BSD Commit Log Search

   xerces-c is no longer BROKEN on sparc64 in the new base-clang era

   ieee80211_chan2mode() was replaced by ieee80211_node_abg_mode()
   ok stsp@

   cbms is no longer BROKEN on sparc64 in the new base-clang era

   libvirt is no longer BROKEN on sparc64 in the new base-clang era

   difftastic is no longer BROKEN on sparc64 in the new base-clang era

   lang/clazy is no longer BROKEN on sparc64 in the new base-clang era

   graphics/mapnik: update to 4.3.1.

   Fix typo (too few 0s) in CLIENT_CONTROL_DISCARD, from someone in GitHub issue
   5622.

   Sync to unbound 1.26.1; heavy lifting by sthen

   devel/proj: better fix from upstream

   cf https://github.com/OSGeo/PROJ/issues/4862

   www/nginx: MFC update to 1.30.5

   www/nextcloud/33: MFC update to 33.0.9

   bpfattach2 -> bpfattach, bpf_mtap2 -> bpf_mtap

   mq_dechain() was removed in uipc_mbuf.c 1.300
   spotted by bluhm@

   wsfont_remove() was removed in wsfont.c rev 1.56

   vn_default_error() was removed in vfs_init.c rev 1.21

   uvm_meter() was removed in uvm_meter.c rev 1.49

   usb_task_add -> usb_add_task

   task_l2set -> rtable_l2set

   srp_gc_finalize() was removed in kern_srp.c rev 1.14

   spllowersoftclock() was removed in 2005

   pmap_collect() was removed in 2025

   pmap_clear_referenced -> pmap_clear_reference

   m_tag_init() was removed in uipc_mbuf2.c rev 1.50

   bpf_tap() was removed in bpf.c rev 1.149

   RBT_FOREACH_SAFE_REVERSE -> RBT_FOREACH_REVERSE_SAFE

   Avoid sleeping while recording SEM_UNDO information

   sys_semop(), by way of semundo_adjust() and semu_alloc(), could sleep in
   pool_get(PR_WAITOK) after it had already applied the semaphore operations.
   Another process can remove the set with IPC_RMID while we sleep; we then write
   sempid and sem_otime through the freed semid_ds_kern and sem_base, and record
   an undo entry for a set semundo_clear() has already swept, which panics
   semexit() when the process exits.

   This was introduced as a fallback in rev 1.14, after moving the
   structures from a static array to a pool made the allocation able to
   fail at all.  semop(2) documents ENOSPC, so drop the fallback with the
   extra complexity to work around the sleep.

   semundo_adjust() must then stop freeing an emptied structure in the middle of
   an operation: sys_semop()'s rollback can need it again, and re-creating it
   there is allowed to fail, which is a panic.  Keep it until done2.  That also
   takes a second sleeping pool_put() out of the done: region.


    [7 lines not shown]

   Daemonize rtrd correctly.
   OK job@

   x509_verify: plug memleak in previous

   CID 656293

   ok joshua

   Import libgudev-237

   libgudev is a library with GObject bindings to libudev <...>
   NOT linked to the build yet

   ok robert@