[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 2/4] Qemu-Xen-vTPM: Register Xen stubdom vTPM fr
From: |
Xu, Quan |
Subject: |
Re: [Qemu-devel] [PATCH 2/4] Qemu-Xen-vTPM: Register Xen stubdom vTPM frontend driver |
Date: |
Thu, 6 Nov 2014 09:03:55 +0000 |
> -----Original Message-----
> From: Stefano Stabellini [mailto:address@hidden
> Sent: Monday, November 03, 2014 7:54 PM
> To: Xu, Quan
> Cc: address@hidden; address@hidden;
> address@hidden
> Subject: Re: [PATCH 2/4] Qemu-Xen-vTPM: Register Xen stubdom vTPM
> frontend driver
>
> On Sun, 2 Nov 2014, Quan Xu wrote:
> > This drvier transfers any request/repond between TPM xenstubdoms
> > driver and Xen vTPM stubdom, and facilitates communications between
> > Xen vTPM stubdom domain and vTPM xenstubdoms driver
> >
> > Signed-off-by: Quan Xu <address@hidden>
>
> Please describe what changes did make to xen_backend.c and why.
> The commit message should contains info on all the changes made by the
> patch below.
>
The following is code process when Qemu is running with Xen.
##code process
[...]
xen_hvm_init()
-->xen_be_register()
-->xenstore_scan()
-->xen_be_check_state()
-->xen_vtpm_register()
ideally, I can register 'vtpm' via xen_vtpm_register() as
+ xen_be_register("console", &xen_console_ops);
+ xen_be_register("vkbd", &xen_kbdmouse_ops);
+ xen_be_register("qdisk", &xen_blkdev_ops);
but there are 2 reasons why I add xen_vtpm_register(), instead
of xen_be_register().
1. The backend of TPM is runing in a Xen stubDom, not Domain 0.
some functions are not working, for example 'setup watch' and
'look for backend' in xenstore_scan()
2. there is a thread runing in Xen stubDom [event_thread()], it will
handle backend status when the frontend is initialized. It is not
compatible with xen_be_check_state(). xen_be_check_state() always tries
to modify the status of backend.
as there is always a tradeoff, if I force to integrate this case into
xen_be_register(), there are maybe a lot of 'if ... else.. '. It will
break the code architecture. Also I should leverage existing source
code with minimum modifcation. i add 'DEVOPS_FLAG_STUBDOM_BE' flag in
include/hw/xen/xen_backend.h to indicate that device backend is Xen
stubDom.
> Please also describe what is the "Xen vTPM stubdom", what is the
> "vTPM xenstubdoms driver" and how the communicate with each others.
>
In previous email, I have explained what is "Xen vTPM stubdom", what is the
" vTPM xenstubdoms driver ".
Let me describe how the communicate with each others.
| | ^ |
| v | |
| vTPM |
| XenStubdoms driver | (new ..)
+-----------------------------+
| ^
v |
+-----------------------------+
| xen_vtpmdev_ops | (new ..)
+-----------------------------+
xen_vtpmdev_ops is initialized with the following process:
xen_hvm_init()
[...]
-->xen_vtpm_register()
[...]
--> vtpm_alloc()
--> vtpm_initialise()
##
vTPM XenStubdoms driver is initialized by Qemu command line options,
"-tpmdev xenstubdoms,id=xenvtpm0 -device tpm-tis,tpmdev=xenvtpm0"
(the communicate with each others in following function.
.. hw/tpm/tpm_xenstubdoms.c
static int tpm_xenstubdoms_unix_transfer(const TPMLocality *locty_data)
{
[...]
xen_vtpm_send(locty_data->w_buffer.buffer, locty_data->w_offset);
xen_vtpm_recv(locty_data->r_buffer.buffer, &rlen);
[...]
}
> Where does the vTPM backend lives?
Xen stubDom.
>
>
> > hw/xen/Makefile.objs | 1 +
> > hw/xen/xen_backend.c | 182 ++++++++++++++++++++++-
> > hw/xen/xen_stubdom_vtpm.c | 333
> +++++++++++++++++++++++++++++++++++++++++++
> > include/hw/xen/xen_backend.h | 11 ++
> > include/hw/xen/xen_common.h | 6 +
> > xen-hvm.c | 13 ++
> > 6 files changed, 544 insertions(+), 2 deletions(-)
> > create mode 100644 hw/xen/xen_stubdom_vtpm.c
> >
> > diff --git a/hw/xen/Makefile.objs b/hw/xen/Makefile.objs
> > index a0ca0aa..724df8d 100644
> > --- a/hw/xen/Makefile.objs
> > +++ b/hw/xen/Makefile.objs
> > @@ -1,5 +1,6 @@
> > # xen backend driver support
> > common-obj-$(CONFIG_XEN_BACKEND) += xen_backend.o
> xen_devconfig.o
> > +common-obj-$(CONFIG_TPM_XENSTUBDOMS) += xen_stubdom_vtpm.o
> >
> > obj-$(CONFIG_XEN_PCI_PASSTHROUGH) += xen-host-pci-device.o
> > obj-$(CONFIG_XEN_PCI_PASSTHROUGH) += xen_pt.o
> xen_pt_config_init.o xen_pt_msi.o
> > diff --git a/hw/xen/xen_backend.c b/hw/xen/xen_backend.c
> > index b2cb22b..45a5778 100644
> > --- a/hw/xen/xen_backend.c
> > +++ b/hw/xen/xen_backend.c
> > @@ -194,6 +194,32 @@ int xen_be_set_state(struct XenDevice *xendev,
> enum xenbus_state state)
> > return 0;
> > }
> >
> > +/*get stubdom backend*/
> > +static char *xen_stubdom_be(const char *type, int dom, int dev)
> > +{
> > + char *val, *domu;
> > + char path[XEN_BUFSIZE];
> > + unsigned int len, ival;
> > +
> > + /*front domu*/
> > + domu = xs_get_domain_path(xenstore, dom);
> > + snprintf(path, sizeof(path), "%s/device/%s/%d/backend-id",
> > + domu, type, dev);
> > + g_free(domu);
> > +
> > + val = xs_read(xenstore, 0, path, &len);
> > + if (!val || 1 != sscanf(val, "%d", &ival)) {
> > + g_free(val);
> > + return NULL;
> > + }
> > + g_free(val);
> > +
> > + /*backend domu*/
> > + domu = xs_get_domain_path(xenstore, ival);
> > +
> > + return domu;
> > +}
> > +
> > /* ------------------------------------------------------------- */
> >
> > struct XenDevice *xen_be_find_xendev(const char *type, int dom, int
> dev)
> > @@ -222,6 +248,7 @@ static struct XenDevice *xen_be_get_xendev(const
> char *type, int dom, int dev,
> > struct XenDevOps *ops)
> > {
> > struct XenDevice *xendev;
> > + char *stub;
> >
> > xendev = xen_be_find_xendev(type, dom, dev);
> > if (xendev) {
> > @@ -235,8 +262,15 @@ static struct XenDevice
> *xen_be_get_xendev(const char *type, int dom, int dev,
> > xendev->dev = dev;
> > xendev->ops = ops;
> >
> > - snprintf(xendev->be, sizeof(xendev->be), "backend/%s/%d/%d",
> > - xendev->type, xendev->dom, xendev->dev);
> > + if (ops->flags & DEVOPS_FLAG_STUBDOM_BE) {
> > + stub = xen_stubdom_be(xendev->type, xendev->dom,
> xendev->dev);
> > + snprintf(xendev->be, sizeof(xendev->be),
> "%s/backend/%s/%d/%d",
> > + stub, xendev->type, xendev->dom, xendev->dev);
> > + g_free(stub);
> > + } else {
> > + snprintf(xendev->be, sizeof(xendev->be), "backend/%s/%d/%d",
> > + xendev->type, xendev->dom, xendev->dev);
> > + }
> > snprintf(xendev->name, sizeof(xendev->name), "%s-%d",
> > xendev->type, xendev->dev);
> >
> > @@ -611,6 +645,47 @@ static int xenstore_scan(const char *type, int
> dom, struct XenDevOps *ops)
> > return 0;
> > }
> >
> > +static void stubdom_update_be(char *watch, char *type, int dom,
> > + struct XenDevOps *ops)
> > +{
> > + struct XenDevice *xendev;
> > + char path[XEN_BUFSIZE];
> > + char *ptr, *bepath;
> > + unsigned int len, dev;
> > +
> > + if (!(ops->flags & DEVOPS_FLAG_STUBDOM_BE)) {
> > + return;
> > + }
> > +
> > + len = snprintf(path, sizeof(path), "backend/%s/%d", type, dom);
> > + ptr = strstr(watch, path);
> > + if (ptr == NULL) {
> > + return;
> > + }
> > +
> > + if (sscanf(ptr+len, "/%u/%255s", &dev, path) != 2) {
> > + strcpy(path, "");
> > + if (sscanf(ptr+len, "/%u", &dev) != 1) {
> > + dev = -1;
> > + }
> > + }
> > +
> > + if (dev == -1) {
> > + return;
> > + }
> > +
> > + xendev = xen_be_get_xendev(type, dom, dev, ops);
> > + if (xendev != NULL) {
> > + 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);
> > + }
> > + }
> > +}
> > +
> > static void xenstore_update_be(char *watch, char *type, int dom,
> > struct XenDevOps *ops)
> > {
> > @@ -681,6 +756,10 @@ static void xenstore_update(void *unused)
> > if (sscanf(vec[XS_WATCH_TOKEN], "fe:%" PRIxPTR, &ptr) == 1) {
> > xenstore_update_fe(vec[XS_WATCH_PATH], (void*)ptr);
> > }
> > + if (sscanf(vec[XS_WATCH_TOKEN], "stub:%" PRIxPTR ":%d:%"
> PRIxPTR,
> > + &type, &dom, &ops) == 3) {
> > + stubdom_update_be(vec[XS_WATCH_PATH], (void *)type, dom,
> (void *)ops);
> > + }
> >
> > cleanup:
> > free(vec);
> > @@ -732,11 +811,74 @@ err:
> > return -1;
> > }
> >
> > +int xen_vtpm_register(struct XenDevOps *ops)
> > +{
> > + struct XenDevice *xendev;
> > + char path[XEN_BUFSIZE], token[XEN_BUFSIZE];
> > + char *domu;
> > + unsigned int cdev, j, rc;
> > + const char *type = "vtpm";
> > + char **dev = NULL;
> > +
> > + /*front domu*/
> > + domu = xs_get_domain_path(xenstore, xen_domid);
> > + snprintf(path, sizeof(path), "%s/device/%s",
> > + domu, type);
> > + free(domu);
> > + dev = xs_directory(xenstore, 0, path, &cdev);
> > + if (dev == NULL) {
> > + return 0;
> > + }
> > +
> > + for (j = 0; j < cdev; j++) {
> > + xendev = xen_be_get_xendev(type, xen_domid, atoi(dev[j]),
> ops);
> > + if (xendev == NULL) {
> > + xen_be_printf(xendev, 0, "xen_vtpm_register xendev is
> NULL.\n");
> > + continue;
> > + }
> > +
> > + if (xendev->ops->initialise) {
> > + rc = xendev->ops->initialise(xendev);
> > +
> > + /*if initialise failed, delete it*/
> > + if (rc != 0) {
> > + xen_be_del_xendev(xen_domid, atoi(dev[j]));
> > + continue;
> > + }
> > + }
> > +
> > + /*setup watch*/
> > + snprintf(token, sizeof(token), "stub:%p:%d:%p",
> > + type, xen_domid, xendev->ops);
> > + if (!xs_watch(xenstore, xendev->be, token)) {
> > + xen_be_printf(xendev, 0, "xen_vtpm_register xs_watch
> failed.\n");
> > + return -1;
> > + }
> > + }
> > +
> > + free(dev);
> > + return 0;
> > +}
>
> What does this function do? I sholdn't need to guess from the code, I
> should be able to tell from the patch description.
>
1. setup watch
2. look for backends
3. initialize the frontend
>
> > int xen_be_register(const char *type, struct XenDevOps *ops)
> > {
> > return xenstore_scan(type, xen_domid, ops);
> > }
> >
> > +int xen_be_alloc_unbound(struct XenDevice *xendev, int dom, int
> remote_dom)
> > +{
> > + xendev->local_port =
> xc_evtchn_bind_unbound_port(xendev->evtchndev,
> > +
> remote_dom);
> > + if (xendev->local_port == -1) {
> > + xen_be_printf(xendev, 0, "xc_evtchn_alloc_unbound failed\n");
> > + return -1;
> > + }
> > + xen_be_printf(xendev, 2, "bind evtchn port %d\n",
> xendev->local_port);
> > + qemu_set_fd_handler(xc_evtchn_fd(xendev->evtchndev),
> > + xen_be_evtchn_event, NULL, xendev);
> > + return 0;
> > +}
> > +
> > int xen_be_bind_evtchn(struct XenDevice *xendev)
> > {
> > if (xendev->local_port != -1) {
> > @@ -770,6 +912,42 @@ int xen_be_send_notify(struct XenDevice
> *xendev)
> > return xc_evtchn_notify(xendev->evtchndev, xendev->local_port);
> > }
> >
> > +int xen_vtpm_send(unsigned char *buf, size_t count)
> > +{
> > + struct XenDevice *xendev;
> > + int rc = -1;
> > +
> > + xendev = xen_be_find_xendev("vtpm", xen_domid, 0);
> > + if (xendev == NULL) {
> > + xen_be_printf(xendev, 0, "Con not find vtpm device\n");
> > + return -1;
> > + }
> > +
> > + if (xendev->ops->send) {
> > + rc = xendev->ops->send(xendev, buf, count);
> > + }
> > +
> > + return rc;
> > +}
> > +
> > +int xen_vtpm_recv(unsigned char *buf, size_t *count)
> > +{
> > + struct XenDevice *xendev;
> > + int rc = -1;
> > +
> > + xendev = xen_be_find_xendev("vtpm", xen_domid, 0);
> > + if (xendev == NULL) {
> > + xen_be_printf(xendev, 0, "Con not find vtpm device\n");
> > + return -1;
> > + }
> > +
> > + if (xendev->ops->recv) {
> > + xendev->ops->recv(xendev, buf, count);
> > + }
> > +
> > + return rc;
> > +}
>
> xen_backend.c is supposed to be generic, so stubdom functions might be
> OK but vtpm specific functions should not be here.
>
OK, I will move it to hw/tpm.
>
> > /*
> > * msg_level:
> > * 0 == errors (stderr + logfile).
> > diff --git a/hw/xen/xen_stubdom_vtpm.c b/hw/xen/xen_stubdom_vtpm.c
> > new file mode 100644
> > index 0000000..0d740c1
> > --- /dev/null
> > +++ b/hw/xen/xen_stubdom_vtpm.c
> > @@ -0,0 +1,333 @@
> > +/*
> > + * Connect to Xen vTPM stubdom domain
> > + *
> > + * Copyright (c) 2014 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
> > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> GNU
> > + * 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 <stdio.h>
> > +#include <stdlib.h>
> > +#include <stdarg.h>
> > +#include <string.h>
> > +#include <unistd.h>
> > +#include <signal.h>
> > +#include <inttypes.h>
> > +#include <time.h>
> > +#include <fcntl.h>
> > +#include <errno.h>
> > +#include <sys/ioctl.h>
> > +#include <sys/types.h>
> > +#include <sys/stat.h>
> > +#include <sys/mman.h>
> > +#include <sys/uio.h>
> > +
> > +#include "hw/hw.h"
> > +#include "block/aio.h"
> > +#include "hw/xen/xen_backend.h"
> > +
> > +enum tpmif_state {
> > + TPMIF_STATE_IDLE, /* no contents / vTPM idle / cancel
> complete */
> > + TPMIF_STATE_SUBMIT, /* request ready / vTPM working */
> > + TPMIF_STATE_FINISH, /* response ready / vTPM idle */
> > + TPMIF_STATE_CANCEL, /* cancel requested / vTPM working */
> > +};
> > +
> > +static AioContext *vtpm_aio_ctx;
> > +
> > +enum status_bits {
> > + VTPM_STATUS_RUNNING = 0x1,
> > + VTPM_STATUS_IDLE = 0x2,
> > + VTPM_STATUS_RESULT = 0x4,
> > + VTPM_STATUS_CANCELED = 0x8,
> > +};
> > +
> > +struct tpmif_shared_page {
> > + uint32_t length; /* request/response length in bytes */
> > +
> > + uint8_t state; /* enum tpmif_state */
> > + uint8_t locality; /* for the current request */
> > + uint8_t pad; /* should be zero */
> > +
> > + uint8_t nr_extra_pages; /* extra pages for long packets; may be
> zero */
> > + uint32_t extra_pages[0]; /* grant IDs; length is actually
> nr_extra_pages */
> > +};
> > +
> > +struct QEMUBH {
> > + AioContext *ctx;
> > + QEMUBHFunc *cb;
> > + void *opaque;
> > + QEMUBH *next;
> > + bool scheduled;
> > + bool idle;
> > + bool deleted;
> > +};
> > +
> > +struct XenVtpmDev {
> > + struct XenDevice xendev; /* must be first */
> > + struct tpmif_shared_page *shr;
> > + xc_gntshr *xen_xcs;
> > + int ring_ref;
> > + int bedomid;
> > + QEMUBH *sr_bh;
> > +};
> > +
> > +static uint8_t vtpm_status(struct XenVtpmDev *vtpmdev)
> > +{
> > + switch (vtpmdev->shr->state) {
> > + case TPMIF_STATE_IDLE:
> > + case TPMIF_STATE_FINISH:
> > + return VTPM_STATUS_IDLE;
> > + case TPMIF_STATE_SUBMIT:
> > + case TPMIF_STATE_CANCEL:
> > + return VTPM_STATUS_RUNNING;
> > + default:
> > + return 0;
> > + }
> > +}
> > +
> > +static int xenbus_switch_state(struct XenDevice *xendev, enum
> xenbus_state xbus)
> > +{
> > + xs_transaction_t xbt = XBT_NULL;
> > +
> > + if (xendev->fe_state == xbus) {
> > + return 0;
> > + }
> > +
> > + xendev->fe_state = xbus;
> > +
> > +retry_transaction:
> > + xbt = xs_transaction_start(xenstore);
> > + if (xbt == XBT_NULL) {
> > + goto abort_transaction;
> > + }
> > +
> > + if (xenstore_write_int(xendev->fe, "state", xbus)) {
> > + goto abort_transaction;
> > + }
> > +
> > + if (!xs_transaction_end(xenstore, xbt, 0)) {
> > + if (errno == EAGAIN) {
> > + goto retry_transaction;
> > + }
> > + }
> > +
> > + return 0;
> > +
> > +abort_transaction:
> > + xs_transaction_end(xenstore, xbt, 1);
> > + return -1;
> > +}
> > +
> > +static int vtpm_aio_wait(QEMUBH *qbh)
> > +{
> > + return aio_poll(qbh->ctx, true);
> > +}
> > +
> > +static void sr_bh_handler(void *opaque)
> > +{
> > +}
> > +
> > +static int vtpm_recv(struct XenDevice *xendev, uint8_t* buf, size_t
> *count)
> > +{
> > + struct XenVtpmDev *vtpmdev = container_of(xendev, struct
> XenVtpmDev,
> > + xendev);
> > + struct tpmif_shared_page *shr = vtpmdev->shr;
> > + unsigned int offset;
> > +
> > + if (shr->state == TPMIF_STATE_IDLE) {
> > + return -ECANCELED;
> > + }
> > +
> > + while (vtpm_status(vtpmdev) != VTPM_STATUS_IDLE) {
> > + vtpm_aio_wait(vtpmdev->sr_bh);
> > + }
> > +
> > + offset = sizeof(*shr) + 4*shr->nr_extra_pages;
> > + memcpy(buf, offset + (uint8_t *)shr, shr->length);
> > + *count = shr->length;
> > +
> > + return 0;
> > +}
> > +
> > +static int vtpm_send(struct XenDevice *xendev, uint8_t* buf, size_t
> count)
> > +{
> > + struct XenVtpmDev *vtpmdev = container_of(xendev, struct
> XenVtpmDev,
> > + xendev);
> > + struct tpmif_shared_page *shr = vtpmdev->shr;
> > + unsigned int offset = sizeof(*shr) + 4*shr->nr_extra_pages;
> > +
> > + while (vtpm_status(vtpmdev) != VTPM_STATUS_IDLE) {
> > + vtpm_aio_wait(vtpmdev->sr_bh);
> > + }
> > +
> > + memcpy(offset + (uint8_t *)shr, buf, count);
> > + shr->length = count;
> > + barrier();
> > + shr->state = TPMIF_STATE_SUBMIT;
> > + xen_wmb();
> > + xen_be_send_notify(&vtpmdev->xendev);
> > +
> > + while (vtpm_status(vtpmdev) != VTPM_STATUS_IDLE) {
> > + vtpm_aio_wait(vtpmdev->sr_bh);
> > + }
> > +
> > + return count;
> > +}
> > +
> > +static int vtpm_initialise(struct XenDevice *xendev)
> > +{
> > + struct XenVtpmDev *vtpmdev = container_of(xendev, struct
> XenVtpmDev,
> > + xendev);
> > + xs_transaction_t xbt = XBT_NULL;
> > + unsigned int ring_ref;
> > +
> > + vtpmdev->xendev.fe = xenstore_read_be_str(&vtpmdev->xendev,
> "frontend");
> > + if (vtpmdev->xendev.fe == NULL) {
> > + return -1;
> > + }
> > +
> > + /* Get backend domid */
> > + if (xenstore_read_fe_int(&vtpmdev->xendev, "backend-id",
> > + &vtpmdev->bedomid)) {
> > + return -1;
> > + }
> > +
> > + /*alloc share page*/
> > + vtpmdev->shr = xc_gntshr_share_pages(vtpmdev->xen_xcs,
> vtpmdev->bedomid, 1,
> > + &ring_ref,
> PROT_READ|PROT_WRITE);
> > + vtpmdev->ring_ref = ring_ref;
> > + if (vtpmdev->shr == NULL) {
> > + return -1;
> > + }
> > +
> > + /*Create event channel */
> > + if (xen_be_alloc_unbound(&vtpmdev->xendev, 0,
> vtpmdev->bedomid)) {
> > + xc_gntshr_munmap(vtpmdev->xen_xcs, vtpmdev->shr, 1);
> > + return -1;
> > + }
> > +
> > + xc_evtchn_unmask(vtpmdev->xendev.evtchndev,
> > + vtpmdev->xendev.local_port);
> > +
> > +again:
> > + xbt = xs_transaction_start(xenstore);
> > + if (xbt == XBT_NULL) {
> > + goto abort_transaction;
> > + }
> > +
> > + if (xenstore_write_int(vtpmdev->xendev.fe, "ring-ref",
> > + vtpmdev->ring_ref)) {
> > + goto abort_transaction;
> > + }
> > +
> > + if (xenstore_write_int(vtpmdev->xendev.fe, "event-channel",
> > + vtpmdev->xendev.local_port)) {
> > + goto abort_transaction;
> > + }
> > +
> > + /* Publish protocol v2 feature */
> > + if (xenstore_write_int(vtpmdev->xendev.fe, "feature-protocol-v2", 1))
> {
> > + goto abort_transaction;
> > + }
> > +
> > + if (!xs_transaction_end(xenstore, xbt, 0)) {
> > + if (errno == EAGAIN) {
> > + goto again;
> > + }
> > + }
> > + /* Tell vtpm backend that we are ready */
> > + xenbus_switch_state(&vtpmdev->xendev, XenbusStateInitialised);
> > +
> > + return 0;
> > +
> > +abort_transaction:
> > + xc_gntshr_munmap(vtpmdev->xen_xcs, vtpmdev->shr, 1);
> > + xs_transaction_end(xenstore, xbt, 1);
> > + return -1;
> > +}
> > +
> > +static void vtpm_backend_changed(struct XenDevice *xendev, const char
> *node)
> > +{
> > + struct XenVtpmDev *vtpmdev = container_of(xendev, struct
> XenVtpmDev,
> > + xendev);
> > + int be_state;
> > +
> > + if (strcmp(node, "state") == 0) {
> > + xenstore_read_be_int(&vtpmdev->xendev, node, &be_state);
> > + switch (be_state) {
> > + case XenbusStateConnected:
> > + /*TODO*/
> > + break;
> > + case XenbusStateClosing:
> > + case XenbusStateClosed:
> > + xenbus_switch_state(&vtpmdev->xendev,
> XenbusStateClosing);
> > + break;
> > + default:
> > + break;
> > + }
> > + }
> > +}
> > +
> > +static int vtpm_free(struct XenDevice *xendev)
> > +{
> > + struct XenVtpmDev *vtpmdev = container_of(xendev, struct
> XenVtpmDev,
> > + xendev);
> > + QEMUBH *qbh = vtpmdev->sr_bh;
> > +
> > + aio_poll(qbh->ctx, false);
> > + qemu_bh_delete(vtpmdev->sr_bh);
> > + if (vtpmdev->shr) {
> > + xc_gntshr_munmap(vtpmdev->xen_xcs, vtpmdev->shr, 1);
> > + }
> > + xc_interface_close(vtpmdev->xen_xcs);
> > + return 0;
> > +}
> > +
> > +static void vtpm_alloc(struct XenDevice *xendev)
> > +{
> > + struct XenVtpmDev *vtpmdev = container_of(xendev, struct
> XenVtpmDev,
> > + xendev);
> > +
> > + vtpm_aio_ctx = aio_context_new(NULL);
> > + if (vtpm_aio_ctx == NULL) {
> > + return;
> > + }
> > + vtpmdev->sr_bh = aio_bh_new(vtpm_aio_ctx, sr_bh_handler,
> vtpmdev);
> > + qemu_bh_schedule(vtpmdev->sr_bh);
> > + vtpmdev->xen_xcs = xen_xc_gntshr_open(0, 0);
> > +}
> > +
> > +static void vtpm_event(struct XenDevice *xendev)
> > +{
> > + struct XenVtpmDev *vtpmdev = container_of(xendev, struct
> XenVtpmDev,
> > + xendev);
> > +
> > + qemu_bh_schedule(vtpmdev->sr_bh);
> > +}
> > +
> > +struct XenDevOps xen_vtpmdev_ops = {
> > + .size = sizeof(struct XenVtpmDev),
> > + .flags = DEVOPS_FLAG_IGNORE_STATE |
> > + DEVOPS_FLAG_STUBDOM_BE,
> > + .event = vtpm_event,
> > + .free = vtpm_free,
> > + .alloc = vtpm_alloc,
> > + .initialise = vtpm_initialise,
> > + .backend_changed = vtpm_backend_changed,
> > + .recv = vtpm_recv,
> > + .send = vtpm_send,
> > +};
>
> Is this the frontend, like the subject line would seem to imply?
> If so, XenDevOps are made for backends, while this is a frontend. In
> fact this is the first PV frontend in QEMU. We need to introduce
> something generic and similar to struct XenDevOps and xen_backend.c but
> for frontends.
>
>
> > diff --git a/include/hw/xen/xen_backend.h
> b/include/hw/xen/xen_backend.h
> > index 3b4125e..45fd6d3 100644
> > --- a/include/hw/xen/xen_backend.h
> > +++ b/include/hw/xen/xen_backend.h
> > @@ -15,6 +15,8 @@ struct XenDevice;
> > #define DEVOPS_FLAG_NEED_GNTDEV 1
> > /* don't expect frontend doing correct state transitions (aka console
> quirk) */
> > #define DEVOPS_FLAG_IGNORE_STATE 2
> > +/*dev backend is stubdom*/
> > +#define DEVOPS_FLAG_STUBDOM_BE 4
> >
> > struct XenDevOps {
> > size_t size;
> > @@ -26,6 +28,8 @@ struct XenDevOps {
> > void (*event)(struct XenDevice *xendev);
> > void (*disconnect)(struct XenDevice *xendev);
> > int (*free)(struct XenDevice *xendev);
> > + int (*send)(struct XenDevice *xendev, uint8_t* buf, size_t
> count);
> > + int (*recv)(struct XenDevice *xendev, uint8_t* buf, size_t
> *count);
> > void (*backend_changed)(struct XenDevice *xendev, const
> char *node);
> > void (*frontend_changed)(struct XenDevice *xendev, const
> char *node);
> > };
> > @@ -91,12 +95,19 @@ int xen_be_send_notify(struct XenDevice
> *xendev);
> > void xen_be_printf(struct XenDevice *xendev, int msg_level, const char
> *fmt, ...)
> > GCC_FMT_ATTR(3, 4);
> >
> > +/*Xen stubdom vtpm*/
> > +int xen_vtpm_register(struct XenDevOps *ops);
> > +int xen_be_alloc_unbound(struct XenDevice *xendev, int dom, int
> remote_dom);
> > +int xen_vtpm_send(unsigned char *buf, size_t count);
> > +int xen_vtpm_recv(unsigned char *buf, size_t *count);
> > +
> > /* actual backend drivers */
> > extern struct XenDevOps xen_console_ops; /* xen_console.c
> */
> > extern struct XenDevOps xen_kbdmouse_ops; /* xen_framebuffer.c
> */
> > extern struct XenDevOps xen_framebuffer_ops; /* xen_framebuffer.c */
> > extern struct XenDevOps xen_blkdev_ops; /* xen_disk.c
> */
> > extern struct XenDevOps xen_netdev_ops; /* xen_nic.c
> */
> > +extern struct XenDevOps xen_vtpmdev_ops; /*
> xen_stubdom_vtpm.c*/
> >
> > void xen_init_display(int domid);
> >
> > diff --git a/include/hw/xen/xen_common.h
> b/include/hw/xen/xen_common.h
> > index 95612a4..fb43084 100644
> > --- a/include/hw/xen/xen_common.h
> > +++ b/include/hw/xen/xen_common.h
> > @@ -130,6 +130,12 @@ static inline XenXC xen_xc_interface_open(void
> *logger, void *dombuild_logger,
> > return xc_interface_open(logger, dombuild_logger, open_flags);
> > }
> >
> > +static inline xc_gntshr *xen_xc_gntshr_open(void *logger,
> > + unsigned int open_flags)
> > +{
> > + return xc_gntshr_open(logger, open_flags);
> > +}
> > +
> > /* FIXME There is now way to have the xen fd */
> > static inline int xc_fd(xc_interface *xen_xc)
> > {
> > diff --git a/xen-hvm.c b/xen-hvm.c
> > index 21f1cbb..c99ace8 100644
> > --- a/xen-hvm.c
> > +++ b/xen-hvm.c
> > @@ -1067,6 +1067,11 @@ int xen_hvm_init(ram_addr_t
> *below_4g_mem_size, ram_addr_t *above_4g_mem_size,
> > int i, rc;
> > unsigned long ioreq_pfn;
> > unsigned long bufioreq_evtchn;
> > +
> > +#ifdef CONFIG_TPM_XENSTUBDOMS
> > + unsigned long stubdom_vtpm = 0;
> > +#endif
> > +
> > XenIOState *state;
> >
> > state = g_malloc0(sizeof (XenIOState));
> > @@ -1169,6 +1174,14 @@ int xen_hvm_init(ram_addr_t
> *below_4g_mem_size, ram_addr_t *above_4g_mem_size,
> > fprintf(stderr, "%s: xen backend core setup failed\n",
> __FUNCTION__);
> > return -1;
> > }
> > +
> > +#ifdef CONFIG_TPM_XENSTUBDOMS
> > + xc_get_hvm_param(xen_xc, xen_domid,
> HVM_PARAM_STUBDOM_VTPM, &stubdom_vtpm);
> > + if (stubdom_vtpm) {
> > + xen_vtpm_register(&xen_vtpmdev_ops);
> > + }
> > +#endif
>
> Given that vtpm is just a PV frontend, can't you just detect whether is
> present on xenstore and initialize it based on that? Like all the
> backend below?
Refer to above explanation.
Thanks
Quan
>
>
> > xen_be_register("console", &xen_console_ops);
> > xen_be_register("vkbd", &xen_kbdmouse_ops);
> > xen_be_register("qdisk", &xen_blkdev_ops);
> > --
> > 1.8.3.2
> >
- [Qemu-devel] [PATCH 2/4] Qemu-Xen-vTPM: Register Xen stubdom vTPM frontend driver, Quan Xu, 2014/11/02
- Re: [Qemu-devel] [PATCH 2/4] Qemu-Xen-vTPM: Register Xen stubdom vTPM frontend driver, Stefano Stabellini, 2014/11/03
- Re: [Qemu-devel] [PATCH 2/4] Qemu-Xen-vTPM: Register Xen stubdom vTPM frontend driver, Xu, Quan, 2014/11/04
- Re: [Qemu-devel] [PATCH 2/4] Qemu-Xen-vTPM: Register Xen stubdom vTPM frontend driver, Stefano Stabellini, 2014/11/05
- Re: [Qemu-devel] [PATCH 2/4] Qemu-Xen-vTPM: Register Xen stubdom vTPM frontend driver, Xu, Quan, 2014/11/05
- Re: [Qemu-devel] [PATCH 2/4] Qemu-Xen-vTPM: Register Xen stubdom vTPM frontend driver, Stefano Stabellini, 2014/11/05
- Re: [Qemu-devel] [PATCH 2/4] Qemu-Xen-vTPM: Register Xen stubdom vTPM frontend driver, Xu, Quan, 2014/11/05
Re: [Qemu-devel] [PATCH 2/4] Qemu-Xen-vTPM: Register Xen stubdom vTPM frontend driver,
Xu, Quan <=
Re: [Qemu-devel] [PATCH 2/4] Qemu-Xen-vTPM: Register Xen stubdom vTPM frontend driver, Paolo Bonzini, 2014/11/03