SIGFPE with no exceptions currently enabled?

[Raymond Toy]

While I've seen this SIGFPE issue before, I never tracked it down. Here is a repeatable case (from Carl):

    (defun test (x y)
      (declare (optimize (speed 3) (safety 0))
               (type (signed-byte 32) x y))
      (the (signed-byte 32) (truncate x y)))

Although the code lies to the compiler (the asserted return type of truncate is incorrect), the point is that (test #x-80000000 -1) causes the SIGFPE exception.

I believe sigfpe-handler was only expecting to handle FP exceptions. In this example, the division causes an overflow which generates the SIGFPE.

[toy.raymond@…]

commit 0174bfebdb2ed8edbdf5b3dade4fd64965d4a9f4 Author: Raymond Toy <toy.raymond@…> Date: Thu Jan 31 21:27:36 2013 -0800

Fix ticket:72 : SIGFPE with no exceptions enabled

code/float-trap.lisp

• In the default case, check FOP to see if it's defined. This means it's a divide exception.

code/x86-vm.lisp

• Also check to see if the offending instruction is a DIV or IDIV, which means we got an integer overflow. Return the appropriate values in this case.