Re: [Boost-build] [EXTERNAL] Re: Boost build vacpp on Linux (PPC)

Subject: Re: [Boost-build] [EXTERNAL] Re: Boost build vacpp on Linux (PPC)
From: Hubert Tong (hstong_at_[hidden])
Date: 2012-03-29 01:31:47

I have confirmed that compiling function.c with optimization is the key to
hitting the issue.
It seems argv/envp can affect how the failure presents itself.

Using xlc option -qinitauto=00 (which will zero otherwise uninitialized
stack variables), I can get:
> bin.aixppc/b2 --help
jambase.c:76: in module scope
*** argument error
* rule find-to-root (  :  )
* called with:
( /home/hstong/b2011/boost-trunk-2012-03-25-12-03-24-EDT/tools/build/v2/engine :
 boost-build.jam )
* extra
argument /home/hstong/b2011/boost-trunk-2012-03-25-12-03-24-EDT/tools/build/v2/engine
jambase.c:11:see definition of rule 'find-to-root' being called
Return:  0x01:1

Here is the debugger trace without -qinitauto=00 (I replaced struct _object
with char to get better output):
Segmentation fault in list_end at line 120 in file "lists.c"
  120           return list_begin( l ) + l->impl.size;
(dbx) where
list_end(l = 0xdeadbeef), line 120 in "lists.c"
type_check(type_name = "", values = (nil), caller = 0x00000020, called =
0x0000c128, arg_name = (nil)), line 2686 in "function.c"
unnamed block in argument_list_push(formal = 0x2ff21640, formal_count = 0,
function = 0x2ff212a0, frame = (nil), s = 0xdeadbeef), line 2809 in
"function.c"
unnamed block in argument_list_push(formal = 0x2ff21640, formal_count = 0,
function = 0x2ff212a0, frame = (nil), s = 0xdeadbeef), line 2809 in
"function.c"
argument_list_push(formal = 0x2ff21640, formal_count = 0, function =
0x2ff212a0, frame = (nil), s = 0xdeadbeef), line 2809 in "function.c"
function_run(function_ = 0x2000e0e0, frame = 0x200046e0, s = 0x2ff214d0),
line 3394 in "function.c"
unnamed block in evaluate_rule(rulename = "find-to-root", frame =
0x2ff21640), line 264 in "compile.c"
evaluate_rule(rulename = "find-to-root", frame = 0x2ff21640), line 264 in
"compile.c"
function_call_rule(function = 0x2000e0e0, frame = 0x2000c128, s =
0x2ff21740, n_args = -265522048, unexpanded = "\200A", file = (nil), line =
22), line 451 in "function.c"
function_run(function_ = 0x2007f888, frame = 0x10030b24, s = 0x10030b54),
line 3938 in "function.c"
unnamed block in parse_file(f = "+", frame = 0x2ff22248), line 57 in
"parse.c"
parse_file(f = "+", frame = 0x2ff22248), line 57 in "parse.c"
unnamed block in main(argc = 0, argv = 0x2ff22418, arg_environ =
0x2ff2241c), line 542 in "jam.c"
unnamed block in main(argc = 0, argv = 0x2ff22418, arg_environ =
0x2ff2241c), line 542 in "jam.c"
main(argc = 0, argv = 0x2ff22418, arg_environ = 0x2ff2241c), line 542 in
"jam.c"
(dbx) frame 8
function_call_rule(function = 0x2000e0e0, frame = 0x2000c128, s =
0x2ff21740, n_args = -265522048, unexpanded = "\200A", file = (nil), line =
22), line 451 in "function.c"
(dbx) print *function
(base = (type = 1718185572, reference_count = 762605357, rulename =
(invalid char ptr (0x726f6f74)), formal_arguments = (nil),
num_formal_arguments = 1684644916), code_size = 0, code = 0x64697200,
num_constants = 61487419, constants = (nil), num_subfunctions = 1735159650,
functions = 0x73000000, num_subactions = 1569788835, actions = 0x20007744,
generic = 0x24286469, file = (invalid char ptr (0x72290000)), line =
-1411044943)
(dbx)

Thanks,

Hubert Tong

Hubert S K Tong
IBM Canada Ltd, C2/YGH/8200/MKM
8200 Warden Ave, Markham ON  L6G 1C7
Canada
E-mail:  hstong_at_[hidden]

From:	Steven Watanabe <watanabesj_at_[hidden]>
To:	"Boost.Build developer's and user's list"
            <boost-build_at_[hidden]>
Date:	28-03-2012 09:55 PM
Subject:	Re: [Boost-build] [EXTERNAL] Re: Boost build vacpp on Linux
            (PPC)
Sent by:	boost-build-bounces_at_[hidden]

AMDG

On 03/28/2012 02:33 PM, Belcourt, K. Noel wrote:
> Hi Hubert,
>
> Looks like the Sandia-vacpp-10.1 tester ran last night using gcc to
> build bjam.  It's beginning to look like a problem with xlc at high
> optimization, does that sound plausible?
>

If it's an optimization problem, it could be a
manifestation of undefined behavior.  I'd really
like to track down the problem anyway, even
if reducing optimization makes it go away.

In Christ,
Steven Watanabe
_______________________________________________
Unsubscribe & other changes:
http://lists.boost.org/mailman/listinfo.cgi/boost-build