Re: FW: [RFC 0/5] perf tools: Add perf data CTF conversion

[Sebastian Andrzej Siewior <bigeasy@linutronix.de>]

I try to get through my ctf mailbox and I hoped I can finish it today
but I don't make it completely…

On 11/06/2014 04:25 AM, Alexandre Montplaisir wrote:
> "mvn clean install". It is the Maven equivalent of "./configure && make" ;)
> 
> Or if you want to build a standalone application (RCP):
> mvn clean install -Pbuild-rcp -Dmaven.test.skip=true

thanks.

> Yes, the state dump is something specific to LTTng. It allows us to know
> about processes that exist on the system, even if they are sleeping for
> the whole duration of the trace (and thus, would not show up in the
> trace at all).
> 
> But even if these events are not present, we can still know about active
> processes when they do shed_switch's for example.

good to know.

> IIRC, compat_sys is for instance for 32-bit system calls on a 64-bit
> kernel. Perhaps the "compat" system calls are recorded as standard
> system call events with perf? We could test it once we get the base
> things working.

I booted a x86-64 in a 32bit userland and I expected to see them
somewhere but nothing. However if the task has a compat flag then it
would be the same information, right?

>> - static final String SYS_CLONE = "sys_clone";
>>    here we have
>>      syscalls:sys_enter_clone
>>      syscalls:sys_exit_clone
>>    I guess the enter is what you are looking for.
>>
>> For the fields, this is one event with alle the members we have. Please
>> note that lttng saves the members with the _ prefix and I haven't seen
>> that prefix in that .java file.
> 
> As Mathieu explained in his reply, in LTTng-CTF they have a _ before
> field names. In our parser, we take out the first character if it is an
> underscore. So it should still work with underscore-less fields.

I see. I think it started working once I added the underline prefix but
I might be wrong. Let me see what Mathieu says if I may leave that
prefix out.

> Cheers,
> Alexandre
Sebastian