[Xen-devel] [XEN][RFC PATCH V2 03/17] hvm-pci: Handle PCI config space in Xen

[Julien Grall <julien.grall@xxxxxxxxxx>]

Add function to register a bdf within a server.
An handler was add to catch (cf8 -> cff) ioport access.
When Xen reveices a PIO for cf8, it's store the value inside the current
vcpu until it receives a PIO for cfc -> cff.
In this case, it checks if the bdf is registered and forge the ioreq
that will be forward to server later.

Signed-off-by: Julien Grall <julien.grall@xxxxxxxxxx>
---
 xen/arch/x86/hvm/Makefile   |    1 +
 xen/arch/x86/hvm/pci_emul.c |  168 +++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 169 insertions(+), 0 deletions(-)
 create mode 100644 xen/arch/x86/hvm/pci_emul.c

diff --git a/xen/arch/x86/hvm/Makefile b/xen/arch/x86/hvm/Makefile
index eea5555..585e9c9 100644
--- a/xen/arch/x86/hvm/Makefile
+++ b/xen/arch/x86/hvm/Makefile
@@ -12,6 +12,7 @@ obj-y += irq.o
 obj-y += mtrr.o
 obj-y += nestedhvm.o
 obj-y += pmtimer.o
+obj-y += pci_emul.o
 obj-y += quirks.o
 obj-y += rtc.o
 obj-y += save.o
diff --git a/xen/arch/x86/hvm/pci_emul.c b/xen/arch/x86/hvm/pci_emul.c
new file mode 100644
index 0000000..48456dd
--- /dev/null
+++ b/xen/arch/x86/hvm/pci_emul.c
@@ -0,0 +1,168 @@
+#include <asm/hvm/support.h>
+#include <xen/hvm/pci_emul.h>
+#include <xen/pci.h>
+#include <xen/sched.h>
+#include <xen/xmalloc.h>
+
+#define PCI_DEBUGSTR "%x:%x.%x"
+#define PCI_DEBUG(bdf) ((bdf) >> 8) & 0xff, ((bdf) >> 3) & 0x1f, ((bdf)) & 0x7
+#define PCI_MASK_BDF(bdf) (((bdf) & 0x00ffff00) >> 8)
+#define PCI_CMP_BDF(Pci, Bdf) ((pci)->bdf == PCI_MASK_BDF(Bdf))
+
+static int handle_config_space(int dir, uint32_t port, uint32_t bytes,
+                               uint32_t *val)
+{
+    uint32_t pci_cf8;
+    struct hvm_ioreq_server *s;
+    ioreq_t *p = get_ioreq(current);
+    int rc = X86EMUL_UNHANDLEABLE;
+    struct vcpu *v = current;
+
+    if ( port == 0xcf8 && bytes == 4 )
+    {
+        if ( dir == IOREQ_READ )
+            *val = v->arch.hvm_vcpu.pci_cf8;
+        else
+            v->arch.hvm_vcpu.pci_cf8 = *val;
+        return X86EMUL_OKAY;
+    }
+    else if ( port < 0xcfc )
+        return X86EMUL_UNHANDLEABLE;
+
+    spin_lock(&v->domain->arch.hvm_domain.pci_root.pci_lock);
+    spin_lock(&v->domain->arch.hvm_domain.ioreq_server_lock);
+
+    pci_cf8 = v->arch.hvm_vcpu.pci_cf8;
+
+    /* Retrieve PCI */
+    s = radix_tree_lookup(&v->domain->arch.hvm_domain.pci_root.pci_list,
+                          PCI_MASK_BDF(pci_cf8));
+
+    if ( unlikely(s == NULL) )
+    {
+        *val = ~0;
+        rc = X86EMUL_OKAY;
+        goto end_handle;
+    }
+
+    /**
+     * We just fill the ioreq, hvm_send_assist_req will send the request
+     * The size is used to find the right access
+     **/
+    /* We use the 16 high-bits for the offset (0 => 0xcfc, 1 => 0xcfd...) */
+    p->size = (p->addr - 0xcfc) << 16 | (p->size & 0xffff);
+    p->type = IOREQ_TYPE_PCI_CONFIG;
+    p->addr = pci_cf8;
+
+    set_ioreq(v, &s->ioreq, p);
+
+end_handle:
+    spin_unlock(&v->domain->arch.hvm_domain.ioreq_server_lock);
+    spin_unlock(&v->domain->arch.hvm_domain.pci_root.pci_lock);
+
+    return rc;
+}
+
+int hvm_register_pcidev(domid_t domid, ioservid_t id,
+                        uint8_t domain, uint8_t bus,
+                        uint8_t device, uint8_t function)
+{
+    struct domain *d;
+    struct hvm_ioreq_server *s;
+    int rc = 0;
+    struct radix_tree_root *tree;
+    uint16_t bdf = 0;
+
+    /* For the moment we don't handle pci when domain != 0 */
+    if ( domain != 0 )
+        return -EINVAL;
+
+    rc = rcu_lock_target_domain_by_id(domid, &d);
+
+    if ( rc != 0 )
+        return rc;
+
+    if ( !is_hvm_domain(d) )
+    {
+        rcu_unlock_domain(d);
+        return -EINVAL;
+    }
+
+    /* Search server */
+    spin_lock(&d->arch.hvm_domain.ioreq_server_lock);
+    s = d->arch.hvm_domain.ioreq_server_list;
+    while ( (s != NULL) && (s->id != id) )
+        s = s->next;
+
+    spin_unlock(&d->arch.hvm_domain.ioreq_server_lock);
+
+    if ( s == NULL )
+    {
+        gdprintk(XENLOG_ERR, "Cannot find server %u\n", id);
+        rc = -ENOENT;
+        goto fail;
+    }
+
+    spin_lock(&d->arch.hvm_domain.pci_root.pci_lock);
+
+    tree = &d->arch.hvm_domain.pci_root.pci_list;
+
+    bdf |= ((uint16_t)bus) << 8;
+    bdf |= ((uint16_t)device & 0x1f) << 3;
+    bdf |= ((uint16_t)function & 0x7);
+
+    if ( radix_tree_lookup(tree, bdf) )
+    {
+        rc = -EEXIST;
+        gdprintk(XENLOG_ERR, "Bdf " PCI_DEBUGSTR " is already allocated\n",
+                 PCI_DEBUG(bdf));
+        goto create_end;
+    }
+
+    rc = radix_tree_insert(tree, bdf, s);
+    if ( rc )
+    {
+        gdprintk(XENLOG_ERR, "Cannot insert the bdf\n");
+        goto create_end;
+    }
+
+create_end:
+    spin_unlock(&d->arch.hvm_domain.pci_root.pci_lock);
+fail:
+    rcu_unlock_domain(d);
+
+    return rc;
+}
+
+void hvm_init_pci_emul(struct domain *d)
+{
+    struct pci_root_emul *root = &d->arch.hvm_domain.pci_root;
+
+    spin_lock_init(&root->pci_lock);
+
+    radix_tree_init(&root->pci_list);
+
+    /* Register the config space handler */
+    register_portio_handler(d, 0xcf8, 8, handle_config_space);
+}
+
+void hvm_destroy_pci_emul(struct domain *d)
+{
+    struct pci_root_emul *root = &d->arch.hvm_domain.pci_root;
+
+    spin_lock(&root->pci_lock);
+
+    radix_tree_destroy(&root->pci_list, NULL);
+
+    spin_unlock(&root->pci_lock);
+}
+
+/*
+ * Local variables:
+ * mode: C
+ * c-set-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
-- 
Julien Grall


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel