From xemacs-m  Sat Mar  1 15:03:49 1997
Received: from mharnois.workgroup.net (pm1-039.sbt.net [206.153.211.39])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id PAA21725
	for <xemacs-beta@xemacs.org>; Sat, 1 Mar 1997 15:03:42 -0600 (CST)
Received: (from mharnois@localhost)
	by mharnois.workgroup.net (8.8.5/8.8.5) id PAA21369;
	Sat, 1 Mar 1997 15:03:52 -0600
Sender: mharnois@sbt.net
To: xemacs-beta@xemacs.org
Subject: crash
Mail-Copies-To: never
X-Face: #ELVc'\6VfS2I%PkOK{H2d%M=]`IfN&$$Wo[x14,.sRozWJ3/dr~cV,0hp}BF%+x0=TYyfE
 B8O;Nrz=1@.mIh_BT>:2N,K|f0)"sy%=AyU1MyXv3)joP%%)IF~+"#!*SRa+hbyFX>?^{f|_Wq/Zs.
 \/@8PsO^^f,p86SJIFi9-6+Tw}?Wc6{i*8Jd^"m^INcjf~0o=Nvrd4XX\u`:0tKebsYKXrtp!v)"OD
 -o2_Jm(M2up1A
X-Attribution: mdh
Mime-Version: 1.0 (generated by tm-edit 7.105)
Content-Type: text/plain; charset=X-CTEXT
From: Michael Harnois <mharnois@sbt.net>
Date: 01 Mar 1997 15:03:47 -0600
Message-ID: <877mjr72a4.fsf@sbt.net>
Lines: 134
X-Mailer: Gnus v5.4.16/XEmacs 20.1

Well, since David seems to be around today (and Steve doesn't), I'll
throw this out to the whole list.

I've compiled with mule on linux-2.1.27, and glibc-2.0.1. Everything works
except for one problem. When attempting to read a message in gnus with
japanese in it, I get 

 Fatal error: assertion failed, file regex.c, line 4060, __dcptr__ - __dcptr2__ == REP_BYTES_BY_FIRST_BYTE (*__dcptr2__)

The lisp backtrace is

  re-search-backward("\n--+ *\n" 1 t)
  # (unwind-protect ...)
  # (unwind-protect ...)
  # (unwind-protect ...)
  # bind (hend p e real-end too-big hack-sig end start)
  highlight-headers(1 1462 t)
  mime-preview/x-face-function-use-highlight-headers()
  run-hooks(mime-viewer/content-header-filter-hook)
  # (unwind-protect ...)
  # bind (end beg)
  mime-preview/display-header(1 1496)
  # bind (subj ne nb e he rcnum encoding params ctype end beg obuf ibuf cinfo content)
  mime-preview/display-content([mime::content-info nil 1 1644 "text/plain" (("charset" . "US-ASCII")) nil nil] [mime::content-info nil 1 1644 "text/plain" (("charset" . "US-ASCII")) nil nil] #<buffer " *Original Article*"> "*Article*")
  # bind (drest mode the-buf dest pcl cinfo obuf)
  mime-viewer/make-preview-buffer("*Article*")
  # bind (obuf ibuf encoding ctl)
  mime-viewer/setup-buffer(nil nil " *Original Article*" "*Article*")
  # bind (mother-keymap obuf ibuf encoding ctl mother)
  mime/viewer-mode(nil nil nil " *Original Article*" "*Article*" #<keymap 354 entries 0x126c>)
  byte-code("\n

                   &e" [mime/viewer-mode nil gnus-original-article-buffer gnus-article-buffer gnus-article-mode-map] 7)
  # (unwind-protect ...)
  # bind (mime-viewer/ignored-field-regexp default-mime-charset)
  gnus-article-preview-mime-message()
  # bind (buffer-read-only)
  # (unwind-protect ...)
  # bind (result group internal-hook summary-buffer gnus-article)
  # (unwind-protect ...)
  # bind (header all-headers article)
  gnus-article-prepare(1926 nil)
  # bind (all-header article)
  gnus-summary-display-article(1926 nil)
  # (unwind-protect ...)
  # bind (article all-headers gnus-summary-display-article-function did article pseudo force all-headers)
  gnus-summary-select-article(nil nil pseudo)
  # bind (lines)
  gnus-summary-scroll-up(1)
  # bind (command-debug-status)
  call-interactively(gnus-summary-scroll-up)
  # (condition-case ... . error)
  # (catch top-level ...)

The gdb trace starts out 

#0  0x402ad991 in ?? () from /lib/libc-2.0.1.so
#1  0x808d8fe in fatal_error_signal (sig=6) at emacs.c:200
#2  0xbfffc728 in ?? ()
#3  0x402ae84f in ?? () from /lib/libc-2.0.1.so
#4  0x808f5ae in assert_failed (file=0x81fb186 "regex.c", line=4060, 
    expr=0x81fb6d5 "__dcptr__ - __dcptr2__ == REP_BYTES_BY_FIRST_BYTE (*__dcptr2__)") at emacs.c:2201
#5  0x81391ad in sys_re_search_2 (bufp=0x82ff840, 
    string1=0x40382000 "Received: from mailhost.sbt.net (mharnois@localhost [127.0.0.1])\n          by mharnois.workgroup.net (8.8.4/8.8.4) with SMTP\n\t  id QAA08630 for <mharnois>; Tue, 21 Jan 1997 16:27:44 -0600\nReceived: fr"..., 
    size1=1070, 
    string2=0x40382caf "@bob.sbt.net\" <steve@miranova.com>\nIn-Reply-To: Michael Harnois's message of 21 Jan 1997 15:17:52 -0600\nMime-Version: 1.0 (generated by tm-edit 7.100)\nContent-Type: text/plain; charset=US-ASCII\nDate: "..., 
    size2=396, startpos=1465, range=-1465, regs=0x82da88c, stop=1465)
    at regex.c:4060
#6  0x8142a3f in search_buffer (buf=0x8889c00, string=943230660, bufpos=1462, 
    buflim=1, n=-1, RE=1, trt=0x8327214 "", inverse_trt=0x8327414 "", posix=0)
    at search.c:1123
#7  0x81423fd in search_command (string=943230660, bound=1, 
    no_error=405839900, count=405839876, buffer=405839876, direction=-1, RE=1, 
    posix=0) at search.c:981
#8  0x8144074 in Fre_search_backward (regexp=943230660, bound=1, 
    no_error=405839900, count=405839876, buffer=405839876) at search.c:1679
#9  0x80947d5 in primitive_funcall (fn=0x8144050 <Fre_search_backward>, 
    nargs=5, args=0xbfffca84) at eval.c:3460
#10 0x8094978 in funcall_subr (subr=0x82cf100, args=0xbfffca84) at eval.c:3481
#11 0x8093ee0 in funcall_recording_as (recorded_as=405946340, nargs=4, 
    args=0xbfffd1ac) at eval.c:3173
#12 0x8093fd3 in Ffuncall (nargs=4, args=0xbfffd1ac) at eval.c:3217
#13 0x8063516 in Fbyte_code (bytestr=943704908, vector=1215971072, maxdepth=6)
    at bytecode.c:418
#14 0x8094d77 in funcall_lambda (fun=410849124, nargs=3, arg_vector=0xbfffd2f8)
    at eval.c:3591
#15 0x8093f8e in funcall_recording_as (recorded_as=406899508, nargs=4, 
    args=0xbfffd2f4) at eval.c:3189
#16 0x8093fd3 in Ffuncall (nargs=4, args=0xbfffd2f4) at eval.c:3217
#17 0x8063516 in Fbyte_code (bytestr=942977904, vector=1214652032, maxdepth=6)
    at bytecode.c:418
#18 0x8094d77 in funcall_lambda (fun=409071460, nargs=0, arg_vector=0xbfffd478)
    at eval.c:3591
#19 0x8093f8e in funcall_recording_as (recorded_as=409349156, nargs=1, 
    args=0xbfffd474) at eval.c:3189
#20 0x8093fd3 in Ffuncall (nargs=1, args=0xbfffd474) at eval.c:3217
#21 0x8095465 in run_hook_with_args_in_buffer (buf=0x8889c00, nargs=1, 
    args=0xbfffd474, cond=RUN_HOOKS_TO_COMPLETION) at eval.c:3772
#22 0x80954ea in run_hook_with_args (nargs=1, args=0xbfffd474, 
    cond=RUN_HOOKS_TO_COMPLETION) at eval.c:3785
#23 0x80950eb in Frun_hooks (nargs=1, args=0xbfffd594) at eval.c:3648
#24 0x8093e85 in funcall_recording_as (recorded_as=405924300, nargs=2, 
    args=0xbfffd590) at eval.c:3156
#25 0x8093fd3 in Ffuncall (nargs=2, args=0xbfffd590) at eval.c:3217
#26 0x8063516 in Fbyte_code (bytestr=945908604, vector=1214056640, maxdepth=4)
    at bytecode.c:418
#27 0x8094d77 in funcall_lambda (fun=408776932, nargs=2, arg_vector=0xbfffdae8)
    at eval.c:3591
#28 0x8093f8e in funcall_recording_as (recorded_as=408727332, nargs=3, 
    args=0xbfffdae4) at eval.c:3189
#29 0x8093fd3 in Ffuncall (nargs=3, args=0xbfffdae4) at eval.c:3217
#30 0x8063516 in Fbyte_code (bytestr=945907968, vector=1214209280, maxdepth=10)
    at bytecode.c:418
#31 0x8094d77 in funcall_lambda (fun=408776900, nargs=4, arg_vector=0xbfffddd4)
    at eval.c:3591
#32 0x8093f8e in funcall_recording_as (recorded_as=408727188, nargs=5, 
    args=0xbfffddd0) at eval.c:3189
#33 0x8093fd3 in Ffuncall (nargs=5, args=0xbfffddd0) at eval.c:3217
#34 0x8063516 in Fbyte_code (bytestr=945796900, vector=1214043136, maxdepth=6)
    at bytecode.c:418
#35 0x8094d77 in funcall_lambda (fun=408776868, nargs=1, arg_vector=0xbfffdf60)
    at eval.c:3591

I have recompiled a number of the shared libraries that xemacs used
against glibc to see if it would help. Last night I also tried
building xemacs with a static libc6 that had the gnu regex.o removed
to see if there was a conflict there. That made no difference
either. The crash is 100% repeatable on my machine.

--
+  Michael D. Harnois              +    If you want to follow Jesus,     +
+  Pastor, computer nerd,          +    you better look good on wood.    +
+  Linux user, Havanese owner      +                                     + 
+  mharnois@sbt.net                +            -- Daniel Berrigan       +

