When running a large number of packets per second with a high data rate
and long A-MPDUs, the packet loss threshold can be reached very quickly
when the link conditions change. This frequently shows up as spurious
disconnects.
Mitigate false positives by using a similar logic for regular stations
as the one being used for TDLS, though with a more aggressive timeout.
Packet loss events are only reported if no ACK was received for a second.
Fix cfg80211_chandef_usable():
consider IEEE80211_VHT_CAP_EXT_NSS_BW when verifying 160/80+80 MHz.
Based on:
"Table 9-272 — Setting of the Supported Channel Width Set subfield and Extended NSS BW
Support subfield at a STA transmitting the VHT Capabilities Information field"
From "Draft P802.11REVmd_D3.0.pdf"
Signed-off-by: Aviad Brikman <aviad.brikman@celeno.com> Signed-off-by: Shay Bar <shay.bar@celeno.com> Link: https://lore.kernel.org/r/20200826143139.25976-1-shay.bar@celeno.com
[reformat the code a bit and use u32_get_bits()] Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Set the skb's network_header before it is passed to the underlying
Ethernet device for transmission.
This patch fixes the following issue:
When we use this driver with AF_PACKET sockets, there would be error
messages of:
protocol 0805 is buggy, dev (Ethernet interface name)
printed in the system "dmesg" log.
This is because skbs passed down to the Ethernet device for transmission
don't have their network_header properly set, and the dev_queue_xmit_nit
function in net/core/dev.c complains about this.
Reason of setting the network_header to this place (at the end of the
Ethernet header, and at the beginning of the Ethernet payload):
Because when this driver receives an skb from the Ethernet device, the
network_header is also set at this place.
Cc: Martin Schiller <ms@dev.tdt.de> Signed-off-by: Xie He <xie.he.0141@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The inode chunk allocation transaction reserves inobt_maxlevels-1
blocks to accommodate a full split of the inode btree. A full split
requires an allocation for every existing level and a new root
block, which means inobt_maxlevels is the worst case block
requirement for a transaction that inserts to the inobt. This can
lead to a transaction block reservation overrun when tmpfile
creation allocates an inode chunk and expands the inobt to its
maximum depth. This problem has been observed in conjunction with
overlayfs, which makes frequent use of tmpfiles internally.
The existing reservation code goes back as far as the Linux git repo
history (v2.6.12). It was likely never observed as a problem because
the traditional file/directory creation transactions also include
worst case block reservation for directory modifications, which most
likely is able to make up for a single block deficiency in the inode
allocation portion of the calculation. tmpfile support is relatively
more recent (v3.15), less heavily used, and only includes the inode
allocation block reservation as tmpfiles aren't linked into the
directory tree on creation.
Fix up the inode alloc block reservation macro and a couple of the
block allocator minleft parameters that enforce an allocation to
leave enough free blocks in the AG for a full inobt split.
Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
when skb->encapsulation is 0, skb->ip_summed is CHECKSUM_PARTIAL
and it is udp packet, which has a dest port as the IANA assigned.
the hardware is expected to do the checksum offload, but the
hardware will not do the checksum offload when udp dest port is
6081.
This patch fixes it by doing the checksum in software.
Reported-by: Li Bing <libing@winhong.com> Signed-off-by: Yi Li <yili@winhong.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
In probe, IRQ is requested before zchan->id is initialized which can be
read in the irq handler. Hence, shift request irq after other initializations
complete.
Found by Linux Driver Verification project (linuxtesting.org).
The WAKEEN bits are used to indicate which bits in the
STATESTS register may cause wake event during the codec
state change request. Configure the WAKEEN register for
the Tegra to detect the wake events.
The Tegra HDA codec HW implementation has an issue related to not
swapping the 2 channel Audio Sample Packet(ASP) channel mapping.
Whatever the FL and FR mapping specified the left channel always
comes out of left speaker and right channel on right speaker. So
add condition to disallow the swapping of FL,FR during the playback.
Link: https://lore.kernel.org/r/20200824033436.45570-1-yebin10@huawei.com Acked-by: Saurav Kashyap <skashyap@marvell.com> Signed-off-by: Ye Bin <yebin10@huawei.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When make_rate() fails, vcc should be freed just
like other error paths in fs_open().
Signed-off-by: Dinghao Liu <dinghao.liu@zju.edu.cn> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
When down_killable() fails, skb_resp should be freed
just like when st95hf_spi_send() fails.
Signed-off-by: Dinghao Liu <dinghao.liu@zju.edu.cn> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The underlying Ethernet device may request necessary tailroom to be
allocated by setting needed_tailroom. This driver should also set
needed_tailroom to request the tailroom needed by the underlying
Ethernet device to be allocated.
Cc: Willem de Bruijn <willemdebruijn.kernel@gmail.com> Cc: Martin Schiller <ms@dev.tdt.de> Signed-off-by: Xie He <xie.he.0141@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Getting creative with nft and omitting the interval_overlap()
check from the set_overlap() function, without omitting
set_overlap() altogether, led to the observation of a partial
overlap that wasn't detected, and would actually result in
replacement of the end element of an existing interval.
This is due to the fact that we'll return -EEXIST on a matching,
pre-existing start element, instead of -ENOTEMPTY, and the error
is cleared by API if NLM_F_EXCL is not given. At this point, we
can insert a matching start, and duplicate the end element as long
as we don't end up into other intervals.
For instance, inserting interval 0 - 2 with an existing 0 - 3
interval would result in a single 0 - 2 interval, and a dangling
'3' end element. This is because nft will proceed after inserting
the '0' start element as no error is reported, and no further
conflicting intervals are detected on insertion of the end element.
This needs a different approach as it's a local condition that can
be detected by looking for duplicate ends coming from left and
right, separately. Track those and directly report -ENOTEMPTY on
duplicated end elements for a matching start.
If an sctp connection gets re-used, heartbeats are flagged as invalid
because their vtag doesn't match.
Handle this in a similar way as TCP conntrack when it suspects that the
endpoints and conntrack are out-of-sync.
When a HEARTBEAT request fails its vtag validation, flag this in the
conntrack state and accept the packet.
When a HEARTBEAT_ACK is received with an invalid vtag in the reverse
direction after we allowed such a HEARTBEAT through, assume we are
out-of-sync and re-set the vtag info.
v2: remove left-over snippet from an older incarnation that moved
new_state/old_state assignments, thats not needed so keep that
as-is.
Do not override ejtag feature to 0 as Loongson 3A1000+ do have ejtag.
For watch, as KVM emulated CPU doesn't have watch feature, we should
not enable it unconditionally.
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com> Reviewed-by: Huacai Chen <chenhc@lemote.com> Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
The acpi_get_table() should be coupled with acpi_put_table() if
the mapped table is not used at runtime to release the table
mapping, put the CSRT table buf after using it.
Newer version of HSDK aka HSDK-4xD (with dual issue HS48x4 CPU) wired up
the perf interrupt, so enable that in DT.
This is OK for old HSDK where this irq is ignored because pct irq is not
wired up in hardware.
Even without in-kernel LAPIC we should allow writing '0' to
MSR_KVM_ASYNC_PF_EN as we're not enabling the mechanism. In
particular, QEMU with 'kernel-irqchip=off' fails to start
a guest with
qemu-system-x86_64: error: failed to set MSR 0x4b564d02 to 0x0
Fixes: 9d3c447c72fb2 ("KVM: X86: Fix async pf caused null-ptr-deref") Reported-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20200911093147.484565-1-vkuznets@redhat.com>
[Actually commit the version proposed by Sean Christopherson. - Paolo] Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently we allocate rx buffers in a single contiguous buffers for
headers (iser and iscsi) and data trailer. This means that most likely the
data starting offset is aligned to 76 bytes (size of both headers).
This worked fine for years, but at some point this broke, resulting in
data corruptions in isert when a command comes with immediate data and the
underlying backend device assumes 512 bytes buffer alignment.
We assume a hard-requirement for all direct I/O buffers to be 512 bytes
aligned. To fix this, we should avoid passing unaligned buffers for I/O.
Instead, we allocate our recv buffers with some extra space such that we
can have the data portion align to 512 byte boundary. This also means that
we cannot reference headers or data using structure but rather
accessors (as they may move based on alignment). Also, get rid of the
wrong __packed annotation from iser_rx_desc as this has only harmful
effects (not aligned to anything).
This affects the rx descriptors for iscsi login and data plane.
Fixes: 3d75ca0adef4 ("block: introduce multi-page bvec helpers") Link: https://lore.kernel.org/r/20200904195039.31687-1-sagi@grimberg.me Reported-by: Stephen Rust <srust@blockbridge.com> Tested-by: Doug Dumitru <doug@dumitru.com> Signed-off-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The device .release function was not being set during the device
initialization. This was leading to the below warning, in error cases when
put_srv was called before device_add was called.
Warning:
Device '(null)' does not have a release() function, it is broken and must
be fixed. See Documentation/kobject.txt.
So, set the device .release function during device initialization in the
__alloc_srv() function.
Fixes: baa5b28b7a47 ("RDMA/rtrs-srv: Replace device_register with device_initialize and device_add") Link: https://lore.kernel.org/r/20200907102216.104041-1-haris.iqbal@cloud.ionos.com Signed-off-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Acked-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
If we hit the UINT_MAX limit of bio->bi_iter.bi_size and so we are anyway
not merging this page in this bio, then it make sense to make same_page
also as false before returning.
Without this patch, we hit below WARNING in iomap.
This mostly happens with very large memory system and / or after tweaking
vm dirty threshold params to delay writeback of dirty data.
Right now we are failing requests based on the controller state (which
is checked inline in nvmf_check_ready) however we should definitely
accept requests if the queue is live.
When entering controller reset, we transition the controller into
NVME_CTRL_RESETTING, and then return BLK_STS_RESOURCE for non-mpath
requests (have blk_noretry_request set).
This is also the case for NVME_REQ_USER for the wrong reason. There
shouldn't be any reason for us to reject this I/O in a controller reset.
We do want to prevent passthru commands on the admin queue because we
need the controller to fully initialize first before we let user passthru
admin commands to be issued.
In a non-mpath setup, this means that the requests will simply be
requeued over and over forever not allowing the q_usage_counter to drop
its final reference, causing controller reset to hang if running
concurrently with heavy I/O.
Fixes: 35897b920c8a ("nvme-fabrics: fix and refine state checks in __nvmf_check_ready") Reviewed-by: James Smart <james.smart@broadcom.com> Signed-off-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
In seccomp_set_mode_filter() with TSYNC | NEW_LISTENER, we first initialize
the listener fd, then check to see if we can actually use it later in
seccomp_may_assign_mode(), which can fail if anyone else in our thread
group has installed a filter and caused some divergence. If we can't, we
partially clean up the newly allocated file: we put the fd, put the file,
but don't actually clean up the *memory* that was allocated at
filter->notif. Let's clean that up too.
To accomplish this, let's hoist the actual "detach a notifier from a
filter" code to its own helper out of seccomp_notify_release(), so that in
case anyone adds stuff to init_listener(), they only have to add the
cleanup code in one spot. This does a bit of extra locking and such on the
failure path when the filter is not attached, but it's a slow failure path
anyway.
timeout_usec value calculation was wrong, the calculated value
was in msec instead of usec.
Fixes: 56a1485b102e ("i2c: npcm7xx: Add Nuvoton NPCM I2C controller driver") Signed-off-by: Tali Perry <tali.perry1@gmail.com> Reviewed-by: Avi Fishman <avifishman70@gmail.com> Reviewed-by: Joel Stanley <joel@jms.id.au> Reviewed-by: Alex Qiu <xqiu@google.com> Signed-off-by: Wolfram Sang <wsa@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
When trying to get a new fs root for a snapshot during the transaction
at transaction.c:create_pending_snapshot(), if btrfs_get_new_fs_root()
fails we leave "pending->snap" pointing to an error pointer, and then
later at ioctl.c:create_snapshot() we dereference that pointer, resulting
in a crash:
(gdb) list *(btrfs_mksubvol+0x438)
0x7c7b8 is in btrfs_mksubvol (fs/btrfs/ioctl.c:858).
853 ret = 0;
854 pending_snapshot->anon_dev = 0;
855 fail:
856 /* Prevent double freeing of anon_dev */
857 if (ret && pending_snapshot->snap)
858 pending_snapshot->snap->anon_dev = 0;
859 btrfs_put_root(pending_snapshot->snap);
860 btrfs_subvolume_release_metadata(root, &pending_snapshot->block_rsv);
861 free_pending:
862 if (pending_snapshot->anon_dev)
So fix this by setting "pending->snap" to NULL if we get an error from the
call to btrfs_get_new_fs_root() at transaction.c:create_pending_snapshot().
Fixes: 2dfb1e43f57dd3 ("btrfs: preallocate anon block device at first phase of snapshot creation") Signed-off-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The 'spi_stm32 44004000.spi: Communication suspended' message means that
when using PIO, the kernel did not read the FIFO fast enough and so the
SPI controller paused the transfer. Currently, this is printed on every
single such event, so if the kernel is busy and the controller is pausing
the transfers often, the kernel will be all the more busy scrolling this
message into the log buffer every few milliseconds. That is not helpful.
Instead, rate-limit the message and print it every once in a while. It is
not possible to use the default dev_warn_ratelimited(), because that is
still too verbose, as it prints 10 lines (DEFAULT_RATELIMIT_BURST) every
5 seconds (DEFAULT_RATELIMIT_INTERVAL). The policy here is to print 1 line
every 50 seconds (DEFAULT_RATELIMIT_INTERVAL * 10), because 1 line is more
than enough and the cycles saved on printing are better left to the CPU to
handle the SPI. However, dev_warn_once() is also not useful, as the user
should be aware that this condition is possibly recurring or ongoing. Thus
the custom rate-limit policy.
Finally, turn the message from dev_warn() to dev_dbg(), since the system
does not suffer any sort of malfunction if this message appears, it is
just slowing down. This further reduces the printing into the log buffer
and frees the CPU to do useful work.
Fixes: dcbe0d84dfa5 ("spi: add driver for STM32 SPI controller") Signed-off-by: Marek Vasut <marex@denx.de> Cc: Alexandre Torgue <alexandre.torgue@st.com> Cc: Amelie Delaunay <amelie.delaunay@st.com> Cc: Antonio Borneo <borneo.antonio@gmail.com> Cc: Mark Brown <broonie@kernel.org> Link: https://lore.kernel.org/r/20200905151913.117775-1-marex@denx.de Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
As the comments in this patch say, if we tune and find all phases are
valid it's _almost_ as bad as no phases being found valid. Probably
all phases are not really reliable but we didn't detect where the
unreliable place is. That means we'll essentially be guessing and
hoping we get a good phase.
This is not just a problem in theory. It was causing real problems on
a real board. On that board, most often phase 10 is found as the only
invalid phase, though sometimes 10 and 11 are invalid and sometimes
just 11. Some percentage of the time, however, all phases are found
to be valid. When this happens, the current logic will decide to use
phase 11. Since phase 11 is sometimes found to be invalid, this is a
bad choice. Sure enough, when phase 11 is picked we often get mmc
errors later in boot.
I have seen cases where all phases were found to be valid 3 times in a
row, so increase the retry count to 10 just to be extra sure.
Fixes: 415b5a75da43 ("mmc: sdhci-msm: Add platform_execute_tuning implementation") Signed-off-by: Douglas Anderson <dianders@chromium.org> Reviewed-by: Veerabhadrarao Badiganti <vbadigan@codeaurora.org> Acked-by: Adrian Hunter <adrian.hunter@intel.com> Link: https://lore.kernel.org/r/20200827075809.1.If179abf5ecb67c963494db79c3bc4247d987419b@changeid Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The commit 61d7437ed1390 ("mmc: sdhci-acpi: Fix HS400 tuning for AMDI0040")
broke resume for eMMC HS400. When the system suspends the eMMC controller
is powered down. So, on resume we need to reinitialize the controller.
Although, amd_sdhci_host was not getting cleared, so the DLL was never
re-enabled on resume. This results in HS400 being non-functional.
To fix the problem, this change clears the tuned_clock flag, clears the
dll_enabled flag and disables the DLL on reset.
Fixes: 61d7437ed1390 ("mmc: sdhci-acpi: Fix HS400 tuning for AMDI0040") Signed-off-by: Raul E Rangel <rrangel@chromium.org> Acked-by: Adrian Hunter <adrian.hunter@intel.com> Link: https://lore.kernel.org/r/20200831150517.1.I93c78bfc6575771bb653c9d3fca5eb018a08417d@changeid Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 196f878a7ac2e (" KVM: arm/arm64: Signal SIGBUS when stage2 discovers
hwpoison memory") modifies user_mem_abort() to send a SIGBUS signal when
the fault IPA maps to a hwpoisoned page. Commit 1559b7583ff6 ("KVM:
arm/arm64: Re-check VMA on detecting a poisoned page") changed
kvm_send_hwpoison_signal() to use the page shift instead of the VMA because
at that point the code had already released the mmap lock, which means
userspace could have modified the VMA.
If userspace uses hugetlbfs for the VM memory, user_mem_abort() tries to
map the guest fault IPA using block mappings in stage 2. That is not always
possible, if, for example, userspace uses dirty page logging for the VM.
Update the page shift appropriately in those cases when we downgrade the
stage 2 entry from a block mapping to a page.
Fixes: 1559b7583ff6 ("KVM: arm/arm64: Re-check VMA on detecting a poisoned page") Signed-off-by: Alexandru Elisei <alexandru.elisei@arm.com> Signed-off-by: Marc Zyngier <maz@kernel.org> Reviewed-by: Gavin Shan <gshan@redhat.com> Link: https://lore.kernel.org/r/20200901133357.52640-2-alexandru.elisei@arm.com Signed-off-by: Sasha Levin <sashal@kernel.org>
Unlike what we previously thought, only the per-pixel alpha is broken on
the lowest plane and the per-plane alpha isn't. Remove the check on the
alpha property being set on the lowest plane to reject a mode.
Fixes: dcf496a6a608 ("drm/sun4i: sun4i: Introduce a quirk for lowest plane alpha support") Signed-off-by: Maxime Ripard <maxime@cerno.tech> Reviewed-by: Chen-Yu Tsai <wens@csie.org> Cc: Paul Kocialkowski <paul.kocialkowski@bootlin.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200728134810.883457-1-maxime@cerno.tech Signed-off-by: Sasha Levin <sashal@kernel.org>
Function sun8i_vi_layer_get_csc_mode() is supposed to return CSC mode
but due to inproper return type (bool instead of u32) it returns just 0
or 1. Colors are wrong for YVU formats because of that.
Fixes: daab3d0e8e2b ("drm/sun4i: de2: csc_mode in de2 format struct is mostly redundant") Reported-by: Roman Stratiienko <r.stratiienko@gmail.com> Signed-off-by: Jernej Skrabec <jernej.skrabec@siol.net> Tested-by: Roman Stratiienko <r.stratiienko@gmail.com> Signed-off-by: Maxime Ripard <maxime@cerno.tech> Link: https://patchwork.freedesktop.org/patch/msgid/20200901220305.6809-1-jernej.skrabec@siol.net Signed-off-by: Sasha Levin <sashal@kernel.org>
padata_do_parallel() takes padata_works_lock with softirqs enabled, so a
deadlock is possible if, on the same CPU, the lock is acquired in
process context and then softirq handling done in an interrupt leads to
the same path.
Fix by leaving softirqs disabled while do_parallel holds
padata_works_lock.
Reported-by: syzbot+f4b9f49e38e25eb4ef52@syzkaller.appspotmail.com Fixes: 4611ce2246889 ("padata: allocate work structures for parallel jobs from a pool") Signed-off-by: Daniel Jordan <daniel.m.jordan@oracle.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: Steffen Klassert <steffen.klassert@secunet.com> Cc: linux-crypto@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Sasha Levin <sashal@kernel.org>
Small BW votes that translate to less than a single BCM unit are
currently truncated to zero. Ensure that non-zero BW requests always
result in at least a vote of 1 to BCM.
Use of the new -flive-patching flag was introduced with the following
commit:
43bd3a95c98e ("kbuild: use -flive-patching when CONFIG_LIVEPATCH is enabled")
This flag has several drawbacks:
- It disables some optimizations, so it can have a negative effect on
performance.
- According to the GCC documentation it's not compatible with LTO, which
will become a compatibility issue as LTO support gets upstreamed in
the kernel.
- It was intended to be used for source-based patch generation tooling,
as opposed to binary-based patch generation tooling (e.g.,
kpatch-build). It probably should have at least been behind a
separate config option so as not to negatively affect other livepatch
users.
- Clang doesn't have the flag, so as far as I can tell, this method of
generating patches is incompatible with Clang, which like LTO is
becoming more mainstream.
- It breaks GCC's implicit noreturn detection for local functions. This
is the cause of several "unreachable instruction" objtool warnings.
- The broken noreturn detection is an obvious GCC regression, but we
haven't yet gotten GCC developers to acknowledge that, which doesn't
inspire confidence in their willingness to keep the feature working as
optimizations are added or changed going forward.
- While there *is* a distro which relies on this flag for their distro
livepatch module builds, there's not a publicly documented way to
create safe livepatch modules with it. Its use seems to be based on
tribal knowledge. It serves no benefit to those who don't know how to
use it.
(In fact, I believe the current livepatch documentation and samples
are misleading and dangerous, and should be corrected. Or at least
amended with a disclaimer. But I don't feel qualified to make such
changes.)
Also, we have an idea for using objtool to detect function changes,
which could potentially obsolete the need for this flag anyway.
At this point the flag has no benefits for upstream which would
counteract the above drawbacks. Revert it until it becomes more ready.
disable_irq() might sleep, replace it with disable_irq_nosync(). For
synchronisation 'irq_poll_scheduled' is sufficient
Fixes: 320e77acb3 scsi: mpt3sas: Irq poll to avoid CPU hard lockups Link: https://lore.kernel.org/r/20200901145026.12174-1-thenzl@redhat.com Signed-off-by: Tomas Henzl <thenzl@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
disable_irq() might sleep. Replace it with disable_irq_nosync() which is
sufficient as irq_poll_scheduled protects against concurrently running
complete_cmd_fusion() from megasas_irqpoll() and megasas_isr_fusion().
Link: https://lore.kernel.org/r/20200827165332.8432-1-thenzl@redhat.com Fixes: a6ffd5bf681 scsi: megaraid_sas: Call disable_irq from process IRQ poll Signed-off-by: Tomas Henzl <thenzl@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When the returned speed from __ethtool_get_link_ksettings() is
SPEED_UNKNOWN this will lead to reporting a wrong speed and width for
providers that uses ib_get_eth_speed(), fix that by defaulting the
netdev_speed to SPEED_1000 in case the returned value from
__ethtool_get_link_ksettings() is SPEED_UNKNOWN.
Fixes: d41861942fc5 ("IB/core: Add generic function to extract IB speed from netdev") Link: https://lore.kernel.org/r/20200902124304.170912-1-kamalheib1@gmail.com Signed-off-by: Kamal Heib <kamalheib1@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
It's not safe to access the next CQ in list_for_each_entry() after
invoking ib_free_cq(), because the CQ has already been freed in current
iteration. It should be replaced by list_for_each_entry_safe().
Fixes: c7ff819aefea ("RDMA/core: Introduce shared CQ pool API") Link: https://lore.kernel.org/r/1598963935-32335-1-git-send-email-liweihang@huawei.com Signed-off-by: Xi Wang <wangxi11@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When going through a disable/enable cycle without changing the
framebuffer the optimization added by commit 3954ff10e06e ("drm/virtio:
skip set_scanout if framebuffer didn't change") causes the screen stay
blank. Add a bool to force an update to fix that.
v2: use drm_atomic_crtc_needs_modeset() (Daniel).
Cc: 1882851@bugs.launchpad.net Fixes: 3954ff10e06e ("drm/virtio: skip set_scanout if framebuffer didn't change") Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> Tested-by: Jiri Slaby <jirislaby@kernel.org> Tested-by: Diego Viola <diego.viola@gmail.com> Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20200818072511.6745-2-kraxel@redhat.com
(cherry picked from commit 1bc371cd0ec907bab870cacb6e898105f9c41dc8) Signed-off-by: Sasha Levin <sashal@kernel.org>
It was discovered that sdparm will fail when attempting to disable write
cache on a SATA disk connected via libsas.
In the ATA command set the write cache state is controlled through the SET
FEATURES operation. This is roughly corresponds to MODE SELECT in SCSI and
the latter command is what is used in the SCSI-ATA translation layer. A
subtle difference is that a MODE SELECT carries data whereas SET FEATURES
is defined as a non-data command in ATA.
Set the DMA data direction to DMA_NONE if the requested ATA command is
identified as non-data.
[mkp: commit desc]
Fixes: fa1c1e8f1ece ("[SCSI] Add SATA support to libsas") Link: https://lore.kernel.org/r/1598426666-54544-1-git-send-email-luojiaxing@huawei.com Reviewed-by: John Garry <john.garry@huawei.com> Reviewed-by: Jason Yan <yanaijie@huawei.com> Signed-off-by: Luo Jiaxing <luojiaxing@huawei.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 98aee70d19a7 ("qla2xxx: Add endianizer to max_payload_size
modifier.") in 2014 broke qla2xxx on sparc64, e.g. as in the Sun Blade 1000
/ 2000. Unbreak by partial revert to fix endianness in nvram firmware
default initialization. Also mark the second frame_payload_size in nvram_t
__le16 to avoid new sparse warnings.
Link: https://lore.kernel.org/r/20200827.222729.1875148247374704975.rene@exactcode.com Fixes: 98aee70d19a7 ("qla2xxx: Add endianizer to max_payload_size modifier.") Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Acked-by: Arun Easi <aeasi@marvell.com> Signed-off-by: René Rebe <rene@exactcode.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
It's writing too much data. regmap_bulk_write expects number of
register sized chunks to write, not a byte sized length of the
bounce buffer. Bounce buffer needs to be padded too, so that
regmap_bulk_write will not read past the end of the buffer.
If sun8i_r40_tcon_tv_set_mux() succeed, sun8i_r40_tcon_tv_set_mux()
doesn't have a corresponding put_device(). Thus add put_device()
to fix the exception handling for this function implementation.
Driver crashes when destroy_qp is re-tried because of an error
returned. This is because the qp entry was removed from the qp list during
the first call.
Remove qp from the list only if destroy_qp returns success.
The driver will still trigger a WARN_ON due to the memory leaking, but at
least it isn't corrupting memory too.
When computing the first psn entry, driver checks for page alignment. If
this address is not page aligned,it attempts to compute the offset in that
page for later use by using ALIGN macro. ALIGN macro does not return
offset bytes but the requested aligned address and hence cannot be used
directly to store as offset. Since driver was using the address itself
instead of offset, it resulted in invalid address when filling the psn
buffer.
Fixed driver to use PAGE_MASK macro to calculate the offset.
At first, driver allocates memory for NQ based on qplib_ctx->cq_count and
qplib_ctx->srqc_count. Later when creating ring, it uses a static value
of 128K -1.
Fixing this with a static value for now.
Fixes: b08fe048a69d ("RDMA/bnxt_re: Refactor net ring allocation function") Link: https://lore.kernel.org/r/1598292876-26529-5-git-send-email-selvin.xavier@broadcom.com Signed-off-by: Naresh Kumar PBS <nareshkumar.pbs@broadcom.com> Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
qp->id can be a value outside the max number of qp. Indexing the qp table
with the id can cause out of bounds crash. So changing the qp table
indexing by (qp->id % max_qp -1).
Allocating one extra entry for QP1. Some adapters create one more than the
max_qp requested to accommodate QP1. If the qp->id is 1, store the
inforamtion in the last entry of the qp table.
QP1 Rx CQE reports transparent VLAN ID in the completion and this is used
while reporting the completion for received MAD packet. Check if the vlan
id is configured before reporting it in the work completion.
To avoid the following kernel panic when calling kmem_cache_create() with
a NULL pointer from pool_cache(), Block the rxe_param_set_add() from
running if the rdma_rxe module is not initialized.
This can be triggered if a user tries to use the 'module option' which is
not actually a real module option but some idiotic (and thankfully no
obsolete) sysfs interface.
There are error cases when we will call free_srv before device kobject is
initialized; in such cases calling put_device generates the following
warning:
kobject: '(null)' (000000009f5445ed): is not initialized, yet
kobject_put() is being called.
So call device_initialize() only once when the server is allocated. If we
end up calling put_srv() and subsequently free_srv(), our call to
put_device() would result in deletion of the obj. Call device_add() later
when we actually have a connection. Correspondingly, call device_del()
instead of device_unregister() when srv->dev_ref falls to 0.
Fixes: 9cb837480424 ("RDMA/rtrs: server: main functionality") Link: https://lore.kernel.org/r/20200811092722.2450-1-haris.iqbal@cloud.ionos.com Suggested-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Reviewed-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Since commit bcf3440c6dd7 ("net: phy: micrel: add phy-mode support for the
KSZ9031 PHY") the networking is broken on the imx7d-zii-rmu2 board.
The end result is that network receive behaviour is marginal with lots of
RX CRC errors experienced and NFS frequently failing.
Quoting the explanation from Andrew Lunn in commit 0672d22a19244
("ARM: dts: imx: Fix the AR803X phy-mode"):
"The problem here is, all the DTs were broken since day 0. However,
because the PHY driver was also broken, nobody noticed and it
worked. Now that the PHY driver has been fixed, all the bugs in the
DTs now become an issue"
Fix it by switching to phy-mode = "rgmii-id".
Fixes: bcf3440c6dd7 ("net: phy: micrel: add phy-mode support for the KSZ9031 PHY") Signed-off-by: Chris Healy <cphealy@gmail.com> Reviewed-by: Fabio Estevam <festevam@gmail.com> Signed-off-by: Shawn Guo <shawnguo@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Correct gpio ranges according to i.MX7ULP pinctrl driver:
gpio_ptc: ONLY pin 0~19 are available;
gpio_ptd: ONLY pin 0~11 are available;
gpio_pte: ONLY pin 0~15 are available;
gpio_ptf: ONLY pin 0~19 are available;
The following 4 tests in timers can take longer than the default 45
seconds that added in commit 852c8cbf34d3 ("selftests/kselftest/runner.sh:
Add 45 second timeout per test") to run:
* nsleep-lat - 2m7.350s
* set-timer-lat - 2m0.66s
* inconsistency-check - 1m45.074s
* raw_skew - 2m0.013s
Thus they will be marked as failed with the current 45s setting:
not ok 3 selftests: timers: nsleep-lat # TIMEOUT
not ok 4 selftests: timers: set-timer-lat # TIMEOUT
not ok 6 selftests: timers: inconsistency-check # TIMEOUT
not ok 7 selftests: timers: raw_skew # TIMEOUT
Disable the timeout setting for timers can make these tests finish
properly:
ok 3 selftests: timers: nsleep-lat
ok 4 selftests: timers: set-timer-lat
ok 6 selftests: timers: inconsistency-check
ok 7 selftests: timers: raw_skew
https://bugs.launchpad.net/bugs/1864626 Fixes: 852c8cbf34d3 ("selftests/kselftest/runner.sh: Add 45 second timeout per test") Signed-off-by: Po-Hsu Lin <po-hsu.lin@canonical.com> Acked-by: John Stultz <john.stultz@linaro.org> Signed-off-by: Shuah Khan <skhan@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Move another allocation out of regulator_list_mutex-protected region, as
reclaim might want to take the same lock.
WARNING: possible circular locking dependency detected
5.7.13+ #534 Not tainted
------------------------------------------------------
kswapd0/383 is trying to acquire lock: c0e5d920 (regulator_list_mutex){+.+.}-{3:3}, at: regulator_lock_dependent+0x54/0x2c0
but task is already holding lock: c0e38518 (fs_reclaim){+.+.}-{0:0}, at: __fs_reclaim_acquire+0x0/0x50
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #1 (fs_reclaim){+.+.}-{0:0}:
fs_reclaim_acquire.part.11+0x40/0x50
fs_reclaim_acquire+0x24/0x28
kmem_cache_alloc_trace+0x40/0x1e8
regulator_register+0x384/0x1630
devm_regulator_register+0x50/0x84
reg_fixed_voltage_probe+0x248/0x35c
[...]
other info that might help us debug this:
*** DEADLOCK ***
[...]
2 locks held by kswapd0/383:
#0: c0e38518 (fs_reclaim){+.+.}-{0:0}, at: __fs_reclaim_acquire+0x0/0x50
#1: cb70e5e0 (hctx->srcu){....}-{0:0}, at: hctx_lock+0x60/0xb8
[...]
Fixes: 541d052d7215 ("regulator: core: Only support passing enable GPIO descriptors")
[this commit only changes context] Fixes: f8702f9e4aa7 ("regulator: core: Use ww_mutex for regulators locking")
[this is when the regulator_list_mutex was introduced in reclaim locking path]
A previous commit removed the panel-dpi driver, which made the
SOM-LV video stop working because it relied on the DPI driver
for setting video timings. Now that the simple-panel driver is
available in omap2plus, this patch migrates the SOM-LV dev kits
to use a similar panel and remove the manual timing requirements.
A similar patch was already done and applied to the Torpedo family.
Fixes: 8bf4b1621178 ("drm/omap: Remove panel-dpi driver") Signed-off-by: Adam Ford <aford173@gmail.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Older versions of U-Boot would pinmux the whole board, but as
the bootloader got updated, it started to only pinmux the pins
it needed, and expected Linux to configure what it needed.
Unfortunately this caused an issue with the audio, because the
mcbsp2 pins were configured in the device tree but never
referenced by the driver. When U-Boot stopped muxing the audio
pins, the audio died.
This patch adds the references to the associate the pin controller
with the mcbsp2 driver which makes audio operate again.
Fixes: 5cb8b0fa55a9 ("ARM: dts: Move most of logicpd-som-lv-37xx-devkit.dts to logicpd-som-lv-baseboard.dtsi") Signed-off-by: Adam Ford <aford173@gmail.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Older versions of U-Boot would pinmux the whole board, but as
the bootloader got updated, it started to only pinmux the pins
it needed, and expected Linux to configure what it needed.
Unfortunately this caused an issue with the audio, because the
mcbsp2 pins were configured in the device tree, they were never
referenced by the driver. When U-Boot stopped muxing the audio
pins, the audio died.
This patch adds the references to the associate the pin controller
with the mcbsp2 driver which makes audio operate again.
Fixes: 739f85bba5ab ("ARM: dts: Move most of logicpd-torpedo-37xx-devkit to logicpd-torpedo-baseboard") Signed-off-by: Adam Ford <aford173@gmail.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
After subflow lock is dropped, more wmem might have been made available.
This fixes a deadlock in mptcp_connect.sh 'mmap' mode: wmem is exhausted.
But as the mptcp socket holds on to already-acked data (for retransmit)
no wakeup will occur.
Using 'goto restart' calls mptcp_clean_una(sk) which will free pages
that have been acked completely in the mean time.
Fixes: fb529e62d3f3 ("mptcp: break and restart in case mptcp sndbuf is full") Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
napi_disable() makes sure to set the NAPI_STATE_NPSVC bit to prevent
netpoll from accessing rings before init is complete. However, the
same is not done for fresh napi instances in netif_napi_add(),
even though we expect NAPI instances to be added as disabled.
This causes crashes during driver reconfiguration (enabling XDP,
changing the channel count) - if there is any printk() after
netif_napi_add() but before napi_enable().
To ensure memory ordering is correct we need to use RCU accessors.
Reported-by: Rob Sherwood <rsher@fb.com> Fixes: 2d8bff12699a ("netpoll: Close race condition between poll_one_napi and napi_disable") Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The 'this_cpu_ptr()' is used to obtain the AEAD key' TFM on the current
CPU for encryption, however the execution can be preemptible since it's
actually user-space context, so the 'using smp_processor_id() in
preemptible' has been observed.
We fix the issue by using the 'get/put_cpu_ptr()' API which consists of
a 'preempt_disable()' instead.
Fixes: fc1b6d6de220 ("tipc: introduce TIPC encryption & authentication") Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
int main(int argc, char *argv[])
{
const int fd = open("/dev/nbd0", 3);
alarm(5);
ioctl(fd, NBD_SET_SOCK, socket(PF_TIPC, SOCK_DGRAM, 0));
ioctl(fd, NBD_DO_IT, 0); /* To be interrupted by SIGALRM. */
return 0;
}
----------
One problem is that wait_for_completion() from flush_workqueue() from
nbd_start_device_ioctl() from nbd_ioctl() cannot be completed when
nbd_start_device_ioctl() received a signal at wait_event_interruptible(),
for tipc_shutdown() from kernel_sock_shutdown(SHUT_RDWR) from
nbd_mark_nsock_dead() from sock_shutdown() from nbd_start_device_ioctl()
is failing to wake up a WQ thread sleeping at wait_woken() from
tipc_wait_for_rcvmsg() from sock_recvmsg() from sock_xmit() from
nbd_read_stat() from recv_work() scheduled by nbd_start_device() from
nbd_start_device_ioctl(). Fix this problem by always invoking
sk->sk_state_change() (like inet_shutdown() does) when tipc_shutdown() is
called.
The other problem is that tipc_wait_for_rcvmsg() cannot return when
tipc_shutdown() is called, for tipc_shutdown() sets sk->sk_shutdown to
SEND_SHUTDOWN (despite "how" is SHUT_RDWR) while tipc_wait_for_rcvmsg()
needs sk->sk_shutdown set to RCV_SHUTDOWN or SHUTDOWN_MASK. Fix this
problem by setting sk->sk_shutdown to SHUTDOWN_MASK (like inet_shutdown()
does) when the socket is connectionless.
Since commit 9c66d1564676 ("taprio: Add support for hardware
offloading") there's a bit of inconsistency when offloading schedules
to the hardware:
In software mode, the gate masks are specified in terms of traffic
classes, so if say "sched-entry S 03 20000", it means that the traffic
classes 0 and 1 are open for 20us; when taprio is offloaded to
hardware, the gate masks are specified in terms of hardware queues.
The idea here is to fix hardware offloading, so schedules in hardware
and software mode have the same behavior. What's needed to do is to
map traffic classes to queues when applying the offload to the driver.
Fixes: 9c66d1564676 ("taprio: Add support for hardware offloading") Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With disabling bh in the whole sctp_get_port_local(), when
snum == 0 and too many ports have been used, the do-while
loop will take the cpu for a long time and cause cpu stuck:
There's no need to disable bh in the whole function of
sctp_get_port_local. So fix this cpu stuck by removing
local_bh_disable() called at the beginning, and using
spin_lock_bh() instead.
The same thing was actually done for inet_csk_get_port() in
Commit ea8add2b1903 ("tcp/dccp: better use of ephemeral
ports in bind()").
Thanks to Marcelo for pointing the buggy code out.
v1->v2:
- use cond_resched() to yield cpu to other tasks if needed,
as Eric noticed.
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: Ying Xu <yinxu@redhat.com> Signed-off-by: Xin Long <lucien.xin@gmail.com> Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Keenetic Plus DSL is a xDSL modem that uses dm9620 as its USB interface.
Signed-off-by: Kamil Lorenc <kamil@re-ws.pl> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch fixes two main problems seen when removing NetLabel
mappings: memory leaks and potentially extra audit noise.
The memory leaks are caused by not properly free'ing the mapping's
address selector struct when free'ing the entire entry as well as
not properly cleaning up a temporary mapping entry when adding new
address selectors to an existing entry. This patch fixes both these
problems such that kmemleak reports no NetLabel associated leaks
after running the SELinux test suite.
The potentially extra audit noise was caused by the auditing code in
netlbl_domhsh_remove_entry() being called regardless of the entry's
validity. If another thread had already marked the entry as invalid,
but not removed/free'd it from the list of mappings, then it was
possible that an additional mapping removal audit record would be
generated. This patch fixes this by returning early from the removal
function when the entry was previously marked invalid. This change
also had the side benefit of improving the code by decreasing the
indentation level of large chunk of code by one (accounting for most
of the diffstat).
Fixes: 63c416887437 ("netlabel: Add network address selectors to the NetLabel/LSM domain mapping") Reported-by: Stephen Smalley <stephen.smalley.work@gmail.com> Signed-off-by: Paul Moore <paul@paul-moore.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cited commit added the possible value of '2', but it cannot be set. Fix
it by adjusting the maximum value to '2'. This is consistent with the
corresponding IPv4 sysctl.
Fixes: d8f74f0975d8 ("ipv6: Support multipath hashing on inner IP pkts") Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: Stephen Suryaputra <ssuryaextr@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fib_info_notify_update() is always called with RTNL held, but not from
an RCU read-side critical section. This leads to the following warning
[1] when the FIB table list is traversed with
hlist_for_each_entry_rcu(), but without a proper lockdep expression.
Since modification of the list is protected by RTNL, silence the warning
by adding a lockdep expression which verifies RTNL is held.
[1]
=============================
WARNING: suspicious RCU usage 5.9.0-rc1-custom-14233-g2f26e122d62f #129 Not tainted
-----------------------------
net/ipv4/fib_trie.c:2124 RCU-list traversed in non-reader section!!
other info that might help us debug this:
rcu_scheduler_active = 2, debug_locks = 1
1 lock held by ip/834:
#0: ffffffff85a3b6b0 (rtnl_mutex){+.+.}-{3:3}, at: rtnetlink_rcv_msg+0x49a/0xbd0
Fixes: 1bff1a0c9bbd ("ipv4: Add function to send route updates") Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When a rdma_cm_id needs to be destroyed after a handler callback fails,
part of the destruction pattern is open coded into each call site.
Unfortunately the blind assignment to state discards important information
needed to do cma_cancel_operation(). This results in active operations
being left running after rdma_destroy_id() completes, and the
use-after-free bugs from KASAN.
Consolidate this entire pattern into destroy_id_handler_unlock() and
manage the locking correctly. The state should be set to
RDMA_CM_DESTROYING under the handler_lock to atomically ensure no futher
handlers are called.
Link: https://lore.kernel.org/r/20200723070707.1771101-5-leon@kernel.org Reported-by: syzbot+08092148130652a6faae@syzkaller.appspotmail.com Reported-by: syzbot+a929647172775e335941@syzkaller.appspotmail.com Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The REQ flows are concerned that once the handler is called on the new
cm_id the ULP can choose to trigger a rdma_destroy_id() concurrently at
any time.
However, this is not true, while the ULP can call rdma_destroy_id(), it
immediately blocks on the handler_mutex which prevents anything harmful
from running concurrently.
Remove the confusing extra locking and refcounts and make the
handler_mutex protecting state during destroy more clear.
Whenever an event is delivered to the handler it should be done under the
handler_mutex and upon any non-zero return from the handler it should
trigger destruction of the cm_id.
cma_process_remove() skips some steps here, it is not necessarily wrong
since the state change should prevent any races, but it is confusing and
unnecessary.
Follow the standard pattern here, with the slight twist that the
transition to RDMA_CM_DEVICE_REMOVAL includes a cma_cancel_operation().
cma_process_remove() triggers an unconditional rdma_destroy_id() for
internal_id's and skips the event deliver and transition through
RDMA_CM_DEVICE_REMOVAL.
This is confusing and unnecessary. internal_id always has
cma_listen_handler() as the handler, have it catch the
RDMA_CM_DEVICE_REMOVAL event and directly consume it and signal removal.
This way the FSM sequence never skips the DEVICE_REMOVAL case and the
logic in this hard to test area is simplified.
The buffer size is 2 Bytes and we expect to receive the same amount of
data. But sometimes we receive less data and run into uninit-was-stored
issue upon read. Hence modify the error check on the return value to match
with the buffer size as a prevention.
Reported-and-tested by: syzbot+a7e220df5a81d1ab400e@syzkaller.appspotmail.com Signed-off-by: Himadri Pandya <himadrispandya@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
collapse_file() in khugepaged passes PAGE_SIZE as the number of pages to
be read to page_cache_sync_readahead(). The intent was probably to read
a single page. Fix it to use the number of pages to the end of the
window instead.
Fixes: 99cb0dbd47a1 ("mm,thp: add read-only THP support for (non-shmem) FS") Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Reviewed-by: Matthew Wilcox (Oracle) <willy@infradead.org> Acked-by: Song Liu <songliubraving@fb.com> Acked-by: Yang Shi <shy828301@gmail.com> Acked-by: Pankaj Gupta <pankaj.gupta.linux@gmail.com> Cc: Eric Biggers <ebiggers@google.com> Link: https://lkml.kernel.org/r/20200903140844.14194-2-willy@infradead.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There is a race between the assignment of `table->data` and write value
to the pointer of `table->data` in the __do_proc_doulongvec_minmax() on
the other thread.
Fix this by duplicating the `table`, and only update the duplicate of
it. And introduce a helper of proc_hugetlb_doulongvec_minmax() to
simplify the code.
Since commit cf11e85fc08c ("mm: hugetlb: optionally allocate gigantic
hugepages using cma"), the gigantic page would be allocated from node
which is not the preferred node, although there are pages available from
that node. The reason is that the nid parameter has been ignored in
alloc_gigantic_page().
Besides, the __GFP_THISNODE also need be checked if user required to
alloc only from the preferred node.
After this patch, the preferred node is tried first before other allowed
nodes, and don't try to allocate from other nodes if __GFP_THISNODE is
specified. If user don't specify the preferred node, the current node
will be used as preferred node, which makes sure consistent behavior of
allocating gigantic and non-gigantic hugetlb page.
Fixes: cf11e85fc08c ("mm: hugetlb: optionally allocate gigantic hugepages using cma") Signed-off-by: Li Xinhai <lixinhai.lxh@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Roman Gushchin <guro@fb.com> Link: https://lkml.kernel.org/r/20200902025016.697260-1-lixinhai.lxh@gmail.com Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
projects / thirdparty / kernel / stable.git / log
