netdev - Re: [PATCH bpf-next 4/4] bpftool: Add struct

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]

Message-ID: <20200317005721.vhruudlmhr637uto@kafai-mbp>
Date:   Mon, 16 Mar 2020 17:57:21 -0700
From:   Martin KaFai Lau <kafai@...com>
To:     Quentin Monnet <quentin@...valent.com>
CC:     <bpf@...r.kernel.org>, Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>, <kernel-team@...com>,
        <netdev@...r.kernel.org>
Subject: Re: [PATCH bpf-next 4/4] bpftool: Add struct_ops support

On Mon, Mar 16, 2020 at 05:24:52PM -0700, Martin KaFai Lau wrote:
> On Mon, Mar 16, 2020 at 11:54:28AM +0000, Quentin Monnet wrote:
> 
> > [...]
> > 
> > > +static int do_unregister(int argc, char **argv)
> > > +{
> > > +	const char *search_type, *search_term;
> > > +	struct res res;
> > > +
> > > +	if (argc != 2)
> > > +		usage();
> > 
> > Or you could reuse the macros in main.h, for more consistency with other
> > subcommands:
> > 
> > 	if (!REQ_ARGS(2))
> > 		return -1;
> Thanks for the review!
> 
> I prefer to print out "usage();" whenever possible but then "-j" gave
> me a 'null' after a json error mesage ...
> 
> # bpftool -j struct_ops unregister
> {"error":"'unregister' needs at least 2 arguments, 0 found"},null
> 
> Then I went without REQ_ARGS(2) which is similar to a few existing
> cases like do_dump(), do_updaate()...etc in map.c.
> 
> That was my consideration.  However, I can go back to use REQ_ARGS(2)
> and return -1 without printing usage.  no strong preference here.
After another look,  I will keep it as is since REQ_ARGS() is a "<"
check.  "argc != 2" is the correct check here.  Otherwise,
allowing 'bpftool struct_ops unregister name cubic dctcp' looks weird.