kernel-tools-debuginfo-4.19.90-2109.1.0.0108.oe1 >  A a2DW^^f|gP4 (?OgHⴊ@aW2u3OL;yV X.@oɐ`clzzf.aLep;G;f? !zȊ'_ i-#,خsK}?Θ VmT#I"W?ti?#+lHs;9uؓ- ׋+rQܭUa눜7 @c)|*[oZۡTe}jMNϏ-h[ 5#18ee1040d67fb1dc4d04068b05d41f463cd00d2dd4af89d3b1f8beb616dcbca3b77955784d5afb2598b4e71a7f0912b162c3353eya2DW^^fJ'YJ1sƕӳ~Q ZѢe10ĥgGl2$c]9Vs.6GPQ%Ɯ BܝPS,*^IW[ u[orM଑Cd5"{C0 < Xee'~Y+b!EYB>p<?d! 3 ^$(Shn   4 p ? P"L3(O8X,9,:,F+GHHIXY\ ]H^blcWdöeûfþltuvT zex|˂Ckernel-tools-debuginfo4.19.902109.1.0.0108.oe1Debug information for package kernel-toolsThis package provides debug information for package kernel-tools. Debug information is useful when developing applications that use this package or when debugging this package.a2ƫobs-worker-0030| openEuler:20.03:LTS:SP2 / standard_aarch64http://openeuler.orgGPLv2http://openeuler.orgDevelopment/Debughttp://www.kernel.org/linuxaarch64b`glfi4 AAA$$$$$$$$A$a2ˌa2ˌ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-4.19.90-2109.1.0.0108.oe1.src.rpmkernel-tools-debuginfokernel-tools-debuginfo(aarch-64)    rpmlib(CompressedFileNames)rpmlib(FileDigests)rpmlib(PayloadFilesHavePrefix)rpmlib(PayloadIsXz)3.0.4-14.6.0-14.0-15.2-14.15.1a/k@a,@a$@a#aaa^aj@aj@a@`@```9@``Ȗ@`g@``!@`!@`8@`@`@`q`}@`Y`e@`e@`e@`x*`x*`x*`x*`x*`x*`x*`x*`x*`x*`x*`x*`x*`u`eCheng Jian - 4.19.90-2109.1.0.0108Cheng Jian - 4.19.90-2108.9.0.0107Cheng Jian - 4.19.90-2108.8.0.0106Cheng Jian - 4.19.90-2108.7.0.0105Cheng Jian - 4.19.90-2108.5.0.0104Cheng Jian - 4.19.90-2108.5.0.0103Cheng Jian - 4.19.90-2108.4.0.0102Gou Hao -4.19.90-2108.4.0.0101Cheng Jian - 4.19.90-2108.4.0.0100Cheng Jian - 4.19.90-2108.1.0.0099Cheng Jian - 4.19.90-2107.5.0.0098Cheng Jian - 4.19.90-2107.4.0.0097Senlin Xia - 4.19.90-2106.3.0.0096Cheng Jian - 4.19.90-2106.3.0.0095Cheng Jian - 4.19.90-2106.2.0.0094Cheng Jian - 4.19.90-2106.1.0.0093Cheng Jian - 4.19.90-2105.9.0.0092Cheng Jian - 4.19.90-2105.8.0.0091Cheng Jian - 4.19.90-2105.6.0.0090Cheng Jian - 4.19.90-2105.5.0.0089Cheng Jian - 4.19.90-2105.4.0.0088Cheng Jian - 4.19.90-2105.3.0.0087Cheng Jian - 4.19.90-2105.2.0.0086Cheng Jian - 4.19.90-2104.26.0.0085Cheng Jian - 4.19.90-2104.26.0.0084Cheng Jian - 4.19.90-2104.21.0.0083Cheng Jian - 4.19.90-2104.19.0.0082Cheng Jian - 4.19.90-2104.18.0.0081Cheng Jian - 4.19.90-2104.17.0.0080Cheng Jian - 4.19.90-2104.16.0.0079Cheng Jian - 4.19.90-2104.15.0.0078Cheng Jian - 4.19.90-2104.14.0.0077Cheng Jian - 4.19.90-2104.13.0.0076Cheng Jian - 4.19.90-2104.12.0.0075Cheng Jian - 4.19.90-2104.11.0.0074Cheng Jian - 4.19.90-2104.10.0.0073Cheng Jian - 4.19.90-2104.9.0.0072Cheng Jian - 4.19.90-2104.8.0.0071Cheng Jian - 4.19.90-2104.7.0.0070Cheng Jian - 4.19.90-2104.6.0.0069Cheng Jian - 4.19.90-2104.5.0.0068Cheng Jian - 4.19.90-2104.4.0.0067Cheng Jian - 4.19.90-2104.3.0.0066Jiachen Fan - 4.19.90-2104.2.0.0065- iommu: smmuv2: Using the SMMU_BYPASS_DEV to bypass SMMU for some SoCs - iommu: dev_bypass: cleanup dev bypass code - arm64: phytium: using MIDR_PHYTIUM_FT2000PLUS instead of ARM_CPU_IMP_PHYTIUM - arm64: Add MIDR encoding for PHYTIUM CPUs - arm64: Add MIDR encoding for HiSilicon Taishan CPUs - sched: Fix sched_fork() access an invalid sched_task_group - KVM: nSVM: avoid picking up unsupported bits from L2 in int_ctl (CVE-2021-3653) - KVM: nSVM: always intercept VMLOAD/VMSAVE when nested (CVE-2021-3656) - Bluetooth: switch to lock_sock in SCO - Bluetooth: avoid circular locks in sco_sock_connect - Bluetooth: schedule SCO timeouts with delayed_work - Bluetooth: defer cleanup of resources in hci_unregister_dev()- tcp_comp: Fix comp_read_size return value - virtio-blk: Add validation for block size in config space - blk-mq: fix divide by zero crash in tg_may_dispatch() - mm, vmscan: guarantee drop_slab_node() termination - jump_label: skip resource release if jump label is not relocated - ext4: prevent getting empty inode buffer - ext4: move ext4_fill_raw_inode() related functions before __ext4_get_inode_loc() - ext4: factor out ext4_fill_raw_inode() - ext4: make the updating inode data procedure atomic - KVM: X86: MMU: Use the correct inherited permissions to get shadow page - x86/config: Enable CONFIG_USERSWAP for openeuler_defconfig - ext4: fix panic when mount failed with parallel flush_stashed_error_work - device core: Consolidate locking and unlocking of parent and device - Revert "ext4: flush s_error_work before journal destroy in ext4_fill_super" - ext2: Strengthen xattr block checks - ext2: Merge loops in ext2_xattr_set() - ext2: introduce helper for xattr entry validation - mm: rmap: explicitly reset vma->anon_vma in unlink_anon_vmas()- bpf: Fix leakage due to insufficient speculative store bypass mitigation - bpf: Introduce BPF nospec instruction for mitigating Spectre v4 - bpf: track spill/fill of constants - bpf/verifier: per-register parent pointers - blk-mq: clear active_queues before clearing BLK_MQ_F_TAG_QUEUE_SHARED- scsi: hisi_sas: Flush workqueue in hisi_sas_v3_remove() - nvme: force complete cancelled requests - blk-mq: blk-mq: provide forced completion method - ext4: flush s_error_work before journal destroy in ext4_fill_super - Revert "net: make get_net_ns return error if NET_NS is disabled" - kthread: Fix PF_KTHREAD vs to_kthread() race - sched/debug: Fix 'sched_debug_lock' undeclared error - Remove MODULE_ALIAS() calls that take undefined macro - scripts/dtc: Remove redundant YYLOC global declaration - x86/boot/compressed: Don't declare __force_order in kaslr_64.c - usb: hso: fix error handling code of hso_create_net_device - hso: fix bailout in error case of probe- spec: fixed the mistake for dates in kernel.spec- iscsi: Fix KABI change for 'Fix conn use after free during resets' - iscsi: alloc iscsi_cls_session with iscsi_cls_session_warpper - iscsi: introduce iscsi_cls_session_warapper and helper - scsi: iscsi: Fix conn use after free during resets - scsi: sr: Return correct event when media event code is 3 - net: xilinx_emaclite: Do not print real IOMEM pointer - sctp: move the active_key update after sh_keys is added - usb: max-3421: Prevent corruption of freed memory - net: ll_temac: Fix bug causing buffer descriptor overrun - tcp_comp: Avoiding the null pointer problem of ctx in comp_stream_read - nbd: add the check to prevent overflow in __nbd_ioctl() - ext4: fix potential uninitialized access to retval in kmmpd - blk-mq: fix kabi broken by "blk-mq: fix hang caused by freeze/unfreeze sequence" - blk-mq: fix hang caused by freeze/unfreeze sequence - config: Enable CONFIG_UCE_KERNEL_RECOVERY by default - EDAC/ghes: Remove intermediate buffer pvt->detail_location - USB: fix some clerical mistakes - uce: pagecache reading scenario support kernel recovery - uce: cow scenario support kernel recovery - selinux: fix NULL dereference in policydb_destroy() - livepatch/x86: Ignore return code of save_stack_trace_tsk_reliable() - mm,hwpoison: return -EHWPOISON to denote that the page has already been poisoned - mm/memory-failure: use a mutex to avoid memory_failure() races - arm64: mm: account for hotplug memory when randomizing the linear region- test modules directory existed when ls- fix rpmbuild error with patches- openeuler_defconfig: Enable ARCH_PHYTIUM and ARM_GIC_PHYTIUM_2500 - config: Enable Phytium FT-2500 support configs for hulk_defconfig - irqchip: phytium-2500: Add interrupt controller driver - mm/vmscan: setup drop_caches_loop_limit in cmdline - mm/memcg: optimize memory.numa_stat like memory.stat - livepatch: Fix crash when access the global variable in hook - timer: Use hlist_unhashed_lockless() in timer_pending() - list: Add hlist_unhashed_lockless() - config: Enable CONFIG_GPIO_HISI by default - gpio: gpio-hisi: Add HiSilicon GPIO support - config: Enable CONFIG_I2C_HISI by default - i2c: add support for HiSilicon I2C controller - i2c: core: add api to provide frequency mode strings - i2c: core: add managed function for adding i2c adapters - blk: reuse lookup_sem to serialize partition operations - Revert "block: take bd_mutex around delete_partitions in del_gendisk" - Revert "block: avoid creating invalid symlink file for patitions" - Revert "block: call bdput() to avoid memleak" - sctp: fix return value check in __sctp_rcv_asconf_lookup - workqueue: fix UAF in pwq_unbound_release_workfn() - exit: Move preemption fixup up, move blocking operations down - Input: joydev - prevent use of not validated data in JSIOCSBTNMAP ioctl - Input: joydev - prevent potential read overflow in ioctl - srcu: Take early exit on memory-allocation failure - Revert "modpost: add read_text_file() and get_line() helpers" - Revert "modpost: use read_text_file() and get_line() for reading text files" - Revert "modpost: remove use of non-standard strsep() in HOSTCC code" - Revert "modpost: explain why we can't use strsep" - cpuidle: fix return type err in haltpoll_switch_governor - mm/slab: add naive detection of double free - mm/mempool: fix a data race in mempool_free() - mm/list_lru: fix a data race in list_lru_count_one - mm/cma.c: fix NULL pointer dereference when cma could not be activated - iommu/amd: Prevent NULL pointer dereference- bcache: always record start time of a sample - bcache: do not collect data insert info created by write_moving - tcp_comp: open configs for tcp compression - tcp_comp: implement recvmsg for tcp compression - tcp_comp: implement sendmsg for tcp compression - tcp_comp: add stub proto ops for tcp compression socket - tcp_comp: allow ignore local tcp connections - tcp_comp: only enable compression for give server ports - tcp_comp: add sysctl for enable/disable compression - tcp_comp: add init and cleanup hook for compression - tcp_comp: add tcp comp option to SYN and SYN-ACK - tcp_comp: add Kconfig for tcp payload compression - tracing: Fix bug in rb_per_cpu_empty() that might cause deadloop. - proc: Avoid mixing integer types in mem_rw() - net: sched: cls_api: Fix the the wrong parameter - sctp: update active_key for asoc when old key is being replaced - nvme-pci: don't WARN_ON in nvme_reset_work if ctrl.state is not RESETTING - net/sched: act_skbmod: Skip non-Ethernet packets - net/tcp_fastopen: fix data races around tfo_active_disable_stamp - scsi: target: Fix protect handling in WRITE SAME(32) - scsi: iscsi: Fix iface sysfs attr detection - nvme-pci: do not call nvme_dev_remove_admin from nvme_remove - ipv6: fix 'disable_policy' for fwd packets - net: ip_tunnel: fix mtu calculation for ETHER tunnel devices - udp: annotate data races around unix_sk(sk)->gso_size - ipv6: tcp: drop silly ICMPv6 packet too big messages - tcp: annotate data races around tp->mtu_info - dma-buf/sync_file: Don't leak fences on merge failure - net: validate lwtstate->data before returning from skb_tunnel_info() - net: send SYNACK packet with accepted fwmark - net: bridge: sync fdb to new unicast-filtering ports - netfilter: ctnetlink: suspicious RCU usage in ctnetlink_dump_helpinfo - dm writecache: fix writing beyond end of underlying device when shrinking - dm writecache: return the exact table values that were set - dm multipath: use updated MPATHF_QUEUE_IO on mapping for bio-based mpath - dm writecache: fix data corruption when reloading the target - dm verity fec: fix hash block number in verity_fec_decode - sched/fair: Fix CFS bandwidth hrtimer expiry type - scsi: libfc: Fix array index out of bound exception - scsi: scsi_dh_alua: Fix signedness bug in alua_rtpg() - net: bridge: multicast: fix PIM hello router port marking race - NFSv4/pNFS: Don't call _nfs4_pnfs_v3_ds_connect multiple times - virtio_net: move tx vq operation under tx queue lock - x86/fpu: Limit xstate copy size in xstateregs_set() - nfs: fix acl memory leak of posix_acl_create() - NFSv4: Initialise connection to the server in nfs4_alloc_client() - PCI/sysfs: Fix dsm_label_utf16s_to_utf8s() buffer overrun - virtio_console: Assure used length from device is limited - virtio_net: Fix error handling in virtnet_restore() - virtio-blk: Fix memory leak among suspend/resume procedure - NFS: nfs_find_open_context() may only select open files - lib/decompress_unlz4.c: correctly handle zero-padding around initrds. - i2c: core: Disable client irq on reboot/shutdown - scsi: qedi: Fix null ref during abort handling - scsi: iscsi: Fix shost->max_id use - scsi: iscsi: Add iscsi_cls_conn refcount helpers - scsi: scsi_dh_alua: Check for negative result value - tracing: Do not reference char * as a string in histograms - scsi: core: Fix bad pointer dereference when ehandler kthread is invalid - seq_buf: Fix overflow in seq_buf_putmem_hex() - ipmi/watchdog: Stop watchdog timer when the current action is 'none' - net: ip: avoid OOM kills with large UDP sends over loopback - vsock: notify server to shutdown when client has pending signal - xfrm: Fix error reporting in xfrm_state_construct. - virtio_net: Remove BUG() to avoid machine dead - dm space maps: don't reset space map allocation cursor when committing - ipv6: use prandom_u32() for ID generation - mm/huge_memory.c: don't discard hugepage if other processes are mapping it - vfio/pci: Handle concurrent vma faults - vfio-pci: Use io_remap_pfn_range() for PCI IO memory - writeback: fix obtain a reference to a freeing memcg css - ipv6: fix out-of-bound access in ip6_parse_tlv() - bpf: Do not change gso_size during bpf_skb_change_proto() - ipv6: exthdrs: do not blindly use init_net - net/ipv4: swap flow ports when validating source - vxlan: add missing rcu_read_lock() in neigh_reduce() - pkt_sched: sch_qfq: fix qfq_change_class() error path - netfilter: nft_tproxy: restrict support to TCP and UDP transport protocols - netfilter: nft_osf: check for TCP packet before further processing - netfilter: nft_exthdr: check for IPv6 packet before further processing - netlabel: Fix memory leak in netlbl_mgmt_add_common - ACPI: sysfs: Fix a buffer overrun problem with description_show() - evm: fix writing /evm overflow - lib: vsprintf: Fix handling of number field widths in vsscanf - ACPI: processor idle: Fix up C-state latency if not ordered - fuse: check connected before queueing on fpq->io - evm: Refuse EVM_ALLOW_METADATA_WRITES only if an HMAC key is loaded - evm: Execute evm_inode_init_security() only when an HMAC key is loaded - seq_buf: Make trace_seq_putmem_hex() support data longer than 8 - ext4: use ext4_grp_locked_error in mb_find_extent - ext4: fix avefreec in find_group_orlov - ext4: remove check for zero nr_to_scan in ext4_es_scan() - ext4: correct the cache_nr in tracepoint ext4_es_shrink_exit - ext4: return error code when ext4_fill_flex_info() fails - ext4: fix kernel infoleak via ext4_extent_header - iov_iter_fault_in_readable() should do nothing in xarray case - scsi: core: Retry I/O for Notify (Enable Spinup) Required error - kthread: prevent deadlock when kthread_mod_delayed_work() races with kthread_cancel_delayed_work_sync() - kthread_worker: split code for canceling the delayed work timer - scsi: sr: Return appropriate error code when disk is ejected - mm, futex: fix shared futex pgoff on shmem huge page - mm/thp: another PVMW_SYNC fix in page_vma_mapped_walk() - mm/thp: fix page_vma_mapped_walk() if THP mapped by ptes - mm: page_vma_mapped_walk(): get vma_address_end() earlier - mm: page_vma_mapped_walk(): use goto instead of while (1) - mm: page_vma_mapped_walk(): add a level of indentation - mm: page_vma_mapped_walk(): crossing page table boundary - mm: page_vma_mapped_walk(): prettify PVMW_MIGRATION block - mm: page_vma_mapped_walk(): use pmde for *pvmw->pmd - mm: page_vma_mapped_walk(): settle PageHuge on entry - mm: page_vma_mapped_walk(): use page for pvmw->page - mm: thp: replace DEBUG_VM BUG with VM_WARN when unmap fails for split - mm/thp: unmap_mapping_page() to fix THP truncate_cleanup_page() - mm/thp: fix page_address_in_vma() on file THP tails - mm/thp: fix vma_address() if virtual address below file offset - mm/thp: try_to_unmap() use TTU_SYNC for safe splitting - mm/thp: make is_huge_zero_pmd() safe and quicker - mm/thp: fix __split_huge_pmd_locked() on shmem migration entry - mm/rmap: use page_not_mapped in try_to_unmap() - mm/rmap: remove unneeded semicolon in page_not_mapped() - mm: add VM_WARN_ON_ONCE_PAGE() macro - sctp: add param size validation for SCTP_PARAM_SET_PRIMARY - sctp: validate chunk size in __rcv_asconf_lookup - stop_machine: Avoid potential race behaviour - KVM: PPC: Book3S: Fix H_RTAS rets buffer overflow - can: raw: fix raw_rcv panic for sock UAF - mm/page_isolation: do not isolate the max order page - mm/zswap: fix passing zero to 'PTR_ERR' warning - mm/page_alloc: speed up the iteration of max_order - mm: hugetlb: fix type of delta parameter and related local variables in gather_surplus_pages() - mm/vmalloc.c:__vmalloc_area_node(): avoid 32-bit overflow - sctp: add size validation when walking chunks - sctp: validate from_addr_param return - jbd2: fix kabi broken in struct journal_s - ext4: inline jbd2_journal_[un]register_shrinker() - jbd2: export jbd2_journal_[un]register_shrinker() - fs: remove bdev_try_to_free_page callback - ext4: remove bdev_try_to_free_page() callback - jbd2: simplify journal_clean_one_cp_list() - jbd2,ext4: add a shrinker to release checkpointed buffers - jbd2: remove redundant buffer io error checks - jbd2: don't abort the journal when freeing buffers - jbd2: ensure abort the journal if detect IO error when writing original buffer back - jbd2: remove the out label in __jbd2_journal_remove_checkpoint() - mm: vmscan: use a new flag to indicate shrinker is registered - Revert "jbd2: remove the out label in __jbd2_journal_remove_checkpoint()" - Revert "jbd2: ensure abort the journal if detect IO error when writing original buffer back" - Revert "jbd2: fix kabi broken in struct journal_s" - Revert "jbd2: don't abort the journal when freeing buffers" - mm/vmscan: add drop_caches_loop_limit to break loop in drop_slab_node - mm/vmscan: fix infinite loop in drop_slab_node - userswap: add a kernel parameter to enable userswap - userfaultfd: fix BUG_ON() in userfaultfd_release() - kprobes: Warn if the kprobe is reregistered - Revert "kretprobe: check re-registration of the same kretprobe earlier"- mm: vmalloc: prevent use after free in _vm_unmap_aliases - PCI/sysfs: Take reference on device to be removed - seq_file: disallow extremely large seq buffer allocations - ARM: footbridge: remove personal server platform - mm: slab: fix kmem_cache_create failed when sysfs node not destroyed - ARM: ensure the signal page contains defined contents - nvme-pci: use atomic bitops to mark a queue enabled - nvme: check the PRINFO bit before deciding the host buffer length - nvme: fix compat address handling in several ioctls - nvme-core: make implicit seed truncation explicit - nvme-core: don't use NVME_NSID_ALL for command effects and supported log - nvme-pci: fix NULL req in completion handler - nvme-pci: cancel nvme device request before disabling - nvme: copy MTFA field from identify controller - nvme-pci: Unblock reset_work on IO failure - nvme-pci: Don't disable on timeout in reset state - nvme-pci: Fix controller freeze wait disabling - block: error out if blk_get_queue() failed in blk_init_rl()- Revert "smp: Fix smp_call_function_single_async prototype" - Revert "vt: Fix character height handling with VT_RESIZEX" - block: only call sched requeue_request() for scheduled requests - KVM: arm: replace WARN_ON with pr_warn for UNKNOWN type - net/packet: annotate accesses to po->ifindex - net/packet: annotate accesses to po->bind - inet: annotate date races around sk->sk_txhash - ping: Check return value of function 'ping_queue_rcv_skb' - net: ethtool: clear heap allocations for ethtool function - x86/fpu: Reset state for all signal restore failures - inet: use bigger hash table for IP ID generation - net: bridge: fix vlan tunnel dst refcnt when egressing - net: bridge: fix vlan tunnel dst null pointer dereference - tracing: Do no increment trace_clock_global() by one - tracing: Do not stop recording comms if the trace file is being read - tracing: Do not stop recording cmdlines when tracing is off - icmp: don't send out ICMP messages with a source address of 0.0.0.0 - net/af_unix: fix a data-race in unix_dgram_sendmsg / unix_release_sock - net: ipv4: fix memory leak in ip_mc_add1_src - net: make get_net_ns return error if NET_NS is disabled - net: add documentation to socket.c - sch_cake: Fix out of bounds when parsing TCP options and header - netfilter: synproxy: Fix out of bounds when parsing TCP options - rtnetlink: Fix regression in bridge VLAN configuration - udp: fix race between close() and udp_abort() - net: ipv4: fix memory leak in netlbl_cipsov4_add_std - fib: Return the correct errno code - net: Return the correct errno code - rtnetlink: Fix missing error code in rtnl_bridge_notify() - net: ipconfig: Don't override command-line hostnames or domains - nvme-loop: check for NVME_LOOP_Q_LIVE in nvme_loop_destroy_admin_queue() - nvme-loop: clear NVME_LOOP_Q_LIVE when nvme_loop_configure_admin_queue() fails - nvme-loop: reset queue count to 1 in nvme_loop_destroy_io_queues() - scsi: target: core: Fix warning on realtime kernels - proc: only require mm_struct for writing - tracing: Correct the length check which causes memory corruption - ftrace: Do not blindly read the ip address in ftrace_bug() - scsi: core: Only put parent device if host state differs from SHOST_CREATED - scsi: core: Put .shost_dev in failure path if host state changes to RUNNING - scsi: core: Fix error handling of scsi_host_alloc() - NFSv4: nfs4_proc_set_acl needs to restore NFS_CAP_UIDGID_NOMAP on error. - NFS: Fix use-after-free in nfs4_init_client() - NFS: Fix a potential NULL dereference in nfs_get_client() - sched/fair: Make sure to update tg contrib for blocked load - perf: Fix data race between pin_count increment/decrement - cgroup1: don't allow ' - wq: handle VM suspension in stall detection - cgroup: disable controllers at parse time - net: mdiobus: get rid of a BUG_ON() - netlink: disable IRQs for netlink_lock_table() - bonding: init notify_work earlier to avoid uninitialized use - proc: Track /proc/$pid/attr/ opener mm_struct - ACPI: EC: Look for ECDT EC after calling acpi_load_tables() - ACPI: probe ECDT before loading AML tables regardless of module-level code flag - mm, hugetlb: fix simple resv_huge_pages underflow on UFFDIO_COPY - x86/apic: Mark _all_ legacy interrupts when IO/APIC is missing - pid: take a reference when initializing `cad_pid` - netfilter: nfnetlink_cthelper: hit EBUSY on updates if size mismatches - ipvs: ignore IP_VS_SVC_F_HASHED flag when adding service - vfio/platform: fix module_put call in error flow - vfio/pci: zap_vma_ptes() needs MMU - vfio/pci: Fix error return code in vfio_ecap_init() - efi: cper: fix snprintf() use in cper_dimm_err_location() - efi: Allow EFI_MEMORY_XP and EFI_MEMORY_RO both to be cleared - lib/clear_user: ensure loop in __arch_clear_user cache-aligned - scsi: core: Treat device offline as a failure - Revert "scsi: check the whole result for reading write protect flag" - ext4: fix WARN_ON_ONCE(!buffer_uptodate) after an error writing the superblock - arm64/config: Set CONFIG_TXGBE=m by default - make bch_btree_check() to be multiple threads - Make compile successful when CONFIG_BCACHE is not set. - Move only dirty data when gc runnning, in order to reducing write amplification. - Add traffic policy for low cache available. - igmp: Add ip_mc_list lock in ip_check_mc_rcu - memcg: fix unsuitable null check after alloc memory - cpuidle: fix a build error when compiling haltpoll into module - config: enable KASAN and UBSAN by default - KVM: x86: expose AVX512_BF16 feature to guest - KVM: cpuid: remove has_leaf_count from struct kvm_cpuid_param - KVM: cpuid: rename do_cpuid_1_ent - KVM: cpuid: set struct kvm_cpuid_entry2 flags in do_cpuid_1_ent - KVM: cpuid: extract do_cpuid_7_mask and support multiple subleafs - KVM: cpuid: do_cpuid_ent works on a whole CPUID function - ext4: fix possible UAF when remounting r/o a mmp-protected file system - locks: Fix UBSAN undefined behaviour in flock64_to_posix_lock - iomap: Mark read blocks uptodate in write_begin - iomap: Clear page error before beginning a write - iomap: move the zeroing case out of iomap_read_page_sync - nbd: handle device refs for DESTROY_ON_DISCONNECT properly - cifs: Fix leak when handling lease break for cached root fid - mm/memcontrol.c: fix kasan slab-out-of-bounds in mem_cgroup_css_alloc - module: limit enabling module.sig_enforce - selftests/bpf: add test_spec_readahead_xfs_file to support specail async readahead - mm: support special async readahead - selftests/bpf: test_xfs_file support to clear FMODE_RANDOM - xfs: let writable tracepoint enable to clear flag of f_mode - jbd2: fix kabi broken in struct journal_s - btrfs: allow btrfs_truncate_block() to fallback to nocow for data space reservation - NFSv4.1: fix kabi for struct rpc_xprt - usb: gadget: rndis: Fix info leak of rndis - once: Fix panic when module unload - SUNRPC: Should wake up the privileged task firstly. - SUNRPC: Fix the batch tasks count wraparound. - bpf: Fix leakage under speculation on mispredicted branches - bpf: Do not mark insn as seen under speculative path verification - bpf: Inherit expanded/patched seen count from old aux data - bpf: Update selftests to reflect new error states - bpf, test_verifier: switch bpf_get_stack's 0 s> r8 test - bpf: Test_verifier, bpf_get_stack return value add <0 - bpf: extend is_branch_taken to registers - selftests/bpf: add selftest part of "bpf: improve verifier branch analysis" - selftests/bpf: Test narrow loads with off > 0 in test_verifier - bpf, selftests: Fix up some test_verifier cases for unprivileged - bpf: fix up selftests after backports were fixed - nvme-rdma: avoid request double completion for concurrent nvme_rdma_timeout - binfmt: Move install_exec_creds after setup_new_exec to match binfmt_elf - ext4: fix memory leak in ext4_fill_super - RDMA/hns: Add support for addressing when hopnum is 0 - RDMA/hns: Optimize hns buffer allocation flow - RDMA/hns: Check if depth of qp is 0 before configure - RDMA/hns: Optimize qp param setup flow - RDMA/hns: Optimize qp buffer allocation flow - RDMA/hns: Optimize qp destroy flow - RDMA/hns: Remove asynchronic QP destroy - RDMA/hns: Bugfix for posting a wqe with sge - RDMA/hns: Delete unnecessary variable max_post - RDMA/hns: optimize the duplicated code for qpc setting flow - RDMA/hns: Prevent undefined behavior in hns_roce_set_user_sq_size() - RDMA/umem: Add rdma_umem_for_each_dma_block() - RDMA/verbs: Add a DMA iterator to return aligned contiguous memory blocks - can: bcm: delay release of struct bcm_op after synchronize_rcu() - etmem_scan: fix memleak in vm_idle_read - x86/uprobes: Do not use prefixes.nbytes when looping over prefixes.bytes - Revert "arm64: capabilities: Merge entries for ARM64_WORKAROUND_CLEAN_CACHE" - Revert "arm64: capabilities: Merge duplicate Cavium erratum entries" - Revert "arm64: capabilities: Merge duplicate entries for Qualcomm erratum 1003" - net: hns3: update hns3 version to 1.9.40.24 - net: hns3: remove redundant assignment to rx_index - net: hns3: Fix potential null pointer defererence of null ae_dev - net: hns3: not reset TQP in the DOWN while VF resetting - net: hns3: remove redundant enum type HNAE3_RESTORE_CLIENT - net: hns3: add stats logging when skb padding fails - net: hns3: add tx send size handling for tso skb - net: hns3: add handling for xmit skb with recursive fraglist - net: hns3: use napi_consume_skb() when cleaning tx desc - net: hns3: use writel() to optimize the barrier operation - net: hns3: optimize the rx clean process - net: hns3: optimize the tx clean process - net: hns3: batch tx doorbell operation - net: hns3: batch the page reference count updates - net: hns3: streaming dma buffer sync between cpu and device - net: hns3: rename buffer-related functions - net: hns3: pointer type of buffer should be void - net: hns3: remove unnecessary devm_kfree - net: hns3: add suspend and resume pm_ops - Revert "net: hns3: add suspend/resume function for hns3 driver" - net: hns3: change flr_prepare/flr_done function names - net: hns3: change hclge_reset_done function name - net: hns3: configure promisc mode for VF asynchronously - kabi: add kabi list for x86_64 - kabi: update kabi list for arm64 - hugetlbfs: hugetlb_fault_mutex_hash() cleanup - ipv6: record frag_max_size in atomic fragments in input path - scsi: libsas: Use _safe() loop in sas_resume_port() - SMB3: incorrect file id in requests compounded with open - NFSv4: Fix v4.0/v4.1 SEEK_DATA return -ENOTSUPP when set NFS_V4_2 config - NFS: Don't corrupt the value of pg_bytes_written in nfs_do_recoalesce() - NFS: fix an incorrect limit in filelayout_decode_layout() - dm snapshot: properly fix a crash when an origin has no snapshots - proc: Check /proc/$pid/attr/ writes against file opener - iommu/vt-d: Fix sysfs leak in alloc_iommu() - NFSv4: Fix a NULL pointer dereference in pnfs_mark_matching_lsegs_return() - cifs: set server->cipher_type to AES-128-CCM for SMB3.0 - tty: vt: always invoke vc->vc_sw->con_resize callback - vt: Fix character height handling with VT_RESIZEX - vgacon: Record video mode changes with VT_RESIZEX - Revert "niu: fix missing checks of niu_pci_eeprom_read" - Revert "qlcnic: Avoid potential NULL pointer dereference" - Revert "rtlwifi: fix a potential NULL pointer dereference" - Revert "media: rcar_drif: fix a memory disclosure" - Revert "gdrom: fix a memory leak bug" - Revert "scsi: ufs: fix a missing check of devm_reset_control_get" - Revert "video: imsttfb: fix potential NULL pointer dereferences" - Revert "hwmon: (lm80) fix a missing check of bus read in lm80 probe" - Revert "leds: lp5523: fix a missing check of return value of lp55xx_read" - Revert "net: stmicro: fix a missing check of clk_prepare" - Revert "video: hgafb: fix potential NULL pointer dereference" - dm snapshot: fix crash with transient storage and zero chunk size - Revert "serial: mvebu-uart: Fix to avoid a potential NULL pointer dereference" - Revert "rapidio: fix a NULL pointer dereference when create_workqueue() fails" - Revert "ALSA: sb8: add a check for request_region" - cifs: fix memory leak in smb2_copychunk_range - locking/mutex: clear MUTEX_FLAGS if wait_list is empty due to signal - nvmet: seset ns->file when open fails - ptrace: make ptrace() fail if the tracee changed its pid unexpectedly - firmware: arm_scpi: Prevent the ternary sign expansion bug - ipv6: remove extra dev_hold() for fallback tunnels - ip6_tunnel: sit: proper dev_{hold|put} in ndo_[un]init methods - sit: proper dev_{hold|put} in ndo_[un]init methods - ip6_gre: proper dev_{hold|put} in ndo_[un]init methods - block: reexpand iov_iter after read/write - scsi: target: tcmu: Return from tcmu_handle_completions() if cmd_id not found - ACPI / hotplug / PCI: Fix reference count leak in enable_slot() - nvme: do not try to reconfigure APST when the controller is not live - netfilter: conntrack: Make global sysctls readonly in non-init netns - kobject_uevent: remove warning in init_uevent_argv() - blk-mq: Swap two calls in blk_mq_exit_queue() - userfaultfd: release page in error path to avoid BUG_ON - netfilter: nftables: avoid overflows in nft_hash_buckets() - kernel: kexec_file: fix error return code of kexec_calculate_store_digests() - sched/fair: Fix unfairness caused by missing load decay - netfilter: nfnetlink_osf: Fix a missing skb_header_pointer() NULL check - net: fix nla_strcmp to handle more then one trailing null character - ksm: fix potential missing rmap_item for stable_node - mm/hugeltb: handle the error case in hugetlb_fix_reserve_counts() - khugepaged: fix wrong result value for trace_mm_collapse_huge_page_isolate() - netfilter: xt_SECMARK: add new revision to fix structure layout - sctp: fix a SCTP_MIB_CURRESTAB leak in sctp_sf_do_dupcook_b - sctp: do asoc update earlier in sctp_sf_do_dupcook_a - NFSv4.2 fix handling of sr_eof in SEEK's reply - pNFS/flexfiles: fix incorrect size check in decode_nfs_fh() - NFS: Deal correctly with attribute generation counter overflow - NFSv4.2: Always flush out writes in nfs42_proc_fallocate() - PCI: Release OF node in pci_scan_device()'s error path - ethtool: ioctl: Fix out-of-bounds warning in store_link_ksettings_for_user() - sctp: Fix out-of-bounds warning in sctp_process_asconf_param() - cuse: prevent clone - ip6_vti: proper dev_{hold|put} in ndo_[un]init methods - tpm: fix error return code in tpm2_get_cc_attrs_tbl() - sctp: delay auto_asconf init until binding the first addr - Revert "net/sctp: fix race condition in sctp_destroy_sock" - smp: Fix smp_call_function_single_async prototype - net: Only allow init netns to set default tcp cong to a restricted algo - mm/memory-failure: unnecessary amount of unmapping - mm/sparse: add the missing sparse_buffer_fini() in error branch - drivers/block/null_blk/main: Fix a double free in null_init. - sched/debug: Fix cgroup_path[] serialization - x86/events/amd/iommu: Fix sysfs type mismatch - vfio/mdev: Do not allow a mdev_type to have a NULL parent pointer - ata: libahci_platform: fix IRQ check - x86/kprobes: Fix to check non boostable prefixes correctly - ACPI: CPPC: Replace cppc_attr with kobj_attribute - irqchip/gic-v3: Fix OF_BAD_ADDR error handling - x86/microcode: Check for offline CPUs before requesting new microcode - ovl: fix missing revert_creds() on error path - x86/cpu: Initialize MSR_TSC_AUX if RDTSCP *or* RDPID is supported - md: Fix missing unused status line of /proc/mdstat - md: md_open returns -EBUSY when entering racing area - md: factor out a mddev_find_locked helper from mddev_find - md: split mddev_find - md-cluster: fix use-after-free issue when removing rdev - md/bitmap: wait for external bitmap writes to complete during tear down - dm rq: fix double free of blk_mq_tag_set in dev remove after table load fails - dm space map common: fix division bug in sm_ll_find_free_block() - dm persistent data: packed struct should have an aligned() attribute too - tracing: Restructure trace_clock_global() to never block - tracing: Map all PIDs to command lines - tty: fix memory leak in vc_deallocate - ext4: fix error code in ext4_commit_super - posix-timers: Preserve return value in clock_adjtime32() - Revert 337f13046ff0 ("futex: Allow FUTEX_CLOCK_REALTIME with FUTEX_WAIT op") - dm raid: fix inconclusive reshape layout on fast raid4/5/6 table reload sequences - md/raid1: properly indicate failure when ending a failed write request - NFSv4: Don't discard segments marked for return in _pnfs_return_layout() - NFS: Don't discard pNFS layout segments that are marked for return - ACPI: GTDT: Don't corrupt interrupt mappings on watchdow probe failure - arm64/vdso: Discard .note.gnu.property sections in vDSO - perf/arm_pmu_platform: Fix error handling - genirq/matrix: Prevent allocation counter corruption - crypto: api - check for ERR pointers in crypto_destroy_tfm() - cifs: Return correct error code from smb2_get_enc_key - ftrace: Handle commands when closing set_ftrace_filter file - ACPI/IORT: Fix 'Number of IDs' handling in iort_id_map() - ext4: do not use extent after put_bh - modpost: explain why we can't use strsep - modpost: remove use of non-standard strsep() in HOSTCC code - modpost: use read_text_file() and get_line() for reading text files - modpost: add read_text_file() and get_line() helpers - arm64: capabilities: Merge duplicate entries for Qualcomm erratum 1003 - arm64: capabilities: Merge duplicate Cavium erratum entries - arm64: capabilities: Merge entries for ARM64_WORKAROUND_CLEAN_CACHE - net: phy: ensure phylib state machine is stopped after calling phy_stop - net: linkwatch: add check for netdevice being present to linkwatch_do_dev - net: phy: call state machine synchronously in phy_stop - of: fix kmemleak crash caused by imbalance in early memory reservation - random: fix soft lockup when trying to read from an uninitialized blocking pool - random: only read from /dev/random after its pool has received 128 bits - block: check queue's limits.discard_granularity in __blkdev_issue_discard() - block: loop: set discard granularity and alignment for block device backed loop - posix-cpu-timers: Stop disabling timers on mt-exec - kprobes: Fix compiler warning for !CONFIG_KPROBES_ON_FTRACE - perf top: Fix stdio interface input handling with glibc 2.28+ - iommu/vt-d: Fix mm reference leak - iommu/dma: Fix for dereferencing before null checking - srcu: Apply *_ONCE() to ->srcu_last_gp_end - arm64: Kconfig: select HAVE_FUTEX_CMPXCHG - kill kernfs_pin_sb() - mm, thp: fix defrag setting if newline is not used - nfsd: Clone should commit src file metadata too - nfsd: Ensure CLONE persists data and metadata changes to the target file - x86/sysfb: Fix check for bad VRAM size - x86/timer: Force PIT initialization when !X86_FEATURE_ARAT - x86/timer: Don't skip PIT setup when APIC is disabled or in legacy mode - x86/timer: Skip PIT initialization on modern chipsets - x86/apic: Rename 'lapic_timer_frequency' to 'lapic_timer_period' - iommu/vt-d: Handle PCI bridge RMRR device scopes in intel_iommu_get_resv_regions - iommu/vt-d: Handle RMRR with PCI bridge device scopes - iommu/vt-d: Introduce is_downstream_to_pci_bridge helper - crypto: x86 - remove SHA multibuffer routines and mcryptd - iommu/vt-d: Duplicate iommu_resv_region objects per device list - memcg: fix kabi broken when memory cgroup enhance - mm: memcontrol: fix NULL-ptr deref in percpu stats flush - mm: memcg: get number of pages on the LRU list in memcgroup base on lru_zone_size - mm: memcontrol: fix percpu vmstats and vmevents flush - mm, memcg: partially revert "mm/memcontrol.c: keep local VM counters in sync with the hierarchical ones" - mm/memcontrol.c: keep local VM counters in sync with the hierarchical ones - mm: memcontrol: flush percpu vmevents before releasing memcg - mm: memcontrol: flush percpu vmstats before releasing memcg - mm/memcontrol: fix wrong statistics in memory.stat - mm: memcontrol: don't batch updates of local VM stats and events - mm: memcontrol: fix NUMA round-robin reclaim at intermediate level - mm: memcontrol: fix recursive statistics correctness & scalabilty - mm: memcontrol: move stat/event counting functions out-of-line - mm: memcontrol: make cgroup stats and events query API explicitly local - mm: memcontrol: quarantine the mem_cgroup_[node_]nr_lru_pages() API - mm, memcg: rename ambiguously named memory.stat counters and functions - mm/memcontrol.c: fix memory.stat item ordering - mm: memcontrol: expose THP events on a per-memcg basis - mm: memcontrol: track LRU counts in the vmstats array - mm: memcontrol: push down mem_cgroup_nr_lru_pages() - mm: memcontrol: push down mem_cgroup_node_nr_lru_pages() - mm: workingset: don't drop refault information prematurely - mm: memcontrol: replace zone summing with lruvec_page_state() - mm: memcontrol: replace node summing with memcg_page_state() - mm, oom: add oom victim's memcg to the oom context information - mm/oom_kill.c: fix uninitialized oc->constraint - mm, oom: reorganize the oom report in dump_header - memcg: update the child's qos_level synchronously in memcg_qos_write() - memcg: Add static key for memcg priority - memcg: fix kabi broken when enable CONFIG_MEMCG_QOS - memcg: enable CONFIG_MEMCG_QOS by default - memcg: support priority for oom - scsi: core: Fix failure handling of scsi_add_host_with_dma() - fuse: fix the ->direct_IO() treatment of iov_iter - bdev: Do not return EBUSY if bdev discard races with write - block: mark flush request as IDLE when it is really finished - blk-mq: mark flush request as IDLE in flush_end_io() - vhost_net: avoid tx queue stuck when sendmsg fails - iommu/vt-d: Add support for ACPI device use physical, node as pci device to establish identity mapping - io_uring: NULL files dereference by SQPOLL - vgacon: remove software scrollback support - block: dio: ensure the memory order between bi_private and bi_css - ext4: fix memory leak in ext4_fill_super - RDMA/ucma: Rework ucma_migrate_id() to avoid races with destroy - RDMA/ucma: Add missing locking around rdma_leave_multicast() - RDMA/ucma: Fix locking for ctx->events_reported - RDMA/ucma: Put a lock around every call to the rdma_cm layer - mm/memory-failure: make sure wait for page writeback in memory_failure - can: bcm: fix infoleak in struct bcm_msg_head - blk-wbt: make sure throttle is enabled properly - blk-wbt: introduce a new disable state to prevent false positive by rwb_enabled() - ext4: stop return ENOSPC from ext4_issue_zeroout - dm btree remove: assign new_root only when removal succeeds - block: call bdput() to avoid memleak - scsi: remove unused kobj map for sd devie to avoid memleak - tools build: Check if gettid() is available before providing helper - tools build feature: Check if eventfd() is available - tools build feature: Check if get_current_dir_name() is available - perf tools: Use - bpf: move new add member to the end of the struct bpf_prog_aux- add buildrequire: perl-devel for with_perf- cpuidle: fix container_of err in cpuidle_device and cpuidle_driver- sched/membarrier: fix NULL poiner in membarrier_global_expedited - writeback: don't warn on an unregistered BDI in __mark_inode_dirty- fs/buffer.c: add checking buffer head stat before clear - Bluetooth: SMP: Fail if remote and local public keys are identical - Bluetooth: use correct lock to prevent UAF of hdev object - Bluetooth: fix the erroneous flush_work() order - iomap: Make sure iomap_end is called after iomap_begin - x86/kvm: Add "nopvspin" parameter to disable PV spinlocks - scsi: libsas: add lun number check in .slave_alloc callback - nfc: fix NULL ptr dereference in llcp_sock_getname() after failed connect - USB:ehci:fix Kunpeng920 ehci hardware problem - nvme: don't warn on block content change effects - block: recalculate segment count for multi-segment discards correctly - nbd: Fix NULL pointer in flush_workqueue - Bluetooth: Fix slab-out-of-bounds read in hci_extended_inquiry_result_evt() - HID: make arrays usage and value to be the same - ath10k: Validate first subframe of A-MSDU before processing the list - mac80211: extend protection against mixed key and fragment cache attacks - mac80211: do not accept/forward invalid EAPOL frames - mac80211: prevent attacks on TKIP/WEP as well - mac80211: check defrag PN against current frame - mac80211: add fragment cache to sta_info - mac80211: drop A-MSDUs on old ciphers - cfg80211: mitigate A-MSDU aggregation attacks - mac80211: properly handle A-MSDUs that start with an RFC 1042 header - mac80211: prevent mixed key and fragment cache attacks - mac80211: assure all fragments are encrypted - mac80211: mark station unauthorized before key removal - block: avoid creating invalid symlink file for patitions - block: take bd_mutex around delete_partitions in del_gendisk - NFSv4: Fix second deadlock in nfs4_evict_inode() - NFSv4: Fix deadlock between nfs4_evict_inode() and nfs4_opendata_get_inode() - NFSv4.1: fix handling of backchannel binding in BIND_CONN_TO_SESSION - NFS: Don't gratuitously clear the inode cache when lookup failed - NFS: Don't revalidate the directory permissions on a lookup failure - NFS: nfs_delegation_find_inode_server must first reference the superblock - nfs4: strengthen error check to avoid unexpected result - NFS: Fix interrupted slots by sending a solo SEQUENCE operation - NFS: Ensure we time out if a delegreturn does not complete - NFSv4.0: nfs4_do_fsinfo() should not do implicit lease renewals - NFS: Use kmemdup_nul() in nfs_readdir_make_qstr() - NFSv3: FIx bug when using chacl and chmod to change acl - NFSv4.x: Handle bad/dead sessions correctly in nfs41_sequence_process() - NFSv4.1: Only reap expired delegations - NFSv4.1: Fix open stateid recovery - NFSv4.1: Don't process the sequence op more than once. - NFS: Ensure NFS writeback allocations don't recurse back into NFS. - nfs_remount(): don't leak, don't ignore LSM options quietly - UACCE backport from mainline - crypto: hisilicon-Cap block size at 2^31 - crypto: hisilicon-hpre add req check when callback - crypto: hisilicon- count send_ref when sending bd - crypto: hisilicon-enhancement of qm DFX - crypto: hisilicon-memory management optimization - net: hns3: update hns3 version to 1.9.38.12 - net: hns3: add match_id to check mailbox response from PF to VF - net: hns3: fix possible mismatches resp of mailbox - net: hns3: fix the logic for clearing resp_msg - net: hns3: fix queue id check error when configure flow director rule by ethtool - net: hns3: add check for HNS3_NIC_STATE_INITED before net open - net: hns3: add waiting time before cmdq memory is released - net: hns3: disable firmware compatible features when uninstall PF - net: hns3: fix change RSS 'hfunc' ineffective issue - net: hns3: fix inconsistent vf id print - net: hns3: remove redundant variable initialization - net: hns3: replace the tab before the left brace with one space - net: hns3: fix hns3_cae_pfc_storm.h missing header guard problem - net: hns3: modify an error type configuration - net: hns3: put off calling register_netdev() until client initialize complete - net: hns3: replace disable_irq by IRQ_NOAUTOEN flag - net: hns3: update rss indirection table after setup tc - net: hns3: don't change tc mqprio configuration when client is unregistered - net: hns3: remove redundant client_setup_tc handle - arm64/mpam: Fix use-after-free in mkdir_resctrl_prepare()- selftests/bpf: add test_xfs_file.c and test_set_xfs_file.c - bpf: add bpf_probe_read_str into bpf_helpers.h - xfs: add writable tracepoint for xfs file buffer read - readahead: introduce FMODE_WILLNEED to read first 2MB of file- tools: libbpf: fix compiler error - bpf: fix kabi for struct bpf_prog_aux and struct bpf_raw_event_map - tools: bpftool: add raw_tracepoint_writable prog type to header - tools: sync bpf.h - bpf: add writable context for raw tracepoints - x86/tsc: Respect tsc command line paraemeter for clocksource_tsc_early- cpuidle: fix kabi broken in cpuidle_device and cpuidle_driver - config: set default value of haltpoll - ARM: cpuidle: Add support for cpuidle-haltpoll driver for ARM - arm64: Add some definitions of kvm_para* - cpuidle-haltpoll: Use arch_cpu_idle() to replace default_idle() - arm64: Optimize ttwu IPI - config: enable CONFIG_CPU_IDLE_GOV_HALTPOLL and CONFIG_HALTPOLL_CPUIDLE default - KVM: polling: add architecture backend to disable polling - cpuidle-haltpoll: Fix small typo - cpuidle: haltpoll: allow force loading on hosts without the REALTIME hint - cpuidle-haltpoll: Enable kvm guest polling when dedicated physical CPUs are available - cpuidle-haltpoll: do not set an owner to allow modunload - cpuidle-haltpoll: return -ENODEV on modinit failure - cpuidle-haltpoll: vcpu hotplug support - cpuidle-haltpoll: set haltpoll as preferred governor - cpuidle: allow governor switch on cpuidle_register_driver() - cpuidle: governor: Add new governors to cpuidle_governors again - cpuidle: Add cpuidle.governor= command line parameter - cpuidle-haltpoll: disable host side polling when kvm virtualized - kvm: x86: add host poll control msrs - cpuidle: add haltpoll governor - governors: unify last_state_idx - cpuidle: use first valid target residency as poll time - cpuidle: header file stubs must be "static inline" - cpuidle: add poll_limit_ns to cpuidle_device structure - add cpuidle-haltpoll driver - cpuidle: poll_state: Fix default time limit - cpuidle: poll_state: Disregard disable idle states - cpuidle: poll_state: Revise loop termination condition - cpuidle: menu: Do not update last_state_idx in menu_select() - bpf: No need to simulate speculative domain for immediates - bpf: Fix mask direction swap upon off reg sign change - bpf: Wrap aux data inside bpf_sanitize_info container- ata: ahci: Disable SXS for Hisilicon Kunpeng920 - fuse: don't ignore errors from fuse_writepages_fill() - NFS: finish_automount() requires us to hold 2 refs to the mount record - NFS: If nfs_mountpoint_expiry_timeout < 0, do not expire submounts - NFS: remove unused check for negative dentry - NFSv3: use nfs_add_or_obtain() to create and reference inodes - NFS: Refactor nfs_instantiate() for dentry referencing callers - sysfs: Remove address alignment constraint in sysfs_emit{_at} - Revert "mm, sl[aou]b: guarantee natural alignment for kmalloc(power-of-two)" - Revert "mm, sl[ou]b: improve memory accounting" - Revert "mm: memcontrol: fix slub memory accounting" - io_uring: truncate lengths larger than MAX_RW_COUNT on provide buffers - arm/ras: Report ARM processor information to userspace - fuse: update attr_version counter on fuse_notify_inval_inode() - alinux: random: speed up the initialization of module - net: mac802154: Fix general protection fault - cipso,calipso: resolve a number of problems with the DOI refcounts - Bluetooth: verify AMP hci_chan before amp_destroy - net/nfc: fix use-after-free llcp_sock_bind/connect - x86: Select HARDIRQS_SW_RESEND on x86 - x86/apic/vector: Force interupt handler invocation to irq context- tools arch x86: Sync asm/cpufeatures.h with the with the kernel - cpufreq: intel_pstate: Also use CPPC nominal_perf for base_frequency - ACPI / CPPC: Fix guaranteed performance handling - perf vendor events: Add JSON metrics for Cascadelake server - perf vendor events: Add stepping in CPUID string for x86 - cpufreq: intel_pstate: Fix compilation for !CONFIG_ACPI - cpufreq: intel_pstate: Add base_frequency attribute - ACPI / CPPC: Add support for guaranteed performance - EDAC, skx: Fix randconfig builds in a better way - EDAC, skx: Fix randconfig builds - EDAC, skx_edac: Add address translation for non-volatile DIMMs - ACPI/ADXL: Add address translation interface using an ACPI DSM - x86/mce: Add macros for the corrected error count bit field - x86/mce: Use BIT_ULL(x) for bit mask definitions - x86/cpufeatures: Enumerate the new AVX512 BFLOAT16 instructions - tools/testing/selftests/exec: fix link error - NFSv4.1: Don't rebind to the same source port when reconnecting to the server - genirq: Sanitize state handling in check_irq_resend() - genirq: Add return value to check_irq_resend() - irqchip/gic-v2, v3: Prevent SW resends entirely - irqchip/git-v3-its: Implement irq_retrigger callback for device-triggered LPIs - irqchip/gic-v2, v3: Implement irq_chip->irq_retrigger() - genirq: Walk the irq_data hierarchy when resending an interrupt - genirq: Add protection against unsafe usage of generic_handle_irq()- jbd2: don't abort the journal when freeing buffers - jbd2: ensure abort the journal if detect IO error when writing original buffer back - jbd2: remove the out label in __jbd2_journal_remove_checkpoint() - x86/unwind/orc: Remove boot-time ORC unwind tables sorting - scripts/sorttable: Implement build-time ORC unwind table sorting - scripts/sorttable: Rename 'sortextable' to 'sorttable' - scripts/sortextable: Refactor the do_func() function - scripts/sortextable: Remove dead code - scripts/sortextable: Clean up the code to meet the kernel coding style better - scripts/sortextable: Rewrite error/success handling - treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 378 - ext4: Fix bug on in ext4_es_cache_extent as ext4_split_extent_at failed - Revert "ext4: Fix bug on in ext4_es_cache_extent as ext4_split_extent_at failed" - nfs4.0: Refetch lease_time after clientid update - nfs4: Rename nfs41_setup_state_renewal - nfs4: Make nfs4_proc_get_lease_time available for nfs4.0 - nfs: Fix copy-and-paste error in debug message - ext4: cleanup in-core orphan list if ext4_truncate() failed to get a transaction handle - bluetooth: eliminate the potential race condition when removing the HCI controller - mm: enhance variables check and sync for pin mem - perf jit: Fix inaccurate DWARF line table - perf jvmti: Remove redundant jitdump line table entries - perf jvmti: Fix demangling Java symbols - perf tests: Add test for the java demangler - perf jvmti: Do not report error when missing debug information - perf jvmti: Fix jitdump for methods without debug info - bpf: Fix leakage of uninitialized bpf stack under speculation - bpf: Fix masking negation logic upon negative dst register - bcache: add readahead cache policy options via sysfs interface - mm/page_alloc: fix managed_pages of zone is incorrect and out of bounds - freezer: Add unsafe version of freezable_schedule_timeout_interruptible() for NFS - NFS: Allow signal interruption of NFS4ERR_DELAYed operations - SUNRPC: Make "no retrans timeout" soft tasks behave like softconn for timeouts - SUNRPC: Don't let RPC_SOFTCONN tasks time out if the transport is connected - ext4: fix check to prevent false positive report of incorrect used inodes - livepatch/x86_64: Fix the deadlock when insmoding livepatch kernel module - tools/testing/selftests: add self-test for verifying load alignment - fs/binfmt_elf: use PT_LOAD p_align values for suitable start address - ext4: introduce ext4_sb_bread_unmovable() to replace sb_bread_unmovable() - ext4: use ext4_sb_bread() instead of sb_bread() - ext4: introduce ext4_sb_breadahead_unmovable() to replace sb_breadahead_unmovable() - ext4: use ext4_buffer_uptodate() in __ext4_get_inode_loc() - ext4: use common helpers in all places reading metadata buffers - ext4: introduce new metadata buffer read helpers - ext4: treat buffers contining write errors as valid in ext4_sb_bread() - bpf: Fix truncation handling for mod32 dst reg wrt zero - bpf: Fix 32 bit src register truncation on div/mod - arm64: bpf: implement jitting of JMP32 - x86_64: bpf: implement jitting of JMP32 - bpf: JIT blinds support JMP32 - bpf: interpreter support for JMP32 - tools: bpftool: teach cfg code about JMP32 - bpf: disassembler support JMP32 - bpf: verifier support JMP32 - bpf: refactor verifier min/max code for condition jump - bpf: allocate 0x06 to new eBPF instruction class JMP32- ovl: allow upperdir inside lowerdir - ACPI: x86: Call acpi_boot_table_init() after acpi_table_upgrade() - ACPI: tables: x86: Reserve memory occupied by ACPI tables - x86/crash: Fix crash_setup_memmap_entries() out-of-bounds access - locking/qrwlock: Fix ordering in queued_write_lock_slowpath() - net: ip6_tunnel: Unregister catch-all devices - netfilter: nft_limit: avoid possible divide error in nft_limit_init - netfilter: conntrack: do not print icmpv6 as unknown via /proc - scsi: libsas: Reset num_scatter if libata marks qc as NODATA - arm64: alternatives: Move length validation in alternative_{insn, endif} - arm64: fix inline asm in load_unaligned_zeropad() - readdir: make sure to verify directory entry for legacy interfaces too - neighbour: Disregard DEAD dst in neigh_update - driver core: Fix locking bug in deferred_probe_timeout_work_func() - netfilter: x_tables: fix compat match/target pad out-of-bound write - workqueue: Move the position of debug_work_activate() in __queue_work() - xfrm: interface: fix ipv4 pmtu check to honor ip header df - net-ipv6: bugfix - raw & sctp - switch to ipv6_can_nonlocal_bind() - net: ensure mac header is set in virtio_net_hdr_to_skb() - fs: direct-io: fix missing sdio->boundary - net: ipv6: check for validity before dereferencing cfg->fc_nlinfo.nlh - cifs: Silently ignore unknown oplock break handle - cifs: revalidate mapping when we open files for SMB1 POSIX - scsi: target: pscsi: Clean up after failure in pscsi_map_sg() - mm: fix race by making init_zero_pfn() early_initcall - tracing: Fix stack trace event size - PM: runtime: Fix ordering in pm_runtime_get_suppliers() - PM: runtime: Fix race getting/putting suppliers at probe - ext4: do not iput inode under running transaction in ext4_rename() - locking/ww_mutex: Simplify use_ww_ctx & ww_ctx handling - thermal/core: Add NULL pointer check before using cooling device stats - scsi: st: Fix a use after free in st_open() - vhost: Fix vhost_vq_reset() - rpc: fix NULL dereference on kmalloc failure - ext4: fix bh ref count on error paths - ipv6: weaken the v4mapped source check - tcp: relookup sock for RST+ACK packets handled by obsolete req sock - nfs: we don't support removing system.nfs4_acl - NFSv4.2: fix return value of _nfs4_get_security_label() - nfs: fix PNFS_FLEXFILE_LAYOUT Kconfig default - pNFS/NFSv4: Try to return invalid layout in pnfs_layout_process() - pNFS/NFSv4: Fix a layout segment leak in pnfs_layout_process() - NFSv4.2: condition READDIR's mask for security label based on LSM state - NFSv4.2: support EXCHGID4_FLAG_SUPP_FENCE_OPS 4.2 EXCHANGE_ID flag - NFS: fix nfs_path in case of a rename retry - NFSv4.1 handle ERR_DELAY error reclaiming locking state on delegation recall - NFS: Don't return layout segments that are in use - NFS: Don't move layouts to plh_return_segs list while in use - SUNRPC reverting d03727b248d0 ("NFSv4 fix CLOSE not waiting for direct IO compeletion") - NFSv4 fix CLOSE not waiting for direct IO compeletion - NFSv4.1 fix rpc_call_done assignment for BIND_CONN_TO_SESSION - nfs: Fix potential posix_acl refcnt leak in nfs3_set_acl - NFSv4/pnfs: Return valid stateids in nfs_layout_find_inode_by_stateid() - NFSv4.1 make cachethis=no for writes - NFS/pnfs: Fix pnfs_generic_prepare_to_resend_writes() - NFS/pnfs: Bulk destroy of layouts needs to be safe w.r.t. umount - cgroup/files: support boot parameter to control if disable files cgroup - efi: Fix a race and a buffer overflow while reading efivars via sysfs - RDMA/hns: Allocate one more recv SGE for HIP08 - mm: memcontrol: fix slub memory accounting - mm, sl[ou]b: improve memory accounting - mm: fix numa stats for thp migration - mm/vmscan: count layzfree pages and fix nr_isolated_* mismatch - SUNRPC: Close a race with transport setup and module put - sunrpc: Change the place of endtime in struct krb5_ctx - bpf: Tighten speculative pointer arithmetic mask - bpf: Move sanitize_val_alu out of op switch - bpf: Refactor and streamline bounds check into helper - bpf: Improve verifier error messages for users - bpf: Rework ptr_limit into alu_limit and add common error path - bpf: Ensure off_reg has no mixed signed bounds for all types - bpf: Move off_reg into sanitize_ptr_alu - bpf: Add sanity check for upper ptr_limit - bpf: Simplify alu_limit masking for pointer arithmetic- add kabi list for aarch64 and x86_64- pid: fix pid recover method kabi change - config: enable kernel hotupgrade features by default - kexec: Add quick kexec support for kernel - arm64: smp: Add support for cpu park - pid: add pid reserve method for checkpoint and restore - mm: add pin memory method for checkpoint add restore - Revert "sched: Introduce qos scheduler for co-location" - Revert "sched: Throttle qos cfs_rq when current cpu is running online task" - Revert "sched: Enable qos scheduler config" - Revert "memcg: support priority for oom" - Revert "memcg: enable CONFIG_MEMCG_QOS by default" - Revert "memcg: fix kabi broken when enable CONFIG_MEMCG_QOS" - f2fs: fix to avoid out-of-bounds memory access - ext4: Reduce ext4 timestamp warnings - livepatch: Restoring code segment permissions after stop_machine completed - livepatch: Delete redundant variable 'flag' - memcg: fix kabi broken when enable CONFIG_MEMCG_QOS - memcg: enable CONFIG_MEMCG_QOS by default - memcg: support priority for oom - sched: Enable qos scheduler config - sched: Throttle qos cfs_rq when current cpu is running online task - sched: Introduce qos scheduler for co-location - ipv6: route: convert comma to semicolon - ipv6/route: Add a missing check on proc_dointvec - netfilter: xtables: avoid BUG_ON - SUNRPC: Test whether the task is queued before grabbing the queue spinlocks - SUNRPC: If there is no reply expected, bail early from call_decode - SUNRPC: Fix backchannel latency metrics - sunrpc: convert to time64_t for expiry - sunrpc: Fix potential leaks in sunrpc_cache_unhash() - SUNRPC: Skip zero-refcount transports - SUNRPC: Fix buffer handling of GSS MIC without slack - SUNRPC: Don't allow compiler optimisation of svc_xprt_release_slot() - SUNRPC/nfs: Fix return value for nfs4_callback_compound() - net/sunrpc: return 0 on attempt to write to "transports" - net/sunrpc: Fix return value for sysctl sunrpc.transports - sunrpc: raise kernel RPC channel buffer size - sunrpc: add missing newline when printing parameter 'pool_mode' by sysfs - xprtrdma: Fix trace point use-after-free race - SUNRPC: Fix backchannel RPC soft lockups - SUNRPC/cache: Fix unsafe traverse caused double-free in cache_purge - nfsd: export upcalls must not return ESTALE when mountd is down - sunrpc/cache: handle missing listeners better. - xprtrdma: Fix handling of RDMA_ERROR replies - xprtrdma: Expose transport header errors - sunrpc: destroy rpc_inode_cachep after unregister_filesystem - xprtrdma: fix incorrect header size calculations - nvme: fix ns removal hang when failing to revalidate due to a transient error - kernel/cputime: do not update cputime when cpu offline - perf/x86: Always store regs->ip in perf_callchain_kernel() - perf/x86: Make perf callchains work without CONFIG_FRAME_POINTER - irqchip/gic-v3: Do not enable irqs when handling spurious interrups - config: enable CONFIG_HW_RANDOM_HISI_V2 by default - hwrng: add data_mode to support rand data with post process - hwrng: add HiSilicon TRNG driver- bcache: Rewrite patch to delay to invalidate cache data - nfc: Avoid endless loops caused by repeated llcp_sock_connect() - nfc: fix memory leak in llcp_sock_connect() - nfc: fix refcount leak in llcp_sock_connect() - nfc: fix refcount leak in llcp_sock_bind()- KVM: arm64: Support the vCPU preemption check - KVM: arm64: Add interface to support vCPU preempted check - KVM: arm64: Support pvsched preempted via shared structure - KVM: arm64: Implement PV_SCHED_FEATURES call - KVM: arm64: Document PV-sched interface - KVM: Check preempted_in_kernel for involuntary preemption - KVM: Boost vCPUs that are delivering interrupts - arm64/spinlock: fix a -Wunused-function warning - locking/osq: Use optimized spinning loop for arm64 - arm/arm64: Make use of the SMCCC 1.1 wrapper - arm/arm64: Provide a wrapper for SMCCC 1.1 calls - KVM: Implement kvm_put_guest() - KVM: arm/arm64: Factor out hypercall handling from PSCI code- bcache: Add a sample of userspace prefetch client - bcache: Delay to invalidate cache data in writearound write - bcache: inflight prefetch requests block overlapped normal requests - bcache: provide a switch to bypass all IO requests - bcache: add a framework to perform prefetch- net/hinic: Fix null pointer dereference in hinic_physical_port_id - kvm: debugfs: Export x86 kvm exits to vcpu_stat - ext4: fix time overflow - ext4: drop legacy pre-1970 encoding workaround - fuse: fix live lock in fuse_iget() - fuse: fix bad inode - net/sctp: fix race condition in sctp_destroy_sock - config: set config hip08 prefetch default value - ext4: do not set SB_ACTIVE in ext4_orphan_cleanup() - RDMA/hns: add eq and cq time cfg compatibility support. - nvme: fix incorrect behavior when BLKROSET is called by the user - nvme-fc: fix error loop in create_hw_io_queues - nvme-fc: Fix wrong return value in __nvme_fc_init_request() - nvme-multipath: fix deadlock between ana_work and scan_work - nvme: fix deadlock caused by ANA update wrong locking - nvme-multipath: Fix memory leak with ana_log_buf - nvme-fc: fix module unloads while lports still pending - ipmi: remve duplicate code in __ipmi_bmc_register() - ipmi_si_intf: Fix race in timer shutdown handling - ipmi_ssif: fix unexpected driver unregister warning - ipmi_si: fix unexpected driver unregister warning - ipmi:ssif: Only unregister the platform driver if it was registered - ipmi: Make ipmi_interfaces_srcu variable static - ipmi: Fix return value when a message is truncated - ipmi: Free the address list on module cleanup - net: hns3: clear VF down state bit before request link status - config: disable config ARM64_BOOTPARAM_HOTPLUG_CPU0 by default - config: disable CONFIG_SATA_ZHAOXIN by default- config/arm64: fix kabi by disable CONFIG_NVME_MULTIPATH- config/x86: enable SHRINK_PAGECACHE - arm64: Add config switch and kernel parameter for CPU0 hotplug- arm64: errata: enable HISILICON_ERRATUM_HIP08_RU_PREFETCH - arm64: errata: fix kabi changed for cpu_errata - arm64: errata: add option to disable cache readunique prefetch on 1620- firewire: nosy: Fix a use-after-free bug in nosy_ioctl()- iommu/arm-smmu-v3: Reduce contention during command-queue insertion - iommu/arm-smmu-v3: Operate directly on low-level queue where possible - iommu/arm-smmu-v3: Move low-level queue fields out of arm_smmu_queue - iommu/arm-smmu-v3: Drop unused 'q' argument from Q_OVF macro - iommu/arm-smmu-v3: Separate s/w and h/w views of prod and cons indexes - iommu/io-pgtable: Rename iommu_gather_ops to iommu_flush_ops - iommu/io-pgtable-arm: Remove redundant call to io_pgtable_tlb_sync() - iommu/arm-smmu-v3: Increase maximum size of queues - iommu/io-pgtable: Replace IO_PGTABLE_QUIRK_NO_DMA with specific flag - iommu: Allow io-pgtable to be used outside of drivers/iommu/ - iommu: Fix flush_tlb_all typo - iommu: Change tlb_range_add to iotlb_range_add and tlb_sync to iotlb_sync- io_uring: order refnode recycling - io_uring: get an active ref_node from files_data - io_uring: fix racy req->flags modification - io_uring: defer file table grabbing request cleanup for locked requests - io_uring: batch put_task_struct() - tasks: add put_task_struct_many() - io_uring: fix missing io_queue_linked_timeout() - io_uring: deduplicate io_grab_files() calls - io_uring: don't do opcode prep twice - io_uring: don't open-code recv kbuf managment - io_uring: extract io_put_kbuf() helper - io_uring: simplify file ref tracking in submission state - io_uring: move BUFFER_SELECT check into *recv[msg] - io_uring: free selected-bufs if error'ed - io_uring: don't forget cflags in io_recv() - io_uring: remove extra checks in send/recv - io_uring: indent left {send,recv}[msg]() - io-wq: update hash bits - io_uring: get rid of atomic FAA for cq_timeouts - io_uring: consolidate *_check_overflow accounting - io_uring: de-unionise io_kiocb - io_uring: follow **iovec idiom in io_import_iovec - io_uring: mark ->work uninitialised after cleanup - io_uring/io-wq: move RLIMIT_FSIZE to io-wq - io_uring: alloc ->io in io_req_defer_prep() - io_uring: inline io_req_work_grab_env() - io_uring: fix racy IOPOLL completions - io_uring: always let io_iopoll_complete() complete polled io - io_uring: don't recurse on tsk->sighand->siglock with signalfd - io_uring: don't use poll handler if file can't be nonblocking read/written - io_uring: fix linked deferred ->files cancellation - io_uring: fix cancel of deferred reqs with ->files - io_uring: flush timeouts that should already have expired - io_uring: find and cancel head link async work on files exit - io_uring: always plug for any number of IOs - io_uring: fix recursive completion locking on oveflow flush - io_uring: enable lookup of links holding inflight files - io_uring: place cflags into completion data - io_uring: remove sequence from io_kiocb - io_uring: use non-intrusive list for defer - io_uring: remove init for unused list - io_uring: add req->timeout.list - io_uring: use completion list for CQ overflow - io_uring: use inflight_entry list for iopoll'ing - io_uring: rename ctx->poll into ctx->iopoll - io_uring: share completion list w/ per-op space - io_uring: get rid of __req_need_defer() - io_uring: only call kfree() for a non-zero pointer - io_uring: fix a use after free in io_async_task_func() - io_uring: remove nr_events arg from iopoll_check() - io_uring: don't delay iopoll'ed req completion - io_uring: fix lost cqe->flags - io_uring: keep queue_sqe()'s fail path separately - io_uring: fix mis-refcounting linked timeouts - io_uring: use new io_req_task_work_add() helper throughout - io_uring: abstract out task work running - io_uring: do grab_env() just before punting - io_uring: factor out grab_env() from defer_prep() - io_uring: do init work in grab_env() - io_uring: don't pass def into io_req_work_grab_env - io_uring: fix function args for !CONFIG_NET - io_uring: set @poll->file after @poll init - io_uring: remove REQ_F_MUST_PUNT - io_uring: remove setting REQ_F_MUST_PUNT in rw - io_uring: optimise io_req_find_next() fast check - io_uring: kill REQ_F_TIMEOUT_NOSEQ - io_uring: kill REQ_F_TIMEOUT - io_uring: replace find_next() out param with ret - io_uring: fix missing io_grab_files() - io_uring: don't mark link's head for_async - io_uring: fix feeding io-wq with uninit reqs - io_uring: fix punting req w/o grabbed env - io_uring: fix req->work corruption - io_uring: simplify io_async_task_func() - io_uring: fix NULL mm in io_poll_task_func() - io_uring: use task_work for links if possible - io_uring: do task_work_run() during iopoll - io_uring: clean up req->result setting by rw - io_uring: cosmetic changes for batch free - io_uring: batch-free linked requests as well - io_uring: dismantle req early and remove need_iter - io_uring: remove inflight batching in free_many() - io_uring: fix refs underflow in io_iopoll_queue() - io_uring: enable READ/WRITE to use deferred completions - io_uring: pass in completion state to appropriate issue side handlers - io_uring: pass down completion state on the issue side - io_uring: add 'io_comp_state' to struct io_submit_state - io_uring: provide generic io_req_complete() helper - io_uring: add missing REQ_F_COMP_LOCKED for nested requests - io_uring: clean up io_kill_linked_timeout() locking - io_uring: deduplicate freeing linked timeouts - io_uring: kill REQ_F_LINK_NEXT - io_uring: fix stalled deferred requests - io_uring: add IORING_OP_OPENAT2 for compatablity - arm64: fix kabi with io_uring interface - x86: fix kabi with io_uring interface - io_uring: fix provide_buffers sign extension - io_uring: ignore double poll add on the same waitqueue head - io_uring: fix SQPOLL IORING_OP_CLOSE cancelation state - io_uring: make ctx cancel on exit targeted to actual ctx - io_uring: fix error path cleanup in io_sqe_files_register() - io_uring: ensure open/openat2 name is cleaned on cancelation - io_uring: sanitize double poll handling - io_uring: fail poll arm on queue proc failure - io_uring: allow non-fixed files with SQPOLL - io_uring: ensure consistent view of original task ->mm from SQPOLL - io_uring: stash ctx task reference for SQPOLL - io_uring: don't miscount pinned memory - io_uring: don't burn CPU for iopoll on exit - io_uring: fix imbalanced sqo_mm accounting - io_uring: return locked and pinned page accounting - io_uring: fix missing ->mm on exit - io_uring: fix NULL-mm for linked reqs - io_uring: account locked memory before potential error case - io_uring: don't touch 'ctx' after installing file descriptor - io_uring: remove dead 'ctx' argument and move forward declaration - io_uring: fix recvmsg setup with compat buf-select - io_uring: fix shift-out-of-bounds when round up cq size - io_uring: round-up cq size before comparing with rounded sq size - io_uring: use type appropriate io_kiocb handler for double poll - io_uring: fix double poll mask init - io_uring: Fix sizeof() mismatch - io_uring: keep a pointer ref_node in file_data - io_uring: refactor *files_register()'s error paths - io_uring: clean file_data access in files_register - io-wq: fix use-after-free in io_wq_worker_running - io_uring: fix potential ABBA deadlock in ->show_fdinfo() - io_uring: always delete double poll wait entry on match - io-wq: fix hang after cancelling pending hashed work - io_uring: fix racy overflow count reporting - io_uring: partially inline io_iopoll_getevents() - io_uring: briefly loose locks while reaping events - io_uring: fix stopping iopoll'ing too early - io_uring: fix potential use after free on fallback request free - io_uring: set table->files[i] to NULL when io_sqe_file_register failed - io_uring: fix removing the wrong file in __io_sqe_files_update() - io_uring: fix IOPOLL -EAGAIN retries - io_uring: clear req->result on IOPOLL re-issue - io_uring: hold 'ctx' reference around task_work queue + execute - io_uring: use TWA_SIGNAL for task_work uncondtionally - io_uring: Fix NULL pointer dereference in loop_rw_iter() - io_uring: clear IORING_SQ_NEED_WAKEUP after executing task works - io_uring: add a helper for async rw iovec prep - io_uring: simplify io_req_map_rw() - io_uring: extract io_sendmsg_copy_hdr() - io_uring: use more specific type in rcv/snd msg cp - io_uring: rename sr->msg into umsg - io_uring: fix sq array offset calculation - io_uring: fix lockup in io_fail_links() - io_uring: fix ->work corruption with poll_add - io_uring: missed req_init_async() for IOSQE_ASYNC - io_uring: always allow drain/link/hardlink/async sqe flags - io_uring: ensure double poll additions work with both request types - io_uring: fix recvmsg memory leak with buffer selection - io_uring: fix not initialised work->flags - io_uring: fix missing msg_name assignment - io_uring: account user memory freed when exit has been queued - io_uring: fix memleak in io_sqe_files_register() - io_uring: fix memleak in __io_sqe_files_update() - io_uring: export cq overflow status to userspace - io_uring: fix regression with always ignoring signals in io_cqring_wait() - io_uring: use signal based task_work running - task_work: teach task_work_add() to do signal_wake_up() - io_uring: fix current->mm NULL dereference on exit - io_uring: fix hanging iopoll in case of -EAGAIN - io_uring: fix io_sq_thread no schedule when busy - io-wq: return next work from ->do_work() directly - io-wq: compact io-wq flags numbers - io_uring: separate reporting of ring pages from registered pages - io_uring: report pinned memory usage - io_uring: rename ctx->account_mem field - io_uring: add wrappers for memory accounting - io_uring: use EPOLLEXCLUSIVE flag to aoid thundering herd type behavior - io_uring: change the poll type to be 32-bits - io_uring: fix possible race condition against REQ_F_NEED_CLEANUP - io_uring: reap poll completions while waiting for refs to drop on exit - io_uring: acquire 'mm' for task_work for SQPOLL - io_uring: add memory barrier to synchronize io_kiocb's result and iopoll_completed - io_uring: don't fail links for EAGAIN error in IOPOLL mode - io_uring: cancel by ->task not pid - io_uring: lazy get task - io_uring: batch cancel in io_uring_cancel_files() - io_uring: cancel all task's requests on exit - io-wq: add an option to cancel all matched reqs - io-wq: reorder cancellation pending -> running - io_uring: fix lazy work init - io_uring: fix io_kiocb.flags modification race in IOPOLL mode - io_uring: check file O_NONBLOCK state for accept - io_uring: avoid unnecessary io_wq_work copy for fast poll feature - io_uring: avoid whole io_wq_work copy for requests completed inline - io_uring: allow O_NONBLOCK async retry - io_wq: add per-wq work handler instead of per work - io_uring: don't arm a timeout through work.func - io_uring: remove custom ->func handlers - io_uring: don't derive close state from ->func - io_uring: use kvfree() in io_sqe_buffer_register() - io_uring: validate the full range of provided buffers for access - io_uring: re-set iov base/len for buffer select retry - io_uring: move send/recv IOPOLL check into prep - io_uring: fix {SQ,IO}POLL with unsupported opcodes - io_uring: disallow close of ring itself - io_uring: fix overflowed reqs cancellation - io_uring: off timeouts based only on completions - io_uring: move timeouts flushing to a helper - statx: hide interfaces no longer used by io_uring - io_uring: call statx directly - statx: allow system call to be invoked from io_uring - io_uring: add io_statx structure - io_uring: get rid of manual punting in io_close - io_uring: separate DRAIN flushing into a cold path - io_uring: don't re-read sqe->off in timeout_prep() - io_uring: simplify io_timeout locking - io_uring: fix flush req->refs underflow - io_uring: don't submit sqes when ctx->refs is dying - io_uring: async task poll trigger cleanup - io_uring: add tee(2) support - splice: export do_tee() - io_uring: don't repeat valid flag list - io_uring: rename io_file_put() - io_uring: remove req->needs_fixed_files - io_uring: cleanup io_poll_remove_one() logic - io_uring: file registration list and lock optimization - io_uring: add IORING_CQ_EVENTFD_DISABLED to the CQ ring flags - io_uring: add 'cq_flags' field for the CQ ring - io_uring: allow POLL_ADD with double poll_wait() users - io_uring: batch reap of dead file registrations - io_uring: name sq thread and ref completions - io_uring: remove duplicate semicolon at the end of line - io_uring: remove obsolete 'state' parameter - io_uring: remove 'fd is io_uring' from close path - io_uring: reset -EBUSY error when io sq thread is waken up - io_uring: don't add non-IO requests to iopoll pending list - io_uring: don't use kiocb.private to store buf_index - io_uring: cancel work if task_work_add() fails - io_uring: remove dead check in io_splice() - io_uring: fix FORCE_ASYNC req preparation - io_uring: don't prepare DRAIN reqs twice - io_uring: initialize ctx->sqo_wait earlier - io_uring: polled fixed file must go through free iteration - io_uring: fix zero len do_splice() - io_uring: don't use 'fd' for openat/openat2/statx - splice: move f_mode checks to do_{splice,tee}() - io_uring: handle -EFAULT properly in io_uring_setup() - io_uring: fix mismatched finish_wait() calls in io_uring_cancel_files() - io_uring: punt splice async because of inode mutex - io_uring: check non-sync defer_list carefully - io_uring: fix extra put in sync_file_range() - io_uring: use cond_resched() in io_ring_ctx_wait_and_kill() - io_uring: use proper references for fallback_req locking - io_uring: only force async punt if poll based retry can't handle it - io_uring: enable poll retry for any file with ->read_iter / ->write_iter - io_uring: statx must grab the file table for valid fd - io_uring: only restore req->work for req that needs do completion - io_uring: don't count rqs failed after current one - io_uring: kill already cached timeout.seq_offset - io_uring: fix cached_sq_head in io_timeout() - io_uring: only post events in io_poll_remove_all() if we completed some - io_uring: io_async_task_func() should check and honor cancelation - io_uring: check for need to re-wait in polled async handling - io_uring: correct O_NONBLOCK check for splice punt - io_uring: restore req->work when canceling poll request - io_uring: move all request init code in one place - io_uring: keep all sqe->flags in req->flags - io_uring: early submission req fail code - io_uring: track mm through current->mm - io_uring: remove obsolete @mm_fault - io_uring: punt final io_ring_ctx wait-and-free to workqueue - io_uring: fix fs cleanup on cqe overflow - io_uring: don't read user-shared sqe flags twice - io_uring: remove req init from io_get_req() - io_uring: alloc req only after getting sqe - io_uring: simplify io_get_sqring - io_uring: do not always copy iovec in io_req_map_rw() - io_uring: ensure openat sets O_LARGEFILE if needed - io_uring: initialize fixed_file_data lock - io_uring: remove redundant variable pointer nxt and io_wq_assign_next call - io_uring: fix ctx refcounting in io_submit_sqes() - io_uring: process requests completed with -EAGAIN on poll list - io_uring: remove bogus RLIMIT_NOFILE check in file registration - io_uring: use io-wq manager as backup task if task is exiting - io_uring: grab task reference for poll requests - io_uring: retry poll if we got woken with non-matching mask - io_uring: add missing finish_wait() in io_sq_thread() - io_uring: refactor file register/unregister/update handling - io_uring: cleanup io_alloc_async_ctx() - io_uring: fix missing 'return' in comment - io-wq: handle hashed writes in chains - io-uring: drop 'free_pfile' in struct io_file_put - io-uring: drop completion when removing file - io_uring: Fix ->data corruption on re-enqueue - io-wq: close cancel gap for hashed linked work - io_uring: make spdxcheck.py happy - io_uring: honor original task RLIMIT_FSIZE - io-wq: hash dependent work - io-wq: split hashing and enqueueing - io-wq: don't resched if there is no work - io-wq: remove duplicated cancel code - io_uring: fix truncated async read/readv and write/writev retry - io_uring: dual license io_uring.h uapi header - io_uring: io_uring_enter(2) don't poll while SETUP_IOPOLL|SETUP_SQPOLL enabled - io_uring: Fix unused function warnings - io_uring: add end-of-bits marker and build time verify it - io_uring: provide means of removing buffers - io_uring: add IOSQE_BUFFER_SELECT support for IORING_OP_RECVMSG - net: abstract out normal and compat msghdr import - io_uring: add IOSQE_BUFFER_SELECT support for IORING_OP_READV - io_uring: support buffer selection for OP_READ and OP_RECV - io_uring: add IORING_OP_PROVIDE_BUFFERS - io_uring: buffer registration infrastructure - io_uring/io-wq: forward submission ref to async - io-wq: optimise out *next_work() double lock - io-wq: optimise locking in io_worker_handle_work() - io-wq: shuffle io_worker_handle_work() code - io_uring: get next work with submission ref drop - io_uring: remove @nxt from handlers - io_uring: make submission ref putting consistent - io_uring: clean up io_close - io_uring: Ensure mask is initialized in io_arm_poll_handler - io_uring: remove io_prep_next_work() - io_uring: remove extra nxt check after punt - io_uring: use poll driven retry for files that support it - io_uring: mark requests that we can do poll async in io_op_defs - io_uring: add per-task callback handler - io_uring: store io_kiocb in wait->private - task_work_run: don't take ->pi_lock unconditionally - io-wq: use BIT for ulong hash - io_uring: remove IO_WQ_WORK_CB - io-wq: remove unused IO_WQ_WORK_HAS_MM - io_uring: extract kmsg copy helper - io_uring: clean io_poll_complete - io_uring: add splice(2) support - io_uring: add interface for getting files - splice: make do_splice public - io_uring: remove req->in_async - io_uring: don't do full *prep_worker() from io-wq - io_uring: don't call work.func from sync ctx - io_uring: io_accept() should hold on to submit reference on retry - io_uring: consider any io_read/write -EAGAIN as final - io_uring: make sure accept honor rlimit nofile - io_uring: make sure openat/openat2 honor rlimit nofile - io_uring: NULL-deref for IOSQE_{ASYNC,DRAIN} - io_uring: ensure RCU callback ordering with rcu_barrier() - io_uring: fix lockup with timeouts - io_uring: free fixed_file_data after RCU grace period - io-wq: remove io_wq_flush and IO_WQ_WORK_INTERNAL - io-wq: fix IO_WQ_WORK_NO_CANCEL cancellation - io_uring: fix 32-bit compatability with sendmsg/recvmsg - io_uring: define and set show_fdinfo only if procfs is enabled - io_uring: drop file set ref put/get on switch - io_uring: import_single_range() returns 0/-ERROR - io_uring: pick up link work on submit reference drop - io-wq: ensure work->task_pid is cleared on init - io-wq: remove spin-for-work optimization - io_uring: fix poll_list race for SETUP_IOPOLL|SETUP_SQPOLL - io_uring: fix personality idr leak - io_uring: handle multiple personalities in link chains - io_uring: fix __io_iopoll_check deadlock in io_sq_thread - io_uring: prevent sq_thread from spinning when it should stop - io_uring: fix use-after-free by io_cleanup_req() - io_uring: remove unnecessary NULL checks - io_uring: add missing io_req_cancelled() - io_uring: prune request from overflow list on flush - io-wq: don't call kXalloc_node() with non-online node - io_uring: retain sockaddr_storage across send/recvmsg async punt - io_uring: cancel pending async work if task exits - io-wq: add io_wq_cancel_pid() to cancel based on a specific pid - io-wq: make io_wqe_cancel_work() take a match handler - io_uring: fix openat/statx's filename leak - io_uring: fix double prep iovec leak - io_uring: fix async close() with f_op->flush() - io_uring: allow AT_FDCWD for non-file openat/openat2/statx - io_uring: grab ->fs as part of async preparation - io-wq: add support for inheriting ->fs - io_uring: retry raw bdev writes if we hit -EOPNOTSUPP - io_uring: add cleanup for openat()/statx() - io_uring: fix iovec leaks - io_uring: remove unused struct io_async_open - io_uring: flush overflowed CQ events in the io_uring_poll() - io_uring: statx/openat/openat2 don't support fixed files - io_uring: fix deferred req iovec leak - io_uring: fix 1-bit bitfields to be unsigned - io_uring: get rid of delayed mm check - io_uring: cleanup fixed file data table references - io_uring: spin for sq thread to idle on shutdown - io_uring: put the flag changing code in the same spot - io_uring: iterate req cache backwards - io_uring: punt even fadvise() WILLNEED to async context - io_uring: fix sporadic double CQE entry for close - io_uring: remove extra ->file check - io_uring: don't map read/write iovec potentially twice - io_uring: use the proper helpers for io_send/recv - io_uring: prevent potential eventfd recursion on poll - io_uring: add BUILD_BUG_ON() to assert the layout of struct io_uring_sqe - io_uring: add ->show_fdinfo() for the io_uring file descriptor - io_uring: add support for epoll_ctl(2) - eventpoll: support non-blocking do_epoll_ctl() calls - eventpoll: abstract out epoll_ctl() handler - io_uring: fix linked command file table usage - io_uring: support using a registered personality for commands - io_uring: allow registering credentials - io_uring: add io-wq workqueue sharing - io-wq: allow grabbing existing io-wq - io_uring/io-wq: don't use static creds/mm assignments - io-wq: make the io_wq ref counted - io_uring: fix refcounting with batched allocations at OOM - io_uring: add comment for drain_next - io_uring: don't attempt to copy iovec for READ/WRITE - io_uring: honor IOSQE_ASYNC for linked reqs - io_uring: prep req when do IOSQE_ASYNC - io_uring: use labeled array init in io_op_defs - io_uring: optimise sqe-to-req flags translation - io_uring: remove REQ_F_IO_DRAINED - io_uring: file switch work needs to get flushed on exit - io_uring: hide uring_fd in ctx - io_uring: remove extra check in __io_commit_cqring - io_uring: optimise use of ctx->drain_next - io_uring: add support for probing opcodes - io_uring: account fixed file references correctly in batch - io_uring: add opcode to issue trace event - io_uring: remove 'fname' from io_open structure - io_uring: enable option to only trigger eventfd for async completions - io_uring: change io_ring_ctx bool fields into bit fields - io_uring: file set registration should use interruptible waits - io_uring: Remove unnecessary null check - io_uring: add support for send(2) and recv(2) - io_uring: remove extra io_wq_current_is_worker() - io_uring: optimise commit_sqring() for common case - io_uring: optimise head checks in io_get_sqring() - io_uring: clamp to_submit in io_submit_sqes() - io_uring: add support for IORING_SETUP_CLAMP - io_uring: extend batch freeing to cover more cases - io_uring: wrap multi-req freeing in struct req_batch - io_uring: batch getting pcpu references - pcpu_ref: add percpu_ref_tryget_many() - io_uring: add IORING_OP_MADVISE - mm: make do_madvise() available internally - io_uring: add IORING_OP_FADVISE - io_uring: allow use of offset == -1 to mean file position - io_uring: add non-vectored read/write commands - io_uring: improve poll completion performance - io_uring: split overflow state into SQ and CQ side - io_uring: add lookup table for various opcode needs - io_uring: remove two unnecessary function declarations - io_uring: move *queue_link_head() from common path - io_uring: rename prev to head - io_uring: add IOSQE_ASYNC - io-wq: support concurrent non-blocking work - io_uring: add support for IORING_OP_STATX - fs: make two stat prep helpers available - io_uring: avoid ring quiesce for fixed file set unregister and update - io_uring: add support for IORING_OP_CLOSE - io-wq: add support for uncancellable work - percpu-refcount: Introduce percpu_ref_resurrect() - percpu_ref: introduce PERCPU_REF_ALLOW_REINIT flag - fs: make filename_lookup available externally - fs: introduce __close_fd_get_file to support IORING_OP_CLOSE for io_uring - io_uring: add support for IORING_OP_OPENAT - fs: make build_open_flags() available internally - io_uring: add support for fallocate() - io_uring: don't cancel all work on process exit - Revert "io_uring: only allow submit from owning task" - io_uring: fix compat for IORING_REGISTER_FILES_UPDATE - io_uring: only allow submit from owning task - io_uring: ensure workqueue offload grabs ring mutex for poll list - io_uring: clear req->result always before issuing a read/write request - io_uring: be consistent in assigning next work from handler - io-wq: cancel work if we fail getting a mm reference - io_uring: don't setup async context for read/write fixed - io_uring: remove punt of short reads to async context - io-wq: add cond_resched() to worker thread - io-wq: remove unused busy list from io_sqe - io_uring: pass in 'sqe' to the prep handlers - io_uring: standardize the prep methods - io_uring: read 'count' for IORING_OP_TIMEOUT in prep handler - io_uring: move all prep state for IORING_OP_{SEND,RECV}_MGS to prep handler - io_uring: move all prep state for IORING_OP_CONNECT to prep handler - io_uring: add and use struct io_rw for read/writes - io_uring: use u64_to_user_ptr() consistently - io_uring: io_wq_submit_work() should not touch req->rw - io_uring: don't wait when under-submitting - io_uring: warn about unhandled opcode - io_uring: read opcode and user_data from SQE exactly once - io_uring: make IORING_OP_TIMEOUT_REMOVE deferrable - io_uring: make IORING_OP_CANCEL_ASYNC deferrable - io_uring: make IORING_POLL_ADD and IORING_POLL_REMOVE deferrable - io_uring: make HARDLINK imply LINK - io_uring: any deferred command must have stable sqe data - io_uring: remove 'sqe' parameter to the OP helpers that take it - io_uring: fix pre-prepped issue with force_nonblock == true - io-wq: re-add io_wq_current_is_worker() - io_uring: fix sporadic -EFAULT from IORING_OP_RECVMSG - io_uring: fix stale comment and a few typos - io_uring: ensure we return -EINVAL on unknown opcode - io_uring: add sockets to list of files that support non-blocking issue - io_uring: only hash regular files for async work execution - io_uring: run next sqe inline if possible - io_uring: don't dynamically allocate poll data - io_uring: deferred send/recvmsg should assign iov - io_uring: sqthread should grab ctx->uring_lock for submissions - io-wq: briefly spin for new work after finishing work - io-wq: remove worker->wait waitqueue - io_uring: allow unbreakable links - io_uring: fix a typo in a comment - io_uring: hook all linked requests via link_list - io_uring: fix error handling in io_queue_link_head - io_uring: use hash table for poll command lookups - io-wq: clear node->next on list deletion - io_uring: ensure deferred timeouts copy necessary data - io_uring: allow IO_SQE_* flags on IORING_OP_TIMEOUT - io_uring: handle connect -EINPROGRESS like -EAGAIN - io_uring: remove io_wq_current_is_worker - io_uring: remove parameter ctx of io_submit_state_start - io_uring: mark us with IORING_FEAT_SUBMIT_STABLE - io_uring: ensure async punted connect requests copy data - io_uring: ensure async punted sendmsg/recvmsg requests copy data - net: disallow ancillary data for __sys_{send,recv}msg_file() - net: separate out the msghdr copy from ___sys_{send,recv}msg() - io_uring: ensure async punted read/write requests copy iovec - io_uring: add general async offload context - io_uring: transform send/recvmsg() -ERESTARTSYS to -EINTR - io_uring: use current task creds instead of allocating a new one - io_uring: fix missing kmap() declaration on powerpc - io_uring: add mapping support for NOMMU archs - io_uring: make poll->wait dynamically allocated - io-wq: shrink io_wq_work a bit - io-wq: fix handling of NUMA node IDs - io_uring: use kzalloc instead of kcalloc for single-element allocations - io_uring: cleanup io_import_fixed() - io_uring: inline struct sqe_submit - io_uring: store timeout's sqe->off in proper place - io_uring: remove superfluous check for sqe->off in io_accept() - io_uring: async workers should inherit the user creds - io-wq: have io_wq_create() take a 'data' argument - io_uring: fix dead-hung for non-iter fixed rw - io_uring: add support for IORING_OP_CONNECT - net: add __sys_connect_file() helper - io_uring: only return -EBUSY for submit on non-flushed backlog - io_uring: only !null ptr to io_issue_sqe() - io_uring: simplify io_req_link_next() - io_uring: pass only !null to io_req_find_next() - io_uring: remove io_free_req_find_next() - io_uring: add likely/unlikely in io_get_sqring() - io_uring: rename __io_submit_sqe() - io_uring: improve trace_io_uring_defer() trace point - io_uring: drain next sqe instead of shadowing - io_uring: close lookup gap for dependent next work - io_uring: allow finding next link independent of req reference count - io_uring: io_allocate_scq_urings() should return a sane state - io_uring: Always REQ_F_FREE_SQE for allocated sqe - io_uring: io_fail_links() should only consider first linked timeout - io_uring: Fix leaking linked timeouts - io_uring: remove redundant check - io_uring: break links for failed defer - io-wq: remove extra space characters - io-wq: wait for io_wq_create() to setup necessary workers - io_uring: request cancellations should break links - io_uring: correct poll cancel and linked timeout expiration completion - io_uring: remove dead REQ_F_SEQ_PREV flag - io_uring: fix sequencing issues with linked timeouts - io_uring: make req->timeout be dynamically allocated - io_uring: make io_double_put_req() use normal completion path - io_uring: cleanup return values from the queueing functions - io_uring: io_async_cancel() should pass in 'nxt' request pointer - io_uring: make POLL_ADD/POLL_REMOVE scale better - io-wq: remove now redundant struct io_wq_nulls_list - io_uring: Fix getting file for non-fd opcodes - io_uring: introduce req_need_defer() - io_uring: clean up io_uring_cancel_files() - io-wq: ensure free/busy list browsing see all items - io_uring: ensure registered buffer import returns the IO length - io-wq: ensure we have a stable view of ->cur_work for cancellations - io_wq: add get/put_work handlers to io_wq_create() - io_uring: Fix getting file for timeout - io_uring: check for validity of ->rings in teardown - io_uring: fix potential deadlock in io_poll_wake() - io_uring: use correct "is IO worker" helper - io_uring: make timeout sequence == 0 mean no sequence - io_uring: fix -ENOENT issue with linked timer with short timeout - io_uring: don't do flush cancel under inflight_lock - io_uring: flag SQPOLL busy condition to userspace - io_uring: make ASYNC_CANCEL work with poll and timeout - io_uring: provide fallback request for OOM situations - io_uring: convert accept4() -ERESTARTSYS into -EINTR - io_uring: fix error clear of ->file_table in io_sqe_files_register() - io_uring: separate the io_free_req and io_free_req_find_next interface - io_uring: keep io_put_req only responsible for release and put req - io_uring: remove passed in 'ctx' function parameter ctx if possible - io_uring: reduce/pack size of io_ring_ctx - io_uring: properly mark async work as bounded vs unbounded - io-wq: add support for bounded vs unbunded work - io-wq: io_wqe_run_queue() doesn't need to use list_empty_careful() - io_uring: add support for backlogged CQ ring - io_uring: pass in io_kiocb to fill/add CQ handlers - io_uring: make io_cqring_events() take 'ctx' as argument - io_uring: add support for linked SQE timeouts - io_uring: abstract out io_async_cancel_one() helper - io_uring: use inlined struct sqe_submit - io_uring: Use submit info inlined into req - io_uring: allocate io_kiocb upfront - io_uring: io_queue_link*() right after submit - io_uring: Merge io_submit_sqes and io_ring_submit - io_uring: kill dead REQ_F_LINK_DONE flag - io_uring: fixup a few spots where link failure isn't flagged - io_uring: enable optimized link handling for IORING_OP_POLL_ADD - io-wq: use proper nesting IRQ disabling spinlocks for cancel - io_uring: add completion trace event - io-wq: use kfree_rcu() to simplify the code - io_uring: set -EINTR directly when a signal wakes up in io_cqring_wait - io_uring: support for generic async request cancel - io_uring: ensure we clear io_kiocb->result before each issue - io_uring: io_wq_create() returns an error pointer, not NULL - io_uring: fix race with canceling timeouts - io_uring: support for larger fixed file sets - io_uring: protect fixed file indexing with array_index_nospec() - io_uring: add support for IORING_OP_ACCEPT - net: add __sys_accept4_file() helper - io_uring: io_uring: add support for async work inheriting files - io_uring: replace workqueue usage with io-wq - io-wq: small threadpool implementation for io_uring - sched/core, workqueues: Distangle worker accounting from rq lock - sched: Remove stale PF_MUTEX_TESTER bit - io_uring: Fix mm_fault with READ/WRITE_FIXED - io_uring: remove index from sqe_submit - io_uring: add set of tracing events - io_uring: add support for canceling timeout requests - io_uring: add support for absolute timeouts - io_uring: replace s->needs_lock with s->in_async - io_uring: allow application controlled CQ ring size - io_uring: add support for IORING_REGISTER_FILES_UPDATE - io_uring: allow sparse fixed file sets - io_uring: run dependent links inline if possible - io_uring: don't touch ctx in setup after ring fd install - io_uring: Fix leaked shadow_req - io_uring: fix bad inflight accounting for SETUP_IOPOLL|SETUP_SQTHREAD - io_uring: used cached copies of sq->dropped and cq->overflow - io_uring: Fix race for sqes with userspace - io_uring: Fix broken links with offloading - io_uring: Fix corrupted user_data - io_uring: correct timeout req sequence when inserting a new entry - io_uring : correct timeout req sequence when waiting timeout - io_uring: revert "io_uring: optimize submit_and_wait API" - io_uring: fix logic error in io_timeout - io_uring: fix up O_NONBLOCK handling for sockets - io_uring: consider the overflow of sequence for timeout req - io_uring: fix sequence logic for timeout requests - io_uring: only flush workqueues on fileset removal - io_uring: remove wait loop spurious wakeups - io_uring: fix reversed nonblock flag for link submission - io_uring: use __kernel_timespec in timeout ABI - io_uring: make CQ ring wakeups be more efficient - io_uring: compare cached_cq_tail with cq.head in_io_uring_poll - io_uring: correctly handle non ->{read,write}_iter() file_operations - io_uring: IORING_OP_TIMEOUT support - io_uring: use cond_resched() in sqthread - io_uring: fix potential crash issue due to io_get_req failure - io_uring: ensure poll commands clear ->sqe - io_uring: fix use-after-free of shadow_req - io_uring: use kmemdup instead of kmalloc and memcpy - io_uring: increase IORING_MAX_ENTRIES to 32K - io_uring: make sqpoll wakeup possible with getevents - io_uring: extend async work merging - io_uring: limit parallelism of buffered writes - io_uring: add io_queue_async_work() helper - io_uring: optimize submit_and_wait API - io_uring: add support for link with drain - io_uring: fix wrong sequence setting logic - io_uring: expose single mmap capability - io_uring: allocate the two rings together - io_uring: add need_resched() check in inner poll loop - io_uring: don't enter poll loop if we have CQEs pending - io_uring: fix potential hang with polled IO - io_uring: fix an issue when IOSQE_IO_LINK is inserted into defer list - io_uring: fix manual setup of iov_iter for fixed buffers - io_uring: fix KASAN use after free in io_sq_wq_submit_work - io_uring: ensure ->list is initialized for poll commands - io_uring: track io length in async_list based on bytes - io_uring: don't use iov_iter_advance() for fixed buffers - io_uring: add a memory barrier before atomic_read - io_uring: fix counter inc/dec mismatch in async_list - io_uring: fix the sequence comparison in io_sequence_defer - io_uring: fix io_sq_thread_stop running in front of io_sq_thread - io_uring: add support for recvmsg() - io_uring: add support for sendmsg() - io_uring: add support for sqe links - io_uring: punt short reads to async context - uio: make import_iovec()/compat_import_iovec() return bytes on success - io_uring: ensure req->file is cleared on allocation - io_uring: fix memory leak of UNIX domain socket inode - io_uring: Fix __io_uring_register() false success - tools/io_uring: sync with liburing - tools/io_uring: fix Makefile for pthread library link - select: shift restore_saved_sigmask_unless() into poll_select_copy_remaining() - select: change do_poll() to return -ERESTARTNOHAND rather than -EINTR - signal: simplify set_user_sigmask/restore_user_sigmask - signal: remove the wrong signal_pending() check in restore_user_sigmask() - io_uring: use wait_event_interruptible for cq_wait conditional wait - io_uring: adjust smp_rmb inside io_cqring_events - io_uring: fix infinite wait in khread_park() on io_finish_async() - io_uring: remove 'ev_flags' argument - io_uring: fix failure to verify SQ_AFF cpu - io_uring: fix race condition reading SQE data - io_uring: use cpu_online() to check p->sq_thread_cpu instead of cpu_possible() - io_uring: fix shadowed variable ret return code being not checked - req->error only used for iopoll - io_uring: add support for eventfd notifications - io_uring: add support for IORING_OP_SYNC_FILE_RANGE - io_uring: add support for marking commands as draining - fs: add sync_file_range() helper - io_uring: avoid page allocation warnings - io_uring: drop req submit reference always in async punt - io_uring: free allocated io_memory once - io_uring: fix SQPOLL cpu validation - io_uring: have submission side sqe errors post a cqe - io_uring: remove unnecessary barrier after unsetting IORING_SQ_NEED_WAKEUP - io_uring: remove unnecessary barrier after incrementing dropped counter - io_uring: remove unnecessary barrier before reading SQ tail - io_uring: remove unnecessary barrier after updating SQ head - io_uring: remove unnecessary barrier before reading cq head - io_uring: remove unnecessary barrier before wq_has_sleeper - io_uring: fix notes on barriers - io_uring: fix handling SQEs requesting NOWAIT - io_uring: remove 'state' argument from io_{read,write} path - io_uring: fix poll full SQ detection - io_uring: fix race condition when sq threads goes sleeping - io_uring: fix race condition reading SQ entries - io_uring: fail io_uring_register(2) on a dying io_uring instance - io_uring: fix CQ overflow condition - io_uring: fix possible deadlock between io_uring_{enter,register} - io_uring: drop io_file_put() 'file' argument - io_uring: only test SQPOLL cpu after we've verified it - io_uring: park SQPOLL thread if it's percpu - io_uring: restrict IORING_SETUP_SQPOLL to root - io_uring: fix double free in case of fileset regitration failure - io_uring: offload write to async worker in case of -EAGAIN - io_uring: fix big-endian compat signal mask handling - io_uring: retry bulk slab allocs as single allocs - io_uring: fix poll races - io_uring: fix fget/fput handling - io_uring: add prepped flag - io_uring: make io_read/write return an integer - io_uring: use regular request ref counts - tools/io_uring: remove IOCQE_FLAG_CACHEHIT - io_uring: add a few test tools - io_uring: allow workqueue item to handle multiple buffered requests - io_uring: add support for IORING_OP_POLL - io_uring: add io_kiocb ref count - io_uring: add submission polling - io_uring: add file set registration - net: split out functions related to registering inflight socket files - io_uring: add support for pre-mapped user IO buffers - io_uring: batch io_kiocb allocation - io_uring: use fget/fput_many() for file references - fs: add fget_many() and fput_many() - io_uring: support for IO polling - io_uring: add fsync support - Add io_uring IO interface - io_pgetevents: use __kernel_timespec - pselect6: use __kernel_timespec - ppoll: use __kernel_timespec - signal: Add restore_user_sigmask() - signal: Add set_user_sigmask() - block: Initialize BIO I/O priority early - block: prevent merging of requests with different priorities - aio: Fix fallback I/O priority value - block: Introduce get_current_ioprio() - aio: Comment use of IOCB_FLAG_IOPRIO aio flag - fs: fix kabi change since add iopoll - fs: add an iopoll method to struct file_operations - signal: Allow cifs and drbd to receive their terminating signals - cifs: fix rmmod regression in cifs.ko caused by force_sig changes - signal/cifs: Fix cifs_put_tcp_session to call send_sig instead of force_sig- bpf, x86: Validate computation of branch displacements for x86-32 - bpf, x86: Validate computation of branch displacements for x86-64- mm/vmalloc.c: fix percpu free VM area search criteria - mm/vmalloc.c: avoid bogus -Wmaybe-uninitialized warning - mm/vmap: add DEBUG_AUGMENT_LOWEST_MATCH_CHECK macro - mm/vmap: add DEBUG_AUGMENT_PROPAGATE_CHECK macro - mm/vmalloc.c: keep track of free blocks for vmap allocation- config: Enable CONFIG_USERSWAP - userswap: support userswap via userfaultfd - userswap: add a new flag 'MAP_REPLACE' for mmap() - mm, mempolicy: fix up gup usage in lookup_node - mm/mempolicy: Allow lookup_node() to handle fatal signal - mm/gup: Let __get_user_pages_locked() return -EINTR for fatal signal - mm/gup: fix fixup_user_fault() on multiple retries - mm/gup: allow VM_FAULT_RETRY for multiple times - mm: allow VM_FAULT_RETRY for multiple times- sched/fair: fix kabi broken due to adding fields in rq and sched_domain_shared - sched/fair: fix try_steal compile error - config: enable CONFIG_SCHED_STEAL by default - sched/fair: introduce SCHED_STEAL - disable stealing by default - sched/fair: Provide idle search schedstats - sched/fair: disable stealing if too many NUMA nodes - sched/fair: Steal work from an overloaded CPU when CPU goes idle - sched/fair: Provide can_migrate_task_llc - sched/fair: Generalize the detach_task interface - sched/fair: Hoist idle_stamp up from idle_balance - sched/fair: Dynamically update cfs_overload_cpus - sched/topology: Provide cfs_overload_cpus bitmap - sched/topology: Provide hooks to allocate data shared per LLC - sched: Provide sparsemask, a reduced contention bitmap- sched/fair: fix kabi broken due to adding idle_h_nr_running in cfs_rq - sched/fair: Make sched-idle CPU selection consistent throughout - sched/fair: Optimize select_idle_cpu - sched/fair: Fall back to sched-idle CPU if idle CPU isn't found - sched/fair: Start tracking SCHED_IDLE tasks count in cfs_rq - sched/core: Create task_has_idle_policy() helper- ext4: add reclaim checks to xattr code - locking/mutex: Fix non debug version of mutex_lock_io_nested() - dm verity: add root hash pkcs#7 signature verification - Revert "netfilter: x_tables: Update remaining dereference to RCU" - netfilter: x_tables: Use correct memory barriers. - Revert "netfilter: x_tables: Switch synchronization to RCU" - arm64: kdump: update ppos when reading elfcorehdr - netfilter: ctnetlink: fix dump of the expect mask attribute - dm ioctl: fix out of bounds array access when no devices - block: Suppress uevent for hidden device when removed - NFS: Correct size calculation for create reply length - cifs: Fix preauth hash corruption - ext4: do not try to set xattr into ea_inode if value is empty - kernel, fs: Introduce and use set_restart_fn() and arch_set_restart_data() - nvme-rdma: fix possible hang when failing to set io queues - sunrpc: fix refcount leak for rpc auth modules - include/linux/sched/mm.h: use rcu_dereference in in_vfork() - hrtimer: Update softirq_expires_next correctly after __hrtimer_get_next_event() - scsi: target: core: Prevent underflow for service actions - scsi: target: core: Add cmd length set before cmd complete - PCI: Fix pci_register_io_range() memory leak - Revert "mm, slub: consider rest of partial list if acquire_slab() fails" - cifs: return proper error code in statfs(2) - tcp: add sanity tests to TCP_QUEUE_SEQ - tcp: annotate tp->write_seq lockless reads - tcp: annotate tp->copied_seq lockless reads - netfilter: x_tables: gpf inside xt_find_revision() - net: Fix gro aggregation for udp encaps with zero csum - dm table: fix zoned iterate_devices based device capability checks - dm table: fix DAX iterate_devices based device capability checks - dm table: fix iterate_devices based device capability checks - dm bufio: subtract the number of initial sectors in dm_bufio_get_device_size - swap: fix swapfile read/write offset - mm/hugetlb.c: fix unnecessary address expansion of pmd sharing - net: fix up truesize of cloned skb in skb_prepare_for_shift() - xfs: Fix assert failure in xfs_setattr_size() - arm64 module: set plt* section addresses to 0x0 - hugetlb: fix update_and_free_page contig page struct assumption - net: icmp: pass zeroed opts from icmp{,v6}_ndo_send before sending - ipv6: silence compilation warning for non-IPV6 builds - ipv6: icmp6: avoid indirect call for icmpv6_send() - xfrm: interface: use icmp_ndo_send helper - sunvnet: use icmp_ndo_send helper - gtp: use icmp_ndo_send helper - icmp: allow icmpv6_ndo_send to work with CONFIG_IPV6=n - icmp: introduce helper for nat'd source address in network device context - dm: fix deadlock when swapping to encrypted device - printk: fix deadlock when kernel panic - module: Ignore _GLOBAL_OFFSET_TABLE_ when warning for undefined symbols - hugetlb: fix copy_huge_page_from_user contig page struct assumption - x86: fix seq_file iteration for pat/memtype.c - ACPI: property: Fix fwnode string properties matching - blk-settings: align max_sectors on "logical_block_size" boundary - mm/rmap: fix potential pte_unmap on an not mapped pte - arm64: Add missing ISB after invalidating TLB in __primary_switch - mm/hugetlb: fix potential double free in hugetlb_register_node() error path - mm/memory.c: fix potential pte_unmap_unlock pte error - ocfs2: fix a use after free on error - tracepoint: Do not fail unregistering a probe due to memory failure - isofs: release buffer head before return - tcp: fix SO_RCVLOWAT related hangs under mem pressure - random: fix the RNDRESEEDCRNG ioctl - bfq: Avoid false bfq queue merging - locking/static_key: Fix false positive warnings on concurrent dec/inc - jump_label/lockdep: Assert we hold the hotplug lock for _cpuslocked() operations - KVM: fix memory leak in kvm_io_bus_unregister_dev() - net: qrtr: fix a kernel-infoleak in qrtr_recvmsg() - xen-blkback: don't leak persistent grants from xen_blkbk_map() - KVM: SVM: Periodically schedule when unregistering regions on destroy - gianfar: fix jumbo packets+napi+rx overrun crash - usbip: fix stub_dev usbip_sockfd_store() races leading to gpf - media: v4l: ioctl: Fix memory leak in video_usercopy - block: only update parent bi_status when bio fail - RDMA/hns: fix timer, gid_type, scc cfg - block: respect queue limit of max discard segment - block: Use non _rcu version of list functions for tag_set_list- ext4: fix potential error in ext4_do_update_inode - mm, sl[aou]b: guarantee natural alignment for kmalloc(power-of-two) - mm,hwpoison: return -EBUSY when migration fails - config: Enable files cgroup on x86 - ext4: Fix unreport netlink message to userspace when fs abort - ext4: don't leak old mountpoint samples - scsi: libiscsi: convert change of struct iscsi_conn to fix KABI - scsi: libiscsi: Reset max/exp cmdsn during recovery - scsi: iscsi_tcp: Fix shost can_queue initialization - scsi: libiscsi: Add helper to calculate max SCSI cmds per session - scsi: libiscsi: Fix iSCSI host workq destruction - scsi: libiscsi: Fix iscsi_task use after free() - scsi: libiscsi: Drop taskqueuelock - scsi: libiscsi: Fix iscsi_prep_scsi_cmd_pdu() error handling - scsi: libiscsi: Fix error count for active session - ext4: fix timer use-after-free on failed mount - loop: fix I/O error on fsync() in detached loop devices - md/bitmap: fix memory leak of temporary bitmap - md: get sysfs entry after redundancy attr group create - md: fix deadlock causing by sysfs_notify - md: fix the checking of wrong work queue - md: flush md_rdev_misc_wq for HOT_ADD_DISK case - md: don't flush workqueue unconditionally in md_open - md: add new workqueue for delete rdev- nvme-fabrics: fix kabi broken due to adding fields in struct nvme_ctrl- Add the option of "with_perf" - Output jvmti plug-in as part of perf buildingobs-worker-0030 1630717611 4.19.90-2109.1.0.0108.oe14.19.90-2109.1.0.0108.oe10-metadata_list-compact_tlv-kernel-tools-debuginfo-4.19.90-2109.1.0.0108.oe1.aarch640-metadata_list-compact-kernel-tools-debuginfo-4.19.90-2109.1.0.0108.oe1.aarch64debugusrbincpupower-4.19.90-2109.1.0.0108.oe1.aarch64.debuggpio-event-mon-4.19.90-2109.1.0.0108.oe1.aarch64.debuggpio-hammer-4.19.90-2109.1.0.0108.oe1.aarch64.debugiio_event_monitor-4.19.90-2109.1.0.0108.oe1.aarch64.debugiio_generic_buffer-4.19.90-2109.1.0.0108.oe1.aarch64.debuglsgpio-4.19.90-2109.1.0.0108.oe1.aarch64.debuglsiio-4.19.90-2109.1.0.0108.oe1.aarch64.debugtmon-4.19.90-2109.1.0.0108.oe1.aarch64.debuglib64libcpupower.so.0.0.1-4.19.90-2109.1.0.0108.oe1.aarch64.debug/etc/ima/digest_lists.tlv//etc/ima/digest_lists//usr/lib//usr/lib/debug//usr/lib/debug/usr//usr/lib/debug/usr/bin//usr/lib/debug/usr/lib64/-O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fstack-protector-strong -grecord-gcc-switches -specs=/usr/lib/rpm/generic-hardened-cc1 -fasynchronous-unwind-tables -fstack-clash-protectionobs://private/openEuler:20.03:LTS:SP2/standard_aarch64/7dc35b8eb2969ac574d107ecca06cf53-kernelcpioxz2aarch64-openEuler-linux-gnu  directoryELF 64-bit LSB shared object, ARM aarch64, version 1 (SYSV), dynamically linked, interpreter \004, BuildID[sha1]=cb60978223b2ee0b8e01541434899c675d0a7f74, for GNU/Linux 3.7.0, with debug_info, not strippedELF 64-bit LSB shared object, ARM aarch64, version 1 (SYSV), dynamically linked, interpreter \004, BuildID[sha1]=bb59c09478076ac2ab03a4cb6b9798ccac9a14c8, for GNU/Linux 3.7.0, with debug_info, not strippedELF 64-bit LSB shared object, ARM aarch64, version 1 (SYSV), dynamically linked, interpreter \004, BuildID[sha1]=9d16050ccc867c52ccb1e81b8d76344511120d08, for GNU/Linux 3.7.0, with debug_info, not strippedELF 64-bit LSB shared object, ARM aarch64, version 1 (SYSV), dynamically linked, interpreter \004, BuildID[sha1]=45aa6163789955ecf96fe3c87d04f8958c944a2f, for GNU/Linux 3.7.0, with debug_info, not strippedELF 64-bit LSB shared object, ARM aarch64, version 1 (SYSV), dynamically linked, interpreter \004, BuildID[sha1]=34a87da92c36f9b6c1eb498b1ce60c50828bfe97, for GNU/Linux 3.7.0, with debug_info, not strippedELF 64-bit LSB shared object, ARM aarch64, version 1 (SYSV), dynamically linked, interpreter \004, BuildID[sha1]=f299eb92823a994b9d13a845c38989a2ca0ce8ba, for GNU/Linux 3.7.0, with debug_info, not strippedELF 64-bit LSB shared object, ARM aarch64, version 1 (SYSV), dynamically linked, interpreter \004, BuildID[sha1]=a45da4f2f3a28aa91c8ef678c281b7cb04caa169, for GNU/Linux 3.7.0, with debug_info, not strippedELF 64-bit LSB shared object, ARM aarch64, version 1 (SYSV), dynamically linked, interpreter \004, BuildID[sha1]=1c2a4d9298c7c53a55f29d03b7208d43587c7c74, for GNU/Linux 3.7.0, not strippedELF 64-bit LSB shared object, ARM aarch64, version 1 (SYSV), dynamically linked, BuildID[sha1]=36d0d7865233997f94c4b69b528bef6b1e5b51fc, with debug_info, not strippedZ/PZO[5utf-8498169800a6a6174c4013c94a9d0b7a150b584e707b299572dda3fc2dd207785?@7zXZ !#,3] b2u Q{LRvK"zix/o}36;fIlnރB+ސvf}"9eU82D]i$̓9aq_m3  x Ji] AsP;Bb(LXكyq9Yc$__O1+x# nȆiCNjbEMt;&6Al A|iM9ZVc8P#:zo> 껉N}ˣ%b,mx@+qm?LMrٔZD;AJ ,j]@w *tAFdQ *Ee!O)>3G8/)9_Nعkߍt~QE%FxO'G?_rub%_ L5s@pňfGpH,%\Q߼K`/ &[ Ջt=nHƞG {1vMESE("&ôǡKUv Z䳡gop`Ŗr|m=+NA81)cVճy4-HuU&3} |-@ &(xǑ8rh u.XKc=5R7sa )=T\+ck'PhNӭP;'K/}`ֵ'hY9 |{@u':p]sSN$s :uD]"_>yy#ږ[^w}<b$Ǵf8*WbO@#NgKyJݝYNgQ0R꟏lU_/މ<̰OZu&6c+YP;("`o@KdB}'/@v9}i2}qMص&L)EY,e֛!9|UB$ &!89ZYv54}-ר4af{Cn$ LCLle!?<=VI)3Fԃ NF?rv.]LAPh#yCA?h;4G(鞀Cf ;8CDQ@e{ Gn:cxJģy'"_u%DLUL$"NYX˶^hJmp)2I1v-$PFVuU= }k3[)D&8smK (,[&nPw>K /-bo9k%}V [3`"w0JX|Ͷ-PaۤūIȷ"nf@Ě]2 ޿o ֤I`? <|1Wef^R^F17oܖh7X!F\C(@ٗk3s^Z6_).B|_4fV*h Zr/(S@Sџ7(}ϲp+|D ;D4W^8F, .;8j ȣMdR*7R@Ἅgp7LW;*ϢL3A[,fT.?Úd6 ^%]3:-qEysnv!!f]_qt-BTӫ4!t_rԙ.'xn+KK~7q,[? Ƙ$WF?\9(4`Ä]L  m,%\a{fs$E,cAOAN/]8RU5r$(!r) O(i =oG:<Ԇ >a_w=WuMaH|C2pEhҎ $>dc,VVq0'0n9Pj5 fJjMYޣ(F J_=ҭo0C_B%wPdJ*#]!@$Ʋuɯb.̋M EHJQ@#QV+h9޴~(UYc1>8,ȩP2QM'to"Xtc |_F~GBig/}5'% #r-be2.D <9WpYպ);eOک@[$ (T#!qi zRi מ#.մ7*Kٵ=hdG\S OɃn*4UEE' 0M; A"ql9aHEIޒrUĢ}Mym׌y̹8-Jr}c;ءU\zӆXTЙ7g-Ոɝ)TZ#$$ȋ%Ԟ}ۉݝڗRw& e FSkF:x/% )t7HG)^{M'颽ӛ{ ֮q=K: 綔fD G ~HP.6y34%KS~(d , zKH .yPAV/ܧP('L53Tk W0!F zH3v@?`q.MNS˹N.ޡ^v=92IW_Gk{ z`8vf^)Y K-Lo?ks (J, EWsg^]I<8s"Dm\60NZ%8f\# p{ThDg/jQl83+MV޻̤b8r<@P,|%V_5/K򾉈]oBLTHFYգwi5rZy8PG-sy!r>桜pd:j,@嗤xXACpdoyޢ>)PR|s}5hbTݺd 6'K51dzNA< JX)%"s(FGV960yU @g& Uc1M7ϛY!Ll&zaLˬ sQ:[zŇʔ<|SHxYς1wg->0dݓX1Op;=4hC'/Y"$i? L3ӉO%m3:j!>~Ck9ua=F)bҬ!N>dH^]霙MoBySZq.kfU -kiH*^A q%]iUS8< :RUMDel#rQ:^ܱ`Ƀ?Iv7l?EKwfQL8:kb.g#A` F& `O3%9TՉ,E9CuB` 6ǫ;4A #; [d2q$)jT3Kj ,ǤbOM} 3dO+8YyYN< %_ Bs%؃Nïua ~p虄+H^a'EcСߐx#ed\L~w )l lng7N.y녢16*-3ePCxIŸ*rvu'>1ɢkoؔPtEbEic9S'+Nmf%}4{uj4@'D3Ê|>%,66Lǟ倰ȎW"m:M',>DRK6)͗η07QXMJZ5b VC!4r=3:x]iSّDXQBnBtzƠr>]Ybʾ; "՚|5V;D7H{LY|3ǃIFA~}LAsZmR6E7n]F -ӎYT/@U\=@n@@4erF%@NGH͡$Q€Ԃ"S>d\\^Q_6!t%-#V*2'&'mGCT `/rFU=4(rqe}mH.FfLp0we_ B_\{ȖFFW0}d(xQAC, )HY~bE猛&adt lvPj"=[ NLaS&/p~n7BoS>5u/[>*RI)o(d'rKOf{L_dZxDA6Bbwd3jhpxu XY Q@{5X}\qߟoB+"4S[gO+*Eau'MP|茟дu<"kFX6bXLiNY-fGa>G7BAE5;,4,*gQ׎P0 P!9i0lĎAc&zAG )2b IKm w:r'a36ћp39^8,P~%d{Ջ8kEIqY0{1VR8h6E@/TuKðNAt'{?k(!GB3쥪tD' -ȱ# Wpu'X1:cPђ'[C ŁRe2{03_o$0RRL-U;o?eQ}N@`w2'ޙ1FL^q\tQBUDФ>IkQ2IwQH/mgX :4Gza1cW(:S*Y ;)J"Dւ~`4@𹺕R˰`ߵo؛sfU0d'$&7L-n<6mWYtDikҕ&?^osρ 7HqSsm:)"$B(L{[ ;6vM|/‚ : }KAHA)8k H#X=_ _Z5AnT7Cƙ>G-9fPև;|>(CL-]X&g/q?E .{]Tێ'Ldr)BdEd%7D$R*Oُ5#|v*jV޻G El[II=$?G]MLv K!(6 p\ZTŒ5#jC6d'Rc/߹/?pj`E]xn|ff(`AxR֬("Vtw]gҕ3K9sYox'!LVB*ƹ%L ůH^\v,n,#{*T'LfQ@fb')e&IXcV;fFx z=txjSK"\K6`&aYe؝>&!_ TvϗI{u }#!ǟfOi 0zuM6YDDCJ%Dǘ/YGq\D1,92Qtޖ^!ʋ0n|wT y%\Pcfl`B{ 7`X0o we !G (|mɢէ.w(#LB2Xc՛b\mU}&PКUfa@Ʌ/?Q思*h|ڋ 9+f[陋]qpΥ]1kܳkFd$~V ?-k6.&8=RPOMZ*v X_֯sK=\ 7;QlbyMn`Lܖ*|l YcYYYGڊXvIэJ}fZ*^>?7_%.Dn5tXs, ӿ0=B6)3^lrqY7:nja4Dj/OwJG+tREpY68K$pk{Γn{Ì섃ƌcVV)@ͳ6D6/ƌ#pRo(>Оg?(z \'\*;!.ֻ}4ȖIF ¬ UL7iշ?Ȏ4^־]Q|p0i)/&dh֐gԵ~Nzn):nQF!*i'H\ U_ Ux㹖(i0Bmiƞ5SMoV<>t(iu5>#Pãlv%L^xa MJ?}_:[,(oV7CāSed/{xOY"]h*3s`Ӈ: ]PTf%U* IlAC{(.Î"OmQTIﵝ+>B&ƄN@Zt_җ+͘!)|3+]i އb&fuér9 LPd;\T n[w8W&3twHXE8> ’=WFӚJÙ\ RTb 'W:}[ڂfbG'ؖH$^6 g|B@4W3:pV OW lsBH9B{Oz #ȴREl;f@$ .b~#G@bvҪ 2@(R,5KȦ)@a|_`AAA nu-y T_imΡ{ <4R<^* AVɾTpy{DZRF$C:?I}%LK}a 22X. PmE:AĪ*~.'޾gȽELh]wU rrfxy*D@X ZJ}}QZt %lR5qJ(3'NDP7{- ] yzR 6}]]e̴g_E7U\+1bw,/#382 +gzt$5A1K\-5$];3Qf'o\_Ùvk ]*z9azYV"ukjoZtAW=]F8N)IEXBFAM)rXuSjq9XE'å٣pױhfy>_- Bld4xE2מK+x_,oV^,+|-b:'wf%fM@-U!i˜wa8&|P)ͶXּE?^+UY5 jx"Cá^t='^ژ(ovA*#e6d"K>J(4iAt`RQ8)dNRYuN$`]󭅅U 7A~br`]vUVX"> -54Rf`ҥ/䤪qsqb25L^)47<xu㡯~ued%7~% ʰtd ${1̮)2 LOY׹ef@nvWt4|ͩ]7 ^=$r6 \ fלD+N]3KrZphҋ{|2=P,&Q$N9`7iiJRlD/Ą Vz[fJFȟo+ȸJ8: [xޑmM{+ߏf٥Ow~m|WCGvWei(ȸY!.b6Y(ޑ$-Dq//%í];d۰_|jvhO>r К2j5IAAzx"lݱ<^G_F=\|n!c=SR*l$a.xZgf'=3RoB~%8tj)Ii*x2Dv ȡȌ&=q(]pSiƄ{Q,'\yʷ4>Lܽl? rg$~L6 ?:cU3R,Hǻk0)O)DlaPK{6ɽ2Jg*HRu*&ȟ|ՄEfHrqX !t˭c%zG1fVp1mG`@;{eNIa$Px"374]rI>l~-:vGW^ʟEA Y1!!ʌ4dwiTsEhŹptǶ`D)k暩9n~UQp]N>ežN ˳q%9Խ|W^01a"#j: `v9f=ܴhߨ ^ fdșkCnCb۸ܿL,+VYF6StEp~]b:ye #!h 2,үGӘi&d ASr5l ?loV_H-a͝dc|ߏ@:phClqO )ԯAnz9z9Xajo\$dx+ ՝wӣw因D{oC]]>`p;Ĩ.prSM3=}ۋW9>?0/Қ ,,iwe52ZE^ۂm*>u= ~2`r#j) [*/+ `7m7؅'A9~RզLonEkI 8@7frJxWUlCce hN*LC7OQ'>Q+{XK'2SCn6*$`UVw'D Zᶁ|.q`p?p$:Ir Shm>kU 7{hė⹇QVXqZdɅVQ'X^p mM6y̹A*D0U=͐0;jc;A\61~aǹ@Y1XKBoG hW]G+Ѹ'0{䵉N\,_KPg_V=DxsL'$Χ\t!Ze@iRJ츚+3O g 5_j`M ?og-kbO'z$jba~Rw(E{+c&='̝dq`mTƳу^q4L8ij}(N#C|<Eߺq&\ Yt>c 7yIzƢdjv#OQ%EV4je2p"Sh)3h3'eHj7̦.h.;}9(FUcP{hdz`1}Uvh:{}ϩ.u~R.\j] \\v J MX^ *k[@vs^+ENH1:ֱV/t:P7ebѐIoUu)&nJzInÒ~QAVG; Tz8swpdzuTva 28B*HDF6nH\ErħAu+㱲A^I+( -?kmlMžۺ RGSnOedE%Ҍ9._6|cj 5<wElxOJ_60zZe0.oBQ(/B_{`/= l~ Ԡl ]r:+d2Rӧ{ ]A{Ҝ<}\My`^W>q}?uw@0\77°Xe$N[ӚR+!9f̆ue>$+4\yx1t`B@Ҥ#orzjANJ†!eCgkVʛQy"7̘ACSڋɚUB%dxZ/ME0@; D /$G'+NX,7r M~c# ː%DO :wN@dRLAfbsDMcM# qz1< ^yfkq+? HDv7rV-pP|۠ GJgQ#pUuɟN!ZNMhh,L~,<웡<1|%mϞ}o rYEu(x<-ܢ\le@ccE~n) ~ԑ@f>Z1`iyb{j]b!2LG({ϯd=y$hK`Z85;Gkϩ D*= /vN䝯Rma@Y5Guʙ)T ܝ5>ByYldgOuXۗ ?K]cB.Lf c/mv.W;B$d'<@Ǹє4Y ifBvx[ orml"tFa~}.r)nq> ^o^xZmYr$FFYWrx7zt,/+$e%uiz_o/ /5dbhim>b۲3z0 e$6吐,ӛ'WAZouw\:#̩FqC@nݻSu~08Hވ#h׬ L/=]sςutx6k fMeʏ[Y/Q 5I*qxaN}?Ci| Ti>rnND܉cC@pԓ7xE]4 'x!X:am}# :uVB4J>}`)^'͗u?ssԚlC0q/=>a&>E!L<, +=əs{9D$SvG/@Dp$ {$\&kX0'ўMޚ:)JS8Mj nd,ʐ[rryL~ ]*غU[f,W84 g]jgTcdASЀIu~%["GHfhP;`LJ(dzY\8RK i(])=q=qPRj̹aPX?`z3w˘֦]'޶NEnoƒZM{=n#i~}zf][`J :H̯0_$0^եP=ۦIT\';eSF'&j ;8pEz_nFIl 4M`TAaH̬%hDwT֓ ޖ%wgõhJA.}A’p/wԥ~jxBmVᯢ5QmɕZwuW ,4}oW:y*LŹunl$ >>W\MAI;# o'̬ 4@z> 'ogYZ[f]'>DLɇ4~*7CFb[ϖԙn|4k4Ý!pf$8Lw(SXt);«#Q9V ic嬩;O⿗y_}wZ&%!Z/`2 awSLT9լl]!`--S]35<RN'.oE1Z({p ;_cnUѽ}}eċ;o%F-A:^MI^u~7F;p6#6K/!yUɕA'#+@ilqwRy+җ%pm6wQC}Z9- O5؂ͷ+hΟ*@%F`Gv{S ӶD2j (@R\,wYj6f1=!~E!Cͳ+Ztt;]+!L,Rmz`1Ⴕcu6eմ0ڽ_BE(#` =NRŸ10ڙ7EFgCt1!AF6)bf3 7># PǷuaJ,&^EUYn f 5>ڑB!o+3iuc@Zݲ; q9Z]kwSlRtTY6u=.sv\<2ވabbC0xU{*1|~;V[%QΚfptC.: {cGeo JmYgw;G 1?1l 0qwJ1L05Wߧo88T ZyZ]G&!iR5yM}Oy]qM JAϸOmPlB̛zfѽ?r]H5Xkr#)Ȑ^9/>`ee>.֬בH2T" 0$AGZrIQ\z"! B$1pȱ(JP223q˷RKz: BOHٵ|Ketg@ tc15SC36?P0{p"-уBzM~ NnYu0K*Bx?GCa`N%rqpYIr({P?^,]'lE#0&zDG y*<B Jڣ2C4Ry$ R,6׶ǃS(ѯ`9;3h݇]$ۤprhPFX[D#yGtvEZA E cK3 EHWZ7Wds!r!:w^z n\:U}4E䭥G UI Uढpgʢ@<~VMcɖ5 5~{Wb  3n?wc- Z(9O(=UFBRfT|Ě  ˆ9 CqK[ponVE+I $zuaN߿d}a36a|٧{h&6PnGz./b<*,asj jKe sd*5c/N_WIatE,7<(}WAz\( Empeg~w^-.ck+ާ{ #t7q3FsOm|~t_ݟ !]F}N-7ڡ =*;qyvm%K~K| -;xVKoh0<8]/pJ]3o$f;ۢ ֨mIJy+9gy27sqZӨT/u8v& >i/@Y7yM@dFtPE|-PK 0Jmdf.zm"mķɲTA=|N$qLӧAG()21&Ee㟲_egxjҲMp>,7?hm,|z%mSF.!ًJi"WRߩh[ Ma5byiac)9ĤTJzUՆXrM ئ|C]Ha//7,ֺS3#š^gigT.6OMǷ;%i_'3EŴ9Y&e2}:Cl4MfԏL qg)^p FW^HF v`Lh{Us"77IҎ >z;hy:Aa܍I^&eJG4~M0r7|hJ xD+Ǎ8q|#y*Wo['[yy!|tLb9fmbn#rLt(Ԃfu=#}?tM!S8LН)uaJ .;9a j<Ô09TJeb⦲*;ћ&̈́ ~JP"F&>.h5 @7U~T+dtydg b9gNI+hMb@- Qe۩"<e|1[wRx*8=w1jB|z`]ӄ_5:Q8naS6Na`^?_D+6=oWN5< D_Eay?.kJ<'ӒmhB;Q\ĕb7s񠇓qr|{$A0*UnoMv3LNÌh镮Y|=ר#LDY3r; "ǖZ` 0[cXJZ~or #a䔟J/HW0ܣVG\br~uB`՝]c))`FK26~ո9aro4>LЄjbvv'RfB|*4k7wK:G7*cOW/qqr?[ ` ztrO`r,Ck4F(sir"d'DL Lr8/˄IkiA)p.Om o s':aj'g޸b k#0%N(%ZE :w3sftǧGcњ"`1~RVcYzW͇Q}پ=xWx aEb6 [DF=! Q^+v 8`7I?rys05JV!MHge]DǨ4K?:4 B ~kQ*z xhΚwqk1i|O߱Tl#&?HX5/f gfc%rCbep}9[wX M?K=igѢ6h; "׭&RUև%@gM6uVGňiC7{ظNaÊn~m1Q,Mgv,x4س쌚wb{! Kx9G6[#z{aiZR\񼵊m=2%/:eW zA%"2 }GW p/BQȸovf/G9ha6'0aMs(WC\j[`I@%aq]e 4^"RK+# +axKm(U4}dzS`-Lo9Rx!lH9!vB Qx? 0`G@OtrifOpI2lF:4w(çTѤ5b3t lsS?-4*4\[":>AMl,AȧjI#NNwqBB ?G)v)G1K`r̟Ѯ Nbԑm7U\ȹKX8o_ cbLpǿ!:YEK%.E .CSK!hSt;ЗִR45xs¾"gBoA AҔ\J$0$N x)O$RĢO}h s4έh3%uMX2"l*omO+}-v˞Hx+KzI^ 4 I#{lUpo엔OLWqY`L@mHtb|1IQ0>`r󑳺9S> 'C&ޟmVFw"ZVW5 1lCf}*~wr:e@appo?X53sZnOZ`V \HXڴ|Z;- }5F n=XQ,YiLb~?N:-4Kl4҃3T'< ,i?$h"bFJNe!0ꔜ93 n :C3}v-] ޾M,hLnD:7JZs# E<:!1z3eő8߸ĶoX땧)4a&r*f~Q)\6oٰ?L#xQո@vѢWt_CNsQt$gXrd~5(dKiȰ{cZz:\/!q\!-+}cMm] wkU?,3 h: 3䶀y>a})p DIuNJ.[(vQTrFfA3;w&5BXu\̽}B,q: R\jhȌ>NSAߗ艹c8|Q 1/# ?k,8Y ?p[B/7+ I]#U vpcy8L'Gqf3~-o~5xnsT3_"Y5+ kY`Zݙ:!ɩ3;c"OYj;c dX#Z5J*Xnu"dP a2۳Pi9&_"/d%-jS & =L~%|TMv>o/5z/‰U5ҵuP4`( ׾n͓i|c*jo~WTb"{jE,yZ&.-EJٯWƒD܆S>z)$:,2fŮD x|xmA\'/r:;?c9j]]sOPשozS^-Ȓ=@`# f]0; N{fQ`_ƍ^"Ps3Bj0!sp4n@\A"y]E A?m}0lJtFߔǞ)El?_;kG)]gc̃n '-@ C Am'kH~':oRga0KCG>l\+*\¿lS7V{:)n> `kly`æa1E,n*H{m3# UGm*0?vŝ7o{o?X (xOܖqM#'LQG "C rZo#d=O7#`#}$%`. n$.2\a7lzyFWC8D[ 7i.BH<7Dn j:f)S\hƀ7C{äDWRQ*.0;7ׁӄ7*G@fQ(Z0$?_lL7,$tPu])ٙ6]r#bOp/܁ zxګsB#C4` ODJE dkZ+M :x?}+z(zcIF/TOtfDpڽ;HxaR9~BBSM[4K֋?[%p>9TqU^Ա*6űͷ&Ɉ%?[0U9ż~Y$ފ%Pshh GTo2nSlFKRwІ#@ n(D~2#Ե s$WJIzrͅ)^8U"fP 3ēVk{6jv`"GO,k[G=t^5kĩxOe8IIN'4<Ɂ("d8Z.v>w.ztnPCۡfkDV3& Q3G:3K2=?S!?ICVi"*r]wyʮ#yaM13؞>d|+xdyhՕ$˷'oAt0n #7wn{J:=99`oq%=/Fvۯ0q\w8B3KX=ir7pԹ7~l_9~ ]"ǕJctKҺZzA53zah<Wyp 2}X6>Cnk[`z Ц|ӢnDZB={zwNR\yPfJ V#T1bc)t}.S"1aw߹O1R>;ȞZ6xGkO>@/7t+䣥J5Nlf|Nd,󶵲j-v/)'jpa61 Y G?s|F-!ZpH 595jIoi͵| j~*X@wt=dO_ZgRK|fڲ-@[a4PiL>ro# YsPyrRS_^9N0Mone˝NnK ϣDxNEh06(*wz4Vw6$&LHU7R%~̋c)O8&R8(Dh)rYvǯ7_*l0] g#_U:4\3->6 K+K jP_}j ^ndOj-~lw1Tp 䦆6'DDZ:W 37h'%de=|\t)Ũ>QkS4evTߞK SVkw&J*$kb(;UJ0Luq=W#z}.>N9!LA5{X][>gPyj!m %o4| *I*|tRp{%* -?fނ7\CF{l߰uZ9qRgj ڂBܹ֖)>ZgnAdrfU"J\ 4 6khiBDay a=O(ȚNBJO/3 +}hRk yYsҺM[$ kexۺv͹}rdHtWpdvVO2a\NS=JraUs"|j1kFvT\=$^Bd>谟o&Y8L!7\}JŦQY毶GJܸ "[| F;'-PǧjYuQ6{2R d5>P=1+[7m``WsO'~ãQf0y+C\g'uEЏ?YP$'ʺaVZ5ZZREK_"s⦓CZuvSD2L(̮4 Lf.lm\ZHr?/[TK3JK!Q:y_#^!)cA֧tQQE&c={%l(&,Xʊ&B5+fWy 5H|ȝYAv-QBVEW3Z+/@S?%;dP/+žCG%5A%9'3,iKK[9P}+ԛ# ٓ%"4Gwt'*'4 Bu =?l z~70j6N.:RSc?Gx`G<>]E8yc+5rēZbe}!]P{~q> 2U@չ4P1@߲:]j=Kti# |O%̇*uOH>6/۩4az-SS 5_'E{hG&rLs#mq(Va=%»c5G4Rrp21rnwL?= Yڐ0y,[wB%+CzCz 6+(aS}_f-~_6܇We|l1p_J%s@ϗ @?6+9Qy&*aK(Ӻl}DhЌXi~]]"8[h9G Gt\K9~+('L\r\(=V8=5[}̲Dl`6Z s_$ Bn%,di'dȨ1d'ף}~UeI#7BB퐣@>8#MOS:.w!XIoNn()V-3eDBh@QQǁ+%S#2g?)4b*;HWP` |f'.Rw4by~8BĻ96#r\Ҩ+yn}>>tgNViTP?h$, ]7,ۤK&U:s/)}T=ϒ`O79"okC^cϓ2_{i~mUB;G[T5cg[P ]8T"FɈZS'HךCZozQ ʼQWq% ;Wȋ_G^:@0}#T@_t,CR֤cwj`%=;k7v"zXCAiTet,&V'>i*cHĤMhC}K[*oՄ2,x2"BԳK= 2١EmeNȘ(fmsŔ6g"K-AF"?U%595RzR[˖^RƠJHU N`DŨNi@v R6z۠Ȭ$D!-PBAsi Lwxپ`"fqMEXw=s/ds'\vzCsJAY#^T7ji;C?G6"Ln'g1X찴"A]N]2!j|~Еey^h2fI9o G| y W^ QiMl_ߡnGƫp8 UQ)d;bvX,Z۟z*{{IHf-JO,D6bҡ Rm#.E2Vvs9& |~#Ckfu]P p>b25`. Wٯ>5{]\_)|FpP _rm_.9e< q?19}򵜇FҰG3O>X:>nba Ӫ='L"m4 U; u1 gJ_־ $%XTf# TJ 3vלe3)+jWex6J}c}aˢ %So'dH.n75}kBpۦN-; a|CTk 4{`Qxo#**Ycg4|˻ZCR L:;d tߠK2sz(؁Tz?&AI24\p?sKS^ƒiDEK]UR!i~;¢Qb~}(Hs=]eЖgTJm<'Hb&&.9nN/UڦI->>peyU#~MƓsFc`ƸǼn_g"g&YkD硆g떙%*MJ 1 oXX5]}娋2~9˧Ғ&5=)\Ffe-FNm4Z.A djKWKH_K o0'{T\PGؿ~8GDfM -Xz1ւ3R+}qu^AmetOiSl~b:CU;޴ 0OXz\Ni>s}v*sF8U&bBG1{ if9{2N%~G Q=JGJj1Pcȃ4v܉y!s|8) ,¾kNy" ?';>NhޑDWAIN'1|; w5l7) M?2'bѸGþaC0 8]19s__>Y 3E(<+ awh=!PqH;1L8 we~JBx]E/en &?̪<=sM3ؚ> aP9{dMj,%H!").sp4!i sƌϟS4P & :r/;O{ruB EC")(p|c U<Gд8|3q)DL$/,͛0Z~w] 5ӑچ7v : aen`O Y sli]KEXW{7oN6Gbu<9y@y;,(]5Imc˲JayTP"|W*{Ų)Fv~3`k* ܬȹ \2yϔ#˫L-tS;o:<| _Ë 8@'߽ %45+m#<5}3.-]\Rer#\J#}o]к5sD:S';U R@. Ӡ\ ͳ^srҍBeӤY>$T;@ˆZ? EZ Z?@ԑ=pvqQ:W1 PkO}œNq'*l%k:QWmtll%=R(s\O[8a 㓸v&0 "?;, SH %4a \=:;29y! z$/uRnn,!s-'@e3]QjZsjUT R :o׫ DF#l .~ _{rѸx=61x^S '7qw*!c<  ;%+xUE:V) ҳgl9w U:rD䷞ŮAf@Lah }ljQ<p\]6N*C~&; Pq}4k`Z+:[+n`U5hȋoH 1&{J*0%M0Hs~&FlŐںyPAX*ڣFc׊<V[d"/ ZXg^U7- 'N*>wi$spuy$Z##%GAJq9H&HO@K7Y 79`5[!l cIߴ"AW Ϊ z=Wv+Ti]b#kMHYP|a 38A-g,:5{HemHL40aH3^~UrO05#jk'az`~ lϩt䌓RlCL;x*"ţ秨1!·.@?iat9 >$lׇDOsA'_ >_tJ fGHm ]zp_jR20+TSjR +28oRV՘0{P٘X"FRNA!ϸ tb<\RJf9f(rϰr5WЇk Ođvu 7zlhzq!'Y J_f#PtsI*׋^'½=jD~}<szp;M[I)9LMF$8*W:liG(rqIKfbzj]~&KJU)%O`ZT4lGӞ(yo}mB)t-{#b3v nL]烂0td 6!$\f"9ѥъȵhY!ae) ?Lx+OT񣼞@. NnA}  VR:U3eTNI¿bP^8qhFTu :Jc>o'ڴ!סY,Oe1E:LiSey!e+V6J6,Le2pLXY:-:|h$R& P"}cN ExlDb@L+M}C#mw@'`lޫG=֑D(aE1@i)uɀTh`}y}7oGClџ)_RR<`9 ^6#vx,'i\T7(Q2( [kJ1U;ժ ».Hg=@pIȥS*ޒ YUݛ%$LvkGCvcߏsY0)B+œEuPyOӴ_UVݽg1?=U  b4|LUx:Y ,H8-+f#!u!ծ7"&w$UQ(<8q_ z*\`&ղj? !.%,.AuyЈʄD7Q?"dZƻ-~d_.=rw[yv t5!' h[9?ԚF^z95ɳŖpxHD^24 u6Ȅ"·PxH,e-a7-W}RFzMvFN(+o qz#|uhb"I68|F"+52rS|-aqp 7fMC3M ap 4 3Yá7e opxU/]`D=EVcy]dy@:ᏕAEOC3y]w^e7;DRYL1`M! '`8B'4&~@:AaτL-T8&+f5  Ts ņͨ=ӫ<tT>cڴO+@>ROS'+hxwLG Xf;>+m9H"'r0~ibY@ٹfg X%-Ѳo_ 0vqiɄP6FO"81ڵM?2ԁC|"qΜX'M+~D5azɬ$Ho-[MEcͨ,>}Vv:[q% W ,rA6+k ^n` _~[`RD_:Hʤ6hf.dĭSO<сP~!`8H#oKcj2) Q^ WZ'`tM;ҚQީ.g&b('3oo]L8A[ҜTpD]/ @qvƿ%a0@ٟSo}PGkSE^׎'X2ClËܟ('P|(t\eǣ5}ƝEY&nS ǟQO*&%8}V˅fqlyH2CKBր7yN6w=ſC1[wj&Ȟ;%cUcxDƼsKsŵkV81%-SSB޺LlA424:{b L$A!8rߣ™yDUuu֡-\2_*cZYSϗ a<]u^B?BXrRJ̣$'|~u} e #B=}ot` Zn-wD-m зRci 6o&Y//?jB*~ufͯs3[v>\9^Eĵ-}#G̋'b'Dw"\fdD/jfI ]ё8* 0>' U =ױAvseߛlq۝;i0e 75dhTRLx*?--q/q)울"+/ϔJzzUTڍe9+V7"06unulU@?Zi%:OwI!mB# Ӯ`vj@ٰe_܋ula %pi#;tqXrȆO^\;"rhX;F,{˯=AyNz6>AF s;ϑ( r_lbq6Gq dbhۡ8aHqA ]cӲY߮? ? #%{Op)f떞g2M6ïHzOyi>Кz(pS9iDk]{8(mrVPyn{I^pb1QY'Ym6 t)TmJ+LLu]?HYgpu+6q[̼G @ Ki2kfQ%)FlIŽ~%HLV}L;|lCz{/55;#\f[sZ2wCΞcu%Hp M\.etdH2ťY33 ;3Ip "|G26yrcnݔْ+$~ ܩOʙGftfֳjWQו&/ "j o}E+/U:ћ xEUe"Ԇt(\}/e-7WֺƬ=RFwRƂk0CES8"*^B?RK zL6gC KFqN53d\73Cr\}[Q o]RvZlt]1'C/t"|M|]QRO< =kh77_}6ؕ0V%S}oMt˼:Da?W4e%Gלt Հ;+9 ˯Rtr 3Gºz?`zoE Yf16AcPXj ʹh=j2¿ADJǓQ~:8v#U v/|cg@#7 u ,t0rz:_W&ŶUP(=몰'hTBW0Rwʞe:ߒAF#^EK5NaWe.Y4ٓҕ{KIUjAnd`UW2g(A,zpYbX< [U)8LaXʋ bK@F䠝OEߦuq3qI+)VBM^uhԥk(Jgb)msђV7d_"7~+I.p^ؓ:1_2ep%)|Њx% ضbeh;HqOyen:]i'D௿DnFP6ŒRx4ǀ_ 2c8 ^/zN:7/R_SD tmU;Y꩷cI-Ʋl ^?S*$y%62WjlߠxpʷRPl$<)lup2b)c|ާ+X=!$?ɴӵWw&8H1tM?<,ÍSkAa΃4TKcr Nl.,gO3zZO +-S+gQƃM$ ?0w u(ZJ.Bp$%5#K nT Y0' mRvd ?5X@:~XzgjIn׫܃NV3{ 1g*Sȵ` ceVNof: EhIHBEo(Q]3]̧EDsD5.'f;9BzLdjTH TeV%wmESt8-Pa1͓Z3&TJ[A oA JpSP6Du33OhDO޺S{4F|ș"/uϻڦBOz $I;[i 7j&g>=i46)uGo/ ΦEUKаv1m&|fӳisB]xN.ufQ>JF8Xim.70+sUoy=cUFhfڈNo,/zm; @saFètN\<E[vBʵ6L@2\Uc=)Bj=La8*IWZj7&h+p.?Iwt$:{`us}F\ЇGk 9Է {H~IDJlF2.$,o@My';Uʯі4?uxw'T{n8bܚM8\mKEWVJӁ2."{Q,o%XՖ;*j"jrm'!^+C@&_.E0c_R2!ކQCr\zvCԅBn]I^ws,$&bX&KE٥ԒDB9;nFJ ͅn]LE.v˺P7 l|Ou0_MP b.5O;X7|<$U_pLD>%߉b>m@s^{ Gt\FKYԵ˸NF2"d/bO?Õa^"~'WH_Dޭ˾ @|`{&qǚ6I _3.%!/O4AEmzw [|!w7xA]Om;o2YV-@Y& %xgZջ2{X;]T!-f6\AP5*tDbr@C'UEN_N^SZbBp@ A(< .Nf-UG%[֬)o5RY`GP>KZ7[v2RcbBќ% qex[|ഹ.Br&l]oʺ0D<aɄt͂p(ƝuBJM4՚BZ0@QQxYNk |8VȮuET@!׼c?NLH lX&?Z0[Lm:Ii"7@Gg1{7: ~v@>9np8}ܕuaXc*Ö0niTcteMH~\aƋ/0TȨew[ʉA;q?ihI* R\yӴ"j )BjkKO[h_4zr 1dCΛrzz5·ԪE0aceat/!ؚ_gTqhXYgTPƇyƭM{]57v`k^5wXgkh4X^z8G3HU2:\^-,`"cH.<7"Np^[RZBܧvxIR**s ^1_$uk^~:c#д/ټZaMf_ .J`^IP[.bY}l ,H. \?R[46J08+uٽY020߉n*H !}BlD :T`E';>S~w[95o9ܦKU+;MM_=$4B6qZo7)+=!)B4r:{.!Kia2R4UFy& aafoOhNVCbPPǫZ,jЌ$'H:zˌ|S1.Xwq)&=g+Мa6>Nt7hq sD?xK( &_RJ 13GY2P/R-#+ fEͶ4r4BN5x&r RS>Km m(.>kRm5~B)ݓ[ue"鲭,CȨl*4Μw;e_shҦ}2B֣-8 2T>xOAoķ 9hZڍQ"~7 s1 G#Pp C2՝"2R`ց!ǟ{G%>{<-\8Nmlՙ{cv4@|Oj淠Um8~4`rxP72-i\ nub 7U 7bເEx=D^ϩ9>CNٔ}e?nV'N"+Sa]d:I1ۺnb3Ps^enhY?YHH ,r$/K{8$#D&jKweE @+A Pg W"F,;>߳9 <6"4D@JV8-GhWLU-Xz5.\Pa;Gnʹc/C?qt}\ۤtș_?[ 2+-4Y :ʻE#0 AP芣B:BY%b,0 Z9w?!UD5:/Ws(DjYj2j23!"Pa*H4\bBGla"y T"2NpTgW|>E}J^ҍ3}w2s& !}hX:%\~7TA_ڶ·ȹ`[fW+s(Q.@ӉCql^d0~?+UsaJ{Xa; [si7y38ұ¨oRz ۋ)=^H1aH}Z3X(($ܴL hwu4&yeHp?Idm wʭ'ՍT 5]a㈔ N 4>5R?{(`ޱ~?v0uC)Lx@ R$eϙDY;rhsO=XעCq,J%;- sϒiL3ai $q>i>Lw&zP[ XDgbA6-g' & :ݔ;מ\f<ڭ^|msGՄ@HCfLըtĈ'tj.$>Ԋԕ;1CꦓV-9LkRF7r\]YEYWD$G`Ή PGZ 0VfC*ցY] ;Y -8!28qg6pғO<z-d@fhXmrwt~!+ R1O(vQ4s fC:rG}ӧC稡ESg3MIF>Tm7^f _`19-/oE]Wb3dk!YD~wf=GhZm8Ŋ4y OCek~=Kqo3u!r>Z *73ݭ l xW6| ayݔJ*hmMf385k %ܲZ`"9ZDDi:pɪOi:}beo =OEZ[NɄ\wRfN/J%0LR7=\uLQ- 0vrM6zk?}HbX(bld?^lA=HZL 2֧9EAƂנmDJԯgl^1Gш ]\7H1=#p p5G"h ߌώvV e5-WvBZ?l{߹t@L d|Յ=YZgr?);Mpj҅DH &,[d}8/%D(؀wsRB:}Ί_/qהB!V 梌f{g]dQG„}G0- 2͸.av9[Tt#DXF?Gs/k1|iESrImv^x(zfB*33AsD8L)3Y1Aq@`N[:݋q590("] h!5D u1fXaIzЛsvTǬ!"nOr%jT"wn`X@ZZkv.:";׷Ump!mmz:E&2æts^}5x]utaiZELnDӪ)~goւQ^,faD0FrpY響״ )@qp*٤ r;]$7!)Wz0L$3W}bθ&יp<2cI]{W_5?B= ؂b'+Z4' z\0V;5gџh_տ * Mպ'uI|nLO9t?Ec4xQD0GApط0 B / STԙ(igVSO\sX&n)fr,a FS7BfnsHFݝ LI{7^d8R:8[JێUOTUx TXP2Fkq>:SY-1!-GXU0GxkȐ@XV6qG\-,!go:y96lN}/j }oY6ښk/2LًZ#ȞkL`m77M2c,`0UWP{41lP-yt$X엡}_5`-=oO o2@Jpˏ7zpX8{`.PIНsQ`A 0bU e1`E# Ǡr縀7)98|!bʏ Ie'r=#*@c҈KaN'Qs1oًm2ee\ HY`Vo;)޵i3'g%ds+^_e!pj*Xb\~drq# 2^uW̵ RiXKU#rݏvZ.YDQx;47[-`3YcM.TLˍOGjĀ 2Q:(QU 5g06&הSjg\/1z3+P:t4-hl|@;7#iisK1,H/ԣSS֗ ՖV"}3ΤJJ*o!) {AJ",AU~kk{==F) b|AT[Pp$R%IPHPQxpjYJ癑fp ѳ!UBxAUd>2A[O;Y^'~c?jth7;Jy3lי>1;סT;d`+%h j%mi\w"V'ўV_7`b#4/iB" z1 : DmAPZؕ(A?P-d_C3Hh"ay Zj[9hf.M}cMAs,4PZ||~SZ8M>9 乆Dܢ:Qm2Fsb\Cj I?lű~Jzg{u?gHm <\:yKih !}+xO{*k?܀h`exJ@xxMm |$%ԌORNo57B7#vOh P6r23N%F Q@SR>>&eR T;\Z @mͤ+P3U7ʃ)K31muЅ`ꐕߥUO1ܲEXU`4\>}550ECP`[FVfVܩ#E,bDJ0p 4S)$tpPsU-*NٽF t4owאӤdcJiP}u$" O9xNaz!dOUF!9chk+bjYzaXLns$aӁ0q5!>^m[yr ?NBQPFw :κcanp,yp8Y9rEmvqt,{hp a޳Qlfxk2DoZL{Uj=(( AqFdD\G=ǝVKgW@%)P*cAgxj[]s{ Z236F"6\DL]|1)gIQj|RUYS@IPc: G6lDJVAv,65.4c)BLz'c-uEfvMƤ"{7{k9Ub7nFЊ[VX•R&HM2VzE "RKZĐԝj5g a5*UX;P=dgq>2n3DmKmNdz`cLChؔs #祥'B?@#E5U{u+jTi‚)=R- Z=Kf.}*aSe>|㷀ޙ ._x9{"TP=%\F.Ͼ.cvPn=3 C0s+9`I'",43IAª3GV2z4lz>Âw8f0c' }$d[@y>"N{c$RV^tNLHFDM2mݨpvY <ԟ!ZR}\ӣ]EgpG|g@EGSxÞ#Ən,>l8uy%Yћ$r\Q=EX >=)˥З ,DZ]XHˌ,{F k&δjm+9T3ӦWյ6s24q.zxy:v*l%VC->Ubg5IH @}}o* p'.VX8iYpTVKkTx|(e M6 lPUQUKSc?4vD_E5[ BJ:H|XyzH>Ҕ6eaW00#Xn7nlmFz:~O,gV5b3."'##5)1?y~\өN4& Pn_&G~!{w{ZciMDPQryaVI5cdNсȹIWںǧbܧ #\$V›8hFCQAo/8#34~;kۼFPQ0.YsZ MQauHբ4w_s|xXY$~ڳX[<QX3>RclΐCZa[ua}֓ vqJob'ݕyܟ;Y^Vä#wMIm dA-^/Ni6ӕ2TY47;1;&۩ wHFsCX riQŷscSAd=5㜒:$c`?C&5h+AP!%dq@sx>hn夞bnJ;lG- #2̃UE󪌭{Gnz)3˃*NXK.'\|pa3NO(1 o zo/N'8N-Jy$ l>wSx]z`/*kEꆎ+P۶/~%[20%&GbZ`~1v <|DˆUD GFR<+#QItշ !9ZzIxd*lU ѹE65 .שMdsP@{MSwžMaph'׵Yxzy:@9 k5([Ln!o7MF=%U@đH(vK R h rwQ5(fri[LTБ:Qj'BO,rQ JY@-0`6Н#:}lIg*v)rM›2,f95=̈́ca/ RJЄ mC~䠏dY* 譩X*;U)dT'%F¹g ||#@o3] v~qi0R2mlϞ>0tӄݒơ6a|cJQ=I=nji~+s5ݨ\0Qr!D7О^%92en[Mz_>kzNP$0mF<\֭}kd@HHf ]tlR@dwNJ1Ra# ^oRyC? @<Ӫ"K"-{Pt D59% A1LK_voWKc Ʌ|(K!"#(e^~KKw 8Y6y˿VaWdL"O0e"٫}yxM"rXMs}pBFHͻlob9AEr*(;hV_6BiN}cjtP[3݇xS)@,|g"?E /hbϥm=36DV(F"fGMcq 6 Ne=a~ %8˼,'naiNԳϺuG;I9!ܕ92 PuKΏL|>.d0˪5fyylɁ7!-ۉf)$/E@$Q Yx~"ʛ@4]=ϽE[a\SU&*6`'rUSDi9 =\heÑ~S8Hi##%[r$4ǽFdJmqYkCu/-M Z g B*KE_"btURKK"ݥ^$N8S %!k9dnJ󼄨qW J*3N 1P nGW C"XLG-1ͯ.&|[=EU&Ȋӿ ].޳&1%OC}_^.}&Q{ι7jjYˠW +YDD *lD^(vVk Th;s4c+1zw-ɉ0>p$9*b,6OyBr&X/_yTŔ}>>Px+3c*ByQCb. ) L ώ-g$kc1j5ŽSy${,C6{&\FR_h+ }\k5`޽ TlzNjmsZɁm(WɲT"` ,diUN%AG<7*shƹH"Z&:^^נ+&~@ wf~GWI hi9JiE rg=.9رL bXY@J_?//Ss] 'kB^J+_5 NDb9Qpx\uKfRG"o MC H !7#F%+!% 6 M3zgiqN0讫ܺ[X1580olW xZpTj'9Ed3bWwKܷȒ~7ĹEkٟz/͚mwC>e,P|L;A^Gz5 2DB lWYJ+bNe#~L ˒1n|;Y 99FBJ&54[[Bc8Pwb~nh0f/f!3>יR4dYE⑆JG8V?k- ,z8+pv"Mčr>0Z= (M/enO.&t7l` /7/t4ÃU$&HH_ydωoely){',6N{hT 4߇/׮W",KÚΤEg^"K@S{9܍ 84wȫ U\w"NWaͨ(b`FA[T0 D@Q/i\a5$JSSYn9-5 ~i*+`d"ŧ~?_͂ xC x5C%6LӚ,IkʫFB#zk/@[ end*p"t%͞V/B i%>]<28S\QͰ%NBڥ˩$2cH-@7Я(J!*m>GDu i _Pp8D/s/jͮZfT,NII\!+yVA#Yf̶U X7wZu}G62pIhN_廵az:&p7EYqspR|)Q[Wg M($vʵFF` %b%YIGja՞{6oƲWV>պlOj7|jZZo9!6T!˪;w ]3zM"l+実6v(ifڂA嵣E@XT7j<k[קϝ"/ 䙳*XE(]p'R!ma/JԎ4Á/R EKYvQ :>ˬAߗF?H?YpB7!;] ֆl֭/Xx1+DP9zi-ߜ݈^)p)V YݡȎ}YҀN:5 )ϭ Gke%ↄ9}0ӿ77JWq! proBc0nhFZq@W[H=9w8O;W_IoYg-8ǀvY` պΧ_"'o=Z7sF_1FRoHܧۯԇyLbd&vETDm{`Ju(.&z͇8[\!g{޼ݚj ,_6,c r:U%Vp ̔ygK K#gpPKT bDž}U(OFH .Pn!\Ut<62#갽,1AZQtJƂOOm2GŶ+,̏i!%%%Q&^__FnVlF"|rb8{sޯUP5/̼Irr5,  BdBh9tahwQx7N>w:/y;<|oӯӽU6lMXC^9/OH!M {bA SjPڿIe"p9yAs* %]aƜ H a|MKc`vً ֦ '" zkJݴK L*\_cڋq *d93Hg nOdmT(;V VC9?ەA`H>[e\2LcEu>3%6*hnVaRޖ>j2Wö(w͑*VfзdjalLafj:!լ?CM{S~k(Y`ĨBDk[ik +הK%s.&>)n^ƫt'n"HR[1]?Qf%7,Txy:J#~'ARh;>Y@YL*ֵOpm{ZP2>p5 "K/] L{dliR,FVgڇnrH*˫ ܞk>ŋK䔹ޢޜkeu5q8Vώd7݋Rvh1ۘ=r.c#7Odl_JVQ+?Ti(ijީ"唹q[:¯"H[]-^o&k81I`+kgh_ 1Z0:dch.cs/M"L4_ko^P^,\Zv.7su20nm-&Nq%qC!ժ7Um؏uHb,ERـ[71"6C4Ld;@WXőja?P5`?\^SwEW3 \`ye2QclH E K< _G!N}/^=rtD%CS=)QMEy췴؟eNFda2C\AzEsi2| gq[Wq)מ Z禹ve#Qz1ٌtTC?(\Qr=~Ұ( g``q+i9,sQڭݻCVH_>Чan`:';l4fўbPECLn( n+~w4YN4V1.v:VWk; aic h*6*ٕA%P*84I6gƓ"MGK$Œ^.q}Vxf` #% |To[|R K)JUW`H2XVL&GӚZgAiO@ oի[0S'_\YWmbr ' )mV25"Ɓ:^ _`3W^7StN<*&WS:+6W{'):\\|9WI[mM1ЖBпIc~ IbI|u˰$ ,8X @ C2JK%-G3G *1;G(rI.t!{C&Ӭu9ʳjlLh[.qȗ2\BBȫXiW90m]ﯸ*C8k{FJ% 8{^խ&TԆ gU&SWaИ;Wl5eګ_$ʉR*׃|,<ԭ ntkE'nz|nk~rE`[tXC!,$ I=Dl9BC=|Ơ3 zS$s@M)thcVK^W%k0#,$~?Y3-M~=z.#8FM? `\#-+$3\3[Xܐȍ6r wNSמ] ZF[-%/>t 90"%v~< ϲ}D_B,@Hh@ͮjشQ A…O &/ahkYq^kK}O! uGB+X)g\IIբ)1BpSmvaki$" ˙.V~Q@F.ĵxWg6m^[kMH}Yޝlc`o텛.wDDkb^bfq̛jDx54l9a:ģ'\]'bpgMϱr:rr_$.*0^ᵉ^kH Zӊ&9`nAq-OO&k(?cC{=r$~þb$݆P6~K aNTM /k,>JNTn>kD`eX,c؏29y8:5wi:V|u2Ml`j95 M6#uIZ;wp[:J<#/R1,Y׬0XKzc,icJ.rYn~Bhޕe|# ,ٓNXH- u@P Ai=ňjKjpĽݲf:?'ޏ<9G+kb.8=Z G)DGg8J\;$Y \I>/G]8a<_RLbo[C,=bvM&8S=4[7KVz35  )/Ji}E#iʮn ~*?;o߄ŀ{@>Q^v<8 p+H$t6R67"bk3=%ʥqN]Bc鯕1[u7;}<^X %>pWfmX6#fmy`~Ol E?ځt*ѯ7{BWzmAL3ؿH<:4@ױ(5CA H\+R4) cr o}qiiauOBK8ctp 0+9Z:U{dm;?clh7ɬL1vϟO s$$ `*!)A)jfCYYǻa9FӼp4> >:\WoZmJAk@#4Sybk-A!du Qa Jfݝ~ԯn;emya-.bV2א`7,DfHmևYp⤧}mM"A߿xeGꬊ` "X x"e{/raC:\@\ .P "iJЧ(.f=ǜ}hQ?4ŸϢCEs5$ {xwv(B솛T?꒝Niqkn*ăeU)1xbޓ*E[ ~~|2!ߎ]c2QO\7Wq6K>02*6uvU3A醞ѡ* ;gN>Td6JDm7 IU2kR1_/ړ%V6G\U|&a@ҕ7РGikfasl[ sōr1NL_D͓!$LfJԖz<.V_8m(@jcv ?C֦o\6O{F!aFKF#λCk0sT)T֔&?5vQ{m>{#&oע*҄h+fh>>-ieF-<6DN8]ſۮAӲT [A0 ;Yq$ 3iz#Z ZzKl$ҠTQj뺃d-YQ,XEؤz{*yH7`=:^$Ǯ/uxy:)~euG/@O|/gSY`VCaQ'<gņ&RxPrL--c0 _ q͹&k;*(Ls!DΆW "k|C[fa~x @wmhp"L bsXccS'.6sBgtfx٠\&~BP\'ϭYl1~On-aL&(S) U);y<'7.  P"xPNMzv#"P&4yQѶRa@-^ c o^ގrZ`d*tY(,>RKfMӺ#Mw]ۦa3TFz2 k@ihn=7+Z̀Tf3|X=Kulo&!(\Թ|D.i$UC[h^Q ڞW\3h:yh\T4oBf[OwS)O$ TΛsj4s,v3Ðf <4p+W^9O[ϧ0_&J= !ią/`L=x=Tl40eD{i~z|Ş3 r ?Ad?xjɀ8e~^l рASֈ_t蜹6½1LU$rSk /T`9 xBoEf ǺlId92A߱ %umpv|˖gP-s%6>M1!}аU?`>|= ˿oabôzF]wiV|4g ˵q&m>l#M\Q ;/oĿď,o,7޴6JE*ʈ0sl|b:ea]59(_ic z.9ٶ%͘g5>)q,I$n?'Я0Xk uc +JwWamq:&ۢ.]7y- h-2qƝ# ȱ^%6Rb$ȴjȑoWLZ mwkGZ0(99 cJ jcfWzd|GuCu{07Rq2=*8NCِfvo F]hS#߱)@52~6۾dDVtFHS&U65(}XLhR2~Pi2Wʤ qs^N\G 2:Pe>*iA~)?n}d$WE^7;H4NS4㸥Ig|gM$T)a˜@|EUGIÖf}N&l |7<O)Ŭl[z]vUK^OU}g5j)J j/GZH &D?ouP{˹|lbiosX<7~c݊>H(DMw-(Ń0UfUj1I̟9n-_Dٜj/-~Wmi ]M|=ӷL,0r>%Q^C%A8RR_ِ&1 a@45EBu;`9lq;ϐ@ovWt8`dYq<2 ,5fk{I;P7|U_ V(^J=tN$rT`b/ ZȄ,3N"4\*}Q57! go'b ,ykRtKEYYc!LPzmXQ~]a %MkDq%էi1^#G]] iر]qc3d["]9R7Qy7ɛU;TSZqQŋ6td唍2r^{>4t<x~ VQmd^|SȫYkb4$S`%xxzS5Ą]=؎"M>?| zݻ-*euiAW m'+y0gs\ M5$0a!a=ToqDHij]`. hf#&W.^ňԒS_CV ɻoSI(_zBEd q7if[ _590}5pqkT.ѸuVx\Ε(7#fi/gMiBi$a}. Bmyp; uAr/#$l#r4*D#؎x[ 8I 9a&1YT+8{;n`AO{2+p!цEҘ4$q0}gY)oh}4Q:43ID:cb/[͊Q}֑--W;CSc?),ݫ?5 Mp4Niq#x{/Sy5>B>OtdT0W/$,7$ͷoGfgڥz=]K$ZT]f0'~σ~|cdV6BTi ǦUJ()z'9kvn'`N>&1\7JƋF٬t˽7[Q~4%ztA=B4s/ ^DWx}ݠ:6γOh"^.&^؜>03Tϱ=dEudѤh9:cSz lPW8+"X&L cVEKFx%͂k37|Ku9L(@]/.0lQՋVgZ2M~06BAZKr֩aմS'El.JY;=`3*,!A3mTE:DzL~^ ?oXbhjj% TH9׳NNAKӫ3 CO!WMT_!pRIrx΀R:+*xDͿw~,3% nwb,#;篳b :t#K_V/UczV޼x<3uڠ=ih+]a6?#$D48 ER 󍅿 C`U GJ\+~rCx=>DpB@SGH<-1HPy GaGnLkkovj(ɞC],zL0}6|+*-q+NJZUxII_BZpf$ܟQJ-*_e59Uih縄Aт,dܝ.65iϽ 5ECm `6}"#x2-hgJ?sYRL97vnRfK)\ɘK}F:<-H"IAG}ʟ9HaLH3Z]$u8L UeHĉCou -*e8'SlM}4ad+y 1y?NɈ-SdO <l9BrX%:VMdTI5mYzp5?R˽VhVD'zX ޿"T^."& Kt4%2Ydh."@g`8H80ֽ?b U*=vPm>wA1h;/D6ِY}1]HSA(3{ZD6PY$.jvgww7!}'!d!?THv l0eX=qgy"f(;2s_V&:‗q6$A{W9E¾P^Ɵ}d<_2{@[ ]!EE s >e'ceVSuarV; ,ĆWGmB1?LohPvf2ys,@#DZm+$[)J3眄=8{BmbRD`8Y#һ+!%k- Iࣺ9Κ%]i<䰯 S1dwX3(Vd~I's ] Nn.5"Z8r)<4n(CՈLJE{GG? oeB>j&~ȕEC RWbFj9xοQxFjzSU.E:- OiۯFn_Ic' lq;,ʌk_T=yf!B>,"%/܏+]!Y%اʼbLʄc hF9(T"`~h@M;6!\nu,%ˉh(qyQHkL-4W>4!a1.?g_l~tj]ʅ4Ĺ.wv d^򎅊wKv'&~fJ[UR׭;uy`W ҟ^/|J~Gbґu5k" }5$ =EY e?0×\AK]PM?e"MJ[#1Er55{L||[ ԰urz ,oİͤ`*jۀ ` 2XWE$u/x 3fmpqa\$v8WREʀU}znY^ /xf7Bv@l<,'ƤS!"['R'@~L, 3h/x +Q*º+1&oE[.Z|[]zۛ`/5?Gf[,} ҈- KDvfuviq>#sb!&.ʙ|"F`.B[Z ix 2H;oK32Һ0دH[Ƌ$Y3KOhV6a2q?piuL |<} K2]I@.Pӥcilm*#O$ltfUލĽ<[ Si8-3qAH@Z M_͘(Q?)#0kg~m;o.I)q^wҢ%AT=R[aÐn:1Q;#u,#.pu%e:_ mlyO'k9{/0W%67D`+b+5i`úǽS'm#Ҷ4b"qښ;ِZUEEڙxPfS@(kR ;$ d{h"{$[yVYE%C )8l7s _EY9 `>vuߪ¢8:\ ݃k5C/'lkȀ~LTIfbpoq̪o-gAFְQC1>- @V p+F\#Um_.Y )|HkկGU D84[IXʞp/Thv{U}#ްo4|)FDS@HtZ1֌\5T傣 t8WWÉT:^)W ^U:DݹY>PĻVVf٘vX G>vSy~0Sp) Bʛsoٝ I(.?Wв5kjAP6NZP&3#/bQJZMSĐ㌁1DͻQET7#.>- &Ճ{S#0ap6"-ԍB#8b8F_ :0mdn6ר#R㡵5EXn0F/npg׳۩$K5rM$Y|+z6Hz:\GDuF qIpݹ3Y/0 .cWΣXӡW _If5ДT(y)+A#X*un cA:w/8i1̀TJ9IO$YG|l /':dzi23C0*<.(!pg0ʴvWE;l xɿ#}_O61ŴovI%:]TyljחO۰;{,6ܜ 4γ29@(K"9/ЌoH:3|V'a&w:-{_7˓8 }*o#I baXFw,UxU){qVC+@տȋA<ȗ[).w^7-kT@ ޗ> DϟuJo7es/&]єO H FVfo0, GNHHRCFnO}7qMX1fJu_U`$0.P ᣬJBN^R:mh5Ian43=yW*#I9"L 3 v{ IV^3iP$j]i4Ag!KU[ Kv`"p@mk)~)jyb[# yu" MTHqIkweۖ:YBbfS)O5vi q`~c Sn+;UZԫ-(1ݜ 18xM5=V -[ 2l 0{8vGe?ѝ^@ez]xVo-AqZG)I8TR`BPMedDpݿG\Usj=VERt'ϐz_BdYU55v2Y[:]j[No\4E/dۤ=,spB^'t!]!G=[EWZRmM1ʞ k RHS30u⻠nr `{.Pm8ـTd'#4H!ďމ3^n3>gD)?ưfSG{h/^FXl4}~j򆰓^oh@m1f-9^+, &?Oө{Ko9R's :U% aߑ5r?m%ɜYBhҚ#C֦v 17SH*uc:ЫsFl=~ Gbr'ٝt |Ax'ӪN>9 Wg__3 s3uz˻K 1Mp9L%O?ў1wX k{uaQ(-J;ͽX/Gi4·`3Ys;L >H"l఺cl]C!;F4|pQ~#}CvvNOxZttxY9{1Ji,g;,ᦁ pzU0ĩl\]ίbz>x.9U0j`>$DIjW2F[5Tv)3zS)2xѾ-O@Atp]|{/e6>JIzBsm>'|?x'hk(75;`$kh]w9y |C<Ɋ<.j:Ffxsi%c3l譳_ Ջz~3eqDj o):xa˄[r=b|/~ 7@ aK ,1 o f)n#JV\i"*_S!d[IZ] GYjI3h\3FW/m8،6<)` lٌ2|A eg$SQ`$Tv \WH->1"{5#pBÔI%f<w=DR4>2 +f0G#HoTcn៤kMӾ&Sź<>%3VF;1,Ǜ8+Kjtw;zJ>Z2xZPxW\|x6l! &eՕ?~F_ZhXX FVi ޽)xSD_e$fbZDH w˪`x̥5^+ ߏ,֊nݶ30צ'g:RGNM^rY/_K8f/* B{jN`w)v}q:'ԑ:(Ux ds&B&\VU廀"l2zuJwئvdG1wcc$IĨ&U4eWuxA8y)$0FFoA>qoPa`Νg5u*bn+9rǥFO"  Uϯl1J` 9\3^ t*8{RkEC;&4}7Ni5v_y,=_ \@@YEU=anm аke2){*dIVA}܌3D@o3,wRzYEx};֪j0E7$/\ Y u:LuQfx)$ҒcnVa(rNϗߌ:Ԏ6F.0:_I[g֣LL\lQK LݨF,.B2dL:KS|VD=<|w@gWaxd/^޺6^P.4#w +l } 4>|A Ӫ? gSa+"xJ4QXʥo ` ڗz̊V[NM&Wrd\^e8SCy~/&2,uڭn@qEۯ^0ht&`Uz#xjbv=AX=NHCfq F <{n!a_SZK@`ߵÍ!_RwY٠M>\nh*!>OlR~2]~8XT7fJ $|Wq7HR$%ڬt˅&fK}--@ QC 0~x+CNK6 B^bN5BC@,lVLlU~ر:ETA 7gsQ8yЂ$^pZB2d5kyB+ͶQ/}z :6̠< (\-`  jE)<<6q~Z~X(tKs+Ι SI J w#' QW3dP tMy f6 a-EVQ;#ѯK/t b\Stt}WEA]/85&({tTJq9"~y%oF5HJ+ux(es#i6k9XGCQ̕6PC_Z`޻ݙ*FosV'h6OImqv3D`j3].hfi9ᆥ K_}p}4!9YFA8aJ?@f~9O~D?rcQ jyźمKyF_; Mh[)tvd89'; o5hC1>xP S)Ḧ8l,_T>~'nY'mOo|7Ɗ_~q[TMO؁zY^3IUK.֟$k1&RF2{ϻl5(Oh?lEN}P(zF7!t3IX9}_d,.u]wt,X*]Rye?KnM%WtS]!v2>Z {FhCK_C Y9 ȹ4 Π !X%DYz\~Kf@YP@O]) C3)ֽ3+.u]rZ:J q/="PA ڠ~&yfʈwd%Qv_Uk.z5!vUe@!;ujnxB)*Oϙq۳݉Mc?p:p~-@;iaɎ!hdR(0t`2 16lkP4_Q|w^[B4 *R-?͏ 5 Kiٶ 4X>P6p uAc¢ZƼ9fWN&$&-hhV.fgå9H,_}*')?-nւ>] ["!3djSpV2K&ؔ›u9M+EwuU8n!;DzlGlA "D٬(Yfǵx!tbvK_zIJMmci ÔTjj%a$z0*J -%g">pJ OU}8T0Ѳ۹[![#3'hw/Zj.xs_ĝ.tK6 Mn]AE-ijfvQci |"rT-'^\pGX*'~hZ!RrJnN[m9"V9Q.0 [ 0 b74)ynAM}}5pSKi6=[D/+DxRo'C~<M={]^z kl=&m%c7CC+dUvQ^q}9?Lf/[EЎ址Ja+V;ʶi,+6ohOBól)oow3]Abc?{i+TM JqCذ0dtj`(eD^E^ʗYB8bL@U$ zrD;(ǙXR6_Õw킝ϗnĿ?HY%/䦯>7G]QK{U-J-;B?7gkJ0n쓜;[3D<J͚ם9+@$: xJpq7*hp"+Z.nxߔc/v_RP <1 *Cwނb-nM:,K%r1I"Co_) /W[ @9B Hq4.ict ^p/aXܨ}쫝Wʌ-;}\j)Zq藆ب3S &`؞˒3Psh͐$zFv0}0EEcGR3)&!DFEʢƎOS /cQ/L [>r3 ޭF&UH/f=mSKC +ƪ6A%otO_k-{| aN[i yFJ.[^bL |x3m"T跪@Dѻ90Bų;` {T9 e$tgt٫>e`%@`k) Og94s~>$:VBT8mzBR+xL=etB} % P'9I=-̮[E?ф:Y`R#nͬGJ;kSs|F[VhIks֋aF?GzI)/E@;4RO(2y ^I wrZ.q(O Țbrs"ji~Z')tdߦ\XT{8U(O& Q+5R_.8JS; RRTZzxtRMj}a:Oo"-\ΒEC@åh؝ds/`Ev>:)PctD^f&G~UC fG_G)f k 1= xKQfD#JS:9pBsE}pPǛB]f鬞3`Z >w4/t`jh-!P+5ܸC%*j6z ^7֙.rDdB R` 䰡_ @7?C'wJG?xY B4<1t>H@J8wua -v,(_r!!sR_thtbhF9(r q`G} fP>X@7@dHT/%3ň(DP_aW,GRp#o3 qu 5R \p?_teZ!g+[f  !>#*[=xXKm9N}:>{=Fc?fTq98 JTe͏?x$ "Q$ {.Mdl;<ﶍ%?R/ XR>_z@k(d'zE<3G q6틠'xf)7OTF'Ų =&料(1]k S?Q@[ih\22 &$"aU/ũM\Aer t_.\]Iە\Xo^UZ,C;6@B?~+Z܁* |CѩDӅp84Qtĸm8Qf{?BT#|q.CpcӒq/OnRQ(xq@'sP &oD{3uw$n>bo6n$놆.yMݯ6R?)M v\hS s Z2tEo`;ʼn|{<ۤ^P8ȼL8g6WiK]7 z4DTgt 4BhցTs :ooeW[?];wݡXk$S`4! %D@{bR}pM"\Niqce}潋X |N`|hXB=a3~`9ǷJP9_1hq Ce>$-|bVᰁCTa9qA7Ll&|uxC'e5U`O4'q 8Pswtd%كg`' Q8 OW攥_"O1NëZ @2|D01^EvC=e.( iB@\bYr#*99 zw<:z* ~nڵ{kb7fP읪.oDr:4bg0t-ɗyRvb>Ud-fX")j'Rf,k]L$ʂ"gS@Mo +>_EnH+_ oijRsusm{?ZW'G m0s-IaD³+ZG'/ls\rް$vvutw&!ΙlԖJ}p?NF_`ptj~1鈧v_ӷ/NҌ,P00CҞOE9{& ];"i[ށzcxmOJPQcJ +Ea(,Y[7"p=%e0>ژOio=Uzr&qԡ=!o=KCކ1+MÞf#ߡV.Og%Ӻ nmH㸍.G h( ցa}q}ƀ )NK( $04? *IaO JVb[iȘ^ܒ>3VSS$ :d_,yϒӥ9չ*YF'ϱ ~u1]7!2ZQz<'$ˀ(i*RgԳ66tGϫ$dGj,6{ǟ/يKH/ZV="$Y>%*ÑAuМq5X)*ܥlѬ?V4CPnN GVY+Z4FA;!'fyogUV+":l(1UՅwةy&=mpC%_fq׃7, Z5a3׮~ ceڙwջ?韽WeCf{@ p쐊tlX OLKEH0a)3jRKh-F kL鄋*.5C a.e%l`B6o⎏2quVCJX[jLTD%6#BdfѪL&B 8/пW"&1Lk1EMhOݾ胙)mhtFAgEDL,(+0$%4OU5EVEa;)}C~ok"SAWhMd6'&^ d gNx`Iw|Ϥ[{izF ΢ 5|ǬcK BuKbM -`Q:oybq9Zyr&gm3TMuxG) 1 p #i]YK%wJ+I")ǯgPv86w6ia#go4)*pQ)X *s $x2Evq)1C[RS^Dc g͖°-6/gC6>ܥk,\b%m=RPg ]>KmtPQ a.|H=(_Xs=H1p=JWo|*EV}S$9u0W| )Xwt􆵖^\=8D ~Hؖ+k혊e+i2܊^a-M19YLt&6fWg۬[UΒ v2O]#ٻrQs=o`$4L|Jw$0C!cӿHȿClJi+Ү ;)- PYǮehEwgzT{hE}{9@qJJVDlP-h,l{I6JpNhS"ЖEO^h7-b1 BK[ 6|x7Q<1 c}! D#5]p:D*uReOJs:ʒ^5&v;bɧE5+Al C cV! '1ppE7*&HCf; Wb:v1@2fZ>/_[S'zbg]JF1*/R/wjZN|nyZSJ&۠.Ƙ#ACaJ:d ,Alm,KPE5*ߘ:w_Hj)TaPNiɢ;rt=qSÚ\f@ghkATܒzwcr1Pb|gMv)lU$wv{do)QG R!X@8 \iP]H_ѫ^q6TA[-u0~e^#H }w22bEd?|a歨ҢMP>p~hqbr?۶"=4 6s)CL2(۷4P}p^Ц^\=ukz7`gy.]%sm;w#D'*Oyш!dɪg~Q&&Wd,€b'|޴@sz )c“PN}(fX-Qԩa 59~g@hט^J_}ec1dRU#`jnҙ@APEZ l*rA;zIʇD%-vYKPNG,Ͻv CW^,w#R/ yrL:v~؅ZMh y5g㏖ZCwkb)> 7Fb`?ZH툳)W;RשhES bBK:C7-$sRb:Ab#TymW]fm 5K^] )3)Izp Jd8Sy=+hĀf9{ϣO?q=|kr v hJվVptZ%4#K:#Гb~wSQ jJ|wKU  K^-)K."Ư$ZX@.U{G醙SJDԟm/ js0 X0/2~_T%W_2H#F{@5pe &0EiɄ=ў"|Vd!Pqh?(?%85UOG= 3dI-%GhЯj&eP.r;p9 1k~ X|iɚ'tEނX | MV᲻T'Qd\Y=@RxME>>as1@R|:a OW]" ۑ0ŸS魓n4JPQ-/  Ԝֱ tS24ȾV 2TАӵb RT!$JĖ,HRA->d2p anEy*ė=Mspд/u"D'(Rn{t!.E`ʼn\s}MgӒ Gc>|X% d }JxwRn DD-j0܁굏zM#%]ҠV(me4 ü=FR0 #ӊ^I4]f]YY.H G<ѳiTWˤrÝYJIcf@ūoR}6>̹PPe_ N';15S='.BkZxH~`6+}|8dzuA] MVr3h63;z >@rg= |gT "BT.D؟id5+0,*f)쮺kY%T ~d-Ԝ|ĭdp_ K ^, f3^9h+57MlLcw.gc1u:-Eh~'H!RDP.1.\ZۻP@ GsCP|_&['hh†'RdūwR/ye.S{5Nr'qjNjTBzQgP Y o/LBb})w@lpY,N3f8ݗ"!c*!qUB =Э[2^zԅbXKD1gkMCUT9>jU?j8{{<3N\[%w Ԋq=o+l;SB>OYaN8 !Mz%I",g*Z@e/U88L:gw ]ƸϚq}2(|`n톯{}OHw-0:b]!K%3f7r+D^@0eu$t6,5Տgr,dOwtSv@^Kա [kKFu`BFyNEDN}r[wQP|Ic,S[K-i%|^X2$dB䷻̌NQ]sV_ 8\|+Mg'+˞-"ګ2m5g^95. g?XǽMtzS3hJF!XղuP`v ,+-cqHj/ue|i4 gDͯ\#66!u3iW+.L59& ! `_6W6:>V'jS!#SAۮ(Ɵ ;Jn ObRʏ-3*:[O-րC)h>_= U僩dVRro}(ĨwX UtZnoj3ʹؑ@\x}YUxdF}a*.hCN#UzJAwϻshPj ɜw <{~_02q+;8<>Kz~بɡm23q_mAdfq8+G7>saw;FL^o$|)d^1aO`:TBLoq4MLF[*MrBYsLu+:C(?](bNu]M]r0U^?l6䦤kJѝ?-*,?e'kb!c^2ɀR)GYr13Q=0imMA0unKcNzrJ >WD-aU)OHh/U]3ߘFiGXxӋ`?ρ-NuiA[iA.edzF n+ A0>W6U R<(+_S+I) />J_!a 9ӻftěi1kKU0Zs?Tl D'XMao dd\ivvi4>%=J 獺) TB7EqNC)7} Z3/Օ5flql\OI=,p.8sjOh]$Kwhغh9˙0ퟆuX1xF#BZvYzl6O.>X$?T3E\y/0|Cnlo?}"*KR xo lihrxI"6Q{ ] H9 d]dѴd=qkǹ| S&۹\Bah@ ˊo .w6 D>]Y1e|S&')]̑.|F˞Ep݋G ȳa˕wI<}^+?#5EDVǩjVq`d~ IKx#iQ}:4~yjL!^yhvdjz:$}K$A97l2#//@;F?Hպ (Jݵ(#lbCetޑ?BNS|"|݃^o-zrw?F776.~bnfcL1lzJ;AA@cMTyH|1& %H8DEo.cUGz>GD8)^f!\WIbB ={IΨ|ވHzEU k2sWN0%uT'!˦daZfItсG 4>`)bqu'MiL@;"bYjioGX =>03S(]׭>|A7QTԊ h[Ju}^Xw'R4Ar H'd.ִZ_X_l<؜j&Tۄ0sW&Xh2Ĭ<9u aX}asJ{S NJuZ)l5,A毷-U4g(aST])EɷdQe8al *흥АN ܎.X8WQ26CsA+ v䕓ɒ(on6kȄ}!嫂gt=y~7Ϯw2vrY$C}y!즋oT4)A{s1Qa&T:+Ϻ 5 }O2]"17^&jg5;'iAAsw?%Wc]q}ct>U_oKd vɰ a$mSqi${S+n'irV }zXά3O²6Y*>>*L6-k]q3ݮ+WJ/ c8eDe%^=XMWWeqn|iB s"D^3]U$b(pw?EHIpe (ti&Fȭ0:eTƤ$.v.>8'GpjEc  bF[\KxNBv z"yxkg2hi} }e1mԐۗ Sw-G] )_b4=LuHL3Fu-aRM@dvHIpR 6\xV|~4ZGXCou +(Ȫ7ǽup=F1k!4.]]kF{cO dFه~&{^2GbNaNئ9F⇏p~dpyB&6=טtb:缊 !MAX:5o' Q]ApNwCl:EwoDVj<n(c:Y:ltX/<$u3u (67(qW+|ns9KT~-ݝ3BN%֮Gh#AO@oj@zKcU@q@A)'6P&`P^S4K:t5%i&n_x>4q,vYA.T{_`a8E"]W v9X @lD~O^};Kp:1 Fy*Mcܹ->^Ol5K'fcaVc-ah }A]D4)xKdV6PᎻJ_cʍ;ըZY h`Pp',5xq[d*e!E$tI: pß@}nw}Sj9r6+Oo{zᚆy U"|u2@ٌbb1O"Vmo)bON=Iuc~+$d.ս⽇VxG%RĘ(h)R^y<[i>{ Sl?LeufZ2KIt v?ь[f:PM >iSʉtPŧ,ա*wEsU <)댻; <] Z},VcSB,N^Z<ķ$[PhE^_=V.1\4ᇷ& $kT.=<`Vaĉ>1>50 p=D·Y\mSW=KCJ^Z Y7L \O{9X VA7.!໦ 1C;YuB,&AZZ_A adGv~!z/ssiEh$xs8r"8ײP- 5`XM.LIdS-peag[9-\LlǜT2H/$DFE'JVj?oBm,N>x𪎞2R@FNZIqlL"!pvEFd(:yoItP [uԛ@_6Q쒲fzmOlcAd,v̐(S%K`ΐOmю ͷԦ602! fWFv0e MȞo !ӘZ|~(oU]xm0IռbL-[srۉ2"fM3iXxCWqQzRf@]΢3^GaW2TLyAՏ\s _ M%>~"ƣƑaDMJ5/!:`Y|x6A WvcS6akd>/[HM$g3D%\/$jlDMR 9 < @UԨ`Ԧ+U͠;3D`׎kX ~=Kd+'hԻCgat$.>BmBacm["m0M_,1 >ؐoJuH 57 # gY\kW!)Eh=7SI@R"yaO0\*XZ%qFJkb p5yNϷ.mK4q $gl<`K'^ ]Dž Q݌jLWL);d%f~;x\HiB6KΕ_Z%GwonT?~m1tu&pQ]uG2]!L~ׇC"kyx]؃o_n6f\(LJEܕ~B{͡JP:;-[ԉ, au/,2-4khJ3eTM·mY*JA/!-OYjKjx S!n N,gBeX,RY8'Ѡ-ȵaeɚz~N X\a2!WXwf$%dcŬ~c b$wAE]ޭzfjntI/=!\$nk4kZ?N`{֟Gq^q0Qԥ|S;yZ;+# 8 >u0#e9X']"VML 6Aӈ _UܑwǗ&k'hݏh)2(g'(3$(IsKZ~Zjk6=M_K4',B:! Vj !Xux' Mn̈}0Ş \%lmms5:".s/OwUt)eK{NiIDdyMa[@Pa{6!M~/W_!".hω7cp٨IvY!B/<ۊ%E,+F@!Mc3:Ig=8eX CJE˪2jo3G{;%N 2X1ӊ3}dr8aS렎m-urlz#m~rLr#TʛTj$R)`'t6aIIlke >'hćIVd3WzX/[mZVw^{U‘6Pʻus.t2?7v-ܾd_4E_D8=ǛOm{c1&:[vޭ ntF1"!r`,`ĸֺB~C$x?0Z<|g:]]WqTw] 0B^?Pc)J#+~bjf rlS[B>?<\Pog2]2I96f}ć͍ʜ(EAS;ީZysn\J3NqW K+nG" B}w5I3Bh߯\K:%[N ,8r ;W|כz2F a~I}0@o*~CNJ0_`twO!J/gXh2ԧu(:1ץt$DQ9t#p!#яIxAAi:Y\ 70KHMGCY ^ߎ흏dǚX4/j?u̝Da^m\] Wn~!aVr-e=!^qjj@ Pjd]/`GX ;W`+l",e|!,Tô3bλP1[64?u&G{AudЀNe!HՄ{&j)f#g-S[8 7~lSeaE2.5eYffoy\=SE YP]S}K KQÛngn1p#/շ:aҚ4!>IVAu?(<)s.9 b`#(%qAG OA:dv# )ڱM W5lGdNuu/UV򫖞?7 -w="Ts#Z68#gGzweӫ5C>wTnTN5B>pk{gĎagy.V]ZsrL*Bk`A^r2b&S|vlۣ[1^B}X}rO0ZA}.Jl I!ڡa+96lYq):pG&3}b$ZXH[Տ#\C|']T,Ot`[Q@?•WPCh#[e59Oz{JҲ!?zYӟHnQ~ V;ȯ2Xw@!cVnRX6 <8u & 4uGaZ!gZN%j?q{Ap4Tыh7CYf^ld-ӦQ;w8fH$4$ px}+oL`M덋Myex8})gc AIWm&JЌ ?$  $gˢl<HOqE }#28GHg4,xg{d_P8 boq` ޠXNHGA3x *Y5ڈD 89 Lt7Qʐ>[:sP-_Eˣ:偺~!-AV1^ׄMhE=tBiRq{eB󐾋, 88̇*6U9|Q1Sd`S)sYZ9!>ۓ~k;Z]0~Rq=0dp$Df|MZ eDBwcͪQ9V. zhׅWaLJĨ3Rl* Ŧq^zVB'k}jBP5\4 VU!AC&JHg=3rfM^Xꠉ^{Q&gԾ}jo:1ev) 1M-9T㗇r?ͨb2ҐfV ɈSmxv0*$̓2N)ypd <^E|?7`:NX{=ɘ/ߌ^ zSK7$O?7$/f Hm$P^O][&Io$&^N ʈI?e즮a3=@[;g4&)׆O*&Hto^4[nPcl)P42DRm7gs>̓ ^QcoinA*|NE=,5Oђ >E:Z8^-JdЄ?{$;ZN_\I; I=+0q]5%|+IOq{b9hՊ^a qw"Љ4Zh#<-Ez"&-|CGc'WaƐ_{`OM䎑/":' ſ#뵥(DڸKY`̳iq:ASP64GJM3Yg\c7 Lw9&hk;Y3记l6A­CCg=~Z3SgIdALS7_3] EK`GT4ŐޣR za\h00 k)!oKpA4kՓ;5*8j 8!'GX[<= aF}MhozSnt*' D y7 MMB6Jae]J˻rѡ pj{`T V!2!w -֘ꪂv@DzHrL$YuioдZi{a`G&ĉ?S_iK-MFK#q< s4G'? ems~> 5 3UfIҪ8ľ)1y;版":Zq}ǰo)0Z 5}ZI"܈VNYCҢ#i3g864-& ZQj(57N`;&_>M8b<68MMl'bC%L] } VLShiN]ac:@NcNz9s;QUKա+cV# *oܑ 9eBW/%v,G-g/MЮ`_+7.d֬0U `=_G|3E-flw!49&*Ŀ/ +# " y5I9ޭPs\S6-P)w=Eh4SG#tVMRA 䜠]֛`UU 2,!eD+\.i(ڨ5+S3wQp}<1Ϥ8ǫS4Ň3蓀zԏYplNG)_[bH %ݙfpt+,FvAE{a#5{К+U+B㥸->Mu#2Mu2Ջw!>nwWbr4$^$t*sX"^\SDvNĺZ[`1Y  LSm ;rV.ugm N/zԸ!}j^De8$=T _ѭo9B#1vBԯ:MFcwI}9A5CS.0cyS;4*?4`Vc wy2KL32MPZN(tvzÃku3~Ną@JaB[_vq^bP;+^s͟dSP*S*/$ 1~UsX>@j4Qe#x 4@{]% oP:zo|G@_f-~#ƿM9Sػa|Ug%U9!GvP"M2Y`z1v'r BvՁR,%u Xpp,#`fSFvB!Kg~}az^ZTFwvM9[,n\{칑ke`,cV%7%68T,Ѥh'M]F0%ᚣ{KB@?8H-ܿ~V[a%W;h(>09 Wɽ@~}31}1WDإSO;h4۲%ho+=#]v@gwGI>FAZiϺtFEv0>;H+: u-ZCy,u#9=  T$qn*[VІS@=ήAu?~Rc Myow1 o~4~cE)R‹p&#ʀEXZņC/Pcf>+2?5{Lp?Y?RMQذȢ| irm^fB Т7E%!pXYe, T\.by> Ǟ Wkp\t<@&I`A-&Q2@);Oq&3`,`>+:1-/T oK QtнW R¦ U;qwVVrՂ$׊բבIj98;C_V?b]Vu'ΡWPHP0f{/Kz1N >r,2!0,%͊qpfdڵ֊^wP$@Ξ(V$E(mQ8ptKHKd8r+?>\;@EX~ 3>[DHFa 'P&;4r d4It n]|[&`)ߩVvn$ TALD㌓hqoIGÝa7$sy(H _-?#K{0Vg4!a öRp0/60.iK/ g+k+D-6DlG 05˼ND68 yUvWnٗXqh+ idwQuKFr:iQ#q;˜y+)v"~x@Br}D28^f8|d0G-[]8ґ,&I"_i@6or ? @}>`󜰂&R4 R$P{`»؜ A_עb<(Ԡ/1*ttRTu2H6I擹f|%:{獍e.xH *oazja[؄38p@H&J.^&;sѝe~X.*,yRoﮕNzkam 3$l'u -y^=nKjP79' qN>'9V`F[./`RKg %Ɯ̒QTC{x}8ߎu@ :?{R`)u_q||@ܤ 0%|"wS nδR⻛hPiA zdς6wϭs_RW'NcNl:tyQѰ~NQIU`q IRG' [/EF4VOܓ;Lu6_k,PEe(7JL\HlZۊ ӌV3krJf,bfVXDhLtNN9T6L|**X5X֠~ҩ6E Cpt^ٽoF|1n,/" u&:Q3O׎m4u=8_\˙?޳ g|' eeN:NX+cֻ;Hk~7,lx܅}kvXˇ(lfG?rs~"'8GQWGj8 П'D lńBR10sCԥVfulf^tR?G?K5~s,Z" 28"b,:_˘Ψ ۂ3^8]GL7R|/8%v ƛnҳ;zV”xB,9JUԶLGx|NGs/#ET;E[z"GB! ښOzM?݋QHSCJ{qI$Pl~&R/mT[=&mCL&3rB\"@jqd/{t%3j1MmC,yH~WҌcw?jGV 9XL1f#{;uٷ*;NuO- cL !p{cA5A\FhP4IPP?dʂG2@bo[Վ4%V(?DV,yC ֢܍[$AXe/UytܗMak*>-klʛfD^A\mUVC7,=! ?86Ik4R6Jl5?0g~w]EiV`9p@ߎa{9ftIv]ht9ѥ`6[f`d`XV(dE\p k[7 B^eƲ߆`&R#&h$#f\N6aTF#Gvn/Զ'-I]oa>g%כ")aә_" Rfm&[nMk;YIe҇25aA*Y̾`ItT~bnhAĄ{ $M [B)J9BN,jO.̈́i8ƛgT~ YZ