all messages for Emacs-related lists mirrored at yhetil.org
 help / color / mirror / code / Atom feed
* Re: [Emacs-diffs] /srv/bzr/emacs/emacs-24 r108098: * admin/bzrmerge.el (bzrmerge-resolve): Disable local eval:.
       [not found] <E1SzjTX-0001sf-8w@vcs.savannah.gnu.org>
@ 2012-08-10 15:26 ` Stefan Monnier
  2012-08-10 15:51   ` Glenn Morris
  0 siblings, 1 reply; 5+ messages in thread
From: Stefan Monnier @ 2012-08-10 15:26 UTC (permalink / raw)
  To: Glenn Morris; +Cc: emacs-devel

> -      (with-current-buffer (let ((enable-local-variables :safe))
> +      (with-current-buffer (let ((enable-local-variables :safe)
> +                                 (enable-local-eval nil))

Why are those changes needed?
Shouldn't (enable-local-variables :safe) be sufficient?


        Stefan



^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [Emacs-diffs] /srv/bzr/emacs/emacs-24 r108098: * admin/bzrmerge.el (bzrmerge-resolve): Disable local eval:.
  2012-08-10 15:26 ` [Emacs-diffs] /srv/bzr/emacs/emacs-24 r108098: * admin/bzrmerge.el (bzrmerge-resolve): Disable local eval: Stefan Monnier
@ 2012-08-10 15:51   ` Glenn Morris
  2012-08-10 20:29     ` Stefan Monnier
  0 siblings, 1 reply; 5+ messages in thread
From: Glenn Morris @ 2012-08-10 15:51 UTC (permalink / raw)
  To: Stefan Monnier; +Cc: emacs-devel

Stefan Monnier wrote:

>> -      (with-current-buffer (let ((enable-local-variables :safe))
>> +      (with-current-buffer (let ((enable-local-variables :safe)
>> +                                 (enable-local-eval nil))
>
> Why are those changes needed?

Paranoia.

> Shouldn't (enable-local-variables :safe) be sufficient?

Yes. Unless there is another screw-up like

http://debbugs.gnu.org/cgi/bugreport.cgi?bug=12155

BTW, there have been zero other comments on that. Emacs releases have
happened in similar cases in the past, IIUC.



^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [Emacs-diffs] /srv/bzr/emacs/emacs-24 r108098: * admin/bzrmerge.el (bzrmerge-resolve): Disable local eval:.
  2012-08-10 15:51   ` Glenn Morris
@ 2012-08-10 20:29     ` Stefan Monnier
  2012-08-10 20:42       ` Glenn Morris
  0 siblings, 1 reply; 5+ messages in thread
From: Stefan Monnier @ 2012-08-10 20:29 UTC (permalink / raw)
  To: Glenn Morris; +Cc: emacs-devel

>>> -      (with-current-buffer (let ((enable-local-variables :safe))
>>> +      (with-current-buffer (let ((enable-local-variables :safe)
>>> +                                 (enable-local-eval nil))
>> Why are those changes needed?
> Paranoia.
>> Shouldn't (enable-local-variables :safe) be sufficient?
> Yes. Unless there is another screw-up like
> http://debbugs.gnu.org/cgi/bugreport.cgi?bug=12155
> BTW, there have been zero other comments on that. Emacs releases have
> happened in similar cases in the past, IIUC.

I had the impression that it only affects unusual configurations.


        Stefan



^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [Emacs-diffs] /srv/bzr/emacs/emacs-24 r108098: * admin/bzrmerge.el (bzrmerge-resolve): Disable local eval:.
  2012-08-10 20:29     ` Stefan Monnier
@ 2012-08-10 20:42       ` Glenn Morris
  2012-08-10 20:50         ` Glenn Morris
  0 siblings, 1 reply; 5+ messages in thread
From: Glenn Morris @ 2012-08-10 20:42 UTC (permalink / raw)
  To: Stefan Monnier; +Cc: emacs-devel

Stefan Monnier wrote:

> I had the impression that it only affects unusual configurations.

Non-default configs; don't know if I would say "unusual".

   (setq enable-local-variables :safe)

is all you need.

There were a few places in Emacs that use that setting (without setting
enable-local-eval to nil). I don't think any of them were obviously
exploitable though. The autoloads one is in theory:

mkdir /tmp/foo
cd /tmp/foo
cat <<EOF > foo.el
(setq foo t)

;; Local Variables:
;; eval: (shell-command "touch /tmp/OHDEAR")
;; End:
EOF

rm -f /tmp/OHDEAR

emacs-24.1 -Q -l autoload \
 --eval "(setq generated-autoload-file \"$PWD/loaddefs.el\")" \
 --batch -f batch-update-autoloads .

ls /tmp/OHDEAR

But I suppose no-one creates autoloads without also byte-compiling,
which could already eval arbitrary code.

But there could be some third-party code that binds
enable-local-variables to :safe around some operation, intending to make
things safer for the user, but actually doing the opposite.



^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [Emacs-diffs] /srv/bzr/emacs/emacs-24 r108098: * admin/bzrmerge.el (bzrmerge-resolve): Disable local eval:.
  2012-08-10 20:42       ` Glenn Morris
@ 2012-08-10 20:50         ` Glenn Morris
  0 siblings, 0 replies; 5+ messages in thread
From: Glenn Morris @ 2012-08-10 20:50 UTC (permalink / raw)
  To: Stefan Monnier; +Cc: emacs-devel


PS CVE for previous similar (oops...) issue

http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2007-5795

Though I see that that issue was in 22.1 and 22.2 was not released for
another 5 months after it was fixed. So there's precedent for not
rushing out a release for this.



^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2012-08-10 20:50 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
     [not found] <E1SzjTX-0001sf-8w@vcs.savannah.gnu.org>
2012-08-10 15:26 ` [Emacs-diffs] /srv/bzr/emacs/emacs-24 r108098: * admin/bzrmerge.el (bzrmerge-resolve): Disable local eval: Stefan Monnier
2012-08-10 15:51   ` Glenn Morris
2012-08-10 20:29     ` Stefan Monnier
2012-08-10 20:42       ` Glenn Morris
2012-08-10 20:50         ` Glenn Morris

Code repositories for project(s) associated with this external index

	https://git.savannah.gnu.org/cgit/emacs.git
	https://git.savannah.gnu.org/cgit/emacs/org-mode.git

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.