Home
last modified time | relevance | path

Searched refs:sse2 (Results 1 – 25 of 1166) sorted by relevance

12345678910>>...47

/external/libjpeg-turbo/
DAndroid.bp91 "simd/i386/jccolor-sse2.asm",
94 "simd/i386/jcgray-sse2.asm",
95 "simd/i386/jchuff-sse2.asm",
96 "simd/i386/jcphuff-sse2.asm",
99 "simd/i386/jcsample-sse2.asm",
102 "simd/i386/jdcolor-sse2.asm",
105 "simd/i386/jdmerge-sse2.asm",
108 "simd/i386/jdsample-sse2.asm",
112 "simd/i386/jfdctfst-sse2.asm",
115 "simd/i386/jfdctint-sse2.asm",
[all …]
/external/XNNPACK/src/xnnpack/
Dparams-init.h94 params.sse2.input_zero_point[i] = (int16_t) (uint16_t) input_zero_point; in xnn_init_q8_gemm_params()
95 params.sse2.kernel_zero_point[i] = (int16_t) (uint16_t) kernel_zero_point; in xnn_init_q8_gemm_params()
97 params.sse2.multiplier[0] = multiplier; in xnn_init_q8_gemm_params()
98 params.sse2.multiplier[1] = multiplier; in xnn_init_q8_gemm_params()
99 params.sse2.multiplier[2] = multiplier; in xnn_init_q8_gemm_params()
100 params.sse2.multiplier[3] = multiplier; in xnn_init_q8_gemm_params()
101 params.sse2.rounding[0] = UINT64_C(0x40000000); in xnn_init_q8_gemm_params()
102 params.sse2.rounding[1] = UINT64_C(0x40000000); in xnn_init_q8_gemm_params()
103 params.sse2.remainder_mask[0] = (int32_t) remainder_mask; in xnn_init_q8_gemm_params()
104 params.sse2.remainder_mask[1] = (int32_t) remainder_mask; in xnn_init_q8_gemm_params()
[all …]
/external/llvm/test/CodeGen/X86/
Dsse2-intrinsics-x86-upgrade.ll2 ; RUN: llc < %s -mtriple=i686-apple-darwin -mattr=+sse2 | FileCheck %s
9 %res = call <2 x i64> @llvm.x86.sse2.psll.dq.bs(<2 x i64> %a0, i32 7) ; <<2 x i64>> [#uses=1]
12 declare <2 x i64> @llvm.x86.sse2.psll.dq.bs(<2 x i64>, i32) nounwind readnone
20 %res = call <2 x i64> @llvm.x86.sse2.psrl.dq.bs(<2 x i64> %a0, i32 7) ; <<2 x i64>> [#uses=1]
23 declare <2 x i64> @llvm.x86.sse2.psrl.dq.bs(<2 x i64>, i32) nounwind readnone
30 %res = call <2 x i64> @llvm.x86.sse2.psll.dq(<2 x i64> %a0, i32 8) ; <<2 x i64>> [#uses=1]
33 declare <2 x i64> @llvm.x86.sse2.psll.dq(<2 x i64>, i32) nounwind readnone
41 %res = call <2 x i64> @llvm.x86.sse2.psrl.dq(<2 x i64> %a0, i32 8) ; <<2 x i64>> [#uses=1]
44 declare <2 x i64> @llvm.x86.sse2.psrl.dq(<2 x i64>, i32) nounwind readnone
52 %res = call <2 x double> @llvm.x86.sse2.cvtdq2pd(<4 x i32> %a0) ; <<2 x double>> [#uses=1]
[all …]
Dsse2-intrinsics-x86.ll2 ; RUN: llc < %s -mtriple=i386-apple-darwin -mattr=-avx,+sse2 | FileCheck %s --check-prefix=SSE
15 …%res = call <2 x double> @llvm.x86.sse2.add.sd(<2 x double> %a0, <2 x double> %a1) ; <<2 x double>…
18 declare <2 x double> @llvm.x86.sse2.add.sd(<2 x double>, <2 x double>) nounwind readnone
31 …%res = call <2 x double> @llvm.x86.sse2.cmp.pd(<2 x double> %a0, <2 x double> %a1, i8 7) ; <<2 x d…
34 declare <2 x double> @llvm.x86.sse2.cmp.pd(<2 x double>, <2 x double>, i8) nounwind readnone
47 …%res = call <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double> %a0, <2 x double> %a1, i8 7) ; <<2 x d…
50 declare <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double>, <2 x double>, i8) nounwind readnone
71 %res = call i32 @llvm.x86.sse2.comieq.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
74 declare i32 @llvm.x86.sse2.comieq.sd(<2 x double>, <2 x double>) nounwind readnone
91 %res = call i32 @llvm.x86.sse2.comige.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
[all …]
Dpic-load-remat.ll1 ; RUN: llc < %s -mtriple=i686-apple-darwin -mattr=+sse2 -relocation-model=pic | grep psllw | grep pb
8 …%tmp4403 = tail call <8 x i16> @llvm.x86.sse2.psubs.w( <8 x i16> zeroinitializer, <8 x i16> zeroin…
9 …%tmp4443 = tail call <8 x i16> @llvm.x86.sse2.padds.w( <8 x i16> zeroinitializer, <8 x i16> zeroin…
10 …%tmp4609 = tail call <8 x i16> @llvm.x86.sse2.psll.w( <8 x i16> zeroinitializer, <8 x i16> bitcast…
12 …%tmp4658 = tail call <8 x i16> @llvm.x86.sse2.psll.w( <8 x i16> %tmp4651, <8 x i16> bitcast (<4 x …
13 …%tmp4669 = tail call <8 x i16> @llvm.x86.sse2.pavg.w( <8 x i16> < i16 -23170, i16 -23170, i16 -231…
14 …%tmp4679 = tail call <8 x i16> @llvm.x86.sse2.padds.w( <8 x i16> %tmp4669, <8 x i16> %tmp4669 ) no…
16 …%tmp4700 = tail call <8 x i16> @llvm.x86.sse2.padds.w( <8 x i16> %tmp4689, <8 x i16> zeroinitializ…
19 …%tmp4779 = tail call <8 x i16> @llvm.x86.sse2.psll.w( <8 x i16> %tmp4772, <8 x i16> bitcast (<4 x …
21 …%tmp4821 = tail call <8 x i16> @llvm.x86.sse2.padds.w( <8 x i16> %tmp4810, <8 x i16> zeroinitializ…
[all …]
Dvec_shift5.ll2 ; RUN: llc < %s -mtriple=i686-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X32
3 ; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X64
19 …%1 = tail call <8 x i16> @llvm.x86.sse2.pslli.w(<8 x i16> <i16 1, i16 2, i16 4, i16 8, i16 1, i16 …
33 …%1 = tail call <8 x i16> @llvm.x86.sse2.psrli.w(<8 x i16> <i16 4, i16 8, i16 16, i16 32, i16 4, i1…
47 …%1 = tail call <8 x i16> @llvm.x86.sse2.psrai.w(<8 x i16> <i16 4, i16 8, i16 16, i16 32, i16 4, i1…
61 %1 = tail call <4 x i32> @llvm.x86.sse2.pslli.d(<4 x i32> <i32 1, i32 2, i32 4, i32 8>, i32 3)
75 %1 = tail call <4 x i32> @llvm.x86.sse2.psrli.d(<4 x i32> <i32 4, i32 8, i32 16, i32 32>, i32 3)
89 %1 = tail call <4 x i32> @llvm.x86.sse2.psrai.d(<4 x i32> <i32 4, i32 8, i32 16, i32 32>, i32 3)
104 %1 = tail call <2 x i64> @llvm.x86.sse2.pslli.q(<2 x i64> <i64 1, i64 2>, i32 3)
119 %1 = tail call <2 x i64> @llvm.x86.sse2.psrli.q(<2 x i64> <i64 8, i64 16>, i32 3)
[all …]
Dvec_shift.ll2 ; RUN: llc < %s -mtriple=i686-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X32
3 ; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X64
18 …%tmp9 = tail call <8 x i16> @llvm.x86.sse2.psll.w( <8 x i16> %tmp8, <8 x i16> %tmp6 ) nounwind rea…
39 …%tmp9 = tail call <8 x i16> @llvm.x86.sse2.psra.w( <8 x i16> %tmp2, <8 x i16> %tmp8 ) ; <<8 x i16…
44 declare <8 x i16> @llvm.x86.sse2.psra.w(<8 x i16>, <8 x i16>) nounwind readnone
57 …%tmp9 = tail call <2 x i64> @llvm.x86.sse2.psrl.q( <2 x i64> %b1, <2 x i64> %c ) nounwind readnone…
61 declare <2 x i64> @llvm.x86.sse2.psrl.q(<2 x i64>, <2 x i64>) nounwind readnone
63 declare <8 x i16> @llvm.x86.sse2.psll.w(<8 x i16>, <8 x i16>) nounwind readnone
Dvec_shift3.ll2 ; RUN: llc < %s -mtriple=i686-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X32
3 ; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X64
18 …%tmp3 = tail call <2 x i64> @llvm.x86.sse2.pslli.q( <2 x i64> %x1, i32 %bits ) nounwind readnone …
33 …%tmp3 = tail call <2 x i64> @llvm.x86.sse2.pslli.q( <2 x i64> %x1, i32 10 ) nounwind readnone ; …
51 …%tmp4 = tail call <8 x i16> @llvm.x86.sse2.psrai.w( <8 x i16> %tmp2, i32 %bits ) nounwind readnone…
56 declare <8 x i16> @llvm.x86.sse2.psrai.w(<8 x i16>, i32) nounwind readnone
57 declare <2 x i64> @llvm.x86.sse2.pslli.q(<2 x i64>, i32) nounwind readnone
Davx-intrinsics-x86_64.ll6 %res = call i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> %a0) ; <i64> [#uses=1]
9 declare i64 @llvm.x86.sse2.cvtsd2si64(<2 x double>) nounwind readnone
14 …%res = call <2 x double> @llvm.x86.sse2.cvtsi642sd(<2 x double> %a0, i64 %a1) ; <<2 x double>> [#u…
17 declare <2 x double> @llvm.x86.sse2.cvtsi642sd(<2 x double>, i64) nounwind readnone
22 %res = call i64 @llvm.x86.sse2.cvttsd2si64(<2 x double> %a0) ; <i64> [#uses=1]
25 declare i64 @llvm.x86.sse2.cvttsd2si64(<2 x double>) nounwind readnone
/external/libvpx/libvpx/vpx_dsp/x86/
Dhighbd_variance_sse2.c261 DECLS(sse2);
277 unsigned int sse2; \
280 &sse2, NULL, NULL); \
282 sse += sse2; \
286 &sse2, NULL, NULL); \
288 sse += sse2; \
291 &sse2, NULL, NULL); \
293 sse += sse2; \
311 uint32_t sse2; \
314 &sse2, NULL, NULL); \
[all …]
Dvariance_sse2.c436 DECLS(sse2, sse2);
450 unsigned int sse2; \
453 ref_stride, h, &sse2, NULL, NULL); \
455 sse_tmp += sse2; \
459 ref_stride, h, &sse2, NULL, NULL); \
461 sse_tmp += sse2; \
464 ref_stride, h, &sse2, NULL, NULL); \
466 sse_tmp += sse2; \
489 FNS(sse2, sse2);
507 DECLS(sse2, sse2);
[all …]
/external/swiftshader/third_party/llvm-7.0/llvm/test/CodeGen/X86/
Dpic-load-remat.ll1 ; RUN: llc < %s -mtriple=i686-apple-darwin -mattr=+sse2 -relocation-model=pic | grep psllw | grep pb
8 …%tmp4403 = tail call <8 x i16> @llvm.x86.sse2.psubs.w( <8 x i16> zeroinitializer, <8 x i16> zeroin…
9 …%tmp4443 = tail call <8 x i16> @llvm.x86.sse2.padds.w( <8 x i16> zeroinitializer, <8 x i16> zeroin…
10 …%tmp4609 = tail call <8 x i16> @llvm.x86.sse2.psll.w( <8 x i16> zeroinitializer, <8 x i16> bitcast…
12 …%tmp4658 = tail call <8 x i16> @llvm.x86.sse2.psll.w( <8 x i16> %tmp4651, <8 x i16> bitcast (<4 x …
13 …%tmp4669 = tail call <8 x i16> @llvm.x86.sse2.pavg.w( <8 x i16> < i16 -23170, i16 -23170, i16 -231…
14 …%tmp4679 = tail call <8 x i16> @llvm.x86.sse2.padds.w( <8 x i16> %tmp4669, <8 x i16> %tmp4669 ) no…
16 …%tmp4700 = tail call <8 x i16> @llvm.x86.sse2.padds.w( <8 x i16> %tmp4689, <8 x i16> zeroinitializ…
19 …%tmp4779 = tail call <8 x i16> @llvm.x86.sse2.psll.w( <8 x i16> %tmp4772, <8 x i16> bitcast (<4 x …
21 …%tmp4821 = tail call <8 x i16> @llvm.x86.sse2.padds.w( <8 x i16> %tmp4810, <8 x i16> zeroinitializ…
[all …]
Dvec_shift5.ll2 ; RUN: llc < %s -mtriple=i686-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X32
3 ; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X64
19 …%1 = tail call <8 x i16> @llvm.x86.sse2.pslli.w(<8 x i16> <i16 1, i16 2, i16 4, i16 8, i16 1, i16 …
33 …%1 = tail call <8 x i16> @llvm.x86.sse2.psrli.w(<8 x i16> <i16 4, i16 8, i16 16, i16 32, i16 4, i1…
47 …%1 = tail call <8 x i16> @llvm.x86.sse2.psrai.w(<8 x i16> <i16 4, i16 8, i16 16, i16 32, i16 4, i1…
61 %1 = tail call <4 x i32> @llvm.x86.sse2.pslli.d(<4 x i32> <i32 1, i32 2, i32 4, i32 8>, i32 3)
75 %1 = tail call <4 x i32> @llvm.x86.sse2.psrli.d(<4 x i32> <i32 4, i32 8, i32 16, i32 32>, i32 3)
89 %1 = tail call <4 x i32> @llvm.x86.sse2.psrai.d(<4 x i32> <i32 4, i32 8, i32 16, i32 32>, i32 3)
103 %1 = tail call <2 x i64> @llvm.x86.sse2.pslli.q(<2 x i64> <i64 1, i64 2>, i32 3)
117 %1 = tail call <2 x i64> @llvm.x86.sse2.psrli.q(<2 x i64> <i64 8, i64 16>, i32 3)
[all …]
Dfptosi-constant.ll11 %r = call i32 @llvm.x86.sse2.cvtsd2si(<2 x double> <double 5.0e+09, double undef>)
20 %r = call i32 @llvm.x86.sse2.cvttsd2si(<2 x double> <double 5.0e+09, double undef>)
29 %r = call i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> <double 5.0e+19, double undef>)
38 %r = call i64 @llvm.x86.sse2.cvttsd2si64(<2 x double> <double 5.0e+19, double undef>)
47 %r = call <4 x i32> @llvm.x86.sse2.cvtpd2dq(<2 x double> <double 5.0e+09, double 5.0e+09>)
56 %r = call <4 x i32> @llvm.x86.sse2.cvttpd2dq(<2 x double> <double 5.0e+09, double 5.0e+09>)
103 …%r = call <4 x i32> @llvm.x86.sse2.cvtps2dq(<4 x float> <float 5.0e+09, float 5.0e+09, float 5.0e+…
112 …%r = call <4 x i32> @llvm.x86.sse2.cvttps2dq(<4 x float> <float 5.0e+09, float 5.0e+09, float 5.0e…
116 declare i32 @llvm.x86.sse2.cvtsd2si(<2 x double>)
117 declare i32 @llvm.x86.sse2.cvttsd2si(<2 x double>)
[all …]
Dsse2-intrinsics-x86.ll2 ; RUN: llc < %s -disable-peephole -mtriple=i386-apple-darwin -mattr=+sse2 -show-mc-encoding | FileC…
5 ; RUN: llc < %s -disable-peephole -mtriple=x86_64-apple-darwin -mattr=+sse2 -show-mc-encoding | Fil…
19 …%res = call <2 x double> @llvm.x86.sse2.cmp.pd(<2 x double> %a0, <2 x double> %a1, i8 7) ; <<2 x d…
22 declare <2 x double> @llvm.x86.sse2.cmp.pd(<2 x double>, <2 x double>, i8) nounwind readnone
35 …%res = call <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double> %a0, <2 x double> %a1, i8 7) ; <<2 x d…
38 declare <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double>, <2 x double>, i8) nounwind readnone
68 %res = call i32 @llvm.x86.sse2.comieq.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
71 declare i32 @llvm.x86.sse2.comieq.sd(<2 x double>, <2 x double>) nounwind readnone
95 %res = call i32 @llvm.x86.sse2.comige.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
98 declare i32 @llvm.x86.sse2.comige.sd(<2 x double>, <2 x double>) nounwind readnone
[all …]
Dvec_shift.ll2 ; RUN: llc < %s -mtriple=i686-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X32
3 ; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X64
18 …%tmp9 = tail call <8 x i16> @llvm.x86.sse2.psll.w( <8 x i16> %tmp8, <8 x i16> %tmp6 ) nounwind rea…
39 …%tmp9 = tail call <8 x i16> @llvm.x86.sse2.psra.w( <8 x i16> %tmp2, <8 x i16> %tmp8 ) ; <<8 x i16…
44 declare <8 x i16> @llvm.x86.sse2.psra.w(<8 x i16>, <8 x i16>) nounwind readnone
57 …%tmp9 = tail call <2 x i64> @llvm.x86.sse2.psrl.q( <2 x i64> %b1, <2 x i64> %c ) nounwind readnone…
61 declare <2 x i64> @llvm.x86.sse2.psrl.q(<2 x i64>, <2 x i64>) nounwind readnone
63 declare <8 x i16> @llvm.x86.sse2.psll.w(<8 x i16>, <8 x i16>) nounwind readnone
Dvec_shift3.ll2 ; RUN: llc < %s -mtriple=i686-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X32
3 ; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X64
18 …%tmp3 = tail call <2 x i64> @llvm.x86.sse2.pslli.q( <2 x i64> %x1, i32 %bits ) nounwind readnone …
33 …%tmp3 = tail call <2 x i64> @llvm.x86.sse2.pslli.q( <2 x i64> %x1, i32 10 ) nounwind readnone ; …
51 …%tmp4 = tail call <8 x i16> @llvm.x86.sse2.psrai.w( <8 x i16> %tmp2, i32 %bits ) nounwind readnone…
56 declare <8 x i16> @llvm.x86.sse2.psrai.w(<8 x i16>, i32) nounwind readnone
57 declare <2 x i64> @llvm.x86.sse2.pslli.q(<2 x i64>, i32) nounwind readnone
/external/llvm/test/Transforms/InstCombine/
Dx86-sse2.ll8 ; CHECK-NEXT: [[TMP2:%.*]] = tail call <2 x double> @llvm.x86.sse2.sqrt.sd(<2 x double> [[TMP1]])
14 %3 = tail call <2 x double> @llvm.x86.sse2.sqrt.sd(<2 x double> %2)
25 %3 = tail call <2 x double> @llvm.x86.sse2.sqrt.sd(<2 x double> %2)
32 ; CHECK-NEXT: [[TMP1:%.*]] = tail call <2 x double> @llvm.x86.sse2.add.sd(<2 x double> %a, <2 x …
36 %2 = tail call <2 x double> @llvm.x86.sse2.add.sd(<2 x double> %a, <2 x double> %1)
49 %5 = tail call <2 x double> @llvm.x86.sse2.add.sd(<2 x double> %2, <2 x double> %4)
62 %5 = tail call <2 x double> @llvm.x86.sse2.add.sd(<2 x double> %2, <2 x double> %4)
69 ; CHECK-NEXT: [[TMP1:%.*]] = tail call <2 x double> @llvm.x86.sse2.sub.sd(<2 x double> %a, <2 x …
73 %2 = tail call <2 x double> @llvm.x86.sse2.sub.sd(<2 x double> %a, <2 x double> %1)
86 %5 = tail call <2 x double> @llvm.x86.sse2.sub.sd(<2 x double> %2, <2 x double> %4)
[all …]
/external/swiftshader/third_party/llvm-7.0/llvm/test/Transforms/InstCombine/X86/
Dx86-sse2.ll12 %3 = tail call <2 x double> @llvm.x86.sse2.sqrt.sd(<2 x double> %2)
23 %3 = tail call <2 x double> @llvm.x86.sse2.sqrt.sd(<2 x double> %2)
37 %5 = tail call <2 x double> @llvm.x86.sse2.add.sd(<2 x double> %2, <2 x double> %4)
50 %5 = tail call <2 x double> @llvm.x86.sse2.add.sd(<2 x double> %2, <2 x double> %4)
64 %5 = tail call <2 x double> @llvm.x86.sse2.sub.sd(<2 x double> %2, <2 x double> %4)
77 %5 = tail call <2 x double> @llvm.x86.sse2.sub.sd(<2 x double> %2, <2 x double> %4)
91 %5 = tail call <2 x double> @llvm.x86.sse2.mul.sd(<2 x double> %2, <2 x double> %4)
104 %5 = tail call <2 x double> @llvm.x86.sse2.mul.sd(<2 x double> %2, <2 x double> %4)
118 %5 = tail call <2 x double> @llvm.x86.sse2.div.sd(<2 x double> %2, <2 x double> %4)
131 %5 = tail call <2 x double> @llvm.x86.sse2.div.sd(<2 x double> %2, <2 x double> %4)
[all …]
/external/XNNPACK/src/q8-vadd/
Dsse2.c23 …const __m128i vzero_point_product = _mm_load_si128((const __m128i*) &params->sse2.zero_point_produ… in xnn_q8_vadd_ukernel__sse2()
24 const __m128i va_multiplier_lo = _mm_load_si128((const __m128i*) &params->sse2.a_multiplier_lo); in xnn_q8_vadd_ukernel__sse2()
25 const __m128i va_multiplier_hi = _mm_load_si128((const __m128i*) &params->sse2.a_multiplier_hi); in xnn_q8_vadd_ukernel__sse2()
26 const __m128i vb_multiplier_lo = _mm_load_si128((const __m128i*) &params->sse2.b_multiplier_lo); in xnn_q8_vadd_ukernel__sse2()
27 const __m128i vb_multiplier_hi = _mm_load_si128((const __m128i*) &params->sse2.b_multiplier_hi); in xnn_q8_vadd_ukernel__sse2()
28 const __m128i vremainder_mask = _mm_load_si128((const __m128i*) params->sse2.remainder_mask); in xnn_q8_vadd_ukernel__sse2()
29 …const __m128i vremainder_threshold = _mm_load_si128((const __m128i*) params->sse2.remainder_thresh… in xnn_q8_vadd_ukernel__sse2()
30 const __m128i vshift = _mm_cvtsi32_si128((int) params->sse2.shift); in xnn_q8_vadd_ukernel__sse2()
68 const __m128i vy_zero_point = _mm_load_si128((const __m128i*) params->sse2.y_zero_point); in xnn_q8_vadd_ukernel__sse2()
71 vy = _mm_max_epu8(vy, _mm_load_si128((const __m128i*) params->sse2.y_min)); in xnn_q8_vadd_ukernel__sse2()
[all …]
/external/libjpeg-turbo/simd/
DCMakeLists.txt107 x86_64/jccolor-sse2.asm x86_64/jcgray-sse2.asm x86_64/jchuff-sse2.asm
108 x86_64/jcphuff-sse2.asm x86_64/jcsample-sse2.asm x86_64/jdcolor-sse2.asm
109 x86_64/jdmerge-sse2.asm x86_64/jdsample-sse2.asm x86_64/jfdctfst-sse2.asm
110 x86_64/jfdctint-sse2.asm x86_64/jidctflt-sse2.asm x86_64/jidctfst-sse2.asm
111 x86_64/jidctint-sse2.asm x86_64/jidctred-sse2.asm x86_64/jquantf-sse2.asm
112 x86_64/jquanti-sse2.asm
124 i386/jccolor-sse2.asm i386/jcgray-sse2.asm i386/jchuff-sse2.asm
125 i386/jcphuff-sse2.asm i386/jcsample-sse2.asm i386/jdcolor-sse2.asm
126 i386/jdmerge-sse2.asm i386/jdsample-sse2.asm i386/jfdctfst-sse2.asm
127 i386/jfdctint-sse2.asm i386/jidctflt-sse2.asm i386/jidctfst-sse2.asm
[all …]
/external/libaom/libaom/aom_dsp/x86/
Dhighbd_variance_sse2.c281 DECLS(sse2);
294 unsigned int sse2; \
300 src, src_stride, x_offset, y_offset, dst, dst_stride, h, &sse2, \
303 sse += sse2; \
307 &sse2, NULL, NULL); \
309 sse += sse2; \
313 dst_stride, h, &sse2, NULL, NULL); \
315 sse += sse2; \
318 dst_stride, h, &sse2, NULL, NULL); \
320 sse += sse2; \
[all …]
Dintrapred_asm_sse2.asm30 INIT_XMM sse2
52 INIT_XMM sse2
73 INIT_XMM sse2
93 INIT_XMM sse2
122 INIT_XMM sse2
148 INIT_XMM sse2
175 INIT_XMM sse2
189 INIT_XMM sse2
208 INIT_XMM sse2
241 INIT_XMM sse2
[all …]
/external/swiftshader/third_party/llvm-7.0/llvm/test/Transforms/ConstProp/
Dsse.ll11 %i2 = call i32 @llvm.x86.sse2.cvtsd2si(<2 x double> <double 7.0, double undef>) nounwind
12 %i3 = call i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> <double 7.0, double undef>) nounwind
31 %i2 = call i32 @llvm.x86.sse2.cvtsd2si(<2 x double> <double 1.75, double undef>) nounwind
32 %i3 = call i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> <double 1.75, double undef>) nounwind
53 %i2 = call i32 @llvm.x86.sse2.cvtsd2si(<2 x double> %dm) nounwind
54 %i3 = call i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> %dm) nounwind
74 %i2 = call i32 @llvm.x86.sse2.cvtsd2si(<2 x double> %dm) nounwind
75 %i3 = call i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> %dm) nounwind
95 %i2 = call i32 @llvm.x86.sse2.cvtsd2si(<2 x double> %dm) nounwind
96 %i3 = call i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> %dm) nounwind
[all …]
/external/libaom/libaom/aom_dsp/arm/
Dvariance_neon.c87 uint32_t sse1, sse2; in aom_variance32x64_neon() local
90 32, 32, &sse2, &sum2); in aom_variance32x64_neon()
91 *sse = sse1 + sse2; in aom_variance32x64_neon()
100 uint32_t sse1, sse2; in aom_variance64x32_neon() local
103 64, 16, &sse2, &sum2); in aom_variance64x32_neon()
104 *sse = sse1 + sse2; in aom_variance64x32_neon()
113 uint32_t sse1, sse2; in aom_variance64x64_neon() local
117 64, 16, &sse2, &sum2); in aom_variance64x64_neon()
118 sse1 += sse2; in aom_variance64x64_neon()
122 b_stride, 64, 16, &sse2, &sum2); in aom_variance64x64_neon()
[all …]

12345678910>>...47