Public bug reported: This is a public version of https://bugs.launchpad.net/bugs/2034985
Backport: 6852c46c783d mm/mmap: move vma operations to mm_struct out of the critical section of file mapping lock (merged upstream in 6.6) to jammy [Impact] The msg_bytes and msg_hdrs atomic counters are frequently updated when IPC msg queue is in heavy use, causing heavy cache bounce and overhead. Change them to percpu_counter greatly improve the performance. Since there is one percpu struct per namespace, additional memory cost is minimal. Reading of the count done in msgctl call, which is infrequent. So the need to sum up the counts in each CPU is infrequent. [Fix] Backport: 6852c46c783d mm/mmap: move vma operations to mm_struct out of the critical section of file mapping lock [Test] I have tested this code. [Where problems could occur] TODO ** Affects: linux (Ubuntu) Importance: Undecided Status: New ** Affects: linux (Ubuntu Jammy) Importance: Undecided Status: New ** Also affects: linux (Ubuntu) Importance: Undecided Status: New ** No longer affects: intel ** Also affects: linux (Ubuntu Jammy) Importance: Undecided Status: New -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/2058488 Title: [SPR] performance: Remove operations to mm_struct out of file mapping lock's critical section Status in linux package in Ubuntu: New Status in linux source package in Jammy: New Bug description: This is a public version of https://bugs.launchpad.net/bugs/2034985 Backport: 6852c46c783d mm/mmap: move vma operations to mm_struct out of the critical section of file mapping lock (merged upstream in 6.6) to jammy [Impact] The msg_bytes and msg_hdrs atomic counters are frequently updated when IPC msg queue is in heavy use, causing heavy cache bounce and overhead. Change them to percpu_counter greatly improve the performance. Since there is one percpu struct per namespace, additional memory cost is minimal. Reading of the count done in msgctl call, which is infrequent. So the need to sum up the counts in each CPU is infrequent. [Fix] Backport: 6852c46c783d mm/mmap: move vma operations to mm_struct out of the critical section of file mapping lock [Test] I have tested this code. [Where problems could occur] TODO To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2058488/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp