kernel/linux.git/tools/sched_ext, branch v7.0.11

tools/sched_ext: Fix off-by-one in scx_sdt payload zeroing

2026-05-23T11:08:54+00:00

[ Upstream commit a3c3fb2f86f8a1f266747622037f90eab58186ad ] scx_alloc_free_idx() zeroes the payload of a freed arena allocation one word at a time. The loop bound was alloc->pool.elem_size / 8, but elem_size includes sizeof(struct sdt_data) (the 8-byte union sdt_id header). This caused the loop to write one extra u64 past the allocation, corrupting the tid field of the adjacent pool element. Fix the loop bound to (elem_size - sizeof(struct sdt_data)) / 8 so only the payload portion is zeroed. Test plan: - Add a temporary sanity check in scx_task_free() before the free call: if (mval->data->tid.idx != mval->tid.idx) scx_bpf_error("tid corruption: arena=%d storage=%d", mval->data->tid.idx, (int)mval->tid.idx); - stress-ng --fork 100 -t 10 & sudo ./build/bin/scx_sdt Without this fix, running scx_sdt under fork-heavy load triggers the corruption error. With the fix applied, the same workload completes without error. Fixes: 36929ebd17ae ("tools/sched_ext: add arena based scheduler") Signed-off-by: Cheng-Yang Chou Reviewed-by: Emil Tsalapatis Signed-off-by: Tejun Heo Signed-off-by: Sasha Levin

tools/sched_ext: scx_pair: fix pair_ctx indexing for CPU pairs

2026-05-23T11:08:48+00:00

[ Upstream commit f546c77038ab898726e7344255217fbec382b97f ] scx_pair sizes pair_ctx to nr_cpu_ids / 2, so valid pair_ctx keys are dense pair indexes in the range [0, nr_cpu_ids / 2). However, the userspace setup code stores pair_id as the first CPU number in each pair. On an 8-CPU system with "-S 1", that produces pair IDs 0, 2, 4 and 6 for pairs [0,1], [2,3], [4,5] and [6,7]. CPUs in the latter half then look up pair_ctx with out-of-range keys and the BPF scheduler aborts with: EXIT: scx_bpf_error (scx_pair.bpf.c:328: failed to lookup pairc and in_pair_mask for cpu[5]) Assign pair_id using a dense pair counter instead so that each CPU pair maps to a valid pair_ctx entry. Besides, reject odd CPU configuration, as scx_pair requires all CPUs to be paired. Fixes: f0262b102c7c ("tools/sched_ext: add scx_pair scheduler") Signed-off-by: Zhao Mengmeng Signed-off-by: Tejun Heo Signed-off-by: Sasha Levin

tools/sched_ext: Add -fms-extensions to bpf build flags

2026-03-03T08:00:23+00:00

Similar to commit 835a50753579 ("selftests/bpf: Add -fms-extensions to bpf build flags") and commit 639f58a0f480 ("bpftool: Fix build warnings due to MS extensions") The kernel is now built with -fms-extensions, therefore generated vmlinux.h contains types like: struct aes_key { struct aes_enckey; union aes_invkey_arch inv_k; }; struct ns_common { ... union { struct ns_tree; struct callback_head ns_rcu; }; }; Which raise warning like below when building scx scheduler: tools/sched_ext/build/include/vmlinux.h:50533:3: warning: declaration does not declare anything [-Wmissing-declarations] 50533 | struct ns_tree; | ^ Fix it by using -fms-extensions and -Wno-microsoft-anon-tag flags to build bpf programs that #include "vmlinux.h" Signed-off-by: Zhao Mengmeng Reviewed-by: Andrea Righi Signed-off-by: Tejun Heo

tools/sched_ext: fix strtoul() misuse in scx_hotplug_seq()

2026-02-27T19:17:44+00:00

scx_hotplug_seq() uses strtoul() but validates the result with a negative check (val < 0), which can never be true for an unsigned return value. Use the endptr mechanism to verify the entire string was consumed, and check errno == ERANGE for overflow detection. Signed-off-by: David Carlier Signed-off-by: Tejun Heo

tools/sched_ext: Add Kconfig to sync with upstream

2026-02-24T17:51:36+00:00

Add the missing Kconfig file to tools/sched_ext/ as referenced in the README. Ref: https://github.com/sched-ext/scx/blob/main/kernel.config Signed-off-by: Cheng-Yang Chou Signed-off-by: Tejun Heo

tools/sched_ext: Sync README.md Kconfig with upstream scx

2026-02-24T17:51:29+00:00

Sync the documentation with the upstream scx repository to reflect the current recommended configuration. Ref: https://github.com/sched-ext/scx/blob/main/README.md#build--install Signed-off-by: Cheng-Yang Chou Signed-off-by: Tejun Heo

tools/sched_ext: scx_sdt: Remove unused '-f' option

2026-02-23T17:45:34+00:00

The '-f' option is defined in getopt() but not handled in the switch statement or documented in the help text. Providing '-f' currently triggers the default error path. Remove it to sync the optstring with the actual implementation. Signed-off-by: Cheng-Yang Chou Signed-off-by: Tejun Heo

tools/sched_ext: scx_central: Remove unused '-p' option

2026-02-23T17:45:30+00:00

The '-p' option is defined in getopt() but not handled in the switch statement or documented in the help text. Providing '-p' currently triggers the default error path. Remove it to sync the optstring with the actual implementation. Signed-off-by: Cheng-Yang Chou Signed-off-by: Tejun Heo

tools/sched_ext: fix getopt not re-parsed on restart

2026-02-21T03:17:38+00:00

After goto restart, optind retains its advanced position from the previous getopt loop, causing getopt() to immediately return -1. This silently drops all command-line options on the restarted skeleton. Reset optind to 1 at the restart label so options are re-parsed. Affected schedulers: scx_simple, scx_central, scx_flatcg, scx_pair, scx_sdt, scx_cpu0. Signed-off-by: David Carlier Signed-off-by: Tejun Heo

tools/sched_ext: scx_userland: fix data races on shared counters

2026-02-21T03:17:31+00:00

The stats thread reads nr_vruntime_enqueues, nr_vruntime_dispatches, nr_vruntime_failed, and nr_curr_enqueued concurrently with the main thread writing them, with no synchronization. Use __atomic builtins with relaxed ordering for all accesses to these counters to eliminate the data races. Only display accuracy is affected, not scheduling correctness. Signed-off-by: David Carlier Signed-off-by: Tejun Heo