#7367: Optimiser / Linker Problem on amd64 -------------------------------+-------------------------------------------- Reporter: wurmli | Owner: Type: bug | Status: new Priority: normal | Milestone: 7.8.1 Component: Build System | Version: 7.6.1 Keywords: | Os: Linux Architecture: x86_64 (amd64) | Failure: Runtime performance bug Difficulty: Unknown | Testcase: Blockedby: | Blocking: Related: | -------------------------------+-------------------------------------------- Changes (by igloo): * milestone: => 7.8.1 Comment: It looks like the "bytes allocated in the heap" dramatically increased between 7.4 and 7.6: {{{ $ ghc-7.4.1 --make -O q -XBangPatterns -threaded -rtsopts [1 of 1] Compiling Main ( q.hs, q.o ) Linking q ... $ time ./q 12 +RTS -N4 -s 3968050 Pfannkuchen(12) = 65 88,592 bytes allocated in the heap 6,040 bytes copied during GC 46,928 bytes maximum residency (1 sample(s)) 43,184 bytes maximum slop 4 MB total memory in use (1 MB lost due to fragmentation) Tot time (elapsed) Avg pause Max pause Gen 0 0 colls, 0 par 0.00s 0.00s 0.0000s 0.0000s Gen 1 1 colls, 0 par 0.00s 0.00s 0.0002s 0.0002s Parallel GC work balance: -nan (0 / 0, ideal 4) MUT time (elapsed) GC time (elapsed) Task 0 (worker) : 78.76s ( 20.00s) 0.00s ( 0.00s) Task 1 (worker) : 78.76s ( 20.00s) 0.00s ( 0.00s) Task 2 (bound) : 78.75s ( 20.00s) 0.00s ( 0.00s) Task 3 (worker) : 78.76s ( 20.00s) 0.00s ( 0.00s) Task 4 (worker) : 78.75s ( 20.00s) 0.00s ( 0.00s) Task 5 (worker) : 78.76s ( 20.00s) 0.00s ( 0.00s) SPARKS: 0 (0 converted, 0 overflowed, 0 dud, 0 GC'd, 0 fizzled) INIT time 0.00s ( 0.00s elapsed) MUT time 78.75s ( 20.00s elapsed) GC time 0.00s ( 0.00s elapsed) EXIT time 0.00s ( 0.00s elapsed) Total time 78.76s ( 20.00s elapsed) Alloc rate 1,124 bytes per MUT second Productivity 100.0% of total user, 393.9% of total elapsed gc_alloc_block_sync: 0 whitehole_spin: 0 gen[0].sync: 0 gen[1].sync: 0 ./q 12 +RTS -N4 -s 78.76s user 0.00s system 393% cpu 19.998 total }}} {{{ $ ghc-7.6.2 --make -O q -XBangPatterns -threaded -rtsopts [1 of 1] Compiling Main ( q.hs, q.o ) Linking q ... $ time ./q 12 +RTS -N4 -s 3968050 Pfannkuchen(12) = 65 10,538,123,264 bytes allocated in the heap 390,688 bytes copied during GC 46,856 bytes maximum residency (2 sample(s)) 51,448 bytes maximum slop 3 MB total memory in use (0 MB lost due to fragmentation) Tot time (elapsed) Avg pause Max pause Gen 0 5843 colls, 5843 par 28.64s 0.38s 0.0001s 0.0131s Gen 1 2 colls, 1 par 0.00s 0.00s 0.0002s 0.0002s Parallel GC work balance: 61.48% (serial 0%, perfect 100%) TASKS: 6 (1 bound, 5 peak workers (5 total), using -N4) SPARKS: 0 (0 converted, 0 overflowed, 0 dud, 0 GC'd, 0 fizzled) INIT time 0.00s ( 0.00s elapsed) MUT time 54.57s ( 20.89s elapsed) GC time 28.64s ( 0.38s elapsed) EXIT time 0.00s ( 0.00s elapsed) Total time 83.21s ( 21.27s elapsed) Alloc rate 193,121,729 bytes per MUT second Productivity 65.6% of total user, 256.6% of total elapsed gc_alloc_block_sync: 28902 whitehole_spin: 0 gen[0].sync: 0 gen[1].sync: 0 ./q 12 +RTS -N4 -s 83.21s user 0.18s system 392% cpu 21.272 total }}} -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7367#comment:2 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler