dnl IA-64 mpn_mul_1, mpn_mul_1c -- Multiply a limb vector with a limb and
dnl store the result in a second limb vector.
dnl Contributed to the GNU project by Torbjorn Granlund.
dnl Copyright 2000-2004, 2006, 2007 Free Software Foundation, Inc.
dnl This file is part of the GNU MP Library.
dnl
dnl The GNU MP Library is free software; you can redistribute it and/or modify
dnl it under the terms of either:
dnl
dnl * the GNU Lesser General Public License as published by the Free
dnl Software Foundation; either version 3 of the License, or (at your
dnl option) any later version.
dnl
dnl or
dnl
dnl * the GNU General Public License as published by the Free Software
dnl Foundation; either version 2 of the License, or (at your option) any
dnl later version.
dnl
dnl or both in parallel, as here.
dnl
dnl The GNU MP Library is distributed in the hope that it will be useful, but
dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
dnl for more details.
dnl
dnl You should have received copies of the GNU General Public License and the
dnl GNU Lesser General Public License along with the GNU MP Library. If not,
dnl see https://www.gnu.org/licenses/.
include(`../config.m4')
C cycles/limb
C Itanium: 4.0
C Itanium 2: 2.0
C TODO
C * Further optimize feed-in and wind-down code, both for speed andcode size.
C * Handle low limb input and results specially, using a common stf8 in the
C epilogue.
C * Use 1 c/l carry propagation scheme in wind-down code.
C * Use extra pointer register for `up' to speed up feed-in loads.
C * Work out final differences with addmul_1.asm.
C INPUT PARAMETERS
define(`rp', `r32')
define(`up', `r33')
define(`n', `r34')
define(`vl', `r35')
define(`cy', `r36') C for mpn_mul_1c
ifdef(`HAVE_ABI_32',
` addp4 rp = 0, rp C M I
addp4 up = 0, up C M I
zxt4 n = n C I ;; ')
{.mfi
adds r15 = -1, n C M I mov f9 = f0 C F mov.i r2 = ar.lc C I0
}
{.mmi
ldf8 f7 = [up], 8 C M nop.m 0 C M and r14 = 3, n C M I ;;
}
.Lcommon:
{.mii
setf.sig f6 = vl C M2 M3 shr.u r31 = r15, 2 C I0 cmp.eq p10, p0 = 0, r14 C M I
}
{.mii cmp.eq p11, p0 = 2, r14 C M I cmp.eq p12, p0 = 3, r14 C M I nop.i 0 C I ;;
}
{.mii cmp.ne p6, p7 = r0, r0 C M I mov.i ar.lc = r31 C I0 cmp.ne p8, p9 = r0, r0 C M I
}
{.bbb
(p10) br.dptk .Lb00 C B
(p11) br.dptk .Lb10 C B
(p12) br.dptk .Lb11 C B ;;
}
.Lb01: mov r20 = 0
br.cloop.dptk .grt1 C B
xma.l f39 = f7, f6, f9 C F
xma.hu f43 = f7, f6, f9 C F ;;
getf.sig r8 = f43 C M2
stf8 [rp] = f39 C M2 M3 mov.i ar.lc = r2 C I0
br.ret.sptk.many b0 C B
ifdef(`HAVE_ABI_32',
` addp4 rp = 0, rp C M I
addp4 up = 0, up C M I
zxt4 n = n C I ;; ')
{.mmi
adds r15 = -1, n C M I
setf.sig f9 = cy C M2 M3 mov.i r2 = ar.lc C I0
}
{.mmb
ldf8 f7 = [up], 8 C M and r14 = 3, n C M I
br.sptk .Lcommon ;;
}
EPILOGUE()
ASM_END()
Messung V0.5
¤ Dauer der Verarbeitung: 0.14 Sekunden
(vorverarbeitet)
¤
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.