x86 asm: intrinsics from memory

README.adoc

@@ -11933,8 +11933,9 @@ where:
 * `<suffix>`: data type:
 ** `ps`: 4 floats (Packed Single)
 ** `pd`: 2 doubles (Packed Double)
-** `ss`: 1 float (Single Single)
+** `ss`: 1 float (Single Single), often the lowest order one
 ** `sd`: 1 double (Single Double)
+** `si128`: 128-bits of integers of any size
 ** `ep<int_type>` integer types, e.g.:
 *** `epi32`: 32 bit signed integers
 *** `epu16`: 16 bit unsigned integers
@@ -11967,7 +11968,7 @@ Present in `gcc-7_3_0-release` tree at: `gcc/config/i386/x86intrin.h`.
 Bibliography:
 
 * https://www.cs.virginia.edu/~cr4bd/3330/S2018/simdref.html
-* https://www.cs.virginia.edu/~cr4bd/3330/S2018/simdref.html
+* https://software.intel.com/en-us/articles/how-to-use-intrinsics
 
 ==== GCC inline assembly register variables
 

userland/arch/x86_64/addpd.S

@@ -6,15 +6,17 @@
 #include <lkmc.h>
 
 LKMC_PROLOGUE
-.bss
-    output:       .skip 16
 .data
+    .align 16
     addps_input0: .float 1.5, 2.5,  3.5,  4.5
     addps_input1: .float 5.5, 6.5,  7.5,  8.5
     addps_expect: .float 7.0, 9.0, 11.0, 13.0
     addpd_input0: .double 1.5, 2.5
     addpd_input1: .double 5.5, 6.5
     addpd_expect: .double 7.0, 9.0
+.bss
+    .align 16
+    output:       .skip 16
 .text
 #define TEST(size) \
     movups addp ## size ## _input0, %xmm0; \

userland/arch/x86_64/intrinsics/addpd.c

@@ -1,11 +1,16 @@
 /* https://github.com/cirosantilli/linux-kernel-module-cheat#gcc-intrinsics */
 
 #include <assert.h>
+#include <string.h>
 
 #include <x86intrin.h>
 
-int main(void) {
+float global_input0[] __attribute__((aligned(16))) = {1.5f, 2.5f, 3.5f, 4.5f};
+float global_input1[] __attribute__((aligned(16))) = {5.5f, 6.5f, 7.5f, 8.5f};
+float global_output[4] __attribute__((aligned(16)));
+float global_expected[] __attribute__((aligned(16))) = {7.0f, 9.0f, 11.0f, 13.0f};
 
+int main(void) {
     /* 32-bit add (addps). */
     {
         __m128 input0 = _mm_set_ps(1.5f, 2.5f, 3.5f, 4.5f);
@@ -33,6 +38,14 @@ int main(void) {
         assert(_mm_cvtss_f32(_mm_shuffle_ps(output, output, 3)) ==  7.0f);
     }
 
+    /* Now from memory. */
+    {
+        __m128 *input0 = (__m128 *)global_input0;
+        __m128 *input1 = (__m128 *)global_input1;
+        _mm_store_ps(global_output, _mm_add_ps(*input0, *input1));
+        assert(!memcmp(global_output, global_expected, sizeof(global_output)));
+    }
+
     /* 64-bit add (addpd). */
     {
         __m128d input0 = _mm_set_pd(1.5, 2.5);

userland/arch/x86_64/intrinsics/paddq.c

@@ -1,9 +1,16 @@
 /* https://github.com/cirosantilli/linux-kernel-module-cheat#gcc-intrinsics */
 
 #include <assert.h>
+#include <inttypes.h>
+#include <string.h>
 
 #include <x86intrin.h>
 
+uint32_t global_input0[] __attribute__((aligned(16))) = {1, 2, 3, 4};
+uint32_t global_input1[] __attribute__((aligned(16))) = {5, 6, 7, 8};
+uint32_t global_output[4] __attribute__((aligned(16)));
+uint32_t global_expected[] __attribute__((aligned(16))) = {6, 8, 10, 12};
+
 int main(void) {
 
     /* 32-bit add hello world. */
@@ -19,6 +26,14 @@ int main(void) {
         assert(_mm_extract_epi32(output, 0) == 12);
     }
 
+    /* Now from memory. */
+    {
+        __m128i *input0 = (__m128i *)global_input0;
+        __m128i *input1 = (__m128i *)global_input1;
+        _mm_store_si128((__m128i *)global_output, _mm_add_epi32(*input0, *input1));
+        assert(!memcmp(global_output, global_expected, sizeof(global_output)));
+    }
+
     /* Now a bunch of other sizes. */
     {
         __m128i input0 = _mm_set_epi32(0xF1F1F1F1, 0xF2F2F2F2, 0xF3F3F3F3, 0xF4F4F4F4);

userland/arch/x86_64/paddq.S

@@ -10,6 +10,7 @@
 
 LKMC_PROLOGUE
 .data
+    .align 16
     input0:       .long 0xF1F1F1F1, 0xF2F2F2F2, 0xF3F3F3F3, 0xF4F4F4F4
     input1:       .long 0x12121212, 0x13131313, 0x14141414, 0x15151515
     paddb_expect: .long 0x03030303, 0x05050505, 0x07070707, 0x09090909
@@ -17,13 +18,14 @@ LKMC_PROLOGUE
     paddd_expect: .long 0x04040403, 0x06060605, 0x08080807, 0x0A0A0A09
     paddq_expect: .long 0x04040403, 0x06060606, 0x08080807, 0x0A0A0A0A
 .bss
+    .align 16
     output:       .skip 16
 .text
-    movups input1, %xmm1
+    movaps input1, %xmm1
 #define TEST(size) \
-    movups input0, %xmm0; \
+    movaps input0, %xmm0; \
     padd ## size %xmm1, %xmm0; \
-    movups %xmm0, output; \
+    movaps %xmm0, output; \
     LKMC_ASSERT_MEMCMP(output, padd ## size ## _expect, $0x10)
 
     /* 16x 8-bit */