Author: aurel32
Date: 2015-11-23 21:30:14 +0000 (Mon, 23 Nov 2015)
New Revision: 6700
Added:
glibc-package/branches/glibc-2.21/debian/patches/hppa/submitted-mathdef.diff
Modified:
glibc-package/branches/glibc-2.21/debian/changelog
glibc-package/branches/glibc-2.21/debian/patches/series
Log:
patches/hppa/submitted-mathdef.diff: new patch from John David Anglin to
define __NO_LONG_DOUBLE_MATH on hppa. Closes: #805836.
Modified: glibc-package/branches/glibc-2.21/debian/changelog
===================================================================
--- glibc-package/branches/glibc-2.21/debian/changelog 2015-11-23 11:31:00 UTC
(rev 6699)
+++ glibc-package/branches/glibc-2.21/debian/changelog 2015-11-23 21:30:14 UTC
(rev 6700)
@@ -9,6 +9,8 @@
* testsuite-checking/expected-results-{arm,mips}*: allow nptl/tst-stack4 to
fail. It's a new test which fails intermitently on the buildds and a known
upstream problem.
+ * patches/hppa/submitted-mathdef.diff: new patch from John David Anglin to
+ define __NO_LONG_DOUBLE_MATH on hppa. Closes: #805836.
-- Aurelien Jarno <[email protected]> Sun, 22 Nov 2015 20:20:54 +0100
Added:
glibc-package/branches/glibc-2.21/debian/patches/hppa/submitted-mathdef.diff
===================================================================
---
glibc-package/branches/glibc-2.21/debian/patches/hppa/submitted-mathdef.diff
(rev 0)
+++
glibc-package/branches/glibc-2.21/debian/patches/hppa/submitted-mathdef.diff
2015-11-23 21:30:14 UTC (rev 6700)
@@ -0,0 +1,22 @@
+2015-11-22 John David Anglin <[email protected]>
+
+ [BZ #19270]
+ * sysdeps/hppa/fpu/bits/mathdef.h (__NO_LONG_DOUBLE_MATH): Define.
+
+diff --git a/sysdeps/hppa/fpu/bits/mathdef.h b/sysdeps/hppa/fpu/bits/mathdef.h
+index d189f4a..443655f 100644
+--- a/sysdeps/hppa/fpu/bits/mathdef.h
++++ b/sysdeps/hppa/fpu/bits/mathdef.h
+@@ -34,5 +34,10 @@ typedef double double_t; /* `double' expressions are
evaluated as
+
+ #endif /* ISO C99 */
+
+-/* On hppa `long double' is 64-bits. */
+-#undef __NO_LONG_DOUBLE_MATH
++#ifndef __NO_LONG_DOUBLE_MATH
++/* On hppa `long double' and `double' are 64-bits. So, libm is built
++ with NO_LONG_DOUBLE defined. The following define ensures the library
++ and headers are consistent. This disables the declaration of all the
++ `long double' function variants. */
++# define __NO_LONG_DOUBLE_MATH 1
++#endif
Modified: glibc-package/branches/glibc-2.21/debian/patches/series
===================================================================
--- glibc-package/branches/glibc-2.21/debian/patches/series 2015-11-23
11:31:00 UTC (rev 6699)
+++ glibc-package/branches/glibc-2.21/debian/patches/series 2015-11-23
21:30:14 UTC (rev 6700)
@@ -72,6 +72,7 @@
hppa/cvs-atomic.diff
hppa/cvs-inline-syscall.diff
hppa/submitted-gmon-start.diff
+hppa/submitted-mathdef.diff
hurd-i386/local-enable-ldconfig.diff
hurd-i386/tg-context_functions.diff
