Added code to print the unregistered OS2 width values using the real instructions and constants used by Core Text.

doc/export/writewidths.c

@@ -0,0 +1,3 @@
+// 22 october 2017
+extern int realMain(void);
+int main(void) { return realMain(); }

doc/export/writewidths.out

@@ -0,0 +1,11 @@
+unregistered OS2 0:	float64as(-0.5, 0xbfe0000000000000)
+unregistered OS2 1:	float64as(-0.4, 0xbfd999999999999a)
+unregistered OS2 2:	float64as(-0.3, 0xbfd3333333333333)
+unregistered OS2 3:	float64as(-0.2, 0xbfc999999999999a)
+unregistered OS2 4:	float64as(-0.1, 0xbfb9999999999998)
+unregistered OS2 5:	float64as(0, 0x0000000000000000)
+unregistered OS2 6:	float64as(0.1, 0x3fb9999999999998)
+unregistered OS2 7:	float64as(0.2, 0x3fc9999999999998)
+unregistered OS2 8:	float64as(0.3, 0x3fd3333333333334)
+unregistered OS2 9:	float64as(0.4, 0x3fd999999999999a)
+unregistered OS2 10:	float64as(0.5, 0x3fe0000000000000)

doc/export/writewidths.s

@@ -0,0 +1,51 @@
+# 22 october 2017
+# clang -o writewidths writewidths.c writewidths.s -g -Wall -Wextra -pedantic -g
+# thanks to:
+# - http://www.idryman.org/blog/2014/12/02/writing-64-bit-assembly-on-mac-os-x/
+# - https://developer.apple.com/library/content/documentation/DeveloperTools/Reference/Assembler/060-i386_Addressing_Modes_and_Assembler_Instructions/i386_intructions.html#//apple_ref/doc/uid/TP30000825-TPXREF101
+# - https://stackoverflow.com/questions/46309041/trivial-macos-assembly-64-bit-program-has-incorrect-stack-alignment
+# - https://www.google.com/search?q=macos+implement+main+in+assembly+-nasm&oq=macos+implement+main+in+assembly+-nasm&gs_l=psy-ab.3...12877.13839.0.13988.6.6.0.0.0.0.117.407.4j1.5.0....0...1.1.64.psy-ab..1.0.0....0.et6MkokjvwA
+# - https://stackoverflow.com/questions/2529185/what-are-cfi-directives-in-gnu-assembler-gas-used-for
+
+.section __DATA,__data
+
+double10:
+	.quad	0x4024000000000000
+doubleNeg05:
+	.quad	0xbfe0000000000000
+
+fmt:
+	.asciz	"unregistered OS2 %ld:\tfloat64as(%g, 0x%016lx)\n"
+
+.section __TEXT,__text
+.globl _realMain
+_realMain:
+	pushq		%rbp
+	movq		%rsp, %rbp
+	addq			$8, %rsp
+
+	xorq			%rcx, %rcx
+loop:
+	pushq		%rcx
+	# the code from core text
+	movzwl		%cx, %ecx
+	xorps		%xmm0, %xmm0
+	cvtsi2sdl		%ecx, %xmm0
+	divsd		double10(%rip), %xmm0
+	addsd		doubleNeg05(%rip), %xmm0
+	# end core text code
+	popq		%rcx
+	pushq		%rcx
+	movd		%xmm0, %rdx
+	movzwq		%cx, %rsi
+	leaq			fmt(%rip), %rdi
+	callq			_printf
+	popq		%rcx
+	incw			%cx
+	cmpw		$10, %cx
+	jbe			loop
+
+	xorq			%rax, %rax
+	subq			$8, %rsp
+	popq		%rbp
+	ret