* [Qemu-devel] [PATCH 0/8] xen/9pfs: introduce the Xen 9pfs backend
@ 2017-03-07 2:12 Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 1/8] xen: import ring.h from xen Stefano Stabellini
0 siblings, 1 reply; 13+ messages in thread
From: Stefano Stabellini @ 2017-03-07 2:12 UTC (permalink / raw)
To: qemu-devel
Cc: xen-devel, sstabellini, anthony.perard, jgross, aneesh.kumar,
groug
Hi all,
This patch series implements a new transport for 9pfs, aimed at Xen
systems.
The transport is based on a traditional Xen frontend and backend drivers
pair. This patch series implements the backend, which typically runs in
Dom0. I sent another series to implement the frontend in Linux
(http://marc.info/?l=linux-kernel&m=148883047125960&w=2).
The backend complies to the Xen transport for 9pfs specification
version 1, available here:
http://xenbits.xen.org/gitweb/?p=xen.git;a=blob_plain;f=docs/misc/9pfs.markdown;hb=HEAD
Stefano Stabellini (8):
xen: import ring.h from xen
xen: introduce the header file for the Xen 9pfs transport protocol
xen/9pfs: introduce Xen 9pfs backend
xen/9pfs: connect to the frontend
xen/9pfs: receive requests from the frontend
xen/9pfs: implement in/out_iov_from_pdu and vmarshal/vunmarshal
xen/9pfs: send responses back to the frontend
xen/9pfs: build and register Xen 9pfs backend
hw/9pfs/Makefile.objs | 1 +
hw/9pfs/xen-9p-backend.c | 403 ++++++++++++++++++++++++++++++++++++++
hw/9pfs/xen_9pfs.h | 20 ++
hw/block/xen_blkif.h | 2 +-
hw/usb/xen-usb.c | 2 +-
hw/xen/xen_backend.c | 1 +
include/hw/xen/io/ring.h | 455 +++++++++++++++++++++++++++++++++++++++++++
include/hw/xen/xen_backend.h | 1 +
8 files changed, 883 insertions(+), 2 deletions(-)
create mode 100644 hw/9pfs/xen-9p-backend.c
create mode 100644 hw/9pfs/xen_9pfs.h
create mode 100644 include/hw/xen/io/ring.h
^ permalink raw reply [flat|nested] 13+ messages in thread
* [Qemu-devel] [PATCH 1/8] xen: import ring.h from xen
2017-03-07 2:12 [Qemu-devel] [PATCH 0/8] xen/9pfs: introduce the Xen 9pfs backend Stefano Stabellini
@ 2017-03-07 2:12 ` Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 2/8] xen: introduce the header file for the Xen 9pfs transport protocol Stefano Stabellini
` (6 more replies)
0 siblings, 7 replies; 13+ messages in thread
From: Stefano Stabellini @ 2017-03-07 2:12 UTC (permalink / raw)
To: qemu-devel
Cc: xen-devel, sstabellini, Stefano Stabellini, anthony.perard,
jgross
Do not use the ring.h header installed on the system. Instead, import
the header into the QEMU codebase. This avoids problems when QEMU is
built against a Xen version too old to provide all the ring macros.
Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
CC: anthony.perard@citrix.com
CC: jgross@suse.com
---
NB: The new macros have not been committed to Xen yet. Do not apply this
patch until they do.
---
---
hw/block/xen_blkif.h | 2 +-
hw/usb/xen-usb.c | 2 +-
include/hw/xen/io/ring.h | 455 +++++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 457 insertions(+), 2 deletions(-)
create mode 100644 include/hw/xen/io/ring.h
diff --git a/hw/block/xen_blkif.h b/hw/block/xen_blkif.h
index 3300b6f..3e6e1ea 100644
--- a/hw/block/xen_blkif.h
+++ b/hw/block/xen_blkif.h
@@ -1,7 +1,7 @@
#ifndef XEN_BLKIF_H
#define XEN_BLKIF_H
-#include <xen/io/ring.h>
+#include "hw/xen/io/ring.h"
#include <xen/io/blkif.h>
#include <xen/io/protocols.h>
diff --git a/hw/usb/xen-usb.c b/hw/usb/xen-usb.c
index 8e676e6..370b3d9 100644
--- a/hw/usb/xen-usb.c
+++ b/hw/usb/xen-usb.c
@@ -33,7 +33,7 @@
#include "qapi/qmp/qint.h"
#include "qapi/qmp/qstring.h"
-#include <xen/io/ring.h>
+#include "hw/xen/io/ring.h"
#include <xen/io/usbif.h>
/*
diff --git a/include/hw/xen/io/ring.h b/include/hw/xen/io/ring.h
new file mode 100644
index 0000000..cf01fc3
--- /dev/null
+++ b/include/hw/xen/io/ring.h
@@ -0,0 +1,455 @@
+/******************************************************************************
+ * ring.h
+ *
+ * Shared producer-consumer ring macros.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Tim Deegan and Andrew Warfield November 2004.
+ */
+
+#ifndef __XEN_PUBLIC_IO_RING_H__
+#define __XEN_PUBLIC_IO_RING_H__
+
+#if __XEN_INTERFACE_VERSION__ < 0x00030208
+#define xen_mb() mb()
+#define xen_rmb() rmb()
+#define xen_wmb() wmb()
+#endif
+
+typedef unsigned int RING_IDX;
+
+/* Round a 32-bit unsigned constant down to the nearest power of two. */
+#define __RD2(_x) (((_x) & 0x00000002) ? 0x2 : ((_x) & 0x1))
+#define __RD4(_x) (((_x) & 0x0000000c) ? __RD2((_x)>>2)<<2 : __RD2(_x))
+#define __RD8(_x) (((_x) & 0x000000f0) ? __RD4((_x)>>4)<<4 : __RD4(_x))
+#define __RD16(_x) (((_x) & 0x0000ff00) ? __RD8((_x)>>8)<<8 : __RD8(_x))
+#define __RD32(_x) (((_x) & 0xffff0000) ? __RD16((_x)>>16)<<16 : __RD16(_x))
+
+/*
+ * Calculate size of a shared ring, given the total available space for the
+ * ring and indexes (_sz), and the name tag of the request/response structure.
+ * A ring contains as many entries as will fit, rounded down to the nearest
+ * power of two (so we can mask with (size-1) to loop around).
+ */
+#define __CONST_RING_SIZE(_s, _sz) \
+ (__RD32(((_sz) - offsetof(struct _s##_sring, ring)) / \
+ sizeof(((struct _s##_sring *)0)->ring[0])))
+/*
+ * The same for passing in an actual pointer instead of a name tag.
+ */
+#define __RING_SIZE(_s, _sz) \
+ (__RD32(((_sz) - (long)(_s)->ring + (long)(_s)) / sizeof((_s)->ring[0])))
+
+/*
+ * Macros to make the correct C datatypes for a new kind of ring.
+ *
+ * To make a new ring datatype, you need to have two message structures,
+ * let's say request_t, and response_t already defined.
+ *
+ * In a header where you want the ring datatype declared, you then do:
+ *
+ * DEFINE_RING_TYPES(mytag, request_t, response_t);
+ *
+ * These expand out to give you a set of types, as you can see below.
+ * The most important of these are:
+ *
+ * mytag_sring_t - The shared ring.
+ * mytag_front_ring_t - The 'front' half of the ring.
+ * mytag_back_ring_t - The 'back' half of the ring.
+ *
+ * To initialize a ring in your code you need to know the location and size
+ * of the shared memory area (PAGE_SIZE, for instance). To initialise
+ * the front half:
+ *
+ * mytag_front_ring_t front_ring;
+ * SHARED_RING_INIT((mytag_sring_t *)shared_page);
+ * FRONT_RING_INIT(&front_ring, (mytag_sring_t *)shared_page, PAGE_SIZE);
+ *
+ * Initializing the back follows similarly (note that only the front
+ * initializes the shared ring):
+ *
+ * mytag_back_ring_t back_ring;
+ * BACK_RING_INIT(&back_ring, (mytag_sring_t *)shared_page, PAGE_SIZE);
+ */
+
+#define DEFINE_RING_TYPES(__name, __req_t, __rsp_t) \
+ \
+/* Shared ring entry */ \
+union __name##_sring_entry { \
+ __req_t req; \
+ __rsp_t rsp; \
+}; \
+ \
+/* Shared ring page */ \
+struct __name##_sring { \
+ RING_IDX req_prod, req_event; \
+ RING_IDX rsp_prod, rsp_event; \
+ union { \
+ struct { \
+ uint8_t smartpoll_active; \
+ } netif; \
+ struct { \
+ uint8_t msg; \
+ } tapif_user; \
+ uint8_t pvt_pad[4]; \
+ } pvt; \
+ uint8_t __pad[44]; \
+ union __name##_sring_entry ring[1]; /* variable-length */ \
+}; \
+ \
+/* "Front" end's private variables */ \
+struct __name##_front_ring { \
+ RING_IDX req_prod_pvt; \
+ RING_IDX rsp_cons; \
+ unsigned int nr_ents; \
+ struct __name##_sring *sring; \
+}; \
+ \
+/* "Back" end's private variables */ \
+struct __name##_back_ring { \
+ RING_IDX rsp_prod_pvt; \
+ RING_IDX req_cons; \
+ unsigned int nr_ents; \
+ struct __name##_sring *sring; \
+}; \
+ \
+/* Syntactic sugar */ \
+typedef struct __name##_sring __name##_sring_t; \
+typedef struct __name##_front_ring __name##_front_ring_t; \
+typedef struct __name##_back_ring __name##_back_ring_t
+
+/*
+ * Macros for manipulating rings.
+ *
+ * FRONT_RING_whatever works on the "front end" of a ring: here
+ * requests are pushed on to the ring and responses taken off it.
+ *
+ * BACK_RING_whatever works on the "back end" of a ring: here
+ * requests are taken off the ring and responses put on.
+ *
+ * N.B. these macros do NO INTERLOCKS OR FLOW CONTROL.
+ * This is OK in 1-for-1 request-response situations where the
+ * requestor (front end) never has more than RING_SIZE()-1
+ * outstanding requests.
+ */
+
+/* Initialising empty rings */
+#define SHARED_RING_INIT(_s) do { \
+ (_s)->req_prod = (_s)->rsp_prod = 0; \
+ (_s)->req_event = (_s)->rsp_event = 1; \
+ (void)memset((_s)->pvt.pvt_pad, 0, sizeof((_s)->pvt.pvt_pad)); \
+ (void)memset((_s)->__pad, 0, sizeof((_s)->__pad)); \
+} while(0)
+
+#define FRONT_RING_INIT(_r, _s, __size) do { \
+ (_r)->req_prod_pvt = 0; \
+ (_r)->rsp_cons = 0; \
+ (_r)->nr_ents = __RING_SIZE(_s, __size); \
+ (_r)->sring = (_s); \
+} while (0)
+
+#define BACK_RING_INIT(_r, _s, __size) do { \
+ (_r)->rsp_prod_pvt = 0; \
+ (_r)->req_cons = 0; \
+ (_r)->nr_ents = __RING_SIZE(_s, __size); \
+ (_r)->sring = (_s); \
+} while (0)
+
+/* How big is this ring? */
+#define RING_SIZE(_r) \
+ ((_r)->nr_ents)
+
+/* Number of free requests (for use on front side only). */
+#define RING_FREE_REQUESTS(_r) \
+ (RING_SIZE(_r) - ((_r)->req_prod_pvt - (_r)->rsp_cons))
+
+/* Test if there is an empty slot available on the front ring.
+ * (This is only meaningful from the front. )
+ */
+#define RING_FULL(_r) \
+ (RING_FREE_REQUESTS(_r) == 0)
+
+/* Test if there are outstanding messages to be processed on a ring. */
+#define RING_HAS_UNCONSUMED_RESPONSES(_r) \
+ ((_r)->sring->rsp_prod - (_r)->rsp_cons)
+
+#ifdef __GNUC__
+#define RING_HAS_UNCONSUMED_REQUESTS(_r) ({ \
+ unsigned int req = (_r)->sring->req_prod - (_r)->req_cons; \
+ unsigned int rsp = RING_SIZE(_r) - \
+ ((_r)->req_cons - (_r)->rsp_prod_pvt); \
+ req < rsp ? req : rsp; \
+})
+#else
+/* Same as above, but without the nice GCC ({ ... }) syntax. */
+#define RING_HAS_UNCONSUMED_REQUESTS(_r) \
+ ((((_r)->sring->req_prod - (_r)->req_cons) < \
+ (RING_SIZE(_r) - ((_r)->req_cons - (_r)->rsp_prod_pvt))) ? \
+ ((_r)->sring->req_prod - (_r)->req_cons) : \
+ (RING_SIZE(_r) - ((_r)->req_cons - (_r)->rsp_prod_pvt)))
+#endif
+
+/* Direct access to individual ring elements, by index. */
+#define RING_GET_REQUEST(_r, _idx) \
+ (&((_r)->sring->ring[((_idx) & (RING_SIZE(_r) - 1))].req))
+
+/*
+ * Get a local copy of a request.
+ *
+ * Use this in preference to RING_GET_REQUEST() so all processing is
+ * done on a local copy that cannot be modified by the other end.
+ *
+ * Note that https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58145 may cause this
+ * to be ineffective where _req is a struct which consists of only bitfields.
+ */
+#define RING_COPY_REQUEST(_r, _idx, _req) do { \
+ /* Use volatile to force the copy into _req. */ \
+ *(_req) = *(volatile typeof(_req))RING_GET_REQUEST(_r, _idx); \
+} while (0)
+
+#define RING_GET_RESPONSE(_r, _idx) \
+ (&((_r)->sring->ring[((_idx) & (RING_SIZE(_r) - 1))].rsp))
+
+/* Loop termination condition: Would the specified index overflow the ring? */
+#define RING_REQUEST_CONS_OVERFLOW(_r, _cons) \
+ (((_cons) - (_r)->rsp_prod_pvt) >= RING_SIZE(_r))
+
+/* Ill-behaved frontend determination: Can there be this many requests? */
+#define RING_REQUEST_PROD_OVERFLOW(_r, _prod) \
+ (((_prod) - (_r)->rsp_prod_pvt) > RING_SIZE(_r))
+
+#define RING_PUSH_REQUESTS(_r) do { \
+ xen_wmb(); /* back sees requests /before/ updated producer index */ \
+ (_r)->sring->req_prod = (_r)->req_prod_pvt; \
+} while (0)
+
+#define RING_PUSH_RESPONSES(_r) do { \
+ xen_wmb(); /* front sees resps /before/ updated producer index */ \
+ (_r)->sring->rsp_prod = (_r)->rsp_prod_pvt; \
+} while (0)
+
+/*
+ * Notification hold-off (req_event and rsp_event):
+ *
+ * When queueing requests or responses on a shared ring, it may not always be
+ * necessary to notify the remote end. For example, if requests are in flight
+ * in a backend, the front may be able to queue further requests without
+ * notifying the back (if the back checks for new requests when it queues
+ * responses).
+ *
+ * When enqueuing requests or responses:
+ *
+ * Use RING_PUSH_{REQUESTS,RESPONSES}_AND_CHECK_NOTIFY(). The second argument
+ * is a boolean return value. True indicates that the receiver requires an
+ * asynchronous notification.
+ *
+ * After dequeuing requests or responses (before sleeping the connection):
+ *
+ * Use RING_FINAL_CHECK_FOR_REQUESTS() or RING_FINAL_CHECK_FOR_RESPONSES().
+ * The second argument is a boolean return value. True indicates that there
+ * are pending messages on the ring (i.e., the connection should not be put
+ * to sleep).
+ *
+ * These macros will set the req_event/rsp_event field to trigger a
+ * notification on the very next message that is enqueued. If you want to
+ * create batches of work (i.e., only receive a notification after several
+ * messages have been enqueued) then you will need to create a customised
+ * version of the FINAL_CHECK macro in your own code, which sets the event
+ * field appropriately.
+ */
+
+#define RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(_r, _notify) do { \
+ RING_IDX __old = (_r)->sring->req_prod; \
+ RING_IDX __new = (_r)->req_prod_pvt; \
+ xen_wmb(); /* back sees requests /before/ updated producer index */ \
+ (_r)->sring->req_prod = __new; \
+ xen_mb(); /* back sees new requests /before/ we check req_event */ \
+ (_notify) = ((RING_IDX)(__new - (_r)->sring->req_event) < \
+ (RING_IDX)(__new - __old)); \
+} while (0)
+
+#define RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(_r, _notify) do { \
+ RING_IDX __old = (_r)->sring->rsp_prod; \
+ RING_IDX __new = (_r)->rsp_prod_pvt; \
+ xen_wmb(); /* front sees resps /before/ updated producer index */ \
+ (_r)->sring->rsp_prod = __new; \
+ xen_mb(); /* front sees new resps /before/ we check rsp_event */ \
+ (_notify) = ((RING_IDX)(__new - (_r)->sring->rsp_event) < \
+ (RING_IDX)(__new - __old)); \
+} while (0)
+
+#define RING_FINAL_CHECK_FOR_REQUESTS(_r, _work_to_do) do { \
+ (_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \
+ if (_work_to_do) break; \
+ (_r)->sring->req_event = (_r)->req_cons + 1; \
+ xen_mb(); \
+ (_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \
+} while (0)
+
+#define RING_FINAL_CHECK_FOR_RESPONSES(_r, _work_to_do) do { \
+ (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \
+ if (_work_to_do) break; \
+ (_r)->sring->rsp_event = (_r)->rsp_cons + 1; \
+ xen_mb(); \
+ (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \
+} while (0)
+
+
+/*
+ * DEFINE_XEN_FLEX_RING_AND_INTF defines two monodirectional rings and
+ * functions to check if there is data on the ring, and to read and
+ * write to them.
+ *
+ * DEFINE_XEN_FLEX_RING is similar to DEFINE_XEN_FLEX_RING_AND_INTF, but
+ * does not define the indexes page. As different protocols can have
+ * extensions to the basic format, this macro allow them to define their
+ * own struct.
+ *
+ * XEN_FLEX_RING_SIZE
+ * Convenience macro to calculate the size of one of the two rings
+ * from the overall order.
+ *
+ * $NAME_mask
+ * Function to apply the size mask to an index, to reduce the index
+ * within the range [0-size].
+ *
+ * $NAME_read_packet
+ * Function to read data from the ring. The amount of data to read is
+ * specified by the "size" argument.
+ *
+ * $NAME_write_packet
+ * Function to write data to the ring. The amount of data to write is
+ * specified by the "size" argument.
+ *
+ * $NAME_get_ring_ptr
+ * Convenience function that returns a pointer to read/write to the
+ * ring at the right location.
+ *
+ * $NAME_data_intf
+ * Indexes page, shared between frontend and backend. It also
+ * contains the array of grant refs.
+ *
+ * $NAME_queued
+ * Function to calculate how many bytes are currently on the ring,
+ * ready to be read. It can also be used to calculate how much free
+ * space is currently on the ring (ring_size - $NAME_queued()).
+ */
+#define XEN_FLEX_RING_SIZE(order) \
+ (1UL << (order + PAGE_SHIFT - 1))
+
+#define DEFINE_XEN_FLEX_RING_AND_INTF(name) \
+struct name##_data_intf { \
+ RING_IDX in_cons, in_prod; \
+ \
+ uint8_t pad1[56]; \
+ \
+ RING_IDX out_cons, out_prod; \
+ \
+ uint8_t pad2[56]; \
+ \
+ RING_IDX ring_order; \
+ grant_ref_t ref[]; \
+}; \
+DEFINE_XEN_FLEX_RING(name);
+
+#define DEFINE_XEN_FLEX_RING(name) \
+static inline RING_IDX name##_mask(RING_IDX idx, RING_IDX ring_size) \
+{ \
+ return (idx & (ring_size - 1)); \
+} \
+ \
+static inline RING_IDX name##_mask_order(RING_IDX idx, RING_IDX ring_order) \
+{ \
+ return (idx & (XEN_FLEX_RING_SIZE(ring_order) - 1)); \
+} \
+ \
+static inline unsigned char* name##_get_ring_ptr(unsigned char *buf, \
+ RING_IDX idx, \
+ RING_IDX ring_order) \
+{ \
+ return buf + name##_mask_order(idx, ring_order); \
+} \
+ \
+static inline void name##_read_packet(const unsigned char *buf, \
+ RING_IDX masked_prod, RING_IDX *masked_cons, \
+ RING_IDX ring_size, void *opaque, size_t size) { \
+ if (*masked_cons < masked_prod || \
+ size <= ring_size - *masked_cons) { \
+ memcpy(opaque, buf + *masked_cons, size); \
+ } else { \
+ memcpy(opaque, buf + *masked_cons, ring_size - *masked_cons); \
+ memcpy((unsigned char *)opaque + ring_size - *masked_cons, buf, \
+ size - (ring_size - *masked_cons)); \
+ } \
+ *masked_cons = name##_mask(*masked_cons + size, ring_size); \
+} \
+ \
+static inline void name##_write_packet(unsigned char *buf, \
+ RING_IDX *masked_prod, RING_IDX masked_cons, \
+ RING_IDX ring_size, const void *opaque, size_t size) { \
+ if (*masked_prod < masked_cons || \
+ size <= ring_size - *masked_prod) { \
+ memcpy(buf + *masked_prod, opaque, size); \
+ } else { \
+ memcpy(buf + *masked_prod, opaque, ring_size - *masked_prod); \
+ memcpy(buf, (unsigned char *)opaque + (ring_size - *masked_prod), \
+ size - (ring_size - *masked_prod)); \
+ } \
+ *masked_prod = name##_mask(*masked_prod + size, ring_size); \
+} \
+ \
+struct name##_data { \
+ unsigned char *in; /* half of the allocation */ \
+ unsigned char *out; /* half of the allocation */ \
+}; \
+ \
+ \
+static inline RING_IDX name##_queued(RING_IDX prod, \
+ RING_IDX cons, RING_IDX ring_size) \
+{ \
+ RING_IDX size; \
+ \
+ if (prod == cons) \
+ return 0; \
+ \
+ prod = name##_mask(prod, ring_size); \
+ cons = name##_mask(cons, ring_size); \
+ \
+ if (prod == cons) \
+ return ring_size; \
+ \
+ if (prod > cons) \
+ size = prod - cons; \
+ else \
+ size = ring_size - (cons - prod); \
+ return size; \
+};
+
+#endif /* __XEN_PUBLIC_IO_RING_H__ */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
--
1.9.1
^ permalink raw reply related [flat|nested] 13+ messages in thread
* [Qemu-devel] [PATCH 2/8] xen: introduce the header file for the Xen 9pfs transport protocol
2017-03-07 2:12 ` [Qemu-devel] [PATCH 1/8] xen: import ring.h from xen Stefano Stabellini
@ 2017-03-07 2:12 ` Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 3/8] xen/9pfs: introduce Xen 9pfs backend Stefano Stabellini
` (5 subsequent siblings)
6 siblings, 0 replies; 13+ messages in thread
From: Stefano Stabellini @ 2017-03-07 2:12 UTC (permalink / raw)
To: qemu-devel
Cc: xen-devel, sstabellini, Stefano Stabellini, anthony.perard,
jgross
It uses the new ring.h macros to declare rings and interfaces.
Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
CC: anthony.perard@citrix.com
CC: jgross@suse.com
---
hw/9pfs/xen_9pfs.h | 20 ++++++++++++++++++++
1 file changed, 20 insertions(+)
create mode 100644 hw/9pfs/xen_9pfs.h
diff --git a/hw/9pfs/xen_9pfs.h b/hw/9pfs/xen_9pfs.h
new file mode 100644
index 0000000..c4e8901
--- /dev/null
+++ b/hw/9pfs/xen_9pfs.h
@@ -0,0 +1,20 @@
+#ifndef XEN_9PFS_H
+#define XEN_9PFS_H
+
+#include "hw/xen/io/ring.h"
+#include <xen/io/protocols.h>
+
+struct xen_9pfs_header {
+ uint32_t size;
+ uint8_t id;
+ uint16_t tag;
+
+ /* uint8_t sdata[]; */
+} __attribute__((packed));
+
+#define PAGE_SHIFT XC_PAGE_SHIFT
+#define XEN_9PFS_RING_ORDER 6
+#define XEN_9PFS_RING_SIZE XEN_FLEX_RING_SIZE(XEN_9PFS_RING_ORDER)
+DEFINE_XEN_FLEX_RING_AND_INTF(xen_9pfs);
+
+#endif
--
1.9.1
^ permalink raw reply related [flat|nested] 13+ messages in thread
* [Qemu-devel] [PATCH 3/8] xen/9pfs: introduce Xen 9pfs backend
2017-03-07 2:12 ` [Qemu-devel] [PATCH 1/8] xen: import ring.h from xen Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 2/8] xen: introduce the header file for the Xen 9pfs transport protocol Stefano Stabellini
@ 2017-03-07 2:12 ` Stefano Stabellini
2017-03-09 17:54 ` Greg Kurz
2017-03-07 2:12 ` [Qemu-devel] [PATCH 4/8] xen/9pfs: connect to the frontend Stefano Stabellini
` (4 subsequent siblings)
6 siblings, 1 reply; 13+ messages in thread
From: Stefano Stabellini @ 2017-03-07 2:12 UTC (permalink / raw)
To: qemu-devel
Cc: xen-devel, sstabellini, Stefano Stabellini, anthony.perard,
jgross, Aneesh Kumar K.V, Greg Kurz
Introduce the Xen 9pfs backend: add struct XenDevOps to register as a
Xen backend and add struct V9fsTransport to register as v9fs transport.
All functions are empty stubs for now.
Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
CC: anthony.perard@citrix.com
CC: jgross@suse.com
CC: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
CC: Greg Kurz <groug@kaod.org>
---
hw/9pfs/xen-9p-backend.c | 96 ++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 96 insertions(+)
create mode 100644 hw/9pfs/xen-9p-backend.c
diff --git a/hw/9pfs/xen-9p-backend.c b/hw/9pfs/xen-9p-backend.c
new file mode 100644
index 0000000..924fb64
--- /dev/null
+++ b/hw/9pfs/xen-9p-backend.c
@@ -0,0 +1,96 @@
+/*
+ * Xen 9p backend
+ *
+ * Copyright Aporeto 2017
+ *
+ * Authors:
+ * Stefano Stabellini <stefano@aporeto.com>
+ *
+ */
+
+#include "qemu/osdep.h"
+
+#include "hw/hw.h"
+#include "hw/9pfs/9p.h"
+#include "hw/xen/xen_backend.h"
+#include "xen_9pfs.h"
+#include "qemu/config-file.h"
+#include "fsdev/qemu-fsdev.h"
+
+struct Xen9pfsDev {
+ struct XenDevice xendev; /* must be first */
+};
+
+static ssize_t xen_9pfs_pdu_vmarshal(V9fsPDU *pdu,
+ size_t offset,
+ const char *fmt,
+ va_list ap)
+{
+ return 0;
+}
+
+static ssize_t xen_9pfs_pdu_vunmarshal(V9fsPDU *pdu,
+ size_t offset,
+ const char *fmt,
+ va_list ap)
+{
+ return 0;
+}
+
+static void xen_9pfs_init_out_iov_from_pdu(V9fsPDU *pdu,
+ struct iovec **piov,
+ unsigned int *pniov)
+{
+}
+
+static void xen_9pfs_init_in_iov_from_pdu(V9fsPDU *pdu,
+ struct iovec **piov,
+ unsigned int *pniov,
+ size_t size)
+{
+}
+
+static void xen_9pfs_push_and_notify(V9fsPDU *pdu)
+{
+}
+
+static const struct V9fsTransport xen_9p_transport = {
+ .pdu_vmarshal = xen_9pfs_pdu_vmarshal,
+ .pdu_vunmarshal = xen_9pfs_pdu_vunmarshal,
+ .init_in_iov_from_pdu = xen_9pfs_init_in_iov_from_pdu,
+ .init_out_iov_from_pdu = xen_9pfs_init_out_iov_from_pdu,
+ .push_and_notify = xen_9pfs_push_and_notify,
+};
+
+static int xen_9pfs_init(struct XenDevice *xendev)
+{
+ return 0;
+}
+
+static int xen_9pfs_free(struct XenDevice *xendev)
+{
+ return -1;
+}
+
+static int xen_9pfs_connect(struct XenDevice *xendev)
+{
+ return 0;
+}
+
+static void xen_9pfs_alloc(struct XenDevice *xendev)
+{
+}
+
+static void xen_9pfs_disconnect(struct XenDevice *xendev)
+{
+}
+
+struct XenDevOps xen_9pfs_ops = {
+ .size = sizeof(struct Xen9pfsDev),
+ .flags = DEVOPS_FLAG_NEED_GNTDEV,
+ .alloc = xen_9pfs_alloc,
+ .init = xen_9pfs_init,
+ .initialise = xen_9pfs_connect,
+ .disconnect = xen_9pfs_disconnect,
+ .free = xen_9pfs_free,
+};
--
1.9.1
^ permalink raw reply related [flat|nested] 13+ messages in thread
* [Qemu-devel] [PATCH 4/8] xen/9pfs: connect to the frontend
2017-03-07 2:12 ` [Qemu-devel] [PATCH 1/8] xen: import ring.h from xen Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 2/8] xen: introduce the header file for the Xen 9pfs transport protocol Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 3/8] xen/9pfs: introduce Xen 9pfs backend Stefano Stabellini
@ 2017-03-07 2:12 ` Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 5/8] xen/9pfs: receive requests from " Stefano Stabellini
` (3 subsequent siblings)
6 siblings, 0 replies; 13+ messages in thread
From: Stefano Stabellini @ 2017-03-07 2:12 UTC (permalink / raw)
To: qemu-devel
Cc: xen-devel, sstabellini, Stefano Stabellini, anthony.perard,
jgross, Aneesh Kumar K.V, Greg Kurz
Write the limits of the backend to xenstore. Connect to the frontend.
Upon connection, allocate the rings according to the protocol
specification.
Initialize a QEMUBH to schedule work upon receiving an event channel
notification from the frontend.
Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
CC: anthony.perard@citrix.com
CC: jgross@suse.com
CC: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
CC: Greg Kurz <groug@kaod.org>
---
hw/9pfs/xen-9p-backend.c | 159 ++++++++++++++++++++++++++++++++++++++++++++++-
1 file changed, 158 insertions(+), 1 deletion(-)
diff --git a/hw/9pfs/xen-9p-backend.c b/hw/9pfs/xen-9p-backend.c
index 924fb64..59e89fd 100644
--- a/hw/9pfs/xen-9p-backend.c
+++ b/hw/9pfs/xen-9p-backend.c
@@ -17,8 +17,35 @@
#include "qemu/config-file.h"
#include "fsdev/qemu-fsdev.h"
+#define VERSIONS "1"
+#define MAX_RINGS 8
+#define MAX_RING_ORDER 8
+
+struct Xen9pfsRing {
+ struct Xen9pfsDev *priv;
+
+ int ref;
+ xenevtchn_handle *evtchndev;
+ int evtchn;
+ int local_port;
+ struct xen_9pfs_data_intf *intf;
+ unsigned char *data;
+ struct xen_9pfs_data ring;
+
+ QEMUBH *bh;
+
+ /* local copies, so that we can read/write PDU data directly from
+ * the ring */
+ RING_IDX out_cons, out_size, in_cons;
+ bool inprogress;
+};
+
struct Xen9pfsDev {
struct XenDevice xendev; /* must be first */
+ V9fsState state;
+
+ int num_rings;
+ struct Xen9pfsRing *rings;
};
static ssize_t xen_9pfs_pdu_vmarshal(V9fsPDU *pdu,
@@ -67,22 +94,152 @@ static int xen_9pfs_init(struct XenDevice *xendev)
return 0;
}
+static void xen_9pfs_bh(void *opaque)
+{
+}
+
+static void xen_9pfs_evtchn_event(void *opaque)
+{
+}
+
static int xen_9pfs_free(struct XenDevice *xendev)
{
- return -1;
+ int i;
+ struct Xen9pfsDev *xen_9pdev = container_of(xendev, struct Xen9pfsDev, xendev);
+
+ for (i = 0; i < xen_9pdev->num_rings; i++) {
+ if (xen_9pdev->rings[i].data != NULL) {
+ xengnttab_unmap(xen_9pdev->xendev.gnttabdev,
+ xen_9pdev->rings[i].data,
+ (1 << XEN_9PFS_RING_ORDER));
+ }
+ if (xen_9pdev->rings[i].intf != NULL) {
+ xengnttab_unmap(xen_9pdev->xendev.gnttabdev,
+ xen_9pdev->rings[i].intf,
+ 1);
+ }
+ if (xen_9pdev->rings[i].evtchndev > 0) {
+ qemu_set_fd_handler(xenevtchn_fd(xen_9pdev->rings[i].evtchndev),
+ NULL, NULL, NULL);
+ xenevtchn_unbind(xen_9pdev->rings[i].evtchndev, xen_9pdev->rings[i].local_port);
+ }
+ if (xen_9pdev->rings[i].bh != NULL) {
+ qemu_bh_delete(xen_9pdev->rings[i].bh);
+ }
+ }
+ g_free(xen_9pdev->rings);
+ return 0;
}
static int xen_9pfs_connect(struct XenDevice *xendev)
{
+ int i;
+ struct Xen9pfsDev *xen_9pdev = container_of(xendev, struct Xen9pfsDev, xendev);
+ V9fsState *s = &xen_9pdev->state;
+ QemuOpts *fsdev;
+ char *security_model, *path;
+
+ if (xenstore_read_fe_int(&xen_9pdev->xendev, "num-rings",
+ &xen_9pdev->num_rings) == -1 ||
+ xen_9pdev->num_rings > MAX_RINGS) {
+ return -1;
+ }
+
+ xen_9pdev->rings = g_malloc0(xen_9pdev->num_rings * sizeof(struct Xen9pfsRing));
+ for (i = 0; i < xen_9pdev->num_rings; i++) {
+ char str[16];
+
+ xen_9pdev->rings[i].priv = xen_9pdev;
+ xen_9pdev->rings[i].evtchn = -1;
+ xen_9pdev->rings[i].local_port = -1;
+
+ sprintf(str, "ring-ref%u", i);
+ if (xenstore_read_fe_int(&xen_9pdev->xendev, str,
+ &xen_9pdev->rings[i].ref) == -1) {
+ goto out;
+ }
+ sprintf(str, "event-channel-%u", i);
+ if (xenstore_read_fe_int(&xen_9pdev->xendev, str,
+ &xen_9pdev->rings[i].evtchn) == -1) {
+ goto out;
+ }
+
+ xen_9pdev->rings[i].intf = xengnttab_map_grant_ref(
+ xen_9pdev->xendev.gnttabdev,
+ xen_9pdev->xendev.dom,
+ xen_9pdev->rings[i].ref,
+ PROT_READ | PROT_WRITE);
+ if (!xen_9pdev->rings[i].intf) {
+ goto out;
+ }
+ xen_9pdev->rings[i].data = xengnttab_map_domain_grant_refs(
+ xen_9pdev->xendev.gnttabdev,
+ (1 << XEN_9PFS_RING_ORDER),
+ xen_9pdev->xendev.dom,
+ xen_9pdev->rings[i].intf->ref,
+ PROT_READ | PROT_WRITE);
+ if (!xen_9pdev->rings[i].data) {
+ goto out;
+ }
+ xen_9pdev->rings[i].ring.in = xen_9pdev->rings[i].data;
+ xen_9pdev->rings[i].ring.out = xen_9pdev->rings[i].data + XEN_9PFS_RING_SIZE;
+
+ xen_9pdev->rings[i].bh = qemu_bh_new(xen_9pfs_bh, &xen_9pdev->rings[i]);
+ xen_9pdev->rings[i].out_cons = 0;
+ xen_9pdev->rings[i].out_size = 0;
+ xen_9pdev->rings[i].inprogress = false;
+
+
+ xen_9pdev->rings[i].evtchndev = xenevtchn_open(NULL, 0);
+ if (xen_9pdev->rings[i].evtchndev == NULL) {
+ goto out;
+ }
+ fcntl(xenevtchn_fd(xen_9pdev->rings[i].evtchndev), F_SETFD, FD_CLOEXEC);
+ xen_9pdev->rings[i].local_port = xenevtchn_bind_interdomain
+ (xen_9pdev->rings[i].evtchndev, xendev->dom, xen_9pdev->rings[i].evtchn);
+ if (xen_9pdev->rings[i].local_port == -1) {
+ xen_pv_printf(xendev, 0, "xenevtchn_bind_interdomain failed port=%d\n",
+ xen_9pdev->rings[i].evtchn);
+ goto out;
+ }
+ xen_pv_printf(xendev, 2, "bind evtchn port %d\n", xendev->local_port);
+ qemu_set_fd_handler(xenevtchn_fd(xen_9pdev->rings[i].evtchndev),
+ xen_9pfs_evtchn_event, NULL, &xen_9pdev->rings[i]);
+ }
+
+ security_model = xenstore_read_be_str(xendev, "security_model");
+ path = xenstore_read_be_str(xendev, "path");
+ s->fsconf.fsdev_id = g_malloc(strlen("xen9p") + 6);
+ sprintf(s->fsconf.fsdev_id, "xen9p%d", xendev->dev);
+ s->fsconf.tag = xenstore_read_fe_str(xendev, "tag");
+ v9fs_register_transport(s, &xen_9p_transport);
+ fsdev = qemu_opts_create(qemu_find_opts("fsdev"),
+ s->fsconf.tag,
+ 1, NULL);
+ qemu_opt_set(fsdev, "fsdriver", "local", NULL);
+ qemu_opt_set(fsdev, "path", path, NULL);
+ qemu_opt_set(fsdev, "security_model", security_model, NULL);
+ qemu_opts_set_id(fsdev, s->fsconf.fsdev_id);
+ qemu_fsdev_add(fsdev);
+ v9fs_device_realize_common(s, NULL);
+
return 0;
+
+out:
+ xen_9pfs_free(xendev);
+ return -1;
}
static void xen_9pfs_alloc(struct XenDevice *xendev)
{
+ xenstore_write_be_str(xendev, "versions", VERSIONS);
+ xenstore_write_be_int(xendev, "max-rings", MAX_RINGS);
+ xenstore_write_be_int(xendev, "max-ring-page-order", MAX_RING_ORDER);
}
static void xen_9pfs_disconnect(struct XenDevice *xendev)
{
+ /* Dynamic hotplug of PV filesystems at runtime is not supported. */
}
struct XenDevOps xen_9pfs_ops = {
--
1.9.1
^ permalink raw reply related [flat|nested] 13+ messages in thread
* [Qemu-devel] [PATCH 5/8] xen/9pfs: receive requests from the frontend
2017-03-07 2:12 ` [Qemu-devel] [PATCH 1/8] xen: import ring.h from xen Stefano Stabellini
` (2 preceding siblings ...)
2017-03-07 2:12 ` [Qemu-devel] [PATCH 4/8] xen/9pfs: connect to the frontend Stefano Stabellini
@ 2017-03-07 2:12 ` Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 6/8] xen/9pfs: implement in/out_iov_from_pdu and vmarshal/vunmarshal Stefano Stabellini
` (2 subsequent siblings)
6 siblings, 0 replies; 13+ messages in thread
From: Stefano Stabellini @ 2017-03-07 2:12 UTC (permalink / raw)
To: qemu-devel
Cc: xen-devel, sstabellini, Stefano Stabellini, anthony.perard,
jgross, Aneesh Kumar K.V, Greg Kurz
Upon receiving an event channel notification from the frontend, schedule
the bottom half. From the bottom half, read one request from the ring,
create a pdu and call pdu_submit to handle it.
For now, only handle one request per ring at a time.
Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
CC: anthony.perard@citrix.com
CC: jgross@suse.com
CC: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
CC: Greg Kurz <groug@kaod.org>
---
hw/9pfs/xen-9p-backend.c | 47 +++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 47 insertions(+)
diff --git a/hw/9pfs/xen-9p-backend.c b/hw/9pfs/xen-9p-backend.c
index 59e89fd..d4c3d36 100644
--- a/hw/9pfs/xen-9p-backend.c
+++ b/hw/9pfs/xen-9p-backend.c
@@ -94,12 +94,59 @@ static int xen_9pfs_init(struct XenDevice *xendev)
return 0;
}
+static int xen_9pfs_receive(struct Xen9pfsRing *ring)
+{
+ struct xen_9pfs_header h;
+ RING_IDX cons, prod, masked_prod, masked_cons;
+ V9fsPDU *pdu;
+
+ if (ring->inprogress) {
+ return 0;
+ }
+
+ cons = ring->intf->out_cons;
+ prod = ring->intf->out_prod;
+ xen_rmb();
+
+ if (xen_9pfs_queued(prod, cons, XEN_9PFS_RING_SIZE) < sizeof(h)) {
+ return 0;
+ }
+ ring->inprogress = true;
+
+ masked_prod = xen_9pfs_mask(prod, XEN_9PFS_RING_SIZE);
+ masked_cons = xen_9pfs_mask(cons, XEN_9PFS_RING_SIZE);
+
+ xen_9pfs_read_packet(ring->ring.out, masked_prod, &masked_cons,
+ XEN_9PFS_RING_SIZE, (uint8_t*) &h, sizeof(h));
+
+ pdu = pdu_alloc(&ring->priv->state);
+ pdu->size = h.size;
+ pdu->id = h.id;
+ pdu->tag = h.tag;
+ ring->out_size = h.size;
+ ring->out_cons = cons + h.size;
+
+ qemu_co_queue_init(&pdu->complete);
+ pdu_submit(pdu);
+
+ return 0;
+}
+
static void xen_9pfs_bh(void *opaque)
{
+ struct Xen9pfsRing *ring = opaque;
+ xen_9pfs_receive(ring);
}
static void xen_9pfs_evtchn_event(void *opaque)
{
+ struct Xen9pfsRing *ring = opaque;
+ evtchn_port_t port;
+
+ port = xenevtchn_pending(ring->evtchndev);
+ xenevtchn_unmask(ring->evtchndev, port);
+
+ qemu_bh_schedule(ring->bh);
}
static int xen_9pfs_free(struct XenDevice *xendev)
--
1.9.1
^ permalink raw reply related [flat|nested] 13+ messages in thread
* [Qemu-devel] [PATCH 6/8] xen/9pfs: implement in/out_iov_from_pdu and vmarshal/vunmarshal
2017-03-07 2:12 ` [Qemu-devel] [PATCH 1/8] xen: import ring.h from xen Stefano Stabellini
` (3 preceding siblings ...)
2017-03-07 2:12 ` [Qemu-devel] [PATCH 5/8] xen/9pfs: receive requests from " Stefano Stabellini
@ 2017-03-07 2:12 ` Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 7/8] xen/9pfs: send responses back to the frontend Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 8/8] xen/9pfs: build and register Xen 9pfs backend Stefano Stabellini
6 siblings, 0 replies; 13+ messages in thread
From: Stefano Stabellini @ 2017-03-07 2:12 UTC (permalink / raw)
To: qemu-devel
Cc: xen-devel, sstabellini, Stefano Stabellini, anthony.perard,
jgross, Aneesh Kumar K.V, Greg Kurz
Implement xen_9pfs_init_in/out_iov_from_pdu and
xen_9pfs_pdu_vmarshal/vunmarshall by creating new sg pointing to the
data on the ring.
This is safe as we only handle one request per ring at any given time.
Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
CC: anthony.perard@citrix.com
CC: jgross@suse.com
CC: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
CC: Greg Kurz <groug@kaod.org>
---
hw/9pfs/xen-9p-backend.c | 91 ++++++++++++++++++++++++++++++++++++++++++++++--
1 file changed, 89 insertions(+), 2 deletions(-)
diff --git a/hw/9pfs/xen-9p-backend.c b/hw/9pfs/xen-9p-backend.c
index d4c3d36..0570e07 100644
--- a/hw/9pfs/xen-9p-backend.c
+++ b/hw/9pfs/xen-9p-backend.c
@@ -48,12 +48,77 @@ struct Xen9pfsDev {
struct Xen9pfsRing *rings;
};
+static void xen_9pfs_in_sg(struct Xen9pfsRing *ring,
+ struct iovec *in_sg,
+ int *num,
+ uint32_t idx,
+ uint32_t size)
+{
+ RING_IDX cons, prod, masked_prod, masked_cons;
+
+ cons = ring->intf->in_cons;
+ prod = ring->intf->in_prod;
+ xen_rmb();
+ masked_prod = xen_9pfs_mask(prod, XEN_9PFS_RING_SIZE);
+ masked_cons = xen_9pfs_mask(cons, XEN_9PFS_RING_SIZE);
+
+ if (masked_prod < masked_cons) {
+ in_sg[0].iov_base = ring->ring.in + masked_prod;
+ in_sg[0].iov_len = masked_cons - masked_prod;
+ *num = 1;
+ } else {
+ in_sg[0].iov_base = ring->ring.in + masked_prod;
+ in_sg[0].iov_len = XEN_9PFS_RING_SIZE - masked_prod;
+ in_sg[1].iov_base = ring->ring.in;
+ in_sg[1].iov_len = masked_cons;
+ *num = 2;
+ }
+}
+
+static void xen_9pfs_out_sg(struct Xen9pfsRing *ring,
+ struct iovec *out_sg,
+ int *num,
+ uint32_t idx)
+{
+ RING_IDX cons, prod, masked_prod, masked_cons;
+
+ cons = ring->intf->out_cons;
+ prod = ring->intf->out_prod;
+ xen_rmb();
+ masked_prod = xen_9pfs_mask(prod, XEN_9PFS_RING_SIZE);
+ masked_cons = xen_9pfs_mask(cons, XEN_9PFS_RING_SIZE);
+
+ if (masked_cons < masked_prod) {
+ out_sg[0].iov_base = ring->ring.out + masked_cons;
+ out_sg[0].iov_len = ring->out_size;
+ *num = 1;
+ } else {
+ if (ring->out_size > (XEN_9PFS_RING_SIZE - masked_cons)) {
+ out_sg[0].iov_base = ring->ring.out + masked_cons;
+ out_sg[0].iov_len = XEN_9PFS_RING_SIZE - masked_cons;
+ out_sg[1].iov_base = ring->ring.out;
+ out_sg[1].iov_len = ring->out_size - (XEN_9PFS_RING_SIZE - masked_cons);
+ *num = 2;
+ } else {
+ out_sg[0].iov_base = ring->ring.out + masked_cons;
+ out_sg[0].iov_len = ring->out_size;
+ *num = 1;
+ }
+ }
+}
+
static ssize_t xen_9pfs_pdu_vmarshal(V9fsPDU *pdu,
size_t offset,
const char *fmt,
va_list ap)
{
- return 0;
+ struct Xen9pfsDev *xen_9pfs = container_of(pdu->s, struct Xen9pfsDev, state);
+ struct iovec in_sg[2];
+ int num;
+
+ xen_9pfs_in_sg(&xen_9pfs->rings[pdu->tag % xen_9pfs->num_rings],
+ in_sg, &num, pdu->idx, ROUND_UP(offset + 128, 512));
+ return v9fs_iov_vmarshal(in_sg, num, offset, 0, fmt, ap);
}
static ssize_t xen_9pfs_pdu_vunmarshal(V9fsPDU *pdu,
@@ -61,13 +126,27 @@ static ssize_t xen_9pfs_pdu_vunmarshal(V9fsPDU *pdu,
const char *fmt,
va_list ap)
{
- return 0;
+ struct Xen9pfsDev *xen_9pfs = container_of(pdu->s, struct Xen9pfsDev, state);
+ struct iovec out_sg[2];
+ int num;
+
+ xen_9pfs_out_sg(&xen_9pfs->rings[pdu->tag % xen_9pfs->num_rings],
+ out_sg, &num, pdu->idx);
+ return v9fs_iov_vunmarshal(out_sg, num, offset, 0, fmt, ap);
}
static void xen_9pfs_init_out_iov_from_pdu(V9fsPDU *pdu,
struct iovec **piov,
unsigned int *pniov)
{
+ struct Xen9pfsDev *xen_9pfs = container_of(pdu->s, struct Xen9pfsDev, state);
+ struct Xen9pfsRing *ring = &xen_9pfs->rings[pdu->tag % xen_9pfs->num_rings];
+ struct iovec *sg = g_malloc0(sizeof(*sg)*2);
+ int num;
+
+ xen_9pfs_out_sg(ring, sg, &num, pdu->idx);
+ *piov = sg;
+ *pniov = num;
}
static void xen_9pfs_init_in_iov_from_pdu(V9fsPDU *pdu,
@@ -75,6 +154,14 @@ static void xen_9pfs_init_in_iov_from_pdu(V9fsPDU *pdu,
unsigned int *pniov,
size_t size)
{
+ struct Xen9pfsDev *xen_9pfs = container_of(pdu->s, struct Xen9pfsDev, state);
+ struct Xen9pfsRing *ring = &xen_9pfs->rings[pdu->tag % xen_9pfs->num_rings];
+ struct iovec *sg = g_malloc0(sizeof(*sg)*2);
+ int num;
+
+ xen_9pfs_in_sg(ring, sg, &num, pdu->idx, size);
+ *piov = sg;
+ *pniov = num;
}
static void xen_9pfs_push_and_notify(V9fsPDU *pdu)
--
1.9.1
^ permalink raw reply related [flat|nested] 13+ messages in thread
* [Qemu-devel] [PATCH 7/8] xen/9pfs: send responses back to the frontend
2017-03-07 2:12 ` [Qemu-devel] [PATCH 1/8] xen: import ring.h from xen Stefano Stabellini
` (4 preceding siblings ...)
2017-03-07 2:12 ` [Qemu-devel] [PATCH 6/8] xen/9pfs: implement in/out_iov_from_pdu and vmarshal/vunmarshal Stefano Stabellini
@ 2017-03-07 2:12 ` Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 8/8] xen/9pfs: build and register Xen 9pfs backend Stefano Stabellini
6 siblings, 0 replies; 13+ messages in thread
From: Stefano Stabellini @ 2017-03-07 2:12 UTC (permalink / raw)
To: qemu-devel
Cc: xen-devel, sstabellini, Stefano Stabellini, anthony.perard,
jgross, Aneesh Kumar K.V, Greg Kurz
Once a request is completed, xen_9pfs_push_and_notify gets called. In
xen_9pfs_push_and_notify, update the indexes (data has already been
copied to the sg by the common code) and send a notification to the
frontend.
Schedule the bottom-half to check if we already have any other requests
pending.
Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
CC: anthony.perard@citrix.com
CC: jgross@suse.com
CC: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
CC: Greg Kurz <groug@kaod.org>
---
hw/9pfs/xen-9p-backend.c | 16 ++++++++++++++++
1 file changed, 16 insertions(+)
diff --git a/hw/9pfs/xen-9p-backend.c b/hw/9pfs/xen-9p-backend.c
index 0570e07..80a8810 100644
--- a/hw/9pfs/xen-9p-backend.c
+++ b/hw/9pfs/xen-9p-backend.c
@@ -166,6 +166,22 @@ static void xen_9pfs_init_in_iov_from_pdu(V9fsPDU *pdu,
static void xen_9pfs_push_and_notify(V9fsPDU *pdu)
{
+ RING_IDX prod;
+ struct Xen9pfsDev *priv = container_of(pdu->s, struct Xen9pfsDev, state);
+ struct Xen9pfsRing *ring = &priv->rings[pdu->tag % priv->num_rings];
+
+ ring->intf->out_cons = ring->out_cons;
+ xen_wmb();
+
+ prod = ring->intf->in_prod;
+ xen_rmb();
+ ring->intf->in_prod = prod + pdu->size;
+ xen_wmb();
+
+ ring->inprogress = false;
+ xenevtchn_notify(ring->evtchndev, ring->local_port);
+
+ qemu_bh_schedule(ring->bh);
}
static const struct V9fsTransport xen_9p_transport = {
--
1.9.1
^ permalink raw reply related [flat|nested] 13+ messages in thread
* [Qemu-devel] [PATCH 8/8] xen/9pfs: build and register Xen 9pfs backend
2017-03-07 2:12 ` [Qemu-devel] [PATCH 1/8] xen: import ring.h from xen Stefano Stabellini
` (5 preceding siblings ...)
2017-03-07 2:12 ` [Qemu-devel] [PATCH 7/8] xen/9pfs: send responses back to the frontend Stefano Stabellini
@ 2017-03-07 2:12 ` Stefano Stabellini
2017-03-09 17:52 ` Greg Kurz
6 siblings, 1 reply; 13+ messages in thread
From: Stefano Stabellini @ 2017-03-07 2:12 UTC (permalink / raw)
To: qemu-devel
Cc: xen-devel, sstabellini, Stefano Stabellini, anthony.perard,
jgross, Aneesh Kumar K.V, Greg Kurz
Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
CC: anthony.perard@citrix.com
CC: jgross@suse.com
CC: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
CC: Greg Kurz <groug@kaod.org>
---
hw/9pfs/Makefile.objs | 1 +
hw/xen/xen_backend.c | 1 +
include/hw/xen/xen_backend.h | 1 +
3 files changed, 3 insertions(+)
diff --git a/hw/9pfs/Makefile.objs b/hw/9pfs/Makefile.objs
index da0ae0c..76a81c3 100644
--- a/hw/9pfs/Makefile.objs
+++ b/hw/9pfs/Makefile.objs
@@ -7,3 +7,4 @@ common-obj-$(CONFIG_OPEN_BY_HANDLE) += 9p-handle.o
common-obj-y += 9p-proxy.o
obj-y += virtio-9p-device.o
+obj-y += xen-9p-backend.o
diff --git a/hw/xen/xen_backend.c b/hw/xen/xen_backend.c
index 6c21c37..7aa347d 100644
--- a/hw/xen/xen_backend.c
+++ b/hw/xen/xen_backend.c
@@ -585,6 +585,7 @@ void xen_be_register_common(void)
xen_be_register("console", &xen_console_ops);
xen_be_register("vkbd", &xen_kbdmouse_ops);
xen_be_register("qdisk", &xen_blkdev_ops);
+ xen_be_register("9pfs", &xen_9pfs_ops);
#ifdef CONFIG_USB_LIBUSB
xen_be_register("qusb", &xen_usb_ops);
#endif
diff --git a/include/hw/xen/xen_backend.h b/include/hw/xen/xen_backend.h
index 4f4799a..84e686c 100644
--- a/include/hw/xen/xen_backend.h
+++ b/include/hw/xen/xen_backend.h
@@ -49,6 +49,7 @@ 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_9pfs_ops; /* xen-9p-backend.c */
extern struct XenDevOps xen_netdev_ops; /* xen_nic.c */
#ifdef CONFIG_USB_LIBUSB
extern struct XenDevOps xen_usb_ops; /* xen-usb.c */
--
1.9.1
^ permalink raw reply related [flat|nested] 13+ messages in thread
* Re: [Qemu-devel] [PATCH 8/8] xen/9pfs: build and register Xen 9pfs backend
2017-03-07 2:12 ` [Qemu-devel] [PATCH 8/8] xen/9pfs: build and register Xen 9pfs backend Stefano Stabellini
@ 2017-03-09 17:52 ` Greg Kurz
0 siblings, 0 replies; 13+ messages in thread
From: Greg Kurz @ 2017-03-09 17:52 UTC (permalink / raw)
To: Stefano Stabellini
Cc: qemu-devel, jgross, Stefano Stabellini, Aneesh Kumar K.V,
anthony.perard, xen-devel
[-- Attachment #1: Type: text/plain, Size: 2383 bytes --]
On Mon, 6 Mar 2017 18:12:48 -0800
Stefano Stabellini <sstabellini@kernel.org> wrote:
> Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
> CC: anthony.perard@citrix.com
> CC: jgross@suse.com
> CC: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
> CC: Greg Kurz <groug@kaod.org>
> ---
> hw/9pfs/Makefile.objs | 1 +
> hw/xen/xen_backend.c | 1 +
> include/hw/xen/xen_backend.h | 1 +
> 3 files changed, 3 insertions(+)
>
> diff --git a/hw/9pfs/Makefile.objs b/hw/9pfs/Makefile.objs
> index da0ae0c..76a81c3 100644
> --- a/hw/9pfs/Makefile.objs
> +++ b/hw/9pfs/Makefile.objs
> @@ -7,3 +7,4 @@ common-obj-$(CONFIG_OPEN_BY_HANDLE) += 9p-handle.o
> common-obj-y += 9p-proxy.o
>
> obj-y += virtio-9p-device.o
> +obj-y += xen-9p-backend.o
The xen-9p-backend.c file seems to be target agnostic: it can be built
only once. Also, it should only be built if QEMU was configured with
Xen support. Hence:
obj-common-$(CONFIG_XEN_BACKEND) += xen-9p-backend.o
> diff --git a/hw/xen/xen_backend.c b/hw/xen/xen_backend.c
> index 6c21c37..7aa347d 100644
> --- a/hw/xen/xen_backend.c
> +++ b/hw/xen/xen_backend.c
> @@ -585,6 +585,7 @@ void xen_be_register_common(void)
> xen_be_register("console", &xen_console_ops);
> xen_be_register("vkbd", &xen_kbdmouse_ops);
> xen_be_register("qdisk", &xen_blkdev_ops);
#ifdef CONFIG_VIRTFS
> + xen_be_register("9pfs", &xen_9pfs_ops);
#endif
> #ifdef CONFIG_USB_LIBUSB
> xen_be_register("qusb", &xen_usb_ops);
> #endif
> diff --git a/include/hw/xen/xen_backend.h b/include/hw/xen/xen_backend.h
> index 4f4799a..84e686c 100644
> --- a/include/hw/xen/xen_backend.h
> +++ b/include/hw/xen/xen_backend.h
> @@ -49,6 +49,7 @@ 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_9pfs_ops; /* xen-9p-backend.c */
ditto
> extern struct XenDevOps xen_netdev_ops; /* xen_nic.c */
> #ifdef CONFIG_USB_LIBUSB
> extern struct XenDevOps xen_usb_ops; /* xen-usb.c */
With the above fixes.
Reviewed-by: Greg Kurz <groug@kaod.org>
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [Qemu-devel] [PATCH 3/8] xen/9pfs: introduce Xen 9pfs backend
2017-03-07 2:12 ` [Qemu-devel] [PATCH 3/8] xen/9pfs: introduce Xen 9pfs backend Stefano Stabellini
@ 2017-03-09 17:54 ` Greg Kurz
2017-03-10 9:48 ` Greg Kurz
0 siblings, 1 reply; 13+ messages in thread
From: Greg Kurz @ 2017-03-09 17:54 UTC (permalink / raw)
To: Stefano Stabellini
Cc: qemu-devel, xen-devel, Stefano Stabellini, anthony.perard, jgross,
Aneesh Kumar K.V
[-- Attachment #1: Type: text/plain, Size: 3776 bytes --]
On Mon, 6 Mar 2017 18:12:43 -0800
Stefano Stabellini <sstabellini@kernel.org> wrote:
> Introduce the Xen 9pfs backend: add struct XenDevOps to register as a
> Xen backend and add struct V9fsTransport to register as v9fs transport.
>
> All functions are empty stubs for now.
>
> Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
> CC: anthony.perard@citrix.com
> CC: jgross@suse.com
> CC: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
> CC: Greg Kurz <groug@kaod.org>
> ---
> hw/9pfs/xen-9p-backend.c | 96 ++++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 96 insertions(+)
> create mode 100644 hw/9pfs/xen-9p-backend.c
>
> diff --git a/hw/9pfs/xen-9p-backend.c b/hw/9pfs/xen-9p-backend.c
> new file mode 100644
> index 0000000..924fb64
> --- /dev/null
> +++ b/hw/9pfs/xen-9p-backend.c
> @@ -0,0 +1,96 @@
> +/*
> + * Xen 9p backend
> + *
> + * Copyright Aporeto 2017
> + *
> + * Authors:
> + * Stefano Stabellini <stefano@aporeto.com>
> + *
> + */
> +
> +#include "qemu/osdep.h"
> +
> +#include "hw/hw.h"
> +#include "hw/9pfs/9p.h"
> +#include "hw/xen/xen_backend.h"
> +#include "xen_9pfs.h"
> +#include "qemu/config-file.h"
> +#include "fsdev/qemu-fsdev.h"
> +
> +struct Xen9pfsDev {
> + struct XenDevice xendev; /* must be first */
> +};
According to HACKING, this should be:
typedef struct Xen9pfsDev {
struct XenDevice xendev; /* must be first */
} Xen9pfsDev;
and...
> +
> +static ssize_t xen_9pfs_pdu_vmarshal(V9fsPDU *pdu,
> + size_t offset,
> + const char *fmt,
> + va_list ap)
> +{
> + return 0;
> +}
> +
> +static ssize_t xen_9pfs_pdu_vunmarshal(V9fsPDU *pdu,
> + size_t offset,
> + const char *fmt,
> + va_list ap)
> +{
> + return 0;
> +}
> +
> +static void xen_9pfs_init_out_iov_from_pdu(V9fsPDU *pdu,
> + struct iovec **piov,
> + unsigned int *pniov)
> +{
> +}
> +
> +static void xen_9pfs_init_in_iov_from_pdu(V9fsPDU *pdu,
> + struct iovec **piov,
> + unsigned int *pniov,
> + size_t size)
> +{
> +}
> +
> +static void xen_9pfs_push_and_notify(V9fsPDU *pdu)
> +{
> +}
> +
> +static const struct V9fsTransport xen_9p_transport = {
> + .pdu_vmarshal = xen_9pfs_pdu_vmarshal,
> + .pdu_vunmarshal = xen_9pfs_pdu_vunmarshal,
> + .init_in_iov_from_pdu = xen_9pfs_init_in_iov_from_pdu,
> + .init_out_iov_from_pdu = xen_9pfs_init_out_iov_from_pdu,
> + .push_and_notify = xen_9pfs_push_and_notify,
> +};
> +
> +static int xen_9pfs_init(struct XenDevice *xendev)
> +{
> + return 0;
> +}
> +
> +static int xen_9pfs_free(struct XenDevice *xendev)
> +{
> + return -1;
> +}
> +
> +static int xen_9pfs_connect(struct XenDevice *xendev)
> +{
> + return 0;
> +}
> +
> +static void xen_9pfs_alloc(struct XenDevice *xendev)
> +{
> +}
> +
> +static void xen_9pfs_disconnect(struct XenDevice *xendev)
> +{
> +}
> +
> +struct XenDevOps xen_9pfs_ops = {
> + .size = sizeof(struct Xen9pfsDev),
... s/struct //
> + .flags = DEVOPS_FLAG_NEED_GNTDEV,
> + .alloc = xen_9pfs_alloc,
> + .init = xen_9pfs_init,
> + .initialise = xen_9pfs_connect,
> + .disconnect = xen_9pfs_disconnect,
> + .free = xen_9pfs_free,
> +};
With the above comments addressed:
Reviewed-by: Greg Kurz <groug@kaod.org>
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [Qemu-devel] [PATCH 3/8] xen/9pfs: introduce Xen 9pfs backend
2017-03-09 17:54 ` Greg Kurz
@ 2017-03-10 9:48 ` Greg Kurz
2017-03-13 23:26 ` Stefano Stabellini
0 siblings, 1 reply; 13+ messages in thread
From: Greg Kurz @ 2017-03-10 9:48 UTC (permalink / raw)
To: Stefano Stabellini
Cc: qemu-devel, xen-devel, Stefano Stabellini, anthony.perard, jgross,
Aneesh Kumar K.V
[-- Attachment #1: Type: text/plain, Size: 5239 bytes --]
On Thu, 9 Mar 2017 18:54:26 +0100
Greg Kurz <groug@kaod.org> wrote:
> On Mon, 6 Mar 2017 18:12:43 -0800
> Stefano Stabellini <sstabellini@kernel.org> wrote:
>
> > Introduce the Xen 9pfs backend: add struct XenDevOps to register as a
> > Xen backend and add struct V9fsTransport to register as v9fs transport.
> >
> > All functions are empty stubs for now.
> >
> > Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
> > CC: anthony.perard@citrix.com
> > CC: jgross@suse.com
> > CC: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
> > CC: Greg Kurz <groug@kaod.org>
> > ---
> > hw/9pfs/xen-9p-backend.c | 96 ++++++++++++++++++++++++++++++++++++++++++++++++
> > 1 file changed, 96 insertions(+)
> > create mode 100644 hw/9pfs/xen-9p-backend.c
> >
> > diff --git a/hw/9pfs/xen-9p-backend.c b/hw/9pfs/xen-9p-backend.c
> > new file mode 100644
> > index 0000000..924fb64
> > --- /dev/null
> > +++ b/hw/9pfs/xen-9p-backend.c
> > @@ -0,0 +1,96 @@
> > +/*
> > + * Xen 9p backend
> > + *
> > + * Copyright Aporeto 2017
> > + *
> > + * Authors:
> > + * Stefano Stabellini <stefano@aporeto.com>
> > + *
> > + */
> > +
> > +#include "qemu/osdep.h"
> > +
> > +#include "hw/hw.h"
> > +#include "hw/9pfs/9p.h"
> > +#include "hw/xen/xen_backend.h"
> > +#include "xen_9pfs.h"
> > +#include "qemu/config-file.h"
> > +#include "fsdev/qemu-fsdev.h"
> > +
> > +struct Xen9pfsDev {
> > + struct XenDevice xendev; /* must be first */
> > +};
>
> According to HACKING, this should be:
>
> typedef struct Xen9pfsDev {
> struct XenDevice xendev; /* must be first */
> } Xen9pfsDev;
>
> and...
>
> > +
> > +static ssize_t xen_9pfs_pdu_vmarshal(V9fsPDU *pdu,
> > + size_t offset,
> > + const char *fmt,
> > + va_list ap)
> > +{
> > + return 0;
> > +}
> > +
> > +static ssize_t xen_9pfs_pdu_vunmarshal(V9fsPDU *pdu,
> > + size_t offset,
> > + const char *fmt,
> > + va_list ap)
> > +{
> > + return 0;
> > +}
> > +
> > +static void xen_9pfs_init_out_iov_from_pdu(V9fsPDU *pdu,
> > + struct iovec **piov,
> > + unsigned int *pniov)
> > +{
> > +}
> > +
> > +static void xen_9pfs_init_in_iov_from_pdu(V9fsPDU *pdu,
> > + struct iovec **piov,
> > + unsigned int *pniov,
> > + size_t size)
> > +{
> > +}
> > +
> > +static void xen_9pfs_push_and_notify(V9fsPDU *pdu)
> > +{
> > +}
> > +
> > +static const struct V9fsTransport xen_9p_transport = {
> > + .pdu_vmarshal = xen_9pfs_pdu_vmarshal,
> > + .pdu_vunmarshal = xen_9pfs_pdu_vunmarshal,
> > + .init_in_iov_from_pdu = xen_9pfs_init_in_iov_from_pdu,
> > + .init_out_iov_from_pdu = xen_9pfs_init_out_iov_from_pdu,
> > + .push_and_notify = xen_9pfs_push_and_notify,
> > +};
> > +
> > +static int xen_9pfs_init(struct XenDevice *xendev)
> > +{
> > + return 0;
> > +}
> > +
> > +static int xen_9pfs_free(struct XenDevice *xendev)
> > +{
> > + return -1;
> > +}
> > +
> > +static int xen_9pfs_connect(struct XenDevice *xendev)
> > +{
> > + return 0;
> > +}
> > +
> > +static void xen_9pfs_alloc(struct XenDevice *xendev)
> > +{
> > +}
> > +
> > +static void xen_9pfs_disconnect(struct XenDevice *xendev)
> > +{
> > +}
> > +
> > +struct XenDevOps xen_9pfs_ops = {
> > + .size = sizeof(struct Xen9pfsDev),
>
> ... s/struct //
>
> > + .flags = DEVOPS_FLAG_NEED_GNTDEV,
> > + .alloc = xen_9pfs_alloc,
> > + .init = xen_9pfs_init,
> > + .initialise = xen_9pfs_connect,
> > + .disconnect = xen_9pfs_disconnect,
> > + .free = xen_9pfs_free,
> > +};
>
> With the above comments addressed:
>
> Reviewed-by: Greg Kurz <groug@kaod.org>
Hmm... there's still a problem actually. This patch breaks build for some
targets with '--enable-xen --enable-virtfs':
LINK cris-softmmu/qemu-system-cris
../hw/xen/xen_backend.o: In function `xen_be_register_common':
/home/greg/Work/qemu/qemu-9p/hw/xen/xen_backend.c:588: undefined reference to `xen_9pfs_ops'
This happens because only targets that support virtio and PCI pull the
9pfs/fsdev object files. This is an effort to have smaller QEMU binaries,
based on the historical dependency of 9pfs on virtio. This patch breaks
this assumption if CONFIG_XEN_BACKEND is defined but the target doesn't
define CONFIG_VIRTIO and CONFIG_PCI.
Since 9pfs can now be used with another transport, I guess that the
dependency on virtio and PCI isn't right anymore. The new condition
for targets to support 9pfs should rather be something like:
CONFIG_VIRTFS == y && CONFIG_SOFTMMU == y && (CONFIG_VIRTIO == y || CONFIG_XEN_BACKEND == y)
This would cause all targets to pull the 9pfs/fsdev object files though.
So I have a question: do all targets need to support the Xen backend, really ?
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [Qemu-devel] [PATCH 3/8] xen/9pfs: introduce Xen 9pfs backend
2017-03-10 9:48 ` Greg Kurz
@ 2017-03-13 23:26 ` Stefano Stabellini
0 siblings, 0 replies; 13+ messages in thread
From: Stefano Stabellini @ 2017-03-13 23:26 UTC (permalink / raw)
To: Greg Kurz
Cc: Stefano Stabellini, jgross, qemu-devel, Stefano Stabellini,
Aneesh Kumar K.V, anthony.perard, xen-devel
On Fri, 10 Mar 2017, Greg Kurz wrote:
> On Thu, 9 Mar 2017 18:54:26 +0100
> Greg Kurz <groug@kaod.org> wrote:
>
> > On Mon, 6 Mar 2017 18:12:43 -0800
> > Stefano Stabellini <sstabellini@kernel.org> wrote:
> >
> > > Introduce the Xen 9pfs backend: add struct XenDevOps to register as a
> > > Xen backend and add struct V9fsTransport to register as v9fs transport.
> > >
> > > All functions are empty stubs for now.
> > >
> > > Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
> > > CC: anthony.perard@citrix.com
> > > CC: jgross@suse.com
> > > CC: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
> > > CC: Greg Kurz <groug@kaod.org>
> > > ---
> > > hw/9pfs/xen-9p-backend.c | 96 ++++++++++++++++++++++++++++++++++++++++++++++++
> > > 1 file changed, 96 insertions(+)
> > > create mode 100644 hw/9pfs/xen-9p-backend.c
> > >
> > > diff --git a/hw/9pfs/xen-9p-backend.c b/hw/9pfs/xen-9p-backend.c
> > > new file mode 100644
> > > index 0000000..924fb64
> > > --- /dev/null
> > > +++ b/hw/9pfs/xen-9p-backend.c
> > > @@ -0,0 +1,96 @@
> > > +/*
> > > + * Xen 9p backend
> > > + *
> > > + * Copyright Aporeto 2017
> > > + *
> > > + * Authors:
> > > + * Stefano Stabellini <stefano@aporeto.com>
> > > + *
> > > + */
> > > +
> > > +#include "qemu/osdep.h"
> > > +
> > > +#include "hw/hw.h"
> > > +#include "hw/9pfs/9p.h"
> > > +#include "hw/xen/xen_backend.h"
> > > +#include "xen_9pfs.h"
> > > +#include "qemu/config-file.h"
> > > +#include "fsdev/qemu-fsdev.h"
> > > +
> > > +struct Xen9pfsDev {
> > > + struct XenDevice xendev; /* must be first */
> > > +};
> >
> > According to HACKING, this should be:
> >
> > typedef struct Xen9pfsDev {
> > struct XenDevice xendev; /* must be first */
> > } Xen9pfsDev;
> >
> > and...
> >
> > > +
> > > +static ssize_t xen_9pfs_pdu_vmarshal(V9fsPDU *pdu,
> > > + size_t offset,
> > > + const char *fmt,
> > > + va_list ap)
> > > +{
> > > + return 0;
> > > +}
> > > +
> > > +static ssize_t xen_9pfs_pdu_vunmarshal(V9fsPDU *pdu,
> > > + size_t offset,
> > > + const char *fmt,
> > > + va_list ap)
> > > +{
> > > + return 0;
> > > +}
> > > +
> > > +static void xen_9pfs_init_out_iov_from_pdu(V9fsPDU *pdu,
> > > + struct iovec **piov,
> > > + unsigned int *pniov)
> > > +{
> > > +}
> > > +
> > > +static void xen_9pfs_init_in_iov_from_pdu(V9fsPDU *pdu,
> > > + struct iovec **piov,
> > > + unsigned int *pniov,
> > > + size_t size)
> > > +{
> > > +}
> > > +
> > > +static void xen_9pfs_push_and_notify(V9fsPDU *pdu)
> > > +{
> > > +}
> > > +
> > > +static const struct V9fsTransport xen_9p_transport = {
> > > + .pdu_vmarshal = xen_9pfs_pdu_vmarshal,
> > > + .pdu_vunmarshal = xen_9pfs_pdu_vunmarshal,
> > > + .init_in_iov_from_pdu = xen_9pfs_init_in_iov_from_pdu,
> > > + .init_out_iov_from_pdu = xen_9pfs_init_out_iov_from_pdu,
> > > + .push_and_notify = xen_9pfs_push_and_notify,
> > > +};
> > > +
> > > +static int xen_9pfs_init(struct XenDevice *xendev)
> > > +{
> > > + return 0;
> > > +}
> > > +
> > > +static int xen_9pfs_free(struct XenDevice *xendev)
> > > +{
> > > + return -1;
> > > +}
> > > +
> > > +static int xen_9pfs_connect(struct XenDevice *xendev)
> > > +{
> > > + return 0;
> > > +}
> > > +
> > > +static void xen_9pfs_alloc(struct XenDevice *xendev)
> > > +{
> > > +}
> > > +
> > > +static void xen_9pfs_disconnect(struct XenDevice *xendev)
> > > +{
> > > +}
> > > +
> > > +struct XenDevOps xen_9pfs_ops = {
> > > + .size = sizeof(struct Xen9pfsDev),
> >
> > ... s/struct //
> >
> > > + .flags = DEVOPS_FLAG_NEED_GNTDEV,
> > > + .alloc = xen_9pfs_alloc,
> > > + .init = xen_9pfs_init,
> > > + .initialise = xen_9pfs_connect,
> > > + .disconnect = xen_9pfs_disconnect,
> > > + .free = xen_9pfs_free,
> > > +};
> >
> > With the above comments addressed:
> >
> > Reviewed-by: Greg Kurz <groug@kaod.org>
Thank you for the quick review and useful comments. I'll address them in
the next version of the series.
> Hmm... there's still a problem actually. This patch breaks build for some
> targets with '--enable-xen --enable-virtfs':
>
> LINK cris-softmmu/qemu-system-cris
> ../hw/xen/xen_backend.o: In function `xen_be_register_common':
> /home/greg/Work/qemu/qemu-9p/hw/xen/xen_backend.c:588: undefined reference to `xen_9pfs_ops'
I could reproduce the problem
> This happens because only targets that support virtio and PCI pull the
> 9pfs/fsdev object files. This is an effort to have smaller QEMU binaries,
> based on the historical dependency of 9pfs on virtio. This patch breaks
> this assumption if CONFIG_XEN_BACKEND is defined but the target doesn't
> define CONFIG_VIRTIO and CONFIG_PCI.
>
> Since 9pfs can now be used with another transport, I guess that the
> dependency on virtio and PCI isn't right anymore. The new condition
> for targets to support 9pfs should rather be something like:
>
> CONFIG_VIRTFS == y && CONFIG_SOFTMMU == y && (CONFIG_VIRTIO == y || CONFIG_XEN_BACKEND == y)
>
> This would cause all targets to pull the 9pfs/fsdev object files though.
>
> So I have a question: do all targets need to support the Xen backend, really ?
Not really: Xen is only supported on x86 and ARM platforms today.
Typically, Xen is only enabled in QEMU for i386 or x86_64 softmmu
targets, even when used in an ARM context (long story short: the Xen PV
machine is arch-agnostic, but the QEMU build doesn't support that, so we
tied it to x86 just because we had to pick one).
In fact, I think the issue is that CONFIG_XEN_BACKEND is set wrongly by
the configure script. It's not an host property, but a target property
like CONFIG_XEN. We need:
diff --git a/configure b/configure
index 6c21975..6d8f752 100755
--- a/configure
+++ b/configure
@@ -5442,7 +5442,6 @@ if test "$virglrenderer" = "yes" ; then
echo "VIRGL_LIBS=$virgl_libs" >> $config_host_mak
fi
if test "$xen" = "yes" ; then
- echo "CONFIG_XEN_BACKEND=y" >> $config_host_mak
echo "CONFIG_XEN_CTRL_INTERFACE_VERSION=$xen_ctrl_version" >> $config_host_mak
if test "$xen_pv_domain_build" = "yes" ; then
echo "CONFIG_XEN_PV_DOMAIN_BUILD=y" >> $config_host_mak
@@ -6028,6 +6027,7 @@ case "$target_name" in
i386|x86_64)
if test "$xen" = "yes" -a "$target_softmmu" = "yes" ; then
echo "CONFIG_XEN=y" >> $config_target_mak
+ echo "CONFIG_XEN_BACKEND=y" >> $config_target_mak
if test "$xen_pci_passthrough" = yes; then
echo "CONFIG_XEN_PCI_PASSTHROUGH=y" >> "$config_target_mak"
fi
With this change, and the other change you suggested:
common-obj-$(CONFIG_XEN_BACKEND) += xen-9p-backend.o
I couldn't see any build issues.
^ permalink raw reply related [flat|nested] 13+ messages in thread
end of thread, other threads:[~2017-03-13 23:26 UTC | newest]
Thread overview: 13+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-03-07 2:12 [Qemu-devel] [PATCH 0/8] xen/9pfs: introduce the Xen 9pfs backend Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 1/8] xen: import ring.h from xen Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 2/8] xen: introduce the header file for the Xen 9pfs transport protocol Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 3/8] xen/9pfs: introduce Xen 9pfs backend Stefano Stabellini
2017-03-09 17:54 ` Greg Kurz
2017-03-10 9:48 ` Greg Kurz
2017-03-13 23:26 ` Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 4/8] xen/9pfs: connect to the frontend Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 5/8] xen/9pfs: receive requests from " Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 6/8] xen/9pfs: implement in/out_iov_from_pdu and vmarshal/vunmarshal Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 7/8] xen/9pfs: send responses back to the frontend Stefano Stabellini
2017-03-07 2:12 ` [Qemu-devel] [PATCH 8/8] xen/9pfs: build and register Xen 9pfs backend Stefano Stabellini
2017-03-09 17:52 ` Greg Kurz
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).