diff options
Diffstat (limited to 'gfx/cairo/libpixman/src/pixman-arm-neon-asm.S')
-rw-r--r-- | gfx/cairo/libpixman/src/pixman-arm-neon-asm.S | 78 |
1 files changed, 37 insertions, 41 deletions
diff --git a/gfx/cairo/libpixman/src/pixman-arm-neon-asm.S b/gfx/cairo/libpixman/src/pixman-arm-neon-asm.S index 35eca116d1..0e092577f1 100644 --- a/gfx/cairo/libpixman/src/pixman-arm-neon-asm.S +++ b/gfx/cairo/libpixman/src/pixman-arm-neon-asm.S @@ -34,12 +34,6 @@ * - pixman_composite_over_n_8_0565_asm_neon */ -#ifdef __clang__ -#define ldrgeb ldrbge -#define subges subsge -#define subpls subspl -#endif - /* Prevent the stack from becoming executable for no reason... */ #if defined(__linux__) && defined(__ELF__) .section .note.GNU-stack,"",%progbits @@ -59,6 +53,8 @@ #include "pixman-arm-asm.h" #include "pixman-arm-neon-asm.h" + pixman_syntax_unified + /* Global configuration options and preferences */ /* @@ -287,12 +283,12 @@ PF subge, PF_X, PF_X, ORIG_W vrshr.u16 q3, q11, #8 vrshr.u16 q15, q12, #8 - PF subges, PF_CTL, PF_CTL, #0x10 + PF subsge, PF_CTL, PF_CTL, #0x10 vsri.u16 q14, q9, #11 - PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! + PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! vraddhn.u16 d20, q10, q13 vraddhn.u16 d23, q11, q3 - PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! + PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! vraddhn.u16 d22, q12, q15 vst1.16 {d28, d29}, [DST_W, :128]! .endm @@ -451,9 +447,9 @@ generate_composite_function \ vshll.u8 q8, d1, #8 vst1.16 {d28, d29}, [DST_W, :128]! PF subge, PF_X, PF_X, ORIG_W - PF subges, PF_CTL, PF_CTL, #0x10 + PF subsge, PF_CTL, PF_CTL, #0x10 vshll.u8 q14, d2, #8 - PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! + PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! vshll.u8 q9, d0, #8 .endm @@ -525,10 +521,10 @@ generate_composite_function \ PF pld, [PF_SRC, PF_X, lsl #src_bpp_shift] PF pld, [PF_DST, PF_X, lsl #dst_bpp_shift] PF subge, PF_X, PF_X, ORIG_W - PF subges, PF_CTL, PF_CTL, #0x10 + PF subsge, PF_CTL, PF_CTL, #0x10 vqadd.u8 q14, q0, q2 - PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! - PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! + PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! + PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! vqadd.u8 q15, q1, q3 .endm @@ -557,10 +553,10 @@ generate_composite_function \ PF pld, [PF_SRC, PF_X, lsl #src_bpp_shift] PF pld, [PF_DST, PF_X, lsl #dst_bpp_shift] PF subge, PF_X, PF_X, ORIG_W - PF subges, PF_CTL, PF_CTL, #0x10 + PF subsge, PF_CTL, PF_CTL, #0x10 vqadd.u8 q14, q0, q2 - PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! - PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! + PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! + PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! vqadd.u8 q15, q1, q3 .endm @@ -631,9 +627,9 @@ generate_composite_function_single_scanline \ vmull.u8 q8, d22, d4 PF subsge, PF_CTL, PF_CTL, #0x10 vmull.u8 q9, d22, d5 - PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! + PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! vmull.u8 q10, d22, d6 - PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! + PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! vmull.u8 q11, d22, d7 .endm @@ -683,11 +679,11 @@ generate_composite_function_single_scanline \ vst4.8 {d28, d29, d30, d31}, [DST_W, :128]! PF subge, PF_X, PF_X, ORIG_W vmull.u8 q8, d22, d4 - PF subges, PF_CTL, PF_CTL, #0x10 + PF subsge, PF_CTL, PF_CTL, #0x10 vmull.u8 q9, d22, d5 - PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! + PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! vmull.u8 q10, d22, d6 - PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! + PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! vmull.u8 q11, d22, d7 .endm @@ -759,9 +755,9 @@ generate_composite_function_single_scanline \ vmull.u8 q9, d24, d5 PF subge, PF_X, PF_X, ORIG_W vmull.u8 q10, d24, d6 - PF subges, PF_CTL, PF_CTL, #0x10 + PF subsge, PF_CTL, PF_CTL, #0x10 vmull.u8 q11, d24, d7 - PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! + PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! vst4.8 {d28, d29, d30, d31}, [DST_W, :128]! .endm @@ -810,10 +806,10 @@ generate_composite_function \ vst4.8 {d28, d29, d30, d31}, [DST_W, :128]! PF subge, PF_X, PF_X, ORIG_W vmull.u8 q8, d22, d4 - PF subges, PF_CTL, PF_CTL, #0x10 + PF subsge, PF_CTL, PF_CTL, #0x10 vmull.u8 q9, d22, d5 vmull.u8 q10, d22, d6 - PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! + PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! vmull.u8 q11, d22, d7 .endm @@ -1265,9 +1261,9 @@ generate_composite_function \ vmull.u8 q9, d24, d1 PF subge, PF_X, PF_X, ORIG_W vmull.u8 q10, d24, d2 - PF subges, PF_CTL, PF_CTL, #0x10 + PF subsge, PF_CTL, PF_CTL, #0x10 vmull.u8 q11, d24, d3 - PF ldrgeb, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]! + PF ldrbge, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]! vst4.8 {d28, d29, d30, d31}, [DST_W, :128]! vrsra.u16 q8, q8, #8 vrsra.u16 q9, q9, #8 @@ -1334,9 +1330,9 @@ generate_composite_function \ vmull.u8 q1, d25, d16 PF subge, PF_X, PF_X, ORIG_W vmull.u8 q2, d26, d16 - PF subges, PF_CTL, PF_CTL, #0x10 + PF subsge, PF_CTL, PF_CTL, #0x10 vmull.u8 q3, d27, d16 - PF ldrgeb, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]! + PF ldrbge, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]! vst1.8 {d28, d29, d30, d31}, [DST_W, :128]! vrsra.u16 q0, q0, #8 vrsra.u16 q1, q1, #8 @@ -1430,11 +1426,11 @@ generate_composite_function \ vmull.u8 q7, d24, d9 PF subge, PF_X, PF_X, ORIG_W vmull.u8 q8, d24, d10 - PF subges, PF_CTL, PF_CTL, #0x10 + PF subsge, PF_CTL, PF_CTL, #0x10 vmull.u8 q9, d24, d11 - PF ldrgeb, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! + PF ldrbge, DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]! vqadd.u8 q14, q0, q14 - PF ldrgeb, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]! + PF ldrbge, DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]! vqadd.u8 q15, q1, q15 vrshr.u16 q10, q6, #8 vrshr.u16 q11, q7, #8 @@ -2444,8 +2440,8 @@ generate_composite_function \ PF cmp, PF_X, ORIG_W PF pld, [PF_SRC, PF_X, lsl #src_bpp_shift] PF subge, PF_X, PF_X, ORIG_W - PF subges, PF_CTL, PF_CTL, #0x10 - PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! + PF subsge, PF_CTL, PF_CTL, #0x10 + PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! .endm generate_composite_function \ @@ -2501,8 +2497,8 @@ generate_composite_function \ PF cmp, PF_X, ORIG_W PF pld, [PF_SRC, PF_X, lsl #src_bpp_shift] PF subge, PF_X, PF_X, ORIG_W - PF subges, PF_CTL, PF_CTL, #0x10 - PF ldrgeb, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! + PF subsge, PF_CTL, PF_CTL, #0x10 + PF ldrbge, DUMMY, [PF_SRC, SRC_STRIDE, lsl #src_bpp_shift]! .endm generate_composite_function \ @@ -3153,7 +3149,7 @@ pixman_asm_function \fname /* ensure good destination alignment */ cmp WIDTH, #1 blt 0f - tst OUT, #(1 << dst_bpp_shift) + tst OUT, #(1 << \dst_bpp_shift) beq 0f vshr.u16 q15, q12, #(16 - BILINEAR_INTERPOLATION_BITS) vadd.u16 q12, q12, q13 @@ -3166,7 +3162,7 @@ pixman_asm_function \fname cmp WIDTH, #2 blt 0f - tst OUT, #(1 << (dst_bpp_shift + 1)) + tst OUT, #(1 << (\dst_bpp_shift + 1)) beq 0f bilinear_interpolate_two_pixels \src_fmt, \dst_fmt sub WIDTH, WIDTH, #2 @@ -3175,7 +3171,7 @@ pixman_asm_function \fname /*********** 8 pixels per iteration *****************/ cmp WIDTH, #4 blt 0f - tst OUT, #(1 << (dst_bpp_shift + 2)) + tst OUT, #(1 << (\dst_bpp_shift + 2)) beq 0f bilinear_interpolate_four_pixels \src_fmt, \dst_fmt sub WIDTH, WIDTH, #4 @@ -3242,7 +3238,7 @@ pixman_asm_function \fname .unreq TMP3 .unreq TMP4 .unreq STRIDE -pixman_end_asm_function + pixman_end_asm_function .endm |