From: Brijesh Singh <brijesh.singh@amd.com>
To: Borislav Petkov <bp@suse.de>
Cc: "Thomas Gleixner" <tglx@linutronix.de>,
"Ingo Molnar" <mingo@redhat.com>,
"H. Peter Anvin" <hpa@zytor.com>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Radim Krčmář" <rkrcmar@redhat.com>,
kvm@vger.kernel.org, x86@kernel.org,
linux-kernel@vger.kernel.org,
"Brijesh Singh" <brijesh.singh@amd.com>
Subject: Re: [PATCH] x86/CPU/AMD, mm: Extend with mem_encrypt=sme option
Date: Sun, 1 Oct 2017 14:45:09 -0500 [thread overview]
Message-ID: <20171001194509.4187-1-brijesh.singh@amd.com> (raw)
In-Reply-To: <20171001171617.uzwfzps7sxowfram@pd.tnic>
>
> So I want to be able to disable SEV and the whole code that comes with
> it in the *host*.
We can add a new variable 'sme_only'. By default this variable should be set
to false. When mem_encrypt=sme is passed then set it to true and
based on sme_only state early_detect_mem_encrypt() can clear X86_FEATURE_SEV
flag.
Here are the changes on top of your patch. I did a quick test in both host and
guest OS and it seems to be working okay. In host OS mem_encrypt=sme disabled
the SEV but in guest its still don't care. I will do more test later...
diff --git a/arch/x86/include/asm/mem_encrypt.h b/arch/x86/include/asm/mem_encrypt.h
index 175310f00202..73a6fb3b14a1 100644
--- a/arch/x86/include/asm/mem_encrypt.h
+++ b/arch/x86/include/asm/mem_encrypt.h
@@ -19,7 +19,7 @@
#include <asm/bootparam.h>
-extern bool sev_enabled;
+extern bool sme_only;
#ifdef CONFIG_AMD_MEM_ENCRYPT
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index d0669f3966a6..a09b02959874 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -33,7 +33,7 @@ static bool cpu_has_amd_erratum(struct cpuinfo_x86 *cpu, const int *erratum);
*/
static u32 nodes_per_socket = 1;
-bool sev_enabled __section(.data) = false;
+bool sme_only __section(.data) = false;
static inline int rdmsrl_amd_safe(unsigned msr, unsigned long long *p)
{
@@ -591,7 +591,7 @@ static void early_detect_mem_encrypt(struct cpuinfo_x86 *c)
if (IS_ENABLED(CONFIG_X86_32))
goto clear_all;
- if (!sev_enabled)
+ if (sme_only)
goto clear_sev;
rdmsrl(MSR_K7_HWCR, msr);
diff --git a/arch/x86/mm/mem_encrypt.c b/arch/x86/mm/mem_encrypt.c
index 9b83bc1be7c0..a135e4497021 100644
--- a/arch/x86/mm/mem_encrypt.c
+++ b/arch/x86/mm/mem_encrypt.c
@@ -45,6 +45,7 @@ u64 sme_me_mask __section(.data) = 0;
EXPORT_SYMBOL_GPL(sme_me_mask);
DEFINE_STATIC_KEY_FALSE(__sev);
EXPORT_SYMBOL_GPL(__sev);
+static bool sev_enabled __section(.data) = false;
/* Buffer used for early in-place encryption by BSP, no locking needed */
static char sme_early_buffer[PAGE_SIZE] __aligned(PAGE_SIZE);
@@ -773,7 +774,6 @@ void __init __nostackprotector sme_enable(struct boot_params *bp)
unsigned long feature_mask;
u64 me_mask, msr;
char buffer[16];
- bool sme_only;
int ret;
/* Check for the SME/SEV support leaf */
@@ -808,6 +808,8 @@ void __init __nostackprotector sme_enable(struct boot_params *bp)
if (!(eax & feature_mask))
return;
+ me_mask = BIT_ULL(ebx & 0x3f);
+
/* For SME, check the SYSCFG MSR */
if (feature_mask == AMD_SME_BIT) {
msr = __rdmsr(MSR_K8_SYSCFG);
@@ -820,9 +822,13 @@ void __init __nostackprotector sme_enable(struct boot_params *bp)
msr = __rdmsr(MSR_AMD64_SEV);
if (!(msr & MSR_AMD64_SEV_ENABLED))
return;
- }
- me_mask = BIT_ULL(ebx & 0x3f);
+ if (feature_mask == AMD_SEV_BIT) {
+ sme_me_mask = me_mask;
+ sev_enabled = true;
+ return;
+ }
+ }
/*
* Fixups have not been applied to phys_base yet and we're running
@@ -847,16 +853,11 @@ void __init __nostackprotector sme_enable(struct boot_params *bp)
} else if (!strncmp(buffer, cmd_on, sizeof(buffer))) {
sme_me_mask = me_mask;
} else if (!strncmp(buffer, cmd_sme, sizeof(buffer))) {
+ sme_me_mask = me_mask;
sme_only = true;
+ return;
}
if (IS_ENABLED(CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT))
sme_me_mask = me_mask;
-
- if (sme_only)
- return;
-
- /* For SEV, check the SEV MSR */
- if (feature_mask == AMD_SEV_BIT)
- sev_enabled = true;
}
--
2.9.5
next prev parent reply other threads:[~2017-10-01 19:45 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-09-27 15:13 [Part1 PATCH v5 00/17] x86: Secure Encrypted Virtualization (AMD) Brijesh Singh
2017-09-27 15:13 ` [Part1 PATCH v5 01/17] Documentation/x86: Add AMD Secure Encrypted Virtualization (SEV) description Brijesh Singh
2017-09-27 15:13 ` [Part1 PATCH v5 02/17] x86/mm: Add Secure Encrypted Virtualization (SEV) support Brijesh Singh
2017-09-28 9:02 ` Borislav Petkov
2017-09-28 18:48 ` Brijesh Singh
2017-09-28 19:23 ` Borislav Petkov
2017-09-29 12:28 ` Brijesh Singh
2017-09-29 14:41 ` Borislav Petkov
2017-09-29 15:54 ` Brijesh Singh
2017-09-29 15:56 ` Borislav Petkov
2017-09-29 23:06 ` [Part1 PATCH v5 18/18] x86/mm: add 'sme' argument in mem_encrypt= Brijesh Singh
2017-09-30 11:56 ` [PATCH] x86/CPU/AMD, mm: Extend with mem_encrypt=sme option Borislav Petkov
2017-09-30 21:17 ` Brijesh Singh
2017-09-30 21:41 ` Borislav Petkov
2017-10-01 17:00 ` Brijesh Singh
2017-10-01 17:16 ` Borislav Petkov
2017-10-01 19:45 ` Brijesh Singh [this message]
2017-10-01 22:02 ` Borislav Petkov
2017-10-02 11:32 ` Brijesh Singh
2017-10-02 12:41 ` Borislav Petkov
2017-10-02 15:07 ` Brijesh Singh
2017-10-03 10:50 ` Paolo Bonzini
2017-10-03 11:20 ` Borislav Petkov
2017-10-02 13:44 ` Tom Lendacky
2017-10-02 13:51 ` Borislav Petkov
2017-10-02 16:35 ` Tom Lendacky
2017-10-03 11:29 ` Borislav Petkov
2017-09-27 15:13 ` [Part1 PATCH v5 03/17] x86/mm: Don't attempt to encrypt initrd under SEV Brijesh Singh
2017-09-27 15:13 ` [Part1 PATCH v5 04/17] x86/realmode: Don't decrypt trampoline area " Brijesh Singh
2017-09-27 15:13 ` [Part1 PATCH v5 05/17] x86/mm: Use encrypted access of boot related data with SEV Brijesh Singh
2017-09-28 11:13 ` Borislav Petkov
2017-09-28 16:20 ` [Part1 PATCH v5.1 " Brijesh Singh
2017-09-27 15:13 ` [Part1 PATCH v5 06/17] x86/mm: Include SEV for encryption memory attribute changes Brijesh Singh
2017-09-27 15:53 ` Brijesh Singh
2017-09-27 17:26 ` Borislav Petkov
2017-09-27 19:17 ` [Part1 PATCH v5.1 " Brijesh Singh
2017-09-27 15:13 ` [Part1 PATCH v5 07/17] x86/efi: Access EFI data as encrypted when SEV is active Brijesh Singh
2017-09-27 15:13 ` [Part1 PATCH v5 08/17] resource: Consolidate resource walking code Brijesh Singh
2017-09-27 15:13 ` [Part1 PATCH v5 09/17] resource: Provide resource struct in resource walk callback Brijesh Singh
2017-09-27 15:13 ` [Part1 PATCH v5 10/17] x86/mm, resource: Use PAGE_KERNEL protection for ioremap of memory pages Brijesh Singh
2017-09-28 16:23 ` Borislav Petkov
2017-09-27 15:13 ` [Part1 PATCH v5 11/17] x86/mm: Add DMA support for SEV memory encryption Brijesh Singh
2017-09-27 15:13 ` [Part1 PATCH v5 12/17] x86/boot: Add early boot support when running with SEV active Brijesh Singh
2017-09-28 17:02 ` Borislav Petkov
2017-09-27 15:13 ` [Part1 PATCH v5 13/17] x86/io: Unroll string I/O when SEV is active Brijesh Singh
2017-09-28 17:51 ` Borislav Petkov
2017-09-27 15:13 ` [Part1 PATCH v5 14/17] x86: Add support for changing memory encryption attribute in early boot Brijesh Singh
2017-09-27 15:13 ` [Part1 PATCH v5 15/17] percpu: Introduce DEFINE_PER_CPU_DECRYPTED Brijesh Singh
2017-09-28 20:32 ` Borislav Petkov
2017-09-27 15:13 ` [Part1 PATCH v5 16/17] X86/KVM: Decrypt shared per-cpu variables when SEV is active Brijesh Singh
2017-09-29 5:51 ` Borislav Petkov
2017-09-27 15:13 ` [Part1 PATCH v5 17/17] X86/KVM: Clear encryption attribute " Brijesh Singh
2017-09-29 6:26 ` [Part1 PATCH v5 00/17] x86: Secure Encrypted Virtualization (AMD) Borislav Petkov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20171001194509.4187-1-brijesh.singh@amd.com \
--to=brijesh.singh@amd.com \
--cc=bp@suse.de \
--cc=hpa@zytor.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=pbonzini@redhat.com \
--cc=rkrcmar@redhat.com \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox