The bpf_fifo_dequeue prog returns a skb which is a pointer. The pointer
is treated as a 32bit value and sign extend to 64bit in epilogue. This
behavior is right for most bpf prog types but wrong for struct ops which
requires LoongArch ABI.
So let's sign extend struct ops return values according to the LoongArch
ABI ([1]) and return value spec in function model.
Hengqi Chen [Thu, 2 Oct 2025 14:39:52 +0000 (22:39 +0800)]
LoongArch: BPF: Don't align trampoline size
Currently, arch_alloc_bpf_trampoline() use bpf_prog_pack_alloc() which
will pack multiple trampolines into a huge page. So, no need to assume
the trampoline size is page aligned.
Cc: stable@vger.kernel.org Tested-by: Vincent Li <vincent.mc.li@gmail.com> Signed-off-by: Hengqi Chen <hengqi.chen@gmail.com> Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
Hengqi Chen [Thu, 2 Oct 2025 14:39:52 +0000 (22:39 +0800)]
LoongArch: BPF: No text_poke() for kernel text
The current implementation of bpf_arch_text_poke() requires 5 nops
at patch site which is not applicable for kernel/module functions.
Because LoongArch reserves ONLY 2 nops at the function entry. With
CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS=y, this can be done by ftrace
instead.
See the following commit for details:
* commit b91e014f078e ("bpf: Make BPF trampoline use register_ftrace_direct() API")
* commit 9cdc3b6a299c ("LoongArch: ftrace: Add direct call support")
Cc: stable@vger.kernel.org Tested-by: Vincent Li <vincent.mc.li@gmail.com> Signed-off-by: Hengqi Chen <hengqi.chen@gmail.com> Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
Huacai Chen [Thu, 2 Oct 2025 14:39:52 +0000 (22:39 +0800)]
LoongArch: BPF: Fix uninitialized symbol 'retval_off'
In __arch_prepare_bpf_trampoline(), retval_off is meaningful only when
save_ret is not 0, so the current logic is correct. But it may cause a
build warning:
arch/loongarch/net/bpf_jit.c:1547 __arch_prepare_bpf_trampoline() error: uninitialized symbol 'retval_off'.
So initialize retval_off unconditionally to fix it.
For 8-bit and 16-bit sign-extention mov instructions, it can use the
native instructions ext.w.b and ext.w.h directly, no need to use the
temporary t1 register, just remove the redundant operations.
Tiezhu Yang [Thu, 2 Oct 2025 14:39:35 +0000 (22:39 +0800)]
LoongArch: Handle new atomic instructions for probes
The atomic instructions sc.q, llacq.{w/d}, screl.{w/d} were newly added
in the LoongArch Reference Manual v1.10, it is necessary to handle them
in insns_not_supported() to avoid putting a breakpoint in the middle of
a ll/sc atomic sequence, otherwise it will loop forever for kprobes and
uprobes.
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn> Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
Youling Tang [Thu, 2 Oct 2025 14:39:08 +0000 (22:39 +0800)]
LoongArch: Automatically disable kaslr if boot from kexec_file
Automatically disable kaslr when the kernel loads from kexec_file.
kexec_file loads the secondary kernel image to a non-linked address,
inherently providing KASLR-like randomization.
However, on LoongArch where System RAM may be non-contiguous, enabling
KASLR for the second kernel may relocate it to an invalid memory region
and cause a boot failure. Thus, we disable KASLR when "kexec_file" is
detected in the command line.
To ensure compatibility with older kernels loaded via kexec_file, this
patch should be backported to stable branches.
Youling Tang [Thu, 2 Oct 2025 14:39:08 +0000 (22:39 +0800)]
LoongArch: Add crash dump support for kexec_file
Enabling crash dump (kdump) includes:
- Prepare contents of ELF header of a core dump file, /proc/vmcore,
using crash_prepare_elf64_headers().
- Add "mem=size@start" parameter to the command line and pass it to the
capture kernel. Limit the runtime memory area of the captured kernel
to avoid disrupting the production kernel's runtime state.
- Add "elfcorehdr=size@start" parameter to the cmdline.
The basic usage for kdump (add the cmdline parameter crashkernel=512M
to grub.cfg for production kernel):
1) Load capture kernel image (vmlinux.efi or vmlinux can both be used):
# kexec -s -p vmlinuz.efi --initrd=initrd.img --reuse-cmdline
2) Do something to crash, like:
# echo c > /proc/sysrq-trigger
phdr->p_paddr should have been a physical address, but it is a virtual
address on the current LoongArch. This will cause kexec_file to fail
when loading the kernel and need to be converted to a physical address.
From the above MemSiz, it can be seen that 0x287a0d8 isn't page aligned.
Although kexec_add_buffer() will perform PAGE_SIZE alignment on kbuf->
memsz, there is still a stampeding in the loaded kernel space and initrd
space. The initrd resolution failed when starting the second kernel.
It can be known from the link script vmlinux.lds.S that,
BSS_SECTION(0, SZ_64K, 8)
. = ALIGN(PECOFF_SEGMENT_ALIGN);
It needs to be aligned according to SZ_64K, so that after alignment, its
size is consistent with _kernel_asize.
Youling Tang [Thu, 2 Oct 2025 14:39:07 +0000 (22:39 +0800)]
LoongArch: Add EFI binary support for kexec_file
This patch creates kexec_efi_ops to load EFI binary file for
kexec_file_load() syscall.
The efi_kexec_load() as two parts:
- the first part loads the kernel image (vmlinuz.efi or vmlinux.efi)
- the second part loads other segments (e.g: initrd, cmdline, etc)
Currently, pez (vmlinuz.efi) and pei (vmlinux.efi) format images are
supported.
Huacai Chen [Thu, 2 Oct 2025 14:39:07 +0000 (22:39 +0800)]
LoongArch: Allow specify SIMD width via kernel parameters
For power saving or debugging purpose, we usually want to limit the SIMD
(LSX/LASX) usage on a rich feature platform. So allow specify SIMD width
via kernel parameters "simd=".
Allowed values of "simd=" are any integers, and recommended values are:
0: Disable all SIMD features;
128: Enable at most 128bit SIMD features;
256: Enable at most 256bit SIMD features;
-1(default): Enable as many as possible SIMD features automatically.
Huacai Chen [Thu, 2 Oct 2025 14:38:57 +0000 (22:38 +0800)]
LoongArch: Init acpi_gbl_use_global_lock to false
Init acpi_gbl_use_global_lock to false, in order to void error messages
during boot phase:
ACPI Error: Could not enable GlobalLock event (20240827/evxfevnt-182)
ACPI Error: No response from Global Lock hardware, disabling lock (20240827/evglock-59)
Huacai Chen [Thu, 2 Oct 2025 14:38:57 +0000 (22:38 +0800)]
LoongArch: Fix build error for LTO with LLVM-18
Commit b15212824a01 ("LoongArch: Make LTO case independent in Makefile")
moves "KBUILD_LDFLAGS += -mllvm --loongarch-annotate-tablejump" out of
CONFIG_CC_HAS_ANNOTATE_TABLEJUMP, which breaks the build for LLVM-18, as
'--loongarch-annotate-tablejump' is unimplemented there:
ld.lld: error: -mllvm: ld.lld: Unknown command line argument '--loongarch-annotate-tablejump'.
Call ld-option to detect '--loongarch-annotate-tablejump' before use, so
as to fix the build error.
Fixes: b15212824a01 ("LoongArch: Make LTO case independent in Makefile") Reported-by: Nathan Chancellor <nathan@kernel.org> Reviewed-by: Nathan Chancellor <nathan@kernel.org> Tested-by: Nathan Chancellor <nathan@kernel.org> # build Suggested-by: WANG Rui <wangrui@loongson.cn> Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
Currently, when compiling with GCC, there is no "break 7" instruction
for zero division due to using the option -mno-check-zero-division, but
the compiler still generates "break 0" instruction for zero division.
Here is a simple example:
$ cat test.c
int div(int a)
{
return a / 0;
}
$ gcc -O2 -S test.c -o test.s
GCC generates "break 0" on LoongArch and "ud2" on x86, objtool decodes
"ud2" as INSN_BUG for x86, so decode "break 0" as INSN_BUG can fix the
objtool warnings for LoongArch, but this is not the intention.
When decoding "break 0" as INSN_TRAP in the previous commit, the aim is
to handle "break 0" as a trap. The generated "break 0" for zero division
by GCC is not proper, it should generate a break instruction with proper
bug type, so add the GCC option -fno-isolate-erroneous-paths-dereference
to avoid generating the unexpected "break 0" instruction for now.
Reported-by: kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/r/202509200413.7uihAxJ5-lkp@intel.com/ Fixes: baad7830ee9a ("objtool/LoongArch: Mark types based on break immediate code") Suggested-by: WANG Rui <wangrui@loongson.cn> Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn> Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
Merge branches 'acpi-thermal', 'acpi-fan', 'acpi-video', 'acpi-tad' and 'acpi-prm'
Merge an ACPI thermal zone driver update, an ACPI fan driver update, an
ACPI backlight (video) driver update, an ACPI TAD (time and alarm
device) driver update, and an ACPI PRM (platform runtime mechanism)
driver update for 6.18-rc1:
- Eliminate a dummy local variable from the ACPI thermal driver (Rafael
Wysocki)
- Fold two simple functions into their only caller in the ACPI fan
driver (Rafael Wysocki)
- Force native backlight on Lenovo 82K8 in the ACPI backlight (video)
driver (Mario Limonciello)
- Add missing sysfs_remove_group() for ACPI_TAD_RT (Daniel Tang)
- Skip PRM handlers with NULL handler_address or NULL VA in the ACPI
PRM driver (Shang song)
* acpi-thermal:
ACPI: thermal: Get rid of a dummy local variable
* acpi-fan:
ACPI: fan: Fold two simple functions into their only caller
* acpi-video:
ACPI: video: force native for Lenovo 82K8
* acpi-tad:
ACPI: TAD: Add missing sysfs_remove_group() for ACPI_TAD_RT
* acpi-prm:
ACPI: PRM: Skip handlers with NULL handler_address or NULL VA
Merge branches 'acpi-property', 'acpi-resource', 'acpi-pm' and 'acpi-tables'
Merge updates of the ACPI device properties management code, ACPI
resources management code, ACPI power management, and ACPI data tables
parsing code for 6.18-rc1:
- Fix ACPI buffer properties extraction for data-only subnodes
represented as _DSD-equivalent packages (Rafael Wysocki)
- Fix handling of ACPI data-only subnodes represented as _DSD-equivalent
packages in the case when they are embedded in larger _DSD-equivalent
packages and clean up acpi_nondev_subnode_extract() (Rafael Wysocki)
- Skip ACPI IRQ override on ASUS Vivobook Pro N6506CU (Sam van Kampen)
- Add power resource init function and use it for introducing an HP
EliteBook 855 G7 WWAN modem power resource quirk (Maciej Szmigiero)
- Add support for DBG2 RISC-V SBI port subtype and Precise Baud Rate
field to the ACPI SPCR table parser (Chen Pei)
* acpi-property:
ACPI: property: Adjust failure handling in acpi_nondev_subnode_extract()
ACPI: property: Do not pass NULL handles to acpi_attach_data()
ACPI: property: Add code comments explaining what is going on
ACPI: property: Disregard references in data-only subnode lists
ACPI: property: Fix buffer properties extraction for subnodes
* acpi-resource:
ACPI: resource: Skip IRQ override on ASUS Vivobook Pro N6506CU
* acpi-pm:
ACPI: PM: Add HP EliteBook 855 G7 WWAN modem power resource quirk
ACPI: PM: Add power resource init function
* acpi-tables:
ACPI: SPCR: Support Precise Baud Rate field
ACPI: SPCR: Add support for DBG2 RISC-V SBI port subtype
- Rearrange local variable definition involving #ifdef in ACPICA to
avoid using uninitialized variables (Zhe Qiao)
- Allow ACPICA to skip Global Lock initialization (Huacai Chen)
- Apply ACPI_NONSTRING in more places in ACPICA and fix two regressions
related to incorrect ACPI_NONSTRING usage (Ahmed Salem)
- Fix printing CDAT table header when dissasebling CDAT AML (Ahmed
Salem)
- Use acpi_ds_clear_operands() in acpi_ds_call_control_method() in
ACPICA (Hans de Goede)
- Update dsmethod.c in ACPICA to address unused variable warning (Saket
Dumbre)
- Print error messages in ACPICA for too few or too many control method
arguments (Saket Dumbre)
- Update ACPICA version to 20250807 (Saket Dumbre)
- Fix largest possible resource descriptor index in ACPICA (Dmitry
Antipov)
- Add Back-Invalidate restriction to CXL Window for CEDT in ACPICA
(Davidlohr Bueso).
- Add the package type to acceptable Arg3 types for _DSM in ACPICA
because ACPI_TYPE_ANY does not cover it (Saket Dumbre)
- Fix return values in ap_is_valid_checksum() in the acpidump utility
in ACPICA (Kaushlendra Kumar)
* acpica:
ACPICA: acpidump: fix return values in ap_is_valid_checksum()
ACPICA: ACPI_TYPE_ANY does not include the package type
ACPICA: CEDT: Add Back-Invalidate restriction to CXL Window
ACPICA: Fix largest possible resource descriptor index
ACPICA: Update version to 20250807
ACPICA: Print error messages for too few or too many arguments
ACPICA: Update dsmethod.c to get rid of unused variable warning
ACPICA: dispatcher: Use acpi_ds_clear_operands() in acpi_ds_call_control_method()
ACPICA: Debugger: drop ACPI_NONSTRING attribute from name_seg
ACPICA: acpidump: drop ACPI_NONSTRING attribute from file_name
ACPICA: iASL: Fix printing CDAT table header
ACPICA: Apply ACPI_NONSTRING
ACPICA: Allow to skip Global Lock initialization
ACPICA: Change the compilation conditions
ACPICA: Remove redundant "#ifdef" definitions
ACPICA: Modify variable definition position
ACPICA: Add SoundWire File Table (SWFT) signature
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rmk/linux
Pull ARM fix from Russell King:
"Just one fix to the module freeing function that was declared __weak
when it should not have been. Thanks to Petr Pavlu for spotting this"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rmk/linux:
ARM: 9458/1: module: Ensure the override of module_arch_freeing_init()
Merge tag 'i2c-for-6.17-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
Pull i2c fixes from Wolfram Sang:
- various MAINTAINERS updates
- fix an off-by-one error in riic
- fix k1 DT schema to allow validation
- rtl9300: fix faulty merge conflict resolution
* tag 'i2c-for-6.17-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: rtl9300: Drop unsupported I2C_FUNC_SMBUS_I2C_BLOCK
MAINTAINERS: add entry for SpacemiT K1 I2C driver
MAINTAINERS: Add me as maintainer of Synopsys DesignWare I2C driver
MAINTAINERS: delete email for Tharun Kumar P
dt-bindings: i2c: spacemit: extend and validate all properties
i2c: riic: Allow setting frequencies lower than 50KHz
MAINTAINERS: Remove myself as Synopsys DesignWare I2C maintainer
MAINTAINERS: Update email address for Qualcomm's I2C GENI maintainers
Merge tag 'trace-v6.17-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull tracing fixes from Steven Rostedt:
- Fix buffer overflow in osnoise_cpu_write()
The allocated buffer to read user space did not add a nul terminating
byte after copying from user the string. It then reads the string,
and if user space did not add a nul byte, the read will continue
beyond the string.
Add a nul terminating byte after reading the string.
- Fix missing check for lockdown on tracing
There's a path from kprobe events or uprobe events that can update
the tracing system even if lockdown on tracing is activate. Add a
check in the dynamic event path.
- Add a recursion check for the function graph return path
Now that fprobes can hook to the function graph tracer and call
different code between the entry and the exit, the exit code may now
call functions that are not called in entry. This means that the exit
handler can possibly trigger recursion that is not caught and cause
the system to crash.
Add the same recursion checks in the function exit handler as exists
in the entry handler path.
* tag 'trace-v6.17-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
tracing: fgraph: Protect return handler from recursion loop
tracing: dynevent: Add a missing lockdown check on dynevent
tracing/osnoise: Fix slab-out-of-bounds in _parse_integer_limit()
Merge tag 'spi-fix-v6.17-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi
Pull spi fixes from Mark Brown:
"A few final driver specific fixes that have been sitting in -next for
a bit.
The OMAP issue is likely to come up very infrequently since mixed
configuration SPI buses are rare and the Cadence issue is specific to
SoCFPGA systems"
* tag 'spi-fix-v6.17-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
spi: omap2-mcspi: drive SPI_CLK on transfer_setup()
spi: cadence-qspi: defer runtime support on socfpga if reset bit is enabled
Merge tag 'mm-hotfixes-stable-2025-09-27-22-35' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull misc fixes from Andrew Morton:
"7 hotfixes. 4 are cc:stable and the remainder address post-6.16 issues
or aren't considered necessary for -stable kernels. 6 of these fixes
are for MM.
All singletons, please see the changelogs for details"
* tag 'mm-hotfixes-stable-2025-09-27-22-35' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
include/linux/pgtable.h: convert arch_enter_lazy_mmu_mode() and friends to static inlines
mm/damon/sysfs: do not ignore callback's return value in damon_sysfs_damon_call()
mailmap: add entry for Bence Csókás
fs/proc/task_mmu: check p->vec_buf for NULL
kmsan: fix out-of-bounds access to shadow memory
mm/hugetlb: fix copy_hugetlb_page_range() to use ->pt_share_count
mm/hugetlb: fix folio is still mapped when deleted
Sven Eckelmann [Sat, 27 Sep 2025 09:52:16 +0000 (11:52 +0200)]
i2c: rtl9300: Drop unsupported I2C_FUNC_SMBUS_I2C_BLOCK
While applying the patch for commit ede965fd555a ("i2c: rtl9300: remove
broken SMBus Quick operation support"), a conflict was incorrectly solved
by adding the I2C_FUNC_SMBUS_I2C_BLOCK feature flag. But the code to handle
I2C_SMBUS_I2C_BLOCK_DATA requests will be added by a separate commit.
Fixes: ede965fd555a ("i2c: rtl9300: remove broken SMBus Quick operation support") Signed-off-by: Sven Eckelmann <sven@narfation.org> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Mika Westerberg [Fri, 26 Sep 2025 11:50:57 +0000 (13:50 +0200)]
MAINTAINERS: Add me as maintainer of Synopsys DesignWare I2C driver
I volunteered as maintainer of the DesignWare I2C driver, so update my
entry from reviewer to maintainer.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Merge tag 'trace-tools-v6.17-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull rtla tool fixes from Steven Rostedt:
- Fix a buffer overflow in actions_parse()
The "trigger_c" variable did not account for the nul byte when
determining its size
- Fix a compare that had the values reversed
actions_destroy() is supposed to reallocate when len is greater than
the current size, but the compare was testing if size is greater than
the new length
* tag 'trace-tools-v6.17-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
rtla/actions: Fix condition for buffer reallocation
rtla: Fix buffer overflow in actions_parse
tracing: fgraph: Protect return handler from recursion loop
function_graph_enter_regs() prevents itself from recursion by
ftrace_test_recursion_trylock(), but __ftrace_return_to_handler(),
which is called at the exit, does not prevent such recursion.
Therefore, while it can prevent recursive calls from
fgraph_ops::entryfunc(), it is not able to prevent recursive calls
to fgraph from fgraph_ops::retfunc(), resulting in a recursive loop.
This can lead an unexpected recursion bug reported by Menglong.
is_endbr() is called in __ftrace_return_to_handler -> fprobe_return
-> kprobe_multi_link_exit_handler -> is_endbr.
To fix this issue, acquire ftrace_test_recursion_trylock() in the
__ftrace_return_to_handler() after unwind the shadow stack to mark
this section must prevent recursive call of fgraph inside user-defined
fgraph_ops::retfunc().
This is essentially a fix to commit 4346ba160409 ("fprobe: Rewrite
fprobe on function-graph tracer"), because before that fgraph was
only used from the function graph tracer. Fprobe allowed user to run
any callbacks from fgraph after that commit.
rtla/actions: Fix condition for buffer reallocation
The condition to check if the actions buffer needs to be resized was
incorrect. The check `self->size >= self->len` would evaluate to
true on almost every call to `actions_new()`, causing the buffer to
be reallocated unnecessarily each time an action was added.
Fix the condition to `self->len >= self.size`, ensuring
that the buffer is only resized when it is actually full.
Cc: John Kacur <jkacur@redhat.com> Cc: Luis Goncalves <lgoncalv@redhat.com> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Chang Yin <cyin@redhat.com> Cc: Costa Shulyupin <costa.shul@redhat.com> Cc: Crystal Wood <crwood@redhat.com> Cc: Gabriele Monaco <gmonaco@redhat.com> Link: https://lore.kernel.org/20250915181101.52513-1-wander@redhat.com Fixes: 6ea082b171e00 ("rtla/timerlat: Add action on threshold feature") Signed-off-by: Wander Lairson Costa <wander@redhat.com> Reviewed-by: Tomas Glozar <tglozar@redhat.com> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Merge tag 'riscv-for-linus-v6.17-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux
Pull RISC-V fixes from Paul Walmsley:
- A race-free implementation of pudp_huge_get_and_clear() (based on the
x86 code)
- A MAINTAINERS update to my E-mail address
* tag 'riscv-for-linus-v6.17-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
MAINTAINERS: Update Paul Walmsley's E-mail address
riscv: Use an atomic xchg in pudp_huge_get_and_clear()
Merge tag 'x86-urgent-2025-09-26' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Ingo Molnar:
"Fix a CPU topology code regression that caused the mishandling of
certain boot command line options, and re-enable CONFIG_PTDUMP on i386
that was mistakenly turned off in the Kconfig"
* tag 'x86-urgent-2025-09-26' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/topology: Implement topology_is_core_online() to address SMT regression
x86/Kconfig: Reenable PTDUMP on i386
Merge tag 'sched-urgent-2025-09-26' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fixes from Ingo Molnar:
"Fix two dl_server regressions: a race that can end up leaving the
dl_server stuck, and a dl_server throttling bug causing lag to fair
tasks"
* tag 'sched-urgent-2025-09-26' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/deadline: Fix dl_server behaviour
sched/deadline: Fix dl_server getting stuck
Merge tag 'locking-urgent-2025-09-26' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull locking fixes from Ingo Molnar:
"Fix a PI-futexes race, and fix a copy_process() futex cleanup bug"
* tag 'locking-urgent-2025-09-26' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
futex: Use correct exit on failure from futex_hash_allocate_default()
futex: Prevent use-after-free during requeue-PI
Merge tag 'v6.17rc7-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6
Pull smb client fixes from Steve French:
- Fix unlink bug
- Fix potential out of bounds access in processing compound requests
* tag 'v6.17rc7-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
smb: client: fix wrong index reference in smb2_compound_op()
smb: client: handle unlink(2) of files open by different clients
Merge tag 'pmdomain-v6.17-rc2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/linux-pm
Pull pmdomain fix from Ulf Hansson:
- mediatek: Make sure MT8195 AUDIO power domain isn't left powered-on
* tag 'pmdomain-v6.17-rc2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/linux-pm:
pmdomain: mediatek: set default off flag for MT8195 AUDIO power domain
Merge tag 'platform-drivers-x86-v6.17-5' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86
Pull x86 platform driver fixes from Ilpo Järvinen:
"Fixes and New HW Supoort
- amd/pmc: Use 8042 quirk for Stellaris Slim Gen6 AMD
- dell: Set USTT mode according to BIOS after reboot
- dell-lis3lv02d: Add Latitude E6530
- lg-laptop: Fix setting the fan mode"
* tag 'platform-drivers-x86-v6.17-5' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86:
platform/x86: lg-laptop: Fix WMAB call in fan_mode_store()
platform/x86: dell-lis3lv02d: Add Latitude E6530
platform/x86/dell: Set USTT mode according to BIOS after reboot
platform/x86/amd/pmc: Add Stellaris Slim Gen6 AMD to spurious 8042 quirks list
Merge tag 'gpio-fixes-for-v6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux
Pull gpio fixes from Bartosz Golaszewski:
- allow looking up GPIOs by the secondary firmware node too
- fix memory leak in gpio-regmap
* tag 'gpio-fixes-for-v6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
gpio: regmap: fix memory leak of gpio_regmap structure
gpiolib: Extend software-node support to support secondary software-nodes
Merge tag 'block-6.17-20250925' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux
Pull block fixes from Jens Axboe:
"A regression fix for this series where an attempt to silence an EOD
error got messed up a bit, and then a change of git trees for the
block and io_uring trees.
Switching the git trees to kernel.org now, as I've just about had it
trying to battle AI bots that bring the box to its knees, continually.
At least I don't have to maintain the kernel.org side"
* tag 'block-6.17-20250925' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux:
MAINTAINERS: update io_uring and block tree git trees
block: fix EOD return for device with nr_sectors == 0
Merge tag 'drm-fixes-2025-09-26' of https://gitlab.freedesktop.org/drm/kernel
Pull drm fixes from Dave Airlie:
"Weekly fixes, some fbcon font handling fixes, then amdgpu/xe/i915 with
a few, and a few misc fixes for other drivers. Seems about right for
this stage, and I don't know of anything outstanding.
fbcon:
- fix OOB access in font allocation
- fix integer overflow in font handling
xe:
- Don't expose sysfs attributes not applicable for VFs
- Fix build with CONFIG_MODULES=n
- Don't copy pinned kernel bos twice on suspend
i915:
- Set O_LARGEFILE in __create_shmem()
- Guard reg_val against a INVALID_TRANSCODER [ddi]
ast:
- sleeps causing cpu stall fix
panthor:
- scheduler race condition fix
gma500:
- NULL ptr deref in hdmi teardown fix"
* tag 'drm-fixes-2025-09-26' of https://gitlab.freedesktop.org/drm/kernel:
drm/panthor: Defer scheduler entitiy destruction to queue release
drm/amd/display: remove output_tf_change flag
drm/amd/display: Init DCN35 clocks from pre-os HW values
drm/amd/display: Use mpc.preblend flag to indicate preblend
drm/amd/display: Only restore backlight after amdgpu_dm_init or dm_resume
fbcon: Fix OOB access in font allocation
drm/i915/ddi: Guard reg_val against a INVALID_TRANSCODER
drm/i915: set O_LARGEFILE in __create_shmem()
drm/xe: Don't copy pinned kernel bos twice on suspend
drm/xe: Fix build with CONFIG_MODULES=n
drm/xe/vf: Don't expose sysfs attributes not applicable for VFs
fbcon: fix integer overflow in fbcon_do_set_font
drm/gma500: Fix null dereference in hdmi teardown
drm/ast: Use msleep instead of mdelay for edid read
smb: client: fix wrong index reference in smb2_compound_op()
In smb2_compound_op(), the loop that processes each command's response
uses wrong indices when accessing response bufferes.
This incorrect indexing leads to improper handling of command results.
Also, if incorrectly computed index is greather than or equal to
MAX_COMPOUND, it can cause out-of-bounds accesses.
Fixes: 3681c74d342d ("smb: client: handle lack of EA support in smb2_query_path_info()") # 6.14 Reviewed-by: Paulo Alcantara (Red Hat) <pc@manguebit.org> Signed-off-by: Sang-Heon Jeon <ekffu200098@gmail.com> Signed-off-by: Steve French <stfrench@microsoft.com>
Max Kellermann [Thu, 25 Sep 2025 13:08:20 +0000 (14:08 +0100)]
netfs: fix reference leak
Commit 20d72b00ca81 ("netfs: Fix the request's work item to not
require a ref") modified netfs_alloc_request() to initialize the
reference counter to 2 instead of 1. The rationale was that the
requet's "work" would release the second reference after completion
(via netfs_{read,write}_collection_worker()). That works most of the
time if all goes well.
However, it leaks this additional reference if the request is released
before the I/O operation has been submitted: the error code path only
decrements the reference counter once and the work item will never be
queued because there will never be a completion.
This has caused outages of our whole server cluster today because
tasks were blocked in netfs_wait_for_outstanding_io(), leading to
deadlocks in Ceph (another bug that I will address soon in another
patch). This was caused by a netfs_pgpriv2_begin_copy_to_cache() call
which failed in fscache_begin_write_operation(). The leaked
netfs_io_request was never completed, leaving `netfs_inode.io_count`
with a positive value forever.
All of this is super-fragile code. Finding out which code paths will
lead to an eventual completion and which do not is hard to see:
- Some functions like netfs_create_write_req() allocate a request, but
will never submit any I/O.
- netfs_unbuffered_read_iter_locked() calls netfs_unbuffered_read()
and then netfs_put_request(); however, netfs_unbuffered_read() can
also fail early before submitting the I/O request, therefore another
netfs_put_request() call must be added there.
A rule of thumb is that functions that return a `netfs_io_request` do
not submit I/O, and all of their callers must be checked.
For my taste, the whole netfs code needs an overhaul to make reference
counting easier to understand and less fragile & obscure. But to fix
this bug here and now and produce a patch that is adequate for a
stable backport, I tried a minimal approach that quickly frees the
request object upon early failure.
I decided against adding a second netfs_put_request() each time
because that would cause code duplication which obscures the code
further. Instead, I added the function netfs_put_failed_request()
which frees such a failed request synchronously under the assumption
that the reference count is exactly 2 (as initially set by
netfs_alloc_request() and never touched), verified by a
WARN_ON_ONCE(). It then deinitializes the request object (without
going through the "cleanup_work" indirection) and frees the allocation
(with RCU protection to protect against concurrent access by
netfs_requests_seq_start()).
All code paths that fail early have been changed to call
netfs_put_failed_request() instead of netfs_put_request().
Additionally, I have added a netfs_put_request() call to
netfs_unbuffered_read() as explained above because the
netfs_put_failed_request() approach does not work there.
Fixes: 20d72b00ca81 ("netfs: Fix the request's work item to not require a ref") Signed-off-by: Max Kellermann <max.kellermann@ionos.com> Signed-off-by: David Howells <dhowells@redhat.com>
cc: Paulo Alcantara <pc@manguebit.org>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: stable@vger.kernel.org Signed-off-by: Christian Brauner <brauner@kernel.org>
Andrew Morton [Sun, 14 Sep 2025 00:03:39 +0000 (17:03 -0700)]
include/linux/pgtable.h: convert arch_enter_lazy_mmu_mode() and friends to static inlines
commit c519c3c0a113 ("mm/kasan: avoid lazy MMU mode hazards") introduced
the use of arch_enter_lazy_mmu_mode(), which results in the compiler
complaining about "statement has no effect", when
__HAVE_ARCH_LAZY_MMU_MODE is not defined in include/linux/pgtable.h
The exact warning/error is:
In file included from ./include/linux/kasan.h:37,
from mm/kasan/shadow.c:14:
mm/kasan/shadow.c: In function kasan_populate_vmalloc_pte:
./include/linux/pgtable.h:247:41: error: statement with no effect [-Werror=unused-value]
247 | #define arch_enter_lazy_mmu_mode() (LAZY_MMU_DEFAULT)
| ^
mm/kasan/shadow.c:322:9: note: in expansion of macro arch_enter_lazy_mmu_mode> 322 | arch_enter_lazy_mmu_mode();
| ^~~~~~~~~~~~~~~~~~~~~~~~
switching these "functions" to static inlines fixes this up.
Akinobu Mita [Sat, 20 Sep 2025 13:25:46 +0000 (22:25 +0900)]
mm/damon/sysfs: do not ignore callback's return value in damon_sysfs_damon_call()
The callback return value is ignored in damon_sysfs_damon_call(), which
means that it is not possible to detect invalid user input when writing
commands such as 'commit' to
/sys/kernel/mm/damon/admin/kdamonds/<K>/state. Fix it.
Link: https://lkml.kernel.org/r/20250920132546.5822-1-akinobu.mita@gmail.com Fixes: f64539dcdb87 ("mm/damon/sysfs: use damon_call() for update_schemes_stats") Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Reviewed-by: SeongJae Park <sj@kernel.org> Cc: <stable@vger.kernel.org> [6.14+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
vec_len = 0 in pagemap_scan_init_bounce_buffer() means no buffers are
allocated and p->vec_buf remains set to NULL.
This breaks an assumption made later in pagemap_scan_backout_range(), that
page_region is always allocated for p->vec_buf_index.
Fix it by explicitly checking p->vec_buf for NULL before dereferencing.
Other sites that might run into same deref-issue are already (directly or
transitively) protected by checking p->vec_buf.
Note:
From PAGEMAP_SCAN man page, it seems vec_len = 0 is valid when no output
is requested and it's only the side effects caller is interested in,
hence it passes check in pagemap_scan_get_args().
This issue was found by syzkaller.
Link: https://lkml.kernel.org/r/20250922082206.6889-1-acsjakub@amazon.de Fixes: 52526ca7fdb9 ("fs/proc/task_mmu: implement IOCTL to get and optionally clear info about PTEs") Signed-off-by: Jakub Acs <acsjakub@amazon.de> Reviewed-by: Muhammad Usama Anjum <usama.anjum@collabora.com> Acked-by: David Hildenbrand <david@redhat.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com> Cc: Jinjiang Tu <tujinjiang@huawei.com> Cc: Suren Baghdasaryan <surenb@google.com> Cc: Penglei Jiang <superman.xpt@gmail.com> Cc: Mark Brown <broonie@kernel.org> Cc: Baolin Wang <baolin.wang@linux.alibaba.com> Cc: Ryan Roberts <ryan.roberts@arm.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: "Michał Mirosław" <mirq-linux@rere.qmqm.pl> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This occurs when memset() is called on a buffer that is not 4-byte aligned
and extends to the end of a guard page, i.e. the next page is unmapped.
The bug is that the loop at the end of kmsan_internal_set_shadow_origin()
accesses the wrong shadow memory bytes when the address is not 4-byte
aligned. Since each 4 bytes are associated with an origin, it rounds the
address and size so that it can access all the origins that contain the
buffer. However, when it checks the corresponding shadow bytes for a
particular origin, it incorrectly uses the original unrounded shadow
address. This results in reads from shadow memory beyond the end of the
buffer's shadow memory, which crashes when that memory is not mapped.
To fix this, correctly align the shadow address before accessing the 4
shadow bytes corresponding to each origin.
Link: https://lkml.kernel.org/r/20250911195858.394235-1-ebiggers@kernel.org Fixes: 2ef3cec44c60 ("kmsan: do not wipe out origin when doing partial unpoisoning") Signed-off-by: Eric Biggers <ebiggers@kernel.org> Tested-by: Alexander Potapenko <glider@google.com> Reviewed-by: Alexander Potapenko <glider@google.com> Cc: Dmitriy Vyukov <dvyukov@google.com> Cc: Marco Elver <elver@google.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Jane Chu [Tue, 16 Sep 2025 00:45:20 +0000 (18:45 -0600)]
mm/hugetlb: fix copy_hugetlb_page_range() to use ->pt_share_count
commit 59d9094df3d79 ("mm: hugetlb: independent PMD page table shared
count") introduced ->pt_share_count dedicated to hugetlb PMD share count
tracking, but omitted fixing copy_hugetlb_page_range(), leaving the
function relying on page_count() for tracking that no longer works.
When lazy page table copy for hugetlb is disabled, that is, revert commit bcd51a3c679d ("hugetlb: lazy page table copies in fork()") fork()'ing with
hugetlb PMD sharing quickly lockup -
There are two options to resolve the potential latent issue:
1. warn against PMD sharing in copy_hugetlb_page_range(),
2. fix it.
This patch opts for the second option.
While at it, simplify the comment, the details are not actually relevant
anymore.
Link: https://lkml.kernel.org/r/20250916004520.1604530-1-jane.chu@oracle.com Fixes: 59d9094df3d7 ("mm: hugetlb: independent PMD page table shared count") Signed-off-by: Jane Chu <jane.chu@oracle.com> Reviewed-by: Harry Yoo <harry.yoo@oracle.com> Acked-by: Oscar Salvador <osalvador@suse.de> Acked-by: David Hildenbrand <david@redhat.com> Cc: Jann Horn <jannh@google.com> Cc: Liu Shixin <liushixin2@huawei.com> Cc: Muchun Song <muchun.song@linux.dev> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Jinjiang Tu [Fri, 12 Sep 2025 07:41:39 +0000 (15:41 +0800)]
mm/hugetlb: fix folio is still mapped when deleted
Migration may be raced with fallocating hole. remove_inode_single_folio
will unmap the folio if the folio is still mapped. However, it's called
without folio lock. If the folio is migrated and the mapped pte has been
converted to migration entry, folio_mapped() returns false, and won't
unmap it. Due to extra refcount held by remove_inode_single_folio,
migration fails, restores migration entry to normal pte, and the folio is
mapped again. As a result, we triggered BUG in filemap_unaccount_folio.
The log is as follows:
BUG: Bad page cache in process hugetlb pfn:156c00
page: refcount:515 mapcount:0 mapping:0000000099fef6e1 index:0x0 pfn:0x156c00
head: order:9 mapcount:1 entire_mapcount:1 nr_pages_mapped:0 pincount:0
aops:hugetlbfs_aops ino:dcc dentry name(?):"my_hugepage_file"
flags: 0x17ffffc00000c1(locked|waiters|head|node=0|zone=2|lastcpupid=0x1fffff)
page_type: f4(hugetlb)
page dumped because: still mapped when deleted
CPU: 1 UID: 0 PID: 395 Comm: hugetlb Not tainted 6.17.0-rc5-00044-g7aac71907bde-dirty #484 NONE
Hardware name: QEMU Ubuntu 24.04 PC (i440FX + PIIX, 1996), BIOS 0.0.0 02/06/2015
Call Trace:
<TASK>
dump_stack_lvl+0x4f/0x70
filemap_unaccount_folio+0xc4/0x1c0
__filemap_remove_folio+0x38/0x1c0
filemap_remove_folio+0x41/0xd0
remove_inode_hugepages+0x142/0x250
hugetlbfs_fallocate+0x471/0x5a0
vfs_fallocate+0x149/0x380
Hold folio lock before checking if the folio is mapped to avold race with
migration.
Link: https://lkml.kernel.org/r/20250912074139.3575005-1-tujinjiang@huawei.com Fixes: 4aae8d1c051e ("mm/hugetlbfs: unmap pages if page fault raced with hole punch") Signed-off-by: Jinjiang Tu <tujinjiang@huawei.com> Cc: David Hildenbrand <david@redhat.com> Cc: Kefeng Wang <wangkefeng.wang@huawei.com> Cc: Matthew Wilcox (Oracle) <willy@infradead.org> Cc: Muchun Song <muchun.song@linux.dev> Cc: Oscar Salvador <osalvador@suse.de> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Merge tag 'net-6.17-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Pull networking fixes from Paolo Abeni:
"Including fixes from Bluetooth, IPsec and CAN.
No known regressions at this point.
Current release - regressions:
- xfrm: xfrm_alloc_spi shouldn't use 0 as SPI
Previous releases - regressions:
- xfrm: fix offloading of cross-family tunnels
- bluetooth: fix several races leading to UaFs
- dsa: lantiq_gswip: fix FDB entries creation for the CPU port
- eth:
- tun: update napi->skb after XDP process
- mlx: fix UAF in flow counter release
Previous releases - always broken:
- core: forbid FDB status change while nexthop is in a group
- smc: fix warning in smc_rx_splice() when calling get_page()
- can: provide missing ndo_change_mtu(), to prevent buffer overflow.
- eth:
- i40e: fix VF config validation
- broadcom: fix support for PTP_EXTTS_REQUEST2 ioctl"
* tag 'net-6.17-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (40 commits)
octeontx2-pf: Fix potential use after free in otx2_tc_add_flow()
net: dsa: lantiq_gswip: suppress -EINVAL errors for bridge FDB entries added to the CPU port
net: dsa: lantiq_gswip: move gswip_add_single_port_br() call to port_setup()
libie: fix string names for AQ error codes
net/mlx5e: Fix missing FEC RS stats for RS_544_514_INTERLEAVED_QUAD
net/mlx5: HWS, ignore flow level for multi-dest table
net/mlx5: fs, fix UAF in flow counter release
selftests: fib_nexthops: Add test cases for FDB status change
selftests: fib_nexthops: Fix creation of non-FDB nexthops
nexthop: Forbid FDB status change while nexthop is in a group
net: allow alloc_skb_with_frags() to use MAX_SKB_FRAGS
bnxt_en: correct offset handling for IPv6 destination address
ptp: document behavior of PTP_STRICT_FLAGS
broadcom: fix support for PTP_EXTTS_REQUEST2 ioctl
broadcom: fix support for PTP_PEROUT_DUTY_CYCLE
Bluetooth: MGMT: Fix possible UAFs
Bluetooth: hci_event: Fix UAF in hci_acl_create_conn_sync
Bluetooth: hci_event: Fix UAF in hci_conn_tx_dequeue
Bluetooth: hci_sync: Fix hci_resume_advertising_sync
Bluetooth: Fix build after header cleanup
...
Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost
Pull virtio fixes from Michael Tsirkin:
"virtio,vhost: last minute fixes
More small fixes. Most notably this fixes crashes and hangs in
vhost-net"
* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
MAINTAINERS, mailmap: Update address for Peter Hilber
virtio_config: clarify output parameters
uapi: vduse: fix typo in comment
vhost: Take a reference on the task in struct vhost_task.
vhost-net: flush batched before enabling notifications
Revert "vhost/net: Defer TX queue re-enable until after sendmsg"
vhost-net: unbreak busy polling
vhost-scsi: fix argument order in tport allocation error message
Daniel Lee [Wed, 24 Sep 2025 18:17:17 +0000 (14:17 -0400)]
platform/x86: lg-laptop: Fix WMAB call in fan_mode_store()
When WMAB is called to set the fan mode, the new mode is read from either
bits 0-1 or bits 4-5 (depending on the value of some other EC register).
Thus when WMAB is called with bits 4-5 zeroed and called again with
bits 0-1 zeroed, the second call undoes the effect of the first call.
This causes writes to /sys/devices/platform/lg-laptop/fan_mode to have
no effect (and causes reads to always report a status of zero).
Fix this by calling WMAB once, with the mode set in bits 0,1 and 4,5.
When the fan mode is returned from WMAB it always has this form, so
there is no need to preserve the other bits. As a bonus, the driver
now supports the "Performance" fan mode seen in the LG-provided Windows
control app, which provides less aggressive CPU throttling but louder
fan noise and shorter battery life.
Also, correct the documentation to reflect that 0 corresponds to the
default mode (what the Windows app calls "Optimal") and 1 corresponds
to the silent mode.
Dan Carpenter [Tue, 23 Sep 2025 11:19:11 +0000 (14:19 +0300)]
octeontx2-pf: Fix potential use after free in otx2_tc_add_flow()
This code calls kfree_rcu(new_node, rcu) and then dereferences "new_node"
and then dereferences it on the next line. Two lines later, we take
a mutex so I don't think this is an RCU safe region. Re-order it to do
the dereferences before queuing up the free.
Fixes: 68fbff68dbea ("octeontx2-pf: Add police action for TC flower") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev> Link: https://patch.msgid.link/aNKCL1jKwK8GRJHh@stanley.mountain Signed-off-by: Paolo Abeni <pabeni@redhat.com>
drm/panthor: Defer scheduler entitiy destruction to queue release
Commit de8548813824 ("drm/panthor: Add the scheduler logical block")
handled destruction of a group's queues' drm scheduler entities early
into the group destruction procedure.
However, that races with the group submit ioctl, because by the time
entities are destroyed (through the group destroy ioctl), the submission
procedure might've already obtained a group handle, and therefore the
ability to push jobs into entities. This is met with a DRM error message
within the drm scheduler core as a situation that should never occur.
Fix by deferring drm scheduler entity destruction to queue release time.
Fixes: de8548813824 ("drm/panthor: Add the scheduler logical block") Signed-off-by: Adrián Larumbe <adrian.larumbe@collabora.com> Reviewed-by: Steven Price <steven.price@arm.com> Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com> Signed-off-by: Steven Price <steven.price@arm.com> Link: https://lore.kernel.org/r/20250919164436.531930-1-adrian.larumbe@collabora.com
Paolo Abeni [Thu, 25 Sep 2025 08:29:22 +0000 (10:29 +0200)]
Merge branch 'lantiq_gswip-fixes'
Vladimir Oltean says:
====================
lantiq_gswip fixes
This is a small set of fixes which I believe should be backported for
the lantiq_gswip driver. Daniel Golle asked me to submit them here:
https://lore.kernel.org/netdev/aLiDfrXUbw1O5Vdi@pidgin.makrotopia.org/
As mentioned there, a merge conflict with net-next is expected, due to
the movement of the driver to the 'drivers/net/dsa/lantiq' folder there.
Good luck :-/
Patch 2/2 fixes an old regression and is the minimal fix for that, as
discussed here:
https://lore.kernel.org/netdev/aJfNMLNoi1VOsPrN@pidgin.makrotopia.org/
Patch 1/2 was identified by me through static analysis, and I consider
it to be a serious deficiency. It needs a test tag.
====================
Vladimir Oltean [Thu, 18 Sep 2025 07:21:42 +0000 (10:21 +0300)]
net: dsa: lantiq_gswip: suppress -EINVAL errors for bridge FDB entries added to the CPU port
The blamed commit and others in that patch set started the trend
of reusing existing DSA driver API for a new purpose: calling
ds->ops->port_fdb_add() on the CPU port.
The lantiq_gswip driver was not prepared to handle that, as can be seen
from the many errors that Daniel presents in the logs:
[ 174.050000] gswip 1e108000.switch: port 2 failed to add fa:aa:72:f4:8b:1e vid 1 to fdb: -22
[ 174.060000] gswip 1e108000.switch lan2: entered promiscuous mode
[ 174.070000] gswip 1e108000.switch: port 2 failed to add 00:01:02:03:04:02 vid 0 to fdb: -22
[ 174.090000] gswip 1e108000.switch: port 2 failed to add 00:01:02:03:04:02 vid 1 to fdb: -22
[ 174.090000] gswip 1e108000.switch: port 2 failed to delete fa:aa:72:f4:8b:1e vid 1 from fdb: -2
The errors are because gswip_port_fdb() wants to get a handle to the
bridge that originated these FDB events, to associate it with a FID.
Absolutely honourable purpose, however this only works for user ports.
To get the bridge that generated an FDB entry for the CPU port, one
would need to look at the db.bridge.dev argument. But this was
introduced in commit c26933639b54 ("net: dsa: request drivers to perform
FDB isolation"), first appeared in v5.18, and when the blamed commit was
introduced in v5.14, no such API existed.
So the core DSA feature was introduced way too soon for lantiq_gswip.
Not acting on these host FDB entries and suppressing any errors has no
other negative effect, and practically returns us to not supporting the
host filtering feature at all - peacefully, this time.
Fixes: 10fae4ac89ce ("net: dsa: include bridge addresses which are local in the host fdb list") Reported-by: Daniel Golle <daniel@makrotopia.org> Closes: https://lore.kernel.org/netdev/aJfNMLNoi1VOsPrN@pidgin.makrotopia.org/ Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Link: https://patch.msgid.link/20250918072142.894692-3-vladimir.oltean@nxp.com Tested-by: Daniel Golle <daniel@makrotopia.org> Reviewed-by: Daniel Golle <daniel@makrotopia.org> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Vladimir Oltean [Thu, 18 Sep 2025 07:21:41 +0000 (10:21 +0300)]
net: dsa: lantiq_gswip: move gswip_add_single_port_br() call to port_setup()
A port added to a "single port bridge" operates as standalone, and this
is mutually exclusive to being part of a Linux bridge. In fact,
gswip_port_bridge_join() calls gswip_add_single_port_br() with
add=false, i.e. removes the port from the "single port bridge" to enable
autonomous forwarding.
The blamed commit seems to have incorrectly thought that ds->ops->port_enable()
is called one time per port, during the setup phase of the switch.
However, it is actually called during the ndo_open() implementation of
DSA user ports, which is to say that this sequence of events:
1. ip link set swp0 down
2. ip link add br0 type bridge
3. ip link set swp0 master br0
4. ip link set swp0 up
would cause swp0 to join back the "single port bridge" which step 3 had
just removed it from.
The correct DSA hook for one-time actions per port at switch init time
is ds->ops->port_setup(). This is what seems to match the coder's
intention; also see the comment at the beginning of the file:
* At the initialization the driver allocates one bridge table entry for
~~~~~~~~~~~~~~~~~~~~~
* each switch port which is used when the port is used without an
* explicit bridge.
Fixes: 8206e0ce96b3 ("net: dsa: lantiq: Add VLAN unaware bridge offloading") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Link: https://patch.msgid.link/20250918072142.894692-2-vladimir.oltean@nxp.com Tested-by: Daniel Golle <daniel@makrotopia.org> Reviewed-by: Daniel Golle <daniel@makrotopia.org> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Peter Zijlstra [Wed, 17 Sep 2025 10:03:20 +0000 (12:03 +0200)]
sched/deadline: Fix dl_server behaviour
John reported undesirable behaviour with the dl_server since commit: cccb45d7c4295 ("sched/deadline: Less agressive dl_server handling").
When starving fair tasks on purpose (starting spinning FIFO tasks),
his fair workload, which often goes (briefly) idle, would delay fair
invocations for a second, running one invocation per second was both
unexpected and terribly slow.
The reason this happens is that when dl_se->server_pick_task() returns
NULL, indicating no runnable tasks, it would yield, pushing any later
jobs out a whole period (1 second).
Instead simply stop the server. This should restore behaviour in that
a later wakeup (which restarts the server) will be able to continue
running (subject to the CBS wakeup rules).
Notably, this does not re-introduce the behaviour cccb45d7c4295 set
out to solve, any start/stop cycle is naturally throttled by the timer
period (no active cancel).
Fixes: cccb45d7c4295 ("sched/deadline: Less agressive dl_server handling") Reported-by: John Stultz <jstultz@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Tested-by: John Stultz <jstultz@google.com>
Peter Zijlstra [Tue, 16 Sep 2025 21:02:41 +0000 (23:02 +0200)]
sched/deadline: Fix dl_server getting stuck
John found it was easy to hit lockup warnings when running locktorture
on a 2 CPU VM, which he bisected down to: commit cccb45d7c429
("sched/deadline: Less agressive dl_server handling").
While debugging it seems there is a chance where we end up with the
dl_server dequeued, with dl_se->dl_server_active. This causes
dl_server_start() to return without enqueueing the dl_server, thus it
fails to run when RT tasks starve the cpu.
When this happens, dl_server_timer() catches the
'!dl_se->server_has_tasks(dl_se)' case, which then calls
replenish_dl_entity() and dl_server_stopped() and finally return
HRTIMER_NO_RESTART.
This ends in no new timer and also no enqueue, leaving the dl_server
'dead', allowing starvation.
What should have happened is for the bandwidth timer to start the
zero-laxity timer, which in turn would enqueue the dl_server and cause
dl_se->server_pick_task() to be called -- which will stop the
dl_server if no fair tasks are observed for a whole period.
IOW, it is totally irrelevant if there are fair tasks at the moment of
bandwidth refresh.
This removes all dl_se->server_has_tasks() users, so remove the whole
thing.
Fixes: cccb45d7c4295 ("sched/deadline: Less agressive dl_server handling") Reported-by: John Stultz <jstultz@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Tested-by: John Stultz <jstultz@google.com>
Zhen Ni [Tue, 23 Sep 2025 07:51:04 +0000 (15:51 +0800)]
afs: Fix potential null pointer dereference in afs_put_server
afs_put_server() accessed server->debug_id before the NULL check, which
could lead to a null pointer dereference. Move the debug_id assignment,
ensuring we never dereference a NULL server pointer.
Fixes: 2757a4dc1849 ("afs: Fix access after dec in put functions") Cc: stable@vger.kernel.org Signed-off-by: Zhen Ni <zhen.ni@easystack.cn> Acked-by: David Howells <dhowells@redhat.com> Reviewed-by: Jeffrey Altman <jaltman@auristor.com> Signed-off-by: Christian Brauner <brauner@kernel.org>
Merge tag 'probes-fixes-v6.17-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull probes fixes from Masami Hiramatsu:
- fprobe: Even if there is a memory allocation failure, try to remove
the addresses recorded until then from the filter. Previously we just
skipped it.
- tracing: dynevent: Add a missing lockdown check on dynevent. This
dynevent is the interface for all probe events. Thus if there is no
check, any probe events can be added after lock down the tracefs.
* tag 'probes-fixes-v6.17-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
tracing: dynevent: Add a missing lockdown check on dynevent
tracing: fprobe: Fix to remove recorded module addresses from filter
Jacob Keller [Tue, 23 Sep 2025 20:56:56 +0000 (13:56 -0700)]
libie: fix string names for AQ error codes
The LIBIE_AQ_STR macro() introduced by commit 5feaa7a07b85 ("libie: add
adminq helper for converting err to str") is used in order to generate
strings for printing human readable error codes. Its definition is missing
the separating underscore ('_') character which makes the resulting strings
difficult to read. Additionally, the string won't match the source code,
preventing search tools from working properly.
Add the missing underscore character, fixing the error string names.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Fixes: 5feaa7a07b85 ("libie: add adminq helper for converting err to str") Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Link: https://patch.msgid.link/20250923205657.846759-1-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Eric Biggers [Wed, 24 Sep 2025 20:18:22 +0000 (13:18 -0700)]
crypto: af_alg - Fix incorrect boolean values in af_alg_ctx
Commit 1b34cbbf4f01 ("crypto: af_alg - Disallow concurrent writes in
af_alg_sendmsg") changed some fields from bool to 1-bit bitfields of
type u32.
However, some assignments to these fields, specifically 'more' and
'merge', assign values greater than 1. These relied on C's implicit
conversion to bool, such that zero becomes false and nonzero becomes
true.
With a 1-bit bitfields of type u32 instead, mod 2 of the value is taken
instead, resulting in 0 being assigned in some cases when 1 was intended.
Fix this by restoring the bool type.
Fixes: 1b34cbbf4f01 ("crypto: af_alg - Disallow concurrent writes in af_alg_sendmsg") Cc: stable@vger.kernel.org Signed-off-by: Eric Biggers <ebiggers@kernel.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge tag 'soc-fixes-6.17-3' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc
Pull SoC fixes from Arnd Bergmann:
"There are a few minor code fixes for tegra firmware, i.MX firmware
and the eyeq reset controller, and a MAINTAINERS update as Alyssa
Rosenzweig moves on to non-kernel projects.
The other changes are all for devicetree files:
- Multiple Marvell Armada SoCs need changes to fix PCIe, audio and
SATA
- A socfpga board fails to probe the ethernet phy
- The two temperature sensors on i.MX8MP are swapped
- Allwinner devicetree files cause build-time warnings
- Two Rockchip based boards need corrections for headphone detection
and SPI flash"
* tag 'soc-fixes-6.17-3' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc:
MAINTAINERS: remove Alyssa Rosenzweig
firmware: tegra: Do not warn on missing memory-region property
arm64: dts: marvell: cn9132-clearfog: fix multi-lane pci x2 and x4 ports
arm64: dts: marvell: cn9132-clearfog: disable eMMC high-speed modes
arm64: dts: marvell: cn913x-solidrun: fix sata ports status
ARM: dts: kirkwood: Fix sound DAI cells for OpenRD clients
arm64: dts: imx8mp: Correct thermal sensor index
ARM: imx: Kconfig: Adjust select after renamed config option
firmware: imx: Add stub functions for SCMI CPU API
firmware: imx: Add stub functions for SCMI LMM API
firmware: imx: Add stub functions for SCMI MISC API
riscv: dts: allwinner: rename devterm i2c-gpio node to comply with binding
arm64: dts: rockchip: Fix the headphone detection on the orangepi 5
arm64: dts: rockchip: Add vcc supply for SPI Flash on NanoPC-T6
ARM: dts: socfpga: sodia: Fix mdio bus probe and PHY address
reset: eyeq: fix OF node leak
ARM64: dts: mcbin: fix SATA ports on Macchiatobin
ARM: dts: armada-370-db: Fix stereo audio input routing on Armada 370
ARM: dts: allwinner: Minor whitespace cleanup
Merge tag 'for-6.17-rc7-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fix from David Sterba:
"One more regression fix for a problem in zoned mode: mounting would
fail if the number of open and active zones reached a common limit
that didn't use to be checked"
* tag 'for-6.17-rc7-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
btrfs: zoned: don't fail mount needlessly due to too many active zones
Merge tag '6.17-rc7-ksmbd-server-fixes' of git://git.samba.org/ksmbd
Pull smb server fixes from Steve French:
- free_transport fix for disconnect races
- minor delayed work fix
* tag '6.17-rc7-ksmbd-server-fixes' of git://git.samba.org/ksmbd:
smb: server: use disable_work_sync in transport_rdma.c
smb: server: don't use delayed_work for post_recv_credits_work
tracing: dynevent: Add a missing lockdown check on dynevent
Since dynamic_events interface on tracefs is compatible with
kprobe_events and uprobe_events, it should also check the lockdown
status and reject if it is set.
tracing: fprobe: Fix to remove recorded module addresses from filter
Even if there is a memory allocation failure in fprobe_addr_list_add(),
there is a partial list of module addresses. So remove the recorded
addresses from filter if exists.
This also removes the redundant ret local variable.
error: cannot jump from this asm goto statement to one of its possible targets
There are other failure scenarios. Shuffling code around slightly makes it
worse and fail even with one instance.
That issue prevents using local labels for a cleanup based user access
mechanism.
After failed attempts to provide a simple enough test case for the 'depends
on' test in Kconfig, the initial cure was to mark ASM goto broken on clang
versions < 17 to get this road block out of the way.
But Nathan pointed out that this is a known clang issue and indeed affects
clang < version 17 in combination with cleanup(). It's not even required to
use local labels for that.
The clang issue tracker has a small enough test case, which can be used as
a test in the 'depends on' section of CC_HAS_ASM_GOTO_OUTPUT:
Add another dependency to config CC_HAS_ASM_GOTO_OUTPUT for it and use the
clang issue tracker test case for detection by condensing it to obfuscated
C-code contest format. This reliably catches the problem on clang < 17 and
did not show any issues on the non broken GCC versions.
That test might be sufficient to catch all issues and therefore could
replace the existing test, but keeping that around does no harm either.
Thanks to Nathan for pointing to the relevant clang issue!
futex: Use correct exit on failure from futex_hash_allocate_default()
copy_process() uses the wrong error exit path from futex_hash_allocate_default().
After exiting from futex_hash_allocate_default(), neither tasklist_lock
nor siglock has been acquired. The exit label bad_fork_core_free unlocks
both of these locks which is wrong.
The next exit label, bad_fork_cancel_cgroup, is the correct exit.
sched_cgroup_fork() did not allocate any resources that need to freed.
Use bad_fork_cancel_cgroup on error exit from futex_hash_allocate_default().
Fixes: 7c4f75a21f636 ("futex: Allow automatic allocation of process wide futex hash") Reported-by: syzbot+80cb3cc5c14fad191a10@syzkaller.appspotmail.com Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org> Closes: https://lore.kernel.org/all/68cb1cbd.050a0220.2ff435.0599.GAE@google.com
Paul Walmsley [Wed, 24 Sep 2025 00:25:52 +0000 (18:25 -0600)]
MAINTAINERS: Update Paul Walmsley's E-mail address
My experiment with using corporate Gmail for Linux kernel list
interaction has come to an end. For my MAINTAINERS entries that
use that E-mail address, let's switch those to use the k.org E-mail
forwarding.
Signed-off-by: Paul Walmsley <paul.walmsley@sifive.com> Signed-off-by: Paul Walmsley <pjw@kernel.org>
riscv: Use an atomic xchg in pudp_huge_get_and_clear()
Make sure we return the right pud value and not a value that could
have been overwritten in between by a different core.
Fixes: c3cc2a4a3a23 ("riscv: Add support for PUD THP") Cc: stable@vger.kernel.org Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250814-dev-alex-thp_pud_xchg-v1-1-b4704dfae206@rivosinc.com
[pjw@kernel.org: use xchg rather than atomic_long_xchg; avoid atomic op for !CONFIG_SMP like x86] Signed-off-by: Paul Walmsley <pjw@kernel.org>
net/mlx5e: Fix missing FEC RS stats for RS_544_514_INTERLEAVED_QUAD
Include MLX5E_FEC_RS_544_514_INTERLEAVED_QUAD in the FEC RS stats
handling. This addresses a gap introduced when adding support for
200G/lane link modes.
Fixes: 4e343c11efbb ("net/mlx5e: Support FEC settings for 200G per lane link modes") Signed-off-by: Carolina Jubran <cjubran@nvidia.com> Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com> Reviewed-by: Yael Chemla <ychemla@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Link: https://patch.msgid.link/1758525094-816583-4-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net/mlx5: HWS, ignore flow level for multi-dest table
When HWS creates multi-dest FW table and adds rules to
forward to other tables, ignore the flow level enforcement
in FW, because HWS is responsible for table levels.
This fixes the following error:
mlx5_core 0000:08:00.0: mlx5_cmd_out_err:818:(pid 192306):
SET_FLOW_TABLE_ENTRY(0x936) op_mod(0x0) failed,
status bad parameter(0x3), syndrome (0x6ae84c), err(-22)
Fix a kernel trace [1] caused by releasing an HWS action of a local flow
counter in mlx5_cmd_hws_delete_fte(), where the HWS action refcount and
mutex were not initialized and the counter struct could already be freed
when deleting the rule.
Fix it by adding the missing initializations and adding refcount for the
local flow counter struct.
selftests: fib_nexthops: Add test cases for FDB status change
Add the following test cases for both IPv4 and IPv6:
* Can change from FDB nexthop to non-FDB nexthop and vice versa.
* Can change FDB nexthop address while in a group.
* Cannot change from FDB nexthop to non-FDB nexthop and vice versa while
in a group.
Output without "nexthop: Forbid FDB status change while nexthop is in a
group":
IPv6 fdb groups functional
--------------------------
[...]
TEST: Replace FDB nexthop to non-FDB nexthop [ OK ]
TEST: Replace non-FDB nexthop to FDB nexthop [ OK ]
TEST: Replace FDB nexthop address while in a group [ OK ]
TEST: Replace FDB nexthop to non-FDB nexthop while in a group [FAIL]
TEST: Replace non-FDB nexthop to FDB nexthop while in a group [FAIL]
[...]
IPv4 fdb groups functional
--------------------------
[...]
TEST: Replace FDB nexthop to non-FDB nexthop [ OK ]
TEST: Replace non-FDB nexthop to FDB nexthop [ OK ]
TEST: Replace FDB nexthop address while in a group [ OK ]
TEST: Replace FDB nexthop to non-FDB nexthop while in a group [FAIL]
TEST: Replace non-FDB nexthop to FDB nexthop while in a group [FAIL]
[...]
Tests passed: 36
Tests failed: 4
Tests skipped: 0
Output with "nexthop: Forbid FDB status change while nexthop is in a
group":
IPv6 fdb groups functional
--------------------------
[...]
TEST: Replace FDB nexthop to non-FDB nexthop [ OK ]
TEST: Replace non-FDB nexthop to FDB nexthop [ OK ]
TEST: Replace FDB nexthop address while in a group [ OK ]
TEST: Replace FDB nexthop to non-FDB nexthop while in a group [ OK ]
TEST: Replace non-FDB nexthop to FDB nexthop while in a group [ OK ]
[...]
IPv4 fdb groups functional
--------------------------
[...]
TEST: Replace FDB nexthop to non-FDB nexthop [ OK ]
TEST: Replace non-FDB nexthop to FDB nexthop [ OK ]
TEST: Replace FDB nexthop address while in a group [ OK ]
TEST: Replace FDB nexthop to non-FDB nexthop while in a group [ OK ]
TEST: Replace non-FDB nexthop to FDB nexthop while in a group [ OK ]
[...]
IPv6 fdb groups functional
--------------------------
[...]
COMMAND: ip -netns me-nRsN3E nexthop add id 63 via 2001:db8:91::4
Error: Device attribute required for non-blackhole and non-fdb nexthops.
COMMAND: ip -netns me-nRsN3E nexthop add id 64 via 2001:db8:91::5
Error: Device attribute required for non-blackhole and non-fdb nexthops.
COMMAND: ip -netns me-nRsN3E nexthop add id 103 group 63/64 fdb
Error: Invalid nexthop id.
TEST: Fdb Nexthop group with non-fdb nexthops [ OK ]
[...]
IPv4 fdb groups functional
--------------------------
[...]
COMMAND: ip -netns me-nRsN3E nexthop add id 14 via 172.16.1.2
Error: Device attribute required for non-blackhole and non-fdb nexthops.
COMMAND: ip -netns me-nRsN3E nexthop add id 15 via 172.16.1.3
Error: Device attribute required for non-blackhole and non-fdb nexthops.
COMMAND: ip -netns me-nRsN3E nexthop add id 103 group 14/15 fdb
Error: Invalid nexthop id.
TEST: Fdb Nexthop group with non-fdb nexthops [ OK ]
COMMAND: ip -netns me-nRsN3E nexthop add id 16 via 172.16.1.2 fdb
COMMAND: ip -netns me-nRsN3E nexthop add id 17 via 172.16.1.3 fdb
COMMAND: ip -netns me-nRsN3E nexthop add id 104 group 14/15
Error: Invalid nexthop id.
TEST: Non-Fdb Nexthop group with fdb nexthops [ OK ]
[...]
COMMAND: ip -netns me-0dlhyd ro add 172.16.0.0/22 nhid 15
Error: Nexthop id does not exist.
TEST: Route add with fdb nexthop [ OK ]
In addition, as can be seen in the above output, a couple of IPv4 test
cases used the non-FDB nexthops (14 and 15) when they intended to use
the FDB nexthops (16 and 17). These test cases only passed because
failure was expected, but they failed for the wrong reason.
Fix the test to create the non-FDB nexthops with a nexthop device and
adjust the IPv4 test cases to use the FDB nexthops instead of the
non-FDB nexthops.
IPv6 fdb groups functional
--------------------------
[...]
COMMAND: ip -netns me-lNzfHP nexthop add id 63 via 2001:db8:91::4 dev veth1
COMMAND: ip -netns me-lNzfHP nexthop add id 64 via 2001:db8:91::5 dev veth1
COMMAND: ip -netns me-lNzfHP nexthop add id 103 group 63/64 fdb
Error: FDB nexthop group can only have fdb nexthops.
TEST: Fdb Nexthop group with non-fdb nexthops [ OK ]
[...]
IPv4 fdb groups functional
--------------------------
[...]
COMMAND: ip -netns me-lNzfHP nexthop add id 14 via 172.16.1.2 dev veth1
COMMAND: ip -netns me-lNzfHP nexthop add id 15 via 172.16.1.3 dev veth1
COMMAND: ip -netns me-lNzfHP nexthop add id 103 group 14/15 fdb
Error: FDB nexthop group can only have fdb nexthops.
TEST: Fdb Nexthop group with non-fdb nexthops [ OK ]
COMMAND: ip -netns me-lNzfHP nexthop add id 16 via 172.16.1.2 fdb
COMMAND: ip -netns me-lNzfHP nexthop add id 17 via 172.16.1.3 fdb
COMMAND: ip -netns me-lNzfHP nexthop add id 104 group 16/17
Error: Non FDB nexthop group cannot have fdb nexthops.
TEST: Non-Fdb Nexthop group with fdb nexthops [ OK ]
[...]
COMMAND: ip -netns me-lNzfHP ro add 172.16.0.0/22 nhid 16
Error: Route cannot point to a fdb nexthop.
TEST: Route add with fdb nexthop [ OK ]
[...]
Tests passed: 30
Tests failed: 0
Tests skipped: 0
nexthop: Forbid FDB status change while nexthop is in a group
The kernel forbids the creation of non-FDB nexthop groups with FDB
nexthops:
# ip nexthop add id 1 via 192.0.2.1 fdb
# ip nexthop add id 2 group 1
Error: Non FDB nexthop group cannot have fdb nexthops.
And vice versa:
# ip nexthop add id 3 via 192.0.2.2 dev dummy1
# ip nexthop add id 4 group 3 fdb
Error: FDB nexthop group can only have fdb nexthops.
However, as long as no routes are pointing to a non-FDB nexthop group,
the kernel allows changing the type of a nexthop from FDB to non-FDB and
vice versa:
# ip nexthop add id 5 via 192.0.2.2 dev dummy1
# ip nexthop add id 6 group 5
# ip nexthop replace id 5 via 192.0.2.2 fdb
# echo $?
0
This configuration is invalid and can result in a NPD [1] since FDB
nexthops are not associated with a nexthop device:
# ip route add 198.51.100.1/32 nhid 6
# ping 198.51.100.1
Fix by preventing nexthop FDB status change while the nexthop is in a
group:
# ip nexthop add id 7 via 192.0.2.2 dev dummy1
# ip nexthop add id 8 group 7
# ip nexthop replace id 7 via 192.0.2.2 fdb
Error: Cannot change nexthop FDB status while in a group.
Jason Baron [Mon, 22 Sep 2025 19:19:57 +0000 (15:19 -0400)]
net: allow alloc_skb_with_frags() to use MAX_SKB_FRAGS
Currently, alloc_skb_with_frags() will only fill (MAX_SKB_FRAGS - 1)
slots. I think it should use all MAX_SKB_FRAGS slots, as callers of
alloc_skb_with_frags() will size their allocation of frags based
on MAX_SKB_FRAGS.
This issue was discovered via a test patch that sets 'order' to 0
in alloc_skb_with_frags(), which effectively tests/simulates high
fragmentation. In this case sendmsg() on unix sockets will fail every
time for large allocations. If the PAGE_SIZE is 4K, then data_len will
request 68K or 17 pages, but alloc_skb_with_frags() can only allocate
64K in this case or 16 pages.
Fixes: 09c2c90705bb ("net: allow alloc_skb_with_frags() to allocate bigger packets") Signed-off-by: Jason Baron <jbaron@akamai.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/20250922191957.2855612-1-jbaron@akamai.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
projects / thirdparty / kernel / stable.git / log
