* rq->clock_update_flags < RQCF_ACT_SKIP in dl_rq_change_utilization()
@ 2025-07-07 13:19 Thomas Walker
0 siblings, 0 replies; only message in thread
From: Thomas Walker @ 2025-07-07 13:19 UTC (permalink / raw)
To: linux-kernel
Hi, we have a couple of workloads running on 6.12 kernels that showed some slightly increased latency tails with fair_server enabled, so we disable it on boot on some number of isolated (nohz_full, etc) cores.
$ echo -1 | sudo tee /proc/sys/kernel/sched_rt_runtime_us
$ for f in $(seq 2 2 16); do echo 0 | sudo tee /sys/kernel/debug/sched/fair_server/cpu${f}/runtime ; done
Which works as expected but, more often than not, triggers the following WARN_ON. Any suggestions for debugging further would be greatly appreciated.
[ 653.430659] rq->clock_update_flags < RQCF_ACT_SKIP
[ 653.430663] WARNING: CPU: 87 PID: 14185 at kernel/sched/sched.h:1649 dl_rq_change_utilization+0x225/0x250
[ 653.430671] Modules linked in: uas usb_storage tcp_diag inet_diag tls nvme_fabrics nvme_keyring dell_rbu xfs dm_crypt rpcrdma sunrpc rdma_ucm ib_iser rdma_cm iw_cm xt_conntrack ib_cm nf_conntrack libiscsi scsi_transport_iscsi nf_defrag_ipv6 intel_rapl_msr nf_defrag_ipv4 intel_rapl_common xt_comment intel_uncore_frequency xt_set intel_uncore_frequency_common intel_ifs xt_tcpudp i10nm_edac ipt_REJECT skx_edac_common nf_reject_ipv4 nfit xt_LOG nf_log_syslog libnvdimm nft_compat nf_tables x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel irdma libcrc32c kvm i40e ib_uverbs ip_set_hash_net irqbypass crct10dif_pclmul ghash_clmulni_intel sha512_ssse3 ip_set_bitmap_port sha256_ssse3 sha1_ssse3 ip_set aesni_intel gf128mul crypto_simd cryptd binfmt_misc nfnetlink ib_core ipmi_ssif snd_pcm rapl intel_cstate drm_shmem_helper snd_timer deflate drm_kms_helper iaa_crypto intel_uncore snd dell_pc nls_ascii idxd pmt_telemetry nls_cp437 evdev intel_sdsi mei_me pmt_class dell_smbios vfat platform_profile iTCO_wdt dax_hmem dcdbas
[ 653.430731] fat soundcore mei intel_pmc_bxt ipmi_si cxl_acpi dell_wmi_descriptor pcspkr wmi_bmof iTCO_vendor_support isst_if_mbox_pci isst_if_mmio acpi_power_meter watchdog acpi_ipmi i2c_algo_bit ipmi_devintf isst_if_common idxd_bus intel_vsec ipmi_msghandler sg button tcp_scalable drm loop dm_mod efi_pstore configfs efivarfs ip_tables x_tables autofs4 ext4 crc32c_generic crc16 mbcache jbd2 sd_mod xhci_pci sfc xhci_hcd mpi3mr ahci nvme libahci ice libata usbcore scsi_transport_sas nvme_core tg3 mtd i2c_i801 crc32_pclmul gnss scsi_mod libphy i2c_ismt nvme_auth crc32c_intel usb_common mdio libie i2c_smbus wmi scsi_common pinctrl_emmitsburg
[ 653.430774] CPU: 87 UID: 0 PID: 14185 Comm: tee Not tainted 6.12.35 #2
[ 653.430776] Hardware name: Dell Inc. PowerEdge R660/XXXXXX, BIOS 2.6.3 03/26/2025
[ 653.430778] RIP: 0010:dl_rq_change_utilization+0x225/0x250
[ 653.430782] Code: 09 00 00 72 d2 e9 7d fe ff ff 80 3d c2 55 cf 01 00 0f 85 9f fe ff ff 48 c7 c7 28 88 11 a2 c6 05 ae 55 cf 01 01 e8 9b 64 f9 ff <0f> 0b e9 85 fe ff ff e8 bf 1e f9 ff e9 e6 fd ff ff 48 c7 c7 88 8d
[ 653.430784] RSP: 0018:ff70e427e8b0fbf0 EFLAGS: 00010082
[ 653.430786] RAX: 0000000000000000 RBX: ff169e20be6b5e40 RCX: 0000000000000027
[ 653.430787] RDX: ff169e60bf5a1748 RSI: 0000000000000001 RDI: ff169e60bf5a1740
[ 653.430788] RBP: ff169e20be6b67a8 R08: 0000000000000000 R09: ff70e427e8b0fa70
[ 653.430789] R10: ff169e60bdf3ffa8 R11: 0000000000000003 R12: ff169e20be6a1680
[ 653.430790] R13: 0000000000000000 R14: ffffffffa0d3fe50 R15: ff169e21c6a3f800
[ 653.430792] FS: 00007facf3351740(0000) GS:ff169e60bf580000(0000) knlGS:0000000000000000
[ 653.430794] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 653.430795] CR2: 000055c54ae470b8 CR3: 00000040e3f92004 CR4: 0000000000f73ef0
[ 653.430797] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 653.430797] DR3: 0000000000000000 DR6: 00000000fffe07f0 DR7: 0000000000000400
[ 653.430799] PKRU: 55555554
[ 653.430800] Call Trace:
[ 653.430801] <TASK>
[ 653.430805] dl_server_apply_params+0x282/0x320
[ 653.430809] sched_fair_server_write.isra.0+0x11a/0x1d0
[ 653.430813] full_proxy_write+0x4e/0x90
[ 653.430816] vfs_write+0xf5/0x450
[ 653.430819] ? vfs_write+0x28d/0x450
[ 653.430822] ksys_write+0x6d/0xf0
[ 653.430824] do_syscall_64+0x82/0x190
[ 653.430828] ? do_syscall_64+0x8e/0x190
[ 653.430830] ? syscall_exit_to_user_mode+0x4d/0x210
[ 653.430832] ? do_syscall_64+0x8e/0x190
[ 653.430833] ? __memcg_slab_free_hook+0xf7/0x140
[ 653.430836] ? __x64_sys_close+0x3c/0x80
[ 653.430839] ? kmem_cache_free+0x3ee/0x440
[ 653.430841] ? syscall_exit_to_user_mode+0x4d/0x210
[ 653.430843] ? do_syscall_64+0x8e/0x190
[ 653.430845] ? exc_page_fault+0x7e/0x180
[ 653.430847] entry_SYSCALL_64_after_hwframe+0x76/0x7e
[ 653.430850] RIP: 0033:0x7facf344c300
[ 653.430852] Code: 40 00 48 8b 15 01 9b 0d 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 80 3d e1 22 0e 00 00 74 17 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 58 c3 0f 1f 80 00 00 00 00 48 83 ec 28 48 89
[ 653.430853] RSP: 002b:00007ffefcf68c88 EFLAGS: 00000202 ORIG_RAX: 0000000000000001
[ 653.430855] RAX: ffffffffffffffda RBX: 0000000000000002 RCX: 00007facf344c300
[ 653.430857] RDX: 0000000000000002 RSI: 00007ffefcf68da0 RDI: 0000000000000003
[ 653.430857] RBP: 00007ffefcf68da0 R08: 0000000000000004 R09: 0000000000000001
[ 653.430858] R10: 00007facf3368f18 R11: 0000000000000202 R12: 0000000000000002
[ 653.430859] R13: 000055c54ae45470 R14: 0000000000000002 R15: 00007facf35229e0
[ 653.430862] </TASK>
[ 653.430863] ---[ end trace 0000000000000000 ]---
[ 653.430865] Fair server disabled in CPU 2, system may crash due to starvation.
[ 653.480471] Fair server disabled in CPU 4, system may crash due to starvation.
[ 653.533134] Fair server disabled in CPU 6, system may crash due to starvation.
[ 653.586481] Fair server disabled in CPU 8, system may crash due to starvation.
[ 653.636169] Fair server disabled in CPU 10, system may crash due to starvation.
[ 653.685718] Fair server disabled in CPU 12, system may crash due to starvation.
[ 653.737015] Fair server disabled in CPU 14, system may crash due to starvation.
[ 653.787139] Fair server disabled in CPU 16, system may crash due to starvation.
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2025-07-07 13:25 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-07-07 13:19 rq->clock_update_flags < RQCF_ACT_SKIP in dl_rq_change_utilization() Thomas Walker
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).