From: 白井彰 <okshirai@gmail.com>
To: Paul Eggert <eggert@cs.ucla.edu>
Cc: Akira Shirai <okshirai@gmail.com>, 26638@debbugs.gnu.org
Subject: bug#26638: 25.2; Fatal error 6: Abortmake[1]: *** [Makefile:737: bootstrap-emacs] Abort (core dumped)
Date: Thu, 28 Sep 2017 00:16:47 +0900 [thread overview]
Message-ID: <A60BD469-0BAC-45B5-8576-FF1B65BB7BCF@gmail.com> (raw)
In-Reply-To: <1083333f-45ba-1d2f-8277-366e44dd1ca1@cs.ucla.edu>
I cannot build emacs-25.2 on Solaris 10 Update 7 and Update 10 with Sun C 5.15,
but on Solaris 10 Update 11 I can build emacs-25.2 with Sun C 5.15 successfully.
Below is the debugger's output for temacs on Solaris 10 Update 7:
t@1 (l@1) program terminated by signal SEGV (no mapping at the fault address)
0xfe5a17ea: rt_dldump+0x07a6: cmpl $0x00000009,0x00000004(%eax)
Current function is unexec
18 if (! dldump (0, new_name, RTLD_MEMORY))
(dbx) down
(dbx) down
(dbx) down
0xfe5a17ea: rt_dldump+0x07a6: cmpl $0x00000009,0x00000004(%eax)
(dbx) where
current thread: t@1
=>[1] rt_dldump(0xfeffdd88, 0x8c0c21c, 0x1000, 0x0), at 0xfe5a17ea
[2] dldump_core(0xfeffb28c, 0x0, 0x8c0c21c, 0x1000), at 0xfefd9fe9
[3] _dldump(0x0, 0x8c0c21c, 0x1000, 0x8bd1b58), at 0xfefda036
[4] unexec(new_name = 0x8c0c21c "/tmp/emacs-25.2/src/emacs", old_name = 0x8c0c250 "/tmp/emacs-25.2/src/temacs"), line 18 in "unexsol.c"
[5] Fdump_emacs(filename = 146611036, symfile = 146610668), line 2115 in "emacs.c"
[6] eval_sub(form = 141049003), line 2175 in "eval.c"
[7] Fprogn(body = 141048979), line 431 in "eval.c"
[8] eval_sub(form = 140739195), line 2125 in "eval.c"
[9] Fif(args = 140739219), line 388 in "eval.c"
[10] eval_sub(form = 140749643), line 2125 in "eval.c"
[11] readevalloop(readcharfun = 11424, stream = 0x8621340, sourcename = 140810868, printflag = 0, unibyte = 0, readfun = 0, start = 0, end = 0), line 1927 in "lread.c"
[12] Fload(file = 140810740, noerror = 0, nomessage = 0, nosuffix = 0, must_suffix = 0), line 1336 in "lread.c"
[13] eval_sub(form = 140745611), line 2188 in "eval.c"
[14] Feval(form = 140745611, lexical = 0), line 1994 in "eval.c"
[15] top_level_2(), line 1121 in "keyboard.c"
[16] internal_condition_case(bfun = 0x82219c0 = &`temacs`keyboard.c`top_level_2(), handlers = 8712, hfun = 0x8221320 = &`temacs`keyboard.c`cmd_error(Lisp_Object data)), line 1315 in "eval.c"
[17] top_level_1(ignore = 0), line 1129 in "keyboard.c"
[18] internal_catch(tag = 21840, func = 0x8221a20 = &`temacs`keyboard.c`top_level_1(Lisp_Object ignore), arg = 0), line 1080 in "eval.c"
[19] command_loop(), line 1090 in "keyboard.c"
[20] recursive_edit_1(), line 697 in "keyboard.c"
[21] Frecursive_edit(), line 768 in "keyboard.c"
[22] main(argc = 5, argv = 0x8047ce0), line 1629 in "emacs.c"
(dbx) regs
current thread: t@1
current frame: [1]
eax 0x00000001
ebx 0xfe5b3000
ecx 0x00001741
edx 0x00000000
esi 0x08047ff2
edi 0xfed61730
ebp 0x080469c4
esp 0x08046794
trapno 0x0000000e 0x00000000
err 0x00000004 0x00000000
eflags 0x00010202 0x00000000
eip 0xfe5a17ea:rt_dldump+0x7a6 cmpl $0x00000009,0x00000004(%eax)
gs 0x000001c3 0x00000000
fs 0x00000043 0x00000000
es 0x00000000 0x00000000
ds 0x00000043 0x00000000
ss 0x00000043 0x00000000
cs 0x0000003b 0x00000000
(dbx) dis 0xfe5a17dc
0xfe5a17dc: rt_dldump+0x0798: cmpl $0x00000005,(%eax)
0xfe5a17df: rt_dldump+0x079b: je rt_dldump+0x812 [ 0xfe5a1856, .+0x77 ]
0xfe5a17e1: rt_dldump+0x079d: movl %eax,%edi
0xfe5a17e3: rt_dldump+0x079f: movl 0x00000008(%edi),%eax
0xfe5a17e6: rt_dldump+0x07a2: testl %eax,%eax
0xfe5a17e8: rt_dldump+0x07a4: je rt_dldump+0x809 [ 0xfe5a184d, .+0x65 ]
0xfe5a17ea: rt_dldump+0x07a6: cmpl $0x00000009,0x00000004(%eax)
0xfe5a17ee: rt_dldump+0x07aa: jne rt_dldump+0x809 [ 0xfe5a184d, .+0x5f ]
0xfe5a17f0: rt_dldump+0x07ac: cmpl $0x00000000,0x0000000c(%eax)
0xfe5a17f4: rt_dldump+0x07b0: je rt_dldump+0x809 [ 0xfe5a184d, .+0x59 ]
Below is librtld.so.1 for each Solaris 10:
Solaris 10 Update 7 -rwxr-xr-x 1 root bin 33884 Sep 14 2009 /lib/librtld.so.1
Solaris 10 Update 8 -rwxr-xr-x 1 root bin 33884 Sep 14 2009 /lib/librtld.so.1
Solaris 10 Update 9 -rwxr-xr-x 1 root bin 33884 Sep 14 2009 /lib/librtld.so.1
Solaris 10 Update 10 -rwxr-xr-x 1 root bin 41720 Jul 5 2011 /lib/librtld.so.1
Solaris 10 Update 11 -rwxr-xr-x 1 root bin 42036 Jan 14 2013 /lib/librtld.so.1
> 2017/09/15 4:14、Paul Eggert <eggert@cs.ucla.edu>のメール:
>
> 白井彰 wrote:
>
>> './configure --with-gif=no CFLAGS="-m64 -g3"' with Sun C 5.15,
>> 'CCLD temacs' failed on my site:
>> make[2]: Leaving directory '/tmp/emacs-25.2/admin/unidata'
>> CCLD temacs
>> ld: fatal: file /usr/sfw/lib/libfreetype.so: wrong ELF class: ELFCLASS32
>> ld: fatal: file /usr/sfw/lib/libXrender.so: wrong ELF class: ELFCLASS32
>> ld: fatal: file /usr/sfw/lib/libfreetype.so: wrong ELF class: ELFCLASS32
>> ld: fatal: File processing errors. No output written to temacs
>
> You'd need to link to 64-bit libraries. If you can't find them, then please disable the the troublesome 32-bitlibraries. E.g.,
>
> ./configure --with-gif=no --with-xft=no CFLAGS="-m64 -g3"
>
> should build without the abovementioned libraries.
>
>> Now I can recreate the segmentation fault of temacs every time with
>> Sun C 5.15 on my site, and the backtrace of temacs seems very alike
>> for each build of emacs-25.2:
>
> That's "encouraging", in some sense. At least you can reproduce the bug.
>
>> % cat /tmp/aaa.txt
>> core '/tmp/emacs-25.2/src/core' of 24956: ./temacs --batch --load loadup bootstrap
>> fe5a17ea rt_dldump (feffdd88, 8c0c21c, 1000, 0) + 7a6
>> fefd9fe9 dldump_core (feffb28c, 0, 8c0c21c, 1000) + 175
>> fefda036 dldump (0, 8c0c21c, 1000, 8bd1b58) + 42
>> 0834b719 unexec (8c0c21c, 8c0c250, 8046ac8, 8215a4d) + 19
>
> This sounds promising. Can you use a debugger to find out what is going wrong inside rt_dldump?
next prev parent reply other threads:[~2017-09-27 15:16 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-04-24 14:20 bug#26638: 25.2; Fatal error 6: Abortmake[1]: *** [Makefile:737: bootstrap-emacs] Abort (core dumped) 白井彰
2017-04-25 0:57 ` Paul Eggert
2017-04-25 15:50 ` 白井彰
2017-04-29 19:21 ` Paul Eggert
2017-09-02 12:31 ` Eli Zaretskii
2017-09-02 13:03 ` 白井彰
2017-09-02 18:18 ` Paul Eggert
2017-09-13 13:17 ` 白井彰
2017-09-13 15:26 ` Paul Eggert
2017-09-14 15:11 ` 白井彰
2017-09-14 19:14 ` Paul Eggert
2017-09-27 15:16 ` 白井彰 [this message]
2017-10-02 6:18 ` Paul Eggert
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://www.gnu.org/software/emacs/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=A60BD469-0BAC-45B5-8576-FF1B65BB7BCF@gmail.com \
--to=okshirai@gmail.com \
--cc=26638@debbugs.gnu.org \
--cc=eggert@cs.ucla.edu \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://git.savannah.gnu.org/cgit/emacs.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).