#709: "Fixup too large" error with -fasm on PowerPC
----------------------------------+-----------------------------------------
Reporter: simonmar | Owner:
Type: bug | Status: patch
Priority: low | Milestone: 6.8.1
Component: Compiler (NCG) | Version: 7.7
Resolution: | Keywords:
Os: Unknown/Multiple | Architecture: powerpc
Failure: Building GHC failed | Difficulty: Moderate (less than a day)
Testcase: | Blockedby:
Blocking: | Related:
----------------------------------+-----------------------------------------
Comment(by simonmar):
Replying to [comment:10 PHO]:
> Replying to [comment:9 simonmar]:
> > Could we not use something better than a hardcoded "5" for the info
table size? We know the size of the info table for a continuation:
`sizeof(StgRetInfoTable)`, and add to that the maximum offset due to
alignment.
>
> The size of `StgRetInfoTable` varies depending on the "way" (e.g. -prof)
so we can't simply grab it from the C compiler. I think it's easy to
> calculate the size of each tables represented as `CmmStatics`.
It also varies based on `tablesNextToCode` and other things, but you only
need the upper bound here. I would add an `stgRetInfoTableMaxWords ::
Int` somewhere, maybe `CmmInfo`.
> > You could also do better than `length blocks`: the actual number of
info tables is available.
>
> Right, but since info tables are scattered around a proc, I couldn't
simply replace {{{length blocks}}} with the actual number of tables.
I don't really understand that. But I don't mind if you want to use
`length blocks`, it's just a bit pessimistic.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/709#comment:11>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
