Re: [PATCHv6 10/16] x86/tdx: Convert shared memory back to private on kexec

From: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
To: "Kalra, Ashish" <ashish.kalra@amd.com>
Cc: Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>,  Borislav Petkov <bp@alien8.de>,
	Dave Hansen <dave.hansen@linux.intel.com>,
	x86@kernel.org,  "Rafael J. Wysocki" <rafael@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	 Adrian Hunter <adrian.hunter@intel.com>,
	 Kuppuswamy Sathyanarayanan
	<sathyanarayanan.kuppuswamy@linux.intel.com>,
	Elena Reshetova <elena.reshetova@intel.com>,
	 Jun Nakajima <jun.nakajima@intel.com>,
	Rick Edgecombe <rick.p.edgecombe@intel.com>,
	 Tom Lendacky <thomas.lendacky@amd.com>,
	Sean Christopherson <seanjc@google.com>,
	 "Huang, Kai" <kai.huang@intel.com>, Baoquan He <bhe@redhat.com>,
	kexec@lists.infradead.org,  linux-coco@lists.linux.dev,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCHv6 10/16] x86/tdx: Convert shared memory back to private on kexec
Date: Mon, 29 Jan 2024 15:34:26 +0200	[thread overview]
Message-ID: <ao36mdmctptya2r25v3lvuocjbhllum2upmskxjcjtt3c7o3ii@nsghte2cnmkx> (raw)
In-Reply-To: <fad515ec-aba3-4f48-b8be-ebc6640899e3@amd.com>

On Mon, Jan 29, 2024 at 07:09:37AM -0600, Kalra, Ashish wrote:
> Hello Kirill,
> 
> On 1/29/2024 4:36 AM, Kirill A. Shutemov wrote:
> > On Mon, Jan 29, 2024 at 04:24:09AM -0600, Kalra, Ashish wrote:
> > > In case of SNP and crash/kdump case, we need to prevent the boot_ghcb being
> > > converted to shared (in snp_kexec_unshare_mem()) as the boot_ghcb is
> > > required to handle all I/O for disabling IO-APIC/lapic, hpet, etc., as the
> > > enc_kexec_unshare_mem() callback is invoked before the apics, hpet, etc. are
> > > disabled.
> > > 
> > > Is there any reason why enc_kexec_unshare_mem() callback is invoked in crash
> > > case before the IO-APIC/lapic, hpet, etc. are shutdown/disabled ?
> > Not really. Either way works for TDX. I've tested the patch below. Is it
> > what you want?
> > 
> > diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
> > index 6585a5f2c2ba..3001f4857ed7 100644
> > --- a/arch/x86/kernel/crash.c
> > +++ b/arch/x86/kernel/crash.c
> > @@ -107,11 +107,6 @@ void native_machine_crash_shutdown(struct pt_regs *regs)
> >   	crash_smp_send_stop();
> > -	if (cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT)) {
> > -		x86_platform.guest.enc_kexec_stop_conversion(true);
> > -		x86_platform.guest.enc_kexec_unshare_mem();
> > -	}
> > -
> >   	cpu_emergency_disable_virtualization();
> >   	/*
> > @@ -129,6 +124,12 @@ void native_machine_crash_shutdown(struct pt_regs *regs)
> >   #ifdef CONFIG_HPET_TIMER
> >   	hpet_disable();
> >   #endif
> > +
> > +	if (cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT)) {
> > +		x86_platform.guest.enc_kexec_stop_conversion(true);
> > +		x86_platform.guest.enc_kexec_unshare_mem();
> > +	}
> > +
> >   	crash_save_cpu(regs, safe_smp_processor_id());
> >   }
> 
> Yes, this will work for SNP.

Okay, good. Will include the change into the next version of the patchset.

-- 
  Kiryl Shutsemau / Kirill A. Shutemov