linux-kernel - [RFC][PATCH 1/4] sched: Force the address order of each sched class descriptor

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20191219214558.510271353@goodmis.org>
Date:   Thu, 19 Dec 2019 16:44:52 -0500
From:   Steven Rostedt <rostedt@...dmis.org>
To:     linux-kernel@...r.kernel.org
Cc:     Kirill Tkhai <tkhai@...dex.ru>,
        Kirill Tkhai <ktkhai@...tuozzo.com>,
        Peter Zijlstra <peterz@...radead.org>,
        "mingo@...hat.com" <mingo@...hat.com>,
        "juri.lelli@...hat.com" <juri.lelli@...hat.com>,
        "vincent.guittot@...aro.org" <vincent.guittot@...aro.org>,
        "dietmar.eggemann@....com" <dietmar.eggemann@....com>,
        "bsegall@...gle.com" <bsegall@...gle.com>,
        "mgorman@...e.de" <mgorman@...e.de>,
        Ingo Molnar <mingo@...nel.org>,
        Andrew Morton <akpm@...ux-foundation.org>
Subject: [RFC][PATCH 1/4] sched: Force the address order of each sched class descriptor

From: "Steven Rostedt (VMware)" <rostedt@...dmis.org>

In order to make a micro optimization in pick_next_task(), the order of the
sched class descriptor address must be in the same order as their priority
to each other. That is:

 &idle_sched_class < &fair_sched_class < &rt_sched_class <
 &dl_sched_class < &stop_sched_class

In order to guarantee this order of the sched class descriptors, add each
one into their own data section and force the order in the linker script.

Link: https://lore.kernel.org/r/157675913272.349305.8936736338884044103.stgit@localhost.localdomain
Signed-off-by: Steven Rostedt (VMware) <rostedt@...dmis.org>
---
 include/asm-generic/vmlinux.lds.h | 19 +++++++++++++++++++
 kernel/sched/deadline.c           |  3 ++-
 kernel/sched/fair.c               |  3 ++-
 kernel/sched/idle.c               |  3 ++-
 kernel/sched/rt.c                 |  3 ++-
 kernel/sched/stop_task.c          |  3 ++-
 6 files changed, 29 insertions(+), 5 deletions(-)

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index e00f41aa8ec4..772d961c69a5 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -108,6 +108,24 @@
 #define SBSS_MAIN .sbss
 #endif
 
+#ifdef CONFIG_SMP
+#define STOP_SCHED_CLASS	*(__stop_sched_class)
+#else
+#define STOP_SCHED_CLASS
+#endif
+
+/*
+ * The order of the sched class addresses are important, as they are
+ * used to determine the order of the priority of each sched class in
+ * relation to each other.
+ */
+#define SCHED_DATA				\
+	*(__idle_sched_class)			\
+	*(__fair_sched_class)			\
+	*(__rt_sched_class)			\
+	*(__dl_sched_class)			\
+	STOP_SCHED_CLASS
+
 /*
  * Align to a 32 byte boundary equal to the
  * alignment gcc 4.5 uses for a struct
@@ -308,6 +326,7 @@
 #define DATA_DATA							\
 	*(.xiptext)							\
 	*(DATA_MAIN)							\
+	SCHED_DATA							\
 	*(.ref.data)							\
 	*(.data..shared_aligned) /* percpu related */			\
 	MEM_KEEP(init.data*)						\
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index 43323f875cb9..5abdbe569f93 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -2428,7 +2428,8 @@ static void prio_changed_dl(struct rq *rq, struct task_struct *p,
 	}
 }
 
-const struct sched_class dl_sched_class = {
+const struct sched_class dl_sched_class
+	__attribute__((section("__dl_sched_class"))) = {
 	.next			= &rt_sched_class,
 	.enqueue_task		= enqueue_task_dl,
 	.dequeue_task		= dequeue_task_dl,
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 08a233e97a01..e745fe0e0cd3 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -10745,7 +10745,8 @@ static unsigned int get_rr_interval_fair(struct rq *rq, struct task_struct *task
 /*
  * All the scheduling class methods:
  */
-const struct sched_class fair_sched_class = {
+const struct sched_class fair_sched_class
+	__attribute__((section("__fair_sched_class"))) = {
 	.next			= &idle_sched_class,
 	.enqueue_task		= enqueue_task_fair,
 	.dequeue_task		= dequeue_task_fair,
diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c
index ffa959e91227..700a9c826f0e 100644
--- a/kernel/sched/idle.c
+++ b/kernel/sched/idle.c
@@ -454,7 +454,8 @@ static void update_curr_idle(struct rq *rq)
 /*
  * Simple, special scheduling class for the per-CPU idle tasks:
  */
-const struct sched_class idle_sched_class = {
+const struct sched_class idle_sched_class
+	__attribute__((section("__idle_sched_class"))) = {
 	/* .next is NULL */
 	/* no enqueue/yield_task for idle tasks */
 
diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index e591d40fd645..5d3f9bcddaeb 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -2354,7 +2354,8 @@ static unsigned int get_rr_interval_rt(struct rq *rq, struct task_struct *task)
 		return 0;
 }
 
-const struct sched_class rt_sched_class = {
+const struct sched_class rt_sched_class
+	__attribute__((section("__rt_sched_class"))) = {
 	.next			= &fair_sched_class,
 	.enqueue_task		= enqueue_task_rt,
 	.dequeue_task		= dequeue_task_rt,
diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c
index 4c9e9975684f..03bc7530ff75 100644
--- a/kernel/sched/stop_task.c
+++ b/kernel/sched/stop_task.c
@@ -115,7 +115,8 @@ static void update_curr_stop(struct rq *rq)
 /*
  * Simple, special scheduling class for the per-CPU stop tasks:
  */
-const struct sched_class stop_sched_class = {
+const struct sched_class stop_sched_class
+	__attribute__((section("__stop_sched_class"))) = {
 	.next			= &dl_sched_class,
 
 	.enqueue_task		= enqueue_task_stop,
-- 
2.24.0