[SCM] Samba Shared Repository - annotated tag ldb-2.6.1 created

Stefan Metzmacher metze at samba.org
Wed Jul 27 13:03:48 UTC 2022

The annotated tag, ldb-2.6.1 has been created
        at  6c21717513fd56a450fdd0b5a21186a1eb555a4d (tag)
   tagging  f4eb4e6478db2b41acf426a7a6ba2e7130b69b29 (commit)
  replaces  tevent-0.13.0
 tagged by  Stefan Metzmacher
        on  Wed Jul 27 15:03:41 2022 +0200

- Log -----------------------------------------------------------------
ldb: tag release ldb-2.6.1


Andrew Bartlett (1):
      CVE-2022-32746 ldb: Release LDB 2.6.1

Joseph Sutton (13):
      CVE-2022-32746 s4/dsdb/objectclass_attrs: Fix typo
      CVE-2022-32746 s4:dsdb:tests: Add test for deleting a disallowed SPN
      CVE-2022-32746 s4/dsdb/partition: Fix LDB flags comparison
      CVE-2022-32746 s4:torture: Fix LDB flags comparison
      CVE-2022-32746 s4/dsdb/acl: Fix LDB flags comparison
      CVE-2022-32746 ldb:rdn_name: Use LDB_FLAG_MOD_TYPE() for flags equality check
      CVE-2022-32746 s4/dsdb/repl_meta_data: Use LDB_FLAG_MOD_TYPE() for flags equality check
      CVE-2022-32746 s4/dsdb/tombstone_reanimate: Use LDB_FLAG_MOD_TYPE() for flags equality check
      CVE-2022-32746 s4/registry: Use LDB_FLAG_MOD_TYPE() for flags equality check
      CVE-2022-32746 ldb: Add flag to mark message element values as shared
      CVE-2022-32746 ldb: Ensure shallow copy modifications do not affect original message
      CVE-2022-32746 ldb: Add functions for appending to an ldb_message
      CVE-2022-32746 ldb: Make use of functions for appending to an ldb_message

Stefan Metzmacher (46):
      lib/messaging: s/getpid/tevent_cached_getpid
      s3:lib/messages*: s/getpid/tevent_cached_getpid
      lib/util: make use of tevent_cached_getpid() in performance critical code
      s3:profile: make use of tevent_cached_getpid() in performance critical code
      smbd: split out smbd_check_access_rights_fname and call it before SMB_VFS_FGET_NT_ACL
      s3:dbwrap_watch: let dbwrap_watched_watch_state_destructor() use DBG_WARNING()
      s3:dbwrap_watch: use value_valid = false during dbwrap_watched_do_locked_fn()
      s3:dbwrap_watch: s/db_watched_subrec/db_watched_record
      s3:dbwrap_watch: s/dbwrap_watched_subrec/dbwrap_watched_record
      s3:dbwrap_watch: rename struct dbwrap_watched_record variables to 'wrec'
      s3:dbwrap_watch: move wakeup_value to struct db_watched_record
      s3:dbwrap_watch: use dbwrap_record_get_db(rec) instead of state->db
      s3:dbwrap_watch: use struct db_watched_record as rec->private_data for do_locked too
      s3:dbwrap_watch: move 'wrec' from dbwrap_watched_do_locked_state to dbwrap_watched_do_locked_fn
      s3:dbwrap_watch: use dbwrap_record_get_key() to access the key
      s3:dbwrap_watch: only pass struct db_watched_record to dbwrap_watched_record_*() functions
      s3:dbwrap_watch: use backend.{rec,initial_value} instead of subrec[_value]
      s3:dbwrap_watch: add db_record_get_watched_record() helper
      s3:dbwrap_watch: move the do_locked optimization to dbwrap_watched_record_wakeup()
      s3:dbwrap_watch: remove unused dbwrap_watched_do_locked_{storev,delete}()
      s3:dbwrap_watch: split out a db_watched_record_init() helper function
      s3:dbwrap_watch: split out db_watched_record_fini() from db_watched_record_destructor()
      s3:dbwrap_watch: also the fetch_locked case only needs to wake waiters just once
      s3:dbwrap_watch: remove dbwrap_watched_record_wakeup_fn() indirection
      s3:dbwrap_watch: split out a dbwrap_watched_watch_add_instance() helper
      s3:dbwrap_watch: move db_record and db_watched_record to dbwrap_watched_do_locked()
      s3:dbwrap_watch: don't use talloc_tos() for messaging_filtered_read_recv()
      s3:dbwrap_watch: let dbwrap_watched_watch_recv() use tevent_req_received()
      s3:dbwrap_watch: remove unused dbwrap_watched_do_locked_state.status
      s3:dbwrap_watch: define/use DBWRAP_MAX_WATCHERS
      s3:dbwrap_watch: prepare dbwrap_watched_record_storev() to store watchers if requested
      s3:dbwrap_watch: filter out records with empty payload during traverse
      s3:dbwrap_watch: let dbwrap_watched_delete() call dbwrap_watched_record_storev(num_dbufs=0)
      s3:dbwrap_watch: use dbwrap_watched_record_storev() to add a new watcher
      s3:dbwrap_watch: remove a watcher via db_watched_record_fini()
      s3:dbwrap_watch: allow callers of dbwrap_watched_watch_send/recv() to manage the watcher instances
      s3:g_lock: avoid calling g_lock_store() from g_lock_cleanup_dead()
      s3:g_lock: always call g_lock_cleanup_shared() before getting stuck on lck.num_shared != 0
      s3:g_lock: avoid a lot of unused overhead using the new dbwrap_watch features
      s3:g_lock: remember an unique_lock_epoch similar to unique_data_epoch
      s3:g_lock: try to keep the watch instance during g_lock_watch_data()
      s3:smbXsrv_client: only change the dbwrap_watch instance when the record has changed
      s3:smbXsrv_session: introduce smb2srv_session_close_previous_cleanup()
      s3:smbXsrv_session: only change the dbwrap_watch instance when the record has changed
      s3:dbwrap_watch: only notify the first waiter
      s3:dbwrap_watch: call dbwrap_watched_trigger_wakeup() outside of the low level record lock


Samba Shared Repository

More information about the samba-cvs mailing list