]> git.ipfire.org Git - thirdparty/linux.git/log
thirdparty/linux.git
10 days agoLoongArch: Adjust default config files for 32BIT/64BIT
Huacai Chen [Mon, 8 Dec 2025 10:09:17 +0000 (18:09 +0800)] 
LoongArch: Adjust default config files for 32BIT/64BIT

Add loongson32_defconfig (for 32BIT) and rename loongson3_defconfig to
loongson64_defconfig (for 64BIT).

Also adjust graphics drivers, such as FB_EFI is replaced with EFIDRM.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
10 days agoLoongArch: Adjust VDSO/VSYSCALL for 32BIT/64BIT
Huacai Chen [Mon, 8 Dec 2025 10:09:17 +0000 (18:09 +0800)] 
LoongArch: Adjust VDSO/VSYSCALL for 32BIT/64BIT

Adjust VDSO/VSYSCALL because read_cpu_id() for 32BIT/64BIT are
different, and LoongArch32 doesn't support GENERIC_GETTIMEOFDAY now
(will be supported in future).

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
10 days agoLoongArch: Adjust misc routines for 32BIT/64BIT
Huacai Chen [Mon, 8 Dec 2025 10:09:17 +0000 (18:09 +0800)] 
LoongArch: Adjust misc routines for 32BIT/64BIT

Adjust misc routines for both 32BIT and 64BIT, including: bitops, bswap,
checksum, string, jump label, unaligned access emulator, suspend/wakeup
routines, etc.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
10 days agoLoongArch: Adjust user accessors for 32BIT/64BIT
Huacai Chen [Mon, 8 Dec 2025 10:09:17 +0000 (18:09 +0800)] 
LoongArch: Adjust user accessors for 32BIT/64BIT

Adjust user accessors for both 32BIT and 64BIT, including: get_user(),
put_user(), copy_user(), clear_user(), etc.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
10 days agoLoongArch: Adjust system call for 32BIT/64BIT
Huacai Chen [Mon, 8 Dec 2025 10:09:17 +0000 (18:09 +0800)] 
LoongArch: Adjust system call for 32BIT/64BIT

Adjust system call for both 32BIT and 64BIT, including: add the uapi
unistd_{32,64}.h and syscall_table_{32,64}.h inclusion, add sys_mmap2()
definition, change the system call entry routines, etc.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
10 days agoLoongArch: Adjust module loader for 32BIT/64BIT
Huacai Chen [Mon, 8 Dec 2025 10:09:17 +0000 (18:09 +0800)] 
LoongArch: Adjust module loader for 32BIT/64BIT

Adjust module loader for both 32BIT and 64BIT, including: change the s64
type to long, change the u64 type to unsigned long, change the plt entry
definition and handling, etc.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
10 days agoLoongArch: Adjust time routines for 32BIT/64BIT
Huacai Chen [Mon, 8 Dec 2025 10:09:17 +0000 (18:09 +0800)] 
LoongArch: Adjust time routines for 32BIT/64BIT

Adjust time routines for both 32BIT and 64BIT, including: rdtime_h() /
rdtime_l() definitions for 32BIT and rdtime_d() definition for 64BIT,
get_cycles() and get_cycles64() definitions for 32BIT/64BIT, show time
frequency info ("CPU MHz" and "BogoMIPS") in /proc/cpuinfo, etc.

Use do_div() for division which works on both 32BIT and 64BIT platforms.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
10 days agoLoongArch: Adjust process management for 32BIT/64BIT
Huacai Chen [Mon, 8 Dec 2025 10:09:17 +0000 (18:09 +0800)] 
LoongArch: Adjust process management for 32BIT/64BIT

Adjust process management for both 32BIT and 64BIT, including: CPU
context switching, FPU loading/restoring, process dumping and process
tracing routines.

Q: Why modify switch.S?
A: LoongArch32 has no ldptr.d/stptr.d instructions, and asm offsets of
   thead_struct members are too large to be filled in the 12b immediate
   field of ld.w/st.w.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
10 days agoLoongArch: Adjust memory management for 32BIT/64BIT
Huacai Chen [Mon, 8 Dec 2025 10:09:17 +0000 (18:09 +0800)] 
LoongArch: Adjust memory management for 32BIT/64BIT

Adjust memory management for both 32BIT and 64BIT, including: address
space definition, DMW CSR definition, page table bits definition, boot
time detection of VA/PA bits, page table init, tlb exception handling,
copy_page/clear_page/dump_tlb libraries, etc.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Yawei Li <liyawei@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
10 days agoLoongArch: Adjust boot & setup for 32BIT/64BIT
Huacai Chen [Mon, 8 Dec 2025 10:09:17 +0000 (18:09 +0800)] 
LoongArch: Adjust boot & setup for 32BIT/64BIT

Adjust boot & setup for both 32BIT and 64BIT, including: efi header
definition, MAX_IO_PICS definition, kernel entry and environment setup
routines, etc.

Add a fallback path in fdt_cpu_clk_init() to avoid 0MHz in /proc/cpuinfo
if there is no valid clock freq from firmware.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
13 days agoLoongArch: Adjust common macro definitions for 32BIT/64BIT
Huacai Chen [Sat, 6 Dec 2025 02:40:32 +0000 (10:40 +0800)] 
LoongArch: Adjust common macro definitions for 32BIT/64BIT

Most common macros are defined in asm.h, asmmacro.h and stackframe.h.
Adjust these macros for both 32BIT and 64BIT.

Add SETUP_TWINS (Setup Trampoline Windows) and SETUP_MODES (Setup CRMD/
PRMD/EUEN) which will be used later.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
13 days agoLoongArch: Add adaptive CSR accessors for 32BIT/64BIT
Huacai Chen [Sat, 6 Dec 2025 02:40:32 +0000 (10:40 +0800)] 
LoongArch: Add adaptive CSR accessors for 32BIT/64BIT

32BIT platforms only have 32bit CSR/IOCSR registers, 64BIT platforms
have both 32bit/64bit CSR/IOCSR registers. Now there are both 32bit and
64bit CSR accessors:

csr_read32()/csr_write32()/csr_xchg32();
csr_read64()/csr_write64()/csr_xchg64();

Some CSR registers (address and timer registers) are 32bit length on
32BIT platform and 64bit length on 64BIT platform. To avoid #ifdefs here
and there, they need adaptive accessors, so we define and use:

csr_read()/csr_write()/csr_xchg();

IOCSR doesn't have a "natural length", which means a 64bit register can
be treated as two 32bit registers, so we just use two 32bit accessors to
emulate a 64bit accessors.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
13 days agoLoongArch: Add atomic operations for 32BIT/64BIT
Huacai Chen [Sat, 6 Dec 2025 02:40:32 +0000 (10:40 +0800)] 
LoongArch: Add atomic operations for 32BIT/64BIT

LoongArch64 has both AMO and LL/SC instructions, while LoongArch32 only
has LL/SC intstructions. So we add a Kconfig option CPU_HAS_AMO here and
implement atomic operations (also including local operations and percpu
operations) for both 32BIT and 64BIT platforms.

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
13 days agoLoongArch: Add new PCI ID for pci_fixup_vgadev()
Huacai Chen [Sat, 6 Dec 2025 02:39:49 +0000 (10:39 +0800)] 
LoongArch: Add new PCI ID for pci_fixup_vgadev()

Loongson-2K3000 has a new PCI ID (0x7a46) for its display controller,
Add it for pci_fixup_vgadev() since we prefer a discrete graphics card
as default boot device if present.

Cc: stable@vger.kernel.org
Signed-off-by: Tianrui Zhao <zhaotianrui@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
13 days agoLoongArch: Add and use some macros for AVEC
Song Gao [Sat, 6 Dec 2025 02:39:49 +0000 (10:39 +0800)] 
LoongArch: Add and use some macros for AVEC

Add and use some macros for AVEC interrupt controller, instead of using
magic numbers.

Signed-off-by: Song Gao <gaosong@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
13 days agoLoongArch: Correct the calculation logic of thread_count
Qiang Ma [Sat, 6 Dec 2025 02:39:49 +0000 (10:39 +0800)] 
LoongArch: Correct the calculation logic of thread_count

For thread_count, the current calculation method has a maximum of 255,
which may not be sufficient in the future. Therefore, we are correcting
it now.

Reference: SMBIOS Specification, 7.5 Processor Information (Type 4)[1]

[1]: https://www.dmtf.org/sites/default/files/standards/documents/DSP0134_3.9.0.pdf

Cc: stable@vger.kernel.org
Signed-off-by: Qiang Ma <maqianga@uniontech.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
13 days agoLoongArch: Use unsigned long for _end and _text
Tiezhu Yang [Sat, 6 Dec 2025 02:39:48 +0000 (10:39 +0800)] 
LoongArch: Use unsigned long for _end and _text

It is better to use unsigned long rather than long for _end and _text to
calculate the kernel length.

Cc: stable@vger.kernel.org # v6.3+
Fixes: e5f02b51fa0c ("LoongArch: Add support for kernel address space layout randomization (KASLR)")
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
13 days agoLoongArch: Use __pmd()/__pte() for swap entry conversions
Yuli Wang [Sat, 6 Dec 2025 02:39:48 +0000 (10:39 +0800)] 
LoongArch: Use __pmd()/__pte() for swap entry conversions

The __pmd() and __pte() helper macros provide the correct initialization
syntax and abstraction for the pmd_t and pte_t types.

Use __pmd() to fix follow warning about __swp_entry_to_pmd() with gcc-15
under specific configs [1] :

  In file included from ./include/linux/pgtable.h:6,
                   from ./include/linux/mm.h:31,
                   from ./include/linux/pagemap.h:8,
                   from arch/loongarch/mm/init.c:14:
  ./include/linux/swapops.h: In function ‘swp_entry_to_pmd’:
  ./arch/loongarch/include/asm/pgtable.h:302:34: error: missing braces around initializer [-Werror=missing-braces]
    302 | #define __swp_entry_to_pmd(x)   ((pmd_t) { (x).val | _PAGE_HUGE })
        |                                  ^
  ./include/linux/swapops.h:559:16: note: in expansion of macro ‘__swp_entry_to_pmd’
    559 |         return __swp_entry_to_pmd(arch_entry);
        |                ^~~~~~~~~~~~~~~~~~
  cc1: all warnings being treated as errors

Also update __swp_entry_to_pte() to use __pte() for consistency.

[1]. https://download.01.org/0day-ci/archive/20251119/202511190316.luI90kAo-lkp@intel.com/config

Cc: stable@vger.kernel.org
Signed-off-by: Yuli Wang <wangyl5933@chinaunicom.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
13 days agoLoongArch: Fix arch_dup_task_struct() for CONFIG_RANDSTRUCT
Huacai Chen [Sat, 6 Dec 2025 02:39:48 +0000 (10:39 +0800)] 
LoongArch: Fix arch_dup_task_struct() for CONFIG_RANDSTRUCT

Now the optimized version of arch_dup_task_struct() for LoongArch
assumes 'thread' is the last member of 'task_struct'. But this is
not true if CONFIG_RANDSTRUCT is enabled after Linux-6.16.

So fix the arch_dup_task_struct() function for CONFIG_RANDSTRUCT by
copying the whole 'task_struct'.

Cc: stable@vger.kernel.org # 6.16+
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
13 days agoLoongArch: Fix build errors for CONFIG_RANDSTRUCT
Huacai Chen [Sat, 6 Dec 2025 02:39:40 +0000 (10:39 +0800)] 
LoongArch: Fix build errors for CONFIG_RANDSTRUCT

When CONFIG_RANDSTRUCT enabled, members of task_struct are randomized.
There is a chance that TASK_STACK_CANARY be out of 12bit immediate's
range and causes build errors. TASK_STACK_CANARY is naturally aligned,
so fix it by replacing ld.d/st.d with ldptr.d/stptr.d which have 14bit
immediates.

Cc: stable@vger.kernel.org
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202511240656.0NaPcJs1-lkp@intel.com/
Suggested-by: Rui Wang <wangrui@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
13 days agoLoongArch: Simplify __arch_bitrev32() implementation
Xi Ruoyao [Sat, 6 Dec 2025 02:39:39 +0000 (10:39 +0800)] 
LoongArch: Simplify __arch_bitrev32() implementation

LoongArch has the bitrev.w instruction to reverse bits in a 32-bit
integer, thus there's no need to reverse the bytes and use bitrev.4b.

Signed-off-by: Xi Ruoyao <xry111@xry111.site>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
13 days agoLoongArch: Select HAVE_ARCH_BITREVERSE in Kconfig
Xi Ruoyao [Sat, 6 Dec 2025 02:39:39 +0000 (10:39 +0800)] 
LoongArch: Select HAVE_ARCH_BITREVERSE in Kconfig

Without HAVE_ARCH_BITREVERSE, the architecture-optimized bit reverse
implementations in arch/loongarch/include/asm/bitrev.h are not used at
all.

Signed-off-by: Xi Ruoyao <xry111@xry111.site>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
2 weeks agoLinux 6.18 v6.18
Linus Torvalds [Sun, 30 Nov 2025 22:42:10 +0000 (14:42 -0800)] 
Linux 6.18

2 weeks agoMerge tag 'timers_urgent_for_v6.18_rc8' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 30 Nov 2025 16:47:10 +0000 (08:47 -0800)] 
Merge tag 'timers_urgent_for_v6.18_rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull timer fix from Borislav Petkov:

 - Have timekeeping aux clocks sysfs interface setup function return an
   error code on failure instead of success

* tag 'timers_urgent_for_v6.18_rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  timekeeping: Fix error code in tk_aux_sysfs_init()

2 weeks agoMerge tag 'mips-fixes_6.18_2' of git://git.kernel.org/pub/scm/linux/kernel/git/mips...
Linus Torvalds [Sat, 29 Nov 2025 23:15:14 +0000 (15:15 -0800)] 
Merge tag 'mips-fixes_6.18_2' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux

Pull MIPS fix from Thomas Bogendoerfer:
 "Fix TLB unification for cores with more than 64 TLB entries"

* tag 'mips-fixes_6.18_2' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux:
  MIPS: mm: kmalloc tlb_vpn array to avoid stack overflow

2 weeks agoMIPS: mm: kmalloc tlb_vpn array to avoid stack overflow
Thomas Bogendoerfer [Fri, 28 Nov 2025 16:53:46 +0000 (16:53 +0000)] 
MIPS: mm: kmalloc tlb_vpn array to avoid stack overflow

Owing to Config4.MMUSizeExt and VTLB/FTLB MMU features later MIPSr2+
cores can have more than 64 TLB entries.  Therefore allocate an array
for uniquification instead of placing too an small array on the stack.

Fixes: 35ad7e181541 ("MIPS: mm: tlb-r4k: Uniquify TLB entries on init")
Co-developed-by: Maciej W. Rozycki <macro@orcam.me.uk>
Signed-off-by: Maciej W. Rozycki <macro@orcam.me.uk>
Cc: stable@vger.kernel.org # v6.17+: 9f048fa48740: MIPS: mm: Prevent a TLB shutdown on initial uniquification
Cc: stable@vger.kernel.org # v6.17+
Tested-by: Gregory CLEMENT <gregory.clement@bootlin.com>
Tested-by: Klara Modin <klarasmodin@gmail.com>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
2 weeks agoafs: Fix uninit var in afs_alloc_anon_key()
David Howells [Sat, 29 Nov 2025 00:40:11 +0000 (00:40 +0000)] 
afs: Fix uninit var in afs_alloc_anon_key()

Fix an uninitialised variable (key) in afs_alloc_anon_key() by setting it
to cell->anonymous_key.  Without this change, the error check may return a
false failure with a bad error number.

Most of the time this is unlikely to happen because the first encounter
with afs_alloc_anon_key() will usually be from (auto)mount, for which all
subsequent operations must wait - apart from other (auto)mounts.  Once the
call->anonymous_key is allocated, all further calls to afs_request_key()
will skip the call to afs_alloc_anon_key() for that cell.

Fixes: d27c71257825 ("afs: Fix delayed allocation of a cell's anonymous key")
Reported-by: Paulo Alcantra <pc@manguebit.org>
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Paulo Alcantara <pc@manguebit.org>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: syzbot+41c68824eefb67cdf00c@syzkaller.appspotmail.com
cc: linux-afs@lists.infradead.org
cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2 weeks agoMerge tag 'spi-fix-v6.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/brooni...
Linus Torvalds [Fri, 28 Nov 2025 22:08:09 +0000 (14:08 -0800)] 
Merge tag 'spi-fix-v6.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi

Pull spi fixes from Mark Brown:
 "A disappointingly large set of device specific fixes that have built
  up since I've been a bit tardy with sending a pull requests as people
  kept sending me new new fixes.

  The bcm63xx and lpspi issues could lead to corruption so the fixes are
  fairly important for the affected parts, the other issues should all
  be relatively minor"

* tag 'spi-fix-v6.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
  spi: nxp-fspi: Propagate fwnode in ACPI case as well
  spi: tegra114: remove Kconfig dependency on TEGRA20_APB_DMA
  spi: amlogic-spifc-a1: Handle devm_pm_runtime_enable() errors
  spi: spi-fsl-lpspi: fix watermark truncation caused by type cast
  spi: cadence-quadspi: Fix cqspi_probe() error handling for runtime pm
  spi: bcm63xx: fix premature CS deassertion on RX-only transactions
  spi: spi-cadence-quadspi: Remove duplicate pm_runtime_put_autosuspend() call
  spi: spi-cadence-quadspi: Enable pm runtime earlier to avoid imbalance

2 weeks agoMerge tag 'regulator-fix-v6.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 28 Nov 2025 22:06:05 +0000 (14:06 -0800)] 
Merge tag 'regulator-fix-v6.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator

Pull regulator fixes from Mark Brown:
 "A couple of fixes for incorrect device descriptions in the rtq2208
  driver"

* tag 'regulator-fix-v6.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:
  regulator: rtq2208: Correct LDO2 logic judgment bits
  regulator: rtq2208: Correct buck group2 phase mapping logic

2 weeks agoMerge tag 'io_uring-6.18-20251128' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 28 Nov 2025 20:42:11 +0000 (12:42 -0800)] 
Merge tag 'io_uring-6.18-20251128' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux

Pull io_uring fixes from Jens Axboe:

 - Ensure that vectored registered buffer imports ties the lifetime of
   those to the zero-copy send notification, not the parent request

 - Fix a bug introduced in this merge window, with the introduction of
   mixed sized CQE support

* tag 'io_uring-6.18-20251128' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux:
  io_uring: fix mixed cqe overflow handling
  io_uring/net: ensure vectored buffer node import is tied to notification

2 weeks agoMerge tag 'vfs-6.18-rc8.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs
Linus Torvalds [Fri, 28 Nov 2025 18:01:24 +0000 (10:01 -0800)] 
Merge tag 'vfs-6.18-rc8.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs

Pull vfs fixes from Christian Brauner:

 - afs: Fix delayed allocation of a cell's anonymous key

   The allocation of a cell's anonymous key is done in a background
   thread along with other cell setup such as doing a DNS upcall. The
   normal key lookup tries to use the key description on the anonymous
   authentication key as the reference for request_key() - but it may
   not yet be set, causing an oops

 - ovl: fail ovl_lock_rename_workdir() if either target is unhashed

   As well as checking that the parent hasn't changed after getting the
   lock, the code needs to check that the dentry hasn't been unhashed.
   Otherwise overlayfs might try to rename something that has been
   removed

 - namespace: fix a reference leak in grab_requested_mnt_ns

   lookup_mnt_ns() already takes a reference on mnt_ns, and so
   grab_requested_mnt_ns() doesn't need to take an extra reference

* tag 'vfs-6.18-rc8.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
  afs: Fix delayed allocation of a cell's anonymous key
  ovl: fail ovl_lock_rename_workdir() if either target is unhashed
  fs/namespace: fix reference leak in grab_requested_mnt_ns

2 weeks agoMerge tag 'soc-fixes-6.18-4' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc
Linus Torvalds [Fri, 28 Nov 2025 17:57:31 +0000 (09:57 -0800)] 
Merge tag 'soc-fixes-6.18-4' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc

Pull SoC fixes from Arnd Bergmann:
 "A few last minute fixes came in this week:

   - interrupt and gpio numbers in foud separate i.MX8 specific
     devicetree files were wrong

   - The vector length property in the C906 CPU description used the
     wrong unit

   - Two bugs with uninitialized stack variables in the tee subsystem

   - Alexander Stein now maintains additional devicetree files"

* tag 'soc-fixes-6.18-4' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc:
  riscv: dts: allwinner: d1: fix vlenb property
  MAINTAINERS: Add entry for TQ-Systems AM335 device trees
  tee: qcomtee: initialize result before use in release worker
  arm64: dts: imx8qm-mek: fix mux-controller select/enable-gpios polarity
  tee: qcomtee: fix uninitialized pointers with free attribute
  ARM: dts: nxp: imx6ul: correct SAI3 interrupt line
  arm64: dts: imx8dxl-ss-conn: swap interrupts number of eqos
  arm64: dts: imx8dxl: Correct pcie-ep interrupt number

2 weeks agoMerge tag 'char-misc-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/gregk...
Linus Torvalds [Fri, 28 Nov 2025 17:44:48 +0000 (09:44 -0800)] 
Merge tag 'char-misc-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc

Pull char / misc / IIO fixes from Greg KH:
 "Here are some much-delayed char/misc/iio driver fixes for 6.18-rc8.

  Fixes in here include:

   - lots of iio driver bugfixes for reported issues.

   - counter driver bugfix

   - slimbus driver bugfix

   - mei tiny bugfix

   - nvmem layout uevent bugfix

  All of these have been in linux-next for a while, but due to travel on
  my side, I haven't had a chance to get them to you"

* tag 'char-misc-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (23 commits)
  nvmem: layouts: fix nvmem_layout_bus_uevent
  iio: accel: bmc150: Fix irq assumption regression
  most: usb: fix double free on late probe failure
  slimbus: ngd: Fix reference count leak in qcom_slim_ngd_notify_slaves
  firmware: stratix10-svc: fix bug in saving controller data
  mei: fix error flow in probe
  iio: st_lsm6dsx: Fixed calibrated timestamp calculation
  iio: humditiy: hdc3020: fix units for thresholds and hysteresis
  iio: humditiy: hdc3020: fix units for temperature and humidity measurement
  iio: imu: st_lsm6dsx: fix array size for st_lsm6dsx_settings fields
  iio: accel: fix ADXL355 startup race condition
  iio: adc: ad7124: fix temperature channel
  iio:common:ssp_sensors: Fix an error handling path ssp_probe()
  iio: adc: ad7280a: fix ad7280_store_balance_timer()
  iio: buffer-dmaengine: enable .get_dma_dev()
  iio: buffer-dma: support getting the DMA channel
  iio: buffer: support getting dma channel from the buffer
  iio: pressure: bmp280: correct meas_time_us calculation
  iio: adc: stm32-dfsdm: fix st,adc-alt-channel property handling
  iio: adc: ad7380: fix SPI offload trigger rate
  ...

2 weeks agoMerge tag 'tty-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
Linus Torvalds [Fri, 28 Nov 2025 17:16:20 +0000 (09:16 -0800)] 
Merge tag 'tty-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty

Pull serial driver fixes from Greg KH:
 "Here are two serial driver fixes for reported issues for 6.18-rc8.

  These are:

   - fix for a much reported symbol build loop that broke the build for
     some kernel configurations

   - amba-pl011 driver bugfix for a reported issue

  Both have been in linux next (the last for weeks, the first for a
  shorter amount of time), with no reported issues"

* tag 'tty-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  serial: 8250: Fix 8250_rsa symbol loop
  serial: amba-pl011: prefer dma_mapping_error() over explicit address checking

2 weeks agoMerge tag 'usb-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Linus Torvalds [Fri, 28 Nov 2025 17:12:40 +0000 (09:12 -0800)] 
Merge tag 'usb-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb

Pull USB/Thunderbolt fixes from Greg KH:
 "Here are some last-minutes USB and Thunderbolt driver fixes and new
  device ids for 6.18-rc8. Included in here are:

   - usb storage quirk fixup

   - xhci driver fixes for reported issues

   - usb gadget driver fixes

   - dwc3 driver fixes

   - UAS driver fixup

   - thunderbolt new device ids

   - usb-serial driver new ids

  All of these have been in linux-next with no reported issues, many for
  many weeks"

* tag 'usb-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (21 commits)
  usb: gadget: renesas_usbf: Handle devm_pm_runtime_enable() errors
  USB: storage: Remove subclass and protocol overrides from Novatek quirk
  usb: uas: fix urb unmapping issue when the uas device is remove during ongoing data transfer
  usb: dwc3: Fix race condition between concurrent dwc3_remove_requests() call paths
  xhci: dbgtty: fix device unregister
  usb: storage: sddr55: Reject out-of-bound new_pba
  USB: serial: option: add support for Rolling RW101R-GL
  usb: typec: ucsi: psy: Set max current to zero when disconnected
  usb: gadget: f_eem: Fix memory leak in eem_unwrap
  usb: dwc3: pci: Sort out the Intel device IDs
  usb: dwc3: pci: add support for the Intel Nova Lake -S
  drivers/usb/dwc3: fix PCI parent check
  usb: storage: Fix memory leak in USB bulk transport
  xhci: sideband: Fix race condition in sideband unregister
  xhci: dbgtty: Fix data corruption when transmitting data form DbC to host
  xhci: fix stale flag preventig URBs after link state error is cleared
  USB: serial: ftdi_sio: add support for u-blox EVK-M101
  usb: cdns3: Fix double resource release in cdns3_pci_probe
  usb: gadget: udc: fix use-after-free in usb_gadget_state_work
  usb: renesas_usbhs: Fix synchronous external abort on unbind
  ...

2 weeks agoMerge tag 'mailbox-fixes-v6.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 28 Nov 2025 17:09:33 +0000 (09:09 -0800)] 
Merge tag 'mailbox-fixes-v6.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/jassibrar/mailbox

Pull mailbox fixes from Jassi Brar:

 - omap: check for pending msgs only when mbox is exclusive

 - mailbox-test: debugfs_create_dir error checking

 - mtk:
     - cmdq: fix DMA address handling
     - gpueb: Add missing 'static' to mailbox ops struct

 - pcc: don't zero error register

 - th1520: fix clock imbalance on probe failure

* tag 'mailbox-fixes-v6.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/jassibrar/mailbox:
  mailbox: th1520: fix clock imbalance on probe failure
  mailbox: pcc: don't zero error register
  mailbox: mtk-gpueb: Add missing 'static' to mailbox ops struct
  mailbox: mtk-cmdq: Refine DMA address handling for the command buffer
  mailbox: mailbox-test: Fix debugfs_create_dir error checking
  mailbox: omap-mailbox: Check for pending msgs only when mbox is exclusive

2 weeks agoMerge tag 'sunxi-fixes-for-6.18' of https://git.kernel.org/pub/scm/linux/kernel/git...
Arnd Bergmann [Fri, 28 Nov 2025 16:37:13 +0000 (17:37 +0100)] 
Merge tag 'sunxi-fixes-for-6.18' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux into arm/fixes

Allwinner fixes for 6.18

Just one fix to correct the "thead,vlenb" property for the RISC-V based
D1 SoC family.

* tag 'sunxi-fixes-for-6.18' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:
  riscv: dts: allwinner: d1: fix vlenb property

2 weeks agoMerge tag 'omap-for-v6.19/maintainers-signed' of git://git.kernel.org/pub/scm/linux...
Arnd Bergmann [Fri, 21 Nov 2025 21:14:04 +0000 (22:14 +0100)] 
Merge tag 'omap-for-v6.19/maintainers-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/khilman/linux-omap into arm/fixes

MAINTAINERS: Add entry for TQ-Systems AM335 device trees

* tag 'omap-for-v6.19/maintainers-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/khilman/linux-omap:
  MAINTAINERS: Add entry for TQ-Systems AM335 device trees

2 weeks agoMerge tag 'mmc-v6.18-rc2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc
Linus Torvalds [Fri, 28 Nov 2025 16:20:14 +0000 (08:20 -0800)] 
Merge tag 'mmc-v6.18-rc2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc

Pull MMC fix from Ulf Hansson:

 - sdhci-of-dwcmshc: Fix reset handling for some variants

* tag 'mmc-v6.18-rc2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc:
  mmc: sdhci-of-dwcmshc: Promote the th1520 reset handling to ip level

2 weeks agoMerge tag 'pmdomain-v6.18-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh...
Linus Torvalds [Fri, 28 Nov 2025 16:08:02 +0000 (08:08 -0800)] 
Merge tag 'pmdomain-v6.18-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/linux-pm

Pull pmdomain fixes from Ulf Hansson:

 - mediatek: Fix spinlock recursion in probe

 - tegra: Use GENPD_FLAG_NO_STAY_ON to restore old behaviour

* tag 'pmdomain-v6.18-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/linux-pm:
  pmdomain: tegra: Add GENPD_FLAG_NO_STAY_ON flag
  pmdomains: mtk-pm-domains: Fix spinlock recursion in probe

2 weeks agomailbox: th1520: fix clock imbalance on probe failure
Johan Hovold [Fri, 17 Oct 2025 05:54:14 +0000 (07:54 +0200)] 
mailbox: th1520: fix clock imbalance on probe failure

The purpose of the devm_add_action_or_reset() helper is to call the
action function in case adding an action ever fails so drop the clock
disable from the error path to avoid disabling the clocks twice.

Fixes: 5d4d263e1c6b ("mailbox: Introduce support for T-head TH1520 Mailbox driver")
Cc: Michal Wilczynski <m.wilczynski@samsung.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Reviewed-by: Michal Wilczynski <m.wilczynski@samsung.com>
Signed-off-by: Jassi Brar <jassisinghbrar@gmail.com>
2 weeks agomailbox: pcc: don't zero error register
Jamie Iles [Wed, 5 Nov 2025 14:42:29 +0000 (14:42 +0000)] 
mailbox: pcc: don't zero error register

The error status mask for a type 3/4 subspace is used for reading the
error status, and the bitwise inverse is used for clearing the error
with the intent being to preserve any of the non-error bits.  However,
we were previously applying the mask to extract the status and then
applying the inverse to the result which ended up clearing all bits.

Instead, store the inverse mask in the preserve mask and then use that
on the original value read from the error status so that only the error
is cleared.

Fixes: c45ded7e1135 ("mailbox: pcc: Add support for PCCT extended PCC subspaces(type 3/4)")
Signed-off-by: Jamie Iles <jamie.iles@oss.qualcomm.com>
Signed-off-by: Punit Agrawal <punit.agrawal@oss.qualcomm.com>
Signed-off-by: Jassi Brar <jassisinghbrar@gmail.com>
2 weeks agomailbox: mtk-gpueb: Add missing 'static' to mailbox ops struct
Nicolas Frattaroli [Mon, 13 Oct 2025 08:09:01 +0000 (10:09 +0200)] 
mailbox: mtk-gpueb: Add missing 'static' to mailbox ops struct

mtk_gpueb_mbox_ops should be declared static. However, due to its const
nature, this specifier was missed, as it compiled fine without it and
with no warning by the compiler.

arc-linux-gcc (GCC) 12.5.0 doesn't seem to like it however, so add the
static to fix that.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202510100629.3nGvrhEU-lkp@intel.com/
Fixes: dbca0eabb821 ("mailbox: add MediaTek GPUEB IPI mailbox")
Signed-off-by: Nicolas Frattaroli <nicolas.frattaroli@collabora.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Jassi Brar <jassisinghbrar@gmail.com>
2 weeks agomailbox: mtk-cmdq: Refine DMA address handling for the command buffer
Jason-JH Lin [Wed, 22 Oct 2025 17:16:30 +0000 (01:16 +0800)] 
mailbox: mtk-cmdq: Refine DMA address handling for the command buffer

GCE can only fetch the command buffer address from a 32-bit register.
Some SoCs support a 35-bit command buffer address for GCE, which
requires a right shift of 3 bits before setting the address into
the 32-bit register. A comment has been added to the header of
cmdq_get_shift_pa() to explain this requirement.

To prevent the GCE command buffer address from being DMA mapped beyond
its supported bit range, the DMA bit mask for the device is set during
initialization.

Additionally, to ensure the correct shift is applied when setting or
reading the register that stores the GCE command buffer address,
new APIs, cmdq_convert_gce_addr() and cmdq_revert_gce_addr(), have
been introduced for consistent operations on this register.

The variable type for the command buffer address has been standardized
to dma_addr_t to prevent handling issues caused by type mismatches.

Fixes: 0858fde496f8 ("mailbox: cmdq: variablize address shift in platform")
Signed-off-by: Jason-JH Lin <jason-jh.lin@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Jassi Brar <jassisinghbrar@gmail.com>
2 weeks agomailbox: mailbox-test: Fix debugfs_create_dir error checking
Haotian Zhang [Thu, 20 Nov 2025 02:40:39 +0000 (10:40 +0800)] 
mailbox: mailbox-test: Fix debugfs_create_dir error checking

The debugfs_create_dir() function returns ERR_PTR() on error, not NULL.
The current null-check fails to catch errors.

Use IS_ERR() to correctly check for errors.

Fixes: 8ea4484d0c2b ("mailbox: Add generic mechanism for testing Mailbox Controllers")
Signed-off-by: Haotian Zhang <vulab@iscas.ac.cn>
Signed-off-by: Jassi Brar <jassisinghbrar@gmail.com>
2 weeks agomailbox: omap-mailbox: Check for pending msgs only when mbox is exclusive
Beleswar Padhi [Mon, 3 Nov 2025 20:11:11 +0000 (01:41 +0530)] 
mailbox: omap-mailbox: Check for pending msgs only when mbox is exclusive

On TI K3 devices, the mailbox resides in the Always-On power domain
(LPSC_main_alwayson) and is shared among multiple processors. The
mailbox is not solely exclusive to Linux.

Currently, the suspend path checks all FIFO queues for pending messages
and blocks suspend if any are present. This behavior is unnecessary for
K3 devices, since some of the FIFOs are used for RTOS<->RTOS
communication and are independent of Linux.

For FIFOs used in Linux<->RTOS communication, any pending message would
trigger an interrupt, which naturally prevents suspend from completing.
Hence, there is no need for the mailbox driver to explicitly check for
pending messages on K3 platforms.

Introduce a device match flag to indicate whether the mailbox instance
is exclusive to Linux, and skip the pending message check for
non-exclusive instances (such as in K3).

Fixes: a49f991e740f ("arm64: dts: ti: k3-am62-verdin: Add missing cfg for TI IPC Firmware")
Closes: https://lore.kernel.org/all/sid7gtg5vay5qgicsl6smnzwg5mnneoa35cempt5ddwjvedaio@hzsgcx6oo74l/
Signed-off-by: Beleswar Padhi <b-padhi@ti.com>
Tested-by: Hiago De Franco <hiago.franco@toradex.com>
Reviewed-by: Andrew Davis <afd@ti.com>
Signed-off-by: Jassi Brar <jassisinghbrar@gmail.com>
2 weeks agoafs: Fix delayed allocation of a cell's anonymous key
David Howells [Fri, 28 Nov 2025 10:19:05 +0000 (10:19 +0000)] 
afs: Fix delayed allocation of a cell's anonymous key

The allocation of a cell's anonymous key is done in a background thread
along with other cell setup such as doing a DNS upcall.  In the reported
bug, this is triggered by afs_parse_source() parsing the device name given
to mount() and calling afs_lookup_cell() with the name of the cell.

The normal key lookup then tries to use the key description on the
anonymous authentication key as the reference for request_key() - but it
may not yet be set and so an oops can happen.

This has been made more likely to happen by the fix for dynamic lookup
failure.

Fix this by firstly allocating a reference name and attaching it to the
afs_cell record when the record is created.  It can share the memory
allocation with the cell name (unfortunately it can't just overlap the cell
name by prepending it with "afs@" as the cell name already has a '.'
prepended for other purposes).  This reference name is then passed to
request_key().

Secondly, the anon key is now allocated on demand at the point a key is
requested in afs_request_key() if it is not already allocated.  A mutex is
used to prevent multiple allocation for a cell.

Thirdly, make afs_request_key_rcu() return NULL if the anonymous key isn't
yet allocated (if we need it) and then the caller can return -ECHILD to
drop out of RCU-mode and afs_request_key() can be called.

Note that the anonymous key is kind of necessary to make the key lookup
cache work as that doesn't currently cache a negative lookup, but it's
probably worth some investigation to see if NULL can be used instead.

Fixes: 330e2c514823 ("afs: Fix dynamic lookup to fail on cell lookup failure")
Reported-by: syzbot+41c68824eefb67cdf00c@syzkaller.appspotmail.com
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://patch.msgid.link/800328.1764325145@warthog.procyon.org.uk
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
2 weeks agoovl: fail ovl_lock_rename_workdir() if either target is unhashed
NeilBrown [Fri, 28 Nov 2025 01:22:35 +0000 (12:22 +1100)] 
ovl: fail ovl_lock_rename_workdir() if either target is unhashed

As well as checking that the parent hasn't changed after getting the
lock we need to check that the dentry hasn't been unhashed.
Otherwise we might try to rename something that has been removed.

Reported-by: syzbot+bfc9a0ccf0de47d04e8c@syzkaller.appspotmail.com
Fixes: d2c995581c7c ("ovl: Call ovl_create_temp() without lock held.")
Signed-off-by: NeilBrown <neil@brown.name>
Link: https://patch.msgid.link/176429295510.634289.1552337113663461690@noble.neil.brown.name
Tested-by: syzbot+bfc9a0ccf0de47d04e8c@syzkaller.appspotmail.com
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
2 weeks agoMerge tag 'drm-fixes-2025-11-28' of https://gitlab.freedesktop.org/drm/kernel
Linus Torvalds [Fri, 28 Nov 2025 05:06:27 +0000 (21:06 -0800)] 
Merge tag 'drm-fixes-2025-11-28' of https://gitlab.freedesktop.org/drm/kernel

Pull drm fixes from Dave Airlie:
 "Last one for this round hopefully, mostly the usual suspects,
  xe/amdgpu, with some single fixes otherwise.

  There is one amdgpu HDMI blackscreen bug that came in late in the
  cycle, but it was bisected and the revert is in here.

  i915:
   - Reject async flips when PSR's selective fetch is enabled

  xe:
   - Fix resource leak in xe_guc_ct_init_noalloc()'s error path
   - Fix stack_depot usage without STACKDEPOT_ALWAYS_INIT
   - Fix overflow in conversion from clock tics to msec

  amdgpu:
   - Unified MES fix
   - HDMI fix
   - Cursor fix
   - Bightness fix
   - EDID reading improvement
   - UserQ fix
   - Cyan Skillfish IP discovery fix

  bridge:
   - sil902x: Fix HDMI detection

  imagination:
   - Update documentation

  sti:
   - Fix leaks in probe

  vga_switcheroo:
   - Avoid race condition during fbcon initialization"

* tag 'drm-fixes-2025-11-28' of https://gitlab.freedesktop.org/drm/kernel:
  drm/amdgpu: fix cyan_skillfish2 gpu info fw handling
  drm/amdgpu: attach tlb fence to the PTs update
  drm/amd/display: Increase EDID read retries
  drm/amd/display: Don't change brightness for disabled connectors
  drm/amd/display: Check NULL before accessing
  Revert "drm/amd/display: Move setup_stream_attribute"
  drm/xe: Fix conversion from clock ticks to milliseconds
  drm/xe/guc: Fix stack_depot usage
  drm/xe/guc: Fix resource leak in xe_guc_ct_init_noalloc()
  drm/i915/psr: Reject async flips when selective fetch is enabled
  drm, fbcon, vga_switcheroo: Avoid race condition in fbcon setup
  drm/amd/amdgpu: reserve vm invalidation engine for uni_mes
  drm: sti: fix device leaks at component probe
  drm/imagination: Document pvr_device.power member
  drm/bridge: sii902x: Fix HDMI detection with DRM_BRIDGE_ATTACH_NO_CONNECTOR

3 weeks agoMerge tag 'drm-xe-fixes-2025-11-27' of https://gitlab.freedesktop.org/drm/xe/kernel...
Dave Airlie [Fri, 28 Nov 2025 02:10:19 +0000 (12:10 +1000)] 
Merge tag 'drm-xe-fixes-2025-11-27' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-fixes

Driver Changes:
 - Fix resource leak in xe_guc_ct_init_noalloc()'s error path (Shuicheng Lin)
 - Fix stack_depot usage without STACKDEPOT_ALWAYS_INIT (Lucas)
 - Fix overflow in conversion from clock tics to msec (Harish Chegondi)

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patch.msgid.link/7ejiqjgthpqybg5svmkind2pszk4fqadxuq7rngchaaw76iept@5pn6sngqj6lk
3 weeks agoMerge tag 'drm-misc-fixes-2025-11-27' of https://gitlab.freedesktop.org/drm/misc...
Dave Airlie [Fri, 28 Nov 2025 02:00:07 +0000 (12:00 +1000)] 
Merge tag 'drm-misc-fixes-2025-11-27' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-fixes

Short summary of fixes pull:

bridge:
- sil902x: Fix HDMI detection

imagination:
- Update documentation

sti:
- Fix leaks in probe

vga_switcheroo:
- Avoid race condition during fbcon initialization

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patch.msgid.link/20251127081007.GA13578@2a02-2454-fd5e-fd00-689d-32c0-780c-bb87.dyn6.pyur.net
3 weeks agoMerge tag 'amd-drm-fixes-6.18-2025-11-26' of https://gitlab.freedesktop.org/agd5f...
Dave Airlie [Fri, 28 Nov 2025 01:56:37 +0000 (11:56 +1000)] 
Merge tag 'amd-drm-fixes-6.18-2025-11-26' of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes

amd-drm-fixes-6.18-2025-11-26:

amdgpu:
- Unified MES fix
- HDMI fix
- Cursor fix
- Bightness fix
- EDID reading improvement
- UserQ fix
- Cyan Skillfish IP discovery fix

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patch.msgid.link/20251126204925.3316684-1-alexander.deucher@amd.com
3 weeks agoMerge tag 'dma-mapping-6.18-2025-11-27' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 28 Nov 2025 01:29:15 +0000 (17:29 -0800)] 
Merge tag 'dma-mapping-6.18-2025-11-27' of git://git.kernel.org/pub/scm/linux/kernel/git/mszyprowski/linux

Pull dma-mapping fixes from Marek Szyprowski:
 "Two last minute fixes for the recently modified DMA API infrastructure:

   - proper handling of DMA_ATTR_MMIO in dma_iova_unlink() function (me)

   - regression fix for the code refactoring related to P2PDMA (Pranjal
     Shrivastava)"

* tag 'dma-mapping-6.18-2025-11-27' of git://git.kernel.org/pub/scm/linux/kernel/git/mszyprowski/linux:
  dma-direct: Fix missing sg_dma_len assignment in P2PDMA bus mappings
  iommu/dma: add missing support for DMA_ATTR_MMIO for dma_iova_unlink()

3 weeks agoMerge tag 'acpi-6.18-rc8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael...
Linus Torvalds [Fri, 28 Nov 2025 01:25:46 +0000 (17:25 -0800)] 
Merge tag 'acpi-6.18-rc8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm

Pull ACPI fix from Rafael Wysocki:
 "One more urgent ACPI support fix for 6.18

  There is one more commit that needs to be reverted after reverting
  problematic commit 7a8c994cbb2d ("ACPI: processor: idle: Optimize ACPI
  idle driver registration"), so revert it"

* tag 'acpi-6.18-rc8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  Revert "ACPI: processor: Update cpuidle driver check in __acpi_processor_start()"

3 weeks agoMerge tag 'drm-intel-fixes-2025-11-26' of https://gitlab.freedesktop.org/drm/i915...
Dave Airlie [Fri, 28 Nov 2025 01:08:54 +0000 (11:08 +1000)] 
Merge tag 'drm-intel-fixes-2025-11-26' of https://gitlab.freedesktop.org/drm/i915/kernel into drm-fixes

- Reject async flips when PSR's selective fetch is enabled (Ville)

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patch.msgid.link/aScgY8QMjmyJRBX2@intel.com
3 weeks agoRevert "ACPI: processor: Update cpuidle driver check in __acpi_processor_start()"
Rafael J. Wysocki [Thu, 27 Nov 2025 19:52:55 +0000 (20:52 +0100)] 
Revert "ACPI: processor: Update cpuidle driver check in __acpi_processor_start()"

Revert commit 8a1b5d412cb4 ("ACPI: processor: Update cpuidle driver
check in __acpi_processor_start()") which depends on commit
7a8c994cbb2d ("ACPI: processor: idle: Optimize ACPI idle driver
registration") that got reverted.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
3 weeks agoMerge tag 'ceph-for-6.18-rc8' of https://github.com/ceph/ceph-client
Linus Torvalds [Thu, 27 Nov 2025 19:11:03 +0000 (11:11 -0800)] 
Merge tag 'ceph-for-6.18-rc8' of https://github.com/ceph/ceph-client

Pull ceph fixes from Ilya Dryomov:
 "A patch to make sparse read handling work in msgr2 secure mode from
  Slava and a couple of fixes from Ziming and myself to avoid operating
  on potentially invalid memory, all marked for stable"

* tag 'ceph-for-6.18-rc8' of https://github.com/ceph/ceph-client:
  libceph: prevent potential out-of-bounds writes in handle_auth_session_key()
  libceph: replace BUG_ON with bounds check for map->max_osd
  ceph: fix crash in process_v2_sparse_read() for encrypted directories
  libceph: drop started parameter of __ceph_open_session()
  libceph: fix potential use-after-free in have_mon_and_osd_map()

3 weeks agoMerge tag 'net-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Linus Torvalds [Thu, 27 Nov 2025 17:18:40 +0000 (09:18 -0800)] 
Merge tag 'net-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Pull networking fixes from Paolo Abeni:
 "Including fixes from bluetooth and CAN. No known outstanding
  regressions.

  Current release - regressions:

   - mptcp: initialize rcv_mss before calling tcp_send_active_reset()

   - eth: mlx5e: fix validation logic in rate limiting

  Previous releases - regressions:

   - xsk: avoid data corruption on cq descriptor number

   - bluetooth:
       - prevent race in socket write iter and sock bind
       - fix not generating mackey and ltk when repairing

   - can:
       - kvaser_usb: fix potential infinite loop in command parsers
       - rcar_canfd: fix CAN-FD mode as default

   - eth:
       - veth: reduce XDP no_direct return section to fix race
       - virtio-net: avoid unnecessary checksum calculation on guest RX

  Previous releases - always broken:

   - sched: fix TCF_LAYER_TRANSPORT handling in tcf_get_base_ptr()

   - bluetooth: mediatek: fix kernel crash when releasing iso interface

   - vhost: rewind next_avail_head while discarding descriptors

   - eth:
       - r8169: fix RTL8127 hang on suspend/shutdown
       - aquantia: add missing descriptor cache invalidation on ATL2

   - dsa: microchip: fix resource releases in error path"

* tag 'net-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (47 commits)
  mptcp: Initialise rcv_mss before calling tcp_send_active_reset() in mptcp_do_fastclose().
  net: fec: do not register PPS event for PEROUT
  net: fec: do not allow enabling PPS and PEROUT simultaneously
  net: fec: do not update PEROUT if it is enabled
  net: fec: cancel perout_timer when PEROUT is disabled
  net: mctp: unconditionally set skb->dev on dst output
  net: atlantic: fix fragment overflow handling in RX path
  MAINTAINERS: separate VIRTIO NET DRIVER and add netdev
  virtio-net: avoid unnecessary checksum calculation on guest RX
  eth: fbnic: Fix counter roll-over issue
  mptcp: clear scheduled subflows on retransmit
  net: dsa: sja1105: fix SGMII linking at 10M or 100M but not passing traffic
  s390/net: list Aswin Karuvally as maintainer
  net: wwan: mhi: Keep modem name match with Foxconn T99W640
  vhost: rewind next_avail_head while discarding descriptors
  net/sched: em_canid: fix uninit-value in em_canid_match
  can: rcar_canfd: Fix CAN-FD mode as default
  xsk: avoid data corruption on cq descriptor number
  r8169: fix RTL8127 hang on suspend/shutdown
  net: sxgbe: fix potential NULL dereference in sxgbe_rx()
  ...

3 weeks agoMerge tag 'platform-drivers-x86-v6.18-5' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Thu, 27 Nov 2025 16:51:36 +0000 (08:51 -0800)] 
Merge tag 'platform-drivers-x86-v6.18-5' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86

Pull platform driver fixes from Ilpo Järvinen:

 - arm64/thinkpad-t14s-ec:
    - Fix IRQ race condition
    - Sleep after EC access

 - intel/punit_ipc: Fix memory corruption

* tag 'platform-drivers-x86-v6.18-5' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86:
  platform/x86: intel: punit_ipc: fix memory corruption
  platform: arm64: thinkpad-t14s-ec: sleep after EC access
  platform: arm64: thinkpad-t14s-ec: fix IRQ race condition

3 weeks agomptcp: Initialise rcv_mss before calling tcp_send_active_reset() in mptcp_do_fastclose().
Kuniyuki Iwashima [Tue, 25 Nov 2025 19:53:29 +0000 (19:53 +0000)] 
mptcp: Initialise rcv_mss before calling tcp_send_active_reset() in mptcp_do_fastclose().

syzbot reported divide-by-zero in __tcp_select_window() by
MPTCP socket. [0]

We had a similar issue for the bare TCP and fixed in commit
499350a5a6e7 ("tcp: initialize rcv_mss to TCP_MIN_MSS instead
of 0").

Let's apply the same fix to mptcp_do_fastclose().

[0]:
Oops: divide error: 0000 [#1] SMP KASAN PTI
CPU: 0 UID: 0 PID: 6068 Comm: syz.0.17 Not tainted syzkaller #0 PREEMPT(full)
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/25/2025
RIP: 0010:__tcp_select_window+0x824/0x1320 net/ipv4/tcp_output.c:3336
Code: ff ff ff 44 89 f1 d3 e0 89 c1 f7 d1 41 01 cc 41 21 c4 e9 a9 00 00 00 e8 ca 49 01 f8 e9 9c 00 00 00 e8 c0 49 01 f8 44 89 e0 99 <f7> 7c 24 1c 41 29 d4 48 bb 00 00 00 00 00 fc ff df e9 80 00 00 00
RSP: 0018:ffffc90003017640 EFLAGS: 00010293
RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff88807b469e40
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
RBP: ffffc90003017730 R08: ffff888033268143 R09: 1ffff1100664d028
R10: dffffc0000000000 R11: ffffed100664d029 R12: 0000000000000000
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
FS:  000055557faa0500(0000) GS:ffff888126135000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f64a1912ff8 CR3: 0000000072122000 CR4: 00000000003526f0
Call Trace:
 <TASK>
 tcp_select_window net/ipv4/tcp_output.c:281 [inline]
 __tcp_transmit_skb+0xbc7/0x3aa0 net/ipv4/tcp_output.c:1568
 tcp_transmit_skb net/ipv4/tcp_output.c:1649 [inline]
 tcp_send_active_reset+0x2d1/0x5b0 net/ipv4/tcp_output.c:3836
 mptcp_do_fastclose+0x27e/0x380 net/mptcp/protocol.c:2793
 mptcp_disconnect+0x238/0x710 net/mptcp/protocol.c:3253
 mptcp_sendmsg_fastopen+0x2f8/0x580 net/mptcp/protocol.c:1776
 mptcp_sendmsg+0x1774/0x1980 net/mptcp/protocol.c:1855
 sock_sendmsg_nosec net/socket.c:727 [inline]
 __sock_sendmsg+0xe5/0x270 net/socket.c:742
 __sys_sendto+0x3bd/0x520 net/socket.c:2244
 __do_sys_sendto net/socket.c:2251 [inline]
 __se_sys_sendto net/socket.c:2247 [inline]
 __x64_sys_sendto+0xde/0x100 net/socket.c:2247
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0xfa/0xfa0 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f66e998f749
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffff9acedb8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c
RAX: ffffffffffffffda RBX: 00007f66e9be5fa0 RCX: 00007f66e998f749
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000003
RBP: 00007ffff9acee10 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000001
R13: 00007f66e9be5fa0 R14: 00007f66e9be5fa0 R15: 0000000000000006
 </TASK>

Fixes: ae155060247b ("mptcp: fix duplicate reset on fastclose")
Reported-by: syzbot+3a92d359bc2ec6255a33@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/netdev/69260882.a70a0220.d98e3.00b4.GAE@google.com/
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20251125195331.309558-1-kuniyu@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 weeks agospi: nxp-fspi: Propagate fwnode in ACPI case as well
Andy Shevchenko [Wed, 26 Nov 2025 20:25:01 +0000 (21:25 +0100)] 
spi: nxp-fspi: Propagate fwnode in ACPI case as well

Propagate fwnode of the ACPI device to the SPI controller Linux device.
Currently only OF case propagates fwnode to the controller.

While at it, replace several calls to dev_fwnode() with a single one
cached in a local variable, and unify checks for fwnode type by using
is_*_node() APIs.

Fixes: 55ab8487e01d ("spi: spi-nxp-fspi: Add ACPI support")
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Haibo Chen <haibo.chen@nxp.com>
Link: https://patch.msgid.link/20251126202501.2319679-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 weeks agoregulator: rtq2208: Correct LDO2 logic judgment bits
ChiYuan Huang [Thu, 27 Nov 2025 02:25:51 +0000 (10:25 +0800)] 
regulator: rtq2208: Correct LDO2 logic judgment bits

The LDO2 judgement bit position should be 7, not 6.

Cc: stable@vger.kernel.org
Reported-by: Yoon Dong Min <dm.youn@telechips.com>
Fixes: b65439d90150 ("regulator: rtq2208: Fix the LDO DVS capability")
Signed-off-by: ChiYuan Huang <cy_huang@richtek.com>
Link: https://patch.msgid.link/faadb009f84b88bfcabe39fc5009c7357b00bbe2.1764209258.git.cy_huang@richtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 weeks agoregulator: rtq2208: Correct buck group2 phase mapping logic
ChiYuan Huang [Thu, 27 Nov 2025 02:25:50 +0000 (10:25 +0800)] 
regulator: rtq2208: Correct buck group2 phase mapping logic

Correct buck group2 H and F mapping logic.

Cc: stable@vger.kernel.org
Reported-by: Yoon Dong Min <dm.youn@telechips.com>
Fixes: 1742e7e978ba ("regulator: rtq2208: Fix incorrect buck converter phase mapping")
Signed-off-by: ChiYuan Huang <cy_huang@richtek.com>
Link: https://patch.msgid.link/8527ae02a72b754d89b7580a5fe7474d6f80f5c3.1764209258.git.cy_huang@richtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>
3 weeks agoMerge branch 'net-fec-fix-some-ptp-related-issues'
Paolo Abeni [Thu, 27 Nov 2025 10:57:47 +0000 (11:57 +0100)] 
Merge branch 'net-fec-fix-some-ptp-related-issues'

Wei Fang says:

====================
net: fec: fix some PTP related issues

There are some issues which were introduced by the commit 350749b909bf
("net: fec: Add support for periodic output signal of PPS"). See each
patch for more details.
====================

Link: https://patch.msgid.link/20251125085210.1094306-1-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 weeks agonet: fec: do not register PPS event for PEROUT
Wei Fang [Tue, 25 Nov 2025 08:52:10 +0000 (16:52 +0800)] 
net: fec: do not register PPS event for PEROUT

There are currently two situations that can trigger the PTP interrupt,
one is the PPS event, the other is the PEROUT event. However, the irq
handler fec_pps_interrupt() does not check the irq event type and
directly registers a PPS event into the system, but the event may be
a PEROUT event. This is incorrect because PEROUT is an output signal,
while PPS is the input of the kernel PPS system. Therefore, add a check
for the event type, if pps_enable is true, it means that the current
event is a PPS event, and then the PPS event is registered.

Fixes: 350749b909bf ("net: fec: Add support for periodic output signal of PPS")
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20251125085210.1094306-5-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 weeks agonet: fec: do not allow enabling PPS and PEROUT simultaneously
Wei Fang [Tue, 25 Nov 2025 08:52:09 +0000 (16:52 +0800)] 
net: fec: do not allow enabling PPS and PEROUT simultaneously

In the current driver, PPS and PEROUT use the same channel to generate
the events, so they cannot be enabled at the same time. Otherwise, the
later configuration will overwrite the earlier configuration. Therefore,
when configuring PPS, the driver will check whether PEROUT is enabled.
Similarly, when configuring PEROUT, the driver will check whether PPS
is enabled.

Fixes: 350749b909bf ("net: fec: Add support for periodic output signal of PPS")
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20251125085210.1094306-4-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 weeks agonet: fec: do not update PEROUT if it is enabled
Wei Fang [Tue, 25 Nov 2025 08:52:08 +0000 (16:52 +0800)] 
net: fec: do not update PEROUT if it is enabled

If the previously set PEROUT is already active, updating it will cause
the new PEROUT to start immediately instead of at the specified time.
This is because fep->reload_period is updated whithout check whether
the PEROUT is enabled, and the old PEROUT is not disabled. Therefore,
the pulse period will be updated immediately in the pulse interrupt
handler fec_pps_interrupt().

Currently, the driver does not support directly updating PEROUT and it
will make the logic be more complicated. To fix the current issue, add
a check before enabling the PEROUT, the driver will return an error if
PEROUT is enabled. If users wants to update a new PEROUT, they should
disable the old PEROUT first.

Fixes: 350749b909bf ("net: fec: Add support for periodic output signal of PPS")
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20251125085210.1094306-3-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 weeks agonet: fec: cancel perout_timer when PEROUT is disabled
Wei Fang [Tue, 25 Nov 2025 08:52:07 +0000 (16:52 +0800)] 
net: fec: cancel perout_timer when PEROUT is disabled

The PEROUT allows the user to set a specified future time to output the
periodic signal. If the future time is far from the current time, the FEC
driver will use hrtimer to configure PEROUT one second before the future
time. However, the hrtimer will not be canceled if the PEROUT is disabled
before the hrtimer expires. So the PEROUT will be configured when the
hrtimer expires, which is not as expected. Therefore, cancel the hrtimer
in fec_ptp_pps_disable() to fix this issue.

Fixes: 350749b909bf ("net: fec: Add support for periodic output signal of PPS")
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20251125085210.1094306-2-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 weeks agonet: mctp: unconditionally set skb->dev on dst output
Jeremy Kerr [Tue, 25 Nov 2025 06:48:54 +0000 (14:48 +0800)] 
net: mctp: unconditionally set skb->dev on dst output

On transmit, we are currently relying on skb->dev being set by
mctp_local_output() when we first set up the skb destination fields.
However, forwarded skbs do not use the local_output path, so will retain
their incoming netdev as their ->dev on tx. This does not work when
we're forwarding between interfaces.

Set skb->dev unconditionally in the transmit path, to allow for proper
forwarding.

We keep the skb->dev initialisation in mctp_local_output(), as we use it
for fragmentation.

Fixes: 269936db5eb3 ("net: mctp: separate routing database from routing operations")
Suggested-by: Vince Chang <vince_chang@aspeedtech.com>
Signed-off-by: Jeremy Kerr <jk@codeconstruct.com.au>
Link: https://patch.msgid.link/20251125-dev-forward-v1-1-54ecffcd0616@codeconstruct.com.au
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
3 weeks agolibceph: prevent potential out-of-bounds writes in handle_auth_session_key()
ziming zhang [Fri, 14 Nov 2025 08:56:10 +0000 (16:56 +0800)] 
libceph: prevent potential out-of-bounds writes in handle_auth_session_key()

The len field originates from untrusted network packets. Boundary
checks have been added to prevent potential out-of-bounds writes when
decrypting the connection secret or processing service tickets.

[ idryomov: changelog ]

Cc: stable@vger.kernel.org
Signed-off-by: ziming zhang <ezrakiez@gmail.com>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
3 weeks agolibceph: replace BUG_ON with bounds check for map->max_osd
ziming zhang [Mon, 17 Nov 2025 10:07:41 +0000 (18:07 +0800)] 
libceph: replace BUG_ON with bounds check for map->max_osd

OSD indexes come from untrusted network packets. Boundary checks are
added to validate these against map->max_osd.

[ idryomov: drop BUG_ON in ceph_get_primary_affinity(), minor cosmetic
  edits ]

Cc: stable@vger.kernel.org
Signed-off-by: ziming zhang <ezrakiez@gmail.com>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
3 weeks agoceph: fix crash in process_v2_sparse_read() for encrypted directories
Viacheslav Dubeyko [Thu, 13 Nov 2025 22:36:24 +0000 (14:36 -0800)] 
ceph: fix crash in process_v2_sparse_read() for encrypted directories

The crash in process_v2_sparse_read() for fscrypt-encrypted directories
has been reported. Issue takes place for Ceph msgr2 protocol in secure
mode. It can be reproduced by the steps:

sudo mount -t ceph :/ /mnt/cephfs/ -o name=admin,fs=cephfs,ms_mode=secure

(1) mkdir /mnt/cephfs/fscrypt-test-3
(2) cp area_decrypted.tar /mnt/cephfs/fscrypt-test-3
(3) fscrypt encrypt --source=raw_key --key=./my.key /mnt/cephfs/fscrypt-test-3
(4) fscrypt lock /mnt/cephfs/fscrypt-test-3
(5) fscrypt unlock --key=my.key /mnt/cephfs/fscrypt-test-3
(6) cat /mnt/cephfs/fscrypt-test-3/area_decrypted.tar
(7) Issue has been triggered

[  408.072247] ------------[ cut here ]------------
[  408.072251] WARNING: CPU: 1 PID: 392 at net/ceph/messenger_v2.c:865
ceph_con_v2_try_read+0x4b39/0x72f0
[  408.072267] Modules linked in: intel_rapl_msr intel_rapl_common
intel_uncore_frequency_common intel_pmc_core pmt_telemetry pmt_discovery
pmt_class intel_pmc_ssram_telemetry intel_vsec kvm_intel joydev kvm irqbypass
polyval_clmulni ghash_clmulni_intel aesni_intel rapl input_leds psmouse
serio_raw i2c_piix4 vga16fb bochs vgastate i2c_smbus floppy mac_hid qemu_fw_cfg
pata_acpi sch_fq_codel rbd msr parport_pc ppdev lp parport efi_pstore
[  408.072304] CPU: 1 UID: 0 PID: 392 Comm: kworker/1:3 Not tainted 6.17.0-rc7+
[  408.072307] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
1.17.0-5.fc42 04/01/2014
[  408.072310] Workqueue: ceph-msgr ceph_con_workfn
[  408.072314] RIP: 0010:ceph_con_v2_try_read+0x4b39/0x72f0
[  408.072317] Code: c7 c1 20 f0 d4 ae 50 31 d2 48 c7 c6 60 27 d5 ae 48 c7 c7 f8
8e 6f b0 68 60 38 d5 ae e8 00 47 61 fe 48 83 c4 18 e9 ac fc ff ff <0f> 0b e9 06
fe ff ff 4c 8b 9d 98 fd ff ff 0f 84 64 e7 ff ff 89 85
[  408.072319] RSP: 0018:ffff88811c3e7a30 EFLAGS: 00010246
[  408.072322] RAX: ffffed1024874c6f RBX: ffffea00042c2b40 RCX: 0000000000000f38
[  408.072324] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
[  408.072325] RBP: ffff88811c3e7ca8 R08: 0000000000000000 R09: 00000000000000c8
[  408.072326] R10: 00000000000000c8 R11: 0000000000000000 R12: 00000000000000c8
[  408.072327] R13: dffffc0000000000 R14: ffff8881243a6030 R15: 0000000000003000
[  408.072329] FS:  0000000000000000(0000) GS:ffff88823eadf000(0000)
knlGS:0000000000000000
[  408.072331] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  408.072332] CR2: 000000c0003c6000 CR3: 000000010c106005 CR4: 0000000000772ef0
[  408.072336] PKRU: 55555554
[  408.072337] Call Trace:
[  408.072338]  <TASK>
[  408.072340]  ? sched_clock_noinstr+0x9/0x10
[  408.072344]  ? __pfx_ceph_con_v2_try_read+0x10/0x10
[  408.072347]  ? _raw_spin_unlock+0xe/0x40
[  408.072349]  ? finish_task_switch.isra.0+0x15d/0x830
[  408.072353]  ? __kasan_check_write+0x14/0x30
[  408.072357]  ? mutex_lock+0x84/0xe0
[  408.072359]  ? __pfx_mutex_lock+0x10/0x10
[  408.072361]  ceph_con_workfn+0x27e/0x10e0
[  408.072364]  ? metric_delayed_work+0x311/0x2c50
[  408.072367]  process_one_work+0x611/0xe20
[  408.072371]  ? __kasan_check_write+0x14/0x30
[  408.072373]  worker_thread+0x7e3/0x1580
[  408.072375]  ? __pfx__raw_spin_lock_irqsave+0x10/0x10
[  408.072378]  ? __pfx_worker_thread+0x10/0x10
[  408.072381]  kthread+0x381/0x7a0
[  408.072383]  ? __pfx__raw_spin_lock_irq+0x10/0x10
[  408.072385]  ? __pfx_kthread+0x10/0x10
[  408.072387]  ? __kasan_check_write+0x14/0x30
[  408.072389]  ? recalc_sigpending+0x160/0x220
[  408.072392]  ? _raw_spin_unlock_irq+0xe/0x50
[  408.072394]  ? calculate_sigpending+0x78/0xb0
[  408.072395]  ? __pfx_kthread+0x10/0x10
[  408.072397]  ret_from_fork+0x2b6/0x380
[  408.072400]  ? __pfx_kthread+0x10/0x10
[  408.072402]  ret_from_fork_asm+0x1a/0x30
[  408.072406]  </TASK>
[  408.072407] ---[ end trace 0000000000000000 ]---
[  408.072418] Oops: general protection fault, probably for non-canonical
address 0xdffffc0000000000: 0000 [#1] SMP KASAN NOPTI
[  408.072984] KASAN: null-ptr-deref in range [0x0000000000000000-
0x0000000000000007]
[  408.073350] CPU: 1 UID: 0 PID: 392 Comm: kworker/1:3 Tainted: G        W
6.17.0-rc7+ #1 PREEMPT(voluntary)
[  408.073886] Tainted: [W]=WARN
[  408.074042] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
1.17.0-5.fc42 04/01/2014
[  408.074468] Workqueue: ceph-msgr ceph_con_workfn
[  408.074694] RIP: 0010:ceph_msg_data_advance+0x79/0x1a80
[  408.074976] Code: fc ff df 49 8d 77 08 48 c1 ee 03 80 3c 16 00 0f 85 07 11 00
00 48 ba 00 00 00 00 00 fc ff df 49 8b 5f 08 48 89 de 48 c1 ee 03 <0f> b6 14 16
84 d2 74 09 80 fa 03 0f 8e 0f 0e 00 00 8b 13 83 fa 03
[  408.075884] RSP: 0018:ffff88811c3e7990 EFLAGS: 00010246
[  408.076305] RAX: ffff8881243a6388 RBX: 0000000000000000 RCX: 0000000000000000
[  408.076909] RDX: dffffc0000000000 RSI: 0000000000000000 RDI: ffff8881243a6378
[  408.077466] RBP: ffff88811c3e7a20 R08: 0000000000000000 R09: 00000000000000c8
[  408.078034] R10: ffff8881243a6388 R11: 0000000000000000 R12: ffffed1024874c71
[  408.078575] R13: dffffc0000000000 R14: ffff8881243a6030 R15: ffff8881243a6378
[  408.079159] FS:  0000000000000000(0000) GS:ffff88823eadf000(0000)
knlGS:0000000000000000
[  408.079736] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  408.080039] CR2: 000000c0003c6000 CR3: 000000010c106005 CR4: 0000000000772ef0
[  408.080376] PKRU: 55555554
[  408.080513] Call Trace:
[  408.080630]  <TASK>
[  408.080729]  ceph_con_v2_try_read+0x49b9/0x72f0
[  408.081115]  ? __pfx_ceph_con_v2_try_read+0x10/0x10
[  408.081348]  ? _raw_spin_unlock+0xe/0x40
[  408.081538]  ? finish_task_switch.isra.0+0x15d/0x830
[  408.081768]  ? __kasan_check_write+0x14/0x30
[  408.081986]  ? mutex_lock+0x84/0xe0
[  408.082160]  ? __pfx_mutex_lock+0x10/0x10
[  408.082343]  ceph_con_workfn+0x27e/0x10e0
[  408.082529]  ? metric_delayed_work+0x311/0x2c50
[  408.082737]  process_one_work+0x611/0xe20
[  408.082948]  ? __kasan_check_write+0x14/0x30
[  408.083156]  worker_thread+0x7e3/0x1580
[  408.083331]  ? __pfx__raw_spin_lock_irqsave+0x10/0x10
[  408.083557]  ? __pfx_worker_thread+0x10/0x10
[  408.083751]  kthread+0x381/0x7a0
[  408.083922]  ? __pfx__raw_spin_lock_irq+0x10/0x10
[  408.084139]  ? __pfx_kthread+0x10/0x10
[  408.084310]  ? __kasan_check_write+0x14/0x30
[  408.084510]  ? recalc_sigpending+0x160/0x220
[  408.084708]  ? _raw_spin_unlock_irq+0xe/0x50
[  408.084917]  ? calculate_sigpending+0x78/0xb0
[  408.085138]  ? __pfx_kthread+0x10/0x10
[  408.085335]  ret_from_fork+0x2b6/0x380
[  408.085525]  ? __pfx_kthread+0x10/0x10
[  408.085720]  ret_from_fork_asm+0x1a/0x30
[  408.085922]  </TASK>
[  408.086036] Modules linked in: intel_rapl_msr intel_rapl_common
intel_uncore_frequency_common intel_pmc_core pmt_telemetry pmt_discovery
pmt_class intel_pmc_ssram_telemetry intel_vsec kvm_intel joydev kvm irqbypass
polyval_clmulni ghash_clmulni_intel aesni_intel rapl input_leds psmouse
serio_raw i2c_piix4 vga16fb bochs vgastate i2c_smbus floppy mac_hid qemu_fw_cfg
pata_acpi sch_fq_codel rbd msr parport_pc ppdev lp parport efi_pstore
[  408.087778] ---[ end trace 0000000000000000 ]---
[  408.088007] RIP: 0010:ceph_msg_data_advance+0x79/0x1a80
[  408.088260] Code: fc ff df 49 8d 77 08 48 c1 ee 03 80 3c 16 00 0f 85 07 11 00
00 48 ba 00 00 00 00 00 fc ff df 49 8b 5f 08 48 89 de 48 c1 ee 03 <0f> b6 14 16
84 d2 74 09 80 fa 03 0f 8e 0f 0e 00 00 8b 13 83 fa 03
[  408.089118] RSP: 0018:ffff88811c3e7990 EFLAGS: 00010246
[  408.089357] RAX: ffff8881243a6388 RBX: 0000000000000000 RCX: 0000000000000000
[  408.089678] RDX: dffffc0000000000 RSI: 0000000000000000 RDI: ffff8881243a6378
[  408.090020] RBP: ffff88811c3e7a20 R08: 0000000000000000 R09: 00000000000000c8
[  408.090360] R10: ffff8881243a6388 R11: 0000000000000000 R12: ffffed1024874c71
[  408.090687] R13: dffffc0000000000 R14: ffff8881243a6030 R15: ffff8881243a6378
[  408.091035] FS:  0000000000000000(0000) GS:ffff88823eadf000(0000)
knlGS:0000000000000000
[  408.091452] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  408.092015] CR2: 000000c0003c6000 CR3: 000000010c106005 CR4: 0000000000772ef0
[  408.092530] PKRU: 55555554
[  417.112915]
==================================================================
[  417.113491] BUG: KASAN: slab-use-after-free in
__mutex_lock.constprop.0+0x1522/0x1610
[  417.114014] Read of size 4 at addr ffff888124870034 by task kworker/2:0/4951

[  417.114587] CPU: 2 UID: 0 PID: 4951 Comm: kworker/2:0 Tainted: G      D W
6.17.0-rc7+ #1 PREEMPT(voluntary)
[  417.114592] Tainted: [D]=DIE, [W]=WARN
[  417.114593] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
1.17.0-5.fc42 04/01/2014
[  417.114596] Workqueue: events handle_timeout
[  417.114601] Call Trace:
[  417.114602]  <TASK>
[  417.114604]  dump_stack_lvl+0x5c/0x90
[  417.114610]  print_report+0x171/0x4dc
[  417.114613]  ? __pfx__raw_spin_lock_irqsave+0x10/0x10
[  417.114617]  ? kasan_complete_mode_report_info+0x80/0x220
[  417.114621]  kasan_report+0xbd/0x100
[  417.114625]  ? __mutex_lock.constprop.0+0x1522/0x1610
[  417.114628]  ? __mutex_lock.constprop.0+0x1522/0x1610
[  417.114630]  __asan_report_load4_noabort+0x14/0x30
[  417.114633]  __mutex_lock.constprop.0+0x1522/0x1610
[  417.114635]  ? queue_con_delay+0x8d/0x200
[  417.114638]  ? __pfx___mutex_lock.constprop.0+0x10/0x10
[  417.114641]  ? __send_subscribe+0x529/0xb20
[  417.114644]  __mutex_lock_slowpath+0x13/0x20
[  417.114646]  mutex_lock+0xd4/0xe0
[  417.114649]  ? __pfx_mutex_lock+0x10/0x10
[  417.114652]  ? ceph_monc_renew_subs+0x2a/0x40
[  417.114654]  ceph_con_keepalive+0x22/0x110
[  417.114656]  handle_timeout+0x6b3/0x11d0
[  417.114659]  ? _raw_spin_unlock_irq+0xe/0x50
[  417.114662]  ? __pfx_handle_timeout+0x10/0x10
[  417.114664]  ? queue_delayed_work_on+0x8e/0xa0
[  417.114669]  process_one_work+0x611/0xe20
[  417.114672]  ? __kasan_check_write+0x14/0x30
[  417.114676]  worker_thread+0x7e3/0x1580
[  417.114678]  ? __pfx__raw_spin_lock_irqsave+0x10/0x10
[  417.114682]  ? __pfx_sched_setscheduler_nocheck+0x10/0x10
[  417.114687]  ? __pfx_worker_thread+0x10/0x10
[  417.114689]  kthread+0x381/0x7a0
[  417.114692]  ? __pfx__raw_spin_lock_irq+0x10/0x10
[  417.114694]  ? __pfx_kthread+0x10/0x10
[  417.114697]  ? __kasan_check_write+0x14/0x30
[  417.114699]  ? recalc_sigpending+0x160/0x220
[  417.114703]  ? _raw_spin_unlock_irq+0xe/0x50
[  417.114705]  ? calculate_sigpending+0x78/0xb0
[  417.114707]  ? __pfx_kthread+0x10/0x10
[  417.114710]  ret_from_fork+0x2b6/0x380
[  417.114713]  ? __pfx_kthread+0x10/0x10
[  417.114715]  ret_from_fork_asm+0x1a/0x30
[  417.114720]  </TASK>

[  417.125171] Allocated by task 2:
[  417.125333]  kasan_save_stack+0x26/0x60
[  417.125522]  kasan_save_track+0x14/0x40
[  417.125742]  kasan_save_alloc_info+0x39/0x60
[  417.125945]  __kasan_slab_alloc+0x8b/0xb0
[  417.126133]  kmem_cache_alloc_node_noprof+0x13b/0x460
[  417.126381]  copy_process+0x320/0x6250
[  417.126595]  kernel_clone+0xb7/0x840
[  417.126792]  kernel_thread+0xd6/0x120
[  417.126995]  kthreadd+0x85c/0xbe0
[  417.127176]  ret_from_fork+0x2b6/0x380
[  417.127378]  ret_from_fork_asm+0x1a/0x30

[  417.127692] Freed by task 0:
[  417.127851]  kasan_save_stack+0x26/0x60
[  417.128057]  kasan_save_track+0x14/0x40
[  417.128267]  kasan_save_free_info+0x3b/0x60
[  417.128491]  __kasan_slab_free+0x6c/0xa0
[  417.128708]  kmem_cache_free+0x182/0x550
[  417.128906]  free_task+0xeb/0x140
[  417.129070]  __put_task_struct+0x1d2/0x4f0
[  417.129259]  __put_task_struct_rcu_cb+0x15/0x20
[  417.129480]  rcu_do_batch+0x3d3/0xe70
[  417.129681]  rcu_core+0x549/0xb30
[  417.129839]  rcu_core_si+0xe/0x20
[  417.130005]  handle_softirqs+0x160/0x570
[  417.130190]  __irq_exit_rcu+0x189/0x1e0
[  417.130369]  irq_exit_rcu+0xe/0x20
[  417.130531]  sysvec_apic_timer_interrupt+0x9f/0xd0
[  417.130768]  asm_sysvec_apic_timer_interrupt+0x1b/0x20

[  417.131082] Last potentially related work creation:
[  417.131305]  kasan_save_stack+0x26/0x60
[  417.131484]  kasan_record_aux_stack+0xae/0xd0
[  417.131695]  __call_rcu_common+0xcd/0x14b0
[  417.131909]  call_rcu+0x31/0x50
[  417.132071]  delayed_put_task_struct+0x128/0x190
[  417.132295]  rcu_do_batch+0x3d3/0xe70
[  417.132478]  rcu_core+0x549/0xb30
[  417.132658]  rcu_core_si+0xe/0x20
[  417.132808]  handle_softirqs+0x160/0x570
[  417.132993]  __irq_exit_rcu+0x189/0x1e0
[  417.133181]  irq_exit_rcu+0xe/0x20
[  417.133353]  sysvec_apic_timer_interrupt+0x9f/0xd0
[  417.133584]  asm_sysvec_apic_timer_interrupt+0x1b/0x20

[  417.133921] Second to last potentially related work creation:
[  417.134183]  kasan_save_stack+0x26/0x60
[  417.134362]  kasan_record_aux_stack+0xae/0xd0
[  417.134566]  __call_rcu_common+0xcd/0x14b0
[  417.134782]  call_rcu+0x31/0x50
[  417.134929]  put_task_struct_rcu_user+0x58/0xb0
[  417.135143]  finish_task_switch.isra.0+0x5d3/0x830
[  417.135366]  __schedule+0xd30/0x5100
[  417.135534]  schedule_idle+0x5a/0x90
[  417.135712]  do_idle+0x25f/0x410
[  417.135871]  cpu_startup_entry+0x53/0x70
[  417.136053]  start_secondary+0x216/0x2c0
[  417.136233]  common_startup_64+0x13e/0x141

[  417.136894] The buggy address belongs to the object at ffff888124870000
                which belongs to the cache task_struct of size 10504
[  417.138122] The buggy address is located 52 bytes inside of
                freed 10504-byte region [ffff888124870000ffff888124872908)

[  417.139465] The buggy address belongs to the physical page:
[  417.140016] page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0
pfn:0x124870
[  417.140789] head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0
pincount:0
[  417.141519] memcg:ffff88811aa20e01
[  417.141874] anon flags:
0x17ffffc0000040(head|node=0|zone=2|lastcpupid=0x1fffff)
[  417.142600] page_type: f5(slab)
[  417.142922] raw: 0017ffffc0000040 ffff88810094f040 0000000000000000
dead000000000001
[  417.143554] raw: 0000000000000000 0000000000030003 00000000f5000000
ffff88811aa20e01
[  417.143954] head: 0017ffffc0000040 ffff88810094f040 0000000000000000
dead000000000001
[  417.144329] head: 0000000000000000 0000000000030003 00000000f5000000
ffff88811aa20e01
[  417.144710] head: 0017ffffc0000003 ffffea0004921c01 00000000ffffffff
00000000ffffffff
[  417.145106] head: ffffffffffffffff 0000000000000000 00000000ffffffff
0000000000000008
[  417.145485] page dumped because: kasan: bad access detected

[  417.145859] Memory state around the buggy address:
[  417.146094]  ffff88812486ff00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
fc
[  417.146439]  ffff88812486ff80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
fc
[  417.146791] >ffff888124870000: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb
fb
[  417.147145]                                      ^
[  417.147387]  ffff888124870080: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
fb
[  417.147751]  ffff888124870100: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
fb
[  417.148123]
==================================================================

First of all, we have warning in get_bvec_at() because
cursor->total_resid contains zero value. And, finally,
we have crash in ceph_msg_data_advance() because
cursor->data is NULL. It means that get_bvec_at()
receives not initialized ceph_msg_data_cursor structure
because data is NULL and total_resid contains zero.

Moreover, we don't have likewise issue for the case of
Ceph msgr1 protocol because ceph_msg_data_cursor_init()
has been called before reading sparse data.

This patch adds calling of ceph_msg_data_cursor_init()
in the beginning of process_v2_sparse_read() with
the goal to guarantee that logic of reading sparse data
works correctly for the case of Ceph msgr2 protocol.

Cc: stable@vger.kernel.org
Link: https://tracker.ceph.com/issues/73152
Signed-off-by: Viacheslav Dubeyko <Slava.Dubeyko@ibm.com>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
3 weeks agoMerge tag 'v6.18rc7-SMB-client-fix' of git://git.samba.org/sfrench/cifs-2.6
Linus Torvalds [Thu, 27 Nov 2025 04:22:20 +0000 (20:22 -0800)] 
Merge tag 'v6.18rc7-SMB-client-fix' of git://git.samba.org/sfrench/cifs-2.6

Pull smb client fix from Steve French:
 "smb client multiuser (with cifscreds) mount fix"

* tag 'v6.18rc7-SMB-client-fix' of git://git.samba.org/sfrench/cifs-2.6:
  smb: client: fix memory leak in cifs_construct_tcon()

3 weeks agoMerge tag 'linux-can-fixes-for-6.18-20251126' of git://git.kernel.org/pub/scm/linux...
Jakub Kicinski [Thu, 27 Nov 2025 03:56:00 +0000 (19:56 -0800)] 
Merge tag 'linux-can-fixes-for-6.18-20251126' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can

Marc Kleine-Budde says:

====================
pull-request: can 2025-11-26

this is a pull request of 8 patches for net/main.

Seungjin Bae provides a patch for the kvaser_usb driver to fix a
potential infinite loop in the USB data stream command parser.

Thomas Mühlbacher's patch for the sja1000 driver IRQ handler's max
loop handling, that might lead to unhandled interrupts.

3 patches by me for the gs_usb driver fix handling of failed transmit
URBs and add checking of the actual length of received URBs before
accessing the data.

The next patch is by me and is a port of Thomas Mühlbacher's patch
(fix IRQ handler's max loop handling, that might lead to unhandled
interrupts.) to the sun4i_can driver.

Biju Das provides a patch for the rcar_canfd driver to fix the CAN-FD
mode setting.

The last patch is by Shaurya Rane for the em_canid filter to ensure
that the complete CAN frame is present in the linear data buffer
before accessing it.

* tag 'linux-can-fixes-for-6.18-20251126' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can:
  net/sched: em_canid: fix uninit-value in em_canid_match
  can: rcar_canfd: Fix CAN-FD mode as default
  can: sun4i_can: sun4i_can_interrupt(): fix max irq loop handling
  can: gs_usb: gs_usb_receive_bulk_callback(): check actual_length before accessing data
  can: gs_usb: gs_usb_receive_bulk_callback(): check actual_length before accessing header
  can: gs_usb: gs_usb_xmit_callback(): fix handling of failed transmitted URBs
  can: sja1000: fix max irq loop handling
  can: kvaser_usb: leaf: Fix potential infinite loop in command parsers
====================

Link: https://patch.msgid.link/20251126155713.217105-1-mkl@pengutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonet: atlantic: fix fragment overflow handling in RX path
Jiefeng Zhang [Wed, 26 Nov 2025 03:22:49 +0000 (11:22 +0800)] 
net: atlantic: fix fragment overflow handling in RX path

The atlantic driver can receive packets with more than MAX_SKB_FRAGS (17)
fragments when handling large multi-descriptor packets. This causes an
out-of-bounds write in skb_add_rx_frag_netmem() leading to kernel panic.

The issue occurs because the driver doesn't check the total number of
fragments before calling skb_add_rx_frag(). When a packet requires more
than MAX_SKB_FRAGS fragments, the fragment index exceeds the array bounds.

Fix by assuming there will be an extra frag if buff->len > AQ_CFG_RX_HDR_SIZE,
then all fragments are accounted for. And reusing the existing check to
prevent the overflow earlier in the code path.

This crash occurred in production with an Aquantia AQC113 10G NIC.

Stack trace from production environment:
```
RIP: 0010:skb_add_rx_frag_netmem+0x29/0xd0
Code: 90 f3 0f 1e fa 0f 1f 44 00 00 48 89 f8 41 89
ca 48 89 d7 48 63 ce 8b 90 c0 00 00 00 48 c1 e1 04 48 01 ca 48 03 90
c8 00 00 00 <48> 89 7a 30 44 89 52 3c 44 89 42 38 40 f6 c7 01 75 74 48
89 fa 83
RSP: 0018:ffffa9bec02a8d50 EFLAGS: 00010287
RAX: ffff925b22e80a00 RBX: ffff925ad38d2700 RCX:
fffffffe0a0c8000
RDX: ffff9258ea95bac0 RSI: ffff925ae0a0c800 RDI:
0000000000037a40
RBP: 0000000000000024 R08: 0000000000000000 R09:
0000000000000021
R10: 0000000000000848 R11: 0000000000000000 R12:
ffffa9bec02a8e24
R13: ffff925ad8615570 R14: 0000000000000000 R15:
ffff925b22e80a00
FS: 0000000000000000(0000)
GS:ffff925e47880000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffff9258ea95baf0 CR3: 0000000166022004 CR4:
0000000000f72ef0
PKRU: 55555554
Call Trace:
<IRQ>
aq_ring_rx_clean+0x175/0xe60 [atlantic]
? aq_ring_rx_clean+0x14d/0xe60 [atlantic]
? aq_ring_tx_clean+0xdf/0x190 [atlantic]
? kmem_cache_free+0x348/0x450
? aq_vec_poll+0x81/0x1d0 [atlantic]
? __napi_poll+0x28/0x1c0
? net_rx_action+0x337/0x420
```

Fixes: 6aecbba12b5c ("net: atlantic: add check for MAX_SKB_FRAGS")
Changes in v4:
- Add Fixes: tag to satisfy patch validation requirements.

Changes in v3:
- Fix by assuming there will be an extra frag if buff->len > AQ_CFG_RX_HDR_SIZE,
  then all fragments are accounted for.

Signed-off-by: Jiefeng Zhang <jiefeng.z.zhang@gmail.com>
Link: https://patch.msgid.link/20251126032249.69358-1-jiefeng.z.zhang@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agoMAINTAINERS: separate VIRTIO NET DRIVER and add netdev
Jon Kohler [Wed, 26 Nov 2025 01:57:48 +0000 (18:57 -0700)] 
MAINTAINERS: separate VIRTIO NET DRIVER and add netdev

Changes to virtio network stack should be cc'd to netdev DL, separate
it into its own group to add netdev in addition to virtualization DL.

Signed-off-by: Jon Kohler <jon@nutanix.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20251126015750.2200267-1-jon@nutanix.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agovirtio-net: avoid unnecessary checksum calculation on guest RX
Jon Kohler [Tue, 25 Nov 2025 22:27:53 +0000 (15:27 -0700)] 
virtio-net: avoid unnecessary checksum calculation on guest RX

Commit a2fb4bc4e2a6 ("net: implement virtio helpers to handle UDP
GSO tunneling.") inadvertently altered checksum offload behavior
for guests not using UDP GSO tunneling.

Before, tun_put_user called tun_vnet_hdr_from_skb, which passed
has_data_valid = true to virtio_net_hdr_from_skb.

After, tun_put_user began calling tun_vnet_hdr_tnl_from_skb instead,
which passes has_data_valid = false into both call sites.

This caused virtio hdr flags to not include VIRTIO_NET_HDR_F_DATA_VALID
for SKBs where skb->ip_summed == CHECKSUM_UNNECESSARY. As a result,
guests are forced to recalculate checksums unnecessarily.

Restore the previous behavior by ensuring has_data_valid = true is
passed in the !tnl_gso_type case, but only from tun side, as
virtio_net_hdr_tnl_from_skb() is used also by the virtio_net driver,
which in turn must not use VIRTIO_NET_HDR_F_DATA_VALID on tx.

cc: stable@vger.kernel.org
Fixes: a2fb4bc4e2a6 ("net: implement virtio helpers to handle UDP GSO tunneling.")
Signed-off-by: Jon Kohler <jon@nutanix.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Link: https://patch.msgid.link/20251125222754.1737443-1-jon@nutanix.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agoeth: fbnic: Fix counter roll-over issue
Mohsin Bashir [Tue, 25 Nov 2025 21:17:04 +0000 (13:17 -0800)] 
eth: fbnic: Fix counter roll-over issue

Fix a potential counter roll-over issue in fbnic_mbx_alloc_rx_msgs()
when calculating descriptor slots. The issue occurs when head - tail
results in a large positive value (unsigned) and the compiler interprets
head - tail - 1 as a signed value.

Since FBNIC_IPC_MBX_DESC_LEN is a power of two, use a masking operation,
which is a common way of avoiding this problem when dealing with these
sort of ring space calculations.

Fixes: da3cde08209e ("eth: fbnic: Add FW communication mechanism")
Signed-off-by: Mohsin Bashir <mohsin.bashr@gmail.com>
Link: https://patch.msgid.link/20251125211704.3222413-1-mohsin.bashr@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agomptcp: clear scheduled subflows on retransmit
Paolo Abeni [Tue, 25 Nov 2025 16:59:11 +0000 (17:59 +0100)] 
mptcp: clear scheduled subflows on retransmit

When __mptcp_retrans() kicks-in, it schedules one or more subflows for
retransmission, but such subflows could be actually left alone if there
is no more data to retransmit and/or in case of concurrent fallback.

Scheduled subflows could be processed much later in time, i.e. when new
data will be transmitted, leading to bad subflow selection.

Explicitly clear all scheduled subflows before leaving the
retransmission function.

Fixes: ee2708aedad0 ("mptcp: use get_retrans wrapper")
Cc: stable@vger.kernel.org
Reported-by: Filip Pokryvka <fpokryvk@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20251125-net-mptcp-clear-sched-rtx-v1-1-1cea4ad2165f@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonet: dsa: sja1105: fix SGMII linking at 10M or 100M but not passing traffic
Vladimir Oltean [Sat, 22 Nov 2025 11:13:24 +0000 (13:13 +0200)] 
net: dsa: sja1105: fix SGMII linking at 10M or 100M but not passing traffic

When using the SGMII PCS as a fixed-link chip-to-chip connection, it is
easy to miss the fact that traffic passes only at 1G, since that's what
any normal such connection would use.

When using the SGMII PCS connected towards an on-board PHY or an SFP
module, it is immediately noticeable that when the link resolves to a
speed other than 1G, traffic from the MAC fails to pass: TX counters
increase, but nothing gets decoded by the other end, and no local RX
counters increase either.

Artificially lowering a fixed-link rate to speed = <100> makes us able
to see the same issue as in the case of having an SGMII PHY.

Some debugging shows that the XPCS configuration is A-OK, but that the
MAC Configuration Table entry for the port has the SPEED bits still set
to 1000Mbps, due to a special condition in the driver. Deleting that
condition, and letting the resolved link speed be programmed directly
into the MAC speed field, results in a functional link at all 3 speeds.

This piece of evidence, based on testing on both generations with SGMII
support (SJA1105S and SJA1110A) directly contradicts the statement from
the blamed commit that "the MAC is fixed at 1 Gbps and we need to
configure the PCS only (if even that)". Worse, that statement is not
backed by any documentation, and no one from NXP knows what it might
refer to.

I am unable to recall sufficient context regarding my testing from March
2020 to understand what led me to draw such a braindead and factually
incorrect conclusion. Yet, there is nothing of value regarding forcing
the MAC speed, either for SGMII or 2500Base-X (introduced at a later
stage), so remove all such logic.

Fixes: ffe10e679cec ("net: dsa: sja1105: Add support for the SGMII port")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://patch.msgid.link/20251122111324.136761-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agos390/net: list Aswin Karuvally as maintainer
Alexandra Winter [Tue, 25 Nov 2025 08:58:29 +0000 (09:58 +0100)] 
s390/net: list Aswin Karuvally as maintainer

Thank you Aswin for taking this responsibility.

Signed-off-by: Alexandra Winter <wintera@linux.ibm.com>
Acked-by: Aswin Karuvally <aswin@linux.ibm.com>
Link: https://patch.msgid.link/20251125085829.3679506-1-wintera@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agonet: wwan: mhi: Keep modem name match with Foxconn T99W640
Slark Xiao [Tue, 25 Nov 2025 07:09:00 +0000 (15:09 +0800)] 
net: wwan: mhi: Keep modem name match with Foxconn T99W640

Correct it since M.2 device T99W640 has updated from T99W515.
We need to align it with MHI side otherwise this modem can't
get the network.

Fixes: ae5a34264354 ("bus: mhi: host: pci_generic: Fix the modem name of Foxconn T99W640")
Signed-off-by: Slark Xiao <slark_xiao@163.com>
Reviewed-by: Loic Poulain <loic.poulain@oss.qualcomm.com>
Link: https://patch.msgid.link/20251125070900.33324-1-slark_xiao@163.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agovhost: rewind next_avail_head while discarding descriptors
Jason Wang [Thu, 20 Nov 2025 02:29:50 +0000 (10:29 +0800)] 
vhost: rewind next_avail_head while discarding descriptors

When discarding descriptors with IN_ORDER, we should rewind
next_avail_head otherwise it would run out of sync with
last_avail_idx. This would cause driver to report
"id X is not a head".

Fixing this by returning the number of descriptors that is used for
each buffer via vhost_get_vq_desc_n() so caller can use the value
while discarding descriptors.

Fixes: 67a873df0c41 ("vhost: basic in order support")
Cc: stable@vger.kernel.org
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20251120022950.10117-1-jasowang@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
3 weeks agolibceph: drop started parameter of __ceph_open_session()
Ilya Dryomov [Mon, 3 Nov 2025 20:34:01 +0000 (21:34 +0100)] 
libceph: drop started parameter of __ceph_open_session()

With the previous commit revamping the timeout handling, started isn't
used anymore.  It could be taken into account by adjusting the initial
value of the timeout, but there is little point as both callers capture
the timestamp shortly before calling __ceph_open_session() -- the only
thing of note that happens in the interim is taking client->mount_mutex
and that isn't expected to take multiple seconds.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Reviewed-by: Viacheslav Dubeyko <Slava.Dubeyko@ibm.com>
3 weeks agolibceph: fix potential use-after-free in have_mon_and_osd_map()
Ilya Dryomov [Mon, 3 Nov 2025 20:34:01 +0000 (21:34 +0100)] 
libceph: fix potential use-after-free in have_mon_and_osd_map()

The wait loop in __ceph_open_session() can race with the client
receiving a new monmap or osdmap shortly after the initial map is
received.  Both ceph_monc_handle_map() and handle_one_map() install
a new map immediately after freeing the old one

    kfree(monc->monmap);
    monc->monmap = monmap;

    ceph_osdmap_destroy(osdc->osdmap);
    osdc->osdmap = newmap;

under client->monc.mutex and client->osdc.lock respectively, but
because neither is taken in have_mon_and_osd_map() it's possible for
client->monc.monmap->epoch and client->osdc.osdmap->epoch arms in

    client->monc.monmap && client->monc.monmap->epoch &&
        client->osdc.osdmap && client->osdc.osdmap->epoch;

condition to dereference an already freed map.  This happens to be
reproducible with generic/395 and generic/397 with KASAN enabled:

    BUG: KASAN: slab-use-after-free in have_mon_and_osd_map+0x56/0x70
    Read of size 4 at addr ffff88811012d810 by task mount.ceph/13305
    CPU: 2 UID: 0 PID: 13305 Comm: mount.ceph Not tainted 6.14.0-rc2-build2+ #1266
    ...
    Call Trace:
    <TASK>
    have_mon_and_osd_map+0x56/0x70
    ceph_open_session+0x182/0x290
    ceph_get_tree+0x333/0x680
    vfs_get_tree+0x49/0x180
    do_new_mount+0x1a3/0x2d0
    path_mount+0x6dd/0x730
    do_mount+0x99/0xe0
    __do_sys_mount+0x141/0x180
    do_syscall_64+0x9f/0x100
    entry_SYSCALL_64_after_hwframe+0x76/0x7e
    </TASK>

    Allocated by task 13305:
    ceph_osdmap_alloc+0x16/0x130
    ceph_osdc_init+0x27a/0x4c0
    ceph_create_client+0x153/0x190
    create_fs_client+0x50/0x2a0
    ceph_get_tree+0xff/0x680
    vfs_get_tree+0x49/0x180
    do_new_mount+0x1a3/0x2d0
    path_mount+0x6dd/0x730
    do_mount+0x99/0xe0
    __do_sys_mount+0x141/0x180
    do_syscall_64+0x9f/0x100
    entry_SYSCALL_64_after_hwframe+0x76/0x7e

    Freed by task 9475:
    kfree+0x212/0x290
    handle_one_map+0x23c/0x3b0
    ceph_osdc_handle_map+0x3c9/0x590
    mon_dispatch+0x655/0x6f0
    ceph_con_process_message+0xc3/0xe0
    ceph_con_v1_try_read+0x614/0x760
    ceph_con_workfn+0x2de/0x650
    process_one_work+0x486/0x7c0
    process_scheduled_works+0x73/0x90
    worker_thread+0x1c8/0x2a0
    kthread+0x2ec/0x300
    ret_from_fork+0x24/0x40
    ret_from_fork_asm+0x1a/0x30

Rewrite the wait loop to check the above condition directly with
client->monc.mutex and client->osdc.lock taken as appropriate.  While
at it, improve the timeout handling (previously mount_timeout could be
exceeded in case wait_event_interruptible_timeout() slept more than
once) and access client->auth_err under client->monc.mutex to match
how it's set in finish_auth().

monmap_show() and osdmap_show() now take the respective lock before
accessing the map as well.

Cc: stable@vger.kernel.org
Reported-by: David Howells <dhowells@redhat.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Reviewed-by: Viacheslav Dubeyko <Slava.Dubeyko@ibm.com>
3 weeks agoMerge tag 'trace-ringbuffer-v6.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Wed, 26 Nov 2025 21:16:22 +0000 (13:16 -0800)] 
Merge tag 'trace-ringbuffer-v6.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace

Pull ring-buffer fix from Steven Rostedt:

 - Do not allow mmapped ring buffer to be split

   When the ring buffer VMA is split by a partial munmap or a MAP_FIXED,
   the kernel calls vm_ops->close() on each portion. This causes the
   ring_buffer_unmap() to be called multiple times. This causes
   subsequent calls to return -ENODEV and triggers a warning.

   There's no reason to allow user space to split up memory mapping of
   the ring buffer. Have it return -EINVAL when that happens.

* tag 'trace-ringbuffer-v6.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  tracing: Fix WARN_ON in tracing_buffers_mmap_close for split VMAs

3 weeks agodma-direct: Fix missing sg_dma_len assignment in P2PDMA bus mappings
Pranjal Shrivastava [Wed, 26 Nov 2025 11:41:12 +0000 (11:41 +0000)] 
dma-direct: Fix missing sg_dma_len assignment in P2PDMA bus mappings

Prior to commit a25e7962db0d7 ("PCI/P2PDMA: Refactor the p2pdma mapping
helpers"), P2P segments were mapped using the pci_p2pdma_map_segment()
helper. This helper was responsible for populating sg->dma_address,
marking the bus address, and also setting sg_dma_len(sg).

The refactor[1] removed this helper and moved the mapping logic directly
into the callers. While iommu_dma_map_sg() was correctly updated to set
the length in the new flow, it was missed in dma_direct_map_sg().

Thus, in dma_direct_map_sg(), the PCI_P2PDMA_MAP_BUS_ADDR case sets the
dma_address and marks the segment, but immediately executes 'continue',
which causes the loop to skip the standard assignment logic at the end:

    sg_dma_len(sg) = sg->length;

As a result, when CONFIG_NEED_SG_DMA_LENGTH is enabled, the dma_length
field remains uninitialized (zero) for P2P bus address mappings. This
breaks upper-layer drivers (for e.g. RDMA/IB) that rely on sg_dma_len()
to determine the transfer size.

Fix this by explicitly setting the DMA length in the
PCI_P2PDMA_MAP_BUS_ADDR case before continuing to the next scatterlist
entry.

Fixes: a25e7962db0d7 ("PCI/P2PDMA: Refactor the p2pdma mapping helpers")
Reported-by: Jacob Moroni <jmoroni@google.com>
Signed-off-by: Pranjal Shrivastava <praan@google.com>
[1]
https://lore.kernel.org/all/ac14a0e94355bf898de65d023ccf8a2ad22a3ece.1746424934.git.leon@kernel.org/

Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Shivaji Kant <shivajikant@google.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Link: https://lore.kernel.org/r/20251126114112.3694469-1-praan@google.com
3 weeks agoMerge tag 'mm-hotfixes-stable-2025-11-26-11-51' of git://git.kernel.org/pub/scm/linux...
Linus Torvalds [Wed, 26 Nov 2025 20:38:05 +0000 (12:38 -0800)] 
Merge tag 'mm-hotfixes-stable-2025-11-26-11-51' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

Pull misc fixes from Andrew Morton:
 "8 hotfixes.  4 are cc:stable, 7 are against mm/.

  All are singletons - please see the respective changelogs for details"

* tag 'mm-hotfixes-stable-2025-11-26-11-51' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
  mm/filemap: fix logic around SIGBUS in filemap_map_pages()
  mm/huge_memory: fix NULL pointer deference when splitting folio
  MAINTAINERS: add test_kho to KHO's entry
  mailmap: add entry for Sam Protsenko
  selftests/mm: fix division-by-zero in uffd-unit-tests
  mm/mmap_lock: reset maple state on lock_vma_under_rcu() retry
  mm/memfd: fix information leak in hugetlb folios
  mm: swap: remove duplicate nr_swap_pages decrement in get_swap_page_of_type()

3 weeks agoFix Intel Dollar Cove TI battery driver 32-bit build error
Linus Torvalds [Wed, 26 Nov 2025 20:31:48 +0000 (12:31 -0800)] 
Fix Intel Dollar Cove TI battery driver 32-bit build error

The driver is doing a 64-bit divide, rather than using the proper
helpers, causing link errors on i386 allyesconfig builds:

  x86_64-linux-ld: drivers/power/supply/intel_dc_ti_battery.o: in function `dc_ti_battery_get_voltage_and_current_now':
  intel_dc_ti_battery.c:(.text+0x5c): undefined reference to `__udivdi3'
  x86_64-linux-ld: intel_dc_ti_battery.c:(.text+0x96): undefined reference to `__udivdi3'

and while fixing that, fix the double rounding: keep the timing
difference in nanoseconds ('ktime'), and then just convert to usecs at
the end.

Not because the timing precision is likely to matter, but because doing
it right also makes the code simpler.

Reported-by: Guenter Roeck <linux@roeck-us.net>
Cc: Hans de Goede <hansg@kernel.org>
Cc: Sebastian Reichel <sebastian.reichel@collabora.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 weeks agoIncrease the default 32-bit build frame size warning limit to 1280 bytes
Linus Torvalds [Wed, 26 Nov 2025 20:11:28 +0000 (12:11 -0800)] 
Increase the default 32-bit build frame size warning limit to 1280 bytes

That was already the limit with KASAN enabled, and the 32-bit x86 build
ends up having a couple of drm cases that have stack frames _just_ over
1kB on my allmodconfig test.  So the minimal fix for this build issue
for now is to just bump the limit and make it independent of KASAN.

[ Side note: XTENSA already used 1.5k and PARISC uses 2k, so 1280 is
  still relatively conservative ]

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 weeks agoMerge tag 'sound-6.18' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Linus Torvalds [Wed, 26 Nov 2025 18:33:32 +0000 (10:33 -0800)] 
Merge tag 'sound-6.18' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "A collection of small fixes. All changes are device-specific and
  trivial, mostly HD-audio and USB-audio quirks and fixups"

* tag 'sound-6.18' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hda/realtek: Add quirk for HP ProBook 450 G8
  ALSA: usb-audio: fix uac2 clock source at terminal parser
  ALSA: hda/realtek: add quirk for HP pavilion aero laptop 13z-be200
  ALSA: hda/cirrus fix cs420x MacPro 6,1 inverted jack detection
  ALSA: usb-audio: Add DSD quirk for LEAK Stereo 230
  ALSA: au88x0: Fix incorrect error handling for PCI config reads

3 weeks agoMerge tag 'acpi-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael...
Linus Torvalds [Wed, 26 Nov 2025 18:17:10 +0000 (10:17 -0800)] 
Merge tag 'acpi-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm

Pull ACPI fix from Rafael Wysocki:
 "Revert a commit that attempted to make the code in the ACPI processor
  driver more straightforward, but it turned out to cause the kernel to
  crash on at least one system, along with some further cleanups on top
  of it"

* tag 'acpi-6.18-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  Revert "ACPI: processor: idle: Optimize ACPI idle driver registration"
  Revert "ACPI: processor: Remove unused empty stubs of some functions"
  Revert "ACPI: processor: idle: Rearrange declarations in header file"
  Revert "ACPI: processor: idle: Redefine two functions as void"
  Revert "ACPI: processor: Do not expose global variable acpi_idle_driver"

3 weeks agodrm/amdgpu: fix cyan_skillfish2 gpu info fw handling
Alex Deucher [Wed, 26 Nov 2025 14:40:31 +0000 (09:40 -0500)] 
drm/amdgpu: fix cyan_skillfish2 gpu info fw handling

If the board supports IP discovery, we don't need to
parse the gpu info firmware.

Backport to 6.18.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4721
Fixes: fa819e3a7c1e ("drm/amdgpu: add support for cyan skillfish gpu_info")
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 5427e32fa3a0ba9a016db83877851ed277b065fb)

3 weeks agodrm/amdgpu: attach tlb fence to the PTs update
Prike Liang [Fri, 31 Oct 2025 09:02:51 +0000 (17:02 +0800)] 
drm/amdgpu: attach tlb fence to the PTs update

Ensure the userq TLB flush is emitted only after
the VM update finishes and the PT BOs have been
annotated with bookkeeping fences.

Suggested-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit f3854e04b708d73276c4488231a8bd66d30b4671)
Cc: stable@vger.kernel.org
3 weeks agodrm/amd/display: Increase EDID read retries
Mario Limonciello (AMD) [Thu, 6 Nov 2025 05:04:54 +0000 (23:04 -0600)] 
drm/amd/display: Increase EDID read retries

[WHY]
When monitor is still booting EDID read can fail while DPCD read
is successful.  In this case no EDID data will be returned, and this
could happen for a while.

[HOW]
Increase number of attempts to read EDID in dm_helpers_read_local_edid()
to 25.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4672
Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit a76d6f2c76c3abac519ba753e2723e6ffe8e461c)
Cc: stable@vger.kernel.org
3 weeks agodrm/amd/display: Don't change brightness for disabled connectors
Mario Limonciello (AMD) [Mon, 3 Nov 2025 22:02:11 +0000 (16:02 -0600)] 
drm/amd/display: Don't change brightness for disabled connectors

[WHY]
When a laptop lid is closed the connector is disabled but userspace
can still try to change brightness.  This doesn't work because the
panel is turned off. It will eventually time out, but there is a lot
of stutter along the way.

[How]
Iterate all connectors to check whether the matching one for the backlight
index is enabled.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4675
Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Ray Wu <ray.wu@amd.com>
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit f6eeab30323d1174a4cc022e769d248fe8241304)
Cc: stable@vger.kernel.org
3 weeks agodrm/amd/display: Check NULL before accessing
Alex Hung [Fri, 7 Nov 2025 22:35:58 +0000 (15:35 -0700)] 
drm/amd/display: Check NULL before accessing

[WHAT]
IGT kms_cursor_legacy's long-nonblocking-modeset-vs-cursor-atomic
fails with NULL pointer dereference. This can be reproduced with
both an eDP panel and a DP monitors connected.

 BUG: kernel NULL pointer dereference, address: 0000000000000000
 #PF: supervisor read access in kernel mode
 #PF: error_code(0x0000) - not-present page
 PGD 0 P4D 0
 Oops: Oops: 0000 [#1] SMP NOPTI
 CPU: 13 UID: 0 PID: 2960 Comm: kms_cursor_lega Not tainted
6.16.0-99-custom #8 PREEMPT(voluntary)
 Hardware name: AMD ........
 RIP: 0010:dc_stream_get_scanoutpos+0x34/0x130 [amdgpu]
 Code: 57 4d 89 c7 41 56 49 89 ce 41 55 49 89 d5 41 54 49
 89 fc 53 48 83 ec 18 48 8b 87 a0 64 00 00 48 89 75 d0 48 c7 c6 e0 41 30
 c2 <48> 8b 38 48 8b 9f 68 06 00 00 e8 8d d7 fd ff 31 c0 48 81 c3 e0 02
 RSP: 0018:ffffd0f3c2bd7608 EFLAGS: 00010292
 RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffd0f3c2bd7668
 RDX: ffffd0f3c2bd7664 RSI: ffffffffc23041e0 RDI: ffff8b32494b8000
 RBP: ffffd0f3c2bd7648 R08: ffffd0f3c2bd766c R09: ffffd0f3c2bd7760
 R10: ffffd0f3c2bd7820 R11: 0000000000000000 R12: ffff8b32494b8000
 R13: ffffd0f3c2bd7664 R14: ffffd0f3c2bd7668 R15: ffffd0f3c2bd766c
 FS:  000071f631b68700(0000) GS:ffff8b399f114000(0000)
knlGS:0000000000000000
 CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
 CR2: 0000000000000000 CR3: 00000001b8105000 CR4: 0000000000f50ef0
 PKRU: 55555554
 Call Trace:
 <TASK>
 dm_crtc_get_scanoutpos+0xd7/0x180 [amdgpu]
 amdgpu_display_get_crtc_scanoutpos+0x86/0x1c0 [amdgpu]
 ? __pfx_amdgpu_crtc_get_scanout_position+0x10/0x10[amdgpu]
 amdgpu_crtc_get_scanout_position+0x27/0x50 [amdgpu]
 drm_crtc_vblank_helper_get_vblank_timestamp_internal+0xf7/0x400
 drm_crtc_vblank_helper_get_vblank_timestamp+0x1c/0x30
 drm_crtc_get_last_vbltimestamp+0x55/0x90
 drm_crtc_next_vblank_start+0x45/0xa0
 drm_atomic_helper_wait_for_fences+0x81/0x1f0
 ...

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 621e55f1919640acab25383362b96e65f2baea3c)
Cc: stable@vger.kernel.org
3 weeks agoRevert "drm/amd/display: Move setup_stream_attribute"
Alex Deucher [Tue, 25 Nov 2025 14:08:45 +0000 (09:08 -0500)] 
Revert "drm/amd/display: Move setup_stream_attribute"

This reverts commit 2681bf4ae8d24df950138b8c9ea9c271cd62e414.

This results in a blank screen on the HDMI port on some systems.
Revert for now so as not to regress 6.18, can be addressed
in 6.19 once the issue is root caused.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4652
Cc: Sunpeng.Li@amd.com
Cc: ivan.lipski@amd.com
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit d0e9de7a81503cdde37fb2d37f1d102f9e0f38fb)

3 weeks agonet/sched: em_canid: fix uninit-value in em_canid_match
Shaurya Rane [Wed, 26 Nov 2025 08:57:18 +0000 (14:27 +0530)] 
net/sched: em_canid: fix uninit-value in em_canid_match

Use pskb_may_pull() to ensure a complete CAN frame is present in the
linear data buffer before reading the CAN ID. A simple skb->len check
is insufficient because it only verifies the total data length but does
not guarantee the data is present in skb->data (it could be in
fragments).

pskb_may_pull() both validates the length and pulls fragmented data
into the linear buffer if necessary, making it safe to directly
access skb->data.

Reported-by: syzbot+5d8269a1e099279152bc@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=5d8269a1e099279152bc
Fixes: f057bbb6f9ed ("net: em_canid: Ematch rule to match CAN frames according to their identifiers")
Signed-off-by: Shaurya Rane <ssrane_b23@ee.vjti.ac.in>
Link: https://patch.msgid.link/20251126085718.50808-1-ssranevjti@gmail.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
3 weeks agocan: rcar_canfd: Fix CAN-FD mode as default
Biju Das [Tue, 18 Nov 2025 12:39:25 +0000 (12:39 +0000)] 
can: rcar_canfd: Fix CAN-FD mode as default

The commit 5cff263606a1 ("can: rcar_canfd: Fix controller mode setting")
has aligned with the flow mentioned in the hardware manual for all SoCs
except R-Car Gen3 and RZ/G2L SoCs. On R-Car Gen4 and RZ/G3E SoCs, due to
the wrong logic in the commit[1] sets the default mode to FD-Only mode
instead of CAN-FD mode.

This patch sets the CAN-FD mode as the default for all SoCs by dropping
the rcar_canfd_set_mode() as some SoC requires mode setting in global
reset mode, and the rest of the SoCs in channel reset mode and update the
rcar_canfd_reset_controller() to take care of these constraints. Moreover,
the RZ/G3E and R-Car Gen4 SoCs support 3 modes compared to 2 modes on the
R-Car Gen3. Use inverted logic in rcar_canfd_reset_controller() to
simplify the code later to support FD-only mode.

[1]
commit 45721c406dcf ("can: rcar_canfd: Add support for r8a779a0 SoC")

Fixes: 5cff263606a1 ("can: rcar_canfd: Fix controller mode setting")
Cc: stable@vger.kernel.org
Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://patch.msgid.link/20251118123926.193445-1-biju.das.jz@bp.renesas.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>