Merge remote-tracking branch 'sstabellini/for_anthony' into staging

* sstabellini/for_anthony:
  xen: introduce an event channel for buffered io event notifications
  xen-mapcache: don't unmap locked entry during mapcache invalidation
  Xen, mapcache: Fix the compute of the size of bucket.
  xen: handle backend deletion from xenstore
  Xen: Add xen-apic support and hook it up.
  Xen: basic HVM MSI injection support.

9 files changed, 171 insertions, 20 deletions

diff --git a/Makefile.target b/Makefile.target
index 61dfe3b79f..84951a09ec 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -239,7 +239,7 @@ QEMU_CFLAGS += $(VNC_PNG_CFLAGS)
 obj-$(CONFIG_XEN) += xen-all.o xen_machine_pv.o xen_domainbuild.o xen-mapcache.o
 obj-$(CONFIG_NO_XEN) += xen-stub.o
 
-obj-i386-$(CONFIG_XEN) += xen_platform.o
+obj-i386-$(CONFIG_XEN) += xen_platform.o xen_apic.o
 
 # Inter-VM PCI shared memory
 CONFIG_IVSHMEM =
diff --git a/hw/pc.c b/hw/pc.c
index 67f0479f40..1f5aacb2d2 100644
--- a/hw/pc.c
+++ b/hw/pc.c
@@ -42,6 +42,7 @@
 #include "sysbus.h"
 #include "sysemu.h"
 #include "kvm.h"
+#include "xen.h"
 #include "blockdev.h"
 #include "ui/qemu-spice.h"
 #include "memory.h"
@@ -891,9 +892,12 @@ static DeviceState *apic_init(void *env, uint8_t apic_id)
 
     if (kvm_irqchip_in_kernel()) {
         dev = qdev_create(NULL, "kvm-apic");
+    } else if (xen_enabled()) {
+        dev = qdev_create(NULL, "xen-apic");
     } else {
         dev = qdev_create(NULL, "apic");
     }
+
     qdev_prop_set_uint8(dev, "id", apic_id);
     qdev_prop_set_ptr(dev, "cpu_env", env);
     qdev_init_nofail(dev);
@@ -912,6 +916,10 @@ static DeviceState *apic_init(void *env, uint8_t apic_id)
         msi_supported = true;
     }
 
+    if (xen_enabled()) {
+        msi_supported = true;
+    }
+
     return dev;
 }
 
diff --git a/hw/xen.h b/hw/xen.h
index b46879c6f7..e5926b7b8a 100644
--- a/hw/xen.h
+++ b/hw/xen.h
@@ -34,6 +34,7 @@ static inline int xen_enabled(void)
 int xen_pci_slot_get_pirq(PCIDevice *pci_dev, int irq_num);
 void xen_piix3_set_irq(void *opaque, int irq_num, int level);
 void xen_piix_pci_write_config_client(uint32_t address, uint32_t val, int len);
+void xen_hvm_inject_msi(uint64_t addr, uint32_t data);
 void xen_cmos_set_s3_resume(void *opaque, int irq, int level);
 
 qemu_irq *xen_interrupt_controller_init(void);
diff --git a/hw/xen_apic.c b/hw/xen_apic.c
new file mode 100644
index 0000000000..1725ff67dd
--- /dev/null
+++ b/hw/xen_apic.c
@@ -0,0 +1,90 @@
+/*
+ * Xen basic APIC support
+ *
+ * Copyright (c) 2012 Citrix
+ *
+ * Authors:
+ *  Wei Liu <wei.liu2@citrix.com>
+ *
+ * This work is licensed under the terms of the GNU GPL version 2 or
+ * later. See the COPYING file in the top-level directory.
+ */
+#include "hw/apic_internal.h"
+#include "hw/msi.h"
+#include "xen.h"
+
+static uint64_t xen_apic_mem_read(void *opaque, target_phys_addr_t addr,
+                                  unsigned size)
+{
+    return ~(uint64_t)0;
+}
+
+static void xen_apic_mem_write(void *opaque, target_phys_addr_t addr,
+                               uint64_t data, unsigned size)
+{
+    if (size != sizeof(uint32_t)) {
+        fprintf(stderr, "Xen: APIC write data size = %d, invalid\n", size);
+        return;
+    }
+
+    xen_hvm_inject_msi(addr, data);
+}
+
+static const MemoryRegionOps xen_apic_io_ops = {
+    .read = xen_apic_mem_read,
+    .write = xen_apic_mem_write,
+    .endianness = DEVICE_NATIVE_ENDIAN,
+};
+
+static void xen_apic_init(APICCommonState *s)
+{
+    memory_region_init_io(&s->io_memory, &xen_apic_io_ops, s, "xen-apic-msi",
+                          MSI_SPACE_SIZE);
+}
+
+static void xen_apic_set_base(APICCommonState *s, uint64_t val)
+{
+}
+
+static void xen_apic_set_tpr(APICCommonState *s, uint8_t val)
+{
+}
+
+static uint8_t xen_apic_get_tpr(APICCommonState *s)
+{
+    return 0;
+}
+
+static void xen_apic_vapic_base_update(APICCommonState *s)
+{
+}
+
+static void xen_apic_external_nmi(APICCommonState *s)
+{
+}
+
+static void xen_apic_class_init(ObjectClass *klass, void *data)
+{
+    APICCommonClass *k = APIC_COMMON_CLASS(klass);
+
+    k->init = xen_apic_init;
+    k->set_base = xen_apic_set_base;
+    k->set_tpr = xen_apic_set_tpr;
+    k->get_tpr = xen_apic_get_tpr;
+    k->vapic_base_update = xen_apic_vapic_base_update;
+    k->external_nmi = xen_apic_external_nmi;
+}
+
+static TypeInfo xen_apic_info = {
+    .name = "xen-apic",
+    .parent = TYPE_APIC_COMMON,
+    .instance_size = sizeof(APICCommonState),
+    .class_init = xen_apic_class_init,
+};
+
+static void xen_apic_register_types(void)
+{
+    type_register_static(&xen_apic_info);
+}
+
+type_init(xen_apic_register_types)
diff --git a/hw/xen_backend.c b/hw/xen_backend.c
index 2673ace185..66cb144397 100644
--- a/hw/xen_backend.c
+++ b/hw/xen_backend.c
@@ -592,7 +592,7 @@ static void xenstore_update_be(char *watch, char *type, int dom,
                                struct XenDevOps *ops)
 {
     struct XenDevice *xendev;
-    char path[XEN_BUFSIZE], *dom0;
+    char path[XEN_BUFSIZE], *dom0, *bepath;
     unsigned int len, dev;
 
     dom0 = xs_get_domain_path(xenstore, 0);
@@ -611,15 +611,16 @@ static void xenstore_update_be(char *watch, char *type, int dom,
         return;
     }
 
-    if (0) {
-        /* FIXME: detect devices being deleted from xenstore ... */
-        xen_be_del_xendev(dom, dev);
-    }
-
     xendev = xen_be_get_xendev(type, dom, dev, ops);
     if (xendev != NULL) {
-        xen_be_backend_changed(xendev, path);
-        xen_be_check_state(xendev);
+        bepath = xs_read(xenstore, 0, xendev->be, &len);
+        if (bepath == NULL) {
+            xen_be_del_xendev(dom, dev);
+        } else {
+            free(bepath);
+            xen_be_backend_changed(xendev, path);
+            xen_be_check_state(xendev);
+        }
     }
 }
 
diff --git a/hw/xen_disk.c b/hw/xen_disk.c
index 9719395b09..22dbd10303 100644
--- a/hw/xen_disk.c
+++ b/hw/xen_disk.c
@@ -745,6 +745,10 @@ static int blk_free(struct XenDevice *xendev)
     struct XenBlkDev *blkdev = container_of(xendev, struct XenBlkDev, xendev);
     struct ioreq *ioreq;
 
+    if (blkdev->bs || blkdev->sring) {
+        blk_disconnect(xendev);
+    }
+
     while (!QLIST_EMPTY(&blkdev->freelist)) {
         ioreq = QLIST_FIRST(&blkdev->freelist);
         QLIST_REMOVE(ioreq, list);
diff --git a/xen-all.c b/xen-all.c
index 3e6de417ec..a08eec0931 100644
--- a/xen-all.c
+++ b/xen-all.c
@@ -59,6 +59,9 @@ static inline ioreq_t *xen_vcpu_ioreq(shared_iopage_t *shared_page, int vcpu)
 }
 #  define FMT_ioreq_size "u"
 #endif
+#ifndef HVM_PARAM_BUFIOREQ_EVTCHN
+#define HVM_PARAM_BUFIOREQ_EVTCHN 26
+#endif
 
 #define BUFFER_IO_MAX_DELAY  100
 
@@ -77,6 +80,8 @@ typedef struct XenIOState {
     QEMUTimer *buffered_io_timer;
     /* the evtchn port for polling the notification, */
     evtchn_port_t *ioreq_local_port;
+    /* evtchn local port for buffered io */
+    evtchn_port_t bufioreq_local_port;
     /* the evtchn fd for polling */
     XenEvtchn xce_handle;
     /* which vcpu we are serving */
@@ -122,6 +127,11 @@ void xen_piix_pci_write_config_client(uint32_t address, uint32_t val, int len)
     }
 }
 
+void xen_hvm_inject_msi(uint64_t addr, uint32_t data)
+{
+    xc_hvm_inject_msi(xen_xc, xen_domid, addr, data);
+}
+
 static void xen_suspend_notifier(Notifier *notifier, void *data)
 {
     xc_set_hvm_param(xen_xc, xen_domid, HVM_PARAM_ACPI_S_STATE, 3);
@@ -624,6 +634,12 @@ static ioreq_t *cpu_get_ioreq(XenIOState *state)
     evtchn_port_t port;
 
     port = xc_evtchn_pending(state->xce_handle);
+    if (port == state->bufioreq_local_port) {
+        qemu_mod_timer(state->buffered_io_timer,
+                BUFFER_IO_MAX_DELAY + qemu_get_clock_ms(rt_clock));
+        return NULL;
+    }
+
     if (port != -1) {
         for (i = 0; i < smp_cpus; i++) {
             if (state->ioreq_local_port[i] == port) {
@@ -772,16 +788,18 @@ static void handle_ioreq(ioreq_t *req)
     }
 }
 
-static void handle_buffered_iopage(XenIOState *state)
+static int handle_buffered_iopage(XenIOState *state)
 {
     buf_ioreq_t *buf_req = NULL;
     ioreq_t req;
     int qw;
 
     if (!state->buffered_io_page) {
-        return;
+        return 0;
     }
 
+    memset(&req, 0x00, sizeof(req));
+
     while (state->buffered_io_page->read_pointer != state->buffered_io_page->write_pointer) {
         buf_req = &state->buffered_io_page->buf_ioreq[
             state->buffered_io_page->read_pointer % IOREQ_BUFFER_SLOT_NUM];
@@ -806,15 +824,21 @@ static void handle_buffered_iopage(XenIOState *state)
         xen_mb();
         state->buffered_io_page->read_pointer += qw ? 2 : 1;
     }
+
+    return req.count;
 }
 
 static void handle_buffered_io(void *opaque)
 {
     XenIOState *state = opaque;
 
-    handle_buffered_iopage(state);
-    qemu_mod_timer(state->buffered_io_timer,
-                   BUFFER_IO_MAX_DELAY + qemu_get_clock_ms(rt_clock));
+    if (handle_buffered_iopage(state)) {
+        qemu_mod_timer(state->buffered_io_timer,
+                BUFFER_IO_MAX_DELAY + qemu_get_clock_ms(rt_clock));
+    } else {
+        qemu_del_timer(state->buffered_io_timer);
+        xc_evtchn_unmask(state->xce_handle, state->bufioreq_local_port);
+    }
 }
 
 static void cpu_handle_ioreq(void *opaque)
@@ -944,7 +968,6 @@ static void xen_main_loop_prepare(XenIOState *state)
 
     state->buffered_io_timer = qemu_new_timer_ms(rt_clock, handle_buffered_io,
                                                  state);
-    qemu_mod_timer(state->buffered_io_timer, qemu_get_clock_ms(rt_clock));
 
     if (evtchn_fd != -1) {
         qemu_set_fd_handler(evtchn_fd, cpu_handle_ioreq, NULL, state);
@@ -1045,6 +1068,7 @@ int xen_hvm_init(void)
 {
     int i, rc;
     unsigned long ioreq_pfn;
+    unsigned long bufioreq_evtchn;
     XenIOState *state;
 
     state = g_malloc0(sizeof (XenIOState));
@@ -1097,6 +1121,20 @@ int xen_hvm_init(void)
         state->ioreq_local_port[i] = rc;
     }
 
+    rc = xc_get_hvm_param(xen_xc, xen_domid, HVM_PARAM_BUFIOREQ_EVTCHN,
+            &bufioreq_evtchn);
+    if (rc < 0) {
+        fprintf(stderr, "failed to get HVM_PARAM_BUFIOREQ_EVTCHN\n");
+        return -1;
+    }
+    rc = xc_evtchn_bind_interdomain(state->xce_handle, xen_domid,
+            (uint32_t)bufioreq_evtchn);
+    if (rc == -1) {
+        fprintf(stderr, "bind interdomain ioctl error %d\n", errno);
+        return -1;
+    }
+    state->bufioreq_local_port = rc;
+
     /* Init RAM management */
     xen_map_cache_init(xen_phys_offset_to_gaddr, state);
     xen_ram_init(ram_size);
diff --git a/xen-mapcache.c b/xen-mapcache.c
index a456479363..59ba085b62 100644
--- a/xen-mapcache.c
+++ b/xen-mapcache.c
@@ -216,12 +216,14 @@ tryagain:
     }
 
     /* size is always a multiple of MCACHE_BUCKET_SIZE */
-    if ((address_offset + (__size % MCACHE_BUCKET_SIZE)) > MCACHE_BUCKET_SIZE)
-        __size += MCACHE_BUCKET_SIZE;
-    if (__size % MCACHE_BUCKET_SIZE)
-        __size += MCACHE_BUCKET_SIZE - (__size % MCACHE_BUCKET_SIZE);
-    if (!__size)
+    if (size) {
+        __size = size + address_offset;
+        if (__size % MCACHE_BUCKET_SIZE) {
+            __size += MCACHE_BUCKET_SIZE - (__size % MCACHE_BUCKET_SIZE);
+        }
+    } else {
         __size = MCACHE_BUCKET_SIZE;
+    }
 
     entry = &mapcache->entry[address_index % mapcache->nr_buckets];
 
@@ -385,6 +387,9 @@ void xen_invalidate_map_cache(void)
         if (entry->vaddr_base == NULL) {
             continue;
         }
+        if (entry->lock > 0) {
+            continue;
+        }
 
         if (munmap(entry->vaddr_base, entry->size) != 0) {
             perror("unmap fails");
diff --git a/xen-stub.c b/xen-stub.c
index 9ea02d435e..8ff2b79ac6 100644
--- a/xen-stub.c
+++ b/xen-stub.c
@@ -29,6 +29,10 @@ void xen_piix_pci_write_config_client(uint32_t address, uint32_t val, int len)
 {
 }
 
+void xen_hvm_inject_msi(uint64_t addr, uint32_t data)
+{
+}
+
 void xen_cmos_set_s3_resume(void *opaque, int irq, int level)
 {
 }