From: Oliver Upton <oliver.upton@linux.dev>
To: kvmarm@lists.linux.dev
Cc: kvm@vger.kernel.org, Marc Zyngier <maz@kernel.org>,
James Morse <james.morse@arm.com>,
Suzuki K Poulose <suzuki.poulose@arm.com>,
Zenghui Yu <yuzenghui@huawei.com>, Will Deacon <will@kernel.org>,
Cornelia Huck <cohuck@redhat.com>,
Oliver Upton <oliver.upton@linux.dev>,
Jing Zhang <jingzhangos@google.com>
Subject: [PATCH v10 06/12] KVM: arm64: Allow userspace to change ID_AA64ISAR{0-2}_EL1
Date: Wed, 20 Sep 2023 18:33:03 +0000 [thread overview]
Message-ID: <20230920183310.1163034-7-oliver.upton@linux.dev> (raw)
In-Reply-To: <20230920183310.1163034-1-oliver.upton@linux.dev>
Almost all of the features described by the ISA registers have no KVM
involvement. Allow userspace to change the value of these registers with
a couple exceptions:
- MOPS is not writable as KVM does not currently virtualize FEAT_MOPS.
- The PAuth fields are not writable as KVM requires both address and
generic authentication be enabled.
- Override the kernel's handling of BC to LOWER_SAFE.
Co-developed-by: Jing Zhang <jingzhangos@google.com>
Signed-off-by: Jing Zhang <jingzhangos@google.com>
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
---
arch/arm64/kvm/sys_regs.c | 42 ++++++++++++++++++++++++++++-----------
1 file changed, 30 insertions(+), 12 deletions(-)
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 10e3e6a736dc..71664bec2808 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -1225,6 +1225,10 @@ static s64 kvm_arm64_ftr_safe_value(u32 id, const struct arm64_ftr_bits *ftrp,
break;
}
break;
+ case SYS_ID_AA64ISAR2_EL1:
+ if (kvm_ftr.shift == ID_AA64ISAR2_EL1_BC_SHIFT)
+ kvm_ftr.type = FTR_LOWER_SAFE;
+ break;
case SYS_ID_DFR0_EL1:
if (kvm_ftr.shift == ID_DFR0_EL1_PerfMon_SHIFT)
kvm_ftr.type = FTR_LOWER_SAFE;
@@ -1851,11 +1855,14 @@ static unsigned int elx2_visibility(const struct kvm_vcpu *vcpu,
* from userspace.
*/
-/* sys_reg_desc initialiser for known cpufeature ID registers */
-#define ID_SANITISED(name) { \
+#define ID_DESC(name) \
SYS_DESC(SYS_##name), \
.access = access_id_reg, \
- .get_user = get_id_reg, \
+ .get_user = get_id_reg \
+
+/* sys_reg_desc initialiser for known cpufeature ID registers */
+#define ID_SANITISED(name) { \
+ ID_DESC(name), \
.set_user = set_id_reg, \
.visibility = id_visibility, \
.reset = kvm_read_sanitised_id_reg, \
@@ -1864,15 +1871,22 @@ static unsigned int elx2_visibility(const struct kvm_vcpu *vcpu,
/* sys_reg_desc initialiser for known cpufeature ID registers */
#define AA32_ID_SANITISED(name) { \
- SYS_DESC(SYS_##name), \
- .access = access_id_reg, \
- .get_user = get_id_reg, \
+ ID_DESC(name), \
.set_user = set_id_reg, \
.visibility = aa32_id_visibility, \
.reset = kvm_read_sanitised_id_reg, \
.val = 0, \
}
+/* sys_reg_desc initialiser for writable ID registers */
+#define ID_WRITABLE(name, mask) { \
+ ID_DESC(name), \
+ .set_user = set_id_reg, \
+ .visibility = id_visibility, \
+ .reset = kvm_read_sanitised_id_reg, \
+ .val = mask, \
+}
+
/*
* sys_reg_desc initialiser for architecturally unallocated cpufeature ID
* register with encoding Op0=3, Op1=0, CRn=0, CRm=crm, Op2=op2
@@ -1894,9 +1908,7 @@ static unsigned int elx2_visibility(const struct kvm_vcpu *vcpu,
* RAZ for the guest.
*/
#define ID_HIDDEN(name) { \
- SYS_DESC(SYS_##name), \
- .access = access_id_reg, \
- .get_user = get_id_reg, \
+ ID_DESC(name), \
.set_user = set_id_reg, \
.visibility = raz_visibility, \
.reset = kvm_read_sanitised_id_reg, \
@@ -2075,9 +2087,15 @@ static const struct sys_reg_desc sys_reg_descs[] = {
ID_UNALLOCATED(5,7),
/* CRm=6 */
- ID_SANITISED(ID_AA64ISAR0_EL1),
- ID_SANITISED(ID_AA64ISAR1_EL1),
- ID_SANITISED(ID_AA64ISAR2_EL1),
+ ID_WRITABLE(ID_AA64ISAR0_EL1, ~ID_AA64ISAR0_EL1_RES0),
+ ID_WRITABLE(ID_AA64ISAR1_EL1, ~(ID_AA64ISAR1_EL1_GPI |
+ ID_AA64ISAR1_EL1_GPA |
+ ID_AA64ISAR1_EL1_API |
+ ID_AA64ISAR1_EL1_APA)),
+ ID_WRITABLE(ID_AA64ISAR2_EL1, ~(ID_AA64ISAR2_EL1_RES0 |
+ ID_AA64ISAR2_EL1_MOPS |
+ ID_AA64ISAR2_EL1_APA3 |
+ ID_AA64ISAR2_EL1_GPA3)),
ID_UNALLOCATED(6,3),
ID_UNALLOCATED(6,4),
ID_UNALLOCATED(6,5),
--
2.42.0.515.g380fc7ccd1-goog
next prev parent reply other threads:[~2023-09-20 18:33 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-20 18:32 [PATCH v10 00/12] KVM: arm64: Enable 'writable' ID registers Oliver Upton
2023-09-20 18:32 ` [PATCH v10 01/12] KVM: arm64: Allow userspace to get the writable masks for feature " Oliver Upton
2023-09-21 9:53 ` Cornelia Huck
2023-09-21 18:20 ` Oliver Upton
2023-09-20 18:32 ` [PATCH v10 02/12] KVM: arm64: Document KVM_ARM_GET_REG_WRITABLE_MASKS Oliver Upton
2023-09-21 9:56 ` Cornelia Huck
2023-09-20 18:33 ` [PATCH v10 03/12] KVM: arm64: Use guest ID register values for the sake of emulation Oliver Upton
2023-09-20 18:33 ` [PATCH v10 04/12] KVM: arm64: Reject attempts to set invalid debug arch version Oliver Upton
2023-09-20 18:33 ` [PATCH v10 05/12] KVM: arm64: Bump up the default KVM sanitised debug version to v8p8 Oliver Upton
2023-09-20 18:33 ` Oliver Upton [this message]
2023-09-22 17:18 ` [PATCH v10 06/12] KVM: arm64: Allow userspace to change ID_AA64ISAR{0-2}_EL1 Kristina Martsenko
2023-09-22 17:20 ` Kristina Martsenko
2023-09-22 17:52 ` Oliver Upton
2023-10-03 19:41 ` Oliver Upton
2023-09-20 18:33 ` [PATCH v10 07/12] KVM: arm64: Allow userspace to change ID_AA64MMFR{0-2}_EL1 Oliver Upton
2023-09-20 18:33 ` [PATCH v10 08/12] KVM: arm64: Allow userspace to change ID_AA64PFR0_EL1 Oliver Upton
2023-09-20 18:33 ` [PATCH v10 09/12] KVM: arm64: Allow userspace to change ID_AA64ZFR0_EL1 Oliver Upton
2023-09-20 18:33 ` [PATCH v10 10/12] KVM: arm64: Document vCPU feature selection UAPIs Oliver Upton
2023-09-27 2:45 ` kernel test robot
2023-10-03 19:37 ` Oliver Upton
2023-09-20 18:33 ` [PATCH v10 11/12] KVM: arm64: selftests: Import automatic generation of sysreg defs Oliver Upton
2023-09-20 18:33 ` [PATCH v10 12/12] KVM: arm64: selftests: Test for setting ID register from usersapce Oliver Upton
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=20230920183310.1163034-7-oliver.upton@linux.dev \
--to=oliver.upton@linux.dev \
--cc=cohuck@redhat.com \
--cc=james.morse@arm.com \
--cc=jingzhangos@google.com \
--cc=kvm@vger.kernel.org \
--cc=kvmarm@lists.linux.dev \
--cc=maz@kernel.org \
--cc=suzuki.poulose@arm.com \
--cc=will@kernel.org \
--cc=yuzenghui@huawei.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