[PATCH 0/3] Resubmit of Sparc T3/T4 patches.

David Miller davem at davemloft.net
Tue Mar 5 23:22:18 CET 2013

From: Richard Henderson <rth at twiddle.net>
Date: Tue, 05 Mar 2013 12:25:53 -0800

> One extra add insn here (copy-paste from addmul)?
> 	addcc	%o5, %g3, %g3
> 	addxccc	%g2, %g1, %g1
> 	addxc	%g0, %o4, %o5

Yep, that works perfectly fine, but makes no difference performance
wise.  I made the same testsuite+try testing as the original code,
all passes.

This is on top of the 3 original patches, thanks Richard:

2013-03-05  David S. Miller  <davem at davemloft.net>

	* mpn/sparc64/ultrasparct3/mul_1.asm (mpn_mul_1): Simplify carrying
	additions (as suggested by Richard Henderson).

diff --git a/mpn/sparc64/ultrasparct3/mul_1.asm b/mpn/sparc64/ultrasparct3/mul_1.asm
index df52647..6a3f193 100644
--- a/mpn/sparc64/ultrasparct3/mul_1.asm
+++ b/mpn/sparc64/ultrasparct3/mul_1.asm
@@ -50,8 +50,7 @@ L(top):
 	umulxhi	%o4, v0, %o4
 	addcc	%o5, %g3, %g3
 	stx	%g3, [rp-16]
-	addxc	%g0, %g2, %o5
-	addcc	%o5, %g1, %g1
+	addxccc	%g2, %g1, %g1
 	addxc	%g0, %o4, %o5
 	brgz	n, L(top)
 	 stx	%g1, [rp-8]

More information about the gmp-devel mailing list