Commit bf5835bcdb96 ("intel_idle: Disable IBRS during long idle")
disables IBRS when the cstate is 6 or lower. However, there are
some use cases where a customer may want to use max_cstate=1 to
lower latency. Such use cases will suffer from the performance
degradation caused by the enabling of IBRS in the sibling idle thread.
Add a "ibrs_off" module parameter to force disable IBRS and the
CPUIDLE_FLAG_IRQ_ENABLE flag if set.
In the case of a Skylake server with max_cstate=1, this new ibrs_off
option will likely increase the IRQ response latency as IRQ will now
be disabled.
When running SPECjbb2015 with cstates set to C1 on a Skylake system.
First test when the kernel is booted with: "intel_idle.ibrs_off"
max-jOPS = 117828, critical-jOPS = 66047
Then retest when the kernel is booted without the "intel_idle.ibrs_off"
added.
max-jOPS = 116408, critical-jOPS = 58958
That means booting with "intel_idle.ibrs_off" improves performance by:
max-jOPS: 1.2%, which could be considered noise range.
critical-jOPS: 12%, which is definitely a solid improvement.
The admin-guide/pm/intel_idle.rst file is updated to add a description
about the new "ibrs_off" module parameter.
Signed-off-by: Waiman Long <[email protected]>
---
Documentation/admin-guide/pm/intel_idle.rst | 17 ++++++++++++++++-
drivers/idle/intel_idle.c | 14 ++++++++++++--
2 files changed, 28 insertions(+), 3 deletions(-)
diff --git a/Documentation/admin-guide/pm/intel_idle.rst b/Documentation/admin-guide/pm/intel_idle.rst
index b799a43da62e..8604e6d1fe2c 100644
--- a/Documentation/admin-guide/pm/intel_idle.rst
+++ b/Documentation/admin-guide/pm/intel_idle.rst
@@ -170,7 +170,7 @@ and ``idle=nomwait``. If any of them is present in the kernel command line, the
``MWAIT`` instruction is not allowed to be used, so the initialization of
``intel_idle`` will fail.
-Apart from that there are four module parameters recognized by ``intel_idle``
+Apart from that there are five module parameters recognized by ``intel_idle``
itself that can be set via the kernel command line (they cannot be updated via
sysfs, so that is the only way to change their values).
@@ -216,6 +216,21 @@ are ignored).
The idle states disabled this way can be enabled (on a per-CPU basis) from user
space via ``sysfs``.
+The ``ibrs_off`` module parameter is a boolean flag (default to false). It is
+used to control if IBRS (Indirect Branch Restricted Speculation) should be
+turned off, if set, when the CPU enters an idle state. This flag will not
+affect CPUs that are using Enhanced IBRS which can remain on with little
+performance impact.
+
+For some CPUs, IBRS will be selected as mitigation for Spectre v2 and Retbleed
+security vulnerabilities by default. Leaving the IBRS mode on while idling may
+have a performance impact on its sibling CPU. The IBRS mode will be turned off
+by default when the CPU enters into a deep idle state, but not in some
+shallower ones. Setting the ``ibrs_off`` module parameter will force the IBRS
+mode to off when the CPU is in any one of the available idle states. This may
+help performance of a sibling CPU at the expense of a slightly higher wakeup
+latency for the idle CPU.
+
.. _intel-idle-core-and-package-idle-states:
diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index e32ff09051af..14874579b396 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -69,6 +69,7 @@ static int max_cstate = CPUIDLE_STATE_MAX - 1;
static unsigned int disabled_states_mask __read_mostly;
static unsigned int preferred_states_mask __read_mostly;
static bool force_irq_on __read_mostly;
+static bool ibrs_off __read_mostly;
static struct cpuidle_device __percpu *intel_idle_cpuidle_devices;
@@ -1919,12 +1920,15 @@ static void state_update_enter_method(struct cpuidle_state *state, int cstate)
}
if (cpu_feature_enabled(X86_FEATURE_KERNEL_IBRS) &&
- state->flags & CPUIDLE_FLAG_IBRS) {
+ ((state->flags & CPUIDLE_FLAG_IBRS) || ibrs_off)) {
/*
* IBRS mitigation requires that C-states are entered
* with interrupts disabled.
*/
- WARN_ON_ONCE(state->flags & CPUIDLE_FLAG_IRQ_ENABLE);
+ if (ibrs_off && (state->flags & CPUIDLE_FLAG_IRQ_ENABLE))
+ state->flags &= ~CPUIDLE_FLAG_IRQ_ENABLE;
+ else
+ WARN_ON_ONCE(state->flags & CPUIDLE_FLAG_IRQ_ENABLE);
state->enter = intel_idle_ibrs;
return;
}
@@ -2346,3 +2350,9 @@ MODULE_PARM_DESC(preferred_cstates, "Mask of preferred idle states");
* 'CPUIDLE_FLAG_INIT_XSTATE' and 'CPUIDLE_FLAG_IBRS' flags.
*/
module_param(force_irq_on, bool, 0444);
+/*
+ * Force the disabling of IBRS when X86_FEATURE_KERNEL_IBRS is on and
+ * CPUIDLE_FLAG_IRQ_ENABLE isn't set.
+ */
+module_param(ibrs_off, bool, 0444);
+MODULE_PARM_DESC(ibrs_off, "Disable IBRS when idle");
--
2.31.1
Hi,
On 6/27/23 19:25, Waiman Long wrote:
> Documentation/admin-guide/pm/intel_idle.rst | 17 ++++++++++++++++-
> drivers/idle/intel_idle.c | 14 ++++++++++++--
> 2 files changed, 28 insertions(+), 3 deletions(-)
>
> diff --git a/Documentation/admin-guide/pm/intel_idle.rst b/Documentation/admin-guide/pm/intel_idle.rst
> index b799a43da62e..8604e6d1fe2c 100644
> --- a/Documentation/admin-guide/pm/intel_idle.rst
> +++ b/Documentation/admin-guide/pm/intel_idle.rst
You may take or leave these comments as you choose.
> @@ -216,6 +216,21 @@ are ignored).
> The idle states disabled this way can be enabled (on a per-CPU basis) from user
> space via ``sysfs``.
>
> +The ``ibrs_off`` module parameter is a boolean flag (default to false). It is
(default false).
or
(defaults to false).
Then I think it reads better if the next sentence begins with: "If set,"
instead of having it in the middle of the sentence.
> +used to control if IBRS (Indirect Branch Restricted Speculation) should be
> +turned off, if set, when the CPU enters an idle state. This flag will not
> +affect CPUs that are using Enhanced IBRS which can remain on with little
> +performance impact.
> +
> +For some CPUs, IBRS will be selected as mitigation for Spectre v2 and Retbleed
> +security vulnerabilities by default. Leaving the IBRS mode on while idling may
> +have a performance impact on its sibling CPU. The IBRS mode will be turned off
> +by default when the CPU enters into a deep idle state, but not in some
> +shallower ones. Setting the ``ibrs_off`` module parameter will force the IBRS
> +mode to off when the CPU is in any one of the available idle states. This may
> +help performance of a sibling CPU at the expense of a slightly higher wakeup
> +latency for the idle CPU.
thanks.
--
~Randy
On 6/27/23 22:58, Randy Dunlap wrote:
> Hi,
>
> On 6/27/23 19:25, Waiman Long wrote:
>> Documentation/admin-guide/pm/intel_idle.rst | 17 ++++++++++++++++-
>> drivers/idle/intel_idle.c | 14 ++++++++++++--
>> 2 files changed, 28 insertions(+), 3 deletions(-)
>>
>> diff --git a/Documentation/admin-guide/pm/intel_idle.rst b/Documentation/admin-guide/pm/intel_idle.rst
>> index b799a43da62e..8604e6d1fe2c 100644
>> --- a/Documentation/admin-guide/pm/intel_idle.rst
>> +++ b/Documentation/admin-guide/pm/intel_idle.rst
> You may take or leave these comments as you choose.
>
>> @@ -216,6 +216,21 @@ are ignored).
>> The idle states disabled this way can be enabled (on a per-CPU basis) from user
>> space via ``sysfs``.
>>
>> +The ``ibrs_off`` module parameter is a boolean flag (default to false). It is
> (default false).
> or
> (defaults to false).
>
> Then I think it reads better if the next sentence begins with: "If set,"
> instead of having it in the middle of the sentence.
>
>> +used to control if IBRS (Indirect Branch Restricted Speculation) should be
>> +turned off, if set, when the CPU enters an idle state. This flag will not
>> +affect CPUs that are using Enhanced IBRS which can remain on with little
>> +performance impact.
>> +
>> +For some CPUs, IBRS will be selected as mitigation for Spectre v2 and Retbleed
>> +security vulnerabilities by default. Leaving the IBRS mode on while idling may
>> +have a performance impact on its sibling CPU. The IBRS mode will be turned off
>> +by default when the CPU enters into a deep idle state, but not in some
>> +shallower ones. Setting the ``ibrs_off`` module parameter will force the IBRS
>> +mode to off when the CPU is in any one of the available idle states. This may
>> +help performance of a sibling CPU at the expense of a slightly higher wakeup
>> +latency for the idle CPU.
> thanks.
Thanks for the suggestion. I will make the change if I have to revise
the patch again for any reason.
Cheers,
Longman
On Tue, Jun 27, 2023 at 10:25:54PM -0400, Waiman Long wrote:
> @@ -69,6 +69,7 @@ static int max_cstate = CPUIDLE_STATE_MAX - 1;
> static unsigned int disabled_states_mask __read_mostly;
> static unsigned int preferred_states_mask __read_mostly;
> static bool force_irq_on __read_mostly;
> +static bool ibrs_off __read_mostly;
>
> static struct cpuidle_device __percpu *intel_idle_cpuidle_devices;
>
> @@ -1919,12 +1920,15 @@ static void state_update_enter_method(struct cpuidle_state *state, int cstate)
> }
>
> if (cpu_feature_enabled(X86_FEATURE_KERNEL_IBRS) &&
> - state->flags & CPUIDLE_FLAG_IBRS) {
> + ((state->flags & CPUIDLE_FLAG_IBRS) || ibrs_off)) {
> /*
> * IBRS mitigation requires that C-states are entered
> * with interrupts disabled.
> */
> - WARN_ON_ONCE(state->flags & CPUIDLE_FLAG_IRQ_ENABLE);
> + if (ibrs_off && (state->flags & CPUIDLE_FLAG_IRQ_ENABLE))
> + state->flags &= ~CPUIDLE_FLAG_IRQ_ENABLE;
> + else
> + WARN_ON_ONCE(state->flags & CPUIDLE_FLAG_IRQ_ENABLE);
If you're respinning this, you can leave out the else and avoid the
indent on the WARN:
+ if (ibrs_off && (state->flags & CPUIDLE_FLAG_IRQ_ENABLE))
+ state->flags &= ~CPUIDLE_FLAG_IRQ_ENABLE;
WARN_ON_ONCE(state->flags & CPUIDLE_FLAG_IRQ_ENABLE);
Same effect, simpler code and all that.
> state->enter = intel_idle_ibrs;
> return;
> }
On 7/3/23 06:32, Peter Zijlstra wrote:
> On Tue, Jun 27, 2023 at 10:25:54PM -0400, Waiman Long wrote:
>
>> @@ -69,6 +69,7 @@ static int max_cstate = CPUIDLE_STATE_MAX - 1;
>> static unsigned int disabled_states_mask __read_mostly;
>> static unsigned int preferred_states_mask __read_mostly;
>> static bool force_irq_on __read_mostly;
>> +static bool ibrs_off __read_mostly;
>>
>> static struct cpuidle_device __percpu *intel_idle_cpuidle_devices;
>>
>> @@ -1919,12 +1920,15 @@ static void state_update_enter_method(struct cpuidle_state *state, int cstate)
>> }
>>
>> if (cpu_feature_enabled(X86_FEATURE_KERNEL_IBRS) &&
>> - state->flags & CPUIDLE_FLAG_IBRS) {
>> + ((state->flags & CPUIDLE_FLAG_IBRS) || ibrs_off)) {
>> /*
>> * IBRS mitigation requires that C-states are entered
>> * with interrupts disabled.
>> */
>> - WARN_ON_ONCE(state->flags & CPUIDLE_FLAG_IRQ_ENABLE);
>> + if (ibrs_off && (state->flags & CPUIDLE_FLAG_IRQ_ENABLE))
>> + state->flags &= ~CPUIDLE_FLAG_IRQ_ENABLE;
>> + else
>> + WARN_ON_ONCE(state->flags & CPUIDLE_FLAG_IRQ_ENABLE);
> If you're respinning this, you can leave out the else and avoid the
> indent on the WARN:
>
> + if (ibrs_off && (state->flags & CPUIDLE_FLAG_IRQ_ENABLE))
> + state->flags &= ~CPUIDLE_FLAG_IRQ_ENABLE;
> WARN_ON_ONCE(state->flags & CPUIDLE_FLAG_IRQ_ENABLE);
>
> Same effect, simpler code and all that.
That is true. I can certainly respin that as there is another suggested
doc change that is pending.
Cheers,
Longman