xen/HVM: atomically access pointers in bufioreq handling

The number of slots per page being 511 (i.e. not a power of two) means
that the (32-bit) read and write indexes going beyond 2^32 will likely
disturb operation. The hypervisor side gets I/O req server creation
extended so we can indicate that we're using suitable atomic accesses
where needed, allowing it to atomically canonicalize both pointers when
both have gone through at least one cycle.

The Xen side counterpart (which is not a functional prereq to this
change, albeit a build one) went in already (commit b7007bc6f9).

Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
---
v3: Check for Xen 4.6 (configure and build time).
v2: Adjust description.

--- a/configure
+++ b/configure
@@ -1882,6 +1882,33 @@ int main(void) {
   xc_gnttab_open(NULL, 0);
   xc_domain_add_to_physmap(0, 0, XENMAPSPACE_gmfn, 0, 0);
   xc_hvm_inject_msi(xc, 0, 0xf0000000, 0x00000000);
+  xc_hvm_create_ioreq_server(xc, 0, HVM_IOREQSRV_BUFIOREQ_ATOMIC, NULL);
+  return 0;
+}
+EOF
+      compile_prog "" "$xen_libs"
+    then
+    xen_ctrl_version=460
+    xen=yes
+
+  # Xen 4.5
+  elif
+      cat > $TMPC <<EOF &&
+#include <xenctrl.h>
+#include <xenstore.h>
+#include <stdint.h>
+#include <xen/hvm/hvm_info_table.h>
+#if !defined(HVM_MAX_VCPUS)
+# error HVM_MAX_VCPUS not defined
+#endif
+int main(void) {
+  xc_interface *xc;
+  xs_daemon_open();
+  xc = xc_interface_open(0, 0, 0);
+  xc_hvm_set_mem_type(0, 0, HVMMEM_ram_ro, 0, 0);
+  xc_gnttab_open(NULL, 0);
+  xc_domain_add_to_physmap(0, 0, XENMAPSPACE_gmfn, 0, 0);
+  xc_hvm_inject_msi(xc, 0, 0xf0000000, 0x00000000);
   xc_hvm_create_ioreq_server(xc, 0, 0, NULL);
   return 0;
 }
--- a/xen-hvm.c
+++ b/xen-hvm.c
@@ -963,19 +963,30 @@ static void handle_ioreq(XenIOState *sta
 
 static int handle_buffered_iopage(XenIOState *state)
 {
+    buffered_iopage_t *buf_page = state->buffered_io_page;
     buf_ioreq_t *buf_req = NULL;
     ioreq_t req;
     int qw;
 
-    if (!state->buffered_io_page) {
+    if (!buf_page) {
         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];
+    for (;;) {
+        uint32_t rdptr = buf_page->read_pointer, wrptr;
+
+        xen_rmb();
+        wrptr = buf_page->write_pointer;
+        xen_rmb();
+        if (rdptr != buf_page->read_pointer) {
+            continue;
+        }
+        if (rdptr == wrptr) {
+            break;
+        }
+        buf_req = &buf_page->buf_ioreq[rdptr % IOREQ_BUFFER_SLOT_NUM];
         req.size = 1UL << buf_req->size;
         req.count = 1;
         req.addr = buf_req->addr;
@@ -987,15 +998,14 @@ static int handle_buffered_iopage(XenIOS
         req.data_is_ptr = 0;
         qw = (req.size == 8);
         if (qw) {
-            buf_req = &state->buffered_io_page->buf_ioreq[
-                (state->buffered_io_page->read_pointer + 1) % IOREQ_BUFFER_SLOT_NUM];
+            buf_req = &buf_page->buf_ioreq[(rdptr + 1) %
+                                           IOREQ_BUFFER_SLOT_NUM];
             req.data |= ((uint64_t)buf_req->data) << 32;
         }
 
         handle_ioreq(state, &req);
 
-        xen_mb();
-        state->buffered_io_page->read_pointer += qw ? 2 : 1;
+        atomic_add(&buf_page->read_pointer, qw + 1);
     }
 
     return req.count;
--- a/include/hw/xen/xen_common.h
+++ b/include/hw/xen/xen_common.h
@@ -186,6 +186,15 @@ static inline int xen_get_vmport_regs_pf
 }
 #endif
 
+/* Xen before 4.6 */
+#if CONFIG_XEN_CTRL_INTERFACE_VERSION < 460
+
+#ifndef HVM_IOREQSRV_BUFIOREQ_ATOMIC
+#define HVM_IOREQSRV_BUFIOREQ_ATOMIC 2
+#endif
+
+#endif
+
 /* Xen before 4.5 */
 #if CONFIG_XEN_CTRL_INTERFACE_VERSION < 450
 
@@ -370,7 +379,8 @@ static inline void xen_unmap_pcidev(XenX
 static inline int xen_create_ioreq_server(XenXC xc, domid_t dom,
                                           ioservid_t *ioservid)
 {
-    int rc = xc_hvm_create_ioreq_server(xc, dom, 1, ioservid);
+    int rc = xc_hvm_create_ioreq_server(xc, dom, HVM_IOREQSRV_BUFIOREQ_ATOMIC,
+                                        ioservid);
 
     if (rc == 0) {
         trace_xen_ioreq_server_create(*ioservid);