qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH 1/5] pnv/xive2: XIVE2 Cache Watch, Cache Flush and Sync Injec


From: Cédric Le Goater
Subject: Re: [PATCH 1/5] pnv/xive2: XIVE2 Cache Watch, Cache Flush and Sync Injection support
Date: Fri, 12 Jul 2024 08:17:28 +0200
User-agent: Mozilla Thunderbird

Hello Mike,

On 7/11/24 22:57, Mike Kowal wrote:
Hello Cedric,

This patch will be split into 5 patches to ease the review.  I also acknowledge 
most of your other comments.

Can you elaborate on the gic_get_current_cpu() comment? Did you want me to use that method name to get a the current CPU?

I meant introducing a similar routine for XIVE to hide the qtest
environment differences from the model. Something like :

static inline int pnv_xive2_get_current_pir(PnvXive2 *xive)
{
    if (!qtest_enabled()) {
        return ppc_cpu_pir(POWERPC_CPU(current_cpu));
    }
    return 0;
}

Also, is this something that you think needs to done at this current time since 
there are many other existing uses to get the CPU,

The other uses of current_cpu are in the TIMA memops. They should be
immune to this problem since the ops should always be done from a CPU.

or can we delay this and an issue for for this to be done at a later time?

It is minor. That said, it is not a big change compared to the split.

Thanks,

C.





reply via email to

[Prev in Thread] Current Thread [Next in Thread]