Message-ID: <effbb6e2-c5a1-af7f-830d-8d7088f57477@amd.com>
Date:   Tue, 7 Mar 2023 10:45:59 -0600
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.7.1
Subject: Re: [PATCH v13 00/11] Parallel CPU bringup for x86_64
Content-Language: en-US
To:     David Woodhouse <dwmw2@infradead.org>,
        Usama Arif <usama.arif@bytedance.com>, tglx@linutronix.de,
        kim.phillips@amd.com, brgerst@gmail.com
Cc:     piotrgorski@cachyos.org, oleksandr@natalenko.name,
        arjan@linux.intel.com, mingo@redhat.com, bp@alien8.de,
        dave.hansen@linux.intel.com, hpa@zytor.com, x86@kernel.org,
        pbonzini@redhat.com, paulmck@kernel.org,
        linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
        rcu@vger.kernel.org, mimoja@mimoja.de, hewenliang4@huawei.com,
        seanjc@google.com, pmenzel@molgen.mpg.de, fam.zheng@bytedance.com,
        punit.agrawal@bytedance.com, simon.evans@bytedance.com,
        liangma@liangbit.com
References: <20230302111227.2102545-1-usama.arif@bytedance.com>
 <faa0eb3bb8ba0326d501516a057ab46eaf1f3c05.camel@infradead.org>
From:   Tom Lendacky <thomas.lendacky@amd.com>
In-Reply-To: <faa0eb3bb8ba0326d501516a057ab46eaf1f3c05.camel@infradead.org>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?RndnWEk0R2lScWpkS2NFWk1IdS94aUdjU0oyRHdVL3V5SG8zcG9NUU5nNUFC?=
 =?utf-8?B?Rk5YMDNmN0M5NURaSVFmZzUyMFMzaW5FNnM5b3dBNjlDYTFkekRPWDRHQTJs?=
 =?utf-8?B?dGJobTZNV01sMk43ODhYRUZmYUZ5NTRuaHRnQ3grOThqd2g2dXRwMWtNcG84?=
 =?utf-8?B?b0JMRVRRR1pCb0svang2N2hoTVVReXhCYnc5ZG5IZmxqUTAxRXRTajd2aGN0?=
 =?utf-8?B?QzhnVnl6Tnd5STI5aUVlaGNqL1kvSXFRVHZKNS9sQy9BakxFSFdLZFl5Tmg3?=
 =?utf-8?B?aXJjWW1yMnJEMzFmVUNsU01BU3BJSDIvdUx6UGpkRGd3cnIvUjdHZ3N2dXYx?=
 =?utf-8?B?T0RmeGtqQlkzdVYzNXd1aktGQnBmUnRJUFhWelE5Tk43SG1tR0RMc2RCTDV4?=
 =?utf-8?B?ZXJhSmZXWGJsamZCRDgwTEtoSm5DUkRxZklTR240ajZ2L1BSdzBia0s5TWpi?=
 =?utf-8?B?RDd2a29melZHYmY0c3BkK3Y0WjYxQ3VabWFSaTBudEd5enpBWGFuWTNwejAw?=
 =?utf-8?B?WlZtanNic2hPN1BGSGw1VUJYREtSb25rMmFmZ1BwR1BodmNRVlRCMkRkV2VY?=
 =?utf-8?B?U2NXaFI1eVBZM2UyVnJBc3Q5Lzd6b0lZRDlwQ01uZldwZXJWM2xIWUtWcnNF?=
 =?utf-8?B?RjBPbDcwMTVrMFBmdHpIdXZiMFFEeEZqejlneWwwYlhPSC9XRDhqYTJQcUdS?=
 =?utf-8?B?YWtPY1hHK3RXU0JBZ1NsQlpJeS9zNWpaTk11VDI5NzFyVnBZR3dQWXpmMHU4?=
 =?utf-8?B?VWZ2S2YxMUlLZXVaaVNlMDN6bUw4WWpSVDlGaldJNHVON2hST21xTFdHZmdv?=
 =?utf-8?B?VlRvcUtvUmdxZEdneWVjMERCUEt1VjBacmY5WSttV0luL3FWdGFWUkhNS200?=
 =?utf-8?B?ZHBFT1VjY3QzaWFxaW1jWmxTL3lGeHFhTzU3VkxGKzVWYVJaUzIzQVF3KzJo?=
 =?utf-8?B?bzFtelVlZENKc0wzWEhQNGRRT2REMmZsUktUMHVqcFU5TnpPa0F4cE1wMk94?=
 =?utf-8?B?UkZGYjBuaEtSWDExR0FHQ2FuY2V0UFVLa0Y5SFgvTmlOVFUwSXBsTGdlbW40?=
 =?utf-8?B?ZWFtaW5uVkF1WmdSOStOb3NoVmZBZURRU3ZLWVZsTXB1RzlkMTVLUklPNVVm?=
 =?utf-8?B?ZjNJdHM1MDA0NzJDUmpTRTQ1TTdldzNyUzVhNTM2bzhoNkhESXVwR29HeFNS?=
 =?utf-8?B?TzZxdkNGVHhzZjBnNXdJVDl2QlZBUkhJNGNVZ3BKZnRwMEFUTTdKeWZpSnB5?=
 =?utf-8?B?WWxWcU44VkxGcUtONVAwVEFCVTdsa2w0eEZLdkRMeFZ0bTJHWnZMRGhOT2dO?=
 =?utf-8?B?Z3drZm1rTjBpaXhIamI4aEFBMS9QSFNqOFBJZHI5WUV4cFhCNnZ6S2tMTS83?=
 =?utf-8?B?SHhxRTk4YWJ1Yy9OaGhhakxDZ3hLWkJqSWc3NFNoekwvc3ZLczB6NVk4TkZU?=
 =?utf-8?B?VWF4T2RZZ09halpkbTF3ZmpYSVBnTnVZZi9RelN3S1RLbmtkL01BWnZEZ1cx?=
 =?utf-8?B?bUM2UzVEUDhScnllbXVTYTNmY3hTenlyOWswYnlCejFtMndXQXhaVlJqREVk?=
 =?utf-8?B?dnp6QXhpM2tPWitjS2FGSmhqekJZVEUvTmFtN282SnY5c0E2Ry81SFFxTnpZ?=
 =?utf-8?B?Zm9jend2Q2hxSE9uSHRENHJTb3B0dGVRcWVYTTllVEljamxtYVpYa1Q2WGM4?=
 =?utf-8?B?VXJlZWM0QVdBNzFwL05uT2VSNGhtR3JMODhiZ0N4MEcrb2lTVlR4OUE5WTRD?=
 =?utf-8?B?TE1nUjFMa09kWEYyeldKdkJIRmRQOTk4N3c2MEFJUE5aNGxSNDkvWEZDY0JY?=
 =?utf-8?B?TzRCSzJjcDJoL0wxazg2Q0RGZmNUb2k3aTRQb2lVVnFKdGJqUnJ6aDdvcGZU?=
 =?utf-8?B?NFBTNHBEa3pZQWVwTnE1cFJBM3V3a3N5Qk9aTjJ0dnNRenNYS2Jjb0QrWDZG?=
 =?utf-8?B?SE4vZ1U1WlFuaUYxSHVPTjRNUnlyajJEV0IzNGZ5OTFwekMvSi8wbFU3ZXJl?=
 =?utf-8?B?VFlqSWV0VU5iT01kSzliVXFOS05rbUtwTEdaNk14ZWcyeVRnd3lxakQwYlc2?=
 =?utf-8?B?THJwZDlvbElOaHFNcGlnK0tLTjllRkdIVXY0N1ZET0lJbVZZMFVYRjh4Skpq?=
 =?utf-8?Q?YVC6Y4yTT0bcBl3JN+MFE935f?=
X-OriginatorOrg: amd.com
X-MS-Exchange-CrossTenant-Network-Message-Id: da57c603-8e0b-4cd9-6c02-08db1f2b6fff
X-MS-Exchange-CrossTenant-AuthSource: DM4PR12MB5229.namprd12.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 07 Mar 2023 16:46:02.7840
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 3dd8961f-e488-4e60-8e11-a82d994e183d
X-MS-Exchange-CrossTenant-MailboxType: HOSTED
X-MS-Exchange-CrossTenant-UserPrincipalName: 9b6W/wR7tiFnWdyZ4mdlezHtWCeo4MlCesUIR5kQ8VKEal+VQKJ4gqpZ/W6Zu9K9bDz21rGjR6U7//JPCy3obQ==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: CY5PR12MB6552
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On 3/7/23 08:42, David Woodhouse wrote:
> On Thu, 2023-03-02 at 11:12 +0000, Usama Arif wrote:
>> The main code change over v12 is to fix the build error when
>> CONFIG_FORCE_NR_CPUS is present.
>>
>> The commit message for removing initial stack has also been improved, typos
>> have been fixed and extra comments have been added to make code clearer.
> 
> Might something like this make it work in parallel with SEV-SNP? If so,
> I can clean it up and adjust the C code to actually invoke it...

This should be ok for both SEV-ES and SEV-SNP.

> 
> diff --git a/arch/x86/include/asm/sev-common.h b/arch/x86/include/asm/sev-common.h
> index b8357d6ecd47..f25df4bd318e 100644
> --- a/arch/x86/include/asm/sev-common.h
> +++ b/arch/x86/include/asm/sev-common.h
> @@ -70,6 +70,7 @@
>   	/* GHCBData[63:12] */				\
>   	(((u64)(v) & GENMASK_ULL(63, 12)) >> 12)
>   
> +#ifndef __ASSEMBLY__
>   /*
>    * SNP Page State Change Operation
>    *
> @@ -160,6 +161,8 @@ struct snp_psc_desc {
>   
>   #define GHCB_RESP_CODE(v)		((v) & GHCB_MSR_INFO_MASK)
>   
> +#endif /* __ASSEMBLY__ */
> +
>   /*
>    * Error codes related to GHCB input that can be communicated back to the guest
>    * by setting the lower 32-bits of the GHCB SW_EXITINFO1 field to 2.
> diff --git a/arch/x86/include/asm/smp.h b/arch/x86/include/asm/smp.h
> index defe76ee9e64..0c26f80f488c 100644
> --- a/arch/x86/include/asm/smp.h
> +++ b/arch/x86/include/asm/smp.h
> @@ -204,6 +204,7 @@ extern unsigned int smpboot_control;
>   /* Control bits for startup_64 */
>   #define STARTUP_APICID_CPUID_0B	0x80000000
>   #define STARTUP_APICID_CPUID_01	0x40000000
> +#define STARTUP_APICID_SEV_SNP	0x20000000
>   
>   #define STARTUP_PARALLEL_MASK (STARTUP_APICID_CPUID_01 | STARTUP_APICID_CPUID_0B)
>   
> diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
> index c35f7c173832..b2571034562c 100644
> --- a/arch/x86/kernel/head_64.S
> +++ b/arch/x86/kernel/head_64.S
> @@ -26,7 +26,7 @@
>   #include <asm/nospec-branch.h>
>   #include <asm/fixmap.h>
>   #include <asm/smp.h>
> -
> +#include <asm/sev-common.h>
>   /*
>    * We are not able to switch in one step to the final KERNEL ADDRESS SPACE
>    * because we need identity-mapped pages.
> @@ -242,6 +242,7 @@ SYM_INNER_LABEL(secondary_startup_64_no_verify, SYM_L_GLOBAL)
>   	 *
>   	 * Bit 31	STARTUP_APICID_CPUID_0B flag (use CPUID 0x0b)
>   	 * Bit 30	STARTUP_APICID_CPUID_01 flag (use CPUID 0x01)
> +	 * Bit 29	STARTUP_APICID_SEV_SNP flag (CPUID 0x0v via GHCB MSR)
>   	 * Bit 0-24	CPU# if STARTUP_APICID_CPUID_xx flags are not set
>   	 */
>   	movl	smpboot_control(%rip), %ecx
> @@ -249,6 +250,8 @@ SYM_INNER_LABEL(secondary_startup_64_no_verify, SYM_L_GLOBAL)
>   	jnz	.Luse_cpuid_0b
>   	testl	$STARTUP_APICID_CPUID_01, %ecx
>   	jnz	.Luse_cpuid_01
> +	testl	$STARTUP_APICID_SEV_SNP, %ecx
> +	jnz	.Luse_sev_cpuid_0b
>   	andl	$0x0FFFFFFF, %ecx
>   	jmp	.Lsetup_cpu
>   
> @@ -259,6 +262,18 @@ SYM_INNER_LABEL(secondary_startup_64_no_verify, SYM_L_GLOBAL)
>   	shr	$24, %edx
>   	jmp	.Lsetup_AP
>   
> +.Luse_sev_cpuid_0b:
> +	movl	$MSR_AMD64_SEV_ES_GHCB, %ecx
> +	# GHCB_CPUID_REQ(0x0b, GHCB_CPUID_REQ_EDX)
> +	movq	$0xc00000040000000b, %rax
> +	xorl	%edx, %edx
> +	vmgexit

According to the GHCB spec, the GHCB MSR protocol is triggered when the 
GHCB value is non-zero in bits 11:0. For the CPUID function, bits 63:32 
hold the CPUID function, bits 31:30 hold the requested register and bits 
11:0 == 0x4. So this should be:

	/* Set the GHCB MSR to request CPUID 0xB_EDX */
	movl	$MSR_AMD64_SEV_ES_GHCB, %ecx
	movl	$0xc0000004, %eax
	movl	$0xb, %edx
	wrmsr

	/* Perform GHCB MSR protocol */
	vmgexit

	/*
	 * Get the result. After the RDMSR:
	 *   EAX should be 0xc0000005
	 *   EDX should have the CPUID register value and since EDX
	 *   is the target register, no need to move the result.
	 */
> +	rdmsr
> +	andl	$GHCB_MSR_INFO_MASK, %eax
> +	cmpl	$GHCB_MSR_CPUID_RESP, %eax
> +	jne	1f
> +	jmp	.Lsetup_AP
> +
>   .Luse_cpuid_0b:
>   	mov	$0x0B, %eax
>   	xorl	%ecx, %ecx
> 

Thanks,
Tom

>