Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Sat, 10 Apr 2021 22:57:41 +0800
From:   Leo Yan <leo.yan@linaro.org>
To:     Adrian Hunter <adrian.hunter@intel.com>
Cc:     Arnaldo Carvalho de Melo <acme@kernel.org>,
        Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>,
        Mark Rutland <mark.rutland@arm.com>,
        Alexander Shishkin <alexander.shishkin@linux.intel.com>,
        Jiri Olsa <jolsa@redhat.com>,
        Namhyung Kim <namhyung@kernel.org>,
        Kan Liang <kan.liang@linux.intel.com>,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH] perf session: Dump PERF_RECORD_TIME_CONV event
Message-ID: <20210410145741.GA6012@leoy-ThinkPad-X240s>
References: <20210410031930.18064-1-leo.yan@linaro.org>
 <9e100c55-16b1-090e-1e81-334a55bd4b15@intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <9e100c55-16b1-090e-1e81-334a55bd4b15@intel.com>
Precedence: bulk

Hi Adrian,

On Sat, Apr 10, 2021 at 11:46:10AM +0300, Adrian Hunter wrote:

[...]

> Hi Leo
> 
> I think there might be some more work related to this.
> 
> Pedantically, shouldn't you cater for backward compatibility and
> not assume the following were in the perf.data file:
>                                                                                                                                            
>        __u64                    time_cycles;                                                                                                                                             
>        __u64                    time_mask;                                                                                                                                               
>        bool                     cap_user_time_zero;                                                                                                                                      
>        bool                     cap_user_time_short;   
> 
> That means checking the event size.
> 
> Also PERF_RECORD_TIME_CONV should have its own byte-swapper instead of  
> perf_event__all64_swap() - also checking event size.
> 
> i.e. fixes for:
> 
>   commit d110162cafc80dad0622cfd40f3113aebb77e1bb
>   Author: Leo Yan <leo.yan@linaro.org>
>   Date:   Mon Sep 14 19:53:09 2020 +0800
> 
>     perf tsc: Support cap_user_time_short for event TIME_CONV

Indeed!  IIUC, should have three fixes with event size checking:

- One fix for dumping TIME_CONV event;
- One fix for byte-swapper (especially for bool values);
- One fix for commit d110162cafc80dad0622cfd40f3113aebb77e1bb;

Will follow up for the suggestions.  Thanks a lot for your insight
review.

Leo