macvtap: advertise link netns via netlink

[ Upstream commit a02192151b7dbf855084c38dca380d77c7658353 ]

Assign rtnl_link_ops->get_link_net() callback so that IFLA_LINK_NETNSID is
added to rtnetlink messages. This fixes iproute2 which otherwise resolved
the link interface to an interface in the wrong namespace.

Test commands:

  ip netns add nst
  ip link add dummy0 type dummy
  ip link add link macvtap0 link dummy0 type macvtap
  ip link set macvtap0 netns nst
  ip -netns nst link show macvtap0

Before:

  10: macvtap0@gre0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 500
      link/ether 5e:8f:ae:1d:60:50 brd ff:ff:ff:ff:ff:ff

After:

  10: macvtap0@if2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 500
      link/ether 5e:8f:ae:1d:60:50 brd ff:ff:ff:ff:ff:ff link-netnsid 0

Reported-by: Leonardo Mörlein <freifunk@irrelefant.net>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Link: https://lore.kernel.org/r/20220228003240.1337426-1-sven@narfation.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mips: ralink: fix a refcount leak in ill_acc_of_setup()

[ Upstream commit 4a0a1436053b17e50b7c88858fb0824326641793 ]

of_node_put(np) needs to be called when pdev == NULL.

Signed-off-by: Hangyu Hua <hbh25y@gmail.com>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net/smc: correct settings of RMB window update limit

[ Upstream commit 6bf536eb5c8ca011d1ff57b5c5f7c57ceac06a37 ]

rmbe_update_limit is used to limit announcing receive
window updating too frequently. RFC7609 request a minimal
increase in the window size of 10% of the receive buffer
space. But current implementation used:

  min_t(int, rmbe_size / 10, SOCK_MIN_SNDBUF / 2)

and SOCK_MIN_SNDBUF / 2 == 2304 Bytes, which is almost
always less then 10% of the receive buffer space.

This causes the receiver always sending CDC message to
update its consumer cursor when it consumes more then 2K
of data. And as a result, we may encounter something like
"TCP silly window syndrome" when sending 2.5~8K message.

This patch fixes this using max(rmbe_size / 10, SOCK_MIN_SNDBUF / 2).

With this patch and SMC autocorking enabled, qperf 2K/4K/8K
tcp_bw test shows 45%/75%/40% increase in throughput respectively.

Signed-off-by: Dust Li <dust.li@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: hisi_sas: Limit users changing debugfs BIST count value

[ Upstream commit 286ce4c65fbdf5eb9d4d5f4e4997c4e32bf1b073 ]

Add a file operation for "cnt" file under bist directory, so users can only
read "cnt" or clear "cnt" to zero, but cannot randomly modify.

Link: https://lore.kernel.org/r/1645703489-87194-6-git-send-email-john.garry@huawei.com
Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: hisi_sas: Free irq vectors in order for v3 HW

[ Upstream commit 554fb72ee34f4732c7f694f56c3c6e67790352a0 ]

If the driver probe fails to request the channel IRQ or fatal IRQ, the
driver will free the IRQ vectors before freeing the IRQs in free_irq(),
and this will cause a kernel BUG like this:

------------[ cut here ]------------
kernel BUG at drivers/pci/msi.c:369!
Internal error: Oops - BUG: 0 [#1] PREEMPT SMP
Call trace:
   free_msi_irqs+0x118/0x13c
   pci_disable_msi+0xfc/0x120
   pci_free_irq_vectors+0x24/0x3c
   hisi_sas_v3_probe+0x360/0x9d0 [hisi_sas_v3_hw]
   local_pci_probe+0x44/0xb0
   work_for_cpu_fn+0x20/0x34
   process_one_work+0x1d0/0x340
   worker_thread+0x2e0/0x460
   kthread+0x180/0x190
   ret_from_fork+0x10/0x20
---[ end trace b88990335b610c11 ]---

So we use devm_add_action() to control the order in which we free the
vectors.

Link: https://lore.kernel.org/r/1645703489-87194-4-git-send-email-john.garry@huawei.com
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: aha152x: Fix aha152x_setup() __setup handler return value

[ Upstream commit cc8294ec4738d25e2bb2d71f7d82a9bf7f4a157b ]

__setup() handlers should return 1 if the command line option is handled
and 0 if not (or maybe never return 0; doing so just pollutes init's
environment with strings that are not init arguments/parameters).

Return 1 from aha152x_setup() to indicate that the boot option has been
handled.

Link: lore.kernel.org/r/64644a2f-4a20-bab3-1e15-3b2cdd0defe3@omprussia.ru
Link: https://lore.kernel.org/r/20220223000623.5920-1-rdunlap@infradead.org
Cc: "Juergen E. Fischer" <fischer@norbit.de>
Cc: "James E.J. Bottomley" <jejb@linux.ibm.com>
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>
Reported-by: Igor Zhbanov <i.zhbanov@omprussia.ru>
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mt76: mt7615: Fix assigning negative values to unsigned variable

[ Upstream commit 9273ffcc9a11942bd586bb42584337ef3962b692 ]

Smatch reports the following:
drivers/net/wireless/mediatek/mt76/mt7615/mac.c:1865
mt7615_mac_adjust_sensitivity() warn: assigning (-110) to unsigned
variable 'def_th'
drivers/net/wireless/mediatek/mt76/mt7615/mac.c:1865
mt7615_mac_adjust_sensitivity() warn: assigning (-98) to unsigned
variable 'def_th'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sasha Levin <sashal@kernel.org>

powerpc/64s/hash: Make hash faults work in NMI context

[ Upstream commit 8b91cee5eadd2021f55e6775f2d50bd56d00c217 ]

Hash faults are not resoved in NMI context, instead causing the access
to fail. This is done because perf interrupts can get backtraces
including walking the user stack, and taking a hash fault on those could
deadlock on the HPTE lock if the perf interrupt hits while the same HPTE
lock is being held by the hash fault code. The user-access for the stack
walking will notice the access failed and deal with that in the perf
code.

The reason to allow perf interrupts in is to better profile hash faults.

The problem with this is any hash fault on a kernel access that happens
in NMI context will crash, because kernel accesses must not fail.

Hard lockups, system reset, machine checks that access vmalloc space
including modules and including stack backtracing and symbol lookup in
modules, per-cpu data, etc could all run into this problem.

Fix this by disallowing perf interrupts in the hash fault code (the
direct hash fault is covered by MSR[EE]=0 so the PMI disable just needs
to extend to the preload case). This simplifies the tricky logic in hash
faults and perf, at the cost of reduced profiling of hash faults.

perf can still latch addresses when interrupts are disabled, it just
won't get the stack trace at that point, so it would still find hot
spots, just sometimes with confusing stack chains.

An alternative could be to allow perf interrupts here but always do the
slowpath stack walk if we are in nmi context, but that slows down all
perf interrupt stack walking on hash though and it does not remove as
much tricky code.

Reported-by: Laurent Dufour <ldufour@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Tested-by: Laurent Dufour <ldufour@linux.ibm.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20220204035348.545435-1-npiggin@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

mt76: mt7915: fix injected MPDU transmission to not use HW A-MSDU

[ Upstream commit 28225a6ef80ebf46c46e5fbd5b1ee231a0b2b5b7 ]

Before, the hardware would be allowed to transmit injected 802.11 MPDUs
as A-MSDU. This resulted in corrupted frames being transmitted. Now,
injected MPDUs are transmitted as-is, without A-MSDU.

The fix was verified with frame injection on MT7915 hardware, both with
and without the injected frame being encrypted.

If the hardware cannot do A-MSDU aggregation on MPDUs, this problem
would also be present in the TX path where mac80211 does the 802.11
encapsulation. However, I have not observed any such problem when
disabling IEEE80211_HW_SUPPORTS_TX_ENCAP_OFFLOAD to force that mode.
Therefore this fix is isolated to injected frames only.

The same A-MSDU logic is also present in the mt7921 driver, so it is
likely that this fix should be applied there too. I do not have access
to mt7921 hardware so I have not been able to test that.

Signed-off-by: Johan Almbladh <johan.almbladh@anyfinetworks.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: pm8001: Fix memory leak in pm8001_chip_fw_flash_update_req()

[ Upstream commit f792a3629f4c4aa4c3703d66b43ce1edcc3ec09a ]

In pm8001_chip_fw_flash_update_build(), if
pm8001_chip_fw_flash_update_build() fails, the struct fw_control_ex
allocated must be freed.

Link: https://lore.kernel.org/r/20220220031810.738362-23-damien.lemoal@opensource.wdc.com
Reviewed-by: Jack Wang <jinpu.wang@ionos.com>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: pm8001: Fix tag leaks on error

[ Upstream commit 4c8f04b1905cd4b776d0b720463c091545478ef7 ]

In pm8001_chip_set_dev_state_req(), pm8001_chip_fw_flash_update_req(),
pm80xx_chip_phy_ctl_req() and pm8001_chip_reg_dev_req() add missing calls
to pm8001_tag_free() to free the allocated tag when pm8001_mpi_build_cmd()
fails.

Similarly, in pm8001_exec_internal_task_abort(), if the chip ->task_abort
method fails, the tag allocated for the abort request task must be
freed. Add the missing call to pm8001_tag_free().

Link: https://lore.kernel.org/r/20220220031810.738362-22-damien.lemoal@opensource.wdc.com
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: pm8001: Fix task leak in pm8001_send_abort_all()

[ Upstream commit f90a74892f3acf0cdec5844e90fc8686ca13e7d7 ]

In pm8001_send_abort_all(), make sure to free the allocated sas task
if pm8001_tag_alloc() or pm8001_mpi_build_cmd() fail.

Link: https://lore.kernel.org/r/20220220031810.738362-21-damien.lemoal@opensource.wdc.com
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: pm8001: Fix pm8001_mpi_task_abort_resp()

[ Upstream commit 7e6b7e740addcea450041b5be8e42f0a4ceece0f ]

The call to pm8001_ccb_task_free() at the end of
pm8001_mpi_task_abort_resp() already frees the ccb tag. So when the device
NCQ_ABORT_ALL_FLAG is set, the tag should not be freed again.  Also change
the hardcoded 0xBFFFFFFF value to ~NCQ_ABORT_ALL_FLAG as it ought to be.

Link: https://lore.kernel.org/r/20220220031810.738362-19-damien.lemoal@opensource.wdc.com
Reviewed-by: Jack Wang <jinpu.wang@ionos.com>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: pm8001: Fix pm80xx_pci_mem_copy() interface

[ Upstream commit 3762d8f6edcdb03994c919f9487fd6d336c06561 ]

The declaration of the local variable destination1 in pm80xx_pci_mem_copy()
as a pointer to a u32 results in the sparse warning:

warning: incorrect type in assignment (different base types)
    expected unsigned int [usertype]
    got restricted __le32 [usertype]

Furthermore, the destination" argument of pm80xx_pci_mem_copy() is wrongly
declared with the const attribute.

Fix both problems by changing the type of the "destination" argument to
"__le32 *" and use this argument directly inside the pm80xx_pci_mem_copy()
function, thus removing the need for the destination1 local variable.

Link: https://lore.kernel.org/r/20220220031810.738362-6-damien.lemoal@opensource.wdc.com
Reviewed-by: Jack Wang <jinpu.wang@ionos.com>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

vfio/pci: Stub vfio_pci_vga_rw when !CONFIG_VFIO_PCI_VGA

[ Upstream commit 6e031ec0e5a2dda53e12e0d2a7e9b15b47a3c502 ]

Resolve build errors reported against UML build for undefined
ioport_map() and ioport_unmap() functions.  Without this config
option a device cannot have vfio_pci_core_device.has_vga set,
so the existing function would always return -EINVAL anyway.

Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Link: https://lore.kernel.org/r/20220123125737.2658758-1-geert@linux-m68k.org
Link: https://lore.kernel.org/r/164306582968.3758255.15192949639574660648.stgit@omen
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/amdkfd: make CRAT table missing message informational only

[ Upstream commit 9dff13f9edf755a15f6507874185a3290c1ae8bb ]

The driver has a fallback so make the message informational
rather than a warning. The driver has a fallback if the
Component Resource Association Table (CRAT) is missing, so
make this informational now.

Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1906
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

dm: requeue IO if mapping table not yet available

[ Upstream commit fa247089de9936a46e290d4724cb5f0b845600f5 ]

Update both bio-based and request-based DM to requeue IO if the
mapping table not available.

This race of IO being submitted before the DM device ready is so
narrow, yet possible for initial table load given that the DM device's
request_queue is created prior, that it best to requeue IO to handle
this unlikely case.

Reported-by: Zhang Yi <yi.zhang@huawei.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

dm ioctl: prevent potential spectre v1 gadget

[ Upstream commit cd9c88da171a62c4b0f1c70e50c75845969fbc18 ]

It appears like cmd could be a Spectre v1 gadget as it's supplied by a
user and used as an array index. Prevent the contents of kernel memory
from being leaked to userspace via speculative execution by using
array_index_nospec.

Signed-off-by: Jordy Zomer <jordy@pwning.systems>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ipv4: Invalidate neighbour for broadcast address upon address addition

[ Upstream commit 0c51e12e218f20b7d976158fdc18019627326f7a ]

In case user space sends a packet destined to a broadcast address when a
matching broadcast route is not configured, the kernel will create a
unicast neighbour entry that will never be resolved [1].

When the broadcast route is configured, the unicast neighbour entry will
not be invalidated and continue to linger, resulting in packets being
dropped.

Solve this by invalidating unresolved neighbour entries for broadcast
addresses after routes for these addresses are internally configured by
the kernel. This allows the kernel to create a broadcast neighbour entry
following the next route lookup.

Another possible solution that is more generic but also more complex is
to have the ARP code register a listener to the FIB notification chain
and invalidate matching neighbour entries upon the addition of broadcast
routes.

It is also possible to wave off the issue as a user space problem, but
it seems a bit excessive to expect user space to be that intimately
familiar with the inner workings of the FIB/neighbour kernel code.

[1] https://lore.kernel.org/netdev/55a04a8f-56f3-f73c-2aea-2195923f09d1@huawei.com/

Reported-by: Wang Hai <wanghai38@huawei.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Tested-by: Wang Hai <wanghai38@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/msm/dsi: Remove spurious IRQF_ONESHOT flag

[ Upstream commit 24b176d8827d167ac3b379317f60c0985f6e95aa ]

Quoting the header comments, IRQF_ONESHOT is "Used by threaded interrupts
which need to keep the irq line disabled until the threaded handler has
been run.". When applied to an interrupt that doesn't request a threaded
irq then IRQF_ONESHOT has a lesser known (undocumented?) side effect,
which it to disable the forced threading of irqs (and for "normal" kernels
it is a nop). In this case I can find no evidence that suppressing forced
threading is intentional. Had it been intentional then a driver must adopt
the raw_spinlock API in order to avoid deadlocks on PREEMPT_RT kernels
(and avoid calling any kernel API that uses regular spinlocks).

Fix this by removing the spurious additional flag.

This change is required for my Snapdragon 7cx Gen2 tablet to boot-to-GUI
with PREEMPT_RT enabled.

Signed-off-by: Daniel Thompson <daniel.thompson@linaro.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Link: https://lore.kernel.org/r/20220201174734.196718-2-daniel.thompson@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

iwlwifi: mvm: move only to an enabled channel

[ Upstream commit e04135c07755d001b5cde61048c69a7cc84bb94b ]

During disassociation we're decreasing the phy's ref count.
If the ref count becomes 0, we're configuring the phy ctxt
to the default channel (the lowest channel which the device
can operate on). Currently we're not checking whether the
the default channel is enabled or not. Fix it by configuring
the phy ctxt to the lowest channel which is enabled.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20220210181930.03f281b6a6bc.I5b63d43ec41996d599e6f37ec3f32e878b3e405e@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

iwlwifi: mvm: Correctly set fragmented EBS

[ Upstream commit d8d4dd26b9e0469baf5017f0544d852fd4e3fb6d ]

Currently, fragmented EBS was set for a channel only if the 'hb_type'
was set to fragmented or balanced scan. However, 'hb_type' is set only
in case of CDB, and thus fragmented EBS is never set for a channel for
non-CDB devices. Fix it.

Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20220204122220.a6165ac9b9d5.I654eafa62fd647030ae6d4f07f32c96c3171decb@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

usb: dwc3: pci: Set the swnode from inside dwc3_pci_quirks()

[ Upstream commit e285cb403994419e997749c9a52b9370884ae0c8 ]

The quirk handling may need to set some different properties
which means using a different swnode, move the setting of the swnode
to inside dwc3_pci_quirks() so that the quirk handling can choose
a different swnode.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20220213130524.18748-4-hdegoede@redhat.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net/mlx5e: Disable TX queues before registering the netdev

[ Upstream commit d08c6e2a4d0308a7922d7ef3b1b3af45d4096aad ]

Normally, the queues are disabled when the channels are deactivated, and
enabled when the channels are activated. However, on register, the
channels are not active, but the queues are enabled by default. This
change fixes it, preventing mlx5e_xmit from running when the channels
are deactivated in the beginning.

Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

power: supply: axp288-charger: Set Vhold to 4.4V

[ Upstream commit 5ac121b81b4051e7fc83d5b3456a5e499d5bd147 ]

The AXP288's recommended and factory default Vhold value (minimum
input voltage below which the input current draw will be reduced)
is 4.4V. This lines up with other charger IC's such as the TI
bq2419x/bq2429x series which use 4.36V or 4.44V.

For some reason some BIOS-es initialize Vhold to 4.6V or even 4.7V
which combined with the typical voltage drop over typically low
wire gauge micro-USB cables leads to the input-current getting
capped below 1A (with a 2A capable dedicated charger) based on Vhold.

This leads to slow charging, or even to the device slowly discharging
if the device is in heavy use.

As the Linux AXP288 drivers use the builtin BC1.2 charger detection
and send the input-current-limit according to the detected charger
there really is no reason not to use the recommended 4.4V Vhold.

Set Vhold to 4.4V to fix the slow charging issue on various devices.

There is one exception, the special-case of the HP X2 2-in-1s which
combine this BC1.2 capable PMIC with a Type-C port and a 5V/3A factory
provided charger with a Type-C plug which does not do BC1.2. These
have their input-current-limit hardcoded to 3A (like under Windows)
and use a higher Vhold on purpose to limit the current when used
with other chargers. To avoid touching Vhold on these HP X2 laptops
the code setting Vhold is added to an else branch of the if checking
for these models.

Note this also fixes the sofar unused VBUS_ISPOUT_VHOLD_SET_MASK
define, which was wrong.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

powerpc/set_memory: Avoid spinlock recursion in change_page_attr()

[ Upstream commit a4c182ecf33584b9b2d1aa9dad073014a504c01f ]

Commit 1f9ad21c3b38 ("powerpc/mm: Implement set_memory() routines")
included a spin_lock() to change_page_attr() in order to
safely perform the three step operations. But then
commit 9f7853d7609d ("powerpc/mm: Fix set_memory_*() against
concurrent accesses") modify it to use pte_update() and do
the operation safely against concurrent access.

In the meantime, Maxime reported some spinlock recursion.

[   15.351649] BUG: spinlock recursion on CPU#0, kworker/0:2/217
[   15.357540]  lock: init_mm+0x3c/0x420, .magic: dead4ead, .owner: kworker/0:2/217, .owner_cpu: 0
[   15.366563] CPU: 0 PID: 217 Comm: kworker/0:2 Not tainted 5.15.0+ #523
[   15.373350] Workqueue: events do_free_init
[   15.377615] Call Trace:
[   15.380232] [e4105ac0] [800946a4] do_raw_spin_lock+0xf8/0x120 (unreliable)
[   15.387340] [e4105ae0] [8001f4ec] change_page_attr+0x40/0x1d4
[   15.393413] [e4105b10] [801424e0] __apply_to_page_range+0x164/0x310
[   15.400009] [e4105b60] [80169620] free_pcp_prepare+0x1e4/0x4a0
[   15.406045] [e4105ba0] [8016c5a0] free_unref_page+0x40/0x2b8
[   15.411979] [e4105be0] [8018724c] kasan_depopulate_vmalloc_pte+0x6c/0x94
[   15.418989] [e4105c00] [801424e0] __apply_to_page_range+0x164/0x310
[   15.425451] [e4105c50] [80187834] kasan_release_vmalloc+0xbc/0x134
[   15.431898] [e4105c70] [8015f7a8] __purge_vmap_area_lazy+0x4e4/0xdd8
[   15.438560] [e4105d30] [80160d10] _vm_unmap_aliases.part.0+0x17c/0x24c
[   15.445283] [e4105d60] [801642d0] __vunmap+0x2f0/0x5c8
[   15.450684] [e4105db0] [800e32d0] do_free_init+0x68/0x94
[   15.456181] [e4105dd0] [8005d094] process_one_work+0x4bc/0x7b8
[   15.462283] [e4105e90] [8005d614] worker_thread+0x284/0x6e8
[   15.468227] [e4105f00] [8006aaec] kthread+0x1f0/0x210
[   15.473489] [e4105f40] [80017148] ret_from_kernel_thread+0x14/0x1c

Remove the read / modify / write sequence to make the operation atomic
and remove the spin_lock() in change_page_attr().

To do the operation atomically, we can't use pte modification helpers
anymore. Because all platforms have different combination of bits, it
is not easy to use those bits directly. But all have the
_PAGE_KERNEL_{RO/ROX/RW/RWX} set of flags. All we need it to compare
two sets to know which bits are set or cleared.

For instance, by comparing _PAGE_KERNEL_ROX and _PAGE_KERNEL_RO you
know which bit gets cleared and which bit get set when changing exec
permission.

Reported-by: Maxime Bizon <mbizon@freebox.fr>
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/all/20211212112152.GA27070@sakura/
Link: https://lore.kernel.org/r/43c3c76a1175ae6dc1a3d3b5c3f7ecb48f683eea.1640344012.git.christophe.leroy@csgroup.eu
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: mpi3mr: Fix memory leaks

[ Upstream commit d44b5fefb22e139408ae12b864da1ecb9ad9d1d2 ]

Fix memory leaks related to operational reply queue's memory segments which
are not getting freed while unloading the driver.

Link: https://lore.kernel.org/r/20220210095817.22828-9-sreekanth.reddy@broadcom.com
Signed-off-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: mpi3mr: Fix reporting of actual data transfer size

[ Upstream commit 9992246127246a27cc7184f05cce6f62ac48f84e ]

The driver is missing to set the residual size while completing an
I/O. Ensure proper data transfer size is reported to the kernel on I/O
completion based on the transfer length reported by the firmware.

Link: https://lore.kernel.org/r/20220210095817.22828-7-sreekanth.reddy@broadcom.com
Signed-off-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

PCI: pciehp: Add Qualcomm quirk for Command Completed erratum

[ Upstream commit 9f72d4757cbe4d1ed669192f6d23817c9e437c4b ]

The Qualcomm PCI bridge device (Device ID 0x0110) found in chipsets such as
SM8450 does not set the Command Completed bit unless writes to the Slot
Command register change "Control" bits.

This results in timeouts like below:

  pcieport 0001:00:00.0: pciehp: Timeout on hotplug command 0x03c0 (issued 2020 msec ago)

Add the device to the Command Completed quirk to mark commands "completed"
immediately unless they change the "Control" bits.

Link: https://lore.kernel.org/r/20220210145003.135907-1-manivannan.sadhasivam@linaro.org
Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

tcp: Don't acquire inet_listen_hashbucket::lock with disabled BH.

[ Upstream commit 4f9bf2a2f5aacf988e6d5e56b961ba45c5a25248 ]

Commit
   9652dc2eb9e40 ("tcp: relax listening_hash operations")

removed the need to disable bottom half while acquiring
listening_hash.lock. There are still two callers left which disable
bottom half before the lock is acquired.

On PREEMPT_RT the softirqs are preemptible and local_bh_disable() acts
as a lock to ensure that resources, that are protected by disabling
bottom halves, remain protected.
This leads to a circular locking dependency if the lock acquired with
disabled bottom halves is also acquired with enabled bottom halves
followed by disabling bottom halves. This is the reverse locking order.
It has been observed with inet_listen_hashbucket::lock:

local_bh_disable() + spin_lock(&ilb->lock):
  inet_listen()
    inet_csk_listen_start()
      sk->sk_prot->hash() := inet_hash()
local_bh_disable()
__inet_hash()
  spin_lock(&ilb->lock);
    acquire(&ilb->lock);

Reverse order: spin_lock(&ilb2->lock) + local_bh_disable():
  tcp_seq_next()
    listening_get_next()
      spin_lock(&ilb2->lock);
acquire(&ilb2->lock);

  tcp4_seq_show()
    get_tcp4_sock()
      sock_i_ino()
read_lock_bh(&sk->sk_callback_lock);
  acquire(softirq_ctrl) // <---- whoops
  acquire(&sk->sk_callback_lock)

Drop local_bh_disable() around __inet_hash() which acquires
listening_hash->lock. Split inet_unhash() and acquire the
listen_hashbucket lock without disabling bottom halves; the inet_ehash
lock with disabled bottom halves.

Reported-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://lkml.kernel.org/r/12d6f9879a97cd56c09fb53dee343cbb14f7f1f7.camel@gmx.de
Link: https://lkml.kernel.org/r/X9CheYjuXWc75Spa@hirez.programming.kicks-ass.net
Link: https://lore.kernel.org/r/YgQOebeZ10eNx1W6@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

PCI: endpoint: Fix alignment fault error in copy tests

[ Upstream commit 829cc0e2ea2d61fb6c54bc3f8a17f86c56e11864 ]

The copy test uses the memcpy() to copy data between IO memory spaces.
This can trigger an alignment fault error (pasted the error logs below)
because memcpy() may use unaligned accesses on a mapped memory that is
just IO, which does not support unaligned memory accesses.

Fix it by using the correct memcpy API to copy from/to IO memory.

Alignment fault error logs:
   Unable to handle kernel paging request at virtual address ffff8000101cd3c1
   Mem abort info:
     ESR = 0x96000021
     EC = 0x25: DABT (current EL), IL = 32 bits
     SET = 0, FnV = 0
     EA = 0, S1PTW = 0
     FSC = 0x21: alignment fault
   Data abort info:
     ISV = 0, ISS = 0x00000021
     CM = 0, WnR = 0
   swapper pgtable: 4k pages, 48-bit VAs, pgdp=0000000081773000
   [ffff8000101cd3c1] pgd=1000000082410003, p4d=1000000082410003, pud=1000000082411003, pmd=1000000082412003, pte=0068004000001f13
   Internal error: Oops: 96000021 [#1] PREEMPT SMP
   Modules linked in:
   CPU: 0 PID: 6 Comm: kworker/0:0H Not tainted 5.15.0-rc1-next-20210914-dirty #2
   Hardware name: LS1012A RDB Board (DT)
   Workqueue: kpcitest pci_epf_test_cmd_handler
   pstate: 80000005 (Nzcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
   pc : __memcpy+0x168/0x230
   lr : pci_epf_test_cmd_handler+0x6f0/0xa68
   sp : ffff80001003bce0
   x29: ffff80001003bce0 x28: ffff800010135000 x27: ffff8000101e5000
   x26: ffff8000101cd000 x25: ffff6cda941cf6c8 x24: 0000000000000000
   x23: ffff6cda863f2000 x22: ffff6cda9096c800 x21: ffff800010135000
   x20: ffff6cda941cf680 x19: ffffaf39fd999000 x18: 0000000000000000
   x17: 0000000000000000 x16: 0000000000000000 x15: ffffaf39fd2b6000
   x14: 0000000000000000 x13: 15f5c8fa2f984d57 x12: 604d132b60275454
   x11: 065cee5e5fb428b6 x10: aae662eb17d0cf3e x9 : 1d97c9a1b4ddef37
   x8 : 7541b65edebf928c x7 : e71937c4fc595de0 x6 : b8a0e09562430d1c
   x5 : ffff8000101e5401 x4 : ffff8000101cd401 x3 : ffff8000101e5380
   x2 : fffffffffffffff1 x1 : ffff8000101cd3c0 x0 : ffff8000101e5000
   Call trace:
    __memcpy+0x168/0x230
    process_one_work+0x1ec/0x370
    worker_thread+0x44/0x478
    kthread+0x154/0x160
    ret_from_fork+0x10/0x20
   Code: a984346c a9c4342c f1010042 54fffee8 (a97c3c8e)
   ---[ end trace 568c28c7b6336335 ]---

Link: https://lore.kernel.org/r/20211217094708.28678-1-Zhiqiang.Hou@nxp.com
Signed-off-by: Hou Zhiqiang <Zhiqiang.Hou@nxp.com>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Reviewed-by: Kishon Vijay Abraham I <kishon@ti.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

usb: ehci: add pci device support for Aspeed platforms

[ Upstream commit c3c9cee592828528fd228b01d312c7526c584a42 ]

Enable Aspeed quirks in commit 7f2d73788d90 ("usb: ehci:
handshake CMD_RUN instead of STS_HALT") to support Aspeed
ehci-pci device.

Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Neal Liu <neal_liu@aspeedtech.com>
Link: https://lore.kernel.org/r/20220208101657.76459-1-neal_liu@aspeedtech.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

iommu/arm-smmu-v3: fix event handling soft lockup

[ Upstream commit 30de2b541af98179780054836b48825fcfba4408 ]

During event processing, events are read from the event queue one
by one until the queue is empty.If the master device continuously
requests address access at the same time and the SMMU generates
events, the cyclic processing of the event takes a long time and
softlockup warnings may be reported.

arm-smmu-v3 arm-smmu-v3.34.auto: event 0x0a received:
arm-smmu-v3 arm-smmu-v3.34.auto:  0x00007f220000280a
arm-smmu-v3 arm-smmu-v3.34.auto:  0x000010000000007e
arm-smmu-v3 arm-smmu-v3.34.auto:  0x00000000034e8670
watchdog: BUG: soft lockup - CPU#0 stuck for 22s! [irq/268-arm-smm:247]
Call trace:
 _dev_info+0x7c/0xa0
 arm_smmu_evtq_thread+0x1c0/0x230
 irq_thread_fn+0x30/0x80
 irq_thread+0x128/0x210
 kthread+0x134/0x138
 ret_from_fork+0x10/0x1c
Kernel panic - not syncing: softlockup: hung tasks

Fix this by calling cond_resched() after the event information is
printed.

Signed-off-by: Zhou Guanghui <zhouguanghui1@huawei.com>
Link: https://lore.kernel.org/r/20220119070754.26528-1-zhouguanghui1@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

PCI: aardvark: Fix support for MSI interrupts

[ Upstream commit b0b0b8b897f8e12b2368e868bd7cdc5742d5c5a9 ]

Aardvark hardware supports Multi-MSI and MSI_FLAG_MULTI_PCI_MSI is already
set for the MSI chip. But when allocating MSI interrupt numbers for
Multi-MSI, the numbers need to be properly aligned, otherwise endpoint
devices send MSI interrupt with incorrect numbers.

Fix this issue by using function bitmap_find_free_region() instead of
bitmap_find_next_zero_area().

To ensure that aligned MSI interrupt numbers are used by endpoint devices,
we cannot use Linux virtual irq numbers (as they are random and not
properly aligned). Instead we need to use the aligned hwirq numbers.

This change fixes receiving MSI interrupts on Armada 3720 boards and
allows using NVMe disks which use Multi-MSI feature with 3 interrupts.

Without this NVMe disks freeze booting as linux nvme-core.c is waiting
60s for an interrupt.

Link: https://lore.kernel.org/r/20220110015018.26359-4-kabel@kernel.org
Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: smartpqi: Fix kdump issue when controller is locked up

[ Upstream commit 3ada501d602abf02353445c03bb3258146445d90 ]

Avoid dropping into shell if the controller is in locked up state.

Driver issues SIS soft reset to bring back the controller to SIS mode while
OS boots into kdump mode.

If the controller is in lockup state, SIS soft reset does not work.

Since the controller lockup code has not been cleared, driver considers the
firmware is no longer up and running. Driver returns back an error code to
OS and the kdump fails.

Link: https://lore.kernel.org/r/164375212337.440833.11955356190354940369.stgit@brunhilda.pdev.net
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Mahesh Rajashekhara <mahesh.rajashekhara@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/amdgpu: Fix recursive locking warning

[ Upstream commit 447c7997b62a5115ba4da846dcdee4fc12298a6a ]

Noticed the below warning while running a pytorch workload on vega10
GPUs. Change to trylock to avoid conflicts with already held reservation
locks.

[  +0.000003] WARNING: possible recursive locking detected
[  +0.000003] 5.13.0-kfd-rajneesh #1030 Not tainted
[  +0.000004] --------------------------------------------
[  +0.000002] python/4822 is trying to acquire lock:
[  +0.000004] ffff932cd9a259f8 (reservation_ww_class_mutex){+.+.}-{3:3},
at: amdgpu_bo_release_notify+0xc4/0x160 [amdgpu]
[  +0.000203]
              but task is already holding lock:
[  +0.000003] ffff932cbb7181f8 (reservation_ww_class_mutex){+.+.}-{3:3},
at: ttm_eu_reserve_buffers+0x270/0x470 [ttm]
[  +0.000017]
              other info that might help us debug this:
[  +0.000002]  Possible unsafe locking scenario:

[  +0.000003]        CPU0
[  +0.000002]        ----
[  +0.000002]   lock(reservation_ww_class_mutex);
[  +0.000004]   lock(reservation_ww_class_mutex);
[  +0.000003]
               *** DEADLOCK ***

[  +0.000002]  May be due to missing lock nesting notation

[  +0.000003] 7 locks held by python/4822:
[  +0.000003]  #0: ffff932c4ac028d0 (&process->mutex){+.+.}-{3:3}, at:
kfd_ioctl_map_memory_to_gpu+0x10b/0x320 [amdgpu]
[  +0.000232]  #1: ffff932c55e830a8 (&info->lock#2){+.+.}-{3:3}, at:
amdgpu_amdkfd_gpuvm_map_memory_to_gpu+0x64/0xf60 [amdgpu]
[  +0.000241]  #2: ffff932cc45b5e68 (&(*mem)->lock){+.+.}-{3:3}, at:
amdgpu_amdkfd_gpuvm_map_memory_to_gpu+0xdf/0xf60 [amdgpu]
[  +0.000236]  #3: ffffb2b35606fd28
(reservation_ww_class_acquire){+.+.}-{0:0}, at:
amdgpu_amdkfd_gpuvm_map_memory_to_gpu+0x232/0xf60 [amdgpu]
[  +0.000235]  #4: ffff932cbb7181f8
(reservation_ww_class_mutex){+.+.}-{3:3}, at:
ttm_eu_reserve_buffers+0x270/0x470 [ttm]
[  +0.000015]  #5: ffffffffc045f700 (*(sspp++)){....}-{0:0}, at:
drm_dev_enter+0x5/0xa0 [drm]
[  +0.000038]  #6: ffff932c52da7078 (&vm->eviction_lock){+.+.}-{3:3},
at: amdgpu_vm_bo_update_mapping+0xd5/0x4f0 [amdgpu]
[  +0.000195]
              stack backtrace:
[  +0.000003] CPU: 11 PID: 4822 Comm: python Not tainted
5.13.0-kfd-rajneesh #1030
[  +0.000005] Hardware name: GIGABYTE MZ01-CE0-00/MZ01-CE0-00, BIOS F02
08/29/2018
[  +0.000003] Call Trace:
[  +0.000003]  dump_stack+0x6d/0x89
[  +0.000010]  __lock_acquire+0xb93/0x1a90
[  +0.000009]  lock_acquire+0x25d/0x2d0
[  +0.000005]  ? amdgpu_bo_release_notify+0xc4/0x160 [amdgpu]
[  +0.000184]  ? lock_is_held_type+0xa2/0x110
[  +0.000006]  ? amdgpu_bo_release_notify+0xc4/0x160 [amdgpu]
[  +0.000184]  __ww_mutex_lock.constprop.17+0xca/0x1060
[  +0.000007]  ? amdgpu_bo_release_notify+0xc4/0x160 [amdgpu]
[  +0.000183]  ? lock_release+0x13f/0x270
[  +0.000005]  ? lock_is_held_type+0xa2/0x110
[  +0.000006]  ? amdgpu_bo_release_notify+0xc4/0x160 [amdgpu]
[  +0.000183]  amdgpu_bo_release_notify+0xc4/0x160 [amdgpu]
[  +0.000185]  ttm_bo_release+0x4c6/0x580 [ttm]
[  +0.000010]  amdgpu_bo_unref+0x1a/0x30 [amdgpu]
[  +0.000183]  amdgpu_vm_free_table+0x76/0xa0 [amdgpu]
[  +0.000189]  amdgpu_vm_free_pts+0xb8/0xf0 [amdgpu]
[  +0.000189]  amdgpu_vm_update_ptes+0x411/0x770 [amdgpu]
[  +0.000191]  amdgpu_vm_bo_update_mapping+0x324/0x4f0 [amdgpu]
[  +0.000191]  amdgpu_vm_bo_update+0x251/0x610 [amdgpu]
[  +0.000191]  update_gpuvm_pte+0xcc/0x290 [amdgpu]
[  +0.000229]  ? amdgpu_vm_bo_map+0xd7/0x130 [amdgpu]
[  +0.000190]  amdgpu_amdkfd_gpuvm_map_memory_to_gpu+0x912/0xf60
[amdgpu]
[  +0.000234]  kfd_ioctl_map_memory_to_gpu+0x182/0x320 [amdgpu]
[  +0.000218]  kfd_ioctl+0x2b9/0x600 [amdgpu]
[  +0.000216]  ? kfd_ioctl_unmap_memory_from_gpu+0x270/0x270 [amdgpu]
[  +0.000216]  ? lock_release+0x13f/0x270
[  +0.000006]  ? __fget_files+0x107/0x1e0
[  +0.000007]  __x64_sys_ioctl+0x8b/0xd0
[  +0.000007]  do_syscall_64+0x36/0x70
[  +0.000004]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[  +0.000007] RIP: 0033:0x7fbff90a7317
[  +0.000004] Code: b3 66 90 48 8b 05 71 4b 2d 00 64 c7 00 26 00 00 00
48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f
05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 41 4b 2d 00 f7 d8 64 89 01 48
[  +0.000005] RSP: 002b:00007fbe301fe648 EFLAGS: 00000246 ORIG_RAX:
0000000000000010
[  +0.000006] RAX: ffffffffffffffda RBX: 00007fbcc402d820 RCX:
00007fbff90a7317
[  +0.000003] RDX: 00007fbe301fe690 RSI: 00000000c0184b18 RDI:
0000000000000004
[  +0.000003] RBP: 00007fbe301fe690 R08: 0000000000000000 R09:
00007fbcc402d880
[  +0.000003] R10: 0000000002001000 R11: 0000000000000246 R12:
00000000c0184b18
[  +0.000003] R13: 0000000000000004 R14: 00007fbf689593a0 R15:
00007fbcc402d820

Cc: Christian König <christian.koenig@amd.com>
Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: Alex Deucher <Alexander.Deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

powerpc: Set crashkernel offset to mid of RMA region

[ Upstream commit 7c5ed82b800d8615cdda00729e7b62e5899f0b13 ]

On large config LPARs (having 192 and more cores), Linux fails to boot
due to insufficient memory in the first memblock. It is due to the
memory reservation for the crash kernel which starts at 128MB offset of
the first memblock. This memory reservation for the crash kernel doesn't
leave enough space in the first memblock to accommodate other essential
system resources.

The crash kernel start address was set to 128MB offset by default to
ensure that the crash kernel get some memory below the RMA region which
is used to be of size 256MB. But given that the RMA region size can be
512MB or more, setting the crash kernel offset to mid of RMA size will
leave enough space for the kernel to allocate memory for other system
resources.

Since the above crash kernel offset change is only applicable to the LPAR
platform, the LPAR feature detection is pushed before the crash kernel
reservation. The rest of LPAR specific initialization will still
be done during pseries_probe_fw_features as usual.

This patch is dependent on changes to paca allocation for boot CPU. It
expect boot CPU to discover 1T segment support which is introduced by
the patch posted here:
https://lists.ozlabs.org/pipermail/linuxppc-dev/2022-January/239175.html

Reported-by: Abdul haleem <abdhalee@linux.vnet.ibm.com>
Signed-off-by: Sourabh Jain <sourabhjain@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20220204085601.107257-1-sourabhjain@linux.ibm.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

net: initialize init_net earlier

[ Upstream commit 9c1be1935fb68b2413796cdc03d019b8cf35ab51 ]

While testing a patch that will follow later
("net: add netns refcount tracker to struct nsproxy")
I found that devtmpfs_init() was called before init_net
was initialized.

This is a bug, because devtmpfs_setup() calls
ksys_unshare(CLONE_NEWNS);

This has the effect of increasing init_net refcount,
which will be later overwritten to 1, as part of setup_net(&init_net)

We had too many prior patches [1] trying to work around the root cause.

Really, make sure init_net is in BSS section, and that net_ns_init()
is called earlier at boot time.

Note that another patch ("vfs: add netns refcount tracker
to struct fs_context") also will need net_ns_init() being called
before vfs_caches_init()

As a bonus, this patch saves around 4KB in .data section.

[1]

f8c46cb39079 ("netns: do not call pernet ops for not yet set up init_net namespace")
b5082df8019a ("net: Initialise init_net.count to 1")
734b65417b24 ("net: Statically initialize init_net.dev_base_head")

v2: fixed a build error reported by kernel build bots (CONFIG_NET=n)

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ipv6: make mc_forwarding atomic

[ Upstream commit 145c7a793838add5e004e7d49a67654dc7eba147 ]

This fixes minor data-races in ip6_mc_input() and
batadv_mcast_mla_rtr_flags_softif_get_ipv6()

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

libbpf: Fix build issue with llvm-readelf

[ Upstream commit 0908a66ad1124c1634c33847ac662106f7f2c198 ]

There are cases where clang compiler is packaged in a way
readelf is a symbolic link to llvm-readelf. In such cases,
llvm-readelf will be used instead of default binutils readelf,
and the following error will appear during libbpf build:

#  Warning: Num of global symbols in
#   /home/yhs/work/bpf-next/tools/testing/selftests/bpf/tools/build/libbpf/sharedobjs/libbpf-in.o (367)
#   does NOT match with num of versioned symbols in
#   /home/yhs/work/bpf-next/tools/testing/selftests/bpf/tools/build/libbpf/libbpf.so libbpf.map (383).
#   Please make sure all LIBBPF_API symbols are versioned in libbpf.map.
#  --- /home/yhs/work/bpf-next/tools/testing/selftests/bpf/tools/build/libbpf/libbpf_global_syms.tmp ...
#  +++ /home/yhs/work/bpf-next/tools/testing/selftests/bpf/tools/build/libbpf/libbpf_versioned_syms.tmp ...
#  @@ -324,6 +324,22 @@
#   btf__str_by_offset
#   btf__type_by_id
#   btf__type_cnt
#  +LIBBPF_0.0.1
#  +LIBBPF_0.0.2
#  +LIBBPF_0.0.3
#  +LIBBPF_0.0.4
#  +LIBBPF_0.0.5
#  +LIBBPF_0.0.6
#  +LIBBPF_0.0.7
#  +LIBBPF_0.0.8
#  +LIBBPF_0.0.9
#  +LIBBPF_0.1.0
#  +LIBBPF_0.2.0
#  +LIBBPF_0.3.0
#  +LIBBPF_0.4.0
#  +LIBBPF_0.5.0
#  +LIBBPF_0.6.0
#  +LIBBPF_0.7.0
#   libbpf_attach_type_by_name
#   libbpf_find_kernel_btf
#   libbpf_find_vmlinux_btf_id
#  make[2]: *** [Makefile:184: check_abi] Error 1
#  make[1]: *** [Makefile:140: all] Error 2

The above failure is due to different printouts for some ABS
versioned symbols. For example, with the same libbpf.so,
  $ /bin/readelf --dyn-syms --wide tools/lib/bpf/libbpf.so | grep "LIBBPF" | grep ABS
     134: 0000000000000000     0 OBJECT  GLOBAL DEFAULT  ABS LIBBPF_0.5.0
     202: 0000000000000000     0 OBJECT  GLOBAL DEFAULT  ABS LIBBPF_0.6.0
     ...
  $ /opt/llvm/bin/readelf --dyn-syms --wide tools/lib/bpf/libbpf.so | grep "LIBBPF" | grep ABS
     134: 0000000000000000     0 OBJECT  GLOBAL DEFAULT   ABS LIBBPF_0.5.0@@LIBBPF_0.5.0
     202: 0000000000000000     0 OBJECT  GLOBAL DEFAULT   ABS LIBBPF_0.6.0@@LIBBPF_0.6.0
     ...
The binutils readelf doesn't print out the symbol LIBBPF_* version and llvm-readelf does.
Such a difference caused libbpf build failure with llvm-readelf.

The proposed fix filters out all ABS symbols as they are not part of the comparison.
This works for both binutils readelf and llvm-readelf.

Reported-by: Delyan Kratunov <delyank@fb.com>
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20220204214355.502108-1-yhs@fb.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

cfg80211: don't add non transmitted BSS to 6GHz scanned channels

[ Upstream commit 5666ee154f4696c011dfa8544aaf5591b6b87515 ]

When adding 6GHz channels to scan request based on reported
co-located APs, don't add channels that have only APs with
"non-transmitted" BSSes if they only match the wildcard SSID since
they will be found by probing the "transmitted" BSS.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20220202104617.f6ddf099f934.I231e55885d3644f292d00dfe0f42653269f2559e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mt76: dma: initialize skip_unmap in mt76_dma_rx_fill

[ Upstream commit 577298ec55dfc8b9aece54520f0258c3f93a6573 ]

Even if it is only a false-positive since skip_buf0/skip_buf1 are only
used in mt76_dma_tx_cleanup_idx routine, initialize skip_unmap in
mt76_dma_rx_fill in order to fix the following UBSAN report:

[   13.924906] UBSAN: invalid-load in linux-5.15.0/drivers/net/wireless/mediatek/mt76/dma.c:162:13
[   13.924909] load of value 225 is not a valid value for type '_Bool'
[   13.924912] CPU: 9 PID: 672 Comm: systemd-udevd Not tainted 5.15.0-18-generic #18-Ubuntu
[   13.924914] Hardware name: LENOVO 21A0000CMX/21A0000CMX, BIOS R1MET43W (1.13 ) 11/05/2021
[   13.924915] Call Trace:
[   13.924917]  <TASK>
[   13.924920]  show_stack+0x52/0x58
[   13.924925]  dump_stack_lvl+0x4a/0x5f
[   13.924931]  dump_stack+0x10/0x12
[   13.924932]  ubsan_epilogue+0x9/0x45
[   13.924934]  __ubsan_handle_load_invalid_value.cold+0x44/0x49
[   13.924935]  ? __iommu_dma_map+0x84/0xf0
[   13.924939]  mt76_dma_add_buf.constprop.0.cold+0x23/0x85 [mt76]
[   13.924949]  mt76_dma_rx_fill.isra.0+0x102/0x1f0 [mt76]
[   13.924954]  mt76_dma_init+0xc9/0x150 [mt76]
[   13.924959]  ? mt7921_dma_enable+0x110/0x110 [mt7921e]
[   13.924966]  mt7921_dma_init+0x1e3/0x260 [mt7921e]
[   13.924970]  mt7921_register_device+0x29d/0x510 [mt7921e]
[   13.924975]  mt7921_pci_probe.part.0+0x17f/0x1b0 [mt7921e]
[   13.924980]  mt7921_pci_probe+0x43/0x60 [mt7921e]
[   13.924984]  local_pci_probe+0x4b/0x90
[   13.924987]  pci_device_probe+0x115/0x1f0
[   13.924989]  really_probe+0x21e/0x420
[   13.924992]  __driver_probe_device+0x115/0x190
[   13.924994]  driver_probe_device+0x23/0xc0
[   13.924996]  __driver_attach+0xbd/0x1d0
[   13.924998]  ? __device_attach_driver+0x110/0x110
[   13.924999]  bus_for_each_dev+0x7e/0xc0
[   13.925001]  driver_attach+0x1e/0x20
[   13.925003]  bus_add_driver+0x135/0x200
[   13.925005]  driver_register+0x95/0xf0
[   13.925008]  ? 0xffffffffc0766000
[   13.925010]  __pci_register_driver+0x68/0x70
[   13.925011]  mt7921_pci_driver_init+0x23/0x1000 [mt7921e]
[   13.925015]  do_one_initcall+0x48/0x1d0
[   13.925019]  ? kmem_cache_alloc_trace+0x19e/0x2e0
[   13.925022]  do_init_module+0x62/0x280
[   13.925025]  load_module+0xac9/0xbb0
[   13.925027]  __do_sys_finit_module+0xbf/0x120
[   13.925029]  __x64_sys_finit_module+0x18/0x20
[   13.925030]  do_syscall_64+0x5c/0xc0
[   13.925033]  ? do_syscall_64+0x69/0xc0
[   13.925034]  ? sysvec_reschedule_ipi+0x78/0xe0
[   13.925036]  ? asm_sysvec_reschedule_ipi+0xa/0x20
[   13.925039]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[   13.925040] RIP: 0033:0x7fbf2b90f94d
[   13.925045] RSP: 002b:00007ffe2ec7e5d8 EFLAGS: 00000246 ORIG_RAX: 0000000000000139
[   13.925047] RAX: ffffffffffffffda RBX: 000056106b0634e0 RCX: 00007fbf2b90f94d
[   13.925048] RDX: 0000000000000000 RSI: 00007fbf2baa3441 RDI: 0000000000000013
[   13.925049] RBP: 0000000000020000 R08: 0000000000000000 R09: 0000000000000002
[   13.925050] R10: 0000000000000013 R11: 0000000000000246 R12: 00007fbf2baa3441
[   13.925051] R13: 000056106b062620 R14: 000056106b0610c0 R15: 000056106b0640d0
[   13.925053]  </TASK>

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mt76: mt7921: fix crash when startup fails.

[ Upstream commit 827e7799c61b978fbc2cc9dac66cb62401b2b3f0 ]

If the nic fails to start, it is possible that the
reset_work has already been scheduled.  Ensure the
work item is canceled so we do not have use-after-free
crash in case cleanup is called before the work item
is executed.

This fixes crash on my x86_64 apu2 when mt7921k radio
fails to work.  Radio still fails, but OS does not
crash.

Signed-off-by: Ben Greear <greearb@candelatech.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sasha Levin <sashal@kernel.org>

power: supply: axp20x_battery: properly report current when discharging

[ Upstream commit d4f408cdcd26921c1268cb8dcbe8ffb6faf837f3 ]

As stated in [1], negative current values are used for discharging
batteries.

AXP PMICs internally have two different ADC channels for shunt current
measurement: one used during charging and one during discharging.
The values reported by these ADCs are unsigned.
While the driver properly selects ADC channel to get the data from,
it doesn't apply negative sign when reporting discharging current.

[1] Documentation/ABI/testing/sysfs-class-power

Signed-off-by: Evgeny Boger <boger@wirenboard.com>
Acked-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/v3d: fix missing unlock

[ Upstream commit e57c1a3bd5e8e0c7181f65ae55581f0236a8f284 ]

[why]
Unlock is needed on the error handling path to prevent dead lock.
v3d_submit_cl_ioctl and v3d_submit_csd_ioctl is missing unlock.

[how]
Fix this by changing goto target on the error handling path. So
changing the goto to target an error handling path
that includes drm_gem_unlock reservations.

Signed-off-by: Yongzhi Liu <lyz_cs@pku.edu.cn>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Signed-off-by: Melissa Wen <melissa.srw@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1643377262-109975-1-git-send-email-lyz_cs@pku.edu.cn
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: bfa: Replace snprintf() with sysfs_emit()

[ Upstream commit 2245ea91fd3a04cafbe2f54911432a8657528c3b ]

coccinelle report:
./drivers/scsi/bfa/bfad_attr.c:908:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:860:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:888:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:853:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:808:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:728:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:822:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:927:9-17:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:900:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:874:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:714:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:839:8-16:
WARNING: use scnprintf or sprintf

Use sysfs_emit() instead of scnprintf() or sprintf().

Link: https://lore.kernel.org/r/def83ff75faec64ba592b867a8499b1367bae303.1643181468.git.yang.guang5@zte.com.cn
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Yang Guang <yang.guang5@zte.com.cn>
Signed-off-by: David Yang <davidcomponentone@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: mvsas: Replace snprintf() with sysfs_emit()

[ Upstream commit 0ad3867b0f13e45cfee5a1298bfd40eef096116c ]

coccinelle report:
./drivers/scsi/mvsas/mv_init.c:699:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/mvsas/mv_init.c:747:8-16:
WARNING: use scnprintf or sprintf

Use sysfs_emit() instead of scnprintf() or sprintf().

Link: https://lore.kernel.org/r/c1711f7cf251730a8ceb5bdfc313bf85662b3395.1643182948.git.yang.guang5@zte.com.cn
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Yang Guang <yang.guang5@zte.com.cn>
Signed-off-by: David Yang <davidcomponentone@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

bpf: Make dst_port field in struct bpf_sock 16-bit wide

[ Upstream commit 4421a582718ab81608d8486734c18083b822390d ]

Menglong Dong reports that the documentation for the dst_port field in
struct bpf_sock is inaccurate and confusing. From the BPF program PoV, the
field is a zero-padded 16-bit integer in network byte order. The value
appears to the BPF user as if laid out in memory as so:

  offsetof(struct bpf_sock, dst_port) + 0  <port MSB>
                                      + 8  <port LSB>
                                      +16  0x00
                                      +24  0x00

32-, 16-, and 8-bit wide loads from the field are all allowed, but only if
the offset into the field is 0.

32-bit wide loads from dst_port are especially confusing. The loaded value,
after converting to host byte order with bpf_ntohl(dst_port), contains the
port number in the upper 16-bits.

Remove the confusion by splitting the field into two 16-bit fields. For
backward compatibility, allow 32-bit wide loads from offsetof(struct
bpf_sock, dst_port).

While at it, allow loads 8-bit loads at offset [0] and [1] from dst_port.

Reported-by: Menglong Dong <imagedong@tencent.com>
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/r/20220130115518.213259-2-jakub@cloudflare.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/bridge: Add missing pm_runtime_put_sync

[ Upstream commit 46f47807738441e354873546dde0b000106c068a ]

pm_runtime_get_sync() will increase the rumtime PM counter
even when it returns an error. Thus a pairing decrement is needed
to prevent refcount leak. Fix this by replacing this API with
pm_runtime_resume_and_get(), which will not change the runtime
PM counter on error. Besides, a matching decrement is needed
on the error handling path to keep the counter balanced.

Signed-off-by: Yongzhi Liu <lyz_cs@pku.edu.cn>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Robert Foss <robert.foss@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/1643008835-73961-1-git-send-email-lyz_cs@pku.edu.cn
Signed-off-by: Sasha Levin <sashal@kernel.org>

net/smc: Send directly when TCP_CORK is cleared

[ Upstream commit ea785a1a573b390a150010b3c5b81e1ccd8c98a8 ]

According to the man page of TCP_CORK [1], if set, don't send out
partial frames. All queued partial frames are sent when option is
cleared again.

When applications call setsockopt to disable TCP_CORK, this call is
protected by lock_sock(), and tries to mod_delayed_work() to 0, in order
to send pending data right now. However, the delayed work smc_tx_work is
also protected by lock_sock(). There introduces lock contention for
sending data.

To fix it, send pending data directly which acts like TCP, without
lock_sock() protected in the context of setsockopt (already lock_sock()ed),
and cancel unnecessary dealyed work, which is protected by lock.

[1] https://linux.die.net/man/7/tcp

Signed-off-by: Tony Lu <tonylu@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ath11k: mhi: use mhi_sync_power_up()

[ Upstream commit 3df6d74aedfdca919cca475d15dfdbc8b05c9e5d ]

If amss.bin was missing ath11k would crash during 'rmmod ath11k_pci'. The
reason for that was that we were using mhi_async_power_up() which does not
check any errors. But mhi_sync_power_up() on the other hand does check for
errors so let's use that to fix the crash.

I was not able to find a reason why an async version was used.
ath11k_mhi_start() (which enables state ATH11K_MHI_POWER_ON) is called from
ath11k_hif_power_up(), which can sleep. So sync version should be safe to use
here.

[  145.569731] general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC KASAN PTI
[  145.569789] KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
[  145.569843] CPU: 2 PID: 1628 Comm: rmmod Kdump: loaded Tainted: G        W         5.16.0-wt-ath+ #567
[  145.569898] Hardware name: Intel(R) Client Systems NUC8i7HVK/NUC8i7HVB, BIOS HNKBLi70.86A.0067.2021.0528.1339 05/28/2021
[  145.569956] RIP: 0010:ath11k_hal_srng_access_begin+0xb5/0x2b0 [ath11k]
[  145.570028] Code: df 48 89 fa 48 c1 ea 03 80 3c 02 00 0f 85 ec 01 00 00 48 8b ab a8 00 00 00 48 b8 00 00 00 00 00 fc ff df 48 89 ea 48 c1 ea 03 <0f> b6 14 02 48 89 e8 83 e0 07 83 c0 03 45 85 ed 75 48 38 d0 7c 08
[  145.570089] RSP: 0018:ffffc900025d7ac0 EFLAGS: 00010246
[  145.570144] RAX: dffffc0000000000 RBX: ffff88814fca2dd8 RCX: 1ffffffff50cb455
[  145.570196] RDX: 0000000000000000 RSI: ffff88814fca2dd8 RDI: ffff88814fca2e80
[  145.570252] RBP: 0000000000000000 R08: 0000000000000000 R09: ffffffffa8659497
[  145.570329] R10: fffffbfff50cb292 R11: 0000000000000001 R12: ffff88814fca0000
[  145.570410] R13: 0000000000000000 R14: ffff88814fca2798 R15: ffff88814fca2dd8
[  145.570465] FS:  00007fa399988540(0000) GS:ffff888233e00000(0000) knlGS:0000000000000000
[  145.570519] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  145.570571] CR2: 00007fa399b51421 CR3: 0000000137898002 CR4: 00000000003706e0
[  145.570623] Call Trace:
[  145.570675]  <TASK>
[  145.570727]  ? ath11k_ce_tx_process_cb+0x34b/0x860 [ath11k]
[  145.570797]  ath11k_ce_tx_process_cb+0x356/0x860 [ath11k]
[  145.570864]  ? tasklet_init+0x150/0x150
[  145.570919]  ? ath11k_ce_alloc_pipes+0x280/0x280 [ath11k]
[  145.570986]  ? tasklet_clear_sched+0x42/0xe0
[  145.571042]  ? tasklet_kill+0xe9/0x1b0
[  145.571095]  ? tasklet_clear_sched+0xe0/0xe0
[  145.571148]  ? irq_has_action+0x120/0x120
[  145.571202]  ath11k_ce_cleanup_pipes+0x45a/0x580 [ath11k]
[  145.571270]  ? ath11k_pci_stop+0x10e/0x170 [ath11k_pci]
[  145.571345]  ath11k_core_stop+0x8a/0xc0 [ath11k]
[  145.571434]  ath11k_core_deinit+0x9e/0x150 [ath11k]
[  145.571499]  ath11k_pci_remove+0xd2/0x260 [ath11k_pci]
[  145.571553]  pci_device_remove+0x9a/0x1c0
[  145.571605]  __device_release_driver+0x332/0x660
[  145.571659]  driver_detach+0x1e7/0x2c0
[  145.571712]  bus_remove_driver+0xe2/0x2d0
[  145.571772]  pci_unregister_driver+0x21/0x250
[  145.571826]  __do_sys_delete_module+0x30a/0x4b0
[  145.571879]  ? free_module+0xac0/0xac0
[  145.571933]  ? lockdep_hardirqs_on_prepare.part.0+0x18c/0x370
[  145.571986]  ? syscall_enter_from_user_mode+0x1d/0x50
[  145.572039]  ? lockdep_hardirqs_on+0x79/0x100
[  145.572097]  do_syscall_64+0x3b/0x90
[  145.572153]  entry_SYSCALL_64_after_hwframe+0x44/0xae

Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03003-QCAHSPSWPL_V1_V2_SILICONZ_LITE-2

Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20220127090117.2024-2-kvalo@kernel.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

ath11k: pci: fix crash on suspend if board file is not found

[ Upstream commit b4f4c56459a5c744f7f066b9fc2b54ea995030c5 ]

Mario reported that the kernel was crashing on suspend if ath11k was not able
to find a board file:

[  473.693286] PM: Suspending system (s2idle)
[  473.693291] printk: Suspending console(s) (use no_console_suspend to debug)
[  474.407787] BUG: unable to handle page fault for address: 0000000000002070
[  474.407791] #PF: supervisor read access in kernel mode
[  474.407794] #PF: error_code(0x0000) - not-present page
[  474.407798] PGD 0 P4D 0
[  474.407801] Oops: 0000 [#1] PREEMPT SMP NOPTI
[  474.407805] CPU: 2 PID: 2350 Comm: kworker/u32:14 Tainted: G        W         5.16.0 #248
[...]
[  474.407868] Call Trace:
[  474.407870]  <TASK>
[  474.407874]  ? _raw_spin_lock_irqsave+0x2a/0x60
[  474.407882]  ? lock_timer_base+0x72/0xa0
[  474.407889]  ? _raw_spin_unlock_irqrestore+0x29/0x3d
[  474.407892]  ? try_to_del_timer_sync+0x54/0x80
[  474.407896]  ath11k_dp_rx_pktlog_stop+0x49/0xc0 [ath11k]
[  474.407912]  ath11k_core_suspend+0x34/0x130 [ath11k]
[  474.407923]  ath11k_pci_pm_suspend+0x1b/0x50 [ath11k_pci]
[  474.407928]  pci_pm_suspend+0x7e/0x170
[  474.407935]  ? pci_pm_freeze+0xc0/0xc0
[  474.407939]  dpm_run_callback+0x4e/0x150
[  474.407947]  __device_suspend+0x148/0x4c0
[  474.407951]  async_suspend+0x20/0x90
dmesg-efi-164255130401001:
Oops#1 Part1
[  474.407955]  async_run_entry_fn+0x33/0x120
[  474.407959]  process_one_work+0x220/0x3f0
[  474.407966]  worker_thread+0x4a/0x3d0
[  474.407971]  kthread+0x17a/0x1a0
[  474.407975]  ? process_one_work+0x3f0/0x3f0
[  474.407979]  ? set_kthread_struct+0x40/0x40
[  474.407983]  ret_from_fork+0x22/0x30
[  474.407991]  </TASK>

The issue here is that board file loading happens after ath11k_pci_probe()
succesfully returns (ath11k initialisation happends asynchronously) and the
suspend handler is still enabled, of course failing as ath11k is not properly
initialised. Fix this by checking ATH11K_FLAG_QMI_FAIL during both suspend and
resume.

Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03003-QCAHSPSWPL_V1_V2_SILICONZ_LITE-2

Reported-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=215504
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20220127090117.2024-1-kvalo@kernel.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

ath11k: fix kernel panic during unload/load ath11k modules

[ Upstream commit 22b59cb965f79ee1accf83172441c9ca0ecb632a ]

Call netif_napi_del() from ath11k_ahb_free_ext_irq() to fix
the following kernel panic when unload/load ath11k modules
for few iterations.

[  971.201365] Unable to handle kernel paging request at virtual address 6d97a208
[  971.204227] pgd = 594c2919
[  971.211478] [6d97a208] *pgd=00000000
[  971.214120] Internal error: Oops: 5 [#1] PREEMPT SMP ARM
[  971.412024] CPU: 2 PID: 4435 Comm: insmod Not tainted 5.4.89 #0
[  971.434256] Hardware name: Generic DT based system
[  971.440165] PC is at napi_by_id+0x10/0x40
[  971.445019] LR is at netif_napi_add+0x160/0x1dc

[  971.743127] (napi_by_id) from [<807d89a0>] (netif_napi_add+0x160/0x1dc)
[  971.751295] (netif_napi_add) from [<7f1209ac>] (ath11k_ahb_config_irq+0xf8/0x414 [ath11k_ahb])
[  971.759164] (ath11k_ahb_config_irq [ath11k_ahb]) from [<7f12135c>] (ath11k_ahb_probe+0x40c/0x51c [ath11k_ahb])
[  971.768567] (ath11k_ahb_probe [ath11k_ahb]) from [<80666864>] (platform_drv_probe+0x48/0x94)
[  971.779670] (platform_drv_probe) from [<80664718>] (really_probe+0x1c8/0x450)
[  971.789389] (really_probe) from [<80664cc4>] (driver_probe_device+0x15c/0x1b8)
[  971.797547] (driver_probe_device) from [<80664f60>] (device_driver_attach+0x44/0x60)
[  971.805795] (device_driver_attach) from [<806650a0>] (__driver_attach+0x124/0x140)
[  971.814822] (__driver_attach) from [<80662adc>] (bus_for_each_dev+0x58/0xa4)
[  971.823328] (bus_for_each_dev) from [<80663a2c>] (bus_add_driver+0xf0/0x1e8)
[  971.831662] (bus_add_driver) from [<806658a4>] (driver_register+0xa8/0xf0)
[  971.839822] (driver_register) from [<8030269c>] (do_one_initcall+0x78/0x1ac)
[  971.847638] (do_one_initcall) from [<80392524>] (do_init_module+0x54/0x200)
[  971.855968] (do_init_module) from [<803945b0>] (load_module+0x1e30/0x1ffc)
[  971.864126] (load_module) from [<803948b0>] (sys_init_module+0x134/0x17c)
[  971.871852] (sys_init_module) from [<80301000>] (ret_fast_syscall+0x0/0x50)

Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.6.0.1-00760-QCAHKSWPL_SILICONZ-1

Signed-off-by: Venkateswara Naralasetty <quic_vnaralas@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/1642583973-21599-1-git-send-email-quic_vnaralas@quicinc.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

powerpc: dts: t104xrdb: fix phy type for FMAN 4/5

[ Upstream commit 17846485dff91acce1ad47b508b633dffc32e838 ]

T1040RDB has two RTL8211E-VB phys which requires setting
of internal delays for correct work.

Changing the phy-connection-type property to `rgmii-id`
will fix this issue.

Signed-off-by: Maxim Kiselev <bigunclemax@gmail.com>
Reviewed-by: Maxim Kochetkov <fido_max@inbox.ru>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211230151123.1258321-1-bigunclemax@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/amdkfd: Don't take process mutex for svm ioctls

[ Upstream commit ac7c48c0cce00d03b3c95fddcccb0a45257e33e3 ]

SVM ioctls take proper svms->lock to handle race conditions, don't need
take process mutex to serialize ioctls. This also fixes circular locking
warning:

WARNING: possible circular locking dependency detected

  Possible unsafe locking scenario:

        CPU0                    CPU1
        ----                    ----
   lock((work_completion)(&svms->deferred_list_work));
                                lock(&process->mutex);
                     lock((work_completion)(&svms->deferred_list_work));
   lock(&process->mutex);

   *** DEADLOCK ***

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ptp: replace snprintf with sysfs_emit

[ Upstream commit e2cf07654efb0fd7bbcb475c6f74be7b5755a8fd ]

coccinelle report:
./drivers/ptp/ptp_sysfs.c:17:8-16:
WARNING: use scnprintf or sprintf
./drivers/ptp/ptp_sysfs.c:390:8-16:
WARNING: use scnprintf or sprintf

Use sysfs_emit instead of scnprintf or sprintf makes more sense.

Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Yang Guang <yang.guang5@zte.com.cn>
Signed-off-by: David Yang <davidcomponentone@gmail.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

usb: cdnsp: fix cdnsp_decode_trb function to properly handle ret value

[ Upstream commit 03db9289b5ab59437e42a111a34545a7cedb5190 ]

Variable ret in function cdnsp_decode_trb is initialized but not
used. To fix this compiler warning patch adds checking whether the
data buffer has not been overflowed.

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Pawel Laszczak <pawell@cadence.com>
Link: https://lore.kernel.org/r/20220112053237.14309-1-pawell@gli-login.cadence.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

usb: gadget: tegra-xudc: Fix control endpoint's definitions

[ Upstream commit 7bd42fb95eb4f98495ccadf467ad15124208ec49 ]

According to the Tegra Technical Reference Manual, the seq_num
field of control endpoint is not [31:24] but [31:27]. Bit 24
is reserved and bit 26 is splitxstate.

The change fixes the wrong control endpoint's definitions.

Signed-off-by: Wayne Chang <waynec@nvidia.com>
Link: https://lore.kernel.org/r/20220107091349.149798-1-waynec@nvidia.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

usb: gadget: tegra-xudc: Do not program SPARAM

[ Upstream commit 62fb61580eb48fc890b7bc9fb5fd263367baeca8 ]

According to the Tegra Technical Reference Manual, SPARAM
is a read-only register and should not be programmed in
the driver.

The change removes the wrong SPARAM usage.

Signed-off-by: Wayne Chang <waynec@nvidia.com>
Link: https://lore.kernel.org/r/20220107090443.149021-1-waynec@nvidia.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/amd/display: Use PSR version selected during set_psr_caps

[ Upstream commit b80ddeb29d9df449f875f0b6f5de08d7537c02b8 ]

[Why]
If the DPCD caps specifies a PSR version newer than PSR_VERSION_1 then
we fallback to using PSR_VERSION_1 in amdgpu_dm_set_psr_caps.

This gets overriden with the raw DPCD value in amdgpu_dm_link_setup_psr,
which can result in DMCUB hanging if we pass in an unsupported PSR
version number.

[How]
Fix the hang by using link->psr_settings.psr_version directly during
amdgpu_dm_link_setup_psr.

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/amd/display: Fix memory leak

[ Upstream commit 5d5c6dba2b43e28845d7d7ed32a36802329a5f52 ]

[why]
Resource release is needed on the error handling path
to prevent memory leak.

[how]
Fix this by adding kfree on the error handling path.

Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Yongzhi Liu <lyz_cs@pku.edu.cn>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/amd/amdgpu/amdgpu_cs: fix refcount leak of a dma_fence obj

[ Upstream commit dfced44f122c500004a48ecc8db516bb6a295a1b ]

This issue takes place in an error path in
amdgpu_cs_fence_to_handle_ioctl(). When `info->in.what` falls into
default case, the function simply returns -EINVAL, forgetting to
decrement the reference count of a dma_fence obj, which is bumped
earlier by amdgpu_cs_get_fence(). This may result in reference count
leaks.

Fix it by decreasing the refcount of specific object before returning
the error code.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Xin Xiong <xiongx18@fudan.edu.cn>
Signed-off-by: Xin Tan <tanxin.ctf@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/amd/display: Add signal type check when verify stream backends same

[ Upstream commit 047db281c026de5971cedb5bb486aa29bd16a39d ]

[Why]
For allow eDP hot-plug feature, the stream signal may change to VIRTUAL
when plug-out and back to eDP when plug-in. OS will still setPathMode
with same timing for each plugging, but eDP gets no stream update as we
don't check signal type changing back as keeping it VIRTUAL. It's also
unsafe for future cases that stream signal is switched with same timing.

[How]
Check stream signal type change include previous HDMI signal case.

Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Acked-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Dale Zhao <dale.zhao@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ath5k: fix OOB in ath5k_eeprom_read_pcal_info_5111

[ Upstream commit 564d4eceb97eaf381dd6ef6470b06377bb50c95a ]

The bug was found during fuzzing. Stacktrace locates it in
ath5k_eeprom_convert_pcal_info_5111.
When none of the curve is selected in the loop, idx can go
up to AR5K_EEPROM_N_PD_CURVES. The line makes pd out of bound.
pd = &chinfo[pier].pd_curves[idx];

There are many OOB writes using pd later in the code. So I
added a sanity check for idx. Checks for other loops involving
AR5K_EEPROM_N_PD_CURVES are not needed as the loop index is not
used outside the loops.

The patch is NOT tested with real device.

The following is the fuzzing report

BUG: KASAN: slab-out-of-bounds in ath5k_eeprom_read_pcal_info_5111+0x126a/0x1390 [ath5k]
Write of size 1 at addr ffff8880174a4d60 by task modprobe/214

CPU: 0 PID: 214 Comm: modprobe Not tainted 5.6.0 #1
Call Trace:
 dump_stack+0x76/0xa0
 print_address_description.constprop.0+0x16/0x200
 ? ath5k_eeprom_read_pcal_info_5111+0x126a/0x1390 [ath5k]
 ? ath5k_eeprom_read_pcal_info_5111+0x126a/0x1390 [ath5k]
 __kasan_report.cold+0x37/0x7c
 ? ath5k_eeprom_read_pcal_info_5111+0x126a/0x1390 [ath5k]
 kasan_report+0xe/0x20
 ath5k_eeprom_read_pcal_info_5111+0x126a/0x1390 [ath5k]
 ? apic_timer_interrupt+0xa/0x20
 ? ath5k_eeprom_init_11a_pcal_freq+0xbc0/0xbc0 [ath5k]
 ? ath5k_pci_eeprom_read+0x228/0x3c0 [ath5k]
 ath5k_eeprom_init+0x2513/0x6290 [ath5k]
 ? ath5k_eeprom_init_11a_pcal_freq+0xbc0/0xbc0 [ath5k]
 ? usleep_range+0xb8/0x100
 ? apic_timer_interrupt+0xa/0x20
 ? ath5k_eeprom_read_pcal_info_2413+0x2f20/0x2f20 [ath5k]
 ath5k_hw_init+0xb60/0x1970 [ath5k]
 ath5k_init_ah+0x6fe/0x2530 [ath5k]
 ? kasprintf+0xa6/0xe0
 ? ath5k_stop+0x140/0x140 [ath5k]
 ? _dev_notice+0xf6/0xf6
 ? apic_timer_interrupt+0xa/0x20
 ath5k_pci_probe.cold+0x29a/0x3d6 [ath5k]
 ? ath5k_pci_eeprom_read+0x3c0/0x3c0 [ath5k]
 ? mutex_lock+0x89/0xd0
 ? ath5k_pci_eeprom_read+0x3c0/0x3c0 [ath5k]
 local_pci_probe+0xd3/0x160
 pci_device_probe+0x23f/0x3e0
 ? pci_device_remove+0x280/0x280
 ? pci_device_remove+0x280/0x280
 really_probe+0x209/0x5d0

Reported-by: Brendan Dolan-Gavitt <brendandg@nyu.edu>
Signed-off-by: Zekun Shen <bruceshenzk@gmail.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/YckvDdj3mtCkDRIt@a-10-27-26-18.dynapool.vpn.nyu.edu
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm: Add orientation quirk for GPD Win Max

[ Upstream commit 0b464ca3e0dd3cec65f28bc6d396d82f19080f69 ]

Panel is 800x1280, but mounted on a laptop form factor, sideways.

Signed-off-by: Anisse Astier <anisse@astier.eu>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211229222200.53128-3-anisse@astier.eu
Signed-off-by: Sasha Levin <sashal@kernel.org>

KVM: x86/emulator: Emulate RDPID only if it is enabled in guest

[ Upstream commit a836839cbfe60dc434c5476a7429cf2bae36415d ]

When RDTSCP is supported but RDPID is not supported in host,
RDPID emulation is available. However, __kvm_get_msr() would
only fail when RDTSCP/RDPID both are disabled in guest, so
the emulator wouldn't inject a #UD when RDPID is disabled but
RDTSCP is enabled in guest.

Fixes: fb6d4d340e05 ("KVM: x86: emulate RDPID")
Signed-off-by: Hou Wenlong <houwenlong.hwl@antgroup.com>
Message-Id: <1dfd46ae5b76d3ed87bde3154d51c64ea64c99c1.1646226788.git.houwenlong.hwl@antgroup.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

KVM: x86/pmu: Fix and isolate TSX-specific performance event logic

[ Upstream commit e644896f5106aa3f6d7e8c7adf2e4dc0fce53555 ]

HSW_IN_TX* bits are used in generic code which are not supported on
AMD. Worse, these bits overlap with AMD EventSelect[11:8] and hence
using HSW_IN_TX* bits unconditionally in generic code is resulting in
unintentional pmu behavior on AMD. For example, if EventSelect[11:8]
is 0x2, pmc_reprogram_counter() wrongly assumes that
HSW_IN_TX_CHECKPOINTED is set and thus forces sampling period to be 0.

Also per the SDM, both bits 32 and 33 "may only be set if the processor
supports HLE or RTM" and for "IN_TXCP (bit 33): this bit may only be set
for IA32_PERFEVTSEL2."

Opportunistically eliminate code redundancy, because if the HSW_IN_TX*
bit is set in pmc->eventsel, it is already set in attr.config.

Reported-by: Ravi Bangoria <ravi.bangoria@amd.com>
Reported-by: Jim Mattson <jmattson@google.com>
Fixes: 103af0a98788 ("perf, kvm: Support the in_tx/in_tx_cp modifiers in KVM arch perfmon emulation v5")
Co-developed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@amd.com>
Signed-off-by: Like Xu <likexu@tencent.com>
Message-Id: <20220309084257.88931-1-likexu@tencent.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

KVM: x86/svm: Clear reserved bits written to PerfEvtSeln MSRs

[ Upstream commit 9b026073db2f1ad0e4d8b61c83316c8497981037 ]

AMD EPYC CPUs never raise a #GP for a WRMSR to a PerfEvtSeln MSR. Some
reserved bits are cleared, and some are not. Specifically, on
Zen3/Milan, bits 19 and 42 are not cleared.

When emulating such a WRMSR, KVM should not synthesize a #GP,
regardless of which bits are set. However, undocumented bits should
not be passed through to the hardware MSR. So, rather than checking
for reserved bits and synthesizing a #GP, just clear the reserved
bits.

This may seem pedantic, but since KVM currently does not support the
"Host/Guest Only" bits (41:40), it is necessary to clear these bits
rather than synthesizing #GP, because some popular guests (e.g Linux)
will set the "Host Only" bit even on CPUs that don't support
EFER.SVME, and they don't expect a #GP.

For example,

root@Ubuntu1804:~# perf stat -e r26 -a sleep 1

 Performance counter stats for 'system wide':

                 0      r26

       1.001070977 seconds time elapsed

Feb 23 03:59:58 Ubuntu1804 kernel: [  405.379957] unchecked MSR access error: WRMSR to 0xc0010200 (tried to write 0x0000020000130026) at rIP: 0xffffffff9b276a28 (native_write_msr+0x8/0x30)
Feb 23 03:59:58 Ubuntu1804 kernel: [  405.379958] Call Trace:
Feb 23 03:59:58 Ubuntu1804 kernel: [  405.379963]  amd_pmu_disable_event+0x27/0x90

Fixes: ca724305a2b0 ("KVM: x86/vPMU: Implement AMD vPMU code for KVM")
Reported-by: Lotus Fenn <lotusf@google.com>
Signed-off-by: Jim Mattson <jmattson@google.com>
Reviewed-by: Like Xu <likexu@tencent.com>
Reviewed-by: David Dunn <daviddunn@google.com>
Message-Id: <20220226234131.2167175-1-jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

KVM: SVM: Fix kvm_cache_regs.h inclusions for is_guest_mode()

[ Upstream commit 4a9e7b9ea252842bc8b14d495706ac6317fafd5d ]

Include kvm_cache_regs.h to pick up the definition of is_guest_mode(),
which is referenced by nested_svm_virtualize_tpr() in svm.h. Remove
include from svm_onhpyerv.c which was done only because of lack of
include in svm.h.

Fixes: 883b0a91f41ab ("KVM: SVM: Move Nested SVM Implementation to nested.c")
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Sean Christopherson <seanjc@google.com>
Cc: kvm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Peter Gonda <pgonda@google.com>
Message-Id: <20220304161032.2270688-1-pgonda@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

KVM: x86/pmu: Use different raw event masks for AMD and Intel

[ Upstream commit 95b065bf5c431c06c68056a03a5853b660640ecc ]

The third nybble of AMD's event select overlaps with Intel's IN_TX and
IN_TXCP bits. Therefore, we can't use AMD64_RAW_EVENT_MASK on Intel
platforms that support TSX.

Declare a raw_event_mask in the kvm_pmu structure, initialize it in
the vendor-specific pmu_refresh() functions, and use that mask for
PERF_TYPE_RAW configurations in reprogram_gp_counter().

Fixes: 710c47651431 ("KVM: x86/pmu: Use AMD64_RAW_EVENT_MASK for PERF_TYPE_RAW")
Signed-off-by: Jim Mattson <jmattson@google.com>
Message-Id: <20220308012452.3468611-1-jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

kfence: limit currently covered allocations when pool nearly full

[ Upstream commit 08f6b10630f284755087f58aa393402e15b92977 ]

One of KFENCE's main design principles is that with increasing uptime,
allocation coverage increases sufficiently to detect previously
undetected bugs.

We have observed that frequent long-lived allocations of the same source
(e.g.  pagecache) tend to permanently fill up the KFENCE pool with
increasing system uptime, thus breaking the above requirement.  The
workaround thus far had been increasing the sample interval and/or
increasing the KFENCE pool size, but is no reliable solution.

To ensure diverse coverage of allocations, limit currently covered
allocations of the same source once pool utilization reaches 75%
(configurable via `kfence.skip_covered_thresh`) or above.  The effect is
retaining reasonable allocation coverage when the pool is close to full.

A side-effect is that this also limits frequent long-lived allocations
of the same source filling up the pool permanently.

Uniqueness of an allocation for coverage purposes is based on its
(partial) allocation stack trace (the source).  A Counting Bloom filter
is used to check if an allocation is covered; if the allocation is
currently covered, the allocation is skipped by KFENCE.

Testing was done using:

(a) a synthetic workload that performs frequent long-lived
    allocations (default config values; sample_interval=1;
    num_objects=63), and

(b) normal desktop workloads on an otherwise idle machine where
    the problem was first reported after a few days of uptime
    (default config values).

In both test cases the sampled allocation rate no longer drops to zero
at any point.  In the case of (b) we observe (after 2 days uptime) 15%
unique allocations in the pool, 77% pool utilization, with 20% "skipped
allocations (covered)".

[elver@google.com: simplify and just use hash_32(), use more random stack_hash_seed]
Link: https://lkml.kernel.org/r/YU3MRGaCaJiYht5g@elver.google.com
[elver@google.com: fix 32 bit]

Link: https://lkml.kernel.org/r/20210923104803.2620285-4-elver@google.com
Signed-off-by: Marco Elver <elver@google.com>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Alexander Potapenko <glider@google.com>
Cc: Aleksandr Nogikh <nogikh@google.com>
Cc: Jann Horn <jannh@google.com>
Cc: Taras Madan <tarasmadan@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

kfence: move saving stack trace of allocations into __kfence_alloc()

[ Upstream commit a9ab52bbcb52df49ec4b30e6741e120588989455 ]

Move the saving of the stack trace of allocations into __kfence_alloc(),
so that the stack entries array can be used outside of
kfence_guarded_alloc() and we avoid potentially unwinding the stack
multiple times.

Link: https://lkml.kernel.org/r/20210923104803.2620285-3-elver@google.com
Signed-off-by: Marco Elver <elver@google.com>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Alexander Potapenko <glider@google.com>
Cc: Aleksandr Nogikh <nogikh@google.com>
Cc: Jann Horn <jannh@google.com>
Cc: Taras Madan <tarasmadan@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

kfence: count unexpectedly skipped allocations

[ Upstream commit 9a19aeb5665068c3e2727230588684aae2cab7ef ]

Maintain a counter to count allocations that are skipped due to being
incompatible (oversized, incompatible gfp flags) or no capacity.

This is to compute the fraction of allocations that could not be
serviced by KFENCE, which we expect to be rare.

Link: https://lkml.kernel.org/r/20210923104803.2620285-2-elver@google.com
Signed-off-by: Marco Elver <elver@google.com>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Alexander Potapenko <glider@google.com>
Cc: Aleksandr Nogikh <nogikh@google.com>
Cc: Jann Horn <jannh@google.com>
Cc: Taras Madan <tarasmadan@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

nbd: fix possible overflow on 'first_minor' in nbd_dev_add()

[ Upstream commit 6d35d04a9e18990040e87d2bbf72689252669d54 ]

When 'index' is a big numbers, it may become negative which forced
to 'int'. then 'index << part_shift' might overflow to a positive
value that is not greater than '0xfffff', then sysfs might complains
about duplicate creation. Because of this, move the 'index' judgment
to the front will fix it and be better.

Fixes: b0d9111a2d53 ("nbd: use an idr to keep track of nbd devices")
Fixes: 940c264984fd ("nbd: fix possible overflow for 'first_minor' in nbd_dev_add()")
Signed-off-by: Zhang Wensheng <zhangwensheng5@huawei.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Link: https://lore.kernel.org/r/20220310093224.4002895-1-zhangwensheng5@huawei.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

nbd: Fix hungtask when nbd_config_put

[ Upstream commit e2daec488c57069a4a431d5b752f50294c4bf273 ]

I got follow issue:
[  247.381177] INFO: task kworker/u10:0:47 blocked for more than 120 seconds.
[  247.382644]       Not tainted 4.19.90-dirty #140
[  247.383502] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[  247.385027] Call Trace:
[  247.388384]  schedule+0xb8/0x3c0
[  247.388966]  schedule_timeout+0x2b4/0x380
[  247.392815]  wait_for_completion+0x367/0x510
[  247.397713]  flush_workqueue+0x32b/0x1340
[  247.402700]  drain_workqueue+0xda/0x3c0
[  247.403442]  destroy_workqueue+0x7b/0x690
[  247.405014]  nbd_config_put.cold+0x2f9/0x5b6
[  247.405823]  recv_work+0x1fd/0x2b0
[  247.406485]  process_one_work+0x70b/0x1610
[  247.407262]  worker_thread+0x5a9/0x1060
[  247.408699]  kthread+0x35e/0x430
[  247.410918]  ret_from_fork+0x1f/0x30

We can reproduce issue as follows:
1. Inject memory fault in nbd_start_device
-1244,10 +1248,18 @@ static int nbd_start_device(struct nbd_device *nbd)
        nbd_dev_dbg_init(nbd);
        for (i = 0; i < num_connections; i++) {
                struct recv_thread_args *args;
-
-               args = kzalloc(sizeof(*args), GFP_KERNEL);
+
+               if (i == 1) {
+                       args = NULL;
+                       printk("%s: inject malloc error\n", __func__);
+               }
+               else
+                       args = kzalloc(sizeof(*args), GFP_KERNEL);
2. Inject delay in recv_work
-757,6 +760,8 @@ static void recv_work(struct work_struct *work)

                blk_mq_complete_request(blk_mq_rq_from_pdu(cmd));
        }
+       printk("%s: comm=%s pid=%d\n", __func__, current->comm, current->pid);
+       mdelay(5 * 1000);
        nbd_config_put(nbd);
        atomic_dec(&config->recv_threads);
        wake_up(&config->recv_wq);
3. Create nbd server
nbd-server 8000 /tmp/disk
4. Create nbd client
nbd-client localhost 8000 /dev/nbd1
Then will trigger above issue.

Reason is when add delay in recv_work, lead to release the last reference
of 'nbd->config_refs'. nbd_config_put will call flush_workqueue to make
all work finish. Obviously, it will lead to deadloop.
To solve this issue, according to Josef's suggestion move 'recv_work'
init from start device to nbd_dev_add, then destroy 'recv_work'when
nbd device teardown.

Signed-off-by: Ye Bin <yebin10@huawei.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Link: https://lore.kernel.org/r/20211102015237.2309763-5-yebin10@huawei.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

nbd: Fix incorrect error handle when first_minor is illegal in nbd_dev_add

[ Upstream commit 69beb62ff0d1723a750eebe1c4d01da573d7cd19 ]

If first_minor is illegal will goto out_free_idr label, this will miss
cleanup disk.

Fixes: b1a811633f73 ("block: nbd: add sanity check for first_minor")
Signed-off-by: Ye Bin <yebin10@huawei.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Link: https://lore.kernel.org/r/20211102015237.2309763-4-yebin10@huawei.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

nbd: add error handling support for add_disk()

[ Upstream commit e1654f413fe08ffbc3292d8d2b8958b2cc5cb5e8 ]

We never checked for errors on add_disk() as this function
returned void. Now that this is fixed, use the shiny new
error handling.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>

rtc: wm8350: Handle error for wm8350_register_irq

[ Upstream commit 43f0269b6b89c1eec4ef83c48035608f4dcdd886 ]

As the potential failure of the wm8350_register_irq(),
it should be better to check it and return error if fails.
Also, it need not free 'wm_rtc->rtc' since it will be freed
automatically.

Fixes: 077eaf5b40ec ("rtc: rtc-wm8350: add support for WM8350 RTC")
Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20220303085030.291793-1-jiasheng@iscas.ac.cn
Signed-off-by: Sasha Levin <sashal@kernel.org>

um: fix and optimize xor select template for CONFIG64 and timetravel mode

[ Upstream commit e3a33af812c611d99756e2ec61e9d7068d466bdf ]

Due to dropped inclusion of asm-generic/xor.h, xor_block_8regs symbol is
missing with CONFIG64 and break compilation, as the asm/xor_64.h also did
not include it. The patch recreate the logic from arch/x86, which check
whether AVX is available and add fallbacks for 32bit and 64bit config of
um.

A very minor additional "fix" is, the return of the macro parameter
instead of NULL, as this is the original intent of the macro, but
this does not change the actual behavior.

Fixes: c0ecca6604b8 ("um: enable the use of optimized xor routines in UML")
Signed-off-by: Benjamin Beichler <benjamin.beichler@uni-rostock.de>
Acked-By: Anton Ivanov <anton.ivanov@cambridgegreys.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Sasha Levin <sashal@kernel.org>

lib/logic_iomem: correct fallback config references

[ Upstream commit 2a6852cb8ff0c8c1363cac648d68489343813212 ]

Due to some renaming, we ended up with the "indirect iomem"
naming in Kconfig, following INDIRECT_PIO. However, clearly
I missed following through on that in the ifdefs, but so far
INDIRECT_IOMEM_FALLBACK isn't used by any architecture.

Reported-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Fixes: ca2e334232b6 ("lib: add iomem emulation (logic_iomem)")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Sasha Levin <sashal@kernel.org>

Linux 5.15.33

Link: https://lore.kernel.org/r/20220405070339.801210740@linuxfoundation.org
Tested-by: Florian Fainelli <f.fainelli@gmail.com>
Tested-by: Slade Watkins <slade@sladewatkins.com>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Tested-by: Jon Hunter <jonathanh@nvidia.com>
Tested-by: Sudip Mukherjee <sudip.mukherjee@codethink.co.uk>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Bagas Sanjaya <bagasdotme@gmail.com>
Tested-by: Fox Chen <foxhlchen@gmail.com>
Tested-by: Ron Economos <re@w6rz.net>
Link: https://lore.kernel.org/r/20220406133055.820319940@linuxfoundation.org
Tested-by: Jon Hunter <jonathanh@nvidia.com>
Tested-by: Shuah Khan <skhan@linuxfoundation.org>
Tested-by: Florian Fainelli <f.fainelli@gmail.com>
Tested-by: Fox Chen <foxhlchen@gmail.com>
Tested-by: Bagas Sanjaya <bagasdotme@gmail.com>
Tested-by: Sudip Mukherjee <sudip.mukherjee@codethink.co.uk>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Tested-by: Ron Economos <re@w6rz.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

PCI: xgene: Revert "PCI: xgene: Use inbound resources for setup"

commit 1874b6d7ab1bdc900e8398026350313ac29caddb upstream.

Commit 6dce5aa59e0b ("PCI: xgene: Use inbound resources for setup")
killed PCIe on my XGene-1 box (a Mustang board). The machine itself
is still alive, but half of its storage (over NVMe) is gone, and the
NVMe driver just times out.

Note that this machine boots with a device tree provided by the
UEFI firmware (2016 vintage), which could well be non conformant
with the spec, hence the breakage.

With the patch reverted, the box boots 5.17-rc8 with flying colors.

Link: https://lore.kernel.org/all/Yf2wTLjmcRj+AbDv@xps13.dannf
Link: https://lore.kernel.org/r/20220321104843.949645-2-maz@kernel.org
Fixes: 6dce5aa59e0b ("PCI: xgene: Use inbound resources for setup")
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Cc: stable@vger.kernel.org
Cc: Rob Herring <robh@kernel.org>
Cc: Toan Le <toan@os.amperecomputing.com>
Cc: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Cc: Krzysztof Wilczyński <kw@linux.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Stéphane Graber <stgraber@ubuntu.com>
Cc: dann frazier <dann.frazier@canonical.com>
[dannf: minor context adjustment]
Signed-off-by: dann frazier <dann.frazier@canonical.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

coredump: Use the vma snapshot in fill_files_note

commit 390031c942116d4733310f0684beb8db19885fe6 upstream.

Matthew Wilcox reported that there is a missing mmap_lock in
file_files_note that could possibly lead to a user after free.

Solve this by using the existing vma snapshot for consistency
and to avoid the need to take the mmap_lock anywhere in the
coredump code except for dump_vma_snapshot.

Update the dump_vma_snapshot to capture vm_pgoff and vm_file
that are neeeded by fill_files_note.

Add free_vma_snapshot to free the captured values of vm_file.

Reported-by: Matthew Wilcox <willy@infradead.org>
Link: https://lkml.kernel.org/r/20220131153740.2396974-1-willy@infradead.org
Cc: stable@vger.kernel.org
Fixes: a07279c9a8cd ("binfmt_elf, binfmt_elf_fdpic: use a VMA list snapshot")
Fixes: 2aa362c49c31 ("coredump: extend core dump note section to contain file names of mapped files")
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

coredump/elf: Pass coredump_params into fill_note_info

commit 9ec7d3230717b4fe9b6c7afeb4811909c23fa1d7 upstream.

Instead of individually passing cprm->siginfo and cprm->regs
into fill_note_info pass all of struct coredump_params.

This is preparation to allow fill_files_note to use the existing
vma snapshot.

Reviewed-by: Jann Horn <jannh@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

coredump: Remove the WARN_ON in dump_vma_snapshot

commit 49c1866348f364478a0c4d3dd13fd08bb82d3a5b upstream.

The condition is impossible and to the best of my knowledge has never
triggered.

We are in deep trouble if that conditions happens and we walk past
the end of our allocated array.

So delete the WARN_ON and the code that makes it look like the kernel
can handle the case of walking past the end of it's vma_meta array.

Reviewed-by: Jann Horn <jannh@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

coredump: Snapshot the vmas in do_coredump

commit 95c5436a4883841588dae86fb0b9325f47ba5ad3 upstream.

Move the call of dump_vma_snapshot and kvfree(vma_meta) out of the
individual coredump routines into do_coredump itself.  This makes
the code less error prone and easier to maintain.

Make the vma snapshot available to the coredump routines
in struct coredump_params.  This makes it easier to
change and update what is captures in the vma snapshot
and will be needed for fixing fill_file_notes.

Reviewed-by: Jann Horn <jannh@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

iommu/dma: Account for min_align_mask w/swiotlb

commit 2cbc61a1b1665c84282dbf2b1747ffa0b6248639 upstream.

Pass the non-aligned size to __iommu_dma_map when using swiotlb bounce
buffers in iommu_dma_map_page, to account for min_align_mask.

To deal with granule alignment, __iommu_dma_map maps iova_align(size +
iova_off) bytes starting at phys - iova_off. If iommu_dma_map_page
passes aligned size when using swiotlb, then this becomes
iova_align(iova_align(orig_size) + iova_off). Normally iova_off will be
zero when using swiotlb. However, this is not the case for devices that
set min_align_mask. When iova_off is non-zero, __iommu_dma_map ends up
mapping an extra page at the end of the buffer. Beyond just being a
security issue, the extra page is not cleaned up by __iommu_dma_unmap.
This causes problems when the IOVA is reused, due to collisions in the
iommu driver.  Just passing the original size is sufficient, since
__iommu_dma_map will take care of granule alignment.

Fixes: 1f221a0d0dbf ("swiotlb: respect min_align_mask")
Signed-off-by: David Stevens <stevensd@chromium.org>
Link: https://lore.kernel.org/r/20210929023300.335969-8-stevensd@google.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

swiotlb: Support aligned swiotlb buffers

commit e81e99bacc9f9347bda7808a949c1ce9fcc2bbf4 upstream.

Add an argument to swiotlb_tbl_map_single that specifies the desired
alignment of the allocated buffer. This is used by dma-iommu to ensure
the buffer is aligned to the iova granule size when using swiotlb with
untrusted sub-granule mappings. This addresses an issue where adjacent
slots could be exposed to the untrusted device if IO_TLB_SIZE < iova
granule < PAGE_SIZE.

Signed-off-by: David Stevens <stevensd@chromium.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210929023300.335969-7-stevensd@google.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Cc: Mario Limonciello <Mario.Limonciello@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

iommu/dma: Check CONFIG_SWIOTLB more broadly

commit 2e727bffbe93750a13d2414f3ce43de2f21600d2 upstream.

Introduce a new dev_use_swiotlb function to guard swiotlb code, instead
of overloading dev_is_untrusted. This allows CONFIG_SWIOTLB to be
checked more broadly, so the swiotlb related code can be removed more
aggressively.

Signed-off-by: David Stevens <stevensd@chromium.org>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210929023300.335969-6-stevensd@google.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Cc: Mario Limonciello <Mario.Limonciello@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

iommu/dma: Fold _swiotlb helpers into callers

commit 9b49bbc2c4dfd0431bf7ff4e862171189cf94b7e upstream.

Fold the _swiotlb helper functions into the respective _page functions,
since recent fixes have moved all logic from the _page functions to the
_swiotlb functions.

Signed-off-by: David Stevens <stevensd@chromium.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/20210929023300.335969-5-stevensd@google.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Cc: Mario Limonciello <Mario.Limonciello@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

iommu/dma: Skip extra sync during unmap w/swiotlb

commit ee9d4097cc145dcaebedf6b113d17c91c21333a0 upstream.

Calling the iommu_dma_sync_*_for_cpu functions during unmap can cause
two copies out of the swiotlb buffer. Do the arch sync directly in
__iommu_dma_unmap_swiotlb instead to avoid this. This makes the call to
iommu_dma_sync_sg_for_cpu for untrusted devices in iommu_dma_unmap_sg no
longer necessary, so move that invocation later in the function.

Signed-off-by: David Stevens <stevensd@chromium.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/20210929023300.335969-4-stevensd@google.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Cc: Mario Limonciello <Mario.Limonciello@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

KVM: x86/mmu: do compare-and-exchange of gPTE via the user address

commit 2a8859f373b0a86f0ece8ec8312607eacf12485d upstream.

FNAME(cmpxchg_gpte) is an inefficient mess.  It is at least decent if it
can go through get_user_pages_fast(), but if it cannot then it tries to
use memremap(); that is not just terribly slow, it is also wrong because
it assumes that the VM_PFNMAP VMA is contiguous.

The right way to do it would be to do the same thing as
hva_to_pfn_remapped() does since commit add6a0cd1c5b ("KVM: MMU: try to
fix up page faults before giving up", 2016-07-05), using follow_pte()
and fixup_user_fault() to determine the correct address to use for
memremap().  To do this, one could for example extract hva_to_pfn()
for use outside virt/kvm/kvm_main.c.  But really there is no reason to
do that either, because there is already a perfectly valid address to
do the cmpxchg() on, only it is a userspace address.  That means doing
user_access_begin()/user_access_end() and writing the code in assembly
to handle exceptions correctly.  Worse, the guest PTE can be 8-byte
even on i686 so there is the extra complication of using cmpxchg8b to
account for.  But at least it is an efficient mess.

(Thanks to Linus for suggesting improvement on the inline assembly).

Reported-by: Qiuhao Li <qiuhao@sysec.org>
Reported-by: Gaoning Pan <pgn@zju.edu.cn>
Reported-by: Yongkang Jia <kangel@zju.edu.cn>
Reported-by: syzbot+6cde2282daa792c49ab8@syzkaller.appspotmail.com
Debugged-by: Tadeusz Struk <tadeusz.struk@linaro.org>
Tested-by: Maxim Levitsky <mlevitsk@redhat.com>
Cc: stable@vger.kernel.org
Fixes: bd53cb35a3e9 ("X86/KVM: Handle PFNs outside of kernel reach when touching GPTEs")
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

mmc: rtsx: Fix build errors/warnings for unused variable

commit 3dd9a926ec2308e49445f22abef149fc64e9332e upstream.

The struct device *dev, is no longer needed at various functions, let's
therefore drop it to fix the build errors/warnings.

Fixes: 7570fb41e450 ("mmc: rtsx: Let MMC core handle runtime PM")
Cc: Kai-Heng Feng <kai.heng.feng@canonical.com>
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Link: https://lore.kernel.org/r/20220301115300.64332-1-ulf.hansson@linaro.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

mmc: rtsx: Let MMC core handle runtime PM

commit 7570fb41e450ba37bf9335fe3751fa9f502c30fa upstream.

Since MMC core handles runtime PM reference counting, we can avoid doing
redundant runtime PM work in the driver. That means the only thing
commit 5b4258f6721f ("misc: rtsx: rts5249 support runtime PM") misses is
to always enable runtime PM, to let its parent driver enable ASPM in the
runtime idle routine.

Fixes: 7499b529d97f ("mmc: rtsx: Use pm_runtime_{get,put}() to handle runtime PM")
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Link: https://lore.kernel.org/r/20220216055435.2335297-1-kai.heng.feng@canonical.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

n64cart: convert bi_disk to bi_bdev->bd_disk fix build

commit b2479de38d8fc7ef13d5c78ff5ded6e5a1a4eac0 upstream.

My kernel robot report below:

  drivers/block/n64cart.c: In function ‘n64cart_submit_bio’:
  drivers/block/n64cart.c:91:26: error: ‘struct bio’ has no member named ‘bi_disk’
     91 |  struct device *dev = bio->bi_disk->private_data;
        |                          ^~
    CC      drivers/slimbus/qcom-ctrl.o
    CC      drivers/auxdisplay/hd44780.o
    CC      drivers/watchdog/watchdog_core.o
    CC      drivers/nvme/host/fault_inject.o
    AR      drivers/accessibility/braille/built-in.a
  make[2]: *** [scripts/Makefile.build:288: drivers/block/n64cart.o] Error 1

Fixes: 309dca309fc3 ("block: store a block_device pointer in struct bio");
Reported-by: k2ci <kernel-bot@kylinos.cn>
Signed-off-by: Jackie Liu <liuyun01@kylinos.cn>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Link: https://lore.kernel.org/r/20220321071216.1549596-1-liu.yun@linux.dev
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

torture: Make torture.sh help message match reality

commit f233673cd32a048f2eed69e56b61174c33fb740b upstream.

This commit fixes a couple of typos: s/--doall/--do-all/ and
s/--doallmodconfig/--do-allmodconfig/.

[ paulmck: Add Fixes: supplied by Paul Menzel. ]

Fixes: a115a775a8d5 ("torture: Add "make allmodconfig" to torture.sh")
Reported-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

openvswitch: Fixed nd target mask field in the flow dump.

commit f19c44452b58a84d95e209b847f5495d91c9983a upstream.

IPv6 nd target mask was not getting populated in flow dump.

In the function __ovs_nla_put_key the icmp code mask field was checked
instead of icmp code key field to classify the flow as neighbour discovery.

ufid:bdfbe3e5-60c2-43b0-a5ff-dfcac1c37328, recirc_id(0),dp_hash(0/0),
skb_priority(0/0),in_port(ovs-nm1),skb_mark(0/0),ct_state(0/0),
ct_zone(0/0),ct_mark(0/0),ct_label(0/0),
eth(src=00:00:00:00:00:00/00:00:00:00:00:00,
dst=00:00:00:00:00:00/00:00:00:00:00:00),
eth_type(0x86dd),
ipv6(src=::/::,dst=::/::,label=0/0,proto=58,tclass=0/0,hlimit=0/0,frag=no),
icmpv6(type=135,code=0),
nd(target=2001::2/::,
sll=00:00:00:00:00:00/00:00:00:00:00:00,
tll=00:00:00:00:00:00/00:00:00:00:00:00),
packets:10, bytes:860, used:0.504s, dp:ovs, actions:ovs-nm2

Fixes: e64457191a25 (openvswitch: Restructure datapath.c and flow.c)
Signed-off-by: Martin Varghese <martin.varghese@nokia.com>
Link: https://lore.kernel.org/r/20220328054148.3057-1-martinvarghesenokia@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

docs: sysctl/kernel: add missing bit to panic_print

commit a1ff1de00db21ecb956213f046b79741b64c6b65 upstream.

Patch series "Some improvements on panic_print".

This is a mix of a documentation fix with some additions to the
"panic_print" syscall / parameter.  The goal here is being able to collect
all CPUs backtraces during a panic event and also to enable "panic_print"
in a kdump event - details of the reasoning and design choices in the
patches.

This patch (of 3):

Commit de6da1e8bcf0 ("panic: add an option to replay all the printk
message in buffer") added a new bit to the sysctl/kernel parameter
"panic_print", but the documentation was added only in
kernel-parameters.txt, not in the sysctl guide.

Fix it here by adding bit 5 to sysctl admin-guide documentation.

[rdunlap@infradead.org: fix table format warning]
Link: https://lkml.kernel.org/r/20220109055635.6999-1-rdunlap@infradead.org
Link: https://lkml.kernel.org/r/20211109202848.610874-1-gpiccoli@igalia.com
Link: https://lkml.kernel.org/r/20211109202848.610874-2-gpiccoli@igalia.com
Fixes: de6da1e8bcf0 ("panic: add an option to replay all the printk message in buffer")
Signed-off-by: Guilherme G. Piccoli <gpiccoli@igalia.com>
Reviewed-by: Feng Tang <feng.tang@intel.com>
Cc: Luis Chamberlain <mcgrof@kernel.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Iurii Zaikin <yzaikin@google.com>
Cc: Samuel Iglesias Gonsalvez <siglesias@igalia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

um: Fix uml_mconsole stop/go

commit 1a3a6a2a035bb6c3a7ef4c788d8fd69a7b2d6284 upstream.

Moving to an EPOLL based IRQ controller broke uml_mconsole stop/go
commands. This fixes it and restores stop/go functionality.

Fixes: ff6a17989c08 ("Epoll based IRQ controller")
Signed-off-by: Anton Ivanov <anton.ivanov@cambridgegreys.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

arm64: dts: ls1046a: Update i2c node dma properties

commit eeeb4f1075d71d67083c75f69247206e9b3d9f4a upstream.

Reorder dmas and dma-names properties for i2c controller node to make it
compliant with bindings.

Fixes: 8126d88162a5 ("arm64: dts: add QorIQ LS1046A SoC support")
Signed-off-by: Kuldeep Singh <singh.kuldeep87k@gmail.com>
CC: soc@kernel.org
Link: https://lore.kernel.org/r/20220326042313.97862-5-singh.kuldeep87k@gmail.com'
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>