linux-kernel - Re: [PATCH v1 02/15] perf report: output trace file name in raw trace dump

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: <a1c21077-0383-aa29-cc79-a8921916c63f@linux.intel.com>
Date:   Tue, 20 Oct 2020 19:31:53 +0300
From:   Alexey Budankov <alexey.budankov@...ux.intel.com>
To:     Andi Kleen <ak@...ux.intel.com>
Cc:     Arnaldo Carvalho de Melo <acme@...nel.org>,
        Jiri Olsa <jolsa@...hat.com>,
        Namhyung Kim <namhyung@...nel.org>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Adrian Hunter <adrian.hunter@...el.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...hat.com>,
        linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v1 02/15] perf report: output trace file name in raw trace
 dump


On 12.10.2020 20:06, Alexey Budankov wrote:
> 
> On 12.10.2020 19:01, Andi Kleen wrote:
>> On Mon, Oct 12, 2020 at 11:54:24AM +0300, Alexey Budankov wrote:
>>> @@ -180,7 +183,8 @@ static int ordered_events__deliver_event(struct ordered_events *oe,
>>>  						    ordered_events);
>>>  
>>>  	return perf_session__deliver_event(session, event->event,
>>> -					   session->tool, event->file_offset);
>>> +					   session->tool, event->file_offset,
>>> +					   event->file_path);
>>
>> Wouldn't it be better to pass "event" around now, which would contain at least
>> four of the arguments.
>>
>> These functions are getting entirely too many arguments.
> 
> Well, either approach is possible, and even shrink of two arguments kept at session object.
> However changing function signature more than posted can cause bigger adjustments all over
> the code. So this needs more evaluation prior implementation. 

After brief evaluation it still doesn't look easy. The simplest thing
I could imagine is to probably combine file_path and file_offset at a
struct object on stack and then pass the object by the reference along
function calls. I expect it will roughly cause the same amount of changes
in the code and saves one argument (GP register). It is not that much
but still. However I don't see issues with passing even 6 args on stack.

Alexei