Some routing socket defines (`RTM_` and `RTA_` ones) clash with the ones
used by the the Netlink.
As some rtsock definitions like interface flags or route flags are used in
both netlink and rtsock, provide a convenient way to include those without
running into the define collision.
Differential Revision: https://reviews.freebsd.org/D38982
MFC after: 2 weeks
Summary:
When iflib was first converted some IfAPI APIs were not yet present, so
were tagged with "XXX" comments. Finish the conversion by using these
new APIs.
Reviewed by: gallatin, erj
Sponsored by: Juniper Networks, Inc
Differential Revision: https://reviews.freebsd.org/D38928
epairs currently shuttle all transmitted packets through a single global
taskqueue thread. To hand packets over to the taskqueue thread, each
epair maintains a pair of ring buffers and a lockless scheme for
notifying the thread of pending work. The implementation can lead to
lost wakeups, causing to-be-transmitted packets to end up stuck in the
queue.
Rather than extending the existing scheme, simply replace it with a
linked list protected by a mutex, and use the mutex to synchronize
wakeups of the taskqueue thread. This appears to give equivalent or
better throughput with >= 16 producer threads and eliminates the lost
wakeups.
Reviewed by: kp
MFC after: 1 week
Sponsored by: Klara, Inc.
Sponsored by: Modirum MDPay
Differential Revision: https://reviews.freebsd.org/D38843
The m_flags field of struct mbuf is 24 bits wide and so gets truncated
in a couple of places in the epair code. Instead of preserving the
entire flag set, just remember whether M_BCAST or M_MCAST is set.
MFC after: 1 week
Sponsored by: Klara, Inc.
Do not pass the pointer to our valid mbuf to pfil(9). Pass an
uninitialized one only. This was unsafe with the old KPI, too,
but for some reason didn't fail.
Fixes: caf32b260a
* Make nhop_set_blackhole() set all necessary properties for the
nexthop
* Make nexthops blackhole/reject based on the rtm_type netlink
property instead of using rtflags.
Reported by: Marek Zarychta <zarychtam@plan-b.pwste.edu.pl>
MFC after: 3 days
Don't define ovpn_find_peer_by_ip() if INET is not set, and do the same
for ovpn_find_peer_by_ip6() and INET6.
Reported by: mjg
Sponsored by: Rubicon Communications, LLC ("Netgate")
These two functions are intended to be used only when allocating or
destroying vnet instances.
No functional change intended.
Reviewed by: kp
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D37955
add_route_flags() uses `rt` prefix data to lookup the the current
rtentry from the routing table. Update rib_add_route_px() to
always pass rtentry regardless of the op_flags.
Reported by: Stefan Grundmann <sg2342@googlemail.com>
MFC after: 1 day
The 0b70e3e78b changed the original design of a single entry point
into pfil(9) chains providing separate functions for the filtering
points that always provide mbufs and know the direction of a flow.
The motivation was to reduce branching. The logical continuation
would be to do the same for the filtering points that always provide
a memory pointer and retire the single entry point.
o Hooks now provide two functions: one for mbufs and optional for
memory pointers.
o pfil_hook_args() has a new member and pfil_add_hook() has a
requirement to zero out uninitialized data. Bump PFIL_VERSION.
o As it was before, a hook function for a memory pointer may realloc
into an mbuf. Such mbuf would be returned via a pointer that must
be provided in argument.
o The only hook that supports memory pointers is ipfw:default-link.
It is rewritten to provide two functions.
o All remaining uses of pfil_run_hooks() are converted to
pfil_mem_in().
o Transparent union of pfil_packet_t and tricks to fix pointer
alignment are retired. Internal pfil_realloc() reduces down to
m_devget() and thus is retired, too.
Reviewed by: mjg, ocochard
Differential revision: https://reviews.freebsd.org/D37977
Summary:
Clean up style issues from IfAPI additions.
Casts to (struct ifnet *) made sense when `if_t` was a `void *`, but
since it's a `struct ifnet *` it no longer makes sense. Fix whitespace
errors, among others.
Reviewed by: kib, glebius
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D38499
Summary:
As a stopgap measure add basic accessors for the if_capabilities2 and
if_capenable2 members to further hide the ifnet details.
Sponsored by: Juniper Networks, Inc.
Reviewed by: glebius, kib
Differential Revision: https://reviews.freebsd.org/D38487
Summary:
Add the following accessors needed by infiniband drivers:
* if_getaddrlen()
* if_setbroadcastaddr()
* if_resolvemulti()
With these accessors, and additional changes on the drivers' side, an
amd64 kernel can be compiled with `struct ifnet` completely hidden.
Reviewed by: melifaro
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D38488
Summary:
Add 2 new APIs for supporting recent mbuf changes:
* 36e0a362ac added the m_snd_tag_alloc() wrapper around
if_snd_tag_alloc(). Push this down to the ifnet level.
* 4d7a1361ef adds the m_rcvif_serialize()/m_rcvif_restore() KPIs to
serialize and restore an ifnet pointer. Add the necessary wrapper to
get the index generation for this.
Reviewed By: jhb
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D38340
Summary:
Sometimes it's useful to iterate over all interfaces in the current
VNET, as the linuxulator does in several places.
Unlike other iterators in the IfAPI this propagates any error received
up to the caller, instead of returning a count.
Sponsored by: Juniper Networks, Inc.
Reviewed by: glebius, melifaro
Differential Revision: https://reviews.freebsd.org/D38348
Summary:
The only user of the ALTQ_IS_ENABLED() in a driver checks against the
ifnet queue. Abstract that all out and present the interface to check
if ALTQ is enabled on the interface.
Sponsored by: Juniper Networks, Inc.
Reviewed By: glebius
Differential Revision: https://reviews.freebsd.org/D38204
Summary:
Firewire is the only device driver that accesses the l2com member, all
other accesses are handled within the netstack itself.
Sponsored by: Juniper Networks, Inc.
Reviewed by: glebius, melifaro
Differential Revision: https://reviews.freebsd.org/D38203
Summary:
* if_setreassignfn for wireguard.
* if_getinputfn() and if_getstartfn() for various drivers. Use the
function descriptor typedefs for these and the setters.
* vlantrunk accessor. This is used by VLAN_CAPABILITIES() used by
several drivers, as well as directly by mxge(4).
* if_pcp member accessor, used by cxgbe.
* accessors for netmap adapter.
Sponsored by: Juniper Networks, Inc.
Reviewed By: glebius
Differential Revision: https://reviews.freebsd.org/D38202
Summary:
Keep TOEDEV() macro for backwards compatibility, and add a SETTOEDEV()
macro to complement with the new accessors.
Sponsored by: Juniper Networks, Inc.
Reviewed by: glebius
Differential Revision: https://reviews.freebsd.org/D38199
Summary:
In preparation of making if_t completely opaque outside of the netstack,
explicitly include the header. <net/if_var.h> will stop including the
header in the future.
Sponsored by: Juniper Networks, Inc.
Reviewed by: glebius, melifaro
Differential Revision: https://reviews.freebsd.org/D38200
Summary:
Port the MAC modules to use the IfAPI APIs as part of this.
Sponsored by: Juniper Networks, Inc.
Reviewed by: glebius
Differential Revision: https://reviews.freebsd.org/D38197
Summary:
Hide more netstack by making the BPF_TAP macros real functions in the
netstack. "struct ifnet" is used in the header instead of "if_t" to
keep header pollution down.
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D38103
When ALTQ is enabled ifnet accessors already need to be called, largely
defeating the purpose of the inline. To that extent, make the ALTQ form
functions in the netstack proper, and make them always available.
Reviewed By: glebius
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D38104
Ever since gtaskqueue_drain() was added to iflib_stop(), a kernel panic
occurs when the ice(4) driver is in recovery mode. Queues are not
initialized in this mode, so gt_taskqueue is not initialized, and
gtaskqueue_drain() will panic.
Fix this by only doing a drain if an RX queue's gt_taskqueue is
initialized.
Signed-off-by: Eric Joyner <erj@FreeBSD.org>
Reviewed by: erj@
MFC after: 1 week
Sponsored by: Intel Corporation
Differential Revision: https://reviews.freebsd.org/D37892
As part of the effort to hide the internals of the ifnet struct, convert
the DEBUGNET_SET() macro to use an accessor instead of directly touching
the methods member.
Reviewed by: glebius (older version)
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D38105
Hide the ifnet structure definition, no user serviceable parts inside,
it's a netstack implementation detail. Include it temporarily in
<net/if_var.h> until all drivers are updated to use the accessors
exclusively.
Reviewed by: glebius
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D38046
Some drivers, like iflib drivers, take a 'context' argument instead of a
ifnet argument, as a single interface may have multiple contexts.
Follow this scheme by passing the context argument down. Most drivers
will likely pass 'ifp' as the context.
Reviewed by: glebius
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D38102
Currently `close(2)` erroneously return `EOPNOTSUPP` for `PF_ROUTE` sockets.
It happened after making rtsock socket implementation self-contained (
36b10ac2cd ). Rtsock code marks socket as connected in `rts_attach()`.
`soclose()` tries to disconnect such socket using `.pr_disconnect` callback.
Rtsock does not implement this callback, resulting in the default method being
substituted. This default method returns `ENOTSUPP`, failing `soclose()` logic.
This diff restores the previous behaviour by adding custom `pr_disconnect()`
returning `ENOTCONN`.
Reviewed by: glebius
Differential Revision: https://reviews.freebsd.org/D38059
Basic scenario: we have a closed connection (In TCPS_FIN_WAIT_2), and
get a new connection (i.e. SYN) re-using the tuple.
Without syncookies we look at the SYN, and completely unlink the old,
closed state on the SYN.
With syncookies we send a generated SYN|ACK back, and drop the SYN,
never looking at the state table.
So when the ACK (i.e. the third step in the three way handshake for
connection setup) turns up, we’ve not actually removed the old state, so
we find it, and don’t do the syncookie dance, or allow the new
connection to get set up.
Explicitly check for this in pf_test_state_tcp(). If we find a state in
TCPS_FIN_WAIT_2 and the syncookie is valid we delete the existing state
so we can set up the new state.
Note that when we verify the syncookie in pf_test_state_tcp() we don't
decrement the number of half-open connections to avoid an incorrect
double decrement.
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D37919
<net/if.h> should be a fully user-facing header, so these APIs don't
belong there. Revert and will find another approach.
This reverts commit fe33e0ab83.
Fixes: fe33e0ab83
Sponsored by: Juniper Networks, Inc.
Summary:
The "public" KPI for ifnet belongs in net/if.h, with net/if_var.h being
implementation details for the netstack. This is the next step in
enforcing that separation.
Reviewed by: melifaro
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D38030
Summary:
A common pattern has been to:
if (foo)
caps = IFCAP_FOO;
ifp->if_capenable &= ~IFCAP_FOO;
ifp->if_capenable |= caps;
which in the new order of things would be:
if (foo)
caps = IF_FOO;
if_setcapenablebits(ifp, 0, IFCAP_FOO);
if_setcapenablebits(ifp, caps, 0);
This change streamlines this into:
if (foo)
caps = IF_FOO;
if_setcapenablebits(ifp, caps, IFCAP_FOO);
Reviewed by: melifaro
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D37993
<net/if_var.h> should be a kernel-only header, but it's included
elsewhere. Until that's addressed expose if_t to userspace to fix the
build.
Fixes: be4315dcbb
Sponsored by: Juniper Networks, Inc.
Summary:
<net/if_var.h> should really be used by the netstack only, not by
drivers. Eventually all the accessors will be moved to <net/if.h> as
well, but for now just move the typedef while the KPI gets sorted and
drivers get converted.
Sponsored by: Juniper Networks, Inc.
Reviewed By: melifaro, glebius
Differential Revision: https://reviews.freebsd.org/D37784