Performing SRU verification. Kim Naru provided Alan Baghumian a bare metal system with high cpu count.
The lscpu is in the realm of: $ lscpu Architecture: x86_64 ... CPU(s): 512 On-line CPU(s) list: 0-511 ... They deployed a Jammy system, enabled -proposed and installed 5.15.0-125-generic: $ uname -a Linux test 5.15.0-125-generic #135-Ubuntu SMP Fri Sep 27 13:53:58 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux $ sudo turbostat turbostat version 21.05.04 - Len Brown <l...@kernel.org> ... current_driver: acpi_idle current_governor: menu current_governor_ro: menu cpu40: POLL: CPUIDLE CORE POLL IDLE cpu40: C1: ACPI FFH MWAIT 0x0 cpu40: C2: ACPI IOPORT 0x814 cpu40: cpufreq driver: acpi-cpufreq cpu40: cpufreq governor: schedutil cpufreq boost: 1 cpu0: MSR_RAPL_PWR_UNIT: 0x000a1000 (1.000000 Watts, 0.000015 Joules, 0.000977 sec.) cpu128: MSR_RAPL_PWR_UNIT: 0x000a1000 (1.000000 Watts, 0.000015 Joules, 0.000977 sec.) Package Die Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz IPC IRQ POLL C1 C2 POLL% C1% C2% CorWatt PkgWatt - - - - 0 0.03 1615 2742 0.51 15320 0 36 14527 0.00 0.00 101.66 0.06 85.53 0 0 0 0 1 0.04 1820 2696 1.50 54 0 0 50 0.00 0.00 99.96 0.00 42.88 ... The full turbostat output works as expected, showing values for all 512 cpu cores. The kernel in -proposed fixes the issue, happy to mark verified for jammy. ** Tags removed: verification-needed-jammy-linux ** Tags added: verification-done-jammy-linux -- 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/2069961 Title: turbostat fails with too many open files on large systems Status in linux package in Ubuntu: Fix Released Status in linux source package in Jammy: Fix Committed Status in linux source package in Mantic: Won't Fix Status in linux source package in Noble: Fix Released Bug description: BugLink: https://bugs.launchpad.net/bugs/2069961 [Impact] On large systems, e.g. with 512 cpus or more, turbostat fails to run due to exceeding the rlimit for number of files. 512 cpus requires 1028 file descriptors, but the current limit is 999. $ lscpu ... CPU(s): 512 On-line CPU(s) list: 0-511 ... $ sudo turbostat ... turbostat: /sys/devices/system/cpu/cpu477/cpuidle/state0/usage: open failed: Too many open files There is no workaround, apart from maybe using powerstat instead. [Fix] The fix is to increase the rlimit to increase the amount of file descriptors that turbostat can open to 2^15, which should be plenty for some time to come. commit 3ac1d14d0583a2de75d49a5234d767e2590384dd Author: Wyes Karny <wyes.ka...@amd.com> Date: Tue Oct 3 05:07:51 2023 +0000 Subject: tools/power turbostat: Increase the limit for fd opened Link: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=3ac1d14d0583a2de75d49a5234d767e2590384dd This landed in 6.9-rc4, and requires a backport for minor context adjustment in the first hunk for jammy. Noble got fixed already through upstream stable. [Testcase] Deploy a bare metal system with 512 or more cpus. Install linux-tools: $ sudo apt install linux-tools-$(uname -r) Run turbostat: $ sudo turbostat ... turbostat: /sys/devices/system/cpu/cpu477/cpuidle/state0/usage: open failed: Too many open files There are test kernels available in the following ppa: https://launchpad.net/~mruffell/+archive/ubuntu/sf388491-test If you install them, you should be able to see normal turbostat output for all cpus installed in the system. [Where problems can occur] We are simply increasing the rlimit for file descriptors that turbostat can open. This should have no impact on any existing systems. If a regression should occur, then turbostat functionality might not work. Users could use powerstat instead as a workaround while things are fixed. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2069961/+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
