rc.d/sendmail: remove a obsolete upgrade seatbelt
This check was in place to aid the transition from sendmail pre-8.10.
8.10 was released in 2000. It's not possible to upgrade directly from
such as system (Freebsd 3?) to FreeBSD 15 so we can drop this.
Reviewed by: gshapiro, jhb
Differential Revision: https://reviews.freebsd.org/D49308
arch.7: Update version reference to 13.0
This document is intended to be a reference for supported FreeBSD
versions, so update text to refer to 13.0 and later.
Sponsored by: The FreeBSD Foundation
(cherry picked from commit 46b91601bb165f20a40529775fffb4d05fe73d15)
Makefile.inc1: Conditionalize some package related variables
In particular, don't invoke git to compute SOURCE_DATE_EPOCH for
unrelated targets like check-old or delete-old. If the git invocation
fails (e.g. when using a git worktree mounted over NFS) it can
generate a lot of irrelevant warning spam.
Reviewed by: emaste
Fixes: 8a3537aaf7c1 ("Makefile.inc1: Make package timestamps reproducible by default")
Differential Revision: https://reviews.freebsd.org/D49278
(cherry picked from commit db6f2bb93a9706963f66d270edb5ee62c37a9296)
Makefile.inc1: Make package timestamps reproducible by default
Set package archive timestamps based on most recent source commit
timestamp (approach suggested by bapt).
I'd like to include git metadata in a file included in src tarballs, so
that the build is reproducible (including the hash shown in uname etc.)
outside of a git checkout. There are still details to be sorted out to
do that, so this is an interim step to improve reproducibility.
Reviewed by: bapt
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D49165
(cherry picked from commit 8a3537aaf7c19f7331fcc160ab42e36fc79e408a)
lockmgr/rmlock/rwlock/sx: Make various assertions more robust
Print pointers to locks instead of their names to avoid a nested panic
if the lock object is corrupted.
Reviewed by: markj
Sponsored by: AFRL, DARPA
Differential Revision: https://reviews.freebsd.org/D49331
inpcb: make sure we don't pass uninitialized faddr to in_pcbladdr()
This very theoretical edge case was discovered by Coverity, not sure if
it was introduced by 2af953b132ee or was there before.
CID: 1593695
Fixes: 2af953b132ee8d2eb4d8d7bb15fc38bf04dde348
procstat.1: correct description of the kstack subcommand after removal of swapping
Reviewed by: markj
Sponsored by: The FreeBSD Foundation
Differential revision: https://reviews.freebsd.org/D49163
kern/sys_eventfd.c: fix includes
Remove redundant includes like sys/types.h and sys/param.h due to
sys/systm.h. Sort alphabetically.
Reviewed by: markj
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D49163
libprocstat: add helper to query knotes for specific kqueue
Reviewed by: markj
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D49163
mlx5en: bump MLX5E_MAX_BUSDMA_RX_SEGS
This is needed to accomodate more data segments in wqes for 64K receive
mbuf chains.
Reviewed by: Ariel Ehrenberg <aehrenberg at nvidia.com>, Slava Shwartsman <slavash at nvidia.com>
Sponsored by: NVidia networking
MFC after: 1 week
mlx5en: fix rq->wqe_sz usage
Define it as the size of the single data segment in wqe.
Reviewed by: Ariel Ehrenberg <aehrenberg at nvidia.com>, Slava Shwartsman <slavash at nvidia.com>
Sponsored by: NVidia networking
MFC after: 1 week
mlx5: bump the max LRO packet size
The belief is that the 7*MCLBYTES limit was set to not hit the segment
limit for wqe busdma tag. But with the current mbuf allocator it is not
possible, and even if it was, the corresponding wqe fill would simply
fail.
Reviewed by: Ariel Ehrenberg <aehrenberg at nvidia.com>, Slava Shwartsman <slavash at nvidia.com>
Sponsored by: NVidia networking
MFC after: 1 week
mlx5en: stop arbitrary limiting max wqe size
Since the times the driver accepts s/g receive buffers, there is no
sense in trying to use pre-existing mbuf clusters sizes. The only
possible optimization is to use full page size if wqe size is greater
than MCLBYTES.
Reviewed by: Ariel Ehrenberg <aehrenberg at nvidia.com>, Slava Shwartsman <slavash at nvidia.com>
Sponsored by: NVidia networking
MFC after: 1 week
mlx5: overwrite only the echo reply timestamp from the last packet in LRO
Reviewed by: Ariel Ehrenberg <aehrenberg at nvidia.com>, Slava Shwartsman <slavash at nvidia.com>
Sponsored by: NVidia networking
MFC after: 1 week
mlx5_en: correct recalculation of the ipv4 checksum for hw lro packet
The call to in_cksum_skip() did not skipped the ethernet header.
Reviewed by: Ariel Ehrenberg <aehrenberg at nvidia.com>, Slava Shwartsman <slavash at nvidia.com>
Sponsored by: NVidia networking
MFC after: 1 week
mlx5en: follow PRM for setting the max hw lro segment size
If the NIC is capable, just pass the full packet size, including L2/L3
headers, as the segment size. Otherwise, decrement the number of
strides by 1 to left the space for L2/IP headers, as it was done before.
But do the arithmetic on the segment number instead of the full packet
size.
Reviewed by: Ariel Ehrenberg <aehrenberg at nvidia.com>, Slava Shwartsman <slavash at nvidia.com>
Sponsored by: NVidia networking
MFC after: 1 week
