Add elision tunables

This patch adds several new tunables to control the behavior of
elision on supported platforms[1].   Since elision now depends
on tunables, we should always *compile* with elision enabled,
and leave the code disabled, but available for runtime
selection.  This gives us *much* better compile-time testing of
the existing code to avoid bit-rot[2].

Tested on ppc, ppc64, ppc64le, s390x and x86_64.

[1] This part of the patch was initially proposed by
Paul Murphy but was "staled" because the framework have changed
since the patch was originally proposed:

https://patchwork.sourceware.org/patch/10342/

[2] This part of the patch was inititally proposed as a RFC by
Carlos O'Donnell.  Make sense to me integrate this on the patch:

https://sourceware.org/ml/libc-alpha/2017-05/msg00335.html

	* elf/dl-tunables.list: Add elision parameters.
	* manual/tunables.texi: Add entries about elision tunable.
	* sysdeps/unix/sysv/linux/powerpc/elision-conf.c:
	Add callback functions to dynamically enable/disable elision.
	Add multiple callbacks functions to set elision parameters.
	Deleted __libc_enable_secure check.
	* sysdeps/unix/sysv/linux/s390/elision-conf.c: Likewise.
	* sysdeps/unix/sysv/linux/x86/elision-conf.c: Likewise.
	* configure: Regenerated.
	* configure.ac: Option enable_lock_elision was deleted.
	* config.h.in: ENABLE_LOCK_ELISION flag was deleted.
	* config.make.in: Remove references to enable_lock_elision.
	* manual/install.texi: Elision configure option was removed.
	* INSTALL: Regenerated to remove enable_lock_elision.
	* nptl/Makefile:
	Disable elision so it can verify error case for destroying a mutex.
	* sysdeps/powerpc/nptl/elide.h:
	Cleanup ENABLE_LOCK_ELISION check.
	Deleted macros for the case when ENABLE_LOCK_ELISION was not defined.
	* sysdeps/s390/configure: Regenerated.
	* sysdeps/s390/configure.ac: Remove references to enable_lock_elision..
	* nptl/tst-mutex8.c:
	Deleted all #ifndef ENABLE_LOCK_ELISION from the test.
	* sysdeps/powerpc/powerpc32/sysdep.h:
	Deleted all ENABLE_LOCK_ELISION checks.
	* sysdeps/powerpc/powerpc64/sysdep.h: Likewise.
	* sysdeps/powerpc/sysdep.h: Likewise.
	* sysdeps/s390/nptl/bits/pthreadtypes-arch.h: Likewise.
	* sysdeps/unix/sysv/linux/powerpc/force-elision.h: Likewise.
	* sysdeps/unix/sysv/linux/s390/elision-conf.h: Likewise.
	* sysdeps/unix/sysv/linux/s390/force-elision.h: Likewise.
	* sysdeps/unix/sysv/linux/s390/lowlevellock.h: Likewise.
	* sysdeps/unix/sysv/linux/s390/Makefile: Remove references to
	enable-lock-elision.

Reviewed-by: Tulio Magno Quites Machado Filho <tuliom@linux.vnet.ibm.com>

ChangeLog

@@ -1,3 +1,42 @@
+2017-12-05  Rogerio A. Cardoso  <rcardoso@linux.vnet.ibm.com>,
+	    Paul E. Murphy  <murphyp@linux.vnet.ibm.com>,
+	    Carlos O'Donnell <carlos@redhat.com>
+
+	* elf/dl-tunables.list: Add elision parameters.
+	* manual/tunables.texi: Add entries about elision tunable.
+	* sysdeps/unix/sysv/linux/powerpc/elision-conf.c:
+	Add callback functions to dynamically enable/disable elision.
+	Add multiple callbacks functions to set elision parameters.
+	Deleted __libc_enable_secure check.
+	* sysdeps/unix/sysv/linux/s390/elision-conf.c: Likewise.
+	* sysdeps/unix/sysv/linux/x86/elision-conf.c: Likewise.
+	* configure: Regenerated.
+	* configure.ac: Option enable_lock_elision was deleted.
+	* config.h.in: ENABLE_LOCK_ELISION flag was deleted.
+	* config.make.in: Remove references to enable_lock_elision.
+	* manual/install.texi: Elision configure option was removed.
+	* INSTALL: Regenerated to remove enable_lock_elision.
+	* nptl/Makefile:
+	Disable elision so it can verify error case for destroying a mutex.
+	* sysdeps/powerpc/nptl/elide.h:
+	Cleanup ENABLE_LOCK_ELISION check.
+	Deleted macros for the case when ENABLE_LOCK_ELISION was not defined.
+	* sysdeps/s390/configure: Regenerated.
+	* sysdeps/s390/configure.ac: Remove references to enable_lock_elision..
+	* nptl/tst-mutex8.c:
+	Deleted all #ifndef ENABLE_LOCK_ELISION from the test.
+	* sysdeps/powerpc/powerpc32/sysdep.h:
+	Deleted all ENABLE_LOCK_ELISION checks.
+	* sysdeps/powerpc/powerpc64/sysdep.h: Likewise.
+	* sysdeps/powerpc/sysdep.h: Likewise.
+	* sysdeps/s390/nptl/bits/pthreadtypes-arch.h: Likewise.
+	* sysdeps/unix/sysv/linux/powerpc/force-elision.h: Likewise.
+	* sysdeps/unix/sysv/linux/s390/elision-conf.h: Likewise.
+	* sysdeps/unix/sysv/linux/s390/force-elision.h: Likewise.
+	* sysdeps/unix/sysv/linux/s390/lowlevellock.h: Likewise.
+	* sysdeps/unix/sysv/linux/s390/Makefile: Remove references to
+	enable-lock-elision.
+
 2017-12-05  Joseph Myers  <joseph@codesourcery.com>
 
 	* stdlib/strtod.c: Include <bits/floatn.h>.

INSTALL

@@ -116,9 +116,6 @@ will be used, and CFLAGS sets optimization options for the compiler.
      formats may change over time.  Consult the 'timezone' subdirectory
      for more details.
 
-'--enable-lock-elision=yes'
-     Enable lock elision for pthread mutexes by default.
-
 '--enable-stack-protector'
 '--enable-stack-protector=strong'
 '--enable-stack-protector=all'

config.h.in

@@ -131,9 +131,6 @@
 /* Define if __stack_chk_guard canary should be randomized at program startup.  */
 #undef ENABLE_STACKGUARD_RANDOMIZE
 
-/* Define if lock elision should be enabled by default.  */
-#undef ENABLE_LOCK_ELISION
-
 /* Package description.  */
 #undef PKGVERSION
 

config.make.in

@@ -99,7 +99,6 @@ build-nscd = @build_nscd@
 use-nscd = @use_nscd@
 build-hardcoded-path-in-tests= @hardcoded_path_in_tests@
 build-pt-chown = @build_pt_chown@
-enable-lock-elision = @enable_lock_elision@
 have-tunables = @have_tunables@
 
 # Build tools.

configure

@@ -679,7 +679,6 @@ enable_werror
 all_warnings
 force_install
 bindnow
-enable_lock_elision
 hardcoded_path_in_tests
 enable_timezone_tools
 use_default_link
@@ -768,7 +767,6 @@ enable_profile
 enable_timezone_tools
 enable_hardcoded_path_in_tests
 enable_stackguard_randomization
-enable_lock_elision
 enable_hidden_plt
 enable_bind_now
 enable_stack_protector
@@ -1428,8 +1426,6 @@ Optional Features:
   --enable-stackguard-randomization
                           initialize __stack_chk_guard canary with a random
                           number at program start
-  --enable-lock-elision=yes/no
-                          Enable lock elision for pthread mutexes by default
   --disable-hidden-plt    do not hide internal function calls to avoid PLT
   --enable-bind-now       disable lazy relocations in DSOs
   --enable-stack-protector=[yes|no|all|strong]
@@ -3395,19 +3391,6 @@ if test "$enable_stackguard_randomize" = yes; then
 
 fi
 
-# Check whether --enable-lock-elision was given.
-if test "${enable_lock_elision+set}" = set; then :
-  enableval=$enable_lock_elision; enable_lock_elision=$enableval
-else
-  enable_lock_elision=no
-fi
-
-
-if test "$enable_lock_elision" = yes ; then
-  $as_echo "#define ENABLE_LOCK_ELISION 1" >>confdefs.h
-
-fi
-
 # Check whether --enable-hidden-plt was given.
 if test "${enable_hidden_plt+set}" = set; then :
   enableval=$enable_hidden_plt; hidden=$enableval

configure.ac

@@ -199,16 +199,6 @@ if test "$enable_stackguard_randomize" = yes; then
   AC_DEFINE(ENABLE_STACKGUARD_RANDOMIZE)
 fi
 
-AC_ARG_ENABLE([lock-elision],
-	      AC_HELP_STRING([--enable-lock-elision[=yes/no]],
-			     [Enable lock elision for pthread mutexes by default]),
-	      [enable_lock_elision=$enableval],
-	      [enable_lock_elision=no])
-AC_SUBST(enable_lock_elision)
-if test "$enable_lock_elision" = yes ; then
-  AC_DEFINE(ENABLE_LOCK_ELISION)
-fi
-
 AC_ARG_ENABLE([hidden-plt],
 	      AC_HELP_STRING([--disable-hidden-plt],
 			     [do not hide internal function calls to avoid PLT]),

elf/dl-tunables.list

@@ -96,4 +96,38 @@ glibc {
       default: HWCAP_IMPORTANT
       }
   }
+
+  elision {
+    enable {
+      type: INT_32
+      minval: 0
+      maxval: 1
+      security_level: SXID_ERASE
+    }
+    skip_lock_busy {
+      type: INT_32
+      default: 3
+      security_level: SXID_ERASE
+    }
+    skip_lock_internal_abort {
+      type: INT_32
+      default: 3
+      security_level: SXID_ERASE
+    }
+    skip_lock_after_retries {
+      type: INT_32
+      default: 3
+      security_level: SXID_ERASE
+    }
+    tries {
+      type: INT_32
+      default: 3
+      security_level: SXID_ERASE
+    }
+    skip_trylock_internal_abort {
+      type: INT_32
+      default: 3
+      security_level: SXID_ERASE
+    }
+  }
 }

manual/install.texi

@@ -145,9 +145,6 @@ Note that you need to make sure the external tools are kept in sync with
 the versions that @theglibc{} expects as the data formats may change over
 time.  Consult the @file{timezone} subdirectory for more details.
 
-@item --enable-lock-elision=yes
-Enable lock elision for pthread mutexes by default.
-
 @item --enable-stack-protector
 @itemx --enable-stack-protector=strong
 @itemx --enable-stack-protector=all

manual/tunables.texi

@@ -31,6 +31,7 @@ their own namespace.
 @menu
 * Tunable names::  The structure of a tunable name
 * Memory Allocation Tunables::  Tunables in the memory allocation subsystem
+* Elision Tunables::  Tunables in elision subsystem
 * Hardware Capability Tunables::  Tunables that modify the hardware
 				  capabilities seen by @theglibc{}
 @end menu
@@ -212,6 +213,74 @@ pre-fill the per-thread cache with.  The default, or when set to zero,
 is no limit.
 @end deftp
 
+@node Elision Tunables
+@section Elision Tunables
+@cindex elision tunables
+@cindex tunables, elision
+
+@deftp {Tunable namespace} glibc.elision
+Contended locks are usually slow and can lead to performance and scalability
+issues in multithread code. Lock elision will use memory transactions to under
+certain conditions, to elide locks and improve performance.
+Elision behavior can be modified by setting the following tunables in
+the @code{elision} namespace:
+@end deftp
+
+@deftp Tunable glibc.elision.enable
+The @code{glibc.elision.enable} tunable enables lock elision if the feature is
+supported by the hardware.  If elision is not supported by the hardware this
+tunable has no effect.
+
+Elision tunables are supported for 64-bit Intel, IBM POWER, and z System
+architectures.
+@end deftp
+
+@deftp Tunable glibc.elision.skip_lock_busy
+The @code{glibc.elision.skip_lock_busy} tunable sets how many times to use a
+non-transactional lock after a transactional failure has occurred because the
+lock is already acquired.  Expressed in number of lock acquisition attempts.
+
+The default value of this tunable is @samp{3}.
+@end deftp
+
+@deftp Tunable glibc.elision.skip_lock_internal_abort
+The @code{glibc.elision.skip_lock_internal_abort} tunable sets how many times
+the thread should avoid using elision if a transaction aborted for any reason
+other than a different thread's memory accesses.  Expressed in number of lock
+acquisition attempts.
+
+The default value of this tunable is @samp{3}.
+@end deftp
+
+@deftp Tunable glibc.elision.skip_lock_after_retries
+The @code{glibc.elision.skip_lock_after_retries} tunable sets how many times
+to try to elide a lock with transactions, that only failed due to a different
+thread's memory accesses, before falling back to regular lock.
+Expressed in number of lock elision attempts.
+
+This tunable is supported only on IBM POWER, and z System architectures.
+
+The default value of this tunable is @samp{3}.
+@end deftp
+
+@deftp Tunable glibc.elision.tries
+The @code{glibc.elision.tries} sets how many times to retry elision if there is
+chance for the transaction to finish execution e.g., it wasn't
+aborted due to the lock being already acquired.  If elision is not supported
+by the hardware this tunable is set to @samp{0} to avoid retries.
+
+The default value of this tunable is @samp{3}.
+@end deftp
+
+@deftp Tunable glibc.elision.skip_trylock_internal_abort
+The @code{glibc.elision.skip_trylock_internal_abort} tunable sets how many
+times the thread should avoid trying the lock if a transaction aborted due to
+reasons other than a different thread's memory accesses.  Expressed in number
+of try lock attempts.
+
+The default value of this tunable is @samp{3}.
+@end deftp
+
 @node Hardware Capability Tunables
 @section Hardware Capability Tunables
 @cindex hardware capability tunables

nptl/Makefile

@@ -714,6 +714,10 @@ endif
 
 $(objpfx)tst-compat-forwarder: $(objpfx)tst-compat-forwarder-mod.so
 
+# Disable elision for tst-mutex8 so it can verify error case for
+# destroying a mutex.
+tst-mutex8-ENV = GLIBC_TUNABLES=glibc.elision.enable=0
+
 # The tests here better do not run in parallel
 ifneq ($(filter %tests,$(MAKECMDGOALS)),)
 .NOTPARALLEL:

nptl/tst-mutex8.c

@@ -127,9 +127,8 @@ check_type (const char *mas, pthread_mutexattr_t *ma)
       return 1;
     }
 
-  /* Elided mutexes don't fail destroy. If elision is not explicitly disabled
-     we don't know, so can also not check this.  */
-#ifndef ENABLE_LOCK_ELISION
+  /* Elided mutexes don't fail destroy, but this test is run with
+     elision disabled so we can test them.  */
   e = pthread_mutex_destroy (m);
   if (e == 0)
     {
@@ -142,7 +141,6 @@ check_type (const char *mas, pthread_mutexattr_t *ma)
 	      mas);
       return 1;
     }
-#endif
 
   if (pthread_mutex_unlock (m) != 0)
     {
@@ -157,7 +155,6 @@ check_type (const char *mas, pthread_mutexattr_t *ma)
     }
 
   /* Elided mutexes don't fail destroy.  */
-#ifndef ENABLE_LOCK_ELISION
   e = pthread_mutex_destroy (m);
   if (e == 0)
     {
@@ -171,7 +168,6 @@ mutex_destroy of self-trylocked mutex did not return EBUSY %s\n",
 	      mas);
       return 1;
     }
-#endif
 
   if (pthread_mutex_unlock (m) != 0)
     {
@@ -207,7 +203,6 @@ mutex_destroy of self-trylocked mutex did not return EBUSY %s\n",
     }
 
   /* Elided mutexes don't fail destroy.  */
-#ifndef ENABLE_LOCK_ELISION
   e = pthread_mutex_destroy (m);
   if (e == 0)
     {
@@ -220,7 +215,6 @@ mutex_destroy of self-trylocked mutex did not return EBUSY %s\n",
 mutex_destroy of condvar-used mutex did not return EBUSY for %s\n", mas);
       return 1;
     }
-#endif
 
   done = true;
   if (pthread_cond_signal (&c) != 0)
@@ -280,7 +274,6 @@ mutex_destroy of condvar-used mutex did not return EBUSY for %s\n", mas);
     }
 
   /* Elided mutexes don't fail destroy.  */
-#ifndef ENABLE_LOCK_ELISION
   e = pthread_mutex_destroy (m);
   if (e == 0)
     {
@@ -295,7 +288,6 @@ mutex_destroy of condvar-used mutex did not return EBUSY for %s\n", mas);
 	      mas);
       return 1;
     }
-#endif
 
   if (pthread_cancel (th) != 0)
     {

sysdeps/powerpc/nptl/elide.h

@@ -19,7 +19,6 @@
 #ifndef ELIDE_PPC_H
 # define ELIDE_PPC_H
 
-#ifdef ENABLE_LOCK_ELISION
 # include <htm.h>
 # include <elision-conf.h>
 
@@ -114,12 +113,4 @@ __elide_unlock (int is_lock_free)
 # define ELIDE_UNLOCK(is_lock_free) \
   __elide_unlock (is_lock_free)
 
-# else
-
-# define ELIDE_LOCK(adapt_count, is_lock_free) 0
-# define ELIDE_TRYLOCK(adapt_count, is_lock_free, write) 0
-# define ELIDE_UNLOCK(is_lock_free) 0
-
-#endif /* ENABLE_LOCK_ELISION  */
-
 #endif

sysdeps/powerpc/powerpc32/sysdep.h

@@ -90,7 +90,7 @@ GOT_LABEL:			;					      \
   cfi_endproc;								      \
   ASM_SIZE_DIRECTIVE(name)
 
-#if ! IS_IN(rtld) && defined (ENABLE_LOCK_ELISION)
+#if ! IS_IN(rtld)
 # define ABORT_TRANSACTION \
     cmpwi    2,0;		\
     beq      1f;		\

sysdeps/powerpc/powerpc64/sysdep.h

@@ -263,7 +263,7 @@ LT_LABELSUFFIX(name,_name_end): ; \
   TRACEBACK_MASK(name,mask);	\
   END_2(name)
 
-#if !IS_IN(rtld) && defined (ENABLE_LOCK_ELISION)
+#if !IS_IN(rtld)
 # define ABORT_TRANSACTION \
     cmpdi    13,0;		\
     beq      1f;		\