[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PULL 19/27] arm: Implement HFNMIENA support for M profile
From: |
Peter Maydell |
Subject: |
[Qemu-devel] [PULL 19/27] arm: Implement HFNMIENA support for M profile MPU |
Date: |
Thu, 1 Jun 2017 18:10:27 +0100 |
Implement HFNMIENA support for the M profile MPU. This bit controls
whether the MPU is treated as enabled when executing at execution
priorities of less than zero (in NMI, HardFault or with the FAULTMASK
bit set).
Doing this requires us to use a different MMU index for "running
at execution priority < 0", because we will have different
access permissions for that case versus the normal case.
Signed-off-by: Peter Maydell <address@hidden>
Message-id: address@hidden
---
target/arm/cpu.h | 24 +++++++++++++++++++++++-
target/arm/helper.c | 18 +++++++++++++++++-
target/arm/translate.c | 1 +
3 files changed, 41 insertions(+), 2 deletions(-)
diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index 5c46c48..13da503 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -2045,6 +2045,18 @@ static inline bool arm_excp_unmasked(CPUState *cs,
unsigned int excp_idx,
* for the accesses done as part of a stage 1 page table walk, rather than
* having to walk the stage 2 page table over and over.)
*
+ * R profile CPUs have an MPU, but can use the same set of MMU indexes
+ * as A profile. They only need to distinguish NS EL0 and NS EL1 (and
+ * NS EL2 if we ever model a Cortex-R52).
+ *
+ * M profile CPUs are rather different as they do not have a true MMU.
+ * They have the following different MMU indexes:
+ * User
+ * Privileged
+ * Execution priority negative (this is like privileged, but the
+ * MPU HFNMIENA bit means that it may have different access permission
+ * check results to normal privileged code, so can't share a TLB).
+ *
* The ARMMMUIdx and the mmu index value used by the core QEMU TLB code
* are not quite the same -- different CPU types (most notably M profile
* vs A/R profile) would like to use MMU indexes with different semantics,
@@ -2080,6 +2092,7 @@ typedef enum ARMMMUIdx {
ARMMMUIdx_S2NS = 6 | ARM_MMU_IDX_A,
ARMMMUIdx_MUser = 0 | ARM_MMU_IDX_M,
ARMMMUIdx_MPriv = 1 | ARM_MMU_IDX_M,
+ ARMMMUIdx_MNegPri = 2 | ARM_MMU_IDX_M,
/* Indexes below here don't have TLBs and are used only for AT system
* instructions or for the first stage of an S12 page table walk.
*/
@@ -2100,6 +2113,7 @@ typedef enum ARMMMUIdxBit {
ARMMMUIdxBit_S2NS = 1 << 6,
ARMMMUIdxBit_MUser = 1 << 0,
ARMMMUIdxBit_MPriv = 1 << 1,
+ ARMMMUIdxBit_MNegPri = 1 << 2,
} ARMMMUIdxBit;
#define MMU_USER_IDX 0
@@ -2125,7 +2139,7 @@ static inline int arm_mmu_idx_to_el(ARMMMUIdx mmu_idx)
case ARM_MMU_IDX_A:
return mmu_idx & 3;
case ARM_MMU_IDX_M:
- return mmu_idx & 1;
+ return mmu_idx == ARMMMUIdx_MUser ? 0 : 1;
default:
g_assert_not_reached();
}
@@ -2139,6 +2153,14 @@ static inline int cpu_mmu_index(CPUARMState *env, bool
ifetch)
if (arm_feature(env, ARM_FEATURE_M)) {
ARMMMUIdx mmu_idx = el == 0 ? ARMMMUIdx_MUser : ARMMMUIdx_MPriv;
+ /* Execution priority is negative if FAULTMASK is set or
+ * we're in a HardFault or NMI handler.
+ */
+ if ((env->v7m.exception > 0 && env->v7m.exception <= 3)
+ || env->daif & PSTATE_F) {
+ return arm_to_core_mmu_idx(ARMMMUIdx_MNegPri);
+ }
+
return arm_to_core_mmu_idx(mmu_idx);
}
diff --git a/target/arm/helper.c b/target/arm/helper.c
index 674b52d..2594faa 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -7037,6 +7037,7 @@ static inline uint32_t regime_el(CPUARMState *env,
ARMMMUIdx mmu_idx)
case ARMMMUIdx_S1NSE0:
case ARMMMUIdx_S1NSE1:
case ARMMMUIdx_MPriv:
+ case ARMMMUIdx_MNegPri:
case ARMMMUIdx_MUser:
return 1;
default:
@@ -7055,6 +7056,7 @@ static inline bool regime_is_secure(CPUARMState *env,
ARMMMUIdx mmu_idx)
case ARMMMUIdx_S1E2:
case ARMMMUIdx_S2NS:
case ARMMMUIdx_MPriv:
+ case ARMMMUIdx_MNegPri:
case ARMMMUIdx_MUser:
return false;
case ARMMMUIdx_S1E3:
@@ -7077,7 +7079,21 @@ static inline bool
regime_translation_disabled(CPUARMState *env,
ARMMMUIdx mmu_idx)
{
if (arm_feature(env, ARM_FEATURE_M)) {
- return !(env->v7m.mpu_ctrl & R_V7M_MPU_CTRL_ENABLE_MASK);
+ switch (env->v7m.mpu_ctrl &
+ (R_V7M_MPU_CTRL_ENABLE_MASK | R_V7M_MPU_CTRL_HFNMIENA_MASK)) {
+ case R_V7M_MPU_CTRL_ENABLE_MASK:
+ /* Enabled, but not for HardFault and NMI */
+ return mmu_idx == ARMMMUIdx_MNegPri;
+ case R_V7M_MPU_CTRL_ENABLE_MASK | R_V7M_MPU_CTRL_HFNMIENA_MASK:
+ /* Enabled for all cases */
+ return false;
+ case 0:
+ default:
+ /* HFNMIENA set and ENABLE clear is UNPREDICTABLE, but
+ * we warned about that in armv7m_nvic.c when the guest set it.
+ */
+ return true;
+ }
}
if (mmu_idx == ARMMMUIdx_S2NS) {
diff --git a/target/arm/translate.c b/target/arm/translate.c
index ac905dd..ae6646c 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -163,6 +163,7 @@ static inline int get_a32_user_mem_index(DisasContext *s)
return arm_to_core_mmu_idx(ARMMMUIdx_S1SE0);
case ARMMMUIdx_MUser:
case ARMMMUIdx_MPriv:
+ case ARMMMUIdx_MNegPri:
return arm_to_core_mmu_idx(ARMMMUIdx_MUser);
case ARMMMUIdx_S2NS:
default:
--
2.7.4
- [Qemu-devel] [PULL 12/27] arm: Don't let no-MPU PMSA cores write to SCTLR.M, (continued)
- [Qemu-devel] [PULL 12/27] arm: Don't let no-MPU PMSA cores write to SCTLR.M, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 11/27] arm: Don't clear ARM_FEATURE_PMSA for no-mpu configs, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 13/27] arm: Remove unnecessary check on cpu->pmsav7_dregion, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 10/27] arm: Clean up handling of no-MPU PMSA CPUs, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 14/27] armv7m: Improve "-d mmu" tracing for PMSAv7 MPU, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 09/27] arm: Use different ARMMMUIdx values for M profile, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 16/27] arm: All M profile cores are PMSA, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 07/27] arm: Use the mmu_idx we're passed in arm_cpu_do_unaligned_access(), Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 05/27] hw/intc/arm_gicv3_cpuif: Fix priority masking for NS BPR1, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 15/27] armv7m: Implement M profile default memory map, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 19/27] arm: Implement HFNMIENA support for M profile MPU,
Peter Maydell <=
- [Qemu-devel] [PULL 21/27] aspeed/i2c: handle LAST command under the RX command, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 06/27] target/arm: clear PMUVER field of AA64DFR0 when vPMU=off, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 02/27] load_uboot_image: don't assume a full header read, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 01/27] libvixl: Correct build failures on NetBSD, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 18/27] arm: add MPU support to M profile CPUs, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 26/27] hw/arm/virt-acpi-build: build SLIT when needed, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 04/27] hw/intc/arm_gicv3_cpuif: Don't let BPR be set below its minimum, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 25/27] aspeed: add a temp sensor device on I2C bus 3, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 17/27] armv7m: Classify faults as MemManage or BusFault, Peter Maydell, 2017/06/01
- [Qemu-devel] [PULL 20/27] aspeed/i2c: improve command handling, Peter Maydell, 2017/06/01