From 0d6526f82c3cdefcca47f73f5fc08dc6f335eac6 Mon Sep 17 00:00:00 2001 From: Tim Chen Date: Mon, 21 Sep 2026 17:37:22 -0700 Subject: sched/cache: Keep nr_pref_llc_running in the runnable domain, to fix LLC mis-scheduling bug alb_break_llc() decides whether to break LLC preference during active load balance. It does so by testing that every runnable fair task on the source rq prefers its LLC: env->src_rq->nr_pref_llc_running == env->src_rq->cfs.h_nr_runnable But the two counters cover different sets. nr_pref_llc_running is updated in account_llc_enqueue()/account_llc_dequeue(), next to cfs_rq->nr_queued, so it follows queued tasks. h_nr_runnable is updated in set_delayed()/ clear_delayed() and drops delay-dequeued tasks. So under DELAY_DEQUEUE, a preferring task that goes to sleep stays counted in nr_pref_llc_running while h_nr_runnable falls. The equality then breaks, alb_break_llc() returns false, and active balance is free to pull a task off its preferred LLC. Active balance only moves runnable tasks, and this is the only LLC check it consults: once the stopper runs, LBF_ACTIVE_LB skips the per-task test in can_migrate_task(). The runnable set is the one we want. Fix it on the counter side. A task should be counted in nr_pref_llc_running exactly while it is both queued on its preferred LLC (pref_llc_queued) and runnable (!sched_delayed). Define that membership once in task_pref_llc_runnable(), and adjust the counter only through pref_llc_running_inc()/pref_llc_running_dec() from the four sites that change either input: account_llc_enqueue(), account_llc_dequeue(), set_delayed() and clear_delayed(). Gating every update on the same predicate keeps the delay, wake and dequeue paths from double-counting or underflowing; see the comments at those sites for the ordering. nr_llc_running and sd->llc_counts are not touched and stay on queued semantics. Fixes: 714059f79ff0 ("sched/cache: Handle moving single tasks to/from their preferred LLC") Closes: https://lore.kernel.org/lkml/20260827135000.735138-1-zhanxusheng@xiaomi.com/ Reported-by: Zhan Xusheng Suggested-by: Chen Yu Signed-off-by: Tim Chen Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Reviewed-by: Kayra Cizmeci Cc: # v7.2.x Link: https://patch.msgid.link/06af61afedac32e6477f57feb4d658f6c411c3af.1790035273.git.tim.c.chen@linux.intel.com --- Documentation/driver-api/infiniband.rst | 109 ++++++++++++++++++++++++++++++++ 1 file changed, 109 insertions(+) create mode 100644 Documentation/driver-api/infiniband.rst (limited to 'Documentation/driver-api/infiniband.rst') diff --git a/Documentation/driver-api/infiniband.rst b/Documentation/driver-api/infiniband.rst new file mode 100644 index 000000000..d48f24677 --- /dev/null +++ b/Documentation/driver-api/infiniband.rst @@ -0,0 +1,109 @@ +=========================================== +InfiniBand and Remote DMA (RDMA) Interfaces +=========================================== + +Introduction and Overview +========================= + +TBD + +InfiniBand core interfaces +========================== + +.. kernel-doc:: drivers/infiniband/core/iwpm_util.h + :internal: + +.. kernel-doc:: drivers/infiniband/core/cq.c + :export: + +.. kernel-doc:: drivers/infiniband/core/cm.c + :export: + +.. kernel-doc:: drivers/infiniband/core/rw.c + :export: + +.. kernel-doc:: drivers/infiniband/core/device.c + :export: + +.. kernel-doc:: drivers/infiniband/core/verbs.c + :export: + +.. kernel-doc:: drivers/infiniband/core/packer.c + :export: + +.. kernel-doc:: drivers/infiniband/core/sa_query.c + :export: + +.. kernel-doc:: drivers/infiniband/core/ud_header.c + :export: + +.. kernel-doc:: drivers/infiniband/core/umem.c + :export: + +.. kernel-doc:: drivers/infiniband/core/umem_odp.c + :export: + +RDMA Verbs transport library +============================ + +.. kernel-doc:: drivers/infiniband/sw/rdmavt/mr.c + :export: + +.. kernel-doc:: drivers/infiniband/sw/rdmavt/rc.c + :export: + +.. kernel-doc:: drivers/infiniband/sw/rdmavt/ah.c + :export: + +.. kernel-doc:: drivers/infiniband/sw/rdmavt/vt.c + :export: + +.. kernel-doc:: drivers/infiniband/sw/rdmavt/cq.c + :export: + +.. kernel-doc:: drivers/infiniband/sw/rdmavt/qp.c + :export: + +.. kernel-doc:: drivers/infiniband/sw/rdmavt/mcast.c + :export: + +Upper Layer Protocols +===================== + +iSCSI Extensions for RDMA (iSER) +-------------------------------- + +.. kernel-doc:: drivers/infiniband/ulp/iser/iscsi_iser.h + :internal: + +.. kernel-doc:: drivers/infiniband/ulp/iser/iscsi_iser.c + :functions: iscsi_iser_pdu_alloc iser_initialize_task_headers + iscsi_iser_task_init iscsi_iser_mtask_xmit iscsi_iser_task_xmit + iscsi_iser_cleanup_task iscsi_iser_check_protection + iscsi_iser_conn_create iscsi_iser_conn_bind + iscsi_iser_conn_start iscsi_iser_conn_stop + iscsi_iser_session_destroy iscsi_iser_session_create + iscsi_iser_set_param iscsi_iser_ep_connect iscsi_iser_ep_poll + iscsi_iser_ep_disconnect + +.. kernel-doc:: drivers/infiniband/ulp/iser/iser_initiator.c + :internal: + +.. kernel-doc:: drivers/infiniband/ulp/iser/iser_verbs.c + :internal: + +InfiniBand SCSI RDMA protocol target support +-------------------------------------------- + +.. kernel-doc:: drivers/infiniband/ulp/srpt/ib_srpt.h + :internal: + +.. kernel-doc:: drivers/infiniband/ulp/srpt/ib_srpt.c + :internal: + +iSCSI Extensions for RDMA (iSER) target support +----------------------------------------------- + +.. kernel-doc:: drivers/infiniband/ulp/isert/ib_isert.c + :internal: + -- cgit v1.3.1