Loading ...
Sorry, an error occurred while loading the content.
 

Re: Test 58 failure on solaris (64bit)

Expand Messages
  • harpchad
    with -g and no optimization: = [1] getout(exitval = 1), line 1280 in main.c [2] deathtrap(sigarg = 11), line 1051 in os_unix.c [3] sigacthandler(0xb, 0x0,
    Message 1 of 11 , Jan 28, 2009
      with -g and no optimization:

      =>[1] getout(exitval = 1), line 1280 in "main.c"
      [2] deathtrap(sigarg = 11), line 1051 in "os_unix.c"
      [3] sigacthandler(0xb, 0x0, 0x100414900, 0xb, 0x0, 0x0), at
      0x7ffefe8a7240
      ---- called from signal handler with signal 11 (SIGSEGV) ------
      [4] mf_sync(mfp = 0x10042e2a0, flags = 4), line 619 in "memfile.c"
      [5] ml_sync_all(check_file = 0, check_char = 0), line 1924 in
      "memline.c"
      [6] preserve_exit(), line 8362 in "misc1.c"
      [7] deathtrap(sigarg = 10), line 1060 in "os_unix.c"
      [8] sigacthandler(0xa, 0x0, 0x100415100, 0xa, 0x0, 0x10044a8cc), at
      0x7ffefe8a7240
      ---- called from signal handler with signal 10 (SIGBUS) ------
      [9] hash_init(ht = 0x10044a7a4), line 72 in "hashtab.c"
      [10] spell_read_aff(spin = 0xffffffff7fffca08, fname =
      0xffffffff7fffd1c4 "Xtest.aff"), line 5328 in "spell.c"
      [11] mkspell(fcount = 2, fnames = 0x100440ec0, ascii = 0, overwrite
      = 1, added_word = 0), line 9362 in "spell.c"
      [12] ex_mkspell(eap = 0xffffffff7fffd7d0), line 8685 in "spell.c"
      [13] do_one_cmd(cmdlinep = 0xffffffff7fffdf40, sourcing = 1, cstack
      = 0xffffffff7fffda88, fgetline = 0x1000d3fd8 = &get_func_line(),
      cookie = 0x100425070), line 2622 in "ex_docmd.c"
      [14] do_cmdline(cmdline = (nil), getline = 0x1000d3fd8 =
      &get_func_line(), cookie = 0x100425070, flags = 7), line 1100 in
      "ex_docmd.c"
      [15] call_user_func(fp = 0x10042b540, argcount = 2, argvars =
      0xffffffff7fffe580, rettv = 0xffffffff7fffe830, firstline = 1,
      lastline = 1, selfdict = (nil)), line 21352 in "eval.c"
      [16] call_func(name = 0x1004233b0 "TestOne", len = 7, rettv =
      0xffffffff7fffe830, argcount = 2, argvars = 0xffffffff7fffe580,
      firstline = 1, lastline = 1, doesrange = 0xffffffff7fffe824, evaluate
      = 1, selfdict = (nil)), line 8071 in "eval.c"
      [17] get_func_tv(name = 0x1004233b0 "TestOne", len = 7, rettv =
      0xffffffff7fffe830, arg = 0xffffffff7fffe860, firstline = 1, lastline
      = 1, doesrange = 0xffffffff7fffe824, evaluate = 1, selfdict = (nil)),
      line 7917 in "eval.c"
      [18] ex_call(eap = 0xffffffff7fffe980), line 3333 in "eval.c"
      [19] do_one_cmd(cmdlinep = 0xffffffff7ffff0f0, sourcing = 0, cstack
      = 0xffffffff7fffec38, fgetline = 0x100119050 = &getexline(), cookie =
      (nil)), line 2622 in "ex_docmd.c"
      [20] do_cmdline(cmdline = (nil), getline = 0x100119050 = &getexline
      (), cookie = (nil), flags = 0), line 1100 in "ex_docmd.c"
      [21] nv_colon(cap = 0xffffffff7ffff308), line 5234 in "normal.c"
      [22] normal_cmd(oap = 0xffffffff7ffff428, toplevel = 1), line 1200
      in "normal.c"
      [23] main_loop(cmdwin = 0, noexmode = 0), line 1180 in "main.c"
      [24] main(argc = 9, argv = 0xffffffff7ffff768), line 939 in "main.c"

      Trying patch now
      --~--~---------~--~----~------------~-------~--~----~
      You received this message from the "vim_dev" maillist.
      For more information, visit http://www.vim.org/maillist.php
      -~----------~----~----~----~------~----~------~--~---
    • Dominique Pelle
      ... I think that s because the debugger did not know how many arguments the function has (because Vim was compiled without debug option), so debugger just
      Message 2 of 11 , Jan 28, 2009
        Bram Moolenaar wrote:

        >> =>[1] hash_init(0x100313ac4, 0x3b0, 0x1003139fc, 0x440, 0x90,
        >> 0x100313aec), at 0x1000bbd00
        >
        > hash_init() only has one argument, I wonder what the numbers mean.


        I think that's because the debugger did not know how
        many arguments the function has (because Vim was
        compiled without debug option), so debugger just prints
        first n registers that the ABI uses to for passing arguments.
        0x100313ac4 was the unique relevant parameter here.

        I see that debugger output when compiled with debug
        is a lot better:

        > [9] hash_init(ht = 0x10044a7a4), line 72 in "hashtab.c"

        Cheers
        -- Dominique

        --~--~---------~--~----~------------~-------~--~----~
        You received this message from the "vim_dev" maillist.
        For more information, visit http://www.vim.org/maillist.php
        -~----------~----~----~----~------~----~------~--~---
      • harpchad
        With patch all test now complete successfully on sparcv9 as well. Thanks for the quick patch :) --~--~---------~--~----~------------~-------~--~----~ You
        Message 3 of 11 , Jan 28, 2009
          With patch all test now complete successfully on sparcv9 as well.

          Thanks for the quick patch :)
          --~--~---------~--~----~------------~-------~--~----~
          You received this message from the "vim_dev" maillist.
          For more information, visit http://www.vim.org/maillist.php
          -~----------~----~----~----~------~----~------~--~---
        Your message has been successfully submitted and would be delivered to recipients shortly.