The original segfault that caused this ticket is closed, I think, but
tcl is still segfaulting on darwin. I think there's another ticket
open; feel free to merge them together.
$ sw_vers
ProductName: Mac OS X
ProductVersion: 10.4.10
BuildVersion: 8R2218
gcc version 4.0.1 (Apple Computer, Inc. build 5363)
svn revision 21166
backtrace on t/cmd_after.t:
Program received signal EXC_BAD_ACCESS, Could not access
memory.Reason: KERN_INVALID_ADDRESS at address: 0x00ec8d14
0x00094e20 in ascii_compare (interp=0x31003b0, lhs=0x35416f0,
rhs=0x189e850) at src/charset/ascii.c:328
328 const int ret_val = memcmp(lhs->strstart, rhs-
>strstart, min_len);
(gdb) bt#0 0x00094e20 in ascii_compare (interp=0x31003b0,
lhs=0x35416f0, rhs=0x189e850) at src/charset/ascii.c:328#1
0x00004572 in string_equal (interp=0x31003b0, s1=0x35416f0,
s2=0x189e850) at src/string.c:1327#2 0x000c354f in STRING_compare
(interp=0x31003b0, search_key=0x35416f0, bucket_key=0x189e850) at src/
hash.c:142
#3 0x000c43e6 in parrot_hash_get_bucket (interp=0x31003b0,
hash=0x3113c30, key=0x35416f0) at src/hash.c:849
#4 0x001aa9e3 in Parrot_Hash_exists_keyed_str (interp=0x31003b0,
pmc=0x188a680, key=0x35416f0) at ./src/pmc/hash.pmc:879
#5 0x000c043a in is_loaded (interp=0x31003b0, path=0x35416f0) at src/
dynext.c:144 #6 0x000c0f57 in Parrot_load_lib (interp=0x31003b0,
lib=0x3522b98, initializer_unused=0x0) at src/dynext.c:471#7
0x0021fbb8 in do_loadlib (interp=0x31003b0, lib=0x31b8ff0
"'tcl_ops'") at compilers/imcc/imcc.y:378#8 0x00221090 in yyparse
(yyscanner=0x319a0a0, interp=0x31003b0) at compilers/imcc/imcc.y:503
#9 0x0022cf55 in compile_string (interp=0x31003b0, s=0x28a2c00 ".HLL
'Tcl', ''\n.loadlib 'tcl_ops'\n.namespace \n.include 'languages/tcl/
src/returncodes.pir'\n.sub '_anon' :anon\n .local pmc colons,
split, epoch\n colons = get_root_global ['_tcl'], 'colons'\n
sp"..., yyscanner=0x319a0a0) at compilers/imcc/imcc.l:1172
#10 0x000b7ce6 in imcc_compile (interp=0x31003b0, s=0x28a2c00 ".HLL
'Tcl', ''\n.loadlib 'tcl_ops'\n.namespace \n.include 'languages/tcl/
src/returncodes.pir'\n.sub '_anon' :anon\n .local pmc colons,
split, epoch\n colons = get_root_global ['_tcl'], 'colons'\n
sp"..., pasm_file=0, error_message=0xbfffee84) at compilers/imcc/
parser_util.c:725#11 0x000b7fe7 in imcc_compile_pir_ex
(interp=0x31003b0, s=0x28a2c00 ".HLL 'Tcl', ''\n.loadlib
'tcl_ops'\n.namespace \n.include 'languages/tcl/src/
returncodes.pir'\n.sub '_anon' :anon\n .local pmc colons, split,
epoch\n colons = get_root_global ['_tcl'], 'colons'\n sp"...)
at compilers/imcc/parser_util.c:835#12 0x000d1357 in pcf_P_Jt
(interp=0x31003b0, self=0x1889eb8) at src/nci.c:2645#13 0x000a3e0e in
Parrot_NCI_invoke (interp=0x31003b0, pmc=0x1889eb8, next=0x321f728)
at ./src/pmc/nci.pmc:168 #14 0x001607cf in Parrot_Compiler_invoke
(interp=0x31003b0, pmc=0x1889eb8, code_ptr=0x321f728) at ./src/pmc/
compiler.pmc:38#15 0x0003426b in Parrot_invokecc_p
(cur_opcode=0x321f720, interp=0x31003b0) at src/ops/core.ops:424
#16 0x000e575e in runops_slow_core (interp=0x31003b0, pc=0x321f720)
at src/runops_cores.c:191#17 0x000c60c3 in runops_int
(interp=0x31003b0, offset=3) at src/interpreter.c:818
#18 0x000146a6 in runops (interp=0x31003b0, offs=3) at src/
inter_run.c:99
#19 0x00014948 in runops_args (interp=0x31003b0, sub=0x1889138,
obj=0x282dc60, meth_unused=0x0, sig=0x23a490 "vP", ap=0xbffff15c "?
~O~H\001?I\021\003L?\n") at src/inter_run.c:213
#20 0x00014a88 in Parrot_runops_fromc_args (interp=0x31003b0,
sub=0x1889138, sig=0x23a490 "vP") at src/inter_run.c:290
#21 0x0000867f in Parrot_runcode (interp=0x31003b0, argc=2,
argv=0xbffff2b8) at src/embed.c:790#22 0x000af44c in imcc_run_pbc
(interp=0x31003b0, obj_file=0, output_file=0x0, argc=2,
argv=0xbffff2b8) at compilers/imcc/main.c:627#23 0x000afed7 in
imcc_run (interp=0x31003b0, sourcefile=0xbffff36b "tcl.pbc", argc=2,
argv=0xbffff2b8) at compilers/imcc/main.c:853#24 0x0000244e in main
(argc=2, argv=0xbffff2b8) at src/main.c:60
On Do. 14. Jun. 2007, 19:10:02, coke wrote:
With recent updates, tcl suite is MUCH faster, but is now segfaulting
in several places. No clue when it last worked.
Hi Coke,
it looks like tcl is working again. Can I close the ticket?
Regards,
Bernhard
--
/* [EMAIL PROTECTED] */
--
Will "Coke" Coleda
[EMAIL PROTECTED]
