[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH v4 4/5] Qemu-Xen-vTPM: Qemu vTPM xenstubdoms bac

From: Stefan Berger
Subject: Re: [Qemu-devel] [PATCH v4 4/5] Qemu-Xen-vTPM: Qemu vTPM xenstubdoms backen.
Date: Mon, 23 Mar 2015 16:06:23 -0400
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0

On 03/23/2015 08:44 AM, Xu, Quan wrote:

-----Original Message-----
From: Stefan Berger [mailto:address@hidden
Sent: Thursday, March 19, 2015 3:17 AM
To: Xu, Quan; address@hidden; address@hidden;
address@hidden; address@hidden; address@hidden;
address@hidden; address@hidden; address@hidden;
address@hidden; address@hidden; address@hidden; address@hidden
Cc: address@hidden
Subject: Re: [PATCH v4 4/5] Qemu-Xen-vTPM: Qemu vTPM xenstubdoms

On 03/10/2015 08:14 AM, Quan Xu wrote:
This Patch provides the glue for the TPM_TIS(Qemu frontend) to Xen
stubdom vTPM domain that provides the actual TPM functionality. It
sends data and TPM commends with xen_vtpm_frontend. It is similar as
another two vTPM backens:
    *vTPM passthrough backen Since QEMU 1.5.
    *vTPM libtpms-based backen.

Some details:
This part of the patch provides support for the spawning of a thread
that will interact with stubdom vTPM domain by the xen_vtpm_frontend.
It expects a signal from the frontend to wake and pick up the TPM
command that is supposed to be processed and delivers the response
packet using a callback function provided by the frontend.

The backend connects itself to the frontend by filling out an
interface structure with pointers to the function implementing support
for various operations.

(QEMU) vTPM XenStubdoms backen is initialized by Qemu command line
        "-tpmdev xenstubdoms,id=xenvtpm0 -device
--Changes in v3:
-Call vtpm_send() and vtpm_recv() directly

--Changes in v4:
-Fix the comment style

Signed-off-by: Quan Xu <address@hidden>
   hw/tpm/Makefile.objs     |   2 +-
   hw/tpm/tpm_xenstubdoms.c | 247
   2 files changed, 248 insertions(+), 1 deletion(-)
   create mode 100644 hw/tpm/tpm_xenstubdoms.c

diff --git a/hw/tpm/Makefile.objs b/hw/tpm/Makefile.objs index
57919fa..190e776 100644
--- a/hw/tpm/Makefile.objs
+++ b/hw/tpm/Makefile.objs
@@ -1,3 +1,3 @@
   common-obj-$(CONFIG_TPM_TIS) += tpm_tis.o
   common-obj-$(CONFIG_TPM_PASSTHROUGH) += tpm_passthrough.o
-common-obj-$(CONFIG_TPM_XENSTUBDOMS) += xen_vtpm_frontend.o
+common-obj-$(CONFIG_TPM_XENSTUBDOMS) += tpm_xenstubdoms.o
diff --git a/hw/tpm/tpm_xenstubdoms.c b/hw/tpm/tpm_xenstubdoms.c new
file mode 100644 index 0000000..6d0dc32
--- /dev/null
+++ b/hw/tpm/tpm_xenstubdoms.c
@@ -0,0 +1,247 @@
+ * Xen Stubdom vTPM driver
+ *
+ *  Copyright (c) 2015 Intel Corporation
+ *  Authors:
+ *    Quan Xu <address@hidden>
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see
+<http://www.gnu.org/licenses/>  */
+#include <dirent.h>
+#include "qemu-common.h"
+#include "qapi/error.h"
+#include "qemu/sockets.h"
+#include "qemu/log.h"
+#include "sysemu/tpm_backend.h"
+#include "tpm_int.h"
+#include "hw/hw.h"
+#include "hw/i386/pc.h"
+#include "hw/xen/xen_backend.h"
+#include "sysemu/tpm_backend_int.h"
+#include "tpm_tis.h"
+#ifdef DEBUG_TPM
+#define DPRINTF(fmt, ...) \
+    do { fprintf(stderr, fmt, ## __VA_ARGS__); } while (0) #else
+#define DPRINTF(fmt, ...) \
+    do { } while (0)
+#define TYPE_TPM_XENSTUBDOMS "tpm-xenstubdoms"
+#define TPM_XENSTUBDOMS(obj) \
+    OBJECT_CHECK(TPMXenstubdomsState, (obj),
+static const TPMDriverOps tpm_xenstubdoms_driver;
+/* Data structures */
+typedef struct TPMXenstubdomsThreadParams {
+    TPMState *tpm_state;
+    TPMRecvDataCB *recv_data_callback;
+    TPMBackend *tb;
+} TPMXenstubdomsThreadParams;
+struct TPMXenstubdomsState {
+    TPMBackend parent;
+    TPMBackendThread tbt;
+    TPMXenstubdomsThreadParams tpm_thread_params;
+    bool had_startup_error;
+typedef struct TPMXenstubdomsState TPMXenstubdomsState;
+/* Functions */
+static void tpm_xenstubdoms_cancel_cmd(TPMBackend *tb);
+static int tpm_xenstubdoms_unix_transfer(const TPMLocality
+*locty_data) {
+    size_t rlen;
+    struct XenDevice *xendev;
+    xendev = xen_find_xendev("vtpm", xen_domid, xenstore_dev);
+    if (xendev == NULL) {
+        xen_be_printf(xendev, 0, "Con not find vtpm device\n");
Con not -> Cannot
+        return -1;
+    }
+    vtpm_send(xendev, locty_data->w_buffer.buffer,
+    vtpm_recv(xendev, locty_data->r_buffer.buffer, &rlen);
+    return 0;
+static void tpm_xenstubdoms_worker_thread(gpointer data,
+                                          gpointer user_data) {
+    TPMXenstubdomsThreadParams *thr_parms = user_data;
+    TPMBackendCmd cmd = (TPMBackendCmd)data;
+    switch (cmd) {
+        /* here need a the cmd process function */
+        thr_parms->recv_data_callback(thr_parms->tpm_state,
+ thr_parms->tpm_state->locty_number);
By now you'll need another parameter here indicating whether the command
that was issues was a selftest and whether it completed successfully. It should 
easy to support this and you could move
tpm_passthrough_is_selftest() into tpm_util.c and rename it to
tpm_util_is_selftest(). But this should be done in a separate patch.

Thanks Stefan.
One question, why should we highlight 'SelfTest' command?
Similar to TPM passthrough Driver backen, worker thread is just for 
distributing process command.
As I know, the command is from guest virtual machine kernel's tpm_tis.ko, if 
'SelfTest' command
Is failed, the guest virtual machine's tpm_tis.ko will return directly.

The TPM TIS 1.3 specification introduced a new flag in the status register in bit 2. Have a look at the specs here:


On page 68 that bit is explained and we need to find out whether to set or clear it.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]