[openib-general] RE: [PATCH] OpenSM - complib fix for branch

Hal Rosenstock halr at voltaire.com
Thu Apr 6 06:18:26 PDT 2006


Hi Yael,

On Thu, 2006-04-06 at 05:01, Yael Kalka wrote:
> Hi Hal,
> 
> > -----Original Message-----
> > From: Hal Rosenstock [mailto:halr at voltaire.com]
> > Sent: Wednesday, April 05, 2006 1:59 PM
> > To: Yael Kalka
> > Cc: openib-general at openib.org; Eitan Zahavi; Sasha Khapyorsky; Ofer
> Gigi
> > Subject: Re: [PATCH] OpenSM - complib fix for branch
> > 
> > Hi Yael,
> > 
> > On Wed, 2006-04-05 at 02:24, Yael Kalka wrote:
> > > Hi Hal,
> > >
> > > I saw that the complib patch (removal of constructor and destructor
> > > attribute), wasn't fully added to the branch.
> > > Attached is a patch for the branch.
> > 
> > Is this needed for 1.0 ? Is this safe to add ? Was there more to it
> than
> > just this ?
> > 
> [Yael Kalka] It isn't needed for 1.0 more than for the trunk, just that
> the minimum differences we have between the branch and the trunk, the
> easier it'll be to handle them.
> I don't see a problem with applying the patch on the branch. 
> The patch also included a change to the osmtest/main.c, but that change
> is already applied to the branch for some reason.

Not sure how it got in there. Anyhow, I made the requested change to the
1.0 branch (r6282).

> Another issue that is not connected to this specific patch - I think we
> should apply the cosmetic changes patches to the branch in order to
> minimize the differences between the trunk and the branch.

That has not been a priority but I agree that this would make things
easier (in terms of a simple diff). I'll work on this in the background
as I have time. As always, you're also welcome to submit patches.

-- Hal

> Yael
> 
> > -- Hal
> > 
> > > Thanks,
> > > Yael
> > >
> > > Signed-off-by:  Yael Kalka <yael at mellanox.co.il>
> > >
> > > Index: complib/cl_complib.c
> > > ===================================================================
> > > --- complib/cl_complib.c	(revision 6203)
> > > +++ complib/cl_complib.c	(working copy)
> > > @@ -65,7 +65,6 @@ __cl_timer_prov_destroy( void );
> > >  cl_spinlock_t cl_atomic_spinlock;
> > >
> > >  void
> > > -__attribute (( constructor ))
> > >  complib_init(void)
> > >  {
> > >  	cl_status_t	status = CL_SUCCESS;
> > > @@ -90,14 +89,6 @@ complib_init(void)
> > >  }
> > >
> > >  void
> > > -__attribute (( destructor ))
> > > -complib_fini(void)
> > > -{
> > > -  __cl_timer_prov_destroy();
> > > -  __cl_user_syshelper_exit();
> > > -}
> > > -
> > > -void
> > >  complib_exit(void)
> > >  {
> > >    __cl_timer_prov_destroy();
> > > Index: opensm/main.c
> > > ===================================================================
> > > --- opensm/main.c	(revision 6203)
> > > +++ opensm/main.c	(working copy)
> > > @@ -44,9 +44,6 @@
> > >   *
> > >   * $Revision: 1.23 $
> > >   */
> > > -#ifdef __WIN__
> > > -#pragma warning(disable : 4996)
> > > -#endif
> > >
> > >  #if HAVE_CONFIG_H
> > >  #  include <config.h>
> > > @@ -557,9 +554,7 @@ main(
> > >        {  NULL,            0, NULL,  0 }  /* Required at the end of
> the
> > array */
> > >      };
> > >
> > > -#ifdef __WIN__
> > >  	complib_init();
> > > -#endif
> > >
> > >    /* Make sure that the opensm and complib were compiled using
> > >       same modes (debug/free) */
> > >
> > 
> 




More information about the general mailing list