Commit 660a0ebd by Jim Wilson Committed by Jim Wilson

Fix glibc math testsuite failures.

	* config/ia64/lib1funcs.asm (__divdi3, __moddi3, __udivdi3, __umoddi3):
	Use .s1 not .s0 for all FP instructions.

From-SVN: r33160
parent f2d5f0cc
Fri Apr 14 16:58:45 2000 Jim Wilson <wilson@cygnus.com>
* config/ia64/lib1funcs.asm (__divdi3, __moddi3, __udivdi3, __umoddi3):
Use .s1 not .s0 for all FP instructions.
2000-04-14 Zack Weinberg <zack@wolery.cumb.org>
* cpplex.c (cpp_output_tokens, cpp_scan_buffer_nooutput): New public
......
......@@ -100,7 +100,7 @@ __divdi3:
fcvt.xf f9 = f9
;;
// Compute the reciprocal approximation.
frcpa f10, p6 = f8, f9
frcpa.s1 f10, p6 = f8, f9
;;
// 3 Newton-Raphson iterations.
(p6) fma.s1 f11 = farg0, f10, f0
......@@ -119,10 +119,10 @@ __divdi3:
;;
(p6) fnma.s1 f8 = f9, f11, f8
;;
(p6) fma f10 = f8, f10, f11
(p6) fma.s1 f10 = f8, f10, f11
;;
// Round quotient to an integer.
fcvt.fx.trunc f8 = f10
fcvt.fx.trunc.s1 f8 = f10
;;
// Transfer result to GP registers.
getf.sig ret0 = f8
......@@ -160,7 +160,7 @@ __moddi3:
fcvt.xf f9 = f9
;;
// Compute the reciprocal approximation.
frcpa f10, p6 = f8, f9
frcpa.s1 f10, p6 = f8, f9
;;
// 3 Newton-Raphson iterations.
(p6) fma.s1 f11 = farg0, f10, f0
......@@ -179,19 +179,19 @@ __moddi3:
;;
(p6) fnma.s1 f12 = f9, f11, f8
;;
(p6) fma f10 = f12, f10, f11
(p6) fma.s1 f10 = f12, f10, f11
;;
// Round quotient to an integer.
fcvt.fx.trunc f10 = f10
fcvt.fx.trunc.s1 f10 = f10
;;
// Renormalize.
fcvt.xf f10 = f10
;;
// Compute remainder.
fnma f8 = f10, f9, f8
fnma.s1 f8 = f10, f9, f8
;;
// Round remainder to an integer.
fcvt.fx.trunc f8 = f8
fcvt.fx.trunc.s1 f8 = f8
;;
// Transfer result to GP registers.
getf.sig ret0 = f8
......@@ -225,11 +225,11 @@ __udivdi3:
setf.sig f9 = in1
;;
// Convert the inputs to FP, to avoid FP software-assist faults.
fcvt.xuf f8 = f8
fcvt.xuf f9 = f9
fcvt.xuf.s1 f8 = f8
fcvt.xuf.s1 f9 = f9
;;
// Compute the reciprocal approximation.
frcpa f10, p6 = f8, f9
frcpa.s1 f10, p6 = f8, f9
;;
// 3 Newton-Raphson iterations.
(p6) fma.s1 f11 = farg0, f10, f0
......@@ -248,10 +248,10 @@ __udivdi3:
;;
(p6) fnma.s1 f8 = f9, f11, f8
;;
(p6) fma f10 = f8, f10, f11
(p6) fma.s1 f10 = f8, f10, f11
;;
// Round quotient to an unsigned integer.
fcvt.fxu.trunc f8 = f10
fcvt.fxu.trunc.s1 f8 = f10
;;
// Transfer result to GP registers.
getf.sig ret0 = f8
......@@ -285,11 +285,11 @@ __umoddi3:
setf.sig f9 = in1
;;
// Convert the inputs to FP, to avoid FP software assist faults.
fcvt.xuf f8 = f8
fcvt.xuf f9 = f9
fcvt.xuf.s1 f8 = f8
fcvt.xuf.s1 f9 = f9
;;
// Compute the reciprocal approximation.
frcpa f10, p6 = f8, f9
frcpa.s1 f10, p6 = f8, f9
;;
// 3 Newton-Raphson iterations.
(p6) fma.s1 f11 = farg0, f10, f0
......@@ -308,19 +308,19 @@ __umoddi3:
;;
(p6) fnma.s1 f12 = f9, f11, f8
;;
(p6) fma f10 = f12, f10, f11
(p6) fma.s1 f10 = f12, f10, f11
;;
// Round quotient to an unsigned integer.
fcvt.fxu.trunc f10 = f10
fcvt.fxu.trunc.s1 f10 = f10
;;
// Renormalize.
fcvt.xuf f10 = f10
fcvt.xuf.s1 f10 = f10
;;
// Compute remainder.
fnma f8 = f10, f9, f8
fnma.s1 f8 = f10, f9, f8
;;
// Round remainder to an integer.
fcvt.fxu.trunc f8 = f8
fcvt.fxu.trunc.s1 f8 = f8
;;
// Transfer result to GP registers.
getf.sig ret0 = f8
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment