sched/headers: Introduce kernel/sched/build_utility.c and build multiple .c files there

Collect all utility functionality source code files into a single kernel/sched/build_utility.c file,
via #include-ing the .c files:

    kernel/sched/clock.c
    kernel/sched/completion.c
    kernel/sched/loadavg.c
    kernel/sched/swait.c
    kernel/sched/wait_bit.c
    kernel/sched/wait.c

CONFIG_CPU_FREQ:
    kernel/sched/cpufreq.c

CONFIG_CPU_FREQ_GOV_SCHEDUTIL:
    kernel/sched/cpufreq_schedutil.c

CONFIG_CGROUP_CPUACCT:
    kernel/sched/cpuacct.c

CONFIG_SCHED_DEBUG:
    kernel/sched/debug.c

CONFIG_SCHEDSTATS:
    kernel/sched/stats.c

CONFIG_SMP:
   kernel/sched/cpupri.c
   kernel/sched/stop_task.c
   kernel/sched/topology.c

CONFIG_SCHED_CORE:
   kernel/sched/core_sched.c

CONFIG_PSI:
   kernel/sched/psi.c

CONFIG_MEMBARRIER:
   kernel/sched/membarrier.c

CONFIG_CPU_ISOLATION:
   kernel/sched/isolation.c

CONFIG_SCHED_AUTOGROUP:
   kernel/sched/autogroup.c

The goal is to amortize the 60+ KLOC header bloat from over a dozen build units into
a single build unit.

The build time of build_utility.c also roughly matches the build time of core.c and
fair.c - allowing better load-balancing of scheduler-only rebuilds.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Peter Zijlstra <peterz@infradead.org>

Author: ingomolnar

kernel/sched/Makefile

@@ -22,18 +22,13 @@ ifneq ($(CONFIG_SCHED_OMIT_FRAME_POINTER),y)
 CFLAGS_core.o := $(PROFILING) -fno-omit-frame-pointer
 endif
 
-obj-y += core.o loadavg.o clock.o cputime.o
-obj-y += idle.o fair.o rt.o deadline.o
-obj-y += wait.o wait_bit.o swait.o completion.o
-
-obj-$(CONFIG_SMP) += cpupri.o cpudeadline.o topology.o stop_task.o pelt.o
-obj-$(CONFIG_SCHED_AUTOGROUP) += autogroup.o
-obj-$(CONFIG_SCHEDSTATS) += stats.o
-obj-$(CONFIG_SCHED_DEBUG) += debug.o
-obj-$(CONFIG_CGROUP_CPUACCT) += cpuacct.o
-obj-$(CONFIG_CPU_FREQ) += cpufreq.o
-obj-$(CONFIG_CPU_FREQ_GOV_SCHEDUTIL) += cpufreq_schedutil.o
-obj-$(CONFIG_MEMBARRIER) += membarrier.o
-obj-$(CONFIG_CPU_ISOLATION) += isolation.o
-obj-$(CONFIG_PSI) += psi.o
-obj-$(CONFIG_SCHED_CORE) += core_sched.o
+#
+# Build efficiency:
+#
+# These compilation units have roughly the same size and complexity - so their
+# build parallelizes well and finishes roughly at once:
+#
+obj-y += core.o
+obj-y += fair.o
+obj-y += build_utility.o
+obj-y += idle.o rt.o deadline.o cputime.o cpudeadline.o pelt.o

kernel/sched/autogroup.c

@@ -1,9 +1,8 @@
 // SPDX-License-Identifier: GPL-2.0
+
 /*
  * Auto-group scheduling implementation:
  */
-#include <linux/nospec.h>
-#include "sched.h"
 
 unsigned int __read_mostly sysctl_sched_autogroup_enabled = 1;
 static struct autogroup autogroup_default;

kernel/sched/build_utility.c

@@ -0,0 +1,70 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * These are various utility functions of the scheduler,
+ * built in a single compilation unit for build efficiency reasons.
+ *
+ * ( Incidentally, the size of the compilation unit is roughly
+ *   comparable to core.c, fair.c, smp.c and policy.c, the other
+ *   big compilation units. This helps balance build time, while
+ *   coalescing source files to amortize header inclusion
+ *   cost. )
+ */
+
+#include "sched.h"
+#include "sched-pelt.h"
+
+#include <linux/sched_clock.h>
+
+#include "clock.c"
+
+#ifdef CONFIG_CGROUP_CPUACCT
+# include "cpuacct.c"
+#endif
+
+#ifdef CONFIG_CPU_FREQ
+# include "cpufreq.c"
+#endif
+
+#ifdef CONFIG_CPU_FREQ_GOV_SCHEDUTIL
+# include "cpufreq_schedutil.c"
+#endif
+
+#ifdef CONFIG_SCHED_DEBUG
+# include "debug.c"
+#endif
+
+#ifdef CONFIG_SCHEDSTATS
+# include "stats.c"
+#endif
+
+#include "loadavg.c"
+#include "completion.c"
+#include "swait.c"
+#include "wait_bit.c"
+#include "wait.c"
+
+#ifdef CONFIG_SMP
+# include "cpupri.c"
+# include "stop_task.c"
+# include "topology.c"
+#endif
+
+#ifdef CONFIG_SCHED_CORE
+# include "core_sched.c"
+#endif
+
+#ifdef CONFIG_PSI
+# include "psi.c"
+#endif
+
+#ifdef CONFIG_MEMBARRIER
+# include "membarrier.c"
+#endif
+
+#ifdef CONFIG_CPU_ISOLATION
+# include "isolation.c"
+#endif
+
+#ifdef CONFIG_SCHED_AUTOGROUP
+# include "autogroup.c"
+#endif

kernel/sched/clock.c

@@ -53,8 +53,6 @@
  * that is otherwise invisible (TSC gets stopped).
  *
  */
-#include "sched.h"
-#include <linux/sched_clock.h>
 
 /*
  * Scheduler clock - returns current time in nanosec units.

kernel/sched/completion.c

@@ -1,4 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
+
 /*
  * Generic wait-for-completion handler;
  *
@@ -11,7 +12,6 @@
  * typically be used for exclusion which gives rise to priority inversion.
  * Waiting for completion is a typically sync point, but not an exclusion point.
  */
-#include "sched.h"
 
 /**
  * complete: - signals a single thread waiting on this completion

kernel/sched/core_sched.c

@@ -1,8 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0-only
 
-#include <linux/prctl.h>
-#include "sched.h"
-
 /*
  * A simple wrapper around refcount. An allocated sched_core_cookie's
  * address is used to compute the cookie of the task.

kernel/sched/cpuacct.c

@@ -1,12 +1,11 @@
 // SPDX-License-Identifier: GPL-2.0
+
 /*
  * CPU accounting code for task groups.
  *
  * Based on the work by Paul Menage (menage@google.com) and Balbir Singh
  * (balbir@in.ibm.com).
  */
-#include <asm/irq_regs.h>
-#include "sched.h"
 
 /* Time spent by the tasks of the CPU accounting group executing in ... */
 enum cpuacct_stat_index {

kernel/sched/cpufreq.c

@@ -5,9 +5,6 @@
  * Copyright (C) 2016, Intel Corporation
  * Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
  */
-#include <linux/cpufreq.h>
-
-#include "sched.h"
 
 DEFINE_PER_CPU(struct update_util_data __rcu *, cpufreq_update_util_data);
 

kernel/sched/cpufreq_schedutil.c

@@ -6,13 +6,6 @@
  * Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
  */
 
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
-#include "sched.h"
-
-#include <linux/sched/cpufreq.h>
-#include <trace/events/power.h>
-
 #define IOWAIT_BOOST_MIN	(SCHED_CAPACITY_SCALE / 8)
 
 struct sugov_tunables {

kernel/sched/cpupri.c

@@ -22,7 +22,6 @@
  *  worst case complexity of O(min(101, nr_domcpus)), though the scenario that
  *  yields the worst case search is fairly contrived.
  */
-#include "sched.h"
 
 /*
  * p->rt_priority   p->prio   newpri   cpupri