The xfrm{4,6}_beet_gso_segment() functions did not correctly set the
SKB_GSO_IPXIP4 and SKB_GSO_IPXIP6 gso types for the address family
tunneling case. Fix this by setting these gso types.
The maximum message size that can be send is bigger than
the maximum site that skb_page_frag_refill can allocate.
So it is possible to write beyond the allocated buffer.
Fix this by doing a fallback to COW in that case.
v2:
Avoid get get_order() costs as suggested by Linus Torvalds.
When vp_vdpa driver is unbind, vp_vdpa is freed in vdpa_unregister_device
and then vp_vdpa->mdev.pci_dev is dereferenced in vp_modern_remove,
triggering use-after-free.
Currently we have a BUG_ON() to make sure the number of sg
list does not exceed queue_max_segments() in virtio_queue_rq().
However, the block layer uses queue_max_discard_segments()
instead of queue_max_segments() to limit the sg list for
discard requests. So the BUG_ON() might be triggered if
virtio-blk device reports a larger value for max discard
segment than queue_max_segments(). To fix it, let's simply
remove the BUG_ON() which has become unnecessary after commit 02746e26c39e("virtio-blk: avoid preallocating big SGL for data").
And the unused vblk->sg_elems can also be removed together.
Fixes: 1f23816b8eb8 ("virtio_blk: add discard and write zeroes support") Suggested-by: Christoph Hellwig <hch@infradead.org> Signed-off-by: Xie Yongji <xieyongji@bytedance.com> Reviewed-by: Max Gurtovoy <mgurtovoy@nvidia.com> Link: https://lore.kernel.org/r/20220304100058.116-2-xieyongji@bytedance.com Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently the value of max_discard_segment will be set to
MAX_DISCARD_SEGMENTS (256) with no basis in hardware if device
set 0 to max_discard_seg in configuration space. It's incorrect
since the device might not be able to handle such large descriptors.
To fix it, let's follow max_segments restrictions in this case.
Fixes: 1f23816b8eb8 ("virtio_blk: add discard and write zeroes support") Signed-off-by: Xie Yongji <xieyongji@bytedance.com> Link: https://lore.kernel.org/r/20220304100058.116-1-xieyongji@bytedance.com Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
In vhost_iotlb_add_range_ctx(), range size can overflow to 0 when
start is 0 and last is ULONG_MAX. One instance where it can happen
is when userspace sends an IOTLB message with iova=size=uaddr=0
(vhost_process_iotlb_msg). So, an entry with size = 0, start = 0,
last = ULONG_MAX ends up in the iotlb. Next time a packet is sent,
iotlb_access_ok() loops indefinitely due to that erroneous entry.
Reported by syzbot at:
https://syzkaller.appspot.com/bug?extid=0abd373e2e50d704db87
To fix this, do two things:
1. Return -EINVAL in vhost_chr_write_iter() when userspace asks to map
a range with size 0.
2. Fix vhost_iotlb_add_range_ctx() to handle the range [0, ULONG_MAX]
by splitting it into two entries.
Fixes: 0bbe30668d89e ("vhost: factor out IOTLB") Reported-by: syzbot+0abd373e2e50d704db87@syzkaller.appspotmail.com Tested-by: syzbot+0abd373e2e50d704db87@syzkaller.appspotmail.com Signed-off-by: Anirudh Rayabharam <mail@anirudhrb.com> Link: https://lore.kernel.org/r/20220305095525.5145-1-mail@anirudhrb.com Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
dsp_pipeline_build() allocates dup pointer by kstrdup(cfg),
but then it updates dup variable by strsep(&dup, "|").
As a result when it calls kfree(dup), the dup variable contains NULL.
Found by Linux Driver Verification project (linuxtesting.org) with SVACE.
Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru> Fixes: 960366cf8dbb ("Add mISDN DSP") Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
This PHY doesn't support a link-up interrupt source. If aneg is enabled
we use the "aneg complete" interrupt for this purpose, but if aneg is
disabled link-up isn't signaled currently.
According to a vendor driver there's an additional "energy detect"
interrupt source that can be used to signal link-up if aneg is disabled.
We can safely ignore this interrupt source if aneg is enabled.
This patch was tested on a TX3 Mini TV box with S905W (even though
boot message says it's a S905D).
This issue has been existing longer, but due to changes in phylib and
the driver the patch applies only from the commit marked as fixed.
Fixes: 84c8f773d2dc ("net: phy: meson-gxl: remove the use of .ack_callback()") Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Link: https://lore.kernel.org/r/04cac530-ea1b-850e-6cfa-144a55c4d75d@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
When control vq receives a VIRTIO_NET_CTRL_MQ_VQ_PAIRS_SET command
request from the driver, presently there is no validation against the
number of queue pairs to configure, or even if multiqueue had been
negotiated or not is unverified. This may lead to kernel panic due to
uninitialized resource for the queues were there any bogus request
sent down by untrusted driver. Tie up the loose ends there.
Fixes: 52893733f2c5 ("vdpa/mlx5: Add multiqueue support") Signed-off-by: Si-Wei Liu <si-wei.liu@oracle.com> Link: https://lore.kernel.org/r/1642206481-30721-4-git-send-email-si-wei.liu@oracle.com Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Eli Cohen <elic@nvidia.com> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The node in question is receiving activate messages in another
thread after changing bearer status to allow message sending/
receiving in current thread:
The function alloc_workqueue() in nintendo_hid_probe() can fail, but
there is no check of its return value. To fix this bug, its return value
should be checked with new error handling code.
The driver creates the top row map sysfs attribute in input_configured()
method; unfortunately we do not have a callback that is executed when HID
interface is unbound, thus we are leaking these sysfs attributes, for
example when device is disconnected.
To fix it let's switch to managed version of adding sysfs attributes which
will ensure that they are destroyed when the driver is unbound.
In commit d687e056a18f ("soc: mediatek: mmsys: Add mt8192 mmsys routing table"),
the mmsys routing table for mt8192 was introduced but the input selector
for DITHER->DSI0 has no value assigned to it.
This means that we are clearing bit 0 instead of setting it, blocking
communication between these two blocks; due to that, any display that
is connected to DSI0 will not work, as no data will go through.
The effect of that issue is that, during bootup, the DRM will block for
some time, while atomically waiting for a vblank that never happens;
later, the situation doesn't get better, leaving the display in a
non-functional state.
To fix this issue, fix the route entry in the table by assigning the
dither input selector to MT8192_DISP_DSI0_SEL_IN.
Fixes: d687e056a18f ("soc: mediatek: mmsys: Add mt8192 mmsys routing table") Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Tested-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com> Reviewed-by: Nícolas F. R. A. Prado <nfraprado@collabora.com> Link: https://lore.kernel.org/r/20220128142056.359900-1-angelogioacchino.delregno@collabora.com Signed-off-by: Matthias Brugger <matthias.bgg@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
On SC7180 we observe black screens because the gdsc is being
enabled/disabled very rapidly and the GDSC FSM state does not work as
expected. This is due to the fact that the GDSC reset value is being
updated from SW.
The recommended transition delay for mdss core gdsc updated for
SC7180/SC7280/SM8250.
Fixes: dd3d06622138 ("clk: qcom: Add display clock controller driver for SC7180") Fixes: 1a00c962f9cd ("clk: qcom: Add display clock controller driver for SC7280") Fixes: 80a18f4a8567 ("clk: qcom: Add display clock controller driver for SM8150 and SM8250") Signed-off-by: Taniya Das <tdas@codeaurora.org> Link: https://lore.kernel.org/r/20220223185606.3941-2-tdas@codeaurora.org Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
[sboyd@kernel.org: lowercase hex] Signed-off-by: Stephen Boyd <sboyd@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
GDSCs have multiple transition delays which are used for the GDSC FSM
states. Older targets/designs required these values to be updated from
gdsc code to certain default values for the FSM state to work as
expected. But on the newer targets/designs the values updated from the
GDSC driver can hamper the FSM state to not work as expected.
On SC7180 we observe black screens because the gdsc is being
enabled/disabled very rapidly and the GDSC FSM state does not work as
expected. This is due to the fact that the GDSC reset value is being
updated from SW.
Thus add support to update the transition delay from the clock
controller gdscs as required.
Fixes: 45dd0e55317cc ("clk: qcom: Add support for GDSCs) Signed-off-by: Taniya Das <tdas@codeaurora.org> Link: https://lore.kernel.org/r/20220223185606.3941-1-tdas@codeaurora.org Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org> Signed-off-by: Stephen Boyd <sboyd@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
While the HVS has the same context memory size in the BCM2711 than in
the previous SoCs, the range allocated to the registers doubled and it
now takes 16k + 16k, compared to 8k + 16k before.
The KMS driver will use the whole context RAM though, eventually
resulting in a pointer dereference error when we access the higher half
of the context memory since it hasn't been mapped.
Fixes: 4564363351e2 ("ARM: dts: bcm2711: Enable the display pipeline") Signed-off-by: Maxime Ripard <maxime@cerno.tech> Signed-off-by: Stefan Wahren <stefan.wahren@i2se.com> Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Syzbot reported an slab-out-of-bounds Read in thrustmaster_probe() bug.
The root case is in missing validation check of actual number of endpoints.
Code should not blindly access usb_host_interface::endpoint array, since
it may contain less endpoints than code expects.
Fix it by adding missing validaion check and print an error if
number of endpoints do not match expected number
Fixes: c49c33637802 ("HID: support for initialization of some Thrustmaster wheels") Reported-and-tested-by: syzbot+35eebd505e97d315d01c@syzkaller.appspotmail.com Signed-off-by: Pavel Skripkin <paskripkin@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 817b8b9c539 ("HID: elo: fix memory leak in elo_probe") introduced
memory leak on error path, but more importantly the whole USB reference
counting is not needed at all in the first place, as the driver itself
doesn't change the reference counting in any way, and the associated
usb_device is guaranteed to be kept around by USB core as long as the
driver binding exists.
Reported-by: Alan Stern <stern@rowland.harvard.edu> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Fixes: fbf42729d0e ("HID: elo: update the reference count of the usb device structure") Fixes: 817b8b9c539 ("HID: elo: fix memory leak in elo_probe") Signed-off-by: Jiri Kosina <jkosina@suse.cz> Signed-off-by: Sasha Levin <sashal@kernel.org>
The introduction of '9a61f813fcc8 ("clk: qcom: regmap-mux: fix parent
clock lookup")' broke UFS support on SM8350.
The cause for this is that the symbol clocks have a specified rate in
the "freq-table-hz" table in the UFS node, which causes the UFS code to
request a rate change, for which the "bi_tcxo" happens to provide the
closest rate. Prior to the change in regmap-mux it was determined
(incorrectly) that no change was needed and everything worked.
The rates of 75 and 300MHz matches the documentation for the symbol
clocks, but we don't represent the parent clocks today. So let's mimic
the configuration found in other platforms, by omitting the rate for the
symbol clocks as well to avoid the rate change.
While at it also fill in the dummy symbol clocks that was dropped from
the GCC driver as it was upstreamed.
When calling gnttab_end_foreign_access_ref() the returned value must
be tested and the reaction to that value should be appropriate.
In case of failure in xennet_get_responses() the reaction should not be
to crash the system, but to disable the network device.
The calls in setup_netfront() can be replaced by calls of
gnttab_end_foreign_access(). While at it avoid double free of ring
pages and grant references via xennet_disconnect_backend() in this case.
This is CVE-2022-23042 / part of XSA-396.
Reported-by: Demi Marie Obenour <demi@invisiblethingslab.com> Signed-off-by: Juergen Gross <jgross@suse.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
gnttab_end_foreign_access() is used to free a grant reference and
optionally to free the associated page. In case the grant is still in
use by the other side processing is being deferred. This leads to a
problem in case no page to be freed is specified by the caller: the
caller doesn't know that the page is still mapped by the other side
and thus should not be used for other purposes.
The correct way to handle this situation is to take an additional
reference to the granted page in case handling is being deferred and
to drop that reference when the grant reference could be freed
finally.
This requires that there are no users of gnttab_end_foreign_access()
left directly repurposing the granted page after the call, as this
might result in clobbered data or information leaks via the not yet
freed grant reference.
This is part of CVE-2022-23041 / XSA-396.
Reported-by: Simon Gaiser <simon@invisiblethingslab.com> Signed-off-by: Juergen Gross <jgross@suse.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Instead of __get_free_pages() and free_pages() use alloc_pages_exact()
and free_pages_exact(). This is in preparation of a change of
gnttab_end_foreign_access() which will prohibit use of high-order
pages.
This is part of CVE-2022-23041 / XSA-396.
Reported-by: Simon Gaiser <simon@invisiblethingslab.com> Signed-off-by: Juergen Gross <jgross@suse.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Instead of __get_free_pages() and free_pages() use alloc_pages_exact()
and free_pages_exact(). This is in preparation of a change of
gnttab_end_foreign_access() which will prohibit use of high-order
pages.
By using the local variable "order" instead of ring->intf->ring_order
in the error path of xen_9pfs_front_alloc_dataring() another bug is
fixed, as the error path can be entered before ring->intf->ring_order
is being set.
By using alloc_pages_exact() the size in bytes is specified for the
allocation, which fixes another bug for the case of
order < (PAGE_SHIFT - XEN_PAGE_SHIFT).
This is part of CVE-2022-23041 / XSA-396.
Reported-by: Simon Gaiser <simon@invisiblethingslab.com> Signed-off-by: Juergen Gross <jgross@suse.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove gnttab_query_foreign_access(), as it is unused and unsafe to
use.
All previous use cases assumed a grant would not be in use after
gnttab_query_foreign_access() returned 0. This information is useless
in best case, as it only refers to a situation in the past, which could
have changed already.
Using gnttab_query_foreign_access() is unsafe, as it is racy by design.
The use case in the gntalloc driver is not needed at all. While at it
replace the call of gnttab_end_foreign_access_ref() with a call of
gnttab_end_foreign_access(), which is what is really wanted there. In
case the grant wasn't used due to an allocation failure, just free the
grant via gnttab_free_grant_reference().
This is CVE-2022-23039 / part of XSA-396.
Reported-by: Demi Marie Obenour <demi@invisiblethingslab.com> Signed-off-by: Juergen Gross <jgross@suse.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It isn't enough to check whether a grant is still being in use by
calling gnttab_query_foreign_access(), as a mapping could be realized
by the other side just after having called that function.
In case the call was done in preparation of revoking a grant it is
better to do so via gnttab_try_end_foreign_access() and check the
success of that operation instead.
This is CVE-2022-23038 / part of XSA-396.
Reported-by: Demi Marie Obenour <demi@invisiblethingslab.com> Signed-off-by: Juergen Gross <jgross@suse.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It isn't enough to check whether a grant is still being in use by
calling gnttab_query_foreign_access(), as a mapping could be realized
by the other side just after having called that function.
In case the call was done in preparation of revoking a grant it is
better to do so via gnttab_end_foreign_access_ref() and check the
success of that operation instead.
This is CVE-2022-23037 / part of XSA-396.
Reported-by: Demi Marie Obenour <demi@invisiblethingslab.com> Signed-off-by: Juergen Gross <jgross@suse.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It isn't enough to check whether a grant is still being in use by
calling gnttab_query_foreign_access(), as a mapping could be realized
by the other side just after having called that function.
In case the call was done in preparation of revoking a grant it is
better to do so via gnttab_end_foreign_access_ref() and check the
success of that operation instead.
For the ring allocation use alloc_pages_exact() in order to avoid
high order pages in case of a multi-page ring.
If a grant wasn't unmapped by the backend without persistent grants
being used, set the device state to "error".
This is CVE-2022-23036 / part of XSA-396.
Reported-by: Demi Marie Obenour <demi@invisiblethingslab.com> Signed-off-by: Juergen Gross <jgross@suse.com> Reviewed-by: Roger Pau Monné <roger.pau@citrix.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add a new grant table function gnttab_try_end_foreign_access(), which
will remove and free a grant if it is not in use.
Its main use case is to either free a grant if it is no longer in use,
or to take some other action if it is still in use. This other action
can be an error exit, or (e.g. in the case of blkfront persistent grant
feature) some special handling.
This is CVE-2022-23036, CVE-2022-23038 / part of XSA-396.
Reported-by: Demi Marie Obenour <demi@invisiblethingslab.com> Signed-off-by: Juergen Gross <jgross@suse.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Letting xenbus_grant_ring() tear down grants in the error case is
problematic, as the other side could already have used these grants.
Calling gnttab_end_foreign_access_ref() without checking success is
resulting in an unclear situation for any caller of xenbus_grant_ring()
as in the error case the memory pages of the ring page might be
partially mapped. Freeing them would risk unwanted foreign access to
them, while not freeing them would leak memory.
In order to remove the need to undo any gnttab_grant_foreign_access()
calls, use gnttab_alloc_grant_references() to make sure no further
error can occur in the loop granting access to the ring pages.
It should be noted that this way of handling removes leaking of
grant entries in the error case, too.
This is CVE-2022-23040 / part of XSA-396.
Reported-by: Demi Marie Obenour <demi@invisiblethingslab.com> Signed-off-by: Juergen Gross <jgross@suse.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When building arm64 defconfig + CONFIG_LTO_CLANG_{FULL,THIN}=y after
commit 558c303c9734 ("arm64: Mitigate spectre style branch history side
channels"), the following error occurs:
Marc figured out that moving "#include <linux/init.h>" in
include/linux/arm-smccc.h into a !__ASSEMBLY__ block resolves it. The
full include chain with CONFIG_LTO=y from include/linux/arm-smccc.h:
The asm/alternative-macros.h include in asm/rwonce.h only happens when
CONFIG_LTO is set, which ultimately casues asm/assembler.h to be
included before the definition of ARM_SMCCC_ARCH_WORKAROUND_3. As a
result, the preprocessor does not expand ARM_SMCCC_ARCH_WORKAROUND_3 in
__mitigate_spectre_bhb_fw, which results in the error above.
Avoid this problem by just avoiding the CONFIG_LTO=y __READ_ONCE() block
in asm/rwonce.h with assembly files, as nothing in that block is useful
to assembly files, which allows ARM_SMCCC_ARCH_WORKAROUND_3 to be
properly expanded with CONFIG_LTO=y builds.
ld.lld does not support the NOCROSSREFS directive at the moment, which
breaks the build after commit b9baf5c8c5c3 ("ARM: Spectre-BHB
workaround"):
ld.lld: error: ./arch/arm/kernel/vmlinux.lds:34: AT expected, but got NOCROSSREFS
Support for this directive will eventually be implemented, at which
point a version check can be added. To avoid breaking the build in the
meantime, just define NOCROSSREFS to nothing when using ld.lld, with a
link to the issue for tracking.
In the recent Spectre BHB patches, there was a typo that is only
exposed in certain configurations: mcr p15,0,XX,c7,r5,4 should have
been mcr p15,0,XX,c7,c5,4
Reported-by: kernel test robot <lkp@intel.com> Fixes: b9baf5c8c5c3 ("ARM: Spectre-BHB workaround") Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Acked-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The mitigations for Spectre-BHB are only applied when an exception is
taken from user-space. The mitigation status is reported via the spectre_v2
sysfs vulnerabilities file.
When unprivileged eBPF is enabled the mitigation in the exception vectors
can be avoided by an eBPF program.
When unprivileged eBPF is enabled, print a warning and report vulnerable
via the sysfs vulnerabilities file.
Future CPUs may implement a clearbhb instruction that is sufficient
to mitigate SpectreBHB. CPUs that implement this instruction, but
not CSV2.3 must be affected by Spectre-BHB.
Add support to use this instruction as the BHB mitigation on CPUs
that support it. The instruction is in the hint space, so it will
be treated by a NOP as older CPUs.
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: James Morse <james.morse@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
KVM allows the guest to discover whether the ARCH_WORKAROUND SMCCC are
implemented, and to preserve that state during migration through its
firmware register interface.
Add the necessary boiler plate for SMCCC_ARCH_WORKAROUND_3.
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: James Morse <james.morse@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Speculation attacks against some high-performance processors can
make use of branch history to influence future speculation.
When taking an exception from user-space, a sequence of branches
or a firmware call overwrites or invalidates the branch history.
The sequence of branches is added to the vectors, and should appear
before the first indirect branch. For systems using KPTI the sequence
is added to the kpti trampoline where it has a free register as the exit
from the trampoline is via a 'ret'. For systems not using KPTI, the same
register tricks are used to free up a register in the vectors.
For the firmware call, arch-workaround-3 clobbers 4 registers, so
there is no choice but to save them to the EL1 stack. This only happens
for entry from EL0, so if we take an exception due to the stack access,
it will not become re-entrant.
For KVM, the existing branch-predictor-hardening vectors are used.
When a spectre version of these vectors is in use, the firmware call
is sufficient to mitigate against Spectre-BHB. For the non-spectre
versions, the sequence of branches is added to the indirect vector.
Speculation attacks against some high-performance processors can
make use of branch history to influence future speculation as part of
a spectre-v2 attack. This is not mitigated by CSV2, meaning CPUs that
previously reported 'Not affected' are now moderately mitigated by CSV2.
Update the value in /sys/devices/system/cpu/vulnerabilities/spectre_v2
to also show the state of the BHB mitigation.
The Spectre-BHB workaround adds a firmware call to the vectors. This
is needed on some CPUs, but not others. To avoid the unaffected CPU in
a big/little pair from making the firmware call, create per cpu vectors.
The per-cpu vectors only apply when returning from EL0.
Systems using KPTI can use the canonical 'full-fat' vectors directly at
EL1, the trampoline exit code will switch to this_cpu_vector on exit to
EL0. Systems not using KPTI should always use this_cpu_vector.
this_cpu_vector will point at a vector in tramp_vecs or
__bp_harden_el1_vectors, depending on whether KPTI is in use.
The trampoline code needs to use the address of symbols in the wider
kernel, e.g. vectors. PC-relative addressing wouldn't work as the
trampoline code doesn't run at the address the linker expected.
tramp_ventry uses a literal pool, unless CONFIG_RANDOMIZE_BASE is
set, in which case it uses the data page as a literal pool because
the data page can be unmapped when running in user-space, which is
required for CPUs vulnerable to meltdown.
Pull this logic out as a macro, instead of adding a third copy
of it.
Some CPUs affected by Spectre-BHB need a sequence of branches, or a
firmware call to be run before any indirect branch. This needs to go
in the vectors. No CPU needs both.
While this can be patched in, it would run on all CPUs as there is a
single set of vectors. If only one part of a big/little combination is
affected, the unaffected CPUs have to run the mitigation too.
Create extra vectors that include the sequence. Subsequent patches will
allow affected CPUs to select this set of vectors. Later patches will
modify the loop count to match what the CPU requires.
Adding a second set of vectors to .entry.tramp.text will make it
larger than a single 4K page.
Allow the trampoline text to occupy up to three pages by adding two
more fixmap slots. Previous changes to tramp_valias allowed it to reach
beyond a single page.
Spectre-BHB needs to add sequences to the vectors. Having one global
set of vectors is a problem for big/little systems where the sequence
is costly on cpus that are not vulnerable.
Making the vectors per-cpu in the style of KVM's bh_harden_hyp_vecs
requires the vectors to be generated by macros.
Make the kpti re-mapping of the kernel optional, so the macros can be
used without kpti.
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: James Morse <james.morse@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The macros for building the kpti trampoline are all behind
CONFIG_UNMAP_KERNEL_AT_EL0, and in a region that outputs to the
.entry.tramp.text section.
Move the macros out so they can be used to generate other kinds of
trampoline. Only the symbols need to be guarded by
CONFIG_UNMAP_KERNEL_AT_EL0 and appear in the .entry.tramp.text section.
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: James Morse <james.morse@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Systems using kpti enter and exit the kernel through a trampoline mapping
that is always mapped, even when the kernel is not. tramp_valias is a macro
to find the address of a symbol in the trampoline mapping.
Adding extra sets of vectors will expand the size of the entry.tramp.text
section to beyond 4K. tramp_valias will be unable to generate addresses
for symbols beyond 4K as it uses the 12 bit immediate of the add
instruction.
As there are now two registers available when tramp_alias is called,
use the extra register to avoid the 4K limit of the 12 bit immediate.
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: James Morse <james.morse@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The trampoline code has a data page that holds the address of the vectors,
which is unmapped when running in user-space. This ensures that with
CONFIG_RANDOMIZE_BASE, the randomised address of the kernel can't be
discovered until after the kernel has been mapped.
If the trampoline text page is extended to include multiple sets of
vectors, it will be larger than a single page, making it tricky to
find the data page without knowing the size of the trampoline text
pages, which will vary with PAGE_SIZE.
Move the data page to appear before the text page. This allows the
data page to be found without knowing the size of the trampoline text
pages. 'tramp_vectors' is used to refer to the beginning of the
.entry.tramp.text section, do that explicitly.
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: James Morse <james.morse@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Subsequent patches will add additional sets of vectors that use
the same tricks as the kpti vectors to reach the full-fat vectors.
The full-fat vectors contain some cleanup for kpti that is patched
in by alternatives when kpti is in use. Once there are additional
vectors, the cleanup will be needed in more cases.
But on big/little systems, the cleanup would be harmful if no
trampoline vector were in use. Instead of forcing CPUs that don't
need a trampoline vector to use one, make the trampoline cleanup
optional.
Entry at the top of the vectors will skip the cleanup. The trampoline
vectors can then skip the first instruction, triggering the cleanup
to run.
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: James Morse <james.morse@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CPUs vulnerable to Spectre-BHB either need to make an SMC-CC firmware
call from the vectors, or run a sequence of branches. This gets added
to the hyp vectors. If there is no support for arch-workaround-1 in
firmware, the indirect vector will be used.
kvm_init_vector_slots() only initialises the two indirect slots if
the platform is vulnerable to Spectre-v3a. pKVM's hyp_map_vectors()
only initialises __hyp_bp_vect_base if the platform is vulnerable to
Spectre-v3a.
As there are about to more users of the indirect vectors, ensure
their entries in hyp_spectre_vector_selector[] are always initialised,
and __hyp_bp_vect_base defaults to the regular VA mapping.
The Spectre-v3a check is moved to a helper
kvm_system_needs_idmapped_vectors(), and merged with the code
that creates the hyp mappings.
The spectre-v4 sequence includes an SMC from the assembly entry code.
spectre_v4_patch_fw_mitigation_conduit is the patching callback that
generates an HVC or SMC depending on the SMCCC conduit type.
As this isn't specific to spectre-v4, rename it
smccc_patch_fw_mitigation_conduit so it can be re-used.
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: James Morse <james.morse@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add a new HWCAP to detect the Increased precision of Reciprocal Estimate
and Reciprocal Square Root Estimate feature (FEAT_RPRES), introduced in Armv8.7.
Also expose this to userspace in the ID_AA64ISAR2_EL1 feature register.
Signed-off-by: Joey Gouly <joey.gouly@arm.com> Cc: Will Deacon <will@kernel.org> Cc: Jonathan Corbet <corbet@lwn.net> Acked-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/20211210165432.8106-4-joey.gouly@arm.com Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The mitigations for Spectre-BHB are only applied when an exception
is taken, but when unprivileged BPF is enabled, userspace can
load BPF programs that can be used to exploit the problem.
When unprivileged BPF is enabled, report the vulnerable status via
the spectre_v2 sysfs file.
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Workaround the Spectre BHB issues for Cortex-A15, Cortex-A57,
Cortex-A72, Cortex-A73 and Cortex-A75. We also include Brahma B15 as
well to be safe, which is affected by Spectre V2 in the same ways as
Cortex-A15.
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
44a3918c8245 ("x86/speculation: Include unprivileged eBPF status in Spectre v2 mitigation reporting")
added a warning for the "eIBRS + unprivileged eBPF" combination, which
has been shown to be vulnerable against Spectre v2 BHB-based attacks.
However, there's no warning about the "eIBRS + LFENCE retpoline +
unprivileged eBPF" combo. The LFENCE adds more protection by shortening
the speculation window after a mispredicted branch. That makes an attack
significantly more difficult, even with unprivileged eBPF. So at least
for now the logic doesn't warn about that combination.
But if you then add SMT into the mix, the SMT attack angle weakens the
effectiveness of the LFENCE considerably.
So extend the "eIBRS + unprivileged eBPF" warning to also include the
"eIBRS + LFENCE + unprivileged eBPF + SMT" case.
it became possible to enable the LFENCE "retpoline" on Intel. However,
Intel doesn't recommend it, as it has some weaknesses compared to
retpoline.
Now AMD doesn't recommend it either.
It can still be left available as a cmdline option. It's faster than
retpoline but is weaker in certain scenarios -- particularly SMT, but
even non-SMT may be vulnerable in some cases.
So just unconditionally warn if the user requests it on the cmdline.
AMD retpoline may be susceptible to speculation. The speculation
execution window for an incorrect indirect branch prediction using
LFENCE/JMP sequence may potentially be large enough to allow
exploitation using Spectre V2.
By default, don't use retpoline,lfence on AMD. Instead, use the
generic retpoline.
Thanks to the chaps at VUsec it is now clear that eIBRS is not
sufficient, therefore allow enabling of retpolines along with eIBRS.
Add spectre_v2=eibrs, spectre_v2=eibrs,lfence and
spectre_v2=eibrs,retpoline options to explicitly pick your preferred
means of mitigation.
Since there's new mitigations there's also user visible changes in
/sys/devices/system/cpu/vulnerabilities/spectre_v2 to reflect these
new mitigations.
[ bp: Massage commit message, trim error messages,
do more precise eIBRS mode checking. ]
Co-developed-by: Josh Poimboeuf <jpoimboe@redhat.com> Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Borislav Petkov <bp@suse.de> Reviewed-by: Patrick Colp <patrick.colp@oracle.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The RETPOLINE_AMD name is unfortunate since it isn't necessarily
AMD only, in fact Hygon also uses it. Furthermore it will likely be
sufficient for some Intel processors. Therefore rename the thing to
RETPOLINE_LFENCE to better describe what it is.
Add the spectre_v2=retpoline,lfence option as an alias to
spectre_v2=retpoline,amd to preserve existing setups. However, the output
of /sys/devices/system/cpu/vulnerabilities/spectre_v2 will be changed.
Just like on the optional mmu_alloc_direct_roots() path, once shadow
path reaches "r = -EIO" somewhere, the caller needs to know the actual
state in order to enter error handling and avoid something worse.
Fixes: 4a38162ee9f1 ("KVM: MMU: load PDPTRs outside mmu_lock") Signed-off-by: Like Xu <likexu@tencent.com> Reviewed-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220301124941.48412-1-likexu@tencent.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ftrace_caller was used for both ftrace_caller and ftrace_regs_caller,
which means that the target address of the hotpatch trampoline was
never updated.
With commit 894979689d3a ("s390/ftrace: provide separate
ftrace_caller/ftrace_regs_caller implementations") a separate
ftrace_regs_caller entry point was implemeted, however it was
forgotten to implement the necessary changes for ftrace_modify_call
and ftrace_make_call, where the branch target has to be modified
accordingly.
Therefore add the missing code now.
Fixes: 894979689d3a ("s390/ftrace: provide separate ftrace_caller/ftrace_regs_caller implementations") Reviewed-by: Sven Schnelle <svens@linux.ibm.com> Acked-by: Ilya Leoshkevich <iii@linux.ibm.com> Signed-off-by: Heiko Carstens <hca@linux.ibm.com> Signed-off-by: Vasily Gorbik <gor@linux.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
arch_ftrace_get_regs is supposed to return a struct pt_regs pointer
only if the pt_regs structure contains all register contents, which
means it must have been populated when created via ftrace_regs_caller.
If it was populated via ftrace_caller the contents are not complete
(the psw mask part is missing), and therefore a NULL pointer needs be
returned.
The current code incorrectly always returns a struct pt_regs pointer.
Fix this by adding another pt_regs flag which indicates if the
contents are complete, and fix arch_ftrace_get_regs accordingly.
Fixes: 894979689d3a ("s390/ftrace: provide separate ftrace_caller/ftrace_regs_caller implementations") Reported-by: Christophe Leroy <christophe.leroy@csgroup.eu> Reported-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Reviewed-by: Sven Schnelle <svens@linux.ibm.com> Acked-by: Ilya Leoshkevich <iii@linux.ibm.com> Signed-off-by: Heiko Carstens <hca@linux.ibm.com> Signed-off-by: Vasily Gorbik <gor@linux.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When Linux runs as an Isolated VM on Hyper-V, it supports AMD SEV-SNP
but it's partially enlightened, i.e. cc_platform_has(
CC_ATTR_GUEST_MEM_ENCRYPT) is true but sev_active() is false.
Commit 4d96f9109109 per se is good, but with it now
kvm_setup_vsyscall_timeinfo() -> kvmclock_init_mem() calls
set_memory_decrypted(), and later gets stuck when trying to zere out
the pages pointed by 'hvclock_mem', if Linux runs as an Isolated VM on
Hyper-V. The cause is that here now the Linux VM should no longer access
the original guest physical addrss (GPA); instead the VM should do
memremap() and access the original GPA + ms_hyperv.shared_gpa_boundary:
see the example code in drivers/hv/connection.c: vmbus_connect() or
drivers/hv/ring_buffer.c: hv_ringbuffer_init(). If the VM tries to
access the original GPA, it keepts getting injected a fault by Hyper-V
and gets stuck there.
Here the issue happens only when the VM has >=65 vCPUs, because the
global static array hv_clock_boot[] can hold 64 "struct
pvclock_vsyscall_time_info" (the sizeof of the struct is 64 bytes), so
kvmclock_init_mem() only allocates memory in the case of vCPUs > 64.
Since the 'hvclock_mem' pages are only useful when the kvm clock is
supported by the underlying hypervisor, fix the issue by returning
early when Linux VM runs on Hyper-V, which doesn't support kvm clock.
Fixes: 4d96f9109109 ("x86/sev: Replace occurrences of sev_active() with cc_platform_has()") Tested-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com> Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com> Signed-off-by: Dexuan Cui <decui@microsoft.com>
Message-Id: <20220225084600.17817-1-decui@microsoft.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Since bit 57 was exported for uffd-wp write-protected (commit fb8e37f35a2f: "mm/pagemap: export uffd-wp protection information"),
fixing it can reduce some unnecessary confusion.
Link: https://lkml.kernel.org/r/20220301044538.3042713-1-yun.zhou@windriver.com Fixes: fb8e37f35a2fe1 ("mm/pagemap: export uffd-wp protection information") Signed-off-by: Yun Zhou <yun.zhou@windriver.com> Reviewed-by: Peter Xu <peterx@redhat.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Tiberiu A Georgescu <tiberiu.georgescu@nutanix.com> Cc: Florian Schmidt <florian.schmidt@nutanix.com> Cc: Ivan Teterevkov <ivan.teterevkov@nutanix.com> Cc: SeongJae Park <sj@kernel.org> Cc: Yang Shi <shy828301@gmail.com> Cc: David Hildenbrand <david@redhat.com> Cc: Axel Rasmussen <axelrasmussen@google.com> Cc: Miaohe Lin <linmiaohe@huawei.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Colin Cross <ccross@google.com> Cc: Alistair Popple <apopple@nvidia.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit b515d2637276a3810d6595e10ab02c13bfd0b63a ("xfrm: xfrm_state_mtu
should return at least 1280 for ipv6") in v5.14 breaks the TCP MSS
calculation in ipsec transport mode, resulting complete stalls of TCP
connections. This happens when the (P)MTU is 1280 or slighly larger.
The desired formula for the MSS is:
MSS = (MTU - ESP_overhead) - IP header - TCP header
However, the above commit clamps the (MTU - ESP_overhead) to a
minimum of 1280, turning the formula into
MSS = max(MTU - ESP overhead, 1280) - IP header - TCP header
With the (P)MTU near 1280, the calculated MSS is too large and the
resulting TCP packets never make it to the destination because they
are over the actual PMTU.
The above commit also causes suboptimal double fragmentation in
xfrm tunnel mode, as described in
https://lore.kernel.org/netdev/20210429202529.codhwpc7w6kbudug@dwarf.suse.cz/
We hit a bug with a recovering relocation on mount for one of our file
systems in production. I reproduced this locally by injecting errors
into snapshot delete with balance running at the same time. This
presented as an error while looking up an extent item
Normally snapshot deletion and relocation are excluded from running at
the same time by the fs_info->cleaner_mutex. However if we had a
pending balance waiting to get the ->cleaner_mutex, and a snapshot
deletion was running, and then the box crashed, we would come up in a
state where we have a half deleted snapshot.
Again, in the normal case the snapshot deletion needs to complete before
relocation can start, but in this case relocation could very well start
before the snapshot deletion completes, as we simply add the root to the
dead roots list and wait for the next time the cleaner runs to clean up
the snapshot.
Fix this by setting a bit on the fs_info if we have any DEAD_ROOT's that
had a pending drop_progress key. If they do then we know we were in the
middle of the drop operation and set a flag on the fs_info. Then
balance can wait until this flag is cleared to start up again.
If there are DEAD_ROOT's that don't have a drop_progress set then we're
safe to start balance right away as we'll be properly protected by the
cleaner_mutex.
CC: stable@vger.kernel.org # 5.10+ Reviewed-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Josef Bacik <josef@toxicpanda.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some users recently reported that MariaDB was getting a read corruption
when using io_uring on top of btrfs. This started to happen in 5.16,
after commit 51bd9563b6783d ("btrfs: fix deadlock due to page faults
during direct IO reads and writes"). That changed btrfs to use the new
iomap flag IOMAP_DIO_PARTIAL and to disable page faults before calling
iomap_dio_rw(). This was necessary to fix deadlocks when the iovector
corresponds to a memory mapped file region. That type of scenario is
exercised by test case generic/647 from fstests.
For this MariaDB scenario, we attempt to read 16K from file offset X
using IOCB_NOWAIT and io_uring. In that range we have 4 extents, each
with a size of 4K, and what happens is the following:
1) btrfs_direct_read() disables page faults and calls iomap_dio_rw();
2) iomap creates a struct iomap_dio object, its reference count is
initialized to 1 and its ->size field is initialized to 0;
3) iomap calls btrfs_dio_iomap_begin() with file offset X, which finds
the first 4K extent, and setups an iomap for this extent consisting
of a single page;
4) At iomap_dio_bio_iter(), we are able to access the first page of the
buffer (struct iov_iter) with bio_iov_iter_get_pages() without
triggering a page fault;
5) iomap submits a bio for this 4K extent
(iomap_dio_submit_bio() -> btrfs_submit_direct()) and increments
the refcount on the struct iomap_dio object to 2; The ->size field
of the struct iomap_dio object is incremented to 4K;
6) iomap calls btrfs_iomap_begin() again, this time with a file
offset of X + 4K. There we setup an iomap for the next extent
that also has a size of 4K;
7) Then at iomap_dio_bio_iter() we call bio_iov_iter_get_pages(),
which tries to access the next page (2nd page) of the buffer.
This triggers a page fault and returns -EFAULT;
8) At __iomap_dio_rw() we see the -EFAULT, but we reset the error
to 0 because we passed the flag IOMAP_DIO_PARTIAL to iomap and
the struct iomap_dio object has a ->size value of 4K (we submitted
a bio for an extent already). The 'wait_for_completion' variable
is not set to true, because our iocb has IOCB_NOWAIT set;
9) At the bottom of __iomap_dio_rw(), we decrement the reference count
of the struct iomap_dio object from 2 to 1. Because we were not
the only ones holding a reference on it and 'wait_for_completion' is
set to false, -EIOCBQUEUED is returned to btrfs_direct_read(), which
just returns it up the callchain, up to io_uring;
10) The bio submitted for the first extent (step 5) completes and its
bio endio function, iomap_dio_bio_end_io(), decrements the last
reference on the struct iomap_dio object, resulting in calling
iomap_dio_complete_work() -> iomap_dio_complete().
11) At iomap_dio_complete() we adjust the iocb->ki_pos from X to X + 4K
and return 4K (the amount of io done) to iomap_dio_complete_work();
12) iomap_dio_complete_work() calls the iocb completion callback,
iocb->ki_complete() with a second argument value of 4K (total io
done) and the iocb with the adjust ki_pos of X + 4K. This results
in completing the read request for io_uring, leaving it with a
result of 4K bytes read, and only the first page of the buffer
filled in, while the remaining 3 pages, corresponding to the other
3 extents, were not filled;
13) For the application, the result is unexpected because if we ask
to read N bytes, it expects to get N bytes read as long as those
N bytes don't cross the EOF (i_size).
MariaDB reports this as an error, as it's not expecting a short read,
since it knows it's asking for read operations fully within the i_size
boundary. This is typical in many applications, but it may also be
questionable if they should react to such short reads by issuing more
read calls to get the remaining data. Nevertheless, the short read
happened due to a change in btrfs regarding how it deals with page
faults while in the middle of a read operation, and there's no reason
why btrfs can't have the previous behaviour of returning the whole data
that was requested by the application.
The problem can also be triggered with the following simple program:
/* Get O_DIRECT */
#ifndef _GNU_SOURCE
#define _GNU_SOURCE
#endif
foo_path = malloc(strlen(argv[1]) + 5);
if (!foo_path) {
fprintf(stderr, "Failed to allocate memory for file path\n");
return 1;
}
strcpy(foo_path, argv[1]);
strcat(foo_path, "/foo");
/*
* Create file foo with 2 extents, each with a size matching
* the page size. Then allocate a buffer to read both extents
* with io_uring, using O_DIRECT and IOCB_NOWAIT. Before doing
* the read with io_uring, access the first page of the buffer
* to fault it in, so that during the read we only trigger a
* page fault when accessing the second page of the buffer.
*/
fd = open(foo_path, O_CREAT | O_TRUNC | O_WRONLY |
O_DIRECT, 0666);
if (fd == -1) {
fprintf(stderr,
"Failed to create file 'foo': %s (errno %d)",
strerror(errno), errno);
return 1;
}
pagesize = sysconf(_SC_PAGE_SIZE);
ret = posix_memalign(&write_buf, pagesize, 2 * pagesize);
if (ret) {
fprintf(stderr, "Failed to allocate write buffer\n");
return 1;
}
/* Create 2 extents, each with a size matching page size. */
for (i = 0; i < 2; i++) {
ret = pwrite(fd, write_buf + i * pagesize, pagesize,
i * pagesize);
if (ret != pagesize) {
fprintf(stderr,
"Failed to write to file, ret = %ld errno %d (%s)\n",
ret, errno, strerror(errno));
return 1;
}
ret = fsync(fd);
if (ret != 0) {
fprintf(stderr, "Failed to fsync file\n");
return 1;
}
}
close(fd);
fd = open(foo_path, O_RDONLY | O_DIRECT);
if (fd == -1) {
fprintf(stderr,
"Failed to open file 'foo': %s (errno %d)",
strerror(errno), errno);
return 1;
}
ret = posix_memalign(&read_buf, pagesize, 2 * pagesize);
if (ret) {
fprintf(stderr, "Failed to allocate read buffer\n");
return 1;
}
/*
* Fault in only the first page of the read buffer.
* We want to trigger a page fault for the 2nd page of the
* read buffer during the read operation with io_uring
* (O_DIRECT and IOCB_NOWAIT).
*/
memset(read_buf, 0, 1);
ret = io_uring_queue_init(1, &ring, 0);
if (ret != 0) {
fprintf(stderr, "Failed to create io_uring queue\n");
return 1;
}
sqe = io_uring_get_sqe(&ring);
if (!sqe) {
fprintf(stderr, "Failed to get io_uring sqe\n");
return 1;
}
After this patch, the read always returns 8192 bytes, with the buffer
filled with the correct data. Although that reproducer always triggers
the bug in my test vms, it's possible that it will not be so reliable
on other environments, as that can happen if the bio for the first
extent completes and decrements the reference on the struct iomap_dio
object before we do the atomic_dec_and_test() on the reference at
__iomap_dio_rw().
Fix this in btrfs by having btrfs_dio_iomap_begin() return -EAGAIN
whenever we try to satisfy a non blocking IO request (IOMAP_NOWAIT flag
set) over a range that spans multiple extents (or a mix of extents and
holes). This avoids returning success to the caller when we only did
partial IO, which is not optimal for writes and for reads it's actually
incorrect, as the caller doesn't expect to get less bytes read than it has
requested (unless EOF is crossed), as previously mentioned. This is also
the type of behaviour that xfs follows (xfs_direct_write_iomap_begin()),
even though it doesn't use IOMAP_DIO_PARTIAL.
During log replay, whenever we need to check if a name (dentry) exists in
a directory we do searches on the subvolume tree for inode references or
or directory entries (BTRFS_DIR_INDEX_KEY keys, and BTRFS_DIR_ITEM_KEY
keys as well, before kernel 5.17). However when during log replay we
unlink a name, through btrfs_unlink_inode(), we may not delete inode
references and dir index keys from a subvolume tree and instead just add
the deletions to the delayed inode's delayed items, which will only be
run when we commit the transaction used for log replay. This means that
after an unlink operation during log replay, if we attempt to search for
the same name during log replay, we will not see that the name was already
deleted, since the deletion is recorded only on the delayed items.
We run delayed items after every unlink operation during log replay,
except at unlink_old_inode_refs() and at add_inode_ref(). This was due
to an overlook, as delayed items should be run after evert unlink, for
the reasons stated above.
So fix those two cases.
Fixes: 0d836392cadd5 ("Btrfs: fix mount failure after fsync due to hard link recreation") Fixes: 1f250e929a9c9 ("Btrfs: fix log replay failure after unlink and link combination") CC: stable@vger.kernel.org # 4.19+ Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The commit e804861bd4e6 ("btrfs: fix deadlock between quota disable and
qgroup rescan worker") by Kawasaki resolves deadlock between quota
disable and qgroup rescan worker. But also there is a deadlock case like
it. It's about enabling or disabling quota and creating or removing
qgroup. It can be reproduced in simple script below.
for i in {1..100}
do
btrfs quota enable /mnt &
btrfs qgroup create 1/0 /mnt &
btrfs qgroup destroy 1/0 /mnt &
btrfs quota disable /mnt &
done
Here's why the deadlock happens:
1) The quota rescan task is running.
2) Task A calls btrfs_quota_disable(), locks the qgroup_ioctl_lock
mutex, and then calls btrfs_qgroup_wait_for_completion(), to wait for
the quota rescan task to complete.
3) Task B calls btrfs_remove_qgroup() and it blocks when trying to lock
the qgroup_ioctl_lock mutex, because it's being held by task A. At that
point task B is holding a transaction handle for the current transaction.
4) The quota rescan task calls btrfs_commit_transaction(). This results
in it waiting for all other tasks to release their handles on the
transaction, but task B is blocked on the qgroup_ioctl_lock mutex
while holding a handle on the transaction, and that mutex is being held
by task A, which is waiting for the quota rescan task to complete,
resulting in a deadlock between these 3 tasks.
To resolve this issue, the thread disabling quota should unlock
qgroup_ioctl_lock before waiting rescan completion. Move
btrfs_qgroup_wait_for_completion() after unlock of qgroup_ioctl_lock.
Fixes: e804861bd4e6 ("btrfs: fix deadlock between quota disable and qgroup rescan worker") CC: stable@vger.kernel.org # 5.4+ Reviewed-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com> Signed-off-by: Sidong Yang <realwakka@gmail.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Whenever we do any extent buffer operations we call
assert_eb_page_uptodate() to complain loudly if we're operating on an
non-uptodate page. Our overnight tests caught this warning earlier this
week
This was partially fixed by c2e39305299f01 ("btrfs: clear extent buffer
uptodate when we fail to write it"), however all that fix did was keep
us from finding extent buffers after a failed writeout. It didn't keep
us from continuing to use a buffer that we already had found.
In this case we're searching the commit root to cache the block group,
so we can start committing the transaction and switch the commit root
and then start writing. After the switch we can look up an extent
buffer that hasn't been written yet and start processing that block
group. Then we fail to write that block out and clear Uptodate on the
page, and then we start spewing these errors.
Normally we're protected by the tree lock to a certain degree here. If
we read a block we have that block read locked, and we block the writer
from locking the block before we submit it for the write. However this
isn't necessarily fool proof because the read could happen before we do
the submit_bio and after we locked and unlocked the extent buffer.
Also in this particular case we have path->skip_locking set, so that
won't save us here. We'll simply get a block that was valid when we
read it, but became invalid while we were using it.
What we really want is to catch the case where we've "read" a block but
it's not marked Uptodate. On read we ClearPageError(), so if we're
!Uptodate and !Error we know we didn't do the right thing for reading
the page.
Fix this by checking !Uptodate && !Error, this way we will not complain
if our buffer gets invalidated while we're using it, and we'll maintain
the spirit of the check which is to make sure we have a fully in-cache
block while we're messing with it.
CC: stable@vger.kernel.org # 5.4+ Signed-off-by: Josef Bacik <josef@toxicpanda.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[BUG]
When looping btrfs/074 with 64K page size and 4K sectorsize, there is a
low chance (1/50~1/100) to crash with the following ASSERT() triggered
in btrfs_subpage_start_writer():
ret = atomic_add_return(nbits, &subpage->writers);
ASSERT(ret == nbits); <<< This one <<<
[CAUSE]
With more debugging output on the parameters of
btrfs_subpage_start_writer(), it shows a very concerning error:
ret=29 nbits=13 start=393216 len=53248
For @nbits it's correct, but @ret which is the returned value from
atomic_add_return(), it's not only larger than nbits, but also larger
than max sectors per page value (for 64K page size and 4K sector size,
it's 16).
This indicates that some call sites are not properly decreasing the value.
And that's exactly the case, in btrfs_page_unlock_writer(), due to the
fact that we can have page locked either by lock_page() or
process_one_page(), we have to check if the subpage has any writer.
If no writers, it's locked by lock_page() and we only need to unlock it.
But unfortunately the check for the writers are completely opposite:
if (atomic_read(&subpage->writers))
/* No writers, locked by plain lock_page() */
return unlock_page(page);
We directly unlock the page if it has writers, which is the completely
opposite what we want.
Thankfully the affected call site is only limited to
extent_write_locked_range(), so it's mostly affecting compressed write.
[FIX]
Just fix the wrong check condition to fix the bug.
We first hit the WARN_ON(rc->block_group->pinned > 0) in
btrfs_relocate_block_group() and then the BUG_ON(!cache) in
unpin_extent_range(). This tells us that we are exiting relocation and
removing the block group with bytes still pinned for that block group.
This is supposed to be impossible: the last thing relocate_block_group()
does is commit the transaction to get rid of pinned extents.
Commit d0c2f4fa555e ("btrfs: make concurrent fsyncs wait less when
waiting for a transaction commit") introduced an optimization so that
commits from fsync don't have to wait for the previous commit to unpin
extents. This was only intended to affect fsync, but it inadvertently
made it possible for any commit to skip waiting for the previous commit
to unpin. This is because if a call to btrfs_commit_transaction() finds
that another thread is already committing the transaction, it waits for
the other thread to complete the commit and then returns. If that other
thread was in fsync, then it completes the commit without completing the
previous commit. This makes the following sequence of events possible:
There are other sequences involving SUPER_COMMITTED transactions that
can cause a similar outcome.
We could fix this by making relocation explicitly wait for unpinning,
but there may be other cases that need it. Josef mentioned ENOSPC
flushing and the free space cache inode as other potential victims.
Rather than playing whack-a-mole, this fix is conservative and makes all
commits not in fsync wait for all previous transactions, which is what
the optimization intended.
Fixes: d0c2f4fa555e ("btrfs: make concurrent fsyncs wait less when waiting for a transaction commit") CC: stable@vger.kernel.org # 5.15+ Reviewed-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Omar Sandoval <osandov@fb.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When doing a full fsync, if we have prealloc extents beyond (or at) eof,
and the leaves that contain them were not modified in the current
transaction, we end up not logging them. This results in losing those
extents when we replay the log after a power failure, since the inode is
truncated to the current value of the logged i_size.
Just like for the fast fsync path, we need to always log all prealloc
extents starting at or beyond i_size. The fast fsync case was fixed in
commit 471d557afed155 ("Btrfs: fix loss of prealloc extents past i_size
after fsync log replay") but it missed the full fsync path. The problem
exists since the very early days, when the log tree was added by
commit e02119d5a7b439 ("Btrfs: Add a write ahead tree log to optimize
synchronous operations").
Example reproducer:
$ mkfs.btrfs -f /dev/sdc
$ mount /dev/sdc /mnt
# Create our test file with many file extent items, so that they span
# several leaves of metadata, even if the node/page size is 64K. Use
# direct IO and not fsync/O_SYNC because it's both faster and it avoids
# clearing the full sync flag from the inode - we want the fsync below
# to trigger the slow full sync code path.
$ xfs_io -f -d -c "pwrite -b 4K 0 16M" /mnt/foo
# Now add two preallocated extents to our file without extending the
# file's size. One right at i_size, and another further beyond, leaving
# a gap between the two prealloc extents.
$ xfs_io -c "falloc -k 16M 1M" /mnt/foo
$ xfs_io -c "falloc -k 20M 1M" /mnt/foo
# Make sure everything is durably persisted and the transaction is
# committed. This makes all created extents to have a generation lower
# than the generation of the transaction used by the next write and
# fsync.
sync
# Now overwrite only the first extent, which will result in modifying
# only the first leaf of metadata for our inode. Then fsync it. This
# fsync will use the slow code path (inode full sync bit is set) because
# it's the first fsync since the inode was created/loaded.
$ xfs_io -c "pwrite 0 4K" -c "fsync" /mnt/foo
__setup() handlers should generally return 1 to indicate that the
boot options have been handled.
Using invalid option values causes the entire kernel boot option
string to be reported as Unknown and added to init's environment
strings, polluting it.
Unknown kernel command line parameters "BOOT_IMAGE=/boot/bzImage-517rc6
kprobe_event=p,syscall_any,$arg1 trace_options=quiet
trace_clock=jiffies", will be passed to user space.
Run /sbin/init as init process
with arguments:
/sbin/init
with environment:
HOME=/
TERM=linux
BOOT_IMAGE=/boot/bzImage-517rc6
kprobe_event=p,syscall_any,$arg1
trace_options=quiet
trace_clock=jiffies
Return 1 from the __setup() handlers so that init's environment is not
polluted with kernel boot options.
When trying to add a histogram against an event with the "cpu" field, it
was impossible due to "cpu" being a keyword to key off of the running CPU.
So to fix this, it was changed to "common_cpu" to match the other generic
fields (like "common_pid"). But since some scripts used "cpu" for keying
off of the CPU (for events that did not have "cpu" as a field, which is
most of them), a backward compatibility trick was added such that if "cpu"
was used as a key, and the event did not have "cpu" as a field name, then
it would fallback and switch over to "common_cpu".
This fix has a couple of subtle bugs. One was that when switching over to
"common_cpu", it did not change the field name, it just set a flag. But
the code still found a "cpu" field. The "cpu" field is used for filtering
and is returned when the event does not have a "cpu" field.
Instead of hard coding the "cpu" checks, take advantage of the fact that
trace_event_field_field() returns a special field for "cpu" and "CPU" if
the event does not have "cpu" as a field. This special field has the
"filter_type" of "FILTER_CPU". Check that to test if the returned field is
of the CPU type instead of doing the string compare.
Also, fix the sorting bug by testing for the hist_field flag of
HIST_FIELD_FL_CPU when setting up the sort routine. Otherwise it will use
the special CPU field to know what compare routine to use, and since that
special field does not have a size, it returns tracing_map_cmp_none.
Cc: stable@vger.kernel.org Fixes: 1e3bac71c505 ("tracing/histogram: Rename "cpu" to "common_cpu"") Reported-by: Daniel Bristot de Oliveira <bristot@kernel.org> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Numerous keyboards are adding dictate keys which allows for text
messages to be dictated by a microphone.
This patch adds a new key definition KEY_DICTATE and maps 0x0c/0x0d8
usage code to this new keycode. Additionally hid-debug is adjusted to
recognize this new usage code as well.
Make the samsung-keypad driver explicitly depend on CONFIG_HAS_IOMEM, as it
calls devm_ioremap(). This prevents compile errors in some configs (e.g,
allyesconfig/randconfig under UML):
/usr/bin/ld: drivers/input/keyboard/samsung-keypad.o: in function `samsung_keypad_probe':
samsung-keypad.c:(.text+0xc60): undefined reference to `devm_ioremap'
Signed-off-by: David Gow <davidgow@google.com> Acked-by: anton ivanov <anton.ivanov@cambridgegreys.com> Link: https://lore.kernel.org/r/20220225041727.1902850-1-davidgow@google.com Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Before these changes elan_suspend() would only disable the regulator
when device_may_wakeup() returns false; whereas elan_resume() would
unconditionally enable it, leading to an enable count imbalance when
device_may_wakeup() returns true.
This triggers the "WARN_ON(regulator->enable_count)" in regulator_put()
when the elan_i2c driver gets unbound, this happens e.g. with the
hot-plugable dock with Elan I2C touchpad for the Asus TF103C 2-in-1.
Fix this by making the regulator_enable() call also be conditional
on device_may_wakeup() returning false.
projects / thirdparty / kernel / stable.git / log
