01-original.hs

#!/usr/bin/env stack
-- stack --resolver lts-7.14 --install-ghc exec -- ghc -O2 -with-rtsopts=-s
import Control.Monad.ST
import Data.STRef

a k x1 x2 x3 x4 x5 =
   do kk <- newSTRef k
      let b = do k <- modifySTRef kk pred >> readSTRef kk; a k b x1 x2 x3 x4
      if k <= 0 then do x3' <- x3; x4' <- x4; return (x3' + x4')
                else do x5' <- x5; b' <- b; return (x5' + b')

main = print (runST (a 22 (return 1) (return (-1)) (return (-1)) (return 1) (return 0)))

02-strict-modify.hs

#!/usr/bin/env stack
-- stack --resolver lts-7.14 --install-ghc exec -- ghc -O2 -with-rtsopts=-s
import Control.Monad.ST
import Data.STRef

a k x1 x2 x3 x4 x5 =
   do kk <- newSTRef k
      let b = do k <- modifySTRef' kk pred >> readSTRef kk; a k b x1 x2 x3 x4
      if k <= 0 then do x3' <- x3; x4' <- x4; return (x3' + x4')
                else do x5' <- x5; b' <- b; return (x5' + b')

main = print (runST (a 22 (return 1) (return (-1)) (return (-1)) (return 1) (return 0)))

03-force-int.hs

#!/usr/bin/env stack
-- stack --resolver lts-7.14 --install-ghc exec -- ghc -O2 -with-rtsopts=-s
import Control.Monad.ST
import Data.STRef

a :: Int
  -> ST s Int
  -> ST s Int
  -> ST s Int
  -> ST s Int
  -> ST s Int
  -> ST s Int
a k x1 x2 x3 x4 x5 =
   do kk <- newSTRef k
      let b = do k <- modifySTRef' kk pred >> readSTRef kk; a k b x1 x2 x3 x4
      if k <= 0 then do x3' <- x3; x4' <- x4; return (x3' + x4')
                else do x5' <- x5; b' <- b; return (x5' + b')

main = print (runST (a 22 (return 1) (return (-1)) (return (-1)) (return 1) (return 0)))

04-use-subtract.hs

#!/usr/bin/env stack
-- stack --resolver lts-7.14 --install-ghc exec -- ghc -O2 -with-rtsopts=-s
import Control.Monad.ST
import Data.STRef

a :: Int
  -> ST s Int
  -> ST s Int
  -> ST s Int
  -> ST s Int
  -> ST s Int
  -> ST s Int
a k x1 x2 x3 x4 x5 =
   do kk <- newSTRef k
      let b = do k <- modifySTRef' kk (subtract 1) >> readSTRef kk; a k b x1 x2 x3 x4
      if k <= 0 then do x3' <- x3; x4' <- x4; return (x3' + x4')
                else do x5' <- x5; b' <- b; return (x5' + b')

main = print (runST (a 22 (return 1) (return (-1)) (return (-1)) (return 1) (return 0)))

05-strict-return.hs

#!/usr/bin/env stack
-- stack --resolver lts-7.14 --install-ghc exec -- ghc -O2 -with-rtsopts=-s
import Control.Monad.ST
import Data.STRef

a :: Int
  -> ST s Int
  -> ST s Int
  -> ST s Int
  -> ST s Int
  -> ST s Int
  -> ST s Int
a k x1 x2 x3 x4 x5 =
   do kk <- newSTRef k
      let b = do k <- modifySTRef' kk (subtract 1) >> readSTRef kk; a k b x1 x2 x3 x4
      if k <= 0 then do x3' <- x3; x4' <- x4; return $! x3' + x4'
                else do x5' <- x5; b' <- b; return $! x5' + b'

main = print (runST (a 22 (return 1) (return (-1)) (return (-1)) (return 1) (return 0)))

allocations.txt

Original version:

-14254067
   4,907,852,080 bytes allocated in the heap
   5,442,629,272 bytes copied during GC
   1,121,103,736 bytes maximum residency (13 sample(s))
      27,788,856 bytes maximum slop
            2201 MB total memory in use (0 MB lost due to fragmentation)

                                     Tot time (elapsed)  Avg pause  Max pause
  Gen  0      9367 colls,     0 par    2.344s   2.428s     0.0003s    0.0049s
  Gen  1        13 colls,     0 par    2.672s   5.098s     0.3922s    2.9496s

  INIT    time    0.000s  (  0.002s elapsed)
  MUT     time    2.413s  (  3.051s elapsed)
  GC      time    5.016s  (  7.527s elapsed)
  EXIT    time    0.021s  (  0.229s elapsed)
  Total   time    7.451s  ( 10.809s elapsed)

  %GC     time      67.3%  (69.6% elapsed)

  Alloc rate    2,033,853,020 bytes per MUT second

  Productivity  32.7% of total user, 22.5% of total elapsed

modifySTRef':

-14254067
   3,932,003,776 bytes allocated in the heap
   5,720,188,216 bytes copied during GC
   1,263,737,656 bytes maximum residency (13 sample(s))
      25,694,112 bytes maximum slop
            2475 MB total memory in use (0 MB lost due to fragmentation)

                                     Tot time (elapsed)  Avg pause  Max pause
  Gen  0      7498 colls,     0 par    2.294s   2.454s     0.0003s    0.0100s
  Gen  1        13 colls,     0 par    2.702s   4.047s     0.3113s    2.1148s

  INIT    time    0.000s  (  0.003s elapsed)
  MUT     time    2.379s  (  3.432s elapsed)
  GC      time    4.997s  (  6.501s elapsed)
  EXIT    time    0.022s  (  0.249s elapsed)
  Total   time    7.399s  ( 10.186s elapsed)

  %GC     time      67.5%  (63.8% elapsed)

  Alloc rate    1,652,871,915 bytes per MUT second

  Productivity  32.5% of total user, 23.6% of total elapsed

Add Int type signature:

-14254067
   4,004,140,832 bytes allocated in the heap
   4,068,947,688 bytes copied during GC
     695,459,136 bytes maximum residency (12 sample(s))
      25,423,360 bytes maximum slop
            1363 MB total memory in use (0 MB lost due to fragmentation)

                                     Tot time (elapsed)  Avg pause  Max pause
  Gen  0      7631 colls,     0 par    2.047s   2.075s     0.0003s    0.0024s
  Gen  1        12 colls,     0 par    1.093s   1.653s     0.1377s    0.7375s

  INIT    time    0.000s  (  0.002s elapsed)
  MUT     time    1.603s  (  1.502s elapsed)
  GC      time    3.140s  (  3.728s elapsed)
  EXIT    time    0.019s  (  0.155s elapsed)
  Total   time    4.762s  (  5.387s elapsed)

  %GC     time      65.9%  (69.2% elapsed)

  Alloc rate    2,497,881,075 bytes per MUT second

  Productivity  34.1% of total user, 30.1% of total elapsed

Replace pred with subtract 1: no change

Strict return

-14254067
   2,703,009,728 bytes allocated in the heap
       7,033,480 bytes copied during GC
          64,544 bytes maximum residency (3 sample(s))
          74,032 bytes maximum slop
               2 MB total memory in use (0 MB lost due to fragmentation)

                                     Tot time (elapsed)  Avg pause  Max pause
  Gen  0      5154 colls,     0 par    0.024s   0.029s     0.0000s    0.0006s
  Gen  1         3 colls,     0 par    0.000s   0.004s     0.0014s    0.0040s

  INIT    time    0.000s  (  0.002s elapsed)
  MUT     time    0.862s  (  0.883s elapsed)
  GC      time    0.025s  (  0.033s elapsed)
  EXIT    time    0.000s  (  0.000s elapsed)
  Total   time    0.887s  (  0.919s elapsed)

  %GC     time       2.8%  (3.6% elapsed)

  Alloc rate    3,137,008,592 bytes per MUT second

  Productivity  97.2% of total user, 93.9% of total elapsed