Perf record: Implement -z,-compression_level option Perf report: Implement perf.data record decompression Perf inject: Enable COMPRESSED record decompression Perf tests: Implement Zstd comp/decomp integration test Perf test zstd: Fixup verbose mode output
Perf vendor events arm64: Remove ] wildcard Perf vendor events arm64: Map Brahma-B53 CPUID to cortex-a53 events Perf vendor events arm64: Add Cortex-A57 and Cortex-A72 events Perf parse-regs: Add generic support for arch_intr/user_reg_mask() Perf regs x86: Add X86 specific arch_intr_reg_mask() Perf intel-pt: Fix instructions sampling rate Perf intel-pt: Fix improved sample timestamp Perf intel-pt: Fix sample timestamp wrt non-taken branches Perf tools: Add a 'percore' event qualifier Perf stat: Factor out aggregate counts printing Perf stat: Support 'percore' event qualifier * 'perf-core-for-linus' of git:///pub/scm/linux/kernel/git/tip/tip: (73 commits) Improve exported-sql-viewer GUI, allowing, for instance, toĬopy'n'paste the trees, useful for e-mailing" UPI replaced the Intel QuickPath Interconnect (QPI) in Add uncore_upi (Intel's "Ultra Path Interconnect" events) JSON Add support for recording and printing XMM registers, available, Add DWARF register mappings for libdw, allowing -call-graph=dwarf Map Brahma-B53 CPUID to cortex-a53 events. now its possible to do this per-event, and have it mixed with This together with other metrics that are collected per hardware We can already do this with -per-core, but it's often useful to do Add a 'percore' event qualifier, e.g.: -eĬpu/event=0,umask=0x3,percore=1/, that sums up the event counts for Remove hist_account_cycles() from callback, speeding up branch Improve -user-regs/intr-regs suggestions to overcome errors When profiling with DWARF-based stacks and tracing of context Saves storage space on larger server systems where perf.data sizeĬan easily reach several tens or even hundreds of GiBs, especially
Perf.data file size reduction on variety of tested workloads what PERF_RECORD_COMPRESSED user space records, resulting in ~3-5x Streaming compression of perf ring buffer into Pull perf tooling updates from Ingo Molnar: Merge branch 'perf-core-for-linus' of git:///pub/scm/linux/kernel/git/tip/tip