[Deepin-Kernel-SIG] [linux 6.6-y] [Upstream] Update kernel base to 6.6.127

[opsiff]

Update kernel base to 6.6.127.

git log --oneline v6.6.126..v6.6.127 | wc
40 348 2850

Merged:
LoongArch: Add writecombine support for DMW-based ioremap()

Alban Bedel (1):
gpiolib: acpi: Fix gpio count with string references

Anatolii Shirykalov (1):
ASoC: amd: yc: Add ASUS ExpertBook PM1503CDA to quirks list

Anil Gurumurthy (1):
scsi: qla2xxx: Fix bsg_done() causing double free

Brahmajit Das (1):
drm/tegra: hdmi: sor: Fix error: variable ‘j’ set but not used

Chao Yu (2):
f2fs: fix to avoid UAF in f2fs_write_end_io()
f2fs: fix to avoid mapping wrong physical block for swapfile

Charles Keepax (1):
ASoC: cs42l43: Correct handling of 3-pole jack load detection

Chelsy Ratnawat (1):
bus: fsl-mc: Replace snprintf and sprintf with sysfs_emit in sysfs
show functions

Chen Ridong (1):
cpuset: Fix missing adaptation for cpuset_is_populated

Christophe JAILLET (1):
PCI: endpoint: Remove unused field in struct pci_epf_group

Daniel Borkmann (1):
Revert "wireguard: device: enable threaded NAPI"

David Hildenbrand (Red Hat) (3):
mm/hugetlb: fix hugetlb_pmd_shared()
mm/hugetlb: fix two comments related to huge_pmd_unshare()
mm/hugetlb: fix excessive IPI broadcasts when unsharing PMD tables
using mmu_gather

Deepanshu Kartikey (1):
romfs: check sb_set_blocksize() return value

Dirk Su (1):
ASoC: amd: yc: Add quirk for HP 200 G2a 16

Fabio Porcedda (1):
USB: serial: option: add Telit FN920C04 RNDIS compositions

Greg Kroah-Hartman (2):
fbdev: smscufx: properly copy ioctl memory to kernelspace
Linux 6.6.127

Guangshuo Li (1):
fbdev: rivafb: fix divide error in nv3_arb()

Gui-Dong Han (1):
bus: fsl-mc: fix use-after-free in driver_override_show()

Jane Chu (1):
mm/hugetlb: fix copy_hugetlb_page_range() to use ->pt_share_count

Kanglong Wang (1):
LoongArch: Add WriteCombine shadow mapping in KASAN

Liu Song (1):
PCI: endpoint: Avoid creating sub-groups asynchronously

Menglong Dong (1):
net: tunnel: make skb_vlan_inet_prepare() return drop reasons

Rafael J. Wysocki (2):
platform/x86: classmate-laptop: Add missing NULL pointer checks
platform/x86: panasonic-laptop: Fix sysfs group leak in error path

Ricardo Rivera-Matos (1):
ASoC: cs35l45: Corrects ASP_TX5 DAPM widget channel

Tagir Garaev (1):
ASoC: Intel: sof_es8336: Add DMI quirk for Huawei BOD-WXX9

Tiezhu Yang (1):
LoongArch: Rework KASAN initialization for PTW-enabled systems

Tim Guttzeit (1):
ALSA: hda/realtek: Fix headset mic for TongFang X6AR55xU

Wenjie Qi (1):
f2fs: fix zoned block device information initialization

Xuewen Yan (1):
gpio: sprd: Change sprd_gpio lock to raw_spin_lock

Yongpeng Yang (2):
f2fs: fix out-of-bounds access in sysfs attribute read/write
f2fs: fix IS_CHECKPOINTED flag inconsistency issue caused by
concurrent atomic commit and checkpoint writes

Zhang Heng (1):
ALSA: hda/realtek: Add quirk for Inspur S14-G1

Zhiguo Niu (1):
f2fs: fix to add gc count stat in f2fs_gc_range

Ziyi Guo (1):
ASoC: fsl_xcvr: fix missing lock in fsl_xcvr_mode_put()

gongqi (1):
platform/x86/amd/pmc: Add quirk for MECHREVO Wujie 15X Pro

Makefile | 2 +-
arch/loongarch/include/asm/kasan.h | 11 +-
arch/loongarch/mm/kasan_init.c | 76 ++++++-----
drivers/bus/fsl-mc/fsl-mc-bus.c | 10 +-
drivers/gpio/gpio-sprd.c | 8 +-
drivers/gpio/gpiolib-acpi.c | 1 +
drivers/gpu/drm/tegra/hdmi.c | 4 +-
drivers/gpu/drm/tegra/sor.c | 4 +-
drivers/net/bareudp.c | 4 +-
drivers/net/geneve.c | 4 +-
drivers/net/wireguard/device.c | 1 -
drivers/pci/endpoint/pci-ep-cfs.c | 16 +--
drivers/platform/x86/amd/pmc/pmc-quirks.c | 7 ++
drivers/platform/x86/classmate-laptop.c | 32 +++++
drivers/platform/x86/panasonic-laptop.c | 4 +-
drivers/scsi/qla2xxx/qla_bsg.c | 28 +++--
drivers/usb/serial/option.c | 6 +
drivers/video/fbdev/riva/riva_hw.c | 3 +
drivers/video/fbdev/smscufx.c | 8 +-
fs/f2fs/data.c | 26 ++--
fs/f2fs/f2fs.h | 1 +
fs/f2fs/gc.c | 1 +
fs/f2fs/node.c | 14 ++-
fs/f2fs/super.c | 27 ++++
fs/f2fs/sysfs.c | 60 +++++++--
fs/romfs/super.c | 5 +-
include/asm-generic/tlb.h | 77 +++++++++++-
include/linux/hugetlb.h | 17 ++-
include/linux/mm_types.h | 6 +
include/net/ip_tunnels.h | 13 +-
kernel/cgroup/cpuset.c | 2 +-
mm/hugetlb.c | 146 ++++++++++++----------
mm/mmu_gather.c | 33 +++++
mm/rmap.c | 25 ++--
sound/pci/hda/patch_realtek.c | 5 +
sound/soc/amd/yc/acp6x-mach.c | 14 +++
sound/soc/codecs/cs35l45.c | 2 +-
sound/soc/codecs/cs42l43-jack.c | 37 +++++-
sound/soc/fsl/fsl_xcvr.c | 3 +
sound/soc/intel/boards/sof_es8336.c | 9 ++
40 files changed, 554 insertions(+), 198 deletions(-)

Summary by Sourcery

Update to Linux 6.6.127 and pull in upstream fixes across memory management, filesystems, device drivers, and platform quirks.

New Features:

• Add LoongArch write-combine (WC) DMW shadow mapping support in KASAN to cover WC segments.
• Expose and track the maximum number of open zones for zoned block devices in F2FS and use it to validate mount/remount options.

Bug Fixes:

• Prevent incorrect handling of shared hugetlb PMD tables and reduce unnecessary IPIs/TLB flushes during unshare, unmap, migration, and protection changes.
• Fix KASAN shadow mapping on LoongArch, including support for write-combine segments and correct early/late shadow handling.
• Resolve multiple F2FS issues including zoned block device max open zone validation, swapfile block mapping alignment, writeback completion ordering, sysfs attribute type safety, and GC statistics accounting.
• Avoid double completion and potential double free in qla2xxx BSG paths by only calling bsg_job_done() on success.
• Fix classmate-laptop and Panasonic laptop platform drivers by adding missing NULL checks and cleaning up sysfs groups on error paths.
• Prevent GPIO race conditions by switching sprd GPIO locking to raw spinlocks and correctly counting ACPI GPIOs with string references.
• Fix various fbdev drivers by avoiding divide-by-zero in Riva hardware arbitration and copying ioctl damage areas from userspace in smscufx.
• Correct WireGuard, BareUDP, and Geneve networking paths to use the new skb_vlan_inet_prepare() drop-reason API and return appropriate errors.
• Ensure ROMFS superblock initialization checks sb_set_blocksize() return value and fails cleanly on error.
• Protect fsl-mc driver_override sysfs access with device locking to avoid use-after-free and use sysfs_emit for sysfs output.
• Fix fsl_xcvr control updates by taking the controls_rwsem while enabling or disabling related controls.

Enhancements:

• Extend mmu_gather and hugetlb infrastructure with VMA-scoped gather helpers and tracking of unshared page tables to centralize TLB and GUP-fast synchronization.
• Improve cs42l43 and cs35l45 codec handling by distinguishing 3-pole vs 4-pole jack load detection, refining impedance thresholds, and correcting ASP_TX5 channel mapping.
• Add new DMI/USB quirks for AMD/Intel audio (ACPI yc boards and sof_es8336), Realtek HDA codecs, AMD PMC, and new Telit FN920C04/FN20C04 RNDIS compositions.
• Refine F2FS atomic fsync behavior to mark dentries for inodes during atomic writes and avoid redundant marking in non-atomic paths.

Build:

• Bump kernel sublevel from 6.6.126 to 6.6.127.

[sourcery-ai]

Reviewer's Guide

Rebases the Deepin 6.6-y kernel to upstream 6.6.127 and pulls in the associated upstream fixes, notably around hugetlb PMD sharing and TLB gathering, LoongArch KASAN write-combine support, multiple f2fs correctness and zoned-device fixes, network tunnel drop-reason handling, GPIO/raw locking, fbdev/ioctl safety, and assorted platform/sound/PCI/USB quirks and bugfixes.

Sequence diagram for huge_pmd_unshare with per-VMA mmu_gather and flush

sequenceDiagram
    participant Caller as move_hugetlb_page_tables
    participant TLB as mmu_gather
    participant VMA as vm_area_struct
    participant HUGEPMD as huge_pmd_unshare
    participant TLBCORE as tlb_unshare_pmd_ptdesc
    participant FLUSH as huge_pmd_unshare_flush

    Caller->>TLB: tlb_gather_mmu_vma(TLB, VMA)
    loop per_huge_pmd
        Caller->>HUGEPMD: huge_pmd_unshare(TLB, VMA, addr, ptep)
        alt PMD table is shared
            HUGEPMD->>TLBCORE: tlb_unshare_pmd_ptdesc(TLB, virt_to_ptdesc(ptep), addr)
            TLBCORE-->>HUGEPMD: updates unshared_tables and fully_unshared_tables
            HUGEPMD-->>Caller: return 1
            Caller->>TLB: tlb_remove_huge_tlb_entry(h, TLB, src_pte, addr)
        else PMD table not shared
            HUGEPMD-->>Caller: return 0
        end
    end

    Caller->>TLB: tlb_flush_mmu_tlbonly(TLB)
    Caller->>FLUSH: huge_pmd_unshare_flush(TLB, VMA)
    FLUSH->>TLB: tlb_flush_unshared_tables(TLB)
    FLUSH-->>Caller: done

    Caller->>TLB: tlb_finish_mmu(TLB)
    TLB-->>Caller: gather finished

Loading

Class diagram for updated hugetlb PMD sharing and mmu_gather integration

classDiagram
    class mmu_gather {
        +mm_struct *mm
        +unsigned int vma_huge
        +unsigned int vma_pfn
        +unsigned int unshared_tables
        +unsigned int fully_unshared_tables
        +unsigned int batch_count
        +void __tlb_reset_range()
        +void tlb_flush_mmu_tlbonly()
        +void tlb_flush_pmd_range(unsigned long addr, unsigned long size)
    }

    class ptdesc {
        +atomic_t pt_share_count
        +void ptdesc_pmd_pts_init()
        +void ptdesc_pmd_pts_inc()
        +void ptdesc_pmd_pts_dec()
        +int ptdesc_pmd_pts_count()
        +bool ptdesc_pmd_is_shared()
    }

    class vm_area_struct {
        +mm_struct *vm_mm
        +struct file *vm_file
    }

    class hugetlb_ops {
        +int huge_pmd_unshare(mmu_gather *tlb, vm_area_struct *vma, unsigned long addr, pte_t *ptep)
        +void huge_pmd_unshare_flush(mmu_gather *tlb, vm_area_struct *vma)
        +unsigned long hugetlb_mask_last_page(struct hstate *h)
    }

    class tlb_api {
        +void tlb_gather_mmu(mmu_gather *tlb, mm_struct *mm)
        +void tlb_gather_mmu_fullmm(mmu_gather *tlb, mm_struct *mm)
        +void tlb_gather_mmu_vma(mmu_gather *tlb, vm_area_struct *vma)
        +void tlb_finish_mmu(mmu_gather *tlb)
        +void tlb_unshare_pmd_ptdesc(mmu_gather *tlb, ptdesc *pt, unsigned long addr)
        +void tlb_flush_unshared_tables(mmu_gather *tlb)
    }

    class mm_hugetlb_c {
        +int copy_hugetlb_page_range(mm_struct *dst, mm_struct *src)
        +int move_hugetlb_page_tables(vm_area_struct *vma)
        +void __unmap_hugepage_range(mmu_gather *tlb, vm_area_struct *vma)
        +long hugetlb_change_protection(vm_area_struct *vma)
        +static void hugetlb_unshare_pmds(vm_area_struct *vma)
    }

    class rmap_c {
        +bool try_to_unmap_one(folio *folio, vm_area_struct *vma)
        +bool try_to_migrate_one(folio *folio, vm_area_struct *vma)
    }

    mmu_gather "1" o-- "*" ptdesc : manages
    vm_area_struct "1" --> "1" mm_struct : vm_mm
    hugetlb_ops ..> mmu_gather : uses
    hugetlb_ops ..> ptdesc : unshare
    tlb_api ..> mmu_gather : initializes
    tlb_api ..> vm_area_struct : vma based
    tlb_api ..> ptdesc : tlb_unshare_pmd_ptdesc

    mm_hugetlb_c ..> hugetlb_ops : calls huge_pmd_unshare
    mm_hugetlb_c ..> tlb_api : uses tlb_gather_mmu_vma tlb_finish_mmu

    rmap_c ..> hugetlb_ops : calls huge_pmd_unshare
    rmap_c ..> tlb_api : uses tlb_gather_mmu_vma tlb_finish_mmu

Loading

Class diagram for f2fs sysfs attributes and zoned device max_open_zones

classDiagram
    class f2fs_attr {
        +struct attribute attr
        +ssize_t (*show)(f2fs_attr *a, f2fs_sb_info *sbi, char *buf)
        +ssize_t (*store)(f2fs_attr *a, f2fs_sb_info *sbi, const char *buf, size_t len)
        +int struct_type
        +int offset
        +int size
        +int id
    }

    class f2fs_sb_info {
        +unsigned int blocks_per_blkz
        +unsigned int max_open_zones
        +atomic_t cp_call_count[ ]
        +wait_queue_head_t cp_wait
    }

    class f2fs_sysfs_c {
        +ssize_t f2fs_sbi_show(f2fs_attr *a, f2fs_sb_info *sbi, char *buf)
        +ssize_t f2fs_sbi_store(f2fs_attr *a, f2fs_sb_info *sbi, const char *buf, size_t len)
        +ssize_t __sbi_show_value(f2fs_attr *a, f2fs_sb_info *sbi, char *buf, unsigned char *value)
        +void __sbi_store_value(f2fs_attr *a, f2fs_sb_info *sbi, unsigned char *ui, unsigned long value)
    }

    class f2fs_super_c {
        +int f2fs_remount(super_block *sb, int *flags, char *data)
        +int f2fs_scan_devices(f2fs_sb_info *sbi)
        +int init_blkz_info(f2fs_sb_info *sbi, int devi)
    }

    class f2fs_data_c {
        +void f2fs_write_end_io(struct bio *bio)
        +int check_swap_activate(struct swap_info_struct *sis, struct file *file, sector_t *span)
    }

    class f2fs_node_c {
        +int __write_node_page(struct page *page, bool atomic, bool *submitted, int compr)
        +int f2fs_fsync_node_pages(f2fs_sb_info *sbi, struct inode *inode, struct writeback_control *wbc, bool atomic)
    }

    f2fs_attr --> f2fs_sb_info : targets struct via struct_type+offset
    f2fs_sysfs_c ..> f2fs_attr : uses size field
    f2fs_sysfs_c ..> f2fs_sb_info : reads and writes fields

    f2fs_super_c ..> f2fs_sb_info : initializes max_open_zones
    f2fs_super_c ..> f2fs_sb_info : validates max_open_zones vs active_logs

    f2fs_data_c ..> f2fs_sb_info : uses cp_wait and cp_call_count
    f2fs_node_c ..> f2fs_sb_info : uses options and inode flags

Loading

File-Level Changes

Change	Details	Files
Refine hugetlb PMD page-table sharing and unsharing, integrating it with mmu_gather and new unshare flush semantics.	Change copy_hugetlb_page_range() to detect shared PMDs via ptdesc_pmd_is_shared() under CONFIG_HUGETLB_PMD_PAGE_TABLE_SHARING instead of page_count() Refactor huge_pmd_unshare() to take an mmu_gather pointer, use tlb_unshare_pmd_ptdesc(), and stop calling tlb_remove_table_sync_one() directly Introduce huge_pmd_unshare_flush() to perform required TLB and GUP-fast synchronization for unshared tables before dropping i_mmap_rwsem Rework move_hugetlb_page_tables(), __unmap_hugepage_range(), hugetlb_change_protection(), hugetlb_unshare_pmds(), and rmap try_to_unmap/try_to_migrate paths to use a local mmu_gather, tlb_remove_huge_tlb_entry(), tlb_flush_mmu_tlbonly(), huge_pmd_unshare_flush(), and tlb_finish_mmu() Extend mmu_gather with unshared_tables/fully_unshared_tables flags and add tlb_unshare_pmd_ptdesc() and tlb_flush_unshared_tables() plus validation in tlb_finish_mmu()	mm/hugetlb.c include/linux/hugetlb.h include/asm-generic/tlb.h mm/mmu_gather.c mm/rmap.c include/linux/mm_types.h
Add tlb_gather_mmu_vma() and related hugetlb helpers to operate on a single VMA with correct page-size and caching semantics.	Implement tlb_gather_mmu_vma() to initialize an on-stack mmu_gather for a single VMA, including tlb_update_vma_flags() and hugepage page_size setup Ensure fully_unshared_tables is reset on gather, and warn in tlb_finish_mmu() if any remain due to missing huge_pmd_unshare_flush() Switch several hugetlb and rmap callers to use tlb_gather_mmu_vma() instead of open-coded mmu_gather initialization	mm/mmu_gather.c include/asm-generic/tlb.h include/linux/mm_types.h mm/hugetlb.c mm/rmap.c
Extend LoongArch KASAN to support write-combine mappings and cleanly separate early vs normal shadow address translation.	Add XKPRANGE_WC segment and its shadow range/offset definitions and adjust VMALLOC KASAN layout to include WC segment Introduce a mem_to_shadow() helper that handles CC, UC, WC, and VC segments and use it for mapping VMALLOC, KFENCE, linear, and module ranges Make kasan_mem_to_shadow() fall back to the early shadow page until kasan_arch_is_ready(), then delegate to mem_to_shadow(), and mark kasan_early_stage=false only at the end of kasan_init() Update kasan_shadow_to_mem() to reverse-map WC shadow addresses back to WC memory	arch/loongarch/include/asm/kasan.h arch/loongarch/mm/kasan_init.c
Harden f2fs behavior around sysfs attributes, writeback completion, swapfile mapping, GC statistics, and zoned-device resource limits.	Extend f2fs_attr with a size field and rework f2fs_sbi_show()/__sbi_store() to correctly handle 1/2/4/8-byte fields via typed helpers, using sizeof_field in attribute macros to avoid out-of-bounds access Move the F2FS_WB_CP_DATA wakeup logic into the per-page completion path in f2fs_write_end_io() to avoid use-after-free races with kill_f2fs_super() Improve swapfile activation alignment logic by tracking last_extent separately and always retrying mapping after GC-induced migration, ensuring correct physical block selection Track sbi->max_open_zones during device scan using bdev_max_open_zones(), enforce that it is at least active_logs, and refuse mount/remount on blkzoned devices when insufficient zones are available Account foreground GC calls in f2fs_gc_range() via stat_inc_gc_call_count() and adjust relevant superblock/ZBD initialization paths	fs/f2fs/sysfs.c fs/f2fs/data.c fs/f2fs/super.c fs/f2fs/gc.c fs/f2fs/node.c fs/f2fs/f2fs.h
Tighten romfs, GPIO, fbdev/ioctl, and cpuset core behavior for safety and correctness.	Check sb_set_blocksize() return value in romfs_fill_super() and fail the mount with an error message when setting blocksize fails Convert sprd_gpio lock from spinlock_t to raw_spinlock_t and adjust locking helpers, suitable for use in IRQ context Fix smscufx framebuffer UFX_IOCTL_REPORT_DAMAGE handling to allocate a dloarea on the kernel side, copy_from_user() the struct, and avoid relying on a user pointer cast Change cpuset’s partition_is_populated() to use cpuset_is_populated() instead of cgroup_is_populated() to respect cpuset-specific semantics	fs/romfs/super.c drivers/gpio/gpio-sprd.c drivers/video/fbdev/smscufx.c kernel/cgroup/cpuset.c
Improve networking tunnel/drop-reason handling and revert WireGuard threaded NAPI, aligning with upstream behavior.	Change skb_vlan_inet_prepare() to return an enum skb_drop_reason using pskb_may_pull_reason(), and update bareudp and geneve IPv4/IPv6 transmit paths to treat any non-SKB_NOT_DROPPED_YET return as an error Include <asm/tlb.h> instead of <asm/tlbflush.h> in rmap to use the new TLB helpers Revert dev_set_threaded(dev, true) for WireGuard devices in wg_newlink(), matching upstream’s decision to disable threaded NAPI for WireGuard	include/net/ip_tunnels.h drivers/net/bareudp.c drivers/net/geneve.c mm/rmap.c drivers/net/wireguard/device.c
Fix platform/sound/PCI/USB quirks and misc driver bugs across several subsystems.	Add NULL-pointer checks to classmate-laptop accelerometer sysfs show/store paths and open callback, returning -ENXIO when devices are not yet bound; fix panasonic-laptop probe error paths to remove sysfs group before returning Guard qla2xxx bsg paths by calling bsg_job_done() only on success to avoid double frees, and ensure error paths skip job completion Add DMI-based quirks for new AMD/Intel laptop models in acp6x-mach, sof_es8336, and amd/pmc (MECHREVO Wujie 15X Pro), plus Realtek HDA pin and codec quirks for specific TongFang and Inspur systems Fix cs42l43 jack load detection by programming different register sequences for 3-pole vs 4-pole jacks and refining impedance thresholds; correct cs35l45 ASP_TX5 DAPM widget channel index Protect fsl_xcvr_mode_put() control updates with card->snd_card->controls_rwsem, add Telit FN920C04/FN20C04 RNDIS USB compositions, and fix Tegra HDMI/SOR infoframe writers and rivafb’s nv3_arb() to avoid unused-variable and divide-by-zero issues Switch fsl-mc sysfs show functions to use sysfs_emit and protect driver_override_show() with device_lock(), and simplify PCI endpoint configfs groups by removing unused fields/work and using configfs_add_default_group() instead of configfs_register_group()	drivers/platform/x86/classmate-laptop.c drivers/platform/x86/panasonic-laptop.c drivers/scsi/qla2xxx/qla_bsg.c sound/soc/amd/yc/acp6x-mach.c sound/soc/intel/boards/sof_es8336.c drivers/platform/x86/amd/pmc/pmc-quirks.c sound/pci/hda/patch_realtek.c sound/soc/codecs/cs42l43-jack.c sound/soc/codecs/cs35l45.c sound/soc/fsl/fsl_xcvr.c drivers/usb/serial/option.c drivers/gpu/drm/tegra/hdmi.c drivers/gpu/drm/tegra/sor.c drivers/video/fbdev/riva/riva_hw.c drivers/bus/fsl-mc/fsl-mc-bus.c drivers/pci/endpoint/pci-ep-cfs.c
Update kernel version metadata to 6.6.127 to track the new upstream base.	Bump SUBLEVEL from 126 to 127 in top-level Makefile to reflect the new 6.6.127 release	Makefile

---

Tips and commands

Interacting with Sourcery

• Trigger a new review: Comment @sourcery-ai review on the pull request.
• Continue discussions: Reply directly to Sourcery's review comments.
• Generate a GitHub issue from a review comment: Ask Sourcery to create an
  issue from a review comment by replying to it. You can also reply to a
  review comment with @sourcery-ai issue to create an issue from it.
• Generate a pull request title: Write @sourcery-ai anywhere in the pull
  request title to generate a title at any time. You can also comment
  @sourcery-ai title on the pull request to (re-)generate the title at any time.
• Generate a pull request summary: Write @sourcery-ai summary anywhere in
  the pull request body to generate a PR summary at any time exactly where you
  want it. You can also comment @sourcery-ai summary on the pull request to
  (re-)generate the summary at any time.
• Generate reviewer's guide: Comment @sourcery-ai guide on the pull
  request to (re-)generate the reviewer's guide at any time.
• Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
  pull request to resolve all Sourcery comments. Useful if you've already
  addressed all the comments and don't want to see them anymore.
• Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
  request to dismiss all existing Sourcery reviews. Especially useful if you
  want to start fresh with a new review - don't forget to comment
  @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

• Enable or disable review features such as the Sourcery-generated pull request
  summary, the reviewer's guide, and others.
• Change the review language.
• Add, remove or edit custom review instructions.
• Adjust other review settings.

Getting Help

• Contact our support team for questions or feedback.
• Visit our documentation for detailed guides and information.
• Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

[deepin-ci-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please ask for approval from opsiff. For more information see the Code Review Process.

The full list of commands accepted by this bot can be found here.

Details

Needs approval from an approver in each of these files:

• deepin/OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[sourcery-ai]

Hey - I've reviewed your changes and they look great!

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}

[Copilot]

Pull request overview

Updates the Deepin kernel tree to upstream Linux 6.6.127, bringing in the corresponding stable fixes across memory management, filesystems, networking, and a broad set of device/platform drivers.

Changes:

• Bump kernel sublevel from 6.6.126 → 6.6.127 and merge upstream stable patches.
• Update MMU gather + hugetlb PMD unsharing/flush handling to reduce unnecessary IPIs and fix correctness issues.
• Pull in targeted fixes/quirks across F2FS, networking tunnels, audio codecs/boards, platform drivers, and assorted device drivers.

Reviewed changes

Copilot reviewed 40 out of 40 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
Makefile	Kernel sublevel bump to 6.6.127.
arch/loongarch/include/asm/kasan.h	Extend LoongArch KASAN shadow layout for WriteCombine range.
arch/loongarch/mm/kasan_init.c	Rework LoongArch KASAN shadow mapping/init sequencing.
drivers/bus/fsl-mc/fsl-mc-bus.c	sysfs_emit conversion + locking in driver_override_show().
drivers/gpio/gpio-sprd.c	Switch SPRD GPIO lock to raw_spinlock.
drivers/gpio/gpiolib-acpi.c	Fix ACPI GPIO counting to handle string references.
drivers/gpu/drm/tegra/hdmi.c	Fix unused variable in HDMI infoframe packing loop.
drivers/gpu/drm/tegra/sor.c	Fix unused variable in SOR HDMI infoframe packing loop.
drivers/net/bareudp.c	Adapt to skb_vlan_inet_prepare() drop-reason API.
drivers/net/geneve.c	Adapt to skb_vlan_inet_prepare() drop-reason API.
drivers/net/wireguard/device.c	Revert threaded-NAPI enabling in wg_newlink().
drivers/pci/endpoint/pci-ep-cfs.c	Avoid async subgroup creation; use configfs default groups.
drivers/platform/x86/amd/pmc/pmc-quirks.c	Add DMI quirk entry for MECHREVO Wujie 15X Pro.
drivers/platform/x86/classmate-laptop.c	Add missing NULL checks for driver data in sysfs ops.
drivers/platform/x86/panasonic-laptop.c	Fix sysfs group cleanup on error path.
drivers/scsi/qla2xxx/qla_bsg.c	Prevent double completion/free by gating bsg_job_done().
drivers/usb/serial/option.c	Add Telit FN20C04/FN920C04 RNDIS compositions.
drivers/video/fbdev/riva/riva_hw.c	Avoid divide-by-zero in nv3 arbitration calculations.
drivers/video/fbdev/smscufx.c	Copy ioctl damage area from userspace safely.
fs/f2fs/data.c	Fix F2FS writeback completion ordering + swapfile mapping behavior.
fs/f2fs/f2fs.h	Track max_open_zones in f2fs_sb_info for zoned devices.
fs/f2fs/gc.c	Account foreground GC range calls in stats.
fs/f2fs/node.c	Refine atomic fsync/dentry-marking behavior.
fs/f2fs/super.c	Zoned device max-open-zones validation on mount/remount; init max_open_zones.
fs/f2fs/sysfs.c	Make sysfs value show/store size-aware (u8/u16/u32/u64).
fs/romfs/super.c	Check sb_set_blocksize() return and fail cleanly.
include/asm-generic/tlb.h	Track/flush unshared page tables via mmu_gather flags/helpers.
include/linux/hugetlb.h	Update huge_pmd_unshare() API + add huge_pmd_unshare_flush().
include/linux/mm_types.h	Add ptdesc_pmd_is_shared() helper + tlb_gather_mmu_vma() decl.
include/net/ip_tunnels.h	Change skb_vlan_inet_prepare() to return skb_drop_reason.
kernel/cgroup/cpuset.c	Use cpuset_is_populated() to fix population detection.
mm/hugetlb.c	Rework hugetlb PMD sharing/unsharing to use mmu_gather and targeted flushes.
mm/mmu_gather.c	Add tlb_gather_mmu_vma() and track fully_unshared_tables.
mm/rmap.c	Use tlb_gather_mmu_vma() + huge_pmd_unshare_flush() for hugetlb unshare paths.
sound/pci/hda/patch_realtek.c	Add new Realtek HDA quirks/pin quirks.
sound/soc/amd/yc/acp6x-mach.c	Add new DMI quirks for AMD YC platforms.
sound/soc/codecs/cs35l45.c	Fix ASP_TX5 DAPM widget channel index.
sound/soc/codecs/cs42l43-jack.c	Refine load-detect patches for 3/4 pole and impedance classification.
sound/soc/fsl/fsl_xcvr.c	Add locking around control activation during mode changes.
sound/soc/intel/boards/sof_es8336.c	Add Huawei BOD-WXX9 DMI quirk.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

__sbi_store_value() takes the value as unsigned long, but attributes can now be size==8 (u64). On 32-bit kernels this truncates and also pairs poorly with kstrtoul() parsing, so writing u64 sysfs fields (e.g., max_io_bytes) won't work correctly. Use u64 for the parsed/stored value (kstrtoull()) when a->size==8 and pass a u64 into __sbi_store_value().

[Copilot]

fsl_xcvr_mode_put() is invoked from ALSA control write paths with card->controls_rwsem already held in write mode (see sound/core/control.c:snd_ctl_elem_write()). Taking down_read() on the same controls_rwsem here can deadlock. The extra locking is also unnecessary for fsl_xcvr_activate_ctl(), which already asserts the lock is held. Drop the down_read()/up_read() here (or replace with a lockdep assertion if needed).

[Copilot]

Early-returning when state->mclk_khz == 0 leaves output state inconsistent: nv3_arb() returns before initializing ainfo->converged and before setting any safe defaults in res_info, but callers (e.g., nv3_get_param()/nv3CalcArbitration) read those fields. Initialize ainfo/res_info to known failure defaults before returning (or fall back to the existing non-converged default path).

Suggested change

	if (!state->mclk_khz)
	return (0);
	if (!state->mclk_khz) {
	/*
	* If the memory clock is zero, we cannot perform a meaningful
	* arbitration calculation. Ensure that the output structures
	* are left in a consistent "failure" state so that callers do
	* not observe uninitialized data.
	*/
	if (res_info)
	memset(res_info, 0, sizeof(*res_info));
	if (ainfo) {
	ainfo->by_gfacc = 0;
	ainfo->wcmocc = 0;
	ainfo->wcgocc = 0;
	ainfo->wcvocc = 0;
	ainfo->wcvlwm = 0;
	ainfo->wcglwm = 0;
	ainfo->mocc = 0;
	ainfo->vocc = 0;
	ainfo->gocc = 0;
	ainfo->engine_en = 0;
	ainfo->converged = 0;
	ainfo->cur = ENGINE;
	ainfo->first_vacc = 0;
	ainfo->first_gacc = 0;
	ainfo->first_macc = 0;
	}
	return 0;
	}