summaryrefslogtreecommitdiffstats
path: root/arch/Kconfig
diff options
context:
space:
mode:
authorFrederic Weisbecker <frederic@kernel.org>2020-01-27 16:41:52 +0100
committerFrederic Weisbecker <frederic@kernel.org>2020-02-14 16:05:04 +0100
commit490f561b783dac2c4825e288e6dbbf83481eea34 (patch)
tree9d51328accd11ec6eafd8e1cc0e0a150c451b764 /arch/Kconfig
parent7c805795307b40af50a45b7db44dd09ac1700947 (diff)
downloadlinux-490f561b783dac2c4825e288e6dbbf83481eea34.tar.bz2
context-tracking: Introduce CONFIG_HAVE_TIF_NOHZ
A few archs (x86, arm, arm64) don't rely anymore on TIF_NOHZ to call into context tracking on user entry/exit but instead use static keys (or not) to optimize those calls. Ideally every arch should migrate to that behaviour in the long run. Settle a config option to let those archs remove their TIF_NOHZ definitions. Signed-off-by: Frederic Weisbecker <frederic@kernel.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Andy Lutomirski <luto@kernel.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Will Deacon <will@kernel.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Paul Burton <paulburton@kernel.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'arch/Kconfig')
-rw-r--r--arch/Kconfig16
1 files changed, 11 insertions, 5 deletions
diff --git a/arch/Kconfig b/arch/Kconfig
index 98de654b79b3..dbf420a9f87b 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -540,11 +540,17 @@ config HAVE_CONTEXT_TRACKING
help
Provide kernel/user boundaries probes necessary for subsystems
that need it, such as userspace RCU extended quiescent state.
- Syscalls need to be wrapped inside user_exit()-user_enter() through
- the slow path using TIF_NOHZ flag. Exceptions handlers must be
- wrapped as well. Irqs are already protected inside
- rcu_irq_enter/rcu_irq_exit() but preemption or signal handling on
- irq exit still need to be protected.
+ Syscalls need to be wrapped inside user_exit()-user_enter(), either
+ optimized behind static key or through the slow path using TIF_NOHZ
+ flag. Exceptions handlers must be wrapped as well. Irqs are already
+ protected inside rcu_irq_enter/rcu_irq_exit() but preemption or signal
+ handling on irq exit still need to be protected.
+
+config HAVE_TIF_NOHZ
+ bool
+ help
+ Arch relies on TIF_NOHZ and syscall slow path to implement context
+ tracking calls to user_enter()/user_exit().
config HAVE_VIRT_CPU_ACCOUNTING
bool