#10043: runtime fails in threaded way on SPARC (bus error -> unaligned access to data) -----------------------------------+---------------------------------- Reporter: kgardas | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.1-rc2 Keywords: | Operating System: Solaris Architecture: sparc | Type of failure: Runtime crash Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -----------------------------------+---------------------------------- Hello, it looks like runtime has changed between 7.8 and 7.10 in a way it no longer works well on SPARC. Tested on SPARC/Solaris 2.11. Both compilers compiled as their are w/o any changes so both unregisterised. On 7.8 testblockalloc runs well, on 7.10.1-rc2 it fails in threaded way with "Bus error" {{{ karel@niagara:~/src/ghc-7.10.0.20150123/testsuite/tests/rts$ ./testblockalloc +RTS -I0 Bus Error (core dumped) }}} When run in debugger it points to assignment of 0 to tso->alloc_limit: {{{ karel@niagara:~/src/ghc-7.10.0.20150123/testsuite/tests/rts$ gdb ./testblockalloc GNU gdb 6.8 Copyright (C) 2008 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later http://gnu.org/licenses/gpl.html This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "sparc-sun-solaris2.11"... set ar(gdb) set args +RTS -I0 (gdb) r Starting program: /home/karel/src/ghc-7.10.0.20150123/testsuite/tests/rts/testblockalloc +RTS -I0 warning: Lowest section in /lib/librt.so.1 is .dynamic at 00000074 warning: Lowest section in /lib/libdl.so.1 is .dynamic at 00000074 warning: Lowest section in /lib/libpthread.so.1 is .dynamic at 00000074 Program received signal SIGSEGV, Segmentation fault. 0x006b62d4 in createThread (cap=0x821ec0, size=256) at rts/Threads.c:113 113 tso->alloc_limit = 0; (gdb) }}} usually bus error is generated by unaligned access to data. I'll have a look at it, but any idea is of course highly appreciated. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/10043 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler