--- title: Linux 7.3 SMP Improvement To Help Reduce Latency,... description: The Linux 7.3 merge window brings SMP scheduling improvements that reduce preemption-disabled sections in smp_call_function*(), targeting scheduling latency... canonical: https://daily.dev/posts/linux-7-3-smp-improvement-to-help-reduce-latency-improve-real-time-performance-yzzvslbvk twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Linux 7.3 SMP Improvement To Help Reduce Latency, Improve Real-Time Performance | daily.dev og:description: The Linux 7.3 merge window brings SMP scheduling improvements that reduce preemption-disabled sections in smp_call_function*(), targeting scheduling latency... og:url: https://daily.dev/posts/linux-7-3-smp-improvement-to-help-reduce-latency-improve-real-time-performance-yzzvslbvk og:image: https://api.daily.dev/og/posts/yzZVSlbvk.png og:image:alt: Linux 7.3 SMP Improvement To Help Reduce Latency, Improve Real-Time Performance og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Linux 7.3 SMP Improvement To Help Reduce Latency, Improve Real-Time Performance **[Phoronix](https://daily.dev/sources/phoronix)** · 2 min read · 0 upvotes · 0 comments ## Summary The Linux 7.3 merge window brings SMP scheduling improvements that reduce preemption-disabled sections in smp_call_function*(), targeting scheduling latency spikes caused by IPI completion waits. Bytedance engineers reworked the code to allow preemption during IPI completion waiting, addressing especially bad behavior for latency-sensitive workloads like DPDK. Thomas Gleixner's pull request notes the change reduces SMP function call induced latency for high priority tasks from roughly 17ms to 1.5ms, a 90% reduction, measured in production fleets. The pull is awaiting merge action from Linus Torvalds and is expected to land as the Linux 7.3 merge window opens. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What SMP change is coming in Linux 7.3 to reduce scheduling latency? Linux 7.3 reworks smp_call_function*() to shrink the preemption-disabled sections around IPI completion waits by giving each task a per-task CPU mask to track which CPUs it is waiting on, allowing preemption to be re-enabled before the wait. Bytedance engineers found this cut SMP function call induced latency for high priority tasks from about 17ms to about 1.5ms, a 90% reduction, in fleet measurements with DPDK workloads. _daily.dev surfaces kernel latency fixes like this for teams tuning real-time or DPDK-based systems._ ### Why does waiting for remote CPU IPI completion cause scheduling latency spikes on Linux? Because the existing smp_call_function*() implementation keeps preemption disabled across the entire operation, including the wait for completion, so if a target CPU is slow to respond to the interprocessor interrupt, the calling task cannot be preempted and latency spikes occur. This is especially costly on x86_64 where TLB flushes rely on IPIs, and process exit or page reclamation can trigger waits on multiple IPI operations at once. _Track kernel scheduling internals like this on daily.dev before they land in a release you rely on._ ## Similar posts on daily.dev - [Linux 7.1 Scheduler Changes May Benefit Some Workloads](https://daily.dev/posts/linux-7-1-scheduler-changes-may-benefit-some-workloads-gjorr1itb) · Phoronix · 0 upvotes · 0 comments - [Linux 7.3 Expected To "Flatten The Pick" For Better Scheduling While Gaming & More](https://daily.dev/posts/linux-7-3-expected-to-flatten-the-pick-for-better-scheduling-while-gaming-more-4sk8p5nyz) · Phoronix · 0 upvotes · 0 comments - [Linux 7.3 To Better Handle Cluster Load Balancing On Intel Hybrid CPUs](https://daily.dev/posts/linux-7-3-to-better-handle-cluster-load-balancing-on-intel-hybrid-cpus-mkqiaouvp) · Phoronix · 0 upvotes · 0 comments --- Tags: [#linux](https://daily.dev/tags/linux), [#real-time-systems](https://daily.dev/tags/real-time-systems) [View this post on daily.dev](https://daily.dev/posts/linux-7-3-smp-improvement-to-help-reduce-latency-improve-real-time-performance-yzzvslbvk) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Linux 7.3 SMP Improvement To Help Reduce Latency, Improve Real-Time Performance","url":"https://daily.dev/posts/linux-7-3-smp-improvement-to-help-reduce-latency-improve-real-time-performance-yzzvslbvk","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/linux-7-3-smp-improvement-to-help-reduce-latency-improve-real-time-performance-yzzvslbvk"},"datePublished":"2026-08-17T14:12:20.609Z","dateModified":"2026-08-21T10:41:24.670Z","description":"The Linux 7.3 merge window brings SMP scheduling improvements that reduce preemption-disabled sections in smp_call_function*(), targeting scheduling latency...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/8f0514ed37333b767a9f3fe115c2ed52?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/8f0514ed37333b767a9f3fe115c2ed52?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Phoronix","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Phoronix","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/6a26ba379dd64932830aa0c9803dd7ee","url":"https://daily.dev/sources/phoronix"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/linux-7-3-smp-improvement-to-help-reduce-latency-improve-real-time-performance-yzzvslbvk","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"linux,real-time-systems","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Phoronix","item":"https://daily.dev/sources/phoronix"},{"@type":"ListItem","position":3,"name":"Linux 7.3 SMP Improvement To Help Reduce Latency, Improve Real-Time Performance"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/linux-7-3-smp-improvement-to-help-reduce-latency-improve-real-time-performance-yzzvslbvk#faq","mainEntity":[{"@type":"Question","name":"What SMP change is coming in Linux 7.3 to reduce scheduling latency?","acceptedAnswer":{"@type":"Answer","text":"Linux 7.3 reworks smp_call_function*() to shrink the preemption-disabled sections around IPI completion waits by giving each task a per-task CPU mask to track which CPUs it is waiting on, allowing preemption to be re-enabled before the wait. Bytedance engineers found this cut SMP function call induced latency for high priority tasks from about 17ms to about 1.5ms, a 90% reduction, in fleet measurements with DPDK workloads. daily.dev surfaces kernel latency fixes like this for teams tuning real-time or DPDK-based systems."}},{"@type":"Question","name":"Why does waiting for remote CPU IPI completion cause scheduling latency spikes on Linux?","acceptedAnswer":{"@type":"Answer","text":"Because the existing smp_call_function*() implementation keeps preemption disabled across the entire operation, including the wait for completion, so if a target CPU is slow to respond to the interprocessor interrupt, the calling task cannot be preempted and latency spikes occur. This is especially costly on x86_64 where TLB flushes rely on IPIs, and process exit or page reclamation can trigger waits on multiple IPI operations at once. Track kernel scheduling internals like this on daily.dev before they land in a release you rely on."}}]} ```