<pre style='margin:0'>
Blair Zajac (blair) pushed a commit to branch blair-py-tensorflow1-mac-10.13-cuda-10.1
in repository macports-ports.

</pre>
<p><a href="https://github.com/macports/macports-ports/commit/78c5dbbe72de56265007bce4f437fcb1522ccf8e">https://github.com/macports/macports-ports/commit/78c5dbbe72de56265007bce4f437fcb1522ccf8e</a></p>
<pre style="white-space: pre; background: #F8F8F8"><span style='display:block; white-space:pre;color:#808000;'>commit 78c5dbbe72de56265007bce4f437fcb1522ccf8e
</span>Author: Blair Zajac <blair@macports.org>
AuthorDate: Tue Aug 27 06:32:01 2019 -0700

<span style='display:block; white-space:pre;color:#404040;'>    blair-v1.14.0-compressed_tuple.h.diff: patch for compressed_tuple.h
</span><span style='display:block; white-space:pre;color:#404040;'>    
</span><span style='display:block; white-space:pre;color:#404040;'>    INFO: From Compiling tensorflow/core/kernels/population_count_op_gpu.cu.cc [for host]:
</span><span style='display:block; white-space:pre;color:#404040;'>    external/com_google_absl/absl/strings/string_view.h(495): warning: expression has no effect
</span><span style='display:block; white-space:pre;color:#404040;'>    
</span><span style='display:block; white-space:pre;color:#404040;'>    ./tensorflow/core/lib/core/status.h(33): warning: attribute "warn_unused_result" does not apply here
</span><span style='display:block; white-space:pre;color:#404040;'>    
</span><span style='display:block; white-space:pre;color:#404040;'>    external/com_google_absl/absl/strings/string_view.h(495): warning: expression has no effect
</span><span style='display:block; white-space:pre;color:#404040;'>    
</span><span style='display:block; white-space:pre;color:#404040;'>    ./tensorflow/core/lib/core/status.h(33): warning: attribute "warn_unused_result" does not apply here
</span><span style='display:block; white-space:pre;color:#404040;'>    
</span><span style='display:block; white-space:pre;color:#404040;'>    external/com_google_absl/absl/container/internal/compressed_tuple.h:170:53: error: use 'template' keyword to treat 'Storage' as a dependent template name
</span><span style='display:block; white-space:pre;color:#404040;'>    return (std::move(*this).internal_compressed_tuple::Storage< CompressedTuple, I> ::get());
</span><span style='display:block; white-space:pre;color:#404040;'>                                                        ^
</span><span style='display:block; white-space:pre;color:#404040;'>                                                        template
</span><span style='display:block; white-space:pre;color:#404040;'>    external/com_google_absl/absl/container/internal/compressed_tuple.h:176:54: error: use 'template' keyword to treat 'Storage' as a dependent template name
</span><span style='display:block; white-space:pre;color:#404040;'>    return (absl::move(*this).internal_compressed_tuple::Storage< CompressedTuple, I> ::get());
</span><span style='display:block; white-space:pre;color:#404040;'>                                                         ^
</span><span style='display:block; white-space:pre;color:#404040;'>                                                         template
</span><span style='display:block; white-space:pre;color:#404040;'>    2 errors generated.
</span><span style='display:block; white-space:pre;color:#404040;'>    ERROR: /opt/local/var/macports/build/_Users_blair_Code_MacPorts_macports-ports.git_python_py-tensorflow1/py37-tensorflow/work/tensorflow-1.14.0/tensorflow/core/kernels/BUILD:3565:1: output 'tensorflow/core/kernels/_objs/population_count_op_gpu/population_count_op_gpu.cu.pic.o' was not created
</span><span style='display:block; white-space:pre;color:#404040;'>    ERROR: /opt/local/var/macports/build/_Users_blair_Code_MacPorts_macports-ports.git_python_py-tensorflow1/py37-tensorflow/work/tensorflow-1.14.0/tensorflow/core/kernels/BUILD:3565:1: not all outputs were created or valid
</span>---
 python/py-tensorflow1/Portfile                         |  4 ++++
 .../files/blair-v1.14.0-compressed_tuple.h.diff        | 18 ++++++++++++++++++
 2 files changed, 22 insertions(+)

<span style='display:block; white-space:pre;color:#808080;'>diff --git a/python/py-tensorflow1/Portfile b/python/py-tensorflow1/Portfile
</span><span style='display:block; white-space:pre;color:#808080;'>index b1994a9..0b9d925 100644
</span><span style='display:block; white-space:pre;background:#e0e0ff;'>--- a/python/py-tensorflow1/Portfile
</span><span style='display:block; white-space:pre;background:#e0e0ff;'>+++ b/python/py-tensorflow1/Portfile
</span><span style='display:block; white-space:pre;background:#e0e0e0;'>@@ -186,6 +186,10 @@ if {${name} ne ${subport}} {
</span>         foreach d {/Developer/NVIDIA/CUDA-10.1/lib /usr/local/cuda/lib /opt/intel/cudnn-10.1-osx-x64-v7.6.3.30/lib} {
             set tf_bazel_build_opts "${tf_bazel_build_opts} --linkopt -Wl,-rpath,${d}"
         }
<span style='display:block; white-space:pre;background:#e0ffe0;'>+        # To unblock the person building the package as quickly as
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+        # possible, build the target that gets a compile failure due
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+        # to an unpatched compressed_tuple.h.
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+        system -W ${worksrcpath} "${tf_bazel_cmd} build ${tf_bazel_build_opts} //tensorflow/core/kernels:fill_functor"
</span>         system -W ${worksrcpath} "${tf_bazel_cmd} build ${tf_bazel_build_opts} //tensorflow/tools/pip_package:build_pip_package"
         # Build the python wheel
         system -W ${worksrcpath} "./bazel-bin/tensorflow/tools/pip_package/build_pip_package ${workpath}"
<span style='display:block; white-space:pre;color:#808080;'>diff --git a/python/py-tensorflow1/files/blair-v1.14.0-compressed_tuple.h.diff b/python/py-tensorflow1/files/blair-v1.14.0-compressed_tuple.h.diff
</span>new file mode 100644
<span style='display:block; white-space:pre;color:#808080;'>index 0000000..30a66b2
</span><span style='display:block; white-space:pre;background:#ffe0e0;'>--- /dev/null
</span><span style='display:block; white-space:pre;background:#e0e0ff;'>+++ b/python/py-tensorflow1/files/blair-v1.14.0-compressed_tuple.h.diff
</span><span style='display:block; white-space:pre;background:#e0e0e0;'>@@ -0,0 +1,18 @@
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+--- compressed_tuple.h.FCS 2019-08-26 20:40:35.000000000 -0700
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>++++ compressed_tuple.h     2019-08-26 20:41:30.000000000 -0700
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+@@ -165,6 +165,7 @@
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+     return internal_compressed_tuple::Storage<CompressedTuple, I>::get();
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+   }
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+ 
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>++#if 0
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+   template <int I>
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+   ElemT<I>&& get() && {
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+     return std::move(*this)
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+@@ -176,6 +177,7 @@
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+     return absl::move(*this)
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+         .internal_compressed_tuple::template Storage<CompressedTuple, I>::get();
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+   }
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>++#endif /* 0 */
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+ };
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+ 
</span><span style='display:block; white-space:pre;background:#e0ffe0;'>+ // Explicit specialization for a zero-element tuple
</span></pre><pre style='margin:0'>

</pre>