1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-11-21 10:55:51 +02:00

x86: Avoid using 'd' as an argument name

x86inc.asm adds defines for <argument_name>{b,w,d,q} which clashes with
the nasm d{b,w,d,q} pseudo-instructions for writing initialized data.
This commit is contained in:
Henrik Gramner 2024-03-16 15:48:28 +01:00
parent 5d7f234e7e
commit 782c4df28d
2 changed files with 18 additions and 18 deletions

View File

@ -113,15 +113,15 @@ cglobal rv34_idct_dc_noround, 1, 2, 0
movd %1, %2
%endmacro
INIT_MMX mmxext
cglobal rv34_idct_add, 3,3,0, d, s, b
ROW_TRANSFORM bq
COL_TRANSFORM [dq], mm0, [pw_col_coeffs+ 0], [pw_col_coeffs+ 8]
mova mm0, [pw_col_coeffs+ 0]
COL_TRANSFORM [dq+sq], mm4, mm0, [pw_col_coeffs+ 8]
mova mm4, [pw_col_coeffs+ 8]
lea dq, [dq + 2*sq]
COL_TRANSFORM [dq], mm6, mm0, mm4
COL_TRANSFORM [dq+sq], mm7, mm0, mm4
cglobal rv34_idct_add, 3, 3, 0, dst, s, b
ROW_TRANSFORM bq
COL_TRANSFORM [dstq], mm0, [pw_col_coeffs+ 0], [pw_col_coeffs+ 8]
mova mm0, [pw_col_coeffs+ 0]
COL_TRANSFORM [dstq+sq], mm4, mm0, [pw_col_coeffs+ 8]
mova mm4, [pw_col_coeffs+ 8]
lea dstq, [dstq + 2*sq]
COL_TRANSFORM [dstq], mm6, mm0, mm4
COL_TRANSFORM [dstq+sq], mm7, mm0, mm4
ret
; ff_rv34_idct_dc_add_sse4(uint8_t *dst, int stride, int dc);

View File

@ -33,7 +33,7 @@ pw_257: times 8 dw 257
SECTION .text
INIT_XMM sse4
cglobal overlay_row_44, 5, 7, 6, 0, d, da, s, a, w, r, x
cglobal overlay_row_44, 5, 7, 6, 0, dst, da, s, a, w, r, x
xor xq, xq
movsxdifnidn wq, wd
mov rq, wq
@ -47,7 +47,7 @@ cglobal overlay_row_44, 5, 7, 6, 0, d, da, s, a, w, r, x
.loop:
pmovzxbw m0, [sq+xq]
pmovzxbw m2, [aq+xq]
pmovzxbw m1, [dq+xq]
pmovzxbw m1, [dstq+xq]
pmullw m0, m2
pxor m2, m3
pmullw m1, m2
@ -55,7 +55,7 @@ cglobal overlay_row_44, 5, 7, 6, 0, d, da, s, a, w, r, x
paddw m0, m1
pmulhuw m0, m5
packuswb m0, m0
movq [dq+xq], m0
movq [dstq+xq], m0
add xq, mmsize/2
cmp xq, wq
jl .loop
@ -65,7 +65,7 @@ cglobal overlay_row_44, 5, 7, 6, 0, d, da, s, a, w, r, x
RET
INIT_XMM sse4
cglobal overlay_row_22, 5, 7, 6, 0, d, da, s, a, w, r, x
cglobal overlay_row_22, 5, 7, 6, 0, dst, da, s, a, w, r, x
xor xq, xq
movsxdifnidn wq, wd
sub wq, 1
@ -85,7 +85,7 @@ cglobal overlay_row_22, 5, 7, 6, 0, d, da, s, a, w, r, x
pavgw m2, m1
pavgw m2, m1
psrlw m2, 8
pmovzxbw m1, [dq+xq]
pmovzxbw m1, [dstq+xq]
pmullw m0, m2
pxor m2, m3
pmullw m1, m2
@ -93,7 +93,7 @@ cglobal overlay_row_22, 5, 7, 6, 0, d, da, s, a, w, r, x
paddw m0, m1
pmulhuw m0, m5
packuswb m0, m0
movq [dq+xq], m0
movq [dstq+xq], m0
add xq, mmsize/2
cmp xq, wq
jl .loop
@ -103,7 +103,7 @@ cglobal overlay_row_22, 5, 7, 6, 0, d, da, s, a, w, r, x
RET
INIT_XMM sse4
cglobal overlay_row_20, 6, 7, 7, 0, d, da, s, a, w, r, x
cglobal overlay_row_20, 6, 7, 7, 0, dst, da, s, a, w, r, x
mov daq, aq
add daq, rmp
xor xq, xq
@ -126,7 +126,7 @@ cglobal overlay_row_20, 6, 7, 7, 0, d, da, s, a, w, r, x
pmaddubsw m1, m6
paddw m2, m1
psrlw m2, 2
pmovzxbw m1, [dq+xq]
pmovzxbw m1, [dstq+xq]
pmullw m0, m2
pxor m2, m3
pmullw m1, m2
@ -134,7 +134,7 @@ cglobal overlay_row_20, 6, 7, 7, 0, d, da, s, a, w, r, x
paddw m0, m1
pmulhuw m0, m5
packuswb m0, m0
movq [dq+xq], m0
movq [dstq+xq], m0
add xq, mmsize/2
cmp xq, wq
jl .loop