Lecture 08 – Multiprocessor Systems and Memory Consistency Problems (PDF)
The Advanced Computer Architectures course main page is https://cw.fel.cvut.cz/wiki/courses/b4m35pap/start
The guidepost to more Czech Technical University in Prague computer architecture teaching materials is provided at https://comparch.edu.cvut.cz/.
There are even Czech language recordings of this course from the Winter Semester 21/22.
Our broader topic knowledge base at https://gitlab.fel.cvut.cz/otrees/org/-/wikis/knowbase.
A link to the article about mentioned problem with atomic operations Linux 6.18 Merges Fix For “Catastrophic Performance Issue” On 64-bit ARM at Phoronix.com
Much more precise and detailed texts (papers and books) about scalability including memory ordering by @paulmckrcu are linked from his page http://www2.rdrop.com/users/paulmck/. The detailed Summary of Memory Ordering choices for current CPU architectures is documented in the Table 15.5 Chapter 15. Advanced Synchronization: Memory Ordering of his book Is Parallel Programming Hard, And If So, What Can You Do About It?.
Links to related Ticket Lock code example for individual architectures:
Kernel Spinlock documentation at LWN.net from test-and-set over ticket-spinlocks to scalable MCS locks and qspinlocks.