Allow dynamic allocation of shared memory segments.

Patch by myself and Amit Kapila.  Design help from Noah Misch.  Review
by Andres Freund.

configure

@@ -8384,6 +8384,180 @@ if test "$ac_res" != no; then
 
 fi
 
+{ $as_echo "$as_me:$LINENO: checking for library containing shm_open" >&5
+$as_echo_n "checking for library containing shm_open... " >&6; }
+if test "${ac_cv_search_shm_open+set}" = set; then
+  $as_echo_n "(cached) " >&6
+else
+  ac_func_search_save_LIBS=$LIBS
+cat >conftest.$ac_ext <<_ACEOF
+/* confdefs.h.  */
+_ACEOF
+cat confdefs.h >>conftest.$ac_ext
+cat >>conftest.$ac_ext <<_ACEOF
+/* end confdefs.h.  */
+
+/* Override any GCC internal prototype to avoid an error.
+   Use char because int might match the return type of a GCC
+   builtin and then its argument prototype would still apply.  */
+#ifdef __cplusplus
+extern "C"
+#endif
+char shm_open ();
+int
+main ()
+{
+return shm_open ();
+  ;
+  return 0;
+}
+_ACEOF
+for ac_lib in '' rt; do
+  if test -z "$ac_lib"; then
+    ac_res="none required"
+  else
+    ac_res=-l$ac_lib
+    LIBS="-l$ac_lib  $ac_func_search_save_LIBS"
+  fi
+  rm -f conftest.$ac_objext conftest$ac_exeext
+if { (ac_try="$ac_link"
+case "(($ac_try" in
+  *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;;
+  *) ac_try_echo=$ac_try;;
+esac
+eval ac_try_echo="\"\$as_me:$LINENO: $ac_try_echo\""
+$as_echo "$ac_try_echo") >&5
+  (eval "$ac_link") 2>conftest.er1
+  ac_status=$?
+  grep -v '^ *+' conftest.er1 >conftest.err
+  rm -f conftest.er1
+  cat conftest.err >&5
+  $as_echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); } && {
+	 test -z "$ac_c_werror_flag" ||
+	 test ! -s conftest.err
+       } && test -s conftest$ac_exeext && {
+	 test "$cross_compiling" = yes ||
+	 $as_test_x conftest$ac_exeext
+       }; then
+  ac_cv_search_shm_open=$ac_res
+else
+  $as_echo "$as_me: failed program was:" >&5
+sed 's/^/| /' conftest.$ac_ext >&5
+
+
+fi
+
+rm -rf conftest.dSYM
+rm -f core conftest.err conftest.$ac_objext conftest_ipa8_conftest.oo \
+      conftest$ac_exeext
+  if test "${ac_cv_search_shm_open+set}" = set; then
+  break
+fi
+done
+if test "${ac_cv_search_shm_open+set}" = set; then
+  :
+else
+  ac_cv_search_shm_open=no
+fi
+rm conftest.$ac_ext
+LIBS=$ac_func_search_save_LIBS
+fi
+{ $as_echo "$as_me:$LINENO: result: $ac_cv_search_shm_open" >&5
+$as_echo "$ac_cv_search_shm_open" >&6; }
+ac_res=$ac_cv_search_shm_open
+if test "$ac_res" != no; then
+  test "$ac_res" = "none required" || LIBS="$ac_res $LIBS"
+
+fi
+
+{ $as_echo "$as_me:$LINENO: checking for library containing shm_unlink" >&5
+$as_echo_n "checking for library containing shm_unlink... " >&6; }
+if test "${ac_cv_search_shm_unlink+set}" = set; then
+  $as_echo_n "(cached) " >&6
+else
+  ac_func_search_save_LIBS=$LIBS
+cat >conftest.$ac_ext <<_ACEOF
+/* confdefs.h.  */
+_ACEOF
+cat confdefs.h >>conftest.$ac_ext
+cat >>conftest.$ac_ext <<_ACEOF
+/* end confdefs.h.  */
+
+/* Override any GCC internal prototype to avoid an error.
+   Use char because int might match the return type of a GCC
+   builtin and then its argument prototype would still apply.  */
+#ifdef __cplusplus
+extern "C"
+#endif
+char shm_unlink ();
+int
+main ()
+{
+return shm_unlink ();
+  ;
+  return 0;
+}
+_ACEOF
+for ac_lib in '' rt; do
+  if test -z "$ac_lib"; then
+    ac_res="none required"
+  else
+    ac_res=-l$ac_lib
+    LIBS="-l$ac_lib  $ac_func_search_save_LIBS"
+  fi
+  rm -f conftest.$ac_objext conftest$ac_exeext
+if { (ac_try="$ac_link"
+case "(($ac_try" in
+  *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;;
+  *) ac_try_echo=$ac_try;;
+esac
+eval ac_try_echo="\"\$as_me:$LINENO: $ac_try_echo\""
+$as_echo "$ac_try_echo") >&5
+  (eval "$ac_link") 2>conftest.er1
+  ac_status=$?
+  grep -v '^ *+' conftest.er1 >conftest.err
+  rm -f conftest.er1
+  cat conftest.err >&5
+  $as_echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); } && {
+	 test -z "$ac_c_werror_flag" ||
+	 test ! -s conftest.err
+       } && test -s conftest$ac_exeext && {
+	 test "$cross_compiling" = yes ||
+	 $as_test_x conftest$ac_exeext
+       }; then
+  ac_cv_search_shm_unlink=$ac_res
+else
+  $as_echo "$as_me: failed program was:" >&5
+sed 's/^/| /' conftest.$ac_ext >&5
+
+
+fi
+
+rm -rf conftest.dSYM
+rm -f core conftest.err conftest.$ac_objext conftest_ipa8_conftest.oo \
+      conftest$ac_exeext
+  if test "${ac_cv_search_shm_unlink+set}" = set; then
+  break
+fi
+done
+if test "${ac_cv_search_shm_unlink+set}" = set; then
+  :
+else
+  ac_cv_search_shm_unlink=no
+fi
+rm conftest.$ac_ext
+LIBS=$ac_func_search_save_LIBS
+fi
+{ $as_echo "$as_me:$LINENO: result: $ac_cv_search_shm_unlink" >&5
+$as_echo "$ac_cv_search_shm_unlink" >&6; }
+ac_res=$ac_cv_search_shm_unlink
+if test "$ac_res" != no; then
+  test "$ac_res" = "none required" || LIBS="$ac_res $LIBS"
+
+fi
+
 # Solaris:
 { $as_echo "$as_me:$LINENO: checking for library containing fdatasync" >&5
 $as_echo_n "checking for library containing fdatasync... " >&6; }
@@ -19763,7 +19937,8 @@ LIBS=`echo "$LIBS" | sed -e 's/-ledit//g' -e 's/-lreadline//g'`
 
 
 
-for ac_func in cbrt dlopen fdatasync getifaddrs getpeerucred getrlimit mbstowcs_l memmove poll pstat readlink setproctitle setsid sigprocmask symlink sync_file_range towlower utime utimes wcstombs wcstombs_l
+
+for ac_func in cbrt dlopen fdatasync getifaddrs getpeerucred getrlimit mbstowcs_l memmove poll pstat readlink setproctitle setsid shm_open sigprocmask symlink sync_file_range towlower utime utimes wcstombs wcstombs_l
 do
 as_ac_var=`$as_echo "ac_cv_func_$ac_func" | $as_tr_sh`
 { $as_echo "$as_me:$LINENO: checking for $ac_func" >&5

configure.in

@@ -883,6 +883,8 @@ case $host_os in
 esac
 AC_SEARCH_LIBS(getopt_long, [getopt gnugetopt])
 AC_SEARCH_LIBS(crypt, crypt)
+AC_SEARCH_LIBS(shm_open, rt)
+AC_SEARCH_LIBS(shm_unlink, rt)
 # Solaris:
 AC_SEARCH_LIBS(fdatasync, [rt posix4])
 # Required for thread_test.c on Solaris 2.5:
@@ -1230,7 +1232,7 @@ PGAC_FUNC_GETTIMEOFDAY_1ARG
 LIBS_including_readline="$LIBS"
 LIBS=`echo "$LIBS" | sed -e 's/-ledit//g' -e 's/-lreadline//g'`
 
-AC_CHECK_FUNCS([cbrt dlopen fdatasync getifaddrs getpeerucred getrlimit mbstowcs_l memmove poll pstat readlink setproctitle setsid sigprocmask symlink sync_file_range towlower utime utimes wcstombs wcstombs_l])
+AC_CHECK_FUNCS([cbrt dlopen fdatasync getifaddrs getpeerucred getrlimit mbstowcs_l memmove poll pstat readlink setproctitle setsid shm_open sigprocmask symlink sync_file_range towlower utime utimes wcstombs wcstombs_l])
 
 AC_REPLACE_FUNCS(fseeko)
 case $host_os in

doc/src/sgml/config.sgml

@@ -1194,6 +1194,32 @@ include 'filename'
       </listitem>
      </varlistentry>
 
+     <varlistentry id="guc-dynamic-shared-memory-type" xreflabel="dynamic_shared_memory_type">
+      <term><varname>dynamic_shared_memory_type</varname> (<type>enum</type>)</term>
+      <indexterm>
+       <primary><varname>dynamic_shared_memory_type</> configuration parameter</primary>
+      </indexterm>
+      <listitem>
+       <para>
+        Specifies the dynamic shared memory implementation that the server
+        should use.  Possible values are <literal>posix</> (for POSIX shared
+        memory allocated using <literal>shm_open</>), <literal>sysv</literal>
+        (for System V shared memory allocated via <literal>shmget</>),
+        <literal>windows</> (for Windows shared memory), <literal>mmap</>
+        (to simulate shared memory using memory-mapped files stored in the
+        data directory), and <literal>none</> (to disable this feature).
+        Not all values are supported on all platforms; the first supported
+        option is the default for that platform.  The use of the
+        <literal>mmap</> option, which is not the default on any platform,
+        is generally discouraged because the operating system may write
+        modified pages back to disk repeatedly, increasing system I/O load;
+        however, it may be useful for debugging, when the
+        <literal>pg_dynshmem</> directory is stored on a RAM disk, or when
+        other shared memory facilities are not available.
+       </para>
+      </listitem>
+     </varlistentry>
+
      </variablelist>
      </sect2>
 

src/backend/port/sysv_shmem.c

@@ -29,6 +29,7 @@
 #endif
 
 #include "miscadmin.h"
+#include "portability/mem.h"
 #include "storage/ipc.h"
 #include "storage/pg_shmem.h"
 
@@ -36,31 +37,6 @@
 typedef key_t IpcMemoryKey;		/* shared memory key passed to shmget(2) */
 typedef int IpcMemoryId;		/* shared memory ID returned by shmget(2) */
 
-#define IPCProtection	(0600)	/* access/modify by user only */
-
-#ifdef SHM_SHARE_MMU			/* use intimate shared memory on Solaris */
-#define PG_SHMAT_FLAGS			SHM_SHARE_MMU
-#else
-#define PG_SHMAT_FLAGS			0
-#endif
-
-/* Linux prefers MAP_ANONYMOUS, but the flag is called MAP_ANON on other systems. */
-#ifndef MAP_ANONYMOUS
-#define MAP_ANONYMOUS			MAP_ANON
-#endif
-
-/* BSD-derived systems have MAP_HASSEMAPHORE, but it's not present (or needed) on Linux. */
-#ifndef MAP_HASSEMAPHORE
-#define MAP_HASSEMAPHORE		0
-#endif
-
-#define PG_MMAP_FLAGS			(MAP_SHARED|MAP_ANONYMOUS|MAP_HASSEMAPHORE)
-
-/* Some really old systems don't define MAP_FAILED. */
-#ifndef MAP_FAILED
-#define MAP_FAILED ((void *) -1)
-#endif
-
 
 unsigned long UsedShmemSegID = 0;
 void	   *UsedShmemSegAddr = NULL;

src/backend/storage/ipc/Makefile

@@ -15,7 +15,7 @@ override CFLAGS+= -fno-inline
 endif
 endif
 
-OBJS = ipc.o ipci.o pmsignal.o procarray.o procsignal.o shmem.o shmqueue.o \
-	sinval.o sinvaladt.o standby.o
+OBJS = dsm_impl.o dsm.o ipc.o ipci.o pmsignal.o procarray.o procsignal.o \
+	shmem.o shmqueue.o sinval.o sinvaladt.o standby.o
 
 include $(top_srcdir)/src/backend/common.mk

src/backend/storage/ipc/ipci.c

@@ -30,6 +30,7 @@
 #include "replication/walreceiver.h"
 #include "replication/walsender.h"
 #include "storage/bufmgr.h"
+#include "storage/dsm.h"
 #include "storage/ipc.h"
 #include "storage/pg_shmem.h"
 #include "storage/pmsignal.h"
@@ -249,6 +250,10 @@ CreateSharedMemoryAndSemaphores(bool makePrivate, int port)
 		ShmemBackendArrayAllocation();
 #endif
 
+	/* Initialize dynamic shared memory facilities. */
+	if (!IsUnderPostmaster)
+		dsm_postmaster_startup();
+
 	/*
 	 * Now give loadable modules a chance to set up their shmem allocations
 	 */

src/backend/utils/misc/guc.c

@@ -61,6 +61,7 @@
 #include "replication/walreceiver.h"
 #include "replication/walsender.h"
 #include "storage/bufmgr.h"
+#include "storage/dsm_impl.h"
 #include "storage/standby.h"
 #include "storage/fd.h"
 #include "storage/proc.h"
@@ -385,6 +386,7 @@ static const struct config_enum_entry synchronous_commit_options[] = {
  */
 extern const struct config_enum_entry wal_level_options[];
 extern const struct config_enum_entry sync_method_options[];
+extern const struct config_enum_entry dynamic_shared_memory_options[];
 
 /*
  * GUC option variables that are exported from this module
@@ -3335,6 +3337,16 @@ static struct config_enum ConfigureNamesEnum[] =
 		NULL, NULL, NULL
 	},
 
+	{
+		{"dynamic_shared_memory_type", PGC_POSTMASTER, RESOURCES_MEM,
+			gettext_noop("Selects the dynamic shared memory implementation used."),
+			NULL
+		},
+		&dynamic_shared_memory_type,
+		DEFAULT_DYNAMIC_SHARED_MEMORY_TYPE, dynamic_shared_memory_options,
+		NULL, NULL, NULL
+	},
+
 	{
 		{"wal_sync_method", PGC_SIGHUP, WAL_SETTINGS,
 			gettext_noop("Selects the method used for forcing WAL updates to disk."),

src/backend/utils/misc/postgresql.conf.sample

@@ -123,6 +123,13 @@
 #work_mem = 1MB				# min 64kB
 #maintenance_work_mem = 16MB		# min 1MB
 #max_stack_depth = 2MB			# min 100kB
+#dynamic_shared_memory_type = posix # the default is the first option
+					# supported by the operating system:
+					#   posix
+					#   sysv
+					#   windows
+					#   mmap
+					# use none to disable dynamic shared memory
 
 # - Disk -
 

src/backend/utils/resowner/resowner.c

@@ -98,6 +98,11 @@ typedef struct ResourceOwnerData
 	int			nfiles;			/* number of owned temporary files */
 	File	   *files;			/* dynamically allocated array */
 	int			maxfiles;		/* currently allocated array size */
+
+	/* We have built-in support for remembering dynamic shmem segments */
+	int			ndsms;			/* number of owned shmem segments */
+	dsm_segment **dsms;			/* dynamically allocated array */
+	int			maxdsms;		/* currently allocated array size */
 }	ResourceOwnerData;
 
 
@@ -132,6 +137,7 @@ static void PrintPlanCacheLeakWarning(CachedPlan *plan);
 static void PrintTupleDescLeakWarning(TupleDesc tupdesc);
 static void PrintSnapshotLeakWarning(Snapshot snapshot);
 static void PrintFileLeakWarning(File file);
+static void PrintDSMLeakWarning(dsm_segment *seg);
 
 
 /*****************************************************************************
@@ -271,6 +277,21 @@ ResourceOwnerReleaseInternal(ResourceOwner owner,
 				PrintRelCacheLeakWarning(owner->relrefs[owner->nrelrefs - 1]);
 			RelationClose(owner->relrefs[owner->nrelrefs - 1]);
 		}
+
+		/*
+		 * Release dynamic shared memory segments.  Note that dsm_detach()
+		 * will remove the segment from my list, so I just have to iterate
+		 * until there are none.
+		 *
+		 * As in the preceding cases, warn if there are leftover at commit
+		 * time.
+		 */
+		while (owner->ndsms > 0)
+		{
+			if (isCommit)
+				PrintDSMLeakWarning(owner->dsms[owner->ndsms - 1]);
+			dsm_detach(owner->dsms[owner->ndsms - 1]);
+		}
 	}
 	else if (phase == RESOURCE_RELEASE_LOCKS)
 	{
@@ -402,6 +423,7 @@ ResourceOwnerDelete(ResourceOwner owner)
 	Assert(owner->ncatrefs == 0);
 	Assert(owner->ncatlistrefs == 0);
 	Assert(owner->nrelrefs == 0);
+	Assert(owner->ndsms == 0);
 	Assert(owner->nplanrefs == 0);
 	Assert(owner->ntupdescs == 0);
 	Assert(owner->nsnapshots == 0);
@@ -438,6 +460,8 @@ ResourceOwnerDelete(ResourceOwner owner)
 		pfree(owner->snapshots);
 	if (owner->files)
 		pfree(owner->files);
+	if (owner->dsms)
+		pfree(owner->dsms);
 
 	pfree(owner);
 }
@@ -1230,3 +1254,88 @@ PrintFileLeakWarning(File file)
 		 "temporary file leak: File %d still referenced",
 		 file);
 }
+
+/*
+ * Make sure there is room for at least one more entry in a ResourceOwner's
+ * dynamic shmem segment reference array.
+ *
+ * This is separate from actually inserting an entry because if we run out
+ * of memory, it's critical to do so *before* acquiring the resource.
+ */
+void
+ResourceOwnerEnlargeDSMs(ResourceOwner owner)
+{
+	int			newmax;
+
+	if (owner->ndsms < owner->maxdsms)
+		return;					/* nothing to do */
+
+	if (owner->dsms == NULL)
+	{
+		newmax = 16;
+		owner->dsms = (dsm_segment **)
+			MemoryContextAlloc(TopMemoryContext,
+							   newmax * sizeof(dsm_segment *));
+		owner->maxdsms = newmax;
+	}
+	else
+	{
+		newmax = owner->maxdsms * 2;
+		owner->dsms = (dsm_segment **)
+			repalloc(owner->dsms, newmax * sizeof(dsm_segment *));
+		owner->maxdsms = newmax;
+	}
+}
+
+/*
+ * Remember that a dynamic shmem segment is owned by a ResourceOwner
+ *
+ * Caller must have previously done ResourceOwnerEnlargeDSMs()
+ */
+void
+ResourceOwnerRememberDSM(ResourceOwner owner, dsm_segment *seg)
+{
+	Assert(owner->ndsms < owner->maxdsms);
+	owner->dsms[owner->ndsms] = seg;
+	owner->ndsms++;
+}
+
+/*
+ * Forget that a temporary file is owned by a ResourceOwner
+ */
+void
+ResourceOwnerForgetDSM(ResourceOwner owner, dsm_segment *seg)
+{
+	dsm_segment **dsms = owner->dsms;
+	int			ns1 = owner->ndsms - 1;
+	int			i;
+
+	for (i = ns1; i >= 0; i--)
+	{
+		if (dsms[i] == seg)
+		{
+			while (i < ns1)
+			{
+				dsms[i] = dsms[i + 1];
+				i++;
+			}
+			owner->ndsms = ns1;
+			return;
+		}
+	}
+	elog(ERROR,
+		 "dynamic shared memory segment %u is not owned by resource owner %s",
+		 dsm_segment_handle(seg), owner->name);
+}
+
+
+/*
+ * Debugging subroutine
+ */
+static void
+PrintDSMLeakWarning(dsm_segment *seg)
+{
+	elog(WARNING,
+		 "dynamic shared memory leak: segment %u still referenced",
+		 dsm_segment_handle(seg));
+}

src/bin/initdb/initdb.c

@@ -182,6 +182,7 @@ const char *subdirs[] = {
 	"pg_xlog",
 	"pg_xlog/archive_status",
 	"pg_clog",
+	"pg_dynshmem",
 	"pg_notify",
 	"pg_serial",
 	"pg_snapshots",

src/include/pg_config.h.in

@@ -424,6 +424,9 @@
 /* Define to 1 if you have the `setsid' function. */
 #undef HAVE_SETSID
 
+/* Define to 1 if you have the `shm_open' function. */
+#undef HAVE_SHM_OPEN
+
 /* Define to 1 if you have the `sigprocmask' function. */
 #undef HAVE_SIGPROCMASK
 

src/include/portability/mem.h

+/*-------------------------------------------------------------------------
+ *
+ * mem.h
+ *	  portability definitions for various memory operations
+ *
+ * Copyright (c) 2001-2013, PostgreSQL Global Development Group
+ *
+ * src/include/portability/mem.h
+ *
+ *-------------------------------------------------------------------------
+ */
+#ifndef MEM_H
+#define MEM_H
+
+#define IPCProtection	(0600)	/* access/modify by user only */
+
+#ifdef SHM_SHARE_MMU			/* use intimate shared memory on Solaris */
+#define PG_SHMAT_FLAGS			SHM_SHARE_MMU
+#else
+#define PG_SHMAT_FLAGS			0
+#endif
+
+/* Linux prefers MAP_ANONYMOUS, but the flag is called MAP_ANON on other systems. */
+#ifndef MAP_ANONYMOUS
+#define MAP_ANONYMOUS			MAP_ANON
+#endif
+
+/* BSD-derived systems have MAP_HASSEMAPHORE, but it's not present (or needed) on Linux. */
+#ifndef MAP_HASSEMAPHORE
+#define MAP_HASSEMAPHORE		0
+#endif
+
+#define PG_MMAP_FLAGS			(MAP_SHARED|MAP_ANONYMOUS|MAP_HASSEMAPHORE)
+
+/* Some really old systems don't define MAP_FAILED. */
+#ifndef MAP_FAILED
+#define MAP_FAILED ((void *) -1)
+#endif
+
+#endif   /* MEM_H */

src/include/storage/dsm.h

+/*-------------------------------------------------------------------------
+ *
+ * dsm.h
+ *	  manage dynamic shared memory segments
+ *
+ * Portions Copyright (c) 1996-2013, PostgreSQL Global Development Group
+ * Portions Copyright (c) 1994, Regents of the University of California
+ *
+ * src/include/storage/dsm.h
+ *
+ *-------------------------------------------------------------------------
+ */
+#ifndef DSM_H
+#define DSM_H
+
+#include "storage/dsm_impl.h"
+
+typedef struct dsm_segment dsm_segment;
+
+/* Initialization function. */
+extern void dsm_postmaster_startup(void);
+
+/* Functions that create, update, or remove mappings. */
+extern dsm_segment *dsm_create(uint64 size);
+extern dsm_segment *dsm_attach(dsm_handle h);
+extern void *dsm_resize(dsm_segment *seg, uint64 size);
+extern void *dsm_remap(dsm_segment *seg);
+extern void dsm_detach(dsm_segment *seg);
+
+/* Resource management functions. */
+extern void dsm_keep_mapping(dsm_segment *seg);
+extern dsm_segment *dsm_find_mapping(dsm_handle h);
+
+/* Informational functions. */
+extern void *dsm_segment_address(dsm_segment *seg);
+extern uint64 dsm_segment_map_length(dsm_segment *seg);
+extern dsm_handle dsm_segment_handle(dsm_segment *seg);
+
+#endif   /* DSM_H */

src/include/storage/dsm_impl.h

+/*-------------------------------------------------------------------------
+ *
+ * dsm_impl.h
+ *	  low-level dynamic shared memory primitives
+ *
+ * Portions Copyright (c) 1996-2013, PostgreSQL Global Development Group
+ * Portions Copyright (c) 1994, Regents of the University of California
+ *
+ * src/include/storage/dsm_impl.h
+ *
+ *-------------------------------------------------------------------------
+ */
+#ifndef DSM_IMPL_H
+#define DSM_IMPL_H
+
+/* Dynamic shared memory implementations. */
+#define DSM_IMPL_NONE			0
+#define DSM_IMPL_POSIX			1
+#define DSM_IMPL_SYSV			2
+#define DSM_IMPL_WINDOWS		3
+#define DSM_IMPL_MMAP			4
+
+/*
+ * Determine which dynamic shared memory implementations will be supported
+ * on this platform, and which one will be the default.
+ */
+#ifdef WIN32
+#define USE_DSM_WINDOWS
+#define DEFAULT_DYNAMIC_SHARED_MEMORY_TYPE		DSM_IMPL_WINDOWS
+#else
+#ifdef HAVE_SHM_OPEN
+#define USE_DSM_POSIX
+#define DEFAULT_DYNAMIC_SHARED_MEMORY_TYPE		DSM_IMPL_POSIX
+#endif
+#define USE_DSM_SYSV
+#ifndef DEFAULT_DYNAMIC_SHARED_MEMORY_TYPE
+#define DEFAULT_DYNAMIC_SHARED_MEMORY_TYPE		DSM_IMPL_SYSV
+#endif
+#define USE_DSM_MMAP
+#endif
+
+/* GUC. */
+extern int dynamic_shared_memory_type;
+
+/*
+ * Directory for on-disk state.
+ *
+ * This is used by all implementations for crash recovery and by the mmap
+ * implementation for storage.
+ */
+#define PG_DYNSHMEM_DIR					"pg_dynshmem"
+#define PG_DYNSHMEM_MMAP_FILE_PREFIX	"mmap."
+
+/* A "name" for a dynamic shared memory segment. */
+typedef uint32 dsm_handle;
+
+/* All the shared-memory operations we know about. */
+typedef enum
+{
+	DSM_OP_CREATE,
+	DSM_OP_ATTACH,
+	DSM_OP_DETACH,
+	DSM_OP_RESIZE,
+	DSM_OP_DESTROY
+} dsm_op;
+
+/* Create, attach to, detach from, resize, or destroy a segment. */
+extern bool dsm_impl_op(dsm_op op, dsm_handle handle, uint64 request_size,
+			void **impl_private, void **mapped_address, uint64 *mapped_size,
+			int elevel);
+
+/* Some implementations cannot resize segments.  Can this one? */
+extern bool dsm_impl_can_resize(void);
+
+#endif   /* DSM_IMPL_H */

src/include/storage/lwlock.h

@@ -80,6 +80,7 @@ typedef enum LWLockId
 	OldSerXidLock,
 	SyncRepLock,
 	BackgroundWorkerLock,
+	DynamicSharedMemoryControlLock,
 	/* Individual lock IDs end here */
 	FirstBufMappingLock,
 	FirstLockMgrLock = FirstBufMappingLock + NUM_BUFFER_PARTITIONS,

src/include/utils/resowner_private.h

@@ -16,6 +16,7 @@
 #ifndef RESOWNER_PRIVATE_H
 #define RESOWNER_PRIVATE_H
 
+#include "storage/dsm.h"
 #include "storage/fd.h"
 #include "storage/lock.h"
 #include "utils/catcache.h"
@@ -80,4 +81,11 @@ extern void ResourceOwnerRememberFile(ResourceOwner owner,
 extern void ResourceOwnerForgetFile(ResourceOwner owner,
 						File file);
 
+/* support for dynamic shared memory management */
+extern void ResourceOwnerEnlargeDSMs(ResourceOwner owner);
+extern void ResourceOwnerRememberDSM(ResourceOwner owner,
+						  dsm_segment *);
+extern void ResourceOwnerForgetDSM(ResourceOwner owner,
+						dsm_segment *);
+
 #endif   /* RESOWNER_PRIVATE_H */