please do Sun On Mon, Mar 7, 2011 at 3:59 PM, Agarwal, Ajit <ajit.agar...@amd.com> wrote: > Hi Sun: > > The code changes for bug Id 723 and 739 also fine with you. > Can I go ahead with the checkin for bug 739 and 723 also. > > Thanks & Regards > Ajit > > -----Original Message----- > From: Sun Chan [mailto:sun.c...@gmail.com] > Sent: Monday, March 07, 2011 1:28 PM > To: Agarwal, Ajit > Cc: Jian-Xin Lai; Open64-devel@lists.sourceforge.net > Subject: Re: [Open64-devel] Code review for bug Id 732 > > ok. TARGET_SL is fine too, but it is immaterial > Sun > > On Mon, Mar 7, 2011 at 3:50 PM, Agarwal, Ajit <ajit.agar...@amd.com> wrote: >> Hi Sun: >> >> I am planning to put the code under x86_64 and along with the >> Fix for 739 and 723. >> >> Please let me know if its okay. Could you please review the >> Changes for bug 739 and 723. >> >> Thanks & Regards >> Ajit >> >> -----Original Message----- >> From: Sun Chan [mailto:sun.c...@gmail.com] >> Sent: Sunday, March 06, 2011 12:02 PM >> To: Jian-Xin Lai >> Cc: Agarwal, Ajit; Open64-devel@lists.sourceforge.net >> Subject: Re: [Open64-devel] Code review for bug Id 732 >> >> or screen out IA64? >> Sun >> >> On Sun, Mar 6, 2011 at 2:22 PM, Jian-Xin Lai <laij...@gmail.com> wrote: >>> This changes in wgen_expr.cxx caused a regression on IA-64. There will be an >>> assertion during CG expansion phase: >>> ### Compiler Error in file case1.cxx (user routine 'main') during >>> Code_Expansion phase: >>> ### Unimplemented function Expand_Expr: MLOAD/MSTORE >>> >>> I'd like to only apply this piece of code on X8664. Any suggestions? >>> >>> 2011/2/28 Agarwal, Ajit <ajit.agar...@amd.com> >>>> >>>> Hello All: >>>> >>>> >>>> >>>> Gatekeeper could you please review the fix for bug Id 732. The changes are >>>> made in wgen_expr.cxx >>>> >>>> And wn_lower.cxx. >>>> >>>> >>>> >>>> In wgen_expr.cxx the copy the -1 preg to a temp area is not done when for >>>> GS_CALL_EXPR and the >>>> >>>> Passed is GS_ADDR_EXPR with operand as GS_VAR_DECL. The MTYPE_M is not >>>> supported on wn_lower.cxx >>>> >>>> Added the changes for that. >>>> >>>> >>>> >>>> Following files are modified. >>>> >>>> Osprey/wgen/wgen_expr.cxx >>>> >>>> Osprey/be/com/wn_lower.cxx >>>> >>>> >>>> >>>> Index: osprey/wgen/wgen_expr.cxx >>>> >>>> =================================================================== >>>> >>>> --- osprey/wgen/wgen_expr.cxx (revision 3495) >>>> >>>> +++ osprey/wgen/wgen_expr.cxx (working copy) >>>> >>>> @@ -7899,6 +7899,7 @@ >>>> >>>> case GS_AGGR_INIT_EXPR: >>>> >>>> case GS_CALL_EXPR: >>>> >>>> { >>>> >>>> + BOOL is_gs_addr_arg = FALSE; >>>> >>>> gs_t arglist = gs_tree_operand (exp, 1); >>>> >>>> TYPE_ID ret_mtype; >>>> >>>> WN *call_wn; >>>> >>>> @@ -9840,11 +9841,24 @@ >>>> >>>> TY_mtype(TY_pointed(ty_idx)) == MTYPE_V) /* pointer to void >>>> */ >>>> >>>> ty_idx = nop_ty_idx; >>>> >>>> #endif >>>> >>>> + if((ret_mtype == MTYPE_M) && (Is_Target_64bit())) >>>> >>>> + { >>>> >>>> + for (list = gs_tree_operand (exp, 1); list; >>>> >>>> + list = gs_tree_chain (list)) { >>>> >>>> + if(gs_tree_code(gs_tree_value (list)) == GS_ADDR_EXPR) >>>> >>>> + { >>>> >>>> + if (gs_tree_code( >>>> >>>> + gs_tree_operand(gs_tree_value (list),0)) == >>>> GS_VAR_DECL ) >>>> >>>> + is_gs_addr_arg = TRUE; >>>> >>>> + } >>>> >>>> + } >>>> >>>> + } >>>> >>>> wn1 = WN_Ldid (ret_mtype, -1, Return_Val_Preg, ty_idx); >>>> >>>> >>>> >>>> - if (ret_mtype == MTYPE_M) { // copy the -1 preg to a temp area >>>> >>>> + if (ret_mtype == MTYPE_M && (!is_gs_addr_arg)) { // copy the -1 >>>> preg to a temp area >>>> >>>> >>>> >>>> TY_IDX ret_ty_idx = ty_idx; >>>> >>>> + is_gs_addr_arg = FALSE; >>>> >>>> #ifndef KEY >>>> >>>> // bug 3735: the compiler cannot arbitrarily change the alignment of >>>> >>>> // individual structures >>>> >>>> >>>> >>>> Index: osprey/be/com/wn_lower.cxx >>>> >>>> =================================================================== >>>> >>>> --- osprey/be/com/wn_lower.cxx (revision 3495) >>>> >>>> +++ osprey/be/com/wn_lower.cxx (working copy) >>>> >>>> @@ -4855,6 +4855,13 @@ >>>> >>>> #endif // TARG_X8664 || VECTOR_MTYPES >>>> >>>> >>>> >>>> case MTYPE_M: >>>> >>>> +#if defined(TARG_X8664) >>>> >>>> + WN_st_idx(tree) = ST_st_idx(MTYPE_To_PREG(MTYPE_U8)); >>>> >>>> + WN_load_offset(tree) = First_Int_Preg_Return_Offset; >>>> >>>> + WN_set_rtype(tree, MTYPE_U8); >>>> >>>> + WN_set_desc(tree, MTYPE_U8); >>>> >>>> + return tree; >>>> >>>> +#endif >>>> >>>> Fail_FmtAssertion ("MLDID of Return_Val_Preg not allowed in middle" >>>> >>>> " of expression"); >>>> >>>> /*NOTREACHED*/ >>>> >>>> @@ -11391,7 +11398,7 @@ >>>> >>>> : parm; >>>> >>>> >>>> >>>> ploc = Get_Output_Parameter_Location(TY_Of_Parameter(parm)); >>>> >>>> - if (MTYPE_is_m(parmType)) >>>> >>>> + if (MTYPE_is_m(parmType)&&(WN_operator(actual) != OPR_LDID)) >>>> >>>> { >>>> >>>> /* >>>> >>>> * structure parameter >>>> >>>> >>>> >>>> Thanks & Regards >>>> >>>> Ajit >>>> >>>> >>>> ------------------------------------------------------------------------------ >>>> Free Software Download: Index, Search & Analyze Logs and other IT data in >>>> Real-Time with Splunk. Collect, index and harness all the fast moving IT >>>> data >>>> generated by your applications, servers and devices whether physical, >>>> virtual >>>> or in the cloud. Deliver compliance at lower cost and gain new business >>>> insights. http://p.sf.net/sfu/splunk-dev2dev >>>> _______________________________________________ >>>> Open64-devel mailing list >>>> Open64-devel@lists.sourceforge.net >>>> https://lists.sourceforge.net/lists/listinfo/open64-devel >>>> >>> >>> >>> >>> -- >>> Regards, >>> Lai Jian-Xin >>> >>> ------------------------------------------------------------------------------ >>> What You Don't Know About Data Connectivity CAN Hurt You >>> This paper provides an overview of data connectivity, details >>> its effect on application quality, and explores various alternative >>> solutions. http://p.sf.net/sfu/progress-d2d >>> _______________________________________________ >>> Open64-devel mailing list >>> Open64-devel@lists.sourceforge.net >>> https://lists.sourceforge.net/lists/listinfo/open64-devel >>> >>> >> >> >> > > >
------------------------------------------------------------------------------ What You Don't Know About Data Connectivity CAN Hurt You This paper provides an overview of data connectivity, details its effect on application quality, and explores various alternative solutions. http://p.sf.net/sfu/progress-d2d _______________________________________________ Open64-devel mailing list Open64-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/open64-devel
