Jiri Olsa [Fri, 12 Jul 2024 13:52:27 +0000 (15:52 +0200)]
uprobe: Change uretprobe syscall scope and number
After discussing with Arnd [1] it's preferable to change uretprobe
syscall number to 467 to omit the merge conflict with xattrat syscalls.
Also changing the ABI to 'common' which will ease up the global
scripts/syscall.tbl management. One consequence is we generate uretprobe
syscall numbers for ABIs that do not support uretprobe syscall, but the
syscall still returns -ENOSYS when called in that ABI.
tracing/kprobes: Fix build error when find_module() is not available
The kernel test robot reported that the find_module() is not available
if CONFIG_MODULES=n.
Fix this error by hiding find_modules() in #ifdef CONFIG_MODULES with
related rcu locks as try_module_get_by_name().
tracing/kprobes: Add symbol counting check when module loads
Currently, kprobe event checks whether the target symbol name is unique
or not, so that it does not put a probe on an unexpected place. But this
skips the check if the target is on a module because the module may not
be loaded.
To fix this issue, this patch checks the number of probe target symbols
in a target module when the module is loaded. If the probe is not on the
unique name symbols in the module, it will be rejected at that point.
Note that the symbol which has a unique name in the target module,
it will be accepted even if there are same-name symbols in the
kernel or other modules,
Andrii Nakryiko [Wed, 22 May 2024 01:38:45 +0000 (18:38 -0700)]
selftests/bpf: add test validating uprobe/uretprobe stack traces
Add a set of tests to validate that stack traces captured from or in the
presence of active uprobes and uretprobes are valid and complete.
For this we use BPF program that are installed either on entry or exit
of user function, plus deep-nested USDT. One of target funtions
(target_1) is recursive to generate two different entries in the stack
trace for the same uprobe/uretprobe, testing potential edge conditions.
If there is no fixes, we get something like this for one of the scenarios:
Entry #3 and #4 (0x7fffffffe000) are uretprobe trampoline addresses
which obscure actual target_1 and another target_1 invocations. Also
note that between entry #0 and entry #1 we are missing an entry for
target_3.
With fixes, we get desired full stack traces:
caller: 0x758fff - 0x7595ab
target_1: 0x758fd5 - 0x758fff
target_2: 0x758fca - 0x758fd5
target_3: 0x758fbf - 0x758fca
target_4: 0x758fb3 - 0x758fbf
ENTRY #0: 0x758fb7 (in target_4)
ENTRY #1: 0x758fc8 (in target_3)
ENTRY #2: 0x758fd3 (in target_2)
ENTRY #3: 0x758ffd (in target_1)
ENTRY #4: 0x758ff3 (in target_1)
ENTRY #5: 0x75922c (in caller)
ENTRY #6: 0x6f8f39
ENTRY #7: 0x6fa6f0
ENTRY #8: 0x7f986adc4cd0
Now there is a logical and complete sequence of function calls.
Andrii Nakryiko [Wed, 22 May 2024 01:38:43 +0000 (18:38 -0700)]
perf,uprobes: fix user stack traces in the presence of pending uretprobes
When kernel has pending uretprobes installed, it hijacks original user
function return address on the stack with a uretprobe trampoline
address. There could be multiple such pending uretprobes (either on
different user functions or on the same recursive one) at any given
time within the same task.
This approach interferes with the user stack trace capture logic, which
would report suprising addresses (like 0x7fffffffe000) that correspond
to a special "[uprobes]" section that kernel installs in the target
process address space for uretprobe trampoline code, while logically it
should be an address somewhere within the calling function of another
traced user function.
This is easy to correct for, though. Uprobes subsystem keeps track of
pending uretprobes and records original return addresses. This patch is
using this to do a post-processing step and restore each trampoline
address entries with correct original return address. This is done only
if there are pending uretprobes for current task.
This is a similar approach to what fprobe/kretprobe infrastructure is
doing when capturing kernel stack traces in the presence of pending
return probes.
tracing/kprobe: Remove cleanup code unrelated to selftest
This cleanup all kprobe events code is not related to the selftest
itself, and it can fail by the reason unrelated to this test.
If the test is successful, the generated events are cleaned up.
And if not, we cannot guarantee that the kprobe events will work
correctly. So, anyway, there is no need to clean it up.
tracing/kprobe: Integrate test warnings into WARN_ONCE
Cleanup the redundant WARN_ON_ONCE(cond) + pr_warn(msg) into
WARN_ONCE(cond, msg). Also add some WARN_ONCE() for hitcount check.
These WARN_ONCE() errors makes it easy to handle errors from ktest.
Jiri Olsa [Tue, 11 Jun 2024 23:44:28 +0000 (08:44 +0900)]
selftests/bpf: Add uretprobe syscall call from user space test
Adding test to verify that when called from outside of the
trampoline provided by kernel, the uretprobe syscall will cause
calling process to receive SIGILL signal and the attached bpf
program is not executed.
Jiri Olsa [Tue, 11 Jun 2024 23:44:28 +0000 (08:44 +0900)]
selftests/bpf: Add uretprobe syscall test for regs changes
Adding test that creates uprobe consumer on uretprobe which changes some
of the registers. Making sure the changed registers are propagated to the
user space when the ureptobe syscall trampoline is used on x86_64.
To be able to do this, adding support to bpf_testmod to create uprobe via
new attribute file:
/sys/kernel/bpf_testmod_uprobe
This file is expecting file offset and creates related uprobe on current
process exe file and removes existing uprobe if offset is 0. The can be
only single uprobe at any time.
The uprobe has specific consumer that changes registers used in ureprobe
syscall trampoline and which are later checked in the test.
Jiri Olsa [Tue, 11 Jun 2024 23:44:28 +0000 (08:44 +0900)]
selftests/bpf: Add uretprobe syscall test for regs integrity
Add uretprobe syscall test that compares register values before
and after the uretprobe is hit. It also compares the register
values seen from attached bpf program.
Jiri Olsa [Tue, 11 Jun 2024 23:44:28 +0000 (08:44 +0900)]
uprobe: Add uretprobe syscall to speed up return probe
Adding uretprobe syscall instead of trap to speed up return probe.
At the moment the uretprobe setup/path is:
- install entry uprobe
- when the uprobe is hit, it overwrites probed function's return address
on stack with address of the trampoline that contains breakpoint
instruction
- the breakpoint trap code handles the uretprobe consumers execution and
jumps back to original return address
This patch replaces the above trampoline's breakpoint instruction with new
ureprobe syscall call. This syscall does exactly the same job as the trap
with some more extra work:
- syscall trampoline must save original value for rax/r11/rcx registers
on stack - rax is set to syscall number and r11/rcx are changed and
used by syscall instruction
- the syscall code reads the original values of those registers and
restore those values in task's pt_regs area
- only caller from trampoline exposed in '[uprobes]' is allowed,
the process will receive SIGILL signal otherwise
Even with some extra work, using the uretprobes syscall shows speed
improvement (compared to using standard breakpoint):
On Intel (11th Gen Intel(R) Core(TM) i7-1165G7 @ 2.80GHz)
with the fix:
uretprobe-nop : 0.860 ± 0.001M/s < 10% speed up
uretprobe-push : 0.818 ± 0.001M/s < 10% speed up
uretprobe-ret : 0.578 ± 0.000M/s < 7% speed up
The performance test spawns a thread that runs loop which triggers
uprobe with attached bpf program that increments the counter that
gets printed in results above.
The uprobe (and uretprobe) kind is determined by which instruction
is being patched with breakpoint instruction. That's also important
for uretprobes, because uprobe is installed for each uretprobe.
The performance test is part of bpf selftests:
tools/testing/selftests/bpf/run_bench_uprobes.sh
Note at the moment uretprobe syscall is supported only for native
64-bit process, compat process still uses standard breakpoint.
Note that when shadow stack is enabled the uretprobe syscall returns
via iret, which is slower than return via sysret, but won't cause the
shadow stack violation.
Jiri Olsa [Tue, 11 Jun 2024 23:44:27 +0000 (08:44 +0900)]
uprobe: Wire up uretprobe system call
Wiring up uretprobe system call, which comes in following changes.
We need to do the wiring before, because the uretprobe implementation
needs the syscall number.
Note at the moment uretprobe syscall is supported only for native
64-bit process.
Jiri Olsa [Tue, 11 Jun 2024 23:44:27 +0000 (08:44 +0900)]
x86/shstk: Make return uprobe work with shadow stack
Currently the application with enabled shadow stack will crash
if it sets up return uprobe. The reason is the uretprobe kernel
code changes the user space task's stack, but does not update
shadow stack accordingly.
Adding new functions to update values on shadow stack and using
them in uprobe code to keep shadow stack in sync with uretprobe
changes to user stack.
make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in samples/kprobes/kprobe_example.o
WARNING: modpost: missing MODULE_DESCRIPTION() in samples/kprobes/kretprobe_example.o
Add the missing invocations of the MODULE_DESCRIPTION() macro.
tracing: Build event generation tests only as modules
The kprobes and synth event generation test modules add events and lock
(get a reference) those event file reference in module init function,
and unlock and delete it in module exit function. This is because those
are designed for playing as modules.
If we make those modules as built-in, those events are left locked in the
kernel, and never be removed. This causes kprobe event self-test failure
as below.
Linus Torvalds [Sun, 9 Jun 2024 16:04:51 +0000 (09:04 -0700)]
Merge tag 'perf-tools-fixes-for-v6.10-2-2024-06-09' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools
Pull perf tools fixes from Arnaldo Carvalho de Melo:
- Update copies of kernel headers, which resulted in support for the
new 'mseal' syscall, SUBVOL statx return mask bit, RISC-V and PPC
prctls, fcntl's DUPFD_QUERY, POSTED_MSI_NOTIFICATION IRQ vector,
'map_shadow_stack' syscall for x86-32.
- Revert perf.data record memory allocation optimization that ended up
causing a regression, work is being done to re-introduce it in the
next merge window.
- Fix handling of minimal vmlinux.h file used with BPF's CO-RE when
interrupting the build.
* tag 'perf-tools-fixes-for-v6.10-2-2024-06-09' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools:
perf bpf: Fix handling of minimal vmlinux.h file when interrupting the build
Revert "perf record: Reduce memory for recording PERF_RECORD_LOST_SAMPLES event"
tools headers arm64: Sync arm64's cputype.h with the kernel sources
tools headers uapi: Sync linux/stat.h with the kernel sources to pick STATX_SUBVOL
tools headers UAPI: Update i915_drm.h with the kernel sources
tools headers UAPI: Sync kvm headers with the kernel sources
tools arch x86: Sync the msr-index.h copy with the kernel sources
tools headers: Update the syscall tables and unistd.h, mostly to support the new 'mseal' syscall
perf trace beauty: Update the arch/x86/include/asm/irq_vectors.h copy with the kernel sources to pick POSTED_MSI_NOTIFICATION
perf beauty: Update copy of linux/socket.h with the kernel sources
tools headers UAPI: Sync fcntl.h with the kernel sources to pick F_DUPFD_QUERY
tools headers UAPI: Sync linux/prctl.h with the kernel sources
tools include UAPI: Sync linux/stat.h with the kernel sources
Linus Torvalds [Sun, 9 Jun 2024 15:49:13 +0000 (08:49 -0700)]
Merge tag 'edac_urgent_for_v6.10_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras
Pull EDAC fixes from Borislav Petkov:
- Convert PCI core error codes to proper error numbers since latter get
propagated all the way up to the module loading functions
* tag 'edac_urgent_for_v6.10_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:
EDAC/igen6: Convert PCIBIOS_* return codes to errnos
EDAC/amd64: Convert PCIBIOS_* return codes to errnos
Linus Torvalds [Sun, 9 Jun 2024 02:14:02 +0000 (19:14 -0700)]
Merge tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux
Pull clk fix from Stephen Boyd:
"One fix for the SiFive PRCI clocks so that the device boots again.
This driver was registering clkdev lookups that were always going to
be useless. This wasn't a problem until clkdev started returning an
error in these cases, causing this driver to fail probe, and thus boot
to fail because clks are essential for most drivers. The fix is
simple, don't use clkdev because this is a DT based system where
clkdev isn't used"
* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
clk: sifive: Do not register clkdevs for PRCI clocks
Linus Torvalds [Sun, 9 Jun 2024 02:07:18 +0000 (19:07 -0700)]
Merge tag '6.10-rc2-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6
Pull smb client fixes from Steve French:
"Two small smb3 client fixes:
- fix deadlock in umount
- minor cleanup due to netfs change"
* tag '6.10-rc2-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
cifs: Don't advance the I/O iterator before terminating subrequest
smb: client: fix deadlock in smb2_find_smb_tcon()
Linus Torvalds [Sat, 8 Jun 2024 17:12:33 +0000 (10:12 -0700)]
Merge tag 'kbuild-fixes-v6.10-2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild
Pull Kbuild fixes from Masahiro Yamada:
- Fix the initial state of the save button in 'make gconfig'
- Improve the Kconfig documentation
- Fix a Kconfig bug regarding property visibility
- Fix build breakage for systems where 'sed' is not installed in /bin
- Fix a false warning about missing MODULE_DESCRIPTION()
* tag 'kbuild-fixes-v6.10-2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
modpost: do not warn about missing MODULE_DESCRIPTION() for vmlinux.o
kbuild: explicitly run mksysmap as sed script from link-vmlinux.sh
kconfig: remove wrong expr_trans_bool()
kconfig: doc: document behavior of 'select' and 'imply' followed by 'if'
kconfig: doc: fix a typo in the note about 'imply'
kconfig: gconf: give a proper initial state to the Save button
kconfig: remove unneeded code for user-supplied values being out of range
Linus Torvalds [Sat, 8 Jun 2024 16:57:09 +0000 (09:57 -0700)]
Merge tag 'media/v6.10-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
Pull media fixes from Mauro Carvalho Chehab:
- fixes for the new ipu6 driver (and related fixes to mei csi driver)
- fix a double debugfs remove logic at mgb4 driver
- a documentation fix
* tag 'media/v6.10-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:
media: intel/ipu6: add csi2 port sanity check in notifier bound
media: intel/ipu6: update the maximum supported csi2 port number to 6
media: mei: csi: Warn less verbosely of a missing device fwnode
media: mei: csi: Put the IPU device reference
media: intel/ipu6: fix the buffer flags caused by wrong parentheses
media: intel/ipu6: Fix an error handling path in isys_probe()
media: intel/ipu6: Move isys_remove() close to isys_probe()
media: intel/ipu6: Fix some redundant resources freeing in ipu6_pci_remove()
media: Documentation: v4l: Fix ACTIVE route flag
media: mgb4: Fix double debugfs remove
Linus Torvalds [Sat, 8 Jun 2024 16:44:50 +0000 (09:44 -0700)]
Merge tag 'irq-urgent-2024-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull irq fixes from Ingo Molnar:
- Fix possible memory leak the riscv-intc irqchip driver load failures
- Fix boot crash in the sifive-plic irqchip driver caused by recently
changed boot initialization order
- Fix race condition in the gic-v3-its irqchip driver
* tag 'irq-urgent-2024-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
irqchip/gic-v3-its: Fix potential race condition in its_vlpi_prop_update()
irqchip/sifive-plic: Chain to parent IRQ after handlers are ready
irqchip/riscv-intc: Prevent memory leak when riscv_intc_init_common() fails
Linus Torvalds [Sat, 8 Jun 2024 16:36:08 +0000 (09:36 -0700)]
Merge tag 'x86-urgent-2024-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Ingo Molnar:
"Miscellaneous fixes:
- Fix kexec() crash if call depth tracking is enabled
- Fix SMN reads on inaccessible registers on certain AMD systems"
* tag 'x86-urgent-2024-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/amd_nb: Check for invalid SMN reads
x86/kexec: Fix bug with call depth tracking
Linus Torvalds [Sat, 8 Jun 2024 16:26:59 +0000 (09:26 -0700)]
Merge tag 'perf-urgent-2024-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf event fix from Ingo Molnar:
"Fix race between perf_event_free_task() and perf_event_release_kernel()
that can result in missed wakeups and hung tasks"
* tag 'perf-urgent-2024-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/core: Fix missing wakeup when waiting for context reference
Linus Torvalds [Sat, 8 Jun 2024 16:03:46 +0000 (09:03 -0700)]
Merge tag 'locking-urgent-2024-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull locking doc fix from Ingo Molnar:
"Fix typos in the kerneldoc of some of the atomic APIs"
* tag 'locking-urgent-2024-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
locking/atomic: scripts: fix ${atomic}_sub_and_test() kerneldoc
Linus Torvalds [Sat, 8 Jun 2024 00:01:10 +0000 (17:01 -0700)]
Merge tag 'mm-hotfixes-stable-2024-06-07-15-24' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull misc fixes from Andrew Morton:
"14 hotfixes, 6 of which are cc:stable.
All except the nilfs2 fix affect MM and all are singletons - see the
chagelogs for details"
* tag 'mm-hotfixes-stable-2024-06-07-15-24' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
nilfs2: fix nilfs_empty_dir() misjudgment and long loop on I/O errors
mm: fix xyz_noprof functions calling profiled functions
codetag: avoid race at alloc_slab_obj_exts
mm/hugetlb: do not call vma_add_reservation upon ENOMEM
mm/ksm: fix ksm_zero_pages accounting
mm/ksm: fix ksm_pages_scanned accounting
kmsan: do not wipe out origin when doing partial unpoisoning
vmalloc: check CONFIG_EXECMEM in is_vmalloc_or_module_addr()
mm: page_alloc: fix highatomic typing in multi-block buddies
nilfs2: fix potential kernel bug due to lack of writeback flag waiting
memcg: remove the lockdep assert from __mod_objcg_mlstate()
mm: arm64: fix the out-of-bounds issue in contpte_clear_young_dirty_ptes
mm: huge_mm: fix undefined reference to `mthp_stats' for CONFIG_SYSFS=n
mm: drop the 'anon_' prefix for swap-out mTHP counters
Linus Torvalds [Fri, 7 Jun 2024 23:54:57 +0000 (16:54 -0700)]
Merge tag 'gpio-fixes-for-v6.10-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux
Pull gpio fixes from Bartosz Golaszewski:
- interrupt handling and Kconfig fixes for gpio-tqmx86
- add a buffer for storing output values in gpio-tqmx86 as reading back
the registers always returns the input values
- add missing MODULE_DESCRIPTION()s to several GPIO drivers
* tag 'gpio-fixes-for-v6.10-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
gpio: add missing MODULE_DESCRIPTION() macros
gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type
gpio: tqmx86: store IRQ trigger type and unmask status separately
gpio: tqmx86: introduce shadow register for GPIO output value
gpio: tqmx86: fix typo in Kconfig label
Linus Torvalds [Fri, 7 Jun 2024 23:45:48 +0000 (16:45 -0700)]
Merge tag 'block-6.10-20240607' of git://git.kernel.dk/linux
Pull block fixes from Jens Axboe:
- Fix for null_blk block size validation (Andreas)
- NVMe pull request via Keith:
- Use reserved tags for special fabrics operations (Chunguang)
- Persistent Reservation status masking fix (Weiwen)
* tag 'block-6.10-20240607' of git://git.kernel.dk/linux:
null_blk: fix validation of block size
nvme: fix nvme_pr_* status code parsing
nvme-fabrics: use reserved tag for reg read/write command
Linus Torvalds [Fri, 7 Jun 2024 23:43:07 +0000 (16:43 -0700)]
Merge tag 'io_uring-6.10-20240607' of git://git.kernel.dk/linux
Pull io_uring fixes from Jens Axboe:
- Fix a locking order issue with setting max async thread workers
(Hagar)
- Fix for a NULL pointer dereference for failed async flagged requests
using ring provided buffers. This doesn't affect the current kernel,
but it does affect older kernels, and is being queued up for 6.10
just to make the stable process easier (me)
- Fix for NAPI timeout calculations for how long to busy poll, and
subsequently how much to sleep post that if a wait timeout is passed
in (me)
- Fix for a regression in this release cycle, where we could end up
using a partially unitialized match value for io-wq (Su)
* tag 'io_uring-6.10-20240607' of git://git.kernel.dk/linux:
io_uring: fix possible deadlock in io_register_iowq_max_workers()
io_uring/io-wq: avoid garbage value of 'match' in io_wq_enqueue()
io_uring/napi: fix timeout calculation
io_uring: check for non-NULL file pointer in io_file_can_poll()
Linus Torvalds [Fri, 7 Jun 2024 22:13:12 +0000 (15:13 -0700)]
Merge tag 'for-6.10-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba:
- fix handling of folio private changes.
The private value holds pointer to our extent buffer structure
representing a metadata range. Release and create of the range was
not properly synchronized when updating the private bit which ended
up in double folio_put, leading to all sorts of breakage
- fix a crash, reported as duplicate key in metadata, but caused by a
race of fsync and size extending write. Requires prealloc target
range + fsync and other conditions (log tree state, timing)
- fix leak of qgroup extent records after transaction abort
* tag 'for-6.10-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: protect folio::private when attaching extent buffer folios
btrfs: fix leak of qgroup extent records after transaction abort
btrfs: fix crash on racing fsync and size-extending write into prealloc
Linus Torvalds [Fri, 7 Jun 2024 21:47:38 +0000 (14:47 -0700)]
Merge tag 'riscv-for-linus-6.10-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux
Pull RISC-V fixes from Palmer Dabbelt:
- Another fix to avoid allocating pages that overlap with ERR_PTR,
which manifests on rv32
- A revert for the badaccess patch I incorrectly picked up an early
version of
* tag 'riscv-for-linus-6.10-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
Revert "riscv: mm: accelerate pagefault when badaccess"
riscv: fix overlap of allocated page and PTR_ERR
Linus Torvalds [Fri, 7 Jun 2024 21:44:53 +0000 (14:44 -0700)]
Merge tag 's390-6.10-3' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Pull s390 fixes from Alexander Gordeev:
- Do not create PT_LOAD program header for the kenel image when the
virtual memory informaton in OS_INFO data is not available. That
fixes stand-alone dump failures against kernels that do not provide
the virtual memory informaton
- Add KVM s390 shared zeropage selftest
* tag 's390-6.10-3' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
KVM: s390x: selftests: Add shared zeropage test
s390/crash: Do not use VM info if os_info does not have it
Linus Torvalds [Fri, 7 Jun 2024 21:36:57 +0000 (14:36 -0700)]
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Will Deacon:
- Fix spurious CPU hotplug warning message from SETEND emulation code
- Fix the build when GCC wasn't inlining our I/O accessor internals
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64/io: add constant-argument check
arm64: armv8_deprecated: Fix warning in isndep cpuhp starting process
Linus Torvalds [Fri, 7 Jun 2024 21:13:46 +0000 (14:13 -0700)]
Merge tag 'platform-drivers-x86-v6.10-3' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86
Pull x86 platform driver fixes from Hans de Goede:
- Default silead touchscreen driver to 10 fingers and drop 10 finger
setting from all DMI quirks. More of a cleanup then a pure fix, but
since the DMI quirks always get updated through the fixes branch
this avoids conflicts.
- Kconfig fix for randconfig builds
- dell-smbios: Fix wrong token data in sysfs
- amd-hsmp: Fix driver poking unsupported hw when loaded manually
* tag 'platform-drivers-x86-v6.10-3' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86:
platform/x86/amd/hsmp: Check HSMP support on AMD family of processors
platform/x86: dell-smbios: Simplify error handling
platform/x86: dell-smbios: Fix wrong token data in sysfs
platform/x86: yt2-1380: add CONFIG_EXTCON dependency
platform/x86: touchscreen_dmi: Use 2-argument strscpy()
platform/x86: touchscreen_dmi: Drop "silead,max-fingers" property
Input: silead - Always support 10 fingers
Linus Torvalds [Fri, 7 Jun 2024 20:34:53 +0000 (13:34 -0700)]
Merge tag 'iommu-fixes-v6.10-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Pull iommu fixes from Joerg Roedel:
"Core:
- Make iommu-dma code recognize 'force_aperture' again
- Fix for potential NULL-ptr dereference from iommu_sva_bind_device()
return value
AMD IOMMU fixes:
- Fix lockdep splat for invalid wait context
- Add feature bit check before enabling PPR
- Make workqueue name fit into buffer
- Fix memory leak in sysfs code"
* tag 'iommu-fixes-v6.10-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
iommu/amd: Fix Invalid wait context issue
iommu/amd: Check EFR[EPHSup] bit before enabling PPR
iommu/amd: Fix workqueue name
iommu: Return right value in iommu_sva_bind_device()
iommu/dma: Fix domain init
iommu/amd: Fix sysfs leak in iommu init
vmwgfx:
- filter modes greater than available graphics memory
- fix 3D vs STDU enable
- remove STDU logic from mode valid
- logging fix
- memcmp pointers fix
- remove unused struct
- screen target lifetime fix
komeda:
- unused struct removal"
* tag 'drm-fixes-2024-06-07' of https://gitlab.freedesktop.org/drm/kernel:
drm/vmwgfx: Don't memcmp equivalent pointers
drm/vmwgfx: remove unused struct 'vmw_stdu_dma'
drm/vmwgfx: Don't destroy Screen Target when CRTC is enabled but inactive
drm/vmwgfx: Standardize use of kibibytes when logging
drm/vmwgfx: Remove STDU logic from generic mode_valid function
drm/vmwgfx: 3D disabled should not effect STDU memory limits
drm/vmwgfx: Filter modes which exceed graphics memory
drm/amdgpu/pptable: Fix UBSAN array-index-out-of-bounds
drm/amd: Fix shutdown (again) on some SMU v13.0.4/11 platforms
drm/xe/pf: Update the LMTT when freeing VF GT config
drm/panel: sitronix-st7789v: Add check for of_drm_get_panel_orientation
drm/komeda: remove unused struct 'gamma_curve_segment'
Johan Hovold [Tue, 7 May 2024 14:48:18 +0000 (16:48 +0200)]
HID: i2c-hid: elan: fix reset suspend current leakage
The Elan eKTH5015M touch controller found on the Lenovo ThinkPad X13s
shares the VCC33 supply with other peripherals that may remain powered
during suspend (e.g. when enabled as wakeup sources).
The reset line is also wired so that it can be left deasserted when the
supply is off.
This is important as it avoids holding the controller in reset for
extended periods of time when it remains powered, which can lead to
increased power consumption, and also avoids leaking current through the
X13s reset circuitry during suspend (and after driver unbind).
Use the new 'no-reset-on-power-off' devicetree property to determine
when reset needs to be asserted on power down.
Notably this also avoids wasting power on machine variants without a
touchscreen for which the driver would otherwise exit probe with reset
asserted.
When the power supply is shared with other peripherals the reset line
can be wired in such a way that it can remain deasserted regardless of
whether the supply is on or not.
This is important as it can be used to avoid holding the controller in
reset for extended periods of time when it remains powered, something
which can lead to increased power consumption. Leaving reset deasserted
also avoids leaking current through the reset circuitry pull-up
resistors.
Add a new 'no-reset-on-power-off' devicetree property which can be used
by the OS to determine when reset needs to be asserted on power down.
Note that this property can also be used when the supply cannot be
turned off by the OS at all.
Johan Hovold [Tue, 7 May 2024 14:48:16 +0000 (16:48 +0200)]
dt-bindings: HID: i2c-hid: elan: add Elan eKTH5015M
Add a compatible string for the Elan eKTH5015M touch controller.
Judging from the current binding and commit bd3cba00dcc6 ("HID: i2c-hid:
elan: Add support for Elan eKTH6915 i2c-hid touchscreens"), eKTH5015M
appears to be compatible with eKTH6915. Notably the power-on sequence is
the same.
While at it, drop a redundant label from the example.
The Ilitek ILI2901 touch screen controller was apparently incorrectly
added to the Elan eKTH6915 schema simply because it also has a reset
gpio and is currently managed by the Elan driver in Linux.
The two controllers are not related even if an unfortunate wording in
the commit message adding the Ilitek compatible made it sound like they
were.
Add a dedicated schema for the ILI2901 which does not specify the I2C
address (which is likely 0x41 rather than 0x10 as for other Ilitek touch
controllers) to avoid cluttering the Elan schema with unrelated devices
and to make it easier to find the correct schema when adding further
Ilitek controllers.
Aseda Aboagye [Tue, 4 Jun 2024 23:16:32 +0000 (23:16 +0000)]
input: Add support for "Do Not Disturb"
HUTRR94 added support for a new usage titled "System Do Not Disturb"
which toggles a system-wide Do Not Disturb setting. This commit simply
adds a new event code for the usage.
Aseda Aboagye [Tue, 4 Jun 2024 23:10:47 +0000 (23:10 +0000)]
input: Add event code for accessibility key
HUTRR116 added support for a new usage titled "System Accessibility
Binding" which toggles a system-wide bound accessibility UI or command.
This commit simply adds a new event code for the usage.
Jeff Johnson [Fri, 7 Jun 2024 03:23:50 +0000 (20:23 -0700)]
gpio: add missing MODULE_DESCRIPTION() macros
On x86, make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/gpio/gpio-gw-pld.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/gpio/gpio-mc33880.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/gpio/gpio-pcf857x.o
Add the missing invocations of the MODULE_DESCRIPTION() macro,
including the one missing in gpio-pl061.c, which is not built for x86.
David Howells [Fri, 31 May 2024 14:53:42 +0000 (15:53 +0100)]
cifs: Don't advance the I/O iterator before terminating subrequest
There's now no need to make sure subreq->io_iter is advanced to match
subreq->transferred before calling one of the netfs subrequest termination
functions as the check has been removed netfslib and the iterator is reset
prior to retrying a subreq.
Richard Acayan [Wed, 5 Jun 2024 17:58:09 +0000 (13:58 -0400)]
kbuild: explicitly run mksysmap as sed script from link-vmlinux.sh
In commit b18b047002b7 ("kbuild: change scripts/mksysmap into sed
script"), the mksysmap script was transformed into a sed script,
made directly executable with "#!/bin/sed -f". Apparently, the path to
sed is different on NixOS.
The shebang can't use the env command, otherwise the "sed -f" command
would be treated as a single argument. This can be solved with the -S
flag, but that is a GNU extension. Explicitly use sed instead of relying
on the executable shebang to fix NixOS builds without breaking build
environments using Busybox.
Linus Torvalds [Thu, 6 Jun 2024 21:40:51 +0000 (14:40 -0700)]
Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI fixes from James Bottomley:
"The core change is to detect unusually large number of VPD pages
(caused by device manufacturers having an endiannes issue) and reject
them rather than trying to parse a huge non-existent array.
The remaining fixes are in drivers the most user visible of which is
the ALUA state transition recognition (leads to intermittent I/O
errors in some situations otherwise)"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: ufs: mcq: Fix error output and clean up ufshcd_mcq_abort()
scsi: core: Handle devices which return an unusually large VPD page count
scsi: mpt3sas: Add missing kerneldoc parameter descriptions
scsi: qedf: Set qed_slowpath_params to zero before use
scsi: qedf: Wait for stag work during unload
scsi: qedf: Don't process stag work during unload and recovery
scsi: sr: Fix unintentional arithmetic wraparound
scsi: core: alua: I/O errors for ALUA state transitions
scsi: mpi3mr: Use proper format specifier in mpi3mr_sas_port_add()
Linus Torvalds [Thu, 6 Jun 2024 21:28:11 +0000 (14:28 -0700)]
Merge tag 'pci-v6.10-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci
Pull pci fix from Bjorn Helgaas:
- Revert lockdep checking on locking that protects device resets from
user-space config accesses; it exposed issues for which fixes are in
the works but are too risky for this cycle (Dan Williams)
* tag 'pci-v6.10-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci:
PCI: Revert the cfg_access_lock lockdep mechanism
[CAUSE]
Commit 09e6cef19c9f ("btrfs: refactor alloc_extent_buffer() to
allocate-then-attach method") changes the sequence when allocating a new
extent buffer.
Previously we always called grab_extent_buffer() under
mapping->i_private_lock, to ensure the safety on modification on
folio::private (which is a pointer to extent buffer for regular
sectorsize).
This can lead to the following race:
Thread A is trying to allocate an extent buffer at bytenr X, with 4
4K pages, meanwhile thread B is trying to release the page at X + 4K
(the second page of the extent buffer at X).
Thread A | Thread B
-----------------------------------+-------------------------------------
| btree_release_folio()
| | This is for the page at X + 4K,
| | Not page X.
| |
alloc_extent_buffer() | |- release_extent_buffer()
|- filemap_add_folio() for the | | |- atomic_dec_and_test(eb->refs)
| page at bytenr X (the first | | |
| page). | | |
| Which returned -EEXIST. | | |
| | | |
|- filemap_lock_folio() | | |
| Returned the first page locked. | | |
| | | |
|- grab_extent_buffer() | | |
| |- atomic_inc_not_zero() | | |
| | Returned false | | |
| |- folio_detach_private() | | |- folio_detach_private() for X
| |- folio_test_private() | | |- folio_test_private()
| Returned true | | | Returned true
|- folio_put() | |- folio_put()
Now there are two puts on the same folio at folio X, leading to refcount
underflow of the folio X, and eventually causing the BUG_ON() on the
page->mapping.
The condition is not that easy to hit:
- The release must be triggered for the middle page of an eb
If the release is on the same first page of an eb, page lock would kick
in and prevent the race.
- folio_detach_private() has a very small race window
It's only between folio_test_private() and folio_clear_private().
That's exactly when mapping->i_private_lock is used to prevent such race,
and commit 09e6cef19c9f ("btrfs: refactor alloc_extent_buffer() to
allocate-then-attach method") screwed that up.
At that time, I thought the page lock would kick in as
filemap_release_folio() also requires the page to be locked, but forgot
the filemap_release_folio() only locks one page, not all pages of an
extent buffer.
[FIX]
Move all the code requiring i_private_lock into
attach_eb_folio_to_filemap(), so that everything is done with proper
lock protection.
Furthermore to prevent future problems, add an extra
lockdep_assert_locked() to ensure we're holding the proper lock.
To reproducer that is able to hit the race (takes a few minutes with
instrumented code inserting delays to alloc_extent_buffer()):
Linus Torvalds [Thu, 6 Jun 2024 16:55:27 +0000 (09:55 -0700)]
Merge tag 'net-6.10-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Pull networking fixes from Jakub Kicinski:
"Including fixes from BPF and big collection of fixes for WiFi core and
drivers.
Current release - regressions:
- vxlan: fix regression when dropping packets due to invalid src
addresses
- bpf: fix a potential use-after-free in bpf_link_free()
- xdp: revert support for redirect to any xsk socket bound to the
same UMEM as it can result in a corruption
- virtio_net:
- add missing lock protection when reading return code from
control_buf
- fix false-positive lockdep splat in DIM
- Revert "wifi: wilc1000: convert list management to RCU"
- wifi: ath11k: fix error path in ath11k_pcic_ext_irq_config
Previous releases - regressions:
- rtnetlink: make the "split" NLM_DONE handling generic, restore the
old behavior for two cases where we started coalescing those
messages with normal messages, breaking sloppily-coded userspace
- wifi:
- cfg80211: validate HE operation element parsing
- cfg80211: fix 6 GHz scan request building
- mt76: mt7615: add missing chanctx ops
- ath11k: move power type check to ASSOC stage, fix connecting to
6 GHz AP
- ath11k: fix WCN6750 firmware crash caused by 17 num_vdevs
- rtlwifi: ignore IEEE80211_CONF_CHANGE_RETRY_LIMITS
- iwlwifi: mvm: fix a crash on 7265
Previous releases - always broken:
- ncsi: prevent multi-threaded channel probing, a spec violation
- vmxnet3: disable rx data ring on dma allocation failure
- ethtool: init tsinfo stats if requested, prevent unintentionally
reporting all-zero stats on devices which don't implement any
- dst_cache: fix possible races in less common IPv6 features
- tcp: auth: don't consider TCP_CLOSE to be in TCP_AO_ESTABLISHED
- ax25: fix two refcounting bugs
- eth: ionic: fix kernel panic in XDP_TX action
Misc:
- tcp: count CLOSE-WAIT sockets for TCP_MIB_CURRESTAB"
* tag 'net-6.10-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (107 commits)
selftests: net: lib: set 'i' as local
selftests: net: lib: avoid error removing empty netns name
selftests: net: lib: support errexit with busywait
net: ethtool: fix the error condition in ethtool_get_phy_stats_ethtool()
ipv6: fix possible race in __fib6_drop_pcpu_from()
af_unix: Annotate data-race of sk->sk_shutdown in sk_diag_fill().
af_unix: Use skb_queue_len_lockless() in sk_diag_show_rqlen().
af_unix: Use skb_queue_empty_lockless() in unix_release_sock().
af_unix: Use unix_recvq_full_lockless() in unix_stream_connect().
af_unix: Annotate data-race of net->unx.sysctl_max_dgram_qlen.
af_unix: Annotate data-races around sk->sk_sndbuf.
af_unix: Annotate data-races around sk->sk_state in UNIX_DIAG.
af_unix: Annotate data-race of sk->sk_state in unix_stream_read_skb().
af_unix: Annotate data-races around sk->sk_state in sendmsg() and recvmsg().
af_unix: Annotate data-race of sk->sk_state in unix_accept().
af_unix: Annotate data-race of sk->sk_state in unix_stream_connect().
af_unix: Annotate data-races around sk->sk_state in unix_write_space() and poll().
af_unix: Annotate data-race of sk->sk_state in unix_inq_len().
af_unix: Annodate data-races around sk->sk_state for writers.
af_unix: Set sk->sk_state under unix_state_lock() for truly disconencted peer.
...
Jakub Kicinski [Thu, 6 Jun 2024 15:23:44 +0000 (08:23 -0700)]
Merge branch 'selftests-net-lib-small-fixes'
Matthieu Baerts says:
====================
selftests: net: lib: small fixes
While looking at using 'lib.sh' for the MPTCP selftests [1], we found
some small issues with 'lib.sh'. Here they are:
- Patch 1: fix 'errexit' (set -e) support with busywait. 'errexit' is
supported in some functions, not all. A fix for v6.8+.
- Patch 2: avoid confusing error messages linked to the cleaning part
when the netns setup fails. A fix for v6.8+.
- Patch 3: set a variable as local to avoid accidentally changing the
value of a another one with the same name on the caller side. A fix
for v6.10-rc1+.
Without this, the 'i' variable declared before could be overridden by
accident, e.g.
for i in "${@}"; do
__ksft_status_merge "${i}" ## 'i' has been modified
foo "${i}" ## using 'i' with an unexpected value
done
After a quick look, it looks like 'i' is currently not used after having
been modified in __ksft_status_merge(), but still, better be safe than
sorry. I saw this while modifying the same file, not because I suspected
an issue somewhere.
selftests: net: lib: avoid error removing empty netns name
If there is an error to create the first netns with 'setup_ns()',
'cleanup_ns()' will be called with an empty string as first parameter.
The consequences is that 'cleanup_ns()' will try to delete an invalid
netns, and wait 20 seconds if the netns list is empty.
Instead of just checking if the name is not empty, convert the string
separated by spaces to an array. Manipulating the array is cleaner, and
calling 'cleanup_ns()' with an empty array will be a no-op.
ata: pata_macio: Fix max_segment_size with PAGE_SIZE == 64K
The pata_macio driver advertises a max_segment_size of 0xff00, because
the hardware doesn't cope with requests >= 64K.
However the SCSI core requires max_segment_size to be at least
PAGE_SIZE, which is a problem for pata_macio when the kernel is built
with 64K pages.
In older kernels the SCSI core would just increase the segment size to
be equal to PAGE_SIZE, however since the commit tagged below it causes a
warning and the device fails to probe:
WARNING: CPU: 0 PID: 26 at block/blk-settings.c:202 .blk_validate_limits+0x2f8/0x35c
CPU: 0 PID: 26 Comm: kworker/u4:1 Not tainted 6.10.0-rc1 #1
Hardware name: PowerMac7,2 PPC970 0x390202 PowerMac
...
NIP .blk_validate_limits+0x2f8/0x35c
LR .blk_alloc_queue+0xc0/0x2f8
Call Trace:
.blk_alloc_queue+0xc0/0x2f8
.blk_mq_alloc_queue+0x60/0xf8
.scsi_alloc_sdev+0x208/0x3c0
.scsi_probe_and_add_lun+0x314/0x52c
.__scsi_add_device+0x170/0x1a4
.ata_scsi_scan_host+0x2bc/0x3e4
.async_port_probe+0x6c/0xa0
.async_run_entry_fn+0x60/0x1bc
.process_one_work+0x228/0x510
.worker_thread+0x360/0x530
.kthread+0x134/0x13c
.start_kernel_thread+0x10/0x14
...
scsi_alloc_sdev: Allocation failure during SCSI scanning, some SCSI devices might not be configured
Although the hardware can't cope with a 64K segment, the driver
already deals with that internally by splitting large requests in
pata_macio_qc_prep(). That is how the driver has managed to function
until now on 64K kernels.
So fix the driver to advertise a max_segment_size of 64K, which avoids
the warning and keeps the SCSI core happy.
Fixes: afd53a3d8528 ("scsi: core: Initialize scsi midlayer limits before allocating the queue") Reported-by: Guenter Roeck <[email protected]> Closes: https://lore.kernel.org/all/[email protected]/ Reported-by: Doru Iorgulescu <[email protected]> Closes: https://bugzilla.kernel.org/show_bug.cgi?id=218858 Signed-off-by: Michael Ellerman <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]> Reviewed-by: Damien Le Moal <[email protected]> Reviewed-by: John Garry <[email protected]> Signed-off-by: Niklas Cassel <[email protected]>
Masahiro Yamada [Mon, 3 Jun 2024 16:19:04 +0000 (01:19 +0900)]
kconfig: remove wrong expr_trans_bool()
expr_trans_bool() performs an incorrect transformation.
[Test Code]
config MODULES
def_bool y
modules
config A
def_bool y
select C if B != n
config B
def_tristate m
config C
tristate
[Result]
CONFIG_MODULES=y
CONFIG_A=y
CONFIG_B=m
CONFIG_C=m
This output is incorrect because CONFIG_C=y is expected.
Documentation/kbuild/kconfig-language.rst clearly explains the function
of the '!=' operator:
If the values of both symbols are equal, it returns 'n',
otherwise 'y'.
Therefore, the statement:
select C if B != n
should be equivalent to:
select C if y
Or, more simply:
select C
Hence, the symbol C should be selected by the value of A, which is 'y'.
However, expr_trans_bool() wrongly transforms it to:
select C if B
Therefore, the symbol C is selected by (A && B), which is 'm'.
The comment block of expr_trans_bool() correctly explains its intention:
* bool FOO!=n => FOO
^^^^
If FOO is bool, FOO!=n can be simplified into FOO. This is correct.
However, the actual code performs this transformation when FOO is
tristate:
if (e->left.sym->type == S_TRISTATE) {
^^^^^^^^^^
While it can be fixed to S_BOOLEAN, there is no point in doing so
because expr_tranform() already transforms FOO!=n to FOO when FOO is
bool. (see the "case E_UNEQUAL" part)
Masahiro Yamada [Mon, 3 Jun 2024 16:19:03 +0000 (01:19 +0900)]
kconfig: doc: document behavior of 'select' and 'imply' followed by 'if'
Documentation/kbuild/kconfig-language.rst explains the behavior of
'select' as follows:
reverse dependencies can be used to force a lower limit of
another symbol. The value of the current menu symbol is used as the
minimal value <symbol> can be set to.
This is not true when the 'select' property is followed by 'if'.
[Test Code]
config MODULES
def_bool y
modules
config A
def_tristate y
select C if B
config B
def_tristate m
config C
tristate
[Result]
CONFIG_MODULES=y
CONFIG_A=y
CONFIG_B=m
CONFIG_C=m
If "the value of A is used as the minimal value C can be set to",
C must be 'y'.
The actual behavior is "C is selected by (A && B)". The lower limit of
C is downgraded due to B being 'm'.
This behavior is kind of weird, and this has arisen several times in
the mailing list.
I do not know whether it is a bug or intended behavior. Anyway, it is
not feasible to change it now because many Kconfig files are written
based on this behavior. The same applies to 'imply'.
Document this (but reserve the possibility for a future change).
Masahiro Yamada [Mon, 3 Jun 2024 16:19:02 +0000 (01:19 +0900)]
kconfig: doc: fix a typo in the note about 'imply'
This sentence does not make sense due to a typo. Fix it.
Fixes: def2fbffe62c ("kconfig: allow symbols implied by y to become m") Signed-off-by: Masahiro Yamada <[email protected]> Reviewed-by: Randy Dunlap <[email protected]>
Masahiro Yamada [Sat, 1 Jun 2024 18:20:38 +0000 (03:20 +0900)]
kconfig: remove unneeded code for user-supplied values being out of range
This is a leftover from commit ce1fc9345a59 ("kconfig: do not clear
SYMBOL_DEF_USER when the value is out of range").
This code is now redundant because if a user-supplied value is out
of range, the value adjusted by sym_validate_range() differs, and
conf_unsaved has already been incremented a few lines above.
Note that with this series there is skb_queue_empty() left in
unix_dgram_disconnected() that needs to be changed to lockless
version, and unix_peer(other) access there should be protected
by unix_state_lock().
This will require some refactoring, so another series will follow.
af_unix: Use skb_queue_empty_lockless() in unix_release_sock().
If the socket type is SOCK_STREAM or SOCK_SEQPACKET, unix_release_sock()
checks the length of the peer socket's recvq under unix_state_lock().
However, unix_stream_read_generic() calls skb_unlink() after releasing
the lock. Also, for SOCK_SEQPACKET, __skb_try_recv_datagram() unlinks
skb without unix_state_lock().
Thues, unix_state_lock() does not protect qlen.
Let's use skb_queue_empty_lockless() in unix_release_sock().
af_unix: Use unix_recvq_full_lockless() in unix_stream_connect().
Once sk->sk_state is changed to TCP_LISTEN, it never changes.
unix_accept() takes advantage of this characteristics; it does not
hold the listener's unix_state_lock() and only acquires recvq lock
to pop one skb.
It means unix_state_lock() does not prevent the queue length from
changing in unix_stream_connect().
Thus, we need to use unix_recvq_full_lockless() to avoid data-race.
Now we remove unix_recvq_full() as no one uses it.
Note that we can remove READ_ONCE() for sk->sk_max_ack_backlog in
unix_recvq_full_lockless() because of the following reasons:
(1) For SOCK_DGRAM, it is a written-once field in unix_create1()
(2) For SOCK_STREAM and SOCK_SEQPACKET, it is changed under the
listener's unix_state_lock() in unix_listen(), and we hold
the lock in unix_stream_connect()
af_unix: Annotate data-races around sk->sk_state in unix_write_space() and poll().
unix_poll() and unix_dgram_poll() read sk->sk_state locklessly and
calls unix_writable() which also reads sk->sk_state without holding
unix_state_lock().
Let's use READ_ONCE() in unix_poll() and unix_dgram_poll() and pass
it to unix_writable().
While at it, we remove TCP_SYN_SENT check in unix_dgram_poll() as
that state does not exist for AF_UNIX socket since the code was added.
Fixes: 1586a5877db9 ("af_unix: do not report POLLOUT on listeners") Fixes: 3c73419c09a5 ("af_unix: fix 'poll for write'/ connected DGRAM sockets") Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Kuniyuki Iwashima <[email protected]> Signed-off-by: Paolo Abeni <[email protected]>
af_unix: Set sk->sk_state under unix_state_lock() for truly disconencted peer.
When a SOCK_DGRAM socket connect()s to another socket, the both sockets'
sk->sk_state are changed to TCP_ESTABLISHED so that we can register them
to BPF SOCKMAP.
When the socket disconnects from the peer by connect(AF_UNSPEC), the state
is set back to TCP_CLOSE.
Then, the peer's state is also set to TCP_CLOSE, but the update is done
locklessly and unconditionally.
Let's say socket A connect()ed to B, B connect()ed to C, and A disconnects
from B.
After the first two connect()s, all three sockets' sk->sk_state are
TCP_ESTABLISHED:
So, when a socket disconnects from the peer, we should not set TCP_CLOSE to
the peer if the peer is connected to yet another socket, and this must be
done under unix_state_lock().
Note that we use WRITE_ONCE() for sk->sk_state as there are many lockless
readers. These data-races will be fixed in the following patches.
Fixes: 83301b5367a9 ("af_unix: Set TCP_ESTABLISHED for datagram sockets too") Signed-off-by: Kuniyuki Iwashima <[email protected]> Signed-off-by: Paolo Abeni <[email protected]>
net: wwan: iosm: Fix tainted pointer delete is case of region creation fail
In case of region creation fail in ipc_devlink_create_region(), previously
created regions delete process starts from tainted pointer which actually
holds error code value.
Fix this bug by decreasing region index before delete.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Ian Forbes [Thu, 28 Mar 2024 19:07:16 +0000 (14:07 -0500)]
drm/vmwgfx: Don't memcmp equivalent pointers
These pointers are frequently the same and memcmp does not compare the
pointers before comparing their contents so this was wasting cycles
comparing 16 KiB of memory which will always be equal.
====================
Intel Wired LAN Driver Updates 2024-05-29 (ice, igc)
This series includes fixes for the ice driver as well as a fix for the igc
driver.
Jacob fixes two issues in the ice driver with reading the NVM for providing
firmware data via devlink info. First, fix an off-by-one error when reading
the Preserved Fields Area, resolving an infinite loop triggered on some
NVMs which lack certain data in the NVM. Second, fix the reading of the NVM
Shadow RAM on newer E830 and E825-C devices which have a variable sized CSS
header rather than assuming this header is always the same fixed size as in
the E810 devices.
Larysa fixes three issues with the ice driver XDP logic that could occur if
the number of queues is changed after enabling an XDP program. First, the
af_xdp_zc_qps bitmap is removed and replaced by simpler logic to track
whether queues are in zero-copy mode. Second, the reset and .ndo_bpf flows
are distinguished to avoid potential races with a PF reset occuring
simultaneously to .ndo_bpf callback from userspace. Third, the logic for
mapping XDP queues to vectors is fixed so that XDP state is restored for
XDP queues after a reconfiguration.
Sasha fixes reporting of Energy Efficient Ethernet support via ethtool in
the igc driver.
Sasha Neftin [Mon, 3 Jun 2024 21:42:35 +0000 (14:42 -0700)]
igc: Fix Energy Efficient Ethernet support declaration
The commit 01cf893bf0f4 ("net: intel: i40e/igc: Remove setting Autoneg in
EEE capabilities") removed SUPPORTED_Autoneg field but left inappropriate
ethtool_keee structure initialization. When "ethtool --show <device>"
(get_eee) invoke, the 'ethtool_keee' structure was accidentally overridden.
Remove the 'ethtool_keee' overriding and add EEE declaration as per IEEE
specification that allows reporting Energy Efficient Ethernet capabilities.
Examples:
Before fix:
ethtool --show-eee enp174s0
EEE settings for enp174s0:
EEE status: not supported
After fix:
EEE settings for enp174s0:
EEE status: disabled
Tx LPI: disabled
Supported EEE link modes: 100baseT/Full
1000baseT/Full
2500baseT/Full
Larysa Zaremba [Mon, 3 Jun 2024 21:42:34 +0000 (14:42 -0700)]
ice: map XDP queues to vectors in ice_vsi_map_rings_to_vectors()
ice_pf_dcb_recfg() re-maps queues to vectors with
ice_vsi_map_rings_to_vectors(), which does not restore the previous
state for XDP queues. This leads to no AF_XDP traffic after rebuild.
Map XDP queues to vectors in ice_vsi_map_rings_to_vectors().
Also, move the code around, so XDP queues are mapped independently only
through .ndo_bpf().
Larysa Zaremba [Mon, 3 Jun 2024 21:42:33 +0000 (14:42 -0700)]
ice: add flag to distinguish reset from .ndo_bpf in XDP rings config
Commit 6624e780a577 ("ice: split ice_vsi_setup into smaller functions")
has placed ice_vsi_free_q_vectors() after ice_destroy_xdp_rings() in
the rebuild process. The behaviour of the XDP rings config functions is
context-dependent, so the change of order has led to
ice_destroy_xdp_rings() doing additional work and removing XDP prog, when
it was supposed to be preserved.
Also, dependency on the PF state reset flags creates an additional,
fortunately less common problem:
* PFR is requested e.g. by tx_timeout handler
* .ndo_bpf() is asked to delete the program, calls ice_destroy_xdp_rings(),
but reset flag is set, so rings are destroyed without deleting the
program
* ice_vsi_rebuild tries to delete non-existent XDP rings, because the
program is still on the VSI
* system crashes
With a similar race, when requested to attach a program,
ice_prepare_xdp_rings() can actually skip setting the program in the VSI
and nevertheless report success.
Instead of reverting to the old order of function calls, add an enum
argument to both ice_prepare_xdp_rings() and ice_destroy_xdp_rings() in
order to distinguish between calls from rebuild and .ndo_bpf().
Larysa Zaremba [Mon, 3 Jun 2024 21:42:32 +0000 (14:42 -0700)]
ice: remove af_xdp_zc_qps bitmap
Referenced commit has introduced a bitmap to distinguish between ZC and
copy-mode AF_XDP queues, because xsk_get_pool_from_qid() does not do this
for us.
The bitmap would be especially useful when restoring previous state after
rebuild, if only it was not reallocated in the process. This leads to e.g.
xdpsock dying after changing number of queues.
Instead of preserving the bitmap during the rebuild, remove it completely
and distinguish between ZC and copy-mode queues based on the presence of
a device associated with the pool.