Add a missing argument to VNASSERT().

Use VNASSERT() instead of KASSERT().

Incorporate review feedback.

Address Florian's comments.

I don't commit it because sabnzbd requires this port with py-paho-mqtt but the newest py-paho-mqtt 2.0.0 isn't compatible with apprise.

• getblk: fail faster with GB_LOCK_NOWAIT
• GB_LOCK_NOWAIT markj feedback

Address @kib 's comment - unmanaged reservations are now distinguished using an invalid vm_object value instead of a separate variable.

In D45278#1032837, @asomers wrote:

FYI, at $WORK the only packages we include in the DVD image are:

• ports-mgmt/pkg
• sysutils/py-salt , because Salt can install everything else
• net-mgmt/lldpd, which helps us debug network problems that would prevent Salt from working.

For desktop use, I really don't think that the packages on the DVD are very important. We can't possibly put every package on the DVD, or even the most popular packages. So what's the point of having any? Just to reduce users' download times?

In D45245#1032860, @rlibby wrote:

In D45245#1032681, @markj wrote:

My only comment is that, reading the code, it's not immediately obvious why we give up right away when failing to acquire the buf lock, but not when we successfully acquire the buf lock and discover that the buffer identity has changed.

There's no reason not to, it just didn't seem likely enough to worry about. My main issue comes from encountering locked bufs with the expected identity. But maybe it's better to be consistent so that the intent is easier to understand. I'll add this to the next diff but I'll take guidance on whether we actually do this or not.

In D45245#1032681, @markj wrote:

My only comment is that, reading the code, it's not immediately obvious why we give up right away when failing to acquire the buf lock, but not when we successfully acquire the buf lock and discover that the buffer identity has changed.

In D45243#1032364, @des wrote:

Have you considered implementing Apple's copyfile API?

I can only comment on markup nits, I don't have this equipment.

FYI, at $WORK the only packages we include in the DVD image are:

• ports-mgmt/pkg
• sysutils/py-salt , because Salt can install everything else
• net-mgmt/lldpd, which helps us debug network problems that would prevent Salt from working.

For non-ports contributions we should really discourage submissions via bugzilla unless it's what a specific committer you're working with wants. Generally, bugzilla is where patches go to die.

• Rename vnode_disk_create_vobject() to vnode_create_disk_vobject().
• Improve assertions.

In D45256#1032818, @dev_submerge.ch wrote:

In D45256#1032314, @christos wrote:

@dev_submerge.ch I am not really sure if we can implement this in SNDCTL_ENGINEINFO as well. In SNDCTL_ENGINEINFO we choose an "engine"/channel on a device that is already opened, so in the case of an unavailable device, we won't be able to open it in the first place.

There's no 1:1 mapping from engines (virtual channels) to pcm devices anyway. I think the only thing we care about with SNDCTL_ENGINEINFO is that applications can iterate them using numaudioengines as expected.

In D45256#1032314, @christos wrote:

@dev_submerge.ch I am not really sure if we can implement this in SNDCTL_ENGINEINFO as well. In SNDCTL_ENGINEINFO we choose an "engine"/channel on a device that is already opened, so in the case of an unavailable device, we won't be able to open it in the first place.

This seems to match what GCC does, though I'm not sure why either need it. Both drivers pass both of -lgcc and -lgcc_s, whether building an executable or shared library (well, testing with GCC 13.2 on arm64 Linux, as that's easily to hand). Can you elaborate on the exact issue you're seeing without this?

Throwing this out for developers and the community to provide feedback on.

Similar to the iavf(4) update here: https://reviews.freebsd.org/D43093, the documentation team here has generated a new man page for ice(4), but I haven't converted it to the FreeBSD upstream format, yet. I'm fine with this version with some fixes, but if you give me a couple days I can convert it and submit it here for review.

It surprises me that GNU ld doesn't want to create canonical PLTs, but avoiding them is reasonable; they are to functions what copy relocations are to data.

I wonder if this would be useful on risc64 which has similar issues. There I have been trying (but not succeeding) to get GCC's libatomic.a to build and install as part of the freebsdN-gcc ports so that it could be linked to.

Merged with D45256.

Merge with D45150.

Use raw maxcnt value

Address Mark's comments.

I'd like to get this into 14.1 if possible -- please review ASAP.

D45275 and D45276.

Note that GB_LOCK_NOWAIT has somewhat subtle meaning, it does not request to not sleep, or even does not request to not sleep on a buffer lock. It only signals to avoid sleeping on buffer lock with the specified identity.

Ok, I have the same patch locally.

They LGTM.