mirror of
https://github.com/rn10950/RetroZilla.git
synced 2024-11-10 01:40:17 +01:00
1031 lines
21 KiB
Plaintext
1031 lines
21 KiB
Plaintext
!
|
|
! ***** BEGIN LICENSE BLOCK *****
|
|
! Version: MPL 1.1/GPL 2.0/LGPL 2.1
|
|
!
|
|
! The contents of this file are subject to the Mozilla Public License Version
|
|
! 1.1 (the "License"); you may not use this file except in compliance with
|
|
! the License. You may obtain a copy of the License at
|
|
! http://www.mozilla.org/MPL/
|
|
!
|
|
! Software distributed under the License is distributed on an "AS IS" basis,
|
|
! WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
|
|
! for the specific language governing rights and limitations under the
|
|
! License.
|
|
!
|
|
! The Original Code is vis inline macros (64 bit). (vis_64.il 3.4).
|
|
!
|
|
! The Initial Developer of the Original Code is
|
|
! Sun Microsystems Inc.
|
|
! Portions created by the Initial Developer are Copyright (C) 1998-2000
|
|
! the Initial Developer. All Rights Reserved.
|
|
!
|
|
! Contributor(s):
|
|
!
|
|
! Alternatively, the contents of this file may be used under the terms of
|
|
! either the GNU General Public License Version 2 or later (the "GPL"), or
|
|
! the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
|
|
! in which case the provisions of the GPL or the LGPL are applicable instead
|
|
! of those above. If you wish to allow use of your version of this file only
|
|
! under the terms of either the GPL or the LGPL, and not to allow others to
|
|
! use your version of this file under the terms of the MPL, indicate your
|
|
! decision by deleting the provisions above and replace them with the notice
|
|
! and other provisions required by the GPL or the LGPL. If you do not delete
|
|
! the provisions above, a recipient may use your version of this file under
|
|
! the terms of any one of the MPL, the GPL or the LGPL.
|
|
!
|
|
! ***** END LICENSE BLOCK *****
|
|
! $Id: vis_64.il,v 1.3 2004/04/27 23:04:36 gerv%gerv.net Exp $
|
|
|
|
! This file is to be used in place of vis.il in 64-bit builds.
|
|
|
|
!--------------------------------------------------------------------
|
|
! Pure edge handling instructions
|
|
!
|
|
! int vis_edge8(void */*frs1*/, void */*frs2*/);
|
|
!
|
|
.inline vis_edge8,16
|
|
edge8 %o0,%o1,%o0
|
|
.end
|
|
!
|
|
! int vis_edge8l(void */*frs1*/, void */*frs2*/);
|
|
!
|
|
.inline vis_edge8l,16
|
|
edge8l %o0,%o1,%o0
|
|
.end
|
|
!
|
|
! int vis_edge16(void */*frs1*/, void */*frs2*/);
|
|
!
|
|
.inline vis_edge16,16
|
|
edge16 %o0,%o1,%o0
|
|
.end
|
|
!
|
|
! int vis_edge16l(void */*frs1*/, void */*frs2*/);
|
|
!
|
|
.inline vis_edge16l,16
|
|
edge16l %o0,%o1,%o0
|
|
.end
|
|
!
|
|
! int vis_edge32(void */*frs1*/, void */*frs2*/);
|
|
!
|
|
.inline vis_edge32,16
|
|
edge32 %o0,%o1,%o0
|
|
.end
|
|
!
|
|
! int vis_edge32l(void */*frs1*/, void */*frs2*/);
|
|
!
|
|
.inline vis_edge32l,16
|
|
edge32l %o0,%o1,%o0
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Edge handling instructions with negative return values if cc set
|
|
!
|
|
! int vis_edge8cc(void */*frs1*/, void */*frs2*/);
|
|
!
|
|
.inline vis_edge8cc,16
|
|
edge8 %o0,%o1,%o0
|
|
mov 0,%o1
|
|
movgu %xcc,-1024,%o1
|
|
or %o1,%o0,%o0
|
|
.end
|
|
!
|
|
! int vis_edge8lcc(void */*frs1*/, void */*frs2*/);
|
|
!
|
|
.inline vis_edge8lcc,16
|
|
edge8l %o0,%o1,%o0
|
|
mov 0,%o1
|
|
movgu %xcc,-1024,%o1
|
|
or %o1,%o0,%o0
|
|
.end
|
|
!
|
|
! int vis_edge16cc(void */*frs1*/, void */*frs2*/);
|
|
!
|
|
.inline vis_edge16cc,16
|
|
edge16 %o0,%o1,%o0
|
|
mov 0,%o1
|
|
movgu %xcc,-1024,%o1
|
|
or %o1,%o0,%o0
|
|
.end
|
|
!
|
|
! int vis_edge16lcc(void */*frs1*/, void */*frs2*/);
|
|
!
|
|
.inline vis_edge16lcc,16
|
|
edge16l %o0,%o1,%o0
|
|
mov 0,%o1
|
|
movgu %xcc,-1024,%o1
|
|
or %o1,%o0,%o0
|
|
.end
|
|
!
|
|
! int vis_edge32cc(void */*frs1*/, void */*frs2*/);
|
|
!
|
|
.inline vis_edge32cc,16
|
|
edge32 %o0,%o1,%o0
|
|
mov 0,%o1
|
|
movgu %xcc,-1024,%o1
|
|
or %o1,%o0,%o0
|
|
.end
|
|
!
|
|
! int vis_edge32lcc(void */*frs1*/, void */*frs2*/);
|
|
!
|
|
.inline vis_edge32lcc,16
|
|
edge32l %o0,%o1,%o0
|
|
mov 0,%o1
|
|
movgu %xcc,-1024,%o1
|
|
or %o1,%o0,%o0
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Alignment instructions
|
|
!
|
|
! void *vis_alignaddr(void */*rs1*/, int /*rs2*/);
|
|
!
|
|
.inline vis_alignaddr,12
|
|
alignaddr %o0,%o1,%o0
|
|
.end
|
|
!
|
|
! void *vis_alignaddrl(void */*rs1*/, int /*rs2*/);
|
|
!
|
|
.inline vis_alignaddrl,12
|
|
alignaddrl %o0,%o1,%o0
|
|
.end
|
|
!
|
|
! double vis_faligndata(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_faligndata,16
|
|
faligndata %f0,%f2,%f0
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Partitioned comparison instructions
|
|
!
|
|
! int vis_fcmple16(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fcmple16,16
|
|
fcmple16 %f0,%f2,%o0
|
|
.end
|
|
!
|
|
! int vis_fcmpne16(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fcmpne16,16
|
|
fcmpne16 %f0,%f2,%o0
|
|
.end
|
|
!
|
|
! int vis_fcmple32(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fcmple32,16
|
|
fcmple32 %f0,%f2,%o0
|
|
.end
|
|
!
|
|
! int vis_fcmpne32(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fcmpne32,16
|
|
fcmpne32 %f0,%f2,%o0
|
|
.end
|
|
!
|
|
! int vis_fcmpgt16(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fcmpgt16,16
|
|
fcmpgt16 %f0,%f2,%o0
|
|
.end
|
|
!
|
|
! int vis_fcmpeq16(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fcmpeq16,16
|
|
fcmpeq16 %f0,%f2,%o0
|
|
.end
|
|
!
|
|
! int vis_fcmpgt32(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fcmpgt32,16
|
|
fcmpgt32 %f0,%f2,%o0
|
|
.end
|
|
!
|
|
! int vis_fcmpeq32(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fcmpeq32,16
|
|
fcmpeq32 %f0,%f2,%o0
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Partitioned arithmetic
|
|
!
|
|
! double vis_fmul8x16(float /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fmul8x16,12
|
|
fmul8x16 %f1,%f2,%f0
|
|
.end
|
|
!
|
|
! double vis_fmul8x16_dummy(float /*frs1*/, int /*dummy*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fmul8x16_dummy,16
|
|
fmul8x16 %f1,%f4,%f0
|
|
.end
|
|
!
|
|
! double vis_fmul8x16au(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fmul8x16au,8
|
|
fmul8x16au %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fmul8x16al(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fmul8x16al,8
|
|
fmul8x16al %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fmul8sux16(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fmul8sux16,16
|
|
fmul8sux16 %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! double vis_fmul8ulx16(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fmul8ulx16,16
|
|
fmul8ulx16 %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! double vis_fmuld8sux16(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fmuld8sux16,8
|
|
fmuld8sux16 %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fmuld8ulx16(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fmuld8ulx16,8
|
|
fmuld8ulx16 %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fpadd16(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fpadd16,16
|
|
fpadd16 %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fpadd16s(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fpadd16s,8
|
|
fpadd16s %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fpadd32(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fpadd32,16
|
|
fpadd32 %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fpadd32s(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fpadd32s,8
|
|
fpadd32s %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fpsub16(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fpsub16,16
|
|
fpsub16 %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fpsub16s(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fpsub16s,8
|
|
fpsub16s %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fpsub32(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fpsub32,16
|
|
fpsub32 %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fpsub32s(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fpsub32s,8
|
|
fpsub32s %f1,%f3,%f0
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Pixel packing
|
|
!
|
|
! float vis_fpack16(double /*frs2*/);
|
|
!
|
|
.inline vis_fpack16,8
|
|
fpack16 %f0,%f0
|
|
.end
|
|
!
|
|
! double vis_fpack16_pair(double /*frs2*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fpack16_pair,16
|
|
fpack16 %f0,%f0
|
|
fpack16 %f2,%f1
|
|
.end
|
|
!
|
|
! void vis_st2_fpack16(double, double, double *)
|
|
!
|
|
.inline vis_st2_fpack16,24
|
|
fpack16 %f0,%f0
|
|
fpack16 %f2,%f1
|
|
st %f0,[%o2+0]
|
|
st %f1,[%o2+4]
|
|
.end
|
|
!
|
|
! void vis_std_fpack16(double, double, double *)
|
|
!
|
|
.inline vis_std_fpack16,24
|
|
fpack16 %f0,%f0
|
|
fpack16 %f2,%f1
|
|
std %f0,[%o2]
|
|
.end
|
|
!
|
|
! void vis_st2_fpackfix(double, double, double *)
|
|
!
|
|
.inline vis_st2_fpackfix,24
|
|
fpackfix %f0,%f0
|
|
fpackfix %f2,%f1
|
|
st %f0,[%o2+0]
|
|
st %f1,[%o2+4]
|
|
.end
|
|
!
|
|
! double vis_fpack16_to_hi(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fpack16_to_hi,16
|
|
fpack16 %f2,%f0
|
|
.end
|
|
|
|
! double vis_fpack16_to_lo(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fpack16_to_lo,16
|
|
fpack16 %f2,%f3
|
|
fmovs %f3,%f1 /* without this, optimizer goes wrong */
|
|
.end
|
|
|
|
!
|
|
! double vis_fpack32(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fpack32,16
|
|
fpack32 %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fpackfix(double /*frs2*/);
|
|
!
|
|
.inline vis_fpackfix,8
|
|
fpackfix %f0,%f0
|
|
.end
|
|
!
|
|
! double vis_fpackfix_pair(double /*frs2*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fpackfix_pair,16
|
|
fpackfix %f0,%f0
|
|
fpackfix %f2,%f1
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Motion estimation
|
|
!
|
|
! double vis_pxldist64(double accum /*frd*/, double pxls1 /*frs1*/,
|
|
! double pxls2 /*frs2*/);
|
|
!
|
|
.inline vis_pxldist64,24
|
|
pdist %f2,%f4,%f0
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Channel merging
|
|
!
|
|
! double vis_fpmerge(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fpmerge,8
|
|
fpmerge %f1,%f3,%f0
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Pixel expansion
|
|
!
|
|
! double vis_fexpand(float /*frs2*/);
|
|
!
|
|
.inline vis_fexpand,4
|
|
fexpand %f1,%f0
|
|
.end
|
|
|
|
! double vis_fexpand_hi(double /*frs2*/);
|
|
!
|
|
.inline vis_fexpand_hi,8
|
|
fexpand %f0,%f0
|
|
.end
|
|
|
|
! double vis_fexpand_lo(double /*frs2*/);
|
|
!
|
|
.inline vis_fexpand_lo,8
|
|
fexpand %f1,%f0
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Bitwise logical operations
|
|
!
|
|
! double vis_fnor(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fnor,16
|
|
fnor %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fnors(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fnors,8
|
|
fnors %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fandnot(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fandnot,16
|
|
fandnot1 %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fandnots(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fandnots,8
|
|
fandnot1s %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fnot(double /*frs1*/);
|
|
!
|
|
.inline vis_fnot,8
|
|
fnot1 %f0,%f0
|
|
.end
|
|
!
|
|
! float vis_fnots(float /*frs1*/);
|
|
!
|
|
.inline vis_fnots,4
|
|
fnot1s %f1,%f0
|
|
.end
|
|
!
|
|
! double vis_fxor(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fxor,16
|
|
fxor %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fxors(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fxors,8
|
|
fxors %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fnand(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fnand,16
|
|
fnand %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fnands(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fnands,8
|
|
fnands %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fand(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fand,16
|
|
fand %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fands(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fands,8
|
|
fands %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fxnor(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fxnor,16
|
|
fxnor %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fxnors(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fxnors,8
|
|
fxnors %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fsrc(double /*frs1*/);
|
|
!
|
|
.inline vis_fsrc,8
|
|
fsrc1 %f0,%f0
|
|
.end
|
|
!
|
|
! float vis_fsrcs(float /*frs1*/);
|
|
!
|
|
.inline vis_fsrcs,4
|
|
fsrc1s %f1,%f0
|
|
.end
|
|
!
|
|
! double vis_fornot(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_fornot,16
|
|
fornot1 %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fornots(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fornots,8
|
|
fornot1s %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_for(double /*frs1*/, double /*frs2*/);
|
|
!
|
|
.inline vis_for,16
|
|
for %f0,%f2,%f0
|
|
.end
|
|
!
|
|
! float vis_fors(float /*frs1*/, float /*frs2*/);
|
|
!
|
|
.inline vis_fors,8
|
|
fors %f1,%f3,%f0
|
|
.end
|
|
!
|
|
! double vis_fzero(/* void */)
|
|
!
|
|
.inline vis_fzero,0
|
|
fzero %f0
|
|
.end
|
|
!
|
|
! float vis_fzeros(/* void */)
|
|
!
|
|
.inline vis_fzeros,0
|
|
fzeros %f0
|
|
.end
|
|
!
|
|
! double vis_fone(/* void */)
|
|
!
|
|
.inline vis_fone,0
|
|
fone %f0
|
|
.end
|
|
!
|
|
! float vis_fones(/* void */)
|
|
!
|
|
.inline vis_fones,0
|
|
fones %f0
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Partial store instructions
|
|
!
|
|
! vis_stdfa_ASI_PST8P(double frd, void *rs1, int rmask)
|
|
!
|
|
.inline vis_stdfa_ASI_PST8P,20
|
|
stda %f0,[%o1]%o2,0xc0 ! ASI_PST8_P
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_PST8PL(double frd, void *rs1, int rmask)
|
|
!
|
|
.inline vis_stdfa_ASI_PST8PL,20
|
|
stda %f0,[%o1]%o2,0xc8 ! ASI_PST8_PL
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_PST8P_int_pair(void *rs1, void *rs2, void *rs3, int rmask);
|
|
!
|
|
.inline vis_stdfa_ASI_PST8P_int_pair,28
|
|
ld [%o0],%f4
|
|
ld [%o1],%f5
|
|
stda %f4,[%o2]%o3,0xc0 ! ASI_PST8_P
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_PST8S(double frd, void *rs1, int rmask)
|
|
!
|
|
.inline vis_stdfa_ASI_PST8S,20
|
|
stda %f0,[%o1]%o2,0xc1 ! ASI_PST8_S
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_PST16P(double frd, void *rs1, int rmask)
|
|
!
|
|
.inline vis_stdfa_ASI_PST16P,20
|
|
stda %f0,[%o1]%o2,0xc2 ! ASI_PST16_P
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_PST16S(double frd, void *rs1, int rmask)
|
|
!
|
|
.inline vis_stdfa_ASI_PST16S,20
|
|
stda %f0,[%o1]%o2,0xc3 ! ASI_PST16_S
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_PST32P(double frd, void *rs1, int rmask)
|
|
!
|
|
.inline vis_stdfa_ASI_PST32P,20
|
|
stda %f0,[%o1]%o2,0xc4 ! ASI_PST32_P
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_PST32S(double frd, void *rs1, int rmask)
|
|
!
|
|
.inline vis_stdfa_ASI_PST32S,20
|
|
stda %f0,[%o1]%o2,0xc5 ! ASI_PST32_S
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Short store instructions
|
|
!
|
|
! vis_stdfa_ASI_FL8P(double frd, void *rs1)
|
|
!
|
|
.inline vis_stdfa_ASI_FL8P,16
|
|
stda %f0,[%o1]0xd0 ! ASI_FL8_P
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_FL8P_index(double frd, void *rs1, long index)
|
|
!
|
|
.inline vis_stdfa_ASI_FL8P_index,24
|
|
stda %f0,[%o1+%o2]0xd0 ! ASI_FL8_P
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_FL8S(double frd, void *rs1)
|
|
!
|
|
.inline vis_stdfa_ASI_FL8S,16
|
|
stda %f0,[%o1]0xd1 ! ASI_FL8_S
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_FL16P(double frd, void *rs1)
|
|
!
|
|
.inline vis_stdfa_ASI_FL16P,16
|
|
stda %f0,[%o1]0xd2 ! ASI_FL16_P
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_FL16P_index(double frd, void *rs1, long index)
|
|
!
|
|
.inline vis_stdfa_ASI_FL16P_index,24
|
|
stda %f0,[%o1+%o2]0xd2 ! ASI_FL16_P
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_FL16S(double frd, void *rs1)
|
|
!
|
|
.inline vis_stdfa_ASI_FL16S,16
|
|
stda %f0,[%o1]0xd3 ! ASI_FL16_S
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_FL8PL(double frd, void *rs1)
|
|
!
|
|
.inline vis_stdfa_ASI_FL8PL,16
|
|
stda %f0,[%o1]0xd8 ! ASI_FL8_PL
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_FL8SL(double frd, void *rs1)
|
|
!
|
|
.inline vis_stdfa_ASI_FL8SL,16
|
|
stda %f0,[%o1]0xd9 ! ASI_FL8_SL
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_FL16PL(double frd, void *rs1)
|
|
!
|
|
.inline vis_stdfa_ASI_FL16PL,16
|
|
stda %f0,[%o1]0xda ! ASI_FL16_PL
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_FL16SL(double frd, void *rs1)
|
|
!
|
|
.inline vis_stdfa_ASI_FL16SL,16
|
|
stda %f0,[%o1]0xdb ! ASI_FL16_SL
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Short load instructions
|
|
!
|
|
! double vis_lddfa_ASI_FL8P(void *rs1)
|
|
!
|
|
.inline vis_lddfa_ASI_FL8P,8
|
|
ldda [%o0]0xd0,%f4 ! ASI_FL8_P
|
|
fmovd %f4,%f0 ! Compiler can clean this up
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL8P_index(void *rs1, long index)
|
|
!
|
|
.inline vis_lddfa_ASI_FL8P_index,16
|
|
ldda [%o0+%o1]0xd0,%f4
|
|
fmovd %f4,%f0
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL8P_hi(void *rs1, unsigned int index)
|
|
!
|
|
.inline vis_lddfa_ASI_FL8P_hi,12
|
|
sra %o1,16,%o1
|
|
ldda [%o0+%o1]0xd0,%f4
|
|
fmovd %f4,%f0
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL8P_lo(void *rs1, unsigned int index)
|
|
!
|
|
.inline vis_lddfa_ASI_FL8P_lo,12
|
|
sll %o1,16,%o1
|
|
sra %o1,16,%o1
|
|
ldda [%o0+%o1]0xd0,%f4
|
|
fmovd %f4,%f0
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL8S(void *rs1)
|
|
!
|
|
.inline vis_lddfa_ASI_FL8S,8
|
|
ldda [%o0]0xd1,%f4 ! ASI_FL8_S
|
|
fmovd %f4,%f0
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL16P(void *rs1)
|
|
!
|
|
.inline vis_lddfa_ASI_FL16P,8
|
|
ldda [%o0]0xd2,%f4 ! ASI_FL16_P
|
|
fmovd %f4,%f0
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL16P_index(void *rs1, long index)
|
|
!
|
|
.inline vis_lddfa_ASI_FL16P_index,16
|
|
ldda [%o0+%o1]0xd2,%f4 ! ASI_FL16_P
|
|
fmovd %f4,%f0
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL16S(void *rs1)
|
|
!
|
|
.inline vis_lddfa_ASI_FL16S,8
|
|
ldda [%o0]0xd3,%f4 ! ASI_FL16_S
|
|
fmovd %f4,%f0
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL8PL(void *rs1)
|
|
!
|
|
.inline vis_lddfa_ASI_FL8PL,8
|
|
ldda [%o0]0xd8,%f4 ! ASI_FL8_PL
|
|
fmovd %f4,%f0
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL8PL_index(void *rs1, long index)
|
|
!
|
|
.inline vis_lddfa_ASI_FL8PL_index,16
|
|
ldda [%o0+%o1]0xd8,%f4 ! ASI_FL8_PL
|
|
fmovd %f4,%f0
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL8SL(void *rs1)
|
|
!
|
|
.inline vis_lddfa_ASI_FL8SL,8
|
|
ldda [%o0]0xd9,%f4 ! ASI_FL8_SL
|
|
fmovd %f4,%f0
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL16PL(void *rs1)
|
|
!
|
|
.inline vis_lddfa_ASI_FL16PL,8
|
|
ldda [%o0]0xda,%f4 ! ASI_FL16_PL
|
|
fmovd %f4,%f0
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL16PL_index(void *rs1, long index)
|
|
!
|
|
.inline vis_lddfa_ASI_FL16PL_index,16
|
|
ldda [%o0+%o1]0xda,%f4 ! ASI_FL16_PL
|
|
fmovd %f4,%f0
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_FL16SL(void *rs1)
|
|
!
|
|
.inline vis_lddfa_ASI_FL16SL,8
|
|
ldda [%o0]0xdb,%f4 ! ASI_FL16_SL
|
|
fmovd %f4,%f0
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Graphics status register
|
|
!
|
|
! unsigned int vis_read_gsr(void)
|
|
!
|
|
.inline vis_read_gsr,0
|
|
rd %gsr,%o0
|
|
.end
|
|
!
|
|
! void vis_write_gsr(unsigned int /* GSR */)
|
|
!
|
|
.inline vis_write_gsr,4
|
|
wr %g0,%o0,%gsr
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Voxel texture mapping
|
|
!
|
|
! unsigned long vis_array8(unsigned long long /*rs1 */, int /*rs2*/)
|
|
!
|
|
.inline vis_array8,12
|
|
array8 %o0,%o1,%o0
|
|
.end
|
|
!
|
|
! unsigned long vis_array16(unsigned long long /*rs1*/, int /*rs2*/)
|
|
!
|
|
.inline vis_array16,12
|
|
array16 %o0,%o1,%o0
|
|
.end
|
|
!
|
|
! unsigned long vis_array32(unsigned long long /*rs1*/, int /*rs2*/)
|
|
!
|
|
.inline vis_array32,12
|
|
array32 %o0,%o1,%o0
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Register aliasing and type casts
|
|
!
|
|
! float vis_read_hi(double /* frs1 */);
|
|
!
|
|
.inline vis_read_hi,8
|
|
fmovs %f0,%f0
|
|
.end
|
|
!
|
|
! float vis_read_lo(double /* frs1 */);
|
|
!
|
|
.inline vis_read_lo,8
|
|
fmovs %f1,%f0 ! %f0 = low word (frs1); return %f0;
|
|
.end
|
|
!
|
|
! double vis_write_hi(double /* frs1 */, float /* frs2 */);
|
|
!
|
|
.inline vis_write_hi,12
|
|
fmovs %f3,%f0 ! %f3 = float frs2; return %f0:f1;
|
|
.end
|
|
!
|
|
! double vis_write_lo(double /* frs1 */, float /* frs2 */);
|
|
!
|
|
.inline vis_write_lo,12
|
|
fmovs %f3,%f1 ! %f3 = float frs2; return %f0:f1;
|
|
.end
|
|
!
|
|
! double vis_freg_pair(float /* frs1 */, float /* frs2 */);
|
|
!
|
|
.inline vis_freg_pair,8
|
|
fmovs %f1,%f0 ! %f1 = float frs1; put in hi;
|
|
fmovs %f3,%f1 ! %f3 = float frs2; put in lo; return %f0:f1;
|
|
.end
|
|
!
|
|
! float vis_to_float(unsigned int /*value*/);
|
|
!
|
|
.inline vis_to_float,4
|
|
st %o0,[%sp+2183]
|
|
ld [%sp+2183],%f0
|
|
.end
|
|
!
|
|
! double vis_to_double(unsigned int /*value1*/, unsigned int /*value2*/);
|
|
!
|
|
.inline vis_to_double,8
|
|
st %o0,[%sp+2183]
|
|
ld [%sp+2183],%f0
|
|
st %o1,[%sp+2183]
|
|
ld [%sp+2183],%f1
|
|
.end
|
|
!
|
|
! double vis_to_double_dup(unsigned int /*value*/);
|
|
!
|
|
.inline vis_to_double_dup,4
|
|
st %o0,[%sp+2183]
|
|
ld [%sp+2183],%f1
|
|
fmovs %f1,%f0 ! duplicate value
|
|
.end
|
|
!
|
|
! double vis_ll_to_double(unsigned long long /*value*/);
|
|
!
|
|
.inline vis_ll_to_double,8
|
|
stx %o0,[%sp+2183]
|
|
ldd [%sp+2183],%f0
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Address space identifier (ASI) register
|
|
!
|
|
! unsigned int vis_read_asi(void)
|
|
!
|
|
.inline vis_read_asi,0
|
|
rd %asi,%o0
|
|
.end
|
|
!
|
|
! void vis_write_asi(unsigned int /* ASI */)
|
|
!
|
|
.inline vis_write_asi,4
|
|
wr %g0,%o0,%asi
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Load/store from/into alternate space
|
|
!
|
|
! float vis_ldfa_ASI_REG(void *rs1)
|
|
!
|
|
.inline vis_ldfa_ASI_REG,8
|
|
lda [%o0+0]%asi,%f4
|
|
fmovs %f4,%f0 ! Compiler can clean this up
|
|
.end
|
|
!
|
|
! float vis_ldfa_ASI_P(void *rs1)
|
|
!
|
|
.inline vis_ldfa_ASI_P,8
|
|
lda [%o0]0x80,%f4 ! ASI_P
|
|
fmovs %f4,%f0 ! Compiler can clean this up
|
|
.end
|
|
!
|
|
! float vis_ldfa_ASI_PL(void *rs1)
|
|
!
|
|
.inline vis_ldfa_ASI_PL,8
|
|
lda [%o0]0x88,%f4 ! ASI_PL
|
|
fmovs %f4,%f0 ! Compiler can clean this up
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_REG(void *rs1)
|
|
!
|
|
.inline vis_lddfa_ASI_REG,8
|
|
ldda [%o0+0]%asi,%f4
|
|
fmovd %f4,%f0 ! Compiler can clean this up
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_P(void *rs1)
|
|
!
|
|
.inline vis_lddfa_ASI_P,8
|
|
ldda [%o0]0x80,%f4 ! ASI_P
|
|
fmovd %f4,%f0 ! Compiler can clean this up
|
|
.end
|
|
!
|
|
! double vis_lddfa_ASI_PL(void *rs1)
|
|
!
|
|
.inline vis_lddfa_ASI_PL,8
|
|
ldda [%o0]0x88,%f4 ! ASI_PL
|
|
fmovd %f4,%f0 ! Compiler can clean this up
|
|
.end
|
|
!
|
|
! vis_stfa_ASI_REG(float frs, void *rs1)
|
|
!
|
|
.inline vis_stfa_ASI_REG,12
|
|
sta %f1,[%o1+0]%asi
|
|
.end
|
|
!
|
|
! vis_stfa_ASI_P(float frs, void *rs1)
|
|
!
|
|
.inline vis_stfa_ASI_P,12
|
|
sta %f1,[%o1]0x80 ! ASI_P
|
|
.end
|
|
!
|
|
! vis_stfa_ASI_PL(float frs, void *rs1)
|
|
!
|
|
.inline vis_stfa_ASI_PL,12
|
|
sta %f1,[%o1]0x88 ! ASI_PL
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_REG(double frd, void *rs1)
|
|
!
|
|
.inline vis_stdfa_ASI_REG,16
|
|
stda %f0,[%o1+0]%asi
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_P(double frd, void *rs1)
|
|
!
|
|
.inline vis_stdfa_ASI_P,16
|
|
stda %f0,[%o1]0x80 ! ASI_P
|
|
.end
|
|
!
|
|
! vis_stdfa_ASI_PL(double frd, void *rs1)
|
|
!
|
|
.inline vis_stdfa_ASI_PL,16
|
|
stda %f0,[%o1]0x88 ! ASI_PL
|
|
.end
|
|
!
|
|
! unsigned short vis_lduha_ASI_REG(void *rs1)
|
|
!
|
|
.inline vis_lduha_ASI_REG,8
|
|
lduha [%o0+0]%asi,%o0
|
|
.end
|
|
!
|
|
! unsigned short vis_lduha_ASI_P(void *rs1)
|
|
!
|
|
.inline vis_lduha_ASI_P,8
|
|
lduha [%o0]0x80,%o0 ! ASI_P
|
|
.end
|
|
!
|
|
! unsigned short vis_lduha_ASI_PL(void *rs1)
|
|
!
|
|
.inline vis_lduha_ASI_PL,8
|
|
lduha [%o0]0x88,%o0 ! ASI_PL
|
|
.end
|
|
!
|
|
! unsigned short vis_lduha_ASI_P_index(void *rs1, long index)
|
|
!
|
|
.inline vis_lduha_ASI_P_index,16
|
|
lduha [%o0+%o1]0x80,%o0 ! ASI_P
|
|
.end
|
|
!
|
|
! unsigned short vis_lduha_ASI_PL_index(void *rs1, long index)
|
|
!
|
|
.inline vis_lduha_ASI_PL_index,16
|
|
lduha [%o0+%o1]0x88,%o0 ! ASI_PL
|
|
.end
|
|
|
|
!--------------------------------------------------------------------
|
|
! Prefetch
|
|
!
|
|
! void vis_prefetch_read(void * /*address*/);
|
|
!
|
|
.inline vis_prefetch_read,8
|
|
prefetch [%o0+0],0
|
|
.end
|
|
!
|
|
! void vis_prefetch_write(void * /*address*/);
|
|
!
|
|
.inline vis_prefetch_write,8
|
|
prefetch [%o0+0],2
|
|
.end
|