Change the declaration of struct varlena so that the length word is

represented as "char ...[4]" not "int32".  Since the length word is never
supposed to be accessed via this struct member anyway, this won't break
any existing code that is following the rules.  The advantage is that C
compilers will no longer assume that a pointer to struct varlena is
word-aligned, which prevents incorrect optimizations in TOAST-pointer
access and perhaps other places.  gcc doesn't seem to do this (at least
not at -O2), but the problem is demonstrable on some other compilers.

I changed struct inet as well, but didn't bother to touch a lot of other
struct definitions in which it wouldn't make any difference because there
were other fields forcing int alignment anyway.  Hopefully none of those
struct definitions are used for accessing unaligned Datums.

diff --git a/doc/src/sgml/xtypes.sgml b/doc/src/sgml/xtypes.sgml
index bc84ccd1c4b016b5bfb8898605a10d861cf5b181..c80126a11952c6202432c048e9fc9840a338af24 100644 (file)
--- a/doc/src/sgml/xtypes.sgml
+++ b/doc/src/sgml/xtypes.sgml
@@ -1,4 +1,4 @@
-<!-- $PostgreSQL: pgsql/doc/src/sgml/xtypes.sgml,v 1.29 2007/05/15 17:39:54 momjian Exp $ -->
+<!-- $PostgreSQL: pgsql/doc/src/sgml/xtypes.sgml,v 1.30 2008/02/23 19:11:45 tgl Exp $ -->
 
  <sect1 id="xtypes">
   <title>User-Defined Types</title>
@@ -219,7 +219,7 @@ CREATE TYPE complex (
   <productname>PostgreSQL</productname> automatically provides support
   for arrays of that
   type.<indexterm><primary>array</primary><secondary>of user-defined
-  type</secondary></indexterm> For historical reasons, the array type
+  type</secondary></indexterm>  The array type typically
   has the same name as the base type with the underscore character
   (<literal>_</>) prepended.
  </para>
@@ -240,15 +240,16 @@ CREATE TYPE complex (
   If the values of your data type vary in size (in internal form), you should
   make the data type <acronym>TOAST</>-able (see <xref
   linkend="storage-toast">). You should do this even if the data are always
-  too small to be compressed or stored externally because
-  <productname>Postgres</> can save space on small data using
-  <acronym>TOAST</> as well. 
+  too small to be compressed or stored externally, because
+  <acronym>TOAST</> can save space on small data too, by reducing header
+  overhead.
  </para>
 
  <para>
   To do this, the internal representation must follow the standard layout for
-  variable-length data: the first four bytes must be an <type>int32</type>
-  which is never accessed directly (customarily named <literal>vl_len_</>). You
+  variable-length data: the first four bytes must be a <type>char[4]</type>
+  field which is never accessed directly (customarily named
+  <structfield>vl_len_</>). You
   must use <function>SET_VARSIZE()</function> to store the size of the datum
   in this field and <function>VARSIZE()</function> to retrieve it. The C
   functions operating on the data type must always be careful to unpack any
@@ -265,12 +266,25 @@ CREATE TYPE complex (
   to avoid some of the overhead of <function>PG_DETOAST_DATUM</>. You can use
   <function>PG_DETOAST_DATUM_PACKED</> instead (customarily hidden by
   defining a <function>GETARG_DATATYPE_PP</> macro) and using the macros
-  <function>VARSIZE_ANY_EXHDR</> and <function>VARDATA_ANY</> macros.
+  <function>VARSIZE_ANY_EXHDR</> and <function>VARDATA_ANY</> to access
+  a potentially-packed datum.
   Again, the data returned by these macros is not aligned even if the data
   type definition specifies an alignment. If the alignment is important you
   must go through the regular <function>PG_DETOAST_DATUM</> interface.
  </para>
 
+ <note>
+  <para>
+   Older code frequently declares <structfield>vl_len_</> as an
+   <type>int32</> field instead of <type>char[4]</>.  This is OK as long as
+   the struct definition has other fields that have at least <type>int32</>
+   alignment.  But it is dangerous to use such a struct definition when
+   working with a potentially unaligned datum; the compiler may take it as
+   license to assume the datum actually is aligned, leading to core dumps on
+   architectures that are strict about alignment.
+  </para>
+ </note>
+
  <para>
   For further details see the description of the
   <xref linkend="sql-createtype" endterm="sql-createtype-title"> command.

diff --git a/src/backend/access/heap/tuptoaster.c b/src/backend/access/heap/tuptoaster.c
index ca49ec9d937be34544fd6e30024c1fbeca11da02..a7fd6d0984d50c3444a088172c24d8d54f915209 100644 (file)
--- a/src/backend/access/heap/tuptoaster.c
+++ b/src/backend/access/heap/tuptoaster.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *       $PostgreSQL: pgsql/src/backend/access/heap/tuptoaster.c,v 1.81 2008/01/01 19:45:46 momjian Exp $
+ *       $PostgreSQL: pgsql/src/backend/access/heap/tuptoaster.c,v 1.82 2008/02/23 19:11:45 tgl Exp $
  *
  *
  * INTERFACE ROUTINES
@@ -65,7 +65,8 @@
 #define VARATT_EXTERNAL_GET_POINTER(toast_pointer, attr) \
 do { \
        varattrib_1b_e *attre = (varattrib_1b_e *) (attr); \
-       Assert(VARSIZE_ANY_EXHDR(attre) == sizeof(toast_pointer)); \
+       Assert(VARATT_IS_EXTERNAL(attre)); \
+       Assert(VARSIZE_EXTERNAL(attre) == sizeof(toast_pointer) + VARHDRSZ_EXTERNAL); \
        memcpy(&(toast_pointer), VARDATA_EXTERNAL(attre), sizeof(toast_pointer)); \
 } while (0)
 

diff --git a/src/include/c.h b/src/include/c.h
index b66006200f8f9ee0bdf9eb1f4cbca6aa6f25e238..07a29d9cdf4e1d0e1356a0d8e0170a384a61c27f 100644 (file)
--- a/src/include/c.h
+++ b/src/include/c.h
@@ -12,7 +12,7 @@
  * Portions Copyright (c) 1996-2008, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/c.h,v 1.222 2008/01/01 19:45:56 momjian Exp $
+ * $PostgreSQL: pgsql/src/include/c.h,v 1.223 2008/02/23 19:11:45 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -410,7 +410,7 @@ typedef struct
  */
 struct varlena
 {
-       int32           vl_len_;                /* Do not touch this field directly! */
+       char            vl_len_[4];             /* Do not touch this field directly! */
        char            vl_dat[1];
 };
 

diff --git a/src/include/utils/inet.h b/src/include/utils/inet.h
index d102573f05f184007e3409a019e3713bd151fdce..793e1207857ca9f1fedde73f495762771a9db86a 100644 (file)
--- a/src/include/utils/inet.h
+++ b/src/include/utils/inet.h
@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1996-2008, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/utils/inet.h,v 1.28 2008/01/01 19:45:59 momjian Exp $
+ * $PostgreSQL: pgsql/src/include/utils/inet.h,v 1.29 2008/02/23 19:11:45 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -49,7 +49,7 @@ typedef struct
  */
 typedef struct
 {
-       int32           vl_len_;                /* Do not touch this field directly! */
+       char            vl_len_[4];             /* Do not touch this field directly! */
        inet_struct inet_data;
 } inet;