This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH 3/9] ENABLE_CHECKING refactoring: Java and Ada


On 10/05/2015 05:30 PM, Mikhail Maltsev wrote:
gcc/java/ChangeLog:

2015-10-05  Mikhail Maltsev<maltsevm@gmail.com>

	* decl.c (java_mark_decl_local): Use flag_checking instead of
	ENABLE_CHECKING.


gcc/ada/ChangeLog:

2015-10-05  Mikhail Maltsev<maltsevm@gmail.com>

	* gcc-interface/decl.c (gnat_to_gnu_entity): Use gcc_checking_assert.
	* gcc-interface/trans.c (assoc_to_constructor): Use flag_checking.
	* gcc-interface/utils.c (relate_alias_sets): Likewise.
	* gcc-interface/utils2.c (build_binary_op, build_unary_op): Use
	gcc_checking_assert



0003-Ada-and-Java-FE-s.patch


 From e8de7d2dc24cff85b6c1e44157dad23c85e435e1 Mon Sep 17 00:00:00 2001
From: Mikhail Maltsev<maltsevm@gmail.com>
Date: Sun, 20 Sep 2015 05:06:01 +0300
Subject: [PATCH 3/9] Ada and Java FE's

---
  gcc/ada/gcc-interface/decl.c   |  4 +---
  gcc/ada/gcc-interface/trans.c  | 11 ++++++-----
  gcc/ada/gcc-interface/utils.c  |  4 +---
  gcc/ada/gcc-interface/utils2.c | 20 ++++++++------------
  gcc/java/decl.c                |  4 +---
  5 files changed, 17 insertions(+), 26 deletions(-)

diff --git a/gcc/ada/gcc-interface/utils.c b/gcc/ada/gcc-interface/utils.c
index 0f3087d..7906495 100644
--- a/gcc/ada/gcc-interface/utils.c
+++ b/gcc/ada/gcc-interface/utils.c
@@ -1499,9 +1499,7 @@ relate_alias_sets (tree gnu_new_type, tree gnu_old_type, enum alias_set_op op)
        /* The alias set shouldn't be copied between array types with different
  	 aliasing settings because this can break the aliasing relationship
  	 between the array type and its element type.  */
-#ifndef ENABLE_CHECKING
-      if (flag_strict_aliasing)
-#endif
+      if (flag_checking || flag_strict_aliasing)
  	gcc_assert (!(TREE_CODE (gnu_new_type) == ARRAY_TYPE
  		      && TREE_CODE (gnu_old_type) == ARRAY_TYPE
  		      && TYPE_NONALIASED_COMPONENT (gnu_new_type)
I wonder who wrote that gem initially. I don't like conditional compilation, I don't like the original code here even more. Having the IF, but not the true/false arms conditionally compiled is just gross.

Having said that, I hope it wasn't me that wrote the code :-)


I've installed this on the trunk.

jeff


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]