Prevent int128 from requiring more than MAXALIGN alignment.

Our initial work with int128 neglected alignment considerations, an
oversight that came back to bite us in bug #14897 from Vincent Lachenal.
It is unsurprising that int128 might have a 16-byte alignment requirement;
what's slightly more surprising is that even notoriously lax Intel chips
sometimes enforce that.

Raising MAXALIGN seems out of the question: the costs in wasted disk and
memory space would be significant, and there would also be an on-disk
compatibility break.  Nor does it seem very practical to try to allow some
data structures to have more-than-MAXALIGN alignment requirement, as we'd
have to push knowledge of that throughout various code that copies data
structures around.

The only way out of the box is to make type int128 conform to the system's
alignment assumptions.  Fortunately, gcc supports that via its
__attribute__(aligned()) pragma; and since we don't currently support
int128 on non-gcc-workalike compilers, we shouldn't be losing any platform
support this way.

Although we could have just done pg_attribute_aligned(MAXIMUM_ALIGNOF) and
called it a day, I did a little bit of extra work to make the code more
portable than that: it will also support int128 on compilers without
__attribute__(aligned()), if the native alignment of their 128-bit-int
type is no more than that of int64.

Add a regression test case that exercises the one known instance of the
problem, in parallel aggregation over a bigint column.

Back-patch of commit 751804998.  The code known to be affected only exists
in 9.6 and later, but we do have some stuff using int128 in 9.5, so patch
back to 9.5.

Discussion: https://postgr.es/m/20171110185747.31519.28038@wrigleys.postgresql.org

1 files changed, 5 insertions, 2 deletions

diff --git a/configure.in b/configure.in
index 1ba41896f39..0b97a532a64 100644
--- a/configure.in
+++ b/configure.in
@@ -1848,7 +1848,10 @@ AC_CHECK_ALIGNOF(double)
 
 # Compute maximum alignment of any basic type.
 # We assume long's alignment is at least as strong as char, short, or int;
-# but we must check long long (if it exists) and double.
+# but we must check long long (if it is being used for int64) and double.
+# Note that we intentionally do not consider any types wider than 64 bits,
+# as allowing MAXIMUM_ALIGNOF to exceed 8 would be too much of a penalty
+# for disk and memory space.
 
 MAX_ALIGNOF=$ac_cv_alignof_long
 if test $MAX_ALIGNOF -lt $ac_cv_alignof_double ; then
@@ -1865,7 +1868,7 @@ AC_DEFINE_UNQUOTED(MAXIMUM_ALIGNOF, $MAX_ALIGNOF, [Define as the maximum alignme
 AC_CHECK_TYPES([int8, uint8, int64, uint64], [], [],
 [#include <stdio.h>])
 
-# Check for extensions offering the integer scalar type __int128.
+# Some compilers offer a 128-bit integer scalar type.
 PGAC_TYPE_128BIT_INT
 
 # Check for various atomic operations now that we have checked how to declare