Sync changes to SLFO-1.2 branch

- ppc64le-revert-power10-strcmp.patch: Revert optimized POWER10 strcmp,
  strncmp implementations (CVE-2025-5745, CVE-2025-5702, bsc#1244184,
  bsc#1244182, BZ #33060, BZ #33056)
- ppc64le-revert-power10-memcmp.patch: Revert optimized POWER10 memcmp
  implementation (BZ #33059) (forwarded request 1286664 from Andreas_Schwab)

OBS-URL: https://build.opensuse.org/request/show/1286665
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/glibc?expand=0&rev=299

gen-tempname-randomness.patch

@@ -0,0 +1,28 @@
+From 5f62cf88c4530c11904482775b7582bd7f6d80d2 Mon Sep 17 00:00:00 2001
+From: Andreas Schwab <schwab@suse.de>
+Date: Wed, 25 Sep 2024 11:49:30 +0200
+Subject: [PATCH] Fix missing randomness in __gen_tempname (bug 32214)
+
+Make sure to update the random value also if getrandom fails.
+
+Fixes: 686d542025 ("posix: Sync tempname with gnulib")
+---
+ sysdeps/posix/tempname.c | 2 ++
+ 1 file changed, 2 insertions(+)
+
+diff --git a/sysdeps/posix/tempname.c b/sysdeps/posix/tempname.c
+index c00fe0c181..fc30958a0c 100644
+--- a/sysdeps/posix/tempname.c
++++ b/sysdeps/posix/tempname.c
+@@ -117,6 +117,8 @@ random_bits (random_value *r, random_value s)
+      succeed.  */
+ #if !_LIBC
+   *r = mix_random_values (v, clock ());
++#else
++  *r = v;
+ #endif
+   return false;
+ }
+-- 
+2.46.2
+

glibc-2.3.90-langpackdir.diff

@@ -1,33 +0,0 @@
-Index: glibc-2.38/intl/loadmsgcat.c
-===================================================================
---- glibc-2.38.orig/intl/loadmsgcat.c
-+++ glibc-2.38/intl/loadmsgcat.c
-@@ -796,8 +796,26 @@ _nl_load_domain (struct loaded_l10nfile
-   if (domain_file->filename == NULL)
-     goto out;
- 
--  /* Try to open the addressed file.  */
--  fd = open (domain_file->filename, O_RDONLY | O_BINARY);
-+  /* Replace /locale/ with /usr/share/locale-langpack/ */
-+  const char *langpackdir = "/usr/share/locale-langpack/";
-+  char *filename_langpack = malloc (strlen (domain_file->filename)
-+				    + strlen (langpackdir));
-+  if (filename_langpack != NULL)
-+    {
-+      char *p = strstr (domain_file->filename, "/locale/");
-+      if (p != NULL)
-+	{
-+	  strcpy (__stpcpy (filename_langpack, langpackdir), p + 8);
-+	  fd = open (filename_langpack, O_RDONLY | O_BINARY);
-+	}
-+
-+      free (filename_langpack);
-+    }
-+
-+  if (fd == -1)
-+    /* Try to open the addressed file.  */
-+    fd = open (domain_file->filename, O_RDONLY | O_BINARY);
-+
-   if (fd == -1)
-     goto out;
- 

glibc.changes

@@ -1,3 +1,103 @@
+-------------------------------------------------------------------
+Thu Jul 24 10:14:52 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- regcomp-double-free.patch: posix: Fix double-free after allocation
+  failure in regcomp (CVE-2025-8058, bsc#1246965, BZ #33185)
+
+-------------------------------------------------------------------
+Tue Jun 17 09:11:26 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- ppc64le-revert-power10-strcmp.patch: Revert optimized POWER10 strcmp,
+  strncmp implementations (CVE-2025-5745, CVE-2025-5702, bsc#1244184,
+  bsc#1244182, BZ #33060, BZ #33056)
+- ppc64le-revert-power10-memcmp.patch: Revert optimized POWER10 memcmp
+  implementation (BZ #33059)
+
+-------------------------------------------------------------------
+Thu Jun  5 09:18:29 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- Add support for userspace livepatching for ppc64le (jsc#PED-11850)
+
+-------------------------------------------------------------------
+Mon Jun  2 12:09:30 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- Filter GLIBC_PRIVATE symbols again
+- Drop ngpt provides
+- Compile functions in libc_nonshared.a as PIC
+
+-------------------------------------------------------------------
+Thu Apr  3 14:43:21 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- pthread-wakeup.patch: pthreads NPTL: lost wakeup fix 2 (bsc#1234128, BZ
+  #25847)
+
+-------------------------------------------------------------------
+Mon Mar 24 15:24:22 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- Mark functions in libc_nonshared.a as hidden (bsc#1239883)
+
+-------------------------------------------------------------------
+Wed Mar 12 10:28:15 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- Do not build libnsl1 (bsc#1239459)
+
+-------------------------------------------------------------------
+Wed Feb 26 15:32:06 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- Bump minimal kernel version to 4.3 to enable use of direct socketcalls
+  on x86-32 and s390x (bsc#1234713)
+
+-------------------------------------------------------------------
+Mon Feb 17 10:37:02 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- Remove nis from nsswitch.conf (bsc#1237210)
+
+-------------------------------------------------------------------
+Mon Feb 10 12:19:00 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- Use rpm.execute when available (bsc#1236869)
+
+-------------------------------------------------------------------
+Mon Jan 20 13:33:58 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- Disable nscd support (bsc#1235247)
+
+-------------------------------------------------------------------
+Tue Jan  7 10:28:36 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- Correctly determine livepatching support
+
+-------------------------------------------------------------------
+Sun Jan  5 11:43:47 UTC 2025 - Andreas Schwab <schwab@suse.de>
+
+- Define _enable_debug_packages for rpm 4.20
+
+-------------------------------------------------------------------
+Mon Dec  2 14:02:08 UTC 2024 - Andreas Schwab <schwab@suse.de>
+
+- Add support for loongarch64
+
+-------------------------------------------------------------------
+Wed Oct  9 07:42:01 UTC 2024 - Andreas Schwab <schwab@suse.de>
+
+- Apply libc_nonshared.a workaround also on s390x and ppc64le (bsc#1231051)
+
+-------------------------------------------------------------------
+Mon Sep 30 20:56:18 UTC 2024 - Stanislav Brabec <sbrabec@suse.com>
+
+- langpacks are no more used. Drop glibc-2.3.90-langpackdir.diff.
+
+-------------------------------------------------------------------
+Thu Sep 26 11:45:57 UTC 2024 - Andreas Schwab <schwab@suse.de>
+
+- gen-tempname-randomness.patch: Fix missing randomness in __gen_tempname
+  (bsc#1230965, BZ #32214)
+
+-------------------------------------------------------------------
+Wed Sep 18 15:11:48 UTC 2024 - Andreas Schwab <schwab@suse.de>
+
+- Use nss-systemd by default also in SLE (bsc#1230638)
+
 -------------------------------------------------------------------
 Mon Jul 29 07:50:31 UTC 2024 - Andreas Schwab <schwab@suse.de>
 
@@ -33,6 +133,7 @@ Mon Jul 29 07:50:31 UTC 2024 - Andreas Schwab <schwab@suse.de>
   s390-clone-error-clobber-r7.patch, sigisemptyset.patch,
   stdbit-builtins.patch, utmp-time-bits.patch: Removed
 - glibc-2.3.90-langpackdir.diff: Rediff
+- bsc#1228041
 
 -------------------------------------------------------------------
 Mon Jun  3 08:24:38 UTC 2024 - Dominique Leuenberger <dimstar@opensuse.org>
@@ -47,6 +148,12 @@ Tue May 28 07:27:01 UTC 2024 - Andreas Schwab <schwab@suse.de>
 
 - Obsolete glibc-locale-base-<targettype> from glibc-<targettype>
 
+-------------------------------------------------------------------
+Thu May 23 11:31:34 UTC 2024 - Andreas Schwab <schwab@suse.de>
+
+- Add workaround for invalid use of libc_nonshared.a with non-SUSE libc
+  (bsc#1221482)
+
 -------------------------------------------------------------------
 Tue May 21 09:44:46 UTC 2024 - Andreas Schwab <schwab@suse.de>
 

glibc.spec

@@ -1,7 +1,7 @@
 #
 # spec file for package glibc
 #
-# Copyright (c) 2024 SUSE LLC
+# Copyright (c) 2025 SUSE LLC
 #
 # All modifications and additions to the file contributed by third parties
 # remain the property of their copyright owners, unless otherwise agreed
@@ -57,15 +57,14 @@
 %bcond_with usrmerged
 %endif
 
-%if 0%{?gcc_version} < 13
-%define with_gcc 13
+%if 0%{?suse_version} >= 1600
+%bcond_with nscd
+%else
+%bcond_without nscd
 %endif
 
-# Enable support for livepatching.
-%ifarch x86_64
-%bcond_without livepatching
-%else
-%bcond_with livepatching
+%if 0%{?gcc_version} < 13
+%define with_gcc 13
 %endif
 
 %bcond_with build_all
@@ -95,6 +94,7 @@ ExclusiveArch:  do_not_build
 %define build_testsuite 0
 %define build_cross 1
 %undefine _build_create_debug
+%define _enable_debug_packages 0
 ExcludeArch:    %{cross_arch}
 %if 0%{?suse_version} < 1600
 ExclusiveArch:  do_not_build
@@ -102,13 +102,39 @@ ExclusiveArch:  do_not_build
 %endif
 %define host_arch %{?cross_cpu}%{!?cross_cpu:%{_target_cpu}}
 
+# Enable support for livepatching.
+%define have_livepatching_support 0
+%if %{build_cross}
+%if "%{cross_arch}" == "x86_64" || "%{cross_arch}" == "ppc64le"
+%define have_livepatching_support 1
+%endif
+%else
+%ifarch x86_64 ppc64le
+%define have_livepatching_support 1
+%endif
+%endif
+
+%if %{have_livepatching_support}
+%bcond_without livepatching
+%else
+%bcond_with livepatching
+%endif
+
+%if %{suse_version} >= 1600
+%define build_libnsl1 0
+%else
+%define build_libnsl1 1
+%endif
+
 %if %{build_main}
 %define name_suffix %{nil}
 %else
 %define name_suffix -%{flavor}-src
 %endif
 
-%define __filter_GLIBC_PRIVATE 1
+# Nothing outside of glibc must use these symbols
+%define __requires_exclude GLIBC_PRIVATE
+%define __provides_exclude GLIBC_PRIVATE
 %ifarch i686
 # For i686 let's only build what's different from i586, so
 # no need to build documentation
@@ -133,20 +159,22 @@ ExclusiveArch:  do_not_build
 %define disable_assert 0
 %define enable_stackguard_randomization 1
 # glibc requires at least kernel 3.2
-%define enablekernel 3.2
+# Bump to 4.3 to enable use of direct socketcalls on x86-32 and s390x
+%define enablekernel 4.3
 # some architectures need a newer kernel
-%ifarch ppc64le
-%define enablekernel 3.10
-%endif
-%ifarch aarch64
-%define enablekernel 3.7
-%endif
-%ifarch ia64
-%define enablekernel 3.2.18
-%endif
 %ifarch riscv64
 %define enablekernel 4.15
 %endif
+%ifarch loongarch64
+%define enablekernel 5.19
+%endif
+
+# Before 2.29
+%define libnsl_archs %ix86 %alpha hppa m68k %mips32 %mips64 %sparc ppc ppc64 ppc64le x86_64 s390 s390x %arm aarch64 riscv64
+# Before 2.34
+%define libutil_archs %libnsl_archs
+# Before 2.35
+%define libanl_archs %libutil_archs
 
 Name:           glibc%{name_suffix}
 Summary:        Standard Shared Libraries (from the GNU C Library)
@@ -170,7 +198,6 @@ Source1:        https://ftp.gnu.org/pub/gnu/glibc/glibc-%{version}.tar.xz.sig
 Source2:        http://savannah.gnu.org/project/memberlist-gpgkeys.php?group=libc&download=1#/glibc.keyring
 Source4:        manpages.tar.bz2
 Source5:        nsswitch.conf
-Source6:        sle-nsswitch.conf
 Source7:        bindresvport.blacklist
 Source9:        glibc.rpmlintrc
 Source10:       baselibs.conf
@@ -180,11 +207,9 @@ Source21:       nscd.service
 Source22:       nscd.sysusers
 
 %if %{build_main}
-# ngpt was used in 8.1 and SLES8
-Obsoletes:      ngpt < 2.2.2
-Obsoletes:      ngpt-devel < 2.2.2
-Provides:       ngpt = 2.2.2
-Provides:       ngpt-devel = 2.2.2
+%if %{without nscd}
+Obsoletes:      nscd <= %{version}
+%endif
 Conflicts:      kernel < %{enablekernel}
 %if %{with usrmerged}
 # make sure we have post-usrmerge filesystem package
@@ -276,8 +301,6 @@ Patch10:        glibc-version.diff
 Patch13:        glibc-2.3.2.no_archive.diff
 # PATCH-FIX-OPENSUSE -- add blacklist for bindresvport
 Patch14:        glibc-bindresvport-blacklist.diff
-# PATCH-FIX-OPENSUSE prefer -lang rpm packages
-Patch15:        glibc-2.3.90-langpackdir.diff
 # PATCH-FEATURE-SLE Use nscd user for nscd
 Patch19:        nscd-server-user.patch
 # PATCH-FEATURE-SLE read nsswich.conf from /usr
@@ -302,6 +325,16 @@ Patch306:       glibc-fix-double-loopback.diff
 %if %{without snapshot}
 ###
 # Patches from upstream
+# PATCH-FIX-UPSTREAM Fix missing randomness in __gen_tempname (BZ #32214)
+Patch1000:      gen-tempname-randomness.patch
+# PATCH-FIX-UPSTREAM pthreads NPTL: lost wakeup fix 2 (BZ #25847)
+Patch1001:      pthread-wakeup.patch
+# PATCH-FIX-UPSTREAM Revert optimized POWER10 strcmp, strncmp implementations (CVE-2025-5745, CVE-2025-5702, BZ #33060, BZ #33056)
+Patch1006:      ppc64le-revert-power10-strcmp.patch
+# PATCH-FIX-UPSTREAM Revert optimized POWER10 memcmp (BZ #33059)
+Patch1007:      ppc64le-revert-power10-memcmp.patch
+# PATCH-FIX-UPSTREAM posix: Fix double-free after allocation failure in regcomp (BZ #33185)
+Patch1008:      regcomp-double-free.patch
 ###
 %endif
 
@@ -704,8 +737,12 @@ profile="--disable-profile"
 %if %{with livepatching}
 	--enable-userspace-livepatch \
 %endif
-	--disable-crypt || \
-  {
+	--disable-crypt \
+%if %{without nscd}
+        --disable-build-nscd \
+        --disable-nscd \
+%endif
+  || {
     rc=$?;
     echo "------- BEGIN config.log ------";
     %{__cat} config.log;
@@ -721,6 +758,46 @@ echo 'CFLAGS-.os += -fdump-ipa-clones' \
 %endif
 
 make %{?_smp_mflags} %{?make_output_sync}
+
+%if %{build_main} && !0%{?is_opensuse}
+%ifarch x86_64 i686 s390x ppc64le
+# Horrible workaround for bsc#1221482
+%ifarch x86_64 i686
+archsub=x86
+%endif
+%ifarch s390x
+archsub=s390
+%endif
+%ifarch ppc64le
+archsub=powerpc
+%endif
+xstatbuild ()
+{
+  gcc -O2 -fpic -I ../sysdeps/unix/sysv/linux/$archsub -xc - -c -o $1stat$2.oS <<EOF
+#include <bits/wordsize.h>
+#include <xstatver.h>
+int __$1xstat$2 (int, $3, void *);
+
+int
+__attribute__ ((visibility ("hidden")))
+$1stat$2 ($3 file, void *buf)
+{
+  return __$1xstat$2 (_STAT_VER, file, buf);
+}
+EOF
+  ar r libc_nonshared.a $1stat$2.oS
+}
+xstatbuild "" "" "const char *"
+xstatbuild f "" int
+xstatbuild l "" "const char *"
+%ifarch i686
+xstatbuild "" 64 "const char *"
+xstatbuild f 64 int
+xstatbuild l 64 "const char *"
+%endif
+%endif
+%endif
+
 cd ..
 
 #
@@ -730,8 +807,10 @@ cd ..
 make %{?_smp_mflags} %{?make_output_sync} -C cc-base html
 %endif
 
+%if %{with nscd}
 # sysusers.d
 %sysusers_generate_pre %{SOURCE22} nscd nscd.conf
+%endif
 
 %check
 %if %{build_testsuite}
@@ -813,6 +892,9 @@ make %{?_smp_mflags} %{?make_output_sync} -C cc-base test t=elf/check-localplt
 %define rtldlib lib
 %define rtld_name ld-linux-riscv64-lp64d.so.1
 %endif
+%ifarch loongarch64
+%define rtld_name ld-linux-loongarch-lp64d.so.1
+%endif
 
 %if %{with usrmerged}
 %define rootsbindir %{_sbindir}
@@ -920,7 +1002,7 @@ install -m 644 %{SOURCE7} %{buildroot}/etc
 %if %suse_version > 1500
 install -D -m 644 %{SOURCE5} %{buildroot}%{_prefix}/etc/nsswitch.conf
 %else
-install -m 644 %{SOURCE6} %{buildroot}/etc/nsswitch.conf
+install -m 644 %{SOURCE5} %{buildroot}/etc/nsswitch.conf
 %endif
 
 %if %{build_html}
@@ -930,6 +1012,7 @@ cp -p cc-base/manual/libc/*.html %{buildroot}%{_datadir}/doc/glibc
 
 cd manpages; make install_root=%{buildroot} install; cd ..
 
+%if %{with nscd}
 # nscd tools:
 
 %ifnarch i686
@@ -939,6 +1022,7 @@ ln -sf %{rootsbindir}/service %{buildroot}%{_sbindir}/rcnscd
 mkdir -p %{buildroot}/run/nscd
 mkdir -p %{buildroot}/var/lib/nscd
 %endif
+%endif
 
 #
 # Create ld.so.conf
@@ -967,6 +1051,11 @@ chmod 644 %{buildroot}%{_bindir}/ldd
 
 rm -f %{buildroot}%{rootsbindir}/sln
 
+%if !%build_libnsl1
+rm -f %{buildroot}%{slibdir}/libnsl.so.1
+%endif
+
+%if %{with nscd}
 %ifnarch i686
 mkdir -p %{buildroot}/usr/lib/tmpfiles.d/
 install -m 644 %{SOURCE20} %{buildroot}/usr/lib/tmpfiles.d/
@@ -975,6 +1064,7 @@ install -m 644 %{SOURCE21} %{buildroot}/usr/lib/systemd/system
 mkdir -p %{buildroot}/usr/lib/sysusers.d/
 install -m 644 %{SOURCE22} %{buildroot}/usr/lib/sysusers.d/nscd.conf
 %endif
+%endif
 
 %if 0%{?rtld_oldname:1}
 # Provide compatibility link
@@ -1001,8 +1091,10 @@ rm -rf %{buildroot}%{_libdir}/audit
 # Remove files from glibc-{extra,info,i18ndata}, nscd
 rm -rf %{buildroot}%{_infodir} %{buildroot}%{_prefix}/share/i18n
 rm -f %{buildroot}%{_bindir}/makedb %{buildroot}/var/lib/misc/Makefile
+%if %{with nscd}
 rm -f %{buildroot}%{_sbindir}/nscd
 %endif
+%endif
 
 %ifnarch i686
 # /var/lib/misc is incompatible with transactional updates (bsc#1138726)
@@ -1099,6 +1191,9 @@ rm %{buildroot}/sbin
 %if %{build_main}
 
 %post -p <lua>
+%if %{suse_version} >= 1600
+exec = rpm.execute
+%else
 function exec(path, ...)
   local pid = posix.fork()
   if pid == 0 then
@@ -1111,6 +1206,7 @@ function exec(path, ...)
   end
   posix.wait(pid)
 end
+%endif
 
 -- First, get rid of platform-optimized libraries. We remove any we have
 -- ever built, since otherwise we might end up using some old leftover
@@ -1220,7 +1316,9 @@ exit 0
 %endif
 
 %{slibdir}/libBrokenLocale.so.1
+%ifarch %libanl_archs
 %{slibdir}/libanl.so.1
+%endif
 %{slibdir}/libc.so.6*
 %{slibdir}/libc_malloc_debug.so.0
 %{slibdir}/libdl.so.2*
@@ -1237,7 +1335,9 @@ exit 0
 %{slibdir}/libresolv.so.2
 %{slibdir}/librt.so.1
 %{slibdir}/libthread_db.so.1
+%ifarch %libutil_archs
 %{slibdir}/libutil.so.1
+%endif
 %dir %attr(0700,root,root) /var/cache/ldconfig
 %{rootsbindir}/ldconfig
 %{_bindir}/gencat
@@ -1317,7 +1417,9 @@ exit 0
 %{_includedir}/*
 %{_libdir}/*.o
 %{_libdir}/libBrokenLocale.so
+%ifarch %libanl_archs
 %{_libdir}/libanl.so
+%endif
 %{_libdir}/libc.so
 %{_libdir}/libc_malloc_debug.so
 %{_libdir}/libm.so
@@ -1345,7 +1447,9 @@ exit 0
 %files devel-static
 %defattr(-,root,root)
 %{_libdir}/libBrokenLocale.a
+%ifarch %libanl_archs
 %{_libdir}/libanl.a
+%endif
 %{_libdir}/libc.a
 %{_libdir}/libm.a
 %ifarch x86_64 aarch64
@@ -1371,6 +1475,7 @@ exit 0
 %defattr(-,root,root)
 %{_prefix}/share/i18n
 
+%if %{with nscd}
 %files -n nscd
 %defattr(-,root,root)
 %config(noreplace) /etc/nscd.conf
@@ -1391,13 +1496,16 @@ exit 0
 %attr(0600,root,root) %verify(not md5 size mtime) %ghost %config(missingok,noreplace) /var/lib/nscd/services
 %attr(0600,root,root) %verify(not md5 size mtime) %ghost %config(missingok,noreplace) /var/lib/nscd/netgroup
 %endif
+%endif
 
 %if %{build_profile}
 %files profile
 %defattr(-,root,root)
 %{_libdir}/libc_p.a
 %{_libdir}/libBrokenLocale_p.a
+%ifarch %libanl_archs
 %{_libdir}/libanl_p.a
+%endif
 %{_libdir}/libm_p.a
 %ifarch x86_64 aarch64
 %{_libdir}/libmvec_p.a
@@ -1405,7 +1513,9 @@ exit 0
 %{_libdir}/libpthread_p.a
 %{_libdir}/libresolv_p.a
 %{_libdir}/librt_p.a
+%ifarch %libutil_archs
 %{_libdir}/libutil_p.a
+%endif
 %{_libdir}/libdl_p.a
 %endif
 
@@ -1419,10 +1529,12 @@ exit 0
 %files lang -f libc.lang
 %endif
 
-%ifarch %ix86 %alpha hppa m68k %mips32 %mips64 %sparc ppc ppc64 ppc64le x86_64 s390 s390x %arm aarch64 riscv64
+%if %build_libnsl1
+%ifarch %libnsl_archs
 %files -n libnsl1
 %{slibdir}/libnsl.so.1
 %endif
+%endif
 
 %endif
 

nsswitch.conf

@@ -64,12 +64,12 @@ networks:	files dns
 aliases:	files usrfiles
 ethers:		files usrfiles
 gshadow:	files usrfiles
-netgroup:	files nis
+netgroup:	files
 protocols:	files usrfiles
 publickey:	files
 rpc:		files usrfiles
 services:	files usrfiles
 
-automount:	files nis
+automount:	files
 bootparams:	files
 netmasks:	files

ppc64le-revert-power10-memcmp.patch

@@ -0,0 +1,417 @@
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/le/power10/memchr.S
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/le/power10/memchr.S
++++ /dev/null
+@@ -1,315 +0,0 @@
+-/* Optimized memchr implementation for POWER10 LE.
+-   Copyright (C) 2021-2024 Free Software Foundation, Inc.
+-   This file is part of the GNU C Library.
+-
+-   The GNU C Library is free software; you can redistribute it and/or
+-   modify it under the terms of the GNU Lesser General Public
+-   License as published by the Free Software Foundation; either
+-   version 2.1 of the License, or (at your option) any later version.
+-
+-   The GNU C Library is distributed in the hope that it will be useful,
+-   but WITHOUT ANY WARRANTY; without even the implied warranty of
+-   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU
+-   Lesser General Public License for more details.
+-
+-   You should have received a copy of the GNU Lesser General Public
+-   License along with the GNU C Library; if not, see
+-   <https://www.gnu.org/licenses/>.  */
+-
+-#include <sysdep.h>
+-
+-# ifndef MEMCHR
+-#  define MEMCHR __memchr
+-# endif
+-# define M_VREG_ZERO v20
+-# define M_OFF_START_LOOP 256
+-# define MEMCHR_SUBTRACT_VECTORS \
+-	vsububm   v4,v4,v18;	    \
+-	vsububm   v5,v5,v18;	    \
+-	vsububm   v6,v6,v18;	    \
+-	vsububm   v7,v7,v18;
+-# define M_TAIL(vreg,increment)	   \
+-	vctzlsbb  r4,vreg;	   \
+-	cmpld     r5,r4;	   \
+-	ble       L(null);	   \
+-	addi	  r4,r4,increment; \
+-	add	  r3,r6,r4;	   \
+-	blr
+-
+-/* TODO: Replace macros by the actual instructions when minimum binutils becomes
+-   >= 2.35.  This is used to keep compatibility with older versions.  */
+-#define M_VEXTRACTBM(rt,vrb)	 \
+-	.long(((4)<<(32-6))	 \
+-	      | ((rt)<<(32-11))	 \
+-	      | ((8)<<(32-16))	 \
+-	      | ((vrb)<<(32-21)) \
+-	      | 1602)
+-
+-#define M_LXVP(xtp,dq,ra)		   \
+-	.long(((6)<<(32-6))		   \
+-	      | ((((xtp)-32)>>1)<<(32-10)) \
+-	      | ((1)<<(32-11))		   \
+-	      | ((ra)<<(32-16))		   \
+-	      | dq)
+-
+-#define CHECK16B(vreg,offset,addr,label) \
+-	lxv	  vreg+32,offset(addr);	\
+-	vcmpequb. vreg,vreg,v18;	\
+-	bne	  cr6,L(label);		\
+-	cmpldi	  r5,16;		\
+-	ble	  L(null);		\
+-	addi	  r5,r5,-16;
+-
+-/* Load 4 quadwords, merge into one VR for speed and check for NULLs.  r6 has #
+-   of bytes already checked.  */
+-#define CHECK64B(offset,addr,label)	    \
+-	M_LXVP(v4+32,offset,addr);	    \
+-	M_LXVP(v6+32,offset+32,addr);	    \
+-	MEMCHR_SUBTRACT_VECTORS;	    \
+-	vminub	  v14,v4,v5;		    \
+-	vminub	  v15,v6,v7;		    \
+-	vminub	  v16,v14,v15;		    \
+-	vcmpequb. v0,v16,M_VREG_ZERO;	    \
+-	beq	  cr6,$+12;		    \
+-	li	  r7,offset;		    \
+-	b     	  L(label);          	    \
+-	cmpldi	  r5,64;		    \
+-	ble	  L(null);		    \
+-	addi	  r5,r5,-64
+-
+-/* Implements the function
+-   void *[r3] memchr (const void *s [r3], int c [r4], size_t n [r5]).  */
+-
+-	.machine power9
+-
+-ENTRY_TOCLESS (MEMCHR)
+-	CALL_MCOUNT 3
+-
+-	cmpldi	r5,0
+-	beq	L(null)
+-	mr	r0,r5
+-	xori	r6,r4,0xff
+-
+-	mtvsrd	v18+32,r4	/* matching char in v18  */
+-	mtvsrd	v19+32,r6	/* non matching char in v19  */
+-
+-	vspltb	v18,v18,7	/* replicate  */
+-	vspltb	v19,v19,7	/* replicate  */
+-	vspltisb  M_VREG_ZERO,0
+-
+-	/* Next 16B-aligned address. Prepare address for L(aligned).  */
+-	addi	  r6,r3,16
+-	clrrdi	  r6,r6,4
+-
+-	/* Align data and fill bytes not loaded with non matching char.	 */
+-	lvx	  v0,0,r3
+-	lvsr	  v1,0,r3
+-	vperm	  v0,v19,v0,v1
+-
+-	vcmpequb. v6,v0,v18
+-	bne	  cr6,L(found)
+-	sub	  r4,r6,r3
+-	cmpld	  r5,r4
+-	ble	  L(null)
+-	sub	  r5,r5,r4
+-
+-	/* Test up to OFF_START_LOOP-16 bytes in 16B chunks.  The main loop is
+-	   optimized for longer strings, so checking the first bytes in 16B
+-	   chunks benefits a lot small strings.  */
+-	.p2align 5
+-L(aligned):
+-	cmpldi	r5,0
+-	beq     L(null)
+-
+-	CHECK16B(v0,0,r6,tail1)
+-	CHECK16B(v1,16,r6,tail2)
+-	CHECK16B(v2,32,r6,tail3)
+-	CHECK16B(v3,48,r6,tail4)
+-	CHECK16B(v4,64,r6,tail5)
+-	CHECK16B(v5,80,r6,tail6)
+-	CHECK16B(v6,96,r6,tail7)
+-	CHECK16B(v7,112,r6,tail8)
+-	CHECK16B(v8,128,r6,tail9)
+-	CHECK16B(v9,144,r6,tail10)
+-	CHECK16B(v10,160,r6,tail11)
+-	CHECK16B(v0,176,r6,tail12)
+-	CHECK16B(v1,192,r6,tail13)
+-	CHECK16B(v2,208,r6,tail14)
+-	CHECK16B(v3,224,r6,tail15)
+-
+-	cmpdi	cr5,r4,0	/* Check if c == 0.  This will be useful to
+-				   choose how we will perform the main loop.  */
+-
+-	/* Prepare address for the loop.  */
+-	addi	  r4,r3,M_OFF_START_LOOP
+-	clrrdi	  r4,r4,6
+-	sub	  r6,r4,r3
+-	sub	  r5,r0,r6
+-	addi	  r6,r4,128
+-
+-	/* If c == 0, use the loop without the vsububm.  */
+-	beq	cr5,L(loop)
+-
+-	/* This is very similar to the block after L(loop), the difference is
+-	   that here MEMCHR_SUBTRACT_VECTORS is not empty, and we subtract
+-	   each byte loaded by the char we are looking for, this way we can keep
+-	   using vminub to merge the results and checking for nulls.  */
+-	.p2align 5
+-L(memchr_loop):
+-	CHECK64B(0,r4,pre_tail_64b)
+-	CHECK64B(64,r4,pre_tail_64b)
+-	addi	r4,r4,256
+-
+-	CHECK64B(0,r6,tail_64b)
+-	CHECK64B(64,r6,tail_64b)
+-	addi	r6,r6,256
+-
+-	CHECK64B(0,r4,pre_tail_64b)
+-	CHECK64B(64,r4,pre_tail_64b)
+-	addi	r4,r4,256
+-
+-	CHECK64B(0,r6,tail_64b)
+-	CHECK64B(64,r6,tail_64b)
+-	addi	r6,r6,256
+-
+-	b	L(memchr_loop)
+-	/* Switch to a more aggressive approach checking 64B each time.  Use 2
+-	   pointers 128B apart and unroll the loop once to make the pointer
+-	   updates and usages separated enough to avoid stalls waiting for
+-	   address calculation.  */
+-	.p2align 5
+-L(loop):
+-#undef MEMCHR_SUBTRACT_VECTORS
+-#define MEMCHR_SUBTRACT_VECTORS /* nothing */
+-	CHECK64B(0,r4,pre_tail_64b)
+-	CHECK64B(64,r4,pre_tail_64b)
+-	addi	  r4,r4,256
+-
+-	CHECK64B(0,r6,tail_64b)
+-	CHECK64B(64,r6,tail_64b)
+-	addi	  r6,r6,256
+-
+-	CHECK64B(0,r4,pre_tail_64b)
+-	CHECK64B(64,r4,pre_tail_64b)
+-	addi      r4,r4,256
+-
+-	CHECK64B(0,r6,tail_64b)
+-	CHECK64B(64,r6,tail_64b)
+-	addi      r6,r6,256
+-
+-	b	  L(loop)
+-
+-	.p2align  5
+-L(pre_tail_64b):
+-	mr	r6,r4
+-L(tail_64b):
+-	/* OK, we found a null byte.  Let's look for it in the current 64-byte
+-	   block and mark it in its corresponding VR.  lxvp vx,0(ry) puts the
+-	   low 16B bytes into vx+1, and the high into vx, so the order here is
+-	   v5, v4, v7, v6.  */
+-	vcmpequb  v1,v5,M_VREG_ZERO
+-	vcmpequb  v2,v4,M_VREG_ZERO
+-	vcmpequb  v3,v7,M_VREG_ZERO
+-	vcmpequb  v4,v6,M_VREG_ZERO
+-
+-	/* Take into account the other 64B blocks we had already checked.  */
+-	add	r6,r6,r7
+-	/* Extract first bit of each byte.  */
+-	M_VEXTRACTBM(r8,v1)
+-	M_VEXTRACTBM(r9,v2)
+-	M_VEXTRACTBM(r10,v3)
+-	M_VEXTRACTBM(r11,v4)
+-
+-	/* Shift each value into their corresponding position.  */
+-	sldi	  r9,r9,16
+-	sldi	  r10,r10,32
+-	sldi	  r11,r11,48
+-
+-	/* Merge the results.  */
+-	or	  r8,r8,r9
+-	or	  r9,r10,r11
+-	or	  r11,r9,r8
+-
+-	cnttzd	  r0,r11	  /* Count trailing zeros before the match.  */
+-	cmpld     r5,r0
+-	ble	  L(null)
+-	add	  r3,r6,r0	  /* Compute final address.  */
+-	blr
+-
+-	.p2align  5
+-L(tail1):
+-	M_TAIL(v0,0)
+-
+-	.p2align  5
+-L(tail2):
+-	M_TAIL(v1,16)
+-
+-	.p2align  5
+-L(tail3):
+-	M_TAIL(v2,32)
+-
+-	.p2align  5
+-L(tail4):
+-	M_TAIL(v3,48)
+-
+-	.p2align  5
+-L(tail5):
+-	M_TAIL(v4,64)
+-
+-	.p2align  5
+-L(tail6):
+-	M_TAIL(v5,80)
+-
+-	.p2align  5
+-L(tail7):
+-	M_TAIL(v6,96)
+-
+-	.p2align  5
+-L(tail8):
+-	M_TAIL(v7,112)
+-
+-	.p2align  5
+-L(tail9):
+-	M_TAIL(v8,128)
+-
+-	.p2align  5
+-L(tail10):
+-	M_TAIL(v9,144)
+-
+-	.p2align  5
+-L(tail11):
+-	M_TAIL(v10,160)
+-
+-	.p2align  5
+-L(tail12):
+-	M_TAIL(v0,176)
+-
+-	.p2align  5
+-L(tail13):
+-	M_TAIL(v1,192)
+-
+-	.p2align  5
+-L(tail14):
+-	M_TAIL(v2,208)
+-
+-	.p2align  5
+-L(tail15):
+-	M_TAIL(v3,224)
+-
+-	.p2align  5
+-L(found):
+-	vctzlsbb  r7,v6
+-	cmpld     r5,r7
+-	ble       L(null)
+-	add       r3,r3,r7
+-	blr
+-
+-	.p2align  5
+-L(null):
+-	li	r3,0
+-	blr
+-
+-END (MEMCHR)
+-
+-weak_alias (__memchr, memchr)
+-libc_hidden_builtin_def (memchr)
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/Makefile
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/multiarch/Makefile
++++ glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/Makefile
+@@ -31,7 +31,7 @@ sysdep_routines += memcpy-power8-cached
+ 		   strncase-power8
+ 
+ ifneq (,$(filter %le,$(config-machine)))
+-sysdep_routines += memchr-power10 memcmp-power10 memcpy-power10 \
++sysdep_routines += memcmp-power10 memcpy-power10 \
+ 		   memmove-power10 memset-power10 rawmemchr-power9 \
+ 		   rawmemchr-power10 strcmp-power9 \
+ 		   strncmp-power9 strcpy-power9 stpcpy-power9 \
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/ifunc-impl-list.c
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/multiarch/ifunc-impl-list.c
++++ glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/ifunc-impl-list.c
+@@ -226,12 +226,6 @@ __libc_ifunc_impl_list (const char *name
+ 
+   /* Support sysdeps/powerpc/powerpc64/multiarch/memchr.c.  */
+   IFUNC_IMPL (i, name, memchr,
+-#ifdef __LITTLE_ENDIAN__
+-	      IFUNC_IMPL_ADD (array, i, memchr,
+-		              hwcap2 & PPC_FEATURE2_ARCH_3_1
+-			      && hwcap & PPC_FEATURE_HAS_VSX,
+-			      __memchr_power10)
+-#endif
+ 	      IFUNC_IMPL_ADD (array, i, memchr,
+ 			      hwcap2 & PPC_FEATURE2_ARCH_2_07
+ 			      && hwcap & PPC_FEATURE_HAS_ALTIVEC,
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/memchr-power10.S
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/multiarch/memchr-power10.S
++++ /dev/null
+@@ -1,28 +0,0 @@
+-/* Optimized memchr implementation for POWER10/PPC64.
+-   Copyright (C) 2016-2024 Free Software Foundation, Inc.
+-   This file is part of the GNU C Library.
+-
+-   The GNU C Library is free software; you can redistribute it and/or
+-   modify it under the terms of the GNU Lesser General Public
+-   License as published by the Free Software Foundation; either
+-   version 2.1 of the License, or (at your option) any later version.
+-
+-   The GNU C Library is distributed in the hope that it will be useful,
+-   but WITHOUT ANY WARRANTY; without even the implied warranty of
+-   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+-   Lesser General Public License for more details.
+-
+-   You should have received a copy of the GNU Lesser General Public
+-   License along with the GNU C Library; if not, see
+-   <https://www.gnu.org/licenses/>.  */
+-
+-#if defined __LITTLE_ENDIAN__ && IS_IN (libc)
+-#define MEMCHR __memchr_power10
+-
+-#undef libc_hidden_builtin_def
+-#define libc_hidden_builtin_def(name)
+-#undef weak_alias
+-#define weak_alias(name,alias)
+-
+-#include <sysdeps/powerpc/powerpc64/le/power10/memchr.S>
+-#endif
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/memchr.c
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/multiarch/memchr.c
++++ glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/memchr.c
+@@ -25,23 +25,15 @@ extern __typeof (__memchr) __memchr_ppc
+ extern __typeof (__memchr) __memchr_power7 attribute_hidden;
+ extern __typeof (__memchr) __memchr_power8 attribute_hidden;
+ 
+-# ifdef __LITTLE_ENDIAN__
+-extern __typeof (__memchr) __memchr_power10 attribute_hidden;
+-# endif
+ /* Avoid DWARF definition DIE on ifunc symbol so that GDB can handle
+    ifunc symbol properly.  */
+ libc_ifunc (__memchr,
+-# ifdef __LITTLE_ENDIAN__
+-	    (hwcap2 & PPC_FEATURE2_ARCH_3_1
+-	     && hwcap & PPC_FEATURE_HAS_VSX)
+-	    ? __memchr_power10 :
+-# endif
+-	      (hwcap2 & PPC_FEATURE2_ARCH_2_07
+-	      && hwcap & PPC_FEATURE_HAS_ALTIVEC)
+-	      ? __memchr_power8 :
+-	        (hwcap & PPC_FEATURE_ARCH_2_06)
+-	        ? __memchr_power7
+-	        : __memchr_ppc);
++	    (hwcap2 & PPC_FEATURE2_ARCH_2_07
++	     && hwcap & PPC_FEATURE_HAS_ALTIVEC)
++	    ? __memchr_power8 :
++	    (hwcap & PPC_FEATURE_ARCH_2_06)
++            ? __memchr_power7
++            : __memchr_ppc);
+ 
+ weak_alias (__memchr, memchr)
+ libc_hidden_builtin_def (memchr)

ppc64le-revert-power10-strcmp.patch

@@ -0,0 +1,658 @@
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/le/power10/strcmp.S
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/le/power10/strcmp.S
++++ /dev/null
+@@ -1,233 +0,0 @@
+-/* Optimized strcmp implementation for PowerPC64/POWER10.
+-   Copyright (C) 2021-2024 Free Software Foundation, Inc.
+-   This file is part of the GNU C Library.
+-
+-   The GNU C Library is free software; you can redistribute it and/or
+-   modify it under the terms of the GNU Lesser General Public
+-   License as published by the Free Software Foundation; either
+-   version 2.1 of the License, or (at your option) any later version.
+-
+-   The GNU C Library is distributed in the hope that it will be useful,
+-   but WITHOUT ANY WARRANTY; without even the implied warranty of
+-   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+-   Lesser General Public License for more details.
+-
+-   You should have received a copy of the GNU Lesser General Public
+-   License along with the GNU C Library; if not, see
+-   <https://www.gnu.org/licenses/>.  */
+-#include <sysdep.h>
+-
+-#ifndef STRCMP
+-# define STRCMP strcmp
+-#endif
+-
+-/* Implements the function
+-   int [r3] strcmp (const char *s1 [r3], const char *s2 [r4]).  */
+-
+-/* TODO: Change this to actual instructions when minimum binutils is upgraded
+-   to 2.27.  Macros are defined below for these newer instructions in order
+-   to maintain compatibility.  */
+-
+-#define LXVP(xtp,dq,ra)		     \
+-	.long(((6)<<(32-6))	     \
+-	| ((((xtp)-32)>>1)<<(32-10)) \
+-	| ((1)<<(32-11))	     \
+-	| ((ra)<<(32-16))	     \
+-	| dq)
+-
+-#define COMPARE_16(vreg1,vreg2,offset)  \
+-	lxv       vreg1+32,offset(r3);  \
+-	lxv       vreg2+32,offset(r4);	\
+-	vcmpnezb. v7,vreg1,vreg2;	\
+-	bne       cr6,L(different);     \
+-
+-#define COMPARE_32(vreg1,vreg2,offset,label1,label2) \
+-	LXVP(vreg1+32,offset,r3);                    \
+-	LXVP(vreg2+32,offset,r4);                    \
+-	vcmpnezb. v7,vreg1+1,vreg2+1;                \
+-	bne	  cr6,L(label1);                     \
+-	vcmpnezb. v7,vreg1,vreg2;                    \
+-	bne	  cr6,L(label2);                     \
+-
+-#define TAIL(vreg1,vreg2)     \
+-	vctzlsbb r6,v7;	      \
+-	vextubrx r5,r6,vreg1; \
+-	vextubrx r4,r6,vreg2; \
+-	subf	 r3,r4,r5;    \
+-	blr;                  \
+-
+-#define CHECK_N_BYTES(reg1,reg2,len_reg) \
+-	sldi	  r0,len_reg,56;         \
+-	lxvl	  32+v4,reg1,r0;         \
+-	lxvl	  32+v5,reg2,r0;         \
+-	add	  reg1,reg1,len_reg;     \
+-	add	  reg2,reg2,len_reg;     \
+-	vcmpnezb  v7,v4,v5;              \
+-	vctzlsbb  r6,v7;                 \
+-	cmpld	  cr7,r6,len_reg;        \
+-	blt	  cr7,L(different);      \
+-
+-	/* TODO: change this to .machine power10 when the minimum required
+-	binutils allows it.  */
+-
+-	.machine  power9
+-ENTRY_TOCLESS (STRCMP, 4)
+-	andi.	r7,r3,4095
+-	andi.	r8,r4,4095
+-	cmpldi	cr0,r7,4096-16
+-	cmpldi	cr1,r8,4096-16
+-	bgt	cr0,L(crosses)
+-	bgt	cr1,L(crosses)
+-	COMPARE_16(v4,v5,0)
+-
+-L(crosses):
+-	andi.	r7,r3,15
+-	subfic	r7,r7,16	/* r7(nalign1) = 16 - (str1 & 15).  */
+-	andi.	r9,r4,15
+-	subfic	r5,r9,16	/* r5(nalign2) = 16 - (str2 & 15).  */
+-	cmpld	cr7,r7,r5
+-	beq	cr7,L(same_aligned)
+-	blt	cr7,L(nalign1_min)
+-
+-	/* nalign2 is minimum and s2 pointer is aligned.  */
+-	CHECK_N_BYTES(r3,r4,r5)
+-	/* Are we on the 64B hunk which crosses a page?  */
+-	andi.	r10,r3,63	/* Determine offset into 64B hunk.  */
+-	andi.	r8,r3,15        /* The offset into the 16B hunk.  */
+-	neg	r7,r3
+-	andi.	r9,r7,15	/* Number of bytes after a 16B cross.  */
+-	rlwinm.	r7,r7,26,0x3F	/* ((r3-4096))>>6&63.  */
+-	beq	L(compare_64_pagecross)
+-	mtctr	r7
+-	b	L(compare_64B_unaligned)
+-
+-	/* nalign1 is minimum and s1 pointer is aligned.  */
+-L(nalign1_min):
+-	CHECK_N_BYTES(r3,r4,r7)
+-	/* Are we on the 64B hunk which crosses a page?  */
+-	andi.	r10,r4,63	/* Determine offset into 64B hunk.  */
+-	andi.	r8,r4,15	/* The offset into the 16B hunk.  */
+-	neg	r7,r4
+-	andi.	r9,r7,15	/* Number of bytes after a 16B cross.  */
+-	rlwinm. r7,r7,26,0x3F	/* ((r4-4096))>>6&63.  */
+-	beq	L(compare_64_pagecross)
+-	mtctr	r7
+-
+-	.p2align 5
+-L(compare_64B_unaligned):
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	COMPARE_16(v4,v5,32)
+-	COMPARE_16(v4,v5,48)
+-	addi	r3,r3,64
+-	addi	r4,r4,64
+-	bdnz	L(compare_64B_unaligned)
+-
+-	/* Cross the page boundary of s2, carefully. Only for first
+-	iteration we have to get the count of 64B blocks to be checked.
+-	From second iteration and beyond, loop counter is always 63.  */
+-L(compare_64_pagecross):
+-	li	r11, 63
+-	mtctr	r11
+-	cmpldi	r10,16
+-	ble	L(cross_4)
+-	cmpldi	r10,32
+-	ble	L(cross_3)
+-	cmpldi	r10,48
+-	ble	L(cross_2)
+-L(cross_1):
+-	CHECK_N_BYTES(r3,r4,r9)
+-	CHECK_N_BYTES(r3,r4,r8)
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	COMPARE_16(v4,v5,32)
+-	addi	r3,r3,48
+-	addi	r4,r4,48
+-	b	L(compare_64B_unaligned)
+-L(cross_2):
+-	COMPARE_16(v4,v5,0)
+-	addi	r3,r3,16
+-	addi	r4,r4,16
+-	CHECK_N_BYTES(r3,r4,r9)
+-	CHECK_N_BYTES(r3,r4,r8)
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	addi	r3,r3,32
+-	addi	r4,r4,32
+-	b	L(compare_64B_unaligned)
+-L(cross_3):
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	addi	r3,r3,32
+-	addi	r4,r4,32
+-	CHECK_N_BYTES(r3,r4,r9)
+-	CHECK_N_BYTES(r3,r4,r8)
+-	COMPARE_16(v4,v5,0)
+-	addi	r3,r3,16
+-	addi	r4,r4,16
+-	b	L(compare_64B_unaligned)
+-L(cross_4):
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	COMPARE_16(v4,v5,32)
+-	addi	r3,r3,48
+-	addi	r4,r4,48
+-	CHECK_N_BYTES(r3,r4,r9)
+-	CHECK_N_BYTES(r3,r4,r8)
+-	b	L(compare_64B_unaligned)
+-
+-L(same_aligned):
+-	CHECK_N_BYTES(r3,r4,r7)
+-        /* Align s1 to 32B and adjust s2 address.
+-	   Use lxvp only if both s1 and s2 are 32B aligned.  */
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	COMPARE_16(v4,v5,32)
+-	COMPARE_16(v4,v5,48)
+-	addi	r3,r3,64
+-	addi	r4,r4,64
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-
+-	clrldi	r6,r3,59
+-	subfic	r5,r6,32
+-	add	r3,r3,r5
+-	add	r4,r4,r5
+-	andi.	r5,r4,0x1F
+-	beq	cr0,L(32B_aligned_loop)
+-
+-	.p2align 5
+-L(16B_aligned_loop):
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	COMPARE_16(v4,v5,32)
+-	COMPARE_16(v4,v5,48)
+-	addi	r3,r3,64
+-	addi	r4,r4,64
+-	b	L(16B_aligned_loop)
+-
+-	/* Calculate and return the difference.  */
+-L(different):
+-	TAIL(v4,v5)
+-
+-	.p2align 5
+-L(32B_aligned_loop):
+-	COMPARE_32(v14,v16,0,tail1,tail2)
+-	COMPARE_32(v18,v20,32,tail3,tail4)
+-	COMPARE_32(v22,v24,64,tail5,tail6)
+-	COMPARE_32(v26,v28,96,tail7,tail8)
+-	addi	r3,r3,128
+-	addi	r4,r4,128
+-	b	L(32B_aligned_loop)
+-
+-L(tail1): TAIL(v15,v17)
+-L(tail2): TAIL(v14,v16)
+-L(tail3): TAIL(v19,v21)
+-L(tail4): TAIL(v18,v20)
+-L(tail5): TAIL(v23,v25)
+-L(tail6): TAIL(v22,v24)
+-L(tail7): TAIL(v27,v29)
+-L(tail8): TAIL(v26,v28)
+-
+-END (STRCMP)
+-libc_hidden_builtin_def (strcmp)
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/le/power10/strncmp.S
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/le/power10/strncmp.S
++++ /dev/null
+@@ -1,271 +0,0 @@
+-/* Optimized strncmp implementation for PowerPC64/POWER10.
+-   Copyright (C) 2024 Free Software Foundation, Inc.
+-   This file is part of the GNU C Library.
+-
+-   The GNU C Library is free software; you can redistribute it and/or
+-   modify it under the terms of the GNU Lesser General Public
+-   License as published by the Free Software Foundation; either
+-   version 2.1 of the License, or (at your option) any later version.
+-
+-   The GNU C Library is distributed in the hope that it will be useful,
+-   but WITHOUT ANY WARRANTY; without even the implied warranty of
+-   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+-   Lesser General Public License for more details.
+-
+-   You should have received a copy of the GNU Lesser General Public
+-   License along with the GNU C Library; if not, see
+-   <https://www.gnu.org/licenses/>.  */
+-
+-#include <sysdep.h>
+-
+-/* Implements the function
+-
+-   int [r3] strncmp (const char *s1 [r3], const char *s2 [r4], size_t [r5] n)
+-
+-   The implementation uses unaligned doubleword access to avoid specialized
+-   code paths depending of data alignment for first 32 bytes and uses
+-   vectorised loops after that.  */
+-
+-#ifndef STRNCMP
+-# define STRNCMP strncmp
+-#endif
+-
+-/* TODO: Change this to actual instructions when minimum binutils is upgraded
+-   to 2.27.  Macros are defined below for these newer instructions in order
+-   to maintain compatibility.  */
+-
+-#define LXVP(xtp,dq,ra)              \
+-	.long(((6)<<(32-6))          \
+-	| ((((xtp)-32)>>1)<<(32-10)) \
+-	| ((1)<<(32-11))             \
+-	| ((ra)<<(32-16))            \
+-	| dq)
+-
+-#define COMPARE_16(vreg1,vreg2,offset) \
+-	lxv	  vreg1+32,offset(r3); \
+-	lxv	  vreg2+32,offset(r4); \
+-	vcmpnezb. v7,vreg1,vreg2;      \
+-	bne	  cr6,L(different);    \
+-	cmpldi	  cr7,r5,16;           \
+-	ble	  cr7,L(ret0);         \
+-	addi	  r5,r5,-16;
+-
+-#define COMPARE_32(vreg1,vreg2,offset,label1,label2) \
+-	LXVP(vreg1+32,offset,r3);                    \
+-	LXVP(vreg2+32,offset,r4);                    \
+-	vcmpnezb. v7,vreg1+1,vreg2+1;                \
+-	bne	  cr6,L(label1);                     \
+-	vcmpnezb. v7,vreg1,vreg2;                    \
+-	bne	  cr6,L(label2);                     \
+-	cmpldi	  cr7,r5,32;                         \
+-	ble	  cr7,L(ret0);                       \
+-	addi	  r5,r5,-32;
+-
+-#define TAIL_FIRST_16B(vreg1,vreg2) \
+-	vctzlsbb r6,v7;             \
+-	cmpld	 cr7,r5,r6;         \
+-	ble	 cr7,L(ret0);       \
+-	vextubrx r5,r6,vreg1;       \
+-	vextubrx r4,r6,vreg2;       \
+-	subf	 r3,r4,r5;          \
+-	blr;
+-
+-#define TAIL_SECOND_16B(vreg1,vreg2) \
+-	vctzlsbb r6,v7;              \
+-	addi	 r0,r6,16;           \
+-	cmpld	 cr7,r5,r0;          \
+-	ble	 cr7,L(ret0);        \
+-	vextubrx r5,r6,vreg1;        \
+-	vextubrx r4,r6,vreg2;        \
+-	subf	 r3,r4,r5;           \
+-	blr;
+-
+-#define CHECK_N_BYTES(reg1,reg2,len_reg) \
+-	sldi	  r6,len_reg,56;	 \
+-	lxvl	  32+v4,reg1,r6;	 \
+-	lxvl	  32+v5,reg2,r6;	 \
+-	add	  reg1,reg1,len_reg;	 \
+-	add	  reg2,reg2,len_reg;	 \
+-	vcmpnezb  v7,v4,v5;		 \
+-	vctzlsbb  r6,v7;		 \
+-	cmpld	  cr7,r6,len_reg;	 \
+-	blt	  cr7,L(different);	 \
+-	cmpld	  cr7,r5,len_reg;	 \
+-	ble	  cr7,L(ret0);		 \
+-	sub	  r5,r5,len_reg;	 \
+-
+-	/* TODO: change this to .machine power10 when the minimum required
+-	 binutils allows it.  */
+-	.machine  power9
+-ENTRY_TOCLESS (STRNCMP, 4)
+-	/* Check if size is 0.  */
+-	cmpdi	 cr0,r5,0
+-	beq	 cr0,L(ret0)
+-	andi.   r7,r3,4095
+-	andi.   r8,r4,4095
+-	cmpldi  cr0,r7,4096-16
+-	cmpldi  cr1,r8,4096-16
+-	bgt     cr0,L(crosses)
+-	bgt     cr1,L(crosses)
+-	COMPARE_16(v4,v5,0)
+-	addi	r3,r3,16
+-	addi	r4,r4,16
+-
+-L(crosses):
+-	andi.	 r7,r3,15
+-	subfic	 r7,r7,16	/* r7(nalign1) = 16 - (str1 & 15).  */
+-	andi.	 r9,r4,15
+-	subfic	 r8,r9,16	/* r8(nalign2) = 16 - (str2 & 15).  */
+-	cmpld	 cr7,r7,r8
+-	beq	 cr7,L(same_aligned)
+-	blt	 cr7,L(nalign1_min)
+-
+-	/* nalign2 is minimum and s2 pointer is aligned.  */
+-	CHECK_N_BYTES(r3,r4,r8)
+-	/* Are we on the 64B hunk which crosses a page?  */
+-	andi.   r10,r3,63       /* Determine offset into 64B hunk.  */
+-	andi.   r8,r3,15        /* The offset into the 16B hunk.  */
+-	neg     r7,r3
+-	andi.   r9,r7,15        /* Number of bytes after a 16B cross.  */
+-	rlwinm. r7,r7,26,0x3F   /* ((r4-4096))>>6&63.  */
+-	beq     L(compare_64_pagecross)
+-	mtctr   r7
+-	b       L(compare_64B_unaligned)
+-
+-	/* nalign1 is minimum and s1 pointer is aligned.  */
+-L(nalign1_min):
+-	CHECK_N_BYTES(r3,r4,r7)
+-	/* Are we on the 64B hunk which crosses a page?  */
+-	andi.   r10,r4,63       /* Determine offset into 64B hunk.  */
+-	andi.   r8,r4,15        /* The offset into the 16B hunk.  */
+-	neg     r7,r4
+-	andi.   r9,r7,15        /* Number of bytes after a 16B cross.  */
+-	rlwinm. r7,r7,26,0x3F   /* ((r4-4096))>>6&63.  */
+-	beq     L(compare_64_pagecross)
+-	mtctr   r7
+-
+-	.p2align 5
+-L(compare_64B_unaligned):
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	COMPARE_16(v4,v5,32)
+-	COMPARE_16(v4,v5,48)
+-	addi    r3,r3,64
+-	addi    r4,r4,64
+-	bdnz    L(compare_64B_unaligned)
+-
+-	/* Cross the page boundary of s2, carefully. Only for first
+-	iteration we have to get the count of 64B blocks to be checked.
+-	From second iteration and beyond, loop counter is always 63.  */
+-L(compare_64_pagecross):
+-	li      r11, 63
+-	mtctr   r11
+-	cmpldi  r10,16
+-	ble     L(cross_4)
+-	cmpldi  r10,32
+-	ble     L(cross_3)
+-	cmpldi  r10,48
+-	ble     L(cross_2)
+-L(cross_1):
+-	CHECK_N_BYTES(r3,r4,r9)
+-	CHECK_N_BYTES(r3,r4,r8)
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	COMPARE_16(v4,v5,32)
+-	addi    r3,r3,48
+-	addi    r4,r4,48
+-	b       L(compare_64B_unaligned)
+-L(cross_2):
+-	COMPARE_16(v4,v5,0)
+-	addi    r3,r3,16
+-	addi    r4,r4,16
+-	CHECK_N_BYTES(r3,r4,r9)
+-	CHECK_N_BYTES(r3,r4,r8)
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	addi    r3,r3,32
+-	addi    r4,r4,32
+-	b       L(compare_64B_unaligned)
+-L(cross_3):
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	addi    r3,r3,32
+-	addi    r4,r4,32
+-	CHECK_N_BYTES(r3,r4,r9)
+-	CHECK_N_BYTES(r3,r4,r8)
+-	COMPARE_16(v4,v5,0)
+-	addi    r3,r3,16
+-	addi    r4,r4,16
+-	b       L(compare_64B_unaligned)
+-L(cross_4):
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	COMPARE_16(v4,v5,32)
+-	addi    r3,r3,48
+-	addi    r4,r4,48
+-	CHECK_N_BYTES(r3,r4,r9)
+-	CHECK_N_BYTES(r3,r4,r8)
+-	b       L(compare_64B_unaligned)
+-
+-L(same_aligned):
+-	CHECK_N_BYTES(r3,r4,r7)
+-	/* Align s1 to 32B and adjust s2 address.
+-	   Use lxvp only if both s1 and s2 are 32B aligned.  */
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	COMPARE_16(v4,v5,32)
+-	COMPARE_16(v4,v5,48)
+-	addi	r3,r3,64
+-	addi	r4,r4,64
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	addi	r5,r5,32
+-
+-	clrldi  r6,r3,59
+-	subfic	r7,r6,32
+-	add	r3,r3,r7
+-	add	r4,r4,r7
+-	subf	r5,r7,r5
+-	andi.	r7,r4,0x1F
+-	beq	cr0,L(32B_aligned_loop)
+-
+-	.p2align 5
+-L(16B_aligned_loop):
+-	COMPARE_16(v4,v5,0)
+-	COMPARE_16(v4,v5,16)
+-	COMPARE_16(v4,v5,32)
+-	COMPARE_16(v4,v5,48)
+-	addi	r3,r3,64
+-	addi	r4,r4,64
+-	b	L(16B_aligned_loop)
+-
+-	/* Calculate and return the difference.  */
+-L(different):
+-	TAIL_FIRST_16B(v4,v5)
+-
+-	.p2align 5
+-L(32B_aligned_loop):
+-	COMPARE_32(v14,v16,0,tail1,tail2)
+-	COMPARE_32(v18,v20,32,tail3,tail4)
+-	COMPARE_32(v22,v24,64,tail5,tail6)
+-	COMPARE_32(v26,v28,96,tail7,tail8)
+-	addi	r3,r3,128
+-	addi	r4,r4,128
+-	b	L(32B_aligned_loop)
+-
+-L(tail1): TAIL_FIRST_16B(v15,v17)
+-L(tail2): TAIL_SECOND_16B(v14,v16)
+-L(tail3): TAIL_FIRST_16B(v19,v21)
+-L(tail4): TAIL_SECOND_16B(v18,v20)
+-L(tail5): TAIL_FIRST_16B(v23,v25)
+-L(tail6): TAIL_SECOND_16B(v22,v24)
+-L(tail7): TAIL_FIRST_16B(v27,v29)
+-L(tail8): TAIL_SECOND_16B(v26,v28)
+-
+-	.p2align 5
+-L(ret0):
+-	li	r3,0
+-	blr
+-
+-END(STRNCMP)
+-libc_hidden_builtin_def(strncmp)
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/Makefile
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/multiarch/Makefile
++++ glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/Makefile
+@@ -33,8 +33,8 @@ sysdep_routines += memcpy-power8-cached
+ ifneq (,$(filter %le,$(config-machine)))
+ sysdep_routines += memchr-power10 memcmp-power10 memcpy-power10 \
+ 		   memmove-power10 memset-power10 rawmemchr-power9 \
+-		   rawmemchr-power10 strcmp-power9 strcmp-power10 \
+-		   strncmp-power9 strncmp-power10 strcpy-power9 stpcpy-power9 \
++		   rawmemchr-power10 strcmp-power9 \
++		   strncmp-power9 strcpy-power9 stpcpy-power9 \
+ 		   strlen-power9 strncpy-power9 stpncpy-power9 strlen-power10
+ endif
+ CFLAGS-strncase-power7.c += -mcpu=power7 -funroll-loops
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/ifunc-impl-list.c
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/multiarch/ifunc-impl-list.c
++++ glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/ifunc-impl-list.c
+@@ -164,9 +164,6 @@ __libc_ifunc_impl_list (const char *name
+   /* Support sysdeps/powerpc/powerpc64/multiarch/strncmp.c.  */
+   IFUNC_IMPL (i, name, strncmp,
+ #ifdef __LITTLE_ENDIAN__
+-	      IFUNC_IMPL_ADD (array, i, strncmp, hwcap2 & PPC_FEATURE2_ARCH_3_1
+-			      && hwcap & PPC_FEATURE_HAS_VSX,
+-			      __strncmp_power10)
+ 	      IFUNC_IMPL_ADD (array, i, strncmp, hwcap2 & PPC_FEATURE2_ARCH_3_00
+ 			      && hwcap & PPC_FEATURE_HAS_ALTIVEC,
+ 			      __strncmp_power9)
+@@ -387,10 +384,6 @@ __libc_ifunc_impl_list (const char *name
+   IFUNC_IMPL (i, name, strcmp,
+ #ifdef __LITTLE_ENDIAN__
+ 	      IFUNC_IMPL_ADD (array, i, strcmp,
+-			      (hwcap2 & PPC_FEATURE2_ARCH_3_1)
+-			      && (hwcap & PPC_FEATURE_HAS_VSX),
+-			      __strcmp_power10)
+-	      IFUNC_IMPL_ADD (array, i, strcmp,
+ 			      hwcap2 & PPC_FEATURE2_ARCH_3_00
+ 			      && hwcap & PPC_FEATURE_HAS_ALTIVEC,
+ 			      __strcmp_power9)
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/strcmp-power10.S
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/multiarch/strcmp-power10.S
++++ /dev/null
+@@ -1,26 +0,0 @@
+-/* Optimized strcmp implementation for POWER10/PPC64.
+-   Copyright (C) 2021-2024 Free Software Foundation, Inc.
+-   This file is part of the GNU C Library.
+-
+-   The GNU C Library is free software; you can redistribute it and/or
+-   modify it under the terms of the GNU Lesser General Public
+-   License as published by the Free Software Foundation; either
+-   version 2.1 of the License, or (at your option) any later version.
+-
+-   The GNU C Library is distributed in the hope that it will be useful,
+-   but WITHOUT ANY WARRANTY; without even the implied warranty of
+-   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+-   Lesser General Public License for more details.
+-
+-   You should have received a copy of the GNU Lesser General Public
+-   License along with the GNU C Library; if not, see
+-   <https://www.gnu.org/licenses/>.  */
+-
+-#if defined __LITTLE_ENDIAN__ && IS_IN (libc)
+-#define STRCMP __strcmp_power10
+-
+-#undef libc_hidden_builtin_def
+-#define libc_hidden_builtin_def(name)
+-
+-#include <sysdeps/powerpc/powerpc64/le/power10/strcmp.S>
+-#endif /* __LITTLE_ENDIAN__ && IS_IN (libc) */
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/strcmp.c
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/multiarch/strcmp.c
++++ glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/strcmp.c
+@@ -29,16 +29,12 @@ extern __typeof (strcmp) __strcmp_power7
+ extern __typeof (strcmp) __strcmp_power8 attribute_hidden;
+ # ifdef __LITTLE_ENDIAN__
+ extern __typeof (strcmp) __strcmp_power9 attribute_hidden;
+-extern __typeof (strcmp) __strcmp_power10 attribute_hidden;
+ # endif
+ 
+ # undef strcmp
+ 
+ libc_ifunc_redirected (__redirect_strcmp, strcmp,
+ # ifdef __LITTLE_ENDIAN__
+-		        (hwcap2 & PPC_FEATURE2_ARCH_3_1
+-			 && hwcap & PPC_FEATURE_HAS_VSX)
+-			? __strcmp_power10 :
+ 			(hwcap2 & PPC_FEATURE2_ARCH_3_00
+ 			 && hwcap & PPC_FEATURE_HAS_ALTIVEC)
+ 			? __strcmp_power9 :
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/strncmp-power10.S
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/multiarch/strncmp-power10.S
++++ /dev/null
+@@ -1,25 +0,0 @@
+-/* Copyright (C) 2024 Free Software Foundation, Inc.
+-   This file is part of the GNU C Library.
+-
+-   The GNU C Library is free software; you can redistribute it and/or
+-   modify it under the terms of the GNU Lesser General Public
+-   License as published by the Free Software Foundation; either
+-   version 2.1 of the License, or (at your option) any later version.
+-
+-   The GNU C Library is distributed in the hope that it will be useful,
+-   but WITHOUT ANY WARRANTY; without even the implied warranty of
+-   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+-   Lesser General Public License for more details.
+-
+-   You should have received a copy of the GNU Lesser General Public
+-   License along with the GNU C Library; if not, see
+-   <https://www.gnu.org/licenses/>.  */
+-
+-#if defined __LITTLE_ENDIAN__ && IS_IN (libc)
+-#define STRNCMP __strncmp_power10
+-
+-#undef libc_hidden_builtin_def
+-#define libc_hidden_builtin_def(name)
+-
+-#include <sysdeps/powerpc/powerpc64/le/power10/strncmp.S>
+-#endif
+Index: glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/strncmp.c
+===================================================================
+--- glibc-2.40.orig/sysdeps/powerpc/powerpc64/multiarch/strncmp.c
++++ glibc-2.40/sysdeps/powerpc/powerpc64/multiarch/strncmp.c
+@@ -29,7 +29,6 @@ extern __typeof (strncmp) __strncmp_ppc
+ extern __typeof (strncmp) __strncmp_power8 attribute_hidden;
+ # ifdef __LITTLE_ENDIAN__
+ extern __typeof (strncmp) __strncmp_power9 attribute_hidden;
+-extern __typeof (strncmp) __strncmp_power10 attribute_hidden;
+ # endif
+ # undef strncmp
+ 
+@@ -37,9 +36,6 @@ extern __typeof (strncmp) __strncmp_powe
+    ifunc symbol properly.  */
+ libc_ifunc_redirected (__redirect_strncmp, strncmp,
+ # ifdef __LITTLE_ENDIAN__
+-			(hwcap2 & PPC_FEATURE2_ARCH_3_1
+-			 && hwcap & PPC_FEATURE_HAS_VSX)
+-			? __strncmp_power10 :
+ 			(hwcap2 & PPC_FEATURE2_ARCH_3_00
+ 			 && hwcap & PPC_FEATURE_HAS_ALTIVEC)
+ 			? __strncmp_power9 :

pthread-wakeup.patch

@@ -0,0 +1,650 @@
+diff --git c/nptl/pthread_cond_broadcast.c w/nptl/pthread_cond_broadcast.c
+index 7409958394..f5793e715f 100644
+--- c/nptl/pthread_cond_broadcast.c
++++ w/nptl/pthread_cond_broadcast.c
+@@ -57,10 +57,10 @@ ___pthread_cond_broadcast (pthread_cond_t *cond)
+     {
+       /* Add as many signals as the remaining size of the group.  */
+       atomic_fetch_add_relaxed (cond->__data.__g_signals + g1,
+-				cond->__data.__g_size[g1] << 1);
++				cond->__data.__g_size[g1]);
+       cond->__data.__g_size[g1] = 0;
+ 
+-      /* We need to wake G1 waiters before we quiesce G1 below.  */
++      /* We need to wake G1 waiters before we switch G1 below.  */
+       /* TODO Only set it if there are indeed futex waiters.  We could
+ 	 also try to move this out of the critical section in cases when
+ 	 G2 is empty (and we don't need to quiesce).  */
+@@ -69,11 +69,11 @@ ___pthread_cond_broadcast (pthread_cond_t *cond)
+ 
+   /* G1 is complete.  Step (2) is next unless there are no waiters in G2, in
+      which case we can stop.  */
+-  if (__condvar_quiesce_and_switch_g1 (cond, wseq, &g1, private))
++  if (__condvar_switch_g1 (cond, wseq, &g1, private))
+     {
+       /* Step (3): Send signals to all waiters in the old G2 / new G1.  */
+       atomic_fetch_add_relaxed (cond->__data.__g_signals + g1,
+-				cond->__data.__g_size[g1] << 1);
++				cond->__data.__g_size[g1]);
+       cond->__data.__g_size[g1] = 0;
+       /* TODO Only set it if there are indeed futex waiters.  */
+       do_futex_wake = true;
+diff --git c/nptl/pthread_cond_common.c w/nptl/pthread_cond_common.c
+index 7440bb18e6..99640968f1 100644
+--- c/nptl/pthread_cond_common.c
++++ w/nptl/pthread_cond_common.c
+@@ -189,19 +189,17 @@ __condvar_get_private (int flags)
+     return FUTEX_SHARED;
+ }
+ 
+-/* This closes G1 (whose index is in G1INDEX), waits for all futex waiters to
+-   leave G1, converts G1 into a fresh G2, and then switches group roles so that
+-   the former G2 becomes the new G1 ending at the current __wseq value when we
+-   eventually make the switch (WSEQ is just an observation of __wseq by the
+-   signaler).
++/* This closes G1 (whose index is in G1INDEX), converts G1 into a fresh G2,
++   and then switches group roles so that the former G2 becomes the new G1
++   ending at the current __wseq value when we eventually make the switch
++   (WSEQ is just an observation of __wseq by the signaler).
+    If G2 is empty, it will not switch groups because then it would create an
+    empty G1 which would require switching groups again on the next signal.
+    Returns false iff groups were not switched because G2 was empty.  */
+ static bool __attribute__ ((unused))
+-__condvar_quiesce_and_switch_g1 (pthread_cond_t *cond, uint64_t wseq,
++__condvar_switch_g1 (pthread_cond_t *cond, uint64_t wseq,
+     unsigned int *g1index, int private)
+ {
+-  const unsigned int maxspin = 0;
+   unsigned int g1 = *g1index;
+ 
+   /* If there is no waiter in G2, we don't do anything.  The expression may
+@@ -210,96 +208,23 @@ __condvar_quiesce_and_switch_g1 (pthread_cond_t *cond, uint64_t wseq,
+      behavior.
+      Note that this works correctly for a zero-initialized condvar too.  */
+   unsigned int old_orig_size = __condvar_get_orig_size (cond);
+-  uint64_t old_g1_start = __condvar_load_g1_start_relaxed (cond) >> 1;
+-  if (((unsigned) (wseq - old_g1_start - old_orig_size)
+-	  + cond->__data.__g_size[g1 ^ 1]) == 0)
++  uint64_t old_g1_start = __condvar_load_g1_start_relaxed (cond);
++  uint64_t new_g1_start = old_g1_start + old_orig_size;
++  if (((unsigned) (wseq - new_g1_start) + cond->__data.__g_size[g1 ^ 1]) == 0)
+ 	return false;
+ 
+-  /* Now try to close and quiesce G1.  We have to consider the following kinds
+-     of waiters:
++  /* We have to consider the following kinds of waiters:
+      * Waiters from less recent groups than G1 are not affected because
+        nothing will change for them apart from __g1_start getting larger.
+      * New waiters arriving concurrently with the group switching will all go
+        into G2 until we atomically make the switch.  Waiters existing in G2
+        are not affected.
+-     * Waiters in G1 will be closed out immediately by setting a flag in
+-       __g_signals, which will prevent waiters from blocking using a futex on
+-       __g_signals and also notifies them that the group is closed.  As a
+-       result, they will eventually remove their group reference, allowing us
+-       to close switch group roles.  */
++     * Waiters in G1 have already received a signal and been woken.  */
+ 
+-  /* First, set the closed flag on __g_signals.  This tells waiters that are
+-     about to wait that they shouldn't do that anymore.  This basically
+-     serves as an advance notification of the upcoming change to __g1_start;
+-     waiters interpret it as if __g1_start was larger than their waiter
+-     sequence position.  This allows us to change __g1_start after waiting
+-     for all existing waiters with group references to leave, which in turn
+-     makes recovery after stealing a signal simpler because it then can be
+-     skipped if __g1_start indicates that the group is closed (otherwise,
+-     we would have to recover always because waiters don't know how big their
+-     groups are).  Relaxed MO is fine.  */
+-  atomic_fetch_or_relaxed (cond->__data.__g_signals + g1, 1);
+-
+-  /* Wait until there are no group references anymore.  The fetch-or operation
+-     injects us into the modification order of __g_refs; release MO ensures
+-     that waiters incrementing __g_refs after our fetch-or see the previous
+-     changes to __g_signals and to __g1_start that had to happen before we can
+-     switch this G1 and alias with an older group (we have two groups, so
+-     aliasing requires switching group roles twice).  Note that nobody else
+-     can have set the wake-request flag, so we do not have to act upon it.
+-
+-     Also note that it is harmless if older waiters or waiters from this G1
+-     get a group reference after we have quiesced the group because it will
+-     remain closed for them either because of the closed flag in __g_signals
+-     or the later update to __g1_start.  New waiters will never arrive here
+-     but instead continue to go into the still current G2.  */
+-  unsigned r = atomic_fetch_or_release (cond->__data.__g_refs + g1, 0);
+-  while ((r >> 1) > 0)
+-    {
+-      for (unsigned int spin = maxspin; ((r >> 1) > 0) && (spin > 0); spin--)
+-	{
+-	  /* TODO Back off.  */
+-	  r = atomic_load_relaxed (cond->__data.__g_refs + g1);
+-	}
+-      if ((r >> 1) > 0)
+-	{
+-	  /* There is still a waiter after spinning.  Set the wake-request
+-	     flag and block.  Relaxed MO is fine because this is just about
+-	     this futex word.
+-
+-	     Update r to include the set wake-request flag so that the upcoming
+-	     futex_wait only blocks if the flag is still set (otherwise, we'd
+-	     violate the basic client-side futex protocol).  */
+-	  r = atomic_fetch_or_relaxed (cond->__data.__g_refs + g1, 1) | 1;
+-
+-	  if ((r >> 1) > 0)
+-	    futex_wait_simple (cond->__data.__g_refs + g1, r, private);
+-	  /* Reload here so we eventually see the most recent value even if we
+-	     do not spin.   */
+-	  r = atomic_load_relaxed (cond->__data.__g_refs + g1);
+-	}
+-    }
+-  /* Acquire MO so that we synchronize with the release operation that waiters
+-     use to decrement __g_refs and thus happen after the waiters we waited
+-     for.  */
+-  atomic_thread_fence_acquire ();
+-
+-  /* Update __g1_start, which finishes closing this group.  The value we add
+-     will never be negative because old_orig_size can only be zero when we
+-     switch groups the first time after a condvar was initialized, in which
+-     case G1 will be at index 1 and we will add a value of 1.  See above for
+-     why this takes place after waiting for quiescence of the group.
+-     Relaxed MO is fine because the change comes with no additional
+-     constraints that others would have to observe.  */
+-  __condvar_add_g1_start_relaxed (cond,
+-      (old_orig_size << 1) + (g1 == 1 ? 1 : - 1));
+-
+-  /* Now reopen the group, thus enabling waiters to again block using the
+-     futex controlled by __g_signals.  Release MO so that observers that see
+-     no signals (and thus can block) also see the write __g1_start and thus
+-     that this is now a new group (see __pthread_cond_wait_common for the
+-     matching acquire MO loads).  */
+-  atomic_store_release (cond->__data.__g_signals + g1, 0);
++  /* Update __g1_start, which closes this group.  Relaxed MO is fine because
++     the change comes with no additional constraints that others would have
++     to observe.  */
++  __condvar_add_g1_start_relaxed (cond, old_orig_size);
+ 
+   /* At this point, the old G1 is now a valid new G2 (but not in use yet).
+      No old waiter can neither grab a signal nor acquire a reference without
+@@ -311,9 +236,13 @@ __condvar_quiesce_and_switch_g1 (pthread_cond_t *cond, uint64_t wseq,
+   g1 ^= 1;
+   *g1index ^= 1;
+ 
++  /* Now advance the new G1 g_signals to the new g1_start, giving it
++     an effective signal count of 0 to start.  */
++  atomic_store_release (cond->__data.__g_signals + g1, (unsigned)new_g1_start);
++
+   /* These values are just observed by signalers, and thus protected by the
+      lock.  */
+-  unsigned int orig_size = wseq - (old_g1_start + old_orig_size);
++  unsigned int orig_size = wseq - new_g1_start;
+   __condvar_set_orig_size (cond, orig_size);
+   /* Use and addition to not loose track of cancellations in what was
+      previously G2.  */
+diff --git c/nptl/pthread_cond_signal.c w/nptl/pthread_cond_signal.c
+index 9ac4bd9f75..bdc1e82466 100644
+--- c/nptl/pthread_cond_signal.c
++++ w/nptl/pthread_cond_signal.c
+@@ -69,19 +69,18 @@ ___pthread_cond_signal (pthread_cond_t *cond)
+   bool do_futex_wake = false;
+ 
+   /* If G1 is still receiving signals, we put the signal there.  If not, we
+-     check if G2 has waiters, and if so, quiesce and switch G1 to the former
+-     G2; if this results in a new G1 with waiters (G2 might have cancellations
+-     already, see __condvar_quiesce_and_switch_g1), we put the signal in the
+-     new G1.  */
++     check if G2 has waiters, and if so, switch G1 to the former G2; if this
++     results in a new G1 with waiters (G2 might have cancellations already,
++     see __condvar_switch_g1), we put the signal in the new G1. */
+   if ((cond->__data.__g_size[g1] != 0)
+-      || __condvar_quiesce_and_switch_g1 (cond, wseq, &g1, private))
++      || __condvar_switch_g1 (cond, wseq, &g1, private))
+     {
+       /* Add a signal.  Relaxed MO is fine because signaling does not need to
+-	 establish a happens-before relation (see above).  We do not mask the
+-	 release-MO store when initializing a group in
+-	 __condvar_quiesce_and_switch_g1 because we use an atomic
+-	 read-modify-write and thus extend that store's release sequence.  */
+-      atomic_fetch_add_relaxed (cond->__data.__g_signals + g1, 2);
++         establish a happens-before relation (see above).  We do not mask the
++         release-MO store when initializing a group in __condvar_switch_g1
++         because we use an atomic read-modify-write and thus extend that
++         store's release sequence.  */
++      atomic_fetch_add_relaxed (cond->__data.__g_signals + g1, 1);
+       cond->__data.__g_size[g1]--;
+       /* TODO Only set it if there are indeed futex waiters.  */
+       do_futex_wake = true;
+diff --git c/nptl/pthread_cond_wait.c w/nptl/pthread_cond_wait.c
+index 806c432d13..d919b3622a 100644
+--- c/nptl/pthread_cond_wait.c
++++ w/nptl/pthread_cond_wait.c
+@@ -84,7 +84,7 @@ __condvar_cancel_waiting (pthread_cond_t *cond, uint64_t seq, unsigned int g,
+      not hold a reference on the group.  */
+   __condvar_acquire_lock (cond, private);
+ 
+-  uint64_t g1_start = __condvar_load_g1_start_relaxed (cond) >> 1;
++  uint64_t g1_start = __condvar_load_g1_start_relaxed (cond);
+   if (g1_start > seq)
+     {
+       /* Our group is closed, so someone provided enough signals for it.
+@@ -143,23 +143,6 @@ __condvar_cancel_waiting (pthread_cond_t *cond, uint64_t seq, unsigned int g,
+     }
+ }
+ 
+-/* Wake up any signalers that might be waiting.  */
+-static void
+-__condvar_dec_grefs (pthread_cond_t *cond, unsigned int g, int private)
+-{
+-  /* Release MO to synchronize-with the acquire load in
+-     __condvar_quiesce_and_switch_g1.  */
+-  if (atomic_fetch_add_release (cond->__data.__g_refs + g, -2) == 3)
+-    {
+-      /* Clear the wake-up request flag before waking up.  We do not need more
+-	 than relaxed MO and it doesn't matter if we apply this for an aliased
+-	 group because we wake all futex waiters right after clearing the
+-	 flag.  */
+-      atomic_fetch_and_relaxed (cond->__data.__g_refs + g, ~(unsigned int) 1);
+-      futex_wake (cond->__data.__g_refs + g, INT_MAX, private);
+-    }
+-}
+-
+ /* Clean-up for cancellation of waiters waiting for normal signals.  We cancel
+    our registration as a waiter, confirm we have woken up, and re-acquire the
+    mutex.  */
+@@ -171,8 +154,6 @@ __condvar_cleanup_waiting (void *arg)
+   pthread_cond_t *cond = cbuffer->cond;
+   unsigned g = cbuffer->wseq & 1;
+ 
+-  __condvar_dec_grefs (cond, g, cbuffer->private);
+-
+   __condvar_cancel_waiting (cond, cbuffer->wseq >> 1, g, cbuffer->private);
+   /* FIXME With the current cancellation implementation, it is possible that
+      a thread is cancelled after it has returned from a syscall.  This could
+@@ -238,9 +219,7 @@ __condvar_cleanup_waiting (void *arg)
+    signaled), and a reference count.
+ 
+    The group reference count is used to maintain the number of waiters that
+-   are using the group's futex.  Before a group can change its role, the
+-   reference count must show that no waiters are using the futex anymore; this
+-   prevents ABA issues on the futex word.
++   are using the group's futex.
+ 
+    To represent which intervals in the waiter sequence the groups cover (and
+    thus also which group slot contains G1 or G2), we use a 64b counter to
+@@ -251,7 +230,7 @@ __condvar_cleanup_waiting (void *arg)
+    figure out whether they are in a group that has already been completely
+    signaled (i.e., if the current G1 starts at a later position that the
+    waiter's position).  Waiters cannot determine whether they are currently
+-   in G2 or G1 -- but they do not have too because all they are interested in
++   in G2 or G1 -- but they do not have to because all they are interested in
+    is whether there are available signals, and they always start in G2 (whose
+    group slot they know because of the bit in the waiter sequence.  Signalers
+    will simply fill the right group until it is completely signaled and can
+@@ -280,7 +259,6 @@ __condvar_cleanup_waiting (void *arg)
+      * Waiters fetch-add while having acquire the mutex associated with the
+        condvar.  Signalers load it and fetch-xor it concurrently.
+    __g1_start: Starting position of G1 (inclusive)
+-     * LSB is index of current G2.
+      * Modified by signalers while having acquired the condvar-internal lock
+        and observed concurrently by waiters.
+    __g1_orig_size: Initial size of G1
+@@ -300,11 +278,10 @@ __condvar_cleanup_waiting (void *arg)
+        last reference.
+      * Reference count used by waiters concurrently with signalers that have
+        acquired the condvar-internal lock.
+-   __g_signals: The number of signals that can still be consumed.
++   __g_signals: The number of signals that can still be consumed, relative to
++     the current g1_start.  (i.e. g1_start with the signal count added)
+      * Used as a futex word by waiters.  Used concurrently by waiters and
+        signalers.
+-     * LSB is true iff this group has been completely signaled (i.e., it is
+-       closed).
+    __g_size: Waiters remaining in this group (i.e., which have not been
+      signaled yet.
+      * Accessed by signalers and waiters that cancel waiting (both do so only
+@@ -328,27 +305,6 @@ __condvar_cleanup_waiting (void *arg)
+    sufficient because if a waiter can see a sufficiently large value, it could
+    have also consume a signal in the waiters group.
+ 
+-   Waiters try to grab a signal from __g_signals without holding a reference
+-   count, which can lead to stealing a signal from a more recent group after
+-   their own group was already closed.  They cannot always detect whether they
+-   in fact did because they do not know when they stole, but they can
+-   conservatively add a signal back to the group they stole from; if they
+-   did so unnecessarily, all that happens is a spurious wake-up.  To make this
+-   even less likely, __g1_start contains the index of the current g2 too,
+-   which allows waiters to check if there aliasing on the group slots; if
+-   there wasn't, they didn't steal from the current G1, which means that the
+-   G1 they stole from must have been already closed and they do not need to
+-   fix anything.
+-
+-   It is essential that the last field in pthread_cond_t is __g_signals[1]:
+-   The previous condvar used a pointer-sized field in pthread_cond_t, so a
+-   PTHREAD_COND_INITIALIZER from that condvar implementation might only
+-   initialize 4 bytes to zero instead of the 8 bytes we need (i.e., 44 bytes
+-   in total instead of the 48 we need).  __g_signals[1] is not accessed before
+-   the first group switch (G2 starts at index 0), which will set its value to
+-   zero after a harmless fetch-or whose return value is ignored.  This
+-   effectively completes initialization.
+-
+ 
+    Limitations:
+    * This condvar isn't designed to allow for more than
+@@ -379,7 +335,6 @@ static __always_inline int
+ __pthread_cond_wait_common (pthread_cond_t *cond, pthread_mutex_t *mutex,
+     clockid_t clockid, const struct __timespec64 *abstime)
+ {
+-  const int maxspin = 0;
+   int err;
+   int result = 0;
+ 
+@@ -396,8 +351,7 @@ __pthread_cond_wait_common (pthread_cond_t *cond, pthread_mutex_t *mutex,
+      because we do not need to establish any happens-before relation with
+      signalers (see __pthread_cond_signal); modification order alone
+      establishes a total order of waiters/signals.  We do need acquire MO
+-     to synchronize with group reinitialization in
+-     __condvar_quiesce_and_switch_g1.  */
++     to synchronize with group reinitialization in __condvar_switch_g1.  */
+   uint64_t wseq = __condvar_fetch_add_wseq_acquire (cond, 2);
+   /* Find our group's index.  We always go into what was G2 when we acquired
+      our position.  */
+@@ -424,178 +378,64 @@ __pthread_cond_wait_common (pthread_cond_t *cond, pthread_mutex_t *mutex,
+       return err;
+     }
+ 
+-  /* Now wait until a signal is available in our group or it is closed.
+-     Acquire MO so that if we observe a value of zero written after group
+-     switching in __condvar_quiesce_and_switch_g1, we synchronize with that
+-     store and will see the prior update of __g1_start done while switching
+-     groups too.  */
+-  unsigned int signals = atomic_load_acquire (cond->__data.__g_signals + g);
+ 
+-  do
++  while (1)
+     {
+-      while (1)
+-	{
+-	  /* Spin-wait first.
+-	     Note that spinning first without checking whether a timeout
+-	     passed might lead to what looks like a spurious wake-up even
+-	     though we should return ETIMEDOUT (e.g., if the caller provides
+-	     an absolute timeout that is clearly in the past).  However,
+-	     (1) spurious wake-ups are allowed, (2) it seems unlikely that a
+-	     user will (ab)use pthread_cond_wait as a check for whether a
+-	     point in time is in the past, and (3) spinning first without
+-	     having to compare against the current time seems to be the right
+-	     choice from a performance perspective for most use cases.  */
+-	  unsigned int spin = maxspin;
+-	  while (signals == 0 && spin > 0)
+-	    {
+-	      /* Check that we are not spinning on a group that's already
+-		 closed.  */
+-	      if (seq < (__condvar_load_g1_start_relaxed (cond) >> 1))
+-		goto done;
+-
+-	      /* TODO Back off.  */
+-
+-	      /* Reload signals.  See above for MO.  */
+-	      signals = atomic_load_acquire (cond->__data.__g_signals + g);
+-	      spin--;
+-	    }
+-
+-	  /* If our group will be closed as indicated by the flag on signals,
+-	     don't bother grabbing a signal.  */
+-	  if (signals & 1)
+-	    goto done;
+-
+-	  /* If there is an available signal, don't block.  */
+-	  if (signals != 0)
+-	    break;
+-
+-	  /* No signals available after spinning, so prepare to block.
+-	     We first acquire a group reference and use acquire MO for that so
+-	     that we synchronize with the dummy read-modify-write in
+-	     __condvar_quiesce_and_switch_g1 if we read from that.  In turn,
+-	     in this case this will make us see the closed flag on __g_signals
+-	     that designates a concurrent attempt to reuse the group's slot.
+-	     We use acquire MO for the __g_signals check to make the
+-	     __g1_start check work (see spinning above).
+-	     Note that the group reference acquisition will not mask the
+-	     release MO when decrementing the reference count because we use
+-	     an atomic read-modify-write operation and thus extend the release
+-	     sequence.  */
+-	  atomic_fetch_add_acquire (cond->__data.__g_refs + g, 2);
+-	  if (((atomic_load_acquire (cond->__data.__g_signals + g) & 1) != 0)
+-	      || (seq < (__condvar_load_g1_start_relaxed (cond) >> 1)))
+-	    {
+-	      /* Our group is closed.  Wake up any signalers that might be
+-		 waiting.  */
+-	      __condvar_dec_grefs (cond, g, private);
+-	      goto done;
+-	    }
+-
+-	  // Now block.
+-	  struct _pthread_cleanup_buffer buffer;
+-	  struct _condvar_cleanup_buffer cbuffer;
+-	  cbuffer.wseq = wseq;
+-	  cbuffer.cond = cond;
+-	  cbuffer.mutex = mutex;
+-	  cbuffer.private = private;
+-	  __pthread_cleanup_push (&buffer, __condvar_cleanup_waiting, &cbuffer);
+-
+-	  err = __futex_abstimed_wait_cancelable64 (
+-	    cond->__data.__g_signals + g, 0, clockid, abstime, private);
+-
+-	  __pthread_cleanup_pop (&buffer, 0);
+-
+-	  if (__glibc_unlikely (err == ETIMEDOUT || err == EOVERFLOW))
+-	    {
+-	      __condvar_dec_grefs (cond, g, private);
+-	      /* If we timed out, we effectively cancel waiting.  Note that
+-		 we have decremented __g_refs before cancellation, so that a
+-		 deadlock between waiting for quiescence of our group in
+-		 __condvar_quiesce_and_switch_g1 and us trying to acquire
+-		 the lock during cancellation is not possible.  */
+-	      __condvar_cancel_waiting (cond, seq, g, private);
+-	      result = err;
+-	      goto done;
+-	    }
+-	  else
+-	    __condvar_dec_grefs (cond, g, private);
+-
+-	  /* Reload signals.  See above for MO.  */
+-	  signals = atomic_load_acquire (cond->__data.__g_signals + g);
++      /* Now wait until a signal is available in our group or it is closed.
++         Acquire MO so that if we observe (signals == lowseq) after group
++         switching in __condvar_switch_g1, we synchronize with that store and
++         will see the prior update of __g1_start done while switching groups
++         too.  */
++      unsigned int signals = atomic_load_acquire (cond->__data.__g_signals + g);
++      uint64_t g1_start = __condvar_load_g1_start_relaxed (cond);
++
++      if (seq < g1_start)
++        {
++          /* If the group is closed already,
++             then this waiter originally had enough extra signals to
++             consume, up until the time its group was closed.  */
++           break;
++        }
++
++      /* If there is an available signal, don't block.
++         If __g1_start has advanced at all, then we must be in G1
++         by now, perhaps in the process of switching back to an older
++         G2, but in either case we're allowed to consume the available
++         signal and should not block anymore.  */
++      if ((int)(signals - (unsigned int)g1_start) > 0)
++        {
++	  /* Try to grab a signal.  See above for MO.  (if we do another loop
++	     iteration we need to see the correct value of g1_start)  */
++	    if (atomic_compare_exchange_weak_acquire (
++			cond->__data.__g_signals + g,
++			&signals, signals - 1))
++	      break;
++	    else
++	      continue;
+ 	}
+ 
++      // Now block.
++      struct _pthread_cleanup_buffer buffer;
++      struct _condvar_cleanup_buffer cbuffer;
++      cbuffer.wseq = wseq;
++      cbuffer.cond = cond;
++      cbuffer.mutex = mutex;
++      cbuffer.private = private;
++      __pthread_cleanup_push (&buffer, __condvar_cleanup_waiting, &cbuffer);
++
++      err = __futex_abstimed_wait_cancelable64 (
++        cond->__data.__g_signals + g, signals, clockid, abstime, private);
++
++      __pthread_cleanup_pop (&buffer, 0);
++
++      if (__glibc_unlikely (err == ETIMEDOUT || err == EOVERFLOW))
++        {
++          /* If we timed out, we effectively cancel waiting.  */
++          __condvar_cancel_waiting (cond, seq, g, private);
++          result = err;
++          break;
++        }
+     }
+-  /* Try to grab a signal.  Use acquire MO so that we see an up-to-date value
+-     of __g1_start below (see spinning above for a similar case).  In
+-     particular, if we steal from a more recent group, we will also see a
+-     more recent __g1_start below.  */
+-  while (!atomic_compare_exchange_weak_acquire (cond->__data.__g_signals + g,
+-						&signals, signals - 2));
+-
+-  /* We consumed a signal but we could have consumed from a more recent group
+-     that aliased with ours due to being in the same group slot.  If this
+-     might be the case our group must be closed as visible through
+-     __g1_start.  */
+-  uint64_t g1_start = __condvar_load_g1_start_relaxed (cond);
+-  if (seq < (g1_start >> 1))
+-    {
+-      /* We potentially stole a signal from a more recent group but we do not
+-	 know which group we really consumed from.
+-	 We do not care about groups older than current G1 because they are
+-	 closed; we could have stolen from these, but then we just add a
+-	 spurious wake-up for the current groups.
+-	 We will never steal a signal from current G2 that was really intended
+-	 for G2 because G2 never receives signals (until it becomes G1).  We
+-	 could have stolen a signal from G2 that was conservatively added by a
+-	 previous waiter that also thought it stole a signal -- but given that
+-	 that signal was added unnecessarily, it's not a problem if we steal
+-	 it.
+-	 Thus, the remaining case is that we could have stolen from the current
+-	 G1, where "current" means the __g1_start value we observed.  However,
+-	 if the current G1 does not have the same slot index as we do, we did
+-	 not steal from it and do not need to undo that.  This is the reason
+-	 for putting a bit with G2's index into__g1_start as well.  */
+-      if (((g1_start & 1) ^ 1) == g)
+-	{
+-	  /* We have to conservatively undo our potential mistake of stealing
+-	     a signal.  We can stop trying to do that when the current G1
+-	     changes because other spinning waiters will notice this too and
+-	     __condvar_quiesce_and_switch_g1 has checked that there are no
+-	     futex waiters anymore before switching G1.
+-	     Relaxed MO is fine for the __g1_start load because we need to
+-	     merely be able to observe this fact and not have to observe
+-	     something else as well.
+-	     ??? Would it help to spin for a little while to see whether the
+-	     current G1 gets closed?  This might be worthwhile if the group is
+-	     small or close to being closed.  */
+-	  unsigned int s = atomic_load_relaxed (cond->__data.__g_signals + g);
+-	  while (__condvar_load_g1_start_relaxed (cond) == g1_start)
+-	    {
+-	      /* Try to add a signal.  We don't need to acquire the lock
+-		 because at worst we can cause a spurious wake-up.  If the
+-		 group is in the process of being closed (LSB is true), this
+-		 has an effect similar to us adding a signal.  */
+-	      if (((s & 1) != 0)
+-		  || atomic_compare_exchange_weak_relaxed
+-		       (cond->__data.__g_signals + g, &s, s + 2))
+-		{
+-		  /* If we added a signal, we also need to add a wake-up on
+-		     the futex.  We also need to do that if we skipped adding
+-		     a signal because the group is being closed because
+-		     while __condvar_quiesce_and_switch_g1 could have closed
+-		     the group, it might still be waiting for futex waiters to
+-		     leave (and one of those waiters might be the one we stole
+-		     the signal from, which cause it to block using the
+-		     futex).  */
+-		  futex_wake (cond->__data.__g_signals + g, 1, private);
+-		  break;
+-		}
+-	      /* TODO Back off.  */
+-	    }
+-	}
+-    }
+-
+- done:
+ 
+   /* Confirm that we have been woken.  We do that before acquiring the mutex
+      to allow for execution of pthread_cond_destroy while having acquired the
+diff --git c/nptl/tst-cond22.c w/nptl/tst-cond22.c
+index 1336e9c79d..bdcb45c536 100644
+--- c/nptl/tst-cond22.c
++++ w/nptl/tst-cond22.c
+@@ -106,13 +106,13 @@ do_test (void)
+       status = 1;
+     }
+ 
+-  printf ("cond = { 0x%x:%x, 0x%x:%x, %u/%u/%u, %u/%u/%u, %u, %u }\n",
++  printf ("cond = { 0x%x:%x, 0x%x:%x, %u/%u, %u/%u, %u, %u }\n",
+ 	  c.__data.__wseq.__value32.__high,
+ 	  c.__data.__wseq.__value32.__low,
+ 	  c.__data.__g1_start.__value32.__high,
+ 	  c.__data.__g1_start.__value32.__low,
+-	  c.__data.__g_signals[0], c.__data.__g_refs[0], c.__data.__g_size[0],
+-	  c.__data.__g_signals[1], c.__data.__g_refs[1], c.__data.__g_size[1],
++	  c.__data.__g_signals[0], c.__data.__g_size[0],
++	  c.__data.__g_signals[1], c.__data.__g_size[1],
+ 	  c.__data.__g1_orig_size, c.__data.__wrefs);
+ 
+   if (pthread_create (&th, NULL, tf, (void *) 1l) != 0)
+@@ -152,13 +152,13 @@ do_test (void)
+       status = 1;
+     }
+ 
+-  printf ("cond = { 0x%x:%x, 0x%x:%x, %u/%u/%u, %u/%u/%u, %u, %u }\n",
++  printf ("cond = { 0x%x:%x, 0x%x:%x, %u/%u, %u/%u, %u, %u }\n",
+ 	  c.__data.__wseq.__value32.__high,
+ 	  c.__data.__wseq.__value32.__low,
+ 	  c.__data.__g1_start.__value32.__high,
+ 	  c.__data.__g1_start.__value32.__low,
+-	  c.__data.__g_signals[0], c.__data.__g_refs[0], c.__data.__g_size[0],
+-	  c.__data.__g_signals[1], c.__data.__g_refs[1], c.__data.__g_size[1],
++	  c.__data.__g_signals[0], c.__data.__g_size[0],
++	  c.__data.__g_signals[1], c.__data.__g_size[1],
+ 	  c.__data.__g1_orig_size, c.__data.__wrefs);
+ 
+   return status;
+diff --git c/sysdeps/nptl/bits/thread-shared-types.h w/sysdeps/nptl/bits/thread-shared-types.h
+index 2de6ff9caf..3fe5d4afc0 100644
+--- c/sysdeps/nptl/bits/thread-shared-types.h
++++ w/sysdeps/nptl/bits/thread-shared-types.h
+@@ -95,11 +95,12 @@ struct __pthread_cond_s
+ {
+   __atomic_wide_counter __wseq;
+   __atomic_wide_counter __g1_start;
+-  unsigned int __g_refs[2] __LOCK_ALIGNMENT;
+-  unsigned int __g_size[2];
++  unsigned int __g_size[2] __LOCK_ALIGNMENT;
+   unsigned int __g1_orig_size;
+   unsigned int __wrefs;
+   unsigned int __g_signals[2];
++  unsigned int __unused_initialized_1;
++  unsigned int __unused_initialized_2;
+ };
+ 
+ typedef unsigned int __tss_t;
+diff --git c/sysdeps/nptl/pthread.h w/sysdeps/nptl/pthread.h
+index 7f65483542..476cd0ed54 100644
+--- c/sysdeps/nptl/pthread.h
++++ w/sysdeps/nptl/pthread.h
+@@ -152,7 +152,7 @@ enum
+ 
+ 
+ /* Conditional variable handling.  */
+-#define PTHREAD_COND_INITIALIZER { { {0}, {0}, {0, 0}, {0, 0}, 0, 0, {0, 0} } }
++#define PTHREAD_COND_INITIALIZER { { {0}, {0}, {0, 0}, 0, 0, {0, 0}, 0, 0 } }
+ 
+ 
+ /* Cleanup buffers */

regcomp-double-free.patch

@@ -0,0 +1,240 @@
+From 2eb180377b96771b8368b0915669c8c7b267e739 Mon Sep 17 00:00:00 2001
+From: Florian Weimer <fweimer@redhat.com>
+Date: Mon, 21 Jul 2025 21:43:49 +0200
+Subject: [PATCH] posix: Fix double-free after allocation failure in regcomp
+ (bug 33185)
+
+If a memory allocation failure occurs during bracket expression
+parsing in regcomp, a double-free error may result.
+
+Reported-by: Anastasia Belova <abelova@astralinux.ru>
+Co-authored-by: Paul Eggert <eggert@cs.ucla.edu>
+Reviewed-by: Andreas K. Huettel <dilfridge@gentoo.org>
+(cherry picked from commit 7ea06e994093fa0bcca0d0ee2c1db271d8d7885d)
+---
+ NEWS                             |   1 +
+ posix/Makefile                   |   1 +
+ posix/regcomp.c                  |   4 +-
+ posix/tst-regcomp-bracket-free.c | 176 +++++++++++++++++++++++++++++++
+ 4 files changed, 181 insertions(+), 1 deletion(-)
+ create mode 100644 posix/tst-regcomp-bracket-free.c
+
+diff --git a/posix/Makefile b/posix/Makefile
+index 2c598cd20a..830278a423 100644
+--- a/posix/Makefile
++++ b/posix/Makefile
+@@ -303,6 +303,7 @@ tests := \
+   tst-posix_spawn-setsid \
+   tst-preadwrite \
+   tst-preadwrite64 \
++  tst-regcomp-bracket-free \
+   tst-regcomp-truncated \
+   tst-regex \
+   tst-regex2 \
+diff --git a/posix/regcomp.c b/posix/regcomp.c
+index 5380d3c7b9..6595bb3c0d 100644
+--- a/posix/regcomp.c
++++ b/posix/regcomp.c
+@@ -3384,6 +3384,7 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token,
+     {
+ #ifdef RE_ENABLE_I18N
+       free_charset (mbcset);
++      mbcset = NULL;
+ #endif
+       /* Build a tree for simple bracket.  */
+       br_token.type = SIMPLE_BRACKET;
+@@ -3399,7 +3400,8 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token,
+  parse_bracket_exp_free_return:
+   re_free (sbcset);
+ #ifdef RE_ENABLE_I18N
+-  free_charset (mbcset);
++  if (__glibc_likely (mbcset != NULL))
++    free_charset (mbcset);
+ #endif /* RE_ENABLE_I18N */
+   return NULL;
+ }
+diff --git a/posix/tst-regcomp-bracket-free.c b/posix/tst-regcomp-bracket-free.c
+new file mode 100644
+index 0000000000..3c091d8c44
+--- /dev/null
++++ b/posix/tst-regcomp-bracket-free.c
+@@ -0,0 +1,176 @@
++/* Test regcomp bracket parsing with injected allocation failures (bug 33185).
++   Copyright (C) 2025 Free Software Foundation, Inc.
++   This file is part of the GNU C Library.
++
++   The GNU C Library is free software; you can redistribute it and/or
++   modify it under the terms of the GNU Lesser General Public
++   License as published by the Free Software Foundation; either
++   version 2.1 of the License, or (at your option) any later version.
++
++   The GNU C Library is distributed in the hope that it will be useful,
++   but WITHOUT ANY WARRANTY; without even the implied warranty of
++   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
++   Lesser General Public License for more details.
++
++   You should have received a copy of the GNU Lesser General Public
++   License along with the GNU C Library; if not, see
++   <https://www.gnu.org/licenses/>.  */
++
++/* This test invokes regcomp multiple times, failing one memory
++   allocation in each call.  The function call should fail with
++   REG_ESPACE (or succeed if it can recover from the allocation
++   failure).  Previously, there was double-free bug.  */
++
++#include <errno.h>
++#include <regex.h>
++#include <stdio.h>
++#include <string.h>
++#include <support/check.h>
++#include <support/namespace.h>
++#include <support/support.h>
++
++/* Data structure allocated via MAP_SHARED, so that writes from the
++   subprocess are visible.  */
++struct shared_data
++{
++  /* Number of tracked allocations performed so far.  */
++  volatile unsigned int allocation_count;
++
++  /* If this number is reached, one allocation fails.  */
++  volatile unsigned int failing_allocation;
++
++  /* The subprocess stores the expected name here.  */
++  char name[100];
++};
++
++/* Allocation count in shared mapping.  */
++static struct shared_data *shared;
++
++/* Returns true if a failure should be injected for this allocation.  */
++static bool
++fail_this_allocation (void)
++{
++  if (shared != NULL)
++    {
++      unsigned int count = shared->allocation_count;
++      shared->allocation_count = count + 1;
++      return count == shared->failing_allocation;
++    }
++  else
++    return false;
++}
++
++/* Failure-injecting wrappers for allocation functions used by glibc.  */
++
++void *
++malloc (size_t size)
++{
++  if (fail_this_allocation ())
++    {
++      errno = ENOMEM;
++      return NULL;
++    }
++  extern __typeof (malloc) __libc_malloc;
++  return __libc_malloc (size);
++}
++
++void *
++calloc (size_t a, size_t b)
++{
++  if (fail_this_allocation ())
++    {
++      errno = ENOMEM;
++      return NULL;
++    }
++  extern __typeof (calloc) __libc_calloc;
++  return __libc_calloc (a, b);
++}
++
++void *
++realloc (void *ptr, size_t size)
++{
++  if (fail_this_allocation ())
++    {
++      errno = ENOMEM;
++      return NULL;
++    }
++  extern __typeof (realloc) __libc_realloc;
++  return __libc_realloc (ptr, size);
++}
++
++/* No-op subprocess to verify that support_isolate_in_subprocess does
++   not perform any heap allocations.  */
++static void
++no_op (void *ignored)
++{
++}
++
++/* Perform a regcomp call in a subprocess.  Used to count its
++   allocations.  */
++static void
++initialize (void *regexp1)
++{
++  const char *regexp = regexp1;
++
++  shared->allocation_count = 0;
++
++  regex_t reg;
++  TEST_COMPARE (regcomp (&reg, regexp, 0), 0);
++}
++
++/* Perform regcomp in a subprocess with fault injection.  */
++static void
++test_in_subprocess (void *regexp1)
++{
++  const char *regexp = regexp1;
++  unsigned int inject_at = shared->failing_allocation;
++
++  regex_t reg;
++  int ret = regcomp (&reg, regexp, 0);
++
++  if (ret != 0)
++    {
++      TEST_COMPARE (ret, REG_ESPACE);
++      printf ("info: allocation %u failure results in return value %d,"
++              " error %s (%d)\n",
++              inject_at, ret, strerrorname_np (errno), errno);
++    }
++}
++
++static int
++do_test (void)
++{
++  char regexp[] = "[:alpha:]";
++
++  shared = support_shared_allocate (sizeof (*shared));
++
++  /* Disable fault injection.  */
++  shared->failing_allocation = ~0U;
++
++  support_isolate_in_subprocess (no_op, NULL);
++  TEST_COMPARE (shared->allocation_count, 0);
++
++  support_isolate_in_subprocess (initialize, regexp);
++
++  /* The number of allocations in the successful case, plus some
++     slack.  Once the number of expected allocations is exceeded,
++     injecting further failures does not make a difference.  */
++  unsigned int maximum_allocation_count = shared->allocation_count;
++  printf ("info: successful call performs %u allocations\n",
++          maximum_allocation_count);
++  maximum_allocation_count += 10;
++
++  for (unsigned int inject_at = 0; inject_at <= maximum_allocation_count;
++       ++inject_at)
++    {
++      shared->allocation_count = 0;
++      shared->failing_allocation = inject_at;
++      support_isolate_in_subprocess (test_in_subprocess, regexp);
++    }
++
++  support_shared_free (shared);
++
++  return 0;
++}
++
++#include <support/test-driver.c>
+-- 
+2.50.1
+

sle-nsswitch.conf

@@ -1,75 +0,0 @@
-#
-# /etc/nsswitch.conf
-#
-# An example Name Service Switch config file. This file should be
-# sorted with the most-used services at the beginning.
-#
-# Valid databases are: aliases, ethers, group, gshadow, hosts,
-# initgroups, netgroup, networks, passwd, protocols, publickey,
-# rpc, services, and shadow.
-#
-# Valid service provider entries include (in alphabetical order):
-#
-#	compat			Use /etc files plus *_compat pseudo-db
-#	db			Use the pre-processed /var/db files
-#	dns			Use DNS (Domain Name Service)
-#	files			Use the local files in /etc
-#	hesiod			Use Hesiod (DNS) for user lookups
-#	nis			Use NIS (NIS version 2), also called YP
-#	nisplus			Use NIS+ (NIS version 3)
-#
-# See `info libc 'NSS Basics'` for more information.
-#
-# Commonly used alternative service providers (may need installation):
-#
-#	ldap			Use LDAP directory server
-#	myhostname		Use systemd host names
-#	mymachines		Use systemd machine names
-#	mdns*, mdns*_minimal	Use Avahi mDNS/DNS-SD
-#	resolve			Use systemd resolved resolver
-#	sss			Use System Security Services Daemon (sssd)
-#	systemd			Use systemd for dynamic user option
-#	winbind			Use Samba winbind support
-#	wins			Use Samba wins support
-#	wrapper			Use wrapper module for testing
-#
-# Notes:
-#
-# 'sssd' performs its own 'files'-based caching, so it should generally
-# come before 'files'.
-#
-# WARNING: Running nscd with a secondary caching service like sssd may
-# 	   lead to unexpected behaviour, especially with how long
-# 	   entries are cached.
-#
-# Installation instructions:
-#
-# To use 'db', install the appropriate package(s) (provide 'makedb' and
-# libnss_db.so.*), and place the 'db' in front of 'files' for entries
-# you want to be looked up first in the databases, like this:
-#
-# passwd:    db files
-# shadow:    db files
-# group:     db files
-
-passwd:		compat
-group:		compat
-shadow:		compat
-# Allow initgroups to default to the setting for group.
-# initgroups:	compat
-
-hosts:  	files dns
-networks:	files dns
-
-aliases:	files usrfiles
-ethers:		files usrfiles
-gshadow:	files usrfiles
-netgroup:	files nis
-protocols:	files usrfiles
-publickey:	files
-rpc:		files usrfiles
-services:	files usrfiles
-
-automount:	files nis
-bootparams:	files
-netmasks:	files

ulp-prologue-into-asm-functions.patch

@@ -1,6 +1,6 @@
-From ec4f0a28ed48c51165e3e72c7427efb0ae14a124 Mon Sep 17 00:00:00 2001
+From 14848b3c532520e71ba2614eb23f173f83c541d2 Mon Sep 17 00:00:00 2001
 From: Giuliano Belinassi <gbelinassi@suse.de>
-Date: Mon, 6 May 2024 20:09:55 -0300
+Date: Thu, 2 Jan 2025 22:40:22 +0100
 Subject: [PATCH] Add Userspace Livepatch prologue into ASM functions
 
 Userspace Live Patching (ULP) refers to the process of applying
@@ -12,38 +12,24 @@ this have to be included manually. This patch does this.
 
 Signed-off-by: Giuliano Belinassi <gbelinassi@suse.de>
 ---
- Makeconfig                               |  5 +++
  config.h.in                              |  3 ++
  config.make.in                           |  1 +
- configure                                | 21 +++++++++
+ configure                                | 22 ++++++++++
  configure.ac                             | 13 ++++++
+ sysdeps/powerpc/powerpc64/le/Makefile    |  5 +++
+ sysdeps/powerpc/powerpc64/sysdep.h       | 55 ++++++++++++++++++++++--
+ sysdeps/x86_64/Makefile                  |  5 +++
  sysdeps/x86_64/multiarch/strcmp-avx2.S   |  5 +--
  sysdeps/x86_64/multiarch/strcmp-evex.S   |  5 +--
  sysdeps/x86_64/multiarch/strcmp-sse4_2.S |  5 +--
- sysdeps/x86_64/sysdep.h                  | 54 ++++++++++++++++++++++--
- 9 files changed, 96 insertions(+), 16 deletions(-)
+ sysdeps/x86_64/sysdep.h                  | 54 +++++++++++++++++++++--
+ 11 files changed, 153 insertions(+), 20 deletions(-)
 
-diff --git a/Makeconfig b/Makeconfig
-index e583765712..b136e10224 100644
---- a/Makeconfig
-+++ b/Makeconfig
-@@ -981,6 +981,11 @@ else
- +cflags += $(no-fortify-source)
- endif
- 
-+# Add flags for Userspace Livepatching support.
-+ifeq (yes,$(enable-userspace-livepatch))
-++cflags	+= -fpatchable-function-entry=16,14
-+endif
-+
- # Each sysdeps directory can contain header files that both will be
- # used to compile and will be installed.  Each can also contain an
- # include/ subdirectory, whose header files will be used to compile
 diff --git a/config.h.in b/config.h.in
-index c4cc7d3b9a..dc70ed03d0 100644
+index f495f11244..6c46990c8d 100644
 --- a/config.h.in
 +++ b/config.h.in
-@@ -205,6 +205,9 @@
+@@ -214,6 +214,9 @@
  /* Define to 1 if libpthread actually resides in libc.  */
  #define PTHREAD_IN_LIBC 0
  
@@ -54,7 +40,7 @@ index c4cc7d3b9a..dc70ed03d0 100644
  #define TIMEOUTFACTOR 1
  
 diff --git a/config.make.in b/config.make.in
-index 55e8b7563b..0f14c05d62 100644
+index 36096881b7..04cf873fad 100644
 --- a/config.make.in
 +++ b/config.make.in
 @@ -81,6 +81,7 @@ mach-interface-list = @mach_interface_list@
@@ -66,10 +52,10 @@ index 55e8b7563b..0f14c05d62 100644
  build-profile = @profile@
  build-static-nss = @static_nss@
 diff --git a/configure b/configure
-index 432e40a592..eb6b203925 100755
+index 1d543548cd..f6275d327e 100755
 --- a/configure
 +++ b/configure
-@@ -622,6 +622,7 @@ LIBOBJS
+@@ -615,6 +615,7 @@ LIBOBJS
  pthread_in_libc
  RELEASE
  VERSION
@@ -77,7 +63,7 @@ index 432e40a592..eb6b203925 100755
  mach_interface_list
  DEFINES
  static_nss
-@@ -812,6 +813,7 @@ enable_cet
+@@ -809,6 +810,7 @@ enable_cet
  enable_scv
  enable_fortify_source
  with_cpu
@@ -85,7 +71,7 @@ index 432e40a592..eb6b203925 100755
  '
        ac_precious_vars='build_alias
  host_alias
-@@ -1490,6 +1492,8 @@ Optional Features:
+@@ -1491,6 +1493,8 @@ Optional Features:
                            Use -D_FORTIFY_SOURCE=[1|2|3] to control code
                            hardening, defaults to highest possible value
                            supported by the build compiler.
@@ -94,7 +80,7 @@ index 432e40a592..eb6b203925 100755
  
  Optional Packages:
    --with-PACKAGE[=ARG]    use PACKAGE [ARG=yes]
-@@ -7867,6 +7871,23 @@ libc_cv_multidir=`${CC-cc} $CFLAGS $CPPFLAGS -print-multi-directory`
+@@ -8096,6 +8100,24 @@ libc_cv_multidir=`${CC-cc} $CFLAGS $CPPFLAGS -print-multi-directory`
  
  
  
@@ -102,8 +88,9 @@ index 432e40a592..eb6b203925 100755
 +if test ${enable_userspace_livepatch+y}
 +then :
 +  enableval=$enable_userspace_livepatch; enable_userspace_livepatch=$enableval
-+else $as_nop
-+  enable_userspace_livepatch=no
++else case e in #(
++  e) enable_userspace_livepatch=no ;;
++esac
 +fi
 +
 +
@@ -119,10 +106,10 @@ index 432e40a592..eb6b203925 100755
  RELEASE=`sed -n -e 's/^#define RELEASE "\([^"]*\)"/\1/p' < $srcdir/version.h`
  
 diff --git a/configure.ac b/configure.ac
-index bdc385d03c..cb4d28b3d8 100644
+index 9cbc0bf68f..4ba19209b7 100644
 --- a/configure.ac
 +++ b/configure.ac
-@@ -1756,6 +1756,19 @@ AC_SUBST(DEFINES)
+@@ -1780,6 +1780,19 @@ AC_SUBST(DEFINES)
  dnl See sysdeps/mach/configure.ac for this variable.
  AC_SUBST(mach_interface_list)
  
@@ -142,6 +129,114 @@ index bdc385d03c..cb4d28b3d8 100644
  VERSION=`sed -n -e 's/^#define VERSION "\([^"]*\)"/\1/p' < $srcdir/version.h`
  RELEASE=`sed -n -e 's/^#define RELEASE "\([^"]*\)"/\1/p' < $srcdir/version.h`
  AC_SUBST(VERSION)
+diff --git a/sysdeps/powerpc/powerpc64/le/Makefile b/sysdeps/powerpc/powerpc64/le/Makefile
+index b77775cf95..2f81e562db 100644
+--- a/sysdeps/powerpc/powerpc64/le/Makefile
++++ b/sysdeps/powerpc/powerpc64/le/Makefile
+@@ -1,3 +1,8 @@
++# Add flags for Userspace Livepatching support.
++ifeq (yes,$(enable-userspace-livepatch))
+++cflags += -fpatchable-function-entry=14,13 -msplit-patch-nops
++endif
++
+ # When building float128 we need to ensure -mfloat128 is
+ # passed to all such object files.
+ type-float128-CFLAGS := -mfloat128
+diff --git a/sysdeps/powerpc/powerpc64/sysdep.h b/sysdeps/powerpc/powerpc64/sysdep.h
+index c363939e1a..33ed236407 100644
+--- a/sysdeps/powerpc/powerpc64/sysdep.h
++++ b/sysdeps/powerpc/powerpc64/sysdep.h
+@@ -162,6 +162,46 @@
+ BODY_LABEL(\name):
+ 	.endm
+ 
++/* Libpulp uses -fpatchable-function-entry to add padding NOPS to the
++   prologue of all functions. This works for C functions. For functions
++   written in ASM, the way we do this is by adding this prologue manually.  */
++
++#if ENABLE_USERSPACE_LIVEPATCH
++
++/* Instructions to be inserted before the function label.  */
++# define ULP_NOPS_PRE_PROLOGUE     .rept 13; nop; .endr
++
++/* Instruction to be inserted after the function label.  */
++# define ULP_NOPS_POST_PROLOGUE    .rept  1; nop; .endr
++
++
++/* this macro expands according to the following condition:
++ * if name = _start, then the prologue is not inserted.
++ * if name = _dl_relocate_static_pie, then the prologue is not inserted.
++ * if name = anything else, then the prologue is inserted.
++ **/
++# define __ULP_PRE_PROLOGUE_dl_relocate_static_pie    ,
++# define __ULP_PRE_PROLOGUE_start                     ,
++# define __ULP_PRE_PROLOGUE(x, y,...)                 y
++# define _ULP_PRE_PROLOGUE(x, ...)                    __ULP_PRE_PROLOGUE(x, __VA_ARGS__)
++# define ULP_PRE_PROLOGUE(name)                       _ULP_PRE_PROLOGUE(__ULP_PRE_PROLOGUE##name, ULP_NOPS_PRE_PROLOGUE,)
++
++/* this macro expands according to the following condition:
++ * if name = _start, then the postlogue is not inserted.
++ * if name = _dl_relocate_static_pie, then the postlogue is not inserted.
++ * if name = anything else, then the postlogue is inserted.
++ **/
++# define __ULP_POST_PROLOGUE_dl_relocate_static_pie   ,
++# define __ULP_POST_PROLOGUE_start                    ,
++# define __ULP_POST_PROLOGUE(x, y,...)                y
++# define _ULP_POST_PROLOGUE(x, ...)                   __ULP_POST_PROLOGUE(x, __VA_ARGS__)
++# define ULP_POST_PROLOGUE(name)                      _ULP_POST_PROLOGUE(__ULP_POST_PROLOGUE##name, ULP_NOPS_POST_PROLOGUE,)
++
++#else
++# define ULP_PRE_PROLOGUE(name)
++# define ULP_POST_PROLOGUE(name)
++#endif
++
+ /* Use ENTRY_TOCLESS for functions that make no use of r2 and
+    guarantee r2 is unchanged on exit.  Any function that has @toc or
+    @got relocs uses r2.  Functions that call other functions via the
+@@ -174,19 +214,26 @@ BODY_LABEL(\name):
+    before the start of the function's code.   */
+ #ifndef PROF
+ #define ENTRY_TOCLESS(name, ...)		\
++	ULP_PRE_PROLOGUE(name);			\
+ 	ENTRY_3 name, ## __VA_ARGS__;		\
+-	cfi_startproc
++	cfi_startproc;				\
++	ULP_POST_PROLOGUE(name)
+ 
+ #define ENTRY(name, ...)			\
+-	ENTRY_TOCLESS(name, ## __VA_ARGS__);	\
+-	LOCALENTRY(name)
++	ULP_PRE_PROLOGUE(name);			\
++	ENTRY_3 name, ## __VA_ARGS__;		\
++	cfi_startproc;				\
++	LOCALENTRY(name);			\
++	ULP_POST_PROLOGUE(name)
+ #else
+ /* The call to _mcount is potentially via the plt, so profiling code
+    is never free of an r2 use.  */
+ #define ENTRY_TOCLESS(name, ...)		\
++	ULP_PRE_PROLOGUE(name);			\
+ 	ENTRY_3 name, ## __VA_ARGS__;		\
+ 	cfi_startproc;				\
+-	LOCALENTRY(name)
++	LOCALENTRY(name);                       \
++	ULP_POST_PROLOGUE(name)
+ 
+ #define ENTRY(name, ...)			\
+ 	ENTRY_TOCLESS(name, ## __VA_ARGS__)
+diff --git a/sysdeps/x86_64/Makefile b/sysdeps/x86_64/Makefile
+index ce949dba27..44e9f38f4e 100644
+--- a/sysdeps/x86_64/Makefile
++++ b/sysdeps/x86_64/Makefile
+@@ -1,3 +1,8 @@
++# Add flags for Userspace Livepatching support.
++ifeq (yes,$(enable-userspace-livepatch))
+++cflags += -fpatchable-function-entry=16,14
++endif
++
+ # The i387 `long double' is a distinct type we support.
+ long-double-fcts = yes
+ 
 diff --git a/sysdeps/x86_64/multiarch/strcmp-avx2.S b/sysdeps/x86_64/multiarch/strcmp-avx2.S
 index 5bc1d90078..3ea96c0aa3 100644
 --- a/sysdeps/x86_64/multiarch/strcmp-avx2.S
@@ -268,5 +363,5 @@ index db6e36b2dd..86a5d1b2be 100644
  /* This macro is for setting proper CFI with DW_CFA_expression describing
     the register as saved relative to %rsp instead of relative to the CFA.
 -- 
-2.44.0
+2.49.0