Sorry. I was confused.

In D43548#993576, @pauamma_gundo.com wrote:

Not sure what you mean by "unrelated to the content", but this is the output of the command above with kern.securelevel having its initial value.

Hugo Documentation: Alias

Restore unintended changes

Removed changes in release and status

bootonly fits in CD

Add comment in sys/conf/kern.mk

Restore share/mk/bsd.sys.mk

Add .error for c94 and c95 flags which are undefined in clang. Other flags are defined by default.

Fix conflict

Fix commit message

style(9)

Remove unnecessary include

Divide _Alignof part to D43409

Change dvd sets to disc copies

I think this should be done by a committer who can commit to the repo without having to upload the patch on phabricator. Closing revision.

This revision is succeeded by D43296. Closing revision

@lwhsu The Github pull request about CI testing is merged now.

Since zstd only uses static_assert, I only took static_assert macro from include/assert.h

Simplifying assert.h

Fixed _Alignof()

I guess this is now ready to go.

$ cd/usr/src/sys
$ grep -r _Alignof
./vm/swap_pager.c:	    NULL, NULL, _Alignof(struct swblk) - 1, 0);
./vm/uma.h:#define	UMA_ALIGNOF(type) (_Alignof(type) - 1)	/* Alignment fit for 'type' */
./riscv/riscv/db_trace.c:			if (!__is_aligned(tf, _Alignof(struct trapframe)) ||
./ufs/ffs/ffs_vnops.c:	(((uintptr_t)(ptr) & (_Alignof(s) - 1)) == 0)
./netinet/libalias/alias_proxy.c:	_Alignas(_Alignof(u_short)) u_char option[OPTION_LEN_BYTES];
./netinet/ip_fw.h:	_Alignas(_Alignof(u_int32_t)) u_int8_t 	opcode;
./arm64/arm64/vfp.c:	    _Alignof(struct vfpstate) - 1, 0);
./arm64/arm64/db_trace.c:			if (!__is_aligned(tf, _Alignof(struct trapframe)) ||
./sys/cdefs.h:#define	_Alignof(x)		alignof(x)
./sys/cdefs.h:#define	_Alignof(x)		__alignof(x)
./sys/_types.h:	long long __max_align1 __aligned(_Alignof(long long));
./sys/_types.h:	long double __max_align2 __aligned(_Alignof(long double));
./amd64/amd64/db_trace.c:	if (!__is_aligned(tf_addr, _Alignof(struct trapframe)) || !INKERNEL(tf_addr)) {
./amd64/amd64/db_trace.c:		if (!__is_aligned(frame, _Alignof(struct amd64_frame)) ||
./amd64/include/vmm.h:_Static_assert(_Alignof(struct vie_op) == 2, "ABI");
./netlink/netlink_message_parser.h:} __aligned(_Alignof(__max_align_t));
./netlink/netlink_message_parser.h:	len = roundup2(len, _Alignof(__max_align_t));
./dev/bhnd/nvram/bhnd_nvram_subr.c:		return (_Alignof(uint8_t));
./dev/bhnd/nvram/bhnd_nvram_subr.c:		return (_Alignof(uint8_t));
./dev/bhnd/nvram/bhnd_nvram_subr.c:		return (_Alignof(uint8_t));
./dev/bhnd/nvram/bhnd_nvram_subr.c:		return (_Alignof(uint16_t));
./dev/bhnd/nvram/bhnd_nvram_subr.c:		return (_Alignof(uint32_t));
./dev/bhnd/nvram/bhnd_nvram_subr.c:		return (_Alignof(uint64_t));
./dev/bhnd/nvram/bhnd_nvram_subr.c:		return (_Alignof(int8_t));
./dev/bhnd/nvram/bhnd_nvram_subr.c:		return (_Alignof(int16_t));
./dev/bhnd/nvram/bhnd_nvram_subr.c:		return (_Alignof(int32_t));
./dev/bhnd/nvram/bhnd_nvram_subr.c:		return (_Alignof(int64_t));

I modified db_trace_c that _Alignof takes struct trapframe instead of *tf.
It seems that db_trace.c is the only file that uses expression in _Alignof.

Differences between c* and gnu* flags are listed here.

Remove unnecessary comment

Fixed pkg-plist

Done in 6d25581fe36672c6ba72c1370d18c1738c57f978. Should be closed

Accepted earlier by @emaste .

In D43253#985801, @agh_riseup.net wrote:

In D43253#985787, @minsoochoo0122_proton.me wrote:

Please add ports group to reviewers, subscribers, and tag otherwise people won't be notified.

I have no idea what that entails. If I Edit Revision via the menu box at the top right, I can find the three fields you mention, however, when I type ports, a number of options are auto populated in a drop down list, none of which are ports.
Is this meant to be standard procedure too?

Remove code for < C99

Restore CFLAGS+= -std=${CSTD}

Removed stdlib division code for < C99

Please add ports group to reviewers, subscribers, and tag otherwise people won't be notified.

In D43237#985747, @imp wrote:

I'd like to see a patch doing what i described for supported version. Needn't be in this review but should be in the series.

In D43237#985660, @imp wrote:

Again, why gnu17 instead of c17 is a question in my mind. Also, does a resulting system work?

In D43237#985661, @imp wrote:

also, I don't see how contrib is omitted.

In D43237#985660, @imp wrote:

Yea, I had vague plans on bumping this to c11 universally as well...

I'd rather we deal with the _Alignof issue as I suggested in the other review.

And we likely need some kind of regression test to ensure that we can still build c89, c99, c11 programs because we have a lot of ports that are at least compiling with c99 and some that do c89 for 'maximum portability' so our headers have to remain compatible over the long haul.

Again, why gnu17 instead of c17 is a question in my mind. Also, does a resulting system work?

Update FREEBSD-upgrade

@brooks Can this revision be accepted?

Resolved build issues

In D43237#985522, @jrtc27 wrote:

What is the minimum Clang version (and GCC version) needed for this?

In D43236#985438, @emaste wrote:

Do you happen to have these changes available in a Git repo somewhere that I can pull from?

Fix function header accordign to style(9)

Please update this revision to zig 0.12.x with LLVM 17.x and add @dch, the maintainer for lang/zig, and portmgr