Sync: introduce Lock and Guard commit, commit, commit, commit, commit, commit, commit, commit, commit, commit Rwbase: Mitigate indefinite writer starvation commit Remove diagnostics and adjust config for CSD lock diagnostics commit, commit, commit, commit Locking: Introduce local_try_cmpxchg commit, commit, commit Pipes: support FMODE_NOWAIT commit, commit Prctl: Add PR_GET_AUXV to copy auxv to userspace commitĪdd Kthread flag to /proc/$pid/status commit Reduce cost of sched_move_task when config autogroup commitįix performance regression introduced by mm_cid commit Interleave cfs bandwidth timers for improved single thread performance at low utilization commit Take advantage of completion batching for uring commands commit Some applications use timeouts to wake someone waiting on completion entries, and this makes it a bit easier to just have a recurring timer rather than needing to rearm it every time commit User mapped provided buffer rings commit, commit, commit, commit, commitĪdd support for multishot timeouts. (FEATURED) PSI: Allow unprivileged PSI polling commit, commit, commit, commit Recommended LWN article: Disabling SELinux's runtime disable Removal of the SELinux runtime disable capabilityįor security reasons, the ability to diable SELinux at runtime has been removed. Recommended LWN article: Generic iterators for BPFġ.8. This takes lazy tlb mm shootdown IPI interrupts from 314 to 3 on a 144 POWERPC CPU system doing a kernel compile.īPF open-coded iterators, support for kptrs in more BPF maps, kptr RCU, shared ownership for local kptrs, support 64-bit pointers to kfuncs, or first class support for attaching uprobes to functions inside ELF objects containers in Android APKs. This release improves scalability of context switching between user and kernel threads on large systems with a threaded process spread across a lot of CPUs More scalable context switching between user and kernel threads Recommended LWN article: Concurrent page-fault handling with per-VMA locksġ.6. For example, startup of Android apps with a large number of threads. This improves scalability in a number of scenarios. The memory management subsystem has added a new kind of locks in each VMA (Virtual Memory Area, a data structure that representes virtual memory) which means that some operations will no longer use the mmap_lock. More scalable memory management with per-VMA locks Recommended LWN article: Process-level kernel samepage merging controlġ.5. The first one allows to enable KSM at the process level and the second one to query the setting This release adds two new options to the prctl system call. To be able to use KSM for more workloads, KSM needs to have the ability to be enabled and disabled at the process / cgroup level. So far KSM can only be enabled by calling madvise() for memory regions, but this API is too insecure. Safer Kernel Samepage Merging with a new process/cgroup API Recommended LWN article: User trace events, one year later.ġ.4. The third is the size of the value at the specified address The second is the bit to set/clear to reflect the event being enabled. The first is the address to update when the event is either enabled or disabled. In this new model during the event registration from user programs 3 new values are specified. This release allows for unprivileged PSI polling, which allow unprivileged tools to provide this information to users.Ī new API has been added for user-space trace events. Recommended LWN article: Support for Intel's Linear Address Maskingġ.2. The capability can be used for efficient address sanitizers (ASAN) implementation and for optimizations in JITs and virtual machines This is similar to ARM's Top Byte Ignore and allows userspace to use some bits of pointers without masking it out before use, allowing sofware to use these untranslated address bits for metadata. Enable Intel's Linear Address MaskingĮnable Address Masking on Intel chips.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |