x86: replace explicit REP_RETs with RETs

From x86inc:
> On AMD cpus <=K10, an ordinary ret is slow if it immediately follows either
> a branch or a branch target. So switch to a 2-byte form of ret in that case.
> We can automatically detect "follows a branch", but not a branch target.
> (SSSE3 is a sufficient condition to know that your cpu doesn't have this problem.)

x86inc can automatically determine whether to use REP_RET rather than
REP in most of these cases, so impact is minimal. Additionally, a few
REP_RETs were used unnecessary, despite the return being nowhere near a
branch.

The only CPUs affected were AMD K10s, made between 2007 and 2011, 16
years ago and 12 years ago, respectively.

In the future, everyone involved with x86inc should consider dropping
REP_RETs altogether.

11 files changed, 22 insertions, 22 deletions

diff --git a/libavfilter/x86/af_volume.asm b/libavfilter/x86/af_volume.asm
index 723ab1f8fb..35a00784a2 100644
--- a/libavfilter/x86/af_volume.asm
+++ b/libavfilter/x86/af_volume.asm
@@ -56,7 +56,7 @@ cglobal scale_samples_s16, 4,4,4, dst, src, len, volume
     mova  [dstq+lenq], m3
     sub       lenq, mmsize
     jge .loop
-    REP_RET
+    RET
 
 ;------------------------------------------------------------------------------
 ; void ff_scale_samples_s32(uint8_t *dst, const uint8_t *src, int len,
@@ -93,7 +93,7 @@ cglobal scale_samples_s32, 4,4,4, dst, src, len, volume
 %endif
     sub            lenq, mmsize
     jge .loop
-    REP_RET
+    RET
 %endmacro
 
 INIT_XMM sse2
@@ -137,4 +137,4 @@ cglobal scale_samples_s32, 4,4,8, dst, src, len, volume
     mova  [dstq+lenq], m0
     sub       lenq, mmsize
     jge .loop
-    REP_RET
+    RET
diff --git a/libavfilter/x86/avf_showcqt.asm b/libavfilter/x86/avf_showcqt.asm
index 63e58408cd..16af0de9b0 100644
--- a/libavfilter/x86/avf_showcqt.asm
+++ b/libavfilter/x86/avf_showcqt.asm
@@ -127,7 +127,7 @@ cglobal showcqt_cqt_calc, 5, 10, 12, dst, src, coeffs, len, fft_len, x, coeffs_v
         lea     dstq, [dstq + 16]
         lea     coeffsq, [coeffsq + 2*Coeffs.sizeof]
         jnz     .loop_k
-        REP_RET
+        RET
         align   16
         .check_loop_a:
         cmp     xd, [coeffsq + Coeffs.len]
@@ -170,7 +170,7 @@ cglobal showcqt_cqt_calc, 4, 7, 8, dst, src, coeffs, len, x, coeffs_val, i
         lea     dstq, [dstq + 8]
         lea     coeffsq, [coeffsq + Coeffs.sizeof]
         jnz     .loop_k
-        REP_RET
+        RET
 %endif ; ARCH_X86_64
 %endmacro ; DECLARE_CQT_CALC
 
diff --git a/libavfilter/x86/scene_sad.asm b/libavfilter/x86/scene_sad.asm
index d38d71ccca..bf7236b3a3 100644
--- a/libavfilter/x86/scene_sad.asm
+++ b/libavfilter/x86/scene_sad.asm
@@ -53,7 +53,7 @@ cglobal scene_sad, 6, 7, 2, src1, stride1, src2, stride2, width, end, x
 
     mov         r0q, r6mp
     movu      [r0q], m1      ; sum
-REP_RET
+RET
 %endmacro
 
 
diff --git a/libavfilter/x86/vf_blend.asm b/libavfilter/x86/vf_blend.asm
index 277b100e4d..362020ec95 100644
--- a/libavfilter/x86/vf_blend.asm
+++ b/libavfilter/x86/vf_blend.asm
@@ -63,7 +63,7 @@ cglobal blend_%1, 5, 7, %2, top, top_linesize, bottom, bottom_linesize, dst, end
     add          dstq, dst_linesizeq
     sub          endd, 1
     jg .nextrow
-REP_RET
+RET
 %endmacro
 
 %macro BLEND_SIMPLE 2-3 0
diff --git a/libavfilter/x86/vf_framerate.asm b/libavfilter/x86/vf_framerate.asm
index 7a30c870bd..b5505b4ff8 100644
--- a/libavfilter/x86/vf_framerate.asm
+++ b/libavfilter/x86/vf_framerate.asm
@@ -84,7 +84,7 @@ cglobal blend_frames%1, 5, 7, 5, src1, src1_linesize, src2, src2_linesize, dst,
     add      dstq, dst_linesizeq
     sub      endd, 1
     jg .nextrow
-REP_RET
+RET
 %endmacro
 
 
diff --git a/libavfilter/x86/vf_gradfun.asm b/libavfilter/x86/vf_gradfun.asm
index 3581f89fe8..d106d52100 100644
--- a/libavfilter/x86/vf_gradfun.asm
+++ b/libavfilter/x86/vf_gradfun.asm
@@ -64,7 +64,7 @@ cglobal gradfun_filter_line, 6, 6
     add       r0, 4
     jl .loop
 .end:
-    REP_RET
+    RET
 
 INIT_XMM ssse3
 cglobal gradfun_filter_line, 6, 6, 8
@@ -78,7 +78,7 @@ cglobal gradfun_filter_line, 6, 6, 8
     FILTER_LINE m4
     add        r0, 8
     jl .loop
-    REP_RET
+    RET
 
 %macro BLUR_LINE 1
 cglobal gradfun_blur_line_%1, 6, 6, 8
@@ -102,7 +102,7 @@ cglobal gradfun_blur_line_%1, 6, 6, 8
     mova   [r3+r0], m0
     add         r0, 16
     jl .loop
-    REP_RET
+    RET
 %endmacro
 
 INIT_XMM sse2
diff --git a/libavfilter/x86/vf_hqdn3d.asm b/libavfilter/x86/vf_hqdn3d.asm
index e3b1bdca53..2c0ca45571 100644
--- a/libavfilter/x86/vf_hqdn3d.asm
+++ b/libavfilter/x86/vf_hqdn3d.asm
@@ -97,7 +97,7 @@ ALIGN 16
     inc    xq
     jl .loop
     je .loop2
-    REP_RET
+    RET
 %endmacro ; HQDN3D_ROW
 
 HQDN3D_ROW 8
diff --git a/libavfilter/x86/vf_interlace.asm b/libavfilter/x86/vf_interlace.asm
index f4a405c754..c28f9fbe3e 100644
--- a/libavfilter/x86/vf_interlace.asm
+++ b/libavfilter/x86/vf_interlace.asm
@@ -73,7 +73,7 @@ SECTION .text
     jl .loop
 
 .end:
-    REP_RET
+    RET
 %endmacro
 
 %macro LOWPASS_LINE 0
@@ -146,7 +146,7 @@ cglobal lowpass_line_complex, 5, 5, 8, dst, h, src, mref, pref
     add srcq, mmsize
     sub hd, mmsize
     jg .loop
-REP_RET
+RET
 
 cglobal lowpass_line_complex_12, 5, 5, 8, 16, dst, h, src, mref, pref, clip_max
     movd m7, DWORD clip_maxm
@@ -208,7 +208,7 @@ cglobal lowpass_line_complex_12, 5, 5, 8, 16, dst, h, src, mref, pref, clip_max
     add srcq, 2*mmsize
     sub hd, mmsize
     jg .loop
-REP_RET
+RET
 %endmacro
 
 INIT_XMM sse2
diff --git a/libavfilter/x86/vf_maskedmerge.asm b/libavfilter/x86/vf_maskedmerge.asm
index 1028299087..d9bd4688fd 100644
--- a/libavfilter/x86/vf_maskedmerge.asm
+++ b/libavfilter/x86/vf_maskedmerge.asm
@@ -81,4 +81,4 @@ cglobal maskedmerge8, 5, 7, 8, bsrc, osrc, msrc, dst, blinesize, w, x
     add          dstq, dlinesizeq
     sub         hd, 1
     jg .nextrow
-REP_RET
+RET
diff --git a/libavfilter/x86/vf_stereo3d.asm b/libavfilter/x86/vf_stereo3d.asm
index a057e495f1..b6a293b18e 100644
--- a/libavfilter/x86/vf_stereo3d.asm
+++ b/libavfilter/x86/vf_stereo3d.asm
@@ -213,4 +213,4 @@ cglobal anaglyph, 3, 6, 8, 2*9*mmsize, dst, lsrc, rsrc, dst_linesize, o, cnt
     add         rsrcq, r_linesizeq
     sub       heightd, 1
     jg .nextrow
-REP_RET
+RET
diff --git a/libavfilter/x86/vf_w3fdif.asm b/libavfilter/x86/vf_w3fdif.asm
index 52628c38d7..3010469f97 100644
--- a/libavfilter/x86/vf_w3fdif.asm
+++ b/libavfilter/x86/vf_w3fdif.asm
@@ -38,7 +38,7 @@ cglobal w3fdif_scale, 3, 3, 2, 0, out_pixel, work_pixel, linesize
     add                 work_pixelq, mmsize*2
     sub                   linesized, mmsize/2
     jg .loop
-REP_RET
+RET
 
 cglobal w3fdif_simple_low, 4, 5, 6, 0, work_line, in_lines_cur0, coef, linesize, offset
     movd                  m1, [coefq]
@@ -63,7 +63,7 @@ cglobal w3fdif_simple_low, 4, 5, 6, 0, work_line, in_lines_cur0, coef, linesize,
     add                               offsetq, mmsize/2
     sub                             linesized, mmsize/2
     jg .loop
-REP_RET
+RET
 
 cglobal w3fdif_complex_low, 4, 7, 8, 0, work_line, in_lines_cur0, coef, linesize
     movq                  m0, [coefq]
@@ -99,7 +99,7 @@ cglobal w3fdif_complex_low, 4, 7, 8, 0, work_line, in_lines_cur0, coef, linesize
     add                               offsetq, mmsize/2
     sub                             linesized, mmsize/2
     jg .loop
-REP_RET
+RET
 
 %if ARCH_X86_64
 cglobal w3fdif_simple_high, 5, 9, 8, 0, work_line, in_lines_cur0, in_lines_adj0, coef, linesize
@@ -179,7 +179,7 @@ cglobal w3fdif_simple_high, 4, 7, 8, 0, work_line, in_lines_cur0, in_lines_adj0,
     add                               offsetq, mmsize/2
     sub                             linesized, mmsize/2
     jg .loop
-REP_RET
+RET
 
 %if ARCH_X86_64
 
@@ -254,6 +254,6 @@ cglobal w3fdif_complex_high, 5, 13, 10, 0, work_line, in_lines_cur0, in_lines_ad
     add                               offsetq, mmsize/2
     sub                             linesized, mmsize/2
     jg .loop
-REP_RET
+RET
 
 %endif