From: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
To: Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
x86@kernel.org, Paolo Bonzini <pbonzini@redhat.com>,
Juergen Gross <jgross@suse.com>, Deep Shah <sdeep@vmware.com>,
VMware Inc <pv-drivers@vmware.com>,
Vitaly Kuznetsov <vkuznets@redhat.com>,
Wanpeng Li <wanpengli@tencent.com>,
Jim Mattson <jmattson@google.com>, Joerg Roedel <joro@8bytes.org>
Cc: Peter H Anvin <hpa@zytor.com>,
Dave Hansen <dave.hansen@intel.com>,
Tony Luck <tony.luck@intel.com>,
Dan Williams <dan.j.williams@intel.com>,
Andi Kleen <ak@linux.intel.com>,
Kirill Shutemov <kirill.shutemov@linux.intel.com>,
Sean Christopherson <seanjc@google.com>,
Kuppuswamy Sathyanarayanan <knsathya@kernel.org>,
linux-kernel@vger.kernel.org
Subject: [PATCH v6 07/11] x86/tdx: Add HLT support for TDX guest
Date: Fri, 3 Sep 2021 10:28:08 -0700 [thread overview]
Message-ID: <20210903172812.1097643-8-sathyanarayanan.kuppuswamy@linux.intel.com> (raw)
In-Reply-To: <20210903172812.1097643-1-sathyanarayanan.kuppuswamy@linux.intel.com>
From: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Per Guest-Host-Communication Interface (GHCI) for Intel Trust
Domain Extensions (Intel TDX) specification, sec 3.8,
TDVMCALL[Instruction.HLT] provides HLT operation. Use it to implement
halt() and safe_halt() paravirtualization calls.
The same TDX hypercall is used to handle #VE exception due to
EXIT_REASON_HLT.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
---
Changes since v5:
* Replaced sti with STI in commit log and comments.
* Added comments for _tdx_hypercall() usage in _tdx_halt().
* Added new helper function _tdx_halt() to contain common
code between tdx_halt() and tdx_safe_halt().
* Renamed tdg_->tdx_.
* Removed BUG_ON() and used WARN_ONCE() for HLT emulation
failure.
Changes since v4:
* Added exception for EXIT_REASON_HLT in __tdx_hypercall() to
enable interrupts using sti.
Changes since v3:
* None
arch/x86/kernel/tdcall.S | 30 +++++++++++++++++++
arch/x86/kernel/tdx.c | 65 ++++++++++++++++++++++++++++++++++++++--
2 files changed, 93 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kernel/tdcall.S b/arch/x86/kernel/tdcall.S
index 2e70133bebf2..1b9649ec2e29 100644
--- a/arch/x86/kernel/tdcall.S
+++ b/arch/x86/kernel/tdcall.S
@@ -40,6 +40,9 @@
*/
#define tdcall .byte 0x66,0x0f,0x01,0xcc
+/* HLT TDVMCALL sub-function ID */
+#define EXIT_REASON_HLT 12
+
/*
* __tdx_module_call() - Helper function used by TDX guests to request
* services from the TDX module (does not include VMM services).
@@ -240,6 +243,33 @@ SYM_FUNC_START(__tdx_hypercall)
movl $TDVMCALL_EXPOSE_REGS_MASK, %ecx
+ /*
+ * For the idle loop STI needs to be called directly before
+ * the TDCALL that enters idle (EXIT_REASON_HLT case). STI
+ * enables interrupts only one instruction later. If there
+ * are any instructions between the STI and the TDCALL for
+ * HLT then an interrupt could happen in that time, but the
+ * code would go back to sleep afterwards, which can cause
+ * longer delays.
+ *
+ * This leads to significant difference in network performance
+ * benchmarks. So add a special case for EXIT_REASON_HLT to
+ * trigger STI before TDCALL. But this change is not required
+ * for all HLT cases. So use R15 register value to identify the
+ * case which needs STI. So, if R11 is EXIT_REASON_HLT and R15
+ * is 1, then call STI before TDCALL instruction. Note that R15
+ * register is not required by TDCALL ABI when triggering the
+ * hypercall for EXIT_REASON_HLT case. So use it in software to
+ * select the STI case.
+ */
+ cmpl $EXIT_REASON_HLT, %r11d
+ jne skip_sti
+ cmpl $1, %r15d
+ jne skip_sti
+ /* Set R15 register to 0, it is unused in EXIT_REASON_HLT case */
+ xor %r15, %r15
+ sti
+skip_sti:
tdcall
/* Restore output pointer to R9 */
diff --git a/arch/x86/kernel/tdx.c b/arch/x86/kernel/tdx.c
index 8d29ed07af1c..ec1008cc42d9 100644
--- a/arch/x86/kernel/tdx.c
+++ b/arch/x86/kernel/tdx.c
@@ -5,6 +5,7 @@
#define pr_fmt(fmt) "x86/tdx: " fmt
#include <asm/tdx.h>
+#include <asm/vmx.h>
/* TDX Module call Leaf IDs */
#define TDGETVEINFO 3
@@ -44,6 +45,52 @@ static inline bool cpuid_has_tdx_guest(void)
return !memcmp("IntelTDX ", sig, 12);
}
+static __cpuidle void _tdx_halt(const bool irq_disabled, const bool do_sti)
+{
+ u64 ret;
+
+ /*
+ * Emulate HLT operation via hypercall. More info about ABI
+ * can be found in TDX Guest-Host-Communication Interface
+ * (GHCI), sec 3.8.
+ *
+ * The VMM uses the "IRQ disabled" param to understand IRQ
+ * enabled status (RFLAGS.IF) of TD guest and determine
+ * whether or not it should schedule the halted vCPU if an
+ * IRQ becomes pending. E.g. if IRQs are disabled the VMM
+ * can keep the vCPU in virtual HLT, even if an IRQ is
+ * pending, without hanging/breaking the guest.
+ *
+ * do_sti parameter is used by __tdx_hypercall() to decide
+ * whether to call STI instruction before executing TDCALL
+ * instruction.
+ */
+ ret = _tdx_hypercall(EXIT_REASON_HLT, irq_disabled, 0, 0, do_sti, NULL);
+
+ /*
+ * Use WARN_ONCE() to report the failure. Since tdx_*halt() calls
+ * are also used in pv_ops, #VE handler error handler cannot be
+ * used to report the failure.
+ */
+ WARN_ONCE(ret, "HLT instruction emulation failed\n");
+}
+
+static __cpuidle void tdx_halt(void)
+{
+ const bool irq_disabled = irqs_disabled();
+ const bool do_sti = false;
+
+ _tdx_halt(irq_disabled, do_sti);
+}
+
+static __cpuidle void tdx_safe_halt(void)
+{
+ const bool irq_disabled = false; /* since sti will be called */
+ const bool do_sti = true;
+
+ _tdx_halt(irq_disabled, do_sti);
+}
+
unsigned long tdx_get_ve_info(struct ve_info *ve)
{
struct tdx_module_output out = {0};
@@ -70,8 +117,19 @@ unsigned long tdx_get_ve_info(struct ve_info *ve)
int tdx_handle_virtualization_exception(struct pt_regs *regs,
struct ve_info *ve)
{
- pr_warn("Unexpected #VE: %lld\n", ve->exit_reason);
- return -EFAULT;
+ switch (ve->exit_reason) {
+ case EXIT_REASON_HLT:
+ tdx_halt();
+ break;
+ default:
+ pr_warn("Unexpected #VE: %lld\n", ve->exit_reason);
+ return -EFAULT;
+ }
+
+ /* After successful #VE handling, move the IP */
+ regs->ip += ve->instr_len;
+
+ return 0;
}
void __init tdx_early_init(void)
@@ -81,5 +139,8 @@ void __init tdx_early_init(void)
setup_force_cpu_cap(X86_FEATURE_TDX_GUEST);
+ pv_ops.irq.safe_halt = tdx_safe_halt;
+ pv_ops.irq.halt = tdx_halt;
+
pr_info("Guest initialized\n");
}
--
2.25.1
next prev parent reply other threads:[~2021-09-03 17:28 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-09-03 17:28 [PATCH v6 00/11] Add TDX Guest Support (Initial support) Kuppuswamy Sathyanarayanan
2021-09-03 17:28 ` [PATCH v6 01/11] x86/paravirt: Move halt paravirt calls under CONFIG_PARAVIRT Kuppuswamy Sathyanarayanan
2021-09-03 18:22 ` Borislav Petkov
2021-09-03 19:03 ` Kuppuswamy, Sathyanarayanan
2021-09-03 20:33 ` Borislav Petkov
2021-09-03 17:28 ` [PATCH v6 02/11] x86/tdx: Introduce INTEL_TDX_GUEST config option Kuppuswamy Sathyanarayanan
2021-09-28 11:37 ` Joerg Roedel
2021-09-28 12:52 ` Kuppuswamy, Sathyanarayanan
2021-09-03 17:28 ` [PATCH v6 03/11] x86/cpufeatures: Add TDX Guest CPU feature Kuppuswamy Sathyanarayanan
2021-09-28 11:39 ` Joerg Roedel
2021-09-28 12:53 ` Kuppuswamy, Sathyanarayanan
2021-09-03 17:28 ` [PATCH v6 04/11] x86/tdx: Add protected guest support for TDX guest Kuppuswamy Sathyanarayanan
2021-09-28 11:46 ` Joerg Roedel
2021-09-28 12:56 ` Kuppuswamy, Sathyanarayanan
2021-09-28 13:04 ` Joerg Roedel
2021-09-28 13:18 ` Borislav Petkov
2021-09-03 17:28 ` [PATCH v6 05/11] x86/tdx: Add __tdx_module_call() and __tdx_hypercall() helper functions Kuppuswamy Sathyanarayanan
2021-09-03 17:28 ` [PATCH v6 06/11] x86/traps: Add #VE support for TDX guest Kuppuswamy Sathyanarayanan
2021-09-28 12:16 ` Joerg Roedel
2021-09-28 14:05 ` Dave Hansen
2021-09-28 15:22 ` Joerg Roedel
2021-09-28 15:25 ` Dave Hansen
2021-09-28 15:23 ` Dave Hansen
2021-09-28 16:59 ` Kuppuswamy, Sathyanarayanan
2021-09-03 17:28 ` Kuppuswamy Sathyanarayanan [this message]
2021-09-03 17:28 ` [PATCH v6 08/11] x86/tdx: Wire up KVM hypercalls Kuppuswamy Sathyanarayanan
2021-09-03 17:28 ` [PATCH v6 09/11] x86/tdx: Add MSR support for TDX guest Kuppuswamy Sathyanarayanan
2021-09-03 17:28 ` [PATCH v6 10/11] x86/tdx: Don't write CSTAR MSR on Intel Kuppuswamy Sathyanarayanan
2021-09-03 17:28 ` [PATCH v6 11/11] x86/tdx: Handle CPUID via #VE Kuppuswamy Sathyanarayanan
2021-09-03 18:35 ` Dave Hansen
2021-09-03 19:14 ` Kuppuswamy, Sathyanarayanan
2021-09-03 23:43 ` Sean Christopherson
2021-09-03 23:54 ` Andi Kleen
2021-09-04 0:00 ` Dave Hansen
2021-09-04 0:05 ` Andi Kleen
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=20210903172812.1097643-8-sathyanarayanan.kuppuswamy@linux.intel.com \
--to=sathyanarayanan.kuppuswamy@linux.intel.com \
--cc=ak@linux.intel.com \
--cc=bp@alien8.de \
--cc=dan.j.williams@intel.com \
--cc=dave.hansen@intel.com \
--cc=hpa@zytor.com \
--cc=jgross@suse.com \
--cc=jmattson@google.com \
--cc=joro@8bytes.org \
--cc=kirill.shutemov@linux.intel.com \
--cc=knsathya@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=pbonzini@redhat.com \
--cc=pv-drivers@vmware.com \
--cc=sdeep@vmware.com \
--cc=seanjc@google.com \
--cc=tglx@linutronix.de \
--cc=tony.luck@intel.com \
--cc=vkuznets@redhat.com \
--cc=wanpengli@tencent.com \
--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;
as well as URLs for NNTP newsgroup(s).