- ghc-tickets - Haskell.org

Re: [GHC] #2940: Do CSE after CorePrep
by GHC 23 Mar '17

23 Mar '17

#2940: Do CSE after CorePrep -------------------------------------+------------------------------------- Reporter: simonpj | Owner: simonpj Type: bug | Status: new Priority: lowest | Milestone: Component: Compiler | Version: 6.10.1 Resolution: | Keywords: CSE Operating System: Unknown/Multiple | Architecture: Type of failure: Runtime | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by ezyang): * keywords: => CSE -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/2940#comment:25> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #5344: CSE should look through coercions
by GHC 23 Mar '17

23 Mar '17

#5344: CSE should look through coercions -------------------------------------+------------------------------------- Reporter: reinerp | Owner: (none) Type: feature request | Status: new Priority: low | Milestone: ⊥ Component: Compiler | Version: 7.0.3 Resolution: | Keywords: CSE Operating System: Unknown/Multiple | Architecture: Type of failure: Runtime | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by ezyang): * keywords: cse => CSE -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/5344#comment:8> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #701: Better CSE optimisation
by GHC 23 Mar '17

23 Mar '17

#701: Better CSE optimisation -------------------------------------+------------------------------------- Reporter: simonmar | Owner: michalt Type: task | Status: new Priority: normal | Milestone: ⊥ Component: Compiler | Version: 6.4.1 Resolution: | Keywords: CSE Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: N/A Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by ezyang): * keywords: => CSE -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/701#comment:15> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #13219: CSE for join points
by GHC 23 Mar '17

23 Mar '17

#13219: CSE for join points -------------------------------------+------------------------------------- Reporter: lukemaurer | Owner: lukemaurer Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.1 Keywords: JoinPoints | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Join points are hazardous for CSE: {{{ join j = e in ... f e ... }}} The temptation is of course to rewrite to {{{ join j = e in ... f (jump j) ... }}} which doesn't typecheck because you can't have a jump in an argument. Similarly: {{{ join j x = e in let y = (... join j2 x = e in ...) in ... }}} can't be rewritten to {{{ join j x = e in let y = (... join j2 x = jump j x ...) in ... }}} because you can't jump out of a value binding. This doesn't seem to come up very often (took a long time for me to trip over it). The simple solution is simply to ignore CSE for join points. That's what we do at the moment. But this is saddening in some cases: {{{ join j x = e in join j2 x = e in ... }}} can perfectly well become {{{ join j x = e in join j2 x = jump j x in ... }}} We just need to track which join points are still currently valid. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13219> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

Re: [GHC] #12913: Port SplitSections to Windows
by GHC 23 Mar '17

23 Mar '17

#12913: Port SplitSections to Windows -------------------------------------+------------------------------------- Reporter: olsner | Owner: (none) Type: task | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 (Linking) | Resolution: | Keywords: Operating System: Windows | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: #8405 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by Phyx-): Ah, great thanks for the info, this saves me a lot of time! Yes I see now what you mean. I can certainly change GHC to output the right section names, that at least lessens the maintenance burden a bit. both it and string merging are on my list for 8.4 (along with dynamic linking). I'll take a look at his blog post and see! -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12913#comment:7> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #12913: Port SplitSections to Windows
by GHC 23 Mar '17

23 Mar '17

#12913: Port SplitSections to Windows -------------------------------------+------------------------------------- Reporter: olsner | Owner: (none) Type: task | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 (Linking) | Resolution: | Keywords: Operating System: Windows | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: #8405 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by awson): Nope, they were only partially checked in. If you follow the commit link in the last post of the [https://sourceware.org/bugzilla/show_bug.cgi?id=19254 discussion] you will understand that only 32-bit bugfix (no new features at all) was landed. The essential part was abandoned because of Simon Brenner's idea that we should fix things on the GHC side making section name generation scheme more "windowsish" which `binutils` already has support for. And yes, I meant string merging when writing about string literals handling. At least my patch solves **sections** merging problem here. To make proper string merging we should, I believe, go entirely different way. AFAIR Walter Bright had met exactly the same problem when developing his `D` compiler and he used `COMDAT` to solve it. He blogged about this a couple of months ago. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12913#comment:6> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7275: Give more detailed information about PINNED data in a heap profile
by GHC 23 Mar '17

23 Mar '17

#7275: Give more detailed information about PINNED data in a heap profile -------------------------------------+------------------------------------- Reporter: edsko | Owner: (none) Type: feature request | Status: new Priority: normal | Milestone: Component: Profiling | Version: 7.6.1 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: Other | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by dfeuer): Replying to [comment:13 varosi]: > Is there a chance that issue to enter GHC 8.2? 8.2 has already branched, so I don't think there is any chance of that. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7275#comment:14> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #12913: Port SplitSections to Windows
by GHC 23 Mar '17

23 Mar '17

#12913: Port SplitSections to Windows -------------------------------------+------------------------------------- Reporter: olsner | Owner: (none) Type: task | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 (Linking) | Resolution: | Keywords: Operating System: Windows | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: #8405 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by Phyx-): Thanks! I thought based on the timing that 2.28 contained them :) thought they were checked in, but probably after the branch? ABout the string literals I assume you mean string merging? GHC now generates a lot more string literals that are identical and should be merged. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12913#comment:5> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #12913: Port SplitSections to Windows
by GHC 23 Mar '17

23 Mar '17

#12913: Port SplitSections to Windows -------------------------------------+------------------------------------- Reporter: olsner | Owner: (none) Type: task | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 (Linking) | Resolution: | Keywords: Operating System: Windows | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: #8405 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by awson): Ah, and if you want already processed linker scripts here they are. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12913#comment:4> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #12913: Port SplitSections to Windows
by GHC 23 Mar '17

23 Mar '17

#12913: Port SplitSections to Windows -------------------------------------+------------------------------------- Reporter: olsner | Owner: (none) Type: task | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 (Linking) | Resolution: | Keywords: Operating System: Windows | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: #8405 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by awson): * Attachment "lscripts.tar.gz" added. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12913> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0