#8199: Get rid of HEAP_ALLOCED ----------------------------+---------------------------------------------- Reporter: ezyang | Owner: ezyang Type: feature | Status: new request | Milestone: 7.10.1 Priority: normal | Version: 7.7 Component: | Keywords: Compiler | Architecture: Unknown/Multiple Resolution: | Difficulty: Project (more than a week) Operating System: | Blocked By: Unknown/Multiple | Related Tickets: Type of failure: | None/Unknown | Test Case: | Blocking: | ----------------------------+---------------------------------------------- Description changed by ezyang: Old description:
This bug is to track progress of removing HEAP_ALLOCED from GHC, promising faster GC, as long as we can keep the cost of indirections down.
The relevant wiki article: http://ghc.haskell.org/trac/ghc/wiki/Commentary/Rts/Storage/HeapAlloced ; we are implementing method 2. The attached patchset is *almost* able to support GHC itself; a ghc-stage2 built this way can compile most files, but is much slower in some cases, and probably has some correctness bugs.
Improvements possible:
* Don't pay for a double indirection when -fPIC is turned on. Probably the easiest way to do this is to *not* bake in the indirections into compiled code when it is -fPIC'd, and instead scribble over the GOT. However, I don't know how to go backwards from a symbol to a GOT entry, so we might need some heinous assembly hacks to get this working.
* Closure tables should have their indirections short-circuited by the initialization code.
* We are paying an indirection when a GC occurs when the closure is not in R1. According to the wiki page, technically this is not needed, but I don't know how to eliminate references to the closure itself from stg_gc_fun.
* Properly handle the Windows DLL case (e.g. SRTs); this should be simpler now that everything is an indirection.
* Bikeshed hs_main API changes (because closures are not valid prior to RTS initialization, so you have to pass in an indirection instead)
* Improve static indirection and stable pointer registration, avoiding binary bloat from `__attribute(constructor)__` stubs.
* Need to implement a new megablock tracking structure so we can free/check for lost blocks
New description: This bug is to track progress of removing HEAP_ALLOCED from GHC, promising faster GC, as long as we can keep the cost of indirections down. The relevant wiki article: http://ghc.haskell.org/trac/ghc/wiki/Commentary/Rts/Storage/HeapAlloced ; we are implementing method 2. The attached patchset is *almost* able to support GHC itself; a ghc-stage2 built this way can compile most files, but is much slower in some cases, and probably has some correctness bugs. Improvements possible: * Don't pay for a double indirection when -fPIC is turned on. Probably the easiest way to do this is to *not* bake in the indirections into compiled code when it is -fPIC'd, and instead scribble over the GOT. However, I don't know how to go backwards from a symbol to a GOT entry, so we might need some heinous assembly hacks to get this working. * Closure tables should have their indirections short-circuited by the initialization code. * We are paying an indirection when a GC occurs when the closure is not in R1. According to the wiki page, technically this is not needed, but I don't know how to eliminate references to the closure itself from stg_gc_fun. * Properly handle the Windows DLL case (e.g. SRTs); this should be simpler now that everything is an indirection. * Bikeshed hs_main API changes (because closures are not valid prior to RTS initialization, so you have to pass in an indirection instead) * Improve static indirection and stable pointer registration, avoiding binary bloat from `__attribute(constructor)__` stubs. * ~~Need to implement a new megablock tracking structure so we can free/check for lost blocks~~. Now that efficient lookup is not necessary, perhaps we can write-optimize the megablock tracking structures. -- -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/8199#comment:3 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler