#12776: Panic Simplifier ticks exhausted since ghc 8 -------------------------------------+------------------------------------- Reporter: sjcjoosten | Owner: Type: bug | Status: new Priority: highest | Milestone: Component: Compiler | Version: 8.0.1 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Compile-time | Unknown/Multiple crash or panic | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by simonpj): Thank you for the small test case. I can see what is happening here. After `SpecConstr`, and occurrence analysis we get {{{ Rec { $slast_s_s1J5 :: forall a_a1BZ. a_a1BZ -> Slist a_a1BZ -> Eq a_a1BZ => a_a1BZ $slast_s_s1J5 = \ (@ a_a1BZ) (sc_s1J0 :: a_a1BZ) (sc_s1J1 :: Slist a_a1BZ) (sc_s1IZ :: Eq a_a1BZ) -> case sc_s1J1 of wild_Xp [Dmd=] { Nil_s -> sc_s1J0; Cons_s _ [Occ=Dead] _ [Occ=Dead] -> let { sc_s1J1 [Occ=Once] :: Slist a_a1BZ [LclId] sc_s1J1 = wild_Xp } in last_s @ a_a1BZ sc_s1IZ sc_s1J1 } last_s [Occ=LoopBreaker] :: forall a_a1kp. Eq a_a1kp => Slist a_a1kp -> a_a1kp [RULES: "SC:last_s0" [ALWAYS] forall (@ a_a1BZ) (sc_s1J0 :: a_a1BZ) (sc_s1J1 :: Slist a_a1BZ) (sc_s1IZ :: Eq a_a1BZ). last_s @ a_a1BZ sc_s1IZ (Cons_s @ a_a1BZ sc_s1J0 sc_s1J1) = $slast_s_s1J5 @ a_a1BZ sc_s1J0 sc_s1J1 sc_s1IZ] last_s = \ (@ a_a1BC) ($dEq_a1BE :: Eq a_a1BC) (ds_d1Do [Occ=Once!] :: Slist a_a1BC) -> case ds_d1Do of { Nil_s -> lvl_s1Fa @ a_a1BC; Cons_s x_a1kB [Occ=Once] xs_a1kC [Occ=Once!, Dmd=] -> case xs_a1kC of { Nil_s -> x_a1kB; Cons_s a1_a1zd [Occ=Once] a2_a1ze [Occ=Once] -> $slast_s_s1J5 @ a_a1BC a1_a1zd a2_a1ze $dEq_a1BE } } end Rec } }}} Then in `last_s`: * `$s_last_s_s1J5` is inlined * The RULE fires, yielding a new cal to `$s_last_s_s1J5` * `$s_last_s_s1J5` is inlined again * and so on. The problem is that `$s_last_s_s1J5` should be a loop breaker. See the extensive notes under `Note [Choosing loop breakers]`. But alas in `occAnalRec` we see this {{{ pairs :: [(Id,CoreExpr)] pairs | isEmptyVarSet weak_fvs = reOrderNodes 0 bndr_set weak_fvs tagged_nodes [] | otherwise = loopBreakNodes 0 bndr_set weak_fvs loop_breaker_edges [] }}} We want to go via the `loopBreakNodes` call, but actually `weak_fvs` is empty so we go via the former call. I'm not yet sure exactly why... just recording breadcrumbs for now. How urgent is this for you, Stef? -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/12776#comment:3 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#12776: Panic Simplifier ticks exhausted since ghc 8 -------------------------------------+------------------------------------- Reporter: sjcjoosten | Owner: Type: bug | Status: new Priority: highest | Milestone: Component: Compiler | Version: 8.0.1 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Compile-time | Unknown/Multiple crash or panic | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by Simon Peyton Jones ): In [changeset:"7fe71636dc14a219c26def5cac2da9fa20689540/ghc" 7fe7163/ghc]: {{{ #!CommitTicketReference repository="ghc" revision="7fe71636dc14a219c26def5cac2da9fa20689540" Adapt the (commented out) pprTrace in OccurAnal I did this while investigating Trac #12776 }}} -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/12776#comment:5 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#12776: Panic Simplifier ticks exhausted since ghc 8 -------------------------------------+------------------------------------- Reporter: sjcjoosten | Owner: Type: bug | Status: new Priority: highest | Milestone: Component: Compiler | Version: 8.0.1 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Compile-time | Unknown/Multiple crash or panic | Test Case: Blocked By: | Blocking: Related Tickets: #12843 #12675 | Differential Rev(s): #12789 | Wiki Page: | -------------------------------------+------------------------------------- Comment (by Simon Peyton Jones ): In [changeset:"6ec2304f46c9a5423943c5bf29bd8a8c062b6560/ghc" 6ec2304f/ghc]: {{{ #!CommitTicketReference repository="ghc" revision="6ec2304f46c9a5423943c5bf29bd8a8c062b6560" Fix an long-standing bug in OccurAnal This bug was beautifully characterised in Trac #12776, which showed a small program for which the inliner went into an infinite loop. Eeek. It turned out to be a genuine and long-standing bug in the occurrence analyer, specifically in the bit that identifies loop breakers. In this line pairs | isEmptyVarSet weak_fvs = reOrderNodes 0 bndr_set weak_fvs tagged_nodes [] | otherwise = loopBreakNodes 0 bndr_set weak_fvs loop_breaker_edges [] the 'tagged_nodes' should be 'loop_breaker_edges'. That's it! The diff looks a lot bigger because I did some work on comments and variable naming, but that's all it is. We were using the wrong set of dependencies! I'm astonished that this bug has not caused more trouble. It dates back to at least 2011 and maybe further. }}} -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/12776#comment:7 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#12776: Panic Simplifier ticks exhausted since ghc 8 -------------------------------------+------------------------------------- Reporter: sjcjoosten | Owner: Type: bug | Status: merge Priority: highest | Milestone: Component: Compiler | Version: 8.0.1 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: Compile-time | Test Case: crash or panic | simplCore/should_compile/T12776 Blocked By: | Blocking: Related Tickets: #12843 #12675 | Differential Rev(s): #12789 | Wiki Page: | -------------------------------------+------------------------------------- Changes (by simonpj): * testcase: => simplCore/should_compile/T12776 * status: new => merge Comment: Big thank you to sjcjoosten for making such a tractable test case. THANKS! It's great to have this one nailed. Worth merging this (or the key part anyway) to 8.0.3 Simon -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/12776#comment:9 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#12776: Panic Simplifier ticks exhausted since ghc 8 -------------------------------------+------------------------------------- Reporter: sjcjoosten | Owner: (none) Type: bug | Status: closed Priority: highest | Milestone: 8.2.1 Component: Compiler | Version: 8.0.1 Resolution: fixed | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: Compile-time | Test Case: crash or panic | simplCore/should_compile/T12776 Blocked By: | Blocking: Related Tickets: #12843 #12675 | Differential Rev(s): #12789 | Wiki Page: | -------------------------------------+------------------------------------- Changes (by bgamari): * status: merge => closed * resolution: => fixed -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/12776#comment:12 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler