usr.sbin.makefs.makefs_cd9660_tests.F_flag fails on a kernel missing 'device cd'

[brd]

Summary

This might not be the best way to go about it, but I couldn't find a simpler way to detect if 'device cd' is present in the kernel.

[brd]

Opps, missed a few other cases that need this..

[imp]

Same thing 4 times... Maybe a routine would be better?

[ngie]

Please use kldstat -m cd instead.

[ngie]

In D10096#208743, @imp wrote:

Same thing 4 times... Maybe a routine would be better?

Agreed. I'll work on standardizing one...

[ngie]

Taking a step back... why is device cd required here ? It seems that the real item that's missing potentially from your kernel is iso9660 support. Can you please note how this is broken when device cd is missing?

[brd]

Good point.. This is what I am seeing when this test runs:

Fail: incorrect exit status: 1, expected: 0
stdout:

stderr:
mount_cd9660: /dev/md4: Operation not supported by device

I'll update the patch.

[brd]

Updated diff to pull in suggestions from @imp and @ngie

[ngie]

I think the "missing cd(4)" -> "failing test" part is a red herring, but I'll have to verify that.

If the issue is happening in cd9660, it's because the vnode provided is either a block or character device. I'm not sure what the intent behind that check is, but it seems like there's an issue between makefs(8) <-> cd9660(4).

Hold on a minute... let me verify some things.

[ngie]

In D10096#208943, @ngie wrote:

I think the "missing cd(4)" -> "failing test" part is a red herring, but I'll have to verify that.

If the issue is happening in cd9660, it's because the vnode provided is either a block or character device. I'm not sure what the intent behind that check is, but it seems like there's an issue between makefs(8) <-> cd9660(4).

Hold on a minute... let me verify some things.

You don't have cd9660 support in your kernel, but you have mount_cd9660 on your system. That's why this is happening.

[brd]

In D10096#208952, @ngie wrote:

You don't have cd9660 support in your kernel, but you have mount_cd9660 on your system. That's why this is happening.

Yeah.. Not sure how to handle that situation. I don't see a way to optionally remove the mount_cd9660 according to ObsoleteFiles.inc. I would appreciate any ideas you have.

[ngie]

In D10096#208988, @brd wrote:

In D10096#208952, @ngie wrote:

You don't have cd9660 support in your kernel, but you have mount_cd9660 on your system. That's why this is happening.

Yeah.. Not sure how to handle that situation. I don't see a way to optionally remove the mount_cd9660 according to ObsoleteFiles.inc. I would appreciate any ideas you have.

You're on the right track. Please rename check_cd_support to check_iso9660_support and check for cd9660 instead of cd.

I'd also state Requires cd9660 filesystem support to be present in the kernel.

(UGH... WHY ISN'T THERE A cd9660 manpage?! I guess I'll have to add it!)

[ngie]

In D10096#209011, @ngie wrote:

...

You're on the right track. Please rename check_cd_support to check_iso9660_support and check for cd9660 instead of cd.

I'd also state Requires cd9660 filesystem support to be present in the kernel.

(UGH... WHY ISN'T THERE A cd9660 manpage?! I guess I'll have to add it!)

Ok, cd9660(5) exists now :)... *phew*

[brd]

Implement @ngie's suggestions.

[ngie]

LGTM -- thanks!