Re: [PATCH] irqchip/gic-v3-its: Don't acquire rt_spin_lock in allocate_vpe_l1_table()

[Waiman Long <llong@redhat.com>]

On 1/21/26 3:38 AM, Marc Zyngier wrote:
> On Mon, 12 Jan 2026 11:20:07 +0000,
> Marc Zyngier <maz@kernel.org> wrote:
>> On Sun, 11 Jan 2026 16:20:45 +0000,
>> Thomas Gleixner <tglx@kernel.org> wrote:
>>> On Sun, Jan 11 2026 at 10:38, Marc Zyngier wrote:
>>>> On Sun, 11 Jan 2026 09:39:07 +0000,
>>>> Thomas Gleixner <tglx@kernel.org> wrote:
>>>>> On Fri, Jan 09 2026 at 16:13, Marc Zyngier wrote:
>>>>>> On Thu, 08 Jan 2026 22:11:33 +0000,
>>>>>> Thomas Gleixner <tglx@kernel.org> wrote:
>>>>>>> At the point where a CPU is brought up, the topology should be known
>>>>>>> already, which means this can be allocated on the control CPU _before_
>>>>>>> the new CPU comes up, no?
>>>>>> No. Each CPU finds *itself* in the forest of redistributors, and from
>>>>>> there tries to find whether it has some shared resource with a CPU
>>>>>> that has booted before it. That's because firmware is absolutely awful
>>>>>> and can't present a consistent view of the system.
>>>>> Groan....
>>>>>
>>>>>> Anyway, I expect it could be solved by moving this part of the init to
>>>>>> an ONLINE HP callback.
>>>>> Which needs to be before CPUHP_AP_IRQ_AFFINITY_ONLINE, but even that
>>>>> might be to late because there are callbacks in the STARTING section,
>>>>> i.e. timer, perf, which might rely on interrupts being accessible.
>>>> Nah. This stuff is only for direct injection of vLPIs into guests, so
>>>> as long as this is done before we can schedule a vcpu on this physical
>>>> CPU, we're good. No physical interrupt is concerned with this code.
>>> That's fine then. vCPUs are considered "user-space" tasks and can't be
>>> scheduled before CPUHP_AP_ACTIVE sets the CPU active for the scheduler.
>> Waiman, can you please give the following hack a go on your box? The
>> machines I have are thankfully limited to a single ITS group, so I
>> can't directly reproduce your issue.
> Have you managed to try this hack? I may be able to spend some time
> addressing the issue in the next cycle if I have an indication that
> I'm on the right track.

Yes, I have tried out your hack patch and the 2-socket Grace test system 
booted up without producing any bug report for a RT debug kernel. I will 
try out your official patch once it come out. So moving the memory 
allocation to a later part of the hotplug bringup pipeline where 
sleeping is allowed should work.

Cheers,
Longman