arm queue; dunno if this will be the last before softfreeze or not, but anyway probably the last large one. New orangepi-pc board model is the big item here. thanks -- PMM The following changes since commit 67d9ef7d541c3d21a25796c51c26da096a433565: Merge remote-tracking branch 'remotes/pmaydell/tags/pull-docs-20200312' into staging (2020-03-12 15:20:52 +0000) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20200312 for you to fetch changes up to aca53be34ac3e7cac5f39396a51a338860a5a837: target/arm: kvm: Inject events at the last stage of sync (2020-03-12 16:31:10 +0000) ---------------------------------------------------------------- target-arm queue: * Fix various bugs that might result in an assert() due to incorrect hflags for M-profile CPUs * Fix Aspeed SMC Controller user-mode select handling * Report correct (with-tag) address in fault address register when TBI is enabled * cubieboard: make sure SOC object isn't leaked * fsl-imx25: Wire up eSDHC controllers * fsl-imx25: Wire up USB controllers * New board model: orangepi-pc (OrangePi PC) * ARM/KVM: if user doesn't select GIC version and the host kernel can only provide GICv3, use that, rather than defaulting to "fail because GICv2 isn't possible" * kvm: Only do KVM_SET_VCPU_EVENTS at the last stage of sync ---------------------------------------------------------------- Beata Michalska (1): target/arm: kvm: Inject events at the last stage of sync Cédric Le Goater (2): aspeed/smc: Add some tracing aspeed/smc: Fix User mode select/unselect scheme Eric Auger (6): hw/arm/virt: Document 'max' value in gic-version property description hw/arm/virt: Introduce VirtGICType enum type hw/arm/virt: Introduce finalize_gic_version() target/arm/kvm: Let kvm_arm_vgic_probe() return a bitmap hw/arm/virt: kvm: Restructure finalize_gic_version() hw/arm/virt: kvm: allow gicv3 by default if v2 cannot work Guenter Roeck (2): hw/arm/fsl-imx25: Wire up eSDHC controllers hw/arm/fsl-imx25: Wire up USB controllers Igor Mammedov (1): hw/arm/cubieboard: make sure SOC object isn't leaked Niek Linnenbank (13): hw/arm: add Allwinner H3 System-on-Chip hw/arm: add Xunlong Orange Pi PC machine hw/arm/allwinner-h3: add Clock Control Unit hw/arm/allwinner-h3: add USB host controller hw/arm/allwinner-h3: add System Control module hw/arm/allwinner: add CPU Configuration module hw/arm/allwinner: add Security Identifier device hw/arm/allwinner: add SD/MMC host controller hw/arm/allwinner-h3: add EMAC ethernet device hw/arm/allwinner-h3: add Boot ROM support hw/arm/allwinner-h3: add SDRAM controller device hw/arm/allwinner: add RTC device support docs: add Orange Pi PC document Peter Maydell (4): hw/intc/armv7m_nvic: Rebuild hflags on reset target/arm: Update hflags in trans_CPS_v7m() target/arm: Recalculate hflags correctly after writes to CONTROL target/arm: Fix some comment typos Philippe Mathieu-Daudé (5): tests/boot_linux_console: Add a quick test for the OrangePi PC board tests/boot_linux_console: Add initrd test for the Orange Pi PC board tests/boot_linux_console: Add a SD card test for the OrangePi PC board tests/boot_linux_console: Add a SLOW test booting Ubuntu on OrangePi PC tests/boot_linux_console: Test booting NetBSD via U-Boot on OrangePi PC Richard Henderson (2): target/arm: Check addresses for disabled regimes target/arm: Disable clean_data_tbi for system mode Makefile.objs | 1 + hw/arm/Makefile.objs | 1 + hw/misc/Makefile.objs | 5 + hw/net/Makefile.objs | 1 + hw/rtc/Makefile.objs | 1 + hw/sd/Makefile.objs | 1 + hw/usb/hcd-ehci.h | 1 + include/hw/arm/allwinner-a10.h | 4 + include/hw/arm/allwinner-h3.h | 161 ++++++ include/hw/arm/fsl-imx25.h | 18 + include/hw/arm/virt.h | 12 +- include/hw/misc/allwinner-cpucfg.h | 52 ++ include/hw/misc/allwinner-h3-ccu.h | 66 +++ include/hw/misc/allwinner-h3-dramc.h | 106 ++++ include/hw/misc/allwinner-h3-sysctrl.h | 67 +++ include/hw/misc/allwinner-sid.h | 60 +++ include/hw/net/allwinner-sun8i-emac.h | 99 ++++ include/hw/rtc/allwinner-rtc.h | 134 +++++ include/hw/sd/allwinner-sdhost.h | 135 +++++ target/arm/helper.h | 1 + target/arm/kvm_arm.h | 3 + hw/arm/allwinner-a10.c | 19 + hw/arm/allwinner-h3.c | 465 ++++++++++++++++++ hw/arm/cubieboard.c | 18 + hw/arm/fsl-imx25.c | 56 +++ hw/arm/imx25_pdk.c | 16 + hw/arm/orangepi.c | 130 +++++ hw/arm/virt.c | 145 ++++-- hw/intc/armv7m_nvic.c | 6 + hw/misc/allwinner-cpucfg.c | 282 +++++++++++ hw/misc/allwinner-h3-ccu.c | 242 +++++++++ hw/misc/allwinner-h3-dramc.c | 358 ++++++++++++++ hw/misc/allwinner-h3-sysctrl.c | 140 ++++++ hw/misc/allwinner-sid.c | 168 +++++++ hw/net/allwinner-sun8i-emac.c | 871 +++++++++++++++++++++++++++++++++ hw/rtc/allwinner-rtc.c | 411 ++++++++++++++++ hw/sd/allwinner-sdhost.c | 854 ++++++++++++++++++++++++++++++++ hw/ssi/aspeed_smc.c | 56 ++- hw/usb/hcd-ehci-sysbus.c | 17 + target/arm/helper.c | 49 +- target/arm/kvm.c | 14 +- target/arm/kvm32.c | 15 +- target/arm/kvm64.c | 15 +- target/arm/translate-a64.c | 11 + target/arm/translate.c | 14 +- MAINTAINERS | 9 + default-configs/arm-softmmu.mak | 1 + docs/system/arm/orangepi.rst | 253 ++++++++++ docs/system/target-arm.rst | 2 + hw/arm/Kconfig | 12 + hw/misc/trace-events | 19 + hw/net/Kconfig | 3 + hw/net/trace-events | 10 + hw/rtc/trace-events | 4 + hw/sd/trace-events | 7 + hw/ssi/trace-events | 10 + tests/acceptance/boot_linux_console.py | 230 +++++++++ 57 files changed, 5787 insertions(+), 74 deletions(-) create mode 100644 include/hw/arm/allwinner-h3.h create mode 100644 include/hw/misc/allwinner-cpucfg.h create mode 100644 include/hw/misc/allwinner-h3-ccu.h create mode 100644 include/hw/misc/allwinner-h3-dramc.h create mode 100644 include/hw/misc/allwinner-h3-sysctrl.h create mode 100644 include/hw/misc/allwinner-sid.h create mode 100644 include/hw/net/allwinner-sun8i-emac.h create mode 100644 include/hw/rtc/allwinner-rtc.h create mode 100644 include/hw/sd/allwinner-sdhost.h create mode 100644 hw/arm/allwinner-h3.c create mode 100644 hw/arm/orangepi.c create mode 100644 hw/misc/allwinner-cpucfg.c create mode 100644 hw/misc/allwinner-h3-ccu.c create mode 100644 hw/misc/allwinner-h3-dramc.c create mode 100644 hw/misc/allwinner-h3-sysctrl.c create mode 100644 hw/misc/allwinner-sid.c create mode 100644 hw/net/allwinner-sun8i-emac.c create mode 100644 hw/rtc/allwinner-rtc.c create mode 100644 hw/sd/allwinner-sdhost.c create mode 100644 docs/system/arm/orangepi.rst create mode 100644 hw/ssi/trace-events Some of an M-profile CPU's cached hflags state depends on state that's in our NVIC object. We already do an hflags rebuild when the NVIC registers are written, but we also need to do this on NVIC reset, because there's no guarantee that this will happen before the CPU reset. This fixes an assertion due to mismatched hflags which happens if the CPU is reset from inside a HardFault handler. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200303174950.3298-2-peter.maydell@linaro.org --- hw/intc/armv7m_nvic.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/hw/intc/armv7m_nvic.c b/hw/intc/armv7m_nvic.c index XXXXXXX..XXXXXXX 100644 --- a/hw/intc/armv7m_nvic.c +++ b/hw/intc/armv7m_nvic.c @@ -XXX,XX +XXX,XX @@ static void armv7m_nvic_reset(DeviceState *dev) s->itns[i] = true; } } + + /* + * We updated state that affects the CPU's MMUidx and thus its hflags; + * and we can't guarantee that we run before the CPU reset function. + */ + arm_rebuild_hflags(&s->cpu->env); } static void nvic_systick_trigger(void *opaque, int n, int level) -- 2.20.1 For M-profile CPUs, the FAULTMASK value affects the CPU's MMU index (it changes the NegPri bit). We update the hflags after calls to the v7m_msr helper in trans_MSR_v7m() but forgot to do so in trans_CPS_v7m(). Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200303174950.3298-3-peter.maydell@linaro.org --- target/arm/translate.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) 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_CPS(DisasContext *s, arg_CPS *a) static bool trans_CPS_v7m(DisasContext *s, arg_CPS_v7m *a) { - TCGv_i32 tmp, addr; + TCGv_i32 tmp, addr, el; if (!arm_dc_feature(s, ARM_FEATURE_M)) { return false; @@ -XXX,XX +XXX,XX @@ static bool trans_CPS_v7m(DisasContext *s, arg_CPS_v7m *a) gen_helper_v7m_msr(cpu_env, addr, tmp); tcg_temp_free_i32(addr); } + el = tcg_const_i32(s->current_el); + gen_helper_rebuild_hflags_m32(cpu_env, el); + tcg_temp_free_i32(el); tcg_temp_free_i32(tmp); gen_lookup_tb(s); return true; -- 2.20.1 A write to the CONTROL register can change our current EL (by writing to the nPRIV bit). That means that we can't assume that s->current_el is still valid in trans_MSR_v7m() when we try to rebuild the hflags. Add a new helper rebuild_hflags_m32_newel() which, like the existing rebuild_hflags_a32_newel(), recalculates the current EL from scratch, and use it in trans_MSR_v7m(). This fixes an assertion about an hflags mismatch when the guest changes privilege by writing to CONTROL. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200303174950.3298-4-peter.maydell@linaro.org --- target/arm/helper.h | 1 + target/arm/helper.c | 12 ++++++++++++ target/arm/translate.c | 7 +++---- 3 files changed, 16 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_4(msr_banked, void, env, i32, i32, i32) DEF_HELPER_2(get_user_reg, i32, env, i32) DEF_HELPER_3(set_user_reg, void, env, i32, i32) +DEF_HELPER_FLAGS_1(rebuild_hflags_m32_newel, TCG_CALL_NO_RWG, void, env) 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) diff --git a/target/arm/helper.c b/target/arm/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.c +++ b/target/arm/helper.c @@ -XXX,XX +XXX,XX @@ void arm_rebuild_hflags(CPUARMState *env) env->hflags = rebuild_hflags_internal(env); } +/* + * If we have triggered a EL state change we can't rely on the + * translator having passed it to us, we need to recompute. + */ +void HELPER(rebuild_hflags_m32_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_m32(env, fp_el, mmu_idx); +} + void HELPER(rebuild_hflags_m32)(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 bool trans_MRS_v7m(DisasContext *s, arg_MRS_v7m *a) static bool trans_MSR_v7m(DisasContext *s, arg_MSR_v7m *a) { - TCGv_i32 addr, reg, el; + TCGv_i32 addr, reg; if (!arm_dc_feature(s, ARM_FEATURE_M)) { return false; @@ -XXX,XX +XXX,XX @@ static bool trans_MSR_v7m(DisasContext *s, arg_MSR_v7m *a) gen_helper_v7m_msr(cpu_env, addr, reg); tcg_temp_free_i32(addr); tcg_temp_free_i32(reg); - el = tcg_const_i32(s->current_el); - gen_helper_rebuild_hflags_m32(cpu_env, el); - tcg_temp_free_i32(el); + /* If we wrote to CONTROL, the EL might have changed */ + gen_helper_rebuild_hflags_m32_newel(cpu_env); gen_lookup_tb(s); return true; } -- 2.20.1 Fix a couple of comment typos. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200303174950.3298-5-peter.maydell@linaro.org --- target/arm/helper.c | 2 +- target/arm/translate.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/target/arm/helper.c b/target/arm/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.c +++ b/target/arm/helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(rebuild_hflags_m32)(CPUARMState *env, int el) /* * If we have triggered a EL state change we can't rely on the - * translator having passed it too us, we need to recompute. + * translator having passed it to us, we need to recompute. */ void HELPER(rebuild_hflags_a32_newel)(CPUARMState *env) { 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 (!isread && !(ri->type & ARM_CP_SUPPRESS_TB_END)) { /* - * A write to any coprocessor regiser that ends a TB + * A write to any coprocessor register that ends a TB * must rebuild the hflags for the next TB. */ TCGv_i32 tcg_el = tcg_const_i32(s->current_el); -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Andrew Jeffery <andrew@aj.id.au> Reviewed-by: Joel Stanley <joel@jms.id.au> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20200206112645.21275-2-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- Makefile.objs | 1 + hw/ssi/aspeed_smc.c | 17 +++++++++++++++++ hw/ssi/trace-events | 9 +++++++++ 3 files changed, 27 insertions(+) create mode 100644 hw/ssi/trace-events diff --git a/Makefile.objs b/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/Makefile.objs +++ b/Makefile.objs @@ -XXX,XX +XXX,XX @@ trace-events-subdirs += hw/scsi trace-events-subdirs += hw/sd trace-events-subdirs += hw/sparc trace-events-subdirs += hw/sparc64 +trace-events-subdirs += hw/ssi trace-events-subdirs += hw/timer trace-events-subdirs += hw/tpm trace-events-subdirs += hw/usb 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 @@ #include "qapi/error.h" #include "exec/address-spaces.h" #include "qemu/units.h" +#include "trace.h" #include "hw/irq.h" #include "hw/qdev-properties.h" @@ -XXX,XX +XXX,XX @@ static void aspeed_smc_flash_set_segment(AspeedSMCState *s, int cs, s->ctrl->reg_to_segment(s, new, &seg); + trace_aspeed_smc_flash_set_segment(cs, new, seg.addr, seg.addr + seg.size); + /* The start address of CS0 is read-only */ if (cs == 0 && seg.addr != s->ctrl->flash_window_base) { qemu_log_mask(LOG_GUEST_ERROR, @@ -XXX,XX +XXX,XX @@ static uint64_t aspeed_smc_flash_read(void *opaque, hwaddr addr, unsigned size) __func__, aspeed_smc_flash_mode(fl)); } + trace_aspeed_smc_flash_read(fl->id, addr, size, ret, + aspeed_smc_flash_mode(fl)); return ret; } @@ -XXX,XX +XXX,XX @@ static bool aspeed_smc_do_snoop(AspeedSMCFlash *fl, uint64_t data, AspeedSMCState *s = fl->controller; uint8_t addr_width = aspeed_smc_flash_is_4byte(fl) ? 4 : 3; + trace_aspeed_smc_do_snoop(fl->id, s->snoop_index, s->snoop_dummies, + (uint8_t) data & 0xff); + if (s->snoop_index == SNOOP_OFF) { return false; /* Do nothing */ @@ -XXX,XX +XXX,XX @@ static void aspeed_smc_flash_write(void *opaque, hwaddr addr, uint64_t data, AspeedSMCState *s = fl->controller; int i; + trace_aspeed_smc_flash_write(fl->id, addr, size, data, + aspeed_smc_flash_mode(fl)); + if (!aspeed_smc_is_writable(fl)) { qemu_log_mask(LOG_GUEST_ERROR, "%s: flash is not writable at 0x%" HWADDR_PRIx "\n", __func__, addr); @@ -XXX,XX +XXX,XX @@ static uint64_t aspeed_smc_read(void *opaque, hwaddr addr, unsigned int size) (s->ctrl->has_dma && addr == R_DMA_CHECKSUM) || (addr >= R_SEG_ADDR0 && addr < R_SEG_ADDR0 + s->ctrl->max_slaves) || (addr >= s->r_ctrl0 && addr < s->r_ctrl0 + s->ctrl->max_slaves)) { + + trace_aspeed_smc_read(addr, size, s->regs[addr]); + return s->regs[addr]; } else { qemu_log_mask(LOG_UNIMP, "%s: not implemented: 0x%" HWADDR_PRIx "\n", @@ -XXX,XX +XXX,XX @@ static void aspeed_smc_dma_checksum(AspeedSMCState *s) __func__, s->regs[R_DMA_FLASH_ADDR]); return; } + trace_aspeed_smc_dma_checksum(s->regs[R_DMA_FLASH_ADDR], data); /* * When the DMA is on-going, the DMA registers are updated @@ -XXX,XX +XXX,XX @@ static void aspeed_smc_write(void *opaque, hwaddr addr, uint64_t data, addr >>= 2; + trace_aspeed_smc_write(addr, size, data); + if (addr == s->r_conf || (addr >= s->r_timings && addr < s->r_timings + s->ctrl->nregs_timings) || diff --git a/hw/ssi/trace-events b/hw/ssi/trace-events new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/ssi/trace-events @@ -XXX,XX +XXX,XX @@ +# aspeed_smc.c + +aspeed_smc_flash_set_segment(int cs, uint64_t reg, uint64_t start, uint64_t end) "CS%d segreg=0x%"PRIx64" [ 0x%"PRIx64" - 0x%"PRIx64" ]" +aspeed_smc_flash_read(int cs, uint64_t addr, uint32_t size, uint64_t data, int mode) "CS%d @0x%" PRIx64 " size %u: 0x%" PRIx64" mode:%d" +aspeed_smc_do_snoop(int cs, int index, int dummies, int data) "CS%d index:0x%x dummies:%d data:0x%x" +aspeed_smc_flash_write(int cs, uint64_t addr, uint32_t size, uint64_t data, int mode) "CS%d @0x%" PRIx64 " size %u: 0x%" PRIx64" mode:%d" +aspeed_smc_read(uint64_t addr, uint32_t size, uint64_t data) "@0x%" PRIx64 " size %u: 0x%" PRIx64 +aspeed_smc_dma_checksum(uint32_t addr, uint32_t data) "0x%08x: 0x%08x" +aspeed_smc_write(uint64_t addr, uint32_t size, uint64_t data) "@0x%" PRIx64 " size %u: 0x%" PRIx64 -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> The Aspeed SMC Controller can operate in different modes : Read, Fast Read, Write and User modes. When the User mode is configured, it selects automatically the SPI slave device until the CE_STOP_ACTIVE bit is set to 1. When any other modes are configured the device is unselected. The HW logic handles the chip select automatically when the flash is accessed through its AHB window. When configuring the CEx Control Register, the User mode logic to select and unselect the slave is incorrect and data corruption can be seen on machines using two chips, witherspoon and romulus. Rework the handler setting the CEx Control Register to fix this issue. Fixes: 7c1c69bca43c ("ast2400: add SMC controllers (FMC and SPI)") Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Andrew Jeffery <andrew@aj.id.au> Message-id: 20200206112645.21275-3-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/ssi/aspeed_smc.c | 39 +++++++++++++++++++++++---------------- hw/ssi/trace-events | 1 + 2 files changed, 24 insertions(+), 16 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 inline int aspeed_smc_flash_is_4byte(const AspeedSMCFlash *fl) } } -static inline bool aspeed_smc_is_ce_stop_active(const AspeedSMCFlash *fl) +static void aspeed_smc_flash_do_select(AspeedSMCFlash *fl, bool unselect) { - const AspeedSMCState *s = fl->controller; + AspeedSMCState *s = fl->controller; - return s->regs[s->r_ctrl0 + fl->id] & CTRL_CE_STOP_ACTIVE; + trace_aspeed_smc_flash_select(fl->id, unselect ? "un" : ""); + + qemu_set_irq(s->cs_lines[fl->id], unselect); } static void aspeed_smc_flash_select(AspeedSMCFlash *fl) { - AspeedSMCState *s = fl->controller; - - s->regs[s->r_ctrl0 + fl->id] &= ~CTRL_CE_STOP_ACTIVE; - qemu_set_irq(s->cs_lines[fl->id], aspeed_smc_is_ce_stop_active(fl)); + aspeed_smc_flash_do_select(fl, false); } static void aspeed_smc_flash_unselect(AspeedSMCFlash *fl) { - AspeedSMCState *s = fl->controller; - - s->regs[s->r_ctrl0 + fl->id] |= CTRL_CE_STOP_ACTIVE; - qemu_set_irq(s->cs_lines[fl->id], aspeed_smc_is_ce_stop_active(fl)); + aspeed_smc_flash_do_select(fl, true); } static uint32_t aspeed_smc_check_segment_addr(const AspeedSMCFlash *fl, @@ -XXX,XX +XXX,XX @@ static const MemoryRegionOps aspeed_smc_flash_ops = { }, }; -static void aspeed_smc_flash_update_cs(AspeedSMCFlash *fl) +static void aspeed_smc_flash_update_ctrl(AspeedSMCFlash *fl, uint32_t value) { AspeedSMCState *s = fl->controller; + bool unselect; - s->snoop_index = aspeed_smc_is_ce_stop_active(fl) ? SNOOP_OFF : SNOOP_START; + /* User mode selects the CS, other modes unselect */ + unselect = (value & CTRL_CMD_MODE_MASK) != CTRL_USERMODE; - qemu_set_irq(s->cs_lines[fl->id], aspeed_smc_is_ce_stop_active(fl)); + /* A change of CTRL_CE_STOP_ACTIVE from 0 to 1, unselects the CS */ + if (!(s->regs[s->r_ctrl0 + fl->id] & CTRL_CE_STOP_ACTIVE) && + value & CTRL_CE_STOP_ACTIVE) { + unselect = true; + } + + s->regs[s->r_ctrl0 + fl->id] = value; + + s->snoop_index = unselect ? SNOOP_OFF : SNOOP_START; + + aspeed_smc_flash_do_select(fl, unselect); } static void aspeed_smc_reset(DeviceState *d) @@ -XXX,XX +XXX,XX @@ static void aspeed_smc_write(void *opaque, hwaddr addr, uint64_t data, s->regs[addr] = value; } else if (addr >= s->r_ctrl0 && addr < s->r_ctrl0 + s->num_cs) { int cs = addr - s->r_ctrl0; - s->regs[addr] = value; - aspeed_smc_flash_update_cs(&s->flashes[cs]); + aspeed_smc_flash_update_ctrl(&s->flashes[cs], value); } else if (addr >= R_SEG_ADDR0 && addr < R_SEG_ADDR0 + s->ctrl->max_slaves) { int cs = addr - R_SEG_ADDR0; diff --git a/hw/ssi/trace-events b/hw/ssi/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/ssi/trace-events +++ b/hw/ssi/trace-events @@ -XXX,XX +XXX,XX @@ aspeed_smc_flash_write(int cs, uint64_t addr, uint32_t size, uint64_t data, int aspeed_smc_read(uint64_t addr, uint32_t size, uint64_t data) "@0x%" PRIx64 " size %u: 0x%" PRIx64 aspeed_smc_dma_checksum(uint32_t addr, uint32_t data) "0x%08x: 0x%08x" aspeed_smc_write(uint64_t addr, uint32_t size, uint64_t data) "@0x%" PRIx64 " size %u: 0x%" PRIx64 +aspeed_smc_flash_select(int cs, const char *prefix) "CS%d %sselect" -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> We fail to validate the upper bits of a virtual address on a translation disabled regime, as per AArch64.TranslateAddressS1Off. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200308012946.16303-2-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 35 ++++++++++++++++++++++++++++++++++- 1 file changed, 34 insertions(+), 1 deletion(-) diff --git a/target/arm/helper.c b/target/arm/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.c +++ b/target/arm/helper.c @@ -XXX,XX +XXX,XX @@ bool get_phys_addr(CPUARMState *env, target_ulong address, /* Definitely a real MMU, not an MPU */ if (regime_translation_disabled(env, mmu_idx)) { - /* MMU disabled. */ + /* + * MMU disabled. S1 addresses within aa64 translation regimes are + * still checked for bounds -- see AArch64.TranslateAddressS1Off. + */ + if (mmu_idx != ARMMMUIdx_Stage2) { + int r_el = regime_el(env, mmu_idx); + if (arm_el_is_aa64(env, r_el)) { + int pamax = arm_pamax(env_archcpu(env)); + uint64_t tcr = env->cp15.tcr_el[r_el].raw_tcr; + int addrtop, tbi; + + tbi = aa64_va_parameter_tbi(tcr, mmu_idx); + if (access_type == MMU_INST_FETCH) { + tbi &= ~aa64_va_parameter_tbid(tcr, mmu_idx); + } + tbi = (tbi >> extract64(address, 55, 1)) & 1; + addrtop = (tbi ? 55 : 63); + + if (extract64(address, pamax, addrtop - pamax + 1) != 0) { + fi->type = ARMFault_AddressSize; + fi->level = 0; + fi->stage2 = false; + return 1; + } + + /* + * When TBI is disabled, we've just validated that all of the + * bits above PAMax are zero, so logically we only need to + * clear the top byte for TBI. But it's clearer to follow + * the pseudocode set of addrdesc.paddress. + */ + address = extract64(address, 0, 52); + } + } *phys_ptr = address; *prot = PAGE_READ | PAGE_WRITE | PAGE_EXEC; *page_size = TARGET_PAGE_SIZE; -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> We must include the tag in the FAR_ELx register when raising an addressing exception. Which means that we should not clear out the tag during translation. We cannot at present comply with this for user mode, so we retain the clean_data_tbi function for the moment, though it no longer does what it says on the tin for system mode. This function is to be replaced with MTE, so don't worry about the slight misnaming. Buglink: https://bugs.launchpad.net/qemu/+bug/1867072 Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200308012946.16303-3-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/translate-a64.c | 11 +++++++++++ 1 file changed, 11 insertions(+) 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 gen_a64_set_pc(DisasContext *s, TCGv_i64 src) static TCGv_i64 clean_data_tbi(DisasContext *s, TCGv_i64 addr) { TCGv_i64 clean = new_tmp_a64(s); + /* + * In order to get the correct value in the FAR_ELx register, + * we must present the memory subsystem with the "dirty" address + * including the TBI. In system mode we can make this work via + * the TLB, dropping the TBI during translation. But for user-only + * mode we don't have that option, and must remove the top byte now. + */ +#ifdef CONFIG_USER_ONLY gen_top_byte_ignore(s, clean, addr, s->tbid); +#else + tcg_gen_mov_i64(clean, addr); +#endif return clean; } -- 2.20.1 From: Igor Mammedov <imammedo@redhat.com> SOC object returned by object_new() is leaked in current code. Set SOC parent explicitly to board and then unref to SOC object to make sure that refererence returned by object_new() is taken care of. The SOC object will be kept alive by its parent (machine) and will be automatically freed when MachineState is destroyed. Signed-off-by: Igor Mammedov <imammedo@redhat.com> Reported-by: Andrew Jones <drjones@redhat.com> Tested-by: Niek Linnenbank <nieklinnenbank@gmail.com> Message-id: 20200303091254.22373-1-imammedo@redhat.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/cubieboard.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/hw/arm/cubieboard.c b/hw/arm/cubieboard.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/cubieboard.c +++ b/hw/arm/cubieboard.c @@ -XXX,XX +XXX,XX @@ static void cubieboard_init(MachineState *machine) } a10 = AW_A10(object_new(TYPE_AW_A10)); + object_property_add_child(OBJECT(machine), "soc", OBJECT(a10), + &error_abort); + object_unref(OBJECT(a10)); object_property_set_int(OBJECT(&a10->emac), 1, "phy-addr", &err); if (err != NULL) { -- 2.20.1 From: Guenter Roeck <linux@roeck-us.net> Wire up eSDHC controllers in fsl-imx25. For imx25-pdk, connect drives provided on the command line to available eSDHC controllers. This patch enables booting the imx25-pdk emulation from SD card. Signed-off-by: Guenter Roeck <linux@roeck-us.net> Message-id: 20200310215146.19688-2-linux@roeck-us.net Reviewed-by: Peter Maydell <peter.maydell@linaro.org> [PMM: made commit subject consistent with other patch] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/fsl-imx25.h | 9 +++++++++ hw/arm/fsl-imx25.c | 32 ++++++++++++++++++++++++++++++++ hw/arm/imx25_pdk.c | 16 ++++++++++++++++ 3 files changed, 57 insertions(+) diff --git a/include/hw/arm/fsl-imx25.h b/include/hw/arm/fsl-imx25.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/fsl-imx25.h +++ b/include/hw/arm/fsl-imx25.h @@ -XXX,XX +XXX,XX @@ #include "hw/misc/imx_rngc.h" #include "hw/i2c/imx_i2c.h" #include "hw/gpio/imx_gpio.h" +#include "hw/sd/sdhci.h" #include "exec/memory.h" #include "target/arm/cpu.h" @@ -XXX,XX +XXX,XX @@ #define FSL_IMX25_NUM_EPITS 2 #define FSL_IMX25_NUM_I2CS 3 #define FSL_IMX25_NUM_GPIOS 4 +#define FSL_IMX25_NUM_ESDHCS 2 typedef struct FslIMX25State { /*< private >*/ @@ -XXX,XX +XXX,XX @@ typedef struct FslIMX25State { IMXRNGCState rngc; IMXI2CState i2c[FSL_IMX25_NUM_I2CS]; IMXGPIOState gpio[FSL_IMX25_NUM_GPIOS]; + SDHCIState esdhc[FSL_IMX25_NUM_ESDHCS]; MemoryRegion rom[2]; MemoryRegion iram; MemoryRegion iram_alias; @@ -XXX,XX +XXX,XX @@ typedef struct FslIMX25State { #define FSL_IMX25_GPIO3_SIZE 0x4000 #define FSL_IMX25_RNGC_ADDR 0x53FB0000 #define FSL_IMX25_RNGC_SIZE 0x4000 +#define FSL_IMX25_ESDHC1_ADDR 0x53FB4000 +#define FSL_IMX25_ESDHC1_SIZE 0x4000 +#define FSL_IMX25_ESDHC2_ADDR 0x53FB8000 +#define FSL_IMX25_ESDHC2_SIZE 0x4000 #define FSL_IMX25_GPIO1_ADDR 0x53FCC000 #define FSL_IMX25_GPIO1_SIZE 0x4000 #define FSL_IMX25_GPIO2_ADDR 0x53FD0000 @@ -XXX,XX +XXX,XX @@ typedef struct FslIMX25State { #define FSL_IMX25_GPIO2_IRQ 51 #define FSL_IMX25_GPIO3_IRQ 16 #define FSL_IMX25_GPIO4_IRQ 23 +#define FSL_IMX25_ESDHC1_IRQ 9 +#define FSL_IMX25_ESDHC2_IRQ 8 #endif /* FSL_IMX25_H */ diff --git a/hw/arm/fsl-imx25.c b/hw/arm/fsl-imx25.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/fsl-imx25.c +++ b/hw/arm/fsl-imx25.c @@ -XXX,XX +XXX,XX @@ #include "hw/qdev-properties.h" #include "chardev/char.h" +#define IMX25_ESDHC_CAPABILITIES 0x07e20000 + static void fsl_imx25_init(Object *obj) { FslIMX25State *s = FSL_IMX25(obj); @@ -XXX,XX +XXX,XX @@ static void fsl_imx25_init(Object *obj) sysbus_init_child_obj(obj, "gpio[*]", &s->gpio[i], sizeof(s->gpio[i]), TYPE_IMX_GPIO); } + + for (i = 0; i < FSL_IMX25_NUM_ESDHCS; i++) { + sysbus_init_child_obj(obj, "sdhc[*]", &s->esdhc[i], sizeof(s->esdhc[i]), + TYPE_IMX_USDHC); + } } static void fsl_imx25_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void fsl_imx25_realize(DeviceState *dev, Error **errp) gpio_table[i].irq)); } + /* Initialize all SDHC */ + for (i = 0; i < FSL_IMX25_NUM_ESDHCS; i++) { + static const struct { + hwaddr addr; + unsigned int irq; + } esdhc_table[FSL_IMX25_NUM_ESDHCS] = { + { FSL_IMX25_ESDHC1_ADDR, FSL_IMX25_ESDHC1_IRQ }, + { FSL_IMX25_ESDHC2_ADDR, FSL_IMX25_ESDHC2_IRQ }, + }; + + object_property_set_uint(OBJECT(&s->esdhc[i]), 2, "sd-spec-version", + &err); + object_property_set_uint(OBJECT(&s->esdhc[i]), IMX25_ESDHC_CAPABILITIES, + "capareg", &err); + object_property_set_bool(OBJECT(&s->esdhc[i]), true, "realized", &err); + if (err) { + error_propagate(errp, err); + return; + } + sysbus_mmio_map(SYS_BUS_DEVICE(&s->esdhc[i]), 0, esdhc_table[i].addr); + sysbus_connect_irq(SYS_BUS_DEVICE(&s->esdhc[i]), 0, + qdev_get_gpio_in(DEVICE(&s->avic), + esdhc_table[i].irq)); + } + /* initialize 2 x 16 KB ROM */ memory_region_init_rom(&s->rom[0], NULL, "imx25.rom0", FSL_IMX25_ROM0_SIZE, &err); diff --git a/hw/arm/imx25_pdk.c b/hw/arm/imx25_pdk.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/imx25_pdk.c +++ b/hw/arm/imx25_pdk.c @@ -XXX,XX +XXX,XX @@ #include "qemu/osdep.h" #include "qapi/error.h" #include "cpu.h" +#include "hw/qdev-properties.h" #include "hw/arm/fsl-imx25.h" #include "hw/boards.h" #include "qemu/error-report.h" @@ -XXX,XX +XXX,XX @@ static void imx25_pdk_init(MachineState *machine) imx25_pdk_binfo.board_id = 1771, imx25_pdk_binfo.nb_cpus = 1; + for (i = 0; i < FSL_IMX25_NUM_ESDHCS; i++) { + BusState *bus; + DeviceState *carddev; + DriveInfo *di; + BlockBackend *blk; + + di = drive_get_next(IF_SD); + blk = di ? blk_by_legacy_dinfo(di) : NULL; + bus = qdev_get_child_bus(DEVICE(&s->soc.esdhc[i]), "sd-bus"); + carddev = qdev_create(bus, TYPE_SD_CARD); + qdev_prop_set_drive(carddev, "drive", blk, &error_fatal); + object_property_set_bool(OBJECT(carddev), true, + "realized", &error_fatal); + } + /* * We test explicitly for qtest here as it is not done (yet?) in * arm_load_kernel(). Without this the "make check" command would -- 2.20.1 From: Guenter Roeck <linux@roeck-us.net> i.MX25 supports two USB controllers. Let's wire them up. With this patch, imx25-pdk can boot from both USB ports. Signed-off-by: Guenter Roeck <linux@roeck-us.net> Message-id: 20200310215146.19688-3-linux@roeck-us.net Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/fsl-imx25.h | 9 +++++++++ hw/arm/fsl-imx25.c | 24 ++++++++++++++++++++++++ 2 files changed, 33 insertions(+) diff --git a/include/hw/arm/fsl-imx25.h b/include/hw/arm/fsl-imx25.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/fsl-imx25.h +++ b/include/hw/arm/fsl-imx25.h @@ -XXX,XX +XXX,XX @@ #include "hw/i2c/imx_i2c.h" #include "hw/gpio/imx_gpio.h" #include "hw/sd/sdhci.h" +#include "hw/usb/chipidea.h" #include "exec/memory.h" #include "target/arm/cpu.h" @@ -XXX,XX +XXX,XX @@ #define FSL_IMX25_NUM_I2CS 3 #define FSL_IMX25_NUM_GPIOS 4 #define FSL_IMX25_NUM_ESDHCS 2 +#define FSL_IMX25_NUM_USBS 2 typedef struct FslIMX25State { /*< private >*/ @@ -XXX,XX +XXX,XX @@ typedef struct FslIMX25State { IMXI2CState i2c[FSL_IMX25_NUM_I2CS]; IMXGPIOState gpio[FSL_IMX25_NUM_GPIOS]; SDHCIState esdhc[FSL_IMX25_NUM_ESDHCS]; + ChipideaState usb[FSL_IMX25_NUM_USBS]; MemoryRegion rom[2]; MemoryRegion iram; MemoryRegion iram_alias; @@ -XXX,XX +XXX,XX @@ typedef struct FslIMX25State { #define FSL_IMX25_GPIO1_SIZE 0x4000 #define FSL_IMX25_GPIO2_ADDR 0x53FD0000 #define FSL_IMX25_GPIO2_SIZE 0x4000 +#define FSL_IMX25_USB1_ADDR 0x53FF4000 +#define FSL_IMX25_USB1_SIZE 0x0200 +#define FSL_IMX25_USB2_ADDR 0x53FF4400 +#define FSL_IMX25_USB2_SIZE 0x0200 #define FSL_IMX25_AVIC_ADDR 0x68000000 #define FSL_IMX25_AVIC_SIZE 0x4000 #define FSL_IMX25_IRAM_ADDR 0x78000000 @@ -XXX,XX +XXX,XX @@ typedef struct FslIMX25State { #define FSL_IMX25_GPIO4_IRQ 23 #define FSL_IMX25_ESDHC1_IRQ 9 #define FSL_IMX25_ESDHC2_IRQ 8 +#define FSL_IMX25_USB1_IRQ 37 +#define FSL_IMX25_USB2_IRQ 35 #endif /* FSL_IMX25_H */ diff --git a/hw/arm/fsl-imx25.c b/hw/arm/fsl-imx25.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/fsl-imx25.c +++ b/hw/arm/fsl-imx25.c @@ -XXX,XX +XXX,XX @@ static void fsl_imx25_init(Object *obj) sysbus_init_child_obj(obj, "sdhc[*]", &s->esdhc[i], sizeof(s->esdhc[i]), TYPE_IMX_USDHC); } + + for (i = 0; i < FSL_IMX25_NUM_USBS; i++) { + sysbus_init_child_obj(obj, "usb[*]", &s->usb[i], sizeof(s->usb[i]), + TYPE_CHIPIDEA); + } + } static void fsl_imx25_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void fsl_imx25_realize(DeviceState *dev, Error **errp) esdhc_table[i].irq)); } + /* USB */ + for (i = 0; i < FSL_IMX25_NUM_USBS; i++) { + static const struct { + hwaddr addr; + unsigned int irq; + } usb_table[FSL_IMX25_NUM_USBS] = { + { FSL_IMX25_USB1_ADDR, FSL_IMX25_USB1_IRQ }, + { FSL_IMX25_USB2_ADDR, FSL_IMX25_USB2_IRQ }, + }; + + object_property_set_bool(OBJECT(&s->usb[i]), true, "realized", + &error_abort); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->usb[i]), 0, usb_table[i].addr); + sysbus_connect_irq(SYS_BUS_DEVICE(&s->usb[i]), 0, + qdev_get_gpio_in(DEVICE(&s->avic), + usb_table[i].irq)); + } + /* initialize 2 x 16 KB ROM */ memory_region_init_rom(&s->rom[0], NULL, "imx25.rom0", FSL_IMX25_ROM0_SIZE, &err); -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> The Allwinner H3 is a System on Chip containing four ARM Cortex A7 processor cores. Features and specifications include DDR2/DDR3 memory, SD/MMC storage cards, 10/100/1000Mbit Ethernet, USB 2.0, HDMI and various I/O modules. This commit adds support for the Allwinner H3 System on Chip. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20200311221854.30370-2-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/Makefile.objs | 1 + include/hw/arm/allwinner-h3.h | 106 +++++++++++ hw/arm/allwinner-h3.c | 327 ++++++++++++++++++++++++++++++++ MAINTAINERS | 7 + default-configs/arm-softmmu.mak | 1 + hw/arm/Kconfig | 8 + 6 files changed, 450 insertions(+) create mode 100644 include/hw/arm/allwinner-h3.h create mode 100644 hw/arm/allwinner-h3.c diff --git a/hw/arm/Makefile.objs b/hw/arm/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/Makefile.objs +++ b/hw/arm/Makefile.objs @@ -XXX,XX +XXX,XX @@ obj-$(CONFIG_DIGIC) += digic.o obj-$(CONFIG_OMAP) += omap1.o omap2.o obj-$(CONFIG_STRONGARM) += strongarm.o obj-$(CONFIG_ALLWINNER_A10) += allwinner-a10.o cubieboard.o +obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3.o obj-$(CONFIG_RASPI) += bcm2835_peripherals.o bcm2836.o raspi.o obj-$(CONFIG_STM32F205_SOC) += stm32f205_soc.o obj-$(CONFIG_STM32F405_SOC) += stm32f405_soc.o diff --git a/include/hw/arm/allwinner-h3.h b/include/hw/arm/allwinner-h3.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/arm/allwinner-h3.h @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner H3 System on Chip emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +/* + * The Allwinner H3 is a System on Chip containing four ARM Cortex A7 + * processor cores. Features and specifications include DDR2/DDR3 memory, + * SD/MMC storage cards, 10/100/1000Mbit Ethernet, USB 2.0, HDMI and + * various I/O modules. + * + * This implementation is based on the following datasheet: + * + * https://linux-sunxi.org/File:Allwinner_H3_Datasheet_V1.2.pdf + * + * The latest datasheet and more info can be found on the Linux Sunxi wiki: + * + * https://linux-sunxi.org/H3 + */ + +#ifndef HW_ARM_ALLWINNER_H3_H +#define HW_ARM_ALLWINNER_H3_H + +#include "qom/object.h" +#include "hw/arm/boot.h" +#include "hw/timer/allwinner-a10-pit.h" +#include "hw/intc/arm_gic.h" +#include "target/arm/cpu.h" + +/** + * Allwinner H3 device list + * + * This enumeration is can be used refer to a particular device in the + * Allwinner H3 SoC. For example, the physical memory base address for + * each device can be found in the AwH3State object in the memmap member + * using the device enum value as index. + * + * @see AwH3State + */ +enum { + AW_H3_SRAM_A1, + AW_H3_SRAM_A2, + AW_H3_SRAM_C, + AW_H3_PIT, + AW_H3_UART0, + AW_H3_UART1, + AW_H3_UART2, + AW_H3_UART3, + AW_H3_GIC_DIST, + AW_H3_GIC_CPU, + AW_H3_GIC_HYP, + AW_H3_GIC_VCPU, + AW_H3_SDRAM +}; + +/** Total number of CPU cores in the H3 SoC */ +#define AW_H3_NUM_CPUS (4) + +/** + * Allwinner H3 object model + * @{ + */ + +/** Object type for the Allwinner H3 SoC */ +#define TYPE_AW_H3 "allwinner-h3" + +/** Convert input object to Allwinner H3 state object */ +#define AW_H3(obj) OBJECT_CHECK(AwH3State, (obj), TYPE_AW_H3) + +/** @} */ + +/** + * Allwinner H3 object + * + * This struct contains the state of all the devices + * which are currently emulated by the H3 SoC code. + */ +typedef struct AwH3State { + /*< private >*/ + DeviceState parent_obj; + /*< public >*/ + + ARMCPU cpus[AW_H3_NUM_CPUS]; + const hwaddr *memmap; + AwA10PITState timer; + GICState gic; + MemoryRegion sram_a1; + MemoryRegion sram_a2; + MemoryRegion sram_c; +} AwH3State; + +#endif /* HW_ARM_ALLWINNER_H3_H */ diff --git a/hw/arm/allwinner-h3.c b/hw/arm/allwinner-h3.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/arm/allwinner-h3.c @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner H3 System on Chip emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "exec/address-spaces.h" +#include "qapi/error.h" +#include "qemu/error-report.h" +#include "qemu/module.h" +#include "qemu/units.h" +#include "hw/qdev-core.h" +#include "cpu.h" +#include "hw/sysbus.h" +#include "hw/char/serial.h" +#include "hw/misc/unimp.h" +#include "sysemu/sysemu.h" +#include "hw/arm/allwinner-h3.h" + +/* Memory map */ +const hwaddr allwinner_h3_memmap[] = { + [AW_H3_SRAM_A1] = 0x00000000, + [AW_H3_SRAM_A2] = 0x00044000, + [AW_H3_SRAM_C] = 0x00010000, + [AW_H3_PIT] = 0x01c20c00, + [AW_H3_UART0] = 0x01c28000, + [AW_H3_UART1] = 0x01c28400, + [AW_H3_UART2] = 0x01c28800, + [AW_H3_UART3] = 0x01c28c00, + [AW_H3_GIC_DIST] = 0x01c81000, + [AW_H3_GIC_CPU] = 0x01c82000, + [AW_H3_GIC_HYP] = 0x01c84000, + [AW_H3_GIC_VCPU] = 0x01c86000, + [AW_H3_SDRAM] = 0x40000000 +}; + +/* List of unimplemented devices */ +struct AwH3Unimplemented { + const char *device_name; + hwaddr base; + hwaddr size; +} unimplemented[] = { + { "d-engine", 0x01000000, 4 * MiB }, + { "d-inter", 0x01400000, 128 * KiB }, + { "syscon", 0x01c00000, 4 * KiB }, + { "dma", 0x01c02000, 4 * KiB }, + { "nfdc", 0x01c03000, 4 * KiB }, + { "ts", 0x01c06000, 4 * KiB }, + { "keymem", 0x01c0b000, 4 * KiB }, + { "lcd0", 0x01c0c000, 4 * KiB }, + { "lcd1", 0x01c0d000, 4 * KiB }, + { "ve", 0x01c0e000, 4 * KiB }, + { "mmc0", 0x01c0f000, 4 * KiB }, + { "mmc1", 0x01c10000, 4 * KiB }, + { "mmc2", 0x01c11000, 4 * KiB }, + { "sid", 0x01c14000, 1 * KiB }, + { "crypto", 0x01c15000, 4 * KiB }, + { "msgbox", 0x01c17000, 4 * KiB }, + { "spinlock", 0x01c18000, 4 * KiB }, + { "usb0-otg", 0x01c19000, 4 * KiB }, + { "usb0-phy", 0x01c1a000, 4 * KiB }, + { "usb1-phy", 0x01c1b000, 4 * KiB }, + { "usb2-phy", 0x01c1c000, 4 * KiB }, + { "usb3-phy", 0x01c1d000, 4 * KiB }, + { "smc", 0x01c1e000, 4 * KiB }, + { "ccu", 0x01c20000, 1 * KiB }, + { "pio", 0x01c20800, 1 * KiB }, + { "owa", 0x01c21000, 1 * KiB }, + { "pwm", 0x01c21400, 1 * KiB }, + { "keyadc", 0x01c21800, 1 * KiB }, + { "pcm0", 0x01c22000, 1 * KiB }, + { "pcm1", 0x01c22400, 1 * KiB }, + { "pcm2", 0x01c22800, 1 * KiB }, + { "audio", 0x01c22c00, 2 * KiB }, + { "smta", 0x01c23400, 1 * KiB }, + { "ths", 0x01c25000, 1 * KiB }, + { "uart0", 0x01c28000, 1 * KiB }, + { "uart1", 0x01c28400, 1 * KiB }, + { "uart2", 0x01c28800, 1 * KiB }, + { "uart3", 0x01c28c00, 1 * KiB }, + { "twi0", 0x01c2ac00, 1 * KiB }, + { "twi1", 0x01c2b000, 1 * KiB }, + { "twi2", 0x01c2b400, 1 * KiB }, + { "scr", 0x01c2c400, 1 * KiB }, + { "emac", 0x01c30000, 64 * KiB }, + { "gpu", 0x01c40000, 64 * KiB }, + { "hstmr", 0x01c60000, 4 * KiB }, + { "dramcom", 0x01c62000, 4 * KiB }, + { "dramctl0", 0x01c63000, 4 * KiB }, + { "dramphy0", 0x01c65000, 4 * KiB }, + { "spi0", 0x01c68000, 4 * KiB }, + { "spi1", 0x01c69000, 4 * KiB }, + { "csi", 0x01cb0000, 320 * KiB }, + { "tve", 0x01e00000, 64 * KiB }, + { "hdmi", 0x01ee0000, 128 * KiB }, + { "rtc", 0x01f00000, 1 * KiB }, + { "r_timer", 0x01f00800, 1 * KiB }, + { "r_intc", 0x01f00c00, 1 * KiB }, + { "r_wdog", 0x01f01000, 1 * KiB }, + { "r_prcm", 0x01f01400, 1 * KiB }, + { "r_twd", 0x01f01800, 1 * KiB }, + { "r_cpucfg", 0x01f01c00, 1 * KiB }, + { "r_cir-rx", 0x01f02000, 1 * KiB }, + { "r_twi", 0x01f02400, 1 * KiB }, + { "r_uart", 0x01f02800, 1 * KiB }, + { "r_pio", 0x01f02c00, 1 * KiB }, + { "r_pwm", 0x01f03800, 1 * KiB }, + { "core-dbg", 0x3f500000, 128 * KiB }, + { "tsgen-ro", 0x3f506000, 4 * KiB }, + { "tsgen-ctl", 0x3f507000, 4 * KiB }, + { "ddr-mem", 0x40000000, 2 * GiB }, + { "n-brom", 0xffff0000, 32 * KiB }, + { "s-brom", 0xffff0000, 64 * KiB } +}; + +/* Per Processor Interrupts */ +enum { + AW_H3_GIC_PPI_MAINT = 9, + AW_H3_GIC_PPI_HYPTIMER = 10, + AW_H3_GIC_PPI_VIRTTIMER = 11, + AW_H3_GIC_PPI_SECTIMER = 13, + AW_H3_GIC_PPI_PHYSTIMER = 14 +}; + +/* Shared Processor Interrupts */ +enum { + AW_H3_GIC_SPI_UART0 = 0, + AW_H3_GIC_SPI_UART1 = 1, + AW_H3_GIC_SPI_UART2 = 2, + AW_H3_GIC_SPI_UART3 = 3, + AW_H3_GIC_SPI_TIMER0 = 18, + AW_H3_GIC_SPI_TIMER1 = 19, +}; + +/* Allwinner H3 general constants */ +enum { + AW_H3_GIC_NUM_SPI = 128 +}; + +static void allwinner_h3_init(Object *obj) +{ + AwH3State *s = AW_H3(obj); + + s->memmap = allwinner_h3_memmap; + + for (int i = 0; i < AW_H3_NUM_CPUS; i++) { + object_initialize_child(obj, "cpu[*]", &s->cpus[i], sizeof(s->cpus[i]), + ARM_CPU_TYPE_NAME("cortex-a7"), + &error_abort, NULL); + } + + sysbus_init_child_obj(obj, "gic", &s->gic, sizeof(s->gic), + TYPE_ARM_GIC); + + sysbus_init_child_obj(obj, "timer", &s->timer, sizeof(s->timer), + TYPE_AW_A10_PIT); + object_property_add_alias(obj, "clk0-freq", OBJECT(&s->timer), + "clk0-freq", &error_abort); + object_property_add_alias(obj, "clk1-freq", OBJECT(&s->timer), + "clk1-freq", &error_abort); +} + +static void allwinner_h3_realize(DeviceState *dev, Error **errp) +{ + AwH3State *s = AW_H3(dev); + unsigned i; + + /* CPUs */ + for (i = 0; i < AW_H3_NUM_CPUS; i++) { + + /* Provide Power State Coordination Interface */ + qdev_prop_set_int32(DEVICE(&s->cpus[i]), "psci-conduit", + QEMU_PSCI_CONDUIT_HVC); + + /* Disable secondary CPUs */ + qdev_prop_set_bit(DEVICE(&s->cpus[i]), "start-powered-off", + i > 0); + + /* All exception levels required */ + qdev_prop_set_bit(DEVICE(&s->cpus[i]), "has_el3", true); + qdev_prop_set_bit(DEVICE(&s->cpus[i]), "has_el2", true); + + /* Mark realized */ + qdev_init_nofail(DEVICE(&s->cpus[i])); + } + + /* Generic Interrupt Controller */ + qdev_prop_set_uint32(DEVICE(&s->gic), "num-irq", AW_H3_GIC_NUM_SPI + + GIC_INTERNAL); + qdev_prop_set_uint32(DEVICE(&s->gic), "revision", 2); + qdev_prop_set_uint32(DEVICE(&s->gic), "num-cpu", AW_H3_NUM_CPUS); + qdev_prop_set_bit(DEVICE(&s->gic), "has-security-extensions", false); + qdev_prop_set_bit(DEVICE(&s->gic), "has-virtualization-extensions", true); + qdev_init_nofail(DEVICE(&s->gic)); + + sysbus_mmio_map(SYS_BUS_DEVICE(&s->gic), 0, s->memmap[AW_H3_GIC_DIST]); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->gic), 1, s->memmap[AW_H3_GIC_CPU]); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->gic), 2, s->memmap[AW_H3_GIC_HYP]); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->gic), 3, s->memmap[AW_H3_GIC_VCPU]); + + /* + * Wire the outputs from each CPU's generic timer and the GICv3 + * maintenance interrupt signal to the appropriate GIC PPI inputs, + * and the GIC's IRQ/FIQ/VIRQ/VFIQ interrupt outputs to the CPU's inputs. + */ + for (i = 0; i < AW_H3_NUM_CPUS; i++) { + DeviceState *cpudev = DEVICE(&s->cpus[i]); + int ppibase = AW_H3_GIC_NUM_SPI + i * GIC_INTERNAL + GIC_NR_SGIS; + int irq; + /* + * Mapping from the output timer irq lines from the CPU to the + * GIC PPI inputs used for this board. + */ + const int timer_irq[] = { + [GTIMER_PHYS] = AW_H3_GIC_PPI_PHYSTIMER, + [GTIMER_VIRT] = AW_H3_GIC_PPI_VIRTTIMER, + [GTIMER_HYP] = AW_H3_GIC_PPI_HYPTIMER, + [GTIMER_SEC] = AW_H3_GIC_PPI_SECTIMER, + }; + + /* Connect CPU timer outputs to GIC PPI inputs */ + for (irq = 0; irq < ARRAY_SIZE(timer_irq); irq++) { + qdev_connect_gpio_out(cpudev, irq, + qdev_get_gpio_in(DEVICE(&s->gic), + ppibase + timer_irq[irq])); + } + + /* Connect GIC outputs to CPU interrupt inputs */ + sysbus_connect_irq(SYS_BUS_DEVICE(&s->gic), i, + qdev_get_gpio_in(cpudev, ARM_CPU_IRQ)); + sysbus_connect_irq(SYS_BUS_DEVICE(&s->gic), i + AW_H3_NUM_CPUS, + qdev_get_gpio_in(cpudev, ARM_CPU_FIQ)); + sysbus_connect_irq(SYS_BUS_DEVICE(&s->gic), i + (2 * AW_H3_NUM_CPUS), + qdev_get_gpio_in(cpudev, ARM_CPU_VIRQ)); + sysbus_connect_irq(SYS_BUS_DEVICE(&s->gic), i + (3 * AW_H3_NUM_CPUS), + qdev_get_gpio_in(cpudev, ARM_CPU_VFIQ)); + + /* GIC maintenance signal */ + sysbus_connect_irq(SYS_BUS_DEVICE(&s->gic), i + (4 * AW_H3_NUM_CPUS), + qdev_get_gpio_in(DEVICE(&s->gic), + ppibase + AW_H3_GIC_PPI_MAINT)); + } + + /* Timer */ + qdev_init_nofail(DEVICE(&s->timer)); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->timer), 0, s->memmap[AW_H3_PIT]); + sysbus_connect_irq(SYS_BUS_DEVICE(&s->timer), 0, + qdev_get_gpio_in(DEVICE(&s->gic), AW_H3_GIC_SPI_TIMER0)); + sysbus_connect_irq(SYS_BUS_DEVICE(&s->timer), 1, + qdev_get_gpio_in(DEVICE(&s->gic), AW_H3_GIC_SPI_TIMER1)); + + /* SRAM */ + memory_region_init_ram(&s->sram_a1, OBJECT(dev), "sram A1", + 64 * KiB, &error_abort); + memory_region_init_ram(&s->sram_a2, OBJECT(dev), "sram A2", + 32 * KiB, &error_abort); + memory_region_init_ram(&s->sram_c, OBJECT(dev), "sram C", + 44 * KiB, &error_abort); + memory_region_add_subregion(get_system_memory(), s->memmap[AW_H3_SRAM_A1], + &s->sram_a1); + memory_region_add_subregion(get_system_memory(), s->memmap[AW_H3_SRAM_A2], + &s->sram_a2); + memory_region_add_subregion(get_system_memory(), s->memmap[AW_H3_SRAM_C], + &s->sram_c); + + /* UART0. For future clocktree API: All UARTS are connected to APB2_CLK. */ + serial_mm_init(get_system_memory(), s->memmap[AW_H3_UART0], 2, + qdev_get_gpio_in(DEVICE(&s->gic), AW_H3_GIC_SPI_UART0), + 115200, serial_hd(0), DEVICE_NATIVE_ENDIAN); + /* UART1 */ + serial_mm_init(get_system_memory(), s->memmap[AW_H3_UART1], 2, + qdev_get_gpio_in(DEVICE(&s->gic), AW_H3_GIC_SPI_UART1), + 115200, serial_hd(1), DEVICE_NATIVE_ENDIAN); + /* UART2 */ + serial_mm_init(get_system_memory(), s->memmap[AW_H3_UART2], 2, + qdev_get_gpio_in(DEVICE(&s->gic), AW_H3_GIC_SPI_UART2), + 115200, serial_hd(2), DEVICE_NATIVE_ENDIAN); + /* UART3 */ + serial_mm_init(get_system_memory(), s->memmap[AW_H3_UART3], 2, + qdev_get_gpio_in(DEVICE(&s->gic), AW_H3_GIC_SPI_UART3), + 115200, serial_hd(3), DEVICE_NATIVE_ENDIAN); + + /* Unimplemented devices */ + for (i = 0; i < ARRAY_SIZE(unimplemented); i++) { + create_unimplemented_device(unimplemented[i].device_name, + unimplemented[i].base, + unimplemented[i].size); + } +} + +static void allwinner_h3_class_init(ObjectClass *oc, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(oc); + + dc->realize = allwinner_h3_realize; + /* Reason: uses serial_hd() in realize function */ + dc->user_creatable = false; +} + +static const TypeInfo allwinner_h3_type_info = { + .name = TYPE_AW_H3, + .parent = TYPE_DEVICE, + .instance_size = sizeof(AwH3State), + .instance_init = allwinner_h3_init, + .class_init = allwinner_h3_class_init, +}; + +static void allwinner_h3_register_types(void) +{ + type_register_static(&allwinner_h3_type_info); +} + +type_init(allwinner_h3_register_types) diff --git a/MAINTAINERS b/MAINTAINERS index XXXXXXX..XXXXXXX 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -XXX,XX +XXX,XX @@ F: hw/*/allwinner* F: include/hw/*/allwinner* F: hw/arm/cubieboard.c +Allwinner-h3 +M: Niek Linnenbank <nieklinnenbank@gmail.com> +L: qemu-arm@nongnu.org +S: Maintained +F: hw/*/allwinner-h3* +F: include/hw/*/allwinner-h3* + ARM PrimeCell and CMSDK devices M: Peter Maydell <peter.maydell@linaro.org> L: qemu-arm@nongnu.org diff --git a/default-configs/arm-softmmu.mak b/default-configs/arm-softmmu.mak index XXXXXXX..XXXXXXX 100644 --- a/default-configs/arm-softmmu.mak +++ b/default-configs/arm-softmmu.mak @@ -XXX,XX +XXX,XX @@ CONFIG_FSL_IMX25=y CONFIG_FSL_IMX7=y CONFIG_FSL_IMX6UL=y CONFIG_SEMIHOSTING=y +CONFIG_ALLWINNER_H3=y diff --git a/hw/arm/Kconfig b/hw/arm/Kconfig index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/Kconfig +++ b/hw/arm/Kconfig @@ -XXX,XX +XXX,XX @@ config ALLWINNER_A10 select SERIAL select UNIMP +config ALLWINNER_H3 + bool + select ALLWINNER_A10_PIT + select SERIAL + select ARM_TIMER + select ARM_GIC + select UNIMP + config RASPI bool select FRAMEBUFFER -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> The Xunlong Orange Pi PC is an Allwinner H3 System on Chip based embedded computer with mainline support in both U-Boot and Linux. The board comes with a Quad Core Cortex A7 @ 1.3GHz, 1GiB RAM, 100Mbit ethernet, USB, SD/MMC, USB, HDMI and various other I/O. This commit add support for the Xunlong Orange Pi PC machine. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Tested-by: KONRAD Frederic <frederic.konrad@adacore.com> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Acked-by: Igor Mammedov <imammedo@redhat.com> Message-id: 20200311221854.30370-3-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/Makefile.objs | 2 +- hw/arm/orangepi.c | 92 ++++++++++++++++++++++++++++++++++++++++++++ MAINTAINERS | 1 + 3 files changed, 94 insertions(+), 1 deletion(-) create mode 100644 hw/arm/orangepi.c diff --git a/hw/arm/Makefile.objs b/hw/arm/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/Makefile.objs +++ b/hw/arm/Makefile.objs @@ -XXX,XX +XXX,XX @@ obj-$(CONFIG_DIGIC) += digic.o obj-$(CONFIG_OMAP) += omap1.o omap2.o obj-$(CONFIG_STRONGARM) += strongarm.o obj-$(CONFIG_ALLWINNER_A10) += allwinner-a10.o cubieboard.o -obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3.o +obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3.o orangepi.o obj-$(CONFIG_RASPI) += bcm2835_peripherals.o bcm2836.o raspi.o obj-$(CONFIG_STM32F205_SOC) += stm32f205_soc.o obj-$(CONFIG_STM32F405_SOC) += stm32f405_soc.o diff --git a/hw/arm/orangepi.c b/hw/arm/orangepi.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/arm/orangepi.c @@ -XXX,XX +XXX,XX @@ +/* + * Orange Pi emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu/units.h" +#include "exec/address-spaces.h" +#include "qapi/error.h" +#include "cpu.h" +#include "hw/sysbus.h" +#include "hw/boards.h" +#include "hw/qdev-properties.h" +#include "hw/arm/allwinner-h3.h" +#include "sysemu/sysemu.h" + +static struct arm_boot_info orangepi_binfo = { + .nb_cpus = AW_H3_NUM_CPUS, +}; + +static void orangepi_init(MachineState *machine) +{ + AwH3State *h3; + + /* BIOS is not supported by this board */ + if (bios_name) { + error_report("BIOS not supported for this machine"); + exit(1); + } + + /* This board has fixed size RAM */ + if (machine->ram_size != 1 * GiB) { + error_report("This machine can only be used with 1GiB of RAM"); + exit(1); + } + + /* Only allow Cortex-A7 for this board */ + if (strcmp(machine->cpu_type, ARM_CPU_TYPE_NAME("cortex-a7")) != 0) { + error_report("This board can only be used with cortex-a7 CPU"); + exit(1); + } + + h3 = AW_H3(object_new(TYPE_AW_H3)); + object_property_add_child(OBJECT(machine), "soc", OBJECT(h3), + &error_abort); + object_unref(OBJECT(h3)); + + /* Setup timer properties */ + object_property_set_int(OBJECT(h3), 32768, "clk0-freq", + &error_abort); + object_property_set_int(OBJECT(h3), 24 * 1000 * 1000, "clk1-freq", + &error_abort); + + /* Mark H3 object realized */ + object_property_set_bool(OBJECT(h3), true, "realized", &error_abort); + + /* SDRAM */ + memory_region_add_subregion(get_system_memory(), h3->memmap[AW_H3_SDRAM], + machine->ram); + + orangepi_binfo.loader_start = h3->memmap[AW_H3_SDRAM]; + orangepi_binfo.ram_size = machine->ram_size; + arm_load_kernel(ARM_CPU(first_cpu), machine, &orangepi_binfo); +} + +static void orangepi_machine_init(MachineClass *mc) +{ + mc->desc = "Orange Pi PC"; + mc->init = orangepi_init; + mc->min_cpus = AW_H3_NUM_CPUS; + mc->max_cpus = AW_H3_NUM_CPUS; + mc->default_cpus = AW_H3_NUM_CPUS; + mc->default_cpu_type = ARM_CPU_TYPE_NAME("cortex-a7"); + mc->default_ram_size = 1 * GiB; + mc->default_ram_id = "orangepi.ram"; +} + +DEFINE_MACHINE("orangepi-pc", orangepi_machine_init) diff --git a/MAINTAINERS b/MAINTAINERS index XXXXXXX..XXXXXXX 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -XXX,XX +XXX,XX @@ L: qemu-arm@nongnu.org S: Maintained F: hw/*/allwinner-h3* F: include/hw/*/allwinner-h3* +F: hw/arm/orangepi.c ARM PrimeCell and CMSDK devices M: Peter Maydell <peter.maydell@linaro.org> -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> The Clock Control Unit is responsible for clock signal generation, configuration and distribution in the Allwinner H3 System on Chip. This commit adds support for the Clock Control Unit which emulates a simple read/write register interface. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20200311221854.30370-4-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/misc/Makefile.objs | 1 + include/hw/arm/allwinner-h3.h | 3 + include/hw/misc/allwinner-h3-ccu.h | 66 ++++++++ hw/arm/allwinner-h3.c | 9 +- hw/misc/allwinner-h3-ccu.c | 242 +++++++++++++++++++++++++++++ 5 files changed, 320 insertions(+), 1 deletion(-) create mode 100644 include/hw/misc/allwinner-h3-ccu.h create mode 100644 hw/misc/allwinner-h3-ccu.c diff --git a/hw/misc/Makefile.objs b/hw/misc/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/Makefile.objs +++ b/hw/misc/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_MACIO) += macio/ common-obj-$(CONFIG_IVSHMEM_DEVICE) += ivshmem.o +common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3-ccu.o common-obj-$(CONFIG_REALVIEW) += arm_sysctl.o common-obj-$(CONFIG_NSERIES) += cbus.o common-obj-$(CONFIG_ECCMEMCTL) += eccmemctl.o diff --git a/include/hw/arm/allwinner-h3.h b/include/hw/arm/allwinner-h3.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-h3.h +++ b/include/hw/arm/allwinner-h3.h @@ -XXX,XX +XXX,XX @@ #include "hw/arm/boot.h" #include "hw/timer/allwinner-a10-pit.h" #include "hw/intc/arm_gic.h" +#include "hw/misc/allwinner-h3-ccu.h" #include "target/arm/cpu.h" /** @@ -XXX,XX +XXX,XX @@ enum { AW_H3_SRAM_A1, AW_H3_SRAM_A2, AW_H3_SRAM_C, + AW_H3_CCU, AW_H3_PIT, AW_H3_UART0, AW_H3_UART1, @@ -XXX,XX +XXX,XX @@ typedef struct AwH3State { ARMCPU cpus[AW_H3_NUM_CPUS]; const hwaddr *memmap; AwA10PITState timer; + AwH3ClockCtlState ccu; GICState gic; MemoryRegion sram_a1; MemoryRegion sram_a2; diff --git a/include/hw/misc/allwinner-h3-ccu.h b/include/hw/misc/allwinner-h3-ccu.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/misc/allwinner-h3-ccu.h @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner H3 Clock Control Unit emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef HW_MISC_ALLWINNER_H3_CCU_H +#define HW_MISC_ALLWINNER_H3_CCU_H + +#include "qom/object.h" +#include "hw/sysbus.h" + +/** + * @name Constants + * @{ + */ + +/** Size of register I/O address space used by CCU device */ +#define AW_H3_CCU_IOSIZE (0x400) + +/** Total number of known registers */ +#define AW_H3_CCU_REGS_NUM (AW_H3_CCU_IOSIZE / sizeof(uint32_t)) + +/** @} */ + +/** + * @name Object model + * @{ + */ + +#define TYPE_AW_H3_CCU "allwinner-h3-ccu" +#define AW_H3_CCU(obj) \ + OBJECT_CHECK(AwH3ClockCtlState, (obj), TYPE_AW_H3_CCU) + +/** @} */ + +/** + * Allwinner H3 CCU object instance state. + */ +typedef struct AwH3ClockCtlState { + /*< private >*/ + SysBusDevice parent_obj; + /*< public >*/ + + /** Maps I/O registers in physical memory */ + MemoryRegion iomem; + + /** Array of hardware registers */ + uint32_t regs[AW_H3_CCU_REGS_NUM]; + +} AwH3ClockCtlState; + +#endif /* HW_MISC_ALLWINNER_H3_CCU_H */ diff --git a/hw/arm/allwinner-h3.c b/hw/arm/allwinner-h3.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/allwinner-h3.c +++ b/hw/arm/allwinner-h3.c @@ -XXX,XX +XXX,XX @@ const hwaddr allwinner_h3_memmap[] = { [AW_H3_SRAM_A1] = 0x00000000, [AW_H3_SRAM_A2] = 0x00044000, [AW_H3_SRAM_C] = 0x00010000, + [AW_H3_CCU] = 0x01c20000, [AW_H3_PIT] = 0x01c20c00, [AW_H3_UART0] = 0x01c28000, [AW_H3_UART1] = 0x01c28400, @@ -XXX,XX +XXX,XX @@ struct AwH3Unimplemented { { "usb2-phy", 0x01c1c000, 4 * KiB }, { "usb3-phy", 0x01c1d000, 4 * KiB }, { "smc", 0x01c1e000, 4 * KiB }, - { "ccu", 0x01c20000, 1 * KiB }, { "pio", 0x01c20800, 1 * KiB }, { "owa", 0x01c21000, 1 * KiB }, { "pwm", 0x01c21400, 1 * KiB }, @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_init(Object *obj) "clk0-freq", &error_abort); object_property_add_alias(obj, "clk1-freq", OBJECT(&s->timer), "clk1-freq", &error_abort); + + sysbus_init_child_obj(obj, "ccu", &s->ccu, sizeof(s->ccu), + TYPE_AW_H3_CCU); } static void allwinner_h3_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_realize(DeviceState *dev, Error **errp) memory_region_add_subregion(get_system_memory(), s->memmap[AW_H3_SRAM_C], &s->sram_c); + /* Clock Control Unit */ + qdev_init_nofail(DEVICE(&s->ccu)); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->ccu), 0, s->memmap[AW_H3_CCU]); + /* UART0. For future clocktree API: All UARTS are connected to APB2_CLK. */ serial_mm_init(get_system_memory(), s->memmap[AW_H3_UART0], 2, qdev_get_gpio_in(DEVICE(&s->gic), AW_H3_GIC_SPI_UART0), diff --git a/hw/misc/allwinner-h3-ccu.c b/hw/misc/allwinner-h3-ccu.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/misc/allwinner-h3-ccu.c @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner H3 Clock Control Unit emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu/units.h" +#include "hw/sysbus.h" +#include "migration/vmstate.h" +#include "qemu/log.h" +#include "qemu/module.h" +#include "hw/misc/allwinner-h3-ccu.h" + +/* CCU register offsets */ +enum { + REG_PLL_CPUX = 0x0000, /* PLL CPUX Control */ + REG_PLL_AUDIO = 0x0008, /* PLL Audio Control */ + REG_PLL_VIDEO = 0x0010, /* PLL Video Control */ + REG_PLL_VE = 0x0018, /* PLL VE Control */ + REG_PLL_DDR = 0x0020, /* PLL DDR Control */ + REG_PLL_PERIPH0 = 0x0028, /* PLL Peripherals 0 Control */ + REG_PLL_GPU = 0x0038, /* PLL GPU Control */ + REG_PLL_PERIPH1 = 0x0044, /* PLL Peripherals 1 Control */ + REG_PLL_DE = 0x0048, /* PLL Display Engine Control */ + REG_CPUX_AXI = 0x0050, /* CPUX/AXI Configuration */ + REG_APB1 = 0x0054, /* ARM Peripheral Bus 1 Config */ + REG_APB2 = 0x0058, /* ARM Peripheral Bus 2 Config */ + REG_DRAM_CFG = 0x00F4, /* DRAM Configuration */ + REG_MBUS = 0x00FC, /* MBUS Reset */ + REG_PLL_TIME0 = 0x0200, /* PLL Stable Time 0 */ + REG_PLL_TIME1 = 0x0204, /* PLL Stable Time 1 */ + REG_PLL_CPUX_BIAS = 0x0220, /* PLL CPUX Bias */ + REG_PLL_AUDIO_BIAS = 0x0224, /* PLL Audio Bias */ + REG_PLL_VIDEO_BIAS = 0x0228, /* PLL Video Bias */ + REG_PLL_VE_BIAS = 0x022C, /* PLL VE Bias */ + REG_PLL_DDR_BIAS = 0x0230, /* PLL DDR Bias */ + REG_PLL_PERIPH0_BIAS = 0x0234, /* PLL Peripherals 0 Bias */ + REG_PLL_GPU_BIAS = 0x023C, /* PLL GPU Bias */ + REG_PLL_PERIPH1_BIAS = 0x0244, /* PLL Peripherals 1 Bias */ + REG_PLL_DE_BIAS = 0x0248, /* PLL Display Engine Bias */ + REG_PLL_CPUX_TUNING = 0x0250, /* PLL CPUX Tuning */ + REG_PLL_DDR_TUNING = 0x0260, /* PLL DDR Tuning */ +}; + +#define REG_INDEX(offset) (offset / sizeof(uint32_t)) + +/* CCU register flags */ +enum { + REG_DRAM_CFG_UPDATE = (1 << 16), +}; + +enum { + REG_PLL_ENABLE = (1 << 31), + REG_PLL_LOCK = (1 << 28), +}; + + +/* CCU register reset values */ +enum { + REG_PLL_CPUX_RST = 0x00001000, + REG_PLL_AUDIO_RST = 0x00035514, + REG_PLL_VIDEO_RST = 0x03006207, + REG_PLL_VE_RST = 0x03006207, + REG_PLL_DDR_RST = 0x00001000, + REG_PLL_PERIPH0_RST = 0x00041811, + REG_PLL_GPU_RST = 0x03006207, + REG_PLL_PERIPH1_RST = 0x00041811, + REG_PLL_DE_RST = 0x03006207, + REG_CPUX_AXI_RST = 0x00010000, + REG_APB1_RST = 0x00001010, + REG_APB2_RST = 0x01000000, + REG_DRAM_CFG_RST = 0x00000000, + REG_MBUS_RST = 0x80000000, + REG_PLL_TIME0_RST = 0x000000FF, + REG_PLL_TIME1_RST = 0x000000FF, + REG_PLL_CPUX_BIAS_RST = 0x08100200, + REG_PLL_AUDIO_BIAS_RST = 0x10100000, + REG_PLL_VIDEO_BIAS_RST = 0x10100000, + REG_PLL_VE_BIAS_RST = 0x10100000, + REG_PLL_DDR_BIAS_RST = 0x81104000, + REG_PLL_PERIPH0_BIAS_RST = 0x10100010, + REG_PLL_GPU_BIAS_RST = 0x10100000, + REG_PLL_PERIPH1_BIAS_RST = 0x10100010, + REG_PLL_DE_BIAS_RST = 0x10100000, + REG_PLL_CPUX_TUNING_RST = 0x0A101000, + REG_PLL_DDR_TUNING_RST = 0x14880000, +}; + +static uint64_t allwinner_h3_ccu_read(void *opaque, hwaddr offset, + unsigned size) +{ + const AwH3ClockCtlState *s = AW_H3_CCU(opaque); + const uint32_t idx = REG_INDEX(offset); + + switch (offset) { + case 0x308 ... AW_H3_CCU_IOSIZE: + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + return 0; + } + + return s->regs[idx]; +} + +static void allwinner_h3_ccu_write(void *opaque, hwaddr offset, + uint64_t val, unsigned size) +{ + AwH3ClockCtlState *s = AW_H3_CCU(opaque); + const uint32_t idx = REG_INDEX(offset); + + switch (offset) { + case REG_DRAM_CFG: /* DRAM Configuration */ + val &= ~REG_DRAM_CFG_UPDATE; + break; + case REG_PLL_CPUX: /* PLL CPUX Control */ + case REG_PLL_AUDIO: /* PLL Audio Control */ + case REG_PLL_VIDEO: /* PLL Video Control */ + case REG_PLL_VE: /* PLL VE Control */ + case REG_PLL_DDR: /* PLL DDR Control */ + case REG_PLL_PERIPH0: /* PLL Peripherals 0 Control */ + case REG_PLL_GPU: /* PLL GPU Control */ + case REG_PLL_PERIPH1: /* PLL Peripherals 1 Control */ + case REG_PLL_DE: /* PLL Display Engine Control */ + if (val & REG_PLL_ENABLE) { + val |= REG_PLL_LOCK; + } + break; + case 0x308 ... AW_H3_CCU_IOSIZE: + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + break; + default: + qemu_log_mask(LOG_UNIMP, "%s: unimplemented write offset 0x%04x\n", + __func__, (uint32_t)offset); + break; + } + + s->regs[idx] = (uint32_t) val; +} + +static const MemoryRegionOps allwinner_h3_ccu_ops = { + .read = allwinner_h3_ccu_read, + .write = allwinner_h3_ccu_write, + .endianness = DEVICE_NATIVE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, + .impl.min_access_size = 4, +}; + +static void allwinner_h3_ccu_reset(DeviceState *dev) +{ + AwH3ClockCtlState *s = AW_H3_CCU(dev); + + /* Set default values for registers */ + s->regs[REG_INDEX(REG_PLL_CPUX)] = REG_PLL_CPUX_RST; + s->regs[REG_INDEX(REG_PLL_AUDIO)] = REG_PLL_AUDIO_RST; + s->regs[REG_INDEX(REG_PLL_VIDEO)] = REG_PLL_VIDEO_RST; + s->regs[REG_INDEX(REG_PLL_VE)] = REG_PLL_VE_RST; + s->regs[REG_INDEX(REG_PLL_DDR)] = REG_PLL_DDR_RST; + s->regs[REG_INDEX(REG_PLL_PERIPH0)] = REG_PLL_PERIPH0_RST; + s->regs[REG_INDEX(REG_PLL_GPU)] = REG_PLL_GPU_RST; + s->regs[REG_INDEX(REG_PLL_PERIPH1)] = REG_PLL_PERIPH1_RST; + s->regs[REG_INDEX(REG_PLL_DE)] = REG_PLL_DE_RST; + s->regs[REG_INDEX(REG_CPUX_AXI)] = REG_CPUX_AXI_RST; + s->regs[REG_INDEX(REG_APB1)] = REG_APB1_RST; + s->regs[REG_INDEX(REG_APB2)] = REG_APB2_RST; + s->regs[REG_INDEX(REG_DRAM_CFG)] = REG_DRAM_CFG_RST; + s->regs[REG_INDEX(REG_MBUS)] = REG_MBUS_RST; + s->regs[REG_INDEX(REG_PLL_TIME0)] = REG_PLL_TIME0_RST; + s->regs[REG_INDEX(REG_PLL_TIME1)] = REG_PLL_TIME1_RST; + s->regs[REG_INDEX(REG_PLL_CPUX_BIAS)] = REG_PLL_CPUX_BIAS_RST; + s->regs[REG_INDEX(REG_PLL_AUDIO_BIAS)] = REG_PLL_AUDIO_BIAS_RST; + s->regs[REG_INDEX(REG_PLL_VIDEO_BIAS)] = REG_PLL_VIDEO_BIAS_RST; + s->regs[REG_INDEX(REG_PLL_VE_BIAS)] = REG_PLL_VE_BIAS_RST; + s->regs[REG_INDEX(REG_PLL_DDR_BIAS)] = REG_PLL_DDR_BIAS_RST; + s->regs[REG_INDEX(REG_PLL_PERIPH0_BIAS)] = REG_PLL_PERIPH0_BIAS_RST; + s->regs[REG_INDEX(REG_PLL_GPU_BIAS)] = REG_PLL_GPU_BIAS_RST; + s->regs[REG_INDEX(REG_PLL_PERIPH1_BIAS)] = REG_PLL_PERIPH1_BIAS_RST; + s->regs[REG_INDEX(REG_PLL_DE_BIAS)] = REG_PLL_DE_BIAS_RST; + s->regs[REG_INDEX(REG_PLL_CPUX_TUNING)] = REG_PLL_CPUX_TUNING_RST; + s->regs[REG_INDEX(REG_PLL_DDR_TUNING)] = REG_PLL_DDR_TUNING_RST; +} + +static void allwinner_h3_ccu_init(Object *obj) +{ + SysBusDevice *sbd = SYS_BUS_DEVICE(obj); + AwH3ClockCtlState *s = AW_H3_CCU(obj); + + /* Memory mapping */ + memory_region_init_io(&s->iomem, OBJECT(s), &allwinner_h3_ccu_ops, s, + TYPE_AW_H3_CCU, AW_H3_CCU_IOSIZE); + sysbus_init_mmio(sbd, &s->iomem); +} + +static const VMStateDescription allwinner_h3_ccu_vmstate = { + .name = "allwinner-h3-ccu", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32_ARRAY(regs, AwH3ClockCtlState, AW_H3_CCU_REGS_NUM), + VMSTATE_END_OF_LIST() + } +}; + +static void allwinner_h3_ccu_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->reset = allwinner_h3_ccu_reset; + dc->vmsd = &allwinner_h3_ccu_vmstate; +} + +static const TypeInfo allwinner_h3_ccu_info = { + .name = TYPE_AW_H3_CCU, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_init = allwinner_h3_ccu_init, + .instance_size = sizeof(AwH3ClockCtlState), + .class_init = allwinner_h3_ccu_class_init, +}; + +static void allwinner_h3_ccu_register(void) +{ + type_register_static(&allwinner_h3_ccu_info); +} + +type_init(allwinner_h3_ccu_register) -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> The Allwinner H3 System on Chip contains multiple USB 2.0 bus connections which provide software access using the Enhanced Host Controller Interface (EHCI) and Open Host Controller Interface (OHCI) interfaces. This commit adds support for both interfaces in the Allwinner H3 System on Chip. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Gerd Hoffmann <kraxel@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20200311221854.30370-5-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/usb/hcd-ehci.h | 1 + include/hw/arm/allwinner-h3.h | 8 +++++++ hw/arm/allwinner-h3.c | 44 +++++++++++++++++++++++++++++++++++ hw/usb/hcd-ehci-sysbus.c | 17 ++++++++++++++ hw/arm/Kconfig | 2 ++ 5 files changed, 72 insertions(+) diff --git a/hw/usb/hcd-ehci.h b/hw/usb/hcd-ehci.h index XXXXXXX..XXXXXXX 100644 --- a/hw/usb/hcd-ehci.h +++ b/hw/usb/hcd-ehci.h @@ -XXX,XX +XXX,XX @@ typedef struct EHCIPCIState { #define TYPE_SYS_BUS_EHCI "sysbus-ehci-usb" #define TYPE_PLATFORM_EHCI "platform-ehci-usb" #define TYPE_EXYNOS4210_EHCI "exynos4210-ehci-usb" +#define TYPE_AW_H3_EHCI "aw-h3-ehci-usb" #define TYPE_TEGRA2_EHCI "tegra2-ehci-usb" #define TYPE_PPC4xx_EHCI "ppc4xx-ehci-usb" #define TYPE_FUSBH200_EHCI "fusbh200-ehci-usb" diff --git a/include/hw/arm/allwinner-h3.h b/include/hw/arm/allwinner-h3.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-h3.h +++ b/include/hw/arm/allwinner-h3.h @@ -XXX,XX +XXX,XX @@ enum { AW_H3_SRAM_A1, AW_H3_SRAM_A2, AW_H3_SRAM_C, + AW_H3_EHCI0, + AW_H3_OHCI0, + AW_H3_EHCI1, + AW_H3_OHCI1, + AW_H3_EHCI2, + AW_H3_OHCI2, + AW_H3_EHCI3, + AW_H3_OHCI3, AW_H3_CCU, AW_H3_PIT, AW_H3_UART0, diff --git a/hw/arm/allwinner-h3.c b/hw/arm/allwinner-h3.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/allwinner-h3.c +++ b/hw/arm/allwinner-h3.c @@ -XXX,XX +XXX,XX @@ #include "hw/sysbus.h" #include "hw/char/serial.h" #include "hw/misc/unimp.h" +#include "hw/usb/hcd-ehci.h" #include "sysemu/sysemu.h" #include "hw/arm/allwinner-h3.h" @@ -XXX,XX +XXX,XX @@ const hwaddr allwinner_h3_memmap[] = { [AW_H3_SRAM_A1] = 0x00000000, [AW_H3_SRAM_A2] = 0x00044000, [AW_H3_SRAM_C] = 0x00010000, + [AW_H3_EHCI0] = 0x01c1a000, + [AW_H3_OHCI0] = 0x01c1a400, + [AW_H3_EHCI1] = 0x01c1b000, + [AW_H3_OHCI1] = 0x01c1b400, + [AW_H3_EHCI2] = 0x01c1c000, + [AW_H3_OHCI2] = 0x01c1c400, + [AW_H3_EHCI3] = 0x01c1d000, + [AW_H3_OHCI3] = 0x01c1d400, [AW_H3_CCU] = 0x01c20000, [AW_H3_PIT] = 0x01c20c00, [AW_H3_UART0] = 0x01c28000, @@ -XXX,XX +XXX,XX @@ enum { AW_H3_GIC_SPI_UART3 = 3, AW_H3_GIC_SPI_TIMER0 = 18, AW_H3_GIC_SPI_TIMER1 = 19, + AW_H3_GIC_SPI_EHCI0 = 72, + AW_H3_GIC_SPI_OHCI0 = 73, + AW_H3_GIC_SPI_EHCI1 = 74, + AW_H3_GIC_SPI_OHCI1 = 75, + AW_H3_GIC_SPI_EHCI2 = 76, + AW_H3_GIC_SPI_OHCI2 = 77, + AW_H3_GIC_SPI_EHCI3 = 78, + AW_H3_GIC_SPI_OHCI3 = 79, }; /* Allwinner H3 general constants */ @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_realize(DeviceState *dev, Error **errp) qdev_init_nofail(DEVICE(&s->ccu)); sysbus_mmio_map(SYS_BUS_DEVICE(&s->ccu), 0, s->memmap[AW_H3_CCU]); + /* Universal Serial Bus */ + sysbus_create_simple(TYPE_AW_H3_EHCI, s->memmap[AW_H3_EHCI0], + qdev_get_gpio_in(DEVICE(&s->gic), + AW_H3_GIC_SPI_EHCI0)); + sysbus_create_simple(TYPE_AW_H3_EHCI, s->memmap[AW_H3_EHCI1], + qdev_get_gpio_in(DEVICE(&s->gic), + AW_H3_GIC_SPI_EHCI1)); + sysbus_create_simple(TYPE_AW_H3_EHCI, s->memmap[AW_H3_EHCI2], + qdev_get_gpio_in(DEVICE(&s->gic), + AW_H3_GIC_SPI_EHCI2)); + sysbus_create_simple(TYPE_AW_H3_EHCI, s->memmap[AW_H3_EHCI3], + qdev_get_gpio_in(DEVICE(&s->gic), + AW_H3_GIC_SPI_EHCI3)); + + sysbus_create_simple("sysbus-ohci", s->memmap[AW_H3_OHCI0], + qdev_get_gpio_in(DEVICE(&s->gic), + AW_H3_GIC_SPI_OHCI0)); + sysbus_create_simple("sysbus-ohci", s->memmap[AW_H3_OHCI1], + qdev_get_gpio_in(DEVICE(&s->gic), + AW_H3_GIC_SPI_OHCI1)); + sysbus_create_simple("sysbus-ohci", s->memmap[AW_H3_OHCI2], + qdev_get_gpio_in(DEVICE(&s->gic), + AW_H3_GIC_SPI_OHCI2)); + sysbus_create_simple("sysbus-ohci", s->memmap[AW_H3_OHCI3], + qdev_get_gpio_in(DEVICE(&s->gic), + AW_H3_GIC_SPI_OHCI3)); + /* UART0. For future clocktree API: All UARTS are connected to APB2_CLK. */ serial_mm_init(get_system_memory(), s->memmap[AW_H3_UART0], 2, qdev_get_gpio_in(DEVICE(&s->gic), AW_H3_GIC_SPI_UART0), diff --git a/hw/usb/hcd-ehci-sysbus.c b/hw/usb/hcd-ehci-sysbus.c index XXXXXXX..XXXXXXX 100644 --- a/hw/usb/hcd-ehci-sysbus.c +++ b/hw/usb/hcd-ehci-sysbus.c @@ -XXX,XX +XXX,XX @@ static const TypeInfo ehci_exynos4210_type_info = { .class_init = ehci_exynos4210_class_init, }; +static void ehci_aw_h3_class_init(ObjectClass *oc, void *data) +{ + SysBusEHCIClass *sec = SYS_BUS_EHCI_CLASS(oc); + DeviceClass *dc = DEVICE_CLASS(oc); + + sec->capsbase = 0x0; + sec->opregbase = 0x10; + set_bit(DEVICE_CATEGORY_USB, dc->categories); +} + +static const TypeInfo ehci_aw_h3_type_info = { + .name = TYPE_AW_H3_EHCI, + .parent = TYPE_SYS_BUS_EHCI, + .class_init = ehci_aw_h3_class_init, +}; + static void ehci_tegra2_class_init(ObjectClass *oc, void *data) { SysBusEHCIClass *sec = SYS_BUS_EHCI_CLASS(oc); @@ -XXX,XX +XXX,XX @@ static void ehci_sysbus_register_types(void) type_register_static(&ehci_type_info); type_register_static(&ehci_platform_type_info); type_register_static(&ehci_exynos4210_type_info); + type_register_static(&ehci_aw_h3_type_info); type_register_static(&ehci_tegra2_type_info); type_register_static(&ehci_ppc4xx_type_info); type_register_static(&ehci_fusbh200_type_info); diff --git a/hw/arm/Kconfig b/hw/arm/Kconfig index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/Kconfig +++ b/hw/arm/Kconfig @@ -XXX,XX +XXX,XX @@ config ALLWINNER_H3 select ARM_TIMER select ARM_GIC select UNIMP + select USB_OHCI + select USB_EHCI_SYSBUS config RASPI bool -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> The Allwinner H3 System on Chip has an System Control module that provides system wide generic controls and device information. This commit adds support for the Allwinner H3 System Control module. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20200311221854.30370-6-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/misc/Makefile.objs | 1 + include/hw/arm/allwinner-h3.h | 3 + include/hw/misc/allwinner-h3-sysctrl.h | 67 ++++++++++++ hw/arm/allwinner-h3.c | 9 +- hw/misc/allwinner-h3-sysctrl.c | 140 +++++++++++++++++++++++++ 5 files changed, 219 insertions(+), 1 deletion(-) create mode 100644 include/hw/misc/allwinner-h3-sysctrl.h create mode 100644 hw/misc/allwinner-h3-sysctrl.c diff --git a/hw/misc/Makefile.objs b/hw/misc/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/Makefile.objs +++ b/hw/misc/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_MACIO) += macio/ common-obj-$(CONFIG_IVSHMEM_DEVICE) += ivshmem.o common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3-ccu.o +common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3-sysctrl.o common-obj-$(CONFIG_REALVIEW) += arm_sysctl.o common-obj-$(CONFIG_NSERIES) += cbus.o common-obj-$(CONFIG_ECCMEMCTL) += eccmemctl.o diff --git a/include/hw/arm/allwinner-h3.h b/include/hw/arm/allwinner-h3.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-h3.h +++ b/include/hw/arm/allwinner-h3.h @@ -XXX,XX +XXX,XX @@ #include "hw/timer/allwinner-a10-pit.h" #include "hw/intc/arm_gic.h" #include "hw/misc/allwinner-h3-ccu.h" +#include "hw/misc/allwinner-h3-sysctrl.h" #include "target/arm/cpu.h" /** @@ -XXX,XX +XXX,XX @@ enum { AW_H3_SRAM_A1, AW_H3_SRAM_A2, AW_H3_SRAM_C, + AW_H3_SYSCTRL, AW_H3_EHCI0, AW_H3_OHCI0, AW_H3_EHCI1, @@ -XXX,XX +XXX,XX @@ typedef struct AwH3State { const hwaddr *memmap; AwA10PITState timer; AwH3ClockCtlState ccu; + AwH3SysCtrlState sysctrl; GICState gic; MemoryRegion sram_a1; MemoryRegion sram_a2; diff --git a/include/hw/misc/allwinner-h3-sysctrl.h b/include/hw/misc/allwinner-h3-sysctrl.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/misc/allwinner-h3-sysctrl.h @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner H3 System Control emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef HW_MISC_ALLWINNER_H3_SYSCTRL_H +#define HW_MISC_ALLWINNER_H3_SYSCTRL_H + +#include "qom/object.h" +#include "hw/sysbus.h" + +/** + * @name Constants + * @{ + */ + +/** Highest register address used by System Control device */ +#define AW_H3_SYSCTRL_REGS_MAXADDR (0x30) + +/** Total number of known registers */ +#define AW_H3_SYSCTRL_REGS_NUM ((AW_H3_SYSCTRL_REGS_MAXADDR / \ + sizeof(uint32_t)) + 1) + +/** @} */ + +/** + * @name Object model + * @{ + */ + +#define TYPE_AW_H3_SYSCTRL "allwinner-h3-sysctrl" +#define AW_H3_SYSCTRL(obj) \ + OBJECT_CHECK(AwH3SysCtrlState, (obj), TYPE_AW_H3_SYSCTRL) + +/** @} */ + +/** + * Allwinner H3 System Control object instance state + */ +typedef struct AwH3SysCtrlState { + /*< private >*/ + SysBusDevice parent_obj; + /*< public >*/ + + /** Maps I/O registers in physical memory */ + MemoryRegion iomem; + + /** Array of hardware registers */ + uint32_t regs[AW_H3_SYSCTRL_REGS_NUM]; + +} AwH3SysCtrlState; + +#endif /* HW_MISC_ALLWINNER_H3_SYSCTRL_H */ diff --git a/hw/arm/allwinner-h3.c b/hw/arm/allwinner-h3.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/allwinner-h3.c +++ b/hw/arm/allwinner-h3.c @@ -XXX,XX +XXX,XX @@ const hwaddr allwinner_h3_memmap[] = { [AW_H3_SRAM_A1] = 0x00000000, [AW_H3_SRAM_A2] = 0x00044000, [AW_H3_SRAM_C] = 0x00010000, + [AW_H3_SYSCTRL] = 0x01c00000, [AW_H3_EHCI0] = 0x01c1a000, [AW_H3_OHCI0] = 0x01c1a400, [AW_H3_EHCI1] = 0x01c1b000, @@ -XXX,XX +XXX,XX @@ struct AwH3Unimplemented { } unimplemented[] = { { "d-engine", 0x01000000, 4 * MiB }, { "d-inter", 0x01400000, 128 * KiB }, - { "syscon", 0x01c00000, 4 * KiB }, { "dma", 0x01c02000, 4 * KiB }, { "nfdc", 0x01c03000, 4 * KiB }, { "ts", 0x01c06000, 4 * KiB }, @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_init(Object *obj) sysbus_init_child_obj(obj, "ccu", &s->ccu, sizeof(s->ccu), TYPE_AW_H3_CCU); + + sysbus_init_child_obj(obj, "sysctrl", &s->sysctrl, sizeof(s->sysctrl), + TYPE_AW_H3_SYSCTRL); } static void allwinner_h3_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_realize(DeviceState *dev, Error **errp) qdev_init_nofail(DEVICE(&s->ccu)); sysbus_mmio_map(SYS_BUS_DEVICE(&s->ccu), 0, s->memmap[AW_H3_CCU]); + /* System Control */ + qdev_init_nofail(DEVICE(&s->sysctrl)); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->sysctrl), 0, s->memmap[AW_H3_SYSCTRL]); + /* Universal Serial Bus */ sysbus_create_simple(TYPE_AW_H3_EHCI, s->memmap[AW_H3_EHCI0], qdev_get_gpio_in(DEVICE(&s->gic), diff --git a/hw/misc/allwinner-h3-sysctrl.c b/hw/misc/allwinner-h3-sysctrl.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/misc/allwinner-h3-sysctrl.c @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner H3 System Control emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu/units.h" +#include "hw/sysbus.h" +#include "migration/vmstate.h" +#include "qemu/log.h" +#include "qemu/module.h" +#include "hw/misc/allwinner-h3-sysctrl.h" + +/* System Control register offsets */ +enum { + REG_VER = 0x24, /* Version */ + REG_EMAC_PHY_CLK = 0x30, /* EMAC PHY Clock */ +}; + +#define REG_INDEX(offset) (offset / sizeof(uint32_t)) + +/* System Control register reset values */ +enum { + REG_VER_RST = 0x0, + REG_EMAC_PHY_CLK_RST = 0x58000, +}; + +static uint64_t allwinner_h3_sysctrl_read(void *opaque, hwaddr offset, + unsigned size) +{ + const AwH3SysCtrlState *s = AW_H3_SYSCTRL(opaque); + const uint32_t idx = REG_INDEX(offset); + + if (idx >= AW_H3_SYSCTRL_REGS_NUM) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + return 0; + } + + return s->regs[idx]; +} + +static void allwinner_h3_sysctrl_write(void *opaque, hwaddr offset, + uint64_t val, unsigned size) +{ + AwH3SysCtrlState *s = AW_H3_SYSCTRL(opaque); + const uint32_t idx = REG_INDEX(offset); + + if (idx >= AW_H3_SYSCTRL_REGS_NUM) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + return; + } + + switch (offset) { + case REG_VER: /* Version */ + break; + default: + s->regs[idx] = (uint32_t) val; + break; + } +} + +static const MemoryRegionOps allwinner_h3_sysctrl_ops = { + .read = allwinner_h3_sysctrl_read, + .write = allwinner_h3_sysctrl_write, + .endianness = DEVICE_NATIVE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, + .impl.min_access_size = 4, +}; + +static void allwinner_h3_sysctrl_reset(DeviceState *dev) +{ + AwH3SysCtrlState *s = AW_H3_SYSCTRL(dev); + + /* Set default values for registers */ + s->regs[REG_INDEX(REG_VER)] = REG_VER_RST; + s->regs[REG_INDEX(REG_EMAC_PHY_CLK)] = REG_EMAC_PHY_CLK_RST; +} + +static void allwinner_h3_sysctrl_init(Object *obj) +{ + SysBusDevice *sbd = SYS_BUS_DEVICE(obj); + AwH3SysCtrlState *s = AW_H3_SYSCTRL(obj); + + /* Memory mapping */ + memory_region_init_io(&s->iomem, OBJECT(s), &allwinner_h3_sysctrl_ops, s, + TYPE_AW_H3_SYSCTRL, 4 * KiB); + sysbus_init_mmio(sbd, &s->iomem); +} + +static const VMStateDescription allwinner_h3_sysctrl_vmstate = { + .name = "allwinner-h3-sysctrl", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32_ARRAY(regs, AwH3SysCtrlState, AW_H3_SYSCTRL_REGS_NUM), + VMSTATE_END_OF_LIST() + } +}; + +static void allwinner_h3_sysctrl_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->reset = allwinner_h3_sysctrl_reset; + dc->vmsd = &allwinner_h3_sysctrl_vmstate; +} + +static const TypeInfo allwinner_h3_sysctrl_info = { + .name = TYPE_AW_H3_SYSCTRL, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_init = allwinner_h3_sysctrl_init, + .instance_size = sizeof(AwH3SysCtrlState), + .class_init = allwinner_h3_sysctrl_class_init, +}; + +static void allwinner_h3_sysctrl_register(void) +{ + type_register_static(&allwinner_h3_sysctrl_info); +} + +type_init(allwinner_h3_sysctrl_register) -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> Various Allwinner System on Chip designs contain multiple processors that can be configured and reset using the generic CPU Configuration module interface. This commit adds support for the Allwinner CPU configuration interface which emulates the following features: * CPU reset * CPU status Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20200311221854.30370-7-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/misc/Makefile.objs | 1 + include/hw/arm/allwinner-h3.h | 3 + include/hw/misc/allwinner-cpucfg.h | 52 ++++++ hw/arm/allwinner-h3.c | 9 +- hw/misc/allwinner-cpucfg.c | 282 +++++++++++++++++++++++++++++ hw/misc/trace-events | 5 + 6 files changed, 351 insertions(+), 1 deletion(-) create mode 100644 include/hw/misc/allwinner-cpucfg.h create mode 100644 hw/misc/allwinner-cpucfg.c diff --git a/hw/misc/Makefile.objs b/hw/misc/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/Makefile.objs +++ b/hw/misc/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_MACIO) += macio/ common-obj-$(CONFIG_IVSHMEM_DEVICE) += ivshmem.o common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3-ccu.o +obj-$(CONFIG_ALLWINNER_H3) += allwinner-cpucfg.o common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3-sysctrl.o common-obj-$(CONFIG_REALVIEW) += arm_sysctl.o common-obj-$(CONFIG_NSERIES) += cbus.o diff --git a/include/hw/arm/allwinner-h3.h b/include/hw/arm/allwinner-h3.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-h3.h +++ b/include/hw/arm/allwinner-h3.h @@ -XXX,XX +XXX,XX @@ #include "hw/timer/allwinner-a10-pit.h" #include "hw/intc/arm_gic.h" #include "hw/misc/allwinner-h3-ccu.h" +#include "hw/misc/allwinner-cpucfg.h" #include "hw/misc/allwinner-h3-sysctrl.h" #include "target/arm/cpu.h" @@ -XXX,XX +XXX,XX @@ enum { AW_H3_GIC_CPU, AW_H3_GIC_HYP, AW_H3_GIC_VCPU, + AW_H3_CPUCFG, AW_H3_SDRAM }; @@ -XXX,XX +XXX,XX @@ typedef struct AwH3State { const hwaddr *memmap; AwA10PITState timer; AwH3ClockCtlState ccu; + AwCpuCfgState cpucfg; AwH3SysCtrlState sysctrl; GICState gic; MemoryRegion sram_a1; diff --git a/include/hw/misc/allwinner-cpucfg.h b/include/hw/misc/allwinner-cpucfg.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/misc/allwinner-cpucfg.h @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner CPU Configuration Module emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef HW_MISC_ALLWINNER_CPUCFG_H +#define HW_MISC_ALLWINNER_CPUCFG_H + +#include "qom/object.h" +#include "hw/sysbus.h" + +/** + * Object model + * @{ + */ + +#define TYPE_AW_CPUCFG "allwinner-cpucfg" +#define AW_CPUCFG(obj) \ + OBJECT_CHECK(AwCpuCfgState, (obj), TYPE_AW_CPUCFG) + +/** @} */ + +/** + * Allwinner CPU Configuration Module instance state + */ +typedef struct AwCpuCfgState { + /*< private >*/ + SysBusDevice parent_obj; + /*< public >*/ + + MemoryRegion iomem; + uint32_t gen_ctrl; + uint32_t super_standby; + uint32_t entry_addr; + +} AwCpuCfgState; + +#endif /* HW_MISC_ALLWINNER_CPUCFG_H */ diff --git a/hw/arm/allwinner-h3.c b/hw/arm/allwinner-h3.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/allwinner-h3.c +++ b/hw/arm/allwinner-h3.c @@ -XXX,XX +XXX,XX @@ const hwaddr allwinner_h3_memmap[] = { [AW_H3_GIC_CPU] = 0x01c82000, [AW_H3_GIC_HYP] = 0x01c84000, [AW_H3_GIC_VCPU] = 0x01c86000, + [AW_H3_CPUCFG] = 0x01f01c00, [AW_H3_SDRAM] = 0x40000000 }; @@ -XXX,XX +XXX,XX @@ struct AwH3Unimplemented { { "r_wdog", 0x01f01000, 1 * KiB }, { "r_prcm", 0x01f01400, 1 * KiB }, { "r_twd", 0x01f01800, 1 * KiB }, - { "r_cpucfg", 0x01f01c00, 1 * KiB }, { "r_cir-rx", 0x01f02000, 1 * KiB }, { "r_twi", 0x01f02400, 1 * KiB }, { "r_uart", 0x01f02800, 1 * KiB }, @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_init(Object *obj) sysbus_init_child_obj(obj, "sysctrl", &s->sysctrl, sizeof(s->sysctrl), TYPE_AW_H3_SYSCTRL); + + sysbus_init_child_obj(obj, "cpucfg", &s->cpucfg, sizeof(s->cpucfg), + TYPE_AW_CPUCFG); } static void allwinner_h3_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_realize(DeviceState *dev, Error **errp) qdev_init_nofail(DEVICE(&s->sysctrl)); sysbus_mmio_map(SYS_BUS_DEVICE(&s->sysctrl), 0, s->memmap[AW_H3_SYSCTRL]); + /* CPU Configuration */ + qdev_init_nofail(DEVICE(&s->cpucfg)); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->cpucfg), 0, s->memmap[AW_H3_CPUCFG]); + /* Universal Serial Bus */ sysbus_create_simple(TYPE_AW_H3_EHCI, s->memmap[AW_H3_EHCI0], qdev_get_gpio_in(DEVICE(&s->gic), diff --git a/hw/misc/allwinner-cpucfg.c b/hw/misc/allwinner-cpucfg.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/misc/allwinner-cpucfg.c @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner CPU Configuration Module emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu/units.h" +#include "hw/sysbus.h" +#include "migration/vmstate.h" +#include "qemu/log.h" +#include "qemu/module.h" +#include "qemu/error-report.h" +#include "qemu/timer.h" +#include "hw/core/cpu.h" +#include "target/arm/arm-powerctl.h" +#include "target/arm/cpu.h" +#include "hw/misc/allwinner-cpucfg.h" +#include "trace.h" + +/* CPUCFG register offsets */ +enum { + REG_CPUS_RST_CTRL = 0x0000, /* CPUs Reset Control */ + REG_CPU0_RST_CTRL = 0x0040, /* CPU#0 Reset Control */ + REG_CPU0_CTRL = 0x0044, /* CPU#0 Control */ + REG_CPU0_STATUS = 0x0048, /* CPU#0 Status */ + REG_CPU1_RST_CTRL = 0x0080, /* CPU#1 Reset Control */ + REG_CPU1_CTRL = 0x0084, /* CPU#1 Control */ + REG_CPU1_STATUS = 0x0088, /* CPU#1 Status */ + REG_CPU2_RST_CTRL = 0x00C0, /* CPU#2 Reset Control */ + REG_CPU2_CTRL = 0x00C4, /* CPU#2 Control */ + REG_CPU2_STATUS = 0x00C8, /* CPU#2 Status */ + REG_CPU3_RST_CTRL = 0x0100, /* CPU#3 Reset Control */ + REG_CPU3_CTRL = 0x0104, /* CPU#3 Control */ + REG_CPU3_STATUS = 0x0108, /* CPU#3 Status */ + REG_CPU_SYS_RST = 0x0140, /* CPU System Reset */ + REG_CLK_GATING = 0x0144, /* CPU Clock Gating */ + REG_GEN_CTRL = 0x0184, /* General Control */ + REG_SUPER_STANDBY = 0x01A0, /* Super Standby Flag */ + REG_ENTRY_ADDR = 0x01A4, /* Reset Entry Address */ + REG_DBG_EXTERN = 0x01E4, /* Debug External */ + REG_CNT64_CTRL = 0x0280, /* 64-bit Counter Control */ + REG_CNT64_LOW = 0x0284, /* 64-bit Counter Low */ + REG_CNT64_HIGH = 0x0288, /* 64-bit Counter High */ +}; + +/* CPUCFG register flags */ +enum { + CPUX_RESET_RELEASED = ((1 << 1) | (1 << 0)), + CPUX_STATUS_SMP = (1 << 0), + CPU_SYS_RESET_RELEASED = (1 << 0), + CLK_GATING_ENABLE = ((1 << 8) | 0xF), +}; + +/* CPUCFG register reset values */ +enum { + REG_CLK_GATING_RST = 0x0000010F, + REG_GEN_CTRL_RST = 0x00000020, + REG_SUPER_STANDBY_RST = 0x0, + REG_CNT64_CTRL_RST = 0x0, +}; + +/* CPUCFG constants */ +enum { + CPU_EXCEPTION_LEVEL_ON_RESET = 3, /* EL3 */ +}; + +static void allwinner_cpucfg_cpu_reset(AwCpuCfgState *s, uint8_t cpu_id) +{ + int ret; + + trace_allwinner_cpucfg_cpu_reset(cpu_id, s->entry_addr); + + ARMCPU *target_cpu = ARM_CPU(arm_get_cpu_by_id(cpu_id)); + if (!target_cpu) { + /* + * Called with a bogus value for cpu_id. Guest error will + * already have been logged, we can simply return here. + */ + return; + } + bool target_aa64 = arm_feature(&target_cpu->env, ARM_FEATURE_AARCH64); + + ret = arm_set_cpu_on(cpu_id, s->entry_addr, 0, + CPU_EXCEPTION_LEVEL_ON_RESET, target_aa64); + if (ret != QEMU_ARM_POWERCTL_RET_SUCCESS) { + error_report("%s: failed to bring up CPU %d: err %d", + __func__, cpu_id, ret); + return; + } +} + +static uint64_t allwinner_cpucfg_read(void *opaque, hwaddr offset, + unsigned size) +{ + const AwCpuCfgState *s = AW_CPUCFG(opaque); + uint64_t val = 0; + + switch (offset) { + case REG_CPUS_RST_CTRL: /* CPUs Reset Control */ + case REG_CPU_SYS_RST: /* CPU System Reset */ + val = CPU_SYS_RESET_RELEASED; + break; + case REG_CPU0_RST_CTRL: /* CPU#0 Reset Control */ + case REG_CPU1_RST_CTRL: /* CPU#1 Reset Control */ + case REG_CPU2_RST_CTRL: /* CPU#2 Reset Control */ + case REG_CPU3_RST_CTRL: /* CPU#3 Reset Control */ + val = CPUX_RESET_RELEASED; + break; + case REG_CPU0_CTRL: /* CPU#0 Control */ + case REG_CPU1_CTRL: /* CPU#1 Control */ + case REG_CPU2_CTRL: /* CPU#2 Control */ + case REG_CPU3_CTRL: /* CPU#3 Control */ + val = 0; + break; + case REG_CPU0_STATUS: /* CPU#0 Status */ + case REG_CPU1_STATUS: /* CPU#1 Status */ + case REG_CPU2_STATUS: /* CPU#2 Status */ + case REG_CPU3_STATUS: /* CPU#3 Status */ + val = CPUX_STATUS_SMP; + break; + case REG_CLK_GATING: /* CPU Clock Gating */ + val = CLK_GATING_ENABLE; + break; + case REG_GEN_CTRL: /* General Control */ + val = s->gen_ctrl; + break; + case REG_SUPER_STANDBY: /* Super Standby Flag */ + val = s->super_standby; + break; + case REG_ENTRY_ADDR: /* Reset Entry Address */ + val = s->entry_addr; + break; + case REG_DBG_EXTERN: /* Debug External */ + case REG_CNT64_CTRL: /* 64-bit Counter Control */ + case REG_CNT64_LOW: /* 64-bit Counter Low */ + case REG_CNT64_HIGH: /* 64-bit Counter High */ + qemu_log_mask(LOG_UNIMP, "%s: unimplemented register at 0x%04x\n", + __func__, (uint32_t)offset); + break; + default: + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + break; + } + + trace_allwinner_cpucfg_read(offset, val, size); + + return val; +} + +static void allwinner_cpucfg_write(void *opaque, hwaddr offset, + uint64_t val, unsigned size) +{ + AwCpuCfgState *s = AW_CPUCFG(opaque); + + trace_allwinner_cpucfg_write(offset, val, size); + + switch (offset) { + case REG_CPUS_RST_CTRL: /* CPUs Reset Control */ + case REG_CPU_SYS_RST: /* CPU System Reset */ + break; + case REG_CPU0_RST_CTRL: /* CPU#0 Reset Control */ + case REG_CPU1_RST_CTRL: /* CPU#1 Reset Control */ + case REG_CPU2_RST_CTRL: /* CPU#2 Reset Control */ + case REG_CPU3_RST_CTRL: /* CPU#3 Reset Control */ + if (val) { + allwinner_cpucfg_cpu_reset(s, (offset - REG_CPU0_RST_CTRL) >> 6); + } + break; + case REG_CPU0_CTRL: /* CPU#0 Control */ + case REG_CPU1_CTRL: /* CPU#1 Control */ + case REG_CPU2_CTRL: /* CPU#2 Control */ + case REG_CPU3_CTRL: /* CPU#3 Control */ + case REG_CPU0_STATUS: /* CPU#0 Status */ + case REG_CPU1_STATUS: /* CPU#1 Status */ + case REG_CPU2_STATUS: /* CPU#2 Status */ + case REG_CPU3_STATUS: /* CPU#3 Status */ + case REG_CLK_GATING: /* CPU Clock Gating */ + break; + case REG_GEN_CTRL: /* General Control */ + s->gen_ctrl = val; + break; + case REG_SUPER_STANDBY: /* Super Standby Flag */ + s->super_standby = val; + break; + case REG_ENTRY_ADDR: /* Reset Entry Address */ + s->entry_addr = val; + break; + case REG_DBG_EXTERN: /* Debug External */ + case REG_CNT64_CTRL: /* 64-bit Counter Control */ + case REG_CNT64_LOW: /* 64-bit Counter Low */ + case REG_CNT64_HIGH: /* 64-bit Counter High */ + qemu_log_mask(LOG_UNIMP, "%s: unimplemented register at 0x%04x\n", + __func__, (uint32_t)offset); + break; + default: + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + break; + } +} + +static const MemoryRegionOps allwinner_cpucfg_ops = { + .read = allwinner_cpucfg_read, + .write = allwinner_cpucfg_write, + .endianness = DEVICE_NATIVE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, + .impl.min_access_size = 4, +}; + +static void allwinner_cpucfg_reset(DeviceState *dev) +{ + AwCpuCfgState *s = AW_CPUCFG(dev); + + /* Set default values for registers */ + s->gen_ctrl = REG_GEN_CTRL_RST; + s->super_standby = REG_SUPER_STANDBY_RST; + s->entry_addr = 0; +} + +static void allwinner_cpucfg_init(Object *obj) +{ + SysBusDevice *sbd = SYS_BUS_DEVICE(obj); + AwCpuCfgState *s = AW_CPUCFG(obj); + + /* Memory mapping */ + memory_region_init_io(&s->iomem, OBJECT(s), &allwinner_cpucfg_ops, s, + TYPE_AW_CPUCFG, 1 * KiB); + sysbus_init_mmio(sbd, &s->iomem); +} + +static const VMStateDescription allwinner_cpucfg_vmstate = { + .name = "allwinner-cpucfg", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32(gen_ctrl, AwCpuCfgState), + VMSTATE_UINT32(super_standby, AwCpuCfgState), + VMSTATE_UINT32(entry_addr, AwCpuCfgState), + VMSTATE_END_OF_LIST() + } +}; + +static void allwinner_cpucfg_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->reset = allwinner_cpucfg_reset; + dc->vmsd = &allwinner_cpucfg_vmstate; +} + +static const TypeInfo allwinner_cpucfg_info = { + .name = TYPE_AW_CPUCFG, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_init = allwinner_cpucfg_init, + .instance_size = sizeof(AwCpuCfgState), + .class_init = allwinner_cpucfg_class_init, +}; + +static void allwinner_cpucfg_register(void) +{ + type_register_static(&allwinner_cpucfg_info); +} + +type_init(allwinner_cpucfg_register) diff --git a/hw/misc/trace-events b/hw/misc/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/trace-events +++ b/hw/misc/trace-events @@ -XXX,XX +XXX,XX @@ # See docs/devel/tracing.txt for syntax documentation. +# allwinner-cpucfg.c +allwinner_cpucfg_cpu_reset(uint8_t cpu_id, uint32_t reset_addr) "id %u, reset_addr 0x%" PRIu32 +allwinner_cpucfg_read(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 +allwinner_cpucfg_write(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 + # eccmemctl.c ecc_mem_writel_mer(uint32_t val) "Write memory enable 0x%08x" ecc_mem_writel_mdr(uint32_t val) "Write memory delay 0x%08x" -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> The Security Identifier device found in various Allwinner System on Chip designs gives applications a per-board unique identifier. This commit adds support for the Allwinner Security Identifier using a 128-bit UUID value as input. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20200311221854.30370-8-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/misc/Makefile.objs | 1 + include/hw/arm/allwinner-h3.h | 3 + include/hw/misc/allwinner-sid.h | 60 ++++++++++++ hw/arm/allwinner-h3.c | 11 ++- hw/arm/orangepi.c | 8 ++ hw/misc/allwinner-sid.c | 168 ++++++++++++++++++++++++++++++++ hw/misc/trace-events | 4 + 7 files changed, 254 insertions(+), 1 deletion(-) create mode 100644 include/hw/misc/allwinner-sid.h create mode 100644 hw/misc/allwinner-sid.c diff --git a/hw/misc/Makefile.objs b/hw/misc/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/Makefile.objs +++ b/hw/misc/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_IVSHMEM_DEVICE) += ivshmem.o common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3-ccu.o obj-$(CONFIG_ALLWINNER_H3) += allwinner-cpucfg.o common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3-sysctrl.o +common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-sid.o common-obj-$(CONFIG_REALVIEW) += arm_sysctl.o common-obj-$(CONFIG_NSERIES) += cbus.o common-obj-$(CONFIG_ECCMEMCTL) += eccmemctl.o diff --git a/include/hw/arm/allwinner-h3.h b/include/hw/arm/allwinner-h3.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-h3.h +++ b/include/hw/arm/allwinner-h3.h @@ -XXX,XX +XXX,XX @@ #include "hw/misc/allwinner-h3-ccu.h" #include "hw/misc/allwinner-cpucfg.h" #include "hw/misc/allwinner-h3-sysctrl.h" +#include "hw/misc/allwinner-sid.h" #include "target/arm/cpu.h" /** @@ -XXX,XX +XXX,XX @@ enum { AW_H3_SRAM_A2, AW_H3_SRAM_C, AW_H3_SYSCTRL, + AW_H3_SID, AW_H3_EHCI0, AW_H3_OHCI0, AW_H3_EHCI1, @@ -XXX,XX +XXX,XX @@ typedef struct AwH3State { AwH3ClockCtlState ccu; AwCpuCfgState cpucfg; AwH3SysCtrlState sysctrl; + AwSidState sid; GICState gic; MemoryRegion sram_a1; MemoryRegion sram_a2; diff --git a/include/hw/misc/allwinner-sid.h b/include/hw/misc/allwinner-sid.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/misc/allwinner-sid.h @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner Security ID emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef HW_MISC_ALLWINNER_SID_H +#define HW_MISC_ALLWINNER_SID_H + +#include "qom/object.h" +#include "hw/sysbus.h" +#include "qemu/uuid.h" + +/** + * Object model + * @{ + */ + +#define TYPE_AW_SID "allwinner-sid" +#define AW_SID(obj) \ + OBJECT_CHECK(AwSidState, (obj), TYPE_AW_SID) + +/** @} */ + +/** + * Allwinner Security ID object instance state + */ +typedef struct AwSidState { + /*< private >*/ + SysBusDevice parent_obj; + /*< public >*/ + + /** Maps I/O registers in physical memory */ + MemoryRegion iomem; + + /** Control register defines how and what to read */ + uint32_t control; + + /** RdKey register contains the data retrieved by the device */ + uint32_t rdkey; + + /** Stores the emulated device identifier */ + QemuUUID identifier; + +} AwSidState; + +#endif /* HW_MISC_ALLWINNER_SID_H */ diff --git a/hw/arm/allwinner-h3.c b/hw/arm/allwinner-h3.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/allwinner-h3.c +++ b/hw/arm/allwinner-h3.c @@ -XXX,XX +XXX,XX @@ const hwaddr allwinner_h3_memmap[] = { [AW_H3_SRAM_A2] = 0x00044000, [AW_H3_SRAM_C] = 0x00010000, [AW_H3_SYSCTRL] = 0x01c00000, + [AW_H3_SID] = 0x01c14000, [AW_H3_EHCI0] = 0x01c1a000, [AW_H3_OHCI0] = 0x01c1a400, [AW_H3_EHCI1] = 0x01c1b000, @@ -XXX,XX +XXX,XX @@ struct AwH3Unimplemented { { "mmc0", 0x01c0f000, 4 * KiB }, { "mmc1", 0x01c10000, 4 * KiB }, { "mmc2", 0x01c11000, 4 * KiB }, - { "sid", 0x01c14000, 1 * KiB }, { "crypto", 0x01c15000, 4 * KiB }, { "msgbox", 0x01c17000, 4 * KiB }, { "spinlock", 0x01c18000, 4 * KiB }, @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_init(Object *obj) sysbus_init_child_obj(obj, "cpucfg", &s->cpucfg, sizeof(s->cpucfg), TYPE_AW_CPUCFG); + + sysbus_init_child_obj(obj, "sid", &s->sid, sizeof(s->sid), + TYPE_AW_SID); + object_property_add_alias(obj, "identifier", OBJECT(&s->sid), + "identifier", &error_abort); } static void allwinner_h3_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_realize(DeviceState *dev, Error **errp) qdev_init_nofail(DEVICE(&s->cpucfg)); sysbus_mmio_map(SYS_BUS_DEVICE(&s->cpucfg), 0, s->memmap[AW_H3_CPUCFG]); + /* Security Identifier */ + qdev_init_nofail(DEVICE(&s->sid)); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->sid), 0, s->memmap[AW_H3_SID]); + /* Universal Serial Bus */ sysbus_create_simple(TYPE_AW_H3_EHCI, s->memmap[AW_H3_EHCI0], qdev_get_gpio_in(DEVICE(&s->gic), diff --git a/hw/arm/orangepi.c b/hw/arm/orangepi.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/orangepi.c +++ b/hw/arm/orangepi.c @@ -XXX,XX +XXX,XX @@ static void orangepi_init(MachineState *machine) object_property_set_int(OBJECT(h3), 24 * 1000 * 1000, "clk1-freq", &error_abort); + /* Setup SID properties. Currently using a default fixed SID identifier. */ + if (qemu_uuid_is_null(&h3->sid.identifier)) { + qdev_prop_set_string(DEVICE(h3), "identifier", + "02c00081-1111-2222-3333-000044556677"); + } else if (ldl_be_p(&h3->sid.identifier.data[0]) != 0x02c00081) { + warn_report("Security Identifier value does not include H3 prefix"); + } + /* Mark H3 object realized */ object_property_set_bool(OBJECT(h3), true, "realized", &error_abort); diff --git a/hw/misc/allwinner-sid.c b/hw/misc/allwinner-sid.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/misc/allwinner-sid.c @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner Security ID emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu/units.h" +#include "hw/sysbus.h" +#include "migration/vmstate.h" +#include "qemu/log.h" +#include "qemu/module.h" +#include "qemu/guest-random.h" +#include "qapi/error.h" +#include "hw/qdev-properties.h" +#include "hw/misc/allwinner-sid.h" +#include "trace.h" + +/* SID register offsets */ +enum { + REG_PRCTL = 0x40, /* Control */ + REG_RDKEY = 0x60, /* Read Key */ +}; + +/* SID register flags */ +enum { + REG_PRCTL_WRITE = 0x0002, /* Unknown write flag */ + REG_PRCTL_OP_LOCK = 0xAC00, /* Lock operation */ +}; + +static uint64_t allwinner_sid_read(void *opaque, hwaddr offset, + unsigned size) +{ + const AwSidState *s = AW_SID(opaque); + uint64_t val = 0; + + switch (offset) { + case REG_PRCTL: /* Control */ + val = s->control; + break; + case REG_RDKEY: /* Read Key */ + val = s->rdkey; + break; + default: + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + return 0; + } + + trace_allwinner_sid_read(offset, val, size); + + return val; +} + +static void allwinner_sid_write(void *opaque, hwaddr offset, + uint64_t val, unsigned size) +{ + AwSidState *s = AW_SID(opaque); + + trace_allwinner_sid_write(offset, val, size); + + switch (offset) { + case REG_PRCTL: /* Control */ + s->control = val; + + if ((s->control & REG_PRCTL_OP_LOCK) && + (s->control & REG_PRCTL_WRITE)) { + uint32_t id = s->control >> 16; + + if (id <= sizeof(QemuUUID) - sizeof(s->rdkey)) { + s->rdkey = ldl_be_p(&s->identifier.data[id]); + } + } + s->control &= ~REG_PRCTL_WRITE; + break; + case REG_RDKEY: /* Read Key */ + break; + default: + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + break; + } +} + +static const MemoryRegionOps allwinner_sid_ops = { + .read = allwinner_sid_read, + .write = allwinner_sid_write, + .endianness = DEVICE_NATIVE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, + .impl.min_access_size = 4, +}; + +static void allwinner_sid_reset(DeviceState *dev) +{ + AwSidState *s = AW_SID(dev); + + /* Set default values for registers */ + s->control = 0; + s->rdkey = 0; +} + +static void allwinner_sid_init(Object *obj) +{ + SysBusDevice *sbd = SYS_BUS_DEVICE(obj); + AwSidState *s = AW_SID(obj); + + /* Memory mapping */ + memory_region_init_io(&s->iomem, OBJECT(s), &allwinner_sid_ops, s, + TYPE_AW_SID, 1 * KiB); + sysbus_init_mmio(sbd, &s->iomem); +} + +static Property allwinner_sid_properties[] = { + DEFINE_PROP_UUID_NODEFAULT("identifier", AwSidState, identifier), + DEFINE_PROP_END_OF_LIST() +}; + +static const VMStateDescription allwinner_sid_vmstate = { + .name = "allwinner-sid", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32(control, AwSidState), + VMSTATE_UINT32(rdkey, AwSidState), + VMSTATE_UINT8_ARRAY_V(identifier.data, AwSidState, sizeof(QemuUUID), 1), + VMSTATE_END_OF_LIST() + } +}; + +static void allwinner_sid_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->reset = allwinner_sid_reset; + dc->vmsd = &allwinner_sid_vmstate; + device_class_set_props(dc, allwinner_sid_properties); +} + +static const TypeInfo allwinner_sid_info = { + .name = TYPE_AW_SID, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_init = allwinner_sid_init, + .instance_size = sizeof(AwSidState), + .class_init = allwinner_sid_class_init, +}; + +static void allwinner_sid_register(void) +{ + type_register_static(&allwinner_sid_info); +} + +type_init(allwinner_sid_register) diff --git a/hw/misc/trace-events b/hw/misc/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/trace-events +++ b/hw/misc/trace-events @@ -XXX,XX +XXX,XX @@ allwinner_cpucfg_cpu_reset(uint8_t cpu_id, uint32_t reset_addr) "id %u, reset_ad allwinner_cpucfg_read(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 allwinner_cpucfg_write(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 +# allwinner-sid.c +allwinner_sid_read(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 +allwinner_sid_write(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 + # eccmemctl.c ecc_mem_writel_mer(uint32_t val) "Write memory enable 0x%08x" ecc_mem_writel_mdr(uint32_t val) "Write memory delay 0x%08x" -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> The Allwinner System on Chip families sun4i and above contain an integrated storage controller for Secure Digital (SD) and Multi Media Card (MMC) interfaces. This commit adds support for the Allwinner SD/MMC storage controller with the following emulated features: * DMA transfers * Direct FIFO I/O * Short/Long format command responses * Auto-Stop command (CMD12) * Insert & remove card detection The following boards are extended with the SD host controller: * Cubieboard (hw/arm/cubieboard.c) * Orange Pi PC (hw/arm/orangepi.c) Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20200311221854.30370-9-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/sd/Makefile.objs | 1 + include/hw/arm/allwinner-a10.h | 2 + include/hw/arm/allwinner-h3.h | 3 + include/hw/sd/allwinner-sdhost.h | 135 +++++ hw/arm/allwinner-a10.c | 11 + hw/arm/allwinner-h3.c | 15 +- hw/arm/cubieboard.c | 15 + hw/arm/orangepi.c | 16 + hw/sd/allwinner-sdhost.c | 854 +++++++++++++++++++++++++++++++ hw/arm/Kconfig | 1 + hw/sd/trace-events | 7 + 11 files changed, 1059 insertions(+), 1 deletion(-) create mode 100644 include/hw/sd/allwinner-sdhost.h create mode 100644 hw/sd/allwinner-sdhost.c diff --git a/hw/sd/Makefile.objs b/hw/sd/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/sd/Makefile.objs +++ b/hw/sd/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_SD) += sd.o core.o sdmmc-internal.o common-obj-$(CONFIG_SDHCI) += sdhci.o common-obj-$(CONFIG_SDHCI_PCI) += sdhci-pci.o +common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-sdhost.o common-obj-$(CONFIG_MILKYMIST) += milkymist-memcard.o common-obj-$(CONFIG_OMAP) += omap_mmc.o common-obj-$(CONFIG_PXA2XX) += pxa2xx_mmci.o diff --git a/include/hw/arm/allwinner-a10.h b/include/hw/arm/allwinner-a10.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-a10.h +++ b/include/hw/arm/allwinner-a10.h @@ -XXX,XX +XXX,XX @@ #include "hw/timer/allwinner-a10-pit.h" #include "hw/intc/allwinner-a10-pic.h" #include "hw/net/allwinner_emac.h" +#include "hw/sd/allwinner-sdhost.h" #include "hw/ide/ahci.h" #include "hw/usb/hcd-ohci.h" #include "hw/usb/hcd-ehci.h" @@ -XXX,XX +XXX,XX @@ typedef struct AwA10State { AwA10PICState intc; AwEmacState emac; AllwinnerAHCIState sata; + AwSdHostState mmc0; MemoryRegion sram_a; EHCISysBusState ehci[AW_A10_NUM_USB]; OHCISysBusState ohci[AW_A10_NUM_USB]; diff --git a/include/hw/arm/allwinner-h3.h b/include/hw/arm/allwinner-h3.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-h3.h +++ b/include/hw/arm/allwinner-h3.h @@ -XXX,XX +XXX,XX @@ #include "hw/misc/allwinner-cpucfg.h" #include "hw/misc/allwinner-h3-sysctrl.h" #include "hw/misc/allwinner-sid.h" +#include "hw/sd/allwinner-sdhost.h" #include "target/arm/cpu.h" /** @@ -XXX,XX +XXX,XX @@ enum { AW_H3_SRAM_A2, AW_H3_SRAM_C, AW_H3_SYSCTRL, + AW_H3_MMC0, AW_H3_SID, AW_H3_EHCI0, AW_H3_OHCI0, @@ -XXX,XX +XXX,XX @@ typedef struct AwH3State { AwCpuCfgState cpucfg; AwH3SysCtrlState sysctrl; AwSidState sid; + AwSdHostState mmc0; GICState gic; MemoryRegion sram_a1; MemoryRegion sram_a2; diff --git a/include/hw/sd/allwinner-sdhost.h b/include/hw/sd/allwinner-sdhost.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/sd/allwinner-sdhost.h @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner (sun4i and above) SD Host Controller emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef HW_SD_ALLWINNER_SDHOST_H +#define HW_SD_ALLWINNER_SDHOST_H + +#include "qom/object.h" +#include "hw/sysbus.h" +#include "hw/sd/sd.h" + +/** + * Object model types + * @{ + */ + +/** Generic Allwinner SD Host Controller (abstract) */ +#define TYPE_AW_SDHOST "allwinner-sdhost" + +/** Allwinner sun4i family (A10, A12) */ +#define TYPE_AW_SDHOST_SUN4I TYPE_AW_SDHOST "-sun4i" + +/** Allwinner sun5i family and newer (A13, H2+, H3, etc) */ +#define TYPE_AW_SDHOST_SUN5I TYPE_AW_SDHOST "-sun5i" + +/** @} */ + +/** + * Object model macros + * @{ + */ + +#define AW_SDHOST(obj) \ + OBJECT_CHECK(AwSdHostState, (obj), TYPE_AW_SDHOST) +#define AW_SDHOST_CLASS(klass) \ + OBJECT_CLASS_CHECK(AwSdHostClass, (klass), TYPE_AW_SDHOST) +#define AW_SDHOST_GET_CLASS(obj) \ + OBJECT_GET_CLASS(AwSdHostClass, (obj), TYPE_AW_SDHOST) + +/** @} */ + +/** + * Allwinner SD Host Controller object instance state. + */ +typedef struct AwSdHostState { + /*< private >*/ + SysBusDevice busdev; + /*< public >*/ + + /** Secure Digital (SD) bus, which connects to SD card (if present) */ + SDBus sdbus; + + /** Maps I/O registers in physical memory */ + MemoryRegion iomem; + + /** Interrupt output signal to notify CPU */ + qemu_irq irq; + + /** Number of bytes left in current DMA transfer */ + uint32_t transfer_cnt; + + /** + * @name Hardware Registers + * @{ + */ + + uint32_t global_ctl; /**< Global Control */ + uint32_t clock_ctl; /**< Clock Control */ + uint32_t timeout; /**< Timeout */ + uint32_t bus_width; /**< Bus Width */ + uint32_t block_size; /**< Block Size */ + uint32_t byte_count; /**< Byte Count */ + + uint32_t command; /**< Command */ + uint32_t command_arg; /**< Command Argument */ + uint32_t response[4]; /**< Command Response */ + + uint32_t irq_mask; /**< Interrupt Mask */ + uint32_t irq_status; /**< Raw Interrupt Status */ + uint32_t status; /**< Status */ + + uint32_t fifo_wlevel; /**< FIFO Water Level */ + uint32_t fifo_func_sel; /**< FIFO Function Select */ + uint32_t debug_enable; /**< Debug Enable */ + uint32_t auto12_arg; /**< Auto Command 12 Argument */ + uint32_t newtiming_set; /**< SD New Timing Set */ + uint32_t newtiming_debug; /**< SD New Timing Debug */ + uint32_t hardware_rst; /**< Hardware Reset */ + uint32_t dmac; /**< Internal DMA Controller Control */ + uint32_t desc_base; /**< Descriptor List Base Address */ + uint32_t dmac_status; /**< Internal DMA Controller Status */ + uint32_t dmac_irq; /**< Internal DMA Controller IRQ Enable */ + uint32_t card_threshold; /**< Card Threshold Control */ + uint32_t startbit_detect; /**< eMMC DDR Start Bit Detection Control */ + uint32_t response_crc; /**< Response CRC */ + uint32_t data_crc[8]; /**< Data CRC */ + uint32_t status_crc; /**< Status CRC */ + + /** @} */ + +} AwSdHostState; + +/** + * Allwinner SD Host Controller class-level struct. + * + * This struct is filled by each sunxi device specific code + * such that the generic code can use this struct to support + * all devices. + */ +typedef struct AwSdHostClass { + /*< private >*/ + SysBusDeviceClass parent_class; + /*< public >*/ + + /** Maximum buffer size in bytes per DMA descriptor */ + size_t max_desc_size; + +} AwSdHostClass; + +#endif /* HW_SD_ALLWINNER_SDHOST_H */ diff --git a/hw/arm/allwinner-a10.c b/hw/arm/allwinner-a10.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/allwinner-a10.c +++ b/hw/arm/allwinner-a10.c @@ -XXX,XX +XXX,XX @@ #include "hw/boards.h" #include "hw/usb/hcd-ohci.h" +#define AW_A10_MMC0_BASE 0x01c0f000 #define AW_A10_PIC_REG_BASE 0x01c20400 #define AW_A10_PIT_REG_BASE 0x01c20c00 #define AW_A10_UART0_REG_BASE 0x01c28000 @@ -XXX,XX +XXX,XX @@ static void aw_a10_init(Object *obj) sizeof(s->ohci[i]), TYPE_SYSBUS_OHCI); } } + + sysbus_init_child_obj(obj, "mmc0", &s->mmc0, sizeof(s->mmc0), + TYPE_AW_SDHOST_SUN4I); } static void aw_a10_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void aw_a10_realize(DeviceState *dev, Error **errp) qdev_get_gpio_in(dev, 64 + i)); } } + + /* SD/MMC */ + qdev_init_nofail(DEVICE(&s->mmc0)); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->mmc0), 0, AW_A10_MMC0_BASE); + sysbus_connect_irq(SYS_BUS_DEVICE(&s->mmc0), 0, qdev_get_gpio_in(dev, 32)); + object_property_add_alias(OBJECT(s), "sd-bus", OBJECT(&s->mmc0), + "sd-bus", &error_abort); } static void aw_a10_class_init(ObjectClass *oc, void *data) diff --git a/hw/arm/allwinner-h3.c b/hw/arm/allwinner-h3.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/allwinner-h3.c +++ b/hw/arm/allwinner-h3.c @@ -XXX,XX +XXX,XX @@ const hwaddr allwinner_h3_memmap[] = { [AW_H3_SRAM_A2] = 0x00044000, [AW_H3_SRAM_C] = 0x00010000, [AW_H3_SYSCTRL] = 0x01c00000, + [AW_H3_MMC0] = 0x01c0f000, [AW_H3_SID] = 0x01c14000, [AW_H3_EHCI0] = 0x01c1a000, [AW_H3_OHCI0] = 0x01c1a400, @@ -XXX,XX +XXX,XX @@ struct AwH3Unimplemented { { "lcd0", 0x01c0c000, 4 * KiB }, { "lcd1", 0x01c0d000, 4 * KiB }, { "ve", 0x01c0e000, 4 * KiB }, - { "mmc0", 0x01c0f000, 4 * KiB }, { "mmc1", 0x01c10000, 4 * KiB }, { "mmc2", 0x01c11000, 4 * KiB }, { "crypto", 0x01c15000, 4 * KiB }, @@ -XXX,XX +XXX,XX @@ enum { AW_H3_GIC_SPI_UART3 = 3, AW_H3_GIC_SPI_TIMER0 = 18, AW_H3_GIC_SPI_TIMER1 = 19, + AW_H3_GIC_SPI_MMC0 = 60, AW_H3_GIC_SPI_EHCI0 = 72, AW_H3_GIC_SPI_OHCI0 = 73, AW_H3_GIC_SPI_EHCI1 = 74, @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_init(Object *obj) TYPE_AW_SID); object_property_add_alias(obj, "identifier", OBJECT(&s->sid), "identifier", &error_abort); + + sysbus_init_child_obj(obj, "mmc0", &s->mmc0, sizeof(s->mmc0), + TYPE_AW_SDHOST_SUN5I); } static void allwinner_h3_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_realize(DeviceState *dev, Error **errp) qdev_init_nofail(DEVICE(&s->sid)); sysbus_mmio_map(SYS_BUS_DEVICE(&s->sid), 0, s->memmap[AW_H3_SID]); + /* SD/MMC */ + qdev_init_nofail(DEVICE(&s->mmc0)); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->mmc0), 0, s->memmap[AW_H3_MMC0]); + sysbus_connect_irq(SYS_BUS_DEVICE(&s->mmc0), 0, + qdev_get_gpio_in(DEVICE(&s->gic), AW_H3_GIC_SPI_MMC0)); + + object_property_add_alias(OBJECT(s), "sd-bus", OBJECT(&s->mmc0), + "sd-bus", &error_abort); + /* Universal Serial Bus */ sysbus_create_simple(TYPE_AW_H3_EHCI, s->memmap[AW_H3_EHCI0], qdev_get_gpio_in(DEVICE(&s->gic), diff --git a/hw/arm/cubieboard.c b/hw/arm/cubieboard.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/cubieboard.c +++ b/hw/arm/cubieboard.c @@ -XXX,XX +XXX,XX @@ #include "sysemu/sysemu.h" #include "hw/sysbus.h" #include "hw/boards.h" +#include "hw/qdev-properties.h" #include "hw/arm/allwinner-a10.h" static struct arm_boot_info cubieboard_binfo = { @@ -XXX,XX +XXX,XX @@ static void cubieboard_init(MachineState *machine) { AwA10State *a10; Error *err = NULL; + DriveInfo *di; + BlockBackend *blk; + BusState *bus; + DeviceState *carddev; /* BIOS is not supported by this board */ if (bios_name) { @@ -XXX,XX +XXX,XX @@ static void cubieboard_init(MachineState *machine) exit(1); } + /* Retrieve SD bus */ + di = drive_get_next(IF_SD); + blk = di ? blk_by_legacy_dinfo(di) : NULL; + bus = qdev_get_child_bus(DEVICE(a10), "sd-bus"); + + /* Plug in SD card */ + carddev = qdev_create(bus, TYPE_SD_CARD); + qdev_prop_set_drive(carddev, "drive", blk, &error_fatal); + object_property_set_bool(OBJECT(carddev), true, "realized", &error_fatal); + memory_region_add_subregion(get_system_memory(), AW_A10_SDRAM_BASE, machine->ram); diff --git a/hw/arm/orangepi.c b/hw/arm/orangepi.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/orangepi.c +++ b/hw/arm/orangepi.c @@ -XXX,XX +XXX,XX @@ static struct arm_boot_info orangepi_binfo = { static void orangepi_init(MachineState *machine) { AwH3State *h3; + DriveInfo *di; + BlockBackend *blk; + BusState *bus; + DeviceState *carddev; /* BIOS is not supported by this board */ if (bios_name) { @@ -XXX,XX +XXX,XX @@ static void orangepi_init(MachineState *machine) /* Mark H3 object realized */ object_property_set_bool(OBJECT(h3), true, "realized", &error_abort); + /* Retrieve SD bus */ + di = drive_get_next(IF_SD); + blk = di ? blk_by_legacy_dinfo(di) : NULL; + bus = qdev_get_child_bus(DEVICE(h3), "sd-bus"); + + /* Plug in SD card */ + carddev = qdev_create(bus, TYPE_SD_CARD); + qdev_prop_set_drive(carddev, "drive", blk, &error_fatal); + object_property_set_bool(OBJECT(carddev), true, "realized", &error_fatal); + /* SDRAM */ memory_region_add_subregion(get_system_memory(), h3->memmap[AW_H3_SDRAM], machine->ram); @@ -XXX,XX +XXX,XX @@ static void orangepi_machine_init(MachineClass *mc) { mc->desc = "Orange Pi PC"; mc->init = orangepi_init; + mc->block_default_type = IF_SD; + mc->units_per_default_bus = 1; mc->min_cpus = AW_H3_NUM_CPUS; mc->max_cpus = AW_H3_NUM_CPUS; mc->default_cpus = AW_H3_NUM_CPUS; diff --git a/hw/sd/allwinner-sdhost.c b/hw/sd/allwinner-sdhost.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/sd/allwinner-sdhost.c @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner (sun4i and above) SD Host Controller emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu/log.h" +#include "qemu/module.h" +#include "qemu/units.h" +#include "sysemu/blockdev.h" +#include "hw/irq.h" +#include "hw/sd/allwinner-sdhost.h" +#include "migration/vmstate.h" +#include "trace.h" + +#define TYPE_AW_SDHOST_BUS "allwinner-sdhost-bus" +#define AW_SDHOST_BUS(obj) \ + OBJECT_CHECK(SDBus, (obj), TYPE_AW_SDHOST_BUS) + +/* SD Host register offsets */ +enum { + REG_SD_GCTL = 0x00, /* Global Control */ + REG_SD_CKCR = 0x04, /* Clock Control */ + REG_SD_TMOR = 0x08, /* Timeout */ + REG_SD_BWDR = 0x0C, /* Bus Width */ + REG_SD_BKSR = 0x10, /* Block Size */ + REG_SD_BYCR = 0x14, /* Byte Count */ + REG_SD_CMDR = 0x18, /* Command */ + REG_SD_CAGR = 0x1C, /* Command Argument */ + REG_SD_RESP0 = 0x20, /* Response Zero */ + REG_SD_RESP1 = 0x24, /* Response One */ + REG_SD_RESP2 = 0x28, /* Response Two */ + REG_SD_RESP3 = 0x2C, /* Response Three */ + REG_SD_IMKR = 0x30, /* Interrupt Mask */ + REG_SD_MISR = 0x34, /* Masked Interrupt Status */ + REG_SD_RISR = 0x38, /* Raw Interrupt Status */ + REG_SD_STAR = 0x3C, /* Status */ + REG_SD_FWLR = 0x40, /* FIFO Water Level */ + REG_SD_FUNS = 0x44, /* FIFO Function Select */ + REG_SD_DBGC = 0x50, /* Debug Enable */ + REG_SD_A12A = 0x58, /* Auto command 12 argument */ + REG_SD_NTSR = 0x5C, /* SD NewTiming Set */ + REG_SD_SDBG = 0x60, /* SD newTiming Set Debug */ + REG_SD_HWRST = 0x78, /* Hardware Reset Register */ + REG_SD_DMAC = 0x80, /* Internal DMA Controller Control */ + REG_SD_DLBA = 0x84, /* Descriptor List Base Address */ + REG_SD_IDST = 0x88, /* Internal DMA Controller Status */ + REG_SD_IDIE = 0x8C, /* Internal DMA Controller IRQ Enable */ + REG_SD_THLDC = 0x100, /* Card Threshold Control */ + REG_SD_DSBD = 0x10C, /* eMMC DDR Start Bit Detection Control */ + REG_SD_RES_CRC = 0x110, /* Response CRC from card/eMMC */ + REG_SD_DATA7_CRC = 0x114, /* CRC Data 7 from card/eMMC */ + REG_SD_DATA6_CRC = 0x118, /* CRC Data 6 from card/eMMC */ + REG_SD_DATA5_CRC = 0x11C, /* CRC Data 5 from card/eMMC */ + REG_SD_DATA4_CRC = 0x120, /* CRC Data 4 from card/eMMC */ + REG_SD_DATA3_CRC = 0x124, /* CRC Data 3 from card/eMMC */ + REG_SD_DATA2_CRC = 0x128, /* CRC Data 2 from card/eMMC */ + REG_SD_DATA1_CRC = 0x12C, /* CRC Data 1 from card/eMMC */ + REG_SD_DATA0_CRC = 0x130, /* CRC Data 0 from card/eMMC */ + REG_SD_CRC_STA = 0x134, /* CRC status from card/eMMC during write */ + REG_SD_FIFO = 0x200, /* Read/Write FIFO */ +}; + +/* SD Host register flags */ +enum { + SD_GCTL_FIFO_AC_MOD = (1 << 31), + SD_GCTL_DDR_MOD_SEL = (1 << 10), + SD_GCTL_CD_DBC_ENB = (1 << 8), + SD_GCTL_DMA_ENB = (1 << 5), + SD_GCTL_INT_ENB = (1 << 4), + SD_GCTL_DMA_RST = (1 << 2), + SD_GCTL_FIFO_RST = (1 << 1), + SD_GCTL_SOFT_RST = (1 << 0), +}; + +enum { + SD_CMDR_LOAD = (1 << 31), + SD_CMDR_CLKCHANGE = (1 << 21), + SD_CMDR_WRITE = (1 << 10), + SD_CMDR_AUTOSTOP = (1 << 12), + SD_CMDR_DATA = (1 << 9), + SD_CMDR_RESPONSE_LONG = (1 << 7), + SD_CMDR_RESPONSE = (1 << 6), + SD_CMDR_CMDID_MASK = (0x3f), +}; + +enum { + SD_RISR_CARD_REMOVE = (1 << 31), + SD_RISR_CARD_INSERT = (1 << 30), + SD_RISR_SDIO_INTR = (1 << 16), + SD_RISR_AUTOCMD_DONE = (1 << 14), + SD_RISR_DATA_COMPLETE = (1 << 3), + SD_RISR_CMD_COMPLETE = (1 << 2), + SD_RISR_NO_RESPONSE = (1 << 1), +}; + +enum { + SD_STAR_CARD_PRESENT = (1 << 8), +}; + +enum { + SD_IDST_INT_SUMMARY = (1 << 8), + SD_IDST_RECEIVE_IRQ = (1 << 1), + SD_IDST_TRANSMIT_IRQ = (1 << 0), + SD_IDST_IRQ_MASK = (1 << 1) | (1 << 0) | (1 << 8), + SD_IDST_WR_MASK = (0x3ff), +}; + +/* SD Host register reset values */ +enum { + REG_SD_GCTL_RST = 0x00000300, + REG_SD_CKCR_RST = 0x0, + REG_SD_TMOR_RST = 0xFFFFFF40, + REG_SD_BWDR_RST = 0x0, + REG_SD_BKSR_RST = 0x00000200, + REG_SD_BYCR_RST = 0x00000200, + REG_SD_CMDR_RST = 0x0, + REG_SD_CAGR_RST = 0x0, + REG_SD_RESP_RST = 0x0, + REG_SD_IMKR_RST = 0x0, + REG_SD_MISR_RST = 0x0, + REG_SD_RISR_RST = 0x0, + REG_SD_STAR_RST = 0x00000100, + REG_SD_FWLR_RST = 0x000F0000, + REG_SD_FUNS_RST = 0x0, + REG_SD_DBGC_RST = 0x0, + REG_SD_A12A_RST = 0x0000FFFF, + REG_SD_NTSR_RST = 0x00000001, + REG_SD_SDBG_RST = 0x0, + REG_SD_HWRST_RST = 0x00000001, + REG_SD_DMAC_RST = 0x0, + REG_SD_DLBA_RST = 0x0, + REG_SD_IDST_RST = 0x0, + REG_SD_IDIE_RST = 0x0, + REG_SD_THLDC_RST = 0x0, + REG_SD_DSBD_RST = 0x0, + REG_SD_RES_CRC_RST = 0x0, + REG_SD_DATA_CRC_RST = 0x0, + REG_SD_CRC_STA_RST = 0x0, + REG_SD_FIFO_RST = 0x0, +}; + +/* Data transfer descriptor for DMA */ +typedef struct TransferDescriptor { + uint32_t status; /* Status flags */ + uint32_t size; /* Data buffer size */ + uint32_t addr; /* Data buffer address */ + uint32_t next; /* Physical address of next descriptor */ +} TransferDescriptor; + +/* Data transfer descriptor flags */ +enum { + DESC_STATUS_HOLD = (1 << 31), /* Set when descriptor is in use by DMA */ + DESC_STATUS_ERROR = (1 << 30), /* Set when DMA transfer error occurred */ + DESC_STATUS_CHAIN = (1 << 4), /* Indicates chained descriptor. */ + DESC_STATUS_FIRST = (1 << 3), /* Set on the first descriptor */ + DESC_STATUS_LAST = (1 << 2), /* Set on the last descriptor */ + DESC_STATUS_NOIRQ = (1 << 1), /* Skip raising interrupt after transfer */ + DESC_SIZE_MASK = (0xfffffffc) +}; + +static void allwinner_sdhost_update_irq(AwSdHostState *s) +{ + uint32_t irq; + + if (s->global_ctl & SD_GCTL_INT_ENB) { + irq = s->irq_status & s->irq_mask; + } else { + irq = 0; + } + + trace_allwinner_sdhost_update_irq(irq); + qemu_set_irq(s->irq, irq); +} + +static void allwinner_sdhost_update_transfer_cnt(AwSdHostState *s, + uint32_t bytes) +{ + if (s->transfer_cnt > bytes) { + s->transfer_cnt -= bytes; + } else { + s->transfer_cnt = 0; + } + + if (!s->transfer_cnt) { + s->irq_status |= SD_RISR_DATA_COMPLETE; + } +} + +static void allwinner_sdhost_set_inserted(DeviceState *dev, bool inserted) +{ + AwSdHostState *s = AW_SDHOST(dev); + + trace_allwinner_sdhost_set_inserted(inserted); + + if (inserted) { + s->irq_status |= SD_RISR_CARD_INSERT; + s->irq_status &= ~SD_RISR_CARD_REMOVE; + s->status |= SD_STAR_CARD_PRESENT; + } else { + s->irq_status &= ~SD_RISR_CARD_INSERT; + s->irq_status |= SD_RISR_CARD_REMOVE; + s->status &= ~SD_STAR_CARD_PRESENT; + } + + allwinner_sdhost_update_irq(s); +} + +static void allwinner_sdhost_send_command(AwSdHostState *s) +{ + SDRequest request; + uint8_t resp[16]; + int rlen; + + /* Auto clear load flag */ + s->command &= ~SD_CMDR_LOAD; + + /* Clock change does not actually interact with the SD bus */ + if (!(s->command & SD_CMDR_CLKCHANGE)) { + + /* Prepare request */ + request.cmd = s->command & SD_CMDR_CMDID_MASK; + request.arg = s->command_arg; + + /* Send request to SD bus */ + rlen = sdbus_do_command(&s->sdbus, &request, resp); + if (rlen < 0) { + goto error; + } + + /* If the command has a response, store it in the response registers */ + if ((s->command & SD_CMDR_RESPONSE)) { + if (rlen == 4 && !(s->command & SD_CMDR_RESPONSE_LONG)) { + s->response[0] = ldl_be_p(&resp[0]); + s->response[1] = s->response[2] = s->response[3] = 0; + + } else if (rlen == 16 && (s->command & SD_CMDR_RESPONSE_LONG)) { + s->response[0] = ldl_be_p(&resp[12]); + s->response[1] = ldl_be_p(&resp[8]); + s->response[2] = ldl_be_p(&resp[4]); + s->response[3] = ldl_be_p(&resp[0]); + } else { + goto error; + } + } + } + + /* Set interrupt status bits */ + s->irq_status |= SD_RISR_CMD_COMPLETE; + return; + +error: + s->irq_status |= SD_RISR_NO_RESPONSE; +} + +static void allwinner_sdhost_auto_stop(AwSdHostState *s) +{ + /* + * The stop command (CMD12) ensures the SD bus + * returns to the transfer state. + */ + if ((s->command & SD_CMDR_AUTOSTOP) && (s->transfer_cnt == 0)) { + /* First save current command registers */ + uint32_t saved_cmd = s->command; + uint32_t saved_arg = s->command_arg; + + /* Prepare stop command (CMD12) */ + s->command &= ~SD_CMDR_CMDID_MASK; + s->command |= 12; /* CMD12 */ + s->command_arg = 0; + + /* Put the command on SD bus */ + allwinner_sdhost_send_command(s); + + /* Restore command values */ + s->command = saved_cmd; + s->command_arg = saved_arg; + + /* Set IRQ status bit for automatic stop done */ + s->irq_status |= SD_RISR_AUTOCMD_DONE; + } +} + +static uint32_t allwinner_sdhost_process_desc(AwSdHostState *s, + hwaddr desc_addr, + TransferDescriptor *desc, + bool is_write, uint32_t max_bytes) +{ + AwSdHostClass *klass = AW_SDHOST_GET_CLASS(s); + uint32_t num_done = 0; + uint32_t num_bytes = max_bytes; + uint8_t buf[1024]; + + /* Read descriptor */ + cpu_physical_memory_read(desc_addr, desc, sizeof(*desc)); + if (desc->size == 0) { + desc->size = klass->max_desc_size; + } else if (desc->size > klass->max_desc_size) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: DMA descriptor buffer size " + " is out-of-bounds: %" PRIu32 " > %zu", + __func__, desc->size, klass->max_desc_size); + desc->size = klass->max_desc_size; + } + if (desc->size < num_bytes) { + num_bytes = desc->size; + } + + trace_allwinner_sdhost_process_desc(desc_addr, desc->size, + is_write, max_bytes); + + while (num_done < num_bytes) { + /* Try to completely fill the local buffer */ + uint32_t buf_bytes = num_bytes - num_done; + if (buf_bytes > sizeof(buf)) { + buf_bytes = sizeof(buf); + } + + /* Write to SD bus */ + if (is_write) { + cpu_physical_memory_read((desc->addr & DESC_SIZE_MASK) + num_done, + buf, buf_bytes); + + for (uint32_t i = 0; i < buf_bytes; i++) { + sdbus_write_data(&s->sdbus, buf[i]); + } + + /* Read from SD bus */ + } else { + for (uint32_t i = 0; i < buf_bytes; i++) { + buf[i] = sdbus_read_data(&s->sdbus); + } + cpu_physical_memory_write((desc->addr & DESC_SIZE_MASK) + num_done, + buf, buf_bytes); + } + num_done += buf_bytes; + } + + /* Clear hold flag and flush descriptor */ + desc->status &= ~DESC_STATUS_HOLD; + cpu_physical_memory_write(desc_addr, desc, sizeof(*desc)); + + return num_done; +} + +static void allwinner_sdhost_dma(AwSdHostState *s) +{ + TransferDescriptor desc; + hwaddr desc_addr = s->desc_base; + bool is_write = (s->command & SD_CMDR_WRITE); + uint32_t bytes_done = 0; + + /* Check if DMA can be performed */ + if (s->byte_count == 0 || s->block_size == 0 || + !(s->global_ctl & SD_GCTL_DMA_ENB)) { + return; + } + + /* + * For read operations, data must be available on the SD bus + * If not, it is an error and we should not act at all + */ + if (!is_write && !sdbus_data_ready(&s->sdbus)) { + return; + } + + /* Process the DMA descriptors until all data is copied */ + while (s->byte_count > 0) { + bytes_done = allwinner_sdhost_process_desc(s, desc_addr, &desc, + is_write, s->byte_count); + allwinner_sdhost_update_transfer_cnt(s, bytes_done); + + if (bytes_done <= s->byte_count) { + s->byte_count -= bytes_done; + } else { + s->byte_count = 0; + } + + if (desc.status & DESC_STATUS_LAST) { + break; + } else { + desc_addr = desc.next; + } + } + + /* Raise IRQ to signal DMA is completed */ + s->irq_status |= SD_RISR_DATA_COMPLETE | SD_RISR_SDIO_INTR; + + /* Update DMAC bits */ + s->dmac_status |= SD_IDST_INT_SUMMARY; + + if (is_write) { + s->dmac_status |= SD_IDST_TRANSMIT_IRQ; + } else { + s->dmac_status |= SD_IDST_RECEIVE_IRQ; + } +} + +static uint64_t allwinner_sdhost_read(void *opaque, hwaddr offset, + unsigned size) +{ + AwSdHostState *s = AW_SDHOST(opaque); + uint32_t res = 0; + + switch (offset) { + case REG_SD_GCTL: /* Global Control */ + res = s->global_ctl; + break; + case REG_SD_CKCR: /* Clock Control */ + res = s->clock_ctl; + break; + case REG_SD_TMOR: /* Timeout */ + res = s->timeout; + break; + case REG_SD_BWDR: /* Bus Width */ + res = s->bus_width; + break; + case REG_SD_BKSR: /* Block Size */ + res = s->block_size; + break; + case REG_SD_BYCR: /* Byte Count */ + res = s->byte_count; + break; + case REG_SD_CMDR: /* Command */ + res = s->command; + break; + case REG_SD_CAGR: /* Command Argument */ + res = s->command_arg; + break; + case REG_SD_RESP0: /* Response Zero */ + res = s->response[0]; + break; + case REG_SD_RESP1: /* Response One */ + res = s->response[1]; + break; + case REG_SD_RESP2: /* Response Two */ + res = s->response[2]; + break; + case REG_SD_RESP3: /* Response Three */ + res = s->response[3]; + break; + case REG_SD_IMKR: /* Interrupt Mask */ + res = s->irq_mask; + break; + case REG_SD_MISR: /* Masked Interrupt Status */ + res = s->irq_status & s->irq_mask; + break; + case REG_SD_RISR: /* Raw Interrupt Status */ + res = s->irq_status; + break; + case REG_SD_STAR: /* Status */ + res = s->status; + break; + case REG_SD_FWLR: /* FIFO Water Level */ + res = s->fifo_wlevel; + break; + case REG_SD_FUNS: /* FIFO Function Select */ + res = s->fifo_func_sel; + break; + case REG_SD_DBGC: /* Debug Enable */ + res = s->debug_enable; + break; + case REG_SD_A12A: /* Auto command 12 argument */ + res = s->auto12_arg; + break; + case REG_SD_NTSR: /* SD NewTiming Set */ + res = s->newtiming_set; + break; + case REG_SD_SDBG: /* SD newTiming Set Debug */ + res = s->newtiming_debug; + break; + case REG_SD_HWRST: /* Hardware Reset Register */ + res = s->hardware_rst; + break; + case REG_SD_DMAC: /* Internal DMA Controller Control */ + res = s->dmac; + break; + case REG_SD_DLBA: /* Descriptor List Base Address */ + res = s->desc_base; + break; + case REG_SD_IDST: /* Internal DMA Controller Status */ + res = s->dmac_status; + break; + case REG_SD_IDIE: /* Internal DMA Controller Interrupt Enable */ + res = s->dmac_irq; + break; + case REG_SD_THLDC: /* Card Threshold Control */ + res = s->card_threshold; + break; + case REG_SD_DSBD: /* eMMC DDR Start Bit Detection Control */ + res = s->startbit_detect; + break; + case REG_SD_RES_CRC: /* Response CRC from card/eMMC */ + res = s->response_crc; + break; + case REG_SD_DATA7_CRC: /* CRC Data 7 from card/eMMC */ + case REG_SD_DATA6_CRC: /* CRC Data 6 from card/eMMC */ + case REG_SD_DATA5_CRC: /* CRC Data 5 from card/eMMC */ + case REG_SD_DATA4_CRC: /* CRC Data 4 from card/eMMC */ + case REG_SD_DATA3_CRC: /* CRC Data 3 from card/eMMC */ + case REG_SD_DATA2_CRC: /* CRC Data 2 from card/eMMC */ + case REG_SD_DATA1_CRC: /* CRC Data 1 from card/eMMC */ + case REG_SD_DATA0_CRC: /* CRC Data 0 from card/eMMC */ + res = s->data_crc[((offset - REG_SD_DATA7_CRC) / sizeof(uint32_t))]; + break; + case REG_SD_CRC_STA: /* CRC status from card/eMMC in write operation */ + res = s->status_crc; + break; + case REG_SD_FIFO: /* Read/Write FIFO */ + if (sdbus_data_ready(&s->sdbus)) { + res = sdbus_read_data(&s->sdbus); + res |= sdbus_read_data(&s->sdbus) << 8; + res |= sdbus_read_data(&s->sdbus) << 16; + res |= sdbus_read_data(&s->sdbus) << 24; + allwinner_sdhost_update_transfer_cnt(s, sizeof(uint32_t)); + allwinner_sdhost_auto_stop(s); + allwinner_sdhost_update_irq(s); + } else { + qemu_log_mask(LOG_GUEST_ERROR, "%s: no data ready on SD bus\n", + __func__); + } + break; + default: + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset %" + HWADDR_PRIx"\n", __func__, offset); + res = 0; + break; + } + + trace_allwinner_sdhost_read(offset, res, size); + return res; +} + +static void allwinner_sdhost_write(void *opaque, hwaddr offset, + uint64_t value, unsigned size) +{ + AwSdHostState *s = AW_SDHOST(opaque); + + trace_allwinner_sdhost_write(offset, value, size); + + switch (offset) { + case REG_SD_GCTL: /* Global Control */ + s->global_ctl = value; + s->global_ctl &= ~(SD_GCTL_DMA_RST | SD_GCTL_FIFO_RST | + SD_GCTL_SOFT_RST); + allwinner_sdhost_update_irq(s); + break; + case REG_SD_CKCR: /* Clock Control */ + s->clock_ctl = value; + break; + case REG_SD_TMOR: /* Timeout */ + s->timeout = value; + break; + case REG_SD_BWDR: /* Bus Width */ + s->bus_width = value; + break; + case REG_SD_BKSR: /* Block Size */ + s->block_size = value; + break; + case REG_SD_BYCR: /* Byte Count */ + s->byte_count = value; + s->transfer_cnt = value; + break; + case REG_SD_CMDR: /* Command */ + s->command = value; + if (value & SD_CMDR_LOAD) { + allwinner_sdhost_send_command(s); + allwinner_sdhost_dma(s); + allwinner_sdhost_auto_stop(s); + } + allwinner_sdhost_update_irq(s); + break; + case REG_SD_CAGR: /* Command Argument */ + s->command_arg = value; + break; + case REG_SD_RESP0: /* Response Zero */ + s->response[0] = value; + break; + case REG_SD_RESP1: /* Response One */ + s->response[1] = value; + break; + case REG_SD_RESP2: /* Response Two */ + s->response[2] = value; + break; + case REG_SD_RESP3: /* Response Three */ + s->response[3] = value; + break; + case REG_SD_IMKR: /* Interrupt Mask */ + s->irq_mask = value; + allwinner_sdhost_update_irq(s); + break; + case REG_SD_MISR: /* Masked Interrupt Status */ + case REG_SD_RISR: /* Raw Interrupt Status */ + s->irq_status &= ~value; + allwinner_sdhost_update_irq(s); + break; + case REG_SD_STAR: /* Status */ + s->status &= ~value; + allwinner_sdhost_update_irq(s); + break; + case REG_SD_FWLR: /* FIFO Water Level */ + s->fifo_wlevel = value; + break; + case REG_SD_FUNS: /* FIFO Function Select */ + s->fifo_func_sel = value; + break; + case REG_SD_DBGC: /* Debug Enable */ + s->debug_enable = value; + break; + case REG_SD_A12A: /* Auto command 12 argument */ + s->auto12_arg = value; + break; + case REG_SD_NTSR: /* SD NewTiming Set */ + s->newtiming_set = value; + break; + case REG_SD_SDBG: /* SD newTiming Set Debug */ + s->newtiming_debug = value; + break; + case REG_SD_HWRST: /* Hardware Reset Register */ + s->hardware_rst = value; + break; + case REG_SD_DMAC: /* Internal DMA Controller Control */ + s->dmac = value; + allwinner_sdhost_update_irq(s); + break; + case REG_SD_DLBA: /* Descriptor List Base Address */ + s->desc_base = value; + break; + case REG_SD_IDST: /* Internal DMA Controller Status */ + s->dmac_status &= (~SD_IDST_WR_MASK) | (~value & SD_IDST_WR_MASK); + allwinner_sdhost_update_irq(s); + break; + case REG_SD_IDIE: /* Internal DMA Controller Interrupt Enable */ + s->dmac_irq = value; + allwinner_sdhost_update_irq(s); + break; + case REG_SD_THLDC: /* Card Threshold Control */ + s->card_threshold = value; + break; + case REG_SD_DSBD: /* eMMC DDR Start Bit Detection Control */ + s->startbit_detect = value; + break; + case REG_SD_FIFO: /* Read/Write FIFO */ + sdbus_write_data(&s->sdbus, value & 0xff); + sdbus_write_data(&s->sdbus, (value >> 8) & 0xff); + sdbus_write_data(&s->sdbus, (value >> 16) & 0xff); + sdbus_write_data(&s->sdbus, (value >> 24) & 0xff); + allwinner_sdhost_update_transfer_cnt(s, sizeof(uint32_t)); + allwinner_sdhost_auto_stop(s); + allwinner_sdhost_update_irq(s); + break; + case REG_SD_RES_CRC: /* Response CRC from card/eMMC */ + case REG_SD_DATA7_CRC: /* CRC Data 7 from card/eMMC */ + case REG_SD_DATA6_CRC: /* CRC Data 6 from card/eMMC */ + case REG_SD_DATA5_CRC: /* CRC Data 5 from card/eMMC */ + case REG_SD_DATA4_CRC: /* CRC Data 4 from card/eMMC */ + case REG_SD_DATA3_CRC: /* CRC Data 3 from card/eMMC */ + case REG_SD_DATA2_CRC: /* CRC Data 2 from card/eMMC */ + case REG_SD_DATA1_CRC: /* CRC Data 1 from card/eMMC */ + case REG_SD_DATA0_CRC: /* CRC Data 0 from card/eMMC */ + case REG_SD_CRC_STA: /* CRC status from card/eMMC in write operation */ + break; + default: + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset %" + HWADDR_PRIx"\n", __func__, offset); + break; + } +} + +static const MemoryRegionOps allwinner_sdhost_ops = { + .read = allwinner_sdhost_read, + .write = allwinner_sdhost_write, + .endianness = DEVICE_NATIVE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, + .impl.min_access_size = 4, +}; + +static const VMStateDescription vmstate_allwinner_sdhost = { + .name = "allwinner-sdhost", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32(global_ctl, AwSdHostState), + VMSTATE_UINT32(clock_ctl, AwSdHostState), + VMSTATE_UINT32(timeout, AwSdHostState), + VMSTATE_UINT32(bus_width, AwSdHostState), + VMSTATE_UINT32(block_size, AwSdHostState), + VMSTATE_UINT32(byte_count, AwSdHostState), + VMSTATE_UINT32(transfer_cnt, AwSdHostState), + VMSTATE_UINT32(command, AwSdHostState), + VMSTATE_UINT32(command_arg, AwSdHostState), + VMSTATE_UINT32_ARRAY(response, AwSdHostState, 4), + VMSTATE_UINT32(irq_mask, AwSdHostState), + VMSTATE_UINT32(irq_status, AwSdHostState), + VMSTATE_UINT32(status, AwSdHostState), + VMSTATE_UINT32(fifo_wlevel, AwSdHostState), + VMSTATE_UINT32(fifo_func_sel, AwSdHostState), + VMSTATE_UINT32(debug_enable, AwSdHostState), + VMSTATE_UINT32(auto12_arg, AwSdHostState), + VMSTATE_UINT32(newtiming_set, AwSdHostState), + VMSTATE_UINT32(newtiming_debug, AwSdHostState), + VMSTATE_UINT32(hardware_rst, AwSdHostState), + VMSTATE_UINT32(dmac, AwSdHostState), + VMSTATE_UINT32(desc_base, AwSdHostState), + VMSTATE_UINT32(dmac_status, AwSdHostState), + VMSTATE_UINT32(dmac_irq, AwSdHostState), + VMSTATE_UINT32(card_threshold, AwSdHostState), + VMSTATE_UINT32(startbit_detect, AwSdHostState), + VMSTATE_UINT32(response_crc, AwSdHostState), + VMSTATE_UINT32_ARRAY(data_crc, AwSdHostState, 8), + VMSTATE_UINT32(status_crc, AwSdHostState), + VMSTATE_END_OF_LIST() + } +}; + +static void allwinner_sdhost_init(Object *obj) +{ + AwSdHostState *s = AW_SDHOST(obj); + + qbus_create_inplace(&s->sdbus, sizeof(s->sdbus), + TYPE_AW_SDHOST_BUS, DEVICE(s), "sd-bus"); + + memory_region_init_io(&s->iomem, obj, &allwinner_sdhost_ops, s, + TYPE_AW_SDHOST, 4 * KiB); + sysbus_init_mmio(SYS_BUS_DEVICE(s), &s->iomem); + sysbus_init_irq(SYS_BUS_DEVICE(s), &s->irq); +} + +static void allwinner_sdhost_reset(DeviceState *dev) +{ + AwSdHostState *s = AW_SDHOST(dev); + + s->global_ctl = REG_SD_GCTL_RST; + s->clock_ctl = REG_SD_CKCR_RST; + s->timeout = REG_SD_TMOR_RST; + s->bus_width = REG_SD_BWDR_RST; + s->block_size = REG_SD_BKSR_RST; + s->byte_count = REG_SD_BYCR_RST; + s->transfer_cnt = 0; + + s->command = REG_SD_CMDR_RST; + s->command_arg = REG_SD_CAGR_RST; + + for (int i = 0; i < ARRAY_SIZE(s->response); i++) { + s->response[i] = REG_SD_RESP_RST; + } + + s->irq_mask = REG_SD_IMKR_RST; + s->irq_status = REG_SD_RISR_RST; + s->status = REG_SD_STAR_RST; + + s->fifo_wlevel = REG_SD_FWLR_RST; + s->fifo_func_sel = REG_SD_FUNS_RST; + s->debug_enable = REG_SD_DBGC_RST; + s->auto12_arg = REG_SD_A12A_RST; + s->newtiming_set = REG_SD_NTSR_RST; + s->newtiming_debug = REG_SD_SDBG_RST; + s->hardware_rst = REG_SD_HWRST_RST; + s->dmac = REG_SD_DMAC_RST; + s->desc_base = REG_SD_DLBA_RST; + s->dmac_status = REG_SD_IDST_RST; + s->dmac_irq = REG_SD_IDIE_RST; + s->card_threshold = REG_SD_THLDC_RST; + s->startbit_detect = REG_SD_DSBD_RST; + s->response_crc = REG_SD_RES_CRC_RST; + + for (int i = 0; i < ARRAY_SIZE(s->data_crc); i++) { + s->data_crc[i] = REG_SD_DATA_CRC_RST; + } + + s->status_crc = REG_SD_CRC_STA_RST; +} + +static void allwinner_sdhost_bus_class_init(ObjectClass *klass, void *data) +{ + SDBusClass *sbc = SD_BUS_CLASS(klass); + + sbc->set_inserted = allwinner_sdhost_set_inserted; +} + +static void allwinner_sdhost_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->reset = allwinner_sdhost_reset; + dc->vmsd = &vmstate_allwinner_sdhost; +} + +static void allwinner_sdhost_sun4i_class_init(ObjectClass *klass, void *data) +{ + AwSdHostClass *sc = AW_SDHOST_CLASS(klass); + sc->max_desc_size = 8 * KiB; +} + +static void allwinner_sdhost_sun5i_class_init(ObjectClass *klass, void *data) +{ + AwSdHostClass *sc = AW_SDHOST_CLASS(klass); + sc->max_desc_size = 64 * KiB; +} + +static TypeInfo allwinner_sdhost_info = { + .name = TYPE_AW_SDHOST, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_init = allwinner_sdhost_init, + .instance_size = sizeof(AwSdHostState), + .class_init = allwinner_sdhost_class_init, + .class_size = sizeof(AwSdHostClass), + .abstract = true, +}; + +static const TypeInfo allwinner_sdhost_sun4i_info = { + .name = TYPE_AW_SDHOST_SUN4I, + .parent = TYPE_AW_SDHOST, + .class_init = allwinner_sdhost_sun4i_class_init, +}; + +static const TypeInfo allwinner_sdhost_sun5i_info = { + .name = TYPE_AW_SDHOST_SUN5I, + .parent = TYPE_AW_SDHOST, + .class_init = allwinner_sdhost_sun5i_class_init, +}; + +static const TypeInfo allwinner_sdhost_bus_info = { + .name = TYPE_AW_SDHOST_BUS, + .parent = TYPE_SD_BUS, + .instance_size = sizeof(SDBus), + .class_init = allwinner_sdhost_bus_class_init, +}; + +static void allwinner_sdhost_register_types(void) +{ + type_register_static(&allwinner_sdhost_info); + type_register_static(&allwinner_sdhost_sun4i_info); + type_register_static(&allwinner_sdhost_sun5i_info); + type_register_static(&allwinner_sdhost_bus_info); +} + +type_init(allwinner_sdhost_register_types) diff --git a/hw/arm/Kconfig b/hw/arm/Kconfig index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/Kconfig +++ b/hw/arm/Kconfig @@ -XXX,XX +XXX,XX @@ config ALLWINNER_H3 select UNIMP select USB_OHCI select USB_EHCI_SYSBUS + select SD config RASPI bool diff --git a/hw/sd/trace-events b/hw/sd/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/sd/trace-events +++ b/hw/sd/trace-events @@ -XXX,XX +XXX,XX @@ # See docs/devel/tracing.txt for syntax documentation. +# allwinner-sdhost.c +allwinner_sdhost_set_inserted(bool inserted) "inserted %u" +allwinner_sdhost_process_desc(uint64_t desc_addr, uint32_t desc_size, bool is_write, uint32_t max_bytes) "desc_addr 0x%" PRIx64 " desc_size %" PRIu32 " is_write %u max_bytes %" PRIu32 +allwinner_sdhost_read(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 +allwinner_sdhost_write(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 +allwinner_sdhost_update_irq(uint32_t irq) "IRQ bits 0x%" PRIx32 + # bcm2835_sdhost.c bcm2835_sdhost_read(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %u" bcm2835_sdhost_write(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %u" -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> The Allwinner Sun8i System on Chip family includes an Ethernet MAC (EMAC) which provides 10M/100M/1000M Ethernet connectivity. This commit adds support for the Allwinner EMAC from the Sun8i family (H2+, H3, A33, etc), including emulation for the following functionality: * DMA transfers * MII interface * Transmit CRC calculation Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20200311221854.30370-10-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/net/Makefile.objs | 1 + include/hw/arm/allwinner-h3.h | 3 + include/hw/net/allwinner-sun8i-emac.h | 99 +++ hw/arm/allwinner-h3.c | 16 +- hw/arm/orangepi.c | 3 + hw/net/allwinner-sun8i-emac.c | 871 ++++++++++++++++++++++++++ hw/arm/Kconfig | 1 + hw/net/Kconfig | 3 + hw/net/trace-events | 10 + 9 files changed, 1006 insertions(+), 1 deletion(-) create mode 100644 include/hw/net/allwinner-sun8i-emac.h create mode 100644 hw/net/allwinner-sun8i-emac.c diff --git a/hw/net/Makefile.objs b/hw/net/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/net/Makefile.objs +++ b/hw/net/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_XGMAC) += xgmac.o common-obj-$(CONFIG_MIPSNET) += mipsnet.o common-obj-$(CONFIG_XILINX_AXI) += xilinx_axienet.o common-obj-$(CONFIG_ALLWINNER_EMAC) += allwinner_emac.o +common-obj-$(CONFIG_ALLWINNER_SUN8I_EMAC) += allwinner-sun8i-emac.o common-obj-$(CONFIG_IMX_FEC) += imx_fec.o common-obj-$(CONFIG_CADENCE) += cadence_gem.o diff --git a/include/hw/arm/allwinner-h3.h b/include/hw/arm/allwinner-h3.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-h3.h +++ b/include/hw/arm/allwinner-h3.h @@ -XXX,XX +XXX,XX @@ #include "hw/misc/allwinner-h3-sysctrl.h" #include "hw/misc/allwinner-sid.h" #include "hw/sd/allwinner-sdhost.h" +#include "hw/net/allwinner-sun8i-emac.h" #include "target/arm/cpu.h" /** @@ -XXX,XX +XXX,XX @@ enum { AW_H3_UART1, AW_H3_UART2, AW_H3_UART3, + AW_H3_EMAC, AW_H3_GIC_DIST, AW_H3_GIC_CPU, AW_H3_GIC_HYP, @@ -XXX,XX +XXX,XX @@ typedef struct AwH3State { AwH3SysCtrlState sysctrl; AwSidState sid; AwSdHostState mmc0; + AwSun8iEmacState emac; GICState gic; MemoryRegion sram_a1; MemoryRegion sram_a2; diff --git a/include/hw/net/allwinner-sun8i-emac.h b/include/hw/net/allwinner-sun8i-emac.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/net/allwinner-sun8i-emac.h @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner Sun8i Ethernet MAC emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef HW_NET_ALLWINNER_SUN8I_EMAC_H +#define HW_NET_ALLWINNER_SUN8I_EMAC_H + +#include "qom/object.h" +#include "net/net.h" +#include "hw/sysbus.h" + +/** + * Object model + * @{ + */ + +#define TYPE_AW_SUN8I_EMAC "allwinner-sun8i-emac" +#define AW_SUN8I_EMAC(obj) \ + OBJECT_CHECK(AwSun8iEmacState, (obj), TYPE_AW_SUN8I_EMAC) + +/** @} */ + +/** + * Allwinner Sun8i EMAC object instance state + */ +typedef struct AwSun8iEmacState { + /*< private >*/ + SysBusDevice parent_obj; + /*< public >*/ + + /** Maps I/O registers in physical memory */ + MemoryRegion iomem; + + /** Interrupt output signal to notify CPU */ + qemu_irq irq; + + /** Generic Network Interface Controller (NIC) for networking API */ + NICState *nic; + + /** Generic Network Interface Controller (NIC) configuration */ + NICConf conf; + + /** + * @name Media Independent Interface (MII) + * @{ + */ + + uint8_t mii_phy_addr; /**< PHY address */ + uint32_t mii_cr; /**< Control */ + uint32_t mii_st; /**< Status */ + uint32_t mii_adv; /**< Advertised Abilities */ + + /** @} */ + + /** + * @name Hardware Registers + * @{ + */ + + uint32_t basic_ctl0; /**< Basic Control 0 */ + uint32_t basic_ctl1; /**< Basic Control 1 */ + uint32_t int_en; /**< Interrupt Enable */ + uint32_t int_sta; /**< Interrupt Status */ + uint32_t frm_flt; /**< Receive Frame Filter */ + + uint32_t rx_ctl0; /**< Receive Control 0 */ + uint32_t rx_ctl1; /**< Receive Control 1 */ + uint32_t rx_desc_head; /**< Receive Descriptor List Address */ + uint32_t rx_desc_curr; /**< Current Receive Descriptor Address */ + + uint32_t tx_ctl0; /**< Transmit Control 0 */ + uint32_t tx_ctl1; /**< Transmit Control 1 */ + uint32_t tx_desc_head; /**< Transmit Descriptor List Address */ + uint32_t tx_desc_curr; /**< Current Transmit Descriptor Address */ + uint32_t tx_flowctl; /**< Transmit Flow Control */ + + uint32_t mii_cmd; /**< Management Interface Command */ + uint32_t mii_data; /**< Management Interface Data */ + + /** @} */ + +} AwSun8iEmacState; + +#endif /* HW_NET_ALLWINNER_SUN8I_H */ diff --git a/hw/arm/allwinner-h3.c b/hw/arm/allwinner-h3.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/allwinner-h3.c +++ b/hw/arm/allwinner-h3.c @@ -XXX,XX +XXX,XX @@ const hwaddr allwinner_h3_memmap[] = { [AW_H3_UART1] = 0x01c28400, [AW_H3_UART2] = 0x01c28800, [AW_H3_UART3] = 0x01c28c00, + [AW_H3_EMAC] = 0x01c30000, [AW_H3_GIC_DIST] = 0x01c81000, [AW_H3_GIC_CPU] = 0x01c82000, [AW_H3_GIC_HYP] = 0x01c84000, @@ -XXX,XX +XXX,XX @@ struct AwH3Unimplemented { { "twi1", 0x01c2b000, 1 * KiB }, { "twi2", 0x01c2b400, 1 * KiB }, { "scr", 0x01c2c400, 1 * KiB }, - { "emac", 0x01c30000, 64 * KiB }, { "gpu", 0x01c40000, 64 * KiB }, { "hstmr", 0x01c60000, 4 * KiB }, { "dramcom", 0x01c62000, 4 * KiB }, @@ -XXX,XX +XXX,XX @@ enum { AW_H3_GIC_SPI_OHCI2 = 77, AW_H3_GIC_SPI_EHCI3 = 78, AW_H3_GIC_SPI_OHCI3 = 79, + AW_H3_GIC_SPI_EMAC = 82 }; /* Allwinner H3 general constants */ @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_init(Object *obj) sysbus_init_child_obj(obj, "mmc0", &s->mmc0, sizeof(s->mmc0), TYPE_AW_SDHOST_SUN5I); + + sysbus_init_child_obj(obj, "emac", &s->emac, sizeof(s->emac), + TYPE_AW_SUN8I_EMAC); } static void allwinner_h3_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_realize(DeviceState *dev, Error **errp) object_property_add_alias(OBJECT(s), "sd-bus", OBJECT(&s->mmc0), "sd-bus", &error_abort); + /* EMAC */ + if (nd_table[0].used) { + qemu_check_nic_model(&nd_table[0], TYPE_AW_SUN8I_EMAC); + qdev_set_nic_properties(DEVICE(&s->emac), &nd_table[0]); + } + qdev_init_nofail(DEVICE(&s->emac)); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->emac), 0, s->memmap[AW_H3_EMAC]); + sysbus_connect_irq(SYS_BUS_DEVICE(&s->emac), 0, + qdev_get_gpio_in(DEVICE(&s->gic), AW_H3_GIC_SPI_EMAC)); + /* Universal Serial Bus */ sysbus_create_simple(TYPE_AW_H3_EHCI, s->memmap[AW_H3_EHCI0], qdev_get_gpio_in(DEVICE(&s->gic), diff --git a/hw/arm/orangepi.c b/hw/arm/orangepi.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/orangepi.c +++ b/hw/arm/orangepi.c @@ -XXX,XX +XXX,XX @@ static void orangepi_init(MachineState *machine) warn_report("Security Identifier value does not include H3 prefix"); } + /* Setup EMAC properties */ + object_property_set_int(OBJECT(&h3->emac), 1, "phy-addr", &error_abort); + /* Mark H3 object realized */ object_property_set_bool(OBJECT(h3), true, "realized", &error_abort); diff --git a/hw/net/allwinner-sun8i-emac.c b/hw/net/allwinner-sun8i-emac.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/net/allwinner-sun8i-emac.c @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner Sun8i Ethernet MAC emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu/units.h" +#include "hw/sysbus.h" +#include "migration/vmstate.h" +#include "net/net.h" +#include "hw/irq.h" +#include "hw/qdev-properties.h" +#include "qemu/log.h" +#include "trace.h" +#include "net/checksum.h" +#include "qemu/module.h" +#include "exec/cpu-common.h" +#include "hw/net/allwinner-sun8i-emac.h" + +/* EMAC register offsets */ +enum { + REG_BASIC_CTL_0 = 0x0000, /* Basic Control 0 */ + REG_BASIC_CTL_1 = 0x0004, /* Basic Control 1 */ + REG_INT_STA = 0x0008, /* Interrupt Status */ + REG_INT_EN = 0x000C, /* Interrupt Enable */ + REG_TX_CTL_0 = 0x0010, /* Transmit Control 0 */ + REG_TX_CTL_1 = 0x0014, /* Transmit Control 1 */ + REG_TX_FLOW_CTL = 0x001C, /* Transmit Flow Control */ + REG_TX_DMA_DESC_LIST = 0x0020, /* Transmit Descriptor List Address */ + REG_RX_CTL_0 = 0x0024, /* Receive Control 0 */ + REG_RX_CTL_1 = 0x0028, /* Receive Control 1 */ + REG_RX_DMA_DESC_LIST = 0x0034, /* Receive Descriptor List Address */ + REG_FRM_FLT = 0x0038, /* Receive Frame Filter */ + REG_RX_HASH_0 = 0x0040, /* Receive Hash Table 0 */ + REG_RX_HASH_1 = 0x0044, /* Receive Hash Table 1 */ + REG_MII_CMD = 0x0048, /* Management Interface Command */ + REG_MII_DATA = 0x004C, /* Management Interface Data */ + REG_ADDR_HIGH = 0x0050, /* MAC Address High */ + REG_ADDR_LOW = 0x0054, /* MAC Address Low */ + REG_TX_DMA_STA = 0x00B0, /* Transmit DMA Status */ + REG_TX_CUR_DESC = 0x00B4, /* Transmit Current Descriptor */ + REG_TX_CUR_BUF = 0x00B8, /* Transmit Current Buffer */ + REG_RX_DMA_STA = 0x00C0, /* Receive DMA Status */ + REG_RX_CUR_DESC = 0x00C4, /* Receive Current Descriptor */ + REG_RX_CUR_BUF = 0x00C8, /* Receive Current Buffer */ + REG_RGMII_STA = 0x00D0, /* RGMII Status */ +}; + +/* EMAC register flags */ +enum { + BASIC_CTL0_100Mbps = (0b11 << 2), + BASIC_CTL0_FD = (1 << 0), + BASIC_CTL1_SOFTRST = (1 << 0), +}; + +enum { + INT_STA_RGMII_LINK = (1 << 16), + INT_STA_RX_EARLY = (1 << 13), + INT_STA_RX_OVERFLOW = (1 << 12), + INT_STA_RX_TIMEOUT = (1 << 11), + INT_STA_RX_DMA_STOP = (1 << 10), + INT_STA_RX_BUF_UA = (1 << 9), + INT_STA_RX = (1 << 8), + INT_STA_TX_EARLY = (1 << 5), + INT_STA_TX_UNDERFLOW = (1 << 4), + INT_STA_TX_TIMEOUT = (1 << 3), + INT_STA_TX_BUF_UA = (1 << 2), + INT_STA_TX_DMA_STOP = (1 << 1), + INT_STA_TX = (1 << 0), +}; + +enum { + INT_EN_RX_EARLY = (1 << 13), + INT_EN_RX_OVERFLOW = (1 << 12), + INT_EN_RX_TIMEOUT = (1 << 11), + INT_EN_RX_DMA_STOP = (1 << 10), + INT_EN_RX_BUF_UA = (1 << 9), + INT_EN_RX = (1 << 8), + INT_EN_TX_EARLY = (1 << 5), + INT_EN_TX_UNDERFLOW = (1 << 4), + INT_EN_TX_TIMEOUT = (1 << 3), + INT_EN_TX_BUF_UA = (1 << 2), + INT_EN_TX_DMA_STOP = (1 << 1), + INT_EN_TX = (1 << 0), +}; + +enum { + TX_CTL0_TX_EN = (1 << 31), + TX_CTL1_TX_DMA_START = (1 << 31), + TX_CTL1_TX_DMA_EN = (1 << 30), + TX_CTL1_TX_FLUSH = (1 << 0), +}; + +enum { + RX_CTL0_RX_EN = (1 << 31), + RX_CTL0_STRIP_FCS = (1 << 28), + RX_CTL0_CRC_IPV4 = (1 << 27), +}; + +enum { + RX_CTL1_RX_DMA_START = (1 << 31), + RX_CTL1_RX_DMA_EN = (1 << 30), + RX_CTL1_RX_MD = (1 << 1), +}; + +enum { + RX_FRM_FLT_DIS_ADDR = (1 << 31), +}; + +enum { + MII_CMD_PHY_ADDR_SHIFT = (12), + MII_CMD_PHY_ADDR_MASK = (0xf000), + MII_CMD_PHY_REG_SHIFT = (4), + MII_CMD_PHY_REG_MASK = (0xf0), + MII_CMD_PHY_RW = (1 << 1), + MII_CMD_PHY_BUSY = (1 << 0), +}; + +enum { + TX_DMA_STA_STOP = (0b000), + TX_DMA_STA_RUN_FETCH = (0b001), + TX_DMA_STA_WAIT_STA = (0b010), +}; + +enum { + RX_DMA_STA_STOP = (0b000), + RX_DMA_STA_RUN_FETCH = (0b001), + RX_DMA_STA_WAIT_FRM = (0b011), +}; + +/* EMAC register reset values */ +enum { + REG_BASIC_CTL_1_RST = 0x08000000, +}; + +/* EMAC constants */ +enum { + AW_SUN8I_EMAC_MIN_PKT_SZ = 64 +}; + +/* Transmit/receive frame descriptor */ +typedef struct FrameDescriptor { + uint32_t status; + uint32_t status2; + uint32_t addr; + uint32_t next; +} FrameDescriptor; + +/* Frame descriptor flags */ +enum { + DESC_STATUS_CTL = (1 << 31), + DESC_STATUS2_BUF_SIZE_MASK = (0x7ff), +}; + +/* Transmit frame descriptor flags */ +enum { + TX_DESC_STATUS_LENGTH_ERR = (1 << 14), + TX_DESC_STATUS2_FIRST_DESC = (1 << 29), + TX_DESC_STATUS2_LAST_DESC = (1 << 30), + TX_DESC_STATUS2_CHECKSUM_MASK = (0x3 << 27), +}; + +/* Receive frame descriptor flags */ +enum { + RX_DESC_STATUS_FIRST_DESC = (1 << 9), + RX_DESC_STATUS_LAST_DESC = (1 << 8), + RX_DESC_STATUS_FRM_LEN_MASK = (0x3fff0000), + RX_DESC_STATUS_FRM_LEN_SHIFT = (16), + RX_DESC_STATUS_NO_BUF = (1 << 14), + RX_DESC_STATUS_HEADER_ERR = (1 << 7), + RX_DESC_STATUS_LENGTH_ERR = (1 << 4), + RX_DESC_STATUS_CRC_ERR = (1 << 1), + RX_DESC_STATUS_PAYLOAD_ERR = (1 << 0), + RX_DESC_STATUS2_RX_INT_CTL = (1 << 31), +}; + +/* MII register offsets */ +enum { + MII_REG_CR = (0x0), /* Control */ + MII_REG_ST = (0x1), /* Status */ + MII_REG_ID_HIGH = (0x2), /* Identifier High */ + MII_REG_ID_LOW = (0x3), /* Identifier Low */ + MII_REG_ADV = (0x4), /* Advertised abilities */ + MII_REG_LPA = (0x5), /* Link partner abilities */ +}; + +/* MII register flags */ +enum { + MII_REG_CR_RESET = (1 << 15), + MII_REG_CR_POWERDOWN = (1 << 11), + MII_REG_CR_10Mbit = (0), + MII_REG_CR_100Mbit = (1 << 13), + MII_REG_CR_1000Mbit = (1 << 6), + MII_REG_CR_AUTO_NEG = (1 << 12), + MII_REG_CR_AUTO_NEG_RESTART = (1 << 9), + MII_REG_CR_FULLDUPLEX = (1 << 8), +}; + +enum { + MII_REG_ST_100BASE_T4 = (1 << 15), + MII_REG_ST_100BASE_X_FD = (1 << 14), + MII_REG_ST_100BASE_X_HD = (1 << 13), + MII_REG_ST_10_FD = (1 << 12), + MII_REG_ST_10_HD = (1 << 11), + MII_REG_ST_100BASE_T2_FD = (1 << 10), + MII_REG_ST_100BASE_T2_HD = (1 << 9), + MII_REG_ST_AUTONEG_COMPLETE = (1 << 5), + MII_REG_ST_AUTONEG_AVAIL = (1 << 3), + MII_REG_ST_LINK_UP = (1 << 2), +}; + +enum { + MII_REG_LPA_10_HD = (1 << 5), + MII_REG_LPA_10_FD = (1 << 6), + MII_REG_LPA_100_HD = (1 << 7), + MII_REG_LPA_100_FD = (1 << 8), + MII_REG_LPA_PAUSE = (1 << 10), + MII_REG_LPA_ASYMPAUSE = (1 << 11), +}; + +/* MII constants */ +enum { + MII_PHY_ID_HIGH = 0x0044, + MII_PHY_ID_LOW = 0x1400, +}; + +static void allwinner_sun8i_emac_mii_set_link(AwSun8iEmacState *s, + bool link_active) +{ + if (link_active) { + s->mii_st |= MII_REG_ST_LINK_UP; + } else { + s->mii_st &= ~MII_REG_ST_LINK_UP; + } +} + +static void allwinner_sun8i_emac_mii_reset(AwSun8iEmacState *s, + bool link_active) +{ + s->mii_cr = MII_REG_CR_100Mbit | MII_REG_CR_AUTO_NEG | + MII_REG_CR_FULLDUPLEX; + s->mii_st = MII_REG_ST_100BASE_T4 | MII_REG_ST_100BASE_X_FD | + MII_REG_ST_100BASE_X_HD | MII_REG_ST_10_FD | MII_REG_ST_10_HD | + MII_REG_ST_100BASE_T2_FD | MII_REG_ST_100BASE_T2_HD | + MII_REG_ST_AUTONEG_COMPLETE | MII_REG_ST_AUTONEG_AVAIL; + s->mii_adv = 0; + + allwinner_sun8i_emac_mii_set_link(s, link_active); +} + +static void allwinner_sun8i_emac_mii_cmd(AwSun8iEmacState *s) +{ + uint8_t addr, reg; + + addr = (s->mii_cmd & MII_CMD_PHY_ADDR_MASK) >> MII_CMD_PHY_ADDR_SHIFT; + reg = (s->mii_cmd & MII_CMD_PHY_REG_MASK) >> MII_CMD_PHY_REG_SHIFT; + + if (addr != s->mii_phy_addr) { + return; + } + + /* Read or write a PHY register? */ + if (s->mii_cmd & MII_CMD_PHY_RW) { + trace_allwinner_sun8i_emac_mii_write_reg(reg, s->mii_data); + + switch (reg) { + case MII_REG_CR: + if (s->mii_data & MII_REG_CR_RESET) { + allwinner_sun8i_emac_mii_reset(s, s->mii_st & + MII_REG_ST_LINK_UP); + } else { + s->mii_cr = s->mii_data & ~(MII_REG_CR_RESET | + MII_REG_CR_AUTO_NEG_RESTART); + } + break; + case MII_REG_ADV: + s->mii_adv = s->mii_data; + break; + case MII_REG_ID_HIGH: + case MII_REG_ID_LOW: + case MII_REG_LPA: + break; + default: + qemu_log_mask(LOG_UNIMP, "allwinner-h3-emac: write access to " + "unknown MII register 0x%x\n", reg); + break; + } + } else { + switch (reg) { + case MII_REG_CR: + s->mii_data = s->mii_cr; + break; + case MII_REG_ST: + s->mii_data = s->mii_st; + break; + case MII_REG_ID_HIGH: + s->mii_data = MII_PHY_ID_HIGH; + break; + case MII_REG_ID_LOW: + s->mii_data = MII_PHY_ID_LOW; + break; + case MII_REG_ADV: + s->mii_data = s->mii_adv; + break; + case MII_REG_LPA: + s->mii_data = MII_REG_LPA_10_HD | MII_REG_LPA_10_FD | + MII_REG_LPA_100_HD | MII_REG_LPA_100_FD | + MII_REG_LPA_PAUSE | MII_REG_LPA_ASYMPAUSE; + break; + default: + qemu_log_mask(LOG_UNIMP, "allwinner-h3-emac: read access to " + "unknown MII register 0x%x\n", reg); + s->mii_data = 0; + break; + } + + trace_allwinner_sun8i_emac_mii_read_reg(reg, s->mii_data); + } +} + +static void allwinner_sun8i_emac_update_irq(AwSun8iEmacState *s) +{ + qemu_set_irq(s->irq, (s->int_sta & s->int_en) != 0); +} + +static uint32_t allwinner_sun8i_emac_next_desc(FrameDescriptor *desc, + size_t min_size) +{ + uint32_t paddr = desc->next; + + cpu_physical_memory_read(paddr, desc, sizeof(*desc)); + + if ((desc->status & DESC_STATUS_CTL) && + (desc->status2 & DESC_STATUS2_BUF_SIZE_MASK) >= min_size) { + return paddr; + } else { + return 0; + } +} + +static uint32_t allwinner_sun8i_emac_get_desc(FrameDescriptor *desc, + uint32_t start_addr, + size_t min_size) +{ + uint32_t desc_addr = start_addr; + + /* Note that the list is a cycle. Last entry points back to the head. */ + while (desc_addr != 0) { + cpu_physical_memory_read(desc_addr, desc, sizeof(*desc)); + + if ((desc->status & DESC_STATUS_CTL) && + (desc->status2 & DESC_STATUS2_BUF_SIZE_MASK) >= min_size) { + return desc_addr; + } else if (desc->next == start_addr) { + break; + } else { + desc_addr = desc->next; + } + } + + return 0; +} + +static uint32_t allwinner_sun8i_emac_rx_desc(AwSun8iEmacState *s, + FrameDescriptor *desc, + size_t min_size) +{ + return allwinner_sun8i_emac_get_desc(desc, s->rx_desc_curr, min_size); +} + +static uint32_t allwinner_sun8i_emac_tx_desc(AwSun8iEmacState *s, + FrameDescriptor *desc, + size_t min_size) +{ + return allwinner_sun8i_emac_get_desc(desc, s->tx_desc_head, min_size); +} + +static void allwinner_sun8i_emac_flush_desc(FrameDescriptor *desc, + uint32_t phys_addr) +{ + cpu_physical_memory_write(phys_addr, desc, sizeof(*desc)); +} + +static int allwinner_sun8i_emac_can_receive(NetClientState *nc) +{ + AwSun8iEmacState *s = qemu_get_nic_opaque(nc); + FrameDescriptor desc; + + return (s->rx_ctl0 & RX_CTL0_RX_EN) && + (allwinner_sun8i_emac_rx_desc(s, &desc, 0) != 0); +} + +static ssize_t allwinner_sun8i_emac_receive(NetClientState *nc, + const uint8_t *buf, + size_t size) +{ + AwSun8iEmacState *s = qemu_get_nic_opaque(nc); + FrameDescriptor desc; + size_t bytes_left = size; + size_t desc_bytes = 0; + size_t pad_fcs_size = 4; + size_t padding = 0; + + if (!(s->rx_ctl0 & RX_CTL0_RX_EN)) { + return -1; + } + + s->rx_desc_curr = allwinner_sun8i_emac_rx_desc(s, &desc, + AW_SUN8I_EMAC_MIN_PKT_SZ); + if (!s->rx_desc_curr) { + s->int_sta |= INT_STA_RX_BUF_UA; + } + + /* Keep filling RX descriptors until the whole frame is written */ + while (s->rx_desc_curr && bytes_left > 0) { + desc.status &= ~DESC_STATUS_CTL; + desc.status &= ~RX_DESC_STATUS_FRM_LEN_MASK; + + if (bytes_left == size) { + desc.status |= RX_DESC_STATUS_FIRST_DESC; + } + + if ((desc.status2 & DESC_STATUS2_BUF_SIZE_MASK) < + (bytes_left + pad_fcs_size)) { + desc_bytes = desc.status2 & DESC_STATUS2_BUF_SIZE_MASK; + desc.status |= desc_bytes << RX_DESC_STATUS_FRM_LEN_SHIFT; + } else { + padding = pad_fcs_size; + if (bytes_left < AW_SUN8I_EMAC_MIN_PKT_SZ) { + padding += (AW_SUN8I_EMAC_MIN_PKT_SZ - bytes_left); + } + + desc_bytes = (bytes_left); + desc.status |= RX_DESC_STATUS_LAST_DESC; + desc.status |= (bytes_left + padding) + << RX_DESC_STATUS_FRM_LEN_SHIFT; + } + + cpu_physical_memory_write(desc.addr, buf, desc_bytes); + allwinner_sun8i_emac_flush_desc(&desc, s->rx_desc_curr); + trace_allwinner_sun8i_emac_receive(s->rx_desc_curr, desc.addr, + desc_bytes); + + /* Check if frame needs to raise the receive interrupt */ + if (!(desc.status2 & RX_DESC_STATUS2_RX_INT_CTL)) { + s->int_sta |= INT_STA_RX; + } + + /* Increment variables */ + buf += desc_bytes; + bytes_left -= desc_bytes; + + /* Move to the next descriptor */ + s->rx_desc_curr = allwinner_sun8i_emac_next_desc(&desc, 64); + if (!s->rx_desc_curr) { + /* Not enough buffer space available */ + s->int_sta |= INT_STA_RX_BUF_UA; + s->rx_desc_curr = s->rx_desc_head; + break; + } + } + + /* Report receive DMA is finished */ + s->rx_ctl1 &= ~RX_CTL1_RX_DMA_START; + allwinner_sun8i_emac_update_irq(s); + + return size; +} + +static void allwinner_sun8i_emac_transmit(AwSun8iEmacState *s) +{ + NetClientState *nc = qemu_get_queue(s->nic); + FrameDescriptor desc; + size_t bytes = 0; + size_t packet_bytes = 0; + size_t transmitted = 0; + static uint8_t packet_buf[2048]; + + s->tx_desc_curr = allwinner_sun8i_emac_tx_desc(s, &desc, 0); + + /* Read all transmit descriptors */ + while (s->tx_desc_curr != 0) { + + /* Read from physical memory into packet buffer */ + bytes = desc.status2 & DESC_STATUS2_BUF_SIZE_MASK; + if (bytes + packet_bytes > sizeof(packet_buf)) { + desc.status |= TX_DESC_STATUS_LENGTH_ERR; + break; + } + cpu_physical_memory_read(desc.addr, packet_buf + packet_bytes, bytes); + packet_bytes += bytes; + desc.status &= ~DESC_STATUS_CTL; + allwinner_sun8i_emac_flush_desc(&desc, s->tx_desc_curr); + + /* After the last descriptor, send the packet */ + if (desc.status2 & TX_DESC_STATUS2_LAST_DESC) { + if (desc.status2 & TX_DESC_STATUS2_CHECKSUM_MASK) { + net_checksum_calculate(packet_buf, packet_bytes); + } + + qemu_send_packet(nc, packet_buf, packet_bytes); + trace_allwinner_sun8i_emac_transmit(s->tx_desc_curr, desc.addr, + bytes); + + packet_bytes = 0; + transmitted++; + } + s->tx_desc_curr = allwinner_sun8i_emac_next_desc(&desc, 0); + } + + /* Raise transmit completed interrupt */ + if (transmitted > 0) { + s->int_sta |= INT_STA_TX; + s->tx_ctl1 &= ~TX_CTL1_TX_DMA_START; + allwinner_sun8i_emac_update_irq(s); + } +} + +static void allwinner_sun8i_emac_reset(DeviceState *dev) +{ + AwSun8iEmacState *s = AW_SUN8I_EMAC(dev); + NetClientState *nc = qemu_get_queue(s->nic); + + trace_allwinner_sun8i_emac_reset(); + + s->mii_cmd = 0; + s->mii_data = 0; + s->basic_ctl0 = 0; + s->basic_ctl1 = REG_BASIC_CTL_1_RST; + s->int_en = 0; + s->int_sta = 0; + s->frm_flt = 0; + s->rx_ctl0 = 0; + s->rx_ctl1 = RX_CTL1_RX_MD; + s->rx_desc_head = 0; + s->rx_desc_curr = 0; + s->tx_ctl0 = 0; + s->tx_ctl1 = 0; + s->tx_desc_head = 0; + s->tx_desc_curr = 0; + s->tx_flowctl = 0; + + allwinner_sun8i_emac_mii_reset(s, !nc->link_down); +} + +static uint64_t allwinner_sun8i_emac_read(void *opaque, hwaddr offset, + unsigned size) +{ + AwSun8iEmacState *s = AW_SUN8I_EMAC(opaque); + uint64_t value = 0; + FrameDescriptor desc; + + switch (offset) { + case REG_BASIC_CTL_0: /* Basic Control 0 */ + value = s->basic_ctl0; + break; + case REG_BASIC_CTL_1: /* Basic Control 1 */ + value = s->basic_ctl1; + break; + case REG_INT_STA: /* Interrupt Status */ + value = s->int_sta; + break; + case REG_INT_EN: /* Interupt Enable */ + value = s->int_en; + break; + case REG_TX_CTL_0: /* Transmit Control 0 */ + value = s->tx_ctl0; + break; + case REG_TX_CTL_1: /* Transmit Control 1 */ + value = s->tx_ctl1; + break; + case REG_TX_FLOW_CTL: /* Transmit Flow Control */ + value = s->tx_flowctl; + break; + case REG_TX_DMA_DESC_LIST: /* Transmit Descriptor List Address */ + value = s->tx_desc_head; + break; + case REG_RX_CTL_0: /* Receive Control 0 */ + value = s->rx_ctl0; + break; + case REG_RX_CTL_1: /* Receive Control 1 */ + value = s->rx_ctl1; + break; + case REG_RX_DMA_DESC_LIST: /* Receive Descriptor List Address */ + value = s->rx_desc_head; + break; + case REG_FRM_FLT: /* Receive Frame Filter */ + value = s->frm_flt; + break; + case REG_RX_HASH_0: /* Receive Hash Table 0 */ + case REG_RX_HASH_1: /* Receive Hash Table 1 */ + break; + case REG_MII_CMD: /* Management Interface Command */ + value = s->mii_cmd; + break; + case REG_MII_DATA: /* Management Interface Data */ + value = s->mii_data; + break; + case REG_ADDR_HIGH: /* MAC Address High */ + value = *(((uint32_t *) (s->conf.macaddr.a)) + 1); + break; + case REG_ADDR_LOW: /* MAC Address Low */ + value = *(uint32_t *) (s->conf.macaddr.a); + break; + case REG_TX_DMA_STA: /* Transmit DMA Status */ + break; + case REG_TX_CUR_DESC: /* Transmit Current Descriptor */ + value = s->tx_desc_curr; + break; + case REG_TX_CUR_BUF: /* Transmit Current Buffer */ + if (s->tx_desc_curr != 0) { + cpu_physical_memory_read(s->tx_desc_curr, &desc, sizeof(desc)); + value = desc.addr; + } else { + value = 0; + } + break; + case REG_RX_DMA_STA: /* Receive DMA Status */ + break; + case REG_RX_CUR_DESC: /* Receive Current Descriptor */ + value = s->rx_desc_curr; + break; + case REG_RX_CUR_BUF: /* Receive Current Buffer */ + if (s->rx_desc_curr != 0) { + cpu_physical_memory_read(s->rx_desc_curr, &desc, sizeof(desc)); + value = desc.addr; + } else { + value = 0; + } + break; + case REG_RGMII_STA: /* RGMII Status */ + break; + default: + qemu_log_mask(LOG_UNIMP, "allwinner-h3-emac: read access to unknown " + "EMAC register 0x" TARGET_FMT_plx "\n", + offset); + } + + trace_allwinner_sun8i_emac_read(offset, value); + return value; +} + +static void allwinner_sun8i_emac_write(void *opaque, hwaddr offset, + uint64_t value, unsigned size) +{ + AwSun8iEmacState *s = AW_SUN8I_EMAC(opaque); + NetClientState *nc = qemu_get_queue(s->nic); + + trace_allwinner_sun8i_emac_write(offset, value); + + switch (offset) { + case REG_BASIC_CTL_0: /* Basic Control 0 */ + s->basic_ctl0 = value; + break; + case REG_BASIC_CTL_1: /* Basic Control 1 */ + if (value & BASIC_CTL1_SOFTRST) { + allwinner_sun8i_emac_reset(DEVICE(s)); + value &= ~BASIC_CTL1_SOFTRST; + } + s->basic_ctl1 = value; + if (allwinner_sun8i_emac_can_receive(nc)) { + qemu_flush_queued_packets(nc); + } + break; + case REG_INT_STA: /* Interrupt Status */ + s->int_sta &= ~value; + allwinner_sun8i_emac_update_irq(s); + break; + case REG_INT_EN: /* Interrupt Enable */ + s->int_en = value; + allwinner_sun8i_emac_update_irq(s); + break; + case REG_TX_CTL_0: /* Transmit Control 0 */ + s->tx_ctl0 = value; + break; + case REG_TX_CTL_1: /* Transmit Control 1 */ + s->tx_ctl1 = value; + if (value & TX_CTL1_TX_DMA_EN) { + allwinner_sun8i_emac_transmit(s); + } + break; + case REG_TX_FLOW_CTL: /* Transmit Flow Control */ + s->tx_flowctl = value; + break; + case REG_TX_DMA_DESC_LIST: /* Transmit Descriptor List Address */ + s->tx_desc_head = value; + s->tx_desc_curr = value; + break; + case REG_RX_CTL_0: /* Receive Control 0 */ + s->rx_ctl0 = value; + break; + case REG_RX_CTL_1: /* Receive Control 1 */ + s->rx_ctl1 = value | RX_CTL1_RX_MD; + if ((value & RX_CTL1_RX_DMA_EN) && + allwinner_sun8i_emac_can_receive(nc)) { + qemu_flush_queued_packets(nc); + } + break; + case REG_RX_DMA_DESC_LIST: /* Receive Descriptor List Address */ + s->rx_desc_head = value; + s->rx_desc_curr = value; + break; + case REG_FRM_FLT: /* Receive Frame Filter */ + s->frm_flt = value; + break; + case REG_RX_HASH_0: /* Receive Hash Table 0 */ + case REG_RX_HASH_1: /* Receive Hash Table 1 */ + break; + case REG_MII_CMD: /* Management Interface Command */ + s->mii_cmd = value & ~MII_CMD_PHY_BUSY; + allwinner_sun8i_emac_mii_cmd(s); + break; + case REG_MII_DATA: /* Management Interface Data */ + s->mii_data = value; + break; + case REG_ADDR_HIGH: /* MAC Address High */ + s->conf.macaddr.a[4] = (value & 0xff); + s->conf.macaddr.a[5] = (value & 0xff00) >> 8; + break; + case REG_ADDR_LOW: /* MAC Address Low */ + s->conf.macaddr.a[0] = (value & 0xff); + s->conf.macaddr.a[1] = (value & 0xff00) >> 8; + s->conf.macaddr.a[2] = (value & 0xff0000) >> 16; + s->conf.macaddr.a[3] = (value & 0xff000000) >> 24; + break; + case REG_TX_DMA_STA: /* Transmit DMA Status */ + case REG_TX_CUR_DESC: /* Transmit Current Descriptor */ + case REG_TX_CUR_BUF: /* Transmit Current Buffer */ + case REG_RX_DMA_STA: /* Receive DMA Status */ + case REG_RX_CUR_DESC: /* Receive Current Descriptor */ + case REG_RX_CUR_BUF: /* Receive Current Buffer */ + case REG_RGMII_STA: /* RGMII Status */ + break; + default: + qemu_log_mask(LOG_UNIMP, "allwinner-h3-emac: write access to unknown " + "EMAC register 0x" TARGET_FMT_plx "\n", + offset); + } +} + +static void allwinner_sun8i_emac_set_link(NetClientState *nc) +{ + AwSun8iEmacState *s = qemu_get_nic_opaque(nc); + + trace_allwinner_sun8i_emac_set_link(!nc->link_down); + allwinner_sun8i_emac_mii_set_link(s, !nc->link_down); +} + +static const MemoryRegionOps allwinner_sun8i_emac_mem_ops = { + .read = allwinner_sun8i_emac_read, + .write = allwinner_sun8i_emac_write, + .endianness = DEVICE_NATIVE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, + .impl.min_access_size = 4, +}; + +static NetClientInfo net_allwinner_sun8i_emac_info = { + .type = NET_CLIENT_DRIVER_NIC, + .size = sizeof(NICState), + .can_receive = allwinner_sun8i_emac_can_receive, + .receive = allwinner_sun8i_emac_receive, + .link_status_changed = allwinner_sun8i_emac_set_link, +}; + +static void allwinner_sun8i_emac_init(Object *obj) +{ + SysBusDevice *sbd = SYS_BUS_DEVICE(obj); + AwSun8iEmacState *s = AW_SUN8I_EMAC(obj); + + memory_region_init_io(&s->iomem, OBJECT(s), &allwinner_sun8i_emac_mem_ops, + s, TYPE_AW_SUN8I_EMAC, 64 * KiB); + sysbus_init_mmio(sbd, &s->iomem); + sysbus_init_irq(sbd, &s->irq); +} + +static void allwinner_sun8i_emac_realize(DeviceState *dev, Error **errp) +{ + AwSun8iEmacState *s = AW_SUN8I_EMAC(dev); + + qemu_macaddr_default_if_unset(&s->conf.macaddr); + s->nic = qemu_new_nic(&net_allwinner_sun8i_emac_info, &s->conf, + object_get_typename(OBJECT(dev)), dev->id, s); + qemu_format_nic_info_str(qemu_get_queue(s->nic), s->conf.macaddr.a); +} + +static Property allwinner_sun8i_emac_properties[] = { + DEFINE_NIC_PROPERTIES(AwSun8iEmacState, conf), + DEFINE_PROP_UINT8("phy-addr", AwSun8iEmacState, mii_phy_addr, 0), + DEFINE_PROP_END_OF_LIST(), +}; + +static int allwinner_sun8i_emac_post_load(void *opaque, int version_id) +{ + AwSun8iEmacState *s = opaque; + + allwinner_sun8i_emac_set_link(qemu_get_queue(s->nic)); + + return 0; +} + +static const VMStateDescription vmstate_aw_emac = { + .name = "allwinner-sun8i-emac", + .version_id = 1, + .minimum_version_id = 1, + .post_load = allwinner_sun8i_emac_post_load, + .fields = (VMStateField[]) { + VMSTATE_UINT8(mii_phy_addr, AwSun8iEmacState), + VMSTATE_UINT32(mii_cmd, AwSun8iEmacState), + VMSTATE_UINT32(mii_data, AwSun8iEmacState), + VMSTATE_UINT32(mii_cr, AwSun8iEmacState), + VMSTATE_UINT32(mii_st, AwSun8iEmacState), + VMSTATE_UINT32(mii_adv, AwSun8iEmacState), + VMSTATE_UINT32(basic_ctl0, AwSun8iEmacState), + VMSTATE_UINT32(basic_ctl1, AwSun8iEmacState), + VMSTATE_UINT32(int_en, AwSun8iEmacState), + VMSTATE_UINT32(int_sta, AwSun8iEmacState), + VMSTATE_UINT32(frm_flt, AwSun8iEmacState), + VMSTATE_UINT32(rx_ctl0, AwSun8iEmacState), + VMSTATE_UINT32(rx_ctl1, AwSun8iEmacState), + VMSTATE_UINT32(rx_desc_head, AwSun8iEmacState), + VMSTATE_UINT32(rx_desc_curr, AwSun8iEmacState), + VMSTATE_UINT32(tx_ctl0, AwSun8iEmacState), + VMSTATE_UINT32(tx_ctl1, AwSun8iEmacState), + VMSTATE_UINT32(tx_desc_head, AwSun8iEmacState), + VMSTATE_UINT32(tx_desc_curr, AwSun8iEmacState), + VMSTATE_UINT32(tx_flowctl, AwSun8iEmacState), + VMSTATE_END_OF_LIST() + } +}; + +static void allwinner_sun8i_emac_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->realize = allwinner_sun8i_emac_realize; + dc->reset = allwinner_sun8i_emac_reset; + dc->vmsd = &vmstate_aw_emac; + device_class_set_props(dc, allwinner_sun8i_emac_properties); +} + +static const TypeInfo allwinner_sun8i_emac_info = { + .name = TYPE_AW_SUN8I_EMAC, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_size = sizeof(AwSun8iEmacState), + .instance_init = allwinner_sun8i_emac_init, + .class_init = allwinner_sun8i_emac_class_init, +}; + +static void allwinner_sun8i_emac_register_types(void) +{ + type_register_static(&allwinner_sun8i_emac_info); +} + +type_init(allwinner_sun8i_emac_register_types) diff --git a/hw/arm/Kconfig b/hw/arm/Kconfig index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/Kconfig +++ b/hw/arm/Kconfig @@ -XXX,XX +XXX,XX @@ config ALLWINNER_A10 config ALLWINNER_H3 bool select ALLWINNER_A10_PIT + select ALLWINNER_SUN8I_EMAC select SERIAL select ARM_TIMER select ARM_GIC diff --git a/hw/net/Kconfig b/hw/net/Kconfig index XXXXXXX..XXXXXXX 100644 --- a/hw/net/Kconfig +++ b/hw/net/Kconfig @@ -XXX,XX +XXX,XX @@ config MIPSNET config ALLWINNER_EMAC bool +config ALLWINNER_SUN8I_EMAC + bool + config IMX_FEC bool diff --git a/hw/net/trace-events b/hw/net/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/net/trace-events +++ b/hw/net/trace-events @@ -XXX,XX +XXX,XX @@ # See docs/devel/tracing.txt for syntax documentation. +# allwinner-sun8i-emac.c +allwinner_sun8i_emac_mii_write_reg(uint32_t reg, uint32_t value) "MII write: reg=0x%" PRIx32 " value=0x%" PRIx32 +allwinner_sun8i_emac_mii_read_reg(uint32_t reg, uint32_t value) "MII read: reg=0x%" PRIx32 " value=0x%" PRIx32 +allwinner_sun8i_emac_receive(uint32_t desc, uint32_t paddr, uint32_t bytes) "RX packet: desc=0x%" PRIx32 " paddr=0x%" PRIx32 " bytes=%" PRIu32 +allwinner_sun8i_emac_transmit(uint32_t desc, uint32_t paddr, uint32_t bytes) "TX packet: desc=0x%" PRIx32 " paddr=0x%" PRIx32 " bytes=%" PRIu32 +allwinner_sun8i_emac_reset(void) "HW reset" +allwinner_sun8i_emac_set_link(bool active) "Set link: active=%u" +allwinner_sun8i_emac_read(uint64_t offset, uint64_t val) "MMIO read: offset=0x%" PRIx64 " value=0x%" PRIx64 +allwinner_sun8i_emac_write(uint64_t offset, uint64_t val) "MMIO write: offset=0x%" PRIx64 " value=0x%" PRIx64 + # etraxfs_eth.c mdio_phy_read(int regnum, uint16_t value) "read phy_reg:%d value:0x%04x" mdio_phy_write(int regnum, uint16_t value) "write phy_reg:%d value:0x%04x" -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> A real Allwinner H3 SoC contains a Boot ROM which is the first code that runs right after the SoC is powered on. The Boot ROM is responsible for loading user code (e.g. a bootloader) from any of the supported external devices and writing the downloaded code to internal SRAM. After loading the SoC begins executing the code written to SRAM. This commits adds emulation of the Boot ROM firmware setup functionality by loading user code from SD card in the A1 SRAM. While the A1 SRAM is 64KiB, we limit the size to 32KiB because the real H3 Boot ROM also rejects sizes larger than 32KiB. For reference, this behaviour is documented by the Linux Sunxi project wiki at: https://linux-sunxi.org/BROM#U-Boot_SPL_limitations Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20200311221854.30370-11-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/allwinner-h3.h | 21 +++++++++++++++++++++ hw/arm/allwinner-h3.c | 17 +++++++++++++++++ hw/arm/orangepi.c | 5 +++++ 3 files changed, 43 insertions(+) diff --git a/include/hw/arm/allwinner-h3.h b/include/hw/arm/allwinner-h3.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-h3.h +++ b/include/hw/arm/allwinner-h3.h @@ -XXX,XX +XXX,XX @@ #include "hw/sd/allwinner-sdhost.h" #include "hw/net/allwinner-sun8i-emac.h" #include "target/arm/cpu.h" +#include "sysemu/block-backend.h" /** * Allwinner H3 device list @@ -XXX,XX +XXX,XX @@ typedef struct AwH3State { MemoryRegion sram_c; } AwH3State; +/** + * Emulate Boot ROM firmware setup functionality. + * + * A real Allwinner H3 SoC contains a Boot ROM + * which is the first code that runs right after + * the SoC is powered on. The Boot ROM is responsible + * for loading user code (e.g. a bootloader) from any + * of the supported external devices and writing the + * downloaded code to internal SRAM. After loading the SoC + * begins executing the code written to SRAM. + * + * This function emulates the Boot ROM by copying 32 KiB + * of data from the given block device and writes it to + * the start of the first internal SRAM memory. + * + * @s: Allwinner H3 state object pointer + * @blk: Block backend device object pointer + */ +void allwinner_h3_bootrom_setup(AwH3State *s, BlockBackend *blk); + #endif /* HW_ARM_ALLWINNER_H3_H */ diff --git a/hw/arm/allwinner-h3.c b/hw/arm/allwinner-h3.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/allwinner-h3.c +++ b/hw/arm/allwinner-h3.c @@ -XXX,XX +XXX,XX @@ #include "hw/char/serial.h" #include "hw/misc/unimp.h" #include "hw/usb/hcd-ehci.h" +#include "hw/loader.h" #include "sysemu/sysemu.h" #include "hw/arm/allwinner-h3.h" @@ -XXX,XX +XXX,XX @@ enum { AW_H3_GIC_NUM_SPI = 128 }; +void allwinner_h3_bootrom_setup(AwH3State *s, BlockBackend *blk) +{ + const int64_t rom_size = 32 * KiB; + g_autofree uint8_t *buffer = g_new0(uint8_t, rom_size); + + if (blk_pread(blk, 8 * KiB, buffer, rom_size) < 0) { + error_setg(&error_fatal, "%s: failed to read BlockBackend data", + __func__); + return; + } + + rom_add_blob("allwinner-h3.bootrom", buffer, rom_size, + rom_size, s->memmap[AW_H3_SRAM_A1], + NULL, NULL, NULL, NULL, false); +} + static void allwinner_h3_init(Object *obj) { AwH3State *s = AW_H3(obj); diff --git a/hw/arm/orangepi.c b/hw/arm/orangepi.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/orangepi.c +++ b/hw/arm/orangepi.c @@ -XXX,XX +XXX,XX @@ static void orangepi_init(MachineState *machine) memory_region_add_subregion(get_system_memory(), h3->memmap[AW_H3_SDRAM], machine->ram); + /* Load target kernel or start using BootROM */ + if (!machine->kernel_filename && blk_is_available(blk)) { + /* Use Boot ROM to copy data from SD card to SRAM */ + allwinner_h3_bootrom_setup(h3, blk); + } orangepi_binfo.loader_start = h3->memmap[AW_H3_SDRAM]; orangepi_binfo.ram_size = machine->ram_size; arm_load_kernel(ARM_CPU(first_cpu), machine, &orangepi_binfo); -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> In the Allwinner H3 SoC the SDRAM controller is responsible for interfacing with the external Synchronous Dynamic Random Access Memory (SDRAM). Types of memory that the SDRAM controller supports are DDR2/DDR3 and capacities of up to 2GiB. This commit adds emulation support of the Allwinner H3 SDRAM controller. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20200311221854.30370-12-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/misc/Makefile.objs | 1 + include/hw/arm/allwinner-h3.h | 5 + include/hw/misc/allwinner-h3-dramc.h | 106 ++++++++ hw/arm/allwinner-h3.c | 19 +- hw/arm/orangepi.c | 6 + hw/misc/allwinner-h3-dramc.c | 358 +++++++++++++++++++++++++++ hw/misc/trace-events | 10 + 7 files changed, 502 insertions(+), 3 deletions(-) create mode 100644 include/hw/misc/allwinner-h3-dramc.h create mode 100644 hw/misc/allwinner-h3-dramc.c diff --git a/hw/misc/Makefile.objs b/hw/misc/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/Makefile.objs +++ b/hw/misc/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_IVSHMEM_DEVICE) += ivshmem.o common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3-ccu.o obj-$(CONFIG_ALLWINNER_H3) += allwinner-cpucfg.o +common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3-dramc.o common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-h3-sysctrl.o common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-sid.o common-obj-$(CONFIG_REALVIEW) += arm_sysctl.o diff --git a/include/hw/arm/allwinner-h3.h b/include/hw/arm/allwinner-h3.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-h3.h +++ b/include/hw/arm/allwinner-h3.h @@ -XXX,XX +XXX,XX @@ #include "hw/intc/arm_gic.h" #include "hw/misc/allwinner-h3-ccu.h" #include "hw/misc/allwinner-cpucfg.h" +#include "hw/misc/allwinner-h3-dramc.h" #include "hw/misc/allwinner-h3-sysctrl.h" #include "hw/misc/allwinner-sid.h" #include "hw/sd/allwinner-sdhost.h" @@ -XXX,XX +XXX,XX @@ enum { AW_H3_UART2, AW_H3_UART3, AW_H3_EMAC, + AW_H3_DRAMCOM, + AW_H3_DRAMCTL, + AW_H3_DRAMPHY, AW_H3_GIC_DIST, AW_H3_GIC_CPU, AW_H3_GIC_HYP, @@ -XXX,XX +XXX,XX @@ typedef struct AwH3State { AwA10PITState timer; AwH3ClockCtlState ccu; AwCpuCfgState cpucfg; + AwH3DramCtlState dramc; AwH3SysCtrlState sysctrl; AwSidState sid; AwSdHostState mmc0; diff --git a/include/hw/misc/allwinner-h3-dramc.h b/include/hw/misc/allwinner-h3-dramc.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/misc/allwinner-h3-dramc.h @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner H3 SDRAM Controller emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef HW_MISC_ALLWINNER_H3_DRAMC_H +#define HW_MISC_ALLWINNER_H3_DRAMC_H + +#include "qom/object.h" +#include "hw/sysbus.h" +#include "exec/hwaddr.h" + +/** + * Constants + * @{ + */ + +/** Highest register address used by DRAMCOM module */ +#define AW_H3_DRAMCOM_REGS_MAXADDR (0x804) + +/** Total number of known DRAMCOM registers */ +#define AW_H3_DRAMCOM_REGS_NUM (AW_H3_DRAMCOM_REGS_MAXADDR / \ + sizeof(uint32_t)) + +/** Highest register address used by DRAMCTL module */ +#define AW_H3_DRAMCTL_REGS_MAXADDR (0x88c) + +/** Total number of known DRAMCTL registers */ +#define AW_H3_DRAMCTL_REGS_NUM (AW_H3_DRAMCTL_REGS_MAXADDR / \ + sizeof(uint32_t)) + +/** Highest register address used by DRAMPHY module */ +#define AW_H3_DRAMPHY_REGS_MAXADDR (0x4) + +/** Total number of known DRAMPHY registers */ +#define AW_H3_DRAMPHY_REGS_NUM (AW_H3_DRAMPHY_REGS_MAXADDR / \ + sizeof(uint32_t)) + +/** @} */ + +/** + * Object model + * @{ + */ + +#define TYPE_AW_H3_DRAMC "allwinner-h3-dramc" +#define AW_H3_DRAMC(obj) \ + OBJECT_CHECK(AwH3DramCtlState, (obj), TYPE_AW_H3_DRAMC) + +/** @} */ + +/** + * Allwinner H3 SDRAM Controller object instance state. + */ +typedef struct AwH3DramCtlState { + /*< private >*/ + SysBusDevice parent_obj; + /*< public >*/ + + /** Physical base address for start of RAM */ + hwaddr ram_addr; + + /** Total RAM size in megabytes */ + uint32_t ram_size; + + /** + * @name Memory Regions + * @{ + */ + + MemoryRegion row_mirror; /**< Simulates rows for RAM size detection */ + MemoryRegion row_mirror_alias; /**< Alias of the row which is mirrored */ + MemoryRegion dramcom_iomem; /**< DRAMCOM module I/O registers */ + MemoryRegion dramctl_iomem; /**< DRAMCTL module I/O registers */ + MemoryRegion dramphy_iomem; /**< DRAMPHY module I/O registers */ + + /** @} */ + + /** + * @name Hardware Registers + * @{ + */ + + uint32_t dramcom[AW_H3_DRAMCOM_REGS_NUM]; /**< Array of DRAMCOM registers */ + uint32_t dramctl[AW_H3_DRAMCTL_REGS_NUM]; /**< Array of DRAMCTL registers */ + uint32_t dramphy[AW_H3_DRAMPHY_REGS_NUM] ;/**< Array of DRAMPHY registers */ + + /** @} */ + +} AwH3DramCtlState; + +#endif /* HW_MISC_ALLWINNER_H3_DRAMC_H */ diff --git a/hw/arm/allwinner-h3.c b/hw/arm/allwinner-h3.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/allwinner-h3.c +++ b/hw/arm/allwinner-h3.c @@ -XXX,XX +XXX,XX @@ const hwaddr allwinner_h3_memmap[] = { [AW_H3_UART2] = 0x01c28800, [AW_H3_UART3] = 0x01c28c00, [AW_H3_EMAC] = 0x01c30000, + [AW_H3_DRAMCOM] = 0x01c62000, + [AW_H3_DRAMCTL] = 0x01c63000, + [AW_H3_DRAMPHY] = 0x01c65000, [AW_H3_GIC_DIST] = 0x01c81000, [AW_H3_GIC_CPU] = 0x01c82000, [AW_H3_GIC_HYP] = 0x01c84000, @@ -XXX,XX +XXX,XX @@ struct AwH3Unimplemented { { "scr", 0x01c2c400, 1 * KiB }, { "gpu", 0x01c40000, 64 * KiB }, { "hstmr", 0x01c60000, 4 * KiB }, - { "dramcom", 0x01c62000, 4 * KiB }, - { "dramctl0", 0x01c63000, 4 * KiB }, - { "dramphy0", 0x01c65000, 4 * KiB }, { "spi0", 0x01c68000, 4 * KiB }, { "spi1", 0x01c69000, 4 * KiB }, { "csi", 0x01cb0000, 320 * KiB }, @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_init(Object *obj) sysbus_init_child_obj(obj, "emac", &s->emac, sizeof(s->emac), TYPE_AW_SUN8I_EMAC); + + sysbus_init_child_obj(obj, "dramc", &s->dramc, sizeof(s->dramc), + TYPE_AW_H3_DRAMC); + object_property_add_alias(obj, "ram-addr", OBJECT(&s->dramc), + "ram-addr", &error_abort); + object_property_add_alias(obj, "ram-size", OBJECT(&s->dramc), + "ram-size", &error_abort); } static void allwinner_h3_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_realize(DeviceState *dev, Error **errp) qdev_get_gpio_in(DEVICE(&s->gic), AW_H3_GIC_SPI_UART3), 115200, serial_hd(3), DEVICE_NATIVE_ENDIAN); + /* DRAMC */ + qdev_init_nofail(DEVICE(&s->dramc)); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->dramc), 0, s->memmap[AW_H3_DRAMCOM]); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->dramc), 1, s->memmap[AW_H3_DRAMCTL]); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->dramc), 2, s->memmap[AW_H3_DRAMPHY]); + /* Unimplemented devices */ for (i = 0; i < ARRAY_SIZE(unimplemented); i++) { create_unimplemented_device(unimplemented[i].device_name, diff --git a/hw/arm/orangepi.c b/hw/arm/orangepi.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/orangepi.c +++ b/hw/arm/orangepi.c @@ -XXX,XX +XXX,XX @@ static void orangepi_init(MachineState *machine) /* Setup EMAC properties */ object_property_set_int(OBJECT(&h3->emac), 1, "phy-addr", &error_abort); + /* DRAMC */ + object_property_set_uint(OBJECT(h3), h3->memmap[AW_H3_SDRAM], + "ram-addr", &error_abort); + object_property_set_int(OBJECT(h3), machine->ram_size / MiB, "ram-size", + &error_abort); + /* Mark H3 object realized */ object_property_set_bool(OBJECT(h3), true, "realized", &error_abort); diff --git a/hw/misc/allwinner-h3-dramc.c b/hw/misc/allwinner-h3-dramc.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/misc/allwinner-h3-dramc.c @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner H3 SDRAM Controller emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu/units.h" +#include "qemu/error-report.h" +#include "hw/sysbus.h" +#include "migration/vmstate.h" +#include "qemu/log.h" +#include "qemu/module.h" +#include "exec/address-spaces.h" +#include "hw/qdev-properties.h" +#include "qapi/error.h" +#include "hw/misc/allwinner-h3-dramc.h" +#include "trace.h" + +#define REG_INDEX(offset) (offset / sizeof(uint32_t)) + +/* DRAMCOM register offsets */ +enum { + REG_DRAMCOM_CR = 0x0000, /* Control Register */ +}; + +/* DRAMCTL register offsets */ +enum { + REG_DRAMCTL_PIR = 0x0000, /* PHY Initialization Register */ + REG_DRAMCTL_PGSR = 0x0010, /* PHY General Status Register */ + REG_DRAMCTL_STATR = 0x0018, /* Status Register */ +}; + +/* DRAMCTL register flags */ +enum { + REG_DRAMCTL_PGSR_INITDONE = (1 << 0), +}; + +enum { + REG_DRAMCTL_STATR_ACTIVE = (1 << 0), +}; + +static void allwinner_h3_dramc_map_rows(AwH3DramCtlState *s, uint8_t row_bits, + uint8_t bank_bits, uint16_t page_size) +{ + /* + * This function simulates row addressing behavior when bootloader + * software attempts to detect the amount of available SDRAM. In U-Boot + * the controller is configured with the widest row addressing available. + * Then a pattern is written to RAM at an offset on the row boundary size. + * If the value read back equals the value read back from the + * start of RAM, the bootloader knows the amount of row bits. + * + * This function inserts a mirrored memory region when the configured row + * bits are not matching the actual emulated memory, to simulate the + * same behavior on hardware as expected by the bootloader. + */ + uint8_t row_bits_actual = 0; + + /* Calculate the actual row bits using the ram_size property */ + for (uint8_t i = 8; i < 12; i++) { + if (1 << i == s->ram_size) { + row_bits_actual = i + 3; + break; + } + } + + if (s->ram_size == (1 << (row_bits - 3))) { + /* When row bits is the expected value, remove the mirror */ + memory_region_set_enabled(&s->row_mirror_alias, false); + trace_allwinner_h3_dramc_rowmirror_disable(); + + } else if (row_bits_actual) { + /* Row bits not matching ram_size, install the rows mirror */ + hwaddr row_mirror = s->ram_addr + ((1 << (row_bits_actual + + bank_bits)) * page_size); + + memory_region_set_enabled(&s->row_mirror_alias, true); + memory_region_set_address(&s->row_mirror_alias, row_mirror); + + trace_allwinner_h3_dramc_rowmirror_enable(row_mirror); + } +} + +static uint64_t allwinner_h3_dramcom_read(void *opaque, hwaddr offset, + unsigned size) +{ + const AwH3DramCtlState *s = AW_H3_DRAMC(opaque); + const uint32_t idx = REG_INDEX(offset); + + if (idx >= AW_H3_DRAMCOM_REGS_NUM) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + return 0; + } + + trace_allwinner_h3_dramcom_read(offset, s->dramcom[idx], size); + + return s->dramcom[idx]; +} + +static void allwinner_h3_dramcom_write(void *opaque, hwaddr offset, + uint64_t val, unsigned size) +{ + AwH3DramCtlState *s = AW_H3_DRAMC(opaque); + const uint32_t idx = REG_INDEX(offset); + + trace_allwinner_h3_dramcom_write(offset, val, size); + + if (idx >= AW_H3_DRAMCOM_REGS_NUM) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + return; + } + + switch (offset) { + case REG_DRAMCOM_CR: /* Control Register */ + allwinner_h3_dramc_map_rows(s, ((val >> 4) & 0xf) + 1, + ((val >> 2) & 0x1) + 2, + 1 << (((val >> 8) & 0xf) + 3)); + break; + default: + break; + }; + + s->dramcom[idx] = (uint32_t) val; +} + +static uint64_t allwinner_h3_dramctl_read(void *opaque, hwaddr offset, + unsigned size) +{ + const AwH3DramCtlState *s = AW_H3_DRAMC(opaque); + const uint32_t idx = REG_INDEX(offset); + + if (idx >= AW_H3_DRAMCTL_REGS_NUM) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + return 0; + } + + trace_allwinner_h3_dramctl_read(offset, s->dramctl[idx], size); + + return s->dramctl[idx]; +} + +static void allwinner_h3_dramctl_write(void *opaque, hwaddr offset, + uint64_t val, unsigned size) +{ + AwH3DramCtlState *s = AW_H3_DRAMC(opaque); + const uint32_t idx = REG_INDEX(offset); + + trace_allwinner_h3_dramctl_write(offset, val, size); + + if (idx >= AW_H3_DRAMCTL_REGS_NUM) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + return; + } + + switch (offset) { + case REG_DRAMCTL_PIR: /* PHY Initialization Register */ + s->dramctl[REG_INDEX(REG_DRAMCTL_PGSR)] |= REG_DRAMCTL_PGSR_INITDONE; + s->dramctl[REG_INDEX(REG_DRAMCTL_STATR)] |= REG_DRAMCTL_STATR_ACTIVE; + break; + default: + break; + } + + s->dramctl[idx] = (uint32_t) val; +} + +static uint64_t allwinner_h3_dramphy_read(void *opaque, hwaddr offset, + unsigned size) +{ + const AwH3DramCtlState *s = AW_H3_DRAMC(opaque); + const uint32_t idx = REG_INDEX(offset); + + if (idx >= AW_H3_DRAMPHY_REGS_NUM) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + return 0; + } + + trace_allwinner_h3_dramphy_read(offset, s->dramphy[idx], size); + + return s->dramphy[idx]; +} + +static void allwinner_h3_dramphy_write(void *opaque, hwaddr offset, + uint64_t val, unsigned size) +{ + AwH3DramCtlState *s = AW_H3_DRAMC(opaque); + const uint32_t idx = REG_INDEX(offset); + + trace_allwinner_h3_dramphy_write(offset, val, size); + + if (idx >= AW_H3_DRAMPHY_REGS_NUM) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + return; + } + + s->dramphy[idx] = (uint32_t) val; +} + +static const MemoryRegionOps allwinner_h3_dramcom_ops = { + .read = allwinner_h3_dramcom_read, + .write = allwinner_h3_dramcom_write, + .endianness = DEVICE_NATIVE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, + .impl.min_access_size = 4, +}; + +static const MemoryRegionOps allwinner_h3_dramctl_ops = { + .read = allwinner_h3_dramctl_read, + .write = allwinner_h3_dramctl_write, + .endianness = DEVICE_NATIVE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, + .impl.min_access_size = 4, +}; + +static const MemoryRegionOps allwinner_h3_dramphy_ops = { + .read = allwinner_h3_dramphy_read, + .write = allwinner_h3_dramphy_write, + .endianness = DEVICE_NATIVE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, + .impl.min_access_size = 4, +}; + +static void allwinner_h3_dramc_reset(DeviceState *dev) +{ + AwH3DramCtlState *s = AW_H3_DRAMC(dev); + + /* Set default values for registers */ + memset(&s->dramcom, 0, sizeof(s->dramcom)); + memset(&s->dramctl, 0, sizeof(s->dramctl)); + memset(&s->dramphy, 0, sizeof(s->dramphy)); +} + +static void allwinner_h3_dramc_realize(DeviceState *dev, Error **errp) +{ + AwH3DramCtlState *s = AW_H3_DRAMC(dev); + + /* Only power of 2 RAM sizes from 256MiB up to 2048MiB are supported */ + for (uint8_t i = 8; i < 13; i++) { + if (1 << i == s->ram_size) { + break; + } else if (i == 12) { + error_report("%s: ram-size %u MiB is not supported", + __func__, s->ram_size); + exit(1); + } + } + + /* Setup row mirror mappings */ + memory_region_init_ram(&s->row_mirror, OBJECT(s), + "allwinner-h3-dramc.row-mirror", + 4 * KiB, &error_abort); + memory_region_add_subregion_overlap(get_system_memory(), s->ram_addr, + &s->row_mirror, 10); + + memory_region_init_alias(&s->row_mirror_alias, OBJECT(s), + "allwinner-h3-dramc.row-mirror-alias", + &s->row_mirror, 0, 4 * KiB); + memory_region_add_subregion_overlap(get_system_memory(), + s->ram_addr + 1 * MiB, + &s->row_mirror_alias, 10); + memory_region_set_enabled(&s->row_mirror_alias, false); +} + +static void allwinner_h3_dramc_init(Object *obj) +{ + SysBusDevice *sbd = SYS_BUS_DEVICE(obj); + AwH3DramCtlState *s = AW_H3_DRAMC(obj); + + /* DRAMCOM registers */ + memory_region_init_io(&s->dramcom_iomem, OBJECT(s), + &allwinner_h3_dramcom_ops, s, + TYPE_AW_H3_DRAMC, 4 * KiB); + sysbus_init_mmio(sbd, &s->dramcom_iomem); + + /* DRAMCTL registers */ + memory_region_init_io(&s->dramctl_iomem, OBJECT(s), + &allwinner_h3_dramctl_ops, s, + TYPE_AW_H3_DRAMC, 4 * KiB); + sysbus_init_mmio(sbd, &s->dramctl_iomem); + + /* DRAMPHY registers */ + memory_region_init_io(&s->dramphy_iomem, OBJECT(s), + &allwinner_h3_dramphy_ops, s, + TYPE_AW_H3_DRAMC, 4 * KiB); + sysbus_init_mmio(sbd, &s->dramphy_iomem); +} + +static Property allwinner_h3_dramc_properties[] = { + DEFINE_PROP_UINT64("ram-addr", AwH3DramCtlState, ram_addr, 0x0), + DEFINE_PROP_UINT32("ram-size", AwH3DramCtlState, ram_size, 256 * MiB), + DEFINE_PROP_END_OF_LIST() +}; + +static const VMStateDescription allwinner_h3_dramc_vmstate = { + .name = "allwinner-h3-dramc", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32_ARRAY(dramcom, AwH3DramCtlState, AW_H3_DRAMCOM_REGS_NUM), + VMSTATE_UINT32_ARRAY(dramctl, AwH3DramCtlState, AW_H3_DRAMCTL_REGS_NUM), + VMSTATE_UINT32_ARRAY(dramphy, AwH3DramCtlState, AW_H3_DRAMPHY_REGS_NUM), + VMSTATE_END_OF_LIST() + } +}; + +static void allwinner_h3_dramc_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->reset = allwinner_h3_dramc_reset; + dc->vmsd = &allwinner_h3_dramc_vmstate; + dc->realize = allwinner_h3_dramc_realize; + device_class_set_props(dc, allwinner_h3_dramc_properties); +} + +static const TypeInfo allwinner_h3_dramc_info = { + .name = TYPE_AW_H3_DRAMC, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_init = allwinner_h3_dramc_init, + .instance_size = sizeof(AwH3DramCtlState), + .class_init = allwinner_h3_dramc_class_init, +}; + +static void allwinner_h3_dramc_register(void) +{ + type_register_static(&allwinner_h3_dramc_info); +} + +type_init(allwinner_h3_dramc_register) diff --git a/hw/misc/trace-events b/hw/misc/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/trace-events +++ b/hw/misc/trace-events @@ -XXX,XX +XXX,XX @@ allwinner_cpucfg_cpu_reset(uint8_t cpu_id, uint32_t reset_addr) "id %u, reset_ad allwinner_cpucfg_read(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 allwinner_cpucfg_write(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 +# allwinner-h3-dramc.c +allwinner_h3_dramc_rowmirror_disable(void) "Disable row mirror" +allwinner_h3_dramc_rowmirror_enable(uint64_t addr) "Enable row mirror: addr 0x%" PRIx64 +allwinner_h3_dramcom_read(uint64_t offset, uint64_t data, unsigned size) "Read: offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 +allwinner_h3_dramcom_write(uint64_t offset, uint64_t data, unsigned size) "Write: offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 +allwinner_h3_dramctl_read(uint64_t offset, uint64_t data, unsigned size) "Read: offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 +allwinner_h3_dramctl_write(uint64_t offset, uint64_t data, unsigned size) "Write: offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 +allwinner_h3_dramphy_read(uint64_t offset, uint64_t data, unsigned size) "Read: offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 +allwinner_h3_dramphy_write(uint64_t offset, uint64_t data, unsigned size) "write: offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 + # allwinner-sid.c allwinner_sid_read(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 allwinner_sid_write(uint64_t offset, uint64_t data, unsigned size) "offset 0x%" PRIx64 " data 0x%" PRIx64 " size %" PRIu32 -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> Allwinner System-on-Chips usually contain a Real Time Clock (RTC) for non-volatile system date and time keeping. This commit adds a generic Allwinner RTC device that supports the RTC devices found in Allwinner SoC family sun4i (A10), sun7i (A20) and sun6i and newer (A31, H2+, H3, etc). The following RTC functionality and features are implemented: * Year-Month-Day read/write * Hour-Minute-Second read/write * General Purpose storage The following boards are extended with the RTC device: * Cubieboard (hw/arm/cubieboard.c) * Orange Pi PC (hw/arm/orangepi.c) Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20200311221854.30370-13-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/rtc/Makefile.objs | 1 + include/hw/arm/allwinner-a10.h | 2 + include/hw/arm/allwinner-h3.h | 3 + include/hw/rtc/allwinner-rtc.h | 134 +++++++++++ hw/arm/allwinner-a10.c | 8 + hw/arm/allwinner-h3.c | 9 +- hw/rtc/allwinner-rtc.c | 411 +++++++++++++++++++++++++++++++++ hw/rtc/trace-events | 4 + 8 files changed, 571 insertions(+), 1 deletion(-) create mode 100644 include/hw/rtc/allwinner-rtc.h create mode 100644 hw/rtc/allwinner-rtc.c diff --git a/hw/rtc/Makefile.objs b/hw/rtc/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/rtc/Makefile.objs +++ b/hw/rtc/Makefile.objs @@ -XXX,XX +XXX,XX @@ obj-$(CONFIG_MC146818RTC) += mc146818rtc.o common-obj-$(CONFIG_SUN4V_RTC) += sun4v-rtc.o common-obj-$(CONFIG_ASPEED_SOC) += aspeed_rtc.o common-obj-$(CONFIG_GOLDFISH_RTC) += goldfish_rtc.o +common-obj-$(CONFIG_ALLWINNER_H3) += allwinner-rtc.o diff --git a/include/hw/arm/allwinner-a10.h b/include/hw/arm/allwinner-a10.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-a10.h +++ b/include/hw/arm/allwinner-a10.h @@ -XXX,XX +XXX,XX @@ #include "hw/ide/ahci.h" #include "hw/usb/hcd-ohci.h" #include "hw/usb/hcd-ehci.h" +#include "hw/rtc/allwinner-rtc.h" #include "target/arm/cpu.h" @@ -XXX,XX +XXX,XX @@ typedef struct AwA10State { AwEmacState emac; AllwinnerAHCIState sata; AwSdHostState mmc0; + AwRtcState rtc; MemoryRegion sram_a; EHCISysBusState ehci[AW_A10_NUM_USB]; OHCISysBusState ohci[AW_A10_NUM_USB]; diff --git a/include/hw/arm/allwinner-h3.h b/include/hw/arm/allwinner-h3.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/allwinner-h3.h +++ b/include/hw/arm/allwinner-h3.h @@ -XXX,XX +XXX,XX @@ #include "hw/misc/allwinner-sid.h" #include "hw/sd/allwinner-sdhost.h" #include "hw/net/allwinner-sun8i-emac.h" +#include "hw/rtc/allwinner-rtc.h" #include "target/arm/cpu.h" #include "sysemu/block-backend.h" @@ -XXX,XX +XXX,XX @@ enum { AW_H3_GIC_CPU, AW_H3_GIC_HYP, AW_H3_GIC_VCPU, + AW_H3_RTC, AW_H3_CPUCFG, AW_H3_SDRAM }; @@ -XXX,XX +XXX,XX @@ typedef struct AwH3State { AwSidState sid; AwSdHostState mmc0; AwSun8iEmacState emac; + AwRtcState rtc; GICState gic; MemoryRegion sram_a1; MemoryRegion sram_a2; diff --git a/include/hw/rtc/allwinner-rtc.h b/include/hw/rtc/allwinner-rtc.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/rtc/allwinner-rtc.h @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner Real Time Clock emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef HW_MISC_ALLWINNER_RTC_H +#define HW_MISC_ALLWINNER_RTC_H + +#include "qom/object.h" +#include "hw/sysbus.h" + +/** + * Constants + * @{ + */ + +/** Highest register address used by RTC device */ +#define AW_RTC_REGS_MAXADDR (0x200) + +/** Total number of known registers */ +#define AW_RTC_REGS_NUM (AW_RTC_REGS_MAXADDR / sizeof(uint32_t)) + +/** @} */ + +/** + * Object model types + * @{ + */ + +/** Generic Allwinner RTC device (abstract) */ +#define TYPE_AW_RTC "allwinner-rtc" + +/** Allwinner RTC sun4i family (A10, A12) */ +#define TYPE_AW_RTC_SUN4I TYPE_AW_RTC "-sun4i" + +/** Allwinner RTC sun6i family and newer (A31, H2+, H3, etc) */ +#define TYPE_AW_RTC_SUN6I TYPE_AW_RTC "-sun6i" + +/** Allwinner RTC sun7i family (A20) */ +#define TYPE_AW_RTC_SUN7I TYPE_AW_RTC "-sun7i" + +/** @} */ + +/** + * Object model macros + * @{ + */ + +#define AW_RTC(obj) \ + OBJECT_CHECK(AwRtcState, (obj), TYPE_AW_RTC) +#define AW_RTC_CLASS(klass) \ + OBJECT_CLASS_CHECK(AwRtcClass, (klass), TYPE_AW_RTC) +#define AW_RTC_GET_CLASS(obj) \ + OBJECT_GET_CLASS(AwRtcClass, (obj), TYPE_AW_RTC) + +/** @} */ + +/** + * Allwinner RTC per-object instance state. + */ +typedef struct AwRtcState { + /*< private >*/ + SysBusDevice parent_obj; + /*< public >*/ + + /** + * Actual year represented by the device when year counter is zero + * + * Can be overridden by the user using the corresponding 'base-year' + * property. The base year used by the target OS driver can vary, for + * example the Linux driver for sun6i uses 1970 while NetBSD uses 2000. + */ + int base_year; + + /** Maps I/O registers in physical memory */ + MemoryRegion iomem; + + /** Array of hardware registers */ + uint32_t regs[AW_RTC_REGS_NUM]; + +} AwRtcState; + +/** + * Allwinner RTC class-level struct. + * + * This struct is filled by each sunxi device specific code + * such that the generic code can use this struct to support + * all devices. + */ +typedef struct AwRtcClass { + /*< private >*/ + SysBusDeviceClass parent_class; + /*< public >*/ + + /** Defines device specific register map */ + const uint8_t *regmap; + + /** Size of the regmap in bytes */ + size_t regmap_size; + + /** + * Read device specific register + * + * @offset: register offset to read + * @return true if register read successful, false otherwise + */ + bool (*read)(AwRtcState *s, uint32_t offset); + + /** + * Write device specific register + * + * @offset: register offset to write + * @data: value to set in register + * @return true if register write successful, false otherwise + */ + bool (*write)(AwRtcState *s, uint32_t offset, uint32_t data); + +} AwRtcClass; + +#endif /* HW_MISC_ALLWINNER_RTC_H */ diff --git a/hw/arm/allwinner-a10.c b/hw/arm/allwinner-a10.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/allwinner-a10.c +++ b/hw/arm/allwinner-a10.c @@ -XXX,XX +XXX,XX @@ #define AW_A10_EHCI_BASE 0x01c14000 #define AW_A10_OHCI_BASE 0x01c14400 #define AW_A10_SATA_BASE 0x01c18000 +#define AW_A10_RTC_BASE 0x01c20d00 static void aw_a10_init(Object *obj) { @@ -XXX,XX +XXX,XX @@ static void aw_a10_init(Object *obj) sysbus_init_child_obj(obj, "mmc0", &s->mmc0, sizeof(s->mmc0), TYPE_AW_SDHOST_SUN4I); + + sysbus_init_child_obj(obj, "rtc", &s->rtc, sizeof(s->rtc), + TYPE_AW_RTC_SUN4I); } static void aw_a10_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void aw_a10_realize(DeviceState *dev, Error **errp) sysbus_connect_irq(SYS_BUS_DEVICE(&s->mmc0), 0, qdev_get_gpio_in(dev, 32)); object_property_add_alias(OBJECT(s), "sd-bus", OBJECT(&s->mmc0), "sd-bus", &error_abort); + + /* RTC */ + qdev_init_nofail(DEVICE(&s->rtc)); + sysbus_mmio_map_overlap(SYS_BUS_DEVICE(&s->rtc), 0, AW_A10_RTC_BASE, 10); } static void aw_a10_class_init(ObjectClass *oc, void *data) diff --git a/hw/arm/allwinner-h3.c b/hw/arm/allwinner-h3.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/allwinner-h3.c +++ b/hw/arm/allwinner-h3.c @@ -XXX,XX +XXX,XX @@ const hwaddr allwinner_h3_memmap[] = { [AW_H3_GIC_CPU] = 0x01c82000, [AW_H3_GIC_HYP] = 0x01c84000, [AW_H3_GIC_VCPU] = 0x01c86000, + [AW_H3_RTC] = 0x01f00000, [AW_H3_CPUCFG] = 0x01f01c00, [AW_H3_SDRAM] = 0x40000000 }; @@ -XXX,XX +XXX,XX @@ struct AwH3Unimplemented { { "csi", 0x01cb0000, 320 * KiB }, { "tve", 0x01e00000, 64 * KiB }, { "hdmi", 0x01ee0000, 128 * KiB }, - { "rtc", 0x01f00000, 1 * KiB }, { "r_timer", 0x01f00800, 1 * KiB }, { "r_intc", 0x01f00c00, 1 * KiB }, { "r_wdog", 0x01f01000, 1 * KiB }, @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_init(Object *obj) "ram-addr", &error_abort); object_property_add_alias(obj, "ram-size", OBJECT(&s->dramc), "ram-size", &error_abort); + + sysbus_init_child_obj(obj, "rtc", &s->rtc, sizeof(s->rtc), + TYPE_AW_RTC_SUN6I); } static void allwinner_h3_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void allwinner_h3_realize(DeviceState *dev, Error **errp) sysbus_mmio_map(SYS_BUS_DEVICE(&s->dramc), 1, s->memmap[AW_H3_DRAMCTL]); sysbus_mmio_map(SYS_BUS_DEVICE(&s->dramc), 2, s->memmap[AW_H3_DRAMPHY]); + /* RTC */ + qdev_init_nofail(DEVICE(&s->rtc)); + sysbus_mmio_map(SYS_BUS_DEVICE(&s->rtc), 0, s->memmap[AW_H3_RTC]); + /* Unimplemented devices */ for (i = 0; i < ARRAY_SIZE(unimplemented); i++) { create_unimplemented_device(unimplemented[i].device_name, diff --git a/hw/rtc/allwinner-rtc.c b/hw/rtc/allwinner-rtc.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/rtc/allwinner-rtc.c @@ -XXX,XX +XXX,XX @@ +/* + * Allwinner Real Time Clock emulation + * + * Copyright (C) 2019 Niek Linnenbank <nieklinnenbank@gmail.com> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu/units.h" +#include "hw/sysbus.h" +#include "migration/vmstate.h" +#include "qemu/log.h" +#include "qemu/module.h" +#include "qemu-common.h" +#include "hw/qdev-properties.h" +#include "hw/rtc/allwinner-rtc.h" +#include "trace.h" + +/* RTC registers */ +enum { + REG_LOSC = 1, /* Low Oscillator Control */ + REG_YYMMDD, /* RTC Year-Month-Day */ + REG_HHMMSS, /* RTC Hour-Minute-Second */ + REG_ALARM1_WKHHMMSS, /* Alarm1 Week Hour-Minute-Second */ + REG_ALARM1_EN, /* Alarm1 Enable */ + REG_ALARM1_IRQ_EN, /* Alarm1 IRQ Enable */ + REG_ALARM1_IRQ_STA, /* Alarm1 IRQ Status */ + REG_GP0, /* General Purpose Register 0 */ + REG_GP1, /* General Purpose Register 1 */ + REG_GP2, /* General Purpose Register 2 */ + REG_GP3, /* General Purpose Register 3 */ + + /* sun4i registers */ + REG_ALARM1_DDHHMMSS, /* Alarm1 Day Hour-Minute-Second */ + REG_CPUCFG, /* CPU Configuration Register */ + + /* sun6i registers */ + REG_LOSC_AUTOSTA, /* LOSC Auto Switch Status */ + REG_INT_OSC_PRE, /* Internal OSC Clock Prescaler */ + REG_ALARM0_COUNTER, /* Alarm0 Counter */ + REG_ALARM0_CUR_VLU, /* Alarm0 Counter Current Value */ + REG_ALARM0_ENABLE, /* Alarm0 Enable */ + REG_ALARM0_IRQ_EN, /* Alarm0 IRQ Enable */ + REG_ALARM0_IRQ_STA, /* Alarm0 IRQ Status */ + REG_ALARM_CONFIG, /* Alarm Config */ + REG_LOSC_OUT_GATING, /* LOSC Output Gating Register */ + REG_GP4, /* General Purpose Register 4 */ + REG_GP5, /* General Purpose Register 5 */ + REG_GP6, /* General Purpose Register 6 */ + REG_GP7, /* General Purpose Register 7 */ + REG_RTC_DBG, /* RTC Debug Register */ + REG_GPL_HOLD_OUT, /* GPL Hold Output Register */ + REG_VDD_RTC, /* VDD RTC Regulate Register */ + REG_IC_CHARA, /* IC Characteristics Register */ +}; + +/* RTC register flags */ +enum { + REG_LOSC_YMD = (1 << 7), + REG_LOSC_HMS = (1 << 8), +}; + +/* RTC sun4i register map (offset to name) */ +const uint8_t allwinner_rtc_sun4i_regmap[] = { + [0x0000] = REG_LOSC, + [0x0004] = REG_YYMMDD, + [0x0008] = REG_HHMMSS, + [0x000C] = REG_ALARM1_DDHHMMSS, + [0x0010] = REG_ALARM1_WKHHMMSS, + [0x0014] = REG_ALARM1_EN, + [0x0018] = REG_ALARM1_IRQ_EN, + [0x001C] = REG_ALARM1_IRQ_STA, + [0x0020] = REG_GP0, + [0x0024] = REG_GP1, + [0x0028] = REG_GP2, + [0x002C] = REG_GP3, + [0x003C] = REG_CPUCFG, +}; + +/* RTC sun6i register map (offset to name) */ +const uint8_t allwinner_rtc_sun6i_regmap[] = { + [0x0000] = REG_LOSC, + [0x0004] = REG_LOSC_AUTOSTA, + [0x0008] = REG_INT_OSC_PRE, + [0x0010] = REG_YYMMDD, + [0x0014] = REG_HHMMSS, + [0x0020] = REG_ALARM0_COUNTER, + [0x0024] = REG_ALARM0_CUR_VLU, + [0x0028] = REG_ALARM0_ENABLE, + [0x002C] = REG_ALARM0_IRQ_EN, + [0x0030] = REG_ALARM0_IRQ_STA, + [0x0040] = REG_ALARM1_WKHHMMSS, + [0x0044] = REG_ALARM1_EN, + [0x0048] = REG_ALARM1_IRQ_EN, + [0x004C] = REG_ALARM1_IRQ_STA, + [0x0050] = REG_ALARM_CONFIG, + [0x0060] = REG_LOSC_OUT_GATING, + [0x0100] = REG_GP0, + [0x0104] = REG_GP1, + [0x0108] = REG_GP2, + [0x010C] = REG_GP3, + [0x0110] = REG_GP4, + [0x0114] = REG_GP5, + [0x0118] = REG_GP6, + [0x011C] = REG_GP7, + [0x0170] = REG_RTC_DBG, + [0x0180] = REG_GPL_HOLD_OUT, + [0x0190] = REG_VDD_RTC, + [0x01F0] = REG_IC_CHARA, +}; + +static bool allwinner_rtc_sun4i_read(AwRtcState *s, uint32_t offset) +{ + /* no sun4i specific registers currently implemented */ + return false; +} + +static bool allwinner_rtc_sun4i_write(AwRtcState *s, uint32_t offset, + uint32_t data) +{ + /* no sun4i specific registers currently implemented */ + return false; +} + +static bool allwinner_rtc_sun6i_read(AwRtcState *s, uint32_t offset) +{ + const AwRtcClass *c = AW_RTC_GET_CLASS(s); + + switch (c->regmap[offset]) { + case REG_GP4: /* General Purpose Register 4 */ + case REG_GP5: /* General Purpose Register 5 */ + case REG_GP6: /* General Purpose Register 6 */ + case REG_GP7: /* General Purpose Register 7 */ + return true; + default: + break; + } + return false; +} + +static bool allwinner_rtc_sun6i_write(AwRtcState *s, uint32_t offset, + uint32_t data) +{ + const AwRtcClass *c = AW_RTC_GET_CLASS(s); + + switch (c->regmap[offset]) { + case REG_GP4: /* General Purpose Register 4 */ + case REG_GP5: /* General Purpose Register 5 */ + case REG_GP6: /* General Purpose Register 6 */ + case REG_GP7: /* General Purpose Register 7 */ + return true; + default: + break; + } + return false; +} + +static uint64_t allwinner_rtc_read(void *opaque, hwaddr offset, + unsigned size) +{ + AwRtcState *s = AW_RTC(opaque); + const AwRtcClass *c = AW_RTC_GET_CLASS(s); + uint64_t val = 0; + + if (offset >= c->regmap_size) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + return 0; + } + + if (!c->regmap[offset]) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: invalid register 0x%04x\n", + __func__, (uint32_t)offset); + return 0; + } + + switch (c->regmap[offset]) { + case REG_LOSC: /* Low Oscillator Control */ + val = s->regs[REG_LOSC]; + s->regs[REG_LOSC] &= ~(REG_LOSC_YMD | REG_LOSC_HMS); + break; + case REG_YYMMDD: /* RTC Year-Month-Day */ + case REG_HHMMSS: /* RTC Hour-Minute-Second */ + case REG_GP0: /* General Purpose Register 0 */ + case REG_GP1: /* General Purpose Register 1 */ + case REG_GP2: /* General Purpose Register 2 */ + case REG_GP3: /* General Purpose Register 3 */ + val = s->regs[c->regmap[offset]]; + break; + default: + if (!c->read(s, offset)) { + qemu_log_mask(LOG_UNIMP, "%s: unimplemented register 0x%04x\n", + __func__, (uint32_t)offset); + } + val = s->regs[c->regmap[offset]]; + break; + } + + trace_allwinner_rtc_read(offset, val); + return val; +} + +static void allwinner_rtc_write(void *opaque, hwaddr offset, + uint64_t val, unsigned size) +{ + AwRtcState *s = AW_RTC(opaque); + const AwRtcClass *c = AW_RTC_GET_CLASS(s); + + if (offset >= c->regmap_size) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: out-of-bounds offset 0x%04x\n", + __func__, (uint32_t)offset); + return; + } + + if (!c->regmap[offset]) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: invalid register 0x%04x\n", + __func__, (uint32_t)offset); + return; + } + + trace_allwinner_rtc_write(offset, val); + + switch (c->regmap[offset]) { + case REG_YYMMDD: /* RTC Year-Month-Day */ + s->regs[REG_YYMMDD] = val; + s->regs[REG_LOSC] |= REG_LOSC_YMD; + break; + case REG_HHMMSS: /* RTC Hour-Minute-Second */ + s->regs[REG_HHMMSS] = val; + s->regs[REG_LOSC] |= REG_LOSC_HMS; + break; + case REG_GP0: /* General Purpose Register 0 */ + case REG_GP1: /* General Purpose Register 1 */ + case REG_GP2: /* General Purpose Register 2 */ + case REG_GP3: /* General Purpose Register 3 */ + s->regs[c->regmap[offset]] = val; + break; + default: + if (!c->write(s, offset, val)) { + qemu_log_mask(LOG_UNIMP, "%s: unimplemented register 0x%04x\n", + __func__, (uint32_t)offset); + } + break; + } +} + +static const MemoryRegionOps allwinner_rtc_ops = { + .read = allwinner_rtc_read, + .write = allwinner_rtc_write, + .endianness = DEVICE_NATIVE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, + .impl.min_access_size = 4, +}; + +static void allwinner_rtc_reset(DeviceState *dev) +{ + AwRtcState *s = AW_RTC(dev); + struct tm now; + + /* Clear registers */ + memset(s->regs, 0, sizeof(s->regs)); + + /* Get current datetime */ + qemu_get_timedate(&now, 0); + + /* Set RTC with current datetime */ + if (s->base_year > 1900) { + s->regs[REG_YYMMDD] = ((now.tm_year + 1900 - s->base_year) << 16) | + ((now.tm_mon + 1) << 8) | + now.tm_mday; + s->regs[REG_HHMMSS] = (((now.tm_wday + 6) % 7) << 29) | + (now.tm_hour << 16) | + (now.tm_min << 8) | + now.tm_sec; + } +} + +static void allwinner_rtc_init(Object *obj) +{ + SysBusDevice *sbd = SYS_BUS_DEVICE(obj); + AwRtcState *s = AW_RTC(obj); + + /* Memory mapping */ + memory_region_init_io(&s->iomem, OBJECT(s), &allwinner_rtc_ops, s, + TYPE_AW_RTC, 1 * KiB); + sysbus_init_mmio(sbd, &s->iomem); +} + +static const VMStateDescription allwinner_rtc_vmstate = { + .name = "allwinner-rtc", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32_ARRAY(regs, AwRtcState, AW_RTC_REGS_NUM), + VMSTATE_END_OF_LIST() + } +}; + +static Property allwinner_rtc_properties[] = { + DEFINE_PROP_INT32("base-year", AwRtcState, base_year, 0), + DEFINE_PROP_END_OF_LIST(), +}; + +static void allwinner_rtc_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->reset = allwinner_rtc_reset; + dc->vmsd = &allwinner_rtc_vmstate; + device_class_set_props(dc, allwinner_rtc_properties); +} + +static void allwinner_rtc_sun4i_init(Object *obj) +{ + AwRtcState *s = AW_RTC(obj); + s->base_year = 2010; +} + +static void allwinner_rtc_sun4i_class_init(ObjectClass *klass, void *data) +{ + AwRtcClass *arc = AW_RTC_CLASS(klass); + + arc->regmap = allwinner_rtc_sun4i_regmap; + arc->regmap_size = sizeof(allwinner_rtc_sun4i_regmap); + arc->read = allwinner_rtc_sun4i_read; + arc->write = allwinner_rtc_sun4i_write; +} + +static void allwinner_rtc_sun6i_init(Object *obj) +{ + AwRtcState *s = AW_RTC(obj); + s->base_year = 1970; +} + +static void allwinner_rtc_sun6i_class_init(ObjectClass *klass, void *data) +{ + AwRtcClass *arc = AW_RTC_CLASS(klass); + + arc->regmap = allwinner_rtc_sun6i_regmap; + arc->regmap_size = sizeof(allwinner_rtc_sun6i_regmap); + arc->read = allwinner_rtc_sun6i_read; + arc->write = allwinner_rtc_sun6i_write; +} + +static void allwinner_rtc_sun7i_init(Object *obj) +{ + AwRtcState *s = AW_RTC(obj); + s->base_year = 1970; +} + +static void allwinner_rtc_sun7i_class_init(ObjectClass *klass, void *data) +{ + AwRtcClass *arc = AW_RTC_CLASS(klass); + allwinner_rtc_sun4i_class_init(klass, arc); +} + +static const TypeInfo allwinner_rtc_info = { + .name = TYPE_AW_RTC, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_init = allwinner_rtc_init, + .instance_size = sizeof(AwRtcState), + .class_init = allwinner_rtc_class_init, + .class_size = sizeof(AwRtcClass), + .abstract = true, +}; + +static const TypeInfo allwinner_rtc_sun4i_info = { + .name = TYPE_AW_RTC_SUN4I, + .parent = TYPE_AW_RTC, + .class_init = allwinner_rtc_sun4i_class_init, + .instance_init = allwinner_rtc_sun4i_init, +}; + +static const TypeInfo allwinner_rtc_sun6i_info = { + .name = TYPE_AW_RTC_SUN6I, + .parent = TYPE_AW_RTC, + .class_init = allwinner_rtc_sun6i_class_init, + .instance_init = allwinner_rtc_sun6i_init, +}; + +static const TypeInfo allwinner_rtc_sun7i_info = { + .name = TYPE_AW_RTC_SUN7I, + .parent = TYPE_AW_RTC, + .class_init = allwinner_rtc_sun7i_class_init, + .instance_init = allwinner_rtc_sun7i_init, +}; + +static void allwinner_rtc_register(void) +{ + type_register_static(&allwinner_rtc_info); + type_register_static(&allwinner_rtc_sun4i_info); + type_register_static(&allwinner_rtc_sun6i_info); + type_register_static(&allwinner_rtc_sun7i_info); +} + +type_init(allwinner_rtc_register) diff --git a/hw/rtc/trace-events b/hw/rtc/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/rtc/trace-events +++ b/hw/rtc/trace-events @@ -XXX,XX +XXX,XX @@ # See docs/devel/tracing.txt for syntax documentation. +# allwinner-rtc.c +allwinner_rtc_read(uint64_t addr, uint64_t value) "addr 0x%" PRIx64 " value 0x%" PRIx64 +allwinner_rtc_write(uint64_t addr, uint64_t value) "addr 0x%" PRIx64 " value 0x%" PRIx64 + # sun4v-rtc.c sun4v_rtc_read(uint64_t addr, uint64_t value) "read: addr 0x%" PRIx64 " value 0x%" PRIx64 sun4v_rtc_write(uint64_t addr, uint64_t value) "write: addr 0x%" PRIx64 " value 0x%" PRIx64 -- 2.20.1 From: Philippe Mathieu-Daudé <f4bug@amsat.org> This test boots a Linux kernel on a OrangePi PC board and verify the serial output is working. The kernel image and DeviceTree blob are built by the Armbian project (based on Debian): https://www.armbian.com/orange-pi-pc/ If ARM is a target being built, "make check-acceptance" will automatically include this test by the use of the "arch:arm" tags. Alternatively, this test can be run using: $ make check-venv $ ./tests/venv/bin/avocado --show=console,app run -t machine:orangepi-pc tests/acceptance/boot_linux_console.py JOB ID : 2e4d15eceb13c33672af406f08171e6e9de1414a JOB LOG : ~/job-results/job-2019-12-17T05.46-2e4d15e/job.log (1/1) tests/acceptance/boot_linux_console.py:BootLinuxConsole.test_arm_orangepi: console: Uncompressing Linux... done, booting the kernel. console: Booting Linux on physical CPU 0x0 console: Linux version 4.20.7-sunxi (root@armbian.com) (gcc version 7.2.1 20171011 (Linaro GCC 7.2-2017.11)) #5.75 SMP Fri Feb 8 09:02:10 CET 2019 console: CPU: ARMv7 Processor [410fc075] revision 5 (ARMv7), cr=50c5387d console: CPU: div instructions available: patching division code console: CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache console: OF: fdt: Machine model: Xunlong Orange Pi PC console: Memory policy: Data cache writealloc console: OF: reserved mem: failed to allocate memory for node 'cma@4a000000' console: cma: Failed to reserve 128 MiB console: psci: probing for conduit method from DT. console: psci: PSCIv0.2 detected in firmware. console: psci: Using standard PSCI v0.2 function IDs console: psci: Trusted OS migration not required console: random: get_random_bytes called from start_kernel+0x8d/0x3c2 with crng_init=0 console: percpu: Embedded 18 pages/cpu @(ptrval) s41228 r8192 d24308 u73728 console: Built 1 zonelists, mobility grouping on. Total pages: 32480 console: Kernel command line: printk.time=0 console=ttyS0,115200 PASS (8.59 s) JOB TIME : 8.81 s Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20200311221854.30370-14-nieklinnenbank@gmail.com [NL: rename in commit message Raspbian to Armbian, remove vm.set_machine()] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/acceptance/boot_linux_console.py | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) diff --git a/tests/acceptance/boot_linux_console.py b/tests/acceptance/boot_linux_console.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/boot_linux_console.py +++ b/tests/acceptance/boot_linux_console.py @@ -XXX,XX +XXX,XX @@ class BootLinuxConsole(Test): exec_command_and_wait_for_pattern(self, 'reboot', 'reboot: Restarting system') + def test_arm_orangepi(self): + """ + :avocado: tags=arch:arm + :avocado: tags=machine:orangepi-pc + """ + deb_url = ('https://apt.armbian.com/pool/main/l/' + 'linux-4.20.7-sunxi/linux-image-dev-sunxi_5.75_armhf.deb') + deb_hash = '1334c29c44d984ffa05ed10de8c3361f33d78315' + deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash) + kernel_path = self.extract_from_deb(deb_path, + '/boot/vmlinuz-4.20.7-sunxi') + dtb_path = '/usr/lib/linux-image-dev-sunxi/sun8i-h3-orangepi-pc.dtb' + dtb_path = self.extract_from_deb(deb_path, dtb_path) + + self.vm.set_console() + kernel_command_line = (self.KERNEL_COMMON_COMMAND_LINE + + 'console=ttyS0,115200n8 ' + 'earlycon=uart,mmio32,0x1c28000') + self.vm.add_args('-kernel', kernel_path, + '-dtb', dtb_path, + '-append', kernel_command_line) + self.vm.launch() + console_pattern = 'Kernel command line: %s' % kernel_command_line + self.wait_for_console_pattern(console_pattern) + def test_s390x_s390_ccw_virtio(self): """ :avocado: tags=arch:s390x -- 2.20.1 From: Philippe Mathieu-Daudé <f4bug@amsat.org> This test boots a Linux kernel on a OrangePi PC board and verify the serial output is working. The kernel image and DeviceTree blob are built by the Armbian project (based on Debian): https://www.armbian.com/orange-pi-pc/ The cpio image used comes from the linux-build-test project: https://github.com/groeck/linux-build-test If ARM is a target being built, "make check-acceptance" will automatically include this test by the use of the "arch:arm" tags. Alternatively, this test can be run using: $ avocado --show=console run -t machine:orangepi-pc tests/acceptance/boot_linux_console.py console: Uncompressing Linux... done, booting the kernel. console: Booting Linux on physical CPU 0x0 console: Linux version 4.20.7-sunxi (root@armbian.com) (gcc version 7.2.1 20171011 (Linaro GCC 7.2-2017.11)) #5.75 SMP Fri Feb 8 09:02:10 CET 2019 console: CPU: ARMv7 Processor [410fc075] revision 5 (ARMv7), cr=50c5387d console: CPU: div instructions available: patching division code console: CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache console: OF: fdt: Machine model: Xunlong Orange Pi PC [...] console: Trying to unpack rootfs image as initramfs... console: Freeing initrd memory: 3256K console: Freeing unused kernel memory: 1024K console: Run /init as init process console: mount: mounting devtmpfs on /dev failed: Device or resource busy console: Starting logging: OK console: Initializing random number generator... random: dd: uninitialized urandom read (512 bytes read) console: done. console: Starting network: OK console: Found console ttyS0 console: Linux version 4.20.7-sunxi (root@armbian.com) (gcc version 7.2.1 20171011 (Linaro GCC 7.2-2017.11)) #5.75 SMP Fri Feb 8 09:02:10 CET 2019 console: Boot successful. console: cat /proc/cpuinfo console: / # cat /proc/cpuinfo console: processor : 0 console: model name : ARMv7 Processor rev 5 (v7l) console: BogoMIPS : 125.00 console: Features : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm console: CPU implementer : 0x41 console: CPU architecture: 7 console: CPU variant : 0x0 console: CPU part : 0xc07 console: CPU revision : 5 [...] console: processor : 3 console: model name : ARMv7 Processor rev 5 (v7l) console: BogoMIPS : 125.00 console: Features : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm console: CPU implementer : 0x41 console: CPU architecture: 7 console: CPU variant : 0x0 console: CPU part : 0xc07 console: CPU revision : 5 console: Hardware : Allwinner sun8i Family console: Revision : 0000 console: Serial : 0000000000000000 console: cat /proc/iomem console: / # cat /proc/iomem console: 01000000-010fffff : clock@1000000 console: 01c00000-01c00fff : system-control@1c00000 console: 01c02000-01c02fff : dma-controller@1c02000 [...] console: reboot console: / # reboot console: / # Found console ttyS0 console: Stopping network: OK console: hrtimer: interrupt took 21852064 ns console: Saving random seed... random: dd: uninitialized urandom read (512 bytes read) console: done. console: Stopping logging: OK console: umount: devtmpfs busy - remounted read-only console: umount: can't unmount /: Invalid argument console: The system is going down NOW! console: Sent SIGTERM to all processes console: Sent SIGKILL to all processes console: Requesting system reboot console: reboot: Restarting system PASS (48.32 s) JOB TIME : 49.16 s Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20200311221854.30370-15-nieklinnenbank@gmail.com [NL: rename in commit message Raspbian to Armbian, remove vm.set_machine()] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/acceptance/boot_linux_console.py | 40 ++++++++++++++++++++++++++ 1 file changed, 40 insertions(+) diff --git a/tests/acceptance/boot_linux_console.py b/tests/acceptance/boot_linux_console.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/boot_linux_console.py +++ b/tests/acceptance/boot_linux_console.py @@ -XXX,XX +XXX,XX @@ class BootLinuxConsole(Test): console_pattern = 'Kernel command line: %s' % kernel_command_line self.wait_for_console_pattern(console_pattern) + def test_arm_orangepi_initrd(self): + """ + :avocado: tags=arch:arm + :avocado: tags=machine:orangepi-pc + """ + deb_url = ('https://apt.armbian.com/pool/main/l/' + 'linux-4.20.7-sunxi/linux-image-dev-sunxi_5.75_armhf.deb') + deb_hash = '1334c29c44d984ffa05ed10de8c3361f33d78315' + deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash) + kernel_path = self.extract_from_deb(deb_path, + '/boot/vmlinuz-4.20.7-sunxi') + dtb_path = '/usr/lib/linux-image-dev-sunxi/sun8i-h3-orangepi-pc.dtb' + dtb_path = self.extract_from_deb(deb_path, dtb_path) + initrd_url = ('https://github.com/groeck/linux-build-test/raw/' + '2eb0a73b5d5a28df3170c546ddaaa9757e1e0848/rootfs/' + 'arm/rootfs-armv7a.cpio.gz') + initrd_hash = '604b2e45cdf35045846b8bbfbf2129b1891bdc9c' + initrd_path_gz = self.fetch_asset(initrd_url, asset_hash=initrd_hash) + initrd_path = os.path.join(self.workdir, 'rootfs.cpio') + archive.gzip_uncompress(initrd_path_gz, initrd_path) + + self.vm.set_console() + kernel_command_line = (self.KERNEL_COMMON_COMMAND_LINE + + 'console=ttyS0,115200 ' + 'panic=-1 noreboot') + self.vm.add_args('-kernel', kernel_path, + '-dtb', dtb_path, + '-initrd', initrd_path, + '-append', kernel_command_line, + '-no-reboot') + self.vm.launch() + self.wait_for_console_pattern('Boot successful.') + + exec_command_and_wait_for_pattern(self, 'cat /proc/cpuinfo', + 'Allwinner sun8i Family') + exec_command_and_wait_for_pattern(self, 'cat /proc/iomem', + 'system-control@1c00000') + exec_command_and_wait_for_pattern(self, 'reboot', + 'reboot: Restarting system') + def test_s390x_s390_ccw_virtio(self): """ :avocado: tags=arch:s390x -- 2.20.1 From: Philippe Mathieu-Daudé <f4bug@amsat.org> The kernel image and DeviceTree blob are built by the Armbian project (based on Debian): https://www.armbian.com/orange-pi-pc/ The SD image is from the kernelci.org project: https://kernelci.org/faq/#the-code If ARM is a target being built, "make check-acceptance" will automatically include this test by the use of the "arch:arm" tags. Alternatively, this test can be run using: $ avocado --show=console run -t machine:orangepi-pc tests/acceptance/boot_linux_console.py console: Uncompressing Linux... done, booting the kernel. console: Booting Linux on physical CPU 0x0 console: Linux version 4.20.7-sunxi (root@armbian.com) (gcc version 7.2.1 20171011 (Linaro GCC 7.2-2017.11)) #5.75 SMP Fri Feb 8 09:02:10 CET 2019 console: CPU: ARMv7 Processor [410fc075] revision 5 (ARMv7), cr=50c5387d [...] console: sunxi-wdt 1c20ca0.watchdog: Watchdog enabled (timeout=16 sec, nowayout=0) console: sunxi-mmc 1c0f000.mmc: Linked as a consumer to regulator.2 console: sunxi-mmc 1c0f000.mmc: Got CD GPIO console: ledtrig-cpu: registered to indicate activity on CPUs console: hidraw: raw HID events driver (C) Jiri Kosina console: usbcore: registered new interface driver usbhid console: usbhid: USB HID core driver console: Initializing XFRM netlink socket console: sunxi-mmc 1c0f000.mmc: initialized, max. request size: 16384 KB console: NET: Registered protocol family 10 console: mmc0: host does not support reading read-only switch, assuming write-enable console: mmc0: Problem switching card into high-speed mode! console: mmc0: new SD card at address 4567 console: mmcblk0: mmc0:4567 QEMU! 60.0 MiB [...] console: EXT4-fs (mmcblk0): mounting ext2 file system using the ext4 subsystem console: EXT4-fs (mmcblk0): mounted filesystem without journal. Opts: (null) console: VFS: Mounted root (ext2 filesystem) on device 179:0. console: Run /sbin/init as init process console: EXT4-fs (mmcblk0): re-mounted. Opts: block_validity,barrier,user_xattr,acl console: Starting syslogd: OK console: Starting klogd: OK console: Populating /dev using udev: udevd[203]: starting version 3.2.7 console: /bin/sh: can't access tty; job control turned off console: cat /proc/partitions console: / # cat /proc/partitions console: major minor #blocks name console: 1 0 4096 ram0 console: 1 1 4096 ram1 console: 1 2 4096 ram2 console: 1 3 4096 ram3 console: 179 0 61440 mmcblk0 console: reboot console: / # reboot console: umount: devtmpfs busy - remounted read-only console: EXT4-fs (mmcblk0): re-mounted. Opts: (null) console: The system is going down NOW! console: Sent SIGTERM to all processes console: Sent SIGKILL to all processes console: Requesting system reboot console: reboot: Restarting system JOB TIME : 68.64 s Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20200311221854.30370-16-nieklinnenbank@gmail.com [NL: rename in commit message Raspbian to Armbian, remove vm.set_machine()] [NL: extend test with ethernet device checks] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/acceptance/boot_linux_console.py | 47 ++++++++++++++++++++++++++ 1 file changed, 47 insertions(+) diff --git a/tests/acceptance/boot_linux_console.py b/tests/acceptance/boot_linux_console.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/boot_linux_console.py +++ b/tests/acceptance/boot_linux_console.py @@ -XXX,XX +XXX,XX @@ class BootLinuxConsole(Test): exec_command_and_wait_for_pattern(self, 'reboot', 'reboot: Restarting system') + def test_arm_orangepi_sd(self): + """ + :avocado: tags=arch:arm + :avocado: tags=machine:orangepi-pc + """ + deb_url = ('https://apt.armbian.com/pool/main/l/' + 'linux-4.20.7-sunxi/linux-image-dev-sunxi_5.75_armhf.deb') + deb_hash = '1334c29c44d984ffa05ed10de8c3361f33d78315' + deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash) + kernel_path = self.extract_from_deb(deb_path, + '/boot/vmlinuz-4.20.7-sunxi') + dtb_path = '/usr/lib/linux-image-dev-sunxi/sun8i-h3-orangepi-pc.dtb' + dtb_path = self.extract_from_deb(deb_path, dtb_path) + rootfs_url = ('http://storage.kernelci.org/images/rootfs/buildroot/' + 'kci-2019.02/armel/base/rootfs.ext2.xz') + rootfs_hash = '692510cb625efda31640d1de0a8d60e26040f061' + rootfs_path_xz = self.fetch_asset(rootfs_url, asset_hash=rootfs_hash) + rootfs_path = os.path.join(self.workdir, 'rootfs.cpio') + archive.lzma_uncompress(rootfs_path_xz, rootfs_path) + + self.vm.set_console() + kernel_command_line = (self.KERNEL_COMMON_COMMAND_LINE + + 'console=ttyS0,115200 ' + 'root=/dev/mmcblk0 rootwait rw ' + 'panic=-1 noreboot') + self.vm.add_args('-kernel', kernel_path, + '-dtb', dtb_path, + '-drive', 'file=' + rootfs_path + ',if=sd,format=raw', + '-append', kernel_command_line, + '-no-reboot') + self.vm.launch() + shell_ready = "/bin/sh: can't access tty; job control turned off" + self.wait_for_console_pattern(shell_ready) + + exec_command_and_wait_for_pattern(self, 'cat /proc/cpuinfo', + 'Allwinner sun8i Family') + exec_command_and_wait_for_pattern(self, 'cat /proc/partitions', + 'mmcblk0') + exec_command_and_wait_for_pattern(self, 'ifconfig eth0 up', + 'eth0: Link is Up') + exec_command_and_wait_for_pattern(self, 'udhcpc eth0', + 'udhcpc: lease of 10.0.2.15 obtained') + exec_command_and_wait_for_pattern(self, 'ping -c 3 10.0.2.2', + '3 packets transmitted, 3 packets received, 0% packet loss') + exec_command_and_wait_for_pattern(self, 'reboot', + 'reboot: Restarting system') + def test_s390x_s390_ccw_virtio(self): """ :avocado: tags=arch:s390x -- 2.20.1 From: Philippe Mathieu-Daudé <f4bug@amsat.org> This test boots Ubuntu Bionic on a OrangePi PC board. As it requires 1GB of storage, and is slow, this test is disabled on automatic CI testing. It is useful for workstation testing. Currently Avocado timeouts too quickly, so we can't run userland commands. The kernel image and DeviceTree blob are built by the Armbian project (based on Debian): https://www.armbian.com/orange-pi-pc/ The Ubuntu image is downloaded from: https://dl.armbian.com/orangepipc/Bionic_current This test can be run using: $ AVOCADO_ALLOW_LARGE_STORAGE=yes \ avocado --show=app,console run -t machine:orangepi-pc \ tests/acceptance/boot_linux_console.py console: U-Boot SPL 2019.04-armbian (Nov 18 2019 - 23:08:35 +0100) console: DRAM: 1024 MiB console: Failed to set core voltage! Can't set CPU frequency console: Trying to boot from MMC1 console: U-Boot 2019.04-armbian (Nov 18 2019 - 23:08:35 +0100) Allwinner Technology console: CPU: Allwinner H3 (SUN8I 0000) console: Model: Xunlong Orange Pi PC console: DRAM: 1 GiB console: MMC: mmc@1c0f000: 0 [...] console: Uncompressing Linux... done, booting the kernel. console: Booting Linux on physical CPU 0x0 console: Linux version 5.3.9-sunxi (root@builder) (gcc version 8.3.0 (GNU Toolchain for the A-profile Architecture 8.3-2019.03 (arm-rel-8.36))) #19.11.3 SMP Mon Nov 18 18:49:43 CET 2019 console: CPU: ARMv7 Processor [410fc075] revision 5 (ARMv7), cr=50c5387d console: CPU: div instructions available: patching division code console: CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache console: OF: fdt: Machine model: Xunlong Orange Pi PC [...] console: EXT4-fs (mmcblk0p1): mounted filesystem with writeback data mode. Opts: (null) console: done. console: Begin: Running /scripts/local-bottom ... done. console: Begin: Running /scripts/init-bottom ... done. console: systemd[1]: systemd 237 running in system mode. (...) console: systemd[1]: Detected architecture arm. console: Welcome to Ubuntu 18.04.3 LTS! console: systemd[1]: Set hostname to <orangepipc>. Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20200311221854.30370-17-nieklinnenbank@gmail.com [NL: rename in commit message Raspbian to Armbian, remove vm.set_machine()] [NL: changed test to boot from SD card via BootROM, added check for 7z] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/acceptance/boot_linux_console.py | 48 ++++++++++++++++++++++++++ 1 file changed, 48 insertions(+) diff --git a/tests/acceptance/boot_linux_console.py b/tests/acceptance/boot_linux_console.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/boot_linux_console.py +++ b/tests/acceptance/boot_linux_console.py @@ -XXX,XX +XXX,XX @@ from avocado_qemu import exec_command_and_wait_for_pattern from avocado_qemu import wait_for_console_pattern from avocado.utils import process from avocado.utils import archive +from avocado.utils.path import find_command, CmdNotFoundError +P7ZIP_AVAILABLE = True +try: + find_command('7z') +except CmdNotFoundError: + P7ZIP_AVAILABLE = False class BootLinuxConsole(Test): """ @@ -XXX,XX +XXX,XX @@ class BootLinuxConsole(Test): exec_command_and_wait_for_pattern(self, 'reboot', 'reboot: Restarting system') + @skipUnless(os.getenv('AVOCADO_ALLOW_LARGE_STORAGE'), 'storage limited') + @skipUnless(P7ZIP_AVAILABLE, '7z not installed') + def test_arm_orangepi_bionic(self): + """ + :avocado: tags=arch:arm + :avocado: tags=machine:orangepi-pc + """ + + # This test download a 196MB compressed image and expand it to 932MB... + image_url = ('https://dl.armbian.com/orangepipc/archive/' + 'Armbian_19.11.3_Orangepipc_bionic_current_5.3.9.7z') + image_hash = '196a8ffb72b0123d92cea4a070894813d305c71e' + image_path_7z = self.fetch_asset(image_url, asset_hash=image_hash) + image_name = 'Armbian_19.11.3_Orangepipc_bionic_current_5.3.9.img' + image_path = os.path.join(self.workdir, image_name) + process.run("7z e -o%s %s" % (self.workdir, image_path_7z)) + + self.vm.set_console() + self.vm.add_args('-drive', 'file=' + image_path + ',if=sd,format=raw', + '-nic', 'user', + '-no-reboot') + self.vm.launch() + + kernel_command_line = (self.KERNEL_COMMON_COMMAND_LINE + + 'console=ttyS0,115200 ' + 'loglevel=7 ' + 'nosmp ' + 'systemd.default_timeout_start_sec=9000 ' + 'systemd.mask=armbian-zram-config.service ' + 'systemd.mask=armbian-ramlog.service') + + self.wait_for_console_pattern('U-Boot SPL') + self.wait_for_console_pattern('Autoboot in ') + exec_command_and_wait_for_pattern(self, ' ', '=>') + exec_command_and_wait_for_pattern(self, "setenv extraargs '" + + kernel_command_line + "'", '=>') + exec_command_and_wait_for_pattern(self, 'boot', 'Starting kernel ...'); + + self.wait_for_console_pattern('systemd[1]: Set hostname ' + + 'to <orangepipc>') + self.wait_for_console_pattern('Starting Load Kernel Modules...') + def test_s390x_s390_ccw_virtio(self): """ :avocado: tags=arch:s390x -- 2.20.1 From: Philippe Mathieu-Daudé <f4bug@amsat.org> This test boots U-Boot then NetBSD (stored on a SD card) on a OrangePi PC board. As it requires ~1.3GB of storage, it is disabled by default. U-Boot is built by the Debian project [1], and the SD card image is provided by the NetBSD organization [2]. Once the compressed SD card image is downloaded (304MB) and extracted, this test is fast: $ AVOCADO_ALLOW_LARGE_STORAGE=yes \ avocado --show=app,console run -t machine:orangepi-pc \ tests/acceptance/boot_linux_console.py console: U-Boot SPL 2020.01+dfsg-1 (Jan 08 2020 - 08:19:44 +0000) console: DRAM: 1024 MiB console: U-Boot 2020.01+dfsg-1 (Jan 08 2020 - 08:19:44 +0000) Allwinner Technology console: CPU: Allwinner H3 (SUN8I 0000) console: scanning bus usb@1c1b000 for devices... 1 USB Device(s) found console: scanning bus usb@1c1d000 for devices... 1 USB Device(s) found console: scanning usb for storage devices... 0 Storage Device(s) found console: Hit any key to stop autoboot: 0 console: => setenv bootargs root=ld0a console: => setenv kernel netbsd-GENERIC.ub console: => setenv fdtfile dtb/sun8i-h3-orangepi-pc.dtb console: => boot console: ## Booting kernel from Legacy Image at 42000000 ... console: Image Name: NetBSD/earmv7hf 9.0_RC1 console: Image Type: ARM Linux Kernel Image (no loading done) (uncompressed) console: XIP Kernel Image (no loading done) console: Loading Device Tree to 49ff6000, end 49fffe01 ... OK console: Starting kernel ... console: [ 1.0000000] NetBSD/evbarm (fdt) booting ... console: [ 1.0000000] NetBSD 9.0 (GENERIC) #0: Fri Feb 14 00:06:28 UTC 2020 console: [ 1.0000000] mkrepro@mkrepro.NetBSD.org:/usr/src/sys/arch/evbarm/compile/GENERIC console: [ 1.0000000] total memory = 1024 MB console: [ 1.0000000] avail memory = 1003 MB console: [ 1.0000000] armfdt0 (root) console: [ 1.0000000] simplebus0 at armfdt0: Xunlong Orange Pi PC console: [ 1.0000000] cpu0 at cpus0: Cortex-A7 r0p5 (Cortex V7A core) console: [ 1.0000000] cpu0: DC enabled IC enabled WB enabled LABT branch prediction enabled console: [ 1.0000000] cpu0: 32KB/64B 2-way L1 VIPT Instruction cache console: [ 1.0000000] cpu0: 32KB/64B 2-way write-back-locking-C L1 PIPT Data cache console: [ 1.0000000] cpu0: 2304KB/64B 16-way write-through L2 PIPT Unified cache console: [ 1.0000000] vfp0 at cpu0: NEON MPE (VFP 3.0+), rounding, NaN propagation, denormals ... console: [ 2.3812082] sdmmc0: SD card status: 4-bit, C0 console: [ 2.3812082] ld0 at sdmmc0: <0xaa:0x5859:QEMU!:0x01:0xdeadbeef:0x062> console: [ 2.4012856] ld0: 1226 MB, 622 cyl, 64 head, 63 sec, 512 bytes/sect x 2511872 sectors console: [ 2.5321222] ld0: 4-bit width, High-Speed/SDR25, 50.000 MHz console: [ 3.1068718] WARNING: 4 errors while detecting hardware; check system log. console: [ 3.1179868] boot device: ld0 console: [ 3.1470623] root on ld0a dumps on ld0b console: [ 3.2464436] root file system type: ffs console: [ 3.2897123] kern.module.path=/stand/evbarm/9.0/modules console: Mon Feb 17 20:33:35 UTC 2020 console: Starting root file system check: PASS (35.96 s) RESULTS : PASS 1 | ERROR 0 | FAIL 0 | SKIP 0 | WARN 0 | INTERRUPT 0 | CANCEL 0 JOB TIME : 36.09 s Note, this test only took ~65 seconds to run on Travis-CI, see: [3]. This test is based on a description from Niek Linnenbank from [4]. [1] https://wiki.debian.org/InstallingDebianOn/Allwinner#Creating_a_bootable_SD_Card_with_u-boot [2] https://wiki.netbsd.org/ports/evbarm/allwinner/ [3] https://travis-ci.org/philmd/qemu/jobs/638823612#L3778 [4] https://www.mail-archive.com/qemu-devel@nongnu.org/msg669347.html Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Tested-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20200311221854.30370-18-nieklinnenbank@gmail.com [NL: changed test to use NetBSD 9.0 final release and -global allwinner-rtc.base-year] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/acceptance/boot_linux_console.py | 70 ++++++++++++++++++++++++++ 1 file changed, 70 insertions(+) diff --git a/tests/acceptance/boot_linux_console.py b/tests/acceptance/boot_linux_console.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/boot_linux_console.py +++ b/tests/acceptance/boot_linux_console.py @@ -XXX,XX +XXX,XX @@ import shutil from avocado import skipUnless from avocado_qemu import Test from avocado_qemu import exec_command_and_wait_for_pattern +from avocado_qemu import interrupt_interactive_console_until_pattern from avocado_qemu import wait_for_console_pattern from avocado.utils import process from avocado.utils import archive @@ -XXX,XX +XXX,XX @@ class BootLinuxConsole(Test): 'to <orangepipc>') self.wait_for_console_pattern('Starting Load Kernel Modules...') + @skipUnless(os.getenv('AVOCADO_ALLOW_LARGE_STORAGE'), 'storage limited') + def test_arm_orangepi_uboot_netbsd9(self): + """ + :avocado: tags=arch:arm + :avocado: tags=machine:orangepi-pc + """ + # This test download a 304MB compressed image and expand it to 1.3GB... + deb_url = ('http://snapshot.debian.org/archive/debian/' + '20200108T145233Z/pool/main/u/u-boot/' + 'u-boot-sunxi_2020.01%2Bdfsg-1_armhf.deb') + deb_hash = 'f67f404a80753ca3d1258f13e38f2b060e13db99' + deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash) + # We use the common OrangePi PC 'plus' build of U-Boot for our secondary + # program loader (SPL). We will then set the path to the more specific + # OrangePi "PC" device tree blob with 'setenv fdtfile' in U-Boot prompt, + # before to boot NetBSD. + uboot_path = '/usr/lib/u-boot/orangepi_plus/u-boot-sunxi-with-spl.bin' + uboot_path = self.extract_from_deb(deb_path, uboot_path) + image_url = ('https://cdn.netbsd.org/pub/NetBSD/NetBSD-9.0/' + 'evbarm-earmv7hf/binary/gzimg/armv7.img.gz') + image_hash = '2babb29d36d8360adcb39c09e31060945259917a' + image_path_gz = self.fetch_asset(image_url, asset_hash=image_hash) + image_path = os.path.join(self.workdir, 'armv7.img') + image_drive_args = 'if=sd,format=raw,snapshot=on,file=' + image_path + archive.gzip_uncompress(image_path_gz, image_path) + + # dd if=u-boot-sunxi-with-spl.bin of=armv7.img bs=1K seek=8 conv=notrunc + with open(uboot_path, 'rb') as f_in: + with open(image_path, 'r+b') as f_out: + f_out.seek(8 * 1024) + shutil.copyfileobj(f_in, f_out) + + # Extend image, to avoid that NetBSD thinks the partition + # inside the image is larger than device size itself + f_out.seek(0, 2) + f_out.seek(64 * 1024 * 1024, 1) + f_out.write(bytearray([0x00])) + + self.vm.set_console() + self.vm.add_args('-nic', 'user', + '-drive', image_drive_args, + '-global', 'allwinner-rtc.base-year=2000', + '-no-reboot') + self.vm.launch() + wait_for_console_pattern(self, 'U-Boot 2020.01+dfsg-1') + interrupt_interactive_console_until_pattern(self, + 'Hit any key to stop autoboot:', + 'switch to partitions #0, OK') + + exec_command_and_wait_for_pattern(self, '', '=>') + cmd = 'setenv bootargs root=ld0a' + exec_command_and_wait_for_pattern(self, cmd, '=>') + cmd = 'setenv kernel netbsd-GENERIC.ub' + exec_command_and_wait_for_pattern(self, cmd, '=>') + cmd = 'setenv fdtfile dtb/sun8i-h3-orangepi-pc.dtb' + exec_command_and_wait_for_pattern(self, cmd, '=>') + cmd = ("setenv bootcmd 'fatload mmc 0:1 ${kernel_addr_r} ${kernel}; " + "fatload mmc 0:1 ${fdt_addr_r} ${fdtfile}; " + "fdt addr ${fdt_addr_r}; " + "bootm ${kernel_addr_r} - ${fdt_addr_r}'") + exec_command_and_wait_for_pattern(self, cmd, '=>') + + exec_command_and_wait_for_pattern(self, 'boot', + 'Booting kernel from Legacy Image') + wait_for_console_pattern(self, 'Starting kernel ...') + wait_for_console_pattern(self, 'NetBSD 9.0 (GENERIC)') + # Wait for user-space + wait_for_console_pattern(self, 'Starting root file system check') + def test_s390x_s390_ccw_virtio(self): """ :avocado: tags=arch:s390x -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> The Xunlong Orange Pi PC machine is a functional ARM machine based on the Allwinner H3 System-on-Chip. It supports mainline Linux, U-Boot, NetBSD and is covered by acceptance tests. This commit adds a documentation text file with a description of the machine and instructions for the user. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20200311221854.30370-19-nieklinnenbank@gmail.com [PMM: moved file into docs/system/arm to match the reorg of the arm target part of the docs; tweaked heading to match other boards] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- MAINTAINERS | 1 + docs/system/arm/orangepi.rst | 253 +++++++++++++++++++++++++++++++++++ docs/system/target-arm.rst | 2 + 3 files changed, 256 insertions(+) create mode 100644 docs/system/arm/orangepi.rst diff --git a/MAINTAINERS b/MAINTAINERS index XXXXXXX..XXXXXXX 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -XXX,XX +XXX,XX @@ S: Maintained F: hw/*/allwinner-h3* F: include/hw/*/allwinner-h3* F: hw/arm/orangepi.c +F: docs/system/orangepi.rst ARM PrimeCell and CMSDK devices M: Peter Maydell <peter.maydell@linaro.org> diff --git a/docs/system/arm/orangepi.rst b/docs/system/arm/orangepi.rst new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/docs/system/arm/orangepi.rst @@ -XXX,XX +XXX,XX @@ +Orange Pi PC (``orangepi-pc``) +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +The Xunlong Orange Pi PC is an Allwinner H3 System on Chip +based embedded computer with mainline support in both U-Boot +and Linux. The board comes with a Quad Core Cortex-A7 @ 1.3GHz, +1GiB RAM, 100Mbit ethernet, USB, SD/MMC, USB, HDMI and +various other I/O. + +Supported devices +""""""""""""""""" + +The Orange Pi PC machine supports the following devices: + + * SMP (Quad Core Cortex-A7) + * Generic Interrupt Controller configuration + * SRAM mappings + * SDRAM controller + * Real Time Clock + * Timer device (re-used from Allwinner A10) + * UART + * SD/MMC storage controller + * EMAC ethernet + * USB 2.0 interfaces + * Clock Control Unit + * System Control module + * Security Identifier device + +Limitations +""""""""""" + +Currently, Orange Pi PC does *not* support the following features: + +- Graphical output via HDMI, GPU and/or the Display Engine +- Audio output +- Hardware Watchdog + +Also see the 'unimplemented' array in the Allwinner H3 SoC module +for a complete list of unimplemented I/O devices: ``./hw/arm/allwinner-h3.c`` + +Boot options +"""""""""""" + +The Orange Pi PC machine can start using the standard -kernel functionality +for loading a Linux kernel or ELF executable. Additionally, the Orange Pi PC +machine can also emulate the BootROM which is present on an actual Allwinner H3 +based SoC, which loads the bootloader from a SD card, specified via the -sd argument +to qemu-system-arm. + +Machine-specific options +"""""""""""""""""""""""" + +The following machine-specific options are supported: + +- allwinner-rtc.base-year=YYYY + + The Allwinner RTC device is automatically created by the Orange Pi PC machine + and uses a default base year value which can be overridden using the 'base-year' property. + The base year is the actual represented year when the RTC year value is zero. + This option can be used in case the target operating system driver uses a different + base year value. The minimum value for the base year is 1900. + +- allwinner-sid.identifier=abcd1122-a000-b000-c000-12345678ffff + + The Security Identifier value can be read by the guest. + For example, U-Boot uses it to determine a unique MAC address. + +The above machine-specific options can be specified in qemu-system-arm +via the '-global' argument, for example: + +.. code-block:: bash + + $ qemu-system-arm -M orangepi-pc -sd mycard.img \ + -global allwinner-rtc.base-year=2000 + +Running mainline Linux +"""""""""""""""""""""" + +Mainline Linux kernels from 4.19 up to latest master are known to work. +To build a Linux mainline kernel that can be booted by the Orange Pi PC machine, +simply configure the kernel using the sunxi_defconfig configuration: + +.. code-block:: bash + + $ ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- make mrproper + $ ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- make sunxi_defconfig + +To be able to use USB storage, you need to manually enable the corresponding +configuration item. Start the kconfig configuration tool: + +.. code-block:: bash + + $ ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- make menuconfig + +Navigate to the following item, enable it and save your configuration: + + Device Drivers > USB support > USB Mass Storage support + +Build the Linux kernel with: + +.. code-block:: bash + + $ ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- make + +To boot the newly build linux kernel in QEMU with the Orange Pi PC machine, use: + +.. code-block:: bash + + $ qemu-system-arm -M orangepi-pc -nic user -nographic \ + -kernel /path/to/linux/arch/arm/boot/zImage \ + -append 'console=ttyS0,115200' \ + -dtb /path/to/linux/arch/arm/boot/dts/sun8i-h3-orangepi-pc.dtb + +Orange Pi PC images +""""""""""""""""""" + +Note that the mainline kernel does not have a root filesystem. You may provide it +with an official Orange Pi PC image from the official website: + + http://www.orangepi.org/downloadresources/ + +Another possibility is to run an Armbian image for Orange Pi PC which +can be downloaded from: + + https://www.armbian.com/orange-pi-pc/ + +Alternatively, you can also choose to build you own image with buildroot +using the orangepi_pc_defconfig. Also see https://buildroot.org for more information. + +You can choose to attach the selected image either as an SD card or as USB mass storage. +For example, to boot using the Orange Pi PC Debian image on SD card, simply add the -sd +argument and provide the proper root= kernel parameter: + +.. code-block:: bash + + $ qemu-system-arm -M orangepi-pc -nic user -nographic \ + -kernel /path/to/linux/arch/arm/boot/zImage \ + -append 'console=ttyS0,115200 root=/dev/mmcblk0p2' \ + -dtb /path/to/linux/arch/arm/boot/dts/sun8i-h3-orangepi-pc.dtb \ + -sd OrangePi_pc_debian_stretch_server_linux5.3.5_v1.0.img + +To attach the image as an USB mass storage device to the machine, +simply append to the command: + +.. code-block:: bash + + -drive if=none,id=stick,file=myimage.img \ + -device usb-storage,bus=usb-bus.0,drive=stick + +Instead of providing a custom Linux kernel via the -kernel command you may also +choose to let the Orange Pi PC machine load the bootloader from SD card, just like +a real board would do using the BootROM. Simply pass the selected image via the -sd +argument and remove the -kernel, -append, -dbt and -initrd arguments: + +.. code-block:: bash + + $ qemu-system-arm -M orangepi-pc -nic user -nographic \ + -sd Armbian_19.11.3_Orangepipc_buster_current_5.3.9.img + +Note that both the official Orange Pi PC images and Armbian images start +a lot of userland programs via systemd. Depending on the host hardware and OS, +they may be slow to emulate, especially due to emulating the 4 cores. +To help reduce the performance slow down due to emulating the 4 cores, you can +give the following kernel parameters via U-Boot (or via -append): + +.. code-block:: bash + + => setenv extraargs 'systemd.default_timeout_start_sec=9000 loglevel=7 nosmp console=ttyS0,115200' + +Running U-Boot +"""""""""""""" + +U-Boot mainline can be build and configured using the orangepi_pc_defconfig +using similar commands as describe above for Linux. Note that it is recommended +for development/testing to select the following configuration setting in U-Boot: + + Device Tree Control > Provider for DTB for DT Control > Embedded DTB + +To start U-Boot using the Orange Pi PC machine, provide the +u-boot binary to the -kernel argument: + +.. code-block:: bash + + $ qemu-system-arm -M orangepi-pc -nic user -nographic \ + -kernel /path/to/uboot/u-boot -sd disk.img + +Use the following U-boot commands to load and boot a Linux kernel from SD card: + +.. code-block:: bash + + => setenv bootargs console=ttyS0,115200 + => ext2load mmc 0 0x42000000 zImage + => ext2load mmc 0 0x43000000 sun8i-h3-orangepi-pc.dtb + => bootz 0x42000000 - 0x43000000 + +Running NetBSD +"""""""""""""" + +The NetBSD operating system also includes support for Allwinner H3 based boards, +including the Orange Pi PC. NetBSD 9.0 is known to work best for the Orange Pi PC +board and provides a fully working system with serial console, networking and storage. +For the Orange Pi PC machine, get the 'evbarm-earmv7hf' based image from: + + https://cdn.netbsd.org/pub/NetBSD/NetBSD-9.0/evbarm-earmv7hf/binary/gzimg/armv7.img.gz + +The image requires manually installing U-Boot in the image. Build U-Boot with +the orangepi_pc_defconfig configuration as described in the previous section. +Next, unzip the NetBSD image and write the U-Boot binary including SPL using: + +.. code-block:: bash + + $ gunzip armv7.img.gz + $ dd if=/path/to/u-boot-sunxi-with-spl.bin of=armv7.img bs=1024 seek=8 conv=notrunc + +Finally, before starting the machine the SD image must be extended such +that the NetBSD kernel will not conclude the NetBSD partition is larger than +the emulated SD card: + +.. code-block:: bash + + $ dd if=/dev/zero bs=1M count=64 >> armv7.img + +Start the machine using the following command: + +.. code-block:: bash + + $ qemu-system-arm -M orangepi-pc -nic user -nographic \ + -sd armv7.img -global allwinner-rtc.base-year=2000 + +At the U-Boot stage, interrupt the automatic boot process by pressing a key +and set the following environment variables before booting: + +.. code-block:: bash + + => setenv bootargs root=ld0a + => setenv kernel netbsd-GENERIC.ub + => setenv fdtfile dtb/sun8i-h3-orangepi-pc.dtb + => setenv bootcmd 'fatload mmc 0:1 ${kernel_addr_r} ${kernel}; fatload mmc 0:1 ${fdt_addr_r} ${fdtfile}; fdt addr ${fdt_addr_r}; bootm ${kernel_addr_r} - ${fdt_addr_r}' + +Optionally you may save the environment variables to SD card with 'saveenv'. +To continue booting simply give the 'boot' command and NetBSD boots. + +Orange Pi PC acceptance tests +""""""""""""""""""""""""""""" + +The Orange Pi PC machine has several acceptance tests included. +To run the whole set of tests, build QEMU from source and simply +provide the following command: + +.. code-block:: bash + + $ AVOCADO_ALLOW_LARGE_STORAGE=yes avocado --show=app,console run \ + -t machine:orangepi-pc tests/acceptance/boot_linux_console.py diff --git a/docs/system/target-arm.rst b/docs/system/target-arm.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/target-arm.rst +++ b/docs/system/target-arm.rst @@ -XXX,XX +XXX,XX @@ undocumented; you can get a complete list by running ``qemu-system-aarch64 --machine help``. .. toctree:: + :maxdepth: 1 arm/integratorcp arm/versatile @@ -XXX,XX +XXX,XX @@ undocumented; you can get a complete list by running arm/stellaris arm/musicpal arm/sx1 + arm/orangepi Arm CPU features ================ -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> Mention 'max' value in the gic-version property description. Signed-off-by: Eric Auger <eric.auger@redhat.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Andrew Jones <drjones@redhat.com> Message-id: 20200311131618.7187-2-eric.auger@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/virt.c | 3 ++- 1 file changed, 2 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 void virt_instance_init(Object *obj) virt_set_gic_version, NULL); object_property_set_description(obj, "gic-version", "Set GIC version. " - "Valid values are 2, 3 and host", NULL); + "Valid values are 2, 3, host and max", + NULL); vms->highmem_ecam = !vmc->no_highmem_ecam; -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> We plan to introduce yet another value for the gic version (nosel). As we already use exotic values such as 0 and -1, let's introduce a dedicated enum type and let vms->gic_version take this type. Signed-off-by: Eric Auger <eric.auger@redhat.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Andrew Jones <drjones@redhat.com> Message-id: 20200311131618.7187-3-eric.auger@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/virt.h | 11 +++++++++-- hw/arm/virt.c | 30 +++++++++++++++--------------- 2 files changed, 24 insertions(+), 17 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 enum VirtIOMMUType { VIRT_IOMMU_VIRTIO, } VirtIOMMUType; +typedef enum VirtGICType { + VIRT_GIC_VERSION_MAX, + VIRT_GIC_VERSION_HOST, + VIRT_GIC_VERSION_2, + VIRT_GIC_VERSION_3, +} VirtGICType; + typedef struct MemMapEntry { hwaddr base; hwaddr size; @@ -XXX,XX +XXX,XX @@ typedef struct { bool highmem_ecam; bool its; bool virt; - int32_t gic_version; + VirtGICType gic_version; VirtIOMMUType iommu; uint16_t virtio_iommu_bdf; struct arm_boot_info bootinfo; @@ -XXX,XX +XXX,XX @@ static inline int virt_gicv3_redist_region_count(VirtMachineState *vms) uint32_t redist0_capacity = vms->memmap[VIRT_GIC_REDIST].size / GICV3_REDIST_SIZE; - assert(vms->gic_version == 3); + assert(vms->gic_version == VIRT_GIC_VERSION_3); return vms->smp_cpus > redist0_capacity ? 2 : 1; } 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_timer_nodes(const VirtMachineState *vms) irqflags = GIC_FDT_IRQ_FLAGS_EDGE_LO_HI; } - if (vms->gic_version == 2) { + if (vms->gic_version == VIRT_GIC_VERSION_2) { irqflags = deposit32(irqflags, GIC_FDT_IRQ_PPI_CPU_START, GIC_FDT_IRQ_PPI_CPU_WIDTH, (1 << vms->smp_cpus) - 1); @@ -XXX,XX +XXX,XX @@ static void fdt_add_gic_node(VirtMachineState *vms) qemu_fdt_setprop_cell(vms->fdt, nodename, "#address-cells", 0x2); qemu_fdt_setprop_cell(vms->fdt, nodename, "#size-cells", 0x2); qemu_fdt_setprop(vms->fdt, nodename, "ranges", NULL, 0); - if (vms->gic_version == 3) { + if (vms->gic_version == VIRT_GIC_VERSION_3) { int nb_redist_regions = virt_gicv3_redist_region_count(vms); qemu_fdt_setprop_string(vms->fdt, nodename, "compatible", @@ -XXX,XX +XXX,XX @@ static void fdt_add_pmu_nodes(const VirtMachineState *vms) } } - if (vms->gic_version == 2) { + if (vms->gic_version == VIRT_GIC_VERSION_2) { irqflags = deposit32(irqflags, GIC_FDT_IRQ_PPI_CPU_START, GIC_FDT_IRQ_PPI_CPU_WIDTH, (1 << vms->smp_cpus) - 1); @@ -XXX,XX +XXX,XX @@ static uint64_t virt_cpu_mp_affinity(VirtMachineState *vms, int idx) * purposes are to make TCG consistent (with 64-bit KVM hosts) * and to improve SGI efficiency. */ - if (vms->gic_version == 3) { + if (vms->gic_version == VIRT_GIC_VERSION_3) { clustersz = GICV3_TARGETLIST_BITS; } else { clustersz = GIC_TARGETLIST_BITS; @@ -XXX,XX +XXX,XX @@ static void machvirt_init(MachineState *machine) /* We can probe only here because during property set * KVM is not available yet */ - if (vms->gic_version <= 0) { - /* "host" or "max" */ + if (vms->gic_version == VIRT_GIC_VERSION_HOST || + vms->gic_version == VIRT_GIC_VERSION_MAX) { if (!kvm_enabled()) { - if (vms->gic_version == 0) { + if (vms->gic_version == VIRT_GIC_VERSION_HOST) { error_report("gic-version=host requires KVM"); exit(1); } else { /* "max": currently means 3 for TCG */ - vms->gic_version = 3; + vms->gic_version = VIRT_GIC_VERSION_3; } } else { vms->gic_version = kvm_arm_vgic_probe(); @@ -XXX,XX +XXX,XX @@ static void machvirt_init(MachineState *machine) /* The maximum number of CPUs depends on the GIC version, or on how * many redistributors we can fit into the memory map. */ - if (vms->gic_version == 3) { + if (vms->gic_version == VIRT_GIC_VERSION_3) { virt_max_cpus = vms->memmap[VIRT_GIC_REDIST].size / GICV3_REDIST_SIZE; virt_max_cpus += @@ -XXX,XX +XXX,XX @@ static void virt_set_its(Object *obj, bool value, Error **errp) static char *virt_get_gic_version(Object *obj, Error **errp) { VirtMachineState *vms = VIRT_MACHINE(obj); - const char *val = vms->gic_version == 3 ? "3" : "2"; + const char *val = vms->gic_version == VIRT_GIC_VERSION_3 ? "3" : "2"; return g_strdup(val); } @@ -XXX,XX +XXX,XX @@ static void virt_set_gic_version(Object *obj, const char *value, Error **errp) VirtMachineState *vms = VIRT_MACHINE(obj); if (!strcmp(value, "3")) { - vms->gic_version = 3; + vms->gic_version = VIRT_GIC_VERSION_3; } else if (!strcmp(value, "2")) { - vms->gic_version = 2; + vms->gic_version = VIRT_GIC_VERSION_2; } else if (!strcmp(value, "host")) { - vms->gic_version = 0; /* Will probe later */ + vms->gic_version = VIRT_GIC_VERSION_HOST; /* Will probe later */ } else if (!strcmp(value, "max")) { - vms->gic_version = -1; /* Will probe later */ + vms->gic_version = VIRT_GIC_VERSION_MAX; /* Will probe later */ } else { error_setg(errp, "Invalid gic-version value"); error_append_hint(errp, "Valid values are 3, 2, host, max.\n"); @@ -XXX,XX +XXX,XX @@ static void virt_instance_init(Object *obj) "physical address space above 32 bits", NULL); /* Default GIC type is v2 */ - vms->gic_version = 2; + vms->gic_version = VIRT_GIC_VERSION_2; object_property_add_str(obj, "gic-version", virt_get_gic_version, virt_set_gic_version, NULL); object_property_set_description(obj, "gic-version", -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> Let's move the code which freezes which gic-version to be applied in a dedicated function. We also now set by default the VIRT_GIC_VERSION_NO_SET. This eventually turns into the legacy v2 choice in the finalize() function. Signed-off-by: Eric Auger <eric.auger@redhat.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Andrew Jones <drjones@redhat.com> Message-id: 20200311131618.7187-4-eric.auger@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/virt.h | 1 + hw/arm/virt.c | 54 ++++++++++++++++++++++++++----------------- 2 files changed, 34 insertions(+), 21 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 enum VirtGICType { VIRT_GIC_VERSION_HOST, VIRT_GIC_VERSION_2, VIRT_GIC_VERSION_3, + VIRT_GIC_VERSION_NOSEL, } VirtGICType; typedef struct MemMapEntry { 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 virt_set_memmap(VirtMachineState *vms) } } +/* + * finalize_gic_version - Determines the final gic_version + * according to the gic-version property + * + * Default GIC type is v2 + */ +static void finalize_gic_version(VirtMachineState *vms) +{ + if (vms->gic_version == VIRT_GIC_VERSION_HOST || + vms->gic_version == VIRT_GIC_VERSION_MAX) { + if (!kvm_enabled()) { + if (vms->gic_version == VIRT_GIC_VERSION_HOST) { + error_report("gic-version=host requires KVM"); + exit(1); + } else { + /* "max": currently means 3 for TCG */ + vms->gic_version = VIRT_GIC_VERSION_3; + } + } else { + vms->gic_version = kvm_arm_vgic_probe(); + if (!vms->gic_version) { + error_report( + "Unable to determine GIC version supported by host"); + exit(1); + } + } + } else if (vms->gic_version == VIRT_GIC_VERSION_NOSEL) { + vms->gic_version = VIRT_GIC_VERSION_2; + } +} + static void machvirt_init(MachineState *machine) { VirtMachineState *vms = VIRT_MACHINE(machine); @@ -XXX,XX +XXX,XX @@ static void machvirt_init(MachineState *machine) /* We can probe only here because during property set * KVM is not available yet */ - if (vms->gic_version == VIRT_GIC_VERSION_HOST || - vms->gic_version == VIRT_GIC_VERSION_MAX) { - if (!kvm_enabled()) { - if (vms->gic_version == VIRT_GIC_VERSION_HOST) { - error_report("gic-version=host requires KVM"); - exit(1); - } else { - /* "max": currently means 3 for TCG */ - vms->gic_version = VIRT_GIC_VERSION_3; - } - } else { - vms->gic_version = kvm_arm_vgic_probe(); - if (!vms->gic_version) { - error_report( - "Unable to determine GIC version supported by host"); - exit(1); - } - } - } + finalize_gic_version(vms); if (!cpu_type_valid(machine->cpu_type)) { error_report("mach-virt: CPU type %s not supported", machine->cpu_type); @@ -XXX,XX +XXX,XX @@ static void virt_instance_init(Object *obj) "Set on/off to enable/disable using " "physical address space above 32 bits", NULL); - /* Default GIC type is v2 */ - vms->gic_version = VIRT_GIC_VERSION_2; + vms->gic_version = VIRT_GIC_VERSION_NOSEL; object_property_add_str(obj, "gic-version", virt_get_gic_version, virt_set_gic_version, NULL); object_property_set_description(obj, "gic-version", -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> Convert kvm_arm_vgic_probe() so that it returns a bitmap of supported in-kernel emulation VGIC versions instead of the max version: at the moment values can be v2 and v3. This allows to expose the case where the host GICv3 also supports GICv2 emulation. This will be useful to choose the default version in KVM accelerated mode. Signed-off-by: Eric Auger <eric.auger@redhat.com> Reviewed-by: Andrew Jones <drjones@redhat.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200311131618.7187-5-eric.auger@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/kvm_arm.h | 3 +++ hw/arm/virt.c | 11 +++++++++-- target/arm/kvm.c | 14 ++++++++------ 3 files changed, 20 insertions(+), 8 deletions(-) diff --git a/target/arm/kvm_arm.h b/target/arm/kvm_arm.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/kvm_arm.h +++ b/target/arm/kvm_arm.h @@ -XXX,XX +XXX,XX @@ #include "exec/memory.h" #include "qemu/error-report.h" +#define KVM_ARM_VGIC_V2 (1 << 0) +#define KVM_ARM_VGIC_V3 (1 << 1) + /** * kvm_arm_vcpu_init: * @cs: CPUState 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 finalize_gic_version(VirtMachineState *vms) vms->gic_version = VIRT_GIC_VERSION_3; } } else { - vms->gic_version = kvm_arm_vgic_probe(); - if (!vms->gic_version) { + int probe_bitmap = kvm_arm_vgic_probe(); + + if (!probe_bitmap) { error_report( "Unable to determine GIC version supported by host"); exit(1); + } else { + if (probe_bitmap & KVM_ARM_VGIC_V3) { + vms->gic_version = VIRT_GIC_VERSION_3; + } else { + vms->gic_version = VIRT_GIC_VERSION_2; + } } } } else if (vms->gic_version == VIRT_GIC_VERSION_NOSEL) { diff --git a/target/arm/kvm.c b/target/arm/kvm.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/kvm.c +++ b/target/arm/kvm.c @@ -XXX,XX +XXX,XX @@ int kvm_arch_irqchip_create(KVMState *s) int kvm_arm_vgic_probe(void) { + int val = 0; + if (kvm_create_device(kvm_state, KVM_DEV_TYPE_ARM_VGIC_V3, true) == 0) { - return 3; - } else if (kvm_create_device(kvm_state, - KVM_DEV_TYPE_ARM_VGIC_V2, true) == 0) { - return 2; - } else { - return 0; + val |= KVM_ARM_VGIC_V3; } + if (kvm_create_device(kvm_state, + KVM_DEV_TYPE_ARM_VGIC_V2, true) == 0) { + val |= KVM_ARM_VGIC_V2; + } + return val; } int kvm_arm_set_irq(int cpu, int irqtype, int irq, int level) -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> Restructure the finalize_gic_version with switch cases and clearly separate the following cases: - KVM mode / in-kernel irqchip - KVM mode / userspace irqchip - TCG mode In KVM mode / in-kernel irqchip , we explictly check whether the chosen version is supported by the host. If the end-user explicitly sets v2/v3 and this is not supported by the host, then the user gets an explicit error message. Note that for old kernels where the CREATE_DEVICE ioctl doesn't exist then we will now fail if the user specifically asked for gicv2, where previously we (probably) would have succeeded. In KVM mode / userspace irqchip we immediatly output an error in case the end-user explicitly selected v3. Also we warn the end-user about the unexpected usage of gic-version=host in that case as only userspace GICv2 is supported. Signed-off-by: Eric Auger <eric.auger@redhat.com> Reviewed-by: Andrew Jones <drjones@redhat.com> Message-id: 20200311131618.7187-6-eric.auger@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/virt.c | 88 +++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 67 insertions(+), 21 deletions(-) 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 virt_set_memmap(VirtMachineState *vms) */ static void finalize_gic_version(VirtMachineState *vms) { - if (vms->gic_version == VIRT_GIC_VERSION_HOST || - vms->gic_version == VIRT_GIC_VERSION_MAX) { - if (!kvm_enabled()) { - if (vms->gic_version == VIRT_GIC_VERSION_HOST) { - error_report("gic-version=host requires KVM"); - exit(1); - } else { - /* "max": currently means 3 for TCG */ - vms->gic_version = VIRT_GIC_VERSION_3; - } - } else { - int probe_bitmap = kvm_arm_vgic_probe(); + if (kvm_enabled()) { + int probe_bitmap; - if (!probe_bitmap) { + if (!kvm_irqchip_in_kernel()) { + switch (vms->gic_version) { + case VIRT_GIC_VERSION_HOST: + warn_report( + "gic-version=host not relevant with kernel-irqchip=off " + "as only userspace GICv2 is supported. Using v2 ..."); + return; + case VIRT_GIC_VERSION_MAX: + case VIRT_GIC_VERSION_NOSEL: + vms->gic_version = VIRT_GIC_VERSION_2; + return; + case VIRT_GIC_VERSION_2: + return; + case VIRT_GIC_VERSION_3: error_report( - "Unable to determine GIC version supported by host"); + "gic-version=3 is not supported with kernel-irqchip=off"); exit(1); - } else { - if (probe_bitmap & KVM_ARM_VGIC_V3) { - vms->gic_version = VIRT_GIC_VERSION_3; - } else { - vms->gic_version = VIRT_GIC_VERSION_2; - } } } - } else if (vms->gic_version == VIRT_GIC_VERSION_NOSEL) { + + probe_bitmap = kvm_arm_vgic_probe(); + if (!probe_bitmap) { + error_report("Unable to determine GIC version supported by host"); + exit(1); + } + + switch (vms->gic_version) { + case VIRT_GIC_VERSION_HOST: + case VIRT_GIC_VERSION_MAX: + if (probe_bitmap & KVM_ARM_VGIC_V3) { + vms->gic_version = VIRT_GIC_VERSION_3; + } else { + vms->gic_version = VIRT_GIC_VERSION_2; + } + return; + case VIRT_GIC_VERSION_NOSEL: + vms->gic_version = VIRT_GIC_VERSION_2; + break; + case VIRT_GIC_VERSION_2: + case VIRT_GIC_VERSION_3: + break; + } + + /* Check chosen version is effectively supported by the host */ + if (vms->gic_version == VIRT_GIC_VERSION_2 && + !(probe_bitmap & KVM_ARM_VGIC_V2)) { + error_report("host does not support in-kernel GICv2 emulation"); + exit(1); + } else if (vms->gic_version == VIRT_GIC_VERSION_3 && + !(probe_bitmap & KVM_ARM_VGIC_V3)) { + error_report("host does not support in-kernel GICv3 emulation"); + exit(1); + } + return; + } + + /* TCG mode */ + switch (vms->gic_version) { + case VIRT_GIC_VERSION_NOSEL: vms->gic_version = VIRT_GIC_VERSION_2; + break; + case VIRT_GIC_VERSION_MAX: + vms->gic_version = VIRT_GIC_VERSION_3; + break; + case VIRT_GIC_VERSION_HOST: + error_report("gic-version=host requires KVM"); + exit(1); + case VIRT_GIC_VERSION_2: + case VIRT_GIC_VERSION_3: + break; } } -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> At the moment if the end-user does not specify the gic-version along with KVM acceleration, v2 is set by default. However most of the systems now have GICv3 and sometimes they do not support GICv2 compatibility. This patch keeps the default v2 selection in all cases except in the KVM accelerated mode when either - the host does not support GICv2 in-kernel emulation or - number of VCPUS exceeds 8. Those cases did not work anyway so we do not break any compatibility. Now we get v3 selected in such a case. Signed-off-by: Eric Auger <eric.auger@redhat.com> Reported-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Andrew Jones <drjones@redhat.com> Message-id: 20200311131618.7187-7-eric.auger@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/virt.c | 17 ++++++++++++++++- 1 file changed, 16 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 void virt_set_memmap(VirtMachineState *vms) */ static void finalize_gic_version(VirtMachineState *vms) { + unsigned int max_cpus = MACHINE(vms)->smp.max_cpus; + if (kvm_enabled()) { int probe_bitmap; @@ -XXX,XX +XXX,XX @@ static void finalize_gic_version(VirtMachineState *vms) } return; case VIRT_GIC_VERSION_NOSEL: - vms->gic_version = VIRT_GIC_VERSION_2; + if ((probe_bitmap & KVM_ARM_VGIC_V2) && max_cpus <= GIC_NCPU) { + vms->gic_version = VIRT_GIC_VERSION_2; + } else if (probe_bitmap & KVM_ARM_VGIC_V3) { + /* + * in case the host does not support v2 in-kernel emulation or + * the end-user requested more than 8 VCPUs we now default + * to v3. In any case defaulting to v2 would be broken. + */ + vms->gic_version = VIRT_GIC_VERSION_3; + } else if (max_cpus > GIC_NCPU) { + error_report("host only supports in-kernel GICv2 emulation " + "but more than 8 vcpus are requested"); + exit(1); + } break; case VIRT_GIC_VERSION_2: case VIRT_GIC_VERSION_3: -- 2.20.1 From: Beata Michalska <beata.michalska@linaro.org> KVM_SET_VCPU_EVENTS might actually lead to vcpu registers being modified. As such this should be the last step of sync to avoid potential overwriting of whatever changes KVM might have done. Signed-off-by: Beata Michalska <beata.michalska@linaro.org> Reviewed-by: Andrew Jones <drjones@redhat.com> Message-id: 20200312003401.29017-2-beata.michalska@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/kvm32.c | 15 ++++++++++----- target/arm/kvm64.c | 15 ++++++++++----- 2 files changed, 20 insertions(+), 10 deletions(-) diff --git a/target/arm/kvm32.c b/target/arm/kvm32.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/kvm32.c +++ b/target/arm/kvm32.c @@ -XXX,XX +XXX,XX @@ int kvm_arch_put_registers(CPUState *cs, int level) return ret; } - ret = kvm_put_vcpu_events(cpu); - if (ret) { - return ret; - } - write_cpustate_to_list(cpu, true); if (!write_list_to_kvmstate(cpu, level)) { return EINVAL; } + /* + * Setting VCPU events should be triggered after syncing the registers + * to avoid overwriting potential changes made by KVM upon calling + * KVM_SET_VCPU_EVENTS ioctl + */ + ret = kvm_put_vcpu_events(cpu); + if (ret) { + return ret; + } + kvm_arm_sync_mpstate_to_kvm(cpu); return ret; diff --git a/target/arm/kvm64.c b/target/arm/kvm64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/kvm64.c +++ b/target/arm/kvm64.c @@ -XXX,XX +XXX,XX @@ int kvm_arch_put_registers(CPUState *cs, int level) return ret; } - ret = kvm_put_vcpu_events(cpu); - if (ret) { - return ret; - } - write_cpustate_to_list(cpu, true); if (!write_list_to_kvmstate(cpu, level)) { return -EINVAL; } + /* + * Setting VCPU events should be triggered after syncing the registers + * to avoid overwriting potential changes made by KVM upon calling + * KVM_SET_VCPU_EVENTS ioctl + */ + ret = kvm_put_vcpu_events(cpu); + if (ret) { + return ret; + } + kvm_arm_sync_mpstate_to_kvm(cpu); return ret; -- 2.20.1 Mostly this is patches from me and RTH cleaning up and doing more decodetree conversion for AArch32 Neon. The major new feature is Dongjiu Geng's patchset to report host memory errors to KVM guests; also a new aspeed board from Patrick Williams. thanks -- PMM The following changes since commit 035b448b84f3557206abc44d786c5d3db2638f7d: Merge remote-tracking branch 'remotes/gkurz/tags/9p-next-2020-05-14' into staging (2020-05-14 10:58:30 +0100) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20200514 for you to fetch changes up to e95485f85657be21135c17a9226e297c21e73360: target/arm: Convert NEON VFMA, VFMS 3-reg-same insns to decodetree (2020-05-14 15:03:09 +0100) ---------------------------------------------------------------- target-arm queue: * target/arm: Use correct GDB XML for M-profile cores * target/arm: Code cleanup to use gvec APIs better * aspeed: Add support for the sonorapass-bmc board * target/arm: Support reporting KVM host memory errors to the guest via ACPI notifications * target/arm: Finish conversion of Neon 3-reg-same insns to decodetree ---------------------------------------------------------------- Dongjiu Geng (10): acpi: nvdimm: change NVDIMM_UUID_LE to a common macro hw/arm/virt: Introduce a RAS machine option docs: APEI GHES generation and CPER record description ACPI: Build related register address fields via hardware error fw_cfg blob ACPI: Build Hardware Error Source Table ACPI: Record the Generic Error Status Block address KVM: Move hwpoison page related functions into kvm-all.c ACPI: Record Generic Error Status Block(GESB) table target-arm: kvm64: handle SIGBUS signal from kernel or KVM MAINTAINERS: Add ACPI/HEST/GHES entries Patrick Williams (1): aspeed: Add support for the sonorapass-bmc board Peter Maydell (18): target/arm: Use correct GDB XML for M-profile cores target/arm: Convert Neon 3-reg-same VQRDMLAH/VQRDMLSH to decodetree target/arm: Convert Neon 3-reg-same SHA to decodetree target/arm: Convert Neon 64-bit element 3-reg-same insns target/arm: Convert Neon VHADD 3-reg-same insns target/arm: Convert Neon VABA/VABD 3-reg-same to decodetree target/arm: Convert Neon VRHADD, VHSUB 3-reg-same insns to decodetree target/arm: Convert Neon VQSHL, VRSHL, VQRSHL 3-reg-same insns to decodetree target/arm: Convert Neon VPMAX/VPMIN 3-reg-same insns to decodetree target/arm: Convert Neon VPADD 3-reg-same insns to decodetree target/arm: Convert Neon VQDMULH/VQRDMULH 3-reg-same to decodetree target/arm: Convert Neon VADD, VSUB, VABD 3-reg-same insns to decodetree target/arm: Convert Neon VPMIN/VPMAX/VPADD float 3-reg-same insns to decodetree target/arm: Convert Neon fp VMUL, VMLA, VMLS 3-reg-same insns to decodetree target/arm: Convert Neon 3-reg-same compare insns to decodetree target/arm: Move 'env' argument of recps_f32 and rsqrts_f32 helpers to usual place target/arm: Convert Neon fp VMAX/VMIN/VMAXNM/VMINNM/VRECPS/VRSQRTS to decodetree target/arm: Convert NEON VFMA, VFMS 3-reg-same insns to decodetree Richard Henderson (16): target/arm: Create gen_gvec_[us]sra target/arm: Create gen_gvec_{u,s}{rshr,rsra} target/arm: Create gen_gvec_{sri,sli} target/arm: Remove unnecessary range check for VSHL target/arm: Tidy handle_vec_simd_shri target/arm: Create gen_gvec_{ceq,clt,cle,cgt,cge}0 target/arm: Create gen_gvec_{mla,mls} target/arm: Swap argument order for VSHL during decode target/arm: Create gen_gvec_{cmtst,ushl,sshl} target/arm: Create gen_gvec_{uqadd, sqadd, uqsub, sqsub} target/arm: Remove fp_status from helper_{recpe, rsqrte}_u32 target/arm: Create gen_gvec_{qrdmla,qrdmls} target/arm: Pass pointer to qc to qrdmla/qrdmls target/arm: Clear tail in gvec_fmul_idx_*, gvec_fmla_idx_* target/arm: Vectorize SABD/UABD target/arm: Vectorize SABA/UABA docs/specs/acpi_hest_ghes.rst | 110 ++ docs/specs/index.rst | 1 + configure | 4 +- default-configs/arm-softmmu.mak | 1 + include/hw/acpi/aml-build.h | 1 + include/hw/acpi/generic_event_device.h | 2 + include/hw/acpi/ghes.h | 74 + include/hw/arm/virt.h | 1 + include/qemu/uuid.h | 27 + include/sysemu/kvm.h | 3 +- include/sysemu/kvm_int.h | 12 + target/arm/cpu.h | 4 + target/arm/helper.h | 78 +- target/arm/internals.h | 5 +- target/arm/translate.h | 84 +- target/i386/cpu.h | 2 + target/arm/neon-dp.decode | 119 +- accel/kvm/kvm-all.c | 36 + hw/acpi/aml-build.c | 2 + hw/acpi/generic_event_device.c | 19 + hw/acpi/ghes.c | 448 ++++++ hw/acpi/nvdimm.c | 10 +- hw/arm/aspeed.c | 78 ++ hw/arm/virt-acpi-build.c | 15 + hw/arm/virt.c | 23 + target/arm/cpu_tcg.c | 1 + target/arm/gdbstub.c | 22 +- target/arm/helper.c | 2 +- target/arm/kvm64.c | 77 ++ target/arm/neon_helper.c | 17 - target/arm/tlb_helper.c | 2 +- target/arm/translate-a64.c | 210 +-- target/arm/translate-neon.inc.c | 682 +++++++++- target/arm/translate.c | 2349 +++++++++++++++++--------------- target/arm/vec_helper.c | 240 +++- target/arm/vfp_helper.c | 9 +- target/i386/kvm.c | 36 - MAINTAINERS | 9 + gdb-xml/arm-m-profile.xml | 27 + hw/acpi/Kconfig | 4 + hw/acpi/Makefile.objs | 1 + 41 files changed, 3402 insertions(+), 1445 deletions(-) create mode 100644 docs/specs/acpi_hest_ghes.rst create mode 100644 include/hw/acpi/ghes.h create mode 100644 hw/acpi/ghes.c create mode 100644 gdb-xml/arm-m-profile.xml GDB's remote protocol requires M-profile cores to use the feature name 'org.gnu.gdb.arm.m-profile' instead of the 'org.gnu.gdb.arm.core' feature used for A- and R-profile cores. We weren't doing this, which meant GDB treated our M-profile cores like A-profile ones. This mostly doesn't matter, but for instance means that it doesn't correctly handle backtraces where an M-profile exception frame is involved. Ship a copy of GDB's arm-m-profile.xml and use it on the M-profile cores. The integer registers have the same offsets as the arm-core.xml, but register 25 is the M-profile XPSR rather than the A-profile CPSR, so we need to update arm_cpu_gdb_read_register() and arm_cpu_gdb_write_register() to handle XSPR reads and writes. Fixes: https://bugs.launchpad.net/qemu/+bug/1877136 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20200507134755.13997-1-peter.maydell@linaro.org --- configure | 4 ++-- target/arm/cpu_tcg.c | 1 + target/arm/gdbstub.c | 22 ++++++++++++++++++---- gdb-xml/arm-m-profile.xml | 27 +++++++++++++++++++++++++++ 4 files changed, 48 insertions(+), 6 deletions(-) create mode 100644 gdb-xml/arm-m-profile.xml diff --git a/configure b/configure index XXXXXXX..XXXXXXX 100755 --- a/configure +++ b/configure @@ -XXX,XX +XXX,XX @@ case "$target_name" in TARGET_SYSTBL_ABI=common,oabi bflt="yes" mttcg="yes" - gdb_xml_files="arm-core.xml arm-vfp.xml arm-vfp3.xml arm-neon.xml" + gdb_xml_files="arm-core.xml arm-vfp.xml arm-vfp3.xml arm-neon.xml arm-m-profile.xml" ;; aarch64|aarch64_be) TARGET_ARCH=aarch64 TARGET_BASE_ARCH=arm bflt="yes" mttcg="yes" - gdb_xml_files="aarch64-core.xml aarch64-fpu.xml arm-core.xml arm-vfp.xml arm-vfp3.xml arm-neon.xml" + gdb_xml_files="aarch64-core.xml aarch64-fpu.xml arm-core.xml arm-vfp.xml arm-vfp3.xml arm-neon.xml arm-m-profile.xml" ;; cris) ;; 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) #endif cc->cpu_exec_interrupt = arm_v7m_cpu_exec_interrupt; + cc->gdb_core_xml_file = "arm-m-profile.xml"; } static const ARMCPUInfo arm_tcg_cpus[] = { diff --git a/target/arm/gdbstub.c b/target/arm/gdbstub.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/gdbstub.c +++ b/target/arm/gdbstub.c @@ -XXX,XX +XXX,XX @@ int arm_cpu_gdb_read_register(CPUState *cs, GByteArray *mem_buf, int n) } return gdb_get_reg32(mem_buf, 0); case 25: - /* CPSR */ - return gdb_get_reg32(mem_buf, cpsr_read(env)); + /* CPSR, or XPSR for M-profile */ + if (arm_feature(env, ARM_FEATURE_M)) { + return gdb_get_reg32(mem_buf, xpsr_read(env)); + } else { + return gdb_get_reg32(mem_buf, cpsr_read(env)); + } } /* Unknown register. */ return 0; @@ -XXX,XX +XXX,XX @@ int arm_cpu_gdb_write_register(CPUState *cs, uint8_t *mem_buf, int n) } return 4; case 25: - /* CPSR */ - cpsr_write(env, tmp, 0xffffffff, CPSRWriteByGDBStub); + /* CPSR, or XPSR for M-profile */ + if (arm_feature(env, ARM_FEATURE_M)) { + /* + * Don't allow writing to XPSR.Exception as it can cause + * a transition into or out of handler mode (it's not + * writeable via the MSR insn so this is a reasonable + * restriction). Other fields are safe to update. + */ + xpsr_write(env, tmp, ~XPSR_EXCP); + } else { + cpsr_write(env, tmp, 0xffffffff, CPSRWriteByGDBStub); + } return 4; } /* Unknown register. */ diff --git a/gdb-xml/arm-m-profile.xml b/gdb-xml/arm-m-profile.xml new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/gdb-xml/arm-m-profile.xml @@ -XXX,XX +XXX,XX @@ +<?xml version="1.0"?> +<!-- Copyright (C) 2010-2020 Free Software Foundation, Inc. + + Copying and distribution of this file, with or without modification, + are permitted in any medium without royalty provided the copyright + notice and this notice are preserved. --> + +<!DOCTYPE feature SYSTEM "gdb-target.dtd"> +<feature name="org.gnu.gdb.arm.m-profile"> + <reg name="r0" bitsize="32"/> + <reg name="r1" bitsize="32"/> + <reg name="r2" bitsize="32"/> + <reg name="r3" bitsize="32"/> + <reg name="r4" bitsize="32"/> + <reg name="r5" bitsize="32"/> + <reg name="r6" bitsize="32"/> + <reg name="r7" bitsize="32"/> + <reg name="r8" bitsize="32"/> + <reg name="r9" bitsize="32"/> + <reg name="r10" bitsize="32"/> + <reg name="r11" bitsize="32"/> + <reg name="r12" bitsize="32"/> + <reg name="sp" bitsize="32" type="data_ptr"/> + <reg name="lr" bitsize="32"/> + <reg name="pc" bitsize="32" type="code_ptr"/> + <reg name="xpsr" bitsize="32" regnum="25"/> +</feature> -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> The functions eliminate duplication of the special cases for this operation. They match up with the GVecGen2iFn typedef. Add out-of-line helpers. We got away with only having inline expanders because the neon vector size is only 16 bytes, and we know that the inline expansion will always succeed. When we reuse this for SVE, tcg-gvec-op may decide to use an out-of-line helper due to longer vector lengths. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-2-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 10 +++ target/arm/translate.h | 7 +- target/arm/translate-a64.c | 15 +--- target/arm/translate.c | 161 ++++++++++++++++++++++--------------- target/arm/vec_helper.c | 25 ++++++ 5 files changed, 139 insertions(+), 79 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_4(gvec_pmull_q, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(neon_pmull_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_ssra_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_ssra_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_ssra_s, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_ssra_d, TCG_CALL_NO_RWG, void, ptr, ptr, i32) + +DEF_HELPER_FLAGS_3(gvec_usra_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_usra_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_usra_s, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_usra_d, TCG_CALL_NO_RWG, void, ptr, ptr, i32) + #ifdef TARGET_AARCH64 #include "helper-a64.h" #include "helper-sve.h" 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 @@ extern const GVecGen3 mls_op[4]; extern const GVecGen3 cmtst_op[4]; extern const GVecGen3 sshl_op[4]; extern const GVecGen3 ushl_op[4]; -extern const GVecGen2i ssra_op[4]; -extern const GVecGen2i usra_op[4]; extern const GVecGen2i sri_op[4]; extern const GVecGen2i sli_op[4]; extern const GVecGen4 uqadd_op[4]; @@ -XXX,XX +XXX,XX @@ void gen_sshl_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b); void gen_ushl_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b); void gen_sshl_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b); +void gen_gvec_ssra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_usra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz); + /* * Forward to the isar_feature_* tests given a DisasContext pointer. */ 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_vec_simd_shri(DisasContext *s, bool is_q, bool is_u, switch (opcode) { case 0x02: /* SSRA / USRA (accumulate) */ - if (is_u) { - /* Shift count same as element size produces zero to add. */ - if (shift == 8 << size) { - goto done; - } - gen_gvec_op2i(s, is_q, rd, rn, shift, &usra_op[size]); - } else { - /* Shift count same as element size produces all sign to add. */ - if (shift == 8 << size) { - shift -= 1; - } - gen_gvec_op2i(s, is_q, rd, rn, shift, &ssra_op[size]); - } + gen_gvec_fn2i(s, is_q, rd, rn, shift, + is_u ? gen_gvec_usra : gen_gvec_ssra, size); return; case 0x08: /* SRI */ /* Shift count same as element size is valid but does nothing. */ diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_ssra_vec(unsigned vece, TCGv_vec d, TCGv_vec a, int64_t sh) tcg_gen_add_vec(vece, d, d, a); } -static const TCGOpcode vecop_list_ssra[] = { - INDEX_op_sari_vec, INDEX_op_add_vec, 0 -}; +void gen_gvec_ssra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_sari_vec, INDEX_op_add_vec, 0 + }; + static const GVecGen2i ops[4] = { + { .fni8 = gen_ssra8_i64, + .fniv = gen_ssra_vec, + .fno = gen_helper_gvec_ssra_b, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fni8 = gen_ssra16_i64, + .fniv = gen_ssra_vec, + .fno = gen_helper_gvec_ssra_h, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = gen_ssra32_i32, + .fniv = gen_ssra_vec, + .fno = gen_helper_gvec_ssra_s, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = gen_ssra64_i64, + .fniv = gen_ssra_vec, + .fno = gen_helper_gvec_ssra_b, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_64 }, + }; -const GVecGen2i ssra_op[4] = { - { .fni8 = gen_ssra8_i64, - .fniv = gen_ssra_vec, - .load_dest = true, - .opt_opc = vecop_list_ssra, - .vece = MO_8 }, - { .fni8 = gen_ssra16_i64, - .fniv = gen_ssra_vec, - .load_dest = true, - .opt_opc = vecop_list_ssra, - .vece = MO_16 }, - { .fni4 = gen_ssra32_i32, - .fniv = gen_ssra_vec, - .load_dest = true, - .opt_opc = vecop_list_ssra, - .vece = MO_32 }, - { .fni8 = gen_ssra64_i64, - .fniv = gen_ssra_vec, - .prefer_i64 = TCG_TARGET_REG_BITS == 64, - .opt_opc = vecop_list_ssra, - .load_dest = true, - .vece = MO_64 }, -}; + /* tszimm encoding produces immediates in the range [1..esize]. */ + tcg_debug_assert(shift > 0); + tcg_debug_assert(shift <= (8 << vece)); + + /* + * Shifts larger than the element size are architecturally valid. + * Signed results in all sign bits. + */ + shift = MIN(shift, (8 << vece) - 1); + tcg_gen_gvec_2i(rd_ofs, rm_ofs, opr_sz, max_sz, shift, &ops[vece]); +} static void gen_usra8_i64(TCGv_i64 d, TCGv_i64 a, int64_t shift) { @@ -XXX,XX +XXX,XX @@ static void gen_usra_vec(unsigned vece, TCGv_vec d, TCGv_vec a, int64_t sh) tcg_gen_add_vec(vece, d, d, a); } -static const TCGOpcode vecop_list_usra[] = { - INDEX_op_shri_vec, INDEX_op_add_vec, 0 -}; +void gen_gvec_usra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_shri_vec, INDEX_op_add_vec, 0 + }; + static const GVecGen2i ops[4] = { + { .fni8 = gen_usra8_i64, + .fniv = gen_usra_vec, + .fno = gen_helper_gvec_usra_b, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_8, }, + { .fni8 = gen_usra16_i64, + .fniv = gen_usra_vec, + .fno = gen_helper_gvec_usra_h, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_16, }, + { .fni4 = gen_usra32_i32, + .fniv = gen_usra_vec, + .fno = gen_helper_gvec_usra_s, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_32, }, + { .fni8 = gen_usra64_i64, + .fniv = gen_usra_vec, + .fno = gen_helper_gvec_usra_d, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_64, }, + }; -const GVecGen2i usra_op[4] = { - { .fni8 = gen_usra8_i64, - .fniv = gen_usra_vec, - .load_dest = true, - .opt_opc = vecop_list_usra, - .vece = MO_8, }, - { .fni8 = gen_usra16_i64, - .fniv = gen_usra_vec, - .load_dest = true, - .opt_opc = vecop_list_usra, - .vece = MO_16, }, - { .fni4 = gen_usra32_i32, - .fniv = gen_usra_vec, - .load_dest = true, - .opt_opc = vecop_list_usra, - .vece = MO_32, }, - { .fni8 = gen_usra64_i64, - .fniv = gen_usra_vec, - .prefer_i64 = TCG_TARGET_REG_BITS == 64, - .load_dest = true, - .opt_opc = vecop_list_usra, - .vece = MO_64, }, -}; + /* tszimm encoding produces immediates in the range [1..esize]. */ + tcg_debug_assert(shift > 0); + tcg_debug_assert(shift <= (8 << vece)); + + /* + * Shifts larger than the element size are architecturally valid. + * Unsigned results in all zeros as input to accumulate: nop. + */ + if (shift < (8 << vece)) { + tcg_gen_gvec_2i(rd_ofs, rm_ofs, opr_sz, max_sz, shift, &ops[vece]); + } else { + /* Nop, but we do need to clear the tail. */ + tcg_gen_gvec_mov(vece, rd_ofs, rd_ofs, opr_sz, max_sz); + } +} static void gen_shr8_ins_i64(TCGv_i64 d, TCGv_i64 a, int64_t shift) { @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case 1: /* VSRA */ /* Right shift comes here negative. */ shift = -shift; - /* Shifts larger than the element size are architecturally - * valid. Unsigned results in all zeros; signed results - * in all sign bits. - */ - if (!u) { - tcg_gen_gvec_2i(rd_ofs, rm_ofs, vec_size, vec_size, - MIN(shift, (8 << size) - 1), - &ssra_op[size]); - } else if (shift >= 8 << size) { - /* rd += 0 */ + if (u) { + gen_gvec_usra(size, rd_ofs, rm_ofs, shift, + vec_size, vec_size); } else { - tcg_gen_gvec_2i(rd_ofs, rm_ofs, vec_size, vec_size, - shift, &usra_op[size]); + gen_gvec_ssra(size, rd_ofs, rm_ofs, shift, + vec_size, vec_size); } return 0; diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(gvec_sqsub_d)(void *vd, void *vq, void *vn, clear_tail(d, oprsz, simd_maxsz(desc)); } + +#define DO_SRA(NAME, TYPE) \ +void HELPER(NAME)(void *vd, void *vn, uint32_t desc) \ +{ \ + intptr_t i, oprsz = simd_oprsz(desc); \ + int shift = simd_data(desc); \ + TYPE *d = vd, *n = vn; \ + for (i = 0; i < oprsz / sizeof(TYPE); i++) { \ + d[i] += n[i] >> shift; \ + } \ + clear_tail(d, oprsz, simd_maxsz(desc)); \ +} + +DO_SRA(gvec_ssra_b, int8_t) +DO_SRA(gvec_ssra_h, int16_t) +DO_SRA(gvec_ssra_s, int32_t) +DO_SRA(gvec_ssra_d, int64_t) + +DO_SRA(gvec_usra_b, uint8_t) +DO_SRA(gvec_usra_h, uint16_t) +DO_SRA(gvec_usra_s, uint32_t) +DO_SRA(gvec_usra_d, uint64_t) + +#undef DO_SRA + /* * Convert float16 to float32, raising no exceptions and * preserving exceptional values, including SNaN. -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Create vectorized versions of handle_shri_with_rndacc for shift+round and shift+round+accumulate. Add out-of-line helpers in preparation for longer vector lengths from SVE. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-3-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 20 ++ target/arm/translate.h | 9 + target/arm/translate-a64.c | 11 +- target/arm/translate.c | 463 +++++++++++++++++++++++++++++++++++-- target/arm/vec_helper.c | 50 ++++ 5 files changed, 527 insertions(+), 26 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_3(gvec_usra_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_3(gvec_usra_s, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_3(gvec_usra_d, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_srshr_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_srshr_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_srshr_s, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_srshr_d, TCG_CALL_NO_RWG, void, ptr, ptr, i32) + +DEF_HELPER_FLAGS_3(gvec_urshr_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_urshr_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_urshr_s, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_urshr_d, TCG_CALL_NO_RWG, void, ptr, ptr, i32) + +DEF_HELPER_FLAGS_3(gvec_srsra_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_srsra_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_srsra_s, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_srsra_d, TCG_CALL_NO_RWG, void, ptr, ptr, i32) + +DEF_HELPER_FLAGS_3(gvec_ursra_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_ursra_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_ursra_s, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_ursra_d, TCG_CALL_NO_RWG, void, ptr, ptr, i32) + #ifdef TARGET_AARCH64 #include "helper-a64.h" #include "helper-sve.h" 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 @@ void gen_gvec_ssra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, void gen_gvec_usra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, int64_t shift, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_srshr(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_urshr(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_srsra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_ursra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz); + /* * Forward to the isar_feature_* tests given a DisasContext pointer. */ 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_vec_simd_shri(DisasContext *s, bool is_q, bool is_u, return; case 0x04: /* SRSHR / URSHR (rounding) */ - break; + gen_gvec_fn2i(s, is_q, rd, rn, shift, + is_u ? gen_gvec_urshr : gen_gvec_srshr, size); + return; + case 0x06: /* SRSRA / URSRA (accum + rounding) */ - accumulate = true; - break; + gen_gvec_fn2i(s, is_q, rd, rn, shift, + is_u ? gen_gvec_ursra : gen_gvec_srsra, size); + return; + default: g_assert_not_reached(); } 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 @@ void gen_gvec_usra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, } } +/* + * Shift one less than the requested amount, and the low bit is + * the rounding bit. For the 8 and 16-bit operations, because we + * mask the low bit, we can perform a normal integer shift instead + * of a vector shift. + */ +static void gen_srshr8_i64(TCGv_i64 d, TCGv_i64 a, int64_t sh) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + tcg_gen_shri_i64(t, a, sh - 1); + tcg_gen_andi_i64(t, t, dup_const(MO_8, 1)); + tcg_gen_vec_sar8i_i64(d, a, sh); + tcg_gen_vec_add8_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_srshr16_i64(TCGv_i64 d, TCGv_i64 a, int64_t sh) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + tcg_gen_shri_i64(t, a, sh - 1); + tcg_gen_andi_i64(t, t, dup_const(MO_16, 1)); + tcg_gen_vec_sar16i_i64(d, a, sh); + tcg_gen_vec_add16_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_srshr32_i32(TCGv_i32 d, TCGv_i32 a, int32_t sh) +{ + TCGv_i32 t = tcg_temp_new_i32(); + + tcg_gen_extract_i32(t, a, sh - 1, 1); + tcg_gen_sari_i32(d, a, sh); + tcg_gen_add_i32(d, d, t); + tcg_temp_free_i32(t); +} + +static void gen_srshr64_i64(TCGv_i64 d, TCGv_i64 a, int64_t sh) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + tcg_gen_extract_i64(t, a, sh - 1, 1); + tcg_gen_sari_i64(d, a, sh); + tcg_gen_add_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_srshr_vec(unsigned vece, TCGv_vec d, TCGv_vec a, int64_t sh) +{ + TCGv_vec t = tcg_temp_new_vec_matching(d); + TCGv_vec ones = tcg_temp_new_vec_matching(d); + + tcg_gen_shri_vec(vece, t, a, sh - 1); + tcg_gen_dupi_vec(vece, ones, 1); + tcg_gen_and_vec(vece, t, t, ones); + tcg_gen_sari_vec(vece, d, a, sh); + tcg_gen_add_vec(vece, d, d, t); + + tcg_temp_free_vec(t); + tcg_temp_free_vec(ones); +} + +void gen_gvec_srshr(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_shri_vec, INDEX_op_sari_vec, INDEX_op_add_vec, 0 + }; + static const GVecGen2i ops[4] = { + { .fni8 = gen_srshr8_i64, + .fniv = gen_srshr_vec, + .fno = gen_helper_gvec_srshr_b, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fni8 = gen_srshr16_i64, + .fniv = gen_srshr_vec, + .fno = gen_helper_gvec_srshr_h, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = gen_srshr32_i32, + .fniv = gen_srshr_vec, + .fno = gen_helper_gvec_srshr_s, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = gen_srshr64_i64, + .fniv = gen_srshr_vec, + .fno = gen_helper_gvec_srshr_d, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .opt_opc = vecop_list, + .vece = MO_64 }, + }; + + /* tszimm encoding produces immediates in the range [1..esize] */ + tcg_debug_assert(shift > 0); + tcg_debug_assert(shift <= (8 << vece)); + + if (shift == (8 << vece)) { + /* + * Shifts larger than the element size are architecturally valid. + * Signed results in all sign bits. With rounding, this produces + * (-1 + 1) >> 1 == 0, or (0 + 1) >> 1 == 0. + * I.e. always zero. + */ + tcg_gen_gvec_dup_imm(vece, rd_ofs, opr_sz, max_sz, 0); + } else { + tcg_gen_gvec_2i(rd_ofs, rm_ofs, opr_sz, max_sz, shift, &ops[vece]); + } +} + +static void gen_srsra8_i64(TCGv_i64 d, TCGv_i64 a, int64_t sh) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + gen_srshr8_i64(t, a, sh); + tcg_gen_vec_add8_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_srsra16_i64(TCGv_i64 d, TCGv_i64 a, int64_t sh) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + gen_srshr16_i64(t, a, sh); + tcg_gen_vec_add16_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_srsra32_i32(TCGv_i32 d, TCGv_i32 a, int32_t sh) +{ + TCGv_i32 t = tcg_temp_new_i32(); + + gen_srshr32_i32(t, a, sh); + tcg_gen_add_i32(d, d, t); + tcg_temp_free_i32(t); +} + +static void gen_srsra64_i64(TCGv_i64 d, TCGv_i64 a, int64_t sh) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + gen_srshr64_i64(t, a, sh); + tcg_gen_add_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_srsra_vec(unsigned vece, TCGv_vec d, TCGv_vec a, int64_t sh) +{ + TCGv_vec t = tcg_temp_new_vec_matching(d); + + gen_srshr_vec(vece, t, a, sh); + tcg_gen_add_vec(vece, d, d, t); + tcg_temp_free_vec(t); +} + +void gen_gvec_srsra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_shri_vec, INDEX_op_sari_vec, INDEX_op_add_vec, 0 + }; + static const GVecGen2i ops[4] = { + { .fni8 = gen_srsra8_i64, + .fniv = gen_srsra_vec, + .fno = gen_helper_gvec_srsra_b, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_8 }, + { .fni8 = gen_srsra16_i64, + .fniv = gen_srsra_vec, + .fno = gen_helper_gvec_srsra_h, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_16 }, + { .fni4 = gen_srsra32_i32, + .fniv = gen_srsra_vec, + .fno = gen_helper_gvec_srsra_s, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_32 }, + { .fni8 = gen_srsra64_i64, + .fniv = gen_srsra_vec, + .fno = gen_helper_gvec_srsra_d, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_64 }, + }; + + /* tszimm encoding produces immediates in the range [1..esize] */ + tcg_debug_assert(shift > 0); + tcg_debug_assert(shift <= (8 << vece)); + + /* + * Shifts larger than the element size are architecturally valid. + * Signed results in all sign bits. With rounding, this produces + * (-1 + 1) >> 1 == 0, or (0 + 1) >> 1 == 0. + * I.e. always zero. With accumulation, this leaves D unchanged. + */ + if (shift == (8 << vece)) { + /* Nop, but we do need to clear the tail. */ + tcg_gen_gvec_mov(vece, rd_ofs, rd_ofs, opr_sz, max_sz); + } else { + tcg_gen_gvec_2i(rd_ofs, rm_ofs, opr_sz, max_sz, shift, &ops[vece]); + } +} + +static void gen_urshr8_i64(TCGv_i64 d, TCGv_i64 a, int64_t sh) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + tcg_gen_shri_i64(t, a, sh - 1); + tcg_gen_andi_i64(t, t, dup_const(MO_8, 1)); + tcg_gen_vec_shr8i_i64(d, a, sh); + tcg_gen_vec_add8_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_urshr16_i64(TCGv_i64 d, TCGv_i64 a, int64_t sh) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + tcg_gen_shri_i64(t, a, sh - 1); + tcg_gen_andi_i64(t, t, dup_const(MO_16, 1)); + tcg_gen_vec_shr16i_i64(d, a, sh); + tcg_gen_vec_add16_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_urshr32_i32(TCGv_i32 d, TCGv_i32 a, int32_t sh) +{ + TCGv_i32 t = tcg_temp_new_i32(); + + tcg_gen_extract_i32(t, a, sh - 1, 1); + tcg_gen_shri_i32(d, a, sh); + tcg_gen_add_i32(d, d, t); + tcg_temp_free_i32(t); +} + +static void gen_urshr64_i64(TCGv_i64 d, TCGv_i64 a, int64_t sh) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + tcg_gen_extract_i64(t, a, sh - 1, 1); + tcg_gen_shri_i64(d, a, sh); + tcg_gen_add_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_urshr_vec(unsigned vece, TCGv_vec d, TCGv_vec a, int64_t shift) +{ + TCGv_vec t = tcg_temp_new_vec_matching(d); + TCGv_vec ones = tcg_temp_new_vec_matching(d); + + tcg_gen_shri_vec(vece, t, a, shift - 1); + tcg_gen_dupi_vec(vece, ones, 1); + tcg_gen_and_vec(vece, t, t, ones); + tcg_gen_shri_vec(vece, d, a, shift); + tcg_gen_add_vec(vece, d, d, t); + + tcg_temp_free_vec(t); + tcg_temp_free_vec(ones); +} + +void gen_gvec_urshr(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_shri_vec, INDEX_op_add_vec, 0 + }; + static const GVecGen2i ops[4] = { + { .fni8 = gen_urshr8_i64, + .fniv = gen_urshr_vec, + .fno = gen_helper_gvec_urshr_b, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fni8 = gen_urshr16_i64, + .fniv = gen_urshr_vec, + .fno = gen_helper_gvec_urshr_h, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = gen_urshr32_i32, + .fniv = gen_urshr_vec, + .fno = gen_helper_gvec_urshr_s, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = gen_urshr64_i64, + .fniv = gen_urshr_vec, + .fno = gen_helper_gvec_urshr_d, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .opt_opc = vecop_list, + .vece = MO_64 }, + }; + + /* tszimm encoding produces immediates in the range [1..esize] */ + tcg_debug_assert(shift > 0); + tcg_debug_assert(shift <= (8 << vece)); + + if (shift == (8 << vece)) { + /* + * Shifts larger than the element size are architecturally valid. + * Unsigned results in zero. With rounding, this produces a + * copy of the most significant bit. + */ + tcg_gen_gvec_shri(vece, rd_ofs, rm_ofs, shift - 1, opr_sz, max_sz); + } else { + tcg_gen_gvec_2i(rd_ofs, rm_ofs, opr_sz, max_sz, shift, &ops[vece]); + } +} + +static void gen_ursra8_i64(TCGv_i64 d, TCGv_i64 a, int64_t sh) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + if (sh == 8) { + tcg_gen_vec_shr8i_i64(t, a, 7); + } else { + gen_urshr8_i64(t, a, sh); + } + tcg_gen_vec_add8_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_ursra16_i64(TCGv_i64 d, TCGv_i64 a, int64_t sh) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + if (sh == 16) { + tcg_gen_vec_shr16i_i64(t, a, 15); + } else { + gen_urshr16_i64(t, a, sh); + } + tcg_gen_vec_add16_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_ursra32_i32(TCGv_i32 d, TCGv_i32 a, int32_t sh) +{ + TCGv_i32 t = tcg_temp_new_i32(); + + if (sh == 32) { + tcg_gen_shri_i32(t, a, 31); + } else { + gen_urshr32_i32(t, a, sh); + } + tcg_gen_add_i32(d, d, t); + tcg_temp_free_i32(t); +} + +static void gen_ursra64_i64(TCGv_i64 d, TCGv_i64 a, int64_t sh) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + if (sh == 64) { + tcg_gen_shri_i64(t, a, 63); + } else { + gen_urshr64_i64(t, a, sh); + } + tcg_gen_add_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_ursra_vec(unsigned vece, TCGv_vec d, TCGv_vec a, int64_t sh) +{ + TCGv_vec t = tcg_temp_new_vec_matching(d); + + if (sh == (8 << vece)) { + tcg_gen_shri_vec(vece, t, a, sh - 1); + } else { + gen_urshr_vec(vece, t, a, sh); + } + tcg_gen_add_vec(vece, d, d, t); + tcg_temp_free_vec(t); +} + +void gen_gvec_ursra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_shri_vec, INDEX_op_add_vec, 0 + }; + static const GVecGen2i ops[4] = { + { .fni8 = gen_ursra8_i64, + .fniv = gen_ursra_vec, + .fno = gen_helper_gvec_ursra_b, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_8 }, + { .fni8 = gen_ursra16_i64, + .fniv = gen_ursra_vec, + .fno = gen_helper_gvec_ursra_h, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_16 }, + { .fni4 = gen_ursra32_i32, + .fniv = gen_ursra_vec, + .fno = gen_helper_gvec_ursra_s, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_32 }, + { .fni8 = gen_ursra64_i64, + .fniv = gen_ursra_vec, + .fno = gen_helper_gvec_ursra_d, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_64 }, + }; + + /* tszimm encoding produces immediates in the range [1..esize] */ + tcg_debug_assert(shift > 0); + tcg_debug_assert(shift <= (8 << vece)); + + tcg_gen_gvec_2i(rd_ofs, rm_ofs, opr_sz, max_sz, shift, &ops[vece]); +} + static void gen_shr8_ins_i64(TCGv_i64 d, TCGv_i64 a, int64_t shift) { uint64_t mask = dup_const(MO_8, 0xff >> shift); @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) } return 0; + case 2: /* VRSHR */ + /* Right shift comes here negative. */ + shift = -shift; + if (u) { + gen_gvec_urshr(size, rd_ofs, rm_ofs, shift, + vec_size, vec_size); + } else { + gen_gvec_srshr(size, rd_ofs, rm_ofs, shift, + vec_size, vec_size); + } + return 0; + + case 3: /* VRSRA */ + /* Right shift comes here negative. */ + shift = -shift; + if (u) { + gen_gvec_ursra(size, rd_ofs, rm_ofs, shift, + vec_size, vec_size); + } else { + gen_gvec_srsra(size, rd_ofs, rm_ofs, shift, + vec_size, vec_size); + } + return 0; + case 4: /* VSRI */ if (!u) { return 1; @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) neon_load_reg64(cpu_V0, rm + pass); tcg_gen_movi_i64(cpu_V1, imm); switch (op) { - case 2: /* VRSHR */ - case 3: /* VRSRA */ - if (u) - gen_helper_neon_rshl_u64(cpu_V0, cpu_V0, cpu_V1); - else - gen_helper_neon_rshl_s64(cpu_V0, cpu_V0, cpu_V1); - break; case 6: /* VQSHLU */ gen_helper_neon_qshlu_s64(cpu_V0, cpu_env, cpu_V0, cpu_V1); @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) default: g_assert_not_reached(); } - if (op == 3) { - /* Accumulate. */ - neon_load_reg64(cpu_V1, rd + pass); - tcg_gen_add_i64(cpu_V0, cpu_V0, cpu_V1); - } neon_store_reg64(cpu_V0, rd + pass); } else { /* size < 3 */ /* Operands in T0 and T1. */ @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) tmp2 = tcg_temp_new_i32(); tcg_gen_movi_i32(tmp2, imm); switch (op) { - case 2: /* VRSHR */ - case 3: /* VRSRA */ - GEN_NEON_INTEGER_OP(rshl); - break; case 6: /* VQSHLU */ switch (size) { case 0: @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) g_assert_not_reached(); } tcg_temp_free_i32(tmp2); - - if (op == 3) { - /* Accumulate. */ - tmp2 = neon_load_reg(rd, pass); - gen_neon_add(size, tmp, tmp2); - tcg_temp_free_i32(tmp2); - } neon_store_reg(rd, pass, tmp); } } /* for pass */ diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ DO_SRA(gvec_usra_d, uint64_t) #undef DO_SRA +#define DO_RSHR(NAME, TYPE) \ +void HELPER(NAME)(void *vd, void *vn, uint32_t desc) \ +{ \ + intptr_t i, oprsz = simd_oprsz(desc); \ + int shift = simd_data(desc); \ + TYPE *d = vd, *n = vn; \ + for (i = 0; i < oprsz / sizeof(TYPE); i++) { \ + TYPE tmp = n[i] >> (shift - 1); \ + d[i] = (tmp >> 1) + (tmp & 1); \ + } \ + clear_tail(d, oprsz, simd_maxsz(desc)); \ +} + +DO_RSHR(gvec_srshr_b, int8_t) +DO_RSHR(gvec_srshr_h, int16_t) +DO_RSHR(gvec_srshr_s, int32_t) +DO_RSHR(gvec_srshr_d, int64_t) + +DO_RSHR(gvec_urshr_b, uint8_t) +DO_RSHR(gvec_urshr_h, uint16_t) +DO_RSHR(gvec_urshr_s, uint32_t) +DO_RSHR(gvec_urshr_d, uint64_t) + +#undef DO_RSHR + +#define DO_RSRA(NAME, TYPE) \ +void HELPER(NAME)(void *vd, void *vn, uint32_t desc) \ +{ \ + intptr_t i, oprsz = simd_oprsz(desc); \ + int shift = simd_data(desc); \ + TYPE *d = vd, *n = vn; \ + for (i = 0; i < oprsz / sizeof(TYPE); i++) { \ + TYPE tmp = n[i] >> (shift - 1); \ + d[i] += (tmp >> 1) + (tmp & 1); \ + } \ + clear_tail(d, oprsz, simd_maxsz(desc)); \ +} + +DO_RSRA(gvec_srsra_b, int8_t) +DO_RSRA(gvec_srsra_h, int16_t) +DO_RSRA(gvec_srsra_s, int32_t) +DO_RSRA(gvec_srsra_d, int64_t) + +DO_RSRA(gvec_ursra_b, uint8_t) +DO_RSRA(gvec_ursra_h, uint16_t) +DO_RSRA(gvec_ursra_s, uint32_t) +DO_RSRA(gvec_ursra_d, uint64_t) + +#undef DO_RSRA + /* * Convert float16 to float32, raising no exceptions and * preserving exceptional values, including SNaN. -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> The functions eliminate duplication of the special cases for this operation. They match up with the GVecGen2iFn typedef. Add out-of-line helpers. We got away with only having inline expanders because the neon vector size is only 16 bytes, and we know that the inline expansion will always succeed. When we reuse this for SVE, tcg-gvec-op may decide to use an out-of-line helper due to longer vector lengths. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-4-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 10 ++ target/arm/translate.h | 7 +- target/arm/translate-a64.c | 20 +--- target/arm/translate.c | 186 +++++++++++++++++++++---------------- target/arm/vec_helper.c | 38 ++++++++ 5 files changed, 160 insertions(+), 101 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_3(gvec_ursra_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_3(gvec_ursra_s, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_3(gvec_ursra_d, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_sri_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_sri_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_sri_s, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_sri_d, TCG_CALL_NO_RWG, void, ptr, ptr, i32) + +DEF_HELPER_FLAGS_3(gvec_sli_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_sli_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_sli_s, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_sli_d, TCG_CALL_NO_RWG, void, ptr, ptr, i32) + #ifdef TARGET_AARCH64 #include "helper-a64.h" #include "helper-sve.h" 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 @@ extern const GVecGen3 mls_op[4]; extern const GVecGen3 cmtst_op[4]; extern const GVecGen3 sshl_op[4]; extern const GVecGen3 ushl_op[4]; -extern const GVecGen2i sri_op[4]; -extern const GVecGen2i sli_op[4]; extern const GVecGen4 uqadd_op[4]; extern const GVecGen4 sqadd_op[4]; extern const GVecGen4 uqsub_op[4]; @@ -XXX,XX +XXX,XX @@ void gen_gvec_srsra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, void gen_gvec_ursra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, int64_t shift, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_sri(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_sli(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz); + /* * Forward to the isar_feature_* tests given a DisasContext pointer. */ 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 gen_gvec_op2(DisasContext *s, bool is_q, int rd, is_q ? 16 : 8, vec_full_reg_size(s), gvec_op); } -/* Expand a 2-operand + immediate AdvSIMD vector operation using - * an op descriptor. - */ -static void gen_gvec_op2i(DisasContext *s, bool is_q, int rd, - int rn, int64_t imm, const GVecGen2i *gvec_op) -{ - tcg_gen_gvec_2i(vec_full_reg_offset(s, rd), vec_full_reg_offset(s, rn), - is_q ? 16 : 8, vec_full_reg_size(s), imm, gvec_op); -} - /* Expand a 3-operand AdvSIMD vector operation using an op descriptor. */ static void gen_gvec_op3(DisasContext *s, bool is_q, int rd, int rn, int rm, const GVecGen3 *gvec_op) @@ -XXX,XX +XXX,XX @@ static void handle_vec_simd_shri(DisasContext *s, bool is_q, bool is_u, gen_gvec_fn2i(s, is_q, rd, rn, shift, is_u ? gen_gvec_usra : gen_gvec_ssra, size); return; + case 0x08: /* SRI */ - /* Shift count same as element size is valid but does nothing. */ - if (shift == 8 << size) { - goto done; - } - gen_gvec_op2i(s, is_q, rd, rn, shift, &sri_op[size]); + gen_gvec_fn2i(s, is_q, rd, rn, shift, gen_gvec_sri, size); return; case 0x00: /* SSHR / USHR */ @@ -XXX,XX +XXX,XX @@ static void handle_vec_simd_shri(DisasContext *s, bool is_q, bool is_u, } tcg_temp_free_i64(tcg_round); - done: clear_vec_high(s, is_q, rd); } @@ -XXX,XX +XXX,XX @@ static void handle_vec_simd_shli(DisasContext *s, bool is_q, bool insert, } if (insert) { - gen_gvec_op2i(s, is_q, rd, rn, shift, &sli_op[size]); + gen_gvec_fn2i(s, is_q, rd, rn, shift, gen_gvec_sli, size); } else { gen_gvec_fn2i(s, is_q, rd, rn, shift, tcg_gen_gvec_shli, size); } diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_shr64_ins_i64(TCGv_i64 d, TCGv_i64 a, int64_t shift) static void gen_shr_ins_vec(unsigned vece, TCGv_vec d, TCGv_vec a, int64_t sh) { - if (sh == 0) { - tcg_gen_mov_vec(d, a); - } else { - TCGv_vec t = tcg_temp_new_vec_matching(d); - TCGv_vec m = tcg_temp_new_vec_matching(d); + TCGv_vec t = tcg_temp_new_vec_matching(d); + TCGv_vec m = tcg_temp_new_vec_matching(d); - tcg_gen_dupi_vec(vece, m, MAKE_64BIT_MASK((8 << vece) - sh, sh)); - tcg_gen_shri_vec(vece, t, a, sh); - tcg_gen_and_vec(vece, d, d, m); - tcg_gen_or_vec(vece, d, d, t); + tcg_gen_dupi_vec(vece, m, MAKE_64BIT_MASK((8 << vece) - sh, sh)); + tcg_gen_shri_vec(vece, t, a, sh); + tcg_gen_and_vec(vece, d, d, m); + tcg_gen_or_vec(vece, d, d, t); - tcg_temp_free_vec(t); - tcg_temp_free_vec(m); - } + tcg_temp_free_vec(t); + tcg_temp_free_vec(m); } -static const TCGOpcode vecop_list_sri[] = { INDEX_op_shri_vec, 0 }; +void gen_gvec_sri(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { INDEX_op_shri_vec, 0 }; + const GVecGen2i ops[4] = { + { .fni8 = gen_shr8_ins_i64, + .fniv = gen_shr_ins_vec, + .fno = gen_helper_gvec_sri_b, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fni8 = gen_shr16_ins_i64, + .fniv = gen_shr_ins_vec, + .fno = gen_helper_gvec_sri_h, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = gen_shr32_ins_i32, + .fniv = gen_shr_ins_vec, + .fno = gen_helper_gvec_sri_s, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = gen_shr64_ins_i64, + .fniv = gen_shr_ins_vec, + .fno = gen_helper_gvec_sri_d, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_64 }, + }; -const GVecGen2i sri_op[4] = { - { .fni8 = gen_shr8_ins_i64, - .fniv = gen_shr_ins_vec, - .load_dest = true, - .opt_opc = vecop_list_sri, - .vece = MO_8 }, - { .fni8 = gen_shr16_ins_i64, - .fniv = gen_shr_ins_vec, - .load_dest = true, - .opt_opc = vecop_list_sri, - .vece = MO_16 }, - { .fni4 = gen_shr32_ins_i32, - .fniv = gen_shr_ins_vec, - .load_dest = true, - .opt_opc = vecop_list_sri, - .vece = MO_32 }, - { .fni8 = gen_shr64_ins_i64, - .fniv = gen_shr_ins_vec, - .prefer_i64 = TCG_TARGET_REG_BITS == 64, - .load_dest = true, - .opt_opc = vecop_list_sri, - .vece = MO_64 }, -}; + /* tszimm encoding produces immediates in the range [1..esize]. */ + tcg_debug_assert(shift > 0); + tcg_debug_assert(shift <= (8 << vece)); + + /* Shift of esize leaves destination unchanged. */ + if (shift < (8 << vece)) { + tcg_gen_gvec_2i(rd_ofs, rm_ofs, opr_sz, max_sz, shift, &ops[vece]); + } else { + /* Nop, but we do need to clear the tail. */ + tcg_gen_gvec_mov(vece, rd_ofs, rd_ofs, opr_sz, max_sz); + } +} static void gen_shl8_ins_i64(TCGv_i64 d, TCGv_i64 a, int64_t shift) { @@ -XXX,XX +XXX,XX @@ static void gen_shl64_ins_i64(TCGv_i64 d, TCGv_i64 a, int64_t shift) static void gen_shl_ins_vec(unsigned vece, TCGv_vec d, TCGv_vec a, int64_t sh) { - if (sh == 0) { - tcg_gen_mov_vec(d, a); - } else { - TCGv_vec t = tcg_temp_new_vec_matching(d); - TCGv_vec m = tcg_temp_new_vec_matching(d); + TCGv_vec t = tcg_temp_new_vec_matching(d); + TCGv_vec m = tcg_temp_new_vec_matching(d); - tcg_gen_dupi_vec(vece, m, MAKE_64BIT_MASK(0, sh)); - tcg_gen_shli_vec(vece, t, a, sh); - tcg_gen_and_vec(vece, d, d, m); - tcg_gen_or_vec(vece, d, d, t); + tcg_gen_shli_vec(vece, t, a, sh); + tcg_gen_dupi_vec(vece, m, MAKE_64BIT_MASK(0, sh)); + tcg_gen_and_vec(vece, d, d, m); + tcg_gen_or_vec(vece, d, d, t); - tcg_temp_free_vec(t); - tcg_temp_free_vec(m); - } + tcg_temp_free_vec(t); + tcg_temp_free_vec(m); } -static const TCGOpcode vecop_list_sli[] = { INDEX_op_shli_vec, 0 }; +void gen_gvec_sli(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { INDEX_op_shli_vec, 0 }; + const GVecGen2i ops[4] = { + { .fni8 = gen_shl8_ins_i64, + .fniv = gen_shl_ins_vec, + .fno = gen_helper_gvec_sli_b, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fni8 = gen_shl16_ins_i64, + .fniv = gen_shl_ins_vec, + .fno = gen_helper_gvec_sli_h, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = gen_shl32_ins_i32, + .fniv = gen_shl_ins_vec, + .fno = gen_helper_gvec_sli_s, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = gen_shl64_ins_i64, + .fniv = gen_shl_ins_vec, + .fno = gen_helper_gvec_sli_d, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_64 }, + }; -const GVecGen2i sli_op[4] = { - { .fni8 = gen_shl8_ins_i64, - .fniv = gen_shl_ins_vec, - .load_dest = true, - .opt_opc = vecop_list_sli, - .vece = MO_8 }, - { .fni8 = gen_shl16_ins_i64, - .fniv = gen_shl_ins_vec, - .load_dest = true, - .opt_opc = vecop_list_sli, - .vece = MO_16 }, - { .fni4 = gen_shl32_ins_i32, - .fniv = gen_shl_ins_vec, - .load_dest = true, - .opt_opc = vecop_list_sli, - .vece = MO_32 }, - { .fni8 = gen_shl64_ins_i64, - .fniv = gen_shl_ins_vec, - .prefer_i64 = TCG_TARGET_REG_BITS == 64, - .load_dest = true, - .opt_opc = vecop_list_sli, - .vece = MO_64 }, -}; + /* tszimm encoding produces immediates in the range [0..esize-1]. */ + tcg_debug_assert(shift >= 0); + tcg_debug_assert(shift < (8 << vece)); + + if (shift == 0) { + tcg_gen_gvec_mov(vece, rd_ofs, rm_ofs, opr_sz, max_sz); + } else { + tcg_gen_gvec_2i(rd_ofs, rm_ofs, opr_sz, max_sz, shift, &ops[vece]); + } +} static void gen_mla8_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) { @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) } /* Right shift comes here negative. */ shift = -shift; - /* Shift out of range leaves destination unchanged. */ - if (shift < 8 << size) { - tcg_gen_gvec_2i(rd_ofs, rm_ofs, vec_size, vec_size, - shift, &sri_op[size]); - } + gen_gvec_sri(size, rd_ofs, rm_ofs, shift, + vec_size, vec_size); return 0; case 5: /* VSHL, VSLI */ if (u) { /* VSLI */ - /* Shift out of range leaves destination unchanged. */ - if (shift < 8 << size) { - tcg_gen_gvec_2i(rd_ofs, rm_ofs, vec_size, - vec_size, shift, &sli_op[size]); - } + gen_gvec_sli(size, rd_ofs, rm_ofs, shift, + vec_size, vec_size); } else { /* VSHL */ /* Shifts larger than the element size are * architecturally valid and results in zero. diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ DO_RSRA(gvec_ursra_d, uint64_t) #undef DO_RSRA +#define DO_SRI(NAME, TYPE) \ +void HELPER(NAME)(void *vd, void *vn, uint32_t desc) \ +{ \ + intptr_t i, oprsz = simd_oprsz(desc); \ + int shift = simd_data(desc); \ + TYPE *d = vd, *n = vn; \ + for (i = 0; i < oprsz / sizeof(TYPE); i++) { \ + d[i] = deposit64(d[i], 0, sizeof(TYPE) * 8 - shift, n[i] >> shift); \ + } \ + clear_tail(d, oprsz, simd_maxsz(desc)); \ +} + +DO_SRI(gvec_sri_b, uint8_t) +DO_SRI(gvec_sri_h, uint16_t) +DO_SRI(gvec_sri_s, uint32_t) +DO_SRI(gvec_sri_d, uint64_t) + +#undef DO_SRI + +#define DO_SLI(NAME, TYPE) \ +void HELPER(NAME)(void *vd, void *vn, uint32_t desc) \ +{ \ + intptr_t i, oprsz = simd_oprsz(desc); \ + int shift = simd_data(desc); \ + TYPE *d = vd, *n = vn; \ + for (i = 0; i < oprsz / sizeof(TYPE); i++) { \ + d[i] = deposit64(d[i], shift, sizeof(TYPE) * 8 - shift, n[i]); \ + } \ + clear_tail(d, oprsz, simd_maxsz(desc)); \ +} + +DO_SLI(gvec_sli_b, uint8_t) +DO_SLI(gvec_sli_h, uint16_t) +DO_SLI(gvec_sli_s, uint32_t) +DO_SLI(gvec_sli_d, uint64_t) + +#undef DO_SLI + /* * Convert float16 to float32, raising no exceptions and * preserving exceptional values, including SNaN. -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> In 1dc8425e551, while converting to gvec, I added an extra range check against the shift count. This was unnecessary because the encoding of the shift count produces 0 to the element size - 1. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-5-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/translate.c | 12 ++---------- 1 file changed, 2 insertions(+), 10 deletions(-) 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_neon_data_insn(DisasContext *s, uint32_t insn) gen_gvec_sli(size, rd_ofs, rm_ofs, shift, vec_size, vec_size); } else { /* VSHL */ - /* Shifts larger than the element size are - * architecturally valid and results in zero. - */ - if (shift >= 8 << size) { - tcg_gen_gvec_dup_imm(size, rd_ofs, - vec_size, vec_size, 0); - } else { - tcg_gen_gvec_shli(size, rd_ofs, rm_ofs, shift, - vec_size, vec_size); - } + tcg_gen_gvec_shli(size, rd_ofs, rm_ofs, shift, + vec_size, vec_size); } return 0; } -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Now that we've converted all cases to gvec, there is quite a bit of dead code at the end of the function. Remove it. Sink the call to gen_gvec_fn2i to the end, loading a function pointer within the switch statement. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-6-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/translate-a64.c | 56 ++++++++++---------------------------- 1 file changed, 14 insertions(+), 42 deletions(-) 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_vec_simd_shri(DisasContext *s, bool is_q, bool is_u, int size = 32 - clz32(immh) - 1; int immhb = immh << 3 | immb; int shift = 2 * (8 << size) - immhb; - bool accumulate = false; - int dsize = is_q ? 128 : 64; - int esize = 8 << size; - int elements = dsize/esize; - MemOp memop = size | (is_u ? 0 : MO_SIGN); - TCGv_i64 tcg_rn = new_tmp_a64(s); - TCGv_i64 tcg_rd = new_tmp_a64(s); - TCGv_i64 tcg_round; - uint64_t round_const; - int i; + GVecGen2iFn *gvec_fn; if (extract32(immh, 3, 1) && !is_q) { unallocated_encoding(s); @@ -XXX,XX +XXX,XX @@ static void handle_vec_simd_shri(DisasContext *s, bool is_q, bool is_u, switch (opcode) { case 0x02: /* SSRA / USRA (accumulate) */ - gen_gvec_fn2i(s, is_q, rd, rn, shift, - is_u ? gen_gvec_usra : gen_gvec_ssra, size); - return; + gvec_fn = is_u ? gen_gvec_usra : gen_gvec_ssra; + break; case 0x08: /* SRI */ - gen_gvec_fn2i(s, is_q, rd, rn, shift, gen_gvec_sri, size); - return; + gvec_fn = gen_gvec_sri; + break; case 0x00: /* SSHR / USHR */ if (is_u) { @@ -XXX,XX +XXX,XX @@ static void handle_vec_simd_shri(DisasContext *s, bool is_q, bool is_u, /* Shift count the same size as element size produces zero. */ tcg_gen_gvec_dup_imm(size, vec_full_reg_offset(s, rd), is_q ? 16 : 8, vec_full_reg_size(s), 0); - } else { - gen_gvec_fn2i(s, is_q, rd, rn, shift, tcg_gen_gvec_shri, size); + return; } + gvec_fn = tcg_gen_gvec_shri; } else { /* Shift count the same size as element size produces all sign. */ if (shift == 8 << size) { shift -= 1; } - gen_gvec_fn2i(s, is_q, rd, rn, shift, tcg_gen_gvec_sari, size); + gvec_fn = tcg_gen_gvec_sari; } - return; + break; case 0x04: /* SRSHR / URSHR (rounding) */ - gen_gvec_fn2i(s, is_q, rd, rn, shift, - is_u ? gen_gvec_urshr : gen_gvec_srshr, size); - return; + gvec_fn = is_u ? gen_gvec_urshr : gen_gvec_srshr; + break; case 0x06: /* SRSRA / URSRA (accum + rounding) */ - gen_gvec_fn2i(s, is_q, rd, rn, shift, - is_u ? gen_gvec_ursra : gen_gvec_srsra, size); - return; + gvec_fn = is_u ? gen_gvec_ursra : gen_gvec_srsra; + break; default: g_assert_not_reached(); } - round_const = 1ULL << (shift - 1); - tcg_round = tcg_const_i64(round_const); - - for (i = 0; i < elements; i++) { - read_vec_element(s, tcg_rn, rn, i, memop); - if (accumulate) { - read_vec_element(s, tcg_rd, rd, i, memop); - } - - handle_shri_with_rndacc(tcg_rd, tcg_rn, tcg_round, - accumulate, is_u, size, shift); - - write_vec_element(s, tcg_rd, rd, i, size); - } - tcg_temp_free_i64(tcg_round); - - clear_vec_high(s, is_q, rd); + gen_gvec_fn2i(s, is_q, rd, rn, shift, gvec_fn, size); } /* SHL/SLI - Vector shift left */ -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Provide a functional interface for the vector expansion. This fits better with the existing set of helpers that we provide for other operations. Macro-ize the 5 nearly identical comparisons. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-7-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/translate.h | 16 ++- target/arm/translate-a64.c | 22 ++-- target/arm/translate.c | 254 ++++++++----------------------------- 3 files changed, 74 insertions(+), 218 deletions(-) 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 @@ static inline void gen_swstep_exception(DisasContext *s, int isv, int ex) uint64_t vfp_expand_imm(int size, uint8_t imm8); /* Vector operations shared between ARM and AArch64. */ -extern const GVecGen2 ceq0_op[4]; -extern const GVecGen2 clt0_op[4]; -extern const GVecGen2 cgt0_op[4]; -extern const GVecGen2 cle0_op[4]; -extern const GVecGen2 cge0_op[4]; +void gen_gvec_ceq0(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_clt0(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_cgt0(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_cle0(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_cge0(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + uint32_t opr_sz, uint32_t max_sz); + extern const GVecGen3 mla_op[4]; extern const GVecGen3 mls_op[4]; extern const GVecGen3 cmtst_op[4]; 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 gen_gvec_fn4(DisasContext *s, bool is_q, int rd, int rn, int rm, is_q ? 16 : 8, vec_full_reg_size(s)); } -/* Expand a 2-operand AdvSIMD vector operation using an op descriptor. */ -static void gen_gvec_op2(DisasContext *s, bool is_q, int rd, - int rn, const GVecGen2 *gvec_op) -{ - tcg_gen_gvec_2(vec_full_reg_offset(s, rd), vec_full_reg_offset(s, rn), - is_q ? 16 : 8, vec_full_reg_size(s), gvec_op); -} - /* Expand a 3-operand AdvSIMD vector operation using an op descriptor. */ static void gen_gvec_op3(DisasContext *s, bool is_q, int rd, int rn, int rm, const GVecGen3 *gvec_op) @@ -XXX,XX +XXX,XX @@ static void disas_simd_two_reg_misc(DisasContext *s, uint32_t insn) } break; case 0x8: /* CMGT, CMGE */ - gen_gvec_op2(s, is_q, rd, rn, u ? &cge0_op[size] : &cgt0_op[size]); + if (u) { + gen_gvec_fn2(s, is_q, rd, rn, gen_gvec_cge0, size); + } else { + gen_gvec_fn2(s, is_q, rd, rn, gen_gvec_cgt0, size); + } return; case 0x9: /* CMEQ, CMLE */ - gen_gvec_op2(s, is_q, rd, rn, u ? &cle0_op[size] : &ceq0_op[size]); + if (u) { + gen_gvec_fn2(s, is_q, rd, rn, gen_gvec_cle0, size); + } else { + gen_gvec_fn2(s, is_q, rd, rn, gen_gvec_ceq0, size); + } return; case 0xa: /* CMLT */ - gen_gvec_op2(s, is_q, rd, rn, &clt0_op[size]); + gen_gvec_fn2(s, is_q, rd, rn, gen_gvec_clt0, size); return; case 0xb: if (u) { /* ABS, NEG */ 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 do_v81_helper(DisasContext *s, gen_helper_gvec_3_ptr *fn, return 1; } -static void gen_ceq0_i32(TCGv_i32 d, TCGv_i32 a) -{ - tcg_gen_setcondi_i32(TCG_COND_EQ, d, a, 0); - tcg_gen_neg_i32(d, d); -} - -static void gen_ceq0_i64(TCGv_i64 d, TCGv_i64 a) -{ - tcg_gen_setcondi_i64(TCG_COND_EQ, d, a, 0); - tcg_gen_neg_i64(d, d); -} - -static void gen_ceq0_vec(unsigned vece, TCGv_vec d, TCGv_vec a) -{ - TCGv_vec zero = tcg_const_zeros_vec_matching(d); - tcg_gen_cmp_vec(TCG_COND_EQ, vece, d, a, zero); - tcg_temp_free_vec(zero); -} +#define GEN_CMP0(NAME, COND) \ + static void gen_##NAME##0_i32(TCGv_i32 d, TCGv_i32 a) \ + { \ + tcg_gen_setcondi_i32(COND, d, a, 0); \ + tcg_gen_neg_i32(d, d); \ + } \ + static void gen_##NAME##0_i64(TCGv_i64 d, TCGv_i64 a) \ + { \ + tcg_gen_setcondi_i64(COND, d, a, 0); \ + tcg_gen_neg_i64(d, d); \ + } \ + static void gen_##NAME##0_vec(unsigned vece, TCGv_vec d, TCGv_vec a) \ + { \ + TCGv_vec zero = tcg_const_zeros_vec_matching(d); \ + tcg_gen_cmp_vec(COND, vece, d, a, zero); \ + tcg_temp_free_vec(zero); \ + } \ + void gen_gvec_##NAME##0(unsigned vece, uint32_t d, uint32_t m, \ + uint32_t opr_sz, uint32_t max_sz) \ + { \ + const GVecGen2 op[4] = { \ + { .fno = gen_helper_gvec_##NAME##0_b, \ + .fniv = gen_##NAME##0_vec, \ + .opt_opc = vecop_list_cmp, \ + .vece = MO_8 }, \ + { .fno = gen_helper_gvec_##NAME##0_h, \ + .fniv = gen_##NAME##0_vec, \ + .opt_opc = vecop_list_cmp, \ + .vece = MO_16 }, \ + { .fni4 = gen_##NAME##0_i32, \ + .fniv = gen_##NAME##0_vec, \ + .opt_opc = vecop_list_cmp, \ + .vece = MO_32 }, \ + { .fni8 = gen_##NAME##0_i64, \ + .fniv = gen_##NAME##0_vec, \ + .opt_opc = vecop_list_cmp, \ + .prefer_i64 = TCG_TARGET_REG_BITS == 64, \ + .vece = MO_64 }, \ + }; \ + tcg_gen_gvec_2(d, m, opr_sz, max_sz, &op[vece]); \ + } static const TCGOpcode vecop_list_cmp[] = { INDEX_op_cmp_vec, 0 }; -const GVecGen2 ceq0_op[4] = { - { .fno = gen_helper_gvec_ceq0_b, - .fniv = gen_ceq0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_8 }, - { .fno = gen_helper_gvec_ceq0_h, - .fniv = gen_ceq0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_16 }, - { .fni4 = gen_ceq0_i32, - .fniv = gen_ceq0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_32 }, - { .fni8 = gen_ceq0_i64, - .fniv = gen_ceq0_vec, - .opt_opc = vecop_list_cmp, - .prefer_i64 = TCG_TARGET_REG_BITS == 64, - .vece = MO_64 }, -}; +GEN_CMP0(ceq, TCG_COND_EQ) +GEN_CMP0(cle, TCG_COND_LE) +GEN_CMP0(cge, TCG_COND_GE) +GEN_CMP0(clt, TCG_COND_LT) +GEN_CMP0(cgt, TCG_COND_GT) -static void gen_cle0_i32(TCGv_i32 d, TCGv_i32 a) -{ - tcg_gen_setcondi_i32(TCG_COND_LE, d, a, 0); - tcg_gen_neg_i32(d, d); -} - -static void gen_cle0_i64(TCGv_i64 d, TCGv_i64 a) -{ - tcg_gen_setcondi_i64(TCG_COND_LE, d, a, 0); - tcg_gen_neg_i64(d, d); -} - -static void gen_cle0_vec(unsigned vece, TCGv_vec d, TCGv_vec a) -{ - TCGv_vec zero = tcg_const_zeros_vec_matching(d); - tcg_gen_cmp_vec(TCG_COND_LE, vece, d, a, zero); - tcg_temp_free_vec(zero); -} - -const GVecGen2 cle0_op[4] = { - { .fno = gen_helper_gvec_cle0_b, - .fniv = gen_cle0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_8 }, - { .fno = gen_helper_gvec_cle0_h, - .fniv = gen_cle0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_16 }, - { .fni4 = gen_cle0_i32, - .fniv = gen_cle0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_32 }, - { .fni8 = gen_cle0_i64, - .fniv = gen_cle0_vec, - .opt_opc = vecop_list_cmp, - .prefer_i64 = TCG_TARGET_REG_BITS == 64, - .vece = MO_64 }, -}; - -static void gen_cge0_i32(TCGv_i32 d, TCGv_i32 a) -{ - tcg_gen_setcondi_i32(TCG_COND_GE, d, a, 0); - tcg_gen_neg_i32(d, d); -} - -static void gen_cge0_i64(TCGv_i64 d, TCGv_i64 a) -{ - tcg_gen_setcondi_i64(TCG_COND_GE, d, a, 0); - tcg_gen_neg_i64(d, d); -} - -static void gen_cge0_vec(unsigned vece, TCGv_vec d, TCGv_vec a) -{ - TCGv_vec zero = tcg_const_zeros_vec_matching(d); - tcg_gen_cmp_vec(TCG_COND_GE, vece, d, a, zero); - tcg_temp_free_vec(zero); -} - -const GVecGen2 cge0_op[4] = { - { .fno = gen_helper_gvec_cge0_b, - .fniv = gen_cge0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_8 }, - { .fno = gen_helper_gvec_cge0_h, - .fniv = gen_cge0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_16 }, - { .fni4 = gen_cge0_i32, - .fniv = gen_cge0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_32 }, - { .fni8 = gen_cge0_i64, - .fniv = gen_cge0_vec, - .opt_opc = vecop_list_cmp, - .prefer_i64 = TCG_TARGET_REG_BITS == 64, - .vece = MO_64 }, -}; - -static void gen_clt0_i32(TCGv_i32 d, TCGv_i32 a) -{ - tcg_gen_setcondi_i32(TCG_COND_LT, d, a, 0); - tcg_gen_neg_i32(d, d); -} - -static void gen_clt0_i64(TCGv_i64 d, TCGv_i64 a) -{ - tcg_gen_setcondi_i64(TCG_COND_LT, d, a, 0); - tcg_gen_neg_i64(d, d); -} - -static void gen_clt0_vec(unsigned vece, TCGv_vec d, TCGv_vec a) -{ - TCGv_vec zero = tcg_const_zeros_vec_matching(d); - tcg_gen_cmp_vec(TCG_COND_LT, vece, d, a, zero); - tcg_temp_free_vec(zero); -} - -const GVecGen2 clt0_op[4] = { - { .fno = gen_helper_gvec_clt0_b, - .fniv = gen_clt0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_8 }, - { .fno = gen_helper_gvec_clt0_h, - .fniv = gen_clt0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_16 }, - { .fni4 = gen_clt0_i32, - .fniv = gen_clt0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_32 }, - { .fni8 = gen_clt0_i64, - .fniv = gen_clt0_vec, - .opt_opc = vecop_list_cmp, - .prefer_i64 = TCG_TARGET_REG_BITS == 64, - .vece = MO_64 }, -}; - -static void gen_cgt0_i32(TCGv_i32 d, TCGv_i32 a) -{ - tcg_gen_setcondi_i32(TCG_COND_GT, d, a, 0); - tcg_gen_neg_i32(d, d); -} - -static void gen_cgt0_i64(TCGv_i64 d, TCGv_i64 a) -{ - tcg_gen_setcondi_i64(TCG_COND_GT, d, a, 0); - tcg_gen_neg_i64(d, d); -} - -static void gen_cgt0_vec(unsigned vece, TCGv_vec d, TCGv_vec a) -{ - TCGv_vec zero = tcg_const_zeros_vec_matching(d); - tcg_gen_cmp_vec(TCG_COND_GT, vece, d, a, zero); - tcg_temp_free_vec(zero); -} - -const GVecGen2 cgt0_op[4] = { - { .fno = gen_helper_gvec_cgt0_b, - .fniv = gen_cgt0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_8 }, - { .fno = gen_helper_gvec_cgt0_h, - .fniv = gen_cgt0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_16 }, - { .fni4 = gen_cgt0_i32, - .fniv = gen_cgt0_vec, - .opt_opc = vecop_list_cmp, - .vece = MO_32 }, - { .fni8 = gen_cgt0_i64, - .fniv = gen_cgt0_vec, - .opt_opc = vecop_list_cmp, - .prefer_i64 = TCG_TARGET_REG_BITS == 64, - .vece = MO_64 }, -}; +#undef GEN_CMP0 static void gen_ssra8_i64(TCGv_i64 d, TCGv_i64 a, int64_t shift) { @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) break; case NEON_2RM_VCEQ0: - tcg_gen_gvec_2(rd_ofs, rm_ofs, vec_size, - vec_size, &ceq0_op[size]); + gen_gvec_ceq0(size, rd_ofs, rm_ofs, vec_size, vec_size); break; case NEON_2RM_VCGT0: - tcg_gen_gvec_2(rd_ofs, rm_ofs, vec_size, - vec_size, &cgt0_op[size]); + gen_gvec_cgt0(size, rd_ofs, rm_ofs, vec_size, vec_size); break; case NEON_2RM_VCLE0: - tcg_gen_gvec_2(rd_ofs, rm_ofs, vec_size, - vec_size, &cle0_op[size]); + gen_gvec_cle0(size, rd_ofs, rm_ofs, vec_size, vec_size); break; case NEON_2RM_VCGE0: - tcg_gen_gvec_2(rd_ofs, rm_ofs, vec_size, - vec_size, &cge0_op[size]); + gen_gvec_cge0(size, rd_ofs, rm_ofs, vec_size, vec_size); break; case NEON_2RM_VCLT0: - tcg_gen_gvec_2(rd_ofs, rm_ofs, vec_size, - vec_size, &clt0_op[size]); + gen_gvec_clt0(size, rd_ofs, rm_ofs, vec_size, vec_size); break; default: -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Provide a functional interface for the vector expansion. This fits better with the existing set of helpers that we provide for other operations. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-8-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/translate.h | 7 +- target/arm/translate-a64.c | 4 +- target/arm/translate-neon.inc.c | 16 +---- target/arm/translate.c | 117 +++++++++++++++++--------------- 4 files changed, 71 insertions(+), 73 deletions(-) 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 @@ void gen_gvec_cle0(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, void gen_gvec_cge0(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); -extern const GVecGen3 mla_op[4]; -extern const GVecGen3 mls_op[4]; +void gen_gvec_mla(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_mls(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); + extern const GVecGen3 cmtst_op[4]; extern const GVecGen3 sshl_op[4]; extern const GVecGen3 ushl_op[4]; 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 disas_simd_3same_int(DisasContext *s, uint32_t insn) return; case 0x12: /* MLA, MLS */ if (u) { - gen_gvec_op3(s, is_q, rd, rn, rm, &mls_op[size]); + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_mls, size); } else { - gen_gvec_op3(s, is_q, rd, rn, rm, &mla_op[size]); + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_mla, size); } return; case 0x11: diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME_NO_SZ_3(VMAX_U, tcg_gen_gvec_umax) DO_3SAME_NO_SZ_3(VMIN_S, tcg_gen_gvec_smin) DO_3SAME_NO_SZ_3(VMIN_U, tcg_gen_gvec_umin) DO_3SAME_NO_SZ_3(VMUL, tcg_gen_gvec_mul) +DO_3SAME_NO_SZ_3(VMLA, gen_gvec_mla) +DO_3SAME_NO_SZ_3(VMLS, gen_gvec_mls) #define DO_3SAME_CMP(INSN, COND) \ static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ @@ -XXX,XX +XXX,XX @@ static bool trans_VMUL_p_3s(DisasContext *s, arg_3same *a) return do_3same(s, a, gen_VMUL_p_3s); } -#define DO_3SAME_GVEC3_NO_SZ_3(INSN, OPARRAY) \ - static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ - uint32_t rn_ofs, uint32_t rm_ofs, \ - uint32_t oprsz, uint32_t maxsz) \ - { \ - tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, \ - oprsz, maxsz, &OPARRAY[vece]); \ - } \ - DO_3SAME_NO_SZ_3(INSN, gen_##INSN##_3s) - - -DO_3SAME_GVEC3_NO_SZ_3(VMLA, mla_op) -DO_3SAME_GVEC3_NO_SZ_3(VMLS, mls_op) - #define DO_3SAME_GVEC3_SHIFT(INSN, OPARRAY) \ static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ uint32_t rn_ofs, uint32_t rm_ofs, \ diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_mls_vec(unsigned vece, TCGv_vec d, TCGv_vec a, TCGv_vec b) /* Note that while NEON does not support VMLA and VMLS as 64-bit ops, * these tables are shared with AArch64 which does support them. */ +void gen_gvec_mla(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_mul_vec, INDEX_op_add_vec, 0 + }; + static const GVecGen3 ops[4] = { + { .fni4 = gen_mla8_i32, + .fniv = gen_mla_vec, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fni4 = gen_mla16_i32, + .fniv = gen_mla_vec, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = gen_mla32_i32, + .fniv = gen_mla_vec, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = gen_mla64_i64, + .fniv = gen_mla_vec, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_64 }, + }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} -static const TCGOpcode vecop_list_mla[] = { - INDEX_op_mul_vec, INDEX_op_add_vec, 0 -}; - -static const TCGOpcode vecop_list_mls[] = { - INDEX_op_mul_vec, INDEX_op_sub_vec, 0 -}; - -const GVecGen3 mla_op[4] = { - { .fni4 = gen_mla8_i32, - .fniv = gen_mla_vec, - .load_dest = true, - .opt_opc = vecop_list_mla, - .vece = MO_8 }, - { .fni4 = gen_mla16_i32, - .fniv = gen_mla_vec, - .load_dest = true, - .opt_opc = vecop_list_mla, - .vece = MO_16 }, - { .fni4 = gen_mla32_i32, - .fniv = gen_mla_vec, - .load_dest = true, - .opt_opc = vecop_list_mla, - .vece = MO_32 }, - { .fni8 = gen_mla64_i64, - .fniv = gen_mla_vec, - .prefer_i64 = TCG_TARGET_REG_BITS == 64, - .load_dest = true, - .opt_opc = vecop_list_mla, - .vece = MO_64 }, -}; - -const GVecGen3 mls_op[4] = { - { .fni4 = gen_mls8_i32, - .fniv = gen_mls_vec, - .load_dest = true, - .opt_opc = vecop_list_mls, - .vece = MO_8 }, - { .fni4 = gen_mls16_i32, - .fniv = gen_mls_vec, - .load_dest = true, - .opt_opc = vecop_list_mls, - .vece = MO_16 }, - { .fni4 = gen_mls32_i32, - .fniv = gen_mls_vec, - .load_dest = true, - .opt_opc = vecop_list_mls, - .vece = MO_32 }, - { .fni8 = gen_mls64_i64, - .fniv = gen_mls_vec, - .prefer_i64 = TCG_TARGET_REG_BITS == 64, - .load_dest = true, - .opt_opc = vecop_list_mls, - .vece = MO_64 }, -}; +void gen_gvec_mls(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_mul_vec, INDEX_op_sub_vec, 0 + }; + static const GVecGen3 ops[4] = { + { .fni4 = gen_mls8_i32, + .fniv = gen_mls_vec, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fni4 = gen_mls16_i32, + .fniv = gen_mls_vec, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = gen_mls32_i32, + .fniv = gen_mls_vec, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = gen_mls64_i64, + .fniv = gen_mls_vec, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .load_dest = true, + .opt_opc = vecop_list, + .vece = MO_64 }, + }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} /* CMTST : test is "if (X & Y != 0)". */ static void gen_cmtst_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Rather than perform the argument swap during code generation, perform it during decode. This means it doesn't have to be special cased later, and we can share code with aarch64 code generation. Hopefully the decode comment addresses any confusion that might arise in between. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-9-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/neon-dp.decode | 17 +++++++++++++++-- target/arm/translate-neon.inc.c | 3 +-- 2 files changed, 16 insertions(+), 4 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VCGT_U_3s 1111 001 1 0 . .. .... .... 0011 . . . 0 .... @3same VCGE_S_3s 1111 001 0 0 . .. .... .... 0011 . . . 1 .... @3same VCGE_U_3s 1111 001 1 0 . .. .... .... 0011 . . . 1 .... @3same -VSHL_S_3s 1111 001 0 0 . .. .... .... 0100 . . . 0 .... @3same -VSHL_U_3s 1111 001 1 0 . .. .... .... 0100 . . . 0 .... @3same +# The _rev suffix indicates that Vn and Vm are reversed. This is +# the case for shifts. In the Arm ARM these insns are documented +# with the Vm and Vn fields in their usual places, but in the +# assembly the operands are listed "backwards", ie in the order +# Dd, Dm, Dn where other insns use Dd, Dn, Dm. For QEMU we choose +# to consider Vm and Vn as being in different fields in the insn, +# which allows us to avoid special-casing shifts in the trans_ +# function code. We would otherwise need to manually swap the operands +# over to call Neon helper functions that are shared with AArch64, +# which does not have this odd reversed-operand situation. +@3same_rev .... ... . . . size:2 .... .... .... . q:1 . . .... \ + &3same vn=%vm_dp vm=%vn_dp vd=%vd_dp + +VSHL_S_3s 1111 001 0 0 . .. .... .... 0100 . . . 0 .... @3same_rev +VSHL_U_3s 1111 001 1 0 . .. .... .... 0100 . . . 0 .... @3same_rev VMAX_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 0 .... @3same VMAX_U_3s 1111 001 1 0 . .. .... .... 0110 . . . 0 .... @3same diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static bool trans_VMUL_p_3s(DisasContext *s, arg_3same *a) uint32_t rn_ofs, uint32_t rm_ofs, \ uint32_t oprsz, uint32_t maxsz) \ { \ - /* Note the operation is vshl vd,vm,vn */ \ - tcg_gen_gvec_3(rd_ofs, rm_ofs, rn_ofs, \ + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, \ oprsz, maxsz, &OPARRAY[vece]); \ } \ DO_3SAME(INSN, gen_##INSN##_3s) -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Provide a functional interface for the vector expansion. This fits better with the existing set of helpers that we provide for other operations. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-10-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/translate.h | 10 ++- target/arm/translate-a64.c | 18 ++-- target/arm/translate-neon.inc.c | 23 +---- target/arm/translate.c | 146 +++++++++++++++++--------------- 4 files changed, 95 insertions(+), 102 deletions(-) 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 @@ void gen_gvec_mla(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, void gen_gvec_mls(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); -extern const GVecGen3 cmtst_op[4]; -extern const GVecGen3 sshl_op[4]; -extern const GVecGen3 ushl_op[4]; +void gen_gvec_cmtst(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_sshl(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_ushl(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); + extern const GVecGen4 uqadd_op[4]; extern const GVecGen4 sqadd_op[4]; extern const GVecGen4 uqsub_op[4]; 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 gen_gvec_fn4(DisasContext *s, bool is_q, int rd, int rn, int rm, is_q ? 16 : 8, vec_full_reg_size(s)); } -/* Expand a 3-operand AdvSIMD vector operation using an op descriptor. */ -static void gen_gvec_op3(DisasContext *s, bool is_q, int rd, - int rn, int rm, const GVecGen3 *gvec_op) -{ - tcg_gen_gvec_3(vec_full_reg_offset(s, rd), vec_full_reg_offset(s, rn), - vec_full_reg_offset(s, rm), is_q ? 16 : 8, - vec_full_reg_size(s), gvec_op); -} - /* Expand a 3-operand operation using an out-of-line helper. */ static void gen_gvec_op3_ool(DisasContext *s, bool is_q, int rd, int rn, int rm, int data, gen_helper_gvec_3 *fn) @@ -XXX,XX +XXX,XX @@ static void disas_simd_3same_int(DisasContext *s, uint32_t insn) (u ? uqsub_op : sqsub_op) + size); return; case 0x08: /* SSHL, USHL */ - gen_gvec_op3(s, is_q, rd, rn, rm, - u ? &ushl_op[size] : &sshl_op[size]); + if (u) { + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_ushl, size); + } else { + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_sshl, size); + } return; case 0x0c: /* SMAX, UMAX */ if (u) { @@ -XXX,XX +XXX,XX @@ static void disas_simd_3same_int(DisasContext *s, uint32_t insn) return; case 0x11: if (!u) { /* CMTST */ - gen_gvec_op3(s, is_q, rd, rn, rm, &cmtst_op[size]); + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_cmtst, size); return; } /* else CMEQ */ diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME(VBIC, tcg_gen_gvec_andc) DO_3SAME(VORR, tcg_gen_gvec_or) DO_3SAME(VORN, tcg_gen_gvec_orc) DO_3SAME(VEOR, tcg_gen_gvec_xor) +DO_3SAME(VSHL_S, gen_gvec_sshl) +DO_3SAME(VSHL_U, gen_gvec_ushl) /* These insns are all gvec_bitsel but with the inputs in various orders. */ #define DO_3SAME_BITSEL(INSN, O1, O2, O3) \ @@ -XXX,XX +XXX,XX @@ DO_3SAME_NO_SZ_3(VMIN_U, tcg_gen_gvec_umin) DO_3SAME_NO_SZ_3(VMUL, tcg_gen_gvec_mul) DO_3SAME_NO_SZ_3(VMLA, gen_gvec_mla) DO_3SAME_NO_SZ_3(VMLS, gen_gvec_mls) +DO_3SAME_NO_SZ_3(VTST, gen_gvec_cmtst) #define DO_3SAME_CMP(INSN, COND) \ static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ @@ -XXX,XX +XXX,XX @@ DO_3SAME_CMP(VCGE_S, TCG_COND_GE) DO_3SAME_CMP(VCGE_U, TCG_COND_GEU) DO_3SAME_CMP(VCEQ, TCG_COND_EQ) -static void gen_VTST_3s(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, - uint32_t rm_ofs, uint32_t oprsz, uint32_t maxsz) -{ - tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, &cmtst_op[vece]); -} -DO_3SAME_NO_SZ_3(VTST, gen_VTST_3s) - #define DO_3SAME_GVEC4(INSN, OPARRAY) \ static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ uint32_t rn_ofs, uint32_t rm_ofs, \ @@ -XXX,XX +XXX,XX @@ static bool trans_VMUL_p_3s(DisasContext *s, arg_3same *a) } return do_3same(s, a, gen_VMUL_p_3s); } - -#define DO_3SAME_GVEC3_SHIFT(INSN, OPARRAY) \ - static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ - uint32_t rn_ofs, uint32_t rm_ofs, \ - uint32_t oprsz, uint32_t maxsz) \ - { \ - tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, \ - oprsz, maxsz, &OPARRAY[vece]); \ - } \ - DO_3SAME(INSN, gen_##INSN##_3s) - -DO_3SAME_GVEC3_SHIFT(VSHL_S, sshl_op) -DO_3SAME_GVEC3_SHIFT(VSHL_U, ushl_op) diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_cmtst_vec(unsigned vece, TCGv_vec d, TCGv_vec a, TCGv_vec b) tcg_gen_cmp_vec(TCG_COND_NE, vece, d, d, a); } -static const TCGOpcode vecop_list_cmtst[] = { INDEX_op_cmp_vec, 0 }; - -const GVecGen3 cmtst_op[4] = { - { .fni4 = gen_helper_neon_tst_u8, - .fniv = gen_cmtst_vec, - .opt_opc = vecop_list_cmtst, - .vece = MO_8 }, - { .fni4 = gen_helper_neon_tst_u16, - .fniv = gen_cmtst_vec, - .opt_opc = vecop_list_cmtst, - .vece = MO_16 }, - { .fni4 = gen_cmtst_i32, - .fniv = gen_cmtst_vec, - .opt_opc = vecop_list_cmtst, - .vece = MO_32 }, - { .fni8 = gen_cmtst_i64, - .fniv = gen_cmtst_vec, - .prefer_i64 = TCG_TARGET_REG_BITS == 64, - .opt_opc = vecop_list_cmtst, - .vece = MO_64 }, -}; +void gen_gvec_cmtst(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { INDEX_op_cmp_vec, 0 }; + static const GVecGen3 ops[4] = { + { .fni4 = gen_helper_neon_tst_u8, + .fniv = gen_cmtst_vec, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fni4 = gen_helper_neon_tst_u16, + .fniv = gen_cmtst_vec, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = gen_cmtst_i32, + .fniv = gen_cmtst_vec, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = gen_cmtst_i64, + .fniv = gen_cmtst_vec, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .opt_opc = vecop_list, + .vece = MO_64 }, + }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} void gen_ushl_i32(TCGv_i32 dst, TCGv_i32 src, TCGv_i32 shift) { @@ -XXX,XX +XXX,XX @@ static void gen_ushl_vec(unsigned vece, TCGv_vec dst, tcg_temp_free_vec(rsh); } -static const TCGOpcode ushl_list[] = { - INDEX_op_neg_vec, INDEX_op_shlv_vec, - INDEX_op_shrv_vec, INDEX_op_cmp_vec, 0 -}; - -const GVecGen3 ushl_op[4] = { - { .fniv = gen_ushl_vec, - .fno = gen_helper_gvec_ushl_b, - .opt_opc = ushl_list, - .vece = MO_8 }, - { .fniv = gen_ushl_vec, - .fno = gen_helper_gvec_ushl_h, - .opt_opc = ushl_list, - .vece = MO_16 }, - { .fni4 = gen_ushl_i32, - .fniv = gen_ushl_vec, - .opt_opc = ushl_list, - .vece = MO_32 }, - { .fni8 = gen_ushl_i64, - .fniv = gen_ushl_vec, - .opt_opc = ushl_list, - .vece = MO_64 }, -}; +void gen_gvec_ushl(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_neg_vec, INDEX_op_shlv_vec, + INDEX_op_shrv_vec, INDEX_op_cmp_vec, 0 + }; + static const GVecGen3 ops[4] = { + { .fniv = gen_ushl_vec, + .fno = gen_helper_gvec_ushl_b, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fniv = gen_ushl_vec, + .fno = gen_helper_gvec_ushl_h, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = gen_ushl_i32, + .fniv = gen_ushl_vec, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = gen_ushl_i64, + .fniv = gen_ushl_vec, + .opt_opc = vecop_list, + .vece = MO_64 }, + }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} void gen_sshl_i32(TCGv_i32 dst, TCGv_i32 src, TCGv_i32 shift) { @@ -XXX,XX +XXX,XX @@ static void gen_sshl_vec(unsigned vece, TCGv_vec dst, tcg_temp_free_vec(tmp); } -static const TCGOpcode sshl_list[] = { - INDEX_op_neg_vec, INDEX_op_umin_vec, INDEX_op_shlv_vec, - INDEX_op_sarv_vec, INDEX_op_cmp_vec, INDEX_op_cmpsel_vec, 0 -}; - -const GVecGen3 sshl_op[4] = { - { .fniv = gen_sshl_vec, - .fno = gen_helper_gvec_sshl_b, - .opt_opc = sshl_list, - .vece = MO_8 }, - { .fniv = gen_sshl_vec, - .fno = gen_helper_gvec_sshl_h, - .opt_opc = sshl_list, - .vece = MO_16 }, - { .fni4 = gen_sshl_i32, - .fniv = gen_sshl_vec, - .opt_opc = sshl_list, - .vece = MO_32 }, - { .fni8 = gen_sshl_i64, - .fniv = gen_sshl_vec, - .opt_opc = sshl_list, - .vece = MO_64 }, -}; +void gen_gvec_sshl(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_neg_vec, INDEX_op_umin_vec, INDEX_op_shlv_vec, + INDEX_op_sarv_vec, INDEX_op_cmp_vec, INDEX_op_cmpsel_vec, 0 + }; + static const GVecGen3 ops[4] = { + { .fniv = gen_sshl_vec, + .fno = gen_helper_gvec_sshl_b, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fniv = gen_sshl_vec, + .fno = gen_helper_gvec_sshl_h, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = gen_sshl_i32, + .fniv = gen_sshl_vec, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = gen_sshl_i64, + .fniv = gen_sshl_vec, + .opt_opc = vecop_list, + .vece = MO_64 }, + }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} static void gen_uqadd_vec(unsigned vece, TCGv_vec t, TCGv_vec sat, TCGv_vec a, TCGv_vec b) -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Provide a functional interface for the vector expansion. This fits better with the existing set of helpers that we provide for other operations. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-11-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/translate.h | 13 +- target/arm/translate-a64.c | 22 ++- target/arm/translate-neon.inc.c | 19 +-- target/arm/translate.c | 228 +++++++++++++++++--------------- 4 files changed, 147 insertions(+), 135 deletions(-) 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 @@ void gen_gvec_sshl(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, void gen_gvec_ushl(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); -extern const GVecGen4 uqadd_op[4]; -extern const GVecGen4 sqadd_op[4]; -extern const GVecGen4 uqsub_op[4]; -extern const GVecGen4 sqsub_op[4]; void gen_cmtst_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b); void gen_ushl_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b); void gen_sshl_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b); void gen_ushl_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b); void gen_sshl_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b); +void gen_gvec_uqadd_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_sqadd_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_uqsub_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_sqsub_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); + void gen_gvec_ssra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, int64_t shift, uint32_t opr_sz, uint32_t max_sz); void gen_gvec_usra(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, 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 disas_simd_3same_int(DisasContext *s, uint32_t insn) switch (opcode) { case 0x01: /* SQADD, UQADD */ - tcg_gen_gvec_4(vec_full_reg_offset(s, rd), - offsetof(CPUARMState, vfp.qc), - vec_full_reg_offset(s, rn), - vec_full_reg_offset(s, rm), - is_q ? 16 : 8, vec_full_reg_size(s), - (u ? uqadd_op : sqadd_op) + size); + if (u) { + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_uqadd_qc, size); + } else { + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_sqadd_qc, size); + } return; case 0x05: /* SQSUB, UQSUB */ - tcg_gen_gvec_4(vec_full_reg_offset(s, rd), - offsetof(CPUARMState, vfp.qc), - vec_full_reg_offset(s, rn), - vec_full_reg_offset(s, rm), - is_q ? 16 : 8, vec_full_reg_size(s), - (u ? uqsub_op : sqsub_op) + size); + if (u) { + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_uqsub_qc, size); + } else { + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_sqsub_qc, size); + } return; case 0x08: /* SSHL, USHL */ if (u) { diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME(VORN, tcg_gen_gvec_orc) DO_3SAME(VEOR, tcg_gen_gvec_xor) DO_3SAME(VSHL_S, gen_gvec_sshl) DO_3SAME(VSHL_U, gen_gvec_ushl) +DO_3SAME(VQADD_S, gen_gvec_sqadd_qc) +DO_3SAME(VQADD_U, gen_gvec_uqadd_qc) +DO_3SAME(VQSUB_S, gen_gvec_sqsub_qc) +DO_3SAME(VQSUB_U, gen_gvec_uqsub_qc) /* These insns are all gvec_bitsel but with the inputs in various orders. */ #define DO_3SAME_BITSEL(INSN, O1, O2, O3) \ @@ -XXX,XX +XXX,XX @@ DO_3SAME_CMP(VCGE_S, TCG_COND_GE) DO_3SAME_CMP(VCGE_U, TCG_COND_GEU) DO_3SAME_CMP(VCEQ, TCG_COND_EQ) -#define DO_3SAME_GVEC4(INSN, OPARRAY) \ - static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ - uint32_t rn_ofs, uint32_t rm_ofs, \ - uint32_t oprsz, uint32_t maxsz) \ - { \ - tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc), \ - rn_ofs, rm_ofs, oprsz, maxsz, &OPARRAY[vece]); \ - } \ - DO_3SAME(INSN, gen_##INSN##_3s) - -DO_3SAME_GVEC4(VQADD_S, sqadd_op) -DO_3SAME_GVEC4(VQADD_U, uqadd_op) -DO_3SAME_GVEC4(VQSUB_S, sqsub_op) -DO_3SAME_GVEC4(VQSUB_U, uqsub_op) - static void gen_VMUL_p_3s(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, uint32_t rm_ofs, uint32_t oprsz, uint32_t maxsz) { diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_uqadd_vec(unsigned vece, TCGv_vec t, TCGv_vec sat, tcg_temp_free_vec(x); } -static const TCGOpcode vecop_list_uqadd[] = { - INDEX_op_usadd_vec, INDEX_op_cmp_vec, INDEX_op_add_vec, 0 -}; - -const GVecGen4 uqadd_op[4] = { - { .fniv = gen_uqadd_vec, - .fno = gen_helper_gvec_uqadd_b, - .write_aofs = true, - .opt_opc = vecop_list_uqadd, - .vece = MO_8 }, - { .fniv = gen_uqadd_vec, - .fno = gen_helper_gvec_uqadd_h, - .write_aofs = true, - .opt_opc = vecop_list_uqadd, - .vece = MO_16 }, - { .fniv = gen_uqadd_vec, - .fno = gen_helper_gvec_uqadd_s, - .write_aofs = true, - .opt_opc = vecop_list_uqadd, - .vece = MO_32 }, - { .fniv = gen_uqadd_vec, - .fno = gen_helper_gvec_uqadd_d, - .write_aofs = true, - .opt_opc = vecop_list_uqadd, - .vece = MO_64 }, -}; +void gen_gvec_uqadd_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_usadd_vec, INDEX_op_cmp_vec, INDEX_op_add_vec, 0 + }; + static const GVecGen4 ops[4] = { + { .fniv = gen_uqadd_vec, + .fno = gen_helper_gvec_uqadd_b, + .write_aofs = true, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fniv = gen_uqadd_vec, + .fno = gen_helper_gvec_uqadd_h, + .write_aofs = true, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fniv = gen_uqadd_vec, + .fno = gen_helper_gvec_uqadd_s, + .write_aofs = true, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fniv = gen_uqadd_vec, + .fno = gen_helper_gvec_uqadd_d, + .write_aofs = true, + .opt_opc = vecop_list, + .vece = MO_64 }, + }; + tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc), + rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} static void gen_sqadd_vec(unsigned vece, TCGv_vec t, TCGv_vec sat, TCGv_vec a, TCGv_vec b) @@ -XXX,XX +XXX,XX @@ static void gen_sqadd_vec(unsigned vece, TCGv_vec t, TCGv_vec sat, tcg_temp_free_vec(x); } -static const TCGOpcode vecop_list_sqadd[] = { - INDEX_op_ssadd_vec, INDEX_op_cmp_vec, INDEX_op_add_vec, 0 -}; - -const GVecGen4 sqadd_op[4] = { - { .fniv = gen_sqadd_vec, - .fno = gen_helper_gvec_sqadd_b, - .opt_opc = vecop_list_sqadd, - .write_aofs = true, - .vece = MO_8 }, - { .fniv = gen_sqadd_vec, - .fno = gen_helper_gvec_sqadd_h, - .opt_opc = vecop_list_sqadd, - .write_aofs = true, - .vece = MO_16 }, - { .fniv = gen_sqadd_vec, - .fno = gen_helper_gvec_sqadd_s, - .opt_opc = vecop_list_sqadd, - .write_aofs = true, - .vece = MO_32 }, - { .fniv = gen_sqadd_vec, - .fno = gen_helper_gvec_sqadd_d, - .opt_opc = vecop_list_sqadd, - .write_aofs = true, - .vece = MO_64 }, -}; +void gen_gvec_sqadd_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_ssadd_vec, INDEX_op_cmp_vec, INDEX_op_add_vec, 0 + }; + static const GVecGen4 ops[4] = { + { .fniv = gen_sqadd_vec, + .fno = gen_helper_gvec_sqadd_b, + .opt_opc = vecop_list, + .write_aofs = true, + .vece = MO_8 }, + { .fniv = gen_sqadd_vec, + .fno = gen_helper_gvec_sqadd_h, + .opt_opc = vecop_list, + .write_aofs = true, + .vece = MO_16 }, + { .fniv = gen_sqadd_vec, + .fno = gen_helper_gvec_sqadd_s, + .opt_opc = vecop_list, + .write_aofs = true, + .vece = MO_32 }, + { .fniv = gen_sqadd_vec, + .fno = gen_helper_gvec_sqadd_d, + .opt_opc = vecop_list, + .write_aofs = true, + .vece = MO_64 }, + }; + tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc), + rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} static void gen_uqsub_vec(unsigned vece, TCGv_vec t, TCGv_vec sat, TCGv_vec a, TCGv_vec b) @@ -XXX,XX +XXX,XX @@ static void gen_uqsub_vec(unsigned vece, TCGv_vec t, TCGv_vec sat, tcg_temp_free_vec(x); } -static const TCGOpcode vecop_list_uqsub[] = { - INDEX_op_ussub_vec, INDEX_op_cmp_vec, INDEX_op_sub_vec, 0 -}; - -const GVecGen4 uqsub_op[4] = { - { .fniv = gen_uqsub_vec, - .fno = gen_helper_gvec_uqsub_b, - .opt_opc = vecop_list_uqsub, - .write_aofs = true, - .vece = MO_8 }, - { .fniv = gen_uqsub_vec, - .fno = gen_helper_gvec_uqsub_h, - .opt_opc = vecop_list_uqsub, - .write_aofs = true, - .vece = MO_16 }, - { .fniv = gen_uqsub_vec, - .fno = gen_helper_gvec_uqsub_s, - .opt_opc = vecop_list_uqsub, - .write_aofs = true, - .vece = MO_32 }, - { .fniv = gen_uqsub_vec, - .fno = gen_helper_gvec_uqsub_d, - .opt_opc = vecop_list_uqsub, - .write_aofs = true, - .vece = MO_64 }, -}; +void gen_gvec_uqsub_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_ussub_vec, INDEX_op_cmp_vec, INDEX_op_sub_vec, 0 + }; + static const GVecGen4 ops[4] = { + { .fniv = gen_uqsub_vec, + .fno = gen_helper_gvec_uqsub_b, + .opt_opc = vecop_list, + .write_aofs = true, + .vece = MO_8 }, + { .fniv = gen_uqsub_vec, + .fno = gen_helper_gvec_uqsub_h, + .opt_opc = vecop_list, + .write_aofs = true, + .vece = MO_16 }, + { .fniv = gen_uqsub_vec, + .fno = gen_helper_gvec_uqsub_s, + .opt_opc = vecop_list, + .write_aofs = true, + .vece = MO_32 }, + { .fniv = gen_uqsub_vec, + .fno = gen_helper_gvec_uqsub_d, + .opt_opc = vecop_list, + .write_aofs = true, + .vece = MO_64 }, + }; + tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc), + rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} static void gen_sqsub_vec(unsigned vece, TCGv_vec t, TCGv_vec sat, TCGv_vec a, TCGv_vec b) @@ -XXX,XX +XXX,XX @@ static void gen_sqsub_vec(unsigned vece, TCGv_vec t, TCGv_vec sat, tcg_temp_free_vec(x); } -static const TCGOpcode vecop_list_sqsub[] = { - INDEX_op_sssub_vec, INDEX_op_cmp_vec, INDEX_op_sub_vec, 0 -}; - -const GVecGen4 sqsub_op[4] = { - { .fniv = gen_sqsub_vec, - .fno = gen_helper_gvec_sqsub_b, - .opt_opc = vecop_list_sqsub, - .write_aofs = true, - .vece = MO_8 }, - { .fniv = gen_sqsub_vec, - .fno = gen_helper_gvec_sqsub_h, - .opt_opc = vecop_list_sqsub, - .write_aofs = true, - .vece = MO_16 }, - { .fniv = gen_sqsub_vec, - .fno = gen_helper_gvec_sqsub_s, - .opt_opc = vecop_list_sqsub, - .write_aofs = true, - .vece = MO_32 }, - { .fniv = gen_sqsub_vec, - .fno = gen_helper_gvec_sqsub_d, - .opt_opc = vecop_list_sqsub, - .write_aofs = true, - .vece = MO_64 }, -}; +void gen_gvec_sqsub_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_sssub_vec, INDEX_op_cmp_vec, INDEX_op_sub_vec, 0 + }; + static const GVecGen4 ops[4] = { + { .fniv = gen_sqsub_vec, + .fno = gen_helper_gvec_sqsub_b, + .opt_opc = vecop_list, + .write_aofs = true, + .vece = MO_8 }, + { .fniv = gen_sqsub_vec, + .fno = gen_helper_gvec_sqsub_h, + .opt_opc = vecop_list, + .write_aofs = true, + .vece = MO_16 }, + { .fniv = gen_sqsub_vec, + .fno = gen_helper_gvec_sqsub_s, + .opt_opc = vecop_list, + .write_aofs = true, + .vece = MO_32 }, + { .fniv = gen_sqsub_vec, + .fno = gen_helper_gvec_sqsub_d, + .opt_opc = vecop_list, + .write_aofs = true, + .vece = MO_64 }, + }; + tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc), + rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} /* Translate a NEON data processing instruction. Return nonzero if the instruction is invalid. -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> These operations do not touch fp_status. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-12-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 4 ++-- target/arm/translate-a64.c | 5 ++--- target/arm/translate.c | 12 ++---------- target/arm/vfp_helper.c | 5 ++--- 4 files changed, 8 insertions(+), 18 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(recpe_f64, TCG_CALL_NO_RWG, f64, f64, ptr) DEF_HELPER_FLAGS_2(rsqrte_f16, TCG_CALL_NO_RWG, f16, f16, ptr) DEF_HELPER_FLAGS_2(rsqrte_f32, TCG_CALL_NO_RWG, f32, f32, ptr) DEF_HELPER_FLAGS_2(rsqrte_f64, TCG_CALL_NO_RWG, f64, f64, ptr) -DEF_HELPER_2(recpe_u32, i32, i32, ptr) -DEF_HELPER_FLAGS_2(rsqrte_u32, TCG_CALL_NO_RWG, i32, i32, ptr) +DEF_HELPER_FLAGS_1(recpe_u32, TCG_CALL_NO_RWG, i32, i32) +DEF_HELPER_FLAGS_1(rsqrte_u32, TCG_CALL_NO_RWG, i32, i32) DEF_HELPER_FLAGS_4(neon_tbl, TCG_CALL_NO_RWG, i32, i32, i32, ptr, i32) DEF_HELPER_3(shl_cc, i32, env, i32, i32) 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_2misc_reciprocal(DisasContext *s, int opcode, switch (opcode) { case 0x3c: /* URECPE */ - gen_helper_recpe_u32(tcg_res, tcg_op, fpst); + gen_helper_recpe_u32(tcg_res, tcg_op); break; case 0x3d: /* FRECPE */ gen_helper_recpe_f32(tcg_res, tcg_op, fpst); @@ -XXX,XX +XXX,XX @@ static void disas_simd_two_reg_misc(DisasContext *s, uint32_t insn) unallocated_encoding(s); return; } - need_fpstatus = true; break; case 0x1e: /* FRINT32Z */ case 0x1f: /* FRINT64Z */ @@ -XXX,XX +XXX,XX @@ static void disas_simd_two_reg_misc(DisasContext *s, uint32_t insn) gen_helper_rints_exact(tcg_res, tcg_op, tcg_fpstatus); break; case 0x7c: /* URSQRTE */ - gen_helper_rsqrte_u32(tcg_res, tcg_op, tcg_fpstatus); + gen_helper_rsqrte_u32(tcg_res, tcg_op); break; case 0x1e: /* FRINT32Z */ case 0x5e: /* FRINT32X */ 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_neon_data_insn(DisasContext *s, uint32_t insn) break; } case NEON_2RM_VRECPE: - { - TCGv_ptr fpstatus = get_fpstatus_ptr(1); - gen_helper_recpe_u32(tmp, tmp, fpstatus); - tcg_temp_free_ptr(fpstatus); + gen_helper_recpe_u32(tmp, tmp); break; - } case NEON_2RM_VRSQRTE: - { - TCGv_ptr fpstatus = get_fpstatus_ptr(1); - gen_helper_rsqrte_u32(tmp, tmp, fpstatus); - tcg_temp_free_ptr(fpstatus); + gen_helper_rsqrte_u32(tmp, tmp); break; - } case NEON_2RM_VRECPE_F: { TCGv_ptr fpstatus = get_fpstatus_ptr(1); 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(rsqrte_f64)(float64 input, void *fpstp) return make_float64(val); } -uint32_t HELPER(recpe_u32)(uint32_t a, void *fpstp) +uint32_t HELPER(recpe_u32)(uint32_t a) { - /* float_status *s = fpstp; */ int input, estimate; if ((a & 0x80000000) == 0) { @@ -XXX,XX +XXX,XX @@ uint32_t HELPER(recpe_u32)(uint32_t a, void *fpstp) return deposit32(0, (32 - 9), 9, estimate); } -uint32_t HELPER(rsqrte_u32)(uint32_t a, void *fpstp) +uint32_t HELPER(rsqrte_u32)(uint32_t a) { int estimate; -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Provide a functional interface for the vector expansion. This fits better with the existing set of helpers that we provide for other operations. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-13-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/translate.h | 5 ++++ target/arm/translate-a64.c | 34 ++---------------------- target/arm/translate.c | 54 +++++++++++++++++++------------------- 3 files changed, 34 insertions(+), 59 deletions(-) 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 @@ void gen_gvec_sri(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, void gen_gvec_sli(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, int64_t shift, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_sqrdmlah_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_sqrdmlsh_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); + /* * Forward to the isar_feature_* tests given a DisasContext pointer. */ 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 gen_gvec_op3_ool(DisasContext *s, bool is_q, int rd, is_q ? 16 : 8, vec_full_reg_size(s), data, fn); } -/* Expand a 3-operand + env pointer operation using - * an out-of-line helper. - */ -static void gen_gvec_op3_env(DisasContext *s, bool is_q, int rd, - int rn, int rm, gen_helper_gvec_3_ptr *fn) -{ - tcg_gen_gvec_3_ptr(vec_full_reg_offset(s, rd), - vec_full_reg_offset(s, rn), - vec_full_reg_offset(s, rm), cpu_env, - is_q ? 16 : 8, vec_full_reg_size(s), 0, fn); -} - /* Expand a 3-operand + fpstatus pointer + simd data value operation using * an out-of-line helper. */ @@ -XXX,XX +XXX,XX @@ static void disas_simd_three_reg_same_extra(DisasContext *s, uint32_t insn) switch (opcode) { case 0x0: /* SQRDMLAH (vector) */ - switch (size) { - case 1: - gen_gvec_op3_env(s, is_q, rd, rn, rm, gen_helper_gvec_qrdmlah_s16); - break; - case 2: - gen_gvec_op3_env(s, is_q, rd, rn, rm, gen_helper_gvec_qrdmlah_s32); - break; - default: - g_assert_not_reached(); - } + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_sqrdmlah_qc, size); return; case 0x1: /* SQRDMLSH (vector) */ - switch (size) { - case 1: - gen_gvec_op3_env(s, is_q, rd, rn, rm, gen_helper_gvec_qrdmlsh_s16); - break; - case 2: - gen_gvec_op3_env(s, is_q, rd, rn, rm, gen_helper_gvec_qrdmlsh_s32); - break; - default: - g_assert_not_reached(); - } + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_sqrdmlsh_qc, size); return; case 0x2: /* SDOT / UDOT */ 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 const uint8_t neon_2rm_sizes[] = { [NEON_2RM_VCVT_UF] = 0x4, }; - -/* Expand v8.1 simd helper. */ -static int do_v81_helper(DisasContext *s, gen_helper_gvec_3_ptr *fn, - int q, int rd, int rn, int rm) +void gen_gvec_sqrdmlah_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) { - if (dc_isar_feature(aa32_rdm, s)) { - int opr_sz = (1 + q) * 8; - tcg_gen_gvec_3_ptr(vfp_reg_offset(1, rd), - vfp_reg_offset(1, rn), - vfp_reg_offset(1, rm), cpu_env, - opr_sz, opr_sz, 0, fn); - return 0; - } - return 1; + static gen_helper_gvec_3_ptr * const fns[2] = { + gen_helper_gvec_qrdmlah_s16, gen_helper_gvec_qrdmlah_s32 + }; + tcg_debug_assert(vece >= 1 && vece <= 2); + tcg_gen_gvec_3_ptr(rd_ofs, rn_ofs, rm_ofs, cpu_env, + opr_sz, max_sz, 0, fns[vece - 1]); +} + +void gen_gvec_sqrdmlsh_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static gen_helper_gvec_3_ptr * const fns[2] = { + gen_helper_gvec_qrdmlsh_s16, gen_helper_gvec_qrdmlsh_s32 + }; + tcg_debug_assert(vece >= 1 && vece <= 2); + tcg_gen_gvec_3_ptr(rd_ofs, rn_ofs, rm_ofs, cpu_env, + opr_sz, max_sz, 0, fns[vece - 1]); } #define GEN_CMP0(NAME, COND) \ @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) break; /* VPADD */ } /* VQRDMLAH */ - switch (size) { - case 1: - return do_v81_helper(s, gen_helper_gvec_qrdmlah_s16, - q, rd, rn, rm); - case 2: - return do_v81_helper(s, gen_helper_gvec_qrdmlah_s32, - q, rd, rn, rm); + if (dc_isar_feature(aa32_rdm, s) && (size == 1 || size == 2)) { + gen_gvec_sqrdmlah_qc(size, rd_ofs, rn_ofs, rm_ofs, + vec_size, vec_size); + return 0; } return 1; @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) break; } /* VQRDMLSH */ - switch (size) { - case 1: - return do_v81_helper(s, gen_helper_gvec_qrdmlsh_s16, - q, rd, rn, rm); - case 2: - return do_v81_helper(s, gen_helper_gvec_qrdmlsh_s32, - q, rd, rn, rm); + if (dc_isar_feature(aa32_rdm, s) && (size == 1 || size == 2)) { + gen_gvec_sqrdmlsh_qc(size, rd_ofs, rn_ofs, rm_ofs, + vec_size, vec_size); + return 0; } return 1; -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Pass a pointer directly to env->vfp.qc[0], rather than env. This will allow SVE2, which does not modify QC, to pass a pointer to dummy storage. Change the return type of inl_qrdml.h_s16 to match the sense of the operation: signed. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-14-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/translate.c | 18 ++++++++--- target/arm/vec_helper.c | 70 +++++++++++++++++++++++------------------ 2 files changed, 54 insertions(+), 34 deletions(-) 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 const uint8_t neon_2rm_sizes[] = { [NEON_2RM_VCVT_UF] = 0x4, }; +static void gen_gvec_fn3_qc(uint32_t rd_ofs, uint32_t rn_ofs, uint32_t rm_ofs, + uint32_t opr_sz, uint32_t max_sz, + gen_helper_gvec_3_ptr *fn) +{ + TCGv_ptr qc_ptr = tcg_temp_new_ptr(); + + tcg_gen_addi_ptr(qc_ptr, cpu_env, offsetof(CPUARMState, vfp.qc)); + tcg_gen_gvec_3_ptr(rd_ofs, rn_ofs, rm_ofs, qc_ptr, + opr_sz, max_sz, 0, fn); + tcg_temp_free_ptr(qc_ptr); +} + void gen_gvec_sqrdmlah_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) { @@ -XXX,XX +XXX,XX @@ void gen_gvec_sqrdmlah_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, gen_helper_gvec_qrdmlah_s16, gen_helper_gvec_qrdmlah_s32 }; tcg_debug_assert(vece >= 1 && vece <= 2); - tcg_gen_gvec_3_ptr(rd_ofs, rn_ofs, rm_ofs, cpu_env, - opr_sz, max_sz, 0, fns[vece - 1]); + gen_gvec_fn3_qc(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, fns[vece - 1]); } void gen_gvec_sqrdmlsh_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, @@ -XXX,XX +XXX,XX @@ void gen_gvec_sqrdmlsh_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, gen_helper_gvec_qrdmlsh_s16, gen_helper_gvec_qrdmlsh_s32 }; tcg_debug_assert(vece >= 1 && vece <= 2); - tcg_gen_gvec_3_ptr(rd_ofs, rn_ofs, rm_ofs, cpu_env, - opr_sz, max_sz, 0, fns[vece - 1]); + gen_gvec_fn3_qc(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, fns[vece - 1]); } #define GEN_CMP0(NAME, COND) \ diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ #define H4(x) (x) #endif -#define SET_QC() env->vfp.qc[0] = 1 - static void clear_tail(void *vd, uintptr_t opr_sz, uintptr_t max_sz) { uint64_t *d = vd + opr_sz; @@ -XXX,XX +XXX,XX @@ static void clear_tail(void *vd, uintptr_t opr_sz, uintptr_t max_sz) } /* Signed saturating rounding doubling multiply-accumulate high half, 16-bit */ -static uint16_t inl_qrdmlah_s16(CPUARMState *env, int16_t src1, - int16_t src2, int16_t src3) +static int16_t inl_qrdmlah_s16(int16_t src1, int16_t src2, + int16_t src3, uint32_t *sat) { /* Simplify: * = ((a3 << 16) + ((e1 * e2) << 1) + (1 << 15)) >> 16 @@ -XXX,XX +XXX,XX @@ static uint16_t inl_qrdmlah_s16(CPUARMState *env, int16_t src1, ret = ((int32_t)src3 << 15) + ret + (1 << 14); ret >>= 15; if (ret != (int16_t)ret) { - SET_QC(); + *sat = 1; ret = (ret < 0 ? -0x8000 : 0x7fff); } return ret; @@ -XXX,XX +XXX,XX @@ static uint16_t inl_qrdmlah_s16(CPUARMState *env, int16_t src1, uint32_t HELPER(neon_qrdmlah_s16)(CPUARMState *env, uint32_t src1, uint32_t src2, uint32_t src3) { - uint16_t e1 = inl_qrdmlah_s16(env, src1, src2, src3); - uint16_t e2 = inl_qrdmlah_s16(env, src1 >> 16, src2 >> 16, src3 >> 16); + uint32_t *sat = &env->vfp.qc[0]; + uint16_t e1 = inl_qrdmlah_s16(src1, src2, src3, sat); + uint16_t e2 = inl_qrdmlah_s16(src1 >> 16, src2 >> 16, src3 >> 16, sat); return deposit32(e1, 16, 16, e2); } void HELPER(gvec_qrdmlah_s16)(void *vd, void *vn, void *vm, - void *ve, uint32_t desc) + void *vq, uint32_t desc) { uintptr_t opr_sz = simd_oprsz(desc); int16_t *d = vd; int16_t *n = vn; int16_t *m = vm; - CPUARMState *env = ve; uintptr_t i; for (i = 0; i < opr_sz / 2; ++i) { - d[i] = inl_qrdmlah_s16(env, n[i], m[i], d[i]); + d[i] = inl_qrdmlah_s16(n[i], m[i], d[i], vq); } clear_tail(d, opr_sz, simd_maxsz(desc)); } /* Signed saturating rounding doubling multiply-subtract high half, 16-bit */ -static uint16_t inl_qrdmlsh_s16(CPUARMState *env, int16_t src1, - int16_t src2, int16_t src3) +static int16_t inl_qrdmlsh_s16(int16_t src1, int16_t src2, + int16_t src3, uint32_t *sat) { /* Similarly, using subtraction: * = ((a3 << 16) - ((e1 * e2) << 1) + (1 << 15)) >> 16 @@ -XXX,XX +XXX,XX @@ static uint16_t inl_qrdmlsh_s16(CPUARMState *env, int16_t src1, ret = ((int32_t)src3 << 15) - ret + (1 << 14); ret >>= 15; if (ret != (int16_t)ret) { - SET_QC(); + *sat = 1; ret = (ret < 0 ? -0x8000 : 0x7fff); } return ret; @@ -XXX,XX +XXX,XX @@ static uint16_t inl_qrdmlsh_s16(CPUARMState *env, int16_t src1, uint32_t HELPER(neon_qrdmlsh_s16)(CPUARMState *env, uint32_t src1, uint32_t src2, uint32_t src3) { - uint16_t e1 = inl_qrdmlsh_s16(env, src1, src2, src3); - uint16_t e2 = inl_qrdmlsh_s16(env, src1 >> 16, src2 >> 16, src3 >> 16); + uint32_t *sat = &env->vfp.qc[0]; + uint16_t e1 = inl_qrdmlsh_s16(src1, src2, src3, sat); + uint16_t e2 = inl_qrdmlsh_s16(src1 >> 16, src2 >> 16, src3 >> 16, sat); return deposit32(e1, 16, 16, e2); } void HELPER(gvec_qrdmlsh_s16)(void *vd, void *vn, void *vm, - void *ve, uint32_t desc) + void *vq, uint32_t desc) { uintptr_t opr_sz = simd_oprsz(desc); int16_t *d = vd; int16_t *n = vn; int16_t *m = vm; - CPUARMState *env = ve; uintptr_t i; for (i = 0; i < opr_sz / 2; ++i) { - d[i] = inl_qrdmlsh_s16(env, n[i], m[i], d[i]); + d[i] = inl_qrdmlsh_s16(n[i], m[i], d[i], vq); } clear_tail(d, opr_sz, simd_maxsz(desc)); } /* Signed saturating rounding doubling multiply-accumulate high half, 32-bit */ -uint32_t HELPER(neon_qrdmlah_s32)(CPUARMState *env, int32_t src1, - int32_t src2, int32_t src3) +static int32_t inl_qrdmlah_s32(int32_t src1, int32_t src2, + int32_t src3, uint32_t *sat) { /* Simplify similarly to int_qrdmlah_s16 above. */ int64_t ret = (int64_t)src1 * src2; ret = ((int64_t)src3 << 31) + ret + (1 << 30); ret >>= 31; if (ret != (int32_t)ret) { - SET_QC(); + *sat = 1; ret = (ret < 0 ? INT32_MIN : INT32_MAX); } return ret; } +uint32_t HELPER(neon_qrdmlah_s32)(CPUARMState *env, int32_t src1, + int32_t src2, int32_t src3) +{ + uint32_t *sat = &env->vfp.qc[0]; + return inl_qrdmlah_s32(src1, src2, src3, sat); +} + void HELPER(gvec_qrdmlah_s32)(void *vd, void *vn, void *vm, - void *ve, uint32_t desc) + void *vq, uint32_t desc) { uintptr_t opr_sz = simd_oprsz(desc); int32_t *d = vd; int32_t *n = vn; int32_t *m = vm; - CPUARMState *env = ve; uintptr_t i; for (i = 0; i < opr_sz / 4; ++i) { - d[i] = helper_neon_qrdmlah_s32(env, n[i], m[i], d[i]); + d[i] = inl_qrdmlah_s32(n[i], m[i], d[i], vq); } clear_tail(d, opr_sz, simd_maxsz(desc)); } /* Signed saturating rounding doubling multiply-subtract high half, 32-bit */ -uint32_t HELPER(neon_qrdmlsh_s32)(CPUARMState *env, int32_t src1, - int32_t src2, int32_t src3) +static int32_t inl_qrdmlsh_s32(int32_t src1, int32_t src2, + int32_t src3, uint32_t *sat) { /* Simplify similarly to int_qrdmlsh_s16 above. */ int64_t ret = (int64_t)src1 * src2; ret = ((int64_t)src3 << 31) - ret + (1 << 30); ret >>= 31; if (ret != (int32_t)ret) { - SET_QC(); + *sat = 1; ret = (ret < 0 ? INT32_MIN : INT32_MAX); } return ret; } +uint32_t HELPER(neon_qrdmlsh_s32)(CPUARMState *env, int32_t src1, + int32_t src2, int32_t src3) +{ + uint32_t *sat = &env->vfp.qc[0]; + return inl_qrdmlsh_s32(src1, src2, src3, sat); +} + void HELPER(gvec_qrdmlsh_s32)(void *vd, void *vn, void *vm, - void *ve, uint32_t desc) + void *vq, uint32_t desc) { uintptr_t opr_sz = simd_oprsz(desc); int32_t *d = vd; int32_t *n = vn; int32_t *m = vm; - CPUARMState *env = ve; uintptr_t i; for (i = 0; i < opr_sz / 4; ++i) { - d[i] = helper_neon_qrdmlsh_s32(env, n[i], m[i], d[i]); + d[i] = inl_qrdmlsh_s32(n[i], m[i], d[i], vq); } clear_tail(d, opr_sz, simd_maxsz(desc)); } -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Must clear the tail for AdvSIMD when SVE is enabled. Fixes: ca40a6e6e39 Cc: qemu-stable@nongnu.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-15-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/vec_helper.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(NAME)(void *vd, void *vn, void *vm, void *stat, uint32_t desc) \ d[i + j] = TYPE##_mul(n[i + j], mm, stat); \ } \ } \ + clear_tail(d, oprsz, simd_maxsz(desc)); \ } DO_MUL_IDX(gvec_fmul_idx_h, float16, H2) @@ -XXX,XX +XXX,XX @@ void HELPER(NAME)(void *vd, void *vn, void *vm, void *va, \ mm, a[i + j], 0, stat); \ } \ } \ + clear_tail(d, oprsz, simd_maxsz(desc)); \ } DO_FMLA_IDX(gvec_fmla_idx_h, float16, H2) -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Include 64-bit element size in preparation for SVE2. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-16-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 10 +++ target/arm/translate.h | 5 ++ target/arm/translate-a64.c | 8 ++- target/arm/translate.c | 133 ++++++++++++++++++++++++++++++++++++- target/arm/vec_helper.c | 24 +++++++ 5 files changed, 176 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_3(gvec_sli_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_3(gvec_sli_s, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_3(gvec_sli_d, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_sabd_b, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_sabd_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_sabd_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_sabd_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) + +DEF_HELPER_FLAGS_4(gvec_uabd_b, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_uabd_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_uabd_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_uabd_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) + #ifdef TARGET_AARCH64 #include "helper-a64.h" #include "helper-sve.h" 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 @@ void gen_gvec_sqrdmlah_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, void gen_gvec_sqrdmlsh_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_sabd(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_uabd(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); + /* * Forward to the isar_feature_* tests given a DisasContext pointer. */ 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 disas_simd_3same_int(DisasContext *s, uint32_t insn) gen_gvec_fn3(s, is_q, rd, rn, rm, tcg_gen_gvec_smin, size); } return; + case 0xe: /* SABD, UABD */ + if (u) { + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_uabd, size); + } else { + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_sabd, size); + } + return; case 0x10: /* ADD, SUB */ if (u) { gen_gvec_fn3(s, is_q, rd, rn, rm, tcg_gen_gvec_sub, size); @@ -XXX,XX +XXX,XX @@ static void disas_simd_3same_int(DisasContext *s, uint32_t insn) genenvfn = fns[size][u]; break; } - case 0xe: /* SABD, UABD */ case 0xf: /* SABA, UABA */ { static NeonGenTwoOpFn * const fns[3][2] = { 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 @@ void gen_gvec_sqsub_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); } +static void gen_sabd_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) +{ + TCGv_i32 t = tcg_temp_new_i32(); + + tcg_gen_sub_i32(t, a, b); + tcg_gen_sub_i32(d, b, a); + tcg_gen_movcond_i32(TCG_COND_LT, d, a, b, d, t); + tcg_temp_free_i32(t); +} + +static void gen_sabd_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + tcg_gen_sub_i64(t, a, b); + tcg_gen_sub_i64(d, b, a); + tcg_gen_movcond_i64(TCG_COND_LT, d, a, b, d, t); + tcg_temp_free_i64(t); +} + +static void gen_sabd_vec(unsigned vece, TCGv_vec d, TCGv_vec a, TCGv_vec b) +{ + TCGv_vec t = tcg_temp_new_vec_matching(d); + + tcg_gen_smin_vec(vece, t, a, b); + tcg_gen_smax_vec(vece, d, a, b); + tcg_gen_sub_vec(vece, d, d, t); + tcg_temp_free_vec(t); +} + +void gen_gvec_sabd(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_sub_vec, INDEX_op_smin_vec, INDEX_op_smax_vec, 0 + }; + static const GVecGen3 ops[4] = { + { .fniv = gen_sabd_vec, + .fno = gen_helper_gvec_sabd_b, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fniv = gen_sabd_vec, + .fno = gen_helper_gvec_sabd_h, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = gen_sabd_i32, + .fniv = gen_sabd_vec, + .fno = gen_helper_gvec_sabd_s, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = gen_sabd_i64, + .fniv = gen_sabd_vec, + .fno = gen_helper_gvec_sabd_d, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .opt_opc = vecop_list, + .vece = MO_64 }, + }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} + +static void gen_uabd_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) +{ + TCGv_i32 t = tcg_temp_new_i32(); + + tcg_gen_sub_i32(t, a, b); + tcg_gen_sub_i32(d, b, a); + tcg_gen_movcond_i32(TCG_COND_LTU, d, a, b, d, t); + tcg_temp_free_i32(t); +} + +static void gen_uabd_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) +{ + TCGv_i64 t = tcg_temp_new_i64(); + + tcg_gen_sub_i64(t, a, b); + tcg_gen_sub_i64(d, b, a); + tcg_gen_movcond_i64(TCG_COND_LTU, d, a, b, d, t); + tcg_temp_free_i64(t); +} + +static void gen_uabd_vec(unsigned vece, TCGv_vec d, TCGv_vec a, TCGv_vec b) +{ + TCGv_vec t = tcg_temp_new_vec_matching(d); + + tcg_gen_umin_vec(vece, t, a, b); + tcg_gen_umax_vec(vece, d, a, b); + tcg_gen_sub_vec(vece, d, d, t); + tcg_temp_free_vec(t); +} + +void gen_gvec_uabd(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_sub_vec, INDEX_op_umin_vec, INDEX_op_umax_vec, 0 + }; + static const GVecGen3 ops[4] = { + { .fniv = gen_uabd_vec, + .fno = gen_helper_gvec_uabd_b, + .opt_opc = vecop_list, + .vece = MO_8 }, + { .fniv = gen_uabd_vec, + .fno = gen_helper_gvec_uabd_h, + .opt_opc = vecop_list, + .vece = MO_16 }, + { .fni4 = gen_uabd_i32, + .fniv = gen_uabd_vec, + .fno = gen_helper_gvec_uabd_s, + .opt_opc = vecop_list, + .vece = MO_32 }, + { .fni8 = gen_uabd_i64, + .fniv = gen_uabd_vec, + .fno = gen_helper_gvec_uabd_d, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .opt_opc = vecop_list, + .vece = MO_64 }, + }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} + /* Translate a NEON data processing instruction. Return nonzero if the instruction is invalid. We process data in a mixture of 32-bit and 64-bit chunks. @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) } return 1; + case NEON_3R_VABD: + if (u) { + gen_gvec_uabd(size, rd_ofs, rn_ofs, rm_ofs, + vec_size, vec_size); + } else { + gen_gvec_sabd(size, rd_ofs, rn_ofs, rm_ofs, + vec_size, vec_size); + } + return 0; + case NEON_3R_VADD_VSUB: case NEON_3R_LOGIC: case NEON_3R_VMAX: @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VQRSHL: GEN_NEON_INTEGER_OP_ENV(qrshl); break; - case NEON_3R_VABD: - GEN_NEON_INTEGER_OP(abd); - break; case NEON_3R_VABA: GEN_NEON_INTEGER_OP(abd); tcg_temp_free_i32(tmp2); diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ DO_CMP0(gvec_cgt0_h, int16_t, >) DO_CMP0(gvec_cge0_h, int16_t, >=) #undef DO_CMP0 + +#define DO_ABD(NAME, TYPE) \ +void HELPER(NAME)(void *vd, void *vn, void *vm, uint32_t desc) \ +{ \ + intptr_t i, opr_sz = simd_oprsz(desc); \ + TYPE *d = vd, *n = vn, *m = vm; \ + \ + for (i = 0; i < opr_sz / sizeof(TYPE); ++i) { \ + d[i] = n[i] < m[i] ? m[i] - n[i] : n[i] - m[i]; \ + } \ + clear_tail(d, opr_sz, simd_maxsz(desc)); \ +} + +DO_ABD(gvec_sabd_b, int8_t) +DO_ABD(gvec_sabd_h, int16_t) +DO_ABD(gvec_sabd_s, int32_t) +DO_ABD(gvec_sabd_d, int64_t) + +DO_ABD(gvec_uabd_b, uint8_t) +DO_ABD(gvec_uabd_h, uint16_t) +DO_ABD(gvec_uabd_s, uint32_t) +DO_ABD(gvec_uabd_d, uint64_t) + +#undef DO_ABD -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Include 64-bit element size in preparation for SVE2. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200513163245.17915-17-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 17 +++-- target/arm/translate.h | 5 ++ target/arm/neon_helper.c | 10 --- target/arm/translate-a64.c | 17 ++--- target/arm/translate.c | 134 +++++++++++++++++++++++++++++++++++-- target/arm/vec_helper.c | 24 +++++++ 6 files changed, 174 insertions(+), 33 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_2(neon_pmax_s8, i32, i32, i32) DEF_HELPER_2(neon_pmax_u16, i32, i32, i32) DEF_HELPER_2(neon_pmax_s16, i32, i32, i32) -DEF_HELPER_2(neon_abd_u8, i32, i32, i32) -DEF_HELPER_2(neon_abd_s8, i32, i32, i32) -DEF_HELPER_2(neon_abd_u16, i32, i32, i32) -DEF_HELPER_2(neon_abd_s16, i32, i32, i32) -DEF_HELPER_2(neon_abd_u32, i32, i32, i32) -DEF_HELPER_2(neon_abd_s32, i32, i32, i32) - DEF_HELPER_2(neon_shl_u16, i32, i32, i32) DEF_HELPER_2(neon_shl_s16, i32, i32, i32) DEF_HELPER_2(neon_rshl_u8, i32, i32, i32) @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_4(gvec_uabd_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(gvec_uabd_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(gvec_uabd_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_saba_b, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_saba_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_saba_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_saba_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) + +DEF_HELPER_FLAGS_4(gvec_uaba_b, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_uaba_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_uaba_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_uaba_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) + #ifdef TARGET_AARCH64 #include "helper-a64.h" #include "helper-sve.h" 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 @@ void gen_gvec_sabd(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, void gen_gvec_uabd(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_saba(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_uaba(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); + /* * Forward to the isar_feature_* tests given a DisasContext pointer. */ diff --git a/target/arm/neon_helper.c b/target/arm/neon_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon_helper.c +++ b/target/arm/neon_helper.c @@ -XXX,XX +XXX,XX @@ NEON_POP(pmax_s16, neon_s16, 2) NEON_POP(pmax_u16, neon_u16, 2) #undef NEON_FN -#define NEON_FN(dest, src1, src2) \ - dest = (src1 > src2) ? (src1 - src2) : (src2 - src1) -NEON_VOP(abd_s8, neon_s8, 4) -NEON_VOP(abd_u8, neon_u8, 4) -NEON_VOP(abd_s16, neon_s16, 2) -NEON_VOP(abd_u16, neon_u16, 2) -NEON_VOP(abd_s32, neon_s32, 1) -NEON_VOP(abd_u32, neon_u32, 1) -#undef NEON_FN - #define NEON_FN(dest, src1, src2) do { \ int8_t tmp; \ tmp = (int8_t)src2; \ 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 disas_simd_3same_int(DisasContext *s, uint32_t insn) gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_sabd, size); } return; + case 0xf: /* SABA, UABA */ + if (u) { + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_uaba, size); + } else { + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_saba, size); + } + return; case 0x10: /* ADD, SUB */ if (u) { gen_gvec_fn3(s, is_q, rd, rn, rm, tcg_gen_gvec_sub, size); @@ -XXX,XX +XXX,XX @@ static void disas_simd_3same_int(DisasContext *s, uint32_t insn) genenvfn = fns[size][u]; break; } - case 0xf: /* SABA, UABA */ - { - static NeonGenTwoOpFn * const fns[3][2] = { - { gen_helper_neon_abd_s8, gen_helper_neon_abd_u8 }, - { gen_helper_neon_abd_s16, gen_helper_neon_abd_u16 }, - { gen_helper_neon_abd_s32, gen_helper_neon_abd_u32 }, - }; - genfn = fns[size][u]; - break; - } case 0x16: /* SQDMULH, SQRDMULH */ { static NeonGenTwoOpEnvFn * const fns[2][2] = { 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 @@ void gen_gvec_uabd(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); } +static void gen_saba_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) +{ + TCGv_i32 t = tcg_temp_new_i32(); + gen_sabd_i32(t, a, b); + tcg_gen_add_i32(d, d, t); + tcg_temp_free_i32(t); +} + +static void gen_saba_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) +{ + TCGv_i64 t = tcg_temp_new_i64(); + gen_sabd_i64(t, a, b); + tcg_gen_add_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_saba_vec(unsigned vece, TCGv_vec d, TCGv_vec a, TCGv_vec b) +{ + TCGv_vec t = tcg_temp_new_vec_matching(d); + gen_sabd_vec(vece, t, a, b); + tcg_gen_add_vec(vece, d, d, t); + tcg_temp_free_vec(t); +} + +void gen_gvec_saba(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_sub_vec, INDEX_op_add_vec, + INDEX_op_smin_vec, INDEX_op_smax_vec, 0 + }; + static const GVecGen3 ops[4] = { + { .fniv = gen_saba_vec, + .fno = gen_helper_gvec_saba_b, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_8 }, + { .fniv = gen_saba_vec, + .fno = gen_helper_gvec_saba_h, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_16 }, + { .fni4 = gen_saba_i32, + .fniv = gen_saba_vec, + .fno = gen_helper_gvec_saba_s, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_32 }, + { .fni8 = gen_saba_i64, + .fniv = gen_saba_vec, + .fno = gen_helper_gvec_saba_d, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_64 }, + }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} + +static void gen_uaba_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) +{ + TCGv_i32 t = tcg_temp_new_i32(); + gen_uabd_i32(t, a, b); + tcg_gen_add_i32(d, d, t); + tcg_temp_free_i32(t); +} + +static void gen_uaba_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) +{ + TCGv_i64 t = tcg_temp_new_i64(); + gen_uabd_i64(t, a, b); + tcg_gen_add_i64(d, d, t); + tcg_temp_free_i64(t); +} + +static void gen_uaba_vec(unsigned vece, TCGv_vec d, TCGv_vec a, TCGv_vec b) +{ + TCGv_vec t = tcg_temp_new_vec_matching(d); + gen_uabd_vec(vece, t, a, b); + tcg_gen_add_vec(vece, d, d, t); + tcg_temp_free_vec(t); +} + +void gen_gvec_uaba(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_sub_vec, INDEX_op_add_vec, + INDEX_op_umin_vec, INDEX_op_umax_vec, 0 + }; + static const GVecGen3 ops[4] = { + { .fniv = gen_uaba_vec, + .fno = gen_helper_gvec_uaba_b, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_8 }, + { .fniv = gen_uaba_vec, + .fno = gen_helper_gvec_uaba_h, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_16 }, + { .fni4 = gen_uaba_i32, + .fniv = gen_uaba_vec, + .fno = gen_helper_gvec_uaba_s, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_32 }, + { .fni8 = gen_uaba_i64, + .fniv = gen_uaba_vec, + .fno = gen_helper_gvec_uaba_d, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_64 }, + }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} + /* Translate a NEON data processing instruction. Return nonzero if the instruction is invalid. We process data in a mixture of 32-bit and 64-bit chunks. @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) } return 0; + case NEON_3R_VABA: + if (u) { + gen_gvec_uaba(size, rd_ofs, rn_ofs, rm_ofs, + vec_size, vec_size); + } else { + gen_gvec_saba(size, rd_ofs, rn_ofs, rm_ofs, + vec_size, vec_size); + } + return 0; + case NEON_3R_VADD_VSUB: case NEON_3R_LOGIC: case NEON_3R_VMAX: @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VQRSHL: GEN_NEON_INTEGER_OP_ENV(qrshl); break; - case NEON_3R_VABA: - GEN_NEON_INTEGER_OP(abd); - tcg_temp_free_i32(tmp2); - tmp2 = neon_load_reg(rd, pass); - gen_neon_add(size, tmp, tmp2); - break; case NEON_3R_VPMAX: GEN_NEON_INTEGER_OP(pmax); break; diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ DO_ABD(gvec_uabd_s, uint32_t) DO_ABD(gvec_uabd_d, uint64_t) #undef DO_ABD + +#define DO_ABA(NAME, TYPE) \ +void HELPER(NAME)(void *vd, void *vn, void *vm, uint32_t desc) \ +{ \ + intptr_t i, opr_sz = simd_oprsz(desc); \ + TYPE *d = vd, *n = vn, *m = vm; \ + \ + for (i = 0; i < opr_sz / sizeof(TYPE); ++i) { \ + d[i] += n[i] < m[i] ? m[i] - n[i] : n[i] - m[i]; \ + } \ + clear_tail(d, opr_sz, simd_maxsz(desc)); \ +} + +DO_ABA(gvec_saba_b, int8_t) +DO_ABA(gvec_saba_h, int16_t) +DO_ABA(gvec_saba_s, int32_t) +DO_ABA(gvec_saba_d, int64_t) + +DO_ABA(gvec_uaba_b, uint8_t) +DO_ABA(gvec_uaba_h, uint16_t) +DO_ABA(gvec_uaba_s, uint32_t) +DO_ABA(gvec_uaba_d, uint64_t) + +#undef DO_ABA -- 2.20.1 From: Patrick Williams <patrick@stwcx.xyz> Sonora Pass is a 2 socket x86 motherboard designed by Facebook and supported by OpenBMC. Strapping configuration was obtained from hardware and i2c configuration is based on dts found at: https://github.com/facebook/openbmc-linux/blob/1633c87b8ba7c162095787c988979b748ba65dc8/arch/arm/boot/dts/aspeed-bmc-facebook-sonorapass.dts Booted a test image of http://github.com/facebook/openbmc to login prompt. Signed-off-by: Patrick Williams <patrick@stwcx.xyz> Reviewed-by: Amithash Prasad <amithash@fb.com> Reviewed-by: Cédric Le Goater <clg@kaod.org> [PMM: fixed block comment style nit] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/aspeed.c | 78 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 78 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 { SCU_AST2500_HW_STRAP_ACPI_ENABLE | \ SCU_HW_STRAP_SPI_MODE(SCU_HW_STRAP_SPI_MASTER)) +/* Sonorapass hardware value: 0xF100D216 */ +#define SONORAPASS_BMC_HW_STRAP1 ( \ + SCU_AST2500_HW_STRAP_SPI_AUTOFETCH_ENABLE | \ + SCU_AST2500_HW_STRAP_GPIO_STRAP_ENABLE | \ + SCU_AST2500_HW_STRAP_UART_DEBUG | \ + SCU_AST2500_HW_STRAP_RESERVED28 | \ + SCU_AST2500_HW_STRAP_DDR4_ENABLE | \ + SCU_HW_STRAP_VGA_CLASS_CODE | \ + SCU_HW_STRAP_LPC_RESET_PIN | \ + SCU_HW_STRAP_SPI_MODE(SCU_HW_STRAP_SPI_MASTER) | \ + SCU_AST2500_HW_STRAP_SET_AXI_AHB_RATIO(AXI_AHB_RATIO_2_1) | \ + SCU_HW_STRAP_VGA_BIOS_ROM | \ + SCU_HW_STRAP_VGA_SIZE_SET(VGA_16M_DRAM) | \ + SCU_AST2500_HW_STRAP_RESERVED1) + /* Swift hardware value: 0xF11AD206 */ #define SWIFT_BMC_HW_STRAP1 ( \ AST2500_HW_STRAP1_DEFAULTS | \ @@ -XXX,XX +XXX,XX @@ static void swift_bmc_i2c_init(AspeedBoardState *bmc) i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 12), "tmp105", 0x4a); } +static void sonorapass_bmc_i2c_init(AspeedBoardState *bmc) +{ + AspeedSoCState *soc = &bmc->soc; + + /* bus 2 : */ + i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 2), "tmp105", 0x48); + i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 2), "tmp105", 0x49); + /* bus 2 : pca9546 @ 0x73 */ + + /* bus 3 : pca9548 @ 0x70 */ + + /* bus 4 : */ + uint8_t *eeprom4_54 = g_malloc0(8 * 1024); + smbus_eeprom_init_one(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 4), 0x54, + eeprom4_54); + /* PCA9539 @ 0x76, but PCA9552 is compatible */ + i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 4), "pca9552", 0x76); + /* PCA9539 @ 0x77, but PCA9552 is compatible */ + i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 4), "pca9552", 0x77); + + /* bus 6 : */ + i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 6), "tmp105", 0x48); + i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 6), "tmp105", 0x49); + /* bus 6 : pca9546 @ 0x73 */ + + /* bus 8 : */ + uint8_t *eeprom8_56 = g_malloc0(8 * 1024); + smbus_eeprom_init_one(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 8), 0x56, + eeprom8_56); + i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 8), "pca9552", 0x60); + i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 8), "pca9552", 0x61); + /* bus 8 : adc128d818 @ 0x1d */ + /* bus 8 : adc128d818 @ 0x1f */ + + /* + * bus 13 : pca9548 @ 0x71 + * - channel 3: + * - tmm421 @ 0x4c + * - tmp421 @ 0x4e + * - tmp421 @ 0x4f + */ + +} + static void witherspoon_bmc_i2c_init(AspeedBoardState *bmc) { AspeedSoCState *soc = &bmc->soc; @@ -XXX,XX +XXX,XX @@ static void aspeed_machine_romulus_class_init(ObjectClass *oc, void *data) mc->default_ram_size = 512 * MiB; }; +static void aspeed_machine_sonorapass_class_init(ObjectClass *oc, void *data) +{ + MachineClass *mc = MACHINE_CLASS(oc); + AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc); + + mc->desc = "OCP SonoraPass BMC (ARM1176)"; + amc->soc_name = "ast2500-a1"; + amc->hw_strap1 = SONORAPASS_BMC_HW_STRAP1; + amc->fmc_model = "mx66l1g45g"; + amc->spi_model = "mx66l1g45g"; + amc->num_cs = 2; + amc->i2c_init = sonorapass_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); @@ -XXX,XX +XXX,XX @@ static const TypeInfo aspeed_machine_types[] = { .name = MACHINE_TYPE_NAME("swift-bmc"), .parent = TYPE_ASPEED_MACHINE, .class_init = aspeed_machine_swift_class_init, + }, { + .name = MACHINE_TYPE_NAME("sonorapass-bmc"), + .parent = TYPE_ASPEED_MACHINE, + .class_init = aspeed_machine_sonorapass_class_init, }, { .name = MACHINE_TYPE_NAME("witherspoon-bmc"), .parent = TYPE_ASPEED_MACHINE, -- 2.20.1 From: Dongjiu Geng <gengdongjiu@huawei.com> The little end UUID is used in many places, so make NVDIMM_UUID_LE to a common macro to convert the UUID to a little end array. Reviewed-by: Xiang Zheng <zhengxiang9@huawei.com> Signed-off-by: Dongjiu Geng <gengdongjiu@huawei.com> Message-id: 20200512030609.19593-2-gengdongjiu@huawei.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/qemu/uuid.h | 27 +++++++++++++++++++++++++++ hw/acpi/nvdimm.c | 10 +++------- 2 files changed, 30 insertions(+), 7 deletions(-) diff --git a/include/qemu/uuid.h b/include/qemu/uuid.h index XXXXXXX..XXXXXXX 100644 --- a/include/qemu/uuid.h +++ b/include/qemu/uuid.h @@ -XXX,XX +XXX,XX @@ typedef struct { }; } QemuUUID; +/** + * UUID_LE - converts the fields of UUID to little-endian array, + * each of parameters is the filed of UUID. + * + * @time_low: The low field of the timestamp + * @time_mid: The middle field of the timestamp + * @time_hi_and_version: The high field of the timestamp + * multiplexed with the version number + * @clock_seq_hi_and_reserved: The high field of the clock + * sequence multiplexed with the variant + * @clock_seq_low: The low field of the clock sequence + * @node0: The spatially unique node0 identifier + * @node1: The spatially unique node1 identifier + * @node2: The spatially unique node2 identifier + * @node3: The spatially unique node3 identifier + * @node4: The spatially unique node4 identifier + * @node5: The spatially unique node5 identifier + */ +#define UUID_LE(time_low, time_mid, time_hi_and_version, \ + clock_seq_hi_and_reserved, clock_seq_low, node0, node1, node2, \ + node3, node4, node5) \ + { (time_low) & 0xff, ((time_low) >> 8) & 0xff, ((time_low) >> 16) & 0xff, \ + ((time_low) >> 24) & 0xff, (time_mid) & 0xff, ((time_mid) >> 8) & 0xff, \ + (time_hi_and_version) & 0xff, ((time_hi_and_version) >> 8) & 0xff, \ + (clock_seq_hi_and_reserved), (clock_seq_low), (node0), (node1), (node2),\ + (node3), (node4), (node5) } + #define UUID_FMT "%02hhx%02hhx%02hhx%02hhx-" \ "%02hhx%02hhx-%02hhx%02hhx-" \ "%02hhx%02hhx-" \ diff --git a/hw/acpi/nvdimm.c b/hw/acpi/nvdimm.c index XXXXXXX..XXXXXXX 100644 --- a/hw/acpi/nvdimm.c +++ b/hw/acpi/nvdimm.c @@ -XXX,XX +XXX,XX @@ */ #include "qemu/osdep.h" +#include "qemu/uuid.h" #include "hw/acpi/acpi.h" #include "hw/acpi/aml-build.h" #include "hw/acpi/bios-linker-loader.h" @@ -XXX,XX +XXX,XX @@ #include "hw/mem/nvdimm.h" #include "qemu/nvdimm-utils.h" -#define NVDIMM_UUID_LE(a, b, c, d0, d1, d2, d3, d4, d5, d6, d7) \ - { (a) & 0xff, ((a) >> 8) & 0xff, ((a) >> 16) & 0xff, ((a) >> 24) & 0xff, \ - (b) & 0xff, ((b) >> 8) & 0xff, (c) & 0xff, ((c) >> 8) & 0xff, \ - (d0), (d1), (d2), (d3), (d4), (d5), (d6), (d7) } - /* * define Byte Addressable Persistent Memory (PM) Region according to * ACPI 6.0: 5.2.25.1 System Physical Address Range Structure. */ static const uint8_t nvdimm_nfit_spa_uuid[] = - NVDIMM_UUID_LE(0x66f0d379, 0xb4f3, 0x4074, 0xac, 0x43, 0x0d, 0x33, - 0x18, 0xb7, 0x8c, 0xdb); + UUID_LE(0x66f0d379, 0xb4f3, 0x4074, 0xac, 0x43, 0x0d, 0x33, + 0x18, 0xb7, 0x8c, 0xdb); /* * NVDIMM Firmware Interface Table -- 2.20.1 From: Dongjiu Geng <gengdongjiu@huawei.com> RAS Virtualization feature is not supported now, so add a RAS machine option and disable it by default. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Dongjiu Geng <gengdongjiu@huawei.com> Signed-off-by: Xiang Zheng <zhengxiang9@huawei.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Message-id: 20200512030609.19593-3-gengdongjiu@huawei.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/virt.h | 1 + hw/arm/virt.c | 23 +++++++++++++++++++++++ 2 files changed, 24 insertions(+) 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 { bool highmem_ecam; bool its; bool virt; + bool ras; OnOffAuto acpi; VirtGICType gic_version; VirtIOMMUType iommu; 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 virt_set_acpi(Object *obj, Visitor *v, const char *name, visit_type_OnOffAuto(v, name, &vms->acpi, errp); } +static bool virt_get_ras(Object *obj, Error **errp) +{ + VirtMachineState *vms = VIRT_MACHINE(obj); + + return vms->ras; +} + +static void virt_set_ras(Object *obj, bool value, Error **errp) +{ + VirtMachineState *vms = VIRT_MACHINE(obj); + + vms->ras = value; +} + static char *virt_get_gic_version(Object *obj, Error **errp) { VirtMachineState *vms = VIRT_MACHINE(obj); @@ -XXX,XX +XXX,XX @@ static void virt_instance_init(Object *obj) "Valid values are none and smmuv3", NULL); + /* Default disallows RAS instantiation */ + vms->ras = false; + object_property_add_bool(obj, "ras", virt_get_ras, + virt_set_ras, NULL); + object_property_set_description(obj, "ras", + "Set on/off to enable/disable reporting host memory errors " + "to a KVM guest using ACPI and guest external abort exceptions", + NULL); + vms->irqmap = a15irqmap; virt_flash_create(vms); -- 2.20.1 From: Dongjiu Geng <gengdongjiu@huawei.com> Add APEI/GHES detailed design document Signed-off-by: Dongjiu Geng <gengdongjiu@huawei.com> Signed-off-by: Xiang Zheng <zhengxiang9@huawei.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Message-id: 20200512030609.19593-4-gengdongjiu@huawei.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/specs/acpi_hest_ghes.rst | 110 ++++++++++++++++++++++++++++++++++ docs/specs/index.rst | 1 + 2 files changed, 111 insertions(+) create mode 100644 docs/specs/acpi_hest_ghes.rst diff --git a/docs/specs/acpi_hest_ghes.rst b/docs/specs/acpi_hest_ghes.rst new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/docs/specs/acpi_hest_ghes.rst @@ -XXX,XX +XXX,XX @@ +APEI tables generating and CPER record +====================================== + +.. + Copyright (c) 2020 HUAWEI TECHNOLOGIES CO., LTD. + + This work is licensed under the terms of the GNU GPL, version 2 or later. + See the COPYING file in the top-level directory. + +Design Details +-------------- + +:: + + etc/acpi/tables etc/hardware_errors + ==================== =============================== + + +--------------------------+ +----------------------------+ + | | HEST | +--------->| error_block_address1 |------+ + | +--------------------------+ | +----------------------------+ | + | | GHES1 | | +------->| error_block_address2 |------+-+ + | +--------------------------+ | | +----------------------------+ | | + | | ................. | | | | .............. | | | + | | error_status_address-----+-+ | -----------------------------+ | | + | | ................. | | +--->| error_block_addressN |------+-+---+ + | | read_ack_register--------+-+ | | +----------------------------+ | | | + | | read_ack_preserve | +-+---+--->| read_ack_register1 | | | | + | | read_ack_write | | | +----------------------------+ | | | + + +--------------------------+ | +-+--->| read_ack_register2 | | | | + | | GHES2 | | | | +----------------------------+ | | | + + +--------------------------+ | | | | ............. | | | | + | | ................. | | | | +----------------------------+ | | | + | | error_status_address-----+---+ | | +->| read_ack_registerN | | | | + | | ................. | | | | +----------------------------+ | | | + | | read_ack_register--------+-----+ | | |Generic Error Status Block 1|<-----+ | | + | | read_ack_preserve | | | |-+------------------------+-+ | | + | | read_ack_write | | | | | CPER | | | | + + +--------------------------| | | | | CPER | | | | + | | ............... | | | | | .... | | | | + + +--------------------------+ | | | | CPER | | | | + | | GHESN | | | |-+------------------------+-| | | + + +--------------------------+ | | |Generic Error Status Block 2|<-------+ | + | | ................. | | | |-+------------------------+-+ | + | | error_status_address-----+-------+ | | | CPER | | | + | | ................. | | | | CPER | | | + | | read_ack_register--------+---------+ | | .... | | | + | | read_ack_preserve | | | CPER | | | + | | read_ack_write | +-+------------------------+-+ | + + +--------------------------+ | .......... | | + |----------------------------+ | + |Generic Error Status Block N |<----------+ + |-+-------------------------+-+ + | | CPER | | + | | CPER | | + | | .... | | + | | CPER | | + +-+-------------------------+-+ + + +(1) QEMU generates the ACPI HEST table. This table goes in the current + "etc/acpi/tables" fw_cfg blob. Each error source has different + notification types. + +(2) A new fw_cfg blob called "etc/hardware_errors" is introduced. QEMU + also needs to populate this blob. The "etc/hardware_errors" fw_cfg blob + contains an address registers table and an Error Status Data Block table. + +(3) The address registers table contains N Error Block Address entries + and N Read Ack Register entries. The size for each entry is 8-byte. + The Error Status Data Block table contains N Error Status Data Block + entries. The size for each entry is 4096(0x1000) bytes. The total size + for the "etc/hardware_errors" fw_cfg blob is (N * 8 * 2 + N * 4096) bytes. + N is the number of the kinds of hardware error sources. + +(4) QEMU generates the ACPI linker/loader script for the firmware. The + firmware pre-allocates memory for "etc/acpi/tables", "etc/hardware_errors" + and copies blob contents there. + +(5) QEMU generates N ADD_POINTER commands, which patch addresses in the + "error_status_address" fields of the HEST table with a pointer to the + corresponding "address registers" in the "etc/hardware_errors" blob. + +(6) QEMU generates N ADD_POINTER commands, which patch addresses in the + "read_ack_register" fields of the HEST table with a pointer to the + corresponding "read_ack_register" within the "etc/hardware_errors" blob. + +(7) QEMU generates N ADD_POINTER commands for the firmware, which patch + addresses in the "error_block_address" fields with a pointer to the + respective "Error Status Data Block" in the "etc/hardware_errors" blob. + +(8) QEMU defines a third and write-only fw_cfg blob which is called + "etc/hardware_errors_addr". Through that blob, the firmware can send back + the guest-side allocation addresses to QEMU. The "etc/hardware_errors_addr" + blob contains a 8-byte entry. QEMU generates a single WRITE_POINTER command + for the firmware. The firmware will write back the start address of + "etc/hardware_errors" blob to the fw_cfg file "etc/hardware_errors_addr". + +(9) When QEMU gets a SIGBUS from the kernel, QEMU writes CPER into corresponding + "Error Status Data Block", guest memory, and then injects platform specific + interrupt (in case of arm/virt machine it's Synchronous External Abort) as a + notification which is necessary for notifying the guest. + +(10) This notification (in virtual hardware) will be handled by the guest + kernel, on receiving notification, guest APEI driver could read the CPER error + and take appropriate action. + +(11) kvm_arch_on_sigbus_vcpu() uses source_id as index in "etc/hardware_errors" to + find out "Error Status Data Block" entry corresponding to error source. So supported + source_id values should be assigned here and not be changed afterwards to make sure + that guest will write error into expected "Error Status Data Block" even if guest was + migrated to a newer QEMU. diff --git a/docs/specs/index.rst b/docs/specs/index.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/specs/index.rst +++ b/docs/specs/index.rst @@ -XXX,XX +XXX,XX @@ Contents: ppc-spapr-xive acpi_hw_reduced_hotplug tpm + acpi_hest_ghes -- 2.20.1 From: Dongjiu Geng <gengdongjiu@huawei.com> This patch builds error_block_address and read_ack_register fields in hardware errors table , the error_block_address points to Generic Error Status Block(GESB) via bios_linker. The max size for one GESB is 1kb, For more detailed information, please refer to document: docs/specs/acpi_hest_ghes.rst Now we only support one Error source, if necessary, we can extend to support more. Suggested-by: Laszlo Ersek <lersek@redhat.com> Signed-off-by: Xiang Zheng <zhengxiang9@huawei.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Dongjiu Geng <gengdongjiu@huawei.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Message-id: 20200512030609.19593-5-gengdongjiu@huawei.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- default-configs/arm-softmmu.mak | 1 + include/hw/acpi/aml-build.h | 1 + include/hw/acpi/ghes.h | 28 +++++++++++ hw/acpi/aml-build.c | 2 + hw/acpi/ghes.c | 89 +++++++++++++++++++++++++++++++++ hw/arm/virt-acpi-build.c | 5 ++ hw/acpi/Kconfig | 4 ++ hw/acpi/Makefile.objs | 1 + 8 files changed, 131 insertions(+) create mode 100644 include/hw/acpi/ghes.h create mode 100644 hw/acpi/ghes.c diff --git a/default-configs/arm-softmmu.mak b/default-configs/arm-softmmu.mak index XXXXXXX..XXXXXXX 100644 --- a/default-configs/arm-softmmu.mak +++ b/default-configs/arm-softmmu.mak @@ -XXX,XX +XXX,XX @@ CONFIG_FSL_IMX7=y CONFIG_FSL_IMX6UL=y CONFIG_SEMIHOSTING=y CONFIG_ALLWINNER_H3=y +CONFIG_ACPI_APEI=y diff --git a/include/hw/acpi/aml-build.h b/include/hw/acpi/aml-build.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/acpi/aml-build.h +++ b/include/hw/acpi/aml-build.h @@ -XXX,XX +XXX,XX @@ struct AcpiBuildTables { GArray *rsdp; GArray *tcpalog; GArray *vmgenid; + GArray *hardware_errors; BIOSLinker *linker; } AcpiBuildTables; diff --git a/include/hw/acpi/ghes.h b/include/hw/acpi/ghes.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/acpi/ghes.h @@ -XXX,XX +XXX,XX @@ +/* + * Support for generating APEI tables and recording CPER for Guests + * + * Copyright (c) 2020 HUAWEI TECHNOLOGIES CO., LTD. + * + * Author: Dongjiu Geng <gengdongjiu@huawei.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + + * You should have received a copy of the GNU General Public License along + * with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef ACPI_GHES_H +#define ACPI_GHES_H + +#include "hw/acpi/bios-linker-loader.h" + +void build_ghes_error_table(GArray *hardware_errors, BIOSLinker *linker); +#endif 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 acpi_build_tables_init(AcpiBuildTables *tables) tables->table_data = g_array_new(false, true /* clear */, 1); tables->tcpalog = g_array_new(false, true /* clear */, 1); tables->vmgenid = g_array_new(false, true /* clear */, 1); + tables->hardware_errors = g_array_new(false, true /* clear */, 1); tables->linker = bios_linker_loader_init(); } @@ -XXX,XX +XXX,XX @@ void acpi_build_tables_cleanup(AcpiBuildTables *tables, bool mfre) g_array_free(tables->table_data, true); g_array_free(tables->tcpalog, mfre); g_array_free(tables->vmgenid, mfre); + g_array_free(tables->hardware_errors, mfre); } /* diff --git a/hw/acpi/ghes.c b/hw/acpi/ghes.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/acpi/ghes.c @@ -XXX,XX +XXX,XX @@ +/* + * Support for generating APEI tables and recording CPER for Guests + * + * Copyright (c) 2020 HUAWEI TECHNOLOGIES CO., LTD. + * + * Author: Dongjiu Geng <gengdongjiu@huawei.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + + * You should have received a copy of the GNU General Public License along + * with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu/units.h" +#include "hw/acpi/ghes.h" +#include "hw/acpi/aml-build.h" + +#define ACPI_GHES_ERRORS_FW_CFG_FILE "etc/hardware_errors" +#define ACPI_GHES_DATA_ADDR_FW_CFG_FILE "etc/hardware_errors_addr" + +/* The max size in bytes for one error block */ +#define ACPI_GHES_MAX_RAW_DATA_LENGTH (1 * KiB) + +/* Now only support ARMv8 SEA notification type error source */ +#define ACPI_GHES_ERROR_SOURCE_COUNT 1 + +/* + * Build table for the hardware error fw_cfg blob. + * Initialize "etc/hardware_errors" and "etc/hardware_errors_addr" fw_cfg blobs. + * See docs/specs/acpi_hest_ghes.rst for blobs format. + */ +void build_ghes_error_table(GArray *hardware_errors, BIOSLinker *linker) +{ + int i, error_status_block_offset; + + /* Build error_block_address */ + for (i = 0; i < ACPI_GHES_ERROR_SOURCE_COUNT; i++) { + build_append_int_noprefix(hardware_errors, 0, sizeof(uint64_t)); + } + + /* Build read_ack_register */ + for (i = 0; i < ACPI_GHES_ERROR_SOURCE_COUNT; i++) { + /* + * Initialize the value of read_ack_register to 1, so GHES can be + * writeable after (re)boot. + * ACPI 6.2: 18.3.2.8 Generic Hardware Error Source version 2 + * (GHESv2 - Type 10) + */ + build_append_int_noprefix(hardware_errors, 1, sizeof(uint64_t)); + } + + /* Generic Error Status Block offset in the hardware error fw_cfg blob */ + error_status_block_offset = hardware_errors->len; + + /* Reserve space for Error Status Data Block */ + acpi_data_push(hardware_errors, + ACPI_GHES_MAX_RAW_DATA_LENGTH * ACPI_GHES_ERROR_SOURCE_COUNT); + + /* Tell guest firmware to place hardware_errors blob into RAM */ + bios_linker_loader_alloc(linker, ACPI_GHES_ERRORS_FW_CFG_FILE, + hardware_errors, sizeof(uint64_t), false); + + for (i = 0; i < ACPI_GHES_ERROR_SOURCE_COUNT; i++) { + /* + * Tell firmware to patch error_block_address entries to point to + * corresponding "Generic Error Status Block" + */ + bios_linker_loader_add_pointer(linker, + ACPI_GHES_ERRORS_FW_CFG_FILE, sizeof(uint64_t) * i, + sizeof(uint64_t), ACPI_GHES_ERRORS_FW_CFG_FILE, + error_status_block_offset + i * ACPI_GHES_MAX_RAW_DATA_LENGTH); + } + + /* + * tell firmware to write hardware_errors GPA into + * hardware_errors_addr fw_cfg, once the former has been initialized. + */ + bios_linker_loader_write_pointer(linker, ACPI_GHES_DATA_ADDR_FW_CFG_FILE, + 0, sizeof(uint64_t), ACPI_GHES_ERRORS_FW_CFG_FILE, 0); +} 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 @@ #include "sysemu/reset.h" #include "kvm_arm.h" #include "migration/vmstate.h" +#include "hw/acpi/ghes.h" #define ARM_SPI_BASE 32 @@ -XXX,XX +XXX,XX @@ void virt_acpi_build(VirtMachineState *vms, AcpiBuildTables *tables) acpi_add_table(table_offsets, tables_blob); build_spcr(tables_blob, tables->linker, vms); + if (vms->ras) { + build_ghes_error_table(tables->hardware_errors, tables->linker); + } + if (ms->numa_state->num_nodes > 0) { acpi_add_table(table_offsets, tables_blob); build_srat(tables_blob, tables->linker, vms); diff --git a/hw/acpi/Kconfig b/hw/acpi/Kconfig index XXXXXXX..XXXXXXX 100644 --- a/hw/acpi/Kconfig +++ b/hw/acpi/Kconfig @@ -XXX,XX +XXX,XX @@ config ACPI_HMAT bool depends on ACPI +config ACPI_APEI + bool + depends on ACPI + config ACPI_PCI bool depends on ACPI && PCI diff --git a/hw/acpi/Makefile.objs b/hw/acpi/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/acpi/Makefile.objs +++ b/hw/acpi/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_ACPI_NVDIMM) += nvdimm.o common-obj-$(CONFIG_ACPI_VMGENID) += vmgenid.o common-obj-$(CONFIG_ACPI_HW_REDUCED) += generic_event_device.o common-obj-$(CONFIG_ACPI_HMAT) += hmat.o +common-obj-$(CONFIG_ACPI_APEI) += ghes.o common-obj-$(call lnot,$(CONFIG_ACPI_X86)) += acpi-stub.o common-obj-$(call lnot,$(CONFIG_PC)) += acpi-x86-stub.o -- 2.20.1 From: Dongjiu Geng <gengdongjiu@huawei.com> This patch builds Hardware Error Source Table(HEST) via fw_cfg blobs. Now it only supports ARMv8 SEA, a type of Generic Hardware Error Source version 2(GHESv2) error source. Afterwards, we can extend the supported types if needed. For the CPER section, currently it is memory section because kernel mainly wants userspace to handle the memory errors. This patch follows the spec ACPI 6.2 to build the Hardware Error Source table. For more detailed information, please refer to document: docs/specs/acpi_hest_ghes.rst build_ghes_hw_error_notification() helper will help to add Hardware Error Notification to ACPI tables without using packed C structures and avoid endianness issues as API doesn't need explicit conversion. Signed-off-by: Xiang Zheng <zhengxiang9@huawei.com> Signed-off-by: Dongjiu Geng <gengdongjiu@huawei.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Message-id: 20200512030609.19593-6-gengdongjiu@huawei.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/acpi/ghes.h | 39 ++++++++++++ hw/acpi/ghes.c | 126 +++++++++++++++++++++++++++++++++++++++ hw/arm/virt-acpi-build.c | 2 + 3 files changed, 167 insertions(+) diff --git a/include/hw/acpi/ghes.h b/include/hw/acpi/ghes.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/acpi/ghes.h +++ b/include/hw/acpi/ghes.h @@ -XXX,XX +XXX,XX @@ #include "hw/acpi/bios-linker-loader.h" +/* + * Values for Hardware Error Notification Type field + */ +enum AcpiGhesNotifyType { + /* Polled */ + ACPI_GHES_NOTIFY_POLLED = 0, + /* External Interrupt */ + ACPI_GHES_NOTIFY_EXTERNAL = 1, + /* Local Interrupt */ + ACPI_GHES_NOTIFY_LOCAL = 2, + /* SCI */ + ACPI_GHES_NOTIFY_SCI = 3, + /* NMI */ + ACPI_GHES_NOTIFY_NMI = 4, + /* CMCI, ACPI 5.0: 18.3.2.7, Table 18-290 */ + ACPI_GHES_NOTIFY_CMCI = 5, + /* MCE, ACPI 5.0: 18.3.2.7, Table 18-290 */ + ACPI_GHES_NOTIFY_MCE = 6, + /* GPIO-Signal, ACPI 6.0: 18.3.2.7, Table 18-332 */ + ACPI_GHES_NOTIFY_GPIO = 7, + /* ARMv8 SEA, ACPI 6.1: 18.3.2.9, Table 18-345 */ + ACPI_GHES_NOTIFY_SEA = 8, + /* ARMv8 SEI, ACPI 6.1: 18.3.2.9, Table 18-345 */ + ACPI_GHES_NOTIFY_SEI = 9, + /* External Interrupt - GSIV, ACPI 6.1: 18.3.2.9, Table 18-345 */ + ACPI_GHES_NOTIFY_GSIV = 10, + /* Software Delegated Exception, ACPI 6.2: 18.3.2.9, Table 18-383 */ + ACPI_GHES_NOTIFY_SDEI = 11, + /* 12 and greater are reserved */ + ACPI_GHES_NOTIFY_RESERVED = 12 +}; + +enum { + ACPI_HEST_SRC_ID_SEA = 0, + /* future ids go here */ + ACPI_HEST_SRC_ID_RESERVED, +}; + void build_ghes_error_table(GArray *hardware_errors, BIOSLinker *linker); +void acpi_build_hest(GArray *table_data, BIOSLinker *linker); #endif diff --git a/hw/acpi/ghes.c b/hw/acpi/ghes.c index XXXXXXX..XXXXXXX 100644 --- a/hw/acpi/ghes.c +++ b/hw/acpi/ghes.c @@ -XXX,XX +XXX,XX @@ #include "qemu/units.h" #include "hw/acpi/ghes.h" #include "hw/acpi/aml-build.h" +#include "qemu/error-report.h" #define ACPI_GHES_ERRORS_FW_CFG_FILE "etc/hardware_errors" #define ACPI_GHES_DATA_ADDR_FW_CFG_FILE "etc/hardware_errors_addr" @@ -XXX,XX +XXX,XX @@ /* Now only support ARMv8 SEA notification type error source */ #define ACPI_GHES_ERROR_SOURCE_COUNT 1 +/* Generic Hardware Error Source version 2 */ +#define ACPI_GHES_SOURCE_GENERIC_ERROR_V2 10 + +/* Address offset in Generic Address Structure(GAS) */ +#define GAS_ADDR_OFFSET 4 + +/* + * Hardware Error Notification + * ACPI 4.0: 17.3.2.7 Hardware Error Notification + * Composes dummy Hardware Error Notification descriptor of specified type + */ +static void build_ghes_hw_error_notification(GArray *table, const uint8_t type) +{ + /* Type */ + build_append_int_noprefix(table, type, 1); + /* + * Length: + * Total length of the structure in bytes + */ + build_append_int_noprefix(table, 28, 1); + /* Configuration Write Enable */ + build_append_int_noprefix(table, 0, 2); + /* Poll Interval */ + build_append_int_noprefix(table, 0, 4); + /* Vector */ + build_append_int_noprefix(table, 0, 4); + /* Switch To Polling Threshold Value */ + build_append_int_noprefix(table, 0, 4); + /* Switch To Polling Threshold Window */ + build_append_int_noprefix(table, 0, 4); + /* Error Threshold Value */ + build_append_int_noprefix(table, 0, 4); + /* Error Threshold Window */ + build_append_int_noprefix(table, 0, 4); +} + /* * Build table for the hardware error fw_cfg blob. * Initialize "etc/hardware_errors" and "etc/hardware_errors_addr" fw_cfg blobs. @@ -XXX,XX +XXX,XX @@ void build_ghes_error_table(GArray *hardware_errors, BIOSLinker *linker) bios_linker_loader_write_pointer(linker, ACPI_GHES_DATA_ADDR_FW_CFG_FILE, 0, sizeof(uint64_t), ACPI_GHES_ERRORS_FW_CFG_FILE, 0); } + +/* Build Generic Hardware Error Source version 2 (GHESv2) */ +static void build_ghes_v2(GArray *table_data, int source_id, BIOSLinker *linker) +{ + uint64_t address_offset; + /* + * Type: + * Generic Hardware Error Source version 2(GHESv2 - Type 10) + */ + build_append_int_noprefix(table_data, ACPI_GHES_SOURCE_GENERIC_ERROR_V2, 2); + /* Source Id */ + build_append_int_noprefix(table_data, source_id, 2); + /* Related Source Id */ + build_append_int_noprefix(table_data, 0xffff, 2); + /* Flags */ + build_append_int_noprefix(table_data, 0, 1); + /* Enabled */ + build_append_int_noprefix(table_data, 1, 1); + + /* Number of Records To Pre-allocate */ + build_append_int_noprefix(table_data, 1, 4); + /* Max Sections Per Record */ + build_append_int_noprefix(table_data, 1, 4); + /* Max Raw Data Length */ + build_append_int_noprefix(table_data, ACPI_GHES_MAX_RAW_DATA_LENGTH, 4); + + address_offset = table_data->len; + /* Error Status Address */ + build_append_gas(table_data, AML_AS_SYSTEM_MEMORY, 0x40, 0, + 4 /* QWord access */, 0); + bios_linker_loader_add_pointer(linker, ACPI_BUILD_TABLE_FILE, + address_offset + GAS_ADDR_OFFSET, sizeof(uint64_t), + ACPI_GHES_ERRORS_FW_CFG_FILE, source_id * sizeof(uint64_t)); + + switch (source_id) { + case ACPI_HEST_SRC_ID_SEA: + /* + * Notification Structure + * Now only enable ARMv8 SEA notification type + */ + build_ghes_hw_error_notification(table_data, ACPI_GHES_NOTIFY_SEA); + break; + default: + error_report("Not support this error source"); + abort(); + } + + /* Error Status Block Length */ + build_append_int_noprefix(table_data, ACPI_GHES_MAX_RAW_DATA_LENGTH, 4); + + /* + * Read Ack Register + * ACPI 6.1: 18.3.2.8 Generic Hardware Error Source + * version 2 (GHESv2 - Type 10) + */ + address_offset = table_data->len; + build_append_gas(table_data, AML_AS_SYSTEM_MEMORY, 0x40, 0, + 4 /* QWord access */, 0); + bios_linker_loader_add_pointer(linker, ACPI_BUILD_TABLE_FILE, + address_offset + GAS_ADDR_OFFSET, + sizeof(uint64_t), ACPI_GHES_ERRORS_FW_CFG_FILE, + (ACPI_GHES_ERROR_SOURCE_COUNT + source_id) * sizeof(uint64_t)); + + /* + * Read Ack Preserve field + * We only provide the first bit in Read Ack Register to OSPM to write + * while the other bits are preserved. + */ + build_append_int_noprefix(table_data, ~0x1ULL, 8); + /* Read Ack Write */ + build_append_int_noprefix(table_data, 0x1, 8); +} + +/* Build Hardware Error Source Table */ +void acpi_build_hest(GArray *table_data, BIOSLinker *linker) +{ + uint64_t hest_start = table_data->len; + + /* Hardware Error Source Table header*/ + acpi_data_push(table_data, sizeof(AcpiTableHeader)); + + /* Error Source Count */ + build_append_int_noprefix(table_data, ACPI_GHES_ERROR_SOURCE_COUNT, 4); + + build_ghes_v2(table_data, ACPI_HEST_SRC_ID_SEA, linker); + + build_header(linker, table_data, (void *)(table_data->data + hest_start), + "HEST", table_data->len - hest_start, 1, NULL, NULL); +} 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 @@ void virt_acpi_build(VirtMachineState *vms, AcpiBuildTables *tables) if (vms->ras) { build_ghes_error_table(tables->hardware_errors, tables->linker); + acpi_add_table(table_offsets, tables_blob); + acpi_build_hest(tables_blob, tables->linker); } if (ms->numa_state->num_nodes > 0) { -- 2.20.1 From: Dongjiu Geng <gengdongjiu@huawei.com> Record the GHEB address via fw_cfg file, when recording a error to CPER, it will use this address to find out Generic Error Data Entries and write the error. In order to avoid migration failure, make hardware error table address to a part of GED device instead of global variable, then this address will be migrated to target QEMU. Acked-by: Xiang Zheng <zhengxiang9@huawei.com> Signed-off-by: Dongjiu Geng <gengdongjiu@huawei.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Message-id: 20200512030609.19593-7-gengdongjiu@huawei.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/acpi/generic_event_device.h | 2 ++ include/hw/acpi/ghes.h | 6 ++++++ hw/acpi/generic_event_device.c | 19 +++++++++++++++++++ hw/acpi/ghes.c | 14 ++++++++++++++ hw/arm/virt-acpi-build.c | 8 ++++++++ 5 files changed, 49 insertions(+) diff --git a/include/hw/acpi/generic_event_device.h b/include/hw/acpi/generic_event_device.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/acpi/generic_event_device.h +++ b/include/hw/acpi/generic_event_device.h @@ -XXX,XX +XXX,XX @@ #include "hw/sysbus.h" #include "hw/acpi/memory_hotplug.h" +#include "hw/acpi/ghes.h" #define ACPI_POWER_BUTTON_DEVICE "PWRB" @@ -XXX,XX +XXX,XX @@ typedef struct AcpiGedState { GEDState ged_state; uint32_t ged_event_bitmap; qemu_irq irq; + AcpiGhesState ghes_state; } AcpiGedState; void build_ged_aml(Aml *table, const char* name, HotplugHandler *hotplug_dev, diff --git a/include/hw/acpi/ghes.h b/include/hw/acpi/ghes.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/acpi/ghes.h +++ b/include/hw/acpi/ghes.h @@ -XXX,XX +XXX,XX @@ enum { ACPI_HEST_SRC_ID_RESERVED, }; +typedef struct AcpiGhesState { + uint64_t ghes_addr_le; +} AcpiGhesState; + void build_ghes_error_table(GArray *hardware_errors, BIOSLinker *linker); void acpi_build_hest(GArray *table_data, BIOSLinker *linker); +void acpi_ghes_add_fw_cfg(AcpiGhesState *vms, FWCfgState *s, + GArray *hardware_errors); #endif diff --git a/hw/acpi/generic_event_device.c b/hw/acpi/generic_event_device.c index XXXXXXX..XXXXXXX 100644 --- a/hw/acpi/generic_event_device.c +++ b/hw/acpi/generic_event_device.c @@ -XXX,XX +XXX,XX @@ static const VMStateDescription vmstate_ged_state = { } }; +static bool ghes_needed(void *opaque) +{ + AcpiGedState *s = opaque; + return s->ghes_state.ghes_addr_le; +} + +static const VMStateDescription vmstate_ghes_state = { + .name = "acpi-ged/ghes", + .version_id = 1, + .minimum_version_id = 1, + .needed = ghes_needed, + .fields = (VMStateField[]) { + VMSTATE_STRUCT(ghes_state, AcpiGedState, 1, + vmstate_ghes_state, AcpiGhesState), + VMSTATE_END_OF_LIST() + } +}; + static const VMStateDescription vmstate_acpi_ged = { .name = "acpi-ged", .version_id = 1, @@ -XXX,XX +XXX,XX @@ static const VMStateDescription vmstate_acpi_ged = { }, .subsections = (const VMStateDescription * []) { &vmstate_memhp_state, + &vmstate_ghes_state, NULL } }; diff --git a/hw/acpi/ghes.c b/hw/acpi/ghes.c index XXXXXXX..XXXXXXX 100644 --- a/hw/acpi/ghes.c +++ b/hw/acpi/ghes.c @@ -XXX,XX +XXX,XX @@ #include "hw/acpi/ghes.h" #include "hw/acpi/aml-build.h" #include "qemu/error-report.h" +#include "hw/acpi/generic_event_device.h" +#include "hw/nvram/fw_cfg.h" #define ACPI_GHES_ERRORS_FW_CFG_FILE "etc/hardware_errors" #define ACPI_GHES_DATA_ADDR_FW_CFG_FILE "etc/hardware_errors_addr" @@ -XXX,XX +XXX,XX @@ void acpi_build_hest(GArray *table_data, BIOSLinker *linker) build_header(linker, table_data, (void *)(table_data->data + hest_start), "HEST", table_data->len - hest_start, 1, NULL, NULL); } + +void acpi_ghes_add_fw_cfg(AcpiGhesState *ags, FWCfgState *s, + GArray *hardware_error) +{ + /* Create a read-only fw_cfg file for GHES */ + fw_cfg_add_file(s, ACPI_GHES_ERRORS_FW_CFG_FILE, hardware_error->data, + hardware_error->len); + + /* Create a read-write fw_cfg file for Address */ + fw_cfg_add_file_callback(s, ACPI_GHES_DATA_ADDR_FW_CFG_FILE, NULL, NULL, + NULL, &(ags->ghes_addr_le), sizeof(ags->ghes_addr_le), false); +} 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 @@ void virt_acpi_setup(VirtMachineState *vms) { AcpiBuildTables tables; AcpiBuildState *build_state; + AcpiGedState *acpi_ged_state; if (!vms->fw_cfg) { trace_virt_acpi_setup(); @@ -XXX,XX +XXX,XX @@ void virt_acpi_setup(VirtMachineState *vms) fw_cfg_add_file(vms->fw_cfg, ACPI_BUILD_TPMLOG_FILE, tables.tcpalog->data, acpi_data_len(tables.tcpalog)); + if (vms->ras) { + assert(vms->acpi_dev); + acpi_ged_state = ACPI_GED(vms->acpi_dev); + acpi_ghes_add_fw_cfg(&acpi_ged_state->ghes_state, + vms->fw_cfg, tables.hardware_errors); + } + build_state->rsdp_mr = acpi_add_rom_blob(virt_acpi_build_update, build_state, tables.rsdp, ACPI_BUILD_RSDP_FILE, 0); -- 2.20.1 From: Dongjiu Geng <gengdongjiu@huawei.com> kvm_hwpoison_page_add() and kvm_unpoison_all() will both be used by X86 and ARM platforms, so moving them into "accel/kvm/kvm-all.c" to avoid duplicate code. For architectures that don't use the poison-list functionality the reset handler will harmlessly do nothing, so let's register the kvm_unpoison_all() function in the generic kvm_init() function. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Dongjiu Geng <gengdongjiu@huawei.com> Signed-off-by: Xiang Zheng <zhengxiang9@huawei.com> Acked-by: Xiang Zheng <zhengxiang9@huawei.com> Message-id: 20200512030609.19593-8-gengdongjiu@huawei.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/sysemu/kvm_int.h | 12 ++++++++++++ accel/kvm/kvm-all.c | 36 ++++++++++++++++++++++++++++++++++++ target/i386/kvm.c | 36 ------------------------------------ 3 files changed, 48 insertions(+), 36 deletions(-) diff --git a/include/sysemu/kvm_int.h b/include/sysemu/kvm_int.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/kvm_int.h +++ b/include/sysemu/kvm_int.h @@ -XXX,XX +XXX,XX @@ void kvm_memory_listener_register(KVMState *s, KVMMemoryListener *kml, AddressSpace *as, int as_id); void kvm_set_max_memslot_size(hwaddr max_slot_size); + +/** + * kvm_hwpoison_page_add: + * + * Parameters: + * @ram_addr: the address in the RAM for the poisoned page + * + * Add a poisoned page to the list + * + * Return: None. + */ +void kvm_hwpoison_page_add(ram_addr_t ram_addr); #endif diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/kvm/kvm-all.c +++ b/accel/kvm/kvm-all.c @@ -XXX,XX +XXX,XX @@ #include "qapi/visitor.h" #include "qapi/qapi-types-common.h" #include "qapi/qapi-visit-common.h" +#include "sysemu/reset.h" #include "hw/boards.h" @@ -XXX,XX +XXX,XX @@ int kvm_vm_check_extension(KVMState *s, unsigned int extension) return ret; } +typedef struct HWPoisonPage { + ram_addr_t ram_addr; + QLIST_ENTRY(HWPoisonPage) list; +} HWPoisonPage; + +static QLIST_HEAD(, HWPoisonPage) hwpoison_page_list = + QLIST_HEAD_INITIALIZER(hwpoison_page_list); + +static void kvm_unpoison_all(void *param) +{ + HWPoisonPage *page, *next_page; + + QLIST_FOREACH_SAFE(page, &hwpoison_page_list, list, next_page) { + QLIST_REMOVE(page, list); + qemu_ram_remap(page->ram_addr, TARGET_PAGE_SIZE); + g_free(page); + } +} + +void kvm_hwpoison_page_add(ram_addr_t ram_addr) +{ + HWPoisonPage *page; + + QLIST_FOREACH(page, &hwpoison_page_list, list) { + if (page->ram_addr == ram_addr) { + return; + } + } + page = g_new(HWPoisonPage, 1); + page->ram_addr = ram_addr; + QLIST_INSERT_HEAD(&hwpoison_page_list, page, list); +} + static uint32_t adjust_ioeventfd_endianness(uint32_t val, uint32_t size) { #if defined(HOST_WORDS_BIGENDIAN) != defined(TARGET_WORDS_BIGENDIAN) @@ -XXX,XX +XXX,XX @@ static int kvm_init(MachineState *ms) s->kernel_irqchip_split = mc->default_kernel_irqchip_split ? ON_OFF_AUTO_ON : ON_OFF_AUTO_OFF; } + qemu_register_reset(kvm_unpoison_all, NULL); + if (s->kernel_irqchip_allowed) { kvm_irqchip_create(s); } diff --git a/target/i386/kvm.c b/target/i386/kvm.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/kvm.c +++ b/target/i386/kvm.c @@ -XXX,XX +XXX,XX @@ #include "sysemu/sysemu.h" #include "sysemu/hw_accel.h" #include "sysemu/kvm_int.h" -#include "sysemu/reset.h" #include "sysemu/runstate.h" #include "kvm_i386.h" #include "hyperv.h" @@ -XXX,XX +XXX,XX @@ uint64_t kvm_arch_get_supported_msr_feature(KVMState *s, uint32_t index) } } - -typedef struct HWPoisonPage { - ram_addr_t ram_addr; - QLIST_ENTRY(HWPoisonPage) list; -} HWPoisonPage; - -static QLIST_HEAD(, HWPoisonPage) hwpoison_page_list = - QLIST_HEAD_INITIALIZER(hwpoison_page_list); - -static void kvm_unpoison_all(void *param) -{ - HWPoisonPage *page, *next_page; - - QLIST_FOREACH_SAFE(page, &hwpoison_page_list, list, next_page) { - QLIST_REMOVE(page, list); - qemu_ram_remap(page->ram_addr, TARGET_PAGE_SIZE); - g_free(page); - } -} - -static void kvm_hwpoison_page_add(ram_addr_t ram_addr) -{ - HWPoisonPage *page; - - QLIST_FOREACH(page, &hwpoison_page_list, list) { - if (page->ram_addr == ram_addr) { - return; - } - } - page = g_new(HWPoisonPage, 1); - page->ram_addr = ram_addr; - QLIST_INSERT_HEAD(&hwpoison_page_list, page, list); -} - static int kvm_get_mce_cap_supported(KVMState *s, uint64_t *mce_cap, int *max_banks) { @@ -XXX,XX +XXX,XX @@ int kvm_arch_init(MachineState *ms, KVMState *s) fprintf(stderr, "e820_add_entry() table is full\n"); return ret; } - qemu_register_reset(kvm_unpoison_all, NULL); shadow_mem = object_property_get_int(OBJECT(s), "kvm-shadow-mem", &error_abort); if (shadow_mem != -1) { -- 2.20.1 From: Dongjiu Geng <gengdongjiu@huawei.com> kvm_arch_on_sigbus_vcpu() error injection uses source_id as index in etc/hardware_errors to find out Error Status Data Block entry corresponding to error source. So supported source_id values should be assigned here and not be changed afterwards to make sure that guest will write error into expected Error Status Data Block. Before QEMU writes a new error to ACPI table, it will check whether previous error has been acknowledged. If not acknowledged, the new errors will be ignored and not be recorded. For the errors section type, QEMU simulate it to memory section error. Signed-off-by: Dongjiu Geng <gengdongjiu@huawei.com> Signed-off-by: Xiang Zheng <zhengxiang9@huawei.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Message-id: 20200512030609.19593-9-gengdongjiu@huawei.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/acpi/ghes.h | 1 + hw/acpi/ghes.c | 219 +++++++++++++++++++++++++++++++++++++++++ 2 files changed, 220 insertions(+) diff --git a/include/hw/acpi/ghes.h b/include/hw/acpi/ghes.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/acpi/ghes.h +++ b/include/hw/acpi/ghes.h @@ -XXX,XX +XXX,XX @@ void build_ghes_error_table(GArray *hardware_errors, BIOSLinker *linker); void acpi_build_hest(GArray *table_data, BIOSLinker *linker); void acpi_ghes_add_fw_cfg(AcpiGhesState *vms, FWCfgState *s, GArray *hardware_errors); +int acpi_ghes_record_errors(uint8_t notify, uint64_t error_physical_addr); #endif diff --git a/hw/acpi/ghes.c b/hw/acpi/ghes.c index XXXXXXX..XXXXXXX 100644 --- a/hw/acpi/ghes.c +++ b/hw/acpi/ghes.c @@ -XXX,XX +XXX,XX @@ #include "qemu/error-report.h" #include "hw/acpi/generic_event_device.h" #include "hw/nvram/fw_cfg.h" +#include "qemu/uuid.h" #define ACPI_GHES_ERRORS_FW_CFG_FILE "etc/hardware_errors" #define ACPI_GHES_DATA_ADDR_FW_CFG_FILE "etc/hardware_errors_addr" @@ -XXX,XX +XXX,XX @@ /* Address offset in Generic Address Structure(GAS) */ #define GAS_ADDR_OFFSET 4 +/* + * The total size of Generic Error Data Entry + * ACPI 6.1/6.2: 18.3.2.7.1 Generic Error Data, + * Table 18-343 Generic Error Data Entry + */ +#define ACPI_GHES_DATA_LENGTH 72 + +/* The memory section CPER size, UEFI 2.6: N.2.5 Memory Error Section */ +#define ACPI_GHES_MEM_CPER_LENGTH 80 + +/* Masks for block_status flags */ +#define ACPI_GEBS_UNCORRECTABLE 1 + +/* + * Total size for Generic Error Status Block except Generic Error Data Entries + * ACPI 6.2: 18.3.2.7.1 Generic Error Data, + * Table 18-380 Generic Error Status Block + */ +#define ACPI_GHES_GESB_SIZE 20 + +/* + * Values for error_severity field + */ +enum AcpiGenericErrorSeverity { + ACPI_CPER_SEV_RECOVERABLE = 0, + ACPI_CPER_SEV_FATAL = 1, + ACPI_CPER_SEV_CORRECTED = 2, + ACPI_CPER_SEV_NONE = 3, +}; + /* * Hardware Error Notification * ACPI 4.0: 17.3.2.7 Hardware Error Notification @@ -XXX,XX +XXX,XX @@ static void build_ghes_hw_error_notification(GArray *table, const uint8_t type) build_append_int_noprefix(table, 0, 4); } +/* + * Generic Error Data Entry + * ACPI 6.1: 18.3.2.7.1 Generic Error Data + */ +static void acpi_ghes_generic_error_data(GArray *table, + const uint8_t *section_type, uint32_t error_severity, + uint8_t validation_bits, uint8_t flags, + uint32_t error_data_length, QemuUUID fru_id, + uint64_t time_stamp) +{ + const uint8_t fru_text[20] = {0}; + + /* Section Type */ + g_array_append_vals(table, section_type, 16); + + /* Error Severity */ + build_append_int_noprefix(table, error_severity, 4); + /* Revision */ + build_append_int_noprefix(table, 0x300, 2); + /* Validation Bits */ + build_append_int_noprefix(table, validation_bits, 1); + /* Flags */ + build_append_int_noprefix(table, flags, 1); + /* Error Data Length */ + build_append_int_noprefix(table, error_data_length, 4); + + /* FRU Id */ + g_array_append_vals(table, fru_id.data, ARRAY_SIZE(fru_id.data)); + + /* FRU Text */ + g_array_append_vals(table, fru_text, sizeof(fru_text)); + + /* Timestamp */ + build_append_int_noprefix(table, time_stamp, 8); +} + +/* + * Generic Error Status Block + * ACPI 6.1: 18.3.2.7.1 Generic Error Data + */ +static void acpi_ghes_generic_error_status(GArray *table, uint32_t block_status, + uint32_t raw_data_offset, uint32_t raw_data_length, + uint32_t data_length, uint32_t error_severity) +{ + /* Block Status */ + build_append_int_noprefix(table, block_status, 4); + /* Raw Data Offset */ + build_append_int_noprefix(table, raw_data_offset, 4); + /* Raw Data Length */ + build_append_int_noprefix(table, raw_data_length, 4); + /* Data Length */ + build_append_int_noprefix(table, data_length, 4); + /* Error Severity */ + build_append_int_noprefix(table, error_severity, 4); +} + +/* UEFI 2.6: N.2.5 Memory Error Section */ +static void acpi_ghes_build_append_mem_cper(GArray *table, + uint64_t error_physical_addr) +{ + /* + * Memory Error Record + */ + + /* Validation Bits */ + build_append_int_noprefix(table, + (1ULL << 14) | /* Type Valid */ + (1ULL << 1) /* Physical Address Valid */, + 8); + /* Error Status */ + build_append_int_noprefix(table, 0, 8); + /* Physical Address */ + build_append_int_noprefix(table, error_physical_addr, 8); + /* Skip all the detailed information normally found in such a record */ + build_append_int_noprefix(table, 0, 48); + /* Memory Error Type */ + build_append_int_noprefix(table, 0 /* Unknown error */, 1); + /* Skip all the detailed information normally found in such a record */ + build_append_int_noprefix(table, 0, 7); +} + +static int acpi_ghes_record_mem_error(uint64_t error_block_address, + uint64_t error_physical_addr) +{ + GArray *block; + + /* Memory Error Section Type */ + const uint8_t uefi_cper_mem_sec[] = + UUID_LE(0xA5BC1114, 0x6F64, 0x4EDE, 0xB8, 0x63, 0x3E, 0x83, \ + 0xED, 0x7C, 0x83, 0xB1); + + /* invalid fru id: ACPI 4.0: 17.3.2.6.1 Generic Error Data, + * Table 17-13 Generic Error Data Entry + */ + QemuUUID fru_id = {}; + uint32_t data_length; + + block = g_array_new(false, true /* clear */, 1); + + /* This is the length if adding a new generic error data entry*/ + data_length = ACPI_GHES_DATA_LENGTH + ACPI_GHES_MEM_CPER_LENGTH; + + /* + * Check whether it will run out of the preallocated memory if adding a new + * generic error data entry + */ + if ((data_length + ACPI_GHES_GESB_SIZE) > ACPI_GHES_MAX_RAW_DATA_LENGTH) { + error_report("Not enough memory to record new CPER!!!"); + g_array_free(block, true); + return -1; + } + + /* Build the new generic error status block header */ + acpi_ghes_generic_error_status(block, ACPI_GEBS_UNCORRECTABLE, + 0, 0, data_length, ACPI_CPER_SEV_RECOVERABLE); + + /* Build this new generic error data entry header */ + acpi_ghes_generic_error_data(block, uefi_cper_mem_sec, + ACPI_CPER_SEV_RECOVERABLE, 0, 0, + ACPI_GHES_MEM_CPER_LENGTH, fru_id, 0); + + /* Build the memory section CPER for above new generic error data entry */ + acpi_ghes_build_append_mem_cper(block, error_physical_addr); + + /* Write the generic error data entry into guest memory */ + cpu_physical_memory_write(error_block_address, block->data, block->len); + + g_array_free(block, true); + + return 0; +} + /* * Build table for the hardware error fw_cfg blob. * Initialize "etc/hardware_errors" and "etc/hardware_errors_addr" fw_cfg blobs. @@ -XXX,XX +XXX,XX @@ void acpi_ghes_add_fw_cfg(AcpiGhesState *ags, FWCfgState *s, fw_cfg_add_file_callback(s, ACPI_GHES_DATA_ADDR_FW_CFG_FILE, NULL, NULL, NULL, &(ags->ghes_addr_le), sizeof(ags->ghes_addr_le), false); } + +int acpi_ghes_record_errors(uint8_t source_id, uint64_t physical_address) +{ + uint64_t error_block_addr, read_ack_register_addr, read_ack_register = 0; + uint64_t start_addr; + bool ret = -1; + AcpiGedState *acpi_ged_state; + AcpiGhesState *ags; + + assert(source_id < ACPI_HEST_SRC_ID_RESERVED); + + acpi_ged_state = ACPI_GED(object_resolve_path_type("", TYPE_ACPI_GED, + NULL)); + g_assert(acpi_ged_state); + ags = &acpi_ged_state->ghes_state; + + start_addr = le64_to_cpu(ags->ghes_addr_le); + + if (physical_address) { + + if (source_id < ACPI_HEST_SRC_ID_RESERVED) { + start_addr += source_id * sizeof(uint64_t); + } + + cpu_physical_memory_read(start_addr, &error_block_addr, + sizeof(error_block_addr)); + + error_block_addr = le64_to_cpu(error_block_addr); + + read_ack_register_addr = start_addr + + ACPI_GHES_ERROR_SOURCE_COUNT * sizeof(uint64_t); + + cpu_physical_memory_read(read_ack_register_addr, + &read_ack_register, sizeof(read_ack_register)); + + /* zero means OSPM does not acknowledge the error */ + if (!read_ack_register) { + error_report("OSPM does not acknowledge previous error," + " so can not record CPER for current error anymore"); + } else if (error_block_addr) { + read_ack_register = cpu_to_le64(0); + /* + * Clear the Read Ack Register, OSPM will write it to 1 when + * it acknowledges this error. + */ + cpu_physical_memory_write(read_ack_register_addr, + &read_ack_register, sizeof(uint64_t)); + + ret = acpi_ghes_record_mem_error(error_block_addr, + physical_address); + } else + error_report("can not find Generic Error Status Block"); + } + + return ret; +} -- 2.20.1 From: Dongjiu Geng <gengdongjiu@huawei.com> Add a SIGBUS signal handler. In this handler, it checks the SIGBUS type, translates the host VA delivered by host to guest PA, then fills this PA to guest APEI GHES memory, then notifies guest according to the SIGBUS type. When guest accesses the poisoned memory, it will generate a Synchronous External Abort(SEA). Then host kernel gets an APEI notification and calls memory_failure() to unmapped the affected page in stage 2, finally returns to guest. Guest continues to access the PG_hwpoison page, it will trap to KVM as stage2 fault, then a SIGBUS_MCEERR_AR synchronous signal is delivered to Qemu, Qemu records this error address into guest APEI GHES memory and notifes guest using Synchronous-External-Abort(SEA). In order to inject a vSEA, we introduce the kvm_inject_arm_sea() function in which we can setup the type of exception and the syndrome information. When switching to guest, the target vcpu will jump to the synchronous external abort vector table entry. The ESR_ELx.DFSC is set to synchronous external abort(0x10), and the ESR_ELx.FnV is set to not valid(0x1), which will tell guest that FAR is not valid and hold an UNKNOWN value. These values will be set to KVM register structures through KVM_SET_ONE_REG IOCTL. Signed-off-by: Dongjiu Geng <gengdongjiu@huawei.com> Signed-off-by: Xiang Zheng <zhengxiang9@huawei.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Xiang Zheng <zhengxiang9@huawei.com> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Message-id: 20200512030609.19593-10-gengdongjiu@huawei.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/sysemu/kvm.h | 3 +- target/arm/cpu.h | 4 +++ target/arm/internals.h | 5 +-- target/i386/cpu.h | 2 ++ target/arm/helper.c | 2 +- target/arm/kvm64.c | 77 +++++++++++++++++++++++++++++++++++++++++ target/arm/tlb_helper.c | 2 +- 7 files changed, 89 insertions(+), 6 deletions(-) diff --git a/include/sysemu/kvm.h b/include/sysemu/kvm.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/kvm.h +++ b/include/sysemu/kvm.h @@ -XXX,XX +XXX,XX @@ bool kvm_vcpu_id_is_valid(int vcpu_id); /* Returns VCPU ID to be used on KVM_CREATE_VCPU ioctl() */ unsigned long kvm_arch_vcpu_id(CPUState *cpu); -#ifdef TARGET_I386 -#define KVM_HAVE_MCE_INJECTION 1 +#ifdef KVM_HAVE_MCE_INJECTION void kvm_arch_on_sigbus_vcpu(CPUState *cpu, int code, void *addr); #endif 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 @@ /* ARM processors have a weak memory model */ #define TCG_GUEST_DEFAULT_MO (0) +#ifdef TARGET_AARCH64 +#define KVM_HAVE_MCE_INJECTION 1 +#endif + #define EXCP_UDEF 1 /* undefined instruction */ #define EXCP_SWI 2 /* software interrupt */ #define EXCP_PREFETCH_ABORT 3 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 uint32_t syn_insn_abort(int same_el, int ea, int s1ptw, int fsc) | ARM_EL_IL | (ea << 9) | (s1ptw << 7) | fsc; } -static inline uint32_t syn_data_abort_no_iss(int same_el, +static inline uint32_t syn_data_abort_no_iss(int same_el, int fnv, int ea, int cm, int s1ptw, int wnr, int fsc) { return (EC_DATAABORT << ARM_EL_EC_SHIFT) | (same_el << ARM_EL_EC_SHIFT) | ARM_EL_IL - | (ea << 9) | (cm << 8) | (s1ptw << 7) | (wnr << 6) | fsc; + | (fnv << 10) | (ea << 9) | (cm << 8) | (s1ptw << 7) + | (wnr << 6) | fsc; } static inline uint32_t syn_data_abort_with_iss(int same_el, diff --git a/target/i386/cpu.h b/target/i386/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/i386/cpu.h +++ b/target/i386/cpu.h @@ -XXX,XX +XXX,XX @@ /* The x86 has a strong memory model with some store-after-load re-ordering */ #define TCG_GUEST_DEFAULT_MO (TCG_MO_ALL & ~TCG_MO_ST_LD) +#define KVM_HAVE_MCE_INJECTION 1 + /* Maximum instruction code size */ #define TARGET_MAX_INSN_SIZE 16 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 do_ats_write(CPUARMState *env, uint64_t value, * Report exception with ESR indicating a fault due to a * translation table walk for a cache maintenance instruction. */ - syn = syn_data_abort_no_iss(current_el == target_el, + syn = syn_data_abort_no_iss(current_el == target_el, 0, fi.ea, 1, fi.s1ptw, 1, fsc); env->exception.vaddress = value; env->exception.fsr = fsr; diff --git a/target/arm/kvm64.c b/target/arm/kvm64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/kvm64.c +++ b/target/arm/kvm64.c @@ -XXX,XX +XXX,XX @@ #include "sysemu/kvm_int.h" #include "kvm_arm.h" #include "internals.h" +#include "hw/acpi/acpi.h" +#include "hw/acpi/ghes.h" +#include "hw/arm/virt.h" static bool have_guest_debug; @@ -XXX,XX +XXX,XX @@ int kvm_arm_cpreg_level(uint64_t regidx) return KVM_PUT_RUNTIME_STATE; } +/* Callers must hold the iothread mutex lock */ +static void kvm_inject_arm_sea(CPUState *c) +{ + ARMCPU *cpu = ARM_CPU(c); + CPUARMState *env = &cpu->env; + CPUClass *cc = CPU_GET_CLASS(c); + uint32_t esr; + bool same_el; + + c->exception_index = EXCP_DATA_ABORT; + env->exception.target_el = 1; + + /* + * Set the DFSC to synchronous external abort and set FnV to not valid, + * this will tell guest the FAR_ELx is UNKNOWN for this abort. + */ + same_el = arm_current_el(env) == env->exception.target_el; + esr = syn_data_abort_no_iss(same_el, 1, 0, 0, 0, 0, 0x10); + + env->exception.syndrome = esr; + + cc->do_interrupt(c); +} + #define AARCH64_CORE_REG(x) (KVM_REG_ARM64 | KVM_REG_SIZE_U64 | \ KVM_REG_ARM_CORE | KVM_REG_ARM_CORE_REG(x)) @@ -XXX,XX +XXX,XX @@ int kvm_arch_get_registers(CPUState *cs) return ret; } +void kvm_arch_on_sigbus_vcpu(CPUState *c, int code, void *addr) +{ + ram_addr_t ram_addr; + hwaddr paddr; + Object *obj = qdev_get_machine(); + VirtMachineState *vms = VIRT_MACHINE(obj); + bool acpi_enabled = virt_is_acpi_enabled(vms); + + assert(code == BUS_MCEERR_AR || code == BUS_MCEERR_AO); + + if (acpi_enabled && addr && + object_property_get_bool(obj, "ras", NULL)) { + ram_addr = qemu_ram_addr_from_host(addr); + if (ram_addr != RAM_ADDR_INVALID && + kvm_physical_memory_addr_from_host(c->kvm_state, addr, &paddr)) { + kvm_hwpoison_page_add(ram_addr); + /* + * If this is a BUS_MCEERR_AR, we know we have been called + * synchronously from the vCPU thread, so we can easily + * synchronize the state and inject an error. + * + * TODO: we currently don't tell the guest at all about + * BUS_MCEERR_AO. In that case we might either be being + * called synchronously from the vCPU thread, or a bit + * later from the main thread, so doing the injection of + * the error would be more complicated. + */ + if (code == BUS_MCEERR_AR) { + kvm_cpu_synchronize_state(c); + if (!acpi_ghes_record_errors(ACPI_HEST_SRC_ID_SEA, paddr)) { + kvm_inject_arm_sea(c); + } else { + error_report("failed to record the error"); + abort(); + } + } + return; + } + if (code == BUS_MCEERR_AO) { + error_report("Hardware memory error at addr %p for memory used by " + "QEMU itself instead of guest system!", addr); + } + } + + if (code == BUS_MCEERR_AR) { + error_report("Hardware memory error!"); + exit(1); + } +} + /* C6.6.29 BRK instruction */ static const uint32_t brk_insn = 0xd4200000; diff --git a/target/arm/tlb_helper.c b/target/arm/tlb_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tlb_helper.c +++ b/target/arm/tlb_helper.c @@ -XXX,XX +XXX,XX @@ static inline uint32_t merge_syn_data_abort(uint32_t template_syn, * ISV field. */ if (!(template_syn & ARM_EL_ISV) || target_el != 2 || s1ptw) { - syn = syn_data_abort_no_iss(same_el, + syn = syn_data_abort_no_iss(same_el, 0, ea, 0, s1ptw, is_write, fsc); } else { /* -- 2.20.1 From: Dongjiu Geng <gengdongjiu@huawei.com> I and Xiang are willing to review the APEI-related patches and volunteer as the reviewers for the HEST/GHES part. Signed-off-by: Dongjiu Geng <gengdongjiu@huawei.com> Signed-off-by: Xiang Zheng <zhengxiang9@huawei.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Message-id: 20200512030609.19593-11-gengdongjiu@huawei.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- MAINTAINERS | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index XXXXXXX..XXXXXXX 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -XXX,XX +XXX,XX @@ F: tests/qtest/bios-tables-test.c F: tests/qtest/acpi-utils.[hc] F: tests/data/acpi/ +ACPI/HEST/GHES +R: Dongjiu Geng <gengdongjiu@huawei.com> +R: Xiang Zheng <zhengxiang9@huawei.com> +L: qemu-arm@nongnu.org +S: Maintained +F: hw/acpi/ghes.c +F: include/hw/acpi/ghes.h +F: docs/specs/acpi_hest_ghes.rst + ppc4xx M: David Gibson <david@gibson.dropbear.id.au> L: qemu-ppc@nongnu.org -- 2.20.1 Convert the Neon VQRDMLAH and VQRDMLSH insns in the 3-reg-same group to decodetree. These don't use do_3same() because they want to operate on VFP double registers, whose offsets are different from the neon_reg_offset() calculations do_3same does. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-2-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 3 +++ target/arm/translate-neon.inc.c | 15 +++++++++++++++ target/arm/translate.c | 14 ++------------ 3 files changed, 20 insertions(+), 12 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VMLS_3s 1111 001 1 0 . .. .... .... 1001 . . . 0 .... @3same VMUL_3s 1111 001 0 0 . .. .... .... 1001 . . . 1 .... @3same VMUL_p_3s 1111 001 1 0 . .. .... .... 1001 . . . 1 .... @3same + +VQRDMLAH_3s 1111 001 1 0 . .. .... .... 1011 ... 1 .... @3same +VQRDMLSH_3s 1111 001 1 0 . .. .... .... 1100 ... 1 .... @3same diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static bool trans_VMUL_p_3s(DisasContext *s, arg_3same *a) } return do_3same(s, a, gen_VMUL_p_3s); } + +#define DO_VQRDMLAH(INSN, FUNC) \ + static bool trans_##INSN##_3s(DisasContext *s, arg_3same *a) \ + { \ + if (!dc_isar_feature(aa32_rdm, s)) { \ + return false; \ + } \ + if (a->size != 1 && a->size != 2) { \ + return false; \ + } \ + return do_3same(s, a, FUNC); \ + } + +DO_VQRDMLAH(VQRDMLAH, gen_gvec_sqrdmlah_qc) +DO_VQRDMLAH(VQRDMLSH, gen_gvec_sqrdmlsh_qc) 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_neon_data_insn(DisasContext *s, uint32_t insn) if (!u) { break; /* VPADD */ } - /* VQRDMLAH */ - if (dc_isar_feature(aa32_rdm, s) && (size == 1 || size == 2)) { - gen_gvec_sqrdmlah_qc(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - return 0; - } + /* VQRDMLAH : handled by decodetree */ return 1; case NEON_3R_VFM_VQRDMLSH: @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) } break; } - /* VQRDMLSH */ - if (dc_isar_feature(aa32_rdm, s) && (size == 1 || size == 2)) { - gen_gvec_sqrdmlsh_qc(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - return 0; - } + /* VQRDMLSH : handled by decodetree */ return 1; case NEON_3R_VABD: -- 2.20.1 Convert the Neon SHA instructions in the 3-reg-same group to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-3-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 10 +++ target/arm/translate-neon.inc.c | 139 ++++++++++++++++++++++++++++++++ target/arm/translate.c | 46 +---------- 3 files changed, 151 insertions(+), 44 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VMUL_3s 1111 001 0 0 . .. .... .... 1001 . . . 1 .... @3same VMUL_p_3s 1111 001 1 0 . .. .... .... 1001 . . . 1 .... @3same VQRDMLAH_3s 1111 001 1 0 . .. .... .... 1011 ... 1 .... @3same + +SHA1_3s 1111 001 0 0 . optype:2 .... .... 1100 . 1 . 0 .... \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp +SHA256H_3s 1111 001 1 0 . 00 .... .... 1100 . 1 . 0 .... \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp +SHA256H2_3s 1111 001 1 0 . 01 .... .... 1100 . 1 . 0 .... \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp +SHA256SU1_3s 1111 001 1 0 . 10 .... .... 1100 . 1 . 0 .... \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp + VQRDMLSH_3s 1111 001 1 0 . .. .... .... 1100 ... 1 .... @3same diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static bool trans_VMUL_p_3s(DisasContext *s, arg_3same *a) DO_VQRDMLAH(VQRDMLAH, gen_gvec_sqrdmlah_qc) DO_VQRDMLAH(VQRDMLSH, gen_gvec_sqrdmlsh_qc) + +static bool trans_SHA1_3s(DisasContext *s, arg_SHA1_3s *a) +{ + TCGv_ptr ptr1, ptr2, ptr3; + TCGv_i32 tmp; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON) || + !dc_isar_feature(aa32_sha1, s)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vn | a->vm) & 0x10)) { + return false; + } + + if ((a->vn | a->vm | a->vd) & 1) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + ptr1 = vfp_reg_ptr(true, a->vd); + ptr2 = vfp_reg_ptr(true, a->vn); + ptr3 = vfp_reg_ptr(true, a->vm); + tmp = tcg_const_i32(a->optype); + gen_helper_crypto_sha1_3reg(ptr1, ptr2, ptr3, tmp); + tcg_temp_free_i32(tmp); + tcg_temp_free_ptr(ptr1); + tcg_temp_free_ptr(ptr2); + tcg_temp_free_ptr(ptr3); + + return true; +} + +static bool trans_SHA256H_3s(DisasContext *s, arg_SHA256H_3s *a) +{ + TCGv_ptr ptr1, ptr2, ptr3; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON) || + !dc_isar_feature(aa32_sha2, s)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vn | a->vm) & 0x10)) { + return false; + } + + if ((a->vn | a->vm | a->vd) & 1) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + ptr1 = vfp_reg_ptr(true, a->vd); + ptr2 = vfp_reg_ptr(true, a->vn); + ptr3 = vfp_reg_ptr(true, a->vm); + gen_helper_crypto_sha256h(ptr1, ptr2, ptr3); + tcg_temp_free_ptr(ptr1); + tcg_temp_free_ptr(ptr2); + tcg_temp_free_ptr(ptr3); + + return true; +} + +static bool trans_SHA256H2_3s(DisasContext *s, arg_SHA256H2_3s *a) +{ + TCGv_ptr ptr1, ptr2, ptr3; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON) || + !dc_isar_feature(aa32_sha2, s)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vn | a->vm) & 0x10)) { + return false; + } + + if ((a->vn | a->vm | a->vd) & 1) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + ptr1 = vfp_reg_ptr(true, a->vd); + ptr2 = vfp_reg_ptr(true, a->vn); + ptr3 = vfp_reg_ptr(true, a->vm); + gen_helper_crypto_sha256h2(ptr1, ptr2, ptr3); + tcg_temp_free_ptr(ptr1); + tcg_temp_free_ptr(ptr2); + tcg_temp_free_ptr(ptr3); + + return true; +} + +static bool trans_SHA256SU1_3s(DisasContext *s, arg_SHA256SU1_3s *a) +{ + TCGv_ptr ptr1, ptr2, ptr3; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON) || + !dc_isar_feature(aa32_sha2, s)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vn | a->vm) & 0x10)) { + return false; + } + + if ((a->vn | a->vm | a->vd) & 1) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + ptr1 = vfp_reg_ptr(true, a->vd); + ptr2 = vfp_reg_ptr(true, a->vn); + ptr3 = vfp_reg_ptr(true, a->vm); + gen_helper_crypto_sha256su1(ptr1, ptr2, ptr3); + tcg_temp_free_ptr(ptr1); + tcg_temp_free_ptr(ptr2); + tcg_temp_free_ptr(ptr3); + + return true; +} 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_neon_data_insn(DisasContext *s, uint32_t insn) int vec_size; uint32_t imm; TCGv_i32 tmp, tmp2, tmp3, tmp4, tmp5; - TCGv_ptr ptr1, ptr2, ptr3; + TCGv_ptr ptr1, ptr2; TCGv_i64 tmp64; if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) return 1; } switch (op) { - case NEON_3R_SHA: - /* The SHA-1/SHA-256 3-register instructions require special - * treatment here, as their size field is overloaded as an - * op type selector, and they all consume their input in a - * single pass. - */ - if (!q) { - return 1; - } - if (!u) { /* SHA-1 */ - if (!dc_isar_feature(aa32_sha1, s)) { - return 1; - } - ptr1 = vfp_reg_ptr(true, rd); - ptr2 = vfp_reg_ptr(true, rn); - ptr3 = vfp_reg_ptr(true, rm); - tmp4 = tcg_const_i32(size); - gen_helper_crypto_sha1_3reg(ptr1, ptr2, ptr3, tmp4); - tcg_temp_free_i32(tmp4); - } else { /* SHA-256 */ - if (!dc_isar_feature(aa32_sha2, s) || size == 3) { - return 1; - } - ptr1 = vfp_reg_ptr(true, rd); - ptr2 = vfp_reg_ptr(true, rn); - ptr3 = vfp_reg_ptr(true, rm); - switch (size) { - case 0: - gen_helper_crypto_sha256h(ptr1, ptr2, ptr3); - break; - case 1: - gen_helper_crypto_sha256h2(ptr1, ptr2, ptr3); - break; - case 2: - gen_helper_crypto_sha256su1(ptr1, ptr2, ptr3); - break; - } - } - tcg_temp_free_ptr(ptr1); - tcg_temp_free_ptr(ptr2); - tcg_temp_free_ptr(ptr3); - return 0; - case NEON_3R_VPADD_VQRDMLAH: if (!u) { break; /* VPADD */ @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VMUL: case NEON_3R_VML: case NEON_3R_VSHL: + case NEON_3R_SHA: /* Already handled by decodetree */ return 1; } -- 2.20.1 Convert the 64-bit element insns in the 3-reg-same group to decodetree. This covers VQSHL, VRSHL and VQRSHL where size==0b11. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-4-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 13 +++++++++++ target/arm/translate-neon.inc.c | 24 +++++++++++++++++++++ target/arm/translate.c | 38 ++------------------------------- 3 files changed, 39 insertions(+), 36 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VCGE_U_3s 1111 001 1 0 . .. .... .... 0011 . . . 1 .... @3same VSHL_S_3s 1111 001 0 0 . .. .... .... 0100 . . . 0 .... @3same_rev VSHL_U_3s 1111 001 1 0 . .. .... .... 0100 . . . 0 .... @3same_rev +# Insns operating on 64-bit elements (size!=0b11 handled elsewhere) +# The _rev suffix indicates that Vn and Vm are reversed (as explained +# by the comment for the @3same_rev format). +@3same_64_rev .... ... . . . 11 .... .... .... . q:1 . . .... \ + &3same vm=%vn_dp vn=%vm_dp vd=%vd_dp size=3 + +VQSHL_S64_3s 1111 001 0 0 . .. .... .... 0100 . . . 1 .... @3same_64_rev +VQSHL_U64_3s 1111 001 1 0 . .. .... .... 0100 . . . 1 .... @3same_64_rev +VRSHL_S64_3s 1111 001 0 0 . .. .... .... 0101 . . . 0 .... @3same_64_rev +VRSHL_U64_3s 1111 001 1 0 . .. .... .... 0101 . . . 0 .... @3same_64_rev +VQRSHL_S64_3s 1111 001 0 0 . .. .... .... 0101 . . . 1 .... @3same_64_rev +VQRSHL_U64_3s 1111 001 1 0 . .. .... .... 0101 . . . 1 .... @3same_64_rev + VMAX_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 0 .... @3same VMAX_U_3s 1111 001 1 0 . .. .... .... 0110 . . . 0 .... @3same VMIN_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 1 .... @3same diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static bool trans_SHA256SU1_3s(DisasContext *s, arg_SHA256SU1_3s *a) return true; } + +#define DO_3SAME_64(INSN, FUNC) \ + static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ + uint32_t rn_ofs, uint32_t rm_ofs, \ + uint32_t oprsz, uint32_t maxsz) \ + { \ + static const GVecGen3 op = { .fni8 = FUNC }; \ + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, &op); \ + } \ + DO_3SAME(INSN, gen_##INSN##_3s) + +#define DO_3SAME_64_ENV(INSN, FUNC) \ + static void gen_##INSN##_elt(TCGv_i64 d, TCGv_i64 n, TCGv_i64 m) \ + { \ + FUNC(d, cpu_env, n, m); \ + } \ + DO_3SAME_64(INSN, gen_##INSN##_elt) + +DO_3SAME_64(VRSHL_S64, gen_helper_neon_rshl_s64) +DO_3SAME_64(VRSHL_U64, gen_helper_neon_rshl_u64) +DO_3SAME_64_ENV(VQSHL_S64, gen_helper_neon_qshl_s64) +DO_3SAME_64_ENV(VQSHL_U64, gen_helper_neon_qshl_u64) +DO_3SAME_64_ENV(VQRSHL_S64, gen_helper_neon_qrshl_s64) +DO_3SAME_64_ENV(VQRSHL_U64, gen_helper_neon_qrshl_u64) 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_neon_data_insn(DisasContext *s, uint32_t insn) } if (size == 3) { - /* 64-bit element instructions. */ - for (pass = 0; pass < (q ? 2 : 1); pass++) { - neon_load_reg64(cpu_V0, rn + pass); - neon_load_reg64(cpu_V1, rm + pass); - switch (op) { - case NEON_3R_VQSHL: - if (u) { - gen_helper_neon_qshl_u64(cpu_V0, cpu_env, - cpu_V1, cpu_V0); - } else { - gen_helper_neon_qshl_s64(cpu_V0, cpu_env, - cpu_V1, cpu_V0); - } - break; - case NEON_3R_VRSHL: - if (u) { - gen_helper_neon_rshl_u64(cpu_V0, cpu_V1, cpu_V0); - } else { - gen_helper_neon_rshl_s64(cpu_V0, cpu_V1, cpu_V0); - } - break; - case NEON_3R_VQRSHL: - if (u) { - gen_helper_neon_qrshl_u64(cpu_V0, cpu_env, - cpu_V1, cpu_V0); - } else { - gen_helper_neon_qrshl_s64(cpu_V0, cpu_env, - cpu_V1, cpu_V0); - } - break; - default: - abort(); - } - neon_store_reg64(cpu_V0, rd + pass); - } - return 0; + /* 64-bit element instructions: handled by decodetree */ + return 1; } pairwise = 0; switch (op) { -- 2.20.1 Convert the Neon VHADD insns in the 3-reg-same group to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-5-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 2 ++ target/arm/translate-neon.inc.c | 24 ++++++++++++++++++++++++ target/arm/translate.c | 4 +--- 3 files changed, 27 insertions(+), 3 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ @3same .... ... . . . size:2 .... .... .... . q:1 . . .... \ &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp +VHADD_S_3s 1111 001 0 0 . .. .... .... 0000 . . . 0 .... @3same +VHADD_U_3s 1111 001 1 0 . .. .... .... 0000 . . . 0 .... @3same VQADD_S_3s 1111 001 0 0 . .. .... .... 0000 . . . 1 .... @3same VQADD_U_3s 1111 001 1 0 . .. .... .... 0000 . . . 1 .... @3same diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME_64_ENV(VQSHL_S64, gen_helper_neon_qshl_s64) DO_3SAME_64_ENV(VQSHL_U64, gen_helper_neon_qshl_u64) DO_3SAME_64_ENV(VQRSHL_S64, gen_helper_neon_qrshl_s64) DO_3SAME_64_ENV(VQRSHL_U64, gen_helper_neon_qrshl_u64) + +#define DO_3SAME_32(INSN, FUNC) \ + static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ + uint32_t rn_ofs, uint32_t rm_ofs, \ + uint32_t oprsz, uint32_t maxsz) \ + { \ + static const GVecGen3 ops[4] = { \ + { .fni4 = gen_helper_neon_##FUNC##8 }, \ + { .fni4 = gen_helper_neon_##FUNC##16 }, \ + { .fni4 = gen_helper_neon_##FUNC##32 }, \ + { 0 }, \ + }; \ + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, &ops[vece]); \ + } \ + static bool trans_##INSN##_3s(DisasContext *s, arg_3same *a) \ + { \ + if (a->size > 2) { \ + return false; \ + } \ + return do_3same(s, a, gen_##INSN##_3s); \ + } + +DO_3SAME_32(VHADD_S, hadd_s) +DO_3SAME_32(VHADD_U, hadd_u) 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_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VML: case NEON_3R_VSHL: case NEON_3R_SHA: + case NEON_3R_VHADD: /* Already handled by decodetree */ return 1; } @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) tmp2 = neon_load_reg(rm, pass); } switch (op) { - case NEON_3R_VHADD: - GEN_NEON_INTEGER_OP(hadd); - break; case NEON_3R_VRHADD: GEN_NEON_INTEGER_OP(rhadd); break; -- 2.20.1 Convert the Neon VABA and VABD insns in the 3-reg-same group to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-6-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 6 ++++++ target/arm/translate-neon.inc.c | 4 ++++ target/arm/translate.c | 22 ++-------------------- 3 files changed, 12 insertions(+), 20 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VMAX_U_3s 1111 001 1 0 . .. .... .... 0110 . . . 0 .... @3same VMIN_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 1 .... @3same VMIN_U_3s 1111 001 1 0 . .. .... .... 0110 . . . 1 .... @3same +VABD_S_3s 1111 001 0 0 . .. .... .... 0111 . . . 0 .... @3same +VABD_U_3s 1111 001 1 0 . .. .... .... 0111 . . . 0 .... @3same + +VABA_S_3s 1111 001 0 0 . .. .... .... 0111 . . . 1 .... @3same +VABA_U_3s 1111 001 1 0 . .. .... .... 0111 . . . 1 .... @3same + VADD_3s 1111 001 0 0 . .. .... .... 1000 . . . 0 .... @3same VSUB_3s 1111 001 1 0 . .. .... .... 1000 . . . 0 .... @3same diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME_NO_SZ_3(VMUL, tcg_gen_gvec_mul) DO_3SAME_NO_SZ_3(VMLA, gen_gvec_mla) DO_3SAME_NO_SZ_3(VMLS, gen_gvec_mls) DO_3SAME_NO_SZ_3(VTST, gen_gvec_cmtst) +DO_3SAME_NO_SZ_3(VABD_S, gen_gvec_sabd) +DO_3SAME_NO_SZ_3(VABA_S, gen_gvec_saba) +DO_3SAME_NO_SZ_3(VABD_U, gen_gvec_uabd) +DO_3SAME_NO_SZ_3(VABA_U, gen_gvec_uaba) #define DO_3SAME_CMP(INSN, COND) \ static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ 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_neon_data_insn(DisasContext *s, uint32_t insn) /* VQRDMLSH : handled by decodetree */ return 1; - case NEON_3R_VABD: - if (u) { - gen_gvec_uabd(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - } else { - gen_gvec_sabd(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - } - return 0; - - case NEON_3R_VABA: - if (u) { - gen_gvec_uaba(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - } else { - gen_gvec_saba(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - } - return 0; - case NEON_3R_VADD_VSUB: case NEON_3R_LOGIC: case NEON_3R_VMAX: @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VSHL: case NEON_3R_SHA: case NEON_3R_VHADD: + case NEON_3R_VABD: + case NEON_3R_VABA: /* Already handled by decodetree */ return 1; } -- 2.20.1 Convert the Neon VRHADD and VHSUB 3-reg-same insns to decodetree. (These are all the other insns in 3-reg-same which were using GEN_NEON_INTEGER_OP() and which are not pairwise or reversed-operands.) Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-7-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 6 ++++++ target/arm/translate-neon.inc.c | 4 ++++ target/arm/translate.c | 8 ++------ 3 files changed, 12 insertions(+), 6 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VHADD_U_3s 1111 001 1 0 . .. .... .... 0000 . . . 0 .... @3same VQADD_S_3s 1111 001 0 0 . .. .... .... 0000 . . . 1 .... @3same VQADD_U_3s 1111 001 1 0 . .. .... .... 0000 . . . 1 .... @3same +VRHADD_S_3s 1111 001 0 0 . .. .... .... 0001 . . . 0 .... @3same +VRHADD_U_3s 1111 001 1 0 . .. .... .... 0001 . . . 0 .... @3same + @3same_logic .... ... . . . .. .... .... .... . q:1 .. .... \ &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp size=0 @@ -XXX,XX +XXX,XX @@ VBSL_3s 1111 001 1 0 . 01 .... .... 0001 ... 1 .... @3same_logic VBIT_3s 1111 001 1 0 . 10 .... .... 0001 ... 1 .... @3same_logic VBIF_3s 1111 001 1 0 . 11 .... .... 0001 ... 1 .... @3same_logic +VHSUB_S_3s 1111 001 0 0 . .. .... .... 0010 . . . 0 .... @3same +VHSUB_U_3s 1111 001 1 0 . .. .... .... 0010 . . . 0 .... @3same + VQSUB_S_3s 1111 001 0 0 . .. .... .... 0010 . . . 1 .... @3same VQSUB_U_3s 1111 001 1 0 . .. .... .... 0010 . . . 1 .... @3same diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME_64_ENV(VQRSHL_U64, gen_helper_neon_qrshl_u64) DO_3SAME_32(VHADD_S, hadd_s) DO_3SAME_32(VHADD_U, hadd_u) +DO_3SAME_32(VHSUB_S, hsub_s) +DO_3SAME_32(VHSUB_U, hsub_u) +DO_3SAME_32(VRHADD_S, rhadd_s) +DO_3SAME_32(VRHADD_U, rhadd_u) 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_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VSHL: case NEON_3R_SHA: case NEON_3R_VHADD: + case NEON_3R_VRHADD: + case NEON_3R_VHSUB: case NEON_3R_VABD: case NEON_3R_VABA: /* Already handled by decodetree */ @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) tmp2 = neon_load_reg(rm, pass); } switch (op) { - case NEON_3R_VRHADD: - GEN_NEON_INTEGER_OP(rhadd); - break; - case NEON_3R_VHSUB: - GEN_NEON_INTEGER_OP(hsub); - break; case NEON_3R_VQSHL: GEN_NEON_INTEGER_OP_ENV(qshl); break; -- 2.20.1 Convert the VQSHL, VRSHL and VQRSHL insns in the 3-reg-same group to decodetree. We have already implemented the size==0b11 case of these insns; this commit handles the remaining sizes. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-8-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 30 ++++++++++++++++++----- target/arm/translate-neon.inc.c | 43 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 22 +++-------------- 3 files changed, 70 insertions(+), 25 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VSHL_U_3s 1111 001 1 0 . .. .... .... 0100 . . . 0 .... @3same_rev @3same_64_rev .... ... . . . 11 .... .... .... . q:1 . . .... \ &3same vm=%vn_dp vn=%vm_dp vd=%vd_dp size=3 -VQSHL_S64_3s 1111 001 0 0 . .. .... .... 0100 . . . 1 .... @3same_64_rev -VQSHL_U64_3s 1111 001 1 0 . .. .... .... 0100 . . . 1 .... @3same_64_rev -VRSHL_S64_3s 1111 001 0 0 . .. .... .... 0101 . . . 0 .... @3same_64_rev -VRSHL_U64_3s 1111 001 1 0 . .. .... .... 0101 . . . 0 .... @3same_64_rev -VQRSHL_S64_3s 1111 001 0 0 . .. .... .... 0101 . . . 1 .... @3same_64_rev -VQRSHL_U64_3s 1111 001 1 0 . .. .... .... 0101 . . . 1 .... @3same_64_rev +{ + VQSHL_S64_3s 1111 001 0 0 . .. .... .... 0100 . . . 1 .... @3same_64_rev + VQSHL_S_3s 1111 001 0 0 . .. .... .... 0100 . . . 1 .... @3same_rev +} +{ + VQSHL_U64_3s 1111 001 1 0 . .. .... .... 0100 . . . 1 .... @3same_64_rev + VQSHL_U_3s 1111 001 1 0 . .. .... .... 0100 . . . 1 .... @3same_rev +} +{ + VRSHL_S64_3s 1111 001 0 0 . .. .... .... 0101 . . . 0 .... @3same_64_rev + VRSHL_S_3s 1111 001 0 0 . .. .... .... 0101 . . . 0 .... @3same_rev +} +{ + VRSHL_U64_3s 1111 001 1 0 . .. .... .... 0101 . . . 0 .... @3same_64_rev + VRSHL_U_3s 1111 001 1 0 . .. .... .... 0101 . . . 0 .... @3same_rev +} +{ + VQRSHL_S64_3s 1111 001 0 0 . .. .... .... 0101 . . . 1 .... @3same_64_rev + VQRSHL_S_3s 1111 001 0 0 . .. .... .... 0101 . . . 1 .... @3same_rev +} +{ + VQRSHL_U64_3s 1111 001 1 0 . .. .... .... 0101 . . . 1 .... @3same_64_rev + VQRSHL_U_3s 1111 001 1 0 . .. .... .... 0101 . . . 1 .... @3same_rev +} VMAX_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 0 .... @3same VMAX_U_3s 1111 001 1 0 . .. .... .... 0110 . . . 0 .... @3same diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME_64_ENV(VQRSHL_U64, gen_helper_neon_qrshl_u64) return do_3same(s, a, gen_##INSN##_3s); \ } +/* + * Some helper functions need to be passed the cpu_env. In order + * to use those with the gvec APIs like tcg_gen_gvec_3() we need + * to create wrapper functions whose prototype is a NeonGenTwoOpFn() + * and which call a NeonGenTwoOpEnvFn(). + */ +#define WRAP_ENV_FN(WRAPNAME, FUNC) \ + static void WRAPNAME(TCGv_i32 d, TCGv_i32 n, TCGv_i32 m) \ + { \ + FUNC(d, cpu_env, n, m); \ + } + +#define DO_3SAME_32_ENV(INSN, FUNC) \ + WRAP_ENV_FN(gen_##INSN##_tramp8, gen_helper_neon_##FUNC##8); \ + WRAP_ENV_FN(gen_##INSN##_tramp16, gen_helper_neon_##FUNC##16); \ + WRAP_ENV_FN(gen_##INSN##_tramp32, gen_helper_neon_##FUNC##32); \ + static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ + uint32_t rn_ofs, uint32_t rm_ofs, \ + uint32_t oprsz, uint32_t maxsz) \ + { \ + static const GVecGen3 ops[4] = { \ + { .fni4 = gen_##INSN##_tramp8 }, \ + { .fni4 = gen_##INSN##_tramp16 }, \ + { .fni4 = gen_##INSN##_tramp32 }, \ + { 0 }, \ + }; \ + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, &ops[vece]); \ + } \ + static bool trans_##INSN##_3s(DisasContext *s, arg_3same *a) \ + { \ + if (a->size > 2) { \ + return false; \ + } \ + return do_3same(s, a, gen_##INSN##_3s); \ + } + DO_3SAME_32(VHADD_S, hadd_s) DO_3SAME_32(VHADD_U, hadd_u) DO_3SAME_32(VHSUB_S, hsub_s) DO_3SAME_32(VHSUB_U, hsub_u) DO_3SAME_32(VRHADD_S, rhadd_s) DO_3SAME_32(VRHADD_U, rhadd_u) +DO_3SAME_32(VRSHL_S, rshl_s) +DO_3SAME_32(VRSHL_U, rshl_u) + +DO_3SAME_32_ENV(VQSHL_S, qshl_s) +DO_3SAME_32_ENV(VQSHL_U, qshl_u) +DO_3SAME_32_ENV(VQRSHL_S, qrshl_s) +DO_3SAME_32_ENV(VQRSHL_U, qrshl_u) 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_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VHSUB: case NEON_3R_VABD: case NEON_3R_VABA: + case NEON_3R_VQSHL: + case NEON_3R_VRSHL: + case NEON_3R_VQRSHL: /* Already handled by decodetree */ return 1; } @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) } pairwise = 0; switch (op) { - case NEON_3R_VQSHL: - case NEON_3R_VRSHL: - case NEON_3R_VQRSHL: - { - int rtmp; - /* Shift instruction operands are reversed. */ - rtmp = rn; - rn = rm; - rm = rtmp; - } - break; case NEON_3R_VPADD_VQRDMLAH: case NEON_3R_VPMAX: case NEON_3R_VPMIN: @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) tmp2 = neon_load_reg(rm, pass); } switch (op) { - case NEON_3R_VQSHL: - GEN_NEON_INTEGER_OP_ENV(qshl); - break; - case NEON_3R_VRSHL: - GEN_NEON_INTEGER_OP(rshl); - break; - case NEON_3R_VQRSHL: - GEN_NEON_INTEGER_OP_ENV(qrshl); break; case NEON_3R_VPMAX: GEN_NEON_INTEGER_OP(pmax); -- 2.20.1 Convert the Neon integer VPMAX and VPMIN 3-reg-same insns to decodetree. These are 'pairwise' operations. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-9-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 9 +++++ target/arm/translate-neon.inc.c | 71 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 17 +------- 3 files changed, 82 insertions(+), 15 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ @3same .... ... . . . size:2 .... .... .... . q:1 . . .... \ &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp +@3same_q0 .... ... . . . size:2 .... .... .... . 0 . . .... \ + &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp q=0 + VHADD_S_3s 1111 001 0 0 . .. .... .... 0000 . . . 0 .... @3same VHADD_U_3s 1111 001 1 0 . .. .... .... 0000 . . . 0 .... @3same VQADD_S_3s 1111 001 0 0 . .. .... .... 0000 . . . 1 .... @3same @@ -XXX,XX +XXX,XX @@ VMLS_3s 1111 001 1 0 . .. .... .... 1001 . . . 0 .... @3same VMUL_3s 1111 001 0 0 . .. .... .... 1001 . . . 1 .... @3same VMUL_p_3s 1111 001 1 0 . .. .... .... 1001 . . . 1 .... @3same +VPMAX_S_3s 1111 001 0 0 . .. .... .... 1010 . . . 0 .... @3same_q0 +VPMAX_U_3s 1111 001 1 0 . .. .... .... 1010 . . . 0 .... @3same_q0 + +VPMIN_S_3s 1111 001 0 0 . .. .... .... 1010 . . . 1 .... @3same_q0 +VPMIN_U_3s 1111 001 1 0 . .. .... .... 1010 . . . 1 .... @3same_q0 + VQRDMLAH_3s 1111 001 1 0 . .. .... .... 1011 ... 1 .... @3same SHA1_3s 1111 001 0 0 . optype:2 .... .... 1100 . 1 . 0 .... \ diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME_32_ENV(VQSHL_S, qshl_s) DO_3SAME_32_ENV(VQSHL_U, qshl_u) DO_3SAME_32_ENV(VQRSHL_S, qrshl_s) DO_3SAME_32_ENV(VQRSHL_U, qrshl_u) + +static bool do_3same_pair(DisasContext *s, arg_3same *a, NeonGenTwoOpFn *fn) +{ + /* Operations handled pairwise 32 bits at a time */ + TCGv_i32 tmp, tmp2, tmp3; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vn | a->vm) & 0x10)) { + return false; + } + + if (a->size == 3) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + assert(a->q == 0); /* enforced by decode patterns */ + + /* + * Note that we have to be careful not to clobber the source operands + * in the "vm == vd" case by storing the result of the first pass too + * early. Since Q is 0 there are always just two passes, so instead + * of a complicated loop over each pass we just unroll. + */ + tmp = neon_load_reg(a->vn, 0); + tmp2 = neon_load_reg(a->vn, 1); + fn(tmp, tmp, tmp2); + tcg_temp_free_i32(tmp2); + + tmp3 = neon_load_reg(a->vm, 0); + tmp2 = neon_load_reg(a->vm, 1); + fn(tmp3, tmp3, tmp2); + tcg_temp_free_i32(tmp2); + + neon_store_reg(a->vd, 0, tmp); + neon_store_reg(a->vd, 1, tmp3); + return true; +} + +#define DO_3SAME_PAIR(INSN, func) \ + static bool trans_##INSN##_3s(DisasContext *s, arg_3same *a) \ + { \ + static NeonGenTwoOpFn * const fns[] = { \ + gen_helper_neon_##func##8, \ + gen_helper_neon_##func##16, \ + gen_helper_neon_##func##32, \ + }; \ + if (a->size > 2) { \ + return false; \ + } \ + return do_3same_pair(s, a, fns[a->size]); \ + } + +/* 32-bit pairwise ops end up the same as the elementwise versions. */ +#define gen_helper_neon_pmax_s32 tcg_gen_smax_i32 +#define gen_helper_neon_pmax_u32 tcg_gen_umax_i32 +#define gen_helper_neon_pmin_s32 tcg_gen_smin_i32 +#define gen_helper_neon_pmin_u32 tcg_gen_umin_i32 + +DO_3SAME_PAIR(VPMAX_S, pmax_s) +DO_3SAME_PAIR(VPMIN_S, pmin_s) +DO_3SAME_PAIR(VPMAX_U, pmax_u) +DO_3SAME_PAIR(VPMIN_U, pmin_u) 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 inline void gen_neon_rsb(int size, TCGv_i32 t0, TCGv_i32 t1) } } -/* 32-bit pairwise ops end up the same as the elementwise versions. */ -#define gen_helper_neon_pmax_s32 tcg_gen_smax_i32 -#define gen_helper_neon_pmax_u32 tcg_gen_umax_i32 -#define gen_helper_neon_pmin_s32 tcg_gen_smin_i32 -#define gen_helper_neon_pmin_u32 tcg_gen_umin_i32 - #define GEN_NEON_INTEGER_OP_ENV(name) do { \ switch ((size << 1) | u) { \ case 0: \ @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VQSHL: case NEON_3R_VRSHL: case NEON_3R_VQRSHL: + case NEON_3R_VPMAX: + case NEON_3R_VPMIN: /* Already handled by decodetree */ return 1; } @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) pairwise = 0; switch (op) { case NEON_3R_VPADD_VQRDMLAH: - case NEON_3R_VPMAX: - case NEON_3R_VPMIN: pairwise = 1; break; case NEON_3R_FLOAT_ARITH: @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) tmp2 = neon_load_reg(rm, pass); } switch (op) { - break; - case NEON_3R_VPMAX: - GEN_NEON_INTEGER_OP(pmax); - break; - case NEON_3R_VPMIN: - GEN_NEON_INTEGER_OP(pmin); - break; case NEON_3R_VQDMULH_VQRDMULH: /* Multiply high. */ if (!u) { /* VQDMULH */ switch (size) { -- 2.20.1 Convert the Neon integer VPADD 3-reg-same insns to decodetree. These are 'pairwise' operations. (Note that VQRDMLAH, which shares the same primary opcode but has U=1, has already been converted.) Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-10-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 2 ++ target/arm/translate-neon.inc.c | 2 ++ target/arm/translate.c | 19 +------------------ 3 files changed, 5 insertions(+), 18 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VPMAX_U_3s 1111 001 1 0 . .. .... .... 1010 . . . 0 .... @3same_q0 VPMIN_S_3s 1111 001 0 0 . .. .... .... 1010 . . . 1 .... @3same_q0 VPMIN_U_3s 1111 001 1 0 . .. .... .... 1010 . . . 1 .... @3same_q0 +VPADD_3s 1111 001 0 0 . .. .... .... 1011 . . . 1 .... @3same_q0 + VQRDMLAH_3s 1111 001 1 0 . .. .... .... 1011 ... 1 .... @3same SHA1_3s 1111 001 0 0 . optype:2 .... .... 1100 . 1 . 0 .... \ diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static bool do_3same_pair(DisasContext *s, arg_3same *a, NeonGenTwoOpFn *fn) #define gen_helper_neon_pmax_u32 tcg_gen_umax_i32 #define gen_helper_neon_pmin_s32 tcg_gen_smin_i32 #define gen_helper_neon_pmin_u32 tcg_gen_umin_i32 +#define gen_helper_neon_padd_u32 tcg_gen_add_i32 DO_3SAME_PAIR(VPMAX_S, pmax_s) DO_3SAME_PAIR(VPMIN_S, pmin_s) DO_3SAME_PAIR(VPMAX_U, pmax_u) DO_3SAME_PAIR(VPMIN_U, pmin_u) +DO_3SAME_PAIR(VPADD, padd_u) 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_neon_data_insn(DisasContext *s, uint32_t insn) return 1; } switch (op) { - case NEON_3R_VPADD_VQRDMLAH: - if (!u) { - break; /* VPADD */ - } - /* VQRDMLAH : handled by decodetree */ - return 1; - case NEON_3R_VFM_VQRDMLSH: if (!u) { /* VFM, VFMS */ @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VQRSHL: case NEON_3R_VPMAX: case NEON_3R_VPMIN: + case NEON_3R_VPADD_VQRDMLAH: /* Already handled by decodetree */ return 1; } @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) } pairwise = 0; switch (op) { - case NEON_3R_VPADD_VQRDMLAH: - pairwise = 1; - break; case NEON_3R_FLOAT_ARITH: pairwise = (u && size < 2); /* if VPADD (float) */ break; @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) } } break; - case NEON_3R_VPADD_VQRDMLAH: - switch (size) { - case 0: gen_helper_neon_padd_u8(tmp, tmp, tmp2); break; - case 1: gen_helper_neon_padd_u16(tmp, tmp, tmp2); break; - case 2: tcg_gen_add_i32(tmp, tmp, tmp2); break; - default: abort(); - } - break; case NEON_3R_FLOAT_ARITH: /* Floating point arithmetic. */ { TCGv_ptr fpstatus = get_fpstatus_ptr(1); -- 2.20.1 Convert the Neon VQDMULH and VQRDMULH 3-reg-same insns to decodetree. These are the last integer operations in the 3-reg-same group. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-11-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 3 +++ target/arm/translate-neon.inc.c | 24 ++++++++++++++++++++++++ target/arm/translate.c | 24 +----------------------- 3 files changed, 28 insertions(+), 23 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VPMAX_U_3s 1111 001 1 0 . .. .... .... 1010 . . . 0 .... @3same_q0 VPMIN_S_3s 1111 001 0 0 . .. .... .... 1010 . . . 1 .... @3same_q0 VPMIN_U_3s 1111 001 1 0 . .. .... .... 1010 . . . 1 .... @3same_q0 +VQDMULH_3s 1111 001 0 0 . .. .... .... 1011 . . . 0 .... @3same +VQRDMULH_3s 1111 001 1 0 . .. .... .... 1011 . . . 0 .... @3same + VPADD_3s 1111 001 0 0 . .. .... .... 1011 . . . 1 .... @3same_q0 VQRDMLAH_3s 1111 001 1 0 . .. .... .... 1011 ... 1 .... @3same diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME_PAIR(VPMIN_S, pmin_s) DO_3SAME_PAIR(VPMAX_U, pmax_u) DO_3SAME_PAIR(VPMIN_U, pmin_u) DO_3SAME_PAIR(VPADD, padd_u) + +#define DO_3SAME_VQDMULH(INSN, FUNC) \ + WRAP_ENV_FN(gen_##INSN##_tramp16, gen_helper_neon_##FUNC##_s16); \ + WRAP_ENV_FN(gen_##INSN##_tramp32, gen_helper_neon_##FUNC##_s32); \ + static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ + uint32_t rn_ofs, uint32_t rm_ofs, \ + uint32_t oprsz, uint32_t maxsz) \ + { \ + static const GVecGen3 ops[2] = { \ + { .fni4 = gen_##INSN##_tramp16 }, \ + { .fni4 = gen_##INSN##_tramp32 }, \ + }; \ + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, &ops[vece - 1]); \ + } \ + static bool trans_##INSN##_3s(DisasContext *s, arg_3same *a) \ + { \ + if (a->size != 1 && a->size != 2) { \ + return false; \ + } \ + return do_3same(s, a, gen_##INSN##_3s); \ + } + +DO_3SAME_VQDMULH(VQDMULH, qdmulh) +DO_3SAME_VQDMULH(VQRDMULH, qrdmulh) 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_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VPMAX: case NEON_3R_VPMIN: case NEON_3R_VPADD_VQRDMLAH: + case NEON_3R_VQDMULH_VQRDMULH: /* Already handled by decodetree */ return 1; } @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) tmp2 = neon_load_reg(rm, pass); } switch (op) { - case NEON_3R_VQDMULH_VQRDMULH: /* Multiply high. */ - if (!u) { /* VQDMULH */ - switch (size) { - case 1: - gen_helper_neon_qdmulh_s16(tmp, cpu_env, tmp, tmp2); - break; - case 2: - gen_helper_neon_qdmulh_s32(tmp, cpu_env, tmp, tmp2); - break; - default: abort(); - } - } else { /* VQRDMULH */ - switch (size) { - case 1: - gen_helper_neon_qrdmulh_s16(tmp, cpu_env, tmp, tmp2); - break; - case 2: - gen_helper_neon_qrdmulh_s32(tmp, cpu_env, tmp, tmp2); - break; - default: abort(); - } - } - break; case NEON_3R_FLOAT_ARITH: /* Floating point arithmetic. */ { TCGv_ptr fpstatus = get_fpstatus_ptr(1); -- 2.20.1 Convert the Neon VADD, VSUB, VABD 3-reg-same insns to decodetree. We already have gvec helpers for addition and subtraction, but must add one for fabd. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-12-peter.maydell@linaro.org --- target/arm/helper.h | 3 ++- target/arm/neon-dp.decode | 8 ++++++++ target/arm/neon_helper.c | 7 ------- target/arm/translate-neon.inc.c | 28 ++++++++++++++++++++++++++++ target/arm/translate.c | 10 +++------- target/arm/vec_helper.c | 7 +++++++ 6 files changed, 48 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_FLAGS_2(neon_qneg_s16, TCG_CALL_NO_RWG, i32, env, i32) DEF_HELPER_FLAGS_2(neon_qneg_s32, TCG_CALL_NO_RWG, i32, env, i32) DEF_HELPER_FLAGS_2(neon_qneg_s64, TCG_CALL_NO_RWG, i64, env, i64) -DEF_HELPER_3(neon_abd_f32, i32, i32, i32, ptr) DEF_HELPER_3(neon_ceq_f32, i32, i32, i32, ptr) DEF_HELPER_3(neon_cge_f32, i32, i32, i32, ptr) DEF_HELPER_3(neon_cgt_f32, i32, i32, i32, ptr) @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_5(gvec_fmul_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_5(gvec_fmul_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_5(gvec_fmul_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_5(gvec_fabd_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) + DEF_HELPER_FLAGS_5(gvec_ftsmul_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_5(gvec_ftsmul_s, TCG_CALL_NO_RWG, diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ @3same_q0 .... ... . . . size:2 .... .... .... . 0 . . .... \ &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp q=0 +# For FP insns the high bit of 'size' is used as part of opcode decode +@3same_fp .... ... . . . . size:1 .... .... .... . q:1 . . .... \ + &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp + VHADD_S_3s 1111 001 0 0 . .. .... .... 0000 . . . 0 .... @3same VHADD_U_3s 1111 001 1 0 . .. .... .... 0000 . . . 0 .... @3same VQADD_S_3s 1111 001 0 0 . .. .... .... 0000 . . . 1 .... @3same @@ -XXX,XX +XXX,XX @@ SHA256SU1_3s 1111 001 1 0 . 10 .... .... 1100 . 1 . 0 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp VQRDMLSH_3s 1111 001 1 0 . .. .... .... 1100 ... 1 .... @3same + +VADD_fp_3s 1111 001 0 0 . 0 . .... .... 1101 ... 0 .... @3same_fp +VSUB_fp_3s 1111 001 0 0 . 1 . .... .... 1101 ... 0 .... @3same_fp +VABD_fp_3s 1111 001 1 0 . 1 . .... .... 1101 ... 0 .... @3same_fp diff --git a/target/arm/neon_helper.c b/target/arm/neon_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon_helper.c +++ b/target/arm/neon_helper.c @@ -XXX,XX +XXX,XX @@ uint64_t HELPER(neon_qneg_s64)(CPUARMState *env, uint64_t x) } /* NEON Float helpers. */ -uint32_t HELPER(neon_abd_f32)(uint32_t a, uint32_t b, void *fpstp) -{ - float_status *fpst = fpstp; - float32 f0 = make_float32(a); - float32 f1 = make_float32(b); - return float32_val(float32_abs(float32_sub(f0, f1, fpst))); -} /* Floating point comparisons produce an integer result. * Note that EQ doesn't signal InvalidOp for QNaNs but GE and GT do. diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME_PAIR(VPADD, padd_u) DO_3SAME_VQDMULH(VQDMULH, qdmulh) DO_3SAME_VQDMULH(VQRDMULH, qrdmulh) + +/* + * For all the functions using this macro, size == 1 means fp16, + * which is an architecture extension we don't implement yet. + */ +#define DO_3S_FP_GVEC(INSN,FUNC) \ + static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ + uint32_t rn_ofs, uint32_t rm_ofs, \ + uint32_t oprsz, uint32_t maxsz) \ + { \ + TCGv_ptr fpst = get_fpstatus_ptr(1); \ + tcg_gen_gvec_3_ptr(rd_ofs, rn_ofs, rm_ofs, fpst, \ + oprsz, maxsz, 0, FUNC); \ + tcg_temp_free_ptr(fpst); \ + } \ + static bool trans_##INSN##_fp_3s(DisasContext *s, arg_3same *a) \ + { \ + if (a->size != 0) { \ + /* TODO fp16 support */ \ + return false; \ + } \ + return do_3same(s, a, gen_##INSN##_3s); \ + } + + +DO_3S_FP_GVEC(VADD, gen_helper_gvec_fadd_s) +DO_3S_FP_GVEC(VSUB, gen_helper_gvec_fsub_s) +DO_3S_FP_GVEC(VABD, gen_helper_gvec_fabd_s) 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_neon_data_insn(DisasContext *s, uint32_t insn) switch (op) { case NEON_3R_FLOAT_ARITH: pairwise = (u && size < 2); /* if VPADD (float) */ + if (!pairwise) { + return 1; /* handled by decodetree */ + } break; case NEON_3R_FLOAT_MINMAX: pairwise = u; /* if VPMIN/VPMAX (float) */ @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) { TCGv_ptr fpstatus = get_fpstatus_ptr(1); switch ((u << 2) | size) { - case 0: /* VADD */ case 4: /* VPADD */ gen_helper_vfp_adds(tmp, tmp, tmp2, fpstatus); break; - case 2: /* VSUB */ - gen_helper_vfp_subs(tmp, tmp, tmp2, fpstatus); - break; - case 6: /* VABD */ - gen_helper_neon_abd_f32(tmp, tmp, tmp2, fpstatus); - break; default: abort(); } diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ static float64 float64_ftsmul(float64 op1, uint64_t op2, float_status *stat) return result; } +static float32 float32_abd(float32 op1, float32 op2, float_status *stat) +{ + return float32_abs(float32_sub(op1, op2, stat)); +} + #define DO_3OP(NAME, FUNC, TYPE) \ void HELPER(NAME)(void *vd, void *vn, void *vm, void *stat, uint32_t desc) \ { \ @@ -XXX,XX +XXX,XX @@ DO_3OP(gvec_ftsmul_h, float16_ftsmul, float16) DO_3OP(gvec_ftsmul_s, float32_ftsmul, float32) DO_3OP(gvec_ftsmul_d, float64_ftsmul, float64) +DO_3OP(gvec_fabd_s, float32_abd, float32) + #ifdef TARGET_AARCH64 DO_3OP(gvec_recps_h, helper_recpsf_f16, float16) -- 2.20.1 Convert the Neon float VPMIN, VPMAX and VPADD 3-reg-same insns to decodetree. These are the only remaining 'pairwise' operations, so we can delete the pairwise-specific bits of the old decoder's for-each-element loop now. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-13-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 5 +++ target/arm/translate-neon.inc.c | 63 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 63 +++++---------------------------- 3 files changed, 76 insertions(+), 55 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ # For FP insns the high bit of 'size' is used as part of opcode decode @3same_fp .... ... . . . . size:1 .... .... .... . q:1 . . .... \ &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp +@3same_fp_q0 .... ... . . . . size:1 .... .... .... . 0 . . .... \ + &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp q=0 VHADD_S_3s 1111 001 0 0 . .. .... .... 0000 . . . 0 .... @3same VHADD_U_3s 1111 001 1 0 . .. .... .... 0000 . . . 0 .... @3same @@ -XXX,XX +XXX,XX @@ VQRDMLSH_3s 1111 001 1 0 . .. .... .... 1100 ... 1 .... @3same VADD_fp_3s 1111 001 0 0 . 0 . .... .... 1101 ... 0 .... @3same_fp VSUB_fp_3s 1111 001 0 0 . 1 . .... .... 1101 ... 0 .... @3same_fp +VPADD_fp_3s 1111 001 1 0 . 0 . .... .... 1101 ... 0 .... @3same_fp_q0 VABD_fp_3s 1111 001 1 0 . 1 . .... .... 1101 ... 0 .... @3same_fp +VPMAX_fp_3s 1111 001 1 0 . 0 . .... .... 1111 ... 0 .... @3same_fp_q0 +VPMIN_fp_3s 1111 001 1 0 . 1 . .... .... 1111 ... 0 .... @3same_fp_q0 diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME_VQDMULH(VQRDMULH, qrdmulh) DO_3S_FP_GVEC(VADD, gen_helper_gvec_fadd_s) DO_3S_FP_GVEC(VSUB, gen_helper_gvec_fsub_s) DO_3S_FP_GVEC(VABD, gen_helper_gvec_fabd_s) + +static bool do_3same_fp_pair(DisasContext *s, arg_3same *a, VFPGen3OpSPFn *fn) +{ + /* FP operations handled pairwise 32 bits at a time */ + TCGv_i32 tmp, tmp2, tmp3; + TCGv_ptr fpstatus; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vn | a->vm) & 0x10)) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + assert(a->q == 0); /* enforced by decode patterns */ + + /* + * Note that we have to be careful not to clobber the source operands + * in the "vm == vd" case by storing the result of the first pass too + * early. Since Q is 0 there are always just two passes, so instead + * of a complicated loop over each pass we just unroll. + */ + fpstatus = get_fpstatus_ptr(1); + tmp = neon_load_reg(a->vn, 0); + tmp2 = neon_load_reg(a->vn, 1); + fn(tmp, tmp, tmp2, fpstatus); + tcg_temp_free_i32(tmp2); + + tmp3 = neon_load_reg(a->vm, 0); + tmp2 = neon_load_reg(a->vm, 1); + fn(tmp3, tmp3, tmp2, fpstatus); + tcg_temp_free_i32(tmp2); + tcg_temp_free_ptr(fpstatus); + + neon_store_reg(a->vd, 0, tmp); + neon_store_reg(a->vd, 1, tmp3); + return true; +} + +/* + * For all the functions using this macro, size == 1 means fp16, + * which is an architecture extension we don't implement yet. + */ +#define DO_3S_FP_PAIR(INSN,FUNC) \ + static bool trans_##INSN##_fp_3s(DisasContext *s, arg_3same *a) \ + { \ + if (a->size != 0) { \ + /* TODO fp16 support */ \ + return false; \ + } \ + return do_3same_fp_pair(s, a, FUNC); \ + } + +DO_3S_FP_PAIR(VPADD, gen_helper_vfp_adds) +DO_3S_FP_PAIR(VPMAX, gen_helper_vfp_maxs) +DO_3S_FP_PAIR(VPMIN, gen_helper_vfp_mins) 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_neon_data_insn(DisasContext *s, uint32_t insn) int shift; int pass; int count; - int pairwise; int u; int vec_size; uint32_t imm; @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VPMIN: case NEON_3R_VPADD_VQRDMLAH: case NEON_3R_VQDMULH_VQRDMULH: + case NEON_3R_FLOAT_ARITH: /* Already handled by decodetree */ return 1; } @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) /* 64-bit element instructions: handled by decodetree */ return 1; } - pairwise = 0; switch (op) { - case NEON_3R_FLOAT_ARITH: - pairwise = (u && size < 2); /* if VPADD (float) */ - if (!pairwise) { - return 1; /* handled by decodetree */ - } - break; case NEON_3R_FLOAT_MINMAX: - pairwise = u; /* if VPMIN/VPMAX (float) */ + if (u) { + return 1; /* VPMIN/VPMAX handled by decodetree */ + } break; case NEON_3R_FLOAT_CMP: if (!u && size) { @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) break; } - if (pairwise && q) { - /* All the pairwise insns UNDEF if Q is set */ - return 1; - } - for (pass = 0; pass < (q ? 4 : 2); pass++) { - if (pairwise) { - /* Pairwise. */ - if (pass < 1) { - tmp = neon_load_reg(rn, 0); - tmp2 = neon_load_reg(rn, 1); - } else { - tmp = neon_load_reg(rm, 0); - tmp2 = neon_load_reg(rm, 1); - } - } else { - /* Elementwise. */ - tmp = neon_load_reg(rn, pass); - tmp2 = neon_load_reg(rm, pass); - } + /* Elementwise. */ + tmp = neon_load_reg(rn, pass); + tmp2 = neon_load_reg(rm, pass); switch (op) { - case NEON_3R_FLOAT_ARITH: /* Floating point arithmetic. */ - { - TCGv_ptr fpstatus = get_fpstatus_ptr(1); - switch ((u << 2) | size) { - case 4: /* VPADD */ - gen_helper_vfp_adds(tmp, tmp, tmp2, fpstatus); - break; - default: - abort(); - } - tcg_temp_free_ptr(fpstatus); - break; - } case NEON_3R_FLOAT_MULTIPLY: { TCGv_ptr fpstatus = get_fpstatus_ptr(1); @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) } tcg_temp_free_i32(tmp2); - /* Save the result. For elementwise operations we can put it - straight into the destination register. For pairwise operations - we have to be careful to avoid clobbering the source operands. */ - if (pairwise && rd == rm) { - neon_store_scratch(pass, tmp); - } else { - neon_store_reg(rd, pass, tmp); - } + neon_store_reg(rd, pass, tmp); } /* for pass */ - if (pairwise && rd == rm) { - for (pass = 0; pass < (q ? 4 : 2); pass++) { - tmp = neon_load_scratch(pass); - neon_store_reg(rd, pass, tmp); - } - } /* End of 3 register same size operations. */ } else if (insn & (1 << 4)) { if ((insn & 0x00380080) != 0) { -- 2.20.1 Convert the Neon integer VMUL, VMLA, and VMLS 3-reg-same inssn to decodetree. We don't have a gvec helper for multiply-accumulate, so VMLA and VMLS need a loop function do_3same_fp(). This takes a reads_vd parameter to do_3same_fp() which tells it to load the old value into vd before calling the callback function, in the same way that the do_vfp_3op_sp() and do_vfp_3op_dp() functions in translate-vfp.inc.c work. (The only uses in this patch pass reads_vd == true, but later commits will use reads_vd == false.) This conversion fixes in passing an underdecoding for VMUL (originally reported by Fredrik Strupe <fredrik@strupe.net>): bit 1 of the 'size' field must be 0. The old decoder didn't enforce this, but the decodetree pattern does. The gen_VMLA_fp_reg() function performs the addition operation with the operands in the opposite order to the old decoder: since Neon sets 'default NaN mode' float32_add operations are commutative so there is no behaviour difference, but putting them this way around matches the Arm ARM pseudocode and the required operation order for the subtraction in gen_VMLS_fp_reg(). Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-14-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 3 ++ target/arm/translate-neon.inc.c | 81 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 17 +------ 3 files changed, 85 insertions(+), 16 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VADD_fp_3s 1111 001 0 0 . 0 . .... .... 1101 ... 0 .... @3same_fp VSUB_fp_3s 1111 001 0 0 . 1 . .... .... 1101 ... 0 .... @3same_fp VPADD_fp_3s 1111 001 1 0 . 0 . .... .... 1101 ... 0 .... @3same_fp_q0 VABD_fp_3s 1111 001 1 0 . 1 . .... .... 1101 ... 0 .... @3same_fp +VMLA_fp_3s 1111 001 0 0 . 0 . .... .... 1101 ... 1 .... @3same_fp +VMLS_fp_3s 1111 001 0 0 . 1 . .... .... 1101 ... 1 .... @3same_fp +VMUL_fp_3s 1111 001 1 0 . 0 . .... .... 1101 ... 1 .... @3same_fp VPMAX_fp_3s 1111 001 1 0 . 0 . .... .... 1111 ... 0 .... @3same_fp_q0 VPMIN_fp_3s 1111 001 1 0 . 1 . .... .... 1111 ... 0 .... @3same_fp_q0 diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME_PAIR(VPADD, padd_u) DO_3SAME_VQDMULH(VQDMULH, qdmulh) DO_3SAME_VQDMULH(VQRDMULH, qrdmulh) +static bool do_3same_fp(DisasContext *s, arg_3same *a, VFPGen3OpSPFn *fn, + bool reads_vd) +{ + /* + * FP operations handled elementwise 32 bits at a time. + * If reads_vd is true then the old value of Vd will be + * loaded before calling the callback function. This is + * used for multiply-accumulate type operations. + */ + TCGv_i32 tmp, tmp2; + int pass; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vn | a->vm) & 0x10)) { + return false; + } + + if ((a->vn | a->vm | a->vd) & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + TCGv_ptr fpstatus = get_fpstatus_ptr(1); + for (pass = 0; pass < (a->q ? 4 : 2); pass++) { + tmp = neon_load_reg(a->vn, pass); + tmp2 = neon_load_reg(a->vm, pass); + if (reads_vd) { + TCGv_i32 tmp_rd = neon_load_reg(a->vd, pass); + fn(tmp_rd, tmp, tmp2, fpstatus); + neon_store_reg(a->vd, pass, tmp_rd); + tcg_temp_free_i32(tmp); + } else { + fn(tmp, tmp, tmp2, fpstatus); + neon_store_reg(a->vd, pass, tmp); + } + tcg_temp_free_i32(tmp2); + } + tcg_temp_free_ptr(fpstatus); + return true; +} + /* * For all the functions using this macro, size == 1 means fp16, * which is an architecture extension we don't implement yet. @@ -XXX,XX +XXX,XX @@ DO_3SAME_VQDMULH(VQRDMULH, qrdmulh) DO_3S_FP_GVEC(VADD, gen_helper_gvec_fadd_s) DO_3S_FP_GVEC(VSUB, gen_helper_gvec_fsub_s) DO_3S_FP_GVEC(VABD, gen_helper_gvec_fabd_s) +DO_3S_FP_GVEC(VMUL, gen_helper_gvec_fmul_s) + +/* + * For all the functions using this macro, size == 1 means fp16, + * which is an architecture extension we don't implement yet. + */ +#define DO_3S_FP(INSN,FUNC,READS_VD) \ + static bool trans_##INSN##_fp_3s(DisasContext *s, arg_3same *a) \ + { \ + if (a->size != 0) { \ + /* TODO fp16 support */ \ + return false; \ + } \ + return do_3same_fp(s, a, FUNC, READS_VD); \ + } + +static void gen_VMLA_fp_3s(TCGv_i32 vd, TCGv_i32 vn, TCGv_i32 vm, + TCGv_ptr fpstatus) +{ + gen_helper_vfp_muls(vn, vn, vm, fpstatus); + gen_helper_vfp_adds(vd, vd, vn, fpstatus); +} + +static void gen_VMLS_fp_3s(TCGv_i32 vd, TCGv_i32 vn, TCGv_i32 vm, + TCGv_ptr fpstatus) +{ + gen_helper_vfp_muls(vn, vn, vm, fpstatus); + gen_helper_vfp_subs(vd, vd, vn, fpstatus); +} + +DO_3S_FP(VMLA, gen_VMLA_fp_3s, true) +DO_3S_FP(VMLS, gen_VMLS_fp_3s, true) static bool do_3same_fp_pair(DisasContext *s, arg_3same *a, VFPGen3OpSPFn *fn) { 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_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VPADD_VQRDMLAH: case NEON_3R_VQDMULH_VQRDMULH: case NEON_3R_FLOAT_ARITH: + case NEON_3R_FLOAT_MULTIPLY: /* Already handled by decodetree */ return 1; } @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) tmp = neon_load_reg(rn, pass); tmp2 = neon_load_reg(rm, pass); switch (op) { - case NEON_3R_FLOAT_MULTIPLY: - { - TCGv_ptr fpstatus = get_fpstatus_ptr(1); - gen_helper_vfp_muls(tmp, tmp, tmp2, fpstatus); - if (!u) { - tcg_temp_free_i32(tmp2); - tmp2 = neon_load_reg(rd, pass); - if (size == 0) { - gen_helper_vfp_adds(tmp, tmp, tmp2, fpstatus); - } else { - gen_helper_vfp_subs(tmp, tmp2, tmp, fpstatus); - } - } - tcg_temp_free_ptr(fpstatus); - break; - } case NEON_3R_FLOAT_CMP: { TCGv_ptr fpstatus = get_fpstatus_ptr(1); -- 2.20.1 Convert the Neon integer 3-reg-same compare insns VCGE, VCGT, VCEQ, VACGE and VACGT to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-15-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 5 +++++ target/arm/translate-neon.inc.c | 6 +++++ target/arm/translate.c | 39 ++------------------------------- 3 files changed, 13 insertions(+), 37 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VABD_fp_3s 1111 001 1 0 . 1 . .... .... 1101 ... 0 .... @3same_fp VMLA_fp_3s 1111 001 0 0 . 0 . .... .... 1101 ... 1 .... @3same_fp VMLS_fp_3s 1111 001 0 0 . 1 . .... .... 1101 ... 1 .... @3same_fp VMUL_fp_3s 1111 001 1 0 . 0 . .... .... 1101 ... 1 .... @3same_fp +VCEQ_fp_3s 1111 001 0 0 . 0 . .... .... 1110 ... 0 .... @3same_fp +VCGE_fp_3s 1111 001 1 0 . 0 . .... .... 1110 ... 0 .... @3same_fp +VACGE_fp_3s 1111 001 1 0 . 0 . .... .... 1110 ... 1 .... @3same_fp +VCGT_fp_3s 1111 001 1 0 . 1 . .... .... 1110 ... 0 .... @3same_fp +VACGT_fp_3s 1111 001 1 0 . 1 . .... .... 1110 ... 1 .... @3same_fp VPMAX_fp_3s 1111 001 1 0 . 0 . .... .... 1111 ... 0 .... @3same_fp_q0 VPMIN_fp_3s 1111 001 1 0 . 1 . .... .... 1111 ... 0 .... @3same_fp_q0 diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3S_FP_GVEC(VMUL, gen_helper_gvec_fmul_s) return do_3same_fp(s, a, FUNC, READS_VD); \ } +DO_3S_FP(VCEQ, gen_helper_neon_ceq_f32, false) +DO_3S_FP(VCGE, gen_helper_neon_cge_f32, false) +DO_3S_FP(VCGT, gen_helper_neon_cgt_f32, false) +DO_3S_FP(VACGE, gen_helper_neon_acge_f32, false) +DO_3S_FP(VACGT, gen_helper_neon_acgt_f32, false) + static void gen_VMLA_fp_3s(TCGv_i32 vd, TCGv_i32 vn, TCGv_i32 vm, TCGv_ptr fpstatus) { 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_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VQDMULH_VQRDMULH: case NEON_3R_FLOAT_ARITH: case NEON_3R_FLOAT_MULTIPLY: + case NEON_3R_FLOAT_CMP: + case NEON_3R_FLOAT_ACMP: /* Already handled by decodetree */ return 1; } @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) return 1; /* VPMIN/VPMAX handled by decodetree */ } break; - case NEON_3R_FLOAT_CMP: - if (!u && size) { - /* no encoding for U=0 C=1x */ - return 1; - } - break; - case NEON_3R_FLOAT_ACMP: - if (!u) { - return 1; - } - break; case NEON_3R_FLOAT_MISC: /* VMAXNM/VMINNM in ARMv8 */ if (u && !arm_dc_feature(s, ARM_FEATURE_V8)) { @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) tmp = neon_load_reg(rn, pass); tmp2 = neon_load_reg(rm, pass); switch (op) { - case NEON_3R_FLOAT_CMP: - { - TCGv_ptr fpstatus = get_fpstatus_ptr(1); - if (!u) { - gen_helper_neon_ceq_f32(tmp, tmp, tmp2, fpstatus); - } else { - if (size == 0) { - gen_helper_neon_cge_f32(tmp, tmp, tmp2, fpstatus); - } else { - gen_helper_neon_cgt_f32(tmp, tmp, tmp2, fpstatus); - } - } - tcg_temp_free_ptr(fpstatus); - break; - } - case NEON_3R_FLOAT_ACMP: - { - TCGv_ptr fpstatus = get_fpstatus_ptr(1); - if (size == 0) { - gen_helper_neon_acge_f32(tmp, tmp, tmp2, fpstatus); - } else { - gen_helper_neon_acgt_f32(tmp, tmp, tmp2, fpstatus); - } - tcg_temp_free_ptr(fpstatus); - break; - } case NEON_3R_FLOAT_MINMAX: { TCGv_ptr fpstatus = get_fpstatus_ptr(1); -- 2.20.1 The usual location for the env argument in the argument list of a TCG helper is immediately after the return-value argument. recps_f32 and rsqrts_f32 differ in that they put it at the end. Move the env argument to its usual place; this will allow us to more easily use these helper functions with the gvec APIs. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-16-peter.maydell@linaro.org --- target/arm/helper.h | 4 ++-- target/arm/translate.c | 4 ++-- target/arm/vfp_helper.c | 4 ++-- 3 files changed, 6 insertions(+), 6 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_3(vfp_fcvt_f64_to_f16, TCG_CALL_NO_RWG, f16, f64, ptr, i32) DEF_HELPER_4(vfp_muladdd, f64, f64, f64, f64, ptr) DEF_HELPER_4(vfp_muladds, f32, f32, f32, f32, ptr) -DEF_HELPER_3(recps_f32, f32, f32, f32, env) -DEF_HELPER_3(rsqrts_f32, f32, f32, f32, env) +DEF_HELPER_3(recps_f32, f32, env, f32, f32) +DEF_HELPER_3(rsqrts_f32, f32, env, f32, f32) DEF_HELPER_FLAGS_2(recpe_f16, TCG_CALL_NO_RWG, f16, f16, ptr) DEF_HELPER_FLAGS_2(recpe_f32, TCG_CALL_NO_RWG, f32, f32, ptr) DEF_HELPER_FLAGS_2(recpe_f64, TCG_CALL_NO_RWG, f64, f64, ptr) 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_neon_data_insn(DisasContext *s, uint32_t insn) tcg_temp_free_ptr(fpstatus); } else { if (size == 0) { - gen_helper_recps_f32(tmp, tmp, tmp2, cpu_env); + gen_helper_recps_f32(tmp, cpu_env, tmp, tmp2); } else { - gen_helper_rsqrts_f32(tmp, tmp, tmp2, cpu_env); + gen_helper_rsqrts_f32(tmp, cpu_env, tmp, tmp2); } } break; 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 @@ uint32_t HELPER(vfp_fcvt_f64_to_f16)(float64 a, void *fpstp, uint32_t ahp_mode) #define float32_three make_float32(0x40400000) #define float32_one_point_five make_float32(0x3fc00000) -float32 HELPER(recps_f32)(float32 a, float32 b, CPUARMState *env) +float32 HELPER(recps_f32)(CPUARMState *env, float32 a, float32 b) { float_status *s = &env->vfp.standard_fp_status; if ((float32_is_infinity(a) && float32_is_zero_or_denormal(b)) || @@ -XXX,XX +XXX,XX @@ float32 HELPER(recps_f32)(float32 a, float32 b, CPUARMState *env) return float32_sub(float32_two, float32_mul(a, b, s), s); } -float32 HELPER(rsqrts_f32)(float32 a, float32 b, CPUARMState *env) +float32 HELPER(rsqrts_f32)(CPUARMState *env, float32 a, float32 b) { float_status *s = &env->vfp.standard_fp_status; float32 product; -- 2.20.1 Convert the Neon fp VMAX/VMIN/VMAXNM/VMINNM/VRECPS/VRSQRTS 3-reg-same insns to decodetree. (These are all the remaining non-accumulation instructions in this group.) Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-17-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 6 +++ target/arm/translate-neon.inc.c | 70 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 42 +------------------- 3 files changed, 78 insertions(+), 40 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VCGE_fp_3s 1111 001 1 0 . 0 . .... .... 1110 ... 0 .... @3same_fp VACGE_fp_3s 1111 001 1 0 . 0 . .... .... 1110 ... 1 .... @3same_fp VCGT_fp_3s 1111 001 1 0 . 1 . .... .... 1110 ... 0 .... @3same_fp VACGT_fp_3s 1111 001 1 0 . 1 . .... .... 1110 ... 1 .... @3same_fp +VMAX_fp_3s 1111 001 0 0 . 0 . .... .... 1111 ... 0 .... @3same_fp +VMIN_fp_3s 1111 001 0 0 . 1 . .... .... 1111 ... 0 .... @3same_fp VPMAX_fp_3s 1111 001 1 0 . 0 . .... .... 1111 ... 0 .... @3same_fp_q0 VPMIN_fp_3s 1111 001 1 0 . 1 . .... .... 1111 ... 0 .... @3same_fp_q0 +VRECPS_fp_3s 1111 001 0 0 . 0 . .... .... 1111 ... 1 .... @3same_fp +VRSQRTS_fp_3s 1111 001 0 0 . 1 . .... .... 1111 ... 1 .... @3same_fp +VMAXNM_fp_3s 1111 001 1 0 . 0 . .... .... 1111 ... 1 .... @3same_fp +VMINNM_fp_3s 1111 001 1 0 . 1 . .... .... 1111 ... 1 .... @3same_fp diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3S_FP(VCGE, gen_helper_neon_cge_f32, false) DO_3S_FP(VCGT, gen_helper_neon_cgt_f32, false) DO_3S_FP(VACGE, gen_helper_neon_acge_f32, false) DO_3S_FP(VACGT, gen_helper_neon_acgt_f32, false) +DO_3S_FP(VMAX, gen_helper_vfp_maxs, false) +DO_3S_FP(VMIN, gen_helper_vfp_mins, false) static void gen_VMLA_fp_3s(TCGv_i32 vd, TCGv_i32 vn, TCGv_i32 vm, TCGv_ptr fpstatus) @@ -XXX,XX +XXX,XX @@ static void gen_VMLS_fp_3s(TCGv_i32 vd, TCGv_i32 vn, TCGv_i32 vm, DO_3S_FP(VMLA, gen_VMLA_fp_3s, true) DO_3S_FP(VMLS, gen_VMLS_fp_3s, true) +static bool trans_VMAXNM_fp_3s(DisasContext *s, arg_3same *a) +{ + if (!arm_dc_feature(s, ARM_FEATURE_V8)) { + return false; + } + + if (a->size != 0) { + /* TODO fp16 support */ + return false; + } + + return do_3same_fp(s, a, gen_helper_vfp_maxnums, false); +} + +static bool trans_VMINNM_fp_3s(DisasContext *s, arg_3same *a) +{ + if (!arm_dc_feature(s, ARM_FEATURE_V8)) { + return false; + } + + if (a->size != 0) { + /* TODO fp16 support */ + return false; + } + + return do_3same_fp(s, a, gen_helper_vfp_minnums, false); +} + +WRAP_ENV_FN(gen_VRECPS_tramp, gen_helper_recps_f32) + +static void gen_VRECPS_fp_3s(unsigned vece, uint32_t rd_ofs, + uint32_t rn_ofs, uint32_t rm_ofs, + uint32_t oprsz, uint32_t maxsz) +{ + static const GVecGen3 ops = { .fni4 = gen_VRECPS_tramp }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, &ops); +} + +static bool trans_VRECPS_fp_3s(DisasContext *s, arg_3same *a) +{ + if (a->size != 0) { + /* TODO fp16 support */ + return false; + } + + return do_3same(s, a, gen_VRECPS_fp_3s); +} + +WRAP_ENV_FN(gen_VRSQRTS_tramp, gen_helper_rsqrts_f32) + +static void gen_VRSQRTS_fp_3s(unsigned vece, uint32_t rd_ofs, + uint32_t rn_ofs, uint32_t rm_ofs, + uint32_t oprsz, uint32_t maxsz) +{ + static const GVecGen3 ops = { .fni4 = gen_VRSQRTS_tramp }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, &ops); +} + +static bool trans_VRSQRTS_fp_3s(DisasContext *s, arg_3same *a) +{ + if (a->size != 0) { + /* TODO fp16 support */ + return false; + } + + return do_3same(s, a, gen_VRSQRTS_fp_3s); +} + static bool do_3same_fp_pair(DisasContext *s, arg_3same *a, VFPGen3OpSPFn *fn) { /* FP operations handled pairwise 32 bits at a time */ 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_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_FLOAT_MULTIPLY: case NEON_3R_FLOAT_CMP: case NEON_3R_FLOAT_ACMP: + case NEON_3R_FLOAT_MINMAX: + case NEON_3R_FLOAT_MISC: /* Already handled by decodetree */ return 1; } @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) return 1; } switch (op) { - case NEON_3R_FLOAT_MINMAX: - if (u) { - return 1; /* VPMIN/VPMAX handled by decodetree */ - } - break; - case NEON_3R_FLOAT_MISC: - /* VMAXNM/VMINNM in ARMv8 */ - if (u && !arm_dc_feature(s, ARM_FEATURE_V8)) { - return 1; - } - break; case NEON_3R_VFM_VQRDMLSH: if (!dc_isar_feature(aa32_simdfmac, s)) { return 1; @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) tmp = neon_load_reg(rn, pass); tmp2 = neon_load_reg(rm, pass); switch (op) { - case NEON_3R_FLOAT_MINMAX: - { - TCGv_ptr fpstatus = get_fpstatus_ptr(1); - if (size == 0) { - gen_helper_vfp_maxs(tmp, tmp, tmp2, fpstatus); - } else { - gen_helper_vfp_mins(tmp, tmp, tmp2, fpstatus); - } - tcg_temp_free_ptr(fpstatus); - break; - } - case NEON_3R_FLOAT_MISC: - if (u) { - /* VMAXNM/VMINNM */ - TCGv_ptr fpstatus = get_fpstatus_ptr(1); - if (size == 0) { - gen_helper_vfp_maxnums(tmp, tmp, tmp2, fpstatus); - } else { - gen_helper_vfp_minnums(tmp, tmp, tmp2, fpstatus); - } - tcg_temp_free_ptr(fpstatus); - } else { - if (size == 0) { - gen_helper_recps_f32(tmp, cpu_env, tmp, tmp2); - } else { - gen_helper_rsqrts_f32(tmp, cpu_env, tmp, tmp2); - } - } - break; case NEON_3R_VFM_VQRDMLSH: { /* VFMA, VFMS: fused multiply-add */ -- 2.20.1 Convert the Neon floating point VFMA and VFMS insn to decodetree. These are the last insns in the 3-reg-same group so we can remove all the support/loop code from the old decoder. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200512163904.10918-18-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 3 + target/arm/translate-neon.inc.c | 41 ++++++++ target/arm/translate.c | 176 +------------------------------- 3 files changed, 46 insertions(+), 174 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ SHA256H2_3s 1111 001 1 0 . 01 .... .... 1100 . 1 . 0 .... \ SHA256SU1_3s 1111 001 1 0 . 10 .... .... 1100 . 1 . 0 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp +VFMA_fp_3s 1111 001 0 0 . 0 . .... .... 1100 ... 1 .... @3same_fp +VFMS_fp_3s 1111 001 0 0 . 1 . .... .... 1100 ... 1 .... @3same_fp + VQRDMLSH_3s 1111 001 1 0 . .. .... .... 1100 ... 1 .... @3same VADD_fp_3s 1111 001 0 0 . 0 . .... .... 1101 ... 0 .... @3same_fp diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static bool trans_VRSQRTS_fp_3s(DisasContext *s, arg_3same *a) return do_3same(s, a, gen_VRSQRTS_fp_3s); } +static void gen_VFMA_fp_3s(TCGv_i32 vd, TCGv_i32 vn, TCGv_i32 vm, + TCGv_ptr fpstatus) +{ + gen_helper_vfp_muladds(vd, vn, vm, vd, fpstatus); +} + +static bool trans_VFMA_fp_3s(DisasContext *s, arg_3same *a) +{ + if (!dc_isar_feature(aa32_simdfmac, s)) { + return false; + } + + if (a->size != 0) { + /* TODO fp16 support */ + return false; + } + + return do_3same_fp(s, a, gen_VFMA_fp_3s, true); +} + +static void gen_VFMS_fp_3s(TCGv_i32 vd, TCGv_i32 vn, TCGv_i32 vm, + TCGv_ptr fpstatus) +{ + gen_helper_vfp_negs(vn, vn); + gen_helper_vfp_muladds(vd, vn, vm, vd, fpstatus); +} + +static bool trans_VFMS_fp_3s(DisasContext *s, arg_3same *a) +{ + if (!dc_isar_feature(aa32_simdfmac, s)) { + return false; + } + + if (a->size != 0) { + /* TODO fp16 support */ + return false; + } + + return do_3same_fp(s, a, gen_VFMS_fp_3s, true); +} + static bool do_3same_fp_pair(DisasContext *s, arg_3same *a, VFPGen3OpSPFn *fn) { /* FP operations handled pairwise 32 bits at a time */ diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_neon_narrow_op(int op, int u, int size, } } -/* Symbolic constants for op fields for Neon 3-register same-length. - * The values correspond to bits [11:8,4]; see the ARM ARM DDI0406B - * table A7-9. - */ -#define NEON_3R_VHADD 0 -#define NEON_3R_VQADD 1 -#define NEON_3R_VRHADD 2 -#define NEON_3R_LOGIC 3 /* VAND,VBIC,VORR,VMOV,VORN,VEOR,VBIF,VBIT,VBSL */ -#define NEON_3R_VHSUB 4 -#define NEON_3R_VQSUB 5 -#define NEON_3R_VCGT 6 -#define NEON_3R_VCGE 7 -#define NEON_3R_VSHL 8 -#define NEON_3R_VQSHL 9 -#define NEON_3R_VRSHL 10 -#define NEON_3R_VQRSHL 11 -#define NEON_3R_VMAX 12 -#define NEON_3R_VMIN 13 -#define NEON_3R_VABD 14 -#define NEON_3R_VABA 15 -#define NEON_3R_VADD_VSUB 16 -#define NEON_3R_VTST_VCEQ 17 -#define NEON_3R_VML 18 /* VMLA, VMLS */ -#define NEON_3R_VMUL 19 -#define NEON_3R_VPMAX 20 -#define NEON_3R_VPMIN 21 -#define NEON_3R_VQDMULH_VQRDMULH 22 -#define NEON_3R_VPADD_VQRDMLAH 23 -#define NEON_3R_SHA 24 /* SHA1C,SHA1P,SHA1M,SHA1SU0,SHA256H{2},SHA256SU1 */ -#define NEON_3R_VFM_VQRDMLSH 25 /* VFMA, VFMS, VQRDMLSH */ -#define NEON_3R_FLOAT_ARITH 26 /* float VADD, VSUB, VPADD, VABD */ -#define NEON_3R_FLOAT_MULTIPLY 27 /* float VMLA, VMLS, VMUL */ -#define NEON_3R_FLOAT_CMP 28 /* float VCEQ, VCGE, VCGT */ -#define NEON_3R_FLOAT_ACMP 29 /* float VACGE, VACGT, VACLE, VACLT */ -#define NEON_3R_FLOAT_MINMAX 30 /* float VMIN, VMAX */ -#define NEON_3R_FLOAT_MISC 31 /* float VRECPS, VRSQRTS, VMAXNM/MINNM */ - -static const uint8_t neon_3r_sizes[] = { - [NEON_3R_VHADD] = 0x7, - [NEON_3R_VQADD] = 0xf, - [NEON_3R_VRHADD] = 0x7, - [NEON_3R_LOGIC] = 0xf, /* size field encodes op type */ - [NEON_3R_VHSUB] = 0x7, - [NEON_3R_VQSUB] = 0xf, - [NEON_3R_VCGT] = 0x7, - [NEON_3R_VCGE] = 0x7, - [NEON_3R_VSHL] = 0xf, - [NEON_3R_VQSHL] = 0xf, - [NEON_3R_VRSHL] = 0xf, - [NEON_3R_VQRSHL] = 0xf, - [NEON_3R_VMAX] = 0x7, - [NEON_3R_VMIN] = 0x7, - [NEON_3R_VABD] = 0x7, - [NEON_3R_VABA] = 0x7, - [NEON_3R_VADD_VSUB] = 0xf, - [NEON_3R_VTST_VCEQ] = 0x7, - [NEON_3R_VML] = 0x7, - [NEON_3R_VMUL] = 0x7, - [NEON_3R_VPMAX] = 0x7, - [NEON_3R_VPMIN] = 0x7, - [NEON_3R_VQDMULH_VQRDMULH] = 0x6, - [NEON_3R_VPADD_VQRDMLAH] = 0x7, - [NEON_3R_SHA] = 0xf, /* size field encodes op type */ - [NEON_3R_VFM_VQRDMLSH] = 0x7, /* For VFM, size bit 1 encodes op */ - [NEON_3R_FLOAT_ARITH] = 0x5, /* size bit 1 encodes op */ - [NEON_3R_FLOAT_MULTIPLY] = 0x5, /* size bit 1 encodes op */ - [NEON_3R_FLOAT_CMP] = 0x5, /* size bit 1 encodes op */ - [NEON_3R_FLOAT_ACMP] = 0x5, /* size bit 1 encodes op */ - [NEON_3R_FLOAT_MINMAX] = 0x5, /* size bit 1 encodes op */ - [NEON_3R_FLOAT_MISC] = 0x5, /* size bit 1 encodes op */ -}; - /* Symbolic constants for op fields for Neon 2-register miscellaneous. * The values correspond to bits [17:16,10:7]; see the ARM ARM DDI0406B * table A7-13. @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) rm_ofs = neon_reg_offset(rm, 0); if ((insn & (1 << 23)) == 0) { - /* Three register same length. */ - op = ((insn >> 7) & 0x1e) | ((insn >> 4) & 1); - /* Catch invalid op and bad size combinations: UNDEF */ - if ((neon_3r_sizes[op] & (1 << size)) == 0) { - return 1; - } - /* All insns of this form UNDEF for either this condition or the - * superset of cases "Q==1"; we catch the latter later. - */ - if (q && ((rd | rn | rm) & 1)) { - return 1; - } - switch (op) { - case NEON_3R_VFM_VQRDMLSH: - if (!u) { - /* VFM, VFMS */ - if (size == 1) { - return 1; - } - break; - } - /* VQRDMLSH : handled by decodetree */ - return 1; - - case NEON_3R_VADD_VSUB: - case NEON_3R_LOGIC: - case NEON_3R_VMAX: - case NEON_3R_VMIN: - case NEON_3R_VTST_VCEQ: - case NEON_3R_VCGT: - case NEON_3R_VCGE: - case NEON_3R_VQADD: - case NEON_3R_VQSUB: - case NEON_3R_VMUL: - case NEON_3R_VML: - case NEON_3R_VSHL: - case NEON_3R_SHA: - case NEON_3R_VHADD: - case NEON_3R_VRHADD: - case NEON_3R_VHSUB: - case NEON_3R_VABD: - case NEON_3R_VABA: - case NEON_3R_VQSHL: - case NEON_3R_VRSHL: - case NEON_3R_VQRSHL: - case NEON_3R_VPMAX: - case NEON_3R_VPMIN: - case NEON_3R_VPADD_VQRDMLAH: - case NEON_3R_VQDMULH_VQRDMULH: - case NEON_3R_FLOAT_ARITH: - case NEON_3R_FLOAT_MULTIPLY: - case NEON_3R_FLOAT_CMP: - case NEON_3R_FLOAT_ACMP: - case NEON_3R_FLOAT_MINMAX: - case NEON_3R_FLOAT_MISC: - /* Already handled by decodetree */ - return 1; - } - - if (size == 3) { - /* 64-bit element instructions: handled by decodetree */ - return 1; - } - switch (op) { - case NEON_3R_VFM_VQRDMLSH: - if (!dc_isar_feature(aa32_simdfmac, s)) { - return 1; - } - break; - default: - break; - } - - for (pass = 0; pass < (q ? 4 : 2); pass++) { - - /* Elementwise. */ - tmp = neon_load_reg(rn, pass); - tmp2 = neon_load_reg(rm, pass); - switch (op) { - case NEON_3R_VFM_VQRDMLSH: - { - /* VFMA, VFMS: fused multiply-add */ - TCGv_ptr fpstatus = get_fpstatus_ptr(1); - TCGv_i32 tmp3 = neon_load_reg(rd, pass); - if (size) { - /* VFMS */ - gen_helper_vfp_negs(tmp, tmp); - } - gen_helper_vfp_muladds(tmp, tmp, tmp2, tmp3, fpstatus); - tcg_temp_free_i32(tmp3); - tcg_temp_free_ptr(fpstatus); - break; - } - default: - abort(); - } - tcg_temp_free_i32(tmp2); - - neon_store_reg(rd, pass, tmp); - - } /* for pass */ - /* End of 3 register same size operations. */ + /* Three register same length: handled by decodetree */ + return 1; } else if (insn & (1 << 4)) { if ((insn & 0x00380080) != 0) { /* Two registers and shift. */ -- 2.20.1