Fix i386 / x86_64 nearbyint exception clearing (bug 15491).

The implementations of nearbyint functions using x87 floating point
(i386 all versions, x86_64 long double only) use the fclex
instruction, which clears any exceptions that were raised before the
function was called.  These functions must not clear exceptions that
were raised before they were called.

This patch fixes these functions to save and restore the whole
floating-point environment (fnstenv / fldenv) as the way of avoiding
raising "inexact" (recall that there isn't an x87 instruction for
loading just the status word, so the whole environment has to be saved
and loaded instead - the code already saved and loaded the control
word, which is now obtained from the saved environment after this
patch, to disable traps on "inexact").  In the case of the long double
functions, any "invalid" exception from frndint (applied to a
signaling NaN) needs merging into the saved state; this issue doesn't
apply to the float and double functions because that exception would
have been raised when the argument is loaded, before the environment
is saved.

	[BZ #15491]
	* sysdeps/i386/fpu/s_nearbyint.S (__nearbyint): Save and restore
	floating-point environment instead of clearing all exceptions.
	* sysdeps/i386/fpu/s_nearbyintf.S (__nearbyintf): Likewise.
	* sysdeps/i386/fpu/s_nearbyintl.S (__nearbyintl): Likewise,
	merging in "invalid" exceptions from frndint.
	* sysdeps/x86_64/fpu/s_nearbyintl.S (__nearbyintl): Likewise.
	* math/test-nearbyint-except.c: New file.
	* math/Makefile (tests): Add test-nearbyint-except.

3 files changed, 30 insertions, 42 deletions

diff --git a/sysdeps/i386/fpu/s_nearbyint.S b/sysdeps/i386/fpu/s_nearbyint.S
index 5290913..8da8ae9 100644
--- a/sysdeps/i386/fpu/s_nearbyint.S
+++ b/sysdeps/i386/fpu/s_nearbyint.S
@@ -8,22 +8,17 @@
 
 ENTRY(__nearbyint)
 	fldl	4(%esp)
-	pushl	%eax
-	cfi_adjust_cfa_offset (4)
-	pushl	%ecx
-	cfi_adjust_cfa_offset (4)
-	fnstcw	(%esp)
-	movl	(%esp), %eax
+	subl	$32, %esp
+	cfi_adjust_cfa_offset (32)
+	fnstenv	4(%esp)
+	movl	4(%esp), %eax
 	orl	$0x20, %eax
-	movl	%eax, 4(%esp)
-	fldcw	4(%esp)
-	frndint
-	fclex
+	movl	%eax, (%esp)
 	fldcw	(%esp)
-	popl	%ecx
-	cfi_adjust_cfa_offset (-4)
-	popl	%eax
-	cfi_adjust_cfa_offset (-4)
+	frndint
+	fldenv	4(%esp)
+	addl	$32, %esp
+	cfi_adjust_cfa_offset (-32)
 	ret
 END (__nearbyint)
 weak_alias (__nearbyint, nearbyint)
diff --git a/sysdeps/i386/fpu/s_nearbyintf.S b/sysdeps/i386/fpu/s_nearbyintf.S
index 6a6b1d6..0c51f72 100644
--- a/sysdeps/i386/fpu/s_nearbyintf.S
+++ b/sysdeps/i386/fpu/s_nearbyintf.S
@@ -8,22 +8,17 @@
 
 ENTRY(__nearbyintf)
 	flds	4(%esp)
-	pushl	%eax
-	cfi_adjust_cfa_offset (4)
-	pushl	%ecx
-	cfi_adjust_cfa_offset (4)
-	fnstcw	(%esp)
-	movl	(%esp), %eax
+	subl	$32, %esp
+	cfi_adjust_cfa_offset (32)
+	fnstenv	4(%esp)
+	movl	4(%esp), %eax
 	orl	$0x20, %eax
-	movl	%eax, 4(%esp)
-	fldcw	4(%esp)
-	frndint
-	fclex
+	movl	%eax, (%esp)
 	fldcw	(%esp)
-	popl	%ecx
-	cfi_adjust_cfa_offset (-4)
-	popl	%eax
-	cfi_adjust_cfa_offset (-4)
+	frndint
+	fldenv	4(%esp)
+	addl	$32, %esp
+	cfi_adjust_cfa_offset (-32)
 	ret
 END (__nearbyintf)
 weak_alias (__nearbyintf, nearbyintf)
diff --git a/sysdeps/i386/fpu/s_nearbyintl.S b/sysdeps/i386/fpu/s_nearbyintl.S
index 819af63..b260ab5 100644
--- a/sysdeps/i386/fpu/s_nearbyintl.S
+++ b/sysdeps/i386/fpu/s_nearbyintl.S
@@ -8,22 +8,20 @@
 
 ENTRY(__nearbyintl)
 	fldt	4(%esp)
-	pushl	%eax
-	cfi_adjust_cfa_offset (4)
-	pushl	%ecx
-	cfi_adjust_cfa_offset (4)
-	fnstcw	(%esp)
-	movl	(%esp), %eax
+	subl	$32, %esp
+	cfi_adjust_cfa_offset (32)
+	fnstenv	4(%esp)
+	movl	4(%esp), %eax
 	orl	$0x20, %eax
-	movl	%eax, 4(%esp)
-	fldcw	4(%esp)
-	frndint
-	fclex
+	movl	%eax, (%esp)
 	fldcw	(%esp)
-	popl	%ecx
-	cfi_adjust_cfa_offset (-4)
-	popl	%eax
-	cfi_adjust_cfa_offset (-4)
+	frndint
+	fnstsw
+	andl	$0x1, %eax
+	orl	%eax, 8(%esp)
+	fldenv	4(%esp)
+	addl	$32, %esp
+	cfi_adjust_cfa_offset (-32)
 	ret
 END (__nearbyintl)
 weak_alias (__nearbyintl, nearbyintl)