procyberian/gcc
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

c++: Fix up constant expression __builtin_convertvector folding [PR104472]

Browse source 
The following testcase ICEs, because due to the -frounding-math
fold_const_call fails, which is it returns NULL, and returning NULL from
cxx_eval* is wrong, all the callers rely on them to either return folded
value or original with *non_constant_p = true.

The following patch does that, and additionally falls through into the
default case where there is diagnostics for the !ctx->quiet case too.

2022-02-11  Jakub Jelinek  <jakub@redhat.com>

	PR c++/104472
	* constexpr.cc (cxx_eval_internal_function) <case IFN_VEC_CONVERT>:
	Only return fold_const_call result if it is non-NULL.  Otherwise
	fall through into the default: case to return t, set *non_constant_p
	and emit diagnostics if needed.

	* g++.dg/cpp0x/constexpr-104472.C: New test.
This commit is contained in:
Jakub Jelinek 2022-02-11 13:52:44 +01:00
parent fb76c0ad35
commit 84993d94e1
2 changed files with 12 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

9
gcc/cp/constexpr.cc
View file

@ -1840,13 +1840,10 @@ cxx_eval_internal_function (const constexpr_ctx *ctx, tree t,
false, non_constant_p,
overflow_p);
if (TREE_CODE (arg) == VECTOR_CST)
return fold_const_call (CFN_VEC_CONVERT, TREE_TYPE (t), arg);
else
{
*non_constant_p = true;
return t;
}
if (tree r = fold_const_call (CFN_VEC_CONVERT, TREE_TYPE (t), arg))
return r;
}
/* FALLTHRU */
default:
if (!ctx->quiet)

9
gcc/testsuite/g++.dg/cpp0x/constexpr-104472.C Normal file
View file

@ -0,0 +1,9 @@
// PR c++/104472
// { dg-options "-O2 -frounding-math" }
// { dg-add-options float16 }
// { dg-require-effective-target float16 }
typedef short __attribute__((__vector_size__ (16))) V;
typedef _Float16 __attribute__((__vector_size__ (16))) F;
V v = __builtin_convertvector (__builtin_convertvector ((V){5534}, F), V) < 8;