[PATCH 0/3] Resubmit of Sparc T3/T4 patches.
davem at davemloft.net
Tue Mar 5 22:27:22 CET 2013
From: Torbjorn Granlund <tg at gmplib.org>
Date: Tue, 05 Mar 2013 22:14:07 +0100
> David Miller <davem at davemloft.net> writes:
> From: Torbjorn Granlund <tg at gmplib.org>
> Date: Tue, 05 Mar 2013 21:35:19 +0100
> > Richard Henderson <rth at twiddle.net> writes:
> > One extra add insn here (copy-paste from addmul)?
> > addcc %o5, %g3, %g3
> > addxccc %g2, %g1, %g1
> > addxc %g0, %o4, %o5
> > Since I cannot test this at all (qemu-system-sparc64 persistenty resists
> > all my usage attempts) I need you to perform the following tests:
> Richard doesn't have access to any addxccc capable machines (virtual
> or otherwise) either :-)
> But you do.
> Did you commit all of these 3 patches to the HG repo? If so I'll
> work on testing the simplification Richard suggested above.
> I've only committed the sparc32 patches so far, since the other arrived
> just a few minutes ago. Also, I cannot run 'try' on these.
The versions I posted passed all of the tests.
Richard's suggestion is an optimization, which I will test once
the original versions go into the tree.
More information about the gmp-devel