Date:   Mon, 20 Mar 2023 10:22:36 +0100
From:   Peter Zijlstra <peterz@infradead.org>
To:     Alexey Klimov <alexey.klimov@linaro.org>
Cc:     draszik@google.com, peter.griffin@linaro.org,
        willmcvicker@google.com, mingo@kernel.org, ulf.hansson@linaro.org,
        tony@atomide.com, linux-block@vger.kernel.org,
        linux-kernel@vger.kernel.org, axboe@kernel.dk,
        alim.akhtar@samsung.com, regressions@lists.linux.dev,
        avri.altman@wdc.com, bvanassche@acm.org, klimova@google.com
Subject: Re: [REGRESSION] CPUIDLE_FLAG_RCU_IDLE, blk_mq_freeze_queue_wait()
 and slow-stuck reboots
Message-ID: <20230320092236.GA2196776@hirez.programming.kicks-ass.net>
References: <20230314230004.961993-1-alexey.klimov@linaro.org>
 <20230315111606.GB2006103@hirez.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20230315111606.GB2006103@hirez.programming.kicks-ass.net>
Precedence: bulk

On Wed, Mar 15, 2023 at 12:16:06PM +0100, Peter Zijlstra wrote:

> So ARM has a weird 'rule' in that idle state 0 (wfi) should not have
> RCU_IDLE set, while others should have.
> 
> Of the dt_init_idle_driver() users:
> 
>  - cpuidle-arm: arm_enter_idle_state()
>  - cpuidle-big_little: bl_enter_powerdown() does ct_cpuidle_{enter,exit}()
>  - cpuidle-psci: psci_enter_idle_state() uses CPU_PM_CPU_IDLE_ENTER_PARAM_RCU()
>  - cpuidle-qcom-spm: spm_enter_idle_state() uses CPU_PM_CPU_IDLE_ENTER_PARAM()
>  - cpuidle-riscv-sbi: sbi_cpuidle_enter_state() uses CPU_PM_CPU_IDLE_ENTER_*_PARAM()
> 
> All of them start on index 1 and hence should have RCU_IDLE set, but at
> least the arm, qcom-spm and riscv-sbi don't actually appear to abide by
> the rules.
> 
> Fixing that gives me the below; does that help?

FWIW..  I got the meaning of CPU_PM_CPU_IDLE_*'s _RCU thing inverted
last week; they're actually ok as is.

Specifically _RCU wants the @low_level_idle_enter to do
ct_cpuidle_{enter,exit}(). And since the ones I audited didn't in fact
do that, they should not be using the _RCU version.