Remove configure --disable-float4-byval and --disable-float8-byval

From e009755a160d3d67900c80c9bc17276e27f79baa Mon Sep 17 00:00:00 2001
From: Peter Eisentraut <peter@eisentraut.org>
Date: Thu, 31 Oct 2019 09:21:38 +0100
Subject: [PATCH 1/3] Remove configure --disable-float4-byval and
 --disable-float8-byval

These build options were only useful to maintain compatibility for version-0
functions, but those are no longer supported, so these options can be
removed.

float4 is now always pass-by-value; the pass-by-reference
code path is completely removed.

float8 and related types are now hardcoded to pass-by-value or
pass-by-reference depending on whether the build is 64- or 32-bit, as
was previously also the default.
---
 configure                               | 118 ------------------------
 configure.in                            |  33 -------
 doc/src/sgml/func.sgml                  |  10 --
 doc/src/sgml/installation.sgml          |  28 ------
 src/backend/access/index/indexam.c      |   5 -
 src/backend/access/transam/xlog.c       |  35 -------
 src/backend/bootstrap/bootstrap.c       |   2 +-
 src/backend/catalog/genbki.pl           |   2 +-
 src/backend/commands/analyze.c          |   2 +-
 src/backend/utils/adt/numeric.c         |   5 +-
 src/backend/utils/fmgr/dfmgr.c          |  18 ----
 src/backend/utils/fmgr/fmgr.c           |  23 +----
 src/backend/utils/misc/pg_controldata.c |  18 +---
 src/bin/initdb/initdb.c                 |   3 -
 src/bin/pg_controldata/pg_controldata.c |   4 -
 src/bin/pg_resetwal/pg_resetwal.c       |   6 --
 src/bin/pg_upgrade/controldata.c        |   7 ++
 src/include/c.h                         |  15 +++
 src/include/catalog/catversion.h        |   2 +-
 src/include/catalog/pg_control.h        |   6 +-
 src/include/catalog/pg_proc.dat         |   6 +-
 src/include/catalog/pg_type.dat         |   2 +-
 src/include/fmgr.h                      |   4 -
 src/include/pg_config.h.in              |  15 ---
 src/include/postgres.h                  |  23 +----
 src/tools/msvc/Solution.pm              |  25 -----
 src/tools/msvc/config_default.pl        |   5 -
 27 files changed, 40 insertions(+), 382 deletions(-)

diff --git a/configure b/configure
index 6b1c779ee3..b38ad1526d 100755
--- a/configure
+++ b/configure
@@ -866,8 +866,6 @@ with_system_tzdata
 with_zlib
 with_gnu_ld
 enable_largefile
-enable_float4_byval
-enable_float8_byval
 '
       ac_precious_vars='build_alias
 host_alias
@@ -1525,8 +1523,6 @@ Optional Features:
   --enable-cassert        enable assertion checks (for debugging)
   --disable-thread-safety disable thread-safety in client libraries
   --disable-largefile     omit support for large files
-  --disable-float4-byval  disable float4 passed by value
-  --disable-float8-byval  disable float8 passed by value
 
 Optional Packages:
   --with-PACKAGE[=ARG]    use PACKAGE [ARG=yes]
@@ -16801,120 +16797,6 @@ _ACEOF
 
 
 
-# Decide whether float4 is passed by value: user-selectable, enabled by default
-{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether to build with float4 passed by value" >&5
-$as_echo_n "checking whether to build with float4 passed by value... " >&6; }
-
-
-# Check whether --enable-float4-byval was given.
-if test "${enable_float4_byval+set}" = set; then :
-  enableval=$enable_float4_byval;
-  case $enableval in
-    yes)
-
-$as_echo "#define USE_FLOAT4_BYVAL 1" >>confdefs.h
-
-               float4passbyval=true
-      ;;
-    no)
-      float4passbyval=false
-      ;;
-    *)
-      as_fn_error $? "no argument expected for --enable-float4-byval option" "$LINENO" 5
-      ;;
-  esac
-
-else
-  enable_float4_byval=yes
-
-$as_echo "#define USE_FLOAT4_BYVAL 1" >>confdefs.h
-
-               float4passbyval=true
-fi
-
-
-{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $enable_float4_byval" >&5
-$as_echo "$enable_float4_byval" >&6; }
-
-cat >>confdefs.h <<_ACEOF
-#define FLOAT4PASSBYVAL $float4passbyval
-_ACEOF
-
-
-# Decide whether float8 is passed by value.
-# Note: this setting also controls int8 and related types such as timestamp.
-# If sizeof(Datum) >= 8, this is user-selectable, enabled by default.
-# If not, trying to select it is an error.
-{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether to build with float8 passed by value" >&5
-$as_echo_n "checking whether to build with float8 passed by value... " >&6; }
-if test $ac_cv_sizeof_void_p -ge 8 ; then
-
-
-# Check whether --enable-float8-byval was given.
-if test "${enable_float8_byval+set}" = set; then :
-  enableval=$enable_float8_byval;
-  case $enableval in
-    yes)
-      :
-      ;;
-    no)
-      :
-      ;;
-    *)
-      as_fn_error $? "no argument expected for --enable-float8-byval option" "$LINENO" 5
-      ;;
-  esac
-
-else
-  enable_float8_byval=yes
-
-fi
-
-
-else
-
-
-# Check whether --enable-float8-byval was given.
-if test "${enable_float8_byval+set}" = set; then :
-  enableval=$enable_float8_byval;
-  case $enableval in
-    yes)
-      :
-      ;;
-    no)
-      :
-      ;;
-    *)
-      as_fn_error $? "no argument expected for --enable-float8-byval option" "$LINENO" 5
-      ;;
-  esac
-
-else
-  enable_float8_byval=no
-
-fi
-
-
-  if test "$enable_float8_byval" = yes ; then
-    as_fn_error $? "--enable-float8-byval is not supported on 32-bit platforms." "$LINENO" 5
-  fi
-fi
-if test "$enable_float8_byval" = yes ; then
-
-$as_echo "#define USE_FLOAT8_BYVAL 1" >>confdefs.h
-
-  float8passbyval=true
-else
-  float8passbyval=false
-fi
-{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $enable_float8_byval" >&5
-$as_echo "$enable_float8_byval" >&6; }
-
-cat >>confdefs.h <<_ACEOF
-#define FLOAT8PASSBYVAL $float8passbyval
-_ACEOF
-
-
 # Determine memory alignment requirements for the basic C data types.
 
 # The cast to long int works around a bug in the HP C Compiler,
diff --git a/configure.in b/configure.in
index 2b9025cac3..73d55ae48d 100644
--- a/configure.in
+++ b/configure.in
@@ -1931,39 +1931,6 @@ AC_CHECK_SIZEOF([void *])
 AC_CHECK_SIZEOF([size_t])
 AC_CHECK_SIZEOF([long])
 
-# Decide whether float4 is passed by value: user-selectable, enabled by default
-AC_MSG_CHECKING([whether to build with float4 passed by value])
-PGAC_ARG_BOOL(enable, float4-byval, yes, [disable float4 passed by value],
-              [AC_DEFINE([USE_FLOAT4_BYVAL], 1,
-                         [Define to 1 if you want float4 values to be passed by value. (--enable-float4-byval)])
-               float4passbyval=true],
-              [float4passbyval=false])
-AC_MSG_RESULT([$enable_float4_byval])
-AC_DEFINE_UNQUOTED([FLOAT4PASSBYVAL], [$float4passbyval], [float4 values are passed by value if 'true', by reference if 'false'])
-
-# Decide whether float8 is passed by value.
-# Note: this setting also controls int8 and related types such as timestamp.
-# If sizeof(Datum) >= 8, this is user-selectable, enabled by default.
-# If not, trying to select it is an error.
-AC_MSG_CHECKING([whether to build with float8 passed by value])
-if test $ac_cv_sizeof_void_p -ge 8 ; then
-  PGAC_ARG_BOOL(enable, float8-byval, yes, [disable float8 passed by value])
-else
-  PGAC_ARG_BOOL(enable, float8-byval, no, [disable float8 passed by value])
-  if test "$enable_float8_byval" = yes ; then
-    AC_MSG_ERROR([--enable-float8-byval is not supported on 32-bit platforms.])
-  fi
-fi
-if test "$enable_float8_byval" = yes ; then
-  AC_DEFINE([USE_FLOAT8_BYVAL], 1,
-            [Define to 1 if you want float8, int8, etc values to be passed by value. (--enable-float8-byval)])
-  float8passbyval=true
-else
-  float8passbyval=false
-fi
-AC_MSG_RESULT([$enable_float8_byval])
-AC_DEFINE_UNQUOTED([FLOAT8PASSBYVAL], [$float8passbyval], [float8, int8, and related values are passed by value if 'true', by reference if 'false'])
-
 # Determine memory alignment requirements for the basic C data types.
 
 AC_CHECK_ALIGNOF(short)
diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 28eb322f3f..e4ae12f874 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -19964,16 +19964,6 @@ <title><function>pg_control_init</function> Columns</title>
        <entry><type>integer</type></entry>
       </row>
 
-      <row>
-       <entry><literal>float4_pass_by_value</literal></entry>
-       <entry><type>boolean</type></entry>
-      </row>
-
-      <row>
-       <entry><literal>float8_pass_by_value</literal></entry>
-       <entry><type>boolean</type></entry>
-      </row>
-
       <row>
        <entry><literal>data_page_checksum_version</literal></entry>
        <entry><type>integer</type></entry>
diff --git a/doc/src/sgml/installation.sgml b/doc/src/sgml/installation.sgml
index d8494e293b..9c10a897f1 100644
--- a/doc/src/sgml/installation.sgml
+++ b/doc/src/sgml/installation.sgml
@@ -1204,34 +1204,6 @@ <title>Anti-Features</title>
        </listitem>
       </varlistentry>
 
-      <varlistentry>
-       <term><option>--disable-float4-byval</option></term>
-       <listitem>
-        <para>
-         Disable passing float4 values <quote>by value</quote>, causing them
-         to be passed <quote>by reference</quote> instead.  This option costs
-         performance, but may be needed for compatibility with very old
-         user-defined functions written in C.
-        </para>
-       </listitem>
-      </varlistentry>
-
-      <varlistentry>
-       <term><option>--disable-float8-byval</option></term>
-       <listitem>
-        <para>
-         Disable passing float8 values <quote>by value</quote>, causing them
-         to be passed <quote>by reference</quote> instead.  This option costs
-         performance, but may be needed for compatibility with very old
-         user-defined functions written in C.
-         Note that this option affects not only float8, but also int8 and some
-         related types such as timestamp.
-         On 32-bit platforms, <option>--disable-float8-byval</option> is the default
-         and it is not allowed to select <option>--enable-float8-byval</option>.
-        </para>
-       </listitem>
-      </varlistentry>
-
       <varlistentry>
        <term><option>--disable-spinlocks</option></term>
        <listitem>
diff --git a/src/backend/access/index/indexam.c b/src/backend/access/index/indexam.c
index 9dfa0ddfbb..4af418287d 100644
--- a/src/backend/access/index/indexam.c
+++ b/src/backend/access/index/indexam.c
@@ -879,11 +879,6 @@ index_store_float8_orderby_distances(IndexScanDesc scan, Oid *orderByTypes,
 		else if (orderByTypes[i] == FLOAT4OID)
 		{
 			/* convert distance function's result to ORDER BY type */
-#ifndef USE_FLOAT4_BYVAL
-			/* must free any old value to avoid memory leakage */
-			if (!scan->xs_orderbynulls[i])
-				pfree(DatumGetPointer(scan->xs_orderbyvals[i]));
-#endif
 			if (distances && !distances[i].isnull)
 			{
 				scan->xs_orderbyvals[i] = Float4GetDatum((float4) distances[i].value);
diff --git a/src/backend/access/transam/xlog.c b/src/backend/access/transam/xlog.c
index 2e3cc51006..27f2c6bbe2 100644
--- a/src/backend/access/transam/xlog.c
+++ b/src/backend/access/transam/xlog.c
@@ -4516,9 +4516,6 @@ WriteControlFile(void)
 	ControlFile->toast_max_chunk_size = TOAST_MAX_CHUNK_SIZE;
 	ControlFile->loblksize = LOBLKSIZE;
 
-	ControlFile->float4ByVal = FLOAT4PASSBYVAL;
-	ControlFile->float8ByVal = FLOAT8PASSBYVAL;
-
 	/* Contents are protected with a CRC */
 	INIT_CRC32C(ControlFile->crc);
 	COMP_CRC32C(ControlFile->crc,
@@ -4720,38 +4717,6 @@ ReadControlFile(void)
 						   ControlFile->loblksize, (int) LOBLKSIZE),
 				 errhint("It looks like you need to recompile or initdb.")));
 
-#ifdef USE_FLOAT4_BYVAL
-	if (ControlFile->float4ByVal != true)
-		ereport(FATAL,
-				(errmsg("database files are incompatible with server"),
-				 errdetail("The database cluster was initialized without USE_FLOAT4_BYVAL"
-						   " but the server was compiled with USE_FLOAT4_BYVAL."),
-				 errhint("It looks like you need to recompile or initdb.")));
-#else
-	if (ControlFile->float4ByVal != false)
-		ereport(FATAL,
-				(errmsg("database files are incompatible with server"),
-				 errdetail("The database cluster was initialized with USE_FLOAT4_BYVAL"
-						   " but the server was compiled without USE_FLOAT4_BYVAL."),
-				 errhint("It looks like you need to recompile or initdb.")));
-#endif
-
-#ifdef USE_FLOAT8_BYVAL
-	if (ControlFile->float8ByVal != true)
-		ereport(FATAL,
-				(errmsg("database files are incompatible with server"),
-				 errdetail("The database cluster was initialized without USE_FLOAT8_BYVAL"
-						   " but the server was compiled with USE_FLOAT8_BYVAL."),
-				 errhint("It looks like you need to recompile or initdb.")));
-#else
-	if (ControlFile->float8ByVal != false)
-		ereport(FATAL,
-				(errmsg("database files are incompatible with server"),
-				 errdetail("The database cluster was initialized with USE_FLOAT8_BYVAL"
-						   " but the server was compiled without USE_FLOAT8_BYVAL."),
-				 errhint("It looks like you need to recompile or initdb.")));
-#endif
-
 	wal_segment_size = ControlFile->xlog_seg_size;
 
 	if (!IsValidWalSegSize(wal_segment_size))
diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c
index 9238fbe98d..8ea033610d 100644
--- a/src/backend/bootstrap/bootstrap.c
+++ b/src/backend/bootstrap/bootstrap.c
@@ -111,7 +111,7 @@ static const struct typinfo TypInfo[] = {
 	F_INT2IN, F_INT2OUT},
 	{"int4", INT4OID, 0, 4, true, 'i', 'p', InvalidOid,
 	F_INT4IN, F_INT4OUT},
-	{"float4", FLOAT4OID, 0, 4, FLOAT4PASSBYVAL, 'i', 'p', InvalidOid,
+	{"float4", FLOAT4OID, 0, 4, true, 'i', 'p', InvalidOid,
 	F_FLOAT4IN, F_FLOAT4OUT},
 	{"name", NAMEOID, CHAROID, NAMEDATALEN, false, 'c', 'p', C_COLLATION_OID,
 	F_NAMEIN, F_NAMEOUT},
diff --git a/src/backend/catalog/genbki.pl b/src/backend/catalog/genbki.pl
index 201d12d358..6eff045bd9 100644
--- a/src/backend/catalog/genbki.pl
+++ b/src/backend/catalog/genbki.pl
@@ -890,7 +890,7 @@ sub morph_row_for_schemapg
 		}
 
 		# Expand booleans from 'f'/'t' to 'false'/'true'.
-		# Some values might be other macros (eg FLOAT4PASSBYVAL),
+		# Some values might be other macros (eg FLOAT8PASSBYVAL),
 		# don't change.
 		elsif ($atttype eq 'bool')
 		{
diff --git a/src/backend/commands/analyze.c b/src/backend/commands/analyze.c
index 7accb950eb..71372ceb16 100644
--- a/src/backend/commands/analyze.c
+++ b/src/backend/commands/analyze.c
@@ -1496,7 +1496,7 @@ update_attstats(Oid relid, bool inh, int natts, VacAttrStats **vacattrstats)
 				/* XXX knows more than it should about type float4: */
 				arry = construct_array(numdatums, nnum,
 									   FLOAT4OID,
-									   sizeof(float4), FLOAT4PASSBYVAL, 'i');
+									   sizeof(float4), true, 'i');
 				values[i++] = PointerGetDatum(arry);	/* stanumbersN */
 			}
 			else
diff --git a/src/backend/utils/adt/numeric.c b/src/backend/utils/adt/numeric.c
index a00db3ce7a..21cc9a8109 100644
--- a/src/backend/utils/adt/numeric.c
+++ b/src/backend/utils/adt/numeric.c
@@ -351,8 +351,9 @@ typedef struct NumericSumAccum
 /*
  * We define our own macros for packing and unpacking abbreviated-key
  * representations for numeric values in order to avoid depending on
- * USE_FLOAT8_BYVAL.  The type of abbreviation we use is based only on
- * the size of a datum, not the argument-passing convention for float8.
+ * USE_FLOAT8_BYVAL, which was previously build-time configurable.  The type
+ * of abbreviation we use is based only on the size of a datum, not the
+ * argument-passing convention for float8.
  */
 #define NUMERIC_ABBREV_BITS (SIZEOF_DATUM * BITS_PER_BYTE)
 #if SIZEOF_DATUM == 8
diff --git a/src/backend/utils/fmgr/dfmgr.c b/src/backend/utils/fmgr/dfmgr.c
index be684786d6..589371942f 100644
--- a/src/backend/utils/fmgr/dfmgr.c
+++ b/src/backend/utils/fmgr/dfmgr.c
@@ -365,24 +365,6 @@ incompatible_module_error(const char *libname,
 						 magic_data.namedatalen,
 						 module_magic_data->namedatalen);
 	}
-	if (module_magic_data->float4byval != magic_data.float4byval)
-	{
-		if (details.len)
-			appendStringInfoChar(&details, '\n');
-		appendStringInfo(&details,
-						 _("Server has FLOAT4PASSBYVAL = %s, library has %s."),
-						 magic_data.float4byval ? "true" : "false",
-						 module_magic_data->float4byval ? "true" : "false");
-	}
-	if (module_magic_data->float8byval != magic_data.float8byval)
-	{
-		if (details.len)
-			appendStringInfoChar(&details, '\n');
-		appendStringInfo(&details,
-						 _("Server has FLOAT8PASSBYVAL = %s, library has %s."),
-						 magic_data.float8byval ? "true" : "false",
-						 module_magic_data->float8byval ? "true" : "false");
-	}
 
 	if (details.len == 0)
 		appendStringInfoString(&details,
diff --git a/src/backend/utils/fmgr/fmgr.c b/src/backend/utils/fmgr/fmgr.c
index 099ebd779b..c2eaa6178d 100644
--- a/src/backend/utils/fmgr/fmgr.c
+++ b/src/backend/utils/fmgr/fmgr.c
@@ -1683,13 +1683,7 @@ OidSendFunctionCall(Oid functionId, Datum val)
 /*-------------------------------------------------------------------------
  *		Support routines for standard maybe-pass-by-reference datatypes
  *
- * int8, float4, and float8 can be passed by value if Datum is wide enough.
- * (For backwards-compatibility reasons, we allow pass-by-ref to be chosen
- * at compile time even if pass-by-val is possible.)
- *
- * Note: there is only one switch controlling the pass-by-value option for
- * both int8 and float8; this is to avoid making things unduly complicated
- * for the timestamp types, which might have either representation.
+ * int8 and float8 can be passed by value if Datum is wide enough.
  *-------------------------------------------------------------------------
  */
 
@@ -1703,21 +1697,6 @@ Int64GetDatum(int64 X)
 	*retval = X;
 	return PointerGetDatum(retval);
 }
-#endif							/* USE_FLOAT8_BYVAL */
-
-#ifndef USE_FLOAT4_BYVAL
-
-Datum
-Float4GetDatum(float4 X)
-{
-	float4	   *retval = (float4 *) palloc(sizeof(float4));
-
-	*retval = X;
-	return PointerGetDatum(retval);
-}
-#endif
-
-#ifndef USE_FLOAT8_BYVAL
 
 Datum
 Float8GetDatum(float8 X)
diff --git a/src/backend/utils/misc/pg_controldata.c b/src/backend/utils/misc/pg_controldata.c
index b42921800b..330c210407 100644
--- a/src/backend/utils/misc/pg_controldata.c
+++ b/src/backend/utils/misc/pg_controldata.c
@@ -264,8 +264,8 @@ pg_control_recovery(PG_FUNCTION_ARGS)
 Datum
 pg_control_init(PG_FUNCTION_ARGS)
 {
-	Datum		values[12];
-	bool		nulls[12];
+	Datum		values[10];
+	bool		nulls[10];
 	TupleDesc	tupdesc;
 	HeapTuple	htup;
 	ControlFileData *ControlFile;
@@ -294,11 +294,7 @@ pg_control_init(PG_FUNCTION_ARGS)
 					   INT4OID, -1, 0);
 	TupleDescInitEntry(tupdesc, (AttrNumber) 9, "large_object_chunk_size",
 					   INT4OID, -1, 0);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 10, "float4_pass_by_value",
-					   BOOLOID, -1, 0);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 11, "float8_pass_by_value",
-					   BOOLOID, -1, 0);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 12, "data_page_checksum_version",
+	TupleDescInitEntry(tupdesc, (AttrNumber) 10, "data_page_checksum_version",
 					   INT4OID, -1, 0);
 	tupdesc = BlessTupleDesc(tupdesc);
 
@@ -335,15 +331,9 @@ pg_control_init(PG_FUNCTION_ARGS)
 	values[8] = Int32GetDatum(ControlFile->loblksize);
 	nulls[8] = false;
 
-	values[9] = BoolGetDatum(ControlFile->float4ByVal);
+	values[9] = Int32GetDatum(ControlFile->data_checksum_version);
 	nulls[9] = false;
 
-	values[10] = BoolGetDatum(ControlFile->float8ByVal);
-	nulls[10] = false;
-
-	values[11] = Int32GetDatum(ControlFile->data_checksum_version);
-	nulls[11] = false;
-
 	htup = heap_form_tuple(tupdesc, values, nulls);
 
 	PG_RETURN_DATUM(HeapTupleGetDatum(htup));
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 88a261d9bd..1f6d8939be 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -1399,9 +1399,6 @@ bootstrap_template1(void)
 	bki_lines = replace_token(bki_lines, "ALIGNOF_POINTER",
 							  (sizeof(Pointer) == 4) ? "i" : "d");
 
-	bki_lines = replace_token(bki_lines, "FLOAT4PASSBYVAL",
-							  FLOAT4PASSBYVAL ? "true" : "false");
-
 	bki_lines = replace_token(bki_lines, "FLOAT8PASSBYVAL",
 							  FLOAT8PASSBYVAL ? "true" : "false");
 
diff --git a/src/bin/pg_controldata/pg_controldata.c b/src/bin/pg_controldata/pg_controldata.c
index b14767f8b6..a2611c253c 100644
--- a/src/bin/pg_controldata/pg_controldata.c
+++ b/src/bin/pg_controldata/pg_controldata.c
@@ -327,10 +327,6 @@ main(int argc, char *argv[])
 	/* This is no longer configurable, but users may still expect to see it: */
 	printf(_("Date/time type storage:               %s\n"),
 		   _("64-bit integers"));
-	printf(_("Float4 argument passing:              %s\n"),
-		   (ControlFile->float4ByVal ? _("by value") : _("by reference")));
-	printf(_("Float8 argument passing:              %s\n"),
-		   (ControlFile->float8ByVal ? _("by value") : _("by reference")));
 	printf(_("Data page checksum version:           %u\n"),
 		   ControlFile->data_checksum_version);
 	printf(_("Mock authentication nonce:            %s\n"),
diff --git a/src/bin/pg_resetwal/pg_resetwal.c b/src/bin/pg_resetwal/pg_resetwal.c
index c4ee0168a9..201fcde2cd 100644
--- a/src/bin/pg_resetwal/pg_resetwal.c
+++ b/src/bin/pg_resetwal/pg_resetwal.c
@@ -722,8 +722,6 @@ GuessControlValues(void)
 	ControlFile.indexMaxKeys = INDEX_MAX_KEYS;
 	ControlFile.toast_max_chunk_size = TOAST_MAX_CHUNK_SIZE;
 	ControlFile.loblksize = LOBLKSIZE;
-	ControlFile.float4ByVal = FLOAT4PASSBYVAL;
-	ControlFile.float8ByVal = FLOAT8PASSBYVAL;
 
 	/*
 	 * XXX eventually, should try to grovel through old XLOG to develop more
@@ -801,10 +799,6 @@ PrintControlValues(bool guessed)
 	/* This is no longer configurable, but users may still expect to see it: */
 	printf(_("Date/time type storage:               %s\n"),
 		   _("64-bit integers"));
-	printf(_("Float4 argument passing:              %s\n"),
-		   (ControlFile.float4ByVal ? _("by value") : _("by reference")));
-	printf(_("Float8 argument passing:              %s\n"),
-		   (ControlFile.float8ByVal ? _("by value") : _("by reference")));
 	printf(_("Data page checksum version:           %u\n"),
 		   ControlFile.data_checksum_version);
 }
diff --git a/src/bin/pg_upgrade/controldata.c b/src/bin/pg_upgrade/controldata.c
index 93f3c34b74..85e22283a6 100644
--- a/src/bin/pg_upgrade/controldata.c
+++ b/src/bin/pg_upgrade/controldata.c
@@ -202,6 +202,13 @@ get_control_data(ClusterInfo *cluster, bool live_check)
 		got_data_checksum_version = true;
 	}
 
+	/* This is no longer configurable. */
+	if (GET_MAJOR_VERSION(cluster->major_version) >= 1300)
+	{
+		cluster->controldata.float8_pass_by_value = FLOAT8PASSBYVAL;
+		got_float8_pass_by_value = true;
+	}
+
 	/* we have the result of cmd in "output". so parse it line by line now */
 	while (fgets(bufin, sizeof(bufin), output))
 	{
diff --git a/src/include/c.h b/src/include/c.h
index d752cc07dc..ab9cb73a4c 100644
--- a/src/include/c.h
+++ b/src/include/c.h
@@ -490,6 +490,21 @@ typedef signed int Offset;
 typedef float float4;
 typedef double float8;
 
+/*
+ * Decide whether 8-byte types are passed by value.  This controls the titular
+ * float8 but also int8 and related types such as timestamp.
+ *
+ * This is defined here instead of, say, postgres.h, so that initdb.c can see
+ * it.
+ */
+#if SIZEOF_VOID_P >= 8
+#define FLOAT8PASSBYVAL true
+#define USE_FLOAT8_BYVAL 1
+#else
+#define FLOAT8PASSBYVAL false
+#undef USE_FLOAT8_BYVAL
+#endif
+
 /*
  * Oid, RegProcedure, TransactionId, SubTransactionId, MultiXactId,
  * CommandId
diff --git a/src/include/catalog/catversion.h b/src/include/catalog/catversion.h
index 1f6de76e9c..37c52ddd38 100644
--- a/src/include/catalog/catversion.h
+++ b/src/include/catalog/catversion.h
@@ -53,6 +53,6 @@
  */
 
 /*							yyyymmddN */
-#define CATALOG_VERSION_NO	201910251
+#define CATALOG_VERSION_NO	201910311
 
 #endif
diff --git a/src/include/catalog/pg_control.h b/src/include/catalog/pg_control.h
index ff98d9e91a..19208aefd8 100644
--- a/src/include/catalog/pg_control.h
+++ b/src/include/catalog/pg_control.h
@@ -22,7 +22,7 @@
 
 
 /* Version identifier for this pg_control format */
-#define PG_CONTROL_VERSION	1201
+#define PG_CONTROL_VERSION	1202
 
 /* Nonce key length, see below */
 #define MOCK_AUTH_NONCE_LEN		32
@@ -214,10 +214,6 @@ typedef struct ControlFileData
 	uint32		toast_max_chunk_size;	/* chunk size in TOAST tables */
 	uint32		loblksize;		/* chunk size in pg_largeobject */
 
-	/* flags indicating pass-by-value status of various types */
-	bool		float4ByVal;	/* float4 pass-by-value? */
-	bool		float8ByVal;	/* float8, int8, etc pass-by-value? */
-
 	/* Are data pages protected by checksums? Zero if no checksum version */
 	uint32		data_checksum_version;
 
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 58ea5b982b..4d4f119268 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10660,9 +10660,9 @@
   descr => 'pg_controldata init state information as a function',
   proname => 'pg_control_init', provolatile => 'v', prorettype => 'record',
   proargtypes => '',
-  proallargtypes => '{int4,int4,int4,int4,int4,int4,int4,int4,int4,bool,bool,int4}',
-  proargmodes => '{o,o,o,o,o,o,o,o,o,o,o,o}',
-  proargnames => '{max_data_alignment,database_block_size,blocks_per_segment,wal_block_size,bytes_per_wal_segment,max_identifier_length,max_index_columns,max_toast_chunk_size,large_object_chunk_size,float4_pass_by_value,float8_pass_by_value,data_page_checksum_version}',
+  proallargtypes => '{int4,int4,int4,int4,int4,int4,int4,int4,int4,int4}',
+  proargmodes => '{o,o,o,o,o,o,o,o,o,o}',
+  proargnames => '{max_data_alignment,database_block_size,blocks_per_segment,wal_block_size,bytes_per_wal_segment,max_identifier_length,max_index_columns,max_toast_chunk_size,large_object_chunk_size,data_page_checksum_version}',
   prosrc => 'pg_control_init' },
 
 # collation management functions
diff --git a/src/include/catalog/pg_type.dat b/src/include/catalog/pg_type.dat
index be49e00114..d9b35af914 100644
--- a/src/include/catalog/pg_type.dat
+++ b/src/include/catalog/pg_type.dat
@@ -215,7 +215,7 @@
 
 { oid => '700', array_type_oid => '1021',
   descr => 'single-precision floating point number, 4-byte storage',
-  typname => 'float4', typlen => '4', typbyval => 'FLOAT4PASSBYVAL',
+  typname => 'float4', typlen => '4', typbyval => 't',
   typcategory => 'N', typinput => 'float4in', typoutput => 'float4out',
   typreceive => 'float4recv', typsend => 'float4send', typalign => 'i' },
 { oid => '701', array_type_oid => '1022',
diff --git a/src/include/fmgr.h b/src/include/fmgr.h
index 29ae4674cc..838d43875a 100644
--- a/src/include/fmgr.h
+++ b/src/include/fmgr.h
@@ -446,8 +446,6 @@ typedef struct
 	int			funcmaxargs;	/* FUNC_MAX_ARGS */
 	int			indexmaxkeys;	/* INDEX_MAX_KEYS */
 	int			namedatalen;	/* NAMEDATALEN */
-	int			float4byval;	/* FLOAT4PASSBYVAL */
-	int			float8byval;	/* FLOAT8PASSBYVAL */
 } Pg_magic_struct;
 
 /* The actual data block contents */
@@ -458,8 +456,6 @@ typedef struct
 	FUNC_MAX_ARGS, \
 	INDEX_MAX_KEYS, \
 	NAMEDATALEN, \
-	FLOAT4PASSBYVAL, \
-	FLOAT8PASSBYVAL \
 }
 
 /*
diff --git a/src/include/pg_config.h.in b/src/include/pg_config.h.in
index 939245db39..97552d63d0 100644
--- a/src/include/pg_config.h.in
+++ b/src/include/pg_config.h.in
@@ -70,13 +70,6 @@
    MSVC and with C++ compilers. */
 #undef FLEXIBLE_ARRAY_MEMBER
 
-/* float4 values are passed by value if 'true', by reference if 'false' */
-#undef FLOAT4PASSBYVAL
-
-/* float8, int8, and related values are passed by value if 'true', by
-   reference if 'false' */
-#undef FLOAT8PASSBYVAL
-
 /* Define to 1 if gettimeofday() takes only 1 argument. */
 #undef GETTIMEOFDAY_1ARG
 
@@ -898,14 +891,6 @@
 /* Define to use /dev/urandom for random number generation */
 #undef USE_DEV_URANDOM
 
-/* Define to 1 if you want float4 values to be passed by value.
-   (--enable-float4-byval) */
-#undef USE_FLOAT4_BYVAL
-
-/* Define to 1 if you want float8, int8, etc values to be passed by value.
-   (--enable-float8-byval) */
-#undef USE_FLOAT8_BYVAL
-
 /* Define to build with ICU support. (--with-icu) */
 #undef USE_ICU
 
diff --git a/src/include/postgres.h b/src/include/postgres.h
index 057a3413ac..f5b7c52f8a 100644
--- a/src/include/postgres.h
+++ b/src/include/postgres.h
@@ -659,11 +659,7 @@ extern Datum Int64GetDatum(int64 X);
 /*
  * DatumGetFloat4
  *		Returns 4-byte floating point value of a datum.
- *
- * Note: this macro hides whether float4 is pass by value or by reference.
  */
-
-#ifdef USE_FLOAT4_BYVAL
 static inline float4
 DatumGetFloat4(Datum X)
 {
@@ -676,18 +672,11 @@ DatumGetFloat4(Datum X)
 	myunion.value = DatumGetInt32(X);
 	return myunion.retval;
 }
-#else
-#define DatumGetFloat4(X) (* ((float4 *) DatumGetPointer(X)))
-#endif
 
 /*
  * Float4GetDatum
  *		Returns datum representation for a 4-byte floating point number.
- *
- * Note: if float4 is pass by reference, this function returns a reference
- * to palloc'd space.
  */
-#ifdef USE_FLOAT4_BYVAL
 static inline Datum
 Float4GetDatum(float4 X)
 {
@@ -700,9 +689,6 @@ Float4GetDatum(float4 X)
 	myunion.value = X;
 	return Int32GetDatum(myunion.retval);
 }
-#else
-extern Datum Float4GetDatum(float4 X);
-#endif
 
 /*
  * DatumGetFloat8
@@ -757,10 +743,9 @@ extern Datum Float8GetDatum(float8 X);
 /*
  * Int64GetDatumFast
  * Float8GetDatumFast
- * Float4GetDatumFast
  *
  * These macros are intended to allow writing code that does not depend on
- * whether int64, float8, float4 are pass-by-reference types, while not
+ * whether int64 and float8 are pass-by-reference types, while not
  * sacrificing performance when they are.  The argument must be a variable
  * that will exist and have the same value for as long as the Datum is needed.
  * In the pass-by-ref case, the address of the variable is taken to use as
@@ -776,10 +761,4 @@ extern Datum Float8GetDatum(float8 X);
 #define Float8GetDatumFast(X) PointerGetDatum(&(X))
 #endif
 
-#ifdef USE_FLOAT4_BYVAL
-#define Float4GetDatumFast(X) Float4GetDatum(X)
-#else
-#define Float4GetDatumFast(X) PointerGetDatum(&(X))
-#endif
-
 #endif							/* POSTGRES_H */
diff --git a/src/tools/msvc/Solution.pm b/src/tools/msvc/Solution.pm
index a6958273ac..2423267b54 100644
--- a/src/tools/msvc/Solution.pm
+++ b/src/tools/msvc/Solution.pm
@@ -31,12 +31,6 @@ sub _new
 	$self->DeterminePlatform();
 	my $bits = $self->{platform} eq 'Win32' ? 32 : 64;
 
-	$options->{float4byval} = 1
-	  unless exists $options->{float4byval};
-	$options->{float8byval} = ($bits == 64)
-	  unless exists $options->{float8byval};
-	die "float8byval not permitted on 32 bit platforms"
-	  if $options->{float8byval} && $bits == 32;
 	if ($options->{xslt} && !$options->{xml})
 	{
 		die "XSLT requires XML\n";
@@ -209,25 +203,6 @@ sub GenerateFiles
 		print $o "#define XLOG_BLCKSZ ",
 		  1024 * $self->{options}->{wal_blocksize}, "\n";
 
-		if ($self->{options}->{float4byval})
-		{
-			print $o "#define USE_FLOAT4_BYVAL 1\n";
-			print $o "#define FLOAT4PASSBYVAL true\n";
-		}
-		else
-		{
-			print $o "#define FLOAT4PASSBYVAL false\n";
-		}
-		if ($self->{options}->{float8byval})
-		{
-			print $o "#define USE_FLOAT8_BYVAL 1\n";
-			print $o "#define FLOAT8PASSBYVAL true\n";
-		}
-		else
-		{
-			print $o "#define FLOAT8PASSBYVAL false\n";
-		}
-
 		if ($self->{options}->{uuid})
 		{
 			print $o "#define HAVE_UUID_OSSP\n";
diff --git a/src/tools/msvc/config_default.pl b/src/tools/msvc/config_default.pl
index 043df4c5e8..d5e010b18a 100644
--- a/src/tools/msvc/config_default.pl
+++ b/src/tools/msvc/config_default.pl
@@ -4,11 +4,6 @@
 
 our $config = {
 	asserts => 0,    # --enable-cassert
-	     # float4byval=>1,         # --disable-float4-byval, on by default
-
-	# float8byval=> $platformbits == 64, # --disable-float8-byval,
-	# off by default on 32 bit platforms, on by default on 64 bit platforms
-
 	# blocksize => 8,         # --with-blocksize, 8kB by default
 	# wal_blocksize => 8,     # --with-wal-blocksize, 8kB by default
 	ldap      => 1,        # --with-ldap

base-commit: 73025140885c889410b9bfc4a30a3866396fc5db
-- 
2.23.0

From 1c20d0217f0fb88a286a61532340b0c07d1a013a Mon Sep 17 00:00:00 2001
From: Peter Eisentraut <peter@eisentraut.org>
Date: Thu, 31 Oct 2019 09:21:38 +0100
Subject: [PATCH 2/3] Replace USE_FLOAT8_BYVAL by FLOAT8PASSBYVAL

It's enough to have one macro for the same thing.
---
 contrib/btree_gist/btree_time.c               |  2 +-
 contrib/btree_gist/btree_ts.c                 |  2 +-
 src/backend/access/heap/heapam_handler.c      |  2 +-
 src/backend/access/index/indexam.c            |  2 +-
 src/backend/utils/adt/int8.c                  |  4 ++--
 src/backend/utils/adt/numeric.c               |  6 +++---
 src/backend/utils/adt/rangetypes_typanalyze.c |  6 +-----
 src/backend/utils/fmgr/fmgr.c                 |  2 +-
 src/include/c.h                               |  2 --
 src/include/postgres.h                        | 14 +++++++-------
 10 files changed, 18 insertions(+), 24 deletions(-)

diff --git a/contrib/btree_gist/btree_time.c b/contrib/btree_gist/btree_time.c
index 90cf6554ea..884ae21ff6 100644
--- a/contrib/btree_gist/btree_time.c
+++ b/contrib/btree_gist/btree_time.c
@@ -30,7 +30,7 @@ PG_FUNCTION_INFO_V1(gbt_time_penalty);
 PG_FUNCTION_INFO_V1(gbt_time_same);
 
 
-#ifdef USE_FLOAT8_BYVAL
+#if FLOAT8PASSBYVAL
 #define TimeADTGetDatumFast(X) TimeADTGetDatum(X)
 #else
 #define TimeADTGetDatumFast(X) PointerGetDatum(&(X))
diff --git a/contrib/btree_gist/btree_ts.c b/contrib/btree_gist/btree_ts.c
index 49d1849d88..30864465d2 100644
--- a/contrib/btree_gist/btree_ts.c
+++ b/contrib/btree_gist/btree_ts.c
@@ -33,7 +33,7 @@ PG_FUNCTION_INFO_V1(gbt_ts_penalty);
 PG_FUNCTION_INFO_V1(gbt_ts_same);
 
 
-#ifdef USE_FLOAT8_BYVAL
+#if FLOAT8PASSBYVAL
 #define TimestampGetDatumFast(X) TimestampGetDatum(X)
 #else
 #define TimestampGetDatumFast(X) PointerGetDatum(&(X))
diff --git a/src/backend/access/heap/heapam_handler.c b/src/backend/access/heap/heapam_handler.c
index 2dd8821fac..1ae4258670 100644
--- a/src/backend/access/heap/heapam_handler.c
+++ b/src/backend/access/heap/heapam_handler.c
@@ -1858,7 +1858,7 @@ heapam_index_validate_scan(Relation heapRelation,
 				indexcursor = &decoded;
 
 				/* If int8 is pass-by-ref, free (encoded) TID Datum memory */
-#ifndef USE_FLOAT8_BYVAL
+#if !FLOAT8PASSBYVAL
 				pfree(DatumGetPointer(ts_val));
 #endif
 			}
diff --git a/src/backend/access/index/indexam.c b/src/backend/access/index/indexam.c
index 4af418287d..6a4a1f935c 100644
--- a/src/backend/access/index/indexam.c
+++ b/src/backend/access/index/indexam.c
@@ -860,7 +860,7 @@ index_store_float8_orderby_distances(IndexScanDesc scan, Oid *orderByTypes,
 	{
 		if (orderByTypes[i] == FLOAT8OID)
 		{
-#ifndef USE_FLOAT8_BYVAL
+#if !FLOAT8PASSBYVAL
 			/* must free any old value to avoid memory leakage */
 			if (!scan->xs_orderbynulls[i])
 				pfree(DatumGetPointer(scan->xs_orderbyvals[i]));
diff --git a/src/backend/utils/adt/int8.c b/src/backend/utils/adt/int8.c
index 0ff9394a2f..165be75da8 100644
--- a/src/backend/utils/adt/int8.c
+++ b/src/backend/utils/adt/int8.c
@@ -679,7 +679,7 @@ int8inc(PG_FUNCTION_ARGS)
 	 * (If int8 is pass-by-value, then of course this is useless as well as
 	 * incorrect, so just ifdef it out.)
 	 */
-#ifndef USE_FLOAT8_BYVAL		/* controls int8 too */
+#if !FLOAT8PASSBYVAL		/* controls int8 too */
 	if (AggCheckCallContext(fcinfo, NULL))
 	{
 		int64	   *arg = (int64 *) PG_GETARG_POINTER(0);
@@ -717,7 +717,7 @@ int8dec(PG_FUNCTION_ARGS)
 	 * (If int8 is pass-by-value, then of course this is useless as well as
 	 * incorrect, so just ifdef it out.)
 	 */
-#ifndef USE_FLOAT8_BYVAL		/* controls int8 too */
+#if !FLOAT8PASSBYVAL		/* controls int8 too */
 	if (AggCheckCallContext(fcinfo, NULL))
 	{
 		int64	   *arg = (int64 *) PG_GETARG_POINTER(0);
diff --git a/src/backend/utils/adt/numeric.c b/src/backend/utils/adt/numeric.c
index 21cc9a8109..318c6a5494 100644
--- a/src/backend/utils/adt/numeric.c
+++ b/src/backend/utils/adt/numeric.c
@@ -351,7 +351,7 @@ typedef struct NumericSumAccum
 /*
  * We define our own macros for packing and unpacking abbreviated-key
  * representations for numeric values in order to avoid depending on
- * USE_FLOAT8_BYVAL, which was previously build-time configurable.  The type
+ * FLOAT8PASSBYVAL, which was previously build-time configurable.  The type
  * of abbreviation we use is based only on the size of a datum, not the
  * argument-passing convention for float8.
  */
@@ -5301,7 +5301,7 @@ int2_sum(PG_FUNCTION_ARGS)
 	 * then of course this is useless as well as incorrect, so just ifdef it
 	 * out.)
 	 */
-#ifndef USE_FLOAT8_BYVAL		/* controls int8 too */
+#if !FLOAT8PASSBYVAL		/* controls int8 too */
 	if (AggCheckCallContext(fcinfo, NULL))
 	{
 		int64	   *oldsum = (int64 *) PG_GETARG_POINTER(0);
@@ -5350,7 +5350,7 @@ int4_sum(PG_FUNCTION_ARGS)
 	 * then of course this is useless as well as incorrect, so just ifdef it
 	 * out.)
 	 */
-#ifndef USE_FLOAT8_BYVAL		/* controls int8 too */
+#if !FLOAT8PASSBYVAL		/* controls int8 too */
 	if (AggCheckCallContext(fcinfo, NULL))
 	{
 		int64	   *oldsum = (int64 *) PG_GETARG_POINTER(0);
diff --git a/src/backend/utils/adt/rangetypes_typanalyze.c b/src/backend/utils/adt/rangetypes_typanalyze.c
index d01d3032cc..b438625951 100644
--- a/src/backend/utils/adt/rangetypes_typanalyze.c
+++ b/src/backend/utils/adt/rangetypes_typanalyze.c
@@ -325,11 +325,7 @@ compute_range_stats(VacAttrStats *stats, AnalyzeAttrFetchFunc fetchfunc,
 		stats->numvalues[slot_idx] = num_hist;
 		stats->statypid[slot_idx] = FLOAT8OID;
 		stats->statyplen[slot_idx] = sizeof(float8);
-#ifdef USE_FLOAT8_BYVAL
-		stats->statypbyval[slot_idx] = true;
-#else
-		stats->statypbyval[slot_idx] = false;
-#endif
+		stats->statypbyval[slot_idx] = FLOAT8PASSBYVAL;
 		stats->statypalign[slot_idx] = 'd';
 
 		/* Store the fraction of empty ranges */
diff --git a/src/backend/utils/fmgr/fmgr.c b/src/backend/utils/fmgr/fmgr.c
index c2eaa6178d..d73fe59a26 100644
--- a/src/backend/utils/fmgr/fmgr.c
+++ b/src/backend/utils/fmgr/fmgr.c
@@ -1687,7 +1687,7 @@ OidSendFunctionCall(Oid functionId, Datum val)
  *-------------------------------------------------------------------------
  */
 
-#ifndef USE_FLOAT8_BYVAL		/* controls int8 too */
+#if !FLOAT8PASSBYVAL		/* controls int8 too */
 
 Datum
 Int64GetDatum(int64 X)
diff --git a/src/include/c.h b/src/include/c.h
index ab9cb73a4c..26d50495e6 100644
--- a/src/include/c.h
+++ b/src/include/c.h
@@ -499,10 +499,8 @@ typedef double float8;
  */
 #if SIZEOF_VOID_P >= 8
 #define FLOAT8PASSBYVAL true
-#define USE_FLOAT8_BYVAL 1
 #else
 #define FLOAT8PASSBYVAL false
-#undef USE_FLOAT8_BYVAL
 #endif
 
 /*
diff --git a/src/include/postgres.h b/src/include/postgres.h
index f5b7c52f8a..bf4b6223f3 100644
--- a/src/include/postgres.h
+++ b/src/include/postgres.h
@@ -601,7 +601,7 @@ typedef struct NullableDatum
  * Note: this macro hides whether int64 is pass by value or by reference.
  */
 
-#ifdef USE_FLOAT8_BYVAL
+#if FLOAT8PASSBYVAL
 #define DatumGetInt64(X) ((int64) (X))
 #else
 #define DatumGetInt64(X) (* ((int64 *) DatumGetPointer(X)))
@@ -615,7 +615,7 @@ typedef struct NullableDatum
  * to palloc'd space.
  */
 
-#ifdef USE_FLOAT8_BYVAL
+#if FLOAT8PASSBYVAL
 #define Int64GetDatum(X) ((Datum) (X))
 #else
 extern Datum Int64GetDatum(int64 X);
@@ -628,7 +628,7 @@ extern Datum Int64GetDatum(int64 X);
  * Note: this macro hides whether int64 is pass by value or by reference.
  */
 
-#ifdef USE_FLOAT8_BYVAL
+#if FLOAT8PASSBYVAL
 #define DatumGetUInt64(X) ((uint64) (X))
 #else
 #define DatumGetUInt64(X) (* ((uint64 *) DatumGetPointer(X)))
@@ -642,7 +642,7 @@ extern Datum Int64GetDatum(int64 X);
  * to palloc'd space.
  */
 
-#ifdef USE_FLOAT8_BYVAL
+#if FLOAT8PASSBYVAL
 #define UInt64GetDatum(X) ((Datum) (X))
 #else
 #define UInt64GetDatum(X) Int64GetDatum((int64) (X))
@@ -697,7 +697,7 @@ Float4GetDatum(float4 X)
  * Note: this macro hides whether float8 is pass by value or by reference.
  */
 
-#ifdef USE_FLOAT8_BYVAL
+#if FLOAT8PASSBYVAL
 static inline float8
 DatumGetFloat8(Datum X)
 {
@@ -722,7 +722,7 @@ DatumGetFloat8(Datum X)
  * to palloc'd space.
  */
 
-#ifdef USE_FLOAT8_BYVAL
+#if FLOAT8PASSBYVAL
 static inline Datum
 Float8GetDatum(float8 X)
 {
@@ -753,7 +753,7 @@ extern Datum Float8GetDatum(float8 X);
  * macros.
  */
 
-#ifdef USE_FLOAT8_BYVAL
+#if FLOAT8PASSBYVAL
 #define Int64GetDatumFast(X)  Int64GetDatum(X)
 #define Float8GetDatumFast(X) Float8GetDatum(X)
 #else
-- 
2.23.0

From 9c38db227b7f2154b5a06c4ff42a43f8835b9929 Mon Sep 17 00:00:00 2001
From: Peter Eisentraut <peter@eisentraut.org>
Date: Thu, 31 Oct 2019 09:21:38 +0100
Subject: [PATCH 3/3] Convert some #if calls to compiler if()s

This ensures that the contained code is looked at by the compiler and
does not bit rot.  This is especially interesting as the affected code
is for the lesser-used 32-bit builds.
---
 src/backend/access/heap/heapam_handler.c |  5 ++---
 src/backend/access/index/indexam.c       | 11 ++++++-----
 src/backend/utils/adt/int8.c             | 12 ++++--------
 src/backend/utils/adt/numeric.c          | 14 ++++----------
 4 files changed, 16 insertions(+), 26 deletions(-)

diff --git a/src/backend/access/heap/heapam_handler.c b/src/backend/access/heap/heapam_handler.c
index 1ae4258670..a002916a73 100644
--- a/src/backend/access/heap/heapam_handler.c
+++ b/src/backend/access/heap/heapam_handler.c
@@ -1858,9 +1858,8 @@ heapam_index_validate_scan(Relation heapRelation,
 				indexcursor = &decoded;
 
 				/* If int8 is pass-by-ref, free (encoded) TID Datum memory */
-#if !FLOAT8PASSBYVAL
-				pfree(DatumGetPointer(ts_val));
-#endif
+				if (!FLOAT8PASSBYVAL)
+					pfree(DatumGetPointer(ts_val));
 			}
 			else
 			{
diff --git a/src/backend/access/index/indexam.c b/src/backend/access/index/indexam.c
index 6a4a1f935c..934c41c1be 100644
--- a/src/backend/access/index/indexam.c
+++ b/src/backend/access/index/indexam.c
@@ -860,11 +860,12 @@ index_store_float8_orderby_distances(IndexScanDesc scan, Oid *orderByTypes,
 	{
 		if (orderByTypes[i] == FLOAT8OID)
 		{
-#if !FLOAT8PASSBYVAL
-			/* must free any old value to avoid memory leakage */
-			if (!scan->xs_orderbynulls[i])
-				pfree(DatumGetPointer(scan->xs_orderbyvals[i]));
-#endif
+			if (!FLOAT8PASSBYVAL)
+			{
+				/* must free any old value to avoid memory leakage */
+				if (!scan->xs_orderbynulls[i])
+					pfree(DatumGetPointer(scan->xs_orderbyvals[i]));
+			}
 			if (distances && !distances[i].isnull)
 			{
 				scan->xs_orderbyvals[i] = Float8GetDatum(distances[i].value);
diff --git a/src/backend/utils/adt/int8.c b/src/backend/utils/adt/int8.c
index 165be75da8..3f25ff6b64 100644
--- a/src/backend/utils/adt/int8.c
+++ b/src/backend/utils/adt/int8.c
@@ -677,10 +677,9 @@ int8inc(PG_FUNCTION_ARGS)
 	 * palloc overhead for COUNT(): when called as an aggregate, we know that
 	 * the argument is modifiable local storage, so just update it in-place.
 	 * (If int8 is pass-by-value, then of course this is useless as well as
-	 * incorrect, so just ifdef it out.)
+	 * incorrect.)
 	 */
-#if !FLOAT8PASSBYVAL		/* controls int8 too */
-	if (AggCheckCallContext(fcinfo, NULL))
+	if (!FLOAT8PASSBYVAL && AggCheckCallContext(fcinfo, NULL))
 	{
 		int64	   *arg = (int64 *) PG_GETARG_POINTER(0);
 
@@ -692,7 +691,6 @@ int8inc(PG_FUNCTION_ARGS)
 		PG_RETURN_POINTER(arg);
 	}
 	else
-#endif
 	{
 		/* Not called as an aggregate, so just do it the dumb way */
 		int64		arg = PG_GETARG_INT64(0);
@@ -715,10 +713,9 @@ int8dec(PG_FUNCTION_ARGS)
 	 * palloc overhead for COUNT(): when called as an aggregate, we know that
 	 * the argument is modifiable local storage, so just update it in-place.
 	 * (If int8 is pass-by-value, then of course this is useless as well as
-	 * incorrect, so just ifdef it out.)
+	 * incorrect.)
 	 */
-#if !FLOAT8PASSBYVAL		/* controls int8 too */
-	if (AggCheckCallContext(fcinfo, NULL))
+	if (!FLOAT8PASSBYVAL && AggCheckCallContext(fcinfo, NULL))
 	{
 		int64	   *arg = (int64 *) PG_GETARG_POINTER(0);
 
@@ -729,7 +726,6 @@ int8dec(PG_FUNCTION_ARGS)
 		PG_RETURN_POINTER(arg);
 	}
 	else
-#endif
 	{
 		/* Not called as an aggregate, so just do it the dumb way */
 		int64		arg = PG_GETARG_INT64(0);
diff --git a/src/backend/utils/adt/numeric.c b/src/backend/utils/adt/numeric.c
index 318c6a5494..d6eae74a35 100644
--- a/src/backend/utils/adt/numeric.c
+++ b/src/backend/utils/adt/numeric.c
@@ -5298,11 +5298,9 @@ int2_sum(PG_FUNCTION_ARGS)
 	 * If we're invoked as an aggregate, we can cheat and modify our first
 	 * parameter in-place to avoid palloc overhead. If not, we need to return
 	 * the new value of the transition variable. (If int8 is pass-by-value,
-	 * then of course this is useless as well as incorrect, so just ifdef it
-	 * out.)
+	 * then of course this is useless as well as incorrect.)
 	 */
-#if !FLOAT8PASSBYVAL		/* controls int8 too */
-	if (AggCheckCallContext(fcinfo, NULL))
+	if (!FLOAT8PASSBYVAL && AggCheckCallContext(fcinfo, NULL))
 	{
 		int64	   *oldsum = (int64 *) PG_GETARG_POINTER(0);
 
@@ -5313,7 +5311,6 @@ int2_sum(PG_FUNCTION_ARGS)
 		PG_RETURN_POINTER(oldsum);
 	}
 	else
-#endif
 	{
 		int64		oldsum = PG_GETARG_INT64(0);
 
@@ -5347,11 +5344,9 @@ int4_sum(PG_FUNCTION_ARGS)
 	 * If we're invoked as an aggregate, we can cheat and modify our first
 	 * parameter in-place to avoid palloc overhead. If not, we need to return
 	 * the new value of the transition variable. (If int8 is pass-by-value,
-	 * then of course this is useless as well as incorrect, so just ifdef it
-	 * out.)
+	 * then of course this is useless as well as incorrect.)
 	 */
-#if !FLOAT8PASSBYVAL		/* controls int8 too */
-	if (AggCheckCallContext(fcinfo, NULL))
+	if (!FLOAT8PASSBYVAL && AggCheckCallContext(fcinfo, NULL))
 	{
 		int64	   *oldsum = (int64 *) PG_GETARG_POINTER(0);
 
@@ -5362,7 +5357,6 @@ int4_sum(PG_FUNCTION_ARGS)
 		PG_RETURN_POINTER(oldsum);
 	}
 	else
-#endif
 	{
 		int64		oldsum = PG_GETARG_INT64(0);
 
-- 
2.23.0