- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I used "ifort -w -g -save-temps -no_pie" to compile my program under Mac os 10.7.2. My intel fortran version is Compose XE 2011. This time idb can be started because of "-no_pie". However, under idb, when I "next/step in" my excutables step by step, it is frequently and unexpectedly interrupted by "Thread received signal ILL" or "Thread received signal BUS", or something like that. Interestingly, the excutables generated by ifort can run smoothly and achieve the correct results without any problems.So then I chose a suse linux system and still use intel fortran but 9.0 version, idb can work smoothly and never receives any signals like "ILL" or "BUS". So I guess it is the problem of Composer 2011 or Mac OS 10.7.Who can help me? Thanks in advance.Jianghui
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
"no_pie helps to load the executable in idb and I can set a break point or run it. However, I don't seem to be able to step it (step or next). It just hangs. When I try stepi or nexti, it doesn't hang, but I only seem to step to the location where I have my breakpoint. If I remove the breakpoint and continue it exits "
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
that's an interesting sighting. Do you see this behavior with simpler applications, you could provide, as well?
If yes, I'd like to ask you to provide the source + binary for which you see those problems. Our debugger engineers can then try to reproduce the problem. They did not hear about problems like that so far.
Thank you very much,
Georg Zitzlsberger
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
thank you!
I'll talk with the responsible engineer this afternoon. Let's hope we can reproduce it on our side. We'll start with the binary you're about to provide have provided.
Best regards,
Georg Zitzlsberger
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
with the binary you provided I was able to reproduce the same problem. However, if I compile it myself (with a slightly older compiler version) it does not appear. I'm currently checking the latest version and try narrow down the problem: I guess it's the combination of Xcode* & compiler that unveils this problem.
While I'm at it: which Xcode* version are you using?
Best regards,
Georg Zitzlsberger
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
after updating Xcode to 4.2.1 I'm not seeing the problem with your binary anymore.
To help me reconstruct your system configuration I'd like to ask you for the following:
i) Please extend fout.f90 by the following:
[fortran]program fout implicit none real*8,allocatable:: a(:,:) integer aerr allocate(a(20,20), stat=aerr) write(*,*) aerr call sps(a) write(*,*) a(1:9,1) call sps(a) write(*,*) a(1:9,1) call sps(a) write(*,*) a(1:9,1) call sps(a) write(*,*) a(1:9,1) deallocate(a) stop end[/fortran]Print-out of aerr should be 0 with normal execution.
ii) What's the output of:
ifort -w -g -save-temps -fpic fout.f90 sps.f90 -o fout -Wl,-no_pie -#
iii) If you start the debugger and repeat the steps you mentioned, what is the value of aerr (first print-out on console) and what's the result of "print a" (print-out of allocated array). The latter should print correct values and not "
Thank you very much,
Georg Zitzlsberger
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
/usr/bin/ifort-2011-base/bin/intel64/fortcom \ -D__INTEL_COMPILER=1210 \ -D__x86_64 \ -D__x86_64__ \ -D_MT \ -D__INTEL_COMPILER_BUILD_DATE=20111207 \ -D__PIC__ \ -D__APPLE__ \ -D__MACH__ \ -D__pentium4 \ -D__pentium4__ \ -D__tune_pentium4__ \ -D__SSE2__ \ -D__SSE3__ \ -D__SSSE3__ \ -D__SSE__ \ -D__MMX__ \ -mGLOB_pack_sort_init_list \ -I. \ -I/usr/bin/ifort-2011-base/compiler/include/intel64 \ -I/usr/bin/ifort-2011-base/compiler/include \ -I/usr/local/include \ -I/Developer/usr/llvm-gcc-4.2/lib/gcc/i686-apple-darwin11/4.2.1/include \ -I/usr/include \ "-fp_modbits honor_f32_conversion" \ "-fp_modbits honor_f64_conversion" \ -O0 \ -g2 \ "-warn none" \ -simd \ -mP1OPT_version=12.1-intel64 \ -mGLOB_diag_file=fout.diag \ -mGLOB_diag_use_message_catalog=FALSE \ -mGLOB_source_language=GLOB_SOURCE_LANGUAGE_F90 \ -mGLOB_tune_for_fort \ -mGLOB_use_fort_dope_vector \ -mP2OPT_static_promotion \ -mP1OPT_print_version=FALSE \ -mCG_use_gas_got_workaround=F \ -mP2OPT_align_option_used=TRUE \ -mGLOB_gcc_version=421 \ "-mGLOB_options_string=-w -g -save-temps -fpic -o fout -WI,-no_pie -#" \ -mGLOB_position_independent_code \ -mGLOB_preemption_model=1 \ -mGLOB_no_compact_unwind=TRUE \ -mGLOB_cxx_limited_range=FALSE \ -mP3OPT_emit_line_numbers \ -mP2OPT_suppress_unreferenced_locals=FALSE \ -mGLOB_debug_target=GLOB_DEBUG_TARGET_ALL \ -mGLOB_debug_format=GLOB_DEBUG_FORMAT_DWARF20 \ -mP3OPT_dwarf_format=P3OPT_DWARF_FORMAT_OBJ_ONLY \ -mGLOB_as_output_backup_file_name=fout_as_.s \ -mGLOB_machine_model=GLOB_MACHINE_MODEL_EFI2 \ -mGLOB_product_id_code=0x22006d8c \ -mGLOB_maintain_errno \ -mGLOB_noinline_intrinsics \ -mGLOB_use_base_pointer \ -mCG_bnl_movbe=T \ -mGLOB_extended_instructions=0x80 \ -mGLOB_advanced_optim=TRUE \ -mP3OPT_use_mspp_call_convention \ -mP2OPT_subs_out_of_bound=FALSE \ -mGLOB_ansi_alias \ -mIPOPT_ninl_limit=0 \ -mIPOPT_ninl_user_level=0 \ -mPGOPTI_value_profile_use=T \ -mP2OPT_gen_linux_const_pooling=FALSE \ -mP2OPT_il0_array_sections=TRUE \ -mGLOB_opt_level=0 \ -mIPOPT_args_in_regs=0 \ -mGLOB_imf_mapping_library=/usr/bin/ifort-2011-base/bin/intel64/libiml_attr.dylib \ -mIPOPT_obj_output_file_name=sps.o \ -mGLOB_long_size_64 \ -mGLOB_routine_pointer_size_64 \ -mGLOB_os_target=GLOB_OS_TARGET_MACH \ -mGLOB_async_unwind_tables=TRUE \ -mGLOB_obj_output_file=sps.o \ -mGLOB_source_dialect=GLOB_SOURCE_DIALECT_FORTRAN \ -mP1OPT_source_file_name=sps.f90 \ -mP2OPT_symtab_type_copy=true \ sps.f90
ld \ -lcrt1.10.6.o \ -dynamic \ -arch \ x86_64 \ -weak_reference_mismatches \ non-weak \ -macosx_version_min \ 10.7.2 \ -no_compact_unwind \ -o \ fout \ /usr/bin/ifort-2011-base/compiler/lib/for_main.o \ -L/usr/bin/ifort-2011-base/compiler/lib \ -L/usr/llvm-gcc-4.2/bin/../lib/gcc/i686-apple-darwin11/4.2.1/x86_64/ \ -L/Developer/usr/llvm-gcc-4.2/lib/gcc/i686-apple-darwin11/4.2.1/x86_64/ \ -L/usr/lib/ \ -L/usr/llvm-gcc-4.2/bin/../lib/gcc/i686-apple-darwin11/4.2.1/x86_64 \ -L/usr/llvm-gcc-4.2/bin/../lib/gcc/i686-apple-darwin11/4.2.1/ \ -L/usr/llvm-gcc-4.2/bin/../lib/gcc/ \ -L/Developer/usr/llvm-gcc-4.2/lib/gcc/i686-apple-darwin11/4.2.1/x86_64 \ -L/Developer/usr/llvm-gcc-4.2/lib/gcc/i686-apple-darwin11/4.2.1/ \ -L/usr/llvm-gcc-4.2/bin/../lib/gcc/i686-apple-darwin11/4.2.1/../../../ \ -L/Developer/usr/llvm-gcc-4.2/lib/gcc/i686-apple-darwin11/4.2.1/../../.. \ fout.o \ sps.o \ /usr/bin/ifort-2011-base/compiler/lib/libifport.a \ /usr/bin/ifort-2011-base/compiler/lib/libifcore.a \ /usr/bin/ifort-2011-base/compiler/lib/libimf.a \ /usr/bin/ifort-2011-base/compiler/lib/libsvml.a \ /usr/bin/ifort-2011-base/compiler/lib/libipgo.a \ -lgcc \ -lgcc_s.10.5 \ -lgcc \ -lSystem \ /usr/bin/ifort-2011-base/compiler/lib/libirc.a \ -lpthread \ -ldl
dsymutil \ fout
rm /var/folders/ds/q_w20grj6p7dv4nbrm37_79400010q/T//ifort8eLYE1libgcc
rm /var/folders/ds/q_w20grj6p7dv4nbrm37_79400010q/T//ifortkpd8Srgnudirs
rm /var/folders/ds/q_w20grj6p7dv4nbrm37_79400010q/T//ifortwbrZquos
rm /var/folders/ds/q_w20grj6p7dv4nbrm37_79400010q/T//ifortdDQZkrtos
rm /var/folders/ds/q_w20grj6p7dv4nbrm37_79400010q/T//ifortJEO0mDdashv
rm /var/folders/ds/q_w20grj6p7dv4nbrm37_79400010q/T//ifortH4ySgtgdb_v
rm /var/folders/ds/q_w20grj6p7dv4nbrm37_79400010q/T//ifortYSSueEarg
rm /var/folders/ds/q_w20grj6p7dv4nbrm37_79400010q/T//ifortpNMEthos
rm /var/folders/ds/q_w20grj6p7dv4nbrm37_79400010q/T//ifortxseZF8tos
rm /var/folders/ds/q_w20grj6p7dv4nbrm37_79400010q/T//ifortgZ7gWFgdb_v
rm /var/folders/ds/q_w20grj6p7dv4nbrm37_79400010q/T//ifortoxf7Zaarg
rm /var/folders/ds/q_w20grj6p7dv4nbrm37_79400010q/T//iforthI17YYgnudirs
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
after talking to our engineers it turned out that it's a bug in IDB. We've filed a defect ticket for it (DPD200223306). I'll let you know as soon as it is fixed.
Best regards,
Georg Zitzlsberger
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page