linux-kernel - Re: [PATCH v3] selftests/livepatch: introduce tests

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 24 Apr 2018 11:50:28 -0400
From:   Joe Lawrence <joe.lawrence@...hat.com>
To:     Libor Pechacek <lpechacek@...e.cz>
Cc:     linux-kernel@...r.kernel.org, linux-kselftest@...r.kernel.org,
        live-patching@...r.kernel.org, Jiri Kosina <jikos@...nel.org>,
        Artem Savkov <asavkov@...hat.com>,
        Josh Poimboeuf <jpoimboe@...hat.com>,
        Petr Mladek <pmladek@...e.com>,
        Miroslav Benes <mbenes@...e.cz>,
        Nicolai Stange <nstange@...e.de>
Subject: Re: [PATCH v3] selftests/livepatch: introduce tests

On 04/23/2018 10:43 AM, Joe Lawrence wrote:
> On Fri, Apr 20, 2018 at 02:56:05PM +0200, Libor Pechacek wrote:
>> On Thu 12-04-18 10:54:31, Joe Lawrence wrote:
>>> +	fi
>>> +	echo "$ret" > /dev/kmsg
>>> +}
>>> +
>>> +# unload_mod(modname) - unload a kernel module
>>> +#	modname - module name to unload
>>> +function unload_mod() {
>>> +	local mod="$1"
>>> +
>>> +	# Wait for module reference count to clear ...
>>> +	local i=0
>>> +	while [[ $(cat /sys/module/"$mod"/refcnt) != "0" ]]; do
>>> +		i=$((i+1))
>>> +		if [[ $i -eq $MAX_RETRIES ]]; then
>>> +			die "failed to unload module $mod (refcnt)"
>>> +		fi
>>> +		sleep $RETRY_INTERVAL
>>> +	done
>>
>> The repeating pattern of "while <some test>; do <count>; if <count beyond max
>> retries>; then <die>..." seems to ask for encapsulation.
>>
> 
> Yeah I definitely agree.  I think at some point I had acquired
> bash-fatigue;  I wasn't sure how to cleanly wrap <some test> around that
> extra logic.  In C, I could do something clever with macros or a
> callback function.  My bash scripting isn't great, so I copied and
> pasted my way through it.  Suggestions welcome.
> 

Okay, here's what I came up with... first off, do you prefer this kind
of transition check vs. looking only at a specific module?

  # check_transition() - verify that no livepatch transition in effect
  function check_transition() {
  	grep -q '^1$' /sys/kernel/livepatch/*/transition 2>/dev/null
  }

then wrap the retry/timeout logic like:

  # retry_cmd(cmd) - loop a command until it is successful or
  #		   $MAX_RETRIES, sleeping $RETRY_INTERVAL in
  #		   between tries
  # 	cmd - command and its arguments to run
  function retry_cmd() {
  	local cmd="$*"
  	local i=0
  	while eval "$cmd"; do
  		i=$((i+1))
  		[[ $i -eq $MAX_RETRIES ]] && return 1
  		sleep $RETRY_INTERVAL
  	done
  	return 0
  }

and the callers to something like:

  # wait_for_transition() - wait until all livepatch transitions clear
  function wait_for_transition() {
  	retry_cmd check_transition ||
  		die "failed to complete transition"
  }


I can create similar check() functions to eval for sysfs file existence,
file content, reference count, etc. to remove all the other
retry/timeout loops.

Regards,

-- Joe