Skip to content
Snippets Groups Projects

Building ghc with "make -jN" gives different base ABI hashes for N<=8 and N>8

    • Closed Issue created by Jens Petersen @juhp

    While building ghc-7.8.4 for Fedora 22 (devel) I found that doing highly parallel make builds of ghc affects the ABI hash of base, etc.

    The threshold seems to be between -j8 and -j12, which likely changes the build order of some modules causing changes to ABI hashes.

    While it is not hard to workaround, it is not really nice so it would be good to track this down IMHO. I haven't seen this kind of issue since 7.4.2 anyway so it may be an indirect buildsystem "regression" with ghc-7.8?

    Edited
    To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information

    Child items ...

    • Activity

    • Jens Petersen changed weight to 5

      changed weight to 5

    • Jens Petersen added Tbug Trac import labels

      added Tbug Trac import labels

    • Jens Petersen changed the description

      changed the description

    • Jens Petersen
      Jens Petersen @juhp ·
      Author Developer
    • Bertram Felgenhauer
      Bertram Felgenhauer @int-e ·
      Reporter

      I could reproduce this with ghc-7.8.4 (for example, I obtained 3 different ABI hashes for the haskell98 package from building ghc 6 times, with -j1, -j4, -j8, -j12, -j16 and -j24).

      On the other hand, ghc-7.10 RC2 produced identical ABI hashes in the same scenario. So this appears to have been fixed, at least for the libraries in ghc.

      PS. These builds were done on an x86_64 linux server with CentOS 6.4, using ghc-7.6.2 for bootstrapping, and gcc-4.4.7. I used the 'perf' settings, so no llvm was involved. So take the results with a grain of salt.

      Trac metadata
      Trac field Value
      Related → #9370
    • Bertram Felgenhauer
      Bertram Felgenhauer @int-e ·
      Reporter
      Trac metadata
      Trac field Value
      Related #9370 → #4012 (closed)
    • Jens Petersen changed title from Building ghc with "make -j12" gives different ABI hashes for base to "make -j8" to Building ghc with "make -jN" gives different base ABI hashes for N<=8 and N>8

      changed title from Building ghc with "make -j12" gives different ABI hashes for base to "make -j8" to Building ghc with "make -jN" gives different base ABI hashes for N<=8 and N>8

    • Jens Petersen
      Jens Petersen @juhp ·
      Author Developer

      Thanks int-e for looking into this - much appreciated. :)

      It is encouraging indeed that this should be fixed with ghc-7.10, hopefully it won't appear again in the future. :)

    • Thomas Miedema closed

      closed

    • Thomas Miedema
      Thomas Miedema @thomie ·

      Replying to [ticket:10090#comment:96357 int-e]:

      1. .. ghc-7.10 RC2 produced identical ABI hashes in the same scenario. So this appears to have been fixed, at least for the libraries in ghc.

      @juhpetersen: please reopen if you still have problems with 7.10.

      Trac metadata
      Trac field Value
      Resolution Unresolved → ResolvedWorksForMe
    • Jens Petersen
      Jens Petersen @juhp ·
      Author Developer

      Okay thanks.

      Just for the record s390 seems to be ABI unstable on -j4.

    • Ben Gamari
      Ben Gamari @bgamari ·
      Maintainer

      The new Hadrian build system has been merged. Relabeling the tickets concerning the legacy make build system to prevent confusion.

      Trac metadata
      Trac field Value
      Component Build System → Build System (make)
    • Ben Gamari added determinism label

      added determinism label

    • Ben Gamari added Pnormal label

      added Pnormal label

    Please register or sign in to reply