Enable optprobe on preepmtive kernel
Masami Hiramatsu
maintainerPATCH v4 1/1 Enable optprobe on preepmtive kernel
This patch enables optprobe even with CONFIG_PREEMPT. The previous patch is https://lkml.org/lkml/2017/9/19/159 .
Sorry for taking so long to update.
I've update the patch description according to Ingo's comment but the code is same.
Thank you,
---
Masami Hiramatsu (1):
kprobes: Use synchronize_rcu_tasks() for optprobe with CONFIG_PREEMPT
arch/Kconfig | 2 +- kernel/kprobes.c | 14 ++++++++------ 2 files changed, 9 insertions(+), 7 deletions(-)
--
Masami Hiramatsu <[email protected]>
Masami Hiramatsu
maintainerPATCH v4 1/1 kprobes: Use synchronize_rcu_tasks() for optprobe with CONFIG_PREEMPT
We want to wait for all potentially preempted kprobes trampoline execution to have completed. This guarantees that any freed trampoline memory is not in use by any task in the system anymore. synchronize_rcu_tasks() gives such a guarantee, so use it. Also, this guarantees to wait for all potentially preempted tasks on the instructions which will be replaced with a jump.
Since this becomes a problem only when CONFIG_PREEMPT=y, enable CONFIG_TASKS_RCU=y for synchronize_rcu_tasks() in that case.
--- arch/Kconfig | 2 +- kernel/kprobes.c | 14 ++++++++------ 2 files changed, 9 insertions(+), 7 deletions(-)
arch/Kconfig +1 -1
@@ -90,7 +90,7 @@ config STATIC_KEYS_SELFTEST config OPTPROBES def_bool y depends on KPROBES && HAVE_OPTPROBES - depends on !PREEMPT + select TASKS_RCU if PREEMPT config KPROBES_ON_FTRACE def_bool y
kernel/kprobes.c +8 -6
@@ -573,13 +573,15 @@ static void kprobe_optimizer(struct work_struct *work)do_unoptimize_kprobes ();/* -* Step 2: Wait for quiesence period to ensure all running interrupts -* are done. Because optprobe may modify multiple instructions -* there is a chance that Nth instruction is interrupted. In that -* case, running interrupt can return to 2nd-Nth byte of jump -* instruction. This wait is for avoiding it. +* Step 2: Wait for quiesence period to ensure all potentially +* preempted tasks to have normally scheduled. Because optprobe +* may modify multiple instructions, there is a chance that Nth +* instruction is preempted. In that case, such tasks can return +* to 2nd-Nth byte of jump instruction. This wait is for avoiding it. +* Note that on non-preemptive kernel, this is transparently converted +* to synchronoze_sched() to wait for all interrupts to have completed. */ -synchronize_sched (); +synchronize_rcu_tasks ();/* Step 3: Optimize kprobes after quiesence period */ do_optimize_kprobes ();
Re: [PATCH -tip v4] kprobes: Use synchronize_rcu_tasks() for optprobe with CONFIG_PREEMPT
Quoted 11 lines · Masami Hiramatsu
We want to wait for all potentially preempted kprobes trampoline execution to have completed. This guarantees that any freed trampoline memory is not in use by any task in the system anymore.
synchronize_rcu_tasks()gives such a guarantee, so use it. Also, this guarantees to wait for all potentially preempted tasks on the instructions which will be replaced with a jump.Since this becomes a problem only when
CONFIG_PREEMPT=y, enableCONFIG_TASKS_RCU=yforsynchronize_rcu_tasks()in that case.Signed-off-by Masami Hiramatsu <[email protected]>
;-)
Quoted 45 lines
--- arch/Kconfig | 2 +- kernel/kprobes.c | 14 ++++++++------ 2 files changed, 9 insertions(+), 7 deletions(-)arch/Kconfig +1
-1@@ -90,7 +90,7 @@ config STATIC_KEYS_SELFTEST config OPTPROBES def_bool y depends on KPROBES && HAVE_OPTPROBES - depends on !PREEMPT + select TASKS_RCU if PREEMPT config KPROBES_ON_FTRACE def_bool ykernel/kprobes.c +8
-6@@ -573,13 +573,15 @@ static void kprobe_optimizer(struct work_struct *work)do_unoptimize_kprobes ();/* -* Step 2: Wait for quiesence period to ensure all running interrupts -* are done. Because optprobe may modify multiple instructions -* there is a chance that Nth instruction is interrupted. In that -* case, running interrupt can return to 2nd-Nth byte of jump -* instruction. This wait is for avoiding it. +* Step 2: Wait for quiesence period to ensure all potentially +* preempted tasks to have normally scheduled. Because optprobe +* may modify multiple instructions, there is a chance that Nth +* instruction is preempted. In that case, such tasks can return +* to 2nd-Nth byte of jump instruction. This wait is for avoiding it. +* Note that on non-preemptive kernel, this is transparently converted +* to synchronoze_sched() to wait for all interrupts to have completed. */ -synchronize_sched (); +synchronize_rcu_tasks ();/* Step 3: Optimize kprobes after quiesence period */ do_optimize_kprobes ();
tip-bot for Masami Hiramatsu
merge bot[tip:perf/core] kprobes: Use synchronize_rcu_tasks() for optprobe with CONFIG_PREEMPT=y
In reply to Masami Hiramatsukprobes: Use synchronize_rcu_tasks() for optprobe with CONFIG_PREEMPT=y
Commit-ID: a30b85df7d599f626973e9cd3056fe755bd778e0 Gitweb: https://git.kernel.org/tip/a30b85df7d599f626973e9cd3056fe755bd778e0 Author: Masami Hiramatsu <[email protected]> AuthorDate: Fri, 20 Oct 2017 08:43:39 +0900 Committer: Ingo Molnar <[email protected]> CommitDate: Fri, 20 Oct 2017 09:45:15 +0200
kprobes: Use synchronize_rcu_tasks() for optprobe with CONFIG_PREEMPT=y
We want to wait for all potentially preempted kprobes trampoline execution to have completed. This guarantees that any freed trampoline memory is not in use by any task in the system anymore. synchronize_rcu_tasks() gives such a guarantee, so use it.
Also, this guarantees to wait for all potentially preempted tasks on the instructions which will be replaced with a jump.
Since this becomes a problem only when CONFIG_PREEMPT=y, enable CONFIG_TASKS_RCU=y for synchronize_rcu_tasks() in that case.
--- arch/Kconfig | 2 +- kernel/kprobes.c | 14 ++++++++------ 2 files changed, 9 insertions(+), 7 deletions(-)
arch/Kconfig +1 -1
@@ -90,7 +90,7 @@ config STATIC_KEYS_SELFTEST config OPTPROBES def_bool y depends on KPROBES && HAVE_OPTPROBES - depends on !PREEMPT + select TASKS_RCU if PREEMPT config KPROBES_ON_FTRACE def_bool y
kernel/kprobes.c +8 -6
@@ -573,13 +573,15 @@ static void kprobe_optimizer(struct work_struct *work)do_unoptimize_kprobes ();/* -* Step 2: Wait for quiesence period to ensure all running interrupts -* are done. Because optprobe may modify multiple instructions -* there is a chance that Nth instruction is interrupted. In that -* case, running interrupt can return to 2nd-Nth byte of jump -* instruction. This wait is for avoiding it. +* Step 2: Wait for quiesence period to ensure all potentially +* preempted tasks to have normally scheduled. Because optprobe +* may modify multiple instructions, there is a chance that Nth +* instruction is preempted. In that case, such tasks can return +* to 2nd-Nth byte of jump instruction. This wait is for avoiding it. +* Note that on non-preemptive kernel, this is transparently converted +* to synchronoze_sched() to wait for all interrupts to have completed. */ -synchronize_sched (); +synchronize_rcu_tasks ();/* Step 3: Optimize kprobes after quiesence period */ do_optimize_kprobes ();