea what
the maximum "boost" frequency is and the frequency tables created by
it cannot be extended in a meaningful way, so in that case make it
ask the arch scale-invariance code to to use the "nominal" performance
level for CPU utilization scaling in order to avoid the issue at hand.
Fixes: db865272d9c4 ("cpufreq: Avoid configuring old governors as default with intel_pstate") Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reviewed-by: Giovanni Gherdovich <ggherdovich@suse.cz> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
cpufreq: ACPI: Extend frequency tables to cover boost frequencies
A severe performance regression on AMD EPYC processors when using
the schedutil scaling governor was discovered by Phoronix.com and
attributed to the following commits:
41ea667227ba ("x86, sched: Calculate frequency invariance for AMD
systems")
976df7e5730e ("x86, sched: Use midpoint of max_boost and max_P for
frequency invariance on AMD EPYC")
The source of the problem is that the maximum performance level taken
for computing the arch_max_freq_ratio value used in the x86 scale-
invariance code is higher than the one corresponding to the
cpuinfo.max_freq value coming from the acpi_cpufreq driver.
This effectively causes the scale-invariant utilization to fall below
100% even if the CPU runs at cpuinfo.max_freq or slightly faster, so
the schedutil governor selects a frequency below cpuinfo.max_freq
then. That frequency corresponds to a frequency table entry below
the maximum performance level necessary to get to the "boost" range
of CPU frequencies.
However, if the cpuinfo.max_freq value coming from acpi_cpufreq was
higher, the schedutil governor would select higher frequencies which
in turn would allow acpi_cpufreq to set more adequate performance
levels and to get to the "boost" range of CPU frequencies more often.
This issue affects any systems where acpi_cpufreq is used and the
"boost" (or "turbo") frequencies are enabled, not just AMD EPYC.
Moreover, commit db865272d9c4 ("cpufreq: Avoid configuring old
governors as default with intel_pstate") from the 5.10 development
cycle made it extremely easy to default to schedutil even if the
preferred driver is acpi_cpufreq as long as intel_pstate is built
too, because the mere presence of the latter effectively removes the
ondemand governor from the defaults. Distro kernels are likely to
include both intel_pstate and acpi_cpufreq on x86, so their users
who cannot use intel_pstate or choose to use acpi_cpufreq may
easily be affectecd by this issue.
To address this issue, extend the frequency table constructed by
acpi_cpufreq for each CPU to cover the entire range of available
frequencies (including the "boost" ones) if CPPC is available and
indicates that "boost" (or "turbo") frequencies are enabled. That
causes cpuinfo.max_freq to become the maximum "boost" frequency of
the given CPU (instead of the maximum frequency returned by the ACPI
_PSS object that corresponds to the "nominal" performance level).
Fixes: 41ea667227ba ("x86, sched: Calculate frequency invariance for AMD systems") Fixes: 976df7e5730e ("x86, sched: Use midpoint of max_boost and max_P for frequency invariance on AMD EPYC") Fixes: db865272d9c4 ("cpufreq: Avoid configuring old governors as default with intel_pstate") Link: https://www.phoronix.com/scan.php?page=article&item=linux511-amd-schedutil&num=1 Link: https://lore.kernel.org/linux-pm/20210203135321.12253-2-ggherdovich@suse.cz/ Reported-by: Michael Larabel <Michael@phoronix.com> Diagnosed-by: Giovanni Gherdovich <ggherdovich@suse.cz> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Tested-by: Giovanni Gherdovich <ggherdovich@suse.cz> Reviewed-by: Giovanni Gherdovich <ggherdovich@suse.cz> Tested-by: Michael Larabel <Michael@phoronix.com>
This reverts commit 842067940a3e3fc008a60fee388e000219b32632.
For some solutions e.g. sound/soc/intel/catpt, DW DMA is part of a
compound device (in that very example, domains: ADSP, SSP0, SSP1, DMA0
and DMA1 are part of a single entity) rather than being a standalone
one. Driver for said device may enlist DMA to transfer data during
suspend or resume sequences.
Manipulating RPM explicitly in dw's DMA request and release channel
functions causes suspend() to also invoke resume() for the exact same
device. Similar situation occurs for resume() sequence. Effectively
renders device dysfunctional after first suspend() attempt. Revert the
change to address the problem.
Linus Torvalds [Sun, 7 Feb 2021 18:45:26 +0000 (10:45 -0800)]
Merge tag 'libnvdimm-fixes-5.11-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm
Pull libnvdimm fixes from Dan Williams:
"A fix for a crash scenario that has been present since the initial
merge, a minor regression in sysfs attribute visibility, and a fix for
some flexible array warnings.
The bulk of this pull is an update to the libnvdimm unit test
infrastructure to test non-ACPI platforms. Given there is zero
regression risk for test updates, and the tests enable validation of
bits headed towards the next merge window, I saw no reason to hold the
new tests back. Santosh originally submitted this before the v5.11
window opened.
Summary:
- Fix a crash when sysfs accesses race 'dimm' driver probe/remove.
- Fix a regression in 'resource' attribute visibility necessary for
mapping badblocks and other physical address interrogations.
- Fix some flexible array warnings
- Expand the unit test infrastructure for non-ACPI platforms"
* tag 'libnvdimm-fixes-5.11-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
libnvdimm/dimm: Avoid race between probe and available_slots_show()
ndtest: Add papr health related flags
ndtest: Add nvdimm control functions
ndtest: Add regions and mappings to the test buses
ndtest: Add dimm attributes
ndtest: Add dimms to the two buses
ndtest: Add compatability string to treat it as PAPR family
testing/nvdimm: Add test module for non-nfit platforms
libnvdimm/namespace: Fix visibility of namespace resource attribute
libnvdimm/pmem: Remove unused header
ACPI: NFIT: Fix flexible_array.cocci warnings
- A fix for MSI activation of PCI endpoints with multiple MSIs
* tag 'irq_urgent_for_v5.11_rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
genirq: Prevent [devm_]irq_alloc_desc from returning irq 0
genirq/msi: Activate Multi-MSI early when MSI_FLAG_ACTIVATE_EARLY is set
Linus Torvalds [Sun, 7 Feb 2021 18:16:24 +0000 (10:16 -0800)]
Merge tag 'core_urgent_for_v5.11_rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull syscall entry fixes from Borislav Petkov:
- For syscall user dispatch, separate prctl operation from syscall
redirection range specification before the API has been made official
in 5.11.
- Ensure tasks using the generic syscall code do trap after returning
from a syscall when single-stepping is requested.
* tag 'core_urgent_for_v5.11_rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
entry: Use different define for selector variable in SUD
entry: Ensure trap after single-step on system call return
Linus Torvalds [Sun, 7 Feb 2021 18:03:43 +0000 (10:03 -0800)]
Merge tag 'sched_urgent_for_v5.11_rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fix from Borislav Petkov:
"Revert an attempt to not spread IRQ threads on isolated CPUs which has
a bunch of problems"
* tag 'sched_urgent_for_v5.11_rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
Revert "lib: Restrict cpumask_local_spread to houskeeping CPUs"
Linus Torvalds [Sun, 7 Feb 2021 17:55:26 +0000 (09:55 -0800)]
Merge tag 'timers_urgent_for_v5.11_rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull timer fixes from Borislav Petkov:
"Two more timers-related fixes for v5.11:
- Use a freezable workqueue for RTC sync because the sync can happen
at any time and trigger suspend assertion checks in the i2c
subsystem.
- Correct a previous RTC validation change to check only bit 6 in
register D because some Intel machines use bits 0-5"
* tag 'timers_urgent_for_v5.11_rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
ntp: Use freezable workqueue for RTC synchronization
rtc: mc146818: Dont test for bit 0-5 in Register D
Linus Torvalds [Sun, 7 Feb 2021 17:40:47 +0000 (09:40 -0800)]
Merge tag 'x86_urgent_for_v5.11_rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Borislav Petkov:
"I hope this is the last batch of x86/urgent updates for this round:
- Remove superfluous EFI PGD range checks which lead to those
assertions failing with certain kernel configs and LLVM.
- Disable setting breakpoints on facilities involved in #DB exception
handling to avoid infinite loops.
- Add extra serialization to non-serializing MSRs (IA32_TSC_DEADLINE
and x2 APIC MSRs) to adhere to SDM's recommendation and avoid any
theoretical issues.
- Re-add the EPB MSR reading on turbostat so that it works on older
kernels which don't have the corresponding EPB sysfs file.
- Add Alder Lake to the list of CPUs which support split lock.
- Fix %dr6 register handling in order to be able to set watchpoints
with gdb again.
- Disable CET instrumentation in the kernel so that gcc doesn't add
ENDBR64 to kernel code and thus confuse tracing"
* tag 'x86_urgent_for_v5.11_rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/efi: Remove EFI PGD build time checks
x86/debug: Prevent data breakpoints on cpu_dr7
x86/debug: Prevent data breakpoints on __per_cpu_offset
x86/apic: Add extra serialization for non-serializing MSRs
tools/power/turbostat: Fallback to an MSR read for EPB
x86/split_lock: Enable the split lock feature on another Alder Lake CPU
x86/debug: Fix DR6 handling
x86/build: Disable CET instrumentation in the kernel
Linus Torvalds [Sat, 6 Feb 2021 23:26:28 +0000 (15:26 -0800)]
Merge tag '5.11-rc6-smb3' of git://git.samba.org/sfrench/cifs-2.6
Pull cifs fixes from Steve French:
"Three small smb3 fixes for stable"
* tag '5.11-rc6-smb3' of git://git.samba.org/sfrench/cifs-2.6:
cifs: report error instead of invalid when revalidating a dentry fails
smb3: fix crediting for compounding when only one request in flight
smb3: Fix out-of-bounds bug in SMB2_negotiate()
Linus Torvalds [Sat, 6 Feb 2021 23:18:10 +0000 (15:18 -0800)]
Merge tag 'riscv-for-linus-5.11-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux
Pull RISC-V fixes from Palmer Dabbelt:
"A handful of fixes for this week:
- A fix to avoid evalating the VA twice in virt_addr_valid, which
fixes some WARNs under DEBUG_VIRTUAL.
- Two fixes related to STRICT_KERNEL_RWX: one that fixes some
permissions when strict is disabled, and one to fix some alignment
issues when strict is enabled.
- A fix to disallow the selection of MAXPHYSMEM_2GB on RV32, which
isn't valid any more but may still show up in some oldconfigs.
We still have the HiFive Unleashed ethernet phy reset regression, so
there will likely be something coming next week"
* tag 'riscv-for-linus-5.11-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
RISC-V: Define MAXPHYSMEM_1GB only for RV32
riscv: Align on L1_CACHE_BYTES when STRICT_KERNEL_RWX
RISC-V: Fix .init section permission update
riscv: virt_addr_valid must check the address belongs to linear mapping
Linus Torvalds [Sat, 6 Feb 2021 23:14:39 +0000 (15:14 -0800)]
Merge tag 'powerpc-5.11-7' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc fixes from Michael Ellerman:
- A fix for a change we made to __kernel_sigtramp_rt64() which confused
glibc's backtrace logic, and also changed the semantics of that
symbol, which was arguably an ABI break.
- A fix for a stack overwrite in our VSX instruction emulation.
- A couple of fixes for the Makefile logic in the new C VDSO.
Thanks to Masahiro Yamada, Naveen N. Rao, Raoni Fassina Firmino, and
Ravi Bangoria.
* tag 'powerpc-5.11-7' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/64/signal: Fix regression in __kernel_sigtramp_rt64() semantics
powerpc/vdso64: remove meaningless vgettimeofday.o build rule
powerpc/vdso: fix unnecessary rebuilds of vgettimeofday.o
powerpc/sstep: Fix array out of bound warning
Linus Torvalds [Sat, 6 Feb 2021 23:07:51 +0000 (15:07 -0800)]
Merge tag 'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm
Pull ARM fixes from Russell King:
- Fix latent bug with DC21285 (Footbridge PCI bridge) configuration
accessors that affects GCC >= 4.9.2
- Fix misplaced tegra_uart_config in decompressor
- Ensure signal page contents are initialised
- Fix kexec oops
* tag 'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm:
ARM: kexec: fix oops after TLB are invalidated
ARM: ensure the signal page contains defined contents
ARM: 9043/1: tegra: Fix misplaced tegra_uart_config in decompressor
ARM: footbridge: fix dc21285 PCI configuration accessors
Shay Agroskin [Fri, 5 Feb 2021 19:51:14 +0000 (21:51 +0200)]
net: ena: Update XDP verdict upon failure
The verdict returned from ena_xdp_execute() is used to determine the
fate of the RX buffer's page. In case of XDP Redirect/TX error the
verdict should be set to XDP_ABORTED, otherwise the page won't be freed.
net/vmw_vsock: improve locking in vsock_connect_timeout()
A possible locking issue in vsock_connect_timeout() was recognized by
Eric Dumazet which might cause a null pointer dereference in
vsock_transport_cancel_pkt(). This patch assures that
vsock_transport_cancel_pkt() will be called within the lock, so a race
condition won't occur which could result in vsk->transport to be set to NULL.
In vsock_stream_connect(), a thread will enter schedule_timeout().
While being scheduled out, another thread can enter vsock_stream_connect()
as well and set vsk->transport to NULL. In case a signal was sent, the
first thread can leave schedule_timeout() and vsock_transport_cancel_pkt()
will be called right after. Inside vsock_transport_cancel_pkt(), a null
dereference will happen on transport->cancel_pkt.
Linus Torvalds [Sat, 6 Feb 2021 22:59:07 +0000 (14:59 -0800)]
Merge tag 'usb-5.11-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Pull USB fixes from Greg KH:
"Here are some small, last-minute, USB driver fixes for 5.11-rc7
They all resolve issues reported, or are a few new device ids for some
drivers. They include:
- new device ids for some usb-serial drivers
- xhci fixes for a variety of reported problems
- dwc3 driver bugfixes
- dwc2 driver bugfixes
- usblp driver bugfix
- thunderbolt bugfix
- few other tiny fixes
All have been in linux-next with no reported issues"
* tag 'usb-5.11-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
usb: dwc2: Fix endpoint direction check in ep_from_windex
usb: dwc3: fix clock issue during resume in OTG mode
xhci: fix bounce buffer usage for non-sg list case
usb: host: xhci: mvebu: make USB 3.0 PHY optional for Armada 3720
usb: xhci-mtk: break loop when find the endpoint to drop
usb: xhci-mtk: skip dropping bandwidth of unchecked endpoints
usb: renesas_usbhs: Clear pipe running flag in usbhs_pkt_pop()
USB: gadget: legacy: fix an error code in eth_bind()
thunderbolt: Fix possible NULL pointer dereference in tb_acpi_add_link()
USB: serial: option: Adding support for Cinterion MV31
usb: xhci-mtk: fix unreleased bandwidth data
usb: gadget: aspeed: add missing of_node_put
USB: usblp: don't call usb_set_interface if there's a single alt
USB: serial: cp210x: add pid/vid for WSDA-200-USB
USB: serial: cp210x: add new VID/PID for supporting Teraoka AD2000
Linus Torvalds [Sat, 6 Feb 2021 22:40:27 +0000 (14:40 -0800)]
Merge tag 'block-5.11-2021-02-05' of git://git.kernel.dk/linux-block
Pull block fixes from Jens Axboe:
"A few small regression fixes:
- NVMe pull request from Christoph:
- more quirks for buggy devices (Thorsten Leemhuis, Claus Stovgaard)
- update the email address for Keith (Keith Busch)
- fix an out of bounds access in nvmet-tcp (Sagi Grimberg)
- Regression fix for BFQ shallow depth calculations introduced in
this merge window (Lin)"
* tag 'block-5.11-2021-02-05' of git://git.kernel.dk/linux-block:
nvmet-tcp: fix out-of-bounds access when receiving multiple h2cdata PDUs
bfq-iosched: Revert "bfq: Fix computation of shallow depth"
update the email address for Keith Bush
nvme-pci: ignore the subsysem NQN on Phison E16
nvme-pci: avoid the deepest sleep state on Kingston A2000 SSDs
Linus Torvalds [Sat, 6 Feb 2021 22:37:24 +0000 (14:37 -0800)]
Merge tag 'io_uring-5.11-2021-02-05' of git://git.kernel.dk/linux-block
Pull io_uring fixes from Jens Axboe:
"Two small fixes that should go into 5.11:
- task_work resource drop fix (Pavel)
- identity COW fix (Xiaoguang)"
* tag 'io_uring-5.11-2021-02-05' of git://git.kernel.dk/linux-block:
io_uring: drop mm/files between task_work_submit
io_uring: don't modify identity's files uncess identity is cowed
ibmvnic: Clear failover_pending if unable to schedule
Normally we clear the failover_pending flag when processing the reset.
But if we are unable to schedule a failover reset we must clear the
flag ourselves. We could fail to schedule the reset if we are in PROBING
state (eg: when booting via kexec) or because we could not allocate memory.
Thanks to Cris Forno for helping isolate the problem and for testing.
Jakub Kicinski [Sat, 6 Feb 2021 17:27:20 +0000 (09:27 -0800)]
Merge tag 'wireless-drivers-2021-02-05' of git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers
Kalle Valo says:
====================
wireless-drivers fixes for v5.11
Third, and most likely the last, set of fixes for v5.11. Two very
small fixes.
ath9k
* fix build regression related to LEDS_CLASS
mt76
* fix a memory leak
* tag 'wireless-drivers-2021-02-05' of git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers:
mt76: dma: fix a possible memory leak in mt76_add_fragment()
ath9k: fix build error with LEDS_CLASS=m
====================
Borislav Petkov [Fri, 5 Feb 2021 10:31:31 +0000 (11:31 +0100)]
x86/efi: Remove EFI PGD build time checks
With CONFIG_X86_5LEVEL, CONFIG_UBSAN and CONFIG_UBSAN_UNSIGNED_OVERFLOW
enabled, clang fails the build with
x86_64-linux-ld: arch/x86/platform/efi/efi_64.o: in function `efi_sync_low_kernel_mappings':
efi_64.c:(.text+0x22c): undefined reference to `__compiletime_assert_354'
which happens due to -fsanitize=unsigned-integer-overflow being enabled:
-fsanitize=unsigned-integer-overflow: Unsigned integer overflow, where
the result of an unsigned integer computation cannot be represented
in its type. Unlike signed integer overflow, this is not undefined
behavior, but it is often unintentional. This sanitizer does not check
for lossy implicit conversions performed before such a computation
(see -fsanitize=implicit-conversion).
and that fires when the (intentional) EFI_VA_START/END defines overflow
an unsigned long, leading to the assertion expressions not getting
optimized away (on GCC they do)...
However, those checks are superfluous: the runtime services mapping
code already makes sure the ranges don't overshoot EFI_VA_END as the
EFI mapping range is hardcoded. On each runtime services call, it is
switched to the EFI-specific PGD and even if mappings manage to escape
that last PGD, this won't remain unnoticed for long.
So rip them out.
See https://github.com/ClangBuiltLinux/linux/issues/256 for more info.
net: stmmac: set TxQ mode back to DCB after disabling CBS
When disable CBS, mode_to_use parameter is not updated even the operation
mode of Tx Queue is changed to Data Centre Bridging (DCB). Therefore,
when tc_setup_cbs() function is called to re-enable CBS, the operation
mode of Tx Queue remains at DCB, which causing CBS fails to work.
This patch updates the value of mode_to_use parameter to MTL_QUEUE_DCB
after operation mode of Tx Queue is changed to DCB in stmmac_dma_qmode()
callback function.
Fixes: 1f705bc61aee ("net: stmmac: Add support for CBS QDISC") Suggested-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Signed-off-by: Mohammad Athari Bin Ismail <mohammad.athari.ismail@intel.com> Signed-off-by: Song, Yoong Siang <yoong.siang.song@intel.com> Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Link: https://lore.kernel.org/r/1612447396-20351-1-git-send-email-yoong.siang.song@intel.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
====================
dpaa_eth: A050385 erratum workaround fixes under XDP
This series addresses issue with the current workaround for the A050385
erratum in XDP scenarios.
The first patch makes sure the xdp_frame structure stored at the start of
new buffers isn't overwritten.
The second patch decreases the required data alignment value, thus
preventing unnecessary realignments.
The third patch moves the data in place to align it, instead of allocating
a new buffer for each frame that breaks the alignment rules, thus bringing
an up to 40% performance increase. With this change, the impact of the
erratum workaround is reduced in many cases to a single digit decrease, and
to lower double digits in single flow scenarios.
Changes in v2:
- guarantee enough tailroom is available for the shared_info in 1/3
====================
Camelia Groza [Thu, 4 Feb 2021 16:49:27 +0000 (18:49 +0200)]
dpaa_eth: reduce data alignment requirements for the A050385 erratum
The 256 byte data alignment is required for preventing DMA transaction
splits when crossing 4K page boundaries. Since XDP deals only with page
sized buffers or less, this restriction isn't needed. Instead, the data
only needs to be aligned to 64 bytes to prevent DMA transaction splits.
These lessened restrictions can increase performance by widening the pool
of permitted data alignments and preventing unnecessary realignments.
Fixes: ae680bcbd06a ("dpaa_eth: implement the A050385 erratum workaround for XDP") Signed-off-by: Camelia Groza <camelia.groza@nxp.com> Acked-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Acked-by: Madalin Bucur <madalin.bucur@oss.nxp.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Camelia Groza [Thu, 4 Feb 2021 16:49:26 +0000 (18:49 +0200)]
dpaa_eth: reserve space for the xdp_frame under the A050385 erratum
When the erratum workaround is triggered, the newly created xdp_frame
structure is stored at the start of the newly allocated buffer. Avoid
the structure from being overwritten by explicitly reserving enough
space in the buffer for storing it.
Account for the fact that the structure's size might increase in time by
aligning the headroom to DPAA_FD_DATA_ALIGNMENT bytes, thus guaranteeing
the data's alignment.
Fixes: ae680bcbd06a ("dpaa_eth: implement the A050385 erratum workaround for XDP") Signed-off-by: Camelia Groza <camelia.groza@nxp.com> Acked-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Acked-by: Madalin Bucur <madalin.bucur@oss.nxp.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Eric Dumazet [Thu, 4 Feb 2021 21:31:46 +0000 (13:31 -0800)]
net: gro: do not keep too many GRO packets in napi->rx_list
Commit c80794323e82 ("net: Fix packet reordering caused by GRO and
listified RX cooperation") had the unfortunate effect of adding
latencies in common workloads.
Before the patch, GRO packets were immediately passed to
upper stacks.
After the patch, we can accumulate quite a lot of GRO
packets (depdending on NAPI budget).
My fix is counting in napi->rx_count number of segments
instead of number of logical packets.
Fixes: c80794323e82 ("net: Fix packet reordering caused by GRO and listified RX cooperation") Signed-off-by: Eric Dumazet <edumazet@google.com> Bisected-by: John Sperbeck <jsperbeck@google.com> Tested-by: Jian Yang <jianyang@google.com> Cc: Maxim Mikityanskiy <maximmi@mellanox.com> Reviewed-by: Saeed Mahameed <saeedm@nvidia.com> Reviewed-by: Edward Cree <ecree.xilinx@gmail.com> Reviewed-by: Alexander Lobakin <alobakin@pm.me> Link: https://lore.kernel.org/r/20210204213146.4192368-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>