diff options
41 files changed, 3414 insertions, 250 deletions
diff --git a/Makefile.target b/Makefile.target index 8b658c0d13..927347bac2 100644 --- a/Makefile.target +++ b/Makefile.target @@ -143,6 +143,9 @@ GENERATED_HEADERS += hmp-commands.h qmp-commands-old.h endif # CONFIG_SOFTMMU +# Workaround for http://gcc.gnu.org/PR55489, see configure. +%/translate.o: QEMU_CFLAGS += $(TRANSLATE_OPT_CFLAGS) + nested-vars += obj-y # This resolves all nested paths, so it must come last @@ -1 +1 @@ -1.2.90 +1.2.91 diff --git a/aio-win32.c b/aio-win32.c index a84eb71246..cec4646635 100644 --- a/aio-win32.c +++ b/aio-win32.c @@ -173,7 +173,7 @@ bool aio_poll(AioContext *ctx, bool blocking) } /* wait until next event */ - for (;;) { + while (count > 0) { int timeout = blocking ? INFINITE : 0; int ret = WaitForMultipleObjects(count, events, FALSE, timeout); @@ -209,6 +209,9 @@ bool aio_poll(AioContext *ctx, bool blocking) g_free(tmp); } } + + /* Try again, but only call each handler once. */ + events[ret - WAIT_OBJECT_0] = events[--count]; } return progress; @@ -1183,6 +1183,21 @@ for flag in $gcc_flags; do fi done +# Workaround for http://gcc.gnu.org/PR55489. Happens with -fPIE/-fPIC and +# large functions that use global variables. The bug is in all releases of +# GCC, but it became particularly acute in 4.6.x and 4.7.x. It is fixed in +# 4.7.3 and 4.8.0. We should be able to delete this at the end of 2013. +cat > $TMPC << EOF +#if __GNUC__ == 4 && (__GNUC_MINOR__ == 6 || (__GNUC_MINOR__ == 7 && __GNUC_PATCHLEVEL__ <= 2)) +int main(void) { return 0; } +#else +#error No bug in this compiler. +#endif +EOF +if compile_prog "-Werror -fno-gcse" "" ; then + TRANSLATE_OPT_CFLAGS=-fno-gcse +fi + if test "$static" = "yes" ; then if test "$pie" = "yes" ; then echo "static and pie are mutually incompatible" @@ -3662,6 +3677,7 @@ echo "LIBS_TOOLS+=$libs_tools" >> $config_host_mak echo "EXESUF=$EXESUF" >> $config_host_mak echo "LIBS_QGA+=$libs_qga" >> $config_host_mak echo "POD2MAN=$POD2MAN" >> $config_host_mak +echo "TRANSLATE_OPT_CFLAGS=$TRANSLATE_OPT_CFLAGS" >> $config_host_mak # generate list of library paths for linker script diff --git a/hw/Makefile.objs b/hw/Makefile.objs index ea46f8128e..d581d8d6d6 100644 --- a/hw/Makefile.objs +++ b/hw/Makefile.objs @@ -10,6 +10,7 @@ common-obj-$(CONFIG_PCI) += shpc.o common-obj-$(CONFIG_PCI) += slotid_cap.o common-obj-$(CONFIG_PCI) += pci_host.o pcie_host.o common-obj-$(CONFIG_PCI) += ioh3420.o xio3130_upstream.o xio3130_downstream.o +common-obj-$(CONFIG_PCI) += i82801b11.o common-obj-y += watchdog.o common-obj-$(CONFIG_ISA_MMIO) += isa_mmio.o common-obj-$(CONFIG_ECC) += ecc.o @@ -28,7 +29,7 @@ common-obj-$(CONFIG_I8254) += i8254_common.o i8254.o common-obj-$(CONFIG_PCSPK) += pcspk.o common-obj-$(CONFIG_PCKBD) += pckbd.o common-obj-$(CONFIG_FDC) += fdc.o -common-obj-$(CONFIG_ACPI) += acpi.o acpi_piix4.o +common-obj-$(CONFIG_ACPI) += acpi.o acpi_piix4.o acpi_ich9.o smbus_ich9.o common-obj-$(CONFIG_APM) += pm_smbus.o apm.o common-obj-$(CONFIG_DMA) += dma.o common-obj-$(CONFIG_I82374) += i82374.o @@ -38,6 +39,7 @@ common-obj-$(CONFIG_SMARTCARD) += ccid-card-passthru.o common-obj-$(CONFIG_SMARTCARD_NSS) += ccid-card-emulated.o common-obj-$(CONFIG_I8259) += i8259_common.o i8259.o common-obj-y += fifo.o +common-obj-y += pam.o # PPC devices common-obj-$(CONFIG_PREP_PCI) += prep_pci.o diff --git a/hw/acpi_ich9.c b/hw/acpi_ich9.c new file mode 100644 index 0000000000..61034d3bd7 --- /dev/null +++ b/hw/acpi_ich9.c @@ -0,0 +1,322 @@ +/* + * ACPI implementation + * + * Copyright (c) 2006 Fabrice Bellard + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License version 2 as published by the Free Software Foundation. + * + * This library 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 + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, see <http://www.gnu.org/licenses/> + */ +/* + * Copyright (c) 2009 Isaku Yamahata <yamahata at valinux co jp> + * VA Linux Systems Japan K.K. + * Copyright (C) 2012 Jason Baron <jbaron@redhat.com> + * + * This is based on acpi.c. + */ +#include "hw.h" +#include "pc.h" +#include "pci.h" +#include "qemu-timer.h" +#include "sysemu.h" +#include "acpi.h" +#include "kvm.h" + +#include "ich9.h" + +//#define DEBUG + +#ifdef DEBUG +#define ICH9_DEBUG(fmt, ...) \ +do { printf("%s "fmt, __func__, ## __VA_ARGS__); } while (0) +#else +#define ICH9_DEBUG(fmt, ...) do { } while (0) +#endif + +static void pm_ioport_write_fallback(void *opaque, uint32_t addr, int len, + uint32_t val); +static uint32_t pm_ioport_read_fallback(void *opaque, uint32_t addr, int len); + +static void pm_update_sci(ICH9LPCPMRegs *pm) +{ + int sci_level, pm1a_sts; + + pm1a_sts = acpi_pm1_evt_get_sts(&pm->acpi_regs); + + sci_level = (((pm1a_sts & pm->acpi_regs.pm1.evt.en) & + (ACPI_BITMASK_RT_CLOCK_ENABLE | + ACPI_BITMASK_POWER_BUTTON_ENABLE | + ACPI_BITMASK_GLOBAL_LOCK_ENABLE | + ACPI_BITMASK_TIMER_ENABLE)) != 0); + qemu_set_irq(pm->irq, sci_level); + + /* schedule a timer interruption if needed */ + acpi_pm_tmr_update(&pm->acpi_regs, + (pm->acpi_regs.pm1.evt.en & ACPI_BITMASK_TIMER_ENABLE) && + !(pm1a_sts & ACPI_BITMASK_TIMER_STATUS)); +} + +static void ich9_pm_update_sci_fn(ACPIREGS *regs) +{ + ICH9LPCPMRegs *pm = container_of(regs, ICH9LPCPMRegs, acpi_regs); + pm_update_sci(pm); +} + +static void pm_ioport_writeb(void *opaque, uint32_t addr, uint32_t val) +{ + ICH9LPCPMRegs *pm = opaque; + + switch (addr & ICH9_PMIO_MASK) { + case ICH9_PMIO_GPE0_STS ... (ICH9_PMIO_GPE0_STS + ICH9_PMIO_GPE0_LEN - 1): + acpi_gpe_ioport_writeb(&pm->acpi_regs, addr, val); + break; + default: + break; + } + + ICH9_DEBUG("port=0x%04x val=0x%04x\n", addr, val); +} + +static uint32_t pm_ioport_readb(void *opaque, uint32_t addr) +{ + ICH9LPCPMRegs *pm = opaque; + uint32_t val = 0; + + switch (addr & ICH9_PMIO_MASK) { + case ICH9_PMIO_GPE0_STS ... (ICH9_PMIO_GPE0_STS + ICH9_PMIO_GPE0_LEN - 1): + val = acpi_gpe_ioport_readb(&pm->acpi_regs, addr); + break; + default: + val = 0; + break; + } + ICH9_DEBUG("port=0x%04x val=0x%04x\n", addr, val); + return val; +} + +static void pm_ioport_writew(void *opaque, uint32_t addr, uint32_t val) +{ + ICH9LPCPMRegs *pm = opaque; + + switch (addr & ICH9_PMIO_MASK) { + case ICH9_PMIO_PM1_STS: + acpi_pm1_evt_write_sts(&pm->acpi_regs, val); + pm_update_sci(pm); + break; + case ICH9_PMIO_PM1_EN: + pm->acpi_regs.pm1.evt.en = val; + pm_update_sci(pm); + break; + case ICH9_PMIO_PM1_CNT: + acpi_pm1_cnt_write(&pm->acpi_regs, val, 0); + break; + default: + pm_ioport_write_fallback(opaque, addr, 2, val); + break; + } + ICH9_DEBUG("port=0x%04x val=0x%04x\n", addr, val); +} + +static uint32_t pm_ioport_readw(void *opaque, uint32_t addr) +{ + ICH9LPCPMRegs *pm = opaque; + uint32_t val; + + switch (addr & ICH9_PMIO_MASK) { + case ICH9_PMIO_PM1_STS: + val = acpi_pm1_evt_get_sts(&pm->acpi_regs); + break; + case ICH9_PMIO_PM1_EN: + val = pm->acpi_regs.pm1.evt.en; + break; + case ICH9_PMIO_PM1_CNT: + val = pm->acpi_regs.pm1.cnt.cnt; + break; + default: + val = pm_ioport_read_fallback(opaque, addr, 2); + break; + } + ICH9_DEBUG("port=0x%04x val=0x%04x\n", addr, val); + return val; +} + +static void pm_ioport_writel(void *opaque, uint32_t addr, uint32_t val) +{ + ICH9LPCPMRegs *pm = opaque; + + switch (addr & ICH9_PMIO_MASK) { + case ICH9_PMIO_SMI_EN: + pm->smi_en = val; + break; + default: + pm_ioport_write_fallback(opaque, addr, 4, val); + break; + } + ICH9_DEBUG("port=0x%04x val=0x%08x\n", addr, val); +} + +static uint32_t pm_ioport_readl(void *opaque, uint32_t addr) +{ + ICH9LPCPMRegs *pm = opaque; + uint32_t val; + + switch (addr & ICH9_PMIO_MASK) { + case ICH9_PMIO_PM1_TMR: + val = acpi_pm_tmr_get(&pm->acpi_regs); + break; + case ICH9_PMIO_SMI_EN: + val = pm->smi_en; + break; + + default: + val = pm_ioport_read_fallback(opaque, addr, 4); + break; + } + ICH9_DEBUG("port=0x%04x val=0x%08x\n", addr, val); + return val; +} + +static void pm_ioport_write_fallback(void *opaque, uint32_t addr, int len, + uint32_t val) + { + int subsize = (len == 4) ? 2 : 1; + IOPortWriteFunc *ioport_write = + (subsize == 2) ? pm_ioport_writew : pm_ioport_writeb; + + int i; + + for (i = 0; i < len; i += subsize) { + ioport_write(opaque, addr, val); + val >>= 8 * subsize; + } +} + +static uint32_t pm_ioport_read_fallback(void *opaque, uint32_t addr, int len) +{ + int subsize = (len == 4) ? 2 : 1; + IOPortReadFunc *ioport_read = + (subsize == 2) ? pm_ioport_readw : pm_ioport_readb; + + uint32_t val; + int i; + + val = 0; + for (i = 0; i < len; i += subsize) { + val <<= 8 * subsize; + val |= ioport_read(opaque, addr); + } + + return val; +} + +void ich9_pm_iospace_update(ICH9LPCPMRegs *pm, uint32_t pm_io_base) +{ + ICH9_DEBUG("to 0x%x\n", pm_io_base); + + assert((pm_io_base & ICH9_PMIO_MASK) == 0); + + if (pm->pm_io_base != 0) { + isa_unassign_ioport(pm->pm_io_base, ICH9_PMIO_SIZE); + } + + /* don't map at 0 */ + if (pm_io_base == 0) { + return; + } + + register_ioport_write(pm_io_base, ICH9_PMIO_SIZE, 1, pm_ioport_writeb, pm); + register_ioport_read(pm_io_base, ICH9_PMIO_SIZE, 1, pm_ioport_readb, pm); + register_ioport_write(pm_io_base, ICH9_PMIO_SIZE, 2, pm_ioport_writew, pm); + register_ioport_read(pm_io_base, ICH9_PMIO_SIZE, 2, pm_ioport_readw, pm); + register_ioport_write(pm_io_base, ICH9_PMIO_SIZE, 4, pm_ioport_writel, pm); + register_ioport_read(pm_io_base, ICH9_PMIO_SIZE, 4, pm_ioport_readl, pm); + + pm->pm_io_base = pm_io_base; + acpi_gpe_blk(&pm->acpi_regs, pm_io_base + ICH9_PMIO_GPE0_STS); +} + +static int ich9_pm_post_load(void *opaque, int version_id) +{ + ICH9LPCPMRegs *pm = opaque; + uint32_t pm_io_base = pm->pm_io_base; + pm->pm_io_base = 0; + ich9_pm_iospace_update(pm, pm_io_base); + return 0; +} + +#define VMSTATE_GPE_ARRAY(_field, _state) \ + { \ + .name = (stringify(_field)), \ + .version_id = 0, \ + .num = ICH9_PMIO_GPE0_LEN, \ + .info = &vmstate_info_uint8, \ + .size = sizeof(uint8_t), \ + .flags = VMS_ARRAY | VMS_POINTER, \ + .offset = vmstate_offset_pointer(_state, _field, uint8_t), \ + } + +const VMStateDescription vmstate_ich9_pm = { + .name = "ich9_pm", + .version_id = 1, + .minimum_version_id = 1, + .minimum_version_id_old = 1, + .post_load = ich9_pm_post_load, + .fields = (VMStateField[]) { + VMSTATE_UINT16(acpi_regs.pm1.evt.sts, ICH9LPCPMRegs), + VMSTATE_UINT16(acpi_regs.pm1.evt.en, ICH9LPCPMRegs), + VMSTATE_UINT16(acpi_regs.pm1.cnt.cnt, ICH9LPCPMRegs), + VMSTATE_TIMER(acpi_regs.tmr.timer, ICH9LPCPMRegs), + VMSTATE_INT64(acpi_regs.tmr.overflow_time, ICH9LPCPMRegs), + VMSTATE_GPE_ARRAY(acpi_regs.gpe.sts, ICH9LPCPMRegs), + VMSTATE_GPE_ARRAY(acpi_regs.gpe.en, ICH9LPCPMRegs), + VMSTATE_UINT32(smi_en, ICH9LPCPMRegs), + VMSTATE_UINT32(smi_sts, ICH9LPCPMRegs), + VMSTATE_END_OF_LIST() + } +}; + +static void pm_reset(void *opaque) +{ + ICH9LPCPMRegs *pm = opaque; + ich9_pm_iospace_update(pm, 0); + + acpi_pm1_evt_reset(&pm->acpi_regs); + acpi_pm1_cnt_reset(&pm->acpi_regs); + acpi_pm_tmr_reset(&pm->acpi_regs); + acpi_gpe_reset(&pm->acpi_regs); + + if (kvm_enabled()) { + /* Mark SMM as already inited to prevent SMM from running. KVM does not + * support SMM mode. */ + pm->smi_en |= ICH9_PMIO_SMI_EN_APMC_EN; + } + + pm_update_sci(pm); +} + +static void pm_powerdown_req(Notifier *n, void *opaque) +{ + ICH9LPCPMRegs *pm = container_of(n, ICH9LPCPMRegs, powerdown_notifier); + + acpi_pm1_evt_power_down(&pm->acpi_regs); +} + +void ich9_pm_init(ICH9LPCPMRegs *pm, qemu_irq sci_irq, qemu_irq cmos_s3) +{ + acpi_pm_tmr_init(&pm->acpi_regs, ich9_pm_update_sci_fn); + acpi_pm1_cnt_init(&pm->acpi_regs); + acpi_gpe_init(&pm->acpi_regs, ICH9_PMIO_GPE0_LEN); + + pm->irq = sci_irq; + qemu_register_reset(pm_reset, pm); + pm->powerdown_notifier.notify = pm_powerdown_req; + qemu_register_powerdown_notifier(&pm->powerdown_notifier); +} diff --git a/hw/acpi_ich9.h b/hw/acpi_ich9.h new file mode 100644 index 0000000000..180c40673b --- /dev/null +++ b/hw/acpi_ich9.h @@ -0,0 +1,47 @@ +/* + * QEMU GMCH/ICH9 LPC PM Emulation + * + * Copyright (c) 2009 Isaku Yamahata <yamahata at valinux co jp> + * VA Linux Systems Japan K.K. + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This library 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 + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, see <http://www.gnu.org/licenses/> + */ + +#ifndef HW_ACPI_ICH9_H +#define HW_ACPI_ICH9_H + +#include "acpi.h" + +typedef struct ICH9LPCPMRegs { + /* + * In ich9 spec says that pm1_cnt register is 32bit width and + * that the upper 16bits are reserved and unused. + * PM1a_CNT_BLK = 2 in FADT so it is defined as uint16_t. + */ + ACPIREGS acpi_regs; + uint32_t smi_en; + uint32_t smi_sts; + + qemu_irq irq; /* SCI */ + + uint32_t pm_io_base; + Notifier powerdown_notifier; +} ICH9LPCPMRegs; + +void ich9_pm_init(ICH9LPCPMRegs *pm, + qemu_irq sci_irq, qemu_irq cmos_s3_resume); +void ich9_pm_iospace_update(ICH9LPCPMRegs *pm, uint32_t pm_io_base); +extern const VMStateDescription vmstate_ich9_pm; + +#endif /* HW_ACPI_ICH9_H */ diff --git a/hw/i386/Makefile.objs b/hw/i386/Makefile.objs index 8c764bbfef..0d3f6a8e84 100644 --- a/hw/i386/Makefile.objs +++ b/hw/i386/Makefile.objs @@ -6,6 +6,7 @@ obj-y += pci-hotplug.o smbios.o wdt_ib700.o obj-y += debugcon.o multiboot.o obj-y += pc_piix.o obj-y += pc_sysfw.o +obj-y += lpc_ich9.o q35.o pc_q35.o obj-$(CONFIG_XEN) += xen_platform.o xen_apic.o obj-$(CONFIG_XEN_PCI_PASSTHROUGH) += xen-host-pci-device.o obj-$(CONFIG_XEN_PCI_PASSTHROUGH) += xen_pt.o xen_pt_config_init.o xen_pt_msi.o diff --git a/hw/i8259_internal.h b/hw/i8259_internal.h index 4137b61703..8785b1da3f 100644 --- a/hw/i8259_internal.h +++ b/hw/i8259_internal.h @@ -33,7 +33,7 @@ typedef struct PICCommonState PICCommonState; #define TYPE_PIC_COMMON "pic-common" #define PIC_COMMON(obj) \ - OBJECT_CHECK(PICCommon, (obj), TYPE_PIC_COMMON) + OBJECT_CHECK(PICCommonState, (obj), TYPE_PIC_COMMON) #define PIC_COMMON_CLASS(klass) \ OBJECT_CLASS_CHECK(PICCommonClass, (klass), TYPE_PIC_COMMON) #define PIC_COMMON_GET_CLASS(obj) \ diff --git a/hw/i82801b11.c b/hw/i82801b11.c new file mode 100644 index 0000000000..3d1f996b2f --- /dev/null +++ b/hw/i82801b11.c @@ -0,0 +1,125 @@ +/* + * Copyright (c) 2006 Fabrice Bellard + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ +/* + * QEMU i82801b11 dmi-to-pci Bridge Emulation + * + * Copyright (c) 2009, 2010, 2011 + * Isaku Yamahata <yamahata at valinux co jp> + * VA Linux Systems Japan K.K. + * Copyright (C) 2012 Jason Baron <jbaron@redhat.com> + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This library 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 + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, see <http://www.gnu.org/licenses/> + */ + +#include "pci.h" +#include "ich9.h" + + +/*****************************************************************************/ +/* ICH9 DMI-to-PCI bridge */ +#define I82801ba_SSVID_OFFSET 0x50 +#define I82801ba_SSVID_SVID 0 +#define I82801ba_SSVID_SSID 0 + +typedef struct I82801b11Bridge { + PCIBridge br; +} I82801b11Bridge; + +static int i82801b11_bridge_initfn(PCIDevice *d) +{ + int rc; + + rc = pci_bridge_initfn(d); + if (rc < 0) { + return rc; + } + + rc = pci_bridge_ssvid_init(d, I82801ba_SSVID_OFFSET, + I82801ba_SSVID_SVID, I82801ba_SSVID_SSID); + if (rc < 0) { + goto err_bridge; + } + pci_config_set_prog_interface(d->config, PCI_CLASS_BRDIGE_PCI_INF_SUB); + return 0; + +err_bridge: + pci_bridge_exitfn(d); + + return rc; +} + +static void i82801b11_bridge_class_init(ObjectClass *klass, void *data) +{ + PCIDeviceClass *k = PCI_DEVICE_CLASS(klass); + + k->is_bridge = 1; + k->vendor_id = PCI_VENDOR_ID_INTEL; + k->device_id = PCI_DEVICE_ID_INTEL_82801BA_11; + k->revision = ICH9_D2P_A2_REVISION; + k->init = i82801b11_bridge_initfn; +} + +static const TypeInfo i82801b11_bridge_info = { + .name = "i82801b11-bridge", + .parent = TYPE_PCI_DEVICE, + .instance_size = sizeof(I82801b11Bridge), + .class_init = i82801b11_bridge_class_init, +}; + +PCIBus *ich9_d2pbr_init(PCIBus *bus, int devfn, int sec_bus) +{ + PCIDevice *d; + PCIBridge *br; + char buf[16]; + DeviceState *qdev; + + d = pci_create_multifunction(bus, devfn, true, "i82801b11-bridge"); + if (!d) { + return NULL; + } + br = DO_UPCAST(PCIBridge, dev, d); + qdev = &br->dev.qdev; + + snprintf(buf, sizeof(buf), "pci.%d", sec_bus); + pci_bridge_map_irq(br, buf, pci_swizzle_map_irq_fn); + qdev_init_nofail(qdev); + + return pci_bridge_get_sec_bus(br); +} + +static void d2pbr_register(void) +{ + type_register_static(&i82801b11_bridge_info); +} + +type_init(d2pbr_register); diff --git a/hw/ich9.h b/hw/ich9.h new file mode 100644 index 0000000000..de491350c3 --- /dev/null +++ b/hw/ich9.h @@ -0,0 +1,207 @@ +#ifndef HW_ICH9_H +#define HW_ICH9_H + +#include "hw.h" +#include "range.h" +#include "isa.h" +#include "sysbus.h" +#include "pc.h" +#include "apm.h" +#include "ioapic.h" +#include "pci.h" +#include "pcie_host.h" +#include "pci_bridge.h" +#include "acpi.h" +#include "acpi_ich9.h" +#include "pam.h" +#include "pci_internals.h" + +void ich9_lpc_set_irq(void *opaque, int irq_num, int level); +int ich9_lpc_map_irq(PCIDevice *pci_dev, int intx); +void ich9_lpc_pm_init(PCIDevice *pci_lpc, qemu_irq cmos_s3); +PCIBus *ich9_d2pbr_init(PCIBus *bus, int devfn, int sec_bus); +i2c_bus *ich9_smb_init(PCIBus *bus, int devfn, uint32_t smb_io_base); + +#define ICH9_CC_SIZE (16 * 1024) /* 16KB */ + +#define TYPE_ICH9_LPC_DEVICE "ICH9 LPC" +#define ICH9_LPC_DEVICE(obj) \ + OBJECT_CHECK(ICH9LPCState, (obj), TYPE_ICH9_LPC_DEVICE) + +typedef struct ICH9LPCState { + /* ICH9 LPC PCI to ISA bridge */ + PCIDevice d; + + /* (pci device, intx) -> pirq + * In real chipset case, the unused slots are never used + * as ICH9 supports only D25-D32 irq routing. + * On the other hand in qemu case, any slot/function can be populated + * via command line option. + * So fallback interrupt routing for any devices in any slots is necessary. + */ + uint8_t irr[PCI_SLOT_MAX][PCI_NUM_PINS]; + + APMState apm; + ICH9LPCPMRegs pm; + uint32_t sci_level; /* track sci level */ + + /* 10.1 Chipset Configuration registers(Memory Space) + which is pointed by RCBA */ + uint8_t chip_config[ICH9_CC_SIZE]; + /* isa bus */ + ISABus *isa_bus; + MemoryRegion rbca_mem; + + qemu_irq *pic; + qemu_irq *ioapic; +} ICH9LPCState; + +#define Q35_MASK(bit, ms_bit, ls_bit) \ +((uint##bit##_t)(((1ULL << ((ms_bit) + 1)) - 1) & ~((1ULL << ls_bit) - 1))) + +/* ICH9: Chipset Configuration Registers */ +#define ICH9_CC_ADDR_MASK (ICH9_CC_SIZE - 1) + +#define ICH9_CC +#define ICH9_CC_D28IP 0x310C +#define ICH9_CC_D28IP_SHIFT 4 +#define ICH9_CC_D28IP_MASK 0xf +#define ICH9_CC_D28IP_DEFAULT 0x00214321 +#define ICH9_CC_D31IR 0x3140 +#define ICH9_CC_D30IR 0x3142 +#define ICH9_CC_D29IR 0x3144 +#define ICH9_CC_D28IR 0x3146 +#define ICH9_CC_D27IR 0x3148 +#define ICH9_CC_D26IR 0x314C +#define ICH9_CC_D25IR 0x3150 +#define ICH9_CC_DIR_DEFAULT 0x3210 +#define ICH9_CC_D30IR_DEFAULT 0x0 +#define ICH9_CC_DIR_SHIFT 4 +#define ICH9_CC_DIR_MASK 0x7 +#define ICH9_CC_OIC 0x31FF +#define ICH9_CC_OIC_AEN 0x1 + +/* D28:F[0-5] */ +#define ICH9_PCIE_DEV 28 +#define ICH9_PCIE_FUNC_MAX 6 + + +/* D29:F0 USB UHCI Controller #1 */ +#define ICH9_USB_UHCI1_DEV 29 +#define ICH9_USB_UHCI1_FUNC 0 + +/* D30:F0 DMI-to-PCI brdige */ +#define ICH9_D2P_BRIDGE "ICH9 D2P BRIDGE" +#define ICH9_D2P_BRIDGE_SAVEVM_VERSION 0 + +#define ICH9_D2P_BRIDGE_DEV 30 +#define ICH9_D2P_BRIDGE_FUNC 0 + +#define ICH9_D2P_SECONDARY_DEFAULT (256 - 8) + +#define ICH9_D2P_A2_REVISION 0x92 + + +/* D31:F1 LPC controller */ +#define ICH9_A2_LPC "ICH9 A2 LPC" +#define ICH9_A2_LPC_SAVEVM_VERSION 0 + +#define ICH9_LPC_DEV 31 +#define ICH9_LPC_FUNC 0 + +#define ICH9_A2_LPC_REVISION 0x2 +#define ICH9_LPC_NB_PIRQS 8 /* PCI A-H */ + +#define ICH9_LPC_PMBASE 0x40 +#define ICH9_LPC_PMBASE_BASE_ADDRESS_MASK Q35_MASK(32, 15, 7) +#define ICH9_LPC_PMBASE_RTE 0x1 +#define ICH9_LPC_PMBASE_DEFAULT 0x1 +#define ICH9_LPC_ACPI_CTRL 0x44 +#define ICH9_LPC_ACPI_CTRL_ACPI_EN 0x80 +#define ICH9_LPC_ACPI_CTRL_SCI_IRQ_SEL_MASK Q35_MASK(8, 2, 0) +#define ICH9_LPC_ACPI_CTRL_9 0x0 +#define ICH9_LPC_ACPI_CTRL_10 0x1 +#define ICH9_LPC_ACPI_CTRL_11 0x2 +#define ICH9_LPC_ACPI_CTRL_20 0x4 +#define ICH9_LPC_ACPI_CTRL_21 0x5 +#define ICH9_LPC_ACPI_CTRL_DEFAULT 0x0 + +#define ICH9_LPC_PIRQA_ROUT 0x60 +#define ICH9_LPC_PIRQB_ROUT 0x61 +#define ICH9_LPC_PIRQC_ROUT 0x62 +#define ICH9_LPC_PIRQD_ROUT 0x63 + +#define ICH9_LPC_PIRQE_ROUT 0x68 +#define ICH9_LPC_PIRQF_ROUT 0x69 +#define ICH9_LPC_PIRQG_ROUT 0x6a +#define ICH9_LPC_PIRQH_ROUT 0x6b + +#define ICH9_LPC_PIRQ_ROUT_IRQEN 0x80 +#define ICH9_LPC_PIRQ_ROUT_MASK Q35_MASK(8, 3, 0) +#define ICH9_LPC_PIRQ_ROUT_DEFAULT 0x80 + +#define ICH9_LPC_RCBA 0xf0 +#define ICH9_LPC_RCBA_BA_MASK Q35_MASK(32, 31, 14) +#define ICH9_LPC_RCBA_EN 0x1 +#define ICH9_LPC_RCBA_DEFAULT 0x0 + +#define ICH9_LPC_PIC_NUM_PINS 16 +#define ICH9_LPC_IOAPIC_NUM_PINS 24 + +/* D31:F2 SATA Controller #1 */ +#define ICH9_SATA1_DEV 31 +#define ICH9_SATA1_FUNC 2 + +/* D30:F1 power management I/O registers + offset from the address ICH9_LPC_PMBASE */ + +/* ICH9 LPC PM I/O registers are 128 ports and 128-aligned */ +#define ICH9_PMIO_SIZE 128 +#define ICH9_PMIO_MASK (ICH9_PMIO_SIZE - 1) + +#define ICH9_PMIO_PM1_STS 0x00 +#define ICH9_PMIO_PM1_EN 0x02 +#define ICH9_PMIO_PM1_CNT 0x04 +#define ICH9_PMIO_PM1_TMR 0x08 +#define ICH9_PMIO_GPE0_STS 0x20 +#define ICH9_PMIO_GPE0_EN 0x28 +#define ICH9_PMIO_GPE0_LEN 16 +#define ICH9_PMIO_SMI_EN 0x30 +#define ICH9_PMIO_SMI_EN_APMC_EN (1 << 5) +#define ICH9_PMIO_SMI_STS 0x34 + +/* FADT ACPI_ENABLE/ACPI_DISABLE */ +#define ICH9_APM_ACPI_ENABLE 0x2 +#define ICH9_APM_ACPI_DISABLE 0x3 + + +/* D31:F3 SMBus controller */ +#define ICH9_A2_SMB_REVISION 0x02 +#define ICH9_SMB_PI 0x00 + +#define ICH9_SMB_SMBMBAR0 0x10 +#define ICH9_SMB_SMBMBAR1 0x14 +#define ICH9_SMB_SMBM_BAR 0 +#define ICH9_SMB_SMBM_SIZE (1 << 8) +#define ICH9_SMB_SMB_BASE 0x20 +#define ICH9_SMB_SMB_BASE_BAR 4 +#define ICH9_SMB_SMB_BASE_SIZE (1 << 5) +#define ICH9_SMB_HOSTC 0x40 +#define ICH9_SMB_HOSTC_SSRESET ((uint8_t)(1 << 3)) +#define ICH9_SMB_HOSTC_I2C_EN ((uint8_t)(1 << 2)) +#define ICH9_SMB_HOSTC_SMB_SMI_EN ((uint8_t)(1 << 1)) +#define ICH9_SMB_HOSTC_HST_EN ((uint8_t)(1 << 0)) + +/* D31:F3 SMBus I/O and memory mapped I/O registers */ +#define ICH9_SMB_DEV 31 +#define ICH9_SMB_FUNC 3 + +#define ICH9_SMB_HST_STS 0x00 +#define ICH9_SMB_HST_CNT 0x02 +#define ICH9_SMB_HST_CMD 0x03 +#define ICH9_SMB_XMIT_SLVA 0x04 +#define ICH9_SMB_HST_D0 0x05 +#define ICH9_SMB_HST_D1 0x06 +#define ICH9_SMB_HOST_BLOCK_DB 0x07 + +#endif /* HW_ICH9_H */ diff --git a/hw/kvm/ioapic.c b/hw/kvm/ioapic.c index 6c3b8fe39a..f95c157591 100644 --- a/hw/kvm/ioapic.c +++ b/hw/kvm/ioapic.c @@ -15,6 +15,46 @@ #include "hw/apic_internal.h" #include "kvm.h" +/* PC Utility function */ +void kvm_pc_setup_irq_routing(bool pci_enabled) +{ + KVMState *s = kvm_state; + int i; + + if (kvm_check_extension(s, KVM_CAP_IRQ_ROUTING)) { + for (i = 0; i < 8; ++i) { + if (i == 2) { + continue; + } + kvm_irqchip_add_irq_route(s, i, KVM_IRQCHIP_PIC_MASTER, i); + } + for (i = 8; i < 16; ++i) { + kvm_irqchip_add_irq_route(s, i, KVM_IRQCHIP_PIC_SLAVE, i - 8); + } + if (pci_enabled) { + for (i = 0; i < 24; ++i) { + if (i == 0) { + kvm_irqchip_add_irq_route(s, i, KVM_IRQCHIP_IOAPIC, 2); + } else if (i != 2) { + kvm_irqchip_add_irq_route(s, i, KVM_IRQCHIP_IOAPIC, i); + } + } + } + } +} + +void kvm_pc_gsi_handler(void *opaque, int n, int level) +{ + GSIState *s = opaque; + + if (n < ISA_NUM_IRQS) { + /* Kernel will forward to both PIC and IOAPIC */ + qemu_set_irq(s->i8259_irq[n], level); + } else { + qemu_set_irq(s->ioapic_irq[n], level); + } +} + typedef struct KVMIOAPICState KVMIOAPICState; struct KVMIOAPICState { diff --git a/hw/lpc_ich9.c b/hw/lpc_ich9.c new file mode 100644 index 0000000000..2fc83a496f --- /dev/null +++ b/hw/lpc_ich9.c @@ -0,0 +1,525 @@ +/* + * Copyright (c) 2006 Fabrice Bellard + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ +/* + * QEMU ICH9 Emulation + * + * Copyright (c) 2009, 2010, 2011 + * Isaku Yamahata <yamahata at valinux co jp> + * VA Linux Systems Japan K.K. + * Copyright (C) 2012 Jason Baron <jbaron@redhat.com> + * + * This is based on piix_pci.c, but heavily modified. + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This library 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 + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, see <http://www.gnu.org/licenses/> + */ + +#include "qemu-common.h" +#include "hw.h" +#include "range.h" +#include "isa.h" +#include "sysbus.h" +#include "pc.h" +#include "apm.h" +#include "ioapic.h" +#include "pci.h" +#include "pcie_host.h" +#include "pci_bridge.h" +#include "ich9.h" +#include "acpi.h" +#include "acpi_ich9.h" +#include "pam.h" +#include "pci_internals.h" +#include "exec-memory.h" + +static int ich9_lpc_sci_irq(ICH9LPCState *lpc); + +/*****************************************************************************/ +/* ICH9 LPC PCI to ISA bridge */ + +static void ich9_lpc_reset(DeviceState *qdev); + +/* chipset configuration register + * to access chipset configuration registers, pci_[sg]et_{byte, word, long} + * are used. + * Although it's not pci configuration space, it's little endian as Intel. + */ + +static void ich9_cc_update_ir(uint8_t irr[PCI_NUM_PINS], uint16_t ir) +{ + int intx; + for (intx = 0; intx < PCI_NUM_PINS; intx++) { + irr[intx] = (ir >> (intx * ICH9_CC_DIR_SHIFT)) & ICH9_CC_DIR_MASK; + } +} + +static void ich9_cc_update(ICH9LPCState *lpc) +{ + int slot; + int pci_intx; + + const int reg_offsets[] = { + ICH9_CC_D25IR, + ICH9_CC_D26IR, + ICH9_CC_D27IR, + ICH9_CC_D28IR, + ICH9_CC_D29IR, + ICH9_CC_D30IR, + ICH9_CC_D31IR, + }; + const int *offset; + + /* D{25 - 31}IR, but D30IR is read only to 0. */ + for (slot = 25, offset = reg_offsets; slot < 32; slot++, offset++) { + if (slot == 30) { + continue; + } + ich9_cc_update_ir(lpc->irr[slot], + pci_get_word(lpc->chip_config + *offset)); + } + + /* + * D30: DMI2PCI bridge + * It is arbitrarily decided how INTx lines of PCI devicesbehind the bridge + * are connected to pirq lines. Our choice is PIRQ[E-H]. + * INT[A-D] are connected to PIRQ[E-H] + */ + for (pci_intx = 0; pci_intx < PCI_NUM_PINS; pci_intx++) { + lpc->irr[30][pci_intx] = pci_intx + 4; + } +} + +static void ich9_cc_init(ICH9LPCState *lpc) +{ + int slot; + int intx; + + /* the default irq routing is arbitrary as long as it matches with + * acpi irq routing table. + * The one that is incompatible with piix_pci(= bochs) one is + * intentionally chosen to let the users know that the different + * board is used. + * + * int[A-D] -> pirq[E-F] + * avoid pirq A-D because they are used for pci express port + */ + for (slot = 0; slot < PCI_SLOT_MAX; slot++) { + for (intx = 0; intx < PCI_NUM_PINS; intx++) { + lpc->irr[slot][intx] = (slot + intx) % 4 + 4; + } + } + ich9_cc_update(lpc); +} + +static void ich9_cc_reset(ICH9LPCState *lpc) +{ + uint8_t *c = lpc->chip_config; + + memset(lpc->chip_config, 0, sizeof(lpc->chip_config)); + + pci_set_long(c + ICH9_CC_D31IR, ICH9_CC_DIR_DEFAULT); + pci_set_long(c + ICH9_CC_D30IR, ICH9_CC_D30IR_DEFAULT); + pci_set_long(c + ICH9_CC_D29IR, ICH9_CC_DIR_DEFAULT); + pci_set_long(c + ICH9_CC_D28IR, ICH9_CC_DIR_DEFAULT); + pci_set_long(c + ICH9_CC_D27IR, ICH9_CC_DIR_DEFAULT); + pci_set_long(c + ICH9_CC_D26IR, ICH9_CC_DIR_DEFAULT); + pci_set_long(c + ICH9_CC_D25IR, ICH9_CC_DIR_DEFAULT); + + ich9_cc_update(lpc); +} + +static void ich9_cc_addr_len(uint64_t *addr, unsigned *len) +{ + *addr &= ICH9_CC_ADDR_MASK; + if (*addr + *len >= ICH9_CC_SIZE) { + *len = ICH9_CC_SIZE - *addr; + } +} + +/* val: little endian */ +static void ich9_cc_write(void *opaque, hwaddr addr, + uint64_t val, unsigned len) +{ + ICH9LPCState *lpc = (ICH9LPCState *)opaque; + + ich9_cc_addr_len(&addr, &len); + memcpy(lpc->chip_config + addr, &val, len); + ich9_cc_update(lpc); +} + +/* return value: little endian */ +static uint64_t ich9_cc_read(void *opaque, hwaddr addr, + unsigned len) +{ + ICH9LPCState *lpc = (ICH9LPCState *)opaque; + + uint32_t val = 0; + ich9_cc_addr_len(&addr, &len); + memcpy(&val, lpc->chip_config + addr, len); + return val; +} + +/* IRQ routing */ +/* */ +static void ich9_lpc_rout(uint8_t pirq_rout, int *pic_irq, int *pic_dis) +{ + *pic_irq = pirq_rout & ICH9_LPC_PIRQ_ROUT_MASK; + *pic_dis = pirq_rout & ICH9_LPC_PIRQ_ROUT_IRQEN; +} + +static void ich9_lpc_pic_irq(ICH9LPCState *lpc, int pirq_num, + int *pic_irq, int *pic_dis) +{ + switch (pirq_num) { + case 0 ... 3: /* A-D */ + ich9_lpc_rout(lpc->d.config[ICH9_LPC_PIRQA_ROUT + pirq_num], + pic_irq, pic_dis); + return; + case 4 ... 7: /* E-H */ + ich9_lpc_rout(lpc->d.config[ICH9_LPC_PIRQE_ROUT + (pirq_num - 4)], + pic_irq, pic_dis); + return; + default: + break; + } + abort(); +} + +/* pic_irq: i8254 irq 0-15 */ +static void ich9_lpc_update_pic(ICH9LPCState *lpc, int pic_irq) +{ + int i, pic_level; + + /* The pic level is the logical OR of all the PCI irqs mapped to it */ + pic_level = 0; + for (i = 0; i < ICH9_LPC_NB_PIRQS; i++) { + int tmp_irq; + int tmp_dis; + ich9_lpc_pic_irq(lpc, i, &tmp_irq, &tmp_dis); + if (!tmp_dis && pic_irq == tmp_irq) { + pic_level |= pci_bus_get_irq_level(lpc->d.bus, i); + } + } + if (pic_irq == ich9_lpc_sci_irq(lpc)) { + pic_level |= lpc->sci_level; + } + + qemu_set_irq(lpc->pic[pic_irq], pic_level); +} + +/* pirq: pirq[A-H] 0-7*/ +static void ich9_lpc_update_by_pirq(ICH9LPCState *lpc, int pirq) +{ + int pic_irq; + int pic_dis; + + ich9_lpc_pic_irq(lpc, pirq, &pic_irq, &pic_dis); + assert(pic_irq < ICH9_LPC_PIC_NUM_PINS); + if (pic_dis) { + return; + } + + ich9_lpc_update_pic(lpc, pic_irq); +} + +/* APIC mode: GSIx: PIRQ[A-H] -> GSI 16, ... no pirq shares same APIC pins. */ +static int ich9_pirq_to_gsi(int pirq) +{ + return pirq + ICH9_LPC_PIC_NUM_PINS; +} + +static int ich9_gsi_to_pirq(int gsi) +{ + return gsi - ICH9_LPC_PIC_NUM_PINS; +} + +static void ich9_lpc_update_apic(ICH9LPCState *lpc, int gsi) +{ + int level = 0; + + if (gsi >= ICH9_LPC_PIC_NUM_PINS) { + level |= pci_bus_get_irq_level(lpc->d.bus, ich9_gsi_to_pirq(gsi)); + } + if (gsi == ich9_lpc_sci_irq(lpc)) { + level |= lpc->sci_level; + } + + qemu_set_irq(lpc->ioapic[gsi], level); +} + +void ich9_lpc_set_irq(void *opaque, int pirq, int level) +{ + ICH9LPCState *lpc = opaque; + + assert(0 <= pirq); + assert(pirq < ICH9_LPC_NB_PIRQS); + + ich9_lpc_update_apic(lpc, ich9_pirq_to_gsi(pirq)); + ich9_lpc_update_by_pirq(lpc, pirq); +} + +/* return the pirq number (PIRQ[A-H]:0-7) corresponding to + * a given device irq pin. + */ +int ich9_lpc_map_irq(PCIDevice *pci_dev, int intx) +{ + BusState *bus = qdev_get_parent_bus(&pci_dev->qdev); + PCIBus *pci_bus = PCI_BUS(bus); + PCIDevice *lpc_pdev = + pci_bus->devices[PCI_DEVFN(ICH9_LPC_DEV, ICH9_LPC_FUNC)]; + ICH9LPCState *lpc = ICH9_LPC_DEVICE(lpc_pdev); + + return lpc->irr[PCI_SLOT(pci_dev->devfn)][intx]; +} + +static int ich9_lpc_sci_irq(ICH9LPCState *lpc) +{ + switch (lpc->d.config[ICH9_LPC_ACPI_CTRL] & + ICH9_LPC_ACPI_CTRL_SCI_IRQ_SEL_MASK) { + case ICH9_LPC_ACPI_CTRL_9: + return 9; + case ICH9_LPC_ACPI_CTRL_10: + return 10; + case ICH9_LPC_ACPI_CTRL_11: + return 11; + case ICH9_LPC_ACPI_CTRL_20: + return 20; + case ICH9_LPC_ACPI_CTRL_21: + return 21; + default: + /* reserved */ + break; + } + return -1; +} + +static void ich9_set_sci(void *opaque, int irq_num, int level) +{ + ICH9LPCState *lpc = opaque; + int irq; + + assert(irq_num == 0); + level = !!level; + if (level == lpc->sci_level) { + return; + } + lpc->sci_level = level; + + irq = ich9_lpc_sci_irq(lpc); + if (irq < 0) { + return; + } + + ich9_lpc_update_apic(lpc, irq); + if (irq < ICH9_LPC_PIC_NUM_PINS) { + ich9_lpc_update_pic(lpc, irq); + } +} + +void ich9_lpc_pm_init(PCIDevice *lpc_pci, qemu_irq cmos_s3) +{ + ICH9LPCState *lpc = ICH9_LPC_DEVICE(lpc_pci); + qemu_irq *sci_irq; + + sci_irq = qemu_allocate_irqs(ich9_set_sci, lpc, 1); + ich9_pm_init(&lpc->pm, sci_irq[0], cmos_s3); + + ich9_lpc_reset(&lpc->d.qdev); +} + +/* APM */ + +static void ich9_apm_ctrl_changed(uint32_t val, void *arg) +{ + ICH9LPCState *lpc = arg; + + /* ACPI specs 3.0, 4.7.2.5 */ + acpi_pm1_cnt_update(&lpc->pm.acpi_regs, + val == ICH9_APM_ACPI_ENABLE, + val == ICH9_APM_ACPI_DISABLE); + + /* SMI_EN = PMBASE + 30. SMI control and enable register */ + if (lpc->pm.smi_en & ICH9_PMIO_SMI_EN_APMC_EN) { + cpu_interrupt(first_cpu, CPU_INTERRUPT_SMI); + } +} + +/* config:PMBASE */ +static void +ich9_lpc_pmbase_update(ICH9LPCState *lpc) +{ + uint32_t pm_io_base = pci_get_long(lpc->d.config + ICH9_LPC_PMBASE); + pm_io_base &= ICH9_LPC_PMBASE_BASE_ADDRESS_MASK; + + ich9_pm_iospace_update(&lpc->pm, pm_io_base); +} + +/* config:RBCA */ +static void ich9_lpc_rcba_update(ICH9LPCState *lpc, uint32_t rbca_old) +{ + uint32_t rbca = pci_get_long(lpc->d.config + ICH9_LPC_RCBA); + + if (rbca_old & ICH9_LPC_RCBA_EN) { + memory_region_del_subregion(get_system_memory(), &lpc->rbca_mem); + } + if (rbca & ICH9_LPC_RCBA_EN) { + memory_region_add_subregion_overlap(get_system_memory(), + rbca & ICH9_LPC_RCBA_BA_MASK, + &lpc->rbca_mem, 1); + } +} + +static int ich9_lpc_post_load(void *opaque, int version_id) +{ + ICH9LPCState *lpc = opaque; + + ich9_lpc_pmbase_update(lpc); + ich9_lpc_rcba_update(lpc, 0 /* disabled ICH9_LPC_RBCA_EN */); + return 0; +} + +static void ich9_lpc_config_write(PCIDevice *d, + uint32_t addr, uint32_t val, int len) +{ + ICH9LPCState *lpc = ICH9_LPC_DEVICE(d); + uint32_t rbca_old = pci_get_long(d->config + ICH9_LPC_RCBA); + + pci_default_write_config(d, addr, val, len); + if (ranges_overlap(addr, len, ICH9_LPC_PMBASE, 4)) { + ich9_lpc_pmbase_update(lpc); + } + if (ranges_overlap(addr, len, ICH9_LPC_RCBA, 4)) { + ich9_lpc_rcba_update(lpc, rbca_old); + } +} + +static void ich9_lpc_reset(DeviceState *qdev) +{ + PCIDevice *d = PCI_DEVICE(qdev); + ICH9LPCState *lpc = ICH9_LPC_DEVICE(d); + uint32_t rbca_old = pci_get_long(d->config + ICH9_LPC_RCBA); + int i; + + for (i = 0; i < 4; i++) { + pci_set_byte(d->config + ICH9_LPC_PIRQA_ROUT + i, + ICH9_LPC_PIRQ_ROUT_DEFAULT); + } + for (i = 0; i < 4; i++) { + pci_set_byte(d->config + ICH9_LPC_PIRQE_ROUT + i, + ICH9_LPC_PIRQ_ROUT_DEFAULT); + } + pci_set_byte(d->config + ICH9_LPC_ACPI_CTRL, ICH9_LPC_ACPI_CTRL_DEFAULT); + + pci_set_long(d->config + ICH9_LPC_PMBASE, ICH9_LPC_PMBASE_DEFAULT); + pci_set_long(d->config + ICH9_LPC_RCBA, ICH9_LPC_RCBA_DEFAULT); + + ich9_cc_reset(lpc); + + ich9_lpc_pmbase_update(lpc); + ich9_lpc_rcba_update(lpc, rbca_old); + + lpc->sci_level = 0; +} + +static const MemoryRegionOps rbca_mmio_ops = { + .read = ich9_cc_read, + .write = ich9_cc_write, + .endianness = DEVICE_LITTLE_ENDIAN, +}; + +static int ich9_lpc_initfn(PCIDevice *d) +{ + ICH9LPCState *lpc = ICH9_LPC_DEVICE(d); + ISABus *isa_bus; + + isa_bus = isa_bus_new(&d->qdev, get_system_io()); + + pci_set_long(d->wmask + ICH9_LPC_PMBASE, + ICH9_LPC_PMBASE_BASE_ADDRESS_MASK); + + memory_region_init_io(&lpc->rbca_mem, &rbca_mmio_ops, lpc, + "lpc-rbca-mmio", ICH9_CC_SIZE); + + lpc->isa_bus = isa_bus; + + ich9_cc_init(lpc); + apm_init(&lpc->apm, ich9_apm_ctrl_changed, lpc); + return 0; +} + +static const VMStateDescription vmstate_ich9_lpc = { + .name = "ICH9LPC", + .version_id = 1, + .minimum_version_id = 1, + .minimum_version_id_old = 1, + .post_load = ich9_lpc_post_load, + .fields = (VMStateField[]) { + VMSTATE_PCI_DEVICE(d, ICH9LPCState), + VMSTATE_STRUCT(apm, ICH9LPCState, 0, vmstate_apm, APMState), + VMSTATE_STRUCT(pm, ICH9LPCState, 0, vmstate_ich9_pm, ICH9LPCPMRegs), + VMSTATE_UINT8_ARRAY(chip_config, ICH9LPCState, ICH9_CC_SIZE), + VMSTATE_UINT32(sci_level, ICH9LPCState), + VMSTATE_END_OF_LIST() + } +}; + +static void ich9_lpc_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + PCIDeviceClass *k = PCI_DEVICE_CLASS(klass); + + dc->reset = ich9_lpc_reset; + k->init = ich9_lpc_initfn; + dc->vmsd = &vmstate_ich9_lpc; + dc->no_user = 1; + k->config_write = ich9_lpc_config_write; + dc->desc = "ICH9 LPC bridge"; + k->vendor_id = PCI_VENDOR_ID_INTEL; + k->device_id = PCI_DEVICE_ID_INTEL_ICH9_8; + k->revision = ICH9_A2_LPC_REVISION; + k->class_id = PCI_CLASS_BRIDGE_ISA; + +} + +static const TypeInfo ich9_lpc_info = { + .name = TYPE_ICH9_LPC_DEVICE, + .parent = TYPE_PCI_DEVICE, + .instance_size = sizeof(struct ICH9LPCState), + .class_init = ich9_lpc_class_init, +}; + +static void ich9_lpc_register(void) +{ + type_register_static(&ich9_lpc_info); +} + +type_init(ich9_lpc_register); diff --git a/hw/mc146818rtc.c b/hw/mc146818rtc.c index 7d84ce3d74..c79fca7d68 100644 --- a/hw/mc146818rtc.c +++ b/hw/mc146818rtc.c @@ -570,7 +570,11 @@ static void rtc_update_time(RTCState *s) guest_nsec = get_guest_rtc_ns(s); guest_sec = guest_nsec / NSEC_PER_SEC; gmtime_r(&guest_sec, &ret); - rtc_set_cmos(s, &ret); + + /* Is SET flag of Register B disabled? */ + if ((s->cmos_data[RTC_REG_B] & REG_B_SET) == 0) { + rtc_set_cmos(s, &ret); + } } static int update_in_progress(RTCState *s) diff --git a/hw/pam.c b/hw/pam.c new file mode 100644 index 0000000000..a95e2cfb07 --- /dev/null +++ b/hw/pam.c @@ -0,0 +1,87 @@ +/* + * QEMU i440FX/PIIX3 PCI Bridge Emulation + * + * Copyright (c) 2006 Fabrice Bellard + * Copyright (c) 2011 Isaku Yamahata <yamahata at valinux co jp> + * VA Linux Systems Japan K.K. + * Copyright (c) 2012 Jason Baron <jbaron@redhat.com> + * + * Split out from piix_pci.c + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ +#include "sysemu.h" +#include "pam.h" + +void smram_update(MemoryRegion *smram_region, uint8_t smram, + uint8_t smm_enabled) +{ + bool smram_enabled; + + smram_enabled = ((smm_enabled && (smram & SMRAM_G_SMRAME)) || + (smram & SMRAM_D_OPEN)); + memory_region_set_enabled(smram_region, !smram_enabled); +} + +void smram_set_smm(uint8_t *host_smm_enabled, int smm, uint8_t smram, + MemoryRegion *smram_region) +{ + uint8_t smm_enabled = (smm != 0); + if (*host_smm_enabled != smm_enabled) { + *host_smm_enabled = smm_enabled; + smram_update(smram_region, smram, *host_smm_enabled); + } +} + +void init_pam(MemoryRegion *ram_memory, MemoryRegion *system_memory, + MemoryRegion *pci_address_space, PAMMemoryRegion *mem, + uint32_t start, uint32_t size) +{ + int i; + + /* RAM */ + memory_region_init_alias(&mem->alias[3], "pam-ram", ram_memory, + start, size); + /* ROM (XXX: not quite correct) */ + memory_region_init_alias(&mem->alias[1], "pam-rom", ram_memory, + start, size); + memory_region_set_readonly(&mem->alias[1], true); + + /* XXX: should distinguish read/write cases */ + memory_region_init_alias(&mem->alias[0], "pam-pci", pci_address_space, + start, size); + memory_region_init_alias(&mem->alias[2], "pam-pci", pci_address_space, + start, size); + + for (i = 0; i < 4; ++i) { + memory_region_set_enabled(&mem->alias[i], false); + memory_region_add_subregion_overlap(system_memory, start, + &mem->alias[i], 1); + } + mem->current = 0; +} + +void pam_update(PAMMemoryRegion *pam, int idx, uint8_t val) +{ + assert(0 <= idx && idx <= 12); + + memory_region_set_enabled(&pam->alias[pam->current], false); + pam->current = (val >> ((!(idx & 1)) * 4)) & PAM_ATTR_MASK; + memory_region_set_enabled(&pam->alias[pam->current], true); +} diff --git a/hw/pam.h b/hw/pam.h new file mode 100644 index 0000000000..2d77ebe0d9 --- /dev/null +++ b/hw/pam.h @@ -0,0 +1,97 @@ +#ifndef QEMU_PAM_H +#define QEMU_PAM_H + +/* + * Copyright (c) 2006 Fabrice Bellard + * Copyright (c) 2011 Isaku Yamahata <yamahata at valinux co jp> + * VA Linux Systems Japan K.K. + * Copyright (c) 2012 Jason Baron <jbaron@redhat.com> + * + * Split out from piix_pci.c + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ + +/* + * SMRAM memory area and PAM memory area in Legacy address range for PC. + * PAM: Programmable Attribute Map registers + * + * 0xa0000 - 0xbffff compatible SMRAM + * + * 0xc0000 - 0xc3fff Expansion area memory segments + * 0xc4000 - 0xc7fff + * 0xc8000 - 0xcbfff + * 0xcc000 - 0xcffff + * 0xd0000 - 0xd3fff + * 0xd4000 - 0xd7fff + * 0xd8000 - 0xdbfff + * 0xdc000 - 0xdffff + * 0xe0000 - 0xe3fff Extended System BIOS Area Memory Segments + * 0xe4000 - 0xe7fff + * 0xe8000 - 0xebfff + * 0xec000 - 0xeffff + * + * 0xf0000 - 0xfffff System BIOS Area Memory Segments + */ + +#include "qemu-common.h" +#include "memory.h" + +#define SMRAM_C_BASE 0xa0000 +#define SMRAM_C_END 0xc0000 +#define SMRAM_C_SIZE 0x20000 + +#define PAM_EXPAN_BASE 0xc0000 +#define PAM_EXPAN_SIZE 0x04000 + +#define PAM_EXBIOS_BASE 0xe0000 +#define PAM_EXBIOS_SIZE 0x04000 + +#define PAM_BIOS_BASE 0xf0000 +#define PAM_BIOS_END 0xfffff +/* 64KB: Intel 3 series express chipset family p. 58*/ +#define PAM_BIOS_SIZE 0x10000 + +/* PAM registers: log nibble and high nibble*/ +#define PAM_ATTR_WE ((uint8_t)2) +#define PAM_ATTR_RE ((uint8_t)1) +#define PAM_ATTR_MASK ((uint8_t)3) + +/* SMRAM register */ +#define SMRAM_D_OPEN ((uint8_t)(1 << 6)) +#define SMRAM_D_CLS ((uint8_t)(1 << 5)) +#define SMRAM_D_LCK ((uint8_t)(1 << 4)) +#define SMRAM_G_SMRAME ((uint8_t)(1 << 3)) +#define SMRAM_C_BASE_SEG_MASK ((uint8_t)0x7) +#define SMRAM_C_BASE_SEG ((uint8_t)0x2) /* hardwired to b010 */ + +typedef struct PAMMemoryRegion { + MemoryRegion alias[4]; /* index = PAM value */ + unsigned current; +} PAMMemoryRegion; + +void smram_update(MemoryRegion *smram_region, uint8_t smram, + uint8_t smm_enabled); +void smram_set_smm(uint8_t *host_smm_enabled, int smm, uint8_t smram, + MemoryRegion *smram_region); +void init_pam(MemoryRegion *ram, MemoryRegion *system, MemoryRegion *pci, + PAMMemoryRegion *mem, uint32_t start, uint32_t size); +void pam_update(PAMMemoryRegion *mem, int idx, uint8_t val); + +#endif /* QEMU_PAM_H */ @@ -1058,6 +1058,21 @@ void pc_basic_device_init(ISABus *isa_bus, qemu_irq *gsi, *floppy = fdctrl_init_isa(isa_bus, fd); } +void pc_nic_init(ISABus *isa_bus, PCIBus *pci_bus) +{ + int i; + + for (i = 0; i < nb_nics; i++) { + NICInfo *nd = &nd_table[i]; + + if (!pci_bus || (nd->model && strcmp(nd->model, "ne2k_isa") == 0)) { + pc_init_ne2k_isa(isa_bus, nd); + } else { + pci_nic_init_nofail(nd, "e1000", NULL); + } + } +} + void pc_pci_device_init(PCIBus *pci_bus) { int max_bus; @@ -1068,3 +1083,27 @@ void pc_pci_device_init(PCIBus *pci_bus) pci_create_simple(pci_bus, -1, "lsi53c895a"); } } + +void ioapic_init_gsi(GSIState *gsi_state, const char *parent_name) +{ + DeviceState *dev; + SysBusDevice *d; + unsigned int i; + + if (kvm_irqchip_in_kernel()) { + dev = qdev_create(NULL, "kvm-ioapic"); + } else { + dev = qdev_create(NULL, "ioapic"); + } + if (parent_name) { + object_property_add_child(object_resolve_path(parent_name, NULL), + "ioapic", OBJECT(dev), NULL); + } + qdev_init_nofail(dev); + d = sysbus_from_qdev(dev); + sysbus_mmio_map(d, 0, 0xfec00000); + + for (i = 0; i < IOAPIC_NUM_PINS; i++) { + gsi_state->ioapic_irq[i] = qdev_get_gpio_in(dev, i); + } +} @@ -98,11 +98,14 @@ void pc_cmos_init(ram_addr_t ram_size, ram_addr_t above_4g_mem_size, const char *boot_device, ISADevice *floppy, BusState *ide0, BusState *ide1, ISADevice *s); +void pc_nic_init(ISABus *isa_bus, PCIBus *pci_bus); void pc_pci_device_init(PCIBus *pci_bus); typedef void (*cpu_set_smm_t)(int smm, void *arg); void cpu_smm_register(cpu_set_smm_t callback, void *arg); +void ioapic_init_gsi(GSIState *gsi_state, const char *parent_name); + /* acpi.c */ extern int acpi_enabled; extern char *acpi_tables; diff --git a/hw/pc_piix.c b/hw/pc_piix.c index cfa839c8b4..aa3e7f40dc 100644 --- a/hw/pc_piix.c +++ b/hw/pc_piix.c @@ -54,70 +54,6 @@ static const int ide_iobase[MAX_IDE_BUS] = { 0x1f0, 0x170 }; static const int ide_iobase2[MAX_IDE_BUS] = { 0x3f6, 0x376 }; static const int ide_irq[MAX_IDE_BUS] = { 14, 15 }; -static void kvm_piix3_setup_irq_routing(bool pci_enabled) -{ -#ifdef CONFIG_KVM - KVMState *s = kvm_state; - int i; - - if (kvm_check_extension(s, KVM_CAP_IRQ_ROUTING)) { - for (i = 0; i < 8; ++i) { - if (i == 2) { - continue; - } - kvm_irqchip_add_irq_route(s, i, KVM_IRQCHIP_PIC_MASTER, i); - } - for (i = 8; i < 16; ++i) { - kvm_irqchip_add_irq_route(s, i, KVM_IRQCHIP_PIC_SLAVE, i - 8); - } - if (pci_enabled) { - for (i = 0; i < 24; ++i) { - if (i == 0) { - kvm_irqchip_add_irq_route(s, i, KVM_IRQCHIP_IOAPIC, 2); - } else if (i != 2) { - kvm_irqchip_add_irq_route(s, i, KVM_IRQCHIP_IOAPIC, i); - } - } - } - } -#endif /* CONFIG_KVM */ -} - -static void kvm_piix3_gsi_handler(void *opaque, int n, int level) -{ - GSIState *s = opaque; - - if (n < ISA_NUM_IRQS) { - /* Kernel will forward to both PIC and IOAPIC */ - qemu_set_irq(s->i8259_irq[n], level); - } else { - qemu_set_irq(s->ioapic_irq[n], level); - } -} - -static void ioapic_init(GSIState *gsi_state) -{ - DeviceState *dev; - SysBusDevice *d; - unsigned int i; - - if (kvm_irqchip_in_kernel()) { - dev = qdev_create(NULL, "kvm-ioapic"); - } else { - dev = qdev_create(NULL, "ioapic"); - } - /* FIXME: this should be under the piix3. */ - object_property_add_child(object_resolve_path("i440fx", NULL), - "ioapic", OBJECT(dev), NULL); - qdev_init_nofail(dev); - d = sysbus_from_qdev(dev); - sysbus_mmio_map(d, 0, 0xfec00000); - - for (i = 0; i < IOAPIC_NUM_PINS; i++) { - gsi_state->ioapic_irq[i] = qdev_get_gpio_in(dev, i); - } -} - /* PC hardware initialisation */ static void pc_init1(MemoryRegion *system_memory, MemoryRegion *system_io, @@ -183,8 +119,8 @@ static void pc_init1(MemoryRegion *system_memory, gsi_state = g_malloc0(sizeof(*gsi_state)); if (kvm_irqchip_in_kernel()) { - kvm_piix3_setup_irq_routing(pci_enabled); - gsi = qemu_allocate_irqs(kvm_piix3_gsi_handler, gsi_state, + kvm_pc_setup_irq_routing(pci_enabled); + gsi = qemu_allocate_irqs(kvm_pc_gsi_handler, gsi_state, GSI_NUM_PINS); } else { gsi = qemu_allocate_irqs(gsi_handler, gsi_state, GSI_NUM_PINS); @@ -221,7 +157,7 @@ static void pc_init1(MemoryRegion *system_memory, gsi_state->i8259_irq[i] = i8259[i]; } if (pci_enabled) { - ioapic_init(gsi_state); + ioapic_init_gsi(gsi_state, "i440fx"); } pc_register_ferr_irq(gsi[13]); @@ -234,14 +170,7 @@ static void pc_init1(MemoryRegion *system_memory, /* init basic PC hardware */ pc_basic_device_init(isa_bus, gsi, &rtc_state, &floppy, xen_enabled()); - for(i = 0; i < nb_nics; i++) { - NICInfo *nd = &nd_table[i]; - - if (!pci_enabled || (nd->model && strcmp(nd->model, "ne2k_isa") == 0)) - pc_init_ne2k_isa(isa_bus, nd); - else - pci_nic_init_nofail(nd, "e1000", NULL); - } + pc_nic_init(isa_bus, pci_bus); ide_drive_get(hd, MAX_IDE_BUS); if (pci_enabled) { diff --git a/hw/pc_q35.c b/hw/pc_q35.c new file mode 100644 index 0000000000..3429a9ae8f --- /dev/null +++ b/hw/pc_q35.c @@ -0,0 +1,223 @@ +/* + * Q35 chipset based pc system emulator + * + * Copyright (c) 2003-2004 Fabrice Bellard + * Copyright (c) 2009, 2010 + * Isaku Yamahata <yamahata at valinux co jp> + * VA Linux Systems Japan K.K. + * Copyright (C) 2012 Jason Baron <jbaron@redhat.com> + * + * This is based on pc.c, but heavily modified. + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ +#include "hw.h" +#include "arch_init.h" +#include "smbus.h" +#include "boards.h" +#include "mc146818rtc.h" +#include "xen.h" +#include "kvm.h" +#include "kvm/clock.h" +#include "q35.h" +#include "exec-memory.h" +#include "ich9.h" +#include "hw/ide/pci.h" +#include "hw/ide/ahci.h" +#include "hw/usb.h" + +/* ICH9 AHCI has 6 ports */ +#define MAX_SATA_PORTS 6 + +/* set CMOS shutdown status register (index 0xF) as S3_resume(0xFE) + * BIOS will read it and start S3 resume at POST Entry */ +static void pc_cmos_set_s3_resume(void *opaque, int irq, int level) +{ + ISADevice *s = opaque; + + if (level) { + rtc_set_memory(s, 0xF, 0xFE); + } +} + +/* PC hardware initialisation */ +static void pc_q35_init(QEMUMachineInitArgs *args) +{ + ram_addr_t ram_size = args->ram_size; + const char *cpu_model = args->cpu_model; + const char *kernel_filename = args->kernel_filename; + const char *kernel_cmdline = args->kernel_cmdline; + const char *initrd_filename = args->initrd_filename; + const char *boot_device = args->boot_device; + ram_addr_t below_4g_mem_size, above_4g_mem_size; + Q35PCIHost *q35_host; + PCIBus *host_bus; + PCIDevice *lpc; + BusState *idebus[MAX_SATA_PORTS]; + ISADevice *rtc_state; + ISADevice *floppy; + MemoryRegion *pci_memory; + MemoryRegion *rom_memory; + MemoryRegion *ram_memory; + GSIState *gsi_state; + ISABus *isa_bus; + int pci_enabled = 1; + qemu_irq *cpu_irq; + qemu_irq *gsi; + qemu_irq *i8259; + int i; + ICH9LPCState *ich9_lpc; + PCIDevice *ahci; + qemu_irq *cmos_s3; + + pc_cpus_init(cpu_model); + + kvmclock_create(); + + if (ram_size >= 0xb0000000) { + above_4g_mem_size = ram_size - 0xb0000000; + below_4g_mem_size = 0xb0000000; + } else { + above_4g_mem_size = 0; + below_4g_mem_size = ram_size; + } + + /* pci enabled */ + if (pci_enabled) { + pci_memory = g_new(MemoryRegion, 1); + memory_region_init(pci_memory, "pci", INT64_MAX); + rom_memory = pci_memory; + } else { + pci_memory = NULL; + rom_memory = get_system_memory(); + } + + /* allocate ram and load rom/bios */ + if (!xen_enabled()) { + pc_memory_init(get_system_memory(), kernel_filename, kernel_cmdline, + initrd_filename, below_4g_mem_size, above_4g_mem_size, + rom_memory, &ram_memory); + } + + /* irq lines */ + gsi_state = g_malloc0(sizeof(*gsi_state)); + if (kvm_irqchip_in_kernel()) { + kvm_pc_setup_irq_routing(pci_enabled); + gsi = qemu_allocate_irqs(kvm_pc_gsi_handler, gsi_state, + GSI_NUM_PINS); + } else { + gsi = qemu_allocate_irqs(gsi_handler, gsi_state, GSI_NUM_PINS); + } + + /* create pci host bus */ + q35_host = Q35_HOST_DEVICE(qdev_create(NULL, TYPE_Q35_HOST_DEVICE)); + + q35_host->mch.ram_memory = ram_memory; + q35_host->mch.pci_address_space = pci_memory; + q35_host->mch.system_memory = get_system_memory(); + q35_host->mch.address_space_io = get_system_io();; + q35_host->mch.below_4g_mem_size = below_4g_mem_size; + q35_host->mch.above_4g_mem_size = above_4g_mem_size; + /* pci */ + qdev_init_nofail(DEVICE(q35_host)); + host_bus = q35_host->host.pci.bus; + /* create ISA bus */ + lpc = pci_create_simple_multifunction(host_bus, PCI_DEVFN(ICH9_LPC_DEV, + ICH9_LPC_FUNC), true, + TYPE_ICH9_LPC_DEVICE); + ich9_lpc = ICH9_LPC_DEVICE(lpc); + ich9_lpc->pic = gsi; + ich9_lpc->ioapic = gsi_state->ioapic_irq; + pci_bus_irqs(host_bus, ich9_lpc_set_irq, ich9_lpc_map_irq, ich9_lpc, + ICH9_LPC_NB_PIRQS); + isa_bus = ich9_lpc->isa_bus; + + /*end early*/ + isa_bus_irqs(isa_bus, gsi); + + if (kvm_irqchip_in_kernel()) { + i8259 = kvm_i8259_init(isa_bus); + } else if (xen_enabled()) { + i8259 = xen_interrupt_controller_init(); + } else { + cpu_irq = pc_allocate_cpu_irq(); + i8259 = i8259_init(isa_bus, cpu_irq[0]); + } + + for (i = 0; i < ISA_NUM_IRQS; i++) { + gsi_state->i8259_irq[i] = i8259[i]; + } + if (pci_enabled) { + ioapic_init_gsi(gsi_state, NULL); + } + + pc_register_ferr_irq(gsi[13]); + + /* init basic PC hardware */ + pc_basic_device_init(isa_bus, gsi, &rtc_state, &floppy, false); + + /* connect pm stuff to lpc */ + cmos_s3 = qemu_allocate_irqs(pc_cmos_set_s3_resume, rtc_state, 1); + ich9_lpc_pm_init(lpc, *cmos_s3); + + /* ahci and SATA device, for q35 1 ahci controller is built-in */ + ahci = pci_create_simple_multifunction(host_bus, + PCI_DEVFN(ICH9_SATA1_DEV, + ICH9_SATA1_FUNC), + true, "ich9-ahci"); + idebus[0] = qdev_get_child_bus(&ahci->qdev, "ide.0"); + idebus[1] = qdev_get_child_bus(&ahci->qdev, "ide.1"); + + if (usb_enabled(false)) { + /* Should we create 6 UHCI according to ich9 spec? */ + ehci_create_ich9_with_companions(host_bus, 0x1d); + } + + /* TODO: Populate SPD eeprom data. */ + smbus_eeprom_init(ich9_smb_init(host_bus, + PCI_DEVFN(ICH9_SMB_DEV, ICH9_SMB_FUNC), + 0xb100), + 8, NULL, 0); + + pc_cmos_init(below_4g_mem_size, above_4g_mem_size, boot_device, + floppy, idebus[0], idebus[1], rtc_state); + + /* the rest devices to which pci devfn is automatically assigned */ + pc_vga_init(isa_bus, host_bus); + audio_init(isa_bus, host_bus); + pc_nic_init(isa_bus, host_bus); + if (pci_enabled) { + pc_pci_device_init(host_bus); + } +} + +static QEMUMachine pc_q35_machine = { + .name = "q35-next", + .alias = "q35", + .desc = "Q35 chipset PC", + .init = pc_q35_init, + .max_cpus = 255, +}; + +static void pc_q35_machine_init(void) +{ + qemu_register_machine(&pc_q35_machine); +} + +machine_init(pc_q35_machine_init); diff --git a/hw/pci-hotplug.c b/hw/pci-hotplug.c index e7fb780a08..0ca5546fc6 100644 --- a/hw/pci-hotplug.c +++ b/hw/pci-hotplug.c @@ -80,7 +80,13 @@ static int scsi_hot_add(Monitor *mon, DeviceState *adapter, SCSIBus *scsibus; SCSIDevice *scsidev; - scsibus = SCSI_BUS(QLIST_FIRST(&adapter->child_bus)); + scsibus = (SCSIBus *) + object_dynamic_cast(OBJECT(QLIST_FIRST(&adapter->child_bus)), + TYPE_SCSI_BUS); + if (!scsibus) { + error_report("Device is not a SCSI adapter"); + return -1; + } /* * drive_init() tries to find a default for dinfo->unit. Doesn't @@ -301,9 +301,9 @@ PCIBus *pci_bus_new(DeviceState *parent, const char *name, PCIBus *bus; bus = g_malloc0(sizeof(*bus)); - bus->qbus.glib_allocated = true; pci_bus_new_inplace(bus, parent, name, address_space_mem, address_space_io, devfn_min); + OBJECT(bus)->free = g_free; return bus; } diff --git a/hw/pci_ids.h b/hw/pci_ids.h index 41f3570fb9..5df7245349 100644 --- a/hw/pci_ids.h +++ b/hw/pci_ids.h @@ -36,6 +36,7 @@ #define PCI_CLASS_BRIDGE_HOST 0x0600 #define PCI_CLASS_BRIDGE_ISA 0x0601 #define PCI_CLASS_BRIDGE_PCI 0x0604 +#define PCI_CLASS_BRDIGE_PCI_INF_SUB 0x01 #define PCI_CLASS_BRIDGE_OTHER 0x0680 #define PCI_CLASS_COMMUNICATION_SERIAL 0x0700 @@ -116,6 +117,17 @@ #define PCI_DEVICE_ID_INTEL_82371AB 0x7111 #define PCI_DEVICE_ID_INTEL_82371AB_2 0x7112 #define PCI_DEVICE_ID_INTEL_82371AB_3 0x7113 + +#define PCI_DEVICE_ID_INTEL_ICH9_0 0x2910 +#define PCI_DEVICE_ID_INTEL_ICH9_1 0x2917 +#define PCI_DEVICE_ID_INTEL_ICH9_2 0x2912 +#define PCI_DEVICE_ID_INTEL_ICH9_3 0x2913 +#define PCI_DEVICE_ID_INTEL_ICH9_4 0x2914 +#define PCI_DEVICE_ID_INTEL_ICH9_5 0x2919 +#define PCI_DEVICE_ID_INTEL_ICH9_6 0x2930 +#define PCI_DEVICE_ID_INTEL_ICH9_7 0x2916 +#define PCI_DEVICE_ID_INTEL_ICH9_8 0x2918 + #define PCI_DEVICE_ID_INTEL_82801I_UHCI1 0x2934 #define PCI_DEVICE_ID_INTEL_82801I_UHCI2 0x2935 #define PCI_DEVICE_ID_INTEL_82801I_UHCI3 0x2936 @@ -126,6 +138,8 @@ #define PCI_DEVICE_ID_INTEL_82801I_EHCI2 0x293c #define PCI_DEVICE_ID_INTEL_82599_SFP_VF 0x10ed +#define PCI_DEVICE_ID_INTEL_Q35_MCH 0x29c0 + #define PCI_VENDOR_ID_XEN 0x5853 #define PCI_DEVICE_ID_XEN_PLATFORM 0x0001 diff --git a/hw/piix_pci.c b/hw/piix_pci.c index 9af5847690..ba1b3de749 100644 --- a/hw/piix_pci.c +++ b/hw/piix_pci.c @@ -30,6 +30,7 @@ #include "sysbus.h" #include "range.h" #include "xen.h" +#include "pam.h" /* * I440FX chipset data sheet. @@ -68,11 +69,6 @@ typedef struct PIIX3State { int32_t pci_irq_levels_vmstate[PIIX_NUM_PIRQS]; } PIIX3State; -typedef struct PAMMemoryRegion { - MemoryRegion alias[4]; /* index = PAM value */ - unsigned current; -} PAMMemoryRegion; - struct PCII440FXState { PCIDevice dev; MemoryRegion *system_memory; @@ -105,52 +101,16 @@ static int pci_slot_get_pirq(PCIDevice *pci_dev, int pci_intx) return (pci_intx + slot_addend) & 3; } -static void init_pam(PCII440FXState *d, PAMMemoryRegion *mem, - uint32_t start, uint32_t size) -{ - int i; - - /* RAM */ - memory_region_init_alias(&mem->alias[3], "pam-ram", d->ram_memory, start, size); - /* ROM (XXX: not quite correct) */ - memory_region_init_alias(&mem->alias[1], "pam-rom", d->ram_memory, start, size); - memory_region_set_readonly(&mem->alias[1], true); - - /* XXX: should distinguish read/write cases */ - memory_region_init_alias(&mem->alias[0], "pam-pci", d->pci_address_space, - start, size); - memory_region_init_alias(&mem->alias[2], "pam-pci", d->pci_address_space, - start, size); - - for (i = 0; i < 4; ++i) { - memory_region_set_enabled(&mem->alias[i], false); - memory_region_add_subregion_overlap(d->system_memory, start, &mem->alias[i], 1); - } - mem->current = 0; -} - -static void update_pam(PAMMemoryRegion *pam, unsigned r) -{ - memory_region_set_enabled(&pam->alias[pam->current], false); - pam->current = r; - memory_region_set_enabled(&pam->alias[pam->current], true); -} - static void i440fx_update_memory_mappings(PCII440FXState *d) { - int i, r; - uint32_t smram; - bool smram_enabled; + int i; memory_region_transaction_begin(); - update_pam(&d->pam_regions[0], (d->dev.config[I440FX_PAM] >> 4) & 3); - for(i = 0; i < 12; i++) { - r = (d->dev.config[(i >> 1) + (I440FX_PAM + 1)] >> ((i & 1) * 4)) & 3; - update_pam(&d->pam_regions[i+1], r); + for (i = 0; i < 13; i++) { + pam_update(&d->pam_regions[i], i, + d->dev.config[I440FX_PAM + ((i + 1) / 2)]); } - smram = d->dev.config[I440FX_SMRAM]; - smram_enabled = (d->smm_enabled && (smram & 0x08)) || (smram & 0x40); - memory_region_set_enabled(&d->smram_region, !smram_enabled); + smram_update(&d->smram_region, d->dev.config[I440FX_SMRAM], d->smm_enabled); memory_region_transaction_commit(); } @@ -158,11 +118,10 @@ static void i440fx_set_smm(int val, void *arg) { PCII440FXState *d = arg; - val = (val != 0); - if (d->smm_enabled != val) { - d->smm_enabled = val; - i440fx_update_memory_mappings(d); - } + memory_region_transaction_begin(); + smram_set_smm(&d->smm_enabled, val, d->dev.config[I440FX_SMRAM], + &d->smram_region); + memory_region_transaction_commit(); } @@ -300,9 +259,12 @@ static PCIBus *i440fx_common_init(const char *device_name, memory_region_add_subregion_overlap(f->system_memory, 0xa0000, &f->smram_region, 1); memory_region_set_enabled(&f->smram_region, false); - init_pam(f, &f->pam_regions[0], 0xf0000, 0x10000); + init_pam(f->ram_memory, f->system_memory, f->pci_address_space, + &f->pam_regions[0], PAM_BIOS_BASE, PAM_BIOS_SIZE); for (i = 0; i < 12; ++i) { - init_pam(f, &f->pam_regions[i+1], 0xc0000 + i * 0x4000, 0x4000); + init_pam(f->ram_memory, f->system_memory, f->pci_address_space, + &f->pam_regions[i+1], PAM_EXPAN_BASE + i * PAM_EXPAN_SIZE, + PAM_EXPAN_SIZE); } /* Xen supports additional interrupt routes from the PCI devices to diff --git a/hw/q35.c b/hw/q35.c new file mode 100644 index 0000000000..efebc2786a --- /dev/null +++ b/hw/q35.c @@ -0,0 +1,309 @@ +/* + * QEMU MCH/ICH9 PCI Bridge Emulation + * + * Copyright (c) 2006 Fabrice Bellard + * Copyright (c) 2009, 2010, 2011 + * Isaku Yamahata <yamahata at valinux co jp> + * VA Linux Systems Japan K.K. + * Copyright (C) 2012 Jason Baron <jbaron@redhat.com> + * + * This is based on piix_pci.c, but heavily modified. + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ +#include "hw.h" +#include "q35.h" + +/**************************************************************************** + * Q35 host + */ + +static int q35_host_init(SysBusDevice *dev) +{ + PCIBus *b; + PCIHostState *pci = FROM_SYSBUS(PCIHostState, dev); + Q35PCIHost *s = Q35_HOST_DEVICE(&dev->qdev); + + memory_region_init_io(&pci->conf_mem, &pci_host_conf_le_ops, pci, + "pci-conf-idx", 4); + sysbus_add_io(dev, MCH_HOST_BRIDGE_CONFIG_ADDR, &pci->conf_mem); + sysbus_init_ioports(&pci->busdev, MCH_HOST_BRIDGE_CONFIG_ADDR, 4); + + memory_region_init_io(&pci->data_mem, &pci_host_data_le_ops, pci, + "pci-conf-data", 4); + sysbus_add_io(dev, MCH_HOST_BRIDGE_CONFIG_DATA, &pci->data_mem); + sysbus_init_ioports(&pci->busdev, MCH_HOST_BRIDGE_CONFIG_DATA, 4); + + if (pcie_host_init(&s->host) < 0) { + return -1; + } + b = pci_bus_new(&s->host.pci.busdev.qdev, "pcie.0", + s->mch.pci_address_space, s->mch.address_space_io, 0); + s->host.pci.bus = b; + qdev_set_parent_bus(DEVICE(&s->mch), BUS(b)); + qdev_init_nofail(DEVICE(&s->mch)); + + return 0; +} + +static Property mch_props[] = { + DEFINE_PROP_UINT64("MCFG", Q35PCIHost, host.base_addr, + MCH_HOST_BRIDGE_PCIEXBAR_DEFAULT), + DEFINE_PROP_END_OF_LIST(), +}; + +static void q35_host_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + SysBusDeviceClass *k = SYS_BUS_DEVICE_CLASS(klass); + + k->init = q35_host_init; + dc->props = mch_props; +} + +static void q35_host_initfn(Object *obj) +{ + Q35PCIHost *s = Q35_HOST_DEVICE(obj); + + object_initialize(&s->mch, TYPE_MCH_PCI_DEVICE); + object_property_add_child(OBJECT(s), "mch", OBJECT(&s->mch), NULL); + qdev_prop_set_uint32(DEVICE(&s->mch), "addr", PCI_DEVFN(0, 0)); + qdev_prop_set_bit(DEVICE(&s->mch), "multifunction", false); +} + +static const TypeInfo q35_host_info = { + .name = TYPE_Q35_HOST_DEVICE, + .parent = TYPE_PCIE_HOST_BRIDGE, + .instance_size = sizeof(Q35PCIHost), + .instance_init = q35_host_initfn, + .class_init = q35_host_class_init, +}; + +/**************************************************************************** + * MCH D0:F0 + */ + +/* PCIe MMCFG */ +static void mch_update_pciexbar(MCHPCIState *mch) +{ + PCIDevice *pci_dev = &mch->d; + BusState *bus = qdev_get_parent_bus(&pci_dev->qdev); + DeviceState *qdev = bus->parent; + Q35PCIHost *s = Q35_HOST_DEVICE(qdev); + + uint64_t pciexbar; + int enable; + uint64_t addr; + uint64_t addr_mask; + uint32_t length; + + pciexbar = pci_get_quad(pci_dev->config + MCH_HOST_BRIDGE_PCIEXBAR); + enable = pciexbar & MCH_HOST_BRIDGE_PCIEXBAREN; + addr_mask = MCH_HOST_BRIDGE_PCIEXBAR_ADMSK; + switch (pciexbar & MCH_HOST_BRIDGE_PCIEXBAR_LENGTH_MASK) { + case MCH_HOST_BRIDGE_PCIEXBAR_LENGTH_256M: + length = 256 * 1024 * 1024; + break; + case MCH_HOST_BRIDGE_PCIEXBAR_LENGTH_128M: + length = 128 * 1024 * 1024; + addr_mask |= MCH_HOST_BRIDGE_PCIEXBAR_128ADMSK | + MCH_HOST_BRIDGE_PCIEXBAR_64ADMSK; + break; + case MCH_HOST_BRIDGE_PCIEXBAR_LENGTH_64M: + length = 64 * 1024 * 1024; + addr_mask |= MCH_HOST_BRIDGE_PCIEXBAR_64ADMSK; + break; + case MCH_HOST_BRIDGE_PCIEXBAR_LENGTH_RVD: + default: + enable = 0; + length = 0; + abort(); + break; + } + addr = pciexbar & addr_mask; + pcie_host_mmcfg_update(&s->host, enable, addr, length); +} + +/* PAM */ +static void mch_update_pam(MCHPCIState *mch) +{ + int i; + + memory_region_transaction_begin(); + for (i = 0; i < 13; i++) { + pam_update(&mch->pam_regions[i], i, + mch->d.config[MCH_HOST_BRIDGE_PAM0 + ((i + 1) / 2)]); + } + memory_region_transaction_commit(); +} + +/* SMRAM */ +static void mch_update_smram(MCHPCIState *mch) +{ + memory_region_transaction_begin(); + smram_update(&mch->smram_region, mch->d.config[MCH_HOST_BRDIGE_SMRAM], + mch->smm_enabled); + memory_region_transaction_commit(); +} + +static void mch_set_smm(int smm, void *arg) +{ + MCHPCIState *mch = arg; + + memory_region_transaction_begin(); + smram_set_smm(&mch->smm_enabled, smm, mch->d.config[MCH_HOST_BRDIGE_SMRAM], + &mch->smram_region); + memory_region_transaction_commit(); +} + +static void mch_write_config(PCIDevice *d, + uint32_t address, uint32_t val, int len) +{ + MCHPCIState *mch = MCH_PCI_DEVICE(d); + + /* XXX: implement SMRAM.D_LOCK */ + pci_default_write_config(d, address, val, len); + + if (ranges_overlap(address, len, MCH_HOST_BRIDGE_PAM0, + MCH_HOST_BRIDGE_PAM_SIZE)) { + mch_update_pam(mch); + } + + if (ranges_overlap(address, len, MCH_HOST_BRIDGE_PCIEXBAR, + MCH_HOST_BRIDGE_PCIEXBAR_SIZE)) { + mch_update_pciexbar(mch); + } + + if (ranges_overlap(address, len, MCH_HOST_BRDIGE_SMRAM, + MCH_HOST_BRDIGE_SMRAM_SIZE)) { + mch_update_smram(mch); + } +} + +static void mch_update(MCHPCIState *mch) +{ + mch_update_pciexbar(mch); + mch_update_pam(mch); + mch_update_smram(mch); +} + +static int mch_post_load(void *opaque, int version_id) +{ + MCHPCIState *mch = opaque; + mch_update(mch); + return 0; +} + +static const VMStateDescription vmstate_mch = { + .name = "mch", + .version_id = 1, + .minimum_version_id = 1, + .minimum_version_id_old = 1, + .post_load = mch_post_load, + .fields = (VMStateField []) { + VMSTATE_PCI_DEVICE(d, MCHPCIState), + VMSTATE_UINT8(smm_enabled, MCHPCIState), + VMSTATE_END_OF_LIST() + } +}; + +static void mch_reset(DeviceState *qdev) +{ + PCIDevice *d = PCI_DEVICE(qdev); + MCHPCIState *mch = MCH_PCI_DEVICE(d); + + pci_set_quad(d->config + MCH_HOST_BRIDGE_PCIEXBAR, + MCH_HOST_BRIDGE_PCIEXBAR_DEFAULT); + + d->config[MCH_HOST_BRDIGE_SMRAM] = MCH_HOST_BRIDGE_SMRAM_DEFAULT; + + mch_update(mch); +} + +static int mch_init(PCIDevice *d) +{ + int i; + hwaddr pci_hole64_size; + MCHPCIState *mch = MCH_PCI_DEVICE(d); + + /* setup pci memory regions */ + memory_region_init_alias(&mch->pci_hole, "pci-hole", + mch->pci_address_space, + mch->below_4g_mem_size, + 0x100000000ULL - mch->below_4g_mem_size); + memory_region_add_subregion(mch->system_memory, mch->below_4g_mem_size, + &mch->pci_hole); + pci_hole64_size = (sizeof(hwaddr) == 4 ? 0 : + ((uint64_t)1 << 62)); + memory_region_init_alias(&mch->pci_hole_64bit, "pci-hole64", + mch->pci_address_space, + 0x100000000ULL + mch->above_4g_mem_size, + pci_hole64_size); + if (pci_hole64_size) { + memory_region_add_subregion(mch->system_memory, + 0x100000000ULL + mch->above_4g_mem_size, + &mch->pci_hole_64bit); + } + /* smram */ + cpu_smm_register(&mch_set_smm, mch); + memory_region_init_alias(&mch->smram_region, "smram-region", + mch->pci_address_space, 0xa0000, 0x20000); + memory_region_add_subregion_overlap(mch->system_memory, 0xa0000, + &mch->smram_region, 1); + memory_region_set_enabled(&mch->smram_region, false); + init_pam(mch->ram_memory, mch->system_memory, mch->pci_address_space, + &mch->pam_regions[0], PAM_BIOS_BASE, PAM_BIOS_SIZE); + for (i = 0; i < 12; ++i) { + init_pam(mch->ram_memory, mch->system_memory, mch->pci_address_space, + &mch->pam_regions[i+1], PAM_EXPAN_BASE + i * PAM_EXPAN_SIZE, + PAM_EXPAN_SIZE); + } + return 0; +} + +static void mch_class_init(ObjectClass *klass, void *data) +{ + PCIDeviceClass *k = PCI_DEVICE_CLASS(klass); + DeviceClass *dc = DEVICE_CLASS(klass); + + k->init = mch_init; + k->config_write = mch_write_config; + dc->reset = mch_reset; + dc->desc = "Host bridge"; + dc->vmsd = &vmstate_mch; + k->vendor_id = PCI_VENDOR_ID_INTEL; + k->device_id = PCI_DEVICE_ID_INTEL_Q35_MCH; + k->revision = MCH_HOST_BRIDGE_REVISION_DEFUALT; + k->class_id = PCI_CLASS_BRIDGE_HOST; +} + +static const TypeInfo mch_info = { + .name = TYPE_MCH_PCI_DEVICE, + .parent = TYPE_PCI_DEVICE, + .instance_size = sizeof(MCHPCIState), + .class_init = mch_class_init, +}; + +static void q35_register(void) +{ + type_register_static(&mch_info); + type_register_static(&q35_host_info); +} + +type_init(q35_register); diff --git a/hw/q35.h b/hw/q35.h new file mode 100644 index 0000000000..e34f7c165f --- /dev/null +++ b/hw/q35.h @@ -0,0 +1,150 @@ +/* + * q35.h + * + * Copyright (c) 2009 Isaku Yamahata <yamahata at valinux co jp> + * VA Linux Systems Japan K.K. + * Copyright (C) 2012 Jason Baron <jbaron@redhat.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 Lesser General Public + * License along with this library; if not, see <http://www.gnu.org/licenses/> + */ + +#ifndef HW_Q35_H +#define HW_Q35_H + +#include "hw.h" +#include "range.h" +#include "isa.h" +#include "sysbus.h" +#include "pc.h" +#include "apm.h" +#include "apic.h" +#include "pci.h" +#include "pcie_host.h" +#include "acpi.h" +#include "acpi_ich9.h" +#include "pam.h" + +#define TYPE_Q35_HOST_DEVICE "q35-pcihost" +#define Q35_HOST_DEVICE(obj) \ + OBJECT_CHECK(Q35PCIHost, (obj), TYPE_Q35_HOST_DEVICE) + +#define TYPE_MCH_PCI_DEVICE "mch" +#define MCH_PCI_DEVICE(obj) \ + OBJECT_CHECK(MCHPCIState, (obj), TYPE_MCH_PCI_DEVICE) + +typedef struct MCHPCIState { + PCIDevice d; + MemoryRegion *ram_memory; + MemoryRegion *pci_address_space; + MemoryRegion *system_memory; + MemoryRegion *address_space_io; + PAMMemoryRegion pam_regions[13]; + MemoryRegion smram_region; + MemoryRegion pci_hole; + MemoryRegion pci_hole_64bit; + uint8_t smm_enabled; + ram_addr_t below_4g_mem_size; + ram_addr_t above_4g_mem_size; +} MCHPCIState; + +typedef struct Q35PCIHost { + PCIExpressHost host; + MCHPCIState mch; +} Q35PCIHost; + +#define Q35_MASK(bit, ms_bit, ls_bit) \ +((uint##bit##_t)(((1ULL << ((ms_bit) + 1)) - 1) & ~((1ULL << ls_bit) - 1))) + +/* + * gmch part + */ + +/* PCI configuration */ +#define MCH_HOST_BRIDGE "MCH" + +#define MCH_HOST_BRIDGE_CONFIG_ADDR 0xcf8 +#define MCH_HOST_BRIDGE_CONFIG_DATA 0xcfc + +/* D0:F0 configuration space */ +#define MCH_HOST_BRIDGE_REVISION_DEFUALT 0x0 + +#define MCH_HOST_BRIDGE_PCIEXBAR 0x60 /* 64bit register */ +#define MCH_HOST_BRIDGE_PCIEXBAR_SIZE 8 /* 64bit register */ +#define MCH_HOST_BRIDGE_PCIEXBAR_DEFAULT 0xb0000000 +#define MCH_HOST_BRIDGE_PCIEXBAR_ADMSK Q35_MASK(64, 35, 28) +#define MCH_HOST_BRIDGE_PCIEXBAR_128ADMSK ((uint64_t)(1 << 26)) +#define MCH_HOST_BRIDGE_PCIEXBAR_64ADMSK ((uint64_t)(1 << 25)) +#define MCH_HOST_BRIDGE_PCIEXBAR_LENGTH_MASK ((uint64_t)(0x3 << 1)) +#define MCH_HOST_BRIDGE_PCIEXBAR_LENGTH_256M ((uint64_t)(0x0 << 1)) +#define MCH_HOST_BRIDGE_PCIEXBAR_LENGTH_128M ((uint64_t)(0x1 << 1)) +#define MCH_HOST_BRIDGE_PCIEXBAR_LENGTH_64M ((uint64_t)(0x2 << 1)) +#define MCH_HOST_BRIDGE_PCIEXBAR_LENGTH_RVD ((uint64_t)(0x3 << 1)) +#define MCH_HOST_BRIDGE_PCIEXBAREN ((uint64_t)1) + +#define MCH_HOST_BRIDGE_PAM_NB 7 +#define MCH_HOST_BRIDGE_PAM_SIZE 7 +#define MCH_HOST_BRIDGE_PAM0 0x90 +#define MCH_HOST_BRIDGE_PAM_BIOS_AREA 0xf0000 +#define MCH_HOST_BRIDGE_PAM_AREA_SIZE 0x10000 /* 16KB */ +#define MCH_HOST_BRIDGE_PAM1 0x91 +#define MCH_HOST_BRIDGE_PAM_EXPAN_AREA 0xc0000 +#define MCH_HOST_BRIDGE_PAM_EXPAN_SIZE 0x04000 +#define MCH_HOST_BRIDGE_PAM2 0x92 +#define MCH_HOST_BRIDGE_PAM3 0x93 +#define MCH_HOST_BRIDGE_PAM4 0x94 +#define MCH_HOST_BRIDGE_PAM_EXBIOS_AREA 0xe0000 +#define MCH_HOST_BRIDGE_PAM_EXBIOS_SIZE 0x04000 +#define MCH_HOST_BRIDGE_PAM5 0x95 +#define MCH_HOST_BRIDGE_PAM6 0x96 +#define MCH_HOST_BRIDGE_PAM_WE_HI ((uint8_t)(0x2 << 4)) +#define MCH_HOST_BRIDGE_PAM_RE_HI ((uint8_t)(0x1 << 4)) +#define MCH_HOST_BRIDGE_PAM_HI_MASK ((uint8_t)(0x3 << 4)) +#define MCH_HOST_BRIDGE_PAM_WE_LO ((uint8_t)0x2) +#define MCH_HOST_BRIDGE_PAM_RE_LO ((uint8_t)0x1) +#define MCH_HOST_BRIDGE_PAM_LO_MASK ((uint8_t)0x3) +#define MCH_HOST_BRIDGE_PAM_WE ((uint8_t)0x2) +#define MCH_HOST_BRIDGE_PAM_RE ((uint8_t)0x1) +#define MCH_HOST_BRIDGE_PAM_MASK ((uint8_t)0x3) + +#define MCH_HOST_BRDIGE_SMRAM 0x9d +#define MCH_HOST_BRDIGE_SMRAM_SIZE 1 +#define MCH_HOST_BRIDGE_SMRAM_DEFAULT ((uint8_t)0x2) +#define MCH_HOST_BRIDGE_SMRAM_D_OPEN ((uint8_t)(1 << 6)) +#define MCH_HOST_BRIDGE_SMRAM_D_CLS ((uint8_t)(1 << 5)) +#define MCH_HOST_BRIDGE_SMRAM_D_LCK ((uint8_t)(1 << 4)) +#define MCH_HOST_BRIDGE_SMRAM_G_SMRAME ((uint8_t)(1 << 3)) +#define MCH_HOST_BRIDGE_SMRAM_C_BASE_SEG_MASK ((uint8_t)0x7) +#define MCH_HOST_BRIDGE_SMRAM_C_BASE_SEG ((uint8_t)0x2) /* hardwired to b010 */ +#define MCH_HOST_BRIDGE_SMRAM_C_BASE 0xa0000 +#define MCH_HOST_BRIDGE_SMRAM_C_END 0xc0000 +#define MCH_HOST_BRIDGE_SMRAM_C_SIZE 0x20000 +#define MCH_HOST_BRIDGE_UPPER_SYSTEM_BIOS_END 0x100000 + +#define MCH_HOST_BRIDGE_ESMRAMC 0x9e +#define MCH_HOST_BRDIGE_ESMRAMC_H_SMRAME ((uint8_t)(1 << 6)) +#define MCH_HOST_BRDIGE_ESMRAMC_E_SMERR ((uint8_t)(1 << 5)) +#define MCH_HOST_BRDIGE_ESMRAMC_SM_CACHE ((uint8_t)(1 << 4)) +#define MCH_HOST_BRDIGE_ESMRAMC_SM_L1 ((uint8_t)(1 << 3)) +#define MCH_HOST_BRDIGE_ESMRAMC_SM_L2 ((uint8_t)(1 << 2)) +#define MCH_HOST_BRDIGE_ESMRAMC_TSEG_SZ_MASK ((uint8_t)(0x3 << 1)) +#define MCH_HOST_BRDIGE_ESMRAMC_TSEG_SZ_1MB ((uint8_t)(0x0 << 1)) +#define MCH_HOST_BRDIGE_ESMRAMC_TSEG_SZ_2MB ((uint8_t)(0x1 << 1)) +#define MCH_HOST_BRDIGE_ESMRAMC_TSEG_SZ_8MB ((uint8_t)(0x2 << 1)) +#define MCH_HOST_BRDIGE_ESMRAMC_T_EN ((uint8_t)1) + +/* D1:F0 PCIE* port*/ +#define MCH_PCIE_DEV 1 +#define MCH_PCIE_FUNC 0 + +#endif /* HW_Q35_H */ diff --git a/hw/qdev-core.h b/hw/qdev-core.h index fce9e2249c..fff7f0f5ab 100644 --- a/hw/qdev-core.h +++ b/hw/qdev-core.h @@ -106,17 +106,12 @@ typedef struct BusChild { /** * BusState: - * @qom_allocated: Indicates whether the object was allocated by QOM. - * @glib_allocated: Indicates whether the object was initialized in-place - * yet is expected to be freed with g_free(). */ struct BusState { Object obj; DeviceState *parent; const char *name; int allow_hotplug; - bool qom_allocated; - bool glib_allocated; int max_index; QTAILQ_HEAD(ChildrenHead, BusChild) children; QLIST_ENTRY(BusState) sibling; @@ -454,7 +454,6 @@ BusState *qbus_create(const char *typename, DeviceState *parent, const char *nam BusState *bus; bus = BUS(object_new(typename)); - bus->qom_allocated = true; bus->parent = parent; bus->name = name ? g_strdup(name) : NULL; @@ -465,14 +464,7 @@ BusState *qbus_create(const char *typename, DeviceState *parent, const char *nam void qbus_free(BusState *bus) { - if (bus->qom_allocated) { - object_delete(OBJECT(bus)); - } else { - object_finalize(OBJECT(bus)); - if (bus->glib_allocated) { - g_free(bus); - } - } + object_delete(OBJECT(bus)); } static char *bus_get_fw_dev_path(BusState *bus, DeviceState *dev) @@ -705,9 +697,6 @@ static void device_finalize(Object *obj) qemu_opts_del(dev->opts); } } - if (dev->parent_bus) { - bus_remove_child(dev->parent_bus, dev); - } } static void device_class_base_init(ObjectClass *class, void *data) @@ -720,6 +709,18 @@ static void device_class_base_init(ObjectClass *class, void *data) klass->props = NULL; } +static void qdev_remove_from_bus(Object *obj) +{ + DeviceState *dev = DEVICE(obj); + + bus_remove_child(dev->parent_bus, dev); +} + +static void device_class_init(ObjectClass *class, void *data) +{ + class->unparent = qdev_remove_from_bus; +} + void device_reset(DeviceState *dev) { DeviceClass *klass = DEVICE_GET_CLASS(dev); @@ -747,6 +748,7 @@ static TypeInfo device_type_info = { .instance_init = device_initfn, .instance_finalize = device_finalize, .class_base_init = device_class_base_init, + .class_init = device_class_init, .abstract = true, .class_size = sizeof(DeviceClass), }; diff --git a/hw/smbus_ich9.c b/hw/smbus_ich9.c new file mode 100644 index 0000000000..6940583bb6 --- /dev/null +++ b/hw/smbus_ich9.c @@ -0,0 +1,159 @@ +/* + * ACPI implementation + * + * Copyright (c) 2006 Fabrice Bellard + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License version 2 as published by the Free Software Foundation. + * + * This library 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 + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, see <http://www.gnu.org/licenses/> + */ +/* + * Copyright (c) 2009 Isaku Yamahata <yamahata at valinux co jp> + * VA Linux Systems Japan K.K. + * Copyright (C) 2012 Jason Baron <jbaron@redhat.com> + * + * This is based on acpi.c, but heavily rewritten. + */ +#include "hw.h" +#include "pc.h" +#include "pm_smbus.h" +#include "pci.h" +#include "sysemu.h" +#include "i2c.h" +#include "smbus.h" + +#include "ich9.h" + +#define TYPE_ICH9_SMB_DEVICE "ICH9 SMB" +#define ICH9_SMB_DEVICE(obj) \ + OBJECT_CHECK(ICH9SMBState, (obj), TYPE_ICH9_SMB_DEVICE) + +typedef struct ICH9SMBState { + PCIDevice dev; + + PMSMBus smb; + MemoryRegion mem_bar; +} ICH9SMBState; + +static const VMStateDescription vmstate_ich9_smbus = { + .name = "ich9_smb", + .version_id = 1, + .minimum_version_id = 1, + .minimum_version_id_old = 1, + .fields = (VMStateField[]) { + VMSTATE_PCI_DEVICE(dev, struct ICH9SMBState), + VMSTATE_END_OF_LIST() + } +}; + +static void ich9_smb_ioport_writeb(void *opaque, hwaddr addr, + uint64_t val, unsigned size) +{ + ICH9SMBState *s = opaque; + uint8_t hostc = s->dev.config[ICH9_SMB_HOSTC]; + + if ((hostc & ICH9_SMB_HOSTC_HST_EN) && !(hostc & ICH9_SMB_HOSTC_I2C_EN)) { + uint64_t offset = addr - s->dev.io_regions[ICH9_SMB_SMB_BASE_BAR].addr; + smb_ioport_writeb(&s->smb, offset, val); + } +} + +static uint64_t ich9_smb_ioport_readb(void *opaque, hwaddr addr, + unsigned size) +{ + ICH9SMBState *s = opaque; + uint8_t hostc = s->dev.config[ICH9_SMB_HOSTC]; + + if ((hostc & ICH9_SMB_HOSTC_HST_EN) && !(hostc & ICH9_SMB_HOSTC_I2C_EN)) { + uint64_t offset = addr - s->dev.io_regions[ICH9_SMB_SMB_BASE_BAR].addr; + return smb_ioport_readb(&s->smb, offset); + } + + return 0xff; +} + +static const MemoryRegionOps lpc_smb_mmio_ops = { + .read = ich9_smb_ioport_readb, + .write = ich9_smb_ioport_writeb, + .endianness = DEVICE_LITTLE_ENDIAN, + .impl = { + .min_access_size = 1, + .max_access_size = 1, + }, +}; + +static int ich9_smbus_initfn(PCIDevice *d) +{ + ICH9SMBState *s = ICH9_SMB_DEVICE(d); + + /* TODO? D31IP.SMIP in chipset configuration space */ + pci_config_set_interrupt_pin(d->config, 0x01); /* interrupt pin 1 */ + + pci_set_byte(d->config + ICH9_SMB_HOSTC, 0); + + /* + * update parameters based on + * paralell_hds[0] + * serial_hds[0] + * serial_hds[0] + * fdc + * + * Is there any OS that depends on them? + */ + + /* TODO smb_io_base */ + pci_set_byte(d->config + ICH9_SMB_HOSTC, 0); + /* TODO bar0, bar1: 64bit BAR support*/ + + memory_region_init_io(&s->mem_bar, &lpc_smb_mmio_ops, s, "ich9-smbus-bar", + ICH9_SMB_SMB_BASE_SIZE); + pci_register_bar(d, ICH9_SMB_SMB_BASE_BAR, PCI_BASE_ADDRESS_SPACE_IO, + &s->mem_bar); + pm_smbus_init(&d->qdev, &s->smb); + return 0; +} + +static void ich9_smb_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + PCIDeviceClass *k = PCI_DEVICE_CLASS(klass); + + k->vendor_id = PCI_VENDOR_ID_INTEL; + k->device_id = PCI_DEVICE_ID_INTEL_ICH9_6; + k->revision = ICH9_A2_SMB_REVISION; + k->class_id = PCI_CLASS_SERIAL_SMBUS; + dc->no_user = 1; + dc->vmsd = &vmstate_ich9_smbus; + dc->desc = "ICH9 SMBUS Bridge"; + k->init = ich9_smbus_initfn; +} + +i2c_bus *ich9_smb_init(PCIBus *bus, int devfn, uint32_t smb_io_base) +{ + PCIDevice *d = + pci_create_simple_multifunction(bus, devfn, true, TYPE_ICH9_SMB_DEVICE); + ICH9SMBState *s = ICH9_SMB_DEVICE(d); + return s->smb.smbus; +} + +static const TypeInfo ich9_smb_info = { + .name = TYPE_ICH9_SMB_DEVICE, + .parent = TYPE_PCI_DEVICE, + .instance_size = sizeof(ICH9SMBState), + .class_init = ich9_smb_class_init, +}; + +static void ich9_smb_register(void) +{ + type_register_static(&ich9_smb_info); +} + +type_init(ich9_smb_register); diff --git a/hw/sysbus.c b/hw/sysbus.c index 4969f06a66..ef8ffb6603 100644 --- a/hw/sysbus.c +++ b/hw/sysbus.c @@ -274,7 +274,7 @@ static void main_system_bus_create(void) main_system_bus = g_malloc0(system_bus_info.instance_size); qbus_create_inplace(main_system_bus, TYPE_SYSTEM_BUS, NULL, "main-system-bus"); - main_system_bus->glib_allocated = true; + OBJECT(main_system_bus)->free = g_free; object_property_add_child(container_get(qdev_get_machine(), "/unattached"), "sysbus", OBJECT(main_system_bus), NULL); diff --git a/hw/virtio-rng.c b/hw/virtio-rng.c index 3ca96c855f..df329f25e1 100644 --- a/hw/virtio-rng.c +++ b/hw/virtio-rng.c @@ -22,14 +22,9 @@ typedef struct VirtIORNG { /* Only one vq - guest puts buffer(s) on it when it needs entropy */ VirtQueue *vq; - VirtQueueElement elem; - /* Config data for the device -- currently only chardev */ VirtIORNGConf *conf; - /* Whether we've popped a vq element into 'elem' above */ - bool popped; - RngBackend *rng; /* We purposefully don't migrate this state. The quota will reset on the @@ -48,17 +43,12 @@ static bool is_guest_ready(VirtIORNG *vrng) return false; } -static size_t pop_an_elem(VirtIORNG *vrng) +static size_t get_request_size(VirtQueue *vq) { - size_t size; - - if (!vrng->popped && !virtqueue_pop(vrng->vq, &vrng->elem)) { - return 0; - } - vrng->popped = true; + unsigned int in, out; - size = iov_size(vrng->elem.in_sg, vrng->elem.in_num); - return size; + virtqueue_get_avail_bytes(vq, &in, &out); + return in; } static void virtio_rng_process(VirtIORNG *vrng); @@ -67,6 +57,7 @@ static void virtio_rng_process(VirtIORNG *vrng); static void chr_read(void *opaque, const void *buf, size_t size) { VirtIORNG *vrng = opaque; + VirtQueueElement elem; size_t len; int offset; @@ -78,43 +69,33 @@ static void chr_read(void *opaque, const void *buf, size_t size) offset = 0; while (offset < size) { - if (!pop_an_elem(vrng)) { + if (!virtqueue_pop(vrng->vq, &elem)) { break; } - len = iov_from_buf(vrng->elem.in_sg, vrng->elem.in_num, + len = iov_from_buf(elem.in_sg, elem.in_num, 0, buf + offset, size - offset); offset += len; - virtqueue_push(vrng->vq, &vrng->elem, len); - vrng->popped = false; + virtqueue_push(vrng->vq, &elem, len); } virtio_notify(&vrng->vdev, vrng->vq); - - /* - * Lastly, if we had multiple elems queued by the guest, and we - * didn't have enough data to fill them all, indicate we want more - * data. - */ - virtio_rng_process(vrng); } static void virtio_rng_process(VirtIORNG *vrng) { - ssize_t size; + size_t size; if (!is_guest_ready(vrng)) { return; } - size = pop_an_elem(vrng); + size = get_request_size(vrng->vq); size = MIN(vrng->quota_remaining, size); - - if (size > 0) { + if (size) { rng_backend_request_entropy(vrng->rng, size, chr_read, vrng); } } - static void handle_input(VirtIODevice *vdev, VirtQueue *vq) { VirtIORNG *vrng = DO_UPCAST(VirtIORNG, vdev, vdev); @@ -131,23 +112,6 @@ static void virtio_rng_save(QEMUFile *f, void *opaque) VirtIORNG *vrng = opaque; virtio_save(&vrng->vdev, f); - - qemu_put_byte(f, vrng->popped); - if (vrng->popped) { - int i; - - qemu_put_be32(f, vrng->elem.index); - - qemu_put_be32(f, vrng->elem.in_num); - for (i = 0; i < vrng->elem.in_num; i++) { - qemu_put_be64(f, vrng->elem.in_addr[i]); - } - - qemu_put_be32(f, vrng->elem.out_num); - for (i = 0; i < vrng->elem.out_num; i++) { - qemu_put_be64(f, vrng->elem.out_addr[i]); - } - } } static int virtio_rng_load(QEMUFile *f, void *opaque, int version_id) @@ -159,34 +123,10 @@ static int virtio_rng_load(QEMUFile *f, void *opaque, int version_id) } virtio_load(&vrng->vdev, f); - vrng->popped = qemu_get_byte(f); - if (vrng->popped) { - int i; - - vrng->elem.index = qemu_get_be32(f); - - vrng->elem.in_num = qemu_get_be32(f); - g_assert(vrng->elem.in_num < VIRTQUEUE_MAX_SIZE); - for (i = 0; i < vrng->elem.in_num; i++) { - vrng->elem.in_addr[i] = qemu_get_be64(f); - } - - vrng->elem.out_num = qemu_get_be32(f); - g_assert(vrng->elem.out_num < VIRTQUEUE_MAX_SIZE); - for (i = 0; i < vrng->elem.out_num; i++) { - vrng->elem.out_addr[i] = qemu_get_be64(f); - } - - virtqueue_map_sg(vrng->elem.in_sg, vrng->elem.in_addr, - vrng->elem.in_num, 1); - virtqueue_map_sg(vrng->elem.out_sg, vrng->elem.out_addr, - vrng->elem.out_num, 0); - } - /* We may have an element ready but couldn't process it due to a quota - limit. Make sure to try again after live migration when the quota may - have been reset. - */ + * limit. Make sure to try again after live migration when the quota may + * have been reset. + */ virtio_rng_process(vrng); return 0; @@ -232,10 +172,9 @@ VirtIODevice *virtio_rng_init(DeviceState *dev, VirtIORNGConf *conf) vrng->qdev = dev; vrng->conf = conf; - vrng->popped = false; - vrng->quota_remaining = vrng->conf->max_bytes; - g_assert_cmpint(vrng->conf->max_bytes, <=, INT64_MAX); + assert(vrng->conf->max_bytes <= INT64_MAX); + vrng->quota_remaining = vrng->conf->max_bytes; vrng->rate_limit_timer = qemu_new_timer_ms(vm_clock, check_rate_limit, vrng); @@ -253,6 +192,8 @@ void virtio_rng_exit(VirtIODevice *vdev) { VirtIORNG *vrng = DO_UPCAST(VirtIORNG, vdev, vdev); + qemu_del_timer(vrng->rate_limit_timer); + qemu_free_timer(vrng->rate_limit_timer); unregister_savevm(vrng->qdev, "virtio-rng", vrng); virtio_cleanup(vdev); } diff --git a/include/qemu/object.h b/include/qemu/object.h index be707f1a36..ed1f47f050 100644 --- a/include/qemu/object.h +++ b/include/qemu/object.h @@ -230,6 +230,23 @@ typedef struct ObjectProperty } ObjectProperty; /** + * ObjectUnparent: + * @obj: the object that is being removed from the composition tree + * + * Called when an object is being removed from the QOM composition tree. + * The function should remove any backlinks from children objects to @obj. + */ +typedef void (ObjectUnparent)(Object *obj); + +/** + * ObjectFree: + * @obj: the object being freed + * + * Called when an object's last reference is removed. + */ +typedef void (ObjectFree)(void *obj); + +/** * ObjectClass: * * The base for all classes. The only thing that #ObjectClass contains is an @@ -240,6 +257,8 @@ struct ObjectClass /*< private >*/ Type type; GSList *interfaces; + + ObjectUnparent *unparent; }; /** @@ -261,6 +280,7 @@ struct Object { /*< private >*/ ObjectClass *class; + ObjectFree *free; QTAILQ_HEAD(, ObjectProperty) properties; uint32_t ref; Object *parent; @@ -485,15 +505,6 @@ void object_initialize_with_type(void *data, Type type); void object_initialize(void *obj, const char *typename); /** - * object_finalize: - * @obj: The object to finalize. - * - * This function destroys and object without freeing the memory associated with - * it. - */ -void object_finalize(void *obj); - -/** * object_dynamic_cast: * @obj: The object to cast. * @typename: The @typename to cast to. diff --git a/include/qemu/rng.h b/include/qemu/rng.h index 7e9d6723ff..d094bf8d4c 100644 --- a/include/qemu/rng.h +++ b/include/qemu/rng.h @@ -61,10 +61,10 @@ struct RngBackend * This function is used by the front-end to request entropy from an entropy * source. This function can be called multiple times before @receive_entropy * is invoked with different values of @receive_entropy and @opaque. The - * backend will queue each request and handle appropriate. + * backend will queue each request and handle appropriately. * * The backend does not need to pass the full amount of data to @receive_entropy - * but will pass at a value greater than 0. + * but will pass a value greater than 0. */ void rng_backend_request_entropy(RngBackend *s, size_t size, EntropyReceiveFunc *receive_entropy, @@ -87,7 +87,7 @@ void rng_backend_cancel_requests(RngBackend *s); * * This function will open the backend if it is not already open. Calling this * function on an already opened backend will not result in an error. - */ + */ void rng_backend_open(RngBackend *s, Error **errp); #endif @@ -275,4 +275,6 @@ void kvm_irqchip_release_virq(KVMState *s, int virq); int kvm_irqchip_add_irqfd_notifier(KVMState *s, EventNotifier *n, int virq); int kvm_irqchip_remove_irqfd_notifier(KVMState *s, EventNotifier *n, int virq); +void kvm_pc_gsi_handler(void *opaque, int n, int level); +void kvm_pc_setup_irq_routing(bool pci_enabled); #endif diff --git a/qapi/qapi-dealloc-visitor.c b/qapi/qapi-dealloc-visitor.c index a154523731..a07b171b86 100644 --- a/qapi/qapi-dealloc-visitor.c +++ b/qapi/qapi-dealloc-visitor.c @@ -132,6 +132,11 @@ static void qapi_dealloc_type_number(Visitor *v, double *obj, const char *name, { } +static void qapi_dealloc_type_size(Visitor *v, size_t *obj, const char *name, + Error **errp) +{ +} + static void qapi_dealloc_type_enum(Visitor *v, int *obj, const char *strings[], const char *kind, const char *name, Error **errp) @@ -164,6 +169,7 @@ QapiDeallocVisitor *qapi_dealloc_visitor_new(void) v->visitor.type_bool = qapi_dealloc_type_bool; v->visitor.type_str = qapi_dealloc_type_str; v->visitor.type_number = qapi_dealloc_type_number; + v->visitor.type_size = qapi_dealloc_type_size; QTAILQ_INIT(&v->stack); diff --git a/qom/object.c b/qom/object.c index d7092b09d8..0739aa2943 100644 --- a/qom/object.c +++ b/qom/object.c @@ -307,6 +307,7 @@ void object_initialize_with_type(void *data, TypeImpl *type) memset(obj, 0, type->instance_size); obj->class = type->class; + object_ref(obj); QTAILQ_INIT(&obj->properties); object_init_with_type(obj, type); } @@ -362,6 +363,9 @@ void object_unparent(Object *obj) if (obj->parent) { object_property_del_child(obj->parent, obj, NULL); } + if (obj->class->unparent) { + (obj->class->unparent)(obj); + } } static void object_deinit(Object *obj, TypeImpl *type) @@ -375,7 +379,7 @@ static void object_deinit(Object *obj, TypeImpl *type) } } -void object_finalize(void *data) +static void object_finalize(void *data) { Object *obj = data; TypeImpl *ti = obj->class->type; @@ -384,6 +388,9 @@ void object_finalize(void *data) object_property_del_all(obj); g_assert(obj->ref == 0); + if (obj->free) { + obj->free(obj); + } } Object *object_new_with_type(Type type) @@ -395,7 +402,7 @@ Object *object_new_with_type(Type type) obj = g_malloc(type->instance_size); object_initialize_with_type(obj, type); - object_ref(obj); + obj->free = g_free; return obj; } @@ -412,12 +419,11 @@ void object_delete(Object *obj) object_unparent(obj); g_assert(obj->ref == 1); object_unref(obj); - g_free(obj); } Object *object_dynamic_cast(Object *obj, const char *typename) { - if (object_class_dynamic_cast(object_get_class(obj), typename)) { + if (obj && object_class_dynamic_cast(object_get_class(obj), typename)) { return obj; } @@ -430,7 +436,7 @@ Object *object_dynamic_cast_assert(Object *obj, const char *typename) inst = object_dynamic_cast(obj, typename); - if (!inst) { + if (!inst && obj) { fprintf(stderr, "Object %p is not an instance of type %s\n", obj, typename); abort(); diff --git a/target-i386/cpu.c b/target-i386/cpu.c index 64c34910a0..c6c2ca03a1 100644 --- a/target-i386/cpu.c +++ b/target-i386/cpu.c @@ -315,7 +315,7 @@ typedef struct x86_def_t { /* missing: CPUID_VME, CPUID_DTS, CPUID_SS, CPUID_HT, CPUID_TM, CPUID_PBE */ #define TCG_EXT_FEATURES (CPUID_EXT_SSE3 | CPUID_EXT_MONITOR | \ - CPUID_EXT_CX16 | CPUID_EXT_POPCNT | \ + CPUID_EXT_SSSE3 | CPUID_EXT_CX16 | CPUID_EXT_POPCNT | \ CPUID_EXT_HYPERVISOR) /* missing: CPUID_EXT_DTES64, CPUID_EXT_DSCPL, CPUID_EXT_VMX, CPUID_EXT_EST, @@ -657,7 +657,7 @@ static x86_def_t builtin_x86_defs[] = { .stepping = 1, .features = CPUID_SSE2 | CPUID_SSE | CPUID_FXSR | CPUID_MMX | CPUID_CLFLUSH | CPUID_PSE36 | CPUID_PAT | CPUID_CMOV | CPUID_MCA | - CPUID_PGE | CPUID_MTRR | CPUID_APIC | CPUID_CX8 | + CPUID_PGE | CPUID_MTRR | CPUID_SEP | CPUID_APIC | CPUID_CX8 | CPUID_MCE | CPUID_PAE | CPUID_MSR | CPUID_TSC | CPUID_PSE | CPUID_DE | CPUID_FP87, .ext_features = CPUID_EXT_AVX | CPUID_EXT_XSAVE | CPUID_EXT_AES | @@ -666,7 +666,8 @@ static x86_def_t builtin_x86_defs[] = { CPUID_EXT_PCLMULQDQ | CPUID_EXT_SSE3 | CPUID_EXT_TSC_DEADLINE_TIMER | CPUID_EXT_FMA | CPUID_EXT_MOVBE | CPUID_EXT_PCID, - .ext2_features = CPUID_EXT2_LM | CPUID_EXT2_NX | CPUID_EXT2_SYSCALL, + .ext2_features = CPUID_EXT2_LM | CPUID_EXT2_RDTSCP | CPUID_EXT2_NX | + CPUID_EXT2_SYSCALL, .ext3_features = CPUID_EXT3_LAHF_LM, .cpuid_7_0_ebx_features = CPUID_7_0_EBX_FSGSBASE | CPUID_7_0_EBX_BMI1 | CPUID_7_0_EBX_HLE | CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP | diff --git a/tests/Makefile b/tests/Makefile index ca680e5644..b60f0fb8f0 100644 --- a/tests/Makefile +++ b/tests/Makefile @@ -15,6 +15,8 @@ check-unit-y += tests/test-string-output-visitor$(EXESUF) check-unit-y += tests/test-coroutine$(EXESUF) check-unit-y += tests/test-visitor-serialization$(EXESUF) check-unit-y += tests/test-iov$(EXESUF) +check-unit-y += tests/test-aio$(EXESUF) +check-unit-y += tests/test-thread-pool$(EXESUF) check-block-$(CONFIG_POSIX) += tests/qemu-iotests-quick.sh @@ -49,6 +51,8 @@ tests/check-qlist$(EXESUF): tests/check-qlist.o qlist.o qint.o tests/check-qfloat$(EXESUF): tests/check-qfloat.o qfloat.o tests/check-qjson$(EXESUF): tests/check-qjson.o $(qobject-obj-y) qemu-tool.o tests/test-coroutine$(EXESUF): tests/test-coroutine.o $(coroutine-obj-y) $(tools-obj-y) $(block-obj-y) iov.o libqemustub.a +tests/test-aio$(EXESUF): tests/test-aio.o $(coroutine-obj-y) $(tools-obj-y) $(block-obj-y) libqemustub.a +tests/test-thread-pool$(EXESUF): tests/test-thread-pool.o $(coroutine-obj-y) $(tools-obj-y) $(block-obj-y) libqemustub.a tests/test-iov$(EXESUF): tests/test-iov.o iov.o tests/test-qapi-types.c tests/test-qapi-types.h :\ diff --git a/tests/rtc-test.c b/tests/rtc-test.c index 7fdc94a3de..02edbf5727 100644 --- a/tests/rtc-test.c +++ b/tests/rtc-test.c @@ -327,6 +327,45 @@ static void fuzz_registers(void) } } +static void register_b_set_flag(void) +{ + /* Enable binary-coded decimal (BCD) mode and SET flag in Register B*/ + cmos_write(RTC_REG_B, (cmos_read(RTC_REG_B) & ~REG_B_DM) | REG_B_SET); + + cmos_write(RTC_REG_A, 0x76); + cmos_write(RTC_YEAR, 0x11); + cmos_write(RTC_CENTURY, 0x20); + cmos_write(RTC_MONTH, 0x02); + cmos_write(RTC_DAY_OF_MONTH, 0x02); + cmos_write(RTC_HOURS, 0x02); + cmos_write(RTC_MINUTES, 0x04); + cmos_write(RTC_SECONDS, 0x58); + cmos_write(RTC_REG_A, 0x26); + + /* Since SET flag is still enabled, these are equality checks. */ + g_assert_cmpint(cmos_read(RTC_HOURS), ==, 0x02); + g_assert_cmpint(cmos_read(RTC_MINUTES), ==, 0x04); + g_assert_cmpint(cmos_read(RTC_SECONDS), ==, 0x58); + g_assert_cmpint(cmos_read(RTC_DAY_OF_MONTH), ==, 0x02); + g_assert_cmpint(cmos_read(RTC_MONTH), ==, 0x02); + g_assert_cmpint(cmos_read(RTC_YEAR), ==, 0x11); + g_assert_cmpint(cmos_read(RTC_CENTURY), ==, 0x20); + + /* Disable SET flag in Register B */ + cmos_write(RTC_REG_B, cmos_read(RTC_REG_B) & ~REG_B_SET); + + g_assert_cmpint(cmos_read(RTC_HOURS), ==, 0x02); + g_assert_cmpint(cmos_read(RTC_MINUTES), ==, 0x04); + + /* Since SET flag is disabled, this is an inequality check. + * We (reasonably) assume that no (sexagesimal) overflow occurs. */ + g_assert_cmpint(cmos_read(RTC_SECONDS), >=, 0x58); + g_assert_cmpint(cmos_read(RTC_DAY_OF_MONTH), ==, 0x02); + g_assert_cmpint(cmos_read(RTC_MONTH), ==, 0x02); + g_assert_cmpint(cmos_read(RTC_YEAR), ==, 0x11); + g_assert_cmpint(cmos_read(RTC_CENTURY), ==, 0x20); +} + int main(int argc, char **argv) { QTestState *s = NULL; @@ -342,6 +381,7 @@ int main(int argc, char **argv) qtest_add_func("/rtc/alarm-time", alarm_time); qtest_add_func("/rtc/set-year/20xx", set_year_20xx); qtest_add_func("/rtc/set-year/1980", set_year_1980); + qtest_add_func("/rtc/register_b_set_flag", register_b_set_flag); qtest_add_func("/rtc/fuzz-registers", fuzz_registers); ret = g_test_run(); diff --git a/tests/test-aio.c b/tests/test-aio.c new file mode 100644 index 0000000000..f53c908707 --- /dev/null +++ b/tests/test-aio.c @@ -0,0 +1,667 @@ +/* + * AioContext tests + * + * Copyright Red Hat, Inc. 2012 + * + * Authors: + * Paolo Bonzini <pbonzini@redhat.com> + * + * This work is licensed under the terms of the GNU LGPL, version 2 or later. + * See the COPYING.LIB file in the top-level directory. + */ + +#include <glib.h> +#include "qemu-aio.h" + +AioContext *ctx; + +/* Simple callbacks for testing. */ + +typedef struct { + QEMUBH *bh; + int n; + int max; +} BHTestData; + +static void bh_test_cb(void *opaque) +{ + BHTestData *data = opaque; + if (++data->n < data->max) { + qemu_bh_schedule(data->bh); + } +} + +static void bh_delete_cb(void *opaque) +{ + BHTestData *data = opaque; + if (++data->n < data->max) { + qemu_bh_schedule(data->bh); + } else { + qemu_bh_delete(data->bh); + data->bh = NULL; + } +} + +typedef struct { + EventNotifier e; + int n; + int active; + bool auto_set; +} EventNotifierTestData; + +static int event_active_cb(EventNotifier *e) +{ + EventNotifierTestData *data = container_of(e, EventNotifierTestData, e); + return data->active > 0; +} + +static void event_ready_cb(EventNotifier *e) +{ + EventNotifierTestData *data = container_of(e, EventNotifierTestData, e); + g_assert(event_notifier_test_and_clear(e)); + data->n++; + if (data->active > 0) { + data->active--; + } + if (data->auto_set && data->active) { + event_notifier_set(e); + } +} + +/* Tests using aio_*. */ + +static void test_notify(void) +{ + g_assert(!aio_poll(ctx, false)); + aio_notify(ctx); + g_assert(!aio_poll(ctx, true)); + g_assert(!aio_poll(ctx, false)); +} + +static void test_flush(void) +{ + g_assert(!aio_poll(ctx, false)); + aio_notify(ctx); + aio_flush(ctx); + g_assert(!aio_poll(ctx, false)); +} + +static void test_bh_schedule(void) +{ + BHTestData data = { .n = 0 }; + data.bh = aio_bh_new(ctx, bh_test_cb, &data); + + qemu_bh_schedule(data.bh); + g_assert_cmpint(data.n, ==, 0); + + g_assert(aio_poll(ctx, true)); + g_assert_cmpint(data.n, ==, 1); + + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 1); + qemu_bh_delete(data.bh); +} + +static void test_bh_schedule10(void) +{ + BHTestData data = { .n = 0, .max = 10 }; + data.bh = aio_bh_new(ctx, bh_test_cb, &data); + + qemu_bh_schedule(data.bh); + g_assert_cmpint(data.n, ==, 0); + + g_assert(aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 1); + + g_assert(aio_poll(ctx, true)); + g_assert_cmpint(data.n, ==, 2); + + aio_flush(ctx); + g_assert_cmpint(data.n, ==, 10); + + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 10); + qemu_bh_delete(data.bh); +} + +static void test_bh_cancel(void) +{ + BHTestData data = { .n = 0 }; + data.bh = aio_bh_new(ctx, bh_test_cb, &data); + + qemu_bh_schedule(data.bh); + g_assert_cmpint(data.n, ==, 0); + + qemu_bh_cancel(data.bh); + g_assert_cmpint(data.n, ==, 0); + + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 0); + qemu_bh_delete(data.bh); +} + +static void test_bh_delete(void) +{ + BHTestData data = { .n = 0 }; + data.bh = aio_bh_new(ctx, bh_test_cb, &data); + + qemu_bh_schedule(data.bh); + g_assert_cmpint(data.n, ==, 0); + + qemu_bh_delete(data.bh); + g_assert_cmpint(data.n, ==, 0); + + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 0); +} + +static void test_bh_delete_from_cb(void) +{ + BHTestData data1 = { .n = 0, .max = 1 }; + + data1.bh = aio_bh_new(ctx, bh_delete_cb, &data1); + + qemu_bh_schedule(data1.bh); + g_assert_cmpint(data1.n, ==, 0); + + aio_flush(ctx); + g_assert_cmpint(data1.n, ==, data1.max); + g_assert(data1.bh == NULL); + + g_assert(!aio_poll(ctx, false)); + g_assert(!aio_poll(ctx, true)); +} + +static void test_bh_delete_from_cb_many(void) +{ + BHTestData data1 = { .n = 0, .max = 1 }; + BHTestData data2 = { .n = 0, .max = 3 }; + BHTestData data3 = { .n = 0, .max = 2 }; + BHTestData data4 = { .n = 0, .max = 4 }; + + data1.bh = aio_bh_new(ctx, bh_delete_cb, &data1); + data2.bh = aio_bh_new(ctx, bh_delete_cb, &data2); + data3.bh = aio_bh_new(ctx, bh_delete_cb, &data3); + data4.bh = aio_bh_new(ctx, bh_delete_cb, &data4); + + qemu_bh_schedule(data1.bh); + qemu_bh_schedule(data2.bh); + qemu_bh_schedule(data3.bh); + qemu_bh_schedule(data4.bh); + g_assert_cmpint(data1.n, ==, 0); + g_assert_cmpint(data2.n, ==, 0); + g_assert_cmpint(data3.n, ==, 0); + g_assert_cmpint(data4.n, ==, 0); + + g_assert(aio_poll(ctx, false)); + g_assert_cmpint(data1.n, ==, 1); + g_assert_cmpint(data2.n, ==, 1); + g_assert_cmpint(data3.n, ==, 1); + g_assert_cmpint(data4.n, ==, 1); + g_assert(data1.bh == NULL); + + aio_flush(ctx); + g_assert_cmpint(data1.n, ==, data1.max); + g_assert_cmpint(data2.n, ==, data2.max); + g_assert_cmpint(data3.n, ==, data3.max); + g_assert_cmpint(data4.n, ==, data4.max); + g_assert(data1.bh == NULL); + g_assert(data2.bh == NULL); + g_assert(data3.bh == NULL); + g_assert(data4.bh == NULL); +} + +static void test_bh_flush(void) +{ + BHTestData data = { .n = 0 }; + data.bh = aio_bh_new(ctx, bh_test_cb, &data); + + qemu_bh_schedule(data.bh); + g_assert_cmpint(data.n, ==, 0); + + aio_flush(ctx); + g_assert_cmpint(data.n, ==, 1); + + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 1); + qemu_bh_delete(data.bh); +} + +static void test_set_event_notifier(void) +{ + EventNotifierTestData data = { .n = 0, .active = 0 }; + event_notifier_init(&data.e, false); + aio_set_event_notifier(ctx, &data.e, event_ready_cb, event_active_cb); + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 0); + + aio_set_event_notifier(ctx, &data.e, NULL, NULL); + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 0); + event_notifier_cleanup(&data.e); +} + +static void test_wait_event_notifier(void) +{ + EventNotifierTestData data = { .n = 0, .active = 1 }; + event_notifier_init(&data.e, false); + aio_set_event_notifier(ctx, &data.e, event_ready_cb, event_active_cb); + g_assert(aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 0); + g_assert_cmpint(data.active, ==, 1); + + event_notifier_set(&data.e); + g_assert(aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 1); + g_assert_cmpint(data.active, ==, 0); + + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 1); + g_assert_cmpint(data.active, ==, 0); + + aio_set_event_notifier(ctx, &data.e, NULL, NULL); + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 1); + + event_notifier_cleanup(&data.e); +} + +static void test_flush_event_notifier(void) +{ + EventNotifierTestData data = { .n = 0, .active = 10, .auto_set = true }; + event_notifier_init(&data.e, false); + aio_set_event_notifier(ctx, &data.e, event_ready_cb, event_active_cb); + g_assert(aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 0); + g_assert_cmpint(data.active, ==, 10); + + event_notifier_set(&data.e); + g_assert(aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 1); + g_assert_cmpint(data.active, ==, 9); + g_assert(aio_poll(ctx, false)); + + aio_flush(ctx); + g_assert_cmpint(data.n, ==, 10); + g_assert_cmpint(data.active, ==, 0); + g_assert(!aio_poll(ctx, false)); + + aio_set_event_notifier(ctx, &data.e, NULL, NULL); + g_assert(!aio_poll(ctx, false)); + event_notifier_cleanup(&data.e); +} + +static void test_wait_event_notifier_noflush(void) +{ + EventNotifierTestData data = { .n = 0 }; + EventNotifierTestData dummy = { .n = 0, .active = 1 }; + + event_notifier_init(&data.e, false); + aio_set_event_notifier(ctx, &data.e, event_ready_cb, NULL); + + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 0); + + /* Until there is an active descriptor, aio_poll may or may not call + * event_ready_cb. Still, it must not block. */ + event_notifier_set(&data.e); + g_assert(!aio_poll(ctx, true)); + data.n = 0; + + /* An active event notifier forces aio_poll to look at EventNotifiers. */ + event_notifier_init(&dummy.e, false); + aio_set_event_notifier(ctx, &dummy.e, event_ready_cb, event_active_cb); + + event_notifier_set(&data.e); + g_assert(aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 1); + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 1); + + event_notifier_set(&data.e); + g_assert(aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 2); + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 2); + + event_notifier_set(&dummy.e); + aio_flush(ctx); + g_assert_cmpint(data.n, ==, 2); + g_assert_cmpint(dummy.n, ==, 1); + g_assert_cmpint(dummy.active, ==, 0); + + aio_set_event_notifier(ctx, &dummy.e, NULL, NULL); + event_notifier_cleanup(&dummy.e); + + aio_set_event_notifier(ctx, &data.e, NULL, NULL); + g_assert(!aio_poll(ctx, false)); + g_assert_cmpint(data.n, ==, 2); + + event_notifier_cleanup(&data.e); +} + +/* Now the same tests, using the context as a GSource. They are + * very similar to the ones above, with g_main_context_iteration + * replacing aio_poll. However: + * - sometimes both the AioContext and the glib main loop wake + * themselves up. Hence, some "g_assert(!aio_poll(ctx, false));" + * are replaced by "while (g_main_context_iteration(NULL, false));". + * - there is no exact replacement for aio_flush's blocking wait. + * "while (g_main_context_iteration(NULL, true)" seems to work, + * but it is not documented _why_ it works. For these tests a + * non-blocking loop like "while (g_main_context_iteration(NULL, false)" + * works well, and that's what I am using. + */ + +static void test_source_notify(void) +{ + while (g_main_context_iteration(NULL, false)); + aio_notify(ctx); + g_assert(g_main_context_iteration(NULL, true)); + g_assert(!g_main_context_iteration(NULL, false)); +} + +static void test_source_flush(void) +{ + g_assert(!g_main_context_iteration(NULL, false)); + aio_notify(ctx); + while (g_main_context_iteration(NULL, false)); + g_assert(!g_main_context_iteration(NULL, false)); +} + +static void test_source_bh_schedule(void) +{ + BHTestData data = { .n = 0 }; + data.bh = aio_bh_new(ctx, bh_test_cb, &data); + + qemu_bh_schedule(data.bh); + g_assert_cmpint(data.n, ==, 0); + + g_assert(g_main_context_iteration(NULL, true)); + g_assert_cmpint(data.n, ==, 1); + + g_assert(!g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 1); + qemu_bh_delete(data.bh); +} + +static void test_source_bh_schedule10(void) +{ + BHTestData data = { .n = 0, .max = 10 }; + data.bh = aio_bh_new(ctx, bh_test_cb, &data); + + qemu_bh_schedule(data.bh); + g_assert_cmpint(data.n, ==, 0); + + g_assert(g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 1); + + g_assert(g_main_context_iteration(NULL, true)); + g_assert_cmpint(data.n, ==, 2); + + while (g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 10); + + g_assert(!g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 10); + qemu_bh_delete(data.bh); +} + +static void test_source_bh_cancel(void) +{ + BHTestData data = { .n = 0 }; + data.bh = aio_bh_new(ctx, bh_test_cb, &data); + + qemu_bh_schedule(data.bh); + g_assert_cmpint(data.n, ==, 0); + + qemu_bh_cancel(data.bh); + g_assert_cmpint(data.n, ==, 0); + + while (g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 0); + qemu_bh_delete(data.bh); +} + +static void test_source_bh_delete(void) +{ + BHTestData data = { .n = 0 }; + data.bh = aio_bh_new(ctx, bh_test_cb, &data); + + qemu_bh_schedule(data.bh); + g_assert_cmpint(data.n, ==, 0); + + qemu_bh_delete(data.bh); + g_assert_cmpint(data.n, ==, 0); + + while (g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 0); +} + +static void test_source_bh_delete_from_cb(void) +{ + BHTestData data1 = { .n = 0, .max = 1 }; + + data1.bh = aio_bh_new(ctx, bh_delete_cb, &data1); + + qemu_bh_schedule(data1.bh); + g_assert_cmpint(data1.n, ==, 0); + + g_main_context_iteration(NULL, true); + g_assert_cmpint(data1.n, ==, data1.max); + g_assert(data1.bh == NULL); + + g_assert(!g_main_context_iteration(NULL, false)); +} + +static void test_source_bh_delete_from_cb_many(void) +{ + BHTestData data1 = { .n = 0, .max = 1 }; + BHTestData data2 = { .n = 0, .max = 3 }; + BHTestData data3 = { .n = 0, .max = 2 }; + BHTestData data4 = { .n = 0, .max = 4 }; + + data1.bh = aio_bh_new(ctx, bh_delete_cb, &data1); + data2.bh = aio_bh_new(ctx, bh_delete_cb, &data2); + data3.bh = aio_bh_new(ctx, bh_delete_cb, &data3); + data4.bh = aio_bh_new(ctx, bh_delete_cb, &data4); + + qemu_bh_schedule(data1.bh); + qemu_bh_schedule(data2.bh); + qemu_bh_schedule(data3.bh); + qemu_bh_schedule(data4.bh); + g_assert_cmpint(data1.n, ==, 0); + g_assert_cmpint(data2.n, ==, 0); + g_assert_cmpint(data3.n, ==, 0); + g_assert_cmpint(data4.n, ==, 0); + + g_assert(g_main_context_iteration(NULL, false)); + g_assert_cmpint(data1.n, ==, 1); + g_assert_cmpint(data2.n, ==, 1); + g_assert_cmpint(data3.n, ==, 1); + g_assert_cmpint(data4.n, ==, 1); + g_assert(data1.bh == NULL); + + while (g_main_context_iteration(NULL, false)); + g_assert_cmpint(data1.n, ==, data1.max); + g_assert_cmpint(data2.n, ==, data2.max); + g_assert_cmpint(data3.n, ==, data3.max); + g_assert_cmpint(data4.n, ==, data4.max); + g_assert(data1.bh == NULL); + g_assert(data2.bh == NULL); + g_assert(data3.bh == NULL); + g_assert(data4.bh == NULL); +} + +static void test_source_bh_flush(void) +{ + BHTestData data = { .n = 0 }; + data.bh = aio_bh_new(ctx, bh_test_cb, &data); + + qemu_bh_schedule(data.bh); + g_assert_cmpint(data.n, ==, 0); + + g_assert(g_main_context_iteration(NULL, true)); + g_assert_cmpint(data.n, ==, 1); + + g_assert(!g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 1); + qemu_bh_delete(data.bh); +} + +static void test_source_set_event_notifier(void) +{ + EventNotifierTestData data = { .n = 0, .active = 0 }; + event_notifier_init(&data.e, false); + aio_set_event_notifier(ctx, &data.e, event_ready_cb, event_active_cb); + while (g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 0); + + aio_set_event_notifier(ctx, &data.e, NULL, NULL); + while (g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 0); + event_notifier_cleanup(&data.e); +} + +static void test_source_wait_event_notifier(void) +{ + EventNotifierTestData data = { .n = 0, .active = 1 }; + event_notifier_init(&data.e, false); + aio_set_event_notifier(ctx, &data.e, event_ready_cb, event_active_cb); + g_assert(g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 0); + g_assert_cmpint(data.active, ==, 1); + + event_notifier_set(&data.e); + g_assert(g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 1); + g_assert_cmpint(data.active, ==, 0); + + while (g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 1); + g_assert_cmpint(data.active, ==, 0); + + aio_set_event_notifier(ctx, &data.e, NULL, NULL); + while (g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 1); + + event_notifier_cleanup(&data.e); +} + +static void test_source_flush_event_notifier(void) +{ + EventNotifierTestData data = { .n = 0, .active = 10, .auto_set = true }; + event_notifier_init(&data.e, false); + aio_set_event_notifier(ctx, &data.e, event_ready_cb, event_active_cb); + g_assert(g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 0); + g_assert_cmpint(data.active, ==, 10); + + event_notifier_set(&data.e); + g_assert(g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 1); + g_assert_cmpint(data.active, ==, 9); + g_assert(g_main_context_iteration(NULL, false)); + + while (g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 10); + g_assert_cmpint(data.active, ==, 0); + g_assert(!g_main_context_iteration(NULL, false)); + + aio_set_event_notifier(ctx, &data.e, NULL, NULL); + while (g_main_context_iteration(NULL, false)); + event_notifier_cleanup(&data.e); +} + +static void test_source_wait_event_notifier_noflush(void) +{ + EventNotifierTestData data = { .n = 0 }; + EventNotifierTestData dummy = { .n = 0, .active = 1 }; + + event_notifier_init(&data.e, false); + aio_set_event_notifier(ctx, &data.e, event_ready_cb, NULL); + + while (g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 0); + + /* Until there is an active descriptor, glib may or may not call + * event_ready_cb. Still, it must not block. */ + event_notifier_set(&data.e); + g_main_context_iteration(NULL, true); + data.n = 0; + + /* An active event notifier forces aio_poll to look at EventNotifiers. */ + event_notifier_init(&dummy.e, false); + aio_set_event_notifier(ctx, &dummy.e, event_ready_cb, event_active_cb); + + event_notifier_set(&data.e); + g_assert(g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 1); + g_assert(!g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 1); + + event_notifier_set(&data.e); + g_assert(g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 2); + g_assert(!g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 2); + + event_notifier_set(&dummy.e); + while (g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 2); + g_assert_cmpint(dummy.n, ==, 1); + g_assert_cmpint(dummy.active, ==, 0); + + aio_set_event_notifier(ctx, &dummy.e, NULL, NULL); + event_notifier_cleanup(&dummy.e); + + aio_set_event_notifier(ctx, &data.e, NULL, NULL); + while (g_main_context_iteration(NULL, false)); + g_assert_cmpint(data.n, ==, 2); + + event_notifier_cleanup(&data.e); +} + +/* End of tests. */ + +int main(int argc, char **argv) +{ + GSource *src; + + ctx = aio_context_new(); + src = aio_get_g_source(ctx); + g_source_attach(src, NULL); + g_source_unref(src); + + while (g_main_context_iteration(NULL, false)); + + g_test_init(&argc, &argv, NULL); + g_test_add_func("/aio/notify", test_notify); + g_test_add_func("/aio/flush", test_flush); + g_test_add_func("/aio/bh/schedule", test_bh_schedule); + g_test_add_func("/aio/bh/schedule10", test_bh_schedule10); + g_test_add_func("/aio/bh/cancel", test_bh_cancel); + g_test_add_func("/aio/bh/delete", test_bh_delete); + g_test_add_func("/aio/bh/callback-delete/one", test_bh_delete_from_cb); + g_test_add_func("/aio/bh/callback-delete/many", test_bh_delete_from_cb_many); + g_test_add_func("/aio/bh/flush", test_bh_flush); + g_test_add_func("/aio/event/add-remove", test_set_event_notifier); + g_test_add_func("/aio/event/wait", test_wait_event_notifier); + g_test_add_func("/aio/event/wait/no-flush-cb", test_wait_event_notifier_noflush); + g_test_add_func("/aio/event/flush", test_flush_event_notifier); + + g_test_add_func("/aio-gsource/notify", test_source_notify); + g_test_add_func("/aio-gsource/flush", test_source_flush); + g_test_add_func("/aio-gsource/bh/schedule", test_source_bh_schedule); + g_test_add_func("/aio-gsource/bh/schedule10", test_source_bh_schedule10); + g_test_add_func("/aio-gsource/bh/cancel", test_source_bh_cancel); + g_test_add_func("/aio-gsource/bh/delete", test_source_bh_delete); + g_test_add_func("/aio-gsource/bh/callback-delete/one", test_source_bh_delete_from_cb); + g_test_add_func("/aio-gsource/bh/callback-delete/many", test_source_bh_delete_from_cb_many); + g_test_add_func("/aio-gsource/bh/flush", test_source_bh_flush); + g_test_add_func("/aio-gsource/event/add-remove", test_source_set_event_notifier); + g_test_add_func("/aio-gsource/event/wait", test_source_wait_event_notifier); + g_test_add_func("/aio-gsource/event/wait/no-flush-cb", test_source_wait_event_notifier_noflush); + g_test_add_func("/aio-gsource/event/flush", test_source_flush_event_notifier); + return g_test_run(); +} diff --git a/tests/test-thread-pool.c b/tests/test-thread-pool.c new file mode 100644 index 0000000000..fea0445fb4 --- /dev/null +++ b/tests/test-thread-pool.c @@ -0,0 +1,216 @@ +#include <glib.h> +#include "qemu-common.h" +#include "qemu-aio.h" +#include "thread-pool.h" +#include "block.h" + +static int active; + +typedef struct { + BlockDriverAIOCB *aiocb; + int n; + int ret; +} WorkerTestData; + +static int worker_cb(void *opaque) +{ + WorkerTestData *data = opaque; + return __sync_fetch_and_add(&data->n, 1); +} + +static int long_cb(void *opaque) +{ + WorkerTestData *data = opaque; + __sync_fetch_and_add(&data->n, 1); + g_usleep(2000000); + __sync_fetch_and_add(&data->n, 1); + return 0; +} + +static void done_cb(void *opaque, int ret) +{ + WorkerTestData *data = opaque; + g_assert_cmpint(data->ret, ==, -EINPROGRESS); + data->ret = ret; + data->aiocb = NULL; + + /* Callbacks are serialized, so no need to use atomic ops. */ + active--; +} + +/* A non-blocking poll of the main AIO context (we cannot use aio_poll + * because we do not know the AioContext). + */ +static void qemu_aio_wait_nonblocking(void) +{ + qemu_notify_event(); + qemu_aio_wait(); +} + +static void test_submit(void) +{ + WorkerTestData data = { .n = 0 }; + thread_pool_submit(worker_cb, &data); + qemu_aio_flush(); + g_assert_cmpint(data.n, ==, 1); +} + +static void test_submit_aio(void) +{ + WorkerTestData data = { .n = 0, .ret = -EINPROGRESS }; + data.aiocb = thread_pool_submit_aio(worker_cb, &data, done_cb, &data); + + /* The callbacks are not called until after the first wait. */ + active = 1; + g_assert_cmpint(data.ret, ==, -EINPROGRESS); + qemu_aio_flush(); + g_assert_cmpint(active, ==, 0); + g_assert_cmpint(data.n, ==, 1); + g_assert_cmpint(data.ret, ==, 0); +} + +static void co_test_cb(void *opaque) +{ + WorkerTestData *data = opaque; + + active = 1; + data->n = 0; + data->ret = -EINPROGRESS; + thread_pool_submit_co(worker_cb, data); + + /* The test continues in test_submit_co, after qemu_coroutine_enter... */ + + g_assert_cmpint(data->n, ==, 1); + data->ret = 0; + active--; + + /* The test continues in test_submit_co, after qemu_aio_flush... */ +} + +static void test_submit_co(void) +{ + WorkerTestData data; + Coroutine *co = qemu_coroutine_create(co_test_cb); + + qemu_coroutine_enter(co, &data); + + /* Back here once the worker has started. */ + + g_assert_cmpint(active, ==, 1); + g_assert_cmpint(data.ret, ==, -EINPROGRESS); + + /* qemu_aio_flush will execute the rest of the coroutine. */ + + qemu_aio_flush(); + + /* Back here after the coroutine has finished. */ + + g_assert_cmpint(active, ==, 0); + g_assert_cmpint(data.ret, ==, 0); +} + +static void test_submit_many(void) +{ + WorkerTestData data[100]; + int i; + + /* Start more work items than there will be threads. */ + for (i = 0; i < 100; i++) { + data[i].n = 0; + data[i].ret = -EINPROGRESS; + thread_pool_submit_aio(worker_cb, &data[i], done_cb, &data[i]); + } + + active = 100; + while (active > 0) { + qemu_aio_wait(); + } + for (i = 0; i < 100; i++) { + g_assert_cmpint(data[i].n, ==, 1); + g_assert_cmpint(data[i].ret, ==, 0); + } +} + +static void test_cancel(void) +{ + WorkerTestData data[100]; + int num_canceled; + int i; + + /* Start more work items than there will be threads, to ensure + * the pool is full. + */ + test_submit_many(); + + /* Start long running jobs, to ensure we can cancel some. */ + for (i = 0; i < 100; i++) { + data[i].n = 0; + data[i].ret = -EINPROGRESS; + data[i].aiocb = thread_pool_submit_aio(long_cb, &data[i], + done_cb, &data[i]); + } + + /* Starting the threads may be left to a bottom half. Let it + * run, but do not waste too much time... + */ + active = 100; + qemu_aio_wait_nonblocking(); + + /* Wait some time for the threads to start, with some sanity + * testing on the behavior of the scheduler... + */ + g_assert_cmpint(active, ==, 100); + g_usleep(1000000); + g_assert_cmpint(active, >, 50); + + /* Cancel the jobs that haven't been started yet. */ + num_canceled = 0; + for (i = 0; i < 100; i++) { + if (__sync_val_compare_and_swap(&data[i].n, 0, 3) == 0) { + data[i].ret = -ECANCELED; + bdrv_aio_cancel(data[i].aiocb); + active--; + num_canceled++; + } + } + g_assert_cmpint(active, >, 0); + g_assert_cmpint(num_canceled, <, 100); + + /* Canceling the others will be a blocking operation. */ + for (i = 0; i < 100; i++) { + if (data[i].n != 3) { + bdrv_aio_cancel(data[i].aiocb); + } + } + + /* Finish execution and execute any remaining callbacks. */ + qemu_aio_flush(); + g_assert_cmpint(active, ==, 0); + for (i = 0; i < 100; i++) { + if (data[i].n == 3) { + g_assert_cmpint(data[i].ret, ==, -ECANCELED); + g_assert(data[i].aiocb != NULL); + } else { + g_assert_cmpint(data[i].n, ==, 2); + g_assert_cmpint(data[i].ret, ==, 0); + g_assert(data[i].aiocb == NULL); + } + } +} + +int main(int argc, char **argv) +{ + /* These should be removed once each AioContext has its thread pool. + * The test should create its own AioContext. + */ + qemu_init_main_loop(); + bdrv_init(); + + g_test_init(&argc, &argv, NULL); + g_test_add_func("/thread-pool/submit", test_submit); + g_test_add_func("/thread-pool/submit-aio", test_submit_aio); + g_test_add_func("/thread-pool/submit-co", test_submit_co); + g_test_add_func("/thread-pool/submit-many", test_submit_many); + g_test_add_func("/thread-pool/cancel", test_cancel); + return g_test_run(); +} |