diff options
author | Hans-Christoph Steiner <eighthave@users.sourceforge.net> | 2005-12-16 01:05:40 +0000 |
---|---|---|
committer | Hans-Christoph Steiner <eighthave@users.sourceforge.net> | 2005-12-16 01:05:40 +0000 |
commit | b694c274836ac8b04d644711ac324eac2e9ab83e (patch) | |
tree | 36b6a5c17f7e1f414f80697210c2ed3e8005035b /system/mmx/pixel_gain_s16.s | |
parent | e28a07fba67af0af818dda6afa4cf67c09700816 (diff) |
checking in pdp 0.12.4 from http://zwizwa.fartit.com/pd/pdp/pdp-0.12.4.tar.gz
svn path=/trunk/externals/pdp/; revision=4232
Diffstat (limited to 'system/mmx/pixel_gain_s16.s')
-rw-r--r-- | system/mmx/pixel_gain_s16.s | 71 |
1 files changed, 71 insertions, 0 deletions
diff --git a/system/mmx/pixel_gain_s16.s b/system/mmx/pixel_gain_s16.s new file mode 100644 index 0000000..adcfdf5 --- /dev/null +++ b/system/mmx/pixel_gain_s16.s @@ -0,0 +1,71 @@ +# Pure Data Packet mmx routine. +# Copyright (c) by Tom Schouten <pdp@zzz.kotnet.org> +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write to the Free Software +# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. +# +.globl pixel_gain_s16 +.type pixel_gain_s16,@function + +# gain is integer, shift count is down +# void pixel_gain_s16(int *buf, int nb_8pixel_vectors, short int gain[4], unsigned long long *shift) + +pixel_gain_s16: + pushl %ebp + movl %esp, %ebp + push %esi + push %edi + + movl 20(%ebp), %edi + movq (%edi), %mm6 # get shift vector + + movl 16(%ebp), %edi + movq (%edi), %mm7 # get gain vector + + movl 8(%ebp), %esi # input array + movl 12(%ebp), %ecx # pixel count + + + .align 16 + .loop_gain: + + movq (%esi), %mm0 # load 4 pixels from memory + movq %mm0, %mm1 + pmulhw %mm7, %mm1 # apply gain (s15.0) fixed point, high word + pmullw %mm7, %mm0 # low word + + movq %mm0, %mm2 # copy + movq %mm1, %mm3 + + punpcklwd %mm1, %mm0 # unpack lsw components + punpckhwd %mm3, %mm2 # unpack msw components + + psrad %mm6, %mm0 # apply signed shift + psrad %mm6, %mm2 + + packssdw %mm2, %mm0 # pack result & saturate + movq %mm0, (%esi) # store result + + + addl $8, %esi # increment source pointer + decl %ecx + jnz .loop_gain # loop + + emms + + pop %edi + pop %esi + leave + ret + |