arm64: Workaround for Cortex-A55 erratum 1530923

Cortex-A55 erratum 1530923 allows TLB entries to be allocated as a
result of a speculative AT instruction. This may happen in the middle of
a guest world switch while the relevant VMSA configuration is in an
inconsistent state, leading to erroneous content being allocated into
TLBs.

The same workaround as is used for Cortex-A76 erratum 1165522
(WORKAROUND_SPECULATIVE_AT_VHE) can be used here. Note that this
mandates the use of VHE on affected parts.

Acked-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Steven Price <steven.price@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>
This commit is contained in:
Steven Price 2019-12-16 11:56:31 +00:00 committed by Will Deacon
parent db0d46a58d
commit 275fa0ea2c
6 changed files with 26 additions and 7 deletions

View file

@ -88,6 +88,8 @@ stable kernels.
+----------------+-----------------+-----------------+-----------------------------+
| ARM | Cortex-A76 | #1463225 | ARM64_ERRATUM_1463225 |
+----------------+-----------------+-----------------+-----------------------------+
| ARM | Cortex-A55 | #1530923 | ARM64_ERRATUM_1530923 |
+----------------+-----------------+-----------------+-----------------------------+
| ARM | Neoverse-N1 | #1188873,1418040| ARM64_ERRATUM_1418040 |
+----------------+-----------------+-----------------+-----------------------------+
| ARM | Neoverse-N1 | #1349291 | N/A |

View file

@ -530,6 +530,19 @@ config ARM64_ERRATUM_1165522
If unsure, say Y.
config ARM64_ERRATUM_1530923
bool "Cortex-A55: Speculative AT instruction using out-of-context translation regime could cause subsequent request to generate an incorrect translation"
default y
select ARM64_WORKAROUND_SPECULATIVE_AT_VHE
help
This option adds a workaround for ARM Cortex-A55 erratum 1530923.
Affected Cortex-A55 cores (r0p0, r0p1, r1p0, r2p0) could end-up with
corrupted TLBs by speculating an AT instruction during a guest
context switch.
If unsure, say Y.
config ARM64_ERRATUM_1286807
bool "Cortex-A76: Modification of the translation table for a virtual address might lead to read-after-read ordering violation"
default y

View file

@ -91,8 +91,8 @@ static __always_inline void __hyp_text __load_guest_stage2(struct kvm *kvm)
write_sysreg(kvm_get_vttbr(kvm), vttbr_el2);
/*
* ARM erratum 1165522 requires the actual execution of the above
* before we can switch to the EL1/EL0 translation regime used by
* ARM errata 1165522 and 1530923 require the actual execution of the
* above before we can switch to the EL1/EL0 translation regime used by
* the guest.
*/
asm(ALTERNATIVE("nop", "isb", ARM64_WORKAROUND_SPECULATIVE_AT_VHE));

View file

@ -762,6 +762,10 @@ static const struct midr_range erratum_speculative_at_vhe_list[] = {
#ifdef CONFIG_ARM64_ERRATUM_1165522
/* Cortex A76 r0p0 to r2p0 */
MIDR_RANGE(MIDR_CORTEX_A76, 0, 0, 2, 0),
#endif
#ifdef CONFIG_ARM64_ERRATUM_1530923
/* Cortex A55 r0p0 to r2p0 */
MIDR_RANGE(MIDR_CORTEX_A55, 0, 0, 2, 0),
#endif
{},
};
@ -895,7 +899,7 @@ const struct arm64_cpu_capabilities arm64_errata[] = {
#endif
#ifdef CONFIG_ARM64_WORKAROUND_SPECULATIVE_AT_VHE
{
.desc = "ARM erratum 1165522",
.desc = "ARM errata 1165522, 1530923",
.capability = ARM64_WORKAROUND_SPECULATIVE_AT_VHE,
ERRATA_MIDR_RANGE_LIST(erratum_speculative_at_vhe_list),
},

View file

@ -158,8 +158,8 @@ static void deactivate_traps_vhe(void)
write_sysreg(HCR_HOST_VHE_FLAGS, hcr_el2);
/*
* ARM erratum 1165522 requires the actual execution of the above
* before we can switch to the EL2/EL0 translation regime used by
* ARM errata 1165522 and 1530923 require the actual execution of the
* above before we can switch to the EL2/EL0 translation regime used by
* the host.
*/
asm(ALTERNATIVE("nop", "isb", ARM64_WORKAROUND_SPECULATIVE_AT_VHE));

View file

@ -25,8 +25,8 @@ static void __hyp_text __tlb_switch_to_guest_vhe(struct kvm *kvm,
if (cpus_have_const_cap(ARM64_WORKAROUND_SPECULATIVE_AT_VHE)) {
/*
* For CPUs that are affected by ARM erratum 1165522, we
* cannot trust stage-1 to be in a correct state at that
* For CPUs that are affected by ARM errata 1165522 or 1530923,
* we cannot trust stage-1 to be in a correct state at that
* point. Since we do not want to force a full load of the
* vcpu state, we prevent the EL1 page-table walker to
* allocate new TLBs. This is done by setting the EPD bits