Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Thomas Gleixner <tglx@linutronix.de>
To:     Yujie Liu <yujie.liu@intel.com>
Cc:     Shanker Donthineni <sdonthineni@nvidia.com>,
        oe-lkp@lists.linux.dev, lkp@intel.com,
        linux-kernel@vger.kernel.org, Marc Zyngier <maz@kernel.org>,
        Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
        Michael Walle <michael@walle.cc>,
        Vikram Sethi <vsethi@nvidia.com>
Subject: Re: [PATCH v3 3/3] genirq: Use the maple tree for IRQ descriptors
 management
In-Reply-To: <ZEsiYbi8dorXTI5t@yujie-X299>
References: <202304251035.19367560-yujie.liu@intel.com>
 <87a5yuzvzd.ffs@tglx> <ZEsiYbi8dorXTI5t@yujie-X299>
Date:   Fri, 28 Apr 2023 12:31:14 +0200
Message-ID: <877ctw5mdp.ffs@tglx>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

Yujie!

On Fri, Apr 28 2023 at 09:33, Yujie Liu wrote:
> On Wed, Apr 26, 2023 at 02:08:54PM +0200, Thomas Gleixner wrote:
>> Can you please retest this on v6.3 and report back when the problem
>> persists?
>
> Thanks for your help looking into this problem.
>
> The problem persists when tested on v6.3, but not 100% reproducible.
> We ran the test on v6.3 and v6.3+patch each for 20 runs. There are 9
> failed runs on v6.3+patch, while v6.3 is all clean. Full dmesg is
> attached.

Under the assumption that the code is correct, then the effect of this
patch is that it changes the timing. Sigh.

  1) Does this happen with a 64-bit kernel too?

  2) Can you enable the irq_vector:vector_*.* tracepoints and provide
     the trace?

And please provide /proc/interrupts from that machine.

Thanks,

        tglx