[Bug 257964] snp(4) manual page is out of sync with the device implementation on 12.2

From: <bugzilla-noreply_at_freebsd.org>
Date: Fri, 20 Aug 2021 09:19:02 +0000
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=257964

            Bug ID: 257964
           Summary: snp(4) manual page is out of sync with the device
                    implementation on 12.2
           Product: Documentation
           Version: Latest
          Hardware: Any
                OS: Any
            Status: New
          Severity: Affects Only Me
          Priority: ---
         Component: Manual Pages
          Assignee: bugs_at_FreeBSD.org
          Reporter: bsdpr_at_phoe.frmug.org
                CC: doc_at_FreeBSD.org

This bug report about the snp(4) manual page is a follow-up to a question I
asked in freebsd-hackers:
https://lists.freebsd.org/archives/freebsd-hackers/2021-August/000216.html

1. The description of the device is inaccurate: in "allows users to attach to
any tty and watch activities on it", the plural on "activities" led me to think
that we could snoop on bidirectional traffic. Instead it should be made clear
that we can only snoop on egress streams, ie. data originating from the host.

2. Special values mentioned for the FIONREAD ioctl() are no longer used, in
snp.c none of those symbols are actually present. They should be removed from
the manual page as well as "This version of snp does not return proper error
codes when calling FIONREAD" in the BUGS section. These symbols could however
still be mentioned as being obsolete in the HISTORY section.

3. Today, disappearance, closure and detachment of the snooped device are now
signaled by an EIO return on read() and by an EIO or ENXIO return on write().
This should be documented.

4. There is no handling of overflow as it used to be. Quoting the source code:
"Right now we don't support any form of flow control". This should be
documented.

-- 
You are receiving this mail because:
You are the assignee for the bug.
Received on Fri Aug 20 2021 - 09:19:02 UTC

Original text of this message