[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250918150959.GEaMwgx78CGCxjGce8@fat_crate.local>
Date: Thu, 18 Sep 2025 17:09:59 +0200
From: Borislav Petkov <bp@...en8.de>
To: Tianyu Lan <ltykernel@...il.com>
Cc: kys@...rosoft.com, haiyangz@...rosoft.com, wei.liu@...nel.org,
decui@...rosoft.com, tglx@...utronix.de, mingo@...hat.com,
dave.hansen@...ux.intel.com, x86@...nel.org, hpa@...or.com,
arnd@...db.de, Neeraj.Upadhyay@....com, tiala@...rosoft.com,
kvijayab@....com, romank@...ux.microsoft.com,
linux-arch@...r.kernel.org, linux-hyperv@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 5/5] x86/Hyper-V: Add Hyper-V specific hvcall to set
backing page
On Thu, Sep 18, 2025 at 11:00:23AM -0400, Tianyu Lan wrote:
> Secure AVIC hardware provides APIC backing page
> to aid the guest in limiting which interrupt
> vectors can be injected into the guest. Hyper-V
> introduces a new register HV_X64_REGISTER_SEV_GPA_PAGE
> to notify hypervisor with APIC backing page and call
> it in Secure AVIC driver.
Why does hyperv needs special handling again and cannot simply adhere to the
secure AVIC spec?
None of that text explains *why* it is absolutely necessary to do something
hyperv-special...
> @@ -361,7 +364,11 @@ static void savic_setup(void)
> * VMRUN, the hypervisor makes use of this information to make sure
> * the APIC backing page is mapped in NPT.
> */
> - res = savic_register_gpa(gpa);
> + if (hv_isolation_type_snp())
> + res = hv_set_savic_backing_page(gfn);
> + else
> + res = savic_register_gpa(gpa);
> +
This is ugly and doesn't belong here.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
Powered by blists - more mailing lists