Landed D55875 after running extensive tests.

Rebased on latest main.

In D56287#1287814, @sjg wrote:

In D56287#1287811, @imp wrote:

I like this, and also @kevans 's suggestion.

So rename it to secure_setenv ?

In D56287#1287811, @imp wrote:

I like this, and also @kevans 's suggestion.

I like this, and also @kevans 's suggestion.

In D56287#1287801, @kevans wrote:

I think the concept is fine, my main nit is: do we want to call this secure_setenv for symmetry with the secure_getenv name we use in libc, for use in security-sensitive getenv() calls?

This one should be in the list of blanket approval.

Thank you!

I think the concept is fine, my main nit is: do we want to call this secure_setenv for symmetry with the secure_getenv name we use in libc, for use in security-sensitive getenv() calls?

In D56228#1286887, @pouria wrote:

@bz
I'd appreciate if your review this again when you have time.
I won't proceed without your approval.

Confirmed: it works for aarch64, finally.

now jail build fine, symbols presented in libm.so.5, try to build math consumers

I've committed the patches together with the update to fwupd 2.1.1. Thanks a lot Sergii!

In D56110#1287730, @ziaee wrote:

Good point. We should probably keep it consistent with those. I think aligning all of them to be more similar should probably be considered over the mailing list and in a different revision, if we decide to go down that route. Should we RELNOTES: yes or discuss this with others?

Good point. We should probably keep it consistent with those. I think aligning all of them to be more similar should probably be considered over the mailing list and in a different revision, if we decide to go down that route. Should we RELNOTES: yes or discuss this with others?

attempt to improve

The version that builds with TARGET=arm64

Are my inline comments visible ? since at the bottom it page it lists them with a "Unsubmitted" label, as in this patch needs revision ? or as in other people can't see them ?

Yes. You must've figured out by now, but in Phabricator, unlike on Github, you must click "Submit" for inline comments to be visible. And BTW if you prefer Github PRs to Phabricator, you can use those too. Both are acceptable ways to submit patches to FreeBSD.

I checked other Broadcom/AVAGO/LSI drivers (mpr, mpt). They all use "driver_version". So I think they should be either consistently "driver_version" or "version".

In D56110#1287717, @ziaee wrote:

it seems more common through the tree to just use "version" here, e.g kern.version or hw.nvidia.version.

it seems more common through the tree to just use "version" here, e.g kern.version or hw.nvidia.version.

Please take https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=293652#c9 into consideration.

Any comments?