We've had CONFIG_MANDATORY_FILE_LOCKING since 2015 and a lot of distros
have disabled it. Warn the stragglers that still use "-o mand" that
we'll be dropping support for that mount option.
Cc: stable@vger.kernel.org Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
We currently check for ret != 0 to indicate error, but '1' is a valid
return and just indicates that the allocation succeeded with a wrap.
Correct the check to be for < 0, like it was before the xarray
conversion.
syzbot hit kernel BUG at fs/hugetlbfs/inode.c:532 as described in [1].
This BUG triggers if the HPageRestoreReserve flag is set on a page in
the page cache. It should never be set, as the routine
huge_add_to_page_cache explicitly clears the flag after adding a page to
the cache.
The only code other than huge page allocation which sets the flag is
restore_reserve_on_error. It will potentially set the flag in rare out
of memory conditions. syzbot was injecting errors to cause memory
allocation errors which exercised this specific path.
The code in restore_reserve_on_error is doing the right thing. However,
there are instances where pages in the page cache were being passed to
restore_reserve_on_error. This is incorrect, as once a page goes into
the cache reservation information will not be modified for the page
until it is removed from the cache. Error paths do not remove pages
from the cache, so even in the case of error, the page will remain in
the cache and no reservation adjustment is needed.
Modify routines that potentially call restore_reserve_on_error with a
page cache page to no longer do so.
Note on fixes tag: Prior to commit 846be08578ed ("mm/hugetlb: expand
restore_reserve_on_error functionality") the routine would not process
page cache pages because the HPageRestoreReserve flag is not set on such
pages. Therefore, this issue could not be trigggered. The code added
by commit 846be08578ed ("mm/hugetlb: expand restore_reserve_on_error
functionality") is needed and correct. It exposed incorrect calls to
restore_reserve_on_error which is the root cause addressed by this
commit.
Originally the addr != NULL check was meant to take care of the case
where __kfence_pool == NULL (KFENCE is disabled). However, this does
not work for addresses where addr > 0 && addr < KFENCE_POOL_SIZE.
This can be the case on NULL-deref where addr > 0 && addr < PAGE_SIZE or
any other faulting access with addr < KFENCE_POOL_SIZE. While the
kernel would likely crash, the stack traces and report might be
confusing due to double faults upon KFENCE's attempt to unprotect such
an address.
Fix it by just checking that __kfence_pool != NULL instead.
Link: https://lkml.kernel.org/r/20210818130300.2482437-1-elver@google.com Fixes: 0ce20dd84089 ("mm: add Kernel Electric-Fence infrastructure") Signed-off-by: Marco Elver <elver@google.com> Reported-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com> Acked-by: Alexander Potapenko <glider@google.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: <stable@vger.kernel.org> [5.12+] 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>
HWPoisonHandlable() sometimes returns false for typical user pages due
to races with average memory events like transfers over LRU lists. This
causes failures in hwpoison handling.
There's retry code for such a case but does not work because the retry
loop reaches the retry limit too quickly before the page settles down to
handlable state. Let get_any_page() call shake_page() to fix it.
[naoya.horiguchi@nec.com: get_any_page(): return -EIO when retry limit reached] Link: https://lkml.kernel.org/r/20210819001958.2365157-1-naoya.horiguchi@linux.dev Link: https://lkml.kernel.org/r/20210817053703.2267588-1-naoya.horiguchi@linux.dev Fixes: 25182f05ffed ("mm,hwpoison: fix race with hugetlb page allocation") Signed-off-by: Naoya Horiguchi <naoya.horiguchi@nec.com> Reported-by: Tony Luck <tony.luck@intel.com> Reviewed-by: Yang Shi <shy828301@gmail.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Muchun Song <songmuchun@bytedance.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Michal Hocko <mhocko@suse.com> Cc: <stable@vger.kernel.org> [5.13+] 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>
__get_hwpoison_page() could fail to grab refcount by some race condition,
so it's helpful if we can handle it by retrying. We already have retry
logic, so make get_hwpoison_page() call get_any_page() when called from
memory_failure().
As a result, get_hwpoison_page() can return negative values (i.e. error
code), so some callers are also changed to handle error cases.
soft_offline_page() does nothing for -EBUSY because that's enough and
users in userspace can easily handle it. unpoison_memory() is also
unchanged because it's broken and need thorough fixes (will be done
later).
Link: https://lkml.kernel.org/r/20210603233632.2964832-3-nao.horiguchi@gmail.com Signed-off-by: Naoya Horiguchi <naoya.horiguchi@nec.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Muchun Song <songmuchun@bytedance.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Tony Luck <tony.luck@intel.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>
We've noticed occasional OOM killing when memory.low settings are in
effect for cgroups. This is unexpected and undesirable as memory.low is
supposed to express non-OOMing memory priorities between cgroups.
The reason for this is proportional memory.low reclaim. When cgroups
are below their memory.low threshold, reclaim passes them over in the
first round, and then retries if it couldn't find pages anywhere else.
But when cgroups are slightly above their memory.low setting, page scan
force is scaled down and diminished in proportion to the overage, to the
point where it can cause reclaim to fail as well - only in that case we
currently don't retry, and instead trigger OOM.
To fix this, hook proportional reclaim into the same retry logic we have
in place for when cgroups are skipped entirely. This way if reclaim
fails and some cgroups were scanned with diminished pressure, we'll try
another full-force cycle before giving up and OOMing.
[akpm@linux-foundation.org: coding-style fixes]
Link: https://lkml.kernel.org/r/20210817180506.220056-1-hannes@cmpxchg.org Fixes: 9783aa9917f8 ("mm, memcg: proportional memory.{low,min} reclaim") Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reported-by: Leon Yang <lnyng@fb.com> Reviewed-by: Rik van Riel <riel@surriel.com> Reviewed-by: Shakeel Butt <shakeelb@google.com> Acked-by: Roman Gushchin <guro@fb.com> Acked-by: Chris Down <chris@chrisdown.name> Acked-by: Michal Hocko <mhocko@suse.com> Cc: <stable@vger.kernel.org> [5.4+] 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>
Function init_resources() allocates a boot memory block to hold an array of
resources which it adds to iomem_resource. The array is filled in from its
end and the function then attempts to free any unused memory at the
beginning. The problem is that size of the unused memory is incorrectly
calculated and this can result in releasing memory which is in use by
active resources. Their data then gets corrupted later when the memory is
reused by a different part of the system.
Fix the size of the released memory to correctly match the number of unused
resource entries.
Fixes: ffe0e5261268 ("RISC-V: Improve init_resources()") Signed-off-by: Petr Pavlu <petr.pavlu@suse.com> Reviewed-by: Sunil V L <sunilvl@ventanamicro.com> Acked-by: Nick Kossifidis <mick@ics.forth.gr> Tested-by: Sunil V L <sunilvl@ventanamicro.com> Cc: stable@vger.kernel.org Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The commit 2e6b836312a4 ("ASoC: intel: atom: Fix reference to PCM
buffer address") changed the reference of PCM buffer address to
substream->runtime->dma_addr as the buffer address may change
dynamically. However, I forgot that the dma_addr field is still not
set up for the CONTINUOUS buffer type (that this driver uses) yet in
5.14 and earlier kernels, and it resulted in garbage I/O. The problem
will be fixed in 5.15, but we need to address it quickly for now.
The fix is to deduce the address again from the DMA pointer with
virt_to_phys(), but from the right one, substream->runtime->dma_area.
The Renoir XHCI controller apparently doesn't resume reliably with the
standard D3hot-to-D0 delay. Increase it to 20ms.
[Alex: I talked to the AMD USB hardware team and the AMD Windows team and
they are not aware of any HW errata or specific issues. The HW works fine
in Windows. I was told Windows uses a rather generous default delay of
100ms for PCI state transitions.]
Link: https://lore.kernel.org/r/20210722025858.220064-1-alexander.deucher@amd.com Signed-off-by: Marcin Bachry <hegel666@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Cc: stable@vger.kernel.org Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Prike Liang <prike.liang@amd.com> Cc: Shyam Sundar S K <shyam-sundar.s-k@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit b5efec00b671 ("powerpc/32s: Move KUEP locking/unlocking in C")
removed the 'isync' instruction after adding/removing NX bit in user
segments. The reasoning behind this change was that when setting the
NX bit we don't mind it taking effect with delay as the kernel never
executes text from userspace, and when clearing the NX bit this is
to return to userspace and then the 'rfi' should synchronise the
context.
However, it looks like on book3s/32 having a hash page table, at least
on the G3 processor, we get an unexpected fault from userspace, then
this is followed by something wrong in the verification of MSR_PR
at end of another interrupt.
This is fixed by adding back the removed isync() following update
of NX bit in user segment registers. Only do it for cores with an
hash table, as 603 cores don't exhibit that problem and the two isync
increase ./null_syscall selftest by 6 cycles on an MPC 832x.
First problem: unexpected WARN_ON() for mysterious PROTFAULT
KUEP implements the update of user segment registers.
Move it into mmu-hash.h in order to use it from other places.
And inline kuep_lock() and kuep_unlock(). Inlining kuep_lock() is
important for system_call_exception(), otherwise system_call_exception()
has to save into stack the system call parameters that are used just
after, and doing that takes more instructions than kuep_lock() itself.
Commit a20dcf53ea98 ("usb: typec: tcpm: Respond Not_Supported if no
snk_vdo"), stops tcpm_pd_data_request() calling tcpm_handle_vdm_request()
when port->nr_snk_vdo is not set. But the VDM might be intended for an
altmode-driver, in which case nr_snk_vdo does not matter.
This change breaks the forwarding of connector hotplug (HPD) events
for displayport altmode on devices which don't set nr_snk_vdo.
tcpm_pd_data_request() is the only caller of tcpm_handle_vdm_request(),
so we can move the nr_snk_vdo check to inside it, at which point we
have already looked up the altmode device so we can check for this too.
Doing this check here also ensures that vdm_state gets set to
VDM_STATE_DONE if it was VDM_STATE_BUSY, even if we end up with
responding with PD_MSG_CTRL_NOT_SUPP later.
Note that tcpm_handle_vdm_request() was already sending
PD_MSG_CTRL_NOT_SUPP in some circumstances, after moving the nr_snk_vdo
check the same error-path is now taken when that check fails. So that
we have only one error-path for this and not two. Replace the
tcpm_queue_message(PD_MSG_CTRL_NOT_SUPP) used by the existing error-path
with the more robust tcpm_pd_handle_msg() from the (now removed) second
error-path.
Fixes: a20dcf53ea98 ("usb: typec: tcpm: Respond Not_Supported if no snk_vdo") Cc: stable <stable@vger.kernel.org> Cc: Kyle Tso <kyletso@google.com> Acked-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Acked-by: Kyle Tso <kyletso@google.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com> Link: https://lore.kernel.org/r/20210816154632.381968-1-hdegoede@redhat.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The struct pci_dev uses reference counting but zPCI assumed erroneously
that the last reference would always be the local reference after
calling pci_stop_and_remove_bus_device(). This is usually the case but
not how reference counting works and thus inherently fragile.
In fact one case where this causes a NULL pointer dereference when on an
SRIOV device the function 0 was hot unplugged before another function of
the same multi-function device. In this case the second function's
pdev->sriov->dev reference keeps the struct pci_dev of function 0 alive
even after the unplug. This bug was previously hidden by the fact that
we were leaking the struct pci_dev which in turn means that it always
outlived the struct zpci_dev. This was fixed in commit 0b13525c20fe
("s390/pci: fix leak of PCI device structure") exposing the broken
behavior.
Fix this by accounting for the long living reference a struct pci_dev
has to its underlying struct zpci_dev via the zbus->function[] array and
only release that in pcibios_release_device() ensuring that the struct
pci_dev is not left with a dangling reference. This is a minimal fix in
the future it would probably better to use fine grained reference
counting for struct zpci_dev.
The event filters are not applied on all of the output, which results in
the flood of printk when using tp_printk. Unfolding
event_trigger_unlock_commit_regs() into trace_event_buffer_commit(), so
the filters can be applied on every output.
Link: https://lkml.kernel.org/r/20210814034538.8428-1-kernelfans@gmail.com Cc: stable@vger.kernel.org Fixes: 0daa2302968c1 ("tracing: Add tp_printk cmdline to have tracepoints go to printk()") Signed-off-by: Pingfan Liu <kernelfans@gmail.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Cross-rename lacks a check when that would prevent exchanging a
directory and subvolume from different parent subvolume. This causes
data inconsistencies and is caught before commit by tree-checker,
turning the filesystem to read-only.
Calling the renameat2 with RENAME_EXCHANGE flags like
The corrupted data will not be written, and filesystem can be unmounted
and mounted again (all changes since the last commit will be lost).
Add the missing check for new_ino so that all non-subvolumes must reside
under the same parent subvolume. There's an exception allowing to
exchange two subvolumes from any parents as the directory representing a
subvolume is only a logical link and does not have any other structures
related to the parent subvolume, unlike files, directories etc, that
are always in the inode namespace of the parent subvolume.
Fixes: cdd1fedf8261 ("btrfs: add support for RENAME_EXCHANGE and RENAME_WHITEOUT") CC: stable@vger.kernel.org # 4.7+ Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: NeilBrown <neilb@suse.de> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The controller doesn't seem to pick-up on clock changes, so set the
SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN flag to query the clock frequency
directly from the clock.
Fixes: f84e411c85be ("mmc: sdhci-iproc: Add support for emmc2 of the BCM2711") Signed-off-by: Nicolas Saenz Julienne <nsaenz@kernel.org> Signed-off-by: Stefan Wahren <stefan.wahren@i2se.com> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/1628334401-6577-6-git-send-email-stefan.wahren@i2se.com Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
There is a known bug on BCM2711's SDHCI core integration where the
controller will hang when the difference between the core clock and the
bus clock is too great. Specifically this can be reproduced under the
following conditions:
- No SD card plugged in, polling thread is running, probing cards at
100 kHz.
- BCM2711's core clock configured at 500MHz or more.
So set 200 kHz as the minimum clock frequency available for that board.
For more information on the issue see this:
https://lore.kernel.org/linux-mmc/20210322185816.27582-1-nsaenz@kernel.org/T/#m11f2783a09b581da6b8a15f302625b43a6ecdeca
Fixes: f84e411c85be ("mmc: sdhci-iproc: Add support for emmc2 of the BCM2711") Signed-off-by: Nicolas Saenz Julienne <nsaenz@kernel.org> Signed-off-by: Stefan Wahren <stefan.wahren@i2se.com> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/1628334401-6577-5-git-send-email-stefan.wahren@i2se.com Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The current_opp is released only when whole OPP table is released,
otherwise it's only marked as removed by dev_pm_opp_remove_table().
Functions like dev_pm_opp_put_clkname() and dev_pm_opp_put_supported_hw()
are checking whether OPP table is empty and it's not if current_opp is
set since it holds the refcount of OPP, this produces a noisy warning
from these functions about busy OPP table. Remove the checks to fix it.
The 2021-model XPS 15 appears to use the same 4-speakers-on-ALC289 audio
setup as the Precision models, so requires the same quirk to enable woofer
output. Tested on my own 9510.
If an SQPOLL based ring is newly created and an application issues an
io_uring_enter(2) system call on it, then we can return a spurious
-EOWNERDEAD error. This happens because there's nothing to submit, and
if the caller doesn't specify any other action, the initial error
assignment of -EOWNERDEAD never gets overwritten. This causes us to
return it directly, even if it isn't valid.
Move the error assignment into the actual failure case instead.
The error handling code in tpci200_register does not free interface_regs
allocated by ioremap and the current version of error handling code is
problematic.
Fix this by refactoring the error handling code and free interface_regs
when necessary.
Fixes: 43986798fd50 ("ipack: add error handling for ioremap_nocache") Cc: stable@vger.kernel.org Reported-by: Dongliang Mu <mudongliangabcd@gmail.com> Signed-off-by: Dongliang Mu <mudongliangabcd@gmail.com> Link: https://lore.kernel.org/r/20210810100323.3938492-2-mudongliangabcd@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The function tpci200_register called by tpci200_install and
tpci200_unregister called by tpci200_uninstall are in pair. However,
tpci200_unregister has some cleanup operations not in the
tpci200_register. So the error handling code of tpci200_pci_probe has
many different double free issues.
Fix this problem by moving those cleanup operations out of
tpci200_unregister, into tpci200_pci_remove and reverting
the previous commit 9272e5d0028d ("ipack/carriers/tpci200:
Fix a double free in tpci200_pci_probe").
Fixes: 9272e5d0028d ("ipack/carriers/tpci200: Fix a double free in tpci200_pci_probe") Cc: stable@vger.kernel.org Reported-by: Dongliang Mu <mudongliangabcd@gmail.com> Signed-off-by: Dongliang Mu <mudongliangabcd@gmail.com> Link: https://lore.kernel.org/r/20210810100323.3938492-1-mudongliangabcd@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
During suspend/resume NGD remote instance is power cycled along
with remotely controlled bam dma engine.
So Reset the dma configuration during this suspend resume path
so that we are not dealing with any stale dma setup.
Without this transactions timeout after first suspend resume path.
In some usecases transaction ids are dynamically allocated inside
the controller driver after sending the messages which have generic
acknowledge responses. So check for this before refcounting pm_runtime.
Without this we would end up imbalancing runtime pm count by
doing pm_runtime_put() in both slim_do_transfer() and slim_msg_response()
for a single pm_runtime_get() in slim_do_transfer()
As tid is unsigned its hard to figure out if the tid is valid or
invalid. So Start the transaction ids from 1 instead of zero
so that we could differentiate between a valid tid and invalid tids
This is useful in cases where controller would add a tid for controller
specific transfers.
The problem was the "trace(read_max,count)" where the "count" should be
"$count" as "onmax()" only handles variables (although it really should be
able to figure out that "count" is a field of sys_enter_read). But there's
a path that does not find the variable and ends up passing a NULL for the
event, which ends up getting passed to "strcmp()".
Add a check for NULL to return and error on the command with:
Commit 2860cd8a2353 ("livepatch: Use the default ftrace_ops instead of
REGS when ARGS is available") intends to enable config LIVEPATCH when
ftrace with ARGS is available. However, the chain of configs to enable
LIVEPATCH is incomplete, as HAVE_DYNAMIC_FTRACE_WITH_ARGS is available,
but the definition of DYNAMIC_FTRACE_WITH_ARGS, combining DYNAMIC_FTRACE
and HAVE_DYNAMIC_FTRACE_WITH_ARGS, needed to enable LIVEPATCH, is missing
in the commit.
Fortunately, ./scripts/checkkconfigsymbols.py detects this and warns:
So, define the config DYNAMIC_FTRACE_WITH_ARGS analogously to the already
existing similar configs, DYNAMIC_FTRACE_WITH_REGS and
DYNAMIC_FTRACE_WITH_DIRECT_CALLS, in ./kernel/trace/Kconfig to connect the
chain of configs.
The original code in the cap_put_caller() function does not
handle correctly the positive values returned from the passed
function for multiple iterations. It means that the change
notifications may be lost.
If rcu_read_lock_sched tracing is enabled, the tracing subsystem can
perform a jump which needs to be checked by CFI. For example, stm_ftrace
source is enabled as a module and hooks into enabled ftrace events. This
can cause an recursive loop where find_shadow_check_fn ->
rcu_read_lock_sched -> (call to stm_ftrace generates cfi slowpath) ->
find_shadow_check_fn -> rcu_read_lock_sched -> ...
To avoid the recursion, either the ftrace codes needs to be marked with
__no_cfi or CFI should not trace. Use the "_notrace" in CFI to avoid
tracing so that CFI can guard ftrace.
Signed-off-by: Elliot Berman <quic_eberman@quicinc.com> Reviewed-by: Sami Tolvanen <samitolvanen@google.com> Cc: stable@vger.kernel.org Fixes: cf68fffb66d6 ("add support for Clang CFI") Signed-off-by: Kees Cook <keescook@chromium.org> Link: https://lore.kernel.org/r/20210811155914.19550-1-quic_eberman@quicinc.com Signed-off-by: Sasha Levin <sashal@kernel.org>
commit a5b8ca97fbf8 ("arm64: do not descend to vdso directories twice")
changes the cleaning behavior of arm64's vdso files, in that vdso.lds,
vdso.so, and vdso.so.dbg are not removed upon a 'make clean/mrproper':
$ make defconfig ARCH=arm64
$ make ARCH=arm64
$ make mrproper ARCH=arm64
$ git clean -nxdf
Would remove arch/arm64/kernel/vdso/vdso.lds
Would remove arch/arm64/kernel/vdso/vdso.so
Would remove arch/arm64/kernel/vdso/vdso.so.dbg
To remedy this, manually descend into arch/arm64/kernel/vdso upon
cleaning.
After this commit:
$ make defconfig ARCH=arm64
$ make ARCH=arm64
$ make mrproper ARCH=arm64
$ git clean -nxdf
<empty>
Similar results are obtained for the vdso32 equivalent.
Signed-off-by: Andrew Delgadillo <adelg@google.com> Cc: stable@vger.kernel.org Fixes: a5b8ca97fbf8 ("arm64: do not descend to vdso directories twice") Link: https://lore.kernel.org/r/20210810231755.1743524-1-adelg@google.com Signed-off-by: Will Deacon <will@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
As GDSCs are registered and found to be already enabled gdsc_init()
ensures that 1) the kernel state matches the hardware state, and 2)
votable GDSCs are properly enabled from this master as well.
But as the (optional) supply regulator is enabled deep into
gdsc_toggle_logic(), which is only executed for votable GDSCs, the
kernel's state of the regulator might not match the hardware. The
regulator might be automatically turned off if no other users are
present or the next call to gdsc_disable() would cause an unbalanced
regulator_disable().
Given that the votable case deals with an already enabled GDSC, most of
gdsc_enable() and gdsc_toggle_logic() can be skipped. Reduce it to just
clearing the SW_COLLAPSE_MASK and enabling hardware control to simply
call regulator_enable() in both cases.
The enablement of hardware control seems to be an independent property
from the GDSC being enabled, so this is moved outside that conditional
segment.
Lastly, as the propagation of ALWAYS_ON to GENPD_FLAG_ALWAYS_ON needs to
happen regardless of the initial state this is grouped together with the
other sc->pd updates at the end of the function.
The earlycon depends on the bootloader setup UART clocks being retained.
There're actually two uart clocks (ipg, per) on MX6QDL,
but the 'Fixes' commit change to register only one which means
another clock may be disabled during booting phase
and result in the earlycon unwork.
Cc: stable@vger.kernel.org # v5.10+ Fixes: 379c9a24cc23 ("clk: imx: Fix reparenting of UARTs not associated with stdout") Signed-off-by: Dong Aisheng <aisheng.dong@nxp.com> Link: https://lore.kernel.org/r/20210702085438.1988087-1-aisheng.dong@nxp.com Reviewed-by: Abel Vesa <abel.vesa@nxp.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Whenever SDHC run at clock rate 50MHZ or below, the hardware data
timeout value will be 21.47secs, which is approx. 22secs and we have
a current software timeout value as 10secs. We have to set software
timeout value more than the hardware data timeout value to avioid seeing
the below register dumps.
If the card has not been power cycled, it may still be using 1.8V
signaling. This situation is detected in mmc_sd_init_card function and
should be handled in mmci stm32 variant. The host->pwr_reg variable is
also correctly protected with spin locks.
When a Data CRC interrupt is received, the driver disables the DMA, then
sends the stop/abort command and then waits for Data Transfer Over.
However, sometimes, when a data CRC error is received in the middle of a
multi-block write transfer, the Data Transfer Over interrupt is never
received, and the driver hangs and never completes the request.
The driver sets the BMOD.SWR bit (SDMMC_IDMAC_SWRESET) when stopping the
DMA, but according to the manual CMD.STOP_ABORT_CMD should be programmed
"before assertion of SWR". Do these operations in the recommended
order. With this change the Data Transfer Over is always received
correctly in my tests.
Currently, iommu_dma_alloc_noncontiguous() allocates a
struct dma_sgt_handle object to hold some state needed for
iommu_dma_free_noncontiguous().
However, the handle is neither freed nor returned explicitly by
the ->alloc_noncontiguous method, and therefore seems leaked.
This was found by code inspection, so please review carefully and test.
As a side note, it appears the struct dma_sgt_handle type is exposed
to users of the DMA-API by linux/dma-map-ops.h, but is has no users
or functions returning the type explicitly.
This may indicate it's a good idea to move the struct dma_sgt_handle type
to drivers/iommu/dma-iommu.c. The decision is left to maintainers :-)
Currently dpaa2_switch_takedown has a funny name and does not do the
opposite of dpaa2_switch_init, which makes probing fail when we need to
handle an -EPROBE_DEFER.
A sketch of what dpaa2_switch_init does:
dpsw_open
dpaa2_switch_detect_features
dpsw_reset
for (i = 0; i < ethsw->sw_attr.num_ifs; i++) {
dpsw_if_disable
dpsw_if_set_stp
dpsw_vlan_remove_if_untagged
dpsw_if_set_tci
dpsw_vlan_remove_if
}
dpsw_vlan_remove
alloc_ordered_workqueue
dpsw_fdb_remove
dpaa2_switch_ctrl_if_setup
When dpaa2_switch_takedown is called from the error path of
dpaa2_switch_probe(), the control interface, enabled by
dpaa2_switch_ctrl_if_setup from dpaa2_switch_init, remains enabled,
because dpaa2_switch_takedown does not call
dpaa2_switch_ctrl_if_teardown.
Since dpaa2_switch_probe might fail due to EPROBE_DEFER of a PHY, this
means that a second probe of the driver will happen with the control
interface directly enabled.
Which if we investigate the /dev/dpaa2_mc_console log, we find out is
caused by:
[E, ctrl_if_set_pools:2211, DPMNG] ctrl_if must be disabled
So make dpaa2_switch_takedown do the opposite of dpaa2_switch_init (in
reasonable limits, no reason to change STP state, re-add VLANs etc), and
rename it to something more conventional, like dpaa2_switch_teardown.
Fixes: 613c0a5810b7 ("staging: dpaa2-switch: enable the control interface") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by: Ioana Ciornei <ioana.ciornei@nxp.com> Link: https://lore.kernel.org/r/20210819141755.1931423-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Cited commit added a check to make sure 'action' is not NULL, but
'action' is already dereferenced before the check, when calling
flow_offload_has_one_action().
Therefore, the check does not make any sense and results in a smatch
warning:
include/net/flow_offload.h:322 flow_action_mixed_hw_stats_check() warn:
variable dereferenced before check 'action' (see line 319)
Fix by reverting this commit.
Cc: gushengxian <gushengxian@yulong.com> Fixes: 9ea3e52c5bc8 ("flow_offload: action should not be NULL when it is referenced") Signed-off-by: Ido Schimmel <idosch@nvidia.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Link: https://lore.kernel.org/r/20210819105842.1315705-1-idosch@idosch.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Make changes to MAC address dependent on the response of PF.
Disallow changes to HW MAC address and MAC filter from untrusted
VF, thanks to that ping is not lost if VF tries to change MAC.
Add a new field in iavf_mac_filter, to indicate whether there
was response from PF for given filter. Based on this field pass
or discard the filter.
If untrusted VF tried to change it's address, it's not changed.
Still filter was changed, because of that ping couldn't go through.
Fixes: c5c922b3e09b ("iavf: fix MAC address setting for VFs when filter is rejected") Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com> Signed-off-by: Sylwester Dziedziuch <sylwesterx.dziedziuch@intel.com> Signed-off-by: Mateusz Palczewski <mateusz.palczewski@intel.com> Tested-by: Gurucharan G <Gurucharanx.g@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Without this patch, ATR does not work. Receive/transmit uses queue
selection based on SW DCB hashing method.
If traffic classes are not configured for PF, then use
netdev_pick_tx function for selecting queue for packet transmission.
Instead of calling i40e_swdcb_skb_tx_hash, call netdev_pick_tx,
which ensures that packet is transmitted/received from CPU that is
running the application.
Reproduction steps:
1. Load i40e driver
2. Map each MSI interrupt of i40e port for each CPU
3. Disable ntuple, enable ATR i.e.:
ethtool -K $interface ntuple off
ethtool --set-priv-flags $interface flow-director-atr
4. Run application that is generating traffic and is bound to a
single CPU, i.e.:
taskset -c 9 netperf -H 1.1.1.1 -t TCP_RR -l 10
5. Observe behavior:
Application's traffic should be restricted to the CPU provided in
taskset.
Fixes: 89ec1f0886c1 ("i40e: Fix queue-to-TC mapping on Tx") Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com> Signed-off-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com> Tested-by: Dave Switzer <david.switzer@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Two legacy PCI sysfs objects "legacy_io" and "legacy_mem" were updated
to use an unified address space in the commit 636b21b50152 ("PCI: Revoke
mappings like devmem"). This allows for revocations to be managed from
a single place when drivers want to take over and mmap() a /dev/mem
range.
Following the update, both of the sysfs objects should leverage the
iomem_get_mapping() function to get an appropriate address range, but
only the "legacy_io" has been correctly updated - the second attribute
seems to be using a wrong variable to pass the iomem_get_mapping()
function to.
Thus, correct the variable name used so that the "legacy_mem" sysfs
object would also correctly call the iomem_get_mapping() function.
Fixes: 636b21b50152 ("PCI: Revoke mappings like devmem") Link: https://lore.kernel.org/r/20210812132144.791268-1-kw@linux.com Signed-off-by: Krzysztof Wilczyński <kw@linux.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Sasha Levin <sashal@kernel.org>
The maximum PLA bp number of RTL8153C is 16, not 8. That is, the
bp 0 ~ 15 are at 0xfc28 ~ 0xfc46, and the bp_en is at 0xfc48.
Fixes: 195aae321c82 ("r8152: support new chips") Signed-off-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The register of USB_BP2_EN is 16 bits, so we should use
ocp_write_word(), not ocp_write_byte().
Fixes: 9370f2d05a2a ("support request_firmware for RTL8153") Signed-off-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
If directly after an MP_CAPABLE 3WHS, the client receives an ADD_ADDR
with HMAC from the server, it is enough to switch to a "fully
established" mode because it has received more MPTCP options.
It was then OK to enable the "fully_established" flag on the MPTCP
socket. Still, best to check if the ADD_ADDR looks valid by looking if
it contains an HMAC (no 'echo' bit). If an ADD_ADDR echo is received
while we are not in "fully established" mode, it is strange and then
we should not switch to this mode now.
But that is not enough. On one hand, the path-manager has be notified
the state has changed. On the other hand, the "fully_established" flag
on the subflow socket should be turned on as well not to re-send the
MP_CAPABLE 3rd ACK content with the next ACK.
Fixes: 84dfe3677a6f ("mptcp: send out dedicated ADD_ADDR packet") Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
We should not require an allocation to cleanup stuff.
Rework the code a bit so that the additional RCU work is no more needed.
Fixes: 1729cf186d8a ("mptcp: create the listening socket for new port") Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently we are unable to ping a bridge on top of a felix switch which
uses the ocelot-8021q tagger. The packets are dropped on the ingress of
the user port and the 'drop_local' counter increments (the counter which
denotes drops due to no valid destinations).
Dumping the PGID tables, it becomes clear that the PGID_SRC of the user
port is zero, so it has no valid destinations.
But looking at the code, the cpu_fwd_mask (the bit mask of DSA tag_8021q
ports) is clearly missing from the forwarding mask of ports that are
under a bridge. So this has always been broken.
Looking at the version history of the patch, in v7
https://patchwork.kernel.org/project/netdevbpf/patch/20210125220333.1004365-12-olteanv@gmail.com/
the code looked like this:
/* Standalone ports forward only to DSA tag_8021q CPU ports */
unsigned long mask = cpu_fwd_mask;
while in v8 (the merged version)
https://patchwork.kernel.org/project/netdevbpf/patch/20210129010009.3959398-12-olteanv@gmail.com/
it looked like this:
[why]
dcn301_calculate_wm_and_dl() causes flickering when external monitor is
connected.
This issue has been fixed before by commit 0e4c0ae59d7e
("drm/amdgpu/display: drop dcn301_calculate_wm_and_dl for now"), however
part of the fix was gone after commit 2cbcb78c9ee5 ("Merge tag 'amd-drm-next-5.13-2021-03-23' of https://gitlab.freedesktop.org/agd5f/linux into drm-next").
[how]
Use dcn30_calculate_wm_and_dlg() instead as in the original fix.
Fixes: 2cbcb78c9ee5 ("Merge tag 'amd-drm-next-5.13-2021-03-23' of https://gitlab.freedesktop.org/agd5f/linux into drm-next") Signed-off-by: Nikola Cornij <nikola.cornij@amd.com> Reviewed-by: Zhan Liu <zhan.liu@amd.com> Tested-by: Zhan Liu <zhan.liu@amd.com> Tested-by: Oliver Logush <oliver.logush@amd.com> Signed-off-by: Zhan Liu <zhan.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
dispcnlunit1_cp_xosc_clkreq clock observed to be active on TGL-H platform
despite Wa_14010685332 original sequence,
thus blocks entry to deeper s0ix state.
The Tweaked Wa_14010685332 sequence fixes this issue, therefore use tweaked
Wa_14010685332 sequence for every PCH since PCH_CNP.
v2:
- removed RKL from comment and simplified condition. [Rodrigo]
Fixes: b896898c7369 ("drm/i915: Tweaked Wa_14010685332 for PCHs used on gen11 platforms") Cc: Matt Roper <matthew.d.roper@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210810113112.31739-2-anshuman.gupta@intel.com
(cherry picked from commit 8b46cc6577f4bbef7e5909bb926da31d705f350f) Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Return ealier in the functions doing interruption setup for GEN8+ also
adding a warning in gen8_de_irq_handler() to let us know that
something else is still missing.
Reviewed-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210408203150.237947-1-jose.souza@intel.com Signed-off-by: Sasha Levin <sashal@kernel.org>
This fixes improper iotlb invalidation in intel_pasid_tear_down_entry().
When a PASID was used as nested mode, released and reused, the following
error message will appear:
[ 180.187556] Unexpected page request in Privilege Mode
[ 180.187565] Unexpected page request in Privilege Mode
[ 180.279933] Unexpected page request in Privilege Mode
[ 180.279937] Unexpected page request in Privilege Mode
Per chapter 6.5.3.3 of VT-d spec 3.3, when tear down a pasid entry, the
software should use Domain selective IOTLB flush if the PGTT of the pasid
entry is SL only or Nested, while for the pasid entries whose PGTT is FL
only or PT using PASID-based IOTLB flush is enough.
Refactor asix_read_phy_addr() to return usable error value directly and
make sure all callers handle this error.
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
fq qdisc requires tstamp to be cleared in the forwarding path. Now ovs
doesn't clear skb->tstamp. We encountered a problem with linux
version 5.4.56 and ovs version 2.14.1, and packets failed to
dequeue from qdisc when fq qdisc was attached to ovs port.
Fixes: fb420d5d91c1 ("tcp/fq: move back to CLOCK_MONOTONIC") Signed-off-by: kaixi.fan <fankaixi.li@bytedance.com> Signed-off-by: xiexiaohui <xiexiaohui.xxh@bytedance.com> Reviewed-by: Cong Wang <cong.wang@bytedance.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
When registering mdiobus children, if we get an -EPROBE_DEFER, we shouldn't
ignore it and continue registering the rest of the mdiobus children. This
would permanently prevent the deferring child mdiobus from working instead
of reattempting it in the future. So, if a child mdiobus needs to be
reattempted in the future, defer the entire mdio-mux initialization.
This fixes the issue where PHYs sitting under the mdio-mux aren't
initialized correctly if the PHY's interrupt controller is not yet ready
when the mdio-mux is being probed. Additional context in the link below.
Fixes: 0ca2997d1452 ("netdev/of/phy: Add MDIO bus multiplexer support.") Link: https://lore.kernel.org/lkml/CAGETcx95kHrv8wA-O+-JtfH7H9biJEGJtijuPVN0V5dUKUAB3A@mail.gmail.com/#t Signed-off-by: Saravana Kannan <saravanak@google.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Acked-by: Marc Zyngier <maz@kernel.org> Tested-by: Marc Zyngier <maz@kernel.org> Acked-by: Kevin Hilman <khilman@baylibre.com> Tested-by: Kevin Hilman <khilman@baylibre.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
If we are seeing memory allocation errors, don't try to continue
registering child mdiobus devices. It's unlikely they'll succeed.
Fixes: 342fa1964439 ("mdio: mux: make child bus walking more permissive and errors more verbose") Signed-off-by: Saravana Kannan <saravanak@google.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Acked-by: Marc Zyngier <maz@kernel.org> Tested-by: Marc Zyngier <maz@kernel.org> Acked-by: Kevin Hilman <khilman@baylibre.com> Tested-by: Kevin Hilman <khilman@baylibre.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
When adding support for using the skb->hash value as the flow hash in CAKE,
I accidentally introduced a logic error that broke the host-only isolation
modes of CAKE (srchost and dsthost keywords). Specifically, the flow_hash
variable should stay initialised to 0 in cake_hash() in pure host-based
hashing mode. Add a check for this before using the skb->hash value as
flow_hash.
Fixes: b0c19ed6088a ("sch_cake: Take advantage of skb->hash where appropriate") Reported-by: Pete Heist <pete@heistp.net> Tested-by: Pete Heist <pete@heistp.net> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
In ixgbe_xsk_pool_enable(), if ixgbe_xsk_wakeup() fails,
We should restore the previous state and clean up the
resources. Add the missing clear af_xdp_zc_qps and unmap dma
to fix this bug.
Fixes: d49e286d354e ("ixgbe: add tracking of AF_XDP zero-copy state for each queue pair") Fixes: 4a9b32f30f80 ("ixgbe: fix potential RX buffer starvation for AF_XDP") Signed-off-by: Wang Hai <wanghai38@huawei.com> Acked-by: Magnus Karlsson <magnus.karlsson@intel.com> Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Link: https://lore.kernel.org/r/20210817203736.3529939-1-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
qlcnic_83xx_unlock_flash() is called on all paths after we call
qlcnic_83xx_lock_flash(), except for one error path on failure
of QLCRD32(), which may cause a deadlock. This bug is suggested
by a static analysis tool, please advise.
Fixes: 81d0aeb0a4fff ("qlcnic: flash template based firmware reset recovery") Signed-off-by: Dinghao Liu <dinghao.liu@zju.edu.cn> Link: https://lore.kernel.org/r/20210816131405.24024-1-dinghao.liu@zju.edu.cn Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit a02e8964eaf92 ("virtio-net: ethtool configurable LRO")
maps LRO to virtio guest offloading features and allows the
administrator to enable and disable those features via ethtool.
This leads to several issues:
- For a device that doesn't support control guest offloads, the "LRO"
can't be disabled triggering WARN in dev_disable_lro() when turning
off LRO or when enabling forwarding bridging etc.
- For a device that supports control guest offloads, the guest
offloads are disabled in cases of bridging, forwarding etc slowing
down the traffic.
Fix this by using NETIF_F_GRO_HW instead. Though the spec does not
guarantee packets to be re-segmented as the original ones,
we can add that to the spec, possibly with a flag for devices to
differentiate between GRO and LRO.
Further, we never advertised LRO historically before a02e8964eaf92
("virtio-net: ethtool configurable LRO") and so bridged/forwarded
configs effectively always relied on virtio receive offloads behaving
like GRO - thus even if this breaks any configs it is at least not
a regression.
Fixes: a02e8964eaf92 ("virtio-net: ethtool configurable LRO") Acked-by: Michael S. Tsirkin <mst@redhat.com> Reported-by: Ivan <ivan@prestigetransportation.com> Tested-by: Ivan <ivan@prestigetransportation.com> Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
When a packet is sent over a VRF, the POST_ROUTING hooks are called
twice: Once from the VRF interface, and once from the "actual"
interface the packet will be sent from:
1) First SNAT: l3mdev_l3_out() -> vrf_l3_out() -> .. -> vrf_output_direct()
This causes the POST_ROUTING hooks to run.
2) Second SNAT: 'ip_output()' calls POST_ROUTING hooks again.
Similarly for replies, first ip_rcv() calls PRE_ROUTING hooks, and
second vrf_l3_rcv() calls them again.
As an example, consider the following SNAT rule:
> iptables -t nat -A POSTROUTING -p udp -m udp --dport 53 -j SNAT --to-source 2.2.2.2 -o vrf_1
In this case sending over a VRF will create 2 conntrack entries.
The first is from the VRF interface, which performs the IP SNAT.
The second will run the SNAT, but since the "expected reply" will remain
the same, conntrack randomizes the source port of the packet:
e..g With a socket bound to 1.1.1.1:10000, sending to 3.3.3.3:53, the conntrack
rules are:
udp 17 29 src=2.2.2.2 dst=3.3.3.3 sport=10000 dport=53 packets=1 bytes=68 [UNREPLIED] src=3.3.3.3 dst=2.2.2.2 sport=53 dport=61033 packets=0 bytes=0 mark=0 use=1
udp 17 29 src=1.1.1.1 dst=3.3.3.3 sport=10000 dport=53 packets=1 bytes=68 [UNREPLIED] src=3.3.3.3 dst=2.2.2.2 sport=53 dport=10000 packets=0 bytes=0 mark=0 use=1
i.e. First SNAT IP from 1.1.1.1 --> 2.2.2.2, and second the src port is
SNAT-ed from 10000 --> 61033.
But when a reply is sent (3.3.3.3:53 -> 2.2.2.2:61033) only the later
conntrack entry is matched:
udp 17 29 src=2.2.2.2 dst=3.3.3.3 sport=10000 dport=53 packets=1 bytes=68 src=3.3.3.3 dst=2.2.2.2 sport=53 dport=61033 packets=1 bytes=49 mark=0 use=1
udp 17 28 src=1.1.1.1 dst=3.3.3.3 sport=10000 dport=53 packets=1 bytes=68 [UNREPLIED] src=3.3.3.3 dst=2.2.2.2 sport=53 dport=10000 packets=0 bytes=0 mark=0 use=1
And a "port 61033 unreachable" ICMP packet is sent back.
The issue is that when PRE_ROUTING hooks are called from vrf_l3_rcv(),
the skb already has a conntrack flow attached to it, which means
nf_conntrack_in() will not resolve the flow again.
This means only the dest port is "reverse-NATed" (61033 -> 10000) but
the dest IP remains 2.2.2.2, and since the socket is bound to 1.1.1.1 it's
not received.
This can be verified by logging the 4-tuple of the packet in '__udp4_lib_rcv()'.
The fix is then to reset the flow when skb is received on a VRF, to let
conntrack resolve the flow again (which now will hit the earlier flow).
To reproduce: (Without the fix "Got pkt_to_nat_port" will not be printed by
running 'bash ./repro'):
$ cat run_in_A1.py
import logging
logging.getLogger("scapy.runtime").setLevel(logging.ERROR)
from scapy.all import *
import argparse
while True:
try:
# Periodically send in order to keep the conntrack entry alive.
s.send(b'a'*40)
resp = s.recvfrom(1024)
msg_name = resp[0].decode('utf-8').split('\0')[0]
print(f"Got {msg_name}", flush=True)
except Exception as e:
pass
$ cat repro.sh
ip netns del A1 2> /dev/null
ip netns del A2 2> /dev/null
ip netns add A1
ip netns add A2
ip -n A1 link add _v0 type veth peer name _v1 netns A2
ip -n A1 link set _v0 up
ip -n A2 link add e00000 type bond
ip -n A2 link add lo0 type dummy
ip -n A2 link add vrf_1 type vrf table 10001
ip -n A2 link set vrf_1 up
ip -n A2 link set e00000 master vrf_1
ip -n A2 addr add 1.1.1.1/24 dev e00000
ip -n A2 link set e00000 up
ip -n A2 link set _v1 master e00000
ip -n A2 link set _v1 up
ip -n A2 link set lo0 up
ip -n A2 addr add 2.2.2.2/32 dev lo0
ip -n A2 neigh add 1.1.1.10 lladdr 77:77:77:77:77:77 dev e00000
ip -n A2 route add 3.3.3.3/32 via 1.1.1.10 dev e00000 table 10001
Each completion ring entry has a valid bit to indicate that the entry
contains a valid completion event. The driver's main poll loop
__bnxt_poll_work() has the proper dma_rmb() to make sure the valid
bit of the next entry has been checked before proceeding further.
But when we call bnxt_rx_pkt() to process the RX event, the RX
completion event consists of two completion entries and only the
first entry has been checked to be valid. We need the same barrier
after checking the next completion entry. Add missing dma_rmb()
barriers in bnxt_rx_pkt() and other similar locations.
Fixes: 67a95e2022c7 ("bnxt_en: Need memory barrier when processing the completion ring.") Reported-by: Lance Richardson <lance.richardson@broadcom.com> Reviewed-by: Andy Gospodarek <gospo@broadcom.com> Reviewed-by: Lance Richardson <lance.richardson@broadcom.com> Signed-off-by: Michael Chan <michael.chan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
212 firmware broke aRFS, so disable it. Traffic may stop after ntuple
filters are inserted and deleted by the 212 firmware.
Fixes: ae10ae740ad2 ("bnxt_en: Add new hardware RFS mode.") Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com> Signed-off-by: Michael Chan <michael.chan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
During the swap dependency on PCH_GBE to selection PTP_1588_CLOCK_PCH
incidentally dropped the implicit dependency on the PCI. Restore it.
Fixes: 18d359ceb044 ("pch_gbe, ptp_pch: Fix the dependency direction between these drivers") Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Syzbot reported slab-out-of bounds write in decode_data().
The problem was in missing validation checks.
Syzbot's reproducer generated malicious input, which caused
decode_data() to be called a lot in sixpack_decode(). Since
rx_count_cooked is only 400 bytes and noone reported before,
that 400 bytes is not enough, let's just check if input is malicious
and complain about buffer overrun.
Fail log:
==================================================================
BUG: KASAN: slab-out-of-bounds in drivers/net/hamradio/6pack.c:843
Write of size 1 at addr ffff888087c5544e by task kworker/u4:0/7
Reported-and-tested-by: syzbot+fc8cd9a673d4577fb2e4@syzkaller.appspotmail.com Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Pavel Skripkin <paskripkin@gmail.com> Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The patch be7ecbd240b2: "soc: fsl: qe: convert QE interrupt
controller to platform_device" from Aug 3, 2021, leads to the
following static checker warning:
drivers/soc/fsl/qe/qe_ic.c:438 qe_ic_init()
warn: unsigned 'qe_ic->virq_low' is never less than zero.
In old variant irq_of_parse_and_map() returns zero if failed so
unsigned int for virq_high/virq_low was ok.
In new variant platform_get_irq() returns negative error codes
if failed so we need to use int for virq_high/virq_low.
Also simplify high_handler checking and remove the curly braces
to make checkpatch happy.
Fixes: be7ecbd240b2 ("soc: fsl: qe: convert QE interrupt controller to platform_device") Signed-off-by: Maxim Kochetkov <fido_max@inbox.ru> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Li Yang <leoyang.li@nxp.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
skbs are freed on error and not put on the ring. We may, however,
be in a situation where we're freeing the last skb of a batch,
and there is a doorbell ring pending because of xmit_more() being
true earlier. Make sure we ring the door bell in such situations.
Since errors are rare don't pay attention to xmit_more() and just
always flush the pending frames.
The busy case should be safe to be left alone because it can
only happen if start_xmit races with completions and they
both enable the queue. In that case the kick can't be pending.
napi schedules DIM, napi has to be disabled first,
then DIM canceled.
Noticed while reading the code.
Fixes: 0bc0b97fca73 ("bnxt_en: cleanup DIM work on device shutdown") Fixes: 6a8788f25625 ("bnxt_en: add support for software dynamic interrupt moderation") Reviewed-by: Michael Chan <michael.chan@broadcom.com> Reviewed-by: Edwin Peer <edwin.peer@broadcom.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
We can't take the tx lock from the napi poll routine, because
netpoll can poll napi at any moment, including with the tx lock
already held.
The tx lock is protecting against two paths - the disable
path, and (as Michael points out) the NETDEV_TX_BUSY case
which may occur if NAPI completions race with start_xmit
and both decide to re-enable the queue.
For the disable/ifdown path use synchronize_net() to make sure
closing the device does not race we restarting the queues.
Annotate accesses to dev_state against data races.
For the NAPI cleanup vs start_xmit path - appropriate barriers
are already in place in the main spot where Tx queue is stopped
but we need to do the same careful dance in the TX_BUSY case.
Fixes: c0c050c58d84 ("bnxt_en: New Broadcom ethernet driver.") Reviewed-by: Michael Chan <michael.chan@broadcom.com> Reviewed-by: Edwin Peer <edwin.peer@broadcom.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
"access skb fields ok" verifier test fails on s390 with the "verifier
bug. zext_dst is set, but no reg is defined" message. The first insns
of the test prog are ...
... and the 3rd one is dead (this does not look intentional to me, but
this is a separate topic).
sanitize_dead_code() converts dead insns into "ja -1", but keeps
zext_dst. When opt_subreg_zext_lo32_rnd_hi32() tries to parse such
an insn, it sees this discrepancy and bails. This problem can be seen
only with JITs whose bpf_jit_needs_zext() returns true.
Fix by clearning dead insns' zext_dst.
The commits that contributed to this problem are:
1. 5aa5bd14c5f8 ("bpf: add initial suite for selftests"), which
introduced the test with the dead code.
2. 5327ed3d44b7 ("bpf: verifier: mark verified-insn with
sub-register zext flag"), which introduced the zext_dst flag.
3. 83a2881903f3 ("bpf: Account for BPF_FETCH in
insn_has_def32()"), which introduced the sanity check.
4. 9183671af6db ("bpf: Fix leakage under speculation on
mispredicted branches"), which bisect points to.
It's best to fix this on stable branches that contain the second one,
since that's the point where the inconsistency was introduced.
Fixes: 5327ed3d44b7 ("bpf: verifier: mark verified-insn with sub-register zext flag") Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20210812151811.184086-2-iii@linux.ibm.com Signed-off-by: Sasha Levin <sashal@kernel.org>
get_queue_type() comments that splict virtqueue is preferred, however,
the actual logic preferred packed virtqueues. Since firmware has not
supported packed virtqueues we ended up using split virtqueues as was
desired.
Since we do not advertise support for packed virtqueues, we add a check
to verify split virtqueues are indeed supported.
Fixes: 1a86b377aa21 ("vdpa/mlx5: Add VDPA driver for supported mlx5 devices") Signed-off-by: Eli Cohen <elic@nvidia.com> Link: https://lore.kernel.org/r/20210811053759.66752-1-elic@nvidia.com Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The current code treats an empty iotlb provdied in set_map() as a
special case and destroy the memory region object. This must not be done
since the virtqueue objects reference this MR. Doing so will cause the
driver unload to emit errors and log timeouts caused by the firmware
complaining on busy resources.
This patch treats an empty iotlb as any other change of mapping. In this
case, mlx5_vdpa_create_mr() will fail and the entire set_map() call to
fail.
This issue has not been encountered before but was seen to occur in a
non-official version of qemu. Since qemu is a userspace program, the
driver must protect against such case.
Fixes: 94abbccdf291 ("vdpa/mlx5: Add shared memory registration code") Signed-off-by: Eli Cohen <elic@nvidia.com> Link: https://lore.kernel.org/r/20210811053713.66658-1-elic@nvidia.com Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The vdpa_alloc_device() returns an error pointer upon
failure, not NULL. To handle the failure correctly, this
replaces NULL check with IS_ERR() check and propagate the
error upwards.
Fixes: 5a2414bc454e ("virtio: Intel IFC VF driver for VDPA") Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Xie Yongji <xieyongji@bytedance.com> Link: https://lore.kernel.org/r/20210715080026.242-3-xieyongji@bytedance.com Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Jason Wang <jasowang@redhat.com> Reviewed-by: Stefano Garzarella <sgarzare@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The vdpa_alloc_device() returns an error pointer upon
failure, not NULL. To handle the failure correctly, this
replaces NULL check with IS_ERR() check and propagate the
error upwards.
Fixes: 64b9f64f80a6 ("vdpa: introduce virtio pci driver") Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Xie Yongji <xieyongji@bytedance.com> Link: https://lore.kernel.org/r/20210715080026.242-2-xieyongji@bytedance.com Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Jason Wang <jasowang@redhat.com> Reviewed-by: Stefano Garzarella <sgarzare@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The vdpa_alloc_device() returns an error pointer upon
failure, not NULL. To handle the failure correctly, this
replaces NULL check with IS_ERR() check and propagate the
error upwards.
Fixes: 2c53d0f64c06 ("vdpasim: vDPA device simulator") Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Xie Yongji <xieyongji@bytedance.com> Link: https://lore.kernel.org/r/20210715080026.242-1-xieyongji@bytedance.com Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Stefano Garzarella <sgarzare@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
This fixes the incorrect calculation for integer overflow
when the last address of iova range is 0xffffffff.
Fixes: ec33d031a14b ("vhost: detect 32 bit integer wrap around") Reported-by: Jason Wang <jasowang@redhat.com> Signed-off-by: Xie Yongji <xieyongji@bytedance.com> Acked-by: Jason Wang <jasowang@redhat.com> Link: https://lore.kernel.org/r/20210728130756.97-2-xieyongji@bytedance.com Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
We have changed the return type for sysc_check_active_timer() from -EBUSY
to -ENXIO, but the gpt12 system timer fix still checks for -EBUSY. We are
also not returning on other errors like we did earlier as noted by
Pavel Machek <pavel@denx.de>.
Commit 3ff340e24c9d ("bus: ti-sysc: Fix gpt12 system timer issue with
reserved status") should have been updated for commit 65fb73676112
("bus: ti-sysc: suppress err msg for timers used as clockevent/source").
Let's fix the issue by checking for -ENXIO and returning on any other
errors as suggested by Pavel Machek <pavel@denx.de>.
Fixes: 3ff340e24c9d ("bus: ti-sysc: Fix gpt12 system timer issue with reserved status")
Depends-on: 65fb73676112 ("bus: ti-sysc: suppress err msg for timers used as clockevent/source") Reported-by: Pavel Machek <pavel@denx.de> Reviewed-by: Pavel Machek (CIP) <pavel@denx.de> Cc: Grygorii Strashko <grygorii.strashko@ti.com> Cc: Jarkko Nikula <jarkko.nikula@bitmer.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
VQs may be accessed to mark the device broken while they are
created/destroyed. Hence protect the access to the vqs list.
Fixes: e2dcdfe95c0b ("virtio: virtio_break_device() to mark all virtqueues broken.") Signed-off-by: Parav Pandit <parav@nvidia.com> Link: https://lore.kernel.org/r/20210721142648.1525924-4-parav@nvidia.com Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
GCC complains about empty macros in an 'if' statement, so convert
them to 'do {} while (0)' macros.
Fixes these build warnings:
net/dccp/output.c: In function 'dccp_xmit_packet':
../net/dccp/output.c:283:71: warning: suggest braces around empty body in an 'if' statement [-Wempty-body]
283 | dccp_pr_debug("transmit_skb() returned err=%d\n", err);
net/dccp/ackvec.c: In function 'dccp_ackvec_update_old':
../net/dccp/ackvec.c:163:80: warning: suggest braces around empty body in an 'else' statement [-Wempty-body]
163 | (unsigned long long)seqno, state);
Fixes: dc841e30eaea ("dccp: Extend CCID packet dequeueing interface") Fixes: 380240864451 ("dccp ccid-2: Update code for the Ack Vector input/registration routine") Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Cc: dccp@vger.kernel.org Cc: "David S. Miller" <davem@davemloft.net> Cc: Jakub Kicinski <kuba@kernel.org> Cc: Gerrit Renker <gerrit@erg.abdn.ac.uk> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The 1.2 GHz variant of the Armada 3720 SOC is unstable with DVFS: when
the SOC boots, the WTMI firmware sets clocks and AVS values that work
correctly with 1.2 GHz CPU frequency, but random crashes occur once
cpufreq driver starts scaling.
We do not know currently what is the reason:
- it may be that the voltage value for L0 for 1.2 GHz variant provided
by the vendor in the OTP is simply incorrect when scaling is used,
- it may be that some delay is needed somewhere,
- it may be something else.
The most sane solution now seems to be to simply forbid the cpufreq
driver on 1.2 GHz variant.
Signed-off-by: Marek Behún <kabel@kernel.org> Fixes: 92ce45fb875d ("cpufreq: Add DVFS support for Armada 37xx") Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Now fw_devlink expects driver to create and probe a struct device
for interrupt controller.
So lets convert this driver to simple platform_device with probe().
Also use platform_get_ and devm_ family function to get/allocate
resources and drop unused .compatible = "qeic".
[1] - https://lore.kernel.org/lkml/CAGETcx9PiX==mLxB9PO8Myyk6u2vhPVwTMsA5NkD-ywH5xhusw@mail.gmail.com Fixes: e590474768f1 ("driver core: Set fw_devlink=on by default") Fixes: ea718c699055 ("Revert "Revert "driver core: Set fw_devlink=on by default""") Signed-off-by: Maxim Kochetkov <fido_max@inbox.ru> Reported-by: kernel test robot <lkp@intel.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Acked-by: Saravana Kannan <saravanak@google.com> Signed-off-by: Li Yang <leoyang.li@nxp.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Due to 14f97f0b8e2b, the rawnand platforms without "secure-regions"
property defined in DT fails to probe. The issue is,
of_get_nand_secure_regions() errors out if
of_property_count_elems_of_size() returns a negative error code.
If the "secure-regions" property is not present in DT, then also we'll
get -EINVAL from of_property_count_elems_of_size() but it should not
be treated as an error for platforms not declaring "secure-regions"
in DT.
So fix this behaviour by checking for the existence of that property in
DT and return 0 if it is not present.
Fixes: 14f97f0b8e2b ("mtd: rawnand: Add a check in of_get_nand_secure_regions()") Reported-by: Martin Kaiser <martin@kaiser.cx> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Reviewed-by: Martin Kaiser <martin@kaiser.cx> Tested-by: Martin Kaiser <martin@kaiser.cx> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com> Link: https://lore.kernel.org/linux-mtd/20210727062813.32619-1-manivannan.sadhasivam@linaro.org Signed-off-by: Sasha Levin <sashal@kernel.org>
Fix the upper guard and the "removed_region", this fixes the random
crashes which used to occur in memory intensive loads. I'm not sure WHY
the upper guard being 0x2000 instead of 0x1000 doesn't fix this, but it
HAS to be 0x1000.
As the default definition breaks booting angler:
[ 1.862561] printk: console [ttyMSM0] enabled
[ 1.872260] msm_serial: driver initialized
D - 15524 - pm_driver_init, Delta
cont_splash_mem was introduced in 74d6d0a145835, but the problem
manifested after commit '86588296acbf ("fdt: Properly handle "no-map"
field in the memory region")'.
Disabling it because Angler's firmware does not report where the memory
is allocated (dmesg from downstream kernel):
[ 0.000000] cma: Found cont_splash_mem@0, memory base 0x0000000000000000, size 16 MiB, limit 0x0000000000000000
[ 0.000000] cma: CMA: reserved 16 MiB at 0x0000000000000000 for cont_splash_mem
Similar issue might be on Google Nexus 5X (lg-bullhead). Other MSM8992/4
are known to report correct address.
Fixes: 74d6d0a145835 ("arm64: dts: qcom: msm8994/8994-kitakami: Fix up the memory map") Suggested-by: Konrad Dybcio <konradybcio@gmail.com> Signed-off-by: Petr Vorel <petr.vorel@gmail.com> Link: https://lore.kernel.org/r/20210622191019.23771-1-petr.vorel@gmail.com Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
If probe_device is failing, iommu_group is not initialized because
iommu_group_add_device is not reached, so freeing it will result
in NULL pointer access.
iommu_bus_init
->bus_iommu_probe
->probe_iommu_group in for each:/* return -22 in fail case */
->iommu_probe_device
->__iommu_probe_device /* return -22 here.*/
-> ops->probe_device /* return -22 here.*/
-> iommu_group_get_for_dev
-> ops->device_group
-> iommu_group_add_device //good case
->remove_iommu_group //in fail case, it will remove group
->iommu_release_device
->iommu_group_remove_device // here we don't have group
In my case ops->probe_device (mtk_iommu_probe_device from
mtk_iommu_v1.c) is due to failing fwspec->ops mismatch.
cont_splash_mem has different memory mapping than generic from msm8994.dtsi:
[ 0.000000] cma: Found cont_splash_mem@0, memory base 0x0000000003400000, size 12 MiB, limit 0xffffffffffffffff
[ 0.000000] cma: CMA: reserved 12 MiB at 0x0000000003400000 for cont_splash_mem
This fixes boot.
Fixes: 976d321f32dc ("arm64: dts: qcom: msm8992: Make the DT an overlay on top of 8994") Signed-off-by: Petr Vorel <petr.vorel@gmail.com> Link: https://lore.kernel.org/r/20210713185734.380-3-petr.vorel@gmail.com Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
WSA881x powerdown pin is connected to GPIO1, GPIO2 not GPIO2 and GPIO3,
so correct this. This was working so far due to a shift bug in gpio driver,
however once that is fixed this will stop working, so fix this!
For some reason we forgot to add this dts change in last merge cycle so
currently audio is broken in 5.13 as the gpio driver fix already landed
in 5.13.
projects / linux-stable / log