Build date: 1770073203 - Mon Feb 2 23:00:03 UTC 2026 Build cvs date: 1770065658 - Mon Feb 2 20:54:18 UTC 2026 Build id: 2026-02-03.1 Build tags: amd64-regress ports sysupgrade Applied the following diff(s): /home/anton/tmp/robsd/src-sys-em.diff /home/anton/tmp/robsd/src-sys-uhidev-sispm.diff /home/anton/tmp/robsd/src-sysupgrade.diff P distrib/sets/lists/comp/mi P distrib/sets/lists/man/mi P lib/libfuse/Makefile P lib/libfuse/fuse_lowlevel_new.3 P lib/libfuse/fuse_opt.3 P share/man/man4/ddb.4 P sys/ddb/db_command.c P sys/ddb/db_interface.h P sys/dev/ic/qwx.c M sys/dev/usb/uhidev.c P sys/kern/kern_proc.c P sys/kern/sysv_sem.c P sys/kern/uipc_mbuf.c P sys/net/if_mpw.c P sys/net/if_tpmr.c P sys/net/if_veb.c P sys/net/pf.c P sys/net/pfvar.h P sys/net/pfvar_priv.h P sys/netinet/in_pcb.h P sys/sys/mbuf.h P usr.bin/netstat/inet.c P usr.bin/tmux/Makefile P usr.bin/tmux/cmd-list-buffers.c P usr.bin/tmux/cmd-list-clients.c P usr.bin/tmux/cmd-list-panes.c P usr.bin/tmux/cmd-list-sessions.c P usr.bin/tmux/cmd-list-windows.c P usr.bin/tmux/cmd-switch-client.c P usr.bin/tmux/format.c P usr.bin/tmux/mode-tree.c P usr.bin/tmux/paste.c P usr.bin/tmux/server-fn.c P usr.bin/tmux/session.c U usr.bin/tmux/sort.c P usr.bin/tmux/tmux.1 P usr.bin/tmux/tmux.h P usr.bin/tmux/window-buffer.c P usr.bin/tmux/window-client.c P usr.bin/tmux/window-customize.c P usr.bin/tmux/window-tree.c M usr.sbin/bgpd/session.c P usr.sbin/httpd/server_http.c commit BvolIfsvL2RjyCiI Author: schwarze Date: 2026/02/02 20:54:18 tweak previous: * make struct fuse_args argument name consistent * change some ellipses into complete sentences * add a missing word * use .Vt for a type name * use .Fa for struct member names * use "or" instead of "and" below RETURN VALUES * add the missing fuse_mount(3) below SEE ALSO lib/libfuse/fuse_lowlevel_new.3 commit QN6om7dpBlq1jFOB Author: deraadt Date: 2026/02/02 17:15:06 sync distrib/sets/lists/man/mi commit ON74ihXvZIZjprdd Author: gnezdo Date: 2026/02/02 16:59:58 qwx: update ni_rssi from ACK frame RSSI in tx completion The hardware reports ack_rssi in the tx completion status when data frames are acknowledged. Update ni_rssi with this value so ifconfig reports accurate signal strength from the data path. Like Linux ath11k, check WMI_TLV_SERVICE_HW_DB2DBM_CONVERSION_SUPPORT to determine if the value is already in dBm or needs noise floor adjustment. ACK frames may be sent with higher power than beacons, providing a more accurate RSSI reading during active data transfer. ok stsp@ sys/dev/ic/qwx.c commit ZUxi7t7IOhGLtkaH Author: claudio Date: 2026/02/02 15:21:28 Document new stop command for processes. OK kettenis@ share/man/man4/ddb.4 commit VCypiqMe33mUim40 Author: claudio Date: 2026/02/02 15:20:51 Implement a ddb stop command that sends a SIGSTOP to the specified pid. SIGSTOP can never be caught or ignored so there is no need for any tricks to make sure the signal makes it. So this may work better in cases where exiting a process may be too harsh. OK kettenis@ sys/ddb/db_command.c sys/ddb/db_interface.h sys/kern/kern_proc.c commit nI37STBs0yVEuSL7 Author: bluhm Date: 2026/02/02 13:44:21 In server_read_httpchunks() do not blindly enable the bufferevent. This leads to a use-after-free since the bev->readcb() call could free the memory holding the bev right before the bufferevent_enable() call. Reported by Pontus Stenetorp. from clauio@; OK florian@ rsadowski@ this is errata/7.7/019_httpd.patch.sig usr.sbin/httpd/server_http.c commit wwU6aQOnaiogKvVV Author: bluhm Date: 2026/02/02 13:43:30 In server_read_httpchunks() do not blindly enable the bufferevent. This leads to a use-after-free since the bev->readcb() call could free the memory holding the bev right before the bufferevent_enable() call. Reported by Pontus Stenetorp. from clauio@; OK florian@ rsadowski@ this is errata/7.8/013_httpd.patch.sig usr.sbin/httpd/server_http.c commit LrSd6ilYiqSrFv35 Author: claudio Date: 2026/02/02 13:37:33 In server_read_httpchunks() do not blindly enable the bufferevent. This leads to a use-after-free since the bev->readcb() call could free the memory holding the bev right before the bufferevent_enable() call. Reported by Pontus Stenetorp. OK florian@ rsadowski@ usr.sbin/httpd/server_http.c commit pagZGO5OBfjOc9vr Author: nicm Date: 2026/02/02 10:08:30 Break sorting out into a common file so formats and modes use the same code. Also add -O for sorting to the list commands. From Dane Jensen in GitHub issue 4813. usr.bin/tmux/Makefile usr.bin/tmux/cmd-list-buffers.c usr.bin/tmux/cmd-list-clients.c usr.bin/tmux/cmd-list-panes.c usr.bin/tmux/cmd-list-sessions.c usr.bin/tmux/cmd-list-windows.c usr.bin/tmux/cmd-switch-client.c usr.bin/tmux/format.c usr.bin/tmux/mode-tree.c usr.bin/tmux/paste.c usr.bin/tmux/server-fn.c usr.bin/tmux/session.c usr.bin/tmux/sort.c usr.bin/tmux/tmux.1 usr.bin/tmux/tmux.h usr.bin/tmux/window-buffer.c usr.bin/tmux/window-client.c usr.bin/tmux/window-customize.c usr.bin/tmux/window-tree.c commit 07DCG5AYH5wU5zLa Author: tb Date: 2026/02/02 09:21:27 sync distrib/sets/lists/comp/mi commit Nu5xQ7cXczzmqL2N Author: helg Date: 2026/02/02 09:20:01 Fix typos. lib/libfuse/fuse_lowlevel_new.3 commit Mdlp5P7z8pV5kWhs Author: helg Date: 2026/02/02 09:17:06 Add low-level FUSE API man pages to Makefile. lib/libfuse/Makefile commit USpNbqHJx2xjp7RF Author: tb Date: 2026/02/02 09:08:31 netstat: print pf state rather than pf state key after in_pcb.h changes unbreaks tree usr.bin/netstat/inet.c commit VRSJpcuDZgGulL3O Author: helg Date: 2026/02/02 07:14:58 Refined markup, clarified function descriptions, and restored a functional example. OK schwarze@ lib/libfuse/fuse_opt.3 commit iE82WDnbDsEiJb6D Author: dlg Date: 2026/02/02 06:37:55 fix a variable name in a comment. no functional change sys/net/pf.c commit zmTaD6vzejO4K3Mf Author: dlg Date: 2026/02/02 06:33:45 improve pf_find_state lookups by checking the direction of the states. if we have the state from the packet coming into the stack we can ensure the state we look up goes in the opposite direction. sys/net/pf.c commit JiYdVRSzj4CDD8Ue Author: dlg Date: 2026/02/02 06:23:40 use pf_states to link mbufs/inpcbs and forwarded connections together this replaces the links between pf_state_keys and mbufs/inpcbs. pf_states represent the actual connection tracked by pf, while pf_state_keys are more general since they only contain the network addresses. the fact that pf_state_keys exist is an implementation detail in pf rather than a fundamentally useful artifact to the rest of the system. the preference would have been to link things to pf_states rather than pf_state_keys in the first place, but there wasn't enough machinery (eg, refcounts and immutable links to pf_state_keys) on pf_states to link to them directly. this means pf still had to iterate over the states hanging off the pf_state_keys to get to the actual pf_state it needed anyway. discussed with henning@ ok sashan@ jmatthew@ sys/kern/uipc_mbuf.c sys/net/if_mpw.c sys/net/if_tpmr.c sys/net/if_veb.c sys/net/pf.c sys/net/pfvar.h sys/net/pfvar_priv.h sys/netinet/in_pcb.h sys/sys/mbuf.h commit oreNUd8zY2NNDjnu Author: dlg Date: 2026/02/02 05:51:50 avoid work in pf_state_insert if pf_state_key_unref didnt swap keys. ok sashan@ jmatthew@ sys/net/pf.c commit ibOVZ03BEzdeQ1Ir Author: dlg Date: 2026/02/02 00:10:04 have state and source limiter state cleanup assert on the right lock. state and source limiters and they pf state links they're wired up with are protected by the pf lock, not the pf state lock. this is asserted correctly when setting up source and state limiters, but i copy and pasted the wrong assert for the cleanup code. this should fix the spurious "splassert: pf_create_state: want 1 have 0" messages i get on my firewalls. sys/net/pf.c commit ffBQUmvSjcBjLi5Q Author: jsg Date: 2026/02/02 00:04:32 In semctl(2), set pointers to NULL after free(). Avoids false positive warnings with a simple use-after-free checker. ok bluhm@ sys/kern/sysv_sem.c