URL
https://opencores.org/ocsvn/openrisc/openrisc/trunk
Subversion Repositories openrisc
[/] [openrisc/] [trunk/] [gnu-stable/] [gcc-4.5.1/] [gcc/] [testsuite/] [gcc.target/] [spu/] [dfcgt-nan.c] - Rev 859
Go to most recent revision | Compare with Previous | Blame | View Log
/* { dg-do compile } */ /* { dg-options "-march=celledp -O1" } */ /* { dg-final { scan-assembler "dfceq" } } */ /* GCC previously transformed an "a <= b" test into "! (a > b)" when compiling with -march=celledp, so that the dfcgt instruction can be used to implement the comparison. However, this transformation violates the IEEE-754 standard in the presence of NaN values. If either a or b is a NaN, a <= b should evaluate to false according to IEEE rules. However, after the transformation, a > b as implemented by dfcgt itself returns false, so the transformed test returns true. Note that the equivalent transformation is valid for single- precision floating-point values on the Cell SPU, because the format does not have NaNs. It is invalid for double-precision, even on Cell, however. */ int test (double a, double b) __attribute__ ((noinline)); int test (double a, double b) { return a <= b; } int main (void) { double x = 0.0; double y = 0.0/0.0; return test (x, y); }
Go to most recent revision | Compare with Previous | Blame | View Log