Pull lsm updates from Paul Moore:
- Move the LSM hook comment blocks into security/security.c
For many years the LSM hook comment blocks were located in a very odd
place, include/linux/lsm_hooks.h, where they lived on their own,
disconnected from both the function prototypes and definitions.
In keeping with current kernel conventions, this moves all of these
comment blocks to the top of the function definitions, transforming
them into the kdoc format in the process. This should make it much
easier to maintain these comments, which are the main source of LSM
hook documentation.
For the most part the comment contents were left as-is, although some
glaring errors were corrected. Expect additional edits in the future
as we slowly update and correct the comment blocks.
This is the bulk of the diffstat.
- Introduce LSM_ORDER_LAST
Similar to how LSM_ORDER_FIRST is used to specify LSMs which should
be ordered before "normal" LSMs, the LSM_ORDER_LAST is used to
specify LSMs which should be ordered after "normal" LSMs.
This is one of the prerequisites for transitioning IMA/EVM to a
proper LSM.
- Remove the security_old_inode_init_security() hook
The security_old_inode_init_security() LSM hook only allows for a
single xattr which is problematic both for LSM stacking and the
IMA/EVM-as-a-LSM effort. This finishes the conversion over to the
security_inode_init_security() hook and removes the single-xattr LSM
hook.
- Fix a reiserfs problem with security xattrs
During the security_old_inode_init_security() removal work it became
clear that reiserfs wasn't handling security xattrs properly so we
fixed it.
* tag 'lsm-pr-20230420' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/lsm: (32 commits)
reiserfs: Add security prefix to xattr name in reiserfs_security_write()
security: Remove security_old_inode_init_security()
ocfs2: Switch to security_inode_init_security()
reiserfs: Switch to security_inode_init_security()
security: Remove integrity from the LSM list in Kconfig
Revert "integrity: double check iint_cache was initialized"
security: Introduce LSM_ORDER_LAST and set it for the integrity LSM
device_cgroup: Fix typo in devcgroup_css_alloc description
lsm: fix a badly named parameter in security_get_getsecurity()
lsm: fix doc warnings in the LSM hook comments
lsm: styling fixes to security/security.c
lsm: move the remaining LSM hook comments to security/security.c
lsm: move the io_uring hook comments to security/security.c
lsm: move the perf hook comments to security/security.c
lsm: move the bpf hook comments to security/security.c
lsm: move the audit hook comments to security/security.c
lsm: move the binder hook comments to security/security.c
lsm: move the sysv hook comments to security/security.c
lsm: move the key hook comments to security/security.c
lsm: move the xfrm hook comments to security/security.c
...
Pull selinux updates from Paul Moore:
- Stop passing the 'selinux_state' pointers as function arguments
As discussed during the end of the last development cycle, passing a
selinux_state pointer through the SELinux code has a noticeable
impact on performance, and with the current code it is not strictly
necessary.
This simplifies things by referring directly to the single
selinux_state global variable which should help improve SELinux
performance.
- Uninline the unlikely portions of avc_has_perm_noaudit()
This change was also based on a discussion from the last development
cycle, and is heavily based on an initial proof of concept patch from
you. The core issue was that avc_has_perm_noaudit() was not able to
be inlined, as intended, due to its size. We solved this issue by
extracting the less frequently hit portions of avc_has_perm_noaudit()
into a separate function, reducing the size of avc_has_perm_noaudit()
to the point where the compiler began inlining the function. We also
took the opportunity to clean up some ugly RCU locking in the code
that became uglier with the change.
- Remove the runtime disable functionality
After several years of work by the userspace and distro folks, we are
finally in a place where we feel comfortable removing the runtime
disable functionality which we initially deprecated at the start of
2020.
There is plenty of information in the kernel's deprecation (now
removal) notice, but the main motivation was to be able to safely
mark the LSM hook structures as '__ro_after_init'.
LWN also wrote a good summary of the deprecation this morning which
offers a more detailed history:
https://lwn.net/SubscriberLink/927463/dcfa0d4ed2872f03
- Remove the checkreqprot functionality
The original checkreqprot deprecation notice stated that the removal
would happen no sooner than June 2021, which means this falls hard
into the "better late than never" bucket.
The Kconfig and deprecation notice has more detail on this setting,
but the basic idea is that we want to ensure that the SELinux policy
allows for the memory protections actually applied by the kernel, and
not those requested by the process.
While we haven't found anyone running a supported distro that is
affected by this deprecation/removal, anyone who is affected would
only need to update their policy to reflect the reality of their
applications' mapping protections.
- Minor Makefile improvements
Some minor Makefile improvements to correct some dependency issues
likely only ever seen by SELinux developers. I expect we will have at
least one more tweak to the Makefile during the next merge window,
but it didn't quite make the cutoff this time around.
* tag 'selinux-pr-20230420' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux:
selinux: ensure av_permissions.h is built when needed
selinux: fix Makefile dependencies of flask.h
selinux: stop returning node from avc_insert()
selinux: clean up dead code after removing runtime disable
selinux: update the file list in MAINTAINERS
selinux: remove the runtime disable functionality
selinux: remove the 'checkreqprot' functionality
selinux: stop passing selinux_state pointers and their offspring
selinux: uninline unlikely parts of avc_has_perm_noaudit()
Merge my x86 user copy updates branch.
This cleans up a lot of our x86 memory copy code, particularly for user
accesses. I've been pushing for microarchitectural support for good
memory copying and clearing for a long while, and it's been visible in
how the kernel has aggressively used 'rep movs' and 'rep stos' whenever
possible.
And that micro-architectural support has been improving over the years,
to the point where on modern CPU's the best option for a memory copy
that would become a function call (as opposed to being something that
can just be turned into individual 'mov' instructions) is now to inline
the string instruction sequence instead.
However, that only makes sense when we have the modern markers for this:
the x86 FSRM and FSRS capabilities ("Fast Short REP MOVS/STOS").
So this cleans up a lot of our historical code, gets rid of the legacy
marker use ("REP_GOOD" and "ERMS") from the memcpy/memset cases, and
replaces it with that modern reality. Note that REP_GOOD and ERMS end
up still being used by the known large cases (ie page copyin gand
clearing).
The reason much of this ends up being about user memory accesses is that
the normal in-kernel cases are done by the compiler (__builtin_memcpy()
and __builtin_memset()) and getting to the point where we can use our
instruction rewriting to inline those to be string instructions will
need some compiler support.
In contrast, the user accessor functions are all entirely controlled by
the kernel code, so we can change those arbitrarily.
Thanks to Borislav Petkov for feedback on the series, and Jens testing
some of this on micro-architectures I didn't personally have access to.
* x86-rep-insns:
x86: rewrite '__copy_user_nocache' function
x86: remove 'zerorest' argument from __copy_user_nocache()
x86: set FSRS automatically on AMD CPUs that have FSRM
x86: improve on the non-rep 'copy_user' function
x86: improve on the non-rep 'clear_user' function
x86: inline the 'rep movs' in user copies for the FSRM case
x86: move stac/clac from user copy routines into callers
x86: don't use REP_GOOD or ERMS for user memory clearing
x86: don't use REP_GOOD or ERMS for user memory copies
x86: don't use REP_GOOD or ERMS for small memory clearing
x86: don't use REP_GOOD or ERMS for small memory copies
Use the same pattern as the compat version of this code does: instead of
copying the whole array to a kernel buffer and then having a separate
phase of verifying it, just do it one entry at a time, verifying as you
go.
On Jens' /dev/zero readv() test this improves performance by ~6%.
[ This was obviously triggered by Jens' ITER_UBUF updates series ]
Reported-and-tested-by: Jens Axboe <axboe@kernel.dk>
Link: https://lore.kernel.org/all/de35d11d-bce7-e976-7372-1f2caf417103@kernel.dk/
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull ITER_UBUF updates from Jens Axboe:
"This turns singe vector imports into ITER_UBUF, rather than
ITER_IOVEC.
The former is more trivial to iterate and advance, and hence a bit
more efficient. From some very unscientific testing, ~60% of all iovec
imports are single vector"
* tag 'iter-ubuf.2-2023-04-21' of git://git.kernel.dk/linux:
iov_iter: Mark copy_compat_iovec_from_user() noinline
iov_iter: import single vector iovecs as ITER_UBUF
iov_iter: convert import_single_range() to ITER_UBUF
iov_iter: overlay struct iovec and ubuf/len
iov_iter: set nr_segs = 1 for ITER_UBUF
iov_iter: remove iov_iter_iovec()
iov_iter: add iter_iov_addr() and iter_iov_len() helpers
ALSA: pcm: check for user backed iterator, not specific iterator type
IB/qib: check for user backed iterator, not specific iterator type
IB/hfi1: check for user backed iterator, not specific iterator type
iov_iter: add iter_iovec() helper
block: ensure bio_alloc_map_data() deals with ITER_UBUF correctly
Pull ARM development updates from Russell King:
"Four changes for v6.4:
- simplify the path to the top vmlinux
- three patches to fix vfp with instrumentation enabled (eg lockdep)"
* tag 'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm:
ARM: 9294/2: vfp: Fix broken softirq handling with instrumentation enabled
ARM: 9293/1: vfp: Pass successful return address via register R3
ARM: 9292/1: vfp: Pass thread_info pointer to vfp_support_entry
ARM: 9291/1: decompressor: simplify the path to the top vmlinux
Intel compiler support has already been completely removed in commit
95207db816 ("Remove Intel compiler support"). However, it appears
that there is still some ICC-related code in scripts/cc-version.sh.
There is no harm in leaving the code as it is, but removing the dead
code makes the codebase a bit cleaner.
Hopefully all ICC-related stuff in the build scripts will be removed
after this commit, given the grep output as below:
(linux/scripts) $ grep -i -w -R 'icc'
cc-version.sh:ICC)
cc-version.sh: min_version=$($min_tool_version icc)
dtc/include-prefixes/arm64/qcom/sm6350.dtsi:#include <dt-bindings/interconnect/qcom,icc.h>
Fixes: 95207db816 ("Remove Intel compiler support")
Signed-off-by: Ruihan Li <lrh2000@pku.edu.cn>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We started disabling '-Warray-bounds' for gcc-12 originally on s390,
because it resulted in some warnings that weren't realistically fixable
(commit 8b202ee218: "s390: disable -Warray-bounds").
That s390-specific issue was then found to be less common elsewhere, but
generic (see f0be87c42c: "gcc-12: disable '-Warray-bounds' universally
for now"), and then later expanded the version check was expanded to
gcc-11 (5a41237ad1: "gcc: disable -Warray-bounds for gcc-11 too").
And it turns out that I was much too optimistic in thinking that it's
all going to go away, and here we are with gcc-13 showing all the same
issues. So instead of expanding this one version at a time, let's just
disable it for gcc-11+, and put an end limit to it only when we actually
find a solution.
Yes, I'm sure some of this is because the kernel just does odd things
(like our "container_of()" use, but also knowingly playing games with
things like linker tables and array layouts).
And yes, some of the warnings are likely signs of real bugs, but when
there are hundreds of false positives, that doesn't really help.
Oh well.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull Kbuild fixes from Masahiro Yamada:
- Fix the prefix in the kernel source tarball
- Fix a typo in the copyright file in Debian package
* tag 'kbuild-fixes-v6.3-4' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
kbuild: use proper prefix for tarballs to fix rpm-pkg build error
kbuild: deb-pkg: Fix a spell typo in mkdebian script
Pull irq fix from Borislav Petkov:
- Remove an over-zealous sanity check of the array of MSI-X vectors to
be allocated for a device
* tag 'irq_urgent_for_v6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
PCI/MSI: Remove over-zealous hardware size check in pci_msix_validate_entries()
Pull x86 fix from Borislav Petkov
- Fix for older binutils which do not support C-syntax constant
suffixes
* tag 'x86_urgent_for_v6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/alternatives: Do not use integer constant suffixes in inline asm
Pull input fixes from Dmitry Torokhov:
- a check in pegasus-notetaker driver to validate the type of pipe when
probing a new device
- a fix for Cypress touch controller to correctly parse maximum number
of touches.
* tag 'input-for-v6.3-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
Input: cyttsp5 - fix sensing configuration data structure
Input: pegasus-notetaker - check pipe type when probing
Since commit f8d94c4e40 ("kbuild: do not create intermediate *.tar
for source tarballs"), 'make rpm-pkg' fails because the prefix of the
source tarball is 'linux.tar/' instead of 'linux/'. $(basename $@)
strips only '.gz' from the filename linux.tar.gz.
You need to strip two suffixes from compressed tarballs and one suffix
from uncompressed tarballs (for example 'perf-6.3.0.tar' generated by
'make perf-tar-src-pkg').
One tricky fix might be --prefix=$(firstword $(subst .tar, ,$@))/
but I think it is better to hard-code the prefix.
Fixes: f8d94c4e40 ("kbuild: do not create intermediate *.tar for source tarballs")
Reported-by: Jiwei Sun <sunjw10@lenovo.com>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>
Pull MIPS fix from Thomas Bogendoerfer:
"Fix for link errors"
* tag 'mips-fixes_6.3_2' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux:
MIPS: Define RUNTIME_DISCARD_EXIT in LD script
Pull kvm fixes from Paolo Bonzini:
"Two serious ARM fixes:
- Plug a buffer overflow due to the use of the user-provided register
width for firmware regs. Outright reject accesses where the user
register width does not match the kernel representation.
- Protect non-atomic RMW operations on vCPU flags against preemption,
as an update to the flags by an intervening preemption could be
lost"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: arm64: Fix buffer overflow in kvm_arm_set_fw_reg()
KVM: arm64: Make vcpu flag updates non-preemptible
Pull cifs fixes from Steve French:
"Three small smb3 client fixes:
- two important fixes for unbuffered read regression with the
iov_iter changes (e.g. read soon after mount in some multichannel
scenarios)
- DFS prefix path fix (also for stable)"
* tag '6.3-rc7-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
cifs: Reapply lost fix from commit 30b2b2196d
cifs: Fix unbuffered read
cifs: avoid dup prefix path in dfs_get_automount_devname()
KVM/arm64 fixes for 6.3, part #4
- Plug a buffer overflow due to the use of the user-provided register
width for firmware regs. Outright reject accesses where the
user register width does not match the kernel representation.
- Protect non-atomic RMW operations on vCPU flags against preemption,
as an update to the flags by an intervening preemption could be lost.
MIPS's exit sections are discarded at runtime as well.
Fixes link error:
`.exit.text' referenced in section `__jump_table' of fs/fuse/inode.o:
defined in discarded section `.exit.text' of fs/fuse/inode.o
Fixes: 99cb0d917f ("arch: fix broken BuildID for arm64 and riscv")
Reported-by: "kernelci.org bot" <bot@kernelci.org>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Pull btrfs fixes from David Sterba:
"Two patches fixing the problem with aync discard.
The default settings had a low IOPS limit and processing a large batch
to discard would take a long time. On laptops this can cause increased
power consumption due to disk activity.
As async discard has been on by default since 6.2 this likely affects
a lot of users.
Summary:
- increase the default IOPS limit 10x which reportedly helped
- setting the sysfs IOPS value to 0 now does not throttle anymore
allowing the discards to be processed at full speed. Previously
there was an arbitrary 6 hour target for processing the pending
batch"
* tag 'for-6.3-rc7-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: reinterpret async discard iops_limit=0 as no delay
btrfs: set default discard iops_limit to 1000
Pull block fix from Jens Axboe:
"Just a single revert of a patch from the 6.3 series"
* tag 'block-6.3-2023-04-21' of git://git.kernel.dk/linux:
Revert "block: Merge bio before checking ->cached_rq"
Pull char/misc driver fixes from Greg KH:
"Here are some last-minute tiny driver fixes for 6.3-final. They
include fixes for some fpga and iio drivers:
- fpga bridge driver fix
- fpga dfl error reporting fix
- fpga m10bmc driver fix
- fpga xilinx driver fix
- iio light driver fix
- iio dac fwhandle leak fix
- iio adc driver fix
All of these have been in linux-next for a few weeks with no reported
problems"
* tag 'char-misc-6.3-final' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
iio: light: tsl2772: fix reading proximity-diodes from device tree
fpga: bridge: properly initialize bridge device before populating children
iio: dac: ad5755: Add missing fwnode_handle_put()
iio: adc: at91-sama5d2_adc: fix an error code in at91_adc_allocate_trigger()
fpga: xilinx-pr-decoupler: Use readl wrapper instead of pure readl
fpga: dfl-pci: Drop redundant pci_enable_pcie_error_reporting()
fpga: m10bmc-sec: Fix rsu_send_data() to return FW_UPLOAD_ERR_HW_ERROR
Pull gpio fixes from Bartosz Golaszewski:
- use raw_spinlocks in regmaps that are used in interrupt context in
gpio-104-idi-48 and gpio-104-dio-48e
* tag 'gpio-fixes-for-v6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
gpio: 104-idi-48: Enable use_raw_spinlock for idi48_regmap_config
gpio: 104-dio-48e: Enable use_raw_spinlock for dio48e_regmap_config
Pull sound fixes from Takashi Iwai:
"Just a few fixes: all small and device-specific (ASoC FSL, SOF, and
HD-audio quirks), should be safe to apply at the last minute"
* tag 'sound-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: hda/realtek: fix mute/micmute LEDs for a HP ProBook
ASoC: fsl_asrc_dma: fix potential null-ptr-deref
ASoC: fsl_sai: Fix pins setting for i.MX8QM platform
ALSA: hda/realtek: Remove specific patch for Dell Precision 3260
ASoC: max98373: change power down sequence for smart amp
ASoC: SOF: pm: Tear down pipelines only if DSP was active
ASoC: SOF: ipc4-topology: Clarify bind failure caused by missing fw_module
Pull drm fixes from Dave Airlie:
"This is the regular and hopefully last round of fixes for 6.3.
Pretty small, a few amdgpu, one i915, one nouveau, one rockchip and
one gpu scheduler fix:
nouveau:
- fix dma-resv timeout
rockchip:
- fix suspend/resume
sched:
- fix timeout handling
i915:
- Fix fast wake AUX sync len
amdgpu:
- GPU reset fix
- DCN 3.1.5 line buffer fix
- Display fix for single channel memory configs
- Fix a possible divide by 0"
* tag 'drm-fixes-2023-04-21' of git://anongit.freedesktop.org/drm/drm:
drm/amd/display: fix a divided-by-zero error
drm/amd/display: limit timing for single dimm memory
drm/amd/display: set dcn315 lb bpp to 48
drm/amdgpu: Fix desktop freezed after gpu-reset
drm/rockchip: vop2: Use regcache_sync() to fix suspend/resume
drm/nouveau: fix incorrect conversion to dma_resv_wait_timeout()
drm/rockchip: vop2: fix suspend/resume
drm/i915: Fix fast wake AUX sync len
drm/sched: Check scheduler ready before calling timeout handling
I didn't really want to do this, but as part of all the other changes to
the user copy loops, I've been looking at this horror.
I tried to clean it up multiple times, but every time I just found more
problems, and the way it's written, it's just too hard to fix them.
For example, the code is written to do quad-word alignment, and will use
regular byte accesses to get to that point. That's fairly simple, but
it means that any initial 8-byte alignment will be done with cached
copies.
However, the code then is very careful to do any 4-byte _tail_ accesses
using an uncached 4-byte write, and that was claimed to be relevant in
commit a82eee7424 ("x86/uaccess/64: Handle the caching of 4-byte
nocache copies properly in __copy_user_nocache()").
So if you do a 4-byte copy using that function, it carefully uses a
4-byte 'movnti' for the destination. But if you were to do a 12-byte
copy that is 4-byte aligned, it would _not_ do a 4-byte 'movnti'
followed by a 8-byte 'movnti' to keep it all uncached.
Instead, it would align the destination to 8 bytes using a
byte-at-a-time loop, and then do a 8-byte 'movnti' for the final 8
bytes.
The main caller that cares is __copy_user_flushcache(), which knows
about this insanity, and has odd cases for it all. But I just can't
deal with looking at this kind of "it does one case right, and another
related case entirely wrong".
And the code really wasn't fixable without hard drugs, which I try to
avoid.
So instead, rewrite it in a form that hopefully not only gets this
right, but is a bit more maintainable. Knock wood.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull pci fix from Bjorn Helgaas:
- Previously we ignored PCI devices if the DT "status" property or the
ACPI _STA method said it was not present.
Per spec, _STA cannot be used for that purpose, and using it that way
caused regressions, so skip the _STA check (Rob Herring)
* tag 'pci-v6.3-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci:
PCI: Restrict device disabled status check to DT
Currently, a limit of 0 results in a hard coded metering over 6 hours.
Since the default is a set limit, I suspect no one truly depends on this
rather arbitrary setting. Repurpose it for an arguably more useful
"unlimited" mode, where the delay is 0.
Note that if block groups are too new, or go fully empty, there is still
a delay associated with those conditions. Those delays implement
heuristics for not trimming a region we are relatively likely to fully
overwrite soon.
CC: stable@vger.kernel.org # 6.2+
Reviewed-by: Neal Gompa <neal@gompa.dev>
Signed-off-by: Boris Burkov <boris@bur.io>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Previously, the default was a relatively conservative 10. This results
in a 100ms delay, so with ~300 discards in a commit, it takes the full
30s till the next commit to finish the discards. On a workstation, this
results in the disk never going idle, wasting power/battery, etc.
Set the default to 1000, which results in using the smallest possible
delay, currently, which is 1ms. This has shown to not pathologically
keep the disk busy by the original reporter.
Link: https://lore.kernel.org/linux-btrfs/Y%2F+n1wS%2F4XAH7X1p@nz/
Link: https://bugzilla.redhat.com/show_bug.cgi?id=2182228
CC: stable@vger.kernel.org # 6.2+
Reviewed-by: Neal Gompa <neal@gompa.dev
Signed-off-by: Boris Burkov <boris@bur.io>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Pull Rust fixes from Miguel Ojeda:
"Most of these are straightforward.
The last one is more complex, but it only touches Rust + GCC builds
which are for the moment best-effort.
- Code: Missing 'extern "C"' fix.
- Scripts: 'is_rust_module.sh' and 'generate_rust_analyzer.py' fixes.
- A couple trivial fixes
- Build: Rust + GCC build fix and 'grep' warning fix"
* tag 'rust-fixes-6.3' of https://github.com/Rust-for-Linux/linux:
rust: allow to use INIT_STACK_ALL_ZERO
rust: fix regexp in scripts/is_rust_module.sh
rust: build: Fix grep warning
scripts: generate_rust_analyzer: Handle sub-modules with no Makefile
rust: kernel: Mark rust_fmt_argument as extern "C"
rust: sort uml documentation arch support table
rust: str: fix requierments->requirements typo
Pull networking fixes from Paolo Abeni:
"Including fixes from netfilter and bpf.
There are a few fixes for new code bugs, including the Mellanox one
noted in the last networking pull. No known regressions outstanding.
Current release - regressions:
- sched: clear actions pointer in miss cookie init fail
- mptcp: fix accept vs worker race
- bpf: fix bpf_arch_text_poke() with new_addr == NULL on s390
- eth: bnxt_en: fix a possible NULL pointer dereference in unload
path
- eth: veth: take into account peer device for
NETDEV_XDP_ACT_NDO_XMIT xdp_features flag
Current release - new code bugs:
- eth: revert "net/mlx5: Enable management PF initialization"
Previous releases - regressions:
- netfilter: fix recent physdev match breakage
- bpf: fix incorrect verifier pruning due to missing register
precision taints
- eth: virtio_net: fix overflow inside xdp_linearize_page()
- eth: cxgb4: fix use after free bugs caused by circular dependency
problem
- eth: mlxsw: pci: fix possible crash during initialization
Previous releases - always broken:
- sched: sch_qfq: prevent slab-out-of-bounds in qfq_activate_agg
- netfilter: validate catch-all set elements
- bridge: don't notify FDB entries with "master dynamic"
- eth: bonding: fix memory leak when changing bond type to ethernet
- eth: i40e: fix accessing vsi->active_filters without holding lock
Misc:
- Mat is back as MPTCP co-maintainer"
* tag 'net-6.3-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (33 commits)
net: bridge: switchdev: don't notify FDB entries with "master dynamic"
Revert "net/mlx5: Enable management PF initialization"
MAINTAINERS: Resume MPTCP co-maintainer role
mailmap: add entries for Mat Martineau
e1000e: Disable TSO on i219-LM card to increase speed
bnxt_en: fix free-runnig PHC mode
net: dsa: microchip: ksz8795: Correctly handle huge frame configuration
bpf: Fix incorrect verifier pruning due to missing register precision taints
hamradio: drop ISA_DMA_API dependency
mlxsw: pci: Fix possible crash during initialization
mptcp: fix accept vs worker race
mptcp: stops worker on unaccepted sockets at listener close
net: rpl: fix rpl header size calculation
net: vmxnet3: Fix NULL pointer dereference in vmxnet3_rq_rx_complete()
bonding: Fix memory leak when changing bond type to Ethernet
veth: take into account peer device for NETDEV_XDP_ACT_NDO_XMIT xdp_features flag
mlxfw: fix null-ptr-deref in mlxfw_mfa2_tlv_next()
bnxt_en: Fix a possible NULL pointer dereference in unload path
bnxt_en: Do not initialize PTP on older P3/P4 chips
netfilter: nf_tables: tighten netlink attribute requirements for catch-all elements
...
There is a structural problem in switchdev, where the flag bits in
struct switchdev_notifier_fdb_info (added_by_user, is_local etc) only
represent a simplified / denatured view of what's in struct
net_bridge_fdb_entry :: flags (BR_FDB_ADDED_BY_USER, BR_FDB_LOCAL etc).
Each time we want to pass more information about struct
net_bridge_fdb_entry :: flags to struct switchdev_notifier_fdb_info
(here, BR_FDB_STATIC), we find that FDB entries were already notified to
switchdev with no regard to this flag, and thus, switchdev drivers had
no indication whether the notified entries were static or not.
For example, this command:
ip link add br0 type bridge && ip link set swp0 master br0
bridge fdb add dev swp0 00:01:02:03:04:05 master dynamic
has never worked as intended with switchdev. It causes a struct
net_bridge_fdb_entry to be passed to br_switchdev_fdb_notify() which has
a single flag set: BR_FDB_ADDED_BY_USER.
This is further passed to the switchdev notifier chain, where interested
drivers have no choice but to assume this is a static (does not age) and
sticky (does not migrate) FDB entry. So currently, all drivers offload
it to hardware as such, as can be seen below ("offload" is set).
bridge fdb get 00:01:02:03:04:05 dev swp0 master
00:01:02:03:04:05 dev swp0 offload master br0
The software FDB entry expires $ageing_time centiseconds after the
kernel last sees a packet with this MAC SA, and the bridge notifies its
deletion as well, so it eventually disappears from hardware too.
This is a problem, because it is actually desirable to start offloading
"master dynamic" FDB entries correctly - they should expire $ageing_time
centiseconds after the *hardware* port last sees a packet with this
MAC SA - and this is how the current incorrect behavior was discovered.
With an offloaded data plane, it can be expected that software only sees
exception path packets, so an otherwise active dynamic FDB entry would
be aged out by software sooner than it should.
With the change in place, these FDB entries are no longer offloaded:
bridge fdb get 00:01:02:03:04:05 dev swp0 master
00:01:02:03:04:05 dev swp0 master br0
and this also constitutes a better way (assuming a backport to stable
kernels) for user space to determine whether the kernel has the
capability of doing something sane with these or not.
As opposed to "master dynamic" FDB entries, on the current behavior of
which no one currently depends on (which can be deduced from the lack of
kselftests), Ido Schimmel explains that entries with the "extern_learn"
flag (BR_FDB_ADDED_BY_EXT_LEARN) should still be notified to switchdev,
since the spectrum driver listens to them (and this is kind of okay,
because although they are treated identically to "static", they are
expected to not age, and to roam).
Fixes: 6b26b51b1d ("net: bridge: Add support for notifying devices about FDB add/del")
Link: https://lore.kernel.org/netdev/20230327115206.jk5q5l753aoelwus@skbuf/
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Tested-by: Ido Schimmel <idosch@nvidia.com>
Link: https://lore.kernel.org/r/20230418155902.898627-1-vladimir.oltean@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Every caller passes in zero, meaning they don't want any partial copy to
zero the remainder of the destination buffer.
Which is just as well, because the implementation of that function
didn't actually even look at that argument, and wasn't even aware it
existed, although some misleading comments did mention it still.
The 'zerorest' thing is a historical artifact of how "copy_from_user()"
worked, in that it would zero the rest of the kernel buffer that it
copied into.
That zeroing still exists, but it's long since been moved to generic
code, and the raw architecture-specific code doesn't do it. See
_copy_from_user() in lib/usercopy.c for this all.
However, while __copy_user_nocache() shares some history and superficial
other similarities with copy_from_user(), it is in many ways also very
different.
In particular, while the code makes it *look* similar to the generic
user copy functions that can copy both to and from user space, and take
faults on both reads and writes as a result, __copy_user_nocache() does
no such thing at all.
__copy_user_nocache() always copies to kernel space, and will never take
a page fault on the destination. What *can* happen, though, is that the
non-temporal stores take a machine check because one of the use cases is
for writing to stable memory, and any memory errors would then take
synchronous faults.
So __copy_user_nocache() does look a lot like copy_from_user(), but has
faulting behavior that is more akin to our old copy_in_user() (which no
longer exists, but copied from user space to user space and could fault
on both source and destination).
And it very much does not have the "zero the end of the destination
buffer", since a problem with the destination buffer is very possibly
the very source of the partial copy.
So this whole thing was just a confusing historical artifact from having
shared some code with a completely different function with completely
different use cases.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Daniel Borkmann says:
====================
bpf 2023-04-19
We've added 3 non-merge commits during the last 6 day(s) which contain
a total of 3 files changed, 34 insertions(+), 9 deletions(-).
The main changes are:
1) Fix a crash on s390's bpf_arch_text_poke() under a NULL new_addr,
from Ilya Leoshkevich.
2) Fix a bug in BPF verifier's precision tracker, from Daniel Borkmann
and Andrii Nakryiko.
3) Fix a regression in veth's xdp_features which led to a broken BPF CI
selftest, from Lorenzo Bianconi.
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
bpf: Fix incorrect verifier pruning due to missing register precision taints
veth: take into account peer device for NETDEV_XDP_ACT_NDO_XMIT xdp_features flag
s390/bpf: Fix bpf_arch_text_poke() with new_addr == NULL
====================
Link: https://lore.kernel.org/r/20230419195847.27060-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2023-04-17 (i40e)
This series contains updates to i40e only.
Alex moves setting of active filters to occur under lock and checks/takes
error path in rebuild if re-initializing the misc interrupt vector
failed.
* '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue:
i40e: fix i40e_setup_misc_vector() error handling
i40e: fix accessing vsi->active_filters without holding lock
====================
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230417205245.1030733-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Pull misc fixes from Andrew Morton:
"22 hotfixes.
19 are cc:stable and the remainder address issues which were
introduced during this merge cycle, or aren't considered suitable for
-stable backporting.
19 are for MM and the remainder are for other subsystems"
* tag 'mm-hotfixes-stable-2023-04-19-16-36' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (22 commits)
nilfs2: initialize unused bytes in segment summary blocks
mm: page_alloc: skip regions with hugetlbfs pages when allocating 1G pages
mm/mmap: regression fix for unmapped_area{_topdown}
maple_tree: fix mas_empty_area() search
maple_tree: make maple state reusable after mas_empty_area_rev()
mm: kmsan: handle alloc failures in kmsan_ioremap_page_range()
mm: kmsan: handle alloc failures in kmsan_vmap_pages_range_noflush()
tools/Makefile: do missed s/vm/mm/
mm: fix memory leak on mm_init error handling
mm/page_alloc: fix potential deadlock on zonelist_update_seq seqlock
kernel/sys.c: fix and improve control flow in __sys_setres[ug]id()
Revert "userfaultfd: don't fail on unrecognized features"
writeback, cgroup: fix null-ptr-deref write in bdi_split_work_to_wbs
maple_tree: fix a potential memory leak, OOB access, or other unpredictable bug
tools/mm/page_owner_sort.c: fix TGID output when cull=tg is used
mailmap: update jtoppins' entry to reference correct email
mm/mempolicy: fix use-after-free of VMA iterator
mm/huge_memory.c: warn with pr_warn_ratelimited instead of VM_WARN_ON_ONCE_FOLIO
mm/mprotect: fix do_mprotect_pkey() return on error
mm/khugepaged: check again on anon uffd-wp during isolation
...
