[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <f898e36f-ba02-4c52-a3be-06caac13323e@linux.ibm.com>
Date: Wed, 24 Jan 2024 13:14:40 +0100
From: Christian Borntraeger <borntraeger@...ux.ibm.com>
To: Yi Wang <up2wing@...il.com>, seanjc@...gle.com, pbonzini@...hat.com,
tglx@...utronix.de, mingo@...hat.com, bp@...en8.de,
dave.hansen@...ux.intel.com, x86@...nel.org, hpa@...or.com,
kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
wanpengli@...cent.com, foxywang@...cent.com, oliver.upton@...ux.dev,
maz@...nel.org, anup@...infault.org, atishp@...shpatra.org,
frankja@...ux.ibm.com, imbrenda@...ux.ibm.com
Subject: Re: [v2 0/4] KVM: irqchip: synchronize srcu only if needed
Am 21.01.24 um 12:17 schrieb Yi Wang:
> From: Yi Wang <foxywang@...cent.com>
>
> We found that it may cost more than 20 milliseconds very accidentally
> to enable cap of KVM_CAP_SPLIT_IRQCHIP on a host which has many vms
> already.
>
> The reason is that when vmm(qemu/CloudHypervisor) invokes
> KVM_CAP_SPLIT_IRQCHIP kvm will call synchronize_srcu_expedited() and
> might_sleep and kworker of srcu may cost some delay during this period.
> One way makes sence is setup empty irq routing when creating vm and
> so that x86/s390 don't need to setup empty/dummy irq routing.
>
> Note: I have no s390 machine so the s390 patch has not been tested.
I just did a quick sniff and it still seems to work. No performance check etc.
Powered by blists - more mailing lists