Series comparison

-[PULL 00/23] target-arm queue
+[PULL 00/14] target-arm queue
-Two small bugfixes, plus most of RTH's refactoring of cpregs
+The following changes since commit 8f6330a807f2642dc2a3cdf33347aa28a4c00a87:
 handling.
--- PMM
+  Merge tag 'pull-maintainer-updates-060324-1' of https://gitlab.com/stsquad/qemu into staging (2024-03-06 16:56:20 +0000)
 The following changes since commit 1fba9dc71a170b3a05b9d3272dd8ecfe7f26e215:
   Merge tag 'pull-request-2022-05-04' of https://gitlab.com/thuth/qemu into staging (2022-05-04 08:07:02 -0700)
 are available in the Git repository at:
-  https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20220505
+  https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20240308
-for you to fetch changes up to 99a50d1a67c602126fc2b3a4812d3000eba9bf34:
+for you to fetch changes up to bbf6c6dbead82292a20951eb1204442a6b838de9:
-  target/arm: read access to performance counters from EL0 (2022-05-05 09:36:22 +0100)
+  target/arm: Move v7m-related code from cpu32.c into a separate file (2024-03-08 14:45:03 +0000)
 ----------------------------------------------------------------
 target-arm queue:
- * Enable read access to performance counters from EL0
+ * Implement FEAT_ECV
- * Enable SCTLR_EL1.BT0 for aarch64-linux-user
+ * STM32L4x5: Implement GPIO device
- * Refactoring of cpreg handling
+ * Fix 32-bit SMOPA
  * Refactor v7m related code from cpu32.c into its own file
  * hw/rtc/sun4v-rtc: Relicense to GPLv2-or-later
 ----------------------------------------------------------------
-Alex Zuepke (1):
+Inès Varhol (3):
-      target/arm: read access to performance counters from EL0
+      hw/gpio: Implement STM32L4x5 GPIO
       hw/arm: Connect STM32L4x5 GPIO to STM32L4x5 SoC
       tests/qtest: Add STM32L4x5 GPIO QTest testcase
-Richard Henderson (22):
+Peter Maydell (9):
-      target/arm: Enable SCTLR_EL1.BT0 for aarch64-linux-user
+      target/arm: Move some register related defines to internals.h
-      target/arm: Split out cpregs.h
+      target/arm: Timer _EL02 registers UNDEF for E2H == 0
-      target/arm: Reorg CPAccessResult and access_check_cp_reg
+      target/arm: use FIELD macro for CNTHCTL bit definitions
-      target/arm: Replace sentinels with ARRAY_SIZE in cpregs.h
+      target/arm: Don't allow RES0 CNTHCTL_EL2 bits to be written
-      target/arm: Make some more cpreg data static const
+      target/arm: Implement new FEAT_ECV trap bits
-      target/arm: Reorg ARMCPRegInfo type field bits
+      target/arm: Define CNTPCTSS_EL0 and CNTVCTSS_EL0
-      target/arm: Avoid bare abort() or assert(0)
+      target/arm: Implement FEAT_ECV CNTPOFF_EL2 handling
-      target/arm: Change cpreg access permissions to enum
+      target/arm: Enable FEAT_ECV for 'max' CPU
-      target/arm: Name CPState type
+      hw/rtc/sun4v-rtc: Relicense to GPLv2-or-later
       target/arm: Name CPSecureState type
       target/arm: Drop always-true test in define_arm_vh_e2h_redirects_aliases
       target/arm: Store cpregs key in the hash table directly
       target/arm: Merge allocation of the cpreg and its name
       target/arm: Hoist computation of key in add_cpreg_to_hashtable
       target/arm: Consolidate cpreg updates in add_cpreg_to_hashtable
       target/arm: Use bool for is64 and ns in add_cpreg_to_hashtable
       target/arm: Hoist isbanked computation in add_cpreg_to_hashtable
       target/arm: Perform override check early in add_cpreg_to_hashtable
       target/arm: Reformat comments in add_cpreg_to_hashtable
       target/arm: Remove HOST_BIG_ENDIAN ifdef in add_cpreg_to_hashtable
       target/arm: Add isar predicates for FEAT_Debugv8p2
       target/arm: Add isar_feature_{aa64,any}_ras
- target/arm/cpregs.h               | 453 ++++++++++++++++++++++++++++++++++++++
+Richard Henderson (1):
- target/arm/cpu.h                  | 393 +++------------------------------
+      target/arm: Fix 32-bit SMOPA
- hw/arm/pxa2xx.c                   |   2 +-
- hw/arm/pxa2xx_pic.c               |   2 +-
+Thomas Huth (1):
- hw/intc/arm_gicv3_cpuif.c         |   6 +-
+      target/arm: Move v7m-related code from cpu32.c into a separate file
- hw/intc/arm_gicv3_kvm.c           |   3 +-
- target/arm/cpu.c                  |  25 +--
+ MAINTAINERS                        |   1 +
- target/arm/cpu64.c                |   2 +-
+ docs/system/arm/b-l475e-iot01a.rst |   2 +-
- target/arm/cpu_tcg.c              |   5 +-
+ docs/system/arm/emulation.rst      |   1 +
- target/arm/gdbstub.c              |   5 +-
+ include/hw/arm/stm32l4x5_soc.h     |   2 +
- target/arm/helper.c               | 358 +++++++++++++-----------------
+ include/hw/gpio/stm32l4x5_gpio.h   |  71 +++++
- target/arm/hvf/hvf.c              |   2 +-
+ include/hw/misc/stm32l4x5_syscfg.h |   3 +-
- target/arm/kvm-stub.c             |   4 +-
+ include/hw/rtc/sun4v-rtc.h         |   2 +-
- target/arm/kvm.c                  |   4 +-
+ target/arm/cpu-features.h          |  10 +
- target/arm/machine.c              |   4 +-
+ target/arm/cpu.h                   | 129 +--------
- target/arm/op_helper.c            |  57 ++---
+ target/arm/internals.h             | 151 ++++++++++
- target/arm/translate-a64.c        |  14 +-
+ hw/arm/stm32l4x5_soc.c             |  71 ++++-
- target/arm/translate-neon.c       |   2 +-
+ hw/gpio/stm32l4x5_gpio.c           | 477 ++++++++++++++++++++++++++++++++
- target/arm/translate.c            |  13 +-
+ hw/misc/stm32l4x5_syscfg.c         |   1 +
- tests/tcg/aarch64/bti-3.c         |  42 ++++
+ hw/rtc/sun4v-rtc.c                 |   2 +-
- tests/tcg/aarch64/Makefile.target |   6 +-
+ target/arm/helper.c                | 189 ++++++++++++-
-files changed, 738 insertions(+), 664 deletions(-)
+ target/arm/tcg/cpu-v7m.c           | 290 +++++++++++++++++++
- create mode 100644 target/arm/cpregs.h
+ target/arm/tcg/cpu32.c             | 261 ------------------
- create mode 100644 tests/tcg/aarch64/bti-3.c
+ target/arm/tcg/cpu64.c             |   1 +
  target/arm/tcg/sme_helper.c        |  77 +++---
  tests/qtest/stm32l4x5_gpio-test.c  | 551 +++++++++++++++++++++++++++++++++++++
  tests/tcg/aarch64/sme-smopa-1.c    |  47 ++++
  tests/tcg/aarch64/sme-smopa-2.c    |  54 ++++
  hw/arm/Kconfig                     |   3 +-
  hw/gpio/Kconfig                    |   3 +
  hw/gpio/meson.build                |   1 +
  hw/gpio/trace-events               |   6 +
  target/arm/meson.build             |   3 +
  target/arm/tcg/meson.build         |   3 +
  target/arm/trace-events            |   1 +
  tests/qtest/meson.build            |   3 +-
  tests/tcg/aarch64/Makefile.target  |   2 +-
 files changed, 1962 insertions(+), 456 deletions(-)
  create mode 100644 include/hw/gpio/stm32l4x5_gpio.h
  create mode 100644 hw/gpio/stm32l4x5_gpio.c
  create mode 100644 target/arm/tcg/cpu-v7m.c
  create mode 100644 tests/qtest/stm32l4x5_gpio-test.c
  create mode 100644 tests/tcg/aarch64/sme-smopa-1.c
  create mode 100644 tests/tcg/aarch64/sme-smopa-2.c

-[PULL 22/23] target/arm: Add isar_feature_{aa64,any}_ras
+[PULL 01/14] target/arm: Move some register related defines to internals.h
-From: Richard Henderson <richard.henderson@linaro.org>
+cpu.h has a lot of #defines relating to CPU register fields.
 Most of these aren't actually used outside target/arm code,
 so there's no point in cluttering up the cpu.h file with them.
 Move some easy ones to internals.h.
-Add the aa64 predicate for detecting RAS support from id registers.
-We already have the aa32 version from the M-profile work.
-Add the 'any' predicate for testing both aa64 and aa32.
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Message-id: 20220501055028.646596-34-richard.henderson@linaro.org
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
+Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
+Message-id: 20240301183219.2424889-2-peter.maydell@linaro.org
 ---
- target/arm/cpu.h | 10 ++++++++++
+ target/arm/cpu.h       | 128 -----------------------------------------
-file changed, 10 insertions(+)
+ target/arm/internals.h | 128 +++++++++++++++++++++++++++++++++++++++++
 files changed, 128 insertions(+), 128 deletions(-)
 diff --git a/target/arm/cpu.h b/target/arm/cpu.h
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/cpu.h
 +++ b/target/arm/cpu.h
-@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_aa32_el1(const ARMISARegisters *id)
+@@ -XXX,XX +XXX,XX @@ typedef struct ARMGenericTimer {
-     return FIELD_EX64(id->id_aa64pfr0, ID_AA64PFR0, EL1) >= 2;
+     uint64_t ctl; /* Timer Control register */
- }
+ } ARMGenericTimer;
-+static inline bool isar_feature_aa64_ras(const ARMISARegisters *id)
+-#define VTCR_NSW (1u << 29)
-+{
+-#define VTCR_NSA (1u << 30)
-+    return FIELD_EX64(id->id_aa64pfr0, ID_AA64PFR0, RAS) != 0;
+-#define VSTCR_SW VTCR_NSW
-+}
+-#define VSTCR_SA VTCR_NSA
-+
+-
- static inline bool isar_feature_aa64_sve(const ARMISARegisters *id)
+ /* Define a maximum sized vector register.
- {
+  * For 32-bit, this is a 128-bit NEON/AdvSIMD register.
-     return FIELD_EX64(id->id_aa64pfr0, ID_AA64PFR0, SVE) != 0;
+  * For 64-bit, this is a 2048-bit SVE register.
-@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_any_debugv8p2(const ARMISARegisters *id)
+@@ -XXX,XX +XXX,XX @@ void pmu_init(ARMCPU *cpu);
-     return isar_feature_aa64_debugv8p2(id) || isar_feature_aa32_debugv8p2(id);
+ #define SCTLR_SPINTMASK (1ULL << 62) /* FEAT_NMI */
- }
+ #define SCTLR_TIDCP   (1ULL << 63) /* FEAT_TIDCP1 */
-+static inline bool isar_feature_any_ras(const ARMISARegisters *id)
+-/* Bit definitions for CPACR (AArch32 only) */
-+{
+-FIELD(CPACR, CP10, 20, 2)
-+    return isar_feature_aa64_ras(id) || isar_feature_aa32_ras(id);
+-FIELD(CPACR, CP11, 22, 2)
-+}
+-FIELD(CPACR, TRCDIS, 28, 1)    /* matches CPACR_EL1.TTA */
-+
+-FIELD(CPACR, D32DIS, 30, 1)    /* up to v7; RAZ in v8 */
- /*
+-FIELD(CPACR, ASEDIS, 31, 1)
-  * Forward to the above feature tests given an ARMCPU pointer.
+-
-  */
+-/* Bit definitions for CPACR_EL1 (AArch64 only) */
 -FIELD(CPACR_EL1, ZEN, 16, 2)
 -FIELD(CPACR_EL1, FPEN, 20, 2)
 -FIELD(CPACR_EL1, SMEN, 24, 2)
 -FIELD(CPACR_EL1, TTA, 28, 1)   /* matches CPACR.TRCDIS */
 -
 -/* Bit definitions for HCPTR (AArch32 only) */
 -FIELD(HCPTR, TCP10, 10, 1)
 -FIELD(HCPTR, TCP11, 11, 1)
 -FIELD(HCPTR, TASE, 15, 1)
 -FIELD(HCPTR, TTA, 20, 1)
 -FIELD(HCPTR, TAM, 30, 1)       /* matches CPTR_EL2.TAM */
 -FIELD(HCPTR, TCPAC, 31, 1)     /* matches CPTR_EL2.TCPAC */
 -
 -/* Bit definitions for CPTR_EL2 (AArch64 only) */
 -FIELD(CPTR_EL2, TZ, 8, 1)      /* !E2H */
 -FIELD(CPTR_EL2, TFP, 10, 1)    /* !E2H, matches HCPTR.TCP10 */
 -FIELD(CPTR_EL2, TSM, 12, 1)    /* !E2H */
 -FIELD(CPTR_EL2, ZEN, 16, 2)    /* E2H */
 -FIELD(CPTR_EL2, FPEN, 20, 2)   /* E2H */
 -FIELD(CPTR_EL2, SMEN, 24, 2)   /* E2H */
 -FIELD(CPTR_EL2, TTA, 28, 1)
 -FIELD(CPTR_EL2, TAM, 30, 1)    /* matches HCPTR.TAM */
 -FIELD(CPTR_EL2, TCPAC, 31, 1)  /* matches HCPTR.TCPAC */
 -
 -/* Bit definitions for CPTR_EL3 (AArch64 only) */
 -FIELD(CPTR_EL3, EZ, 8, 1)
 -FIELD(CPTR_EL3, TFP, 10, 1)
 -FIELD(CPTR_EL3, ESM, 12, 1)
 -FIELD(CPTR_EL3, TTA, 20, 1)
 -FIELD(CPTR_EL3, TAM, 30, 1)
 -FIELD(CPTR_EL3, TCPAC, 31, 1)
 -
 -#define MDCR_MTPME    (1U << 28)
 -#define MDCR_TDCC     (1U << 27)
 -#define MDCR_HLP      (1U << 26)  /* MDCR_EL2 */
 -#define MDCR_SCCD     (1U << 23)  /* MDCR_EL3 */
 -#define MDCR_HCCD     (1U << 23)  /* MDCR_EL2 */
 -#define MDCR_EPMAD    (1U << 21)
 -#define MDCR_EDAD     (1U << 20)
 -#define MDCR_TTRF     (1U << 19)
 -#define MDCR_STE      (1U << 18)  /* MDCR_EL3 */
 -#define MDCR_SPME     (1U << 17)  /* MDCR_EL3 */
 -#define MDCR_HPMD     (1U << 17)  /* MDCR_EL2 */
 -#define MDCR_SDD      (1U << 16)
 -#define MDCR_SPD      (3U << 14)
 -#define MDCR_TDRA     (1U << 11)
 -#define MDCR_TDOSA    (1U << 10)
 -#define MDCR_TDA      (1U << 9)
 -#define MDCR_TDE      (1U << 8)
 -#define MDCR_HPME     (1U << 7)
 -#define MDCR_TPM      (1U << 6)
 -#define MDCR_TPMCR    (1U << 5)
 -#define MDCR_HPMN     (0x1fU)
 -
 -/* Not all of the MDCR_EL3 bits are present in the 32-bit SDCR */
 -#define SDCR_VALID_MASK (MDCR_MTPME | MDCR_TDCC | MDCR_SCCD | \
 -                         MDCR_EPMAD | MDCR_EDAD | MDCR_TTRF | \
 -                         MDCR_STE | MDCR_SPME | MDCR_SPD)
 -
  #define CPSR_M (0x1fU)
  #define CPSR_T (1U << 5)
  #define CPSR_F (1U << 6)
@@ -XXX,XX +XXX,XX @@ FIELD(CPTR_EL3, TCPAC, 31, 1)
  #define XPSR_NZCV CPSR_NZCV
  #define XPSR_IT CPSR_IT
 -#define TTBCR_N      (7U << 0) /* TTBCR.EAE==0 */
 -#define TTBCR_T0SZ   (7U << 0) /* TTBCR.EAE==1 */
 -#define TTBCR_PD0    (1U << 4)
 -#define TTBCR_PD1    (1U << 5)
 -#define TTBCR_EPD0   (1U << 7)
 -#define TTBCR_IRGN0  (3U << 8)
 -#define TTBCR_ORGN0  (3U << 10)
 -#define TTBCR_SH0    (3U << 12)
 -#define TTBCR_T1SZ   (3U << 16)
 -#define TTBCR_A1     (1U << 22)
 -#define TTBCR_EPD1   (1U << 23)
 -#define TTBCR_IRGN1  (3U << 24)
 -#define TTBCR_ORGN1  (3U << 26)
 -#define TTBCR_SH1    (1U << 28)
 -#define TTBCR_EAE    (1U << 31)
 -
 -FIELD(VTCR, T0SZ, 0, 6)
 -FIELD(VTCR, SL0, 6, 2)
 -FIELD(VTCR, IRGN0, 8, 2)
 -FIELD(VTCR, ORGN0, 10, 2)
 -FIELD(VTCR, SH0, 12, 2)
 -FIELD(VTCR, TG0, 14, 2)
 -FIELD(VTCR, PS, 16, 3)
 -FIELD(VTCR, VS, 19, 1)
 -FIELD(VTCR, HA, 21, 1)
 -FIELD(VTCR, HD, 22, 1)
 -FIELD(VTCR, HWU59, 25, 1)
 -FIELD(VTCR, HWU60, 26, 1)
 -FIELD(VTCR, HWU61, 27, 1)
 -FIELD(VTCR, HWU62, 28, 1)
 -FIELD(VTCR, NSW, 29, 1)
 -FIELD(VTCR, NSA, 30, 1)
 -FIELD(VTCR, DS, 32, 1)
 -FIELD(VTCR, SL2, 33, 1)
 -
  /* Bit definitions for ARMv8 SPSR (PSTATE) format.
   * Only these are valid when in AArch64 mode; in
   * AArch32 mode SPSRs are basically CPSR-format.
@@ -XXX,XX +XXX,XX @@ static inline void xpsr_write(CPUARMState *env, uint32_t val, uint32_t mask)
  #define HCR_TWEDEN    (1ULL << 59)
  #define HCR_TWEDEL    MAKE_64BIT_MASK(60, 4)
 -#define HCRX_ENAS0    (1ULL << 0)
 -#define HCRX_ENALS    (1ULL << 1)
 -#define HCRX_ENASR    (1ULL << 2)
 -#define HCRX_FNXS     (1ULL << 3)
 -#define HCRX_FGTNXS   (1ULL << 4)
 -#define HCRX_SMPME    (1ULL << 5)
 -#define HCRX_TALLINT  (1ULL << 6)
 -#define HCRX_VINMI    (1ULL << 7)
 -#define HCRX_VFNMI    (1ULL << 8)
 -#define HCRX_CMOW     (1ULL << 9)
 -#define HCRX_MCE2     (1ULL << 10)
 -#define HCRX_MSCEN    (1ULL << 11)
 -
 -#define HPFAR_NS      (1ULL << 63)
 -
  #define SCR_NS                (1ULL << 0)
  #define SCR_IRQ               (1ULL << 1)
  #define SCR_FIQ               (1ULL << 2)
@@ -XXX,XX +XXX,XX @@ static inline void xpsr_write(CPUARMState *env, uint32_t val, uint32_t mask)
  #define SCR_GPF               (1ULL << 48)
  #define SCR_NSE               (1ULL << 62)
 -#define HSTR_TTEE (1 << 16)
 -#define HSTR_TJDBX (1 << 17)
 -
 -#define CNTHCTL_CNTVMASK      (1 << 18)
 -#define CNTHCTL_CNTPMASK      (1 << 19)
 -
  /* Return the current FPSCR value.  */
  uint32_t vfp_get_fpscr(CPUARMState *env);
  void vfp_set_fpscr(CPUARMState *env, uint32_t val);
 diff --git a/target/arm/internals.h b/target/arm/internals.h
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/internals.h
 +++ b/target/arm/internals.h
@@ -XXX,XX +XXX,XX @@ FIELD(DBGWCR, WT, 20, 1)
  FIELD(DBGWCR, MASK, 24, 5)
  FIELD(DBGWCR, SSCE, 29, 1)
 +#define VTCR_NSW (1u << 29)
 +#define VTCR_NSA (1u << 30)
 +#define VSTCR_SW VTCR_NSW
 +#define VSTCR_SA VTCR_NSA
 +
 +/* Bit definitions for CPACR (AArch32 only) */
 +FIELD(CPACR, CP10, 20, 2)
 +FIELD(CPACR, CP11, 22, 2)
 +FIELD(CPACR, TRCDIS, 28, 1)    /* matches CPACR_EL1.TTA */
 +FIELD(CPACR, D32DIS, 30, 1)    /* up to v7; RAZ in v8 */
 +FIELD(CPACR, ASEDIS, 31, 1)
 +
 +/* Bit definitions for CPACR_EL1 (AArch64 only) */
 +FIELD(CPACR_EL1, ZEN, 16, 2)
 +FIELD(CPACR_EL1, FPEN, 20, 2)
 +FIELD(CPACR_EL1, SMEN, 24, 2)
 +FIELD(CPACR_EL1, TTA, 28, 1)   /* matches CPACR.TRCDIS */
 +
 +/* Bit definitions for HCPTR (AArch32 only) */
 +FIELD(HCPTR, TCP10, 10, 1)
 +FIELD(HCPTR, TCP11, 11, 1)
 +FIELD(HCPTR, TASE, 15, 1)
 +FIELD(HCPTR, TTA, 20, 1)
 +FIELD(HCPTR, TAM, 30, 1)       /* matches CPTR_EL2.TAM */
 +FIELD(HCPTR, TCPAC, 31, 1)     /* matches CPTR_EL2.TCPAC */
 +
 +/* Bit definitions for CPTR_EL2 (AArch64 only) */
 +FIELD(CPTR_EL2, TZ, 8, 1)      /* !E2H */
 +FIELD(CPTR_EL2, TFP, 10, 1)    /* !E2H, matches HCPTR.TCP10 */
 +FIELD(CPTR_EL2, TSM, 12, 1)    /* !E2H */
 +FIELD(CPTR_EL2, ZEN, 16, 2)    /* E2H */
 +FIELD(CPTR_EL2, FPEN, 20, 2)   /* E2H */
 +FIELD(CPTR_EL2, SMEN, 24, 2)   /* E2H */
 +FIELD(CPTR_EL2, TTA, 28, 1)
 +FIELD(CPTR_EL2, TAM, 30, 1)    /* matches HCPTR.TAM */
 +FIELD(CPTR_EL2, TCPAC, 31, 1)  /* matches HCPTR.TCPAC */
 +
 +/* Bit definitions for CPTR_EL3 (AArch64 only) */
 +FIELD(CPTR_EL3, EZ, 8, 1)
 +FIELD(CPTR_EL3, TFP, 10, 1)
 +FIELD(CPTR_EL3, ESM, 12, 1)
 +FIELD(CPTR_EL3, TTA, 20, 1)
 +FIELD(CPTR_EL3, TAM, 30, 1)
 +FIELD(CPTR_EL3, TCPAC, 31, 1)
 +
 +#define MDCR_MTPME    (1U << 28)
 +#define MDCR_TDCC     (1U << 27)
 +#define MDCR_HLP      (1U << 26)  /* MDCR_EL2 */
 +#define MDCR_SCCD     (1U << 23)  /* MDCR_EL3 */
 +#define MDCR_HCCD     (1U << 23)  /* MDCR_EL2 */
 +#define MDCR_EPMAD    (1U << 21)
 +#define MDCR_EDAD     (1U << 20)
 +#define MDCR_TTRF     (1U << 19)
 +#define MDCR_STE      (1U << 18)  /* MDCR_EL3 */
 +#define MDCR_SPME     (1U << 17)  /* MDCR_EL3 */
 +#define MDCR_HPMD     (1U << 17)  /* MDCR_EL2 */
 +#define MDCR_SDD      (1U << 16)
 +#define MDCR_SPD      (3U << 14)
 +#define MDCR_TDRA     (1U << 11)
 +#define MDCR_TDOSA    (1U << 10)
 +#define MDCR_TDA      (1U << 9)
 +#define MDCR_TDE      (1U << 8)
 +#define MDCR_HPME     (1U << 7)
 +#define MDCR_TPM      (1U << 6)
 +#define MDCR_TPMCR    (1U << 5)
 +#define MDCR_HPMN     (0x1fU)
 +
 +/* Not all of the MDCR_EL3 bits are present in the 32-bit SDCR */
 +#define SDCR_VALID_MASK (MDCR_MTPME | MDCR_TDCC | MDCR_SCCD | \
 +                         MDCR_EPMAD | MDCR_EDAD | MDCR_TTRF | \
 +                         MDCR_STE | MDCR_SPME | MDCR_SPD)
 +
 +#define TTBCR_N      (7U << 0) /* TTBCR.EAE==0 */
 +#define TTBCR_T0SZ   (7U << 0) /* TTBCR.EAE==1 */
 +#define TTBCR_PD0    (1U << 4)
 +#define TTBCR_PD1    (1U << 5)
 +#define TTBCR_EPD0   (1U << 7)
 +#define TTBCR_IRGN0  (3U << 8)
 +#define TTBCR_ORGN0  (3U << 10)
 +#define TTBCR_SH0    (3U << 12)
 +#define TTBCR_T1SZ   (3U << 16)
 +#define TTBCR_A1     (1U << 22)
 +#define TTBCR_EPD1   (1U << 23)
 +#define TTBCR_IRGN1  (3U << 24)
 +#define TTBCR_ORGN1  (3U << 26)
 +#define TTBCR_SH1    (1U << 28)
 +#define TTBCR_EAE    (1U << 31)
 +
 +FIELD(VTCR, T0SZ, 0, 6)
 +FIELD(VTCR, SL0, 6, 2)
 +FIELD(VTCR, IRGN0, 8, 2)
 +FIELD(VTCR, ORGN0, 10, 2)
 +FIELD(VTCR, SH0, 12, 2)
 +FIELD(VTCR, TG0, 14, 2)
 +FIELD(VTCR, PS, 16, 3)
 +FIELD(VTCR, VS, 19, 1)
 +FIELD(VTCR, HA, 21, 1)
 +FIELD(VTCR, HD, 22, 1)
 +FIELD(VTCR, HWU59, 25, 1)
 +FIELD(VTCR, HWU60, 26, 1)
 +FIELD(VTCR, HWU61, 27, 1)
 +FIELD(VTCR, HWU62, 28, 1)
 +FIELD(VTCR, NSW, 29, 1)
 +FIELD(VTCR, NSA, 30, 1)
 +FIELD(VTCR, DS, 32, 1)
 +FIELD(VTCR, SL2, 33, 1)
 +
 +#define HCRX_ENAS0    (1ULL << 0)
 +#define HCRX_ENALS    (1ULL << 1)
 +#define HCRX_ENASR    (1ULL << 2)
 +#define HCRX_FNXS     (1ULL << 3)
 +#define HCRX_FGTNXS   (1ULL << 4)
 +#define HCRX_SMPME    (1ULL << 5)
 +#define HCRX_TALLINT  (1ULL << 6)
 +#define HCRX_VINMI    (1ULL << 7)
 +#define HCRX_VFNMI    (1ULL << 8)
 +#define HCRX_CMOW     (1ULL << 9)
 +#define HCRX_MCE2     (1ULL << 10)
 +#define HCRX_MSCEN    (1ULL << 11)
 +
 +#define HPFAR_NS      (1ULL << 63)
 +
 +#define HSTR_TTEE (1 << 16)
 +#define HSTR_TJDBX (1 << 17)
 +
 +#define CNTHCTL_CNTVMASK      (1 << 18)
 +#define CNTHCTL_CNTPMASK      (1 << 19)
 +
  /* We use a few fake FSR values for internal purposes in M profile.
   * M profile cores don't have A/R format FSRs, but currently our
   * get_phys_addr() code assumes A/R profile and reports failures via
 --
-.25.1
+.34.1

-[PULL 23/23] target/arm: read access to performance counters from EL0
+[PULL 02/14] target/arm: Timer _EL02 registers UNDEF for E2H == 0
-From: Alex Zuepke <alex.zuepke@tum.de>
+The timer _EL02 registers should UNDEF for invalid accesses from EL2
 or EL3 when HCR_EL2.E2H == 0, not take a cp access trap.  We were
 delivering the exception to EL2 with the wrong syndrome.
-The ARMv8 manual defines that PMUSERENR_EL0.ER enables read-access
+Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
 to both PMXEVCNTR_EL0 and PMEVCNTR<n>_EL0 registers, however,
 we only use it for PMXEVCNTR_EL0. Extend to PMEVCNTR<n>_EL0 as well.
 Signed-off-by: Alex Zuepke <alex.zuepke@tum.de>
 Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
-Message-id: 20220428132717.84190-1-alex.zuepke@tum.de
+Message-id: 20240301183219.2424889-3-peter.maydell@linaro.org
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
 ---
- target/arm/helper.c | 4 ++--
+ target/arm/helper.c | 2 +-
-file changed, 2 insertions(+), 2 deletions(-)
+file changed, 1 insertion(+), 1 deletion(-)
 diff --git a/target/arm/helper.c b/target/arm/helper.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/helper.c
 +++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ static void define_pmu_regs(ARMCPU *cpu)
+@@ -XXX,XX +XXX,XX @@ static CPAccessResult e2h_access(CPUARMState *env, const ARMCPRegInfo *ri,
-               .crm = 8 | (3 & (i >> 3)), .opc1 = 0, .opc2 = i & 7,
+         return CP_ACCESS_OK;
-               .access = PL0_RW, .type = ARM_CP_IO | ARM_CP_ALIAS,
+     }
-               .readfn = pmevcntr_readfn, .writefn = pmevcntr_writefn,
+     if (!(arm_hcr_el2_eff(env) & HCR_E2H)) {
--              .accessfn = pmreg_access },
+-        return CP_ACCESS_TRAP;
-+              .accessfn = pmreg_access_xevcntr },
++        return CP_ACCESS_TRAP_UNCATEGORIZED;
-             { .name = pmevcntr_el0_name, .state = ARM_CP_STATE_AA64,
+     }
-               .opc0 = 3, .opc1 = 3, .crn = 14, .crm = 8 | (3 & (i >> 3)),
+     return CP_ACCESS_OK;
--              .opc2 = i & 7, .access = PL0_RW, .accessfn = pmreg_access,
+ }
 +              .opc2 = i & 7, .access = PL0_RW, .accessfn = pmreg_access_xevcntr,
                .type = ARM_CP_IO,
                .readfn = pmevcntr_readfn, .writefn = pmevcntr_writefn,
                .raw_readfn = pmevcntr_rawread,
 --
-.25.1
+.34.1

-[PULL 05/23] target/arm: Make some more cpreg data static const
+[PULL 03/14] target/arm: use FIELD macro for CNTHCTL bit definitions
-From: Richard Henderson <richard.henderson@linaro.org>
+We prefer the FIELD macro over ad-hoc #defines for register bits;
 switch CNTHCTL to that style before we add any more bits.
-These particular data structures are not modified at runtime.
+Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
 Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
 Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
 Message-id: 20240301183219.2424889-4-peter.maydell@linaro.org
 ---
  target/arm/internals.h | 27 +++++++++++++++++++++++++--
  target/arm/helper.c    |  9 ++++-----
 files changed, 29 insertions(+), 7 deletions(-)
-Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
+diff --git a/target/arm/internals.h b/target/arm/internals.h
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
+index XXXXXXX..XXXXXXX 100644
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
+--- a/target/arm/internals.h
-Message-id: 20220501055028.646596-5-richard.henderson@linaro.org
++++ b/target/arm/internals.h
-Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
+@@ -XXX,XX +XXX,XX @@ FIELD(VTCR, SL2, 33, 1)
----
+ #define HSTR_TTEE (1 << 16)
- target/arm/helper.c | 16 ++++++++--------
+ #define HSTR_TJDBX (1 << 17)
-file changed, 8 insertions(+), 8 deletions(-)
+-#define CNTHCTL_CNTVMASK      (1 << 18)
 -#define CNTHCTL_CNTPMASK      (1 << 19)
 +/*
 + * Depending on the value of HCR_EL2.E2H, bits 0 and 1
 + * have different bit definitions, and EL1PCTEN might be
 + * bit 0 or bit 10. We use _E2H1 and _E2H0 suffixes to
 + * disambiguate if necessary.
 + */
 +FIELD(CNTHCTL, EL0PCTEN_E2H1, 0, 1)
 +FIELD(CNTHCTL, EL0VCTEN_E2H1, 1, 1)
 +FIELD(CNTHCTL, EL1PCTEN_E2H0, 0, 1)
 +FIELD(CNTHCTL, EL1PCEN_E2H0, 1, 1)
 +FIELD(CNTHCTL, EVNTEN, 2, 1)
 +FIELD(CNTHCTL, EVNTDIR, 3, 1)
 +FIELD(CNTHCTL, EVNTI, 4, 4)
 +FIELD(CNTHCTL, EL0VTEN, 8, 1)
 +FIELD(CNTHCTL, EL0PTEN, 9, 1)
 +FIELD(CNTHCTL, EL1PCTEN_E2H1, 10, 1)
 +FIELD(CNTHCTL, EL1PTEN, 11, 1)
 +FIELD(CNTHCTL, ECV, 12, 1)
 +FIELD(CNTHCTL, EL1TVT, 13, 1)
 +FIELD(CNTHCTL, EL1TVCT, 14, 1)
 +FIELD(CNTHCTL, EL1NVPCT, 15, 1)
 +FIELD(CNTHCTL, EL1NVVCT, 16, 1)
 +FIELD(CNTHCTL, EVNTIS, 17, 1)
 +FIELD(CNTHCTL, CNTVMASK, 18, 1)
 +FIELD(CNTHCTL, CNTPMASK, 19, 1)
  /* We use a few fake FSR values for internal purposes in M profile.
   * M profile cores don't have A/R format FSRs, but currently our
 diff --git a/target/arm/helper.c b/target/arm/helper.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/helper.c
 +++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
+@@ -XXX,XX +XXX,XX @@ static void gt_update_irq(ARMCPU *cpu, int timeridx)
-               .resetvalue = cpu->pmceid1 },
+      * It is RES0 in Secure and NonSecure state.
          };
  #ifdef CONFIG_USER_ONLY
 -        ARMCPRegUserSpaceInfo v8_user_idregs[] = {
 +        static const ARMCPRegUserSpaceInfo v8_user_idregs[] = {
              { .name = "ID_AA64PFR0_EL1",
                .exported_bits = 0x000f000f00ff0000,
                .fixed_bits    = 0x0000000000000011 },
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
       */
-     if (arm_feature(env, ARM_FEATURE_EL3)) {
+     if ((ss == ARMSS_Root || ss == ARMSS_Realm) &&
-         if (arm_feature(env, ARM_FEATURE_AARCH64)) {
+-        ((timeridx == GTIMER_VIRT && (cnthctl & CNTHCTL_CNTVMASK)) ||
--            ARMCPRegInfo nsacr = {
+-         (timeridx == GTIMER_PHYS && (cnthctl & CNTHCTL_CNTPMASK)))) {
-+            static const ARMCPRegInfo nsacr = {
++        ((timeridx == GTIMER_VIRT && (cnthctl & R_CNTHCTL_CNTVMASK_MASK)) ||
-                 .name = "NSACR", .type = ARM_CP_CONST,
++         (timeridx == GTIMER_PHYS && (cnthctl & R_CNTHCTL_CNTPMASK_MASK)))) {
-                 .cp = 15, .opc1 = 0, .crn = 1, .crm = 1, .opc2 = 2,
+         irqstate = 0;
                  .access = PL1_RW, .accessfn = nsacr_access,
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
              };
              define_one_arm_cp_reg(cpu, &nsacr);
          } else {
 -            ARMCPRegInfo nsacr = {
 +            static const ARMCPRegInfo nsacr = {
                  .name = "NSACR",
                  .cp = 15, .opc1 = 0, .crn = 1, .crm = 1, .opc2 = 2,
                  .access = PL3_RW | PL1_R,
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
          }
      } else {
          if (arm_feature(env, ARM_FEATURE_V8)) {
 -            ARMCPRegInfo nsacr = {
 +            static const ARMCPRegInfo nsacr = {
                  .name = "NSACR", .type = ARM_CP_CONST,
                  .cp = 15, .opc1 = 0, .crn = 1, .crm = 1, .opc2 = 2,
                  .access = PL1_R,
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                .access = PL1_R, .type = ARM_CP_CONST,
                .resetvalue = cpu->pmsav7_dregion << 8
          };
 -        ARMCPRegInfo crn0_wi_reginfo = {
 +        static const ARMCPRegInfo crn0_wi_reginfo = {
              .name = "CRN0_WI", .cp = 15, .crn = 0, .crm = CP_ANY,
              .opc1 = CP_ANY, .opc2 = CP_ANY, .access = PL1_W,
              .type = ARM_CP_NOP | ARM_CP_OVERRIDE
          };
  #ifdef CONFIG_USER_ONLY
 -        ARMCPRegUserSpaceInfo id_v8_user_midr_cp_reginfo[] = {
 +        static const ARMCPRegUserSpaceInfo id_v8_user_midr_cp_reginfo[] = {
              { .name = "MIDR_EL1",
                .exported_bits = 0x00000000ffffffff },
              { .name = "REVIDR_EL1"                },
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                .access = PL1_R, .readfn = mpidr_read, .type = ARM_CP_NO_RAW },
          };
  #ifdef CONFIG_USER_ONLY
 -        ARMCPRegUserSpaceInfo mpidr_user_cp_reginfo[] = {
 +        static const ARMCPRegUserSpaceInfo mpidr_user_cp_reginfo[] = {
              { .name = "MPIDR_EL1",
                .fixed_bits = 0x0000000080000000 },
          };
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
      }
-     if (arm_feature(env, ARM_FEATURE_VBAR)) {
+@@ -XXX,XX +XXX,XX @@ static void gt_cnthctl_write(CPUARMState *env, const ARMCPRegInfo *ri,
--        ARMCPRegInfo vbar_cp_reginfo[] = {
+ {
-+        static const ARMCPRegInfo vbar_cp_reginfo[] = {
+     ARMCPU *cpu = env_archcpu(env);
-             { .name = "VBAR", .state = ARM_CP_STATE_BOTH,
+     uint32_t oldval = env->cp15.cnthctl_el2;
-               .opc0 = 3, .crn = 12, .crm = 0, .opc1 = 0, .opc2 = 0,
+-
-               .access = PL1_RW, .writefn = vbar_write,
+     raw_write(env, ri, value);
 -    if ((oldval ^ value) & CNTHCTL_CNTVMASK) {
 +    if ((oldval ^ value) & R_CNTHCTL_CNTVMASK_MASK) {
          gt_update_irq(cpu, GTIMER_VIRT);
 -    } else if ((oldval ^ value) & CNTHCTL_CNTPMASK) {
 +    } else if ((oldval ^ value) & R_CNTHCTL_CNTPMASK_MASK) {
          gt_update_irq(cpu, GTIMER_PHYS);
      }
  }
 --
-.25.1
+.34.1

-[PULL 18/23] target/arm: Perform override check early in add_cpreg_to_hashtable
+[PULL 04/14] target/arm: Don't allow RES0 CNTHCTL_EL2 bits to be written
-From: Richard Henderson <richard.henderson@linaro.org>
+Don't allow the guest to write CNTHCTL_EL2 bits which don't exist.
 This is not strictly architecturally required, but it is how we've
 tended to implement registers more recently.
-Perform the override check early, so that it is still done
+In particular, bits [19:18] are only present with FEAT_RME,
-even when we decide to discard an unreachable cpreg.
+and bits [17:12] will only be present with FEAT_ECV.
-Use assert not printf+abort.
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Message-id: 20220501055028.646596-18-richard.henderson@linaro.org
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
+Message-id: 20240301183219.2424889-5-peter.maydell@linaro.org
 ---
- target/arm/helper.c | 22 ++++++++--------------
+ target/arm/helper.c | 18 ++++++++++++++++++
-file changed, 8 insertions(+), 14 deletions(-)
+file changed, 18 insertions(+)
 diff --git a/target/arm/helper.c b/target/arm/helper.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/helper.c
 +++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
+@@ -XXX,XX +XXX,XX @@ static void gt_cnthctl_write(CPUARMState *env, const ARMCPRegInfo *ri,
-         g_assert_not_reached();
+ {
-     }
+     ARMCPU *cpu = env_archcpu(env);
+     uint32_t oldval = env->cp15.cnthctl_el2;
-+    /* Overriding of an existing definition must be explicitly requested. */
++    uint32_t valid_mask =
-+    if (!(r->type & ARM_CP_OVERRIDE)) {
++        R_CNTHCTL_EL0PCTEN_E2H1_MASK |
-+        const ARMCPRegInfo *oldreg = get_arm_cp_reginfo(cpu->cp_regs, key);
++        R_CNTHCTL_EL0VCTEN_E2H1_MASK |
-+        if (oldreg) {
++        R_CNTHCTL_EVNTEN_MASK |
-+            assert(oldreg->type & ARM_CP_OVERRIDE);
++        R_CNTHCTL_EVNTDIR_MASK |
-+        }
++        R_CNTHCTL_EVNTI_MASK |
 +        R_CNTHCTL_EL0VTEN_MASK |
 +        R_CNTHCTL_EL0PTEN_MASK |
 +        R_CNTHCTL_EL1PCTEN_E2H1_MASK |
 +        R_CNTHCTL_EL1PTEN_MASK;
 +
 +    if (cpu_isar_feature(aa64_rme, cpu)) {
 +        valid_mask |= R_CNTHCTL_CNTVMASK_MASK | R_CNTHCTL_CNTPMASK_MASK;
 +    }
 +
-     /* Combine cpreg and name into one allocation. */
++    /* Clear RES0 bits */
-     name_len = strlen(name) + 1;
++    value &= valid_mask;
-     r2 = g_malloc(sizeof(*r2) + name_len);
++
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
+     raw_write(env, ri, value);
-         assert(!raw_accessors_invalid(r2));
-     }
+     if ((oldval ^ value) & R_CNTHCTL_CNTVMASK_MASK) {
 -    /* Overriding of an existing definition must be explicitly
 -     * requested.
 -     */
 -    if (!(r->type & ARM_CP_OVERRIDE)) {
 -        const ARMCPRegInfo *oldreg = get_arm_cp_reginfo(cpu->cp_regs, key);
 -        if (oldreg && !(oldreg->type & ARM_CP_OVERRIDE)) {
 -            fprintf(stderr, "Register redefined: cp=%d %d bit "
 -                    "crn=%d crm=%d opc1=%d opc2=%d, "
 -                    "was %s, now %s\n", r2->cp, 32 + 32 * is64,
 -                    r2->crn, r2->crm, r2->opc1, r2->opc2,
 -                    oldreg->name, r2->name);
 -            g_assert_not_reached();
 -        }
 -    }
      g_hash_table_insert(cpu->cp_regs, (gpointer)(uintptr_t)key, r2);
  }
 --
-.25.1
+.34.1

-[PULL 07/23] target/arm: Avoid bare abort() or assert(0)
+[PULL 05/14] target/arm: Implement new FEAT_ECV trap bits
-From: Richard Henderson <richard.henderson@linaro.org>
+The functionality defined by ID_AA64MMFR0_EL1.ECV == 1 is:
  * four new trap bits for various counter and timer registers
  * the CNTHCTL_EL2.EVNTIS and CNTKCTL_EL1.EVNTIS bits which control
    scaling of the event stream. This is a no-op for us, because we don't
    implement the event stream (our WFE is a NOP): all we need to do is
    allow CNTHCTL_EL2.ENVTIS to be read and written.
  * extensions to PMSCR_EL1.PCT, PMSCR_EL2.PCT, TRFCR_EL1.TS and
    TRFCR_EL2.TS: these are all no-ops for us, because we don't implement
    FEAT_SPE or FEAT_TRF.
  * new registers CNTPCTSS_EL0 and NCTVCTSS_EL0 which are
    "self-sychronizing" views of the CNTPCT_EL0 and CNTVCT_EL0, meaning
    that no barriers are needed around their accesses. For us these
    are just the same as the normal views, because all our sysregs are
    inherently self-sychronizing.
-Standardize on g_assert_not_reached() for "should not happen".
+In this commit we implement the trap handling and permit the new
-Retain abort() when preceeded by fprintf or error_report.
+CNTHCTL_EL2 bits to be written.
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Message-id: 20220501055028.646596-7-richard.henderson@linaro.org
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
+Message-id: 20240301183219.2424889-6-peter.maydell@linaro.org
 ---
- target/arm/helper.c         | 7 +++----
+ target/arm/cpu-features.h |  5 ++++
- target/arm/hvf/hvf.c        | 2 +-
+ target/arm/helper.c       | 51 +++++++++++++++++++++++++++++++++++----
- target/arm/kvm-stub.c       | 4 ++--
+files changed, 51 insertions(+), 5 deletions(-)
  target/arm/kvm.c            | 4 ++--
  target/arm/machine.c        | 4 ++--
  target/arm/translate-a64.c  | 4 ++--
  target/arm/translate-neon.c | 2 +-
  target/arm/translate.c      | 4 ++--
 files changed, 15 insertions(+), 16 deletions(-)
+diff --git a/target/arm/cpu-features.h b/target/arm/cpu-features.h
+index XXXXXXX..XXXXXXX 100644
+--- a/target/arm/cpu-features.h
++++ b/target/arm/cpu-features.h
+@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_fgt(const ARMISARegisters *id)
+     return FIELD_EX64(id->id_aa64mmfr0, ID_AA64MMFR0, FGT) != 0;
+ }
++static inline bool isar_feature_aa64_ecv_traps(const ARMISARegisters *id)
++{
++    return FIELD_EX64(id->id_aa64mmfr0, ID_AA64MMFR0, ECV) > 0;
++}
++
+ static inline bool isar_feature_aa64_vh(const ARMISARegisters *id)
+ {
+     return FIELD_EX64(id->id_aa64mmfr1, ID_AA64MMFR1, VH) != 0;
 diff --git a/target/arm/helper.c b/target/arm/helper.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/helper.c
 +++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
+@@ -XXX,XX +XXX,XX @@ static CPAccessResult gt_counter_access(CPUARMState *env, int timeridx,
-             break;
+              : !extract32(env->cp15.cnthctl_el2, 0, 1))) {
-         default:
+             return CP_ACCESS_TRAP_EL2;
              /* broken reginfo with out-of-range opc1 */
 -            assert(false);
 -            break;
 +            g_assert_not_reached();
          }
-         /* assert our permissions are not too lax (stricter is fine) */
++        if (has_el2 && timeridx == GTIMER_VIRT) {
-         assert((r->access & ~mask) == 0);
++            if (FIELD_EX64(env->cp15.cnthctl_el2, CNTHCTL, EL1TVCT)) {
-@@ -XXX,XX +XXX,XX @@ static bool get_phys_addr_v5(CPUARMState *env, uint32_t address,
++                return CP_ACCESS_TRAP_EL2;
-             break;
++            }
-         default:
++        }
              /* Never happens, but compiler isn't smart enough to tell.  */
 -            abort();
 +            g_assert_not_reached();
          }
      }
      *prot = ap_to_rw_prot(env, mmu_idx, ap, domain_prot);
@@ -XXX,XX +XXX,XX @@ static bool get_phys_addr_v6(CPUARMState *env, uint32_t address,
              break;
          default:
              /* Never happens, but compiler isn't smart enough to tell.  */
 -            abort();
 +            g_assert_not_reached();
          }
      }
      if (domain_prot == 3) {
 diff --git a/target/arm/hvf/hvf.c b/target/arm/hvf/hvf.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/hvf/hvf.c
 +++ b/target/arm/hvf/hvf.c
@@ -XXX,XX +XXX,XX @@ int hvf_vcpu_exec(CPUState *cpu)
          /* we got kicked, no exit to process */
          return 0;
      default:
 -        assert(0);
 +        g_assert_not_reached();
      }
      hvf_sync_vtimer(cpu);
 diff --git a/target/arm/kvm-stub.c b/target/arm/kvm-stub.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/kvm-stub.c
 +++ b/target/arm/kvm-stub.c
@@ -XXX,XX +XXX,XX @@
  bool write_kvmstate_to_list(ARMCPU *cpu)
  {
 -    abort();
 +    g_assert_not_reached();
  }
  bool write_list_to_kvmstate(ARMCPU *cpu, int level)
  {
 -    abort();
 +    g_assert_not_reached();
  }
 diff --git a/target/arm/kvm.c b/target/arm/kvm.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/kvm.c
 +++ b/target/arm/kvm.c
@@ -XXX,XX +XXX,XX @@ bool write_kvmstate_to_list(ARMCPU *cpu)
              ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &r);
              break;
          default:
 -            abort();
 +            g_assert_not_reached();
          }
          if (ret) {
              ok = false;
@@ -XXX,XX +XXX,XX @@ bool write_list_to_kvmstate(ARMCPU *cpu, int level)
              r.addr = (uintptr_t)(cpu->cpreg_values + i);
              break;
          default:
 -            abort();
 +            g_assert_not_reached();
          }
          ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &r);
          if (ret) {
 diff --git a/target/arm/machine.c b/target/arm/machine.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/machine.c
 +++ b/target/arm/machine.c
@@ -XXX,XX +XXX,XX @@ static int cpu_pre_save(void *opaque)
      if (kvm_enabled()) {
          if (!write_kvmstate_to_list(cpu)) {
              /* This should never fail */
 -            abort();
 +            g_assert_not_reached();
          }
          /*
@@ -XXX,XX +XXX,XX @@ static int cpu_pre_save(void *opaque)
      } else {
          if (!write_cpustate_to_list(cpu, false)) {
              /* This should never fail. */
 -            abort();
 +            g_assert_not_reached();
          }
      }
 diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/translate-a64.c
 +++ b/target/arm/translate-a64.c
@@ -XXX,XX +XXX,XX @@ static void handle_fp_1src_half(DisasContext *s, int opcode, int rd, int rn)
          gen_helper_advsimd_rinth(tcg_res, tcg_op, fpst);
          break;
      default:
 -        abort();
 +        g_assert_not_reached();
      }
      write_fp_sreg(s, rd, tcg_res);
@@ -XXX,XX +XXX,XX @@ static void handle_fp_fcvt(DisasContext *s, int opcode,
          break;
      }
-     default:
+     return CP_ACCESS_OK;
--        abort();
+@@ -XXX,XX +XXX,XX @@ static CPAccessResult gt_timer_access(CPUARMState *env, int timeridx,
-+        g_assert_not_reached();
+                 }
              }
          }
 +        if (has_el2 && timeridx == GTIMER_VIRT) {
 +            if (FIELD_EX64(env->cp15.cnthctl_el2, CNTHCTL, EL1TVT)) {
 +                return CP_ACCESS_TRAP_EL2;
 +            }
 +        }
          break;
      }
+     return CP_ACCESS_OK;
+@@ -XXX,XX +XXX,XX @@ static void gt_cnthctl_write(CPUARMState *env, const ARMCPRegInfo *ri,
+     if (cpu_isar_feature(aa64_rme, cpu)) {
+         valid_mask |= R_CNTHCTL_CNTVMASK_MASK | R_CNTHCTL_CNTPMASK_MASK;
+     }
++    if (cpu_isar_feature(aa64_ecv_traps, cpu)) {
++        valid_mask |=
++            R_CNTHCTL_EL1TVT_MASK |
++            R_CNTHCTL_EL1TVCT_MASK |
++            R_CNTHCTL_EL1NVPCT_MASK |
++            R_CNTHCTL_EL1NVVCT_MASK |
++            R_CNTHCTL_EVNTIS_MASK;
++    }
+     /* Clear RES0 bits */
+     value &= valid_mask;
+@@ -XXX,XX +XXX,XX @@ static CPAccessResult e2h_access(CPUARMState *env, const ARMCPRegInfo *ri,
+ {
+     if (arm_current_el(env) == 1) {
+         /* This must be a FEAT_NV access */
+-        /* TODO: FEAT_ECV will need to check CNTHCTL_EL2 here */
+         return CP_ACCESS_OK;
+     }
+     if (!(arm_hcr_el2_eff(env) & HCR_E2H)) {
+@@ -XXX,XX +XXX,XX @@ static CPAccessResult e2h_access(CPUARMState *env, const ARMCPRegInfo *ri,
+     return CP_ACCESS_OK;
  }
-diff --git a/target/arm/translate-neon.c b/target/arm/translate-neon.c
++static CPAccessResult access_el1nvpct(CPUARMState *env, const ARMCPRegInfo *ri,
-index XXXXXXX..XXXXXXX 100644
++                                      bool isread)
---- a/target/arm/translate-neon.c
++{
-+++ b/target/arm/translate-neon.c
++    if (arm_current_el(env) == 1) {
-@@ -XXX,XX +XXX,XX @@ static bool trans_VLDST_single(DisasContext *s, arg_VLDST_single *a)
++        /* This must be a FEAT_NV access with NVx == 101 */
-         }
++        if (FIELD_EX64(env->cp15.cnthctl_el2, CNTHCTL, EL1NVPCT)) {
-         break;
++            return CP_ACCESS_TRAP_EL2;
-     default:
++        }
--        abort();
++    }
-+        g_assert_not_reached();
++    return e2h_access(env, ri, isread);
-     }
++}
-     if ((vd + a->stride * (nregs - 1)) > 31) {
++
-         /*
++static CPAccessResult access_el1nvvct(CPUARMState *env, const ARMCPRegInfo *ri,
-diff --git a/target/arm/translate.c b/target/arm/translate.c
++                                      bool isread)
-index XXXXXXX..XXXXXXX 100644
++{
---- a/target/arm/translate.c
++    if (arm_current_el(env) == 1) {
-+++ b/target/arm/translate.c
++        /* This must be a FEAT_NV access with NVx == 101 */
-@@ -XXX,XX +XXX,XX @@ static void gen_srs(DisasContext *s,
++        if (FIELD_EX64(env->cp15.cnthctl_el2, CNTHCTL, EL1NVVCT)) {
-         offset = 4;
++            return CP_ACCESS_TRAP_EL2;
-         break;
++        }
-     default:
++    }
--        abort();
++    return e2h_access(env, ri, isread);
-+        g_assert_not_reached();
++}
-     }
++
-     tcg_gen_addi_i32(addr, addr, offset);
+ /* Test if system register redirection is to occur in the current state.  */
-     tmp = load_reg(s, 14);
+ static bool redirect_for_e2h(CPUARMState *env)
-@@ -XXX,XX +XXX,XX @@ static void gen_srs(DisasContext *s,
+ {
-             offset = 0;
+@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo vhe_reginfo[] = {
-             break;
+     { .name = "CNTP_CTL_EL02", .state = ARM_CP_STATE_AA64,
-         default:
+       .opc0 = 3, .opc1 = 5, .crn = 14, .crm = 2, .opc2 = 1,
--            abort();
+       .type = ARM_CP_IO | ARM_CP_ALIAS,
-+            g_assert_not_reached();
+-      .access = PL2_RW, .accessfn = e2h_access,
-         }
++      .access = PL2_RW, .accessfn = access_el1nvpct,
-         tcg_gen_addi_i32(addr, addr, offset);
+       .nv2_redirect_offset = 0x180 | NV2_REDIR_NO_NV1,
-         gen_helper_set_r13_banked(cpu_env, tcg_constant_i32(mode), addr);
+       .fieldoffset = offsetof(CPUARMState, cp15.c14_timer[GTIMER_PHYS].ctl),
        .writefn = gt_phys_ctl_write, .raw_writefn = raw_write },
      { .name = "CNTV_CTL_EL02", .state = ARM_CP_STATE_AA64,
        .opc0 = 3, .opc1 = 5, .crn = 14, .crm = 3, .opc2 = 1,
        .type = ARM_CP_IO | ARM_CP_ALIAS,
 -      .access = PL2_RW, .accessfn = e2h_access,
 +      .access = PL2_RW, .accessfn = access_el1nvvct,
        .nv2_redirect_offset = 0x170 | NV2_REDIR_NO_NV1,
        .fieldoffset = offsetof(CPUARMState, cp15.c14_timer[GTIMER_VIRT].ctl),
        .writefn = gt_virt_ctl_write, .raw_writefn = raw_write },
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo vhe_reginfo[] = {
        .type = ARM_CP_IO | ARM_CP_ALIAS,
        .fieldoffset = offsetof(CPUARMState, cp15.c14_timer[GTIMER_PHYS].cval),
        .nv2_redirect_offset = 0x178 | NV2_REDIR_NO_NV1,
 -      .access = PL2_RW, .accessfn = e2h_access,
 +      .access = PL2_RW, .accessfn = access_el1nvpct,
        .writefn = gt_phys_cval_write, .raw_writefn = raw_write },
      { .name = "CNTV_CVAL_EL02", .state = ARM_CP_STATE_AA64,
        .opc0 = 3, .opc1 = 5, .crn = 14, .crm = 3, .opc2 = 2,
        .type = ARM_CP_IO | ARM_CP_ALIAS,
        .nv2_redirect_offset = 0x168 | NV2_REDIR_NO_NV1,
        .fieldoffset = offsetof(CPUARMState, cp15.c14_timer[GTIMER_VIRT].cval),
 -      .access = PL2_RW, .accessfn = e2h_access,
 +      .access = PL2_RW, .accessfn = access_el1nvvct,
        .writefn = gt_virt_cval_write, .raw_writefn = raw_write },
  #endif
  };
 --
-.25.1
+.34.1

-[PULL 19/23] target/arm: Reformat comments in add_cpreg_to_hashtable
+[PULL 06/14] target/arm: Define CNTPCTSS_EL0 and CNTVCTSS_EL0
-From: Richard Henderson <richard.henderson@linaro.org>
+For FEAT_ECV, new registers CNTPCTSS_EL0 and CNTVCTSS_EL0 are
 defined, which are "self-synchronized" views of the physical and
 virtual counts as seen in the CNTPCT_EL0 and CNTVCT_EL0 registers
 (meaning that no barriers are needed around accesses to them to
 ensure that reads of them do not occur speculatively and out-of-order
 with other instructions).
-Put the block comments into the current coding style.
+For QEMU, all our system registers are self-synchronized, so we can
 simply copy the existing implementation of CNTPCT_EL0 and CNTVCT_EL0
 to the new register encodings.
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
+This means we now implement all the functionality required for
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
+ID_AA64MMFR0_EL1.ECV == 0b0001.
-Message-id: 20220501055028.646596-19-richard.henderson@linaro.org
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
+Message-id: 20240301183219.2424889-7-peter.maydell@linaro.org
 ---
- target/arm/helper.c | 24 +++++++++++++++---------
+ target/arm/helper.c | 43 +++++++++++++++++++++++++++++++++++++++++++
-file changed, 15 insertions(+), 9 deletions(-)
+file changed, 43 insertions(+)
 diff --git a/target/arm/helper.c b/target/arm/helper.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/helper.c
 +++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ CpuDefinitionInfoList *qmp_query_cpu_definitions(Error **errp)
+@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo generic_timer_cp_reginfo[] = {
-     return cpu_list;
+     },
- }
+ };
 +/*
-+ * Private utility function for define_one_arm_cp_reg_with_opaque():
++ * FEAT_ECV adds extra views of CNTVCT_EL0 and CNTPCT_EL0 which
-+ * add a single reginfo struct to the hash table.
++ * are "self-synchronizing". For QEMU all sysregs are self-synchronizing,
 + * so our implementations here are identical to the normal registers.
 + */
- static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
++static const ARMCPRegInfo gen_timer_ecv_cp_reginfo[] = {
-                                    void *opaque, CPState state,
++    { .name = "CNTVCTSS", .cp = 15, .crm = 14, .opc1 = 9,
-                                    CPSecureState secstate,
++      .access = PL0_R, .type = ARM_CP_64BIT | ARM_CP_NO_RAW | ARM_CP_IO,
-                                    int crm, int opc1, int opc2,
++      .accessfn = gt_vct_access,
-                                    const char *name)
++      .readfn = gt_virt_cnt_read, .resetfn = arm_cp_reset_ignore,
- {
++    },
--    /* Private utility function for define_one_arm_cp_reg_with_opaque():
++    { .name = "CNTVCTSS_EL0", .state = ARM_CP_STATE_AA64,
--     * add a single reginfo struct to the hash table.
++      .opc0 = 3, .opc1 = 3, .crn = 14, .crm = 0, .opc2 = 6,
--     */
++      .access = PL0_R, .type = ARM_CP_NO_RAW | ARM_CP_IO,
-     uint32_t key;
++      .accessfn = gt_vct_access, .readfn = gt_virt_cnt_read,
-     ARMCPRegInfo *r2;
++    },
-     bool is64 = r->type & ARM_CP_64BIT;
++    { .name = "CNTPCTSS", .cp = 15, .crm = 14, .opc1 = 8,
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
++      .access = PL0_R, .type = ARM_CP_64BIT | ARM_CP_NO_RAW | ARM_CP_IO,
++      .accessfn = gt_pct_access,
-     isbanked = r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1];
++      .readfn = gt_cnt_read, .resetfn = arm_cp_reset_ignore,
-     if (isbanked) {
++    },
--        /* Register is banked (using both entries in array).
++    { .name = "CNTPCTSS_EL0", .state = ARM_CP_STATE_AA64,
-+        /*
++      .opc0 = 3, .opc1 = 3, .crn = 14, .crm = 0, .opc2 = 5,
-+         * Register is banked (using both entries in array).
++      .access = PL0_R, .type = ARM_CP_NO_RAW | ARM_CP_IO,
-          * Overwriting fieldoffset as the array is only used to define
++      .accessfn = gt_pct_access, .readfn = gt_cnt_read,
-          * banked registers but later only fieldoffset is used.
++    },
-          */
++};
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
++
+ #else
-     if (state == ARM_CP_STATE_AA32) {
-         if (isbanked) {
+ /*
--            /* If the register is banked then we don't need to migrate or
+@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo generic_timer_cp_reginfo[] = {
-+            /*
+     },
-+             * If the register is banked then we don't need to migrate or
+ };
-              * reset the 32-bit instance in certain cases:
-              *
++/*
-              * 1) If the register has both 32-bit and 64-bit instances then we
++ * CNTVCTSS_EL0 has the same trap conditions as CNTVCT_EL0, so it also
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
++ * is exposed to userspace by Linux.
-                 r2->type |= ARM_CP_ALIAS;
++ */
-             }
++static const ARMCPRegInfo gen_timer_ecv_cp_reginfo[] = {
-         } else if ((secstate != r->secure) && !ns) {
++    { .name = "CNTVCTSS_EL0", .state = ARM_CP_STATE_AA64,
--            /* The register is not banked so we only want to allow migration of
++      .opc0 = 3, .opc1 = 3, .crn = 14, .crm = 0, .opc2 = 6,
--             * the non-secure instance.
++      .access = PL0_R, .type = ARM_CP_NO_RAW | ARM_CP_IO,
-+            /*
++      .readfn = gt_virt_cnt_read,
-+             * The register is not banked so we only want to allow migration
++    },
-+             * of the non-secure instance.
++};
-              */
++
-             r2->type |= ARM_CP_ALIAS;
+ #endif
-         }
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
+ static void par_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value)
-         }
+@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
      if (arm_feature(env, ARM_FEATURE_GENERIC_TIMER)) {
          define_arm_cp_regs(cpu, generic_timer_cp_reginfo);
      }
++    if (cpu_isar_feature(aa64_ecv_traps, cpu)) {
--    /* By convention, for wildcarded registers only the first
++        define_arm_cp_regs(cpu, gen_timer_ecv_cp_reginfo);
-+    /*
++    }
-+     * By convention, for wildcarded registers only the first
+     if (arm_feature(env, ARM_FEATURE_VAPA)) {
-      * entry is used for migration; the others are marked as
+         ARMCPRegInfo vapa_cp_reginfo[] = {
-      * ALIAS so we don't try to transfer the register
+             { .name = "PAR", .cp = 15, .crn = 7, .crm = 4, .opc1 = 0, .opc2 = 0,
       * multiple times. Special registers (ie NOP/WFI) are
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
          r2->type |= ARM_CP_ALIAS | ARM_CP_NO_GDB;
      }
 -    /* Check that raw accesses are either forbidden or handled. Note that
 +    /*
 +     * Check that raw accesses are either forbidden or handled. Note that
       * we can't assert this earlier because the setup of fieldoffset for
       * banked registers has to be done first.
       */
 --
-.25.1
+.34.1

-[PULL 21/23] target/arm: Add isar predicates for FEAT_Debugv8p2
+[PULL 07/14] target/arm: Implement FEAT_ECV CNTPOFF_EL2 handling
-From: Richard Henderson <richard.henderson@linaro.org>
+When ID_AA64MMFR0_EL1.ECV is 0b0010, a new register CNTPOFF_EL2 is
 implemented.  This is similar to the existing CNTVOFF_EL2, except
 that it controls a hypervisor-adjustable offset made to the physical
 counter and timer.
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
+Implement the handling for this register, which includes control/trap
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
+bits in SCR_EL3 and CNTHCTL_EL2.
-Message-id: 20220501055028.646596-24-richard.henderson@linaro.org
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
+Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
+Message-id: 20240301183219.2424889-8-peter.maydell@linaro.org
 ---
- target/arm/cpu.h | 15 +++++++++++++++
+ target/arm/cpu-features.h |  5 +++
-file changed, 15 insertions(+)
+ target/arm/cpu.h          |  1 +
  target/arm/helper.c       | 68 +++++++++++++++++++++++++++++++++++++--
  target/arm/trace-events   |  1 +
 files changed, 73 insertions(+), 2 deletions(-)
+diff --git a/target/arm/cpu-features.h b/target/arm/cpu-features.h
+index XXXXXXX..XXXXXXX 100644
+--- a/target/arm/cpu-features.h
++++ b/target/arm/cpu-features.h
+@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_ecv_traps(const ARMISARegisters *id)
+     return FIELD_EX64(id->id_aa64mmfr0, ID_AA64MMFR0, ECV) > 0;
+ }
++static inline bool isar_feature_aa64_ecv(const ARMISARegisters *id)
++{
++    return FIELD_EX64(id->id_aa64mmfr0, ID_AA64MMFR0, ECV) > 1;
++}
++
+ static inline bool isar_feature_aa64_vh(const ARMISARegisters *id)
+ {
+     return FIELD_EX64(id->id_aa64mmfr1, ID_AA64MMFR1, VH) != 0;
 diff --git a/target/arm/cpu.h b/target/arm/cpu.h
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/cpu.h
 +++ b/target/arm/cpu.h
-@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa32_ssbs(const ARMISARegisters *id)
+@@ -XXX,XX +XXX,XX @@ typedef struct CPUArchState {
-     return FIELD_EX32(id->id_pfr2, ID_PFR2, SSBS) != 0;
+         uint64_t c14_cntkctl; /* Timer Control register */
          uint64_t cnthctl_el2; /* Counter/Timer Hyp Control register */
          uint64_t cntvoff_el2; /* Counter Virtual Offset register */
 +        uint64_t cntpoff_el2; /* Counter Physical Offset register */
          ARMGenericTimer c14_timer[NUM_GTIMERS];
          uint32_t c15_cpar; /* XScale Coprocessor Access Register */
          uint32_t c15_ticonfig; /* TI925T configuration byte.  */
 diff --git a/target/arm/helper.c b/target/arm/helper.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/helper.c
 +++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static void scr_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value)
          if (cpu_isar_feature(aa64_rme, cpu)) {
              valid_mask |= SCR_NSE | SCR_GPF;
          }
 +        if (cpu_isar_feature(aa64_ecv, cpu)) {
 +            valid_mask |= SCR_ECVEN;
 +        }
      } else {
          valid_mask &= ~(SCR_RW | SCR_ST);
          if (cpu_isar_feature(aa32_ras, cpu)) {
@@ -XXX,XX +XXX,XX @@ void gt_rme_post_el_change(ARMCPU *cpu, void *ignored)
      gt_update_irq(cpu, GTIMER_PHYS);
  }
-+static inline bool isar_feature_aa32_debugv8p2(const ARMISARegisters *id)
++static uint64_t gt_phys_raw_cnt_offset(CPUARMState *env)
 +{
-+    return FIELD_EX32(id->id_dfr0, ID_DFR0, COPDBG) >= 8;
++    if ((env->cp15.scr_el3 & SCR_ECVEN) &&
 +        FIELD_EX64(env->cp15.cnthctl_el2, CNTHCTL, ECV) &&
 +        arm_is_el2_enabled(env) &&
 +        (arm_hcr_el2_eff(env) & (HCR_E2H | HCR_TGE)) != (HCR_E2H | HCR_TGE)) {
 +        return env->cp15.cntpoff_el2;
 +    }
 +    return 0;
 +}
 +
- /*
++static uint64_t gt_phys_cnt_offset(CPUARMState *env)
   * 64-bit feature tests via id registers.
   */
@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_ssbs(const ARMISARegisters *id)
      return FIELD_EX64(id->id_aa64pfr1, ID_AA64PFR1, SSBS) != 0;
  }
 +static inline bool isar_feature_aa64_debugv8p2(const ARMISARegisters *id)
 +{
-+    return FIELD_EX64(id->id_aa64dfr0, ID_AA64DFR0, DEBUGVER) >= 8;
++    if (arm_current_el(env) >= 2) {
 +        return 0;
 +    }
 +    return gt_phys_raw_cnt_offset(env);
 +}
 +
- static inline bool isar_feature_aa64_sve2(const ARMISARegisters *id)
+ static void gt_recalc_timer(ARMCPU *cpu, int timeridx)
  {
-     return FIELD_EX64(id->id_aa64zfr0, ID_AA64ZFR0, SVEVER) != 0;
+     ARMGenericTimer *gt = &cpu->env.cp15.c14_timer[timeridx];
-@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_any_tts2uxn(const ARMISARegisters *id)
+@@ -XXX,XX +XXX,XX @@ static void gt_recalc_timer(ARMCPU *cpu, int timeridx)
-     return isar_feature_aa64_tts2uxn(id) || isar_feature_aa32_tts2uxn(id);
+          * reset timer to when ISTATUS next has to change
           */
          uint64_t offset = timeridx == GTIMER_VIRT ?
 -                                      cpu->env.cp15.cntvoff_el2 : 0;
 +            cpu->env.cp15.cntvoff_el2 : gt_phys_raw_cnt_offset(&cpu->env);
          uint64_t count = gt_get_countervalue(&cpu->env);
          /* Note that this must be unsigned 64 bit arithmetic: */
          int istatus = count - offset >= gt->cval;
@@ -XXX,XX +XXX,XX @@ static void gt_timer_reset(CPUARMState *env, const ARMCPRegInfo *ri,
  static uint64_t gt_cnt_read(CPUARMState *env, const ARMCPRegInfo *ri)
  {
 -    return gt_get_countervalue(env);
 +    return gt_get_countervalue(env) - gt_phys_cnt_offset(env);
  }
-+static inline bool isar_feature_any_debugv8p2(const ARMISARegisters *id)
+ static uint64_t gt_virt_cnt_offset(CPUARMState *env)
@@ -XXX,XX +XXX,XX @@ static uint64_t gt_tval_read(CPUARMState *env, const ARMCPRegInfo *ri,
      case GTIMER_HYPVIRT:
          offset = gt_virt_cnt_offset(env);
          break;
 +    case GTIMER_PHYS:
 +        offset = gt_phys_cnt_offset(env);
 +        break;
      }
      return (uint32_t)(env->cp15.c14_timer[timeridx].cval -
@@ -XXX,XX +XXX,XX @@ static void gt_tval_write(CPUARMState *env, const ARMCPRegInfo *ri,
      case GTIMER_HYPVIRT:
          offset = gt_virt_cnt_offset(env);
          break;
 +    case GTIMER_PHYS:
 +        offset = gt_phys_cnt_offset(env);
 +        break;
      }
      trace_arm_gt_tval_write(timeridx, value);
@@ -XXX,XX +XXX,XX @@ static void gt_cnthctl_write(CPUARMState *env, const ARMCPRegInfo *ri,
              R_CNTHCTL_EL1NVVCT_MASK |
              R_CNTHCTL_EVNTIS_MASK;
      }
 +    if (cpu_isar_feature(aa64_ecv, cpu)) {
 +        valid_mask |= R_CNTHCTL_ECV_MASK;
 +    }
      /* Clear RES0 bits */
      value &= valid_mask;
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo gen_timer_ecv_cp_reginfo[] = {
      },
  };
 +static CPAccessResult gt_cntpoff_access(CPUARMState *env,
 +                                        const ARMCPRegInfo *ri,
 +                                        bool isread)
 +{
-+    return isar_feature_aa64_debugv8p2(id) || isar_feature_aa32_debugv8p2(id);
++    if (arm_current_el(env) == 2 && !(env->cp15.scr_el3 & SCR_ECVEN)) {
 +        return CP_ACCESS_TRAP_EL3;
 +    }
 +    return CP_ACCESS_OK;
 +}
 +
++static void gt_cntpoff_write(CPUARMState *env, const ARMCPRegInfo *ri,
++                              uint64_t value)
++{
++    ARMCPU *cpu = env_archcpu(env);
++
++    trace_arm_gt_cntpoff_write(value);
++    raw_write(env, ri, value);
++    gt_recalc_timer(cpu, GTIMER_PHYS);
++}
++
++static const ARMCPRegInfo gen_timer_cntpoff_reginfo = {
++    .name = "CNTPOFF_EL2", .state = ARM_CP_STATE_AA64,
++    .opc0 = 3, .opc1 = 4, .crn = 14, .crm = 0, .opc2 = 6,
++    .access = PL2_RW, .type = ARM_CP_IO, .resetvalue = 0,
++    .accessfn = gt_cntpoff_access, .writefn = gt_cntpoff_write,
++    .nv2_redirect_offset = 0x1a8,
++    .fieldoffset = offsetof(CPUARMState, cp15.cntpoff_el2),
++};
+ #else
  /*
-  * Forward to the above feature tests given an ARMCPU pointer.
+@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
-  */
+     if (cpu_isar_feature(aa64_ecv_traps, cpu)) {
          define_arm_cp_regs(cpu, gen_timer_ecv_cp_reginfo);
      }
 +#ifndef CONFIG_USER_ONLY
 +    if (cpu_isar_feature(aa64_ecv, cpu)) {
 +        define_one_arm_cp_reg(cpu, &gen_timer_cntpoff_reginfo);
 +    }
 +#endif
      if (arm_feature(env, ARM_FEATURE_VAPA)) {
          ARMCPRegInfo vapa_cp_reginfo[] = {
              { .name = "PAR", .cp = 15, .crn = 7, .crm = 4, .opc1 = 0, .opc2 = 0,
 diff --git a/target/arm/trace-events b/target/arm/trace-events
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/trace-events
 +++ b/target/arm/trace-events
@@ -XXX,XX +XXX,XX @@ arm_gt_tval_write(int timer, uint64_t value) "gt_tval_write: timer %d value 0x%"
  arm_gt_ctl_write(int timer, uint64_t value) "gt_ctl_write: timer %d value 0x%" PRIx64
  arm_gt_imask_toggle(int timer) "gt_ctl_write: timer %d IMASK toggle"
  arm_gt_cntvoff_write(uint64_t value) "gt_cntvoff_write: value 0x%" PRIx64
 +arm_gt_cntpoff_write(uint64_t value) "gt_cntpoff_write: value 0x%" PRIx64
  arm_gt_update_irq(int timer, int irqstate) "gt_update_irq: timer %d irqstate %d"
  # kvm.c
 --
-.25.1
+.34.1

-[PULL 09/23] target/arm: Name CPState type
+[PULL 08/14] target/arm: Enable FEAT_ECV for 'max' CPU
-From: Richard Henderson <richard.henderson@linaro.org>
+Enable all FEAT_ECV features on the 'max' CPU.
-Give this enum a name and use in ARMCPRegInfo,
+Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
-add_cpreg_to_hashtable and define_one_arm_cp_reg_with_opaque.
+Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
 Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
 Message-id: 20240301183219.2424889-9-peter.maydell@linaro.org
 ---
  docs/system/arm/emulation.rst | 1 +
  target/arm/tcg/cpu64.c        | 1 +
 files changed, 2 insertions(+)
-Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
+diff --git a/docs/system/arm/emulation.rst b/docs/system/arm/emulation.rst
 Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
 Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
 Message-id: 20220501055028.646596-9-richard.henderson@linaro.org
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
 ---
  target/arm/cpregs.h | 6 +++---
  target/arm/helper.c | 6 ++++--
 files changed, 7 insertions(+), 5 deletions(-)
 diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
 index XXXXXXX..XXXXXXX 100644
---- a/target/arm/cpregs.h
+--- a/docs/system/arm/emulation.rst
-+++ b/target/arm/cpregs.h
++++ b/docs/system/arm/emulation.rst
-@@ -XXX,XX +XXX,XX @@ enum {
+@@ -XXX,XX +XXX,XX @@ the following architecture extensions:
-  * Note that we rely on the values of these enums as we iterate through
+ - FEAT_DotProd (Advanced SIMD dot product instructions)
-  * the various states in some places.
+ - FEAT_DoubleFault (Double Fault Extension)
-  */
+ - FEAT_E0PD (Preventing EL0 access to halves of address maps)
--enum {
++- FEAT_ECV (Enhanced Counter Virtualization)
-+typedef enum {
+ - FEAT_EPAC (Enhanced pointer authentication)
-     ARM_CP_STATE_AA32 = 0,
+ - FEAT_ETS (Enhanced Translation Synchronization)
-     ARM_CP_STATE_AA64 = 1,
+ - FEAT_EVT (Enhanced Virtualization Traps)
-     ARM_CP_STATE_BOTH = 2,
+diff --git a/target/arm/tcg/cpu64.c b/target/arm/tcg/cpu64.c
 -};
 +} CPState;
  /*
   * ARM CP register secure state flags.  These flags identify security state
@@ -XXX,XX +XXX,XX @@ struct ARMCPRegInfo {
      uint8_t opc1;
      uint8_t opc2;
      /* Execution state in which this register is visible: ARM_CP_STATE_* */
 -    int state;
 +    CPState state;
      /* Register type: ARM_CP_* bits/values */
      int type;
      /* Access rights: PL*_[RW] */
 diff --git a/target/arm/helper.c b/target/arm/helper.c
 index XXXXXXX..XXXXXXX 100644
---- a/target/arm/helper.c
+--- a/target/arm/tcg/cpu64.c
-+++ b/target/arm/helper.c
++++ b/target/arm/tcg/cpu64.c
-@@ -XXX,XX +XXX,XX @@ CpuDefinitionInfoList *qmp_query_cpu_definitions(Error **errp)
+@@ -XXX,XX +XXX,XX @@ void aarch64_max_tcg_initfn(Object *obj)
- }
+     t = FIELD_DP64(t, ID_AA64MMFR0, TGRAN64_2, 2); /* 64k stage2 supported */
+     t = FIELD_DP64(t, ID_AA64MMFR0, TGRAN4_2, 2);  /*  4k stage2 supported */
- static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
+     t = FIELD_DP64(t, ID_AA64MMFR0, FGT, 1);       /* FEAT_FGT */
--                                   void *opaque, int state, int secstate,
++    t = FIELD_DP64(t, ID_AA64MMFR0, ECV, 2);       /* FEAT_ECV */
-+                                   void *opaque, CPState state, int secstate,
+     cpu->isar.id_aa64mmfr0 = t;
-                                    int crm, int opc1, int opc2,
-                                    const char *name)
+     t = cpu->isar.id_aa64mmfr1;
  {
@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
       * bits; the ARM_CP_64BIT* flag applies only to the AArch32 view of
       * the register, if any.
       */
 -    int crm, opc1, opc2, state;
 +    int crm, opc1, opc2;
      int crmmin = (r->crm == CP_ANY) ? 0 : r->crm;
      int crmmax = (r->crm == CP_ANY) ? 15 : r->crm;
      int opc1min = (r->opc1 == CP_ANY) ? 0 : r->opc1;
      int opc1max = (r->opc1 == CP_ANY) ? 7 : r->opc1;
      int opc2min = (r->opc2 == CP_ANY) ? 0 : r->opc2;
      int opc2max = (r->opc2 == CP_ANY) ? 7 : r->opc2;
 +    CPState state;
 +
      /* 64 bit registers have only CRm and Opc1 fields */
      assert(!((r->type & ARM_CP_64BIT) && (r->opc2 || r->crn)));
      /* op0 only exists in the AArch64 encodings */
 --
-.25.1
+.34.1

-[PULL 02/23] target/arm: Split out cpregs.h
+[PULL 09/14] hw/gpio: Implement STM32L4x5 GPIO
-From: Richard Henderson <richard.henderson@linaro.org>
+From: Inès Varhol <ines.varhol@telecom-paris.fr>
-Move ARMCPRegInfo and all related declarations to a new
+Features supported :
-internal header, out of the public cpu.h.
+- the 8 STM32L4x5 GPIOs are initialized with their reset values
     (except IDR, see below)
 - input mode : setting a pin in input mode "externally" (using input
     irqs) results in an out irq (transmitted to SYSCFG)
 - output mode : setting a bit in ODR sets the corresponding out irq
     (if this line is configured in output mode)
 - pull-up, pull-down
 - push-pull, open-drain
-Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
+Difference with the real GPIOs :
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
+- Alternate Function and Analog mode aren't implemented :
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
+    pins in AF/Analog behave like pins in input mode
-Message-id: 20220501055028.646596-2-richard.henderson@linaro.org
+- floating pins stay at their last value
 - register IDR reset values differ from the real one :
     values are coherent with the other registers reset values
     and the fact that AF/Analog modes aren't implemented
 - setting I/O output speed isn't supported
 - locking port bits isn't supported
 - ADC function isn't supported
 - GPIOH has 16 pins instead of 2 pins
 - writing to registers LCKR, AFRL, AFRH and ASCR is ineffective
 Signed-off-by: Arnaud Minier <arnaud.minier@telecom-paris.fr>
 Signed-off-by: Inès Varhol <ines.varhol@telecom-paris.fr>
 Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
 Acked-by: Alistair Francis <alistair.francis@wdc.com>
 Message-id: 20240305210444.310665-2-ines.varhol@telecom-paris.fr
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
 ---
- target/arm/cpregs.h        | 413 +++++++++++++++++++++++++++++++++++++
+ MAINTAINERS                        |   1 +
- target/arm/cpu.h           | 368 ---------------------------------
+ docs/system/arm/b-l475e-iot01a.rst |   2 +-
- hw/arm/pxa2xx.c            |   1 +
+ include/hw/gpio/stm32l4x5_gpio.h   |  70 +++++
- hw/arm/pxa2xx_pic.c        |   1 +
+ hw/gpio/stm32l4x5_gpio.c           | 477 +++++++++++++++++++++++++++++
- hw/intc/arm_gicv3_cpuif.c  |   1 +
+ hw/gpio/Kconfig                    |   3 +
- hw/intc/arm_gicv3_kvm.c    |   2 +
+ hw/gpio/meson.build                |   1 +
- target/arm/cpu.c           |   1 +
+ hw/gpio/trace-events               |   6 +
- target/arm/cpu64.c         |   1 +
+files changed, 559 insertions(+), 1 deletion(-)
- target/arm/cpu_tcg.c       |   1 +
+ create mode 100644 include/hw/gpio/stm32l4x5_gpio.h
- target/arm/gdbstub.c       |   3 +-
+ create mode 100644 hw/gpio/stm32l4x5_gpio.c
  target/arm/helper.c        |   1 +
  target/arm/op_helper.c     |   1 +
  target/arm/translate-a64.c |   4 +-
  target/arm/translate.c     |   3 +-
 files changed, 427 insertions(+), 374 deletions(-)
  create mode 100644 target/arm/cpregs.h
-diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
+diff --git a/MAINTAINERS b/MAINTAINERS
 index XXXXXXX..XXXXXXX 100644
 --- a/MAINTAINERS
 +++ b/MAINTAINERS
@@ -XXX,XX +XXX,XX @@ F: hw/arm/stm32l4x5_soc.c
  F: hw/misc/stm32l4x5_exti.c
  F: hw/misc/stm32l4x5_syscfg.c
  F: hw/misc/stm32l4x5_rcc.c
 +F: hw/gpio/stm32l4x5_gpio.c
  F: include/hw/*/stm32l4x5_*.h
  B-L475E-IOT01A IoT Node
 diff --git a/docs/system/arm/b-l475e-iot01a.rst b/docs/system/arm/b-l475e-iot01a.rst
 index XXXXXXX..XXXXXXX 100644
 --- a/docs/system/arm/b-l475e-iot01a.rst
 +++ b/docs/system/arm/b-l475e-iot01a.rst
@@ -XXX,XX +XXX,XX @@ Currently B-L475E-IOT01A machine's only supports the following devices:
  - STM32L4x5 EXTI (Extended interrupts and events controller)
  - STM32L4x5 SYSCFG (System configuration controller)
  - STM32L4x5 RCC (Reset and clock control)
 +- STM32L4x5 GPIOs (General-purpose I/Os)
  Missing devices
  """""""""""""""
@@ -XXX,XX +XXX,XX @@ Missing devices
  The B-L475E-IOT01A does *not* support the following devices:
  - Serial ports (UART)
 -- General-purpose I/Os (GPIO)
  - Analog to Digital Converter (ADC)
  - SPI controller
  - Timer controller (TIMER)
 diff --git a/include/hw/gpio/stm32l4x5_gpio.h b/include/hw/gpio/stm32l4x5_gpio.h
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
-+++ b/target/arm/cpregs.h
++++ b/include/hw/gpio/stm32l4x5_gpio.h
 @@ -XXX,XX +XXX,XX @@
 +/*
-+ * QEMU ARM CP Register access and descriptions
++ * STM32L4x5 GPIO (General Purpose Input/Ouput)
 + *
-+ * Copyright (c) 2022 Linaro Ltd
++ * Copyright (c) 2024 Arnaud Minier <arnaud.minier@telecom-paris.fr>
 + * Copyright (c) 2024 Inès Varhol <ines.varhol@telecom-paris.fr>
 + *
-+ * This program is free software; you can redistribute it and/or
++ * SPDX-License-Identifier: GPL-2.0-or-later
 + * modify it under the terms of the GNU General Public License
 + * as published by the Free Software Foundation; either version 2
 + * of the License, or (at your option) any later version.
 + *
-+ * This program is distributed in the hope that it will be useful,
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * See the COPYING file in the top-level directory.
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
++ */
-+ * GNU General Public License for more details.
++
 +/*
 + * The reference used is the STMicroElectronics RM0351 Reference manual
 + * for STM32L4x5 and STM32L4x6 advanced Arm ® -based 32-bit MCUs.
 + * https://www.st.com/en/microcontrollers-microprocessors/stm32l4x5/documentation.html
 + */
 +
 +#ifndef HW_STM32L4X5_GPIO_H
 +#define HW_STM32L4X5_GPIO_H
 +
 +#include "hw/sysbus.h"
 +#include "qom/object.h"
 +
 +#define TYPE_STM32L4X5_GPIO "stm32l4x5-gpio"
 +OBJECT_DECLARE_SIMPLE_TYPE(Stm32l4x5GpioState, STM32L4X5_GPIO)
 +
 +#define GPIO_NUM_PINS 16
 +
 +struct Stm32l4x5GpioState {
 +    SysBusDevice parent_obj;
 +
 +    MemoryRegion mmio;
 +
 +    /* GPIO registers */
 +    uint32_t moder;
 +    uint32_t otyper;
 +    uint32_t ospeedr;
 +    uint32_t pupdr;
 +    uint32_t idr;
 +    uint32_t odr;
 +    uint32_t lckr;
 +    uint32_t afrl;
 +    uint32_t afrh;
 +    uint32_t ascr;
 +
 +    /* GPIO registers reset values */
 +    uint32_t moder_reset;
 +    uint32_t ospeedr_reset;
 +    uint32_t pupdr_reset;
 +
 +    /*
 +     * External driving of pins.
 +     * The pins can be set externally through the device
 +     * anonymous input GPIOs lines under certain conditions.
 +     * The pin must not be in push-pull output mode,
 +     * and can't be set high in open-drain mode.
 +     * Pins driven externally and configured to
 +     * output mode will in general be "disconnected"
 +     * (see `get_gpio_pinmask_to_disconnect()`)
 +     */
 +    uint16_t disconnected_pins;
 +    uint16_t pins_connected_high;
 +
 +    char *name;
 +    Clock *clk;
 +    qemu_irq pin[GPIO_NUM_PINS];
 +};
 +
 +#endif
 diff --git a/hw/gpio/stm32l4x5_gpio.c b/hw/gpio/stm32l4x5_gpio.c
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
 +++ b/hw/gpio/stm32l4x5_gpio.c
@@ -XXX,XX +XXX,XX @@
 +/*
 + * STM32L4x5 GPIO (General Purpose Input/Ouput)
 + *
-+ * You should have received a copy of the GNU General Public License
++ * Copyright (c) 2024 Arnaud Minier <arnaud.minier@telecom-paris.fr>
-+ * along with this program; if not, see
++ * Copyright (c) 2024 Inès Varhol <ines.varhol@telecom-paris.fr>
-+ * <http://www.gnu.org/licenses/gpl-2.0.html>
++ *
 + * SPDX-License-Identifier: GPL-2.0-or-later
 + *
 + * This work is licensed under the terms of the GNU GPL, version 2 or later.
 + * See the COPYING file in the top-level directory.
 + */
 +
-+#ifndef TARGET_ARM_CPREGS_H
-+#define TARGET_ARM_CPREGS_H
-+
 +/*
-+ * ARMCPRegInfo type field bits. If the SPECIAL bit is set this is a
++ * The reference used is the STMicroElectronics RM0351 Reference manual
-+ * special-behaviour cp reg and bits [11..8] indicate what behaviour
++ * for STM32L4x5 and STM32L4x6 advanced Arm ® -based 32-bit MCUs.
-+ * it has. Otherwise it is a simple cp reg, where CONST indicates that
++ * https://www.st.com/en/microcontrollers-microprocessors/stm32l4x5/documentation.html
 + * TCG can assume the value to be constant (ie load at translate time)
 + * and 64BIT indicates a 64 bit wide coprocessor register. SUPPRESS_TB_END
 + * indicates that the TB should not be ended after a write to this register
 + * (the default is that the TB ends after cp writes). OVERRIDE permits
 + * a register definition to override a previous definition for the
 + * same (cp, is64, crn, crm, opc1, opc2) tuple: either the new or the
 + * old must have the OVERRIDE bit set.
 + * ALIAS indicates that this register is an alias view of some underlying
 + * state which is also visible via another register, and that the other
 + * register is handling migration and reset; registers marked ALIAS will not be
 + * migrated but may have their state set by syncing of register state from KVM.
 + * NO_RAW indicates that this register has no underlying state and does not
 + * support raw access for state saving/loading; it will not be used for either
 + * migration or KVM state synchronization. (Typically this is for "registers"
 + * which are actually used as instructions for cache maintenance and so on.)
 + * IO indicates that this register does I/O and therefore its accesses
 + * need to be marked with gen_io_start() and also end the TB. In particular,
 + * registers which implement clocks or timers require this.
 + * RAISES_EXC is for when the read or write hook might raise an exception;
 + * the generated code will synchronize the CPU state before calling the hook
 + * so that it is safe for the hook to call raise_exception().
 + * NEWEL is for writes to registers that might change the exception
 + * level - typically on older ARM chips. For those cases we need to
 + * re-read the new el when recomputing the translation flags.
 + */
-+#define ARM_CP_SPECIAL           0x0001
++
-+#define ARM_CP_CONST             0x0002
++#include "qemu/osdep.h"
-+#define ARM_CP_64BIT             0x0004
++#include "qemu/log.h"
-+#define ARM_CP_SUPPRESS_TB_END   0x0008
++#include "hw/gpio/stm32l4x5_gpio.h"
-+#define ARM_CP_OVERRIDE          0x0010
++#include "hw/irq.h"
-+#define ARM_CP_ALIAS             0x0020
++#include "hw/qdev-clock.h"
-+#define ARM_CP_IO                0x0040
++#include "hw/qdev-properties.h"
-+#define ARM_CP_NO_RAW            0x0080
++#include "qapi/visitor.h"
-+#define ARM_CP_NOP               (ARM_CP_SPECIAL | 0x0100)
++#include "qapi/error.h"
-+#define ARM_CP_WFI               (ARM_CP_SPECIAL | 0x0200)
++#include "migration/vmstate.h"
-+#define ARM_CP_NZCV              (ARM_CP_SPECIAL | 0x0300)
++#include "trace.h"
-+#define ARM_CP_CURRENTEL         (ARM_CP_SPECIAL | 0x0400)
++
-+#define ARM_CP_DC_ZVA            (ARM_CP_SPECIAL | 0x0500)
++#define GPIO_MODER 0x00
-+#define ARM_CP_DC_GVA            (ARM_CP_SPECIAL | 0x0600)
++#define GPIO_OTYPER 0x04
-+#define ARM_CP_DC_GZVA           (ARM_CP_SPECIAL | 0x0700)
++#define GPIO_OSPEEDR 0x08
-+#define ARM_LAST_SPECIAL         ARM_CP_DC_GZVA
++#define GPIO_PUPDR 0x0C
-+#define ARM_CP_FPU               0x1000
++#define GPIO_IDR 0x10
-+#define ARM_CP_SVE               0x2000
++#define GPIO_ODR 0x14
-+#define ARM_CP_NO_GDB            0x4000
++#define GPIO_BSRR 0x18
-+#define ARM_CP_RAISES_EXC        0x8000
++#define GPIO_LCKR 0x1C
-+#define ARM_CP_NEWEL             0x10000
++#define GPIO_AFRL 0x20
-+/* Used only as a terminator for ARMCPRegInfo lists */
++#define GPIO_AFRH 0x24
-+#define ARM_CP_SENTINEL          0xfffff
++#define GPIO_BRR 0x28
-+/* Mask of only the flag bits in a type field */
++#define GPIO_ASCR 0x2C
-+#define ARM_CP_FLAG_MASK         0x1f0ff
++
 +/* 0b11111111_11111111_00000000_00000000 */
 +#define RESERVED_BITS_MASK 0xFFFF0000
 +
 +static void update_gpio_idr(Stm32l4x5GpioState *s);
 +
 +static bool is_pull_up(Stm32l4x5GpioState *s, unsigned pin)
 +{
 +    return extract32(s->pupdr, 2 * pin, 2) == 1;
 +}
 +
 +static bool is_pull_down(Stm32l4x5GpioState *s, unsigned pin)
 +{
 +    return extract32(s->pupdr, 2 * pin, 2) == 2;
 +}
 +
 +static bool is_output(Stm32l4x5GpioState *s, unsigned pin)
 +{
 +    return extract32(s->moder, 2 * pin, 2) == 1;
 +}
 +
 +static bool is_open_drain(Stm32l4x5GpioState *s, unsigned pin)
 +{
 +    return extract32(s->otyper, pin, 1) == 1;
 +}
 +
 +static bool is_push_pull(Stm32l4x5GpioState *s, unsigned pin)
 +{
 +    return extract32(s->otyper, pin, 1) == 0;
 +}
 +
 +static void stm32l4x5_gpio_reset_hold(Object *obj)
 +{
 +    Stm32l4x5GpioState *s = STM32L4X5_GPIO(obj);
 +
 +    s->moder = s->moder_reset;
 +    s->otyper = 0x00000000;
 +    s->ospeedr = s->ospeedr_reset;
 +    s->pupdr = s->pupdr_reset;
 +    s->idr = 0x00000000;
 +    s->odr = 0x00000000;
 +    s->lckr = 0x00000000;
 +    s->afrl = 0x00000000;
 +    s->afrh = 0x00000000;
 +    s->ascr = 0x00000000;
 +
 +    s->disconnected_pins = 0xFFFF;
 +    s->pins_connected_high = 0x0000;
 +    update_gpio_idr(s);
 +}
 +
 +static void stm32l4x5_gpio_set(void *opaque, int line, int level)
 +{
 +    Stm32l4x5GpioState *s = opaque;
 +    /*
 +     * The pin isn't set if line is configured in output mode
 +     * except if level is 0 and the output is open-drain.
 +     * This way there will be no short-circuit prone situations.
 +     */
 +    if (is_output(s, line) && !(is_open_drain(s, line) && (level == 0))) {
 +        qemu_log_mask(LOG_GUEST_ERROR, "Line %d can't be driven externally\n",
 +                      line);
 +        return;
 +    }
 +
 +    s->disconnected_pins &= ~(1 << line);
 +    if (level) {
 +        s->pins_connected_high |= (1 << line);
 +    } else {
 +        s->pins_connected_high &= ~(1 << line);
 +    }
 +    trace_stm32l4x5_gpio_pins(s->name, s->disconnected_pins,
 +                              s->pins_connected_high);
 +    update_gpio_idr(s);
 +}
 +
 +
 +static void update_gpio_idr(Stm32l4x5GpioState *s)
 +{
 +    uint32_t new_idr_mask = 0;
 +    uint32_t new_idr = s->odr;
 +    uint32_t old_idr = s->idr;
 +    int new_pin_state, old_pin_state;
 +
 +    for (int i = 0; i < GPIO_NUM_PINS; i++) {
 +        if (is_output(s, i)) {
 +            if (is_push_pull(s, i)) {
 +                new_idr_mask |= (1 << i);
 +            } else if (!(s->odr & (1 << i))) {
 +                /* open-drain ODR 0 */
 +                new_idr_mask |= (1 << i);
 +            /* open-drain ODR 1 */
 +            } else if (!(s->disconnected_pins & (1 << i)) &&
 +                       !(s->pins_connected_high & (1 << i))) {
 +                /* open-drain ODR 1 with pin connected low */
 +                new_idr_mask |= (1 << i);
 +                new_idr &= ~(1 << i);
 +            /* open-drain ODR 1 with unactive pin */
 +            } else if (is_pull_up(s, i)) {
 +                new_idr_mask |= (1 << i);
 +            } else if (is_pull_down(s, i)) {
 +                new_idr_mask |= (1 << i);
 +                new_idr &= ~(1 << i);
 +            }
 +            /*
 +             * The only case left is for open-drain ODR 1
 +             * with unactive pin without pull-up or pull-down :
 +             * the value is floating.
 +             */
 +        /* input or analog mode with connected pin */
 +        } else if (!(s->disconnected_pins & (1 << i))) {
 +            if (s->pins_connected_high & (1 << i)) {
 +                /* pin high */
 +                new_idr_mask |= (1 << i);
 +                new_idr |= (1 << i);
 +            } else {
 +                /* pin low */
 +                new_idr_mask |= (1 << i);
 +                new_idr &= ~(1 << i);
 +            }
 +        /* input or analog mode with disconnected pin */
 +        } else {
 +            if (is_pull_up(s, i)) {
 +                /* pull-up */
 +                new_idr_mask |= (1 << i);
 +                new_idr |= (1 << i);
 +            } else if (is_pull_down(s, i)) {
 +                /* pull-down */
 +                new_idr_mask |= (1 << i);
 +                new_idr &= ~(1 << i);
 +            }
 +            /*
 +             * The only case left is for a disconnected pin
 +             * without pull-up or pull-down :
 +             * the value is floating.
 +             */
 +        }
 +    }
 +
 +    s->idr = (old_idr & ~new_idr_mask) | (new_idr & new_idr_mask);
 +    trace_stm32l4x5_gpio_update_idr(s->name, old_idr, s->idr);
 +
 +    for (int i = 0; i < GPIO_NUM_PINS; i++) {
 +        if (new_idr_mask & (1 << i)) {
 +            new_pin_state = (new_idr & (1 << i)) > 0;
 +            old_pin_state = (old_idr & (1 << i)) > 0;
 +            if (new_pin_state > old_pin_state) {
 +                qemu_irq_raise(s->pin[i]);
 +            } else if (new_pin_state < old_pin_state) {
 +                qemu_irq_lower(s->pin[i]);
 +            }
 +        }
 +    }
 +}
 +
 +/*
-+ * Valid values for ARMCPRegInfo state field, indicating which of
++ * Return mask of pins that are both configured in output
-+ * the AArch32 and AArch64 execution states this register is visible in.
++ * mode and externally driven (except pins in open-drain
-+ * If the reginfo doesn't explicitly specify then it is AArch32 only.
++ * mode externally set to 0).
 + * If the reginfo is declared to be visible in both states then a second
 + * reginfo is synthesised for the AArch32 view of the AArch64 register,
 + * such that the AArch32 view is the lower 32 bits of the AArch64 one.
 + * Note that we rely on the values of these enums as we iterate through
 + * the various states in some places.
 + */
-+enum {
++static uint32_t get_gpio_pinmask_to_disconnect(Stm32l4x5GpioState *s)
-+    ARM_CP_STATE_AA32 = 0,
++{
-+    ARM_CP_STATE_AA64 = 1,
++    uint32_t pins_to_disconnect = 0;
-+    ARM_CP_STATE_BOTH = 2,
++    for (int i = 0; i < GPIO_NUM_PINS; i++) {
 +        /* for each connected pin in output mode */
 +        if (!(s->disconnected_pins & (1 << i)) && is_output(s, i)) {
 +            /* if either push-pull or high level */
 +            if (is_push_pull(s, i) || s->pins_connected_high & (1 << i)) {
 +                pins_to_disconnect |= (1 << i);
 +                qemu_log_mask(LOG_GUEST_ERROR,
 +                              "Line %d can't be driven externally\n",
 +                              i);
 +            }
 +        }
 +    }
 +    return pins_to_disconnect;
 +}
 +
 +/*
 + * Set field `disconnected_pins` and call `update_gpio_idr()`
 + */
 +static void disconnect_gpio_pins(Stm32l4x5GpioState *s, uint16_t lines)
 +{
 +    s->disconnected_pins |= lines;
 +    trace_stm32l4x5_gpio_pins(s->name, s->disconnected_pins,
 +                              s->pins_connected_high);
 +    update_gpio_idr(s);
 +}
 +
 +static void disconnected_pins_set(Object *obj, Visitor *v,
 +    const char *name, void *opaque, Error **errp)
 +{
 +    Stm32l4x5GpioState *s = STM32L4X5_GPIO(obj);
 +    uint16_t value;
 +    if (!visit_type_uint16(v, name, &value, errp)) {
 +        return;
 +    }
 +    disconnect_gpio_pins(s, value);
 +}
 +
 +static void disconnected_pins_get(Object *obj, Visitor *v,
 +    const char *name, void *opaque, Error **errp)
 +{
 +    visit_type_uint16(v, name, (uint16_t *)opaque, errp);
 +}
 +
 +static void clock_freq_get(Object *obj, Visitor *v,
 +    const char *name, void *opaque, Error **errp)
 +{
 +    Stm32l4x5GpioState *s = STM32L4X5_GPIO(obj);
 +    uint32_t clock_freq_hz = clock_get_hz(s->clk);
 +    visit_type_uint32(v, name, &clock_freq_hz, errp);
 +}
 +
 +static void stm32l4x5_gpio_write(void *opaque, hwaddr addr,
 +                                 uint64_t val64, unsigned int size)
 +{
 +    Stm32l4x5GpioState *s = opaque;
 +
 +    uint32_t value = val64;
 +    trace_stm32l4x5_gpio_write(s->name, addr, val64);
 +
 +    switch (addr) {
 +    case GPIO_MODER:
 +        s->moder = value;
 +        disconnect_gpio_pins(s, get_gpio_pinmask_to_disconnect(s));
 +        qemu_log_mask(LOG_UNIMP,
 +                      "%s: Analog and AF modes aren't supported\n\
 +                       Analog and AF mode behave like input mode\n",
 +                      __func__);
 +        return;
 +    case GPIO_OTYPER:
 +        s->otyper = value & ~RESERVED_BITS_MASK;
 +        disconnect_gpio_pins(s, get_gpio_pinmask_to_disconnect(s));
 +        return;
 +    case GPIO_OSPEEDR:
 +        qemu_log_mask(LOG_UNIMP,
 +                      "%s: Changing I/O output speed isn't supported\n\
 +                       I/O speed is already maximal\n",
 +                      __func__);
 +        s->ospeedr = value;
 +        return;
 +    case GPIO_PUPDR:
 +        s->pupdr = value;
 +        update_gpio_idr(s);
 +        return;
 +    case GPIO_IDR:
 +        qemu_log_mask(LOG_UNIMP,
 +                      "%s: GPIO->IDR is read-only\n",
 +                      __func__);
 +        return;
 +    case GPIO_ODR:
 +        s->odr = value & ~RESERVED_BITS_MASK;
 +        update_gpio_idr(s);
 +        return;
 +    case GPIO_BSRR: {
 +        uint32_t bits_to_reset = (value & RESERVED_BITS_MASK) >> GPIO_NUM_PINS;
 +        uint32_t bits_to_set = value & ~RESERVED_BITS_MASK;
 +        /* If both BSx and BRx are set, BSx has priority.*/
 +        s->odr &= ~bits_to_reset;
 +        s->odr |= bits_to_set;
 +        update_gpio_idr(s);
 +        return;
 +    }
 +    case GPIO_LCKR:
 +        qemu_log_mask(LOG_UNIMP,
 +                      "%s: Locking port bits configuration isn't supported\n",
 +                      __func__);
 +        s->lckr = value & ~RESERVED_BITS_MASK;
 +        return;
 +    case GPIO_AFRL:
 +        qemu_log_mask(LOG_UNIMP,
 +                      "%s: Alternate functions aren't supported\n",
 +                      __func__);
 +        s->afrl = value;
 +        return;
 +    case GPIO_AFRH:
 +        qemu_log_mask(LOG_UNIMP,
 +                      "%s: Alternate functions aren't supported\n",
 +                      __func__);
 +        s->afrh = value;
 +        return;
 +    case GPIO_BRR: {
 +        uint32_t bits_to_reset = value & ~RESERVED_BITS_MASK;
 +        s->odr &= ~bits_to_reset;
 +        update_gpio_idr(s);
 +        return;
 +    }
 +    case GPIO_ASCR:
 +        qemu_log_mask(LOG_UNIMP,
 +                      "%s: ADC function isn't supported\n",
 +                      __func__);
 +        s->ascr = value & ~RESERVED_BITS_MASK;
 +        return;
 +    default:
 +        qemu_log_mask(LOG_GUEST_ERROR,
 +                      "%s: Bad offset 0x%" HWADDR_PRIx "\n", __func__, addr);
 +    }
 +}
 +
 +static uint64_t stm32l4x5_gpio_read(void *opaque, hwaddr addr,
 +                                    unsigned int size)
 +{
 +    Stm32l4x5GpioState *s = opaque;
 +
 +    trace_stm32l4x5_gpio_read(s->name, addr);
 +
 +    switch (addr) {
 +    case GPIO_MODER:
 +        return s->moder;
 +    case GPIO_OTYPER:
 +        return s->otyper;
 +    case GPIO_OSPEEDR:
 +        return s->ospeedr;
 +    case GPIO_PUPDR:
 +        return s->pupdr;
 +    case GPIO_IDR:
 +        return s->idr;
 +    case GPIO_ODR:
 +        return s->odr;
 +    case GPIO_BSRR:
 +        return 0;
 +    case GPIO_LCKR:
 +        return s->lckr;
 +    case GPIO_AFRL:
 +        return s->afrl;
 +    case GPIO_AFRH:
 +        return s->afrh;
 +    case GPIO_BRR:
 +        return 0;
 +    case GPIO_ASCR:
 +        return s->ascr;
 +    default:
 +        qemu_log_mask(LOG_GUEST_ERROR,
 +                      "%s: Bad offset 0x%" HWADDR_PRIx "\n", __func__, addr);
 +        return 0;
 +    }
 +}
 +
 +static const MemoryRegionOps stm32l4x5_gpio_ops = {
 +    .read = stm32l4x5_gpio_read,
 +    .write = stm32l4x5_gpio_write,
 +    .endianness = DEVICE_NATIVE_ENDIAN,
 +    .impl = {
 +        .min_access_size = 4,
 +        .max_access_size = 4,
 +        .unaligned = false,
 +    },
 +    .valid = {
 +        .min_access_size = 4,
 +        .max_access_size = 4,
 +        .unaligned = false,
 +    },
 +};
 +
-+/*
++static void stm32l4x5_gpio_init(Object *obj)
-+ * ARM CP register secure state flags.  These flags identify security state
++{
-+ * attributes for a given CP register entry.
++    Stm32l4x5GpioState *s = STM32L4X5_GPIO(obj);
-+ * The existence of both or neither secure and non-secure flags indicates that
++
-+ * the register has both a secure and non-secure hash entry.  A single one of
++    memory_region_init_io(&s->mmio, obj, &stm32l4x5_gpio_ops, s,
-+ * these flags causes the register to only be hashed for the specified
++                          TYPE_STM32L4X5_GPIO, 0x400);
-+ * security state.
++
-+ * Although definitions may have any combination of the S/NS bits, each
++    sysbus_init_mmio(SYS_BUS_DEVICE(obj), &s->mmio);
-+ * registered entry will only have one to identify whether the entry is secure
++
-+ * or non-secure.
++    qdev_init_gpio_out(DEVICE(obj), s->pin, GPIO_NUM_PINS);
-+ */
++    qdev_init_gpio_in(DEVICE(obj), stm32l4x5_gpio_set, GPIO_NUM_PINS);
-+enum {
++
-+    ARM_CP_SECSTATE_S =   (1 << 0), /* bit[0]: Secure state register */
++    s->clk = qdev_init_clock_in(DEVICE(s), "clk", NULL, s, 0);
-+    ARM_CP_SECSTATE_NS =  (1 << 1), /* bit[1]: Non-secure state register */
++
 +    object_property_add(obj, "disconnected-pins", "uint16",
 +                        disconnected_pins_get, disconnected_pins_set,
 +                        NULL, &s->disconnected_pins);
 +    object_property_add(obj, "clock-freq-hz", "uint32",
 +                        clock_freq_get, NULL, NULL, NULL);
 +}
 +
 +static void stm32l4x5_gpio_realize(DeviceState *dev, Error **errp)
 +{
 +    Stm32l4x5GpioState *s = STM32L4X5_GPIO(dev);
 +    if (!clock_has_source(s->clk)) {
 +        error_setg(errp, "GPIO: clk input must be connected");
 +        return;
 +    }
 +}
 +
 +static const VMStateDescription vmstate_stm32l4x5_gpio = {
 +    .name = TYPE_STM32L4X5_GPIO,
 +    .version_id = 1,
 +    .minimum_version_id = 1,
 +    .fields = (VMStateField[]){
 +        VMSTATE_UINT32(moder, Stm32l4x5GpioState),
 +        VMSTATE_UINT32(otyper, Stm32l4x5GpioState),
 +        VMSTATE_UINT32(ospeedr, Stm32l4x5GpioState),
 +        VMSTATE_UINT32(pupdr, Stm32l4x5GpioState),
 +        VMSTATE_UINT32(idr, Stm32l4x5GpioState),
 +        VMSTATE_UINT32(odr, Stm32l4x5GpioState),
 +        VMSTATE_UINT32(lckr, Stm32l4x5GpioState),
 +        VMSTATE_UINT32(afrl, Stm32l4x5GpioState),
 +        VMSTATE_UINT32(afrh, Stm32l4x5GpioState),
 +        VMSTATE_UINT32(ascr, Stm32l4x5GpioState),
 +        VMSTATE_UINT16(disconnected_pins, Stm32l4x5GpioState),
 +        VMSTATE_UINT16(pins_connected_high, Stm32l4x5GpioState),
 +        VMSTATE_END_OF_LIST()
 +    }
 +};
 +
-+/*
++static Property stm32l4x5_gpio_properties[] = {
-+ * Return true if cptype is a valid type field. This is used to try to
++    DEFINE_PROP_STRING("name", Stm32l4x5GpioState, name),
-+ * catch errors where the sentinel has been accidentally left off the end
++    DEFINE_PROP_UINT32("mode-reset", Stm32l4x5GpioState, moder_reset, 0),
-+ * of a list of registers.
++    DEFINE_PROP_UINT32("ospeed-reset", Stm32l4x5GpioState, ospeedr_reset, 0),
-+ */
++    DEFINE_PROP_UINT32("pupd-reset", Stm32l4x5GpioState, pupdr_reset, 0),
-+static inline bool cptype_valid(int cptype)
++    DEFINE_PROP_END_OF_LIST(),
 +{
 +    return ((cptype & ~ARM_CP_FLAG_MASK) == 0)
 +        || ((cptype & ARM_CP_SPECIAL) &&
 +            ((cptype & ~ARM_CP_FLAG_MASK) <= ARM_LAST_SPECIAL));
 +}
 +
 +/*
 + * Access rights:
 + * We define bits for Read and Write access for what rev C of the v7-AR ARM ARM
 + * defines as PL0 (user), PL1 (fiq/irq/svc/abt/und/sys, ie privileged), and
 + * PL2 (hyp). The other level which has Read and Write bits is Secure PL1
 + * (ie any of the privileged modes in Secure state, or Monitor mode).
 + * If a register is accessible in one privilege level it's always accessible
 + * in higher privilege levels too. Since "Secure PL1" also follows this rule
 + * (ie anything visible in PL2 is visible in S-PL1, some things are only
 + * visible in S-PL1) but "Secure PL1" is a bit of a mouthful, we bend the
 + * terminology a little and call this PL3.
 + * In AArch64 things are somewhat simpler as the PLx bits line up exactly
 + * with the ELx exception levels.
 + *
 + * If access permissions for a register are more complex than can be
 + * described with these bits, then use a laxer set of restrictions, and
 + * do the more restrictive/complex check inside a helper function.
 + */
 +#define PL3_R 0x80
 +#define PL3_W 0x40
 +#define PL2_R (0x20 | PL3_R)
 +#define PL2_W (0x10 | PL3_W)
 +#define PL1_R (0x08 | PL2_R)
 +#define PL1_W (0x04 | PL2_W)
 +#define PL0_R (0x02 | PL1_R)
 +#define PL0_W (0x01 | PL1_W)
 +
 +/*
 + * For user-mode some registers are accessible to EL0 via a kernel
 + * trap-and-emulate ABI. In this case we define the read permissions
 + * as actually being PL0_R. However some bits of any given register
 + * may still be masked.
 + */
 +#ifdef CONFIG_USER_ONLY
 +#define PL0U_R PL0_R
 +#else
 +#define PL0U_R PL1_R
 +#endif
 +
 +#define PL3_RW (PL3_R | PL3_W)
 +#define PL2_RW (PL2_R | PL2_W)
 +#define PL1_RW (PL1_R | PL1_W)
 +#define PL0_RW (PL0_R | PL0_W)
 +
 +typedef enum CPAccessResult {
 +    /* Access is permitted */
 +    CP_ACCESS_OK = 0,
 +    /*
 +     * Access fails due to a configurable trap or enable which would
 +     * result in a categorized exception syndrome giving information about
 +     * the failing instruction (ie syndrome category 0x3, 0x4, 0x5, 0x6,
 +     * 0xc or 0x18). The exception is taken to the usual target EL (EL1 or
 +     * PL1 if in EL0, otherwise to the current EL).
 +     */
 +    CP_ACCESS_TRAP = 1,
 +    /*
 +     * Access fails and results in an exception syndrome 0x0 ("uncategorized").
 +     * Note that this is not a catch-all case -- the set of cases which may
 +     * result in this failure is specifically defined by the architecture.
 +     */
 +    CP_ACCESS_TRAP_UNCATEGORIZED = 2,
 +    /* As CP_ACCESS_TRAP, but for traps directly to EL2 or EL3 */
 +    CP_ACCESS_TRAP_EL2 = 3,
 +    CP_ACCESS_TRAP_EL3 = 4,
 +    /* As CP_ACCESS_UNCATEGORIZED, but for traps directly to EL2 or EL3 */
 +    CP_ACCESS_TRAP_UNCATEGORIZED_EL2 = 5,
 +    CP_ACCESS_TRAP_UNCATEGORIZED_EL3 = 6,
 +} CPAccessResult;
 +
 +typedef struct ARMCPRegInfo ARMCPRegInfo;
 +
 +/*
 + * Access functions for coprocessor registers. These cannot fail and
 + * may not raise exceptions.
 + */
 +typedef uint64_t CPReadFn(CPUARMState *env, const ARMCPRegInfo *opaque);
 +typedef void CPWriteFn(CPUARMState *env, const ARMCPRegInfo *opaque,
 +                       uint64_t value);
 +/* Access permission check functions for coprocessor registers. */
 +typedef CPAccessResult CPAccessFn(CPUARMState *env,
 +                                  const ARMCPRegInfo *opaque,
 +                                  bool isread);
 +/* Hook function for register reset */
 +typedef void CPResetFn(CPUARMState *env, const ARMCPRegInfo *opaque);
 +
 +#define CP_ANY 0xff
 +
 +/* Definition of an ARM coprocessor register */
 +struct ARMCPRegInfo {
 +    /* Name of register (useful mainly for debugging, need not be unique) */
 +    const char *name;
 +    /*
 +     * Location of register: coprocessor number and (crn,crm,opc1,opc2)
 +     * tuple. Any of crm, opc1 and opc2 may be CP_ANY to indicate a
 +     * 'wildcard' field -- any value of that field in the MRC/MCR insn
 +     * will be decoded to this register. The register read and write
 +     * callbacks will be passed an ARMCPRegInfo with the crn/crm/opc1/opc2
 +     * used by the program, so it is possible to register a wildcard and
 +     * then behave differently on read/write if necessary.
 +     * For 64 bit registers, only crm and opc1 are relevant; crn and opc2
 +     * must both be zero.
 +     * For AArch64-visible registers, opc0 is also used.
 +     * Since there are no "coprocessors" in AArch64, cp is purely used as a
 +     * way to distinguish (for KVM's benefit) guest-visible system registers
 +     * from demuxed ones provided to preserve the "no side effects on
 +     * KVM register read/write from QEMU" semantics. cp==0x13 is guest
 +     * visible (to match KVM's encoding); cp==0 will be converted to
 +     * cp==0x13 when the ARMCPRegInfo is registered, for convenience.
 +     */
 +    uint8_t cp;
 +    uint8_t crn;
 +    uint8_t crm;
 +    uint8_t opc0;
 +    uint8_t opc1;
 +    uint8_t opc2;
 +    /* Execution state in which this register is visible: ARM_CP_STATE_* */
 +    int state;
 +    /* Register type: ARM_CP_* bits/values */
 +    int type;
 +    /* Access rights: PL*_[RW] */
 +    int access;
 +    /* Security state: ARM_CP_SECSTATE_* bits/values */
 +    int secure;
 +    /*
 +     * The opaque pointer passed to define_arm_cp_regs_with_opaque() when
 +     * this register was defined: can be used to hand data through to the
 +     * register read/write functions, since they are passed the ARMCPRegInfo*.
 +     */
 +    void *opaque;
 +    /*
 +     * Value of this register, if it is ARM_CP_CONST. Otherwise, if
 +     * fieldoffset is non-zero, the reset value of the register.
 +     */
 +    uint64_t resetvalue;
 +    /*
 +     * Offset of the field in CPUARMState for this register.
 +     * This is not needed if either:
 +     *  1. type is ARM_CP_CONST or one of the ARM_CP_SPECIALs
 +     *  2. both readfn and writefn are specified
 +     */
 +    ptrdiff_t fieldoffset; /* offsetof(CPUARMState, field) */
 +
 +    /*
 +     * Offsets of the secure and non-secure fields in CPUARMState for the
 +     * register if it is banked.  These fields are only used during the static
 +     * registration of a register.  During hashing the bank associated
 +     * with a given security state is copied to fieldoffset which is used from
 +     * there on out.
 +     *
 +     * It is expected that register definitions use either fieldoffset or
 +     * bank_fieldoffsets in the definition but not both.  It is also expected
 +     * that both bank offsets are set when defining a banked register.  This
 +     * use indicates that a register is banked.
 +     */
 +    ptrdiff_t bank_fieldoffsets[2];
 +
 +    /*
 +     * Function for making any access checks for this register in addition to
 +     * those specified by the 'access' permissions bits. If NULL, no extra
 +     * checks required. The access check is performed at runtime, not at
 +     * translate time.
 +     */
 +    CPAccessFn *accessfn;
 +    /*
 +     * Function for handling reads of this register. If NULL, then reads
 +     * will be done by loading from the offset into CPUARMState specified
 +     * by fieldoffset.
 +     */
 +    CPReadFn *readfn;
 +    /*
 +     * Function for handling writes of this register. If NULL, then writes
 +     * will be done by writing to the offset into CPUARMState specified
 +     * by fieldoffset.
 +     */
 +    CPWriteFn *writefn;
 +    /*
 +     * Function for doing a "raw" read; used when we need to copy
 +     * coprocessor state to the kernel for KVM or out for
 +     * migration. This only needs to be provided if there is also a
 +     * readfn and it has side effects (for instance clear-on-read bits).
 +     */
 +    CPReadFn *raw_readfn;
 +    /*
 +     * Function for doing a "raw" write; used when we need to copy KVM
 +     * kernel coprocessor state into userspace, or for inbound
 +     * migration. This only needs to be provided if there is also a
 +     * writefn and it masks out "unwritable" bits or has write-one-to-clear
 +     * or similar behaviour.
 +     */
 +    CPWriteFn *raw_writefn;
 +    /*
 +     * Function for resetting the register. If NULL, then reset will be done
 +     * by writing resetvalue to the field specified in fieldoffset. If
 +     * fieldoffset is 0 then no reset will be done.
 +     */
 +    CPResetFn *resetfn;
 +
 +    /*
 +     * "Original" writefn and readfn.
 +     * For ARMv8.1-VHE register aliases, we overwrite the read/write
 +     * accessor functions of various EL1/EL0 to perform the runtime
 +     * check for which sysreg should actually be modified, and then
 +     * forwards the operation.  Before overwriting the accessors,
 +     * the original function is copied here, so that accesses that
 +     * really do go to the EL1/EL0 version proceed normally.
 +     * (The corresponding EL2 register is linked via opaque.)
 +     */
 +    CPReadFn *orig_readfn;
 +    CPWriteFn *orig_writefn;
 +};
 +
-+/*
++static void stm32l4x5_gpio_class_init(ObjectClass *klass, void *data)
-+ * Macros which are lvalues for the field in CPUARMState for the
++{
-+ * ARMCPRegInfo *ri.
++    DeviceClass *dc = DEVICE_CLASS(klass);
-+ */
++    ResettableClass *rc = RESETTABLE_CLASS(klass);
-+#define CPREG_FIELD32(env, ri) \
++
-+    (*(uint32_t *)((char *)(env) + (ri)->fieldoffset))
++    device_class_set_props(dc, stm32l4x5_gpio_properties);
-+#define CPREG_FIELD64(env, ri) \
++    dc->vmsd = &vmstate_stm32l4x5_gpio;
-+    (*(uint64_t *)((char *)(env) + (ri)->fieldoffset))
++    dc->realize = stm32l4x5_gpio_realize;
-+
++    rc->phases.hold = stm32l4x5_gpio_reset_hold;
-+#define REGINFO_SENTINEL { .type = ARM_CP_SENTINEL }
++}
 +
-+void define_arm_cp_regs_with_opaque(ARMCPU *cpu,
++static const TypeInfo stm32l4x5_gpio_types[] = {
-+                                    const ARMCPRegInfo *regs, void *opaque);
++    {
-+void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
++        .name = TYPE_STM32L4X5_GPIO,
-+                                       const ARMCPRegInfo *regs, void *opaque);
++        .parent = TYPE_SYS_BUS_DEVICE,
-+static inline void define_arm_cp_regs(ARMCPU *cpu, const ARMCPRegInfo *regs)
++        .instance_size = sizeof(Stm32l4x5GpioState),
-+{
++        .instance_init = stm32l4x5_gpio_init,
-+    define_arm_cp_regs_with_opaque(cpu, regs, 0);
++        .class_init = stm32l4x5_gpio_class_init,
-+}
++    },
-+static inline void define_one_arm_cp_reg(ARMCPU *cpu, const ARMCPRegInfo *regs)
++};
-+{
++
-+    define_one_arm_cp_reg_with_opaque(cpu, regs, 0);
++DEFINE_TYPES(stm32l4x5_gpio_types)
-+}
+diff --git a/hw/gpio/Kconfig b/hw/gpio/Kconfig
 +const ARMCPRegInfo *get_arm_cp_reginfo(GHashTable *cpregs, uint32_t encoded_cp);
 +
 +/*
 + * Definition of an ARM co-processor register as viewed from
 + * userspace. This is used for presenting sanitised versions of
 + * registers to userspace when emulating the Linux AArch64 CPU
 + * ID/feature ABI (advertised as HWCAP_CPUID).
 + */
 +typedef struct ARMCPRegUserSpaceInfo {
 +    /* Name of register */
 +    const char *name;
 +
 +    /* Is the name actually a glob pattern */
 +    bool is_glob;
 +
 +    /* Only some bits are exported to user space */
 +    uint64_t exported_bits;
 +
 +    /* Fixed bits are applied after the mask */
 +    uint64_t fixed_bits;
 +} ARMCPRegUserSpaceInfo;
 +
 +#define REGUSERINFO_SENTINEL { .name = NULL }
 +
 +void modify_arm_cp_regs(ARMCPRegInfo *regs, const ARMCPRegUserSpaceInfo *mods);
 +
 +/* CPWriteFn that can be used to implement writes-ignored behaviour */
 +void arm_cp_write_ignore(CPUARMState *env, const ARMCPRegInfo *ri,
 +                         uint64_t value);
 +/* CPReadFn that can be used for read-as-zero behaviour */
 +uint64_t arm_cp_read_zero(CPUARMState *env, const ARMCPRegInfo *ri);
 +
 +/*
 + * CPResetFn that does nothing, for use if no reset is required even
 + * if fieldoffset is non zero.
 + */
 +void arm_cp_reset_ignore(CPUARMState *env, const ARMCPRegInfo *opaque);
 +
 +/*
 + * Return true if this reginfo struct's field in the cpu state struct
 + * is 64 bits wide.
 + */
 +static inline bool cpreg_field_is_64bit(const ARMCPRegInfo *ri)
 +{
 +    return (ri->state == ARM_CP_STATE_AA64) || (ri->type & ARM_CP_64BIT);
 +}
 +
 +static inline bool cp_access_ok(int current_el,
 +                                const ARMCPRegInfo *ri, int isread)
 +{
 +    return (ri->access >> ((current_el * 2) + isread)) & 1;
 +}
 +
 +/* Raw read of a coprocessor register (as needed for migration, etc) */
 +uint64_t read_raw_cp_reg(CPUARMState *env, const ARMCPRegInfo *ri);
 +
 +#endif /* TARGET_ARM_CPREGS_H */
 diff --git a/target/arm/cpu.h b/target/arm/cpu.h
 index XXXXXXX..XXXXXXX 100644
---- a/target/arm/cpu.h
+--- a/hw/gpio/Kconfig
-+++ b/target/arm/cpu.h
++++ b/hw/gpio/Kconfig
-@@ -XXX,XX +XXX,XX @@ static inline uint64_t cpreg_to_kvm_id(uint32_t cpregid)
+@@ -XXX,XX +XXX,XX @@ config GPIO_PWR
-     return kvmid;
- }
+ config SIFIVE_GPIO
+     bool
--/* ARMCPRegInfo type field bits. If the SPECIAL bit is set this is a
++
-- * special-behaviour cp reg and bits [11..8] indicate what behaviour
++config STM32L4X5_GPIO
-- * it has. Otherwise it is a simple cp reg, where CONST indicates that
++    bool
-- * TCG can assume the value to be constant (ie load at translate time)
+diff --git a/hw/gpio/meson.build b/hw/gpio/meson.build
 - * and 64BIT indicates a 64 bit wide coprocessor register. SUPPRESS_TB_END
 - * indicates that the TB should not be ended after a write to this register
 - * (the default is that the TB ends after cp writes). OVERRIDE permits
 - * a register definition to override a previous definition for the
 - * same (cp, is64, crn, crm, opc1, opc2) tuple: either the new or the
 - * old must have the OVERRIDE bit set.
 - * ALIAS indicates that this register is an alias view of some underlying
 - * state which is also visible via another register, and that the other
 - * register is handling migration and reset; registers marked ALIAS will not be
 - * migrated but may have their state set by syncing of register state from KVM.
 - * NO_RAW indicates that this register has no underlying state and does not
 - * support raw access for state saving/loading; it will not be used for either
 - * migration or KVM state synchronization. (Typically this is for "registers"
 - * which are actually used as instructions for cache maintenance and so on.)
 - * IO indicates that this register does I/O and therefore its accesses
 - * need to be marked with gen_io_start() and also end the TB. In particular,
 - * registers which implement clocks or timers require this.
 - * RAISES_EXC is for when the read or write hook might raise an exception;
 - * the generated code will synchronize the CPU state before calling the hook
 - * so that it is safe for the hook to call raise_exception().
 - * NEWEL is for writes to registers that might change the exception
 - * level - typically on older ARM chips. For those cases we need to
 - * re-read the new el when recomputing the translation flags.
 - */
 -#define ARM_CP_SPECIAL           0x0001
 -#define ARM_CP_CONST             0x0002
 -#define ARM_CP_64BIT             0x0004
 -#define ARM_CP_SUPPRESS_TB_END   0x0008
 -#define ARM_CP_OVERRIDE          0x0010
 -#define ARM_CP_ALIAS             0x0020
 -#define ARM_CP_IO                0x0040
 -#define ARM_CP_NO_RAW            0x0080
 -#define ARM_CP_NOP               (ARM_CP_SPECIAL | 0x0100)
 -#define ARM_CP_WFI               (ARM_CP_SPECIAL | 0x0200)
 -#define ARM_CP_NZCV              (ARM_CP_SPECIAL | 0x0300)
 -#define ARM_CP_CURRENTEL         (ARM_CP_SPECIAL | 0x0400)
 -#define ARM_CP_DC_ZVA            (ARM_CP_SPECIAL | 0x0500)
 -#define ARM_CP_DC_GVA            (ARM_CP_SPECIAL | 0x0600)
 -#define ARM_CP_DC_GZVA           (ARM_CP_SPECIAL | 0x0700)
 -#define ARM_LAST_SPECIAL         ARM_CP_DC_GZVA
 -#define ARM_CP_FPU               0x1000
 -#define ARM_CP_SVE               0x2000
 -#define ARM_CP_NO_GDB            0x4000
 -#define ARM_CP_RAISES_EXC        0x8000
 -#define ARM_CP_NEWEL             0x10000
 -/* Used only as a terminator for ARMCPRegInfo lists */
 -#define ARM_CP_SENTINEL          0xfffff
 -/* Mask of only the flag bits in a type field */
 -#define ARM_CP_FLAG_MASK         0x1f0ff
 -
 -/* Valid values for ARMCPRegInfo state field, indicating which of
 - * the AArch32 and AArch64 execution states this register is visible in.
 - * If the reginfo doesn't explicitly specify then it is AArch32 only.
 - * If the reginfo is declared to be visible in both states then a second
 - * reginfo is synthesised for the AArch32 view of the AArch64 register,
 - * such that the AArch32 view is the lower 32 bits of the AArch64 one.
 - * Note that we rely on the values of these enums as we iterate through
 - * the various states in some places.
 - */
 -enum {
 -    ARM_CP_STATE_AA32 = 0,
 -    ARM_CP_STATE_AA64 = 1,
 -    ARM_CP_STATE_BOTH = 2,
 -};
 -
 -/* ARM CP register secure state flags.  These flags identify security state
 - * attributes for a given CP register entry.
 - * The existence of both or neither secure and non-secure flags indicates that
 - * the register has both a secure and non-secure hash entry.  A single one of
 - * these flags causes the register to only be hashed for the specified
 - * security state.
 - * Although definitions may have any combination of the S/NS bits, each
 - * registered entry will only have one to identify whether the entry is secure
 - * or non-secure.
 - */
 -enum {
 -    ARM_CP_SECSTATE_S =   (1 << 0), /* bit[0]: Secure state register */
 -    ARM_CP_SECSTATE_NS =  (1 << 1), /* bit[1]: Non-secure state register */
 -};
 -
 -/* Return true if cptype is a valid type field. This is used to try to
 - * catch errors where the sentinel has been accidentally left off the end
 - * of a list of registers.
 - */
 -static inline bool cptype_valid(int cptype)
 -{
 -    return ((cptype & ~ARM_CP_FLAG_MASK) == 0)
 -        || ((cptype & ARM_CP_SPECIAL) &&
 -            ((cptype & ~ARM_CP_FLAG_MASK) <= ARM_LAST_SPECIAL));
 -}
 -
 -/* Access rights:
 - * We define bits for Read and Write access for what rev C of the v7-AR ARM ARM
 - * defines as PL0 (user), PL1 (fiq/irq/svc/abt/und/sys, ie privileged), and
 - * PL2 (hyp). The other level which has Read and Write bits is Secure PL1
 - * (ie any of the privileged modes in Secure state, or Monitor mode).
 - * If a register is accessible in one privilege level it's always accessible
 - * in higher privilege levels too. Since "Secure PL1" also follows this rule
 - * (ie anything visible in PL2 is visible in S-PL1, some things are only
 - * visible in S-PL1) but "Secure PL1" is a bit of a mouthful, we bend the
 - * terminology a little and call this PL3.
 - * In AArch64 things are somewhat simpler as the PLx bits line up exactly
 - * with the ELx exception levels.
 - *
 - * If access permissions for a register are more complex than can be
 - * described with these bits, then use a laxer set of restrictions, and
 - * do the more restrictive/complex check inside a helper function.
 - */
 -#define PL3_R 0x80
 -#define PL3_W 0x40
 -#define PL2_R (0x20 | PL3_R)
 -#define PL2_W (0x10 | PL3_W)
 -#define PL1_R (0x08 | PL2_R)
 -#define PL1_W (0x04 | PL2_W)
 -#define PL0_R (0x02 | PL1_R)
 -#define PL0_W (0x01 | PL1_W)
 -
 -/*
 - * For user-mode some registers are accessible to EL0 via a kernel
 - * trap-and-emulate ABI. In this case we define the read permissions
 - * as actually being PL0_R. However some bits of any given register
 - * may still be masked.
 - */
 -#ifdef CONFIG_USER_ONLY
 -#define PL0U_R PL0_R
 -#else
 -#define PL0U_R PL1_R
 -#endif
 -
 -#define PL3_RW (PL3_R | PL3_W)
 -#define PL2_RW (PL2_R | PL2_W)
 -#define PL1_RW (PL1_R | PL1_W)
 -#define PL0_RW (PL0_R | PL0_W)
 -
  /* Return the highest implemented Exception Level */
  static inline int arm_highest_el(CPUARMState *env)
  {
@@ -XXX,XX +XXX,XX @@ static inline int arm_current_el(CPUARMState *env)
      }
  }
 -typedef struct ARMCPRegInfo ARMCPRegInfo;
 -
 -typedef enum CPAccessResult {
 -    /* Access is permitted */
 -    CP_ACCESS_OK = 0,
 -    /* Access fails due to a configurable trap or enable which would
 -     * result in a categorized exception syndrome giving information about
 -     * the failing instruction (ie syndrome category 0x3, 0x4, 0x5, 0x6,
 -     * 0xc or 0x18). The exception is taken to the usual target EL (EL1 or
 -     * PL1 if in EL0, otherwise to the current EL).
 -     */
 -    CP_ACCESS_TRAP = 1,
 -    /* Access fails and results in an exception syndrome 0x0 ("uncategorized").
 -     * Note that this is not a catch-all case -- the set of cases which may
 -     * result in this failure is specifically defined by the architecture.
 -     */
 -    CP_ACCESS_TRAP_UNCATEGORIZED = 2,
 -    /* As CP_ACCESS_TRAP, but for traps directly to EL2 or EL3 */
 -    CP_ACCESS_TRAP_EL2 = 3,
 -    CP_ACCESS_TRAP_EL3 = 4,
 -    /* As CP_ACCESS_UNCATEGORIZED, but for traps directly to EL2 or EL3 */
 -    CP_ACCESS_TRAP_UNCATEGORIZED_EL2 = 5,
 -    CP_ACCESS_TRAP_UNCATEGORIZED_EL3 = 6,
 -} CPAccessResult;
 -
 -/* Access functions for coprocessor registers. These cannot fail and
 - * may not raise exceptions.
 - */
 -typedef uint64_t CPReadFn(CPUARMState *env, const ARMCPRegInfo *opaque);
 -typedef void CPWriteFn(CPUARMState *env, const ARMCPRegInfo *opaque,
 -                       uint64_t value);
 -/* Access permission check functions for coprocessor registers. */
 -typedef CPAccessResult CPAccessFn(CPUARMState *env,
 -                                  const ARMCPRegInfo *opaque,
 -                                  bool isread);
 -/* Hook function for register reset */
 -typedef void CPResetFn(CPUARMState *env, const ARMCPRegInfo *opaque);
 -
 -#define CP_ANY 0xff
 -
 -/* Definition of an ARM coprocessor register */
 -struct ARMCPRegInfo {
 -    /* Name of register (useful mainly for debugging, need not be unique) */
 -    const char *name;
 -    /* Location of register: coprocessor number and (crn,crm,opc1,opc2)
 -     * tuple. Any of crm, opc1 and opc2 may be CP_ANY to indicate a
 -     * 'wildcard' field -- any value of that field in the MRC/MCR insn
 -     * will be decoded to this register. The register read and write
 -     * callbacks will be passed an ARMCPRegInfo with the crn/crm/opc1/opc2
 -     * used by the program, so it is possible to register a wildcard and
 -     * then behave differently on read/write if necessary.
 -     * For 64 bit registers, only crm and opc1 are relevant; crn and opc2
 -     * must both be zero.
 -     * For AArch64-visible registers, opc0 is also used.
 -     * Since there are no "coprocessors" in AArch64, cp is purely used as a
 -     * way to distinguish (for KVM's benefit) guest-visible system registers
 -     * from demuxed ones provided to preserve the "no side effects on
 -     * KVM register read/write from QEMU" semantics. cp==0x13 is guest
 -     * visible (to match KVM's encoding); cp==0 will be converted to
 -     * cp==0x13 when the ARMCPRegInfo is registered, for convenience.
 -     */
 -    uint8_t cp;
 -    uint8_t crn;
 -    uint8_t crm;
 -    uint8_t opc0;
 -    uint8_t opc1;
 -    uint8_t opc2;
 -    /* Execution state in which this register is visible: ARM_CP_STATE_* */
 -    int state;
 -    /* Register type: ARM_CP_* bits/values */
 -    int type;
 -    /* Access rights: PL*_[RW] */
 -    int access;
 -    /* Security state: ARM_CP_SECSTATE_* bits/values */
 -    int secure;
 -    /* The opaque pointer passed to define_arm_cp_regs_with_opaque() when
 -     * this register was defined: can be used to hand data through to the
 -     * register read/write functions, since they are passed the ARMCPRegInfo*.
 -     */
 -    void *opaque;
 -    /* Value of this register, if it is ARM_CP_CONST. Otherwise, if
 -     * fieldoffset is non-zero, the reset value of the register.
 -     */
 -    uint64_t resetvalue;
 -    /* Offset of the field in CPUARMState for this register.
 -     *
 -     * This is not needed if either:
 -     *  1. type is ARM_CP_CONST or one of the ARM_CP_SPECIALs
 -     *  2. both readfn and writefn are specified
 -     */
 -    ptrdiff_t fieldoffset; /* offsetof(CPUARMState, field) */
 -
 -    /* Offsets of the secure and non-secure fields in CPUARMState for the
 -     * register if it is banked.  These fields are only used during the static
 -     * registration of a register.  During hashing the bank associated
 -     * with a given security state is copied to fieldoffset which is used from
 -     * there on out.
 -     *
 -     * It is expected that register definitions use either fieldoffset or
 -     * bank_fieldoffsets in the definition but not both.  It is also expected
 -     * that both bank offsets are set when defining a banked register.  This
 -     * use indicates that a register is banked.
 -     */
 -    ptrdiff_t bank_fieldoffsets[2];
 -
 -    /* Function for making any access checks for this register in addition to
 -     * those specified by the 'access' permissions bits. If NULL, no extra
 -     * checks required. The access check is performed at runtime, not at
 -     * translate time.
 -     */
 -    CPAccessFn *accessfn;
 -    /* Function for handling reads of this register. If NULL, then reads
 -     * will be done by loading from the offset into CPUARMState specified
 -     * by fieldoffset.
 -     */
 -    CPReadFn *readfn;
 -    /* Function for handling writes of this register. If NULL, then writes
 -     * will be done by writing to the offset into CPUARMState specified
 -     * by fieldoffset.
 -     */
 -    CPWriteFn *writefn;
 -    /* Function for doing a "raw" read; used when we need to copy
 -     * coprocessor state to the kernel for KVM or out for
 -     * migration. This only needs to be provided if there is also a
 -     * readfn and it has side effects (for instance clear-on-read bits).
 -     */
 -    CPReadFn *raw_readfn;
 -    /* Function for doing a "raw" write; used when we need to copy KVM
 -     * kernel coprocessor state into userspace, or for inbound
 -     * migration. This only needs to be provided if there is also a
 -     * writefn and it masks out "unwritable" bits or has write-one-to-clear
 -     * or similar behaviour.
 -     */
 -    CPWriteFn *raw_writefn;
 -    /* Function for resetting the register. If NULL, then reset will be done
 -     * by writing resetvalue to the field specified in fieldoffset. If
 -     * fieldoffset is 0 then no reset will be done.
 -     */
 -    CPResetFn *resetfn;
 -
 -    /*
 -     * "Original" writefn and readfn.
 -     * For ARMv8.1-VHE register aliases, we overwrite the read/write
 -     * accessor functions of various EL1/EL0 to perform the runtime
 -     * check for which sysreg should actually be modified, and then
 -     * forwards the operation.  Before overwriting the accessors,
 -     * the original function is copied here, so that accesses that
 -     * really do go to the EL1/EL0 version proceed normally.
 -     * (The corresponding EL2 register is linked via opaque.)
 -     */
 -    CPReadFn *orig_readfn;
 -    CPWriteFn *orig_writefn;
 -};
 -
 -/* Macros which are lvalues for the field in CPUARMState for the
 - * ARMCPRegInfo *ri.
 - */
 -#define CPREG_FIELD32(env, ri) \
 -    (*(uint32_t *)((char *)(env) + (ri)->fieldoffset))
 -#define CPREG_FIELD64(env, ri) \
 -    (*(uint64_t *)((char *)(env) + (ri)->fieldoffset))
 -
 -#define REGINFO_SENTINEL { .type = ARM_CP_SENTINEL }
 -
 -void define_arm_cp_regs_with_opaque(ARMCPU *cpu,
 -                                    const ARMCPRegInfo *regs, void *opaque);
 -void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
 -                                       const ARMCPRegInfo *regs, void *opaque);
 -static inline void define_arm_cp_regs(ARMCPU *cpu, const ARMCPRegInfo *regs)
 -{
 -    define_arm_cp_regs_with_opaque(cpu, regs, 0);
 -}
 -static inline void define_one_arm_cp_reg(ARMCPU *cpu, const ARMCPRegInfo *regs)
 -{
 -    define_one_arm_cp_reg_with_opaque(cpu, regs, 0);
 -}
 -const ARMCPRegInfo *get_arm_cp_reginfo(GHashTable *cpregs, uint32_t encoded_cp);
 -
 -/*
 - * Definition of an ARM co-processor register as viewed from
 - * userspace. This is used for presenting sanitised versions of
 - * registers to userspace when emulating the Linux AArch64 CPU
 - * ID/feature ABI (advertised as HWCAP_CPUID).
 - */
 -typedef struct ARMCPRegUserSpaceInfo {
 -    /* Name of register */
 -    const char *name;
 -
 -    /* Is the name actually a glob pattern */
 -    bool is_glob;
 -
 -    /* Only some bits are exported to user space */
 -    uint64_t exported_bits;
 -
 -    /* Fixed bits are applied after the mask */
 -    uint64_t fixed_bits;
 -} ARMCPRegUserSpaceInfo;
 -
 -#define REGUSERINFO_SENTINEL { .name = NULL }
 -
 -void modify_arm_cp_regs(ARMCPRegInfo *regs, const ARMCPRegUserSpaceInfo *mods);
 -
 -/* CPWriteFn that can be used to implement writes-ignored behaviour */
 -void arm_cp_write_ignore(CPUARMState *env, const ARMCPRegInfo *ri,
 -                         uint64_t value);
 -/* CPReadFn that can be used for read-as-zero behaviour */
 -uint64_t arm_cp_read_zero(CPUARMState *env, const ARMCPRegInfo *ri);
 -
 -/* CPResetFn that does nothing, for use if no reset is required even
 - * if fieldoffset is non zero.
 - */
 -void arm_cp_reset_ignore(CPUARMState *env, const ARMCPRegInfo *opaque);
 -
 -/* Return true if this reginfo struct's field in the cpu state struct
 - * is 64 bits wide.
 - */
 -static inline bool cpreg_field_is_64bit(const ARMCPRegInfo *ri)
 -{
 -    return (ri->state == ARM_CP_STATE_AA64) || (ri->type & ARM_CP_64BIT);
 -}
 -
 -static inline bool cp_access_ok(int current_el,
 -                                const ARMCPRegInfo *ri, int isread)
 -{
 -    return (ri->access >> ((current_el * 2) + isread)) & 1;
 -}
 -
 -/* Raw read of a coprocessor register (as needed for migration, etc) */
 -uint64_t read_raw_cp_reg(CPUARMState *env, const ARMCPRegInfo *ri);
 -
  /**
   * write_list_to_cpustate
   * @cpu: ARMCPU
 diff --git a/hw/arm/pxa2xx.c b/hw/arm/pxa2xx.c
 index XXXXXXX..XXXXXXX 100644
---- a/hw/arm/pxa2xx.c
+--- a/hw/gpio/meson.build
-+++ b/hw/arm/pxa2xx.c
++++ b/hw/gpio/meson.build
-@@ -XXX,XX +XXX,XX @@
+@@ -XXX,XX +XXX,XX @@ system_ss.add(when: 'CONFIG_RASPI', if_true: files(
- #include "qemu/cutils.h"
+     'bcm2835_gpio.c',
- #include "qemu/log.h"
+     'bcm2838_gpio.c'
- #include "qom/object.h"
+ ))
-+#include "target/arm/cpregs.h"
++system_ss.add(when: 'CONFIG_STM32L4X5_SOC', if_true: files('stm32l4x5_gpio.c'))
+ system_ss.add(when: 'CONFIG_ASPEED_SOC', if_true: files('aspeed_gpio.c'))
- static struct {
+ system_ss.add(when: 'CONFIG_SIFIVE_GPIO', if_true: files('sifive_gpio.c'))
-     hwaddr io_base;
+diff --git a/hw/gpio/trace-events b/hw/gpio/trace-events
 diff --git a/hw/arm/pxa2xx_pic.c b/hw/arm/pxa2xx_pic.c
 index XXXXXXX..XXXXXXX 100644
---- a/hw/arm/pxa2xx_pic.c
+--- a/hw/gpio/trace-events
-+++ b/hw/arm/pxa2xx_pic.c
++++ b/hw/gpio/trace-events
-@@ -XXX,XX +XXX,XX @@
+@@ -XXX,XX +XXX,XX @@ sifive_gpio_update_output_irq(int64_t line, int64_t value) "line %" PRIi64 " val
- #include "hw/sysbus.h"
+ # aspeed_gpio.c
- #include "migration/vmstate.h"
+ aspeed_gpio_read(uint64_t offset, uint64_t value) "offset: 0x%" PRIx64 " value 0x%" PRIx64
- #include "qom/object.h"
+ aspeed_gpio_write(uint64_t offset, uint64_t value) "offset: 0x%" PRIx64 " value 0x%" PRIx64
-+#include "target/arm/cpregs.h"
++
++# stm32l4x5_gpio.c
- #define ICIP    0x00    /* Interrupt Controller IRQ Pending register */
++stm32l4x5_gpio_read(char *gpio, uint64_t addr) "GPIO%s addr: 0x%" PRIx64 " "
- #define ICMR    0x04    /* Interrupt Controller Mask register */
++stm32l4x5_gpio_write(char *gpio, uint64_t addr, uint64_t data) "GPIO%s addr: 0x%" PRIx64 " val: 0x%" PRIx64 ""
-diff --git a/hw/intc/arm_gicv3_cpuif.c b/hw/intc/arm_gicv3_cpuif.c
++stm32l4x5_gpio_update_idr(char *gpio, uint32_t old_idr, uint32_t new_idr) "GPIO%s from: 0x%x to: 0x%x"
-index XXXXXXX..XXXXXXX 100644
++stm32l4x5_gpio_pins(char *gpio, uint16_t disconnected, uint16_t high) "GPIO%s disconnected pins: 0x%x levels: 0x%x"
 --- a/hw/intc/arm_gicv3_cpuif.c
 +++ b/hw/intc/arm_gicv3_cpuif.c
@@ -XXX,XX +XXX,XX @@
  #include "gicv3_internal.h"
  #include "hw/irq.h"
  #include "cpu.h"
 +#include "target/arm/cpregs.h"
  /*
   * Special case return value from hppvi_index(); must be larger than
 diff --git a/hw/intc/arm_gicv3_kvm.c b/hw/intc/arm_gicv3_kvm.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/intc/arm_gicv3_kvm.c
 +++ b/hw/intc/arm_gicv3_kvm.c
@@ -XXX,XX +XXX,XX @@
  #include "vgic_common.h"
  #include "migration/blocker.h"
  #include "qom/object.h"
 +#include "target/arm/cpregs.h"
 +
  #ifdef DEBUG_GICV3_KVM
  #define DPRINTF(fmt, ...) \
 diff --git a/target/arm/cpu.c b/target/arm/cpu.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/cpu.c
 +++ b/target/arm/cpu.c
@@ -XXX,XX +XXX,XX @@
  #include "kvm_arm.h"
  #include "disas/capstone.h"
  #include "fpu/softfloat.h"
 +#include "cpregs.h"
  static void arm_cpu_set_pc(CPUState *cs, vaddr value)
  {
 diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/cpu64.c
 +++ b/target/arm/cpu64.c
@@ -XXX,XX +XXX,XX @@
  #include "hvf_arm.h"
  #include "qapi/visitor.h"
  #include "hw/qdev-properties.h"
 +#include "cpregs.h"
  #ifndef CONFIG_USER_ONLY
 diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/cpu_tcg.c
 +++ b/target/arm/cpu_tcg.c
@@ -XXX,XX +XXX,XX @@
  #if !defined(CONFIG_USER_ONLY)
  #include "hw/boards.h"
  #endif
 +#include "cpregs.h"
  /* CPU models. These are not needed for the AArch64 linux-user build. */
  #if !defined(CONFIG_USER_ONLY) || !defined(TARGET_AARCH64)
 diff --git a/target/arm/gdbstub.c b/target/arm/gdbstub.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/gdbstub.c
 +++ b/target/arm/gdbstub.c
@@ -XXX,XX +XXX,XX @@
   */
  #include "qemu/osdep.h"
  #include "cpu.h"
 -#include "internals.h"
  #include "exec/gdbstub.h"
 +#include "internals.h"
 +#include "cpregs.h"
  typedef struct RegisterSysregXmlParam {
      CPUState *cs;
 diff --git a/target/arm/helper.c b/target/arm/helper.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/helper.c
 +++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@
  #include "exec/cpu_ldst.h"
  #include "semihosting/common-semi.h"
  #endif
 +#include "cpregs.h"
  #define ARM_CPU_FREQ 1000000000 /* FIXME: 1 GHz, should be configurable */
  #define PMCR_NUM_COUNTERS 4 /* QEMU IMPDEF choice */
 diff --git a/target/arm/op_helper.c b/target/arm/op_helper.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/op_helper.c
 +++ b/target/arm/op_helper.c
@@ -XXX,XX +XXX,XX @@
  #include "internals.h"
  #include "exec/exec-all.h"
  #include "exec/cpu_ldst.h"
 +#include "cpregs.h"
  #define SIGNBIT (uint32_t)0x80000000
  #define SIGNBIT64 ((uint64_t)1 << 63)
 diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/translate-a64.c
 +++ b/target/arm/translate-a64.c
@@ -XXX,XX +XXX,XX @@
  #include "translate.h"
  #include "internals.h"
  #include "qemu/host-utils.h"
 -
  #include "semihosting/semihost.h"
  #include "exec/gen-icount.h"
 -
  #include "exec/helper-proto.h"
  #include "exec/helper-gen.h"
  #include "exec/log.h"
 -
 +#include "cpregs.h"
  #include "translate-a64.h"
  #include "qemu/atomic128.h"
 diff --git a/target/arm/translate.c b/target/arm/translate.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/translate.c
 +++ b/target/arm/translate.c
@@ -XXX,XX +XXX,XX @@
  #include "qemu/bitops.h"
  #include "arm_ldst.h"
  #include "semihosting/semihost.h"
 -
  #include "exec/helper-proto.h"
  #include "exec/helper-gen.h"
 -
  #include "exec/log.h"
 +#include "cpregs.h"
  #define ENABLE_ARCH_4T    arm_dc_feature(s, ARM_FEATURE_V4T)
 --
-.25.1
+.34.1

-[PULL 06/23] target/arm: Reorg ARMCPRegInfo type field bits
+[PULL 10/14] hw/arm: Connect STM32L4x5 GPIO to STM32L4x5 SoC
-From: Richard Henderson <richard.henderson@linaro.org>
+From: Inès Varhol <ines.varhol@telecom-paris.fr>
-Instead of defining ARM_CP_FLAG_MASK to remove flags,
+Signed-off-by: Arnaud Minier <arnaud.minier@telecom-paris.fr>
-define ARM_CP_SPECIAL_MASK to isolate special cases.
+Signed-off-by: Inès Varhol <ines.varhol@telecom-paris.fr>
-Sort the specials to the low bits. Use an enum.
+Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
+Acked-by: Alistair Francis <alistair.francis@wdc.com>
-Split the large comment block so as to document each
+Message-id: 20240305210444.310665-3-ines.varhol@telecom-paris.fr
 value separately.
 Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
 Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
 Message-id: 20220501055028.646596-6-richard.henderson@linaro.org
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
 ---
- target/arm/cpregs.h        | 130 +++++++++++++++++++++++--------------
+ include/hw/arm/stm32l4x5_soc.h     |  2 +
- target/arm/cpu.c           |   4 +-
+ include/hw/gpio/stm32l4x5_gpio.h   |  1 +
- target/arm/helper.c        |   4 +-
+ include/hw/misc/stm32l4x5_syscfg.h |  3 +-
- target/arm/translate-a64.c |   6 +-
+ hw/arm/stm32l4x5_soc.c             | 71 +++++++++++++++++++++++-------
- target/arm/translate.c     |   6 +-
+ hw/misc/stm32l4x5_syscfg.c         |  1 +
-files changed, 92 insertions(+), 58 deletions(-)
+ hw/arm/Kconfig                     |  3 +-
+files changed, 63 insertions(+), 18 deletions(-)
-diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
-index XXXXXXX..XXXXXXX 100644
+diff --git a/include/hw/arm/stm32l4x5_soc.h b/include/hw/arm/stm32l4x5_soc.h
---- a/target/arm/cpregs.h
+index XXXXXXX..XXXXXXX 100644
-+++ b/target/arm/cpregs.h
+--- a/include/hw/arm/stm32l4x5_soc.h
-@@ -XXX,XX +XXX,XX @@
++++ b/include/hw/arm/stm32l4x5_soc.h
- #define TARGET_ARM_CPREGS_H
+@@ -XXX,XX +XXX,XX @@
+ #include "hw/misc/stm32l4x5_syscfg.h"
- /*
+ #include "hw/misc/stm32l4x5_exti.h"
-- * ARMCPRegInfo type field bits. If the SPECIAL bit is set this is a
+ #include "hw/misc/stm32l4x5_rcc.h"
-- * special-behaviour cp reg and bits [11..8] indicate what behaviour
++#include "hw/gpio/stm32l4x5_gpio.h"
-- * it has. Otherwise it is a simple cp reg, where CONST indicates that
+ #include "qom/object.h"
-- * TCG can assume the value to be constant (ie load at translate time)
-- * and 64BIT indicates a 64 bit wide coprocessor register. SUPPRESS_TB_END
+ #define TYPE_STM32L4X5_SOC "stm32l4x5-soc"
-- * indicates that the TB should not be ended after a write to this register
+@@ -XXX,XX +XXX,XX @@ struct Stm32l4x5SocState {
-- * (the default is that the TB ends after cp writes). OVERRIDE permits
+     OrIRQState exti_or_gates[NUM_EXTI_OR_GATES];
-- * a register definition to override a previous definition for the
+     Stm32l4x5SyscfgState syscfg;
-- * same (cp, is64, crn, crm, opc1, opc2) tuple: either the new or the
+     Stm32l4x5RccState rcc;
-- * old must have the OVERRIDE bit set.
++    Stm32l4x5GpioState gpio[NUM_GPIOS];
-- * ALIAS indicates that this register is an alias view of some underlying
-- * state which is also visible via another register, and that the other
+     MemoryRegion sram1;
-- * register is handling migration and reset; registers marked ALIAS will not be
+     MemoryRegion sram2;
-- * migrated but may have their state set by syncing of register state from KVM.
+diff --git a/include/hw/gpio/stm32l4x5_gpio.h b/include/hw/gpio/stm32l4x5_gpio.h
-- * NO_RAW indicates that this register has no underlying state and does not
+index XXXXXXX..XXXXXXX 100644
-- * support raw access for state saving/loading; it will not be used for either
+--- a/include/hw/gpio/stm32l4x5_gpio.h
-- * migration or KVM state synchronization. (Typically this is for "registers"
++++ b/include/hw/gpio/stm32l4x5_gpio.h
-- * which are actually used as instructions for cache maintenance and so on.)
+@@ -XXX,XX +XXX,XX @@
-- * IO indicates that this register does I/O and therefore its accesses
+ #define TYPE_STM32L4X5_GPIO "stm32l4x5-gpio"
-- * need to be marked with gen_io_start() and also end the TB. In particular,
+ OBJECT_DECLARE_SIMPLE_TYPE(Stm32l4x5GpioState, STM32L4X5_GPIO)
-- * registers which implement clocks or timers require this.
-- * RAISES_EXC is for when the read or write hook might raise an exception;
++#define NUM_GPIOS 8
-- * the generated code will synchronize the CPU state before calling the hook
+ #define GPIO_NUM_PINS 16
-- * so that it is safe for the hook to call raise_exception().
-- * NEWEL is for writes to registers that might change the exception
+ struct Stm32l4x5GpioState {
-- * level - typically on older ARM chips. For those cases we need to
+diff --git a/include/hw/misc/stm32l4x5_syscfg.h b/include/hw/misc/stm32l4x5_syscfg.h
-- * re-read the new el when recomputing the translation flags.
+index XXXXXXX..XXXXXXX 100644
-+ * ARMCPRegInfo type field bits:
+--- a/include/hw/misc/stm32l4x5_syscfg.h
-  */
++++ b/include/hw/misc/stm32l4x5_syscfg.h
--#define ARM_CP_SPECIAL           0x0001
+@@ -XXX,XX +XXX,XX @@
--#define ARM_CP_CONST             0x0002
--#define ARM_CP_64BIT             0x0004
+ #include "hw/sysbus.h"
--#define ARM_CP_SUPPRESS_TB_END   0x0008
+ #include "qom/object.h"
--#define ARM_CP_OVERRIDE          0x0010
++#include "hw/gpio/stm32l4x5_gpio.h"
--#define ARM_CP_ALIAS             0x0020
--#define ARM_CP_IO                0x0040
+ #define TYPE_STM32L4X5_SYSCFG "stm32l4x5-syscfg"
--#define ARM_CP_NO_RAW            0x0080
+ OBJECT_DECLARE_SIMPLE_TYPE(Stm32l4x5SyscfgState, STM32L4X5_SYSCFG)
--#define ARM_CP_NOP               (ARM_CP_SPECIAL | 0x0100)
--#define ARM_CP_WFI               (ARM_CP_SPECIAL | 0x0200)
+-#define NUM_GPIOS 8
--#define ARM_CP_NZCV              (ARM_CP_SPECIAL | 0x0300)
+-#define GPIO_NUM_PINS 16
--#define ARM_CP_CURRENTEL         (ARM_CP_SPECIAL | 0x0400)
+ #define SYSCFG_NUM_EXTICR 4
--#define ARM_CP_DC_ZVA            (ARM_CP_SPECIAL | 0x0500)
--#define ARM_CP_DC_GVA            (ARM_CP_SPECIAL | 0x0600)
+ struct Stm32l4x5SyscfgState {
--#define ARM_CP_DC_GZVA           (ARM_CP_SPECIAL | 0x0700)
+diff --git a/hw/arm/stm32l4x5_soc.c b/hw/arm/stm32l4x5_soc.c
--#define ARM_LAST_SPECIAL         ARM_CP_DC_GZVA
+index XXXXXXX..XXXXXXX 100644
--#define ARM_CP_FPU               0x1000
+--- a/hw/arm/stm32l4x5_soc.c
--#define ARM_CP_SVE               0x2000
++++ b/hw/arm/stm32l4x5_soc.c
--#define ARM_CP_NO_GDB            0x4000
+@@ -XXX,XX +XXX,XX @@
--#define ARM_CP_RAISES_EXC        0x8000
+ #include "sysemu/sysemu.h"
--#define ARM_CP_NEWEL             0x10000
+ #include "hw/or-irq.h"
--/* Mask of only the flag bits in a type field */
+ #include "hw/arm/stm32l4x5_soc.h"
--#define ARM_CP_FLAG_MASK         0x1f0ff
++#include "hw/gpio/stm32l4x5_gpio.h"
-+enum {
+ #include "hw/qdev-clock.h"
-+    /*
+ #include "hw/misc/unimp.h"
-+     * Register must be handled specially during translation.
-+     * The method is one of the values below:
+@@ -XXX,XX +XXX,XX @@ static const int exti_or_gate1_lines_in[EXTI_OR_GATE1_NUM_LINES_IN] = {
-+     */
+, 35, 36, 37, 38,
-+    ARM_CP_SPECIAL_MASK          = 0x000f,
+ };
-+    /* Special: no change to PE state: writes ignored, reads ignored. */
-+    ARM_CP_NOP                   = 0x0001,
++static const struct {
-+    /* Special: sysreg is WFI, for v5 and v6. */
++    uint32_t addr;
-+    ARM_CP_WFI                   = 0x0002,
++    uint32_t moder_reset;
-+    /* Special: sysreg is NZCV. */
++    uint32_t ospeedr_reset;
-+    ARM_CP_NZCV                  = 0x0003,
++    uint32_t pupdr_reset;
-+    /* Special: sysreg is CURRENTEL. */
++} stm32l4x5_gpio_cfg[NUM_GPIOS] = {
-+    ARM_CP_CURRENTEL             = 0x0004,
++    { 0x48000000, 0xABFFFFFF, 0x0C000000, 0x64000000 },
-+    /* Special: sysreg is DC ZVA or similar. */
++    { 0x48000400, 0xFFFFFEBF, 0x00000000, 0x00000100 },
-+    ARM_CP_DC_ZVA                = 0x0005,
++    { 0x48000800, 0xFFFFFFFF, 0x00000000, 0x00000000 },
-+    ARM_CP_DC_GVA                = 0x0006,
++    { 0x48000C00, 0xFFFFFFFF, 0x00000000, 0x00000000 },
-+    ARM_CP_DC_GZVA               = 0x0007,
++    { 0x48001000, 0xFFFFFFFF, 0x00000000, 0x00000000 },
-+
++    { 0x48001400, 0xFFFFFFFF, 0x00000000, 0x00000000 },
-+    /* Flag: reads produce resetvalue; writes ignored. */
++    { 0x48001800, 0xFFFFFFFF, 0x00000000, 0x00000000 },
-+    ARM_CP_CONST                 = 1 << 4,
++    { 0x48001C00, 0x0000000F, 0x00000000, 0x00000000 },
 +    /* Flag: For ARM_CP_STATE_AA32, sysreg is 64-bit. */
 +    ARM_CP_64BIT                 = 1 << 5,
 +    /*
 +     * Flag: TB should not be ended after a write to this register
 +     * (the default is that the TB ends after cp writes).
 +     */
 +    ARM_CP_SUPPRESS_TB_END       = 1 << 6,
 +    /*
 +     * Flag: Permit a register definition to override a previous definition
 +     * for the same (cp, is64, crn, crm, opc1, opc2) tuple: either the new
 +     * or the old must have the ARM_CP_OVERRIDE bit set.
 +     */
 +    ARM_CP_OVERRIDE              = 1 << 7,
 +    /*
 +     * Flag: Register is an alias view of some underlying state which is also
 +     * visible via another register, and that the other register is handling
 +     * migration and reset; registers marked ARM_CP_ALIAS will not be migrated
 +     * but may have their state set by syncing of register state from KVM.
 +     */
 +    ARM_CP_ALIAS                 = 1 << 8,
 +    /*
 +     * Flag: Register does I/O and therefore its accesses need to be marked
 +     * with gen_io_start() and also end the TB. In particular, registers which
 +     * implement clocks or timers require this.
 +     */
 +    ARM_CP_IO                    = 1 << 9,
 +    /*
 +     * Flag: Register has no underlying state and does not support raw access
 +     * for state saving/loading; it will not be used for either migration or
 +     * KVM state synchronization. Typically this is for "registers" which are
 +     * actually used as instructions for cache maintenance and so on.
 +     */
 +    ARM_CP_NO_RAW                = 1 << 10,
 +    /*
 +     * Flag: The read or write hook might raise an exception; the generated
 +     * code will synchronize the CPU state before calling the hook so that it
 +     * is safe for the hook to call raise_exception().
 +     */
 +    ARM_CP_RAISES_EXC            = 1 << 11,
 +    /*
 +     * Flag: Writes to the sysreg might change the exception level - typically
 +     * on older ARM chips. For those cases we need to re-read the new el when
 +     * recomputing the translation flags.
 +     */
 +    ARM_CP_NEWEL                 = 1 << 12,
 +    /*
 +     * Flag: Access check for this sysreg is identical to accessing FPU state
 +     * from an instruction: use translation fp_access_check().
 +     */
 +    ARM_CP_FPU                   = 1 << 13,
 +    /*
 +     * Flag: Access check for this sysreg is identical to accessing SVE state
 +     * from an instruction: use translation sve_access_check().
 +     */
 +    ARM_CP_SVE                   = 1 << 14,
 +    /* Flag: Do not expose in gdb sysreg xml. */
 +    ARM_CP_NO_GDB                = 1 << 15,
 +};
++
- /*
+ static void stm32l4x5_soc_initfn(Object *obj)
-  * Valid values for ARMCPRegInfo state field, indicating which of
+ {
-diff --git a/target/arm/cpu.c b/target/arm/cpu.c
+     Stm32l4x5SocState *s = STM32L4X5_SOC(obj);
-index XXXXXXX..XXXXXXX 100644
+@@ -XXX,XX +XXX,XX @@ static void stm32l4x5_soc_initfn(Object *obj)
---- a/target/arm/cpu.c
+     }
-+++ b/target/arm/cpu.c
+     object_initialize_child(obj, "syscfg", &s->syscfg, TYPE_STM32L4X5_SYSCFG);
-@@ -XXX,XX +XXX,XX @@ static void cp_reg_reset(gpointer key, gpointer value, gpointer opaque)
+     object_initialize_child(obj, "rcc", &s->rcc, TYPE_STM32L4X5_RCC);
-     ARMCPRegInfo *ri = value;
++
-     ARMCPU *cpu = opaque;
++    for (unsigned i = 0; i < NUM_GPIOS; i++) {
++        g_autofree char *name = g_strdup_printf("gpio%c", 'a' + i);
--    if (ri->type & (ARM_CP_SPECIAL | ARM_CP_ALIAS)) {
++        object_initialize_child(obj, name, &s->gpio[i], TYPE_STM32L4X5_GPIO);
-+    if (ri->type & (ARM_CP_SPECIAL_MASK | ARM_CP_ALIAS)) {
++    }
  }
  static void stm32l4x5_soc_realize(DeviceState *dev_soc, Error **errp)
@@ -XXX,XX +XXX,XX @@ static void stm32l4x5_soc_realize(DeviceState *dev_soc, Error **errp)
      Stm32l4x5SocState *s = STM32L4X5_SOC(dev_soc);
      const Stm32l4x5SocClass *sc = STM32L4X5_SOC_GET_CLASS(dev_soc);
      MemoryRegion *system_memory = get_system_memory();
 -    DeviceState *armv7m;
 +    DeviceState *armv7m, *dev;
      SysBusDevice *busdev;
 +    uint32_t pin_index;
      if (!memory_region_init_rom(&s->flash, OBJECT(dev_soc), "flash",
                                  sc->flash_size, errp)) {
@@ -XXX,XX +XXX,XX @@ static void stm32l4x5_soc_realize(DeviceState *dev_soc, Error **errp)
          return;
      }
-@@ -XXX,XX +XXX,XX @@ static void cp_reg_check_reset(gpointer key, gpointer value,  gpointer opaque)
++    /* GPIOs */
-     ARMCPU *cpu = opaque;
++    for (unsigned i = 0; i < NUM_GPIOS; i++) {
-     uint64_t oldvalue, newvalue;
++        g_autofree char *name = g_strdup_printf("%c", 'A' + i);
++        dev = DEVICE(&s->gpio[i]);
--    if (ri->type & (ARM_CP_SPECIAL | ARM_CP_ALIAS | ARM_CP_NO_RAW)) {
++        qdev_prop_set_string(dev, "name", name);
-+    if (ri->type & (ARM_CP_SPECIAL_MASK | ARM_CP_ALIAS | ARM_CP_NO_RAW)) {
++        qdev_prop_set_uint32(dev, "mode-reset",
 +                             stm32l4x5_gpio_cfg[i].moder_reset);
 +        qdev_prop_set_uint32(dev, "ospeed-reset",
 +                             stm32l4x5_gpio_cfg[i].ospeedr_reset);
 +        qdev_prop_set_uint32(dev, "pupd-reset",
 +                            stm32l4x5_gpio_cfg[i].pupdr_reset);
 +        busdev = SYS_BUS_DEVICE(&s->gpio[i]);
 +        g_free(name);
 +        name = g_strdup_printf("gpio%c-out", 'a' + i);
 +        qdev_connect_clock_in(DEVICE(&s->gpio[i]), "clk",
 +            qdev_get_clock_out(DEVICE(&(s->rcc)), name));
 +        if (!sysbus_realize(busdev, errp)) {
 +            return;
 +        }
 +        sysbus_mmio_map(busdev, 0, stm32l4x5_gpio_cfg[i].addr);
 +    }
 +
      /* System configuration controller */
      busdev = SYS_BUS_DEVICE(&s->syscfg);
      if (!sysbus_realize(busdev, errp)) {
          return;
      }
+     sysbus_mmio_map(busdev, 0, SYSCFG_ADDR);
-diff --git a/target/arm/helper.c b/target/arm/helper.c
+-    /*
-index XXXXXXX..XXXXXXX 100644
+-     * TODO: when the GPIO device is implemented, connect it
---- a/target/arm/helper.c
+-     * to SYCFG using `qdev_connect_gpio_out`, NUM_GPIOS and
-+++ b/target/arm/helper.c
+-     * GPIO_NUM_PINS.
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
+-     */
-      * multiple times. Special registers (ie NOP/WFI) are
++
-      * never migratable and not even raw-accessible.
++    for (unsigned i = 0; i < NUM_GPIOS; i++) {
-      */
++        for (unsigned j = 0; j < GPIO_NUM_PINS; j++) {
--    if ((r->type & ARM_CP_SPECIAL)) {
++            pin_index = GPIO_NUM_PINS * i + j;
-+    if (r->type & ARM_CP_SPECIAL_MASK) {
++            qdev_connect_gpio_out(DEVICE(&s->gpio[i]), j,
-         r2->type |= ARM_CP_NO_RAW;
++                                  qdev_get_gpio_in(DEVICE(&s->syscfg),
-     }
++                                  pin_index));
-     if (((r->crm == CP_ANY) && crm != 0) ||
++        }
-@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
++    }
-     /* Check that the register definition has enough info to handle
-      * reads and writes if they are permitted.
+     /* EXTI device */
-      */
+     busdev = SYS_BUS_DEVICE(&s->exti);
--    if (!(r->type & (ARM_CP_SPECIAL|ARM_CP_CONST))) {
+@@ -XXX,XX +XXX,XX @@ static void stm32l4x5_soc_realize(DeviceState *dev_soc, Error **errp)
 +    if (!(r->type & (ARM_CP_SPECIAL_MASK | ARM_CP_CONST))) {
          if (r->access & PL3_R) {
              assert((r->fieldoffset ||
                     (r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1])) ||
 diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/translate-a64.c
 +++ b/target/arm/translate-a64.c
@@ -XXX,XX +XXX,XX @@ static void handle_sys(DisasContext *s, uint32_t insn, bool isread,
      }
      /* Handle special cases first */
 -    switch (ri->type & ~(ARM_CP_FLAG_MASK & ~ARM_CP_SPECIAL)) {
 +    switch (ri->type & ARM_CP_SPECIAL_MASK) {
 +    case 0:
 +        break;
      case ARM_CP_NOP:
          return;
      case ARM_CP_NZCV:
@@ -XXX,XX +XXX,XX @@ static void handle_sys(DisasContext *s, uint32_t insn, bool isread,
          }
-         return;
+     }
-     default:
--        break;
+-    for (unsigned i = 0; i < 16; i++) {
-+        g_assert_not_reached();
++    for (unsigned i = 0; i < GPIO_NUM_PINS; i++) {
-     }
+         qdev_connect_gpio_out(DEVICE(&s->syscfg), i,
-     if ((ri->type & ARM_CP_FPU) && !fp_access_check(s)) {
+                               qdev_get_gpio_in(DEVICE(&s->exti), i));
-         return;
+     }
-diff --git a/target/arm/translate.c b/target/arm/translate.c
+@@ -XXX,XX +XXX,XX @@ static void stm32l4x5_soc_realize(DeviceState *dev_soc, Error **errp)
-index XXXXXXX..XXXXXXX 100644
+     /* RESERVED:    0x40024400, 0x7FDBC00 */
---- a/target/arm/translate.c
-+++ b/target/arm/translate.c
+     /* AHB2 BUS */
-@@ -XXX,XX +XXX,XX @@ static void do_coproc_insn(DisasContext *s, int cpnum, int is64,
+-    create_unimplemented_device("GPIOA",     0x48000000, 0x400);
-         }
+-    create_unimplemented_device("GPIOB",     0x48000400, 0x400);
+-    create_unimplemented_device("GPIOC",     0x48000800, 0x400);
-         /* Handle special cases first */
+-    create_unimplemented_device("GPIOD",     0x48000C00, 0x400);
--        switch (ri->type & ~(ARM_CP_FLAG_MASK & ~ARM_CP_SPECIAL)) {
+-    create_unimplemented_device("GPIOE",     0x48001000, 0x400);
-+        switch (ri->type & ARM_CP_SPECIAL_MASK) {
+-    create_unimplemented_device("GPIOF",     0x48001400, 0x400);
-+        case 0:
+-    create_unimplemented_device("GPIOG",     0x48001800, 0x400);
-+            break;
+-    create_unimplemented_device("GPIOH",     0x48001C00, 0x400);
-         case ARM_CP_NOP:
+     /* RESERVED:    0x48002000, 0x7FDBC00 */
-             return;
+     create_unimplemented_device("OTG_FS",    0x50000000, 0x40000);
-         case ARM_CP_WFI:
+     create_unimplemented_device("ADC",       0x50040000, 0x400);
-@@ -XXX,XX +XXX,XX @@ static void do_coproc_insn(DisasContext *s, int cpnum, int is64,
+diff --git a/hw/misc/stm32l4x5_syscfg.c b/hw/misc/stm32l4x5_syscfg.c
-             s->base.is_jmp = DISAS_WFI;
+index XXXXXXX..XXXXXXX 100644
-             return;
+--- a/hw/misc/stm32l4x5_syscfg.c
-         default:
++++ b/hw/misc/stm32l4x5_syscfg.c
--            break;
+@@ -XXX,XX +XXX,XX @@
-+            g_assert_not_reached();
+ #include "hw/irq.h"
-         }
+ #include "migration/vmstate.h"
+ #include "hw/misc/stm32l4x5_syscfg.h"
-         if ((tb_cflags(s->base.tb) & CF_USE_ICOUNT) && (ri->type & ARM_CP_IO)) {
++#include "hw/gpio/stm32l4x5_gpio.h"
  #define SYSCFG_MEMRMP 0x00
  #define SYSCFG_CFGR1 0x04
 diff --git a/hw/arm/Kconfig b/hw/arm/Kconfig
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/arm/Kconfig
 +++ b/hw/arm/Kconfig
@@ -XXX,XX +XXX,XX @@ config STM32L4X5_SOC
      bool
      select ARM_V7M
      select OR_IRQ
 -    select STM32L4X5_SYSCFG
      select STM32L4X5_EXTI
 +    select STM32L4X5_SYSCFG
      select STM32L4X5_RCC
 +    select STM32L4X5_GPIO
  config XLNX_ZYNQMP_ARM
      bool
 --
-.25.1
+.34.1

-[PULL 04/23] target/arm: Replace sentinels with ARRAY_SIZE in cpregs.h
+[PULL 11/14] tests/qtest: Add STM32L4x5 GPIO QTest testcase
-From: Richard Henderson <richard.henderson@linaro.org>
+From: Inès Varhol <ines.varhol@telecom-paris.fr>
-Remove a possible source of error by removing REGINFO_SENTINEL
+The testcase contains :
-and using ARRAY_SIZE (convinently hidden inside a macro) to
+- `test_idr_reset_value()` :
-find the end of the set of regs being registered or modified.
+Checks the reset values of MODER, OTYPER, PUPDR, ODR and IDR.
 - `test_gpio_output_mode()` :
 Checks that writing a bit in register ODR results in the corresponding
 pin rising or lowering, if this pin is configured in output mode.
 - `test_gpio_input_mode()` :
 Checks that a input pin set high or low externally results
 in the pin rising and lowering.
 - `test_pull_up_pull_down()` :
 Checks that a floating pin in pull-up/down mode is actually high/down.
 - `test_push_pull()` :
 Checks that a pin set externally is disconnected when configured in
 push-pull output mode, and can't be set externally while in this mode.
 - `test_open_drain()` :
 Checks that a pin set externally high is disconnected when configured
 in open-drain output mode, and can't be set high while in this mode.
 - `test_bsrr_brr()` :
 Checks that writing to BSRR and BRR has the desired result in ODR.
 - `test_clock_enable()` :
 Checks that GPIO clock is at the right frequency after enabling it.
-The space saved by not having the extra array element reduces
+Acked-by: Thomas Huth <thuth@redhat.com>
-the executable's .data.rel.ro section by about 9k.
+Signed-off-by: Arnaud Minier <arnaud.minier@telecom-paris.fr>
+Signed-off-by: Inès Varhol <ines.varhol@telecom-paris.fr>
-Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
+Message-id: 20240305210444.310665-4-ines.varhol@telecom-paris.fr
 Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
 Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
 Message-id: 20220501055028.646596-4-richard.henderson@linaro.org
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
 ---
- target/arm/cpregs.h       |  53 +++++++++---------
+ tests/qtest/stm32l4x5_gpio-test.c | 551 ++++++++++++++++++++++++++++++
- hw/arm/pxa2xx.c           |   1 -
+ tests/qtest/meson.build           |   3 +-
- hw/arm/pxa2xx_pic.c       |   1 -
+files changed, 553 insertions(+), 1 deletion(-)
- hw/intc/arm_gicv3_cpuif.c |   5 --
+ create mode 100644 tests/qtest/stm32l4x5_gpio-test.c
  hw/intc/arm_gicv3_kvm.c   |   1 -
  target/arm/cpu64.c        |   1 -
  target/arm/cpu_tcg.c      |   4 --
  target/arm/helper.c       | 111 ++++++++------------------------------
 files changed, 48 insertions(+), 129 deletions(-)
-diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
+diff --git a/tests/qtest/stm32l4x5_gpio-test.c b/tests/qtest/stm32l4x5_gpio-test.c
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
 +++ b/tests/qtest/stm32l4x5_gpio-test.c
@@ -XXX,XX +XXX,XX @@
 +/*
 + * QTest testcase for STM32L4x5_GPIO
 + *
 + * Copyright (c) 2024 Arnaud Minier <arnaud.minier@telecom-paris.fr>
 + * Copyright (c) 2024 Inès Varhol <ines.varhol@telecom-paris.fr>
 + *
 + * This work is licensed under the terms of the GNU GPL, version 2 or later.
 + * See the COPYING file in the top-level directory.
 + */
 +
 +#include "qemu/osdep.h"
 +#include "libqtest-single.h"
 +
 +#define GPIO_BASE_ADDR 0x48000000
 +#define GPIO_SIZE      0x400
 +#define NUM_GPIOS      8
 +#define NUM_GPIO_PINS  16
 +
 +#define GPIO_A 0x48000000
 +#define GPIO_B 0x48000400
 +#define GPIO_C 0x48000800
 +#define GPIO_D 0x48000C00
 +#define GPIO_E 0x48001000
 +#define GPIO_F 0x48001400
 +#define GPIO_G 0x48001800
 +#define GPIO_H 0x48001C00
 +
 +#define MODER 0x00
 +#define OTYPER 0x04
 +#define PUPDR 0x0C
 +#define IDR 0x10
 +#define ODR 0x14
 +#define BSRR 0x18
 +#define BRR 0x28
 +
 +#define MODER_INPUT 0
 +#define MODER_OUTPUT 1
 +
 +#define PUPDR_NONE 0
 +#define PUPDR_PULLUP 1
 +#define PUPDR_PULLDOWN 2
 +
 +#define OTYPER_PUSH_PULL 0
 +#define OTYPER_OPEN_DRAIN 1
 +
 +const uint32_t moder_reset[NUM_GPIOS] = {
 +    0xABFFFFFF,
 +    0xFFFFFEBF,
 +    0xFFFFFFFF,
 +    0xFFFFFFFF,
 +    0xFFFFFFFF,
 +    0xFFFFFFFF,
 +    0xFFFFFFFF,
 +    0x0000000F
 +};
 +
 +const uint32_t pupdr_reset[NUM_GPIOS] = {
 +    0x64000000,
 +    0x00000100,
 +    0x00000000,
 +    0x00000000,
 +    0x00000000,
 +    0x00000000,
 +    0x00000000,
 +    0x00000000
 +};
 +
 +const uint32_t idr_reset[NUM_GPIOS] = {
 +    0x0000A000,
 +    0x00000010,
 +    0x00000000,
 +    0x00000000,
 +    0x00000000,
 +    0x00000000,
 +    0x00000000,
 +    0x00000000
 +};
 +
 +static uint32_t gpio_readl(unsigned int gpio, unsigned int offset)
 +{
 +    return readl(gpio + offset);
 +}
 +
 +static void gpio_writel(unsigned int gpio, unsigned int offset, uint32_t value)
 +{
 +    writel(gpio + offset, value);
 +}
 +
 +static void gpio_set_bit(unsigned int gpio, unsigned int reg,
 +                         unsigned int pin, uint32_t value)
 +{
 +    uint32_t mask = 0xFFFFFFFF & ~(0x1 << pin);
 +    gpio_writel(gpio, reg, (gpio_readl(gpio, reg) & mask) | value << pin);
 +}
 +
 +static void gpio_set_2bits(unsigned int gpio, unsigned int reg,
 +                           unsigned int pin, uint32_t value)
 +{
 +    uint32_t offset = 2 * pin;
 +    uint32_t mask = 0xFFFFFFFF & ~(0x3 << offset);
 +    gpio_writel(gpio, reg, (gpio_readl(gpio, reg) & mask) | value << offset);
 +}
 +
 +static unsigned int get_gpio_id(uint32_t gpio_addr)
 +{
 +    return (gpio_addr - GPIO_BASE_ADDR) / GPIO_SIZE;
 +}
 +
 +static void gpio_set_irq(unsigned int gpio, int num, int level)
 +{
 +    g_autofree char *name = g_strdup_printf("/machine/soc/gpio%c",
 +                                            get_gpio_id(gpio) + 'a');
 +    qtest_set_irq_in(global_qtest, name, NULL, num, level);
 +}
 +
 +static void disconnect_all_pins(unsigned int gpio)
 +{
 +    g_autofree char *path = g_strdup_printf("/machine/soc/gpio%c",
 +                                            get_gpio_id(gpio) + 'a');
 +    QDict *r;
 +
 +    r = qtest_qmp(global_qtest, "{ 'execute': 'qom-set', 'arguments': "
 +        "{ 'path': %s, 'property': 'disconnected-pins', 'value': %d } }",
 +        path, 0xFFFF);
 +    g_assert_false(qdict_haskey(r, "error"));
 +    qobject_unref(r);
 +}
 +
 +static uint32_t get_disconnected_pins(unsigned int gpio)
 +{
 +    g_autofree char *path = g_strdup_printf("/machine/soc/gpio%c",
 +                                            get_gpio_id(gpio) + 'a');
 +    uint32_t disconnected_pins = 0;
 +    QDict *r;
 +
 +    r = qtest_qmp(global_qtest, "{ 'execute': 'qom-get', 'arguments':"
 +        " { 'path': %s, 'property': 'disconnected-pins'} }", path);
 +    g_assert_false(qdict_haskey(r, "error"));
 +    disconnected_pins = qdict_get_int(r, "return");
 +    qobject_unref(r);
 +    return disconnected_pins;
 +}
 +
 +static uint32_t reset(uint32_t gpio, unsigned int offset)
 +{
 +    switch (offset) {
 +    case MODER:
 +        return moder_reset[get_gpio_id(gpio)];
 +    case PUPDR:
 +        return pupdr_reset[get_gpio_id(gpio)];
 +    case IDR:
 +        return idr_reset[get_gpio_id(gpio)];
 +    }
 +    return 0x0;
 +}
 +
 +static void system_reset(void)
 +{
 +    QDict *r;
 +    r = qtest_qmp(global_qtest, "{'execute': 'system_reset'}");
 +    g_assert_false(qdict_haskey(r, "error"));
 +    qobject_unref(r);
 +}
 +
 +static void test_idr_reset_value(void)
 +{
 +    /*
 +     * Checks that the values in MODER, OTYPER, PUPDR and ODR
 +     * after reset are correct, and that the value in IDR is
 +     * coherent.
 +     * Since AF and analog modes aren't implemented, IDR reset
 +     * values aren't the same as with a real board.
 +     *
 +     * Register IDR contains the actual values of all GPIO pins.
 +     * Its value depends on the pins' configuration
 +     * (intput/output/analog : register MODER, push-pull/open-drain :
 +     * register OTYPER, pull-up/pull-down/none : register PUPDR)
 +     * and on the values stored in register ODR
 +     * (in case the pin is in output mode).
 +     */
 +
 +    gpio_writel(GPIO_A, MODER, 0xDEADBEEF);
 +    gpio_writel(GPIO_A, ODR, 0xDEADBEEF);
 +    gpio_writel(GPIO_A, OTYPER, 0xDEADBEEF);
 +    gpio_writel(GPIO_A, PUPDR, 0xDEADBEEF);
 +
 +    gpio_writel(GPIO_B, MODER, 0xDEADBEEF);
 +    gpio_writel(GPIO_B, ODR, 0xDEADBEEF);
 +    gpio_writel(GPIO_B, OTYPER, 0xDEADBEEF);
 +    gpio_writel(GPIO_B, PUPDR, 0xDEADBEEF);
 +
 +    gpio_writel(GPIO_C, MODER, 0xDEADBEEF);
 +    gpio_writel(GPIO_C, ODR, 0xDEADBEEF);
 +    gpio_writel(GPIO_C, OTYPER, 0xDEADBEEF);
 +    gpio_writel(GPIO_C, PUPDR, 0xDEADBEEF);
 +
 +    gpio_writel(GPIO_H, MODER, 0xDEADBEEF);
 +    gpio_writel(GPIO_H, ODR, 0xDEADBEEF);
 +    gpio_writel(GPIO_H, OTYPER, 0xDEADBEEF);
 +    gpio_writel(GPIO_H, PUPDR, 0xDEADBEEF);
 +
 +    system_reset();
 +
 +    uint32_t moder = gpio_readl(GPIO_A, MODER);
 +    uint32_t odr = gpio_readl(GPIO_A, ODR);
 +    uint32_t otyper = gpio_readl(GPIO_A, OTYPER);
 +    uint32_t pupdr = gpio_readl(GPIO_A, PUPDR);
 +    uint32_t idr = gpio_readl(GPIO_A, IDR);
 +    /* 15: AF, 14: AF, 13: AF, 12: Analog ... */
 +    /* here AF is the same as Analog and Input mode */
 +    g_assert_cmphex(moder, ==, reset(GPIO_A, MODER));
 +    g_assert_cmphex(odr, ==, reset(GPIO_A, ODR));
 +    g_assert_cmphex(otyper, ==, reset(GPIO_A, OTYPER));
 +    /* 15: pull-up, 14: pull-down, 13: pull-up, 12: neither ... */
 +    g_assert_cmphex(pupdr, ==, reset(GPIO_A, PUPDR));
 +    /* 15 : 1, 14: 0, 13: 1, 12 : reset value ... */
 +    g_assert_cmphex(idr, ==, reset(GPIO_A, IDR));
 +
 +    moder = gpio_readl(GPIO_B, MODER);
 +    odr = gpio_readl(GPIO_B, ODR);
 +    otyper = gpio_readl(GPIO_B, OTYPER);
 +    pupdr = gpio_readl(GPIO_B, PUPDR);
 +    idr = gpio_readl(GPIO_B, IDR);
 +    /* ... 5: Analog, 4: AF, 3: AF, 2: Analog ... */
 +    /* here AF is the same as Analog and Input mode */
 +    g_assert_cmphex(moder, ==, reset(GPIO_B, MODER));
 +    g_assert_cmphex(odr, ==, reset(GPIO_B, ODR));
 +    g_assert_cmphex(otyper, ==, reset(GPIO_B, OTYPER));
 +    /* ... 5: neither, 4: pull-up, 3: neither ... */
 +    g_assert_cmphex(pupdr, ==, reset(GPIO_B, PUPDR));
 +    /* ... 5 : reset value, 4 : 1, 3 : reset value ... */
 +    g_assert_cmphex(idr, ==, reset(GPIO_B, IDR));
 +
 +    moder = gpio_readl(GPIO_C, MODER);
 +    odr = gpio_readl(GPIO_C, ODR);
 +    otyper = gpio_readl(GPIO_C, OTYPER);
 +    pupdr = gpio_readl(GPIO_C, PUPDR);
 +    idr = gpio_readl(GPIO_C, IDR);
 +    /* Analog, same as Input mode*/
 +    g_assert_cmphex(moder, ==, reset(GPIO_C, MODER));
 +    g_assert_cmphex(odr, ==, reset(GPIO_C, ODR));
 +    g_assert_cmphex(otyper, ==, reset(GPIO_C, OTYPER));
 +    /* no pull-up or pull-down */
 +    g_assert_cmphex(pupdr, ==, reset(GPIO_C, PUPDR));
 +    /* reset value */
 +    g_assert_cmphex(idr, ==, reset(GPIO_C, IDR));
 +
 +    moder = gpio_readl(GPIO_H, MODER);
 +    odr = gpio_readl(GPIO_H, ODR);
 +    otyper = gpio_readl(GPIO_H, OTYPER);
 +    pupdr = gpio_readl(GPIO_H, PUPDR);
 +    idr = gpio_readl(GPIO_H, IDR);
 +    /* Analog, same as Input mode */
 +    g_assert_cmphex(moder, ==, reset(GPIO_H, MODER));
 +    g_assert_cmphex(odr, ==, reset(GPIO_H, ODR));
 +    g_assert_cmphex(otyper, ==, reset(GPIO_H, OTYPER));
 +    /* no pull-up or pull-down */
 +    g_assert_cmphex(pupdr, ==, reset(GPIO_H, PUPDR));
 +    /* reset value */
 +    g_assert_cmphex(idr, ==, reset(GPIO_H, IDR));
 +}
 +
 +static void test_gpio_output_mode(const void *data)
 +{
 +    /*
 +     * Checks that setting a bit in ODR sets the corresponding
 +     * GPIO line high : it should set the right bit in IDR
 +     * and send an irq to syscfg.
 +     * Additionally, it checks that values written to ODR
 +     * when not in output mode are stored and not discarded.
 +     */
 +    unsigned int pin = ((uint64_t)data) & 0xF;
 +    uint32_t gpio = ((uint64_t)data) >> 32;
 +    unsigned int gpio_id = get_gpio_id(gpio);
 +
 +    qtest_irq_intercept_in(global_qtest, "/machine/soc/syscfg");
 +
 +    /* Set a bit in ODR and check nothing happens */
 +    gpio_set_bit(gpio, ODR, pin, 1);
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR));
 +    g_assert_false(get_irq(gpio_id * NUM_GPIO_PINS + pin));
 +
 +    /* Configure the relevant line as output and check the pin is high */
 +    gpio_set_2bits(gpio, MODER, pin, MODER_OUTPUT);
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) | (1 << pin));
 +    g_assert_true(get_irq(gpio_id * NUM_GPIO_PINS + pin));
 +
 +    /* Reset the bit in ODR and check the pin is low */
 +    gpio_set_bit(gpio, ODR, pin, 0);
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
 +    g_assert_false(get_irq(gpio_id * NUM_GPIO_PINS + pin));
 +
 +    /* Clean the test */
 +    gpio_writel(gpio, ODR, reset(gpio, ODR));
 +    gpio_writel(gpio, MODER, reset(gpio, MODER));
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR));
 +    g_assert_false(get_irq(gpio_id * NUM_GPIO_PINS + pin));
 +}
 +
 +static void test_gpio_input_mode(const void *data)
 +{
 +    /*
 +     * Test that setting a line high/low externally sets the
 +     * corresponding GPIO line high/low : it should set the
 +     * right bit in IDR and send an irq to syscfg.
 +     */
 +    unsigned int pin = ((uint64_t)data) & 0xF;
 +    uint32_t gpio = ((uint64_t)data) >> 32;
 +    unsigned int gpio_id = get_gpio_id(gpio);
 +
 +    qtest_irq_intercept_in(global_qtest, "/machine/soc/syscfg");
 +
 +    /* Configure a line as input, raise it, and check that the pin is high */
 +    gpio_set_2bits(gpio, MODER, pin, MODER_INPUT);
 +    gpio_set_irq(gpio, pin, 1);
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) | (1 << pin));
 +    g_assert_true(get_irq(gpio_id * NUM_GPIO_PINS + pin));
 +
 +    /* Lower the line and check that the pin is low */
 +    gpio_set_irq(gpio, pin, 0);
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
 +    g_assert_false(get_irq(gpio_id * NUM_GPIO_PINS + pin));
 +
 +    /* Clean the test */
 +    gpio_writel(gpio, MODER, reset(gpio, MODER));
 +    disconnect_all_pins(gpio);
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR));
 +}
 +
 +static void test_pull_up_pull_down(const void *data)
 +{
 +    /*
 +     * Test that a floating pin with pull-up sets the pin
 +     * high and vice-versa.
 +     */
 +    unsigned int pin = ((uint64_t)data) & 0xF;
 +    uint32_t gpio = ((uint64_t)data) >> 32;
 +    unsigned int gpio_id = get_gpio_id(gpio);
 +
 +    qtest_irq_intercept_in(global_qtest, "/machine/soc/syscfg");
 +
 +    /* Configure a line as input with pull-up, check the line is set high */
 +    gpio_set_2bits(gpio, MODER, pin, MODER_INPUT);
 +    gpio_set_2bits(gpio, PUPDR, pin, PUPDR_PULLUP);
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) | (1 << pin));
 +    g_assert_true(get_irq(gpio_id * NUM_GPIO_PINS + pin));
 +
 +    /* Configure the line with pull-down, check the line is low */
 +    gpio_set_2bits(gpio, PUPDR, pin, PUPDR_PULLDOWN);
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
 +    g_assert_false(get_irq(gpio_id * NUM_GPIO_PINS + pin));
 +
 +    /* Clean the test */
 +    gpio_writel(gpio, MODER, reset(gpio, MODER));
 +    gpio_writel(gpio, PUPDR, reset(gpio, PUPDR));
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR));
 +}
 +
 +static void test_push_pull(const void *data)
 +{
 +    /*
 +     * Test that configuring a line in push-pull output mode
 +     * disconnects the pin, that the pin can't be set or reset
 +     * externally afterwards.
 +     */
 +    unsigned int pin = ((uint64_t)data) & 0xF;
 +    uint32_t gpio = ((uint64_t)data) >> 32;
 +    uint32_t gpio2 = GPIO_BASE_ADDR + (GPIO_H - gpio);
 +
 +    qtest_irq_intercept_in(global_qtest, "/machine/soc/syscfg");
 +
 +    /* Setting a line high externally, configuring it in push-pull output */
 +    /* And checking the pin was disconnected */
 +    gpio_set_irq(gpio, pin, 1);
 +    gpio_set_2bits(gpio, MODER, pin, MODER_OUTPUT);
 +    g_assert_cmphex(get_disconnected_pins(gpio), ==, 0xFFFF);
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
 +
 +    /* Setting a line low externally, configuring it in push-pull output */
 +    /* And checking the pin was disconnected */
 +    gpio_set_irq(gpio2, pin, 0);
 +    gpio_set_bit(gpio2, ODR, pin, 1);
 +    gpio_set_2bits(gpio2, MODER, pin, MODER_OUTPUT);
 +    g_assert_cmphex(get_disconnected_pins(gpio2), ==, 0xFFFF);
 +    g_assert_cmphex(gpio_readl(gpio2, IDR), ==, reset(gpio2, IDR) | (1 << pin));
 +
 +    /* Trying to set a push-pull output pin, checking it doesn't work */
 +    gpio_set_irq(gpio, pin, 1);
 +    g_assert_cmphex(get_disconnected_pins(gpio), ==, 0xFFFF);
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
 +
 +    /* Trying to reset a push-pull output pin, checking it doesn't work */
 +    gpio_set_irq(gpio2, pin, 0);
 +    g_assert_cmphex(get_disconnected_pins(gpio2), ==, 0xFFFF);
 +    g_assert_cmphex(gpio_readl(gpio2, IDR), ==, reset(gpio2, IDR) | (1 << pin));
 +
 +    /* Clean the test */
 +    gpio_writel(gpio, MODER, reset(gpio, MODER));
 +    gpio_writel(gpio2, ODR, reset(gpio2, ODR));
 +    gpio_writel(gpio2, MODER, reset(gpio2, MODER));
 +}
 +
 +static void test_open_drain(const void *data)
 +{
 +    /*
 +     * Test that configuring a line in open-drain output mode
 +     * disconnects a pin set high externally and that the pin
 +     * can't be set high externally while configured in open-drain.
 +     *
 +     * However a pin set low externally shouldn't be disconnected,
 +     * and it can be set low externally when in open-drain mode.
 +     */
 +    unsigned int pin = ((uint64_t)data) & 0xF;
 +    uint32_t gpio = ((uint64_t)data) >> 32;
 +    uint32_t gpio2 = GPIO_BASE_ADDR + (GPIO_H - gpio);
 +
 +    qtest_irq_intercept_in(global_qtest, "/machine/soc/syscfg");
 +
 +    /* Setting a line high externally, configuring it in open-drain output */
 +    /* And checking the pin was disconnected */
 +    gpio_set_irq(gpio, pin, 1);
 +    gpio_set_bit(gpio, OTYPER, pin, OTYPER_OPEN_DRAIN);
 +    gpio_set_2bits(gpio, MODER, pin, MODER_OUTPUT);
 +    g_assert_cmphex(get_disconnected_pins(gpio), ==, 0xFFFF);
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
 +
 +    /* Setting a line low externally, configuring it in open-drain output */
 +    /* And checking the pin wasn't disconnected */
 +    gpio_set_irq(gpio2, pin, 0);
 +    gpio_set_bit(gpio2, ODR, pin, 1);
 +    gpio_set_bit(gpio2, OTYPER, pin, OTYPER_OPEN_DRAIN);
 +    gpio_set_2bits(gpio2, MODER, pin, MODER_OUTPUT);
 +    g_assert_cmphex(get_disconnected_pins(gpio2), ==, 0xFFFF & ~(1 << pin));
 +    g_assert_cmphex(gpio_readl(gpio2, IDR), ==,
 +                               reset(gpio2, IDR) & ~(1 << pin));
 +
 +    /* Trying to set a open-drain output pin, checking it doesn't work */
 +    gpio_set_irq(gpio, pin, 1);
 +    g_assert_cmphex(get_disconnected_pins(gpio), ==, 0xFFFF);
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
 +
 +    /* Trying to reset a open-drain output pin, checking it works */
 +    gpio_set_bit(gpio, ODR, pin, 1);
 +    gpio_set_irq(gpio, pin, 0);
 +    g_assert_cmphex(get_disconnected_pins(gpio2), ==, 0xFFFF & ~(1 << pin));
 +    g_assert_cmphex(gpio_readl(gpio2, IDR), ==,
 +                               reset(gpio2, IDR) & ~(1 << pin));
 +
 +    /* Clean the test */
 +    disconnect_all_pins(gpio2);
 +    gpio_writel(gpio2, OTYPER, reset(gpio2, OTYPER));
 +    gpio_writel(gpio2, ODR, reset(gpio2, ODR));
 +    gpio_writel(gpio2, MODER, reset(gpio2, MODER));
 +    g_assert_cmphex(gpio_readl(gpio2, IDR), ==, reset(gpio2, IDR));
 +    disconnect_all_pins(gpio);
 +    gpio_writel(gpio, OTYPER, reset(gpio, OTYPER));
 +    gpio_writel(gpio, ODR, reset(gpio, ODR));
 +    gpio_writel(gpio, MODER, reset(gpio, MODER));
 +    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR));
 +}
 +
 +static void test_bsrr_brr(const void *data)
 +{
 +    /*
 +     * Test that writing a '1' in BSS and BSRR
 +     * has the desired effect on ODR.
 +     * In BSRR, BSx has priority over BRx.
 +     */
 +    unsigned int pin = ((uint64_t)data) & 0xF;
 +    uint32_t gpio = ((uint64_t)data) >> 32;
 +
 +    gpio_writel(gpio, BSRR, (1 << pin));
 +    g_assert_cmphex(gpio_readl(gpio, ODR), ==, reset(gpio, ODR) | (1 << pin));
 +
 +    gpio_writel(gpio, BSRR, (1 << (pin + NUM_GPIO_PINS)));
 +    g_assert_cmphex(gpio_readl(gpio, ODR), ==, reset(gpio, ODR));
 +
 +    gpio_writel(gpio, BSRR, (1 << pin));
 +    g_assert_cmphex(gpio_readl(gpio, ODR), ==, reset(gpio, ODR) | (1 << pin));
 +
 +    gpio_writel(gpio, BRR, (1 << pin));
 +    g_assert_cmphex(gpio_readl(gpio, ODR), ==, reset(gpio, ODR));
 +
 +    /* BSx should have priority over BRx */
 +    gpio_writel(gpio, BSRR, (1 << pin) | (1 << (pin + NUM_GPIO_PINS)));
 +    g_assert_cmphex(gpio_readl(gpio, ODR), ==, reset(gpio, ODR) | (1 << pin));
 +
 +    gpio_writel(gpio, BRR, (1 << pin));
 +    g_assert_cmphex(gpio_readl(gpio, ODR), ==, reset(gpio, ODR));
 +
 +    gpio_writel(gpio, ODR, reset(gpio, ODR));
 +}
 +
 +int main(int argc, char **argv)
 +{
 +    int ret;
 +
 +    g_test_init(&argc, &argv, NULL);
 +    g_test_set_nonfatal_assertions();
 +    qtest_add_func("stm32l4x5/gpio/test_idr_reset_value",
 +                   test_idr_reset_value);
 +    /*
 +     * The inputs for the tests (gpio and pin) can be changed,
 +     * but the tests don't work for pins that are high at reset
 +     * (GPIOA15, GPIO13 and GPIOB5).
 +     * Specifically, rising the pin then checking `get_irq()`
 +     * is problematic since the pin was already high.
 +     */
 +    qtest_add_data_func("stm32l4x5/gpio/test_gpioc5_output_mode",
 +                        (void *)((uint64_t)GPIO_C << 32 | 5),
 +                        test_gpio_output_mode);
 +    qtest_add_data_func("stm32l4x5/gpio/test_gpioh3_output_mode",
 +                        (void *)((uint64_t)GPIO_H << 32 | 3),
 +                        test_gpio_output_mode);
 +    qtest_add_data_func("stm32l4x5/gpio/test_gpio_input_mode1",
 +                        (void *)((uint64_t)GPIO_D << 32 | 6),
 +                        test_gpio_input_mode);
 +    qtest_add_data_func("stm32l4x5/gpio/test_gpio_input_mode2",
 +                        (void *)((uint64_t)GPIO_C << 32 | 10),
 +                        test_gpio_input_mode);
 +    qtest_add_data_func("stm32l4x5/gpio/test_gpio_pull_up_pull_down1",
 +                        (void *)((uint64_t)GPIO_B << 32 | 5),
 +                        test_pull_up_pull_down);
 +    qtest_add_data_func("stm32l4x5/gpio/test_gpio_pull_up_pull_down2",
 +                        (void *)((uint64_t)GPIO_F << 32 | 1),
 +                        test_pull_up_pull_down);
 +    qtest_add_data_func("stm32l4x5/gpio/test_gpio_push_pull1",
 +                        (void *)((uint64_t)GPIO_G << 32 | 6),
 +                        test_push_pull);
 +    qtest_add_data_func("stm32l4x5/gpio/test_gpio_push_pull2",
 +                        (void *)((uint64_t)GPIO_H << 32 | 3),
 +                        test_push_pull);
 +    qtest_add_data_func("stm32l4x5/gpio/test_gpio_open_drain1",
 +                        (void *)((uint64_t)GPIO_C << 32 | 4),
 +                        test_open_drain);
 +    qtest_add_data_func("stm32l4x5/gpio/test_gpio_open_drain2",
 +                        (void *)((uint64_t)GPIO_E << 32 | 11),
 +                        test_open_drain);
 +    qtest_add_data_func("stm32l4x5/gpio/test_bsrr_brr1",
 +                        (void *)((uint64_t)GPIO_A << 32 | 12),
 +                        test_bsrr_brr);
 +    qtest_add_data_func("stm32l4x5/gpio/test_bsrr_brr2",
 +                        (void *)((uint64_t)GPIO_D << 32 | 0),
 +                        test_bsrr_brr);
 +
 +    qtest_start("-machine b-l475e-iot01a");
 +    ret = g_test_run();
 +    qtest_end();
 +
 +    return ret;
 +}
 diff --git a/tests/qtest/meson.build b/tests/qtest/meson.build
 index XXXXXXX..XXXXXXX 100644
---- a/target/arm/cpregs.h
+--- a/tests/qtest/meson.build
-+++ b/target/arm/cpregs.h
++++ b/tests/qtest/meson.build
-@@ -XXX,XX +XXX,XX @@
+@@ -XXX,XX +XXX,XX @@ qtests_aspeed = \
- #define ARM_CP_NO_GDB            0x4000
+ qtests_stm32l4x5 = \
- #define ARM_CP_RAISES_EXC        0x8000
+   ['stm32l4x5_exti-test',
- #define ARM_CP_NEWEL             0x10000
+    'stm32l4x5_syscfg-test',
--/* Used only as a terminator for ARMCPRegInfo lists */
+-   'stm32l4x5_rcc-test']
--#define ARM_CP_SENTINEL          0xfffff
++   'stm32l4x5_rcc-test',
- /* Mask of only the flag bits in a type field */
++   'stm32l4x5_gpio-test']
- #define ARM_CP_FLAG_MASK         0x1f0ff
+ qtests_arm = \
-@@ -XXX,XX +XXX,XX @@ enum {
+   (config_all_devices.has_key('CONFIG_MPS2') ? ['sse-timer-test'] : []) + \
      ARM_CP_SECSTATE_NS =  (1 << 1), /* bit[1]: Non-secure state register */
  };
 -/*
 - * Return true if cptype is a valid type field. This is used to try to
 - * catch errors where the sentinel has been accidentally left off the end
 - * of a list of registers.
 - */
 -static inline bool cptype_valid(int cptype)
 -{
 -    return ((cptype & ~ARM_CP_FLAG_MASK) == 0)
 -        || ((cptype & ARM_CP_SPECIAL) &&
 -            ((cptype & ~ARM_CP_FLAG_MASK) <= ARM_LAST_SPECIAL));
 -}
 -
  /*
   * Access rights:
   * We define bits for Read and Write access for what rev C of the v7-AR ARM ARM
@@ -XXX,XX +XXX,XX @@ struct ARMCPRegInfo {
  #define CPREG_FIELD64(env, ri) \
      (*(uint64_t *)((char *)(env) + (ri)->fieldoffset))
 -#define REGINFO_SENTINEL { .type = ARM_CP_SENTINEL }
 +void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu, const ARMCPRegInfo *reg,
 +                                       void *opaque);
 -void define_arm_cp_regs_with_opaque(ARMCPU *cpu,
 -                                    const ARMCPRegInfo *regs, void *opaque);
 -void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
 -                                       const ARMCPRegInfo *regs, void *opaque);
 -static inline void define_arm_cp_regs(ARMCPU *cpu, const ARMCPRegInfo *regs)
 -{
 -    define_arm_cp_regs_with_opaque(cpu, regs, 0);
 -}
  static inline void define_one_arm_cp_reg(ARMCPU *cpu, const ARMCPRegInfo *regs)
  {
 -    define_one_arm_cp_reg_with_opaque(cpu, regs, 0);
 +    define_one_arm_cp_reg_with_opaque(cpu, regs, NULL);
  }
 +
 +void define_arm_cp_regs_with_opaque_len(ARMCPU *cpu, const ARMCPRegInfo *regs,
 +                                        void *opaque, size_t len);
 +
 +#define define_arm_cp_regs_with_opaque(CPU, REGS, OPAQUE)               \
 +    do {                                                                \
 +        QEMU_BUILD_BUG_ON(ARRAY_SIZE(REGS) == 0);                       \
 +        define_arm_cp_regs_with_opaque_len(CPU, REGS, OPAQUE,           \
 +                                           ARRAY_SIZE(REGS));           \
 +    } while (0)
 +
 +#define define_arm_cp_regs(CPU, REGS) \
 +    define_arm_cp_regs_with_opaque(CPU, REGS, NULL)
 +
  const ARMCPRegInfo *get_arm_cp_reginfo(GHashTable *cpregs, uint32_t encoded_cp);
  /*
@@ -XXX,XX +XXX,XX @@ typedef struct ARMCPRegUserSpaceInfo {
      uint64_t fixed_bits;
  } ARMCPRegUserSpaceInfo;
 -#define REGUSERINFO_SENTINEL { .name = NULL }
 +void modify_arm_cp_regs_with_len(ARMCPRegInfo *regs, size_t regs_len,
 +                                 const ARMCPRegUserSpaceInfo *mods,
 +                                 size_t mods_len);
 -void modify_arm_cp_regs(ARMCPRegInfo *regs, const ARMCPRegUserSpaceInfo *mods);
 +#define modify_arm_cp_regs(REGS, MODS)                                  \
 +    do {                                                                \
 +        QEMU_BUILD_BUG_ON(ARRAY_SIZE(REGS) == 0);                       \
 +        QEMU_BUILD_BUG_ON(ARRAY_SIZE(MODS) == 0);                       \
 +        modify_arm_cp_regs_with_len(REGS, ARRAY_SIZE(REGS),             \
 +                                    MODS, ARRAY_SIZE(MODS));            \
 +    } while (0)
  /* CPWriteFn that can be used to implement writes-ignored behaviour */
  void arm_cp_write_ignore(CPUARMState *env, const ARMCPRegInfo *ri,
 diff --git a/hw/arm/pxa2xx.c b/hw/arm/pxa2xx.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/arm/pxa2xx.c
 +++ b/hw/arm/pxa2xx.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo pxa_cp_reginfo[] = {
      { .name = "PWRMODE", .cp = 14, .crn = 7, .crm = 0, .opc1 = 0, .opc2 = 0,
        .access = PL1_RW, .type = ARM_CP_IO,
        .readfn = arm_cp_read_zero, .writefn = pxa2xx_pwrmode_write },
 -    REGINFO_SENTINEL
  };
  static void pxa2xx_setup_cp14(PXA2xxState *s)
 diff --git a/hw/arm/pxa2xx_pic.c b/hw/arm/pxa2xx_pic.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/arm/pxa2xx_pic.c
 +++ b/hw/arm/pxa2xx_pic.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo pxa_pic_cp_reginfo[] = {
      REGINFO_FOR_PIC_CP("ICLR2", 8),
      REGINFO_FOR_PIC_CP("ICFP2", 9),
      REGINFO_FOR_PIC_CP("ICPR2", 0xa),
 -    REGINFO_SENTINEL
  };
  static const MemoryRegionOps pxa2xx_pic_ops = {
 diff --git a/hw/intc/arm_gicv3_cpuif.c b/hw/intc/arm_gicv3_cpuif.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/intc/arm_gicv3_cpuif.c
 +++ b/hw/intc/arm_gicv3_cpuif.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo gicv3_cpuif_reginfo[] = {
        .readfn = icc_igrpen1_el3_read,
        .writefn = icc_igrpen1_el3_write,
      },
 -    REGINFO_SENTINEL
  };
  static uint64_t ich_ap_read(CPUARMState *env, const ARMCPRegInfo *ri)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo gicv3_cpuif_hcr_reginfo[] = {
        .readfn = ich_vmcr_read,
        .writefn = ich_vmcr_write,
      },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo gicv3_cpuif_ich_apxr1_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo gicv3_cpuif_ich_apxr1_reginfo[] = {
        .readfn = ich_ap_read,
        .writefn = ich_ap_write,
      },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo gicv3_cpuif_ich_apxr23_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo gicv3_cpuif_ich_apxr23_reginfo[] = {
        .readfn = ich_ap_read,
        .writefn = ich_ap_write,
      },
 -    REGINFO_SENTINEL
  };
  static void gicv3_cpuif_el_change_hook(ARMCPU *cpu, void *opaque)
@@ -XXX,XX +XXX,XX @@ void gicv3_init_cpuif(GICv3State *s)
                        .readfn = ich_lr_read,
                        .writefn = ich_lr_write,
                      },
 -                    REGINFO_SENTINEL
                  };
                  define_arm_cp_regs(cpu, lr_regset);
              }
 diff --git a/hw/intc/arm_gicv3_kvm.c b/hw/intc/arm_gicv3_kvm.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/intc/arm_gicv3_kvm.c
 +++ b/hw/intc/arm_gicv3_kvm.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo gicv3_cpuif_reginfo[] = {
         */
        .resetfn = arm_gicv3_icc_reset,
      },
 -    REGINFO_SENTINEL
  };
  /**
 diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/cpu64.c
 +++ b/target/arm/cpu64.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cortex_a72_a57_a53_cp_reginfo[] = {
      { .name = "L2MERRSR",
        .cp = 15, .opc1 = 3, .crm = 15,
        .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  static void aarch64_a57_initfn(Object *obj)
 diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/cpu_tcg.c
 +++ b/target/arm/cpu_tcg.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cortexa8_cp_reginfo[] = {
        .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 },
      { .name = "L2AUXCR", .cp = 15, .crn = 9, .crm = 0, .opc1 = 1, .opc2 = 2,
        .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  static void cortex_a8_initfn(Object *obj)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cortexa9_cp_reginfo[] = {
        .access = PL1_RW, .resetvalue = 0, .type = ARM_CP_CONST },
      { .name = "TLB_ATTR", .cp = 15, .crn = 15, .crm = 7, .opc1 = 5, .opc2 = 2,
        .access = PL1_RW, .resetvalue = 0, .type = ARM_CP_CONST },
 -    REGINFO_SENTINEL
  };
  static void cortex_a9_initfn(Object *obj)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cortexa15_cp_reginfo[] = {
  #endif
      { .name = "L2ECTLR", .cp = 15, .crn = 9, .crm = 0, .opc1 = 1, .opc2 = 3,
        .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  static void cortex_a7_initfn(Object *obj)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cortexr5_cp_reginfo[] = {
        .access = PL1_RW, .type = ARM_CP_CONST },
      { .name = "DCACHE_INVAL", .cp = 15, .opc1 = 0, .crn = 15, .crm = 5,
        .opc2 = 0, .access = PL1_W, .type = ARM_CP_NOP },
 -    REGINFO_SENTINEL
  };
  static void cortex_r5_initfn(Object *obj)
 diff --git a/target/arm/helper.c b/target/arm/helper.c
 index XXXXXXX..XXXXXXX 100644
 --- a/target/arm/helper.c
 +++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cp_reginfo[] = {
        .secure = ARM_CP_SECSTATE_S,
        .fieldoffset = offsetof(CPUARMState, cp15.contextidr_s),
        .resetvalue = 0, .writefn = contextidr_write, .raw_writefn = raw_write, },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo not_v8_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo not_v8_cp_reginfo[] = {
      { .name = "CACHEMAINT", .cp = 15, .crn = 7, .crm = CP_ANY,
        .opc1 = 0, .opc2 = CP_ANY, .access = PL1_W,
        .type = ARM_CP_NOP | ARM_CP_OVERRIDE },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo not_v6_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo not_v6_cp_reginfo[] = {
       */
      { .name = "WFI_v5", .cp = 15, .crn = 7, .crm = 8, .opc1 = 0, .opc2 = 2,
        .access = PL1_W, .type = ARM_CP_WFI },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo not_v7_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo not_v7_cp_reginfo[] = {
        .opc1 = 0, .opc2 = 0, .access = PL1_RW, .type = ARM_CP_NOP },
      { .name = "NMRR", .cp = 15, .crn = 10, .crm = 2,
        .opc1 = 0, .opc2 = 1, .access = PL1_RW, .type = ARM_CP_NOP },
 -    REGINFO_SENTINEL
  };
  static void cpacr_write(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v6_cp_reginfo[] = {
        .crn = 1, .crm = 0, .opc1 = 0, .opc2 = 2, .accessfn = cpacr_access,
        .access = PL1_RW, .fieldoffset = offsetof(CPUARMState, cp15.cpacr_el1),
        .resetfn = cpacr_reset, .writefn = cpacr_write, .readfn = cpacr_read },
 -    REGINFO_SENTINEL
  };
  typedef struct pm_event {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v7_cp_reginfo[] = {
      { .name = "TLBIMVAA", .cp = 15, .opc1 = 0, .crn = 8, .crm = 7, .opc2 = 3,
        .type = ARM_CP_NO_RAW, .access = PL1_W, .accessfn = access_ttlb,
        .writefn = tlbimvaa_write },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo v7mp_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v7mp_cp_reginfo[] = {
      { .name = "TLBIMVAAIS", .cp = 15, .opc1 = 0, .crn = 8, .crm = 3, .opc2 = 3,
        .type = ARM_CP_NO_RAW, .access = PL1_W, .accessfn = access_ttlb,
        .writefn = tlbimvaa_is_write },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo pmovsset_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo pmovsset_cp_reginfo[] = {
        .fieldoffset = offsetof(CPUARMState, cp15.c9_pmovsr),
        .writefn = pmovsset_write,
        .raw_writefn = raw_write },
 -    REGINFO_SENTINEL
  };
  static void teecr_write(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo t2ee_cp_reginfo[] = {
      { .name = "TEEHBR", .cp = 14, .crn = 1, .crm = 0, .opc1 = 6, .opc2 = 0,
        .access = PL0_RW, .fieldoffset = offsetof(CPUARMState, teehbr),
        .accessfn = teehbr_access, .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo v6k_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v6k_cp_reginfo[] = {
        .bank_fieldoffsets = { offsetoflow32(CPUARMState, cp15.tpidrprw_s),
                               offsetoflow32(CPUARMState, cp15.tpidrprw_ns) },
        .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  #ifndef CONFIG_USER_ONLY
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo generic_timer_cp_reginfo[] = {
        .fieldoffset = offsetof(CPUARMState, cp15.c14_timer[GTIMER_SEC].cval),
        .writefn = gt_sec_cval_write, .raw_writefn = raw_write,
      },
 -    REGINFO_SENTINEL
  };
  static CPAccessResult e2h_access(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo generic_timer_cp_reginfo[] = {
        .access = PL0_R, .type = ARM_CP_NO_RAW | ARM_CP_IO,
        .readfn = gt_virt_cnt_read,
      },
 -    REGINFO_SENTINEL
  };
  #endif
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo vapa_cp_reginfo[] = {
        .access = PL1_W, .accessfn = ats_access,
        .writefn = ats_write, .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC },
  #endif
 -    REGINFO_SENTINEL
  };
  /* Return basic MPU access permission bits.  */
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo pmsav7_cp_reginfo[] = {
        .fieldoffset = offsetof(CPUARMState, pmsav7.rnr[M_REG_NS]),
        .writefn = pmsav7_rgnr_write,
        .resetfn = arm_cp_reset_ignore },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo pmsav5_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo pmsav5_cp_reginfo[] = {
      { .name = "946_PRBS7", .cp = 15, .crn = 6, .crm = 7, .opc1 = 0,
        .opc2 = CP_ANY, .access = PL1_RW, .resetvalue = 0,
        .fieldoffset = offsetof(CPUARMState, cp15.c6_region[7]) },
 -    REGINFO_SENTINEL
  };
  static void vmsa_ttbcr_raw_write(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo vmsa_pmsa_cp_reginfo[] = {
        .access = PL1_RW, .accessfn = access_tvm_trvm,
        .fieldoffset = offsetof(CPUARMState, cp15.far_el[1]),
        .resetvalue = 0, },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo vmsa_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo vmsa_cp_reginfo[] = {
        /* No offsetoflow32 -- pass the entire TCR to writefn/raw_writefn. */
        .bank_fieldoffsets = { offsetof(CPUARMState, cp15.tcr_el[3]),
                               offsetof(CPUARMState, cp15.tcr_el[1])} },
 -    REGINFO_SENTINEL
  };
  /* Note that unlike TTBCR, writing to TTBCR2 does not require flushing
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo omap_cp_reginfo[] = {
      { .name = "C9", .cp = 15, .crn = 9,
        .crm = CP_ANY, .opc1 = CP_ANY, .opc2 = CP_ANY, .access = PL1_RW,
        .type = ARM_CP_CONST | ARM_CP_OVERRIDE, .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  static void xscale_cpar_write(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo xscale_cp_reginfo[] = {
      { .name = "XSCALE_UNLOCK_DCACHE",
        .cp = 15, .opc1 = 0, .crn = 9, .crm = 2, .opc2 = 1,
        .access = PL1_W, .type = ARM_CP_NOP },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo dummy_c15_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo dummy_c15_cp_reginfo[] = {
        .access = PL1_RW,
        .type = ARM_CP_CONST | ARM_CP_NO_RAW | ARM_CP_OVERRIDE,
        .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo cache_dirty_status_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cache_dirty_status_cp_reginfo[] = {
      { .name = "CDSR", .cp = 15, .crn = 7, .crm = 10, .opc1 = 0, .opc2 = 6,
        .access = PL1_R, .type = ARM_CP_CONST | ARM_CP_NO_RAW,
        .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo cache_block_ops_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cache_block_ops_cp_reginfo[] = {
        .access = PL0_W, .type = ARM_CP_NOP|ARM_CP_64BIT },
      { .name = "CIDCR", .cp = 15, .crm = 14, .opc1 = 0,
        .access = PL1_W, .type = ARM_CP_NOP|ARM_CP_64BIT },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo cache_test_clean_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cache_test_clean_cp_reginfo[] = {
      { .name = "TCI_DCACHE", .cp = 15, .crn = 7, .crm = 14, .opc1 = 0, .opc2 = 3,
        .access = PL0_R, .type = ARM_CP_CONST | ARM_CP_NO_RAW,
        .resetvalue = (1 << 30) },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo strongarm_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo strongarm_cp_reginfo[] = {
        .crm = CP_ANY, .opc1 = CP_ANY, .opc2 = CP_ANY,
        .access = PL1_RW, .resetvalue = 0,
        .type = ARM_CP_CONST | ARM_CP_OVERRIDE | ARM_CP_NO_RAW },
 -    REGINFO_SENTINEL
  };
  static uint64_t midr_read(CPUARMState *env, const ARMCPRegInfo *ri)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo lpae_cp_reginfo[] = {
        .bank_fieldoffsets = { offsetof(CPUARMState, cp15.ttbr1_s),
                               offsetof(CPUARMState, cp15.ttbr1_ns) },
        .writefn = vmsa_ttbr_write, },
 -    REGINFO_SENTINEL
  };
  static uint64_t aa64_fpcr_read(CPUARMState *env, const ARMCPRegInfo *ri)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v8_cp_reginfo[] = {
        .access = PL1_RW, .accessfn = access_trap_aa32s_el1,
        .writefn = sdcr_write,
        .fieldoffset = offsetoflow32(CPUARMState, cp15.mdcr_el3) },
 -    REGINFO_SENTINEL
  };
  /* Used to describe the behaviour of EL2 regs when EL2 does not exist.  */
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el3_no_el2_cp_reginfo[] = {
        .type = ARM_CP_CONST,
        .cp = 15, .opc1 = 4, .crn = 6, .crm = 0, .opc2 = 2,
        .access = PL2_RW, .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  /* Ditto, but for registers which exist in ARMv8 but not v7 */
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el3_no_el2_v8_cp_reginfo[] = {
        .cp = 15, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 4,
        .access = PL2_RW,
        .type = ARM_CP_CONST, .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  static void do_hcr_write(CPUARMState *env, uint64_t value, uint64_t valid_mask)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el2_cp_reginfo[] = {
        .cp = 15, .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 3,
        .access = PL2_RW,
        .fieldoffset = offsetof(CPUARMState, cp15.hstr_el2) },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo el2_v8_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el2_v8_cp_reginfo[] = {
        .access = PL2_RW,
        .fieldoffset = offsetofhigh32(CPUARMState, cp15.hcr_el2),
        .writefn = hcr_writehigh },
 -    REGINFO_SENTINEL
  };
  static CPAccessResult sel2_access(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el2_sec_cp_reginfo[] = {
        .opc0 = 3, .opc1 = 4, .crn = 2, .crm = 6, .opc2 = 2,
        .access = PL2_RW, .accessfn = sel2_access,
        .fieldoffset = offsetof(CPUARMState, cp15.vstcr_el2) },
 -    REGINFO_SENTINEL
  };
  static CPAccessResult nsacr_access(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el3_cp_reginfo[] = {
        .opc0 = 1, .opc1 = 6, .crn = 8, .crm = 7, .opc2 = 5,
        .access = PL3_W, .type = ARM_CP_NO_RAW,
        .writefn = tlbi_aa64_vae3_write },
 -    REGINFO_SENTINEL
  };
  #ifndef CONFIG_USER_ONLY
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo debug_cp_reginfo[] = {
        .cp = 14, .opc0 = 2, .opc1 = 0, .crn = 0, .crm = 2, .opc2 = 0,
        .access = PL1_RW, .accessfn = access_tda,
        .type = ARM_CP_NOP },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo debug_lpae_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo debug_lpae_cp_reginfo[] = {
        .access = PL0_R, .type = ARM_CP_CONST|ARM_CP_64BIT, .resetvalue = 0 },
      { .name = "DBGDSAR", .cp = 14, .crm = 2, .opc1 = 0,
        .access = PL0_R, .type = ARM_CP_CONST|ARM_CP_64BIT, .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  /* Return the exception level to which exceptions should be taken
@@ -XXX,XX +XXX,XX @@ static void define_debug_regs(ARMCPU *cpu)
                .fieldoffset = offsetof(CPUARMState, cp15.dbgbcr[i]),
                .writefn = dbgbcr_write, .raw_writefn = raw_write
              },
 -            REGINFO_SENTINEL
          };
          define_arm_cp_regs(cpu, dbgregs);
      }
@@ -XXX,XX +XXX,XX @@ static void define_debug_regs(ARMCPU *cpu)
                .fieldoffset = offsetof(CPUARMState, cp15.dbgwcr[i]),
                .writefn = dbgwcr_write, .raw_writefn = raw_write
              },
 -            REGINFO_SENTINEL
          };
          define_arm_cp_regs(cpu, dbgregs);
      }
@@ -XXX,XX +XXX,XX @@ static void define_pmu_regs(ARMCPU *cpu)
                .type = ARM_CP_IO,
                .readfn = pmevtyper_readfn, .writefn = pmevtyper_writefn,
                .raw_writefn = pmevtyper_rawwrite },
 -            REGINFO_SENTINEL
          };
          define_arm_cp_regs(cpu, pmev_regs);
          g_free(pmevcntr_name);
@@ -XXX,XX +XXX,XX @@ static void define_pmu_regs(ARMCPU *cpu)
                .cp = 15, .opc1 = 0, .crn = 9, .crm = 14, .opc2 = 5,
                .access = PL0_R, .accessfn = pmreg_access, .type = ARM_CP_CONST,
                .resetvalue = extract64(cpu->pmceid1, 32, 32) },
 -            REGINFO_SENTINEL
          };
          define_arm_cp_regs(cpu, v81_pmu_regs);
      }
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo lor_reginfo[] = {
        .opc0 = 3, .opc1 = 0, .crn = 10, .crm = 4, .opc2 = 7,
        .access = PL1_R, .accessfn = access_lor_ns,
        .type = ARM_CP_CONST, .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  #ifdef TARGET_AARCH64
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo pauth_reginfo[] = {
        .opc0 = 3, .opc1 = 0, .crn = 2, .crm = 1, .opc2 = 3,
        .access = PL1_RW, .accessfn = access_pauth,
        .fieldoffset = offsetof(CPUARMState, keys.apib.hi) },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo tlbirange_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo tlbirange_reginfo[] = {
        .opc0 = 1, .opc1 = 6, .crn = 8, .crm = 6, .opc2 = 5,
        .access = PL3_W, .type = ARM_CP_NO_RAW,
        .writefn = tlbi_aa64_rvae3_write },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo tlbios_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo tlbios_reginfo[] = {
        .opc0 = 1, .opc1 = 6, .crn = 8, .crm = 1, .opc2 = 5,
        .access = PL3_W, .type = ARM_CP_NO_RAW,
        .writefn = tlbi_aa64_vae3is_write },
 -    REGINFO_SENTINEL
  };
  static uint64_t rndr_readfn(CPUARMState *env, const ARMCPRegInfo *ri)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo rndr_reginfo[] = {
        .type = ARM_CP_NO_RAW | ARM_CP_SUPPRESS_TB_END | ARM_CP_IO,
        .opc0 = 3, .opc1 = 3, .crn = 2, .crm = 4, .opc2 = 1,
        .access = PL0_R, .readfn = rndr_readfn },
 -    REGINFO_SENTINEL
  };
  #ifndef CONFIG_USER_ONLY
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo dcpop_reg[] = {
        .opc0 = 1, .opc1 = 3, .crn = 7, .crm = 12, .opc2 = 1,
        .access = PL0_W, .type = ARM_CP_NO_RAW | ARM_CP_SUPPRESS_TB_END,
        .accessfn = aa64_cacheop_poc_access, .writefn = dccvap_writefn },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo dcpodp_reg[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo dcpodp_reg[] = {
        .opc0 = 1, .opc1 = 3, .crn = 7, .crm = 13, .opc2 = 1,
        .access = PL0_W, .type = ARM_CP_NO_RAW | ARM_CP_SUPPRESS_TB_END,
        .accessfn = aa64_cacheop_poc_access, .writefn = dccvap_writefn },
 -    REGINFO_SENTINEL
  };
  #endif /*CONFIG_USER_ONLY*/
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo mte_reginfo[] = {
      { .name = "DC_CIGDSW", .state = ARM_CP_STATE_AA64,
        .opc0 = 1, .opc1 = 0, .crn = 7, .crm = 14, .opc2 = 6,
        .type = ARM_CP_NOP, .access = PL1_W, .accessfn = access_tsw },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo mte_tco_ro_reginfo[] = {
      { .name = "TCO", .state = ARM_CP_STATE_AA64,
        .opc0 = 3, .opc1 = 3, .crn = 4, .crm = 2, .opc2 = 7,
        .type = ARM_CP_CONST, .access = PL0_RW, },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo mte_el0_cacheop_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo mte_el0_cacheop_reginfo[] = {
        .accessfn = aa64_zva_access,
  #endif
      },
 -    REGINFO_SENTINEL
  };
  #endif
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo predinv_reginfo[] = {
      { .name = "CPPRCTX", .state = ARM_CP_STATE_AA32,
        .cp = 15, .opc1 = 0, .crn = 7, .crm = 3, .opc2 = 7,
        .type = ARM_CP_NOP, .access = PL0_W, .accessfn = access_predinv },
 -    REGINFO_SENTINEL
  };
  static uint64_t ccsidr2_read(CPUARMState *env, const ARMCPRegInfo *ri)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo ccsidr2_reginfo[] = {
        .access = PL1_R,
        .accessfn = access_aa64_tid2,
        .readfn = ccsidr2_read, .type = ARM_CP_NO_RAW },
 -    REGINFO_SENTINEL
  };
  static CPAccessResult access_aa64_tid3(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo jazelle_regs[] = {
        .cp = 14, .crn = 2, .crm = 0, .opc1 = 7, .opc2 = 0,
        .accessfn = access_joscr_jmcr,
        .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo vhe_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo vhe_reginfo[] = {
        .access = PL2_RW, .accessfn = e2h_access,
        .writefn = gt_virt_cval_write, .raw_writefn = raw_write },
  #endif
 -    REGINFO_SENTINEL
  };
  #ifndef CONFIG_USER_ONLY
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo ats1e1_reginfo[] = {
        .opc0 = 1, .opc1 = 0, .crn = 7, .crm = 9, .opc2 = 1,
        .access = PL1_W, .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC,
        .writefn = ats_write64 },
 -    REGINFO_SENTINEL
  };
  static const ARMCPRegInfo ats1cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo ats1cp_reginfo[] = {
        .cp = 15, .opc1 = 0, .crn = 7, .crm = 9, .opc2 = 1,
        .access = PL1_W, .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC,
        .writefn = ats_write },
 -    REGINFO_SENTINEL
  };
  #endif
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo actlr2_hactlr2_reginfo[] = {
        .cp = 15, .opc1 = 4, .crn = 1, .crm = 0, .opc2 = 3,
        .access = PL2_RW, .type = ARM_CP_CONST,
        .resetvalue = 0 },
 -    REGINFO_SENTINEL
  };
  void register_cp_regs_for_features(ARMCPU *cpu)
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                .access = PL1_R, .type = ARM_CP_CONST,
                .accessfn = access_aa32_tid3,
                .resetvalue = cpu->isar.id_isar6 },
 -            REGINFO_SENTINEL
          };
          define_arm_cp_regs(cpu, v6_idregs);
          define_arm_cp_regs(cpu, v6_cp_reginfo);
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                .opc0 = 3, .opc1 = 3, .crn = 9, .crm = 12, .opc2 = 7,
                .access = PL0_R, .accessfn = pmreg_access, .type = ARM_CP_CONST,
                .resetvalue = cpu->pmceid1 },
 -            REGINFO_SENTINEL
          };
  #ifdef CONFIG_USER_ONLY
          ARMCPRegUserSpaceInfo v8_user_idregs[] = {
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                .exported_bits = 0x000000f0ffffffff },
              { .name = "ID_AA64ISAR*_EL1_RESERVED",
                .is_glob = true                     },
 -            REGUSERINFO_SENTINEL
          };
          modify_arm_cp_regs(v8_idregs, v8_user_idregs);
  #endif
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                .access = PL2_RW,
                .resetvalue = vmpidr_def,
                .fieldoffset = offsetof(CPUARMState, cp15.vmpidr_el2) },
 -            REGINFO_SENTINEL
          };
          define_arm_cp_regs(cpu, vpidr_regs);
          define_arm_cp_regs(cpu, el2_cp_reginfo);
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                    .access = PL2_RW, .accessfn = access_el3_aa32ns,
                    .type = ARM_CP_NO_RAW,
                    .writefn = arm_cp_write_ignore, .readfn = mpidr_read },
 -                REGINFO_SENTINEL
              };
              define_arm_cp_regs(cpu, vpidr_regs);
              define_arm_cp_regs(cpu, el3_no_el2_cp_reginfo);
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                .raw_writefn = raw_write, .writefn = sctlr_write,
                .fieldoffset = offsetof(CPUARMState, cp15.sctlr_el[3]),
                .resetvalue = cpu->reset_sctlr },
 -            REGINFO_SENTINEL
          };
          define_arm_cp_regs(cpu, el3_regs);
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
              { .name = "DUMMY",
                .cp = 15, .crn = 0, .crm = 7, .opc1 = 0, .opc2 = CP_ANY,
                .access = PL1_R, .type = ARM_CP_CONST, .resetvalue = 0 },
 -            REGINFO_SENTINEL
          };
          ARMCPRegInfo id_v8_midr_cp_reginfo[] = {
              { .name = "MIDR_EL1", .state = ARM_CP_STATE_BOTH,
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                .access = PL1_R,
                .accessfn = access_aa64_tid1,
                .type = ARM_CP_CONST, .resetvalue = cpu->revidr },
 -            REGINFO_SENTINEL
          };
          ARMCPRegInfo id_cp_reginfo[] = {
              /* These are common to v8 and pre-v8 */
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                .access = PL1_R,
                .accessfn = access_aa32_tid1,
                .type = ARM_CP_CONST, .resetvalue = 0 },
 -            REGINFO_SENTINEL
          };
          /* TLBTR is specific to VMSA */
          ARMCPRegInfo id_tlbtr_reginfo = {
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
              { .name = "MIDR_EL1",
                .exported_bits = 0x00000000ffffffff },
              { .name = "REVIDR_EL1"                },
 -            REGUSERINFO_SENTINEL
          };
          modify_arm_cp_regs(id_v8_midr_cp_reginfo, id_v8_user_midr_cp_reginfo);
  #endif
          if (arm_feature(env, ARM_FEATURE_OMAPCP) ||
              arm_feature(env, ARM_FEATURE_STRONGARM)) {
 -            ARMCPRegInfo *r;
 +            size_t i;
              /* Register the blanket "writes ignored" value first to cover the
               * whole space. Then update the specific ID registers to allow write
               * access, so that they ignore writes rather than causing them to
               * UNDEF.
               */
              define_one_arm_cp_reg(cpu, &crn0_wi_reginfo);
 -            for (r = id_pre_v8_midr_cp_reginfo;
 -                 r->type != ARM_CP_SENTINEL; r++) {
 -                r->access = PL1_RW;
 +            for (i = 0; i < ARRAY_SIZE(id_pre_v8_midr_cp_reginfo); ++i) {
 +                id_pre_v8_midr_cp_reginfo[i].access = PL1_RW;
              }
 -            for (r = id_cp_reginfo; r->type != ARM_CP_SENTINEL; r++) {
 -                r->access = PL1_RW;
 +            for (i = 0; i < ARRAY_SIZE(id_cp_reginfo); ++i) {
 +                id_cp_reginfo[i].access = PL1_RW;
              }
              id_mpuir_reginfo.access = PL1_RW;
              id_tlbtr_reginfo.access = PL1_RW;
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
              { .name = "MPIDR_EL1", .state = ARM_CP_STATE_BOTH,
                .opc0 = 3, .crn = 0, .crm = 0, .opc1 = 0, .opc2 = 5,
                .access = PL1_R, .readfn = mpidr_read, .type = ARM_CP_NO_RAW },
 -            REGINFO_SENTINEL
          };
  #ifdef CONFIG_USER_ONLY
          ARMCPRegUserSpaceInfo mpidr_user_cp_reginfo[] = {
              { .name = "MPIDR_EL1",
                .fixed_bits = 0x0000000080000000 },
 -            REGUSERINFO_SENTINEL
          };
          modify_arm_cp_regs(mpidr_cp_reginfo, mpidr_user_cp_reginfo);
  #endif
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                .opc0 = 3, .opc1 = 6, .crn = 1, .crm = 0, .opc2 = 1,
                .access = PL3_RW, .type = ARM_CP_CONST,
                .resetvalue = 0 },
 -            REGINFO_SENTINEL
          };
          define_arm_cp_regs(cpu, auxcr_reginfo);
          if (cpu_isar_feature(aa32_ac2, cpu)) {
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                    .type = ARM_CP_CONST,
                    .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 3, .opc2 = 0,
                    .access = PL1_R, .resetvalue = cpu->reset_cbar },
 -                REGINFO_SENTINEL
              };
              /* We don't implement a r/w 64 bit CBAR currently */
              assert(arm_feature(env, ARM_FEATURE_CBAR_RO));
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                .bank_fieldoffsets = { offsetof(CPUARMState, cp15.vbar_s),
                                       offsetof(CPUARMState, cp15.vbar_ns) },
                .resetvalue = 0 },
 -            REGINFO_SENTINEL
          };
          define_arm_cp_regs(cpu, vbar_cp_reginfo);
      }
@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
                     r->writefn);
          }
      }
 -    /* Bad type field probably means missing sentinel at end of reg list */
 -    assert(cptype_valid(r->type));
 +
      for (crm = crmmin; crm <= crmmax; crm++) {
          for (opc1 = opc1min; opc1 <= opc1max; opc1++) {
              for (opc2 = opc2min; opc2 <= opc2max; opc2++) {
@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
      }
  }
 -void define_arm_cp_regs_with_opaque(ARMCPU *cpu,
 -                                    const ARMCPRegInfo *regs, void *opaque)
 +/* Define a whole list of registers */
 +void define_arm_cp_regs_with_opaque_len(ARMCPU *cpu, const ARMCPRegInfo *regs,
 +                                        void *opaque, size_t len)
  {
 -    /* Define a whole list of registers */
 -    const ARMCPRegInfo *r;
 -    for (r = regs; r->type != ARM_CP_SENTINEL; r++) {
 -        define_one_arm_cp_reg_with_opaque(cpu, r, opaque);
 +    size_t i;
 +    for (i = 0; i < len; ++i) {
 +        define_one_arm_cp_reg_with_opaque(cpu, regs + i, opaque);
      }
  }
@@ -XXX,XX +XXX,XX @@ void define_arm_cp_regs_with_opaque(ARMCPU *cpu,
   * user-space cannot alter any values and dynamic values pertaining to
   * execution state are hidden from user space view anyway.
   */
 -void modify_arm_cp_regs(ARMCPRegInfo *regs, const ARMCPRegUserSpaceInfo *mods)
 +void modify_arm_cp_regs_with_len(ARMCPRegInfo *regs, size_t regs_len,
 +                                 const ARMCPRegUserSpaceInfo *mods,
 +                                 size_t mods_len)
  {
 -    const ARMCPRegUserSpaceInfo *m;
 -    ARMCPRegInfo *r;
 -
 -    for (m = mods; m->name; m++) {
 +    for (size_t mi = 0; mi < mods_len; ++mi) {
 +        const ARMCPRegUserSpaceInfo *m = mods + mi;
          GPatternSpec *pat = NULL;
 +
          if (m->is_glob) {
              pat = g_pattern_spec_new(m->name);
          }
 -        for (r = regs; r->type != ARM_CP_SENTINEL; r++) {
 +        for (size_t ri = 0; ri < regs_len; ++ri) {
 +            ARMCPRegInfo *r = regs + ri;
 +
              if (pat && g_pattern_match_string(pat, r->name)) {
                  r->type = ARM_CP_CONST;
                  r->access = PL0U_R;
 --
-.25.1
+.34.1

-[PULL 01/23] target/arm: Enable SCTLR_EL1.BT0 for aarch64-linux-user
+[PULL 12/14] target/arm: Fix 32-bit SMOPA
 From: Richard Henderson <richard.henderson@linaro.org>
-This controls whether the PACI{A,B}SP instructions trap with BTYPE=3
+While the 8-bit input elements are sequential in the input vector,
-(indirect branch from register other than x16/x17).  The linux kernel
+the 32-bit output elements are not sequential in the output matrix.
-sets this in bti_enable().
+Do not attempt to compute 2 32-bit outputs at the same time.
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/998
+Cc: qemu-stable@nongnu.org
 Fixes: 23a5e3859f5 ("target/arm: Implement SME integer outer product")
 Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2083
 Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
+Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
-Message-id: 20220427042312.294300-1-richard.henderson@linaro.org
+Message-id: 20240305163931.242795-1-richard.henderson@linaro.org
 [PMM: remove stray change to makefile comment]
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
 ---
- target/arm/cpu.c                  |  2 ++
+ target/arm/tcg/sme_helper.c       | 77 ++++++++++++++++++-------------
- tests/tcg/aarch64/bti-3.c         | 42 +++++++++++++++++++++++++++++++
+ tests/tcg/aarch64/sme-smopa-1.c   | 47 +++++++++++++++++++
- tests/tcg/aarch64/Makefile.target |  6 ++---
+ tests/tcg/aarch64/sme-smopa-2.c   | 54 ++++++++++++++++++++++
-files changed, 47 insertions(+), 3 deletions(-)
+ tests/tcg/aarch64/Makefile.target |  2 +-
- create mode 100644 tests/tcg/aarch64/bti-3.c
+files changed, 147 insertions(+), 33 deletions(-)
+ create mode 100644 tests/tcg/aarch64/sme-smopa-1.c
-diff --git a/target/arm/cpu.c b/target/arm/cpu.c
+ create mode 100644 tests/tcg/aarch64/sme-smopa-2.c
 diff --git a/target/arm/tcg/sme_helper.c b/target/arm/tcg/sme_helper.c
 index XXXXXXX..XXXXXXX 100644
---- a/target/arm/cpu.c
+--- a/target/arm/tcg/sme_helper.c
-+++ b/target/arm/cpu.c
++++ b/target/arm/tcg/sme_helper.c
-@@ -XXX,XX +XXX,XX @@ static void arm_cpu_reset(DeviceState *dev)
+@@ -XXX,XX +XXX,XX @@ void HELPER(sme_bfmopa)(void *vza, void *vzn, void *vzm, void *vpn,
-         /* Enable all PAC keys.  */
+     }
-         env->cp15.sctlr_el[1] |= (SCTLR_EnIA | SCTLR_EnIB |
+ }
-                                   SCTLR_EnDA | SCTLR_EnDB);
-+        /* Trap on btype=3 for PACIxSP. */
+-typedef uint64_t IMOPFn(uint64_t, uint64_t, uint64_t, uint8_t, bool);
-+        env->cp15.sctlr_el[1] |= SCTLR_BT0;
++typedef uint32_t IMOPFn32(uint32_t, uint32_t, uint32_t, uint8_t, bool);
-         /* and to the FP/Neon instructions */
++static inline void do_imopa_s(uint32_t *za, uint32_t *zn, uint32_t *zm,
-         env->cp15.cpacr_el1 = deposit64(env->cp15.cpacr_el1, 20, 2, 3);
++                              uint8_t *pn, uint8_t *pm,
-         /* and to the SVE instructions */
++                              uint32_t desc, IMOPFn32 *fn)
-diff --git a/tests/tcg/aarch64/bti-3.c b/tests/tcg/aarch64/bti-3.c
++{
 +    intptr_t row, col, oprsz = simd_oprsz(desc) / 4;
 +    bool neg = simd_data(desc);
 -static inline void do_imopa(uint64_t *za, uint64_t *zn, uint64_t *zm,
 -                            uint8_t *pn, uint8_t *pm,
 -                            uint32_t desc, IMOPFn *fn)
 +    for (row = 0; row < oprsz; ++row) {
 +        uint8_t pa = (pn[H1(row >> 1)] >> ((row & 1) * 4)) & 0xf;
 +        uint32_t *za_row = &za[tile_vslice_index(row)];
 +        uint32_t n = zn[H4(row)];
 +
 +        for (col = 0; col < oprsz; ++col) {
 +            uint8_t pb = pm[H1(col >> 1)] >> ((col & 1) * 4);
 +            uint32_t *a = &za_row[H4(col)];
 +
 +            *a = fn(n, zm[H4(col)], *a, pa & pb, neg);
 +        }
 +    }
 +}
 +
 +typedef uint64_t IMOPFn64(uint64_t, uint64_t, uint64_t, uint8_t, bool);
 +static inline void do_imopa_d(uint64_t *za, uint64_t *zn, uint64_t *zm,
 +                              uint8_t *pn, uint8_t *pm,
 +                              uint32_t desc, IMOPFn64 *fn)
  {
      intptr_t row, col, oprsz = simd_oprsz(desc) / 8;
      bool neg = simd_data(desc);
@@ -XXX,XX +XXX,XX @@ static inline void do_imopa(uint64_t *za, uint64_t *zn, uint64_t *zm,
  }
  #define DEF_IMOP_32(NAME, NTYPE, MTYPE) \
 -static uint64_t NAME(uint64_t n, uint64_t m, uint64_t a, uint8_t p, bool neg) \
 +static uint32_t NAME(uint32_t n, uint32_t m, uint32_t a, uint8_t p, bool neg) \
  {                                                                           \
 -    uint32_t sum0 = 0, sum1 = 0;                                            \
 +    uint32_t sum = 0;                                                       \
      /* Apply P to N as a mask, making the inactive elements 0. */           \
      n &= expand_pred_b(p);                                                  \
 -    sum0 += (NTYPE)(n >> 0) * (MTYPE)(m >> 0);                              \
 -    sum0 += (NTYPE)(n >> 8) * (MTYPE)(m >> 8);                              \
 -    sum0 += (NTYPE)(n >> 16) * (MTYPE)(m >> 16);                            \
 -    sum0 += (NTYPE)(n >> 24) * (MTYPE)(m >> 24);                            \
 -    sum1 += (NTYPE)(n >> 32) * (MTYPE)(m >> 32);                            \
 -    sum1 += (NTYPE)(n >> 40) * (MTYPE)(m >> 40);                            \
 -    sum1 += (NTYPE)(n >> 48) * (MTYPE)(m >> 48);                            \
 -    sum1 += (NTYPE)(n >> 56) * (MTYPE)(m >> 56);                            \
 -    if (neg) {                                                              \
 -        sum0 = (uint32_t)a - sum0, sum1 = (uint32_t)(a >> 32) - sum1;       \
 -    } else {                                                                \
 -        sum0 = (uint32_t)a + sum0, sum1 = (uint32_t)(a >> 32) + sum1;       \
 -    }                                                                       \
 -    return ((uint64_t)sum1 << 32) | sum0;                                   \
 +    sum += (NTYPE)(n >> 0) * (MTYPE)(m >> 0);                               \
 +    sum += (NTYPE)(n >> 8) * (MTYPE)(m >> 8);                               \
 +    sum += (NTYPE)(n >> 16) * (MTYPE)(m >> 16);                             \
 +    sum += (NTYPE)(n >> 24) * (MTYPE)(m >> 24);                             \
 +    return neg ? a - sum : a + sum;                                         \
  }
  #define DEF_IMOP_64(NAME, NTYPE, MTYPE) \
@@ -XXX,XX +XXX,XX @@ DEF_IMOP_64(umopa_d, uint16_t, uint16_t)
  DEF_IMOP_64(sumopa_d, int16_t, uint16_t)
  DEF_IMOP_64(usmopa_d, uint16_t, int16_t)
 -#define DEF_IMOPH(NAME) \
 -    void HELPER(sme_##NAME)(void *vza, void *vzn, void *vzm, void *vpn,      \
 -                            void *vpm, uint32_t desc)                        \
 -    { do_imopa(vza, vzn, vzm, vpn, vpm, desc, NAME); }
 +#define DEF_IMOPH(NAME, S) \
 +    void HELPER(sme_##NAME##_##S)(void *vza, void *vzn, void *vzm,          \
 +                                  void *vpn, void *vpm, uint32_t desc)      \
 +    { do_imopa_##S(vza, vzn, vzm, vpn, vpm, desc, NAME##_##S); }
 -DEF_IMOPH(smopa_s)
 -DEF_IMOPH(umopa_s)
 -DEF_IMOPH(sumopa_s)
 -DEF_IMOPH(usmopa_s)
 -DEF_IMOPH(smopa_d)
 -DEF_IMOPH(umopa_d)
 -DEF_IMOPH(sumopa_d)
 -DEF_IMOPH(usmopa_d)
 +DEF_IMOPH(smopa, s)
 +DEF_IMOPH(umopa, s)
 +DEF_IMOPH(sumopa, s)
 +DEF_IMOPH(usmopa, s)
 +
 +DEF_IMOPH(smopa, d)
 +DEF_IMOPH(umopa, d)
 +DEF_IMOPH(sumopa, d)
 +DEF_IMOPH(usmopa, d)
 diff --git a/tests/tcg/aarch64/sme-smopa-1.c b/tests/tcg/aarch64/sme-smopa-1.c
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
-+++ b/tests/tcg/aarch64/bti-3.c
++++ b/tests/tcg/aarch64/sme-smopa-1.c
 @@ -XXX,XX +XXX,XX @@
-+/*
++#include <stdio.h>
-+ * BTI vs PACIASP
++#include <string.h>
 + */
 +
 +#include "bti-crt.inc.c"
 +
 +static void skip2_sigill(int sig, siginfo_t *info, ucontext_t *uc)
 +{
 +    uc->uc_mcontext.pc += 8;
 +    uc->uc_mcontext.pstate = 1;
 +}
 +
 +#define BTYPE_1() \
 +    asm("mov %0,#1; adr x16, 1f; br x16; 1: hint #25; mov %0,#0" \
 +        : "=r"(skipped) : : "x16", "x30")
 +
 +#define BTYPE_2() \
 +    asm("mov %0,#1; adr x16, 1f; blr x16; 1: hint #25; mov %0,#0" \
 +        : "=r"(skipped) : : "x16", "x30")
 +
 +#define BTYPE_3() \
 +    asm("mov %0,#1; adr x15, 1f; br x15; 1: hint #25; mov %0,#0" \
 +        : "=r"(skipped) : : "x15", "x30")
 +
 +#define TEST(WHICH, EXPECT) \
 +    do { WHICH(); fail += skipped ^ EXPECT; } while (0)
 +
 +int main()
 +{
-+    int fail = 0;
++    static const int cmp[4][4] = {
-+    int skipped;
++        {  110,  134,  158,  182 },
-+
++        {  390,  478,  566,  654 },
-+    /* Signal-like with SA_SIGINFO.  */
++        {  670,  822,  974, 1126 },
-+    signal_info(SIGILL, skip2_sigill);
++        {  950, 1166, 1382, 1598 }
-+
++    };
-+    /* With SCTLR_EL1.BT0 set, PACIASP is not compatible with type=3. */
++    int dst[4][4];
-+    TEST(BTYPE_1, 0);
++    int *tmp = &dst[0][0];
-+    TEST(BTYPE_2, 0);
++
-+    TEST(BTYPE_3, 1);
++    asm volatile(
-+
++        ".arch armv8-r+sme\n\t"
-+    return fail;
++        "smstart\n\t"
 +        "index z0.b, #0, #1\n\t"
 +        "movprfx z1, z0\n\t"
 +        "add z1.b, z1.b, #16\n\t"
 +        "ptrue p0.b\n\t"
 +        "smopa za0.s, p0/m, p0/m, z0.b, z1.b\n\t"
 +        "ptrue p0.s, vl4\n\t"
 +        "mov w12, #0\n\t"
 +        "st1w { za0h.s[w12, #0] }, p0, [%0]\n\t"
 +        "add %0, %0, #16\n\t"
 +        "st1w { za0h.s[w12, #1] }, p0, [%0]\n\t"
 +        "add %0, %0, #16\n\t"
 +        "st1w { za0h.s[w12, #2] }, p0, [%0]\n\t"
 +        "add %0, %0, #16\n\t"
 +        "st1w { za0h.s[w12, #3] }, p0, [%0]\n\t"
 +        "smstop"
 +        : "+r"(tmp) : : "memory");
 +
 +    if (memcmp(cmp, dst, sizeof(dst)) == 0) {
 +        return 0;
 +    }
 +
 +    /* See above for correct results. */
 +    for (int i = 0; i < 4; ++i) {
 +        for (int j = 0; j < 4; ++j) {
 +            printf("%6d", dst[i][j]);
 +        }
 +        printf("\n");
 +    }
 +    return 1;
 +}
 diff --git a/tests/tcg/aarch64/sme-smopa-2.c b/tests/tcg/aarch64/sme-smopa-2.c
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
 +++ b/tests/tcg/aarch64/sme-smopa-2.c
@@ -XXX,XX +XXX,XX @@
 +#include <stdio.h>
 +#include <string.h>
 +
 +int main()
 +{
 +    static const long cmp[4][4] = {
 +        {  110,  134,  158,  182 },
 +        {  390,  478,  566,  654 },
 +        {  670,  822,  974, 1126 },
 +        {  950, 1166, 1382, 1598 }
 +    };
 +    long dst[4][4];
 +    long *tmp = &dst[0][0];
 +    long svl;
 +
 +    /* Validate that we have a wide enough vector for 4 elements. */
 +    asm(".arch armv8-r+sme-i64\n\trdsvl %0, #1" : "=r"(svl));
 +    if (svl < 32) {
 +        return 0;
 +    }
 +
 +    asm volatile(
 +        "smstart\n\t"
 +        "index z0.h, #0, #1\n\t"
 +        "movprfx z1, z0\n\t"
 +        "add z1.h, z1.h, #16\n\t"
 +        "ptrue p0.b\n\t"
 +        "smopa za0.d, p0/m, p0/m, z0.h, z1.h\n\t"
 +        "ptrue p0.d, vl4\n\t"
 +        "mov w12, #0\n\t"
 +        "st1d { za0h.d[w12, #0] }, p0, [%0]\n\t"
 +        "add %0, %0, #32\n\t"
 +        "st1d { za0h.d[w12, #1] }, p0, [%0]\n\t"
 +        "mov w12, #2\n\t"
 +        "add %0, %0, #32\n\t"
 +        "st1d { za0h.d[w12, #0] }, p0, [%0]\n\t"
 +        "add %0, %0, #32\n\t"
 +        "st1d { za0h.d[w12, #1] }, p0, [%0]\n\t"
 +        "smstop"
 +        : "+r"(tmp) : : "memory");
 +
 +    if (memcmp(cmp, dst, sizeof(dst)) == 0) {
 +        return 0;
 +    }
 +
 +    /* See above for correct results. */
 +    for (int i = 0; i < 4; ++i) {
 +        for (int j = 0; j < 4; ++j) {
 +            printf("%6ld", dst[i][j]);
 +        }
 +        printf("\n");
 +    }
 +    return 1;
 +}
 diff --git a/tests/tcg/aarch64/Makefile.target b/tests/tcg/aarch64/Makefile.target
 index XXXXXXX..XXXXXXX 100644
 --- a/tests/tcg/aarch64/Makefile.target
 +++ b/tests/tcg/aarch64/Makefile.target
 @@ -XXX,XX +XXX,XX @@ endif
- # BTI Tests
- # bti-1 tests the elf notes, so we require special compiler support.
+ # SME Tests
- ifneq ($(CROSS_CC_HAS_ARMV8_BTI),)
+ ifneq ($(CROSS_AS_HAS_ARMV9_SME),)
--AARCH64_TESTS += bti-1
+-AARCH64_TESTS += sme-outprod1
--bti-1: CFLAGS += -mbranch-protection=standard
++AARCH64_TESTS += sme-outprod1 sme-smopa-1 sme-smopa-2
 -bti-1: LDFLAGS += -nostdlib
 +AARCH64_TESTS += bti-1 bti-3
 +bti-1 bti-3: CFLAGS += -mbranch-protection=standard
 +bti-1 bti-3: LDFLAGS += -nostdlib
  endif
- # bti-2 tests PROT_BTI, so no special compiler support required.
- AARCH64_TESTS += bti-2
+ # System Registers Tests
 --
-.25.1
+.34.1

-[PULL 03/23] target/arm: Reorg CPAccessResult and access_check_cp_reg
+[PULL 13/14] hw/rtc/sun4v-rtc: Relicense to GPLv2-or-later
-From: Richard Henderson <richard.henderson@linaro.org>
+The sun4v RTC device model added under commit a0e893039cf2ce0 in 2016
 was unfortunately added with a license of GPL-v3-or-later, which is
 not compatible with other QEMU code which has a GPL-v2-only license.
-Rearrange the values of the enumerators of CPAccessResult
+Relicense the code in the .c and the .h file to GPL-v2-or-later,
-so that we may directly extract the target el. For the two
+to make it compatible with the rest of QEMU.
 special cases in access_check_cp_reg, use CPAccessResult.
-Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
+Cc: qemu-stable@nongnu.org
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
+Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
+Signed-off-by: Paolo Bonzini (for Red Hat) <pbonzini@redhat.com>
-Message-id: 20220501055028.646596-3-richard.henderson@linaro.org
+Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com>
 Signed-off-by: Markus Armbruster <armbru@redhat.com>
 Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
 Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
 Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
 Acked-by: Alex Bennée <alex.bennee@linaro.org>
 Message-id: 20240223161300.938542-1-peter.maydell@linaro.org
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
 ---
- target/arm/cpregs.h    | 26 ++++++++++++--------
+ include/hw/rtc/sun4v-rtc.h | 2 +-
- target/arm/op_helper.c | 56 +++++++++++++++++++++---------------------
+ hw/rtc/sun4v-rtc.c         | 2 +-
-files changed, 44 insertions(+), 38 deletions(-)
+files changed, 2 insertions(+), 2 deletions(-)
-diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
+diff --git a/include/hw/rtc/sun4v-rtc.h b/include/hw/rtc/sun4v-rtc.h
 index XXXXXXX..XXXXXXX 100644
---- a/target/arm/cpregs.h
+--- a/include/hw/rtc/sun4v-rtc.h
-+++ b/target/arm/cpregs.h
++++ b/include/hw/rtc/sun4v-rtc.h
-@@ -XXX,XX +XXX,XX @@ static inline bool cptype_valid(int cptype)
+@@ -XXX,XX +XXX,XX @@
- typedef enum CPAccessResult {
+  *
-     /* Access is permitted */
+  * Copyright (c) 2016 Artyom Tarasenko
-     CP_ACCESS_OK = 0,
+  *
-+
+- * This code is licensed under the GNU GPL v3 or (at your option) any later
-+    /*
++ * This code is licensed under the GNU GPL v2 or (at your option) any later
-+     * Combined with one of the following, the low 2 bits indicate the
+  * version.
-+     * target exception level.  If 0, the exception is taken to the usual
+  */
-+     * target EL (EL1 or PL1 if in EL0, otherwise to the current EL).
-+     */
+diff --git a/hw/rtc/sun4v-rtc.c b/hw/rtc/sun4v-rtc.c
 +    CP_ACCESS_EL_MASK = 3,
 +
      /*
       * Access fails due to a configurable trap or enable which would
       * result in a categorized exception syndrome giving information about
       * the failing instruction (ie syndrome category 0x3, 0x4, 0x5, 0x6,
 -     * 0xc or 0x18). The exception is taken to the usual target EL (EL1 or
 -     * PL1 if in EL0, otherwise to the current EL).
 +     * 0xc or 0x18).
       */
 -    CP_ACCESS_TRAP = 1,
 +    CP_ACCESS_TRAP = (1 << 2),
 +    CP_ACCESS_TRAP_EL2 = CP_ACCESS_TRAP | 2,
 +    CP_ACCESS_TRAP_EL3 = CP_ACCESS_TRAP | 3,
 +
      /*
       * Access fails and results in an exception syndrome 0x0 ("uncategorized").
       * Note that this is not a catch-all case -- the set of cases which may
       * result in this failure is specifically defined by the architecture.
       */
 -    CP_ACCESS_TRAP_UNCATEGORIZED = 2,
 -    /* As CP_ACCESS_TRAP, but for traps directly to EL2 or EL3 */
 -    CP_ACCESS_TRAP_EL2 = 3,
 -    CP_ACCESS_TRAP_EL3 = 4,
 -    /* As CP_ACCESS_UNCATEGORIZED, but for traps directly to EL2 or EL3 */
 -    CP_ACCESS_TRAP_UNCATEGORIZED_EL2 = 5,
 -    CP_ACCESS_TRAP_UNCATEGORIZED_EL3 = 6,
 +    CP_ACCESS_TRAP_UNCATEGORIZED = (2 << 2),
 +    CP_ACCESS_TRAP_UNCATEGORIZED_EL2 = CP_ACCESS_TRAP_UNCATEGORIZED | 2,
 +    CP_ACCESS_TRAP_UNCATEGORIZED_EL3 = CP_ACCESS_TRAP_UNCATEGORIZED | 3,
  } CPAccessResult;
  typedef struct ARMCPRegInfo ARMCPRegInfo;
 diff --git a/target/arm/op_helper.c b/target/arm/op_helper.c
 index XXXXXXX..XXXXXXX 100644
---- a/target/arm/op_helper.c
+--- a/hw/rtc/sun4v-rtc.c
-+++ b/target/arm/op_helper.c
++++ b/hw/rtc/sun4v-rtc.c
-@@ -XXX,XX +XXX,XX @@ void HELPER(access_check_cp_reg)(CPUARMState *env, void *rip, uint32_t syndrome,
+@@ -XXX,XX +XXX,XX @@
-                                  uint32_t isread)
+  *
- {
+  * Copyright (c) 2016 Artyom Tarasenko
-     const ARMCPRegInfo *ri = rip;
+  *
-+    CPAccessResult res = CP_ACCESS_OK;
+- * This code is licensed under the GNU GPL v3 or (at your option) any later
-     int target_el;
++ * This code is licensed under the GNU GPL v2 or (at your option) any later
+  * version.
-     if (arm_feature(env, ARM_FEATURE_XSCALE) && ri->cp < 14
+  */
          && extract32(env->cp15.c15_cpar, ri->cp, 1) == 0) {
 -        raise_exception(env, EXCP_UDEF, syndrome, exception_target_el(env));
 +        res = CP_ACCESS_TRAP;
 +        goto fail;
      }
      /*
@@ -XXX,XX +XXX,XX @@ void HELPER(access_check_cp_reg)(CPUARMState *env, void *rip, uint32_t syndrome,
          mask &= ~((1 << 4) | (1 << 14));
          if (env->cp15.hstr_el2 & mask) {
 -            target_el = 2;
 -            goto exept;
 +            res = CP_ACCESS_TRAP_EL2;
 +            goto fail;
          }
      }
 -    if (!ri->accessfn) {
 +    if (ri->accessfn) {
 +        res = ri->accessfn(env, ri, isread);
 +    }
 +    if (likely(res == CP_ACCESS_OK)) {
          return;
      }
 -    switch (ri->accessfn(env, ri, isread)) {
 -    case CP_ACCESS_OK:
 -        return;
 + fail:
 +    switch (res & ~CP_ACCESS_EL_MASK) {
      case CP_ACCESS_TRAP:
 -        target_el = exception_target_el(env);
 -        break;
 -    case CP_ACCESS_TRAP_EL2:
 -        /* Requesting a trap to EL2 when we're in EL3 is
 -         * a bug in the access function.
 -         */
 -        assert(arm_current_el(env) != 3);
 -        target_el = 2;
 -        break;
 -    case CP_ACCESS_TRAP_EL3:
 -        target_el = 3;
          break;
      case CP_ACCESS_TRAP_UNCATEGORIZED:
 -        target_el = exception_target_el(env);
 -        syndrome = syn_uncategorized();
 -        break;
 -    case CP_ACCESS_TRAP_UNCATEGORIZED_EL2:
 -        target_el = 2;
 -        syndrome = syn_uncategorized();
 -        break;
 -    case CP_ACCESS_TRAP_UNCATEGORIZED_EL3:
 -        target_el = 3;
          syndrome = syn_uncategorized();
          break;
      default:
          g_assert_not_reached();
      }
 -exept:
 +    target_el = res & CP_ACCESS_EL_MASK;
 +    switch (target_el) {
 +    case 0:
 +        target_el = exception_target_el(env);
 +        break;
 +    case 2:
 +        assert(arm_current_el(env) != 3);
 +        assert(arm_is_el2_enabled(env));
 +        break;
 +    case 3:
 +        assert(arm_feature(env, ARM_FEATURE_EL3));
 +        break;
 +    default:
 +        /* No "direct" traps to EL1 */
 +        g_assert_not_reached();
 +    }
 +
      raise_exception(env, EXCP_UDEF, syndrome, target_el);
  }
 --
-.25.1
+.34.1

-[PULL 08/23] target/arm: Change cpreg access permissions to enum
+Deleted patch
-From: Richard Henderson <richard.henderson@linaro.org>
-Create a typedef as well, and use it in ARMCPRegInfo.
-This won't be perfect for debugging, but it'll nicely
-display the most common cases.
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Message-id: 20220501055028.646596-8-richard.henderson@linaro.org
-Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
----
- target/arm/cpregs.h | 44 +++++++++++++++++++++++---------------------
- target/arm/helper.c |  2 +-
-files changed, 24 insertions(+), 22 deletions(-)
-diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/cpregs.h
-+++ b/target/arm/cpregs.h
-@@ -XXX,XX +XXX,XX @@ enum {
-  * described with these bits, then use a laxer set of restrictions, and
-  * do the more restrictive/complex check inside a helper function.
-  */
--#define PL3_R 0x80
--#define PL3_W 0x40
--#define PL2_R (0x20 | PL3_R)
--#define PL2_W (0x10 | PL3_W)
--#define PL1_R (0x08 | PL2_R)
--#define PL1_W (0x04 | PL2_W)
--#define PL0_R (0x02 | PL1_R)
--#define PL0_W (0x01 | PL1_W)
-+typedef enum {
-+    PL3_R = 0x80,
-+    PL3_W = 0x40,
-+    PL2_R = 0x20 | PL3_R,
-+    PL2_W = 0x10 | PL3_W,
-+    PL1_R = 0x08 | PL2_R,
-+    PL1_W = 0x04 | PL2_W,
-+    PL0_R = 0x02 | PL1_R,
-+    PL0_W = 0x01 | PL1_W,
--/*
-- * For user-mode some registers are accessible to EL0 via a kernel
-- * trap-and-emulate ABI. In this case we define the read permissions
-- * as actually being PL0_R. However some bits of any given register
-- * may still be masked.
-- */
-+    /*
-+     * For user-mode some registers are accessible to EL0 via a kernel
-+     * trap-and-emulate ABI. In this case we define the read permissions
-+     * as actually being PL0_R. However some bits of any given register
-+     * may still be masked.
-+     */
- #ifdef CONFIG_USER_ONLY
--#define PL0U_R PL0_R
-+    PL0U_R = PL0_R,
- #else
--#define PL0U_R PL1_R
-+    PL0U_R = PL1_R,
- #endif
--#define PL3_RW (PL3_R | PL3_W)
--#define PL2_RW (PL2_R | PL2_W)
--#define PL1_RW (PL1_R | PL1_W)
--#define PL0_RW (PL0_R | PL0_W)
-+    PL3_RW = PL3_R | PL3_W,
-+    PL2_RW = PL2_R | PL2_W,
-+    PL1_RW = PL1_R | PL1_W,
-+    PL0_RW = PL0_R | PL0_W,
-+} CPAccessRights;
- typedef enum CPAccessResult {
-     /* Access is permitted */
-@@ -XXX,XX +XXX,XX @@ struct ARMCPRegInfo {
-     /* Register type: ARM_CP_* bits/values */
-     int type;
-     /* Access rights: PL*_[RW] */
--    int access;
-+    CPAccessRights access;
-     /* Security state: ARM_CP_SECSTATE_* bits/values */
-     int secure;
-     /*
-diff --git a/target/arm/helper.c b/target/arm/helper.c
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/helper.c
-+++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
-      * to encompass the generic architectural permission check.
-      */
-     if (r->state != ARM_CP_STATE_AA32) {
--        int mask = 0;
-+        CPAccessRights mask;
-         switch (r->opc1) {
-         case 0:
-             /* min_EL EL1, but some accessible to EL0 via kernel ABI */
---
-.25.1

-[PULL 10/23] target/arm: Name CPSecureState type
+Deleted patch
-From: Richard Henderson <richard.henderson@linaro.org>
-Give this enum a name and use in ARMCPRegInfo and add_cpreg_to_hashtable.
-Add the enumerator ARM_CP_SECSTATE_BOTH to clarify how 0
-is handled in define_one_arm_cp_reg_with_opaque.
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Message-id: 20220501055028.646596-10-richard.henderson@linaro.org
-Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
----
- target/arm/cpregs.h | 7 ++++---
- target/arm/helper.c | 7 +++++--
-files changed, 9 insertions(+), 5 deletions(-)
-diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/cpregs.h
-+++ b/target/arm/cpregs.h
-@@ -XXX,XX +XXX,XX @@ typedef enum {
-  * registered entry will only have one to identify whether the entry is secure
-  * or non-secure.
-  */
--enum {
-+typedef enum {
-+    ARM_CP_SECSTATE_BOTH = 0,       /* define one cpreg for each secstate */
-     ARM_CP_SECSTATE_S =   (1 << 0), /* bit[0]: Secure state register */
-     ARM_CP_SECSTATE_NS =  (1 << 1), /* bit[1]: Non-secure state register */
--};
-+} CPSecureState;
- /*
-  * Access rights:
-@@ -XXX,XX +XXX,XX @@ struct ARMCPRegInfo {
-     /* Access rights: PL*_[RW] */
-     CPAccessRights access;
-     /* Security state: ARM_CP_SECSTATE_* bits/values */
--    int secure;
-+    CPSecureState secure;
-     /*
-      * The opaque pointer passed to define_arm_cp_regs_with_opaque() when
-      * this register was defined: can be used to hand data through to the
-diff --git a/target/arm/helper.c b/target/arm/helper.c
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/helper.c
-+++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ CpuDefinitionInfoList *qmp_query_cpu_definitions(Error **errp)
- }
- static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
--                                   void *opaque, CPState state, int secstate,
-+                                   void *opaque, CPState state,
-+                                   CPSecureState secstate,
-                                    int crm, int opc1, int opc2,
-                                    const char *name)
- {
-@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
-                                                    r->secure, crm, opc1, opc2,
-                                                    r->name);
-                             break;
--                        default:
-+                        case ARM_CP_SECSTATE_BOTH:
-                             name = g_strdup_printf("%s_S", r->name);
-                             add_cpreg_to_hashtable(cpu, r, opaque, state,
-                                                    ARM_CP_SECSTATE_S,
-@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
-                                                    ARM_CP_SECSTATE_NS,
-                                                    crm, opc1, opc2, r->name);
-                             break;
-+                        default:
-+                            g_assert_not_reached();
-                         }
-                     } else {
-                         /* AArch64 registers get mapped to non-secure instance
---
-.25.1

-[PULL 11/23] target/arm: Drop always-true test in define_arm_vh_e2h_redirects_aliases
+Deleted patch
-From: Richard Henderson <richard.henderson@linaro.org>
-The new_key field is always non-zero -- drop the if.
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Message-id: 20220501055028.646596-11-richard.henderson@linaro.org
-[PMM: reinstated dropped PL3_RW mask]
-Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
----
- target/arm/helper.c | 23 +++++++++++------------
-file changed, 11 insertions(+), 12 deletions(-)
-diff --git a/target/arm/helper.c b/target/arm/helper.c
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/helper.c
-+++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ static void define_arm_vh_e2h_redirects_aliases(ARMCPU *cpu)
-     for (i = 0; i < ARRAY_SIZE(aliases); i++) {
-         const struct E2HAlias *a = &aliases[i];
--        ARMCPRegInfo *src_reg, *dst_reg;
-+        ARMCPRegInfo *src_reg, *dst_reg, *new_reg;
-+        uint32_t *new_key;
-+        bool ok;
-         if (a->feature && !a->feature(&cpu->isar)) {
-             continue;
-@@ -XXX,XX +XXX,XX @@ static void define_arm_vh_e2h_redirects_aliases(ARMCPU *cpu)
-         g_assert(src_reg->opaque == NULL);
-         /* Create alias before redirection so we dup the right data. */
--        if (a->new_key) {
--            ARMCPRegInfo *new_reg = g_memdup(src_reg, sizeof(ARMCPRegInfo));
--            uint32_t *new_key = g_memdup(&a->new_key, sizeof(uint32_t));
--            bool ok;
-+        new_reg = g_memdup(src_reg, sizeof(ARMCPRegInfo));
-+        new_key = g_memdup(&a->new_key, sizeof(uint32_t));
--            new_reg->name = a->new_name;
--            new_reg->type |= ARM_CP_ALIAS;
--            /* Remove PL1/PL0 access, leaving PL2/PL3 R/W in place.  */
--            new_reg->access &= PL2_RW | PL3_RW;
-+        new_reg->name = a->new_name;
-+        new_reg->type |= ARM_CP_ALIAS;
-+        /* Remove PL1/PL0 access, leaving PL2/PL3 R/W in place.  */
-+        new_reg->access &= PL2_RW | PL3_RW;
--            ok = g_hash_table_insert(cpu->cp_regs, new_key, new_reg);
--            g_assert(ok);
--        }
-+        ok = g_hash_table_insert(cpu->cp_regs, new_key, new_reg);
-+        g_assert(ok);
-         src_reg->opaque = dst_reg;
-         src_reg->orig_readfn = src_reg->readfn ?: raw_read;
---
-.25.1

-[PULL 12/23] target/arm: Store cpregs key in the hash table directly
+Deleted patch
-From: Richard Henderson <richard.henderson@linaro.org>
-Cast the uint32_t key into a gpointer directly, which
-allows us to avoid allocating storage for each key.
-Use g_hash_table_lookup when we already have a gpointer
-(e.g. for callbacks like count_cpreg), or when using
-get_arm_cp_reginfo would require casting away const.
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Message-id: 20220501055028.646596-12-richard.henderson@linaro.org
-Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
----
- target/arm/cpu.c     |  4 ++--
- target/arm/gdbstub.c |  2 +-
- target/arm/helper.c  | 41 ++++++++++++++++++-----------------------
-files changed, 21 insertions(+), 26 deletions(-)
-diff --git a/target/arm/cpu.c b/target/arm/cpu.c
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/cpu.c
-+++ b/target/arm/cpu.c
-@@ -XXX,XX +XXX,XX @@ static void arm_cpu_initfn(Object *obj)
-     ARMCPU *cpu = ARM_CPU(obj);
-     cpu_set_cpustate_pointers(cpu);
--    cpu->cp_regs = g_hash_table_new_full(g_int_hash, g_int_equal,
--                                         g_free, cpreg_hashtable_data_destroy);
-+    cpu->cp_regs = g_hash_table_new_full(g_direct_hash, g_direct_equal,
-+                                         NULL, cpreg_hashtable_data_destroy);
-     QLIST_INIT(&cpu->pre_el_change_hooks);
-     QLIST_INIT(&cpu->el_change_hooks);
-diff --git a/target/arm/gdbstub.c b/target/arm/gdbstub.c
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/gdbstub.c
-+++ b/target/arm/gdbstub.c
-@@ -XXX,XX +XXX,XX @@ static void arm_gen_one_xml_sysreg_tag(GString *s, DynamicGDBXMLInfo *dyn_xml,
- static void arm_register_sysreg_for_xml(gpointer key, gpointer value,
-                                         gpointer p)
- {
--    uint32_t ri_key = *(uint32_t *)key;
-+    uint32_t ri_key = (uintptr_t)key;
-     ARMCPRegInfo *ri = value;
-     RegisterSysregXmlParam *param = (RegisterSysregXmlParam *)p;
-     GString *s = param->s;
-diff --git a/target/arm/helper.c b/target/arm/helper.c
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/helper.c
-+++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ bool write_list_to_cpustate(ARMCPU *cpu)
- static void add_cpreg_to_list(gpointer key, gpointer opaque)
- {
-     ARMCPU *cpu = opaque;
--    uint64_t regidx;
--    const ARMCPRegInfo *ri;
--
--    regidx = *(uint32_t *)key;
--    ri = get_arm_cp_reginfo(cpu->cp_regs, regidx);
-+    uint32_t regidx = (uintptr_t)key;
-+    const ARMCPRegInfo *ri = get_arm_cp_reginfo(cpu->cp_regs, regidx);
-     if (!(ri->type & (ARM_CP_NO_RAW|ARM_CP_ALIAS))) {
-         cpu->cpreg_indexes[cpu->cpreg_array_len] = cpreg_to_kvm_id(regidx);
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_list(gpointer key, gpointer opaque)
- static void count_cpreg(gpointer key, gpointer opaque)
- {
-     ARMCPU *cpu = opaque;
--    uint64_t regidx;
-     const ARMCPRegInfo *ri;
--    regidx = *(uint32_t *)key;
--    ri = get_arm_cp_reginfo(cpu->cp_regs, regidx);
-+    ri = g_hash_table_lookup(cpu->cp_regs, key);
-     if (!(ri->type & (ARM_CP_NO_RAW|ARM_CP_ALIAS))) {
-         cpu->cpreg_array_len++;
-@@ -XXX,XX +XXX,XX @@ static void count_cpreg(gpointer key, gpointer opaque)
- static gint cpreg_key_compare(gconstpointer a, gconstpointer b)
- {
--    uint64_t aidx = cpreg_to_kvm_id(*(uint32_t *)a);
--    uint64_t bidx = cpreg_to_kvm_id(*(uint32_t *)b);
-+    uint64_t aidx = cpreg_to_kvm_id((uintptr_t)a);
-+    uint64_t bidx = cpreg_to_kvm_id((uintptr_t)b);
-     if (aidx > bidx) {
-         return 1;
-@@ -XXX,XX +XXX,XX @@ static void define_arm_vh_e2h_redirects_aliases(ARMCPU *cpu)
-     for (i = 0; i < ARRAY_SIZE(aliases); i++) {
-         const struct E2HAlias *a = &aliases[i];
-         ARMCPRegInfo *src_reg, *dst_reg, *new_reg;
--        uint32_t *new_key;
-         bool ok;
-         if (a->feature && !a->feature(&cpu->isar)) {
-             continue;
-         }
--        src_reg = g_hash_table_lookup(cpu->cp_regs, &a->src_key);
--        dst_reg = g_hash_table_lookup(cpu->cp_regs, &a->dst_key);
-+        src_reg = g_hash_table_lookup(cpu->cp_regs,
-+                                      (gpointer)(uintptr_t)a->src_key);
-+        dst_reg = g_hash_table_lookup(cpu->cp_regs,
-+                                      (gpointer)(uintptr_t)a->dst_key);
-         g_assert(src_reg != NULL);
-         g_assert(dst_reg != NULL);
-@@ -XXX,XX +XXX,XX @@ static void define_arm_vh_e2h_redirects_aliases(ARMCPU *cpu)
-         /* Create alias before redirection so we dup the right data. */
-         new_reg = g_memdup(src_reg, sizeof(ARMCPRegInfo));
--        new_key = g_memdup(&a->new_key, sizeof(uint32_t));
-         new_reg->name = a->new_name;
-         new_reg->type |= ARM_CP_ALIAS;
-         /* Remove PL1/PL0 access, leaving PL2/PL3 R/W in place.  */
-         new_reg->access &= PL2_RW | PL3_RW;
--        ok = g_hash_table_insert(cpu->cp_regs, new_key, new_reg);
-+        ok = g_hash_table_insert(cpu->cp_regs,
-+                                 (gpointer)(uintptr_t)a->new_key, new_reg);
-         g_assert(ok);
-         src_reg->opaque = dst_reg;
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-     /* Private utility function for define_one_arm_cp_reg_with_opaque():
-      * add a single reginfo struct to the hash table.
-      */
--    uint32_t *key = g_new(uint32_t, 1);
-+    uint32_t key;
-     ARMCPRegInfo *r2 = g_memdup(r, sizeof(ARMCPRegInfo));
-     int is64 = (r->type & ARM_CP_64BIT) ? 1 : 0;
-     int ns = (secstate & ARM_CP_SECSTATE_NS) ? 1 : 0;
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-         if (r->cp == 0 || r->state == ARM_CP_STATE_BOTH) {
-             r2->cp = CP_REG_ARM64_SYSREG_CP;
-         }
--        *key = ENCODE_AA64_CP_REG(r2->cp, r2->crn, crm,
--                                  r2->opc0, opc1, opc2);
-+        key = ENCODE_AA64_CP_REG(r2->cp, r2->crn, crm,
-+                                 r2->opc0, opc1, opc2);
-     } else {
--        *key = ENCODE_CP_REG(r2->cp, is64, ns, r2->crn, crm, opc1, opc2);
-+        key = ENCODE_CP_REG(r2->cp, is64, ns, r2->crn, crm, opc1, opc2);
-     }
-     if (opaque) {
-         r2->opaque = opaque;
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-      * requested.
-      */
-     if (!(r->type & ARM_CP_OVERRIDE)) {
--        ARMCPRegInfo *oldreg;
--        oldreg = g_hash_table_lookup(cpu->cp_regs, key);
-+        const ARMCPRegInfo *oldreg = get_arm_cp_reginfo(cpu->cp_regs, key);
-         if (oldreg && !(oldreg->type & ARM_CP_OVERRIDE)) {
-             fprintf(stderr, "Register redefined: cp=%d %d bit "
-                     "crn=%d crm=%d opc1=%d opc2=%d, "
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-             g_assert_not_reached();
-         }
-     }
--    g_hash_table_insert(cpu->cp_regs, key, r2);
-+    g_hash_table_insert(cpu->cp_regs, (gpointer)(uintptr_t)key, r2);
- }
-@@ -XXX,XX +XXX,XX @@ void modify_arm_cp_regs_with_len(ARMCPRegInfo *regs, size_t regs_len,
- const ARMCPRegInfo *get_arm_cp_reginfo(GHashTable *cpregs, uint32_t encoded_cp)
- {
--    return g_hash_table_lookup(cpregs, &encoded_cp);
-+    return g_hash_table_lookup(cpregs, (gpointer)(uintptr_t)encoded_cp);
- }
- void arm_cp_write_ignore(CPUARMState *env, const ARMCPRegInfo *ri,
---
-.25.1

-[PULL 13/23] target/arm: Merge allocation of the cpreg and its name
+[PULL 14/14] target/arm: Move v7m-related code from cpu32.c into a separate file
-From: Richard Henderson <richard.henderson@linaro.org>
+From: Thomas Huth <thuth@redhat.com>
-Simplify freeing cp_regs hash table entries by using a single
+Move the code to a separate file so that we do not have to compile
-allocation for the entire value.
+it anymore if CONFIG_ARM_V7M is not set.
-This fixes a theoretical bug if we were to ever free the entire
+Signed-off-by: Thomas Huth <thuth@redhat.com>
-hash table, because we've been installing string literal constants
+Message-id: 20240308141051.536599-2-thuth@redhat.com
 into the cpreg structure in define_arm_vh_e2h_redirects_aliases.
 However, at present we only free entries created for AArch32
 wildcard cpregs which get overwritten by more specific cpregs,
 so this bug is never exposed.
 Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
 Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Message-id: 20220501055028.646596-13-richard.henderson@linaro.org
 Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
 ---
- target/arm/cpu.c    | 16 +---------------
+ target/arm/tcg/cpu-v7m.c   | 290 +++++++++++++++++++++++++++++++++++++
- target/arm/helper.c | 10 ++++++++--
+ target/arm/tcg/cpu32.c     | 261 ---------------------------------
-files changed, 9 insertions(+), 17 deletions(-)
+ target/arm/meson.build     |   3 +
  target/arm/tcg/meson.build |   3 +
 files changed, 296 insertions(+), 261 deletions(-)
  create mode 100644 target/arm/tcg/cpu-v7m.c
-diff --git a/target/arm/cpu.c b/target/arm/cpu.c
+diff --git a/target/arm/tcg/cpu-v7m.c b/target/arm/tcg/cpu-v7m.c
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
 +++ b/target/arm/tcg/cpu-v7m.c
@@ -XXX,XX +XXX,XX @@
 +/*
 + * QEMU ARMv7-M TCG-only CPUs.
 + *
 + * Copyright (c) 2012 SUSE LINUX Products GmbH
 + *
 + * This code is licensed under the GNU GPL v2 or later.
 + *
 + * SPDX-License-Identifier: GPL-2.0-or-later
 + */
 +
 +#include "qemu/osdep.h"
 +#include "cpu.h"
 +#include "hw/core/tcg-cpu-ops.h"
 +#include "internals.h"
 +
 +#if !defined(CONFIG_USER_ONLY)
 +
 +#include "hw/intc/armv7m_nvic.h"
 +
 +static bool arm_v7m_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
 +{
 +    CPUClass *cc = CPU_GET_CLASS(cs);
 +    ARMCPU *cpu = ARM_CPU(cs);
 +    CPUARMState *env = &cpu->env;
 +    bool ret = false;
 +
 +    /*
 +     * ARMv7-M interrupt masking works differently than -A or -R.
 +     * There is no FIQ/IRQ distinction. Instead of I and F bits
 +     * masking FIQ and IRQ interrupts, an exception is taken only
 +     * if it is higher priority than the current execution priority
 +     * (which depends on state like BASEPRI, FAULTMASK and the
 +     * currently active exception).
 +     */
 +    if (interrupt_request & CPU_INTERRUPT_HARD
 +        && (armv7m_nvic_can_take_pending_exception(env->nvic))) {
 +        cs->exception_index = EXCP_IRQ;
 +        cc->tcg_ops->do_interrupt(cs);
 +        ret = true;
 +    }
 +    return ret;
 +}
 +
 +#endif /* !CONFIG_USER_ONLY */
 +
 +static void cortex_m0_initfn(Object *obj)
 +{
 +    ARMCPU *cpu = ARM_CPU(obj);
 +    set_feature(&cpu->env, ARM_FEATURE_V6);
 +    set_feature(&cpu->env, ARM_FEATURE_M);
 +
 +    cpu->midr = 0x410cc200;
 +
 +    /*
 +     * These ID register values are not guest visible, because
 +     * we do not implement the Main Extension. They must be set
 +     * to values corresponding to the Cortex-M0's implemented
 +     * features, because QEMU generally controls its emulation
 +     * by looking at ID register fields. We use the same values as
 +     * for the M3.
 +     */
 +    cpu->isar.id_pfr0 = 0x00000030;
 +    cpu->isar.id_pfr1 = 0x00000200;
 +    cpu->isar.id_dfr0 = 0x00100000;
 +    cpu->id_afr0 = 0x00000000;
 +    cpu->isar.id_mmfr0 = 0x00000030;
 +    cpu->isar.id_mmfr1 = 0x00000000;
 +    cpu->isar.id_mmfr2 = 0x00000000;
 +    cpu->isar.id_mmfr3 = 0x00000000;
 +    cpu->isar.id_isar0 = 0x01141110;
 +    cpu->isar.id_isar1 = 0x02111000;
 +    cpu->isar.id_isar2 = 0x21112231;
 +    cpu->isar.id_isar3 = 0x01111110;
 +    cpu->isar.id_isar4 = 0x01310102;
 +    cpu->isar.id_isar5 = 0x00000000;
 +    cpu->isar.id_isar6 = 0x00000000;
 +}
 +
 +static void cortex_m3_initfn(Object *obj)
 +{
 +    ARMCPU *cpu = ARM_CPU(obj);
 +    set_feature(&cpu->env, ARM_FEATURE_V7);
 +    set_feature(&cpu->env, ARM_FEATURE_M);
 +    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
 +    cpu->midr = 0x410fc231;
 +    cpu->pmsav7_dregion = 8;
 +    cpu->isar.id_pfr0 = 0x00000030;
 +    cpu->isar.id_pfr1 = 0x00000200;
 +    cpu->isar.id_dfr0 = 0x00100000;
 +    cpu->id_afr0 = 0x00000000;
 +    cpu->isar.id_mmfr0 = 0x00000030;
 +    cpu->isar.id_mmfr1 = 0x00000000;
 +    cpu->isar.id_mmfr2 = 0x00000000;
 +    cpu->isar.id_mmfr3 = 0x00000000;
 +    cpu->isar.id_isar0 = 0x01141110;
 +    cpu->isar.id_isar1 = 0x02111000;
 +    cpu->isar.id_isar2 = 0x21112231;
 +    cpu->isar.id_isar3 = 0x01111110;
 +    cpu->isar.id_isar4 = 0x01310102;
 +    cpu->isar.id_isar5 = 0x00000000;
 +    cpu->isar.id_isar6 = 0x00000000;
 +}
 +
 +static void cortex_m4_initfn(Object *obj)
 +{
 +    ARMCPU *cpu = ARM_CPU(obj);
 +
 +    set_feature(&cpu->env, ARM_FEATURE_V7);
 +    set_feature(&cpu->env, ARM_FEATURE_M);
 +    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
 +    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
 +    cpu->midr = 0x410fc240; /* r0p0 */
 +    cpu->pmsav7_dregion = 8;
 +    cpu->isar.mvfr0 = 0x10110021;
 +    cpu->isar.mvfr1 = 0x11000011;
 +    cpu->isar.mvfr2 = 0x00000000;
 +    cpu->isar.id_pfr0 = 0x00000030;
 +    cpu->isar.id_pfr1 = 0x00000200;
 +    cpu->isar.id_dfr0 = 0x00100000;
 +    cpu->id_afr0 = 0x00000000;
 +    cpu->isar.id_mmfr0 = 0x00000030;
 +    cpu->isar.id_mmfr1 = 0x00000000;
 +    cpu->isar.id_mmfr2 = 0x00000000;
 +    cpu->isar.id_mmfr3 = 0x00000000;
 +    cpu->isar.id_isar0 = 0x01141110;
 +    cpu->isar.id_isar1 = 0x02111000;
 +    cpu->isar.id_isar2 = 0x21112231;
 +    cpu->isar.id_isar3 = 0x01111110;
 +    cpu->isar.id_isar4 = 0x01310102;
 +    cpu->isar.id_isar5 = 0x00000000;
 +    cpu->isar.id_isar6 = 0x00000000;
 +}
 +
 +static void cortex_m7_initfn(Object *obj)
 +{
 +    ARMCPU *cpu = ARM_CPU(obj);
 +
 +    set_feature(&cpu->env, ARM_FEATURE_V7);
 +    set_feature(&cpu->env, ARM_FEATURE_M);
 +    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
 +    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
 +    cpu->midr = 0x411fc272; /* r1p2 */
 +    cpu->pmsav7_dregion = 8;
 +    cpu->isar.mvfr0 = 0x10110221;
 +    cpu->isar.mvfr1 = 0x12000011;
 +    cpu->isar.mvfr2 = 0x00000040;
 +    cpu->isar.id_pfr0 = 0x00000030;
 +    cpu->isar.id_pfr1 = 0x00000200;
 +    cpu->isar.id_dfr0 = 0x00100000;
 +    cpu->id_afr0 = 0x00000000;
 +    cpu->isar.id_mmfr0 = 0x00100030;
 +    cpu->isar.id_mmfr1 = 0x00000000;
 +    cpu->isar.id_mmfr2 = 0x01000000;
 +    cpu->isar.id_mmfr3 = 0x00000000;
 +    cpu->isar.id_isar0 = 0x01101110;
 +    cpu->isar.id_isar1 = 0x02112000;
 +    cpu->isar.id_isar2 = 0x20232231;
 +    cpu->isar.id_isar3 = 0x01111131;
 +    cpu->isar.id_isar4 = 0x01310132;
 +    cpu->isar.id_isar5 = 0x00000000;
 +    cpu->isar.id_isar6 = 0x00000000;
 +}
 +
 +static void cortex_m33_initfn(Object *obj)
 +{
 +    ARMCPU *cpu = ARM_CPU(obj);
 +
 +    set_feature(&cpu->env, ARM_FEATURE_V8);
 +    set_feature(&cpu->env, ARM_FEATURE_M);
 +    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
 +    set_feature(&cpu->env, ARM_FEATURE_M_SECURITY);
 +    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
 +    cpu->midr = 0x410fd213; /* r0p3 */
 +    cpu->pmsav7_dregion = 16;
 +    cpu->sau_sregion = 8;
 +    cpu->isar.mvfr0 = 0x10110021;
 +    cpu->isar.mvfr1 = 0x11000011;
 +    cpu->isar.mvfr2 = 0x00000040;
 +    cpu->isar.id_pfr0 = 0x00000030;
 +    cpu->isar.id_pfr1 = 0x00000210;
 +    cpu->isar.id_dfr0 = 0x00200000;
 +    cpu->id_afr0 = 0x00000000;
 +    cpu->isar.id_mmfr0 = 0x00101F40;
 +    cpu->isar.id_mmfr1 = 0x00000000;
 +    cpu->isar.id_mmfr2 = 0x01000000;
 +    cpu->isar.id_mmfr3 = 0x00000000;
 +    cpu->isar.id_isar0 = 0x01101110;
 +    cpu->isar.id_isar1 = 0x02212000;
 +    cpu->isar.id_isar2 = 0x20232232;
 +    cpu->isar.id_isar3 = 0x01111131;
 +    cpu->isar.id_isar4 = 0x01310132;
 +    cpu->isar.id_isar5 = 0x00000000;
 +    cpu->isar.id_isar6 = 0x00000000;
 +    cpu->clidr = 0x00000000;
 +    cpu->ctr = 0x8000c000;
 +}
 +
 +static void cortex_m55_initfn(Object *obj)
 +{
 +    ARMCPU *cpu = ARM_CPU(obj);
 +
 +    set_feature(&cpu->env, ARM_FEATURE_V8);
 +    set_feature(&cpu->env, ARM_FEATURE_V8_1M);
 +    set_feature(&cpu->env, ARM_FEATURE_M);
 +    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
 +    set_feature(&cpu->env, ARM_FEATURE_M_SECURITY);
 +    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
 +    cpu->midr = 0x410fd221; /* r0p1 */
 +    cpu->revidr = 0;
 +    cpu->pmsav7_dregion = 16;
 +    cpu->sau_sregion = 8;
 +    /* These are the MVFR* values for the FPU + full MVE configuration */
 +    cpu->isar.mvfr0 = 0x10110221;
 +    cpu->isar.mvfr1 = 0x12100211;
 +    cpu->isar.mvfr2 = 0x00000040;
 +    cpu->isar.id_pfr0 = 0x20000030;
 +    cpu->isar.id_pfr1 = 0x00000230;
 +    cpu->isar.id_dfr0 = 0x10200000;
 +    cpu->id_afr0 = 0x00000000;
 +    cpu->isar.id_mmfr0 = 0x00111040;
 +    cpu->isar.id_mmfr1 = 0x00000000;
 +    cpu->isar.id_mmfr2 = 0x01000000;
 +    cpu->isar.id_mmfr3 = 0x00000011;
 +    cpu->isar.id_isar0 = 0x01103110;
 +    cpu->isar.id_isar1 = 0x02212000;
 +    cpu->isar.id_isar2 = 0x20232232;
 +    cpu->isar.id_isar3 = 0x01111131;
 +    cpu->isar.id_isar4 = 0x01310132;
 +    cpu->isar.id_isar5 = 0x00000000;
 +    cpu->isar.id_isar6 = 0x00000000;
 +    cpu->clidr = 0x00000000; /* caches not implemented */
 +    cpu->ctr = 0x8303c003;
 +}
 +
 +static const TCGCPUOps arm_v7m_tcg_ops = {
 +    .initialize = arm_translate_init,
 +    .synchronize_from_tb = arm_cpu_synchronize_from_tb,
 +    .debug_excp_handler = arm_debug_excp_handler,
 +    .restore_state_to_opc = arm_restore_state_to_opc,
 +
 +#ifdef CONFIG_USER_ONLY
 +    .record_sigsegv = arm_cpu_record_sigsegv,
 +    .record_sigbus = arm_cpu_record_sigbus,
 +#else
 +    .tlb_fill = arm_cpu_tlb_fill,
 +    .cpu_exec_interrupt = arm_v7m_cpu_exec_interrupt,
 +    .do_interrupt = arm_v7m_cpu_do_interrupt,
 +    .do_transaction_failed = arm_cpu_do_transaction_failed,
 +    .do_unaligned_access = arm_cpu_do_unaligned_access,
 +    .adjust_watchpoint_address = arm_adjust_watchpoint_address,
 +    .debug_check_watchpoint = arm_debug_check_watchpoint,
 +    .debug_check_breakpoint = arm_debug_check_breakpoint,
 +#endif /* !CONFIG_USER_ONLY */
 +};
 +
 +static void arm_v7m_class_init(ObjectClass *oc, void *data)
 +{
 +    ARMCPUClass *acc = ARM_CPU_CLASS(oc);
 +    CPUClass *cc = CPU_CLASS(oc);
 +
 +    acc->info = data;
 +    cc->tcg_ops = &arm_v7m_tcg_ops;
 +    cc->gdb_core_xml_file = "arm-m-profile.xml";
 +}
 +
 +static const ARMCPUInfo arm_v7m_cpus[] = {
 +    { .name = "cortex-m0",   .initfn = cortex_m0_initfn,
 +                             .class_init = arm_v7m_class_init },
 +    { .name = "cortex-m3",   .initfn = cortex_m3_initfn,
 +                             .class_init = arm_v7m_class_init },
 +    { .name = "cortex-m4",   .initfn = cortex_m4_initfn,
 +                             .class_init = arm_v7m_class_init },
 +    { .name = "cortex-m7",   .initfn = cortex_m7_initfn,
 +                             .class_init = arm_v7m_class_init },
 +    { .name = "cortex-m33",  .initfn = cortex_m33_initfn,
 +                             .class_init = arm_v7m_class_init },
 +    { .name = "cortex-m55",  .initfn = cortex_m55_initfn,
 +                             .class_init = arm_v7m_class_init },
 +};
 +
 +static void arm_v7m_cpu_register_types(void)
 +{
 +    size_t i;
 +
 +    for (i = 0; i < ARRAY_SIZE(arm_v7m_cpus); ++i) {
 +        arm_cpu_register(&arm_v7m_cpus[i]);
 +    }
 +}
 +
 +type_init(arm_v7m_cpu_register_types)
 diff --git a/target/arm/tcg/cpu32.c b/target/arm/tcg/cpu32.c
 index XXXXXXX..XXXXXXX 100644
---- a/target/arm/cpu.c
+--- a/target/arm/tcg/cpu32.c
-+++ b/target/arm/cpu.c
++++ b/target/arm/tcg/cpu32.c
-@@ -XXX,XX +XXX,XX @@ uint64_t arm_cpu_mp_affinity(int idx, uint8_t clustersz)
+@@ -XXX,XX +XXX,XX @@
-     return (Aff1 << ARM_AFF1_SHIFT) | Aff0;
+ #include "hw/boards.h"
- }
+ #endif
+ #include "cpregs.h"
--static void cpreg_hashtable_data_destroy(gpointer data)
+-#if !defined(CONFIG_USER_ONLY) && defined(CONFIG_TCG)
--{
+-#include "hw/intc/armv7m_nvic.h"
 -#endif
  /* Share AArch32 -cpu max features with AArch64. */
@@ -XXX,XX +XXX,XX @@ void aa32_max_features(ARMCPU *cpu)
  /* CPU models. These are not needed for the AArch64 linux-user build. */
  #if !defined(CONFIG_USER_ONLY) || !defined(TARGET_AARCH64)
 -#if !defined(CONFIG_USER_ONLY)
 -static bool arm_v7m_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
 -{
 -    CPUClass *cc = CPU_GET_CLASS(cs);
 -    ARMCPU *cpu = ARM_CPU(cs);
 -    CPUARMState *env = &cpu->env;
 -    bool ret = false;
 -
 -    /*
--     * Destroy function for cpu->cp_regs hashtable data entries.
+-     * ARMv7-M interrupt masking works differently than -A or -R.
--     * We must free the name string because it was g_strdup()ed in
+-     * There is no FIQ/IRQ distinction. Instead of I and F bits
--     * add_cpreg_to_hashtable(). It's OK to cast away the 'const'
+-     * masking FIQ and IRQ interrupts, an exception is taken only
--     * from r->name because we know we definitely allocated it.
+-     * if it is higher priority than the current execution priority
 -     * (which depends on state like BASEPRI, FAULTMASK and the
 -     * currently active exception).
 -     */
--    ARMCPRegInfo *r = data;
+-    if (interrupt_request & CPU_INTERRUPT_HARD
--
+-        && (armv7m_nvic_can_take_pending_exception(env->nvic))) {
--    g_free((void *)r->name);
+-        cs->exception_index = EXCP_IRQ;
--    g_free(r);
+-        cc->tcg_ops->do_interrupt(cs);
--}
+-        ret = true;
--
+-    }
- static void arm_cpu_initfn(Object *obj)
+-    return ret;
 -}
 -#endif /* !CONFIG_USER_ONLY */
 -
  static void arm926_initfn(Object *obj)
  {
      ARMCPU *cpu = ARM_CPU(obj);
+@@ -XXX,XX +XXX,XX @@ static void cortex_a15_initfn(Object *obj)
-     cpu_set_cpustate_pointers(cpu);
+     define_arm_cp_regs(cpu, cortexa15_cp_reginfo);
-     cpu->cp_regs = g_hash_table_new_full(g_direct_hash, g_direct_equal,
+ }
--                                         NULL, cpreg_hashtable_data_destroy);
-+                                         NULL, g_free);
+-static void cortex_m0_initfn(Object *obj)
+-{
-     QLIST_INIT(&cpu->pre_el_change_hooks);
+-    ARMCPU *cpu = ARM_CPU(obj);
-     QLIST_INIT(&cpu->el_change_hooks);
+-    set_feature(&cpu->env, ARM_FEATURE_V6);
-diff --git a/target/arm/helper.c b/target/arm/helper.c
+-    set_feature(&cpu->env, ARM_FEATURE_M);
 -
 -    cpu->midr = 0x410cc200;
 -
 -    /*
 -     * These ID register values are not guest visible, because
 -     * we do not implement the Main Extension. They must be set
 -     * to values corresponding to the Cortex-M0's implemented
 -     * features, because QEMU generally controls its emulation
 -     * by looking at ID register fields. We use the same values as
 -     * for the M3.
 -     */
 -    cpu->isar.id_pfr0 = 0x00000030;
 -    cpu->isar.id_pfr1 = 0x00000200;
 -    cpu->isar.id_dfr0 = 0x00100000;
 -    cpu->id_afr0 = 0x00000000;
 -    cpu->isar.id_mmfr0 = 0x00000030;
 -    cpu->isar.id_mmfr1 = 0x00000000;
 -    cpu->isar.id_mmfr2 = 0x00000000;
 -    cpu->isar.id_mmfr3 = 0x00000000;
 -    cpu->isar.id_isar0 = 0x01141110;
 -    cpu->isar.id_isar1 = 0x02111000;
 -    cpu->isar.id_isar2 = 0x21112231;
 -    cpu->isar.id_isar3 = 0x01111110;
 -    cpu->isar.id_isar4 = 0x01310102;
 -    cpu->isar.id_isar5 = 0x00000000;
 -    cpu->isar.id_isar6 = 0x00000000;
 -}
 -
 -static void cortex_m3_initfn(Object *obj)
 -{
 -    ARMCPU *cpu = ARM_CPU(obj);
 -    set_feature(&cpu->env, ARM_FEATURE_V7);
 -    set_feature(&cpu->env, ARM_FEATURE_M);
 -    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
 -    cpu->midr = 0x410fc231;
 -    cpu->pmsav7_dregion = 8;
 -    cpu->isar.id_pfr0 = 0x00000030;
 -    cpu->isar.id_pfr1 = 0x00000200;
 -    cpu->isar.id_dfr0 = 0x00100000;
 -    cpu->id_afr0 = 0x00000000;
 -    cpu->isar.id_mmfr0 = 0x00000030;
 -    cpu->isar.id_mmfr1 = 0x00000000;
 -    cpu->isar.id_mmfr2 = 0x00000000;
 -    cpu->isar.id_mmfr3 = 0x00000000;
 -    cpu->isar.id_isar0 = 0x01141110;
 -    cpu->isar.id_isar1 = 0x02111000;
 -    cpu->isar.id_isar2 = 0x21112231;
 -    cpu->isar.id_isar3 = 0x01111110;
 -    cpu->isar.id_isar4 = 0x01310102;
 -    cpu->isar.id_isar5 = 0x00000000;
 -    cpu->isar.id_isar6 = 0x00000000;
 -}
 -
 -static void cortex_m4_initfn(Object *obj)
 -{
 -    ARMCPU *cpu = ARM_CPU(obj);
 -
 -    set_feature(&cpu->env, ARM_FEATURE_V7);
 -    set_feature(&cpu->env, ARM_FEATURE_M);
 -    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
 -    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
 -    cpu->midr = 0x410fc240; /* r0p0 */
 -    cpu->pmsav7_dregion = 8;
 -    cpu->isar.mvfr0 = 0x10110021;
 -    cpu->isar.mvfr1 = 0x11000011;
 -    cpu->isar.mvfr2 = 0x00000000;
 -    cpu->isar.id_pfr0 = 0x00000030;
 -    cpu->isar.id_pfr1 = 0x00000200;
 -    cpu->isar.id_dfr0 = 0x00100000;
 -    cpu->id_afr0 = 0x00000000;
 -    cpu->isar.id_mmfr0 = 0x00000030;
 -    cpu->isar.id_mmfr1 = 0x00000000;
 -    cpu->isar.id_mmfr2 = 0x00000000;
 -    cpu->isar.id_mmfr3 = 0x00000000;
 -    cpu->isar.id_isar0 = 0x01141110;
 -    cpu->isar.id_isar1 = 0x02111000;
 -    cpu->isar.id_isar2 = 0x21112231;
 -    cpu->isar.id_isar3 = 0x01111110;
 -    cpu->isar.id_isar4 = 0x01310102;
 -    cpu->isar.id_isar5 = 0x00000000;
 -    cpu->isar.id_isar6 = 0x00000000;
 -}
 -
 -static void cortex_m7_initfn(Object *obj)
 -{
 -    ARMCPU *cpu = ARM_CPU(obj);
 -
 -    set_feature(&cpu->env, ARM_FEATURE_V7);
 -    set_feature(&cpu->env, ARM_FEATURE_M);
 -    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
 -    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
 -    cpu->midr = 0x411fc272; /* r1p2 */
 -    cpu->pmsav7_dregion = 8;
 -    cpu->isar.mvfr0 = 0x10110221;
 -    cpu->isar.mvfr1 = 0x12000011;
 -    cpu->isar.mvfr2 = 0x00000040;
 -    cpu->isar.id_pfr0 = 0x00000030;
 -    cpu->isar.id_pfr1 = 0x00000200;
 -    cpu->isar.id_dfr0 = 0x00100000;
 -    cpu->id_afr0 = 0x00000000;
 -    cpu->isar.id_mmfr0 = 0x00100030;
 -    cpu->isar.id_mmfr1 = 0x00000000;
 -    cpu->isar.id_mmfr2 = 0x01000000;
 -    cpu->isar.id_mmfr3 = 0x00000000;
 -    cpu->isar.id_isar0 = 0x01101110;
 -    cpu->isar.id_isar1 = 0x02112000;
 -    cpu->isar.id_isar2 = 0x20232231;
 -    cpu->isar.id_isar3 = 0x01111131;
 -    cpu->isar.id_isar4 = 0x01310132;
 -    cpu->isar.id_isar5 = 0x00000000;
 -    cpu->isar.id_isar6 = 0x00000000;
 -}
 -
 -static void cortex_m33_initfn(Object *obj)
 -{
 -    ARMCPU *cpu = ARM_CPU(obj);
 -
 -    set_feature(&cpu->env, ARM_FEATURE_V8);
 -    set_feature(&cpu->env, ARM_FEATURE_M);
 -    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
 -    set_feature(&cpu->env, ARM_FEATURE_M_SECURITY);
 -    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
 -    cpu->midr = 0x410fd213; /* r0p3 */
 -    cpu->pmsav7_dregion = 16;
 -    cpu->sau_sregion = 8;
 -    cpu->isar.mvfr0 = 0x10110021;
 -    cpu->isar.mvfr1 = 0x11000011;
 -    cpu->isar.mvfr2 = 0x00000040;
 -    cpu->isar.id_pfr0 = 0x00000030;
 -    cpu->isar.id_pfr1 = 0x00000210;
 -    cpu->isar.id_dfr0 = 0x00200000;
 -    cpu->id_afr0 = 0x00000000;
 -    cpu->isar.id_mmfr0 = 0x00101F40;
 -    cpu->isar.id_mmfr1 = 0x00000000;
 -    cpu->isar.id_mmfr2 = 0x01000000;
 -    cpu->isar.id_mmfr3 = 0x00000000;
 -    cpu->isar.id_isar0 = 0x01101110;
 -    cpu->isar.id_isar1 = 0x02212000;
 -    cpu->isar.id_isar2 = 0x20232232;
 -    cpu->isar.id_isar3 = 0x01111131;
 -    cpu->isar.id_isar4 = 0x01310132;
 -    cpu->isar.id_isar5 = 0x00000000;
 -    cpu->isar.id_isar6 = 0x00000000;
 -    cpu->clidr = 0x00000000;
 -    cpu->ctr = 0x8000c000;
 -}
 -
 -static void cortex_m55_initfn(Object *obj)
 -{
 -    ARMCPU *cpu = ARM_CPU(obj);
 -
 -    set_feature(&cpu->env, ARM_FEATURE_V8);
 -    set_feature(&cpu->env, ARM_FEATURE_V8_1M);
 -    set_feature(&cpu->env, ARM_FEATURE_M);
 -    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
 -    set_feature(&cpu->env, ARM_FEATURE_M_SECURITY);
 -    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
 -    cpu->midr = 0x410fd221; /* r0p1 */
 -    cpu->revidr = 0;
 -    cpu->pmsav7_dregion = 16;
 -    cpu->sau_sregion = 8;
 -    /* These are the MVFR* values for the FPU + full MVE configuration */
 -    cpu->isar.mvfr0 = 0x10110221;
 -    cpu->isar.mvfr1 = 0x12100211;
 -    cpu->isar.mvfr2 = 0x00000040;
 -    cpu->isar.id_pfr0 = 0x20000030;
 -    cpu->isar.id_pfr1 = 0x00000230;
 -    cpu->isar.id_dfr0 = 0x10200000;
 -    cpu->id_afr0 = 0x00000000;
 -    cpu->isar.id_mmfr0 = 0x00111040;
 -    cpu->isar.id_mmfr1 = 0x00000000;
 -    cpu->isar.id_mmfr2 = 0x01000000;
 -    cpu->isar.id_mmfr3 = 0x00000011;
 -    cpu->isar.id_isar0 = 0x01103110;
 -    cpu->isar.id_isar1 = 0x02212000;
 -    cpu->isar.id_isar2 = 0x20232232;
 -    cpu->isar.id_isar3 = 0x01111131;
 -    cpu->isar.id_isar4 = 0x01310132;
 -    cpu->isar.id_isar5 = 0x00000000;
 -    cpu->isar.id_isar6 = 0x00000000;
 -    cpu->clidr = 0x00000000; /* caches not implemented */
 -    cpu->ctr = 0x8303c003;
 -}
 -
  static const ARMCPRegInfo cortexr5_cp_reginfo[] = {
      /* Dummy the TCM region regs for the moment */
      { .name = "ATCM", .cp = 15, .opc1 = 0, .crn = 9, .crm = 1, .opc2 = 0,
@@ -XXX,XX +XXX,XX @@ static void pxa270c5_initfn(Object *obj)
      cpu->reset_sctlr = 0x00000078;
  }
 -static const TCGCPUOps arm_v7m_tcg_ops = {
 -    .initialize = arm_translate_init,
 -    .synchronize_from_tb = arm_cpu_synchronize_from_tb,
 -    .debug_excp_handler = arm_debug_excp_handler,
 -    .restore_state_to_opc = arm_restore_state_to_opc,
 -
 -#ifdef CONFIG_USER_ONLY
 -    .record_sigsegv = arm_cpu_record_sigsegv,
 -    .record_sigbus = arm_cpu_record_sigbus,
 -#else
 -    .tlb_fill = arm_cpu_tlb_fill,
 -    .cpu_exec_interrupt = arm_v7m_cpu_exec_interrupt,
 -    .do_interrupt = arm_v7m_cpu_do_interrupt,
 -    .do_transaction_failed = arm_cpu_do_transaction_failed,
 -    .do_unaligned_access = arm_cpu_do_unaligned_access,
 -    .adjust_watchpoint_address = arm_adjust_watchpoint_address,
 -    .debug_check_watchpoint = arm_debug_check_watchpoint,
 -    .debug_check_breakpoint = arm_debug_check_breakpoint,
 -#endif /* !CONFIG_USER_ONLY */
 -};
 -
 -static void arm_v7m_class_init(ObjectClass *oc, void *data)
 -{
 -    ARMCPUClass *acc = ARM_CPU_CLASS(oc);
 -    CPUClass *cc = CPU_CLASS(oc);
 -
 -    acc->info = data;
 -    cc->tcg_ops = &arm_v7m_tcg_ops;
 -    cc->gdb_core_xml_file = "arm-m-profile.xml";
 -}
 -
  #ifndef TARGET_AARCH64
  /*
   * -cpu max: a CPU with as many features enabled as our emulation supports.
@@ -XXX,XX +XXX,XX @@ static const ARMCPUInfo arm_tcg_cpus[] = {
      { .name = "cortex-a8",   .initfn = cortex_a8_initfn },
      { .name = "cortex-a9",   .initfn = cortex_a9_initfn },
      { .name = "cortex-a15",  .initfn = cortex_a15_initfn },
 -    { .name = "cortex-m0",   .initfn = cortex_m0_initfn,
 -                             .class_init = arm_v7m_class_init },
 -    { .name = "cortex-m3",   .initfn = cortex_m3_initfn,
 -                             .class_init = arm_v7m_class_init },
 -    { .name = "cortex-m4",   .initfn = cortex_m4_initfn,
 -                             .class_init = arm_v7m_class_init },
 -    { .name = "cortex-m7",   .initfn = cortex_m7_initfn,
 -                             .class_init = arm_v7m_class_init },
 -    { .name = "cortex-m33",  .initfn = cortex_m33_initfn,
 -                             .class_init = arm_v7m_class_init },
 -    { .name = "cortex-m55",  .initfn = cortex_m55_initfn,
 -                             .class_init = arm_v7m_class_init },
      { .name = "cortex-r5",   .initfn = cortex_r5_initfn },
      { .name = "cortex-r5f",  .initfn = cortex_r5f_initfn },
      { .name = "cortex-r52",  .initfn = cortex_r52_initfn },
 diff --git a/target/arm/meson.build b/target/arm/meson.build
 index XXXXXXX..XXXXXXX 100644
---- a/target/arm/helper.c
+--- a/target/arm/meson.build
-+++ b/target/arm/helper.c
++++ b/target/arm/meson.build
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
+@@ -XXX,XX +XXX,XX @@ arm_system_ss.add(files(
-      * add a single reginfo struct to the hash table.
+   'ptw.c',
-      */
+ ))
-     uint32_t key;
--    ARMCPRegInfo *r2 = g_memdup(r, sizeof(ARMCPRegInfo));
++arm_user_ss = ss.source_set()
-+    ARMCPRegInfo *r2;
++
-     int is64 = (r->type & ARM_CP_64BIT) ? 1 : 0;
+ subdir('hvf')
-     int ns = (secstate & ARM_CP_SECSTATE_NS) ? 1 : 0;
-+    size_t name_len;
+ if 'CONFIG_TCG' in config_all_accel
-+
+@@ -XXX,XX +XXX,XX @@ endif
-+    /* Combine cpreg and name into one allocation. */
-+    name_len = strlen(name) + 1;
+ target_arch += {'arm': arm_ss}
-+    r2 = g_malloc(sizeof(*r2) + name_len);
+ target_system_arch += {'arm': arm_system_ss}
-+    *r2 = *r;
++target_user_arch += {'arm': arm_user_ss}
-+    r2->name = memcpy(r2 + 1, name, name_len);
+diff --git a/target/arm/tcg/meson.build b/target/arm/tcg/meson.build
+index XXXXXXX..XXXXXXX 100644
--    r2->name = g_strdup(name);
+--- a/target/arm/tcg/meson.build
-     /* Reset the secure state to the specific incoming state.  This is
++++ b/target/arm/tcg/meson.build
-      * necessary as the register may have been defined with both states.
+@@ -XXX,XX +XXX,XX @@ arm_ss.add(when: 'TARGET_AARCH64', if_true: files(
-      */
+ arm_system_ss.add(files(
    'psci.c',
  ))
 +
 +arm_system_ss.add(when: 'CONFIG_ARM_V7M', if_true: files('cpu-v7m.c'))
 +arm_user_ss.add(when: 'TARGET_AARCH64', if_false: files('cpu-v7m.c'))
 --
-.25.1
+.34.1

-[PULL 14/23] target/arm: Hoist computation of key in add_cpreg_to_hashtable
+Deleted patch
-From: Richard Henderson <richard.henderson@linaro.org>
-Move the computation of key to the top of the function.
-Hoist the resolution of cp as well, as an input to the
-computation of key.
-This will be required by a subsequent patch.
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Message-id: 20220501055028.646596-14-richard.henderson@linaro.org
-Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
----
- target/arm/helper.c | 49 +++++++++++++++++++++++++--------------------
-file changed, 27 insertions(+), 22 deletions(-)
-diff --git a/target/arm/helper.c b/target/arm/helper.c
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/helper.c
-+++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-     ARMCPRegInfo *r2;
-     int is64 = (r->type & ARM_CP_64BIT) ? 1 : 0;
-     int ns = (secstate & ARM_CP_SECSTATE_NS) ? 1 : 0;
-+    int cp = r->cp;
-     size_t name_len;
-+    switch (state) {
-+    case ARM_CP_STATE_AA32:
-+        /* We assume it is a cp15 register if the .cp field is left unset. */
-+        if (cp == 0 && r->state == ARM_CP_STATE_BOTH) {
-+            cp = 15;
-+        }
-+        key = ENCODE_CP_REG(cp, is64, ns, r->crn, crm, opc1, opc2);
-+        break;
-+    case ARM_CP_STATE_AA64:
-+        /*
-+         * To allow abbreviation of ARMCPRegInfo definitions, we treat
-+         * cp == 0 as equivalent to the value for "standard guest-visible
-+         * sysreg".  STATE_BOTH definitions are also always "standard sysreg"
-+         * in their AArch64 view (the .cp value may be non-zero for the
-+         * benefit of the AArch32 view).
-+         */
-+        if (cp == 0 || r->state == ARM_CP_STATE_BOTH) {
-+            cp = CP_REG_ARM64_SYSREG_CP;
-+        }
-+        key = ENCODE_AA64_CP_REG(cp, r->crn, crm, r->opc0, opc1, opc2);
-+        break;
-+    default:
-+        g_assert_not_reached();
-+    }
-+
-     /* Combine cpreg and name into one allocation. */
-     name_len = strlen(name) + 1;
-     r2 = g_malloc(sizeof(*r2) + name_len);
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-         }
-         if (r->state == ARM_CP_STATE_BOTH) {
--            /* We assume it is a cp15 register if the .cp field is left unset.
--             */
--            if (r2->cp == 0) {
--                r2->cp = 15;
--            }
--
- #if HOST_BIG_ENDIAN
-             if (r2->fieldoffset) {
-                 r2->fieldoffset += sizeof(uint32_t);
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
- #endif
-         }
-     }
--    if (state == ARM_CP_STATE_AA64) {
--        /* To allow abbreviation of ARMCPRegInfo
--         * definitions, we treat cp == 0 as equivalent to
--         * the value for "standard guest-visible sysreg".
--         * STATE_BOTH definitions are also always "standard
--         * sysreg" in their AArch64 view (the .cp value may
--         * be non-zero for the benefit of the AArch32 view).
--         */
--        if (r->cp == 0 || r->state == ARM_CP_STATE_BOTH) {
--            r2->cp = CP_REG_ARM64_SYSREG_CP;
--        }
--        key = ENCODE_AA64_CP_REG(r2->cp, r2->crn, crm,
--                                 r2->opc0, opc1, opc2);
--    } else {
--        key = ENCODE_CP_REG(r2->cp, is64, ns, r2->crn, crm, opc1, opc2);
--    }
-     if (opaque) {
-         r2->opaque = opaque;
-     }
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-     /* Make sure reginfo passed to helpers for wildcarded regs
-      * has the correct crm/opc1/opc2 for this reg, not CP_ANY:
-      */
-+    r2->cp = cp;
-     r2->crm = crm;
-     r2->opc1 = opc1;
-     r2->opc2 = opc2;
---
-.25.1

-[PULL 15/23] target/arm: Consolidate cpreg updates in add_cpreg_to_hashtable
+Deleted patch
-From: Richard Henderson <richard.henderson@linaro.org>
-Put most of the value writeback to the same place,
-and improve the comment that goes with them.
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Message-id: 20220501055028.646596-15-richard.henderson@linaro.org
-Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
----
- target/arm/helper.c | 28 ++++++++++++----------------
-file changed, 12 insertions(+), 16 deletions(-)
-diff --git a/target/arm/helper.c b/target/arm/helper.c
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/helper.c
-+++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-     *r2 = *r;
-     r2->name = memcpy(r2 + 1, name, name_len);
--    /* Reset the secure state to the specific incoming state.  This is
--     * necessary as the register may have been defined with both states.
-+    /*
-+     * Update fields to match the instantiation, overwiting wildcards
-+     * such as CP_ANY, ARM_CP_STATE_BOTH, or ARM_CP_SECSTATE_BOTH.
-      */
-+    r2->cp = cp;
-+    r2->crm = crm;
-+    r2->opc1 = opc1;
-+    r2->opc2 = opc2;
-+    r2->state = state;
-     r2->secure = secstate;
-+    if (opaque) {
-+        r2->opaque = opaque;
-+    }
-     if (r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1]) {
-         /* Register is banked (using both entries in array).
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
- #endif
-         }
-     }
--    if (opaque) {
--        r2->opaque = opaque;
--    }
--    /* reginfo passed to helpers is correct for the actual access,
--     * and is never ARM_CP_STATE_BOTH:
--     */
--    r2->state = state;
--    /* Make sure reginfo passed to helpers for wildcarded regs
--     * has the correct crm/opc1/opc2 for this reg, not CP_ANY:
--     */
--    r2->cp = cp;
--    r2->crm = crm;
--    r2->opc1 = opc1;
--    r2->opc2 = opc2;
-+
-     /* By convention, for wildcarded registers only the first
-      * entry is used for migration; the others are marked as
-      * ALIAS so we don't try to transfer the register
---
-.25.1

-[PULL 16/23] target/arm: Use bool for is64 and ns in add_cpreg_to_hashtable
+Deleted patch
-From: Richard Henderson <richard.henderson@linaro.org>
-Bool is a more appropriate type for these variables.
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Message-id: 20220501055028.646596-16-richard.henderson@linaro.org
-Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
----
- target/arm/helper.c | 4 ++--
-file changed, 2 insertions(+), 2 deletions(-)
-diff --git a/target/arm/helper.c b/target/arm/helper.c
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/helper.c
-+++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-      */
-     uint32_t key;
-     ARMCPRegInfo *r2;
--    int is64 = (r->type & ARM_CP_64BIT) ? 1 : 0;
--    int ns = (secstate & ARM_CP_SECSTATE_NS) ? 1 : 0;
-+    bool is64 = r->type & ARM_CP_64BIT;
-+    bool ns = secstate & ARM_CP_SECSTATE_NS;
-     int cp = r->cp;
-     size_t name_len;
---
-.25.1

-[PULL 17/23] target/arm: Hoist isbanked computation in add_cpreg_to_hashtable
+Deleted patch
-From: Richard Henderson <richard.henderson@linaro.org>
-Computing isbanked only once makes the code
-a bit easier to read.
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Message-id: 20220501055028.646596-17-richard.henderson@linaro.org
-Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
----
- target/arm/helper.c | 6 ++++--
-file changed, 4 insertions(+), 2 deletions(-)
-diff --git a/target/arm/helper.c b/target/arm/helper.c
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/helper.c
-+++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-     bool is64 = r->type & ARM_CP_64BIT;
-     bool ns = secstate & ARM_CP_SECSTATE_NS;
-     int cp = r->cp;
-+    bool isbanked;
-     size_t name_len;
-     switch (state) {
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-         r2->opaque = opaque;
-     }
--    if (r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1]) {
-+    isbanked = r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1];
-+    if (isbanked) {
-         /* Register is banked (using both entries in array).
-          * Overwriting fieldoffset as the array is only used to define
-          * banked registers but later only fieldoffset is used.
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-     }
-     if (state == ARM_CP_STATE_AA32) {
--        if (r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1]) {
-+        if (isbanked) {
-             /* If the register is banked then we don't need to migrate or
-              * reset the 32-bit instance in certain cases:
-              *
---
-.25.1

-[PULL 20/23] target/arm: Remove HOST_BIG_ENDIAN ifdef in add_cpreg_to_hashtable
+Deleted patch
-From: Richard Henderson <richard.henderson@linaro.org>
-Since e03b56863d2bc, our host endian indicator is unconditionally
-set, which means that we can use a normal C condition.
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
-Message-id: 20220501055028.646596-20-richard.henderson@linaro.org
-[PMM: quote correct git hash in commit message]
-Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
----
- target/arm/helper.c | 9 +++------
-file changed, 3 insertions(+), 6 deletions(-)
-diff --git a/target/arm/helper.c b/target/arm/helper.c
-index XXXXXXX..XXXXXXX 100644
---- a/target/arm/helper.c
-+++ b/target/arm/helper.c
-@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-             r2->type |= ARM_CP_ALIAS;
-         }
--        if (r->state == ARM_CP_STATE_BOTH) {
--#if HOST_BIG_ENDIAN
--            if (r2->fieldoffset) {
--                r2->fieldoffset += sizeof(uint32_t);
--            }
--#endif
-+        if (HOST_BIG_ENDIAN &&
-+            r->state == ARM_CP_STATE_BOTH && r2->fieldoffset) {
-+            r2->fieldoffset += sizeof(uint32_t);
-         }
-     }
---
-.25.1

Two small bugfixes, plus most of RTH's refactoring of cpregs
handling.

-- PMM

The following changes since commit 1fba9dc71a170b3a05b9d3272dd8ecfe7f26e215:

Merge tag 'pull-request-2022-05-04' of https://gitlab.com/thuth/qemu into staging (2022-05-04 08:07:02 -0700)

are available in the Git repository at:

https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20220505

for you to fetch changes up to 99a50d1a67c602126fc2b3a4812d3000eba9bf34:

target/arm: read access to performance counters from EL0 (2022-05-05 09:36:22 +0100)

----------------------------------------------------------------
target-arm queue:
 * Enable read access to performance counters from EL0
 * Enable SCTLR_EL1.BT0 for aarch64-linux-user
 * Refactoring of cpreg handling

----------------------------------------------------------------
Alex Zuepke (1):
      target/arm: read access to performance counters from EL0

Richard Henderson (22):
      target/arm: Enable SCTLR_EL1.BT0 for aarch64-linux-user
      target/arm: Split out cpregs.h
      target/arm: Reorg CPAccessResult and access_check_cp_reg
      target/arm: Replace sentinels with ARRAY_SIZE in cpregs.h
      target/arm: Make some more cpreg data static const
      target/arm: Reorg ARMCPRegInfo type field bits
      target/arm: Avoid bare abort() or assert(0)
      target/arm: Change cpreg access permissions to enum
      target/arm: Name CPState type
      target/arm: Name CPSecureState type
      target/arm: Drop always-true test in define_arm_vh_e2h_redirects_aliases
      target/arm: Store cpregs key in the hash table directly
      target/arm: Merge allocation of the cpreg and its name
      target/arm: Hoist computation of key in add_cpreg_to_hashtable
      target/arm: Consolidate cpreg updates in add_cpreg_to_hashtable
      target/arm: Use bool for is64 and ns in add_cpreg_to_hashtable
      target/arm: Hoist isbanked computation in add_cpreg_to_hashtable
      target/arm: Perform override check early in add_cpreg_to_hashtable
      target/arm: Reformat comments in add_cpreg_to_hashtable
      target/arm: Remove HOST_BIG_ENDIAN ifdef in add_cpreg_to_hashtable
      target/arm: Add isar predicates for FEAT_Debugv8p2
      target/arm: Add isar_feature_{aa64,any}_ras

From: Richard Henderson <richard.henderson@linaro.org>

This controls whether the PACI{A,B}SP instructions trap with BTYPE=3
(indirect branch from register other than x16/x17).  The linux kernel
sets this in bti_enable().

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/998
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20220427042312.294300-1-richard.henderson@linaro.org
[PMM: remove stray change to makefile comment]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/cpu.c                  |  2 ++
 tests/tcg/aarch64/bti-3.c         | 42 +++++++++++++++++++++++++++++++
 tests/tcg/aarch64/Makefile.target |  6 ++---
 3 files changed, 47 insertions(+), 3 deletions(-)
 create mode 100644 tests/tcg/aarch64/bti-3.c

diff --git a/target/arm/cpu.c b/target/arm/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu.c
+++ b/target/arm/cpu.c
@@ -XXX,XX +XXX,XX @@ static void arm_cpu_reset(DeviceState *dev)
         /* Enable all PAC keys.  */
         env->cp15.sctlr_el[1] |= (SCTLR_EnIA | SCTLR_EnIB |
                                   SCTLR_EnDA | SCTLR_EnDB);
+        /* Trap on btype=3 for PACIxSP. */
+        env->cp15.sctlr_el[1] |= SCTLR_BT0;
         /* and to the FP/Neon instructions */
         env->cp15.cpacr_el1 = deposit64(env->cp15.cpacr_el1, 20, 2, 3);
         /* and to the SVE instructions */
diff --git a/tests/tcg/aarch64/bti-3.c b/tests/tcg/aarch64/bti-3.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/tests/tcg/aarch64/bti-3.c
@@ -XXX,XX +XXX,XX @@
+/*
+ * BTI vs PACIASP
+ */
+
+#include "bti-crt.inc.c"
+
+static void skip2_sigill(int sig, siginfo_t *info, ucontext_t *uc)
+{
+    uc->uc_mcontext.pc += 8;
+    uc->uc_mcontext.pstate = 1;
+}
+
+#define BTYPE_1() \
+    asm("mov %0,#1; adr x16, 1f; br x16; 1: hint #25; mov %0,#0" \
+        : "=r"(skipped) : : "x16", "x30")
+
+#define BTYPE_2() \
+    asm("mov %0,#1; adr x16, 1f; blr x16; 1: hint #25; mov %0,#0" \
+        : "=r"(skipped) : : "x16", "x30")
+
+#define BTYPE_3() \
+    asm("mov %0,#1; adr x15, 1f; br x15; 1: hint #25; mov %0,#0" \
+        : "=r"(skipped) : : "x15", "x30")
+
+#define TEST(WHICH, EXPECT) \
+    do { WHICH(); fail += skipped ^ EXPECT; } while (0)
+
+int main()
+{
+    int fail = 0;
+    int skipped;
+
+    /* Signal-like with SA_SIGINFO.  */
+    signal_info(SIGILL, skip2_sigill);
+
+    /* With SCTLR_EL1.BT0 set, PACIASP is not compatible with type=3. */
+    TEST(BTYPE_1, 0);
+    TEST(BTYPE_2, 0);
+    TEST(BTYPE_3, 1);
+
+    return fail;
+}
diff --git a/tests/tcg/aarch64/Makefile.target b/tests/tcg/aarch64/Makefile.target
index XXXXXXX..XXXXXXX 100644
--- a/tests/tcg/aarch64/Makefile.target
+++ b/tests/tcg/aarch64/Makefile.target
@@ -XXX,XX +XXX,XX @@ endif
 # BTI Tests
 # bti-1 tests the elf notes, so we require special compiler support.
 ifneq ($(CROSS_CC_HAS_ARMV8_BTI),)
-AARCH64_TESTS += bti-1
-bti-1: CFLAGS += -mbranch-protection=standard
-bti-1: LDFLAGS += -nostdlib
+AARCH64_TESTS += bti-1 bti-3
+bti-1 bti-3: CFLAGS += -mbranch-protection=standard
+bti-1 bti-3: LDFLAGS += -nostdlib
 endif
 # bti-2 tests PROT_BTI, so no special compiler support required.
 AARCH64_TESTS += bti-2
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Move ARMCPRegInfo and all related declarations to a new
internal header, out of the public cpu.h.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220501055028.646596-2-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/cpregs.h        | 413 +++++++++++++++++++++++++++++++++++++
 target/arm/cpu.h           | 368 ---------------------------------
 hw/arm/pxa2xx.c            |   1 +
 hw/arm/pxa2xx_pic.c        |   1 +
 hw/intc/arm_gicv3_cpuif.c  |   1 +
 hw/intc/arm_gicv3_kvm.c    |   2 +
 target/arm/cpu.c           |   1 +
 target/arm/cpu64.c         |   1 +
 target/arm/cpu_tcg.c       |   1 +
 target/arm/gdbstub.c       |   3 +-
 target/arm/helper.c        |   1 +
 target/arm/op_helper.c     |   1 +
 target/arm/translate-a64.c |   4 +-
 target/arm/translate.c     |   3 +-
 14 files changed, 427 insertions(+), 374 deletions(-)
 create mode 100644 target/arm/cpregs.h

diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/target/arm/cpregs.h
@@ -XXX,XX +XXX,XX @@
+/*
+ * QEMU ARM CP Register access and descriptions
+ *
+ * Copyright (c) 2022 Linaro Ltd
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see
+ * <http://www.gnu.org/licenses/gpl-2.0.html>
+ */
+
+#ifndef TARGET_ARM_CPREGS_H
+#define TARGET_ARM_CPREGS_H
+
+/*
+ * ARMCPRegInfo type field bits. If the SPECIAL bit is set this is a
+ * special-behaviour cp reg and bits [11..8] indicate what behaviour
+ * it has. Otherwise it is a simple cp reg, where CONST indicates that
+ * TCG can assume the value to be constant (ie load at translate time)
+ * and 64BIT indicates a 64 bit wide coprocessor register. SUPPRESS_TB_END
+ * indicates that the TB should not be ended after a write to this register
+ * (the default is that the TB ends after cp writes). OVERRIDE permits
+ * a register definition to override a previous definition for the
+ * same (cp, is64, crn, crm, opc1, opc2) tuple: either the new or the
+ * old must have the OVERRIDE bit set.
+ * ALIAS indicates that this register is an alias view of some underlying
+ * state which is also visible via another register, and that the other
+ * register is handling migration and reset; registers marked ALIAS will not be
+ * migrated but may have their state set by syncing of register state from KVM.
+ * NO_RAW indicates that this register has no underlying state and does not
+ * support raw access for state saving/loading; it will not be used for either
+ * migration or KVM state synchronization. (Typically this is for "registers"
+ * which are actually used as instructions for cache maintenance and so on.)
+ * IO indicates that this register does I/O and therefore its accesses
+ * need to be marked with gen_io_start() and also end the TB. In particular,
+ * registers which implement clocks or timers require this.
+ * RAISES_EXC is for when the read or write hook might raise an exception;
+ * the generated code will synchronize the CPU state before calling the hook
+ * so that it is safe for the hook to call raise_exception().
+ * NEWEL is for writes to registers that might change the exception
+ * level - typically on older ARM chips. For those cases we need to
+ * re-read the new el when recomputing the translation flags.
+ */
+#define ARM_CP_SPECIAL           0x0001
+#define ARM_CP_CONST             0x0002
+#define ARM_CP_64BIT             0x0004
+#define ARM_CP_SUPPRESS_TB_END   0x0008
+#define ARM_CP_OVERRIDE          0x0010
+#define ARM_CP_ALIAS             0x0020
+#define ARM_CP_IO                0x0040
+#define ARM_CP_NO_RAW            0x0080
+#define ARM_CP_NOP               (ARM_CP_SPECIAL | 0x0100)
+#define ARM_CP_WFI               (ARM_CP_SPECIAL | 0x0200)
+#define ARM_CP_NZCV              (ARM_CP_SPECIAL | 0x0300)
+#define ARM_CP_CURRENTEL         (ARM_CP_SPECIAL | 0x0400)
+#define ARM_CP_DC_ZVA            (ARM_CP_SPECIAL | 0x0500)
+#define ARM_CP_DC_GVA            (ARM_CP_SPECIAL | 0x0600)
+#define ARM_CP_DC_GZVA           (ARM_CP_SPECIAL | 0x0700)
+#define ARM_LAST_SPECIAL         ARM_CP_DC_GZVA
+#define ARM_CP_FPU               0x1000
+#define ARM_CP_SVE               0x2000
+#define ARM_CP_NO_GDB            0x4000
+#define ARM_CP_RAISES_EXC        0x8000
+#define ARM_CP_NEWEL             0x10000
+/* Used only as a terminator for ARMCPRegInfo lists */
+#define ARM_CP_SENTINEL          0xfffff
+/* Mask of only the flag bits in a type field */
+#define ARM_CP_FLAG_MASK         0x1f0ff
+
+/*
+ * Valid values for ARMCPRegInfo state field, indicating which of
+ * the AArch32 and AArch64 execution states this register is visible in.
+ * If the reginfo doesn't explicitly specify then it is AArch32 only.
+ * If the reginfo is declared to be visible in both states then a second
+ * reginfo is synthesised for the AArch32 view of the AArch64 register,
+ * such that the AArch32 view is the lower 32 bits of the AArch64 one.
+ * Note that we rely on the values of these enums as we iterate through
+ * the various states in some places.
+ */
+enum {
+    ARM_CP_STATE_AA32 = 0,
+    ARM_CP_STATE_AA64 = 1,
+    ARM_CP_STATE_BOTH = 2,
+};
+
+/*
+ * ARM CP register secure state flags.  These flags identify security state
+ * attributes for a given CP register entry.
+ * The existence of both or neither secure and non-secure flags indicates that
+ * the register has both a secure and non-secure hash entry.  A single one of
+ * these flags causes the register to only be hashed for the specified
+ * security state.
+ * Although definitions may have any combination of the S/NS bits, each
+ * registered entry will only have one to identify whether the entry is secure
+ * or non-secure.
+ */
+enum {
+    ARM_CP_SECSTATE_S =   (1 << 0), /* bit[0]: Secure state register */
+    ARM_CP_SECSTATE_NS =  (1 << 1), /* bit[1]: Non-secure state register */
+};
+
+/*
+ * Return true if cptype is a valid type field. This is used to try to
+ * catch errors where the sentinel has been accidentally left off the end
+ * of a list of registers.
+ */
+static inline bool cptype_valid(int cptype)
+{
+    return ((cptype & ~ARM_CP_FLAG_MASK) == 0)
+        || ((cptype & ARM_CP_SPECIAL) &&
+            ((cptype & ~ARM_CP_FLAG_MASK) <= ARM_LAST_SPECIAL));
+}
+
+/*
+ * Access rights:
+ * We define bits for Read and Write access for what rev C of the v7-AR ARM ARM
+ * defines as PL0 (user), PL1 (fiq/irq/svc/abt/und/sys, ie privileged), and
+ * PL2 (hyp). The other level which has Read and Write bits is Secure PL1
+ * (ie any of the privileged modes in Secure state, or Monitor mode).
+ * If a register is accessible in one privilege level it's always accessible
+ * in higher privilege levels too. Since "Secure PL1" also follows this rule
+ * (ie anything visible in PL2 is visible in S-PL1, some things are only
+ * visible in S-PL1) but "Secure PL1" is a bit of a mouthful, we bend the
+ * terminology a little and call this PL3.
+ * In AArch64 things are somewhat simpler as the PLx bits line up exactly
+ * with the ELx exception levels.
+ *
+ * If access permissions for a register are more complex than can be
+ * described with these bits, then use a laxer set of restrictions, and
+ * do the more restrictive/complex check inside a helper function.
+ */
+#define PL3_R 0x80
+#define PL3_W 0x40
+#define PL2_R (0x20 | PL3_R)
+#define PL2_W (0x10 | PL3_W)
+#define PL1_R (0x08 | PL2_R)
+#define PL1_W (0x04 | PL2_W)
+#define PL0_R (0x02 | PL1_R)
+#define PL0_W (0x01 | PL1_W)
+
+/*
+ * For user-mode some registers are accessible to EL0 via a kernel
+ * trap-and-emulate ABI. In this case we define the read permissions
+ * as actually being PL0_R. However some bits of any given register
+ * may still be masked.
+ */
+#ifdef CONFIG_USER_ONLY
+#define PL0U_R PL0_R
+#else
+#define PL0U_R PL1_R
+#endif
+
+#define PL3_RW (PL3_R | PL3_W)
+#define PL2_RW (PL2_R | PL2_W)
+#define PL1_RW (PL1_R | PL1_W)
+#define PL0_RW (PL0_R | PL0_W)
+
+typedef enum CPAccessResult {
+    /* Access is permitted */
+    CP_ACCESS_OK = 0,
+    /*
+     * Access fails due to a configurable trap or enable which would
+     * result in a categorized exception syndrome giving information about
+     * the failing instruction (ie syndrome category 0x3, 0x4, 0x5, 0x6,
+     * 0xc or 0x18). The exception is taken to the usual target EL (EL1 or
+     * PL1 if in EL0, otherwise to the current EL).
+     */
+    CP_ACCESS_TRAP = 1,
+    /*
+     * Access fails and results in an exception syndrome 0x0 ("uncategorized").
+     * Note that this is not a catch-all case -- the set of cases which may
+     * result in this failure is specifically defined by the architecture.
+     */
+    CP_ACCESS_TRAP_UNCATEGORIZED = 2,
+    /* As CP_ACCESS_TRAP, but for traps directly to EL2 or EL3 */
+    CP_ACCESS_TRAP_EL2 = 3,
+    CP_ACCESS_TRAP_EL3 = 4,
+    /* As CP_ACCESS_UNCATEGORIZED, but for traps directly to EL2 or EL3 */
+    CP_ACCESS_TRAP_UNCATEGORIZED_EL2 = 5,
+    CP_ACCESS_TRAP_UNCATEGORIZED_EL3 = 6,
+} CPAccessResult;
+
+typedef struct ARMCPRegInfo ARMCPRegInfo;
+
+/*
+ * Access functions for coprocessor registers. These cannot fail and
+ * may not raise exceptions.
+ */
+typedef uint64_t CPReadFn(CPUARMState *env, const ARMCPRegInfo *opaque);
+typedef void CPWriteFn(CPUARMState *env, const ARMCPRegInfo *opaque,
+                       uint64_t value);
+/* Access permission check functions for coprocessor registers. */
+typedef CPAccessResult CPAccessFn(CPUARMState *env,
+                                  const ARMCPRegInfo *opaque,
+                                  bool isread);
+/* Hook function for register reset */
+typedef void CPResetFn(CPUARMState *env, const ARMCPRegInfo *opaque);
+
+#define CP_ANY 0xff
+
+/* Definition of an ARM coprocessor register */
+struct ARMCPRegInfo {
+    /* Name of register (useful mainly for debugging, need not be unique) */
+    const char *name;
+    /*
+     * Location of register: coprocessor number and (crn,crm,opc1,opc2)
+     * tuple. Any of crm, opc1 and opc2 may be CP_ANY to indicate a
+     * 'wildcard' field -- any value of that field in the MRC/MCR insn
+     * will be decoded to this register. The register read and write
+     * callbacks will be passed an ARMCPRegInfo with the crn/crm/opc1/opc2
+     * used by the program, so it is possible to register a wildcard and
+     * then behave differently on read/write if necessary.
+     * For 64 bit registers, only crm and opc1 are relevant; crn and opc2
+     * must both be zero.
+     * For AArch64-visible registers, opc0 is also used.
+     * Since there are no "coprocessors" in AArch64, cp is purely used as a
+     * way to distinguish (for KVM's benefit) guest-visible system registers
+     * from demuxed ones provided to preserve the "no side effects on
+     * KVM register read/write from QEMU" semantics. cp==0x13 is guest
+     * visible (to match KVM's encoding); cp==0 will be converted to
+     * cp==0x13 when the ARMCPRegInfo is registered, for convenience.
+     */
+    uint8_t cp;
+    uint8_t crn;
+    uint8_t crm;
+    uint8_t opc0;
+    uint8_t opc1;
+    uint8_t opc2;
+    /* Execution state in which this register is visible: ARM_CP_STATE_* */
+    int state;
+    /* Register type: ARM_CP_* bits/values */
+    int type;
+    /* Access rights: PL*_[RW] */
+    int access;
+    /* Security state: ARM_CP_SECSTATE_* bits/values */
+    int secure;
+    /*
+     * The opaque pointer passed to define_arm_cp_regs_with_opaque() when
+     * this register was defined: can be used to hand data through to the
+     * register read/write functions, since they are passed the ARMCPRegInfo*.
+     */
+    void *opaque;
+    /*
+     * Value of this register, if it is ARM_CP_CONST. Otherwise, if
+     * fieldoffset is non-zero, the reset value of the register.
+     */
+    uint64_t resetvalue;
+    /*
+     * Offset of the field in CPUARMState for this register.
+     * This is not needed if either:
+     *  1. type is ARM_CP_CONST or one of the ARM_CP_SPECIALs
+     *  2. both readfn and writefn are specified
+     */
+    ptrdiff_t fieldoffset; /* offsetof(CPUARMState, field) */
+
+    /*
+     * Offsets of the secure and non-secure fields in CPUARMState for the
+     * register if it is banked.  These fields are only used during the static
+     * registration of a register.  During hashing the bank associated
+     * with a given security state is copied to fieldoffset which is used from
+     * there on out.
+     *
+     * It is expected that register definitions use either fieldoffset or
+     * bank_fieldoffsets in the definition but not both.  It is also expected
+     * that both bank offsets are set when defining a banked register.  This
+     * use indicates that a register is banked.
+     */
+    ptrdiff_t bank_fieldoffsets[2];
+
+    /*
+     * Function for making any access checks for this register in addition to
+     * those specified by the 'access' permissions bits. If NULL, no extra
+     * checks required. The access check is performed at runtime, not at
+     * translate time.
+     */
+    CPAccessFn *accessfn;
+    /*
+     * Function for handling reads of this register. If NULL, then reads
+     * will be done by loading from the offset into CPUARMState specified
+     * by fieldoffset.
+     */
+    CPReadFn *readfn;
+    /*
+     * Function for handling writes of this register. If NULL, then writes
+     * will be done by writing to the offset into CPUARMState specified
+     * by fieldoffset.
+     */
+    CPWriteFn *writefn;
+    /*
+     * Function for doing a "raw" read; used when we need to copy
+     * coprocessor state to the kernel for KVM or out for
+     * migration. This only needs to be provided if there is also a
+     * readfn and it has side effects (for instance clear-on-read bits).
+     */
+    CPReadFn *raw_readfn;
+    /*
+     * Function for doing a "raw" write; used when we need to copy KVM
+     * kernel coprocessor state into userspace, or for inbound
+     * migration. This only needs to be provided if there is also a
+     * writefn and it masks out "unwritable" bits or has write-one-to-clear
+     * or similar behaviour.
+     */
+    CPWriteFn *raw_writefn;
+    /*
+     * Function for resetting the register. If NULL, then reset will be done
+     * by writing resetvalue to the field specified in fieldoffset. If
+     * fieldoffset is 0 then no reset will be done.
+     */
+    CPResetFn *resetfn;
+
+    /*
+     * "Original" writefn and readfn.
+     * For ARMv8.1-VHE register aliases, we overwrite the read/write
+     * accessor functions of various EL1/EL0 to perform the runtime
+     * check for which sysreg should actually be modified, and then
+     * forwards the operation.  Before overwriting the accessors,
+     * the original function is copied here, so that accesses that
+     * really do go to the EL1/EL0 version proceed normally.
+     * (The corresponding EL2 register is linked via opaque.)
+     */
+    CPReadFn *orig_readfn;
+    CPWriteFn *orig_writefn;
+};
+
+/*
+ * Macros which are lvalues for the field in CPUARMState for the
+ * ARMCPRegInfo *ri.
+ */
+#define CPREG_FIELD32(env, ri) \
+    (*(uint32_t *)((char *)(env) + (ri)->fieldoffset))
+#define CPREG_FIELD64(env, ri) \
+    (*(uint64_t *)((char *)(env) + (ri)->fieldoffset))
+
+#define REGINFO_SENTINEL { .type = ARM_CP_SENTINEL }
+
+void define_arm_cp_regs_with_opaque(ARMCPU *cpu,
+                                    const ARMCPRegInfo *regs, void *opaque);
+void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
+                                       const ARMCPRegInfo *regs, void *opaque);
+static inline void define_arm_cp_regs(ARMCPU *cpu, const ARMCPRegInfo *regs)
+{
+    define_arm_cp_regs_with_opaque(cpu, regs, 0);
+}
+static inline void define_one_arm_cp_reg(ARMCPU *cpu, const ARMCPRegInfo *regs)
+{
+    define_one_arm_cp_reg_with_opaque(cpu, regs, 0);
+}
+const ARMCPRegInfo *get_arm_cp_reginfo(GHashTable *cpregs, uint32_t encoded_cp);
+
+/*
+ * Definition of an ARM co-processor register as viewed from
+ * userspace. This is used for presenting sanitised versions of
+ * registers to userspace when emulating the Linux AArch64 CPU
+ * ID/feature ABI (advertised as HWCAP_CPUID).
+ */
+typedef struct ARMCPRegUserSpaceInfo {
+    /* Name of register */
+    const char *name;
+
+    /* Is the name actually a glob pattern */
+    bool is_glob;
+
+    /* Only some bits are exported to user space */
+    uint64_t exported_bits;
+
+    /* Fixed bits are applied after the mask */
+    uint64_t fixed_bits;
+} ARMCPRegUserSpaceInfo;
+
+#define REGUSERINFO_SENTINEL { .name = NULL }
+
+void modify_arm_cp_regs(ARMCPRegInfo *regs, const ARMCPRegUserSpaceInfo *mods);
+
+/* CPWriteFn that can be used to implement writes-ignored behaviour */
+void arm_cp_write_ignore(CPUARMState *env, const ARMCPRegInfo *ri,
+                         uint64_t value);
+/* CPReadFn that can be used for read-as-zero behaviour */
+uint64_t arm_cp_read_zero(CPUARMState *env, const ARMCPRegInfo *ri);
+
+/*
+ * CPResetFn that does nothing, for use if no reset is required even
+ * if fieldoffset is non zero.
+ */
+void arm_cp_reset_ignore(CPUARMState *env, const ARMCPRegInfo *opaque);
+
+/*
+ * Return true if this reginfo struct's field in the cpu state struct
+ * is 64 bits wide.
+ */
+static inline bool cpreg_field_is_64bit(const ARMCPRegInfo *ri)
+{
+    return (ri->state == ARM_CP_STATE_AA64) || (ri->type & ARM_CP_64BIT);
+}
+
+static inline bool cp_access_ok(int current_el,
+                                const ARMCPRegInfo *ri, int isread)
+{
+    return (ri->access >> ((current_el * 2) + isread)) & 1;
+}
+
+/* Raw read of a coprocessor register (as needed for migration, etc) */
+uint64_t read_raw_cp_reg(CPUARMState *env, const ARMCPRegInfo *ri);
+
+#endif /* TARGET_ARM_CPREGS_H */
diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -XXX,XX +XXX,XX @@ static inline uint64_t cpreg_to_kvm_id(uint32_t cpregid)
     return kvmid;
 }
 
-/* ARMCPRegInfo type field bits. If the SPECIAL bit is set this is a
- * special-behaviour cp reg and bits [11..8] indicate what behaviour
- * it has. Otherwise it is a simple cp reg, where CONST indicates that
- * TCG can assume the value to be constant (ie load at translate time)
- * and 64BIT indicates a 64 bit wide coprocessor register. SUPPRESS_TB_END
- * indicates that the TB should not be ended after a write to this register
- * (the default is that the TB ends after cp writes). OVERRIDE permits
- * a register definition to override a previous definition for the
- * same (cp, is64, crn, crm, opc1, opc2) tuple: either the new or the
- * old must have the OVERRIDE bit set.
- * ALIAS indicates that this register is an alias view of some underlying
- * state which is also visible via another register, and that the other
- * register is handling migration and reset; registers marked ALIAS will not be
- * migrated but may have their state set by syncing of register state from KVM.
- * NO_RAW indicates that this register has no underlying state and does not
- * support raw access for state saving/loading; it will not be used for either
- * migration or KVM state synchronization. (Typically this is for "registers"
- * which are actually used as instructions for cache maintenance and so on.)
- * IO indicates that this register does I/O and therefore its accesses
- * need to be marked with gen_io_start() and also end the TB. In particular,
- * registers which implement clocks or timers require this.
- * RAISES_EXC is for when the read or write hook might raise an exception;
- * the generated code will synchronize the CPU state before calling the hook
- * so that it is safe for the hook to call raise_exception().
- * NEWEL is for writes to registers that might change the exception
- * level - typically on older ARM chips. For those cases we need to
- * re-read the new el when recomputing the translation flags.
- */
-#define ARM_CP_SPECIAL           0x0001
-#define ARM_CP_CONST             0x0002
-#define ARM_CP_64BIT             0x0004
-#define ARM_CP_SUPPRESS_TB_END   0x0008
-#define ARM_CP_OVERRIDE          0x0010
-#define ARM_CP_ALIAS             0x0020
-#define ARM_CP_IO                0x0040
-#define ARM_CP_NO_RAW            0x0080
-#define ARM_CP_NOP               (ARM_CP_SPECIAL | 0x0100)
-#define ARM_CP_WFI               (ARM_CP_SPECIAL | 0x0200)
-#define ARM_CP_NZCV              (ARM_CP_SPECIAL | 0x0300)
-#define ARM_CP_CURRENTEL         (ARM_CP_SPECIAL | 0x0400)
-#define ARM_CP_DC_ZVA            (ARM_CP_SPECIAL | 0x0500)
-#define ARM_CP_DC_GVA            (ARM_CP_SPECIAL | 0x0600)
-#define ARM_CP_DC_GZVA           (ARM_CP_SPECIAL | 0x0700)
-#define ARM_LAST_SPECIAL         ARM_CP_DC_GZVA
-#define ARM_CP_FPU               0x1000
-#define ARM_CP_SVE               0x2000
-#define ARM_CP_NO_GDB            0x4000
-#define ARM_CP_RAISES_EXC        0x8000
-#define ARM_CP_NEWEL             0x10000
-/* Used only as a terminator for ARMCPRegInfo lists */
-#define ARM_CP_SENTINEL          0xfffff
-/* Mask of only the flag bits in a type field */
-#define ARM_CP_FLAG_MASK         0x1f0ff
-
-/* Valid values for ARMCPRegInfo state field, indicating which of
- * the AArch32 and AArch64 execution states this register is visible in.
- * If the reginfo doesn't explicitly specify then it is AArch32 only.
- * If the reginfo is declared to be visible in both states then a second
- * reginfo is synthesised for the AArch32 view of the AArch64 register,
- * such that the AArch32 view is the lower 32 bits of the AArch64 one.
- * Note that we rely on the values of these enums as we iterate through
- * the various states in some places.
- */
-enum {
-    ARM_CP_STATE_AA32 = 0,
-    ARM_CP_STATE_AA64 = 1,
-    ARM_CP_STATE_BOTH = 2,
-};
-
-/* ARM CP register secure state flags.  These flags identify security state
- * attributes for a given CP register entry.
- * The existence of both or neither secure and non-secure flags indicates that
- * the register has both a secure and non-secure hash entry.  A single one of
- * these flags causes the register to only be hashed for the specified
- * security state.
- * Although definitions may have any combination of the S/NS bits, each
- * registered entry will only have one to identify whether the entry is secure
- * or non-secure.
- */
-enum {
-    ARM_CP_SECSTATE_S =   (1 << 0), /* bit[0]: Secure state register */
-    ARM_CP_SECSTATE_NS =  (1 << 1), /* bit[1]: Non-secure state register */
-};
-
-/* Return true if cptype is a valid type field. This is used to try to
- * catch errors where the sentinel has been accidentally left off the end
- * of a list of registers.
- */
-static inline bool cptype_valid(int cptype)
-{
-    return ((cptype & ~ARM_CP_FLAG_MASK) == 0)
-        || ((cptype & ARM_CP_SPECIAL) &&
-            ((cptype & ~ARM_CP_FLAG_MASK) <= ARM_LAST_SPECIAL));
-}
-
-/* Access rights:
- * We define bits for Read and Write access for what rev C of the v7-AR ARM ARM
- * defines as PL0 (user), PL1 (fiq/irq/svc/abt/und/sys, ie privileged), and
- * PL2 (hyp). The other level which has Read and Write bits is Secure PL1
- * (ie any of the privileged modes in Secure state, or Monitor mode).
- * If a register is accessible in one privilege level it's always accessible
- * in higher privilege levels too. Since "Secure PL1" also follows this rule
- * (ie anything visible in PL2 is visible in S-PL1, some things are only
- * visible in S-PL1) but "Secure PL1" is a bit of a mouthful, we bend the
- * terminology a little and call this PL3.
- * In AArch64 things are somewhat simpler as the PLx bits line up exactly
- * with the ELx exception levels.
- *
- * If access permissions for a register are more complex than can be
- * described with these bits, then use a laxer set of restrictions, and
- * do the more restrictive/complex check inside a helper function.
- */
-#define PL3_R 0x80
-#define PL3_W 0x40
-#define PL2_R (0x20 | PL3_R)
-#define PL2_W (0x10 | PL3_W)
-#define PL1_R (0x08 | PL2_R)
-#define PL1_W (0x04 | PL2_W)
-#define PL0_R (0x02 | PL1_R)
-#define PL0_W (0x01 | PL1_W)
-
-/*
- * For user-mode some registers are accessible to EL0 via a kernel
- * trap-and-emulate ABI. In this case we define the read permissions
- * as actually being PL0_R. However some bits of any given register
- * may still be masked.
- */
-#ifdef CONFIG_USER_ONLY
-#define PL0U_R PL0_R
-#else
-#define PL0U_R PL1_R
-#endif
-
-#define PL3_RW (PL3_R | PL3_W)
-#define PL2_RW (PL2_R | PL2_W)
-#define PL1_RW (PL1_R | PL1_W)
-#define PL0_RW (PL0_R | PL0_W)
-
 /* Return the highest implemented Exception Level */
 static inline int arm_highest_el(CPUARMState *env)
 {
@@ -XXX,XX +XXX,XX @@ static inline int arm_current_el(CPUARMState *env)
     }
 }
 
-typedef struct ARMCPRegInfo ARMCPRegInfo;
-
-typedef enum CPAccessResult {
-    /* Access is permitted */
-    CP_ACCESS_OK = 0,
-    /* Access fails due to a configurable trap or enable which would
-     * result in a categorized exception syndrome giving information about
-     * the failing instruction (ie syndrome category 0x3, 0x4, 0x5, 0x6,
-     * 0xc or 0x18). The exception is taken to the usual target EL (EL1 or
-     * PL1 if in EL0, otherwise to the current EL).
-     */
-    CP_ACCESS_TRAP = 1,
-    /* Access fails and results in an exception syndrome 0x0 ("uncategorized").
-     * Note that this is not a catch-all case -- the set of cases which may
-     * result in this failure is specifically defined by the architecture.
-     */
-    CP_ACCESS_TRAP_UNCATEGORIZED = 2,
-    /* As CP_ACCESS_TRAP, but for traps directly to EL2 or EL3 */
-    CP_ACCESS_TRAP_EL2 = 3,
-    CP_ACCESS_TRAP_EL3 = 4,
-    /* As CP_ACCESS_UNCATEGORIZED, but for traps directly to EL2 or EL3 */
-    CP_ACCESS_TRAP_UNCATEGORIZED_EL2 = 5,
-    CP_ACCESS_TRAP_UNCATEGORIZED_EL3 = 6,
-} CPAccessResult;
-
-/* Access functions for coprocessor registers. These cannot fail and
- * may not raise exceptions.
- */
-typedef uint64_t CPReadFn(CPUARMState *env, const ARMCPRegInfo *opaque);
-typedef void CPWriteFn(CPUARMState *env, const ARMCPRegInfo *opaque,
-                       uint64_t value);
-/* Access permission check functions for coprocessor registers. */
-typedef CPAccessResult CPAccessFn(CPUARMState *env,
-                                  const ARMCPRegInfo *opaque,
-                                  bool isread);
-/* Hook function for register reset */
-typedef void CPResetFn(CPUARMState *env, const ARMCPRegInfo *opaque);
-
-#define CP_ANY 0xff
-
-/* Definition of an ARM coprocessor register */
-struct ARMCPRegInfo {
-    /* Name of register (useful mainly for debugging, need not be unique) */
-    const char *name;
-    /* Location of register: coprocessor number and (crn,crm,opc1,opc2)
-     * tuple. Any of crm, opc1 and opc2 may be CP_ANY to indicate a
-     * 'wildcard' field -- any value of that field in the MRC/MCR insn
-     * will be decoded to this register. The register read and write
-     * callbacks will be passed an ARMCPRegInfo with the crn/crm/opc1/opc2
-     * used by the program, so it is possible to register a wildcard and
-     * then behave differently on read/write if necessary.
-     * For 64 bit registers, only crm and opc1 are relevant; crn and opc2
-     * must both be zero.
-     * For AArch64-visible registers, opc0 is also used.
-     * Since there are no "coprocessors" in AArch64, cp is purely used as a
-     * way to distinguish (for KVM's benefit) guest-visible system registers
-     * from demuxed ones provided to preserve the "no side effects on
-     * KVM register read/write from QEMU" semantics. cp==0x13 is guest
-     * visible (to match KVM's encoding); cp==0 will be converted to
-     * cp==0x13 when the ARMCPRegInfo is registered, for convenience.
-     */
-    uint8_t cp;
-    uint8_t crn;
-    uint8_t crm;
-    uint8_t opc0;
-    uint8_t opc1;
-    uint8_t opc2;
-    /* Execution state in which this register is visible: ARM_CP_STATE_* */
-    int state;
-    /* Register type: ARM_CP_* bits/values */
-    int type;
-    /* Access rights: PL*_[RW] */
-    int access;
-    /* Security state: ARM_CP_SECSTATE_* bits/values */
-    int secure;
-    /* The opaque pointer passed to define_arm_cp_regs_with_opaque() when
-     * this register was defined: can be used to hand data through to the
-     * register read/write functions, since they are passed the ARMCPRegInfo*.
-     */
-    void *opaque;
-    /* Value of this register, if it is ARM_CP_CONST. Otherwise, if
-     * fieldoffset is non-zero, the reset value of the register.
-     */
-    uint64_t resetvalue;
-    /* Offset of the field in CPUARMState for this register.
-     *
-     * This is not needed if either:
-     *  1. type is ARM_CP_CONST or one of the ARM_CP_SPECIALs
-     *  2. both readfn and writefn are specified
-     */
-    ptrdiff_t fieldoffset; /* offsetof(CPUARMState, field) */
-
-    /* Offsets of the secure and non-secure fields in CPUARMState for the
-     * register if it is banked.  These fields are only used during the static
-     * registration of a register.  During hashing the bank associated
-     * with a given security state is copied to fieldoffset which is used from
-     * there on out.
-     *
-     * It is expected that register definitions use either fieldoffset or
-     * bank_fieldoffsets in the definition but not both.  It is also expected
-     * that both bank offsets are set when defining a banked register.  This
-     * use indicates that a register is banked.
-     */
-    ptrdiff_t bank_fieldoffsets[2];
-
-    /* Function for making any access checks for this register in addition to
-     * those specified by the 'access' permissions bits. If NULL, no extra
-     * checks required. The access check is performed at runtime, not at
-     * translate time.
-     */
-    CPAccessFn *accessfn;
-    /* Function for handling reads of this register. If NULL, then reads
-     * will be done by loading from the offset into CPUARMState specified
-     * by fieldoffset.
-     */
-    CPReadFn *readfn;
-    /* Function for handling writes of this register. If NULL, then writes
-     * will be done by writing to the offset into CPUARMState specified
-     * by fieldoffset.
-     */
-    CPWriteFn *writefn;
-    /* Function for doing a "raw" read; used when we need to copy
-     * coprocessor state to the kernel for KVM or out for
-     * migration. This only needs to be provided if there is also a
-     * readfn and it has side effects (for instance clear-on-read bits).
-     */
-    CPReadFn *raw_readfn;
-    /* Function for doing a "raw" write; used when we need to copy KVM
-     * kernel coprocessor state into userspace, or for inbound
-     * migration. This only needs to be provided if there is also a
-     * writefn and it masks out "unwritable" bits or has write-one-to-clear
-     * or similar behaviour.
-     */
-    CPWriteFn *raw_writefn;
-    /* Function for resetting the register. If NULL, then reset will be done
-     * by writing resetvalue to the field specified in fieldoffset. If
-     * fieldoffset is 0 then no reset will be done.
-     */
-    CPResetFn *resetfn;
-
-    /*
-     * "Original" writefn and readfn.
-     * For ARMv8.1-VHE register aliases, we overwrite the read/write
-     * accessor functions of various EL1/EL0 to perform the runtime
-     * check for which sysreg should actually be modified, and then
-     * forwards the operation.  Before overwriting the accessors,
-     * the original function is copied here, so that accesses that
-     * really do go to the EL1/EL0 version proceed normally.
-     * (The corresponding EL2 register is linked via opaque.)
-     */
-    CPReadFn *orig_readfn;
-    CPWriteFn *orig_writefn;
-};
-
-/* Macros which are lvalues for the field in CPUARMState for the
- * ARMCPRegInfo *ri.
- */
-#define CPREG_FIELD32(env, ri) \
-    (*(uint32_t *)((char *)(env) + (ri)->fieldoffset))
-#define CPREG_FIELD64(env, ri) \
-    (*(uint64_t *)((char *)(env) + (ri)->fieldoffset))
-
-#define REGINFO_SENTINEL { .type = ARM_CP_SENTINEL }
-
-void define_arm_cp_regs_with_opaque(ARMCPU *cpu,
-                                    const ARMCPRegInfo *regs, void *opaque);
-void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
-                                       const ARMCPRegInfo *regs, void *opaque);
-static inline void define_arm_cp_regs(ARMCPU *cpu, const ARMCPRegInfo *regs)
-{
-    define_arm_cp_regs_with_opaque(cpu, regs, 0);
-}
-static inline void define_one_arm_cp_reg(ARMCPU *cpu, const ARMCPRegInfo *regs)
-{
-    define_one_arm_cp_reg_with_opaque(cpu, regs, 0);
-}
-const ARMCPRegInfo *get_arm_cp_reginfo(GHashTable *cpregs, uint32_t encoded_cp);
-
-/*
- * Definition of an ARM co-processor register as viewed from
- * userspace. This is used for presenting sanitised versions of
- * registers to userspace when emulating the Linux AArch64 CPU
- * ID/feature ABI (advertised as HWCAP_CPUID).
- */
-typedef struct ARMCPRegUserSpaceInfo {
-    /* Name of register */
-    const char *name;
-
-    /* Is the name actually a glob pattern */
-    bool is_glob;
-
-    /* Only some bits are exported to user space */
-    uint64_t exported_bits;
-
-    /* Fixed bits are applied after the mask */
-    uint64_t fixed_bits;
-} ARMCPRegUserSpaceInfo;
-
-#define REGUSERINFO_SENTINEL { .name = NULL }
-
-void modify_arm_cp_regs(ARMCPRegInfo *regs, const ARMCPRegUserSpaceInfo *mods);
-
-/* CPWriteFn that can be used to implement writes-ignored behaviour */
-void arm_cp_write_ignore(CPUARMState *env, const ARMCPRegInfo *ri,
-                         uint64_t value);
-/* CPReadFn that can be used for read-as-zero behaviour */
-uint64_t arm_cp_read_zero(CPUARMState *env, const ARMCPRegInfo *ri);
-
-/* CPResetFn that does nothing, for use if no reset is required even
- * if fieldoffset is non zero.
- */
-void arm_cp_reset_ignore(CPUARMState *env, const ARMCPRegInfo *opaque);
-
-/* Return true if this reginfo struct's field in the cpu state struct
- * is 64 bits wide.
- */
-static inline bool cpreg_field_is_64bit(const ARMCPRegInfo *ri)
-{
-    return (ri->state == ARM_CP_STATE_AA64) || (ri->type & ARM_CP_64BIT);
-}
-
-static inline bool cp_access_ok(int current_el,
-                                const ARMCPRegInfo *ri, int isread)
-{
-    return (ri->access >> ((current_el * 2) + isread)) & 1;
-}
-
-/* Raw read of a coprocessor register (as needed for migration, etc) */
-uint64_t read_raw_cp_reg(CPUARMState *env, const ARMCPRegInfo *ri);
-
 /**
  * write_list_to_cpustate
  * @cpu: ARMCPU
diff --git a/hw/arm/pxa2xx.c b/hw/arm/pxa2xx.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/arm/pxa2xx.c
+++ b/hw/arm/pxa2xx.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/cutils.h"
 #include "qemu/log.h"
 #include "qom/object.h"
+#include "target/arm/cpregs.h"
 
 static struct {
     hwaddr io_base;
diff --git a/hw/arm/pxa2xx_pic.c b/hw/arm/pxa2xx_pic.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/arm/pxa2xx_pic.c
+++ b/hw/arm/pxa2xx_pic.c
@@ -XXX,XX +XXX,XX @@
 #include "hw/sysbus.h"
 #include "migration/vmstate.h"
 #include "qom/object.h"
+#include "target/arm/cpregs.h"
 
 #define ICIP	0x00	/* Interrupt Controller IRQ Pending register */
 #define ICMR	0x04	/* Interrupt Controller Mask register */
diff --git a/hw/intc/arm_gicv3_cpuif.c b/hw/intc/arm_gicv3_cpuif.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/intc/arm_gicv3_cpuif.c
+++ b/hw/intc/arm_gicv3_cpuif.c
@@ -XXX,XX +XXX,XX @@
 #include "gicv3_internal.h"
 #include "hw/irq.h"
 #include "cpu.h"
+#include "target/arm/cpregs.h"
 
 /*
  * Special case return value from hppvi_index(); must be larger than
diff --git a/hw/intc/arm_gicv3_kvm.c b/hw/intc/arm_gicv3_kvm.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/intc/arm_gicv3_kvm.c
+++ b/hw/intc/arm_gicv3_kvm.c
@@ -XXX,XX +XXX,XX @@
 #include "vgic_common.h"
 #include "migration/blocker.h"
 #include "qom/object.h"
+#include "target/arm/cpregs.h"
+
 
 #ifdef DEBUG_GICV3_KVM
 #define DPRINTF(fmt, ...) \
diff --git a/target/arm/cpu.c b/target/arm/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu.c
+++ b/target/arm/cpu.c
@@ -XXX,XX +XXX,XX @@
 #include "kvm_arm.h"
 #include "disas/capstone.h"
 #include "fpu/softfloat.h"
+#include "cpregs.h"
 
 static void arm_cpu_set_pc(CPUState *cs, vaddr value)
 {
diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu64.c
+++ b/target/arm/cpu64.c
@@ -XXX,XX +XXX,XX @@
 #include "hvf_arm.h"
 #include "qapi/visitor.h"
 #include "hw/qdev-properties.h"
+#include "cpregs.h"
 
 
 #ifndef CONFIG_USER_ONLY
diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu_tcg.c
+++ b/target/arm/cpu_tcg.c
@@ -XXX,XX +XXX,XX @@
 #if !defined(CONFIG_USER_ONLY)
 #include "hw/boards.h"
 #endif
+#include "cpregs.h"
 
 /* CPU models. These are not needed for the AArch64 linux-user build. */
 #if !defined(CONFIG_USER_ONLY) || !defined(TARGET_AARCH64)
diff --git a/target/arm/gdbstub.c b/target/arm/gdbstub.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/gdbstub.c
+++ b/target/arm/gdbstub.c
@@ -XXX,XX +XXX,XX @@
  */
 #include "qemu/osdep.h"
 #include "cpu.h"
-#include "internals.h"
 #include "exec/gdbstub.h"
+#include "internals.h"
+#include "cpregs.h"
 
 typedef struct RegisterSysregXmlParam {
     CPUState *cs;
diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@
 #include "exec/cpu_ldst.h"
 #include "semihosting/common-semi.h"
 #endif
+#include "cpregs.h"
 
 #define ARM_CPU_FREQ 1000000000 /* FIXME: 1 GHz, should be configurable */
 #define PMCR_NUM_COUNTERS 4 /* QEMU IMPDEF choice */
diff --git a/target/arm/op_helper.c b/target/arm/op_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/op_helper.c
+++ b/target/arm/op_helper.c
@@ -XXX,XX +XXX,XX @@
 #include "internals.h"
 #include "exec/exec-all.h"
 #include "exec/cpu_ldst.h"
+#include "cpregs.h"
 
 #define SIGNBIT (uint32_t)0x80000000
 #define SIGNBIT64 ((uint64_t)1 << 63)
diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/translate-a64.c
+++ b/target/arm/translate-a64.c
@@ -XXX,XX +XXX,XX @@
 #include "translate.h"
 #include "internals.h"
 #include "qemu/host-utils.h"
-
 #include "semihosting/semihost.h"
 #include "exec/gen-icount.h"
-
 #include "exec/helper-proto.h"
 #include "exec/helper-gen.h"
 #include "exec/log.h"
-
+#include "cpregs.h"
 #include "translate-a64.h"
 #include "qemu/atomic128.h"
 
diff --git a/target/arm/translate.c b/target/arm/translate.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/bitops.h"
 #include "arm_ldst.h"
 #include "semihosting/semihost.h"
-
 #include "exec/helper-proto.h"
 #include "exec/helper-gen.h"
-
 #include "exec/log.h"
+#include "cpregs.h"
 
 
 #define ENABLE_ARCH_4T    arm_dc_feature(s, ARM_FEATURE_V4T)
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Rearrange the values of the enumerators of CPAccessResult
so that we may directly extract the target el. For the two
special cases in access_check_cp_reg, use CPAccessResult.

diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpregs.h
+++ b/target/arm/cpregs.h
@@ -XXX,XX +XXX,XX @@ static inline bool cptype_valid(int cptype)
 typedef enum CPAccessResult {
     /* Access is permitted */
     CP_ACCESS_OK = 0,
+
+    /*
+     * Combined with one of the following, the low 2 bits indicate the
+     * target exception level.  If 0, the exception is taken to the usual
+     * target EL (EL1 or PL1 if in EL0, otherwise to the current EL).
+     */
+    CP_ACCESS_EL_MASK = 3,
+
     /*
      * Access fails due to a configurable trap or enable which would
      * result in a categorized exception syndrome giving information about
      * the failing instruction (ie syndrome category 0x3, 0x4, 0x5, 0x6,
-     * 0xc or 0x18). The exception is taken to the usual target EL (EL1 or
-     * PL1 if in EL0, otherwise to the current EL).
+     * 0xc or 0x18).
      */
-    CP_ACCESS_TRAP = 1,
+    CP_ACCESS_TRAP = (1 << 2),
+    CP_ACCESS_TRAP_EL2 = CP_ACCESS_TRAP | 2,
+    CP_ACCESS_TRAP_EL3 = CP_ACCESS_TRAP | 3,
+
     /*
      * Access fails and results in an exception syndrome 0x0 ("uncategorized").
      * Note that this is not a catch-all case -- the set of cases which may
      * result in this failure is specifically defined by the architecture.
      */
-    CP_ACCESS_TRAP_UNCATEGORIZED = 2,
-    /* As CP_ACCESS_TRAP, but for traps directly to EL2 or EL3 */
-    CP_ACCESS_TRAP_EL2 = 3,
-    CP_ACCESS_TRAP_EL3 = 4,
-    /* As CP_ACCESS_UNCATEGORIZED, but for traps directly to EL2 or EL3 */
-    CP_ACCESS_TRAP_UNCATEGORIZED_EL2 = 5,
-    CP_ACCESS_TRAP_UNCATEGORIZED_EL3 = 6,
+    CP_ACCESS_TRAP_UNCATEGORIZED = (2 << 2),
+    CP_ACCESS_TRAP_UNCATEGORIZED_EL2 = CP_ACCESS_TRAP_UNCATEGORIZED | 2,
+    CP_ACCESS_TRAP_UNCATEGORIZED_EL3 = CP_ACCESS_TRAP_UNCATEGORIZED | 3,
 } CPAccessResult;
 
 typedef struct ARMCPRegInfo ARMCPRegInfo;
diff --git a/target/arm/op_helper.c b/target/arm/op_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/op_helper.c
+++ b/target/arm/op_helper.c
@@ -XXX,XX +XXX,XX @@ void HELPER(access_check_cp_reg)(CPUARMState *env, void *rip, uint32_t syndrome,
                                  uint32_t isread)
 {
     const ARMCPRegInfo *ri = rip;
+    CPAccessResult res = CP_ACCESS_OK;
     int target_el;
 
     if (arm_feature(env, ARM_FEATURE_XSCALE) && ri->cp < 14
         && extract32(env->cp15.c15_cpar, ri->cp, 1) == 0) {
-        raise_exception(env, EXCP_UDEF, syndrome, exception_target_el(env));
+        res = CP_ACCESS_TRAP;
+        goto fail;
     }
 
     /*
@@ -XXX,XX +XXX,XX @@ void HELPER(access_check_cp_reg)(CPUARMState *env, void *rip, uint32_t syndrome,
         mask &= ~((1 << 4) | (1 << 14));
 
         if (env->cp15.hstr_el2 & mask) {
-            target_el = 2;
-            goto exept;
+            res = CP_ACCESS_TRAP_EL2;
+            goto fail;
         }
     }
 
-    if (!ri->accessfn) {
+    if (ri->accessfn) {
+        res = ri->accessfn(env, ri, isread);
+    }
+    if (likely(res == CP_ACCESS_OK)) {
         return;
     }
 
-    switch (ri->accessfn(env, ri, isread)) {
-    case CP_ACCESS_OK:
-        return;
+ fail:
+    switch (res & ~CP_ACCESS_EL_MASK) {
     case CP_ACCESS_TRAP:
-        target_el = exception_target_el(env);
-        break;
-    case CP_ACCESS_TRAP_EL2:
-        /* Requesting a trap to EL2 when we're in EL3 is
-         * a bug in the access function.
-         */
-        assert(arm_current_el(env) != 3);
-        target_el = 2;
-        break;
-    case CP_ACCESS_TRAP_EL3:
-        target_el = 3;
         break;
     case CP_ACCESS_TRAP_UNCATEGORIZED:
-        target_el = exception_target_el(env);
-        syndrome = syn_uncategorized();
-        break;
-    case CP_ACCESS_TRAP_UNCATEGORIZED_EL2:
-        target_el = 2;
-        syndrome = syn_uncategorized();
-        break;
-    case CP_ACCESS_TRAP_UNCATEGORIZED_EL3:
-        target_el = 3;
         syndrome = syn_uncategorized();
         break;
     default:
         g_assert_not_reached();
     }
 
-exept:
+    target_el = res & CP_ACCESS_EL_MASK;
+    switch (target_el) {
+    case 0:
+        target_el = exception_target_el(env);
+        break;
+    case 2:
+        assert(arm_current_el(env) != 3);
+        assert(arm_is_el2_enabled(env));
+        break;
+    case 3:
+        assert(arm_feature(env, ARM_FEATURE_EL3));
+        break;
+    default:
+        /* No "direct" traps to EL1 */
+        g_assert_not_reached();
+    }
+
     raise_exception(env, EXCP_UDEF, syndrome, target_el);
 }
 
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Remove a possible source of error by removing REGINFO_SENTINEL
and using ARRAY_SIZE (convinently hidden inside a macro) to
find the end of the set of regs being registered or modified.

The space saved by not having the extra array element reduces
the executable's .data.rel.ro section by about 9k.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220501055028.646596-4-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/cpregs.h       |  53 +++++++++---------
 hw/arm/pxa2xx.c           |   1 -
 hw/arm/pxa2xx_pic.c       |   1 -
 hw/intc/arm_gicv3_cpuif.c |   5 --
 hw/intc/arm_gicv3_kvm.c   |   1 -
 target/arm/cpu64.c        |   1 -
 target/arm/cpu_tcg.c      |   4 --
 target/arm/helper.c       | 111 ++++++++------------------------------
 8 files changed, 48 insertions(+), 129 deletions(-)

diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpregs.h
+++ b/target/arm/cpregs.h
@@ -XXX,XX +XXX,XX @@
 #define ARM_CP_NO_GDB            0x4000
 #define ARM_CP_RAISES_EXC        0x8000
 #define ARM_CP_NEWEL             0x10000
-/* Used only as a terminator for ARMCPRegInfo lists */
-#define ARM_CP_SENTINEL          0xfffff
 /* Mask of only the flag bits in a type field */
 #define ARM_CP_FLAG_MASK         0x1f0ff
 
@@ -XXX,XX +XXX,XX @@ enum {
     ARM_CP_SECSTATE_NS =  (1 << 1), /* bit[1]: Non-secure state register */
 };
 
-/*
- * Return true if cptype is a valid type field. This is used to try to
- * catch errors where the sentinel has been accidentally left off the end
- * of a list of registers.
- */
-static inline bool cptype_valid(int cptype)
-{
-    return ((cptype & ~ARM_CP_FLAG_MASK) == 0)
-        || ((cptype & ARM_CP_SPECIAL) &&
-            ((cptype & ~ARM_CP_FLAG_MASK) <= ARM_LAST_SPECIAL));
-}
-
 /*
  * Access rights:
  * We define bits for Read and Write access for what rev C of the v7-AR ARM ARM
@@ -XXX,XX +XXX,XX @@ struct ARMCPRegInfo {
 #define CPREG_FIELD64(env, ri) \
     (*(uint64_t *)((char *)(env) + (ri)->fieldoffset))
 
-#define REGINFO_SENTINEL { .type = ARM_CP_SENTINEL }
+void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu, const ARMCPRegInfo *reg,
+                                       void *opaque);
 
-void define_arm_cp_regs_with_opaque(ARMCPU *cpu,
-                                    const ARMCPRegInfo *regs, void *opaque);
-void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
-                                       const ARMCPRegInfo *regs, void *opaque);
-static inline void define_arm_cp_regs(ARMCPU *cpu, const ARMCPRegInfo *regs)
-{
-    define_arm_cp_regs_with_opaque(cpu, regs, 0);
-}
 static inline void define_one_arm_cp_reg(ARMCPU *cpu, const ARMCPRegInfo *regs)
 {
-    define_one_arm_cp_reg_with_opaque(cpu, regs, 0);
+    define_one_arm_cp_reg_with_opaque(cpu, regs, NULL);
 }
+
+void define_arm_cp_regs_with_opaque_len(ARMCPU *cpu, const ARMCPRegInfo *regs,
+                                        void *opaque, size_t len);
+
+#define define_arm_cp_regs_with_opaque(CPU, REGS, OPAQUE)               \
+    do {                                                                \
+        QEMU_BUILD_BUG_ON(ARRAY_SIZE(REGS) == 0);                       \
+        define_arm_cp_regs_with_opaque_len(CPU, REGS, OPAQUE,           \
+                                           ARRAY_SIZE(REGS));           \
+    } while (0)
+
+#define define_arm_cp_regs(CPU, REGS) \
+    define_arm_cp_regs_with_opaque(CPU, REGS, NULL)
+
 const ARMCPRegInfo *get_arm_cp_reginfo(GHashTable *cpregs, uint32_t encoded_cp);
 
 /*
@@ -XXX,XX +XXX,XX @@ typedef struct ARMCPRegUserSpaceInfo {
     uint64_t fixed_bits;
 } ARMCPRegUserSpaceInfo;
 
-#define REGUSERINFO_SENTINEL { .name = NULL }
+void modify_arm_cp_regs_with_len(ARMCPRegInfo *regs, size_t regs_len,
+                                 const ARMCPRegUserSpaceInfo *mods,
+                                 size_t mods_len);
 
-void modify_arm_cp_regs(ARMCPRegInfo *regs, const ARMCPRegUserSpaceInfo *mods);
+#define modify_arm_cp_regs(REGS, MODS)                                  \
+    do {                                                                \
+        QEMU_BUILD_BUG_ON(ARRAY_SIZE(REGS) == 0);                       \
+        QEMU_BUILD_BUG_ON(ARRAY_SIZE(MODS) == 0);                       \
+        modify_arm_cp_regs_with_len(REGS, ARRAY_SIZE(REGS),             \
+                                    MODS, ARRAY_SIZE(MODS));            \
+    } while (0)
 
 /* CPWriteFn that can be used to implement writes-ignored behaviour */
 void arm_cp_write_ignore(CPUARMState *env, const ARMCPRegInfo *ri,
diff --git a/hw/arm/pxa2xx.c b/hw/arm/pxa2xx.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/arm/pxa2xx.c
+++ b/hw/arm/pxa2xx.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo pxa_cp_reginfo[] = {
     { .name = "PWRMODE", .cp = 14, .crn = 7, .crm = 0, .opc1 = 0, .opc2 = 0,
       .access = PL1_RW, .type = ARM_CP_IO,
       .readfn = arm_cp_read_zero, .writefn = pxa2xx_pwrmode_write },
-    REGINFO_SENTINEL
 };
 
 static void pxa2xx_setup_cp14(PXA2xxState *s)
diff --git a/hw/arm/pxa2xx_pic.c b/hw/arm/pxa2xx_pic.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/arm/pxa2xx_pic.c
+++ b/hw/arm/pxa2xx_pic.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo pxa_pic_cp_reginfo[] = {
     REGINFO_FOR_PIC_CP("ICLR2", 8),
     REGINFO_FOR_PIC_CP("ICFP2", 9),
     REGINFO_FOR_PIC_CP("ICPR2", 0xa),
-    REGINFO_SENTINEL
 };
 
 static const MemoryRegionOps pxa2xx_pic_ops = {
diff --git a/hw/intc/arm_gicv3_cpuif.c b/hw/intc/arm_gicv3_cpuif.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/intc/arm_gicv3_cpuif.c
+++ b/hw/intc/arm_gicv3_cpuif.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo gicv3_cpuif_reginfo[] = {
       .readfn = icc_igrpen1_el3_read,
       .writefn = icc_igrpen1_el3_write,
     },
-    REGINFO_SENTINEL
 };
 
 static uint64_t ich_ap_read(CPUARMState *env, const ARMCPRegInfo *ri)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo gicv3_cpuif_hcr_reginfo[] = {
       .readfn = ich_vmcr_read,
       .writefn = ich_vmcr_write,
     },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo gicv3_cpuif_ich_apxr1_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo gicv3_cpuif_ich_apxr1_reginfo[] = {
       .readfn = ich_ap_read,
       .writefn = ich_ap_write,
     },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo gicv3_cpuif_ich_apxr23_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo gicv3_cpuif_ich_apxr23_reginfo[] = {
       .readfn = ich_ap_read,
       .writefn = ich_ap_write,
     },
-    REGINFO_SENTINEL
 };
 
 static void gicv3_cpuif_el_change_hook(ARMCPU *cpu, void *opaque)
@@ -XXX,XX +XXX,XX @@ void gicv3_init_cpuif(GICv3State *s)
                       .readfn = ich_lr_read,
                       .writefn = ich_lr_write,
                     },
-                    REGINFO_SENTINEL
                 };
                 define_arm_cp_regs(cpu, lr_regset);
             }
diff --git a/hw/intc/arm_gicv3_kvm.c b/hw/intc/arm_gicv3_kvm.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/intc/arm_gicv3_kvm.c
+++ b/hw/intc/arm_gicv3_kvm.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo gicv3_cpuif_reginfo[] = {
        */
       .resetfn = arm_gicv3_icc_reset,
     },
-    REGINFO_SENTINEL
 };
 
 /**
diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu64.c
+++ b/target/arm/cpu64.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cortex_a72_a57_a53_cp_reginfo[] = {
     { .name = "L2MERRSR",
       .cp = 15, .opc1 = 3, .crm = 15,
       .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 static void aarch64_a57_initfn(Object *obj)
diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu_tcg.c
+++ b/target/arm/cpu_tcg.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cortexa8_cp_reginfo[] = {
       .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 },
     { .name = "L2AUXCR", .cp = 15, .crn = 9, .crm = 0, .opc1 = 1, .opc2 = 2,
       .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 static void cortex_a8_initfn(Object *obj)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cortexa9_cp_reginfo[] = {
       .access = PL1_RW, .resetvalue = 0, .type = ARM_CP_CONST },
     { .name = "TLB_ATTR", .cp = 15, .crn = 15, .crm = 7, .opc1 = 5, .opc2 = 2,
       .access = PL1_RW, .resetvalue = 0, .type = ARM_CP_CONST },
-    REGINFO_SENTINEL
 };
 
 static void cortex_a9_initfn(Object *obj)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cortexa15_cp_reginfo[] = {
 #endif
     { .name = "L2ECTLR", .cp = 15, .crn = 9, .crm = 0, .opc1 = 1, .opc2 = 3,
       .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 static void cortex_a7_initfn(Object *obj)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cortexr5_cp_reginfo[] = {
       .access = PL1_RW, .type = ARM_CP_CONST },
     { .name = "DCACHE_INVAL", .cp = 15, .opc1 = 0, .crn = 15, .crm = 5,
       .opc2 = 0, .access = PL1_W, .type = ARM_CP_NOP },
-    REGINFO_SENTINEL
 };
 
 static void cortex_r5_initfn(Object *obj)
diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cp_reginfo[] = {
       .secure = ARM_CP_SECSTATE_S,
       .fieldoffset = offsetof(CPUARMState, cp15.contextidr_s),
       .resetvalue = 0, .writefn = contextidr_write, .raw_writefn = raw_write, },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo not_v8_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo not_v8_cp_reginfo[] = {
     { .name = "CACHEMAINT", .cp = 15, .crn = 7, .crm = CP_ANY,
       .opc1 = 0, .opc2 = CP_ANY, .access = PL1_W,
       .type = ARM_CP_NOP | ARM_CP_OVERRIDE },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo not_v6_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo not_v6_cp_reginfo[] = {
      */
     { .name = "WFI_v5", .cp = 15, .crn = 7, .crm = 8, .opc1 = 0, .opc2 = 2,
       .access = PL1_W, .type = ARM_CP_WFI },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo not_v7_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo not_v7_cp_reginfo[] = {
       .opc1 = 0, .opc2 = 0, .access = PL1_RW, .type = ARM_CP_NOP },
     { .name = "NMRR", .cp = 15, .crn = 10, .crm = 2,
       .opc1 = 0, .opc2 = 1, .access = PL1_RW, .type = ARM_CP_NOP },
-    REGINFO_SENTINEL
 };
 
 static void cpacr_write(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v6_cp_reginfo[] = {
       .crn = 1, .crm = 0, .opc1 = 0, .opc2 = 2, .accessfn = cpacr_access,
       .access = PL1_RW, .fieldoffset = offsetof(CPUARMState, cp15.cpacr_el1),
       .resetfn = cpacr_reset, .writefn = cpacr_write, .readfn = cpacr_read },
-    REGINFO_SENTINEL
 };
 
 typedef struct pm_event {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v7_cp_reginfo[] = {
     { .name = "TLBIMVAA", .cp = 15, .opc1 = 0, .crn = 8, .crm = 7, .opc2 = 3,
       .type = ARM_CP_NO_RAW, .access = PL1_W, .accessfn = access_ttlb,
       .writefn = tlbimvaa_write },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo v7mp_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v7mp_cp_reginfo[] = {
     { .name = "TLBIMVAAIS", .cp = 15, .opc1 = 0, .crn = 8, .crm = 3, .opc2 = 3,
       .type = ARM_CP_NO_RAW, .access = PL1_W, .accessfn = access_ttlb,
       .writefn = tlbimvaa_is_write },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo pmovsset_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo pmovsset_cp_reginfo[] = {
       .fieldoffset = offsetof(CPUARMState, cp15.c9_pmovsr),
       .writefn = pmovsset_write,
       .raw_writefn = raw_write },
-    REGINFO_SENTINEL
 };
 
 static void teecr_write(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo t2ee_cp_reginfo[] = {
     { .name = "TEEHBR", .cp = 14, .crn = 1, .crm = 0, .opc1 = 6, .opc2 = 0,
       .access = PL0_RW, .fieldoffset = offsetof(CPUARMState, teehbr),
       .accessfn = teehbr_access, .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo v6k_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v6k_cp_reginfo[] = {
       .bank_fieldoffsets = { offsetoflow32(CPUARMState, cp15.tpidrprw_s),
                              offsetoflow32(CPUARMState, cp15.tpidrprw_ns) },
       .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 #ifndef CONFIG_USER_ONLY
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo generic_timer_cp_reginfo[] = {
       .fieldoffset = offsetof(CPUARMState, cp15.c14_timer[GTIMER_SEC].cval),
       .writefn = gt_sec_cval_write, .raw_writefn = raw_write,
     },
-    REGINFO_SENTINEL
 };
 
 static CPAccessResult e2h_access(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo generic_timer_cp_reginfo[] = {
       .access = PL0_R, .type = ARM_CP_NO_RAW | ARM_CP_IO,
       .readfn = gt_virt_cnt_read,
     },
-    REGINFO_SENTINEL
 };
 
 #endif
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo vapa_cp_reginfo[] = {
       .access = PL1_W, .accessfn = ats_access,
       .writefn = ats_write, .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC },
 #endif
-    REGINFO_SENTINEL
 };
 
 /* Return basic MPU access permission bits.  */
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo pmsav7_cp_reginfo[] = {
       .fieldoffset = offsetof(CPUARMState, pmsav7.rnr[M_REG_NS]),
       .writefn = pmsav7_rgnr_write,
       .resetfn = arm_cp_reset_ignore },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo pmsav5_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo pmsav5_cp_reginfo[] = {
     { .name = "946_PRBS7", .cp = 15, .crn = 6, .crm = 7, .opc1 = 0,
       .opc2 = CP_ANY, .access = PL1_RW, .resetvalue = 0,
       .fieldoffset = offsetof(CPUARMState, cp15.c6_region[7]) },
-    REGINFO_SENTINEL
 };
 
 static void vmsa_ttbcr_raw_write(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo vmsa_pmsa_cp_reginfo[] = {
       .access = PL1_RW, .accessfn = access_tvm_trvm,
       .fieldoffset = offsetof(CPUARMState, cp15.far_el[1]),
       .resetvalue = 0, },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo vmsa_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo vmsa_cp_reginfo[] = {
       /* No offsetoflow32 -- pass the entire TCR to writefn/raw_writefn. */
       .bank_fieldoffsets = { offsetof(CPUARMState, cp15.tcr_el[3]),
                              offsetof(CPUARMState, cp15.tcr_el[1])} },
-    REGINFO_SENTINEL
 };
 
 /* Note that unlike TTBCR, writing to TTBCR2 does not require flushing
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo omap_cp_reginfo[] = {
     { .name = "C9", .cp = 15, .crn = 9,
       .crm = CP_ANY, .opc1 = CP_ANY, .opc2 = CP_ANY, .access = PL1_RW,
       .type = ARM_CP_CONST | ARM_CP_OVERRIDE, .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 static void xscale_cpar_write(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo xscale_cp_reginfo[] = {
     { .name = "XSCALE_UNLOCK_DCACHE",
       .cp = 15, .opc1 = 0, .crn = 9, .crm = 2, .opc2 = 1,
       .access = PL1_W, .type = ARM_CP_NOP },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo dummy_c15_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo dummy_c15_cp_reginfo[] = {
       .access = PL1_RW,
       .type = ARM_CP_CONST | ARM_CP_NO_RAW | ARM_CP_OVERRIDE,
       .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo cache_dirty_status_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cache_dirty_status_cp_reginfo[] = {
     { .name = "CDSR", .cp = 15, .crn = 7, .crm = 10, .opc1 = 0, .opc2 = 6,
       .access = PL1_R, .type = ARM_CP_CONST | ARM_CP_NO_RAW,
       .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo cache_block_ops_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cache_block_ops_cp_reginfo[] = {
       .access = PL0_W, .type = ARM_CP_NOP|ARM_CP_64BIT },
     { .name = "CIDCR", .cp = 15, .crm = 14, .opc1 = 0,
       .access = PL1_W, .type = ARM_CP_NOP|ARM_CP_64BIT },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo cache_test_clean_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo cache_test_clean_cp_reginfo[] = {
     { .name = "TCI_DCACHE", .cp = 15, .crn = 7, .crm = 14, .opc1 = 0, .opc2 = 3,
       .access = PL0_R, .type = ARM_CP_CONST | ARM_CP_NO_RAW,
       .resetvalue = (1 << 30) },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo strongarm_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo strongarm_cp_reginfo[] = {
       .crm = CP_ANY, .opc1 = CP_ANY, .opc2 = CP_ANY,
       .access = PL1_RW, .resetvalue = 0,
       .type = ARM_CP_CONST | ARM_CP_OVERRIDE | ARM_CP_NO_RAW },
-    REGINFO_SENTINEL
 };
 
 static uint64_t midr_read(CPUARMState *env, const ARMCPRegInfo *ri)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo lpae_cp_reginfo[] = {
       .bank_fieldoffsets = { offsetof(CPUARMState, cp15.ttbr1_s),
                              offsetof(CPUARMState, cp15.ttbr1_ns) },
       .writefn = vmsa_ttbr_write, },
-    REGINFO_SENTINEL
 };
 
 static uint64_t aa64_fpcr_read(CPUARMState *env, const ARMCPRegInfo *ri)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v8_cp_reginfo[] = {
       .access = PL1_RW, .accessfn = access_trap_aa32s_el1,
       .writefn = sdcr_write,
       .fieldoffset = offsetoflow32(CPUARMState, cp15.mdcr_el3) },
-    REGINFO_SENTINEL
 };
 
 /* Used to describe the behaviour of EL2 regs when EL2 does not exist.  */
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el3_no_el2_cp_reginfo[] = {
       .type = ARM_CP_CONST,
       .cp = 15, .opc1 = 4, .crn = 6, .crm = 0, .opc2 = 2,
       .access = PL2_RW, .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 /* Ditto, but for registers which exist in ARMv8 but not v7 */
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el3_no_el2_v8_cp_reginfo[] = {
       .cp = 15, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 4,
       .access = PL2_RW,
       .type = ARM_CP_CONST, .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 static void do_hcr_write(CPUARMState *env, uint64_t value, uint64_t valid_mask)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el2_cp_reginfo[] = {
       .cp = 15, .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 3,
       .access = PL2_RW,
       .fieldoffset = offsetof(CPUARMState, cp15.hstr_el2) },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo el2_v8_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el2_v8_cp_reginfo[] = {
       .access = PL2_RW,
       .fieldoffset = offsetofhigh32(CPUARMState, cp15.hcr_el2),
       .writefn = hcr_writehigh },
-    REGINFO_SENTINEL
 };
 
 static CPAccessResult sel2_access(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el2_sec_cp_reginfo[] = {
       .opc0 = 3, .opc1 = 4, .crn = 2, .crm = 6, .opc2 = 2,
       .access = PL2_RW, .accessfn = sel2_access,
       .fieldoffset = offsetof(CPUARMState, cp15.vstcr_el2) },
-    REGINFO_SENTINEL
 };
 
 static CPAccessResult nsacr_access(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el3_cp_reginfo[] = {
       .opc0 = 1, .opc1 = 6, .crn = 8, .crm = 7, .opc2 = 5,
       .access = PL3_W, .type = ARM_CP_NO_RAW,
       .writefn = tlbi_aa64_vae3_write },
-    REGINFO_SENTINEL
 };
 
 #ifndef CONFIG_USER_ONLY
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo debug_cp_reginfo[] = {
       .cp = 14, .opc0 = 2, .opc1 = 0, .crn = 0, .crm = 2, .opc2 = 0,
       .access = PL1_RW, .accessfn = access_tda,
       .type = ARM_CP_NOP },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo debug_lpae_cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo debug_lpae_cp_reginfo[] = {
       .access = PL0_R, .type = ARM_CP_CONST|ARM_CP_64BIT, .resetvalue = 0 },
     { .name = "DBGDSAR", .cp = 14, .crm = 2, .opc1 = 0,
       .access = PL0_R, .type = ARM_CP_CONST|ARM_CP_64BIT, .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 /* Return the exception level to which exceptions should be taken
@@ -XXX,XX +XXX,XX @@ static void define_debug_regs(ARMCPU *cpu)
               .fieldoffset = offsetof(CPUARMState, cp15.dbgbcr[i]),
               .writefn = dbgbcr_write, .raw_writefn = raw_write
             },
-            REGINFO_SENTINEL
         };
         define_arm_cp_regs(cpu, dbgregs);
     }
@@ -XXX,XX +XXX,XX @@ static void define_debug_regs(ARMCPU *cpu)
               .fieldoffset = offsetof(CPUARMState, cp15.dbgwcr[i]),
               .writefn = dbgwcr_write, .raw_writefn = raw_write
             },
-            REGINFO_SENTINEL
         };
         define_arm_cp_regs(cpu, dbgregs);
     }
@@ -XXX,XX +XXX,XX @@ static void define_pmu_regs(ARMCPU *cpu)
               .type = ARM_CP_IO,
               .readfn = pmevtyper_readfn, .writefn = pmevtyper_writefn,
               .raw_writefn = pmevtyper_rawwrite },
-            REGINFO_SENTINEL
         };
         define_arm_cp_regs(cpu, pmev_regs);
         g_free(pmevcntr_name);
@@ -XXX,XX +XXX,XX @@ static void define_pmu_regs(ARMCPU *cpu)
               .cp = 15, .opc1 = 0, .crn = 9, .crm = 14, .opc2 = 5,
               .access = PL0_R, .accessfn = pmreg_access, .type = ARM_CP_CONST,
               .resetvalue = extract64(cpu->pmceid1, 32, 32) },
-            REGINFO_SENTINEL
         };
         define_arm_cp_regs(cpu, v81_pmu_regs);
     }
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo lor_reginfo[] = {
       .opc0 = 3, .opc1 = 0, .crn = 10, .crm = 4, .opc2 = 7,
       .access = PL1_R, .accessfn = access_lor_ns,
       .type = ARM_CP_CONST, .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 #ifdef TARGET_AARCH64
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo pauth_reginfo[] = {
       .opc0 = 3, .opc1 = 0, .crn = 2, .crm = 1, .opc2 = 3,
       .access = PL1_RW, .accessfn = access_pauth,
       .fieldoffset = offsetof(CPUARMState, keys.apib.hi) },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo tlbirange_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo tlbirange_reginfo[] = {
       .opc0 = 1, .opc1 = 6, .crn = 8, .crm = 6, .opc2 = 5,
       .access = PL3_W, .type = ARM_CP_NO_RAW,
       .writefn = tlbi_aa64_rvae3_write },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo tlbios_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo tlbios_reginfo[] = {
       .opc0 = 1, .opc1 = 6, .crn = 8, .crm = 1, .opc2 = 5,
       .access = PL3_W, .type = ARM_CP_NO_RAW,
       .writefn = tlbi_aa64_vae3is_write },
-    REGINFO_SENTINEL
 };
 
 static uint64_t rndr_readfn(CPUARMState *env, const ARMCPRegInfo *ri)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo rndr_reginfo[] = {
       .type = ARM_CP_NO_RAW | ARM_CP_SUPPRESS_TB_END | ARM_CP_IO,
       .opc0 = 3, .opc1 = 3, .crn = 2, .crm = 4, .opc2 = 1,
       .access = PL0_R, .readfn = rndr_readfn },
-    REGINFO_SENTINEL
 };
 
 #ifndef CONFIG_USER_ONLY
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo dcpop_reg[] = {
       .opc0 = 1, .opc1 = 3, .crn = 7, .crm = 12, .opc2 = 1,
       .access = PL0_W, .type = ARM_CP_NO_RAW | ARM_CP_SUPPRESS_TB_END,
       .accessfn = aa64_cacheop_poc_access, .writefn = dccvap_writefn },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo dcpodp_reg[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo dcpodp_reg[] = {
       .opc0 = 1, .opc1 = 3, .crn = 7, .crm = 13, .opc2 = 1,
       .access = PL0_W, .type = ARM_CP_NO_RAW | ARM_CP_SUPPRESS_TB_END,
       .accessfn = aa64_cacheop_poc_access, .writefn = dccvap_writefn },
-    REGINFO_SENTINEL
 };
 #endif /*CONFIG_USER_ONLY*/
 
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo mte_reginfo[] = {
     { .name = "DC_CIGDSW", .state = ARM_CP_STATE_AA64,
       .opc0 = 1, .opc1 = 0, .crn = 7, .crm = 14, .opc2 = 6,
       .type = ARM_CP_NOP, .access = PL1_W, .accessfn = access_tsw },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo mte_tco_ro_reginfo[] = {
     { .name = "TCO", .state = ARM_CP_STATE_AA64,
       .opc0 = 3, .opc1 = 3, .crn = 4, .crm = 2, .opc2 = 7,
       .type = ARM_CP_CONST, .access = PL0_RW, },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo mte_el0_cacheop_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo mte_el0_cacheop_reginfo[] = {
       .accessfn = aa64_zva_access,
 #endif
     },
-    REGINFO_SENTINEL
 };
 
 #endif
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo predinv_reginfo[] = {
     { .name = "CPPRCTX", .state = ARM_CP_STATE_AA32,
       .cp = 15, .opc1 = 0, .crn = 7, .crm = 3, .opc2 = 7,
       .type = ARM_CP_NOP, .access = PL0_W, .accessfn = access_predinv },
-    REGINFO_SENTINEL
 };
 
 static uint64_t ccsidr2_read(CPUARMState *env, const ARMCPRegInfo *ri)
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo ccsidr2_reginfo[] = {
       .access = PL1_R,
       .accessfn = access_aa64_tid2,
       .readfn = ccsidr2_read, .type = ARM_CP_NO_RAW },
-    REGINFO_SENTINEL
 };
 
 static CPAccessResult access_aa64_tid3(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo jazelle_regs[] = {
       .cp = 14, .crn = 2, .crm = 0, .opc1 = 7, .opc2 = 0,
       .accessfn = access_joscr_jmcr,
       .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo vhe_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo vhe_reginfo[] = {
       .access = PL2_RW, .accessfn = e2h_access,
       .writefn = gt_virt_cval_write, .raw_writefn = raw_write },
 #endif
-    REGINFO_SENTINEL
 };
 
 #ifndef CONFIG_USER_ONLY
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo ats1e1_reginfo[] = {
       .opc0 = 1, .opc1 = 0, .crn = 7, .crm = 9, .opc2 = 1,
       .access = PL1_W, .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC,
       .writefn = ats_write64 },
-    REGINFO_SENTINEL
 };
 
 static const ARMCPRegInfo ats1cp_reginfo[] = {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo ats1cp_reginfo[] = {
       .cp = 15, .opc1 = 0, .crn = 7, .crm = 9, .opc2 = 1,
       .access = PL1_W, .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC,
       .writefn = ats_write },
-    REGINFO_SENTINEL
 };
 #endif
 
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo actlr2_hactlr2_reginfo[] = {
       .cp = 15, .opc1 = 4, .crn = 1, .crm = 0, .opc2 = 3,
       .access = PL2_RW, .type = ARM_CP_CONST,
       .resetvalue = 0 },
-    REGINFO_SENTINEL
 };
 
 void register_cp_regs_for_features(ARMCPU *cpu)
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
               .access = PL1_R, .type = ARM_CP_CONST,
               .accessfn = access_aa32_tid3,
               .resetvalue = cpu->isar.id_isar6 },
-            REGINFO_SENTINEL
         };
         define_arm_cp_regs(cpu, v6_idregs);
         define_arm_cp_regs(cpu, v6_cp_reginfo);
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
               .opc0 = 3, .opc1 = 3, .crn = 9, .crm = 12, .opc2 = 7,
               .access = PL0_R, .accessfn = pmreg_access, .type = ARM_CP_CONST,
               .resetvalue = cpu->pmceid1 },
-            REGINFO_SENTINEL
         };
 #ifdef CONFIG_USER_ONLY
         ARMCPRegUserSpaceInfo v8_user_idregs[] = {
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
               .exported_bits = 0x000000f0ffffffff },
             { .name = "ID_AA64ISAR*_EL1_RESERVED",
               .is_glob = true                     },
-            REGUSERINFO_SENTINEL
         };
         modify_arm_cp_regs(v8_idregs, v8_user_idregs);
 #endif
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
               .access = PL2_RW,
               .resetvalue = vmpidr_def,
               .fieldoffset = offsetof(CPUARMState, cp15.vmpidr_el2) },
-            REGINFO_SENTINEL
         };
         define_arm_cp_regs(cpu, vpidr_regs);
         define_arm_cp_regs(cpu, el2_cp_reginfo);
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                   .access = PL2_RW, .accessfn = access_el3_aa32ns,
                   .type = ARM_CP_NO_RAW,
                   .writefn = arm_cp_write_ignore, .readfn = mpidr_read },
-                REGINFO_SENTINEL
             };
             define_arm_cp_regs(cpu, vpidr_regs);
             define_arm_cp_regs(cpu, el3_no_el2_cp_reginfo);
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
               .raw_writefn = raw_write, .writefn = sctlr_write,
               .fieldoffset = offsetof(CPUARMState, cp15.sctlr_el[3]),
               .resetvalue = cpu->reset_sctlr },
-            REGINFO_SENTINEL
         };
 
         define_arm_cp_regs(cpu, el3_regs);
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
             { .name = "DUMMY",
               .cp = 15, .crn = 0, .crm = 7, .opc1 = 0, .opc2 = CP_ANY,
               .access = PL1_R, .type = ARM_CP_CONST, .resetvalue = 0 },
-            REGINFO_SENTINEL
         };
         ARMCPRegInfo id_v8_midr_cp_reginfo[] = {
             { .name = "MIDR_EL1", .state = ARM_CP_STATE_BOTH,
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
               .access = PL1_R,
               .accessfn = access_aa64_tid1,
               .type = ARM_CP_CONST, .resetvalue = cpu->revidr },
-            REGINFO_SENTINEL
         };
         ARMCPRegInfo id_cp_reginfo[] = {
             /* These are common to v8 and pre-v8 */
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
               .access = PL1_R,
               .accessfn = access_aa32_tid1,
               .type = ARM_CP_CONST, .resetvalue = 0 },
-            REGINFO_SENTINEL
         };
         /* TLBTR is specific to VMSA */
         ARMCPRegInfo id_tlbtr_reginfo = {
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
             { .name = "MIDR_EL1",
               .exported_bits = 0x00000000ffffffff },
             { .name = "REVIDR_EL1"                },
-            REGUSERINFO_SENTINEL
         };
         modify_arm_cp_regs(id_v8_midr_cp_reginfo, id_v8_user_midr_cp_reginfo);
 #endif
         if (arm_feature(env, ARM_FEATURE_OMAPCP) ||
             arm_feature(env, ARM_FEATURE_STRONGARM)) {
-            ARMCPRegInfo *r;
+            size_t i;
             /* Register the blanket "writes ignored" value first to cover the
              * whole space. Then update the specific ID registers to allow write
              * access, so that they ignore writes rather than causing them to
              * UNDEF.
              */
             define_one_arm_cp_reg(cpu, &crn0_wi_reginfo);
-            for (r = id_pre_v8_midr_cp_reginfo;
-                 r->type != ARM_CP_SENTINEL; r++) {
-                r->access = PL1_RW;
+            for (i = 0; i < ARRAY_SIZE(id_pre_v8_midr_cp_reginfo); ++i) {
+                id_pre_v8_midr_cp_reginfo[i].access = PL1_RW;
             }
-            for (r = id_cp_reginfo; r->type != ARM_CP_SENTINEL; r++) {
-                r->access = PL1_RW;
+            for (i = 0; i < ARRAY_SIZE(id_cp_reginfo); ++i) {
+                id_cp_reginfo[i].access = PL1_RW;
             }
             id_mpuir_reginfo.access = PL1_RW;
             id_tlbtr_reginfo.access = PL1_RW;
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
             { .name = "MPIDR_EL1", .state = ARM_CP_STATE_BOTH,
               .opc0 = 3, .crn = 0, .crm = 0, .opc1 = 0, .opc2 = 5,
               .access = PL1_R, .readfn = mpidr_read, .type = ARM_CP_NO_RAW },
-            REGINFO_SENTINEL
         };
 #ifdef CONFIG_USER_ONLY
         ARMCPRegUserSpaceInfo mpidr_user_cp_reginfo[] = {
             { .name = "MPIDR_EL1",
               .fixed_bits = 0x0000000080000000 },
-            REGUSERINFO_SENTINEL
         };
         modify_arm_cp_regs(mpidr_cp_reginfo, mpidr_user_cp_reginfo);
 #endif
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
               .opc0 = 3, .opc1 = 6, .crn = 1, .crm = 0, .opc2 = 1,
               .access = PL3_RW, .type = ARM_CP_CONST,
               .resetvalue = 0 },
-            REGINFO_SENTINEL
         };
         define_arm_cp_regs(cpu, auxcr_reginfo);
         if (cpu_isar_feature(aa32_ac2, cpu)) {
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
                   .type = ARM_CP_CONST,
                   .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 3, .opc2 = 0,
                   .access = PL1_R, .resetvalue = cpu->reset_cbar },
-                REGINFO_SENTINEL
             };
             /* We don't implement a r/w 64 bit CBAR currently */
             assert(arm_feature(env, ARM_FEATURE_CBAR_RO));
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
               .bank_fieldoffsets = { offsetof(CPUARMState, cp15.vbar_s),
                                      offsetof(CPUARMState, cp15.vbar_ns) },
               .resetvalue = 0 },
-            REGINFO_SENTINEL
         };
         define_arm_cp_regs(cpu, vbar_cp_reginfo);
     }
@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
                    r->writefn);
         }
     }
-    /* Bad type field probably means missing sentinel at end of reg list */
-    assert(cptype_valid(r->type));
+
     for (crm = crmmin; crm <= crmmax; crm++) {
         for (opc1 = opc1min; opc1 <= opc1max; opc1++) {
             for (opc2 = opc2min; opc2 <= opc2max; opc2++) {
@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
     }
 }
 
-void define_arm_cp_regs_with_opaque(ARMCPU *cpu,
-                                    const ARMCPRegInfo *regs, void *opaque)
+/* Define a whole list of registers */
+void define_arm_cp_regs_with_opaque_len(ARMCPU *cpu, const ARMCPRegInfo *regs,
+                                        void *opaque, size_t len)
 {
-    /* Define a whole list of registers */
-    const ARMCPRegInfo *r;
-    for (r = regs; r->type != ARM_CP_SENTINEL; r++) {
-        define_one_arm_cp_reg_with_opaque(cpu, r, opaque);
+    size_t i;
+    for (i = 0; i < len; ++i) {
+        define_one_arm_cp_reg_with_opaque(cpu, regs + i, opaque);
     }
 }
 
@@ -XXX,XX +XXX,XX @@ void define_arm_cp_regs_with_opaque(ARMCPU *cpu,
  * user-space cannot alter any values and dynamic values pertaining to
  * execution state are hidden from user space view anyway.
  */
-void modify_arm_cp_regs(ARMCPRegInfo *regs, const ARMCPRegUserSpaceInfo *mods)
+void modify_arm_cp_regs_with_len(ARMCPRegInfo *regs, size_t regs_len,
+                                 const ARMCPRegUserSpaceInfo *mods,
+                                 size_t mods_len)
 {
-    const ARMCPRegUserSpaceInfo *m;
-    ARMCPRegInfo *r;
-
-    for (m = mods; m->name; m++) {
+    for (size_t mi = 0; mi < mods_len; ++mi) {
+        const ARMCPRegUserSpaceInfo *m = mods + mi;
         GPatternSpec *pat = NULL;
+
         if (m->is_glob) {
             pat = g_pattern_spec_new(m->name);
         }
-        for (r = regs; r->type != ARM_CP_SENTINEL; r++) {
+        for (size_t ri = 0; ri < regs_len; ++ri) {
+            ARMCPRegInfo *r = regs + ri;
+
             if (pat && g_pattern_match_string(pat, r->name)) {
                 r->type = ARM_CP_CONST;
                 r->access = PL0U_R;
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

These particular data structures are not modified at runtime.

diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
               .resetvalue = cpu->pmceid1 },
         };
 #ifdef CONFIG_USER_ONLY
-        ARMCPRegUserSpaceInfo v8_user_idregs[] = {
+        static const ARMCPRegUserSpaceInfo v8_user_idregs[] = {
             { .name = "ID_AA64PFR0_EL1",
               .exported_bits = 0x000f000f00ff0000,
               .fixed_bits    = 0x0000000000000011 },
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
      */
     if (arm_feature(env, ARM_FEATURE_EL3)) {
         if (arm_feature(env, ARM_FEATURE_AARCH64)) {
-            ARMCPRegInfo nsacr = {
+            static const ARMCPRegInfo nsacr = {
                 .name = "NSACR", .type = ARM_CP_CONST,
                 .cp = 15, .opc1 = 0, .crn = 1, .crm = 1, .opc2 = 2,
                 .access = PL1_RW, .accessfn = nsacr_access,
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
             };
             define_one_arm_cp_reg(cpu, &nsacr);
         } else {
-            ARMCPRegInfo nsacr = {
+            static const ARMCPRegInfo nsacr = {
                 .name = "NSACR",
                 .cp = 15, .opc1 = 0, .crn = 1, .crm = 1, .opc2 = 2,
                 .access = PL3_RW | PL1_R,
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
         }
     } else {
         if (arm_feature(env, ARM_FEATURE_V8)) {
-            ARMCPRegInfo nsacr = {
+            static const ARMCPRegInfo nsacr = {
                 .name = "NSACR", .type = ARM_CP_CONST,
                 .cp = 15, .opc1 = 0, .crn = 1, .crm = 1, .opc2 = 2,
                 .access = PL1_R,
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
               .access = PL1_R, .type = ARM_CP_CONST,
               .resetvalue = cpu->pmsav7_dregion << 8
         };
-        ARMCPRegInfo crn0_wi_reginfo = {
+        static const ARMCPRegInfo crn0_wi_reginfo = {
             .name = "CRN0_WI", .cp = 15, .crn = 0, .crm = CP_ANY,
             .opc1 = CP_ANY, .opc2 = CP_ANY, .access = PL1_W,
             .type = ARM_CP_NOP | ARM_CP_OVERRIDE
         };
 #ifdef CONFIG_USER_ONLY
-        ARMCPRegUserSpaceInfo id_v8_user_midr_cp_reginfo[] = {
+        static const ARMCPRegUserSpaceInfo id_v8_user_midr_cp_reginfo[] = {
             { .name = "MIDR_EL1",
               .exported_bits = 0x00000000ffffffff },
             { .name = "REVIDR_EL1"                },
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
               .access = PL1_R, .readfn = mpidr_read, .type = ARM_CP_NO_RAW },
         };
 #ifdef CONFIG_USER_ONLY
-        ARMCPRegUserSpaceInfo mpidr_user_cp_reginfo[] = {
+        static const ARMCPRegUserSpaceInfo mpidr_user_cp_reginfo[] = {
             { .name = "MPIDR_EL1",
               .fixed_bits = 0x0000000080000000 },
         };
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
     }
 
     if (arm_feature(env, ARM_FEATURE_VBAR)) {
-        ARMCPRegInfo vbar_cp_reginfo[] = {
+        static const ARMCPRegInfo vbar_cp_reginfo[] = {
             { .name = "VBAR", .state = ARM_CP_STATE_BOTH,
               .opc0 = 3, .crn = 12, .crm = 0, .opc1 = 0, .opc2 = 0,
               .access = PL1_RW, .writefn = vbar_write,
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Instead of defining ARM_CP_FLAG_MASK to remove flags,
define ARM_CP_SPECIAL_MASK to isolate special cases.
Sort the specials to the low bits. Use an enum.

Split the large comment block so as to document each
value separately.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20220501055028.646596-6-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/cpregs.h        | 130 +++++++++++++++++++++++--------------
 target/arm/cpu.c           |   4 +-
 target/arm/helper.c        |   4 +-
 target/arm/translate-a64.c |   6 +-
 target/arm/translate.c     |   6 +-
 5 files changed, 92 insertions(+), 58 deletions(-)

diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpregs.h
+++ b/target/arm/cpregs.h
@@ -XXX,XX +XXX,XX @@
 #define TARGET_ARM_CPREGS_H
 
 /*
- * ARMCPRegInfo type field bits. If the SPECIAL bit is set this is a
- * special-behaviour cp reg and bits [11..8] indicate what behaviour
- * it has. Otherwise it is a simple cp reg, where CONST indicates that
- * TCG can assume the value to be constant (ie load at translate time)
- * and 64BIT indicates a 64 bit wide coprocessor register. SUPPRESS_TB_END
- * indicates that the TB should not be ended after a write to this register
- * (the default is that the TB ends after cp writes). OVERRIDE permits
- * a register definition to override a previous definition for the
- * same (cp, is64, crn, crm, opc1, opc2) tuple: either the new or the
- * old must have the OVERRIDE bit set.
- * ALIAS indicates that this register is an alias view of some underlying
- * state which is also visible via another register, and that the other
- * register is handling migration and reset; registers marked ALIAS will not be
- * migrated but may have their state set by syncing of register state from KVM.
- * NO_RAW indicates that this register has no underlying state and does not
- * support raw access for state saving/loading; it will not be used for either
- * migration or KVM state synchronization. (Typically this is for "registers"
- * which are actually used as instructions for cache maintenance and so on.)
- * IO indicates that this register does I/O and therefore its accesses
- * need to be marked with gen_io_start() and also end the TB. In particular,
- * registers which implement clocks or timers require this.
- * RAISES_EXC is for when the read or write hook might raise an exception;
- * the generated code will synchronize the CPU state before calling the hook
- * so that it is safe for the hook to call raise_exception().
- * NEWEL is for writes to registers that might change the exception
- * level - typically on older ARM chips. For those cases we need to
- * re-read the new el when recomputing the translation flags.
+ * ARMCPRegInfo type field bits:
  */
-#define ARM_CP_SPECIAL           0x0001
-#define ARM_CP_CONST             0x0002
-#define ARM_CP_64BIT             0x0004
-#define ARM_CP_SUPPRESS_TB_END   0x0008
-#define ARM_CP_OVERRIDE          0x0010
-#define ARM_CP_ALIAS             0x0020
-#define ARM_CP_IO                0x0040
-#define ARM_CP_NO_RAW            0x0080
-#define ARM_CP_NOP               (ARM_CP_SPECIAL | 0x0100)
-#define ARM_CP_WFI               (ARM_CP_SPECIAL | 0x0200)
-#define ARM_CP_NZCV              (ARM_CP_SPECIAL | 0x0300)
-#define ARM_CP_CURRENTEL         (ARM_CP_SPECIAL | 0x0400)
-#define ARM_CP_DC_ZVA            (ARM_CP_SPECIAL | 0x0500)
-#define ARM_CP_DC_GVA            (ARM_CP_SPECIAL | 0x0600)
-#define ARM_CP_DC_GZVA           (ARM_CP_SPECIAL | 0x0700)
-#define ARM_LAST_SPECIAL         ARM_CP_DC_GZVA
-#define ARM_CP_FPU               0x1000
-#define ARM_CP_SVE               0x2000
-#define ARM_CP_NO_GDB            0x4000
-#define ARM_CP_RAISES_EXC        0x8000
-#define ARM_CP_NEWEL             0x10000
-/* Mask of only the flag bits in a type field */
-#define ARM_CP_FLAG_MASK         0x1f0ff
+enum {
+    /*
+     * Register must be handled specially during translation.
+     * The method is one of the values below:
+     */
+    ARM_CP_SPECIAL_MASK          = 0x000f,
+    /* Special: no change to PE state: writes ignored, reads ignored. */
+    ARM_CP_NOP                   = 0x0001,
+    /* Special: sysreg is WFI, for v5 and v6. */
+    ARM_CP_WFI                   = 0x0002,
+    /* Special: sysreg is NZCV. */
+    ARM_CP_NZCV                  = 0x0003,
+    /* Special: sysreg is CURRENTEL. */
+    ARM_CP_CURRENTEL             = 0x0004,
+    /* Special: sysreg is DC ZVA or similar. */
+    ARM_CP_DC_ZVA                = 0x0005,
+    ARM_CP_DC_GVA                = 0x0006,
+    ARM_CP_DC_GZVA               = 0x0007,
+
+    /* Flag: reads produce resetvalue; writes ignored. */
+    ARM_CP_CONST                 = 1 << 4,
+    /* Flag: For ARM_CP_STATE_AA32, sysreg is 64-bit. */
+    ARM_CP_64BIT                 = 1 << 5,
+    /*
+     * Flag: TB should not be ended after a write to this register
+     * (the default is that the TB ends after cp writes).
+     */
+    ARM_CP_SUPPRESS_TB_END       = 1 << 6,
+    /*
+     * Flag: Permit a register definition to override a previous definition
+     * for the same (cp, is64, crn, crm, opc1, opc2) tuple: either the new
+     * or the old must have the ARM_CP_OVERRIDE bit set.
+     */
+    ARM_CP_OVERRIDE              = 1 << 7,
+    /*
+     * Flag: Register is an alias view of some underlying state which is also
+     * visible via another register, and that the other register is handling
+     * migration and reset; registers marked ARM_CP_ALIAS will not be migrated
+     * but may have their state set by syncing of register state from KVM.
+     */
+    ARM_CP_ALIAS                 = 1 << 8,
+    /*
+     * Flag: Register does I/O and therefore its accesses need to be marked
+     * with gen_io_start() and also end the TB. In particular, registers which
+     * implement clocks or timers require this.
+     */
+    ARM_CP_IO                    = 1 << 9,
+    /*
+     * Flag: Register has no underlying state and does not support raw access
+     * for state saving/loading; it will not be used for either migration or
+     * KVM state synchronization. Typically this is for "registers" which are
+     * actually used as instructions for cache maintenance and so on.
+     */
+    ARM_CP_NO_RAW                = 1 << 10,
+    /*
+     * Flag: The read or write hook might raise an exception; the generated
+     * code will synchronize the CPU state before calling the hook so that it
+     * is safe for the hook to call raise_exception().
+     */
+    ARM_CP_RAISES_EXC            = 1 << 11,
+    /*
+     * Flag: Writes to the sysreg might change the exception level - typically
+     * on older ARM chips. For those cases we need to re-read the new el when
+     * recomputing the translation flags.
+     */
+    ARM_CP_NEWEL                 = 1 << 12,
+    /*
+     * Flag: Access check for this sysreg is identical to accessing FPU state
+     * from an instruction: use translation fp_access_check().
+     */
+    ARM_CP_FPU                   = 1 << 13,
+    /*
+     * Flag: Access check for this sysreg is identical to accessing SVE state
+     * from an instruction: use translation sve_access_check().
+     */
+    ARM_CP_SVE                   = 1 << 14,
+    /* Flag: Do not expose in gdb sysreg xml. */
+    ARM_CP_NO_GDB                = 1 << 15,
+};
 
 /*
  * Valid values for ARMCPRegInfo state field, indicating which of
diff --git a/target/arm/cpu.c b/target/arm/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu.c
+++ b/target/arm/cpu.c
@@ -XXX,XX +XXX,XX @@ static void cp_reg_reset(gpointer key, gpointer value, gpointer opaque)
     ARMCPRegInfo *ri = value;
     ARMCPU *cpu = opaque;
 
-    if (ri->type & (ARM_CP_SPECIAL | ARM_CP_ALIAS)) {
+    if (ri->type & (ARM_CP_SPECIAL_MASK | ARM_CP_ALIAS)) {
         return;
     }
 
@@ -XXX,XX +XXX,XX @@ static void cp_reg_check_reset(gpointer key, gpointer value,  gpointer opaque)
     ARMCPU *cpu = opaque;
     uint64_t oldvalue, newvalue;
 
-    if (ri->type & (ARM_CP_SPECIAL | ARM_CP_ALIAS | ARM_CP_NO_RAW)) {
+    if (ri->type & (ARM_CP_SPECIAL_MASK | ARM_CP_ALIAS | ARM_CP_NO_RAW)) {
         return;
     }
 
diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
      * multiple times. Special registers (ie NOP/WFI) are
      * never migratable and not even raw-accessible.
      */
-    if ((r->type & ARM_CP_SPECIAL)) {
+    if (r->type & ARM_CP_SPECIAL_MASK) {
         r2->type |= ARM_CP_NO_RAW;
     }
     if (((r->crm == CP_ANY) && crm != 0) ||
@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
     /* Check that the register definition has enough info to handle
      * reads and writes if they are permitted.
      */
-    if (!(r->type & (ARM_CP_SPECIAL|ARM_CP_CONST))) {
+    if (!(r->type & (ARM_CP_SPECIAL_MASK | ARM_CP_CONST))) {
         if (r->access & PL3_R) {
             assert((r->fieldoffset ||
                    (r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1])) ||
diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/translate-a64.c
+++ b/target/arm/translate-a64.c
@@ -XXX,XX +XXX,XX @@ static void handle_sys(DisasContext *s, uint32_t insn, bool isread,
     }
 
     /* Handle special cases first */
-    switch (ri->type & ~(ARM_CP_FLAG_MASK & ~ARM_CP_SPECIAL)) {
+    switch (ri->type & ARM_CP_SPECIAL_MASK) {
+    case 0:
+        break;
     case ARM_CP_NOP:
         return;
     case ARM_CP_NZCV:
@@ -XXX,XX +XXX,XX @@ static void handle_sys(DisasContext *s, uint32_t insn, bool isread,
         }
         return;
     default:
-        break;
+        g_assert_not_reached();
     }
     if ((ri->type & ARM_CP_FPU) && !fp_access_check(s)) {
         return;
diff --git a/target/arm/translate.c b/target/arm/translate.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -XXX,XX +XXX,XX @@ static void do_coproc_insn(DisasContext *s, int cpnum, int is64,
         }
 
         /* Handle special cases first */
-        switch (ri->type & ~(ARM_CP_FLAG_MASK & ~ARM_CP_SPECIAL)) {
+        switch (ri->type & ARM_CP_SPECIAL_MASK) {
+        case 0:
+            break;
         case ARM_CP_NOP:
             return;
         case ARM_CP_WFI:
@@ -XXX,XX +XXX,XX @@ static void do_coproc_insn(DisasContext *s, int cpnum, int is64,
             s->base.is_jmp = DISAS_WFI;
             return;
         default:
-            break;
+            g_assert_not_reached();
         }
 
         if ((tb_cflags(s->base.tb) & CF_USE_ICOUNT) && (ri->type & ARM_CP_IO)) {
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Standardize on g_assert_not_reached() for "should not happen".
Retain abort() when preceeded by fprintf or error_report.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20220501055028.646596-7-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/helper.c         | 7 +++----
 target/arm/hvf/hvf.c        | 2 +-
 target/arm/kvm-stub.c       | 4 ++--
 target/arm/kvm.c            | 4 ++--
 target/arm/machine.c        | 4 ++--
 target/arm/translate-a64.c  | 4 ++--
 target/arm/translate-neon.c | 2 +-
 target/arm/translate.c      | 4 ++--
 8 files changed, 15 insertions(+), 16 deletions(-)

diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
             break;
         default:
             /* broken reginfo with out-of-range opc1 */
-            assert(false);
-            break;
+            g_assert_not_reached();
         }
         /* assert our permissions are not too lax (stricter is fine) */
         assert((r->access & ~mask) == 0);
@@ -XXX,XX +XXX,XX @@ static bool get_phys_addr_v5(CPUARMState *env, uint32_t address,
             break;
         default:
             /* Never happens, but compiler isn't smart enough to tell.  */
-            abort();
+            g_assert_not_reached();
         }
     }
     *prot = ap_to_rw_prot(env, mmu_idx, ap, domain_prot);
@@ -XXX,XX +XXX,XX @@ static bool get_phys_addr_v6(CPUARMState *env, uint32_t address,
             break;
         default:
             /* Never happens, but compiler isn't smart enough to tell.  */
-            abort();
+            g_assert_not_reached();
         }
     }
     if (domain_prot == 3) {
diff --git a/target/arm/hvf/hvf.c b/target/arm/hvf/hvf.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/hvf/hvf.c
+++ b/target/arm/hvf/hvf.c
@@ -XXX,XX +XXX,XX @@ int hvf_vcpu_exec(CPUState *cpu)
         /* we got kicked, no exit to process */
         return 0;
     default:
-        assert(0);
+        g_assert_not_reached();
     }
 
     hvf_sync_vtimer(cpu);
diff --git a/target/arm/kvm-stub.c b/target/arm/kvm-stub.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/kvm-stub.c
+++ b/target/arm/kvm-stub.c
@@ -XXX,XX +XXX,XX @@
 
 bool write_kvmstate_to_list(ARMCPU *cpu)
 {
-    abort();
+    g_assert_not_reached();
 }
 
 bool write_list_to_kvmstate(ARMCPU *cpu, int level)
 {
-    abort();
+    g_assert_not_reached();
 }
diff --git a/target/arm/kvm.c b/target/arm/kvm.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/kvm.c
+++ b/target/arm/kvm.c
@@ -XXX,XX +XXX,XX @@ bool write_kvmstate_to_list(ARMCPU *cpu)
             ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &r);
             break;
         default:
-            abort();
+            g_assert_not_reached();
         }
         if (ret) {
             ok = false;
@@ -XXX,XX +XXX,XX @@ bool write_list_to_kvmstate(ARMCPU *cpu, int level)
             r.addr = (uintptr_t)(cpu->cpreg_values + i);
             break;
         default:
-            abort();
+            g_assert_not_reached();
         }
         ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &r);
         if (ret) {
diff --git a/target/arm/machine.c b/target/arm/machine.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/machine.c
+++ b/target/arm/machine.c
@@ -XXX,XX +XXX,XX @@ static int cpu_pre_save(void *opaque)
     if (kvm_enabled()) {
         if (!write_kvmstate_to_list(cpu)) {
             /* This should never fail */
-            abort();
+            g_assert_not_reached();
         }
 
         /*
@@ -XXX,XX +XXX,XX @@ static int cpu_pre_save(void *opaque)
     } else {
         if (!write_cpustate_to_list(cpu, false)) {
             /* This should never fail. */
-            abort();
+            g_assert_not_reached();
         }
     }
 
diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/translate-a64.c
+++ b/target/arm/translate-a64.c
@@ -XXX,XX +XXX,XX @@ static void handle_fp_1src_half(DisasContext *s, int opcode, int rd, int rn)
         gen_helper_advsimd_rinth(tcg_res, tcg_op, fpst);
         break;
     default:
-        abort();
+        g_assert_not_reached();
     }
 
     write_fp_sreg(s, rd, tcg_res);
@@ -XXX,XX +XXX,XX @@ static void handle_fp_fcvt(DisasContext *s, int opcode,
         break;
     }
     default:
-        abort();
+        g_assert_not_reached();
     }
 }
 
diff --git a/target/arm/translate-neon.c b/target/arm/translate-neon.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/translate-neon.c
+++ b/target/arm/translate-neon.c
@@ -XXX,XX +XXX,XX @@ static bool trans_VLDST_single(DisasContext *s, arg_VLDST_single *a)
         }
         break;
     default:
-        abort();
+        g_assert_not_reached();
     }
     if ((vd + a->stride * (nregs - 1)) > 31) {
         /*
diff --git a/target/arm/translate.c b/target/arm/translate.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -XXX,XX +XXX,XX @@ static void gen_srs(DisasContext *s,
         offset = 4;
         break;
     default:
-        abort();
+        g_assert_not_reached();
     }
     tcg_gen_addi_i32(addr, addr, offset);
     tmp = load_reg(s, 14);
@@ -XXX,XX +XXX,XX @@ static void gen_srs(DisasContext *s,
             offset = 0;
             break;
         default:
-            abort();
+            g_assert_not_reached();
         }
         tcg_gen_addi_i32(addr, addr, offset);
         gen_helper_set_r13_banked(cpu_env, tcg_constant_i32(mode), addr);
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Create a typedef as well, and use it in ARMCPRegInfo.
This won't be perfect for debugging, but it'll nicely
display the most common cases.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220501055028.646596-8-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/cpregs.h | 44 +++++++++++++++++++++++---------------------
 target/arm/helper.c |  2 +-
 2 files changed, 24 insertions(+), 22 deletions(-)

diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpregs.h
+++ b/target/arm/cpregs.h
@@ -XXX,XX +XXX,XX @@ enum {
  * described with these bits, then use a laxer set of restrictions, and
  * do the more restrictive/complex check inside a helper function.
  */
-#define PL3_R 0x80
-#define PL3_W 0x40
-#define PL2_R (0x20 | PL3_R)
-#define PL2_W (0x10 | PL3_W)
-#define PL1_R (0x08 | PL2_R)
-#define PL1_W (0x04 | PL2_W)
-#define PL0_R (0x02 | PL1_R)
-#define PL0_W (0x01 | PL1_W)
+typedef enum {
+    PL3_R = 0x80,
+    PL3_W = 0x40,
+    PL2_R = 0x20 | PL3_R,
+    PL2_W = 0x10 | PL3_W,
+    PL1_R = 0x08 | PL2_R,
+    PL1_W = 0x04 | PL2_W,
+    PL0_R = 0x02 | PL1_R,
+    PL0_W = 0x01 | PL1_W,
 
-/*
- * For user-mode some registers are accessible to EL0 via a kernel
- * trap-and-emulate ABI. In this case we define the read permissions
- * as actually being PL0_R. However some bits of any given register
- * may still be masked.
- */
+    /*
+     * For user-mode some registers are accessible to EL0 via a kernel
+     * trap-and-emulate ABI. In this case we define the read permissions
+     * as actually being PL0_R. However some bits of any given register
+     * may still be masked.
+     */
 #ifdef CONFIG_USER_ONLY
-#define PL0U_R PL0_R
+    PL0U_R = PL0_R,
 #else
-#define PL0U_R PL1_R
+    PL0U_R = PL1_R,
 #endif
 
-#define PL3_RW (PL3_R | PL3_W)
-#define PL2_RW (PL2_R | PL2_W)
-#define PL1_RW (PL1_R | PL1_W)
-#define PL0_RW (PL0_R | PL0_W)
+    PL3_RW = PL3_R | PL3_W,
+    PL2_RW = PL2_R | PL2_W,
+    PL1_RW = PL1_R | PL1_W,
+    PL0_RW = PL0_R | PL0_W,
+} CPAccessRights;
 
 typedef enum CPAccessResult {
     /* Access is permitted */
@@ -XXX,XX +XXX,XX @@ struct ARMCPRegInfo {
     /* Register type: ARM_CP_* bits/values */
     int type;
     /* Access rights: PL*_[RW] */
-    int access;
+    CPAccessRights access;
     /* Security state: ARM_CP_SECSTATE_* bits/values */
     int secure;
     /*
diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
      * to encompass the generic architectural permission check.
      */
     if (r->state != ARM_CP_STATE_AA32) {
-        int mask = 0;
+        CPAccessRights mask;
         switch (r->opc1) {
         case 0:
             /* min_EL EL1, but some accessible to EL0 via kernel ABI */
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Give this enum a name and use in ARMCPRegInfo,
add_cpreg_to_hashtable and define_one_arm_cp_reg_with_opaque.

diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpregs.h
+++ b/target/arm/cpregs.h
@@ -XXX,XX +XXX,XX @@ enum {
  * Note that we rely on the values of these enums as we iterate through
  * the various states in some places.
  */
-enum {
+typedef enum {
     ARM_CP_STATE_AA32 = 0,
     ARM_CP_STATE_AA64 = 1,
     ARM_CP_STATE_BOTH = 2,
-};
+} CPState;
 
 /*
  * ARM CP register secure state flags.  These flags identify security state
@@ -XXX,XX +XXX,XX @@ struct ARMCPRegInfo {
     uint8_t opc1;
     uint8_t opc2;
     /* Execution state in which this register is visible: ARM_CP_STATE_* */
-    int state;
+    CPState state;
     /* Register type: ARM_CP_* bits/values */
     int type;
     /* Access rights: PL*_[RW] */
diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ CpuDefinitionInfoList *qmp_query_cpu_definitions(Error **errp)
 }
 
 static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-                                   void *opaque, int state, int secstate,
+                                   void *opaque, CPState state, int secstate,
                                    int crm, int opc1, int opc2,
                                    const char *name)
 {
@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
      * bits; the ARM_CP_64BIT* flag applies only to the AArch32 view of
      * the register, if any.
      */
-    int crm, opc1, opc2, state;
+    int crm, opc1, opc2;
     int crmmin = (r->crm == CP_ANY) ? 0 : r->crm;
     int crmmax = (r->crm == CP_ANY) ? 15 : r->crm;
     int opc1min = (r->opc1 == CP_ANY) ? 0 : r->opc1;
     int opc1max = (r->opc1 == CP_ANY) ? 7 : r->opc1;
     int opc2min = (r->opc2 == CP_ANY) ? 0 : r->opc2;
     int opc2max = (r->opc2 == CP_ANY) ? 7 : r->opc2;
+    CPState state;
+
     /* 64 bit registers have only CRm and Opc1 fields */
     assert(!((r->type & ARM_CP_64BIT) && (r->opc2 || r->crn)));
     /* op0 only exists in the AArch64 encodings */
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Give this enum a name and use in ARMCPRegInfo and add_cpreg_to_hashtable.
Add the enumerator ARM_CP_SECSTATE_BOTH to clarify how 0
is handled in define_one_arm_cp_reg_with_opaque.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220501055028.646596-10-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/cpregs.h | 7 ++++---
 target/arm/helper.c | 7 +++++--
 2 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/target/arm/cpregs.h b/target/arm/cpregs.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpregs.h
+++ b/target/arm/cpregs.h
@@ -XXX,XX +XXX,XX @@ typedef enum {
  * registered entry will only have one to identify whether the entry is secure
  * or non-secure.
  */
-enum {
+typedef enum {
+    ARM_CP_SECSTATE_BOTH = 0,       /* define one cpreg for each secstate */
     ARM_CP_SECSTATE_S =   (1 << 0), /* bit[0]: Secure state register */
     ARM_CP_SECSTATE_NS =  (1 << 1), /* bit[1]: Non-secure state register */
-};
+} CPSecureState;
 
 /*
  * Access rights:
@@ -XXX,XX +XXX,XX @@ struct ARMCPRegInfo {
     /* Access rights: PL*_[RW] */
     CPAccessRights access;
     /* Security state: ARM_CP_SECSTATE_* bits/values */
-    int secure;
+    CPSecureState secure;
     /*
      * The opaque pointer passed to define_arm_cp_regs_with_opaque() when
      * this register was defined: can be used to hand data through to the
diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ CpuDefinitionInfoList *qmp_query_cpu_definitions(Error **errp)
 }
 
 static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
-                                   void *opaque, CPState state, int secstate,
+                                   void *opaque, CPState state,
+                                   CPSecureState secstate,
                                    int crm, int opc1, int opc2,
                                    const char *name)
 {
@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
                                                    r->secure, crm, opc1, opc2,
                                                    r->name);
                             break;
-                        default:
+                        case ARM_CP_SECSTATE_BOTH:
                             name = g_strdup_printf("%s_S", r->name);
                             add_cpreg_to_hashtable(cpu, r, opaque, state,
                                                    ARM_CP_SECSTATE_S,
@@ -XXX,XX +XXX,XX @@ void define_one_arm_cp_reg_with_opaque(ARMCPU *cpu,
                                                    ARM_CP_SECSTATE_NS,
                                                    crm, opc1, opc2, r->name);
                             break;
+                        default:
+                            g_assert_not_reached();
                         }
                     } else {
                         /* AArch64 registers get mapped to non-secure instance
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

The new_key field is always non-zero -- drop the if.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20220501055028.646596-11-richard.henderson@linaro.org
[PMM: reinstated dropped PL3_RW mask]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/helper.c | 23 +++++++++++------------
 1 file changed, 11 insertions(+), 12 deletions(-)

diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static void define_arm_vh_e2h_redirects_aliases(ARMCPU *cpu)
 
     for (i = 0; i < ARRAY_SIZE(aliases); i++) {
         const struct E2HAlias *a = &aliases[i];
-        ARMCPRegInfo *src_reg, *dst_reg;
+        ARMCPRegInfo *src_reg, *dst_reg, *new_reg;
+        uint32_t *new_key;
+        bool ok;
 
         if (a->feature && !a->feature(&cpu->isar)) {
             continue;
@@ -XXX,XX +XXX,XX @@ static void define_arm_vh_e2h_redirects_aliases(ARMCPU *cpu)
         g_assert(src_reg->opaque == NULL);
 
         /* Create alias before redirection so we dup the right data. */
-        if (a->new_key) {
-            ARMCPRegInfo *new_reg = g_memdup(src_reg, sizeof(ARMCPRegInfo));
-            uint32_t *new_key = g_memdup(&a->new_key, sizeof(uint32_t));
-            bool ok;
+        new_reg = g_memdup(src_reg, sizeof(ARMCPRegInfo));
+        new_key = g_memdup(&a->new_key, sizeof(uint32_t));
 
-            new_reg->name = a->new_name;
-            new_reg->type |= ARM_CP_ALIAS;
-            /* Remove PL1/PL0 access, leaving PL2/PL3 R/W in place.  */
-            new_reg->access &= PL2_RW | PL3_RW;
+        new_reg->name = a->new_name;
+        new_reg->type |= ARM_CP_ALIAS;
+        /* Remove PL1/PL0 access, leaving PL2/PL3 R/W in place.  */
+        new_reg->access &= PL2_RW | PL3_RW;
 
-            ok = g_hash_table_insert(cpu->cp_regs, new_key, new_reg);
-            g_assert(ok);
-        }
+        ok = g_hash_table_insert(cpu->cp_regs, new_key, new_reg);
+        g_assert(ok);
 
         src_reg->opaque = dst_reg;
         src_reg->orig_readfn = src_reg->readfn ?: raw_read;
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Cast the uint32_t key into a gpointer directly, which
allows us to avoid allocating storage for each key.

Use g_hash_table_lookup when we already have a gpointer
(e.g. for callbacks like count_cpreg), or when using
get_arm_cp_reginfo would require casting away const.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20220501055028.646596-12-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/cpu.c     |  4 ++--
 target/arm/gdbstub.c |  2 +-
 target/arm/helper.c  | 41 ++++++++++++++++++-----------------------
 3 files changed, 21 insertions(+), 26 deletions(-)

diff --git a/target/arm/cpu.c b/target/arm/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu.c
+++ b/target/arm/cpu.c
@@ -XXX,XX +XXX,XX @@ static void arm_cpu_initfn(Object *obj)
     ARMCPU *cpu = ARM_CPU(obj);
 
     cpu_set_cpustate_pointers(cpu);
-    cpu->cp_regs = g_hash_table_new_full(g_int_hash, g_int_equal,
-                                         g_free, cpreg_hashtable_data_destroy);
+    cpu->cp_regs = g_hash_table_new_full(g_direct_hash, g_direct_equal,
+                                         NULL, cpreg_hashtable_data_destroy);
 
     QLIST_INIT(&cpu->pre_el_change_hooks);
     QLIST_INIT(&cpu->el_change_hooks);
diff --git a/target/arm/gdbstub.c b/target/arm/gdbstub.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/gdbstub.c
+++ b/target/arm/gdbstub.c
@@ -XXX,XX +XXX,XX @@ static void arm_gen_one_xml_sysreg_tag(GString *s, DynamicGDBXMLInfo *dyn_xml,
 static void arm_register_sysreg_for_xml(gpointer key, gpointer value,
                                         gpointer p)
 {
-    uint32_t ri_key = *(uint32_t *)key;
+    uint32_t ri_key = (uintptr_t)key;
     ARMCPRegInfo *ri = value;
     RegisterSysregXmlParam *param = (RegisterSysregXmlParam *)p;
     GString *s = param->s;
diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ bool write_list_to_cpustate(ARMCPU *cpu)
 static void add_cpreg_to_list(gpointer key, gpointer opaque)
 {
     ARMCPU *cpu = opaque;
-    uint64_t regidx;
-    const ARMCPRegInfo *ri;
-
-    regidx = *(uint32_t *)key;
-    ri = get_arm_cp_reginfo(cpu->cp_regs, regidx);
+    uint32_t regidx = (uintptr_t)key;
+    const ARMCPRegInfo *ri = get_arm_cp_reginfo(cpu->cp_regs, regidx);
 
     if (!(ri->type & (ARM_CP_NO_RAW|ARM_CP_ALIAS))) {
         cpu->cpreg_indexes[cpu->cpreg_array_len] = cpreg_to_kvm_id(regidx);
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_list(gpointer key, gpointer opaque)
 static void count_cpreg(gpointer key, gpointer opaque)
 {
     ARMCPU *cpu = opaque;
-    uint64_t regidx;
     const ARMCPRegInfo *ri;
 
-    regidx = *(uint32_t *)key;
-    ri = get_arm_cp_reginfo(cpu->cp_regs, regidx);
+    ri = g_hash_table_lookup(cpu->cp_regs, key);
 
     if (!(ri->type & (ARM_CP_NO_RAW|ARM_CP_ALIAS))) {
         cpu->cpreg_array_len++;
@@ -XXX,XX +XXX,XX @@ static void count_cpreg(gpointer key, gpointer opaque)
 
 static gint cpreg_key_compare(gconstpointer a, gconstpointer b)
 {
-    uint64_t aidx = cpreg_to_kvm_id(*(uint32_t *)a);
-    uint64_t bidx = cpreg_to_kvm_id(*(uint32_t *)b);
+    uint64_t aidx = cpreg_to_kvm_id((uintptr_t)a);
+    uint64_t bidx = cpreg_to_kvm_id((uintptr_t)b);
 
     if (aidx > bidx) {
         return 1;
@@ -XXX,XX +XXX,XX @@ static void define_arm_vh_e2h_redirects_aliases(ARMCPU *cpu)
     for (i = 0; i < ARRAY_SIZE(aliases); i++) {
         const struct E2HAlias *a = &aliases[i];
         ARMCPRegInfo *src_reg, *dst_reg, *new_reg;
-        uint32_t *new_key;
         bool ok;
 
         if (a->feature && !a->feature(&cpu->isar)) {
             continue;
         }
 
-        src_reg = g_hash_table_lookup(cpu->cp_regs, &a->src_key);
-        dst_reg = g_hash_table_lookup(cpu->cp_regs, &a->dst_key);
+        src_reg = g_hash_table_lookup(cpu->cp_regs,
+                                      (gpointer)(uintptr_t)a->src_key);
+        dst_reg = g_hash_table_lookup(cpu->cp_regs,
+                                      (gpointer)(uintptr_t)a->dst_key);
         g_assert(src_reg != NULL);
         g_assert(dst_reg != NULL);
 
@@ -XXX,XX +XXX,XX @@ static void define_arm_vh_e2h_redirects_aliases(ARMCPU *cpu)
 
         /* Create alias before redirection so we dup the right data. */
         new_reg = g_memdup(src_reg, sizeof(ARMCPRegInfo));
-        new_key = g_memdup(&a->new_key, sizeof(uint32_t));
 
         new_reg->name = a->new_name;
         new_reg->type |= ARM_CP_ALIAS;
         /* Remove PL1/PL0 access, leaving PL2/PL3 R/W in place.  */
         new_reg->access &= PL2_RW | PL3_RW;
 
-        ok = g_hash_table_insert(cpu->cp_regs, new_key, new_reg);
+        ok = g_hash_table_insert(cpu->cp_regs,
+                                 (gpointer)(uintptr_t)a->new_key, new_reg);
         g_assert(ok);
 
         src_reg->opaque = dst_reg;
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
     /* Private utility function for define_one_arm_cp_reg_with_opaque():
      * add a single reginfo struct to the hash table.
      */
-    uint32_t *key = g_new(uint32_t, 1);
+    uint32_t key;
     ARMCPRegInfo *r2 = g_memdup(r, sizeof(ARMCPRegInfo));
     int is64 = (r->type & ARM_CP_64BIT) ? 1 : 0;
     int ns = (secstate & ARM_CP_SECSTATE_NS) ? 1 : 0;
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
         if (r->cp == 0 || r->state == ARM_CP_STATE_BOTH) {
             r2->cp = CP_REG_ARM64_SYSREG_CP;
         }
-        *key = ENCODE_AA64_CP_REG(r2->cp, r2->crn, crm,
-                                  r2->opc0, opc1, opc2);
+        key = ENCODE_AA64_CP_REG(r2->cp, r2->crn, crm,
+                                 r2->opc0, opc1, opc2);
     } else {
-        *key = ENCODE_CP_REG(r2->cp, is64, ns, r2->crn, crm, opc1, opc2);
+        key = ENCODE_CP_REG(r2->cp, is64, ns, r2->crn, crm, opc1, opc2);
     }
     if (opaque) {
         r2->opaque = opaque;
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
      * requested.
      */
     if (!(r->type & ARM_CP_OVERRIDE)) {
-        ARMCPRegInfo *oldreg;
-        oldreg = g_hash_table_lookup(cpu->cp_regs, key);
+        const ARMCPRegInfo *oldreg = get_arm_cp_reginfo(cpu->cp_regs, key);
         if (oldreg && !(oldreg->type & ARM_CP_OVERRIDE)) {
             fprintf(stderr, "Register redefined: cp=%d %d bit "
                     "crn=%d crm=%d opc1=%d opc2=%d, "
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
             g_assert_not_reached();
         }
     }
-    g_hash_table_insert(cpu->cp_regs, key, r2);
+    g_hash_table_insert(cpu->cp_regs, (gpointer)(uintptr_t)key, r2);
 }
 
 
@@ -XXX,XX +XXX,XX @@ void modify_arm_cp_regs_with_len(ARMCPRegInfo *regs, size_t regs_len,
 
 const ARMCPRegInfo *get_arm_cp_reginfo(GHashTable *cpregs, uint32_t encoded_cp)
 {
-    return g_hash_table_lookup(cpregs, &encoded_cp);
+    return g_hash_table_lookup(cpregs, (gpointer)(uintptr_t)encoded_cp);
 }
 
 void arm_cp_write_ignore(CPUARMState *env, const ARMCPRegInfo *ri,
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Simplify freeing cp_regs hash table entries by using a single
allocation for the entire value.

This fixes a theoretical bug if we were to ever free the entire
hash table, because we've been installing string literal constants
into the cpreg structure in define_arm_vh_e2h_redirects_aliases.
However, at present we only free entries created for AArch32
wildcard cpregs which get overwritten by more specific cpregs,
so this bug is never exposed.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20220501055028.646596-13-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/cpu.c    | 16 +---------------
 target/arm/helper.c | 10 ++++++++--
 2 files changed, 9 insertions(+), 17 deletions(-)

diff --git a/target/arm/cpu.c b/target/arm/cpu.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu.c
+++ b/target/arm/cpu.c
@@ -XXX,XX +XXX,XX @@ uint64_t arm_cpu_mp_affinity(int idx, uint8_t clustersz)
     return (Aff1 << ARM_AFF1_SHIFT) | Aff0;
 }
 
-static void cpreg_hashtable_data_destroy(gpointer data)
-{
-    /*
-     * Destroy function for cpu->cp_regs hashtable data entries.
-     * We must free the name string because it was g_strdup()ed in
-     * add_cpreg_to_hashtable(). It's OK to cast away the 'const'
-     * from r->name because we know we definitely allocated it.
-     */
-    ARMCPRegInfo *r = data;
-
-    g_free((void *)r->name);
-    g_free(r);
-}
-
 static void arm_cpu_initfn(Object *obj)
 {
     ARMCPU *cpu = ARM_CPU(obj);
 
     cpu_set_cpustate_pointers(cpu);
     cpu->cp_regs = g_hash_table_new_full(g_direct_hash, g_direct_equal,
-                                         NULL, cpreg_hashtable_data_destroy);
+                                         NULL, g_free);
 
     QLIST_INIT(&cpu->pre_el_change_hooks);
     QLIST_INIT(&cpu->el_change_hooks);
diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
      * add a single reginfo struct to the hash table.
      */
     uint32_t key;
-    ARMCPRegInfo *r2 = g_memdup(r, sizeof(ARMCPRegInfo));
+    ARMCPRegInfo *r2;
     int is64 = (r->type & ARM_CP_64BIT) ? 1 : 0;
     int ns = (secstate & ARM_CP_SECSTATE_NS) ? 1 : 0;
+    size_t name_len;
+
+    /* Combine cpreg and name into one allocation. */
+    name_len = strlen(name) + 1;
+    r2 = g_malloc(sizeof(*r2) + name_len);
+    *r2 = *r;
+    r2->name = memcpy(r2 + 1, name, name_len);
 
-    r2->name = g_strdup(name);
     /* Reset the secure state to the specific incoming state.  This is
      * necessary as the register may have been defined with both states.
      */
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Move the computation of key to the top of the function.
Hoist the resolution of cp as well, as an input to the
computation of key.

This will be required by a subsequent patch.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20220501055028.646596-14-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/helper.c | 49 +++++++++++++++++++++++++--------------------
 1 file changed, 27 insertions(+), 22 deletions(-)

diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
     ARMCPRegInfo *r2;
     int is64 = (r->type & ARM_CP_64BIT) ? 1 : 0;
     int ns = (secstate & ARM_CP_SECSTATE_NS) ? 1 : 0;
+    int cp = r->cp;
     size_t name_len;
 
+    switch (state) {
+    case ARM_CP_STATE_AA32:
+        /* We assume it is a cp15 register if the .cp field is left unset. */
+        if (cp == 0 && r->state == ARM_CP_STATE_BOTH) {
+            cp = 15;
+        }
+        key = ENCODE_CP_REG(cp, is64, ns, r->crn, crm, opc1, opc2);
+        break;
+    case ARM_CP_STATE_AA64:
+        /*
+         * To allow abbreviation of ARMCPRegInfo definitions, we treat
+         * cp == 0 as equivalent to the value for "standard guest-visible
+         * sysreg".  STATE_BOTH definitions are also always "standard sysreg"
+         * in their AArch64 view (the .cp value may be non-zero for the
+         * benefit of the AArch32 view).
+         */
+        if (cp == 0 || r->state == ARM_CP_STATE_BOTH) {
+            cp = CP_REG_ARM64_SYSREG_CP;
+        }
+        key = ENCODE_AA64_CP_REG(cp, r->crn, crm, r->opc0, opc1, opc2);
+        break;
+    default:
+        g_assert_not_reached();
+    }
+
     /* Combine cpreg and name into one allocation. */
     name_len = strlen(name) + 1;
     r2 = g_malloc(sizeof(*r2) + name_len);
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
         }
 
         if (r->state == ARM_CP_STATE_BOTH) {
-            /* We assume it is a cp15 register if the .cp field is left unset.
-             */
-            if (r2->cp == 0) {
-                r2->cp = 15;
-            }
-
 #if HOST_BIG_ENDIAN
             if (r2->fieldoffset) {
                 r2->fieldoffset += sizeof(uint32_t);
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
 #endif
         }
     }
-    if (state == ARM_CP_STATE_AA64) {
-        /* To allow abbreviation of ARMCPRegInfo
-         * definitions, we treat cp == 0 as equivalent to
-         * the value for "standard guest-visible sysreg".
-         * STATE_BOTH definitions are also always "standard
-         * sysreg" in their AArch64 view (the .cp value may
-         * be non-zero for the benefit of the AArch32 view).
-         */
-        if (r->cp == 0 || r->state == ARM_CP_STATE_BOTH) {
-            r2->cp = CP_REG_ARM64_SYSREG_CP;
-        }
-        key = ENCODE_AA64_CP_REG(r2->cp, r2->crn, crm,
-                                 r2->opc0, opc1, opc2);
-    } else {
-        key = ENCODE_CP_REG(r2->cp, is64, ns, r2->crn, crm, opc1, opc2);
-    }
     if (opaque) {
         r2->opaque = opaque;
     }
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
     /* Make sure reginfo passed to helpers for wildcarded regs
      * has the correct crm/opc1/opc2 for this reg, not CP_ANY:
      */
+    r2->cp = cp;
     r2->crm = crm;
     r2->opc1 = opc1;
     r2->opc2 = opc2;
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Put most of the value writeback to the same place,
and improve the comment that goes with them.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20220501055028.646596-15-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/helper.c | 28 ++++++++++++----------------
 1 file changed, 12 insertions(+), 16 deletions(-)

diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
     *r2 = *r;
     r2->name = memcpy(r2 + 1, name, name_len);
 
-    /* Reset the secure state to the specific incoming state.  This is
-     * necessary as the register may have been defined with both states.
+    /*
+     * Update fields to match the instantiation, overwiting wildcards
+     * such as CP_ANY, ARM_CP_STATE_BOTH, or ARM_CP_SECSTATE_BOTH.
      */
+    r2->cp = cp;
+    r2->crm = crm;
+    r2->opc1 = opc1;
+    r2->opc2 = opc2;
+    r2->state = state;
     r2->secure = secstate;
+    if (opaque) {
+        r2->opaque = opaque;
+    }
 
     if (r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1]) {
         /* Register is banked (using both entries in array).
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
 #endif
         }
     }
-    if (opaque) {
-        r2->opaque = opaque;
-    }
-    /* reginfo passed to helpers is correct for the actual access,
-     * and is never ARM_CP_STATE_BOTH:
-     */
-    r2->state = state;
-    /* Make sure reginfo passed to helpers for wildcarded regs
-     * has the correct crm/opc1/opc2 for this reg, not CP_ANY:
-     */
-    r2->cp = cp;
-    r2->crm = crm;
-    r2->opc1 = opc1;
-    r2->opc2 = opc2;
+
     /* By convention, for wildcarded registers only the first
      * entry is used for migration; the others are marked as
      * ALIAS so we don't try to transfer the register
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Computing isbanked only once makes the code
a bit easier to read.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20220501055028.646596-17-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/helper.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

From: Richard Henderson <richard.henderson@linaro.org>

Perform the override check early, so that it is still done
even when we decide to discard an unreachable cpreg.

Use assert not printf+abort.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20220501055028.646596-18-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/helper.c | 22 ++++++++--------------
 1 file changed, 8 insertions(+), 14 deletions(-)

diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
         g_assert_not_reached();
     }
 
+    /* Overriding of an existing definition must be explicitly requested. */
+    if (!(r->type & ARM_CP_OVERRIDE)) {
+        const ARMCPRegInfo *oldreg = get_arm_cp_reginfo(cpu->cp_regs, key);
+        if (oldreg) {
+            assert(oldreg->type & ARM_CP_OVERRIDE);
+        }
+    }
+
     /* Combine cpreg and name into one allocation. */
     name_len = strlen(name) + 1;
     r2 = g_malloc(sizeof(*r2) + name_len);
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
         assert(!raw_accessors_invalid(r2));
     }
 
-    /* Overriding of an existing definition must be explicitly
-     * requested.
-     */
-    if (!(r->type & ARM_CP_OVERRIDE)) {
-        const ARMCPRegInfo *oldreg = get_arm_cp_reginfo(cpu->cp_regs, key);
-        if (oldreg && !(oldreg->type & ARM_CP_OVERRIDE)) {
-            fprintf(stderr, "Register redefined: cp=%d %d bit "
-                    "crn=%d crm=%d opc1=%d opc2=%d, "
-                    "was %s, now %s\n", r2->cp, 32 + 32 * is64,
-                    r2->crn, r2->crm, r2->opc1, r2->opc2,
-                    oldreg->name, r2->name);
-            g_assert_not_reached();
-        }
-    }
     g_hash_table_insert(cpu->cp_regs, (gpointer)(uintptr_t)key, r2);
 }
 
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Put the block comments into the current coding style.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20220501055028.646596-19-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/helper.c | 24 +++++++++++++++---------
 1 file changed, 15 insertions(+), 9 deletions(-)

diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ CpuDefinitionInfoList *qmp_query_cpu_definitions(Error **errp)
     return cpu_list;
 }
 
+/*
+ * Private utility function for define_one_arm_cp_reg_with_opaque():
+ * add a single reginfo struct to the hash table.
+ */
 static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
                                    void *opaque, CPState state,
                                    CPSecureState secstate,
                                    int crm, int opc1, int opc2,
                                    const char *name)
 {
-    /* Private utility function for define_one_arm_cp_reg_with_opaque():
-     * add a single reginfo struct to the hash table.
-     */
     uint32_t key;
     ARMCPRegInfo *r2;
     bool is64 = r->type & ARM_CP_64BIT;
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
 
     isbanked = r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1];
     if (isbanked) {
-        /* Register is banked (using both entries in array).
+        /*
+         * Register is banked (using both entries in array).
          * Overwriting fieldoffset as the array is only used to define
          * banked registers but later only fieldoffset is used.
          */
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
 
     if (state == ARM_CP_STATE_AA32) {
         if (isbanked) {
-            /* If the register is banked then we don't need to migrate or
+            /*
+             * If the register is banked then we don't need to migrate or
              * reset the 32-bit instance in certain cases:
              *
              * 1) If the register has both 32-bit and 64-bit instances then we
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
                 r2->type |= ARM_CP_ALIAS;
             }
         } else if ((secstate != r->secure) && !ns) {
-            /* The register is not banked so we only want to allow migration of
-             * the non-secure instance.
+            /*
+             * The register is not banked so we only want to allow migration
+             * of the non-secure instance.
              */
             r2->type |= ARM_CP_ALIAS;
         }
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
         }
     }
 
-    /* By convention, for wildcarded registers only the first
+    /*
+     * By convention, for wildcarded registers only the first
      * entry is used for migration; the others are marked as
      * ALIAS so we don't try to transfer the register
      * multiple times. Special registers (ie NOP/WFI) are
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
         r2->type |= ARM_CP_ALIAS | ARM_CP_NO_GDB;
     }
 
-    /* Check that raw accesses are either forbidden or handled. Note that
+    /*
+     * Check that raw accesses are either forbidden or handled. Note that
      * we can't assert this earlier because the setup of fieldoffset for
      * banked registers has to be done first.
      */
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Since e03b56863d2bc, our host endian indicator is unconditionally
set, which means that we can use a normal C condition.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20220501055028.646596-20-richard.henderson@linaro.org
[PMM: quote correct git hash in commit message]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/helper.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r,
             r2->type |= ARM_CP_ALIAS;
         }
 
-        if (r->state == ARM_CP_STATE_BOTH) {
-#if HOST_BIG_ENDIAN
-            if (r2->fieldoffset) {
-                r2->fieldoffset += sizeof(uint32_t);
-            }
-#endif
+        if (HOST_BIG_ENDIAN &&
+            r->state == ARM_CP_STATE_BOTH && r2->fieldoffset) {
+            r2->fieldoffset += sizeof(uint32_t);
         }
     }
 
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220501055028.646596-24-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/cpu.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa32_ssbs(const ARMISARegisters *id)
     return FIELD_EX32(id->id_pfr2, ID_PFR2, SSBS) != 0;
 }
 
+static inline bool isar_feature_aa32_debugv8p2(const ARMISARegisters *id)
+{
+    return FIELD_EX32(id->id_dfr0, ID_DFR0, COPDBG) >= 8;
+}
+
 /*
  * 64-bit feature tests via id registers.
  */
@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_ssbs(const ARMISARegisters *id)
     return FIELD_EX64(id->id_aa64pfr1, ID_AA64PFR1, SSBS) != 0;
 }
 
+static inline bool isar_feature_aa64_debugv8p2(const ARMISARegisters *id)
+{
+    return FIELD_EX64(id->id_aa64dfr0, ID_AA64DFR0, DEBUGVER) >= 8;
+}
+
 static inline bool isar_feature_aa64_sve2(const ARMISARegisters *id)
 {
     return FIELD_EX64(id->id_aa64zfr0, ID_AA64ZFR0, SVEVER) != 0;
@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_any_tts2uxn(const ARMISARegisters *id)
     return isar_feature_aa64_tts2uxn(id) || isar_feature_aa32_tts2uxn(id);
 }
 
+static inline bool isar_feature_any_debugv8p2(const ARMISARegisters *id)
+{
+    return isar_feature_aa64_debugv8p2(id) || isar_feature_aa32_debugv8p2(id);
+}
+
 /*
  * Forward to the above feature tests given an ARMCPU pointer.
  */
-- 
2.25.1

From: Richard Henderson <richard.henderson@linaro.org>

Add the aa64 predicate for detecting RAS support from id registers.
We already have the aa32 version from the M-profile work.
Add the 'any' predicate for testing both aa64 and aa32.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220501055028.646596-34-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/cpu.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_aa32_el1(const ARMISARegisters *id)
     return FIELD_EX64(id->id_aa64pfr0, ID_AA64PFR0, EL1) >= 2;
 }
 
+static inline bool isar_feature_aa64_ras(const ARMISARegisters *id)
+{
+    return FIELD_EX64(id->id_aa64pfr0, ID_AA64PFR0, RAS) != 0;
+}
+
 static inline bool isar_feature_aa64_sve(const ARMISARegisters *id)
 {
     return FIELD_EX64(id->id_aa64pfr0, ID_AA64PFR0, SVE) != 0;
@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_any_debugv8p2(const ARMISARegisters *id)
     return isar_feature_aa64_debugv8p2(id) || isar_feature_aa32_debugv8p2(id);
 }
 
+static inline bool isar_feature_any_ras(const ARMISARegisters *id)
+{
+    return isar_feature_aa64_ras(id) || isar_feature_aa32_ras(id);
+}
+
 /*
  * Forward to the above feature tests given an ARMCPU pointer.
  */
-- 
2.25.1

From: Alex Zuepke <alex.zuepke@tum.de>

The ARMv8 manual defines that PMUSERENR_EL0.ER enables read-access
to both PMXEVCNTR_EL0 and PMEVCNTR<n>_EL0 registers, however,
we only use it for PMXEVCNTR_EL0. Extend to PMEVCNTR<n>_EL0 as well.

Signed-off-by: Alex Zuepke <alex.zuepke@tum.de>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220428132717.84190-1-alex.zuepke@tum.de
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/helper.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static void define_pmu_regs(ARMCPU *cpu)
               .crm = 8 | (3 & (i >> 3)), .opc1 = 0, .opc2 = i & 7,
               .access = PL0_RW, .type = ARM_CP_IO | ARM_CP_ALIAS,
               .readfn = pmevcntr_readfn, .writefn = pmevcntr_writefn,
-              .accessfn = pmreg_access },
+              .accessfn = pmreg_access_xevcntr },
             { .name = pmevcntr_el0_name, .state = ARM_CP_STATE_AA64,
               .opc0 = 3, .opc1 = 3, .crn = 14, .crm = 8 | (3 & (i >> 3)),
-              .opc2 = i & 7, .access = PL0_RW, .accessfn = pmreg_access,
+              .opc2 = i & 7, .access = PL0_RW, .accessfn = pmreg_access_xevcntr,
               .type = ARM_CP_IO,
               .readfn = pmevcntr_readfn, .writefn = pmevcntr_writefn,
               .raw_readfn = pmevcntr_rawread,
-- 
2.25.1

The following changes since commit 8f6330a807f2642dc2a3cdf33347aa28a4c00a87:

Merge tag 'pull-maintainer-updates-060324-1' of https://gitlab.com/stsquad/qemu into staging (2024-03-06 16:56:20 +0000)

are available in the Git repository at:

https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20240308

for you to fetch changes up to bbf6c6dbead82292a20951eb1204442a6b838de9:

target/arm: Move v7m-related code from cpu32.c into a separate file (2024-03-08 14:45:03 +0000)

----------------------------------------------------------------
target-arm queue:
 * Implement FEAT_ECV
 * STM32L4x5: Implement GPIO device
 * Fix 32-bit SMOPA
 * Refactor v7m related code from cpu32.c into its own file
 * hw/rtc/sun4v-rtc: Relicense to GPLv2-or-later

----------------------------------------------------------------
Inès Varhol (3):
      hw/gpio: Implement STM32L4x5 GPIO
      hw/arm: Connect STM32L4x5 GPIO to STM32L4x5 SoC
      tests/qtest: Add STM32L4x5 GPIO QTest testcase

Peter Maydell (9):
      target/arm: Move some register related defines to internals.h
      target/arm: Timer _EL02 registers UNDEF for E2H == 0
      target/arm: use FIELD macro for CNTHCTL bit definitions
      target/arm: Don't allow RES0 CNTHCTL_EL2 bits to be written
      target/arm: Implement new FEAT_ECV trap bits
      target/arm: Define CNTPCTSS_EL0 and CNTVCTSS_EL0
      target/arm: Implement FEAT_ECV CNTPOFF_EL2 handling
      target/arm: Enable FEAT_ECV for 'max' CPU
      hw/rtc/sun4v-rtc: Relicense to GPLv2-or-later

Richard Henderson (1):
      target/arm: Fix 32-bit SMOPA

Thomas Huth (1):
      target/arm: Move v7m-related code from cpu32.c into a separate file

MAINTAINERS                        |   1 +
 docs/system/arm/b-l475e-iot01a.rst |   2 +-
 docs/system/arm/emulation.rst      |   1 +
 include/hw/arm/stm32l4x5_soc.h     |   2 +
 include/hw/gpio/stm32l4x5_gpio.h   |  71 +++++
 include/hw/misc/stm32l4x5_syscfg.h |   3 +-
 include/hw/rtc/sun4v-rtc.h         |   2 +-
 target/arm/cpu-features.h          |  10 +
 target/arm/cpu.h                   | 129 +--------
 target/arm/internals.h             | 151 ++++++++++
 hw/arm/stm32l4x5_soc.c             |  71 ++++-
 hw/gpio/stm32l4x5_gpio.c           | 477 ++++++++++++++++++++++++++++++++
 hw/misc/stm32l4x5_syscfg.c         |   1 +
 hw/rtc/sun4v-rtc.c                 |   2 +-
 target/arm/helper.c                | 189 ++++++++++++-
 target/arm/tcg/cpu-v7m.c           | 290 +++++++++++++++++++
 target/arm/tcg/cpu32.c             | 261 ------------------
 target/arm/tcg/cpu64.c             |   1 +
 target/arm/tcg/sme_helper.c        |  77 +++---
 tests/qtest/stm32l4x5_gpio-test.c  | 551 +++++++++++++++++++++++++++++++++++++
 tests/tcg/aarch64/sme-smopa-1.c    |  47 ++++
 tests/tcg/aarch64/sme-smopa-2.c    |  54 ++++
 hw/arm/Kconfig                     |   3 +-
 hw/gpio/Kconfig                    |   3 +
 hw/gpio/meson.build                |   1 +
 hw/gpio/trace-events               |   6 +
 target/arm/meson.build             |   3 +
 target/arm/tcg/meson.build         |   3 +
 target/arm/trace-events            |   1 +
 tests/qtest/meson.build            |   3 +-
 tests/tcg/aarch64/Makefile.target  |   2 +-
 31 files changed, 1962 insertions(+), 456 deletions(-)
 create mode 100644 include/hw/gpio/stm32l4x5_gpio.h
 create mode 100644 hw/gpio/stm32l4x5_gpio.c
 create mode 100644 target/arm/tcg/cpu-v7m.c
 create mode 100644 tests/qtest/stm32l4x5_gpio-test.c
 create mode 100644 tests/tcg/aarch64/sme-smopa-1.c
 create mode 100644 tests/tcg/aarch64/sme-smopa-2.c

cpu.h has a lot of #defines relating to CPU register fields.
Most of these aren't actually used outside target/arm code,
so there's no point in cluttering up the cpu.h file with them.
Move some easy ones to internals.h.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20240301183219.2424889-2-peter.maydell@linaro.org
---
 target/arm/cpu.h       | 128 -----------------------------------------
 target/arm/internals.h | 128 +++++++++++++++++++++++++++++++++++++++++
 2 files changed, 128 insertions(+), 128 deletions(-)

diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -XXX,XX +XXX,XX @@ typedef struct ARMGenericTimer {
     uint64_t ctl; /* Timer Control register */
 } ARMGenericTimer;
 
-#define VTCR_NSW (1u << 29)
-#define VTCR_NSA (1u << 30)
-#define VSTCR_SW VTCR_NSW
-#define VSTCR_SA VTCR_NSA
-
 /* Define a maximum sized vector register.
  * For 32-bit, this is a 128-bit NEON/AdvSIMD register.
  * For 64-bit, this is a 2048-bit SVE register.
@@ -XXX,XX +XXX,XX @@ void pmu_init(ARMCPU *cpu);
 #define SCTLR_SPINTMASK (1ULL << 62) /* FEAT_NMI */
 #define SCTLR_TIDCP   (1ULL << 63) /* FEAT_TIDCP1 */
 
-/* Bit definitions for CPACR (AArch32 only) */
-FIELD(CPACR, CP10, 20, 2)
-FIELD(CPACR, CP11, 22, 2)
-FIELD(CPACR, TRCDIS, 28, 1)    /* matches CPACR_EL1.TTA */
-FIELD(CPACR, D32DIS, 30, 1)    /* up to v7; RAZ in v8 */
-FIELD(CPACR, ASEDIS, 31, 1)
-
-/* Bit definitions for CPACR_EL1 (AArch64 only) */
-FIELD(CPACR_EL1, ZEN, 16, 2)
-FIELD(CPACR_EL1, FPEN, 20, 2)
-FIELD(CPACR_EL1, SMEN, 24, 2)
-FIELD(CPACR_EL1, TTA, 28, 1)   /* matches CPACR.TRCDIS */
-
-/* Bit definitions for HCPTR (AArch32 only) */
-FIELD(HCPTR, TCP10, 10, 1)
-FIELD(HCPTR, TCP11, 11, 1)
-FIELD(HCPTR, TASE, 15, 1)
-FIELD(HCPTR, TTA, 20, 1)
-FIELD(HCPTR, TAM, 30, 1)       /* matches CPTR_EL2.TAM */
-FIELD(HCPTR, TCPAC, 31, 1)     /* matches CPTR_EL2.TCPAC */
-
-/* Bit definitions for CPTR_EL2 (AArch64 only) */
-FIELD(CPTR_EL2, TZ, 8, 1)      /* !E2H */
-FIELD(CPTR_EL2, TFP, 10, 1)    /* !E2H, matches HCPTR.TCP10 */
-FIELD(CPTR_EL2, TSM, 12, 1)    /* !E2H */
-FIELD(CPTR_EL2, ZEN, 16, 2)    /* E2H */
-FIELD(CPTR_EL2, FPEN, 20, 2)   /* E2H */
-FIELD(CPTR_EL2, SMEN, 24, 2)   /* E2H */
-FIELD(CPTR_EL2, TTA, 28, 1)
-FIELD(CPTR_EL2, TAM, 30, 1)    /* matches HCPTR.TAM */
-FIELD(CPTR_EL2, TCPAC, 31, 1)  /* matches HCPTR.TCPAC */
-
-/* Bit definitions for CPTR_EL3 (AArch64 only) */
-FIELD(CPTR_EL3, EZ, 8, 1)
-FIELD(CPTR_EL3, TFP, 10, 1)
-FIELD(CPTR_EL3, ESM, 12, 1)
-FIELD(CPTR_EL3, TTA, 20, 1)
-FIELD(CPTR_EL3, TAM, 30, 1)
-FIELD(CPTR_EL3, TCPAC, 31, 1)
-
-#define MDCR_MTPME    (1U << 28)
-#define MDCR_TDCC     (1U << 27)
-#define MDCR_HLP      (1U << 26)  /* MDCR_EL2 */
-#define MDCR_SCCD     (1U << 23)  /* MDCR_EL3 */
-#define MDCR_HCCD     (1U << 23)  /* MDCR_EL2 */
-#define MDCR_EPMAD    (1U << 21)
-#define MDCR_EDAD     (1U << 20)
-#define MDCR_TTRF     (1U << 19)
-#define MDCR_STE      (1U << 18)  /* MDCR_EL3 */
-#define MDCR_SPME     (1U << 17)  /* MDCR_EL3 */
-#define MDCR_HPMD     (1U << 17)  /* MDCR_EL2 */
-#define MDCR_SDD      (1U << 16)
-#define MDCR_SPD      (3U << 14)
-#define MDCR_TDRA     (1U << 11)
-#define MDCR_TDOSA    (1U << 10)
-#define MDCR_TDA      (1U << 9)
-#define MDCR_TDE      (1U << 8)
-#define MDCR_HPME     (1U << 7)
-#define MDCR_TPM      (1U << 6)
-#define MDCR_TPMCR    (1U << 5)
-#define MDCR_HPMN     (0x1fU)
-
-/* Not all of the MDCR_EL3 bits are present in the 32-bit SDCR */
-#define SDCR_VALID_MASK (MDCR_MTPME | MDCR_TDCC | MDCR_SCCD | \
-                         MDCR_EPMAD | MDCR_EDAD | MDCR_TTRF | \
-                         MDCR_STE | MDCR_SPME | MDCR_SPD)
-
 #define CPSR_M (0x1fU)
 #define CPSR_T (1U << 5)
 #define CPSR_F (1U << 6)
@@ -XXX,XX +XXX,XX @@ FIELD(CPTR_EL3, TCPAC, 31, 1)
 #define XPSR_NZCV CPSR_NZCV
 #define XPSR_IT CPSR_IT
 
-#define TTBCR_N      (7U << 0) /* TTBCR.EAE==0 */
-#define TTBCR_T0SZ   (7U << 0) /* TTBCR.EAE==1 */
-#define TTBCR_PD0    (1U << 4)
-#define TTBCR_PD1    (1U << 5)
-#define TTBCR_EPD0   (1U << 7)
-#define TTBCR_IRGN0  (3U << 8)
-#define TTBCR_ORGN0  (3U << 10)
-#define TTBCR_SH0    (3U << 12)
-#define TTBCR_T1SZ   (3U << 16)
-#define TTBCR_A1     (1U << 22)
-#define TTBCR_EPD1   (1U << 23)
-#define TTBCR_IRGN1  (3U << 24)
-#define TTBCR_ORGN1  (3U << 26)
-#define TTBCR_SH1    (1U << 28)
-#define TTBCR_EAE    (1U << 31)
-
-FIELD(VTCR, T0SZ, 0, 6)
-FIELD(VTCR, SL0, 6, 2)
-FIELD(VTCR, IRGN0, 8, 2)
-FIELD(VTCR, ORGN0, 10, 2)
-FIELD(VTCR, SH0, 12, 2)
-FIELD(VTCR, TG0, 14, 2)
-FIELD(VTCR, PS, 16, 3)
-FIELD(VTCR, VS, 19, 1)
-FIELD(VTCR, HA, 21, 1)
-FIELD(VTCR, HD, 22, 1)
-FIELD(VTCR, HWU59, 25, 1)
-FIELD(VTCR, HWU60, 26, 1)
-FIELD(VTCR, HWU61, 27, 1)
-FIELD(VTCR, HWU62, 28, 1)
-FIELD(VTCR, NSW, 29, 1)
-FIELD(VTCR, NSA, 30, 1)
-FIELD(VTCR, DS, 32, 1)
-FIELD(VTCR, SL2, 33, 1)
-
 /* Bit definitions for ARMv8 SPSR (PSTATE) format.
  * Only these are valid when in AArch64 mode; in
  * AArch32 mode SPSRs are basically CPSR-format.
@@ -XXX,XX +XXX,XX @@ static inline void xpsr_write(CPUARMState *env, uint32_t val, uint32_t mask)
 #define HCR_TWEDEN    (1ULL << 59)
 #define HCR_TWEDEL    MAKE_64BIT_MASK(60, 4)
 
-#define HCRX_ENAS0    (1ULL << 0)
-#define HCRX_ENALS    (1ULL << 1)
-#define HCRX_ENASR    (1ULL << 2)
-#define HCRX_FNXS     (1ULL << 3)
-#define HCRX_FGTNXS   (1ULL << 4)
-#define HCRX_SMPME    (1ULL << 5)
-#define HCRX_TALLINT  (1ULL << 6)
-#define HCRX_VINMI    (1ULL << 7)
-#define HCRX_VFNMI    (1ULL << 8)
-#define HCRX_CMOW     (1ULL << 9)
-#define HCRX_MCE2     (1ULL << 10)
-#define HCRX_MSCEN    (1ULL << 11)
-
-#define HPFAR_NS      (1ULL << 63)
-
 #define SCR_NS                (1ULL << 0)
 #define SCR_IRQ               (1ULL << 1)
 #define SCR_FIQ               (1ULL << 2)
@@ -XXX,XX +XXX,XX @@ static inline void xpsr_write(CPUARMState *env, uint32_t val, uint32_t mask)
 #define SCR_GPF               (1ULL << 48)
 #define SCR_NSE               (1ULL << 62)
 
-#define HSTR_TTEE (1 << 16)
-#define HSTR_TJDBX (1 << 17)
-
-#define CNTHCTL_CNTVMASK      (1 << 18)
-#define CNTHCTL_CNTPMASK      (1 << 19)
-
 /* Return the current FPSCR value.  */
 uint32_t vfp_get_fpscr(CPUARMState *env);
 void vfp_set_fpscr(CPUARMState *env, uint32_t val);
diff --git a/target/arm/internals.h b/target/arm/internals.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/internals.h
+++ b/target/arm/internals.h
@@ -XXX,XX +XXX,XX @@ FIELD(DBGWCR, WT, 20, 1)
 FIELD(DBGWCR, MASK, 24, 5)
 FIELD(DBGWCR, SSCE, 29, 1)
 
+#define VTCR_NSW (1u << 29)
+#define VTCR_NSA (1u << 30)
+#define VSTCR_SW VTCR_NSW
+#define VSTCR_SA VTCR_NSA
+
+/* Bit definitions for CPACR (AArch32 only) */
+FIELD(CPACR, CP10, 20, 2)
+FIELD(CPACR, CP11, 22, 2)
+FIELD(CPACR, TRCDIS, 28, 1)    /* matches CPACR_EL1.TTA */
+FIELD(CPACR, D32DIS, 30, 1)    /* up to v7; RAZ in v8 */
+FIELD(CPACR, ASEDIS, 31, 1)
+
+/* Bit definitions for CPACR_EL1 (AArch64 only) */
+FIELD(CPACR_EL1, ZEN, 16, 2)
+FIELD(CPACR_EL1, FPEN, 20, 2)
+FIELD(CPACR_EL1, SMEN, 24, 2)
+FIELD(CPACR_EL1, TTA, 28, 1)   /* matches CPACR.TRCDIS */
+
+/* Bit definitions for HCPTR (AArch32 only) */
+FIELD(HCPTR, TCP10, 10, 1)
+FIELD(HCPTR, TCP11, 11, 1)
+FIELD(HCPTR, TASE, 15, 1)
+FIELD(HCPTR, TTA, 20, 1)
+FIELD(HCPTR, TAM, 30, 1)       /* matches CPTR_EL2.TAM */
+FIELD(HCPTR, TCPAC, 31, 1)     /* matches CPTR_EL2.TCPAC */
+
+/* Bit definitions for CPTR_EL2 (AArch64 only) */
+FIELD(CPTR_EL2, TZ, 8, 1)      /* !E2H */
+FIELD(CPTR_EL2, TFP, 10, 1)    /* !E2H, matches HCPTR.TCP10 */
+FIELD(CPTR_EL2, TSM, 12, 1)    /* !E2H */
+FIELD(CPTR_EL2, ZEN, 16, 2)    /* E2H */
+FIELD(CPTR_EL2, FPEN, 20, 2)   /* E2H */
+FIELD(CPTR_EL2, SMEN, 24, 2)   /* E2H */
+FIELD(CPTR_EL2, TTA, 28, 1)
+FIELD(CPTR_EL2, TAM, 30, 1)    /* matches HCPTR.TAM */
+FIELD(CPTR_EL2, TCPAC, 31, 1)  /* matches HCPTR.TCPAC */
+
+/* Bit definitions for CPTR_EL3 (AArch64 only) */
+FIELD(CPTR_EL3, EZ, 8, 1)
+FIELD(CPTR_EL3, TFP, 10, 1)
+FIELD(CPTR_EL3, ESM, 12, 1)
+FIELD(CPTR_EL3, TTA, 20, 1)
+FIELD(CPTR_EL3, TAM, 30, 1)
+FIELD(CPTR_EL3, TCPAC, 31, 1)
+
+#define MDCR_MTPME    (1U << 28)
+#define MDCR_TDCC     (1U << 27)
+#define MDCR_HLP      (1U << 26)  /* MDCR_EL2 */
+#define MDCR_SCCD     (1U << 23)  /* MDCR_EL3 */
+#define MDCR_HCCD     (1U << 23)  /* MDCR_EL2 */
+#define MDCR_EPMAD    (1U << 21)
+#define MDCR_EDAD     (1U << 20)
+#define MDCR_TTRF     (1U << 19)
+#define MDCR_STE      (1U << 18)  /* MDCR_EL3 */
+#define MDCR_SPME     (1U << 17)  /* MDCR_EL3 */
+#define MDCR_HPMD     (1U << 17)  /* MDCR_EL2 */
+#define MDCR_SDD      (1U << 16)
+#define MDCR_SPD      (3U << 14)
+#define MDCR_TDRA     (1U << 11)
+#define MDCR_TDOSA    (1U << 10)
+#define MDCR_TDA      (1U << 9)
+#define MDCR_TDE      (1U << 8)
+#define MDCR_HPME     (1U << 7)
+#define MDCR_TPM      (1U << 6)
+#define MDCR_TPMCR    (1U << 5)
+#define MDCR_HPMN     (0x1fU)
+
+/* Not all of the MDCR_EL3 bits are present in the 32-bit SDCR */
+#define SDCR_VALID_MASK (MDCR_MTPME | MDCR_TDCC | MDCR_SCCD | \
+                         MDCR_EPMAD | MDCR_EDAD | MDCR_TTRF | \
+                         MDCR_STE | MDCR_SPME | MDCR_SPD)
+
+#define TTBCR_N      (7U << 0) /* TTBCR.EAE==0 */
+#define TTBCR_T0SZ   (7U << 0) /* TTBCR.EAE==1 */
+#define TTBCR_PD0    (1U << 4)
+#define TTBCR_PD1    (1U << 5)
+#define TTBCR_EPD0   (1U << 7)
+#define TTBCR_IRGN0  (3U << 8)
+#define TTBCR_ORGN0  (3U << 10)
+#define TTBCR_SH0    (3U << 12)
+#define TTBCR_T1SZ   (3U << 16)
+#define TTBCR_A1     (1U << 22)
+#define TTBCR_EPD1   (1U << 23)
+#define TTBCR_IRGN1  (3U << 24)
+#define TTBCR_ORGN1  (3U << 26)
+#define TTBCR_SH1    (1U << 28)
+#define TTBCR_EAE    (1U << 31)
+
+FIELD(VTCR, T0SZ, 0, 6)
+FIELD(VTCR, SL0, 6, 2)
+FIELD(VTCR, IRGN0, 8, 2)
+FIELD(VTCR, ORGN0, 10, 2)
+FIELD(VTCR, SH0, 12, 2)
+FIELD(VTCR, TG0, 14, 2)
+FIELD(VTCR, PS, 16, 3)
+FIELD(VTCR, VS, 19, 1)
+FIELD(VTCR, HA, 21, 1)
+FIELD(VTCR, HD, 22, 1)
+FIELD(VTCR, HWU59, 25, 1)
+FIELD(VTCR, HWU60, 26, 1)
+FIELD(VTCR, HWU61, 27, 1)
+FIELD(VTCR, HWU62, 28, 1)
+FIELD(VTCR, NSW, 29, 1)
+FIELD(VTCR, NSA, 30, 1)
+FIELD(VTCR, DS, 32, 1)
+FIELD(VTCR, SL2, 33, 1)
+
+#define HCRX_ENAS0    (1ULL << 0)
+#define HCRX_ENALS    (1ULL << 1)
+#define HCRX_ENASR    (1ULL << 2)
+#define HCRX_FNXS     (1ULL << 3)
+#define HCRX_FGTNXS   (1ULL << 4)
+#define HCRX_SMPME    (1ULL << 5)
+#define HCRX_TALLINT  (1ULL << 6)
+#define HCRX_VINMI    (1ULL << 7)
+#define HCRX_VFNMI    (1ULL << 8)
+#define HCRX_CMOW     (1ULL << 9)
+#define HCRX_MCE2     (1ULL << 10)
+#define HCRX_MSCEN    (1ULL << 11)
+
+#define HPFAR_NS      (1ULL << 63)
+
+#define HSTR_TTEE (1 << 16)
+#define HSTR_TJDBX (1 << 17)
+
+#define CNTHCTL_CNTVMASK      (1 << 18)
+#define CNTHCTL_CNTPMASK      (1 << 19)
+
 /* We use a few fake FSR values for internal purposes in M profile.
  * M profile cores don't have A/R format FSRs, but currently our
  * get_phys_addr() code assumes A/R profile and reports failures via
-- 
2.34.1

We prefer the FIELD macro over ad-hoc #defines for register bits;
switch CNTHCTL to that style before we add any more bits.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20240301183219.2424889-4-peter.maydell@linaro.org
---
 target/arm/internals.h | 27 +++++++++++++++++++++++++--
 target/arm/helper.c    |  9 ++++-----
 2 files changed, 29 insertions(+), 7 deletions(-)

diff --git a/target/arm/internals.h b/target/arm/internals.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/internals.h
+++ b/target/arm/internals.h
@@ -XXX,XX +XXX,XX @@ FIELD(VTCR, SL2, 33, 1)
 #define HSTR_TTEE (1 << 16)
 #define HSTR_TJDBX (1 << 17)
 
-#define CNTHCTL_CNTVMASK      (1 << 18)
-#define CNTHCTL_CNTPMASK      (1 << 19)
+/*
+ * Depending on the value of HCR_EL2.E2H, bits 0 and 1
+ * have different bit definitions, and EL1PCTEN might be
+ * bit 0 or bit 10. We use _E2H1 and _E2H0 suffixes to
+ * disambiguate if necessary.
+ */
+FIELD(CNTHCTL, EL0PCTEN_E2H1, 0, 1)
+FIELD(CNTHCTL, EL0VCTEN_E2H1, 1, 1)
+FIELD(CNTHCTL, EL1PCTEN_E2H0, 0, 1)
+FIELD(CNTHCTL, EL1PCEN_E2H0, 1, 1)
+FIELD(CNTHCTL, EVNTEN, 2, 1)
+FIELD(CNTHCTL, EVNTDIR, 3, 1)
+FIELD(CNTHCTL, EVNTI, 4, 4)
+FIELD(CNTHCTL, EL0VTEN, 8, 1)
+FIELD(CNTHCTL, EL0PTEN, 9, 1)
+FIELD(CNTHCTL, EL1PCTEN_E2H1, 10, 1)
+FIELD(CNTHCTL, EL1PTEN, 11, 1)
+FIELD(CNTHCTL, ECV, 12, 1)
+FIELD(CNTHCTL, EL1TVT, 13, 1)
+FIELD(CNTHCTL, EL1TVCT, 14, 1)
+FIELD(CNTHCTL, EL1NVPCT, 15, 1)
+FIELD(CNTHCTL, EL1NVVCT, 16, 1)
+FIELD(CNTHCTL, EVNTIS, 17, 1)
+FIELD(CNTHCTL, CNTVMASK, 18, 1)
+FIELD(CNTHCTL, CNTPMASK, 19, 1)
 
 /* We use a few fake FSR values for internal purposes in M profile.
  * M profile cores don't have A/R format FSRs, but currently our
diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static void gt_update_irq(ARMCPU *cpu, int timeridx)
      * It is RES0 in Secure and NonSecure state.
      */
     if ((ss == ARMSS_Root || ss == ARMSS_Realm) &&
-        ((timeridx == GTIMER_VIRT && (cnthctl & CNTHCTL_CNTVMASK)) ||
-         (timeridx == GTIMER_PHYS && (cnthctl & CNTHCTL_CNTPMASK)))) {
+        ((timeridx == GTIMER_VIRT && (cnthctl & R_CNTHCTL_CNTVMASK_MASK)) ||
+         (timeridx == GTIMER_PHYS && (cnthctl & R_CNTHCTL_CNTPMASK_MASK)))) {
         irqstate = 0;
     }
 
@@ -XXX,XX +XXX,XX @@ static void gt_cnthctl_write(CPUARMState *env, const ARMCPRegInfo *ri,
 {
     ARMCPU *cpu = env_archcpu(env);
     uint32_t oldval = env->cp15.cnthctl_el2;
-
     raw_write(env, ri, value);
 
-    if ((oldval ^ value) & CNTHCTL_CNTVMASK) {
+    if ((oldval ^ value) & R_CNTHCTL_CNTVMASK_MASK) {
         gt_update_irq(cpu, GTIMER_VIRT);
-    } else if ((oldval ^ value) & CNTHCTL_CNTPMASK) {
+    } else if ((oldval ^ value) & R_CNTHCTL_CNTPMASK_MASK) {
         gt_update_irq(cpu, GTIMER_PHYS);
     }
 }
-- 
2.34.1

Don't allow the guest to write CNTHCTL_EL2 bits which don't exist.
This is not strictly architecturally required, but it is how we've
tended to implement registers more recently.

In particular, bits [19:18] are only present with FEAT_RME,
and bits [17:12] will only be present with FEAT_ECV.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20240301183219.2424889-5-peter.maydell@linaro.org
---
 target/arm/helper.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static void gt_cnthctl_write(CPUARMState *env, const ARMCPRegInfo *ri,
 {
     ARMCPU *cpu = env_archcpu(env);
     uint32_t oldval = env->cp15.cnthctl_el2;
+    uint32_t valid_mask =
+        R_CNTHCTL_EL0PCTEN_E2H1_MASK |
+        R_CNTHCTL_EL0VCTEN_E2H1_MASK |
+        R_CNTHCTL_EVNTEN_MASK |
+        R_CNTHCTL_EVNTDIR_MASK |
+        R_CNTHCTL_EVNTI_MASK |
+        R_CNTHCTL_EL0VTEN_MASK |
+        R_CNTHCTL_EL0PTEN_MASK |
+        R_CNTHCTL_EL1PCTEN_E2H1_MASK |
+        R_CNTHCTL_EL1PTEN_MASK;
+
+    if (cpu_isar_feature(aa64_rme, cpu)) {
+        valid_mask |= R_CNTHCTL_CNTVMASK_MASK | R_CNTHCTL_CNTPMASK_MASK;
+    }
+
+    /* Clear RES0 bits */
+    value &= valid_mask;
+
     raw_write(env, ri, value);
 
     if ((oldval ^ value) & R_CNTHCTL_CNTVMASK_MASK) {
-- 
2.34.1

The functionality defined by ID_AA64MMFR0_EL1.ECV == 1 is:
 * four new trap bits for various counter and timer registers
 * the CNTHCTL_EL2.EVNTIS and CNTKCTL_EL1.EVNTIS bits which control
   scaling of the event stream. This is a no-op for us, because we don't
   implement the event stream (our WFE is a NOP): all we need to do is
   allow CNTHCTL_EL2.ENVTIS to be read and written.
 * extensions to PMSCR_EL1.PCT, PMSCR_EL2.PCT, TRFCR_EL1.TS and
   TRFCR_EL2.TS: these are all no-ops for us, because we don't implement
   FEAT_SPE or FEAT_TRF.
 * new registers CNTPCTSS_EL0 and NCTVCTSS_EL0 which are
   "self-sychronizing" views of the CNTPCT_EL0 and CNTVCT_EL0, meaning
   that no barriers are needed around their accesses. For us these
   are just the same as the normal views, because all our sysregs are
   inherently self-sychronizing.

In this commit we implement the trap handling and permit the new
CNTHCTL_EL2 bits to be written.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20240301183219.2424889-6-peter.maydell@linaro.org
---
 target/arm/cpu-features.h |  5 ++++
 target/arm/helper.c       | 51 +++++++++++++++++++++++++++++++++++----
 2 files changed, 51 insertions(+), 5 deletions(-)

diff --git a/target/arm/cpu-features.h b/target/arm/cpu-features.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu-features.h
+++ b/target/arm/cpu-features.h
@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_fgt(const ARMISARegisters *id)
     return FIELD_EX64(id->id_aa64mmfr0, ID_AA64MMFR0, FGT) != 0;
 }
 
+static inline bool isar_feature_aa64_ecv_traps(const ARMISARegisters *id)
+{
+    return FIELD_EX64(id->id_aa64mmfr0, ID_AA64MMFR0, ECV) > 0;
+}
+
 static inline bool isar_feature_aa64_vh(const ARMISARegisters *id)
 {
     return FIELD_EX64(id->id_aa64mmfr1, ID_AA64MMFR1, VH) != 0;
diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static CPAccessResult gt_counter_access(CPUARMState *env, int timeridx,
              : !extract32(env->cp15.cnthctl_el2, 0, 1))) {
             return CP_ACCESS_TRAP_EL2;
         }
+        if (has_el2 && timeridx == GTIMER_VIRT) {
+            if (FIELD_EX64(env->cp15.cnthctl_el2, CNTHCTL, EL1TVCT)) {
+                return CP_ACCESS_TRAP_EL2;
+            }
+        }
         break;
     }
     return CP_ACCESS_OK;
@@ -XXX,XX +XXX,XX @@ static CPAccessResult gt_timer_access(CPUARMState *env, int timeridx,
                 }
             }
         }
+        if (has_el2 && timeridx == GTIMER_VIRT) {
+            if (FIELD_EX64(env->cp15.cnthctl_el2, CNTHCTL, EL1TVT)) {
+                return CP_ACCESS_TRAP_EL2;
+            }
+        }
         break;
     }
     return CP_ACCESS_OK;
@@ -XXX,XX +XXX,XX @@ static void gt_cnthctl_write(CPUARMState *env, const ARMCPRegInfo *ri,
     if (cpu_isar_feature(aa64_rme, cpu)) {
         valid_mask |= R_CNTHCTL_CNTVMASK_MASK | R_CNTHCTL_CNTPMASK_MASK;
     }
+    if (cpu_isar_feature(aa64_ecv_traps, cpu)) {
+        valid_mask |=
+            R_CNTHCTL_EL1TVT_MASK |
+            R_CNTHCTL_EL1TVCT_MASK |
+            R_CNTHCTL_EL1NVPCT_MASK |
+            R_CNTHCTL_EL1NVVCT_MASK |
+            R_CNTHCTL_EVNTIS_MASK;
+    }
 
     /* Clear RES0 bits */
     value &= valid_mask;
@@ -XXX,XX +XXX,XX @@ static CPAccessResult e2h_access(CPUARMState *env, const ARMCPRegInfo *ri,
 {
     if (arm_current_el(env) == 1) {
         /* This must be a FEAT_NV access */
-        /* TODO: FEAT_ECV will need to check CNTHCTL_EL2 here */
         return CP_ACCESS_OK;
     }
     if (!(arm_hcr_el2_eff(env) & HCR_E2H)) {
@@ -XXX,XX +XXX,XX @@ static CPAccessResult e2h_access(CPUARMState *env, const ARMCPRegInfo *ri,
     return CP_ACCESS_OK;
 }
 
+static CPAccessResult access_el1nvpct(CPUARMState *env, const ARMCPRegInfo *ri,
+                                      bool isread)
+{
+    if (arm_current_el(env) == 1) {
+        /* This must be a FEAT_NV access with NVx == 101 */
+        if (FIELD_EX64(env->cp15.cnthctl_el2, CNTHCTL, EL1NVPCT)) {
+            return CP_ACCESS_TRAP_EL2;
+        }
+    }
+    return e2h_access(env, ri, isread);
+}
+
+static CPAccessResult access_el1nvvct(CPUARMState *env, const ARMCPRegInfo *ri,
+                                      bool isread)
+{
+    if (arm_current_el(env) == 1) {
+        /* This must be a FEAT_NV access with NVx == 101 */
+        if (FIELD_EX64(env->cp15.cnthctl_el2, CNTHCTL, EL1NVVCT)) {
+            return CP_ACCESS_TRAP_EL2;
+        }
+    }
+    return e2h_access(env, ri, isread);
+}
+
 /* Test if system register redirection is to occur in the current state.  */
 static bool redirect_for_e2h(CPUARMState *env)
 {
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo vhe_reginfo[] = {
     { .name = "CNTP_CTL_EL02", .state = ARM_CP_STATE_AA64,
       .opc0 = 3, .opc1 = 5, .crn = 14, .crm = 2, .opc2 = 1,
       .type = ARM_CP_IO | ARM_CP_ALIAS,
-      .access = PL2_RW, .accessfn = e2h_access,
+      .access = PL2_RW, .accessfn = access_el1nvpct,
       .nv2_redirect_offset = 0x180 | NV2_REDIR_NO_NV1,
       .fieldoffset = offsetof(CPUARMState, cp15.c14_timer[GTIMER_PHYS].ctl),
       .writefn = gt_phys_ctl_write, .raw_writefn = raw_write },
     { .name = "CNTV_CTL_EL02", .state = ARM_CP_STATE_AA64,
       .opc0 = 3, .opc1 = 5, .crn = 14, .crm = 3, .opc2 = 1,
       .type = ARM_CP_IO | ARM_CP_ALIAS,
-      .access = PL2_RW, .accessfn = e2h_access,
+      .access = PL2_RW, .accessfn = access_el1nvvct,
       .nv2_redirect_offset = 0x170 | NV2_REDIR_NO_NV1,
       .fieldoffset = offsetof(CPUARMState, cp15.c14_timer[GTIMER_VIRT].ctl),
       .writefn = gt_virt_ctl_write, .raw_writefn = raw_write },
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo vhe_reginfo[] = {
       .type = ARM_CP_IO | ARM_CP_ALIAS,
       .fieldoffset = offsetof(CPUARMState, cp15.c14_timer[GTIMER_PHYS].cval),
       .nv2_redirect_offset = 0x178 | NV2_REDIR_NO_NV1,
-      .access = PL2_RW, .accessfn = e2h_access,
+      .access = PL2_RW, .accessfn = access_el1nvpct,
       .writefn = gt_phys_cval_write, .raw_writefn = raw_write },
     { .name = "CNTV_CVAL_EL02", .state = ARM_CP_STATE_AA64,
       .opc0 = 3, .opc1 = 5, .crn = 14, .crm = 3, .opc2 = 2,
       .type = ARM_CP_IO | ARM_CP_ALIAS,
       .nv2_redirect_offset = 0x168 | NV2_REDIR_NO_NV1,
       .fieldoffset = offsetof(CPUARMState, cp15.c14_timer[GTIMER_VIRT].cval),
-      .access = PL2_RW, .accessfn = e2h_access,
+      .access = PL2_RW, .accessfn = access_el1nvvct,
       .writefn = gt_virt_cval_write, .raw_writefn = raw_write },
 #endif
 };
-- 
2.34.1

For FEAT_ECV, new registers CNTPCTSS_EL0 and CNTVCTSS_EL0 are
defined, which are "self-synchronized" views of the physical and
virtual counts as seen in the CNTPCT_EL0 and CNTVCT_EL0 registers
(meaning that no barriers are needed around accesses to them to
ensure that reads of them do not occur speculatively and out-of-order
with other instructions).

For QEMU, all our system registers are self-synchronized, so we can
simply copy the existing implementation of CNTPCT_EL0 and CNTVCT_EL0
to the new register encodings.

This means we now implement all the functionality required for
ID_AA64MMFR0_EL1.ECV == 0b0001.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20240301183219.2424889-7-peter.maydell@linaro.org
---
 target/arm/helper.c | 43 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 43 insertions(+)

diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo generic_timer_cp_reginfo[] = {
     },
 };
 
+/*
+ * FEAT_ECV adds extra views of CNTVCT_EL0 and CNTPCT_EL0 which
+ * are "self-synchronizing". For QEMU all sysregs are self-synchronizing,
+ * so our implementations here are identical to the normal registers.
+ */
+static const ARMCPRegInfo gen_timer_ecv_cp_reginfo[] = {
+    { .name = "CNTVCTSS", .cp = 15, .crm = 14, .opc1 = 9,
+      .access = PL0_R, .type = ARM_CP_64BIT | ARM_CP_NO_RAW | ARM_CP_IO,
+      .accessfn = gt_vct_access,
+      .readfn = gt_virt_cnt_read, .resetfn = arm_cp_reset_ignore,
+    },
+    { .name = "CNTVCTSS_EL0", .state = ARM_CP_STATE_AA64,
+      .opc0 = 3, .opc1 = 3, .crn = 14, .crm = 0, .opc2 = 6,
+      .access = PL0_R, .type = ARM_CP_NO_RAW | ARM_CP_IO,
+      .accessfn = gt_vct_access, .readfn = gt_virt_cnt_read,
+    },
+    { .name = "CNTPCTSS", .cp = 15, .crm = 14, .opc1 = 8,
+      .access = PL0_R, .type = ARM_CP_64BIT | ARM_CP_NO_RAW | ARM_CP_IO,
+      .accessfn = gt_pct_access,
+      .readfn = gt_cnt_read, .resetfn = arm_cp_reset_ignore,
+    },
+    { .name = "CNTPCTSS_EL0", .state = ARM_CP_STATE_AA64,
+      .opc0 = 3, .opc1 = 3, .crn = 14, .crm = 0, .opc2 = 5,
+      .access = PL0_R, .type = ARM_CP_NO_RAW | ARM_CP_IO,
+      .accessfn = gt_pct_access, .readfn = gt_cnt_read,
+    },
+};
+
 #else
 
 /*
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo generic_timer_cp_reginfo[] = {
     },
 };
 
+/*
+ * CNTVCTSS_EL0 has the same trap conditions as CNTVCT_EL0, so it also
+ * is exposed to userspace by Linux.
+ */
+static const ARMCPRegInfo gen_timer_ecv_cp_reginfo[] = {
+    { .name = "CNTVCTSS_EL0", .state = ARM_CP_STATE_AA64,
+      .opc0 = 3, .opc1 = 3, .crn = 14, .crm = 0, .opc2 = 6,
+      .access = PL0_R, .type = ARM_CP_NO_RAW | ARM_CP_IO,
+      .readfn = gt_virt_cnt_read,
+    },
+};
+
 #endif
 
 static void par_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value)
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
     if (arm_feature(env, ARM_FEATURE_GENERIC_TIMER)) {
         define_arm_cp_regs(cpu, generic_timer_cp_reginfo);
     }
+    if (cpu_isar_feature(aa64_ecv_traps, cpu)) {
+        define_arm_cp_regs(cpu, gen_timer_ecv_cp_reginfo);
+    }
     if (arm_feature(env, ARM_FEATURE_VAPA)) {
         ARMCPRegInfo vapa_cp_reginfo[] = {
             { .name = "PAR", .cp = 15, .crn = 7, .crm = 4, .opc1 = 0, .opc2 = 0,
-- 
2.34.1

When ID_AA64MMFR0_EL1.ECV is 0b0010, a new register CNTPOFF_EL2 is
implemented.  This is similar to the existing CNTVOFF_EL2, except
that it controls a hypervisor-adjustable offset made to the physical
counter and timer.

Implement the handling for this register, which includes control/trap
bits in SCR_EL3 and CNTHCTL_EL2.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20240301183219.2424889-8-peter.maydell@linaro.org
---
 target/arm/cpu-features.h |  5 +++
 target/arm/cpu.h          |  1 +
 target/arm/helper.c       | 68 +++++++++++++++++++++++++++++++++++++--
 target/arm/trace-events   |  1 +
 4 files changed, 73 insertions(+), 2 deletions(-)

diff --git a/target/arm/cpu-features.h b/target/arm/cpu-features.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu-features.h
+++ b/target/arm/cpu-features.h
@@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_ecv_traps(const ARMISARegisters *id)
     return FIELD_EX64(id->id_aa64mmfr0, ID_AA64MMFR0, ECV) > 0;
 }
 
+static inline bool isar_feature_aa64_ecv(const ARMISARegisters *id)
+{
+    return FIELD_EX64(id->id_aa64mmfr0, ID_AA64MMFR0, ECV) > 1;
+}
+
 static inline bool isar_feature_aa64_vh(const ARMISARegisters *id)
 {
     return FIELD_EX64(id->id_aa64mmfr1, ID_AA64MMFR1, VH) != 0;
diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -XXX,XX +XXX,XX @@ typedef struct CPUArchState {
         uint64_t c14_cntkctl; /* Timer Control register */
         uint64_t cnthctl_el2; /* Counter/Timer Hyp Control register */
         uint64_t cntvoff_el2; /* Counter Virtual Offset register */
+        uint64_t cntpoff_el2; /* Counter Physical Offset register */
         ARMGenericTimer c14_timer[NUM_GTIMERS];
         uint32_t c15_cpar; /* XScale Coprocessor Access Register */
         uint32_t c15_ticonfig; /* TI925T configuration byte.  */
diff --git a/target/arm/helper.c b/target/arm/helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -XXX,XX +XXX,XX @@ static void scr_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value)
         if (cpu_isar_feature(aa64_rme, cpu)) {
             valid_mask |= SCR_NSE | SCR_GPF;
         }
+        if (cpu_isar_feature(aa64_ecv, cpu)) {
+            valid_mask |= SCR_ECVEN;
+        }
     } else {
         valid_mask &= ~(SCR_RW | SCR_ST);
         if (cpu_isar_feature(aa32_ras, cpu)) {
@@ -XXX,XX +XXX,XX @@ void gt_rme_post_el_change(ARMCPU *cpu, void *ignored)
     gt_update_irq(cpu, GTIMER_PHYS);
 }
 
+static uint64_t gt_phys_raw_cnt_offset(CPUARMState *env)
+{
+    if ((env->cp15.scr_el3 & SCR_ECVEN) &&
+        FIELD_EX64(env->cp15.cnthctl_el2, CNTHCTL, ECV) &&
+        arm_is_el2_enabled(env) &&
+        (arm_hcr_el2_eff(env) & (HCR_E2H | HCR_TGE)) != (HCR_E2H | HCR_TGE)) {
+        return env->cp15.cntpoff_el2;
+    }
+    return 0;
+}
+
+static uint64_t gt_phys_cnt_offset(CPUARMState *env)
+{
+    if (arm_current_el(env) >= 2) {
+        return 0;
+    }
+    return gt_phys_raw_cnt_offset(env);
+}
+
 static void gt_recalc_timer(ARMCPU *cpu, int timeridx)
 {
     ARMGenericTimer *gt = &cpu->env.cp15.c14_timer[timeridx];
@@ -XXX,XX +XXX,XX @@ static void gt_recalc_timer(ARMCPU *cpu, int timeridx)
          * reset timer to when ISTATUS next has to change
          */
         uint64_t offset = timeridx == GTIMER_VIRT ?
-                                      cpu->env.cp15.cntvoff_el2 : 0;
+            cpu->env.cp15.cntvoff_el2 : gt_phys_raw_cnt_offset(&cpu->env);
         uint64_t count = gt_get_countervalue(&cpu->env);
         /* Note that this must be unsigned 64 bit arithmetic: */
         int istatus = count - offset >= gt->cval;
@@ -XXX,XX +XXX,XX @@ static void gt_timer_reset(CPUARMState *env, const ARMCPRegInfo *ri,
 
 static uint64_t gt_cnt_read(CPUARMState *env, const ARMCPRegInfo *ri)
 {
-    return gt_get_countervalue(env);
+    return gt_get_countervalue(env) - gt_phys_cnt_offset(env);
 }
 
 static uint64_t gt_virt_cnt_offset(CPUARMState *env)
@@ -XXX,XX +XXX,XX @@ static uint64_t gt_tval_read(CPUARMState *env, const ARMCPRegInfo *ri,
     case GTIMER_HYPVIRT:
         offset = gt_virt_cnt_offset(env);
         break;
+    case GTIMER_PHYS:
+        offset = gt_phys_cnt_offset(env);
+        break;
     }
 
     return (uint32_t)(env->cp15.c14_timer[timeridx].cval -
@@ -XXX,XX +XXX,XX @@ static void gt_tval_write(CPUARMState *env, const ARMCPRegInfo *ri,
     case GTIMER_HYPVIRT:
         offset = gt_virt_cnt_offset(env);
         break;
+    case GTIMER_PHYS:
+        offset = gt_phys_cnt_offset(env);
+        break;
     }
 
     trace_arm_gt_tval_write(timeridx, value);
@@ -XXX,XX +XXX,XX @@ static void gt_cnthctl_write(CPUARMState *env, const ARMCPRegInfo *ri,
             R_CNTHCTL_EL1NVVCT_MASK |
             R_CNTHCTL_EVNTIS_MASK;
     }
+    if (cpu_isar_feature(aa64_ecv, cpu)) {
+        valid_mask |= R_CNTHCTL_ECV_MASK;
+    }
 
     /* Clear RES0 bits */
     value &= valid_mask;
@@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo gen_timer_ecv_cp_reginfo[] = {
     },
 };
 
+static CPAccessResult gt_cntpoff_access(CPUARMState *env,
+                                        const ARMCPRegInfo *ri,
+                                        bool isread)
+{
+    if (arm_current_el(env) == 2 && !(env->cp15.scr_el3 & SCR_ECVEN)) {
+        return CP_ACCESS_TRAP_EL3;
+    }
+    return CP_ACCESS_OK;
+}
+
+static void gt_cntpoff_write(CPUARMState *env, const ARMCPRegInfo *ri,
+                              uint64_t value)
+{
+    ARMCPU *cpu = env_archcpu(env);
+
+    trace_arm_gt_cntpoff_write(value);
+    raw_write(env, ri, value);
+    gt_recalc_timer(cpu, GTIMER_PHYS);
+}
+
+static const ARMCPRegInfo gen_timer_cntpoff_reginfo = {
+    .name = "CNTPOFF_EL2", .state = ARM_CP_STATE_AA64,
+    .opc0 = 3, .opc1 = 4, .crn = 14, .crm = 0, .opc2 = 6,
+    .access = PL2_RW, .type = ARM_CP_IO, .resetvalue = 0,
+    .accessfn = gt_cntpoff_access, .writefn = gt_cntpoff_write,
+    .nv2_redirect_offset = 0x1a8,
+    .fieldoffset = offsetof(CPUARMState, cp15.cntpoff_el2),
+};
 #else
 
 /*
@@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu)
     if (cpu_isar_feature(aa64_ecv_traps, cpu)) {
         define_arm_cp_regs(cpu, gen_timer_ecv_cp_reginfo);
     }
+#ifndef CONFIG_USER_ONLY
+    if (cpu_isar_feature(aa64_ecv, cpu)) {
+        define_one_arm_cp_reg(cpu, &gen_timer_cntpoff_reginfo);
+    }
+#endif
     if (arm_feature(env, ARM_FEATURE_VAPA)) {
         ARMCPRegInfo vapa_cp_reginfo[] = {
             { .name = "PAR", .cp = 15, .crn = 7, .crm = 4, .opc1 = 0, .opc2 = 0,
diff --git a/target/arm/trace-events b/target/arm/trace-events
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/trace-events
+++ b/target/arm/trace-events
@@ -XXX,XX +XXX,XX @@ arm_gt_tval_write(int timer, uint64_t value) "gt_tval_write: timer %d value 0x%"
 arm_gt_ctl_write(int timer, uint64_t value) "gt_ctl_write: timer %d value 0x%" PRIx64
 arm_gt_imask_toggle(int timer) "gt_ctl_write: timer %d IMASK toggle"
 arm_gt_cntvoff_write(uint64_t value) "gt_cntvoff_write: value 0x%" PRIx64
+arm_gt_cntpoff_write(uint64_t value) "gt_cntpoff_write: value 0x%" PRIx64
 arm_gt_update_irq(int timer, int irqstate) "gt_update_irq: timer %d irqstate %d"
 
 # kvm.c
-- 
2.34.1

Enable all FEAT_ECV features on the 'max' CPU.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20240301183219.2424889-9-peter.maydell@linaro.org
---
 docs/system/arm/emulation.rst | 1 +
 target/arm/tcg/cpu64.c        | 1 +
 2 files changed, 2 insertions(+)

diff --git a/docs/system/arm/emulation.rst b/docs/system/arm/emulation.rst
index XXXXXXX..XXXXXXX 100644
--- a/docs/system/arm/emulation.rst
+++ b/docs/system/arm/emulation.rst
@@ -XXX,XX +XXX,XX @@ the following architecture extensions:
 - FEAT_DotProd (Advanced SIMD dot product instructions)
 - FEAT_DoubleFault (Double Fault Extension)
 - FEAT_E0PD (Preventing EL0 access to halves of address maps)
+- FEAT_ECV (Enhanced Counter Virtualization)
 - FEAT_EPAC (Enhanced pointer authentication)
 - FEAT_ETS (Enhanced Translation Synchronization)
 - FEAT_EVT (Enhanced Virtualization Traps)
diff --git a/target/arm/tcg/cpu64.c b/target/arm/tcg/cpu64.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/tcg/cpu64.c
+++ b/target/arm/tcg/cpu64.c
@@ -XXX,XX +XXX,XX @@ void aarch64_max_tcg_initfn(Object *obj)
     t = FIELD_DP64(t, ID_AA64MMFR0, TGRAN64_2, 2); /* 64k stage2 supported */
     t = FIELD_DP64(t, ID_AA64MMFR0, TGRAN4_2, 2);  /*  4k stage2 supported */
     t = FIELD_DP64(t, ID_AA64MMFR0, FGT, 1);       /* FEAT_FGT */
+    t = FIELD_DP64(t, ID_AA64MMFR0, ECV, 2);       /* FEAT_ECV */
     cpu->isar.id_aa64mmfr0 = t;
 
     t = cpu->isar.id_aa64mmfr1;
-- 
2.34.1

From: Inès Varhol <ines.varhol@telecom-paris.fr>

Features supported :
- the 8 STM32L4x5 GPIOs are initialized with their reset values
    (except IDR, see below)
- input mode : setting a pin in input mode "externally" (using input
    irqs) results in an out irq (transmitted to SYSCFG)
- output mode : setting a bit in ODR sets the corresponding out irq
    (if this line is configured in output mode)
- pull-up, pull-down
- push-pull, open-drain

Difference with the real GPIOs :
- Alternate Function and Analog mode aren't implemented :
    pins in AF/Analog behave like pins in input mode
- floating pins stay at their last value
- register IDR reset values differ from the real one :
    values are coherent with the other registers reset values
    and the fact that AF/Analog modes aren't implemented
- setting I/O output speed isn't supported
- locking port bits isn't supported
- ADC function isn't supported
- GPIOH has 16 pins instead of 2 pins
- writing to registers LCKR, AFRL, AFRH and ASCR is ineffective

Signed-off-by: Arnaud Minier <arnaud.minier@telecom-paris.fr>
Signed-off-by: Inès Varhol <ines.varhol@telecom-paris.fr>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 20240305210444.310665-2-ines.varhol@telecom-paris.fr
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 MAINTAINERS                        |   1 +
 docs/system/arm/b-l475e-iot01a.rst |   2 +-
 include/hw/gpio/stm32l4x5_gpio.h   |  70 +++++
 hw/gpio/stm32l4x5_gpio.c           | 477 +++++++++++++++++++++++++++++
 hw/gpio/Kconfig                    |   3 +
 hw/gpio/meson.build                |   1 +
 hw/gpio/trace-events               |   6 +
 7 files changed, 559 insertions(+), 1 deletion(-)
 create mode 100644 include/hw/gpio/stm32l4x5_gpio.h
 create mode 100644 hw/gpio/stm32l4x5_gpio.c

diff --git a/MAINTAINERS b/MAINTAINERS
index XXXXXXX..XXXXXXX 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -XXX,XX +XXX,XX @@ F: hw/arm/stm32l4x5_soc.c
 F: hw/misc/stm32l4x5_exti.c
 F: hw/misc/stm32l4x5_syscfg.c
 F: hw/misc/stm32l4x5_rcc.c
+F: hw/gpio/stm32l4x5_gpio.c
 F: include/hw/*/stm32l4x5_*.h
 
 B-L475E-IOT01A IoT Node
diff --git a/docs/system/arm/b-l475e-iot01a.rst b/docs/system/arm/b-l475e-iot01a.rst
index XXXXXXX..XXXXXXX 100644
--- a/docs/system/arm/b-l475e-iot01a.rst
+++ b/docs/system/arm/b-l475e-iot01a.rst
@@ -XXX,XX +XXX,XX @@ Currently B-L475E-IOT01A machine's only supports the following devices:
 - STM32L4x5 EXTI (Extended interrupts and events controller)
 - STM32L4x5 SYSCFG (System configuration controller)
 - STM32L4x5 RCC (Reset and clock control)
+- STM32L4x5 GPIOs (General-purpose I/Os)
 
 Missing devices
 """""""""""""""
@@ -XXX,XX +XXX,XX @@ Missing devices
 The B-L475E-IOT01A does *not* support the following devices:
 
 - Serial ports (UART)
-- General-purpose I/Os (GPIO)
 - Analog to Digital Converter (ADC)
 - SPI controller
 - Timer controller (TIMER)
diff --git a/include/hw/gpio/stm32l4x5_gpio.h b/include/hw/gpio/stm32l4x5_gpio.h
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/include/hw/gpio/stm32l4x5_gpio.h
@@ -XXX,XX +XXX,XX @@
+/*
+ * STM32L4x5 GPIO (General Purpose Input/Ouput)
+ *
+ * Copyright (c) 2024 Arnaud Minier <arnaud.minier@telecom-paris.fr>
+ * Copyright (c) 2024 Inès Varhol <ines.varhol@telecom-paris.fr>
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+/*
+ * The reference used is the STMicroElectronics RM0351 Reference manual
+ * for STM32L4x5 and STM32L4x6 advanced Arm ® -based 32-bit MCUs.
+ * https://www.st.com/en/microcontrollers-microprocessors/stm32l4x5/documentation.html
+ */
+
+#ifndef HW_STM32L4X5_GPIO_H
+#define HW_STM32L4X5_GPIO_H
+
+#include "hw/sysbus.h"
+#include "qom/object.h"
+
+#define TYPE_STM32L4X5_GPIO "stm32l4x5-gpio"
+OBJECT_DECLARE_SIMPLE_TYPE(Stm32l4x5GpioState, STM32L4X5_GPIO)
+
+#define GPIO_NUM_PINS 16
+
+struct Stm32l4x5GpioState {
+    SysBusDevice parent_obj;
+
+    MemoryRegion mmio;
+
+    /* GPIO registers */
+    uint32_t moder;
+    uint32_t otyper;
+    uint32_t ospeedr;
+    uint32_t pupdr;
+    uint32_t idr;
+    uint32_t odr;
+    uint32_t lckr;
+    uint32_t afrl;
+    uint32_t afrh;
+    uint32_t ascr;
+
+    /* GPIO registers reset values */
+    uint32_t moder_reset;
+    uint32_t ospeedr_reset;
+    uint32_t pupdr_reset;
+
+    /*
+     * External driving of pins.
+     * The pins can be set externally through the device
+     * anonymous input GPIOs lines under certain conditions.
+     * The pin must not be in push-pull output mode,
+     * and can't be set high in open-drain mode.
+     * Pins driven externally and configured to
+     * output mode will in general be "disconnected"
+     * (see `get_gpio_pinmask_to_disconnect()`)
+     */
+    uint16_t disconnected_pins;
+    uint16_t pins_connected_high;
+
+    char *name;
+    Clock *clk;
+    qemu_irq pin[GPIO_NUM_PINS];
+};
+
+#endif
diff --git a/hw/gpio/stm32l4x5_gpio.c b/hw/gpio/stm32l4x5_gpio.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/hw/gpio/stm32l4x5_gpio.c
@@ -XXX,XX +XXX,XX @@
+/*
+ * STM32L4x5 GPIO (General Purpose Input/Ouput)
+ *
+ * Copyright (c) 2024 Arnaud Minier <arnaud.minier@telecom-paris.fr>
+ * Copyright (c) 2024 Inès Varhol <ines.varhol@telecom-paris.fr>
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+/*
+ * The reference used is the STMicroElectronics RM0351 Reference manual
+ * for STM32L4x5 and STM32L4x6 advanced Arm ® -based 32-bit MCUs.
+ * https://www.st.com/en/microcontrollers-microprocessors/stm32l4x5/documentation.html
+ */
+
+#include "qemu/osdep.h"
+#include "qemu/log.h"
+#include "hw/gpio/stm32l4x5_gpio.h"
+#include "hw/irq.h"
+#include "hw/qdev-clock.h"
+#include "hw/qdev-properties.h"
+#include "qapi/visitor.h"
+#include "qapi/error.h"
+#include "migration/vmstate.h"
+#include "trace.h"
+
+#define GPIO_MODER 0x00
+#define GPIO_OTYPER 0x04
+#define GPIO_OSPEEDR 0x08
+#define GPIO_PUPDR 0x0C
+#define GPIO_IDR 0x10
+#define GPIO_ODR 0x14
+#define GPIO_BSRR 0x18
+#define GPIO_LCKR 0x1C
+#define GPIO_AFRL 0x20
+#define GPIO_AFRH 0x24
+#define GPIO_BRR 0x28
+#define GPIO_ASCR 0x2C
+
+/* 0b11111111_11111111_00000000_00000000 */
+#define RESERVED_BITS_MASK 0xFFFF0000
+
+static void update_gpio_idr(Stm32l4x5GpioState *s);
+
+static bool is_pull_up(Stm32l4x5GpioState *s, unsigned pin)
+{
+    return extract32(s->pupdr, 2 * pin, 2) == 1;
+}
+
+static bool is_pull_down(Stm32l4x5GpioState *s, unsigned pin)
+{
+    return extract32(s->pupdr, 2 * pin, 2) == 2;
+}
+
+static bool is_output(Stm32l4x5GpioState *s, unsigned pin)
+{
+    return extract32(s->moder, 2 * pin, 2) == 1;
+}
+
+static bool is_open_drain(Stm32l4x5GpioState *s, unsigned pin)
+{
+    return extract32(s->otyper, pin, 1) == 1;
+}
+
+static bool is_push_pull(Stm32l4x5GpioState *s, unsigned pin)
+{
+    return extract32(s->otyper, pin, 1) == 0;
+}
+
+static void stm32l4x5_gpio_reset_hold(Object *obj)
+{
+    Stm32l4x5GpioState *s = STM32L4X5_GPIO(obj);
+
+    s->moder = s->moder_reset;
+    s->otyper = 0x00000000;
+    s->ospeedr = s->ospeedr_reset;
+    s->pupdr = s->pupdr_reset;
+    s->idr = 0x00000000;
+    s->odr = 0x00000000;
+    s->lckr = 0x00000000;
+    s->afrl = 0x00000000;
+    s->afrh = 0x00000000;
+    s->ascr = 0x00000000;
+
+    s->disconnected_pins = 0xFFFF;
+    s->pins_connected_high = 0x0000;
+    update_gpio_idr(s);
+}
+
+static void stm32l4x5_gpio_set(void *opaque, int line, int level)
+{
+    Stm32l4x5GpioState *s = opaque;
+    /*
+     * The pin isn't set if line is configured in output mode
+     * except if level is 0 and the output is open-drain.
+     * This way there will be no short-circuit prone situations.
+     */
+    if (is_output(s, line) && !(is_open_drain(s, line) && (level == 0))) {
+        qemu_log_mask(LOG_GUEST_ERROR, "Line %d can't be driven externally\n",
+                      line);
+        return;
+    }
+
+    s->disconnected_pins &= ~(1 << line);
+    if (level) {
+        s->pins_connected_high |= (1 << line);
+    } else {
+        s->pins_connected_high &= ~(1 << line);
+    }
+    trace_stm32l4x5_gpio_pins(s->name, s->disconnected_pins,
+                              s->pins_connected_high);
+    update_gpio_idr(s);
+}
+
+
+static void update_gpio_idr(Stm32l4x5GpioState *s)
+{
+    uint32_t new_idr_mask = 0;
+    uint32_t new_idr = s->odr;
+    uint32_t old_idr = s->idr;
+    int new_pin_state, old_pin_state;
+
+    for (int i = 0; i < GPIO_NUM_PINS; i++) {
+        if (is_output(s, i)) {
+            if (is_push_pull(s, i)) {
+                new_idr_mask |= (1 << i);
+            } else if (!(s->odr & (1 << i))) {
+                /* open-drain ODR 0 */
+                new_idr_mask |= (1 << i);
+            /* open-drain ODR 1 */
+            } else if (!(s->disconnected_pins & (1 << i)) &&
+                       !(s->pins_connected_high & (1 << i))) {
+                /* open-drain ODR 1 with pin connected low */
+                new_idr_mask |= (1 << i);
+                new_idr &= ~(1 << i);
+            /* open-drain ODR 1 with unactive pin */
+            } else if (is_pull_up(s, i)) {
+                new_idr_mask |= (1 << i);
+            } else if (is_pull_down(s, i)) {
+                new_idr_mask |= (1 << i);
+                new_idr &= ~(1 << i);
+            }
+            /*
+             * The only case left is for open-drain ODR 1
+             * with unactive pin without pull-up or pull-down :
+             * the value is floating.
+             */
+        /* input or analog mode with connected pin */
+        } else if (!(s->disconnected_pins & (1 << i))) {
+            if (s->pins_connected_high & (1 << i)) {
+                /* pin high */
+                new_idr_mask |= (1 << i);
+                new_idr |= (1 << i);
+            } else {
+                /* pin low */
+                new_idr_mask |= (1 << i);
+                new_idr &= ~(1 << i);
+            }
+        /* input or analog mode with disconnected pin */
+        } else {
+            if (is_pull_up(s, i)) {
+                /* pull-up */
+                new_idr_mask |= (1 << i);
+                new_idr |= (1 << i);
+            } else if (is_pull_down(s, i)) {
+                /* pull-down */
+                new_idr_mask |= (1 << i);
+                new_idr &= ~(1 << i);
+            }
+            /*
+             * The only case left is for a disconnected pin
+             * without pull-up or pull-down :
+             * the value is floating.
+             */
+        }
+    }
+
+    s->idr = (old_idr & ~new_idr_mask) | (new_idr & new_idr_mask);
+    trace_stm32l4x5_gpio_update_idr(s->name, old_idr, s->idr);
+
+    for (int i = 0; i < GPIO_NUM_PINS; i++) {
+        if (new_idr_mask & (1 << i)) {
+            new_pin_state = (new_idr & (1 << i)) > 0;
+            old_pin_state = (old_idr & (1 << i)) > 0;
+            if (new_pin_state > old_pin_state) {
+                qemu_irq_raise(s->pin[i]);
+            } else if (new_pin_state < old_pin_state) {
+                qemu_irq_lower(s->pin[i]);
+            }
+        }
+    }
+}
+
+/*
+ * Return mask of pins that are both configured in output
+ * mode and externally driven (except pins in open-drain
+ * mode externally set to 0).
+ */
+static uint32_t get_gpio_pinmask_to_disconnect(Stm32l4x5GpioState *s)
+{
+    uint32_t pins_to_disconnect = 0;
+    for (int i = 0; i < GPIO_NUM_PINS; i++) {
+        /* for each connected pin in output mode */
+        if (!(s->disconnected_pins & (1 << i)) && is_output(s, i)) {
+            /* if either push-pull or high level */
+            if (is_push_pull(s, i) || s->pins_connected_high & (1 << i)) {
+                pins_to_disconnect |= (1 << i);
+                qemu_log_mask(LOG_GUEST_ERROR,
+                              "Line %d can't be driven externally\n",
+                              i);
+            }
+        }
+    }
+    return pins_to_disconnect;
+}
+
+/*
+ * Set field `disconnected_pins` and call `update_gpio_idr()`
+ */
+static void disconnect_gpio_pins(Stm32l4x5GpioState *s, uint16_t lines)
+{
+    s->disconnected_pins |= lines;
+    trace_stm32l4x5_gpio_pins(s->name, s->disconnected_pins,
+                              s->pins_connected_high);
+    update_gpio_idr(s);
+}
+
+static void disconnected_pins_set(Object *obj, Visitor *v,
+    const char *name, void *opaque, Error **errp)
+{
+    Stm32l4x5GpioState *s = STM32L4X5_GPIO(obj);
+    uint16_t value;
+    if (!visit_type_uint16(v, name, &value, errp)) {
+        return;
+    }
+    disconnect_gpio_pins(s, value);
+}
+
+static void disconnected_pins_get(Object *obj, Visitor *v,
+    const char *name, void *opaque, Error **errp)
+{
+    visit_type_uint16(v, name, (uint16_t *)opaque, errp);
+}
+
+static void clock_freq_get(Object *obj, Visitor *v,
+    const char *name, void *opaque, Error **errp)
+{
+    Stm32l4x5GpioState *s = STM32L4X5_GPIO(obj);
+    uint32_t clock_freq_hz = clock_get_hz(s->clk);
+    visit_type_uint32(v, name, &clock_freq_hz, errp);
+}
+
+static void stm32l4x5_gpio_write(void *opaque, hwaddr addr,
+                                 uint64_t val64, unsigned int size)
+{
+    Stm32l4x5GpioState *s = opaque;
+
+    uint32_t value = val64;
+    trace_stm32l4x5_gpio_write(s->name, addr, val64);
+
+    switch (addr) {
+    case GPIO_MODER:
+        s->moder = value;
+        disconnect_gpio_pins(s, get_gpio_pinmask_to_disconnect(s));
+        qemu_log_mask(LOG_UNIMP,
+                      "%s: Analog and AF modes aren't supported\n\
+                       Analog and AF mode behave like input mode\n",
+                      __func__);
+        return;
+    case GPIO_OTYPER:
+        s->otyper = value & ~RESERVED_BITS_MASK;
+        disconnect_gpio_pins(s, get_gpio_pinmask_to_disconnect(s));
+        return;
+    case GPIO_OSPEEDR:
+        qemu_log_mask(LOG_UNIMP,
+                      "%s: Changing I/O output speed isn't supported\n\
+                       I/O speed is already maximal\n",
+                      __func__);
+        s->ospeedr = value;
+        return;
+    case GPIO_PUPDR:
+        s->pupdr = value;
+        update_gpio_idr(s);
+        return;
+    case GPIO_IDR:
+        qemu_log_mask(LOG_UNIMP,
+                      "%s: GPIO->IDR is read-only\n",
+                      __func__);
+        return;
+    case GPIO_ODR:
+        s->odr = value & ~RESERVED_BITS_MASK;
+        update_gpio_idr(s);
+        return;
+    case GPIO_BSRR: {
+        uint32_t bits_to_reset = (value & RESERVED_BITS_MASK) >> GPIO_NUM_PINS;
+        uint32_t bits_to_set = value & ~RESERVED_BITS_MASK;
+        /* If both BSx and BRx are set, BSx has priority.*/
+        s->odr &= ~bits_to_reset;
+        s->odr |= bits_to_set;
+        update_gpio_idr(s);
+        return;
+    }
+    case GPIO_LCKR:
+        qemu_log_mask(LOG_UNIMP,
+                      "%s: Locking port bits configuration isn't supported\n",
+                      __func__);
+        s->lckr = value & ~RESERVED_BITS_MASK;
+        return;
+    case GPIO_AFRL:
+        qemu_log_mask(LOG_UNIMP,
+                      "%s: Alternate functions aren't supported\n",
+                      __func__);
+        s->afrl = value;
+        return;
+    case GPIO_AFRH:
+        qemu_log_mask(LOG_UNIMP,
+                      "%s: Alternate functions aren't supported\n",
+                      __func__);
+        s->afrh = value;
+        return;
+    case GPIO_BRR: {
+        uint32_t bits_to_reset = value & ~RESERVED_BITS_MASK;
+        s->odr &= ~bits_to_reset;
+        update_gpio_idr(s);
+        return;
+    }
+    case GPIO_ASCR:
+        qemu_log_mask(LOG_UNIMP,
+                      "%s: ADC function isn't supported\n",
+                      __func__);
+        s->ascr = value & ~RESERVED_BITS_MASK;
+        return;
+    default:
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "%s: Bad offset 0x%" HWADDR_PRIx "\n", __func__, addr);
+    }
+}
+
+static uint64_t stm32l4x5_gpio_read(void *opaque, hwaddr addr,
+                                    unsigned int size)
+{
+    Stm32l4x5GpioState *s = opaque;
+
+    trace_stm32l4x5_gpio_read(s->name, addr);
+
+    switch (addr) {
+    case GPIO_MODER:
+        return s->moder;
+    case GPIO_OTYPER:
+        return s->otyper;
+    case GPIO_OSPEEDR:
+        return s->ospeedr;
+    case GPIO_PUPDR:
+        return s->pupdr;
+    case GPIO_IDR:
+        return s->idr;
+    case GPIO_ODR:
+        return s->odr;
+    case GPIO_BSRR:
+        return 0;
+    case GPIO_LCKR:
+        return s->lckr;
+    case GPIO_AFRL:
+        return s->afrl;
+    case GPIO_AFRH:
+        return s->afrh;
+    case GPIO_BRR:
+        return 0;
+    case GPIO_ASCR:
+        return s->ascr;
+    default:
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "%s: Bad offset 0x%" HWADDR_PRIx "\n", __func__, addr);
+        return 0;
+    }
+}
+
+static const MemoryRegionOps stm32l4x5_gpio_ops = {
+    .read = stm32l4x5_gpio_read,
+    .write = stm32l4x5_gpio_write,
+    .endianness = DEVICE_NATIVE_ENDIAN,
+    .impl = {
+        .min_access_size = 4,
+        .max_access_size = 4,
+        .unaligned = false,
+    },
+    .valid = {
+        .min_access_size = 4,
+        .max_access_size = 4,
+        .unaligned = false,
+    },
+};
+
+static void stm32l4x5_gpio_init(Object *obj)
+{
+    Stm32l4x5GpioState *s = STM32L4X5_GPIO(obj);
+
+    memory_region_init_io(&s->mmio, obj, &stm32l4x5_gpio_ops, s,
+                          TYPE_STM32L4X5_GPIO, 0x400);
+
+    sysbus_init_mmio(SYS_BUS_DEVICE(obj), &s->mmio);
+
+    qdev_init_gpio_out(DEVICE(obj), s->pin, GPIO_NUM_PINS);
+    qdev_init_gpio_in(DEVICE(obj), stm32l4x5_gpio_set, GPIO_NUM_PINS);
+
+    s->clk = qdev_init_clock_in(DEVICE(s), "clk", NULL, s, 0);
+
+    object_property_add(obj, "disconnected-pins", "uint16",
+                        disconnected_pins_get, disconnected_pins_set,
+                        NULL, &s->disconnected_pins);
+    object_property_add(obj, "clock-freq-hz", "uint32",
+                        clock_freq_get, NULL, NULL, NULL);
+}
+
+static void stm32l4x5_gpio_realize(DeviceState *dev, Error **errp)
+{
+    Stm32l4x5GpioState *s = STM32L4X5_GPIO(dev);
+    if (!clock_has_source(s->clk)) {
+        error_setg(errp, "GPIO: clk input must be connected");
+        return;
+    }
+}
+
+static const VMStateDescription vmstate_stm32l4x5_gpio = {
+    .name = TYPE_STM32L4X5_GPIO,
+    .version_id = 1,
+    .minimum_version_id = 1,
+    .fields = (VMStateField[]){
+        VMSTATE_UINT32(moder, Stm32l4x5GpioState),
+        VMSTATE_UINT32(otyper, Stm32l4x5GpioState),
+        VMSTATE_UINT32(ospeedr, Stm32l4x5GpioState),
+        VMSTATE_UINT32(pupdr, Stm32l4x5GpioState),
+        VMSTATE_UINT32(idr, Stm32l4x5GpioState),
+        VMSTATE_UINT32(odr, Stm32l4x5GpioState),
+        VMSTATE_UINT32(lckr, Stm32l4x5GpioState),
+        VMSTATE_UINT32(afrl, Stm32l4x5GpioState),
+        VMSTATE_UINT32(afrh, Stm32l4x5GpioState),
+        VMSTATE_UINT32(ascr, Stm32l4x5GpioState),
+        VMSTATE_UINT16(disconnected_pins, Stm32l4x5GpioState),
+        VMSTATE_UINT16(pins_connected_high, Stm32l4x5GpioState),
+        VMSTATE_END_OF_LIST()
+    }
+};
+
+static Property stm32l4x5_gpio_properties[] = {
+    DEFINE_PROP_STRING("name", Stm32l4x5GpioState, name),
+    DEFINE_PROP_UINT32("mode-reset", Stm32l4x5GpioState, moder_reset, 0),
+    DEFINE_PROP_UINT32("ospeed-reset", Stm32l4x5GpioState, ospeedr_reset, 0),
+    DEFINE_PROP_UINT32("pupd-reset", Stm32l4x5GpioState, pupdr_reset, 0),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void stm32l4x5_gpio_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    ResettableClass *rc = RESETTABLE_CLASS(klass);
+
+    device_class_set_props(dc, stm32l4x5_gpio_properties);
+    dc->vmsd = &vmstate_stm32l4x5_gpio;
+    dc->realize = stm32l4x5_gpio_realize;
+    rc->phases.hold = stm32l4x5_gpio_reset_hold;
+}
+
+static const TypeInfo stm32l4x5_gpio_types[] = {
+    {
+        .name = TYPE_STM32L4X5_GPIO,
+        .parent = TYPE_SYS_BUS_DEVICE,
+        .instance_size = sizeof(Stm32l4x5GpioState),
+        .instance_init = stm32l4x5_gpio_init,
+        .class_init = stm32l4x5_gpio_class_init,
+    },
+};
+
+DEFINE_TYPES(stm32l4x5_gpio_types)
diff --git a/hw/gpio/Kconfig b/hw/gpio/Kconfig
index XXXXXXX..XXXXXXX 100644
--- a/hw/gpio/Kconfig
+++ b/hw/gpio/Kconfig
@@ -XXX,XX +XXX,XX @@ config GPIO_PWR
 
 config SIFIVE_GPIO
     bool
+
+config STM32L4X5_GPIO
+    bool
diff --git a/hw/gpio/meson.build b/hw/gpio/meson.build
index XXXXXXX..XXXXXXX 100644
--- a/hw/gpio/meson.build
+++ b/hw/gpio/meson.build
@@ -XXX,XX +XXX,XX @@ system_ss.add(when: 'CONFIG_RASPI', if_true: files(
     'bcm2835_gpio.c',
     'bcm2838_gpio.c'
 ))
+system_ss.add(when: 'CONFIG_STM32L4X5_SOC', if_true: files('stm32l4x5_gpio.c'))
 system_ss.add(when: 'CONFIG_ASPEED_SOC', if_true: files('aspeed_gpio.c'))
 system_ss.add(when: 'CONFIG_SIFIVE_GPIO', if_true: files('sifive_gpio.c'))
diff --git a/hw/gpio/trace-events b/hw/gpio/trace-events
index XXXXXXX..XXXXXXX 100644
--- a/hw/gpio/trace-events
+++ b/hw/gpio/trace-events
@@ -XXX,XX +XXX,XX @@ sifive_gpio_update_output_irq(int64_t line, int64_t value) "line %" PRIi64 " val
 # aspeed_gpio.c
 aspeed_gpio_read(uint64_t offset, uint64_t value) "offset: 0x%" PRIx64 " value 0x%" PRIx64
 aspeed_gpio_write(uint64_t offset, uint64_t value) "offset: 0x%" PRIx64 " value 0x%" PRIx64
+
+# stm32l4x5_gpio.c
+stm32l4x5_gpio_read(char *gpio, uint64_t addr) "GPIO%s addr: 0x%" PRIx64 " "
+stm32l4x5_gpio_write(char *gpio, uint64_t addr, uint64_t data) "GPIO%s addr: 0x%" PRIx64 " val: 0x%" PRIx64 ""
+stm32l4x5_gpio_update_idr(char *gpio, uint32_t old_idr, uint32_t new_idr) "GPIO%s from: 0x%x to: 0x%x"
+stm32l4x5_gpio_pins(char *gpio, uint16_t disconnected, uint16_t high) "GPIO%s disconnected pins: 0x%x levels: 0x%x"
-- 
2.34.1

From: Inès Varhol <ines.varhol@telecom-paris.fr>

Signed-off-by: Arnaud Minier <arnaud.minier@telecom-paris.fr>
Signed-off-by: Inès Varhol <ines.varhol@telecom-paris.fr>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 20240305210444.310665-3-ines.varhol@telecom-paris.fr
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 include/hw/arm/stm32l4x5_soc.h     |  2 +
 include/hw/gpio/stm32l4x5_gpio.h   |  1 +
 include/hw/misc/stm32l4x5_syscfg.h |  3 +-
 hw/arm/stm32l4x5_soc.c             | 71 +++++++++++++++++++++++-------
 hw/misc/stm32l4x5_syscfg.c         |  1 +
 hw/arm/Kconfig                     |  3 +-
 6 files changed, 63 insertions(+), 18 deletions(-)

diff --git a/include/hw/arm/stm32l4x5_soc.h b/include/hw/arm/stm32l4x5_soc.h
index XXXXXXX..XXXXXXX 100644
--- a/include/hw/arm/stm32l4x5_soc.h
+++ b/include/hw/arm/stm32l4x5_soc.h
@@ -XXX,XX +XXX,XX @@
 #include "hw/misc/stm32l4x5_syscfg.h"
 #include "hw/misc/stm32l4x5_exti.h"
 #include "hw/misc/stm32l4x5_rcc.h"
+#include "hw/gpio/stm32l4x5_gpio.h"
 #include "qom/object.h"
 
 #define TYPE_STM32L4X5_SOC "stm32l4x5-soc"
@@ -XXX,XX +XXX,XX @@ struct Stm32l4x5SocState {
     OrIRQState exti_or_gates[NUM_EXTI_OR_GATES];
     Stm32l4x5SyscfgState syscfg;
     Stm32l4x5RccState rcc;
+    Stm32l4x5GpioState gpio[NUM_GPIOS];
 
     MemoryRegion sram1;
     MemoryRegion sram2;
diff --git a/include/hw/gpio/stm32l4x5_gpio.h b/include/hw/gpio/stm32l4x5_gpio.h
index XXXXXXX..XXXXXXX 100644
--- a/include/hw/gpio/stm32l4x5_gpio.h
+++ b/include/hw/gpio/stm32l4x5_gpio.h
@@ -XXX,XX +XXX,XX @@
 #define TYPE_STM32L4X5_GPIO "stm32l4x5-gpio"
 OBJECT_DECLARE_SIMPLE_TYPE(Stm32l4x5GpioState, STM32L4X5_GPIO)
 
+#define NUM_GPIOS 8
 #define GPIO_NUM_PINS 16
 
 struct Stm32l4x5GpioState {
diff --git a/include/hw/misc/stm32l4x5_syscfg.h b/include/hw/misc/stm32l4x5_syscfg.h
index XXXXXXX..XXXXXXX 100644
--- a/include/hw/misc/stm32l4x5_syscfg.h
+++ b/include/hw/misc/stm32l4x5_syscfg.h
@@ -XXX,XX +XXX,XX @@
 
 #include "hw/sysbus.h"
 #include "qom/object.h"
+#include "hw/gpio/stm32l4x5_gpio.h"
 
 #define TYPE_STM32L4X5_SYSCFG "stm32l4x5-syscfg"
 OBJECT_DECLARE_SIMPLE_TYPE(Stm32l4x5SyscfgState, STM32L4X5_SYSCFG)
 
-#define NUM_GPIOS 8
-#define GPIO_NUM_PINS 16
 #define SYSCFG_NUM_EXTICR 4
 
 struct Stm32l4x5SyscfgState {
diff --git a/hw/arm/stm32l4x5_soc.c b/hw/arm/stm32l4x5_soc.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/arm/stm32l4x5_soc.c
+++ b/hw/arm/stm32l4x5_soc.c
@@ -XXX,XX +XXX,XX @@
 #include "sysemu/sysemu.h"
 #include "hw/or-irq.h"
 #include "hw/arm/stm32l4x5_soc.h"
+#include "hw/gpio/stm32l4x5_gpio.h"
 #include "hw/qdev-clock.h"
 #include "hw/misc/unimp.h"
 
@@ -XXX,XX +XXX,XX @@ static const int exti_or_gate1_lines_in[EXTI_OR_GATE1_NUM_LINES_IN] = {
     16, 35, 36, 37, 38,
 };
 
+static const struct {
+    uint32_t addr;
+    uint32_t moder_reset;
+    uint32_t ospeedr_reset;
+    uint32_t pupdr_reset;
+} stm32l4x5_gpio_cfg[NUM_GPIOS] = {
+    { 0x48000000, 0xABFFFFFF, 0x0C000000, 0x64000000 },
+    { 0x48000400, 0xFFFFFEBF, 0x00000000, 0x00000100 },
+    { 0x48000800, 0xFFFFFFFF, 0x00000000, 0x00000000 },
+    { 0x48000C00, 0xFFFFFFFF, 0x00000000, 0x00000000 },
+    { 0x48001000, 0xFFFFFFFF, 0x00000000, 0x00000000 },
+    { 0x48001400, 0xFFFFFFFF, 0x00000000, 0x00000000 },
+    { 0x48001800, 0xFFFFFFFF, 0x00000000, 0x00000000 },
+    { 0x48001C00, 0x0000000F, 0x00000000, 0x00000000 },
+};
+
 static void stm32l4x5_soc_initfn(Object *obj)
 {
     Stm32l4x5SocState *s = STM32L4X5_SOC(obj);
@@ -XXX,XX +XXX,XX @@ static void stm32l4x5_soc_initfn(Object *obj)
     }
     object_initialize_child(obj, "syscfg", &s->syscfg, TYPE_STM32L4X5_SYSCFG);
     object_initialize_child(obj, "rcc", &s->rcc, TYPE_STM32L4X5_RCC);
+
+    for (unsigned i = 0; i < NUM_GPIOS; i++) {
+        g_autofree char *name = g_strdup_printf("gpio%c", 'a' + i);
+        object_initialize_child(obj, name, &s->gpio[i], TYPE_STM32L4X5_GPIO);
+    }
 }
 
 static void stm32l4x5_soc_realize(DeviceState *dev_soc, Error **errp)
@@ -XXX,XX +XXX,XX @@ static void stm32l4x5_soc_realize(DeviceState *dev_soc, Error **errp)
     Stm32l4x5SocState *s = STM32L4X5_SOC(dev_soc);
     const Stm32l4x5SocClass *sc = STM32L4X5_SOC_GET_CLASS(dev_soc);
     MemoryRegion *system_memory = get_system_memory();
-    DeviceState *armv7m;
+    DeviceState *armv7m, *dev;
     SysBusDevice *busdev;
+    uint32_t pin_index;
 
     if (!memory_region_init_rom(&s->flash, OBJECT(dev_soc), "flash",
                                 sc->flash_size, errp)) {
@@ -XXX,XX +XXX,XX @@ static void stm32l4x5_soc_realize(DeviceState *dev_soc, Error **errp)
         return;
     }
 
+    /* GPIOs */
+    for (unsigned i = 0; i < NUM_GPIOS; i++) {
+        g_autofree char *name = g_strdup_printf("%c", 'A' + i);
+        dev = DEVICE(&s->gpio[i]);
+        qdev_prop_set_string(dev, "name", name);
+        qdev_prop_set_uint32(dev, "mode-reset",
+                             stm32l4x5_gpio_cfg[i].moder_reset);
+        qdev_prop_set_uint32(dev, "ospeed-reset",
+                             stm32l4x5_gpio_cfg[i].ospeedr_reset);
+        qdev_prop_set_uint32(dev, "pupd-reset",
+                            stm32l4x5_gpio_cfg[i].pupdr_reset);
+        busdev = SYS_BUS_DEVICE(&s->gpio[i]);
+        g_free(name);
+        name = g_strdup_printf("gpio%c-out", 'a' + i);
+        qdev_connect_clock_in(DEVICE(&s->gpio[i]), "clk",
+            qdev_get_clock_out(DEVICE(&(s->rcc)), name));
+        if (!sysbus_realize(busdev, errp)) {
+            return;
+        }
+        sysbus_mmio_map(busdev, 0, stm32l4x5_gpio_cfg[i].addr);
+    }
+
     /* System configuration controller */
     busdev = SYS_BUS_DEVICE(&s->syscfg);
     if (!sysbus_realize(busdev, errp)) {
         return;
     }
     sysbus_mmio_map(busdev, 0, SYSCFG_ADDR);
-    /*
-     * TODO: when the GPIO device is implemented, connect it
-     * to SYCFG using `qdev_connect_gpio_out`, NUM_GPIOS and
-     * GPIO_NUM_PINS.
-     */
+
+    for (unsigned i = 0; i < NUM_GPIOS; i++) {
+        for (unsigned j = 0; j < GPIO_NUM_PINS; j++) {
+            pin_index = GPIO_NUM_PINS * i + j;
+            qdev_connect_gpio_out(DEVICE(&s->gpio[i]), j,
+                                  qdev_get_gpio_in(DEVICE(&s->syscfg),
+                                  pin_index));
+        }
+    }
 
     /* EXTI device */
     busdev = SYS_BUS_DEVICE(&s->exti);
@@ -XXX,XX +XXX,XX @@ static void stm32l4x5_soc_realize(DeviceState *dev_soc, Error **errp)
         }
     }
 
-    for (unsigned i = 0; i < 16; i++) {
+    for (unsigned i = 0; i < GPIO_NUM_PINS; i++) {
         qdev_connect_gpio_out(DEVICE(&s->syscfg), i,
                               qdev_get_gpio_in(DEVICE(&s->exti), i));
     }
@@ -XXX,XX +XXX,XX @@ static void stm32l4x5_soc_realize(DeviceState *dev_soc, Error **errp)
     /* RESERVED:    0x40024400, 0x7FDBC00 */
 
     /* AHB2 BUS */
-    create_unimplemented_device("GPIOA",     0x48000000, 0x400);
-    create_unimplemented_device("GPIOB",     0x48000400, 0x400);
-    create_unimplemented_device("GPIOC",     0x48000800, 0x400);
-    create_unimplemented_device("GPIOD",     0x48000C00, 0x400);
-    create_unimplemented_device("GPIOE",     0x48001000, 0x400);
-    create_unimplemented_device("GPIOF",     0x48001400, 0x400);
-    create_unimplemented_device("GPIOG",     0x48001800, 0x400);
-    create_unimplemented_device("GPIOH",     0x48001C00, 0x400);
     /* RESERVED:    0x48002000, 0x7FDBC00 */
     create_unimplemented_device("OTG_FS",    0x50000000, 0x40000);
     create_unimplemented_device("ADC",       0x50040000, 0x400);
diff --git a/hw/misc/stm32l4x5_syscfg.c b/hw/misc/stm32l4x5_syscfg.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/misc/stm32l4x5_syscfg.c
+++ b/hw/misc/stm32l4x5_syscfg.c
@@ -XXX,XX +XXX,XX @@
 #include "hw/irq.h"
 #include "migration/vmstate.h"
 #include "hw/misc/stm32l4x5_syscfg.h"
+#include "hw/gpio/stm32l4x5_gpio.h"
 
 #define SYSCFG_MEMRMP 0x00
 #define SYSCFG_CFGR1 0x04
diff --git a/hw/arm/Kconfig b/hw/arm/Kconfig
index XXXXXXX..XXXXXXX 100644
--- a/hw/arm/Kconfig
+++ b/hw/arm/Kconfig
@@ -XXX,XX +XXX,XX @@ config STM32L4X5_SOC
     bool
     select ARM_V7M
     select OR_IRQ
-    select STM32L4X5_SYSCFG
     select STM32L4X5_EXTI
+    select STM32L4X5_SYSCFG
     select STM32L4X5_RCC
+    select STM32L4X5_GPIO
 
 config XLNX_ZYNQMP_ARM
     bool
-- 
2.34.1

From: Inès Varhol <ines.varhol@telecom-paris.fr>

The testcase contains :
- `test_idr_reset_value()` :
Checks the reset values of MODER, OTYPER, PUPDR, ODR and IDR.
- `test_gpio_output_mode()` :
Checks that writing a bit in register ODR results in the corresponding
pin rising or lowering, if this pin is configured in output mode.
- `test_gpio_input_mode()` :
Checks that a input pin set high or low externally results
in the pin rising and lowering.
- `test_pull_up_pull_down()` :
Checks that a floating pin in pull-up/down mode is actually high/down.
- `test_push_pull()` :
Checks that a pin set externally is disconnected when configured in
push-pull output mode, and can't be set externally while in this mode.
- `test_open_drain()` :
Checks that a pin set externally high is disconnected when configured
in open-drain output mode, and can't be set high while in this mode.
- `test_bsrr_brr()` :
Checks that writing to BSRR and BRR has the desired result in ODR.
- `test_clock_enable()` :
Checks that GPIO clock is at the right frequency after enabling it.

Acked-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Arnaud Minier <arnaud.minier@telecom-paris.fr>
Signed-off-by: Inès Varhol <ines.varhol@telecom-paris.fr>
Message-id: 20240305210444.310665-4-ines.varhol@telecom-paris.fr
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 tests/qtest/stm32l4x5_gpio-test.c | 551 ++++++++++++++++++++++++++++++
 tests/qtest/meson.build           |   3 +-
 2 files changed, 553 insertions(+), 1 deletion(-)
 create mode 100644 tests/qtest/stm32l4x5_gpio-test.c

diff --git a/tests/qtest/stm32l4x5_gpio-test.c b/tests/qtest/stm32l4x5_gpio-test.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/tests/qtest/stm32l4x5_gpio-test.c
@@ -XXX,XX +XXX,XX @@
+/*
+ * QTest testcase for STM32L4x5_GPIO
+ *
+ * Copyright (c) 2024 Arnaud Minier <arnaud.minier@telecom-paris.fr>
+ * Copyright (c) 2024 Inès Varhol <ines.varhol@telecom-paris.fr>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+#include "qemu/osdep.h"
+#include "libqtest-single.h"
+
+#define GPIO_BASE_ADDR 0x48000000
+#define GPIO_SIZE      0x400
+#define NUM_GPIOS      8
+#define NUM_GPIO_PINS  16
+
+#define GPIO_A 0x48000000
+#define GPIO_B 0x48000400
+#define GPIO_C 0x48000800
+#define GPIO_D 0x48000C00
+#define GPIO_E 0x48001000
+#define GPIO_F 0x48001400
+#define GPIO_G 0x48001800
+#define GPIO_H 0x48001C00
+
+#define MODER 0x00
+#define OTYPER 0x04
+#define PUPDR 0x0C
+#define IDR 0x10
+#define ODR 0x14
+#define BSRR 0x18
+#define BRR 0x28
+
+#define MODER_INPUT 0
+#define MODER_OUTPUT 1
+
+#define PUPDR_NONE 0
+#define PUPDR_PULLUP 1
+#define PUPDR_PULLDOWN 2
+
+#define OTYPER_PUSH_PULL 0
+#define OTYPER_OPEN_DRAIN 1
+
+const uint32_t moder_reset[NUM_GPIOS] = {
+    0xABFFFFFF,
+    0xFFFFFEBF,
+    0xFFFFFFFF,
+    0xFFFFFFFF,
+    0xFFFFFFFF,
+    0xFFFFFFFF,
+    0xFFFFFFFF,
+    0x0000000F
+};
+
+const uint32_t pupdr_reset[NUM_GPIOS] = {
+    0x64000000,
+    0x00000100,
+    0x00000000,
+    0x00000000,
+    0x00000000,
+    0x00000000,
+    0x00000000,
+    0x00000000
+};
+
+const uint32_t idr_reset[NUM_GPIOS] = {
+    0x0000A000,
+    0x00000010,
+    0x00000000,
+    0x00000000,
+    0x00000000,
+    0x00000000,
+    0x00000000,
+    0x00000000
+};
+
+static uint32_t gpio_readl(unsigned int gpio, unsigned int offset)
+{
+    return readl(gpio + offset);
+}
+
+static void gpio_writel(unsigned int gpio, unsigned int offset, uint32_t value)
+{
+    writel(gpio + offset, value);
+}
+
+static void gpio_set_bit(unsigned int gpio, unsigned int reg,
+                         unsigned int pin, uint32_t value)
+{
+    uint32_t mask = 0xFFFFFFFF & ~(0x1 << pin);
+    gpio_writel(gpio, reg, (gpio_readl(gpio, reg) & mask) | value << pin);
+}
+
+static void gpio_set_2bits(unsigned int gpio, unsigned int reg,
+                           unsigned int pin, uint32_t value)
+{
+    uint32_t offset = 2 * pin;
+    uint32_t mask = 0xFFFFFFFF & ~(0x3 << offset);
+    gpio_writel(gpio, reg, (gpio_readl(gpio, reg) & mask) | value << offset);
+}
+
+static unsigned int get_gpio_id(uint32_t gpio_addr)
+{
+    return (gpio_addr - GPIO_BASE_ADDR) / GPIO_SIZE;
+}
+
+static void gpio_set_irq(unsigned int gpio, int num, int level)
+{
+    g_autofree char *name = g_strdup_printf("/machine/soc/gpio%c",
+                                            get_gpio_id(gpio) + 'a');
+    qtest_set_irq_in(global_qtest, name, NULL, num, level);
+}
+
+static void disconnect_all_pins(unsigned int gpio)
+{
+    g_autofree char *path = g_strdup_printf("/machine/soc/gpio%c",
+                                            get_gpio_id(gpio) + 'a');
+    QDict *r;
+
+    r = qtest_qmp(global_qtest, "{ 'execute': 'qom-set', 'arguments': "
+        "{ 'path': %s, 'property': 'disconnected-pins', 'value': %d } }",
+        path, 0xFFFF);
+    g_assert_false(qdict_haskey(r, "error"));
+    qobject_unref(r);
+}
+
+static uint32_t get_disconnected_pins(unsigned int gpio)
+{
+    g_autofree char *path = g_strdup_printf("/machine/soc/gpio%c",
+                                            get_gpio_id(gpio) + 'a');
+    uint32_t disconnected_pins = 0;
+    QDict *r;
+
+    r = qtest_qmp(global_qtest, "{ 'execute': 'qom-get', 'arguments':"
+        " { 'path': %s, 'property': 'disconnected-pins'} }", path);
+    g_assert_false(qdict_haskey(r, "error"));
+    disconnected_pins = qdict_get_int(r, "return");
+    qobject_unref(r);
+    return disconnected_pins;
+}
+
+static uint32_t reset(uint32_t gpio, unsigned int offset)
+{
+    switch (offset) {
+    case MODER:
+        return moder_reset[get_gpio_id(gpio)];
+    case PUPDR:
+        return pupdr_reset[get_gpio_id(gpio)];
+    case IDR:
+        return idr_reset[get_gpio_id(gpio)];
+    }
+    return 0x0;
+}
+
+static void system_reset(void)
+{
+    QDict *r;
+    r = qtest_qmp(global_qtest, "{'execute': 'system_reset'}");
+    g_assert_false(qdict_haskey(r, "error"));
+    qobject_unref(r);
+}
+
+static void test_idr_reset_value(void)
+{
+    /*
+     * Checks that the values in MODER, OTYPER, PUPDR and ODR
+     * after reset are correct, and that the value in IDR is
+     * coherent.
+     * Since AF and analog modes aren't implemented, IDR reset
+     * values aren't the same as with a real board.
+     *
+     * Register IDR contains the actual values of all GPIO pins.
+     * Its value depends on the pins' configuration
+     * (intput/output/analog : register MODER, push-pull/open-drain :
+     * register OTYPER, pull-up/pull-down/none : register PUPDR)
+     * and on the values stored in register ODR
+     * (in case the pin is in output mode).
+     */
+
+    gpio_writel(GPIO_A, MODER, 0xDEADBEEF);
+    gpio_writel(GPIO_A, ODR, 0xDEADBEEF);
+    gpio_writel(GPIO_A, OTYPER, 0xDEADBEEF);
+    gpio_writel(GPIO_A, PUPDR, 0xDEADBEEF);
+
+    gpio_writel(GPIO_B, MODER, 0xDEADBEEF);
+    gpio_writel(GPIO_B, ODR, 0xDEADBEEF);
+    gpio_writel(GPIO_B, OTYPER, 0xDEADBEEF);
+    gpio_writel(GPIO_B, PUPDR, 0xDEADBEEF);
+
+    gpio_writel(GPIO_C, MODER, 0xDEADBEEF);
+    gpio_writel(GPIO_C, ODR, 0xDEADBEEF);
+    gpio_writel(GPIO_C, OTYPER, 0xDEADBEEF);
+    gpio_writel(GPIO_C, PUPDR, 0xDEADBEEF);
+
+    gpio_writel(GPIO_H, MODER, 0xDEADBEEF);
+    gpio_writel(GPIO_H, ODR, 0xDEADBEEF);
+    gpio_writel(GPIO_H, OTYPER, 0xDEADBEEF);
+    gpio_writel(GPIO_H, PUPDR, 0xDEADBEEF);
+
+    system_reset();
+
+    uint32_t moder = gpio_readl(GPIO_A, MODER);
+    uint32_t odr = gpio_readl(GPIO_A, ODR);
+    uint32_t otyper = gpio_readl(GPIO_A, OTYPER);
+    uint32_t pupdr = gpio_readl(GPIO_A, PUPDR);
+    uint32_t idr = gpio_readl(GPIO_A, IDR);
+    /* 15: AF, 14: AF, 13: AF, 12: Analog ... */
+    /* here AF is the same as Analog and Input mode */
+    g_assert_cmphex(moder, ==, reset(GPIO_A, MODER));
+    g_assert_cmphex(odr, ==, reset(GPIO_A, ODR));
+    g_assert_cmphex(otyper, ==, reset(GPIO_A, OTYPER));
+    /* 15: pull-up, 14: pull-down, 13: pull-up, 12: neither ... */
+    g_assert_cmphex(pupdr, ==, reset(GPIO_A, PUPDR));
+    /* 15 : 1, 14: 0, 13: 1, 12 : reset value ... */
+    g_assert_cmphex(idr, ==, reset(GPIO_A, IDR));
+
+    moder = gpio_readl(GPIO_B, MODER);
+    odr = gpio_readl(GPIO_B, ODR);
+    otyper = gpio_readl(GPIO_B, OTYPER);
+    pupdr = gpio_readl(GPIO_B, PUPDR);
+    idr = gpio_readl(GPIO_B, IDR);
+    /* ... 5: Analog, 4: AF, 3: AF, 2: Analog ... */
+    /* here AF is the same as Analog and Input mode */
+    g_assert_cmphex(moder, ==, reset(GPIO_B, MODER));
+    g_assert_cmphex(odr, ==, reset(GPIO_B, ODR));
+    g_assert_cmphex(otyper, ==, reset(GPIO_B, OTYPER));
+    /* ... 5: neither, 4: pull-up, 3: neither ... */
+    g_assert_cmphex(pupdr, ==, reset(GPIO_B, PUPDR));
+    /* ... 5 : reset value, 4 : 1, 3 : reset value ... */
+    g_assert_cmphex(idr, ==, reset(GPIO_B, IDR));
+
+    moder = gpio_readl(GPIO_C, MODER);
+    odr = gpio_readl(GPIO_C, ODR);
+    otyper = gpio_readl(GPIO_C, OTYPER);
+    pupdr = gpio_readl(GPIO_C, PUPDR);
+    idr = gpio_readl(GPIO_C, IDR);
+    /* Analog, same as Input mode*/
+    g_assert_cmphex(moder, ==, reset(GPIO_C, MODER));
+    g_assert_cmphex(odr, ==, reset(GPIO_C, ODR));
+    g_assert_cmphex(otyper, ==, reset(GPIO_C, OTYPER));
+    /* no pull-up or pull-down */
+    g_assert_cmphex(pupdr, ==, reset(GPIO_C, PUPDR));
+    /* reset value */
+    g_assert_cmphex(idr, ==, reset(GPIO_C, IDR));
+
+    moder = gpio_readl(GPIO_H, MODER);
+    odr = gpio_readl(GPIO_H, ODR);
+    otyper = gpio_readl(GPIO_H, OTYPER);
+    pupdr = gpio_readl(GPIO_H, PUPDR);
+    idr = gpio_readl(GPIO_H, IDR);
+    /* Analog, same as Input mode */
+    g_assert_cmphex(moder, ==, reset(GPIO_H, MODER));
+    g_assert_cmphex(odr, ==, reset(GPIO_H, ODR));
+    g_assert_cmphex(otyper, ==, reset(GPIO_H, OTYPER));
+    /* no pull-up or pull-down */
+    g_assert_cmphex(pupdr, ==, reset(GPIO_H, PUPDR));
+    /* reset value */
+    g_assert_cmphex(idr, ==, reset(GPIO_H, IDR));
+}
+
+static void test_gpio_output_mode(const void *data)
+{
+    /*
+     * Checks that setting a bit in ODR sets the corresponding
+     * GPIO line high : it should set the right bit in IDR
+     * and send an irq to syscfg.
+     * Additionally, it checks that values written to ODR
+     * when not in output mode are stored and not discarded.
+     */
+    unsigned int pin = ((uint64_t)data) & 0xF;
+    uint32_t gpio = ((uint64_t)data) >> 32;
+    unsigned int gpio_id = get_gpio_id(gpio);
+
+    qtest_irq_intercept_in(global_qtest, "/machine/soc/syscfg");
+
+    /* Set a bit in ODR and check nothing happens */
+    gpio_set_bit(gpio, ODR, pin, 1);
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR));
+    g_assert_false(get_irq(gpio_id * NUM_GPIO_PINS + pin));
+
+    /* Configure the relevant line as output and check the pin is high */
+    gpio_set_2bits(gpio, MODER, pin, MODER_OUTPUT);
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) | (1 << pin));
+    g_assert_true(get_irq(gpio_id * NUM_GPIO_PINS + pin));
+
+    /* Reset the bit in ODR and check the pin is low */
+    gpio_set_bit(gpio, ODR, pin, 0);
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
+    g_assert_false(get_irq(gpio_id * NUM_GPIO_PINS + pin));
+
+    /* Clean the test */
+    gpio_writel(gpio, ODR, reset(gpio, ODR));
+    gpio_writel(gpio, MODER, reset(gpio, MODER));
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR));
+    g_assert_false(get_irq(gpio_id * NUM_GPIO_PINS + pin));
+}
+
+static void test_gpio_input_mode(const void *data)
+{
+    /*
+     * Test that setting a line high/low externally sets the
+     * corresponding GPIO line high/low : it should set the
+     * right bit in IDR and send an irq to syscfg.
+     */
+    unsigned int pin = ((uint64_t)data) & 0xF;
+    uint32_t gpio = ((uint64_t)data) >> 32;
+    unsigned int gpio_id = get_gpio_id(gpio);
+
+    qtest_irq_intercept_in(global_qtest, "/machine/soc/syscfg");
+
+    /* Configure a line as input, raise it, and check that the pin is high */
+    gpio_set_2bits(gpio, MODER, pin, MODER_INPUT);
+    gpio_set_irq(gpio, pin, 1);
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) | (1 << pin));
+    g_assert_true(get_irq(gpio_id * NUM_GPIO_PINS + pin));
+
+    /* Lower the line and check that the pin is low */
+    gpio_set_irq(gpio, pin, 0);
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
+    g_assert_false(get_irq(gpio_id * NUM_GPIO_PINS + pin));
+
+    /* Clean the test */
+    gpio_writel(gpio, MODER, reset(gpio, MODER));
+    disconnect_all_pins(gpio);
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR));
+}
+
+static void test_pull_up_pull_down(const void *data)
+{
+    /*
+     * Test that a floating pin with pull-up sets the pin
+     * high and vice-versa.
+     */
+    unsigned int pin = ((uint64_t)data) & 0xF;
+    uint32_t gpio = ((uint64_t)data) >> 32;
+    unsigned int gpio_id = get_gpio_id(gpio);
+
+    qtest_irq_intercept_in(global_qtest, "/machine/soc/syscfg");
+
+    /* Configure a line as input with pull-up, check the line is set high */
+    gpio_set_2bits(gpio, MODER, pin, MODER_INPUT);
+    gpio_set_2bits(gpio, PUPDR, pin, PUPDR_PULLUP);
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) | (1 << pin));
+    g_assert_true(get_irq(gpio_id * NUM_GPIO_PINS + pin));
+
+    /* Configure the line with pull-down, check the line is low */
+    gpio_set_2bits(gpio, PUPDR, pin, PUPDR_PULLDOWN);
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
+    g_assert_false(get_irq(gpio_id * NUM_GPIO_PINS + pin));
+
+    /* Clean the test */
+    gpio_writel(gpio, MODER, reset(gpio, MODER));
+    gpio_writel(gpio, PUPDR, reset(gpio, PUPDR));
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR));
+}
+
+static void test_push_pull(const void *data)
+{
+    /*
+     * Test that configuring a line in push-pull output mode
+     * disconnects the pin, that the pin can't be set or reset
+     * externally afterwards.
+     */
+    unsigned int pin = ((uint64_t)data) & 0xF;
+    uint32_t gpio = ((uint64_t)data) >> 32;
+    uint32_t gpio2 = GPIO_BASE_ADDR + (GPIO_H - gpio);
+
+    qtest_irq_intercept_in(global_qtest, "/machine/soc/syscfg");
+
+    /* Setting a line high externally, configuring it in push-pull output */
+    /* And checking the pin was disconnected */
+    gpio_set_irq(gpio, pin, 1);
+    gpio_set_2bits(gpio, MODER, pin, MODER_OUTPUT);
+    g_assert_cmphex(get_disconnected_pins(gpio), ==, 0xFFFF);
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
+
+    /* Setting a line low externally, configuring it in push-pull output */
+    /* And checking the pin was disconnected */
+    gpio_set_irq(gpio2, pin, 0);
+    gpio_set_bit(gpio2, ODR, pin, 1);
+    gpio_set_2bits(gpio2, MODER, pin, MODER_OUTPUT);
+    g_assert_cmphex(get_disconnected_pins(gpio2), ==, 0xFFFF);
+    g_assert_cmphex(gpio_readl(gpio2, IDR), ==, reset(gpio2, IDR) | (1 << pin));
+
+    /* Trying to set a push-pull output pin, checking it doesn't work */
+    gpio_set_irq(gpio, pin, 1);
+    g_assert_cmphex(get_disconnected_pins(gpio), ==, 0xFFFF);
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
+
+    /* Trying to reset a push-pull output pin, checking it doesn't work */
+    gpio_set_irq(gpio2, pin, 0);
+    g_assert_cmphex(get_disconnected_pins(gpio2), ==, 0xFFFF);
+    g_assert_cmphex(gpio_readl(gpio2, IDR), ==, reset(gpio2, IDR) | (1 << pin));
+
+    /* Clean the test */
+    gpio_writel(gpio, MODER, reset(gpio, MODER));
+    gpio_writel(gpio2, ODR, reset(gpio2, ODR));
+    gpio_writel(gpio2, MODER, reset(gpio2, MODER));
+}
+
+static void test_open_drain(const void *data)
+{
+    /*
+     * Test that configuring a line in open-drain output mode
+     * disconnects a pin set high externally and that the pin
+     * can't be set high externally while configured in open-drain.
+     *
+     * However a pin set low externally shouldn't be disconnected,
+     * and it can be set low externally when in open-drain mode.
+     */
+    unsigned int pin = ((uint64_t)data) & 0xF;
+    uint32_t gpio = ((uint64_t)data) >> 32;
+    uint32_t gpio2 = GPIO_BASE_ADDR + (GPIO_H - gpio);
+
+    qtest_irq_intercept_in(global_qtest, "/machine/soc/syscfg");
+
+    /* Setting a line high externally, configuring it in open-drain output */
+    /* And checking the pin was disconnected */
+    gpio_set_irq(gpio, pin, 1);
+    gpio_set_bit(gpio, OTYPER, pin, OTYPER_OPEN_DRAIN);
+    gpio_set_2bits(gpio, MODER, pin, MODER_OUTPUT);
+    g_assert_cmphex(get_disconnected_pins(gpio), ==, 0xFFFF);
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
+
+    /* Setting a line low externally, configuring it in open-drain output */
+    /* And checking the pin wasn't disconnected */
+    gpio_set_irq(gpio2, pin, 0);
+    gpio_set_bit(gpio2, ODR, pin, 1);
+    gpio_set_bit(gpio2, OTYPER, pin, OTYPER_OPEN_DRAIN);
+    gpio_set_2bits(gpio2, MODER, pin, MODER_OUTPUT);
+    g_assert_cmphex(get_disconnected_pins(gpio2), ==, 0xFFFF & ~(1 << pin));
+    g_assert_cmphex(gpio_readl(gpio2, IDR), ==,
+                               reset(gpio2, IDR) & ~(1 << pin));
+
+    /* Trying to set a open-drain output pin, checking it doesn't work */
+    gpio_set_irq(gpio, pin, 1);
+    g_assert_cmphex(get_disconnected_pins(gpio), ==, 0xFFFF);
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR) & ~(1 << pin));
+
+    /* Trying to reset a open-drain output pin, checking it works */
+    gpio_set_bit(gpio, ODR, pin, 1);
+    gpio_set_irq(gpio, pin, 0);
+    g_assert_cmphex(get_disconnected_pins(gpio2), ==, 0xFFFF & ~(1 << pin));
+    g_assert_cmphex(gpio_readl(gpio2, IDR), ==,
+                               reset(gpio2, IDR) & ~(1 << pin));
+
+    /* Clean the test */
+    disconnect_all_pins(gpio2);
+    gpio_writel(gpio2, OTYPER, reset(gpio2, OTYPER));
+    gpio_writel(gpio2, ODR, reset(gpio2, ODR));
+    gpio_writel(gpio2, MODER, reset(gpio2, MODER));
+    g_assert_cmphex(gpio_readl(gpio2, IDR), ==, reset(gpio2, IDR));
+    disconnect_all_pins(gpio);
+    gpio_writel(gpio, OTYPER, reset(gpio, OTYPER));
+    gpio_writel(gpio, ODR, reset(gpio, ODR));
+    gpio_writel(gpio, MODER, reset(gpio, MODER));
+    g_assert_cmphex(gpio_readl(gpio, IDR), ==, reset(gpio, IDR));
+}
+
+static void test_bsrr_brr(const void *data)
+{
+    /*
+     * Test that writing a '1' in BSS and BSRR
+     * has the desired effect on ODR.
+     * In BSRR, BSx has priority over BRx.
+     */
+    unsigned int pin = ((uint64_t)data) & 0xF;
+    uint32_t gpio = ((uint64_t)data) >> 32;
+
+    gpio_writel(gpio, BSRR, (1 << pin));
+    g_assert_cmphex(gpio_readl(gpio, ODR), ==, reset(gpio, ODR) | (1 << pin));
+
+    gpio_writel(gpio, BSRR, (1 << (pin + NUM_GPIO_PINS)));
+    g_assert_cmphex(gpio_readl(gpio, ODR), ==, reset(gpio, ODR));
+
+    gpio_writel(gpio, BSRR, (1 << pin));
+    g_assert_cmphex(gpio_readl(gpio, ODR), ==, reset(gpio, ODR) | (1 << pin));
+
+    gpio_writel(gpio, BRR, (1 << pin));
+    g_assert_cmphex(gpio_readl(gpio, ODR), ==, reset(gpio, ODR));
+
+    /* BSx should have priority over BRx */
+    gpio_writel(gpio, BSRR, (1 << pin) | (1 << (pin + NUM_GPIO_PINS)));
+    g_assert_cmphex(gpio_readl(gpio, ODR), ==, reset(gpio, ODR) | (1 << pin));
+
+    gpio_writel(gpio, BRR, (1 << pin));
+    g_assert_cmphex(gpio_readl(gpio, ODR), ==, reset(gpio, ODR));
+
+    gpio_writel(gpio, ODR, reset(gpio, ODR));
+}
+
+int main(int argc, char **argv)
+{
+    int ret;
+
+    g_test_init(&argc, &argv, NULL);
+    g_test_set_nonfatal_assertions();
+    qtest_add_func("stm32l4x5/gpio/test_idr_reset_value",
+                   test_idr_reset_value);
+    /*
+     * The inputs for the tests (gpio and pin) can be changed,
+     * but the tests don't work for pins that are high at reset
+     * (GPIOA15, GPIO13 and GPIOB5).
+     * Specifically, rising the pin then checking `get_irq()`
+     * is problematic since the pin was already high.
+     */
+    qtest_add_data_func("stm32l4x5/gpio/test_gpioc5_output_mode",
+                        (void *)((uint64_t)GPIO_C << 32 | 5),
+                        test_gpio_output_mode);
+    qtest_add_data_func("stm32l4x5/gpio/test_gpioh3_output_mode",
+                        (void *)((uint64_t)GPIO_H << 32 | 3),
+                        test_gpio_output_mode);
+    qtest_add_data_func("stm32l4x5/gpio/test_gpio_input_mode1",
+                        (void *)((uint64_t)GPIO_D << 32 | 6),
+                        test_gpio_input_mode);
+    qtest_add_data_func("stm32l4x5/gpio/test_gpio_input_mode2",
+                        (void *)((uint64_t)GPIO_C << 32 | 10),
+                        test_gpio_input_mode);
+    qtest_add_data_func("stm32l4x5/gpio/test_gpio_pull_up_pull_down1",
+                        (void *)((uint64_t)GPIO_B << 32 | 5),
+                        test_pull_up_pull_down);
+    qtest_add_data_func("stm32l4x5/gpio/test_gpio_pull_up_pull_down2",
+                        (void *)((uint64_t)GPIO_F << 32 | 1),
+                        test_pull_up_pull_down);
+    qtest_add_data_func("stm32l4x5/gpio/test_gpio_push_pull1",
+                        (void *)((uint64_t)GPIO_G << 32 | 6),
+                        test_push_pull);
+    qtest_add_data_func("stm32l4x5/gpio/test_gpio_push_pull2",
+                        (void *)((uint64_t)GPIO_H << 32 | 3),
+                        test_push_pull);
+    qtest_add_data_func("stm32l4x5/gpio/test_gpio_open_drain1",
+                        (void *)((uint64_t)GPIO_C << 32 | 4),
+                        test_open_drain);
+    qtest_add_data_func("stm32l4x5/gpio/test_gpio_open_drain2",
+                        (void *)((uint64_t)GPIO_E << 32 | 11),
+                        test_open_drain);
+    qtest_add_data_func("stm32l4x5/gpio/test_bsrr_brr1",
+                        (void *)((uint64_t)GPIO_A << 32 | 12),
+                        test_bsrr_brr);
+    qtest_add_data_func("stm32l4x5/gpio/test_bsrr_brr2",
+                        (void *)((uint64_t)GPIO_D << 32 | 0),
+                        test_bsrr_brr);
+
+    qtest_start("-machine b-l475e-iot01a");
+    ret = g_test_run();
+    qtest_end();
+
+    return ret;
+}
diff --git a/tests/qtest/meson.build b/tests/qtest/meson.build
index XXXXXXX..XXXXXXX 100644
--- a/tests/qtest/meson.build
+++ b/tests/qtest/meson.build
@@ -XXX,XX +XXX,XX @@ qtests_aspeed = \
 qtests_stm32l4x5 = \
   ['stm32l4x5_exti-test',
    'stm32l4x5_syscfg-test',
-   'stm32l4x5_rcc-test']
+   'stm32l4x5_rcc-test',
+   'stm32l4x5_gpio-test']
 
 qtests_arm = \
   (config_all_devices.has_key('CONFIG_MPS2') ? ['sse-timer-test'] : []) + \
-- 
2.34.1

From: Richard Henderson <richard.henderson@linaro.org>

While the 8-bit input elements are sequential in the input vector,
the 32-bit output elements are not sequential in the output matrix.
Do not attempt to compute 2 32-bit outputs at the same time.

Cc: qemu-stable@nongnu.org
Fixes: 23a5e3859f5 ("target/arm: Implement SME integer outer product")
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2083
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-id: 20240305163931.242795-1-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/tcg/sme_helper.c       | 77 ++++++++++++++++++-------------
 tests/tcg/aarch64/sme-smopa-1.c   | 47 +++++++++++++++++++
 tests/tcg/aarch64/sme-smopa-2.c   | 54 ++++++++++++++++++++++
 tests/tcg/aarch64/Makefile.target |  2 +-
 4 files changed, 147 insertions(+), 33 deletions(-)
 create mode 100644 tests/tcg/aarch64/sme-smopa-1.c
 create mode 100644 tests/tcg/aarch64/sme-smopa-2.c

diff --git a/target/arm/tcg/sme_helper.c b/target/arm/tcg/sme_helper.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/tcg/sme_helper.c
+++ b/target/arm/tcg/sme_helper.c
@@ -XXX,XX +XXX,XX @@ void HELPER(sme_bfmopa)(void *vza, void *vzn, void *vzm, void *vpn,
     }
 }
 
-typedef uint64_t IMOPFn(uint64_t, uint64_t, uint64_t, uint8_t, bool);
+typedef uint32_t IMOPFn32(uint32_t, uint32_t, uint32_t, uint8_t, bool);
+static inline void do_imopa_s(uint32_t *za, uint32_t *zn, uint32_t *zm,
+                              uint8_t *pn, uint8_t *pm,
+                              uint32_t desc, IMOPFn32 *fn)
+{
+    intptr_t row, col, oprsz = simd_oprsz(desc) / 4;
+    bool neg = simd_data(desc);
 
-static inline void do_imopa(uint64_t *za, uint64_t *zn, uint64_t *zm,
-                            uint8_t *pn, uint8_t *pm,
-                            uint32_t desc, IMOPFn *fn)
+    for (row = 0; row < oprsz; ++row) {
+        uint8_t pa = (pn[H1(row >> 1)] >> ((row & 1) * 4)) & 0xf;
+        uint32_t *za_row = &za[tile_vslice_index(row)];
+        uint32_t n = zn[H4(row)];
+
+        for (col = 0; col < oprsz; ++col) {
+            uint8_t pb = pm[H1(col >> 1)] >> ((col & 1) * 4);
+            uint32_t *a = &za_row[H4(col)];
+
+            *a = fn(n, zm[H4(col)], *a, pa & pb, neg);
+        }
+    }
+}
+
+typedef uint64_t IMOPFn64(uint64_t, uint64_t, uint64_t, uint8_t, bool);
+static inline void do_imopa_d(uint64_t *za, uint64_t *zn, uint64_t *zm,
+                              uint8_t *pn, uint8_t *pm,
+                              uint32_t desc, IMOPFn64 *fn)
 {
     intptr_t row, col, oprsz = simd_oprsz(desc) / 8;
     bool neg = simd_data(desc);
@@ -XXX,XX +XXX,XX @@ static inline void do_imopa(uint64_t *za, uint64_t *zn, uint64_t *zm,
 }
 
 #define DEF_IMOP_32(NAME, NTYPE, MTYPE) \
-static uint64_t NAME(uint64_t n, uint64_t m, uint64_t a, uint8_t p, bool neg) \
+static uint32_t NAME(uint32_t n, uint32_t m, uint32_t a, uint8_t p, bool neg) \
 {                                                                           \
-    uint32_t sum0 = 0, sum1 = 0;                                            \
+    uint32_t sum = 0;                                                       \
     /* Apply P to N as a mask, making the inactive elements 0. */           \
     n &= expand_pred_b(p);                                                  \
-    sum0 += (NTYPE)(n >> 0) * (MTYPE)(m >> 0);                              \
-    sum0 += (NTYPE)(n >> 8) * (MTYPE)(m >> 8);                              \
-    sum0 += (NTYPE)(n >> 16) * (MTYPE)(m >> 16);                            \
-    sum0 += (NTYPE)(n >> 24) * (MTYPE)(m >> 24);                            \
-    sum1 += (NTYPE)(n >> 32) * (MTYPE)(m >> 32);                            \
-    sum1 += (NTYPE)(n >> 40) * (MTYPE)(m >> 40);                            \
-    sum1 += (NTYPE)(n >> 48) * (MTYPE)(m >> 48);                            \
-    sum1 += (NTYPE)(n >> 56) * (MTYPE)(m >> 56);                            \
-    if (neg) {                                                              \
-        sum0 = (uint32_t)a - sum0, sum1 = (uint32_t)(a >> 32) - sum1;       \
-    } else {                                                                \
-        sum0 = (uint32_t)a + sum0, sum1 = (uint32_t)(a >> 32) + sum1;       \
-    }                                                                       \
-    return ((uint64_t)sum1 << 32) | sum0;                                   \
+    sum += (NTYPE)(n >> 0) * (MTYPE)(m >> 0);                               \
+    sum += (NTYPE)(n >> 8) * (MTYPE)(m >> 8);                               \
+    sum += (NTYPE)(n >> 16) * (MTYPE)(m >> 16);                             \
+    sum += (NTYPE)(n >> 24) * (MTYPE)(m >> 24);                             \
+    return neg ? a - sum : a + sum;                                         \
 }
 
 #define DEF_IMOP_64(NAME, NTYPE, MTYPE) \
@@ -XXX,XX +XXX,XX @@ DEF_IMOP_64(umopa_d, uint16_t, uint16_t)
 DEF_IMOP_64(sumopa_d, int16_t, uint16_t)
 DEF_IMOP_64(usmopa_d, uint16_t, int16_t)
 
-#define DEF_IMOPH(NAME) \
-    void HELPER(sme_##NAME)(void *vza, void *vzn, void *vzm, void *vpn,      \
-                            void *vpm, uint32_t desc)                        \
-    { do_imopa(vza, vzn, vzm, vpn, vpm, desc, NAME); }
+#define DEF_IMOPH(NAME, S) \
+    void HELPER(sme_##NAME##_##S)(void *vza, void *vzn, void *vzm,          \
+                                  void *vpn, void *vpm, uint32_t desc)      \
+    { do_imopa_##S(vza, vzn, vzm, vpn, vpm, desc, NAME##_##S); }
 
-DEF_IMOPH(smopa_s)
-DEF_IMOPH(umopa_s)
-DEF_IMOPH(sumopa_s)
-DEF_IMOPH(usmopa_s)
-DEF_IMOPH(smopa_d)
-DEF_IMOPH(umopa_d)
-DEF_IMOPH(sumopa_d)
-DEF_IMOPH(usmopa_d)
+DEF_IMOPH(smopa, s)
+DEF_IMOPH(umopa, s)
+DEF_IMOPH(sumopa, s)
+DEF_IMOPH(usmopa, s)
+
+DEF_IMOPH(smopa, d)
+DEF_IMOPH(umopa, d)
+DEF_IMOPH(sumopa, d)
+DEF_IMOPH(usmopa, d)
diff --git a/tests/tcg/aarch64/sme-smopa-1.c b/tests/tcg/aarch64/sme-smopa-1.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/tests/tcg/aarch64/sme-smopa-1.c
@@ -XXX,XX +XXX,XX @@
+#include <stdio.h>
+#include <string.h>
+
+int main()
+{
+    static const int cmp[4][4] = {
+        {  110,  134,  158,  182 },
+        {  390,  478,  566,  654 },
+        {  670,  822,  974, 1126 },
+        {  950, 1166, 1382, 1598 }
+    };
+    int dst[4][4];
+    int *tmp = &dst[0][0];
+
+    asm volatile(
+        ".arch armv8-r+sme\n\t"
+        "smstart\n\t"
+        "index z0.b, #0, #1\n\t"
+        "movprfx z1, z0\n\t"
+        "add z1.b, z1.b, #16\n\t"
+        "ptrue p0.b\n\t"
+        "smopa za0.s, p0/m, p0/m, z0.b, z1.b\n\t"
+        "ptrue p0.s, vl4\n\t"
+        "mov w12, #0\n\t"
+        "st1w { za0h.s[w12, #0] }, p0, [%0]\n\t"
+        "add %0, %0, #16\n\t"
+        "st1w { za0h.s[w12, #1] }, p0, [%0]\n\t"
+        "add %0, %0, #16\n\t"
+        "st1w { za0h.s[w12, #2] }, p0, [%0]\n\t"
+        "add %0, %0, #16\n\t"
+        "st1w { za0h.s[w12, #3] }, p0, [%0]\n\t"
+        "smstop"
+        : "+r"(tmp) : : "memory");
+
+    if (memcmp(cmp, dst, sizeof(dst)) == 0) {
+        return 0;
+    }
+
+    /* See above for correct results. */
+    for (int i = 0; i < 4; ++i) {
+        for (int j = 0; j < 4; ++j) {
+            printf("%6d", dst[i][j]);
+        }
+        printf("\n");
+    }
+    return 1;
+}
diff --git a/tests/tcg/aarch64/sme-smopa-2.c b/tests/tcg/aarch64/sme-smopa-2.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/tests/tcg/aarch64/sme-smopa-2.c
@@ -XXX,XX +XXX,XX @@
+#include <stdio.h>
+#include <string.h>
+
+int main()
+{
+    static const long cmp[4][4] = {
+        {  110,  134,  158,  182 },
+        {  390,  478,  566,  654 },
+        {  670,  822,  974, 1126 },
+        {  950, 1166, 1382, 1598 }
+    };
+    long dst[4][4];
+    long *tmp = &dst[0][0];
+    long svl;
+
+    /* Validate that we have a wide enough vector for 4 elements. */
+    asm(".arch armv8-r+sme-i64\n\trdsvl %0, #1" : "=r"(svl));
+    if (svl < 32) {
+        return 0;
+    }
+
+    asm volatile(
+        "smstart\n\t"
+        "index z0.h, #0, #1\n\t"
+        "movprfx z1, z0\n\t"
+        "add z1.h, z1.h, #16\n\t"
+        "ptrue p0.b\n\t"
+        "smopa za0.d, p0/m, p0/m, z0.h, z1.h\n\t"
+        "ptrue p0.d, vl4\n\t"
+        "mov w12, #0\n\t"
+        "st1d { za0h.d[w12, #0] }, p0, [%0]\n\t"
+        "add %0, %0, #32\n\t"
+        "st1d { za0h.d[w12, #1] }, p0, [%0]\n\t"
+        "mov w12, #2\n\t"
+        "add %0, %0, #32\n\t"
+        "st1d { za0h.d[w12, #0] }, p0, [%0]\n\t"
+        "add %0, %0, #32\n\t"
+        "st1d { za0h.d[w12, #1] }, p0, [%0]\n\t"
+        "smstop"
+        : "+r"(tmp) : : "memory");
+
+    if (memcmp(cmp, dst, sizeof(dst)) == 0) {
+        return 0;
+    }
+
+    /* See above for correct results. */
+    for (int i = 0; i < 4; ++i) {
+        for (int j = 0; j < 4; ++j) {
+            printf("%6ld", dst[i][j]);
+        }
+        printf("\n");
+    }
+    return 1;
+}
diff --git a/tests/tcg/aarch64/Makefile.target b/tests/tcg/aarch64/Makefile.target
index XXXXXXX..XXXXXXX 100644
--- a/tests/tcg/aarch64/Makefile.target
+++ b/tests/tcg/aarch64/Makefile.target
@@ -XXX,XX +XXX,XX @@ endif
 
 # SME Tests
 ifneq ($(CROSS_AS_HAS_ARMV9_SME),)
-AARCH64_TESTS += sme-outprod1
+AARCH64_TESTS += sme-outprod1 sme-smopa-1 sme-smopa-2
 endif
 
 # System Registers Tests
-- 
2.34.1

The sun4v RTC device model added under commit a0e893039cf2ce0 in 2016
was unfortunately added with a license of GPL-v3-or-later, which is
not compatible with other QEMU code which has a GPL-v2-only license.

Relicense the code in the .c and the .h file to GPL-v2-or-later,
to make it compatible with the rest of QEMU.

Cc: qemu-stable@nongnu.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Paolo Bonzini (for Red Hat) <pbonzini@redhat.com>
Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Alex Bennée <alex.bennee@linaro.org>
Message-id: 20240223161300.938542-1-peter.maydell@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 include/hw/rtc/sun4v-rtc.h | 2 +-
 hw/rtc/sun4v-rtc.c         | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/hw/rtc/sun4v-rtc.h b/include/hw/rtc/sun4v-rtc.h
index XXXXXXX..XXXXXXX 100644
--- a/include/hw/rtc/sun4v-rtc.h
+++ b/include/hw/rtc/sun4v-rtc.h
@@ -XXX,XX +XXX,XX @@
  *
  * Copyright (c) 2016 Artyom Tarasenko
  *
- * This code is licensed under the GNU GPL v3 or (at your option) any later
+ * This code is licensed under the GNU GPL v2 or (at your option) any later
  * version.
  */
 
diff --git a/hw/rtc/sun4v-rtc.c b/hw/rtc/sun4v-rtc.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/rtc/sun4v-rtc.c
+++ b/hw/rtc/sun4v-rtc.c
@@ -XXX,XX +XXX,XX @@
  *
  * Copyright (c) 2016 Artyom Tarasenko
  *
- * This code is licensed under the GNU GPL v3 or (at your option) any later
+ * This code is licensed under the GNU GPL v2 or (at your option) any later
  * version.
  */
 
-- 
2.34.1

From: Thomas Huth <thuth@redhat.com>

Move the code to a separate file so that we do not have to compile
it anymore if CONFIG_ARM_V7M is not set.

Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-id: 20240308141051.536599-2-thuth@redhat.com
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/tcg/cpu-v7m.c   | 290 +++++++++++++++++++++++++++++++++++++
 target/arm/tcg/cpu32.c     | 261 ---------------------------------
 target/arm/meson.build     |   3 +
 target/arm/tcg/meson.build |   3 +
 4 files changed, 296 insertions(+), 261 deletions(-)
 create mode 100644 target/arm/tcg/cpu-v7m.c

diff --git a/target/arm/tcg/cpu-v7m.c b/target/arm/tcg/cpu-v7m.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/target/arm/tcg/cpu-v7m.c
@@ -XXX,XX +XXX,XX @@
+/*
+ * QEMU ARMv7-M TCG-only CPUs.
+ *
+ * Copyright (c) 2012 SUSE LINUX Products GmbH
+ *
+ * This code is licensed under the GNU GPL v2 or later.
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "hw/core/tcg-cpu-ops.h"
+#include "internals.h"
+
+#if !defined(CONFIG_USER_ONLY)
+
+#include "hw/intc/armv7m_nvic.h"
+
+static bool arm_v7m_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
+{
+    CPUClass *cc = CPU_GET_CLASS(cs);
+    ARMCPU *cpu = ARM_CPU(cs);
+    CPUARMState *env = &cpu->env;
+    bool ret = false;
+
+    /*
+     * ARMv7-M interrupt masking works differently than -A or -R.
+     * There is no FIQ/IRQ distinction. Instead of I and F bits
+     * masking FIQ and IRQ interrupts, an exception is taken only
+     * if it is higher priority than the current execution priority
+     * (which depends on state like BASEPRI, FAULTMASK and the
+     * currently active exception).
+     */
+    if (interrupt_request & CPU_INTERRUPT_HARD
+        && (armv7m_nvic_can_take_pending_exception(env->nvic))) {
+        cs->exception_index = EXCP_IRQ;
+        cc->tcg_ops->do_interrupt(cs);
+        ret = true;
+    }
+    return ret;
+}
+
+#endif /* !CONFIG_USER_ONLY */
+
+static void cortex_m0_initfn(Object *obj)
+{
+    ARMCPU *cpu = ARM_CPU(obj);
+    set_feature(&cpu->env, ARM_FEATURE_V6);
+    set_feature(&cpu->env, ARM_FEATURE_M);
+
+    cpu->midr = 0x410cc200;
+
+    /*
+     * These ID register values are not guest visible, because
+     * we do not implement the Main Extension. They must be set
+     * to values corresponding to the Cortex-M0's implemented
+     * features, because QEMU generally controls its emulation
+     * by looking at ID register fields. We use the same values as
+     * for the M3.
+     */
+    cpu->isar.id_pfr0 = 0x00000030;
+    cpu->isar.id_pfr1 = 0x00000200;
+    cpu->isar.id_dfr0 = 0x00100000;
+    cpu->id_afr0 = 0x00000000;
+    cpu->isar.id_mmfr0 = 0x00000030;
+    cpu->isar.id_mmfr1 = 0x00000000;
+    cpu->isar.id_mmfr2 = 0x00000000;
+    cpu->isar.id_mmfr3 = 0x00000000;
+    cpu->isar.id_isar0 = 0x01141110;
+    cpu->isar.id_isar1 = 0x02111000;
+    cpu->isar.id_isar2 = 0x21112231;
+    cpu->isar.id_isar3 = 0x01111110;
+    cpu->isar.id_isar4 = 0x01310102;
+    cpu->isar.id_isar5 = 0x00000000;
+    cpu->isar.id_isar6 = 0x00000000;
+}
+
+static void cortex_m3_initfn(Object *obj)
+{
+    ARMCPU *cpu = ARM_CPU(obj);
+    set_feature(&cpu->env, ARM_FEATURE_V7);
+    set_feature(&cpu->env, ARM_FEATURE_M);
+    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
+    cpu->midr = 0x410fc231;
+    cpu->pmsav7_dregion = 8;
+    cpu->isar.id_pfr0 = 0x00000030;
+    cpu->isar.id_pfr1 = 0x00000200;
+    cpu->isar.id_dfr0 = 0x00100000;
+    cpu->id_afr0 = 0x00000000;
+    cpu->isar.id_mmfr0 = 0x00000030;
+    cpu->isar.id_mmfr1 = 0x00000000;
+    cpu->isar.id_mmfr2 = 0x00000000;
+    cpu->isar.id_mmfr3 = 0x00000000;
+    cpu->isar.id_isar0 = 0x01141110;
+    cpu->isar.id_isar1 = 0x02111000;
+    cpu->isar.id_isar2 = 0x21112231;
+    cpu->isar.id_isar3 = 0x01111110;
+    cpu->isar.id_isar4 = 0x01310102;
+    cpu->isar.id_isar5 = 0x00000000;
+    cpu->isar.id_isar6 = 0x00000000;
+}
+
+static void cortex_m4_initfn(Object *obj)
+{
+    ARMCPU *cpu = ARM_CPU(obj);
+
+    set_feature(&cpu->env, ARM_FEATURE_V7);
+    set_feature(&cpu->env, ARM_FEATURE_M);
+    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
+    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
+    cpu->midr = 0x410fc240; /* r0p0 */
+    cpu->pmsav7_dregion = 8;
+    cpu->isar.mvfr0 = 0x10110021;
+    cpu->isar.mvfr1 = 0x11000011;
+    cpu->isar.mvfr2 = 0x00000000;
+    cpu->isar.id_pfr0 = 0x00000030;
+    cpu->isar.id_pfr1 = 0x00000200;
+    cpu->isar.id_dfr0 = 0x00100000;
+    cpu->id_afr0 = 0x00000000;
+    cpu->isar.id_mmfr0 = 0x00000030;
+    cpu->isar.id_mmfr1 = 0x00000000;
+    cpu->isar.id_mmfr2 = 0x00000000;
+    cpu->isar.id_mmfr3 = 0x00000000;
+    cpu->isar.id_isar0 = 0x01141110;
+    cpu->isar.id_isar1 = 0x02111000;
+    cpu->isar.id_isar2 = 0x21112231;
+    cpu->isar.id_isar3 = 0x01111110;
+    cpu->isar.id_isar4 = 0x01310102;
+    cpu->isar.id_isar5 = 0x00000000;
+    cpu->isar.id_isar6 = 0x00000000;
+}
+
+static void cortex_m7_initfn(Object *obj)
+{
+    ARMCPU *cpu = ARM_CPU(obj);
+
+    set_feature(&cpu->env, ARM_FEATURE_V7);
+    set_feature(&cpu->env, ARM_FEATURE_M);
+    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
+    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
+    cpu->midr = 0x411fc272; /* r1p2 */
+    cpu->pmsav7_dregion = 8;
+    cpu->isar.mvfr0 = 0x10110221;
+    cpu->isar.mvfr1 = 0x12000011;
+    cpu->isar.mvfr2 = 0x00000040;
+    cpu->isar.id_pfr0 = 0x00000030;
+    cpu->isar.id_pfr1 = 0x00000200;
+    cpu->isar.id_dfr0 = 0x00100000;
+    cpu->id_afr0 = 0x00000000;
+    cpu->isar.id_mmfr0 = 0x00100030;
+    cpu->isar.id_mmfr1 = 0x00000000;
+    cpu->isar.id_mmfr2 = 0x01000000;
+    cpu->isar.id_mmfr3 = 0x00000000;
+    cpu->isar.id_isar0 = 0x01101110;
+    cpu->isar.id_isar1 = 0x02112000;
+    cpu->isar.id_isar2 = 0x20232231;
+    cpu->isar.id_isar3 = 0x01111131;
+    cpu->isar.id_isar4 = 0x01310132;
+    cpu->isar.id_isar5 = 0x00000000;
+    cpu->isar.id_isar6 = 0x00000000;
+}
+
+static void cortex_m33_initfn(Object *obj)
+{
+    ARMCPU *cpu = ARM_CPU(obj);
+
+    set_feature(&cpu->env, ARM_FEATURE_V8);
+    set_feature(&cpu->env, ARM_FEATURE_M);
+    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
+    set_feature(&cpu->env, ARM_FEATURE_M_SECURITY);
+    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
+    cpu->midr = 0x410fd213; /* r0p3 */
+    cpu->pmsav7_dregion = 16;
+    cpu->sau_sregion = 8;
+    cpu->isar.mvfr0 = 0x10110021;
+    cpu->isar.mvfr1 = 0x11000011;
+    cpu->isar.mvfr2 = 0x00000040;
+    cpu->isar.id_pfr0 = 0x00000030;
+    cpu->isar.id_pfr1 = 0x00000210;
+    cpu->isar.id_dfr0 = 0x00200000;
+    cpu->id_afr0 = 0x00000000;
+    cpu->isar.id_mmfr0 = 0x00101F40;
+    cpu->isar.id_mmfr1 = 0x00000000;
+    cpu->isar.id_mmfr2 = 0x01000000;
+    cpu->isar.id_mmfr3 = 0x00000000;
+    cpu->isar.id_isar0 = 0x01101110;
+    cpu->isar.id_isar1 = 0x02212000;
+    cpu->isar.id_isar2 = 0x20232232;
+    cpu->isar.id_isar3 = 0x01111131;
+    cpu->isar.id_isar4 = 0x01310132;
+    cpu->isar.id_isar5 = 0x00000000;
+    cpu->isar.id_isar6 = 0x00000000;
+    cpu->clidr = 0x00000000;
+    cpu->ctr = 0x8000c000;
+}
+
+static void cortex_m55_initfn(Object *obj)
+{
+    ARMCPU *cpu = ARM_CPU(obj);
+
+    set_feature(&cpu->env, ARM_FEATURE_V8);
+    set_feature(&cpu->env, ARM_FEATURE_V8_1M);
+    set_feature(&cpu->env, ARM_FEATURE_M);
+    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
+    set_feature(&cpu->env, ARM_FEATURE_M_SECURITY);
+    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
+    cpu->midr = 0x410fd221; /* r0p1 */
+    cpu->revidr = 0;
+    cpu->pmsav7_dregion = 16;
+    cpu->sau_sregion = 8;
+    /* These are the MVFR* values for the FPU + full MVE configuration */
+    cpu->isar.mvfr0 = 0x10110221;
+    cpu->isar.mvfr1 = 0x12100211;
+    cpu->isar.mvfr2 = 0x00000040;
+    cpu->isar.id_pfr0 = 0x20000030;
+    cpu->isar.id_pfr1 = 0x00000230;
+    cpu->isar.id_dfr0 = 0x10200000;
+    cpu->id_afr0 = 0x00000000;
+    cpu->isar.id_mmfr0 = 0x00111040;
+    cpu->isar.id_mmfr1 = 0x00000000;
+    cpu->isar.id_mmfr2 = 0x01000000;
+    cpu->isar.id_mmfr3 = 0x00000011;
+    cpu->isar.id_isar0 = 0x01103110;
+    cpu->isar.id_isar1 = 0x02212000;
+    cpu->isar.id_isar2 = 0x20232232;
+    cpu->isar.id_isar3 = 0x01111131;
+    cpu->isar.id_isar4 = 0x01310132;
+    cpu->isar.id_isar5 = 0x00000000;
+    cpu->isar.id_isar6 = 0x00000000;
+    cpu->clidr = 0x00000000; /* caches not implemented */
+    cpu->ctr = 0x8303c003;
+}
+
+static const TCGCPUOps arm_v7m_tcg_ops = {
+    .initialize = arm_translate_init,
+    .synchronize_from_tb = arm_cpu_synchronize_from_tb,
+    .debug_excp_handler = arm_debug_excp_handler,
+    .restore_state_to_opc = arm_restore_state_to_opc,
+
+#ifdef CONFIG_USER_ONLY
+    .record_sigsegv = arm_cpu_record_sigsegv,
+    .record_sigbus = arm_cpu_record_sigbus,
+#else
+    .tlb_fill = arm_cpu_tlb_fill,
+    .cpu_exec_interrupt = arm_v7m_cpu_exec_interrupt,
+    .do_interrupt = arm_v7m_cpu_do_interrupt,
+    .do_transaction_failed = arm_cpu_do_transaction_failed,
+    .do_unaligned_access = arm_cpu_do_unaligned_access,
+    .adjust_watchpoint_address = arm_adjust_watchpoint_address,
+    .debug_check_watchpoint = arm_debug_check_watchpoint,
+    .debug_check_breakpoint = arm_debug_check_breakpoint,
+#endif /* !CONFIG_USER_ONLY */
+};
+
+static void arm_v7m_class_init(ObjectClass *oc, void *data)
+{
+    ARMCPUClass *acc = ARM_CPU_CLASS(oc);
+    CPUClass *cc = CPU_CLASS(oc);
+
+    acc->info = data;
+    cc->tcg_ops = &arm_v7m_tcg_ops;
+    cc->gdb_core_xml_file = "arm-m-profile.xml";
+}
+
+static const ARMCPUInfo arm_v7m_cpus[] = {
+    { .name = "cortex-m0",   .initfn = cortex_m0_initfn,
+                             .class_init = arm_v7m_class_init },
+    { .name = "cortex-m3",   .initfn = cortex_m3_initfn,
+                             .class_init = arm_v7m_class_init },
+    { .name = "cortex-m4",   .initfn = cortex_m4_initfn,
+                             .class_init = arm_v7m_class_init },
+    { .name = "cortex-m7",   .initfn = cortex_m7_initfn,
+                             .class_init = arm_v7m_class_init },
+    { .name = "cortex-m33",  .initfn = cortex_m33_initfn,
+                             .class_init = arm_v7m_class_init },
+    { .name = "cortex-m55",  .initfn = cortex_m55_initfn,
+                             .class_init = arm_v7m_class_init },
+};
+
+static void arm_v7m_cpu_register_types(void)
+{
+    size_t i;
+
+    for (i = 0; i < ARRAY_SIZE(arm_v7m_cpus); ++i) {
+        arm_cpu_register(&arm_v7m_cpus[i]);
+    }
+}
+
+type_init(arm_v7m_cpu_register_types)
diff --git a/target/arm/tcg/cpu32.c b/target/arm/tcg/cpu32.c
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/tcg/cpu32.c
+++ b/target/arm/tcg/cpu32.c
@@ -XXX,XX +XXX,XX @@
 #include "hw/boards.h"
 #endif
 #include "cpregs.h"
-#if !defined(CONFIG_USER_ONLY) && defined(CONFIG_TCG)
-#include "hw/intc/armv7m_nvic.h"
-#endif
 
 
 /* Share AArch32 -cpu max features with AArch64. */
@@ -XXX,XX +XXX,XX @@ void aa32_max_features(ARMCPU *cpu)
 /* CPU models. These are not needed for the AArch64 linux-user build. */
 #if !defined(CONFIG_USER_ONLY) || !defined(TARGET_AARCH64)
 
-#if !defined(CONFIG_USER_ONLY)
-static bool arm_v7m_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
-{
-    CPUClass *cc = CPU_GET_CLASS(cs);
-    ARMCPU *cpu = ARM_CPU(cs);
-    CPUARMState *env = &cpu->env;
-    bool ret = false;
-
-    /*
-     * ARMv7-M interrupt masking works differently than -A or -R.
-     * There is no FIQ/IRQ distinction. Instead of I and F bits
-     * masking FIQ and IRQ interrupts, an exception is taken only
-     * if it is higher priority than the current execution priority
-     * (which depends on state like BASEPRI, FAULTMASK and the
-     * currently active exception).
-     */
-    if (interrupt_request & CPU_INTERRUPT_HARD
-        && (armv7m_nvic_can_take_pending_exception(env->nvic))) {
-        cs->exception_index = EXCP_IRQ;
-        cc->tcg_ops->do_interrupt(cs);
-        ret = true;
-    }
-    return ret;
-}
-#endif /* !CONFIG_USER_ONLY */
-
 static void arm926_initfn(Object *obj)
 {
     ARMCPU *cpu = ARM_CPU(obj);
@@ -XXX,XX +XXX,XX @@ static void cortex_a15_initfn(Object *obj)
     define_arm_cp_regs(cpu, cortexa15_cp_reginfo);
 }
 
-static void cortex_m0_initfn(Object *obj)
-{
-    ARMCPU *cpu = ARM_CPU(obj);
-    set_feature(&cpu->env, ARM_FEATURE_V6);
-    set_feature(&cpu->env, ARM_FEATURE_M);
-
-    cpu->midr = 0x410cc200;
-
-    /*
-     * These ID register values are not guest visible, because
-     * we do not implement the Main Extension. They must be set
-     * to values corresponding to the Cortex-M0's implemented
-     * features, because QEMU generally controls its emulation
-     * by looking at ID register fields. We use the same values as
-     * for the M3.
-     */
-    cpu->isar.id_pfr0 = 0x00000030;
-    cpu->isar.id_pfr1 = 0x00000200;
-    cpu->isar.id_dfr0 = 0x00100000;
-    cpu->id_afr0 = 0x00000000;
-    cpu->isar.id_mmfr0 = 0x00000030;
-    cpu->isar.id_mmfr1 = 0x00000000;
-    cpu->isar.id_mmfr2 = 0x00000000;
-    cpu->isar.id_mmfr3 = 0x00000000;
-    cpu->isar.id_isar0 = 0x01141110;
-    cpu->isar.id_isar1 = 0x02111000;
-    cpu->isar.id_isar2 = 0x21112231;
-    cpu->isar.id_isar3 = 0x01111110;
-    cpu->isar.id_isar4 = 0x01310102;
-    cpu->isar.id_isar5 = 0x00000000;
-    cpu->isar.id_isar6 = 0x00000000;
-}
-
-static void cortex_m3_initfn(Object *obj)
-{
-    ARMCPU *cpu = ARM_CPU(obj);
-    set_feature(&cpu->env, ARM_FEATURE_V7);
-    set_feature(&cpu->env, ARM_FEATURE_M);
-    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
-    cpu->midr = 0x410fc231;
-    cpu->pmsav7_dregion = 8;
-    cpu->isar.id_pfr0 = 0x00000030;
-    cpu->isar.id_pfr1 = 0x00000200;
-    cpu->isar.id_dfr0 = 0x00100000;
-    cpu->id_afr0 = 0x00000000;
-    cpu->isar.id_mmfr0 = 0x00000030;
-    cpu->isar.id_mmfr1 = 0x00000000;
-    cpu->isar.id_mmfr2 = 0x00000000;
-    cpu->isar.id_mmfr3 = 0x00000000;
-    cpu->isar.id_isar0 = 0x01141110;
-    cpu->isar.id_isar1 = 0x02111000;
-    cpu->isar.id_isar2 = 0x21112231;
-    cpu->isar.id_isar3 = 0x01111110;
-    cpu->isar.id_isar4 = 0x01310102;
-    cpu->isar.id_isar5 = 0x00000000;
-    cpu->isar.id_isar6 = 0x00000000;
-}
-
-static void cortex_m4_initfn(Object *obj)
-{
-    ARMCPU *cpu = ARM_CPU(obj);
-
-    set_feature(&cpu->env, ARM_FEATURE_V7);
-    set_feature(&cpu->env, ARM_FEATURE_M);
-    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
-    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
-    cpu->midr = 0x410fc240; /* r0p0 */
-    cpu->pmsav7_dregion = 8;
-    cpu->isar.mvfr0 = 0x10110021;
-    cpu->isar.mvfr1 = 0x11000011;
-    cpu->isar.mvfr2 = 0x00000000;
-    cpu->isar.id_pfr0 = 0x00000030;
-    cpu->isar.id_pfr1 = 0x00000200;
-    cpu->isar.id_dfr0 = 0x00100000;
-    cpu->id_afr0 = 0x00000000;
-    cpu->isar.id_mmfr0 = 0x00000030;
-    cpu->isar.id_mmfr1 = 0x00000000;
-    cpu->isar.id_mmfr2 = 0x00000000;
-    cpu->isar.id_mmfr3 = 0x00000000;
-    cpu->isar.id_isar0 = 0x01141110;
-    cpu->isar.id_isar1 = 0x02111000;
-    cpu->isar.id_isar2 = 0x21112231;
-    cpu->isar.id_isar3 = 0x01111110;
-    cpu->isar.id_isar4 = 0x01310102;
-    cpu->isar.id_isar5 = 0x00000000;
-    cpu->isar.id_isar6 = 0x00000000;
-}
-
-static void cortex_m7_initfn(Object *obj)
-{
-    ARMCPU *cpu = ARM_CPU(obj);
-
-    set_feature(&cpu->env, ARM_FEATURE_V7);
-    set_feature(&cpu->env, ARM_FEATURE_M);
-    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
-    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
-    cpu->midr = 0x411fc272; /* r1p2 */
-    cpu->pmsav7_dregion = 8;
-    cpu->isar.mvfr0 = 0x10110221;
-    cpu->isar.mvfr1 = 0x12000011;
-    cpu->isar.mvfr2 = 0x00000040;
-    cpu->isar.id_pfr0 = 0x00000030;
-    cpu->isar.id_pfr1 = 0x00000200;
-    cpu->isar.id_dfr0 = 0x00100000;
-    cpu->id_afr0 = 0x00000000;
-    cpu->isar.id_mmfr0 = 0x00100030;
-    cpu->isar.id_mmfr1 = 0x00000000;
-    cpu->isar.id_mmfr2 = 0x01000000;
-    cpu->isar.id_mmfr3 = 0x00000000;
-    cpu->isar.id_isar0 = 0x01101110;
-    cpu->isar.id_isar1 = 0x02112000;
-    cpu->isar.id_isar2 = 0x20232231;
-    cpu->isar.id_isar3 = 0x01111131;
-    cpu->isar.id_isar4 = 0x01310132;
-    cpu->isar.id_isar5 = 0x00000000;
-    cpu->isar.id_isar6 = 0x00000000;
-}
-
-static void cortex_m33_initfn(Object *obj)
-{
-    ARMCPU *cpu = ARM_CPU(obj);
-
-    set_feature(&cpu->env, ARM_FEATURE_V8);
-    set_feature(&cpu->env, ARM_FEATURE_M);
-    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
-    set_feature(&cpu->env, ARM_FEATURE_M_SECURITY);
-    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
-    cpu->midr = 0x410fd213; /* r0p3 */
-    cpu->pmsav7_dregion = 16;
-    cpu->sau_sregion = 8;
-    cpu->isar.mvfr0 = 0x10110021;
-    cpu->isar.mvfr1 = 0x11000011;
-    cpu->isar.mvfr2 = 0x00000040;
-    cpu->isar.id_pfr0 = 0x00000030;
-    cpu->isar.id_pfr1 = 0x00000210;
-    cpu->isar.id_dfr0 = 0x00200000;
-    cpu->id_afr0 = 0x00000000;
-    cpu->isar.id_mmfr0 = 0x00101F40;
-    cpu->isar.id_mmfr1 = 0x00000000;
-    cpu->isar.id_mmfr2 = 0x01000000;
-    cpu->isar.id_mmfr3 = 0x00000000;
-    cpu->isar.id_isar0 = 0x01101110;
-    cpu->isar.id_isar1 = 0x02212000;
-    cpu->isar.id_isar2 = 0x20232232;
-    cpu->isar.id_isar3 = 0x01111131;
-    cpu->isar.id_isar4 = 0x01310132;
-    cpu->isar.id_isar5 = 0x00000000;
-    cpu->isar.id_isar6 = 0x00000000;
-    cpu->clidr = 0x00000000;
-    cpu->ctr = 0x8000c000;
-}
-
-static void cortex_m55_initfn(Object *obj)
-{
-    ARMCPU *cpu = ARM_CPU(obj);
-
-    set_feature(&cpu->env, ARM_FEATURE_V8);
-    set_feature(&cpu->env, ARM_FEATURE_V8_1M);
-    set_feature(&cpu->env, ARM_FEATURE_M);
-    set_feature(&cpu->env, ARM_FEATURE_M_MAIN);
-    set_feature(&cpu->env, ARM_FEATURE_M_SECURITY);
-    set_feature(&cpu->env, ARM_FEATURE_THUMB_DSP);
-    cpu->midr = 0x410fd221; /* r0p1 */
-    cpu->revidr = 0;
-    cpu->pmsav7_dregion = 16;
-    cpu->sau_sregion = 8;
-    /* These are the MVFR* values for the FPU + full MVE configuration */
-    cpu->isar.mvfr0 = 0x10110221;
-    cpu->isar.mvfr1 = 0x12100211;
-    cpu->isar.mvfr2 = 0x00000040;
-    cpu->isar.id_pfr0 = 0x20000030;
-    cpu->isar.id_pfr1 = 0x00000230;
-    cpu->isar.id_dfr0 = 0x10200000;
-    cpu->id_afr0 = 0x00000000;
-    cpu->isar.id_mmfr0 = 0x00111040;
-    cpu->isar.id_mmfr1 = 0x00000000;
-    cpu->isar.id_mmfr2 = 0x01000000;
-    cpu->isar.id_mmfr3 = 0x00000011;
-    cpu->isar.id_isar0 = 0x01103110;
-    cpu->isar.id_isar1 = 0x02212000;
-    cpu->isar.id_isar2 = 0x20232232;
-    cpu->isar.id_isar3 = 0x01111131;
-    cpu->isar.id_isar4 = 0x01310132;
-    cpu->isar.id_isar5 = 0x00000000;
-    cpu->isar.id_isar6 = 0x00000000;
-    cpu->clidr = 0x00000000; /* caches not implemented */
-    cpu->ctr = 0x8303c003;
-}
-
 static const ARMCPRegInfo cortexr5_cp_reginfo[] = {
     /* Dummy the TCM region regs for the moment */
     { .name = "ATCM", .cp = 15, .opc1 = 0, .crn = 9, .crm = 1, .opc2 = 0,
@@ -XXX,XX +XXX,XX @@ static void pxa270c5_initfn(Object *obj)
     cpu->reset_sctlr = 0x00000078;
 }
 
-static const TCGCPUOps arm_v7m_tcg_ops = {
-    .initialize = arm_translate_init,
-    .synchronize_from_tb = arm_cpu_synchronize_from_tb,
-    .debug_excp_handler = arm_debug_excp_handler,
-    .restore_state_to_opc = arm_restore_state_to_opc,
-
-#ifdef CONFIG_USER_ONLY
-    .record_sigsegv = arm_cpu_record_sigsegv,
-    .record_sigbus = arm_cpu_record_sigbus,
-#else
-    .tlb_fill = arm_cpu_tlb_fill,
-    .cpu_exec_interrupt = arm_v7m_cpu_exec_interrupt,
-    .do_interrupt = arm_v7m_cpu_do_interrupt,
-    .do_transaction_failed = arm_cpu_do_transaction_failed,
-    .do_unaligned_access = arm_cpu_do_unaligned_access,
-    .adjust_watchpoint_address = arm_adjust_watchpoint_address,
-    .debug_check_watchpoint = arm_debug_check_watchpoint,
-    .debug_check_breakpoint = arm_debug_check_breakpoint,
-#endif /* !CONFIG_USER_ONLY */
-};
-
-static void arm_v7m_class_init(ObjectClass *oc, void *data)
-{
-    ARMCPUClass *acc = ARM_CPU_CLASS(oc);
-    CPUClass *cc = CPU_CLASS(oc);
-
-    acc->info = data;
-    cc->tcg_ops = &arm_v7m_tcg_ops;
-    cc->gdb_core_xml_file = "arm-m-profile.xml";
-}
-
 #ifndef TARGET_AARCH64
 /*
  * -cpu max: a CPU with as many features enabled as our emulation supports.
@@ -XXX,XX +XXX,XX @@ static const ARMCPUInfo arm_tcg_cpus[] = {
     { .name = "cortex-a8",   .initfn = cortex_a8_initfn },
     { .name = "cortex-a9",   .initfn = cortex_a9_initfn },
     { .name = "cortex-a15",  .initfn = cortex_a15_initfn },
-    { .name = "cortex-m0",   .initfn = cortex_m0_initfn,
-                             .class_init = arm_v7m_class_init },
-    { .name = "cortex-m3",   .initfn = cortex_m3_initfn,
-                             .class_init = arm_v7m_class_init },
-    { .name = "cortex-m4",   .initfn = cortex_m4_initfn,
-                             .class_init = arm_v7m_class_init },
-    { .name = "cortex-m7",   .initfn = cortex_m7_initfn,
-                             .class_init = arm_v7m_class_init },
-    { .name = "cortex-m33",  .initfn = cortex_m33_initfn,
-                             .class_init = arm_v7m_class_init },
-    { .name = "cortex-m55",  .initfn = cortex_m55_initfn,
-                             .class_init = arm_v7m_class_init },
     { .name = "cortex-r5",   .initfn = cortex_r5_initfn },
     { .name = "cortex-r5f",  .initfn = cortex_r5f_initfn },
     { .name = "cortex-r52",  .initfn = cortex_r52_initfn },
diff --git a/target/arm/meson.build b/target/arm/meson.build
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/meson.build
+++ b/target/arm/meson.build
@@ -XXX,XX +XXX,XX @@ arm_system_ss.add(files(
   'ptw.c',
 ))
 
+arm_user_ss = ss.source_set()
+
 subdir('hvf')
 
 if 'CONFIG_TCG' in config_all_accel
@@ -XXX,XX +XXX,XX @@ endif
 
 target_arch += {'arm': arm_ss}
 target_system_arch += {'arm': arm_system_ss}
+target_user_arch += {'arm': arm_user_ss}
diff --git a/target/arm/tcg/meson.build b/target/arm/tcg/meson.build
index XXXXXXX..XXXXXXX 100644
--- a/target/arm/tcg/meson.build
+++ b/target/arm/tcg/meson.build
@@ -XXX,XX +XXX,XX @@ arm_ss.add(when: 'TARGET_AARCH64', if_true: files(
 arm_system_ss.add(files(
   'psci.c',
 ))
+
+arm_system_ss.add(when: 'CONFIG_ARM_V7M', if_true: files('cpu-v7m.c'))
+arm_user_ss.add(when: 'TARGET_AARCH64', if_false: files('cpu-v7m.c'))
-- 
2.34.1