In D55751#1276024, @cperciva wrote:

Thanks, I fixed the typo. Also I added periods at the end of sentences.

Note that https://bugs.freebsd.org/293654 has been solved by D55713. (I'm not clear on how to insure it is included in 15.1)

Incorporate @kevans 's clever fix that works before and after
lfs.DT_LNK is defined.

In D55751#1275970, @ziaee wrote:

Your version leaves out "regression" which I included because this use worked before.

I'm not attached to my version, we just need to have something that's clear. It's fine to include the word regression, however the sole point of this section is listing regressions.

Update issue number

In D55753#1275581, @vladlen wrote:

In D55753#1275568, @leres wrote:

if we like to keep the style from 15.0 Errata, it would be good to place every sentence in new line in the final output. It needs '+' in the end of every line.

(D55751 does not have this issue)

Are these reviews (D55751 and D55753) considered as two different issues (loader(8) and loader_lua(8)) linked to the same PR? Should we choose only one of them?

if we like to keep the style from 15.0 Errata, it would be good to place every sentence in new line in the final output. It needs '+' in the end of every line.

(I already created D55751)

Address feedback and use ftype to permit symlinks.
Unfortunately lfs.DT_DIR is not defined so use magic value 10

Address feedback and use ftype to permit symlinks.
Unfortunately lfs.DT_DIR is not defined so use magic value 10

Humm... I found that the value for lfs.DT_LNK will be 10 but I suck at git and my attempt to update this review didn't work. I tried:

Address feedback and use ftype to permit symlinks.
Unfortunately lfs.DT_DIR is not defined so use magic value 10

At first I tried to find a DL_* that would identify a symlink but could not (I think I tried DT_LNK). I also tried debugging the logic using lua from the command line but wasn't easily able to figure out how to use the luafilesystem port from the command line...

Also see https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=293654

I meant to include an example of multiple -e flags:

I'm trying to create an arm64.aarch64 poudriere jail (on a amd64 host) from 16.0-CURRENT@44f3e9f7f6c (this approach has worked for me in the past) but d4dab32448ce doesn't appear to be sufficient for the cross-compile case:

I don't have time this week but will definitely look into this next week, including adding a test.