[continued from previous message]   
      
        - bpf: Check netfilter ctx accesses are aligned   
        - apparmor: ensure WB_HISTORY_SIZE value is a power of 2   
        - apparmor: fix loop detection used in conflicting attachment resolution   
        - apparmor: Fix unaligned memory accesses in KUnit test   
        - module: Restore the moduleparam prefix length check   
        - ucount: fix atomic_long_inc_below() argument type   
        - rtc: ds1307: fix incorrect maximum clock rate handling   
        - rtc: hym8563: fix incorrect maximum clock rate handling   
        - rtc: nct3018y: fix incorrect maximum clock rate handling   
        - rtc: pcf85063: fix incorrect maximum clock rate handling   
        - rtc: pcf8563: fix incorrect maximum clock rate handling   
        - rtc: rv3028: fix incorrect maximum clock rate handling   
        - f2fs: turn off one_time when forcibly set to foreground GC   
        - f2fs: fix bio memleak when committing super block   
        - f2fs: fix KMSAN uninit-value in extent_info usage   
        - f2fs: fix to check upper boundary for value of gc_boost_zoned_gc_percent   
        - f2fs: fix to check upper boundary for gc_valid_thresh_ratio   
        - f2fs: fix to check upper boundary for gc_no_zoned_gc_percent   
        - f2fs: doc: fix wrong quota mount option description   
        - f2fs: fix to avoid UAF in f2fs_sync_inode_meta()   
        - f2fs: fix to avoid panic in f2fs_evict_inode   
        - f2fs: fix to avoid out-of-boundary access in devs.path   
        - f2fs: vm_unmap_ram() may be called from an invalid context   
        - f2fs: fix to update upper_p in __get_secs_required() correctly   
        - f2fs: fix to calculate dirty data during has_not_enough_free_secs()   
        - f2fs: fix to trigger foreground gc during f2fs_map_blocks() in lfs mode   
        - exfat: fdatasync flag should be same like generic_write_sync()   
        - i2c: muxes: mule: Fix an error handling path in mule_i2c_mux_probe()   
        - vfio: Fix unbalanced vfio_df_close call in no-iommu mode   
        - vfio: Prevent open_count decrement to negative   
        - vfio/pds: Fix missing detach_ioas op   
        - vfio/pci: Separate SR-IOV VF dev_set   
        - scsi: mpt3sas: Fix a fw_event memory leak   
        - scsi: Revert "scsi: iscsi: Fix HW conn removal use after free"   
        - scsi: ufs: core: Use link recovery when h8 exit fails during runtime   
          resume   
        - scsi: sd: Make sd shutdown issue START STOP UNIT appropriately   
        - kconfig: qconf: fix ConfigList::updateListAllforAll()   
        - sched/psi: Fix psi_seq initialization   
        - PCI: pnv_php: Clean up allocated IRQs on unplug   
        - PCI: pnv_php: Work around switches with broken presence detection   
        - [powerpc*] eeh: Export eeh_unfreeze_pe()   
        - [powerpc*] eeh: Make EEH driver device hotplug safe   
        - PCI: pnv_php: Fix surprise plug detection and recovery   
        - pNFS/flexfiles: don't attempt pnfs on fatal DS errors   
        - sched: Add test_and_clear_wake_up_bit() and atomic_dec_and_wake_up()   
        - NFS: Fix wakeup of __nfs_lookup_revalidate() in unblock_revalidate()   
        - NFS: Fix filehandle bounds checking in nfs_fh_to_dentry()   
        - NFSv4.2: another fix for listxattr   
        - NFS: Fixup allocation flags for nfsiod's __GFP_NORETRY   
        - md/md-cluster: handle REMOVE message earlier   
        - netpoll: prevent hanging NAPI when netcons gets enabled   
        - phy: mscc: Fix parsing of unicast frames   
        - net: ipa: add IPA v5.1 and v5.5 to ipa_version_string()   
        - pptp: ensure minimal skb length in pptp_xmit()   
        - nvmet: initialize discovery subsys after debugfs is initialized   
        - [s390x] ap: Unmask SLCF bit in card and queue ap functions sysfs   
        - netlink: specs: ethtool: fix module EEPROM input/output arguments   
        - block: Fix default IO priority if there is no IO context   
        - block: ensure discard_granularity is zero when discard is not supported   
        - ASoC: tas2781: Fix the wrong step for TLV on tas2781   
        - [amd64] spi: cs42l43: Property entry should be a null-terminated array   
        - net/mlx5: Correctly set gso_segs when LRO is used   
        - ipv6: reject malicious packets in ipv6_gso_segment()   
        - net: mdio: mdio-bcm-unimac: Correct rate fallback logic   
        - net: drop UFO packets in udp_rcv_segment()   
        - net/sched: taprio: enforce minimum value for picos_per_byte   
        - sunrpc: fix client side handling of tls alerts   
        - [x86] irq: Plug vector setup race   
        - benet: fix BUG when creating VFs   
        - net/sched: mqprio: fix stack out-of-bounds write in tc entry parsing   
        - [s390x] mm: Allocate page table with PAGE_SIZE granularity   
        - eth: fbnic: remove the debugging trick of super high page bias   
        - irqchip: Build IMX_MU_MSI only on ARM   
        - ALSA: hda/ca0132: Fix missing error handling in ca0132_alt_select_out()   
        - smb: server: remove separate empty_recvmsg_queue   
        - smb: server: make sure we call ib_dma_unmap_single() only if we called   
          ib_dma_map_single already   
        - smb: server: let recv_done() consistently call   
          put_recvmsg/smb_direct_disconnect_rdma_connection   
        - smb: server: let recv_done() avoid touching data_transfer after   
          cleanup/move   
        - smb: client: remove separate empty_packet_queue   
        - smb: client: make sure we call ib_dma_unmap_single() only if we called   
          ib_dma_map_single already   
        - smb: client: let recv_done() cleanup before notifying the callers.   
        - smb: client: let recv_done() avoid touching data_transfer after   
          cleanup/move   
        - nvmet: exit debugfs after discovery subsystem exits   
        - pptp: fix pptp_xmit() error path   
        - smb: client: return an error if rdma_connect does not return within 5   
          seconds   
        - sunrpc: fix handling of server side tls alerts   
        - perf/core: Don't leak AUX buffer refcount on allocation failure   
        - perf/core: Exit early on perf_mmap() fail   
        - perf/core: Prevent VMA split of buffer mappings   
        - selftests/perf_events: Add a mmap() correctness test   
        - net/packet: fix a race in packet_set_ring() and packet_notifier()   
        - vsock: Do not allow binding to VMADDR_PORT_ANY   
        - [amd64] accel/ivpu: Fix reset_engine debugfs file logic   
        - Revert "bcache: remove heap-related macros and switch to generic   
   min_heap"   
        - ice/ptp: fix crosstimestamp reporting   
        - [amd64] drm/i915/ddi: change intel_ddi_init_{dp, hdmi}_connector()   
   return   
          type   
        - [amd64] drm/i915/hdmi: propagate errors from intel_hdmi_init_connector()   
        - [amd64] drm/i915/hdmi: add error handling in g4x_hdmi_init()   
        - [amd64] drm/i915/ddi: gracefully handle errors from   
          intel_ddi_init_hdmi_connector()   
        - [amd64] drm/i915/display: add intel_encoder_is_hdmi()   
        - [amd64] drm/i915/ddi: only call shutdown hooks for valid encoders   
        - ksmbd: fix null pointer dereference error in generate_encryptionkey   
        - ksmbd: fix Preauh_HashValue race condition   
        - ksmbd: fix corrupted mtime and ctime in smb2_open   
        - ksmbd: limit repeated connections from clients with the same IP   
          (CVE-2025-38501)   
        - smb: server: Fix extension string in ksmbd_extract_shortname()   
        - USB: serial: option: add Foxconn T99W709   
        - PCI/ASPM: Save parent L1SS config in pci_save_aspm_l1ss_state()   
        - PCI/ASPM: Fix L1SS saving   
        - Bluetooth: btusb: Add USB ID 3625:010b for TP-LINK Archer TX10UB Nano   
        - net: usbnet: Avoid potential RCU stall on LINK_CHANGE event   
        - net: usbnet: Fix the wrong netif_carrier_on() call   
        - [x86] sev: Evict cache lines during SNP memory validation   
   (CVE-2024-36331)   
      
   [continued in next message]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)