summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2024-11-14sched_ext: Replace scx_next_task_picked() with switch_class() in commentZhao Mengmeng
scx_next_task_picked() has been replaced with siwtch_class(), but comment is still referencing old one, so replace it. Signed-off-by: Zhao Mengmeng <zhaomengmeng@kylinos.cn> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-11-11sched_ext: Rename scx_bpf_dispatch[_vtime]_from_dsq*() -> ↵Tejun Heo
scx_bpf_dsq_move[_vtime]*() In sched_ext API, a repeatedly reported pain point is the overuse of the verb "dispatch" and confusion around "consume": - ops.dispatch() - scx_bpf_dispatch[_vtime]() - scx_bpf_consume() - scx_bpf_dispatch[_vtime]_from_dsq*() This overloading of the term is historical. Originally, there were only built-in DSQs and moving a task into a DSQ always dispatched it for execution. Using the verb "dispatch" for the kfuncs to move tasks into these DSQs made sense. Later, user DSQs were added and scx_bpf_dispatch[_vtime]() updated to be able to insert tasks into any DSQ. The only allowed DSQ to DSQ transfer was from a non-local DSQ to a local DSQ and this operation was named "consume". This was already confusing as a task could be dispatched to a user DSQ from ops.enqueue() and then the DSQ would have to be consumed in ops.dispatch(). Later addition of scx_bpf_dispatch_from_dsq*() made the confusion even worse as "dispatch" in this context meant moving a task to an arbitrary DSQ from a user DSQ. Clean up the API with the following renames: 1. scx_bpf_dispatch[_vtime]() -> scx_bpf_dsq_insert[_vtime]() 2. scx_bpf_consume() -> scx_bpf_dsq_move_to_local() 3. scx_bpf_dispatch[_vtime]_from_dsq*() -> scx_bpf_dsq_move[_vtime]*() This patch performs the third set of renames. Compatibility is maintained by: - The previous kfunc names are still provided by the kernel so that old binaries can run. Kernel generates a warning when the old names are used. - compat.bpf.h provides wrappers for the new names which automatically fall back to the old names when running on older kernels. They also trigger build error if old names are used for new builds. - scx_bpf_dispatch[_vtime]_from_dsq*() were already wrapped in __COMPAT macros as they were introduced during v6.12 cycle. Wrap new API in __COMPAT macros too and trigger build errors on both __COMPAT prefixed and naked usages of the old names. The compat features will be dropped after v6.15. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: Andrea Righi <arighi@nvidia.com> Acked-by: Changwoo Min <changwoo@igalia.com> Acked-by: Johannes Bechberger <me@mostlynerdless.de> Acked-by: Giovanni Gherdovich <ggherdovich@suse.com> Cc: Dan Schatzberg <dschatzberg@meta.com> Cc: Ming Yang <yougmark94@gmail.com>
2024-11-11sched_ext: Rename scx_bpf_consume() to scx_bpf_dsq_move_to_local()Tejun Heo
In sched_ext API, a repeatedly reported pain point is the overuse of the verb "dispatch" and confusion around "consume": - ops.dispatch() - scx_bpf_dispatch[_vtime]() - scx_bpf_consume() - scx_bpf_dispatch[_vtime]_from_dsq*() This overloading of the term is historical. Originally, there were only built-in DSQs and moving a task into a DSQ always dispatched it for execution. Using the verb "dispatch" for the kfuncs to move tasks into these DSQs made sense. Later, user DSQs were added and scx_bpf_dispatch[_vtime]() updated to be able to insert tasks into any DSQ. The only allowed DSQ to DSQ transfer was from a non-local DSQ to a local DSQ and this operation was named "consume". This was already confusing as a task could be dispatched to a user DSQ from ops.enqueue() and then the DSQ would have to be consumed in ops.dispatch(). Later addition of scx_bpf_dispatch_from_dsq*() made the confusion even worse as "dispatch" in this context meant moving a task to an arbitrary DSQ from a user DSQ. Clean up the API with the following renames: 1. scx_bpf_dispatch[_vtime]() -> scx_bpf_dsq_insert[_vtime]() 2. scx_bpf_consume() -> scx_bpf_dsq_move_to_local() 3. scx_bpf_dispatch[_vtime]_from_dsq*() -> scx_bpf_dsq_move[_vtime]*() This patch performs the second rename. Compatibility is maintained by: - The previous kfunc names are still provided by the kernel so that old binaries can run. Kernel generates a warning when the old names are used. - compat.bpf.h provides wrappers for the new names which automatically fall back to the old names when running on older kernels. They also trigger build error if old names are used for new builds. The compat features will be dropped after v6.15. v2: Comment and documentation updates. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: Andrea Righi <arighi@nvidia.com> Acked-by: Changwoo Min <changwoo@igalia.com> Acked-by: Johannes Bechberger <me@mostlynerdless.de> Acked-by: Giovanni Gherdovich <ggherdovich@suse.com> Cc: Dan Schatzberg <dschatzberg@meta.com> Cc: Ming Yang <yougmark94@gmail.com>
2024-11-11sched_ext: Rename scx_bpf_dispatch[_vtime]() to scx_bpf_dsq_insert[_vtime]()Tejun Heo
In sched_ext API, a repeatedly reported pain point is the overuse of the verb "dispatch" and confusion around "consume": - ops.dispatch() - scx_bpf_dispatch[_vtime]() - scx_bpf_consume() - scx_bpf_dispatch[_vtime]_from_dsq*() This overloading of the term is historical. Originally, there were only built-in DSQs and moving a task into a DSQ always dispatched it for execution. Using the verb "dispatch" for the kfuncs to move tasks into these DSQs made sense. Later, user DSQs were added and scx_bpf_dispatch[_vtime]() updated to be able to insert tasks into any DSQ. The only allowed DSQ to DSQ transfer was from a non-local DSQ to a local DSQ and this operation was named "consume". This was already confusing as a task could be dispatched to a user DSQ from ops.enqueue() and then the DSQ would have to be consumed in ops.dispatch(). Later addition of scx_bpf_dispatch_from_dsq*() made the confusion even worse as "dispatch" in this context meant moving a task to an arbitrary DSQ from a user DSQ. Clean up the API with the following renames: 1. scx_bpf_dispatch[_vtime]() -> scx_bpf_dsq_insert[_vtime]() 2. scx_bpf_consume() -> scx_bpf_dsq_move_to_local() 3. scx_bpf_dispatch[_vtime]_from_dsq*() -> scx_bpf_dsq_move[_vtime]*() This patch performs the first set of renames. Compatibility is maintained by: - The previous kfunc names are still provided by the kernel so that old binaries can run. Kernel generates a warning when the old names are used. - compat.bpf.h provides wrappers for the new names which automatically fall back to the old names when running on older kernels. They also trigger build error if old names are used for new builds. The compat features will be dropped after v6.15. v2: Documentation updates. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: Andrea Righi <arighi@nvidia.com> Acked-by: Changwoo Min <changwoo@igalia.com> Acked-by: Johannes Bechberger <me@mostlynerdless.de> Acked-by: Giovanni Gherdovich <ggherdovich@suse.com> Cc: Dan Schatzberg <dschatzberg@meta.com> Cc: Ming Yang <yougmark94@gmail.com>
2024-11-09sched_ext: scx_bpf_dispatch_from_dsq_set_*() are allowed from unlocked contextTejun Heo
4c30f5ce4f7a ("sched_ext: Implement scx_bpf_dispatch[_vtime]_from_dsq()") added four kfuncs for dispatching while iterating. They are allowed from the dispatch and unlocked contexts but two of the kfuncs were only added in the dispatch section. Add missing declarations in the unlocked section. Signed-off-by: Tejun Heo <tj@kernel.org> Fixes: 4c30f5ce4f7a ("sched_ext: Implement scx_bpf_dispatch[_vtime]_from_dsq()")
2024-11-09sched_ext: add a missing rcu_read_lock/unlock pair at scx_select_cpu_dfl()Changwoo Min
When getting an LLC CPU mask in the default CPU selection policy, scx_select_cpu_dfl(), a pointer to the sched_domain is dereferenced using rcu_read_lock() without holding rcu_read_lock(). Such an unprotected dereference often causes the following warning and can cause an invalid memory access in the worst case. Therefore, protect dereference of a sched_domain pointer using a pair of rcu_read_lock() and unlock(). [ 20.996135] ============================= [ 20.996345] WARNING: suspicious RCU usage [ 20.996563] 6.11.0-virtme #17 Tainted: G W [ 20.996576] ----------------------------- [ 20.996576] kernel/sched/ext.c:3323 suspicious rcu_dereference_check() usage! [ 20.996576] [ 20.996576] other info that might help us debug this: [ 20.996576] [ 20.996576] [ 20.996576] rcu_scheduler_active = 2, debug_locks = 1 [ 20.996576] 4 locks held by kworker/8:1/140: [ 20.996576] #0: ffff8b18c00dd348 ((wq_completion)pm){+.+.}-{0:0}, at: process_one_work+0x4a0/0x590 [ 20.996576] #1: ffffb3da01f67e58 ((work_completion)(&dev->power.work)){+.+.}-{0:0}, at: process_one_work+0x1ba/0x590 [ 20.996576] #2: ffffffffa316f9f0 (&rcu_state.gp_wq){..-.}-{2:2}, at: swake_up_one+0x15/0x60 [ 20.996576] #3: ffff8b1880398a60 (&p->pi_lock){-.-.}-{2:2}, at: try_to_wake_up+0x59/0x7d0 [ 20.996576] [ 20.996576] stack backtrace: [ 20.996576] CPU: 8 UID: 0 PID: 140 Comm: kworker/8:1 Tainted: G W 6.11.0-virtme #17 [ 20.996576] Tainted: [W]=WARN [ 20.996576] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Arch Linux 1.16.3-1-1 04/01/2014 [ 20.996576] Workqueue: pm pm_runtime_work [ 20.996576] Sched_ext: simple (disabling+all), task: runnable_at=-6ms [ 20.996576] Call Trace: [ 20.996576] <IRQ> [ 20.996576] dump_stack_lvl+0x6f/0xb0 [ 20.996576] lockdep_rcu_suspicious.cold+0x4e/0x96 [ 20.996576] scx_select_cpu_dfl+0x234/0x260 [ 20.996576] select_task_rq_scx+0xfb/0x190 [ 20.996576] select_task_rq+0x47/0x110 [ 20.996576] try_to_wake_up+0x110/0x7d0 [ 20.996576] swake_up_one+0x39/0x60 [ 20.996576] rcu_core+0xb08/0xe50 [ 20.996576] ? srso_alias_return_thunk+0x5/0xfbef5 [ 20.996576] ? mark_held_locks+0x40/0x70 [ 20.996576] handle_softirqs+0xd3/0x410 [ 20.996576] irq_exit_rcu+0x78/0xa0 [ 20.996576] sysvec_apic_timer_interrupt+0x73/0x80 [ 20.996576] </IRQ> [ 20.996576] <TASK> [ 20.996576] asm_sysvec_apic_timer_interrupt+0x1a/0x20 [ 20.996576] RIP: 0010:_raw_spin_unlock_irqrestore+0x36/0x70 [ 20.996576] Code: f5 53 48 8b 74 24 10 48 89 fb 48 83 c7 18 e8 11 b4 36 ff 48 89 df e8 99 0d 37 ff f7 c5 00 02 00 00 75 17 9c 58 f6 c4 02 75 2b <65> ff 0d 5b 55 3c 5e 74 16 5b 5d e9 95 8e 28 00 e8 a5 ee 44 ff 9c [ 20.996576] RSP: 0018:ffffb3da01f67d20 EFLAGS: 00000246 [ 20.996576] RAX: 0000000000000002 RBX: ffffffffa4640220 RCX: 0000000000000040 [ 20.996576] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffffffffa1c7b27b [ 20.996576] RBP: 0000000000000246 R08: 0000000000000001 R09: 0000000000000000 [ 20.996576] R10: 0000000000000001 R11: 000000000000021c R12: 0000000000000246 [ 20.996576] R13: ffff8b1881363958 R14: 0000000000000000 R15: ffff8b1881363800 [ 20.996576] ? _raw_spin_unlock_irqrestore+0x4b/0x70 [ 20.996576] serial_port_runtime_resume+0xd4/0x1a0 [ 20.996576] ? __pfx_serial_port_runtime_resume+0x10/0x10 [ 20.996576] __rpm_callback+0x44/0x170 [ 20.996576] ? __pfx_serial_port_runtime_resume+0x10/0x10 [ 20.996576] rpm_callback+0x55/0x60 [ 20.996576] ? __pfx_serial_port_runtime_resume+0x10/0x10 [ 20.996576] rpm_resume+0x582/0x7b0 [ 20.996576] pm_runtime_work+0x7c/0xb0 [ 20.996576] process_one_work+0x1fb/0x590 [ 20.996576] worker_thread+0x18e/0x350 [ 20.996576] ? __pfx_worker_thread+0x10/0x10 [ 20.996576] kthread+0xe2/0x110 [ 20.996576] ? __pfx_kthread+0x10/0x10 [ 20.996576] ret_from_fork+0x34/0x50 [ 20.996576] ? __pfx_kthread+0x10/0x10 [ 20.996576] ret_from_fork_asm+0x1a/0x30 [ 20.996576] </TASK> [ 21.056592] sched_ext: BPF scheduler "simple" disabled (unregistered from user space) Signed-off-by: Changwoo Min <changwoo@igalia.com> Acked-by: Andrea Righi <arighi@nvidia.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-11-08sched_ext: Clarify sched_ext_ops table for userland schedulerChangwoo Min
Update the comments in sched_ext_ops to clarify this table is for a BPF scheduler and a userland scheduler should also rely on the sched_ext_ops table through the BPF scheduler. Signed-off-by: Changwoo Min <changwoo@igalia.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-11-08sched_ext: Enable the ops breather and eject BPF scheduler on softlockupTejun Heo
On 2 x Intel Sapphire Rapids machines with 224 logical CPUs, a poorly behaving BPF scheduler can live-lock the system by making multiple CPUs bang on the same DSQ to the point where soft-lockup detection triggers before SCX's own watchdog can take action. It also seems possible that the machine can be live-locked enough to prevent scx_ops_helper, which is an RT task, from running in a timely manner. Implement scx_softlockup() which is called when three quarters of soft-lockup threshold has passed. The function immediately enables the ops breather and triggers an ops error to initiate ejection of the BPF scheduler. The previous and this patch combined enable the kernel to reliably recover the system from live-lock conditions that can be triggered by a poorly behaving BPF scheduler on Intel dual socket systems. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Douglas Anderson <dianders@chromium.org> Cc: Andrew Morton <akpm@linux-foundation.org>
2024-11-08sched_ext: Avoid live-locking bypass mode switchingTejun Heo
A poorly behaving BPF scheduler can live-lock the system by e.g. incessantly banging on the same DSQ on a large NUMA system to the point where switching to the bypass mode can take a long time. Turning on the bypass mode requires dequeueing and re-enqueueing currently runnable tasks, if the DSQs that they are on are live-locked, this can take tens of seconds cascading into other failures. This was observed on 2 x Intel Sapphire Rapids machines with 224 logical CPUs. Inject artifical delays while the bypass mode is switching to guarantee timely completion. While at it, move __scx_ops_bypass_lock into scx_ops_bypass() and rename it to bypass_lock. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Valentin Andrei <vandrei@meta.com> Reported-by: Patrick Lu <patlu@meta.com>
2024-11-08Merge branch 'for-6.12-fixes' into for-6.13Tejun Heo
Pull sched_ext/for-6.12-fixes to receive 0e7ffff1b811 ("scx: Fix raciness in scx_ops_bypass()"). Planned updates for scx_ops_bypass() depends on it. Signed-off-by: Tejun Heo <tj@kernel.org>
2024-11-08sched_ext: Fix incorrect use of bitwise ANDAndrea Righi
There is no reason to use a bitwise AND when checking the conditions to enable NUMA optimization for the built-in CPU idle selection policy, so use a logical AND instead. Fixes: f6ce6b949304 ("sched_ext: Do not enable LLC/NUMA optimizations when domains overlap") Reported-by: Nathan Chancellor <nathan@kernel.org> Closes: https://lore.kernel.org/lkml/20241108181753.GA2681424@thelio-3990X/ Signed-off-by: Andrea Righi <arighi@nvidia.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-11-07sched_ext: Do not enable LLC/NUMA optimizations when domains overlapAndrea Righi
When the LLC and NUMA domains fully overlap, enabling both optimizations in the built-in idle CPU selection policy is redundant, as it leads to searching for an idle CPU within the same domain twice. Likewise, if all online CPUs are within a single LLC domain, LLC optimization is unnecessary. Therefore, detect overlapping domains and enable topology optimizations only when necessary. Moreover, rely on the online CPUs for this detection logic, instead of using the possible CPUs. Fixes: 860a45219bce ("sched_ext: Introduce NUMA awareness to the default idle selection policy") Signed-off-by: Andrea Righi <arighi@nvidia.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-11-05sched_ext: Update scx_show_state.py to match scx_ops_bypass_depth's new typeTejun Heo
0e7ffff1b811 ("scx: Fix raciness in scx_ops_bypass()") converted scx_ops_bypass_depth from an atomic to an int. Update scx_show_state.py accordingly. Signed-off-by: Tejun Heo <tj@kernel.org> Fixes: 0e7ffff1b811 ("scx: Fix raciness in scx_ops_bypass()")
2024-11-05sched_ext: Add a missing newline at the end of an error messageTejun Heo
Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-29sched_ext: Introduce NUMA awareness to the default idle selection policyAndrea Righi
Similarly to commit dfa4ed29b18c ("sched_ext: Introduce LLC awareness to the default idle selection policy"), extend the built-in idle CPU selection policy to also prioritize CPUs within the same NUMA node. With this change applied, the built-in CPU idle selection policy follows this logic: - always prioritize CPUs from fully idle SMT cores, - select the same CPU if possible, - select a CPU within the same LLC domain, - select a CPU within the same NUMA node. Both NUMA and LLC awareness features are enabled only when the system has multiple NUMA nodes or multiple LLC domains. In the future, we may want to improve the NUMA node selection to account the node distance from prev_cpu. Currently, the logic only tries to keep tasks running on the same NUMA node. If all CPUs within a node are busy, the next NUMA node is chosen randomly. Signed-off-by: Andrea Righi <arighi@nvidia.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-25sched_ext: Fix enq_last_no_enq_fails selftestTejun Heo
cc9877fb7677 ("sched_ext: Improve error reporting during loading") changed how load failures are reported so that more error context can be communicated. This breaks the enq_last_no_enq_fails test as attach no longer fails. The scheduler is guaranteed to be ejected on attach completion with full error information. Update enq_last_no_enq_fails so that it checks that the scheduler is ejected using ops.exit(). Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Vishal Chourasia <vishalc@linux.ibm.com> Link: http://lkml.kernel.org/r/Zxknp7RAVNjmdJSc@linux.ibm.com Fixes: cc9877fb7677 ("sched_ext: Improve error reporting during loading")
2024-10-25sched_ext: Make cast_mask() inlineTejun Heo
cast_mask() doesn't do any actual work and is defined in a header file. Force it to be inline. When it is not inlined and the function is not used, it can cause verificaiton failures like the following: # tools/testing/selftests/sched_ext/runner -t minimal ===== START ===== TEST: minimal DESCRIPTION: Verify we can load a fully minimal scheduler OUTPUT: libbpf: prog 'cast_mask': missing BPF prog type, check ELF section name '.text' libbpf: prog 'cast_mask': failed to load: -22 libbpf: failed to load object 'minimal' libbpf: failed to load BPF skeleton 'minimal': -22 ERR: minimal.c:20 Failed to open and load skel not ok 1 minimal # ===== END ===== Signed-off-by: Tejun Heo <tj@kernel.org> Fixes: a748db0c8c6a ("tools/sched_ext: Receive misc updates from SCX repo")
2024-10-25scx: Fix raciness in scx_ops_bypass()David Vernet
scx_ops_bypass() can currently race on the ops enable / disable path as follows: 1. scx_ops_bypass(true) called on enable path, bypass depth is set to 1 2. An op on the init path exits, which schedules scx_ops_disable_workfn() 3. scx_ops_bypass(false) is called on the disable path, and bypass depth is decremented to 0 4. kthread is scheduled to execute scx_ops_disable_workfn() 5. scx_ops_bypass(true) called, bypass depth set to 1 6. scx_ops_bypass() races when iterating over CPUs While it's not safe to take any blocking locks on the bypass path, it is safe to take a raw spinlock which cannot be preempted. This patch therefore updates scx_ops_bypass() to use a raw spinlock to synchronize, and changes scx_ops_bypass_depth to be a regular int. Without this change, we observe the following warnings when running the 'exit' sched_ext selftest (sometimes requires a couple of runs): .[root@virtme-ng sched_ext]# ./runner -t exit ===== START ===== TEST: exit ... [ 14.935078] WARNING: CPU: 2 PID: 360 at kernel/sched/ext.c:4332 scx_ops_bypass+0x1ca/0x280 [ 14.935126] Modules linked in: [ 14.935150] CPU: 2 UID: 0 PID: 360 Comm: sched_ext_ops_h Not tainted 6.11.0-virtme #24 [ 14.935192] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Arch Linux 1.16.3-1-1 04/01/2014 [ 14.935242] Sched_ext: exit (enabling+all) [ 14.935244] RIP: 0010:scx_ops_bypass+0x1ca/0x280 [ 14.935300] Code: ff ff ff e8 48 96 10 00 fb e9 08 ff ff ff c6 05 7b 34 e8 01 01 90 48 c7 c7 89 86 88 87 e8 be 1d f8 ff 90 0f 0b 90 90 eb 95 90 <0f> 0b 90 41 8b 84 24 24 0a 00 00 eb 97 90 0f 0b 90 41 8b 84 24 24 [ 14.935394] RSP: 0018:ffffb706c0957ce0 EFLAGS: 00010002 [ 14.935424] RAX: 0000000000000009 RBX: 0000000000000001 RCX: 00000000e3fb8b2a [ 14.935465] RDX: 0000000000000001 RSI: 0000000000000004 RDI: ffffffff88a4c080 [ 14.935512] RBP: 0000000000009b56 R08: 0000000000000004 R09: 00000003f12e520a [ 14.935555] R10: ffffffff863a9795 R11: 0000000000000000 R12: ffff8fc5fec31300 [ 14.935598] R13: ffff8fc5fec31318 R14: 0000000000000286 R15: 0000000000000018 [ 14.935642] FS: 0000000000000000(0000) GS:ffff8fc5fe680000(0000) knlGS:0000000000000000 [ 14.935684] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 14.935721] CR2: 0000557d92890b88 CR3: 000000002464a000 CR4: 0000000000750ef0 [ 14.935765] PKRU: 55555554 [ 14.935782] Call Trace: [ 14.935802] <TASK> [ 14.935823] ? __warn+0xce/0x220 [ 14.935850] ? scx_ops_bypass+0x1ca/0x280 [ 14.935881] ? report_bug+0xc1/0x160 [ 14.935909] ? handle_bug+0x61/0x90 [ 14.935934] ? exc_invalid_op+0x1a/0x50 [ 14.935959] ? asm_exc_invalid_op+0x1a/0x20 [ 14.935984] ? raw_spin_rq_lock_nested+0x15/0x30 [ 14.936019] ? scx_ops_bypass+0x1ca/0x280 [ 14.936046] ? srso_alias_return_thunk+0x5/0xfbef5 [ 14.936081] ? __pfx_scx_ops_disable_workfn+0x10/0x10 [ 14.936111] scx_ops_disable_workfn+0x146/0xac0 [ 14.936142] ? finish_task_switch+0xa9/0x2c0 [ 14.936172] ? srso_alias_return_thunk+0x5/0xfbef5 [ 14.936211] ? __pfx_scx_ops_disable_workfn+0x10/0x10 [ 14.936244] kthread_worker_fn+0x101/0x2c0 [ 14.936268] ? __pfx_kthread_worker_fn+0x10/0x10 [ 14.936299] kthread+0xec/0x110 [ 14.936327] ? __pfx_kthread+0x10/0x10 [ 14.936351] ret_from_fork+0x37/0x50 [ 14.936374] ? __pfx_kthread+0x10/0x10 [ 14.936400] ret_from_fork_asm+0x1a/0x30 [ 14.936427] </TASK> [ 14.936443] irq event stamp: 21002 [ 14.936467] hardirqs last enabled at (21001): [<ffffffff863aa35f>] resched_cpu+0x9f/0xd0 [ 14.936521] hardirqs last disabled at (21002): [<ffffffff863dd0ba>] scx_ops_bypass+0x11a/0x280 [ 14.936571] softirqs last enabled at (20642): [<ffffffff863683d7>] __irq_exit_rcu+0x67/0xd0 [ 14.936622] softirqs last disabled at (20637): [<ffffffff863683d7>] __irq_exit_rcu+0x67/0xd0 [ 14.936672] ---[ end trace 0000000000000000 ]--- [ 14.953282] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 14.953352] ------------[ cut here ]------------ [ 14.953383] WARNING: CPU: 2 PID: 360 at kernel/sched/ext.c:4335 scx_ops_bypass+0x1d8/0x280 [ 14.953428] Modules linked in: [ 14.953453] CPU: 2 UID: 0 PID: 360 Comm: sched_ext_ops_h Tainted: G W 6.11.0-virtme #24 [ 14.953505] Tainted: [W]=WARN [ 14.953527] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Arch Linux 1.16.3-1-1 04/01/2014 [ 14.953574] RIP: 0010:scx_ops_bypass+0x1d8/0x280 [ 14.953603] Code: c6 05 7b 34 e8 01 01 90 48 c7 c7 89 86 88 87 e8 be 1d f8 ff 90 0f 0b 90 90 eb 95 90 0f 0b 90 41 8b 84 24 24 0a 00 00 eb 97 90 <0f> 0b 90 41 8b 84 24 24 0a 00 00 eb 92 f3 0f 1e fa 49 8d 84 24 f0 [ 14.953693] RSP: 0018:ffffb706c0957ce0 EFLAGS: 00010046 [ 14.953722] RAX: 0000000000000001 RBX: 0000000000000000 RCX: 0000000000000001 [ 14.953763] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff8fc5fec31318 [ 14.953804] RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000 [ 14.953845] R10: ffffffff863a9795 R11: 0000000000000000 R12: ffff8fc5fec31300 [ 14.953888] R13: ffff8fc5fec31318 R14: 0000000000000286 R15: 0000000000000018 [ 14.953934] FS: 0000000000000000(0000) GS:ffff8fc5fe680000(0000) knlGS:0000000000000000 [ 14.953974] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 14.954009] CR2: 0000557d92890b88 CR3: 000000002464a000 CR4: 0000000000750ef0 [ 14.954052] PKRU: 55555554 [ 14.954068] Call Trace: [ 14.954085] <TASK> [ 14.954102] ? __warn+0xce/0x220 [ 14.954126] ? scx_ops_bypass+0x1d8/0x280 [ 14.954150] ? report_bug+0xc1/0x160 [ 14.954178] ? handle_bug+0x61/0x90 [ 14.954203] ? exc_invalid_op+0x1a/0x50 [ 14.954226] ? asm_exc_invalid_op+0x1a/0x20 [ 14.954250] ? raw_spin_rq_lock_nested+0x15/0x30 [ 14.954285] ? scx_ops_bypass+0x1d8/0x280 [ 14.954311] ? __mutex_unlock_slowpath+0x3a/0x260 [ 14.954343] scx_ops_disable_workfn+0xa3e/0xac0 [ 14.954381] ? __pfx_scx_ops_disable_workfn+0x10/0x10 [ 14.954413] kthread_worker_fn+0x101/0x2c0 [ 14.954442] ? __pfx_kthread_worker_fn+0x10/0x10 [ 14.954479] kthread+0xec/0x110 [ 14.954507] ? __pfx_kthread+0x10/0x10 [ 14.954530] ret_from_fork+0x37/0x50 [ 14.954553] ? __pfx_kthread+0x10/0x10 [ 14.954576] ret_from_fork_asm+0x1a/0x30 [ 14.954603] </TASK> [ 14.954621] irq event stamp: 21002 [ 14.954644] hardirqs last enabled at (21001): [<ffffffff863aa35f>] resched_cpu+0x9f/0xd0 [ 14.954686] hardirqs last disabled at (21002): [<ffffffff863dd0ba>] scx_ops_bypass+0x11a/0x280 [ 14.954735] softirqs last enabled at (20642): [<ffffffff863683d7>] __irq_exit_rcu+0x67/0xd0 [ 14.954782] softirqs last disabled at (20637): [<ffffffff863683d7>] __irq_exit_rcu+0x67/0xd0 [ 14.954829] ---[ end trace 0000000000000000 ]--- [ 15.022283] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 15.092282] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 15.149282] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) ok 1 exit # ===== END ===== And with it, the test passes without issue after 1000s of runs: .[root@virtme-ng sched_ext]# ./runner -t exit ===== START ===== TEST: exit DESCRIPTION: Verify we can cleanly exit a scheduler in multiple places OUTPUT: [ 7.412856] sched_ext: BPF scheduler "exit" enabled [ 7.427924] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 7.466677] sched_ext: BPF scheduler "exit" enabled [ 7.475923] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 7.512803] sched_ext: BPF scheduler "exit" enabled [ 7.532924] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 7.586809] sched_ext: BPF scheduler "exit" enabled [ 7.595926] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 7.661923] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 7.723923] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) ok 1 exit # ===== END ===== ============================= RESULTS: PASSED: 1 SKIPPED: 0 FAILED: 0 Fixes: f0e1a0643a59 ("sched_ext: Implement BPF extensible scheduler class") Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-25scx: Fix exit selftest to use custom DSQDavid Vernet
In commit 63fb3ec80516 ("sched_ext: Allow only user DSQs for scx_bpf_consume(), scx_bpf_dsq_nr_queued() and bpf_iter_scx_dsq_new()"), we updated the consume path to only accept user DSQs, thus making it invalid to consume SCX_DSQ_GLOBAL. This selftest was doing that, so let's create a custom DSQ and use that instead. The test now passes: [root@virtme-ng sched_ext]# ./runner -t exit ===== START ===== TEST: exit DESCRIPTION: Verify we can cleanly exit a scheduler in multiple places OUTPUT: [ 12.387229] sched_ext: BPF scheduler "exit" enabled [ 12.406064] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 12.453325] sched_ext: BPF scheduler "exit" enabled [ 12.474064] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 12.515241] sched_ext: BPF scheduler "exit" enabled [ 12.532064] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 12.592063] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 12.654063] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) [ 12.715062] sched_ext: BPF scheduler "exit" disabled (unregistered from BPF) ok 1 exit # ===== END ===== Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-24sched_ext: Replace set_arg_maybe_null() with __nullable CFI stub tagsTejun Heo
ops.dispatch() and ops.yield() may be fed a NULL task_struct pointer. set_arg_maybe_null() is used to tell the verifier that they should be NULL checked before being dereferenced. BPF now has an a lot prettier way to express this - tagging arguments in CFI stubs with __nullable. Replace set_arg_maybe_null() with __nullable CFI stub tags. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: David Vernet <void@manifault.com> Acked-by: Martin KaFai Lau <martin.lau@kernel.org> Acked-by: Alexei Starovoitov <ast@kernel.org>
2024-10-24sched_ext: Rename CFI stubs to names that are recognized by BPFTejun Heo
CFI stubs can be used to tag arguments with __nullable (and possibly other tags in the future) but for that to work the CFI stubs must have names that are recognized by BPF. Rename them. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: David Vernet <void@manifault.com> Acked-by: Martin KaFai Lau <martin.lau@kernel.org> Acked-by: Alexei Starovoitov <ast@kernel.org>
2024-10-24sched_ext: Fix function pointer type mismatches in BPF selftestsVishal Chourasia
Fix incompatible function pointer type warnings in sched_ext BPF selftests by explicitly casting the function pointers when initializing struct_ops. This addresses multiple -Wincompatible-function-pointer-types warnings from the clang compiler where function signatures didn't match exactly. The void * cast ensures the compiler accepts the function pointer assignment despite minor type differences in the parameters. Signed-off-by: Vishal Chourasia <vishalc@linux.ibm.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-23sched_ext: Introduce LLC awareness to the default idle selection policyAndrea Righi
Rely on the scheduler topology information to implement basic LLC awareness in the sched_ext build-in idle selection policy. This allows schedulers using the built-in policy to make more informed decisions when selecting an idle CPU in systems with multiple LLCs, such as NUMA systems or chiplet-based architectures, and it helps keep tasks within the same LLC domain, thereby improving cache locality. For efficiency, LLC awareness is applied only to tasks that can run on all the CPUs in the system for now. If a task's affinity is modified from user space, it's the responsibility of user space to choose the appropriate optimized scheduling domain. Signed-off-by: Andrea Righi <arighi@nvidia.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-23sched_ext: Clarify ops.select_cpu() for single-CPU tasksAndrea Righi
Update ops.select_cpu() documentation to clarify that this method is not called for tasks that are restricted to run on a single CPU, as these tasks do not have the option to select a different CPU. Signed-off-by: Andrea Righi <arighi@nvidia.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-23selftests/sched_ext: add order-only dependency of runner.o on BPFOBJIhor Solodrai
The runner.o may start building before libbpf headers are installed, and as a result build fails. This happened a couple of times on libbpf/ci test jobs: * https://github.com/libbpf/ci/actions/runs/11447667257/job/31849533100 * https://github.com/theihor/libbpf-ci/actions/runs/11445162764/job/31841649552 Headers are installed in a recipe for $(BPFOBJ) target, and adding an order-only dependency should ensure this doesn't happen. Signed-off-by: Ihor Solodrai <ihor.solodrai@pm.me> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-18sched_ext: improve WAKE_SYNC behavior for default idle CPU selectionAndrea Righi
In the sched_ext built-in idle CPU selection logic, when handling a WF_SYNC wakeup, we always attempt to migrate the task to the waker's CPU, as the waker is expected to yield the CPU after waking the task. However, it may be preferable to keep the task on its previous CPU if the waker's CPU is cache-affine. The same approach is also used by the fair class and in other scx schedulers, like scx_rusty and scx_bpfland. Therefore, apply the same logic to the built-in idle CPU selection policy as well. Signed-off-by: Andrea Righi <andrea.righi@linux.dev> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-17sched_ext: Use btf_ids to resolve task_structTianchen Ding
Save the searching time during bpf_scx_init. Signed-off-by: Tianchen Ding <dtcccc@linux.alibaba.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-14sched_ext: Remove unnecessary cpu_relax()David Vernet
As described in commit b07996c7abac ("sched_ext: Don't hold scx_tasks_lock for too long"), we're doing a cond_resched() every 32 calls to scx_task_iter_next() to avoid RCU and other stalls. That commit also added a cpu_relax() to the codepath where we drop and reacquire the lock, but as Waiman described in [0], cpu_relax() should only be necessary in busy loops to avoid pounding on a cacheline (or to allow a hypertwin to more fully utilize a core). Let's remove the unnecessary cpu_relax(). [0]: https://lore.kernel.org/all/35b3889b-904a-4d26-981f-c8aa1557a7c7@redhat.com/ Cc: Waiman Long <llong@redhat.com> Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-10sched_ext: Don't hold scx_tasks_lock for too longTejun Heo
While enabling and disabling a BPF scheduler, every task is iterated a couple times by walking scx_tasks. Except for one, all iterations keep holding scx_tasks_lock. On multi-socket systems under heavy rq lock contention and high number of threads, this can can lead to RCU and other stalls. The following is triggered on a 2 x AMD EPYC 7642 system (192 logical CPUs) running `stress-ng --workload 150 --workload-threads 10` with >400k idle threads and RCU stall period reduced to 5s: rcu: INFO: rcu_preempt detected stalls on CPUs/tasks: rcu: 91-...!: (10 ticks this GP) idle=0754/1/0x4000000000000000 softirq=18204/18206 fqs=17 rcu: 186-...!: (17 ticks this GP) idle=ec54/1/0x4000000000000000 softirq=25863/25866 fqs=17 rcu: (detected by 80, t=10042 jiffies, g=89305, q=33 ncpus=192) Sending NMI from CPU 80 to CPUs 91: NMI backtrace for cpu 91 CPU: 91 UID: 0 PID: 284038 Comm: sched_ext_ops_h Kdump: loaded Not tainted 6.12.0-rc2-work-g6bf5681f7ee2-dirty #471 Hardware name: Supermicro Super Server/H11DSi, BIOS 2.8 12/14/2023 Sched_ext: simple (disabling+all) RIP: 0010:queued_spin_lock_slowpath+0x17b/0x2f0 Code: 02 c0 10 03 00 83 79 08 00 75 08 f3 90 83 79 08 00 74 f8 48 8b 11 48 85 d2 74 09 0f 0d 0a eb 0a 31 d2 eb 06 31 d2 eb 02 f3 90 <8b> 07 66 85 c0 75 f7 39 d8 75 0d be 01 00 00 00 89 d8 f0 0f b1 37 RSP: 0018:ffffc9000fadfcb8 EFLAGS: 00000002 RAX: 0000000001700001 RBX: 0000000001700000 RCX: ffff88bfcaaf10c0 RDX: 0000000000000000 RSI: 0000000000000101 RDI: ffff88bfca8f0080 RBP: 0000000001700000 R08: 0000000000000090 R09: ffffffffffffffff R10: ffff88a74761b268 R11: 0000000000000000 R12: ffff88a6b6765460 R13: ffffc9000fadfd60 R14: ffff88bfca8f0080 R15: ffff88bfcaac0000 FS: 0000000000000000(0000) GS:ffff88bfcaac0000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f5c55f526a0 CR3: 0000000afd474000 CR4: 0000000000350eb0 Call Trace: <NMI> </NMI> <TASK> do_raw_spin_lock+0x9c/0xb0 task_rq_lock+0x50/0x190 scx_task_iter_next_locked+0x157/0x170 scx_ops_disable_workfn+0x2c2/0xbf0 kthread_worker_fn+0x108/0x2a0 kthread+0xeb/0x110 ret_from_fork+0x36/0x40 ret_from_fork_asm+0x1a/0x30 </TASK> Sending NMI from CPU 80 to CPUs 186: NMI backtrace for cpu 186 CPU: 186 UID: 0 PID: 51248 Comm: fish Kdump: loaded Not tainted 6.12.0-rc2-work-g6bf5681f7ee2-dirty #471 scx_task_iter can safely drop locks while iterating. Make scx_task_iter_next() drop scx_tasks_lock every 32 iterations to avoid stalls. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: David Vernet <void@manifault.com>
2024-10-10sched_ext: Move scx_tasks_lock handling into scx_task_iter helpersTejun Heo
Iterating with scx_task_iter involves scx_tasks_lock and optionally the rq lock of the task being iterated. Both locks can be released during iteration and the iteration can be continued after re-grabbing scx_tasks_lock. Currently, all lock handling is pushed to the caller which is a bit cumbersome and makes it difficult to add lock-aware behaviors. Make the scx_task_iter helpers handle scx_tasks_lock. - scx_task_iter_init/scx_taks_iter_exit() now grabs and releases scx_task_lock, respectively. Renamed to scx_task_iter_start/scx_task_iter_stop() to more clearly indicate that there are non-trivial side-effects. - Add __ prefix to scx_task_iter_rq_unlock() to indicate that the function is internal. - Add scx_task_iter_unlock/relock(). The former drops both rq lock (if held) and scx_tasks_lock and the latter re-locks only scx_tasks_lock. This doesn't cause behavior changes and will be used to implement stall avoidance. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: David Vernet <void@manifault.com>
2024-10-10sched_ext: bypass mode shouldn't depend on ops.select_cpu()Tejun Heo
Bypass mode was depending on ops.select_cpu() which can't be trusted as with the rest of the BPF scheduler. Always enable and use scx_select_cpu_dfl() in bypass mode. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: David Vernet <void@manifault.com>
2024-10-10sched_ext: Move scx_buildin_idle_enabled check to scx_bpf_select_cpu_dfl()Tejun Heo
Move the sanity check from the inner function scx_select_cpu_dfl() to the exported kfunc scx_bpf_select_cpu_dfl(). This doesn't cause behavior differences and will allow using scx_select_cpu_dfl() in bypass mode regardless of scx_builtin_idle_enabled. Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-10sched_ext: Start schedulers with consistent p->scx.slice valuesTejun Heo
The disable path caps p->scx.slice to SCX_SLICE_DFL. As the field is already being ignored at this stage during disable, the only effect this has is that when the next BPF scheduler is loaded, it won't see unreasonable left-over slices. Ultimately, this shouldn't matter but it's better to start in a known state. Drop p->scx.slice capping from the disable path and instead reset it to SCX_SLICE_DFL in the enable path. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: David Vernet <void@manifault.com>
2024-10-10Revert "sched_ext: Use shorter slice while bypassing"Tejun Heo
This reverts commit 6f34d8d382d64e7d8e77f5a9ddfd06f4c04937b0. Slice length is ignored while bypassing and tasks are switched on every tick and thus the patch does not make any difference. The perceived difference was from test noise. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: David Vernet <void@manifault.com>
2024-10-10sched_ext: use correct function name in pick_task_scx() warning messageHonglei Wang
pick_next_task_scx() was turned into pick_task_scx() since commit 753e2836d139 ("sched_ext: Unify regular and core-sched pick task paths"). Update the outdated message. Signed-off-by: Honglei Wang <jameshongleiwang@126.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-09selftests: sched_ext: Add sched_ext as proper selftest targetBjörn Töpel
The sched_ext selftests is missing proper cross-compilation support, a proper target entry, and out-of-tree build support. When building the kselftest suite, e.g.: make ARCH=riscv CROSS_COMPILE=riscv64-linux-gnu- \ TARGETS=sched_ext SKIP_TARGETS="" O=/output/foo \ -C tools/testing/selftests install or: make ARCH=arm64 LLVM=1 TARGETS=sched_ext SKIP_TARGETS="" \ O=/output/foo -C tools/testing/selftests install The expectation is that the sched_ext is included, cross-built, the correct toolchain is picked up, and placed into /output/foo. In contrast to the BPF selftests, the sched_ext suite does not use bpftool at test run-time, so it is sufficient to build bpftool for the build host only. Add ARCH, CROSS_COMPILE, OUTPUT, and TARGETS support to the sched_ext selftest. Also, remove some variables that were unused by the Makefile. Signed-off-by: Björn Töpel <bjorn@rivosinc.com> Reviewed-by: Shuah Khan <skhan@linuxfoundation.org> Acked-by: David Vernet <void@manifault.com> Tested-by: Mark Brown <broonie@kernel.org> Reviewed-by: Mark Brown <broonie@kernel.org> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-08sched_ext: Documentation: Update instructions for running example schedulersDevaansh-Kumar
Since the artifact paths for tools changed, we need to update the documentation to reflect that path. Signed-off-by: Devaansh-Kumar <devaanshk840@gmail.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-10-07sched_ext, scx_qmap: Add and use SCX_ENQ_CPU_SELECTEDTejun Heo
scx_qmap and other schedulers in the SCX repo are using SCX_ENQ_WAKEUP to tell whether ops.select_cpu() was called. This is incorrect as ops.select_cpu() can be skipped in the wakeup path and leads to e.g. incorrectly skipping direct dispatch for tasks that are bound to a single CPU. sched core has been updated to specify ENQUEUE_RQ_SELECTED if ->select_task_rq() was called. Map it to SCX_ENQ_CPU_SELECTED and update scx_qmap to test it instead of SCX_ENQ_WAKEUP. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: David Vernet <void@manifault.com> Cc: Daniel Hodges <hodges.daniel.scott@gmail.com> Cc: Changwoo Min <multics69@gmail.com> Cc: Andrea Righi <andrea.righi@linux.dev> Cc: Dan Schatzberg <schatzberg.dan@gmail.com>
2024-10-07sched/core: Add ENQUEUE_RQ_SELECTED to indicate whether ->select_task_rq() ↵Tejun Heo
was called During ttwu, ->select_task_rq() can be skipped if only one CPU is allowed or migration is disabled. sched_ext schedulers may perform operations such as direct dispatch from ->select_task_rq() path and it is useful for them to know whether ->select_task_rq() was skipped in the ->enqueue_task() path. Currently, sched_ext schedulers are using ENQUEUE_WAKEUP for this purpose and end up assuming incorrectly that ->select_task_rq() was called for tasks that are bound to a single CPU or migration disabled. Make select_task_rq() indicate whether ->select_task_rq() was called by setting WF_RQ_SELECTED in *wake_flags and make ttwu_do_activate() map that to ENQUEUE_RQ_SELECTED for ->enqueue_task(). This will be used by sched_ext to fix ->select_task_rq() skip detection. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: David Vernet <void@manifault.com>
2024-10-07sched/core: Make select_task_rq() take the pointer to wake_flags instead of ↵Tejun Heo
value This will be used to allow select_task_rq() to indicate whether ->select_task_rq() was called by modifying *wake_flags. This makes try_to_wake_up() call all functions that take wake_flags with WF_TTWU set. Previously, only select_task_rq() was. Using the same flags is more consistent, and, as the flag is only tested by ->select_task_rq() implementations, it doesn't cause any behavior differences. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: David Vernet <void@manifault.com>
2024-10-04sched_ext: scx_cgroup_exit() may be called without successful scx_cgroup_init()Tejun Heo
568894edbe48 ("sched_ext: Add scx_cgroup_enabled to gate cgroup operations and fix scx_tg_online()") assumed that scx_cgroup_exit() is only called after scx_cgroup_init() finished successfully. This isn't true. scx_cgroup_exit() can be called without scx_cgroup_init() being called at all or after scx_cgroup_init() failed in the middle. As init state is tracked per cgroup, scx_cgroup_exit() can be used safely to clean up in all cases. Remove the incorrect WARN_ON_ONCE(). Signed-off-by: Tejun Heo <tj@kernel.org> Fixes: 568894edbe48 ("sched_ext: Add scx_cgroup_enabled to gate cgroup operations and fix scx_tg_online()")
2024-10-04sched_ext: Improve error reporting during loadingTejun Heo
When the BPF scheduler fails, ops.exit() allows rich error reporting through scx_exit_info. Use scx.exit() path consistently for all failures which can be caused by the BPF scheduler: - scx_ops_error() is called after ops.init() and ops.cgroup_init() failure to record error information. - ops.init_task() failure now uses scx_ops_error() instead of pr_err(). - The err_disable path updated to automatically trigger scx_ops_error() to cover cases that the error message hasn't already been generated and always return 0 indicating init success so that the error is reported through ops.exit(). Signed-off-by: Tejun Heo <tj@kernel.org> Cc: David Vernet <void@manifault.com> Cc: Daniel Hodges <hodges.daniel.scott@gmail.com> Cc: Changwoo Min <multics69@gmail.com> Cc: Andrea Righi <andrea.righi@linux.dev> Cc: Dan Schatzberg <schatzberg.dan@gmail.com>
2024-10-02sched_ext: Add __weak markers to BPF helper function decalarationsVishal Chourasia
Fix build errors by adding __weak markers to BPF helper function declarations in header files. This resolves static assertion failures in scx_qmap.bpf.c and scx_flatcg.bpf.c where functions like scx_bpf_dispatch_from_dsq_set_slice, scx_bpf_dispatch_from_dsq_set_vtime, and scx_bpf_task_cgroup were missing the __weak attribute. [1] https://lore.kernel.org/all/ZvvfUqRNM4-jYQzH@linux.ibm.com Signed-off-by: Vishal Chourasia <vishalc@linux.ibm.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-09-27sched/ext: Use tg_cgroup() to elieminate duplicate codeZhang Qiao
Use tg_cgroup() to eliminate duplicate code patterns in scx_bpf_task_cgroup(). Signed-off-by: Zhang Qiao <zhangqiao22@huawei.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-09-27sched/ext: Fix unmatch trailing comment of CONFIG_EXT_GROUP_SCHEDZhang Qiao
The #endif trailing comment of CONFIG_EXT_GROUP_SCHED is unmatched, so fix it. Signed-off-by: Zhang Qiao <zhangqiao22@huawei.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-09-27sched_ext: Factor out move_task_between_dsqs() from scx_dispatch_from_dsq()Tejun Heo
Pure reorganization. No functional changes. Signed-off-by: Tejun Heo <tj@kernel.org>
2024-09-27sched_ext: Remove redundant p->nr_cpus_allowed checkerZhang Qiao
select_rq_task() already checked that 'p->nr_cpus_allowed > 1', 'p->nr_cpus_allowed == 1' checker in scx_select_cpu_dfl() is redundant. Signed-off-by: Zhang Qiao <zhangqiao22@huawei.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-09-27sched_ext: Decouple locks in scx_ops_enable()Tejun Heo
The enable path uses three big locks - scx_fork_rwsem, scx_cgroup_rwsem and cpus_read_lock. Currently, the locks are grabbed together which is prone to locking order problems. For example, currently, there is a possible deadlock involving scx_fork_rwsem and cpus_read_lock. cpus_read_lock has to nest inside scx_fork_rwsem due to locking order existing in other subsystems. However, there exists a dependency in the other direction during hotplug if hotplug needs to fork a new task, which happens in some cases. This leads to the following deadlock: scx_ops_enable() hotplug percpu_down_write(&cpu_hotplug_lock) percpu_down_write(&scx_fork_rwsem) block on cpu_hotplug_lock kthread_create() waits for kthreadd kthreadd blocks on scx_fork_rwsem Note that this doesn't trigger lockdep because the hotplug side dependency bounces through kthreadd. With the preceding scx_cgroup_enabled change, this can be solved by decoupling cpus_read_lock, which is needed for static_key manipulations, from the other two locks. - Move the first block of static_key manipulations outside of scx_fork_rwsem and scx_cgroup_rwsem. This is now safe with the preceding scx_cgroup_enabled change. - Drop scx_cgroup_rwsem and scx_fork_rwsem between the two task iteration blocks so that __scx_ops_enabled static_key enabling is outside the two rwsems. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-and-tested-by: Aboorva Devarajan <aboorvad@linux.ibm.com> Link: http://lkml.kernel.org/r/8cd0ec0c4c7c1bc0119e61fbef0bee9d5e24022d.camel@linux.ibm.com
2024-09-27sched_ext: Decouple locks in scx_ops_disable_workfn()Tejun Heo
The disable path uses three big locks - scx_fork_rwsem, scx_cgroup_rwsem and cpus_read_lock. Currently, the locks are grabbed together which is prone to locking order problems. With the preceding scx_cgroup_enabled change, we can decouple them: - As cgroup disabling no longer requires modifying a static_key which requires cpus_read_lock(), no need to grab cpus_read_lock() before grabbing scx_cgroup_rwsem. - cgroup can now be independently disabled before tasks are moved back to the fair class. Relocate scx_cgroup_exit() invocation before scx_fork_rwsem is grabbed, drop now unnecessary cpus_read_lock() and move static_key operations out of scx_fork_rwsem. This decouples all three locks in the disable path. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-and-tested-by: Aboorva Devarajan <aboorvad@linux.ibm.com> Link: http://lkml.kernel.org/r/8cd0ec0c4c7c1bc0119e61fbef0bee9d5e24022d.camel@linux.ibm.com
2024-09-27sched_ext: Add scx_cgroup_enabled to gate cgroup operations and fix ↵Tejun Heo
scx_tg_online() If the BPF scheduler does not implement ops.cgroup_init(), scx_tg_online() didn't set SCX_TG_INITED which meant that ops.cgroup_exit(), even if implemented, won't be called from scx_tg_offline(). This is because SCX_HAS_OP(cgroupt_init) is used to test both whether SCX cgroup operations are enabled and ops.cgroup_init() exists. Fix it by introducing a separate bool scx_cgroup_enabled to gate cgroup operations and use SCX_HAS_OP(cgroup_init) only to test whether ops.cgroup_init() exists. Make all cgroup operations consistently use scx_cgroup_enabled to test whether cgroup operations are enabled. scx_cgroup_enabled is added instead of using scx_enabled() to ease planned locking updates. Signed-off-by: Tejun Heo <tj@kernel.org>
generated by cgit (git 2.34.1) at 2025-09-07 08:58:13 +0000