BSD Commit Log Search

   Remove unused isa_dmamap_destroy().

   Remove mrt6_mrouter_ver since it is unused and will never be used.

   Our MRT6 code has no compat fallback for the old (> 25year) mrt6msg
   format. We only use the fixed version with the alignment fix.
   So there is no need to track mrt6_mrouter_ver since it is always MRT6_INIT
   and that will never change. So burn this bridge for good.

   OK bluhm@

   fallback to PCI reset on x86

   Attempt system reset with writes to a memory region used for PCI.
   Both Intel and AMD document reset at the 0xcf9 address.

   Adapted from FreeBSD by Gert-Jan Vons to fix reset on a
   Fujitsu Futro S940.

   ok deraadt@

   Don't free the io queue memory when suspending, so it doesn't have to be
   reallocated on resume.  This removes a potential sleep point in the resume
   path.   The queue is still recreated on the controller, so its software state
   has to be reset instead.

   ok dlg@ deraadt@

   Use nvme_ccb_get/put directly for queue create and delete commands.
   These are issued in situations where we know all ccbs are on the free list,
   and we wouldn't be able to sleep if they weren't, so going through the io
   pool doesn't help.

   ok dlg@ deraadt@

   Avoid using a large stack buffer by referencing the FDT blob directly in
   OF_getindex().

   ok deraadt@, phessler@

   Take the kernel lock around pgo_refernce() call; not all of our uvm_object
   implementations are mpsafe (in particular the vnode one).  From Acts1631.

   ok claudio@

   Avoid using malloc(9) when walking lists of clocks, PHYs, mailboxes,
   hardware locks and power domains.  This helps avoiding sleeping in
   DVACT_SUSPEND and DVACT_RESUME.

   ok phessler@

   Implement a different fix for the use-after-free in filt_procattach
   from rev 1.207.

   Instead of shuffling locks around take a process reference in this
   code to prevent the process from exiting under us while sleeping
   for the kqueue_ps_list_lock rwlock. This is the same solution that
   was used for a similar issue in fork1.

   Reported by Acts1631 and OpenAI
   Patch provided by OpenAI
   OK mvs@

   Plug another mbuf leak. In ip6_forward in the ip6_forwarding == 2 case
   the mbuf is not freed if it was not put through ipsec.
   Unlike the similar IPv4 code in ip_output the version in ip6_forward
   needs an m_freem call before the goto senderr.

   OK jan@ mvs@

   Do not expose msg_first and msg_last kernel pointers from msqid_ds
   structure to the userland.

   ok deraadt millert

   Move pf_sysctl() out of sysctl_vslock and out of kernel lock. We do
   copyout() from local variable, the pf_status content is mp-safe.

   ok deraadt sashan

   add Q flag/key to show pledge/chroot/unveil usage

   Show promises in COMMAND and ps(1) state letters pertaining to pledge,
   chroot and unveil usage in PRIV instead of TIME.

   To quickly identify programs (un)aware of their privileges and thus
   in extend the system's security.

   A non-root user it not enough and alternatives like ps(1) -o state,pledge
   in tree-mode or wrapped in watch(1) are not nearly as ergonomic as top(1).

   p and P are already in use, Q has no meaning, but is free.

   OK deraadt

   Sync with libc bsearch.c 1.9; now also matches the copy in
   dev/pci/drm/include/linux/bsearch.h

   Remove badaddr_read{,1,2,4} - unused since OpenBSD/armish removal 10 years ago.

   vio: fix control queue handling during reset

   At virtio_reset(), the features are reset, so we need to check for
   control queue availability in a different way.

   After reset, we first need to re-enable the control-queue before we can
   use it for configuring multiqueue.

   ok jan@

   Remove duplicate fortune.

   From Egor Gushcha

   Fix mbuf leak in rip_output when in_pcbselsrc fails.

   Reported by Tom Smyth
   Ok henning@

   bail out in fetchname() if out_of_memory

   fetchname() is the only caller of savestr() that doesn't handle the
   return NULL + set out_of_mem.

   ok kirill@

   swap conditional in another_hunk() to deindent; ok deraadt@ kirill@

   For MRT_INIT the value v must be 1. Any other value results in an error.

   OK deraadt@

   Work around RTL8125 and RTL8125B TX hang on short UDP frames.

   These chips look for PTP messages in short UDP frames without regard
   to IP fragmentation.  A frame shorter than 175 bytes that reads port
   319 or 320 at the UDP destination port offset with fewer than 47 bytes
   of UDP data following hangs the TX.

   Pad such frames to 47 bytes of UDP data in software.  Also pad every
   frame shorter than 60 bytes, since the chip doesn't checksum those
   correctly, and compute the checksums in software for padded frames.

   Problem reported and fix tested by Tomas Slobodnik.

   httpd: remove MAX_IMSG_DATA_SIZE

   Discussed and OK claudio@

   httpd: convert imsg handling to imsg_get_ and remove IMSG_SIZE_CHECK macro

   Second attempt, since the first one still had a bug. I overlooked the
   fact that in "config_getserver", the imsg data was passed through
   again to "config_getserver_config" and read there once more.

   OK claudio@

   fix up how aggr tries to scavange a flowid for the mbuf.

   Samuel Thibault on misc@ noticed that aggr(4) wasn't balancing
   traffic over ports like trunk(4) did, because aggr(4) relies on
   mbuf flowids where trunk(4) parses and hashes packets onto ports
   as a matter of course.

   if the ports don't do RSS hashing and set a flowid, aggr(4) was
   supposed to use the port interface index as a cheap way to populate
   a flowid, which it did, but i forgot to also
   SET(m->m_pkthdr.csum_flags, M_FLOWID).

   this tweaks the use of flowids in the transmit path too. if a flowid
   isnt set it'll fall back to the mbuf receive interface index. locally
   generated traffic should have a flowid set most of the time cos
   most traffic is tcp, and tcp generates flowid.

   pf(4) pf_state_import() can be bit more paranoid

   Add sanity check for queue priority assigned to state
   via 'set prio' rule action. The priority must fit
   to bounds used by kernel (<= IFQ_MAXPRIO)

   Issue reported and patch submitted by Acts1631

   OK renaud@, OK henning@

   +smtpd

   import initial smtpd regress suite

   based on bluhm@'s relayd regress; bulk of the work done by doom (tagram
   at gmail) with tweaks by me.

   unveil: Limit count of visited covers to ps_uvvcount, this avoids an
   infinite loop with a particular construction of unveils.

   from Acts1631
   from dgl@

   this is errata/8.0/009_unveil.patch.sig

   unveil: Limit count of visited covers to ps_uvvcount, this avoids an
   infinite loop with a particular construction of unveils.

   from Acts1631
   from dgl@

   this is errata/7.9/044_unveil.patch.sig