Merge tag 'for-upstream' of https://gitlab.com/bonzini/qemu into staging

* Fixes for SGX
* force_rcu notifiers

# gpg: Signature made Wed 10 Nov 2021 10:57:48 PM CET
# gpg:                using RSA key F13338574B662389866C7682BFFBD25F78C7AE83
# gpg:                issuer "pbonzini@redhat.com"
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>" [full]
# gpg:                 aka "Paolo Bonzini <pbonzini@redhat.com>" [full]

* tag 'for-upstream' of https://gitlab.com/bonzini/qemu:
  sgx: Reset the vEPC regions during VM reboot
  numa: avoid crash with SGX and "info numa"
  accel/tcg: Register a force_rcu notifier
  rcu: Introduce force_rcu notifier
  target/i386: sgx: mark device not user creatable

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

3 files changed, 58 insertions, 0 deletions

diff --git a/hw/core/numa.c b/hw/core/numa.c
index 510d096..e6050b2 100644
--- a/hw/core/numa.c
+++ b/hw/core/numa.c
@@ -756,6 +756,7 @@ static void numa_stat_memory_devices(NumaNodeMem node_mem[])
     PCDIMMDeviceInfo     *pcdimm_info;
     VirtioPMEMDeviceInfo *vpi;
     VirtioMEMDeviceInfo *vmi;
+    SgxEPCDeviceInfo *se;
 
     for (info = info_list; info; info = info->next) {
         MemoryDeviceInfo *value = info->value;
@@ -781,6 +782,12 @@ static void numa_stat_memory_devices(NumaNodeMem node_mem[])
                 node_mem[vmi->node].node_mem += vmi->size;
                 node_mem[vmi->node].node_plugged_mem += vmi->size;
                 break;
+            case MEMORY_DEVICE_INFO_KIND_SGX_EPC:
+                se = value->u.sgx_epc.data;
+                /* TODO: once we support numa, assign to right node */
+                node_mem[0].node_mem += se->size;
+                node_mem[0].node_plugged_mem += se->size;
+                break;
             default:
                 g_assert_not_reached();
             }
diff --git a/hw/i386/sgx-epc.c b/hw/i386/sgx-epc.c
index 55e2217..e508827 100644
--- a/hw/i386/sgx-epc.c
+++ b/hw/i386/sgx-epc.c
@@ -154,6 +154,7 @@ static void sgx_epc_class_init(ObjectClass *oc, void *data)
     dc->realize = sgx_epc_realize;
     dc->unrealize = sgx_epc_unrealize;
     dc->desc = "SGX EPC section";
+    dc->user_creatable = false;
     device_class_set_props(dc, sgx_epc_properties);
 
     mdc->get_addr = sgx_epc_md_get_addr;
diff --git a/hw/i386/sgx.c b/hw/i386/sgx.c
index 1160756..8fef3dd 100644
--- a/hw/i386/sgx.c
+++ b/hw/i386/sgx.c
@@ -21,6 +21,8 @@
 #include "qapi/qapi-commands-misc-target.h"
 #include "exec/address-spaces.h"
 #include "sysemu/hw_accel.h"
+#include "sysemu/reset.h"
+#include <sys/ioctl.h>
 
 #define SGX_MAX_EPC_SECTIONS            8
 #define SGX_CPUID_EPC_INVALID           0x0
@@ -29,6 +31,11 @@
 #define SGX_CPUID_EPC_SECTION           0x1
 #define SGX_CPUID_EPC_MASK              0xF
 
+#define SGX_MAGIC 0xA4
+#define SGX_IOC_VEPC_REMOVE_ALL       _IO(SGX_MAGIC, 0x04)
+
+#define RETRY_NUM                       2
+
 static uint64_t sgx_calc_section_metric(uint64_t low, uint64_t high)
 {
     return (low & MAKE_64BIT_MASK(12, 20)) +
@@ -59,6 +66,46 @@ static uint64_t sgx_calc_host_epc_section_size(void)
     return size;
 }
 
+static void sgx_epc_reset(void *opaque)
+{
+    PCMachineState *pcms = PC_MACHINE(qdev_get_machine());
+    HostMemoryBackend *hostmem;
+    SGXEPCDevice *epc;
+    int failures;
+    int fd, i, j, r;
+    static bool warned = false;
+
+    /*
+     * The second pass is needed to remove SECS pages that could not
+     * be removed during the first.
+     */
+    for (i = 0; i < RETRY_NUM; i++) {
+        failures = 0;
+        for (j = 0; j < pcms->sgx_epc.nr_sections; j++) {
+            epc = pcms->sgx_epc.sections[j];
+            hostmem = MEMORY_BACKEND(epc->hostmem);
+            fd = memory_region_get_fd(host_memory_backend_get_memory(hostmem));
+
+            r = ioctl(fd, SGX_IOC_VEPC_REMOVE_ALL);
+            if (r == -ENOTTY && !warned) {
+                warned = true;
+                warn_report("kernel does not support SGX_IOC_VEPC_REMOVE_ALL");
+                warn_report("SGX might operate incorrectly in the guest after reset");
+                break;
+            } else if (r > 0) {
+                /* SECS pages remain */
+                failures++;
+                if (i == 1) {
+                    error_report("cannot reset vEPC section %d", j);
+                }
+            }
+        }
+        if (!failures) {
+            break;
+        }
+     }
+}
+
 SGXInfo *qmp_query_sgx_capabilities(Error **errp)
 {
     SGXInfo *info = NULL;
@@ -190,4 +237,7 @@ void pc_machine_init_sgx_epc(PCMachineState *pcms)
     }
 
     memory_region_set_size(&sgx_epc->mr, sgx_epc->size);
+
+    /* register the reset callback for sgx epc */
+    qemu_register_reset(sgx_epc_reset, NULL);
 }