Linux-PM Archive on lore.kernel.org
 help / color / Atom feed
From: Qian Cai <cai@lca.pw>
To: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: Linux PM <linux-pm@vger.kernel.org>,
	Len Brown <len.brown@intel.com>,
	LKML <linux-kernel@vger.kernel.org>,
	Zhang Rui <rui.zhang@intel.com>,
	"Rafael J. Wysocki" <rafael@kernel.org>,
	Chen Yu <yu.c.chen@intel.com>,
	clang-built-linux@googlegroups.com
Subject: Re: [PATCH 3/9] intel_idle: Relocate definitions of cpuidle callbacks
Date: Fri, 22 May 2020 20:04:27 -0400
Message-ID: <20200523000427.GF1337@Qians-MacBook-Air.local> (raw)
In-Reply-To: <2912140.PDVJEUYNKe@kreacher>

On Thu, Feb 13, 2020 at 11:00:26PM +0100, Rafael J. Wysocki wrote:
> From: "Rafael J. Wysocki" <rafael.j.wysocki@intel.com>
> 
> Move the definitions of intel_idle() and intel_idle_s2idle() before
> the definitions of cpuidle_state structures referring to them to
> avoid having to use additional declarations of them (and drop those
> declarations).
> 
> No functional impact.
> 
> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> ---
>  drivers/idle/intel_idle.c | 154 ++++++++++++++++++++++------------------------
>  1 file changed, 75 insertions(+), 79 deletions(-)
> 
> diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
> index 5adc058c705d..e0332d567735 100644
> --- a/drivers/idle/intel_idle.c
> +++ b/drivers/idle/intel_idle.c
[]
> +/**
> + * intel_idle - Ask the processor to enter the given idle state.
> + * @dev: cpuidle device of the target CPU.
> + * @drv: cpuidle driver (assumed to point to intel_idle_driver).
> + * @index: Target idle state index.
> + *
> + * Use the MWAIT instruction to notify the processor that the CPU represented by
> + * @dev is idle and it can try to enter the idle state corresponding to @index.
> + *
> + * If the local APIC timer is not known to be reliable in the target idle state,
> + * enable one-shot tick broadcasting for the target CPU before executing MWAIT.
> + *
> + * Optionally call leave_mm() for the target CPU upfront to avoid wakeups due to
> + * flushing user TLBs.
> + *
> + * Must be called under local_irq_disable().
> + */
> +static __cpuidle int intel_idle(struct cpuidle_device *dev,
> +				struct cpuidle_driver *drv, int index)
> +{
> +	struct cpuidle_state *state = &drv->states[index];
> +	unsigned long eax = flg2MWAIT(state->flags);
> +	unsigned long ecx = 1; /* break on interrupt flag */
> +	bool uninitialized_var(tick);

This will generate an UBSAN warning because Clang could poison all
uninitialized stack variables to 0xAA due to CONFIG_INIT_STACK_ALL=y, so one
issue is that,

bool uninitialized_var(x);

would always broken on Clang like this,

[   92.140611] UBSAN: invalid-load in drivers/idle/intel_idle.c:135:7
[   92.143111] load of value 170 is not a valid value for type 'bool' (aka '_Bool')
[   92.145657] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.7.0-rc6-next-20200522+ #3
[   92.147424] Hardware name: HP ProLiant BL660c Gen9, BIOS I38 10/17/2018
[   92.149869] Call Trace:
[   92.149869]  dump_stack+0x10b/0x17f
[   92.149869]  __ubsan_handle_load_invalid_value+0xd2/0x110
[   92.149869]  intel_idle+0x54/0xf0
[   92.156202]  cpuidle_enter_state+0x120/0x4f0
[   92.156202]  cpuidle_enter+0x5b/0xa0
[   92.156202]  call_cpuidle+0x25/0x50
[   92.156202]  do_idle+0x1eb/0x2c0
[   92.156202]  cpu_startup_entry+0x25/0x30
[   92.156202]  rest_init+0x26f/0x280
[   92.156202]  arch_call_rest_init+0x17/0x1e
[   92.156202]  start_kernel+0x598/0x633
[   92.156202]  x86_64_start_reservations+0x24/0x26
[   92.156202]  x86_64_start_kernel+0x116/0x1c1
[   92.156202]  secondary_startup_64+0xb6/0xc0

However, I am wondering if it is correct to let "tick" uninitialized to begin
with. If this condition is true,

!static_cpu_has(X86_FEATURE_ARAT) && lapic_timer_always_reliable

Then, we could in the final branch to use the uninitialized value.

if (!static_cpu_has(X86_FEATURE_ARAT) && tick)

Isn't that possible?

> +	int cpu = smp_processor_id();
> +
> +	/*
> +	 * leave_mm() to avoid costly and often unnecessary wakeups
> +	 * for flushing the user TLB's associated with the active mm.
> +	 */
> +	if (state->flags & CPUIDLE_FLAG_TLB_FLUSHED)
> +		leave_mm(cpu);
> +
> +	if (!static_cpu_has(X86_FEATURE_ARAT) && !lapic_timer_always_reliable) {
> +		/*
> +		 * Switch over to one-shot tick broadcast if the target C-state
> +		 * is deeper than C1.
> +		 */
> +		if ((eax >> MWAIT_SUBSTATE_SIZE) & MWAIT_CSTATE_MASK) {
> +			tick = true;
> +			tick_broadcast_enter();
> +		} else {
> +			tick = false;
> +		}
> +	}
> +
> +	mwait_idle_with_hints(eax, ecx);
> +
> +	if (!static_cpu_has(X86_FEATURE_ARAT) && tick)
> +		tick_broadcast_exit();
> +
> +	return index;
> +}

  reply index

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-13 21:58 [PATCH 0/9] intel_idle: More assorted cleanups Rafael J. Wysocki
2020-02-13 21:59 ` [PATCH 1/9] intel_idle: Simplify LAPIC timer reliability checks Rafael J. Wysocki
2020-02-13 21:59 ` [PATCH 2/9] intel_idle: Clean up definitions of cpuidle callbacks Rafael J. Wysocki
2020-02-13 22:00 ` [PATCH 3/9] intel_idle: Relocate " Rafael J. Wysocki
2020-05-23  0:04   ` Qian Cai [this message]
2020-02-13 22:01 ` [PATCH 4/9] intel_idle: Add __initdata annotations to init time variables Rafael J. Wysocki
2020-02-13 22:01 ` [PATCH 5/9] intel_idle: Annotate init time data structures Rafael J. Wysocki
2020-02-13 22:02 ` [PATCH 6/9] intel_idle: Reorder declarations of static variables Rafael J. Wysocki
2020-02-13 22:03 ` [PATCH 7/9] intel_idle: Clean up kerneldoc comments of multiple functions Rafael J. Wysocki
2020-02-13 22:03 ` [PATCH 8/9] intel_idle: Define CPUIDLE_FLAG_TLB_FLUSHED as BIT(16) Rafael J. Wysocki
2020-02-13 22:04 ` [PATCH 9/9] intel_idle: Update copyright notice, known limitations and version Rafael J. Wysocki

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200523000427.GF1337@Qians-MacBook-Air.local \
    --to=cai@lca.pw \
    --cc=clang-built-linux@googlegroups.com \
    --cc=len.brown@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=rafael@kernel.org \
    --cc=rjw@rjwysocki.net \
    --cc=rui.zhang@intel.com \
    --cc=yu.c.chen@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

Linux-PM Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-pm/0 linux-pm/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-pm linux-pm/ https://lore.kernel.org/linux-pm \
		linux-pm@vger.kernel.org
	public-inbox-index linux-pm

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-pm


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git