Re: [PATCH 4/9 net-next] ipv6: replace if/BUG by BUG_ON

[Fabian Frederick <fabf@skynet.be>]

> On 31 March 2015 at 17:17 David Miller <davem@davemloft.net> wrote:
>
>
> From: YOSHIFUJI Hideaki <hideaki.yoshifuji@miraclelinux.com>
> Date: Tue, 31 Mar 2015 12:50:27 +0900
>
> > Hi,
> >
> > Fabian Frederick wrote:
> >> Signed-off-by: Fabian Frederick <fabf@skynet.be>
> >> ---
> >>  net/ipv6/addrconf.c                     | 3 +--
> >>  net/ipv6/esp6.c                         | 3 +--
> >>  net/ipv6/netfilter/nf_conntrack_reasm.c | 3 +--
> >>  3 files changed, 3 insertions(+), 6 deletions(-)
> >>
> >> diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
> >> index 2660263..e205918 100644
> >> --- a/net/ipv6/addrconf.c
> >> +++ b/net/ipv6/addrconf.c
> >> @@ -4805,8 +4805,7 @@ static int inet6_set_link_af(struct net_device *dev,
> >> const struct nlattr *nla)
> >>    if (!idev)
> >>            return -EAFNOSUPPORT;
> >> 
> >> -  if (nla_parse_nested(tb, IFLA_INET6_MAX, nla, NULL) < 0)
> >> -          BUG();
> >> +  BUG_ON(nla_parse_nested(tb, IFLA_INET6_MAX, nla, NULL) < 0);
> >> 
> >
> > NACK, I do not prefer using BUG_ON() with side effects.
>
> Agreed, I do not like these changes at all and will not be
> applying them.

I guess patches like these should be reverted then ?

e3f0b86b996d
"ipv6: Use BUG_ON"

8242fc33925c
"net: ipv6: Use BUG_ON"

Some existing BUG_ON() as well ? ; eg second one in net/decnet/dn_rules.c:
void __init dn_fib_rules_init(void)
{
        dn_fib_rules_ops =
                fib_rules_register(&dn_fib_rules_ops_template, &init_net);
        BUG_ON(IS_ERR(dn_fib_rules_ops));
        BUG_ON(fib_default_rule_add(dn_fib_rules_ops, 0x7fff,
                                    RT_TABLE_MAIN, 0));
}

Regards,
Fabian