unofficial mirror of bug-gnu-emacs@gnu.org 
 help / color / mirror / code / Atom feed
* bug#41680: Recent compiled byte code changes broke loading
@ 2020-06-03  0:26 Daniel Colascione
  2020-06-03  1:43 ` Paul Eggert
  0 siblings, 1 reply; 2+ messages in thread
From: Daniel Colascione @ 2020-06-03  0:26 UTC (permalink / raw)
  To: 41680; +Cc: eggert

f0b0105d913a94c66f230874c9269b19dbbc83bd changes how we generate and
evaluate byte-code objects. It causes a weird wrong-type-argument when
loading some elc files, however:

Debugger entered--Lisp error: (wrong-type-argument stringp
("/home/dancol/emacs/conf/gnus.elc" . 1556))
  read(get-file-char)
  load("~/emacs/conf/gnus" nil t)
  (lambda nil (load "~/emacs/conf/gnus" nil t))()
  eval-after-load-helper("/home/dancol/edev/debug/lisp/gnus/gnus.elc")
  run-hook-with-args(eval-after-load-helper
"/home/dancol/edev/debug/lisp/gnus/gnus.elc")
  do-after-load-evaluation("/home/dancol/edev/debug/lisp/gnus/gnus.elc")

I did a bit of debugging. We're passing some kind of cons cell to the
Fstring_as_unibyte call in read1. (The CDR seems invalid too somehow?)

There's a difference in the generated elc file ("gnus.elc" above) that
seems to drive the error.

Bad: (defalias 'qtmstr-setup-summary-mode #[nil #@38 ("\300\301!\207" .
[hl-line-mode 1])\x1f(#$ . 1556) nil 2])

Good: (defalias 'qtmstr-setup-summary-mode #[nil "\300\301!\207"
[hl-line-mode 1] 2])

Reverting f0b0105d913a94c66f230874c9269b19dbbc83bd seems to fix the
problem for now.






^ permalink raw reply	[flat|nested] 2+ messages in thread

* bug#41680: Recent compiled byte code changes broke loading
  2020-06-03  0:26 bug#41680: Recent compiled byte code changes broke loading Daniel Colascione
@ 2020-06-03  1:43 ` Paul Eggert
  0 siblings, 0 replies; 2+ messages in thread
From: Paul Eggert @ 2020-06-03  1:43 UTC (permalink / raw)
  To: Daniel Colascione; +Cc: 41680-done

[-- Attachment #1: Type: text/plain, Size: 122 bytes --]

Thanks for reporting that bug. I installed the attached patch, which should fix
it, and am boldly closing the bug report.

[-- Attachment #2: 0001-Fix-bug-in-recent-byte-code-checking-hoist.patch --]
[-- Type: text/x-patch, Size: 2447 bytes --]

From 00613f4c4a464c16d6d50a34e1b8b68fc50f2d2e Mon Sep 17 00:00:00 2001
From: Paul Eggert <eggert@cs.ucla.edu>
Date: Tue, 2 Jun 2020 18:40:10 -0700
Subject: [PATCH] Fix bug in recent byte-code checking hoist

Problem reported by Daniel Colascione (Bug#41680).
* src/lread.c (read1): Check that AREF (tmp, COMPILED_BYTECODE)
is a string before subjecting it to STRING_MULTIBYTE.
Be more consistent about using AREF in the neighborhood,
to help prevent this sort of problem from recurring.
---
 src/lread.c | 21 +++++++++++----------
 1 file changed, 11 insertions(+), 10 deletions(-)

diff --git a/src/lread.c b/src/lread.c
index 29deddaf15..8064bf4d0e 100644
--- a/src/lread.c
+++ b/src/lread.c
@@ -2966,17 +2966,18 @@ read1 (Lisp_Object readcharfun, int *pch, bool first_in_list)
 	  struct Lisp_Vector *vec;
 	  tmp = read_vector (readcharfun, 1);
 	  vec = XVECTOR (tmp);
-	  if (! (COMPILED_STACK_DEPTH < vec->header.size
-		 && (FIXNUMP (vec->contents[COMPILED_ARGLIST])
-		     || CONSP (vec->contents[COMPILED_ARGLIST])
-		     || NILP (vec->contents[COMPILED_ARGLIST]))
-		 && ((STRINGP (vec->contents[COMPILED_BYTECODE])
-		      && VECTORP (vec->contents[COMPILED_CONSTANTS]))
-		     || CONSP (vec->contents[COMPILED_BYTECODE]))
-		 && FIXNATP (vec->contents[COMPILED_STACK_DEPTH])))
+	  if (! (COMPILED_STACK_DEPTH < ASIZE (tmp)
+		 && (FIXNUMP (AREF (tmp, COMPILED_ARGLIST))
+		     || CONSP (AREF (tmp, COMPILED_ARGLIST))
+		     || NILP (AREF (tmp, COMPILED_ARGLIST)))
+		 && ((STRINGP (AREF (tmp, COMPILED_BYTECODE))
+		      && VECTORP (AREF (tmp, COMPILED_CONSTANTS)))
+		     || CONSP (AREF (tmp, COMPILED_BYTECODE)))
+		 && FIXNATP (AREF (tmp, COMPILED_STACK_DEPTH))))
 	    invalid_syntax ("Invalid byte-code object");
 
-	  if (STRING_MULTIBYTE (AREF (tmp, COMPILED_BYTECODE)))
+	  if (STRINGP (AREF (tmp, COMPILED_BYTECODE))
+	      && STRING_MULTIBYTE (AREF (tmp, COMPILED_BYTECODE)))
 	    {
 	      /* BYTESTR must have been produced by Emacs 20.2 or earlier
 		 because it produced a raw 8-bit string for byte-code and
@@ -2987,7 +2988,7 @@ read1 (Lisp_Object readcharfun, int *pch, bool first_in_list)
 		    Fstring_as_unibyte (AREF (tmp, COMPILED_BYTECODE)));
 	    }
 
-	  if (COMPILED_DOC_STRING < vec->header.size
+	  if (COMPILED_DOC_STRING < ASIZE (tmp)
 	      && EQ (AREF (tmp, COMPILED_DOC_STRING), make_fixnum (0)))
 	    {
 	      /* read_list found a docstring like '(#$ . 5521)' and treated it
-- 
2.17.1


^ permalink raw reply related	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2020-06-03  1:43 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2020-06-03  0:26 bug#41680: Recent compiled byte code changes broke loading Daniel Colascione
2020-06-03  1:43 ` Paul Eggert

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).