netdev - Re: [PATCH iproute2-next 1/8] iplink: Return from function instead of calling exit()

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, 20 Feb 2018 13:27:21 -0700
From:   David Ahern <dsahern@...il.com>
To:     Serhey Popovych <serhe.popovych@...il.com>,
        Stephen Hemminger <stephen@...workplumber.org>
Cc:     netdev@...r.kernel.org
Subject: Re: [PATCH iproute2-next 1/8] iplink: Return from function instead of
 calling exit()

On 2/20/18 1:17 PM, Serhey Popovych wrote:
> Stephen Hemminger wrote:
>> On Tue, 20 Feb 2018 21:39:51 +0200
>> Serhey Popovych <serhe.popovych@...il.com> wrote:
>>
>>> Signed-off-by: Serhey Popovych <serhe.popovych@...il.com>
>>> ---
>>>  ip/iplink.c |   12 ++++++------
>>>  1 file changed, 6 insertions(+), 6 deletions(-)
>>>
>>> diff --git a/ip/iplink.c b/ip/iplink.c
>>> index 74c377c..a2c8108 100644
>>> --- a/ip/iplink.c
>>> +++ b/ip/iplink.c
>>> @@ -653,7 +653,7 @@ int iplink_parse(int argc, char **argv, struct iplink_req *req,
>>>  			NEXT_ARG();
>>>  			if (xdp_parse(&argc, &argv, req, dev_index,
>>>  				      generic, drv, offload))
>>> -				exit(-1);
>>> +				return -1;
>>>  		} else if (strcmp(*argv, "netns") == 0) {
>>>  			NEXT_ARG();
>>>  			if (netns != -1)
>>> @@ -972,12 +972,12 @@ static int iplink_modify(int cmd, unsigned int flags, int argc, char **argv)
>>>  		if (!dev) {
>>>  			fprintf(stderr,
>>>  				"Not enough information: \"dev\" argument is required.\n");
>>> -			exit(-1);
>>> +			return -1;
>>>  		}
>>>  		if (cmd == RTM_NEWLINK && index) {
>>>  			fprintf(stderr,
>>>  				"index can be used only when creating devices.\n");
>>> -			exit(-1);
>>> +			return -1;
>>>  		}
>>>  
>>>  		req.i.ifi_index = ll_name_to_index(dev);
>>> @@ -1392,7 +1392,7 @@ static int do_set(int argc, char **argv)
>>>  	if (!dev) {
>>>  		fprintf(stderr,
>>>  			"Not enough of information: \"dev\" argument is required.\n");
>>> -		exit(-1);
>>> +		return -1;
>>>  	}
>>>  
>>>  	if (newaddr || newbrd) {
>>> @@ -1553,7 +1553,7 @@ static int iplink_afstats(int argc, char **argv)
>>>  			fprintf(stderr,
>>>  				"Command \"%s\" is unknown, try \"ip link help\".\n",
>>>  				*argv);
>>> -			exit(-1);
>>> +			return -1;
>>>  		}
>>>  
>>>  		argv++; argc--;
>>> @@ -1648,5 +1648,5 @@ int do_iplink(int argc, char **argv)
>>>  
>>>  	fprintf(stderr, "Command \"%s\" is unknown, try \"ip link help\".\n",
>>>  		*argv);
>>> -	exit(-1);
>>> +	return -1;
>>>  }
>>
>> Not sure I like this. If given bad input in batch it is better to stop and exit
>> rather than continuing with more bad data.
> 
> When preparing this change I think in opposite direction: we want to
> continue batch mode if single line is broken.
> 

batch mode needs to stop on the line that fails. That said, batch still
fails with the /exit/return/ change

$ cat /tmp/ip.batch
li sh
li foo
li add veth1 type veth peer name veth2

Current command
$ ip -batch /tmp/ip.batch
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode
DEFAULT group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00

<link list snipped>
Command "foo" is unknown, try "ip link help".
$ echo $?
255

$ ip/ip -batch /tmp/ip.batch
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode
DEFAULT group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
<link list snipped>
Command "foo" is unknown, try "ip link help".
Command failed /tmp/ip.batch:2
dsa@...ny:~/iproute2/iproute2-next.git$ echo $?
1

I like that better because it tells me the line that fails.