Latest attempt to get this working on 0.7 #65
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
This breaks 0.6, but the point of pushing this now is that it reveals a segfault in Julia master
This was referenced Jul 13, 2018
Closed
With proposed PRs for TiledIterations this gets through the test suite. A tricky bounds violation and some inference failures remain.
Closed
A few more patches for v0.7
Minor adjustments for 0.7
Codecov Report
@@ Coverage Diff @@
## master #65 +/- ##
==========================================
+ Coverage 91.51% 94.49% +2.97%
==========================================
Files 8 8
Lines 1096 690 -406
==========================================
- Hits 1003 652 -351
+ Misses 93 38 -55
Continue to review full report at Codecov.
|
|
@timholy I wasn't sure whether you meant to just merge into this branch or fully into master. |
Earlier versions of Julia supported two types of lispy tuple programming:
one in which arguments shrink (e.g., via Base.tail) and the output gets
appended, and the other in which the result is appended to one of the
arguments (and therefore grows). For 0.7, inference has prioritized
either (1) proving that inference will converge or (2) exiting quickly.
Proving convergence is apparently much harder if some of the arguments
grow in size, so inference doesn't try very hard to get the right ansswer
anymore. Consequently inference was broken in our previous implementation.
This slight disadvantage is more than compensated by the fact that we
can now do inferrable arithmetic on type parameters, as exemplified here
by `ReshapedOneD{N,N-M}(arg)`.
|
Woot! Thanks to all who helped this happen. |
Closed
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.
This incorporates #41 and #64. As I am rushing to leave town for two weeks, most likely without any internet whatsoever, this is quick.
Most of the tests are working on 0.7. You need a fresh build of Julia and you also want JuliaLang/julia#28100 and JuliaLang/julia#28084. While I am gone I would urge members of the Images community to keep trying to get the JuliaImages packages working on 0.7: porting these has shaken out something like a dozen critical bugs in Base Julia and its stdlibs. Having put so much work into Julia and JuliaImages so far, it would be something of a shame if 1.0 got released and we couldn't really run JuliaImages packages.
Discourse advertised a slack channel on Fridays to help people port packages. Good luck!