May 2010 - Ad-dev - mailman3.cels.anl.gov

[openAD] #200: introduce scope per CFG node
by OpenAD 18 Nov '14

18 Nov '14

#200: introduce scope per CFG node -------------------------+-------------------------------------------------- Reporter: utke | Owner: snarayan Type: enhancement | Status: new Priority: major | Milestone: Component: xaif | Keywords: -------------------------+-------------------------------------------------- for C++, with default scope being the same scope as the top level cfg scope for C/fortran -- Ticket URL: <https://trac.mcs.anl.gov/projects/openAD/ticket/200> OpenAD <http://www.mcs.anl.gov/openad> Bug tracking for OpenAD components

1 5

[openAD] #213: postProcess.py needs to add value selector (and do nothing else) for original function definitions that have been turned into subroutines
by OpenAD 31 Mar '11

31 Mar '11

#213: postProcess.py needs to add value selector (and do nothing else) for original function definitions that have been turned into subroutines ------------------------------+--------------------------------------------- Reporter: lyonsam | Owner: utke Type: defect | Status: new Priority: major | Milestone: scale1 Component: SourceProcessing | Keywords: ------------------------------+--------------------------------------------- The issue is this: you know how we have postprocess and transformfile? I have a situation where there's a function in centrm (in centrm/psget_I.f90) that gets subroutinized by the preprocessor, so that both the function and the new subroutine go through the tool. As is usually the case when we do this, the subroutine will have active statements whereas the function will not (because we've replaced all the function calls that we've seen with subroutine calls, so the function isn't used in an active context). So far so good. In this case, though, the function uses a module variable that happens to be active, like {{{ passive1 = active*passive2 }}} where we would need {{{ passive1 = active%v*passive2 }}} So here's what we need: for function units, the postprocessor should do exactly what transformfile does (like adding the %v whenever an active variable is referenced), and none of the stuff that the postprocessor does on non-function units. -- Ticket URL: <http://trac.mcs.anl.gov/projects/openAD/ticket/213> OpenAD <http://www.mcs.anl.gov/openad> Bug tracking for OpenAD components

1 2

[openAD] #202: insufficient parsing of rewind statements
by OpenAD 09 Sep '10

09 Sep '10

#202: insufficient parsing of rewind statements ------------------------------+--------------------------------------------- Reporter: lyonsam | Owner: lyonsam Type: defect | Status: new Priority: minor | Milestone: Component: SourceProcessing | Keywords: ------------------------------+--------------------------------------------- occuring in centrm/pxlib_read_M.f90 (we were able to get around this for now because we dont really need to transform this file): rewind ['(', 'linpxs', ')'] -- Ticket URL: <https://trac.mcs.anl.gov/projects/openAD/ticket/202> OpenAD <http://www.mcs.anl.gov/openad> Bug tracking for OpenAD components

1 4

[openAD] #212: option --removeFunction fails in preProcess.py
by OpenAD 27 May '10

27 May '10

#212: option --removeFunction fails in preProcess.py ------------------------------+--------------------------------------------- Reporter: lyonsam | Owner: utke Type: defect | Status: new Priority: major | Milestone: scale1 Component: SourceProcessing | Keywords: ------------------------------+--------------------------------------------- This flag is supposed to discard function definitions that are subroutinized, leaving only the new subroutine. The problem reveals itself when you try this flag on SCALE: {{{ lyonsam@churn:/sandbox/lyonsam/hgBranch/vecOp_Scale/Scale/scale6-new/src/mods/ [15:43:10]$ make # ================================================= PREPROCESS ================================================= /sandbox/lyonsam/hgBranch/vecOp_Scale/OpenAD/OpenADFortTk/tools/SourceProcessing/preProcess.py --timing --inputFormat=free --outputFormat=free --removeFunction -o numCore1.pre.f90 scalelib/Vast_kind_param_M.f90 scalelib/wot8_I.f90 scalelib/ki3_I.f90 scalelib/cmps_I.f90 scalelib/jll1_I.f90 scalelib/mwlist_I.f90 scalelib/e3_I.f90 scalelib/e3.f90 scalelib/erf_M.f90 scalelib/shelc_I.f90 scalelib/shelc.f90 scalelibDummies/Vtime_M.f90 scalelibDummies/common_unit_C.f90 scalelibDummies/erro_I.f90 scalelibDummies/Error_functions_M.f90 scalelibDummies/direct_access_M.f90 scalelibDummies/fido_M.f90 scalelibDummies/fill_M.f90 scalelibDummies/ffpun_I.f90 scalelibDummies/io_I.f90 scalelibDummies/mgcwrd_M.f90 scalelibDummies/separator_character_M.f90 scalelibDummies/opnfil_I.f90 centrmDummies/pxretrv_I.f90 centrmDummies/pxlib_read_M.f90 centrmDummies/calcrng_I.f90 centrmDummies/xtract_I.f90 centrm/data_centrm_C.f90 centrm/pesc_I.f90 centrm/asym_flx_I.f90 centrm/balanc_C.f90 centrm/dflt_C.f90 centrm/drtacs_centrm_C.f90 centrm/inels2_C.f90 centrm/prbttl_C.f90 centrm/pxdata_C.f90 centrm/pxparm_C.f90 centrm/sources_C.f90 centrm/terp_M.f90 centrm/thermal_C.f90 centrm/tots_C.f90 centrm/units_C.f90 centrm/rebaln_I.f90 centrm/accel_I.f90 centrm/cell_I.f90 centrm/clean_I.f90 centrm/convrg_I.f90 centrm/cumintn_I.f90 centrm/difcof_I.f90 centrm/drevers_I.f90 centrm/dt_I.f90 centrm/fitter_I.f90 centrm/gcoef_M.f90 centrm/getdfc_I.f90 centrm/getins_I.f90 centrm/h_factors_M.f90 centrm/w_factors_M.f90 centrm/inelas_I.f90 centrm/inner_I.f90 centrm/revers_I.f90 centrm/linrzq_I.f90 centrm/scatsc_I.f90 centrm/mgsrc_I.f90 centrm/mgsum_I.f90 centrm/nchi_I.f90 centrm/nmspt_I.f90 centrm/postn_I.f90 centrm/prepout_I.f90 centrm/vintn_I.f90 centrm/vintni_I.f90 centrm/pxbs_I.f90 centrm/prepxc_I.f90 centrm/psget_I.f90 centrm/pxcon1_I.f90 centrm/pxcon2_I.f90 centrm/pxsec_I.f90 centrm/pxseci_I.f90 centrm/pxsormt_I.f90 centrm/pxtsor_I.f90 centrm/initial_I.f90 centrm/firstp_I.f90 centrm/range_M.f90 centrm/flxguess_I.f90 centrm/rittemp_I.f90 centrm/setcons_I.f90 centrm/setmbyz_I.f90 centrm/storxs_I.f90 centrm/sumbal_I.f90 centrm/thinit_I.f90 centrm/thtolmr_I.f90 centrm/thermit_I.f90 centrm/update_I.f90 centrm/clear_dflt_I.f90 centrm/evap_I.f90 centrm/pxsorm_I.f90 centrm/pxcal_I.f90 centrm/pxtomg_I.f90 centrm/epitoth_I.f90 centrm/fixsrc_I.f90 centrm/geomf_I.f90 centrm/zhomo_I.f90 centrm/solver_I.f90 centrm/pxbound_I.f90 centrm/pxget.f90 centrm/pxzas_I.f90 centrm/zmsh_I.f90 centrm/pw_chi_I.f90 centrm/lumper_I.f90 centrm/linriz_I.f90 centrm/pxpts_I.f90 centrm/pxread_I.f90 centrm/free_gas_M.f90 centrm/pxsset_I.f90 centrm/united_I.f90 centrm/emsh_I.f90 centrm/calc_I.f90 centrm/pxarr_I.f90 centrm/mix_centrm_I.f90 centrm/getmom_I.f90 centrmDummies/prtf_I.f90 centrmDummies/wrtf_I.f90 centrm/mgout_I.f90 centrmDummies/redf_I.f90 centrmDummies/quadrature_M.f90 centrmDummies/pncnst_I.f90 centrmDummies/mgarr_I.f90 centrmDummies/drtran_I.f90 centrmDummies/contrl_I.f90 Traceback (most recent call last): File "/sandbox/lyonsam/hgBranch/vecOp_Scale/OpenAD/OpenADFortTk/tools/SourceProcessing/preProcess.py", line 293, in <module> sys.exit(main()) File "/sandbox/lyonsam/hgBranch/vecOp_Scale/OpenAD/OpenADFortTk/tools/SourceProcessing/preProcess.py", line 220, in main UnitCanonicalizer(aUnit).canonicalizeUnit().printit(out) File "/sandbox/lyonsam/hgBranch/vecOp_Scale/OpenAD/OpenADFortTk/tools/SourceProcessing/Canon/canon.py", line 694, in canonicalizeUnit newUnit = UnitCanonicalizer(subUnit,self.__SRmoduleUsed).canonicalizeUnit() File "/sandbox/lyonsam/hgBranch/vecOp_Scale/OpenAD/OpenADFortTk/tools/SourceProcessing/Canon/canon.py", line 724, in canonicalizeUnit self.__myUnit = self.__createNewSubroutine(self.__myUnit,subroutineDecls) File "/sandbox/lyonsam/hgBranch/vecOp_Scale/OpenAD/OpenADFortTk/tools/SourceProcessing/Canon/canon.py", line 640, in __createNewSubroutine aUnit,self.__myNewExecs,declList,self._keepFunctionDecl) NameError: global name 'declList' is not defined make: *** [numCore1.pre.f90] Error 1 }}} -- Ticket URL: <https://trac.mcs.anl.gov/projects/openAD/ticket/212> OpenAD <http://www.mcs.anl.gov/openad> Bug tracking for OpenAD components

1 2

[openAD] #219: in typechange : test for overlap of typechanged arguments with others (f77 style calls)
by OpenAD 25 May '10

25 May '10

#219: in typechange : test for overlap of typechanged arguments with others (f77 style calls) -------------------------+-------------------------------------------------- Reporter: utke | Owner: utke Type: defect | Status: new Priority: major | Milestone: Component: xaifBooster | Keywords: -------------------------+-------------------------------------------------- test case illustrates potential overlap between typechanged and not typechanged arguments when full array copies are made for f77 style calls leading to potential mutual overwrites. -- Ticket URL: <https://trac.mcs.anl.gov/projects/openAD/ticket/219> OpenAD <http://www.mcs.anl.gov/openad> Bug tracking for OpenAD components

1 2

[openAD] #218: full centrm activation
by OpenAD 25 May '10

25 May '10

#218: full centrm activation ---------------------------------+------------------------------------------ Reporter: utke | Owner: Type: defect | Status: new Priority: major | Milestone: Component: Application - SCALE | Keywords: ---------------------------------+------------------------------------------ The true dependent variable is 'pxj' but it is currently not declared dependent because of output discrepancies that are obderved in the testcases: centrm-pwr Instead currently a number of the variables that are conveyed via file i/o between the two centrm phases are marked as dependent. These variables are !$openad DEPENDENT(bkern) !$openad DEPENDENT(coh_kern) !$openad DEPENDENT(pxt) !$openad DEPENDENT(pxabs) !$openad DEPENDENT(pxss) !$openad DEPENDENT(pxis) !$openad DEPENDENT(pddxs) !$openad DEPENDENT(coh_xs) !$openad DEPENDENT(crx) (see src/mods/centrmDummies/contrl_I.f90 ) and compare to the output of scale5.1/centrmst> grep read_scratch *.f90 'crx' has been added upon possibly incorrect recommendation from scale development. If we add pxj and remove the above DEPENDENT scratch variables then a subset of the scratch variables becomes active. Therefore to retain dependency we have to retain some scratch vars as (IN)DEPENDENT. -- Ticket URL: <https://trac.mcs.anl.gov/projects/openAD/ticket/218> OpenAD <http://www.mcs.anl.gov/openad> Bug tracking for OpenAD components

1 2

[openAD] #217: example for self adjoint call template
by OpenAD 24 May '10

24 May '10

#217: example for self adjoint call template ---------------------------------+------------------------------------------ Reporter: utke | Owner: utke Type: task | Status: new Priority: major | Milestone: Component: multiple components | Keywords: ---------------------------------+------------------------------------------ -- Ticket URL: <https://trac.mcs.anl.gov/projects/openAD/ticket/217> OpenAD <http://www.mcs.anl.gov/openad> Bug tracking for OpenAD components

1 1

[openAD] #216: MacOS build
by OpenAD 21 May '10

21 May '10

#216: MacOS build ---------------------------------+------------------------------------------ Reporter: utke | Owner: utke Type: enhancement | Status: new Priority: major | Milestone: Component: multiple components | Keywords: ---------------------------------+------------------------------------------ -- Ticket URL: <https://trac.mcs.anl.gov/projects/openAD/ticket/216> OpenAD <http://www.mcs.anl.gov/openad> Bug tracking for OpenAD components

1 14

[ADIC] #39: build fails with gcc version 4.2.4 (Ubuntu 4.2.4-1ubuntu4)
by ADIC 17 May '10

17 May '10

#39: build fails with gcc version 4.2.4 (Ubuntu 4.2.4-1ubuntu4) --------------------------+------------------------------------------------- Reporter: utke | Owner: narayanan Type: defect | Status: new Priority: major | Milestone: Component: OpenAnalysis | Version: Keywords: | --------------------------+------------------------------------------------- gives link error in ADIC: libtool: link: g++ -shared -nostdlib /usr/lib/gcc/x86_64-linux- gnu/4.2.4/../../../../lib/crti.o /usr/lib/gcc/x86_64-linux- gnu/4.2.4/crtbeginS.o .libs/liboarose_la-Sage2OA.o .libs/liboarose_la- SageOACallGraph.o .libs/liboarose_la-common.o .libs/liboarose_la- MemSage2OA.o .libs/liboarose_la-OATest.o -Wl,-rpath -Wl,/disks/utke/Apps/rose_inst/lib -Wl,-rpath -Wl,/disks/utke/Apps/rose_inst/lib -L/disks/utke/Argonne/Dev/RoseFE/OpenAD/OpenAnalysis/x86_64-Linux/lib -lOAul -L/disks/utke/Apps/rose_inst/lib /disks/utke/Apps/rose_inst/lib/librose.so -L/usr/lib/jvm/java-6-sun-1.6.0.17/jre/lib/amd64/server -L/home/derivs/software/linux-Ubuntu_8.04-x86_64/boost_1_36_0/lib -L/disks/utke/Apps/rose_build/src/3rdPartyLibraries/libharu-2.1.0/src -ljvm -ldl -lboost_date_time-gcc41-mt-1_36 -lboost_thread-gcc41-mt-1_36 -lboost_filesystem-gcc41-mt-1_36 -lboost_program_options-gcc41-mt-1_36 -lboost_regex-gcc41-mt-1_36 -lboost_system-gcc41-mt-1_36 -lboost_wave- gcc41-mt-1_36 /disks/utke/Apps/rose_inst/lib/libhpdf.so -lpthread -lrt -L/usr/lib/gcc/x86_64-linux-gnu/4.2.4 -L/usr/lib/gcc/x86_64-linux- gnu/4.2.4/../../../../lib -L/lib/../lib -L/usr/lib/../lib -L/usr/lib/gcc/x86_64-linux-gnu/4.2.4/../../.. -lstdc++ -lm -lc -lgcc_s /usr/lib/gcc/x86_64-linux-gnu/4.2.4/crtendS.o /usr/lib/gcc/x86_64-linux- gnu/4.2.4/../../../../lib/crtn.o -Wl,-soname -Wl,liboarose.so.0 -o .libs/liboarose.so.0.0.0 /usr/bin/ld: /disks/utke/Argonne/Dev/RoseFE/OpenAD/OpenAnalysis/x86_64-Linux/lib/libOAul.a (libOAul_a-Annotation.o): relocation R_X86_64_32 against `a local symbol' can not be used when making a shared object; recompile with -fPIC But cannot compile OA with -fPIC because that leads to an ICE there. This is just a placeholder for the issue. -- Ticket URL: <http://trac.mcs.anl.gov/projects/ADIC/ticket/39> ADIC <http://trac.mcs.anl.gov/projects/ADIC> ADIC: Automatic Differentiation for C and C++

1 0

[openAD] #215: ANYis missing
by OpenAD 10 May '10

10 May '10

#215: ANYis missing --------------------------+------------------------------------------------- Reporter: utke | Owner: utke Type: defect | Status: new Priority: major | Milestone: Component: OpenADFortTk | Keywords: --------------------------+------------------------------------------------- From: Jonathan Mc Govern <mcgovej(a)gmail.com> .... call to function: ANY near line 502 is not supported! This should either be recognized as an intrinsic or should have been canonicalized into a subroutine call -- Ticket URL: <https://trac.mcs.anl.gov/projects/openAD/ticket/215> OpenAD <http://www.mcs.anl.gov/openad> Bug tracking for OpenAD components

1 4