diff options
| author | Simon Peyton Jones <simonpj@microsoft.com> | 2018-04-26 15:59:13 +0100 |
|---|---|---|
| committer | Simon Peyton Jones <simonpj@microsoft.com> | 2018-04-26 17:44:02 +0100 |
| commit | 3d38e8284b7382844f9862e8d8afbae9c7248b09 (patch) | |
| tree | bfb23aa121d7a0192f2a73874b0796767e88d6d7 /testsuite/tests/ghci.debugger | |
| parent | 625eea9332532666a35b4cc680ef4f7e51d4b2d1 (diff) | |
| download | haskell-3d38e8284b7382844f9862e8d8afbae9c7248b09.tar.gz | |
Do not unpack class dictionaries with INLINABLE
Matthew Pickering uncovered a bad performance hole in the way
that single-method dictionaries work, described in Trac #14955.
See Note [Do not unpack class dictionaries] in WwLib.
I tried to fix this 6 years ago, but got it slightly wrong. This patch
fixes it, which makes a dramatic improvement in the test case.
Nofib highlights: not much happening:
Program Size Allocs Runtime Elapsed TotalMem
-----------------------------------------------------------------
VSM -0.3% +2.7% -7.4% -7.4% 0.0%
cacheprof -0.0% +0.1% +0.3% +0.7% 0.0%
integer -0.0% +1.1% +7.5% +7.5% 0.0%
tak -0.1% -0.2% 0.024 0.024 0.0%
-----------------------------------------------------------------
Min -4.4% -0.2% -7.4% -7.4% -8.0%
Max +0.6% +2.7% +7.5% +7.5% 0.0%
Geom Mean -0.1% +0.0% +0.1% +0.1% -0.2%
I investigated VSM. The patch unpacks class dictionaries a bit more
than before (i.e. does so if there is no INLINABLE pragma). And that
gives better code in VSM (less dictionary selection etc), but one closure
gets one word bigger.
I'll accept these changes in exchange for more robust performance.
Some ghci.debugger output wobbled around (order of bindings
being displayed). I have no idea why; but I accepted the changes.
Diffstat (limited to 'testsuite/tests/ghci.debugger')
| -rw-r--r-- | testsuite/tests/ghci.debugger/scripts/break006.stdout | 4 | ||||
| -rw-r--r-- | testsuite/tests/ghci.debugger/scripts/hist001.stdout | 2 | ||||
| -rw-r--r-- | testsuite/tests/ghci.debugger/scripts/hist002.stdout | 2 |
3 files changed, 4 insertions, 4 deletions
diff --git a/testsuite/tests/ghci.debugger/scripts/break006.stdout b/testsuite/tests/ghci.debugger/scripts/break006.stdout index 8a2463fa4e..cd9f6ea1fe 100644 --- a/testsuite/tests/ghci.debugger/scripts/break006.stdout +++ b/testsuite/tests/ghci.debugger/scripts/break006.stdout @@ -4,14 +4,14 @@ f :: Integer -> a = _ x :: Integer = 1 xs :: [Integer] = [2,3] xs :: [Integer] = [2,3] -x :: Integer = 1 f :: Integer -> a = _ +x :: Integer = 1 _result :: [a] = _ y = (_t1::a) y = 2 xs :: [Integer] = [2,3] -x :: Integer = 1 f :: Integer -> Integer = _ +x :: Integer = 1 _result :: [Integer] = _ y :: Integer = 2 _t1 :: Integer = 2 diff --git a/testsuite/tests/ghci.debugger/scripts/hist001.stdout b/testsuite/tests/ghci.debugger/scripts/hist001.stdout index a19a34f315..b52e8aa5fe 100644 --- a/testsuite/tests/ghci.debugger/scripts/hist001.stdout +++ b/testsuite/tests/ghci.debugger/scripts/hist001.stdout @@ -20,8 +20,8 @@ _result :: a f :: Integer -> a x :: Integer xs :: [t] = [] -x :: Integer = 2 f :: Integer -> a = _ +x :: Integer = 2 _result :: a = _ _result = 3 Logged breakpoint at Test3.hs:2:18-31 diff --git a/testsuite/tests/ghci.debugger/scripts/hist002.stdout b/testsuite/tests/ghci.debugger/scripts/hist002.stdout index a19a34f315..b52e8aa5fe 100644 --- a/testsuite/tests/ghci.debugger/scripts/hist002.stdout +++ b/testsuite/tests/ghci.debugger/scripts/hist002.stdout @@ -20,8 +20,8 @@ _result :: a f :: Integer -> a x :: Integer xs :: [t] = [] -x :: Integer = 2 f :: Integer -> a = _ +x :: Integer = 2 _result :: a = _ _result = 3 Logged breakpoint at Test3.hs:2:18-31 |