90057151bb
- s390x/sclp: Simplify get_sclp_device() - target/ppc: Remove msr_pr macro - docs/system/target-arm: Re-alphabetize board list - migration: Extend query-migrate to provide dirty page limit info - migration: Implement dirty-limit convergence algo - migration: Put the detection logic before auto-converge checking - migration: Refactor auto-converge capability logic - migration: Introduce dirty-limit capability - qapi/migration: Introduce vcpu-dirty-limit parameters - qapi/migration: Introduce x-vcpu-dirty-limit-period parameter - Change the value of no_ged from true to false - Allow UNIX socket option for VNC websocket - tpm_emulator: Avoid double initialization during - chardev/char-socket: Update AF_UNIX for Windows - KVM: dirty ring: add missing memory barrier - i386: reset KVM nested state upon CPU reset - esp: Handle CMD_BUSRESET by resetting the SCSI bus - dbus-vmstate: Restrict error checks to registered proxies in dbus_get_proxies - vfio/pci: Add Ascend310b scend910b support - target/i386: Export RFDS bit to guests - target/i386: Add new CPU model SierraForest - target/i386: Introduce Icelake-Server-v7 to enable TSX - hw/virtio: handle un-configured shutdown in virtio-pci - target/s390x: display deprecation status in '-cpu help' - target/i386: display deprecation status in '-cpu help' - pc-bios/s390-ccw: Fix booting with logical block size < physical block size Signed-off-by: Jiabo Feng <fengjiabo1@huawei.com>
65 lines
2.4 KiB
Diff
65 lines
2.4 KiB
Diff
From 2f722c0308c72223e1fe13caf1a3e5843635c98e Mon Sep 17 00:00:00 2001
|
|
From: Zhenzhong Duan <zhenzhong.duan@intel.com>
|
|
Date: Wed, 20 Mar 2024 17:31:38 +0800
|
|
Subject: [PATCH] target/i386: Introduce Icelake-Server-v7 to enable TSX
|
|
|
|
commit c895fa54e3060c5ac6f3888dce96c9b78626072b upstream.
|
|
|
|
When start L2 guest with both L1/L2 using Icelake-Server-v3 or above,
|
|
QEMU reports below warning:
|
|
|
|
"warning: host doesn't support requested feature: MSR(10AH).taa-no [bit 8]"
|
|
|
|
Reason is QEMU Icelake-Server-v3 has TSX feature disabled but enables taa-no
|
|
bit. It's meaningless that TSX isn't supported but still claim TSX is secure.
|
|
So L1 KVM doesn't expose taa-no to L2 if TSX is unsupported, then starting L2
|
|
triggers the warning.
|
|
|
|
Fix it by introducing a new version Icelake-Server-v7 which has both TSX
|
|
and taa-no features. Then guest can use TSX securely when it see taa-no.
|
|
|
|
This matches the production Icelake which supports TSX and isn't susceptible
|
|
to TSX Async Abort (TAA) vulnerabilities, a.k.a, taa-no.
|
|
|
|
Ideally, TSX should have being enabled together with taa-no since v3, but for
|
|
compatibility, we'd better to add v7 to enable it.
|
|
|
|
Fixes: d965dc35592d ("target/i386: Add ARCH_CAPABILITIES related bits into Icelake-Server CPU model")
|
|
Intel-SIG: commit c895fa54e306 target/i386: Introduce Icelake-Server-v7 to enable TSX.
|
|
6.2.0-Add SRF CPU module support
|
|
|
|
Tested-by: Xiangfei Ma <xiangfeix.ma@intel.com>
|
|
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
|
|
Message-ID: <20240320093138.80267-2-zhenzhong.duan@intel.com>
|
|
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
[ Quanxian Wang: amend commit log ]
|
|
Signed-off-by: Quanxian Wang <quanxian.wang@intel.com>
|
|
---
|
|
target/i386/cpu.c | 10 ++++++++++
|
|
1 file changed, 10 insertions(+)
|
|
|
|
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
|
|
index 62ac5ed005..b5bf5ce7aa 100644
|
|
--- a/target/i386/cpu.c
|
|
+++ b/target/i386/cpu.c
|
|
@@ -3606,6 +3606,16 @@ static const X86CPUDefinition builtin_x86_defs[] = {
|
|
{ /* end of list */ }
|
|
},
|
|
},
|
|
+ {
|
|
+ .version = 7,
|
|
+ .note = "TSX, taa-no",
|
|
+ .props = (PropValue[]) {
|
|
+ /* Restore TSX features removed by -v2 above */
|
|
+ { "hle", "on" },
|
|
+ { "rtm", "on" },
|
|
+ { /* end of list */ }
|
|
+ },
|
|
+ },
|
|
{ /* end of list */ }
|
|
}
|
|
},
|
|
--
|
|
2.27.0
|
|
|