arm64: errata: Speculative SSBS workaround
cover arm64: errata: Speculative SSBS workaround
Hi,
This series is a v6.6-only backport (based on v6.6.44) of the upstream workaround for SSBS errata on Arm Ltd CPUs, as affected parts are likely to be used with stable kernels. This does not apply to earlier stable trees, which will receive a separate backport.
The errata mean that an MSR to the SSBS special-purpose register does not affect subsequent speculative instructions, permitting speculative store bypassing for a window of time.
The upstream support was original posted as:
* https://lore.kernel.org/linux-arm-kernel/[email protected]/
"arm64: errata: Add workaround for Arm errata 3194386 and 3312417" Present in v6.10
* https://lore.kernel.org/linux-arm-kernel/[email protected]/
"arm64: errata: Expand speculative SSBS workaround"
Present in v6.11-rc1
* https://lore.kernel.org/linux-arm-kernel/[email protected]/
"arm64: errata: Expand speculative SSBS workaround (again)"
Present in v6.11-rc2
This backport applies the patches which are not present in v6.6.y, and as prerequisites backports the addition of the Neoverse-V2 MIDR values and the restoration of the spec_bar() macro.
I have tested the backport (when applied to v6.6.44), ensuring that the detection logic works and that the HWCAP and string in /proc/cpuinfo are both hidden when the relevant errata are detected.
Mark.Besar Wicaksono (1):
arm64: Add Neoverse-V2 part
Mark Rutland (12):
arm64: barrier: Restore spec_bar() macro
arm64: cputype: Add Cortex-X4 definitions
arm64: cputype: Add Neoverse-V3 definitions
arm64: errata: Add workaround for Arm errata 3194386 and 3312417
arm64: cputype: Add Cortex-X3 definitions
arm64: cputype: Add Cortex-A720 definitions
arm64: cputype: Add Cortex-X925 definitions
arm64: errata: Unify speculative SSBS errata logic
arm64: errata: Expand speculative SSBS workaround
arm64: cputype: Add Cortex-X1C definitions
arm64: cputype: Add Cortex-A725 definitions
arm64: errata: Expand speculative SSBS workaround (again)
Documentation/arch/arm64/silicon-errata.rst | 36 +++++++++++++++++++ arch/arm64/Kconfig | 38 +++++++++++++++++++++ arch/arm64/include/asm/barrier.h | 4 +++ arch/arm64/include/asm/cputype.h | 16 +++++++++ arch/arm64/kernel/cpu_errata.c | 31 +++++++++++++++++ arch/arm64/kernel/cpufeature.c | 12 +++++++ arch/arm64/kernel/proton-pack.c | 12 +++++++ arch/arm64/tools/cpucaps | 1 + 8 files changed, 150 insertions(+)
Signature · 1 line
2.30.2
PATCH 1/13 arm64: Add Neoverse-V2 part
From: Besar Wicaksono <[email protected]>
[ Upstream commit f4d9d9dcc70b96b5e5d7801bd5fbf8491b07b13d ]
Add the part number and MIDR for Neoverse-V2
--- arch/arm64/include/asm/cputype.h | 2 ++ 1 file changed, 2 insertions(+)
arch/arm64/include/asm/cputype.h +2 -0
@@ -86,6 +86,7 @@#define ARM_CPU_PART_CORTEX_X2 0xD48#define ARM_CPU_PART_NEOVERSE_N2 0xD49#define ARM_CPU_PART_CORTEX_A78C 0xD4B +#define ARM_CPU_PART_NEOVERSE_V2 0xD4F#define APM_CPU_PART_XGENE 0x000#define APM_CPU_VAR_POTENZA 0x00
@@ -159,6 +160,7 @@#define MIDR_CORTEX_X2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X2)#define MIDR_NEOVERSE_N2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_N2)#define MIDR_CORTEX_A78C MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A78C) +#define MIDR_NEOVERSE_V2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V2)#define MIDR_THUNDERX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX)#define MIDR_THUNDERX_81XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_81XX)#define MIDR_THUNDERX_83XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_83XX)
Signature · 1 line
2.30.2
PATCH 2/13 arm64: barrier: Restore spec_bar() macro
In reply to Mark Rutland[ Upstream commit ebfc726eae3f31bdb5fae1bbd74ef235d71046ca ]
Upcoming errata workarounds will need to use SB from C code. Restore the spec_bar() macro so that we can use SB.
This is effectively a revert of commit:
4f30ba1cce36d413 ("arm64: barrier: Remove spec_bar() macro")
--- arch/arm64/include/asm/barrier.h | 4 ++++ 1 file changed, 4 insertions(+)
arch/arm64/include/asm/barrier.h +4 -0
@@ -40,6 +40,10 @@*/ #define dgh () asm volatile("hint #6" : : : "memory") +#define spec_bar () asm volatile(ALTERNATIVE("dsb nsh\nisb\n", \ + SB_BARRIER_INSN"nop\n", \ + ARM64_HAS_SB)) +#ifdef CONFIG_ARM64_PSEUDO_NMI #define pmr_sync () \ do { \
Signature · 1 line
2.30.2
PATCH 3/13 arm64: cputype: Add Cortex-X4 definitions
In reply to Mark Rutland[ Upstream commit 02a0a04676fa7796d9cbc9eb5ca120aaa194d2dd ]
Add cputype definitions for Cortex-X4. These will be used for errata detection in subsequent patches.
These values can be found in Table B-249 ("MIDR_EL1 bit descriptions") in issue 0002-05 of the Cortex-X4 TRM, which can be found at:
https://developer.arm.com/documentation/102484/0002/?lang=en
--- arch/arm64/include/asm/cputype.h | 2 ++ 1 file changed, 2 insertions(+)
arch/arm64/include/asm/cputype.h +2 -0
@@ -87,6 +87,7 @@#define ARM_CPU_PART_NEOVERSE_N2 0xD49#define ARM_CPU_PART_CORTEX_A78C 0xD4B#define ARM_CPU_PART_NEOVERSE_V2 0xD4F +#define ARM_CPU_PART_CORTEX_X4 0xD82#define APM_CPU_PART_XGENE 0x000#define APM_CPU_VAR_POTENZA 0x00
@@ -161,6 +162,7 @@#define MIDR_NEOVERSE_N2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_N2)#define MIDR_CORTEX_A78C MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A78C)#define MIDR_NEOVERSE_V2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V2) +#define MIDR_CORTEX_X4 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X4)#define MIDR_THUNDERX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX)#define MIDR_THUNDERX_81XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_81XX)#define MIDR_THUNDERX_83XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_83XX)
Signature · 1 line
2.30.2
PATCH 4/13 arm64: cputype: Add Neoverse-V3 definitions
In reply to Mark Rutland[ Upstream commit 0ce85db6c2141b7ffb95709d76fc55a27ff3cdc1 ]
Add cputype definitions for Neoverse-V3. These will be used for errata detection in subsequent patches.
These values can be found in Table B-249 ("MIDR_EL1 bit descriptions") in issue 0001-04 of the Neoverse-V3 TRM, which can be found at:
https://developer.arm.com/documentation/107734/0001/?lang=en
--- arch/arm64/include/asm/cputype.h | 2 ++ 1 file changed, 2 insertions(+)
arch/arm64/include/asm/cputype.h +2 -0
@@ -88,6 +88,7 @@#define ARM_CPU_PART_CORTEX_A78C 0xD4B#define ARM_CPU_PART_NEOVERSE_V2 0xD4F#define ARM_CPU_PART_CORTEX_X4 0xD82 +#define ARM_CPU_PART_NEOVERSE_V3 0xD84#define APM_CPU_PART_XGENE 0x000#define APM_CPU_VAR_POTENZA 0x00
@@ -163,6 +164,7 @@#define MIDR_CORTEX_A78C MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A78C)#define MIDR_NEOVERSE_V2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V2)#define MIDR_CORTEX_X4 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X4) +#define MIDR_NEOVERSE_V3 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V3)#define MIDR_THUNDERX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX)#define MIDR_THUNDERX_81XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_81XX)#define MIDR_THUNDERX_83XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_83XX)
Signature · 1 line
2.30.2
PATCH 5/13 arm64: errata: Add workaround for Arm errata 3194386 and 3312417
In reply to Mark Rutland[ Upstream commit 7187bb7d0b5c7dfa18ca82e9e5c75e13861b1d88 ]
Cortex-X4 and Neoverse-V3 suffer from errata whereby an MSR to the SSBS special-purpose register does not affect subsequent speculative instructions, permitting speculative store bypassing for a window of time. This is described in their Software Developer Errata Notice (SDEN) documents:
* Cortex-X4 SDEN v8.0, erratum 3194386:
https://developer.arm.com/documentation/SDEN-2432808/0800/
* Neoverse-V3 SDEN v6.0, erratum 3312417:
https://developer.arm.com/documentation/SDEN-2891958/0600/
To workaround these errata, it is necessary to place a speculation barrier (SB) after MSR to the SSBS special-purpose register. This patch adds the requisite SB after writes to SSBS within the kernel, and hides the presence of SSBS from EL0 such that userspace software which cares about SSBS will manipulate this via prctl(PR_GET_SPECULATION_CTRL, ...).
--- Documentation/arch/arm64/silicon-errata.rst | 4 ++ arch/arm64/Kconfig | 42 +++++++++++++++++++++ arch/arm64/kernel/cpu_errata.c | 19 ++++++++++ arch/arm64/kernel/cpufeature.c | 12 ++++++ arch/arm64/kernel/proton-pack.c | 12 ++++++ arch/arm64/tools/cpucaps | 1 + 6 files changed, 90 insertions(+)
Documentation/arch/arm64/silicon-errata.rst +4 -0
@@ -133,6 +133,8 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-X2 | #2224489 | ARM64_ERRATUM_2224489 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-X4 | #3194386 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N1 | #1188873,1418040| ARM64_ERRATUM_1418040 | +----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N1 | #1349291 | N/A |
@@ -145,6 +147,8 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N2 | #2253138 | ARM64_ERRATUM_2253138 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Neoverse-V3 | #3312417 | ARM64_ERRATUM_3312417 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | MMU-500 | #841119,826419 | N/A | +----------------+-----------------+-----------------+-----------------------------+ | ARM | MMU-600 | #1076982,1209401| N/A |
arch/arm64/Kconfig +42 -0
48 lines not on this page · raw message
arch/arm64/kernel/cpu_errata.c +19 -0
@@ -448,6 +448,18 @@ static const struct midr_range erratum_spec_unpriv_load_list[] = { };#endif +#ifdef CONFIG_ARM64_WORKAROUND_SPECULATIVE_SSBS +static const struct midr_range erratum_spec_ssbs_list[] = { +#ifdef CONFIG_ARM64_ERRATUM_3194386 +MIDR_ALL_VERSIONS (MIDR_CORTEX_X4 ), +#endif +#ifdef CONFIG_ARM64_ERRATUM_3312417 +MIDR_ALL_VERSIONS (MIDR_NEOVERSE_V3 ), +#endif + {} +}; +#endif + const struct arm64_cpu_capabilities arm64_errata[] = {#ifdef CONFIG_ARM64_WORKAROUND_CLEAN_CACHE {
@@ -746,6 +758,13 @@ const struct arm64_cpu_capabilities arm64_errata[] = { .cpu_enable = cpu_clear_bf16_from_user_emulation, },#endif +#ifdef CONFIG_ARM64_WORKAROUND_SPECULATIVE_SSBS + { + .desc = "ARM errata 3194386, 3312417", + .capability = ARM64_WORKAROUND_SPECULATIVE_SSBS, + ERRATA_MIDR_RANGE_LIST(erratum_spec_ssbs_list), + }, +#endif #ifdef CONFIG_ARM64_WORKAROUND_SPECULATIVE_UNPRIV_LOAD { .desc = "ARM errata 2966298, 3117295",
arch/arm64/kernel/cpufeature.c +12 -0
@@ -2190,6 +2190,17 @@ static void cpu_enable_mte(struct arm64_cpu_capabilities const *cap) }#endif /* CONFIG_ARM64_MTE */ +static void user_feature_fixup (void ) +{ +if (cpus_have_cap (ARM64_WORKAROUND_SPECULATIVE_SSBS )) { +struct arm64_ftr_reg *regp; + + regp =get_arm64_ftr_reg (SYS_ID_AA64PFR1_EL1 ); +if (regp) + regp->user_mask &= ~ID_AA64PFR1_EL1_SSBS_MASK ; + } +} +static void elf_hwcap_fixup (void ) {#ifdef CONFIG_ARM64_ERRATUM_1742098
@@ -3345,6 +3356,7 @@ void __init setup_cpu_features(void)u32 cwg;setup_system_capabilities (); +user_feature_fixup ();setup_elf_hwcaps (arm64_elf_hwcaps);if (system_supports_32bit_el0 ()) {
arch/arm64/kernel/proton-pack.c +12 -0
@@ -558,6 +558,18 @@ static enum mitigation_state spectre_v4_enable_hw_mitigation(void)/* SCTLR_EL1.DSSBS was initialised to 0 during boot */ set_pstate_ssbs (0 ); + +/* +* SSBS is self-synchronizing and is intended to affect subsequent +* speculative instructions, but some CPUs can speculate with a stale +* value of SSBS. +* +* Mitigate this with an unconditional speculation barrier, as CPUs +* could mis-speculate branches and bypass a conditional barrier. +*/ +if (IS_ENABLED (CONFIG_ARM64_WORKAROUND_SPECULATIVE_SSBS )) +spec_bar (); +return SPECTRE_MITIGATED ; }
arch/arm64/tools/cpucaps +1 -0
@@ -99,4 +99,5 @@ WORKAROUND_NVIDIA_CARMEL_CNP WORKAROUND_QCOM_FALKOR_E1003 WORKAROUND_REPEAT_TLBI WORKAROUND_SPECULATIVE_AT +WORKAROUND_SPECULATIVE_SSBS WORKAROUND_SPECULATIVE_UNPRIV_LOAD
Signature · 1 line
2.30.2
PATCH 6/13 arm64: cputype: Add Cortex-X3 definitions
In reply to Mark Rutland[ Upstream commit be5a6f238700f38b534456608588723fba96c5ab ]
Add cputype definitions for Cortex-X3. These will be used for errata detection in subsequent patches.
These values can be found in Table A-263 ("MIDR_EL1 bit descriptions") in issue 07 of the Cortex-X3 TRM, which can be found at:
https://developer.arm.com/documentation/101593/0102/?lang=en
--- arch/arm64/include/asm/cputype.h | 2 ++ 1 file changed, 2 insertions(+)
arch/arm64/include/asm/cputype.h +2 -0
@@ -86,6 +86,7 @@#define ARM_CPU_PART_CORTEX_X2 0xD48#define ARM_CPU_PART_NEOVERSE_N2 0xD49#define ARM_CPU_PART_CORTEX_A78C 0xD4B +#define ARM_CPU_PART_CORTEX_X3 0xD4E#define ARM_CPU_PART_NEOVERSE_V2 0xD4F#define ARM_CPU_PART_CORTEX_X4 0xD82#define ARM_CPU_PART_NEOVERSE_V3 0xD84
@@ -162,6 +163,7 @@#define MIDR_CORTEX_X2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X2)#define MIDR_NEOVERSE_N2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_N2)#define MIDR_CORTEX_A78C MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A78C) +#define MIDR_CORTEX_X3 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X3)#define MIDR_NEOVERSE_V2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V2)#define MIDR_CORTEX_X4 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X4)#define MIDR_NEOVERSE_V3 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V3)
Signature · 1 line
2.30.2
PATCH 7/13 arm64: cputype: Add Cortex-A720 definitions
In reply to Mark Rutland[ Upstream commit add332c40328cf06fe35e4b3cde8ec315c4629e5 ]
Add cputype definitions for Cortex-A720. These will be used for errata detection in subsequent patches.
These values can be found in Table A-186 ("MIDR_EL1 bit descriptions") in issue 0002-05 of the Cortex-A720 TRM, which can be found at:
https://developer.arm.com/documentation/102530/0002/?lang=en
--- arch/arm64/include/asm/cputype.h | 2 ++ 1 file changed, 2 insertions(+)
arch/arm64/include/asm/cputype.h +2 -0
@@ -88,6 +88,7 @@#define ARM_CPU_PART_CORTEX_A78C 0xD4B#define ARM_CPU_PART_CORTEX_X3 0xD4E#define ARM_CPU_PART_NEOVERSE_V2 0xD4F +#define ARM_CPU_PART_CORTEX_A720 0xD81#define ARM_CPU_PART_CORTEX_X4 0xD82#define ARM_CPU_PART_NEOVERSE_V3 0xD84
@@ -165,6 +166,7 @@#define MIDR_CORTEX_A78C MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A78C)#define MIDR_CORTEX_X3 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X3)#define MIDR_NEOVERSE_V2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V2) +#define MIDR_CORTEX_A720 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A720)#define MIDR_CORTEX_X4 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X4)#define MIDR_NEOVERSE_V3 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V3)#define MIDR_THUNDERX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX)
Signature · 1 line
2.30.2
PATCH 8/13 arm64: cputype: Add Cortex-X925 definitions
In reply to Mark Rutland[ Upstream commit fd2ff5f0b320f418288e7a1f919f648fbc8a0dfc ]
Add cputype definitions for Cortex-X925. These will be used for errata detection in subsequent patches.
These values can be found in Table A-285 ("MIDR_EL1 bit descriptions") in issue 0001-05 of the Cortex-X925 TRM, which can be found at:
https://developer.arm.com/documentation/102807/0001/?lang=en
--- arch/arm64/include/asm/cputype.h | 2 ++ 1 file changed, 2 insertions(+)
arch/arm64/include/asm/cputype.h +2 -0
@@ -91,6 +91,7 @@#define ARM_CPU_PART_CORTEX_A720 0xD81#define ARM_CPU_PART_CORTEX_X4 0xD82#define ARM_CPU_PART_NEOVERSE_V3 0xD84 +#define ARM_CPU_PART_CORTEX_X925 0xD85#define APM_CPU_PART_XGENE 0x000#define APM_CPU_VAR_POTENZA 0x00
@@ -169,6 +170,7 @@#define MIDR_CORTEX_A720 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A720)#define MIDR_CORTEX_X4 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X4)#define MIDR_NEOVERSE_V3 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V3) +#define MIDR_CORTEX_X925 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X925)#define MIDR_THUNDERX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX)#define MIDR_THUNDERX_81XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_81XX)#define MIDR_THUNDERX_83XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_83XX)
Signature · 1 line
2.30.2
PATCH 9/13 arm64: errata: Unify speculative SSBS errata logic
In reply to Mark Rutland[ Upstream commit ec768766608092087dfb5c1fc45a16a6f524dee2 ]
Cortex-X4 erratum 3194386 and Neoverse-V3 erratum 3312417 are identical, with duplicate Kconfig text and some unsightly ifdeffery. While we try to share code behind CONFIG_ARM64_WORKAROUND_SPECULATIVE_SSBS, having separate options results in a fair amount of boilerplate code, and this will only get worse as we expand the set of affected CPUs.
To reduce this boilerplate, unify the two behind a common Kconfig option. This removes the duplicate text and Kconfig logic, and removes the need for the intermediate ARM64_WORKAROUND_SPECULATIVE_SSBS option. The set of affected CPUs is described as a list so that this can easily be extended.
I've used ARM64_ERRATUM_3194386 (matching the Neoverse-V3 erratum ID) as the common option, matching the way we use ARM64_ERRATUM_1319367 to cover Cortex-A57 erratum 1319537 and Cortex-A72 erratum 1319367.
--- Documentation/arch/arm64/silicon-errata.rst | 2 +- arch/arm64/Kconfig | 29 +++------------------ arch/arm64/kernel/cpu_errata.c | 8 ++---- arch/arm64/kernel/proton-pack.c | 2 +- 4 files changed, 8 insertions(+), 33 deletions(-)
Documentation/arch/arm64/silicon-errata.rst +1 -1
@@ -147,7 +147,7 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N2 | #2253138 | ARM64_ERRATUM_2253138 | +----------------+-----------------+-----------------+-----------------------------+ -| ARM | Neoverse-V3 | #3312417 | ARM64_ERRATUM_3312417 | +| ARM | Neoverse-V3 | #3312417 | ARM64_ERRATUM_3194386 | +----------------+-----------------+-----------------+-----------------------------+ | ARM | MMU-500 | #841119,826419 | N/A | +----------------+-----------------+-----------------+-----------------------------+
arch/arm64/Kconfig +4 -25
@@ -1068,34 +1068,14 @@ config ARM64_ERRATUM_3117295 If unsure, say Y. -config ARM64_WORKAROUND_SPECULATIVE_SSBS - bool - config ARM64_ERRATUM_3194386 - bool "Cortex-X4: 3194386: workaround for MSR SSBS not self-synchronizing" - select ARM64_WORKAROUND_SPECULATIVE_SSBS + bool "Cortex-X4/Neoverse-V3: workaround for MSR SSBS not self-synchronizing" default y help - This option adds the workaround for ARM Cortex-X4 erratum 3194386. + This option adds the workaround for the following errata: - On affected cores "MSR SSBS, #0" instructions may not affect - subsequent speculative instructions, which may permit unexepected - speculative store bypassing. - - Work around this problem by placing a speculation barrier after - kernel changes to SSBS. The presence of the SSBS special-purpose - register is hidden from hwcaps and EL0 reads of ID_AA64PFR1_EL1, such - that userspace will use the PR_SPEC_STORE_BYPASS prctl to change - SSBS. - - If unsure, say Y. - -config ARM64_ERRATUM_3312417 - bool "Neoverse-V3: 3312417: workaround for MSR SSBS not self-synchronizing" - select ARM64_WORKAROUND_SPECULATIVE_SSBS - default y - help - This option adds the workaround for ARM Neoverse-V3 erratum 3312417. + * ARM Cortex-X4 erratum 3194386 + * ARM Neoverse-V3 erratum 3312417 On affected cores "MSR SSBS, #0" instructions may not affect subsequent speculative instructions, which may permit unexepected
@@ -1109,7 +1089,6 @@ config ARM64_ERRATUM_3312417 If unsure, say Y. - config CAVIUM_ERRATUM_22375 bool "Cavium erratum 22375, 24313" default y
arch/arm64/kernel/cpu_errata.c +2 -6
@@ -448,14 +448,10 @@ static const struct midr_range erratum_spec_unpriv_load_list[] = { };#endif -#ifdef CONFIG_ARM64_WORKAROUND_SPECULATIVE_SSBS -static const struct midr_range erratum_spec_ssbs_list[] = {#ifdef CONFIG_ARM64_ERRATUM_3194386 +static const struct midr_range erratum_spec_ssbs_list[] = {MIDR_ALL_VERSIONS (MIDR_CORTEX_X4 ), -#endif -#ifdef CONFIG_ARM64_ERRATUM_3312417MIDR_ALL_VERSIONS (MIDR_NEOVERSE_V3 ), -#endif {} };#endif
@@ -758,7 +754,7 @@ const struct arm64_cpu_capabilities arm64_errata[] = { .cpu_enable = cpu_clear_bf16_from_user_emulation, },#endif -#ifdef CONFIG_ARM64_WORKAROUND_SPECULATIVE_SSBS +#ifdef CONFIG_ARM64_ERRATUM_3194386 { .desc = "ARM errata 3194386, 3312417", .capability = ARM64_WORKAROUND_SPECULATIVE_SSBS,
arch/arm64/kernel/proton-pack.c +1 -1
@@ -567,7 +567,7 @@ static enum mitigation_state spectre_v4_enable_hw_mitigation(void)* Mitigate this with an unconditional speculation barrier, as CPUs * could mis-speculate branches and bypass a conditional barrier. */ -if (IS_ENABLED (CONFIG_ARM64_WORKAROUND_SPECULATIVE_SSBS )) +if (IS_ENABLED (CONFIG_ARM64_ERRATUM_3194386 ))spec_bar ();return SPECTRE_MITIGATED ;
Signature · 1 line
2.30.2
PATCH 10/13 arm64: errata: Expand speculative SSBS workaround
In reply to Mark Rutland[ Upstream commit 75b3c43eab594bfbd8184ec8ee1a6b820950819a ]
A number of Arm Ltd CPUs suffer from errata whereby an MSR to the SSBS special-purpose register does not affect subsequent speculative instructions, permitting speculative store bypassing for a window of time.
We worked around this for Cortex-X4 and Neoverse-V3, in commit:
7187bb7d0b5c7dfa ("arm64: errata: Add workaround for Arm errata 3194386 and 3312417")
... as per their Software Developer Errata Notice (SDEN) documents:
* Cortex-X4 SDEN v8.0, erratum 3194386:
https://developer.arm.com/documentation/SDEN-2432808/0800/
* Neoverse-V3 SDEN v6.0, erratum 3312417:
https://developer.arm.com/documentation/SDEN-2891958/0600/
Since then, similar errata have been published for a number of other Arm Ltd CPUs, for which the mitigation is the same. This is described in their respective SDEN documents:
* Cortex-A710 SDEN v19.0, errataum 3324338
https://developer.arm.com/documentation/SDEN-1775101/1900/?lang=en
* Cortex-A720 SDEN v11.0, erratum 3456091
https://developer.arm.com/documentation/SDEN-2439421/1100/?lang=en
* Cortex-X2 SDEN v19.0, erratum 3324338
https://developer.arm.com/documentation/SDEN-1775100/1900/?lang=en
* Cortex-X3 SDEN v14.0, erratum 3324335
https://developer.arm.com/documentation/SDEN-2055130/1400/?lang=en
* Cortex-X925 SDEN v8.0, erratum 3324334
https://developer.arm.com/documentation/109108/800/?lang=en
* Neoverse-N2 SDEN v17.0, erratum 3324339
https://developer.arm.com/documentation/SDEN-1982442/1700/?lang=en
* Neoverse-V2 SDEN v9.0, erratum 3324336
https://developer.arm.com/documentation/SDEN-2332927/900/?lang=en
Note that due to shared design lineage, some CPUs share the same erratum number.
Add these to the existing mitigation under CONFIG_ARM64_ERRATUM_3194386. As listing all of the erratum IDs in the runtime description would be unwieldy, this is reduced to:
"SSBS not fully self-synchronizing"
... matching the description of the errata in all of the SDENs.
--- Documentation/arch/arm64/silicon-errata.rst | 14 ++++++++++++++ arch/arm64/Kconfig | 9 ++++++++- arch/arm64/kernel/cpu_errata.c | 9 ++++++++- 3 files changed, 30 insertions(+), 2 deletions(-)
Documentation/arch/arm64/silicon-errata.rst +14 -0
@@ -127,14 +127,24 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-A710 | #2224489 | ARM64_ERRATUM_2224489 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A710 | #3324338 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-A715 | #2645198 | ARM64_ERRATUM_2645198 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A720 | #3456091 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-X2 | #2119858 | ARM64_ERRATUM_2119858 | +----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-X2 | #2224489 | ARM64_ERRATUM_2224489 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-X2 | #3324338 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-X3 | #3324335 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-X4 | #3194386 | ARM64_ERRATUM_3194386 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-X925 | #3324334 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N1 | #1188873,1418040| ARM64_ERRATUM_1418040 | +----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N1 | #1349291 | N/A |
@@ -147,6 +157,10 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N2 | #2253138 | ARM64_ERRATUM_2253138 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Neoverse-N2 | #3324339 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Neoverse-V2 | #3324336 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-V3 | #3312417 | ARM64_ERRATUM_3194386 | +----------------+-----------------+-----------------+-----------------------------+ | ARM | MMU-500 | #841119,826419 | N/A |
arch/arm64/Kconfig +8 -1
@@ -1069,12 +1069,19 @@ config ARM64_ERRATUM_3117295 If unsure, say Y. config ARM64_ERRATUM_3194386 - bool "Cortex-X4/Neoverse-V3: workaround for MSR SSBS not self-synchronizing" + bool "Cortex-{A720,X4,X925}/Neoverse-V3: workaround for MSR SSBS not self-synchronizing" default y help This option adds the workaround for the following errata: + * ARM Cortex-A710 erratam 3324338 + * ARM Cortex-A720 erratum 3456091 + * ARM Cortex-X2 erratum 3324338 + * ARM Cortex-X3 erratum 3324335 * ARM Cortex-X4 erratum 3194386 + * ARM Cortex-X925 erratum 3324334 + * ARM Neoverse N2 erratum 3324339 + * ARM Neoverse V2 erratum 3324336 * ARM Neoverse-V3 erratum 3312417 On affected cores "MSR SSBS, #0" instructions may not affect
arch/arm64/kernel/cpu_errata.c +8 -1
@@ -450,8 +450,15 @@ static const struct midr_range erratum_spec_unpriv_load_list[] = {#ifdef CONFIG_ARM64_ERRATUM_3194386 static const struct midr_range erratum_spec_ssbs_list[] = { +MIDR_ALL_VERSIONS (MIDR_CORTEX_A710 ), +MIDR_ALL_VERSIONS (MIDR_CORTEX_A720 ), +MIDR_ALL_VERSIONS (MIDR_CORTEX_X2 ), +MIDR_ALL_VERSIONS (MIDR_CORTEX_X3 ),MIDR_ALL_VERSIONS (MIDR_CORTEX_X4 ), +MIDR_ALL_VERSIONS (MIDR_CORTEX_X925 ), +MIDR_ALL_VERSIONS (MIDR_NEOVERSE_N2 ),MIDR_ALL_VERSIONS (MIDR_NEOVERSE_V3 ), +MIDR_ALL_VERSIONS (MIDR_NEOVERSE_V2 ), {} };#endif
@@ -756,7 +763,7 @@ const struct arm64_cpu_capabilities arm64_errata[] = {#endif #ifdef CONFIG_ARM64_ERRATUM_3194386 { - .desc = "ARM errata 3194386, 3312417", + .desc = "SSBS not fully self-synchronizing", .capability = ARM64_WORKAROUND_SPECULATIVE_SSBS, ERRATA_MIDR_RANGE_LIST(erratum_spec_ssbs_list), },
Signature · 1 line
2.30.2
PATCH 11/13 arm64: cputype: Add Cortex-X1C definitions
In reply to Mark Rutland[ Upstream commit 58d245e03c324d083a0ec3b9ab8ebd46ec9848d7 ]
Add cputype definitions for Cortex-X1C. These will be used for errata detection in subsequent patches.
These values can be found in the Cortex-X1C TRM:
https://developer.arm.com/documentation/101968/0002/
... in section B2.107 ("MIDR_EL1, Main ID Register, EL1").
--- arch/arm64/include/asm/cputype.h | 2 ++ 1 file changed, 2 insertions(+)
arch/arm64/include/asm/cputype.h +2 -0
@@ -86,6 +86,7 @@#define ARM_CPU_PART_CORTEX_X2 0xD48#define ARM_CPU_PART_NEOVERSE_N2 0xD49#define ARM_CPU_PART_CORTEX_A78C 0xD4B +#define ARM_CPU_PART_CORTEX_X1C 0xD4C#define ARM_CPU_PART_CORTEX_X3 0xD4E#define ARM_CPU_PART_NEOVERSE_V2 0xD4F#define ARM_CPU_PART_CORTEX_A720 0xD81
@@ -165,6 +166,7 @@#define MIDR_CORTEX_X2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X2)#define MIDR_NEOVERSE_N2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_N2)#define MIDR_CORTEX_A78C MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A78C) +#define MIDR_CORTEX_X1C MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X1C)#define MIDR_CORTEX_X3 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X3)#define MIDR_NEOVERSE_V2 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V2)#define MIDR_CORTEX_A720 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A720)
Signature · 1 line
2.30.2
PATCH 12/13 arm64: cputype: Add Cortex-A725 definitions
In reply to Mark Rutland[ Upstream commit 9ef54a384526911095db465e77acc1cb5266b32c ]
Add cputype definitions for Cortex-A725. These will be used for errata detection in subsequent patches.
These values can be found in the Cortex-A725 TRM:
https://developer.arm.com/documentation/107652/0001/
... in table A-247 ("MIDR_EL1 bit descriptions").
--- arch/arm64/include/asm/cputype.h | 2 ++ 1 file changed, 2 insertions(+)
arch/arm64/include/asm/cputype.h +2 -0
@@ -93,6 +93,7 @@#define ARM_CPU_PART_CORTEX_X4 0xD82#define ARM_CPU_PART_NEOVERSE_V3 0xD84#define ARM_CPU_PART_CORTEX_X925 0xD85 +#define ARM_CPU_PART_CORTEX_A725 0xD87#define APM_CPU_PART_XGENE 0x000#define APM_CPU_VAR_POTENZA 0x00
@@ -173,6 +174,7 @@#define MIDR_CORTEX_X4 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X4)#define MIDR_NEOVERSE_V3 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V3)#define MIDR_CORTEX_X925 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X925) +#define MIDR_CORTEX_A725 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A725)#define MIDR_THUNDERX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX)#define MIDR_THUNDERX_81XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_81XX)#define MIDR_THUNDERX_83XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_83XX)
Signature · 1 line
2.30.2
PATCH 13/13 arm64: errata: Expand speculative SSBS workaround (again)
In reply to Mark Rutland[ Upstream commit b0672bbe133ebb6f7be21fce1d742d52f25bcdc7 ]
A number of Arm Ltd CPUs suffer from errata whereby an MSR to the SSBS special-purpose register does not affect subsequent speculative instructions, permitting speculative store bypassing for a window of time.
We worked around this for a number of CPUs in commits:
* 7187bb7d0b5c7dfa ("arm64: errata: Add workaround for Arm errata 3194386 and 3312417")
* 75b3c43eab594bfb ("arm64: errata: Expand speculative SSBS workaround")
Since then, similar errata have been published for a number of other Arm Ltd CPUs, for which the same mitigation is sufficient. This is described in their respective Software Developer Errata Notice (SDEN) documents:
* Cortex-A76 (MP052) SDEN v31.0, erratum 3324349
https://developer.arm.com/documentation/SDEN-885749/3100/
* Cortex-A77 (MP074) SDEN v19.0, erratum 3324348
https://developer.arm.com/documentation/SDEN-1152370/1900/
* Cortex-A78 (MP102) SDEN v21.0, erratum 3324344
https://developer.arm.com/documentation/SDEN-1401784/2100/
* Cortex-A78C (MP138) SDEN v16.0, erratum 3324346
https://developer.arm.com/documentation/SDEN-1707916/1600/
* Cortex-A78C (MP154) SDEN v10.0, erratum 3324347
https://developer.arm.com/documentation/SDEN-2004089/1000/
* Cortex-A725 (MP190) SDEN v5.0, erratum 3456106
https://developer.arm.com/documentation/SDEN-2832921/0500/
* Cortex-X1 (MP077) SDEN v21.0, erratum 3324344
https://developer.arm.com/documentation/SDEN-1401782/2100/
* Cortex-X1C (MP136) SDEN v16.0, erratum 3324346
https://developer.arm.com/documentation/SDEN-1707914/1600/
* Neoverse-N1 (MP050) SDEN v32.0, erratum 3324349
https://developer.arm.com/documentation/SDEN-885747/3200/
* Neoverse-V1 (MP076) SDEN v19.0, erratum 3324341
https://developer.arm.com/documentation/SDEN-1401781/1900/
Note that due to the manner in which Arm develops IP and tracks errata, some CPUs share a common erratum number and some CPUs have multiple erratum numbers for the same HW issue.
On parts without SB, it is necessary to use ISB for the workaround. The spec_bar() macro used in the mitigation will expand to a "DSB SY; ISB" sequence in this case, which is sufficient on all affected parts.
Enable the existing mitigation by adding the relevant MIDRs to erratum_spec_ssbs_list. The list is sorted alphanumerically (involving moving Neoverse-V3 after Neoverse-V2) so that this is easy to audit and potentially extend again in future. The Kconfig text is also updated to clarify the set of affected parts and the mitigation.
--- Documentation/arch/arm64/silicon-errata.rst | 18 +++++++++++++++++ arch/arm64/Kconfig | 22 +++++++++++++++------ arch/arm64/kernel/cpu_errata.c | 11 ++++++++++- 3 files changed, 44 insertions(+), 7 deletions(-)
Documentation/arch/arm64/silicon-errata.rst +18 -0
@@ -119,8 +119,16 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-A76 | #1463225 | ARM64_ERRATUM_1463225 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A76 | #3324349 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-A77 | #1508412 | ARM64_ERRATUM_1508412 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A77 | #3324348 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A78 | #3324344 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A78C | #3324346,3324347| ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-A710 | #2119858 | ARM64_ERRATUM_2119858 | +----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-A710 | #2054223 | ARM64_ERRATUM_2054223 |
@@ -133,6 +141,12 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-A720 | #3456091 | ARM64_ERRATUM_3194386 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A725 | #3456106 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-X1 | #3324344 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-X1C | #3324346 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-X2 | #2119858 | ARM64_ERRATUM_2119858 | +----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-X2 | #2224489 | ARM64_ERRATUM_2224489 |
@@ -151,6 +165,8 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N1 | #1542419 | ARM64_ERRATUM_1542419 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Neoverse-N1 | #3324349 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N2 | #2139208 | ARM64_ERRATUM_2139208 | +----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N2 | #2067961 | ARM64_ERRATUM_2067961 |
@@ -159,6 +175,8 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N2 | #3324339 | ARM64_ERRATUM_3194386 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Neoverse-V1 | #3324341 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-V2 | #3324336 | ARM64_ERRATUM_3194386 | +----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-V3 | #3312417 | ARM64_ERRATUM_3194386 |
arch/arm64/Kconfig +16 -6
@@ -1069,18 +1069,28 @@ config ARM64_ERRATUM_3117295 If unsure, say Y. config ARM64_ERRATUM_3194386 - bool "Cortex-{A720,X4,X925}/Neoverse-V3: workaround for MSR SSBS not self-synchronizing" + bool "Cortex-*/Neoverse-*: workaround for MSR SSBS not self-synchronizing" default y help This option adds the workaround for the following errata: + * ARM Cortex-A76 erratum 3324349 + * ARM Cortex-A77 erratum 3324348 + * ARM Cortex-A78 erratum 3324344 + * ARM Cortex-A78C erratum 3324346 + * ARM Cortex-A78C erratum 3324347 * ARM Cortex-A710 erratam 3324338 * ARM Cortex-A720 erratum 3456091 + * ARM Cortex-A725 erratum 3456106 + * ARM Cortex-X1 erratum 3324344 + * ARM Cortex-X1C erratum 3324346 * ARM Cortex-X2 erratum 3324338 * ARM Cortex-X3 erratum 3324335 * ARM Cortex-X4 erratum 3194386 * ARM Cortex-X925 erratum 3324334 + * ARM Neoverse-N1 erratum 3324349 * ARM Neoverse N2 erratum 3324339 + * ARM Neoverse-V1 erratum 3324341 * ARM Neoverse V2 erratum 3324336 * ARM Neoverse-V3 erratum 3312417
@@ -1088,11 +1098,11 @@ config ARM64_ERRATUM_3194386 subsequent speculative instructions, which may permit unexepected speculative store bypassing. - Work around this problem by placing a speculation barrier after - kernel changes to SSBS. The presence of the SSBS special-purpose - register is hidden from hwcaps and EL0 reads of ID_AA64PFR1_EL1, such - that userspace will use the PR_SPEC_STORE_BYPASS prctl to change - SSBS. + Work around this problem by placing a Speculation Barrier (SB) or + Instruction Synchronization Barrier (ISB) after kernel changes to + SSBS. The presence of the SSBS special-purpose register is hidden + from hwcaps and EL0 reads of ID_AA64PFR1_EL1, such that userspace + will use the PR_SPEC_STORE_BYPASS prctl to change SSBS. If unsure, say Y.
arch/arm64/kernel/cpu_errata.c +10 -1
@@ -450,15 +450,24 @@ static const struct midr_range erratum_spec_unpriv_load_list[] = {#ifdef CONFIG_ARM64_ERRATUM_3194386 static const struct midr_range erratum_spec_ssbs_list[] = { +MIDR_ALL_VERSIONS (MIDR_CORTEX_A76 ), +MIDR_ALL_VERSIONS (MIDR_CORTEX_A77 ), +MIDR_ALL_VERSIONS (MIDR_CORTEX_A78 ), +MIDR_ALL_VERSIONS (MIDR_CORTEX_A78C ),MIDR_ALL_VERSIONS (MIDR_CORTEX_A710 ),MIDR_ALL_VERSIONS (MIDR_CORTEX_A720 ), +MIDR_ALL_VERSIONS (MIDR_CORTEX_A725 ), +MIDR_ALL_VERSIONS (MIDR_CORTEX_X1 ), +MIDR_ALL_VERSIONS (MIDR_CORTEX_X1C ),MIDR_ALL_VERSIONS (MIDR_CORTEX_X2 ),MIDR_ALL_VERSIONS (MIDR_CORTEX_X3 ),MIDR_ALL_VERSIONS (MIDR_CORTEX_X4 ),MIDR_ALL_VERSIONS (MIDR_CORTEX_X925 ), +MIDR_ALL_VERSIONS (MIDR_NEOVERSE_N1 ),MIDR_ALL_VERSIONS (MIDR_NEOVERSE_N2 ), -MIDR_ALL_VERSIONS (MIDR_NEOVERSE_V3 ), +MIDR_ALL_VERSIONS (MIDR_NEOVERSE_V1 ),MIDR_ALL_VERSIONS (MIDR_NEOVERSE_V2 ), +MIDR_ALL_VERSIONS (MIDR_NEOVERSE_V3 ), {} };#endif
Signature · 1 line
2.30.2
Greg KH
maintainerRe: [PATCH 6.6.y 13/13] arm64: errata: Expand speculative SSBS workaround (again)
Quoted 1 line · Mark Rutland
[ Upstream commit
b0672bbe133ebb6f7be21fce1d742d52f25bcdc7]
Now I figured it out, this is the wrong git id, and is not in any tree anywhere. It should be adeec61a4723fd3e39da68db4cc4d924e6d7f641.
I'll go hand-edit it now...
greg k-h
Re: [PATCH 6.6.y 13/13] arm64: errata: Expand speculative SSBS workaround (again)
Quoted 7 lines · Greg KH
Quoted 1 line · Mark Rutland
[ Upstream commit
b0672bbe133ebb6f7be21fce1d742d52f25bcdc7]Now I figured it out, this is the wrong git id, and is not in any tree anywhere. It should be
adeec61a4723fd3e39da68db4cc4d924e6d7f641.I'll go hand-edit it now...
Thanks, and sorry about that.
Looking in my local tree, the incorrect commit id is a copy of adeec61a4723fd3e39da68db4cc4d924e6d7f641 where the earlier commits had been updated with "[ Upstream commit XXX ]" wording. Evidently I grabbed the commit ID from the wrong branch when I was folding that in.
I'll try to make sure that doesn't happen again. Is there any public tooling for sanity-checking this?
Mark.
Greg KH
maintainerRe: [PATCH 6.6.y 13/13] arm64: errata: Expand speculative SSBS workaround (again)
Quoted 11 lines · Mark Rutland
Quoted 7 lines · Greg KH
Quoted 1 line · Mark Rutland
[ Upstream commit
b0672bbe133ebb6f7be21fce1d742d52f25bcdc7]Now I figured it out, this is the wrong git id, and is not in any tree anywhere. It should be
adeec61a4723fd3e39da68db4cc4d924e6d7f641.I'll go hand-edit it now...
Thanks, and sorry about that.
Quoted 7 lines
Looking in my local tree, the incorrect commit id is a copy of
adeec61a4723fd3e39da68db4cc4d924e6d7f641where the earlier commits had been updated with "[ Upstream commit XXX ]" wording. Evidently I grabbed the commit ID from the wrong branch when I was folding that in.I'll try to make sure that doesn't happen again. Is there any public tooling for sanity-checking this?
Not that I know of :(