First arm pullreq of 5.0! The following changes since commit 084a398bf8aa7634738e6c6c0103236ee1b3b72f: Merge remote-tracking branch 'remotes/stefanha/tags/block-pull-request' into staging (2019-12-13 18:14:07 +0000) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20191216-1 for you to fetch changes up to f80741d107673f162e3b097fc76a1590036cc9d1: target/arm: ensure we use current exception state after SCR update (2019-12-16 10:52:58 +0000) ---------------------------------------------------------------- target-arm queue: * Add support for Cortex-M7 CPU * exynos4210_gic: Suppress gcc9 format-truncation warnings * aspeed: Various minor bug fixes and improvements * aspeed: Add support for the tacoma-bmc board * Honour HCR_EL32.TID1 and .TID2 trapping requirements * Handle trapping to EL2 of AArch32 VMRS instructions * Handle AArch32 CP15 trapping via HSTR_EL2 * Add support for missing Jazelle system registers * arm/arm-powerctl: set NSACR.{CP11, CP10} bits in arm_set_cpu_on * Add support for DC CVAP & DC CVADP instructions * Fix assertion when SCR.NS is changed in Secure-SVC &c * enable SHPC native hot plug in arm ACPI ---------------------------------------------------------------- Alex Bennée (1): target/arm: ensure we use current exception state after SCR update Beata Michalska (4): tcg: cputlb: Add probe_read Memory: Enable writeback for given memory region migration: ram: Switch to ram block writeback target/arm: Add support for DC CVAP & DC CVADP ins Christophe Lyon (1): target/arm: Add support for cortex-m7 CPU Cédric Le Goater (12): aspeed/i2c: Add support for pool buffer transfers aspeed/i2c: Check SRAM enablement on AST2500 aspeed: Add a DRAM memory region at the SoC level aspeed/i2c: Add support for DMA transfers aspeed/i2c: Add trace events aspeed/smc: Restore default AHB window mapping at reset aspeed/smc: Do not map disabled segment on the AST2600 aspeed/smc: Add AST2600 timings registers aspeed: Remove AspeedBoardConfig array and use AspeedMachineClass aspeed: Add support for the tacoma-bmc board aspeed: Change the "scu" property definition aspeed: Change the "nic" property definition David Gibson (1): exynos4210_gic: Suppress gcc9 format-truncation warnings Heyi Guo (2): hw/arm/acpi: simplify AML bit and/or statement hw/arm/acpi: enable SHPC native hot plug Joel Stanley (4): aspeed/sdmc: Make ast2600 default 1G aspeed/scu: Fix W1C behavior watchdog/aspeed: Improve watchdog timeout message watchdog/aspeed: Fix AST2600 frequency behaviour Marc Zyngier (5): target/arm: Honor HCR_EL2.TID2 trapping requirements target/arm: Honor HCR_EL2.TID1 trapping requirements target/arm: Handle trapping to EL2 of AArch32 VMRS instructions target/arm: Handle AArch32 CP15 trapping via HSTR_EL2 target/arm: Add support for missing Jazelle system registers Niek Linnenbank (1): arm/arm-powerctl: set NSACR.{CP11, CP10} bits in arm_set_cpu_on() PanNengyuan (1): gpio: fix memory leak in aspeed_gpio_init() Philippe Mathieu-Daudé (2): hw/arm/sbsa-ref: Simplify by moving the gic in the machine state hw/arm/virt: Simplify by moving the gic in the machine state include/exec/exec-all.h | 6 + include/exec/memory.h | 6 + include/exec/ram_addr.h | 8 + include/hw/arm/aspeed.h | 24 +-- include/hw/arm/aspeed_soc.h | 1 + include/hw/arm/virt.h | 1 + include/hw/i2c/aspeed_i2c.h | 16 ++ include/hw/ssi/aspeed_smc.h | 1 + include/hw/watchdog/wdt_aspeed.h | 1 + include/qemu/cutils.h | 1 + target/arm/cpu.h | 20 +- target/arm/helper.h | 3 + target/arm/translate.h | 2 + exec.c | 36 ++++ hw/arm/aspeed.c | 271 +++++++++++++---------- hw/arm/aspeed_ast2600.c | 25 ++- hw/arm/aspeed_soc.c | 22 +- hw/arm/sbsa-ref.c | 86 ++++---- hw/arm/virt-acpi-build.c | 21 +- hw/arm/virt.c | 109 +++++----- hw/gpio/aspeed_gpio.c | 1 + hw/i2c/aspeed_i2c.c | 439 +++++++++++++++++++++++++++++++++++--- hw/intc/exynos4210_gic.c | 9 +- hw/misc/aspeed_scu.c | 19 +- hw/misc/aspeed_sdmc.c | 6 +- hw/net/ftgmac100.c | 19 +- hw/ssi/aspeed_smc.c | 63 ++++-- hw/timer/aspeed_timer.c | 17 +- hw/watchdog/wdt_aspeed.c | 41 ++-- linux-user/elfload.c | 2 + memory.c | 12 ++ migration/ram.c | 5 +- target/arm/arm-powerctl.c | 3 + target/arm/cpu.c | 33 +++ target/arm/cpu64.c | 1 + target/arm/helper.c | 170 ++++++++++++++- target/arm/op_helper.c | 22 ++ target/arm/translate-vfp.inc.c | 20 +- target/arm/translate.c | 9 +- target/arm/vfp_helper.c | 29 +++ util/cutils.c | 38 ++++ hw/i2c/trace-events | 9 + tests/data/acpi/virt/DSDT | Bin 18470 -> 18462 bytes tests/data/acpi/virt/DSDT.memhp | Bin 19807 -> 19799 bytes tests/data/acpi/virt/DSDT.numamem | Bin 18470 -> 18462 bytes 45 files changed, 1273 insertions(+), 354 deletions(-) From: Christophe Lyon <christophe.lyon@linaro.org> This is derived from cortex-m4 description, adding DP support and FPv5 instructions with the corresponding flags in isar and mvfr2. Checked that it could successfully execute vrinta.f32 s15, s15 while cortex-m4 emulation rejects it with "illegal instruction". Signed-off-by: Christophe Lyon <christophe.lyon@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20191025090841.10299-1-christophe.lyon@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.c | 33 +++++++++++++++++++++++++++++++++ 1 file changed, 33 insertions(+) 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 cortex_m4_initfn(Object *obj) 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); + set_feature(&cpu->env, ARM_FEATURE_VFP4); + cpu->midr = 0x411fc272; /* r1p2 */ + cpu->pmsav7_dregion = 8; + cpu->isar.mvfr0 = 0x10110221; + cpu->isar.mvfr1 = 0x12000011; + cpu->isar.mvfr2 = 0x00000040; + cpu->id_pfr0 = 0x00000030; + cpu->id_pfr1 = 0x00000200; + cpu->id_dfr0 = 0x00100000; + cpu->id_afr0 = 0x00000000; + cpu->id_mmfr0 = 0x00100030; + cpu->id_mmfr1 = 0x00000000; + cpu->id_mmfr2 = 0x01000000; + cpu->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); @@ -XXX,XX +XXX,XX @@ static const ARMCPUInfo arm_cpus[] = { .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-r5", .initfn = cortex_r5_initfn }, -- 2.20.1 From: David Gibson <david@gibson.dropbear.id.au> exynos4210_gic_realize() prints the number of cpus into some temporary buffers, but it only allows 3 bytes space for it. That's plenty: existing machines will only ever set this value to EXYNOS4210_NCPUS (2). But the compiler can't always figure that out, so some[*] gcc9 versions emit -Wformat-truncation warnings. We can fix that by hinting the constraint to the compiler with a suitably placed assert(). [*] The bizarre thing here, is that I've long gotten these warnings compiling in a 32-bit x86 container as host - Fedora 30 with gcc-9.2.1-1.fc30.i686 - but it compiles just fine on my normal x86_64 host - Fedora 30 with and gcc-9.2.1-1.fc30.x86_64. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> [PMM: deleted stray blank line] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/intc/exynos4210_gic.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/hw/intc/exynos4210_gic.c b/hw/intc/exynos4210_gic.c index XXXXXXX..XXXXXXX 100644 --- a/hw/intc/exynos4210_gic.c +++ b/hw/intc/exynos4210_gic.c @@ -XXX,XX +XXX,XX @@ static void exynos4210_gic_realize(DeviceState *dev, Error **errp) char cpu_alias_name[sizeof(cpu_prefix) + 3]; char dist_alias_name[sizeof(cpu_prefix) + 3]; SysBusDevice *gicbusdev; + uint32_t n = s->num_cpu; uint32_t i; s->gic = qdev_create(NULL, "arm_gic"); @@ -XXX,XX +XXX,XX @@ static void exynos4210_gic_realize(DeviceState *dev, Error **errp) memory_region_init(&s->dist_container, obj, "exynos4210-dist-container", EXYNOS4210_EXT_GIC_DIST_REGION_SIZE); - for (i = 0; i < s->num_cpu; i++) { + /* + * This clues in gcc that our on-stack buffers do, in fact have + * enough room for the cpu numbers. gcc 9.2.1 on 32-bit x86 + * doesn't figure this out, otherwise and gives spurious warnings. + */ + assert(n <= EXYNOS4210_NCPUS); + for (i = 0; i < n; i++) { /* Map CPU interface per SMP Core */ sprintf(cpu_alias_name, "%s%x", cpu_prefix, i); memory_region_init_alias(&s->cpu_alias[i], obj, -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> The Aspeed I2C controller can operate in different transfer modes : - Byte Buffer mode, using a dedicated register to transfer a byte. This is what the model supports today. - Pool Buffer mode, using an internal SRAM to transfer multiple bytes in the same command sequence. Each SoC has different SRAM characteristics. On the AST2400, 2048 bytes of SRAM are available at offset 0x800 of the controller AHB window. The pool buffer can be configured from 1 to 256 bytes per bus. On the AST2500, the SRAM is at offset 0x200 and the pool buffer is of 16 bytes per bus. On the AST2600, the SRAM is at offset 0xC00 and the pool buffer is of 32 bytes per bus. It can be splitted in two for TX and RX but the current model does not add support for it as it it unused by known drivers. Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Tested-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-2-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/i2c/aspeed_i2c.h | 8 ++ hw/i2c/aspeed_i2c.c | 197 ++++++++++++++++++++++++++++++++---- 2 files changed, 186 insertions(+), 19 deletions(-) diff --git a/include/hw/i2c/aspeed_i2c.h b/include/hw/i2c/aspeed_i2c.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/i2c/aspeed_i2c.h +++ b/include/hw/i2c/aspeed_i2c.h @@ -XXX,XX +XXX,XX @@ OBJECT_CHECK(AspeedI2CState, (obj), TYPE_ASPEED_I2C) #define ASPEED_I2C_NR_BUSSES 16 +#define ASPEED_I2C_MAX_POOL_SIZE 0x800 struct AspeedI2CState; @@ -XXX,XX +XXX,XX @@ typedef struct AspeedI2CBus { uint32_t intr_status; uint32_t cmd; uint32_t buf; + uint32_t pool_ctrl; } AspeedI2CBus; typedef struct AspeedI2CState { @@ -XXX,XX +XXX,XX @@ typedef struct AspeedI2CState { qemu_irq irq; uint32_t intr_status; + MemoryRegion pool_iomem; + uint8_t pool[ASPEED_I2C_MAX_POOL_SIZE]; AspeedI2CBus busses[ASPEED_I2C_NR_BUSSES]; } AspeedI2CState; @@ -XXX,XX +XXX,XX @@ typedef struct AspeedI2CClass { uint8_t reg_size; uint8_t gap; qemu_irq (*bus_get_irq)(AspeedI2CBus *); + + uint64_t pool_size; + hwaddr pool_base; + uint8_t *(*bus_pool_base)(AspeedI2CBus *); } AspeedI2CClass; I2CBus *aspeed_i2c_get_bus(DeviceState *dev, int busnr); diff --git a/hw/i2c/aspeed_i2c.c b/hw/i2c/aspeed_i2c.c index XXXXXXX..XXXXXXX 100644 --- a/hw/i2c/aspeed_i2c.c +++ b/hw/i2c/aspeed_i2c.c @@ -XXX,XX +XXX,XX @@ /* I2C Device (Bus) Register */ #define I2CD_FUN_CTRL_REG 0x00 /* I2CD Function Control */ -#define I2CD_BUFF_SEL_MASK (0x7 << 20) -#define I2CD_BUFF_SEL(x) (x << 20) +#define I2CD_POOL_PAGE_SEL(x) (((x) >> 20) & 0x7) /* AST2400 */ #define I2CD_M_SDA_LOCK_EN (0x1 << 16) #define I2CD_MULTI_MASTER_DIS (0x1 << 15) #define I2CD_M_SCL_DRIVE_EN (0x1 << 14) @@ -XXX,XX +XXX,XX @@ #define I2CD_SCL_O_OUT_DIR (0x1 << 12) #define I2CD_BUS_RECOVER_CMD_EN (0x1 << 11) #define I2CD_S_ALT_EN (0x1 << 10) -#define I2CD_RX_DMA_ENABLE (0x1 << 9) -#define I2CD_TX_DMA_ENABLE (0x1 << 8) /* Command Bit */ +#define I2CD_RX_DMA_ENABLE (0x1 << 9) +#define I2CD_TX_DMA_ENABLE (0x1 << 8) +#define I2CD_RX_BUFF_ENABLE (0x1 << 7) +#define I2CD_TX_BUFF_ENABLE (0x1 << 6) #define I2CD_M_STOP_CMD (0x1 << 5) #define I2CD_M_S_RX_CMD_LAST (0x1 << 4) #define I2CD_M_RX_CMD (0x1 << 3) @@ -XXX,XX +XXX,XX @@ #define I2CD_M_START_CMD (0x1) #define I2CD_DEV_ADDR_REG 0x18 /* Slave Device Address */ -#define I2CD_BUF_CTRL_REG 0x1c /* Pool Buffer Control */ +#define I2CD_POOL_CTRL_REG 0x1c /* Pool Buffer Control */ +#define I2CD_POOL_RX_COUNT(x) (((x) >> 24) & 0xff) +#define I2CD_POOL_RX_SIZE(x) ((((x) >> 16) & 0xff) + 1) +#define I2CD_POOL_TX_COUNT(x) ((((x) >> 8) & 0xff) + 1) +#define I2CD_POOL_OFFSET(x) (((x) & 0x3f) << 2) /* AST2400 */ #define I2CD_BYTE_BUF_REG 0x20 /* Transmit/Receive Byte Buffer */ #define I2CD_BYTE_BUF_TX_SHIFT 0 #define I2CD_BYTE_BUF_TX_MASK 0xff @@ -XXX,XX +XXX,XX @@ static uint64_t aspeed_i2c_bus_read(void *opaque, hwaddr offset, return bus->intr_ctrl; case I2CD_INTR_STS_REG: return bus->intr_status; + case I2CD_POOL_CTRL_REG: + return bus->pool_ctrl; case I2CD_BYTE_BUF_REG: return bus->buf; case I2CD_CMD_REG: @@ -XXX,XX +XXX,XX @@ static uint8_t aspeed_i2c_get_state(AspeedI2CBus *bus) return (bus->cmd >> I2CD_TX_STATE_SHIFT) & I2CD_TX_STATE_MASK; } +static int aspeed_i2c_bus_send(AspeedI2CBus *bus, uint8_t pool_start) +{ + AspeedI2CClass *aic = ASPEED_I2C_GET_CLASS(bus->controller); + int ret = -1; + int i; + + if (bus->cmd & I2CD_TX_BUFF_ENABLE) { + for (i = pool_start; i < I2CD_POOL_TX_COUNT(bus->pool_ctrl); i++) { + uint8_t *pool_base = aic->bus_pool_base(bus); + + ret = i2c_send(bus->bus, pool_base[i]); + if (ret) { + break; + } + } + bus->cmd &= ~I2CD_TX_BUFF_ENABLE; + } else { + ret = i2c_send(bus->bus, bus->buf); + } + + return ret; +} + +static void aspeed_i2c_bus_recv(AspeedI2CBus *bus) +{ + AspeedI2CState *s = bus->controller; + AspeedI2CClass *aic = ASPEED_I2C_GET_CLASS(s); + uint8_t data; + int i; + + if (bus->cmd & I2CD_RX_BUFF_ENABLE) { + uint8_t *pool_base = aic->bus_pool_base(bus); + + for (i = 0; i < I2CD_POOL_RX_SIZE(bus->pool_ctrl); i++) { + pool_base[i] = i2c_recv(bus->bus); + } + + /* Update RX count */ + bus->pool_ctrl &= ~(0xff << 24); + bus->pool_ctrl |= (i & 0xff) << 24; + bus->cmd &= ~I2CD_RX_BUFF_ENABLE; + } else { + data = i2c_recv(bus->bus); + bus->buf = (data & I2CD_BYTE_BUF_RX_MASK) << I2CD_BYTE_BUF_RX_SHIFT; + } +} + static void aspeed_i2c_handle_rx_cmd(AspeedI2CBus *bus) { - uint8_t ret; - aspeed_i2c_set_state(bus, I2CD_MRXD); - ret = i2c_recv(bus->bus); + aspeed_i2c_bus_recv(bus); bus->intr_status |= I2CD_INTR_RX_DONE; - bus->buf = (ret & I2CD_BYTE_BUF_RX_MASK) << I2CD_BYTE_BUF_RX_SHIFT; if (bus->cmd & I2CD_M_S_RX_CMD_LAST) { i2c_nack(bus->bus); } @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_handle_rx_cmd(AspeedI2CBus *bus) aspeed_i2c_set_state(bus, I2CD_MACTIVE); } +static uint8_t aspeed_i2c_get_addr(AspeedI2CBus *bus) +{ + AspeedI2CClass *aic = ASPEED_I2C_GET_CLASS(bus->controller); + + if (bus->cmd & I2CD_TX_BUFF_ENABLE) { + uint8_t *pool_base = aic->bus_pool_base(bus); + + return pool_base[0]; + } else { + return bus->buf; + } +} + /* * The state machine needs some refinement. It is only used to track * invalid STOP commands for the moment. */ static void aspeed_i2c_bus_handle_cmd(AspeedI2CBus *bus, uint64_t value) { + uint8_t pool_start = 0; + bus->cmd &= ~0xFFFF; bus->cmd |= value & 0xFFFF; if (bus->cmd & I2CD_M_START_CMD) { uint8_t state = aspeed_i2c_get_state(bus) & I2CD_MACTIVE ? I2CD_MSTARTR : I2CD_MSTART; + uint8_t addr; aspeed_i2c_set_state(bus, state); - if (i2c_start_transfer(bus->bus, extract32(bus->buf, 1, 7), - extract32(bus->buf, 0, 1))) { + addr = aspeed_i2c_get_addr(bus); + + if (i2c_start_transfer(bus->bus, extract32(addr, 1, 7), + extract32(addr, 0, 1))) { bus->intr_status |= I2CD_INTR_TX_NAK; } else { bus->intr_status |= I2CD_INTR_TX_ACK; } - /* START command is also a TX command, as the slave address is - * sent on the bus */ - bus->cmd &= ~(I2CD_M_START_CMD | I2CD_M_TX_CMD); + bus->cmd &= ~I2CD_M_START_CMD; + + /* + * The START command is also a TX command, as the slave + * address is sent on the bus. Drop the TX flag if nothing + * else needs to be sent in this sequence. + */ + if (bus->cmd & I2CD_TX_BUFF_ENABLE) { + if (I2CD_POOL_TX_COUNT(bus->pool_ctrl) == 1) { + bus->cmd &= ~I2CD_M_TX_CMD; + } else { + /* + * Increase the start index in the TX pool buffer to + * skip the address byte. + */ + pool_start++; + } + } else { + bus->cmd &= ~I2CD_M_TX_CMD; + } /* No slave found */ if (!i2c_bus_busy(bus->bus)) { @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_bus_handle_cmd(AspeedI2CBus *bus, uint64_t value) if (bus->cmd & I2CD_M_TX_CMD) { aspeed_i2c_set_state(bus, I2CD_MTXD); - if (i2c_send(bus->bus, bus->buf)) { + if (aspeed_i2c_bus_send(bus, pool_start)) { bus->intr_status |= (I2CD_INTR_TX_NAK); i2c_end_transfer(bus->bus); } else { @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_bus_write(void *opaque, hwaddr offset, qemu_log_mask(LOG_UNIMP, "%s: slave mode not implemented\n", __func__); break; + case I2CD_POOL_CTRL_REG: + bus->pool_ctrl &= ~0xffffff; + bus->pool_ctrl |= (value & 0xffffff); + break; + case I2CD_BYTE_BUF_REG: bus->buf = (value & I2CD_BYTE_BUF_TX_MASK) << I2CD_BYTE_BUF_TX_SHIFT; break; @@ -XXX,XX +XXX,XX @@ static const MemoryRegionOps aspeed_i2c_ctrl_ops = { .endianness = DEVICE_LITTLE_ENDIAN, }; +static uint64_t aspeed_i2c_pool_read(void *opaque, hwaddr offset, + unsigned size) +{ + AspeedI2CState *s = opaque; + uint64_t ret = 0; + int i; + + for (i = 0; i < size; i++) { + ret |= (uint64_t) s->pool[offset + i] << (8 * i); + } + + return ret; +} + +static void aspeed_i2c_pool_write(void *opaque, hwaddr offset, + uint64_t value, unsigned size) +{ + AspeedI2CState *s = opaque; + int i; + + for (i = 0; i < size; i++) { + s->pool[offset + i] = (value >> (8 * i)) & 0xFF; + } +} + +static const MemoryRegionOps aspeed_i2c_pool_ops = { + .read = aspeed_i2c_pool_read, + .write = aspeed_i2c_pool_write, + .endianness = DEVICE_LITTLE_ENDIAN, + .valid = { + .min_access_size = 1, + .max_access_size = 4, + }, +}; + static const VMStateDescription aspeed_i2c_bus_vmstate = { .name = TYPE_ASPEED_I2C, - .version_id = 1, - .minimum_version_id = 1, + .version_id = 2, + .minimum_version_id = 2, .fields = (VMStateField[]) { VMSTATE_UINT8(id, AspeedI2CBus), VMSTATE_UINT32(ctrl, AspeedI2CBus), @@ -XXX,XX +XXX,XX @@ static const VMStateDescription aspeed_i2c_bus_vmstate = { VMSTATE_UINT32(intr_status, AspeedI2CBus), VMSTATE_UINT32(cmd, AspeedI2CBus), VMSTATE_UINT32(buf, AspeedI2CBus), + VMSTATE_UINT32(pool_ctrl, AspeedI2CBus), VMSTATE_END_OF_LIST() } }; static const VMStateDescription aspeed_i2c_vmstate = { .name = TYPE_ASPEED_I2C, - .version_id = 1, - .minimum_version_id = 1, + .version_id = 2, + .minimum_version_id = 2, .fields = (VMStateField[]) { VMSTATE_UINT32(intr_status, AspeedI2CState), VMSTATE_STRUCT_ARRAY(busses, AspeedI2CState, ASPEED_I2C_NR_BUSSES, 1, aspeed_i2c_bus_vmstate, AspeedI2CBus), + VMSTATE_UINT8_ARRAY(pool, AspeedI2CState, ASPEED_I2C_MAX_POOL_SIZE), VMSTATE_END_OF_LIST() } }; @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_realize(DeviceState *dev, Error **errp) memory_region_add_subregion(&s->iomem, aic->reg_size * (i + offset), &s->busses[i].mr); } + + memory_region_init_io(&s->pool_iomem, OBJECT(s), &aspeed_i2c_pool_ops, s, + "aspeed.i2c-pool", aic->pool_size); + memory_region_add_subregion(&s->iomem, aic->pool_base, &s->pool_iomem); } static void aspeed_i2c_class_init(ObjectClass *klass, void *data) @@ -XXX,XX +XXX,XX @@ static qemu_irq aspeed_2400_i2c_bus_get_irq(AspeedI2CBus *bus) return bus->controller->irq; } +static uint8_t *aspeed_2400_i2c_bus_pool_base(AspeedI2CBus *bus) +{ + uint8_t *pool_page = + &bus->controller->pool[I2CD_POOL_PAGE_SEL(bus->ctrl) * 0x100]; + + return &pool_page[I2CD_POOL_OFFSET(bus->pool_ctrl)]; +} + static void aspeed_2400_i2c_class_init(ObjectClass *klass, void *data) { DeviceClass *dc = DEVICE_CLASS(klass); @@ -XXX,XX +XXX,XX @@ static void aspeed_2400_i2c_class_init(ObjectClass *klass, void *data) aic->reg_size = 0x40; aic->gap = 7; aic->bus_get_irq = aspeed_2400_i2c_bus_get_irq; + aic->pool_size = 0x800; + aic->pool_base = 0x800; + aic->bus_pool_base = aspeed_2400_i2c_bus_pool_base; } static const TypeInfo aspeed_2400_i2c_info = { @@ -XXX,XX +XXX,XX @@ static qemu_irq aspeed_2500_i2c_bus_get_irq(AspeedI2CBus *bus) return bus->controller->irq; } +static uint8_t *aspeed_2500_i2c_bus_pool_base(AspeedI2CBus *bus) +{ + return &bus->controller->pool[bus->id * 0x10]; +} + static void aspeed_2500_i2c_class_init(ObjectClass *klass, void *data) { DeviceClass *dc = DEVICE_CLASS(klass); @@ -XXX,XX +XXX,XX @@ static void aspeed_2500_i2c_class_init(ObjectClass *klass, void *data) aic->reg_size = 0x40; aic->gap = 7; aic->bus_get_irq = aspeed_2500_i2c_bus_get_irq; + aic->pool_size = 0x100; + aic->pool_base = 0x200; + aic->bus_pool_base = aspeed_2500_i2c_bus_pool_base; } static const TypeInfo aspeed_2500_i2c_info = { @@ -XXX,XX +XXX,XX @@ static qemu_irq aspeed_2600_i2c_bus_get_irq(AspeedI2CBus *bus) return bus->irq; } +static uint8_t *aspeed_2600_i2c_bus_pool_base(AspeedI2CBus *bus) +{ + return &bus->controller->pool[bus->id * 0x20]; +} + static void aspeed_2600_i2c_class_init(ObjectClass *klass, void *data) { DeviceClass *dc = DEVICE_CLASS(klass); @@ -XXX,XX +XXX,XX @@ static void aspeed_2600_i2c_class_init(ObjectClass *klass, void *data) aic->reg_size = 0x80; aic->gap = -1; /* no gap */ aic->bus_get_irq = aspeed_2600_i2c_bus_get_irq; + aic->pool_size = 0x200; + aic->pool_base = 0xC00; + aic->bus_pool_base = aspeed_2600_i2c_bus_pool_base; } static const TypeInfo aspeed_2600_i2c_info = { -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> The SRAM must be enabled before using the Buffer Pool mode or the DMA mode. This is not required on other SoCs. Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Tested-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-3-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/i2c/aspeed_i2c.h | 3 +++ hw/i2c/aspeed_i2c.c | 37 +++++++++++++++++++++++++++++++++++++ 2 files changed, 40 insertions(+) diff --git a/include/hw/i2c/aspeed_i2c.h b/include/hw/i2c/aspeed_i2c.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/i2c/aspeed_i2c.h +++ b/include/hw/i2c/aspeed_i2c.h @@ -XXX,XX +XXX,XX @@ typedef struct AspeedI2CState { qemu_irq irq; uint32_t intr_status; + uint32_t ctrl_global; MemoryRegion pool_iomem; uint8_t pool[ASPEED_I2C_MAX_POOL_SIZE]; @@ -XXX,XX +XXX,XX @@ typedef struct AspeedI2CClass { uint64_t pool_size; hwaddr pool_base; uint8_t *(*bus_pool_base)(AspeedI2CBus *); + bool check_sram; + } AspeedI2CClass; I2CBus *aspeed_i2c_get_bus(DeviceState *dev, int busnr); diff --git a/hw/i2c/aspeed_i2c.c b/hw/i2c/aspeed_i2c.c index XXXXXXX..XXXXXXX 100644 --- a/hw/i2c/aspeed_i2c.c +++ b/hw/i2c/aspeed_i2c.c @@ -XXX,XX +XXX,XX @@ #define I2C_CTRL_STATUS 0x00 /* Device Interrupt Status */ #define I2C_CTRL_ASSIGN 0x08 /* Device Interrupt Target Assignment */ +#define I2C_CTRL_GLOBAL 0x0C /* Global Control Register */ +#define I2C_CTRL_SRAM_EN BIT(0) /* I2C Device (Bus) Register */ @@ -XXX,XX +XXX,XX @@ static uint8_t aspeed_i2c_get_addr(AspeedI2CBus *bus) } } +static bool aspeed_i2c_check_sram(AspeedI2CBus *bus) +{ + AspeedI2CState *s = bus->controller; + AspeedI2CClass *aic = ASPEED_I2C_GET_CLASS(s); + + if (!aic->check_sram) { + return true; + } + + /* + * AST2500: SRAM must be enabled before using the Buffer Pool or + * DMA mode. + */ + if (!(s->ctrl_global & I2C_CTRL_SRAM_EN) && + (bus->cmd & (I2CD_RX_DMA_ENABLE | I2CD_TX_DMA_ENABLE | + I2CD_RX_BUFF_ENABLE | I2CD_TX_BUFF_ENABLE))) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: SRAM is not enabled\n", __func__); + return false; + } + + return true; +} + /* * The state machine needs some refinement. It is only used to track * invalid STOP commands for the moment. @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_bus_handle_cmd(AspeedI2CBus *bus, uint64_t value) bus->cmd &= ~0xFFFF; bus->cmd |= value & 0xFFFF; + if (!aspeed_i2c_check_sram(bus)) { + return; + } + if (bus->cmd & I2CD_M_START_CMD) { uint8_t state = aspeed_i2c_get_state(bus) & I2CD_MACTIVE ? I2CD_MSTARTR : I2CD_MSTART; @@ -XXX,XX +XXX,XX @@ static uint64_t aspeed_i2c_ctrl_read(void *opaque, hwaddr offset, switch (offset) { case I2C_CTRL_STATUS: return s->intr_status; + case I2C_CTRL_GLOBAL: + return s->ctrl_global; default: qemu_log_mask(LOG_GUEST_ERROR, "%s: Bad offset 0x%" HWADDR_PRIx "\n", __func__, offset); @@ -XXX,XX +XXX,XX @@ static uint64_t aspeed_i2c_ctrl_read(void *opaque, hwaddr offset, static void aspeed_i2c_ctrl_write(void *opaque, hwaddr offset, uint64_t value, unsigned size) { + AspeedI2CState *s = opaque; + switch (offset) { + case I2C_CTRL_GLOBAL: + s->ctrl_global = value; + break; case I2C_CTRL_STATUS: default: qemu_log_mask(LOG_GUEST_ERROR, "%s: Bad offset 0x%" HWADDR_PRIx "\n", @@ -XXX,XX +XXX,XX @@ static void aspeed_2500_i2c_class_init(ObjectClass *klass, void *data) aic->pool_size = 0x100; aic->pool_base = 0x200; aic->bus_pool_base = aspeed_2500_i2c_bus_pool_base; + aic->check_sram = true; } static const TypeInfo aspeed_2500_i2c_info = { -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> Currently, we link the DRAM memory region to the FMC model (for DMAs) through a property alias at the SoC level. The I2C model will need a similar region for DMA support, add a DRAM region property at the SoC level for both model to use. Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Tested-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-4-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/aspeed_soc.h | 1 + hw/arm/aspeed_ast2600.c | 7 +++++-- hw/arm/aspeed_soc.c | 9 +++++++-- 3 files changed, 13 insertions(+), 4 deletions(-) diff --git a/include/hw/arm/aspeed_soc.h b/include/hw/arm/aspeed_soc.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/aspeed_soc.h +++ b/include/hw/arm/aspeed_soc.h @@ -XXX,XX +XXX,XX @@ typedef struct AspeedSoCState { ARMCPU cpu[ASPEED_CPUS_NUM]; uint32_t num_cpus; A15MPPrivState a7mpcore; + MemoryRegion *dram_mr; MemoryRegion sram; AspeedVICState vic; AspeedRtcState rtc; diff --git a/hw/arm/aspeed_ast2600.c b/hw/arm/aspeed_ast2600.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/aspeed_ast2600.c +++ b/hw/arm/aspeed_ast2600.c @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_ast2600_init(Object *obj) typename); object_property_add_alias(obj, "num-cs", OBJECT(&s->fmc), "num-cs", &error_abort); - object_property_add_alias(obj, "dram", OBJECT(&s->fmc), "dram", - &error_abort); for (i = 0; i < sc->spis_num; i++) { snprintf(typename, sizeof(typename), "aspeed.spi%d-%s", i + 1, socname); @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_ast2600_realize(DeviceState *dev, Error **errp) } /* FMC, The number of CS is set at the board level */ + object_property_set_link(OBJECT(&s->fmc), OBJECT(s->dram_mr), "dram", &err); + if (err) { + error_propagate(errp, err); + return; + } object_property_set_int(OBJECT(&s->fmc), sc->memmap[ASPEED_SDRAM], "sdram-base", &err); if (err) { diff --git a/hw/arm/aspeed_soc.c b/hw/arm/aspeed_soc.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/aspeed_soc.c +++ b/hw/arm/aspeed_soc.c @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_init(Object *obj) typename); object_property_add_alias(obj, "num-cs", OBJECT(&s->fmc), "num-cs", &error_abort); - object_property_add_alias(obj, "dram", OBJECT(&s->fmc), "dram", - &error_abort); for (i = 0; i < sc->spis_num; i++) { snprintf(typename, sizeof(typename), "aspeed.spi%d-%s", i + 1, socname); @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp) aspeed_soc_get_irq(s, ASPEED_I2C)); /* FMC, The number of CS is set at the board level */ + object_property_set_link(OBJECT(&s->fmc), OBJECT(s->dram_mr), "dram", &err); + if (err) { + error_propagate(errp, err); + return; + } object_property_set_int(OBJECT(&s->fmc), sc->memmap[ASPEED_SDRAM], "sdram-base", &err); if (err) { @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp) } static Property aspeed_soc_properties[] = { DEFINE_PROP_UINT32("num-cpus", AspeedSoCState, num_cpus, 0), + DEFINE_PROP_LINK("dram", AspeedSoCState, dram_mr, TYPE_MEMORY_REGION, + MemoryRegion *), DEFINE_PROP_END_OF_LIST(), }; -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> The I2C controller of the Aspeed AST2500 and AST2600 SoCs supports DMA transfers to and from DRAM. A pair of registers defines the buffer address and the length of the DMA transfer. The address should be aligned on 4 bytes and the maximum length should not exceed 4K. The receive or transmit DMA transfer can then be initiated with specific bits in the Command/Status register of the controller. Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Tested-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-5-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/i2c/aspeed_i2c.h | 5 ++ hw/arm/aspeed_ast2600.c | 5 ++ hw/arm/aspeed_soc.c | 5 ++ hw/i2c/aspeed_i2c.c | 126 +++++++++++++++++++++++++++++++++++- 4 files changed, 138 insertions(+), 3 deletions(-) diff --git a/include/hw/i2c/aspeed_i2c.h b/include/hw/i2c/aspeed_i2c.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/i2c/aspeed_i2c.h +++ b/include/hw/i2c/aspeed_i2c.h @@ -XXX,XX +XXX,XX @@ typedef struct AspeedI2CBus { uint32_t cmd; uint32_t buf; uint32_t pool_ctrl; + uint32_t dma_addr; + uint32_t dma_len; } AspeedI2CBus; typedef struct AspeedI2CState { @@ -XXX,XX +XXX,XX @@ typedef struct AspeedI2CState { uint8_t pool[ASPEED_I2C_MAX_POOL_SIZE]; AspeedI2CBus busses[ASPEED_I2C_NR_BUSSES]; + MemoryRegion *dram_mr; + AddressSpace dram_as; } AspeedI2CState; #define ASPEED_I2C_CLASS(klass) \ @@ -XXX,XX +XXX,XX @@ typedef struct AspeedI2CClass { hwaddr pool_base; uint8_t *(*bus_pool_base)(AspeedI2CBus *); bool check_sram; + bool has_dma; } AspeedI2CClass; diff --git a/hw/arm/aspeed_ast2600.c b/hw/arm/aspeed_ast2600.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/aspeed_ast2600.c +++ b/hw/arm/aspeed_ast2600.c @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_ast2600_realize(DeviceState *dev, Error **errp) } /* I2C */ + object_property_set_link(OBJECT(&s->i2c), OBJECT(s->dram_mr), "dram", &err); + if (err) { + error_propagate(errp, err); + return; + } object_property_set_bool(OBJECT(&s->i2c), true, "realized", &err); if (err) { error_propagate(errp, err); diff --git a/hw/arm/aspeed_soc.c b/hw/arm/aspeed_soc.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/aspeed_soc.c +++ b/hw/arm/aspeed_soc.c @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp) } /* I2C */ + object_property_set_link(OBJECT(&s->i2c), OBJECT(s->dram_mr), "dram", &err); + if (err) { + error_propagate(errp, err); + return; + } object_property_set_bool(OBJECT(&s->i2c), true, "realized", &err); if (err) { error_propagate(errp, err); diff --git a/hw/i2c/aspeed_i2c.c b/hw/i2c/aspeed_i2c.c index XXXXXXX..XXXXXXX 100644 --- a/hw/i2c/aspeed_i2c.c +++ b/hw/i2c/aspeed_i2c.c @@ -XXX,XX +XXX,XX @@ #include "migration/vmstate.h" #include "qemu/log.h" #include "qemu/module.h" +#include "qemu/error-report.h" +#include "qapi/error.h" #include "hw/i2c/aspeed_i2c.h" #include "hw/irq.h" +#include "hw/qdev-properties.h" /* I2C Global Register */ @@ -XXX,XX +XXX,XX @@ #define I2CD_BYTE_BUF_TX_MASK 0xff #define I2CD_BYTE_BUF_RX_SHIFT 8 #define I2CD_BYTE_BUF_RX_MASK 0xff - +#define I2CD_DMA_ADDR 0x24 /* DMA Buffer Address */ +#define I2CD_DMA_LEN 0x28 /* DMA Transfer Length < 4KB */ static inline bool aspeed_i2c_bus_is_master(AspeedI2CBus *bus) { @@ -XXX,XX +XXX,XX @@ static uint64_t aspeed_i2c_bus_read(void *opaque, hwaddr offset, unsigned size) { AspeedI2CBus *bus = opaque; + AspeedI2CClass *aic = ASPEED_I2C_GET_CLASS(bus->controller); switch (offset) { case I2CD_FUN_CTRL_REG: @@ -XXX,XX +XXX,XX @@ static uint64_t aspeed_i2c_bus_read(void *opaque, hwaddr offset, return bus->buf; case I2CD_CMD_REG: return bus->cmd | (i2c_bus_busy(bus->bus) << 16); + case I2CD_DMA_ADDR: + if (!aic->has_dma) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: No DMA support\n", __func__); + return -1; + } + return bus->dma_addr; + case I2CD_DMA_LEN: + if (!aic->has_dma) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: No DMA support\n", __func__); + return -1; + } + return bus->dma_len; default: qemu_log_mask(LOG_GUEST_ERROR, "%s: Bad offset 0x%" HWADDR_PRIx "\n", __func__, offset); @@ -XXX,XX +XXX,XX @@ static uint8_t aspeed_i2c_get_state(AspeedI2CBus *bus) return (bus->cmd >> I2CD_TX_STATE_SHIFT) & I2CD_TX_STATE_MASK; } +static int aspeed_i2c_dma_read(AspeedI2CBus *bus, uint8_t *data) +{ + MemTxResult result; + AspeedI2CState *s = bus->controller; + + result = address_space_read(&s->dram_as, bus->dma_addr, + MEMTXATTRS_UNSPECIFIED, data, 1); + if (result != MEMTX_OK) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: DRAM read failed @%08x\n", + __func__, bus->dma_addr); + return -1; + } + + bus->dma_addr++; + bus->dma_len--; + return 0; +} + static int aspeed_i2c_bus_send(AspeedI2CBus *bus, uint8_t pool_start) { AspeedI2CClass *aic = ASPEED_I2C_GET_CLASS(bus->controller); @@ -XXX,XX +XXX,XX @@ static int aspeed_i2c_bus_send(AspeedI2CBus *bus, uint8_t pool_start) } } bus->cmd &= ~I2CD_TX_BUFF_ENABLE; + } else if (bus->cmd & I2CD_TX_DMA_ENABLE) { + while (bus->dma_len) { + uint8_t data; + aspeed_i2c_dma_read(bus, &data); + ret = i2c_send(bus->bus, data); + if (ret) { + break; + } + } + bus->cmd &= ~I2CD_TX_DMA_ENABLE; } else { ret = i2c_send(bus->bus, bus->buf); } @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_bus_recv(AspeedI2CBus *bus) bus->pool_ctrl &= ~(0xff << 24); bus->pool_ctrl |= (i & 0xff) << 24; bus->cmd &= ~I2CD_RX_BUFF_ENABLE; + } else if (bus->cmd & I2CD_RX_DMA_ENABLE) { + uint8_t data; + + while (bus->dma_len) { + MemTxResult result; + + data = i2c_recv(bus->bus); + result = address_space_write(&s->dram_as, bus->dma_addr, + MEMTXATTRS_UNSPECIFIED, &data, 1); + if (result != MEMTX_OK) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: DRAM write failed @%08x\n", + __func__, bus->dma_addr); + return; + } + bus->dma_addr++; + bus->dma_len--; + } + bus->cmd &= ~I2CD_RX_DMA_ENABLE; } else { data = i2c_recv(bus->bus); bus->buf = (data & I2CD_BYTE_BUF_RX_MASK) << I2CD_BYTE_BUF_RX_SHIFT; @@ -XXX,XX +XXX,XX @@ static uint8_t aspeed_i2c_get_addr(AspeedI2CBus *bus) uint8_t *pool_base = aic->bus_pool_base(bus); return pool_base[0]; + } else if (bus->cmd & I2CD_TX_DMA_ENABLE) { + uint8_t data; + + aspeed_i2c_dma_read(bus, &data); + return data; } else { return bus->buf; } @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_bus_handle_cmd(AspeedI2CBus *bus, uint64_t value) */ pool_start++; } + } else if (bus->cmd & I2CD_TX_DMA_ENABLE) { + if (bus->dma_len == 0) { + bus->cmd &= ~I2CD_M_TX_CMD; + } } else { bus->cmd &= ~I2CD_M_TX_CMD; } @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_bus_write(void *opaque, hwaddr offset, break; } + if (!aic->has_dma && + value & (I2CD_RX_DMA_ENABLE | I2CD_TX_DMA_ENABLE)) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: No DMA support\n", __func__); + break; + } + aspeed_i2c_bus_handle_cmd(bus, value); aspeed_i2c_bus_raise_interrupt(bus); break; + case I2CD_DMA_ADDR: + if (!aic->has_dma) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: No DMA support\n", __func__); + break; + } + + bus->dma_addr = value & 0xfffffffc; + break; + + case I2CD_DMA_LEN: + if (!aic->has_dma) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: No DMA support\n", __func__); + break; + } + + bus->dma_len = value & 0xfff; + if (!bus->dma_len) { + qemu_log_mask(LOG_UNIMP, "%s: invalid DMA length\n", __func__); + } + break; default: qemu_log_mask(LOG_GUEST_ERROR, "%s: Bad offset 0x%" HWADDR_PRIx "\n", @@ -XXX,XX +XXX,XX @@ static const MemoryRegionOps aspeed_i2c_pool_ops = { static const VMStateDescription aspeed_i2c_bus_vmstate = { .name = TYPE_ASPEED_I2C, - .version_id = 2, - .minimum_version_id = 2, + .version_id = 3, + .minimum_version_id = 3, .fields = (VMStateField[]) { VMSTATE_UINT8(id, AspeedI2CBus), VMSTATE_UINT32(ctrl, AspeedI2CBus), @@ -XXX,XX +XXX,XX @@ static const VMStateDescription aspeed_i2c_bus_vmstate = { VMSTATE_UINT32(cmd, AspeedI2CBus), VMSTATE_UINT32(buf, AspeedI2CBus), VMSTATE_UINT32(pool_ctrl, AspeedI2CBus), + VMSTATE_UINT32(dma_addr, AspeedI2CBus), + VMSTATE_UINT32(dma_len, AspeedI2CBus), VMSTATE_END_OF_LIST() } }; @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_reset(DeviceState *dev) s->busses[i].intr_status = 0; s->busses[i].cmd = 0; s->busses[i].buf = 0; + s->busses[i].dma_addr = 0; + s->busses[i].dma_len = 0; i2c_end_transfer(s->busses[i].bus); } } @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_realize(DeviceState *dev, Error **errp) memory_region_init_io(&s->pool_iomem, OBJECT(s), &aspeed_i2c_pool_ops, s, "aspeed.i2c-pool", aic->pool_size); memory_region_add_subregion(&s->iomem, aic->pool_base, &s->pool_iomem); + + if (aic->has_dma) { + if (!s->dram_mr) { + error_setg(errp, TYPE_ASPEED_I2C ": 'dram' link not set"); + return; + } + + address_space_init(&s->dram_as, s->dram_mr, "dma-dram"); + } } +static Property aspeed_i2c_properties[] = { + DEFINE_PROP_LINK("dram", AspeedI2CState, dram_mr, + TYPE_MEMORY_REGION, MemoryRegion *), + DEFINE_PROP_END_OF_LIST(), +}; + static void aspeed_i2c_class_init(ObjectClass *klass, void *data) { DeviceClass *dc = DEVICE_CLASS(klass); dc->vmsd = &aspeed_i2c_vmstate; dc->reset = aspeed_i2c_reset; + dc->props = aspeed_i2c_properties; dc->realize = aspeed_i2c_realize; dc->desc = "Aspeed I2C Controller"; } @@ -XXX,XX +XXX,XX @@ static void aspeed_2500_i2c_class_init(ObjectClass *klass, void *data) aic->pool_base = 0x200; aic->bus_pool_base = aspeed_2500_i2c_bus_pool_base; aic->check_sram = true; + aic->has_dma = true; } static const TypeInfo aspeed_2500_i2c_info = { @@ -XXX,XX +XXX,XX @@ static void aspeed_2600_i2c_class_init(ObjectClass *klass, void *data) aic->pool_size = 0x200; aic->pool_base = 0xC00; aic->bus_pool_base = aspeed_2600_i2c_bus_pool_base; + aic->has_dma = true; } static const TypeInfo aspeed_2600_i2c_info = { -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Tested-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-6-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/i2c/aspeed_i2c.c | 93 ++++++++++++++++++++++++++++++++++++++------- hw/i2c/trace-events | 9 +++++ 2 files changed, 89 insertions(+), 13 deletions(-) diff --git a/hw/i2c/aspeed_i2c.c b/hw/i2c/aspeed_i2c.c index XXXXXXX..XXXXXXX 100644 --- a/hw/i2c/aspeed_i2c.c +++ b/hw/i2c/aspeed_i2c.c @@ -XXX,XX +XXX,XX @@ #include "hw/i2c/aspeed_i2c.h" #include "hw/irq.h" #include "hw/qdev-properties.h" +#include "trace.h" /* I2C Global Register */ @@ -XXX,XX +XXX,XX @@ static inline void aspeed_i2c_bus_raise_interrupt(AspeedI2CBus *bus) { AspeedI2CClass *aic = ASPEED_I2C_GET_CLASS(bus->controller); + trace_aspeed_i2c_bus_raise_interrupt(bus->intr_status, + bus->intr_status & I2CD_INTR_TX_NAK ? "nak|" : "", + bus->intr_status & I2CD_INTR_TX_ACK ? "ack|" : "", + bus->intr_status & I2CD_INTR_RX_DONE ? "done|" : "", + bus->intr_status & I2CD_INTR_NORMAL_STOP ? "normal|" : "", + bus->intr_status & I2CD_INTR_ABNORMAL ? "abnormal" : ""); + bus->intr_status &= bus->intr_ctrl; if (bus->intr_status) { bus->controller->intr_status |= 1 << bus->id; @@ -XXX,XX +XXX,XX @@ static uint64_t aspeed_i2c_bus_read(void *opaque, hwaddr offset, { AspeedI2CBus *bus = opaque; AspeedI2CClass *aic = ASPEED_I2C_GET_CLASS(bus->controller); + uint64_t value = -1; switch (offset) { case I2CD_FUN_CTRL_REG: - return bus->ctrl; + value = bus->ctrl; + break; case I2CD_AC_TIMING_REG1: - return bus->timing[0]; + value = bus->timing[0]; + break; case I2CD_AC_TIMING_REG2: - return bus->timing[1]; + value = bus->timing[1]; + break; case I2CD_INTR_CTRL_REG: - return bus->intr_ctrl; + value = bus->intr_ctrl; + break; case I2CD_INTR_STS_REG: - return bus->intr_status; + value = bus->intr_status; + break; case I2CD_POOL_CTRL_REG: - return bus->pool_ctrl; + value = bus->pool_ctrl; + break; case I2CD_BYTE_BUF_REG: - return bus->buf; + value = bus->buf; + break; case I2CD_CMD_REG: - return bus->cmd | (i2c_bus_busy(bus->bus) << 16); + value = bus->cmd | (i2c_bus_busy(bus->bus) << 16); + break; case I2CD_DMA_ADDR: if (!aic->has_dma) { qemu_log_mask(LOG_GUEST_ERROR, "%s: No DMA support\n", __func__); - return -1; + break; } - return bus->dma_addr; + value = bus->dma_addr; + break; case I2CD_DMA_LEN: if (!aic->has_dma) { qemu_log_mask(LOG_GUEST_ERROR, "%s: No DMA support\n", __func__); - return -1; + break; } - return bus->dma_len; + value = bus->dma_len; + break; + default: qemu_log_mask(LOG_GUEST_ERROR, "%s: Bad offset 0x%" HWADDR_PRIx "\n", __func__, offset); - return -1; + value = -1; + break; } + + trace_aspeed_i2c_bus_read(bus->id, offset, size, value); + return value; } static void aspeed_i2c_set_state(AspeedI2CBus *bus, uint8_t state) @@ -XXX,XX +XXX,XX @@ static int aspeed_i2c_bus_send(AspeedI2CBus *bus, uint8_t pool_start) for (i = pool_start; i < I2CD_POOL_TX_COUNT(bus->pool_ctrl); i++) { uint8_t *pool_base = aic->bus_pool_base(bus); + trace_aspeed_i2c_bus_send("BUF", i + 1, + I2CD_POOL_TX_COUNT(bus->pool_ctrl), + pool_base[i]); ret = i2c_send(bus->bus, pool_base[i]); if (ret) { break; @@ -XXX,XX +XXX,XX @@ static int aspeed_i2c_bus_send(AspeedI2CBus *bus, uint8_t pool_start) while (bus->dma_len) { uint8_t data; aspeed_i2c_dma_read(bus, &data); + trace_aspeed_i2c_bus_send("DMA", bus->dma_len, bus->dma_len, data); ret = i2c_send(bus->bus, data); if (ret) { break; @@ -XXX,XX +XXX,XX @@ static int aspeed_i2c_bus_send(AspeedI2CBus *bus, uint8_t pool_start) } bus->cmd &= ~I2CD_TX_DMA_ENABLE; } else { + trace_aspeed_i2c_bus_send("BYTE", pool_start, 1, bus->buf); ret = i2c_send(bus->bus, bus->buf); } @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_bus_recv(AspeedI2CBus *bus) for (i = 0; i < I2CD_POOL_RX_SIZE(bus->pool_ctrl); i++) { pool_base[i] = i2c_recv(bus->bus); + trace_aspeed_i2c_bus_recv("BUF", i + 1, + I2CD_POOL_RX_SIZE(bus->pool_ctrl), + pool_base[i]); } /* Update RX count */ @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_bus_recv(AspeedI2CBus *bus) MemTxResult result; data = i2c_recv(bus->bus); + trace_aspeed_i2c_bus_recv("DMA", bus->dma_len, bus->dma_len, data); result = address_space_write(&s->dram_as, bus->dma_addr, MEMTXATTRS_UNSPECIFIED, &data, 1); if (result != MEMTX_OK) { @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_bus_recv(AspeedI2CBus *bus) bus->cmd &= ~I2CD_RX_DMA_ENABLE; } else { data = i2c_recv(bus->bus); + trace_aspeed_i2c_bus_recv("BYTE", 1, 1, bus->buf); bus->buf = (data & I2CD_BYTE_BUF_RX_MASK) << I2CD_BYTE_BUF_RX_SHIFT; } } @@ -XXX,XX +XXX,XX @@ static bool aspeed_i2c_check_sram(AspeedI2CBus *bus) return true; } +static void aspeed_i2c_bus_cmd_dump(AspeedI2CBus *bus) +{ + g_autofree char *cmd_flags; + uint32_t count; + + if (bus->cmd & (I2CD_RX_BUFF_ENABLE | I2CD_RX_BUFF_ENABLE)) { + count = I2CD_POOL_TX_COUNT(bus->pool_ctrl); + } else if (bus->cmd & (I2CD_RX_DMA_ENABLE | I2CD_RX_DMA_ENABLE)) { + count = bus->dma_len; + } else { /* BYTE mode */ + count = 1; + } + + cmd_flags = g_strdup_printf("%s%s%s%s%s%s%s%s%s", + bus->cmd & I2CD_M_START_CMD ? "start|" : "", + bus->cmd & I2CD_RX_DMA_ENABLE ? "rxdma|" : "", + bus->cmd & I2CD_TX_DMA_ENABLE ? "txdma|" : "", + bus->cmd & I2CD_RX_BUFF_ENABLE ? "rxbuf|" : "", + bus->cmd & I2CD_TX_BUFF_ENABLE ? "txbuf|" : "", + bus->cmd & I2CD_M_TX_CMD ? "tx|" : "", + bus->cmd & I2CD_M_RX_CMD ? "rx|" : "", + bus->cmd & I2CD_M_S_RX_CMD_LAST ? "last|" : "", + bus->cmd & I2CD_M_STOP_CMD ? "stop" : ""); + + trace_aspeed_i2c_bus_cmd(bus->cmd, cmd_flags, count, bus->intr_status); +} + /* * The state machine needs some refinement. It is only used to track * invalid STOP commands for the moment. @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_bus_handle_cmd(AspeedI2CBus *bus, uint64_t value) return; } + if (trace_event_get_state_backends(TRACE_ASPEED_I2C_BUS_CMD)) { + aspeed_i2c_bus_cmd_dump(bus); + } + if (bus->cmd & I2CD_M_START_CMD) { uint8_t state = aspeed_i2c_get_state(bus) & I2CD_MACTIVE ? I2CD_MSTARTR : I2CD_MSTART; @@ -XXX,XX +XXX,XX @@ static void aspeed_i2c_bus_write(void *opaque, hwaddr offset, AspeedI2CClass *aic = ASPEED_I2C_GET_CLASS(bus->controller); bool handle_rx; + trace_aspeed_i2c_bus_write(bus->id, offset, size, value); + switch (offset) { case I2CD_FUN_CTRL_REG: if (value & I2CD_SLAVE_EN) { diff --git a/hw/i2c/trace-events b/hw/i2c/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/i2c/trace-events +++ b/hw/i2c/trace-events @@ -XXX,XX +XXX,XX @@ i2c_event(const char *event, uint8_t address) "%s(addr:0x%02x)" i2c_send(uint8_t address, uint8_t data) "send(addr:0x%02x) data:0x%02x" i2c_recv(uint8_t address, uint8_t data) "recv(addr:0x%02x) data:0x%02x" + +# aspeed_i2c.c + +aspeed_i2c_bus_cmd(uint32_t cmd, const char *cmd_flags, uint32_t count, uint32_t intr_status) "handling cmd=0x%x %s count=%d intr=0x%x" +aspeed_i2c_bus_raise_interrupt(uint32_t intr_status, const char *str1, const char *str2, const char *str3, const char *str4, const char *str5) "handled intr=0x%x %s%s%s%s%s" +aspeed_i2c_bus_read(uint32_t busid, uint64_t offset, unsigned size, uint64_t value) "bus[%d]: To 0x%" PRIx64 " of size %u: 0x%" PRIx64 +aspeed_i2c_bus_write(uint32_t busid, uint64_t offset, unsigned size, uint64_t value) "bus[%d]: To 0x%" PRIx64 " of size %u: 0x%" PRIx64 +aspeed_i2c_bus_send(const char *mode, int i, int count, uint8_t byte) "%s send %d/%d 0x%02x" +aspeed_i2c_bus_recv(const char *mode, int i, int count, uint8_t byte) "%s recv %d/%d 0x%02x" -- 2.20.1 From: Joel Stanley <joel@jms.id.au> Most boards have this much. Reviewed-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-7-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/misc/aspeed_sdmc.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/hw/misc/aspeed_sdmc.c b/hw/misc/aspeed_sdmc.c index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/aspeed_sdmc.c +++ b/hw/misc/aspeed_sdmc.c @@ -XXX,XX +XXX,XX @@ static int ast2600_rambits(AspeedSDMCState *s) } /* use a common default */ - warn_report("Invalid RAM size 0x%" PRIx64 ". Using default 512M", + warn_report("Invalid RAM size 0x%" PRIx64 ". Using default 1024M", s->ram_size); - s->ram_size = 512 << 20; - return ASPEED_SDMC_AST2600_512MB; + s->ram_size = 1024 << 20; + return ASPEED_SDMC_AST2600_1024MB; } static void aspeed_sdmc_reset(DeviceState *dev) -- 2.20.1 From: Joel Stanley <joel@jms.id.au> This models the clock write one to clear registers, and fixes up some incorrect behavior in all of the write to clear registers. There was also a typo in one of the register definitions. Reviewed-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-8-clg@kaod.org [clg: checkpatch.pl fixes ] Signed-off-by: Cédric Le Goater <clg@kaod.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/misc/aspeed_scu.c | 19 ++++++++++++++----- 1 file changed, 14 insertions(+), 5 deletions(-) diff --git a/hw/misc/aspeed_scu.c b/hw/misc/aspeed_scu.c index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/aspeed_scu.c +++ b/hw/misc/aspeed_scu.c @@ -XXX,XX +XXX,XX @@ #define AST2600_CLK_STOP_CTRL TO_REG(0x80) #define AST2600_CLK_STOP_CTRL_CLR TO_REG(0x84) #define AST2600_CLK_STOP_CTRL2 TO_REG(0x90) -#define AST2600_CLK_STOP_CTR2L_CLR TO_REG(0x94) +#define AST2600_CLK_STOP_CTRL2_CLR TO_REG(0x94) #define AST2600_SDRAM_HANDSHAKE TO_REG(0x100) #define AST2600_HPLL_PARAM TO_REG(0x200) #define AST2600_HPLL_EXT TO_REG(0x204) @@ -XXX,XX +XXX,XX @@ static uint64_t aspeed_ast2600_scu_read(void *opaque, hwaddr offset, return s->regs[reg]; } -static void aspeed_ast2600_scu_write(void *opaque, hwaddr offset, uint64_t data, - unsigned size) +static void aspeed_ast2600_scu_write(void *opaque, hwaddr offset, + uint64_t data64, unsigned size) { AspeedSCUState *s = ASPEED_SCU(opaque); int reg = TO_REG(offset); + /* Truncate here so bitwise operations below behave as expected */ + uint32_t data = data64; if (reg >= ASPEED_AST2600_SCU_NR_REGS) { qemu_log_mask(LOG_GUEST_ERROR, @@ -XXX,XX +XXX,XX @@ static void aspeed_ast2600_scu_write(void *opaque, hwaddr offset, uint64_t data, /* fall through */ case AST2600_SYS_RST_CTRL: case AST2600_SYS_RST_CTRL2: + case AST2600_CLK_STOP_CTRL: + case AST2600_CLK_STOP_CTRL2: /* W1S (Write 1 to set) registers */ s->regs[reg] |= data; return; case AST2600_SYS_RST_CTRL_CLR: case AST2600_SYS_RST_CTRL2_CLR: + case AST2600_CLK_STOP_CTRL_CLR: + case AST2600_CLK_STOP_CTRL2_CLR: case AST2600_HW_STRAP1_CLR: case AST2600_HW_STRAP2_CLR: - /* W1C (Write 1 to clear) registers */ - s->regs[reg] &= ~data; + /* + * W1C (Write 1 to clear) registers are offset by one address from + * the data register + */ + s->regs[reg - 1] &= ~data; return; case AST2600_RNG_DATA: -- 2.20.1 From: Joel Stanley <joel@jms.id.au> Users benefit from knowing which watchdog timer has expired. The address of the watchdog's registers unambiguously indicates which has expired, so log that. Reviewed-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-9-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/watchdog/wdt_aspeed.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/hw/watchdog/wdt_aspeed.c b/hw/watchdog/wdt_aspeed.c index XXXXXXX..XXXXXXX 100644 --- a/hw/watchdog/wdt_aspeed.c +++ b/hw/watchdog/wdt_aspeed.c @@ -XXX,XX +XXX,XX @@ static void aspeed_wdt_timer_expired(void *dev) return; } - qemu_log_mask(CPU_LOG_RESET, "Watchdog timer expired.\n"); + qemu_log_mask(CPU_LOG_RESET, "Watchdog timer %" HWADDR_PRIx " expired.\n", + s->iomem.addr); watchdog_perform_action(); timer_del(s->timer); } -- 2.20.1 From: Joel Stanley <joel@jms.id.au> The AST2600 control register sneakily changed the meaning of bit 4 without anyone noticing. It no longer controls the 1MHz vs APB clock select, and instead always runs at 1MHz. The AST2500 was always 1MHz too, but it retained bit 4, making it read only. We can model both using the same fixed 1MHz calculation. Fixes: 6b2b2a703cad ("hw: wdt_aspeed: Add AST2600 support") Reviewed-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-10-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/watchdog/wdt_aspeed.h | 1 + hw/watchdog/wdt_aspeed.c | 21 +++++++++++++++++---- 2 files changed, 18 insertions(+), 4 deletions(-) diff --git a/include/hw/watchdog/wdt_aspeed.h b/include/hw/watchdog/wdt_aspeed.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/watchdog/wdt_aspeed.h +++ b/include/hw/watchdog/wdt_aspeed.h @@ -XXX,XX +XXX,XX @@ typedef struct AspeedWDTClass { uint32_t ext_pulse_width_mask; uint32_t reset_ctrl_reg; void (*reset_pulse)(AspeedWDTState *s, uint32_t property); + void (*wdt_reload)(AspeedWDTState *s); } AspeedWDTClass; #endif /* WDT_ASPEED_H */ diff --git a/hw/watchdog/wdt_aspeed.c b/hw/watchdog/wdt_aspeed.c index XXXXXXX..XXXXXXX 100644 --- a/hw/watchdog/wdt_aspeed.c +++ b/hw/watchdog/wdt_aspeed.c @@ -XXX,XX +XXX,XX @@ static uint64_t aspeed_wdt_read(void *opaque, hwaddr offset, unsigned size) } -static void aspeed_wdt_reload(AspeedWDTState *s, bool pclk) +static void aspeed_wdt_reload(AspeedWDTState *s) { uint64_t reload; - if (pclk) { + if (!(s->regs[WDT_CTRL] & WDT_CTRL_1MHZ_CLK)) { reload = muldiv64(s->regs[WDT_RELOAD_VALUE], NANOSECONDS_PER_SECOND, s->pclk_freq); } else { @@ -XXX,XX +XXX,XX @@ static void aspeed_wdt_reload(AspeedWDTState *s, bool pclk) } } +static void aspeed_wdt_reload_1mhz(AspeedWDTState *s) +{ + uint64_t reload = s->regs[WDT_RELOAD_VALUE] * 1000ULL; + + if (aspeed_wdt_is_enabled(s)) { + timer_mod(s->timer, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) + reload); + } +} + + static void aspeed_wdt_write(void *opaque, hwaddr offset, uint64_t data, unsigned size) { @@ -XXX,XX +XXX,XX @@ static void aspeed_wdt_write(void *opaque, hwaddr offset, uint64_t data, case WDT_RESTART: if ((data & 0xFFFF) == WDT_RESTART_MAGIC) { s->regs[WDT_STATUS] = s->regs[WDT_RELOAD_VALUE]; - aspeed_wdt_reload(s, !(s->regs[WDT_CTRL] & WDT_CTRL_1MHZ_CLK)); + awc->wdt_reload(s); } break; case WDT_CTRL: if (enable && !aspeed_wdt_is_enabled(s)) { s->regs[WDT_CTRL] = data; - aspeed_wdt_reload(s, !(data & WDT_CTRL_1MHZ_CLK)); + awc->wdt_reload(s); } else if (!enable && aspeed_wdt_is_enabled(s)) { s->regs[WDT_CTRL] = data; timer_del(s->timer); @@ -XXX,XX +XXX,XX @@ static void aspeed_2400_wdt_class_init(ObjectClass *klass, void *data) awc->offset = 0x20; awc->ext_pulse_width_mask = 0xff; awc->reset_ctrl_reg = SCU_RESET_CONTROL1; + awc->wdt_reload = aspeed_wdt_reload; } static const TypeInfo aspeed_2400_wdt_info = { @@ -XXX,XX +XXX,XX @@ static void aspeed_2500_wdt_class_init(ObjectClass *klass, void *data) awc->ext_pulse_width_mask = 0xfffff; awc->reset_ctrl_reg = SCU_RESET_CONTROL1; awc->reset_pulse = aspeed_2500_wdt_reset_pulse; + awc->wdt_reload = aspeed_wdt_reload_1mhz; } static const TypeInfo aspeed_2500_wdt_info = { @@ -XXX,XX +XXX,XX @@ static void aspeed_2600_wdt_class_init(ObjectClass *klass, void *data) awc->ext_pulse_width_mask = 0xfffff; /* TODO */ awc->reset_ctrl_reg = AST2600_SCU_RESET_CONTROL1; awc->reset_pulse = aspeed_2500_wdt_reset_pulse; + awc->wdt_reload = aspeed_wdt_reload_1mhz; } static const TypeInfo aspeed_2600_wdt_info = { -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> The current model only restores the Segment Register values but leaves the previous CS mapping behind. Introduce a helper setting the register value and mapping the region at the requested address. Use this helper when a Segment register is set and at reset. Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-11-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/ssi/aspeed_smc.c | 32 +++++++++++++++++++++----------- 1 file changed, 21 insertions(+), 11 deletions(-) diff --git a/hw/ssi/aspeed_smc.c b/hw/ssi/aspeed_smc.c index XXXXXXX..XXXXXXX 100644 --- a/hw/ssi/aspeed_smc.c +++ b/hw/ssi/aspeed_smc.c @@ -XXX,XX +XXX,XX @@ static bool aspeed_smc_flash_overlap(const AspeedSMCState *s, return false; } +static void aspeed_smc_flash_set_segment_region(AspeedSMCState *s, int cs, + uint64_t regval) +{ + AspeedSMCFlash *fl = &s->flashes[cs]; + AspeedSegments seg; + + s->ctrl->reg_to_segment(s, regval, &seg); + + memory_region_transaction_begin(); + memory_region_set_size(&fl->mmio, seg.size); + memory_region_set_address(&fl->mmio, seg.addr - s->ctrl->flash_window_base); + memory_region_set_enabled(&fl->mmio, true); + memory_region_transaction_commit(); + + s->regs[R_SEG_ADDR0 + cs] = regval; +} + static void aspeed_smc_flash_set_segment(AspeedSMCState *s, int cs, uint64_t new) { - AspeedSMCFlash *fl = &s->flashes[cs]; AspeedSegments seg; s->ctrl->reg_to_segment(s, new, &seg); @@ -XXX,XX +XXX,XX @@ static void aspeed_smc_flash_set_segment(AspeedSMCState *s, int cs, aspeed_smc_flash_overlap(s, &seg, cs); /* All should be fine now to move the region */ - memory_region_transaction_begin(); - memory_region_set_size(&fl->mmio, seg.size); - memory_region_set_address(&fl->mmio, seg.addr - s->ctrl->flash_window_base); - memory_region_set_enabled(&fl->mmio, true); - memory_region_transaction_commit(); - - s->regs[R_SEG_ADDR0 + cs] = new; + aspeed_smc_flash_set_segment_region(s, cs, new); } static uint64_t aspeed_smc_flash_default_read(void *opaque, hwaddr addr, @@ -XXX,XX +XXX,XX @@ static void aspeed_smc_reset(DeviceState *d) qemu_set_irq(s->cs_lines[i], true); } - /* setup default segment register values for all */ + /* setup the default segment register values and regions for all */ for (i = 0; i < s->ctrl->max_slaves; ++i) { - s->regs[R_SEG_ADDR0 + i] = - s->ctrl->segment_to_reg(s, &s->ctrl->segments[i]); + aspeed_smc_flash_set_segment_region(s, i, + s->ctrl->segment_to_reg(s, &s->ctrl->segments[i])); } /* HW strapping flash type for the AST2600 controllers */ -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> The segments can be disabled on the AST2600 (zero register value). CS0 is open by default but not the other CS. This is closing the access to the flash device in user mode and forbids scanning. In the model, check the segment size and disable the associated region when the value is zero. Fixes: bcaa8ddd081c ("aspeed/smc: Add AST2600 support") Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-12-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/ssi/aspeed_smc.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/hw/ssi/aspeed_smc.c b/hw/ssi/aspeed_smc.c index XXXXXXX..XXXXXXX 100644 --- a/hw/ssi/aspeed_smc.c +++ b/hw/ssi/aspeed_smc.c @@ -XXX,XX +XXX,XX @@ static void aspeed_2600_smc_reg_to_segment(const AspeedSMCState *s, uint32_t start_offset = (reg << 16) & AST2600_SEG_ADDR_MASK; uint32_t end_offset = reg & AST2600_SEG_ADDR_MASK; - seg->addr = s->ctrl->flash_window_base + start_offset; - seg->size = end_offset + MiB - start_offset; + if (reg) { + seg->addr = s->ctrl->flash_window_base + start_offset; + seg->size = end_offset + MiB - start_offset; + } else { + seg->addr = s->ctrl->flash_window_base; + seg->size = 0; + } } static bool aspeed_smc_flash_overlap(const AspeedSMCState *s, @@ -XXX,XX +XXX,XX @@ static void aspeed_smc_flash_set_segment_region(AspeedSMCState *s, int cs, memory_region_transaction_begin(); memory_region_set_size(&fl->mmio, seg.size); memory_region_set_address(&fl->mmio, seg.addr - s->ctrl->flash_window_base); - memory_region_set_enabled(&fl->mmio, true); + memory_region_set_enabled(&fl->mmio, !!seg.size); memory_region_transaction_commit(); s->regs[R_SEG_ADDR0 + cs] = regval; @@ -XXX,XX +XXX,XX @@ static void aspeed_smc_flash_set_segment(AspeedSMCState *s, int cs, } /* Keep the segment in the overall flash window */ - if (seg.addr + seg.size <= s->ctrl->flash_window_base || - seg.addr > s->ctrl->flash_window_base + s->ctrl->flash_window_size) { + if (seg.size && + (seg.addr + seg.size <= s->ctrl->flash_window_base || + seg.addr > s->ctrl->flash_window_base + s->ctrl->flash_window_size)) { qemu_log_mask(LOG_GUEST_ERROR, "%s: new segment for CS%d is invalid : " "[ 0x%"HWADDR_PRIx" - 0x%"HWADDR_PRIx" ]\n", s->ctrl->name, cs, seg.addr, seg.addr + seg.size); -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> Each CS has its own Read Timing Compensation Register on newer SoCs. Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-13-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/ssi/aspeed_smc.h | 1 + hw/ssi/aspeed_smc.c | 17 ++++++++++++++--- 2 files changed, 15 insertions(+), 3 deletions(-) diff --git a/include/hw/ssi/aspeed_smc.h b/include/hw/ssi/aspeed_smc.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/ssi/aspeed_smc.h +++ b/include/hw/ssi/aspeed_smc.h @@ -XXX,XX +XXX,XX @@ typedef struct AspeedSMCController { uint8_t r_ce_ctrl; uint8_t r_ctrl0; uint8_t r_timings; + uint8_t nregs_timings; uint8_t conf_enable_w0; uint8_t max_slaves; const AspeedSegments *segments; diff --git a/hw/ssi/aspeed_smc.c b/hw/ssi/aspeed_smc.c index XXXXXXX..XXXXXXX 100644 --- a/hw/ssi/aspeed_smc.c +++ b/hw/ssi/aspeed_smc.c @@ -XXX,XX +XXX,XX @@ /* Checksum Calculation Result */ #define R_DMA_CHECKSUM (0x90 / 4) -/* Misc Control Register #2 */ +/* Read Timing Compensation Register */ #define R_TIMINGS (0x94 / 4) /* SPI controller registers and bits (AST2400) */ @@ -XXX,XX +XXX,XX @@ static const AspeedSMCController controllers[] = { .r_ce_ctrl = R_CE_CTRL, .r_ctrl0 = R_CTRL0, .r_timings = R_TIMINGS, + .nregs_timings = 1, .conf_enable_w0 = CONF_ENABLE_W0, .max_slaves = 5, .segments = aspeed_segments_legacy, @@ -XXX,XX +XXX,XX @@ static const AspeedSMCController controllers[] = { .r_ce_ctrl = R_CE_CTRL, .r_ctrl0 = R_CTRL0, .r_timings = R_TIMINGS, + .nregs_timings = 1, .conf_enable_w0 = CONF_ENABLE_W0, .max_slaves = 5, .segments = aspeed_segments_fmc, @@ -XXX,XX +XXX,XX @@ static const AspeedSMCController controllers[] = { .r_ce_ctrl = 0xff, .r_ctrl0 = R_SPI_CTRL0, .r_timings = R_SPI_TIMINGS, + .nregs_timings = 1, .conf_enable_w0 = SPI_CONF_ENABLE_W0, .max_slaves = 1, .segments = aspeed_segments_spi, @@ -XXX,XX +XXX,XX @@ static const AspeedSMCController controllers[] = { .r_ce_ctrl = R_CE_CTRL, .r_ctrl0 = R_CTRL0, .r_timings = R_TIMINGS, + .nregs_timings = 1, .conf_enable_w0 = CONF_ENABLE_W0, .max_slaves = 3, .segments = aspeed_segments_ast2500_fmc, @@ -XXX,XX +XXX,XX @@ static const AspeedSMCController controllers[] = { .r_ce_ctrl = R_CE_CTRL, .r_ctrl0 = R_CTRL0, .r_timings = R_TIMINGS, + .nregs_timings = 1, .conf_enable_w0 = CONF_ENABLE_W0, .max_slaves = 2, .segments = aspeed_segments_ast2500_spi1, @@ -XXX,XX +XXX,XX @@ static const AspeedSMCController controllers[] = { .r_ce_ctrl = R_CE_CTRL, .r_ctrl0 = R_CTRL0, .r_timings = R_TIMINGS, + .nregs_timings = 1, .conf_enable_w0 = CONF_ENABLE_W0, .max_slaves = 2, .segments = aspeed_segments_ast2500_spi2, @@ -XXX,XX +XXX,XX @@ static const AspeedSMCController controllers[] = { .r_ce_ctrl = R_CE_CTRL, .r_ctrl0 = R_CTRL0, .r_timings = R_TIMINGS, + .nregs_timings = 1, .conf_enable_w0 = CONF_ENABLE_W0, .max_slaves = 3, .segments = aspeed_segments_ast2600_fmc, @@ -XXX,XX +XXX,XX @@ static const AspeedSMCController controllers[] = { .r_ce_ctrl = R_CE_CTRL, .r_ctrl0 = R_CTRL0, .r_timings = R_TIMINGS, + .nregs_timings = 2, .conf_enable_w0 = CONF_ENABLE_W0, .max_slaves = 2, .segments = aspeed_segments_ast2600_spi1, @@ -XXX,XX +XXX,XX @@ static const AspeedSMCController controllers[] = { .r_ce_ctrl = R_CE_CTRL, .r_ctrl0 = R_CTRL0, .r_timings = R_TIMINGS, + .nregs_timings = 3, .conf_enable_w0 = CONF_ENABLE_W0, .max_slaves = 3, .segments = aspeed_segments_ast2600_spi2, @@ -XXX,XX +XXX,XX @@ static uint64_t aspeed_smc_read(void *opaque, hwaddr addr, unsigned int size) addr >>= 2; if (addr == s->r_conf || - addr == s->r_timings || + (addr >= s->r_timings && + addr < s->r_timings + s->ctrl->nregs_timings) || addr == s->r_ce_ctrl || addr == R_INTR_CTRL || addr == R_DUMMY_DATA || @@ -XXX,XX +XXX,XX @@ static void aspeed_smc_write(void *opaque, hwaddr addr, uint64_t data, addr >>= 2; if (addr == s->r_conf || - addr == s->r_timings || + (addr >= s->r_timings && + addr < s->r_timings + s->ctrl->nregs_timings) || addr == s->r_ce_ctrl) { s->regs[addr] = value; } else if (addr >= s->r_ctrl0 && addr < s->r_ctrl0 + s->num_cs) { -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> AspeedBoardConfig is a redundant way to define class attributes and it complexifies the machine definition and initialization. Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-14-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/aspeed.h | 24 ++-- hw/arm/aspeed.c | 243 ++++++++++++++++++++++------------------ 2 files changed, 143 insertions(+), 124 deletions(-) diff --git a/include/hw/arm/aspeed.h b/include/hw/arm/aspeed.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/aspeed.h +++ b/include/hw/arm/aspeed.h @@ -XXX,XX +XXX,XX @@ typedef struct AspeedBoardState AspeedBoardState; -typedef struct AspeedBoardConfig { - const char *name; - const char *desc; - const char *soc_name; - uint32_t hw_strap1; - uint32_t hw_strap2; - const char *fmc_model; - const char *spi_model; - uint32_t num_cs; - void (*i2c_init)(AspeedBoardState *bmc); - uint32_t ram; -} AspeedBoardConfig; - #define TYPE_ASPEED_MACHINE MACHINE_TYPE_NAME("aspeed") #define ASPEED_MACHINE(obj) \ OBJECT_CHECK(AspeedMachine, (obj), TYPE_ASPEED_MACHINE) @@ -XXX,XX +XXX,XX @@ typedef struct AspeedMachine { typedef struct AspeedMachineClass { MachineClass parent_obj; - const AspeedBoardConfig *board; + + const char *name; + const char *desc; + const char *soc_name; + uint32_t hw_strap1; + uint32_t hw_strap2; + const char *fmc_model; + const char *spi_model; + uint32_t num_cs; + void (*i2c_init)(AspeedBoardState *bmc); } AspeedMachineClass; diff --git a/hw/arm/aspeed.c b/hw/arm/aspeed.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/aspeed.c +++ b/hw/arm/aspeed.c @@ -XXX,XX +XXX,XX @@ static void aspeed_board_init_flashes(AspeedSMCState *s, const char *flashtype, } } -static void aspeed_board_init(MachineState *machine, - const AspeedBoardConfig *cfg) +static void aspeed_machine_init(MachineState *machine) { AspeedBoardState *bmc; + AspeedMachineClass *amc = ASPEED_MACHINE_GET_CLASS(machine); AspeedSoCClass *sc; DriveInfo *drive0 = drive_get(IF_MTD, 0, 0); ram_addr_t max_ram_size; @@ -XXX,XX +XXX,XX @@ static void aspeed_board_init(MachineState *machine, UINT32_MAX); object_initialize_child(OBJECT(machine), "soc", &bmc->soc, - (sizeof(bmc->soc)), cfg->soc_name, &error_abort, + (sizeof(bmc->soc)), amc->soc_name, &error_abort, NULL); sc = ASPEED_SOC_GET_CLASS(&bmc->soc); object_property_set_uint(OBJECT(&bmc->soc), ram_size, "ram-size", &error_abort); - object_property_set_int(OBJECT(&bmc->soc), cfg->hw_strap1, "hw-strap1", + object_property_set_int(OBJECT(&bmc->soc), amc->hw_strap1, "hw-strap1", &error_abort); - object_property_set_int(OBJECT(&bmc->soc), cfg->hw_strap2, "hw-strap2", + object_property_set_int(OBJECT(&bmc->soc), amc->hw_strap2, "hw-strap2", &error_abort); - object_property_set_int(OBJECT(&bmc->soc), cfg->num_cs, "num-cs", + object_property_set_int(OBJECT(&bmc->soc), amc->num_cs, "num-cs", &error_abort); object_property_set_int(OBJECT(&bmc->soc), machine->smp.cpus, "num-cpus", &error_abort); @@ -XXX,XX +XXX,XX @@ static void aspeed_board_init(MachineState *machine, "max_ram", max_ram_size - ram_size); memory_region_add_subregion(&bmc->ram_container, ram_size, &bmc->max_ram); - aspeed_board_init_flashes(&bmc->soc.fmc, cfg->fmc_model, &error_abort); - aspeed_board_init_flashes(&bmc->soc.spi[0], cfg->spi_model, &error_abort); + aspeed_board_init_flashes(&bmc->soc.fmc, amc->fmc_model, &error_abort); + aspeed_board_init_flashes(&bmc->soc.spi[0], amc->spi_model, &error_abort); /* Install first FMC flash content as a boot rom. */ if (drive0) { @@ -XXX,XX +XXX,XX @@ static void aspeed_board_init(MachineState *machine, aspeed_board_binfo.loader_start = sc->memmap[ASPEED_SDRAM]; aspeed_board_binfo.nb_cpus = bmc->soc.num_cpus; - if (cfg->i2c_init) { - cfg->i2c_init(bmc); + if (amc->i2c_init) { + amc->i2c_init(bmc); } for (i = 0; i < ARRAY_SIZE(bmc->soc.sdhci.slots); i++) { @@ -XXX,XX +XXX,XX @@ static void witherspoon_bmc_i2c_init(AspeedBoardState *bmc) 0x60); } -static void aspeed_machine_init(MachineState *machine) -{ - AspeedMachineClass *amc = ASPEED_MACHINE_GET_CLASS(machine); - - aspeed_board_init(machine, amc->board); -} - static void aspeed_machine_class_init(ObjectClass *oc, void *data) { MachineClass *mc = MACHINE_CLASS(oc); - AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc); - const AspeedBoardConfig *board = data; - mc->desc = board->desc; mc->init = aspeed_machine_init; mc->max_cpus = ASPEED_CPUS_NUM; mc->no_floppy = 1; mc->no_cdrom = 1; mc->no_parallel = 1; - if (board->ram) { - mc->default_ram_size = board->ram; - } - amc->board = board; } -static const TypeInfo aspeed_machine_type = { - .name = TYPE_ASPEED_MACHINE, - .parent = TYPE_MACHINE, - .instance_size = sizeof(AspeedMachine), - .class_size = sizeof(AspeedMachineClass), - .abstract = true, -}; - -static const AspeedBoardConfig aspeed_boards[] = { - { - .name = MACHINE_TYPE_NAME("palmetto-bmc"), - .desc = "OpenPOWER Palmetto BMC (ARM926EJ-S)", - .soc_name = "ast2400-a1", - .hw_strap1 = PALMETTO_BMC_HW_STRAP1, - .fmc_model = "n25q256a", - .spi_model = "mx25l25635e", - .num_cs = 1, - .i2c_init = palmetto_bmc_i2c_init, - .ram = 256 * MiB, - }, { - .name = MACHINE_TYPE_NAME("ast2500-evb"), - .desc = "Aspeed AST2500 EVB (ARM1176)", - .soc_name = "ast2500-a1", - .hw_strap1 = AST2500_EVB_HW_STRAP1, - .fmc_model = "w25q256", - .spi_model = "mx25l25635e", - .num_cs = 1, - .i2c_init = ast2500_evb_i2c_init, - .ram = 512 * MiB, - }, { - .name = MACHINE_TYPE_NAME("romulus-bmc"), - .desc = "OpenPOWER Romulus BMC (ARM1176)", - .soc_name = "ast2500-a1", - .hw_strap1 = ROMULUS_BMC_HW_STRAP1, - .fmc_model = "n25q256a", - .spi_model = "mx66l1g45g", - .num_cs = 2, - .i2c_init = romulus_bmc_i2c_init, - .ram = 512 * MiB, - }, { - .name = MACHINE_TYPE_NAME("swift-bmc"), - .desc = "OpenPOWER Swift BMC (ARM1176)", - .soc_name = "ast2500-a1", - .hw_strap1 = SWIFT_BMC_HW_STRAP1, - .fmc_model = "mx66l1g45g", - .spi_model = "mx66l1g45g", - .num_cs = 2, - .i2c_init = swift_bmc_i2c_init, - .ram = 512 * MiB, - }, { - .name = MACHINE_TYPE_NAME("witherspoon-bmc"), - .desc = "OpenPOWER Witherspoon BMC (ARM1176)", - .soc_name = "ast2500-a1", - .hw_strap1 = WITHERSPOON_BMC_HW_STRAP1, - .fmc_model = "mx25l25635e", - .spi_model = "mx66l1g45g", - .num_cs = 2, - .i2c_init = witherspoon_bmc_i2c_init, - .ram = 512 * MiB, - }, { - .name = MACHINE_TYPE_NAME("ast2600-evb"), - .desc = "Aspeed AST2600 EVB (Cortex A7)", - .soc_name = "ast2600-a0", - .hw_strap1 = AST2600_EVB_HW_STRAP1, - .hw_strap2 = AST2600_EVB_HW_STRAP2, - .fmc_model = "w25q512jv", - .spi_model = "mx66u51235f", - .num_cs = 1, - .i2c_init = ast2600_evb_i2c_init, - .ram = 1 * GiB, - }, -}; - -static void aspeed_machine_types(void) +static void aspeed_machine_palmetto_class_init(ObjectClass *oc, void *data) { - int i; + MachineClass *mc = MACHINE_CLASS(oc); + AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc); - type_register_static(&aspeed_machine_type); - for (i = 0; i < ARRAY_SIZE(aspeed_boards); ++i) { - TypeInfo ti = { - .name = aspeed_boards[i].name, - .parent = TYPE_ASPEED_MACHINE, - .class_init = aspeed_machine_class_init, - .class_data = (void *)&aspeed_boards[i], - }; - type_register(&ti); + mc->desc = "OpenPOWER Palmetto BMC (ARM926EJ-S)"; + amc->soc_name = "ast2400-a1"; + amc->hw_strap1 = PALMETTO_BMC_HW_STRAP1; + amc->fmc_model = "n25q256a"; + amc->spi_model = "mx25l25635e"; + amc->num_cs = 1; + amc->i2c_init = palmetto_bmc_i2c_init; + mc->default_ram_size = 256 * MiB; +}; + +static void aspeed_machine_ast2500_evb_class_init(ObjectClass *oc, void *data) +{ + MachineClass *mc = MACHINE_CLASS(oc); + AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc); + + mc->desc = "Aspeed AST2500 EVB (ARM1176)"; + amc->soc_name = "ast2500-a1"; + amc->hw_strap1 = AST2500_EVB_HW_STRAP1; + amc->fmc_model = "w25q256"; + amc->spi_model = "mx25l25635e"; + amc->num_cs = 1; + amc->i2c_init = ast2500_evb_i2c_init; + mc->default_ram_size = 512 * MiB; +}; + +static void aspeed_machine_romulus_class_init(ObjectClass *oc, void *data) +{ + MachineClass *mc = MACHINE_CLASS(oc); + AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc); + + mc->desc = "OpenPOWER Romulus BMC (ARM1176)"; + amc->soc_name = "ast2500-a1"; + amc->hw_strap1 = ROMULUS_BMC_HW_STRAP1; + amc->fmc_model = "n25q256a"; + amc->spi_model = "mx66l1g45g"; + amc->num_cs = 2; + amc->i2c_init = romulus_bmc_i2c_init; + mc->default_ram_size = 512 * MiB; +}; + +static void aspeed_machine_swift_class_init(ObjectClass *oc, void *data) +{ + MachineClass *mc = MACHINE_CLASS(oc); + AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc); + + mc->desc = "OpenPOWER Swift BMC (ARM1176)"; + amc->soc_name = "ast2500-a1"; + amc->hw_strap1 = SWIFT_BMC_HW_STRAP1; + amc->fmc_model = "mx66l1g45g"; + amc->spi_model = "mx66l1g45g"; + amc->num_cs = 2; + amc->i2c_init = swift_bmc_i2c_init; + mc->default_ram_size = 512 * MiB; +}; + +static void aspeed_machine_witherspoon_class_init(ObjectClass *oc, void *data) +{ + MachineClass *mc = MACHINE_CLASS(oc); + AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc); + + mc->desc = "OpenPOWER Witherspoon BMC (ARM1176)"; + amc->soc_name = "ast2500-a1"; + amc->hw_strap1 = WITHERSPOON_BMC_HW_STRAP1; + amc->fmc_model = "mx25l25635e"; + amc->spi_model = "mx66l1g45g"; + amc->num_cs = 2; + amc->i2c_init = witherspoon_bmc_i2c_init; + mc->default_ram_size = 512 * MiB; +}; + +static void aspeed_machine_ast2600_evb_class_init(ObjectClass *oc, void *data) +{ + MachineClass *mc = MACHINE_CLASS(oc); + AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc); + + mc->desc = "Aspeed AST2600 EVB (Cortex A7)"; + amc->soc_name = "ast2600-a0"; + amc->hw_strap1 = AST2600_EVB_HW_STRAP1; + amc->hw_strap2 = AST2600_EVB_HW_STRAP2; + amc->fmc_model = "w25q512jv"; + amc->spi_model = "mx66u51235f"; + amc->num_cs = 1; + amc->i2c_init = ast2600_evb_i2c_init; + mc->default_ram_size = 1 * GiB; +}; + +static const TypeInfo aspeed_machine_types[] = { + { + .name = MACHINE_TYPE_NAME("palmetto-bmc"), + .parent = TYPE_ASPEED_MACHINE, + .class_init = aspeed_machine_palmetto_class_init, + }, { + .name = MACHINE_TYPE_NAME("ast2500-evb"), + .parent = TYPE_ASPEED_MACHINE, + .class_init = aspeed_machine_ast2500_evb_class_init, + }, { + .name = MACHINE_TYPE_NAME("romulus-bmc"), + .parent = TYPE_ASPEED_MACHINE, + .class_init = aspeed_machine_romulus_class_init, + }, { + .name = MACHINE_TYPE_NAME("swift-bmc"), + .parent = TYPE_ASPEED_MACHINE, + .class_init = aspeed_machine_swift_class_init, + }, { + .name = MACHINE_TYPE_NAME("witherspoon-bmc"), + .parent = TYPE_ASPEED_MACHINE, + .class_init = aspeed_machine_witherspoon_class_init, + }, { + .name = MACHINE_TYPE_NAME("ast2600-evb"), + .parent = TYPE_ASPEED_MACHINE, + .class_init = aspeed_machine_ast2600_evb_class_init, + }, { + .name = TYPE_ASPEED_MACHINE, + .parent = TYPE_MACHINE, + .instance_size = sizeof(AspeedMachine), + .class_size = sizeof(AspeedMachineClass), + .class_init = aspeed_machine_class_init, + .abstract = true, } -} +}; -type_init(aspeed_machine_types) +DEFINE_TYPES(aspeed_machine_types) -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> The Tacoma BMC board is replacement board for the BMC of the OpenPOWER Witherspoon system. It uses a AST2600 SoC instead of a AST2500 and the I2C layout is the same as it controls the same main board. Used for HW bringup. Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-15-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/aspeed.c | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/hw/arm/aspeed.c b/hw/arm/aspeed.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/aspeed.c +++ b/hw/arm/aspeed.c @@ -XXX,XX +XXX,XX @@ struct AspeedBoardState { #define AST2600_EVB_HW_STRAP1 0x000000C0 #define AST2600_EVB_HW_STRAP2 0x00000003 +/* Tacoma hardware value */ +#define TACOMA_BMC_HW_STRAP1 0x00000000 +#define TACOMA_BMC_HW_STRAP2 0x00000000 + /* * The max ram region is for firmwares that scan the address space * with load/store to guess how much RAM the SoC has. @@ -XXX,XX +XXX,XX @@ static void witherspoon_bmc_i2c_init(AspeedBoardState *bmc) AspeedSoCState *soc = &bmc->soc; uint8_t *eeprom_buf = g_malloc0(8 * 1024); + /* Bus 3: TODO bmp280@77 */ + /* Bus 3: TODO max31785@52 */ + /* Bus 3: TODO dps310@76 */ i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 3), TYPE_PCA9552, 0x60); @@ -XXX,XX +XXX,XX @@ static void witherspoon_bmc_i2c_init(AspeedBoardState *bmc) eeprom_buf); i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 11), TYPE_PCA9552, 0x60); + /* Bus 11: TODO ucd90160@64 */ } static void aspeed_machine_class_init(ObjectClass *oc, void *data) @@ -XXX,XX +XXX,XX @@ static void aspeed_machine_ast2600_evb_class_init(ObjectClass *oc, void *data) mc->default_ram_size = 1 * GiB; }; +static void aspeed_machine_tacoma_class_init(ObjectClass *oc, void *data) +{ + MachineClass *mc = MACHINE_CLASS(oc); + AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc); + + mc->desc = "Aspeed AST2600 EVB (Cortex A7)"; + amc->soc_name = "ast2600-a0"; + amc->hw_strap1 = TACOMA_BMC_HW_STRAP1; + amc->hw_strap2 = TACOMA_BMC_HW_STRAP2; + amc->fmc_model = "mx66l1g45g"; + amc->spi_model = "mx66l1g45g"; + amc->num_cs = 2; + amc->i2c_init = witherspoon_bmc_i2c_init; /* Same board layout */ + mc->default_ram_size = 1 * GiB; +}; + static const TypeInfo aspeed_machine_types[] = { { .name = MACHINE_TYPE_NAME("palmetto-bmc"), @@ -XXX,XX +XXX,XX @@ static const TypeInfo aspeed_machine_types[] = { .name = MACHINE_TYPE_NAME("ast2600-evb"), .parent = TYPE_ASPEED_MACHINE, .class_init = aspeed_machine_ast2600_evb_class_init, + }, { + .name = MACHINE_TYPE_NAME("tacoma-bmc"), + .parent = TYPE_ASPEED_MACHINE, + .class_init = aspeed_machine_tacoma_class_init, }, { .name = TYPE_ASPEED_MACHINE, .parent = TYPE_MACHINE, -- 2.20.1 From: PanNengyuan <pannengyuan@huawei.com> Address Sanitizer shows memory leak in hw/gpio/aspeed_gpio.c:875 Reported-by: Euler Robot <euler.robot@huawei.com> Signed-off-by: PanNengyuan <pannengyuan@huawei.com> Reviewed-by: Cédric Le Goater <clg@kaod.org> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-16-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/gpio/aspeed_gpio.c | 1 + 1 file changed, 1 insertion(+) diff --git a/hw/gpio/aspeed_gpio.c b/hw/gpio/aspeed_gpio.c index XXXXXXX..XXXXXXX 100644 --- a/hw/gpio/aspeed_gpio.c +++ b/hw/gpio/aspeed_gpio.c @@ -XXX,XX +XXX,XX @@ static void aspeed_gpio_init(Object *obj) pin_idx % GPIOS_PER_GROUP); object_property_add(obj, name, "bool", aspeed_gpio_get_pin, aspeed_gpio_set_pin, NULL, NULL, NULL); + g_free(name); } } -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> The Aspeed Watchdog and Timer models have a link pointing to the SCU controller model of the machine. Change the "scu" property definition so that it explicitly sets the pointer. The property isn't optional : not being able to set the link is a bug and QEMU should rather abort than exit in this case. Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Greg Kurz <groug@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-17-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/aspeed_ast2600.c | 8 ++++---- hw/arm/aspeed_soc.c | 8 ++++---- hw/timer/aspeed_timer.c | 17 +++++++++-------- hw/watchdog/wdt_aspeed.c | 17 ++++++++--------- 4 files changed, 25 insertions(+), 25 deletions(-) diff --git a/hw/arm/aspeed_ast2600.c b/hw/arm/aspeed_ast2600.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/aspeed_ast2600.c +++ b/hw/arm/aspeed_ast2600.c @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_ast2600_init(Object *obj) snprintf(typename, sizeof(typename), "aspeed.timer-%s", socname); sysbus_init_child_obj(obj, "timerctrl", OBJECT(&s->timerctrl), sizeof(s->timerctrl), typename); - object_property_add_const_link(OBJECT(&s->timerctrl), "scu", - OBJECT(&s->scu), &error_abort); snprintf(typename, sizeof(typename), "aspeed.i2c-%s", socname); sysbus_init_child_obj(obj, "i2c", OBJECT(&s->i2c), sizeof(s->i2c), @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_ast2600_init(Object *obj) snprintf(typename, sizeof(typename), "aspeed.wdt-%s", socname); sysbus_init_child_obj(obj, "wdt[*]", OBJECT(&s->wdt[i]), sizeof(s->wdt[i]), typename); - object_property_add_const_link(OBJECT(&s->wdt[i]), "scu", - OBJECT(&s->scu), &error_abort); } for (i = 0; i < sc->macs_num; i++) { @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_ast2600_realize(DeviceState *dev, Error **errp) aspeed_soc_get_irq(s, ASPEED_RTC)); /* Timer */ + object_property_set_link(OBJECT(&s->timerctrl), + OBJECT(&s->scu), "scu", &error_abort); object_property_set_bool(OBJECT(&s->timerctrl), true, "realized", &err); if (err) { error_propagate(errp, err); @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_ast2600_realize(DeviceState *dev, Error **errp) for (i = 0; i < sc->wdts_num; i++) { AspeedWDTClass *awc = ASPEED_WDT_GET_CLASS(&s->wdt[i]); + object_property_set_link(OBJECT(&s->wdt[i]), + OBJECT(&s->scu), "scu", &error_abort); object_property_set_bool(OBJECT(&s->wdt[i]), true, "realized", &err); if (err) { error_propagate(errp, err); diff --git a/hw/arm/aspeed_soc.c b/hw/arm/aspeed_soc.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/aspeed_soc.c +++ b/hw/arm/aspeed_soc.c @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_init(Object *obj) snprintf(typename, sizeof(typename), "aspeed.timer-%s", socname); sysbus_init_child_obj(obj, "timerctrl", OBJECT(&s->timerctrl), sizeof(s->timerctrl), typename); - object_property_add_const_link(OBJECT(&s->timerctrl), "scu", - OBJECT(&s->scu), &error_abort); snprintf(typename, sizeof(typename), "aspeed.i2c-%s", socname); sysbus_init_child_obj(obj, "i2c", OBJECT(&s->i2c), sizeof(s->i2c), @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_init(Object *obj) snprintf(typename, sizeof(typename), "aspeed.wdt-%s", socname); sysbus_init_child_obj(obj, "wdt[*]", OBJECT(&s->wdt[i]), sizeof(s->wdt[i]), typename); - object_property_add_const_link(OBJECT(&s->wdt[i]), "scu", - OBJECT(&s->scu), &error_abort); } for (i = 0; i < sc->macs_num; i++) { @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp) aspeed_soc_get_irq(s, ASPEED_RTC)); /* Timer */ + object_property_set_link(OBJECT(&s->timerctrl), + OBJECT(&s->scu), "scu", &error_abort); object_property_set_bool(OBJECT(&s->timerctrl), true, "realized", &err); if (err) { error_propagate(errp, err); @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp) for (i = 0; i < sc->wdts_num; i++) { AspeedWDTClass *awc = ASPEED_WDT_GET_CLASS(&s->wdt[i]); + object_property_set_link(OBJECT(&s->wdt[i]), + OBJECT(&s->scu), "scu", &error_abort); object_property_set_bool(OBJECT(&s->wdt[i]), true, "realized", &err); if (err) { error_propagate(errp, err); diff --git a/hw/timer/aspeed_timer.c b/hw/timer/aspeed_timer.c index XXXXXXX..XXXXXXX 100644 --- a/hw/timer/aspeed_timer.c +++ b/hw/timer/aspeed_timer.c @@ -XXX,XX +XXX,XX @@ #include "qemu/timer.h" #include "qemu/log.h" #include "qemu/module.h" +#include "hw/qdev-properties.h" #include "trace.h" #define TIMER_NR_REGS 4 @@ -XXX,XX +XXX,XX @@ static void aspeed_timer_realize(DeviceState *dev, Error **errp) int i; SysBusDevice *sbd = SYS_BUS_DEVICE(dev); AspeedTimerCtrlState *s = ASPEED_TIMER(dev); - Object *obj; - Error *err = NULL; - obj = object_property_get_link(OBJECT(dev), "scu", &err); - if (!obj) { - error_propagate_prepend(errp, err, "required link 'scu' not found: "); - return; - } - s->scu = ASPEED_SCU(obj); + assert(s->scu); for (i = 0; i < ASPEED_TIMER_NR_TIMERS; i++) { aspeed_init_one_timer(s, i); @@ -XXX,XX +XXX,XX @@ static const VMStateDescription vmstate_aspeed_timer_state = { } }; +static Property aspeed_timer_properties[] = { + DEFINE_PROP_LINK("scu", AspeedTimerCtrlState, scu, TYPE_ASPEED_SCU, + AspeedSCUState *), + DEFINE_PROP_END_OF_LIST(), +}; + static void timer_class_init(ObjectClass *klass, void *data) { DeviceClass *dc = DEVICE_CLASS(klass); @@ -XXX,XX +XXX,XX @@ static void timer_class_init(ObjectClass *klass, void *data) dc->reset = aspeed_timer_reset; dc->desc = "ASPEED Timer"; dc->vmsd = &vmstate_aspeed_timer_state; + dc->props = aspeed_timer_properties; } static const TypeInfo aspeed_timer_info = { diff --git a/hw/watchdog/wdt_aspeed.c b/hw/watchdog/wdt_aspeed.c index XXXXXXX..XXXXXXX 100644 --- a/hw/watchdog/wdt_aspeed.c +++ b/hw/watchdog/wdt_aspeed.c @@ -XXX,XX +XXX,XX @@ static void aspeed_wdt_realize(DeviceState *dev, Error **errp) { SysBusDevice *sbd = SYS_BUS_DEVICE(dev); AspeedWDTState *s = ASPEED_WDT(dev); - Error *err = NULL; - Object *obj; - obj = object_property_get_link(OBJECT(dev), "scu", &err); - if (!obj) { - error_propagate(errp, err); - error_prepend(errp, "required link 'scu' not found: "); - return; - } - s->scu = ASPEED_SCU(obj); + assert(s->scu); s->timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, aspeed_wdt_timer_expired, dev); @@ -XXX,XX +XXX,XX @@ static void aspeed_wdt_realize(DeviceState *dev, Error **errp) sysbus_init_mmio(sbd, &s->iomem); } +static Property aspeed_wdt_properties[] = { + DEFINE_PROP_LINK("scu", AspeedWDTState, scu, TYPE_ASPEED_SCU, + AspeedSCUState *), + DEFINE_PROP_END_OF_LIST(), +}; + static void aspeed_wdt_class_init(ObjectClass *klass, void *data) { DeviceClass *dc = DEVICE_CLASS(klass); @@ -XXX,XX +XXX,XX @@ static void aspeed_wdt_class_init(ObjectClass *klass, void *data) dc->reset = aspeed_wdt_reset; set_bit(DEVICE_CATEGORY_MISC, dc->categories); dc->vmsd = &vmstate_aspeed_wdt; + dc->props = aspeed_wdt_properties; } static const TypeInfo aspeed_wdt_info = { -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> The Aspeed MII model has a link pointing to its associated FTGMAC100 NIC in the machine. Change the "nic" property definition so that it explicitly sets the pointer. The property isn't optional : not being able to set the link is a bug and QEMU should rather abort than exit in this case. Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Greg Kurz <groug@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Cédric Le Goater <clg@kaod.org> Message-id: 20191119141211.25716-18-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/aspeed_ast2600.c | 5 ++--- hw/net/ftgmac100.c | 19 +++++++++---------- 2 files changed, 11 insertions(+), 13 deletions(-) diff --git a/hw/arm/aspeed_ast2600.c b/hw/arm/aspeed_ast2600.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/aspeed_ast2600.c +++ b/hw/arm/aspeed_ast2600.c @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_ast2600_init(Object *obj) sysbus_init_child_obj(obj, "mii[*]", &s->mii[i], sizeof(s->mii[i]), TYPE_ASPEED_MII); - object_property_add_const_link(OBJECT(&s->mii[i]), "nic", - OBJECT(&s->ftgmac100[i]), - &error_abort); } sysbus_init_child_obj(obj, "xdma", OBJECT(&s->xdma), sizeof(s->xdma), @@ -XXX,XX +XXX,XX @@ static void aspeed_soc_ast2600_realize(DeviceState *dev, Error **errp) sysbus_connect_irq(SYS_BUS_DEVICE(&s->ftgmac100[i]), 0, aspeed_soc_get_irq(s, ASPEED_ETH1 + i)); + object_property_set_link(OBJECT(&s->mii[i]), OBJECT(&s->ftgmac100[i]), + "nic", &error_abort); object_property_set_bool(OBJECT(&s->mii[i]), true, "realized", &err); if (err) { diff --git a/hw/net/ftgmac100.c b/hw/net/ftgmac100.c index XXXXXXX..XXXXXXX 100644 --- a/hw/net/ftgmac100.c +++ b/hw/net/ftgmac100.c @@ -XXX,XX +XXX,XX @@ static void aspeed_mii_realize(DeviceState *dev, Error **errp) { AspeedMiiState *s = ASPEED_MII(dev); SysBusDevice *sbd = SYS_BUS_DEVICE(dev); - Object *obj; - Error *local_err = NULL; - obj = object_property_get_link(OBJECT(dev), "nic", &local_err); - if (!obj) { - error_propagate(errp, local_err); - error_prepend(errp, "required link 'nic' not found: "); - return; - } - - s->nic = FTGMAC100(obj); + assert(s->nic); memory_region_init_io(&s->iomem, OBJECT(dev), &aspeed_mii_ops, s, TYPE_ASPEED_MII, 0x8); @@ -XXX,XX +XXX,XX @@ static const VMStateDescription vmstate_aspeed_mii = { VMSTATE_END_OF_LIST() } }; + +static Property aspeed_mii_properties[] = { + DEFINE_PROP_LINK("nic", AspeedMiiState, nic, TYPE_FTGMAC100, + FTGMAC100State *), + DEFINE_PROP_END_OF_LIST(), +}; + static void aspeed_mii_class_init(ObjectClass *klass, void *data) { DeviceClass *dc = DEVICE_CLASS(klass); @@ -XXX,XX +XXX,XX @@ static void aspeed_mii_class_init(ObjectClass *klass, void *data) dc->reset = aspeed_mii_reset; dc->realize = aspeed_mii_realize; dc->desc = "Aspeed MII controller"; + dc->props = aspeed_mii_properties; } static const TypeInfo aspeed_mii_info = { -- 2.20.1 From: Marc Zyngier <maz@kernel.org> HCR_EL2.TID2 mandates that access from EL1 to CTR_EL0, CCSIDR_EL1, CCSIDR2_EL1, CLIDR_EL1, CSSELR_EL1 are trapped to EL2, and QEMU completely ignores it, making it impossible for hypervisors to virtualize the cache hierarchy. Do the right thing by trapping to EL2 if HCR_EL2.TID2 is set. Signed-off-by: Marc Zyngier <maz@kernel.org> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20191201122018.25808-2-maz@kernel.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 31 +++++++++++++++++++++++++++---- 1 file changed, 27 insertions(+), 4 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 scr_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) raw_write(env, ri, value); } +static CPAccessResult access_aa64_tid2(CPUARMState *env, + const ARMCPRegInfo *ri, + bool isread) +{ + if (arm_current_el(env) == 1 && (arm_hcr_el2_eff(env) & HCR_TID2)) { + return CP_ACCESS_TRAP_EL2; + } + + return CP_ACCESS_OK; +} + static uint64_t ccsidr_read(CPUARMState *env, const ARMCPRegInfo *ri) { ARMCPU *cpu = env_archcpu(env); @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v7_cp_reginfo[] = { .writefn = pmintenclr_write }, { .name = "CCSIDR", .state = ARM_CP_STATE_BOTH, .opc0 = 3, .crn = 0, .crm = 0, .opc1 = 1, .opc2 = 0, - .access = PL1_R, .readfn = ccsidr_read, .type = ARM_CP_NO_RAW }, + .access = PL1_R, + .accessfn = access_aa64_tid2, + .readfn = ccsidr_read, .type = ARM_CP_NO_RAW }, { .name = "CSSELR", .state = ARM_CP_STATE_BOTH, .opc0 = 3, .crn = 0, .crm = 0, .opc1 = 2, .opc2 = 0, - .access = PL1_RW, .writefn = csselr_write, .resetvalue = 0, + .access = PL1_RW, + .accessfn = access_aa64_tid2, + .writefn = csselr_write, .resetvalue = 0, .bank_fieldoffsets = { offsetof(CPUARMState, cp15.csselr_s), offsetof(CPUARMState, cp15.csselr_ns) } }, /* Auxiliary ID register: this actually has an IMPDEF value but for now @@ -XXX,XX +XXX,XX @@ static CPAccessResult ctr_el0_access(CPUARMState *env, const ARMCPRegInfo *ri, if (arm_current_el(env) == 0 && !(env->cp15.sctlr_el[1] & SCTLR_UCT)) { return CP_ACCESS_TRAP; } + + if (arm_current_el(env) < 2 && arm_hcr_el2_eff(env) & HCR_TID2) { + return CP_ACCESS_TRAP_EL2; + } + return CP_ACCESS_OK; } @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) ARMCPRegInfo clidr = { .name = "CLIDR", .state = ARM_CP_STATE_BOTH, .opc0 = 3, .crn = 0, .crm = 0, .opc1 = 1, .opc2 = 1, - .access = PL1_R, .type = ARM_CP_CONST, .resetvalue = cpu->clidr + .access = PL1_R, .type = ARM_CP_CONST, + .accessfn = access_aa64_tid2, + .resetvalue = cpu->clidr }; define_one_arm_cp_reg(cpu, &clidr); define_arm_cp_regs(cpu, v7_cp_reginfo); @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) /* These are common to v8 and pre-v8 */ { .name = "CTR", .cp = 15, .crn = 0, .crm = 0, .opc1 = 0, .opc2 = 1, - .access = PL1_R, .type = ARM_CP_CONST, .resetvalue = cpu->ctr }, + .access = PL1_R, .accessfn = ctr_el0_access, + .type = ARM_CP_CONST, .resetvalue = cpu->ctr }, { .name = "CTR_EL0", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 3, .opc2 = 1, .crn = 0, .crm = 0, .access = PL0_R, .accessfn = ctr_el0_access, -- 2.20.1 From: Marc Zyngier <maz@kernel.org> HCR_EL2.TID1 mandates that access from EL1 to REVIDR_EL1, AIDR_EL1 (and their 32bit equivalents) as well as TCMTR, TLBTR are trapped to EL2. QEMU ignores it, making it harder for a hypervisor to virtualize the HW (though to be fair, no known hypervisor actually cares). Do the right thing by trapping to EL2 if HCR_EL2.TID1 is set. Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Signed-off-by: Marc Zyngier <maz@kernel.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20191201122018.25808-3-maz@kernel.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 36 ++++++++++++++++++++++++++++++++---- 1 file changed, 32 insertions(+), 4 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 uint64_t isr_read(CPUARMState *env, const ARMCPRegInfo *ri) return ret; } +static CPAccessResult access_aa64_tid1(CPUARMState *env, const ARMCPRegInfo *ri, + bool isread) +{ + if (arm_current_el(env) == 1 && (arm_hcr_el2_eff(env) & HCR_TID1)) { + return CP_ACCESS_TRAP_EL2; + } + + return CP_ACCESS_OK; +} + +static CPAccessResult access_aa32_tid1(CPUARMState *env, const ARMCPRegInfo *ri, + bool isread) +{ + if (arm_feature(env, ARM_FEATURE_V8)) { + return access_aa64_tid1(env, ri, isread); + } + + return CP_ACCESS_OK; +} + static const ARMCPRegInfo v7_cp_reginfo[] = { /* the old v6 WFI, UNPREDICTABLE in v7 but we choose to NOP */ { .name = "NOP", .cp = 15, .crn = 7, .crm = 0, .opc1 = 0, .opc2 = 4, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v7_cp_reginfo[] = { */ { .name = "AIDR", .state = ARM_CP_STATE_BOTH, .opc0 = 3, .opc1 = 1, .crn = 0, .crm = 0, .opc2 = 7, - .access = PL1_R, .type = ARM_CP_CONST, .resetvalue = 0 }, + .access = PL1_R, .type = ARM_CP_CONST, + .accessfn = access_aa64_tid1, + .resetvalue = 0 }, /* Auxiliary fault status registers: these also are IMPDEF, and we * choose to RAZ/WI for all cores. */ @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) .access = PL1_R, .resetvalue = cpu->midr }, { .name = "REVIDR_EL1", .state = ARM_CP_STATE_BOTH, .opc0 = 3, .opc1 = 0, .crn = 0, .crm = 0, .opc2 = 6, - .access = PL1_R, .type = ARM_CP_CONST, .resetvalue = cpu->revidr }, + .access = PL1_R, + .accessfn = access_aa64_tid1, + .type = ARM_CP_CONST, .resetvalue = cpu->revidr }, REGINFO_SENTINEL }; ARMCPRegInfo id_cp_reginfo[] = { @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) /* TCMTR and TLBTR exist in v8 but have no 64-bit versions */ { .name = "TCMTR", .cp = 15, .crn = 0, .crm = 0, .opc1 = 0, .opc2 = 2, - .access = PL1_R, .type = ARM_CP_CONST, .resetvalue = 0 }, + .access = PL1_R, + .accessfn = access_aa32_tid1, + .type = ARM_CP_CONST, .resetvalue = 0 }, REGINFO_SENTINEL }; /* TLBTR is specific to VMSA */ ARMCPRegInfo id_tlbtr_reginfo = { .name = "TLBTR", .cp = 15, .crn = 0, .crm = 0, .opc1 = 0, .opc2 = 3, - .access = PL1_R, .type = ARM_CP_CONST, .resetvalue = 0, + .access = PL1_R, + .accessfn = access_aa32_tid1, + .type = ARM_CP_CONST, .resetvalue = 0, }; /* MPUIR is specific to PMSA V6+ */ ARMCPRegInfo id_mpuir_reginfo = { -- 2.20.1 From: Marc Zyngier <maz@kernel.org> HCR_EL2.TID3 requires that AArch32 reads of MVFR[012] are trapped to EL2, and HCR_EL2.TID0 does the same for reads of FPSID. In order to handle this, introduce a new TCG helper function that checks for these control bits before executing the VMRC instruction. Tested with a hacked-up version of KVM/arm64 that sets the control bits for 32bit guests. Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Signed-off-by: Marc Zyngier <maz@kernel.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20191201122018.25808-4-maz@kernel.org [PMM: move helper declaration to helper.h; make it TCG_CALL_NO_WG] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 2 ++ target/arm/translate-vfp.inc.c | 20 ++++++++++++++++---- target/arm/vfp_helper.c | 29 +++++++++++++++++++++++++++++ 3 files changed, 47 insertions(+), 4 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_2(rintd, TCG_CALL_NO_RWG, f64, f64, ptr) DEF_HELPER_FLAGS_2(vjcvt, TCG_CALL_NO_RWG, i32, f64, env) DEF_HELPER_FLAGS_2(fjcvtzs, TCG_CALL_NO_RWG, i64, f64, ptr) +DEF_HELPER_FLAGS_3(check_hcr_el2_trap, TCG_CALL_NO_WG, void, env, i32, i32) + /* neon_helper.c */ DEF_HELPER_FLAGS_3(neon_qadd_u8, TCG_CALL_NO_RWG, i32, env, i32, i32) DEF_HELPER_FLAGS_3(neon_qadd_s8, TCG_CALL_NO_RWG, i32, env, i32, i32) diff --git a/target/arm/translate-vfp.inc.c b/target/arm/translate-vfp.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-vfp.inc.c +++ b/target/arm/translate-vfp.inc.c @@ -XXX,XX +XXX,XX @@ static bool trans_VMSR_VMRS(DisasContext *s, arg_VMSR_VMRS *a) if (a->l) { /* VMRS, move VFP special register to gp register */ switch (a->reg) { - case ARM_VFP_FPSID: - case ARM_VFP_FPEXC: - case ARM_VFP_FPINST: - case ARM_VFP_FPINST2: case ARM_VFP_MVFR0: case ARM_VFP_MVFR1: case ARM_VFP_MVFR2: + case ARM_VFP_FPSID: + if (s->current_el == 1) { + TCGv_i32 tcg_reg, tcg_rt; + + gen_set_condexec(s); + gen_set_pc_im(s, s->pc_curr); + tcg_reg = tcg_const_i32(a->reg); + tcg_rt = tcg_const_i32(a->rt); + gen_helper_check_hcr_el2_trap(cpu_env, tcg_rt, tcg_reg); + tcg_temp_free_i32(tcg_reg); + tcg_temp_free_i32(tcg_rt); + } + /* fall through */ + case ARM_VFP_FPEXC: + case ARM_VFP_FPINST: + case ARM_VFP_FPINST2: tmp = load_cpu_field(vfp.xregs[a->reg]); break; case ARM_VFP_FPSCR: diff --git a/target/arm/vfp_helper.c b/target/arm/vfp_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vfp_helper.c +++ b/target/arm/vfp_helper.c @@ -XXX,XX +XXX,XX @@ float64 HELPER(frint64_d)(float64 f, void *fpst) return frint_d(f, fpst, 64); } +void HELPER(check_hcr_el2_trap)(CPUARMState *env, uint32_t rt, uint32_t reg) +{ + uint32_t syndrome; + + switch (reg) { + case ARM_VFP_MVFR0: + case ARM_VFP_MVFR1: + case ARM_VFP_MVFR2: + if (!(arm_hcr_el2_eff(env) & HCR_TID3)) { + return; + } + break; + case ARM_VFP_FPSID: + if (!(arm_hcr_el2_eff(env) & HCR_TID0)) { + return; + } + break; + default: + g_assert_not_reached(); + } + + syndrome = ((EC_FPIDTRAP << ARM_EL_EC_SHIFT) + | ARM_EL_IL + | (1 << 24) | (0xe << 20) | (7 << 14) + | (reg << 10) | (rt << 5) | 1); + + raise_exception(env, EXCP_HYP_TRAP, syndrome, 2); +} + #endif -- 2.20.1 From: Marc Zyngier <maz@kernel.org> HSTR_EL2 offers a way to trap ranges of CP15 system register accesses to EL2, and it looks like this register is completely ignored by QEMU. To avoid adding extra .accessfn filters all over the place (which would have a direct performance impact), let's add a new TB flag that gets set whenever HSTR_EL2 is non-zero and that QEMU translates a context where this trap has a chance to apply, and only generate the extra access check if the hypervisor is actively using this feature. Tested with a hand-crafted KVM guest accessing CBAR. Signed-off-by: Marc Zyngier <maz@kernel.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20191201122018.25808-5-maz@kernel.org [PMM: use is_a64(); fix comment syntax] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.h | 2 ++ target/arm/translate.h | 2 ++ target/arm/helper.c | 6 ++++++ target/arm/op_helper.c | 22 ++++++++++++++++++++++ target/arm/translate.c | 3 ++- 5 files changed, 34 insertions(+), 1 deletion(-) 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 @@ FIELD(TBFLAG_A32, NS, 6, 1) FIELD(TBFLAG_A32, VFPEN, 7, 1) /* Partially cached, minus FPEXC. */ FIELD(TBFLAG_A32, CONDEXEC, 8, 8) /* Not cached. */ FIELD(TBFLAG_A32, SCTLR_B, 16, 1) +FIELD(TBFLAG_A32, HSTR_ACTIVE, 17, 1) + /* For M profile only, set if FPCCR.LSPACT is set */ FIELD(TBFLAG_A32, LSPACT, 18, 1) /* Not cached. */ /* For M profile only, set if we must create a new FP context */ diff --git a/target/arm/translate.h b/target/arm/translate.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.h +++ b/target/arm/translate.h @@ -XXX,XX +XXX,XX @@ typedef struct DisasContext { bool pauth_active; /* True with v8.5-BTI and SCTLR_ELx.BT* set. */ bool bt; + /* True if any CP15 access is trapped by HSTR_EL2 */ + bool hstr_active; /* * >= 0, a copy of PSTATE.BTYPE, which will be 0 without v8.5-BTI. * < 0, set by the current instruction. 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 uint32_t rebuild_hflags_a32(CPUARMState *env, int fp_el, if (arm_el_is_aa64(env, 1)) { flags = FIELD_DP32(flags, TBFLAG_A32, VFPEN, 1); } + + if (arm_current_el(env) < 2 && env->cp15.hstr_el2 && + (arm_hcr_el2_eff(env) & (HCR_E2H | HCR_TGE)) != (HCR_E2H | HCR_TGE)) { + flags = FIELD_DP32(flags, TBFLAG_A32, HSTR_ACTIVE, 1); + } + return rebuild_hflags_common_32(env, fp_el, mmu_idx, flags); } 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, raise_exception(env, EXCP_UDEF, syndrome, exception_target_el(env)); } + /* + * Check for an EL2 trap due to HSTR_EL2. We expect EL0 accesses + * to sysregs non accessible at EL0 to have UNDEF-ed already. + */ + if (!is_a64(env) && arm_current_el(env) < 2 && ri->cp == 15 && + (arm_hcr_el2_eff(env) & (HCR_E2H | HCR_TGE)) != (HCR_E2H | HCR_TGE)) { + uint32_t mask = 1 << ri->crn; + + if (ri->type & ARM_CP_64BIT) { + mask = 1 << ri->crm; + } + + /* T4 and T14 are RES0 */ + mask &= ~((1 << 4) | (1 << 14)); + + if (env->cp15.hstr_el2 & mask) { + target_el = 2; + goto exept; + } + } + if (!ri->accessfn) { return; } @@ -XXX,XX +XXX,XX @@ void HELPER(access_check_cp_reg)(CPUARMState *env, void *rip, uint32_t syndrome, g_assert_not_reached(); } +exept: raise_exception(env, EXCP_UDEF, syndrome, target_el); } 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 int disas_coproc_insn(DisasContext *s, uint32_t insn) return 1; } - if (ri->accessfn || + if (s->hstr_active || ri->accessfn || (arm_dc_feature(s, ARM_FEATURE_XSCALE) && cpnum < 14)) { /* Emit code to perform further access permissions checks at * runtime; this may result in an exception. @@ -XXX,XX +XXX,XX @@ static void arm_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs) !arm_el_is_aa64(env, 3); dc->thumb = FIELD_EX32(tb_flags, TBFLAG_A32, THUMB); dc->sctlr_b = FIELD_EX32(tb_flags, TBFLAG_A32, SCTLR_B); + dc->hstr_active = FIELD_EX32(tb_flags, TBFLAG_A32, HSTR_ACTIVE); dc->be_data = FIELD_EX32(tb_flags, TBFLAG_ANY, BE_DATA) ? MO_BE : MO_LE; condexec = FIELD_EX32(tb_flags, TBFLAG_A32, CONDEXEC); dc->condexec_mask = (condexec & 0xf) << 1; -- 2.20.1 From: Marc Zyngier <maz@kernel.org> QEMU lacks the minimum Jazelle implementation that is required by the architecture (everything is RAZ or RAZ/WI). Add it together with the HCR_EL2.TID0 trapping that goes with it. Signed-off-by: Marc Zyngier <maz@kernel.org> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20191201122018.25808-6-maz@kernel.org [PMM: moved ARMCPRegInfo array to file scope, marked it 'static global', moved new condition down in register_cp_regs_for_features() to go with other feature things rather than up with the v6/v7/v8 stuff] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 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 CPAccessResult access_aa32_tid3(CPUARMState *env, const ARMCPRegInfo *ri, return CP_ACCESS_OK; } +static CPAccessResult access_jazelle(CPUARMState *env, const ARMCPRegInfo *ri, + bool isread) +{ + if (arm_current_el(env) == 1 && (arm_hcr_el2_eff(env) & HCR_TID0)) { + return CP_ACCESS_TRAP_EL2; + } + + return CP_ACCESS_OK; +} + +static const ARMCPRegInfo jazelle_regs[] = { + { .name = "JIDR", + .cp = 14, .crn = 0, .crm = 0, .opc1 = 7, .opc2 = 0, + .access = PL1_R, .accessfn = access_jazelle, + .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "JOSCR", + .cp = 14, .crn = 1, .crm = 0, .opc1 = 7, .opc2 = 0, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "JMCR", + .cp = 14, .crn = 2, .crm = 0, .opc1 = 7, .opc2 = 0, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + REGINFO_SENTINEL +}; + void register_cp_regs_for_features(ARMCPU *cpu) { /* Register all the coprocessor registers based on feature bits */ @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) if (arm_feature(env, ARM_FEATURE_LPAE)) { define_arm_cp_regs(cpu, lpae_cp_reginfo); } + if (cpu_isar_feature(jazelle, cpu)) { + define_arm_cp_regs(cpu, jazelle_regs); + } /* Slightly awkwardly, the OMAP and StrongARM cores need all of * cp15 crn=0 to be writes-ignored, whereas for other cores they should * be read-only (ie write causes UNDEF exception). -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> This change ensures that the FPU can be accessed in Non-Secure mode when the CPU core is reset using the arm_set_cpu_on() function call. The NSACR.{CP11,CP10} bits define the exception level required to access the FPU in Non-Secure mode. Without these bits set, the CPU will give an undefined exception trap on the first FPU access for the secondary cores under Linux. This is necessary because in this power-control codepath QEMU is effectively emulating a bit of EL3 firmware, and has to set the CPU up as the EL3 firmware would. Fixes: fc1120a7f5 Cc: qemu-stable@nongnu.org Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> [PMM: added clarifying para to commit message] Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/arm-powerctl.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/target/arm/arm-powerctl.c b/target/arm/arm-powerctl.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/arm-powerctl.c +++ b/target/arm/arm-powerctl.c @@ -XXX,XX +XXX,XX @@ static void arm_set_cpu_on_async_work(CPUState *target_cpu_state, /* Processor is not in secure mode */ target_cpu->env.cp15.scr_el3 |= SCR_NS; + /* Set NSACR.{CP11,CP10} so NS can access the FPU */ + target_cpu->env.cp15.nsacr |= 3 << 10; + /* * If QEMU is providing the equivalent of EL3 firmware, then we need * to make sure a CPU targeting EL2 comes out of reset with a -- 2.20.1 From: Beata Michalska <beata.michalska@linaro.org> Add probe_read alongside the write probing equivalent. Signed-off-by: Beata Michalska <beata.michalska@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20191121000843.24844-2-beata.michalska@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/exec/exec-all.h | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ static inline void *probe_write(CPUArchState *env, target_ulong addr, int size, return probe_access(env, addr, size, MMU_DATA_STORE, mmu_idx, retaddr); } +static inline void *probe_read(CPUArchState *env, target_ulong addr, int size, + int mmu_idx, uintptr_t retaddr) +{ + return probe_access(env, addr, size, MMU_DATA_LOAD, mmu_idx, retaddr); +} + #define CODE_GEN_ALIGN 16 /* must be >= of the size of a icache line */ /* Estimated block size for TB allocation. */ -- 2.20.1 From: Beata Michalska <beata.michalska@linaro.org> Add an option to trigger memory writeback to sync given memory region with the corresponding backing store, case one is available. This extends the support for persistent memory, allowing syncing on-demand. Signed-off-by: Beata Michalska <beata.michalska@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20191121000843.24844-3-beata.michalska@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/exec/memory.h | 6 ++++++ include/exec/ram_addr.h | 8 ++++++++ include/qemu/cutils.h | 1 + exec.c | 36 ++++++++++++++++++++++++++++++++++++ memory.c | 12 ++++++++++++ util/cutils.c | 38 ++++++++++++++++++++++++++++++++++++++ 6 files changed, 101 insertions(+) diff --git a/include/exec/memory.h b/include/exec/memory.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/memory.h +++ b/include/exec/memory.h @@ -XXX,XX +XXX,XX @@ void *memory_region_get_ram_ptr(MemoryRegion *mr); */ void memory_region_ram_resize(MemoryRegion *mr, ram_addr_t newsize, Error **errp); +/** + * memory_region_do_writeback: Trigger writeback for selected address range + * [addr, addr + size] + * + */ +void memory_region_do_writeback(MemoryRegion *mr, hwaddr addr, hwaddr size); /** * memory_region_set_log: Turn dirty logging on or off for a region. diff --git a/include/exec/ram_addr.h b/include/exec/ram_addr.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/ram_addr.h +++ b/include/exec/ram_addr.h @@ -XXX,XX +XXX,XX @@ void qemu_ram_free(RAMBlock *block); int qemu_ram_resize(RAMBlock *block, ram_addr_t newsize, Error **errp); +void qemu_ram_writeback(RAMBlock *block, ram_addr_t start, ram_addr_t length); + +/* Clear whole block of mem */ +static inline void qemu_ram_block_writeback(RAMBlock *block) +{ + qemu_ram_writeback(block, 0, block->used_length); +} + #define DIRTY_CLIENTS_ALL ((1 << DIRTY_MEMORY_NUM) - 1) #define DIRTY_CLIENTS_NOCODE (DIRTY_CLIENTS_ALL & ~(1 << DIRTY_MEMORY_CODE)) diff --git a/include/qemu/cutils.h b/include/qemu/cutils.h index XXXXXXX..XXXXXXX 100644 --- a/include/qemu/cutils.h +++ b/include/qemu/cutils.h @@ -XXX,XX +XXX,XX @@ const char *qemu_strchrnul(const char *s, int c); #endif time_t mktimegm(struct tm *tm); int qemu_fdatasync(int fd); +int qemu_msync(void *addr, size_t length, int fd); int fcntl_setfl(int fd, int flag); int qemu_parse_fd(const char *param); int qemu_strtoi(const char *nptr, const char **endptr, int base, diff --git a/exec.c b/exec.c index XXXXXXX..XXXXXXX 100644 --- a/exec.c +++ b/exec.c @@ -XXX,XX +XXX,XX @@ #include "exec/ram_addr.h" #include "exec/log.h" +#include "qemu/pmem.h" + #include "migration/vmstate.h" #include "qemu/range.h" @@ -XXX,XX +XXX,XX @@ int qemu_ram_resize(RAMBlock *block, ram_addr_t newsize, Error **errp) return 0; } +/* + * Trigger sync on the given ram block for range [start, start + length] + * with the backing store if one is available. + * Otherwise no-op. + * @Note: this is supposed to be a synchronous op. + */ +void qemu_ram_writeback(RAMBlock *block, ram_addr_t start, ram_addr_t length) +{ + void *addr = ramblock_ptr(block, start); + + /* The requested range should fit in within the block range */ + g_assert((start + length) <= block->used_length); + +#ifdef CONFIG_LIBPMEM + /* The lack of support for pmem should not block the sync */ + if (ramblock_is_pmem(block)) { + pmem_persist(addr, length); + return; + } +#endif + if (block->fd >= 0) { + /** + * Case there is no support for PMEM or the memory has not been + * specified as persistent (or is not one) - use the msync. + * Less optimal but still achieves the same goal + */ + if (qemu_msync(addr, length, block->fd)) { + warn_report("%s: failed to sync memory range: start: " + RAM_ADDR_FMT " length: " RAM_ADDR_FMT, + __func__, start, length); + } + } +} + /* Called with ram_list.mutex held */ static void dirty_memory_extend(ram_addr_t old_ram_size, ram_addr_t new_ram_size) diff --git a/memory.c b/memory.c index XXXXXXX..XXXXXXX 100644 --- a/memory.c +++ b/memory.c @@ -XXX,XX +XXX,XX @@ void memory_region_ram_resize(MemoryRegion *mr, ram_addr_t newsize, Error **errp qemu_ram_resize(mr->ram_block, newsize, errp); } + +void memory_region_do_writeback(MemoryRegion *mr, hwaddr addr, hwaddr size) +{ + /* + * Might be extended case needed to cover + * different types of memory regions + */ + if (mr->ram_block && mr->dirty_log_mask) { + qemu_ram_writeback(mr->ram_block, addr, size); + } +} + /* * Call proper memory listeners about the change on the newly * added/removed CoalescedMemoryRange. diff --git a/util/cutils.c b/util/cutils.c index XXXXXXX..XXXXXXX 100644 --- a/util/cutils.c +++ b/util/cutils.c @@ -XXX,XX +XXX,XX @@ int qemu_fdatasync(int fd) #endif } +/** + * Sync changes made to the memory mapped file back to the backing + * storage. For POSIX compliant systems this will fallback + * to regular msync call. Otherwise it will trigger whole file sync + * (including the metadata case there is no support to skip that otherwise) + * + * @addr - start of the memory area to be synced + * @length - length of the are to be synced + * @fd - file descriptor for the file to be synced + * (mandatory only for POSIX non-compliant systems) + */ +int qemu_msync(void *addr, size_t length, int fd) +{ +#ifdef CONFIG_POSIX + size_t align_mask = ~(qemu_real_host_page_size - 1); + + /** + * There are no strict reqs as per the length of mapping + * to be synced. Still the length needs to follow the address + * alignment changes. Additionally - round the size to the multiple + * of PAGE_SIZE + */ + length += ((uintptr_t)addr & (qemu_real_host_page_size - 1)); + length = (length + ~align_mask) & align_mask; + + addr = (void *)((uintptr_t)addr & align_mask); + + return msync(addr, length, MS_SYNC); +#else /* CONFIG_POSIX */ + /** + * Perform the sync based on the file descriptor + * The sync range will most probably be wider than the one + * requested - but it will still get the job done + */ + return qemu_fdatasync(fd); +#endif /* CONFIG_POSIX */ +} + #ifndef _WIN32 /* Sets a specific flag */ int fcntl_setfl(int fd, int flag) -- 2.20.1 From: Beata Michalska <beata.michalska@linaro.org> Switch to ram block writeback for pmem migration. Signed-off-by: Beata Michalska <beata.michalska@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Message-id: 20191121000843.24844-4-beata.michalska@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- migration/ram.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/migration/ram.c b/migration/ram.c index XXXXXXX..XXXXXXX 100644 --- a/migration/ram.c +++ b/migration/ram.c @@ -XXX,XX +XXX,XX @@ #include "qemu/bitops.h" #include "qemu/bitmap.h" #include "qemu/main-loop.h" -#include "qemu/pmem.h" #include "xbzrle.h" #include "ram.h" #include "migration.h" @@ -XXX,XX +XXX,XX @@ static int ram_load_cleanup(void *opaque) RAMBlock *rb; RAMBLOCK_FOREACH_NOT_IGNORED(rb) { - if (ramblock_is_pmem(rb)) { - pmem_persist(rb->host, rb->used_length); - } + qemu_ram_block_writeback(rb); } xbzrle_load_cleanup(); -- 2.20.1 From: Beata Michalska <beata.michalska@linaro.org> ARMv8.2 introduced support for Data Cache Clean instructions to PoP (point-of-persistence) - DC CVAP and PoDP (point-of-deep-persistence) - DV CVADP. Both specify conceptual points in a memory system where all writes that are to reach them are considered persistent. The support provided considers both to be actually the same so there is no distinction between the two. If none is available (there is no backing store for given memory) both will result in Data Cache Clean up to the point of coherency. Otherwise sync for the specified range shall be performed. Signed-off-by: Beata Michalska <beata.michalska@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20191121000843.24844-5-beata.michalska@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.h | 10 ++++++++ linux-user/elfload.c | 2 ++ target/arm/cpu64.c | 1 + target/arm/helper.c | 56 ++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 69 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_frint(const ARMISARegisters *id) return FIELD_EX64(id->id_aa64isar1, ID_AA64ISAR1, FRINTTS) != 0; } +static inline bool isar_feature_aa64_dcpop(const ARMISARegisters *id) +{ + return FIELD_EX64(id->id_aa64isar1, ID_AA64ISAR1, DPB) != 0; +} + +static inline bool isar_feature_aa64_dcpodp(const ARMISARegisters *id) +{ + return FIELD_EX64(id->id_aa64isar1, ID_AA64ISAR1, DPB) >= 2; +} + static inline bool isar_feature_aa64_fp16(const ARMISARegisters *id) { /* We always set the AdvSIMD and FP fields identically wrt FP16. */ diff --git a/linux-user/elfload.c b/linux-user/elfload.c index XXXXXXX..XXXXXXX 100644 --- a/linux-user/elfload.c +++ b/linux-user/elfload.c @@ -XXX,XX +XXX,XX @@ static uint32_t get_elf_hwcap(void) GET_FEATURE_ID(aa64_jscvt, ARM_HWCAP_A64_JSCVT); GET_FEATURE_ID(aa64_sb, ARM_HWCAP_A64_SB); GET_FEATURE_ID(aa64_condm_4, ARM_HWCAP_A64_FLAGM); + GET_FEATURE_ID(aa64_dcpop, ARM_HWCAP_A64_DCPOP); return hwcaps; } @@ -XXX,XX +XXX,XX @@ static uint32_t get_elf_hwcap2(void) ARMCPU *cpu = ARM_CPU(thread_cpu); uint32_t hwcaps = 0; + GET_FEATURE_ID(aa64_dcpodp, ARM_HWCAP2_A64_DCPODP); GET_FEATURE_ID(aa64_condm_5, ARM_HWCAP2_A64_FLAGM2); GET_FEATURE_ID(aa64_frint, ARM_HWCAP2_A64_FRINT); 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 void aarch64_max_initfn(Object *obj) cpu->isar.id_aa64isar0 = t; t = cpu->isar.id_aa64isar1; + t = FIELD_DP64(t, ID_AA64ISAR1, DPB, 2); t = FIELD_DP64(t, ID_AA64ISAR1, JSCVT, 1); t = FIELD_DP64(t, ID_AA64ISAR1, FCMA, 1); t = FIELD_DP64(t, ID_AA64ISAR1, APA, 1); /* PAuth, architected only */ 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 rndr_reginfo[] = { .access = PL0_R, .readfn = rndr_readfn }, REGINFO_SENTINEL }; + +#ifndef CONFIG_USER_ONLY +static void dccvap_writefn(CPUARMState *env, const ARMCPRegInfo *opaque, + uint64_t value) +{ + ARMCPU *cpu = env_archcpu(env); + /* CTR_EL0 System register -> DminLine, bits [19:16] */ + uint64_t dline_size = 4 << ((cpu->ctr >> 16) & 0xF); + uint64_t vaddr_in = (uint64_t) value; + uint64_t vaddr = vaddr_in & ~(dline_size - 1); + void *haddr; + int mem_idx = cpu_mmu_index(env, false); + + /* This won't be crossing page boundaries */ + haddr = probe_read(env, vaddr, dline_size, mem_idx, GETPC()); + if (haddr) { + + ram_addr_t offset; + MemoryRegion *mr; + + /* RCU lock is already being held */ + mr = memory_region_from_host(haddr, &offset); + + if (mr) { + memory_region_do_writeback(mr, offset, dline_size); + } + } +} + +static const ARMCPRegInfo dcpop_reg[] = { + { .name = "DC_CVAP", .state = ARM_CP_STATE_AA64, + .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_access, .writefn = dccvap_writefn }, + REGINFO_SENTINEL +}; + +static const ARMCPRegInfo dcpodp_reg[] = { + { .name = "DC_CVADP", .state = ARM_CP_STATE_AA64, + .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_access, .writefn = dccvap_writefn }, + REGINFO_SENTINEL +}; +#endif /*CONFIG_USER_ONLY*/ + #endif static CPAccessResult access_predinv(CPUARMState *env, const ARMCPRegInfo *ri, @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) if (cpu_isar_feature(aa64_rndr, cpu)) { define_arm_cp_regs(cpu, rndr_reginfo); } +#ifndef CONFIG_USER_ONLY + /* Data Cache clean instructions up to PoP */ + if (cpu_isar_feature(aa64_dcpop, cpu)) { + define_one_arm_cp_reg(cpu, dcpop_reg); + + if (cpu_isar_feature(aa64_dcpodp, cpu)) { + define_one_arm_cp_reg(cpu, dcpodp_reg); + } + } +#endif /*CONFIG_USER_ONLY*/ #endif /* -- 2.20.1 From: Philippe Mathieu-Daudé <philmd@redhat.com> Make the gic a field in the machine state, and instead of filling an array of qemu_irq and passing it around, directly call qdev_get_gpio_in() on the gic field. Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20191206162303.30338-1-philmd@redhat.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/sbsa-ref.c | 86 +++++++++++++++++++++++------------------------ 1 file changed, 42 insertions(+), 44 deletions(-) diff --git a/hw/arm/sbsa-ref.c b/hw/arm/sbsa-ref.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/sbsa-ref.c +++ b/hw/arm/sbsa-ref.c @@ -XXX,XX +XXX,XX @@ typedef struct { void *fdt; int fdt_size; int psci_conduit; + DeviceState *gic; PFlashCFI01 *flash[2]; } SBSAMachineState; @@ -XXX,XX +XXX,XX @@ static void create_secure_ram(SBSAMachineState *sms, memory_region_add_subregion(secure_sysmem, base, secram); } -static void create_gic(SBSAMachineState *sms, qemu_irq *pic) +static void create_gic(SBSAMachineState *sms) { unsigned int smp_cpus = MACHINE(sms)->smp.cpus; - DeviceState *gicdev; SysBusDevice *gicbusdev; const char *gictype; uint32_t redist0_capacity, redist0_count; @@ -XXX,XX +XXX,XX @@ static void create_gic(SBSAMachineState *sms, qemu_irq *pic) gictype = gicv3_class_name(); - gicdev = qdev_create(NULL, gictype); - qdev_prop_set_uint32(gicdev, "revision", 3); - qdev_prop_set_uint32(gicdev, "num-cpu", smp_cpus); + sms->gic = qdev_create(NULL, gictype); + qdev_prop_set_uint32(sms->gic, "revision", 3); + qdev_prop_set_uint32(sms->gic, "num-cpu", smp_cpus); /* * Note that the num-irq property counts both internal and external * interrupts; there are always 32 of the former (mandated by GIC spec). */ - qdev_prop_set_uint32(gicdev, "num-irq", NUM_IRQS + 32); - qdev_prop_set_bit(gicdev, "has-security-extensions", true); + qdev_prop_set_uint32(sms->gic, "num-irq", NUM_IRQS + 32); + qdev_prop_set_bit(sms->gic, "has-security-extensions", true); redist0_capacity = sbsa_ref_memmap[SBSA_GIC_REDIST].size / GICV3_REDIST_SIZE; redist0_count = MIN(smp_cpus, redist0_capacity); - qdev_prop_set_uint32(gicdev, "len-redist-region-count", 1); - qdev_prop_set_uint32(gicdev, "redist-region-count[0]", redist0_count); + qdev_prop_set_uint32(sms->gic, "len-redist-region-count", 1); + qdev_prop_set_uint32(sms->gic, "redist-region-count[0]", redist0_count); - qdev_init_nofail(gicdev); - gicbusdev = SYS_BUS_DEVICE(gicdev); + qdev_init_nofail(sms->gic); + gicbusdev = SYS_BUS_DEVICE(sms->gic); sysbus_mmio_map(gicbusdev, 0, sbsa_ref_memmap[SBSA_GIC_DIST].base); sysbus_mmio_map(gicbusdev, 1, sbsa_ref_memmap[SBSA_GIC_REDIST].base); @@ -XXX,XX +XXX,XX @@ static void create_gic(SBSAMachineState *sms, qemu_irq *pic) for (irq = 0; irq < ARRAY_SIZE(timer_irq); irq++) { qdev_connect_gpio_out(cpudev, irq, - qdev_get_gpio_in(gicdev, + qdev_get_gpio_in(sms->gic, ppibase + timer_irq[irq])); } qdev_connect_gpio_out_named(cpudev, "gicv3-maintenance-interrupt", 0, - qdev_get_gpio_in(gicdev, ppibase + qdev_get_gpio_in(sms->gic, ppibase + ARCH_GIC_MAINT_IRQ)); qdev_connect_gpio_out_named(cpudev, "pmu-interrupt", 0, - qdev_get_gpio_in(gicdev, ppibase + qdev_get_gpio_in(sms->gic, ppibase + VIRTUAL_PMU_IRQ)); sysbus_connect_irq(gicbusdev, i, qdev_get_gpio_in(cpudev, ARM_CPU_IRQ)); @@ -XXX,XX +XXX,XX @@ static void create_gic(SBSAMachineState *sms, qemu_irq *pic) sysbus_connect_irq(gicbusdev, i + 3 * smp_cpus, qdev_get_gpio_in(cpudev, ARM_CPU_VFIQ)); } - - for (i = 0; i < NUM_IRQS; i++) { - pic[i] = qdev_get_gpio_in(gicdev, i); - } } -static void create_uart(const SBSAMachineState *sms, qemu_irq *pic, int uart, +static void create_uart(const SBSAMachineState *sms, int uart, MemoryRegion *mem, Chardev *chr) { hwaddr base = sbsa_ref_memmap[uart].base; @@ -XXX,XX +XXX,XX @@ static void create_uart(const SBSAMachineState *sms, qemu_irq *pic, int uart, qdev_init_nofail(dev); memory_region_add_subregion(mem, base, sysbus_mmio_get_region(s, 0)); - sysbus_connect_irq(s, 0, pic[irq]); + sysbus_connect_irq(s, 0, qdev_get_gpio_in(sms->gic, irq)); } -static void create_rtc(const SBSAMachineState *sms, qemu_irq *pic) +static void create_rtc(const SBSAMachineState *sms) { hwaddr base = sbsa_ref_memmap[SBSA_RTC].base; int irq = sbsa_ref_irqmap[SBSA_RTC]; - sysbus_create_simple("pl031", base, pic[irq]); + sysbus_create_simple("pl031", base, qdev_get_gpio_in(sms->gic, irq)); } static DeviceState *gpio_key_dev; @@ -XXX,XX +XXX,XX @@ static Notifier sbsa_ref_powerdown_notifier = { .notify = sbsa_ref_powerdown_req }; -static void create_gpio(const SBSAMachineState *sms, qemu_irq *pic) +static void create_gpio(const SBSAMachineState *sms) { DeviceState *pl061_dev; hwaddr base = sbsa_ref_memmap[SBSA_GPIO].base; int irq = sbsa_ref_irqmap[SBSA_GPIO]; - pl061_dev = sysbus_create_simple("pl061", base, pic[irq]); + pl061_dev = sysbus_create_simple("pl061", base, + qdev_get_gpio_in(sms->gic, irq)); gpio_key_dev = sysbus_create_simple("gpio-key", -1, qdev_get_gpio_in(pl061_dev, 3)); @@ -XXX,XX +XXX,XX @@ static void create_gpio(const SBSAMachineState *sms, qemu_irq *pic) qemu_register_powerdown_notifier(&sbsa_ref_powerdown_notifier); } -static void create_ahci(const SBSAMachineState *sms, qemu_irq *pic) +static void create_ahci(const SBSAMachineState *sms) { hwaddr base = sbsa_ref_memmap[SBSA_AHCI].base; int irq = sbsa_ref_irqmap[SBSA_AHCI]; @@ -XXX,XX +XXX,XX @@ static void create_ahci(const SBSAMachineState *sms, qemu_irq *pic) qdev_prop_set_uint32(dev, "num-ports", NUM_SATA_PORTS); qdev_init_nofail(dev); sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, base); - sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, pic[irq]); + sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, qdev_get_gpio_in(sms->gic, irq)); sysahci = SYSBUS_AHCI(dev); ahci = &sysahci->ahci; @@ -XXX,XX +XXX,XX @@ static void create_ahci(const SBSAMachineState *sms, qemu_irq *pic) } } -static void create_ehci(const SBSAMachineState *sms, qemu_irq *pic) +static void create_ehci(const SBSAMachineState *sms) { hwaddr base = sbsa_ref_memmap[SBSA_EHCI].base; int irq = sbsa_ref_irqmap[SBSA_EHCI]; - sysbus_create_simple("platform-ehci-usb", base, pic[irq]); + sysbus_create_simple("platform-ehci-usb", base, + qdev_get_gpio_in(sms->gic, irq)); } -static void create_smmu(const SBSAMachineState *sms, qemu_irq *pic, - PCIBus *bus) +static void create_smmu(const SBSAMachineState *sms, PCIBus *bus) { hwaddr base = sbsa_ref_memmap[SBSA_SMMU].base; int irq = sbsa_ref_irqmap[SBSA_SMMU]; @@ -XXX,XX +XXX,XX @@ static void create_smmu(const SBSAMachineState *sms, qemu_irq *pic, qdev_init_nofail(dev); sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, base); for (i = 0; i < NUM_SMMU_IRQS; i++) { - sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, pic[irq + i]); + sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, + qdev_get_gpio_in(sms->gic, irq + 1)); } } -static void create_pcie(SBSAMachineState *sms, qemu_irq *pic) +static void create_pcie(SBSAMachineState *sms) { hwaddr base_ecam = sbsa_ref_memmap[SBSA_PCIE_ECAM].base; hwaddr size_ecam = sbsa_ref_memmap[SBSA_PCIE_ECAM].size; @@ -XXX,XX +XXX,XX @@ static void create_pcie(SBSAMachineState *sms, qemu_irq *pic) sysbus_mmio_map(SYS_BUS_DEVICE(dev), 2, base_pio); for (i = 0; i < GPEX_NUM_IRQS; i++) { - sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, pic[irq + i]); + sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, + qdev_get_gpio_in(sms->gic, irq + 1)); gpex_set_irq_num(GPEX_HOST(dev), i, irq + i); } @@ -XXX,XX +XXX,XX @@ static void create_pcie(SBSAMachineState *sms, qemu_irq *pic) pci_create_simple(pci->bus, -1, "VGA"); - create_smmu(sms, pic, pci->bus); + create_smmu(sms, pci->bus); } static void *sbsa_ref_dtb(const struct arm_boot_info *binfo, int *fdt_size) @@ -XXX,XX +XXX,XX @@ static void sbsa_ref_init(MachineState *machine) bool firmware_loaded; const CPUArchIdList *possible_cpus; int n, sbsa_max_cpus; - qemu_irq pic[NUM_IRQS]; if (strcmp(machine->cpu_type, ARM_CPU_TYPE_NAME("cortex-a57"))) { error_report("sbsa-ref: CPU type other than the built-in " @@ -XXX,XX +XXX,XX @@ static void sbsa_ref_init(MachineState *machine) create_secure_ram(sms, secure_sysmem); - create_gic(sms, pic); + create_gic(sms); - create_uart(sms, pic, SBSA_UART, sysmem, serial_hd(0)); - create_uart(sms, pic, SBSA_SECURE_UART, secure_sysmem, serial_hd(1)); + create_uart(sms, SBSA_UART, sysmem, serial_hd(0)); + create_uart(sms, SBSA_SECURE_UART, secure_sysmem, serial_hd(1)); /* Second secure UART for RAS and MM from EL0 */ - create_uart(sms, pic, SBSA_SECURE_UART_MM, secure_sysmem, serial_hd(2)); + create_uart(sms, SBSA_SECURE_UART_MM, secure_sysmem, serial_hd(2)); - create_rtc(sms, pic); + create_rtc(sms); - create_gpio(sms, pic); + create_gpio(sms); - create_ahci(sms, pic); + create_ahci(sms); - create_ehci(sms, pic); + create_ehci(sms); - create_pcie(sms, pic); + create_pcie(sms); sms->bootinfo.ram_size = machine->ram_size; sms->bootinfo.nb_cpus = smp_cpus; -- 2.20.1 From: Heyi Guo <guoheyi@huawei.com> The last argument of AML bit and/or statement is the target variable, so we don't need to use a NULL target and then an additional store operation; using just aml_and() or aml_or() statement is enough. Also update tests/data/acpi/virt/DSDT* to pass "make check". Cc: Shannon Zhao <shannon.zhaosl@gmail.com> Cc: Peter Maydell <peter.maydell@linaro.org> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Igor Mammedov <imammedo@redhat.com> Suggested-by: Igor Mammedov <imammedo@redhat.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Heyi Guo <guoheyi@huawei.com> Message-id: 20191209063719.23086-2-guoheyi@huawei.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/virt-acpi-build.c | 16 ++++++++-------- tests/data/acpi/virt/DSDT | Bin 18470 -> 18462 bytes tests/data/acpi/virt/DSDT.memhp | Bin 19807 -> 19799 bytes tests/data/acpi/virt/DSDT.numamem | Bin 18470 -> 18462 bytes 4 files changed, 8 insertions(+), 8 deletions(-) diff --git a/hw/arm/virt-acpi-build.c b/hw/arm/virt-acpi-build.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt-acpi-build.c +++ b/hw/arm/virt-acpi-build.c @@ -XXX,XX +XXX,XX @@ static void acpi_dsdt_add_pci(Aml *scope, const MemMapEntry *memmap, aml_create_dword_field(aml_arg(3), aml_int(8), "CDW3")); aml_append(ifctx, aml_store(aml_name("CDW2"), aml_name("SUPP"))); aml_append(ifctx, aml_store(aml_name("CDW3"), aml_name("CTRL"))); - aml_append(ifctx, aml_store(aml_and(aml_name("CTRL"), aml_int(0x1D), NULL), - aml_name("CTRL"))); + aml_append(ifctx, aml_and(aml_name("CTRL"), aml_int(0x1D), + aml_name("CTRL"))); ifctx1 = aml_if(aml_lnot(aml_equal(aml_arg(1), aml_int(0x1)))); - aml_append(ifctx1, aml_store(aml_or(aml_name("CDW1"), aml_int(0x08), NULL), - aml_name("CDW1"))); + aml_append(ifctx1, aml_or(aml_name("CDW1"), aml_int(0x08), + aml_name("CDW1"))); aml_append(ifctx, ifctx1); ifctx1 = aml_if(aml_lnot(aml_equal(aml_name("CDW3"), aml_name("CTRL")))); - aml_append(ifctx1, aml_store(aml_or(aml_name("CDW1"), aml_int(0x10), NULL), - aml_name("CDW1"))); + aml_append(ifctx1, aml_or(aml_name("CDW1"), aml_int(0x10), + aml_name("CDW1"))); aml_append(ifctx, ifctx1); aml_append(ifctx, aml_store(aml_name("CTRL"), aml_name("CDW3"))); @@ -XXX,XX +XXX,XX @@ static void acpi_dsdt_add_pci(Aml *scope, const MemMapEntry *memmap, aml_append(method, ifctx); elsectx = aml_else(); - aml_append(elsectx, aml_store(aml_or(aml_name("CDW1"), aml_int(4), NULL), - aml_name("CDW1"))); + aml_append(elsectx, aml_or(aml_name("CDW1"), aml_int(4), + aml_name("CDW1"))); aml_append(elsectx, aml_return(aml_arg(3))); aml_append(method, elsectx); aml_append(dev, method); diff --git a/tests/data/acpi/virt/DSDT b/tests/data/acpi/virt/DSDT index XXXXXXX..XXXXXXX 100644 GIT binary patch delta 133 zcmZ2BfpOjhMlP3Nmk>D*1_q|2iCof5o%I{lJ2{y;?{412x!p#<jWgaq*qNm(o59&7 z+;D-%<VrV7_iE>mARjJS5V=5L(&S9WT970c2Uv;Nq{%?q7$gZ1761tsfcPNsCD{x4 MAmS{W8QoPG0j8@bzW@LL delta 141 zcmbO?fpOUcMlP3Nmk>1%1_q`n6S<_B8XGpMcXBc{-rKy1bGwazA7{LOuro_nHiNTE zxZwi7$(3%F{sq;}AwfP|vJ4<<fzYJMnT!RsAbBnhh%$*ulYv}gkTg_604z}e5&_99 R$zCV`m0@An{L@X95dZ+BD!u>! diff --git a/tests/data/acpi/virt/DSDT.memhp b/tests/data/acpi/virt/DSDT.memhp index XXXXXXX..XXXXXXX 100644 GIT binary patch delta 132 zcmcaVi}Cs_MlP3NmymE@1_ma@iCof*O&is^IGH-{Zr;SX-A2HTGu}VgnWZb6!PzC; zaDm6<N;gaQYUhw3A1+xCxj<mj<V?m|kR%reSc%xA$w1l|Bnc4~00|d>_#p8m*$ep~ L;w+mP-Q(B*s{AMU delta 140 zcmcaUi}C&}MlP3Nmymd01_maViCof*T^rT9IGGynZQjJW-A2HVGu}VgnWZb6!PzC; zaDm_CN;gaYf@<fGARjJS1`xGCXwu|N#)4XqJQoK<nZ%^YK&~-J8Y&?GmM8#;fMk|r QFBE{vurO@?=@!QZ00dYn_y7O^ diff --git a/tests/data/acpi/virt/DSDT.numamem b/tests/data/acpi/virt/DSDT.numamem index XXXXXXX..XXXXXXX 100644 GIT binary patch delta 133 zcmZ2BfpOjhMlP3Nmk>D*1_q|2iCof5o%I{lJ2{y;?{412x!p#<jWgaq*qNm(o59&7 z+;D-%<VrV7_iE>mARjJS5V=5L(&S9WT970c2Uv;Nq{%?q7$gZ1761tsfcPNsCD{x4 MAmS{W8QoPG0j8@bzW@LL delta 141 zcmbO?fpOUcMlP3Nmk>1%1_q`n6S<_B8XGpMcXBc{-rKy1bGwazA7{LOuro_nHiNTE zxZwi7$(3%F{sq;}AwfP|vJ4<<fzYJMnT!RsAbBnhh%$*ulYv}gkTg_604z}e5&_99 R$zCV`m0@An{L@X95dZ+BD!u>! -- 2.20.1 From: Heyi Guo <guoheyi@huawei.com> After the introduction of generic PCIe root port and PCIe-PCI bridge, we will also have SHPC controller on ARM, so just enable SHPC native hot plug. Also update tests/data/acpi/virt/DSDT* to pass "make check". Cc: Shannon Zhao <shannon.zhaosl@gmail.com> Cc: Peter Maydell <peter.maydell@linaro.org> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Igor Mammedov <imammedo@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Heyi Guo <guoheyi@huawei.com> Message-id: 20191209063719.23086-3-guoheyi@huawei.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/virt-acpi-build.c | 7 ++++++- tests/data/acpi/virt/DSDT | Bin 18462 -> 18462 bytes tests/data/acpi/virt/DSDT.memhp | Bin 19799 -> 19799 bytes tests/data/acpi/virt/DSDT.numamem | Bin 18462 -> 18462 bytes 4 files changed, 6 insertions(+), 1 deletion(-) diff --git a/hw/arm/virt-acpi-build.c b/hw/arm/virt-acpi-build.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt-acpi-build.c +++ b/hw/arm/virt-acpi-build.c @@ -XXX,XX +XXX,XX @@ static void acpi_dsdt_add_pci(Aml *scope, const MemMapEntry *memmap, aml_create_dword_field(aml_arg(3), aml_int(8), "CDW3")); aml_append(ifctx, aml_store(aml_name("CDW2"), aml_name("SUPP"))); aml_append(ifctx, aml_store(aml_name("CDW3"), aml_name("CTRL"))); - aml_append(ifctx, aml_and(aml_name("CTRL"), aml_int(0x1D), + + /* + * Allow OS control for all 5 features: + * PCIeHotplug SHPCHotplug PME AER PCIeCapability. + */ + aml_append(ifctx, aml_and(aml_name("CTRL"), aml_int(0x1F), aml_name("CTRL"))); ifctx1 = aml_if(aml_lnot(aml_equal(aml_arg(1), aml_int(0x1)))); diff --git a/tests/data/acpi/virt/DSDT b/tests/data/acpi/virt/DSDT index XXXXXXX..XXXXXXX 100644 GIT binary patch delta 28 kcmbO?fpOjhMlP3Nmk>D*1_q{tja=*8809zbbW3Ff0C~9xM*si- delta 28 kcmbO?fpOjhMlP3Nmk>D*1_q|2ja=*87-cu_bW3Ff0C~j-M*si- diff --git a/tests/data/acpi/virt/DSDT.memhp b/tests/data/acpi/virt/DSDT.memhp index XXXXXXX..XXXXXXX 100644 GIT binary patch delta 28 kcmcaUi}Cs_MlP3NmymE@1_mbija=*8809zbbeqQp0Eq|*2mk;8 delta 28 kcmcaUi}Cs_MlP3NmymE@1_ma@ja=*87-cu_beqQp0ErX{2mk;8 diff --git a/tests/data/acpi/virt/DSDT.numamem b/tests/data/acpi/virt/DSDT.numamem index XXXXXXX..XXXXXXX 100644 GIT binary patch delta 28 kcmbO?fpOjhMlP3Nmk>D*1_q{tja=*8809zbbW3Ff0C~9xM*si- delta 28 kcmbO?fpOjhMlP3Nmk>D*1_q|2ja=*87-cu_bW3Ff0C~j-M*si- -- 2.20.1 From: Philippe Mathieu-Daudé <philmd@redhat.com> Make the gic a field in the machine state, and instead of filling an array of qemu_irq and passing it around, directly call qdev_get_gpio_in() on the gic field. Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Luc Michel <luc.michel@greensocs.com> Message-id: 20191209090306.20433-1-philmd@redhat.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/virt.h | 1 + hw/arm/virt.c | 109 +++++++++++++++++++++--------------------- 2 files changed, 55 insertions(+), 55 deletions(-) diff --git a/include/hw/arm/virt.h b/include/hw/arm/virt.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/virt.h +++ b/include/hw/arm/virt.h @@ -XXX,XX +XXX,XX @@ typedef struct { uint32_t iommu_phandle; int psci_conduit; hwaddr highest_gpa; + DeviceState *gic; DeviceState *acpi_dev; Notifier powerdown_notifier; } VirtMachineState; diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ static void fdt_add_pmu_nodes(const VirtMachineState *vms) } } -static inline DeviceState *create_acpi_ged(VirtMachineState *vms, qemu_irq *pic) +static inline DeviceState *create_acpi_ged(VirtMachineState *vms) { DeviceState *dev; MachineState *ms = MACHINE(vms); @@ -XXX,XX +XXX,XX @@ static inline DeviceState *create_acpi_ged(VirtMachineState *vms, qemu_irq *pic) sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, vms->memmap[VIRT_ACPI_GED].base); sysbus_mmio_map(SYS_BUS_DEVICE(dev), 1, vms->memmap[VIRT_PCDIMM_ACPI].base); - sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, pic[irq]); + sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, qdev_get_gpio_in(vms->gic, irq)); qdev_init_nofail(dev); return dev; } -static void create_its(VirtMachineState *vms, DeviceState *gicdev) +static void create_its(VirtMachineState *vms) { const char *itsclass = its_class_name(); DeviceState *dev; @@ -XXX,XX +XXX,XX @@ static void create_its(VirtMachineState *vms, DeviceState *gicdev) dev = qdev_create(NULL, itsclass); - object_property_set_link(OBJECT(dev), OBJECT(gicdev), "parent-gicv3", + object_property_set_link(OBJECT(dev), OBJECT(vms->gic), "parent-gicv3", &error_abort); qdev_init_nofail(dev); sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, vms->memmap[VIRT_GIC_ITS].base); @@ -XXX,XX +XXX,XX @@ static void create_its(VirtMachineState *vms, DeviceState *gicdev) fdt_add_its_gic_node(vms); } -static void create_v2m(VirtMachineState *vms, qemu_irq *pic) +static void create_v2m(VirtMachineState *vms) { int i; int irq = vms->irqmap[VIRT_GIC_V2M]; @@ -XXX,XX +XXX,XX @@ static void create_v2m(VirtMachineState *vms, qemu_irq *pic) qdev_init_nofail(dev); for (i = 0; i < NUM_GICV2M_SPIS; i++) { - sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, pic[irq + i]); + sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, + qdev_get_gpio_in(vms->gic, irq + i)); } fdt_add_v2m_gic_node(vms); } -static void create_gic(VirtMachineState *vms, qemu_irq *pic) +static void create_gic(VirtMachineState *vms) { MachineState *ms = MACHINE(vms); /* We create a standalone GIC */ - DeviceState *gicdev; SysBusDevice *gicbusdev; const char *gictype; int type = vms->gic_version, i; @@ -XXX,XX +XXX,XX @@ static void create_gic(VirtMachineState *vms, qemu_irq *pic) gictype = (type == 3) ? gicv3_class_name() : gic_class_name(); - gicdev = qdev_create(NULL, gictype); - qdev_prop_set_uint32(gicdev, "revision", type); - qdev_prop_set_uint32(gicdev, "num-cpu", smp_cpus); + vms->gic = qdev_create(NULL, gictype); + qdev_prop_set_uint32(vms->gic, "revision", type); + qdev_prop_set_uint32(vms->gic, "num-cpu", smp_cpus); /* Note that the num-irq property counts both internal and external * interrupts; there are always 32 of the former (mandated by GIC spec). */ - qdev_prop_set_uint32(gicdev, "num-irq", NUM_IRQS + 32); + qdev_prop_set_uint32(vms->gic, "num-irq", NUM_IRQS + 32); if (!kvm_irqchip_in_kernel()) { - qdev_prop_set_bit(gicdev, "has-security-extensions", vms->secure); + qdev_prop_set_bit(vms->gic, "has-security-extensions", vms->secure); } if (type == 3) { @@ -XXX,XX +XXX,XX @@ static void create_gic(VirtMachineState *vms, qemu_irq *pic) nb_redist_regions = virt_gicv3_redist_region_count(vms); - qdev_prop_set_uint32(gicdev, "len-redist-region-count", + qdev_prop_set_uint32(vms->gic, "len-redist-region-count", nb_redist_regions); - qdev_prop_set_uint32(gicdev, "redist-region-count[0]", redist0_count); + qdev_prop_set_uint32(vms->gic, "redist-region-count[0]", redist0_count); if (nb_redist_regions == 2) { uint32_t redist1_capacity = vms->memmap[VIRT_HIGH_GIC_REDIST2].size / GICV3_REDIST_SIZE; - qdev_prop_set_uint32(gicdev, "redist-region-count[1]", + qdev_prop_set_uint32(vms->gic, "redist-region-count[1]", MIN(smp_cpus - redist0_count, redist1_capacity)); } } else { if (!kvm_irqchip_in_kernel()) { - qdev_prop_set_bit(gicdev, "has-virtualization-extensions", + qdev_prop_set_bit(vms->gic, "has-virtualization-extensions", vms->virt); } } - qdev_init_nofail(gicdev); - gicbusdev = SYS_BUS_DEVICE(gicdev); + qdev_init_nofail(vms->gic); + gicbusdev = SYS_BUS_DEVICE(vms->gic); sysbus_mmio_map(gicbusdev, 0, vms->memmap[VIRT_GIC_DIST].base); if (type == 3) { sysbus_mmio_map(gicbusdev, 1, vms->memmap[VIRT_GIC_REDIST].base); @@ -XXX,XX +XXX,XX @@ static void create_gic(VirtMachineState *vms, qemu_irq *pic) for (irq = 0; irq < ARRAY_SIZE(timer_irq); irq++) { qdev_connect_gpio_out(cpudev, irq, - qdev_get_gpio_in(gicdev, + qdev_get_gpio_in(vms->gic, ppibase + timer_irq[irq])); } if (type == 3) { - qemu_irq irq = qdev_get_gpio_in(gicdev, + qemu_irq irq = qdev_get_gpio_in(vms->gic, ppibase + ARCH_GIC_MAINT_IRQ); qdev_connect_gpio_out_named(cpudev, "gicv3-maintenance-interrupt", 0, irq); } else if (vms->virt) { - qemu_irq irq = qdev_get_gpio_in(gicdev, + qemu_irq irq = qdev_get_gpio_in(vms->gic, ppibase + ARCH_GIC_MAINT_IRQ); sysbus_connect_irq(gicbusdev, i + 4 * smp_cpus, irq); } qdev_connect_gpio_out_named(cpudev, "pmu-interrupt", 0, - qdev_get_gpio_in(gicdev, ppibase + qdev_get_gpio_in(vms->gic, ppibase + VIRTUAL_PMU_IRQ)); sysbus_connect_irq(gicbusdev, i, qdev_get_gpio_in(cpudev, ARM_CPU_IRQ)); @@ -XXX,XX +XXX,XX @@ static void create_gic(VirtMachineState *vms, qemu_irq *pic) qdev_get_gpio_in(cpudev, ARM_CPU_VFIQ)); } - for (i = 0; i < NUM_IRQS; i++) { - pic[i] = qdev_get_gpio_in(gicdev, i); - } - fdt_add_gic_node(vms); if (type == 3 && vms->its) { - create_its(vms, gicdev); + create_its(vms); } else if (type == 2) { - create_v2m(vms, pic); + create_v2m(vms); } } -static void create_uart(const VirtMachineState *vms, qemu_irq *pic, int uart, +static void create_uart(const VirtMachineState *vms, int uart, MemoryRegion *mem, Chardev *chr) { char *nodename; @@ -XXX,XX +XXX,XX @@ static void create_uart(const VirtMachineState *vms, qemu_irq *pic, int uart, qdev_init_nofail(dev); memory_region_add_subregion(mem, base, sysbus_mmio_get_region(s, 0)); - sysbus_connect_irq(s, 0, pic[irq]); + sysbus_connect_irq(s, 0, qdev_get_gpio_in(vms->gic, irq)); nodename = g_strdup_printf("/pl011@%" PRIx64, base); qemu_fdt_add_subnode(vms->fdt, nodename); @@ -XXX,XX +XXX,XX @@ static void create_uart(const VirtMachineState *vms, qemu_irq *pic, int uart, g_free(nodename); } -static void create_rtc(const VirtMachineState *vms, qemu_irq *pic) +static void create_rtc(const VirtMachineState *vms) { char *nodename; hwaddr base = vms->memmap[VIRT_RTC].base; @@ -XXX,XX +XXX,XX @@ static void create_rtc(const VirtMachineState *vms, qemu_irq *pic) int irq = vms->irqmap[VIRT_RTC]; const char compat[] = "arm,pl031\0arm,primecell"; - sysbus_create_simple("pl031", base, pic[irq]); + sysbus_create_simple("pl031", base, qdev_get_gpio_in(vms->gic, irq)); nodename = g_strdup_printf("/pl031@%" PRIx64, base); qemu_fdt_add_subnode(vms->fdt, nodename); @@ -XXX,XX +XXX,XX @@ static void virt_powerdown_req(Notifier *n, void *opaque) } } -static void create_gpio(const VirtMachineState *vms, qemu_irq *pic) +static void create_gpio(const VirtMachineState *vms) { char *nodename; DeviceState *pl061_dev; @@ -XXX,XX +XXX,XX @@ static void create_gpio(const VirtMachineState *vms, qemu_irq *pic) int irq = vms->irqmap[VIRT_GPIO]; const char compat[] = "arm,pl061\0arm,primecell"; - pl061_dev = sysbus_create_simple("pl061", base, pic[irq]); + pl061_dev = sysbus_create_simple("pl061", base, + qdev_get_gpio_in(vms->gic, irq)); uint32_t phandle = qemu_fdt_alloc_phandle(vms->fdt); nodename = g_strdup_printf("/pl061@%" PRIx64, base); @@ -XXX,XX +XXX,XX @@ static void create_gpio(const VirtMachineState *vms, qemu_irq *pic) g_free(nodename); } -static void create_virtio_devices(const VirtMachineState *vms, qemu_irq *pic) +static void create_virtio_devices(const VirtMachineState *vms) { int i; hwaddr size = vms->memmap[VIRT_MMIO].size; @@ -XXX,XX +XXX,XX @@ static void create_virtio_devices(const VirtMachineState *vms, qemu_irq *pic) int irq = vms->irqmap[VIRT_MMIO] + i; hwaddr base = vms->memmap[VIRT_MMIO].base + i * size; - sysbus_create_simple("virtio-mmio", base, pic[irq]); + sysbus_create_simple("virtio-mmio", base, + qdev_get_gpio_in(vms->gic, irq)); } /* We add dtb nodes in reverse order so that they appear in the finished @@ -XXX,XX +XXX,XX @@ static void create_pcie_irq_map(const VirtMachineState *vms, 0x7 /* PCI irq */); } -static void create_smmu(const VirtMachineState *vms, qemu_irq *pic, +static void create_smmu(const VirtMachineState *vms, PCIBus *bus) { char *node; @@ -XXX,XX +XXX,XX @@ static void create_smmu(const VirtMachineState *vms, qemu_irq *pic, qdev_init_nofail(dev); sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, base); for (i = 0; i < NUM_SMMU_IRQS; i++) { - sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, pic[irq + i]); + sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, + qdev_get_gpio_in(vms->gic, irq + i)); } node = g_strdup_printf("/smmuv3@%" PRIx64, base); @@ -XXX,XX +XXX,XX @@ static void create_smmu(const VirtMachineState *vms, qemu_irq *pic, g_free(node); } -static void create_pcie(VirtMachineState *vms, qemu_irq *pic) +static void create_pcie(VirtMachineState *vms) { hwaddr base_mmio = vms->memmap[VIRT_PCIE_MMIO].base; hwaddr size_mmio = vms->memmap[VIRT_PCIE_MMIO].size; @@ -XXX,XX +XXX,XX @@ static void create_pcie(VirtMachineState *vms, qemu_irq *pic) sysbus_mmio_map(SYS_BUS_DEVICE(dev), 2, base_pio); for (i = 0; i < GPEX_NUM_IRQS; i++) { - sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, pic[irq + i]); + sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, + qdev_get_gpio_in(vms->gic, irq + i)); gpex_set_irq_num(GPEX_HOST(dev), i, irq + i); } @@ -XXX,XX +XXX,XX @@ static void create_pcie(VirtMachineState *vms, qemu_irq *pic) if (vms->iommu) { vms->iommu_phandle = qemu_fdt_alloc_phandle(vms->fdt); - create_smmu(vms, pic, pci->bus); + create_smmu(vms, pci->bus); qemu_fdt_setprop_cells(vms->fdt, nodename, "iommu-map", 0x0, vms->iommu_phandle, 0x0, 0x10000); @@ -XXX,XX +XXX,XX @@ static void create_pcie(VirtMachineState *vms, qemu_irq *pic) g_free(nodename); } -static void create_platform_bus(VirtMachineState *vms, qemu_irq *pic) +static void create_platform_bus(VirtMachineState *vms) { DeviceState *dev; SysBusDevice *s; @@ -XXX,XX +XXX,XX @@ static void create_platform_bus(VirtMachineState *vms, qemu_irq *pic) s = SYS_BUS_DEVICE(dev); for (i = 0; i < PLATFORM_BUS_NUM_IRQS; i++) { - int irqn = vms->irqmap[VIRT_PLATFORM_BUS] + i; - sysbus_connect_irq(s, i, pic[irqn]); + int irq = vms->irqmap[VIRT_PLATFORM_BUS] + i; + sysbus_connect_irq(s, i, qdev_get_gpio_in(vms->gic, irq)); } memory_region_add_subregion(sysmem, @@ -XXX,XX +XXX,XX @@ static void machvirt_init(MachineState *machine) VirtMachineClass *vmc = VIRT_MACHINE_GET_CLASS(machine); MachineClass *mc = MACHINE_GET_CLASS(machine); const CPUArchIdList *possible_cpus; - qemu_irq pic[NUM_IRQS]; MemoryRegion *sysmem = get_system_memory(); MemoryRegion *secure_sysmem = NULL; int n, virt_max_cpus; @@ -XXX,XX +XXX,XX @@ static void machvirt_init(MachineState *machine) virt_flash_fdt(vms, sysmem, secure_sysmem ?: sysmem); - create_gic(vms, pic); + create_gic(vms); fdt_add_pmu_nodes(vms); - create_uart(vms, pic, VIRT_UART, sysmem, serial_hd(0)); + create_uart(vms, VIRT_UART, sysmem, serial_hd(0)); if (vms->secure) { create_secure_ram(vms, secure_sysmem); - create_uart(vms, pic, VIRT_SECURE_UART, secure_sysmem, serial_hd(1)); + create_uart(vms, VIRT_SECURE_UART, secure_sysmem, serial_hd(1)); } vms->highmem_ecam &= vms->highmem && (!firmware_loaded || aarch64); - create_rtc(vms, pic); + create_rtc(vms); - create_pcie(vms, pic); + create_pcie(vms); if (has_ged && aarch64 && firmware_loaded && acpi_enabled) { - vms->acpi_dev = create_acpi_ged(vms, pic); + vms->acpi_dev = create_acpi_ged(vms); } else { - create_gpio(vms, pic); + create_gpio(vms); } /* connect powerdown request */ @@ -XXX,XX +XXX,XX @@ static void machvirt_init(MachineState *machine) * (which will be automatically plugged in to the transports). If * no backend is created the transport will just sit harmlessly idle. */ - create_virtio_devices(vms, pic); + create_virtio_devices(vms); vms->fw_cfg = create_fw_cfg(vms, &address_space_memory); rom_set_fw(vms->fw_cfg); - create_platform_bus(vms, pic); + create_platform_bus(vms); vms->bootinfo.ram_size = machine->ram_size; vms->bootinfo.nb_cpus = smp_cpus; -- 2.20.1 From: Alex Bennée <alex.bennee@linaro.org> A write to the SCR can change the effective EL by droppping the system from secure to non-secure mode. However if we use a cached current_el from before the change we'll rebuild the flags incorrectly. To fix this we introduce the ARM_CP_NEWEL CP flag to indicate the new EL should be used when recomputing the flags. Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20191212114734.6962-1-alex.bennee@linaro.org Cc: Richard Henderson <richard.henderson@linaro.org> Message-Id: <20191209143723.6368-1-alex.bennee@linaro.org> Cc: qemu-stable@nongnu.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.h | 8 ++++++-- target/arm/helper.h | 1 + target/arm/helper.c | 14 +++++++++++++- target/arm/translate.c | 6 +++++- 4 files changed, 25 insertions(+), 4 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 uint64_t cpreg_to_kvm_id(uint32_t cpregid) * 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 @@ -XXX,XX +XXX,XX @@ static inline uint64_t cpreg_to_kvm_id(uint32_t cpregid) #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 0xffff +#define ARM_CP_SENTINEL 0xfffff /* Mask of only the flag bits in a type field */ -#define ARM_CP_FLAG_MASK 0xf0ff +#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. diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_2(get_user_reg, i32, env, i32) DEF_HELPER_3(set_user_reg, void, env, i32, i32) DEF_HELPER_FLAGS_2(rebuild_hflags_m32, TCG_CALL_NO_RWG, void, env, int) +DEF_HELPER_FLAGS_1(rebuild_hflags_a32_newel, TCG_CALL_NO_RWG, void, env) DEF_HELPER_FLAGS_2(rebuild_hflags_a32, TCG_CALL_NO_RWG, void, env, int) DEF_HELPER_FLAGS_2(rebuild_hflags_a64, TCG_CALL_NO_RWG, void, env, int) 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 el3_cp_reginfo[] = { .opc0 = 3, .opc1 = 6, .crn = 1, .crm = 1, .opc2 = 0, .access = PL3_RW, .fieldoffset = offsetof(CPUARMState, cp15.scr_el3), .resetvalue = 0, .writefn = scr_write }, - { .name = "SCR", .type = ARM_CP_ALIAS, + { .name = "SCR", .type = ARM_CP_ALIAS | ARM_CP_NEWEL, .cp = 15, .opc1 = 0, .crn = 1, .crm = 1, .opc2 = 0, .access = PL1_RW, .accessfn = access_trap_aa32s_el1, .fieldoffset = offsetoflow32(CPUARMState, cp15.scr_el3), @@ -XXX,XX +XXX,XX @@ void HELPER(rebuild_hflags_m32)(CPUARMState *env, int el) env->hflags = rebuild_hflags_m32(env, fp_el, mmu_idx); } +/* + * If we have triggered a EL state change we can't rely on the + * translator having passed it too us, we need to recompute. + */ +void HELPER(rebuild_hflags_a32_newel)(CPUARMState *env) +{ + int el = arm_current_el(env); + int fp_el = fp_exception_el(env, el); + ARMMMUIdx mmu_idx = arm_mmu_idx_el(env, el); + env->hflags = rebuild_hflags_a32(env, fp_el, mmu_idx); +} + void HELPER(rebuild_hflags_a32)(CPUARMState *env, int el) { int fp_el = fp_exception_el(env, el); 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 int disas_coproc_insn(DisasContext *s, uint32_t insn) if (arm_dc_feature(s, ARM_FEATURE_M)) { gen_helper_rebuild_hflags_m32(cpu_env, tcg_el); } else { - gen_helper_rebuild_hflags_a32(cpu_env, tcg_el); + if (ri->type & ARM_CP_NEWEL) { + gen_helper_rebuild_hflags_a32_newel(cpu_env); + } else { + gen_helper_rebuild_hflags_a32(cpu_env, tcg_el); + } } tcg_temp_free_i32(tcg_el); /* -- 2.20.1 target-arm queue: the big stuff here is the final part of rth's patches for Cortex-A76 and Neoverse-N1 support; also present are Gavin's NUMA series and a few other things. thanks -- PMM The following changes since commit 554623226f800acf48a2ed568900c1c968ec9a8b: Merge tag 'qemu-sparc-20220508' of https://github.com/mcayland/qemu into staging (2022-05-08 17:03:26 -0500) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20220509 for you to fetch changes up to ae9141d4a3265553503bf07d3574b40f84615a34: hw/acpi/aml-build: Use existing CPU topology to build PPTT table (2022-05-09 11:47:55 +0100) ---------------------------------------------------------------- target-arm queue: * MAINTAINERS/.mailmap: update email for Leif Lindholm * hw/arm: add version information to sbsa-ref machine DT * Enable new features for -cpu max: FEAT_Debugv8p2, FEAT_Debugv8p4, FEAT_RAS (minimal version only), FEAT_IESB, FEAT_CSV2, FEAT_CSV2_2, FEAT_CSV3, FEAT_DGH * Emulate Cortex-A76 * Emulate Neoverse-N1 * Fix the virt board default NUMA topology ---------------------------------------------------------------- Gavin Shan (6): qapi/machine.json: Add cluster-id qtest/numa-test: Specify CPU topology in aarch64_numa_cpu() hw/arm/virt: Consider SMP configuration in CPU topology qtest/numa-test: Correct CPU and NUMA association in aarch64_numa_cpu() hw/arm/virt: Fix CPU's default NUMA node ID hw/acpi/aml-build: Use existing CPU topology to build PPTT table Leif Lindholm (2): MAINTAINERS/.mailmap: update email for Leif Lindholm hw/arm: add versioning to sbsa-ref machine DT Richard Henderson (24): target/arm: Handle cpreg registration for missing EL target/arm: Drop EL3 no EL2 fallbacks target/arm: Merge zcr reginfo target/arm: Adjust definition of CONTEXTIDR_EL2 target/arm: Move cortex impdef sysregs to cpu_tcg.c target/arm: Update qemu-system-arm -cpu max to cortex-a57 target/arm: Set ID_DFR0.PerfMon for qemu-system-arm -cpu max target/arm: Split out aa32_max_features target/arm: Annotate arm_max_initfn with FEAT identifiers target/arm: Use field names for manipulating EL2 and EL3 modes target/arm: Enable FEAT_Debugv8p2 for -cpu max target/arm: Enable FEAT_Debugv8p4 for -cpu max target/arm: Add minimal RAS registers target/arm: Enable SCR and HCR bits for RAS target/arm: Implement virtual SError exceptions target/arm: Implement ESB instruction target/arm: Enable FEAT_RAS for -cpu max target/arm: Enable FEAT_IESB for -cpu max target/arm: Enable FEAT_CSV2 for -cpu max target/arm: Enable FEAT_CSV2_2 for -cpu max target/arm: Enable FEAT_CSV3 for -cpu max target/arm: Enable FEAT_DGH for -cpu max target/arm: Define cortex-a76 target/arm: Define neoverse-n1 docs/system/arm/emulation.rst | 10 + docs/system/arm/virt.rst | 2 + qapi/machine.json | 6 +- target/arm/cpregs.h | 11 + target/arm/cpu.h | 23 ++ target/arm/helper.h | 1 + target/arm/internals.h | 16 ++ target/arm/syndrome.h | 5 + target/arm/a32.decode | 16 +- target/arm/t32.decode | 18 +- hw/acpi/aml-build.c | 111 ++++---- hw/arm/sbsa-ref.c | 16 ++ hw/arm/virt.c | 21 +- hw/core/machine-hmp-cmds.c | 4 + hw/core/machine.c | 16 ++ target/arm/cpu.c | 66 ++++- target/arm/cpu64.c | 353 ++++++++++++++----------- target/arm/cpu_tcg.c | 227 +++++++++++----- target/arm/helper.c | 600 +++++++++++++++++++++++++----------------- target/arm/op_helper.c | 43 +++ target/arm/translate-a64.c | 18 ++ target/arm/translate.c | 23 ++ tests/qtest/numa-test.c | 19 +- .mailmap | 3 +- MAINTAINERS | 2 +- 25 files changed, 1068 insertions(+), 562 deletions(-) From: Leif Lindholm <quic_llindhol@quicinc.com> NUVIA was acquired by Qualcomm in March 2021, but kept functioning on separate infrastructure for a transitional period. We've now switched over to contributing as Qualcomm Innovation Center (quicinc), so update my email address to reflect this. Signed-off-by: Leif Lindholm <quic_llindhol@quicinc.com> Message-id: 20220505113740.75565-1-quic_llindhol@quicinc.com Cc: Leif Lindholm <leif@nuviainc.com> Cc: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> [Fixed commit message typo] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- .mailmap | 3 ++- MAINTAINERS | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/.mailmap b/.mailmap index XXXXXXX..XXXXXXX 100644 --- a/.mailmap +++ b/.mailmap @@ -XXX,XX +XXX,XX @@ Greg Kurz <groug@kaod.org> <gkurz@linux.vnet.ibm.com> Huacai Chen <chenhuacai@kernel.org> <chenhc@lemote.com> Huacai Chen <chenhuacai@kernel.org> <chenhuacai@loongson.cn> James Hogan <jhogan@kernel.org> <james.hogan@imgtec.com> -Leif Lindholm <leif@nuviainc.com> <leif.lindholm@linaro.org> +Leif Lindholm <quic_llindhol@quicinc.com> <leif.lindholm@linaro.org> +Leif Lindholm <quic_llindhol@quicinc.com> <leif@nuviainc.com> Radoslaw Biernacki <rad@semihalf.com> <radoslaw.biernacki@linaro.org> Paul Burton <paulburton@kernel.org> <paul.burton@mips.com> Paul Burton <paulburton@kernel.org> <paul.burton@imgtec.com> diff --git a/MAINTAINERS b/MAINTAINERS index XXXXXXX..XXXXXXX 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -XXX,XX +XXX,XX @@ F: include/hw/ssi/imx_spi.h SBSA-REF M: Radoslaw Biernacki <rad@semihalf.com> M: Peter Maydell <peter.maydell@linaro.org> -R: Leif Lindholm <leif@nuviainc.com> +R: Leif Lindholm <quic_llindhol@quicinc.com> L: qemu-arm@nongnu.org S: Maintained F: hw/arm/sbsa-ref.c -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> More gracefully handle cpregs when EL2 and/or EL3 are missing. If the reg is entirely inaccessible, do not register it at all. If the reg is for EL2, and EL3 is present but EL2 is not, either discard, squash to res0, const, or keep unchanged. Per rule RJFFP, mark the 4 aarch32 hypervisor access registers with ARM_CP_EL3_NO_EL2_KEEP, and mark all of the EL2 address translation and tlb invalidation "regs" ARM_CP_EL3_NO_EL2_UNDEF. Mark the 2 virtualization processor id regs ARM_CP_EL3_NO_EL2_C_NZ. This will simplify cpreg registration for conditional arm features. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-2-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpregs.h | 11 +++ target/arm/helper.c | 178 ++++++++++++++++++++++++++++++-------------- 2 files changed, 133 insertions(+), 56 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 { ARM_CP_SVE = 1 << 14, /* Flag: Do not expose in gdb sysreg xml. */ ARM_CP_NO_GDB = 1 << 15, + /* + * Flags: If EL3 but not EL2... + * - UNDEF: discard the cpreg, + * - KEEP: retain the cpreg as is, + * - C_NZ: set const on the cpreg, but retain resetvalue, + * - else: set const on the cpreg, zero resetvalue, aka RES0. + * See rule RJFFP in section D1.1.3 of DDI0487H.a. + */ + ARM_CP_EL3_NO_EL2_UNDEF = 1 << 16, + ARM_CP_EL3_NO_EL2_KEEP = 1 << 17, + ARM_CP_EL3_NO_EL2_C_NZ = 1 << 18, }; /* 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 v8_cp_reginfo[] = { .access = PL1_RW, .readfn = spsel_read, .writefn = spsel_write }, { .name = "FPEXC32_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 5, .crm = 3, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_ALIAS | ARM_CP_FPU, + .access = PL2_RW, + .type = ARM_CP_ALIAS | ARM_CP_FPU | ARM_CP_EL3_NO_EL2_KEEP, .fieldoffset = offsetof(CPUARMState, vfp.xregs[ARM_VFP_FPEXC]) }, { .name = "DACR32_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 3, .crm = 0, .opc2 = 0, - .access = PL2_RW, .resetvalue = 0, + .access = PL2_RW, .resetvalue = 0, .type = ARM_CP_EL3_NO_EL2_KEEP, .writefn = dacr_write, .raw_writefn = raw_write, .fieldoffset = offsetof(CPUARMState, cp15.dacr32_el2) }, { .name = "IFSR32_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 5, .crm = 0, .opc2 = 1, - .access = PL2_RW, .resetvalue = 0, + .access = PL2_RW, .resetvalue = 0, .type = ARM_CP_EL3_NO_EL2_KEEP, .fieldoffset = offsetof(CPUARMState, cp15.ifsr32_el2) }, { .name = "SPSR_IRQ", .state = ARM_CP_STATE_AA64, .type = ARM_CP_ALIAS, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el2_cp_reginfo[] = { .writefn = tlbimva_hyp_is_write }, { .name = "TLBI_ALLE2", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 7, .opc2 = 0, - .type = ARM_CP_NO_RAW, .access = PL2_W, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_alle2_write }, { .name = "TLBI_VAE2", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 7, .opc2 = 1, - .type = ARM_CP_NO_RAW, .access = PL2_W, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_vae2_write }, { .name = "TLBI_VALE2", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 7, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_vae2_write }, { .name = "TLBI_ALLE2IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 3, .opc2 = 0, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_alle2is_write }, { .name = "TLBI_VAE2IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 3, .opc2 = 1, - .type = ARM_CP_NO_RAW, .access = PL2_W, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_vae2is_write }, { .name = "TLBI_VALE2IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 3, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_vae2is_write }, #ifndef CONFIG_USER_ONLY /* Unlike the other EL2-related AT operations, these must @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo el2_cp_reginfo[] = { { .name = "AT_S1E2R", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 7, .crm = 8, .opc2 = 0, .access = PL2_W, .accessfn = at_s1e2_access, - .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC, .writefn = ats_write64 }, + .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC | ARM_CP_EL3_NO_EL2_UNDEF, + .writefn = ats_write64 }, { .name = "AT_S1E2W", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 7, .crm = 8, .opc2 = 1, .access = PL2_W, .accessfn = at_s1e2_access, - .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC, .writefn = ats_write64 }, + .type = ARM_CP_NO_RAW | ARM_CP_RAISES_EXC | ARM_CP_EL3_NO_EL2_UNDEF, + .writefn = ats_write64 }, /* The AArch32 ATS1H* operations are CONSTRAINED UNPREDICTABLE * if EL2 is not implemented; we choose to UNDEF. Behaviour at EL3 * with SCR.NS == 0 outside Monitor mode is UNPREDICTABLE; we choose @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo debug_cp_reginfo[] = { { .name = "DBGVCR32_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 2, .opc1 = 4, .crn = 0, .crm = 7, .opc2 = 0, .access = PL2_RW, .accessfn = access_tda, - .type = ARM_CP_NOP }, + .type = ARM_CP_NOP | ARM_CP_EL3_NO_EL2_KEEP }, /* Dummy MDCCINT_EL1, since we don't implement the Debug Communications * Channel but Linux may try to access this register. The 32-bit * alias is DBGDCCINT. @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo tlbirange_reginfo[] = { .access = PL2_W, .type = ARM_CP_NOP }, { .name = "TLBI_RVAE2IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 2, .opc2 = 1, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_rvae2is_write }, { .name = "TLBI_RVALE2IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 2, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_rvae2is_write }, { .name = "TLBI_RIPAS2E1", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 4, .opc2 = 2, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo tlbirange_reginfo[] = { .access = PL2_W, .type = ARM_CP_NOP }, { .name = "TLBI_RVAE2OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 5, .opc2 = 1, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_rvae2is_write }, { .name = "TLBI_RVALE2OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 5, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_rvae2is_write }, { .name = "TLBI_RVAE2", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 6, .opc2 = 1, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_rvae2_write }, { .name = "TLBI_RVALE2", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 6, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_rvae2_write }, { .name = "TLBI_RVAE3IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 6, .crn = 8, .crm = 2, .opc2 = 1, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo tlbios_reginfo[] = { .writefn = tlbi_aa64_vae1is_write }, { .name = "TLBI_ALLE2OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 1, .opc2 = 0, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_alle2is_write }, { .name = "TLBI_VAE2OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 1, .opc2 = 1, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_vae2is_write }, { .name = "TLBI_ALLE1OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 1, .opc2 = 4, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo tlbios_reginfo[] = { .writefn = tlbi_aa64_alle1is_write }, { .name = "TLBI_VALE2OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 1, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, + .access = PL2_W, .type = ARM_CP_NO_RAW | ARM_CP_EL3_NO_EL2_UNDEF, .writefn = tlbi_aa64_vae2is_write }, { .name = "TLBI_VMALLS12E1OS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 1, .opc2 = 6, @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) { .name = "VPIDR", .state = ARM_CP_STATE_AA32, .cp = 15, .opc1 = 4, .crn = 0, .crm = 0, .opc2 = 0, .access = PL2_RW, .accessfn = access_el3_aa32ns, - .resetvalue = cpu->midr, .type = ARM_CP_ALIAS, + .resetvalue = cpu->midr, + .type = ARM_CP_ALIAS | ARM_CP_EL3_NO_EL2_C_NZ, .fieldoffset = offsetoflow32(CPUARMState, cp15.vpidr_el2) }, { .name = "VPIDR_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 0, .crm = 0, .opc2 = 0, .access = PL2_RW, .resetvalue = cpu->midr, + .type = ARM_CP_EL3_NO_EL2_C_NZ, .fieldoffset = offsetof(CPUARMState, cp15.vpidr_el2) }, { .name = "VMPIDR", .state = ARM_CP_STATE_AA32, .cp = 15, .opc1 = 4, .crn = 0, .crm = 0, .opc2 = 5, .access = PL2_RW, .accessfn = access_el3_aa32ns, - .resetvalue = vmpidr_def, .type = ARM_CP_ALIAS, + .resetvalue = vmpidr_def, + .type = ARM_CP_ALIAS | ARM_CP_EL3_NO_EL2_C_NZ, .fieldoffset = offsetoflow32(CPUARMState, cp15.vmpidr_el2) }, { .name = "VMPIDR_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 0, .crm = 0, .opc2 = 5, - .access = PL2_RW, - .resetvalue = vmpidr_def, + .access = PL2_RW, .resetvalue = vmpidr_def, + .type = ARM_CP_EL3_NO_EL2_C_NZ, .fieldoffset = offsetof(CPUARMState, cp15.vmpidr_el2) }, }; define_arm_cp_regs(cpu, vpidr_regs); @@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r, int crm, int opc1, int opc2, const char *name) { + CPUARMState *env = &cpu->env; uint32_t key; ARMCPRegInfo *r2; bool is64 = r->type & ARM_CP_64BIT; bool ns = secstate & ARM_CP_SECSTATE_NS; int cp = r->cp; - bool isbanked; size_t name_len; + bool make_const; switch (state) { case ARM_CP_STATE_AA32: @@ -XXX,XX +XXX,XX @@ static void add_cpreg_to_hashtable(ARMCPU *cpu, const ARMCPRegInfo *r, } } + /* + * Eliminate registers that are not present because the EL is missing. + * Doing this here makes it easier to put all registers for a given + * feature into the same ARMCPRegInfo array and define them all at once. + */ + make_const = false; + if (arm_feature(env, ARM_FEATURE_EL3)) { + /* + * An EL2 register without EL2 but with EL3 is (usually) RES0. + * See rule RJFFP in section D1.1.3 of DDI0487H.a. + */ + int min_el = ctz32(r->access) / 2; + if (min_el == 2 && !arm_feature(env, ARM_FEATURE_EL2)) { + if (r->type & ARM_CP_EL3_NO_EL2_UNDEF) { + return; + } + make_const = !(r->type & ARM_CP_EL3_NO_EL2_KEEP); + } + } else { + CPAccessRights max_el = (arm_feature(env, ARM_FEATURE_EL2) + ? PL2_RW : PL1_RW); + if ((r->access & max_el) == 0) { + return; + } + } + /* 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, r2->opaque = opaque; } - isbanked = r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1]; - if (isbanked) { + if (make_const) { + /* This should not have been a very special register to begin. */ + int old_special = r2->type & ARM_CP_SPECIAL_MASK; + assert(old_special == 0 || old_special == ARM_CP_NOP); /* - * 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. + * Set the special function to CONST, retaining the other flags. + * This is important for e.g. ARM_CP_SVE so that we still + * take the SVE trap if CPTR_EL3.EZ == 0. */ - r2->fieldoffset = r->bank_fieldoffsets[ns]; - } + r2->type = (r2->type & ~ARM_CP_SPECIAL_MASK) | ARM_CP_CONST; + /* + * Usually, these registers become RES0, but there are a few + * special cases like VPIDR_EL2 which have a constant non-zero + * value with writes ignored. + */ + if (!(r->type & ARM_CP_EL3_NO_EL2_C_NZ)) { + r2->resetvalue = 0; + } + /* + * ARM_CP_CONST has precedence, so removing the callbacks and + * offsets are not strictly necessary, but it is potentially + * less confusing to debug later. + */ + r2->readfn = NULL; + r2->writefn = NULL; + r2->raw_readfn = NULL; + r2->raw_writefn = NULL; + r2->resetfn = NULL; + r2->fieldoffset = 0; + r2->bank_fieldoffsets[0] = 0; + r2->bank_fieldoffsets[1] = 0; + } else { + bool isbanked = r->bank_fieldoffsets[0] && r->bank_fieldoffsets[1]; - if (state == ARM_CP_STATE_AA32) { if (isbanked) { /* - * 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 - * can count on the 64-bit instance taking care of the - * non-secure bank. - * 2) If ARMv8 is enabled then we can count on a 64-bit version - * taking care of the secure bank. This requires that separate - * 32 and 64-bit definitions are provided. + * 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. */ - if ((r->state == ARM_CP_STATE_BOTH && ns) || - (arm_feature(&cpu->env, ARM_FEATURE_V8) && !ns)) { + r2->fieldoffset = r->bank_fieldoffsets[ns]; + } + if (state == ARM_CP_STATE_AA32) { + if (isbanked) { + /* + * 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 can count on the 64-bit instance taking care + * of the non-secure bank. + * 2) If ARMv8 is enabled then we can count on a 64-bit + * version taking care of the secure bank. This requires + * that separate 32 and 64-bit definitions are provided. + */ + if ((r->state == ARM_CP_STATE_BOTH && ns) || + (arm_feature(env, ARM_FEATURE_V8) && !ns)) { + 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. + */ 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. - */ - r2->type |= ARM_CP_ALIAS; - } - if (HOST_BIG_ENDIAN && - r->state == ARM_CP_STATE_BOTH && r2->fieldoffset) { - r2->fieldoffset += sizeof(uint32_t); + if (HOST_BIG_ENDIAN && + r->state == ARM_CP_STATE_BOTH && r2->fieldoffset) { + r2->fieldoffset += sizeof(uint32_t); + } } } @@ -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_MASK) { + if (r2->type & ARM_CP_SPECIAL_MASK) { r2->type |= ARM_CP_NO_RAW; } if (((r->crm == CP_ANY) && crm != 0) || -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Drop el3_no_el2_cp_reginfo, el3_no_el2_v8_cp_reginfo, and the local vpidr_regs definition, and rely on the squashing to ARM_CP_CONST while registering for v8. This is a behavior change for v7 cpus with Security Extensions and without Virtualization Extensions, in that the virtualization cpregs are now correctly not present. This would be a migration compatibility break, except that we have an existing bug in which migration of 32-bit cpus with Security Extensions enabled does not work. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-3-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 158 ++++---------------------------------------- 1 file changed, 13 insertions(+), 145 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 const ARMCPRegInfo v8_cp_reginfo[] = { .fieldoffset = offsetoflow32(CPUARMState, cp15.mdcr_el3) }, }; -/* Used to describe the behaviour of EL2 regs when EL2 does not exist. */ -static const ARMCPRegInfo el3_no_el2_cp_reginfo[] = { - { .name = "VBAR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 12, .crm = 0, .opc2 = 0, - .access = PL2_RW, - .readfn = arm_cp_read_zero, .writefn = arm_cp_write_ignore }, - { .name = "HCR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 0, - .access = PL2_RW, - .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "HACR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 7, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "ESR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 5, .crm = 2, .opc2 = 0, - .access = PL2_RW, - .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CPTR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 2, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "MAIR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 10, .crm = 2, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "HMAIR1", .state = ARM_CP_STATE_AA32, - .cp = 15, .opc1 = 4, .crn = 10, .crm = 2, .opc2 = 1, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "AMAIR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 10, .crm = 3, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "HAMAIR1", .state = ARM_CP_STATE_AA32, - .cp = 15, .opc1 = 4, .crn = 10, .crm = 3, .opc2 = 1, - .access = PL2_RW, .type = ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "AFSR0_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 5, .crm = 1, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "AFSR1_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 5, .crm = 1, .opc2 = 1, - .access = PL2_RW, .type = ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "TCR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 2, .crm = 0, .opc2 = 2, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "VTCR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 2, .crm = 1, .opc2 = 2, - .access = PL2_RW, .accessfn = access_el3_aa32ns, - .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "VTTBR", .state = ARM_CP_STATE_AA32, - .cp = 15, .opc1 = 6, .crm = 2, - .access = PL2_RW, .accessfn = access_el3_aa32ns, - .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, - { .name = "VTTBR_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 2, .crm = 1, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "SCTLR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 0, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "TPIDR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 13, .crm = 0, .opc2 = 2, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "TTBR0_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 2, .crm = 0, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "HTTBR", .cp = 15, .opc1 = 4, .crm = 2, - .access = PL2_RW, .type = ARM_CP_64BIT | ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "CNTHCTL_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 14, .crm = 1, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CNTVOFF_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 14, .crm = 0, .opc2 = 3, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CNTVOFF", .cp = 15, .opc1 = 4, .crm = 14, - .access = PL2_RW, .type = ARM_CP_64BIT | ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "CNTHP_CVAL_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 14, .crm = 2, .opc2 = 2, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CNTHP_CVAL", .cp = 15, .opc1 = 6, .crm = 14, - .access = PL2_RW, .type = ARM_CP_64BIT | ARM_CP_CONST, - .resetvalue = 0 }, - { .name = "CNTHP_TVAL_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 14, .crm = 2, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CNTHP_CTL_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 14, .crm = 2, .opc2 = 1, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "MDCR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 1, - .access = PL2_RW, .accessfn = access_tda, - .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "HPFAR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 6, .crm = 0, .opc2 = 4, - .access = PL2_RW, .accessfn = access_el3_aa32ns, - .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "HSTR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 3, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "FAR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 6, .crm = 0, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "HIFAR", .state = ARM_CP_STATE_AA32, - .type = ARM_CP_CONST, - .cp = 15, .opc1 = 4, .crn = 6, .crm = 0, .opc2 = 2, - .access = PL2_RW, .resetvalue = 0 }, -}; - -/* Ditto, but for registers which exist in ARMv8 but not v7 */ -static const ARMCPRegInfo el3_no_el2_v8_cp_reginfo[] = { - { .name = "HCR2", .state = ARM_CP_STATE_AA32, - .cp = 15, .opc1 = 4, .crn = 1, .crm = 1, .opc2 = 4, - .access = PL2_RW, - .type = ARM_CP_CONST, .resetvalue = 0 }, -}; - static void do_hcr_write(CPUARMState *env, uint64_t value, uint64_t valid_mask) { ARMCPU *cpu = env_archcpu(env); @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) define_arm_cp_regs(cpu, v8_idregs); define_arm_cp_regs(cpu, v8_cp_reginfo); } - if (arm_feature(env, ARM_FEATURE_EL2)) { + + /* + * Register the base EL2 cpregs. + * Pre v8, these registers are implemented only as part of the + * Virtualization Extensions (EL2 present). Beginning with v8, + * if EL2 is missing but EL3 is enabled, mostly these become + * RES0 from EL3, with some specific exceptions. + */ + if (arm_feature(env, ARM_FEATURE_EL2) + || (arm_feature(env, ARM_FEATURE_EL3) + && arm_feature(env, ARM_FEATURE_V8))) { uint64_t vmpidr_def = mpidr_read_val(env); ARMCPRegInfo vpidr_regs[] = { { .name = "VPIDR", .state = ARM_CP_STATE_AA32, @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) }; define_one_arm_cp_reg(cpu, &rvbar); } - } else { - /* If EL2 is missing but higher ELs are enabled, we need to - * register the no_el2 reginfos. - */ - if (arm_feature(env, ARM_FEATURE_EL3)) { - /* When EL3 exists but not EL2, VPIDR and VMPIDR take the value - * of MIDR_EL1 and MPIDR_EL1. - */ - ARMCPRegInfo vpidr_regs[] = { - { .name = "VPIDR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 0, .crm = 0, .opc2 = 0, - .access = PL2_RW, .accessfn = access_el3_aa32ns, - .type = ARM_CP_CONST, .resetvalue = cpu->midr, - .fieldoffset = offsetof(CPUARMState, cp15.vpidr_el2) }, - { .name = "VMPIDR_EL2", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 4, .crn = 0, .crm = 0, .opc2 = 5, - .access = PL2_RW, .accessfn = access_el3_aa32ns, - .type = ARM_CP_NO_RAW, - .writefn = arm_cp_write_ignore, .readfn = mpidr_read }, - }; - define_arm_cp_regs(cpu, vpidr_regs); - define_arm_cp_regs(cpu, el3_no_el2_cp_reginfo); - if (arm_feature(env, ARM_FEATURE_V8)) { - define_arm_cp_regs(cpu, el3_no_el2_v8_cp_reginfo); - } - } } + + /* Register the base EL3 cpregs. */ if (arm_feature(env, ARM_FEATURE_EL3)) { define_arm_cp_regs(cpu, el3_cp_reginfo); ARMCPRegInfo el3_regs[] = { -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Drop zcr_no_el2_reginfo and merge the 3 registers into one array, now that ZCR_EL2 can be squashed to RES0 and ZCR_EL3 dropped while registering. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-4-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 55 ++++++++++++++------------------------------- 1 file changed, 17 insertions(+), 38 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 zcr_write(CPUARMState *env, const ARMCPRegInfo *ri, } } -static const ARMCPRegInfo zcr_el1_reginfo = { - .name = "ZCR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 0, .crn = 1, .crm = 2, .opc2 = 0, - .access = PL1_RW, .type = ARM_CP_SVE, - .fieldoffset = offsetof(CPUARMState, vfp.zcr_el[1]), - .writefn = zcr_write, .raw_writefn = raw_write -}; - -static const ARMCPRegInfo zcr_el2_reginfo = { - .name = "ZCR_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 2, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_SVE, - .fieldoffset = offsetof(CPUARMState, vfp.zcr_el[2]), - .writefn = zcr_write, .raw_writefn = raw_write -}; - -static const ARMCPRegInfo zcr_no_el2_reginfo = { - .name = "ZCR_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 2, .opc2 = 0, - .access = PL2_RW, .type = ARM_CP_SVE, - .readfn = arm_cp_read_zero, .writefn = arm_cp_write_ignore -}; - -static const ARMCPRegInfo zcr_el3_reginfo = { - .name = "ZCR_EL3", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 6, .crn = 1, .crm = 2, .opc2 = 0, - .access = PL3_RW, .type = ARM_CP_SVE, - .fieldoffset = offsetof(CPUARMState, vfp.zcr_el[3]), - .writefn = zcr_write, .raw_writefn = raw_write +static const ARMCPRegInfo zcr_reginfo[] = { + { .name = "ZCR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 0, .crn = 1, .crm = 2, .opc2 = 0, + .access = PL1_RW, .type = ARM_CP_SVE, + .fieldoffset = offsetof(CPUARMState, vfp.zcr_el[1]), + .writefn = zcr_write, .raw_writefn = raw_write }, + { .name = "ZCR_EL2", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 4, .crn = 1, .crm = 2, .opc2 = 0, + .access = PL2_RW, .type = ARM_CP_SVE, + .fieldoffset = offsetof(CPUARMState, vfp.zcr_el[2]), + .writefn = zcr_write, .raw_writefn = raw_write }, + { .name = "ZCR_EL3", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 6, .crn = 1, .crm = 2, .opc2 = 0, + .access = PL3_RW, .type = ARM_CP_SVE, + .fieldoffset = offsetof(CPUARMState, vfp.zcr_el[3]), + .writefn = zcr_write, .raw_writefn = raw_write }, }; void hw_watchpoint_update(ARMCPU *cpu, int n) @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) } if (cpu_isar_feature(aa64_sve, cpu)) { - define_one_arm_cp_reg(cpu, &zcr_el1_reginfo); - if (arm_feature(env, ARM_FEATURE_EL2)) { - define_one_arm_cp_reg(cpu, &zcr_el2_reginfo); - } else { - define_one_arm_cp_reg(cpu, &zcr_no_el2_reginfo); - } - if (arm_feature(env, ARM_FEATURE_EL3)) { - define_one_arm_cp_reg(cpu, &zcr_el3_reginfo); - } + define_arm_cp_regs(cpu, zcr_reginfo); } #ifdef TARGET_AARCH64 -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> This register is present for either VHE or Debugv8p2. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-5-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 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 const ARMCPRegInfo jazelle_regs[] = { .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, }; +static const ARMCPRegInfo contextidr_el2 = { + .name = "CONTEXTIDR_EL2", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 4, .crn = 13, .crm = 0, .opc2 = 1, + .access = PL2_RW, + .fieldoffset = offsetof(CPUARMState, cp15.contextidr_el[2]) +}; + static const ARMCPRegInfo vhe_reginfo[] = { - { .name = "CONTEXTIDR_EL2", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 4, .crn = 13, .crm = 0, .opc2 = 1, - .access = PL2_RW, - .fieldoffset = offsetof(CPUARMState, cp15.contextidr_el[2]) }, { .name = "TTBR1_EL2", .state = ARM_CP_STATE_AA64, .opc0 = 3, .opc1 = 4, .crn = 2, .crm = 0, .opc2 = 1, .access = PL2_RW, .writefn = vmsa_tcr_ttbr_el2_write, @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) define_one_arm_cp_reg(cpu, &ssbs_reginfo); } + if (cpu_isar_feature(aa64_vh, cpu) || + cpu_isar_feature(aa64_debugv8p2, cpu)) { + define_one_arm_cp_reg(cpu, &contextidr_el2); + } if (arm_feature(env, ARM_FEATURE_EL2) && cpu_isar_feature(aa64_vh, cpu)) { define_arm_cp_regs(cpu, vhe_reginfo); } -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Previously we were defining some of these in user-only mode, but none of them are accessible from user-only, therefore define them only in system mode. This will shortly be used from cpu_tcg.c also. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-6-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/internals.h | 6 ++++ target/arm/cpu64.c | 64 +++--------------------------------------- target/arm/cpu_tcg.c | 59 ++++++++++++++++++++++++++++++++++++++ 3 files changed, 69 insertions(+), 60 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 @@ int aarch64_fpu_gdb_get_reg(CPUARMState *env, GByteArray *buf, int reg); int aarch64_fpu_gdb_set_reg(CPUARMState *env, uint8_t *buf, int reg); #endif +#ifdef CONFIG_USER_ONLY +static inline void define_cortex_a72_a57_a53_cp_reginfo(ARMCPU *cpu) { } +#else +void define_cortex_a72_a57_a53_cp_reginfo(ARMCPU *cpu); +#endif + #endif 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" +#include "internals.h" -#ifndef CONFIG_USER_ONLY -static uint64_t a57_a53_l2ctlr_read(CPUARMState *env, const ARMCPRegInfo *ri) -{ - ARMCPU *cpu = env_archcpu(env); - - /* Number of cores is in [25:24]; otherwise we RAZ */ - return (cpu->core_count - 1) << 24; -} -#endif - -static const ARMCPRegInfo cortex_a72_a57_a53_cp_reginfo[] = { -#ifndef CONFIG_USER_ONLY - { .name = "L2CTLR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 11, .crm = 0, .opc2 = 2, - .access = PL1_RW, .readfn = a57_a53_l2ctlr_read, - .writefn = arm_cp_write_ignore }, - { .name = "L2CTLR", - .cp = 15, .opc1 = 1, .crn = 9, .crm = 0, .opc2 = 2, - .access = PL1_RW, .readfn = a57_a53_l2ctlr_read, - .writefn = arm_cp_write_ignore }, -#endif - { .name = "L2ECTLR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 11, .crm = 0, .opc2 = 3, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "L2ECTLR", - .cp = 15, .opc1 = 1, .crn = 9, .crm = 0, .opc2 = 3, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "L2ACTLR", .state = ARM_CP_STATE_BOTH, - .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 0, .opc2 = 0, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CPUACTLR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 0, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CPUACTLR", - .cp = 15, .opc1 = 0, .crm = 15, - .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, - { .name = "CPUECTLR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 1, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CPUECTLR", - .cp = 15, .opc1 = 1, .crm = 15, - .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, - { .name = "CPUMERRSR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 2, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "CPUMERRSR", - .cp = 15, .opc1 = 2, .crm = 15, - .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, - { .name = "L2MERRSR_EL1", .state = ARM_CP_STATE_AA64, - .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 3, - .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, - { .name = "L2MERRSR", - .cp = 15, .opc1 = 3, .crm = 15, - .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, -}; - static void aarch64_a57_initfn(Object *obj) { ARMCPU *cpu = ARM_CPU(obj); @@ -XXX,XX +XXX,XX @@ static void aarch64_a57_initfn(Object *obj) cpu->gic_num_lrs = 4; cpu->gic_vpribits = 5; cpu->gic_vprebits = 5; - define_arm_cp_regs(cpu, cortex_a72_a57_a53_cp_reginfo); + define_cortex_a72_a57_a53_cp_reginfo(cpu); } static void aarch64_a53_initfn(Object *obj) @@ -XXX,XX +XXX,XX @@ static void aarch64_a53_initfn(Object *obj) cpu->gic_num_lrs = 4; cpu->gic_vpribits = 5; cpu->gic_vprebits = 5; - define_arm_cp_regs(cpu, cortex_a72_a57_a53_cp_reginfo); + define_cortex_a72_a57_a53_cp_reginfo(cpu); } static void aarch64_a72_initfn(Object *obj) @@ -XXX,XX +XXX,XX @@ static void aarch64_a72_initfn(Object *obj) cpu->gic_num_lrs = 4; cpu->gic_vpribits = 5; cpu->gic_vprebits = 5; - define_arm_cp_regs(cpu, cortex_a72_a57_a53_cp_reginfo); + define_cortex_a72_a57_a53_cp_reginfo(cpu); } void arm_cpu_sve_finalize(ARMCPU *cpu, Error **errp) 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 @@ #endif #include "cpregs.h" +#ifndef CONFIG_USER_ONLY +static uint64_t l2ctlr_read(CPUARMState *env, const ARMCPRegInfo *ri) +{ + ARMCPU *cpu = env_archcpu(env); + + /* Number of cores is in [25:24]; otherwise we RAZ */ + return (cpu->core_count - 1) << 24; +} + +static const ARMCPRegInfo cortex_a72_a57_a53_cp_reginfo[] = { + { .name = "L2CTLR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 11, .crm = 0, .opc2 = 2, + .access = PL1_RW, .readfn = l2ctlr_read, + .writefn = arm_cp_write_ignore }, + { .name = "L2CTLR", + .cp = 15, .opc1 = 1, .crn = 9, .crm = 0, .opc2 = 2, + .access = PL1_RW, .readfn = l2ctlr_read, + .writefn = arm_cp_write_ignore }, + { .name = "L2ECTLR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 11, .crm = 0, .opc2 = 3, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "L2ECTLR", + .cp = 15, .opc1 = 1, .crn = 9, .crm = 0, .opc2 = 3, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "L2ACTLR", .state = ARM_CP_STATE_BOTH, + .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 0, .opc2 = 0, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "CPUACTLR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 0, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "CPUACTLR", + .cp = 15, .opc1 = 0, .crm = 15, + .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, + { .name = "CPUECTLR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 1, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "CPUECTLR", + .cp = 15, .opc1 = 1, .crm = 15, + .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, + { .name = "CPUMERRSR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 2, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "CPUMERRSR", + .cp = 15, .opc1 = 2, .crm = 15, + .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, + { .name = "L2MERRSR_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 1, .crn = 15, .crm = 2, .opc2 = 3, + .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "L2MERRSR", + .cp = 15, .opc1 = 3, .crm = 15, + .access = PL1_RW, .type = ARM_CP_CONST | ARM_CP_64BIT, .resetvalue = 0 }, +}; + +void define_cortex_a72_a57_a53_cp_reginfo(ARMCPU *cpu) +{ + define_arm_cp_regs(cpu, cortex_a72_a57_a53_cp_reginfo); +} +#endif /* !CONFIG_USER_ONLY */ + /* CPU models. These are not needed for the AArch64 linux-user build. */ #if !defined(CONFIG_USER_ONLY) || !defined(TARGET_AARCH64) -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Instead of starting with cortex-a15 and adding v8 features to a v7 cpu, begin with a v8 cpu stripped of its aarch64 features. This fixes the long-standing to-do where we only enabled v8 features for user-only. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-7-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu_tcg.c | 151 ++++++++++++++++++++++++++----------------- 1 file changed, 92 insertions(+), 59 deletions(-) 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 void arm_v7m_class_init(ObjectClass *oc, void *data) static void arm_max_initfn(Object *obj) { ARMCPU *cpu = ARM_CPU(obj); + uint32_t t; - cortex_a15_initfn(obj); + /* aarch64_a57_initfn, advertising none of the aarch64 features */ + cpu->dtb_compatible = "arm,cortex-a57"; + set_feature(&cpu->env, ARM_FEATURE_V8); + set_feature(&cpu->env, ARM_FEATURE_NEON); + set_feature(&cpu->env, ARM_FEATURE_GENERIC_TIMER); + set_feature(&cpu->env, ARM_FEATURE_CBAR_RO); + set_feature(&cpu->env, ARM_FEATURE_EL2); + set_feature(&cpu->env, ARM_FEATURE_EL3); + set_feature(&cpu->env, ARM_FEATURE_PMU); + cpu->midr = 0x411fd070; + cpu->revidr = 0x00000000; + cpu->reset_fpsid = 0x41034070; + cpu->isar.mvfr0 = 0x10110222; + cpu->isar.mvfr1 = 0x12111111; + cpu->isar.mvfr2 = 0x00000043; + cpu->ctr = 0x8444c004; + cpu->reset_sctlr = 0x00c50838; + cpu->isar.id_pfr0 = 0x00000131; + cpu->isar.id_pfr1 = 0x00011011; + cpu->isar.id_dfr0 = 0x03010066; + cpu->id_afr0 = 0x00000000; + cpu->isar.id_mmfr0 = 0x10101105; + cpu->isar.id_mmfr1 = 0x40000000; + cpu->isar.id_mmfr2 = 0x01260000; + cpu->isar.id_mmfr3 = 0x02102211; + cpu->isar.id_isar0 = 0x02101110; + cpu->isar.id_isar1 = 0x13112111; + cpu->isar.id_isar2 = 0x21232042; + cpu->isar.id_isar3 = 0x01112131; + cpu->isar.id_isar4 = 0x00011142; + cpu->isar.id_isar5 = 0x00011121; + cpu->isar.id_isar6 = 0; + cpu->isar.dbgdidr = 0x3516d000; + cpu->clidr = 0x0a200023; + cpu->ccsidr[0] = 0x701fe00a; /* 32KB L1 dcache */ + cpu->ccsidr[1] = 0x201fe012; /* 48KB L1 icache */ + cpu->ccsidr[2] = 0x70ffe07a; /* 2048KB L2 cache */ + define_cortex_a72_a57_a53_cp_reginfo(cpu); - /* old-style VFP short-vector support */ - cpu->isar.mvfr0 = FIELD_DP32(cpu->isar.mvfr0, MVFR0, FPSHVEC, 1); + /* Add additional features supported by QEMU */ + t = cpu->isar.id_isar5; + t = FIELD_DP32(t, ID_ISAR5, AES, 2); + t = FIELD_DP32(t, ID_ISAR5, SHA1, 1); + t = FIELD_DP32(t, ID_ISAR5, SHA2, 1); + t = FIELD_DP32(t, ID_ISAR5, CRC32, 1); + t = FIELD_DP32(t, ID_ISAR5, RDM, 1); + t = FIELD_DP32(t, ID_ISAR5, VCMA, 1); + cpu->isar.id_isar5 = t; + + t = cpu->isar.id_isar6; + t = FIELD_DP32(t, ID_ISAR6, JSCVT, 1); + t = FIELD_DP32(t, ID_ISAR6, DP, 1); + t = FIELD_DP32(t, ID_ISAR6, FHM, 1); + t = FIELD_DP32(t, ID_ISAR6, SB, 1); + t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); + t = FIELD_DP32(t, ID_ISAR6, BF16, 1); + t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); + cpu->isar.id_isar6 = t; + + t = cpu->isar.mvfr1; + t = FIELD_DP32(t, MVFR1, FPHP, 3); /* v8.2-FP16 */ + t = FIELD_DP32(t, MVFR1, SIMDHP, 2); /* v8.2-FP16 */ + cpu->isar.mvfr1 = t; + + t = cpu->isar.mvfr2; + t = FIELD_DP32(t, MVFR2, SIMDMISC, 3); /* SIMD MaxNum */ + t = FIELD_DP32(t, MVFR2, FPMISC, 4); /* FP MaxNum */ + cpu->isar.mvfr2 = t; + + t = cpu->isar.id_mmfr3; + t = FIELD_DP32(t, ID_MMFR3, PAN, 2); /* ATS1E1 */ + cpu->isar.id_mmfr3 = t; + + t = cpu->isar.id_mmfr4; + t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* AA32HPD */ + t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ + t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* TTCNP */ + t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* TTS2UXN */ + cpu->isar.id_mmfr4 = t; + + t = cpu->isar.id_pfr0; + t = FIELD_DP32(t, ID_PFR0, DIT, 1); + cpu->isar.id_pfr0 = t; + + t = cpu->isar.id_pfr2; + t = FIELD_DP32(t, ID_PFR2, SSBS, 1); + cpu->isar.id_pfr2 = t; #ifdef CONFIG_USER_ONLY /* - * We don't set these in system emulation mode for the moment, - * since we don't correctly set (all of) the ID registers to - * advertise them. + * Break with true ARMv8 and add back old-style VFP short-vector support. + * Only do this for user-mode, where -cpu max is the default, so that + * older v6 and v7 programs are more likely to work without adjustment. */ - set_feature(&cpu->env, ARM_FEATURE_V8); - { - uint32_t t; - - t = cpu->isar.id_isar5; - t = FIELD_DP32(t, ID_ISAR5, AES, 2); - t = FIELD_DP32(t, ID_ISAR5, SHA1, 1); - t = FIELD_DP32(t, ID_ISAR5, SHA2, 1); - t = FIELD_DP32(t, ID_ISAR5, CRC32, 1); - t = FIELD_DP32(t, ID_ISAR5, RDM, 1); - t = FIELD_DP32(t, ID_ISAR5, VCMA, 1); - cpu->isar.id_isar5 = t; - - t = cpu->isar.id_isar6; - t = FIELD_DP32(t, ID_ISAR6, JSCVT, 1); - t = FIELD_DP32(t, ID_ISAR6, DP, 1); - t = FIELD_DP32(t, ID_ISAR6, FHM, 1); - t = FIELD_DP32(t, ID_ISAR6, SB, 1); - t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); - t = FIELD_DP32(t, ID_ISAR6, BF16, 1); - t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); - cpu->isar.id_isar6 = t; - - t = cpu->isar.mvfr1; - t = FIELD_DP32(t, MVFR1, FPHP, 3); /* v8.2-FP16 */ - t = FIELD_DP32(t, MVFR1, SIMDHP, 2); /* v8.2-FP16 */ - cpu->isar.mvfr1 = t; - - t = cpu->isar.mvfr2; - t = FIELD_DP32(t, MVFR2, SIMDMISC, 3); /* SIMD MaxNum */ - t = FIELD_DP32(t, MVFR2, FPMISC, 4); /* FP MaxNum */ - cpu->isar.mvfr2 = t; - - t = cpu->isar.id_mmfr3; - t = FIELD_DP32(t, ID_MMFR3, PAN, 2); /* ATS1E1 */ - cpu->isar.id_mmfr3 = t; - - t = cpu->isar.id_mmfr4; - t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* AA32HPD */ - t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ - t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* TTCNP */ - t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* TTS2UXN */ - cpu->isar.id_mmfr4 = t; - - t = cpu->isar.id_pfr0; - t = FIELD_DP32(t, ID_PFR0, DIT, 1); - cpu->isar.id_pfr0 = t; - - t = cpu->isar.id_pfr2; - t = FIELD_DP32(t, ID_PFR2, SSBS, 1); - cpu->isar.id_pfr2 = t; - } -#endif /* CONFIG_USER_ONLY */ + cpu->isar.mvfr0 = FIELD_DP32(cpu->isar.mvfr0, MVFR0, FPSHVEC, 1); +#endif } #endif /* !TARGET_AARCH64 */ -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> We set this for qemu-system-aarch64, but failed to do so for the strictly 32-bit emulation. Fixes: 3bec78447a9 ("target/arm: Provide ARMv8.4-PMU in '-cpu max'") Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-8-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu_tcg.c | 4 ++++ 1 file changed, 4 insertions(+) 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 void arm_max_initfn(Object *obj) t = FIELD_DP32(t, ID_PFR2, SSBS, 1); cpu->isar.id_pfr2 = t; + t = cpu->isar.id_dfr0; + t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* v8.4-PMU */ + cpu->isar.id_dfr0 = t; + #ifdef CONFIG_USER_ONLY /* * Break with true ARMv8 and add back old-style VFP short-vector support. -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Share the code to set AArch32 max features so that we no longer have code drift between qemu{-system,}-{arm,aarch64}. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-9-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/internals.h | 2 + target/arm/cpu64.c | 50 +----------------- target/arm/cpu_tcg.c | 114 ++++++++++++++++++++++------------------- 3 files changed, 65 insertions(+), 101 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 @@ static inline void define_cortex_a72_a57_a53_cp_reginfo(ARMCPU *cpu) { } void define_cortex_a72_a57_a53_cp_reginfo(ARMCPU *cpu); #endif +void aa32_max_features(ARMCPU *cpu); + #endif 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 void aarch64_max_initfn(Object *obj) { ARMCPU *cpu = ARM_CPU(obj); uint64_t t; - uint32_t u; if (kvm_enabled() || hvf_enabled()) { /* With KVM or HVF, '-cpu max' is identical to '-cpu host' */ @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64ZFR0, F64MM, 1); cpu->isar.id_aa64zfr0 = t; - /* Replicate the same data to the 32-bit id registers. */ - u = cpu->isar.id_isar5; - u = FIELD_DP32(u, ID_ISAR5, AES, 2); /* AES + PMULL */ - u = FIELD_DP32(u, ID_ISAR5, SHA1, 1); - u = FIELD_DP32(u, ID_ISAR5, SHA2, 1); - u = FIELD_DP32(u, ID_ISAR5, CRC32, 1); - u = FIELD_DP32(u, ID_ISAR5, RDM, 1); - u = FIELD_DP32(u, ID_ISAR5, VCMA, 1); - cpu->isar.id_isar5 = u; - - u = cpu->isar.id_isar6; - u = FIELD_DP32(u, ID_ISAR6, JSCVT, 1); - u = FIELD_DP32(u, ID_ISAR6, DP, 1); - u = FIELD_DP32(u, ID_ISAR6, FHM, 1); - u = FIELD_DP32(u, ID_ISAR6, SB, 1); - u = FIELD_DP32(u, ID_ISAR6, SPECRES, 1); - u = FIELD_DP32(u, ID_ISAR6, BF16, 1); - u = FIELD_DP32(u, ID_ISAR6, I8MM, 1); - cpu->isar.id_isar6 = u; - - u = cpu->isar.id_pfr0; - u = FIELD_DP32(u, ID_PFR0, DIT, 1); - cpu->isar.id_pfr0 = u; - - u = cpu->isar.id_pfr2; - u = FIELD_DP32(u, ID_PFR2, SSBS, 1); - cpu->isar.id_pfr2 = u; - - u = cpu->isar.id_mmfr3; - u = FIELD_DP32(u, ID_MMFR3, PAN, 2); /* ATS1E1 */ - cpu->isar.id_mmfr3 = u; - - u = cpu->isar.id_mmfr4; - u = FIELD_DP32(u, ID_MMFR4, HPDS, 1); /* AA32HPD */ - u = FIELD_DP32(u, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ - u = FIELD_DP32(u, ID_MMFR4, CNP, 1); /* TTCNP */ - u = FIELD_DP32(u, ID_MMFR4, XNX, 1); /* TTS2UXN */ - cpu->isar.id_mmfr4 = u; - t = cpu->isar.id_aa64dfr0; t = FIELD_DP64(t, ID_AA64DFR0, PMUVER, 5); /* v8.4-PMU */ cpu->isar.id_aa64dfr0 = t; - u = cpu->isar.id_dfr0; - u = FIELD_DP32(u, ID_DFR0, PERFMON, 5); /* v8.4-PMU */ - cpu->isar.id_dfr0 = u; - - u = cpu->isar.mvfr1; - u = FIELD_DP32(u, MVFR1, FPHP, 3); /* v8.2-FP16 */ - u = FIELD_DP32(u, MVFR1, SIMDHP, 2); /* v8.2-FP16 */ - cpu->isar.mvfr1 = u; + /* Replicate the same data to the 32-bit id registers. */ + aa32_max_features(cpu); #ifdef 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 @@ #endif #include "cpregs.h" + +/* Share AArch32 -cpu max features with AArch64. */ +void aa32_max_features(ARMCPU *cpu) +{ + uint32_t t; + + /* Add additional features supported by QEMU */ + t = cpu->isar.id_isar5; + t = FIELD_DP32(t, ID_ISAR5, AES, 2); + t = FIELD_DP32(t, ID_ISAR5, SHA1, 1); + t = FIELD_DP32(t, ID_ISAR5, SHA2, 1); + t = FIELD_DP32(t, ID_ISAR5, CRC32, 1); + t = FIELD_DP32(t, ID_ISAR5, RDM, 1); + t = FIELD_DP32(t, ID_ISAR5, VCMA, 1); + cpu->isar.id_isar5 = t; + + t = cpu->isar.id_isar6; + t = FIELD_DP32(t, ID_ISAR6, JSCVT, 1); + t = FIELD_DP32(t, ID_ISAR6, DP, 1); + t = FIELD_DP32(t, ID_ISAR6, FHM, 1); + t = FIELD_DP32(t, ID_ISAR6, SB, 1); + t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); + t = FIELD_DP32(t, ID_ISAR6, BF16, 1); + t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); + cpu->isar.id_isar6 = t; + + t = cpu->isar.mvfr1; + t = FIELD_DP32(t, MVFR1, FPHP, 3); /* v8.2-FP16 */ + t = FIELD_DP32(t, MVFR1, SIMDHP, 2); /* v8.2-FP16 */ + cpu->isar.mvfr1 = t; + + t = cpu->isar.mvfr2; + t = FIELD_DP32(t, MVFR2, SIMDMISC, 3); /* SIMD MaxNum */ + t = FIELD_DP32(t, MVFR2, FPMISC, 4); /* FP MaxNum */ + cpu->isar.mvfr2 = t; + + t = cpu->isar.id_mmfr3; + t = FIELD_DP32(t, ID_MMFR3, PAN, 2); /* ATS1E1 */ + cpu->isar.id_mmfr3 = t; + + t = cpu->isar.id_mmfr4; + t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* AA32HPD */ + t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ + t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* TTCNP */ + t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* TTS2UXN */ + cpu->isar.id_mmfr4 = t; + + t = cpu->isar.id_pfr0; + t = FIELD_DP32(t, ID_PFR0, DIT, 1); + cpu->isar.id_pfr0 = t; + + t = cpu->isar.id_pfr2; + t = FIELD_DP32(t, ID_PFR2, SSBS, 1); + cpu->isar.id_pfr2 = t; + + t = cpu->isar.id_dfr0; + t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* v8.4-PMU */ + cpu->isar.id_dfr0 = t; +} + #ifndef CONFIG_USER_ONLY static uint64_t l2ctlr_read(CPUARMState *env, const ARMCPRegInfo *ri) { @@ -XXX,XX +XXX,XX @@ static void arm_v7m_class_init(ObjectClass *oc, void *data) static void arm_max_initfn(Object *obj) { ARMCPU *cpu = ARM_CPU(obj); - uint32_t t; /* aarch64_a57_initfn, advertising none of the aarch64 features */ cpu->dtb_compatible = "arm,cortex-a57"; @@ -XXX,XX +XXX,XX @@ static void arm_max_initfn(Object *obj) cpu->ccsidr[2] = 0x70ffe07a; /* 2048KB L2 cache */ define_cortex_a72_a57_a53_cp_reginfo(cpu); - /* Add additional features supported by QEMU */ - t = cpu->isar.id_isar5; - t = FIELD_DP32(t, ID_ISAR5, AES, 2); - t = FIELD_DP32(t, ID_ISAR5, SHA1, 1); - t = FIELD_DP32(t, ID_ISAR5, SHA2, 1); - t = FIELD_DP32(t, ID_ISAR5, CRC32, 1); - t = FIELD_DP32(t, ID_ISAR5, RDM, 1); - t = FIELD_DP32(t, ID_ISAR5, VCMA, 1); - cpu->isar.id_isar5 = t; - - t = cpu->isar.id_isar6; - t = FIELD_DP32(t, ID_ISAR6, JSCVT, 1); - t = FIELD_DP32(t, ID_ISAR6, DP, 1); - t = FIELD_DP32(t, ID_ISAR6, FHM, 1); - t = FIELD_DP32(t, ID_ISAR6, SB, 1); - t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); - t = FIELD_DP32(t, ID_ISAR6, BF16, 1); - t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); - cpu->isar.id_isar6 = t; - - t = cpu->isar.mvfr1; - t = FIELD_DP32(t, MVFR1, FPHP, 3); /* v8.2-FP16 */ - t = FIELD_DP32(t, MVFR1, SIMDHP, 2); /* v8.2-FP16 */ - cpu->isar.mvfr1 = t; - - t = cpu->isar.mvfr2; - t = FIELD_DP32(t, MVFR2, SIMDMISC, 3); /* SIMD MaxNum */ - t = FIELD_DP32(t, MVFR2, FPMISC, 4); /* FP MaxNum */ - cpu->isar.mvfr2 = t; - - t = cpu->isar.id_mmfr3; - t = FIELD_DP32(t, ID_MMFR3, PAN, 2); /* ATS1E1 */ - cpu->isar.id_mmfr3 = t; - - t = cpu->isar.id_mmfr4; - t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* AA32HPD */ - t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ - t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* TTCNP */ - t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* TTS2UXN */ - cpu->isar.id_mmfr4 = t; - - t = cpu->isar.id_pfr0; - t = FIELD_DP32(t, ID_PFR0, DIT, 1); - cpu->isar.id_pfr0 = t; - - t = cpu->isar.id_pfr2; - t = FIELD_DP32(t, ID_PFR2, SSBS, 1); - cpu->isar.id_pfr2 = t; - - t = cpu->isar.id_dfr0; - t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* v8.4-PMU */ - cpu->isar.id_dfr0 = t; + aa32_max_features(cpu); #ifdef CONFIG_USER_ONLY /* -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Update the legacy feature names to the current names. Provide feature names for id changes that were not marked. Sort the field updates into increasing bitfield order. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-10-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu64.c | 100 +++++++++++++++++++++---------------------- target/arm/cpu_tcg.c | 48 ++++++++++----------- 2 files changed, 74 insertions(+), 74 deletions(-) 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 void aarch64_max_initfn(Object *obj) cpu->midr = t; t = cpu->isar.id_aa64isar0; - t = FIELD_DP64(t, ID_AA64ISAR0, AES, 2); /* AES + PMULL */ - t = FIELD_DP64(t, ID_AA64ISAR0, SHA1, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, SHA2, 2); /* SHA512 */ + t = FIELD_DP64(t, ID_AA64ISAR0, AES, 2); /* FEAT_PMULL */ + t = FIELD_DP64(t, ID_AA64ISAR0, SHA1, 1); /* FEAT_SHA1 */ + t = FIELD_DP64(t, ID_AA64ISAR0, SHA2, 2); /* FEAT_SHA512 */ t = FIELD_DP64(t, ID_AA64ISAR0, CRC32, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, ATOMIC, 2); - t = FIELD_DP64(t, ID_AA64ISAR0, RDM, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, SHA3, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, SM3, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, SM4, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, DP, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, FHM, 1); - t = FIELD_DP64(t, ID_AA64ISAR0, TS, 2); /* v8.5-CondM */ - t = FIELD_DP64(t, ID_AA64ISAR0, TLB, 2); /* FEAT_TLBIRANGE */ - t = FIELD_DP64(t, ID_AA64ISAR0, RNDR, 1); + t = FIELD_DP64(t, ID_AA64ISAR0, ATOMIC, 2); /* FEAT_LSE */ + t = FIELD_DP64(t, ID_AA64ISAR0, RDM, 1); /* FEAT_RDM */ + t = FIELD_DP64(t, ID_AA64ISAR0, SHA3, 1); /* FEAT_SHA3 */ + t = FIELD_DP64(t, ID_AA64ISAR0, SM3, 1); /* FEAT_SM3 */ + t = FIELD_DP64(t, ID_AA64ISAR0, SM4, 1); /* FEAT_SM4 */ + t = FIELD_DP64(t, ID_AA64ISAR0, DP, 1); /* FEAT_DotProd */ + t = FIELD_DP64(t, ID_AA64ISAR0, FHM, 1); /* FEAT_FHM */ + t = FIELD_DP64(t, ID_AA64ISAR0, TS, 2); /* FEAT_FlagM2 */ + t = FIELD_DP64(t, ID_AA64ISAR0, TLB, 2); /* FEAT_TLBIRANGE */ + t = FIELD_DP64(t, ID_AA64ISAR0, RNDR, 1); /* FEAT_RNG */ cpu->isar.id_aa64isar0 = t; t = cpu->isar.id_aa64isar1; - t = FIELD_DP64(t, ID_AA64ISAR1, DPB, 2); - t = FIELD_DP64(t, ID_AA64ISAR1, JSCVT, 1); - t = FIELD_DP64(t, ID_AA64ISAR1, FCMA, 1); - t = FIELD_DP64(t, ID_AA64ISAR1, SB, 1); - t = FIELD_DP64(t, ID_AA64ISAR1, SPECRES, 1); - t = FIELD_DP64(t, ID_AA64ISAR1, BF16, 1); - t = FIELD_DP64(t, ID_AA64ISAR1, FRINTTS, 1); - t = FIELD_DP64(t, ID_AA64ISAR1, LRCPC, 2); /* ARMv8.4-RCPC */ - t = FIELD_DP64(t, ID_AA64ISAR1, I8MM, 1); + t = FIELD_DP64(t, ID_AA64ISAR1, DPB, 2); /* FEAT_DPB2 */ + t = FIELD_DP64(t, ID_AA64ISAR1, JSCVT, 1); /* FEAT_JSCVT */ + t = FIELD_DP64(t, ID_AA64ISAR1, FCMA, 1); /* FEAT_FCMA */ + t = FIELD_DP64(t, ID_AA64ISAR1, LRCPC, 2); /* FEAT_LRCPC2 */ + t = FIELD_DP64(t, ID_AA64ISAR1, FRINTTS, 1); /* FEAT_FRINTTS */ + t = FIELD_DP64(t, ID_AA64ISAR1, SB, 1); /* FEAT_SB */ + t = FIELD_DP64(t, ID_AA64ISAR1, SPECRES, 1); /* FEAT_SPECRES */ + t = FIELD_DP64(t, ID_AA64ISAR1, BF16, 1); /* FEAT_BF16 */ + t = FIELD_DP64(t, ID_AA64ISAR1, I8MM, 1); /* FEAT_I8MM */ cpu->isar.id_aa64isar1 = t; t = cpu->isar.id_aa64pfr0; + t = FIELD_DP64(t, ID_AA64PFR0, FP, 1); /* FEAT_FP16 */ + t = FIELD_DP64(t, ID_AA64PFR0, ADVSIMD, 1); /* FEAT_FP16 */ t = FIELD_DP64(t, ID_AA64PFR0, SVE, 1); - t = FIELD_DP64(t, ID_AA64PFR0, FP, 1); - t = FIELD_DP64(t, ID_AA64PFR0, ADVSIMD, 1); - t = FIELD_DP64(t, ID_AA64PFR0, SEL2, 1); - t = FIELD_DP64(t, ID_AA64PFR0, DIT, 1); + t = FIELD_DP64(t, ID_AA64PFR0, SEL2, 1); /* FEAT_SEL2 */ + t = FIELD_DP64(t, ID_AA64PFR0, DIT, 1); /* FEAT_DIT */ cpu->isar.id_aa64pfr0 = t; t = cpu->isar.id_aa64pfr1; - t = FIELD_DP64(t, ID_AA64PFR1, BT, 1); - t = FIELD_DP64(t, ID_AA64PFR1, SSBS, 2); + t = FIELD_DP64(t, ID_AA64PFR1, BT, 1); /* FEAT_BTI */ + t = FIELD_DP64(t, ID_AA64PFR1, SSBS, 2); /* FEAT_SSBS2 */ /* * Begin with full support for MTE. This will be downgraded to MTE=0 * during realize if the board provides no tag memory, much like * we do for EL2 with the virtualization=on property. */ - t = FIELD_DP64(t, ID_AA64PFR1, MTE, 3); + t = FIELD_DP64(t, ID_AA64PFR1, MTE, 3); /* FEAT_MTE3 */ cpu->isar.id_aa64pfr1 = t; t = cpu->isar.id_aa64mmfr0; @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) cpu->isar.id_aa64mmfr0 = t; t = cpu->isar.id_aa64mmfr1; - t = FIELD_DP64(t, ID_AA64MMFR1, HPDS, 1); /* HPD */ - t = FIELD_DP64(t, ID_AA64MMFR1, LO, 1); - t = FIELD_DP64(t, ID_AA64MMFR1, VH, 1); - t = FIELD_DP64(t, ID_AA64MMFR1, PAN, 2); /* ATS1E1 */ - t = FIELD_DP64(t, ID_AA64MMFR1, VMIDBITS, 2); /* VMID16 */ - t = FIELD_DP64(t, ID_AA64MMFR1, XNX, 1); /* TTS2UXN */ + t = FIELD_DP64(t, ID_AA64MMFR1, VMIDBITS, 2); /* FEAT_VMID16 */ + t = FIELD_DP64(t, ID_AA64MMFR1, VH, 1); /* FEAT_VHE */ + t = FIELD_DP64(t, ID_AA64MMFR1, HPDS, 1); /* FEAT_HPDS */ + t = FIELD_DP64(t, ID_AA64MMFR1, LO, 1); /* FEAT_LOR */ + t = FIELD_DP64(t, ID_AA64MMFR1, PAN, 2); /* FEAT_PAN2 */ + t = FIELD_DP64(t, ID_AA64MMFR1, XNX, 1); /* FEAT_XNX */ cpu->isar.id_aa64mmfr1 = t; t = cpu->isar.id_aa64mmfr2; - t = FIELD_DP64(t, ID_AA64MMFR2, UAO, 1); - t = FIELD_DP64(t, ID_AA64MMFR2, CNP, 1); /* TTCNP */ - t = FIELD_DP64(t, ID_AA64MMFR2, ST, 1); /* TTST */ - t = FIELD_DP64(t, ID_AA64MMFR2, VARANGE, 1); /* FEAT_LVA */ - t = FIELD_DP64(t, ID_AA64MMFR2, TTL, 1); /* FEAT_TTL */ - t = FIELD_DP64(t, ID_AA64MMFR2, BBM, 2); /* FEAT_BBM at level 2 */ + t = FIELD_DP64(t, ID_AA64MMFR2, CNP, 1); /* FEAT_TTCNP */ + t = FIELD_DP64(t, ID_AA64MMFR2, UAO, 1); /* FEAT_UAO */ + t = FIELD_DP64(t, ID_AA64MMFR2, VARANGE, 1); /* FEAT_LVA */ + t = FIELD_DP64(t, ID_AA64MMFR2, ST, 1); /* FEAT_TTST */ + t = FIELD_DP64(t, ID_AA64MMFR2, TTL, 1); /* FEAT_TTL */ + t = FIELD_DP64(t, ID_AA64MMFR2, BBM, 2); /* FEAT_BBM at level 2 */ cpu->isar.id_aa64mmfr2 = t; t = cpu->isar.id_aa64zfr0; t = FIELD_DP64(t, ID_AA64ZFR0, SVEVER, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, AES, 2); /* PMULL */ - t = FIELD_DP64(t, ID_AA64ZFR0, BITPERM, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, BFLOAT16, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, SHA3, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, SM4, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, I8MM, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, F32MM, 1); - t = FIELD_DP64(t, ID_AA64ZFR0, F64MM, 1); + t = FIELD_DP64(t, ID_AA64ZFR0, AES, 2); /* FEAT_SVE_PMULL128 */ + t = FIELD_DP64(t, ID_AA64ZFR0, BITPERM, 1); /* FEAT_SVE_BitPerm */ + t = FIELD_DP64(t, ID_AA64ZFR0, BFLOAT16, 1); /* FEAT_BF16 */ + t = FIELD_DP64(t, ID_AA64ZFR0, SHA3, 1); /* FEAT_SVE_SHA3 */ + t = FIELD_DP64(t, ID_AA64ZFR0, SM4, 1); /* FEAT_SVE_SM4 */ + t = FIELD_DP64(t, ID_AA64ZFR0, I8MM, 1); /* FEAT_I8MM */ + t = FIELD_DP64(t, ID_AA64ZFR0, F32MM, 1); /* FEAT_F32MM */ + t = FIELD_DP64(t, ID_AA64ZFR0, F64MM, 1); /* FEAT_F64MM */ cpu->isar.id_aa64zfr0 = t; t = cpu->isar.id_aa64dfr0; - t = FIELD_DP64(t, ID_AA64DFR0, PMUVER, 5); /* v8.4-PMU */ + t = FIELD_DP64(t, ID_AA64DFR0, PMUVER, 5); /* FEAT_PMUv3p4 */ cpu->isar.id_aa64dfr0 = t; /* Replicate the same data to the 32-bit id registers. */ 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 @@ void aa32_max_features(ARMCPU *cpu) /* Add additional features supported by QEMU */ t = cpu->isar.id_isar5; - t = FIELD_DP32(t, ID_ISAR5, AES, 2); - t = FIELD_DP32(t, ID_ISAR5, SHA1, 1); - t = FIELD_DP32(t, ID_ISAR5, SHA2, 1); + t = FIELD_DP32(t, ID_ISAR5, AES, 2); /* FEAT_PMULL */ + t = FIELD_DP32(t, ID_ISAR5, SHA1, 1); /* FEAT_SHA1 */ + t = FIELD_DP32(t, ID_ISAR5, SHA2, 1); /* FEAT_SHA256 */ t = FIELD_DP32(t, ID_ISAR5, CRC32, 1); - t = FIELD_DP32(t, ID_ISAR5, RDM, 1); - t = FIELD_DP32(t, ID_ISAR5, VCMA, 1); + t = FIELD_DP32(t, ID_ISAR5, RDM, 1); /* FEAT_RDM */ + t = FIELD_DP32(t, ID_ISAR5, VCMA, 1); /* FEAT_FCMA */ cpu->isar.id_isar5 = t; t = cpu->isar.id_isar6; - t = FIELD_DP32(t, ID_ISAR6, JSCVT, 1); - t = FIELD_DP32(t, ID_ISAR6, DP, 1); - t = FIELD_DP32(t, ID_ISAR6, FHM, 1); - t = FIELD_DP32(t, ID_ISAR6, SB, 1); - t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); - t = FIELD_DP32(t, ID_ISAR6, BF16, 1); - t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); + t = FIELD_DP32(t, ID_ISAR6, JSCVT, 1); /* FEAT_JSCVT */ + t = FIELD_DP32(t, ID_ISAR6, DP, 1); /* Feat_DotProd */ + t = FIELD_DP32(t, ID_ISAR6, FHM, 1); /* FEAT_FHM */ + t = FIELD_DP32(t, ID_ISAR6, SB, 1); /* FEAT_SB */ + t = FIELD_DP32(t, ID_ISAR6, SPECRES, 1); /* FEAT_SPECRES */ + t = FIELD_DP32(t, ID_ISAR6, BF16, 1); /* FEAT_AA32BF16 */ + t = FIELD_DP32(t, ID_ISAR6, I8MM, 1); /* FEAT_AA32I8MM */ cpu->isar.id_isar6 = t; t = cpu->isar.mvfr1; - t = FIELD_DP32(t, MVFR1, FPHP, 3); /* v8.2-FP16 */ - t = FIELD_DP32(t, MVFR1, SIMDHP, 2); /* v8.2-FP16 */ + t = FIELD_DP32(t, MVFR1, FPHP, 3); /* FEAT_FP16 */ + t = FIELD_DP32(t, MVFR1, SIMDHP, 2); /* FEAT_FP16 */ cpu->isar.mvfr1 = t; t = cpu->isar.mvfr2; - t = FIELD_DP32(t, MVFR2, SIMDMISC, 3); /* SIMD MaxNum */ - t = FIELD_DP32(t, MVFR2, FPMISC, 4); /* FP MaxNum */ + t = FIELD_DP32(t, MVFR2, SIMDMISC, 3); /* SIMD MaxNum */ + t = FIELD_DP32(t, MVFR2, FPMISC, 4); /* FP MaxNum */ cpu->isar.mvfr2 = t; t = cpu->isar.id_mmfr3; - t = FIELD_DP32(t, ID_MMFR3, PAN, 2); /* ATS1E1 */ + t = FIELD_DP32(t, ID_MMFR3, PAN, 2); /* FEAT_PAN2 */ cpu->isar.id_mmfr3 = t; t = cpu->isar.id_mmfr4; - t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* AA32HPD */ - t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ - t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* TTCNP */ - t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* TTS2UXN */ + t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* FEAT_AA32HPD */ + t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ + t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* FEAT_TTCNP */ + t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* FEAT_XNX*/ cpu->isar.id_mmfr4 = t; t = cpu->isar.id_pfr0; - t = FIELD_DP32(t, ID_PFR0, DIT, 1); + t = FIELD_DP32(t, ID_PFR0, DIT, 1); /* FEAT_DIT */ cpu->isar.id_pfr0 = t; t = cpu->isar.id_pfr2; - t = FIELD_DP32(t, ID_PFR2, SSBS, 1); + t = FIELD_DP32(t, ID_PFR2, SSBS, 1); /* FEAT_SSBS */ cpu->isar.id_pfr2 = t; t = cpu->isar.id_dfr0; - t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* v8.4-PMU */ + t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* FEAT_PMUv3p4 */ cpu->isar.id_dfr0 = t; } -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Use FIELD_DP{32,64} to manipulate id_pfr1 and id_aa64pfr0 during arm_cpu_realizefn. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-11-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.c | 22 +++++++++++++--------- 1 file changed, 13 insertions(+), 9 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_realizefn(DeviceState *dev, Error **errp) */ unset_feature(env, ARM_FEATURE_EL3); - /* Disable the security extension feature bits in the processor feature - * registers as well. These are id_pfr1[7:4] and id_aa64pfr0[15:12]. + /* + * Disable the security extension feature bits in the processor + * feature registers as well. */ - cpu->isar.id_pfr1 &= ~0xf0; - cpu->isar.id_aa64pfr0 &= ~0xf000; + cpu->isar.id_pfr1 = FIELD_DP32(cpu->isar.id_pfr1, ID_PFR1, SECURITY, 0); + cpu->isar.id_aa64pfr0 = FIELD_DP64(cpu->isar.id_aa64pfr0, + ID_AA64PFR0, EL3, 0); } if (!cpu->has_el2) { @@ -XXX,XX +XXX,XX @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp) } if (!arm_feature(env, ARM_FEATURE_EL2)) { - /* Disable the hypervisor feature bits in the processor feature - * registers if we don't have EL2. These are id_pfr1[15:12] and - * id_aa64pfr0_el1[11:8]. + /* + * Disable the hypervisor feature bits in the processor feature + * registers if we don't have EL2. */ - cpu->isar.id_aa64pfr0 &= ~0xf00; - cpu->isar.id_pfr1 &= ~0xf000; + cpu->isar.id_aa64pfr0 = FIELD_DP64(cpu->isar.id_aa64pfr0, + ID_AA64PFR0, EL2, 0); + cpu->isar.id_pfr1 = FIELD_DP32(cpu->isar.id_pfr1, + ID_PFR1, VIRTUALIZATION, 0); } #ifndef CONFIG_USER_ONLY -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> The only portion of FEAT_Debugv8p2 that is relevant to QEMU is CONTEXTIDR_EL2, which is also conditionally implemented with FEAT_VHE. The rest of the debug extension concerns the External debug interface, which is outside the scope of QEMU. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-12-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu.c | 1 + target/arm/cpu64.c | 1 + target/arm/cpu_tcg.c | 2 ++ 4 files changed, 5 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_BTI (Branch Target Identification) - FEAT_DIT (Data Independent Timing instructions) - FEAT_DPB (DC CVAP instruction) +- FEAT_Debugv8p2 (Debug changes for v8.2) - FEAT_DotProd (Advanced SIMD dot product instructions) - FEAT_FCMA (Floating-point complex number instructions) - FEAT_FHM (Floating-point half-precision multiplication instructions) 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_realizefn(DeviceState *dev, Error **errp) * feature registers as well. */ cpu->isar.id_pfr1 = FIELD_DP32(cpu->isar.id_pfr1, ID_PFR1, SECURITY, 0); + cpu->isar.id_dfr0 = FIELD_DP32(cpu->isar.id_dfr0, ID_DFR0, COPSDBG, 0); cpu->isar.id_aa64pfr0 = FIELD_DP64(cpu->isar.id_aa64pfr0, ID_AA64PFR0, EL3, 0); } 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 void aarch64_max_initfn(Object *obj) cpu->isar.id_aa64zfr0 = t; t = cpu->isar.id_aa64dfr0; + t = FIELD_DP64(t, ID_AA64DFR0, DEBUGVER, 8); /* FEAT_Debugv8p2 */ t = FIELD_DP64(t, ID_AA64DFR0, PMUVER, 5); /* FEAT_PMUv3p4 */ cpu->isar.id_aa64dfr0 = t; 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 @@ void aa32_max_features(ARMCPU *cpu) cpu->isar.id_pfr2 = t; t = cpu->isar.id_dfr0; + t = FIELD_DP32(t, ID_DFR0, COPDBG, 8); /* FEAT_Debugv8p2 */ + t = FIELD_DP32(t, ID_DFR0, COPSDBG, 8); /* FEAT_Debugv8p2 */ t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* FEAT_PMUv3p4 */ cpu->isar.id_dfr0 = t; } -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> This extension concerns changes to the External Debug interface, with Secure and Non-secure access to the debug registers, and all of it is outside the scope of QEMU. Indicating support for this is mandatory with FEAT_SEL2, which we do implement. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-13-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu64.c | 2 +- target/arm/cpu_tcg.c | 4 ++-- 3 files changed, 4 insertions(+), 3 deletions(-) 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_DIT (Data Independent Timing instructions) - FEAT_DPB (DC CVAP instruction) - FEAT_Debugv8p2 (Debug changes for v8.2) +- FEAT_Debugv8p4 (Debug changes for v8.4) - FEAT_DotProd (Advanced SIMD dot product instructions) - FEAT_FCMA (Floating-point complex number instructions) - FEAT_FHM (Floating-point half-precision multiplication instructions) 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 void aarch64_max_initfn(Object *obj) cpu->isar.id_aa64zfr0 = t; t = cpu->isar.id_aa64dfr0; - t = FIELD_DP64(t, ID_AA64DFR0, DEBUGVER, 8); /* FEAT_Debugv8p2 */ + t = FIELD_DP64(t, ID_AA64DFR0, DEBUGVER, 9); /* FEAT_Debugv8p4 */ t = FIELD_DP64(t, ID_AA64DFR0, PMUVER, 5); /* FEAT_PMUv3p4 */ cpu->isar.id_aa64dfr0 = t; 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 @@ void aa32_max_features(ARMCPU *cpu) cpu->isar.id_pfr2 = t; t = cpu->isar.id_dfr0; - t = FIELD_DP32(t, ID_DFR0, COPDBG, 8); /* FEAT_Debugv8p2 */ - t = FIELD_DP32(t, ID_DFR0, COPSDBG, 8); /* FEAT_Debugv8p2 */ + t = FIELD_DP32(t, ID_DFR0, COPDBG, 9); /* FEAT_Debugv8p4 */ + t = FIELD_DP32(t, ID_DFR0, COPSDBG, 9); /* FEAT_Debugv8p4 */ t = FIELD_DP32(t, ID_DFR0, PERFMON, 5); /* FEAT_PMUv3p4 */ cpu->isar.id_dfr0 = t; } -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Add only the system registers required to implement zero error records. This means that all values for ERRSELR are out of range, which means that it and all of the indexed error record registers need not be implemented. Add the EL2 registers required for injecting virtual SError. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-14-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.h | 5 +++ target/arm/helper.c | 84 +++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 89 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 @@ typedef struct CPUArchState { uint64_t tfsr_el[4]; /* tfsre0_el1 is index 0. */ uint64_t gcr_el1; uint64_t rgsr_el1; + + /* Minimal RAS registers */ + uint64_t disr_el1; + uint64_t vdisr_el2; + uint64_t vsesr_el2; } cp15; struct { 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 debug_lpae_cp_reginfo[] = { .access = PL0_R, .type = ARM_CP_CONST|ARM_CP_64BIT, .resetvalue = 0 }, }; +/* + * Check for traps to RAS registers, which are controlled + * by HCR_EL2.TERR and SCR_EL3.TERR. + */ +static CPAccessResult access_terr(CPUARMState *env, const ARMCPRegInfo *ri, + bool isread) +{ + int el = arm_current_el(env); + + if (el < 2 && (arm_hcr_el2_eff(env) & HCR_TERR)) { + return CP_ACCESS_TRAP_EL2; + } + if (el < 3 && (env->cp15.scr_el3 & SCR_TERR)) { + return CP_ACCESS_TRAP_EL3; + } + return CP_ACCESS_OK; +} + +static uint64_t disr_read(CPUARMState *env, const ARMCPRegInfo *ri) +{ + int el = arm_current_el(env); + + if (el < 2 && (arm_hcr_el2_eff(env) & HCR_AMO)) { + return env->cp15.vdisr_el2; + } + if (el < 3 && (env->cp15.scr_el3 & SCR_EA)) { + return 0; /* RAZ/WI */ + } + return env->cp15.disr_el1; +} + +static void disr_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t val) +{ + int el = arm_current_el(env); + + if (el < 2 && (arm_hcr_el2_eff(env) & HCR_AMO)) { + env->cp15.vdisr_el2 = val; + return; + } + if (el < 3 && (env->cp15.scr_el3 & SCR_EA)) { + return; /* RAZ/WI */ + } + env->cp15.disr_el1 = val; +} + +/* + * Minimal RAS implementation with no Error Records. + * Which means that all of the Error Record registers: + * ERXADDR_EL1 + * ERXCTLR_EL1 + * ERXFR_EL1 + * ERXMISC0_EL1 + * ERXMISC1_EL1 + * ERXMISC2_EL1 + * ERXMISC3_EL1 + * ERXPFGCDN_EL1 (RASv1p1) + * ERXPFGCTL_EL1 (RASv1p1) + * ERXPFGF_EL1 (RASv1p1) + * ERXSTATUS_EL1 + * and + * ERRSELR_EL1 + * may generate UNDEFINED, which is the effect we get by not + * listing them at all. + */ +static const ARMCPRegInfo minimal_ras_reginfo[] = { + { .name = "DISR_EL1", .state = ARM_CP_STATE_BOTH, + .opc0 = 3, .opc1 = 0, .crn = 12, .crm = 1, .opc2 = 1, + .access = PL1_RW, .fieldoffset = offsetof(CPUARMState, cp15.disr_el1), + .readfn = disr_read, .writefn = disr_write, .raw_writefn = raw_write }, + { .name = "ERRIDR_EL1", .state = ARM_CP_STATE_BOTH, + .opc0 = 3, .opc1 = 0, .crn = 5, .crm = 3, .opc2 = 0, + .access = PL1_R, .accessfn = access_terr, + .type = ARM_CP_CONST, .resetvalue = 0 }, + { .name = "VDISR_EL2", .state = ARM_CP_STATE_BOTH, + .opc0 = 3, .opc1 = 4, .crn = 12, .crm = 1, .opc2 = 1, + .access = PL2_RW, .fieldoffset = offsetof(CPUARMState, cp15.vdisr_el2) }, + { .name = "VSESR_EL2", .state = ARM_CP_STATE_BOTH, + .opc0 = 3, .opc1 = 4, .crn = 5, .crm = 2, .opc2 = 3, + .access = PL2_RW, .fieldoffset = offsetof(CPUARMState, cp15.vsesr_el2) }, +}; + /* Return the exception level to which exceptions should be taken * via SVEAccessTrap. If an exception should be routed through * AArch64.AdvSIMDFPAccessTrap, return 0; fp_exception_el should @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) if (cpu_isar_feature(aa64_ssbs, cpu)) { define_one_arm_cp_reg(cpu, &ssbs_reginfo); } + if (cpu_isar_feature(any_ras, cpu)) { + define_arm_cp_regs(cpu, minimal_ras_reginfo); + } if (cpu_isar_feature(aa64_vh, cpu) || cpu_isar_feature(aa64_debugv8p2, cpu)) { -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Enable writes to the TERR and TEA bits when RAS is enabled. These bits are otherwise RES0. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-15-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 9 +++++++++ 1 file changed, 9 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 scr_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) } valid_mask &= ~SCR_NET; + if (cpu_isar_feature(aa64_ras, cpu)) { + valid_mask |= SCR_TERR; + } if (cpu_isar_feature(aa64_lor, cpu)) { valid_mask |= SCR_TLOR; } @@ -XXX,XX +XXX,XX @@ static void scr_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) } } else { valid_mask &= ~(SCR_RW | SCR_ST); + if (cpu_isar_feature(aa32_ras, cpu)) { + valid_mask |= SCR_TERR; + } } if (!arm_feature(env, ARM_FEATURE_EL2)) { @@ -XXX,XX +XXX,XX @@ static void do_hcr_write(CPUARMState *env, uint64_t value, uint64_t valid_mask) if (cpu_isar_feature(aa64_vh, cpu)) { valid_mask |= HCR_E2H; } + if (cpu_isar_feature(aa64_ras, cpu)) { + valid_mask |= HCR_TERR | HCR_TEA; + } if (cpu_isar_feature(aa64_lor, cpu)) { valid_mask |= HCR_TLOR; } -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Virtual SError exceptions are raised by setting HCR_EL2.VSE, and are routed to EL1 just like other virtual exceptions. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-16-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.h | 2 ++ target/arm/internals.h | 8 ++++++++ target/arm/syndrome.h | 5 +++++ target/arm/cpu.c | 38 +++++++++++++++++++++++++++++++++++++- target/arm/helper.c | 40 +++++++++++++++++++++++++++++++++++++++- 5 files changed, 91 insertions(+), 2 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 @@ #define EXCP_LSERR 21 /* v8M LSERR SecureFault */ #define EXCP_UNALIGNED 22 /* v7M UNALIGNED UsageFault */ #define EXCP_DIVBYZERO 23 /* v7M DIVBYZERO UsageFault */ +#define EXCP_VSERR 24 /* NB: add new EXCP_ defines to the array in arm_log_exception() too */ #define ARMV7M_EXCP_RESET 1 @@ -XXX,XX +XXX,XX @@ enum { #define CPU_INTERRUPT_FIQ CPU_INTERRUPT_TGT_EXT_1 #define CPU_INTERRUPT_VIRQ CPU_INTERRUPT_TGT_EXT_2 #define CPU_INTERRUPT_VFIQ CPU_INTERRUPT_TGT_EXT_3 +#define CPU_INTERRUPT_VSERR CPU_INTERRUPT_TGT_INT_0 /* The usual mapping for an AArch64 system register to its AArch32 * counterpart is for the 32 bit world to have access to the lower 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 @@ void arm_cpu_update_virq(ARMCPU *cpu); */ void arm_cpu_update_vfiq(ARMCPU *cpu); +/** + * arm_cpu_update_vserr: Update CPU_INTERRUPT_VSERR bit + * + * Update the CPU_INTERRUPT_VSERR bit in cs->interrupt_request, + * following a change to the HCR_EL2.VSE bit. + */ +void arm_cpu_update_vserr(ARMCPU *cpu); + /** * arm_mmu_idx_el: * @env: The cpu environment diff --git a/target/arm/syndrome.h b/target/arm/syndrome.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/syndrome.h +++ b/target/arm/syndrome.h @@ -XXX,XX +XXX,XX @@ static inline uint32_t syn_pcalignment(void) return (EC_PCALIGNMENT << ARM_EL_EC_SHIFT) | ARM_EL_IL; } +static inline uint32_t syn_serror(uint32_t extra) +{ + return (EC_SERROR << ARM_EL_EC_SHIFT) | ARM_EL_IL | extra; +} + #endif /* TARGET_ARM_SYNDROME_H */ 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 bool arm_cpu_has_work(CPUState *cs) return (cpu->power_state != PSCI_OFF) && cs->interrupt_request & (CPU_INTERRUPT_FIQ | CPU_INTERRUPT_HARD - | CPU_INTERRUPT_VFIQ | CPU_INTERRUPT_VIRQ + | CPU_INTERRUPT_VFIQ | CPU_INTERRUPT_VIRQ | CPU_INTERRUPT_VSERR | CPU_INTERRUPT_EXITTB); } @@ -XXX,XX +XXX,XX @@ static inline bool arm_excp_unmasked(CPUState *cs, unsigned int excp_idx, return false; } return !(env->daif & PSTATE_I); + case EXCP_VSERR: + if (!(hcr_el2 & HCR_AMO) || (hcr_el2 & HCR_TGE)) { + /* VIRQs are only taken when hypervized. */ + return false; + } + return !(env->daif & PSTATE_A); default: g_assert_not_reached(); } @@ -XXX,XX +XXX,XX @@ static bool arm_cpu_exec_interrupt(CPUState *cs, int interrupt_request) goto found; } } + if (interrupt_request & CPU_INTERRUPT_VSERR) { + excp_idx = EXCP_VSERR; + target_el = 1; + if (arm_excp_unmasked(cs, excp_idx, target_el, + cur_el, secure, hcr_el2)) { + /* Taking a virtual abort clears HCR_EL2.VSE */ + env->cp15.hcr_el2 &= ~HCR_VSE; + cpu_reset_interrupt(cs, CPU_INTERRUPT_VSERR); + goto found; + } + } return false; found: @@ -XXX,XX +XXX,XX @@ void arm_cpu_update_vfiq(ARMCPU *cpu) } } +void arm_cpu_update_vserr(ARMCPU *cpu) +{ + /* + * Update the interrupt level for VSERR, which is the HCR_EL2.VSE bit. + */ + CPUARMState *env = &cpu->env; + CPUState *cs = CPU(cpu); + + bool new_state = env->cp15.hcr_el2 & HCR_VSE; + + if (new_state != ((cs->interrupt_request & CPU_INTERRUPT_VSERR) != 0)) { + if (new_state) { + cpu_interrupt(cs, CPU_INTERRUPT_VSERR); + } else { + cpu_reset_interrupt(cs, CPU_INTERRUPT_VSERR); + } + } +} + #ifndef CONFIG_USER_ONLY static void arm_cpu_set_irq(void *opaque, int irq, int level) { 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 uint64_t isr_read(CPUARMState *env, const ARMCPRegInfo *ri) } } - /* External aborts are not possible in QEMU so A bit is always clear */ + if (hcr_el2 & HCR_AMO) { + if (cs->interrupt_request & CPU_INTERRUPT_VSERR) { + ret |= CPSR_A; + } + } + return ret; } @@ -XXX,XX +XXX,XX @@ static void do_hcr_write(CPUARMState *env, uint64_t value, uint64_t valid_mask) g_assert(qemu_mutex_iothread_locked()); arm_cpu_update_virq(cpu); arm_cpu_update_vfiq(cpu); + arm_cpu_update_vserr(cpu); } static void hcr_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) @@ -XXX,XX +XXX,XX @@ void arm_log_exception(CPUState *cs) [EXCP_LSERR] = "v8M LSERR UsageFault", [EXCP_UNALIGNED] = "v7M UNALIGNED UsageFault", [EXCP_DIVBYZERO] = "v7M DIVBYZERO UsageFault", + [EXCP_VSERR] = "Virtual SERR", }; if (idx >= 0 && idx < ARRAY_SIZE(excnames)) { @@ -XXX,XX +XXX,XX @@ static void arm_cpu_do_interrupt_aarch32(CPUState *cs) mask = CPSR_A | CPSR_I | CPSR_F; offset = 4; break; + case EXCP_VSERR: + { + /* + * Note that this is reported as a data abort, but the DFAR + * has an UNKNOWN value. Construct the SError syndrome from + * AET and ExT fields. + */ + ARMMMUFaultInfo fi = { .type = ARMFault_AsyncExternal, }; + + if (extended_addresses_enabled(env)) { + env->exception.fsr = arm_fi_to_lfsc(&fi); + } else { + env->exception.fsr = arm_fi_to_sfsc(&fi); + } + env->exception.fsr |= env->cp15.vsesr_el2 & 0xd000; + A32_BANKED_CURRENT_REG_SET(env, dfsr, env->exception.fsr); + qemu_log_mask(CPU_LOG_INT, "...with IFSR 0x%x\n", + env->exception.fsr); + + new_mode = ARM_CPU_MODE_ABT; + addr = 0x10; + mask = CPSR_A | CPSR_I; + offset = 8; + } + break; case EXCP_SMC: new_mode = ARM_CPU_MODE_MON; addr = 0x08; @@ -XXX,XX +XXX,XX @@ static void arm_cpu_do_interrupt_aarch64(CPUState *cs) case EXCP_VFIQ: addr += 0x100; break; + case EXCP_VSERR: + addr += 0x180; + /* Construct the SError syndrome from IDS and ISS fields. */ + env->exception.syndrome = syn_serror(env->cp15.vsesr_el2 & 0x1ffffff); + env->cp15.esr_el[new_el] = env->exception.syndrome; + break; default: cpu_abort(cs, "Unhandled exception 0x%x\n", cs->exception_index); } -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Check for and defer any pending virtual SError. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-17-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 1 + target/arm/a32.decode | 16 ++++++++------ target/arm/t32.decode | 18 ++++++++-------- target/arm/op_helper.c | 43 ++++++++++++++++++++++++++++++++++++++ target/arm/translate-a64.c | 17 +++++++++++++++ target/arm/translate.c | 23 ++++++++++++++++++++ 6 files changed, 103 insertions(+), 15 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_1(wfe, void, env) DEF_HELPER_1(yield, void, env) DEF_HELPER_1(pre_hvc, void, env) DEF_HELPER_2(pre_smc, void, env, i32) +DEF_HELPER_1(vesb, void, env) DEF_HELPER_3(cpsr_write, void, env, i32, i32) DEF_HELPER_2(cpsr_write_eret, void, env, i32) diff --git a/target/arm/a32.decode b/target/arm/a32.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/a32.decode +++ b/target/arm/a32.decode @@ -XXX,XX +XXX,XX @@ SMULTT .... 0001 0110 .... 0000 .... 1110 .... @rd0mn { { - YIELD ---- 0011 0010 0000 1111 ---- 0000 0001 - WFE ---- 0011 0010 0000 1111 ---- 0000 0010 - WFI ---- 0011 0010 0000 1111 ---- 0000 0011 + [ + YIELD ---- 0011 0010 0000 1111 ---- 0000 0001 + WFE ---- 0011 0010 0000 1111 ---- 0000 0010 + WFI ---- 0011 0010 0000 1111 ---- 0000 0011 - # TODO: Implement SEV, SEVL; may help SMP performance. - # SEV ---- 0011 0010 0000 1111 ---- 0000 0100 - # SEVL ---- 0011 0010 0000 1111 ---- 0000 0101 + # TODO: Implement SEV, SEVL; may help SMP performance. + # SEV ---- 0011 0010 0000 1111 ---- 0000 0100 + # SEVL ---- 0011 0010 0000 1111 ---- 0000 0101 + + ESB ---- 0011 0010 0000 1111 ---- 0001 0000 + ] # The canonical nop ends in 00000000, but the whole of the # rest of the space executes as nop if otherwise unsupported. diff --git a/target/arm/t32.decode b/target/arm/t32.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/t32.decode +++ b/target/arm/t32.decode @@ -XXX,XX +XXX,XX @@ CLZ 1111 1010 1011 ---- 1111 .... 1000 .... @rdm [ # Hints, and CPS { - YIELD 1111 0011 1010 1111 1000 0000 0000 0001 - WFE 1111 0011 1010 1111 1000 0000 0000 0010 - WFI 1111 0011 1010 1111 1000 0000 0000 0011 + [ + YIELD 1111 0011 1010 1111 1000 0000 0000 0001 + WFE 1111 0011 1010 1111 1000 0000 0000 0010 + WFI 1111 0011 1010 1111 1000 0000 0000 0011 - # TODO: Implement SEV, SEVL; may help SMP performance. - # SEV 1111 0011 1010 1111 1000 0000 0000 0100 - # SEVL 1111 0011 1010 1111 1000 0000 0000 0101 + # TODO: Implement SEV, SEVL; may help SMP performance. + # SEV 1111 0011 1010 1111 1000 0000 0000 0100 + # SEVL 1111 0011 1010 1111 1000 0000 0000 0101 - # For M-profile minimal-RAS ESB can be a NOP, which is the - # default behaviour since it is in the hint space. - # ESB 1111 0011 1010 1111 1000 0000 0001 0000 + ESB 1111 0011 1010 1111 1000 0000 0001 0000 + ] # The canonical nop ends in 0000 0000, but the whole rest # of the space is "reserved hint, behaves as nop". 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(probe_access)(CPUARMState *env, target_ulong ptr, access_type, mmu_idx, ra); } } + +/* + * This function corresponds to AArch64.vESBOperation(). + * Note that the AArch32 version is not functionally different. + */ +void HELPER(vesb)(CPUARMState *env) +{ + /* + * The EL2Enabled() check is done inside arm_hcr_el2_eff, + * and will return HCR_EL2.VSE == 0, so nothing happens. + */ + uint64_t hcr = arm_hcr_el2_eff(env); + bool enabled = !(hcr & HCR_TGE) && (hcr & HCR_AMO); + bool pending = enabled && (hcr & HCR_VSE); + bool masked = (env->daif & PSTATE_A); + + /* If VSE pending and masked, defer the exception. */ + if (pending && masked) { + uint32_t syndrome; + + if (arm_el_is_aa64(env, 1)) { + /* Copy across IDS and ISS from VSESR. */ + syndrome = env->cp15.vsesr_el2 & 0x1ffffff; + } else { + ARMMMUFaultInfo fi = { .type = ARMFault_AsyncExternal }; + + if (extended_addresses_enabled(env)) { + syndrome = arm_fi_to_lfsc(&fi); + } else { + syndrome = arm_fi_to_sfsc(&fi); + } + /* Copy across AET and ExT from VSESR. */ + syndrome |= env->cp15.vsesr_el2 & 0xd000; + } + + /* Set VDISR_EL2.A along with the syndrome. */ + env->cp15.vdisr_el2 = syndrome | (1u << 31); + + /* Clear pending virtual SError */ + env->cp15.hcr_el2 &= ~HCR_VSE; + cpu_reset_interrupt(env_cpu(env), CPU_INTERRUPT_VSERR); + } +} 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_hint(DisasContext *s, uint32_t insn, gen_helper_autib(cpu_X[17], cpu_env, cpu_X[17], cpu_X[16]); } break; + case 0b10000: /* ESB */ + /* Without RAS, we must implement this as NOP. */ + if (dc_isar_feature(aa64_ras, s)) { + /* + * QEMU does not have a source of physical SErrors, + * so we are only concerned with virtual SErrors. + * The pseudocode in the ARM for this case is + * if PSTATE.EL IN {EL0, EL1} && EL2Enabled() then + * AArch64.vESBOperation(); + * Most of the condition can be evaluated at translation time. + * Test for EL2 present, and defer test for SEL2 to runtime. + */ + if (s->current_el <= 1 && arm_dc_feature(s, ARM_FEATURE_EL2)) { + gen_helper_vesb(cpu_env); + } + } + break; case 0b11000: /* PACIAZ */ if (s->pauth_active) { gen_helper_pacia(cpu_X[30], cpu_env, cpu_X[30], 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 bool trans_WFI(DisasContext *s, arg_WFI *a) return true; } +static bool trans_ESB(DisasContext *s, arg_ESB *a) +{ + /* + * For M-profile, minimal-RAS ESB can be a NOP. + * Without RAS, we must implement this as NOP. + */ + if (!arm_dc_feature(s, ARM_FEATURE_M) && dc_isar_feature(aa32_ras, s)) { + /* + * QEMU does not have a source of physical SErrors, + * so we are only concerned with virtual SErrors. + * The pseudocode in the ARM for this case is + * if PSTATE.EL IN {EL0, EL1} && EL2Enabled() then + * AArch32.vESBOperation(); + * Most of the condition can be evaluated at translation time. + * Test for EL2 present, and defer test for SEL2 to runtime. + */ + if (s->current_el <= 1 && arm_dc_feature(s, ARM_FEATURE_EL2)) { + gen_helper_vesb(cpu_env); + } + } + return true; +} + static bool trans_NOP(DisasContext *s, arg_NOP *a) { return true; -- 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: 20220506180242.216785-18-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu64.c | 1 + target/arm/cpu_tcg.c | 1 + 3 files changed, 3 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_PMULL (PMULL, PMULL2 instructions) - FEAT_PMUv3p1 (PMU Extensions v3.1) - FEAT_PMUv3p4 (PMU Extensions v3.4) +- FEAT_RAS (Reliability, availability, and serviceability) - FEAT_RDM (Advanced SIMD rounding double multiply accumulate instructions) - FEAT_RNG (Random number generator) - FEAT_SB (Speculation Barrier) 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 void aarch64_max_initfn(Object *obj) t = cpu->isar.id_aa64pfr0; t = FIELD_DP64(t, ID_AA64PFR0, FP, 1); /* FEAT_FP16 */ t = FIELD_DP64(t, ID_AA64PFR0, ADVSIMD, 1); /* FEAT_FP16 */ + t = FIELD_DP64(t, ID_AA64PFR0, RAS, 1); /* FEAT_RAS */ t = FIELD_DP64(t, ID_AA64PFR0, SVE, 1); t = FIELD_DP64(t, ID_AA64PFR0, SEL2, 1); /* FEAT_SEL2 */ t = FIELD_DP64(t, ID_AA64PFR0, DIT, 1); /* FEAT_DIT */ 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 @@ void aa32_max_features(ARMCPU *cpu) t = cpu->isar.id_pfr0; t = FIELD_DP32(t, ID_PFR0, DIT, 1); /* FEAT_DIT */ + t = FIELD_DP32(t, ID_PFR0, RAS, 1); /* FEAT_RAS */ cpu->isar.id_pfr0 = t; t = cpu->isar.id_pfr2; -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> This feature is AArch64 only, and applies to physical SErrors, which QEMU does not implement, thus the feature is a nop. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-19-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/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_FlagM2 (Enhancements to flag manipulation instructions) - FEAT_HPDS (Hierarchical permission disables) - FEAT_I8MM (AArch64 Int8 matrix multiplication instructions) +- FEAT_IESB (Implicit error synchronization event) - FEAT_JSCVT (JavaScript conversion instructions) - FEAT_LOR (Limited ordering regions) - FEAT_LPA (Large Physical Address space) 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 void aarch64_max_initfn(Object *obj) t = cpu->isar.id_aa64mmfr2; t = FIELD_DP64(t, ID_AA64MMFR2, CNP, 1); /* FEAT_TTCNP */ t = FIELD_DP64(t, ID_AA64MMFR2, UAO, 1); /* FEAT_UAO */ + t = FIELD_DP64(t, ID_AA64MMFR2, IESB, 1); /* FEAT_IESB */ t = FIELD_DP64(t, ID_AA64MMFR2, VARANGE, 1); /* FEAT_LVA */ t = FIELD_DP64(t, ID_AA64MMFR2, ST, 1); /* FEAT_TTST */ t = FIELD_DP64(t, ID_AA64MMFR2, TTL, 1); /* FEAT_TTL */ -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> This extension concerns branch speculation, which TCG does not implement. Thus we can trivially enable this feature. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-20-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu64.c | 1 + target/arm/cpu_tcg.c | 1 + 3 files changed, 3 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_BBM at level 2 (Translation table break-before-make levels) - FEAT_BF16 (AArch64 BFloat16 instructions) - FEAT_BTI (Branch Target Identification) +- FEAT_CSV2 (Cache speculation variant 2) - FEAT_DIT (Data Independent Timing instructions) - FEAT_DPB (DC CVAP instruction) - FEAT_Debugv8p2 (Debug changes for v8.2) 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 void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64PFR0, SVE, 1); t = FIELD_DP64(t, ID_AA64PFR0, SEL2, 1); /* FEAT_SEL2 */ t = FIELD_DP64(t, ID_AA64PFR0, DIT, 1); /* FEAT_DIT */ + t = FIELD_DP64(t, ID_AA64PFR0, CSV2, 1); /* FEAT_CSV2 */ cpu->isar.id_aa64pfr0 = t; t = cpu->isar.id_aa64pfr1; 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 @@ void aa32_max_features(ARMCPU *cpu) cpu->isar.id_mmfr4 = t; t = cpu->isar.id_pfr0; + t = FIELD_DP32(t, ID_PFR0, CSV2, 2); /* FEAT_CVS2 */ t = FIELD_DP32(t, ID_PFR0, DIT, 1); /* FEAT_DIT */ t = FIELD_DP32(t, ID_PFR0, RAS, 1); /* FEAT_RAS */ cpu->isar.id_pfr0 = t; -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> There is no branch prediction in TCG, therefore there is no need to actually include the context number into the predictor. Therefore all we need to do is add the state for SCXTNUM_ELx. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-21-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 3 ++ target/arm/cpu.h | 16 +++++++++ target/arm/cpu.c | 5 +++ target/arm/cpu64.c | 3 +- target/arm/helper.c | 61 ++++++++++++++++++++++++++++++++++- 5 files changed, 86 insertions(+), 2 deletions(-) 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_BF16 (AArch64 BFloat16 instructions) - FEAT_BTI (Branch Target Identification) - FEAT_CSV2 (Cache speculation variant 2) +- FEAT_CSV2_1p1 (Cache speculation variant 2, version 1.1) +- FEAT_CSV2_1p2 (Cache speculation variant 2, version 1.2) +- FEAT_CSV2_2 (Cache speculation variant 2, version 2) - FEAT_DIT (Data Independent Timing instructions) - FEAT_DPB (DC CVAP instruction) - FEAT_Debugv8p2 (Debug changes for v8.2) 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 { ARMPACKey apdb; ARMPACKey apga; } keys; + + uint64_t scxtnum_el[4]; #endif #if defined(CONFIG_USER_ONLY) @@ -XXX,XX +XXX,XX @@ void pmu_init(ARMCPU *cpu); #define SCTLR_WXN (1U << 19) #define SCTLR_ST (1U << 20) /* up to ??, RAZ in v6 */ #define SCTLR_UWXN (1U << 20) /* v7 onward, AArch32 only */ +#define SCTLR_TSCXT (1U << 20) /* FEAT_CSV2_1p2, AArch64 only */ #define SCTLR_FI (1U << 21) /* up to v7, v8 RES0 */ #define SCTLR_IESB (1U << 21) /* v8.2-IESB, AArch64 only */ #define SCTLR_U (1U << 22) /* up to v6, RAO in v7 */ @@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_dit(const ARMISARegisters *id) return FIELD_EX64(id->id_aa64pfr0, ID_AA64PFR0, DIT) != 0; } +static inline bool isar_feature_aa64_scxtnum(const ARMISARegisters *id) +{ + int key = FIELD_EX64(id->id_aa64pfr0, ID_AA64PFR0, CSV2); + if (key >= 2) { + return true; /* FEAT_CSV2_2 */ + } + if (key == 1) { + key = FIELD_EX64(id->id_aa64pfr1, ID_AA64PFR1, CSV2_FRAC); + return key >= 2; /* FEAT_CSV2_1p2 */ + } + return false; +} + static inline bool isar_feature_aa64_ssbs(const ARMISARegisters *id) { return FIELD_EX64(id->id_aa64pfr1, ID_AA64PFR1, SSBS) != 0; 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) */ env->cp15.gcr_el1 = 0x1ffff; } + /* + * Disable access to SCXTNUM_EL0 from CSV2_1p2. + * This is not yet exposed from the Linux kernel in any way. + */ + env->cp15.sctlr_el[1] |= SCTLR_TSCXT; #else /* Reset into the highest available EL */ if (arm_feature(env, ARM_FEATURE_EL3)) { 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 void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64PFR0, SVE, 1); t = FIELD_DP64(t, ID_AA64PFR0, SEL2, 1); /* FEAT_SEL2 */ t = FIELD_DP64(t, ID_AA64PFR0, DIT, 1); /* FEAT_DIT */ - t = FIELD_DP64(t, ID_AA64PFR0, CSV2, 1); /* FEAT_CSV2 */ + t = FIELD_DP64(t, ID_AA64PFR0, CSV2, 2); /* FEAT_CSV2_2 */ cpu->isar.id_aa64pfr0 = t; t = cpu->isar.id_aa64pfr1; @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) * we do for EL2 with the virtualization=on property. */ t = FIELD_DP64(t, ID_AA64PFR1, MTE, 3); /* FEAT_MTE3 */ + t = FIELD_DP64(t, ID_AA64PFR1, CSV2_FRAC, 0); /* FEAT_CSV2_2 */ cpu->isar.id_aa64pfr1 = t; t = cpu->isar.id_aa64mmfr0; 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_mte, cpu)) { valid_mask |= SCR_ATA; } + if (cpu_isar_feature(aa64_scxtnum, cpu)) { + valid_mask |= SCR_ENSCXT; + } } else { valid_mask &= ~(SCR_RW | SCR_ST); if (cpu_isar_feature(aa32_ras, cpu)) { @@ -XXX,XX +XXX,XX @@ static void do_hcr_write(CPUARMState *env, uint64_t value, uint64_t valid_mask) if (cpu_isar_feature(aa64_mte, cpu)) { valid_mask |= HCR_ATA | HCR_DCT | HCR_TID5; } + if (cpu_isar_feature(aa64_scxtnum, cpu)) { + valid_mask |= HCR_ENSCXT; + } } /* Clear RES0 bits. */ @@ -XXX,XX +XXX,XX @@ static void define_arm_vh_e2h_redirects_aliases(ARMCPU *cpu) { K(3, 0, 5, 6, 0), K(3, 4, 5, 6, 0), K(3, 5, 5, 6, 0), "TFSR_EL1", "TFSR_EL2", "TFSR_EL12", isar_feature_aa64_mte }, + { K(3, 0, 13, 0, 7), K(3, 4, 13, 0, 7), K(3, 5, 13, 0, 7), + "SCXTNUM_EL1", "SCXTNUM_EL2", "SCXTNUM_EL12", + isar_feature_aa64_scxtnum }, + /* TODO: ARMv8.2-SPE -- PMSCR_EL2 */ /* TODO: ARMv8.4-Trace -- TRFCR_EL2 */ }; @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo mte_el0_cacheop_reginfo[] = { }, }; -#endif +static CPAccessResult access_scxtnum(CPUARMState *env, const ARMCPRegInfo *ri, + bool isread) +{ + uint64_t hcr = arm_hcr_el2_eff(env); + int el = arm_current_el(env); + + if (el == 0 && !((hcr & HCR_E2H) && (hcr & HCR_TGE))) { + if (env->cp15.sctlr_el[1] & SCTLR_TSCXT) { + if (hcr & HCR_TGE) { + return CP_ACCESS_TRAP_EL2; + } + return CP_ACCESS_TRAP; + } + } else if (el < 2 && (env->cp15.sctlr_el[2] & SCTLR_TSCXT)) { + return CP_ACCESS_TRAP_EL2; + } + if (el < 2 && arm_is_el2_enabled(env) && !(hcr & HCR_ENSCXT)) { + return CP_ACCESS_TRAP_EL2; + } + if (el < 3 + && arm_feature(env, ARM_FEATURE_EL3) + && !(env->cp15.scr_el3 & SCR_ENSCXT)) { + return CP_ACCESS_TRAP_EL3; + } + return CP_ACCESS_OK; +} + +static const ARMCPRegInfo scxtnum_reginfo[] = { + { .name = "SCXTNUM_EL0", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 3, .crn = 13, .crm = 0, .opc2 = 7, + .access = PL0_RW, .accessfn = access_scxtnum, + .fieldoffset = offsetof(CPUARMState, scxtnum_el[0]) }, + { .name = "SCXTNUM_EL1", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 0, .crn = 13, .crm = 0, .opc2 = 7, + .access = PL1_RW, .accessfn = access_scxtnum, + .fieldoffset = offsetof(CPUARMState, scxtnum_el[1]) }, + { .name = "SCXTNUM_EL2", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 4, .crn = 13, .crm = 0, .opc2 = 7, + .access = PL2_RW, .accessfn = access_scxtnum, + .fieldoffset = offsetof(CPUARMState, scxtnum_el[2]) }, + { .name = "SCXTNUM_EL3", .state = ARM_CP_STATE_AA64, + .opc0 = 3, .opc1 = 6, .crn = 13, .crm = 0, .opc2 = 7, + .access = PL3_RW, + .fieldoffset = offsetof(CPUARMState, scxtnum_el[3]) }, +}; +#endif /* TARGET_AARCH64 */ static CPAccessResult access_predinv(CPUARMState *env, const ARMCPRegInfo *ri, bool isread) @@ -XXX,XX +XXX,XX @@ void register_cp_regs_for_features(ARMCPU *cpu) define_arm_cp_regs(cpu, mte_tco_ro_reginfo); define_arm_cp_regs(cpu, mte_el0_cacheop_reginfo); } + + if (cpu_isar_feature(aa64_scxtnum, cpu)) { + define_arm_cp_regs(cpu, scxtnum_reginfo); + } #endif if (cpu_isar_feature(any_predinv, cpu)) { -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> This extension concerns cache speculation, which TCG does not implement. Thus we can trivially enable this feature. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-22-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu64.c | 1 + target/arm/cpu_tcg.c | 1 + 3 files changed, 3 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_CSV2_1p1 (Cache speculation variant 2, version 1.1) - FEAT_CSV2_1p2 (Cache speculation variant 2, version 1.2) - FEAT_CSV2_2 (Cache speculation variant 2, version 2) +- FEAT_CSV3 (Cache speculation variant 3) - FEAT_DIT (Data Independent Timing instructions) - FEAT_DPB (DC CVAP instruction) - FEAT_Debugv8p2 (Debug changes for v8.2) 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 void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64PFR0, SEL2, 1); /* FEAT_SEL2 */ t = FIELD_DP64(t, ID_AA64PFR0, DIT, 1); /* FEAT_DIT */ t = FIELD_DP64(t, ID_AA64PFR0, CSV2, 2); /* FEAT_CSV2_2 */ + t = FIELD_DP64(t, ID_AA64PFR0, CSV3, 1); /* FEAT_CSV3 */ cpu->isar.id_aa64pfr0 = t; t = cpu->isar.id_aa64pfr1; 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 @@ void aa32_max_features(ARMCPU *cpu) cpu->isar.id_pfr0 = t; t = cpu->isar.id_pfr2; + t = FIELD_DP32(t, ID_PFR2, CSV3, 1); /* FEAT_CSV3 */ t = FIELD_DP32(t, ID_PFR2, SSBS, 1); /* FEAT_SSBS */ cpu->isar.id_pfr2 = t; -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> This extension concerns not merging memory access, which TCG does not implement. Thus we can trivially enable this feature. Add a comment to handle_hint for the DGH instruction, but no code. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-23-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/emulation.rst | 1 + target/arm/cpu64.c | 1 + target/arm/translate-a64.c | 1 + 3 files changed, 3 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_CSV2_1p2 (Cache speculation variant 2, version 1.2) - FEAT_CSV2_2 (Cache speculation variant 2, version 2) - FEAT_CSV3 (Cache speculation variant 3) +- FEAT_DGH (Data gathering hint) - FEAT_DIT (Data Independent Timing instructions) - FEAT_DPB (DC CVAP instruction) - FEAT_Debugv8p2 (Debug changes for v8.2) 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 void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64ISAR1, SB, 1); /* FEAT_SB */ t = FIELD_DP64(t, ID_AA64ISAR1, SPECRES, 1); /* FEAT_SPECRES */ t = FIELD_DP64(t, ID_AA64ISAR1, BF16, 1); /* FEAT_BF16 */ + t = FIELD_DP64(t, ID_AA64ISAR1, DGH, 1); /* FEAT_DGH */ t = FIELD_DP64(t, ID_AA64ISAR1, I8MM, 1); /* FEAT_I8MM */ cpu->isar.id_aa64isar1 = t; 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_hint(DisasContext *s, uint32_t insn, break; case 0b00100: /* SEV */ case 0b00101: /* SEVL */ + case 0b00110: /* DGH */ /* we treat all as NOP at least for now */ break; case 0b00111: /* XPACLRI */ -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Enable the a76 for virt and sbsa board use. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-24-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/virt.rst | 1 + hw/arm/sbsa-ref.c | 1 + hw/arm/virt.c | 1 + target/arm/cpu64.c | 66 ++++++++++++++++++++++++++++++++++++++++ 4 files changed, 69 insertions(+) diff --git a/docs/system/arm/virt.rst b/docs/system/arm/virt.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/virt.rst +++ b/docs/system/arm/virt.rst @@ -XXX,XX +XXX,XX @@ Supported guest CPU types: - ``cortex-a53`` (64-bit) - ``cortex-a57`` (64-bit) - ``cortex-a72`` (64-bit) +- ``cortex-a76`` (64-bit) - ``a64fx`` (64-bit) - ``host`` (with KVM only) - ``max`` (same as ``host`` for KVM; best possible emulation with TCG) diff --git a/hw/arm/sbsa-ref.c b/hw/arm/sbsa-ref.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/sbsa-ref.c +++ b/hw/arm/sbsa-ref.c @@ -XXX,XX +XXX,XX @@ static const int sbsa_ref_irqmap[] = { static const char * const valid_cpus[] = { ARM_CPU_TYPE_NAME("cortex-a57"), ARM_CPU_TYPE_NAME("cortex-a72"), + ARM_CPU_TYPE_NAME("cortex-a76"), ARM_CPU_TYPE_NAME("max"), }; diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ static const char *valid_cpus[] = { ARM_CPU_TYPE_NAME("cortex-a53"), ARM_CPU_TYPE_NAME("cortex-a57"), ARM_CPU_TYPE_NAME("cortex-a72"), + ARM_CPU_TYPE_NAME("cortex-a76"), ARM_CPU_TYPE_NAME("a64fx"), ARM_CPU_TYPE_NAME("host"), ARM_CPU_TYPE_NAME("max"), 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 void aarch64_a72_initfn(Object *obj) define_cortex_a72_a57_a53_cp_reginfo(cpu); } +static void aarch64_a76_initfn(Object *obj) +{ + ARMCPU *cpu = ARM_CPU(obj); + + cpu->dtb_compatible = "arm,cortex-a76"; + set_feature(&cpu->env, ARM_FEATURE_V8); + set_feature(&cpu->env, ARM_FEATURE_NEON); + set_feature(&cpu->env, ARM_FEATURE_GENERIC_TIMER); + set_feature(&cpu->env, ARM_FEATURE_AARCH64); + set_feature(&cpu->env, ARM_FEATURE_CBAR_RO); + set_feature(&cpu->env, ARM_FEATURE_EL2); + set_feature(&cpu->env, ARM_FEATURE_EL3); + set_feature(&cpu->env, ARM_FEATURE_PMU); + + /* Ordered by B2.4 AArch64 registers by functional group */ + cpu->clidr = 0x82000023; + cpu->ctr = 0x8444C004; + cpu->dcz_blocksize = 4; + cpu->isar.id_aa64dfr0 = 0x0000000010305408ull; + cpu->isar.id_aa64isar0 = 0x0000100010211120ull; + cpu->isar.id_aa64isar1 = 0x0000000000100001ull; + cpu->isar.id_aa64mmfr0 = 0x0000000000101122ull; + cpu->isar.id_aa64mmfr1 = 0x0000000010212122ull; + cpu->isar.id_aa64mmfr2 = 0x0000000000001011ull; + cpu->isar.id_aa64pfr0 = 0x1100000010111112ull; /* GIC filled in later */ + cpu->isar.id_aa64pfr1 = 0x0000000000000010ull; + cpu->id_afr0 = 0x00000000; + cpu->isar.id_dfr0 = 0x04010088; + cpu->isar.id_isar0 = 0x02101110; + cpu->isar.id_isar1 = 0x13112111; + cpu->isar.id_isar2 = 0x21232042; + cpu->isar.id_isar3 = 0x01112131; + cpu->isar.id_isar4 = 0x00010142; + cpu->isar.id_isar5 = 0x01011121; + cpu->isar.id_isar6 = 0x00000010; + cpu->isar.id_mmfr0 = 0x10201105; + cpu->isar.id_mmfr1 = 0x40000000; + cpu->isar.id_mmfr2 = 0x01260000; + cpu->isar.id_mmfr3 = 0x02122211; + cpu->isar.id_mmfr4 = 0x00021110; + cpu->isar.id_pfr0 = 0x10010131; + cpu->isar.id_pfr1 = 0x00010000; /* GIC filled in later */ + cpu->isar.id_pfr2 = 0x00000011; + cpu->midr = 0x414fd0b1; /* r4p1 */ + cpu->revidr = 0; + + /* From B2.18 CCSIDR_EL1 */ + cpu->ccsidr[0] = 0x701fe01a; /* 64KB L1 dcache */ + cpu->ccsidr[1] = 0x201fe01a; /* 64KB L1 icache */ + cpu->ccsidr[2] = 0x707fe03a; /* 512KB L2 cache */ + + /* From B2.93 SCTLR_EL3 */ + cpu->reset_sctlr = 0x30c50838; + + /* From B4.23 ICH_VTR_EL2 */ + cpu->gic_num_lrs = 4; + cpu->gic_vpribits = 5; + cpu->gic_vprebits = 5; + + /* From B5.1 AdvSIMD AArch64 register summary */ + cpu->isar.mvfr0 = 0x10110222; + cpu->isar.mvfr1 = 0x13211111; + cpu->isar.mvfr2 = 0x00000043; +} + void arm_cpu_sve_finalize(ARMCPU *cpu, Error **errp) { /* @@ -XXX,XX +XXX,XX @@ static const ARMCPUInfo aarch64_cpus[] = { { .name = "cortex-a57", .initfn = aarch64_a57_initfn }, { .name = "cortex-a53", .initfn = aarch64_a53_initfn }, { .name = "cortex-a72", .initfn = aarch64_a72_initfn }, + { .name = "cortex-a76", .initfn = aarch64_a76_initfn }, { .name = "a64fx", .initfn = aarch64_a64fx_initfn }, { .name = "max", .initfn = aarch64_max_initfn }, #if defined(CONFIG_KVM) || defined(CONFIG_HVF) -- 2.25.1 From: Richard Henderson <richard.henderson@linaro.org> Enable the n1 for virt and sbsa board use. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20220506180242.216785-25-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/virt.rst | 1 + hw/arm/sbsa-ref.c | 1 + hw/arm/virt.c | 1 + target/arm/cpu64.c | 66 ++++++++++++++++++++++++++++++++++++++++ 4 files changed, 69 insertions(+) diff --git a/docs/system/arm/virt.rst b/docs/system/arm/virt.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/virt.rst +++ b/docs/system/arm/virt.rst @@ -XXX,XX +XXX,XX @@ Supported guest CPU types: - ``cortex-a76`` (64-bit) - ``a64fx`` (64-bit) - ``host`` (with KVM only) +- ``neoverse-n1`` (64-bit) - ``max`` (same as ``host`` for KVM; best possible emulation with TCG) Note that the default is ``cortex-a15``, so for an AArch64 guest you must diff --git a/hw/arm/sbsa-ref.c b/hw/arm/sbsa-ref.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/sbsa-ref.c +++ b/hw/arm/sbsa-ref.c @@ -XXX,XX +XXX,XX @@ static const char * const valid_cpus[] = { ARM_CPU_TYPE_NAME("cortex-a57"), ARM_CPU_TYPE_NAME("cortex-a72"), ARM_CPU_TYPE_NAME("cortex-a76"), + ARM_CPU_TYPE_NAME("neoverse-n1"), ARM_CPU_TYPE_NAME("max"), }; diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ static const char *valid_cpus[] = { ARM_CPU_TYPE_NAME("cortex-a72"), ARM_CPU_TYPE_NAME("cortex-a76"), ARM_CPU_TYPE_NAME("a64fx"), + ARM_CPU_TYPE_NAME("neoverse-n1"), ARM_CPU_TYPE_NAME("host"), ARM_CPU_TYPE_NAME("max"), }; 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 void aarch64_a76_initfn(Object *obj) cpu->isar.mvfr2 = 0x00000043; } +static void aarch64_neoverse_n1_initfn(Object *obj) +{ + ARMCPU *cpu = ARM_CPU(obj); + + cpu->dtb_compatible = "arm,neoverse-n1"; + set_feature(&cpu->env, ARM_FEATURE_V8); + set_feature(&cpu->env, ARM_FEATURE_NEON); + set_feature(&cpu->env, ARM_FEATURE_GENERIC_TIMER); + set_feature(&cpu->env, ARM_FEATURE_AARCH64); + set_feature(&cpu->env, ARM_FEATURE_CBAR_RO); + set_feature(&cpu->env, ARM_FEATURE_EL2); + set_feature(&cpu->env, ARM_FEATURE_EL3); + set_feature(&cpu->env, ARM_FEATURE_PMU); + + /* Ordered by B2.4 AArch64 registers by functional group */ + cpu->clidr = 0x82000023; + cpu->ctr = 0x8444c004; + cpu->dcz_blocksize = 4; + cpu->isar.id_aa64dfr0 = 0x0000000110305408ull; + cpu->isar.id_aa64isar0 = 0x0000100010211120ull; + cpu->isar.id_aa64isar1 = 0x0000000000100001ull; + cpu->isar.id_aa64mmfr0 = 0x0000000000101125ull; + cpu->isar.id_aa64mmfr1 = 0x0000000010212122ull; + cpu->isar.id_aa64mmfr2 = 0x0000000000001011ull; + cpu->isar.id_aa64pfr0 = 0x1100000010111112ull; /* GIC filled in later */ + cpu->isar.id_aa64pfr1 = 0x0000000000000020ull; + cpu->id_afr0 = 0x00000000; + cpu->isar.id_dfr0 = 0x04010088; + cpu->isar.id_isar0 = 0x02101110; + cpu->isar.id_isar1 = 0x13112111; + cpu->isar.id_isar2 = 0x21232042; + cpu->isar.id_isar3 = 0x01112131; + cpu->isar.id_isar4 = 0x00010142; + cpu->isar.id_isar5 = 0x01011121; + cpu->isar.id_isar6 = 0x00000010; + cpu->isar.id_mmfr0 = 0x10201105; + cpu->isar.id_mmfr1 = 0x40000000; + cpu->isar.id_mmfr2 = 0x01260000; + cpu->isar.id_mmfr3 = 0x02122211; + cpu->isar.id_mmfr4 = 0x00021110; + cpu->isar.id_pfr0 = 0x10010131; + cpu->isar.id_pfr1 = 0x00010000; /* GIC filled in later */ + cpu->isar.id_pfr2 = 0x00000011; + cpu->midr = 0x414fd0c1; /* r4p1 */ + cpu->revidr = 0; + + /* From B2.23 CCSIDR_EL1 */ + cpu->ccsidr[0] = 0x701fe01a; /* 64KB L1 dcache */ + cpu->ccsidr[1] = 0x201fe01a; /* 64KB L1 icache */ + cpu->ccsidr[2] = 0x70ffe03a; /* 1MB L2 cache */ + + /* From B2.98 SCTLR_EL3 */ + cpu->reset_sctlr = 0x30c50838; + + /* From B4.23 ICH_VTR_EL2 */ + cpu->gic_num_lrs = 4; + cpu->gic_vpribits = 5; + cpu->gic_vprebits = 5; + + /* From B5.1 AdvSIMD AArch64 register summary */ + cpu->isar.mvfr0 = 0x10110222; + cpu->isar.mvfr1 = 0x13211111; + cpu->isar.mvfr2 = 0x00000043; +} + void arm_cpu_sve_finalize(ARMCPU *cpu, Error **errp) { /* @@ -XXX,XX +XXX,XX @@ static const ARMCPUInfo aarch64_cpus[] = { { .name = "cortex-a72", .initfn = aarch64_a72_initfn }, { .name = "cortex-a76", .initfn = aarch64_a76_initfn }, { .name = "a64fx", .initfn = aarch64_a64fx_initfn }, + { .name = "neoverse-n1", .initfn = aarch64_neoverse_n1_initfn }, { .name = "max", .initfn = aarch64_max_initfn }, #if defined(CONFIG_KVM) || defined(CONFIG_HVF) { .name = "host", .initfn = aarch64_host_initfn }, -- 2.25.1 From: Leif Lindholm <quic_llindhol@quicinc.com> The sbsa-ref machine is continuously evolving. Some of the changes we want to make in the near future, to align with real components (e.g. the GIC-700), will break compatibility for existing firmware. Introduce two new properties to the DT generated on machine generation: - machine-version-major To be incremented when a platform change makes the machine incompatible with existing firmware. - machine-version-minor To be incremented when functionality is added to the machine without causing incompatibility with existing firmware. to be reset to 0 when machine-version-major is incremented. This versioning scheme is *neither*: - A QEMU versioned machine type; a given version of QEMU will emulate a given version of the platform. - A reflection of level of SBSA (now SystemReady SR) support provided. The version will increment on guest-visible functional changes only, akin to a revision ID register found on a physical platform. These properties are both introduced with the value 0. (Hence, a machine where the DT is lacking these nodes is equivalent to version 0.0.) Signed-off-by: Leif Lindholm <quic_llindhol@quicinc.com> Message-id: 20220505113947.75714-1-quic_llindhol@quicinc.com Cc: Peter Maydell <peter.maydell@linaro.org> Cc: Radoslaw Biernacki <rad@semihalf.com> Cc: Cédric Le Goater <clg@kaod.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/sbsa-ref.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/hw/arm/sbsa-ref.c b/hw/arm/sbsa-ref.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/sbsa-ref.c +++ b/hw/arm/sbsa-ref.c @@ -XXX,XX +XXX,XX @@ static void create_fdt(SBSAMachineState *sms) qemu_fdt_setprop_cell(fdt, "/", "#address-cells", 0x2); qemu_fdt_setprop_cell(fdt, "/", "#size-cells", 0x2); + /* + * This versioning scheme is for informing platform fw only. It is neither: + * - A QEMU versioned machine type; a given version of QEMU will emulate + * a given version of the platform. + * - A reflection of level of SBSA (now SystemReady SR) support provided. + * + * machine-version-major: updated when changes breaking fw compatibility + * are introduced. + * machine-version-minor: updated when features are added that don't break + * fw compatibility. + */ + qemu_fdt_setprop_cell(fdt, "/", "machine-version-major", 0); + qemu_fdt_setprop_cell(fdt, "/", "machine-version-minor", 0); + if (ms->numa_state->have_numa_distance) { int size = nb_numa_nodes * nb_numa_nodes * 3 * sizeof(uint32_t); uint32_t *matrix = g_malloc0(size); -- 2.25.1 From: Gavin Shan <gshan@redhat.com> This adds cluster-id in CPU instance properties, which will be used by arm/virt machine. Besides, the cluster-id is also verified or dumped in various spots: * hw/core/machine.c::machine_set_cpu_numa_node() to associate CPU with its NUMA node. * hw/core/machine.c::machine_numa_finish_cpu_init() to record CPU slots with no NUMA mapping set. * hw/core/machine-hmp-cmds.c::hmp_hotpluggable_cpus() to dump cluster-id. Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Yanan Wang <wangyanan55@huawei.com> Acked-by: Igor Mammedov <imammedo@redhat.com> Message-id: 20220503140304.855514-2-gshan@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- qapi/machine.json | 6 ++++-- hw/core/machine-hmp-cmds.c | 4 ++++ hw/core/machine.c | 16 ++++++++++++++++ 3 files changed, 24 insertions(+), 2 deletions(-) diff --git a/qapi/machine.json b/qapi/machine.json index XXXXXXX..XXXXXXX 100644 --- a/qapi/machine.json +++ b/qapi/machine.json @@ -XXX,XX +XXX,XX @@ # @node-id: NUMA node ID the CPU belongs to # @socket-id: socket number within node/board the CPU belongs to # @die-id: die number within socket the CPU belongs to (since 4.1) -# @core-id: core number within die the CPU belongs to +# @cluster-id: cluster number within die the CPU belongs to (since 7.1) +# @core-id: core number within cluster the CPU belongs to # @thread-id: thread number within core the CPU belongs to # -# Note: currently there are 5 properties that could be present +# Note: currently there are 6 properties that could be present # but management should be prepared to pass through other # properties with device_add command to allow for future # interface extension. This also requires the filed names to be kept in @@ -XXX,XX +XXX,XX @@ 'data': { '*node-id': 'int', '*socket-id': 'int', '*die-id': 'int', + '*cluster-id': 'int', '*core-id': 'int', '*thread-id': 'int' } diff --git a/hw/core/machine-hmp-cmds.c b/hw/core/machine-hmp-cmds.c index XXXXXXX..XXXXXXX 100644 --- a/hw/core/machine-hmp-cmds.c +++ b/hw/core/machine-hmp-cmds.c @@ -XXX,XX +XXX,XX @@ void hmp_hotpluggable_cpus(Monitor *mon, const QDict *qdict) if (c->has_die_id) { monitor_printf(mon, " die-id: \"%" PRIu64 "\"\n", c->die_id); } + if (c->has_cluster_id) { + monitor_printf(mon, " cluster-id: \"%" PRIu64 "\"\n", + c->cluster_id); + } if (c->has_core_id) { monitor_printf(mon, " core-id: \"%" PRIu64 "\"\n", c->core_id); } diff --git a/hw/core/machine.c b/hw/core/machine.c index XXXXXXX..XXXXXXX 100644 --- a/hw/core/machine.c +++ b/hw/core/machine.c @@ -XXX,XX +XXX,XX @@ void machine_set_cpu_numa_node(MachineState *machine, return; } + if (props->has_cluster_id && !slot->props.has_cluster_id) { + error_setg(errp, "cluster-id is not supported"); + return; + } + if (props->has_socket_id && !slot->props.has_socket_id) { error_setg(errp, "socket-id is not supported"); return; @@ -XXX,XX +XXX,XX @@ void machine_set_cpu_numa_node(MachineState *machine, continue; } + if (props->has_cluster_id && + props->cluster_id != slot->props.cluster_id) { + continue; + } + if (props->has_die_id && props->die_id != slot->props.die_id) { continue; } @@ -XXX,XX +XXX,XX @@ static char *cpu_slot_to_string(const CPUArchId *cpu) } g_string_append_printf(s, "die-id: %"PRId64, cpu->props.die_id); } + if (cpu->props.has_cluster_id) { + if (s->len) { + g_string_append_printf(s, ", "); + } + g_string_append_printf(s, "cluster-id: %"PRId64, cpu->props.cluster_id); + } if (cpu->props.has_core_id) { if (s->len) { g_string_append_printf(s, ", "); -- 2.25.1 From: Gavin Shan <gshan@redhat.com> The CPU topology isn't enabled on arm/virt machine yet, but we're going to do it in next patch. After the CPU topology is enabled by next patch, "thread-id=1" becomes invalid because the CPU core is preferred on arm/virt machine. It means these two CPUs have 0/1 as their core IDs, but their thread IDs are all 0. It will trigger test failure as the following message indicates: [14/21 qemu:qtest+qtest-aarch64 / qtest-aarch64/numa-test ERROR 1.48s killed by signal 6 SIGABRT >>> G_TEST_DBUS_DAEMON=/home/gavin/sandbox/qemu.main/tests/dbus-vmstate-daemon.sh \ QTEST_QEMU_STORAGE_DAEMON_BINARY=./storage-daemon/qemu-storage-daemon \ QTEST_QEMU_BINARY=./qemu-system-aarch64 \ QTEST_QEMU_IMG=./qemu-img MALLOC_PERTURB_=83 \ /home/gavin/sandbox/qemu.main/build/tests/qtest/numa-test --tap -k ―――――――――――――――――――――――――――――――――――――――――――――― stderr: qemu-system-aarch64: -numa cpu,node-id=0,thread-id=1: no match found This fixes the issue by providing comprehensive SMP configurations in aarch64_numa_cpu(). The SMP configurations aren't used before the CPU topology is enabled in next patch. Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Yanan Wang <wangyanan55@huawei.com> Message-id: 20220503140304.855514-3-gshan@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/qtest/numa-test.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/qtest/numa-test.c b/tests/qtest/numa-test.c index XXXXXXX..XXXXXXX 100644 --- a/tests/qtest/numa-test.c +++ b/tests/qtest/numa-test.c @@ -XXX,XX +XXX,XX @@ static void aarch64_numa_cpu(const void *data) QTestState *qts; g_autofree char *cli = NULL; - cli = make_cli(data, "-machine smp.cpus=2 " + cli = make_cli(data, "-machine " + "smp.cpus=2,smp.sockets=1,smp.clusters=1,smp.cores=1,smp.threads=2 " "-numa node,nodeid=0,memdev=ram -numa node,nodeid=1 " "-numa cpu,node-id=1,thread-id=0 " "-numa cpu,node-id=0,thread-id=1"); -- 2.25.1 From: Gavin Shan <gshan@redhat.com> Currently, the SMP configuration isn't considered when the CPU topology is populated. In this case, it's impossible to provide the default CPU-to-NUMA mapping or association based on the socket ID of the given CPU. This takes account of SMP configuration when the CPU topology is populated. The die ID for the given CPU isn't assigned since it's not supported on arm/virt machine. Besides, the used SMP configuration in qtest/numa-test/aarch64_numa_cpu() is corrcted to avoid testing failure Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Yanan Wang <wangyanan55@huawei.com> Acked-by: Igor Mammedov <imammedo@redhat.com> Message-id: 20220503140304.855514-4-gshan@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/virt.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ static const CPUArchIdList *virt_possible_cpu_arch_ids(MachineState *ms) int n; unsigned int max_cpus = ms->smp.max_cpus; VirtMachineState *vms = VIRT_MACHINE(ms); + MachineClass *mc = MACHINE_GET_CLASS(vms); if (ms->possible_cpus) { assert(ms->possible_cpus->len == max_cpus); @@ -XXX,XX +XXX,XX @@ static const CPUArchIdList *virt_possible_cpu_arch_ids(MachineState *ms) ms->possible_cpus->cpus[n].type = ms->cpu_type; ms->possible_cpus->cpus[n].arch_id = virt_cpu_mp_affinity(vms, n); + + assert(!mc->smp_props.dies_supported); + ms->possible_cpus->cpus[n].props.has_socket_id = true; + ms->possible_cpus->cpus[n].props.socket_id = + n / (ms->smp.clusters * ms->smp.cores * ms->smp.threads); + ms->possible_cpus->cpus[n].props.has_cluster_id = true; + ms->possible_cpus->cpus[n].props.cluster_id = + (n / (ms->smp.cores * ms->smp.threads)) % ms->smp.clusters; + ms->possible_cpus->cpus[n].props.has_core_id = true; + ms->possible_cpus->cpus[n].props.core_id = + (n / ms->smp.threads) % ms->smp.cores; ms->possible_cpus->cpus[n].props.has_thread_id = true; - ms->possible_cpus->cpus[n].props.thread_id = n; + ms->possible_cpus->cpus[n].props.thread_id = + n % ms->smp.threads; } return ms->possible_cpus; } -- 2.25.1 From: Gavin Shan <gshan@redhat.com> In aarch64_numa_cpu(), the CPU and NUMA association is something like below. Two threads in the same core/cluster/socket are associated with two individual NUMA nodes, which is unreal as Igor Mammedov mentioned. We don't expect the association to break NUMA-to-socket boundary, which matches with the real world. NUMA-node socket cluster core thread ------------------------------------------ 0 0 0 0 0 1 0 0 0 1 This corrects the topology for CPUs and their association with NUMA nodes. After this patch is applied, the CPU and NUMA association becomes something like below, which looks real. Besides, socket/cluster/core/thread IDs are all checked when the NUMA node IDs are verified. It helps to check if the CPU topology is properly populated or not. NUMA-node socket cluster core thread ------------------------------------------ 0 1 0 0 0 1 0 0 0 0 Suggested-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Gavin Shan <gshan@redhat.com> Acked-by: Igor Mammedov <imammedo@redhat.com> Message-id: 20220503140304.855514-5-gshan@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/qtest/numa-test.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/tests/qtest/numa-test.c b/tests/qtest/numa-test.c index XXXXXXX..XXXXXXX 100644 --- a/tests/qtest/numa-test.c +++ b/tests/qtest/numa-test.c @@ -XXX,XX +XXX,XX @@ static void aarch64_numa_cpu(const void *data) g_autofree char *cli = NULL; cli = make_cli(data, "-machine " - "smp.cpus=2,smp.sockets=1,smp.clusters=1,smp.cores=1,smp.threads=2 " + "smp.cpus=2,smp.sockets=2,smp.clusters=1,smp.cores=1,smp.threads=1 " "-numa node,nodeid=0,memdev=ram -numa node,nodeid=1 " - "-numa cpu,node-id=1,thread-id=0 " - "-numa cpu,node-id=0,thread-id=1"); + "-numa cpu,node-id=0,socket-id=1,cluster-id=0,core-id=0,thread-id=0 " + "-numa cpu,node-id=1,socket-id=0,cluster-id=0,core-id=0,thread-id=0"); qts = qtest_init(cli); cpus = get_cpus(qts, &resp); g_assert(cpus); while ((e = qlist_pop(cpus))) { QDict *cpu, *props; - int64_t thread, node; + int64_t socket, cluster, core, thread, node; cpu = qobject_to(QDict, e); g_assert(qdict_haskey(cpu, "props")); @@ -XXX,XX +XXX,XX @@ static void aarch64_numa_cpu(const void *data) g_assert(qdict_haskey(props, "node-id")); node = qdict_get_int(props, "node-id"); + g_assert(qdict_haskey(props, "socket-id")); + socket = qdict_get_int(props, "socket-id"); + g_assert(qdict_haskey(props, "cluster-id")); + cluster = qdict_get_int(props, "cluster-id"); + g_assert(qdict_haskey(props, "core-id")); + core = qdict_get_int(props, "core-id"); g_assert(qdict_haskey(props, "thread-id")); thread = qdict_get_int(props, "thread-id"); - if (thread == 0) { + if (socket == 0 && cluster == 0 && core == 0 && thread == 0) { g_assert_cmpint(node, ==, 1); - } else if (thread == 1) { + } else if (socket == 1 && cluster == 0 && core == 0 && thread == 0) { g_assert_cmpint(node, ==, 0); } else { g_assert(false); -- 2.25.1 From: Gavin Shan <gshan@redhat.com> When CPU-to-NUMA association isn't explicitly provided by users, the default one is given by mc->get_default_cpu_node_id(). However, the CPU topology isn't fully considered in the default association and this causes CPU topology broken warnings on booting Linux guest. For example, the following warning messages are observed when the Linux guest is booted with the following command lines. /home/gavin/sandbox/qemu.main/build/qemu-system-aarch64 \ -accel kvm -machine virt,gic-version=host \ -cpu host \ -smp 6,sockets=2,cores=3,threads=1 \ -m 1024M,slots=16,maxmem=64G \ -object memory-backend-ram,id=mem0,size=128M \ -object memory-backend-ram,id=mem1,size=128M \ -object memory-backend-ram,id=mem2,size=128M \ -object memory-backend-ram,id=mem3,size=128M \ -object memory-backend-ram,id=mem4,size=128M \ -object memory-backend-ram,id=mem4,size=384M \ -numa node,nodeid=0,memdev=mem0 \ -numa node,nodeid=1,memdev=mem1 \ -numa node,nodeid=2,memdev=mem2 \ -numa node,nodeid=3,memdev=mem3 \ -numa node,nodeid=4,memdev=mem4 \ -numa node,nodeid=5,memdev=mem5 : alternatives: patching kernel code BUG: arch topology borken the CLS domain not a subset of the MC domain <the above error log repeats> BUG: arch topology borken the DIE domain not a subset of the NODE domain With current implementation of mc->get_default_cpu_node_id(), CPU#0 to CPU#5 are associated with NODE#0 to NODE#5 separately. That's incorrect because CPU#0/1/2 should be associated with same NUMA node because they're seated in same socket. This fixes the issue by considering the socket ID when the default CPU-to-NUMA association is provided in virt_possible_cpu_arch_ids(). With this applied, no more CPU topology broken warnings are seen from the Linux guest. The 6 CPUs are associated with NODE#0/1, but there are no CPUs associated with NODE#2/3/4/5. Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Reviewed-by: Yanan Wang <wangyanan55@huawei.com> Message-id: 20220503140304.855514-6-gshan@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/virt.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ virt_cpu_index_to_props(MachineState *ms, unsigned cpu_index) static int64_t virt_get_default_cpu_node_id(const MachineState *ms, int idx) { - return idx % ms->numa_state->num_nodes; + int64_t socket_id = ms->possible_cpus->cpus[idx].props.socket_id; + + return socket_id % ms->numa_state->num_nodes; } static const CPUArchIdList *virt_possible_cpu_arch_ids(MachineState *ms) -- 2.25.1 From: Gavin Shan <gshan@redhat.com> When the PPTT table is built, the CPU topology is re-calculated, but it's unecessary because the CPU topology has been populated in virt_possible_cpu_arch_ids() on arm/virt machine. This reworks build_pptt() to avoid by reusing the existing IDs in ms->possible_cpus. Currently, the only user of build_pptt() is arm/virt machine. Signed-off-by: Gavin Shan <gshan@redhat.com> Tested-by: Yanan Wang <wangyanan55@huawei.com> Reviewed-by: Yanan Wang <wangyanan55@huawei.com> Acked-by: Igor Mammedov <imammedo@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Message-id: 20220503140304.855514-7-gshan@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/acpi/aml-build.c | 111 +++++++++++++++++++------------------------- 1 file changed, 48 insertions(+), 63 deletions(-) diff --git a/hw/acpi/aml-build.c b/hw/acpi/aml-build.c index XXXXXXX..XXXXXXX 100644 --- a/hw/acpi/aml-build.c +++ b/hw/acpi/aml-build.c @@ -XXX,XX +XXX,XX @@ void build_pptt(GArray *table_data, BIOSLinker *linker, MachineState *ms, const char *oem_id, const char *oem_table_id) { MachineClass *mc = MACHINE_GET_CLASS(ms); - GQueue *list = g_queue_new(); - guint pptt_start = table_data->len; - guint parent_offset; - guint length, i; - int uid = 0; - int socket; + CPUArchIdList *cpus = ms->possible_cpus; + int64_t socket_id = -1, cluster_id = -1, core_id = -1; + uint32_t socket_offset = 0, cluster_offset = 0, core_offset = 0; + uint32_t pptt_start = table_data->len; + int n; AcpiTable table = { .sig = "PPTT", .rev = 2, .oem_id = oem_id, .oem_table_id = oem_table_id }; acpi_table_begin(&table, table_data); - for (socket = 0; socket < ms->smp.sockets; socket++) { - g_queue_push_tail(list, - GUINT_TO_POINTER(table_data->len - pptt_start)); - build_processor_hierarchy_node( - table_data, - /* - * Physical package - represents the boundary - * of a physical package - */ - (1 << 0), - 0, socket, NULL, 0); - } - - if (mc->smp_props.clusters_supported) { - length = g_queue_get_length(list); - for (i = 0; i < length; i++) { - int cluster; - - parent_offset = GPOINTER_TO_UINT(g_queue_pop_head(list)); - for (cluster = 0; cluster < ms->smp.clusters; cluster++) { - g_queue_push_tail(list, - GUINT_TO_POINTER(table_data->len - pptt_start)); - build_processor_hierarchy_node( - table_data, - (0 << 0), /* not a physical package */ - parent_offset, cluster, NULL, 0); - } + /* + * This works with the assumption that cpus[n].props.*_id has been + * sorted from top to down levels in mc->possible_cpu_arch_ids(). + * Otherwise, the unexpected and duplicated containers will be + * created. + */ + for (n = 0; n < cpus->len; n++) { + if (cpus->cpus[n].props.socket_id != socket_id) { + assert(cpus->cpus[n].props.socket_id > socket_id); + socket_id = cpus->cpus[n].props.socket_id; + cluster_id = -1; + core_id = -1; + socket_offset = table_data->len - pptt_start; + build_processor_hierarchy_node(table_data, + (1 << 0), /* Physical package */ + 0, socket_id, NULL, 0); } - } - length = g_queue_get_length(list); - for (i = 0; i < length; i++) { - int core; - - parent_offset = GPOINTER_TO_UINT(g_queue_pop_head(list)); - for (core = 0; core < ms->smp.cores; core++) { - if (ms->smp.threads > 1) { - g_queue_push_tail(list, - GUINT_TO_POINTER(table_data->len - pptt_start)); - build_processor_hierarchy_node( - table_data, - (0 << 0), /* not a physical package */ - parent_offset, core, NULL, 0); - } else { - build_processor_hierarchy_node( - table_data, - (1 << 1) | /* ACPI Processor ID valid */ - (1 << 3), /* Node is a Leaf */ - parent_offset, uid++, NULL, 0); + if (mc->smp_props.clusters_supported) { + if (cpus->cpus[n].props.cluster_id != cluster_id) { + assert(cpus->cpus[n].props.cluster_id > cluster_id); + cluster_id = cpus->cpus[n].props.cluster_id; + core_id = -1; + cluster_offset = table_data->len - pptt_start; + build_processor_hierarchy_node(table_data, + (0 << 0), /* Not a physical package */ + socket_offset, cluster_id, NULL, 0); } + } else { + cluster_offset = socket_offset; } - } - length = g_queue_get_length(list); - for (i = 0; i < length; i++) { - int thread; + if (ms->smp.threads == 1) { + build_processor_hierarchy_node(table_data, + (1 << 1) | /* ACPI Processor ID valid */ + (1 << 3), /* Node is a Leaf */ + cluster_offset, n, NULL, 0); + } else { + if (cpus->cpus[n].props.core_id != core_id) { + assert(cpus->cpus[n].props.core_id > core_id); + core_id = cpus->cpus[n].props.core_id; + core_offset = table_data->len - pptt_start; + build_processor_hierarchy_node(table_data, + (0 << 0), /* Not a physical package */ + cluster_offset, core_id, NULL, 0); + } - parent_offset = GPOINTER_TO_UINT(g_queue_pop_head(list)); - for (thread = 0; thread < ms->smp.threads; thread++) { - build_processor_hierarchy_node( - table_data, + build_processor_hierarchy_node(table_data, (1 << 1) | /* ACPI Processor ID valid */ (1 << 2) | /* Processor is a Thread */ (1 << 3), /* Node is a Leaf */ - parent_offset, uid++, NULL, 0); + core_offset, n, NULL, 0); } } - g_queue_free(list); acpi_table_end(linker, &table); } -- 2.25.1