[v1,00/11] perf: Fix errors detected by Smatch

Message ID 20190702103420.27540-1-leo.yan@linaro.org
Headers show
Series
  • perf: Fix errors detected by Smatch
Related show

Message

Leo Yan July 2, 2019, 10:34 a.m.
When I used static checker Smatch for perf building, the main target is
to check if there have any potential issues in Arm cs-etm code.  So
finally I get many reporting for errors/warnings.

I used below command for using static checker with perf building:

  # make VF=1 CORESIGHT=1 -C tools/perf/ \
    CHECK="/root/Work/smatch/smatch --full-path" \
    CC=/root/Work/smatch/cgcc | tee smatch_reports.txt

I reviewed the errors one by one, if I understood some of these errors
so changed the code as I can, this patch set is the working result; but
still leave some errors due to I don't know what's the best way to fix
it.  There also have many inconsistent indenting warnings.  So I firstly
send out this patch set and let's see what's the feedback from public
reviewing.

Leo Yan (11):
  perf report: Smatch: Fix potential NULL pointer dereference
  perf stat: Smatch: Fix use-after-freed pointer
  perf top: Smatch: Fix potential NULL pointer dereference
  perf annotate: Smatch: Fix dereferencing freed memory
  perf trace: Smatch: Fix potential NULL pointer dereference
  perf hists: Smatch: Fix potential NULL pointer dereference
  perf map: Smatch: Fix potential NULL pointer dereference
  perf session: Smatch: Fix potential NULL pointer dereference
  perf intel-bts: Smatch: Fix potential NULL pointer dereference
  perf intel-pt: Smatch: Fix potential NULL pointer dereference
  perf cs-etm: Smatch: Fix potential NULL pointer dereference

 tools/perf/builtin-report.c    |  4 ++--
 tools/perf/builtin-stat.c      |  2 +-
 tools/perf/builtin-top.c       |  8 ++++++--
 tools/perf/builtin-trace.c     |  5 +++--
 tools/perf/ui/browsers/hists.c | 13 +++++++++----
 tools/perf/util/annotate.c     |  6 ++----
 tools/perf/util/cs-etm.c       |  2 +-
 tools/perf/util/intel-bts.c    |  5 ++---
 tools/perf/util/intel-pt.c     |  5 ++---
 tools/perf/util/map.c          |  7 +++++--
 tools/perf/util/session.c      |  3 +++
 11 files changed, 36 insertions(+), 24 deletions(-)

-- 
2.17.1

Comments

Jiri Olsa July 2, 2019, 11:07 a.m. | #1
On Tue, Jul 02, 2019 at 06:34:09PM +0800, Leo Yan wrote:
> When I used static checker Smatch for perf building, the main target is

> to check if there have any potential issues in Arm cs-etm code.  So

> finally I get many reporting for errors/warnings.

> 

> I used below command for using static checker with perf building:

> 

>   # make VF=1 CORESIGHT=1 -C tools/perf/ \

>     CHECK="/root/Work/smatch/smatch --full-path" \

>     CC=/root/Work/smatch/cgcc | tee smatch_reports.txt

> 

> I reviewed the errors one by one, if I understood some of these errors

> so changed the code as I can, this patch set is the working result; but

> still leave some errors due to I don't know what's the best way to fix

> it.  There also have many inconsistent indenting warnings.  So I firstly

> send out this patch set and let's see what's the feedback from public

> reviewing.

> 

> Leo Yan (11):

>   perf report: Smatch: Fix potential NULL pointer dereference

>   perf stat: Smatch: Fix use-after-freed pointer

>   perf top: Smatch: Fix potential NULL pointer dereference

>   perf annotate: Smatch: Fix dereferencing freed memory

>   perf trace: Smatch: Fix potential NULL pointer dereference

>   perf hists: Smatch: Fix potential NULL pointer dereference

>   perf map: Smatch: Fix potential NULL pointer dereference

>   perf session: Smatch: Fix potential NULL pointer dereference

>   perf intel-bts: Smatch: Fix potential NULL pointer dereference

>   perf intel-pt: Smatch: Fix potential NULL pointer dereference

>   perf cs-etm: Smatch: Fix potential NULL pointer dereference


from quick look it all looks good to me, nice tool ;-)

Acked-by: Jiri Olsa <jolsa@kernel.org>


jirka

> 

>  tools/perf/builtin-report.c    |  4 ++--

>  tools/perf/builtin-stat.c      |  2 +-

>  tools/perf/builtin-top.c       |  8 ++++++--

>  tools/perf/builtin-trace.c     |  5 +++--

>  tools/perf/ui/browsers/hists.c | 13 +++++++++----

>  tools/perf/util/annotate.c     |  6 ++----

>  tools/perf/util/cs-etm.c       |  2 +-

>  tools/perf/util/intel-bts.c    |  5 ++---

>  tools/perf/util/intel-pt.c     |  5 ++---

>  tools/perf/util/map.c          |  7 +++++--

>  tools/perf/util/session.c      |  3 +++

>  11 files changed, 36 insertions(+), 24 deletions(-)

> 

> -- 

> 2.17.1

>
Leo Yan July 3, 2019, 1:48 a.m. | #2
On Tue, Jul 02, 2019 at 01:07:43PM +0200, Jiri Olsa wrote:
> On Tue, Jul 02, 2019 at 06:34:09PM +0800, Leo Yan wrote:

> > When I used static checker Smatch for perf building, the main target is

> > to check if there have any potential issues in Arm cs-etm code.  So

> > finally I get many reporting for errors/warnings.

> > 

> > I used below command for using static checker with perf building:

> > 

> >   # make VF=1 CORESIGHT=1 -C tools/perf/ \

> >     CHECK="/root/Work/smatch/smatch --full-path" \

> >     CC=/root/Work/smatch/cgcc | tee smatch_reports.txt

> > 

> > I reviewed the errors one by one, if I understood some of these errors

> > so changed the code as I can, this patch set is the working result; but

> > still leave some errors due to I don't know what's the best way to fix

> > it.  There also have many inconsistent indenting warnings.  So I firstly

> > send out this patch set and let's see what's the feedback from public

> > reviewing.

> > 

> > Leo Yan (11):

> >   perf report: Smatch: Fix potential NULL pointer dereference

> >   perf stat: Smatch: Fix use-after-freed pointer

> >   perf top: Smatch: Fix potential NULL pointer dereference

> >   perf annotate: Smatch: Fix dereferencing freed memory

> >   perf trace: Smatch: Fix potential NULL pointer dereference

> >   perf hists: Smatch: Fix potential NULL pointer dereference

> >   perf map: Smatch: Fix potential NULL pointer dereference

> >   perf session: Smatch: Fix potential NULL pointer dereference

> >   perf intel-bts: Smatch: Fix potential NULL pointer dereference

> >   perf intel-pt: Smatch: Fix potential NULL pointer dereference

> >   perf cs-etm: Smatch: Fix potential NULL pointer dereference

> 

> from quick look it all looks good to me, nice tool ;-)

> 

> Acked-by: Jiri Olsa <jolsa@kernel.org>


Thanks for reviewing, Jiri.

@Arnaldo, Just want to check, will you firstly pick up 01~05, 07,
08/11 patches if you think they are okay?  Or you want to wait me to
spin new patch set with all patches after I gather all comments?

Thanks,
Leo Yan

> >  tools/perf/builtin-report.c    |  4 ++--

> >  tools/perf/builtin-stat.c      |  2 +-

> >  tools/perf/builtin-top.c       |  8 ++++++--

> >  tools/perf/builtin-trace.c     |  5 +++--

> >  tools/perf/ui/browsers/hists.c | 13 +++++++++----

> >  tools/perf/util/annotate.c     |  6 ++----

> >  tools/perf/util/cs-etm.c       |  2 +-

> >  tools/perf/util/intel-bts.c    |  5 ++---

> >  tools/perf/util/intel-pt.c     |  5 ++---

> >  tools/perf/util/map.c          |  7 +++++--

> >  tools/perf/util/session.c      |  3 +++

> >  11 files changed, 36 insertions(+), 24 deletions(-)

> > 

> > -- 

> > 2.17.1

> >
Arnaldo Carvalho de Melo July 3, 2019, 6:18 p.m. | #3
Em Wed, Jul 03, 2019 at 09:48:08AM +0800, Leo Yan escreveu:
> On Tue, Jul 02, 2019 at 01:07:43PM +0200, Jiri Olsa wrote:

> > On Tue, Jul 02, 2019 at 06:34:09PM +0800, Leo Yan wrote:

> > > When I used static checker Smatch for perf building, the main target is

> > > to check if there have any potential issues in Arm cs-etm code.  So

> > > finally I get many reporting for errors/warnings.

> > > 

> > > I used below command for using static checker with perf building:

> > > 

> > >   # make VF=1 CORESIGHT=1 -C tools/perf/ \

> > >     CHECK="/root/Work/smatch/smatch --full-path" \

> > >     CC=/root/Work/smatch/cgcc | tee smatch_reports.txt

> > > 

> > > I reviewed the errors one by one, if I understood some of these errors

> > > so changed the code as I can, this patch set is the working result; but

> > > still leave some errors due to I don't know what's the best way to fix

> > > it.  There also have many inconsistent indenting warnings.  So I firstly

> > > send out this patch set and let's see what's the feedback from public

> > > reviewing.

> > > 

> > > Leo Yan (11):

> > >   perf report: Smatch: Fix potential NULL pointer dereference

> > >   perf stat: Smatch: Fix use-after-freed pointer

> > >   perf top: Smatch: Fix potential NULL pointer dereference

> > >   perf annotate: Smatch: Fix dereferencing freed memory

> > >   perf trace: Smatch: Fix potential NULL pointer dereference

> > >   perf hists: Smatch: Fix potential NULL pointer dereference

> > >   perf map: Smatch: Fix potential NULL pointer dereference

> > >   perf session: Smatch: Fix potential NULL pointer dereference

> > >   perf intel-bts: Smatch: Fix potential NULL pointer dereference

> > >   perf intel-pt: Smatch: Fix potential NULL pointer dereference

> > >   perf cs-etm: Smatch: Fix potential NULL pointer dereference

> > 

> > from quick look it all looks good to me, nice tool ;-)

> > 

> > Acked-by: Jiri Olsa <jolsa@kernel.org>

> 

> Thanks for reviewing, Jiri.

> 

> @Arnaldo, Just want to check, will you firstly pick up 01~05, 07,

> 08/11 patches if you think they are okay?  Or you want to wait me to

> spin new patch set with all patches after I gather all comments?


I'm picking up the uncontrovertial, will push to my perf/core branch,
continue from there, please.

- Arnaldo
 
> Thanks,

> Leo Yan

> 

> > >  tools/perf/builtin-report.c    |  4 ++--

> > >  tools/perf/builtin-stat.c      |  2 +-

> > >  tools/perf/builtin-top.c       |  8 ++++++--

> > >  tools/perf/builtin-trace.c     |  5 +++--

> > >  tools/perf/ui/browsers/hists.c | 13 +++++++++----

> > >  tools/perf/util/annotate.c     |  6 ++----

> > >  tools/perf/util/cs-etm.c       |  2 +-

> > >  tools/perf/util/intel-bts.c    |  5 ++---

> > >  tools/perf/util/intel-pt.c     |  5 ++---

> > >  tools/perf/util/map.c          |  7 +++++--

> > >  tools/perf/util/session.c      |  3 +++

> > >  11 files changed, 36 insertions(+), 24 deletions(-)

> > > 

> > > -- 

> > > 2.17.1

> > > 


-- 

- Arnaldo
Leo Yan July 4, 2019, 7:29 a.m. | #4
Hi Arnaldo,

On Wed, Jul 03, 2019 at 03:18:15PM -0300, Arnaldo Carvalho de Melo wrote:
> Em Wed, Jul 03, 2019 at 09:48:08AM +0800, Leo Yan escreveu:

> > On Tue, Jul 02, 2019 at 01:07:43PM +0200, Jiri Olsa wrote:

> > > On Tue, Jul 02, 2019 at 06:34:09PM +0800, Leo Yan wrote:

> > > > When I used static checker Smatch for perf building, the main target is

> > > > to check if there have any potential issues in Arm cs-etm code.  So

> > > > finally I get many reporting for errors/warnings.

> > > > 

> > > > I used below command for using static checker with perf building:

> > > > 

> > > >   # make VF=1 CORESIGHT=1 -C tools/perf/ \

> > > >     CHECK="/root/Work/smatch/smatch --full-path" \

> > > >     CC=/root/Work/smatch/cgcc | tee smatch_reports.txt

> > > > 

> > > > I reviewed the errors one by one, if I understood some of these errors

> > > > so changed the code as I can, this patch set is the working result; but

> > > > still leave some errors due to I don't know what's the best way to fix

> > > > it.  There also have many inconsistent indenting warnings.  So I firstly

> > > > send out this patch set and let's see what's the feedback from public

> > > > reviewing.

> > > > 

> > > > Leo Yan (11):

> > > >   perf report: Smatch: Fix potential NULL pointer dereference

> > > >   perf stat: Smatch: Fix use-after-freed pointer

> > > >   perf top: Smatch: Fix potential NULL pointer dereference

> > > >   perf annotate: Smatch: Fix dereferencing freed memory

> > > >   perf trace: Smatch: Fix potential NULL pointer dereference

> > > >   perf hists: Smatch: Fix potential NULL pointer dereference

> > > >   perf map: Smatch: Fix potential NULL pointer dereference

> > > >   perf session: Smatch: Fix potential NULL pointer dereference

> > > >   perf intel-bts: Smatch: Fix potential NULL pointer dereference

> > > >   perf intel-pt: Smatch: Fix potential NULL pointer dereference

> > > >   perf cs-etm: Smatch: Fix potential NULL pointer dereference

> > > 

> > > from quick look it all looks good to me, nice tool ;-)

> > > 

> > > Acked-by: Jiri Olsa <jolsa@kernel.org>

> > 

> > Thanks for reviewing, Jiri.

> > 

> > @Arnaldo, Just want to check, will you firstly pick up 01~05, 07,

> > 08/11 patches if you think they are okay?  Or you want to wait me to

> > spin new patch set with all patches after I gather all comments?

> 

> I'm picking up the uncontrovertial, will push to my perf/core branch,

> continue from there, please.


Will do it.  Thank you much for amendment in these patches.

Leo.