Mei, I think you are right. The assertion about RID is a hint of something else wrong (RID stands for region id). Sun
On Sat, Aug 28, 2010 at 12:36 AM, Ye, Mei <mei...@amd.com> wrote: > My guess is that by adding a label statement you may have prevented the > elimination of an empty block or something similar which then preserves the > label-BB map and thus avoids the assertion. > > > > -Mei > > > > ________________________________ > > From: Ye, Mei [mailto:mei...@amd.com] > Sent: Friday, August 27, 2010 9:25 AM > To: ruifen Shen; open64-devel > Subject: Re: [Open64-devel] One bug in add_one_if_stmt, Can gatekeeper > review this patch? Thx very much > > > > Hello Ruifen > > > > I am not convinced that a label statement is needed here. > > It is likely that somewhere downstream after some sort of transformations a > different bug misses a mapping between a label and a BB which then triggers > the assertion. > > > > -Mei > > > > ________________________________ > > From: ruifen Shen [mailto:shenr...@gmail.com] > Sent: Friday, August 27, 2010 2:36 AM > To: open64-devel > Subject: [Open64-devel] One bug in add_one_if_stmt, Can gatekeeper review > this patch? Thx very much > > > > hi, all. > > > > We found One bug in add_one_if_stmt (opt_cfg.cxx). Can gatekeeper give a > reivew? Thanks very much. > > > > When create the else_bb, it just create label_map without label_stmt. > > Then in the later phase. if there is check_label for BB(such as dce, > reconstruction-cfg), the label_stmt will be added. > > Othere wise, the label will be missed at cg phase. Assertion will be > reported like followed. > > > > ### Assertion failure at line 5392 of > /home/test/regression/open64-SL/osprey/be/cg/whirl2ops.cxx: > ### Compiler Error in file a.c during Code_Expansion phase: > ### RID == NULL, label 2050 doesn't have a matching target > > > > I would like to fix the bug as followed. > > > > Index: opt_cfg.cxx > =================================================================== > --- opt_cfg.cxx (revision 3323) > +++ opt_cfg.cxx (working copy) > @@ -2298,8 +2298,7 @@ > CFG::Add_one_if_stmt( WN *wn, END_BLOCK *ends_bb ) > { > // create, but do not connect, the "else" block > - BB_NODE *else_bb = Create_bb(); > - Append_label_map(Alloc_label(), else_bb); > + BB_NODE *else_bb = Create_labelled_bb(); > > // create if bb > WN *end_cond = WN_CreateFalsebr(else_bb->Labnam(), WN_if_test(wn)); > > > > -- > Best Regards. > > Shen Ruifen > > tel: 010-51266989-226 > > ------------------------------------------------------------------------------ > Sell apps to millions through the Intel(R) Atom(Tm) Developer Program > Be part of this innovative community and reach millions of netbook users > worldwide. Take advantage of special opportunities to increase revenue and > speed time-to-market. Join now, and jumpstart your future. > http://p.sf.net/sfu/intel-atom-d2d > _______________________________________________ > Open64-devel mailing list > Open64-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/open64-devel > > ------------------------------------------------------------------------------ Sell apps to millions through the Intel(R) Atom(Tm) Developer Program Be part of this innovative community and reach millions of netbook users worldwide. Take advantage of special opportunities to increase revenue and speed time-to-market. Join now, and jumpstart your future. http://p.sf.net/sfu/intel-atom-d2d _______________________________________________ Open64-devel mailing list Open64-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/open64-devel
