Instruction-based sampling support for upcoming AMD Zen 4 processors under Linux review
The previous and current patch series for the new AMD Zen 4 family have had basic and lackluster terminology. With the latest patches under review, users can rest assured that the Zen 4 Linux support is currently taking effect and will be apparent in all-new patches and updates to the open-source operating system. Zen 4 is set to enhance instruction-based sampling by creating an additional data source extension and a new L3 cache miss filtering ability. The new Zen 4 IBS features commented on the higher priority modifications: Linux users will be able to read through the mailing list on the official Linux kernel site about the new Zen 4 IBS patch information. IBS L3 miss filtering works by tagging an instruction on IBS counter overflow and generating an NMI if the tagged instruction causes an L3 miss. Samples without an L3 miss are discarded and counter is reset with random value (between 1-15 for fetch pmu and 1-127 for op pmu). This helps in reducing sampling overhead when user is interested only in such samples. One of the use case of such filtered samples is to feed data to page-migration daemon in tiered memory systems. Add support for L3 miss filtering in IBS driver via new pmu attribute “l3missonly”. In summary, outside of perf instruction-based sampling, the new submission for selection is also beneficial for a gradually increasing number of features for compiling to give the ability to feed the perf hardware sampling outcomes back to the compiler for aiding in designing profile-based and optimized binaries. Intel has been more aggressive with functionality and hardware performance counters developed under Linux’s upcoming operating system. Along with seeing utilization for profiling possible optimization and issues that will need debugging, the enablement by AMD for their IBS Zen 4 capability is excellent for the company. Source: Phoronix
