#4464: RTS options broken for dynamic libraries
---------------------------------+------------------------------------------
Reporter: rl | Owner:
Type: bug | Status: new
Priority: normal | Milestone: 7.2.1
Component: Runtime System | Version: 7.1
Keywords: | Testcase:
Blockedby: | Difficulty:
Os: Unknown/Multiple | Blocking:
Architecture: Unknown/Multiple | Failure: None/Unknown
---------------------------------+------------------------------------------
Changes (by igloo):
* owner: igloo =>
Comment:
Why is not copying the arguments bad?
And regarding `hs_init` aborting, currently we have:
* `barf` calls `fatalInternalErrorFn` which calls `abort()`, but you can
override `fatalInternalErrorFn`. However, `barf` then calls `stg_exit` if
`fatalInternalErrorFn` returns.
* There are a number of other places that call `stg_exit` (It might make
sense to make a `die` function similar to `barf` for these. Perhaps even
define the default `fatalInternalErrorFn` as calling `die`)
but I'm not really sure what we should do instead. Should we drop the
fall-through `stg_exit` in `barf`/`die`, and turn every single
`barf`/`stg_exit` usage site into
{{{
barf(...); return NULL /* or 0 or -1 or whatever makes sense to indicate
failure */
}}}
? That could presumably also mean a lot of more checking for errors by RTS
functions that call other RTS functions (which may transitively
`barf`/`die`).
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4464#comment:9>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
