[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#682908: marked as done (Freeze exception request: emacs24/24.1+1-4)

Your message dated Fri, 5 Oct 2012 09:49:09 +0200
with message-id <20121005074909.GB31434@radis.cristau.org>
and subject line Re: Bug#682908: Freeze exception request: emacs24/24.1+1-4
has caused the Debian Bug report #682908,
regarding Freeze exception request: emacs24/24.1+1-4
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
682908: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=682908
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: release.debian.org
Severity: normal

Emacs 24.1 was recently released upstream, and emacs24 24.1+1-1 made it
for the freeze, but just barely.

Then it turned out there were some arch-specific FTBS problems that I
managed to track down, with help, during DebConf.  So now I wanted to
see if you thought that emacs24 might still be appropriate for Wheezy,
or if it just needs to be withdrawn.

I uploaded the last FTBS fix last week:

  https://buildd.debian.org/status/package.php?p=emacs24&suite=sid

and since the original 24.4+1-1, there have been three changes:

  * Restrict libselinux1-dev build dependency to [linux-any] (#679679).
  * Don't check errno after fabs() -- fixes i386 builds. (upstream patch).
  * Use -O1 on ia64 (fix FTBS) -- this is a reversion to the emacs23 behavior.

Here are the significant bits of the -1 to -4 diff:


--- a/debian/control
+++ b/debian/control
@@ -8,7 +8,7 @@ Build-Depends: bsd-mailx | mailx, libncurses5-dev, texinfo, liblockfile-dev, lib
  libgpm-dev [linux-any], libdbus-1-dev,
  autoconf, automake, autotools-dev, dpkg-dev (>> 1.10.0), quilt (>= 0.42),
  debhelper (>= 7.0.50~), libxaw7-dev, sharutils, imagemagick, libgtk2.0-dev,
- libgnutls-dev, libxml2-dev, libselinux1-dev, libmagick++-dev,
+ libgnutls-dev, libxml2-dev, libselinux1-dev [linux-any], libmagick++-dev,
  libgconf2-dev, libasound2-dev [!hurd-i386 !kfreebsd-i386 !kfreebsd-amd64]
 Homepage: http://www.gnu.org/software/emacs/
 Standards-Version: 3.7.2
--- a/debian/rules
+++ b/debian/rules
@@ -105,10 +105,12 @@ CFLAGS += -Wall
 LDFLAGS = `dpkg-buildflags --get LDFLAGS`
 CPPFLAGS = `dpkg-buildflags --get CPPFLAGS`
 
-LDFLAGS   += -g
-
 ifeq (,$(findstring noopt,$(DEB_BUILD_OPTIONS)))
-  CFLAGS += -O2
+  ifneq (,$(filter $(DEB_HOST_ARCH),ia64))
+    # Fix a problem with newer versions of gcc on ia64.
+    # See bugs #582439 and #679986.
+    CFLAGS += -O1
+  endif # eq ia64
 endif
 
 joblimit := $(patsubst parallel=%,%,$(filter parallel=%,$(DEB_BUILD_OPTIONS)))
--- a/src/ChangeLog
+++ b/src/ChangeLog
@@ -1,3 +1,9 @@
+2012-07-17  Paul Eggert  <eggert@cs.ucla.edu>
+
+	* floatfns.c (Fabs): Do not wrap fabs inside IN_FLOAT (Bug#11913).
+	Unlike the other wrapped functions, fabs has an unspecified
+	effect on errno.
+
 2012-06-01  Chong Yidong  <cyd@gnu.org>
 
 	* Version 24.1 released.
--- a/src/floatfns.c
+++ b/src/floatfns.c
@@ -676,7 +676,7 @@ DEFUN ("abs", Fabs, Sabs, 1, 1, 0,
   CHECK_NUMBER_OR_FLOAT (arg);
 
   if (FLOATP (arg))
-    IN_FLOAT (arg = make_float (fabs (XFLOAT_DATA (arg))), "abs", arg);
+    arg = make_float (fabs (XFLOAT_DATA (arg)));
   else if (XINT (arg) < 0)
     XSETINT (arg, - XINT (arg));
 

and here's the full debdiff:


diff -Nru emacs24-24.1+1/debian/.git-dpm emacs24-24.1+1/debian/.git-dpm
--- emacs24-24.1+1/debian/.git-dpm	2012-06-27 21:41:01.000000000 -0500
+++ emacs24-24.1+1/debian/.git-dpm	2012-07-17 19:23:17.000000000 -0500
@@ -1,6 +1,6 @@
 # see git-dpm(1) from git-dpm package
-c29cccb1d825523faf484808a699959766d71348
-c29cccb1d825523faf484808a699959766d71348
+6ac9d5dbfded9610d8208378a3be49b1ccd8fe9a
+6ac9d5dbfded9610d8208378a3be49b1ccd8fe9a
 dccb7aff653035782f95e87102338cabcd318122
 dccb7aff653035782f95e87102338cabcd318122
 emacs24_24.1+1.orig.tar.bz2
diff -Nru emacs24-24.1+1/debian/changelog emacs24-24.1+1/debian/changelog
--- emacs24-24.1+1/debian/changelog	2012-06-29 16:16:58.000000000 -0500
+++ emacs24-24.1+1/debian/changelog	2012-07-20 12:51:38.000000000 -0500
@@ -1,3 +1,27 @@
+emacs24 (24.1+1-4) unstable; urgency=low
+
+  * Fix FTBS on ia64 with -O1.  Remove redundant LDFLAGS+=-g and
+    CFLAGS+=-O2.  See also: #582439. (Closes: #679986)
+
+ -- Rob Browning <rlb@defaultvalue.org>  Fri, 20 Jul 2012 12:51:32 -0500
+
+emacs24 (24.1+1-3) unstable; urgency=low
+
+  * Add 0011-Don-t-use-IN_FLOAT-when-calling-fabs-since-it-may-cl.patch.
+    Fix FTBS on i386 by adding an upstream patch to skip fabs() errno
+    checking.  Since fabs() has no error conditions, it doesn't guarantee
+    that it won't clobber errno.
+
+ -- Rob Browning <rlb@defaultvalue.org>  Tue, 17 Jul 2012 19:25:59 -0500
+
+emacs24 (24.1+1-2) unstable; urgency=low
+
+  * Restrict libselinux1-dev build dependency to [linux-any].
+    Thanks to James McCoy <jamessan@debian.org> for the report.
+    (Closes: #679679)
+
+ -- Rob Browning <rlb@defaultvalue.org>  Wed, 04 Jul 2012 19:16:00 -0500
+
 emacs24 (24.1+1-1) unstable; urgency=low
 
   * Upgrade to upstream version 24.1.
diff -Nru emacs24-24.1+1/debian/control emacs24-24.1+1/debian/control
--- emacs24-24.1+1/debian/control	2012-06-29 16:18:17.000000000 -0500
+++ emacs24-24.1+1/debian/control	2012-07-20 12:53:48.000000000 -0500
@@ -8,7 +8,7 @@
  libgpm-dev [linux-any], libdbus-1-dev,
  autoconf, automake, autotools-dev, dpkg-dev (>> 1.10.0), quilt (>= 0.42),
  debhelper (>= 7.0.50~), libxaw7-dev, sharutils, imagemagick, libgtk2.0-dev,
- libgnutls-dev, libxml2-dev, libselinux1-dev, libmagick++-dev,
+ libgnutls-dev, libxml2-dev, libselinux1-dev [linux-any], libmagick++-dev,
  libgconf2-dev, libasound2-dev [!hurd-i386 !kfreebsd-i386 !kfreebsd-amd64]
 Homepage: http://www.gnu.org/software/emacs/
 Standards-Version: 3.7.2
diff -Nru emacs24-24.1+1/debian/control.in emacs24-24.1+1/debian/control.in
--- emacs24-24.1+1/debian/control.in	2012-06-27 21:41:01.000000000 -0500
+++ emacs24-24.1+1/debian/control.in	2012-07-17 19:23:16.000000000 -0500
@@ -8,7 +8,7 @@
  libgpm-dev [linux-any], libdbus-1-dev,
  autoconf, automake, autotools-dev, dpkg-dev (>> 1.10.0), quilt (>= 0.42),
  debhelper (>= 7.0.50~), libxaw7-dev, sharutils, imagemagick, libgtk2.0-dev,
- libgnutls-dev, libxml2-dev, libselinux1-dev, libmagick++-dev,
+ libgnutls-dev, libxml2-dev, libselinux1-dev [linux-any], libmagick++-dev,
  libgconf2-dev, libasound2-dev [!hurd-i386 !kfreebsd-i386 !kfreebsd-amd64]
 Homepage: http://www.gnu.org/software/emacs/
 Standards-Version: 3.7.2
diff -Nru emacs24-24.1+1/debian/patches/0011-Don-t-use-IN_FLOAT-when-calling-fabs-since-it-may-cl.patch emacs24-24.1+1/debian/patches/0011-Don-t-use-IN_FLOAT-when-calling-fabs-since-it-may-cl.patch
--- emacs24-24.1+1/debian/patches/0011-Don-t-use-IN_FLOAT-when-calling-fabs-since-it-may-cl.patch	1969-12-31 18:00:00.000000000 -0600
+++ emacs24-24.1+1/debian/patches/0011-Don-t-use-IN_FLOAT-when-calling-fabs-since-it-may-cl.patch	2012-07-17 19:23:17.000000000 -0500
@@ -0,0 +1,47 @@
+From 6ac9d5dbfded9610d8208378a3be49b1ccd8fe9a Mon Sep 17 00:00:00 2001
+From: Paul Eggert <eggert@cs.ucla.edu>
+Date: Mon, 16 Jul 2012 19:56:00 -0700
+Subject: Don't use IN_FLOAT() when calling fabs() since it may clobber errno.
+
+The emacs build shouldn't fail sometimes with an "Arithmetic error:".
+
+Previously Emacs assumed that fabs() would not modify errno
+unless there was an error, but that isn't guaranteed since fabs()
+has no error conditions.
+
+Origin: upstream, commit: 8e0e7a92f5ae99ce2461fc0f0b606d4cec3efb81
+Added-by: Rob Browning <rlb@defaultvalue.org>
+Provided-By: Paul Eggert  <eggert@cs.ucla.edu>
+Bug: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=11913
+---
+ src/ChangeLog  |    6 ++++++
+ src/floatfns.c |    2 +-
+ 2 files changed, 7 insertions(+), 1 deletion(-)
+
+diff --git a/src/ChangeLog b/src/ChangeLog
+index 98c4124..32631e7 100644
+--- a/src/ChangeLog
++++ b/src/ChangeLog
+@@ -1,3 +1,9 @@
++2012-07-17  Paul Eggert  <eggert@cs.ucla.edu>
++
++	* floatfns.c (Fabs): Do not wrap fabs inside IN_FLOAT (Bug#11913).
++	Unlike the other wrapped functions, fabs has an unspecified
++	effect on errno.
++
+ 2012-06-01  Chong Yidong  <cyd@gnu.org>
+ 
+ 	* Version 24.1 released.
+diff --git a/src/floatfns.c b/src/floatfns.c
+index 305c78c..fa672c7 100644
+--- a/src/floatfns.c
++++ b/src/floatfns.c
+@@ -676,7 +676,7 @@ DEFUN ("abs", Fabs, Sabs, 1, 1, 0,
+   CHECK_NUMBER_OR_FLOAT (arg);
+ 
+   if (FLOATP (arg))
+-    IN_FLOAT (arg = make_float (fabs (XFLOAT_DATA (arg))), "abs", arg);
++    arg = make_float (fabs (XFLOAT_DATA (arg)));
+   else if (XINT (arg) < 0)
+     XSETINT (arg, - XINT (arg));
+ 
diff -Nru emacs24-24.1+1/debian/patches/series emacs24-24.1+1/debian/patches/series
--- emacs24-24.1+1/debian/patches/series	2012-06-27 21:41:01.000000000 -0500
+++ emacs24-24.1+1/debian/patches/series	2012-07-17 19:23:17.000000000 -0500
@@ -8,3 +8,4 @@
 0008-Fix-startup-hang-on-Debian-GNU-kFreeBSD.patch
 0009-Restore-doc-emacs-emacsver.texi.patch
 0010-Rename-infodir-to-buildinfodir-in-doc-Makefile.in-GN.patch
+0011-Don-t-use-IN_FLOAT-when-calling-fabs-since-it-may-cl.patch
diff -Nru emacs24-24.1+1/debian/rules emacs24-24.1+1/debian/rules
--- emacs24-24.1+1/debian/rules	2012-06-29 16:15:15.000000000 -0500
+++ emacs24-24.1+1/debian/rules	2012-07-20 12:49:20.000000000 -0500
@@ -105,10 +105,12 @@
 LDFLAGS = `dpkg-buildflags --get LDFLAGS`
 CPPFLAGS = `dpkg-buildflags --get CPPFLAGS`
 
-LDFLAGS   += -g
-
 ifeq (,$(findstring noopt,$(DEB_BUILD_OPTIONS)))
-  CFLAGS += -O2
+  ifneq (,$(filter $(DEB_HOST_ARCH),ia64))
+    # Fix a problem with newer versions of gcc on ia64.
+    # See bugs #582439 and #679986.
+    CFLAGS += -O1
+  endif # eq ia64
 endif
 
 joblimit := $(patsubst parallel=%,%,$(filter parallel=%,$(DEB_BUILD_OPTIONS)))

If you think emacs24's not ready, that's certainly fine, and please let
me know if you'd like any further information.

Thanks
-- 
Rob Browning
rlb @defaultvalue.org and @debian.org
GPG as of 2002-11-03 14DD 432F AE39 534D B592 F9A0 25C8 D377 8C7E 73A4

--- End Message ---
--- Begin Message --- 
On Fri, Jul 27, 2012 at 12:00:07 -0500, Rob Browning wrote:

> Julien Cristau <jcristau@debian.org> writes:
> 
> > On Fri, Jul 27, 2012 at 09:59:25 -0500, Rob Browning wrote:
> >
> >> Just to double-check, do I need to do anything to have emacs24 pulled
> >> from testing (or is that documented somewhere I should read)?
> >> 
> > It's not in testing, and has never been there.
> 
> Oh, OK.  I must have just misunderstood the initial circumstances for
> -1.
> 
Closing this bug, we'll stay with just 23 in wheezy.  Thanks for your
work.

Cheers,
Julien

Attachment: signature.asc
Description: Digital signature


--- End Message ---
Reply to: