Variables off and len typed as uint32 in rndis_query function
are controlled by incoming RNDIS response message thus their
value may be manipulated. Setting off to a unexpectetly large
value will cause the sum with len and 8 to overflow and pass
the implemented validation step. Consequently the response
pointer will be referring to a location past the expected
buffer boundaries allowing information leakage e.g. via
RNDIS_OID_802_3_PERMANENT_ADDRESS OID.
Fixes: ddda08624013 ("USB: rndis_host, various cleanups") Signed-off-by: Szymon Heidrich <szymon.heidrich@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Current code uses per_cpu pointer to get the lmtst_id mapped to
the core on which aura_free() is executed. Using per_cpu pointer
without preemption disable causing mismatch between lmtst_id and
core on which pointer gets freed. This patch fixes the issue by
disabling preemption around aura_free.
Fixes: ef6c8da71eaf ("octeontx2-pf: cn10K: Reserve LMTST lines per core") Signed-off-by: Sunil Goutham <sgoutham@marvell.com> Signed-off-by: Geetha sowjanya <gakula@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
syzbot is reporting hung task at do_user_addr_fault() [1], for there is
a silent deadlock between PG_locked bit and ni_lock lock.
Since filemap_update_page() calls filemap_read_folio() after calling
folio_trylock() which will set PG_locked bit, ntfs_truncate() must not
call truncate_setsize() which will wait for PG_locked bit to be cleared
when holding ni_lock lock.
ipu_src_rect_width() was introduced to support odd screen resolutions
such as 1366x768 by internally rounding up primary plane width to a
multiple of 8 and compensating with reduced horizontal blanking.
This also caused overlay plane width to be rounded up, which was not
intended. Fix overlay plane width by limiting the rounding up to the
primary plane.
drm_rect_width(&new_state->src) >> 16 is the same value as
drm_rect_width(dst) because there is no plane scaling support.
When adding/deleting large number of elements in one step in ipset, it can
take a reasonable amount of time and can result in soft lockup errors. The
patch 5f7b51bf09ba ("netfilter: ipset: Limit the maximal range of
consecutive elements to add/delete") tried to fix it by limiting the max
elements to process at all. However it was not enough, it is still possible
that we get hung tasks. Lowering the limit is not reasonable, so the
approach in this patch is as follows: rely on the method used at resizing
sets and save the state when we reach a smaller internal batch limit,
unlock/lock and proceed from the saved state. Thus we can avoid long
continuous tasks and at the same time removed the limit to add/delete large
number of elements in one step.
The nfnl mutex is held during the whole operation which prevents one to
issue other ipset commands in parallel.
Fixes: 5f7b51bf09ba ("netfilter: ipset: Limit the maximal range of consecutive elements to add/delete") Reported-by: syzbot+9204e7399656300bf271@syzkaller.appspotmail.com Signed-off-by: Jozsef Kadlecsik <kadlec@netfilter.org> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The hash:net,port,net set type supports /0 subnets. However, the patch
commit 5f7b51bf09baca8e titled "netfilter: ipset: Limit the maximal range
of consecutive elements to add/delete" did not take into account it and
resulted in an endless loop. The bug is actually older but the patch 5f7b51bf09baca8e brings it out earlier.
Handle /0 subnets properly in hash:net,port,net set types.
Fixes: 5f7b51bf09ba ("netfilter: ipset: Limit the maximal range of consecutive elements to add/delete") Reported-by: Марк Коренберг <socketpair@gmail.com> Signed-off-by: Jozsef Kadlecsik <kadlec@netfilter.org> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
There is an issue with the checking of the return value of
'of_get_mac_address', which returns 0 on success and negative value on
failure. The driver interpretated the result the opposite way. Therefore
if there was a MAC address defined in the DT, then the driver was
generating a random MAC address otherwise it would use address 0.
Fix this by checking correctly the return value of 'of_get_mac_address'
Fixes: b74ef9f9cb91 ("net: sparx5: Do not use mac_addr uninitialized in mchp_sparx5_probe()") Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
If asked to drop a packet via TC_ACT_SHOT it is unsafe to assume
res.class contains a valid pointer Fixes: b0188d4dbe5f ("[NET_SCHED]: sch_atm: Lindent") Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
of_irq_find_parent() returns a node pointer with refcount incremented,
We should use of_node_put() on it when not needed anymore.
Add missing of_node_put() to avoid refcount leak.
For the POSIX locks they are using the same owner, which is the
thread id. And multiple POSIX locks could be merged into single one,
so when checking whether the 'file' has locks may fail.
For a file where some openers use locking and others don't is a
really odd usage pattern though. Locks are like stoplights -- they
only work if everyone pays attention to them.
Just switch ceph_get_caps() to check whether any locks are set on
the inode. If there are POSIX/OFD/FLOCK locks on the file at the
time, we should set CHECK_FILELOCK, regardless of what fd was used
to set the lock.
Fixes: ff5d913dfc71 ("ceph: return -EIO if read/write against filp that lost file locks") Signed-off-by: Xiubo Li <xiubli@redhat.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Ilya Dryomov <idryomov@gmail.com> Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Ceph has a need to know whether a particular inode has any locks set on
it. It's currently tracking that by a num_locks field in its
filp->private_data, but that's problematic as it tries to decrement this
field when releasing locks and that can race with the file being torn
down.
Add a new vfs_inode_has_locks helper that just returns whether any locks
are currently held on the inode.
Reviewed-by: Xiubo Li <xiubli@redhat.com> Reviewed-by: Christoph Hellwig <hch@infradead.org> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Stable-dep-of: 461ab10ef7e6 ("ceph: switch to vfs_inode_has_locks() to fix file lock bug") Signed-off-by: Sasha Levin <sashal@kernel.org>
Having a bigger number of FIFO lines held after vsync is only useful to
SoCs using AFBC to give time to the AFBC decoder to be reset, configured
and enabled again.
For SoCs not using AFBC this, on the contrary, is causing on some
displays issues and a few pixels vertical offset in the displayed image.
Conditionally increase the number of lines held after vsync only for
SoCs using AFBC, leaving the default value for all the others.
Fixes: 24e0d4058eff ("drm/meson: hold 32 lines after vsync to give time for AFBC start") Signed-off-by: Carlo Caione <ccaione@baylibre.com> Acked-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com> Acked-by: Neil Armstrong <neil.armstrong@linaro.org>
[narmstrong: added fixes tag] Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org> Link: https://patchwork.freedesktop.org/patch/msgid/20221216-afbc_s905x-v1-0-033bebf780d9@baylibre.com Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently, when modifying DC, we validate max_rd_atomic user attribute
against the RC cap, validate against DC. RC and DC QP types have different
device limitations.
This can cause userspace created DC QPs to malfunction.
Currently, when mlx5_ib_get_hw_stats() is used for device (port_num = 0),
there is a special handling in order to use the correct counters, but,
port_num is being passed down the stack without any change. Also, some
functions assume that port_num >=1. As a result, the following oops can
occur.
of_phy_find_device() return device node with refcount incremented.
Call put_device() to relese it when not needed anymore.
Fixes: ab4e6ee578e8 ("net: phy: xgmiitorgmii: Check phy_driver ready before accessing") Signed-off-by: Miaoqian Lin <linmq006@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The device supports a PCIe optimization hint, which indicates on
which NUMA the queue is currently processed. This hint is utilized
by PCIe in order to reduce its access time by accessing the
correct NUMA resources and maintaining cache coherence.
The driver calls the register update for the hint (called TPH -
TLP Processing Hint) during the NAPI loop.
Though the update is expected upon a NUMA change (when a queue
is moved from one NUMA to the other), the current logic performs
a register update when the queue is moved to a different CPU,
but the CPU is not necessarily in a different NUMA.
The changes include:
1. Performing the TPH update only when the queue has switched
a NUMA node.
2. Moving the TPH update call to be triggered only when NAPI was
scheduled from interrupt context, as opposed to a busy-polling loop.
This is due to the fact that during busy-polling, the frequency
of CPU switches for a particular queue is significantly higher,
thus, the likelihood to switch NUMA is much higher. Therefore,
providing the frequent updates to the device upon a NUMA update
are unlikely to be beneficial.
Fixes: 1738cd3ed342 ("net: ena: Add a driver for Amazon Elastic Network Adapters (ENA)") Signed-off-by: David Arinzon <darinzon@amazon.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
RX ring can be NULL in XDP use cases where only TX queues
are configured. In this scenario, the RX interrupt moderation
value sent to the device remains in its default value of 0.
In this change, setting the default value of the RX interrupt
moderation to be the same as of the TX.
Fixes: 548c4940b9f1 ("net: ena: Implement XDP_TX action") Signed-off-by: David Arinzon <darinzon@amazon.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Make the upper bound on rx_copybreak tighter, by
making sure it is smaller than the minimum of mtu and
ENA_PAGE_SIZE. With the current upper bound of mtu,
rx_copybreak can be larger than a page. Such large
rx_copybreak will not bring any performance benefit to
the user and therefore makes no sense.
In addition, the value update was only reflected in
the adapter structure, but not applied for each ring,
causing it to not take effect.
Fixes: 1738cd3ed342 ("net: ena: Add a driver for Amazon Elastic Network Adapters (ENA)") Signed-off-by: Osama Abboud <osamaabb@amazon.com> Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com> Signed-off-by: David Arinzon <darinzon@amazon.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Redirecting packets with XDP Redirect is done in two phases:
1. A packet is passed by the driver to the kernel using
xdp_do_redirect().
2. After finishing polling for new packets the driver lets the kernel
know that it can now process the redirected packet using
xdp_do_flush_map().
The packets' redirection is handled in the napi context of the
queue that called xdp_do_redirect()
To avoid calling xdp_do_flush_map() each time the driver first checks
whether any packets were redirected, using
xdp_flags |= xdp_verdict;
and
if (xdp_flags & XDP_REDIRECT)
xdp_do_flush_map()
essentially treating XDP instructions as a bitmask, which isn't the case:
enum xdp_action {
XDP_ABORTED = 0,
XDP_DROP,
XDP_PASS,
XDP_TX,
XDP_REDIRECT,
};
Given the current possible values of xdp_action, the current design
doesn't have a bug (since XDP_REDIRECT = 100b), but it is still
flawed.
This patch makes the driver use a bitmask instead, to avoid future
issues.
Fixes: a318c70ad152 ("net: ena: introduce XDP redirect implementation") Signed-off-by: Shay Agroskin <shayagr@amazon.com> Signed-off-by: David Arinzon <darinzon@amazon.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
On driver initialization, RSS hash initial value is set to zero,
instead of the default value. This happens because we pass NULL as
the RSS key parameter, which caused us to never initialize
the RSS hash value.
This patch fixes it by making sure the initial value is set, no matter
what the value of the RSS key is.
Fixes: 91a65b7d3ed8 ("net: ena: fix potential crash when rxfh key is NULL") Signed-off-by: Nati Koler <nkoler@amazon.com> Signed-off-by: David Arinzon <darinzon@amazon.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The driver does not call tasklet_kill in several places.
Add the calls to fix it.
Fixes: 85b85c853401 ("amd-xgbe: Re-issue interrupt if interrupt status not cleared") Signed-off-by: Jiguang Xiao <jiguang.xiao@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Current xdp xmit functions logic (mlx5e_xmit_xdp_frame_mpwqe or
mlx5e_xmit_xdp_frame), validates xdp packet length by comparing it to
hw mtu (configured at xdp sq allocation) before xmiting it. This check
does not account for ethernet fcs length (calculated and filled by the
nic). Hence, when we try sending packets with length > (hw-mtu -
ethernet-fcs-size), the device port drops it and tx_errors_phy is
incremented. Desired behavior is to catch these packets and drop them
by the driver.
Fix this behavior in XDP SQ allocation function (mlx5e_alloc_xdpsq) by
subtracting ethernet FCS header size (4 Bytes) from current hw mtu
value, since ethernet FCS is calculated and written to ethernet frames
by the nic.
The cited commit introduced a bug for multiple encapsulations flow.
If one dest encap becomes invalid, the flow is set slow path flag.
But when other dests encap become invalid, they are not cleared due
to slow path flag of the flow. When neigh-update-add is running, it
will use invalid encap.
Fix it by checking slow path flag after clearing dest encap.
Fixes: 9a5f9cc794e1 ("net/mlx5e: Fix possible use-after-free deleting fdb rule") Signed-off-by: Chris Mi <cmi@nvidia.com> Reviewed-by: Roi Dayan <roid@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
In later commit we are going to instantiate multiple attr instances
for flow instead of single attr.
Make sure mlx5e_tc_add_flow_mod_hdr() use the correct attr and not flow->attr.
Signed-off-by: Roi Dayan <roid@nvidia.com> Reviewed-by: Oz Shlomo <ozsh@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Stable-dep-of: 2951b2e142ec ("net/mlx5e: Always clear dest encap in neigh-update-del") Signed-off-by: Sasha Levin <sashal@kernel.org>
mlx5e_build_nic_params will turn CQE compression on if the hardware
capability is enabled and the slow_pci_heuristic condition is detected.
As IPoIB doesn't support CQE compression, make sure to disable the
feature in the IPoIB profile init.
Please note that the feature is not exposed to the user for IPoIB
interfaces, so it can't be subsequently turned on.
Fixes: b797a684b0dd ("net/mlx5e: Enable CQE compression when PCI is slower than link") Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com> Reviewed-by: Gal Pressman <gal@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently, recovery is done without considering whether the device is
still in probe flow.
This may lead to recovery before device have finished probed
successfully. e.g.: while mlx5_init_one() is running. Recovery flow is
using functionality that is loaded only by mlx5_init_one(), and there
is no point in running recovery without mlx5_init_one() finished
successfully.
Fix it by waiting for probe flow to finish and checking whether the
device is probed before trying to perform recovery.
Fix SRIOV VST mode behavior to insert cvlan when a guest tag is already
present in the frame. Previous VST mode behavior was to drop packets or
override existing tag, depending on the device version.
In this patch we fix this behavior by correctly building the HW steering
rule with a push vlan action, or for older devices we ask the FW to stack
the vlan when a vlan is already present.
When we initialize vringh, we should pass the features and the
number of elements in the virtqueue negotiated with the driver,
otherwise operations with vringh may fail.
This was discovered in a case where the driver sets a number of
elements in the virtqueue different from the value returned by
.get_vq_num_max().
In vdpasim_vq_reset() is safe to initialize the vringh with
default values, since the virtqueue will not be used until
vdpasim_queue_ready() is called again.
Fixes: 2c53d0f64c06 ("vdpasim: vDPA device simulator") Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20221110141335.62171-1-sgarzare@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Jason Wang <jasowang@redhat.com> Acked-by: Eugenio Pérez <eperezma@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
vhost_iotlb_itree_first() requires `start` and `last` parameters
to search for a mapping that overlaps the range.
In translate_desc() we cyclically call vhost_iotlb_itree_first(),
incrementing `addr` by the amount already translated, so rightly
we move the `start` parameter passed to vhost_iotlb_itree_first(),
but we should hold the `last` parameter constant.
Let's fix it by saving the `last` parameter value before incrementing
`addr` in the loop.
Fixes: a9709d6874d5 ("vhost: convert pre sorted vhost memory array to interval tree") Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20221109102503.18816-3-sgarzare@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
vhost_iotlb_itree_first() requires `start` and `last` parameters
to search for a mapping that overlaps the range.
In iotlb_translate() we cyclically call vhost_iotlb_itree_first(),
incrementing `addr` by the amount already translated, so rightly
we move the `start` parameter passed to vhost_iotlb_itree_first(),
but we should hold the `last` parameter constant.
Let's fix it by saving the `last` parameter value before incrementing
`addr` in the loop.
Fixes: 9ad9c49cfe97 ("vringh: IOTLB support") Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20221109102503.18816-2-sgarzare@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
A problem about modprobe vhost_vsock failed is triggered with the
following log given:
modprobe: ERROR: could not insert 'vhost_vsock': Device or resource busy
The reason is that vhost_vsock_init() returns misc_register() directly
without checking its return value, if misc_register() failed, it returns
without calling vsock_core_unregister() on vhost_transport, resulting the
vhost_vsock can never be installed later.
A simple call graph is shown as below:
vhost_vsock_init()
vsock_core_register() # register vhost_transport
misc_register()
device_create_with_groups()
device_create_groups_vargs()
dev = kzalloc(...) # OOM happened
# return without unregister vhost_transport
Fix by calling vsock_core_unregister() when misc_register() returns error.
Fixes: 433fc58e6bf2 ("VSOCK: Introduce vhost_vsock.ko") Signed-off-by: Yuan Can <yuancan@huawei.com>
Message-Id: <20221108101705.45981-1-yuancan@huawei.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Stefano Garzarella <sgarzare@redhat.com> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Inject fault while probing module, if device_register() fails in
vdpasim_net_init() or vdpasim_blk_init(), but the refcount of kobject is
not decreased to 0, the name allocated in dev_set_name() is leaked.
Fix this by calling put_device(), so that name can be freed in
callback function kobject_cleanup().
Fixes: 899c4d187f6a ("vdpa_sim_blk: add support for vdpa management tool") Fixes: a3c06ae158dd ("vdpa_sim_net: Add support for user supported devices") Signed-off-by: ruanjinjie <ruanjinjie@huawei.com> Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20221110082348.4105476-1-ruanjinjie@huawei.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
nfc_get_device() take reference for the device, add missing
nfc_put_device() to release it when not need anymore.
Also fix the style warnning by use error EOPNOTSUPP instead of
ENOTSUPP.
Fixes: 5ce3f32b5264 ("NFC: netlink: SE API implementation") Fixes: 29e76924cf08 ("nfc: netlink: Add capability to reply to vendor_cmd with data") Signed-off-by: Miaoqian Lin <linmq006@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
PTP hardware timestamping related objects are not linked when PTP
support for MV88E6xxx (NET_DSA_MV88E6XXX_PTP) is disabled, therefore
NET_DSA_MV88E6XXX should not depend on PTP_1588_CLOCK_OPTIONAL
regardless of NET_DSA_MV88E6XXX_PTP.
Instead, condition more strictly on how NET_DSA_MV88E6XXX_PTP's
dependencies are met, making sure that it cannot be enabled when
NET_DSA_MV88E6XXX=y and PTP_1588_CLOCK=m.
In other words, this commit allows NET_DSA_MV88E6XXX to be built-in
while PTP_1588_CLOCK is a module, as long as NET_DSA_MV88E6XXX_PTP is
prevented from being enabled.
Fixes: e5f31552674e ("ethernet: fix PTP_1588_CLOCK dependencies") Signed-off-by: Johnny S. Lee <foss@jsl.io> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
adapter->dcb would get silently freed inside qlcnic_dcb_enable() in
case qlcnic_dcb_attach() would return an error, which always happens
under OOM conditions. This would lead to use-after-free because both
of the existing callers invoke qlcnic_dcb_get_info() on the obtained
pointer, which is potentially freed at that point.
Propagate errors from qlcnic_dcb_enable(), and instead free the dcb
pointer at callsite using qlcnic_dcb_free(). This also removes the now
unused qlcnic_clear_dcb_ops() helper, which was a simple wrapper around
kfree() also causing memory leaks for partially initialized dcb.
Found by Linux Verification Center (linuxtesting.org) with the SVACE
static analysis tool.
Fixes: 3c44bba1d270 ("qlcnic: Disable DCB operations from SR-IOV VFs") Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Signed-off-by: Daniil Tatianin <d-tatianin@yandex-team.ru> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Kernel uses tcindex_change() to change an existing
filter properties.
Yet the problem is that, during the process of changing,
if `old_r` is retrieved from `p->perfect`, then
kernel uses tcindex_alloc_perfect_hash() to newly
allocate filter results, uses tcindex_filter_result_init()
to clear the old filter result, without destroying
its tcf_exts structure, which triggers the above memory leak.
To be more specific, there are only two source for the `old_r`,
according to the tcindex_lookup(). `old_r` is retrieved from
`p->perfect`, or `old_r` is retrieved from `p->h`.
* If `old_r` is retrieved from `p->perfect`, kernel uses
tcindex_alloc_perfect_hash() to newly allocate the
filter results. Then `r` is assigned with `cp->perfect + handle`,
which is newly allocated. So condition `old_r && old_r != r` is
true in this situation, and kernel uses tcindex_filter_result_init()
to clear the old filter result, without destroying
its tcf_exts structure
* If `old_r` is retrieved from `p->h`, then `p->perfect` is NULL
according to the tcindex_lookup(). Considering that `cp->h`
is directly copied from `p->h` and `p->perfect` is NULL,
`r` is assigned with `tcindex_lookup(cp, handle)`, whose value
should be the same as `old_r`, so condition `old_r && old_r != r`
is false in this situation, kernel ignores using
tcindex_filter_result_init() to clear the old filter result.
So only when `old_r` is retrieved from `p->perfect` does kernel use
tcindex_filter_result_init() to clear the old filter result, which
triggers the above memory leak.
Considering that there already exists a tc_filter_wq workqueue
to destroy the old tcindex_data by tcindex_partial_destroy_work()
at the end of tcindex_set_parms(), this patch solves
this memory leak bug by removing this old filter result
clearing part and delegating it to the tc_filter_wq workqueue.
Note that this patch doesn't introduce any other issues. If
`old_r` is retrieved from `p->perfect`, this patch just
delegates old filter result clearing part to the
tc_filter_wq workqueue; If `old_r` is retrieved from `p->h`,
kernel doesn't reach the old filter result clearing part, so
removing this part has no effect.
[Thanks to the suggestion from Jakub Kicinski, Cong Wang, Paolo Abeni
and Dmitry Vyukov]
Fixes: b9a24bb76bf6 ("net_sched: properly handle failure case of tcf_exts_init()") Link: https://lore.kernel.org/all/0000000000001de5c505ebc9ec59@google.com/ Reported-by: syzbot+232ebdbd36706c965ebf@syzkaller.appspotmail.com Tested-by: syzbot+232ebdbd36706c965ebf@syzkaller.appspotmail.com Cc: Cong Wang <cong.wang@bytedance.com> Cc: Jakub Kicinski <kuba@kernel.org> Cc: Paolo Abeni <pabeni@redhat.com> Cc: Dmitry Vyukov <dvyukov@google.com> Acked-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Hawkins Jiawei <yin31149@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently, it missed set HCLGE_VPORT_STATE_PROMISC_CHANGE
flag for VF when vport->overflow_promisc_flags changed.
So the VF won't check whether to update promisc mode in
this case. So add it.
Fixes: 1e6e76101fd9 ("net: hns3: configure promisc mode for VF asynchronously") Signed-off-by: Jian Shen <shenjian15@huawei.com> Signed-off-by: Hao Lan <lanhao@huawei.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
For device supports RXD advanced layout, the driver will
return directly if the hardware finish the checksum
calculate. It cause missing L3E checking for ip packets.
Fixes it.
Fixes: 1ddc028ac849 ("net: hns3: refactor out RX completion checksum") Signed-off-by: Jian Shen <shenjian15@huawei.com> Signed-off-by: Hao Lan <lanhao@huawei.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
As the code to update ring stats is alike for different ring stats
type, this patch extract macro to simplify ring stats update code.
Signed-off-by: Peng Li <lipeng321@huawei.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Stable-dep-of: 7d89b53cea1a ("net: hns3: fix miss L3E checking for rx packet") Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently keep alive message between PF and VF may be lost and the VF is
unalive in PF. So the VF will not do reset during PF FLR reset process.
This would make the allocated interrupt resources of VF invalid and VF
would't receive or respond to PF any more.
So this patch adds VF interrupts re-initialization during VF FLR for VF
recovery in above cases.
Fixes: 862d969a3a4d ("net: hns3: do VF's pci re-initialization while PF doing FLR") Signed-off-by: Jie Wang <wangjie125@huawei.com> Signed-off-by: Hao Lan <lanhao@huawei.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently, we shut down the filecache before trying to clean up the
stateids that depend on it. This leads to the kernel trying to free an
nfsd_file twice, and a refcount overput on the nf_mark.
Change the shutdown procedure to tear down all of the stateids prior
to shutting down the filecache.
Reported-and-tested-by: Wang Yugui <wangyugui@e16-tech.com> Signed-off-by: Jeff Layton <jlayton@kernel.org> Fixes: 5e113224c17e ("nfsd: nfsd_file cache entries should be per net namespace") Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When AF_XDP is used on on a veth interface the RX ring is updated in two
steps. veth_xdp_rcv() removes packet descriptors from the FILL ring
fills them and places them in the RX ring updating the cached_prod
pointer. Later xdp_do_flush() syncs the RX ring prod pointer with the
cached_prod pointer allowing user-space to see the recently filled in
descriptors. The rings are intended to be SPSC, however the existing
order in veth_poll allows the xdp_do_flush() to run concurrently with
another CPU creating a race condition that allows user-space to see old
or uninitialized descriptors in the RX ring. This bug has been observed
in production systems.
To summarize, we are expecting this ordering:
CPU 0 __xsk_rcv_zc()
CPU 0 __xsk_map_flush()
CPU 2 __xsk_rcv_zc()
CPU 2 __xsk_map_flush()
But we are seeing this order:
CPU 0 __xsk_rcv_zc()
CPU 2 __xsk_rcv_zc()
CPU 0 __xsk_map_flush()
CPU 2 __xsk_map_flush()
This occurs because we rely on NAPI to ensure that only one napi_poll
handler is running at a time for the given veth receive queue.
napi_schedule_prep() will prevent multiple instances from getting
scheduled. However calling napi_complete_done() signals that this
napi_poll is complete and allows subsequent calls to
napi_schedule_prep() and __napi_schedule() to succeed in scheduling a
concurrent napi_poll before the xdp_do_flush() has been called. For the
veth driver a concurrent call to napi_schedule_prep() and
__napi_schedule() can occur on a different CPU because the veth xmit
path can additionally schedule a napi_poll creating the race.
The fix as suggested by Magnus Karlsson, is to simply move the
xdp_do_flush() call before napi_complete_done(). This syncs the
producer ring pointers before another instance of napi_poll can be
scheduled on another CPU. It will also slightly improve performance by
moving the flush closer to when the descriptors were placed in the
RX ring.
Fixes: d1396004dd86 ("veth: Add XDP TX and REDIRECT") Suggested-by: Magnus Karlsson <magnus.karlsson@gmail.com> Signed-off-by: Shawn Bohrer <sbohrer@cloudflare.com> Link: https://lore.kernel.org/r/20221220185903.1105011-1-sbohrer@cloudflare.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Set timeout and garbage collection interval updates are ignored on
updates. Add transaction to update global set element timeout and
garbage collection interval.
Commit dacce2be3312 ("vmxnet3: add geneve and vxlan tunnel offload
support") added support for encapsulation offload. However, the
pathc did not report correctly the csum_level for encapsulated packet.
This patch fixes this issue by reporting correct csum level for the
encapsulated packet.
Fixes: dacce2be3312 ("vmxnet3: add geneve and vxlan tunnel offload support") Signed-off-by: Ronak Doshi <doshir@vmware.com> Acked-by: Peng Li <lpeng@vmware.com> Link: https://lore.kernel.org/r/20221220202556.24421-1-doshir@vmware.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
If a ruleset declares a set name that matches an existing set in the
kernel, then validate that this declaration really refers to the same
set, otherwise bail out with EEXIST.
Currently, the kernel reports success when adding a set that already
exists in the kernel. This usually results in EINVAL errors at a later
stage, when the user adds elements to the set, if the set declaration
mismatches the existing set representation in the kernel.
Add a new function to check that the set declaration really refers to
the same existing set in the kernel.
panfrost_gem_create_with_handle() previously returned a BO but with the
only reference being from the handle, which user space could in theory
guess and release, causing a use-after-free. Additionally if the call to
panfrost_gem_mapping_get() in panfrost_ioctl_create_bo() failed then
a(nother) reference on the BO was dropped.
The _create_with_handle() is a problematic pattern, so ditch it and
instead create the handle in panfrost_ioctl_create_bo(). If the call to
panfrost_gem_mapping_get() fails then this means that user space has
indeed gone behind our back and freed the handle. In which case just
return an error code.
Anand hit a BUG() when pulling off headers on egress to a SW tunnel.
We get to skb_checksum_help() with an invalid checksum offset
(commit d7ea0d9df2a6 ("net: remove two BUG() from skb_checksum_help()")
converted those BUGs to WARN_ONs()).
He points out oddness in how skb_postpull_rcsum() gets used.
Indeed looks like we should pull before "postpull", otherwise
the CHECKSUM_PARTIAL fixup from skb_postpull_rcsum() will not
be able to do its job:
Fixes: 6702ed490ca0 ("Btrfs: Add run time btree defrag, and an ioctl to force btree defrag") Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 9130b8dbc6ac ("SUNRPC: allow for upcalls for the same uid
but different gss service") introduced `auth` argument to
__gss_find_upcall(), but in gss_pipe_downcall() it was left as NULL
since it (and auth->service) was not (yet) determined.
When multiple upcalls with the same uid and different service are
ongoing, it could happen that __gss_find_upcall(), which returns the
first match found in the pipe->in_downcall list, could not find the
correct gss_msg corresponding to the downcall we are looking for.
Moreover, it might return a msg which is not sent to rpc.gssd yet.
We could see mount.nfs process hung in D state with multiple mount.nfs
are executed in parallel. The call trace below is of CentOS 7.9
kernel-3.10.0-1160.24.1.el7.x86_64 but we observed the same hang w/
elrepo kernel-ml-6.0.7-1.el7.
The scenario is like this. Let's say there are two upcalls for
services A and B, A -> B in pipe->in_downcall, B -> A in pipe->pipe.
When rpc.gssd reads pipe to get the upcall msg corresponding to
service B from pipe->pipe and then writes the response, in
gss_pipe_downcall the msg corresponding to service A will be picked
because only uid is used to find the msg and it is before the one for
B in pipe->in_downcall. And the process waiting for the msg
corresponding to service A will be woken up.
Actual scheduing of that process might be after rpc.gssd processes the
next msg. In rpc_pipe_generic_upcall it clears msg->errno (for A).
The process is scheduled to see gss_msg->ctx == NULL and
gss_msg->msg.errno == 0, therefore it cannot break the loop in
gss_create_upcall and is never woken up after that.
This patch adds a simple check to ensure that a msg which is not
sent to rpc.gssd yet is not chosen as the matching upcall upon
receiving a downcall.
Signed-off-by: minoura makoto <minoura@valinux.co.jp> Signed-off-by: Hiroshi Shimamoto <h-shimamoto@nec.com> Tested-by: Hiroshi Shimamoto <h-shimamoto@nec.com> Cc: Trond Myklebust <trondmy@hammerspace.com> Fixes: 9130b8dbc6ac ("SUNRPC: allow for upcalls for same uid but different gss service") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The scratch page might not be allocated in LMEM(like on DG2), so instead
of using that as the deciding factor for where the paging structures
live, let's just query the pt before mapping it.
When manipulating xattr blocks, we can deadlock infinitely looping
inside ext4_xattr_block_set() where we constantly keep finding xattr
block for reuse in mbcache but we are unable to reuse it because its
reference count is too big. This happens because cache entry for the
xattr block is marked as reusable (e_reusable set) although its
reference count is too big. When this inconsistency happens, this
inconsistent state is kept indefinitely and so ext4_xattr_block_set()
keeps retrying indefinitely.
The inconsistent state is caused by non-atomic update of e_reusable bit.
e_reusable is part of a bitfield and e_reusable update can race with
update of e_referenced bit in the same bitfield resulting in loss of one
of the updates. Fix the problem by using atomic bitops instead.
This bug has been around for many years, but it became *much* easier
to hit after commit 65f8b80053a1 ("ext4: fix race when reusing xattr
blocks").
Cc: stable@vger.kernel.org Fixes: 6048c64b2609 ("mbcache: add reusable flag to cache entries") Fixes: 65f8b80053a1 ("ext4: fix race when reusing xattr blocks") Reported-and-tested-by: Jeremi Piotrowski <jpiotrowski@linux.microsoft.com> Reported-by: Thilo Fromm <t-lo@linux.microsoft.com> Link: https://lore.kernel.org/r/c77bf00f-4618-7149-56f1-b8d1664b9d07@linux.microsoft.com/ Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Andreas Dilger <adilger@dilger.ca> Link: https://lore.kernel.org/r/20221123193950.16758-1-jack@suse.cz Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Sasha Levin <sashal@kernel.org>
Use the fact that entries with elevated refcount are not removed from
the hash and just move removal of the entry from the hash to the entry
freeing time. When doing this we also change the generic code to hold
one reference to the cache entry, not two of them, which makes code
somewhat more obvious.
When we used the journal_async_commit mounting option in nojournal mode,
the kernel told me that "can't mount with journal_checksum", was very
confusing. I find that when we mount with journal_async_commit, both the
JOURNAL_ASYNC_COMMIT and EXPLICIT_JOURNAL_CHECKSUM flags are set. However,
in the error branch, CHECKSUM is checked before ASYNC_COMMIT. As a result,
the above inconsistency occurs, and the ASYNC_COMMIT branch becomes dead
code that cannot be executed. Therefore, we exchange the positions of the
two judgments to make the error msg more accurate.
Before these two branches neither loaded the journal nor created the
xattr cache. So the right label to goto is 'failed_mount3a'. Although
this did not cause any issues because the error handler validated if the
pointer is null. However this still made me confused when reading
the code. So it's still worth to modify to goto the right label.
Signed-off-by: Jason Yan <yanaijie@huawei.com> Reviewed-by: Jan Kara <jack@suse.cz> Reviewed-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com> Link: https://lore.kernel.org/r/20220916141527.1012715-2-yanaijie@huawei.com Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Stable-dep-of: 89481b5fa8c0 ("ext4: correct inconsistent error msg in nojournal mode") Signed-off-by: Sasha Levin <sashal@kernel.org>
Consider the call to unregister_netdev()
unregister_netdev->unregister_netdevice_queue->rollback_registered_many
that calls the below functions which access the registers after
pm_runtime_put_sync()
1) ravb_get_stats
2) ravb_close
Fixes: c156633f1353 ("Renesas Ethernet AVB driver proper") Cc: stable@vger.kernel.org Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Link: https://lore.kernel.org/r/20221214105118.2495313-1-biju.das.jz@bp.renesas.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Any debugging information entry representing the declaration of an object,
module, subprogram or type may have DW_AT_decl_file, DW_AT_decl_line and
DW_AT_decl_column attributes, each of whose value is an unsigned integer
constant.
So it should be an unsigned integer data. Also, even though the standard
doesn't clearly say the DW_AT_call_file is signed or unsigned, the
elfutils (eu-readelf) interprets it as unsigned integer data and it is
natural to handle it as unsigned integer data as same as DW_AT_decl_file.
This changes the DW_AT_call_file as unsigned integer data too.
Fixes: 3f4460a28fb2f73d ("perf probe: Filter out redundant inline-instances") Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: stable@vger.kernel.org Cc: Steven Rostedt (VMware) <rostedt@goodmis.org> Link: https://lore.kernel.org/r/166761727445.480106.3738447577082071942.stgit@devnote3 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Use dwarf_attr_integrate() instead of dwarf_attr() for generic attribute
acccessor functions, so that it can find the specified attribute from
abstact origin DIE etc.
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Steven Rostedt (VMware) <rostedt@goodmis.org> Link: https://lore.kernel.org/r/166731051988.2100653.13595339994343449770.stgit@devnote3 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Stable-dep-of: a9dfc46c67b5 ("perf probe: Fix to get the DW_AT_decl_file and DW_AT_call_file as unsinged data") Signed-off-by: Sasha Levin <sashal@kernel.org>
During error on CLOSE_INSTANCE command, ctx_work_bits was not getting
cleared. During consequent mfc execution NULL pointer dereferencing of
this context led to kernel panic. This patch fixes this issue by making
sure to clear ctx_work_bits always.
On receiving last buffer driver puts MFC to MFCINST_FINISHING state which
in turn skips transferring of frame from SRC to REF queue. This causes
driver to stop MFC encoding and last frame is lost.
This patch guarantees safe handling of frames during MFCINST_FINISHING and
correct clearing of workbit to avoid early stopping of encoding.
AMD's MCA Thresholding feature counts errors of all severity levels, not
just correctable errors. If a deferred error causes the threshold limit
to be reached (it was the error that caused the overflow), then both a
deferred error interrupt and a thresholding interrupt will be triggered.
The order of the interrupts is not guaranteed. If the threshold
interrupt handler is executed first, then it will clear MCA_STATUS for
the error. It will not check or clear MCA_DESTAT which also holds a copy
of the deferred error. When the deferred error interrupt handler runs it
will not find an error in MCA_STATUS, but it will find the error in
MCA_DESTAT. This will cause two errors to be logged.
Check for deferred errors when handling a threshold interrupt. If a bank
contains a deferred error, then clear the bank's MCA_DESTAT register.
Define a new helper function to do the deferred error check and clearing
of MCA_DESTAT.
[ bp: Simplify, convert comment to passive voice. ]
Store the error code before freeing the extent_map. Though it's
reference counted structure, in that function it's the first and last
allocation so this would lead to a potential use-after-free.
The error can happen eg. when chunk is stored on a missing device and
the degraded mount option is missing.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=216721 Reported-by: eriri <1527030098@qq.com> Fixes: adfb69af7d8c ("btrfs: add_missing_dev() should return the actual error") CC: stable@vger.kernel.org # 4.9+ Signed-off-by: void0red <void0red@gmail.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
This simplifies the code flow in read_one_chunk and makes error handling
when handling missing devices a bit simpler by reducing it to a single
check if something went wrong. No functional changes.
Reviewed-by: Su Yue <l@damenly.su> Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Stable-dep-of: 1742e1c90c3d ("btrfs: fix extent map use-after-free when handling missing device in read_one_chunk") Signed-off-by: Sasha Levin <sashal@kernel.org>
The SM8250 only uses three clocks but the DP configuration erroneously
described four clocks.
In case the DP part of the PHY is initialised before the USB part, this
would lead to uninitialised memory beyond the bulk-clocks array to be
treated as a clock pointer as the clocks are requested based on the USB
configuration.
We want to ensure that the mask related to calling do_work_pending()
is within the first 16 bits. Move bits unrelated to that outside of
that range, to avoid spuriously calling do_work_pending() when we don't
need to.
Cc: stable@vger.kernel.org Fixes: 32d59773da38 ("arm: add support for TIF_NOTIFY_SIGNAL") Reported-and-tested-by: Hui Tang <tanghui20@huawei.com> Suggested-by: Russell King (Oracle) <linux@armlinux.org.uk> Link: https://lore.kernel.org/lkml/7ecb8f3c-2aeb-a905-0d4a-aa768b9649b5@huawei.com/ Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Due to several different off-by-one errors, or perhaps due to a late
change in design that wasn't fully reflected in the code that was
actually merged, there are several very strange constraints on how
fast-commit blocks are filled with tlv entries:
- tlvs must start at least 10 bytes before the end of the block, even
though the minimum tlv length is 8. Otherwise, the replay code will
ignore them. (BUG: ext4_fc_reserve_space() could violate this
requirement if called with a len of blocksize - 9 or blocksize - 8.
Fortunately, this doesn't seem to happen currently.)
- tlvs must end at least 1 byte before the end of the block. Otherwise
the replay code will consider them to be invalid. This quirk
contributed to a bug (fixed by an earlier commit) where uninitialized
memory was being leaked to disk in the last byte of blocks.
Also, strangely these constraints don't apply to the replay code in
e2fsprogs, which will accept any tlvs in the blocks (with no bounds
checks at all, but that is a separate issue...).
Given that this all seems to be a bug, let's fix it by just filling
blocks with tlv entries in the natural way.
Note that old kernels will be unable to replay fast-commit journals
created by kernels that have this commit.
As is done elsewhere in the file, build the struct ext4_fc_tl on the
stack and memcpy() it into the buffer, rather than directly writing it
to a potentially-unaligned location in the buffer.
Validate the inode and filename lengths in fast-commit journal records
so that a malicious fast-commit journal cannot cause a crash by having
invalid values for these. Also validate EXT4_FC_TAG_DEL_RANGE.
Commit a80f7fcf1867 ("ext4: fixup ext4_fc_track_* functions' signature")
extended the scope of the transaction in ext4_unlink() too far, making
it include the call to ext4_find_entry(). However, ext4_find_entry()
can deadlock when called from within a transaction because it may need
to set up the directory's encryption key.
Fix this by restoring the transaction to its original scope.
fast-commit of create, link, and unlink operations in encrypted
directories is completely broken because the unencrypted filenames are
being written to the fast-commit journal instead of the encrypted
filenames. These operations can't be replayed, as encryption keys
aren't present at journal replay time. It is also an information leak.
Until if/when we can get this working properly, make encrypted directory
operations ineligible for fast-commit.
Note that fast-commit operations on encrypted regular files continue to
be allowed, as they seem to work.
For scan loop must ensure that at least EXT4_FC_TAG_BASE_LEN space. If remain
space less than EXT4_FC_TAG_BASE_LEN which will lead to out of bound read
when mounting corrupt file system image.
ADD_RANGE/HEAD/TAIL is needed to add extra check when do journal scan, as this
three tags will read data during scan, tag length couldn't less than data length
which will read.
Cc: stable@kernel.org Signed-off-by: Ye Bin <yebin10@huawei.com> Link: https://lore.kernel.org/r/20220924075233.2315259-4-yebin10@huawei.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We use jbd_debug() in some places in ext4. It seems a bit strange to use
jbd2 debugging output function for ext4 code. Also these days
ext4_debug() uses dynamic printk so each debug message can be enabled /
disabled on its own so the time when it made some sense to have these
combined (to allow easier common selecting of messages to report) has
passed. Just convert all jbd_debug() uses in ext4 to ext4_debug().
Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Lukas Czerner <lczerner@redhat.com> Link: https://lore.kernel.org/r/20220608112355.4397-1-jack@suse.cz Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The maximum number of synthetic fields supported is defined as
SYNTH_FIELDS_MAX which value currently is 64, but it actually fails
when try to generate a synthetic event with 64 fields by executing like:
# echo "my_synth_event int v1; int v2; int v3; int v4; int v5; int v6;\
int v7; int v8; int v9; int v10; int v11; int v12; int v13; int v14;\
int v15; int v16; int v17; int v18; int v19; int v20; int v21; int v22;\
int v23; int v24; int v25; int v26; int v27; int v28; int v29; int v30;\
int v31; int v32; int v33; int v34; int v35; int v36; int v37; int v38;\
int v39; int v40; int v41; int v42; int v43; int v44; int v45; int v46;\
int v47; int v48; int v49; int v50; int v51; int v52; int v53; int v54;\
int v55; int v56; int v57; int v58; int v59; int v60; int v61; int v62;\
int v63; int v64" >> /sys/kernel/tracing/synthetic_events
Correct the field counting to fix it.
Link: https://lore.kernel.org/linux-trace-kernel/20221207091557.3137904-1-zhengyejian1@huawei.com Cc: <mhiramat@kernel.org> Cc: <zanussi@kernel.org> Cc: stable@vger.kernel.org Fixes: c9e759b1e845 ("tracing: Rework synthetic event command parsing") Signed-off-by: Zheng Yejian <zhengyejian1@huawei.com> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
[Fix conflict due to lack of c24be24aed405d64ebcf04526614c13b2adfb1d2] Signed-off-by: Zheng Yejian <zhengyejian1@huawei.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
dd_finish_request() tests if the per prio fifo_list is not empty to
determine if request dispatching must be restarted for handling blocked
write requests to zoned devices with a call to
blk_mq_sched_mark_restart_hctx(). While simple, this implementation has
2 problems:
1) Only the priority level of the completed request is considered.
However, writes to a zone may be blocked due to other writes to the
same zone using a different priority level. While this is unlikely to
happen in practice, as writing a zone with different IO priorirites
does not make sense, nothing in the code prevents this from
happening.
2) The use of list_empty() is dangerous as dd_finish_request() does not
take dd->lock and may run concurrently with the insert and dispatch
code.
Fix these 2 problems by testing the write fifo list of all priority
levels using the new helper dd_has_write_work(), and by testing each
fifo list using list_empty_careful().
Fixes: c807ab520fc3 ("block/mq-deadline: Add I/O priority support") Cc: <stable@vger.kernel.org> Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com> Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Link: https://lore.kernel.org/r/20221124021208.242541-2-damien.lemoal@opensource.wdc.com Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Only apply the static threshold for Stoney and Carrizo.
This hardware has certain requirements that don't allow
mixing of GTT and VRAM. Newer asics do not have these
requirements so we should be able to be more flexible
with where buffers end up.
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2270
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2291
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2255 Acked-by: Luben Tuikov <luben.tuikov@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Now, extended attribute value maximum length is 64K. The memory
requested here does not need continuous physical addresses, so it is
appropriate to use kvmalloc to request memory. At the same time, it
can also cope with the situation that the extended attribute will
become longer in the future.
Signed-off-by: Ye Bin <yebin10@huawei.com> Reviewed-by: Jan Kara <jack@suse.cz> Link: https://lore.kernel.org/r/20221208023233.1231330-3-yebin@huaweicloud.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> Cc: stable@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When expanding inode space in ext4_expand_extra_isize_ea() we may need
to allocate external xattr block. If quota is not initialized for the
inode, the block allocation will not be accounted into quota usage. Make
sure the quota is initialized before we try to expand inode space.
Make sure we initialize quotas before possibly expanding inode space
(and thus maybe needing to allocate external xattr block) in
ext4_ioctl_setproject(). This prevents not accounting the necessary
block allocation.
This occurs in 'ext4_xattr_inode_create()'. If 'ext4_mark_inode_dirty()'
fails, dropping i_nlink of the inode is needed. Or will lead to inode leak.
Signed-off-by: Ye Bin <yebin10@huawei.com> Reviewed-by: Jan Kara <jack@suse.cz> Link: https://lore.kernel.org/r/20221208023233.1231330-5-yebin@huaweicloud.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> Cc: stable@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Above issue may happens as follows:
ext4_da_write_begin
ext4_da_write_inline_data_begin
ext4_da_convert_inline_data_to_extent
ext4_clear_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA);
ext4_da_write_end
ext4_run_li_request
ext4_mb_prefetch
ext4_read_block_bitmap_nowait
ext4_validate_block_bitmap
ext4_mark_group_bitmap_corrupted(sb, block_group, EXT4_GROUP_INFO_BBITMAP_CORRUPT)
percpu_counter_sub(&sbi->s_freeclusters_counter,grp->bb_free);
-> sbi->s_freeclusters_counter become zero
ext4_da_write_begin
if (ext4_nonda_switch(inode->i_sb)) -> As freeclusters_counter is zero will return true
*fsdata = (void *)FALL_BACK_TO_NONDELALLOC;
ext4_write_begin
ext4_da_write_end
if (write_mode == FALL_BACK_TO_NONDELALLOC)
ext4_write_end
if (inline_data)
ext4_write_inline_data_end
ext4_write_inline_data
BUG_ON(pos + len > EXT4_I(inode)->i_inline_size);
-> As inode is already convert to extent, so 'pos + len' > inline_size
-> then trigger BUG.
To solve this issue, instead of checking ext4_has_inline_data() which
is only cleared after data has been written back, check the
EXT4_STATE_MAY_INLINE_DATA flag in ext4_write_end().
Fixes: f19d5870cbf7 ("ext4: add normal write support for inline data") Reported-by: syzbot+4faa160fa96bfba639f8@syzkaller.appspotmail.com Reported-by: Jun Nie <jun.nie@linaro.org> Signed-off-by: Ye Bin <yebin10@huawei.com> Link: https://lore.kernel.org/r/20221206144134.1919987-1-yebin@huaweicloud.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> Cc: stable@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit fb0a387dcdcd ("ext4: limit block allocations for indirect-block
files to < 2^32") added code to try to allocate xattr block with 32-bit
block number for indirect block based files on the grounds that these
files cannot use larger block numbers. It also added BUG_ON when
allocated block could not fit into 32 bits. This is however bogus
reasoning because xattr block is stored in inode->i_file_acl and
inode->i_file_acl_hi and as such even indirect block based files can
happily use full 48 bits for xattr block number. The proper handling
seems to be there basically since 64-bit block number support was added.
So remove the bogus limitation and BUG_ON.
Cc: Eric Sandeen <sandeen@redhat.com> Fixes: fb0a387dcdcd ("ext4: limit block allocations for indirect-block files to < 2^32") Signed-off-by: Jan Kara <jack@suse.cz> Link: https://lore.kernel.org/r/20221121130929.32031-1-jack@suse.cz Signed-off-by: Theodore Ts'o <tytso@mit.edu> Cc: stable@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If a block is out of range in ext4_get_branch(), -ENOMEM will be returned
to user-space. Obviously, this error code isn't really useful. This
patch fixes it by making sure the right error code (-EFSCORRUPTED) is
propagated to user-space. EUCLEAN is more informative than ENOMEM.
When a backup superblock is updated in update_backups(), the primary
superblock's offset in the group (that is, sbi->s_sbh->b_blocknr) is used
as the backup superblock's offset in its group. However, when the block
size is 1K and bigalloc is enabled, the two offsets are not equal. This
causes the backup group descriptors to be overwritten by the superblock
in update_backups(). Moreover, if meta_bg is enabled, the file system will
be corrupted because this feature uses backup group descriptors.
To solve this issue, we use a more accurate ext4_group_first_block_no() as
the offset of the backup superblock in its group.
Fixes: d77147ff443b ("ext4: add support for online resizing with bigalloc") Signed-off-by: Baokun Li <libaokun1@huawei.com> Reviewed-by: Jan Kara <jack@suse.cz> Cc: stable@kernel.org Link: https://lore.kernel.org/r/20221117040341.1380702-4-libaokun1@huawei.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
projects / linux.git / log