pixman_composite_over_n_8_0565_process_pixblock_tail
vst1.16 {d28, d29}, [DST_W, :128]!
vld1.16 {d4, d5}, [DST_R, :128]!
- vld1.8 {d24}, [MASK]!
+ fetch_mask_pixblock
cache_preload 8, 8
pixman_composite_over_n_8_0565_process_pixblock_head
.endm
pixman_composite_over_n_8_0565_process_pixblock_tail
fetch_src_pixblock
cache_preload 8, 8
- vld1.8 {d24}, [MASK]!
+ fetch_mask_pixblock
pixman_composite_over_n_8_0565_process_pixblock_head
vst1.16 {d28, d29}, [DST_W, :128]!
.endm
pixman_composite_over_n_8_8888_process_pixblock_tail
vst4.8 {d28, d29, d30, d31}, [DST_W, :128]!
vld4.8 {d4, d5, d6, d7}, [DST_R, :128]!
- vld1.8 {d24}, [MASK]!
+ fetch_mask_pixblock
cache_preload 8, 8
pixman_composite_over_n_8_8888_process_pixblock_head
.endm
.macro pixman_composite_over_n_8_8_process_pixblock_tail_head
vld1.8 {d4, d5, d6, d7}, [DST_R, :128]!
pixman_composite_over_n_8_8_process_pixblock_tail
- vld1.8 {d24, d25, d26, d27}, [MASK]!
+ fetch_mask_pixblock
cache_preload 32, 32
vst1.8 {d28, d29, d30, d31}, [DST_W, :128]!
pixman_composite_over_n_8_8_process_pixblock_head
vraddhn.u16 d29, q15, q9
vraddhn.u16 d30, q6, q10
vraddhn.u16 d31, q7, q11
- vld4.8 {d24, d25, d26, d27}, [MASK]!
+ fetch_mask_pixblock
vqadd.u8 q14, q0, q14
vqadd.u8 q15, q1, q15
cache_preload 8, 8
pixman_composite_add_n_8_8_process_pixblock_tail
vst1.8 {d28, d29, d30, d31}, [DST_W, :128]!
vld1.8 {d4, d5, d6, d7}, [DST_R, :128]!
- vld1.8 {d24, d25, d26, d27}, [MASK]!
+ fetch_mask_pixblock
cache_preload 32, 32
pixman_composite_add_n_8_8_process_pixblock_head
.endm
pixman_composite_add_8_8_8_process_pixblock_tail
vst1.8 {d28, d29, d30, d31}, [DST_W, :128]!
vld1.8 {d4, d5, d6, d7}, [DST_R, :128]!
- vld1.8 {d24, d25, d26, d27}, [MASK]!
+ fetch_mask_pixblock
fetch_src_pixblock
cache_preload 32, 32
pixman_composite_add_8_8_8_process_pixblock_head
pixman_composite_add_8888_8888_8888_process_pixblock_tail
vst4.8 {d28, d29, d30, d31}, [DST_W, :128]!
vld4.8 {d4, d5, d6, d7}, [DST_R, :128]!
- vld4.8 {d24, d25, d26, d27}, [MASK]!
+ fetch_mask_pixblock
fetch_src_pixblock
cache_preload 8, 8
pixman_composite_add_8888_8888_8888_process_pixblock_head
pixman_composite_out_reverse_8888_n_8888_process_pixblock_tail
fetch_src_pixblock
cache_preload 8, 8
- vld4.8 {d12, d13, d14, d15}, [MASK]!
+ fetch_mask_pixblock
pixman_composite_out_reverse_8888_n_8888_process_pixblock_head
vst4.8 {d28, d29, d30, d31}, [DST_W, :128]!
.endm
pixman_composite_over_8888_n_8888_process_pixblock_tail
fetch_src_pixblock
cache_preload 8, 8
- vld4.8 {d12, d13, d14, d15}, [MASK]!
+ fetch_mask_pixblock
pixman_composite_over_8888_n_8888_process_pixblock_head
vst4.8 {d28, d29, d30, d31}, [DST_W, :128]!
.endm
pixman_composite_over_8888_n_8888_process_pixblock_tail
fetch_src_pixblock
cache_preload 8, 8
- vld1.8 {d15}, [MASK]!
+ fetch_mask_pixblock
pixman_composite_over_8888_n_8888_process_pixblock_head
vst4.8 {d28, d29, d30, d31}, [DST_W, :128]!
.endm
/* TODO: expand macros and do better instructions scheduling */
.macro pixman_composite_over_0565_8_0565_process_pixblock_tail_head
- vld1.8 {d15}, [MASK]!
+ fetch_mask_pixblock
pixman_composite_over_0565_8_0565_process_pixblock_tail
fetch_src_pixblock
vld1.16 {d10, d11}, [DST_R, :128]!
/* TODO: expand macros and do better instructions scheduling */
.macro pixman_composite_add_0565_8_0565_process_pixblock_tail_head
- vld1.8 {d15}, [MASK]!
+ fetch_mask_pixblock
pixman_composite_add_0565_8_0565_process_pixblock_tail
fetch_src_pixblock
vld1.16 {d10, d11}, [DST_R, :128]!