Re: [PATCH 21/22] perf tools: Add trace-event global object for tracepoint interface

[Namhyung Kim <namhyung@kernel.org>]

2013-11-27 (수), 16:27 +0100, Jiri Olsa:
> On Wed, Nov 27, 2013 at 04:10:45PM +0100, Jiri Olsa wrote:
> > On Wed, Nov 27, 2013 at 10:46:00PM +0900, Namhyung Kim wrote:
> > > Hi Jiri,
> > > 
> > > 2013-11-25 (월), 17:03 +0100, Jiri Olsa:
> > > > On Sat, Nov 23, 2013 at 12:47:14AM +0900, Namhyung Kim wrote:
> > > > > 2013-11-21 (목), 12:01 +0100, Jiri Olsa:
> > > > > > In order to get the proper plugins processing we need to
> > > > > > use full trace-event interface when creating tracepoint
> > > > > > events. So far we were using shortcut to get the parsed
> > > > > > format.
> > > > > > 
> > > > > > Moving current 'event_format__new' function into trace-event
> > > > > > object as 'trace_event__tp_format'.
> > > > > > 
> > > > > > This function uses properly initialized global trace-event
> > > > > > object, ensuring proper plugins processing.
> > > > > 
> > > > > I see this when I compile it on my old ubuntu 10.04 box (with gcc 4.4.3)
> > > > > 
> > > > >   CC       util/trace-event.o
> > > > > cc1: warnings being treated as errors
> > > > > util/trace-event.c: In function 'trace_event__tp_format':
> > > > > util/trace-event.c:63: error: not protecting function: no buffer at
> > > > > least 8 bytes long
> > > > > make[1]: *** [util/trace-event.o] Error 1
> > > > 
> > > > hum, what was the stackprotector status?
> > > > 
> > > > ...                stackprotector: [ on  ]
> > > > ...            stackprotector-all: [ on  ]
> > > 
> > > Right. Both of them were turned on.
> > > 
> > > Auto-detecting system features:
> > > ...                     backtrace: [ on  ]
> > > ...                         dwarf: [ on  ]
> > > ...                fortify-source: [ on  ]
> > > ...                         glibc: [ on  ]
> > > ...                          gtk2: [ on  ]
> > > ...                  gtk2-infobar: [ on  ]
> > > ...                      libaudit: [ on  ]
> > > ...                        libbfd: [ on  ]
> > > ...                        libelf: [ on  ]
> > > ...             libelf-getphdrnum: [ OFF ]
> > > ...                   libelf-mmap: [ on  ]
> > > ...                       libnuma: [ OFF ]
> > > ...                       libperl: [ on  ]
> > > ...                     libpython: [ on  ]
> > > ...             libpython-version: [ on  ]
> > > ...                      libslang: [ on  ]
> > > ...                     libunwind: [ OFF ]
> > > ...                       on-exit: [ on  ]
> > > ...                stackprotector: [ on  ]
> > > ...            stackprotector-all: [ on  ]
> > > ...                       timerfd: [ on  ]
> > 
> > I found that -fstack-protector might not always work:
> > https://bugs.launchpad.net/ubuntu/+source/gcc-4.5/+bug/691722
> > 
> > they say it was fixed in gcc 4.5.0 which I was not
> > able to prove in gcc release notes..
> > 
> > we could disable the warning like in the below
> > for gcc < 4.5.0, could you please test?

Works well for me, thanks.

> > 
> > any other idea? ;-)
> 
> or remove -Wstack-protector option completely, as it's just informative
> seems there's nothing you could do about the warning anyway

Agreed.  It seems just adding -fstack-protector is enough.

Thanks,
Namhyung