diff options
author | Andres Freund | 2017-12-16 20:49:41 +0000 |
---|---|---|
committer | Andres Freund | 2017-12-16 20:49:41 +0000 |
commit | c04d35f442a8c4fd5a20103b31839ec52fce3046 (patch) | |
tree | 7b39615fb8aac2b41fe17c02f91f95981d717341 /config/c-compiler.m4 | |
parent | c757a3da0af0e5eb636eeee2af6602d279162b0a (diff) |
Try to detect runtime unavailability of __builtin_mul_overflow(int64).
On some systems the results of 64 bit __builtin_mul_overflow()
operations can be computed at compile time, but not at runtime. The
known cases are arm buildfar animals using clang where the runtime
operation is implemented in a unavailable function.
Try to avoid compile-time computation by using volatile arguments to
__builtin_mul_overflow(). In that case we hopefully will get a link
error when unavailable, similar to what buildfarm animals dangomushi
and gull are reporting.
Author: Andres Freund
Discussion: https://postgr.es/m/20171213213754.pydkyjs6bt2hvsdb@alap3.anarazel.de
Diffstat (limited to 'config/c-compiler.m4')
-rw-r--r-- | config/c-compiler.m4 | 12 |
1 files changed, 8 insertions, 4 deletions
diff --git a/config/c-compiler.m4 b/config/c-compiler.m4 index 67323ade124..b35436481c5 100644 --- a/config/c-compiler.m4 +++ b/config/c-compiler.m4 @@ -310,13 +310,17 @@ fi])# PGAC_C_BUILTIN_CONSTANT_P # and define HAVE__BUILTIN_OP_OVERFLOW if so. # # Check for the most complicated case, 64 bit multiplication, as a -# proxy for all of the operations. Have to link to be sure to -# recognize a missing __builtin_mul_overflow. +# proxy for all of the operations. Use volatile variables to avoid the +# compiler computing result at compile time, even though the runtime +# might not supply operation. Have to link to be sure to recognize a +# missing __builtin_mul_overflow. AC_DEFUN([PGAC_C_BUILTIN_OP_OVERFLOW], [AC_CACHE_CHECK(for __builtin_mul_overflow, pgac_cv__builtin_op_overflow, [AC_LINK_IFELSE([AC_LANG_PROGRAM([], -[PG_INT64_TYPE result; -__builtin_mul_overflow((PG_INT64_TYPE) 1, (PG_INT64_TYPE) 2, &result);] +[PG_INT64_TYPE a = 1; +PG_INT64_TYPE b = 1; +PG_INT64_TYPE result; +__builtin_mul_overflow(*(volatile PG_INT64_TYPE *) a, *(volatile PG_INT64_TYPE *) b, &result);] )], [pgac_cv__builtin_op_overflow=yes], [pgac_cv__builtin_op_overflow=no])]) |