From: Xiaoyao Li <xiaoyao.li@intel.com>
To: "Paolo Bonzini" <pbonzini@redhat.com>,
"Philippe Mathieu-Daudé" <f4bug@amsat.org>,
"Richard Henderson" <richard.henderson@linaro.org>,
"Michael S. Tsirkin" <mst@redhat.com>,
"Marcel Apfelbaum" <marcel.apfelbaum@gmail.com>,
"Cornelia Huck" <cohuck@redhat.com>,
"Daniel P. Berrangé" <berrange@redhat.com>,
"Marcelo Tosatti" <mtosatti@redhat.com>,
"Laszlo Ersek" <lersek@redhat.com>,
"Gerd Hoffmann" <kraxel@redhat.com>,
"Eric Blake" <eblake@redhat.com>
Cc: isaku.yamahata@intel.com, kvm@vger.kernel.org,
Connor Kuehl <ckuehl@redhat.com>,
seanjc@google.com, xiaoyao.li@intel.com, qemu-devel@nongnu.org,
erdemaktas@google.com
Subject: [RFC PATCH v3 08/36] i386/tdx: Adjust get_supported_cpuid() for TDX VM
Date: Thu, 17 Mar 2022 21:58:45 +0800 [thread overview]
Message-ID: <20220317135913.2166202-9-xiaoyao.li@intel.com> (raw)
In-Reply-To: <20220317135913.2166202-1-xiaoyao.li@intel.com>
For TDX, the allowable CPUID configuration differs from what KVM
reports for KVM scope via KVM_GET_SUPPORTED_CPUID.
- Some CPUID bits are not supported for TDX VM while KVM reports the
support. Mask them off for TDX VM. e.g., CPUID_EXT_VMX, some PV
featues.
- The supported XCR0 and XSS bits needs to be caped by tdx_caps, because
KVM uses them to setup XFAM of TD.
Introduce tdx_get_supported_cpuid() to adjust the
kvm_arch_get_supported_cpuid() for TDX VM.
Signed-off-by: Xiaoyao Li <xiaoyao.li@intel.com>
---
target/i386/cpu.h | 5 +++++
target/i386/kvm/kvm.c | 4 ++++
target/i386/kvm/tdx.c | 39 +++++++++++++++++++++++++++++++++++++++
target/i386/kvm/tdx.h | 2 ++
4 files changed, 50 insertions(+)
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 5e406088a91a..7fa30f4ed7db 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -566,6 +566,11 @@ typedef enum X86Seg {
#define ESA_FEATURE_XFD_MASK (1U << ESA_FEATURE_XFD_BIT)
+#define XCR0_MASK (XSTATE_FP_MASK | XSTATE_SSE_MASK | XSTATE_YMM_MASK | \
+ XSTATE_BNDREGS_MASK | XSTATE_BNDCSR_MASK | \
+ XSTATE_OPMASK_MASK | XSTATE_ZMM_Hi256_MASK | \
+ XSTATE_Hi16_ZMM_MASK | XSTATE_PKRU_MASK)
+
/* CPUID feature words */
typedef enum FeatureWord {
FEAT_1_EDX, /* CPUID[1].EDX */
diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c
index 26ed5faf07b8..ddbe8f64fadb 100644
--- a/target/i386/kvm/kvm.c
+++ b/target/i386/kvm/kvm.c
@@ -486,6 +486,10 @@ uint32_t kvm_arch_get_supported_cpuid(KVMState *s, uint32_t function,
ret |= 1U << KVM_HINTS_REALTIME;
}
+ if (is_tdx_vm()) {
+ tdx_get_supported_cpuid(function, index, reg, &ret);
+ }
+
return ret;
}
diff --git a/target/i386/kvm/tdx.c b/target/i386/kvm/tdx.c
index 846511b299f4..e4ee55f30c79 100644
--- a/target/i386/kvm/tdx.c
+++ b/target/i386/kvm/tdx.c
@@ -14,6 +14,7 @@
#include "qemu/osdep.h"
#include "qapi/error.h"
#include "qom/object_interfaces.h"
+#include "standard-headers/asm-x86/kvm_para.h"
#include "sysemu/kvm.h"
#include "hw/i386/x86.h"
@@ -110,6 +111,44 @@ int tdx_kvm_init(MachineState *ms, Error **errp)
return 0;
}
+void tdx_get_supported_cpuid(uint32_t function, uint32_t index, int reg,
+ uint32_t *ret)
+{
+ switch (function) {
+ case 1:
+ if (reg == R_ECX) {
+ *ret &= ~CPUID_EXT_VMX;
+ }
+ break;
+ case 0xd:
+ if (index == 0) {
+ if (reg == R_EAX) {
+ *ret &= (uint32_t)tdx_caps->xfam_fixed0 & XCR0_MASK;
+ *ret |= (uint32_t)tdx_caps->xfam_fixed1 & XCR0_MASK;
+ } else if (reg == R_EDX) {
+ *ret &= (tdx_caps->xfam_fixed0 & XCR0_MASK) >> 32;
+ *ret |= (tdx_caps->xfam_fixed1 & XCR0_MASK) >> 32;
+ }
+ } else if (index == 1) {
+ /* TODO: Adjust XSS when it's supported. */
+ }
+ break;
+ case KVM_CPUID_FEATURES:
+ if (reg == R_EAX) {
+ *ret &= ~((1ULL << KVM_FEATURE_CLOCKSOURCE) |
+ (1ULL << KVM_FEATURE_CLOCKSOURCE2) |
+ (1ULL << KVM_FEATURE_CLOCKSOURCE_STABLE_BIT) |
+ (1ULL << KVM_FEATURE_ASYNC_PF) |
+ (1ULL << KVM_FEATURE_ASYNC_PF_VMEXIT) |
+ (1ULL << KVM_FEATURE_ASYNC_PF_INT));
+ }
+ break;
+ default:
+ /* TODO: Use tdx_caps to adjust CPUID leafs. */
+ break;
+ }
+}
+
/* tdx guest */
OBJECT_DEFINE_TYPE_WITH_INTERFACES(TdxGuest,
tdx_guest,
diff --git a/target/i386/kvm/tdx.h b/target/i386/kvm/tdx.h
index 4036ca2f3f99..06599b65b827 100644
--- a/target/i386/kvm/tdx.h
+++ b/target/i386/kvm/tdx.h
@@ -27,5 +27,7 @@ bool is_tdx_vm(void);
#endif /* CONFIG_TDX */
int tdx_kvm_init(MachineState *ms, Error **errp);
+void tdx_get_supported_cpuid(uint32_t function, uint32_t index, int reg,
+ uint32_t *ret);
#endif /* QEMU_I386_TDX_H */
--
2.27.0
next prev parent reply other threads:[~2022-03-17 14:08 UTC|newest]
Thread overview: 77+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-03-17 13:58 [RFC PATCH v3 00/36] TDX QEMU support Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 01/36] *** HACK *** linux-headers: Update headers to pull in TDX API changes Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 02/36] i386: Introduce tdx-guest object Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 03/36] target/i386: Implement mc->kvm_type() to get VM type Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 04/36] target/i386: Introduce kvm_confidential_guest_init() Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 05/36] i386/tdx: Implement tdx_kvm_init() to initialize TDX VM context Xiaoyao Li
2022-03-18 2:07 ` Isaku Yamahata
2022-03-21 5:35 ` Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 06/36] i386/tdx: Get tdx_capabilities via KVM_TDX_CAPABILITIES Xiaoyao Li
2022-03-18 2:08 ` Isaku Yamahata
2022-03-21 6:56 ` Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 07/36] i386/tdx: Introduce is_tdx_vm() helper and cache tdx_guest object Xiaoyao Li
2022-03-17 13:58 ` Xiaoyao Li [this message]
2022-03-18 16:55 ` [RFC PATCH v3 08/36] i386/tdx: Adjust get_supported_cpuid() for TDX VM Isaku Yamahata
2022-03-21 5:37 ` Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 09/36] KVM: Introduce kvm_arch_pre_create_vcpu() Xiaoyao Li
2022-03-18 16:56 ` Isaku Yamahata
2022-03-21 7:02 ` Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 10/36] i386/kvm: Move architectural CPUID leaf generation to separate helper Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 11/36] i386/tdx: Initialize TDX before creating TD vcpus Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 12/36] i386/tdx: Add property sept-ve-disable for tdx-guest object Xiaoyao Li
2022-03-22 9:02 ` Gerd Hoffmann
2022-03-24 6:52 ` Xiaoyao Li
2022-03-24 7:57 ` Gerd Hoffmann
2022-03-24 8:08 ` Xiaoyao Li
2022-03-24 9:37 ` Gerd Hoffmann
2022-03-24 14:36 ` Xiaoyao Li
2022-03-25 1:35 ` Isaku Yamahata
2022-03-17 13:58 ` [RFC PATCH v3 13/36] i386/tdx: Wire CPU features up with attributes of TD guest Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 14/36] i386/tdx: Validate TD attributes Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 15/36] i386/tdx: Implement user specified tsc frequency Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 16/36] i386/tdx: Set kvm_readonly_mem_enabled to false for TDX VM Xiaoyao Li
2022-03-18 17:11 ` Isaku Yamahata
2022-03-21 8:15 ` Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 17/36] pflash_cfi01/tdx: Introduce ram_mode of pflash for TDVF Xiaoyao Li
2022-03-18 14:07 ` Philippe Mathieu-Daudé
2022-03-21 8:54 ` Xiaoyao Li
2022-03-21 22:06 ` Isaku Yamahata
2022-03-22 9:21 ` Gerd Hoffmann
2022-03-22 9:29 ` Daniel P. Berrangé
2022-03-22 10:35 ` Gerd Hoffmann
2022-03-22 10:51 ` Daniel P. Berrangé
2022-03-22 12:20 ` Gerd Hoffmann
2022-03-24 8:35 ` Gerd Hoffmann
2022-03-31 6:57 ` Xiaoyao Li
2022-03-24 6:13 ` Xiaoyao Li
2022-03-24 7:58 ` Gerd Hoffmann
2022-03-24 8:18 ` Xiaoyao Li
2022-03-24 8:52 ` Daniel P. Berrangé
2022-03-22 9:27 ` Daniel P. Berrangé
2022-03-31 8:51 ` Xiaoyao Li
2022-03-31 9:00 ` Daniel P. Berrangé
2022-03-31 14:50 ` Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 18/36] i386/tdvf: Introduce function to parse TDVF metadata Xiaoyao Li
2022-03-18 17:19 ` Isaku Yamahata
2022-03-21 6:11 ` Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 19/36] i386/tdx: Parse TDVF metadata for TDX VM Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 20/36] i386/tdx: Get and store the mem_ptr of TDVF firmware Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 21/36] i386/tdx: Track mem_ptr for each firmware entry of TDVF Xiaoyao Li
2022-03-17 13:58 ` [RFC PATCH v3 22/36] i386/tdx: Track RAM entries for TDX VM Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 23/36] i386/tdx: Create the TD HOB list upon machine init done Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 24/36] i386/tdx: Call KVM_TDX_INIT_VCPU to initialize TDX vcpu Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 25/36] i386/tdx: Add TDVF memory via KVM_TDX_INIT_MEM_REGION Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 26/36] i386/tdx: Finalize TDX VM Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 27/36] i386/tdx: Disable SMM for TDX VMs Xiaoyao Li
2022-03-21 6:51 ` Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 28/36] i386/tdx: Disable PIC " Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 29/36] i386/tdx: Don't allow system reset " Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 30/36] hw/i386: add eoi_intercept_unsupported member to X86MachineState Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 31/36] hw/i386: add option to forcibly report edge trigger in acpi tables Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 32/36] i386/tdx: Don't synchronize guest tsc for TDs Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 33/36] i386/tdx: Only configure MSR_IA32_UCODE_REV in kvm_init_msrs() " Xiaoyao Li
2022-03-18 17:31 ` Isaku Yamahata
2022-03-21 6:08 ` Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 34/36] i386/tdx: Skip kvm_put_apicbase() " Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 35/36] i386/tdx: Don't get/put guest state for TDX VMs Xiaoyao Li
2022-03-17 13:59 ` [RFC PATCH v3 36/36] docs: Add TDX documentation Xiaoyao Li
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=20220317135913.2166202-9-xiaoyao.li@intel.com \
--to=xiaoyao.li@intel.com \
--cc=berrange@redhat.com \
--cc=ckuehl@redhat.com \
--cc=cohuck@redhat.com \
--cc=eblake@redhat.com \
--cc=erdemaktas@google.com \
--cc=f4bug@amsat.org \
--cc=isaku.yamahata@intel.com \
--cc=kraxel@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=lersek@redhat.com \
--cc=marcel.apfelbaum@gmail.com \
--cc=mst@redhat.com \
--cc=mtosatti@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=richard.henderson@linaro.org \
--cc=seanjc@google.com \
/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;
as well as URLs for NNTP newsgroup(s).