From bca964bcea65e69895ff7200ca0b5ddd01b87d24 Mon Sep 17 00:00:00 2001 From: Sergio Lopez Date: Thu, 26 Sep 2019 15:02:22 +0200 Subject: hw/virtio: Factorize virtio-mmio headers Put QOM and main struct definition in a separate header file, so it can be accessed from other components. Signed-off-by: Sergio Lopez Reviewed-by: Philippe Mathieu-Daudé Tested-by: Philippe Mathieu-Daudé Reviewed-by: Michael S. Tsirkin --- include/hw/virtio/virtio-mmio.h | 73 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 73 insertions(+) create mode 100644 include/hw/virtio/virtio-mmio.h (limited to 'include') diff --git a/include/hw/virtio/virtio-mmio.h b/include/hw/virtio/virtio-mmio.h new file mode 100644 index 0000000000..7dbfd03dcf --- /dev/null +++ b/include/hw/virtio/virtio-mmio.h @@ -0,0 +1,73 @@ +/* + * Virtio MMIO bindings + * + * Copyright (c) 2011 Linaro Limited + * + * Author: + * Peter Maydell + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, see . + */ + +#ifndef HW_VIRTIO_MMIO_H +#define HW_VIRTIO_MMIO_H + +#include "hw/virtio/virtio-bus.h" + +/* QOM macros */ +/* virtio-mmio-bus */ +#define TYPE_VIRTIO_MMIO_BUS "virtio-mmio-bus" +#define VIRTIO_MMIO_BUS(obj) \ + OBJECT_CHECK(VirtioBusState, (obj), TYPE_VIRTIO_MMIO_BUS) +#define VIRTIO_MMIO_BUS_GET_CLASS(obj) \ + OBJECT_GET_CLASS(VirtioBusClass, (obj), TYPE_VIRTIO_MMIO_BUS) +#define VIRTIO_MMIO_BUS_CLASS(klass) \ + OBJECT_CLASS_CHECK(VirtioBusClass, (klass), TYPE_VIRTIO_MMIO_BUS) + +/* virtio-mmio */ +#define TYPE_VIRTIO_MMIO "virtio-mmio" +#define VIRTIO_MMIO(obj) \ + OBJECT_CHECK(VirtIOMMIOProxy, (obj), TYPE_VIRTIO_MMIO) + +#define VIRT_MAGIC 0x74726976 /* 'virt' */ +#define VIRT_VERSION 2 +#define VIRT_VERSION_LEGACY 1 +#define VIRT_VENDOR 0x554D4551 /* 'QEMU' */ + +typedef struct VirtIOMMIOQueue { + uint16_t num; + bool enabled; + uint32_t desc[2]; + uint32_t avail[2]; + uint32_t used[2]; +} VirtIOMMIOQueue; + +typedef struct { + /* Generic */ + SysBusDevice parent_obj; + MemoryRegion iomem; + qemu_irq irq; + bool legacy; + /* Guest accessible state needing migration and reset */ + uint32_t host_features_sel; + uint32_t guest_features_sel; + uint32_t guest_page_shift; + /* virtio-bus */ + VirtioBusState bus; + bool format_transport_address; + /* Fields only used for non-legacy (v2) devices */ + uint32_t guest_features[2]; + VirtIOMMIOQueue vqs[VIRTIO_QUEUE_MAX]; +} VirtIOMMIOProxy; + +#endif -- cgit v1.2.3-55-g7522 From 81ef68e44c84377fe66e0b27626f2f562f558a6d Mon Sep 17 00:00:00 2001 From: Sergio Lopez Date: Thu, 26 Sep 2019 15:26:16 +0200 Subject: hw/i386/pc: rename functions shared with non-PC machines The following functions are named *pc* but are not PC-machine specific but generic to the X86 architecture, rename them: load_linux -> x86_load_linux pc_new_cpu -> x86_new_cpu pc_cpus_init -> x86_cpus_init pc_cpu_index_to_props -> x86_cpu_index_to_props pc_get_default_cpu_node_id -> x86_get_default_cpu_node_id pc_possible_cpu_arch_ids -> x86_possible_cpu_arch_ids old_pc_system_rom_init -> x86_system_rom_init Signed-off-by: Sergio Lopez Reviewed-by: Philippe Mathieu-Daudé Tested-by: Philippe Mathieu-Daudé Reviewed-by: Stefano Garzarella Reviewed-by: Michael S. Tsirkin --- hw/i386/pc.c | 28 ++++++++++++++-------------- hw/i386/pc_piix.c | 2 +- hw/i386/pc_q35.c | 2 +- hw/i386/pc_sysfw.c | 6 +++--- include/hw/i386/pc.h | 2 +- 5 files changed, 20 insertions(+), 20 deletions(-) (limited to 'include') diff --git a/hw/i386/pc.c b/hw/i386/pc.c index 4b1904237e..1c27223774 100644 --- a/hw/i386/pc.c +++ b/hw/i386/pc.c @@ -1019,8 +1019,8 @@ static bool load_elfboot(const char *kernel_filename, return true; } -static void load_linux(PCMachineState *pcms, - FWCfgState *fw_cfg) +static void x86_load_linux(PCMachineState *pcms, + FWCfgState *fw_cfg) { uint16_t protocol; int setup_size, kernel_size, cmdline_size; @@ -1374,7 +1374,7 @@ void pc_acpi_smi_interrupt(void *opaque, int irq, int level) } } -static void pc_new_cpu(PCMachineState *pcms, int64_t apic_id, Error **errp) +static void x86_cpu_new(PCMachineState *pcms, int64_t apic_id, Error **errp) { Object *cpu = NULL; Error *local_err = NULL; @@ -1490,14 +1490,14 @@ void pc_hot_add_cpu(MachineState *ms, const int64_t id, Error **errp) return; } - pc_new_cpu(PC_MACHINE(ms), apic_id, &local_err); + x86_cpu_new(PC_MACHINE(ms), apic_id, &local_err); if (local_err) { error_propagate(errp, local_err); return; } } -void pc_cpus_init(PCMachineState *pcms) +void x86_cpus_init(PCMachineState *pcms) { int i; const CPUArchIdList *possible_cpus; @@ -1518,7 +1518,7 @@ void pc_cpus_init(PCMachineState *pcms) ms->smp.max_cpus - 1) + 1; possible_cpus = mc->possible_cpu_arch_ids(ms); for (i = 0; i < ms->smp.cpus; i++) { - pc_new_cpu(pcms, possible_cpus->cpus[i].arch_id, &error_fatal); + x86_cpu_new(pcms, possible_cpus->cpus[i].arch_id, &error_fatal); } } @@ -1621,7 +1621,7 @@ void xen_load_linux(PCMachineState *pcms) fw_cfg_add_i16(fw_cfg, FW_CFG_NB_CPUS, pcms->boot_cpus); rom_set_fw(fw_cfg); - load_linux(pcms, fw_cfg); + x86_load_linux(pcms, fw_cfg); for (i = 0; i < nb_option_roms; i++) { assert(!strcmp(option_rom[i].name, "linuxboot.bin") || !strcmp(option_rom[i].name, "linuxboot_dma.bin") || @@ -1756,7 +1756,7 @@ void pc_memory_init(PCMachineState *pcms, } if (linux_boot) { - load_linux(pcms, fw_cfg); + x86_load_linux(pcms, fw_cfg); } for (i = 0; i < nb_option_roms; i++) { @@ -2681,7 +2681,7 @@ static void pc_machine_wakeup(MachineState *machine) } static CpuInstanceProperties -pc_cpu_index_to_props(MachineState *ms, unsigned cpu_index) +x86_cpu_index_to_props(MachineState *ms, unsigned cpu_index) { MachineClass *mc = MACHINE_GET_CLASS(ms); const CPUArchIdList *possible_cpus = mc->possible_cpu_arch_ids(ms); @@ -2690,7 +2690,7 @@ pc_cpu_index_to_props(MachineState *ms, unsigned cpu_index) return possible_cpus->cpus[cpu_index].props; } -static int64_t pc_get_default_cpu_node_id(const MachineState *ms, int idx) +static int64_t x86_get_default_cpu_node_id(const MachineState *ms, int idx) { X86CPUTopoInfo topo; PCMachineState *pcms = PC_MACHINE(ms); @@ -2702,7 +2702,7 @@ static int64_t pc_get_default_cpu_node_id(const MachineState *ms, int idx) return topo.pkg_id % ms->numa_state->num_nodes; } -static const CPUArchIdList *pc_possible_cpu_arch_ids(MachineState *ms) +static const CPUArchIdList *x86_possible_cpu_arch_ids(MachineState *ms) { PCMachineState *pcms = PC_MACHINE(ms); int i; @@ -2804,9 +2804,9 @@ static void pc_machine_class_init(ObjectClass *oc, void *data) assert(!mc->get_hotplug_handler); mc->get_hotplug_handler = pc_get_hotplug_handler; mc->hotplug_allowed = pc_hotplug_allowed; - mc->cpu_index_to_instance_props = pc_cpu_index_to_props; - mc->get_default_cpu_node_id = pc_get_default_cpu_node_id; - mc->possible_cpu_arch_ids = pc_possible_cpu_arch_ids; + mc->cpu_index_to_instance_props = x86_cpu_index_to_props; + mc->get_default_cpu_node_id = x86_get_default_cpu_node_id; + mc->possible_cpu_arch_ids = x86_possible_cpu_arch_ids; mc->auto_enable_numa_with_memhp = true; mc->has_hotpluggable_cpus = true; mc->default_boot_order = "cad"; diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c index 6824b72124..de09e076cd 100644 --- a/hw/i386/pc_piix.c +++ b/hw/i386/pc_piix.c @@ -152,7 +152,7 @@ static void pc_init1(MachineState *machine, } } - pc_cpus_init(pcms); + x86_cpus_init(pcms); if (kvm_enabled() && pcmc->kvmclock_enabled) { kvmclock_create(); diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c index 8fad20f314..894989b64e 100644 --- a/hw/i386/pc_q35.c +++ b/hw/i386/pc_q35.c @@ -179,7 +179,7 @@ static void pc_q35_init(MachineState *machine) xen_hvm_init(pcms, &ram_memory); } - pc_cpus_init(pcms); + x86_cpus_init(pcms); kvmclock_create(); diff --git a/hw/i386/pc_sysfw.c b/hw/i386/pc_sysfw.c index a9983f0bfb..28cb1f63c9 100644 --- a/hw/i386/pc_sysfw.c +++ b/hw/i386/pc_sysfw.c @@ -211,7 +211,7 @@ static void pc_system_flash_map(PCMachineState *pcms, } } -static void old_pc_system_rom_init(MemoryRegion *rom_memory, bool isapc_ram_fw) +static void x86_bios_rom_init(MemoryRegion *rom_memory, bool isapc_ram_fw) { char *filename; MemoryRegion *bios, *isa_bios; @@ -272,7 +272,7 @@ void pc_system_firmware_init(PCMachineState *pcms, BlockBackend *pflash_blk[ARRAY_SIZE(pcms->flash)]; if (!pcmc->pci_enabled) { - old_pc_system_rom_init(rom_memory, true); + x86_bios_rom_init(rom_memory, true); return; } @@ -293,7 +293,7 @@ void pc_system_firmware_init(PCMachineState *pcms, if (!pflash_blk[0]) { /* Machine property pflash0 not set, use ROM mode */ - old_pc_system_rom_init(rom_memory, false); + x86_bios_rom_init(rom_memory, false); } else { if (kvm_enabled() && !kvm_readonly_mem_enabled()) { /* diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h index 37bfd95113..7082b06c51 100644 --- a/include/hw/i386/pc.h +++ b/include/hw/i386/pc.h @@ -198,7 +198,7 @@ bool pc_machine_is_smm_enabled(PCMachineState *pcms); void pc_register_ferr_irq(qemu_irq irq); void pc_acpi_smi_interrupt(void *opaque, int irq, int level); -void pc_cpus_init(PCMachineState *pcms); +void x86_cpus_init(PCMachineState *pcms); void pc_hot_add_cpu(MachineState *ms, const int64_t id, Error **errp); void pc_smp_parse(MachineState *ms, QemuOpts *opts); -- cgit v1.2.3-55-g7522 From 549e984e67d8b3ea868be4ba935cecb9c1e753dc Mon Sep 17 00:00:00 2001 From: Sergio Lopez Date: Tue, 8 Oct 2019 11:56:49 +0200 Subject: hw/i386/pc: move shared x86 functions to x86.c and export them Move x86 functions that will be shared between PC and non-PC machine types to x86.c, along with their helpers. Signed-off-by: Sergio Lopez Reviewed-by: Philippe Mathieu-Daudé Tested-by: Philippe Mathieu-Daudé Reviewed-by: Stefano Garzarella Reviewed-by: Michael S. Tsirkin --- hw/i386/Makefile.objs | 1 + hw/i386/pc.c | 587 +----------------------------------------- hw/i386/pc_piix.c | 1 + hw/i386/pc_q35.c | 1 + hw/i386/pc_sysfw.c | 56 +--- hw/i386/x86.c | 690 ++++++++++++++++++++++++++++++++++++++++++++++++++ include/hw/i386/pc.h | 1 - include/hw/i386/x86.h | 35 +++ 8 files changed, 730 insertions(+), 642 deletions(-) create mode 100644 hw/i386/x86.c create mode 100644 include/hw/i386/x86.h (limited to 'include') diff --git a/hw/i386/Makefile.objs b/hw/i386/Makefile.objs index d3374e0831..7ed80a4853 100644 --- a/hw/i386/Makefile.objs +++ b/hw/i386/Makefile.objs @@ -1,5 +1,6 @@ obj-$(CONFIG_KVM) += kvm/ obj-y += e820_memory_layout.o multiboot.o +obj-y += x86.o obj-y += pc.o obj-$(CONFIG_I440FX) += pc_piix.o obj-$(CONFIG_Q35) += pc_q35.o diff --git a/hw/i386/pc.c b/hw/i386/pc.c index 903bc05f8f..4a1bc95ccb 100644 --- a/hw/i386/pc.c +++ b/hw/i386/pc.c @@ -24,6 +24,7 @@ #include "qemu/osdep.h" #include "qemu/units.h" +#include "hw/i386/x86.h" #include "hw/i386/pc.h" #include "hw/char/serial.h" #include "hw/char/parallel.h" @@ -103,9 +104,6 @@ struct hpet_fw_config hpet_cfg = {.count = UINT8_MAX}; -/* Physical Address of PVH entry point read from kernel ELF NOTE */ -static size_t pvh_start_addr; - GlobalProperty pc_compat_4_1[] = {}; const size_t pc_compat_4_1_len = G_N_ELEMENTS(pc_compat_4_1); @@ -867,481 +865,6 @@ static void handle_a20_line_change(void *opaque, int irq, int level) x86_cpu_set_a20(cpu, level); } -/* - * Calculates initial APIC ID for a specific CPU index - * - * Currently we need to be able to calculate the APIC ID from the CPU index - * alone (without requiring a CPU object), as the QEMU<->Seabios interfaces have - * no concept of "CPU index", and the NUMA tables on fw_cfg need the APIC ID of - * all CPUs up to max_cpus. - */ -static uint32_t x86_cpu_apic_id_from_index(PCMachineState *pcms, - unsigned int cpu_index) -{ - MachineState *ms = MACHINE(pcms); - PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); - uint32_t correct_id; - static bool warned; - - correct_id = x86_apicid_from_cpu_idx(pcms->smp_dies, ms->smp.cores, - ms->smp.threads, cpu_index); - if (pcmc->compat_apic_id_mode) { - if (cpu_index != correct_id && !warned && !qtest_enabled()) { - error_report("APIC IDs set in compatibility mode, " - "CPU topology won't match the configuration"); - warned = true; - } - return cpu_index; - } else { - return correct_id; - } -} - -static long get_file_size(FILE *f) -{ - long where, size; - - /* XXX: on Unix systems, using fstat() probably makes more sense */ - - where = ftell(f); - fseek(f, 0, SEEK_END); - size = ftell(f); - fseek(f, where, SEEK_SET); - - return size; -} - -struct setup_data { - uint64_t next; - uint32_t type; - uint32_t len; - uint8_t data[0]; -} __attribute__((packed)); - - -/* - * The entry point into the kernel for PVH boot is different from - * the native entry point. The PVH entry is defined by the x86/HVM - * direct boot ABI and is available in an ELFNOTE in the kernel binary. - * - * This function is passed to load_elf() when it is called from - * load_elfboot() which then additionally checks for an ELF Note of - * type XEN_ELFNOTE_PHYS32_ENTRY and passes it to this function to - * parse the PVH entry address from the ELF Note. - * - * Due to trickery in elf_opts.h, load_elf() is actually available as - * load_elf32() or load_elf64() and this routine needs to be able - * to deal with being called as 32 or 64 bit. - * - * The address of the PVH entry point is saved to the 'pvh_start_addr' - * global variable. (although the entry point is 32-bit, the kernel - * binary can be either 32-bit or 64-bit). - */ -static uint64_t read_pvh_start_addr(void *arg1, void *arg2, bool is64) -{ - size_t *elf_note_data_addr; - - /* Check if ELF Note header passed in is valid */ - if (arg1 == NULL) { - return 0; - } - - if (is64) { - struct elf64_note *nhdr64 = (struct elf64_note *)arg1; - uint64_t nhdr_size64 = sizeof(struct elf64_note); - uint64_t phdr_align = *(uint64_t *)arg2; - uint64_t nhdr_namesz = nhdr64->n_namesz; - - elf_note_data_addr = - ((void *)nhdr64) + nhdr_size64 + - QEMU_ALIGN_UP(nhdr_namesz, phdr_align); - } else { - struct elf32_note *nhdr32 = (struct elf32_note *)arg1; - uint32_t nhdr_size32 = sizeof(struct elf32_note); - uint32_t phdr_align = *(uint32_t *)arg2; - uint32_t nhdr_namesz = nhdr32->n_namesz; - - elf_note_data_addr = - ((void *)nhdr32) + nhdr_size32 + - QEMU_ALIGN_UP(nhdr_namesz, phdr_align); - } - - pvh_start_addr = *elf_note_data_addr; - - return pvh_start_addr; -} - -static bool load_elfboot(const char *kernel_filename, - int kernel_file_size, - uint8_t *header, - size_t pvh_xen_start_addr, - FWCfgState *fw_cfg) -{ - uint32_t flags = 0; - uint32_t mh_load_addr = 0; - uint32_t elf_kernel_size = 0; - uint64_t elf_entry; - uint64_t elf_low, elf_high; - int kernel_size; - - if (ldl_p(header) != 0x464c457f) { - return false; /* no elfboot */ - } - - bool elf_is64 = header[EI_CLASS] == ELFCLASS64; - flags = elf_is64 ? - ((Elf64_Ehdr *)header)->e_flags : ((Elf32_Ehdr *)header)->e_flags; - - if (flags & 0x00010004) { /* LOAD_ELF_HEADER_HAS_ADDR */ - error_report("elfboot unsupported flags = %x", flags); - exit(1); - } - - uint64_t elf_note_type = XEN_ELFNOTE_PHYS32_ENTRY; - kernel_size = load_elf(kernel_filename, read_pvh_start_addr, - NULL, &elf_note_type, &elf_entry, - &elf_low, &elf_high, 0, I386_ELF_MACHINE, - 0, 0); - - if (kernel_size < 0) { - error_report("Error while loading elf kernel"); - exit(1); - } - mh_load_addr = elf_low; - elf_kernel_size = elf_high - elf_low; - - if (pvh_start_addr == 0) { - error_report("Error loading uncompressed kernel without PVH ELF Note"); - exit(1); - } - fw_cfg_add_i32(fw_cfg, FW_CFG_KERNEL_ENTRY, pvh_start_addr); - fw_cfg_add_i32(fw_cfg, FW_CFG_KERNEL_ADDR, mh_load_addr); - fw_cfg_add_i32(fw_cfg, FW_CFG_KERNEL_SIZE, elf_kernel_size); - - return true; -} - -static void x86_load_linux(PCMachineState *pcms, - FWCfgState *fw_cfg) -{ - uint16_t protocol; - int setup_size, kernel_size, cmdline_size; - int dtb_size, setup_data_offset; - uint32_t initrd_max; - uint8_t header[8192], *setup, *kernel; - hwaddr real_addr, prot_addr, cmdline_addr, initrd_addr = 0; - FILE *f; - char *vmode; - MachineState *machine = MACHINE(pcms); - PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); - struct setup_data *setup_data; - const char *kernel_filename = machine->kernel_filename; - const char *initrd_filename = machine->initrd_filename; - const char *dtb_filename = machine->dtb; - const char *kernel_cmdline = machine->kernel_cmdline; - - /* Align to 16 bytes as a paranoia measure */ - cmdline_size = (strlen(kernel_cmdline) + 16) & ~15; - - /* load the kernel header */ - f = fopen(kernel_filename, "rb"); - if (!f) { - fprintf(stderr, "qemu: could not open kernel file '%s': %s\n", - kernel_filename, strerror(errno)); - exit(1); - } - - kernel_size = get_file_size(f); - if (!kernel_size || - fread(header, 1, MIN(ARRAY_SIZE(header), kernel_size), f) != - MIN(ARRAY_SIZE(header), kernel_size)) { - fprintf(stderr, "qemu: could not load kernel '%s': %s\n", - kernel_filename, strerror(errno)); - exit(1); - } - - /* kernel protocol version */ - if (ldl_p(header + 0x202) == 0x53726448) { - protocol = lduw_p(header + 0x206); - } else { - /* - * This could be a multiboot kernel. If it is, let's stop treating it - * like a Linux kernel. - * Note: some multiboot images could be in the ELF format (the same of - * PVH), so we try multiboot first since we check the multiboot magic - * header before to load it. - */ - if (load_multiboot(fw_cfg, f, kernel_filename, initrd_filename, - kernel_cmdline, kernel_size, header)) { - return; - } - /* - * Check if the file is an uncompressed kernel file (ELF) and load it, - * saving the PVH entry point used by the x86/HVM direct boot ABI. - * If load_elfboot() is successful, populate the fw_cfg info. - */ - if (pcmc->pvh_enabled && - load_elfboot(kernel_filename, kernel_size, - header, pvh_start_addr, fw_cfg)) { - fclose(f); - - fw_cfg_add_i32(fw_cfg, FW_CFG_CMDLINE_SIZE, - strlen(kernel_cmdline) + 1); - fw_cfg_add_string(fw_cfg, FW_CFG_CMDLINE_DATA, kernel_cmdline); - - fw_cfg_add_i32(fw_cfg, FW_CFG_SETUP_SIZE, sizeof(header)); - fw_cfg_add_bytes(fw_cfg, FW_CFG_SETUP_DATA, - header, sizeof(header)); - - /* load initrd */ - if (initrd_filename) { - GMappedFile *mapped_file; - gsize initrd_size; - gchar *initrd_data; - GError *gerr = NULL; - - mapped_file = g_mapped_file_new(initrd_filename, false, &gerr); - if (!mapped_file) { - fprintf(stderr, "qemu: error reading initrd %s: %s\n", - initrd_filename, gerr->message); - exit(1); - } - pcms->initrd_mapped_file = mapped_file; - - initrd_data = g_mapped_file_get_contents(mapped_file); - initrd_size = g_mapped_file_get_length(mapped_file); - initrd_max = pcms->below_4g_mem_size - pcmc->acpi_data_size - 1; - if (initrd_size >= initrd_max) { - fprintf(stderr, "qemu: initrd is too large, cannot support." - "(max: %"PRIu32", need %"PRId64")\n", - initrd_max, (uint64_t)initrd_size); - exit(1); - } - - initrd_addr = (initrd_max - initrd_size) & ~4095; - - fw_cfg_add_i32(fw_cfg, FW_CFG_INITRD_ADDR, initrd_addr); - fw_cfg_add_i32(fw_cfg, FW_CFG_INITRD_SIZE, initrd_size); - fw_cfg_add_bytes(fw_cfg, FW_CFG_INITRD_DATA, initrd_data, - initrd_size); - } - - option_rom[nb_option_roms].bootindex = 0; - option_rom[nb_option_roms].name = "pvh.bin"; - nb_option_roms++; - - return; - } - protocol = 0; - } - - if (protocol < 0x200 || !(header[0x211] & 0x01)) { - /* Low kernel */ - real_addr = 0x90000; - cmdline_addr = 0x9a000 - cmdline_size; - prot_addr = 0x10000; - } else if (protocol < 0x202) { - /* High but ancient kernel */ - real_addr = 0x90000; - cmdline_addr = 0x9a000 - cmdline_size; - prot_addr = 0x100000; - } else { - /* High and recent kernel */ - real_addr = 0x10000; - cmdline_addr = 0x20000; - prot_addr = 0x100000; - } - - /* highest address for loading the initrd */ - if (protocol >= 0x20c && - lduw_p(header + 0x236) & XLF_CAN_BE_LOADED_ABOVE_4G) { - /* - * Linux has supported initrd up to 4 GB for a very long time (2007, - * long before XLF_CAN_BE_LOADED_ABOVE_4G which was added in 2013), - * though it only sets initrd_max to 2 GB to "work around bootloader - * bugs". Luckily, QEMU firmware(which does something like bootloader) - * has supported this. - * - * It's believed that if XLF_CAN_BE_LOADED_ABOVE_4G is set, initrd can - * be loaded into any address. - * - * In addition, initrd_max is uint32_t simply because QEMU doesn't - * support the 64-bit boot protocol (specifically the ext_ramdisk_image - * field). - * - * Therefore here just limit initrd_max to UINT32_MAX simply as well. - */ - initrd_max = UINT32_MAX; - } else if (protocol >= 0x203) { - initrd_max = ldl_p(header + 0x22c); - } else { - initrd_max = 0x37ffffff; - } - - if (initrd_max >= pcms->below_4g_mem_size - pcmc->acpi_data_size) { - initrd_max = pcms->below_4g_mem_size - pcmc->acpi_data_size - 1; - } - - fw_cfg_add_i32(fw_cfg, FW_CFG_CMDLINE_ADDR, cmdline_addr); - fw_cfg_add_i32(fw_cfg, FW_CFG_CMDLINE_SIZE, strlen(kernel_cmdline) + 1); - fw_cfg_add_string(fw_cfg, FW_CFG_CMDLINE_DATA, kernel_cmdline); - - if (protocol >= 0x202) { - stl_p(header + 0x228, cmdline_addr); - } else { - stw_p(header + 0x20, 0xA33F); - stw_p(header + 0x22, cmdline_addr - real_addr); - } - - /* handle vga= parameter */ - vmode = strstr(kernel_cmdline, "vga="); - if (vmode) { - unsigned int video_mode; - int ret; - /* skip "vga=" */ - vmode += 4; - if (!strncmp(vmode, "normal", 6)) { - video_mode = 0xffff; - } else if (!strncmp(vmode, "ext", 3)) { - video_mode = 0xfffe; - } else if (!strncmp(vmode, "ask", 3)) { - video_mode = 0xfffd; - } else { - ret = qemu_strtoui(vmode, NULL, 0, &video_mode); - if (ret != 0) { - fprintf(stderr, "qemu: can't parse 'vga' parameter: %s\n", - strerror(-ret)); - exit(1); - } - } - stw_p(header + 0x1fa, video_mode); - } - - /* loader type */ - /* - * High nybble = B reserved for QEMU; low nybble is revision number. - * If this code is substantially changed, you may want to consider - * incrementing the revision. - */ - if (protocol >= 0x200) { - header[0x210] = 0xB0; - } - /* heap */ - if (protocol >= 0x201) { - header[0x211] |= 0x80; /* CAN_USE_HEAP */ - stw_p(header + 0x224, cmdline_addr - real_addr - 0x200); - } - - /* load initrd */ - if (initrd_filename) { - GMappedFile *mapped_file; - gsize initrd_size; - gchar *initrd_data; - GError *gerr = NULL; - - if (protocol < 0x200) { - fprintf(stderr, "qemu: linux kernel too old to load a ram disk\n"); - exit(1); - } - - mapped_file = g_mapped_file_new(initrd_filename, false, &gerr); - if (!mapped_file) { - fprintf(stderr, "qemu: error reading initrd %s: %s\n", - initrd_filename, gerr->message); - exit(1); - } - pcms->initrd_mapped_file = mapped_file; - - initrd_data = g_mapped_file_get_contents(mapped_file); - initrd_size = g_mapped_file_get_length(mapped_file); - if (initrd_size >= initrd_max) { - fprintf(stderr, "qemu: initrd is too large, cannot support." - "(max: %"PRIu32", need %"PRId64")\n", - initrd_max, (uint64_t)initrd_size); - exit(1); - } - - initrd_addr = (initrd_max - initrd_size) & ~4095; - - fw_cfg_add_i32(fw_cfg, FW_CFG_INITRD_ADDR, initrd_addr); - fw_cfg_add_i32(fw_cfg, FW_CFG_INITRD_SIZE, initrd_size); - fw_cfg_add_bytes(fw_cfg, FW_CFG_INITRD_DATA, initrd_data, initrd_size); - - stl_p(header + 0x218, initrd_addr); - stl_p(header + 0x21c, initrd_size); - } - - /* load kernel and setup */ - setup_size = header[0x1f1]; - if (setup_size == 0) { - setup_size = 4; - } - setup_size = (setup_size + 1) * 512; - if (setup_size > kernel_size) { - fprintf(stderr, "qemu: invalid kernel header\n"); - exit(1); - } - kernel_size -= setup_size; - - setup = g_malloc(setup_size); - kernel = g_malloc(kernel_size); - fseek(f, 0, SEEK_SET); - if (fread(setup, 1, setup_size, f) != setup_size) { - fprintf(stderr, "fread() failed\n"); - exit(1); - } - if (fread(kernel, 1, kernel_size, f) != kernel_size) { - fprintf(stderr, "fread() failed\n"); - exit(1); - } - fclose(f); - - /* append dtb to kernel */ - if (dtb_filename) { - if (protocol < 0x209) { - fprintf(stderr, "qemu: Linux kernel too old to load a dtb\n"); - exit(1); - } - - dtb_size = get_image_size(dtb_filename); - if (dtb_size <= 0) { - fprintf(stderr, "qemu: error reading dtb %s: %s\n", - dtb_filename, strerror(errno)); - exit(1); - } - - setup_data_offset = QEMU_ALIGN_UP(kernel_size, 16); - kernel_size = setup_data_offset + sizeof(struct setup_data) + dtb_size; - kernel = g_realloc(kernel, kernel_size); - - stq_p(header + 0x250, prot_addr + setup_data_offset); - - setup_data = (struct setup_data *)(kernel + setup_data_offset); - setup_data->next = 0; - setup_data->type = cpu_to_le32(SETUP_DTB); - setup_data->len = cpu_to_le32(dtb_size); - - load_image_size(dtb_filename, setup_data->data, dtb_size); - } - - memcpy(setup, header, MIN(sizeof(header), setup_size)); - - fw_cfg_add_i32(fw_cfg, FW_CFG_KERNEL_ADDR, prot_addr); - fw_cfg_add_i32(fw_cfg, FW_CFG_KERNEL_SIZE, kernel_size); - fw_cfg_add_bytes(fw_cfg, FW_CFG_KERNEL_DATA, kernel, kernel_size); - - fw_cfg_add_i32(fw_cfg, FW_CFG_SETUP_ADDR, real_addr); - fw_cfg_add_i32(fw_cfg, FW_CFG_SETUP_SIZE, setup_size); - fw_cfg_add_bytes(fw_cfg, FW_CFG_SETUP_DATA, setup, setup_size); - - option_rom[nb_option_roms].bootindex = 0; - option_rom[nb_option_roms].name = "linuxboot.bin"; - if (pcmc->linuxboot_dma_enabled && fw_cfg_dma_enabled(fw_cfg)) { - option_rom[nb_option_roms].name = "linuxboot_dma.bin"; - } - nb_option_roms++; -} - #define NE2000_NB_MAX 6 static const int ne2000_io[NE2000_NB_MAX] = { 0x300, 0x320, 0x340, 0x360, @@ -1378,24 +901,6 @@ void pc_acpi_smi_interrupt(void *opaque, int irq, int level) } } -static void x86_cpu_new(PCMachineState *pcms, int64_t apic_id, Error **errp) -{ - Object *cpu = NULL; - Error *local_err = NULL; - CPUX86State *env = NULL; - - cpu = object_new(MACHINE(pcms)->cpu_type); - - env = &X86_CPU(cpu)->env; - env->nr_dies = pcms->smp_dies; - - object_property_set_uint(cpu, apic_id, "apic-id", &local_err); - object_property_set_bool(cpu, true, "realized", &local_err); - - object_unref(cpu); - error_propagate(errp, local_err); -} - /* * This function is very similar to smp_parse() * in hw/core/machine.c but includes CPU die support. @@ -1501,32 +1006,6 @@ void pc_hot_add_cpu(MachineState *ms, const int64_t id, Error **errp) } } -void x86_cpus_init(PCMachineState *pcms) -{ - int i; - const CPUArchIdList *possible_cpus; - MachineState *ms = MACHINE(pcms); - MachineClass *mc = MACHINE_GET_CLASS(pcms); - PCMachineClass *pcmc = PC_MACHINE_CLASS(mc); - - x86_cpu_set_default_version(pcmc->default_cpu_version); - - /* - * Calculates the limit to CPU APIC ID values - * - * Limit for the APIC ID value, so that all - * CPU APIC IDs are < pcms->apic_id_limit. - * - * This is used for FW_CFG_MAX_CPUS. See comments on fw_cfg_arch_create(). - */ - pcms->apic_id_limit = x86_cpu_apic_id_from_index(pcms, - ms->smp.max_cpus - 1) + 1; - possible_cpus = mc->possible_cpu_arch_ids(ms); - for (i = 0; i < ms->smp.cpus; i++) { - x86_cpu_new(pcms, possible_cpus->cpus[i].arch_id, &error_fatal); - } -} - static void rtc_set_cpus_count(ISADevice *rtc, uint16_t cpus_count) { if (cpus_count > 0xff) { @@ -2685,70 +2164,6 @@ static void pc_machine_wakeup(MachineState *machine) cpu_synchronize_all_post_reset(); } -static CpuInstanceProperties -x86_cpu_index_to_props(MachineState *ms, unsigned cpu_index) -{ - MachineClass *mc = MACHINE_GET_CLASS(ms); - const CPUArchIdList *possible_cpus = mc->possible_cpu_arch_ids(ms); - - assert(cpu_index < possible_cpus->len); - return possible_cpus->cpus[cpu_index].props; -} - -static int64_t x86_get_default_cpu_node_id(const MachineState *ms, int idx) -{ - X86CPUTopoInfo topo; - PCMachineState *pcms = PC_MACHINE(ms); - - assert(idx < ms->possible_cpus->len); - x86_topo_ids_from_apicid(ms->possible_cpus->cpus[idx].arch_id, - pcms->smp_dies, ms->smp.cores, - ms->smp.threads, &topo); - return topo.pkg_id % ms->numa_state->num_nodes; -} - -static const CPUArchIdList *x86_possible_cpu_arch_ids(MachineState *ms) -{ - PCMachineState *pcms = PC_MACHINE(ms); - int i; - unsigned int max_cpus = ms->smp.max_cpus; - - if (ms->possible_cpus) { - /* - * make sure that max_cpus hasn't changed since the first use, i.e. - * -smp hasn't been parsed after it - */ - assert(ms->possible_cpus->len == max_cpus); - return ms->possible_cpus; - } - - ms->possible_cpus = g_malloc0(sizeof(CPUArchIdList) + - sizeof(CPUArchId) * max_cpus); - ms->possible_cpus->len = max_cpus; - for (i = 0; i < ms->possible_cpus->len; i++) { - X86CPUTopoInfo topo; - - ms->possible_cpus->cpus[i].type = ms->cpu_type; - ms->possible_cpus->cpus[i].vcpus_count = 1; - ms->possible_cpus->cpus[i].arch_id = - x86_cpu_apic_id_from_index(pcms, i); - x86_topo_ids_from_apicid(ms->possible_cpus->cpus[i].arch_id, - pcms->smp_dies, ms->smp.cores, - ms->smp.threads, &topo); - ms->possible_cpus->cpus[i].props.has_socket_id = true; - ms->possible_cpus->cpus[i].props.socket_id = topo.pkg_id; - if (pcms->smp_dies > 1) { - ms->possible_cpus->cpus[i].props.has_die_id = true; - ms->possible_cpus->cpus[i].props.die_id = topo.die_id; - } - ms->possible_cpus->cpus[i].props.has_core_id = true; - ms->possible_cpus->cpus[i].props.core_id = topo.core_id; - ms->possible_cpus->cpus[i].props.has_thread_id = true; - ms->possible_cpus->cpus[i].props.thread_id = topo.smt_id; - } - return ms->possible_cpus; -} - static void x86_nmi(NMIState *n, int cpu_index, Error **errp) { /* cpu index isn't used */ diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c index de09e076cd..1396451abf 100644 --- a/hw/i386/pc_piix.c +++ b/hw/i386/pc_piix.c @@ -27,6 +27,7 @@ #include "qemu/units.h" #include "hw/loader.h" +#include "hw/i386/x86.h" #include "hw/i386/pc.h" #include "hw/i386/apic.h" #include "hw/display/ramfb.h" diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c index 894989b64e..8920bd8978 100644 --- a/hw/i386/pc_q35.c +++ b/hw/i386/pc_q35.c @@ -41,6 +41,7 @@ #include "hw/pci-host/q35.h" #include "hw/qdev-properties.h" #include "exec/address-spaces.h" +#include "hw/i386/x86.h" #include "hw/i386/pc.h" #include "hw/i386/ich9.h" #include "hw/i386/amd_iommu.h" diff --git a/hw/i386/pc_sysfw.c b/hw/i386/pc_sysfw.c index 28cb1f63c9..f5f3f466b0 100644 --- a/hw/i386/pc_sysfw.c +++ b/hw/i386/pc_sysfw.c @@ -31,6 +31,7 @@ #include "qemu/option.h" #include "qemu/units.h" #include "hw/sysbus.h" +#include "hw/i386/x86.h" #include "hw/i386/pc.h" #include "hw/loader.h" #include "hw/qdev-properties.h" @@ -38,8 +39,6 @@ #include "hw/block/flash.h" #include "sysemu/kvm.h" -#define BIOS_FILENAME "bios.bin" - /* * We don't have a theoretically justifiable exact lower bound on the base * address of any flash mapping. In practice, the IO-APIC MMIO range is @@ -211,59 +210,6 @@ static void pc_system_flash_map(PCMachineState *pcms, } } -static void x86_bios_rom_init(MemoryRegion *rom_memory, bool isapc_ram_fw) -{ - char *filename; - MemoryRegion *bios, *isa_bios; - int bios_size, isa_bios_size; - int ret; - - /* BIOS load */ - if (bios_name == NULL) { - bios_name = BIOS_FILENAME; - } - filename = qemu_find_file(QEMU_FILE_TYPE_BIOS, bios_name); - if (filename) { - bios_size = get_image_size(filename); - } else { - bios_size = -1; - } - if (bios_size <= 0 || - (bios_size % 65536) != 0) { - goto bios_error; - } - bios = g_malloc(sizeof(*bios)); - memory_region_init_ram(bios, NULL, "pc.bios", bios_size, &error_fatal); - if (!isapc_ram_fw) { - memory_region_set_readonly(bios, true); - } - ret = rom_add_file_fixed(bios_name, (uint32_t)(-bios_size), -1); - if (ret != 0) { - bios_error: - fprintf(stderr, "qemu: could not load PC BIOS '%s'\n", bios_name); - exit(1); - } - g_free(filename); - - /* map the last 128KB of the BIOS in ISA space */ - isa_bios_size = MIN(bios_size, 128 * KiB); - isa_bios = g_malloc(sizeof(*isa_bios)); - memory_region_init_alias(isa_bios, NULL, "isa-bios", bios, - bios_size - isa_bios_size, isa_bios_size); - memory_region_add_subregion_overlap(rom_memory, - 0x100000 - isa_bios_size, - isa_bios, - 1); - if (!isapc_ram_fw) { - memory_region_set_readonly(isa_bios, true); - } - - /* map all the bios at the top of memory */ - memory_region_add_subregion(rom_memory, - (uint32_t)(-bios_size), - bios); -} - void pc_system_firmware_init(PCMachineState *pcms, MemoryRegion *rom_memory) { diff --git a/hw/i386/x86.c b/hw/i386/x86.c new file mode 100644 index 0000000000..71fb093983 --- /dev/null +++ b/hw/i386/x86.c @@ -0,0 +1,690 @@ +/* + * Copyright (c) 2003-2004 Fabrice Bellard + * Copyright (c) 2019 Red Hat, Inc. + * + * 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 "qemu/osdep.h" +#include "qemu/error-report.h" +#include "qemu/option.h" +#include "qemu/cutils.h" +#include "qemu/units.h" +#include "qemu-common.h" +#include "qapi/error.h" +#include "qapi/qmp/qerror.h" +#include "qapi/qapi-visit-common.h" +#include "qapi/visitor.h" +#include "sysemu/qtest.h" +#include "sysemu/numa.h" +#include "sysemu/replay.h" +#include "sysemu/sysemu.h" + +#include "hw/i386/x86.h" +#include "hw/i386/pc.h" +#include "target/i386/cpu.h" +#include "hw/i386/topology.h" +#include "hw/i386/fw_cfg.h" + +#include "hw/acpi/cpu_hotplug.h" +#include "hw/nmi.h" +#include "hw/loader.h" +#include "multiboot.h" +#include "elf.h" +#include "standard-headers/asm-x86/bootparam.h" + +#define BIOS_FILENAME "bios.bin" + +/* Physical Address of PVH entry point read from kernel ELF NOTE */ +static size_t pvh_start_addr; + +/* + * Calculates initial APIC ID for a specific CPU index + * + * Currently we need to be able to calculate the APIC ID from the CPU index + * alone (without requiring a CPU object), as the QEMU<->Seabios interfaces have + * no concept of "CPU index", and the NUMA tables on fw_cfg need the APIC ID of + * all CPUs up to max_cpus. + */ +uint32_t x86_cpu_apic_id_from_index(PCMachineState *pcms, + unsigned int cpu_index) +{ + MachineState *ms = MACHINE(pcms); + PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); + uint32_t correct_id; + static bool warned; + + correct_id = x86_apicid_from_cpu_idx(pcms->smp_dies, ms->smp.cores, + ms->smp.threads, cpu_index); + if (pcmc->compat_apic_id_mode) { + if (cpu_index != correct_id && !warned && !qtest_enabled()) { + error_report("APIC IDs set in compatibility mode, " + "CPU topology won't match the configuration"); + warned = true; + } + return cpu_index; + } else { + return correct_id; + } +} + +void x86_cpu_new(PCMachineState *pcms, int64_t apic_id, Error **errp) +{ + Object *cpu = NULL; + Error *local_err = NULL; + CPUX86State *env = NULL; + + cpu = object_new(MACHINE(pcms)->cpu_type); + + env = &X86_CPU(cpu)->env; + env->nr_dies = pcms->smp_dies; + + object_property_set_uint(cpu, apic_id, "apic-id", &local_err); + object_property_set_bool(cpu, true, "realized", &local_err); + + object_unref(cpu); + error_propagate(errp, local_err); +} + +void x86_cpus_init(PCMachineState *pcms) +{ + int i; + const CPUArchIdList *possible_cpus; + MachineState *ms = MACHINE(pcms); + MachineClass *mc = MACHINE_GET_CLASS(pcms); + PCMachineClass *pcmc = PC_MACHINE_CLASS(mc); + + x86_cpu_set_default_version(pcmc->default_cpu_version); + + /* + * Calculates the limit to CPU APIC ID values + * + * Limit for the APIC ID value, so that all + * CPU APIC IDs are < pcms->apic_id_limit. + * + * This is used for FW_CFG_MAX_CPUS. See comments on fw_cfg_arch_create(). + */ + pcms->apic_id_limit = x86_cpu_apic_id_from_index(pcms, + ms->smp.max_cpus - 1) + 1; + possible_cpus = mc->possible_cpu_arch_ids(ms); + for (i = 0; i < ms->smp.cpus; i++) { + x86_cpu_new(pcms, possible_cpus->cpus[i].arch_id, &error_fatal); + } +} + +CpuInstanceProperties +x86_cpu_index_to_props(MachineState *ms, unsigned cpu_index) +{ + MachineClass *mc = MACHINE_GET_CLASS(ms); + const CPUArchIdList *possible_cpus = mc->possible_cpu_arch_ids(ms); + + assert(cpu_index < possible_cpus->len); + return possible_cpus->cpus[cpu_index].props; +} + +int64_t x86_get_default_cpu_node_id(const MachineState *ms, int idx) +{ + X86CPUTopoInfo topo; + PCMachineState *pcms = PC_MACHINE(ms); + + assert(idx < ms->possible_cpus->len); + x86_topo_ids_from_apicid(ms->possible_cpus->cpus[idx].arch_id, + pcms->smp_dies, ms->smp.cores, + ms->smp.threads, &topo); + return topo.pkg_id % ms->numa_state->num_nodes; +} + +const CPUArchIdList *x86_possible_cpu_arch_ids(MachineState *ms) +{ + PCMachineState *pcms = PC_MACHINE(ms); + int i; + unsigned int max_cpus = ms->smp.max_cpus; + + if (ms->possible_cpus) { + /* + * make sure that max_cpus hasn't changed since the first use, i.e. + * -smp hasn't been parsed after it + */ + assert(ms->possible_cpus->len == max_cpus); + return ms->possible_cpus; + } + + ms->possible_cpus = g_malloc0(sizeof(CPUArchIdList) + + sizeof(CPUArchId) * max_cpus); + ms->possible_cpus->len = max_cpus; + for (i = 0; i < ms->possible_cpus->len; i++) { + X86CPUTopoInfo topo; + + ms->possible_cpus->cpus[i].type = ms->cpu_type; + ms->possible_cpus->cpus[i].vcpus_count = 1; + ms->possible_cpus->cpus[i].arch_id = + x86_cpu_apic_id_from_index(pcms, i); + x86_topo_ids_from_apicid(ms->possible_cpus->cpus[i].arch_id, + pcms->smp_dies, ms->smp.cores, + ms->smp.threads, &topo); + ms->possible_cpus->cpus[i].props.has_socket_id = true; + ms->possible_cpus->cpus[i].props.socket_id = topo.pkg_id; + if (pcms->smp_dies > 1) { + ms->possible_cpus->cpus[i].props.has_die_id = true; + ms->possible_cpus->cpus[i].props.die_id = topo.die_id; + } + ms->possible_cpus->cpus[i].props.has_core_id = true; + ms->possible_cpus->cpus[i].props.core_id = topo.core_id; + ms->possible_cpus->cpus[i].props.has_thread_id = true; + ms->possible_cpus->cpus[i].props.thread_id = topo.smt_id; + } + return ms->possible_cpus; +} + +static long get_file_size(FILE *f) +{ + long where, size; + + /* XXX: on Unix systems, using fstat() probably makes more sense */ + + where = ftell(f); + fseek(f, 0, SEEK_END); + size = ftell(f); + fseek(f, where, SEEK_SET); + + return size; +} + +struct setup_data { + uint64_t next; + uint32_t type; + uint32_t len; + uint8_t data[0]; +} __attribute__((packed)); + + +/* + * The entry point into the kernel for PVH boot is different from + * the native entry point. The PVH entry is defined by the x86/HVM + * direct boot ABI and is available in an ELFNOTE in the kernel binary. + * + * This function is passed to load_elf() when it is called from + * load_elfboot() which then additionally checks for an ELF Note of + * type XEN_ELFNOTE_PHYS32_ENTRY and passes it to this function to + * parse the PVH entry address from the ELF Note. + * + * Due to trickery in elf_opts.h, load_elf() is actually available as + * load_elf32() or load_elf64() and this routine needs to be able + * to deal with being called as 32 or 64 bit. + * + * The address of the PVH entry point is saved to the 'pvh_start_addr' + * global variable. (although the entry point is 32-bit, the kernel + * binary can be either 32-bit or 64-bit). + */ +static uint64_t read_pvh_start_addr(void *arg1, void *arg2, bool is64) +{ + size_t *elf_note_data_addr; + + /* Check if ELF Note header passed in is valid */ + if (arg1 == NULL) { + return 0; + } + + if (is64) { + struct elf64_note *nhdr64 = (struct elf64_note *)arg1; + uint64_t nhdr_size64 = sizeof(struct elf64_note); + uint64_t phdr_align = *(uint64_t *)arg2; + uint64_t nhdr_namesz = nhdr64->n_namesz; + + elf_note_data_addr = + ((void *)nhdr64) + nhdr_size64 + + QEMU_ALIGN_UP(nhdr_namesz, phdr_align); + } else { + struct elf32_note *nhdr32 = (struct elf32_note *)arg1; + uint32_t nhdr_size32 = sizeof(struct elf32_note); + uint32_t phdr_align = *(uint32_t *)arg2; + uint32_t nhdr_namesz = nhdr32->n_namesz; + + elf_note_data_addr = + ((void *)nhdr32) + nhdr_size32 + + QEMU_ALIGN_UP(nhdr_namesz, phdr_align); + } + + pvh_start_addr = *elf_note_data_addr; + + return pvh_start_addr; +} + +static bool load_elfboot(const char *kernel_filename, + int kernel_file_size, + uint8_t *header, + size_t pvh_xen_start_addr, + FWCfgState *fw_cfg) +{ + uint32_t flags = 0; + uint32_t mh_load_addr = 0; + uint32_t elf_kernel_size = 0; + uint64_t elf_entry; + uint64_t elf_low, elf_high; + int kernel_size; + + if (ldl_p(header) != 0x464c457f) { + return false; /* no elfboot */ + } + + bool elf_is64 = header[EI_CLASS] == ELFCLASS64; + flags = elf_is64 ? + ((Elf64_Ehdr *)header)->e_flags : ((Elf32_Ehdr *)header)->e_flags; + + if (flags & 0x00010004) { /* LOAD_ELF_HEADER_HAS_ADDR */ + error_report("elfboot unsupported flags = %x", flags); + exit(1); + } + + uint64_t elf_note_type = XEN_ELFNOTE_PHYS32_ENTRY; + kernel_size = load_elf(kernel_filename, read_pvh_start_addr, + NULL, &elf_note_type, &elf_entry, + &elf_low, &elf_high, 0, I386_ELF_MACHINE, + 0, 0); + + if (kernel_size < 0) { + error_report("Error while loading elf kernel"); + exit(1); + } + mh_load_addr = elf_low; + elf_kernel_size = elf_high - elf_low; + + if (pvh_start_addr == 0) { + error_report("Error loading uncompressed kernel without PVH ELF Note"); + exit(1); + } + fw_cfg_add_i32(fw_cfg, FW_CFG_KERNEL_ENTRY, pvh_start_addr); + fw_cfg_add_i32(fw_cfg, FW_CFG_KERNEL_ADDR, mh_load_addr); + fw_cfg_add_i32(fw_cfg, FW_CFG_KERNEL_SIZE, elf_kernel_size); + + return true; +} + +void x86_load_linux(PCMachineState *pcms, + FWCfgState *fw_cfg) +{ + uint16_t protocol; + int setup_size, kernel_size, cmdline_size; + int dtb_size, setup_data_offset; + uint32_t initrd_max; + uint8_t header[8192], *setup, *kernel; + hwaddr real_addr, prot_addr, cmdline_addr, initrd_addr = 0; + FILE *f; + char *vmode; + MachineState *machine = MACHINE(pcms); + PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); + struct setup_data *setup_data; + const char *kernel_filename = machine->kernel_filename; + const char *initrd_filename = machine->initrd_filename; + const char *dtb_filename = machine->dtb; + const char *kernel_cmdline = machine->kernel_cmdline; + + /* Align to 16 bytes as a paranoia measure */ + cmdline_size = (strlen(kernel_cmdline) + 16) & ~15; + + /* load the kernel header */ + f = fopen(kernel_filename, "rb"); + if (!f) { + fprintf(stderr, "qemu: could not open kernel file '%s': %s\n", + kernel_filename, strerror(errno)); + exit(1); + } + + kernel_size = get_file_size(f); + if (!kernel_size || + fread(header, 1, MIN(ARRAY_SIZE(header), kernel_size), f) != + MIN(ARRAY_SIZE(header), kernel_size)) { + fprintf(stderr, "qemu: could not load kernel '%s': %s\n", + kernel_filename, strerror(errno)); + exit(1); + } + + /* kernel protocol version */ + if (ldl_p(header + 0x202) == 0x53726448) { + protocol = lduw_p(header + 0x206); + } else { + /* + * This could be a multiboot kernel. If it is, let's stop treating it + * like a Linux kernel. + * Note: some multiboot images could be in the ELF format (the same of + * PVH), so we try multiboot first since we check the multiboot magic + * header before to load it. + */ + if (load_multiboot(fw_cfg, f, kernel_filename, initrd_filename, + kernel_cmdline, kernel_size, header)) { + return; + } + /* + * Check if the file is an uncompressed kernel file (ELF) and load it, + * saving the PVH entry point used by the x86/HVM direct boot ABI. + * If load_elfboot() is successful, populate the fw_cfg info. + */ + if (pcmc->pvh_enabled && + load_elfboot(kernel_filename, kernel_size, + header, pvh_start_addr, fw_cfg)) { + fclose(f); + + fw_cfg_add_i32(fw_cfg, FW_CFG_CMDLINE_SIZE, + strlen(kernel_cmdline) + 1); + fw_cfg_add_string(fw_cfg, FW_CFG_CMDLINE_DATA, kernel_cmdline); + + fw_cfg_add_i32(fw_cfg, FW_CFG_SETUP_SIZE, sizeof(header)); + fw_cfg_add_bytes(fw_cfg, FW_CFG_SETUP_DATA, + header, sizeof(header)); + + /* load initrd */ + if (initrd_filename) { + GMappedFile *mapped_file; + gsize initrd_size; + gchar *initrd_data; + GError *gerr = NULL; + + mapped_file = g_mapped_file_new(initrd_filename, false, &gerr); + if (!mapped_file) { + fprintf(stderr, "qemu: error reading initrd %s: %s\n", + initrd_filename, gerr->message); + exit(1); + } + pcms->initrd_mapped_file = mapped_file; + + initrd_data = g_mapped_file_get_contents(mapped_file); + initrd_size = g_mapped_file_get_length(mapped_file); + initrd_max = pcms->below_4g_mem_size - pcmc->acpi_data_size - 1; + if (initrd_size >= initrd_max) { + fprintf(stderr, "qemu: initrd is too large, cannot support." + "(max: %"PRIu32", need %"PRId64")\n", + initrd_max, (uint64_t)initrd_size); + exit(1); + } + + initrd_addr = (initrd_max - initrd_size) & ~4095; + + fw_cfg_add_i32(fw_cfg, FW_CFG_INITRD_ADDR, initrd_addr); + fw_cfg_add_i32(fw_cfg, FW_CFG_INITRD_SIZE, initrd_size); + fw_cfg_add_bytes(fw_cfg, FW_CFG_INITRD_DATA, initrd_data, + initrd_size); + } + + option_rom[nb_option_roms].bootindex = 0; + option_rom[nb_option_roms].name = "pvh.bin"; + nb_option_roms++; + + return; + } + protocol = 0; + } + + if (protocol < 0x200 || !(header[0x211] & 0x01)) { + /* Low kernel */ + real_addr = 0x90000; + cmdline_addr = 0x9a000 - cmdline_size; + prot_addr = 0x10000; + } else if (protocol < 0x202) { + /* High but ancient kernel */ + real_addr = 0x90000; + cmdline_addr = 0x9a000 - cmdline_size; + prot_addr = 0x100000; + } else { + /* High and recent kernel */ + real_addr = 0x10000; + cmdline_addr = 0x20000; + prot_addr = 0x100000; + } + + /* highest address for loading the initrd */ + if (protocol >= 0x20c && + lduw_p(header + 0x236) & XLF_CAN_BE_LOADED_ABOVE_4G) { + /* + * Linux has supported initrd up to 4 GB for a very long time (2007, + * long before XLF_CAN_BE_LOADED_ABOVE_4G which was added in 2013), + * though it only sets initrd_max to 2 GB to "work around bootloader + * bugs". Luckily, QEMU firmware(which does something like bootloader) + * has supported this. + * + * It's believed that if XLF_CAN_BE_LOADED_ABOVE_4G is set, initrd can + * be loaded into any address. + * + * In addition, initrd_max is uint32_t simply because QEMU doesn't + * support the 64-bit boot protocol (specifically the ext_ramdisk_image + * field). + * + * Therefore here just limit initrd_max to UINT32_MAX simply as well. + */ + initrd_max = UINT32_MAX; + } else if (protocol >= 0x203) { + initrd_max = ldl_p(header + 0x22c); + } else { + initrd_max = 0x37ffffff; + } + + if (initrd_max >= pcms->below_4g_mem_size - pcmc->acpi_data_size) { + initrd_max = pcms->below_4g_mem_size - pcmc->acpi_data_size - 1; + } + + fw_cfg_add_i32(fw_cfg, FW_CFG_CMDLINE_ADDR, cmdline_addr); + fw_cfg_add_i32(fw_cfg, FW_CFG_CMDLINE_SIZE, strlen(kernel_cmdline) + 1); + fw_cfg_add_string(fw_cfg, FW_CFG_CMDLINE_DATA, kernel_cmdline); + + if (protocol >= 0x202) { + stl_p(header + 0x228, cmdline_addr); + } else { + stw_p(header + 0x20, 0xA33F); + stw_p(header + 0x22, cmdline_addr - real_addr); + } + + /* handle vga= parameter */ + vmode = strstr(kernel_cmdline, "vga="); + if (vmode) { + unsigned int video_mode; + int ret; + /* skip "vga=" */ + vmode += 4; + if (!strncmp(vmode, "normal", 6)) { + video_mode = 0xffff; + } else if (!strncmp(vmode, "ext", 3)) { + video_mode = 0xfffe; + } else if (!strncmp(vmode, "ask", 3)) { + video_mode = 0xfffd; + } else { + ret = qemu_strtoui(vmode, NULL, 0, &video_mode); + if (ret != 0) { + fprintf(stderr, "qemu: can't parse 'vga' parameter: %s\n", + strerror(-ret)); + exit(1); + } + } + stw_p(header + 0x1fa, video_mode); + } + + /* loader type */ + /* + * High nybble = B reserved for QEMU; low nybble is revision number. + * If this code is substantially changed, you may want to consider + * incrementing the revision. + */ + if (protocol >= 0x200) { + header[0x210] = 0xB0; + } + /* heap */ + if (protocol >= 0x201) { + header[0x211] |= 0x80; /* CAN_USE_HEAP */ + stw_p(header + 0x224, cmdline_addr - real_addr - 0x200); + } + + /* load initrd */ + if (initrd_filename) { + GMappedFile *mapped_file; + gsize initrd_size; + gchar *initrd_data; + GError *gerr = NULL; + + if (protocol < 0x200) { + fprintf(stderr, "qemu: linux kernel too old to load a ram disk\n"); + exit(1); + } + + mapped_file = g_mapped_file_new(initrd_filename, false, &gerr); + if (!mapped_file) { + fprintf(stderr, "qemu: error reading initrd %s: %s\n", + initrd_filename, gerr->message); + exit(1); + } + pcms->initrd_mapped_file = mapped_file; + + initrd_data = g_mapped_file_get_contents(mapped_file); + initrd_size = g_mapped_file_get_length(mapped_file); + if (initrd_size >= initrd_max) { + fprintf(stderr, "qemu: initrd is too large, cannot support." + "(max: %"PRIu32", need %"PRId64")\n", + initrd_max, (uint64_t)initrd_size); + exit(1); + } + + initrd_addr = (initrd_max - initrd_size) & ~4095; + + fw_cfg_add_i32(fw_cfg, FW_CFG_INITRD_ADDR, initrd_addr); + fw_cfg_add_i32(fw_cfg, FW_CFG_INITRD_SIZE, initrd_size); + fw_cfg_add_bytes(fw_cfg, FW_CFG_INITRD_DATA, initrd_data, initrd_size); + + stl_p(header + 0x218, initrd_addr); + stl_p(header + 0x21c, initrd_size); + } + + /* load kernel and setup */ + setup_size = header[0x1f1]; + if (setup_size == 0) { + setup_size = 4; + } + setup_size = (setup_size + 1) * 512; + if (setup_size > kernel_size) { + fprintf(stderr, "qemu: invalid kernel header\n"); + exit(1); + } + kernel_size -= setup_size; + + setup = g_malloc(setup_size); + kernel = g_malloc(kernel_size); + fseek(f, 0, SEEK_SET); + if (fread(setup, 1, setup_size, f) != setup_size) { + fprintf(stderr, "fread() failed\n"); + exit(1); + } + if (fread(kernel, 1, kernel_size, f) != kernel_size) { + fprintf(stderr, "fread() failed\n"); + exit(1); + } + fclose(f); + + /* append dtb to kernel */ + if (dtb_filename) { + if (protocol < 0x209) { + fprintf(stderr, "qemu: Linux kernel too old to load a dtb\n"); + exit(1); + } + + dtb_size = get_image_size(dtb_filename); + if (dtb_size <= 0) { + fprintf(stderr, "qemu: error reading dtb %s: %s\n", + dtb_filename, strerror(errno)); + exit(1); + } + + setup_data_offset = QEMU_ALIGN_UP(kernel_size, 16); + kernel_size = setup_data_offset + sizeof(struct setup_data) + dtb_size; + kernel = g_realloc(kernel, kernel_size); + + stq_p(header + 0x250, prot_addr + setup_data_offset); + + setup_data = (struct setup_data *)(kernel + setup_data_offset); + setup_data->next = 0; + setup_data->type = cpu_to_le32(SETUP_DTB); + setup_data->len = cpu_to_le32(dtb_size); + + load_image_size(dtb_filename, setup_data->data, dtb_size); + } + + memcpy(setup, header, MIN(sizeof(header), setup_size)); + + fw_cfg_add_i32(fw_cfg, FW_CFG_KERNEL_ADDR, prot_addr); + fw_cfg_add_i32(fw_cfg, FW_CFG_KERNEL_SIZE, kernel_size); + fw_cfg_add_bytes(fw_cfg, FW_CFG_KERNEL_DATA, kernel, kernel_size); + + fw_cfg_add_i32(fw_cfg, FW_CFG_SETUP_ADDR, real_addr); + fw_cfg_add_i32(fw_cfg, FW_CFG_SETUP_SIZE, setup_size); + fw_cfg_add_bytes(fw_cfg, FW_CFG_SETUP_DATA, setup, setup_size); + + option_rom[nb_option_roms].bootindex = 0; + option_rom[nb_option_roms].name = "linuxboot.bin"; + if (pcmc->linuxboot_dma_enabled && fw_cfg_dma_enabled(fw_cfg)) { + option_rom[nb_option_roms].name = "linuxboot_dma.bin"; + } + nb_option_roms++; +} + +void x86_bios_rom_init(MemoryRegion *rom_memory, bool isapc_ram_fw) +{ + char *filename; + MemoryRegion *bios, *isa_bios; + int bios_size, isa_bios_size; + int ret; + + /* BIOS load */ + if (bios_name == NULL) { + bios_name = BIOS_FILENAME; + } + filename = qemu_find_file(QEMU_FILE_TYPE_BIOS, bios_name); + if (filename) { + bios_size = get_image_size(filename); + } else { + bios_size = -1; + } + if (bios_size <= 0 || + (bios_size % 65536) != 0) { + goto bios_error; + } + bios = g_malloc(sizeof(*bios)); + memory_region_init_ram(bios, NULL, "pc.bios", bios_size, &error_fatal); + if (!isapc_ram_fw) { + memory_region_set_readonly(bios, true); + } + ret = rom_add_file_fixed(bios_name, (uint32_t)(-bios_size), -1); + if (ret != 0) { + bios_error: + fprintf(stderr, "qemu: could not load PC BIOS '%s'\n", bios_name); + exit(1); + } + g_free(filename); + + /* map the last 128KB of the BIOS in ISA space */ + isa_bios_size = MIN(bios_size, 128 * KiB); + isa_bios = g_malloc(sizeof(*isa_bios)); + memory_region_init_alias(isa_bios, NULL, "isa-bios", bios, + bios_size - isa_bios_size, isa_bios_size); + memory_region_add_subregion_overlap(rom_memory, + 0x100000 - isa_bios_size, + isa_bios, + 1); + if (!isapc_ram_fw) { + memory_region_set_readonly(isa_bios, true); + } + + /* map all the bios at the top of memory */ + memory_region_add_subregion(rom_memory, + (uint32_t)(-bios_size), + bios); +} diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h index 7082b06c51..f135ae8c07 100644 --- a/include/hw/i386/pc.h +++ b/include/hw/i386/pc.h @@ -198,7 +198,6 @@ bool pc_machine_is_smm_enabled(PCMachineState *pcms); void pc_register_ferr_irq(qemu_irq irq); void pc_acpi_smi_interrupt(void *opaque, int irq, int level); -void x86_cpus_init(PCMachineState *pcms); void pc_hot_add_cpu(MachineState *ms, const int64_t id, Error **errp); void pc_smp_parse(MachineState *ms, QemuOpts *opts); diff --git a/include/hw/i386/x86.h b/include/hw/i386/x86.h new file mode 100644 index 0000000000..71e2b6985d --- /dev/null +++ b/include/hw/i386/x86.h @@ -0,0 +1,35 @@ +/* + * Copyright (c) 2019 Red Hat, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2 or later, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see . + */ + +#ifndef HW_I386_X86_H +#define HW_I386_X86_H + +#include "hw/boards.h" + +uint32_t x86_cpu_apic_id_from_index(PCMachineState *pcms, + unsigned int cpu_index); +void x86_cpu_new(PCMachineState *pcms, int64_t apic_id, Error **errp); +void x86_cpus_init(PCMachineState *pcms); +CpuInstanceProperties x86_cpu_index_to_props(MachineState *ms, + unsigned cpu_index); +int64_t x86_get_default_cpu_node_id(const MachineState *ms, int idx); +const CPUArchIdList *x86_possible_cpu_arch_ids(MachineState *ms); + +void x86_bios_rom_init(MemoryRegion *rom_memory, bool isapc_ram_fw); + +void x86_load_linux(PCMachineState *x86ms, FWCfgState *fw_cfg); + +#endif -- cgit v1.2.3-55-g7522 From f0bb276bf8d5b3df57697357b802ca76e4cdf05f Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Tue, 22 Oct 2019 09:39:50 +0200 Subject: hw/i386: split PCMachineState deriving X86MachineState from it Split up PCMachineState and PCMachineClass and derive X86MachineState and X86MachineClass from them. This allows sharing code with non-PC x86 machine types. Signed-off-by: Sergio Lopez Reviewed-by: Philippe Mathieu-Daudé Tested-by: Philippe Mathieu-Daudé Reviewed-by: Michael S. Tsirkin --- hw/acpi/cpu_hotplug.c | 10 +-- hw/i386/acpi-build.c | 29 ++++---- hw/i386/amd_iommu.c | 3 +- hw/i386/intel_iommu.c | 3 +- hw/i386/pc.c | 178 +++++++++++++++++--------------------------------- hw/i386/pc_piix.c | 43 ++++++------ hw/i386/pc_q35.c | 35 +++++----- hw/i386/x86.c | 140 ++++++++++++++++++++++++++++++++++----- hw/i386/xen/xen-hvm.c | 28 ++++---- hw/intc/ioapic.c | 2 +- include/hw/i386/pc.h | 27 +------- include/hw/i386/x86.h | 58 +++++++++++++++- 12 files changed, 326 insertions(+), 230 deletions(-) (limited to 'include') diff --git a/hw/acpi/cpu_hotplug.c b/hw/acpi/cpu_hotplug.c index 6e8293aac9..3ac2045a95 100644 --- a/hw/acpi/cpu_hotplug.c +++ b/hw/acpi/cpu_hotplug.c @@ -128,7 +128,7 @@ void build_legacy_cpu_hotplug_aml(Aml *ctx, MachineState *machine, Aml *one = aml_int(1); MachineClass *mc = MACHINE_GET_CLASS(machine); const CPUArchIdList *apic_ids = mc->possible_cpu_arch_ids(machine); - PCMachineState *pcms = PC_MACHINE(machine); + X86MachineState *x86ms = X86_MACHINE(machine); /* * _MAT method - creates an madt apic buffer @@ -236,9 +236,9 @@ void build_legacy_cpu_hotplug_aml(Aml *ctx, MachineState *machine, /* The current AML generator can cover the APIC ID range [0..255], * inclusive, for VCPU hotplug. */ QEMU_BUILD_BUG_ON(ACPI_CPU_HOTPLUG_ID_LIMIT > 256); - if (pcms->apic_id_limit > ACPI_CPU_HOTPLUG_ID_LIMIT) { + if (x86ms->apic_id_limit > ACPI_CPU_HOTPLUG_ID_LIMIT) { error_report("max_cpus is too large. APIC ID of last CPU is %u", - pcms->apic_id_limit - 1); + x86ms->apic_id_limit - 1); exit(1); } @@ -315,8 +315,8 @@ void build_legacy_cpu_hotplug_aml(Aml *ctx, MachineState *machine, * ith up to 255 elements. Windows guests up to win2k8 fail when * VarPackageOp is used. */ - pkg = pcms->apic_id_limit <= 255 ? aml_package(pcms->apic_id_limit) : - aml_varpackage(pcms->apic_id_limit); + pkg = x86ms->apic_id_limit <= 255 ? aml_package(x86ms->apic_id_limit) : + aml_varpackage(x86ms->apic_id_limit); for (i = 0, apic_idx = 0; i < apic_ids->len; i++) { int apic_id = apic_ids->cpus[i].arch_id; diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c index 1d077a7cb7..bc0256e1e0 100644 --- a/hw/i386/acpi-build.c +++ b/hw/i386/acpi-build.c @@ -361,6 +361,7 @@ static void build_madt(GArray *table_data, BIOSLinker *linker, PCMachineState *pcms) { MachineClass *mc = MACHINE_GET_CLASS(pcms); + X86MachineState *x86ms = X86_MACHINE(pcms); const CPUArchIdList *apic_ids = mc->possible_cpu_arch_ids(MACHINE(pcms)); int madt_start = table_data->len; AcpiDeviceIfClass *adevc = ACPI_DEVICE_IF_GET_CLASS(pcms->acpi_dev); @@ -390,7 +391,7 @@ build_madt(GArray *table_data, BIOSLinker *linker, PCMachineState *pcms) io_apic->address = cpu_to_le32(IO_APIC_DEFAULT_ADDRESS); io_apic->interrupt = cpu_to_le32(0); - if (pcms->apic_xrupt_override) { + if (x86ms->apic_xrupt_override) { intsrcovr = acpi_data_push(table_data, sizeof *intsrcovr); intsrcovr->type = ACPI_APIC_XRUPT_OVERRIDE; intsrcovr->length = sizeof(*intsrcovr); @@ -1831,6 +1832,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker, CrsRangeSet crs_range_set; PCMachineState *pcms = PC_MACHINE(machine); PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(machine); + X86MachineState *x86ms = X86_MACHINE(machine); AcpiMcfgInfo mcfg; uint32_t nr_mem = machine->ram_slots; int root_bus_limit = 0xFF; @@ -2103,7 +2105,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker, * with half of the 16-bit control register. Hence, the total size * of the i/o region used is FW_CFG_CTL_SIZE; when using DMA, the * DMA control register is located at FW_CFG_DMA_IO_BASE + 4 */ - uint8_t io_size = object_property_get_bool(OBJECT(pcms->fw_cfg), + uint8_t io_size = object_property_get_bool(OBJECT(x86ms->fw_cfg), "dma_enabled", NULL) ? ROUND_UP(FW_CFG_CTL_SIZE, 4) + sizeof(dma_addr_t) : FW_CFG_CTL_SIZE; @@ -2336,6 +2338,7 @@ build_srat(GArray *table_data, BIOSLinker *linker, MachineState *machine) int srat_start, numa_start, slots; uint64_t mem_len, mem_base, next_base; MachineClass *mc = MACHINE_GET_CLASS(machine); + X86MachineState *x86ms = X86_MACHINE(machine); const CPUArchIdList *apic_ids = mc->possible_cpu_arch_ids(machine); PCMachineState *pcms = PC_MACHINE(machine); ram_addr_t hotplugabble_address_space_size = @@ -2406,16 +2409,16 @@ build_srat(GArray *table_data, BIOSLinker *linker, MachineState *machine) } /* Cut out the ACPI_PCI hole */ - if (mem_base <= pcms->below_4g_mem_size && - next_base > pcms->below_4g_mem_size) { - mem_len -= next_base - pcms->below_4g_mem_size; + if (mem_base <= x86ms->below_4g_mem_size && + next_base > x86ms->below_4g_mem_size) { + mem_len -= next_base - x86ms->below_4g_mem_size; if (mem_len > 0) { numamem = acpi_data_push(table_data, sizeof *numamem); build_srat_memory(numamem, mem_base, mem_len, i - 1, MEM_AFFINITY_ENABLED); } mem_base = 1ULL << 32; - mem_len = next_base - pcms->below_4g_mem_size; + mem_len = next_base - x86ms->below_4g_mem_size; next_base = mem_base + mem_len; } @@ -2634,6 +2637,7 @@ void acpi_build(AcpiBuildTables *tables, MachineState *machine) { PCMachineState *pcms = PC_MACHINE(machine); PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); + X86MachineState *x86ms = X86_MACHINE(machine); GArray *table_offsets; unsigned facs, dsdt, rsdt, fadt; AcpiPmInfo pm; @@ -2795,7 +2799,7 @@ void acpi_build(AcpiBuildTables *tables, MachineState *machine) */ int legacy_aml_len = pcmc->legacy_acpi_table_size + - ACPI_BUILD_LEGACY_CPU_AML_SIZE * pcms->apic_id_limit; + ACPI_BUILD_LEGACY_CPU_AML_SIZE * x86ms->apic_id_limit; int legacy_table_size = ROUND_UP(tables_blob->len - aml_len + legacy_aml_len, ACPI_BUILD_ALIGN_SIZE); @@ -2885,13 +2889,14 @@ void acpi_setup(void) { PCMachineState *pcms = PC_MACHINE(qdev_get_machine()); PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); + X86MachineState *x86ms = X86_MACHINE(pcms); AcpiBuildTables tables; AcpiBuildState *build_state; Object *vmgenid_dev; TPMIf *tpm; static FwCfgTPMConfig tpm_config; - if (!pcms->fw_cfg) { + if (!x86ms->fw_cfg) { ACPI_BUILD_DPRINTF("No fw cfg. Bailing out.\n"); return; } @@ -2922,7 +2927,7 @@ void acpi_setup(void) acpi_add_rom_blob(acpi_build_update, build_state, tables.linker->cmd_blob, "etc/table-loader", 0); - fw_cfg_add_file(pcms->fw_cfg, ACPI_BUILD_TPMLOG_FILE, + fw_cfg_add_file(x86ms->fw_cfg, ACPI_BUILD_TPMLOG_FILE, tables.tcpalog->data, acpi_data_len(tables.tcpalog)); tpm = tpm_find(); @@ -2932,13 +2937,13 @@ void acpi_setup(void) .tpm_version = tpm_get_version(tpm), .tpmppi_version = TPM_PPI_VERSION_1_30 }; - fw_cfg_add_file(pcms->fw_cfg, "etc/tpm/config", + fw_cfg_add_file(x86ms->fw_cfg, "etc/tpm/config", &tpm_config, sizeof tpm_config); } vmgenid_dev = find_vmgenid_dev(); if (vmgenid_dev) { - vmgenid_add_fw_cfg(VMGENID(vmgenid_dev), pcms->fw_cfg, + vmgenid_add_fw_cfg(VMGENID(vmgenid_dev), x86ms->fw_cfg, tables.vmgenid); } @@ -2951,7 +2956,7 @@ void acpi_setup(void) uint32_t rsdp_size = acpi_data_len(tables.rsdp); build_state->rsdp = g_memdup(tables.rsdp->data, rsdp_size); - fw_cfg_add_file_callback(pcms->fw_cfg, ACPI_BUILD_RSDP_FILE, + fw_cfg_add_file_callback(x86ms->fw_cfg, ACPI_BUILD_RSDP_FILE, acpi_build_update, NULL, build_state, build_state->rsdp, rsdp_size, true); build_state->rsdp_mr = NULL; diff --git a/hw/i386/amd_iommu.c b/hw/i386/amd_iommu.c index d3726361dd..d55dbf07fc 100644 --- a/hw/i386/amd_iommu.c +++ b/hw/i386/amd_iommu.c @@ -1540,6 +1540,7 @@ static void amdvi_realize(DeviceState *dev, Error **err) X86IOMMUState *x86_iommu = X86_IOMMU_DEVICE(dev); MachineState *ms = MACHINE(qdev_get_machine()); PCMachineState *pcms = PC_MACHINE(ms); + X86MachineState *x86ms = X86_MACHINE(ms); PCIBus *bus = pcms->bus; s->iotlb = g_hash_table_new_full(amdvi_uint64_hash, @@ -1568,7 +1569,7 @@ static void amdvi_realize(DeviceState *dev, Error **err) } /* Pseudo address space under root PCI bus. */ - pcms->ioapic_as = amdvi_host_dma_iommu(bus, s, AMDVI_IOAPIC_SB_DEVID); + x86ms->ioapic_as = amdvi_host_dma_iommu(bus, s, AMDVI_IOAPIC_SB_DEVID); /* set up MMIO */ memory_region_init_io(&s->mmio, OBJECT(s), &mmio_mem_ops, s, "amdvi-mmio", diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c index 771bed25c9..14e4e6ad62 100644 --- a/hw/i386/intel_iommu.c +++ b/hw/i386/intel_iommu.c @@ -3733,6 +3733,7 @@ static void vtd_realize(DeviceState *dev, Error **errp) { MachineState *ms = MACHINE(qdev_get_machine()); PCMachineState *pcms = PC_MACHINE(ms); + X86MachineState *x86ms = X86_MACHINE(ms); PCIBus *bus = pcms->bus; IntelIOMMUState *s = INTEL_IOMMU_DEVICE(dev); X86IOMMUState *x86_iommu = X86_IOMMU_DEVICE(dev); @@ -3773,7 +3774,7 @@ static void vtd_realize(DeviceState *dev, Error **errp) sysbus_mmio_map(SYS_BUS_DEVICE(s), 0, Q35_HOST_BRIDGE_IOMMU_ADDR); pci_setup_iommu(bus, vtd_host_dma_iommu, dev); /* Pseudo address space under root PCI bus. */ - pcms->ioapic_as = vtd_host_dma_iommu(bus, s, Q35_PSEUDO_DEVFN_IOAPIC); + x86ms->ioapic_as = vtd_host_dma_iommu(bus, s, Q35_PSEUDO_DEVFN_IOAPIC); qemu_add_machine_init_done_notifier(&vtd_machine_done_notify); } diff --git a/hw/i386/pc.c b/hw/i386/pc.c index 4a1bc95ccb..e6ad1f1f36 100644 --- a/hw/i386/pc.c +++ b/hw/i386/pc.c @@ -79,7 +79,6 @@ #include "qapi/qapi-visit-common.h" #include "qapi/visitor.h" #include "hw/core/cpu.h" -#include "hw/nmi.h" #include "hw/usb.h" #include "hw/i386/intel_iommu.h" #include "hw/net/ne2000-isa.h" @@ -680,17 +679,18 @@ void pc_cmos_init(PCMachineState *pcms, { int val; static pc_cmos_init_late_arg arg; + X86MachineState *x86ms = X86_MACHINE(pcms); /* various important CMOS locations needed by PC/Bochs bios */ /* memory size */ /* base memory (first MiB) */ - val = MIN(pcms->below_4g_mem_size / KiB, 640); + val = MIN(x86ms->below_4g_mem_size / KiB, 640); rtc_set_memory(s, 0x15, val); rtc_set_memory(s, 0x16, val >> 8); /* extended memory (next 64MiB) */ - if (pcms->below_4g_mem_size > 1 * MiB) { - val = (pcms->below_4g_mem_size - 1 * MiB) / KiB; + if (x86ms->below_4g_mem_size > 1 * MiB) { + val = (x86ms->below_4g_mem_size - 1 * MiB) / KiB; } else { val = 0; } @@ -701,8 +701,8 @@ void pc_cmos_init(PCMachineState *pcms, rtc_set_memory(s, 0x30, val); rtc_set_memory(s, 0x31, val >> 8); /* memory between 16MiB and 4GiB */ - if (pcms->below_4g_mem_size > 16 * MiB) { - val = (pcms->below_4g_mem_size - 16 * MiB) / (64 * KiB); + if (x86ms->below_4g_mem_size > 16 * MiB) { + val = (x86ms->below_4g_mem_size - 16 * MiB) / (64 * KiB); } else { val = 0; } @@ -711,14 +711,14 @@ void pc_cmos_init(PCMachineState *pcms, rtc_set_memory(s, 0x34, val); rtc_set_memory(s, 0x35, val >> 8); /* memory above 4GiB */ - val = pcms->above_4g_mem_size / 65536; + val = x86ms->above_4g_mem_size / 65536; rtc_set_memory(s, 0x5b, val); rtc_set_memory(s, 0x5c, val >> 8); rtc_set_memory(s, 0x5d, val >> 16); object_property_add_link(OBJECT(pcms), "rtc_state", TYPE_ISA_DEVICE, - (Object **)&pcms->rtc, + (Object **)&x86ms->rtc, object_property_allow_set_link, OBJ_PROP_LINK_STRONG, &error_abort); object_property_set_link(OBJECT(pcms), OBJECT(s), @@ -907,7 +907,7 @@ void pc_acpi_smi_interrupt(void *opaque, int irq, int level) */ void pc_smp_parse(MachineState *ms, QemuOpts *opts) { - PCMachineState *pcms = PC_MACHINE(ms); + X86MachineState *x86ms = X86_MACHINE(ms); if (opts) { unsigned cpus = qemu_opt_get_number(opts, "cpus", 0); @@ -971,7 +971,7 @@ void pc_smp_parse(MachineState *ms, QemuOpts *opts) ms->smp.cpus = cpus; ms->smp.cores = cores; ms->smp.threads = threads; - pcms->smp_dies = dies; + x86ms->smp_dies = dies; } if (ms->smp.cpus > 1) { @@ -1024,10 +1024,11 @@ void pc_machine_done(Notifier *notifier, void *data) { PCMachineState *pcms = container_of(notifier, PCMachineState, machine_done); + X86MachineState *x86ms = X86_MACHINE(pcms); PCIBus *bus = pcms->bus; /* set the number of CPUs */ - rtc_set_cpus_count(pcms->rtc, pcms->boot_cpus); + rtc_set_cpus_count(x86ms->rtc, x86ms->boot_cpus); if (bus) { int extra_hosts = 0; @@ -1038,23 +1039,23 @@ void pc_machine_done(Notifier *notifier, void *data) extra_hosts++; } } - if (extra_hosts && pcms->fw_cfg) { + if (extra_hosts && x86ms->fw_cfg) { uint64_t *val = g_malloc(sizeof(*val)); *val = cpu_to_le64(extra_hosts); - fw_cfg_add_file(pcms->fw_cfg, + fw_cfg_add_file(x86ms->fw_cfg, "etc/extra-pci-roots", val, sizeof(*val)); } } acpi_setup(); - if (pcms->fw_cfg) { - fw_cfg_build_smbios(MACHINE(pcms), pcms->fw_cfg); - fw_cfg_build_feature_control(MACHINE(pcms), pcms->fw_cfg); + if (x86ms->fw_cfg) { + fw_cfg_build_smbios(MACHINE(pcms), x86ms->fw_cfg); + fw_cfg_build_feature_control(MACHINE(pcms), x86ms->fw_cfg); /* update FW_CFG_NB_CPUS to account for -device added CPUs */ - fw_cfg_modify_i16(pcms->fw_cfg, FW_CFG_NB_CPUS, pcms->boot_cpus); + fw_cfg_modify_i16(x86ms->fw_cfg, FW_CFG_NB_CPUS, x86ms->boot_cpus); } - if (pcms->apic_id_limit > 255 && !xen_enabled()) { + if (x86ms->apic_id_limit > 255 && !xen_enabled()) { IntelIOMMUState *iommu = INTEL_IOMMU_DEVICE(x86_iommu_get_default()); if (!iommu || !x86_iommu_ir_supported(X86_IOMMU_DEVICE(iommu)) || @@ -1072,8 +1073,9 @@ void pc_guest_info_init(PCMachineState *pcms) { int i; MachineState *ms = MACHINE(pcms); + X86MachineState *x86ms = X86_MACHINE(pcms); - pcms->apic_xrupt_override = kvm_allows_irq0_override(); + x86ms->apic_xrupt_override = kvm_allows_irq0_override(); pcms->numa_nodes = ms->numa_state->num_nodes; pcms->node_mem = g_malloc0(pcms->numa_nodes * sizeof *pcms->node_mem); @@ -1098,11 +1100,12 @@ void xen_load_linux(PCMachineState *pcms) { int i; FWCfgState *fw_cfg; + X86MachineState *x86ms = X86_MACHINE(pcms); assert(MACHINE(pcms)->kernel_filename != NULL); fw_cfg = fw_cfg_init_io(FW_CFG_IO_BASE); - fw_cfg_add_i16(fw_cfg, FW_CFG_NB_CPUS, pcms->boot_cpus); + fw_cfg_add_i16(fw_cfg, FW_CFG_NB_CPUS, x86ms->boot_cpus); rom_set_fw(fw_cfg); x86_load_linux(pcms, fw_cfg); @@ -1113,7 +1116,7 @@ void xen_load_linux(PCMachineState *pcms) !strcmp(option_rom[i].name, "multiboot.bin")); rom_add_option(option_rom[i].name, option_rom[i].bootindex); } - pcms->fw_cfg = fw_cfg; + x86ms->fw_cfg = fw_cfg; } void pc_memory_init(PCMachineState *pcms, @@ -1128,9 +1131,10 @@ void pc_memory_init(PCMachineState *pcms, MachineState *machine = MACHINE(pcms); MachineClass *mc = MACHINE_GET_CLASS(machine); PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); + X86MachineState *x86ms = X86_MACHINE(pcms); - assert(machine->ram_size == pcms->below_4g_mem_size + - pcms->above_4g_mem_size); + assert(machine->ram_size == x86ms->below_4g_mem_size + + x86ms->above_4g_mem_size); linux_boot = (machine->kernel_filename != NULL); @@ -1144,17 +1148,17 @@ void pc_memory_init(PCMachineState *pcms, *ram_memory = ram; ram_below_4g = g_malloc(sizeof(*ram_below_4g)); memory_region_init_alias(ram_below_4g, NULL, "ram-below-4g", ram, - 0, pcms->below_4g_mem_size); + 0, x86ms->below_4g_mem_size); memory_region_add_subregion(system_memory, 0, ram_below_4g); - e820_add_entry(0, pcms->below_4g_mem_size, E820_RAM); - if (pcms->above_4g_mem_size > 0) { + e820_add_entry(0, x86ms->below_4g_mem_size, E820_RAM); + if (x86ms->above_4g_mem_size > 0) { ram_above_4g = g_malloc(sizeof(*ram_above_4g)); memory_region_init_alias(ram_above_4g, NULL, "ram-above-4g", ram, - pcms->below_4g_mem_size, - pcms->above_4g_mem_size); + x86ms->below_4g_mem_size, + x86ms->above_4g_mem_size); memory_region_add_subregion(system_memory, 0x100000000ULL, ram_above_4g); - e820_add_entry(0x100000000ULL, pcms->above_4g_mem_size, E820_RAM); + e820_add_entry(0x100000000ULL, x86ms->above_4g_mem_size, E820_RAM); } if (!pcmc->has_reserved_memory && @@ -1188,7 +1192,7 @@ void pc_memory_init(PCMachineState *pcms, } machine->device_memory->base = - ROUND_UP(0x100000000ULL + pcms->above_4g_mem_size, 1 * GiB); + ROUND_UP(0x100000000ULL + x86ms->above_4g_mem_size, 1 * GiB); if (pcmc->enforce_aligned_dimm) { /* size device region assuming 1G page max alignment per slot */ @@ -1223,7 +1227,7 @@ void pc_memory_init(PCMachineState *pcms, 1); fw_cfg = fw_cfg_arch_create(machine, - pcms->boot_cpus, pcms->apic_id_limit); + x86ms->boot_cpus, x86ms->apic_id_limit); rom_set_fw(fw_cfg); @@ -1246,10 +1250,10 @@ void pc_memory_init(PCMachineState *pcms, for (i = 0; i < nb_option_roms; i++) { rom_add_option(option_rom[i].name, option_rom[i].bootindex); } - pcms->fw_cfg = fw_cfg; + x86ms->fw_cfg = fw_cfg; /* Init default IOAPIC address space */ - pcms->ioapic_as = &address_space_memory; + x86ms->ioapic_as = &address_space_memory; /* Init ACPI memory hotplug IO base address */ pcms->memhp_io_base = ACPI_MEMORY_HOTPLUG_BASE; @@ -1264,6 +1268,7 @@ uint64_t pc_pci_hole64_start(void) PCMachineState *pcms = PC_MACHINE(qdev_get_machine()); PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); MachineState *ms = MACHINE(pcms); + X86MachineState *x86ms = X86_MACHINE(pcms); uint64_t hole64_start = 0; if (pcmc->has_reserved_memory && ms->device_memory->base) { @@ -1272,7 +1277,7 @@ uint64_t pc_pci_hole64_start(void) hole64_start += memory_region_size(&ms->device_memory->mr); } } else { - hole64_start = 0x100000000ULL + pcms->above_4g_mem_size; + hole64_start = 0x100000000ULL + x86ms->above_4g_mem_size; } return ROUND_UP(hole64_start, 1 * GiB); @@ -1611,6 +1616,7 @@ static void pc_cpu_plug(HotplugHandler *hotplug_dev, Error *local_err = NULL; X86CPU *cpu = X86_CPU(dev); PCMachineState *pcms = PC_MACHINE(hotplug_dev); + X86MachineState *x86ms = X86_MACHINE(pcms); if (pcms->acpi_dev) { hotplug_handler_plug(HOTPLUG_HANDLER(pcms->acpi_dev), dev, &local_err); @@ -1620,12 +1626,12 @@ static void pc_cpu_plug(HotplugHandler *hotplug_dev, } /* increment the number of CPUs */ - pcms->boot_cpus++; - if (pcms->rtc) { - rtc_set_cpus_count(pcms->rtc, pcms->boot_cpus); + x86ms->boot_cpus++; + if (x86ms->rtc) { + rtc_set_cpus_count(x86ms->rtc, x86ms->boot_cpus); } - if (pcms->fw_cfg) { - fw_cfg_modify_i16(pcms->fw_cfg, FW_CFG_NB_CPUS, pcms->boot_cpus); + if (x86ms->fw_cfg) { + fw_cfg_modify_i16(x86ms->fw_cfg, FW_CFG_NB_CPUS, x86ms->boot_cpus); } found_cpu = pc_find_cpu_slot(MACHINE(pcms), cpu->apic_id, NULL); @@ -1671,6 +1677,7 @@ static void pc_cpu_unplug_cb(HotplugHandler *hotplug_dev, Error *local_err = NULL; X86CPU *cpu = X86_CPU(dev); PCMachineState *pcms = PC_MACHINE(hotplug_dev); + X86MachineState *x86ms = X86_MACHINE(pcms); hotplug_handler_unplug(HOTPLUG_HANDLER(pcms->acpi_dev), dev, &local_err); if (local_err) { @@ -1682,10 +1689,10 @@ static void pc_cpu_unplug_cb(HotplugHandler *hotplug_dev, object_property_set_bool(OBJECT(dev), false, "realized", NULL); /* decrement the number of CPUs */ - pcms->boot_cpus--; + x86ms->boot_cpus--; /* Update the number of CPUs in CMOS */ - rtc_set_cpus_count(pcms->rtc, pcms->boot_cpus); - fw_cfg_modify_i16(pcms->fw_cfg, FW_CFG_NB_CPUS, pcms->boot_cpus); + rtc_set_cpus_count(x86ms->rtc, x86ms->boot_cpus); + fw_cfg_modify_i16(x86ms->fw_cfg, FW_CFG_NB_CPUS, x86ms->boot_cpus); out: error_propagate(errp, local_err); } @@ -1701,6 +1708,7 @@ static void pc_cpu_pre_plug(HotplugHandler *hotplug_dev, CPUX86State *env = &cpu->env; MachineState *ms = MACHINE(hotplug_dev); PCMachineState *pcms = PC_MACHINE(hotplug_dev); + X86MachineState *x86ms = X86_MACHINE(pcms); unsigned int smp_cores = ms->smp.cores; unsigned int smp_threads = ms->smp.threads; @@ -1710,7 +1718,7 @@ static void pc_cpu_pre_plug(HotplugHandler *hotplug_dev, return; } - env->nr_dies = pcms->smp_dies; + env->nr_dies = x86ms->smp_dies; /* * If APIC ID is not set, @@ -1718,13 +1726,13 @@ static void pc_cpu_pre_plug(HotplugHandler *hotplug_dev, */ if (cpu->apic_id == UNASSIGNED_APIC_ID) { int max_socket = (ms->smp.max_cpus - 1) / - smp_threads / smp_cores / pcms->smp_dies; + smp_threads / smp_cores / x86ms->smp_dies; /* * die-id was optional in QEMU 4.0 and older, so keep it optional * if there's only one die per socket. */ - if (cpu->die_id < 0 && pcms->smp_dies == 1) { + if (cpu->die_id < 0 && x86ms->smp_dies == 1) { cpu->die_id = 0; } @@ -1739,9 +1747,9 @@ static void pc_cpu_pre_plug(HotplugHandler *hotplug_dev, if (cpu->die_id < 0) { error_setg(errp, "CPU die-id is not set"); return; - } else if (cpu->die_id > pcms->smp_dies - 1) { + } else if (cpu->die_id > x86ms->smp_dies - 1) { error_setg(errp, "Invalid CPU die-id: %u must be in range 0:%u", - cpu->die_id, pcms->smp_dies - 1); + cpu->die_id, x86ms->smp_dies - 1); return; } if (cpu->core_id < 0) { @@ -1765,7 +1773,7 @@ static void pc_cpu_pre_plug(HotplugHandler *hotplug_dev, topo.die_id = cpu->die_id; topo.core_id = cpu->core_id; topo.smt_id = cpu->thread_id; - cpu->apic_id = apicid_from_topo_ids(pcms->smp_dies, smp_cores, + cpu->apic_id = apicid_from_topo_ids(x86ms->smp_dies, smp_cores, smp_threads, &topo); } @@ -1773,7 +1781,7 @@ static void pc_cpu_pre_plug(HotplugHandler *hotplug_dev, if (!cpu_slot) { MachineState *ms = MACHINE(pcms); - x86_topo_ids_from_apicid(cpu->apic_id, pcms->smp_dies, + x86_topo_ids_from_apicid(cpu->apic_id, x86ms->smp_dies, smp_cores, smp_threads, &topo); error_setg(errp, "Invalid CPU [socket: %u, die: %u, core: %u, thread: %u] with" @@ -1795,7 +1803,7 @@ static void pc_cpu_pre_plug(HotplugHandler *hotplug_dev, /* TODO: move socket_id/core_id/thread_id checks into x86_cpu_realizefn() * once -smp refactoring is complete and there will be CPU private * CPUState::nr_cores and CPUState::nr_threads fields instead of globals */ - x86_topo_ids_from_apicid(cpu->apic_id, pcms->smp_dies, + x86_topo_ids_from_apicid(cpu->apic_id, x86ms->smp_dies, smp_cores, smp_threads, &topo); if (cpu->socket_id != -1 && cpu->socket_id != topo.pkg_id) { error_setg(errp, "property socket-id: %u doesn't match set apic-id:" @@ -1977,45 +1985,6 @@ pc_machine_get_device_memory_region_size(Object *obj, Visitor *v, visit_type_int(v, name, &value, errp); } -static void pc_machine_get_max_ram_below_4g(Object *obj, Visitor *v, - const char *name, void *opaque, - Error **errp) -{ - PCMachineState *pcms = PC_MACHINE(obj); - uint64_t value = pcms->max_ram_below_4g; - - visit_type_size(v, name, &value, errp); -} - -static void pc_machine_set_max_ram_below_4g(Object *obj, Visitor *v, - const char *name, void *opaque, - Error **errp) -{ - PCMachineState *pcms = PC_MACHINE(obj); - Error *error = NULL; - uint64_t value; - - visit_type_size(v, name, &value, &error); - if (error) { - error_propagate(errp, error); - return; - } - if (value > 4 * GiB) { - error_setg(&error, - "Machine option 'max-ram-below-4g=%"PRIu64 - "' expects size less than or equal to 4G", value); - error_propagate(errp, error); - return; - } - - if (value < 1 * MiB) { - warn_report("Only %" PRIu64 " bytes of RAM below the 4GiB boundary," - "BIOS may not work with less than 1MiB", value); - } - - pcms->max_ram_below_4g = value; -} - static void pc_machine_get_vmport(Object *obj, Visitor *v, const char *name, void *opaque, Error **errp) { @@ -2121,7 +2090,6 @@ static void pc_machine_initfn(Object *obj) { PCMachineState *pcms = PC_MACHINE(obj); - pcms->max_ram_below_4g = 0; /* use default */ pcms->smm = ON_OFF_AUTO_AUTO; #ifdef CONFIG_VMPORT pcms->vmport = ON_OFF_AUTO_AUTO; @@ -2133,7 +2101,6 @@ static void pc_machine_initfn(Object *obj) pcms->smbus_enabled = true; pcms->sata_enabled = true; pcms->pit_enabled = true; - pcms->smp_dies = 1; pc_system_flash_create(pcms); } @@ -2164,23 +2131,6 @@ static void pc_machine_wakeup(MachineState *machine) cpu_synchronize_all_post_reset(); } -static void x86_nmi(NMIState *n, int cpu_index, Error **errp) -{ - /* cpu index isn't used */ - CPUState *cs; - - CPU_FOREACH(cs) { - X86CPU *cpu = X86_CPU(cs); - - if (!cpu->apic_state) { - cpu_interrupt(cs, CPU_INTERRUPT_NMI); - } else { - apic_deliver_nmi(cpu->apic_state); - } - } -} - - static bool pc_hotplug_allowed(MachineState *ms, DeviceState *dev, Error **errp) { X86IOMMUState *iommu = x86_iommu_get_default(); @@ -2205,7 +2155,6 @@ static void pc_machine_class_init(ObjectClass *oc, void *data) MachineClass *mc = MACHINE_CLASS(oc); PCMachineClass *pcmc = PC_MACHINE_CLASS(oc); HotplugHandlerClass *hc = HOTPLUG_HANDLER_CLASS(oc); - NMIClass *nc = NMI_CLASS(oc); pcmc->pci_enabled = true; pcmc->has_acpi_build = true; @@ -2241,7 +2190,6 @@ static void pc_machine_class_init(ObjectClass *oc, void *data) hc->plug = pc_machine_device_plug_cb; hc->unplug_request = pc_machine_device_unplug_request_cb; hc->unplug = pc_machine_device_unplug_cb; - nc->nmi_monitor_handler = x86_nmi; mc->default_cpu_type = TARGET_DEFAULT_CPU_TYPE; mc->nvdimm_supported = true; mc->numa_mem_supported = true; @@ -2250,13 +2198,6 @@ static void pc_machine_class_init(ObjectClass *oc, void *data) pc_machine_get_device_memory_region_size, NULL, NULL, NULL, &error_abort); - object_class_property_add(oc, PC_MACHINE_MAX_RAM_BELOW_4G, "size", - pc_machine_get_max_ram_below_4g, pc_machine_set_max_ram_below_4g, - NULL, NULL, &error_abort); - - object_class_property_set_description(oc, PC_MACHINE_MAX_RAM_BELOW_4G, - "Maximum ram below the 4G boundary (32bit boundary)", &error_abort); - object_class_property_add(oc, PC_MACHINE_SMM, "OnOffAuto", pc_machine_get_smm, pc_machine_set_smm, NULL, NULL, &error_abort); @@ -2281,7 +2222,7 @@ static void pc_machine_class_init(ObjectClass *oc, void *data) static const TypeInfo pc_machine_info = { .name = TYPE_PC_MACHINE, - .parent = TYPE_MACHINE, + .parent = TYPE_X86_MACHINE, .abstract = true, .instance_size = sizeof(PCMachineState), .instance_init = pc_machine_initfn, @@ -2289,7 +2230,6 @@ static const TypeInfo pc_machine_info = { .class_init = pc_machine_class_init, .interfaces = (InterfaceInfo[]) { { TYPE_HOTPLUG_HANDLER }, - { TYPE_NMI }, { } }, }; diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c index 1396451abf..0afa8fe6ea 100644 --- a/hw/i386/pc_piix.c +++ b/hw/i386/pc_piix.c @@ -74,6 +74,7 @@ static void pc_init1(MachineState *machine, { PCMachineState *pcms = PC_MACHINE(machine); PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); + X86MachineState *x86ms = X86_MACHINE(machine); MemoryRegion *system_memory = get_system_memory(); MemoryRegion *system_io = get_system_io(); int i; @@ -126,11 +127,11 @@ static void pc_init1(MachineState *machine, if (xen_enabled()) { xen_hvm_init(pcms, &ram_memory); } else { - if (!pcms->max_ram_below_4g) { - pcms->max_ram_below_4g = 0xe0000000; /* default: 3.5G */ + if (!x86ms->max_ram_below_4g) { + x86ms->max_ram_below_4g = 0xe0000000; /* default: 3.5G */ } - lowmem = pcms->max_ram_below_4g; - if (machine->ram_size >= pcms->max_ram_below_4g) { + lowmem = x86ms->max_ram_below_4g; + if (machine->ram_size >= x86ms->max_ram_below_4g) { if (pcmc->gigabyte_align) { if (lowmem > 0xc0000000) { lowmem = 0xc0000000; @@ -139,17 +140,17 @@ static void pc_init1(MachineState *machine, warn_report("Large machine and max_ram_below_4g " "(%" PRIu64 ") not a multiple of 1G; " "possible bad performance.", - pcms->max_ram_below_4g); + x86ms->max_ram_below_4g); } } } if (machine->ram_size >= lowmem) { - pcms->above_4g_mem_size = machine->ram_size - lowmem; - pcms->below_4g_mem_size = lowmem; + x86ms->above_4g_mem_size = machine->ram_size - lowmem; + x86ms->below_4g_mem_size = lowmem; } else { - pcms->above_4g_mem_size = 0; - pcms->below_4g_mem_size = machine->ram_size; + x86ms->above_4g_mem_size = 0; + x86ms->below_4g_mem_size = machine->ram_size; } } @@ -191,19 +192,19 @@ static void pc_init1(MachineState *machine, gsi_state = g_malloc0(sizeof(*gsi_state)); if (kvm_ioapic_in_kernel()) { kvm_pc_setup_irq_routing(pcmc->pci_enabled); - pcms->gsi = qemu_allocate_irqs(kvm_pc_gsi_handler, gsi_state, + x86ms->gsi = qemu_allocate_irqs(kvm_pc_gsi_handler, gsi_state, GSI_NUM_PINS); } else { - pcms->gsi = qemu_allocate_irqs(gsi_handler, gsi_state, GSI_NUM_PINS); + x86ms->gsi = qemu_allocate_irqs(gsi_handler, gsi_state, GSI_NUM_PINS); } if (pcmc->pci_enabled) { pci_bus = i440fx_init(host_type, pci_type, - &i440fx_state, &piix3_devfn, &isa_bus, pcms->gsi, + &i440fx_state, &piix3_devfn, &isa_bus, x86ms->gsi, system_memory, system_io, machine->ram_size, - pcms->below_4g_mem_size, - pcms->above_4g_mem_size, + x86ms->below_4g_mem_size, + x86ms->above_4g_mem_size, pci_memory, ram_memory); pcms->bus = pci_bus; } else { @@ -213,7 +214,7 @@ static void pc_init1(MachineState *machine, &error_abort); no_hpet = 1; } - isa_bus_irqs(isa_bus, pcms->gsi); + isa_bus_irqs(isa_bus, x86ms->gsi); if (kvm_pic_in_kernel()) { i8259 = kvm_i8259_init(isa_bus); @@ -231,7 +232,7 @@ static void pc_init1(MachineState *machine, ioapic_init_gsi(gsi_state, "i440fx"); } - pc_register_ferr_irq(pcms->gsi[13]); + pc_register_ferr_irq(x86ms->gsi[13]); pc_vga_init(isa_bus, pcmc->pci_enabled ? pci_bus : NULL); @@ -241,7 +242,7 @@ static void pc_init1(MachineState *machine, } /* init basic PC hardware */ - pc_basic_device_init(isa_bus, pcms->gsi, &rtc_state, true, + pc_basic_device_init(isa_bus, x86ms->gsi, &rtc_state, true, (pcms->vmport != ON_OFF_AUTO_ON), pcms->pit_enabled, 0x4); @@ -288,7 +289,7 @@ else { smi_irq = qemu_allocate_irq(pc_acpi_smi_interrupt, first_cpu, 0); /* TODO: Populate SPD eeprom data. */ pcms->smbus = piix4_pm_init(pci_bus, piix3_devfn + 3, 0xb100, - pcms->gsi[9], smi_irq, + x86ms->gsi[9], smi_irq, pc_machine_is_smm_enabled(pcms), &piix4_pm); smbus_eeprom_init(pcms->smbus, 8, NULL, 0); @@ -304,7 +305,7 @@ else { if (machine->nvdimms_state->is_enabled) { nvdimm_init_acpi_state(machine->nvdimms_state, system_io, - pcms->fw_cfg, OBJECT(pcms)); + x86ms->fw_cfg, OBJECT(pcms)); } } @@ -729,7 +730,7 @@ DEFINE_I440FX_MACHINE(v1_4, "pc-i440fx-1.4", pc_compat_1_4_fn, static void pc_i440fx_1_3_machine_options(MachineClass *m) { - PCMachineClass *pcmc = PC_MACHINE_CLASS(m); + X86MachineClass *x86mc = X86_MACHINE_CLASS(m); static GlobalProperty compat[] = { PC_CPU_MODEL_IDS("1.3.0") { "usb-tablet", "usb_version", "1" }, @@ -740,7 +741,7 @@ static void pc_i440fx_1_3_machine_options(MachineClass *m) pc_i440fx_1_4_machine_options(m); m->hw_version = "1.3.0"; - pcmc->compat_apic_id_mode = true; + x86mc->compat_apic_id_mode = true; compat_props_add(m->compat_props, compat, G_N_ELEMENTS(compat)); } diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c index 8920bd8978..374ac6c068 100644 --- a/hw/i386/pc_q35.c +++ b/hw/i386/pc_q35.c @@ -116,6 +116,7 @@ static void pc_q35_init(MachineState *machine) { PCMachineState *pcms = PC_MACHINE(machine); PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); + X86MachineState *x86ms = X86_MACHINE(machine); Q35PCIHost *q35_host; PCIHostState *phb; PCIBus *host_bus; @@ -153,27 +154,27 @@ static void pc_q35_init(MachineState *machine) /* Handle the machine opt max-ram-below-4g. It is basically doing * min(qemu limit, user limit). */ - if (!pcms->max_ram_below_4g) { - pcms->max_ram_below_4g = 1ULL << 32; /* default: 4G */; + if (!x86ms->max_ram_below_4g) { + x86ms->max_ram_below_4g = 4 * GiB; } - if (lowmem > pcms->max_ram_below_4g) { - lowmem = pcms->max_ram_below_4g; + if (lowmem > x86ms->max_ram_below_4g) { + lowmem = x86ms->max_ram_below_4g; if (machine->ram_size - lowmem > lowmem && lowmem & (1 * GiB - 1)) { warn_report("There is possibly poor performance as the ram size " " (0x%" PRIx64 ") is more then twice the size of" " max-ram-below-4g (%"PRIu64") and" " max-ram-below-4g is not a multiple of 1G.", - (uint64_t)machine->ram_size, pcms->max_ram_below_4g); + (uint64_t)machine->ram_size, x86ms->max_ram_below_4g); } } if (machine->ram_size >= lowmem) { - pcms->above_4g_mem_size = machine->ram_size - lowmem; - pcms->below_4g_mem_size = lowmem; + x86ms->above_4g_mem_size = machine->ram_size - lowmem; + x86ms->below_4g_mem_size = lowmem; } else { - pcms->above_4g_mem_size = 0; - pcms->below_4g_mem_size = machine->ram_size; + x86ms->above_4g_mem_size = 0; + x86ms->below_4g_mem_size = machine->ram_size; } if (xen_enabled()) { @@ -214,10 +215,10 @@ static void pc_q35_init(MachineState *machine) gsi_state = g_malloc0(sizeof(*gsi_state)); if (kvm_ioapic_in_kernel()) { kvm_pc_setup_irq_routing(pcmc->pci_enabled); - pcms->gsi = qemu_allocate_irqs(kvm_pc_gsi_handler, gsi_state, + x86ms->gsi = qemu_allocate_irqs(kvm_pc_gsi_handler, gsi_state, GSI_NUM_PINS); } else { - pcms->gsi = qemu_allocate_irqs(gsi_handler, gsi_state, GSI_NUM_PINS); + x86ms->gsi = qemu_allocate_irqs(gsi_handler, gsi_state, GSI_NUM_PINS); } /* create pci host bus */ @@ -232,9 +233,9 @@ static void pc_q35_init(MachineState *machine) MCH_HOST_PROP_SYSTEM_MEM, NULL); object_property_set_link(OBJECT(q35_host), OBJECT(system_io), MCH_HOST_PROP_IO_MEM, NULL); - object_property_set_int(OBJECT(q35_host), pcms->below_4g_mem_size, + object_property_set_int(OBJECT(q35_host), x86ms->below_4g_mem_size, PCI_HOST_BELOW_4G_MEM_SIZE, NULL); - object_property_set_int(OBJECT(q35_host), pcms->above_4g_mem_size, + object_property_set_int(OBJECT(q35_host), x86ms->above_4g_mem_size, PCI_HOST_ABOVE_4G_MEM_SIZE, NULL); /* pci */ qdev_init_nofail(DEVICE(q35_host)); @@ -256,7 +257,7 @@ static void pc_q35_init(MachineState *machine) ich9_lpc = ICH9_LPC_DEVICE(lpc); lpc_dev = DEVICE(lpc); for (i = 0; i < GSI_NUM_PINS; i++) { - qdev_connect_gpio_out_named(lpc_dev, ICH9_GPIO_GSI, i, pcms->gsi[i]); + qdev_connect_gpio_out_named(lpc_dev, ICH9_GPIO_GSI, i, x86ms->gsi[i]); } pci_bus_irqs(host_bus, ich9_lpc_set_irq, ich9_lpc_map_irq, ich9_lpc, ICH9_LPC_NB_PIRQS); @@ -280,7 +281,7 @@ static void pc_q35_init(MachineState *machine) ioapic_init_gsi(gsi_state, "q35"); } - pc_register_ferr_irq(pcms->gsi[13]); + pc_register_ferr_irq(x86ms->gsi[13]); assert(pcms->vmport != ON_OFF_AUTO__MAX); if (pcms->vmport == ON_OFF_AUTO_AUTO) { @@ -288,7 +289,7 @@ static void pc_q35_init(MachineState *machine) } /* init basic PC hardware */ - pc_basic_device_init(isa_bus, pcms->gsi, &rtc_state, !mc->no_floppy, + pc_basic_device_init(isa_bus, x86ms->gsi, &rtc_state, !mc->no_floppy, (pcms->vmport != ON_OFF_AUTO_ON), pcms->pit_enabled, 0xff0104); @@ -331,7 +332,7 @@ static void pc_q35_init(MachineState *machine) if (machine->nvdimms_state->is_enabled) { nvdimm_init_acpi_state(machine->nvdimms_state, system_io, - pcms->fw_cfg, OBJECT(pcms)); + x86ms->fw_cfg, OBJECT(pcms)); } } diff --git a/hw/i386/x86.c b/hw/i386/x86.c index 71fb093983..de4fed0164 100644 --- a/hw/i386/x86.c +++ b/hw/i386/x86.c @@ -65,13 +65,14 @@ uint32_t x86_cpu_apic_id_from_index(PCMachineState *pcms, unsigned int cpu_index) { MachineState *ms = MACHINE(pcms); - PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); + X86MachineState *x86ms = X86_MACHINE(pcms); + X86MachineClass *x86mc = X86_MACHINE_GET_CLASS(x86ms); uint32_t correct_id; static bool warned; - correct_id = x86_apicid_from_cpu_idx(pcms->smp_dies, ms->smp.cores, + correct_id = x86_apicid_from_cpu_idx(x86ms->smp_dies, ms->smp.cores, ms->smp.threads, cpu_index); - if (pcmc->compat_apic_id_mode) { + if (x86mc->compat_apic_id_mode) { if (cpu_index != correct_id && !warned && !qtest_enabled()) { error_report("APIC IDs set in compatibility mode, " "CPU topology won't match the configuration"); @@ -88,11 +89,12 @@ void x86_cpu_new(PCMachineState *pcms, int64_t apic_id, Error **errp) Object *cpu = NULL; Error *local_err = NULL; CPUX86State *env = NULL; + X86MachineState *x86ms = X86_MACHINE(pcms); cpu = object_new(MACHINE(pcms)->cpu_type); env = &X86_CPU(cpu)->env; - env->nr_dies = pcms->smp_dies; + env->nr_dies = x86ms->smp_dies; object_property_set_uint(cpu, apic_id, "apic-id", &local_err); object_property_set_bool(cpu, true, "realized", &local_err); @@ -108,6 +110,7 @@ void x86_cpus_init(PCMachineState *pcms) MachineState *ms = MACHINE(pcms); MachineClass *mc = MACHINE_GET_CLASS(pcms); PCMachineClass *pcmc = PC_MACHINE_CLASS(mc); + X86MachineState *x86ms = X86_MACHINE(pcms); x86_cpu_set_default_version(pcmc->default_cpu_version); @@ -119,8 +122,8 @@ void x86_cpus_init(PCMachineState *pcms) * * This is used for FW_CFG_MAX_CPUS. See comments on fw_cfg_arch_create(). */ - pcms->apic_id_limit = x86_cpu_apic_id_from_index(pcms, - ms->smp.max_cpus - 1) + 1; + x86ms->apic_id_limit = x86_cpu_apic_id_from_index(pcms, + ms->smp.max_cpus - 1) + 1; possible_cpus = mc->possible_cpu_arch_ids(ms); for (i = 0; i < ms->smp.cpus; i++) { x86_cpu_new(pcms, possible_cpus->cpus[i].arch_id, &error_fatal); @@ -140,11 +143,11 @@ x86_cpu_index_to_props(MachineState *ms, unsigned cpu_index) int64_t x86_get_default_cpu_node_id(const MachineState *ms, int idx) { X86CPUTopoInfo topo; - PCMachineState *pcms = PC_MACHINE(ms); + X86MachineState *x86ms = X86_MACHINE(ms); assert(idx < ms->possible_cpus->len); x86_topo_ids_from_apicid(ms->possible_cpus->cpus[idx].arch_id, - pcms->smp_dies, ms->smp.cores, + x86ms->smp_dies, ms->smp.cores, ms->smp.threads, &topo); return topo.pkg_id % ms->numa_state->num_nodes; } @@ -152,6 +155,7 @@ int64_t x86_get_default_cpu_node_id(const MachineState *ms, int idx) const CPUArchIdList *x86_possible_cpu_arch_ids(MachineState *ms) { PCMachineState *pcms = PC_MACHINE(ms); + X86MachineState *x86ms = X86_MACHINE(ms); int i; unsigned int max_cpus = ms->smp.max_cpus; @@ -175,11 +179,11 @@ const CPUArchIdList *x86_possible_cpu_arch_ids(MachineState *ms) ms->possible_cpus->cpus[i].arch_id = x86_cpu_apic_id_from_index(pcms, i); x86_topo_ids_from_apicid(ms->possible_cpus->cpus[i].arch_id, - pcms->smp_dies, ms->smp.cores, + x86ms->smp_dies, ms->smp.cores, ms->smp.threads, &topo); ms->possible_cpus->cpus[i].props.has_socket_id = true; ms->possible_cpus->cpus[i].props.socket_id = topo.pkg_id; - if (pcms->smp_dies > 1) { + if (x86ms->smp_dies > 1) { ms->possible_cpus->cpus[i].props.has_die_id = true; ms->possible_cpus->cpus[i].props.die_id = topo.die_id; } @@ -191,6 +195,22 @@ const CPUArchIdList *x86_possible_cpu_arch_ids(MachineState *ms) return ms->possible_cpus; } +static void x86_nmi(NMIState *n, int cpu_index, Error **errp) +{ + /* cpu index isn't used */ + CPUState *cs; + + CPU_FOREACH(cs) { + X86CPU *cpu = X86_CPU(cs); + + if (!cpu->apic_state) { + cpu_interrupt(cs, CPU_INTERRUPT_NMI); + } else { + apic_deliver_nmi(cpu->apic_state); + } + } +} + static long get_file_size(FILE *f) { long where, size; @@ -328,6 +348,7 @@ void x86_load_linux(PCMachineState *pcms, char *vmode; MachineState *machine = MACHINE(pcms); PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); + X86MachineState *x86ms = X86_MACHINE(pcms); struct setup_data *setup_data; const char *kernel_filename = machine->kernel_filename; const char *initrd_filename = machine->initrd_filename; @@ -400,11 +421,12 @@ void x86_load_linux(PCMachineState *pcms, initrd_filename, gerr->message); exit(1); } - pcms->initrd_mapped_file = mapped_file; + x86ms->initrd_mapped_file = mapped_file; initrd_data = g_mapped_file_get_contents(mapped_file); initrd_size = g_mapped_file_get_length(mapped_file); - initrd_max = pcms->below_4g_mem_size - pcmc->acpi_data_size - 1; + initrd_max = + x86ms->below_4g_mem_size - pcmc->acpi_data_size - 1; if (initrd_size >= initrd_max) { fprintf(stderr, "qemu: initrd is too large, cannot support." "(max: %"PRIu32", need %"PRId64")\n", @@ -472,8 +494,8 @@ void x86_load_linux(PCMachineState *pcms, initrd_max = 0x37ffffff; } - if (initrd_max >= pcms->below_4g_mem_size - pcmc->acpi_data_size) { - initrd_max = pcms->below_4g_mem_size - pcmc->acpi_data_size - 1; + if (initrd_max >= x86ms->below_4g_mem_size - pcmc->acpi_data_size) { + initrd_max = x86ms->below_4g_mem_size - pcmc->acpi_data_size - 1; } fw_cfg_add_i32(fw_cfg, FW_CFG_CMDLINE_ADDR, cmdline_addr); @@ -544,7 +566,7 @@ void x86_load_linux(PCMachineState *pcms, initrd_filename, gerr->message); exit(1); } - pcms->initrd_mapped_file = mapped_file; + x86ms->initrd_mapped_file = mapped_file; initrd_data = g_mapped_file_get_contents(mapped_file); initrd_size = g_mapped_file_get_length(mapped_file); @@ -688,3 +710,91 @@ void x86_bios_rom_init(MemoryRegion *rom_memory, bool isapc_ram_fw) (uint32_t)(-bios_size), bios); } + +static void x86_machine_get_max_ram_below_4g(Object *obj, Visitor *v, + const char *name, void *opaque, + Error **errp) +{ + X86MachineState *x86ms = X86_MACHINE(obj); + uint64_t value = x86ms->max_ram_below_4g; + + visit_type_size(v, name, &value, errp); +} + +static void x86_machine_set_max_ram_below_4g(Object *obj, Visitor *v, + const char *name, void *opaque, + Error **errp) +{ + X86MachineState *x86ms = X86_MACHINE(obj); + Error *error = NULL; + uint64_t value; + + visit_type_size(v, name, &value, &error); + if (error) { + error_propagate(errp, error); + return; + } + if (value > 4 * GiB) { + error_setg(&error, + "Machine option 'max-ram-below-4g=%"PRIu64 + "' expects size less than or equal to 4G", value); + error_propagate(errp, error); + return; + } + + if (value < 1 * MiB) { + warn_report("Only %" PRIu64 " bytes of RAM below the 4GiB boundary," + "BIOS may not work with less than 1MiB", value); + } + + x86ms->max_ram_below_4g = value; +} + +static void x86_machine_initfn(Object *obj) +{ + X86MachineState *x86ms = X86_MACHINE(obj); + + x86ms->max_ram_below_4g = 0; /* use default */ + x86ms->smp_dies = 1; +} + +static void x86_machine_class_init(ObjectClass *oc, void *data) +{ + MachineClass *mc = MACHINE_CLASS(oc); + X86MachineClass *x86mc = X86_MACHINE_CLASS(oc); + NMIClass *nc = NMI_CLASS(oc); + + mc->cpu_index_to_instance_props = x86_cpu_index_to_props; + mc->get_default_cpu_node_id = x86_get_default_cpu_node_id; + mc->possible_cpu_arch_ids = x86_possible_cpu_arch_ids; + x86mc->compat_apic_id_mode = false; + nc->nmi_monitor_handler = x86_nmi; + + object_class_property_add(oc, X86_MACHINE_MAX_RAM_BELOW_4G, "size", + x86_machine_get_max_ram_below_4g, x86_machine_set_max_ram_below_4g, + NULL, NULL, &error_abort); + + object_class_property_set_description(oc, X86_MACHINE_MAX_RAM_BELOW_4G, + "Maximum ram below the 4G boundary (32bit boundary)", &error_abort); +} + +static const TypeInfo x86_machine_info = { + .name = TYPE_X86_MACHINE, + .parent = TYPE_MACHINE, + .abstract = true, + .instance_size = sizeof(X86MachineState), + .instance_init = x86_machine_initfn, + .class_size = sizeof(X86MachineClass), + .class_init = x86_machine_class_init, + .interfaces = (InterfaceInfo[]) { + { TYPE_NMI }, + { } + }, +}; + +static void x86_machine_register_types(void) +{ + type_register_static(&x86_machine_info); +} + +type_init(x86_machine_register_types) diff --git a/hw/i386/xen/xen-hvm.c b/hw/i386/xen/xen-hvm.c index 6b5e5bb7f5..95f23a263c 100644 --- a/hw/i386/xen/xen-hvm.c +++ b/hw/i386/xen/xen-hvm.c @@ -197,11 +197,13 @@ qemu_irq *xen_interrupt_controller_init(void) static void xen_ram_init(PCMachineState *pcms, ram_addr_t ram_size, MemoryRegion **ram_memory_p) { + X86MachineState *x86ms = X86_MACHINE(pcms); MemoryRegion *sysmem = get_system_memory(); ram_addr_t block_len; - uint64_t user_lowmem = object_property_get_uint(qdev_get_machine(), - PC_MACHINE_MAX_RAM_BELOW_4G, - &error_abort); + uint64_t user_lowmem = + object_property_get_uint(qdev_get_machine(), + X86_MACHINE_MAX_RAM_BELOW_4G, + &error_abort); /* Handle the machine opt max-ram-below-4g. It is basically doing * min(xen limit, user limit). @@ -214,20 +216,20 @@ static void xen_ram_init(PCMachineState *pcms, } if (ram_size >= user_lowmem) { - pcms->above_4g_mem_size = ram_size - user_lowmem; - pcms->below_4g_mem_size = user_lowmem; + x86ms->above_4g_mem_size = ram_size - user_lowmem; + x86ms->below_4g_mem_size = user_lowmem; } else { - pcms->above_4g_mem_size = 0; - pcms->below_4g_mem_size = ram_size; + x86ms->above_4g_mem_size = 0; + x86ms->below_4g_mem_size = ram_size; } - if (!pcms->above_4g_mem_size) { + if (!x86ms->above_4g_mem_size) { block_len = ram_size; } else { /* * Xen does not allocate the memory continuously, it keeps a * hole of the size computed above or passed in. */ - block_len = (1ULL << 32) + pcms->above_4g_mem_size; + block_len = (1ULL << 32) + x86ms->above_4g_mem_size; } memory_region_init_ram(&ram_memory, NULL, "xen.ram", block_len, &error_fatal); @@ -244,12 +246,12 @@ static void xen_ram_init(PCMachineState *pcms, */ memory_region_init_alias(&ram_lo, NULL, "xen.ram.lo", &ram_memory, 0xc0000, - pcms->below_4g_mem_size - 0xc0000); + x86ms->below_4g_mem_size - 0xc0000); memory_region_add_subregion(sysmem, 0xc0000, &ram_lo); - if (pcms->above_4g_mem_size > 0) { + if (x86ms->above_4g_mem_size > 0) { memory_region_init_alias(&ram_hi, NULL, "xen.ram.hi", &ram_memory, 0x100000000ULL, - pcms->above_4g_mem_size); + x86ms->above_4g_mem_size); memory_region_add_subregion(sysmem, 0x100000000ULL, &ram_hi); } } @@ -265,7 +267,7 @@ void xen_ram_alloc(ram_addr_t ram_addr, ram_addr_t size, MemoryRegion *mr, /* RAM already populated in Xen */ fprintf(stderr, "%s: do not alloc "RAM_ADDR_FMT " bytes of ram at "RAM_ADDR_FMT" when runstate is INMIGRATE\n", - __func__, size, ram_addr); + __func__, size, ram_addr); return; } diff --git a/hw/intc/ioapic.c b/hw/intc/ioapic.c index 1ede055387..ead14e1888 100644 --- a/hw/intc/ioapic.c +++ b/hw/intc/ioapic.c @@ -89,7 +89,7 @@ static void ioapic_entry_parse(uint64_t entry, struct ioapic_entry_info *info) static void ioapic_service(IOAPICCommonState *s) { - AddressSpace *ioapic_as = PC_MACHINE(qdev_get_machine())->ioapic_as; + AddressSpace *ioapic_as = X86_MACHINE(qdev_get_machine())->ioapic_as; struct ioapic_entry_info info; uint8_t i; uint32_t mask; diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h index f135ae8c07..13c4eac0f6 100644 --- a/include/hw/i386/pc.h +++ b/include/hw/i386/pc.h @@ -8,6 +8,7 @@ #include "hw/block/flash.h" #include "net/net.h" #include "hw/i386/ioapic.h" +#include "hw/i386/x86.h" #include "qemu/range.h" #include "qemu/bitmap.h" @@ -27,7 +28,7 @@ */ struct PCMachineState { /*< private >*/ - MachineState parent_obj; + X86MachineState parent_obj; /* */ @@ -36,16 +37,11 @@ struct PCMachineState { /* Pointers to devices and objects: */ HotplugHandler *acpi_dev; - ISADevice *rtc; PCIBus *bus; I2CBus *smbus; - FWCfgState *fw_cfg; - qemu_irq *gsi; PFlashCFI01 *flash[2]; - GMappedFile *initrd_mapped_file; /* Configuration options: */ - uint64_t max_ram_below_4g; OnOffAuto vmport; OnOffAuto smm; @@ -54,30 +50,16 @@ struct PCMachineState { bool sata_enabled; bool pit_enabled; - /* RAM information (sizes, addresses, configuration): */ - ram_addr_t below_4g_mem_size, above_4g_mem_size; - - /* CPU and apic information: */ - bool apic_xrupt_override; - unsigned apic_id_limit; - uint16_t boot_cpus; - unsigned smp_dies; - /* NUMA information: */ uint64_t numa_nodes; uint64_t *node_mem; - /* Address space used by IOAPIC device. All IOAPIC interrupts - * will be translated to MSI messages in the address space. */ - AddressSpace *ioapic_as; - /* ACPI Memory hotplug IO base address */ hwaddr memhp_io_base; }; #define PC_MACHINE_ACPI_DEVICE_PROP "acpi-device" #define PC_MACHINE_DEVMEM_REGION_SIZE "device-memory-region-size" -#define PC_MACHINE_MAX_RAM_BELOW_4G "max-ram-below-4g" #define PC_MACHINE_VMPORT "vmport" #define PC_MACHINE_SMM "smm" #define PC_MACHINE_SMBUS "smbus" @@ -102,7 +84,7 @@ struct PCMachineState { */ typedef struct PCMachineClass { /*< private >*/ - MachineClass parent_class; + X86MachineClass parent_class; /*< public >*/ @@ -144,9 +126,6 @@ typedef struct PCMachineClass { /* use PVH to load kernels that support this feature */ bool pvh_enabled; - - /* Enables contiguous-apic-ID mode */ - bool compat_apic_id_mode; } PCMachineClass; #define TYPE_PC_MACHINE "generic-pc-machine" diff --git a/include/hw/i386/x86.h b/include/hw/i386/x86.h index 71e2b6985d..d15713e92e 100644 --- a/include/hw/i386/x86.h +++ b/include/hw/i386/x86.h @@ -17,7 +17,63 @@ #ifndef HW_I386_X86_H #define HW_I386_X86_H +#include "qemu-common.h" +#include "exec/hwaddr.h" +#include "qemu/notify.h" + #include "hw/boards.h" +#include "hw/nmi.h" + +typedef struct { + /*< private >*/ + MachineClass parent; + + /*< public >*/ + + /* Enables contiguous-apic-ID mode */ + bool compat_apic_id_mode; +} X86MachineClass; + +typedef struct { + /*< private >*/ + MachineState parent; + + /*< public >*/ + + /* Pointers to devices and objects: */ + ISADevice *rtc; + FWCfgState *fw_cfg; + qemu_irq *gsi; + GMappedFile *initrd_mapped_file; + + /* Configuration options: */ + uint64_t max_ram_below_4g; + + /* RAM information (sizes, addresses, configuration): */ + ram_addr_t below_4g_mem_size, above_4g_mem_size; + + /* CPU and apic information: */ + bool apic_xrupt_override; + unsigned apic_id_limit; + uint16_t boot_cpus; + unsigned smp_dies; + + /* + * Address space used by IOAPIC device. All IOAPIC interrupts + * will be translated to MSI messages in the address space. + */ + AddressSpace *ioapic_as; +} X86MachineState; + +#define X86_MACHINE_MAX_RAM_BELOW_4G "max-ram-below-4g" + +#define TYPE_X86_MACHINE MACHINE_TYPE_NAME("x86") +#define X86_MACHINE(obj) \ + OBJECT_CHECK(X86MachineState, (obj), TYPE_X86_MACHINE) +#define X86_MACHINE_GET_CLASS(obj) \ + OBJECT_GET_CLASS(X86MachineClass, obj, TYPE_X86_MACHINE) +#define X86_MACHINE_CLASS(class) \ + OBJECT_CLASS_CHECK(X86MachineClass, class, TYPE_X86_MACHINE) uint32_t x86_cpu_apic_id_from_index(PCMachineState *pcms, unsigned int cpu_index); @@ -30,6 +86,6 @@ const CPUArchIdList *x86_possible_cpu_arch_ids(MachineState *ms); void x86_bios_rom_init(MemoryRegion *rom_memory, bool isapc_ram_fw); -void x86_load_linux(PCMachineState *x86ms, FWCfgState *fw_cfg); +void x86_load_linux(PCMachineState *pcms, FWCfgState *fw_cfg); #endif -- cgit v1.2.3-55-g7522 From 703a548aa9f148186ed8647092e8645e5a85f373 Mon Sep 17 00:00:00 2001 From: Sergio Lopez Date: Mon, 30 Sep 2019 17:26:29 +0200 Subject: hw/i386: make x86.c independent from PCMachineState As a last step into splitting PCMachineState and deriving X86MachineState from it, make the functions previously extracted from pc.c to x86.c independent from PCMachineState, using X86MachineState instead. Signed-off-by: Sergio Lopez Reviewed-by: Philippe Mathieu-Daudé Tested-by: Philippe Mathieu-Daudé Reviewed-by: Michael S. Tsirkin --- hw/i386/pc.c | 14 +++++++++----- hw/i386/pc_piix.c | 2 +- hw/i386/pc_q35.c | 2 +- hw/i386/x86.c | 53 +++++++++++++++++++++++---------------------------- include/hw/i386/x86.h | 13 +++++++++---- 5 files changed, 44 insertions(+), 40 deletions(-) (limited to 'include') diff --git a/hw/i386/pc.c b/hw/i386/pc.c index e6ad1f1f36..a8888dd622 100644 --- a/hw/i386/pc.c +++ b/hw/i386/pc.c @@ -983,8 +983,8 @@ void pc_smp_parse(MachineState *ms, QemuOpts *opts) void pc_hot_add_cpu(MachineState *ms, const int64_t id, Error **errp) { - PCMachineState *pcms = PC_MACHINE(ms); - int64_t apic_id = x86_cpu_apic_id_from_index(pcms, id); + X86MachineState *x86ms = X86_MACHINE(ms); + int64_t apic_id = x86_cpu_apic_id_from_index(x86ms, id); Error *local_err = NULL; if (id < 0) { @@ -999,7 +999,8 @@ void pc_hot_add_cpu(MachineState *ms, const int64_t id, Error **errp) return; } - x86_cpu_new(PC_MACHINE(ms), apic_id, &local_err); + + x86_cpu_new(X86_MACHINE(ms), apic_id, &local_err); if (local_err) { error_propagate(errp, local_err); return; @@ -1100,6 +1101,7 @@ void xen_load_linux(PCMachineState *pcms) { int i; FWCfgState *fw_cfg; + PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); X86MachineState *x86ms = X86_MACHINE(pcms); assert(MACHINE(pcms)->kernel_filename != NULL); @@ -1108,7 +1110,8 @@ void xen_load_linux(PCMachineState *pcms) fw_cfg_add_i16(fw_cfg, FW_CFG_NB_CPUS, x86ms->boot_cpus); rom_set_fw(fw_cfg); - x86_load_linux(pcms, fw_cfg); + x86_load_linux(x86ms, fw_cfg, pcmc->acpi_data_size, + pcmc->pvh_enabled, pcmc->linuxboot_dma_enabled); for (i = 0; i < nb_option_roms; i++) { assert(!strcmp(option_rom[i].name, "linuxboot.bin") || !strcmp(option_rom[i].name, "linuxboot_dma.bin") || @@ -1244,7 +1247,8 @@ void pc_memory_init(PCMachineState *pcms, } if (linux_boot) { - x86_load_linux(pcms, fw_cfg); + x86_load_linux(x86ms, fw_cfg, pcmc->acpi_data_size, + pcmc->pvh_enabled, pcmc->linuxboot_dma_enabled); } for (i = 0; i < nb_option_roms; i++) { diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c index 0afa8fe6ea..a86317cdff 100644 --- a/hw/i386/pc_piix.c +++ b/hw/i386/pc_piix.c @@ -154,7 +154,7 @@ static void pc_init1(MachineState *machine, } } - x86_cpus_init(pcms); + x86_cpus_init(x86ms, pcmc->default_cpu_version); if (kvm_enabled() && pcmc->kvmclock_enabled) { kvmclock_create(); diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c index 374ac6c068..75c8caf7c2 100644 --- a/hw/i386/pc_q35.c +++ b/hw/i386/pc_q35.c @@ -181,7 +181,7 @@ static void pc_q35_init(MachineState *machine) xen_hvm_init(pcms, &ram_memory); } - x86_cpus_init(pcms); + x86_cpus_init(x86ms, pcmc->default_cpu_version); kvmclock_create(); diff --git a/hw/i386/x86.c b/hw/i386/x86.c index de4fed0164..fd84b23124 100644 --- a/hw/i386/x86.c +++ b/hw/i386/x86.c @@ -36,7 +36,6 @@ #include "sysemu/sysemu.h" #include "hw/i386/x86.h" -#include "hw/i386/pc.h" #include "target/i386/cpu.h" #include "hw/i386/topology.h" #include "hw/i386/fw_cfg.h" @@ -61,11 +60,10 @@ static size_t pvh_start_addr; * no concept of "CPU index", and the NUMA tables on fw_cfg need the APIC ID of * all CPUs up to max_cpus. */ -uint32_t x86_cpu_apic_id_from_index(PCMachineState *pcms, +uint32_t x86_cpu_apic_id_from_index(X86MachineState *x86ms, unsigned int cpu_index) { - MachineState *ms = MACHINE(pcms); - X86MachineState *x86ms = X86_MACHINE(pcms); + MachineState *ms = MACHINE(x86ms); X86MachineClass *x86mc = X86_MACHINE_GET_CLASS(x86ms); uint32_t correct_id; static bool warned; @@ -84,14 +82,14 @@ uint32_t x86_cpu_apic_id_from_index(PCMachineState *pcms, } } -void x86_cpu_new(PCMachineState *pcms, int64_t apic_id, Error **errp) + +void x86_cpu_new(X86MachineState *x86ms, int64_t apic_id, Error **errp) { Object *cpu = NULL; Error *local_err = NULL; CPUX86State *env = NULL; - X86MachineState *x86ms = X86_MACHINE(pcms); - cpu = object_new(MACHINE(pcms)->cpu_type); + cpu = object_new(MACHINE(x86ms)->cpu_type); env = &X86_CPU(cpu)->env; env->nr_dies = x86ms->smp_dies; @@ -103,30 +101,28 @@ void x86_cpu_new(PCMachineState *pcms, int64_t apic_id, Error **errp) error_propagate(errp, local_err); } -void x86_cpus_init(PCMachineState *pcms) +void x86_cpus_init(X86MachineState *x86ms, int default_cpu_version) { int i; const CPUArchIdList *possible_cpus; - MachineState *ms = MACHINE(pcms); - MachineClass *mc = MACHINE_GET_CLASS(pcms); - PCMachineClass *pcmc = PC_MACHINE_CLASS(mc); - X86MachineState *x86ms = X86_MACHINE(pcms); + MachineState *ms = MACHINE(x86ms); + MachineClass *mc = MACHINE_GET_CLASS(x86ms); - x86_cpu_set_default_version(pcmc->default_cpu_version); + x86_cpu_set_default_version(default_cpu_version); /* * Calculates the limit to CPU APIC ID values * * Limit for the APIC ID value, so that all - * CPU APIC IDs are < pcms->apic_id_limit. + * CPU APIC IDs are < x86ms->apic_id_limit. * * This is used for FW_CFG_MAX_CPUS. See comments on fw_cfg_arch_create(). */ - x86ms->apic_id_limit = x86_cpu_apic_id_from_index(pcms, + x86ms->apic_id_limit = x86_cpu_apic_id_from_index(x86ms, ms->smp.max_cpus - 1) + 1; possible_cpus = mc->possible_cpu_arch_ids(ms); for (i = 0; i < ms->smp.cpus; i++) { - x86_cpu_new(pcms, possible_cpus->cpus[i].arch_id, &error_fatal); + x86_cpu_new(x86ms, possible_cpus->cpus[i].arch_id, &error_fatal); } } @@ -154,7 +150,6 @@ int64_t x86_get_default_cpu_node_id(const MachineState *ms, int idx) const CPUArchIdList *x86_possible_cpu_arch_ids(MachineState *ms) { - PCMachineState *pcms = PC_MACHINE(ms); X86MachineState *x86ms = X86_MACHINE(ms); int i; unsigned int max_cpus = ms->smp.max_cpus; @@ -177,7 +172,7 @@ const CPUArchIdList *x86_possible_cpu_arch_ids(MachineState *ms) ms->possible_cpus->cpus[i].type = ms->cpu_type; ms->possible_cpus->cpus[i].vcpus_count = 1; ms->possible_cpus->cpus[i].arch_id = - x86_cpu_apic_id_from_index(pcms, i); + x86_cpu_apic_id_from_index(x86ms, i); x86_topo_ids_from_apicid(ms->possible_cpus->cpus[i].arch_id, x86ms->smp_dies, ms->smp.cores, ms->smp.threads, &topo); @@ -335,8 +330,11 @@ static bool load_elfboot(const char *kernel_filename, return true; } -void x86_load_linux(PCMachineState *pcms, - FWCfgState *fw_cfg) +void x86_load_linux(X86MachineState *x86ms, + FWCfgState *fw_cfg, + int acpi_data_size, + bool pvh_enabled, + bool linuxboot_dma_enabled) { uint16_t protocol; int setup_size, kernel_size, cmdline_size; @@ -346,9 +344,7 @@ void x86_load_linux(PCMachineState *pcms, hwaddr real_addr, prot_addr, cmdline_addr, initrd_addr = 0; FILE *f; char *vmode; - MachineState *machine = MACHINE(pcms); - PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms); - X86MachineState *x86ms = X86_MACHINE(pcms); + MachineState *machine = MACHINE(x86ms); struct setup_data *setup_data; const char *kernel_filename = machine->kernel_filename; const char *initrd_filename = machine->initrd_filename; @@ -395,7 +391,7 @@ void x86_load_linux(PCMachineState *pcms, * saving the PVH entry point used by the x86/HVM direct boot ABI. * If load_elfboot() is successful, populate the fw_cfg info. */ - if (pcmc->pvh_enabled && + if (pvh_enabled && load_elfboot(kernel_filename, kernel_size, header, pvh_start_addr, fw_cfg)) { fclose(f); @@ -425,8 +421,7 @@ void x86_load_linux(PCMachineState *pcms, initrd_data = g_mapped_file_get_contents(mapped_file); initrd_size = g_mapped_file_get_length(mapped_file); - initrd_max = - x86ms->below_4g_mem_size - pcmc->acpi_data_size - 1; + initrd_max = x86ms->below_4g_mem_size - acpi_data_size - 1; if (initrd_size >= initrd_max) { fprintf(stderr, "qemu: initrd is too large, cannot support." "(max: %"PRIu32", need %"PRId64")\n", @@ -494,8 +489,8 @@ void x86_load_linux(PCMachineState *pcms, initrd_max = 0x37ffffff; } - if (initrd_max >= x86ms->below_4g_mem_size - pcmc->acpi_data_size) { - initrd_max = x86ms->below_4g_mem_size - pcmc->acpi_data_size - 1; + if (initrd_max >= x86ms->below_4g_mem_size - acpi_data_size) { + initrd_max = x86ms->below_4g_mem_size - acpi_data_size - 1; } fw_cfg_add_i32(fw_cfg, FW_CFG_CMDLINE_ADDR, cmdline_addr); @@ -652,7 +647,7 @@ void x86_load_linux(PCMachineState *pcms, option_rom[nb_option_roms].bootindex = 0; option_rom[nb_option_roms].name = "linuxboot.bin"; - if (pcmc->linuxboot_dma_enabled && fw_cfg_dma_enabled(fw_cfg)) { + if (linuxboot_dma_enabled && fw_cfg_dma_enabled(fw_cfg)) { option_rom[nb_option_roms].name = "linuxboot_dma.bin"; } nb_option_roms++; diff --git a/include/hw/i386/x86.h b/include/hw/i386/x86.h index d15713e92e..82d09fd7d0 100644 --- a/include/hw/i386/x86.h +++ b/include/hw/i386/x86.h @@ -75,10 +75,11 @@ typedef struct { #define X86_MACHINE_CLASS(class) \ OBJECT_CLASS_CHECK(X86MachineClass, class, TYPE_X86_MACHINE) -uint32_t x86_cpu_apic_id_from_index(PCMachineState *pcms, +uint32_t x86_cpu_apic_id_from_index(X86MachineState *pcms, unsigned int cpu_index); -void x86_cpu_new(PCMachineState *pcms, int64_t apic_id, Error **errp); -void x86_cpus_init(PCMachineState *pcms); + +void x86_cpu_new(X86MachineState *pcms, int64_t apic_id, Error **errp); +void x86_cpus_init(X86MachineState *pcms, int default_cpu_version); CpuInstanceProperties x86_cpu_index_to_props(MachineState *ms, unsigned cpu_index); int64_t x86_get_default_cpu_node_id(const MachineState *ms, int idx); @@ -86,6 +87,10 @@ const CPUArchIdList *x86_possible_cpu_arch_ids(MachineState *ms); void x86_bios_rom_init(MemoryRegion *rom_memory, bool isapc_ram_fw); -void x86_load_linux(PCMachineState *pcms, FWCfgState *fw_cfg); +void x86_load_linux(X86MachineState *x86ms, + FWCfgState *fw_cfg, + int acpi_data_size, + bool pvh_enabled, + bool linuxboot_dma_enabled); #endif -- cgit v1.2.3-55-g7522 From e5f6aa319f7636168469b2c797249680fa5e0413 Mon Sep 17 00:00:00 2001 From: Sergio Lopez Date: Tue, 24 Sep 2019 11:38:18 +0200 Subject: fw_cfg: add "modify" functions for all types This allows to alter the contents of an already added item. Signed-off-by: Sergio Lopez Reviewed-by: Michael S. Tsirkin Reviewed-by: Marc-André Lureau --- hw/nvram/fw_cfg.c | 29 +++++++++++++++++++++++++++++ include/hw/nvram/fw_cfg.h | 42 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 71 insertions(+) (limited to 'include') diff --git a/hw/nvram/fw_cfg.c b/hw/nvram/fw_cfg.c index 7dc3ac378e..aef1727250 100644 --- a/hw/nvram/fw_cfg.c +++ b/hw/nvram/fw_cfg.c @@ -690,6 +690,15 @@ void fw_cfg_add_string(FWCfgState *s, uint16_t key, const char *value) fw_cfg_add_bytes(s, key, g_memdup(value, sz), sz); } +void fw_cfg_modify_string(FWCfgState *s, uint16_t key, const char *value) +{ + size_t sz = strlen(value) + 1; + char *old; + + old = fw_cfg_modify_bytes_read(s, key, g_memdup(value, sz), sz); + g_free(old); +} + void fw_cfg_add_i16(FWCfgState *s, uint16_t key, uint16_t value) { uint16_t *copy; @@ -720,6 +729,16 @@ void fw_cfg_add_i32(FWCfgState *s, uint16_t key, uint32_t value) fw_cfg_add_bytes(s, key, copy, sizeof(value)); } +void fw_cfg_modify_i32(FWCfgState *s, uint16_t key, uint32_t value) +{ + uint32_t *copy, *old; + + copy = g_malloc(sizeof(value)); + *copy = cpu_to_le32(value); + old = fw_cfg_modify_bytes_read(s, key, copy, sizeof(value)); + g_free(old); +} + void fw_cfg_add_i64(FWCfgState *s, uint16_t key, uint64_t value) { uint64_t *copy; @@ -730,6 +749,16 @@ void fw_cfg_add_i64(FWCfgState *s, uint16_t key, uint64_t value) fw_cfg_add_bytes(s, key, copy, sizeof(value)); } +void fw_cfg_modify_i64(FWCfgState *s, uint16_t key, uint64_t value) +{ + uint64_t *copy, *old; + + copy = g_malloc(sizeof(value)); + *copy = cpu_to_le64(value); + old = fw_cfg_modify_bytes_read(s, key, copy, sizeof(value)); + g_free(old); +} + void fw_cfg_set_order_override(FWCfgState *s, int order) { assert(s->fw_cfg_order_override == 0); diff --git a/include/hw/nvram/fw_cfg.h b/include/hw/nvram/fw_cfg.h index 80e435d303..b5291eefad 100644 --- a/include/hw/nvram/fw_cfg.h +++ b/include/hw/nvram/fw_cfg.h @@ -98,6 +98,20 @@ void fw_cfg_add_bytes(FWCfgState *s, uint16_t key, void *data, size_t len); */ void fw_cfg_add_string(FWCfgState *s, uint16_t key, const char *value); +/** + * fw_cfg_modify_string: + * @s: fw_cfg device being modified + * @key: selector key value for new fw_cfg item + * @value: NUL-terminated ascii string + * + * Replace the fw_cfg item available by selecting the given key. The new + * data will consist of a dynamically allocated copy of the provided string, + * including its NUL terminator. The data being replaced, assumed to have + * been dynamically allocated during an earlier call to either + * fw_cfg_add_string() or fw_cfg_modify_string(), is freed before returning. + */ +void fw_cfg_modify_string(FWCfgState *s, uint16_t key, const char *value); + /** * fw_cfg_add_i16: * @s: fw_cfg device being modified @@ -136,6 +150,20 @@ void fw_cfg_modify_i16(FWCfgState *s, uint16_t key, uint16_t value); */ void fw_cfg_add_i32(FWCfgState *s, uint16_t key, uint32_t value); +/** + * fw_cfg_modify_i32: + * @s: fw_cfg device being modified + * @key: selector key value for new fw_cfg item + * @value: 32-bit integer + * + * Replace the fw_cfg item available by selecting the given key. The new + * data will consist of a dynamically allocated copy of the given 32-bit + * value, converted to little-endian representation. The data being replaced, + * assumed to have been dynamically allocated during an earlier call to + * either fw_cfg_add_i32() or fw_cfg_modify_i32(), is freed before returning. + */ +void fw_cfg_modify_i32(FWCfgState *s, uint16_t key, uint32_t value); + /** * fw_cfg_add_i64: * @s: fw_cfg device being modified @@ -148,6 +176,20 @@ void fw_cfg_add_i32(FWCfgState *s, uint16_t key, uint32_t value); */ void fw_cfg_add_i64(FWCfgState *s, uint16_t key, uint64_t value); +/** + * fw_cfg_modify_i64: + * @s: fw_cfg device being modified + * @key: selector key value for new fw_cfg item + * @value: 64-bit integer + * + * Replace the fw_cfg item available by selecting the given key. The new + * data will consist of a dynamically allocated copy of the given 64-bit + * value, converted to little-endian representation. The data being replaced, + * assumed to have been dynamically allocated during an earlier call to + * either fw_cfg_add_i64() or fw_cfg_modify_i64(), is freed before returning. + */ +void fw_cfg_modify_i64(FWCfgState *s, uint16_t key, uint64_t value); + /** * fw_cfg_add_file: * @s: fw_cfg device being modified -- cgit v1.2.3-55-g7522 From 0ebf007ddacf15a4483a4e740bf6c9dd688b5b5a Mon Sep 17 00:00:00 2001 From: Sergio Lopez Date: Wed, 2 Oct 2019 08:03:19 +0200 Subject: hw/i386: Introduce the microvm machine type microvm is a machine type inspired by Firecracker and constructed after its machine model. It's a minimalist machine type without PCI nor ACPI support, designed for short-lived guests. microvm also establishes a baseline for benchmarking and optimizing both QEMU and guest operating systems, since it is optimized for both boot time and footprint. Signed-off-by: Sergio Lopez Reviewed-by: Michael S. Tsirkin --- default-configs/i386-softmmu.mak | 1 + hw/i386/Kconfig | 10 + hw/i386/Makefile.objs | 1 + hw/i386/microvm.c | 572 +++++++++++++++++++++++++++++++++++++++ include/hw/i386/microvm.h | 71 +++++ 5 files changed, 655 insertions(+) create mode 100644 hw/i386/microvm.c create mode 100644 include/hw/i386/microvm.h (limited to 'include') diff --git a/default-configs/i386-softmmu.mak b/default-configs/i386-softmmu.mak index 4229900f57..4cc64dafa2 100644 --- a/default-configs/i386-softmmu.mak +++ b/default-configs/i386-softmmu.mak @@ -28,3 +28,4 @@ CONFIG_ISAPC=y CONFIG_I440FX=y CONFIG_Q35=y +CONFIG_MICROVM=y diff --git a/hw/i386/Kconfig b/hw/i386/Kconfig index c5c9d4900e..b25bb6d78a 100644 --- a/hw/i386/Kconfig +++ b/hw/i386/Kconfig @@ -92,6 +92,16 @@ config Q35 select SMBIOS select FW_CFG_DMA +config MICROVM + bool + imply SERIAL_ISA + select ISA_BUS + select APIC + select IOAPIC + select I8259 + select MC146818RTC + select VIRTIO_MMIO + config VTD bool diff --git a/hw/i386/Makefile.objs b/hw/i386/Makefile.objs index 7ed80a4853..0d195b5210 100644 --- a/hw/i386/Makefile.objs +++ b/hw/i386/Makefile.objs @@ -4,6 +4,7 @@ obj-y += x86.o obj-y += pc.o obj-$(CONFIG_I440FX) += pc_piix.o obj-$(CONFIG_Q35) += pc_q35.o +obj-$(CONFIG_MICROVM) += microvm.o obj-y += fw_cfg.o pc_sysfw.o obj-y += x86-iommu.o obj-$(CONFIG_VTD) += intel_iommu.o diff --git a/hw/i386/microvm.c b/hw/i386/microvm.c new file mode 100644 index 0000000000..20d2189ea8 --- /dev/null +++ b/hw/i386/microvm.c @@ -0,0 +1,572 @@ +/* + * Copyright (c) 2018 Intel Corporation + * Copyright (c) 2019 Red Hat, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2 or later, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see . + */ + +#include "qemu/osdep.h" +#include "qemu/error-report.h" +#include "qemu/cutils.h" +#include "qemu/units.h" +#include "qapi/error.h" +#include "qapi/visitor.h" +#include "qapi/qapi-visit-common.h" +#include "sysemu/sysemu.h" +#include "sysemu/cpus.h" +#include "sysemu/numa.h" +#include "sysemu/reset.h" + +#include "hw/loader.h" +#include "hw/irq.h" +#include "hw/kvm/clock.h" +#include "hw/i386/microvm.h" +#include "hw/i386/x86.h" +#include "hw/i386/pc.h" +#include "target/i386/cpu.h" +#include "hw/timer/i8254.h" +#include "hw/timer/mc146818rtc.h" +#include "hw/char/serial.h" +#include "hw/i386/topology.h" +#include "hw/i386/e820_memory_layout.h" +#include "hw/i386/fw_cfg.h" +#include "hw/virtio/virtio-mmio.h" + +#include "cpu.h" +#include "elf.h" +#include "kvm_i386.h" +#include "hw/xen/start_info.h" + +#define MICROVM_BIOS_FILENAME "bios-microvm.bin" + +static void microvm_set_rtc(MicrovmMachineState *mms, ISADevice *s) +{ + X86MachineState *x86ms = X86_MACHINE(mms); + int val; + + val = MIN(x86ms->below_4g_mem_size / KiB, 640); + rtc_set_memory(s, 0x15, val); + rtc_set_memory(s, 0x16, val >> 8); + /* extended memory (next 64MiB) */ + if (x86ms->below_4g_mem_size > 1 * MiB) { + val = (x86ms->below_4g_mem_size - 1 * MiB) / KiB; + } else { + val = 0; + } + if (val > 65535) { + val = 65535; + } + rtc_set_memory(s, 0x17, val); + rtc_set_memory(s, 0x18, val >> 8); + rtc_set_memory(s, 0x30, val); + rtc_set_memory(s, 0x31, val >> 8); + /* memory between 16MiB and 4GiB */ + if (x86ms->below_4g_mem_size > 16 * MiB) { + val = (x86ms->below_4g_mem_size - 16 * MiB) / (64 * KiB); + } else { + val = 0; + } + if (val > 65535) { + val = 65535; + } + rtc_set_memory(s, 0x34, val); + rtc_set_memory(s, 0x35, val >> 8); + /* memory above 4GiB */ + val = x86ms->above_4g_mem_size / 65536; + rtc_set_memory(s, 0x5b, val); + rtc_set_memory(s, 0x5c, val >> 8); + rtc_set_memory(s, 0x5d, val >> 16); +} + +static void microvm_gsi_handler(void *opaque, int n, int level) +{ + GSIState *s = opaque; + + qemu_set_irq(s->ioapic_irq[n], level); +} + +static void microvm_devices_init(MicrovmMachineState *mms) +{ + X86MachineState *x86ms = X86_MACHINE(mms); + ISABus *isa_bus; + ISADevice *rtc_state; + GSIState *gsi_state; + int i; + + /* Core components */ + + gsi_state = g_malloc0(sizeof(*gsi_state)); + if (mms->pic == ON_OFF_AUTO_ON || mms->pic == ON_OFF_AUTO_AUTO) { + x86ms->gsi = qemu_allocate_irqs(gsi_handler, gsi_state, GSI_NUM_PINS); + } else { + x86ms->gsi = qemu_allocate_irqs(microvm_gsi_handler, + gsi_state, GSI_NUM_PINS); + } + + isa_bus = isa_bus_new(NULL, get_system_memory(), get_system_io(), + &error_abort); + isa_bus_irqs(isa_bus, x86ms->gsi); + + ioapic_init_gsi(gsi_state, "machine"); + + kvmclock_create(); + + for (i = 0; i < VIRTIO_NUM_TRANSPORTS; i++) { + sysbus_create_simple("virtio-mmio", + VIRTIO_MMIO_BASE + i * 512, + x86ms->gsi[VIRTIO_IRQ_BASE + i]); + } + + /* Optional and legacy devices */ + + if (mms->pic == ON_OFF_AUTO_ON || mms->pic == ON_OFF_AUTO_AUTO) { + qemu_irq *i8259; + + i8259 = i8259_init(isa_bus, pc_allocate_cpu_irq()); + for (i = 0; i < ISA_NUM_IRQS; i++) { + gsi_state->i8259_irq[i] = i8259[i]; + } + g_free(i8259); + } + + if (mms->pit == ON_OFF_AUTO_ON || mms->pit == ON_OFF_AUTO_AUTO) { + if (kvm_pit_in_kernel()) { + kvm_pit_init(isa_bus, 0x40); + } else { + i8254_pit_init(isa_bus, 0x40, 0, NULL); + } + } + + if (mms->rtc == ON_OFF_AUTO_ON || + (mms->rtc == ON_OFF_AUTO_AUTO && !kvm_enabled())) { + rtc_state = mc146818_rtc_init(isa_bus, 2000, NULL); + microvm_set_rtc(mms, rtc_state); + } + + if (mms->isa_serial) { + serial_hds_isa_init(isa_bus, 0, 1); + } + + if (bios_name == NULL) { + bios_name = MICROVM_BIOS_FILENAME; + } + x86_bios_rom_init(get_system_memory(), true); +} + +static void microvm_memory_init(MicrovmMachineState *mms) +{ + MachineState *machine = MACHINE(mms); + X86MachineState *x86ms = X86_MACHINE(mms); + MemoryRegion *ram, *ram_below_4g, *ram_above_4g; + MemoryRegion *system_memory = get_system_memory(); + FWCfgState *fw_cfg; + ram_addr_t lowmem; + int i; + + /* + * Check whether RAM fits below 4G (leaving 1/2 GByte for IO memory + * and 256 Mbytes for PCI Express Enhanced Configuration Access Mapping + * also known as MMCFG). + * If it doesn't, we need to split it in chunks below and above 4G. + * In any case, try to make sure that guest addresses aligned at + * 1G boundaries get mapped to host addresses aligned at 1G boundaries. + */ + if (machine->ram_size >= 0xb0000000) { + lowmem = 0x80000000; + } else { + lowmem = 0xb0000000; + } + + /* + * Handle the machine opt max-ram-below-4g. It is basically doing + * min(qemu limit, user limit). + */ + if (!x86ms->max_ram_below_4g) { + x86ms->max_ram_below_4g = 4 * GiB; + } + if (lowmem > x86ms->max_ram_below_4g) { + lowmem = x86ms->max_ram_below_4g; + if (machine->ram_size - lowmem > lowmem && + lowmem & (1 * GiB - 1)) { + warn_report("There is possibly poor performance as the ram size " + " (0x%" PRIx64 ") is more then twice the size of" + " max-ram-below-4g (%"PRIu64") and" + " max-ram-below-4g is not a multiple of 1G.", + (uint64_t)machine->ram_size, x86ms->max_ram_below_4g); + } + } + + if (machine->ram_size > lowmem) { + x86ms->above_4g_mem_size = machine->ram_size - lowmem; + x86ms->below_4g_mem_size = lowmem; + } else { + x86ms->above_4g_mem_size = 0; + x86ms->below_4g_mem_size = machine->ram_size; + } + + ram = g_malloc(sizeof(*ram)); + memory_region_allocate_system_memory(ram, NULL, "microvm.ram", + machine->ram_size); + + ram_below_4g = g_malloc(sizeof(*ram_below_4g)); + memory_region_init_alias(ram_below_4g, NULL, "ram-below-4g", ram, + 0, x86ms->below_4g_mem_size); + memory_region_add_subregion(system_memory, 0, ram_below_4g); + + e820_add_entry(0, x86ms->below_4g_mem_size, E820_RAM); + + if (x86ms->above_4g_mem_size > 0) { + ram_above_4g = g_malloc(sizeof(*ram_above_4g)); + memory_region_init_alias(ram_above_4g, NULL, "ram-above-4g", ram, + x86ms->below_4g_mem_size, + x86ms->above_4g_mem_size); + memory_region_add_subregion(system_memory, 0x100000000ULL, + ram_above_4g); + e820_add_entry(0x100000000ULL, x86ms->above_4g_mem_size, E820_RAM); + } + + fw_cfg = fw_cfg_init_io_dma(FW_CFG_IO_BASE, FW_CFG_IO_BASE + 4, + &address_space_memory); + + fw_cfg_add_i16(fw_cfg, FW_CFG_NB_CPUS, machine->smp.cpus); + fw_cfg_add_i16(fw_cfg, FW_CFG_MAX_CPUS, machine->smp.max_cpus); + fw_cfg_add_i64(fw_cfg, FW_CFG_RAM_SIZE, (uint64_t)machine->ram_size); + fw_cfg_add_i32(fw_cfg, FW_CFG_IRQ0_OVERRIDE, kvm_allows_irq0_override()); + fw_cfg_add_bytes(fw_cfg, FW_CFG_E820_TABLE, + &e820_reserve, sizeof(e820_reserve)); + fw_cfg_add_file(fw_cfg, "etc/e820", e820_table, + sizeof(struct e820_entry) * e820_get_num_entries()); + + rom_set_fw(fw_cfg); + + if (machine->kernel_filename != NULL) { + x86_load_linux(x86ms, fw_cfg, 0, true, true); + } + + if (mms->option_roms) { + for (i = 0; i < nb_option_roms; i++) { + rom_add_option(option_rom[i].name, option_rom[i].bootindex); + } + } + + x86ms->fw_cfg = fw_cfg; + x86ms->ioapic_as = &address_space_memory; +} + +static gchar *microvm_get_mmio_cmdline(gchar *name) +{ + gchar *cmdline; + gchar *separator; + long int index; + int ret; + + separator = g_strrstr(name, "."); + if (!separator) { + return NULL; + } + + if (qemu_strtol(separator + 1, NULL, 10, &index) != 0) { + return NULL; + } + + cmdline = g_malloc0(VIRTIO_CMDLINE_MAXLEN); + ret = g_snprintf(cmdline, VIRTIO_CMDLINE_MAXLEN, + " virtio_mmio.device=512@0x%lx:%ld", + VIRTIO_MMIO_BASE + index * 512, + VIRTIO_IRQ_BASE + index); + if (ret < 0 || ret >= VIRTIO_CMDLINE_MAXLEN) { + g_free(cmdline); + return NULL; + } + + return cmdline; +} + +static void microvm_fix_kernel_cmdline(MachineState *machine) +{ + X86MachineState *x86ms = X86_MACHINE(machine); + BusState *bus; + BusChild *kid; + char *cmdline; + + /* + * Find MMIO transports with attached devices, and add them to the kernel + * command line. + * + * Yes, this is a hack, but one that heavily improves the UX without + * introducing any significant issues. + */ + cmdline = g_strdup(machine->kernel_cmdline); + bus = sysbus_get_default(); + QTAILQ_FOREACH(kid, &bus->children, sibling) { + DeviceState *dev = kid->child; + ObjectClass *class = object_get_class(OBJECT(dev)); + + if (class == object_class_by_name(TYPE_VIRTIO_MMIO)) { + VirtIOMMIOProxy *mmio = VIRTIO_MMIO(OBJECT(dev)); + VirtioBusState *mmio_virtio_bus = &mmio->bus; + BusState *mmio_bus = &mmio_virtio_bus->parent_obj; + + if (!QTAILQ_EMPTY(&mmio_bus->children)) { + gchar *mmio_cmdline = microvm_get_mmio_cmdline(mmio_bus->name); + if (mmio_cmdline) { + char *newcmd = g_strjoin(NULL, cmdline, mmio_cmdline, NULL); + g_free(mmio_cmdline); + g_free(cmdline); + cmdline = newcmd; + } + } + } + } + + fw_cfg_modify_i32(x86ms->fw_cfg, FW_CFG_CMDLINE_SIZE, strlen(cmdline) + 1); + fw_cfg_modify_string(x86ms->fw_cfg, FW_CFG_CMDLINE_DATA, cmdline); +} + +static void microvm_machine_state_init(MachineState *machine) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(machine); + X86MachineState *x86ms = X86_MACHINE(machine); + Error *local_err = NULL; + + microvm_memory_init(mms); + + x86_cpus_init(x86ms, CPU_VERSION_LATEST); + if (local_err) { + error_report_err(local_err); + exit(1); + } + + microvm_devices_init(mms); +} + +static void microvm_machine_reset(MachineState *machine) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(machine); + CPUState *cs; + X86CPU *cpu; + + if (machine->kernel_filename != NULL && + mms->auto_kernel_cmdline && !mms->kernel_cmdline_fixed) { + microvm_fix_kernel_cmdline(machine); + mms->kernel_cmdline_fixed = true; + } + + qemu_devices_reset(); + + CPU_FOREACH(cs) { + cpu = X86_CPU(cs); + + if (cpu->apic_state) { + device_reset(cpu->apic_state); + } + } +} + +static void microvm_machine_get_pic(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + OnOffAuto pic = mms->pic; + + visit_type_OnOffAuto(v, name, &pic, errp); +} + +static void microvm_machine_set_pic(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + + visit_type_OnOffAuto(v, name, &mms->pic, errp); +} + +static void microvm_machine_get_pit(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + OnOffAuto pit = mms->pit; + + visit_type_OnOffAuto(v, name, &pit, errp); +} + +static void microvm_machine_set_pit(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + + visit_type_OnOffAuto(v, name, &mms->pit, errp); +} + +static void microvm_machine_get_rtc(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + OnOffAuto rtc = mms->rtc; + + visit_type_OnOffAuto(v, name, &rtc, errp); +} + +static void microvm_machine_set_rtc(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + + visit_type_OnOffAuto(v, name, &mms->rtc, errp); +} + +static bool microvm_machine_get_isa_serial(Object *obj, Error **errp) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + + return mms->isa_serial; +} + +static void microvm_machine_set_isa_serial(Object *obj, bool value, + Error **errp) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + + mms->isa_serial = value; +} + +static bool microvm_machine_get_option_roms(Object *obj, Error **errp) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + + return mms->option_roms; +} + +static void microvm_machine_set_option_roms(Object *obj, bool value, + Error **errp) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + + mms->option_roms = value; +} + +static bool microvm_machine_get_auto_kernel_cmdline(Object *obj, Error **errp) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + + return mms->auto_kernel_cmdline; +} + +static void microvm_machine_set_auto_kernel_cmdline(Object *obj, bool value, + Error **errp) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + + mms->auto_kernel_cmdline = value; +} + +static void microvm_machine_initfn(Object *obj) +{ + MicrovmMachineState *mms = MICROVM_MACHINE(obj); + + /* Configuration */ + mms->pic = ON_OFF_AUTO_AUTO; + mms->pit = ON_OFF_AUTO_AUTO; + mms->rtc = ON_OFF_AUTO_AUTO; + mms->isa_serial = true; + mms->option_roms = true; + mms->auto_kernel_cmdline = true; + + /* State */ + mms->kernel_cmdline_fixed = false; +} + +static void microvm_class_init(ObjectClass *oc, void *data) +{ + MachineClass *mc = MACHINE_CLASS(oc); + + mc->init = microvm_machine_state_init; + + mc->family = "microvm_i386"; + mc->desc = "microvm (i386)"; + mc->units_per_default_bus = 1; + mc->no_floppy = 1; + mc->max_cpus = 288; + mc->has_hotpluggable_cpus = false; + mc->auto_enable_numa_with_memhp = false; + mc->default_cpu_type = TARGET_DEFAULT_CPU_TYPE; + mc->nvdimm_supported = false; + + /* Avoid relying too much on kernel components */ + mc->default_kernel_irqchip_split = true; + + /* Machine class handlers */ + mc->reset = microvm_machine_reset; + + object_class_property_add(oc, MICROVM_MACHINE_PIC, "OnOffAuto", + microvm_machine_get_pic, + microvm_machine_set_pic, + NULL, NULL, &error_abort); + object_class_property_set_description(oc, MICROVM_MACHINE_PIC, + "Enable i8259 PIC", &error_abort); + + object_class_property_add(oc, MICROVM_MACHINE_PIT, "OnOffAuto", + microvm_machine_get_pit, + microvm_machine_set_pit, + NULL, NULL, &error_abort); + object_class_property_set_description(oc, MICROVM_MACHINE_PIT, + "Enable i8254 PIT", &error_abort); + + object_class_property_add(oc, MICROVM_MACHINE_RTC, "OnOffAuto", + microvm_machine_get_rtc, + microvm_machine_set_rtc, + NULL, NULL, &error_abort); + object_class_property_set_description(oc, MICROVM_MACHINE_RTC, + "Enable MC146818 RTC", &error_abort); + + object_class_property_add_bool(oc, MICROVM_MACHINE_ISA_SERIAL, + microvm_machine_get_isa_serial, + microvm_machine_set_isa_serial, + &error_abort); + object_class_property_set_description(oc, MICROVM_MACHINE_ISA_SERIAL, + "Set off to disable the instantiation an ISA serial port", + &error_abort); + + object_class_property_add_bool(oc, MICROVM_MACHINE_OPTION_ROMS, + microvm_machine_get_option_roms, + microvm_machine_set_option_roms, + &error_abort); + object_class_property_set_description(oc, MICROVM_MACHINE_OPTION_ROMS, + "Set off to disable loading option ROMs", &error_abort); + + object_class_property_add_bool(oc, MICROVM_MACHINE_AUTO_KERNEL_CMDLINE, + microvm_machine_get_auto_kernel_cmdline, + microvm_machine_set_auto_kernel_cmdline, + &error_abort); + object_class_property_set_description(oc, + MICROVM_MACHINE_AUTO_KERNEL_CMDLINE, + "Set off to disable adding virtio-mmio devices to the kernel cmdline", + &error_abort); +} + +static const TypeInfo microvm_machine_info = { + .name = TYPE_MICROVM_MACHINE, + .parent = TYPE_X86_MACHINE, + .instance_size = sizeof(MicrovmMachineState), + .instance_init = microvm_machine_initfn, + .class_size = sizeof(MicrovmMachineClass), + .class_init = microvm_class_init, + .interfaces = (InterfaceInfo[]) { + { } + }, +}; + +static void microvm_machine_init(void) +{ + type_register_static(µvm_machine_info); +} +type_init(microvm_machine_init); diff --git a/include/hw/i386/microvm.h b/include/hw/i386/microvm.h new file mode 100644 index 0000000000..ba68d1f22b --- /dev/null +++ b/include/hw/i386/microvm.h @@ -0,0 +1,71 @@ +/* + * Copyright (c) 2018 Intel Corporation + * Copyright (c) 2019 Red Hat, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2 or later, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see . + */ + +#ifndef HW_I386_MICROVM_H +#define HW_I386_MICROVM_H + +#include "qemu-common.h" +#include "exec/hwaddr.h" +#include "qemu/notify.h" + +#include "hw/boards.h" +#include "hw/i386/x86.h" + +/* Platform virtio definitions */ +#define VIRTIO_MMIO_BASE 0xc0000000 +#define VIRTIO_IRQ_BASE 5 +#define VIRTIO_NUM_TRANSPORTS 8 +#define VIRTIO_CMDLINE_MAXLEN 64 + +/* Machine type options */ +#define MICROVM_MACHINE_PIT "pit" +#define MICROVM_MACHINE_PIC "pic" +#define MICROVM_MACHINE_RTC "rtc" +#define MICROVM_MACHINE_ISA_SERIAL "isa-serial" +#define MICROVM_MACHINE_OPTION_ROMS "x-option-roms" +#define MICROVM_MACHINE_AUTO_KERNEL_CMDLINE "auto-kernel-cmdline" + +typedef struct { + X86MachineClass parent; + HotplugHandler *(*orig_hotplug_handler)(MachineState *machine, + DeviceState *dev); +} MicrovmMachineClass; + +typedef struct { + X86MachineState parent; + + /* Machine type options */ + OnOffAuto pic; + OnOffAuto pit; + OnOffAuto rtc; + bool isa_serial; + bool option_roms; + bool auto_kernel_cmdline; + + /* Machine state */ + bool kernel_cmdline_fixed; +} MicrovmMachineState; + +#define TYPE_MICROVM_MACHINE MACHINE_TYPE_NAME("microvm") +#define MICROVM_MACHINE(obj) \ + OBJECT_CHECK(MicrovmMachineState, (obj), TYPE_MICROVM_MACHINE) +#define MICROVM_MACHINE_GET_CLASS(obj) \ + OBJECT_GET_CLASS(MicrovmMachineClass, obj, TYPE_MICROVM_MACHINE) +#define MICROVM_MACHINE_CLASS(class) \ + OBJECT_CLASS_CHECK(MicrovmMachineClass, class, TYPE_MICROVM_MACHINE) + +#endif -- cgit v1.2.3-55-g7522 From 417258f139e61899511d6a99c11b276f12bbbd86 Mon Sep 17 00:00:00 2001 From: Philippe Mathieu-Daudé Date: Fri, 18 Oct 2019 15:59:06 +0200 Subject: hw/i386/pc: Extract pc_gsi_create() The GSI creation code is common to all PC machines, extract the common code. Reviewed-by: Aleksandar Markovic Reviewed-by: Thomas Huth Signed-off-by: Philippe Mathieu-Daudé Message-Id: <20191018135910.24286-2-philmd@redhat.com> Signed-off-by: Paolo Bonzini --- hw/i386/pc.c | 15 +++++++++++++++ hw/i386/pc_piix.c | 9 +-------- hw/i386/pc_q35.c | 9 +-------- include/hw/i386/pc.h | 2 ++ 4 files changed, 19 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/hw/i386/pc.c b/hw/i386/pc.c index a8888dd622..e8a54acc38 100644 --- a/hw/i386/pc.c +++ b/hw/i386/pc.c @@ -355,6 +355,21 @@ void gsi_handler(void *opaque, int n, int level) qemu_set_irq(s->ioapic_irq[n], level); } +GSIState *pc_gsi_create(qemu_irq **irqs, bool pci_enabled) +{ + GSIState *s; + + s = g_new0(GSIState, 1); + if (kvm_ioapic_in_kernel()) { + kvm_pc_setup_irq_routing(pci_enabled); + *irqs = qemu_allocate_irqs(kvm_pc_gsi_handler, s, GSI_NUM_PINS); + } else { + *irqs = qemu_allocate_irqs(gsi_handler, s, GSI_NUM_PINS); + } + + return s; +} + static void ioport80_write(void *opaque, hwaddr addr, uint64_t data, unsigned size) { diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c index a86317cdff..0cc951a0b5 100644 --- a/hw/i386/pc_piix.c +++ b/hw/i386/pc_piix.c @@ -189,14 +189,7 @@ static void pc_init1(MachineState *machine, xen_load_linux(pcms); } - gsi_state = g_malloc0(sizeof(*gsi_state)); - if (kvm_ioapic_in_kernel()) { - kvm_pc_setup_irq_routing(pcmc->pci_enabled); - x86ms->gsi = qemu_allocate_irqs(kvm_pc_gsi_handler, gsi_state, - GSI_NUM_PINS); - } else { - x86ms->gsi = qemu_allocate_irqs(gsi_handler, gsi_state, GSI_NUM_PINS); - } + gsi_state = pc_gsi_create(&x86ms->gsi, pcmc->pci_enabled); if (pcmc->pci_enabled) { pci_bus = i440fx_init(host_type, diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c index 75c8caf7c2..255c803688 100644 --- a/hw/i386/pc_q35.c +++ b/hw/i386/pc_q35.c @@ -212,14 +212,7 @@ static void pc_q35_init(MachineState *machine) } /* irq lines */ - gsi_state = g_malloc0(sizeof(*gsi_state)); - if (kvm_ioapic_in_kernel()) { - kvm_pc_setup_irq_routing(pcmc->pci_enabled); - x86ms->gsi = qemu_allocate_irqs(kvm_pc_gsi_handler, gsi_state, - GSI_NUM_PINS); - } else { - x86ms->gsi = qemu_allocate_irqs(gsi_handler, gsi_state, GSI_NUM_PINS); - } + gsi_state = pc_gsi_create(&x86ms->gsi, pcmc->pci_enabled); /* create pci host bus */ q35_host = Q35_HOST_DEVICE(qdev_create(NULL, TYPE_Q35_HOST_DEVICE)); diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h index 13c4eac0f6..8c5dc39d84 100644 --- a/include/hw/i386/pc.h +++ b/include/hw/i386/pc.h @@ -157,6 +157,8 @@ typedef struct GSIState { void gsi_handler(void *opaque, int n, int level); +GSIState *pc_gsi_create(qemu_irq **irqs, bool pci_enabled); + /* vmport.c */ #define TYPE_VMPORT "vmport" typedef uint32_t (VMPortReadFunc)(void *opaque, uint32_t address); -- cgit v1.2.3-55-g7522 From 4501d317b50e52de090192aab6244060fbe42da0 Mon Sep 17 00:00:00 2001 From: Philippe Mathieu-Daudé Date: Fri, 18 Oct 2019 15:59:09 +0200 Subject: hw/i386/pc: Extract pc_i8259_create() The i8259 creation code is common to all PC machines, extract the common code. Signed-off-by: Philippe Mathieu-Daudé Message-Id: <20191018135910.24286-5-philmd@redhat.com> Signed-off-by: Paolo Bonzini --- hw/i386/pc.c | 19 +++++++++++++++++++ hw/i386/pc_piix.c | 13 +------------ hw/i386/pc_q35.c | 14 +------------- include/hw/i386/pc.h | 1 + 4 files changed, 22 insertions(+), 25 deletions(-) (limited to 'include') diff --git a/hw/i386/pc.c b/hw/i386/pc.c index e8a54acc38..5fce60c856 100644 --- a/hw/i386/pc.c +++ b/hw/i386/pc.c @@ -1474,6 +1474,25 @@ void pc_nic_init(PCMachineClass *pcmc, ISABus *isa_bus, PCIBus *pci_bus) rom_reset_order_override(); } +void pc_i8259_create(ISABus *isa_bus, qemu_irq *i8259_irqs) +{ + qemu_irq *i8259; + + if (kvm_pic_in_kernel()) { + i8259 = kvm_i8259_init(isa_bus); + } else if (xen_enabled()) { + i8259 = xen_interrupt_controller_init(); + } else { + i8259 = i8259_init(isa_bus, pc_allocate_cpu_irq()); + } + + for (size_t i = 0; i < ISA_NUM_IRQS; i++) { + i8259_irqs[i] = i8259[i]; + } + + g_free(i8259); +} + void ioapic_init_gsi(GSIState *gsi_state, const char *parent_name) { DeviceState *dev; diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c index 0cc951a0b5..648dc9ab2d 100644 --- a/hw/i386/pc_piix.c +++ b/hw/i386/pc_piix.c @@ -82,7 +82,6 @@ static void pc_init1(MachineState *machine, ISABus *isa_bus; PCII440FXState *i440fx_state; int piix3_devfn = -1; - qemu_irq *i8259; qemu_irq smi_irq; GSIState *gsi_state; DriveInfo *hd[MAX_IDE_BUS * MAX_IDE_DEVS]; @@ -209,18 +208,8 @@ static void pc_init1(MachineState *machine, } isa_bus_irqs(isa_bus, x86ms->gsi); - if (kvm_pic_in_kernel()) { - i8259 = kvm_i8259_init(isa_bus); - } else if (xen_enabled()) { - i8259 = xen_interrupt_controller_init(); - } else { - i8259 = i8259_init(isa_bus, pc_allocate_cpu_irq()); - } + pc_i8259_create(isa_bus, gsi_state->i8259_irq); - for (i = 0; i < ISA_NUM_IRQS; i++) { - gsi_state->i8259_irq[i] = i8259[i]; - } - g_free(i8259); if (pcmc->pci_enabled) { ioapic_init_gsi(gsi_state, "i440fx"); } diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c index 6cf12b792b..157d1daa60 100644 --- a/hw/i386/pc_q35.c +++ b/hw/i386/pc_q35.c @@ -130,7 +130,6 @@ static void pc_q35_init(MachineState *machine) MemoryRegion *ram_memory; GSIState *gsi_state; ISABus *isa_bus; - qemu_irq *i8259; int i; ICH9LPCState *ich9_lpc; PCIDevice *ahci; @@ -257,18 +256,7 @@ static void pc_q35_init(MachineState *machine) pci_bus_set_route_irq_fn(host_bus, ich9_route_intx_pin_to_irq); isa_bus = ich9_lpc->isa_bus; - if (kvm_pic_in_kernel()) { - i8259 = kvm_i8259_init(isa_bus); - } else if (xen_enabled()) { - i8259 = xen_interrupt_controller_init(); - } else { - i8259 = i8259_init(isa_bus, pc_allocate_cpu_irq()); - } - - for (i = 0; i < ISA_NUM_IRQS; i++) { - gsi_state->i8259_irq[i] = i8259[i]; - } - g_free(i8259); + pc_i8259_create(isa_bus, gsi_state->i8259_irq); if (pcmc->pci_enabled) { ioapic_init_gsi(gsi_state, "q35"); diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h index 8c5dc39d84..5923318ea5 100644 --- a/include/hw/i386/pc.h +++ b/include/hw/i386/pc.h @@ -219,6 +219,7 @@ void pc_pci_device_init(PCIBus *pci_bus); typedef void (*cpu_set_smm_t)(int smm, void *arg); +void pc_i8259_create(ISABus *isa_bus, qemu_irq *i8259_irqs); void ioapic_init_gsi(GSIState *gsi_state, const char *parent_name); ISADevice *pc_find_fdc0(void); -- cgit v1.2.3-55-g7522 From 53e4b8018e600c256bd60e5898f6747eddbc2131 Mon Sep 17 00:00:00 2001 From: Hervé Poussineau Date: Fri, 18 Oct 2019 15:35:44 +0200 Subject: mc146818rtc: move structure to header file We are now able to embed a timer in another object. Acked-by: Michael S. Tsirkin Acked-by: Paolo Bonzini Signed-off-by: Hervé Poussineau Message-Id: <20171216090228.28505-4-hpoussin@reactos.org> Reviewed-by: Aleksandar Markovic Signed-off-by: Philippe Mathieu-Daudé Message-Id: <20191018133547.10936-2-philmd@redhat.com> Signed-off-by: Paolo Bonzini --- hw/timer/mc146818rtc.c | 30 ------------------------------ include/hw/timer/mc146818rtc.h | 33 +++++++++++++++++++++++++++++++++ 2 files changed, 33 insertions(+), 30 deletions(-) (limited to 'include') diff --git a/hw/timer/mc146818rtc.c b/hw/timer/mc146818rtc.c index c979db0992..09edb934e5 100644 --- a/hw/timer/mc146818rtc.c +++ b/hw/timer/mc146818rtc.c @@ -71,36 +71,6 @@ #define RTC_CLOCK_RATE 32768 #define UIP_HOLD_LENGTH (8 * NANOSECONDS_PER_SECOND / 32768) -#define MC146818_RTC(obj) OBJECT_CHECK(RTCState, (obj), TYPE_MC146818_RTC) - -typedef struct RTCState { - ISADevice parent_obj; - - MemoryRegion io; - MemoryRegion coalesced_io; - uint8_t cmos_data[128]; - uint8_t cmos_index; - int32_t base_year; - uint64_t base_rtc; - uint64_t last_update; - int64_t offset; - qemu_irq irq; - int it_shift; - /* periodic timer */ - QEMUTimer *periodic_timer; - int64_t next_periodic_time; - /* update-ended timer */ - QEMUTimer *update_timer; - uint64_t next_alarm_time; - uint16_t irq_reinject_on_ack_count; - uint32_t irq_coalesced; - uint32_t period; - QEMUTimer *coalesced_timer; - LostTickPolicy lost_tick_policy; - Notifier suspend_notifier; - QLIST_ENTRY(RTCState) link; -} RTCState; - static void rtc_set_time(RTCState *s); static void rtc_update_time(RTCState *s); static void rtc_set_cmos(RTCState *s, const struct tm *tm); diff --git a/include/hw/timer/mc146818rtc.h b/include/hw/timer/mc146818rtc.h index fe6ed63f71..0f1c886e5b 100644 --- a/include/hw/timer/mc146818rtc.h +++ b/include/hw/timer/mc146818rtc.h @@ -1,10 +1,43 @@ #ifndef MC146818RTC_H #define MC146818RTC_H +#include "qapi/qapi-types-misc.h" +#include "qemu/queue.h" +#include "qemu/timer.h" #include "hw/isa/isa.h" #include "hw/timer/mc146818rtc_regs.h" #define TYPE_MC146818_RTC "mc146818rtc" +#define MC146818_RTC(obj) OBJECT_CHECK(RTCState, (obj), TYPE_MC146818_RTC) + +typedef struct RTCState { + ISADevice parent_obj; + + MemoryRegion io; + MemoryRegion coalesced_io; + uint8_t cmos_data[128]; + uint8_t cmos_index; + int32_t base_year; + uint64_t base_rtc; + uint64_t last_update; + int64_t offset; + qemu_irq irq; + int it_shift; + /* periodic timer */ + QEMUTimer *periodic_timer; + int64_t next_periodic_time; + /* update-ended timer */ + QEMUTimer *update_timer; + uint64_t next_alarm_time; + uint16_t irq_reinject_on_ack_count; + uint32_t irq_coalesced; + uint32_t period; + QEMUTimer *coalesced_timer; + Notifier clock_reset_notifier; + LostTickPolicy lost_tick_policy; + Notifier suspend_notifier; + QLIST_ENTRY(RTCState) link; +} RTCState; ISADevice *mc146818_rtc_init(ISABus *bus, int base_year, qemu_irq intercept_irq); -- cgit v1.2.3-55-g7522 From d578f7dfb4e4b911fd7c6418b9344dc3dccad7f3 Mon Sep 17 00:00:00 2001 From: Philippe Mathieu-Daudé Date: Fri, 18 Oct 2019 15:35:45 +0200 Subject: mc146818rtc: Move RTC_ISA_IRQ definition The ISA default number for the RTC devices is not related to its registers neither. Move this definition to "hw/timer/mc146818rtc.h". Signed-off-by: Philippe Mathieu-Daudé Message-Id: <20191018133547.10936-3-philmd@redhat.com> Reviewed-by: Thomas Huth Signed-off-by: Paolo Bonzini Signed-off-by: Philippe Mathieu-Daudé --- include/hw/timer/mc146818rtc.h | 2 ++ include/hw/timer/mc146818rtc_regs.h | 2 -- tests/rtc-test.c | 1 + 3 files changed, 3 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/hw/timer/mc146818rtc.h b/include/hw/timer/mc146818rtc.h index 0f1c886e5b..17761cf6d9 100644 --- a/include/hw/timer/mc146818rtc.h +++ b/include/hw/timer/mc146818rtc.h @@ -39,6 +39,8 @@ typedef struct RTCState { QLIST_ENTRY(RTCState) link; } RTCState; +#define RTC_ISA_IRQ 8 + ISADevice *mc146818_rtc_init(ISABus *bus, int base_year, qemu_irq intercept_irq); void rtc_set_memory(ISADevice *dev, int addr, int val); diff --git a/include/hw/timer/mc146818rtc_regs.h b/include/hw/timer/mc146818rtc_regs.h index bfbb57e570..631f71cfd9 100644 --- a/include/hw/timer/mc146818rtc_regs.h +++ b/include/hw/timer/mc146818rtc_regs.h @@ -27,8 +27,6 @@ #include "qemu/timer.h" -#define RTC_ISA_IRQ 8 - #define RTC_SECONDS 0 #define RTC_SECONDS_ALARM 1 #define RTC_MINUTES 2 diff --git a/tests/rtc-test.c b/tests/rtc-test.c index 6309b0ef6c..18f895690f 100644 --- a/tests/rtc-test.c +++ b/tests/rtc-test.c @@ -15,6 +15,7 @@ #include "libqtest-single.h" #include "qemu/timer.h" +#include "hw/timer/mc146818rtc.h" #include "hw/timer/mc146818rtc_regs.h" #define UIP_HOLD_LENGTH (8 * NANOSECONDS_PER_SECOND / 32768) -- cgit v1.2.3-55-g7522 From d93884e89bb6bf26ebffdee5887907cf940b3090 Mon Sep 17 00:00:00 2001 From: Philippe Mathieu-Daudé Date: Fri, 18 Oct 2019 15:35:46 +0200 Subject: mc146818rtc: Include mc146818rtc_regs.h directly in mc146818rtc.c Devices/boards wanting to use the MC146818 RTC don't need the knowledge its internal registers. Move the "mc146818rtc_regs.h" inclusion to mc146818rtc.c where it is required. We can not move this file from include/hw/timer/ to hw/timer/ for local inclusion because the ACPI FADT table use the RTC_CENTURY register address. Signed-off-by: Philippe Mathieu-Daudé Message-Id: <20191018133547.10936-4-philmd@redhat.com> Signed-off-by: Paolo Bonzini Signed-off-by: Philippe Mathieu-Daudé --- hw/timer/mc146818rtc.c | 1 + include/hw/timer/mc146818rtc.h | 1 - 2 files changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/hw/timer/mc146818rtc.c b/hw/timer/mc146818rtc.c index 09edb934e5..dce4dae483 100644 --- a/hw/timer/mc146818rtc.c +++ b/hw/timer/mc146818rtc.c @@ -40,6 +40,7 @@ #include "qapi/qapi-events-misc-target.h" #include "qapi/visitor.h" #include "exec/address-spaces.h" +#include "hw/timer/mc146818rtc_regs.h" #ifdef TARGET_I386 #include "qapi/qapi-commands-misc-target.h" diff --git a/include/hw/timer/mc146818rtc.h b/include/hw/timer/mc146818rtc.h index 17761cf6d9..a857dcdc69 100644 --- a/include/hw/timer/mc146818rtc.h +++ b/include/hw/timer/mc146818rtc.h @@ -5,7 +5,6 @@ #include "qemu/queue.h" #include "qemu/timer.h" #include "hw/isa/isa.h" -#include "hw/timer/mc146818rtc_regs.h" #define TYPE_MC146818_RTC "mc146818rtc" #define MC146818_RTC(obj) OBJECT_CHECK(RTCState, (obj), TYPE_MC146818_RTC) -- cgit v1.2.3-55-g7522 From 038adc2f5850e32019bda06c559d0301be436eae Mon Sep 17 00:00:00 2001 From: Wei Yang Date: Sun, 13 Oct 2019 10:11:45 +0800 Subject: core: replace getpagesize() with qemu_real_host_page_size There are three page size in qemu: real host page size host page size target page size All of them have dedicate variable to represent. For the last two, we use the same form in the whole qemu project, while for the first one we use two forms: qemu_real_host_page_size and getpagesize(). qemu_real_host_page_size is defined to be a replacement of getpagesize(), so let it serve the role. [Note] Not fully tested for some arch or device. Signed-off-by: Wei Yang Message-Id: <20191013021145.16011-3-richardw.yang@linux.intel.com> Signed-off-by: Paolo Bonzini --- accel/kvm/kvm-all.c | 6 +++--- backends/hostmem.c | 2 +- block.c | 4 ++-- block/file-posix.c | 9 +++++---- block/io.c | 2 +- block/parallels.c | 2 +- block/qcow2-cache.c | 2 +- contrib/vhost-user-gpu/vugbm.c | 2 +- exec.c | 6 +++--- hw/intc/s390_flic_kvm.c | 2 +- hw/ppc/mac_newworld.c | 2 +- hw/ppc/spapr_pci.c | 2 +- hw/rdma/vmw/pvrdma_main.c | 2 +- hw/vfio/spapr.c | 7 ++++--- include/exec/ram_addr.h | 2 +- include/qemu/osdep.h | 4 ++-- migration/migration.c | 2 +- migration/postcopy-ram.c | 4 ++-- monitor/misc.c | 2 +- target/ppc/kvm.c | 2 +- tests/vhost-user-bridge.c | 8 ++++---- util/mmap-alloc.c | 10 +++++----- util/oslib-posix.c | 4 ++-- util/oslib-win32.c | 2 +- util/vfio-helpers.c | 12 ++++++------ 25 files changed, 52 insertions(+), 50 deletions(-) (limited to 'include') diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c index d2d96d73e8..140b0bd8f6 100644 --- a/accel/kvm/kvm-all.c +++ b/accel/kvm/kvm-all.c @@ -52,7 +52,7 @@ /* KVM uses PAGE_SIZE in its definition of KVM_COALESCED_MMIO_MAX. We * need to use the real host PAGE_SIZE, as that's what KVM will use. */ -#define PAGE_SIZE getpagesize() +#define PAGE_SIZE qemu_real_host_page_size //#define DEBUG_KVM @@ -507,7 +507,7 @@ static int kvm_get_dirty_pages_log_range(MemoryRegionSection *section, { ram_addr_t start = section->offset_within_region + memory_region_get_ram_addr(section->mr); - ram_addr_t pages = int128_get64(section->size) / getpagesize(); + ram_addr_t pages = int128_get64(section->size) / qemu_real_host_page_size; cpu_physical_memory_set_dirty_lebitmap(bitmap, start, pages); return 0; @@ -1841,7 +1841,7 @@ static int kvm_init(MachineState *ms) * even with KVM. TARGET_PAGE_SIZE is assumed to be the minimum * page size for the system though. */ - assert(TARGET_PAGE_SIZE <= getpagesize()); + assert(TARGET_PAGE_SIZE <= qemu_real_host_page_size); s->sigmask_len = 8; diff --git a/backends/hostmem.c b/backends/hostmem.c index 6d333dc23c..e773bdfa6e 100644 --- a/backends/hostmem.c +++ b/backends/hostmem.c @@ -304,7 +304,7 @@ size_t host_memory_backend_pagesize(HostMemoryBackend *memdev) #else size_t host_memory_backend_pagesize(HostMemoryBackend *memdev) { - return getpagesize(); + return qemu_real_host_page_size; } #endif diff --git a/block.c b/block.c index dad5a3d8e0..4cffc2bc35 100644 --- a/block.c +++ b/block.c @@ -106,7 +106,7 @@ size_t bdrv_opt_mem_align(BlockDriverState *bs) { if (!bs || !bs->drv) { /* page size or 4k (hdd sector size) should be on the safe side */ - return MAX(4096, getpagesize()); + return MAX(4096, qemu_real_host_page_size); } return bs->bl.opt_mem_alignment; @@ -116,7 +116,7 @@ size_t bdrv_min_mem_align(BlockDriverState *bs) { if (!bs || !bs->drv) { /* page size or 4k (hdd sector size) should be on the safe side */ - return MAX(4096, getpagesize()); + return MAX(4096, qemu_real_host_page_size); } return bs->bl.min_mem_alignment; diff --git a/block/file-posix.c b/block/file-posix.c index 695fcf740d..5d1995a07c 100644 --- a/block/file-posix.c +++ b/block/file-posix.c @@ -327,7 +327,7 @@ static void raw_probe_alignment(BlockDriverState *bs, int fd, Error **errp) { BDRVRawState *s = bs->opaque; char *buf; - size_t max_align = MAX(MAX_BLOCKSIZE, getpagesize()); + size_t max_align = MAX(MAX_BLOCKSIZE, qemu_real_host_page_size); size_t alignments[] = {1, 512, 1024, 2048, 4096}; /* For SCSI generic devices the alignment is not really used. @@ -1136,13 +1136,14 @@ static void raw_refresh_limits(BlockDriverState *bs, Error **errp) ret = sg_get_max_segments(s->fd); if (ret > 0) { - bs->bl.max_transfer = MIN(bs->bl.max_transfer, ret * getpagesize()); + bs->bl.max_transfer = MIN(bs->bl.max_transfer, + ret * qemu_real_host_page_size); } } raw_probe_alignment(bs, s->fd, errp); bs->bl.min_mem_alignment = s->buf_align; - bs->bl.opt_mem_alignment = MAX(s->buf_align, getpagesize()); + bs->bl.opt_mem_alignment = MAX(s->buf_align, qemu_real_host_page_size); } static int check_for_dasd(int fd) @@ -1705,7 +1706,7 @@ static int allocate_first_block(int fd, size_t max_size) size_t write_size = (max_size < MAX_BLOCKSIZE) ? BDRV_SECTOR_SIZE : MAX_BLOCKSIZE; - size_t max_align = MAX(MAX_BLOCKSIZE, getpagesize()); + size_t max_align = MAX(MAX_BLOCKSIZE, qemu_real_host_page_size); void *buf; ssize_t n; int ret; diff --git a/block/io.c b/block/io.c index f0b86c1d19..e46d9e8b97 100644 --- a/block/io.c +++ b/block/io.c @@ -160,7 +160,7 @@ void bdrv_refresh_limits(BlockDriverState *bs, Error **errp) bdrv_merge_limits(&bs->bl, &bs->file->bs->bl); } else { bs->bl.min_mem_alignment = 512; - bs->bl.opt_mem_alignment = getpagesize(); + bs->bl.opt_mem_alignment = qemu_real_host_page_size; /* Safe default since most protocols use readv()/writev()/etc */ bs->bl.max_iov = IOV_MAX; diff --git a/block/parallels.c b/block/parallels.c index 7cd2714b69..f1dfb03eef 100644 --- a/block/parallels.c +++ b/block/parallels.c @@ -847,7 +847,7 @@ static int parallels_open(BlockDriverState *bs, QDict *options, int flags, } } - s->bat_dirty_block = 4 * getpagesize(); + s->bat_dirty_block = 4 * qemu_real_host_page_size; s->bat_dirty_bmap = bitmap_new(DIV_ROUND_UP(s->header_size, s->bat_dirty_block)); diff --git a/block/qcow2-cache.c b/block/qcow2-cache.c index d29b038a67..7444b9c4ab 100644 --- a/block/qcow2-cache.c +++ b/block/qcow2-cache.c @@ -74,7 +74,7 @@ static void qcow2_cache_table_release(Qcow2Cache *c, int i, int num_tables) /* Using MADV_DONTNEED to discard memory is a Linux-specific feature */ #ifdef CONFIG_LINUX void *t = qcow2_cache_get_table_addr(c, i); - int align = getpagesize(); + int align = qemu_real_host_page_size; size_t mem_size = (size_t) c->table_size * num_tables; size_t offset = QEMU_ALIGN_UP((uintptr_t) t, align) - (uintptr_t) t; size_t length = QEMU_ALIGN_DOWN(mem_size - offset, align); diff --git a/contrib/vhost-user-gpu/vugbm.c b/contrib/vhost-user-gpu/vugbm.c index d3bb82ff0e..9c357b6399 100644 --- a/contrib/vhost-user-gpu/vugbm.c +++ b/contrib/vhost-user-gpu/vugbm.c @@ -52,7 +52,7 @@ struct udmabuf_create { static size_t udmabuf_get_size(struct vugbm_buffer *buf) { - return ROUND_UP(buf->width * buf->height * 4, getpagesize()); + return ROUND_UP(buf->width * buf->height * 4, qemu_real_host_page_size); } static bool diff --git a/exec.c b/exec.c index fb0943cfed..9fc0d01895 100644 --- a/exec.c +++ b/exec.c @@ -1756,11 +1756,11 @@ long qemu_maxrampagesize(void) #else long qemu_minrampagesize(void) { - return getpagesize(); + return qemu_real_host_page_size; } long qemu_maxrampagesize(void) { - return getpagesize(); + return qemu_real_host_page_size; } #endif @@ -2417,7 +2417,7 @@ RAMBlock *qemu_ram_alloc_internal(ram_addr_t size, ram_addr_t max_size, new_block->max_length = max_size; assert(max_size >= size); new_block->fd = -1; - new_block->page_size = getpagesize(); + new_block->page_size = qemu_real_host_page_size; new_block->host = host; if (host) { new_block->flags |= RAM_PREALLOC; diff --git a/hw/intc/s390_flic_kvm.c b/hw/intc/s390_flic_kvm.c index cedccba8a9..c9ee80eaae 100644 --- a/hw/intc/s390_flic_kvm.c +++ b/hw/intc/s390_flic_kvm.c @@ -25,7 +25,7 @@ #include "migration/qemu-file-types.h" #include "trace.h" -#define FLIC_SAVE_INITIAL_SIZE getpagesize() +#define FLIC_SAVE_INITIAL_SIZE qemu_real_host_page_size #define FLIC_FAILED (-1UL) #define FLIC_SAVEVM_VERSION 1 diff --git a/hw/ppc/mac_newworld.c b/hw/ppc/mac_newworld.c index c5bbcc7433..3594517f0c 100644 --- a/hw/ppc/mac_newworld.c +++ b/hw/ppc/mac_newworld.c @@ -439,7 +439,7 @@ static void ppc_core99_init(MachineState *machine) } /* The NewWorld NVRAM is not located in the MacIO device */ - if (kvm_enabled() && getpagesize() > 4096) { + if (kvm_enabled() && qemu_real_host_page_size > 4096) { /* We can't combine read-write and read-only in a single page, so move the NVRAM out of ROM again for KVM */ nvram_addr = 0xFFE00000; diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c index cc0e7829b6..f6fbcf99ed 100644 --- a/hw/ppc/spapr_pci.c +++ b/hw/ppc/spapr_pci.c @@ -1942,7 +1942,7 @@ static void spapr_phb_realize(DeviceState *dev, Error **errp) * our memory slot is of page size granularity. */ if (kvm_enabled()) { - msi_window_size = getpagesize(); + msi_window_size = qemu_real_host_page_size; } memory_region_init_io(&sphb->msiwindow, OBJECT(sphb), &spapr_msi_ops, spapr, diff --git a/hw/rdma/vmw/pvrdma_main.c b/hw/rdma/vmw/pvrdma_main.c index 3e36e13013..3722d9e772 100644 --- a/hw/rdma/vmw/pvrdma_main.c +++ b/hw/rdma/vmw/pvrdma_main.c @@ -601,7 +601,7 @@ static void pvrdma_realize(PCIDevice *pdev, Error **errp) rdma_info_report("Initializing device %s %x.%x", pdev->name, PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn)); - if (TARGET_PAGE_SIZE != getpagesize()) { + if (TARGET_PAGE_SIZE != qemu_real_host_page_size) { error_setg(errp, "Target page size must be the same as host page size"); return; } diff --git a/hw/vfio/spapr.c b/hw/vfio/spapr.c index e853eebe11..33692fc86f 100644 --- a/hw/vfio/spapr.c +++ b/hw/vfio/spapr.c @@ -196,14 +196,15 @@ int vfio_spapr_create_window(VFIOContainer *container, * bits_per_level is a safe guess of how much we can allocate per level: * 8 is the current minimum for CONFIG_FORCE_MAX_ZONEORDER and MAX_ORDER * is usually bigger than that. - * Below we look at getpagesize() as TCEs are allocated from system pages. + * Below we look at qemu_real_host_page_size as TCEs are allocated from + * system pages. */ - bits_per_level = ctz64(getpagesize()) + 8; + bits_per_level = ctz64(qemu_real_host_page_size) + 8; create.levels = bits_total / bits_per_level; if (bits_total % bits_per_level) { ++create.levels; } - max_levels = (64 - create.page_shift) / ctz64(getpagesize()); + max_levels = (64 - create.page_shift) / ctz64(qemu_real_host_page_size); for ( ; create.levels <= max_levels; ++create.levels) { ret = ioctl(container->fd, VFIO_IOMMU_SPAPR_TCE_CREATE, &create); if (!ret) { diff --git a/include/exec/ram_addr.h b/include/exec/ram_addr.h index ad158bb247..bed0554f4d 100644 --- a/include/exec/ram_addr.h +++ b/include/exec/ram_addr.h @@ -373,7 +373,7 @@ static inline void cpu_physical_memory_set_dirty_lebitmap(unsigned long *bitmap, hwaddr addr; ram_addr_t ram_addr; unsigned long len = (pages + HOST_LONG_BITS - 1) / HOST_LONG_BITS; - unsigned long hpratio = getpagesize() / TARGET_PAGE_SIZE; + unsigned long hpratio = qemu_real_host_page_size / TARGET_PAGE_SIZE; unsigned long page = BIT_WORD(start >> TARGET_PAGE_BITS); /* start address is aligned at the start of a word? */ diff --git a/include/qemu/osdep.h b/include/qemu/osdep.h index c7d242f476..0f97d68586 100644 --- a/include/qemu/osdep.h +++ b/include/qemu/osdep.h @@ -423,9 +423,9 @@ void qemu_anon_ram_free(void *ptr, size_t size); # define QEMU_VMALLOC_ALIGN (256 * 4096) #elif defined(__linux__) && defined(__sparc__) #include -# define QEMU_VMALLOC_ALIGN MAX(getpagesize(), SHMLBA) +# define QEMU_VMALLOC_ALIGN MAX(qemu_real_host_page_size, SHMLBA) #else -# define QEMU_VMALLOC_ALIGN getpagesize() +# define QEMU_VMALLOC_ALIGN qemu_real_host_page_size #endif #ifdef CONFIG_POSIX diff --git a/migration/migration.c b/migration/migration.c index 3febd0f8f3..4133ed2684 100644 --- a/migration/migration.c +++ b/migration/migration.c @@ -2284,7 +2284,7 @@ static struct rp_cmd_args { static void migrate_handle_rp_req_pages(MigrationState *ms, const char* rbname, ram_addr_t start, size_t len) { - long our_host_ps = getpagesize(); + long our_host_ps = qemu_real_host_page_size; trace_migrate_handle_rp_req_pages(rbname, start, len); diff --git a/migration/postcopy-ram.c b/migration/postcopy-ram.c index abccafc8c8..a36402722b 100644 --- a/migration/postcopy-ram.c +++ b/migration/postcopy-ram.c @@ -308,7 +308,7 @@ static bool ufd_check_and_apply(int ufd, MigrationIncomingState *mis) return false; } - if (getpagesize() != ram_pagesize_summary()) { + if (qemu_real_host_page_size != ram_pagesize_summary()) { bool have_hp = false; /* We've got a huge page */ #ifdef UFFD_FEATURE_MISSING_HUGETLBFS @@ -346,7 +346,7 @@ static int test_ramblock_postcopiable(RAMBlock *rb, void *opaque) */ bool postcopy_ram_supported_by_host(MigrationIncomingState *mis) { - long pagesize = getpagesize(); + long pagesize = qemu_real_host_page_size; int ufd = -1; bool ret = false; /* Error unless we change it */ void *testarea = NULL; diff --git a/monitor/misc.c b/monitor/misc.c index aef16f6cfb..3baa15f3bf 100644 --- a/monitor/misc.c +++ b/monitor/misc.c @@ -862,7 +862,7 @@ static uint64_t vtop(void *ptr, Error **errp) uint64_t pinfo; uint64_t ret = -1; uintptr_t addr = (uintptr_t) ptr; - uintptr_t pagesize = getpagesize(); + uintptr_t pagesize = qemu_real_host_page_size; off_t offset = addr / pagesize * sizeof(pinfo); int fd; diff --git a/target/ppc/kvm.c b/target/ppc/kvm.c index 820724cc7d..7d2e8969ac 100644 --- a/target/ppc/kvm.c +++ b/target/ppc/kvm.c @@ -411,7 +411,7 @@ void kvm_check_mmu(PowerPCCPU *cpu, Error **errp) * will be a normal mapping, not a special hugepage one used * for RAM. */ - if (getpagesize() < 0x10000) { + if (qemu_real_host_page_size < 0x10000) { error_setg(errp, "KVM can't supply 64kiB CI pages, which guest expects"); } diff --git a/tests/vhost-user-bridge.c b/tests/vhost-user-bridge.c index c4e350e1f5..6c3d490611 100644 --- a/tests/vhost-user-bridge.c +++ b/tests/vhost-user-bridge.c @@ -468,8 +468,8 @@ vubr_queue_set_started(VuDev *dev, int qidx, bool started) if (started && vubr->notifier.fd >= 0) { vu_set_queue_host_notifier(dev, vq, vubr->notifier.fd, - getpagesize(), - qidx * getpagesize()); + qemu_real_host_page_size, + qidx * qemu_real_host_page_size); } if (qidx % 2 == 1) { @@ -594,7 +594,7 @@ static void *notifier_thread(void *arg) { VuDev *dev = (VuDev *)arg; VubrDev *vubr = container_of(dev, VubrDev, vudev); - int pagesize = getpagesize(); + int pagesize = qemu_real_host_page_size; int qidx; while (true) { @@ -630,7 +630,7 @@ vubr_host_notifier_setup(VubrDev *dev) void *addr; int fd; - length = getpagesize() * VHOST_USER_BRIDGE_MAX_QUEUES; + length = qemu_real_host_page_size * VHOST_USER_BRIDGE_MAX_QUEUES; fd = mkstemp(template); if (fd < 0) { diff --git a/util/mmap-alloc.c b/util/mmap-alloc.c index f7f177d0ea..27dcccd8ec 100644 --- a/util/mmap-alloc.c +++ b/util/mmap-alloc.c @@ -48,7 +48,7 @@ size_t qemu_fd_getpagesize(int fd) #endif #endif - return getpagesize(); + return qemu_real_host_page_size; } size_t qemu_mempath_getpagesize(const char *mem_path) @@ -79,7 +79,7 @@ size_t qemu_mempath_getpagesize(const char *mem_path) #endif #endif - return getpagesize(); + return qemu_real_host_page_size; } void *qemu_ram_mmap(int fd, @@ -114,7 +114,7 @@ void *qemu_ram_mmap(int fd, */ flags = MAP_PRIVATE; pagesize = qemu_fd_getpagesize(fd); - if (fd == -1 || pagesize == getpagesize()) { + if (fd == -1 || pagesize == qemu_real_host_page_size) { guardfd = -1; flags |= MAP_ANONYMOUS; } else { @@ -123,7 +123,7 @@ void *qemu_ram_mmap(int fd, } #else guardfd = -1; - pagesize = getpagesize(); + pagesize = qemu_real_host_page_size; flags = MAP_PRIVATE | MAP_ANONYMOUS; #endif @@ -205,7 +205,7 @@ void qemu_ram_munmap(int fd, void *ptr, size_t size) #if defined(__powerpc64__) && defined(__linux__) pagesize = qemu_fd_getpagesize(fd); #else - pagesize = getpagesize(); + pagesize = qemu_real_host_page_size; #endif munmap(ptr, size + pagesize); } diff --git a/util/oslib-posix.c b/util/oslib-posix.c index f8693384fc..5a291cc982 100644 --- a/util/oslib-posix.c +++ b/util/oslib-posix.c @@ -617,7 +617,7 @@ void *qemu_alloc_stack(size_t *sz) #ifdef CONFIG_DEBUG_STACK_USAGE void *ptr2; #endif - size_t pagesz = getpagesize(); + size_t pagesz = qemu_real_host_page_size; #ifdef _SC_THREAD_STACK_MIN /* avoid stacks smaller than _SC_THREAD_STACK_MIN */ long min_stack_sz = sysconf(_SC_THREAD_STACK_MIN); @@ -679,7 +679,7 @@ void qemu_free_stack(void *stack, size_t sz) unsigned int usage; void *ptr; - for (ptr = stack + getpagesize(); ptr < stack + sz; + for (ptr = stack + qemu_real_host_page_size; ptr < stack + sz; ptr += sizeof(uint32_t)) { if (*(uint32_t *)ptr != 0xdeadbeaf) { break; diff --git a/util/oslib-win32.c b/util/oslib-win32.c index 886e400d6a..e9b14ab178 100644 --- a/util/oslib-win32.c +++ b/util/oslib-win32.c @@ -554,7 +554,7 @@ void os_mem_prealloc(int fd, char *area, size_t memory, int smp_cpus, Error **errp) { int i; - size_t pagesize = getpagesize(); + size_t pagesize = qemu_real_host_page_size; memory = (memory + pagesize - 1) & -pagesize; for (i = 0; i < memory / pagesize; i++) { diff --git a/util/vfio-helpers.c b/util/vfio-helpers.c index 26ffd0d6b5..813f7ec564 100644 --- a/util/vfio-helpers.c +++ b/util/vfio-helpers.c @@ -514,9 +514,9 @@ static IOVAMapping *qemu_vfio_add_mapping(QEMUVFIOState *s, IOVAMapping m = {.host = host, .size = size, .iova = iova}; IOVAMapping *insert; - assert(QEMU_IS_ALIGNED(size, getpagesize())); - assert(QEMU_IS_ALIGNED(s->low_water_mark, getpagesize())); - assert(QEMU_IS_ALIGNED(s->high_water_mark, getpagesize())); + assert(QEMU_IS_ALIGNED(size, qemu_real_host_page_size)); + assert(QEMU_IS_ALIGNED(s->low_water_mark, qemu_real_host_page_size)); + assert(QEMU_IS_ALIGNED(s->high_water_mark, qemu_real_host_page_size)); trace_qemu_vfio_new_mapping(s, host, size, index, iova); assert(index >= 0); @@ -567,7 +567,7 @@ static void qemu_vfio_undo_mapping(QEMUVFIOState *s, IOVAMapping *mapping, index = mapping - s->mappings; assert(mapping->size > 0); - assert(QEMU_IS_ALIGNED(mapping->size, getpagesize())); + assert(QEMU_IS_ALIGNED(mapping->size, qemu_real_host_page_size)); assert(index >= 0 && index < s->nr_mappings); if (ioctl(s->container, VFIO_IOMMU_UNMAP_DMA, &unmap)) { error_setg(errp, "VFIO_UNMAP_DMA failed: %d", -errno); @@ -613,8 +613,8 @@ int qemu_vfio_dma_map(QEMUVFIOState *s, void *host, size_t size, IOVAMapping *mapping; uint64_t iova0; - assert(QEMU_PTR_IS_ALIGNED(host, getpagesize())); - assert(QEMU_IS_ALIGNED(size, getpagesize())); + assert(QEMU_PTR_IS_ALIGNED(host, qemu_real_host_page_size)); + assert(QEMU_IS_ALIGNED(size, qemu_real_host_page_size)); trace_qemu_vfio_dma_map(s, host, size, temporary, iova); qemu_mutex_lock(&s->lock); mapping = qemu_vfio_find_mapping(s, host, &index); -- cgit v1.2.3-55-g7522 From 6f529b7534c534afe2f2b834199191d8b4cc07ca Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Wed, 16 Oct 2019 10:18:10 +0200 Subject: target/i386: move FERR handling to target/i386 Move it out of pc.c since it is strictly tied to TCG. This is almost exclusively code movement, the next patch will implement IGNNE. Signed-off-by: Paolo Bonzini --- hw/i386/pc.c | 17 +++-------------- hw/i386/pc_piix.c | 4 +++- hw/i386/pc_q35.c | 4 +++- include/hw/i386/pc.h | 1 - target/i386/cpu.h | 3 ++- target/i386/fpu_helper.c | 26 +++++++++++++++++++++++++- 6 files changed, 36 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/hw/i386/pc.c b/hw/i386/pc.c index b8f02c6f3f..c1a39de59f 100644 --- a/hw/i386/pc.c +++ b/hw/i386/pc.c @@ -381,23 +381,12 @@ static uint64_t ioport80_read(void *opaque, hwaddr addr, unsigned size) } /* MSDOS compatibility mode FPU exception support */ -static qemu_irq ferr_irq; - -void pc_register_ferr_irq(qemu_irq irq) -{ - ferr_irq = irq; -} - -/* XXX: add IGNNE support */ -void cpu_set_ferr(CPUX86State *s) -{ - qemu_irq_raise(ferr_irq); -} - static void ioportF0_write(void *opaque, hwaddr addr, uint64_t data, unsigned size) { - qemu_irq_lower(ferr_irq); + if (tcg_enabled()) { + cpu_clear_ferr(); + } } static uint64_t ioportF0_read(void *opaque, hwaddr addr, unsigned size) diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c index 3a4a64a38d..c15929a1f5 100644 --- a/hw/i386/pc_piix.c +++ b/hw/i386/pc_piix.c @@ -213,7 +213,9 @@ static void pc_init1(MachineState *machine, ioapic_init_gsi(gsi_state, "i440fx"); } - pc_register_ferr_irq(x86ms->gsi[13]); + if (tcg_enabled()) { + x86_register_ferr_irq(x86ms->gsi[13]); + } pc_vga_init(isa_bus, pcmc->pci_enabled ? pci_bus : NULL); diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c index d8b4c48021..d51f524727 100644 --- a/hw/i386/pc_q35.c +++ b/hw/i386/pc_q35.c @@ -261,7 +261,9 @@ static void pc_q35_init(MachineState *machine) ioapic_init_gsi(gsi_state, "q35"); } - pc_register_ferr_irq(x86ms->gsi[13]); + if (tcg_enabled()) { + x86_register_ferr_irq(x86ms->gsi[13]); + } assert(pcms->vmport != ON_OFF_AUTO__MAX); if (pcms->vmport == ON_OFF_AUTO_AUTO) { diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h index 5923318ea5..f040a72095 100644 --- a/include/hw/i386/pc.h +++ b/include/hw/i386/pc.h @@ -176,7 +176,6 @@ void vmmouse_set_data(const uint32_t *data); extern int fd_bootchk; bool pc_machine_is_smm_enabled(PCMachineState *pcms); -void pc_register_ferr_irq(qemu_irq irq); void pc_acpi_smi_interrupt(void *opaque, int irq, int level); void pc_hot_add_cpu(MachineState *ms, const int64_t id, Error **errp); diff --git a/target/i386/cpu.h b/target/i386/cpu.h index b772e82476..01e052b3ba 100644 --- a/target/i386/cpu.h +++ b/target/i386/cpu.h @@ -1761,7 +1761,8 @@ int cpu_x86_support_mca_broadcast(CPUX86State *env); int cpu_get_pic_interrupt(CPUX86State *s); /* MSDOS compatibility mode FPU exception support */ -void cpu_set_ferr(CPUX86State *s); +void x86_register_ferr_irq(qemu_irq irq); +void cpu_clear_ferr(void); /* mpx_helper.c */ void cpu_sync_bndcs_hflags(CPUX86State *env); diff --git a/target/i386/fpu_helper.c b/target/i386/fpu_helper.c index 005f1f68f8..4db0059676 100644 --- a/target/i386/fpu_helper.c +++ b/target/i386/fpu_helper.c @@ -26,6 +26,10 @@ #include "exec/cpu_ldst.h" #include "fpu/softfloat.h" +#ifdef CONFIG_SOFTMMU +#include "hw/irq.h" +#endif + #define FPU_RC_MASK 0xc00 #define FPU_RC_NEAR 0x000 #define FPU_RC_DOWN 0x400 @@ -58,6 +62,26 @@ #define floatx80_l2e make_floatx80(0x3fff, 0xb8aa3b295c17f0bcLL) #define floatx80_l2t make_floatx80(0x4000, 0xd49a784bcd1b8afeLL) +#if !defined(CONFIG_USER_ONLY) +static qemu_irq ferr_irq; + +void x86_register_ferr_irq(qemu_irq irq) +{ + ferr_irq = irq; +} + +void cpu_clear_ferr(void) +{ + qemu_irq_lower(ferr_irq); +} + +static void cpu_set_ferr(void) +{ + qemu_irq_raise(ferr_irq); +} +#endif + + static inline void fpush(CPUX86State *env) { env->fpstt = (env->fpstt - 1) & 7; @@ -137,7 +161,7 @@ static void fpu_raise_exception(CPUX86State *env, uintptr_t retaddr) } #if !defined(CONFIG_USER_ONLY) else { - cpu_set_ferr(env); + cpu_set_ferr(); } #endif } -- cgit v1.2.3-55-g7522