ktrdump(8): Capsicumify

[cem]

Summary

We restrict the (optional) input file and output files. It would be
nice to restrict the KVM files, but that's up to libkvm.

We wait until after kvm_nlist() is invoked to cap_enter() because
kldsym() isn't supported in the Capsicum sandbox.

Test Plan

• Someone™ needs to test this with a KTR-enabled kernel. My GENERIC kernel doesn't have KTR and I don't want to reboot into a different kernel quite yet.

[emaste]

usr.bin/ktrdump/ktrdump.c

31–32 ↗

(On Diff #20419)

style(9) as noted elsewhere

94 ↗

(On Diff #20419)

I think cap_rights_t should sort with structs

144–146 ↗

(On Diff #20419)

what's up here?

[cem]

usr.bin/ktrdump/ktrdump.c

31–32 ↗

(On Diff #20419)

Well, not style(9) :-).

94 ↗

(On Diff #20419)

What specific ordering do you want here? It is adjacent to the only other struct variable.

144–146 ↗

(On Diff #20419)

The fopen() should have stayed fopen(optarg). (Initially, I had the strlcpy first. After I moved it below, I forgot to change the fopen argument.)

[imp]

usr.bin/ktrdump/ktrdump.c

31–32 ↗

(On Diff #20419)

Yes, style(9).

[emaste]

usr.bin/ktrdump/ktrdump.c

31–32 ↗

(On Diff #20419)

style, not style(9) -- it is our style, it's just not documented as such in the man page

94 ↗

(On Diff #20419)

there's struct ktr_entry above; I didn't notice the other struct below.

[imp]

usr.bin/ktrdump/ktrdump.c

31–32 ↗

(On Diff #20419)

style(9) is an ambiguous term. It is also the shorthand for our style as (poorly) documented in /usr/src/share/man/man9/style.9 and tribal knowledge of the group. It's also an emotionally charged topic that elicits PTSD type behavior and responses.

[cem]

usr.bin/ktrdump/ktrdump.c

31–32 ↗

(On Diff #20419)

I think imp's reading into style(9) a suggestion that isn't really there. It should be explicit in the document. To me "style(9)" refers specifically to the document.

94 ↗

(On Diff #20419)

That's a pointer. I have no idea how the sorting is supposed to work, though.

[emaste]

usr.bin/ktrdump/ktrdump.c

31–32 ↗

(On Diff #20419)

Added in rS306010

[imp]

usr.bin/ktrdump/ktrdump.c

31–32 ↗

(On Diff #20419)

I've made some sensible changes to style(9) that you should find favorable. Or at least not unfavorable...

[emaste]

usr.bin/ktrdump/ktrdump.c

94 ↗

(On Diff #20419)

Sigh, so it is.

style(9) does actually have guidance here: "When declaring variables in functions declare them sorted by size, then in alphabetical order"

[cem]

• Sort headers correctly
• Unbreak '-o' fopen(3) call

usr.bin/ktrdump/ktrdump.c

94 ↗

(On Diff #20419)

Left as-is due to lack of any apparent prevailing order. :-)

[emaste]

to be updated with capsicum helpers

[cem]

Use capsicum helper for opening NLS catpages.