aboutsummaryrefslogtreecommitdiffstats
path: root/tools/perf/scripts/python/event_analyzing_sample.py
diff options
context:
space:
mode:
authorYan Zhai <yan@cloudflare.com>2024-03-19 13:44:40 -0700
committerJakub Kicinski <kuba@kernel.org>2024-03-20 21:05:43 -0700
commit00bf63122459e87193ee7f1bc6161c83a525569f (patch)
treebb54c5de6214b985d103ec0d400e6b8d9527e10f /tools/perf/scripts/python/event_analyzing_sample.py
parentnet: report RCU QS on threaded NAPI repolling (diff)
downloadlinux-00bf63122459e87193ee7f1bc6161c83a525569f.tar.gz
linux-00bf63122459e87193ee7f1bc6161c83a525569f.zip
bpf: report RCU QS in cpumap kthread
When there are heavy load, cpumap kernel threads can be busy polling packets from redirect queues and block out RCU tasks from reaching quiescent states. It is insufficient to just call cond_resched() in such context. Periodically raise a consolidated RCU QS before cond_resched fixes the problem. Fixes: 6710e1126934 ("bpf: introduce new bpf cpu map type BPF_MAP_TYPE_CPUMAP") Reviewed-by: Jesper Dangaard Brouer <hawk@kernel.org> Signed-off-by: Yan Zhai <yan@cloudflare.com> Acked-by: Paul E. McKenney <paulmck@kernel.org> Acked-by: Jesper Dangaard Brouer <hawk@kernel.org> Link: https://lore.kernel.org/r/c17b9f1517e19d813da3ede5ed33ee18496bb5d8.1710877680.git.yan@cloudflare.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Diffstat (limited to 'tools/perf/scripts/python/event_analyzing_sample.py')
0 files changed, 0 insertions, 0 deletions
ions'>-1/+137 2017-11-08x86/xen/time: set pvclock flags on xen_time_init()Joao Martins1-0/+9 2017-11-08x86/pvclock: add setter for pvclock_pvti_cpu0_vaJoao Martins5-17/+27 2017-11-08ptp_kvm: probe for kvm guest availabilityJoao Martins1-0/+3 2017-11-08xen/privcmd: remove unused variable pageidxColin Ian King1-3/+0 2017-11-06xen: select grant interface versionJuergen Gross1-2/+33 2017-11-06xen: update arch/x86/include/asm/xen/cpuid.hJuergen Gross1-10/+32 2017-11-06xen: add grant interface version dependent constants to gnttab_opsJuergen Gross1-30/+43 2017-11-06xen: limit grant v2 interface to the v1 functionalityJuergen Gross2-175/+0 2017-11-06xen: re-introduce support for grant v2 interfaceJuergen Gross4-13/+398 2017-11-03xen: support priv-mapping in an HVM tools domainPaul Durrant2-2/+36 2017-11-03xen/pvcalls: remove redundant check for irq >= 0Colin Ian King1-3/+1 2017-11-03xen/pvcalls: fix unsigned less than zero error checkColin Ian King1-4/+3 2017-11-03xen/time: Return -ENODEV from xen_get_wallclock()Boris Ostrovsky1-1/+1 2017-11-03xen/pvcalls-front: mark expected switch fall-throughGustavo A. R. Silva1-1/+2 2017-11-03xen: xenbus_probe_frontend: mark expected switch fall-throughsGustavo A. R. Silva1-0/+2 2017-11-02xen/time: do not decrease steal time after live migration on xenDongli Zhang3-7/+73 2017-10-31xen: support 52 bit physical addresses in pv guestsJuergen Gross2-3/+12 2017-10-31xen: introduce a Kconfig option to enable the pvcalls frontendStefano Stabellini2-0/+12 2017-10-31xen/pvcalls: implement release commandStefano Stabellini2-0/+99 2017-10-31xen/pvcalls: implement poll commandStefano Stabellini2-9/+138 2017-10-31xen/pvcalls: implement recvmsgStefano Stabellini2-0/+115 2017-10-31xen/pvcalls: implement sendmsgStefano Stabellini2-0/+124 2017-10-31xen/pvcalls: implement accept commandStefano Stabellini2-0/+148 2017-10-31xen/pvcalls: implement listen commandStefano Stabellini2-0/+58 2017-10-31xen/pvcalls: implement bind commandStefano Stabellini2-0/+69 2017-10-31xen/pvcalls: implement connect commandStefano Stabellini2-0/+160 2017-10-31xen/pvcalls: implement socket command and handle eventsStefano Stabellini2-0/+139 2017-10-31xen/pvcalls: connect to the backendStefano Stabellini1-0/+132 2017-10-31xen/pvcalls: implement frontend disconnectStefano Stabellini1-0/+71