#1372: Recompilation checker should consider package versions (and other 
factors)
----------------------+-----------------------------------------------------
 Reporter:  bringert  |          Owner:  simonmar  
     Type:  bug       |         Status:  new       
 Priority:  normal    |      Milestone:  6.8 branch
Component:  Compiler  |        Version:  6.8.1     
 Severity:  normal    |     Resolution:            
 Keywords:            |     Difficulty:  Unknown   
 Testcase:            |   Architecture:  Unknown   
       Os:  Unknown   |  
----------------------+-----------------------------------------------------
Comment (by dons):

 We continue to get user reports of linker failures, and funny runtime
 behaviour, when:

 * a user updates xmonad darcs library
 * then rebuilds a previously built xmonad-contrib library (the extension
 repo)

 since we can't bump xmonad's version number on every internal change,
 cabal doesn't spot
 that 'xmonad', the library, has updated, and the result is linker errors
 or worse.
 For example, today, one user reports:

 {{{
 10:13  gwern>
 '/home/gwern/bin/lib/xmonad-0.4/ghc-6.8.1.20071117/libHSxmonad-0.4.a(Main.o):
 In function `stEm_info':
 10:14  gwern> (.text+0x8990): undefined reference to
               `X11zm1zi4zi0_GraphicsziX11ziXlibziExtras_a181_info'
 10:14  gwern> rebuilding and install x11 and using --make didn't work...
               better try cleaning everything
 10:15  dons> oh, clean everything.
 10:16  gwern> yes, that worked. geez.
 }}}

 Another reported massive memory use, that disappeared on cleaning.

 My suggestion would be that cabal requires a 'clean' to be run if the
 global or user package.conf has changed since the last build. That would
 solve this problem, wouldn't it?

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/1372#comment:22>
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

Reply via email to