Skip to contentSkip to conversation

arm64: Add initial support for E0PD

PATCH In review Mark Brown 3 participants 6 messages Jun 27, 2019 – Jun 29, 2019 arm64

lore
Share & tools
lore

Mark Brown

author
#0 lore raw

PATCH v2 1/1 arm64: Add initial support for E0PD

Kernel Page Table Isolation (KPTI) is used to mitigate some speculation based security issues by ensuring that the kernel is not mapped when userspace is running but this approach is expensive and is incompatible with SPE. E0PD, introduced in the ARMv8.5 extensions, provides an alternative to this which ensures that accesses from userspace to the kernel's half of the memory map to always fault with constant time, preventing timing attacks without requiring constant unmapping and remapping or preventing legitimate accesses.

This initial patch does not yet integrate with KPTI, a followup patch will ensure that by default we don't use KPTI on CPUs where E0PD is present which will provide a much greater benefit in general configurations.

---

- Reword commit message
- Change to detection as ARM64_CPUCAP_WEAK_LOCAL_CPU_FEATURE.

 arch/arm64/Kconfig                     | 14 ++++++++++++++
 arch/arm64/include/asm/cpucaps.h       |  3 ++-
 arch/arm64/include/asm/pgtable-hwdef.h |  2 ++
 arch/arm64/include/asm/sysreg.h        |  1 +
 arch/arm64/kernel/cpufeature.c         | 21 +++++++++++++++++++++
 5 files changed, 40 insertions(+), 1 deletion(-)
arch/arm64/Kconfig +14 -0
index 0758d89524d0..6bb0d6b578d3 100644
@@ -1382,6 +1382,20 @@ config ARM64_PTR_AUTH
 
 endmenu
 
+menu "ARMv8.5 architectural features"
+
+config ARM64_E0PD
+	bool "Enable support for E0PD"
+	default y
+	help
+	   E0PD (part of the ARMv8.5 extensions) ensures that EL0
+	   accesses made via TTBR1 always fault in constant time,
+	   providing the same guarantees as KPTI with lower overhead.
+
+	   This option enables E0PD where available.
+
+endmenu
+
 config ARM64_SVE
 	bool "ARM Scalable Vector Extension support"
 	default y
arch/arm64/include/asm/cpucaps.h +2 -1
index f19fe4b9acc4..f25388981075 100644
@@ -52,7 +52,8 @@
 #define ARM64_HAS_IRQ_PRIO_MASKING		42
 #define ARM64_HAS_DCPODP			43
 #define ARM64_WORKAROUND_1463225		44
+#define ARM64_HAS_E0PD				45
 
-#define ARM64_NCAPS				45
+#define ARM64_NCAPS				46
 
 #endif /* __ASM_CPUCAPS_H */
arch/arm64/include/asm/pgtable-hwdef.h +2 -0
index e2f8c6b09717..195a01156460 100644
@@ -294,6 +294,8 @@
 #define TCR_HD			(UL(1) << 40)
 #define TCR_NFD0		(UL(1) << 53)
 #define TCR_NFD1		(UL(1) << 54)
+#define TCR_E0PD0		(UL(1) << 55)
+#define TCR_E0PD1		(UL(1) << 56)
 
 /*
  * TTBR.
arch/arm64/include/asm/sysreg.h +1 -0
index b069b673494f..2f3672c186dc 100644
@@ -650,6 +650,7 @@
 #define ID_AA64MMFR1_VMIDBITS_16	2
 
 /* id_aa64mmfr2 */
+#define ID_AA64MMFR2_E0PD_SHIFT		60
 #define ID_AA64MMFR2_FWB_SHIFT		40
 #define ID_AA64MMFR2_AT_SHIFT		32
 #define ID_AA64MMFR2_LVA_SHIFT		16
arch/arm64/kernel/cpufeature.c +21 -0
index f29f36a65175..676ec3db35a3 100644
@@ -211,6 +211,7 @@ static const struct arm64_ftr_bits ftr_id_aa64mmfr1[] = {
 };
 
 static const struct arm64_ftr_bits ftr_id_aa64mmfr2[] = {
+	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64MMFR2_E0PD_SHIFT, 4, 0),
 	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64MMFR2_FWB_SHIFT, 4, 0),
 	ARM64_FTR_BITS(FTR_VISIBLE, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64MMFR2_AT_SHIFT, 4, 0),
 	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64MMFR2_LVA_SHIFT, 4, 0),
@@ -1232,6 +1233,13 @@ static void cpu_enable_address_auth(struct arm64_cpu_capabilities const *cap)
 }
 #endif /* CONFIG_ARM64_PTR_AUTH */
 
+#ifdef CONFIG_ARM64_E0PD
+static void cpu_enable_e0pd(struct arm64_cpu_capabilities const *cap)
+{
+	sysreg_clear_set(tcr_el1, 0, TCR_E0PD1);
+}
+#endif /* CONFIG_ARM64_E0PD */
+
 #ifdef CONFIG_ARM64_PSEUDO_NMI
 static bool enable_pseudo_nmi;
 
@@ -1547,6 +1555,19 @@ static const struct arm64_cpu_capabilities arm64_features[] = {
 		.sign = FTR_UNSIGNED,
 		.min_field_value = 1,
 	},
+#endif
+#ifdef CONFIG_ARM64_E0PD
+	{
+		.desc = "E0PD",
+		.capability = ARM64_HAS_E0PD,
+		.type = ARM64_CPUCAP_WEAK_LOCAL_CPU_FEATURE,
+		.sys_reg = SYS_ID_AA64MMFR2_EL1,
+		.sign = FTR_UNSIGNED,
+		.field_pos = ID_AA64MMFR2_E0PD_SHIFT,
+		.matches = has_cpuid_feature,
+		.min_field_value = 1,
+		.cpu_enable = cpu_enable_e0pd,
+	},
 #endif
 	{},
 };
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
maintainer
#1 lore raw
Quoted 14 lines · Mark Brown

Kernel Page Table Isolation (KPTI) is used to mitigate some speculation based security issues by ensuring that the kernel is not mapped when userspace is running but this approach is expensive and is incompatible with SPE. E0PD, introduced in the ARMv8.5 extensions, provides an alternative to this which ensures that accesses from userspace to the kernel's half of the memory map to always fault with constant time, preventing timing attacks without requiring constant unmapping and remapping or preventing legitimate accesses.

This initial patch does not yet integrate with KPTI, a followup patch will ensure that by default we don't use KPTI on CPUs where E0PD is present which will provide a much greater benefit in general configurations.

Quoted 5 lines

---

- Reword commit message
- Change to detection as ARM64_CPUCAP_WEAK_LOCAL_CPU_FEATURE.

Thanks for the quick v2, although I still don't see the rush to merge this without the associated kaslr/kpti changes we've been discussing. It's not like there's a whole load of 8.5 silicon we're rushing to support.

Will

_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

author
#2 lore raw
Quoted 3 lines · Will Deacon

Thanks for the quick v2, although I still don't see the rush to merge this without the associated kaslr/kpti changes we've been discussing. It's not like there's a whole load of 8.5 silicon we're rushing to support.

It's largely on the general theory that it's better to carry less out of tree code - there's less diff to manage, less chance of collisions with other work (in either direction), and less pending review to worry about. So long as something represents forward progress I tend to work on the basis of why not rather than why.

_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

maintainer
#3 lore raw
Quoted 10 lines · Mark Brown
Quoted 3 lines · Will Deacon

Thanks for the quick v2, although I still don't see the rush to merge this without the associated kaslr/kpti changes we've been discussing. It's not like there's a whole load of 8.5 silicon we're rushing to support.

It's largely on the general theory that it's better to carry less out of tree code - there's less diff to manage, less chance of collisions with other work (in either direction), and less pending review to worry about. So long as something represents forward progress I tend to work on the basis of why not rather than why.

A reason is that once we add the support for disabling kpti based on the E0PD feature, this patch may turn out to be slightly different (for example, you may add a common has_e0pd() check that is called from both unmap_kernel_at_el0() and the E0PD arm64_features[] entry). Given that both patches would be relatively small, I agree with Will that there is no rush to merge them independently.

Signature · 6 lines
Catalin

_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author
#4 lore raw
Quoted 8 lines · Catalin Marinas
Quoted 5 lines · Mark Brown

It's largely on the general theory that it's better to carry less out of tree code - there's less diff to manage, less chance of collisions with other work (in either direction), and less pending review to worry about. So long as something represents forward progress I tend to work on the basis of why not rather than why.

Quoted 4 lines

A reason is that once we add the support for disabling kpti based on the E0PD feature, this patch may turn out to be slightly different (for example, you may add a common has_e0pd() check that is called from both unmap_kernel_at_el0() and the E0PD arm64_features[] entry). Given

Hrm, I don't really get that - incremental patches can always be done (indeed they're often really helpful for people trying to understand how the code got to be the way it is). Obviously you can go too far in the other direction as well, a middle ground is generally best.

Quoted 2 lines

that both patches would be relatively small, I agree with Will that there is no rush to merge them independently.

I guess.

_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

maintainer
#5 lore raw

Hi Mark,

Quoted 9 lines · Mark Brown
Quoted 7 lines · Catalin Marinas
Quoted 5 lines · Mark Brown

It's largely on the general theory that it's better to carry less out of tree code - there's less diff to manage, less chance of collisions with other work (in either direction), and less pending review to worry about. So long as something represents forward progress I tend to work on the basis of why not rather than why.

Quoted 8 lines
Quoted 4 lines

A reason is that once we add the support for disabling kpti based on the E0PD feature, this patch may turn out to be slightly different (for example, you may add a common has_e0pd() check that is called from both unmap_kernel_at_el0() and the E0PD arm64_features[] entry). Given

Hrm, I don't really get that - incremental patches can always be done (indeed they're often really helpful for people trying to understand how the code got to be the way it is).

Sorry, I wasn't clear enough. I'm perfectly fine to incremental patches and encourage them as part of a series achieving an end goal. Here the end goal (and the reason behind this architecture feature) is to be able to safely disable KPTI in favour of a hardware-based mitigation. Your one-patch series does not achieve this yet, so I suggest we don't merge this until we have the other part of the series.

Signature · 6 lines
Catalin

_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel