diff options
author | Jason Zaugg <jzaugg@gmail.com> | 2014-04-24 10:53:20 +0200 |
---|---|---|
committer | Jason Zaugg <jzaugg@gmail.com> | 2014-05-08 10:33:45 +0200 |
commit | b154269b0e7447cfbf7fdb73c8c4efcdca545b11 (patch) | |
tree | 662076793ed1333d8d0393efac365f9fe3cd80b9 /test/files/pos/t8531/Test.scala | |
parent | aa3a3dd6a420f57b0fe815bfcb2b23fbb4284468 (diff) | |
download | scala-b154269b0e7447cfbf7fdb73c8c4efcdca545b11.tar.gz scala-b154269b0e7447cfbf7fdb73c8c4efcdca545b11.tar.bz2 scala-b154269b0e7447cfbf7fdb73c8c4efcdca545b11.zip |
SI-8531 Better space efficiency for patmat analysis
By adding logging to `clause`, I found that the majority of
calls provide 0 or 1 elements. In SI-7020 / 69557da55, we changed this
method to use a `LinkedHashSet` to have deterministic results
for clauses with more elements. But I suspect that this
contributes to higher memory usage from the pattern matcher.
The enclosed test case, carefully whittled down by @oxbowlakes,
used to consume an inordinate amount of memory and time.
After this patch, it is back to 2.10.4 performance.
I have run `neg/t7020.scala` in a loop and it still is deterministic.
Diffstat (limited to 'test/files/pos/t8531/Test.scala')
-rw-r--r-- | test/files/pos/t8531/Test.scala | 24 |
1 files changed, 24 insertions, 0 deletions
diff --git a/test/files/pos/t8531/Test.scala b/test/files/pos/t8531/Test.scala new file mode 100644 index 0000000000..59861435a6 --- /dev/null +++ b/test/files/pos/t8531/Test.scala @@ -0,0 +1,24 @@ +package test + +// takes > 50s and > 800M heap to compile under 2.11.0 +import foobar._ +class `SI-8531` { + //https://issues.scala-lang.org/browse/SI-8531 + + import MyEnum._ + def foo(e1: MyEnum, e2: MyEnum) = (e1, e2) match { + case (A1, x) => "a1" + case (x, A1) => "a1" + case (A2, x) => "a2" + case (x, A2) => "a2" + case (A3, x) => "a3" + case (x, A3) => "a3" + case (A4, x) => "a4" + case (x, A4) => "a4" + case (A5, x) => "a5" + case (x, A5) => "a5" + case (A6, x) => "a6" + case (x, A6) => "a6" + case (a, b) => "ab" + } +} |