[Bug 463767] Review Request: cloog - The Chunky Loop Generator

bugzilla at redhat.com bugzilla at redhat.com
Tue Dec 16 16:13:29 UTC 2008


Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug.


https://bugzilla.redhat.com/show_bug.cgi?id=463767





--- Comment #13 from Mamoru Tasaka <mtasaka at ioa.s.u-tokyo.ac.jp>  2008-12-16 11:13:28 EDT ---
Well, would you change the release number every time you modify your
spec file?

(In reply to comment #12)
> cloog.x86_64: W: shared-lib-calls-exit /usr/lib64/libcloog.so.0.0.0
> exit at GLIBC_2.2.5

- This library actually calls exit() in the library (e.g.
  source/names.c). This is unusual situation. Usually when some unexpected
  behavior happens in a function in a library, the function should
  return a value which tells the error or so and should call
  exit() ($ rpmlint -I shared-lib-calls-exit shows the explanation).

  Would you contact upstream?

> cloog-devel.x86_64: W: no-documentation
- This warning can be ignored.
> I'd appreciate any help there.

> >? Some header files design flaw
> >  - Well, for example the head of %_includedir/cloog/cloog.h
> >    says:
> >----------------------------------------------------------
> >    40  #ifndef CLOOG_H
> >    41  #define CLOOG_H
> >    42  
> >    43  #ifdef CLOOG_PPL_BACKEND
> >    44  # define GNUMP
> >    45  # include<cloog/ppl_backend.h>
> >    46  #else
> >    47  # include <polylib/missing.h>
> >    48  # include<cloog/polylib_backend.h>
> >    49  #endif
> >----------------------------------------------------------
> >    However, where can we tell if CLOOG_PPL_BACKEND is defined
> >    or not (when this package was built) (i.e whether this header
> >    file includes ppl_backend.h or polylib_backend.h)?
> >    build.log shows that when rebuilding this package -DCLOOG_PPL_BACKEND
> >    is used, however installed header files does not save such
> >    information....
> 
> How can I fix this ? I mean this is an upstream problem. Do you mean I should
> append a patch to the package ?
> I could as well ship the package as is, and submit a patch upstream to fix it ?

- For this issue, I don't think this issue can be unresolved (I won't
  approve this package unless this is fixed). At least a patch should be
appended
  or so and also this must be fixed upstream.

-- 
Configure bugmail: https://bugzilla.redhat.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.




More information about the package-review mailing list