[PATCH 01/17] sched_ext: Add ext_types.h for early subsystem-wide defs

[PATCH 01/17] sched_ext: Add ext_types.h for early subsystem-wide defs

[Tejun Heo]

Introduce kernel/sched/ext_types.h as the early-def header for the
sched_ext compilation unit. Included from kernel/sched/build_policy.c
before ext_internal.h so every later header and source in the unit
sees its content without re-inclusion. Later patches add their types
here (struct scx_cid_topo, scx_cmask, scx_cid_shard, etc.) so the
subsystem has one place to stash types shared across the TU.

Move enum scx_consts (SCX_DSP_DFL_MAX_BATCH, SCX_WATCHDOG_MAX_TIMEOUT,
SCX_SUB_MAX_DEPTH, etc.) here as the initial content. Ops-facing
content stays in ext_internal.h.

Signed-off-by: Tejun Heo <tj@kernel.org>
---
 kernel/sched/build_policy.c |  1 +
 kernel/sched/ext_internal.h | 32 ---------------------------
 kernel/sched/ext_types.h    | 43 +++++++++++++++++++++++++++++++++++++
 3 files changed, 44 insertions(+), 32 deletions(-)
 create mode 100644 kernel/sched/ext_types.h

diff --git a/kernel/sched/build_policy.c b/kernel/sched/build_policy.c
index 755883faf751..180ade38625e 100644
--- a/kernel/sched/build_policy.c
+++ b/kernel/sched/build_policy.c
@@ -58,6 +58,7 @@
 #include "deadline.c"
 
 #ifdef CONFIG_SCHED_CLASS_EXT
+# include "ext_types.h"
 # include "ext_internal.h"
 # include "ext.c"
 # include "ext_idle.c"
diff --git a/kernel/sched/ext_internal.h b/kernel/sched/ext_internal.h
index 4a7ffc7f55d2..c35098668fb1 100644
--- a/kernel/sched/ext_internal.h
+++ b/kernel/sched/ext_internal.h
@@ -8,38 +8,6 @@
 #define SCX_OP_IDX(op)		(offsetof(struct sched_ext_ops, op) / sizeof(void (*)(void)))
 #define SCX_MOFF_IDX(moff)	((moff) / sizeof(void (*)(void)))
 
-enum scx_consts {
-	SCX_DSP_DFL_MAX_BATCH		= 32,
-	SCX_DSP_MAX_LOOPS		= 32,
-	SCX_WATCHDOG_MAX_TIMEOUT	= 30 * HZ,
-
-	/* per-CPU chunk size for p->scx.tid allocation, see scx_alloc_tid() */
-	SCX_TID_CHUNK			= 1024,
-
-	SCX_EXIT_BT_LEN			= 64,
-	SCX_EXIT_MSG_LEN		= 1024,
-	SCX_EXIT_DUMP_DFL_LEN		= 32768,
-
-	SCX_CPUPERF_ONE			= SCHED_CAPACITY_SCALE,
-
-	/*
-	 * Iterating all tasks may take a while. Periodically drop
-	 * scx_tasks_lock to avoid causing e.g. CSD and RCU stalls.
-	 */
-	SCX_TASK_ITER_BATCH		= 32,
-
-	SCX_BYPASS_HOST_NTH		= 2,
-
-	SCX_BYPASS_LB_DFL_INTV_US	= 500 * USEC_PER_MSEC,
-	SCX_BYPASS_LB_DONOR_PCT		= 125,
-	SCX_BYPASS_LB_MIN_DELTA_DIV	= 4,
-	SCX_BYPASS_LB_BATCH		= 256,
-
-	SCX_REENQ_LOCAL_MAX_REPEAT	= 256,
-
-	SCX_SUB_MAX_DEPTH		= 4,
-};
-
 enum scx_exit_kind {
 	SCX_EXIT_NONE,
 	SCX_EXIT_DONE,
diff --git a/kernel/sched/ext_types.h b/kernel/sched/ext_types.h
new file mode 100644
index 000000000000..19299ec3920e
--- /dev/null
+++ b/kernel/sched/ext_types.h
@@ -0,0 +1,43 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Early sched_ext type definitions.
+ *
+ * Copyright (c) 2026 Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2026 Tejun Heo <tj@kernel.org>
+ */
+#ifndef _KERNEL_SCHED_EXT_TYPES_H
+#define _KERNEL_SCHED_EXT_TYPES_H
+
+enum scx_consts {
+	SCX_DSP_DFL_MAX_BATCH		= 32,
+	SCX_DSP_MAX_LOOPS		= 32,
+	SCX_WATCHDOG_MAX_TIMEOUT	= 30 * HZ,
+
+	/* per-CPU chunk size for p->scx.tid allocation, see scx_alloc_tid() */
+	SCX_TID_CHUNK			= 1024,
+
+	SCX_EXIT_BT_LEN			= 64,
+	SCX_EXIT_MSG_LEN		= 1024,
+	SCX_EXIT_DUMP_DFL_LEN		= 32768,
+
+	SCX_CPUPERF_ONE			= SCHED_CAPACITY_SCALE,
+
+	/*
+	 * Iterating all tasks may take a while. Periodically drop
+	 * scx_tasks_lock to avoid causing e.g. CSD and RCU stalls.
+	 */
+	SCX_TASK_ITER_BATCH		= 32,
+
+	SCX_BYPASS_HOST_NTH		= 2,
+
+	SCX_BYPASS_LB_DFL_INTV_US	= 500 * USEC_PER_MSEC,
+	SCX_BYPASS_LB_DONOR_PCT		= 125,
+	SCX_BYPASS_LB_MIN_DELTA_DIV	= 4,
+	SCX_BYPASS_LB_BATCH		= 256,
+
+	SCX_REENQ_LOCAL_MAX_REPEAT	= 256,
+
+	SCX_SUB_MAX_DEPTH		= 4,
+};
+
+#endif /* _KERNEL_SCHED_EXT_TYPES_H */
-- 
2.53.0

[PATCHSET v2 REPOST sched_ext/for-7.2] sched_ext: Topological CPU IDs and cid-form struct_ops

[Tejun Heo]

Hello,

Reposting v2 because the original send was not properly threaded -
each patch went out as a standalone top-level message. Content is
unchanged from the original v2.

Original v2: https://lore.kernel.org/r/20260424013220.2923402-1-tj@kernel.org

v2 of https://lore.kernel.org/r/20260421071945.3110084-1-tj@kernel.org

v2:
- Add ext-types.h first patch for early subsystem-wide type defs.
- cid: publish the cid tables with WRITE_ONCE / read with READ_ONCE;
  document the visibility contract.
- cid-kfuncs: NULL-guard scx_bpf_this_cid / scx_bpf_task_cid for
  TRACING/SYSCALL callers before any SCX sched has enabled.
- cid-struct-ops: use struct_size() for the set_cmask_scratch percpu
  alloc; cluster __scx_is_cid_type disable with __scx_enabled disable
  in scx_root_disable().
- cid-kfunc-filter: sync per-entry kfunc flags with each kfunc's
  primary BTF_ID_FLAGS() declaration (Zhao). pahole intersects flags
  across occurrences; omitting them drops the flags globally - the
  visible symptom was KF_IMPLICIT_ARGS getting cleared on
  scx_bpf_kick_cpu, leaking bpf_prog_aux into vmlinux.h.
- cmask: narrow to the helpers this series actually uses;
  cmask_copy_from_kernel contract and runtime guard.

This patchset introduces topological CPU IDs (cids) - dense,
topology-ordered cpu identifiers - and an alternative cid-form struct_ops
type that lets BPF schedulers operate in cid space directly.

Key pieces:

- cid space: scx_cid_init() walks nodes * LLCs * cores * threads and packs
  a dense cid mapping. The mapping can be overridden via
  scx_bpf_cid_override(). See "Topological CPU IDs" in ext_cid.h for the
  model.

- cmask: a base-windowed bitmap over cid space. Kernel and BPF helpers with
  identical semantics. Used by scx_qmap for per-task affinity and idle-cid
  tracking; meant to be the substrate for sub-sched cid allocation.

- bpf_sched_ext_ops_cid: a parallel struct_ops type whose callbacks take
  cids/cmasks instead of cpus/cpumasks. Kernel translates at the boundary
  via scx_cpu_arg() / scx_cpu_ret(); the two struct types share offsets up
  through @priv (verified by BUILD_BUG_ON) so the union view in scx_sched
  works without function-pointer casts. Sub-sched support is tied to
  cid-form: validate_ops() rejects cpu-form sub-scheds and cpu-form roots
  that expose sub_attach / sub_detach.

- cid-form kfuncs: scx_bpf_kick_cid, scx_bpf_cidperf_{cap,cur,set},
  scx_bpf_cid_curr, scx_bpf_task_cid, scx_bpf_this_cid,
  scx_bpf_nr_{cids,online_cids}, scx_bpf_cid_to_cpu, scx_bpf_cpu_to_cid.
  A cid-form program may not call cpu-only kfuncs (enforced at verifier
  load via scx_kfunc_context_filter); the reverse is intentionally
  permissive to ease migration.

- scx_qmap port: scx_qmap is converted to cid-form. It uses the cmask-based
  idle picker, per-task cid-space cpus_allowed, and cid-form kfuncs
  throughout. Sub-sched dispatching via scx_bpf_sub_dispatch() continues to
  work.

v2 re-tested on the 16-cpu QEMU: cid-form scx_qmap, cpu-form scx_simple,
cid<->cpu cycling, scx_qmap under stress-ng, hotplug auto-restart, and
sub-sched (root scx_qmap + cgroup-scoped scx_qmap child). Clean.

Based on sched_ext/for-7.2 (c2929bc21dce).

 0001-sched_ext-Add-ext_types.h-for-early-subsystem-wide-d.patch
 0002-sched_ext-Rename-ops_cpu_valid-to-scx_cpu_valid-and-.patch
 0003-sched_ext-Move-scx_exit-scx_error-and-friends-to-ext.patch
 0004-sched_ext-Shift-scx_kick_cpu-validity-check-to-scx_b.patch
 0005-sched_ext-Relocate-cpu_acquire-cpu_release-to-end-of.patch
 0006-sched_ext-Make-scx_enable-take-scx_enable_cmd.patch
 0007-sched_ext-Add-topological-CPU-IDs-cids.patch
 0008-sched_ext-Add-scx_bpf_cid_override-kfunc.patch
 0009-tools-sched_ext-Add-struct_size-helpers-to-common.bp.patch
 0010-sched_ext-Add-cmask-a-base-windowed-bitmap-over-cid-.patch
 0011-sched_ext-Add-cid-form-kfunc-wrappers-alongside-cpu-.patch
 0012-sched_ext-Add-bpf_sched_ext_ops_cid-struct_ops-type.patch
 0013-sched_ext-Forbid-cpu-form-kfuncs-from-cid-form-sched.patch
 0014-tools-sched_ext-scx_qmap-Restart-on-hotplug-instead-.patch
 0015-tools-sched_ext-scx_qmap-Add-cmask-based-idle-tracki.patch
 0016-tools-sched_ext-scx_qmap-Port-to-cid-form-struct_ops.patch
 0017-sched_ext-Require-cid-form-struct_ops-for-sub-sched-.patch

Git tree: git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext.git scx-cid-v2

 kernel/sched/build_policy.c              |   2 +
 kernel/sched/ext.c                       | 650 +++++++++++++++++++++++++----
 kernel/sched/ext_cid.c                   | 417 ++++++++++++++++++++
 kernel/sched/ext_cid.h                   | 164 ++++++++
 kernel/sched/ext_idle.c                  |   8 +-
 kernel/sched/ext_internal.h              | 203 +++++++---
 kernel/sched/ext_types.h                 | 104 +++++
 tools/sched_ext/include/scx/cid.bpf.h    | 597 ++++++++++++++++++++++++++++
 tools/sched_ext/include/scx/common.bpf.h |  23 ++
 tools/sched_ext/include/scx/compat.bpf.h |  24 ++
 tools/sched_ext/scx_qmap.bpf.c           | 306 ++++++++-------
 tools/sched_ext/scx_qmap.c               |  25 +-
 tools/sched_ext/scx_qmap.h               |   2 +-
 13 files changed, 2240 insertions(+), 285 deletions(-)

--
tejun

[PATCH 01/17] sched_ext: Add ext_types.h for early subsystem-wide defs

[Tejun Heo]

Introduce kernel/sched/ext_types.h as the early-def header for the
sched_ext compilation unit. Included from kernel/sched/build_policy.c
before ext_internal.h so every later header and source in the unit
sees its content without re-inclusion. Later patches add their types
here (struct scx_cid_topo, scx_cmask, scx_cid_shard, etc.) so the
subsystem has one place to stash types shared across the TU.

Move enum scx_consts (SCX_DSP_DFL_MAX_BATCH, SCX_WATCHDOG_MAX_TIMEOUT,
SCX_SUB_MAX_DEPTH, etc.) here as the initial content. Ops-facing
content stays in ext_internal.h.

Signed-off-by: Tejun Heo <tj@kernel.org>
---
 kernel/sched/build_policy.c |  1 +
 kernel/sched/ext_internal.h | 32 ---------------------------
 kernel/sched/ext_types.h    | 43 +++++++++++++++++++++++++++++++++++++
 3 files changed, 44 insertions(+), 32 deletions(-)
 create mode 100644 kernel/sched/ext_types.h

diff --git a/kernel/sched/build_policy.c b/kernel/sched/build_policy.c
index 755883faf751..180ade38625e 100644
--- a/kernel/sched/build_policy.c
+++ b/kernel/sched/build_policy.c
@@ -58,6 +58,7 @@
 #include "deadline.c"
 
 #ifdef CONFIG_SCHED_CLASS_EXT
+# include "ext_types.h"
 # include "ext_internal.h"
 # include "ext.c"
 # include "ext_idle.c"
diff --git a/kernel/sched/ext_internal.h b/kernel/sched/ext_internal.h
index 4a7ffc7f55d2..c35098668fb1 100644
--- a/kernel/sched/ext_internal.h
+++ b/kernel/sched/ext_internal.h
@@ -8,38 +8,6 @@
 #define SCX_OP_IDX(op)		(offsetof(struct sched_ext_ops, op) / sizeof(void (*)(void)))
 #define SCX_MOFF_IDX(moff)	((moff) / sizeof(void (*)(void)))
 
-enum scx_consts {
-	SCX_DSP_DFL_MAX_BATCH		= 32,
-	SCX_DSP_MAX_LOOPS		= 32,
-	SCX_WATCHDOG_MAX_TIMEOUT	= 30 * HZ,
-
-	/* per-CPU chunk size for p->scx.tid allocation, see scx_alloc_tid() */
-	SCX_TID_CHUNK			= 1024,
-
-	SCX_EXIT_BT_LEN			= 64,
-	SCX_EXIT_MSG_LEN		= 1024,
-	SCX_EXIT_DUMP_DFL_LEN		= 32768,
-
-	SCX_CPUPERF_ONE			= SCHED_CAPACITY_SCALE,
-
-	/*
-	 * Iterating all tasks may take a while. Periodically drop
-	 * scx_tasks_lock to avoid causing e.g. CSD and RCU stalls.
-	 */
-	SCX_TASK_ITER_BATCH		= 32,
-
-	SCX_BYPASS_HOST_NTH		= 2,
-
-	SCX_BYPASS_LB_DFL_INTV_US	= 500 * USEC_PER_MSEC,
-	SCX_BYPASS_LB_DONOR_PCT		= 125,
-	SCX_BYPASS_LB_MIN_DELTA_DIV	= 4,
-	SCX_BYPASS_LB_BATCH		= 256,
-
-	SCX_REENQ_LOCAL_MAX_REPEAT	= 256,
-
-	SCX_SUB_MAX_DEPTH		= 4,
-};
-
 enum scx_exit_kind {
 	SCX_EXIT_NONE,
 	SCX_EXIT_DONE,
diff --git a/kernel/sched/ext_types.h b/kernel/sched/ext_types.h
new file mode 100644
index 000000000000..19299ec3920e
--- /dev/null
+++ b/kernel/sched/ext_types.h
@@ -0,0 +1,43 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Early sched_ext type definitions.
+ *
+ * Copyright (c) 2026 Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2026 Tejun Heo <tj@kernel.org>
+ */
+#ifndef _KERNEL_SCHED_EXT_TYPES_H
+#define _KERNEL_SCHED_EXT_TYPES_H
+
+enum scx_consts {
+	SCX_DSP_DFL_MAX_BATCH		= 32,
+	SCX_DSP_MAX_LOOPS		= 32,
+	SCX_WATCHDOG_MAX_TIMEOUT	= 30 * HZ,
+
+	/* per-CPU chunk size for p->scx.tid allocation, see scx_alloc_tid() */
+	SCX_TID_CHUNK			= 1024,
+
+	SCX_EXIT_BT_LEN			= 64,
+	SCX_EXIT_MSG_LEN		= 1024,
+	SCX_EXIT_DUMP_DFL_LEN		= 32768,
+
+	SCX_CPUPERF_ONE			= SCHED_CAPACITY_SCALE,
+
+	/*
+	 * Iterating all tasks may take a while. Periodically drop
+	 * scx_tasks_lock to avoid causing e.g. CSD and RCU stalls.
+	 */
+	SCX_TASK_ITER_BATCH		= 32,
+
+	SCX_BYPASS_HOST_NTH		= 2,
+
+	SCX_BYPASS_LB_DFL_INTV_US	= 500 * USEC_PER_MSEC,
+	SCX_BYPASS_LB_DONOR_PCT		= 125,
+	SCX_BYPASS_LB_MIN_DELTA_DIV	= 4,
+	SCX_BYPASS_LB_BATCH		= 256,
+
+	SCX_REENQ_LOCAL_MAX_REPEAT	= 256,
+
+	SCX_SUB_MAX_DEPTH		= 4,
+};
+
+#endif /* _KERNEL_SCHED_EXT_TYPES_H */
-- 
2.53.0

[PATCHSET v3 sched_ext/for-7.2] sched_ext: Topological CPU IDs and cid-form struct_ops

[Tejun Heo]

Hello,

v3 (all from the Sashiko AI review at
https://sashiko.dev/#/patchset/20260424172721.3458520-1-tj%40kernel.org):

- cid: drop leaked cpus_read_lock() on scx_cid_init() failure;
  BUILD_BUG_ON tightened to NR_CPUS<=8192 to match the BPF cmask
  helpers' CMASK_MAX_WORDS coverage.
- bpf-struct-size: use offsetof() in struct_size() to match the
  kernel <linux/overflow.h> macro semantics (no inflation from
  trailing struct padding).
- cmask: cmask_copy_from_kernel() validates src->base==0 via
  probe-read; nr_bits check is bit-level rather than rounded-up
  word-count.
- cid-qmap-idle: qmap_init() refuses to load when scx_bpf_nr_cids()
  exceeds SCX_QMAP_MAX_CPUS; the task_ctx flex array would otherwise
  overflow into the next slab entry.

v2: https://lore.kernel.org/r/20260424172721.3458520-1-tj@kernel.org
v1: https://lore.kernel.org/r/20260421071945.3110084-1-tj@kernel.org

This patchset introduces topological CPU IDs (cids) - dense,
topology-ordered cpu identifiers - and an alternative cid-form struct_ops
type that lets BPF schedulers operate in cid space directly.

Key pieces:

- cid space: scx_cid_init() walks nodes * LLCs * cores * threads and packs
  a dense cid mapping. The mapping can be overridden via
  scx_bpf_cid_override(). See "Topological CPU IDs" in ext_cid.h for the
  model.

- cmask: a base-windowed bitmap over cid space. Kernel and BPF helpers with
  identical semantics. Used by scx_qmap for per-task affinity and idle-cid
  tracking; meant to be the substrate for sub-sched cid allocation.

- bpf_sched_ext_ops_cid: a parallel struct_ops type whose callbacks take
  cids/cmasks instead of cpus/cpumasks. Kernel translates at the boundary
  via scx_cpu_arg() / scx_cpu_ret(); the two struct types share offsets up
  through @priv (verified by BUILD_BUG_ON) so the union view in scx_sched
  works without function-pointer casts. Sub-sched support is tied to
  cid-form: validate_ops() rejects cpu-form sub-scheds and cpu-form roots
  that expose sub_attach / sub_detach.

- cid-form kfuncs: scx_bpf_kick_cid, scx_bpf_cidperf_{cap,cur,set},
  scx_bpf_cid_curr, scx_bpf_task_cid, scx_bpf_this_cid,
  scx_bpf_nr_{cids,online_cids}, scx_bpf_cid_to_cpu, scx_bpf_cpu_to_cid.
  A cid-form program may not call cpu-only kfuncs (enforced at verifier
  load via scx_kfunc_context_filter); the reverse is intentionally
  permissive to ease migration.

- scx_qmap port: scx_qmap is converted to cid-form. It uses the cmask-based
  idle picker, per-task cid-space cpus_allowed, and cid-form kfuncs
  throughout. Sub-sched dispatching via scx_bpf_sub_dispatch() continues to
  work.

v3 re-tested on the 16-cpu QEMU: cid-form scx_qmap under stress-ng plus
reload cycles, hotplug auto-restart, and sub-sched (root scx_qmap +
cgroup-scoped scx_qmap child). Clean.

Based on sched_ext/for-7.2 (4939721aad2e).

 0001-sched_ext-Add-ext_types.h-for-early-subsystem-wide-d.patch
 0002-sched_ext-Rename-ops_cpu_valid-to-scx_cpu_valid-and-.patch
 0003-sched_ext-Move-scx_exit-scx_error-and-friends-to-ext.patch
 0004-sched_ext-Shift-scx_kick_cpu-validity-check-to-scx_b.patch
 0005-sched_ext-Relocate-cpu_acquire-cpu_release-to-end-of.patch
 0006-sched_ext-Make-scx_enable-take-scx_enable_cmd.patch
 0007-sched_ext-Add-topological-CPU-IDs-cids.patch
 0008-sched_ext-Add-scx_bpf_cid_override-kfunc.patch
 0009-tools-sched_ext-Add-struct_size-helpers-to-common.bp.patch
 0010-sched_ext-Add-cmask-a-base-windowed-bitmap-over-cid-.patch
 0011-sched_ext-Add-cid-form-kfunc-wrappers-alongside-cpu-.patch
 0012-sched_ext-Add-bpf_sched_ext_ops_cid-struct_ops-type.patch
 0013-sched_ext-Forbid-cpu-form-kfuncs-from-cid-form-sched.patch
 0014-tools-sched_ext-scx_qmap-Restart-on-hotplug-instead-.patch
 0015-tools-sched_ext-scx_qmap-Add-cmask-based-idle-tracki.patch
 0016-tools-sched_ext-scx_qmap-Port-to-cid-form-struct_ops.patch
 0017-sched_ext-Require-cid-form-struct_ops-for-sub-sched-.patch

Git tree: git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext.git scx-cid-v3

 kernel/sched/build_policy.c              |   3 +
 kernel/sched/ext.c                       | 651 ++++++++++++++++++++++++++----
 kernel/sched/ext_cid.c                   | 409 +++++++++++++++++++
 kernel/sched/ext_cid.h                   | 164 ++++++++
 kernel/sched/ext_idle.c                  |   8 +-
 kernel/sched/ext_internal.h              | 205 +++++++---
 kernel/sched/ext_types.h                 | 104 +++++
 tools/sched_ext/include/scx/cid.bpf.h    | 667 +++++++++++++++++++++++++++++++
 tools/sched_ext/include/scx/common.bpf.h |  23 ++
 tools/sched_ext/include/scx/compat.bpf.h |  24 ++
 tools/sched_ext/scx_qmap.bpf.c           | 346 +++++++++-------
 tools/sched_ext/scx_qmap.c               |  70 +++-
 tools/sched_ext/scx_qmap.h               |   2 +-
 13 files changed, 2391 insertions(+), 285 deletions(-)

Thanks.

--
tejun

[PATCH 01/17] sched_ext: Add ext_types.h for early subsystem-wide defs

[Tejun Heo]

Introduce kernel/sched/ext_types.h as the early-def header for the
sched_ext compilation unit. Included from kernel/sched/build_policy.c
before ext_internal.h so every later header and source in the unit
sees its content without re-inclusion. Later patches add their types
here (struct scx_cid_topo, scx_cmask, scx_cid_shard, etc.) so the
subsystem has one place to stash types shared across the TU.

Move enum scx_consts (SCX_DSP_DFL_MAX_BATCH, SCX_WATCHDOG_MAX_TIMEOUT,
SCX_SUB_MAX_DEPTH, etc.) here as the initial content. Ops-facing
content stays in ext_internal.h.

Signed-off-by: Tejun Heo <tj@kernel.org>
---
 kernel/sched/build_policy.c |  1 +
 kernel/sched/ext_internal.h | 32 ---------------------------
 kernel/sched/ext_types.h    | 43 +++++++++++++++++++++++++++++++++++++
 3 files changed, 44 insertions(+), 32 deletions(-)
 create mode 100644 kernel/sched/ext_types.h

diff --git a/kernel/sched/build_policy.c b/kernel/sched/build_policy.c
index ffb386889218..1d92f7d7a19f 100644
--- a/kernel/sched/build_policy.c
+++ b/kernel/sched/build_policy.c
@@ -59,6 +59,7 @@
 
 #ifdef CONFIG_SCHED_CLASS_EXT
 # include <linux/btf_ids.h>
+# include "ext_types.h"
 # include "ext_internal.h"
 # include "ext_idle.h"
 # include "ext.c"
diff --git a/kernel/sched/ext_internal.h b/kernel/sched/ext_internal.h
index a54903bb74b3..1b2ea6fa9fd6 100644
--- a/kernel/sched/ext_internal.h
+++ b/kernel/sched/ext_internal.h
@@ -8,38 +8,6 @@
 #define SCX_OP_IDX(op)		(offsetof(struct sched_ext_ops, op) / sizeof(void (*)(void)))
 #define SCX_MOFF_IDX(moff)	((moff) / sizeof(void (*)(void)))
 
-enum scx_consts {
-	SCX_DSP_DFL_MAX_BATCH		= 32,
-	SCX_DSP_MAX_LOOPS		= 32,
-	SCX_WATCHDOG_MAX_TIMEOUT	= 30 * HZ,
-
-	/* per-CPU chunk size for p->scx.tid allocation, see scx_alloc_tid() */
-	SCX_TID_CHUNK			= 1024,
-
-	SCX_EXIT_BT_LEN			= 64,
-	SCX_EXIT_MSG_LEN		= 1024,
-	SCX_EXIT_DUMP_DFL_LEN		= 32768,
-
-	SCX_CPUPERF_ONE			= SCHED_CAPACITY_SCALE,
-
-	/*
-	 * Iterating all tasks may take a while. Periodically drop
-	 * scx_tasks_lock to avoid causing e.g. CSD and RCU stalls.
-	 */
-	SCX_TASK_ITER_BATCH		= 32,
-
-	SCX_BYPASS_HOST_NTH		= 2,
-
-	SCX_BYPASS_LB_DFL_INTV_US	= 500 * USEC_PER_MSEC,
-	SCX_BYPASS_LB_DONOR_PCT		= 125,
-	SCX_BYPASS_LB_MIN_DELTA_DIV	= 4,
-	SCX_BYPASS_LB_BATCH		= 256,
-
-	SCX_REENQ_LOCAL_MAX_REPEAT	= 256,
-
-	SCX_SUB_MAX_DEPTH		= 4,
-};
-
 enum scx_exit_kind {
 	SCX_EXIT_NONE,
 	SCX_EXIT_DONE,
diff --git a/kernel/sched/ext_types.h b/kernel/sched/ext_types.h
new file mode 100644
index 000000000000..19299ec3920e
--- /dev/null
+++ b/kernel/sched/ext_types.h
@@ -0,0 +1,43 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Early sched_ext type definitions.
+ *
+ * Copyright (c) 2026 Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2026 Tejun Heo <tj@kernel.org>
+ */
+#ifndef _KERNEL_SCHED_EXT_TYPES_H
+#define _KERNEL_SCHED_EXT_TYPES_H
+
+enum scx_consts {
+	SCX_DSP_DFL_MAX_BATCH		= 32,
+	SCX_DSP_MAX_LOOPS		= 32,
+	SCX_WATCHDOG_MAX_TIMEOUT	= 30 * HZ,
+
+	/* per-CPU chunk size for p->scx.tid allocation, see scx_alloc_tid() */
+	SCX_TID_CHUNK			= 1024,
+
+	SCX_EXIT_BT_LEN			= 64,
+	SCX_EXIT_MSG_LEN		= 1024,
+	SCX_EXIT_DUMP_DFL_LEN		= 32768,
+
+	SCX_CPUPERF_ONE			= SCHED_CAPACITY_SCALE,
+
+	/*
+	 * Iterating all tasks may take a while. Periodically drop
+	 * scx_tasks_lock to avoid causing e.g. CSD and RCU stalls.
+	 */
+	SCX_TASK_ITER_BATCH		= 32,
+
+	SCX_BYPASS_HOST_NTH		= 2,
+
+	SCX_BYPASS_LB_DFL_INTV_US	= 500 * USEC_PER_MSEC,
+	SCX_BYPASS_LB_DONOR_PCT		= 125,
+	SCX_BYPASS_LB_MIN_DELTA_DIV	= 4,
+	SCX_BYPASS_LB_BATCH		= 256,
+
+	SCX_REENQ_LOCAL_MAX_REPEAT	= 256,
+
+	SCX_SUB_MAX_DEPTH		= 4,
+};
+
+#endif /* _KERNEL_SCHED_EXT_TYPES_H */
-- 
2.54.0

[PATCHSET v4 sched_ext/for-7.2] sched_ext: Topological CPU IDs and cid-form struct_ops

[Tejun Heo]

Hello,

v4:
- cmask: bump CMASK_CAS_TRIES to (1U << 23) so abort fires only after
  seconds of real spinning, not on plausible contention. The kfunc
  slow-path Changwoo suggested would let BPF loops keep banging a
  contended cacheline indefinitely - on multi-socket SPRs that path
  can stall the machine into hard lockups, so failing hard is the
  right behavior. A follow-up patch will add a kfunc to bail the BPF
  CAS loops immediately when sch->aborting is set. Switch
  __builtin_ctzll() to the ctzll() wrapper for clang compat.
- cid-qmap-port: cid-shard handling was wired against a future kfunc
  signature that didn't make it into v3, leaving the snapshot broken.
  Drop the shard test plumbing for v4, match the 2-arg
  scx_bpf_cid_override(), bound nr_cpu_ids for the verifier, and
  rename mode 3 from bad-mono to bad-range. (Changwoo, Andrea)
- Rebased over the exit_cpu plumbing in for-7.2:
    scx-error-header: scx_exit() and scx_verror() are macros now;
    move both plus the underlying __scx_exit() / scx_vexit()
    declarations to ext_internal.h.
    cid-struct-ops: dump_cpu callsite shifted into scx_dump_cpu()
    helper; the scx_cpu_arg() wrap moved with it.

v3: https://lore.kernel.org/r/20260428203545.181052-1-tj@kernel.org
v2: https://lore.kernel.org/r/20260424172721.3458520-1-tj@kernel.org
v1: https://lore.kernel.org/r/20260421071945.3110084-1-tj@kernel.org

This patchset introduces topological CPU IDs (cids) - dense,
topology-ordered cpu identifiers - and an alternative cid-form struct_ops
type that lets BPF schedulers operate in cid space directly.

Key pieces:

- cid space: scx_cid_init() walks nodes * LLCs * cores * threads and packs
  a dense cid mapping. The mapping can be overridden via
  scx_bpf_cid_override(). See "Topological CPU IDs" in ext_cid.h for the
  model.

- cmask: a base-windowed bitmap over cid space. Kernel and BPF helpers with
  identical semantics. Used by scx_qmap for per-task affinity and idle-cid
  tracking; meant to be the substrate for sub-sched cid allocation.

- bpf_sched_ext_ops_cid: a parallel struct_ops type whose callbacks take
  cids/cmasks instead of cpus/cpumasks. Kernel translates at the boundary
  via scx_cpu_arg() / scx_cpu_ret(); the two struct types share offsets up
  through @priv (verified by BUILD_BUG_ON) so the union view in scx_sched
  works without function-pointer casts. Sub-sched support is tied to
  cid-form: validate_ops() rejects cpu-form sub-scheds and cpu-form roots
  that expose sub_attach / sub_detach.

- cid-form kfuncs: scx_bpf_kick_cid, scx_bpf_cidperf_{cap,cur,set},
  scx_bpf_cid_curr, scx_bpf_task_cid, scx_bpf_this_cid,
  scx_bpf_nr_{cids,online_cids}, scx_bpf_cid_to_cpu, scx_bpf_cpu_to_cid.
  A cid-form program may not call cpu-only kfuncs (enforced at verifier
  load via scx_kfunc_context_filter); the reverse is intentionally
  permissive to ease migration.

- scx_qmap port: scx_qmap is converted to cid-form. It uses the cmask-based
  idle picker, per-task cid-space cpus_allowed, and cid-form kfuncs
  throughout. Sub-sched dispatching via scx_bpf_sub_dispatch() continues to
  work.

v4 re-tested on the 16-cpu QEMU VM with the v3 cut (only the 17 cid
patches applied): basic load + stress, cid-override modes
(shuffle/bad-dup/bad-range), and three enable/disable cycles all clean.
No BUG/WARNING/panic in the dump.

Based on sched_ext/for-7.2 (ee8391ba1164).

  0001-sched_ext-Add-ext_types.h-for-early-subsystem-wide-d.patch
  0002-sched_ext-Rename-ops_cpu_valid-to-scx_cpu_valid-and-.patch
  0003-sched_ext-Move-scx_exit-scx_error-and-friends-to-ext.patch
  0004-sched_ext-Shift-scx_kick_cpu-validity-check-to-scx_b.patch
  0005-sched_ext-Relocate-cpu_acquire-cpu_release-to-end-of.patch
  0006-sched_ext-Make-scx_enable-take-scx_enable_cmd.patch
  0007-sched_ext-Add-topological-CPU-IDs-cids.patch
  0008-sched_ext-Add-scx_bpf_cid_override-kfunc.patch
  0009-tools-sched_ext-Add-struct_size-helpers-to-common.bp.patch
  0010-sched_ext-Add-cmask-a-base-windowed-bitmap-over-cid-.patch
  0011-sched_ext-Add-cid-form-kfunc-wrappers-alongside-cpu-.patch
  0012-sched_ext-Add-bpf_sched_ext_ops_cid-struct_ops-type.patch
  0013-sched_ext-Forbid-cpu-form-kfuncs-from-cid-form-sched.patch
  0014-tools-sched_ext-scx_qmap-Restart-on-hotplug-instead-.patch
  0015-tools-sched_ext-scx_qmap-Add-cmask-based-idle-tracki.patch
  0016-tools-sched_ext-scx_qmap-Port-to-cid-form-struct_ops.patch
  0017-sched_ext-Require-cid-form-struct_ops-for-sub-sched-.patch

Git tree: git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext.git scx-cid-v4

 kernel/sched/build_policy.c              |   3 +
 kernel/sched/ext.c                       | 660 ++++++++++++++++++++++++++----
 kernel/sched/ext_cid.c                   | 409 +++++++++++++++++++
 kernel/sched/ext_cid.h                   | 164 ++++++++
 kernel/sched/ext_idle.c                  |   8 +-
 kernel/sched/ext_internal.h              | 209 +++++++---
 kernel/sched/ext_types.h                 | 104 +++++
 tools/sched_ext/include/scx/cid.bpf.h    | 666 +++++++++++++++++++++++++++++++
 tools/sched_ext/include/scx/common.bpf.h |  23 ++
 tools/sched_ext/include/scx/compat.bpf.h |  24 ++
 tools/sched_ext/scx_qmap.bpf.c           | 350 +++++++++-------
 tools/sched_ext/scx_qmap.c               |  57 ++-
 tools/sched_ext/scx_qmap.h               |   2 +-
 13 files changed, 2387 insertions(+), 292 deletions(-)

Thanks.

--
tejun

[PATCH 01/17] sched_ext: Add ext_types.h for early subsystem-wide defs

[Tejun Heo]

Introduce kernel/sched/ext_types.h as the early-def header for the
sched_ext compilation unit. Included from kernel/sched/build_policy.c
before ext_internal.h so every later header and source in the unit
sees its content without re-inclusion. Later patches add their types
here (struct scx_cid_topo, scx_cmask, scx_cid_shard, etc.) so the
subsystem has one place to stash types shared across the TU.

Move enum scx_consts (SCX_DSP_DFL_MAX_BATCH, SCX_WATCHDOG_MAX_TIMEOUT,
SCX_SUB_MAX_DEPTH, etc.) here as the initial content. Ops-facing
content stays in ext_internal.h.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
---
 kernel/sched/build_policy.c |  1 +
 kernel/sched/ext_internal.h | 32 ---------------------------
 kernel/sched/ext_types.h    | 43 +++++++++++++++++++++++++++++++++++++
 3 files changed, 44 insertions(+), 32 deletions(-)
 create mode 100644 kernel/sched/ext_types.h

diff --git a/kernel/sched/build_policy.c b/kernel/sched/build_policy.c
index ffb386889218..1d92f7d7a19f 100644
--- a/kernel/sched/build_policy.c
+++ b/kernel/sched/build_policy.c
@@ -59,6 +59,7 @@
 
 #ifdef CONFIG_SCHED_CLASS_EXT
 # include <linux/btf_ids.h>
+# include "ext_types.h"
 # include "ext_internal.h"
 # include "ext_idle.h"
 # include "ext.c"
diff --git a/kernel/sched/ext_internal.h b/kernel/sched/ext_internal.h
index 54c6ed43b6c7..26668bbce2b3 100644
--- a/kernel/sched/ext_internal.h
+++ b/kernel/sched/ext_internal.h
@@ -8,38 +8,6 @@
 #define SCX_OP_IDX(op)		(offsetof(struct sched_ext_ops, op) / sizeof(void (*)(void)))
 #define SCX_MOFF_IDX(moff)	((moff) / sizeof(void (*)(void)))
 
-enum scx_consts {
-	SCX_DSP_DFL_MAX_BATCH		= 32,
-	SCX_DSP_MAX_LOOPS		= 32,
-	SCX_WATCHDOG_MAX_TIMEOUT	= 30 * HZ,
-
-	/* per-CPU chunk size for p->scx.tid allocation, see scx_alloc_tid() */
-	SCX_TID_CHUNK			= 1024,
-
-	SCX_EXIT_BT_LEN			= 64,
-	SCX_EXIT_MSG_LEN		= 1024,
-	SCX_EXIT_DUMP_DFL_LEN		= 32768,
-
-	SCX_CPUPERF_ONE			= SCHED_CAPACITY_SCALE,
-
-	/*
-	 * Iterating all tasks may take a while. Periodically drop
-	 * scx_tasks_lock to avoid causing e.g. CSD and RCU stalls.
-	 */
-	SCX_TASK_ITER_BATCH		= 32,
-
-	SCX_BYPASS_HOST_NTH		= 2,
-
-	SCX_BYPASS_LB_DFL_INTV_US	= 500 * USEC_PER_MSEC,
-	SCX_BYPASS_LB_DONOR_PCT		= 125,
-	SCX_BYPASS_LB_MIN_DELTA_DIV	= 4,
-	SCX_BYPASS_LB_BATCH		= 256,
-
-	SCX_REENQ_LOCAL_MAX_REPEAT	= 256,
-
-	SCX_SUB_MAX_DEPTH		= 4,
-};
-
 enum scx_exit_kind {
 	SCX_EXIT_NONE,
 	SCX_EXIT_DONE,
diff --git a/kernel/sched/ext_types.h b/kernel/sched/ext_types.h
new file mode 100644
index 000000000000..19299ec3920e
--- /dev/null
+++ b/kernel/sched/ext_types.h
@@ -0,0 +1,43 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Early sched_ext type definitions.
+ *
+ * Copyright (c) 2026 Meta Platforms, Inc. and affiliates.
+ * Copyright (c) 2026 Tejun Heo <tj@kernel.org>
+ */
+#ifndef _KERNEL_SCHED_EXT_TYPES_H
+#define _KERNEL_SCHED_EXT_TYPES_H
+
+enum scx_consts {
+	SCX_DSP_DFL_MAX_BATCH		= 32,
+	SCX_DSP_MAX_LOOPS		= 32,
+	SCX_WATCHDOG_MAX_TIMEOUT	= 30 * HZ,
+
+	/* per-CPU chunk size for p->scx.tid allocation, see scx_alloc_tid() */
+	SCX_TID_CHUNK			= 1024,
+
+	SCX_EXIT_BT_LEN			= 64,
+	SCX_EXIT_MSG_LEN		= 1024,
+	SCX_EXIT_DUMP_DFL_LEN		= 32768,
+
+	SCX_CPUPERF_ONE			= SCHED_CAPACITY_SCALE,
+
+	/*
+	 * Iterating all tasks may take a while. Periodically drop
+	 * scx_tasks_lock to avoid causing e.g. CSD and RCU stalls.
+	 */
+	SCX_TASK_ITER_BATCH		= 32,
+
+	SCX_BYPASS_HOST_NTH		= 2,
+
+	SCX_BYPASS_LB_DFL_INTV_US	= 500 * USEC_PER_MSEC,
+	SCX_BYPASS_LB_DONOR_PCT		= 125,
+	SCX_BYPASS_LB_MIN_DELTA_DIV	= 4,
+	SCX_BYPASS_LB_BATCH		= 256,
+
+	SCX_REENQ_LOCAL_MAX_REPEAT	= 256,
+
+	SCX_SUB_MAX_DEPTH		= 4,
+};
+
+#endif /* _KERNEL_SCHED_EXT_TYPES_H */
-- 
2.54.0