LTTng driver submission for inclusion in staging (resend)

Mathieu Desnoyers mathieu.desnoyers at efficios.com
Sun Nov 27 10:07:42 UTC 2011


* Greg KH (greg at kroah.com) wrote:
> On Sat, Nov 26, 2011 at 10:03:25AM -0500, Mathieu Desnoyers wrote:
> > 
> > Hi,
> > 
> > I hereby propose the LTTng kernel tracer for inclusion into the staging
> > area of the Linux kernel. For those eager to try it out, and to get
> > information about the features of this tracer, you should grab the
> > control/consumer/viewer tools at: http://lttng.org/lttng2.0
> > 
> > Feedback is very welcome!
> 
> I applied this, but it failed to build:
> 
>   CC [M]  drivers/staging/lttng/ltt-ring-buffer-client-discard.o
> In file included from drivers/staging/lttng/ltt-tracer.h:24:0,
>                  from drivers/staging/lttng/ltt-ring-buffer-client-discard.c:12:
> drivers/staging/lttng/wrapper/trace-clock.h: In function ‘get_trace_clock’:
> drivers/staging/lttng/wrapper/trace-clock.h:66:2: warning: #warning "LTTng: Using mainline kernel monotonic clock. NMIs will not be traced, and expect significant performance degradation compared to the LTTng trace clocks. Integration of the LTTng 0.x trace clocks into LTTng 2.0 is planned in a near future." [-Wcpp]
> In file included from drivers/staging/lttng/wrapper/ringbuffer/frontend_types.h:1:0,
>                  from drivers/staging/lttng/ltt-ring-buffer-client.h:18,
>                  from drivers/staging/lttng/ltt-ring-buffer-client-discard.c:17:
> drivers/staging/lttng/wrapper/ringbuffer/../../lib/ringbuffer/frontend_types.h: At top level:
> drivers/staging/lttng/wrapper/ringbuffer/../../lib/ringbuffer/frontend_types.h:23:86: fatal error: ../../lib/prio_heap/lttng_prio_heap.h: No such file or directory
> compilation terminated.
> 
> 
> The build warning is a bit harsh, and should be a Kconfig option or a run-time
> kernel log message, please fix that.  No new code should ever add build
> warnings, especially ones like this.
> 
> The build error isn't ok, so I can't apply this series, sorry :(
> 
> Care to fix this up and resend?

No problem. I'll look into what happened in the git-to-patch transition,
remove the lgpl/gpl files, and move the #warning messages into the TODO
file, then resend.

Thanks,

Mathieu

> 
> thanks,
> 
> greg k-h

-- 
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com



More information about the devel mailing list