Allow dynamic allocation of shared memory segments.

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

Author: robertmhaas

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

@@ -29,38 +29,14 @@
 #endif
 
 #include "miscadmin.h"
+#include "portability/mem.h"
 #include "storage/ipc.h"
 #include "storage/pg_shmem.h"
 
 
 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