diff --git a/README.adoc b/README.adoc
index 26d8fb0..cbfe67b 100644
--- a/README.adoc
+++ b/README.adoc
@@ -11742,23 +11742,20 @@ As a quick reminder, the fastest setups to get started are:
 
 However, as usual, it is saner to build your toolchain as explained at: <<qemu-user-mode-getting-started>>.
 
-The first example that you want to run for each arch is:
+The first examples that you want to run for each arch are:
 
-....
-./run --userland userland/arch/<arch>/add.S
-....
+* how to move data between registers and memory
+* how to add two numbers!
 
-e.g.:
+These examples are located at:
 
-....
-./run --userland userland/arch/x86_64/add.S
-....
-
-Sources:
-
-* link:userland/arch/x86_64/add.S[]
-* link:userland/arch/arm/add.S[]
-* link:userland/arch/aarch64/add.S[]
+* x86
+** link:userland/arch/x86_64/add.S[]
+** link:userland/arch/x86_64/mov.S[]
+* arm
+** <<arm-mov-instruction>>
+** link:userland/arch/arm/add.S[]
+** link:userland/arch/aarch64/add.S[]
 
 These examples use the venerable ADD instruction to:
 
@@ -12288,15 +12285,7 @@ Bibliography: https://stackoverflow.com/questions/27147043/n-suffix-to-branch-in
 
 Arch agnostic infrastructure getting started at: <<userland-assembly>>.
 
-=== x86 userland assembly getting started
-
-These are the main concepts and instructions that you should learn to be able to understand what is going on.
-
-Once those are done, everything else left on userland is just to learn a huge list of instructions: <<x86-userland-assembly-instructions>>
-
-=== x86 userland assembly instructions
-
-==== x86 SIMD
+=== x86 SIMD
 
 History:
 
@@ -12309,15 +12298,15 @@ History:
 * AVX2:2013
 * AVX-512: 2016. 512-bit ZMM registers. Extension of YMM.
 
-===== x86 SSE2
+==== x86 SSE2
 
-====== x86 addpd instruction
+===== x86 addpd instruction
 
 link:userland/arch/x86_64/addpd.S[]: `addps`, `addpd`
 
 Good first instruction to learn SIMD: <<simd-assembly>>
 
-====== x86 paddq instruction
+===== x86 paddq instruction
 
 link:userland/arch/x86_64/paddq.S[]: `paddq`, `paddl`, `paddw`, `paddb`
 
diff --git a/userland/arch/x86_64/mov.S b/userland/arch/x86_64/mov.S
new file mode 100644
index 0000000..f25b143
--- /dev/null
+++ b/userland/arch/x86_64/mov.S
@@ -0,0 +1,37 @@
+/* https://github.com/cirosantilli/linux-kernel-module-cheat#userland-assembly */
+
+#include <lkmc.h>
+
+.data
+    myint: .long 0x12345678
+LKMC_PROLOGUE
+    /* Immediate and register. */
+    mov $0, %rax
+    mov $1, %rax
+    LKMC_ASSERT_EQ(%rax, $1)
+
+    /* Register and register. */
+    mov $0, %rax
+    mov $1, %rbx
+    mov %rbx, %rax
+    LKMC_ASSERT_EQ(%rax, $1)
+
+    /* Memory and register. */
+    mov myint, %rax
+    LKMC_ASSERT_EQ(%rax, $0x12345678)
+
+    /* Memory and immediate. */
+    movl $0x9ABCDEF0, myint
+    LKMC_ASSERT_EQ(myint, $0x9ABCDEF0)
+
+    /* Memory via pointer to address. */
+    /* eax = &myint */
+    mov $myint, %rax
+    movl $0x11112222, (%rax)
+    LKMC_ASSERT_EQ(myint, $0x11112222)
+
+    /* Possible to move on itself, seems like a NOP and way to clear 32 high bits in x86-64:
+     * http://stackoverflow.com/questions/11910501/why-did-gcc-generate-mov-eax-eax-and-what-does-it-mean
+     */
+    mov %rax, %rax
+LKMC_EPILOGUE