netdev - Re: [PATCH net-next 1/3] perf: generalize perf

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] [day] [month] [year] [list]

Message-ID: <56CF398A.8020905@fb.com>
Date:	Thu, 25 Feb 2016 09:27:38 -0800
From:	Alexei Starovoitov <ast@...com>
To:	Peter Zijlstra <peterz@...radead.org>
CC:	"David S. Miller" <davem@...emloft.net>,
	Ingo Molnar <mingo@...nel.org>,
	Steven Rostedt <rostedt@...dmis.org>,
	Wang Nan <wangnan0@...wei.com>,
	Daniel Borkmann <daniel@...earbox.net>,
	Brendan Gregg <brendan.d.gregg@...il.com>,
	<netdev@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH net-next 1/3] perf: generalize perf_callchain

On 2/25/16 8:47 AM, Peter Zijlstra wrote:
> On Wed, Feb 17, 2016 at 07:58:57PM -0800, Alexei Starovoitov wrote:
>> +static inline int perf_callchain_store(struct perf_callchain_entry *entry, u64 ip)
>>   {
>> +	if (entry->nr < PERF_MAX_STACK_DEPTH) {
>>   		entry->ip[entry->nr++] = ip;
>> +		return 0;
>> +	} else {
>> +		return -1; /* no more room, stop walking the stack */
>> +	}
>>   }
>
> Why 0 and -1 ?

because that's the interface you had for callbacks in
'struct stacktrace_ops' including a comment there:
/* On negative return stop dumping */

> What's wrong with something like:
>
> static inline bool perf_callchain_store(struct perf_callchain_entry *entry, u64 ip)
> {
> 	if (entry->nr < PERF_MAX_STACK_DEPTH) {
> 		entry->ip[entry->nr++] = ip;
> 		return true;
> 	}
> 	return false;
> }

I would prefer something like this as well, but it would look
inconsistent with what is already there. To make it bool
one would need to change struct stacktrace_ops for all archs
and touch a lot of files all over the tree.
Way more than 51 insertions(+), 29 deletions(-) as this patch did
for no real gain.
It's better to be consistent with existing code.