Re: [PATCH v3 3/6] perf tools: Decompress kernel module when reading DSO data

[Namhyung Kim <namhyung@kernel.org>]

Hi Arnaldo,

On Wed, Jun 07, 2017 at 07:23:05PM -0300, Arnaldo Carvalho de Melo wrote:
> Em Thu, Jun 08, 2017 at 12:39:00AM +0900, Namhyung Kim escreveu:
> > Currently perf decompresses kernel modules when loading symbol table but
> > it missed to do it when reading raw data.
> > 
> > Signed-off-by: Namhyung Kim <namhyung@kernel.org>
> > ---
> >  tools/perf/util/dso.c | 16 +++++++++++++++-
> >  1 file changed, 15 insertions(+), 1 deletion(-)
> > 
> > diff --git a/tools/perf/util/dso.c b/tools/perf/util/dso.c
> > index 59257ce4580e..2a0c689227f8 100644
> > --- a/tools/perf/util/dso.c
> > +++ b/tools/perf/util/dso.c
> > @@ -463,8 +463,22 @@ static int __open_dso(struct dso *dso, struct machine *machine)
> >  		return -EINVAL;
> >  	}
> >  
> > -	if (!is_regular_file(name))
> > +	if (!is_regular_file(name)) {
> > +		free(name);
> >  		return -EINVAL;
> > +	}
> 
> Humm, this looks like something for a separate patch? Its a pre-existing
> leak, separate issue.

OK, will separate.

> 
> That error path in __open_dso() could be consolidated, but the above
> hunk is the minimal fix for perf/urgent, where I think your series
> should go from what I've read so far, agreed?

Agreed.

Thanks,
Namhyung


> 
> > +
> > +	if (dso__needs_decompress(dso)) {
> > +		char newpath[KMOD_DECOMP_LEN];
> > +		size_t len = sizeof(newpath);
> > +
> > +		if (dso__decompress_kmodule_path(dso, name, newpath, len) < 0) {
> > +			free(name);
> > +			return -1;
> > +		}
> > +
> > +		strcpy(name, newpath);
> > +	}
> >  
> >  	fd = do_open(name);
> >  	free(name);
> > -- 
> > 2.13.0