netpgp_t is a structure

(sevan)

Update modules' required lists to take advantage of the new unlimited
number of requirements permitted.

(pgoyette)

Adapt to new dynamically-allocate required-module array

(pgoyette)

Remove fixed limit on number of modules which can be "required" by
a module.  Instead, allocate the array dynamically, and expand it
when needed.

XXX Note the modctl(2)'s MODCTL_STAT routine needs to be updated
(and versioned) to handle the unlimited size.  For now, the kernel
will simply truncate the required list if it doesn't fit in the
existing export structure.

(pgoyette)

add commented out debugging options.

(christos)

Avoid double frees (thanks asan)

(christos)

make debugrm compile again.

(christos)

Add a round_rate() callback for the sunxi clock domain.
Add a sunxi_ccu_display.c file with helpers for setting up display engine
clocks.
for fractional clocks, rename frac_en to div_en, I got the logic inverted.
Adjust tcon0-ch0, tcon0-ch1, tcon1-ch0 and tcon1-ch1 definitions to
automatically select a parent. tcon0 hardcoded to pll3 and tcon1 to pll7.
Define a round_rate() callback for these clocks, as well as fractional clocks.
Hardcode debe clocks parent to pll5.

(bouyer)

As discussed on tech-kern@ 10 days ago, add a clk_round_rate() method,
which returns the rate that would be used by this clock if clk_set_rate()
was called. Used by drivers (or other clocks) which have their own divider
and need to know the parent's clock capabilities to compute the best
parameters.

(bouyer)

Fix lvds output: PD10->19 are for lvds1, not lvds0. But they may be used
by tcon0 when it is set to dual-LVDS mode.

(bouyer)

handle log being before timestamp...

(christos)

Handle new timestamp sequences.

(christos)

Add the ability to prepend a timestamp [ sec.nsec] relative to boottime
in kernel messages if KLOG_TIMESTAMP is enabled.

(christos)

Instead of expanding the syslog level into <level> early, defer expansion
inside putchar; extract the actual character addition function to a separate
function.

(christos)

Move the add char portion to its own function.

(christos)

PR misc/53145 (Bruce Lilly)

Use ps -A instead of ps -ax (-A means -ax, but -A is posix, -x is not)
Use ps -o args instead of ps -o command (same reason).

This makes no difference when the ps used is /bin/ps on NetBSD, but
can make a difference when some other ps command gets invoked instead.

XXX pullup -8

(kre)

fix build error

(ryo)

Change the check to be <= instead of <. This fixes one occurrence of an
apparently widespread division-by-zero bug in our TCP code: if a user adds
huge IPv6 options with setsockopt, and if the total size of the options
happens to be equal to the available space calculated for the TCP payload,
t_segsz gets set to zero, and given that we then divide several things by
it, the kernel crashes.

(maxv)

Reorder and style, for clarity.

(maxv)

In the data returned by MODCTL_STAT, include alias information for
modules that are still on the built-in list.

(pgoyette)

Tickets #1540 and #1541

(martin)

Pull up following revision(s) (requested by maxv in ticket #1541):

sys/netinet6/raw_ip6.c: revision 1.161

Fix use-after-free, the first m_copyback_cow may have freed the mbuf, so
it is wrong to read ip6->ip6_nxt.

(martin)

Tickets #1540 and #1541

(martin)

Pull up following revision(s) (requested by maxv in ticket #1541):

sys/netinet6/raw_ip6.c: revision 1.161

Fix use-after-free, the first m_copyback_cow may have freed the mbuf, so
it is wrong to read ip6->ip6_nxt.

(martin)

Tickets #1540 and #1541

(martin)

Pull up following revision(s) (requested by maxv in ticket #1541):

sys/netinet6/raw_ip6.c: revision 1.161

Fix use-after-free, the first m_copyback_cow may have freed the mbuf, so
it is wrong to read ip6->ip6_nxt.

(martin)

Pull up following revision(s) (requested by maxv in ticket #1540):

sys/netinet6/ip6_forward.c: revision 1.91 (via patch)

Fix two pretty bad mistakes. If ipsec6_check_policy fails m is not freed,
and a 'goto out' is missing after ipsec6_process_packet.

(martin)

Pull up following revision(s) (requested by maxv in ticket #1540):

sys/netinet6/ip6_forward.c: revision 1.91 (via patch)

Fix two pretty bad mistakes. If ipsec6_check_policy fails m is not freed,
and a 'goto out' is missing after ipsec6_process_packet.

(martin)

Pull up following revision(s) (requested by maxv in ticket #1540):

sys/netinet6/ip6_forward.c: revision 1.91 (via patch)

Fix two pretty bad mistakes. If ipsec6_check_policy fails m is not freed,
and a 'goto out' is missing after ipsec6_process_packet.

(martin)

Tickets #1590 and #1591

(martin)

Pull up following revision(s) (requested by maxv in ticket #1591):

sys/netinet6/raw_ip6.c: revision 1.161

Fix use-after-free, the first m_copyback_cow may have freed the mbuf, so
it is wrong to read ip6->ip6_nxt.

(martin)

Tickets #1590 and #1591

(martin)

Pull up following revision(s) (requested by maxv in ticket #1591):

sys/netinet6/raw_ip6.c: revision 1.161

Fix use-after-free, the first m_copyback_cow may have freed the mbuf, so
it is wrong to read ip6->ip6_nxt.

(martin)

Tickets #1590 and #1591

(martin)

Pull up following revision(s) (requested by maxv in ticket #1591):

sys/netinet6/raw_ip6.c: revision 1.161

Fix use-after-free, the first m_copyback_cow may have freed the mbuf, so
it is wrong to read ip6->ip6_nxt.

(martin)

Pull up following revision(s) (requested by maxv in ticket #1590):

sys/netinet6/ip6_forward.c: revision 1.91 (via patch)

Fix two pretty bad mistakes. If ipsec6_check_policy fails m is not freed,
and a 'goto out' is missing after ipsec6_process_packet.

(martin)

Pull up following revision(s) (requested by maxv in ticket #1590):

sys/netinet6/ip6_forward.c: revision 1.91 (via patch)

Fix two pretty bad mistakes. If ipsec6_check_policy fails m is not freed,
and a 'goto out' is missing after ipsec6_process_packet.

(martin)

Pull up following revision(s) (requested by maxv in ticket #1590):

sys/netinet6/ip6_forward.c: revision 1.91 (via patch)

Fix two pretty bad mistakes. If ipsec6_check_policy fails m is not freed,
and a 'goto out' is missing after ipsec6_process_packet.

(martin)

Tickets #679, #680, #681, #682

(martin)

Pull up following revision(s) (requested by mrg in ticket #682):

tools/gcc/gcc-version.mk: revision 1.10 (via patch)
external/gpl3/gcc.old/dist/gcc/config/i386/i386.c: revision 1.10
external/gpl3/gcc.old/dist/gcc/config/i386/i386-protos.h: revision 1.6
external/gpl3/gcc.old/dist/gcc/config/i386/constraints.md: revision 1.6
external/gpl3/gcc.old/dist/gcc/doc/invoke.texi: revision 1.7
external/gpl3/gcc.old/dist/gcc/config/i386/i386.md: revision 1.7
external/gpl3/gcc.old/dist/gcc/doc/extend.texi: revision 1.7
external/gpl3/gcc.old/dist/gcc/config/i386/i386-opts.h: revision 1.4
external/gpl3/gcc.old/dist/gcc/config/i386/predicates.md: revision 1.6
external/gpl3/gcc.old/dist/gcc/config/i386/i386.h: revision 1.7
external/gpl3/gcc.old/dist/gcc/config/i386/i386.opt: revision 1.7
(with external/gpl3/gcc.old/ -> external/gpl3/gcc/)

add the spectre mitigation options for x86:

  -mindirect-branch=<choice>
  -mfunction-return=<choice>
  -mindirect-branch-register

the values for 'choice' are "keep" (default, existing behaviour),
"thunk", "thunk-inline", and "thunk-extern".

as taken from the Ubuntu port of these changes in their
ubuntu:gcc-5_5.5.0-8ubuntu1.diff.  i've also included the doc
updates that are missing from ubuntu from gcc itself.
i've tested both i386 and amd64 fairly heavily with these options
enabled in both kernels and userland, atf runs and hundreds of
package builds.

bump the NetBSD GCC version.  both GCC 5 and 6 got x86 spectre
mitigation code, and both are now "nb2 20180327".

(martin)

Pull up following revision(s) (requested by maxv in ticket #681):
sys/arch/x86/include/cpu.h: revision 1.90
sys/arch/x86/x86/identcpu.c: revision 1.71
Retrieve cpuid.7:%edx.

(martin)

Pull up following revision(s) (requested by maxv in ticket #680):

sys/netipsec/xform_ah.c: revision 1.87
sys/netipsec/xform_ah.c: revision 1.77

Reinforce and clarify.

Reinforce this area, make sure the length field fits the option. Normally
it always does because the options were already sanitized earlier.

(martin)

Pull up following revision(s) (requested by kamil in ticket #679):

sys/kern/kern_proc.c: revision 1.211

Make sysctl_doeproc() more predictable

Swap the order of looking into zombie and all process lists, start now
with the zombie one. This prevents a race observed previously that the
same process could be detected on both lists during a single polling call.

While there:
- Short-circuit break for KERN_PROC_PID, once a pid has been detected.
- Removal of redundant "if (kbuf)" and "if (marker)" checks.
- Update of comments regarding potential optimization, explaining why we
  don't want to it as of now. Performance gain from lookup call vs
  iteration over a list is neglible on a regular system.
- Return ESRCH when no results have been found. This allows more easily
  to implement a retry or abandon algorithm.

This corrects races observed in the existing ATF ptrace(2) tests, related
to await_zombie(). This function was expecting to check whether a process
has been transformed into a zombie, however it was causing occasional
crashes as it was overflowing the return buffer, returning the same pid
twice: once from allproc list and the second time from zombieproc one.

Fix suggested by <christos>
Short-circuit break suggested by <kre>

Discussed on tech-kern.

Sponsored by <The NetBSD Foundation>

(martin)

Add initial support for ARMv8 (AARCH64) (by nisimura@ and ryo@)

- sys/arch/evbarm64 is gone and integrated into sys/arch/evbarm. (by skrll@)
- add support fdt. evbarm/conf/GENERIC64 fdt (bcm2837,sunxi,tegra) based generic 64bit kernel config. (by skrll@, jmcneill@)

(ryo)

factor out some repeated code and simplify the logputchar function.

(christos)

typo in comments

(maxv)

There was only one G3 PowerBook with Firewire support, the Pismo. Issue is not
exclusive to the Pismo, confirmed on a G4 Titanium PowerBook. Chip is a Texas
Instruments TSB41AB1.

(sevan)

PG3_FW device has been renamed UNINORTH_FW in pcidevs.

(sevan)

Listed UniNorth device is in fact the UniNorth 2 interface, rename.
https://pci-ids.ucw.cz/read/PC/106b/0031

(sevan)

regen

(sevan)

What we have listed as the PowerBook G3 Firwire device is actually the UniNorth
Firwire interface which is not exclusive to the PowerBook G3.
It can be found in the G3 iMac DV for example
http://dmesgd.nycbug.org/index.cgi?do=view&id=1124

Rename the PBG3_FW device UNINORTH_FW & set the previously listed UNINORTH_FW to
UNINORTH2_FW which is what the device actually is.

Verified using https://pci-ids.ucw.cz/read/PC/106b

(sevan)

Drop the USE_LIBTRE build option

libtre cannot be used any more as a replacement for regex(3).
Tt does not build and the library development is stalled in upstream.

Proposed on mailing list.

Sponsored by <The NetBSD Foundation>

(kamil)

Only need to #include sysctl.h once

(pgoyette)

Better entry for 659

(bouyer)

ticket 659

(bouyer)

Pull up following revision(s) (requested by martin in ticket #659):
lib/libm/src/s_scalbn.c: revision 1.19
Add missing alias for ldexpl on ! __HAVE_LONG_DOUBLE architectures.
Pointed out by Hal Murray on port-arm.

(bouyer)

Tickets #665, #674 - #678

(martin)

Pull up following revision(s) (requested by maxv in ticket #678):

sys/arch/x86/include/specialreg.h: revision 1.115-1.117,1.120

Add IC_CFG.DIS_IND: "Disable Indirect Branch Predictor". Available (at
least) on AMD Families 10h, 12h and 16h.

Add the IBRS and STIBP MSRs.

... and also add IBPB ...

Add RDCL_NO and IBRS_ALL.

(martin)

Pull up following revision(s) (requested by maxv in ticket #677):

sys/netipsec/ipsec_input.c: revision 1.55

Fix the iteration: IPPROTO_FRAGMENT options are special, in the sense
that they don't have a 'length' field. It is therefore incorrect to
read ip6e.ip6e_len, it contains garbage.

I'm not sure whether this an exploitable vulnerability. Because of this
bug you could theoretically craft 'protoff', which means that you can
have the kernel patch the nxt value at the wrong place once the packet
is decrypted. Perhaps it can be used in some unusual MITM - a router that
happens to be between two IPsec hosts adds a frag6 option in the outer
IPv6 header to trigger the bug in the receiver -, but I couldn't come up
with anything worrying.

(martin)

Pull up following revision(s) (requested by maxv in ticket #676):

sys/netinet/in_proto.c: revision 1.127
sys/netinet6/in6_proto.c: revision 1.122

Add the PR_LASTHDR flag on the PFsync and CARP entries. Otherwise a
"require" IPsec policy is not enforced on them, and unauthenticated
packets will be accepted.

Tested with a require-AH configuration. Sent on tech-net@, no comment.

(martin)

Pull up following revision(s) (requested by maxv in ticket #675):

sys/netinet/ip_icmp.c: revision 1.168

Fix a possible buffer overflow in the IPv4 _ctlinput functions.

In _icmp_input we are guaranteeing that the ICMP_ADVLENMIN-byte area
starting from 'icp' is contiguous.

        ICMP_ADVLENMIN = 8 + sizeof(struct ip) + 8 = 36

But the _ctlinput functions (eg udp_ctlinput) expect the area to be
larger. These functions read at:

        (uint8_t *)icp + 8 + (icp->icmp_ip.ip_hl << 2)

which can be crafted to be:

        (uint8_t *)icp + 68

So we end up reading 'icp+68' while the valid area ended at 'icp+36'.

Having said that, it seems pretty complicated to trigger this bug; it
would have to be a fragmented packet with half of the ICMP header in the
first fragment, and we would need to have a driver that did not allocate
a cluster for the first mbuf of the chain.

The check of icmplen against ICMP_ADVLEN(icp) was not sufficient: while it
did guarantee that the ICMP header fit the chain, it did not guarantee
that it fit 'm'.

Fix this bug by pulling up to hlen+ICMP_ADVLEN(icp). No need to log an
error. Rebase the pointers afterwards.

(martin)

Pull up following revision(s) (requested by sevan in ticket #674):

share/man/man4/man4.macppc/snapper.4: revision 1.5

Add email addresses to author names.

snapper(4) was enabled by default in r1.208 of GENERIC over 12 years
ago, drop the bugs section about needing more testing.

(martin)

Pull up following revision(s) (requested by maxv in ticket #665):

sys/netinet6/icmp6.c: revision 1.215

Style, and four fixes:

* Remove the (disabled) IPPROTO_ESP check. If the packet was decrypted it
  will have M_DECRYPTED, and this is already checked.
* Memory leaks in icmp6_error2. They seem hardly triggerable.
* Fix miscomputation in _icmp6_input, the ICMP6 header is not guaranteed
  to be located right after the IP6 header. ok mlelstv@
* Memory leak in _icmp6_input. This one seems to be impossible to trigger.

(martin)

Add compat_12 to the alias list, and add compat_80 entry points to the
init-fini list.

(pgoyette)

create the compat_12 module

(pgoyette)

Rename spectreV2 -> spectre_v2, and introduce spectre_v1 (which defaults
to not-mitigated).

This gives the user an easy way to find out whether the system is
vulnerable:

machdep.spectre_v1.mitigated
machdep.spectre_v2.mitigated

They are also available on i386.

(maxv)

Rather than allocating and freeing the recursion stack entry from the pool,
just use an entry allocated on the procedure's stack.  The recursion entry
is very short-lived anyway, and always gets freed before the procedure
exits.

(pgoyette)

Reorganize to simplify.

(maxv)

Add #ifdef, for i386 not to panic.

(maxv)

Shrink MAXDSIZ to 512MB.
XXX this should be depending of amount of physical memory.

(ragge)

Set max/dfl size for vm a process may have to the same as for data.
This avoids a problem where mmap may lock processes in the system,
and solves PR port-vax/28379.

(ragge)

Regen from HADES.in rev 1.16:

> Remove duplicated devices already defined in the common GENERIC.in.

(tsutsui)

Remove duplicated devices already defined in the common GENERIC.in.

(tsutsui)

Regen to denote GENERIC.in rev 1.115 in "Created from:" lines.

(tsutsui)

Re-enable inclusion of CPU_UCODE for i386/amd64.  Even though it won't
build correctly on XEN (won't use the XEN ABI), it works fine on real
hardware.

(pgoyette)

Create the compat_util module for use by version-specific compat code.
(It's automatically included in the monolithic compat module, as well
as in kernels with built-in compat code.)

(pgoyette)

Import fixes from HEAD

(pgoyette)

Use an SLIST-based stack rather than a statically-allocated array for
recursion control.  This eliminates the MAXDEPTH constraint.

XXX We still have a static limit on MAXMODDEPS

(pgoyette)

correct typo: and and -> and

from chris28.

(maya)

correct typo: and and -> and (comments only)

heads up on this being a common typo from chris28.

(maya)

Improve the detection. Future generations of Intel CPUs will have a bit to
say they are not affected by Meltdown.

(maxv)

Retrieve cpuid.7:%edx.

(maxv)

Add RDCL_NO and IBRS_ALL.

(maxv)

Add email addresses to author names.
snapper(4) was enabled by default in r1.208 of GENERIC over 12 years ago, drop
the bugs section about needing more testing.

(sevan)

Add email address for author.

(sevan)

Remove Tn, comment out mediabay(4) because it doesn't exist.

(wiz)

Use An macro for author name.
awacs(4) has qualified to be enabled by default in GENERIC config as of r1.343.
It should get a whole lot more testing so drop the note in BUGS section.

(sevan)

awacs(4) is enabled by default in GENERIC as of r1.343.

(sevan)

Reset fail counter when link goes down so that next session starts
with the correct retry count.

(mlelstv)

add prototypes, validate ifm_change and ifm_status vectors. NFC.

(mlelstv)

Handle sector sizes != 512, check partition overlaps, improve guessing
of filesystem parameters.

(mlelstv)

PR/51418: Jose Luis Rodriguez Garcia: Fix incore src/sbin/fsck_lfs/bufcache.c
XXX: pullup-8, pullup-7

(christos)

Tickets #660-#664, #666-#673

(martin)

Pull up following revision(s) (requested by msaitoh in ticket #673):

sys/dev/pci/ixgbe/if_sriov.c: revision 1.2
sys/dev/pci/ixgbe/ixgbe.c: revision 1.135
sys/dev/pci/ixgbe/ixgbe.c: revision 1.136
sys/dev/pci/ixgbe/ixgbe.c: revision 1.137
sys/dev/pci/ixgbe/ixgbe_common.c: revision 1.17
sys/dev/pci/ixgbe/if_bypass.c: revision 1.3
sys/dev/pci/ixgbe/ixgbe_common.c: revision 1.18
sys/dev/pci/ixgbe/ixgbe_common.h: revision 1.10
sys/dev/pci/ixgbe/ixgbe.h: revision 1.36
sys/dev/pci/ixgbe/ixgbe_osdep.c: revision 1.3
sys/dev/pci/ixgbe/ixgbe.h: revision 1.37
sys/dev/pci/ixgbe/ix_txrx.c: revision 1.36
sys/dev/pci/ixgbe/ixgbe_type.h: revision 1.32
sys/dev/pci/ixgbe/ixgbe_vf.h: revision 1.12
sys/dev/pci/ixgbe/ixgbe_sriov.h: revision 1.2
sys/dev/pci/ixgbe/ixgbe_osdep.h: revision 1.20
sys/dev/pci/ixgbe/ixv.c: revision 1.88
sys/dev/pci/ixgbe/ixgbe_phy.c: revision 1.15
sys/dev/pci/ixgbe/ixv.c: revision 1.89
sys/dev/pci/ixgbe/ixgbe_x540.c: revision 1.13
sys/dev/pci/ixgbe/ixgbe_x550.c: revision 1.8
sys/dev/pci/ixgbe/ixgbe_x550.c: revision 1.9

Add some changes from ix-3.2.17.tar.gz and r328265. Not fully synchronized.

Some others (e.g. sfp cage interrupt and bypass adapter  stuff) will be merged
later:

- Initialize firmware command buffer correctly in ixgbe_read_ee_hostif_X550()
  and ixgbe_read_ee_hostif_buffer_X550(). These functions are used when reading
  NVM.
- Fix a bug that ixgbe_mng_present() misunderstand management capability on
  X550 and newer on some environment. X550 changed FWSM bit definition.
  See X540 document and X550's document and compare them.
- Fix checksum calculation in ixgbe_set_fw_drv_ver_generic(). This function is
  not used in NetBSD.
- Add some unused funtions.
- Whitespace fix.
- Check offset correctly in ixgbe_get_oem_prod_version(). Note that this
  function is not used.
- Set PHY correctly in ixgbe_setup_mac_link_sfp_x550a() if a device is a
  C3000 KR SFP+.

Fix race about writing adapter->link_active for ixg(4).
adapter->link_active is updated by ixgbe_update_link_status() only.

The function is called from the following four functions.
    - ixgbe_media_status()
    - ixgbe_local_timer1()
    - ixgbe_stop()
    - ixgbe_handle_link()

The functions other than ixgbe_handle_link() call ixgbe_update_link_status()
with holding IXGBE_CORE_LOCK, however ixgbe_handle_link() calls it without
holding IXGBE_CORE_LOCK. That can cause race. So, add IXGBE_CORE_LOCK to
ixgbe_handle_link().

Tested by msaitoh@n.o and me.

Fix race about writing adapter->link_active for ixv(4).
adapter->link_active is updated by ixv_update_link_status() only.

The function is called from the following two functions.
    - ixv_media_status()
    - ixv_handle_link()

ixv_media_status() calls ixv_update_link_status() with holding
IXGBE_CORE_LOCK, however ixv_handle_link() calls it without
holding IXGBE_CORE_LOCK, the same as ixg(4).

ok by msaitoh@n.o.

- Add link related softint's counter.
- Fix indent.

(martin)

Pull up following revision(s) (requested by maxv in ticket #672):
sys/netinet6/in6_proto.c: revision 1.120
Change ip6_hdrnestlimit to be 15 instead of 50. I couldn't find any
reference in RFCs about what a correct limit should be, but FreeBSD already
uses 15.
If an IPv6 packet has 50 options, there is clearly something wrong with it.

(martin)

Pull up following revision(s) (requested by maxv in ticket #671):

sys/netinet6/ip6_forward.c: revision 1.91

Fix two pretty bad mistakes. If ipsec6_check_policy fails m is not freed,
and a 'goto out' is missing after ipsec6_process_packet.

(martin)

Pull up following revision(s) (requested by maxv in ticket #670):

sys/netipsec/xform_esp.c: revision 1.73

Add missing NULL check. Normally that's not triggerable remotely, since we
are guaranteed that 8 bytes are valid at mbuf+skip.

(martin)

Pull up following revision(s) (requested by maxv in ticket #669):

sys/netipsec/ipsec.c: revision 1.134

Fix ipsec4_get_ulp(). We should do "goto done" instead of "return",
otherwise the port fields of spidx are uninitialized.

ok mlelstv@

(martin)

Pull up following revision(s) (requested by maxv in ticket #667):

sys/netipsec/ipsec_input.c: revision 1.54

ipsec4_fixup_checksum calls m_pullup, so don't forget to do mtod() again,
to prevent use-after-free.

In fact, the m_pullup call is never reached: it is impossible for 'skip'
to be zero in this function, so add an XXX for now.

(martin)

Pull up following revision(s) (requested by maxv in ticket #666):

sys/netinet6/raw_ip6.c: revision 1.161

Fix use-after-free, the first m_copyback_cow may have freed the mbuf, so
it is wrong to read ip6->ip6_nxt.

(martin)

Update list of modules which are included in the monolithic compat_netbsd
module.

(pgoyette)

Pull up following revision(s) (requested by maxv in ticket #664):

sys/netinet6/dest6.c: revision 1.21

Fix the calculation of the ICMP6 error pointer. It is not correct to use

        pointer = opt - mtod(m, u_int8_t *)

because m may have gone through m_pulldown, and it is possible that
m->m_data is no longer the beginning of the packet.

(martin)

Pull up following revision(s) (requested by maxv in ticket #663):

sys/netinet6/frag6.c: revision 1.64

Kick zero-sized fragments. We can't allow them to enter; two fragments
could be put at the same offset.

(martin)

Add compat_13 module

(pgoyette)

Pull up following revision(s) (requested by maxv in ticket #662):

sys/netinet/tcp_input.c: revision 1.383 (via patch)

Revert rev1.183 (2003).

It was intended as an optimization, but it increases the attack surface:

the IPsec policy is not enforced on RST packets when the socket is in the
LISTEN state, and an (unauthenticated) attacker could jam the connection
between two IPsec hosts by sending RST packets between the client's SYN
and ACK packets.

Discussed with ozaki-r@.

(martin)

Forgot to commit this previously.

(pgoyette)

Pull up following revision(s) (requested by maxv in ticket #661):

sys/netinet/ip_encap.c: revision 1.67

Fix memory leak, found by Mootja.

(martin)

Pull up following revision(s) (requested by maxv in ticket #668):
sys/netinet/ip_reass.c: revision 1.12

Add one more check in ip_reass_packet(): make sure that the end of each
fragment does not exceed IP_MAXPACKET.

In ip_reass(), we only check the final length of the reassembled packet
against IP_MAXPACKET.

But there is an integer overflow that can happen a little earlier. We
are doing:

        i = ntohs(p->ipqe_ip->ip_off) + ntohs(p->ipqe_ip->ip_len) -
            ntohs(ip->ip_off);
        [...]
        ip->ip_off = htons(ntohs(ip->ip_off) + i);

It is possible that

        ntohs(p->ipqe_ip->ip_off) + ntohs(p->ipqe_ip->ip_len) > 65535

so the computation of ip_off wraps to zero. This breaks an assumption in
the reassembler - it expects the list of fragments to be ordered by
offset, and here it's not ordered anymore. (Un)Fortunately I couldn't
turn this into anything exploitable.

With the new check, it is guaranteed that ip_off+ip_len<=65535.

(martin)

Pull up following revision(s) (requested by nonaka in ticket #660):
usr.sbin/makefs/cd9660/cd9660_eltorito.c: revision 1.22
usr.sbin/makefs/cd9660/cd9660_eltorito.c: revision 1.23

Initialize boot_catalog_entry's entry_type properly.

This had been missing but the type was used in cd9660_setup_boot().
From OpenBSD usr.sbin/makefs/cd9660/cd9660_eltorito.c r1.10.

Correctly mark the last El Torito section header.
Pointed out by Benno Rice via DM.

(martin)

Extract compat_14 stuff into its own module

(pgoyette)

Fix warning when compiling Xen; FLAT_RING3_CS64 is defined in a child of
xen.h, which is already included in genassym.cf. So don't redefine it.

(maxv)

Add #ifndef XEN, xen doesn't have speculation_barrier.

(maxv)

Add Some bit definitions of AMD Fn80000001 %edx:
- MMX
- FXSR

(msaitoh)

From the latest Intel SDM:
- Add Intel Fn0000_0006 %eax new bit 14-20 (HWP stuff).
- Intel Fn0000_0007 %ecx bit 22 is for both RDPID and IA32_TSC_AUX.

(msaitoh)

From the latest Intel SDM:
- Add Xeon Phi 7215, 7285 and 7295
- Add Coffee Lake

(msaitoh)

Remove dead code. It was introduced in rev1 (25 years ago), and is
irrelevant today.

(maxv)

Style, use NULL for pointers, use KASSERT, and don't inline huge functions,
we want to debug them with DDB (and not just with GPROF).

(maxv)

Allow allocation of PTE space to fail in pmap_enter if PMAP_CANFAIL is set.
This fixes the panic part of PR port-vax/28379.

(ragge)

Fix the log. mtod never returns NULL, so 'ip' is always non-NULL, and the
'ip6' branch is never taken. As a result we log garbage on IPv6 packets.

Use ip_v instead.

(maxv)

Increase UVM read ahead window limit a bit to match concurrency of reading
from the raw device.

(mlelstv)

Use consttime_memequal instead of memcmp, to prevent side channels. This
functions returns 1 when the buffers are equal, contrary to memcmp, hence
the !.

(maxv)

-  Add missing IFM_NONE support. If a interface support linkdown,
  "ifconfig ixgN media none" drpos link.  Not all interface can do link down.

Tested:
82598 AT2 (T)
92599 SF+(SFI) (X520-DA2)
X540
X550-T1
X550EM_x (X10SDV-8C-TLN4F)
X550EM_a (A2SDi-H-TP4F port 0, 1 (T))

Doesn't work:
X550EM_a (A2SDi-H-TP4F port 2, 3 (SFP+ (KR)))
X550EM_a (MA10-ST0 port 2, 3 (SFP+ (SFI)))
(Denverton SFP+ can't force link down because SFP+'s TX_DISABLE pin is
pull down. Is there a way to shutdown SFP+ cage's power?)
Not tested:
82598 fiber.

- Change some functions static.

(msaitoh)

typo in version dependency

(pgoyette)

Resolve conflicts between branch and HEAD

(pgoyette)

Forgot a file!

(pgoyette)

Create and build a compat_16 module

(pgoyette)

Don't write EIMC directly. It is required to manage with struct ix_queue status.

XXX pullup-8

(knakahara)

Fix the problem between eitr and link_speed.

In ixgbe_msix_que(), que->eitr_setting is limited to IXGBE_MIN_RSC_EITR_10G1G
when link_speed is 1Gbps or 10Gbps. However, que->eitr_setting is set to EITR
register in the *next* Tx/Rx interrupt. If link_speed changes from 100Mbps to
1Gbps ro 10Gbps, que->eitr_setting which is not limited can be set to EITR
register, that is, the problem fixed by ixgbe.c:r1.124 can occur in this case.

To fix this case, que->eitr_setting should be clear when link_speed is changed
or link state is changed.

Furthermore, expand the variants used for AIM (txr->bytes, txr->packets,
rxr->bytes and rxr->packets) from u32 to u64 to avoid wraparound which causes
que->eitr_setting calculation mistake.

XXX pullup-8

(knakahara)

Add compat_20 to the list of aliases

(pgoyette)

Create and build the compat_20 module

(pgoyette)

Add compat_20 module to sets lists

(pgoyette)

Rename usb_30_subr.c --> usb_subr_30.c for consistency with others

(pgoyette)

tcp_urp_drop: fix a bug introduced in 1.390 rev (hi maxv@).

(rmind)

audio works now, so enable it

(macallan)

Remove TCPREASS_DEBUG. It was introduced 20 years ago when the reassembler
was being developed, but it's irrelevant today. Makes the code clearer.

(maxv)

explicitly enable the i2s part in the keylargo FCR
now this works on my PCI-X G5

(macallan)

Reorder/Fix comments to clarify.

(maxv)

Remove two more 'else' branches.

(maxv)

Fix memory leak, we may reallocate 'tcp_saveti' after 'findpcb'. It's not
a tragic bug, because it happens only on sockets with debug enabled.

(maxv)

Remove 'else', makes it clearer that we leave.

(maxv)

Clarify with KASSERT.

(maxv)

Simplify the computation:

  m->m_pkthdr.len -  sizeof(struct tcphdr) - optlen - hlen
= m->m_pkthdr.len - (sizeof(struct tcphdr) + optlen + hlen)
= m->m_pkthdr.len - [tcp_len]
= toff

(maxv)

fix build for 32bit non-bridge SMP kernels

(macallan)

Use same major for rx floppies for both b and c devices.
Works around a bug in the mscp disk driver, and solves PR port-vax/26138.

(ragge)

Bump image size again for new bloat.

(joerg)

Move the complex logic for dynamically writing branches from ld.elf_so
into a header for reuse in crt0.o for static ifunc support. Change the
existing logic for sparc64 to use the Bicc variant of ba,a as it allows
+-8MB displacement compared to the BPcc variant's +-1MB. Teach the sparc
variant the same trick for using ba,a and not sethi+jmp when possible.

(joerg)

We no longer need to force compat_30 - it will be automatically
invoked if the appropriate compat module is loaded.

(pgoyette)

Split out the usb compat_30 code and add it to the module

(pgoyette)

Revert previous.  Instead, we'll just move usbd_printBCD() to usbdi.h

(pgoyette)

Protect against multiple-inclusion

(pgoyette)

Misc changes; no real functional change.

(maxv)