Agner's CPU blog

I have done some research about this topic. For aligned fundamental data type, it is atomic to access. In the document《Arm® Architecture Reference Manual Armv8, for A-profile architecture》, it is called single-copy atomic.

From INTEL CPU manual, it says "nonaligned data accesses will seriously impact the performance of the processor". Then I do a test in order to prove it. But the result is that aligned and nonaligned data accesses have the same performance. Why??? Could someone help? My code is shown below: #include ...

As we know, access aligned fundamental data types in INTEL X86 architecture is atomic. How about ARMV8？ I have tried to get the result from Arm Architecture Reference Manual Armv8, for A-profile architecture, I did find something related to atomicity. ARMV8 is other-multi-copy atomic. It promises th...