Git Product home page Git Product logo

Comments (3)

yongtang avatar yongtang commented on August 9, 2024

When trying tensorflow-io's build with TF 1.13, I noticed that the generated binary for 1.13 may not be compatible with TF 2.0. Here is some core dump:

(gdb) where
#0  __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
#1  0x00007fd30249c801 in __GI_abort () at abort.c:79
#2  0x00007fd3024e5897 in __libc_message (action=action@entry=do_abort, fmt=fmt@entry=0x7fd302612b9a "%s\n") at ../sysdeps/posix/libc_fatal.c:181
#3  0x00007fd3024ec90a in malloc_printerr (str=str@entry=0x7fd302610d88 "free(): invalid pointer") at malloc.c:5350
#4  0x00007fd3024f3e1c in _int_free (have_lock=0, p=0x25f6798, av=0x7fd302847c40 <main_arena>) at malloc.c:4157
#5  __GI___libc_free (mem=0x25f67a8) at malloc.c:3124
#6  0x00007fd2e127b8d8 in __gnu_cxx::new_allocator<std::string>::deallocate(std::string*, unsigned long) ()
   from /usr/local/lib/python3.6/dist-packages/tensorflow_io/image/python/ops/_image_ops.so
#7  0x00007fd2e1279c28 in std::_Vector_base<std::string, std::allocator<std::string> >::_M_deallocate(std::string*, unsigned long) ()
   from /usr/local/lib/python3.6/dist-packages/tensorflow_io/image/python/ops/_image_ops.so
#8  0x00007fd2e12789fd in std::_Vector_base<std::string, std::allocator<std::string> >::~_Vector_base() ()
   from /usr/local/lib/python3.6/dist-packages/tensorflow_io/image/python/ops/_image_ops.so
#9  0x00007fd2e12765e7 in std::vector<std::string, std::allocator<std::string> >::~vector() ()
   from /usr/local/lib/python3.6/dist-packages/tensorflow_io/image/python/ops/_image_ops.so
#10 0x00007fd2e1280d32 in tensorflow::OpDefBuilder::~OpDefBuilder() ()
   from /usr/local/lib/python3.6/dist-packages/tensorflow_io/image/python/ops/_image_ops.so
#11 0x00007fd2e1280f68 in tensorflow::register_op::OpDefBuilderWrapper<true>::~OpDefBuilderWrapper() ()
   from /usr/local/lib/python3.6/dist-packages/tensorflow_io/image/python/ops/_image_ops.so
#12 0x00007fd2e128082a in __static_initialization_and_destruction_0 ()
   from /usr/local/lib/python3.6/dist-packages/tensorflow_io/image/python/ops/_image_ops.so
#13 0x00007fd2e1280a88 in _GLOBAL__sub_I_dataset_ops.cc () from /usr/local/lib/python3.6/dist-packages/tensorflow_io/image/python/ops/_image_ops.so
#14 0x00007fd30285d733 in call_init (env=0x1c71a30, argv=0x7ffe8a9b6188, argc=1, l=<optimized out>) at dl-init.c:72
#15 _dl_init (main_map=main_map@entry=0x201fc00, argc=1, argv=0x7ffe8a9b6188, env=0x1c71a30) at dl-init.c:119
#16 0x00007fd3028621ff in dl_open_worker (a=a@entry=0x7ffe8a9b3340) at dl-open.c:522
#17 0x00007fd3025c32df in __GI__dl_catch_exception (exception=exception@entry=0x7ffe8a9b3320, operate=operate@entry=0x7fd302861dc0 <dl_open_worker>, 
    args=args@entry=0x7ffe8a9b3340) at dl-error-skeleton.c:196
#18 0x00007fd3028617ca in _dl_open (file=0x3a4b840 "/usr/local/lib/python3.6/dist-packages/tensorflow_io/image/python/ops/_image_ops.so", 
    mode=-2147483646, caller_dlopen=0x7fd2e9a1728a <tensorflow::internal::LoadLibrary(char const*, void**)+26>, nsid=<optimized out>, argc=1, 
---Type <return> to continue, or q <return> to quit--- 

from io.

yongtang avatar yongtang commented on August 9, 2024

Likely we may have to rebuild tensorflow-io with TF 1.13 and 2.0 separately and release separate versions for them.

from io.

yongtang avatar yongtang commented on August 9, 2024

With #106 merged, we have the TF 2.0 supports now. The R still has some issues but it is likely caused by upstream R package (tfdatasets of tensorflow). We could follow up to address those issues later.

from io.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.