Re: [PATCH] perf tools: session: avoid infinite loop

[Adrian Hunter <adrian.hunter@intel.com>]

On 18/09/15 18:29, Arnaldo Carvalho de Melo wrote:
> Em Fri, Sep 18, 2015 at 04:18:51PM +0100, Mark Rutland escreveu:
>> On Fri, Sep 18, 2015 at 04:00:18PM +0100, Arnaldo Carvalho de Melo wrote:
>>> After applying your patch it works, had to tweak the commit log to avoid
>>> starting lines with ---, breaks git scripts, also added a commiter log,
>>> check it, patch is below, after the one I used to not process any
>>> samples.
>  
>> Sorry for the '---' problem, I'll bear that in mind in future.
>  
>> Your commit log looks fine, though I'm slightly confused by the
>> Reported-by line -- did you mean to add that?
> 
> Sorry, I mean Tested-by:, will replace, guess i can replace the one for
> Adrian from Cc: to Tested-by too, right?

Yes

> 
> - Arnaldo
>  
>> Mark.
>>
>>> commit dd486ec4aa33cfca2fd912ef501d49909005de79
>>> Author: Mark Rutland <mark.rutland@arm.com>
>>> Date:   Wed Sep 16 18:18:49 2015 +0100
>>>
>>>     perf record: Avoid infinite loop at buildid processing with no samples
>>>     
>>>     If a session contains no events, we can get stuck in an infinite loop in
>>>     __perf_session__process_events, with a non-zero file_size and data_offset, but
>>>     a zero data_size.
>>>     
>>>     In this case, we can mmap the entirety of the file (consisting of the file and
>>>     attribute headers), and fetch_mmaped_event will correctly refuse to read any
>>>     (unmapped and non-existent) event headers. This causes
>>>     __perf_session__process_events to unmap the file and retry with the exact same
>>>     parameters, getting stuck in an infinite loop.
>>>     
>>>     This has been observed to result in an exit-time hang when counting
>>>     rare/unschedulable events with perf record, and can be triggered artificially
>>>     with the script below:
>>>     
>>>       ----
>>>       #!/bin/sh
>>>       printf "REPRO: launching perf\n";
>>>       ./perf record -e software/config=9/ sleep 1 &
>>>       PERF_PID=$!;
>>>       sleep 0.002;
>>>       kill -2 $PERF_PID;
>>>       printf "REPRO: waiting for perf (%d) to exit...\n" "$PERF_PID";
>>>       wait $PERF_PID;
>>>       printf "REPRO: perf exited\n";
>>>       ----
>>>     
>>>     To avoid this, have __perf_session__process_events bail out early when
>>>     the file has no data (i.e. it has no events).
>>>     
>>>     Commiter note:
>>>     
>>>     I only managed to reproduce this when setting
>>>     /proc/sys/kernel/kptr_restrict to '1' and changing the code to
>>>     purposefully not process any samples and no synthesized samples, i.e.
>>>     kptr_restrict prevents 'record' from synthesizing the kernel mmaps for
>>>     vmlinux + modules and since it is a workload started from perf, we don't
>>>     synthesize mmap/comm records for existing threads.
>>>     
>>>     Adrian Hunter managed to reproduce it in his environment tho.
>>>     
>>>     Reported-by: Arnaldo Carvalho de Melo <acme@kernel.org>
>>>     Signed-off-by: Mark Rutland <mark.rutland@arm.com>
>>>     Cc: Adrian Hunter <adrian.hunter@intel.com>
>>>     Cc: Jiri Olsa <jolsa@redhat.com>
>>>     Cc: Peter Zijlstra <peterz@infradead.org>
>>>     Link: http://lkml.kernel.org/r/1442423929-12253-1-git-send-email-mark.rutland@arm.com
>>>     Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
>>>
>>> diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
>>> index 8a4537ee9bc3..fc3f7c922f99 100644
>>> --- a/tools/perf/util/session.c
>>> +++ b/tools/perf/util/session.c
>>> @@ -1580,7 +1580,10 @@ static int __perf_session__process_events(struct perf_session *session,
>>>  	file_offset = page_offset;
>>>  	head = data_offset - page_offset;
>>>  
>>> -	if (data_size && (data_offset + data_size < file_size))
>>> +	if (data_size == 0)
>>> +		goto out;
>>> +
>>> +	if (data_offset + data_size < file_size)
>>>  		file_size = data_offset + data_size;
>>>  
>>>  	ui_progress__init(&prog, file_size, "Processing events...");
>>>
>