mirrors/flac
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fix bug in SSE version of ...autocorrelation...

Browse Source
This commit is contained in:
Josh Coalson 2001-05-30 19:08:50 +00:00
parent 714216128b
commit 7fc68ffc81
1 changed files with 28 additions and 35 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

63
src/libFLAC/i386/lpc_asm.s
View File

@ -57,10 +57,10 @@ cglobal FLAC__lpc_restore_signal_asm_i386_mmx
ALIGN 16
cident FLAC__lpc_compute_autocorrelation_asm_i386
; esp + 32 == autoc[]
; esp + 28 == lag
; esp + 24 == data_len
; esp + 20 == data[]
;[esp + 32] == autoc[]
;[esp + 28] == lag
;[esp + 24] == data_len
;[esp + 20] == data[]
push ebp
push ebx
@ -208,10 +208,10 @@ cident FLAC__lpc_compute_autocorrelation_asm_i386
ALIGN 16
cident FLAC__lpc_compute_autocorrelation_asm_i386_sse
; esp + 16 == autoc[]
; esp + 12 == lag
; esp + 8 == data_len
; esp + 4 == data[]
;[esp + 16] == autoc[]
;[esp + 12] == lag
;[esp + 8] == data_len
;[esp + 4] == data[]
; for(coeff = 0; coeff < lag; coeff++)
; autoc[coeff] = 0.0;
@ -227,45 +227,38 @@ cident FLAC__lpc_compute_autocorrelation_asm_i386_sse
shufps xmm0, xmm0, 0 ; xmm0 = data[0],data[0],data[0],data[0]
movaps xmm1, xmm0 ; xmm1 = data[0],data[0],data[0],data[0]
xorps xmm3, xmm3 ; xmm3 = 0,0,0,0
.warmup: ; xmm3:xmm2 = data[sample-[7..0]]
movaps xmm4, xmm0
movaps xmm5, xmm1 ; xmm5:xmm4 = xmm1:xmm0 = data[sample]*8
mulps xmm4, xmm2
mulps xmm5, xmm3 ; xmm5:xmm4 = xmm1:xmm0 * xmm3:xmm2
addps xmm6, xmm4
addps xmm7, xmm5 ; xmm7:xmm6 += xmm1:xmm0 * xmm3:xmm2
.warmup: ; xmm3:xmm2 == data[sample-7],data[sample-6],...,data[sample]
mulps xmm0, xmm2
mulps xmm1, xmm3 ; xmm5:xmm4 = xmm1:xmm0 * xmm3:xmm2
addps xmm6, xmm0
addps xmm7, xmm1 ; xmm7:xmm6 += xmm1:xmm0 * xmm3:xmm2
dec edx
;* there's no need to even check for this because we know that lag == 8
;* and data_len >= lag, so our 1-sample warmup cannot finish the loop
; jz .loop_end
ALIGN 16
.loop_8:
; read the next sample
; start by reading the next sample
movss xmm0, [eax] ; xmm0 = 0,0,0,data[sample]
add eax, 4
shufps xmm0, xmm0, 0 ; xmm0 = data[sample],data[sample],data[sample],data[sample]
movaps xmm1, xmm0 ; xmm1 = data[sample],data[sample],data[sample],data[sample]
; now shift the lagged samples
movaps xmm4, xmm2
movaps xmm5, xmm3
shufps xmm2, xmm4, 93h ; 93h=2-1-0-3 => xmm2 gets rotated left by one float
shufps xmm3, xmm5, 93h ; 93h=2-1-0-3 => xmm3 gets rotated left by one float
movss xmm3, xmm2
movss xmm2, xmm0
movaps xmm4, xmm0
movaps xmm5, xmm1 ; xmm5:xmm4 = xmm1:xmm0 = data[sample]*8
mulps xmm4, xmm2
mulps xmm5, xmm3 ; xmm5:xmm4 = xmm1:xmm0 * xmm3:xmm2
addps xmm6, xmm4
addps xmm7, xmm5 ; xmm7:xmm6 += xmm1:xmm0 * xmm3:xmm2
; here we reorder the instructions; see the (#) indexes for a logical order
shufps xmm2, xmm2, 93h ; (3) 93h=2-1-0-3 => xmm2 gets rotated left by one float
add eax, 4 ; (0)
shufps xmm3, xmm3, 93h ; (4) 93h=2-1-0-3 => xmm3 gets rotated left by one float
shufps xmm0, xmm0, 0 ; (1) xmm0 = data[sample],data[sample],data[sample],data[sample]
movss xmm3, xmm2 ; (5)
movaps xmm1, xmm0 ; (2) xmm1 = data[sample],data[sample],data[sample],data[sample]
movss xmm2, xmm0 ; (6)
mulps xmm1, xmm3 ; (8)
mulps xmm0, xmm2 ; (7) xmm5:xmm4 = xmm1:xmm0 * xmm3:xmm2
addps xmm7, xmm1 ; (10)
addps xmm6, xmm0 ; (9) xmm7:xmm6 += xmm1:xmm0 * xmm3:xmm2
dec edx
jnz .loop_8
.loop_end:
; store autoc
mov edx, [esp + 16] ; edx == autoc
movups xmm6, [edx]
movups xmm7, [edx + 4]
movups [edx], xmm6
movups [edx + 4], xmm7
.end:
ret