[Bug c/27051] New: Compiler generates .sdata when -mno-sdata specified

[jsworley at qwest dot net]

Given the following code:

float
abc(float a, float b)
{
    return ((a - b) * 0.1);
}

Compiled with the -mno-sdata flag, the compiler still defines an .sdata item
for the floating constant 0.1, which causes problems for embedded systems
that rely on -mno-sdata meaning NO .sdata

This bug can be circumvented as follows:

float
abc(float a, float b)
{
    static float r = 0.1;

    return ((a - b) * r);
}


-- 
           Summary: Compiler generates .sdata when -mno-sdata specified
           Product: gcc
           Version: 3.4.5
            Status: UNCONFIRMED
          Severity: minor
          Priority: P3
         Component: c
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: jsworley at qwest dot net
 GCC build triplet: ia64-unknown-linux-gnu
  GCC host triplet: ia64-redhat-linux
GCC target triplet: ia64-unknown-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27051