man/man3type/sockaddr{,_un}.3type: Split struct sockaddr_un from sockaddr(3type)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3type/sockaddr{,_in,_in6}.3type: Split struct sockaddr_in6 from sockaddr_in(3type)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3type/sockaddr_in.3type: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3type/sockaddr{,_in,_in6}.3type: Split IP types from sockaddr(3type)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_*,SO_*}.2const: Make sashimi of IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IPPROTO_IP.2const: Tweak after making sashimi of this page

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/SO_PEERSEC.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,SO_PEERSEC}.2const: Split SO_PEERSEC from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_UNBLOCK_SOURCE.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_UNBLOCK_SOURCE}.2const: Split IP_UNBLOCK_SOURCE from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_TTL.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_TTL}.2const: Split IP_TTL from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_TRANSPARENT.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_TRANSPARENT}.2const: Split IP_TRANSPARENT from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_TOS.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_TOS}.2const: Split IP_TOS from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_ROUTER_ALERT.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_ROUTER_ALERT}.2const: Split IP_ROUTER_ALERT from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_RETOPTS.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_RETOPTS}.2const: Split IP_RETOPTS from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_RECVTTL.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_RECVTTL}.2const: Split IP_RECVTTL from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_RECVTOS.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_RECVTOS}.2const: Split IP_RECVTOS from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_RECVORIGDSTADDR.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_RECVORIGDSTADDR}.2const: Split IP_RECVORIGDSTADDR from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_RECVOPTS.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_RECVOPTS}.2const: Split IP_RECVOPTS from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_RECVERR.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_RECVERR}.2const: Split IP_RECVERR from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_PKTINFO.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_PKTINFO}.2const: Split IP_PKTINFO from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_PASSSEC.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_PASSSEC}.2const: Split IP_PASSSEC from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_OPTIONS.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_OPTIONS}.2const: Split IP_OPTIONS from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_NODEFRAG.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_NODEFRAG}.2const: Split IP_NODEFRAG from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_MULTICAST_TTL.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_MULTICAST_TTL}.2const: Split IP_MULTICAST_TTL from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_MULTICAST_LOOP.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_MULTICAST_LOOP}.2const: Split IP_MULTICAST_LOOP from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_MULTICAST_IF.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_MULTICAST_IF}.2const: Split IP_MULTICAST_IF from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_MULTICAST_ALL.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_MULTICAST_ALL}.2const: Split IP_MULTICAST_ALL from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_MTU_DISCOVER.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_MTU_DISCOVER}.2const: Split IP_MTU_DISCOVER from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_MTU.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_MTU}.2const: Split IP_MTU from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_MSFILTER.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_MSFILTER}.2const: Split IP_MSFILTER from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_LOCAL_PORT_RANGE.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_LOCAL_PORT_RANGE}.2const: Split IP_LOCAL_PORT_RANGE from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_HDRINCL.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_HDRINCL}.2const: Split IP_HDRINCL from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_FREEBIND.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_FREEBIND}.2const: Split IP_FREEBIND from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_DROP_SOURCE_MEMBERSHIP.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_DROP_SOURCE_MEMBERSHIP}.2const: Split IP_DROP_SOURCE_MEMBERSHIP from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_DROP_MEMBERSHIP.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_DROP_MEMBERSHIP}.2const: Split IP_DROP_MEMBERSHIP from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_BLOCK_SOURCE.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_BLOCK_SOURCE}.2const: Split IP_BLOCK_SOURCE from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_BIND_ADDRESS_NO_PORT.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_BIND_ADDRESS_NO_PORT}.2const: Split IP_BIND_ADDRESS_NO_PORT from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_ADD_SOURCE_MEMBERSHIP}.2const, man/man2type/ip_mreq_source.2type: Split struct ip_mreq_source from IP_ADD_SOURCE_MEMBERSHIP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_ADD_SOURCE_MEMBERSHIP.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_ADD_SOURCE_MEMBERSHIP}.2const: Split IP_ADD_SOURCE_MEMBERSHIP from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2type/ip_mreq.2type: Add link page

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_ADD_MEMBERSHIP}.2const, man/man2type/ip_mreqn.2type: Split struct ip_mreqn from IP_ADD_MEMBERSHIP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IP_ADD_MEMBERSHIP.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/{IPPROTO_IP,IP_ADD_MEMBERSHIP}.2const: Split IP_ADD_MEMBERSHIP from IPPROTO_IP(2const)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man5/proc_sys_net.5: Add reference to proc_sys_net_ipv4(5)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man5/proc_sys_net_ipv4.5: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/ip.7, man/man2const/IPPROTO_IP.2const: Split IPPROTO_IP from ip(7)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/ip.7, man/man5/proc_sys_net_ipv4.5: Split /proc/sys/net/ipv4/ from ip(7)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/: {str,wcs}n{cmp,len}(3): Remove array parameter length expressions

The arguments to these APIs may be shorter than that, as long as they
are null terminated strings.

Reported-by: Florian Weimer <fweimer@redhat.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/mremap.2: Fix incorrect reference to VM_MIXEDMAP

We actually disallow VM_DONTEXPAND and VM_PFNMAP for MREMAP_DONTUNMAP,
the manpage incorrectly references VM_MIXEDMAP.

Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Message-ID: <20251121081609.52462-1-lorenzo.stoakes@oracle.com>
Cc: <linux-mm@kvack.org>
Cc: <linux-kernel@vger.kernel.org>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Jann Horn <jannh@google.com>
Cc: Pedro Falcato <pfalcato@suse.de>
Cc: Rik van Riel <riel@surriel.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/ip.7: srcfix

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/ip.7: Clarify IP_PKTINFO's semantics depending on packet direction

For recvmsg(2), ipi_spec_dst is set by ipv4_pktinfo_prepare() to the
result of fib_compute_sec_dst().  The latter was introduced in
linux.git 35ebf65e851c6d97 ("ipv4: Create and use fib_compute_spec_dst() helper.").

Quoting its commit message:

> The specific destination is the host we direct unicast replies to.
> Usually this is the original packet source address, but if we are
> responding to a multicast or broadcast packet we have to use something
> different.
>
> Specifically we must use the source address we would use if we were to
> send a packet to the unicast source of the original packet.

Experimentation seems to confirm that behavior.

As for the note about ipi_spec_dst being on a different interface:
- For unicast packets (for which ipi_spec_dst is the original
  destination address), I believe this is trivially true because Linux
  uses the weak host model (unless there's some interaction with
  RTCF_LOCAL that I'm missing).
- For multicast/broadcast packets, fib_compute_sec_dst() only passes the
  original interface to the lookup in the context of L3M.  In
  particular, the original implementation (cited above) set iif and oof
  to 0. Also, citing
linux.git e7372197e15856ec ("net/ipv4: Set oif in fib_compute_spec_dst"),
  > If the device is not enslaved, oif is still 0 so no affect.

It doesn't seem like using an address specifically from the interface
the packet was received on was ever the intention.  I've also confirmed
this behavior (sending a multicast packet from another machine, whose IP
I've routed to a dummy interface).

I'm focusing on this because that's a misconception I've had before
digging into the code - the sendmsg behavior explained in the same
paragraph made me think ipi_spec_dst was the (primary?) address of
ipi_ifindex.  I think this is worth clarifying.

I've made it explicit that ipi_addr isn't used by sendmsg because that's
another possible misconception.

The (first) extra comma in sendmsg's ipi_spec_dst's description is meant
to emphasize that it's used as the local source address _and_ for the
routing table lookup, as opposed to just affecting the routing table
lookup.
Stylistically it might be a bit weird but idk how to convey this better.

Apart from the cited commits I was referencing the linux-6.17.7 tarball.

__fib_validate_source (and the comment near it) might also be of
interest to people trying to figure out what "specific destinations"
are, exactly.

Signed-off-by: Jakub Głogowski <not@dzwdz.net>
Message-ID: <fb3980b64d1c827ad59726bb30761d735396e109.1763130571.git.not@dzwdz.net>
Cc: <linux-kernel@vger.kernel.org>
Cc: <linux-api@vger.kernel.org>
Cc: Jan Engelhardt <ej@inai.de>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3attr/gnu::warning.3attr: Add page

Signed-off-by: Alejandro Colomar <alx@kernel.org>

CONTRIBUTING.d/ai: Add guidelines banning AI for contributing

This policy is based on the Gentoo policy (see link below).
However, I've modified our text to be more restrictive.

Link: <https://wiki.gentoo.org/wiki/Project:Council/AI_policy>
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Cc: Collin Funk <collin.funk1@gmail.com>
Cc: Sam James <sam@gentoo.org>
Cc: "G. Branden Robinson" <branden@debian.org>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/flock.2: Mention non-atomicity w.r.t close

Ideally one should be able to use flock to synchronize with another
process (or thread) closing that file, for instance before attempting
to execve it (as execve of a file open for writing fails with ETXTBSY).

Unfortunately, on Linux it is not reliable, because in the process of
closing a file its locks are dropped before the refcounts of the file
(as well as its underlying filesystem) are decremented, creating a race
window where execve of the just-unlocked file sees it as if still open.

Linux developers have indicated that it is not easy to fix, and the
appropriate course of action for now is to document this limitation.

Link: <https://lore.kernel.org/linux-fsdevel/68c99812-e933-ce93-17c0-3fe3ab01afb8@ispras.ru/>

Signed-off-by: Alexander Monakov <amonakov@ispras.ru>
Message-ID: <181d561860e52955b29fe388ad089bde4f67241a.1760627023.git.amonakov@ispras.ru>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/copy_file_range.2: glibc no longer provides fallback after 2.30

Closes: <https://bugzilla.kernel.org/show_bug.cgi?id=220489>
Signed-off-by: Wes Gibbs <wg21908@gmail.com>
Message-ID: <20251102000330.155591-1-wg21908@gmail.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/readdir.3, man/man3type/stat.3type: Improve documentation about .d_ino and .st_ino

Suggested-by: Pali Rohár <pali@kernel.org>
Co-authored-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Pali Rohár <pali@kernel.org>
Co-authored-by: Jan Kara <jack@suse.cz>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: "Darrick J. Wong" <djwong@kernel.org>
Cc: "G. Branden Robinson" <branden@debian.org>
Cc: <linux-fsdevel@vger.kernel.org>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: Use '.member' notation for referring to struct/union members

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/madvise.2, man/man2const/PR_[GS]ET_THP_DISABLE.2const: document addition of PR_THP_DISABLE_EXCEPT_ADVISED

PR_THP_DISABLE_EXCEPT_ADVISED extended PR_SET_THP_DISABLE to only provide
THPs when advised.  IOW, it allows individual processes to opt-out of
THP = "always" into THP = "madvise", without affecting other workloads
on the system.  The series has been merged in [1].  Before [1], the
following 2 calls were allowed with PR_SET_THP_DISABLE:

prctl(PR_SET_THP_DISABLE, 0, 0, 0, 0);  // reset THP setting.
prctl(PR_SET_THP_DISABLE, 1, 0, 0, 0);  // disable THPs completely.

Now in addition to the 2 calls above, you can do:

// disable THPs except madvise.
prctl(PR_SET_THP_DISABLE, 1, PR_THP_DISABLE_EXCEPT_ADVISED, 0, 0);

This patch documents the changes introduced due to the addition of
PR_THP_DISABLE_EXCEPT_ADVISED flag:
-  PR_GET_THP_DISABLE returns a value whose bits indicate how
   THP-disable is configured for the calling thread (with or without
   PR_THP_DISABLE_EXCEPT_ADVISED).
-  PR_SET_THP_DISABLE now uses arg3 to specify whether to disable THP
   completely for the process, or disable except madvise
   (PR_THP_DISABLE_EXCEPT_ADVISED).

Link: [1] <https://github.com/torvalds/linux/commit/9dc21bbd62edeae6f63e6f25e1edb7167452457b>
Signed-off-by: Usama Arif <usamaarif642@gmail.com>
Message-ID: <20251105134811.3170745-1-usamaarif642@gmail.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

src/bin/grepc: Remove unnecessary 'sort|uniq' call

This was necessary when we used one pcre2grep(1) call per regex.
but long ago we switched to a single pcre2grep(1) call for all regexes,
which only prints each file once, and thus we don't need uniq(1).

Signed-off-by: Alejandro Colomar <alx@kernel.org>

src/bin/grepc, man/man1/grepc.1: -r: Remove flag

Colorizing is imperfect, so don't do it.
Let the user colorize it by piping to grep(1), if needed.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

etc/cpplint/cpplint.cfg: Ignore build/include_what_you_use

We already run iwyu(1) on a separate target (lint-c-iwyu).

Signed-off-by: Alejandro Colomar <alx@kernel.org>

src/bin/, man/man1/grepc{,_c}.1: Various improvements to the scripts; mainly grepc(1)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

src/bin/grepc_c: Use non-capturing groups where possible

Signed-off-by: Alejandro Colomar <alx@kernel.org>

src/bin/grepc_c: -te, -tue: Reuse grepc_c_t_sue_decl_()

Signed-off-by: Alejandro Colomar <alx@kernel.org>

src/bin/grepc_c: -tt: Add support for enums with underlying type

Signed-off-by: Alejandro Colomar <alx@kernel.org>

src/bin/grepc_c: -tt, -tut: Reuse code

Signed-off-by: Alejandro Colomar <alx@kernel.org>

src/bin/grepc_c: -tt: Remove negative lookahead

It was useful when we had separate calls to pcre2grep(1) for each
pattern, but now that we call it once, it's fine to have overlapping
regexes.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

src/bin/grepc_c: -tut: Merge grepc_c_ut_su() and grepc_c_ut_td_su()

Signed-off-by: Alejandro Colomar <alx@kernel.org>

src/bin/grepc_c: Split grepc_c_body_() and grepc_c_u_body_()

grepc_c_u_body_() consumes a lot of resources unnecessarily.
This is especially true within grepc_c_t_td_braced(), which was crashing
with relatively small code.  The reason is that grepc_c_t_td_braced()
searches for code where the identifier is last, so it needs to store
a lot of code just in case.  For that, an optimized grepc_c_body_() is
more appropriate.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

src/bin/grepc_c: -tt, -tut: Reuse grepc_c_body_()

In the last case, it removes part of the regex, but I believe that
wasn't necessary.

Also, while at this, make the trailing of the regexes to be consistently
'[^;];', which is simpler, and seems to work just fine.  If we ever see
some false positives or negatives with this regex, we should fix them
all at once, and maybe put them in a helper function.

Signed-off-by: Alejandro Colomar <alx@kernel.org>