Skip to content
GitLab
Explore
Projects
Groups
Topics
Snippets
Projects
Groups
Topics
Snippets
/
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
BC
public
external
libvpx
Commits
4cefb443
Commit
4cefb443
authored
14 years ago
by
Yunqing Wang
Browse files
Options
Download
Patches
Plain Diff
Add MMWORD PTR/XMMWORD PTR in subtract_sse2.asm
Change-Id: Ia649b500ef020225d8bbf611799d0f47658dc2ac
parent
31752f2f
v1.14.0-linphone
1.4.X
aylesbury
bali
cayuga
eider
experimental
feature/update_to_v1.9.0-linphone
feature/uwp_nuget
forest
frame_parallel
highbitdepth
indianrunnerduck
javanwhistlingduck
khakicampbell
linphone
linphone-android
linphone-old
longtailedduck
m29-baseline
m31-baseline
m49-2623
m52-2743
m54-2840
m56-2924
m66-3359
m68-3440
mandarinduck
mcw
mcw2
nextgen
nextgenv2
pcs-2013
playground
sandbox/Jingning/experimental
sandbox/Jingning/transcode
sandbox/Jingning/vpx
sandbox/aconverse@google.com/ansbench
sandbox/atna/dec_sem_sync
sandbox/awatry/initial_opencl_implementation
sandbox/debargha/playground
sandbox/hkuang/frame_parallel
sandbox/hkuang@google.com/decode
sandbox/hlundin/error-concealment
sandbox/holmer/error-concealment
sandbox/jimbankoski@google.com/proposed-aom
sandbox/jingning@google.com/decoder_test_suite
sandbox/jingning@google.com/experimental
sandbox/jkoleszar/cached-multibit
sandbox/jkoleszar/experimental-knobs
sandbox/jkoleszar/new-rate-control
sandbox/jkoleszar/new-rtcd
sandbox/jkoleszar/reuse-modemv
sandbox/jkoleszar/use-memcpy
sandbox/jzern@google.com/test
sandbox/slavarnway/test
sandbox/wangch@google.com/vp9
sandbox/yaowu@google.com/mergeaom
stable-vp9-decoder
v1.12.0-linphone
v1.6.1_linphone
v1.7.0-linphone
v1.9.0-linphone
vp9-preview
v1.9.0
v1.9.0-rc1
v1.8.2
v1.8.1
v1.8.0
v1.7.0
v1.6.1
v1.6.0
v1.5.0
v1.4.0
v1.3.0
v1.2.0
v1.1.0
v1.0.0
v0.9.7
v0.9.7-p1
v0.9.6
v0.9.5
No related merge requests found
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
vp8/encoder/x86/subtract_sse2.asm
+52
-52
vp8/encoder/x86/subtract_sse2.asm
with
52 additions
and
52 deletions
vp8/encoder/x86/subtract_sse2.asm
+
52
−
52
View file @
4cefb443
...
@@ -35,21 +35,21 @@ sym(vp8_subtract_b_sse2_impl):
...
@@ -35,21 +35,21 @@ sym(vp8_subtract_b_sse2_impl):
punpcklbw
mm0
,
mm7
punpcklbw
mm0
,
mm7
punpcklbw
mm1
,
mm7
punpcklbw
mm1
,
mm7
psubw
mm0
,
mm1
psubw
mm0
,
mm1
movq
[
rdi
],
mm0
movq
MMWORD
PTR
[
rdi
],
mm0
movd
mm0
,
[
rsi
+
rdx
]
movd
mm0
,
[
rsi
+
rdx
]
movd
mm1
,
[
rax
+
rcx
]
movd
mm1
,
[
rax
+
rcx
]
punpcklbw
mm0
,
mm7
punpcklbw
mm0
,
mm7
punpcklbw
mm1
,
mm7
punpcklbw
mm1
,
mm7
psubw
mm0
,
mm1
psubw
mm0
,
mm1
movq
[
rdi
+
rcx
*
2
],
mm0
movq
MMWORD
PTR
[
rdi
+
rcx
*
2
],
mm0
movd
mm0
,
[
rsi
+
rdx
*
2
]
movd
mm0
,
[
rsi
+
rdx
*
2
]
movd
mm1
,
[
rax
+
rcx
*
2
]
movd
mm1
,
[
rax
+
rcx
*
2
]
punpcklbw
mm0
,
mm7
punpcklbw
mm0
,
mm7
punpcklbw
mm1
,
mm7
punpcklbw
mm1
,
mm7
psubw
mm0
,
mm1
psubw
mm0
,
mm1
movq
[
rdi
+
rcx
*
4
],
mm0
movq
MMWORD
PTR
[
rdi
+
rcx
*
4
],
mm0
lea
rsi
,
[
rsi
+
rdx
*
2
]
lea
rsi
,
[
rsi
+
rdx
*
2
]
lea
rcx
,
[
rcx
+
rcx
*
2
]
lea
rcx
,
[
rcx
+
rcx
*
2
]
...
@@ -59,7 +59,7 @@ sym(vp8_subtract_b_sse2_impl):
...
@@ -59,7 +59,7 @@ sym(vp8_subtract_b_sse2_impl):
punpcklbw
mm0
,
mm7
punpcklbw
mm0
,
mm7
punpcklbw
mm1
,
mm7
punpcklbw
mm1
,
mm7
psubw
mm0
,
mm1
psubw
mm0
,
mm1
movq
[
rdi
+
rcx
*
2
],
mm0
movq
MMWORD
PTR
[
rdi
+
rcx
*
2
],
mm0
; begin epilog
; begin epilog
pop
rdi
pop
rdi
...
@@ -88,8 +88,8 @@ sym(vp8_subtract_mby_sse2):
...
@@ -88,8 +88,8 @@ sym(vp8_subtract_mby_sse2):
mov
rcx
,
8
; do two lines at one time
mov
rcx
,
8
; do two lines at one time
submby_loop:
submby_loop:
movdqa
xmm0
,
[
rsi
]
; src
movdqa
xmm0
,
XMMWORD
PTR
[
rsi
]
; src
movdqa
xmm1
,
[
rax
]
; pred
movdqa
xmm1
,
XMMWORD
PTR
[
rax
]
; pred
movdqa
xmm2
,
xmm0
movdqa
xmm2
,
xmm0
psubb
xmm0
,
xmm1
psubb
xmm0
,
xmm1
...
@@ -103,11 +103,11 @@ submby_loop:
...
@@ -103,11 +103,11 @@ submby_loop:
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
movdqa
[
rdi
],
xmm0
movdqa
XMMWORD
PTR
[
rdi
],
xmm0
movdqa
[
rdi
+
16
],
xmm2
movdqa
XMMWORD
PTR
[
rdi
+
16
],
xmm2
movdqa
xmm4
,
[
rsi
+
rdx
]
movdqa
xmm4
,
XMMWORD
PTR
[
rsi
+
rdx
]
movdqa
xmm5
,
[
rax
+
16
]
movdqa
xmm5
,
XMMWORD
PTR
[
rax
+
16
]
movdqa
xmm6
,
xmm4
movdqa
xmm6
,
xmm4
psubb
xmm4
,
xmm5
psubb
xmm4
,
xmm5
...
@@ -121,8 +121,8 @@ submby_loop:
...
@@ -121,8 +121,8 @@ submby_loop:
punpcklbw
xmm4
,
xmm5
; put sign back to subtraction
punpcklbw
xmm4
,
xmm5
; put sign back to subtraction
punpckhbw
xmm6
,
xmm7
; put sign back to subtraction
punpckhbw
xmm6
,
xmm7
; put sign back to subtraction
movdqa
[
rdi
+
32
],
xmm4
movdqa
XMMWORD
PTR
[
rdi
+
32
],
xmm4
movdqa
[
rdi
+
48
],
xmm6
movdqa
XMMWORD
PTR
[
rdi
+
48
],
xmm6
add
rdi
,
64
add
rdi
,
64
add
rax
,
32
add
rax
,
32
...
@@ -159,9 +159,9 @@ sym(vp8_subtract_mbuv_sse2):
...
@@ -159,9 +159,9 @@ sym(vp8_subtract_mbuv_sse2):
;u
;u
;line 0 1
;line 0 1
movq
xmm0
,
[
rsi
]
; src
movq
xmm0
,
MMWORD
PTR
[
rsi
]
; src
movq
xmm2
,
[
rsi
+
rdx
]
movq
xmm2
,
MMWORD
PTR
[
rsi
+
rdx
]
movdqa
xmm1
,
[
rax
]
; pred
movdqa
xmm1
,
XMMWORD
PTR
[
rax
]
; pred
punpcklqdq
xmm0
,
xmm2
punpcklqdq
xmm0
,
xmm2
movdqa
xmm2
,
xmm0
movdqa
xmm2
,
xmm0
...
@@ -176,13 +176,13 @@ sym(vp8_subtract_mbuv_sse2):
...
@@ -176,13 +176,13 @@ sym(vp8_subtract_mbuv_sse2):
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
movdqa
[
rdi
],
xmm0
movdqa
XMMWORD
PTR
[
rdi
],
xmm0
movdqa
[
rdi
+
16
],
xmm2
movdqa
XMMWORD
PTR
[
rdi
+
16
],
xmm2
;line 2 3
;line 2 3
movq
xmm0
,
[
rsi
+
rdx
*
2
]
; src
movq
xmm0
,
MMWORD
PTR
[
rsi
+
rdx
*
2
]
; src
movq
xmm2
,
[
rsi
+
rcx
]
movq
xmm2
,
MMWORD
PTR
[
rsi
+
rcx
]
movdqa
xmm1
,
[
rax
+
16
]
; pred
movdqa
xmm1
,
XMMWORD
PTR
[
rax
+
16
]
; pred
punpcklqdq
xmm0
,
xmm2
punpcklqdq
xmm0
,
xmm2
movdqa
xmm2
,
xmm0
movdqa
xmm2
,
xmm0
...
@@ -197,15 +197,15 @@ sym(vp8_subtract_mbuv_sse2):
...
@@ -197,15 +197,15 @@ sym(vp8_subtract_mbuv_sse2):
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
movdqa
[
rdi
+
32
],
xmm0
movdqa
XMMWORD
PTR
[
rdi
+
32
],
xmm0
movdqa
[
rdi
+
48
],
xmm2
movdqa
XMMWORD
PTR
[
rdi
+
48
],
xmm2
;line 4 5
;line 4 5
lea
rsi
,
[
rsi
+
rdx
*
4
]
lea
rsi
,
[
rsi
+
rdx
*
4
]
movq
xmm0
,
[
rsi
]
; src
movq
xmm0
,
MMWORD
PTR
[
rsi
]
; src
movq
xmm2
,
[
rsi
+
rdx
]
movq
xmm2
,
MMWORD
PTR
[
rsi
+
rdx
]
movdqa
xmm1
,
[
rax
+
32
]
; pred
movdqa
xmm1
,
XMMWORD
PTR
[
rax
+
32
]
; pred
punpcklqdq
xmm0
,
xmm2
punpcklqdq
xmm0
,
xmm2
movdqa
xmm2
,
xmm0
movdqa
xmm2
,
xmm0
...
@@ -220,13 +220,13 @@ sym(vp8_subtract_mbuv_sse2):
...
@@ -220,13 +220,13 @@ sym(vp8_subtract_mbuv_sse2):
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
movdqa
[
rdi
+
64
],
xmm0
movdqa
XMMWORD
PTR
[
rdi
+
64
],
xmm0
movdqa
[
rdi
+
80
],
xmm2
movdqa
XMMWORD
PTR
[
rdi
+
80
],
xmm2
;line 6 7
;line 6 7
movq
xmm0
,
[
rsi
+
rdx
*
2
]
; src
movq
xmm0
,
MMWORD
PTR
[
rsi
+
rdx
*
2
]
; src
movq
xmm2
,
[
rsi
+
rcx
]
movq
xmm2
,
MMWORD
PTR
[
rsi
+
rcx
]
movdqa
xmm1
,
[
rax
+
48
]
; pred
movdqa
xmm1
,
XMMWORD
PTR
[
rax
+
48
]
; pred
punpcklqdq
xmm0
,
xmm2
punpcklqdq
xmm0
,
xmm2
movdqa
xmm2
,
xmm0
movdqa
xmm2
,
xmm0
...
@@ -241,8 +241,8 @@ sym(vp8_subtract_mbuv_sse2):
...
@@ -241,8 +241,8 @@ sym(vp8_subtract_mbuv_sse2):
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
movdqa
[
rdi
+
96
],
xmm0
movdqa
XMMWORD
PTR
[
rdi
+
96
],
xmm0
movdqa
[
rdi
+
112
],
xmm2
movdqa
XMMWORD
PTR
[
rdi
+
112
],
xmm2
;v
;v
mov
rsi
,
arg
(
2
)
;z = vsrc
mov
rsi
,
arg
(
2
)
;z = vsrc
...
@@ -250,9 +250,9 @@ sym(vp8_subtract_mbuv_sse2):
...
@@ -250,9 +250,9 @@ sym(vp8_subtract_mbuv_sse2):
add
rax
,
64
;Predictor = pred + 320
add
rax
,
64
;Predictor = pred + 320
;line 0 1
;line 0 1
movq
xmm0
,
[
rsi
]
; src
movq
xmm0
,
MMWORD
PTR
[
rsi
]
; src
movq
xmm2
,
[
rsi
+
rdx
]
movq
xmm2
,
MMWORD
PTR
[
rsi
+
rdx
]
movdqa
xmm1
,
[
rax
]
; pred
movdqa
xmm1
,
XMMWORD
PTR
[
rax
]
; pred
punpcklqdq
xmm0
,
xmm2
punpcklqdq
xmm0
,
xmm2
movdqa
xmm2
,
xmm0
movdqa
xmm2
,
xmm0
...
@@ -267,13 +267,13 @@ sym(vp8_subtract_mbuv_sse2):
...
@@ -267,13 +267,13 @@ sym(vp8_subtract_mbuv_sse2):
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
movdqa
[
rdi
],
xmm0
movdqa
XMMWORD
PTR
[
rdi
],
xmm0
movdqa
[
rdi
+
16
],
xmm2
movdqa
XMMWORD
PTR
[
rdi
+
16
],
xmm2
;line 2 3
;line 2 3
movq
xmm0
,
[
rsi
+
rdx
*
2
]
; src
movq
xmm0
,
MMWORD
PTR
[
rsi
+
rdx
*
2
]
; src
movq
xmm2
,
[
rsi
+
rcx
]
movq
xmm2
,
MMWORD
PTR
[
rsi
+
rcx
]
movdqa
xmm1
,
[
rax
+
16
]
; pred
movdqa
xmm1
,
XMMWORD
PTR
[
rax
+
16
]
; pred
punpcklqdq
xmm0
,
xmm2
punpcklqdq
xmm0
,
xmm2
movdqa
xmm2
,
xmm0
movdqa
xmm2
,
xmm0
...
@@ -288,15 +288,15 @@ sym(vp8_subtract_mbuv_sse2):
...
@@ -288,15 +288,15 @@ sym(vp8_subtract_mbuv_sse2):
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
movdqa
[
rdi
+
32
],
xmm0
movdqa
XMMWORD
PTR
[
rdi
+
32
],
xmm0
movdqa
[
rdi
+
48
],
xmm2
movdqa
XMMWORD
PTR
[
rdi
+
48
],
xmm2
;line 4 5
;line 4 5
lea
rsi
,
[
rsi
+
rdx
*
4
]
lea
rsi
,
[
rsi
+
rdx
*
4
]
movq
xmm0
,
[
rsi
]
; src
movq
xmm0
,
MMWORD
PTR
[
rsi
]
; src
movq
xmm2
,
[
rsi
+
rdx
]
movq
xmm2
,
MMWORD
PTR
[
rsi
+
rdx
]
movdqa
xmm1
,
[
rax
+
32
]
; pred
movdqa
xmm1
,
XMMWORD
PTR
[
rax
+
32
]
; pred
punpcklqdq
xmm0
,
xmm2
punpcklqdq
xmm0
,
xmm2
movdqa
xmm2
,
xmm0
movdqa
xmm2
,
xmm0
...
@@ -311,13 +311,13 @@ sym(vp8_subtract_mbuv_sse2):
...
@@ -311,13 +311,13 @@ sym(vp8_subtract_mbuv_sse2):
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
movdqa
[
rdi
+
64
],
xmm0
movdqa
XMMWORD
PTR
[
rdi
+
64
],
xmm0
movdqa
[
rdi
+
80
],
xmm2
movdqa
XMMWORD
PTR
[
rdi
+
80
],
xmm2
;line 6 7
;line 6 7
movq
xmm0
,
[
rsi
+
rdx
*
2
]
; src
movq
xmm0
,
MMWORD
PTR
[
rsi
+
rdx
*
2
]
; src
movq
xmm2
,
[
rsi
+
rcx
]
movq
xmm2
,
MMWORD
PTR
[
rsi
+
rcx
]
movdqa
xmm1
,
[
rax
+
48
]
; pred
movdqa
xmm1
,
XMMWORD
PTR
[
rax
+
48
]
; pred
punpcklqdq
xmm0
,
xmm2
punpcklqdq
xmm0
,
xmm2
movdqa
xmm2
,
xmm0
movdqa
xmm2
,
xmm0
...
@@ -332,8 +332,8 @@ sym(vp8_subtract_mbuv_sse2):
...
@@ -332,8 +332,8 @@ sym(vp8_subtract_mbuv_sse2):
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpcklbw
xmm0
,
xmm1
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
punpckhbw
xmm2
,
xmm3
; put sign back to subtraction
movdqa
[
rdi
+
96
],
xmm0
movdqa
XMMWORD
PTR
[
rdi
+
96
],
xmm0
movdqa
[
rdi
+
112
],
xmm2
movdqa
XMMWORD
PTR
[
rdi
+
112
],
xmm2
; begin epilog
; begin epilog
pop
rdi
pop
rdi
...
...
This diff is collapsed.
Click to expand it.
Preview
Supports
Markdown
0%
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment
Menu
Explore
Projects
Groups
Topics
Snippets