diff options
Diffstat (limited to 'hw/xen')
-rw-r--r-- | hw/xen/meson.build | 4 | ||||
-rw-r--r-- | hw/xen/trace-events | 6 | ||||
-rw-r--r-- | hw/xen/xen-bus-helper.c | 37 | ||||
-rw-r--r-- | hw/xen/xen-bus.c | 35 | ||||
-rw-r--r-- | hw/xen/xen-hvm-common.c | 114 | ||||
-rw-r--r-- | hw/xen/xen-legacy-backend.c | 50 | ||||
-rw-r--r-- | hw/xen/xen-mapcache.c | 37 | ||||
-rw-r--r-- | hw/xen/xen-pvh-common.c | 399 | ||||
-rw-r--r-- | hw/xen/xen_devconfig.c | 12 | ||||
-rw-r--r-- | hw/xen/xen_pt.c | 67 | ||||
-rw-r--r-- | hw/xen/xen_pt_graphics.c | 4 | ||||
-rw-r--r-- | hw/xen/xen_pvdev.c | 8 | ||||
-rw-r--r-- | hw/xen/xen_stubs.c | 51 |
13 files changed, 682 insertions, 142 deletions
diff --git a/hw/xen/meson.build b/hw/xen/meson.build index d887fa9..a1850e7 100644 --- a/hw/xen/meson.build +++ b/hw/xen/meson.build @@ -9,12 +9,16 @@ system_ss.add(when: ['CONFIG_XEN_BUS'], if_true: files( system_ss.add(when: ['CONFIG_XEN', xen], if_true: files( 'xen-operations.c', +), +if_false: files( + 'xen_stubs.c', )) xen_specific_ss = ss.source_set() xen_specific_ss.add(files( 'xen-mapcache.c', 'xen-hvm-common.c', + 'xen-pvh-common.c', )) if have_xen_pci_passthrough xen_specific_ss.add(files( diff --git a/hw/xen/trace-events b/hw/xen/trace-events index d1b27f6..b67942d 100644 --- a/hw/xen/trace-events +++ b/hw/xen/trace-events @@ -38,7 +38,7 @@ xen_device_remove_watch(const char *type, char *name, const char *node, const ch xs_node_create(const char *node) "%s" xs_node_destroy(const char *node) "%s" xs_node_vprintf(char *path, char *value) "%s %s" -xs_node_vscanf(char *path, char *value) "%s %s" +xs_node_read(const char *path, const char *value) "%s %s" xs_node_watch(char *path) "%s" xs_node_unwatch(char *path) "%s" @@ -64,6 +64,10 @@ destroy_hvm_domain_cannot_acquire_handle(void) "Cannot acquire xenctrl handle" destroy_hvm_domain_failed_action(const char *action, int sts, char *errno_s) "xc_domain_shutdown failed to issue %s, sts %d, %s" destroy_hvm_domain_action(int xen_domid, const char *action) "Issued domain %d %s" +# xen-pvh-common.c +xen_create_virtio_mmio_devices(int i, int irq, uint64_t base) "Created virtio-mmio device %d: irq %d base 0x%"PRIx64 +xen_enable_tpm(uint64_t addr) "Connected tpmdev at address 0x%"PRIx64 + # xen-mapcache.c xen_map_cache(uint64_t phys_addr) "want 0x%"PRIx64 xen_remap_bucket(uint64_t index) "index 0x%"PRIx64 diff --git a/hw/xen/xen-bus-helper.c b/hw/xen/xen-bus-helper.c index b2b2cc9..288fad4 100644 --- a/hw/xen/xen-bus-helper.c +++ b/hw/xen/xen-bus-helper.c @@ -105,25 +105,22 @@ int xs_node_vscanf(struct qemu_xs_handle *h, xs_transaction_t tid, const char *node, const char *key, Error **errp, const char *fmt, va_list ap) { - char *path, *value; + char *value; int rc; - path = (strlen(node) != 0) ? g_strdup_printf("%s/%s", node, key) : - g_strdup(key); - value = qemu_xen_xs_read(h, tid, path, NULL); - - trace_xs_node_vscanf(path, value); + if (node && strlen(node) != 0) { + value = xs_node_read(h, tid, NULL, errp, "%s/%s", node, key); + } else { + value = xs_node_read(h, tid, NULL, errp, "%s", key); + } if (value) { rc = vsscanf(value, fmt, ap); } else { - error_setg_errno(errp, errno, "failed to read from '%s'", - path); rc = EOF; } free(value); - g_free(path); return rc; } @@ -142,6 +139,28 @@ int xs_node_scanf(struct qemu_xs_handle *h, xs_transaction_t tid, return rc; } +char *xs_node_read(struct qemu_xs_handle *h, xs_transaction_t tid, + unsigned int *len, Error **errp, + const char *path_fmt, ...) +{ + char *path, *value; + va_list ap; + + va_start(ap, path_fmt); + path = g_strdup_vprintf(path_fmt, ap); + va_end(ap); + + value = qemu_xen_xs_read(h, tid, path, len); + trace_xs_node_read(path, value); + if (!value) { + error_setg_errno(errp, errno, "failed to read from '%s'", path); + } + + g_free(path); + + return value; +} + struct qemu_xs_watch *xs_node_watch(struct qemu_xs_handle *h, const char *node, const char *key, xs_watch_fn fn, void *opaque, Error **errp) diff --git a/hw/xen/xen-bus.c b/hw/xen/xen-bus.c index 95b207a..6bd2e54 100644 --- a/hw/xen/xen-bus.c +++ b/hw/xen/xen-bus.c @@ -18,8 +18,8 @@ #include "hw/xen/xen-bus-helper.h" #include "monitor/monitor.h" #include "qapi/error.h" -#include "qapi/qmp/qdict.h" -#include "sysemu/sysemu.h" +#include "qobject/qdict.h" +#include "system/system.h" #include "net/net.h" #include "trace.h" @@ -156,8 +156,8 @@ again: !strcmp(key[i], "hotplug-status")) continue; - if (xs_node_scanf(xenbus->xsh, tid, path, key[i], NULL, "%ms", - &val) == 1) { + val = xs_node_read(xenbus->xsh, tid, NULL, NULL, "%s/%s", path, key[i]); + if (val) { qdict_put_str(opts, key[i], val); free(val); } @@ -353,10 +353,10 @@ static void xen_bus_realize(BusState *bus, Error **errp) xs_node_watch(xenbus->xsh, node, key, xen_bus_backend_changed, xenbus, &local_err); if (local_err) { - /* This need not be treated as a hard error so don't propagate */ - error_reportf_err(local_err, - "failed to set up '%s' enumeration watch: ", - type[i]); + warn_reportf_err(local_err, + "failed to set up '%s' enumeration watch: ", + type[i]); + local_err = NULL; } g_free(node); @@ -380,7 +380,7 @@ static void xen_bus_unplug_request(HotplugHandler *hotplug, xen_device_unplug(xendev, errp); } -static void xen_bus_class_init(ObjectClass *class, void *data) +static void xen_bus_class_init(ObjectClass *class, const void *data) { BusClass *bus_class = BUS_CLASS(class); HotplugHandlerClass *hotplug_class = HOTPLUG_HANDLER_CLASS(class); @@ -399,7 +399,7 @@ static const TypeInfo xen_bus_type_info = { .instance_size = sizeof(XenBus), .class_size = sizeof(XenBusClass), .class_init = xen_bus_class_init, - .interfaces = (InterfaceInfo[]) { + .interfaces = (const InterfaceInfo[]) { { TYPE_HOTPLUG_HANDLER }, { } }, @@ -650,6 +650,16 @@ int xen_device_frontend_scanf(XenDevice *xendev, const char *key, return rc; } +char *xen_device_frontend_read(XenDevice *xendev, const char *key) +{ + XenBus *xenbus = XEN_BUS(qdev_get_parent_bus(DEVICE(xendev))); + + g_assert(xenbus->xsh); + + return xs_node_read(xenbus->xsh, XBT_NULL, NULL, NULL, "%s/%s", + xendev->frontend_path, key); +} + static void xen_device_frontend_set_state(XenDevice *xendev, enum xenbus_state state, bool publish) @@ -1092,13 +1102,12 @@ unrealize: xen_device_unrealize(dev); } -static Property xen_device_props[] = { +static const Property xen_device_props[] = { DEFINE_PROP_UINT16("frontend-id", XenDevice, frontend_id, DOMID_INVALID), - DEFINE_PROP_END_OF_LIST() }; -static void xen_device_class_init(ObjectClass *class, void *data) +static void xen_device_class_init(ObjectClass *class, const void *data) { DeviceClass *dev_class = DEVICE_CLASS(class); diff --git a/hw/xen/xen-hvm-common.c b/hw/xen/xen-hvm-common.c index 3a9d6f9..78e0bc8 100644 --- a/hw/xen/xen-hvm-common.c +++ b/hw/xen/xen-hvm-common.c @@ -1,14 +1,21 @@ #include "qemu/osdep.h" #include "qemu/units.h" +#include "qemu/error-report.h" #include "qapi/error.h" +#include "exec/target_long.h" #include "exec/target_page.h" #include "trace.h" +#include "hw/hw.h" #include "hw/pci/pci_host.h" #include "hw/xen/xen-hvm-common.h" #include "hw/xen/xen-bus.h" #include "hw/boards.h" #include "hw/xen/arch_hvm.h" +#include "system/runstate.h" +#include "system/system.h" +#include "system/xen.h" +#include "system/xen-mapcache.h" MemoryRegion xen_memory, xen_grants; @@ -667,6 +674,8 @@ static int xen_map_ioreq_server(XenIOState *state) xen_pfn_t ioreq_pfn; xen_pfn_t bufioreq_pfn; evtchn_port_t bufioreq_evtchn; + unsigned long num_frames = 1; + unsigned long frame = 1; int rc; /* @@ -675,65 +684,85 @@ static int xen_map_ioreq_server(XenIOState *state) */ QEMU_BUILD_BUG_ON(XENMEM_resource_ioreq_server_frame_bufioreq != 0); QEMU_BUILD_BUG_ON(XENMEM_resource_ioreq_server_frame_ioreq(0) != 1); + + if (state->has_bufioreq) { + frame = 0; + num_frames = 2; + } state->fres = xenforeignmemory_map_resource(xen_fmem, xen_domid, XENMEM_resource_ioreq_server, - state->ioservid, 0, 2, + state->ioservid, + frame, num_frames, &addr, PROT_READ | PROT_WRITE, 0); if (state->fres != NULL) { trace_xen_map_resource_ioreq(state->ioservid, addr); - state->buffered_io_page = addr; - state->shared_page = addr + XC_PAGE_SIZE; + state->shared_page = addr; + if (state->has_bufioreq) { + state->buffered_io_page = addr; + state->shared_page = addr + XC_PAGE_SIZE; + } } else if (errno != EOPNOTSUPP) { error_report("failed to map ioreq server resources: error %d handle=%p", errno, xen_xc); return -1; } - rc = xen_get_ioreq_server_info(xen_domid, state->ioservid, - (state->shared_page == NULL) ? - &ioreq_pfn : NULL, - (state->buffered_io_page == NULL) ? - &bufioreq_pfn : NULL, - &bufioreq_evtchn); - if (rc < 0) { - error_report("failed to get ioreq server info: error %d handle=%p", - errno, xen_xc); - return rc; - } + /* + * If we fail to map the shared page with xenforeignmemory_map_resource() + * or if we're using buffered ioreqs, we need xen_get_ioreq_server_info() + * to provide the addresses to map the shared page and/or to get the + * event-channel port for buffered ioreqs. + */ + if (state->shared_page == NULL || state->has_bufioreq) { + rc = xen_get_ioreq_server_info(xen_domid, state->ioservid, + (state->shared_page == NULL) ? + &ioreq_pfn : NULL, + (state->has_bufioreq && + state->buffered_io_page == NULL) ? + &bufioreq_pfn : NULL, + &bufioreq_evtchn); + if (rc < 0) { + error_report("failed to get ioreq server info: error %d handle=%p", + errno, xen_xc); + return rc; + } - if (state->shared_page == NULL) { - trace_xen_map_ioreq_server_shared_page(ioreq_pfn); + if (state->shared_page == NULL) { + trace_xen_map_ioreq_server_shared_page(ioreq_pfn); - state->shared_page = xenforeignmemory_map(xen_fmem, xen_domid, - PROT_READ | PROT_WRITE, - 1, &ioreq_pfn, NULL); + state->shared_page = xenforeignmemory_map(xen_fmem, xen_domid, + PROT_READ | PROT_WRITE, + 1, &ioreq_pfn, NULL); + } if (state->shared_page == NULL) { error_report("map shared IO page returned error %d handle=%p", errno, xen_xc); } - } - if (state->buffered_io_page == NULL) { - trace_xen_map_ioreq_server_buffered_io_page(bufioreq_pfn); + if (state->has_bufioreq && state->buffered_io_page == NULL) { + trace_xen_map_ioreq_server_buffered_io_page(bufioreq_pfn); - state->buffered_io_page = xenforeignmemory_map(xen_fmem, xen_domid, - PROT_READ | PROT_WRITE, - 1, &bufioreq_pfn, - NULL); - if (state->buffered_io_page == NULL) { - error_report("map buffered IO page returned error %d", errno); - return -1; + state->buffered_io_page = xenforeignmemory_map(xen_fmem, xen_domid, + PROT_READ | PROT_WRITE, + 1, &bufioreq_pfn, + NULL); + if (state->buffered_io_page == NULL) { + error_report("map buffered IO page returned error %d", errno); + return -1; + } } } - if (state->shared_page == NULL || state->buffered_io_page == NULL) { + if (state->shared_page == NULL || + (state->has_bufioreq && state->buffered_io_page == NULL)) { return -1; } - trace_xen_map_ioreq_server_buffered_io_evtchn(bufioreq_evtchn); - - state->bufioreq_remote_port = bufioreq_evtchn; + if (state->has_bufioreq) { + trace_xen_map_ioreq_server_buffered_io_evtchn(bufioreq_evtchn); + state->bufioreq_remote_port = bufioreq_evtchn; + } return 0; } @@ -830,14 +859,15 @@ static void xen_do_ioreq_register(XenIOState *state, state->ioreq_local_port[i] = rc; } - rc = qemu_xen_evtchn_bind_interdomain(state->xce_handle, xen_domid, - state->bufioreq_remote_port); - if (rc == -1) { - error_report("buffered evtchn bind error %d", errno); - goto err; + if (state->has_bufioreq) { + rc = qemu_xen_evtchn_bind_interdomain(state->xce_handle, xen_domid, + state->bufioreq_remote_port); + if (rc == -1) { + error_report("buffered evtchn bind error %d", errno); + goto err; + } + state->bufioreq_local_port = rc; } - state->bufioreq_local_port = rc; - /* Init RAM management */ #ifdef XEN_COMPAT_PHYSMAP xen_map_cache_init(xen_phys_offset_to_gaddr, state); @@ -865,6 +895,7 @@ err: } void xen_register_ioreq(XenIOState *state, unsigned int max_cpus, + uint8_t handle_bufioreq, const MemoryListener *xen_memory_listener) { int rc; @@ -883,7 +914,8 @@ void xen_register_ioreq(XenIOState *state, unsigned int max_cpus, goto err; } - rc = xen_create_ioreq_server(xen_domid, &state->ioservid); + state->has_bufioreq = handle_bufioreq != HVM_IOREQSRV_BUFIOREQ_OFF; + rc = xen_create_ioreq_server(xen_domid, handle_bufioreq, &state->ioservid); if (!rc) { xen_do_ioreq_register(state, max_cpus, xen_memory_listener); } else { diff --git a/hw/xen/xen-legacy-backend.c b/hw/xen/xen-legacy-backend.c index 5514184..5ed53f8 100644 --- a/hw/xen/xen-legacy-backend.c +++ b/hw/xen/xen-legacy-backend.c @@ -147,24 +147,6 @@ void xen_be_unmap_grant_refs(struct XenLegacyDevice *xendev, void *ptr, } } -int xen_be_copy_grant_refs(struct XenLegacyDevice *xendev, - bool to_domain, - XenGrantCopySegment segs[], - unsigned int nr_segs) -{ - int rc; - - assert(xendev->ops->flags & DEVOPS_FLAG_NEED_GNTDEV); - - rc = qemu_xen_gnttab_grant_copy(xendev->gnttabdev, to_domain, xen_domid, - segs, nr_segs, NULL); - if (rc) { - xen_pv_printf(xendev, 0, "xengnttab_grant_copy failed: %s\n", - strerror(-rc)); - } - return rc; -} - /* * get xen backend device, allocate a new one if it doesn't exist. */ @@ -181,7 +163,7 @@ static struct XenLegacyDevice *xen_be_get_xendev(const char *type, int dom, /* init new xendev */ xendev = g_malloc0(ops->size); - object_initialize(&xendev->qdev, ops->size, TYPE_XENBACKEND); + object_initialize(xendev, ops->size, TYPE_XENBACKEND); OBJECT(xendev)->free = g_free; qdev_set_id(DEVICE(xendev), g_strdup_printf("xen-%s-%d", type, dev), &error_fatal); @@ -653,29 +635,22 @@ int xen_be_bind_evtchn(struct XenLegacyDevice *xendev) } -static Property xendev_properties[] = { - DEFINE_PROP_END_OF_LIST(), -}; - -static void xendev_class_init(ObjectClass *klass, void *data) +static void xendev_class_init(ObjectClass *klass, const void *data) { DeviceClass *dc = DEVICE_CLASS(klass); - device_class_set_props(dc, xendev_properties); set_bit(DEVICE_CATEGORY_MISC, dc->categories); - /* xen-backend devices can be plugged/unplugged dynamically */ - dc->user_creatable = true; dc->bus_type = TYPE_XENSYSBUS; } static const TypeInfo xendev_type_info = { .name = TYPE_XENBACKEND, - .parent = TYPE_DEVICE, + .parent = TYPE_DYNAMIC_SYS_BUS_DEVICE, .class_init = xendev_class_init, - .instance_size = sizeof(struct XenLegacyDevice), + .instance_size = sizeof(XenLegacyDevice), }; -static void xen_sysbus_class_init(ObjectClass *klass, void *data) +static void xen_sysbus_class_init(ObjectClass *klass, const void *data) { HotplugHandlerClass *hc = HOTPLUG_HANDLER_CLASS(klass); @@ -686,28 +661,15 @@ static const TypeInfo xensysbus_info = { .name = TYPE_XENSYSBUS, .parent = TYPE_BUS, .class_init = xen_sysbus_class_init, - .interfaces = (InterfaceInfo[]) { + .interfaces = (const InterfaceInfo[]) { { TYPE_HOTPLUG_HANDLER }, { } } }; -static Property xen_sysdev_properties[] = { - {/* end of property list */}, -}; - -static void xen_sysdev_class_init(ObjectClass *klass, void *data) -{ - DeviceClass *dc = DEVICE_CLASS(klass); - - device_class_set_props(dc, xen_sysdev_properties); -} - static const TypeInfo xensysdev_info = { .name = TYPE_XENSYSDEV, .parent = TYPE_SYS_BUS_DEVICE, - .instance_size = sizeof(SysBusDevice), - .class_init = xen_sysdev_class_init, }; static void xenbe_register_types(void) diff --git a/hw/xen/xen-mapcache.c b/hw/xen/xen-mapcache.c index 18ba7b1..e31d379 100644 --- a/hw/xen/xen-mapcache.c +++ b/hw/xen/xen-mapcache.c @@ -18,8 +18,8 @@ #include "hw/xen/xen_native.h" #include "qemu/bitmap.h" -#include "sysemu/runstate.h" -#include "sysemu/xen-mapcache.h" +#include "system/runstate.h" +#include "system/xen-mapcache.h" #include "trace.h" #include <xenevtchn.h> @@ -75,7 +75,8 @@ typedef struct MapCache { } MapCache; static MapCache *mapcache; -static MapCache *mapcache_grants; +static MapCache *mapcache_grants_ro; +static MapCache *mapcache_grants_rw; static xengnttab_handle *xen_region_gnttabdev; static inline void mapcache_lock(MapCache *mc) @@ -176,9 +177,12 @@ void xen_map_cache_init(phys_offset_to_gaddr_t f, void *opaque) * Grant mappings must use XC_PAGE_SIZE granularity since we can't * map anything beyond the number of pages granted to us. */ - mapcache_grants = xen_map_cache_init_single(f, opaque, - XC_PAGE_SHIFT, - max_mcache_size); + mapcache_grants_ro = xen_map_cache_init_single(f, opaque, + XC_PAGE_SHIFT, + max_mcache_size); + mapcache_grants_rw = xen_map_cache_init_single(f, opaque, + XC_PAGE_SHIFT, + max_mcache_size); setrlimit(RLIMIT_AS, &rlimit_as); } @@ -376,12 +380,12 @@ tryagain: entry = &mc->entry[address_index % mc->nr_buckets]; - while (entry && (lock || entry->lock) && entry->vaddr_base && - (entry->paddr_index != address_index || entry->size != cache_size || + while (entry && (!entry->vaddr_base || + entry->paddr_index != address_index || entry->size != cache_size || !test_bits(address_offset >> XC_PAGE_SHIFT, test_bit_size >> XC_PAGE_SHIFT, entry->valid_mapping))) { - if (!free_entry && !entry->lock) { + if (!free_entry && (!entry->lock || !entry->vaddr_base)) { free_entry = entry; free_pentry = pentry; } @@ -456,9 +460,13 @@ uint8_t *xen_map_cache(MemoryRegion *mr, bool is_write) { bool grant = xen_mr_is_grants(mr); - MapCache *mc = grant ? mapcache_grants : mapcache; + MapCache *mc = mapcache; uint8_t *p; + if (grant) { + mc = is_write ? mapcache_grants_rw : mapcache_grants_ro; + } + if (grant && !lock) { /* * Grants are only supported via address_space_map(). Anything @@ -523,7 +531,10 @@ ram_addr_t xen_ram_addr_from_mapcache(void *ptr) addr = xen_ram_addr_from_mapcache_single(mapcache, ptr); if (addr == RAM_ADDR_INVALID) { - addr = xen_ram_addr_from_mapcache_single(mapcache_grants, ptr); + addr = xen_ram_addr_from_mapcache_single(mapcache_grants_ro, ptr); + } + if (addr == RAM_ADDR_INVALID) { + addr = xen_ram_addr_from_mapcache_single(mapcache_grants_rw, ptr); } return addr; @@ -626,7 +637,8 @@ static void xen_invalidate_map_cache_entry_single(MapCache *mc, uint8_t *buffer) static void xen_invalidate_map_cache_entry_all(uint8_t *buffer) { xen_invalidate_map_cache_entry_single(mapcache, buffer); - xen_invalidate_map_cache_entry_single(mapcache_grants, buffer); + xen_invalidate_map_cache_entry_single(mapcache_grants_ro, buffer); + xen_invalidate_map_cache_entry_single(mapcache_grants_rw, buffer); } static void xen_invalidate_map_cache_entry_bh(void *opaque) @@ -700,7 +712,6 @@ void xen_invalidate_map_cache(void) bdrv_drain_all(); xen_invalidate_map_cache_single(mapcache); - xen_invalidate_map_cache_single(mapcache_grants); } static uint8_t *xen_replace_cache_entry_unlocked(MapCache *mc, diff --git a/hw/xen/xen-pvh-common.c b/hw/xen/xen-pvh-common.c new file mode 100644 index 0000000..b93ff80 --- /dev/null +++ b/hw/xen/xen-pvh-common.c @@ -0,0 +1,399 @@ +/* + * QEMU Xen PVH machine - common code. + * + * Copyright (c) 2024 Advanced Micro Devices, Inc. + * + * SPDX-License-Identifier: GPL-2.0-or-later + */ + +#include "qemu/osdep.h" +#include "qemu/error-report.h" +#include "qemu/units.h" +#include "qapi/visitor.h" +#include "hw/boards.h" +#include "hw/irq.h" +#include "system/tpm.h" +#include "system/tpm_backend.h" +#include "system/runstate.h" +#include "hw/xen/xen-pvh-common.h" +#include "trace.h" + +static const MemoryListener xen_memory_listener = { + .region_add = xen_region_add, + .region_del = xen_region_del, + .log_start = NULL, + .log_stop = NULL, + .log_sync = NULL, + .log_global_start = NULL, + .log_global_stop = NULL, + .priority = MEMORY_LISTENER_PRIORITY_ACCEL, +}; + +static void xen_pvh_init_ram(XenPVHMachineState *s, + MemoryRegion *sysmem) +{ + MachineState *ms = MACHINE(s); + ram_addr_t block_len, ram_size[2]; + + if (ms->ram_size <= s->cfg.ram_low.size) { + ram_size[0] = ms->ram_size; + ram_size[1] = 0; + block_len = s->cfg.ram_low.base + ram_size[0]; + } else { + ram_size[0] = s->cfg.ram_low.size; + ram_size[1] = ms->ram_size - s->cfg.ram_low.size; + block_len = s->cfg.ram_high.base + ram_size[1]; + } + + memory_region_init_ram(&xen_memory, NULL, "xen.ram", block_len, + &error_fatal); + + memory_region_init_alias(&s->ram.low, NULL, "xen.ram.lo", &xen_memory, + s->cfg.ram_low.base, ram_size[0]); + memory_region_add_subregion(sysmem, s->cfg.ram_low.base, &s->ram.low); + if (ram_size[1] > 0) { + memory_region_init_alias(&s->ram.high, NULL, "xen.ram.hi", &xen_memory, + s->cfg.ram_high.base, ram_size[1]); + memory_region_add_subregion(sysmem, s->cfg.ram_high.base, &s->ram.high); + } + + /* Setup support for grants. */ + memory_region_init_ram(&xen_grants, NULL, "xen.grants", block_len, + &error_fatal); + memory_region_add_subregion(sysmem, XEN_GRANT_ADDR_OFF, &xen_grants); +} + +static void xen_set_irq(void *opaque, int irq, int level) +{ + if (xendevicemodel_set_irq_level(xen_dmod, xen_domid, irq, level)) { + error_report("xendevicemodel_set_irq_level failed"); + } +} + +static void xen_create_virtio_mmio_devices(XenPVHMachineState *s) +{ + int i; + + /* + * We create the transports in reverse order. Since qbus_realize() + * prepends (not appends) new child buses, the decrementing loop below will + * create a list of virtio-mmio buses with increasing base addresses. + * + * When a -device option is processed from the command line, + * qbus_find_recursive() picks the next free virtio-mmio bus in forwards + * order. + * + * This is what the Xen tools expect. + */ + for (i = s->cfg.virtio_mmio_num - 1; i >= 0; i--) { + hwaddr base = s->cfg.virtio_mmio.base + i * s->cfg.virtio_mmio.size; + qemu_irq irq = qemu_allocate_irq(xen_set_irq, NULL, + s->cfg.virtio_mmio_irq_base + i); + + sysbus_create_simple("virtio-mmio", base, irq); + + trace_xen_create_virtio_mmio_devices(i, + s->cfg.virtio_mmio_irq_base + i, + base); + } +} + +#ifdef CONFIG_TPM +static void xen_enable_tpm(XenPVHMachineState *s) +{ + Error *errp = NULL; + DeviceState *dev; + SysBusDevice *busdev; + + TPMBackend *be = qemu_find_tpm_be("tpm0"); + if (be == NULL) { + error_report("Couldn't find tmp0 backend"); + return; + } + dev = qdev_new(TYPE_TPM_TIS_SYSBUS); + object_property_set_link(OBJECT(dev), "tpmdev", OBJECT(be), &errp); + object_property_set_str(OBJECT(dev), "tpmdev", be->id, &errp); + busdev = SYS_BUS_DEVICE(dev); + sysbus_realize_and_unref(busdev, &error_fatal); + sysbus_mmio_map(busdev, 0, s->cfg.tpm.base); + + trace_xen_enable_tpm(s->cfg.tpm.base); +} +#endif + +/* + * We use the GPEX PCIe controller with its internal INTX PCI interrupt + * swizzling. This swizzling is emulated in QEMU and routes all INTX + * interrupts from endpoints down to only 4 INTX interrupts. + * See include/hw/pci/pci.h : pci_swizzle() + */ +static inline void xenpvh_gpex_init(XenPVHMachineState *s, + XenPVHMachineClass *xpc, + MemoryRegion *sysmem) +{ + MemoryRegion *ecam_reg; + MemoryRegion *mmio_reg; + DeviceState *dev; + int i; + + object_initialize_child(OBJECT(s), "gpex", &s->pci.gpex, + TYPE_GPEX_HOST); + dev = DEVICE(&s->pci.gpex); + sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), &error_fatal); + + ecam_reg = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0); + memory_region_add_subregion(sysmem, s->cfg.pci_ecam.base, ecam_reg); + + mmio_reg = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 1); + + if (s->cfg.pci_mmio.size) { + memory_region_init_alias(&s->pci.mmio_alias, OBJECT(dev), "pcie-mmio", + mmio_reg, + s->cfg.pci_mmio.base, s->cfg.pci_mmio.size); + memory_region_add_subregion(sysmem, s->cfg.pci_mmio.base, + &s->pci.mmio_alias); + } + + if (s->cfg.pci_mmio_high.size) { + memory_region_init_alias(&s->pci.mmio_high_alias, OBJECT(dev), + "pcie-mmio-high", + mmio_reg, s->cfg.pci_mmio_high.base, s->cfg.pci_mmio_high.size); + memory_region_add_subregion(sysmem, s->cfg.pci_mmio_high.base, + &s->pci.mmio_high_alias); + } + + /* + * PVH implementations with PCI enabled must provide set_pci_intx_irq() + * and optionally an implementation of set_pci_link_route(). + */ + assert(xpc->set_pci_intx_irq); + + for (i = 0; i < PCI_NUM_PINS; i++) { + qemu_irq irq = qemu_allocate_irq(xpc->set_pci_intx_irq, s, i); + + sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, irq); + gpex_set_irq_num(GPEX_HOST(dev), i, s->cfg.pci_intx_irq_base + i); + if (xpc->set_pci_link_route) { + xpc->set_pci_link_route(i, s->cfg.pci_intx_irq_base + i); + } + } +} + +static void xen_pvh_init(MachineState *ms) +{ + XenPVHMachineState *s = XEN_PVH_MACHINE(ms); + XenPVHMachineClass *xpc = XEN_PVH_MACHINE_GET_CLASS(s); + MemoryRegion *sysmem = get_system_memory(); + + if (ms->ram_size == 0) { + warn_report("%s: ram size not specified. QEMU machine started" + " without IOREQ (no emulated devices including virtio)", + MACHINE_CLASS(object_get_class(OBJECT(ms)))->desc); + return; + } + + xen_pvh_init_ram(s, sysmem); + xen_register_ioreq(&s->ioreq, ms->smp.max_cpus, + xpc->handle_bufioreq, + &xen_memory_listener); + + if (s->cfg.virtio_mmio_num) { + xen_create_virtio_mmio_devices(s); + } + +#ifdef CONFIG_TPM + if (xpc->has_tpm) { + if (s->cfg.tpm.base) { + xen_enable_tpm(s); + } else { + warn_report("tpm-base-addr is not set. TPM will not be enabled"); + } + } +#endif + + /* Non-zero pci-ecam-size enables PCI. */ + if (s->cfg.pci_ecam.size) { + if (s->cfg.pci_ecam.size != 256 * MiB) { + error_report("pci-ecam-size only supports values 0 or 0x10000000"); + exit(EXIT_FAILURE); + } + if (!s->cfg.pci_intx_irq_base) { + error_report("PCI enabled but pci-intx-irq-base not set"); + exit(EXIT_FAILURE); + } + + xenpvh_gpex_init(s, xpc, sysmem); + } + + /* Call the implementation specific init. */ + if (xpc->init) { + xpc->init(ms); + } +} + +#define XEN_PVH_PROP_MEMMAP_SETTER(n, f) \ +static void xen_pvh_set_ ## n ## _ ## f(Object *obj, Visitor *v, \ + const char *name, void *opaque, \ + Error **errp) \ +{ \ + XenPVHMachineState *xp = XEN_PVH_MACHINE(obj); \ + uint64_t value; \ + \ + if (!visit_type_size(v, name, &value, errp)) { \ + return; \ + } \ + xp->cfg.n.f = value; \ +} + +#define XEN_PVH_PROP_MEMMAP_GETTER(n, f) \ +static void xen_pvh_get_ ## n ## _ ## f(Object *obj, Visitor *v, \ + const char *name, void *opaque, \ + Error **errp) \ +{ \ + XenPVHMachineState *xp = XEN_PVH_MACHINE(obj); \ + uint64_t value = xp->cfg.n.f; \ + \ + visit_type_uint64(v, name, &value, errp); \ +} + +#define XEN_PVH_PROP_MEMMAP_BASE(n) \ + XEN_PVH_PROP_MEMMAP_SETTER(n, base) \ + XEN_PVH_PROP_MEMMAP_GETTER(n, base) \ + +#define XEN_PVH_PROP_MEMMAP_SIZE(n) \ + XEN_PVH_PROP_MEMMAP_SETTER(n, size) \ + XEN_PVH_PROP_MEMMAP_GETTER(n, size) + +#define XEN_PVH_PROP_MEMMAP(n) \ + XEN_PVH_PROP_MEMMAP_BASE(n) \ + XEN_PVH_PROP_MEMMAP_SIZE(n) + +XEN_PVH_PROP_MEMMAP(ram_low) +XEN_PVH_PROP_MEMMAP(ram_high) +/* TPM only has a base-addr option. */ +XEN_PVH_PROP_MEMMAP_BASE(tpm) +XEN_PVH_PROP_MEMMAP(virtio_mmio) +XEN_PVH_PROP_MEMMAP(pci_ecam) +XEN_PVH_PROP_MEMMAP(pci_mmio) +XEN_PVH_PROP_MEMMAP(pci_mmio_high) + +static void xen_pvh_set_pci_intx_irq_base(Object *obj, Visitor *v, + const char *name, void *opaque, + Error **errp) +{ + XenPVHMachineState *xp = XEN_PVH_MACHINE(obj); + uint32_t value; + + if (!visit_type_uint32(v, name, &value, errp)) { + return; + } + + xp->cfg.pci_intx_irq_base = value; +} + +static void xen_pvh_get_pci_intx_irq_base(Object *obj, Visitor *v, + const char *name, void *opaque, + Error **errp) +{ + XenPVHMachineState *xp = XEN_PVH_MACHINE(obj); + uint32_t value = xp->cfg.pci_intx_irq_base; + + visit_type_uint32(v, name, &value, errp); +} + +void xen_pvh_class_setup_common_props(XenPVHMachineClass *xpc) +{ + ObjectClass *oc = OBJECT_CLASS(xpc); + MachineClass *mc = MACHINE_CLASS(xpc); + +#define OC_MEMMAP_PROP_BASE(c, prop_name, name) \ +do { \ + object_class_property_add(c, prop_name "-base", "uint64_t", \ + xen_pvh_get_ ## name ## _base, \ + xen_pvh_set_ ## name ## _base, NULL, NULL); \ + object_class_property_set_description(oc, prop_name "-base", \ + "Set base address for " prop_name); \ +} while (0) + +#define OC_MEMMAP_PROP_SIZE(c, prop_name, name) \ +do { \ + object_class_property_add(c, prop_name "-size", "uint64_t", \ + xen_pvh_get_ ## name ## _size, \ + xen_pvh_set_ ## name ## _size, NULL, NULL); \ + object_class_property_set_description(oc, prop_name "-size", \ + "Set memory range size for " prop_name); \ +} while (0) + +#define OC_MEMMAP_PROP(c, prop_name, name) \ +do { \ + OC_MEMMAP_PROP_BASE(c, prop_name, name); \ + OC_MEMMAP_PROP_SIZE(c, prop_name, name); \ +} while (0) + + /* + * We provide memmap properties to allow Xen to move things to other + * addresses for example when users need to accomodate the memory-map + * for 1:1 mapped devices/memory. + */ + OC_MEMMAP_PROP(oc, "ram-low", ram_low); + OC_MEMMAP_PROP(oc, "ram-high", ram_high); + + if (xpc->has_virtio_mmio) { + OC_MEMMAP_PROP(oc, "virtio-mmio", virtio_mmio); + } + + if (xpc->has_pci) { + OC_MEMMAP_PROP(oc, "pci-ecam", pci_ecam); + OC_MEMMAP_PROP(oc, "pci-mmio", pci_mmio); + OC_MEMMAP_PROP(oc, "pci-mmio-high", pci_mmio_high); + + object_class_property_add(oc, "pci-intx-irq-base", "uint32_t", + xen_pvh_get_pci_intx_irq_base, + xen_pvh_set_pci_intx_irq_base, + NULL, NULL); + object_class_property_set_description(oc, "pci-intx-irq-base", + "Set PCI INTX interrupt base line."); + } + +#ifdef CONFIG_TPM + if (xpc->has_tpm) { + object_class_property_add(oc, "tpm-base-addr", "uint64_t", + xen_pvh_get_tpm_base, + xen_pvh_set_tpm_base, + NULL, NULL); + object_class_property_set_description(oc, "tpm-base-addr", + "Set Base address for TPM device."); + + machine_class_allow_dynamic_sysbus_dev(mc, TYPE_TPM_TIS_SYSBUS); + } +#endif +} + +static void xen_pvh_class_init(ObjectClass *oc, const void *data) +{ + MachineClass *mc = MACHINE_CLASS(oc); + + mc->init = xen_pvh_init; + + mc->desc = "Xen PVH machine"; + mc->max_cpus = 1; + mc->default_machine_opts = "accel=xen"; + /* Set to zero to make sure that the real ram size is passed. */ + mc->default_ram_size = 0; +} + +static const TypeInfo xen_pvh_info = { + .name = TYPE_XEN_PVH_MACHINE, + .parent = TYPE_MACHINE, + .abstract = true, + .instance_size = sizeof(XenPVHMachineState), + .class_size = sizeof(XenPVHMachineClass), + .class_init = xen_pvh_class_init, +}; + +static void xen_pvh_register_types(void) +{ + type_register_static(&xen_pvh_info); +} + +type_init(xen_pvh_register_types); diff --git a/hw/xen/xen_devconfig.c b/hw/xen/xen_devconfig.c index 2150869..5698cc7 100644 --- a/hw/xen/xen_devconfig.c +++ b/hw/xen/xen_devconfig.c @@ -1,8 +1,8 @@ #include "qemu/osdep.h" #include "hw/xen/xen-legacy-backend.h" #include "qemu/option.h" -#include "sysemu/blockdev.h" -#include "sysemu/sysemu.h" +#include "system/blockdev.h" +#include "system/system.h" /* ------------------------------------------------------------- */ @@ -66,11 +66,3 @@ int xen_config_dev_vkbd(int vdev) xen_config_dev_dirs("vkbd", "vkbd", vdev, fe, be, sizeof(fe)); return xen_config_dev_all(fe, be); } - -int xen_config_dev_console(int vdev) -{ - char fe[256], be[256]; - - xen_config_dev_dirs("console", "console", vdev, fe, be, sizeof(fe)); - return xen_config_dev_all(fe, be); -} diff --git a/hw/xen/xen_pt.c b/hw/xen/xen_pt.c index 3635d1b..9d16644 100644 --- a/hw/xen/xen_pt.c +++ b/hw/xen/xen_pt.c @@ -766,6 +766,57 @@ static void xen_pt_destroy(PCIDevice *d) { } /* init */ +#if CONFIG_XEN_CTRL_INTERFACE_VERSION >= 42000 +static bool xen_pt_need_gsi(void) +{ + FILE *fp; + int len; + /* + * The max length of guest_type is "PVH"+'\n'+'\0', it is 5, + * so here set the length of type to be twice. + */ + char type[10]; + const char *guest_type = "/sys/hypervisor/guest_type"; + + fp = fopen(guest_type, "r"); + if (!fp) { + error_report("Cannot open %s: %s", guest_type, strerror(errno)); + return false; + } + + if (fgets(type, sizeof(type), fp)) { + len = strlen(type); + if (len) { + type[len - 1] = '\0'; + if (!strcmp(type, "PVH")) { + fclose(fp); + return true; + } + } + } + + fclose(fp); + return false; +} + +static int xen_pt_map_pirq_for_gsi(PCIDevice *d, int *pirq) +{ + int gsi; + XenPCIPassthroughState *s = XEN_PT_DEVICE(d); + + gsi = xc_pcidev_get_gsi(xen_xc, + PCI_SBDF(s->real_device.domain, + s->real_device.bus, + s->real_device.dev, + s->real_device.func)); + if (gsi >= 0) { + return xc_physdev_map_pirq_gsi(xen_xc, xen_domid, gsi, pirq); + } + + return gsi; +} +#endif + static void xen_pt_realize(PCIDevice *d, Error **errp) { ERRP_GUARD(); @@ -847,7 +898,16 @@ static void xen_pt_realize(PCIDevice *d, Error **errp) goto out; } +#if CONFIG_XEN_CTRL_INTERFACE_VERSION >= 42000 + if (xen_pt_need_gsi()) { + rc = xen_pt_map_pirq_for_gsi(d, &pirq); + } else { + rc = xc_physdev_map_pirq(xen_xc, xen_domid, machine_irq, &pirq); + } +#else rc = xc_physdev_map_pirq(xen_xc, xen_domid, machine_irq, &pirq); +#endif + if (rc < 0) { XEN_PT_ERR(d, "Mapping machine irq %u to pirq %i failed, (err: %d)\n", machine_irq, pirq, errno); @@ -931,10 +991,9 @@ static void xen_pt_unregister_device(PCIDevice *d) xen_pt_destroy(d); } -static Property xen_pci_passthrough_properties[] = { +static const Property xen_pci_passthrough_properties[] = { DEFINE_PROP_PCI_HOST_DEVADDR("hostaddr", XenPCIPassthroughState, hostaddr), DEFINE_PROP_BOOL("permissive", XenPCIPassthroughState, permissive, false), - DEFINE_PROP_END_OF_LIST(), }; static void xen_pci_passthrough_instance_init(Object *obj) @@ -988,7 +1047,7 @@ static void xen_igd_clear_slot(DeviceState *qdev, Error **errp) xpdc->pci_qdev_realize(qdev, errp); } -static void xen_pci_passthrough_class_init(ObjectClass *klass, void *data) +static void xen_pci_passthrough_class_init(ObjectClass *klass, const void *data) { DeviceClass *dc = DEVICE_CLASS(klass); PCIDeviceClass *k = PCI_DEVICE_CLASS(klass); @@ -1020,7 +1079,7 @@ static const TypeInfo xen_pci_passthrough_info = { .class_init = xen_pci_passthrough_class_init, .class_size = sizeof(XenPTDeviceClass), .instance_init = xen_pci_passthrough_instance_init, - .interfaces = (InterfaceInfo[]) { + .interfaces = (const InterfaceInfo[]) { { INTERFACE_CONVENTIONAL_PCI_DEVICE }, { INTERFACE_PCIE_DEVICE }, { }, diff --git a/hw/xen/xen_pt_graphics.c b/hw/xen/xen_pt_graphics.c index 6c2e3f4..2c0cec9 100644 --- a/hw/xen/xen_pt_graphics.c +++ b/hw/xen/xen_pt_graphics.c @@ -347,7 +347,7 @@ static const IGDDeviceIDInfo igd_combo_id_infos[] = { {0x162D, 0x9cc3, 0x03}, /* BDWGT3SRVR, BDWM_w7 */ }; -static void isa_bridge_class_init(ObjectClass *klass, void *data) +static void isa_bridge_class_init(ObjectClass *klass, const void *data) { DeviceClass *dc = DEVICE_CLASS(klass); PCIDeviceClass *k = PCI_DEVICE_CLASS(klass); @@ -363,7 +363,7 @@ static const TypeInfo isa_bridge_info = { .parent = TYPE_PCI_DEVICE, .instance_size = sizeof(PCIDevice), .class_init = isa_bridge_class_init, - .interfaces = (InterfaceInfo[]) { + .interfaces = (const InterfaceInfo[]) { { INTERFACE_CONVENTIONAL_PCI_DEVICE }, { }, }, diff --git a/hw/xen/xen_pvdev.c b/hw/xen/xen_pvdev.c index c5ad71e..fe95b62 100644 --- a/hw/xen/xen_pvdev.c +++ b/hw/xen/xen_pvdev.c @@ -22,6 +22,7 @@ #include "qemu/main-loop.h" #include "hw/qdev-core.h" #include "hw/xen/xen-legacy-backend.h" +#include "hw/xen/xen-bus-helper.h" #include "hw/xen/xen_pvdev.h" /* private */ @@ -81,12 +82,9 @@ int xenstore_write_str(const char *base, const char *node, const char *val) char *xenstore_read_str(const char *base, const char *node) { - char abspath[XEN_BUFSIZE]; - unsigned int len; char *str, *ret = NULL; - snprintf(abspath, sizeof(abspath), "%s/%s", base, node); - str = qemu_xen_xs_read(xenstore, 0, abspath, &len); + str = xs_node_read(xenstore, 0, NULL, NULL, "%s/%s", base, node); if (str != NULL) { /* move to qemu-allocated memory to make sure * callers can safely g_free() stuff. */ @@ -275,7 +273,7 @@ void xen_pv_del_xendev(struct XenLegacyDevice *xendev) QTAILQ_REMOVE(&xendevs, xendev, next); - qdev_unplug(&xendev->qdev, NULL); + qdev_unplug(DEVICE(xendev), NULL); } void xen_pv_insert_xendev(struct XenLegacyDevice *xendev) diff --git a/hw/xen/xen_stubs.c b/hw/xen/xen_stubs.c new file mode 100644 index 0000000..5e565df --- /dev/null +++ b/hw/xen/xen_stubs.c @@ -0,0 +1,51 @@ +/* + * Various stubs for xen functions + * + * Those functions are used only if xen_enabled(). This file is linked only if + * CONFIG_XEN is not set, so they should never be called. + * + * Copyright (c) 2025 Linaro, Ltd. + * + * SPDX-License-Identifier: GPL-2.0-or-later + */ + +#include "qemu/osdep.h" +#include "system/xen.h" +#include "system/xen-mapcache.h" + +void xen_hvm_modified_memory(ram_addr_t start, ram_addr_t length) +{ + g_assert_not_reached(); +} + +void xen_ram_alloc(ram_addr_t ram_addr, ram_addr_t size, + struct MemoryRegion *mr, Error **errp) +{ + g_assert_not_reached(); +} + +bool xen_mr_is_memory(MemoryRegion *mr) +{ + g_assert_not_reached(); +} + +void xen_invalidate_map_cache_entry(uint8_t *buffer) +{ + g_assert_not_reached(); +} + +ram_addr_t xen_ram_addr_from_mapcache(void *ptr) +{ + g_assert_not_reached(); +} + +uint8_t *xen_map_cache(MemoryRegion *mr, + hwaddr phys_addr, + hwaddr size, + ram_addr_t ram_addr_offset, + uint8_t lock, + bool dma, + bool is_write) +{ + g_assert_not_reached(); +} |