diff options
Diffstat (limited to 'libavcodec/aarch64/vp9lpf_16bpp_neon.S')
| -rw-r--r-- | libavcodec/aarch64/vp9lpf_16bpp_neon.S | 38 |
1 files changed, 19 insertions, 19 deletions
diff --git a/libavcodec/aarch64/vp9lpf_16bpp_neon.S b/libavcodec/aarch64/vp9lpf_16bpp_neon.S index 9869614a29..a092617b92 100644 --- a/libavcodec/aarch64/vp9lpf_16bpp_neon.S +++ b/libavcodec/aarch64/vp9lpf_16bpp_neon.S @@ -57,7 +57,7 @@ mov x12, v4.d[1] adds x11, x11, x12 b.ne 1f - br x10 + ret x10 1: .if \wd >= 8 @@ -193,7 +193,7 @@ b.eq 6f .else b.ne 1f - br x13 + ret x13 1: .endif @@ -252,7 +252,7 @@ b.ne 1f // If no pixels needed flat8in nor flat8out, jump to a // writeout of the inner 4 pixels - br x14 + ret x14 1: mov x11, v7.d[0] @@ -260,7 +260,7 @@ adds x11, x11, x12 b.ne 1f // If no pixels need flat8out, jump to a writeout of the inner 6 pixels - br x15 + ret x15 1: // flat8out @@ -434,7 +434,7 @@ function ff_\func\()_\bpp\()_neon, export=1 ldp d10, d11, [sp], 0x10 ldp d12, d13, [sp], 0x10 ldp d14, d15, [sp], 0x10 - br x16 + ret x16 .else b \func\()_16_neon .endif @@ -474,7 +474,7 @@ function ff_\func\()_\suffix\()_\bpp\()_neon, export=1 ldp d12, d13, [sp], 0x10 ldp d14, d15, [sp], 0x10 .endif - br x16 + ret x16 endfunc .endm @@ -508,7 +508,7 @@ function ff_vp9_loop_filter_\dir\()_\wd1\()\wd2\()_16_\bpp\()_neon, export=1 lsl w3, w14, #\bpp - 8 lsl w4, w15, #\bpp - 8 bl vp9_loop_filter_\dir\()_\wd2\()_8_16_neon - br x16 + ret x16 endfunc .endm @@ -541,7 +541,7 @@ function vp9_loop_filter_v_4_8_16_neon st1 {v25.8h}, [x0], x1 sub x0, x0, x1, lsl #1 - br x10 + ret x10 endfunc bpp_frontends vp9_loop_filter_v_4_8 @@ -589,7 +589,7 @@ function vp9_loop_filter_h_4_8_16_neon sub x0, x0, x1, lsl #3 add x0, x0, #4 - br x10 + ret x10 endfunc bpp_frontends vp9_loop_filter_h_4_8 @@ -620,7 +620,7 @@ function vp9_loop_filter_v_8_8_16_neon sub x0, x0, x1, lsl #1 sub x0, x0, x1 - br x10 + ret x10 6: sub x9, x0, x1, lsl #1 st1 {v22.8h}, [x9], x1 @@ -628,7 +628,7 @@ function vp9_loop_filter_v_8_8_16_neon st1 {v23.8h}, [x9], x1 st1 {v25.8h}, [x0], x1 sub x0, x0, x1, lsl #1 - br x10 + ret x10 endfunc bpp_frontends vp9_loop_filter_v_8_8 @@ -671,7 +671,7 @@ function vp9_loop_filter_h_8_8_16_neon sub x0, x0, x1, lsl #3 add x0, x0, #8 - br x10 + ret x10 6: // If we didn't need to do the flat8in part, we use the same writeback // as in loop_filter_h_4_8. @@ -688,7 +688,7 @@ function vp9_loop_filter_h_8_8_16_neon st1 {v25.d}[1], [x0], x1 sub x0, x0, x1, lsl #3 add x0, x0, #4 - br x10 + ret x10 endfunc bpp_frontends vp9_loop_filter_h_8_8 @@ -743,7 +743,7 @@ function vp9_loop_filter_v_16_8_16_neon sub x0, x0, x1, lsl #3 add x0, x0, x1 - br x10 + ret x10 8: add x9, x9, x1, lsl #2 // If we didn't do the flat8out part, the output is left in the @@ -756,7 +756,7 @@ function vp9_loop_filter_v_16_8_16_neon st1 {v26.8h}, [x0], x1 sub x0, x0, x1, lsl #1 sub x0, x0, x1 - br x10 + ret x10 7: sub x9, x0, x1, lsl #1 st1 {v22.8h}, [x9], x1 @@ -764,7 +764,7 @@ function vp9_loop_filter_v_16_8_16_neon st1 {v23.8h}, [x9], x1 st1 {v25.8h}, [x0], x1 sub x0, x0, x1, lsl #1 - br x10 + ret x10 endfunc bpp_frontends vp9_loop_filter_v_16_8, push=1 @@ -821,7 +821,7 @@ function vp9_loop_filter_h_16_8_16_neon st1 {v31.8h}, [x0], x1 sub x0, x0, x1, lsl #3 - br x10 + ret x10 8: // The same writeback as in loop_filter_h_8_8 sub x9, x0, #8 @@ -838,7 +838,7 @@ function vp9_loop_filter_h_16_8_16_neon st1 {v27.8h}, [x0], x1 sub x0, x0, x1, lsl #3 add x0, x0, #8 - br x10 + ret x10 7: // The same writeback as in loop_filter_h_4_8 sub x9, x0, #4 @@ -854,7 +854,7 @@ function vp9_loop_filter_h_16_8_16_neon st1 {v25.d}[1], [x0], x1 sub x0, x0, x1, lsl #3 add x0, x0, #4 - br x10 + ret x10 endfunc bpp_frontends vp9_loop_filter_h_16_8, push=1 |