python.pkgs.tensorflow: wheel version #41259
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
GrahamcOfBorg
added
6.topic: python
FRidh
requested changes
May 30, 2018
| @@ -192,6 +192,13 @@ in { | |||
| gcc = gcc5; | |||
| }; | |||
|
|
|||
| cudatoolkit90 = common { | |||
There was a problem hiding this comment.
cudatoolkit90 while we have cudatoolkit9 is very confusing.
Would you mind rewriting these all to cudatoolkit_X_X or cudatoolkit_X in a separate commit, which will form the base for this change?
| }; | ||
| src = let dls = import ./tf1.7.1-hashes.nix; | ||
| in | ||
| fetchurl ( |
There was a problem hiding this comment.
That's ugly. Maybe the following?
let
version = python.majorVersion if stdenv.isDarwin else "${python.majorVersion}{python.minorVersion}";
platform = "mac" if stdenv.isDarwin else "linux";
unit = "gpu" if cudaSupport else "cpu";
key = "${platform}_py_${version}_${unit}";
in dls.${key}We don't support Python 3.4 and 3.5 so I think this should work.
| # patchelf --shrink-rpath will remove the cuda libraries. | ||
| postFixup = let | ||
| rpath = stdenv.lib.makeLibraryPath | ||
| (if cudaSupport then |
There was a problem hiding this comment.
use lib.optionals for the extra items.
| '' | ||
| rrPath="$out/${python.sitePackages}/tensorflow/:${rpath}" | ||
| internalLibPath="$out/${python.sitePackages}/tensorflow/python/_pywrap_tensorflow_internal.so" | ||
| find $out -name '*.so' -exec patchelf --set-rpath "$rrPath" {} \; |
There was a problem hiding this comment.
.so is linux only, so use .${stdenv.hostPlatform.extensions.sharedLibrary}
There was a problem hiding this comment.
patchelf is also linux only so the whole find commands needs to be made conditional.
| meta = with stdenv.lib; { | ||
| description = "Computation using data flow graphs for scalable machine learning"; | ||
| homepage = http://tensorflow.org; | ||
| license = licenses.asl20; | ||
| maintainers = with maintainers; [ jyp abbradar ]; | ||
| platforms = platforms.darwin; | ||
| platforms = with platforms; if cudaSupport then linux else linux ++ darwin; |
| sha256 = "0s5dy956jvwazqflc90v15i912zvhwsbzlf0cl8k7isq52j6g3kp"; | ||
| }; | ||
| } | ||
| # This file, prefetcher.sh, was generated using the following script: |
There was a problem hiding this comment.
just check in the script in a separate file
jyp
force-pushed
the
tf-1.7.1-bin
branch
3 times, most recently
from
55eae50 to
442d3e6
Compare
|
@FRidh Thanks for the review! I think I've applied all your comments in the current version. |
|
Indeed. In master no fixup is done on darwin; i've updated the commit as
you suggest.
…On Thu, May 31, 2018 at 9:21 AM, Jörg Thalheim ***@***.***> wrote:
***@***.**** commented on this pull request.
------------------------------
In pkgs/development/python-modules/tensorflow/bin.nix
<#41259 (comment)>:
> + # Note that we need to run *after* the fixup phase because the
+ # libraries are loaded at runtime. If we run in preFixup then
+ # patchelf --shrink-rpath will remove the cuda libraries.
+ postFixup = let
+ rpath = stdenv.lib.makeLibraryPath
+ (if cudaSupport then
+ [ stdenv.cc.cc.lib zlib cudatoolkit_joined cudnn
+ nvidia_x11 ]
+ else
+ [ stdenv.cc.cc.lib zlib ]
+ );
+ in
+ ''
+ rrPath="$out/${python.sitePackages}/tensorflow/:${rpath}"
+ internalLibPath="$out/${python.sitePackages}/tensorflow/python/_pywrap_tensorflow_internal.so"
+ find $out -name '*.so' -exec patchelf --set-rpath "$rrPath" {} \;
patchelf is also linux only so the whole find commands needs to be made
conditional.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#41259 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AABsY6k9s-WemyqkUMqcE9DhZEldVr8Uks5t35n1gaJpZM4UTa2q>
.
|
This version is a dependency of tensorflow binaries
Re-instates binary build for all versions.
GrahamcOfBorg
added
the
8.has: clean-up
|
@FRidh I believe that the PR satisfies your suggestions. What do you think? |
|
I'm getting |
|
@dotlambda If you're targetting py 3.6 it'll work, but not with eager graph execution. Would you care submitting a bug report? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Motivation for this change
Tensorflow is broken in master.
Things done
build-use-sandboxinnix.confon non-NixOS)nix-shell -p nox --run "nox-review wip"./result/bin/)