Skip to contentSkip to conversation

package/x11r7/xlib_libxshmfence: fix build with glibc 2.27

PATCH New Romain Naour 6 participants 20 messages Feb 5, 2018 – Feb 6, 2018

lore
Share & tools
lore
#0 lore raw

PATCH 1/5 package/x11r7/xlib_libxshmfence: fix build with glibc 2.27

With glibc 2.27 memfd_create() is inside a _GNU_SOURCE guard, so call AC_USE_SYSTEM_EXTENSIONS to get this defined.

---
 ...onfigure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch | 30 ++++++++++++++++++++++
 .../x11r7/xlib_libxshmfence/xlib_libxshmfence.mk   |  3 +++
 2 files changed, 33 insertions(+)
 create mode 100644 package/x11r7/xlib_libxshmfence/0001-configure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch
package/x11r7/xlib_libxshmfence/0001-configure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch new file +30 -0
new file mode 100644
index 0000000000..f24b1da5dd
@@ -0,0 +1,30 @@
+From 0b550a4e7acf02d3478602848f6afbfcbfb0d4b2 Mon Sep 17 00:00:00 2001
+From: Ross Burton <[email protected]>
+Date: Mon, 29 Jan 2018 16:24:36 +0000
+Subject: [PATCH] configure.ac: call AC_USE_SYSTEM_EXTENSIONS
+
+With glibc 2.27 memfd_create() is inside a _GNU_SOURCE guard, so call
+AC_USE_SYSTEM_EXTENSIONS to get this defined.
+
+Signed-off-by: Ross Burton <[email protected]>
+Signed-off-by: Romain Naour <[email protected]>
+---
+ configure.ac | 2 ++
+ 1 file changed, 2 insertions(+)
+
+diff --git a/configure.ac b/configure.ac
+index 55772d0..ed77e6d 100644
+--- a/configure.ac
++++ b/configure.ac
+@@ -28,6 +28,8 @@ AC_INIT([libxshmfence], [1.2],
+ AC_CONFIG_SRCDIR([Makefile.am])
+ AC_CONFIG_HEADERS([config.h])
+ 
++AC_USE_SYSTEM_EXTENSIONS
++
+ # Initialize Automake
+ AM_INIT_AUTOMAKE([foreign dist-bzip2])
+ 
+-- 
+2.14.3
+
package/x11r7/xlib_libxshmfence/xlib_libxshmfence.mk +3 -0
index 3a2f32623a..9686c408c1 100644
@@ -12,4 +12,7 @@ XLIB_LIBXSHMFENCE_LICENSE_FILES = COPYING
 XLIB_LIBXSHMFENCE_INSTALL_STAGING = YES
 XLIB_LIBXSHMFENCE_DEPENDENCIES = host-pkgconf xproto_xproto
 
+# 0001-configure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch
+XLIB_LIBXSHMFENCE_AUTORECONF = YES
+
 $(eval $(autotools-package))
Signature · 1 line
2.14.3
#1 lore raw

PATCH 2/5 package/pulseaudio: only define memfd_create() if not already defined

Before glibc version 2.27 there was no wrapper for memfd_create(2).

---
 ...ers-only-define-memfd_create-if-not-alrea.patch | 64 ++++++++++++++++++++++
 package/pulseaudio/pulseaudio.mk                   |  3 +
 2 files changed, 67 insertions(+)
 create mode 100644 package/pulseaudio/0002-memfd-wrappers-only-define-memfd_create-if-not-alrea.patch
package/pulseaudio/0002-memfd-wrappers-only-define-memfd_create-if-not-alrea.patch new file +64 -0
new file mode 100644
index 0000000000..8f103bf201
@@ -0,0 +1,64 @@
+From dfb0460fb4743aec047cdf755a660a9ac2d0f3fb Mon Sep 17 00:00:00 2001
+From: Tanu Kaskinen <[email protected]>
+Date: Wed, 24 Jan 2018 03:51:49 +0200
+Subject: [PATCH] memfd-wrappers: only define memfd_create() if not already
+ defined
+
+glibc 2.27 is to be released soon, and it will provide memfd_create().
+If glibc provides the function, we must not define it ourselves,
+otherwise building fails due to conflict between the two implementations
+of the same function.
+
+BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=104733
+Signed-off-by: Romain Naour <[email protected]>
+---
+ configure.ac                   | 3 +++
+ src/pulsecore/memfd-wrappers.h | 7 ++++---
+ 2 files changed, 7 insertions(+), 3 deletions(-)
+
+diff --git a/configure.ac b/configure.ac
+index 0084c86e..0eb44b08 100644
+--- a/configure.ac
++++ b/configure.ac
+@@ -610,6 +610,9 @@ AS_IF([test "x$enable_memfd" = "xyes" && test "x$HAVE_MEMFD" = "x0"],
+     [AC_MSG_ERROR([*** Your Linux kernel does not support memfd shared memory.
+                   *** Use linux v3.17 or higher for such a feature.])])
+ 
++AS_IF([test "x$HAVE_MEMFD" = "x1"],
++    AC_CHECK_FUNCS([memfd_create]))
++
+ AC_SUBST(HAVE_MEMFD)
+ AM_CONDITIONAL([HAVE_MEMFD], [test "x$HAVE_MEMFD" = x1])
+ AS_IF([test "x$HAVE_MEMFD" = "x1"], AC_DEFINE([HAVE_MEMFD], 1, [Have memfd shared memory.]))
+diff --git a/src/pulsecore/memfd-wrappers.h b/src/pulsecore/memfd-wrappers.h
+index 3bed9b2b..c7aadfd3 100644
+--- a/src/pulsecore/memfd-wrappers.h
++++ b/src/pulsecore/memfd-wrappers.h
+@@ -20,13 +20,14 @@
+   License along with PulseAudio; if not, see <http://www.gnu.org/licenses/>.
+ ***/
+ 
+-#ifdef HAVE_MEMFD
++#if defined(HAVE_MEMFD) && !defined(HAVE_MEMFD_CREATE)
+ 
+ #include <sys/syscall.h>
+ #include <fcntl.h>
+ 
+ /*
+- * No glibc wrappers exist for memfd_create(2), so provide our own.
++ * Before glibc version 2.27 there was no wrapper for memfd_create(2),
++ * so we have to provide our own.
+  *
+  * Also define memfd fcntl sealing macros. While they are already
+  * defined in the kernel header file <linux/fcntl.h>, that file as
+@@ -63,6 +64,6 @@ static inline int memfd_create(const char *name, unsigned int flags) {
+ #define F_SEAL_WRITE    0x0008  /* prevent writes */
+ #endif
+ 
+-#endif /* HAVE_MEMFD */
++#endif /* HAVE_MEMFD && !HAVE_MEMFD_CREATE */
+ 
+ #endif
+-- 
+2.14.3
+
package/pulseaudio/pulseaudio.mk +3 -0
index 13fcec52a3..f59828d43d 100644
@@ -15,6 +15,9 @@ PULSEAUDIO_CONF_OPTS = \
 	--disable-legacy-database-entry-format \
 	--disable-manpages
 
+# 0001-memfd-wrappers-only-define-memfd_create-if-not-alrea.patch
+PULSEAUDIO_AUTORECONF = YES
+
 PULSEAUDIO_DEPENDENCIES = \
 	host-pkgconf libtool libsndfile speex host-intltool \
 	$(if $(BR2_PACKAGE_LIBGLIB2),libglib2) \
Signature · 1 line
2.14.3
#2 lore raw

PATCH 3/5 package/libraw: rename internal powf64

In reply to Romain Naour

Starting with Glibc 2.27, powf64() is provided by the C library and colide with the one defined dy libraw.

Rename internal powf64 to libraw_powf64.

---
 ...ernal-dcraw_common-rename-internal-powf64.patch | 348 +++++++++++++++++++++
 1 file changed, 348 insertions(+)
 create mode 100644 package/libraw/0003-internal-dcraw_common-rename-internal-powf64.patch
package/libraw/0003-internal-dcraw_common-rename-internal-powf64.patch new file +348 -0
new file mode 100644
index 0000000000..ef3b81aa24
@@ -0,0 +1,348 @@
+From 045c57d9d030075575409cf50cd7f6ee635df5c0 Mon Sep 17 00:00:00 2001
+From: Romain Naour <[email protected]>
+Date: Mon, 5 Feb 2018 16:22:39 +0100
+Subject: [PATCH] internal/dcraw_common: rename internal powf64()
+
+Starting with Glibc 2.27, powf64() is provided by the C library
+and colide with the one defined dy libraw.
+
+Rename internal powf64 to libraw_powf64.
+
+Signed-off-by: Romain Naour <[email protected]>
+---
+ internal/dcraw_common.cpp | 98 +++++++++++++++++++++++------------------------
+ 1 file changed, 49 insertions(+), 49 deletions(-)
+
+diff --git a/internal/dcraw_common.cpp b/internal/dcraw_common.cpp
+index 0c7c07a..cf26865 100644
+--- a/internal/dcraw_common.cpp
++++ b/internal/dcraw_common.cpp
+@@ -5537,7 +5537,7 @@ static float powf_lim(float a, float b, float limup)
+ {
+   return (b>limup || b < -limup)?0.f:powf(a,b);
+ }
+-static float powf64(float a, float b)
++static float libraw_powf64(float a, float b)
+ {
+   return powf_lim(a,b,64.f);
+ }
+@@ -5561,7 +5561,7 @@ static float my_roundf(float x) {
+ static float _CanonConvertAperture(ushort in)
+ {
+   if ((in == (ushort)0xffe0) || (in == (ushort)0x7fff)) return 0.0f;
+-  return powf64(2.0, in/64.0);
++  return libraw_powf64(2.0, in/64.0);
+ }
+ 
+ void CLASS setCanonBodyFeatures (unsigned id)
+@@ -5902,21 +5902,21 @@ void CLASS processNikonLensData (uchar *LensData, unsigned len)
+     if (fabsf(imgdata.lens.makernotes.MinFocal) < 1.1f)
+     {
+       if ((imgdata.lens.nikon.NikonLensType ^ (uchar)0x01) || LensData[i + 2])
+-        imgdata.lens.makernotes.MinFocal = 5.0f * powf64(2.0f, (float)LensData[i + 2] / 24.0f);
++        imgdata.lens.makernotes.MinFocal = 5.0f * libraw_powf64(2.0f, (float)LensData[i + 2] / 24.0f);
+       if ((imgdata.lens.nikon.NikonLensType ^ (uchar)0x01) || LensData[i + 3])
+-        imgdata.lens.makernotes.MaxFocal = 5.0f * powf64(2.0f, (float)LensData[i + 3] / 24.0f);
++        imgdata.lens.makernotes.MaxFocal = 5.0f * libraw_powf64(2.0f, (float)LensData[i + 3] / 24.0f);
+       if ((imgdata.lens.nikon.NikonLensType ^ (uchar)0x01) || LensData[i + 4])
+-        imgdata.lens.makernotes.MaxAp4MinFocal = powf64(2.0f, (float)LensData[i + 4] / 24.0f);
++        imgdata.lens.makernotes.MaxAp4MinFocal = libraw_powf64(2.0f, (float)LensData[i + 4] / 24.0f);
+       if ((imgdata.lens.nikon.NikonLensType ^ (uchar)0x01) || LensData[i + 5])
+-        imgdata.lens.makernotes.MaxAp4MaxFocal = powf64(2.0f, (float)LensData[i + 5] / 24.0f);
++        imgdata.lens.makernotes.MaxAp4MaxFocal = libraw_powf64(2.0f, (float)LensData[i + 5] / 24.0f);
+     }
+     imgdata.lens.nikon.NikonMCUVersion = LensData[i + 6];
+     if (i != 2)
+       {
+         if ((LensData[i - 1]) &&
+             (fabsf(imgdata.lens.makernotes.CurFocal) < 1.1f))
+-          imgdata.lens.makernotes.CurFocal = 5.0f * powf64(2.0f, (float)LensData[i - 1] / 24.0f);
+-        if (LensData[i + 7]) imgdata.lens.nikon.NikonEffectiveMaxAp = powf64(2.0f, (float)LensData[i + 7] / 24.0f);
++          imgdata.lens.makernotes.CurFocal = 5.0f * libraw_powf64(2.0f, (float)LensData[i - 1] / 24.0f);
++        if (LensData[i + 7]) imgdata.lens.nikon.NikonEffectiveMaxAp = libraw_powf64(2.0f, (float)LensData[i + 7] / 24.0f);
+       }
+     imgdata.lens.makernotes.LensID =
+       (unsigned long long) LensData[i] << 56 |
+@@ -6088,13 +6088,13 @@ void CLASS PentaxLensInfo (unsigned id, unsigned len)	// tag 0x0207
+ 		if (table_buf[iLensData+9] &&
+ 			(fabs(imgdata.lens.makernotes.CurFocal) < 0.1f))
+ 		  imgdata.lens.makernotes.CurFocal =
+-			10*(table_buf[iLensData+9]>>2) * powf64(4, (table_buf[iLensData+9] & 0x03)-2);
++			10*(table_buf[iLensData+9]>>2) * libraw_powf64(4, (table_buf[iLensData+9] & 0x03)-2);
+ 		if (table_buf[iLensData+10] & 0xf0)
+ 		  imgdata.lens.makernotes.MaxAp4CurFocal =
+-			powf64(2.0f, (float)((table_buf[iLensData+10] & 0xf0) >>4)/4.0f);
++			libraw_powf64(2.0f, (float)((table_buf[iLensData+10] & 0xf0) >>4)/4.0f);
+ 		if (table_buf[iLensData+10] & 0x0f)
+ 		  imgdata.lens.makernotes.MinAp4CurFocal =
+-			powf64(2.0f, (float)((table_buf[iLensData+10] & 0x0f) + 10)/4.0f);
++			libraw_powf64(2.0f, (float)((table_buf[iLensData+10] & 0x0f) + 10)/4.0f);
+ 
+ 		if (iLensData != 12)
+ 		  {
+@@ -6111,14 +6111,14 @@ void CLASS PentaxLensInfo (unsigned id, unsigned len)	// tag 0x0207
+ 			if ((table_buf[iLensData+14] > 1) &&
+ 				(fabs(imgdata.lens.makernotes.MaxAp4CurFocal) < 0.7f))
+ 			  imgdata.lens.makernotes.MaxAp4CurFocal =
+-				powf64(2.0f, (float)((table_buf[iLensData+14] & 0x7f) -1)/32.0f);
++				libraw_powf64(2.0f, (float)((table_buf[iLensData+14] & 0x7f) -1)/32.0f);
+ 		  }
+ 		else if ((id != 0x12e76) &&	// K-5
+ 				 (table_buf[iLensData+15] > 1) &&
+ 				 (fabs(imgdata.lens.makernotes.MaxAp4CurFocal) < 0.7f))
+ 		  {
+ 			imgdata.lens.makernotes.MaxAp4CurFocal =
+-			  powf64(2.0f, (float)((table_buf[iLensData+15] & 0x7f) -1)/32.0f);
++			  libraw_powf64(2.0f, (float)((table_buf[iLensData+15] & 0x7f) -1)/32.0f);
+ 		  }
+ 	  }
+ 	free(table_buf);
+@@ -6525,11 +6525,11 @@ void CLASS process_Sony_0x9050 (uchar * buf, unsigned id)
+     {
+       if (buf[0])
+         imgdata.lens.makernotes.MaxAp4CurFocal =
+-          my_roundf(powf64(2.0f, ((float)SonySubstitution[buf[0]] / 8.0 - 1.06f) / 2.0f)*10.0f) / 10.0f;
++          my_roundf(libraw_powf64(2.0f, ((float)SonySubstitution[buf[0]] / 8.0 - 1.06f) / 2.0f)*10.0f) / 10.0f;
+ 
+       if (buf[1])
+         imgdata.lens.makernotes.MinAp4CurFocal =
+-          my_roundf(powf64(2.0f, ((float)SonySubstitution[buf[1]] / 8.0 - 1.06f) / 2.0f)*10.0f) / 10.0f;
++          my_roundf(libraw_powf64(2.0f, ((float)SonySubstitution[buf[1]] / 8.0 - 1.06f) / 2.0f)*10.0f) / 10.0f;
+     }
+ 
+   if (imgdata.lens.makernotes.CameraMount != LIBRAW_MOUNT_FixedLens)
+@@ -6539,7 +6539,7 @@ void CLASS process_Sony_0x9050 (uchar * buf, unsigned id)
+           lid = SonySubstitution[buf[0x3d]] << 8 |
+             SonySubstitution[buf[0x3c]];
+           imgdata.lens.makernotes.CurAp =
+-            powf64(2.0f, ((float)lid/256.0f - 16.0f) / 2.0f);
++            libraw_powf64(2.0f, ((float)lid/256.0f - 16.0f) / 2.0f);
+         }
+       if (buf[0x105] && (imgdata.lens.makernotes.LensMount != LIBRAW_MOUNT_Canon_EF))
+         imgdata.lens.makernotes.LensMount =
+@@ -6924,7 +6924,7 @@ void CLASS parse_makernote_0xc634(int base, int uptag, unsigned dng_writer)
+           {
+             unsigned char cc;
+             fread(&cc, 1, 1, ifp);
+-            iso_speed = (int)(100.0 * powf64(2.0, (double)(cc) / 12.0 - 5.0));
++            iso_speed = (int)(100.0 * libraw_powf64(2.0, (double)(cc) / 12.0 - 5.0));
+             break;
+           }
+       }
+@@ -6957,7 +6957,7 @@ void CLASS parse_makernote_0xc634(int base, int uptag, unsigned dng_writer)
+           }
+           break;
+         case 0x1002:
+-          imgdata.lens.makernotes.CurAp = powf64(2.0f, getreal(type)/2);
++          imgdata.lens.makernotes.CurAp = libraw_powf64(2.0f, getreal(type)/2);
+           break;
+         case 0x20100201:
+           imgdata.lens.makernotes.LensID =
+@@ -6977,10 +6977,10 @@ void CLASS parse_makernote_0xc634(int base, int uptag, unsigned dng_writer)
+           fread(imgdata.lens.makernotes.Lens, MIN(len,127), 1, ifp);
+           break;
+         case 0x20100205:
+-          imgdata.lens.makernotes.MaxAp4MinFocal = powf64(sqrt(2.0f), get2() / 256.0f);
++          imgdata.lens.makernotes.MaxAp4MinFocal = libraw_powf64(sqrt(2.0f), get2() / 256.0f);
+           break;
+         case 0x20100206:
+-          imgdata.lens.makernotes.MaxAp4MaxFocal = powf64(sqrt(2.0f), get2() / 256.0f);
++          imgdata.lens.makernotes.MaxAp4MaxFocal = libraw_powf64(sqrt(2.0f), get2() / 256.0f);
+           break;
+         case 0x20100207:
+           imgdata.lens.makernotes.MinFocal = (float)get2();
+@@ -6991,7 +6991,7 @@ void CLASS parse_makernote_0xc634(int base, int uptag, unsigned dng_writer)
+             imgdata.lens.makernotes.MaxFocal = imgdata.lens.makernotes.MinFocal;
+           break;
+         case 0x2010020a:
+-          imgdata.lens.makernotes.MaxAp4CurFocal = powf64(sqrt(2.0f), get2() / 256.0f);
++          imgdata.lens.makernotes.MaxAp4CurFocal = libraw_powf64(sqrt(2.0f), get2() / 256.0f);
+           break;
+         case 0x20100301:
+           imgdata.lens.makernotes.TeleconverterID = fgetc(ifp) << 8;
+@@ -7195,7 +7195,7 @@ void CLASS parse_makernote_0xc634(int base, int uptag, unsigned dng_writer)
+                   lid = (((ushort)table_buf[2])<<8) |
+                     ((ushort)table_buf[3]);
+                   imgdata.lens.makernotes.CurAp =
+-                    powf64(2.0f, ((float)lid/8.0f-1.0f)/2.0f);
++                    libraw_powf64(2.0f, ((float)lid/8.0f-1.0f)/2.0f);
+                 }
+               break;
+             case 1536:
+@@ -7669,7 +7669,7 @@ void CLASS parse_makernote (int base, int uptag)
+           }
+           break;
+         case 0x1002:
+-          imgdata.lens.makernotes.CurAp = powf64(2.0f, getreal(type)/2);
++          imgdata.lens.makernotes.CurAp = libraw_powf64(2.0f, getreal(type)/2);
+           break;
+         case 0x20100201:
+           {
+@@ -7694,10 +7694,10 @@ void CLASS parse_makernote (int base, int uptag)
+           fread(imgdata.lens.makernotes.Lens, MIN(len,127), 1, ifp);
+           break;
+         case 0x20100205:
+-          imgdata.lens.makernotes.MaxAp4MinFocal = powf64(sqrt(2.0f), get2() / 256.0f);
++          imgdata.lens.makernotes.MaxAp4MinFocal = libraw_powf64(sqrt(2.0f), get2() / 256.0f);
+           break;
+         case 0x20100206:
+-          imgdata.lens.makernotes.MaxAp4MaxFocal = powf64(sqrt(2.0f), get2() / 256.0f);
++          imgdata.lens.makernotes.MaxAp4MaxFocal = libraw_powf64(sqrt(2.0f), get2() / 256.0f);
+           break;
+         case 0x20100207:
+           imgdata.lens.makernotes.MinFocal = (float)get2();
+@@ -7708,7 +7708,7 @@ void CLASS parse_makernote (int base, int uptag)
+             imgdata.lens.makernotes.MaxFocal = imgdata.lens.makernotes.MinFocal;
+           break;
+         case 0x2010020a:
+-          imgdata.lens.makernotes.MaxAp4CurFocal = powf64(sqrt(2.0f), get2() / 256.0f);
++          imgdata.lens.makernotes.MaxAp4CurFocal = libraw_powf64(sqrt(2.0f), get2() / 256.0f);
+           break;
+         case 0x20100301:
+           imgdata.lens.makernotes.TeleconverterID = fgetc(ifp) << 8;
+@@ -7977,7 +7977,7 @@ void CLASS parse_makernote (int base, int uptag)
+                   lid = (((ushort)table_buf[2])<<8) |
+                     ((ushort)table_buf[3]);
+                   imgdata.lens.makernotes.CurAp =
+-                    powf64(2.0f, ((float)lid/8.0f-1.0f)/2.0f);
++                    libraw_powf64(2.0f, ((float)lid/8.0f-1.0f)/2.0f);
+                 }
+               break;
+             case 1536:
+@@ -8064,19 +8064,19 @@ void CLASS parse_makernote (int base, int uptag)
+       {
+         unsigned char cc;
+         fread(&cc,1,1,ifp);
+-        iso_speed = int(100.0 * powf64(2.0f,float(cc)/12.0-5.0));
++        iso_speed = int(100.0 * libraw_powf64(2.0f,float(cc)/12.0-5.0));
+       }
+     if (tag == 4 && len > 26 && len < 35) {
+       if ((i=(get4(),get2())) != 0x7fff && (!iso_speed || iso_speed == 65535))
+-	iso_speed = 50 * powf64(2.0, i/32.0 - 4);
++	iso_speed = 50 * libraw_powf64(2.0, i/32.0 - 4);
+ #ifdef LIBRAW_LIBRARY_BUILD
+       get4();
+ #else
+       if ((i=(get2(),get2())) != 0x7fff && !aperture)
+-	aperture = powf64(2.0, i/64.0);
++	aperture = libraw_powf64(2.0, i/64.0);
+ #endif
+       if ((i=get2()) != 0xffff && !shutter)
+-	shutter = powf64(2.0, (short) i/-32.0);
++	shutter = libraw_powf64(2.0, (short) i/-32.0);
+       wbi = (get2(),get2());
+       shot_order = (get2(),get2());
+     }
+@@ -8579,7 +8579,7 @@ void CLASS parse_exif (int base)
+         imgdata.lens.Lens[0] = 0;
+       break;
+     case 0x9205:
+-      imgdata.lens.EXIF_MaxAp = powf64(2.0f, (getreal(type) / 2.0f));
++      imgdata.lens.EXIF_MaxAp = libraw_powf64(2.0f, (getreal(type) / 2.0f));
+       break;
+ #endif
+       case 33434:  shutter = getreal(type);		break;
+@@ -8592,10 +8592,10 @@ void CLASS parse_exif (int base)
+       case 36867:
+       case 36868:  get_timestamp(0);			break;
+       case 37377:  if ((expo = -getreal(type)) < 128 && shutter == 0.)
+-          shutter = powf64(2.0, expo);		break;
++          shutter = libraw_powf64(2.0, expo);		break;
+       case 37378:                                       // 0x9202 ApertureValue
+         if ((fabs(ape = getreal(type))<256.0) && (!aperture))
+-          aperture = powf64(2.0, ape/2);
++          aperture = libraw_powf64(2.0, ape/2);
+         break;
+       case 37385:  flash_used = getreal(type);          break;
+       case 37386:  focal_len = getreal(type);		break;
+@@ -9196,7 +9196,7 @@ int CLASS parse_tiff_ifd (int base)
+         imgdata.lens.Lens[0] = 0;
+       break;
+     case 0x9205:
+-				imgdata.lens.EXIF_MaxAp = powf64(2.0f, (getreal(type) / 2.0f));
++				imgdata.lens.EXIF_MaxAp = libraw_powf64(2.0f, (getreal(type) / 2.0f));
+       break;
+ // IB end
+ #endif
+@@ -9970,22 +9970,22 @@ void CLASS parse_ciff (int offset, int length, int depth)
+       thumb_length = len;
+     }
+     if (type == 0x1818) {
+-      shutter = powf64(2.0f, -int_to_float((get4(),get4())));
+-      aperture = powf64(2.0f, int_to_float(get4())/2);
++      shutter = libraw_powf64(2.0f, -int_to_float((get4(),get4())));
++      aperture = libraw_powf64(2.0f, int_to_float(get4())/2);
+ #ifdef LIBRAW_LIBRARY_BUILD
+ 			imgdata.lens.makernotes.CurAp = aperture;
+ #endif
+     }
+     if (type == 0x102a) {
+ 			//      iso_speed = pow (2.0, (get4(),get2())/32.0 - 4) * 50;
+-      iso_speed = powf64(2.0f, ((get2(),get2()) + get2())/32.0f - 5.0f) * 100.0f;
++      iso_speed = libraw_powf64(2.0f, ((get2(),get2()) + get2())/32.0f - 5.0f) * 100.0f;
+ #ifdef LIBRAW_LIBRARY_BUILD
+       aperture  = _CanonConvertAperture((get2(),get2()));
+       imgdata.lens.makernotes.CurAp = aperture;
+ #else
+-      aperture  = powf64(2.0, (get2(),(short)get2())/64.0);
++      aperture  = libraw_powf64(2.0, (get2(),(short)get2())/64.0);
+ #endif
+-      shutter   = powf64(2.0,-((short)get2())/32.0);
++      shutter   = libraw_powf64(2.0,-((short)get2())/32.0);
+       wbi = (get2(),get2());
+       if (wbi > 17) wbi = 0;
+       fseek (ifp, 32, SEEK_CUR);
+@@ -10189,8 +10189,8 @@ void CLASS parse_phase_one (int base)
+       setPhaseOneFeatures(unique_id);
+       break;
+     case 0x0401:
+-      if (type == 4) imgdata.lens.makernotes.CurAp =  powf64(2.0f, (int_to_float(data)/2.0f));
+-      else imgdata.lens.makernotes.CurAp = powf64(2.0f, (getreal(type)/2.0f));
++      if (type == 4) imgdata.lens.makernotes.CurAp =  libraw_powf64(2.0f, (int_to_float(data)/2.0f));
++      else imgdata.lens.makernotes.CurAp = libraw_powf64(2.0f, (getreal(type)/2.0f));
+       break;
+     case 0x0403:
+       if (type == 4) imgdata.lens.makernotes.CurFocal =  int_to_float(data);
+@@ -10204,16 +10204,16 @@ void CLASS parse_phase_one (int base)
+       break;
+     case 0x0414:
+       if (type == 4) {
+-      	imgdata.lens.makernotes.MaxAp4CurFocal = powf64(2.0f, (int_to_float(data)/2.0f));
++      	imgdata.lens.makernotes.MaxAp4CurFocal = libraw_powf64(2.0f, (int_to_float(data)/2.0f));
+       } else {
+-        imgdata.lens.makernotes.MaxAp4CurFocal = powf64(2.0f, (getreal(type) / 2.0f));
++        imgdata.lens.makernotes.MaxAp4CurFocal = libraw_powf64(2.0f, (getreal(type) / 2.0f));
+       }
+       break;
+     case 0x0415:
+       if (type == 4) {
+-      	imgdata.lens.makernotes.MinAp4CurFocal = powf64(2.0f, (int_to_float(data)/2.0f));
++      	imgdata.lens.makernotes.MinAp4CurFocal = libraw_powf64(2.0f, (int_to_float(data)/2.0f));
+       } else {
+-        imgdata.lens.makernotes.MinAp4CurFocal = powf64(2.0f, (getreal(type) / 2.0f));
++        imgdata.lens.makernotes.MinAp4CurFocal = libraw_powf64(2.0f, (getreal(type) / 2.0f));
+       }
+       break;
+     case 0x0416:
+@@ -12119,15 +12119,15 @@ void CLASS identify()
+       case 18: iso_speed = 320; break;
+       case 19: iso_speed = 400; break;
+     }
+-    shutter = powf64(2.0f, (((float)get4())/8.0f)) / 16000.0f;
++    shutter = libraw_powf64(2.0f, (((float)get4())/8.0f)) / 16000.0f;
+     FORC4 cam_mul[c ^ (c >> 1)] = get4();
+     fseek (ifp, 88, SEEK_SET);
+-    aperture = powf64(2.0f, ((float)get4())/16.0f);
++    aperture = libraw_powf64(2.0f, ((float)get4())/16.0f);
+     fseek (ifp, 112, SEEK_SET);
+     focal_len = get4();
+ #ifdef LIBRAW_LIBRARY_BUILD
+     fseek (ifp, 104, SEEK_SET);
+-    imgdata.lens.makernotes.MaxAp4CurFocal = powf64(2.0f, ((float)get4())/16.0f);
++    imgdata.lens.makernotes.MaxAp4CurFocal = libraw_powf64(2.0f, ((float)get4())/16.0f);
+     fseek (ifp, 124, SEEK_SET);
+     fread(imgdata.lens.makernotes.Lens, 32, 1, ifp);
+     imgdata.lens.makernotes.CameraMount = LIBRAW_MOUNT_Contax_N;
+-- 
+2.14.3
+
Signature · 1 line
2.14.3
#3 lore raw

PATCH 4/5 package/glibc: remove GLIBC_SRC_SUBDIR

In reply to Romain Naour

GLIBC_SRC_SUBDIR was needed when Buildroot supported eglibc which stored all sources in a sub-directory. It was not removed by the commit removing eglibc support [1].

[1] 500de2598a4902a0e9ec53f7b475046ff17c6681

---
 package/glibc/glibc.mk | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
package/glibc/glibc.mk +2 -4
index fd79d84f15..cf4bdec065 100644
@@ -20,10 +20,8 @@ GLIBC_VERSION = glibc-2.26-107-g73a92363619e52c458146e903dfb9b1ba823aa40
 GLIBC_SITE = $(call github,bminor,glibc,$(GLIBC_VERSION))
 endif
 
-GLIBC_SRC_SUBDIR = .
-
 GLIBC_LICENSE = GPL-2.0+ (programs), LGPL-2.1+, BSD-3-Clause, MIT (library)
-GLIBC_LICENSE_FILES = $(addprefix $(GLIBC_SRC_SUBDIR)/,COPYING COPYING.LIB LICENSES)
+GLIBC_LICENSE_FILES = COPYING COPYING.LIB LICENSES
 
 # glibc is part of the toolchain so disable the toolchain dependency
 GLIBC_ADD_TOOLCHAIN_DEPENDENCY = NO
@@ -86,7 +84,7 @@ define GLIBC_CONFIGURE_CMDS
 		$(TARGET_CONFIGURE_OPTS) \
 		CFLAGS="-O2 $(GLIBC_EXTRA_CFLAGS)" CPPFLAGS="" \
 		CXXFLAGS="-O2 $(GLIBC_EXTRA_CFLAGS)" \
-		$(SHELL) $(@D)/$(GLIBC_SRC_SUBDIR)/configure \
+		$(SHELL) $(@D)/configure \
 		ac_cv_path_BASH_SHELL=/bin/bash \
 		libc_cv_forced_unwind=yes \
 		libc_cv_ssp=no \
Signature · 1 line
2.14.3
#4 lore raw

PATCH 4/5 package/glibc: remove GLIBC_SRC_SUBDIR

Romain, All,

On 2018-02-05 21:57 +0100, Romain Naour spake thusly:

Quoted 7 lines

GLIBC_SRC_SUBDIR was needed when Buildroot supported eglibc which stored all sources in a sub-directory. It was not removed by the commit removing eglibc support [1].

[1] 500de2598a4902a0e9ec53f7b475046ff17c6681

This can also be applied to master, probably...

Regards,
Yann E. MORIN.

Quoted 36 lines
---
 package/glibc/glibc.mk | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
package/glibc/glibc.mk +2 -4
index fd79d84f15..cf4bdec065 100644
@@ -20,10 +20,8 @@ GLIBC_VERSION = glibc-2.26-107-g73a92363619e52c458146e903dfb9b1ba823aa40
 GLIBC_SITE = $(call github,bminor,glibc,$(GLIBC_VERSION))
 endif
 
-GLIBC_SRC_SUBDIR = .
-
 GLIBC_LICENSE = GPL-2.0+ (programs), LGPL-2.1+, BSD-3-Clause, MIT (library)
-GLIBC_LICENSE_FILES = $(addprefix $(GLIBC_SRC_SUBDIR)/,COPYING COPYING.LIB LICENSES)
+GLIBC_LICENSE_FILES = COPYING COPYING.LIB LICENSES
 
 # glibc is part of the toolchain so disable the toolchain dependency
 GLIBC_ADD_TOOLCHAIN_DEPENDENCY = NO
@@ -86,7 +84,7 @@ define GLIBC_CONFIGURE_CMDS
 		$(TARGET_CONFIGURE_OPTS) \
 		CFLAGS="-O2 $(GLIBC_EXTRA_CFLAGS)" CPPFLAGS="" \
 		CXXFLAGS="-O2 $(GLIBC_EXTRA_CFLAGS)" \
-		$(SHELL) $(@D)/$(GLIBC_SRC_SUBDIR)/configure \
+		$(SHELL) $(@D)/configure \
 		ac_cv_path_BASH_SHELL=/bin/bash \
 		libc_cv_forced_unwind=yes \
 		libc_cv_ssp=no \
Signature · 6 lines
2.14.3

_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot
Signature · 6 lines
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'
#5 lore raw

PATCH 5/5 package/glibc: bump to 2.27

In reply to Romain Naour

See: https://sourceware.org/ml/libc-announce/2018/msg00000.html
https://sourceware.org/glibc/wiki/Release/2.27

Add hash files

---
 package/glibc/glibc.hash | 6 +++++-
 package/glibc/glibc.mk   | 2 +-
 2 files changed, 6 insertions(+), 2 deletions(-)
package/glibc/glibc.hash +5 -1
index f3a6577d2a..86d3bb56dd 100644
@@ -1,4 +1,8 @@
 # Locally calculated (fetched from Github)
-sha256  0766875391224153502c5542a71b6e46db53b44691078b3130e1a0df41586430     glibc-glibc-2.26-107-g73a92363619e52c458146e903dfb9b1ba823aa40.tar.gz
+sha256  a74489d14f4017bee6a6c6fe76f1de0dbf7d66c8695116de5aadd141c4757892     glibc-glibc-2.27.tar.gz
 # Locally calculated (fetched from Github)
 sha256  5aa9adeac09727db0b8a52794186563771e74d70410e9fd86431e339953fd4bb     glibc-arc-2017.09-release.tar.gz
+
+sha256  8177f97513213526df2cf6184d8ff986c675afb514d4e68a404010521b880643  COPYING
+sha256  dc626520dcd53a22f727af3ee42c770e56c97a64fe3adb063799d8ab032fe551  COPYING.LIB
+sha256  61abdd6930c9c599062d89e916b3e7968783879b6be0ee1c6229dd6169def431  LICENSES
package/glibc/glibc.mk +1 -1
index cf4bdec065..b674191b22 100644
@@ -10,7 +10,7 @@ GLIBC_SITE = $(call github,foss-for-synopsys-dwc-arc-processors,glibc,$(GLIBC_VE
 else
 # Generate version string using:
 #   git describe --match 'glibc-*' --abbrev=40 origin/release/MAJOR.MINOR/master
-GLIBC_VERSION = glibc-2.26-107-g73a92363619e52c458146e903dfb9b1ba823aa40
+GLIBC_VERSION = glibc-2.27
 # Upstream doesn't officially provide an https download link.
 # There is one (https://sourceware.org/git/glibc.git) but it's not reliable,
 # sometimes the connection times out. So use an unofficial github mirror.
Signature · 1 line
2.14.3
#6 lore raw

PATCH 5/5 package/glibc: bump to 2.27

Hi Romain,

Quoted 2 lines · Romain Naour

See: https://sourceware.org/ml/libc-announce/2018/msg00000.html
https://sourceware.org/glibc/wiki/Release/2.27

Note that this is a security bump fixing CVE-2017-1000408, CVE-2017-1000409, CVE-2017-16997, CVE-2018-1000001, and CVE-2018-6485.

Quoted 1 line

Add hash files

You mean license files hash, right?

baruch

Quoted 33 lines
---
 package/glibc/glibc.hash | 6 +++++-
 package/glibc/glibc.mk   | 2 +-
 2 files changed, 6 insertions(+), 2 deletions(-)
package/glibc/glibc.hash +5 -1
index f3a6577d2a..86d3bb56dd 100644
@@ -1,4 +1,8 @@
 # Locally calculated (fetched from Github)
-sha256  0766875391224153502c5542a71b6e46db53b44691078b3130e1a0df41586430     glibc-glibc-2.26-107-g73a92363619e52c458146e903dfb9b1ba823aa40.tar.gz
+sha256  a74489d14f4017bee6a6c6fe76f1de0dbf7d66c8695116de5aadd141c4757892     glibc-glibc-2.27.tar.gz
 # Locally calculated (fetched from Github)
 sha256  5aa9adeac09727db0b8a52794186563771e74d70410e9fd86431e339953fd4bb     glibc-arc-2017.09-release.tar.gz
+
+sha256  8177f97513213526df2cf6184d8ff986c675afb514d4e68a404010521b880643  COPYING
+sha256  dc626520dcd53a22f727af3ee42c770e56c97a64fe3adb063799d8ab032fe551  COPYING.LIB
+sha256  61abdd6930c9c599062d89e916b3e7968783879b6be0ee1c6229dd6169def431  LICENSES
package/glibc/glibc.mk +1 -1
index cf4bdec065..b674191b22 100644
@@ -10,7 +10,7 @@ GLIBC_SITE = $(call github,foss-for-synopsys-dwc-arc-processors,glibc,$(GLIBC_VE
 else
 # Generate version string using:
 #   git describe --match 'glibc-*' --abbrev=40 origin/release/MAJOR.MINOR/master
-GLIBC_VERSION = glibc-2.26-107-g73a92363619e52c458146e903dfb9b1ba823aa40
+GLIBC_VERSION = glibc-2.27
 # Upstream doesn't officially provide an https download link.
 # There is one (https://sourceware.org/git/glibc.git) but it's not reliable,
 # sometimes the connection times out. So use an unofficial github mirror.
Signature · 3 lines
     http://baruch.siach.name/blog/                  ~. .~   Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
   - baruch at tkos.co.il - tel: +972.52.368.4656, http://www.tkos.co.il -
#7 lore raw

PATCH 5/5 package/glibc: bump to 2.27

Hi Baruch,

Le 05/02/2018 ? 22:01, Baruch Siach a ?crit?:

Quoted 8 lines

Hi Romain,

Quoted 2 lines · Romain Naour

See: https://sourceware.org/ml/libc-announce/2018/msg00000.html
https://sourceware.org/glibc/wiki/Release/2.27

Note that this is a security bump fixing CVE-2017-1000408, CVE-2017-1000409, CVE-2017-16997, CVE-2018-1000001, and CVE-2018-6485.

Ok thanks, I'll add them.

Quoted 4 lines
Quoted 2 lines

Add hash files

You mean license files hash, right?

Yes Indeed.

Best regards,
Romain

Quoted 37 lines

baruch

Quoted 33 lines
---
 package/glibc/glibc.hash | 6 +++++-
 package/glibc/glibc.mk   | 2 +-
 2 files changed, 6 insertions(+), 2 deletions(-)
package/glibc/glibc.hash +5 -1
index f3a6577d2a..86d3bb56dd 100644
@@ -1,4 +1,8 @@
 # Locally calculated (fetched from Github)
-sha256  0766875391224153502c5542a71b6e46db53b44691078b3130e1a0df41586430     glibc-glibc-2.26-107-g73a92363619e52c458146e903dfb9b1ba823aa40.tar.gz
+sha256  a74489d14f4017bee6a6c6fe76f1de0dbf7d66c8695116de5aadd141c4757892     glibc-glibc-2.27.tar.gz
 # Locally calculated (fetched from Github)
 sha256  5aa9adeac09727db0b8a52794186563771e74d70410e9fd86431e339953fd4bb     glibc-arc-2017.09-release.tar.gz
+
+sha256  8177f97513213526df2cf6184d8ff986c675afb514d4e68a404010521b880643  COPYING
+sha256  dc626520dcd53a22f727af3ee42c770e56c97a64fe3adb063799d8ab032fe551  COPYING.LIB
+sha256  61abdd6930c9c599062d89e916b3e7968783879b6be0ee1c6229dd6169def431  LICENSES
package/glibc/glibc.mk +1 -1
index cf4bdec065..b674191b22 100644
@@ -10,7 +10,7 @@ GLIBC_SITE = $(call github,foss-for-synopsys-dwc-arc-processors,glibc,$(GLIBC_VE
 else
 # Generate version string using:
 #   git describe --match 'glibc-*' --abbrev=40 origin/release/MAJOR.MINOR/master
-GLIBC_VERSION = glibc-2.26-107-g73a92363619e52c458146e903dfb9b1ba823aa40
+GLIBC_VERSION = glibc-2.27
 # Upstream doesn't officially provide an https download link.
 # There is one (https://sourceware.org/git/glibc.git) but it's not reliable,
 # sometimes the connection times out. So use an unofficial github mirror.
#8 lore raw

PATCH 5/5 package/glibc: bump to 2.27

In reply to Baruch Siach

Baruch, All,

On 2018-02-05 23:01 +0200, Baruch Siach spake thusly:

Quoted 5 lines · Romain Naour
Quoted 2 lines

See: https://sourceware.org/ml/libc-announce/2018/msg00000.html
https://sourceware.org/glibc/wiki/Release/2.27

Note that this is a security bump fixing CVE-2017-1000408, CVE-2017-1000409, CVE-2017-16997, CVE-2018-1000001, and CVE-2018-6485.

There are 10 CVE listed in the release annoucement mail, but you list only five here. Why only those?

Do we want to list all the CVEs fixed in a release? And if we don't list all, why do we even list only a subset?

I don't think we should, especially since the release mail is linked to the commit log and has all the details.

Regards,
Yann E. MORIN.

Quoted 48 lines
Quoted 1 line

Add hash files

You mean license files hash, right?

baruch

Quoted 33 lines
---
 package/glibc/glibc.hash | 6 +++++-
 package/glibc/glibc.mk   | 2 +-
 2 files changed, 6 insertions(+), 2 deletions(-)
package/glibc/glibc.hash +5 -1
index f3a6577d2a..86d3bb56dd 100644
@@ -1,4 +1,8 @@
 # Locally calculated (fetched from Github)
-sha256  0766875391224153502c5542a71b6e46db53b44691078b3130e1a0df41586430     glibc-glibc-2.26-107-g73a92363619e52c458146e903dfb9b1ba823aa40.tar.gz
+sha256  a74489d14f4017bee6a6c6fe76f1de0dbf7d66c8695116de5aadd141c4757892     glibc-glibc-2.27.tar.gz
 # Locally calculated (fetched from Github)
 sha256  5aa9adeac09727db0b8a52794186563771e74d70410e9fd86431e339953fd4bb     glibc-arc-2017.09-release.tar.gz
+
+sha256  8177f97513213526df2cf6184d8ff986c675afb514d4e68a404010521b880643  COPYING
+sha256  dc626520dcd53a22f727af3ee42c770e56c97a64fe3adb063799d8ab032fe551  COPYING.LIB
+sha256  61abdd6930c9c599062d89e916b3e7968783879b6be0ee1c6229dd6169def431  LICENSES
package/glibc/glibc.mk +1 -1
index cf4bdec065..b674191b22 100644
@@ -10,7 +10,7 @@ GLIBC_SITE = $(call github,foss-for-synopsys-dwc-arc-processors,glibc,$(GLIBC_VE
 else
 # Generate version string using:
 #   git describe --match 'glibc-*' --abbrev=40 origin/release/MAJOR.MINOR/master
-GLIBC_VERSION = glibc-2.26-107-g73a92363619e52c458146e903dfb9b1ba823aa40
+GLIBC_VERSION = glibc-2.27
 # Upstream doesn't officially provide an https download link.
 # There is one (https://sourceware.org/git/glibc.git) but it's not reliable,
 # sometimes the connection times out. So use an unofficial github mirror.
Signature · 7 lines
     http://baruch.siach.name/blog/                  ~. .~   Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
   - baruch at tkos.co.il - tel: +972.52.368.4656, http://www.tkos.co.il -
_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot
Signature · 6 lines
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'
#9 lore raw

PATCH 5/5 package/glibc: bump to 2.27

Quoted 10 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Yann" == Yann E MORIN <[email protected]> writes:

Baruch, All,
On 2018-02-05 23:01 +0200, Baruch Siach spake thusly:

Quoted 5 lines · Romain Naour
Quoted 2 lines

See: https://sourceware.org/ml/libc-announce/2018/msg00000.html
https://sourceware.org/glibc/wiki/Release/2.27

Note that this is a security bump fixing CVE-2017-1000408, CVE-2017-1000409, CVE-2017-16997, CVE-2018-1000001, and CVE-2018-6485.

Quoted 8 lines

There are 10 CVE listed in the release annoucement mail, but you list only five here. Why only those?

Do we want to list all the CVEs fixed in a release? And if we don't list all, why do we even list only a subset?

I don't think we should, especially since the release mail is linked to the commit log and has all the details.

For security bumps we normally DO list the CVEs for ease of use, E.G.:

commit d52cd750c762c78ebcf8623fab8a9c43c9419625

Author: Peter Korsgaard <[email protected]>
Date:   Sun Jan 28 20:23:02 2018 +0100
    wireshark: security bump to version 2.2.12

    Fixes the following security issues:

    CVE-2017-17997: MRDISC dissector crash
    https://www.wireshark.org/security/wnpa-sec-2018-02.html

    CVE-2018-5334: IxVeriWave file parser crash
    https://www.wireshark.org/security/wnpa-sec-2018-03.html

    CVE-2018-5335: WCP dissector crash
    https://www.wireshark.org/security/wnpa-sec-2018-04.html

    CVE-2018-5336: Multiple dissectors could crash
    https://www.wireshark.org/security/wnpa-sec-2018-01.html

    For more information, see the release notes:
    https://www.wireshark.org/docs/relnotes/wireshark-2.2.12.html

    While we are at it, also add as hash for license file.

    Signed-off-by: Peter Korsgaard <[email protected]>

So I prefer to do that as well here.

Signature · 1 line
Bye, Peter Korsgaard
#10 lore raw

PATCH 5/5 package/glibc: bump to 2.27

In reply to Yann E. MORIN

Hi Yann,

Quoted 9 lines · Yann E. MORIN

On 2018-02-05 23:01 +0200, Baruch Siach spake thusly:

Quoted 5 lines · Romain Naour
Quoted 2 lines

See: https://sourceware.org/ml/libc-announce/2018/msg00000.html
https://sourceware.org/glibc/wiki/Release/2.27

Note that this is a security bump fixing CVE-2017-1000408, CVE-2017-1000409, CVE-2017-16997, CVE-2018-1000001, and CVE-2018-6485.

There are 10 CVE listed in the release annoucement mail, but you list only five here. Why only those?

Because all other CVEs were fixed in the 2.26 stable branch that we currently use. See 971ed9653e (glibc: security bump to the latest 2.26 branch).

Quoted 2 lines

Do we want to list all the CVEs fixed in a release? And if we don't list all, why do we even list only a subset?

We do list all. See above.

Quoted 2 lines

I don't think we should, especially since the release mail is linked to the commit log and has all the details.

baruch

Signature · 3 lines
     http://baruch.siach.name/blog/                  ~. .~   Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
   - baruch at tkos.co.il - tel: +972.52.368.4656, http://www.tkos.co.il -
#11 lore raw

PATCH 5/5 package/glibc: bump to 2.27

In reply to Baruch Siach
Quoted 7 lines · Baruch Siach

Hi Romain,

Quoted 2 lines · Romain Naour

See: https://sourceware.org/ml/libc-announce/2018/msg00000.html
https://sourceware.org/glibc/wiki/Release/2.27

Note that this is a security bump fixing CVE-2017-1000408, CVE-2017-1000409, CVE-2017-16997, CVE-2018-1000001, and CVE-2018-6485.

Even though this release fixes a number of CVEs, I wouldn't call it a security
bump. Indeed, it also makes a number of potentially breaking feature updates, cfr. the memfd_create() change.

So, I would indeed mention the CVE numbers in the commit message, but not put
"security bump" in the title so that it doesn't mindlessly get applied to LTS branches.

Now, in this particular case we made enough noise about it that it really
doesn't matter what goes into the subject line :-) However I think it's good to converge on some conventions on how to tag LTS things.

Regards,
Arnout

Signature · 6 lines
Arnout Vandecappelle                          arnout at mind be
Senior Embedded Software Architect            +32-16-286500
Essensium/Mind                                http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium           BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint:  7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF
#12 lore raw

PATCH 5/5 package/glibc: bump to 2.27

Hi Arnout,

Quoted 11 lines · Arnout Vandecappelle
Quoted 5 lines · Baruch Siach
Quoted 2 lines · Romain Naour

See: https://sourceware.org/ml/libc-announce/2018/msg00000.html
https://sourceware.org/glibc/wiki/Release/2.27

Note that this is a security bump fixing CVE-2017-1000408, CVE-2017-1000409, CVE-2017-16997, CVE-2018-1000001, and CVE-2018-6485.

Even though this release fixes a number of CVEs, I wouldn't call it a security
bump. Indeed, it also makes a number of potentially breaking feature updates, cfr. the memfd_create() change.

Quoted 7 lines

So, I would indeed mention the CVE numbers in the commit message, but not put
"security bump" in the title so that it doesn't mindlessly get applied to LTS branches.

Now, in this particular case we made enough noise about it that it really
doesn't matter what goes into the subject line :-) However I think it's good to converge on some conventions on how to tag LTS things.

I agree that this bump might not be suitable for the maintenance branches. But I think we should consider it for the master branch, especially since we are still early in the -rc cycle.

baruch

Signature · 3 lines
     http://baruch.siach.name/blog/                  ~. .~   Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
   - baruch at tkos.co.il - tel: +972.52.368.4656, http://www.tkos.co.il -
#13 lore raw

PATCH 5/5 package/glibc: bump to 2.27

Quoted 18 lines · Baruch Siach

Hi Arnout,

Quoted 15 lines · Arnout Vandecappelle
Quoted 5 lines · Baruch Siach
Quoted 2 lines · Romain Naour

See: https://sourceware.org/ml/libc-announce/2018/msg00000.html
https://sourceware.org/glibc/wiki/Release/2.27

Note that this is a security bump fixing CVE-2017-1000408, CVE-2017-1000409, CVE-2017-16997, CVE-2018-1000001, and CVE-2018-6485.

Even though this release fixes a number of CVEs, I wouldn't call it a security
bump. Indeed, it also makes a number of potentially breaking feature updates, cfr. the memfd_create() change.

So, I would indeed mention the CVE numbers in the commit message, but not put
"security bump" in the title so that it doesn't mindlessly get applied to LTS branches.

Quoted 7 lines
Quoted 3 lines

Now, in this particular case we made enough noise about it that it really
doesn't matter what goes into the subject line :-) However I think it's good to converge on some conventions on how to tag LTS things.

I agree that this bump might not be suitable for the maintenance branches. But I think we should consider it for the master branch, especially since we are still early in the -rc cycle.

After a bit of discussion at the BR developer meeting: if it fixes CVEs, it
should be called a "security bump". So ignore my earlier comment. It something that should be backported to LTS.

Regards,
Arnout

Signature · 6 lines
Arnout Vandecappelle                          arnout at mind be
Senior Embedded Software Architect            +32-16-286500
Essensium/Mind                                http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium           BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint:  7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF
#14 lore raw

PATCH 5/5 package/glibc: bump to 2.27

In reply to Baruch Siach
Quoted 19 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Baruch" == Baruch Siach <[email protected]> writes:

Hi Arnout,

Quoted 15 lines · Arnout Vandecappelle
Quoted 5 lines · Baruch Siach
Quoted 2 lines · Romain Naour

See: https://sourceware.org/ml/libc-announce/2018/msg00000.html
https://sourceware.org/glibc/wiki/Release/2.27

Note that this is a security bump fixing CVE-2017-1000408, CVE-2017-1000409, CVE-2017-16997, CVE-2018-1000001, and CVE-2018-6485.

Even though this release fixes a number of CVEs, I wouldn't call it a security bump. Indeed, it also makes a number of potentially breaking feature updates, cfr. the memfd_create() change.

So, I would indeed mention the CVE numbers in the commit message, but not put "security bump" in the title so that it doesn't mindlessly get applied to LTS branches.

Quoted 7 lines
Quoted 3 lines

Now, in this particular case we made enough noise about it that it really doesn't matter what goes into the subject line :-) However I think it's good to converge on some conventions on how to tag LTS things.

I agree that this bump might not be suitable for the maintenance branches. But I think we should consider it for the master branch, especially since we are still early in the -rc cycle.

Possibly, yes. Lets see how much blows up on next. Do you know if (some of) these issues are also fixed on the 2.26 branch?

Signature · 1 line
Bye, Peter Korsgaard
#15 lore raw

PATCH 5/5 package/glibc: bump to 2.27

Hi Peter,

Quoted 22 lines · Peter Korsgaard
Quoted 21 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Baruch" == Baruch Siach <[email protected]> writes:

Quoted 18 lines · Arnout Vandecappelle
Quoted 5 lines · Baruch Siach
Quoted 2 lines · Romain Naour

See: https://sourceware.org/ml/libc-announce/2018/msg00000.html
https://sourceware.org/glibc/wiki/Release/2.27

Note that this is a security bump fixing CVE-2017-1000408, CVE-2017-1000409, CVE-2017-16997, CVE-2018-1000001, and CVE-2018-6485.

Even though this release fixes a number of CVEs, I wouldn't call it a security bump. Indeed, it also makes a number of potentially breaking feature updates, cfr. the memfd_create() change.

So, I would indeed mention the CVE numbers in the commit message, but not put "security bump" in the title so that it doesn't mindlessly get applied to LTS branches.

Now, in this particular case we made enough noise about it that it really doesn't matter what goes into the subject line :-) However I think it's good to converge on some conventions on how to tag LTS things.

Quoted 6 lines
Quoted 3 lines

I agree that this bump might not be suitable for the maintenance branches. But I think we should consider it for the master branch, especially since we are still early in the -rc cycle.

Possibly, yes. Lets see how much blows up on next. Do you know if (some of) these issues are also fixed on the 2.26 branch?

As far as I know all these issues are fixed in the 2.26 stable branch. See the NEWS file in that branch.

baruch

Signature · 3 lines
     http://baruch.siach.name/blog/                  ~. .~   Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
   - baruch at tkos.co.il - tel: +972.52.368.4656, http://www.tkos.co.il -
#16 lore raw

PATCH 5/5 package/glibc: bump to 2.27

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Baruch" == Baruch Siach <[email protected]> writes:

Hi,

Quoted 5 lines
Quoted 2 lines

Possibly, yes. Lets see how much blows up on next. Do you know if (some of) these issues are also fixed on the 2.26 branch?

As far as I know all these issues are fixed in the 2.26 stable branch. See the NEWS file in that branch.

Ok, but only when we bump to the latest version on the 2.26 branch
- E.G.:

git diff 73a92363619e52c458146e903dfb9b1ba823aa40.. -- NEWS

CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads to the allocation of too much memory. (This is not a security bug per se, it is mentioned here only because of the CVE assignment.) Reported by Qualys.

CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation of the number of search path components. (This is not a security vulnerability per se because no trust boundary is crossed if the fix for CVE-2017-1000366 has been applied, but it is mentioned here only because of the CVE assignment.) Reported by Qualys.

CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN for AT_SECURE or SUID binaries could be used to load libraries from the current directory.

CVE-2018-1000001: Buffer underflow in realpath function when getcwd function succeeds without returning an absolute path due to unexpected behaviour of the Linux kernel getcwd syscall. Reported by halfdog.

I don't see any reference to CVE-2018-6485 though.

I'll send a patch to bump the version.

Signature · 1 line
Bye, Peter Korsgaard
#17 lore raw

PATCH 5/5 package/glibc: bump to 2.27

Hi Peter,

Quoted 30 lines · Peter Korsgaard
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Baruch" == Baruch Siach <[email protected]> writes:

Hi,

Quoted 5 lines
Quoted 2 lines

Possibly, yes. Lets see how much blows up on next. Do you know if (some of) these issues are also fixed on the 2.26 branch?

As far as I know all these issues are fixed in the 2.26 stable branch. See the NEWS file in that branch.

Ok, but only when we bump to the latest version on the 2.26 branch
- E.G.:

git diff 73a92363619e52c458146e903dfb9b1ba823aa40.. -- NEWS

CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads to the allocation of too much memory. (This is not a security bug per se, it is mentioned here only because of the CVE assignment.) Reported by Qualys.

CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation of the number of search path components. (This is not a security vulnerability per se because no trust boundary is crossed if the fix for CVE-2017-1000366 has been applied, but it is mentioned here only because of the CVE assignment.) Reported by Qualys.

CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN for AT_SECURE or SUID binaries could be used to load libraries from the current directory.

Quoted 5 lines

CVE-2018-1000001: Buffer underflow in realpath function when getcwd function succeeds without returning an absolute path due to unexpected behaviour of the Linux kernel getcwd syscall. Reported by halfdog.

I don't see any reference to CVE-2018-6485 though.

Right. CVE-2018-6485 and CVE-2018-6551 are fixed in 2.27, but not (yet?) in the 2.26 release branch.

Quoted 1 line

I'll send a patch to bump the version.

baruch

Signature · 3 lines
     http://baruch.siach.name/blog/                  ~. .~   Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
   - baruch at tkos.co.il - tel: +972.52.368.4656, http://www.tkos.co.il -
#18 lore raw

PATCH 1/5 package/x11r7/xlib_libxshmfence: fix build with glibc 2.27

In reply to Romain Naour

Romain, All,

On 2018-02-05 21:57 +0100, Romain Naour spake thusly:

Quoted 16 lines

With glibc 2.27 memfd_create() is inside a _GNU_SOURCE guard, so call AC_USE_SYSTEM_EXTENSIONS to get this defined.

---
 ...onfigure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch | 30 ++++++++++++++++++++++
 .../x11r7/xlib_libxshmfence/xlib_libxshmfence.mk   |  3 +++
 2 files changed, 33 insertions(+)
 create mode 100644 package/x11r7/xlib_libxshmfence/0001-configure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch
package/x11r7/xlib_libxshmfence/0001-configure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch new file +0 -0
new file mode 100644
index 0000000000..f24b1da5dd

package/x11r7/xlib_libxshmfence/0001-configure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch line 10

Quoted 10 lines
package/x11r7/xlib_libxshmfence/0001-configure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch new file +10 -0
@@ -0,0 +1,30 @@
+From 0b550a4e7acf02d3478602848f6afbfcbfb0d4b2 Mon Sep 17 00:00:00 2001
+From: Ross Burton <[email protected]>
+Date: Mon, 29 Jan 2018 16:24:36 +0000
+Subject: [PATCH] configure.ac: call AC_USE_SYSTEM_EXTENSIONS
+
+With glibc 2.27 memfd_create() is inside a _GNU_SOURCE guard, so call
+AC_USE_SYSTEM_EXTENSIONS to get this defined.
+
+Signed-off-by: Ross Burton <[email protected]>
+Signed-off-by: Romain Naour <[email protected]>

This looks liek it is a backport from upstream, right? If so, please say so; in such a case I like to add something like:

[yann: backport from upstream URL-or-commit]
Signed-off-by ME

Regards,
Yann E. MORIN.

Quoted 39 lines
package/x11r7/xlib_libxshmfence/0001-configure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch new file +20 -0
+---
+ configure.ac | 2 ++
+ 1 file changed, 2 insertions(+)
+
+diff --git a/configure.ac b/configure.ac
+index 55772d0..ed77e6d 100644
+--- a/configure.ac
++++ b/configure.ac
+@@ -28,6 +28,8 @@ AC_INIT([libxshmfence], [1.2],
+ AC_CONFIG_SRCDIR([Makefile.am])
+ AC_CONFIG_HEADERS([config.h])
+ 
++AC_USE_SYSTEM_EXTENSIONS
++
+ # Initialize Automake
+ AM_INIT_AUTOMAKE([foreign dist-bzip2])
+ 
+-- 
+2.14.3
+
package/x11r7/xlib_libxshmfence/xlib_libxshmfence.mk +3 -0
index 3a2f32623a..9686c408c1 100644
@@ -12,4 +12,7 @@ XLIB_LIBXSHMFENCE_LICENSE_FILES = COPYING
 XLIB_LIBXSHMFENCE_INSTALL_STAGING = YES
 XLIB_LIBXSHMFENCE_DEPENDENCIES = host-pkgconf xproto_xproto
 
+# 0001-configure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch
+XLIB_LIBXSHMFENCE_AUTORECONF = YES
+
 $(eval $(autotools-package))
Signature · 6 lines
2.14.3

_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot
Signature · 6 lines
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'
#19 lore raw

PATCH 1/5 package/x11r7/xlib_libxshmfence: fix build with glibc 2.27

In reply to Romain Naour

Hello,

Quoted 9 lines · Romain Naour

With glibc 2.27 memfd_create() is inside a _GNU_SOURCE guard, so call AC_USE_SYSTEM_EXTENSIONS to get this defined.

---
 ...onfigure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch | 30 ++++++++++++++++++++++
 .../x11r7/xlib_libxshmfence/xlib_libxshmfence.mk   |  3 +++
 2 files changed, 33 insertions(+)
 create mode 100644 package/x11r7/xlib_libxshmfence/0001-configure.ac-call-AC_USE_SYSTEM_EXTENSIONS.patch

Thanks, I've applied the series to the next branch.

For PATCH 1, I added a reference to the upstream commit you've backported, as suggested by Yann.

For PATCH 5, I added the references to the CVEs being fixed, tweaked the commit log about license file hashes, and changed the title to "security bump".

Thanks!

Thomas

Signature · 3 lines
Thomas Petazzoni, CTO, Bootlin (formerly Free Electrons)
Embedded Linux and Kernel engineering
https://bootlin.com