[stgt] [PATCH] sheepdog backing store support

FUJITA Tomonori fujita.tomonori at lab.ntt.co.jp
Tue Sep 18 02:37:24 CEST 2012


Chen Xiaoxi asked me to send the following patch.

=
>From 186aaea8ed077832d26b30543d1fee86875d50b6 Mon Sep 17 00:00:00 2001
From: Chen Xiaoxi <xiaoxi.chen at intel.com>
Date: Wed, 12 Sep 2012 15:09:43 +0800
Subject: [PATCH] multi-thread sheepdog backend


diff --git a/usr/bs_sheepdog.c b/usr/bs_sheepdog.c
new file mode 100644
index 0000000..5946d84
--- /dev/null
+++ b/usr/bs_sheepdog.c
@@ -0,0 +1,183 @@
+/*
+ * Copyright (C) 2010 FUJITA Tomonori <tomof at acm.org>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation, version 2 of the
+ * License.
+ *
+ * This program 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
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
+ * 02110-1301 USA
+ */
+#include <errno.h>
+#include <fcntl.h>
+#include <inttypes.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
+#include <linux/fs.h>
+#include <sys/epoll.h>
+
+#include "list.h"
+#include "util.h"
+#include "tgtd.h"
+#include "scsi.h"
+#include "bs_thread.h"
+#include "sheepdog.h"
+
+static void set_medium_error(int *result, uint8_t *key, uint16_t *asc)
+{
+	*result = SAM_STAT_CHECK_CONDITION;
+	*key = MEDIUM_ERROR;
+	*asc = ASC_READ_ERROR;
+}
+
+static void bs_sheepdog_request(struct scsi_cmd *cmd)
+{
+	int ret;
+	uint32_t length;
+	int result = SAM_STAT_GOOD;
+	uint8_t key;
+	uint16_t asc;
+	struct bs_thread_info *info = BS_THREAD_I(cmd->dev);
+	//xiaoxi added here
+	pthread_t myThreadId=pthread_self();
+	int myTID=0;
+//	eprintf("in bs_sheepdog_request,systid=%u\n",myThreadId);
+	int i;
+	for(i = 0; i < info->nr_worker_threads; i++)
+	{
+		dprintf("in for:TID=%d:sysid=%u,myid=%u\n",i,info->worker_thread[i],myThreadId);
+		if((uint32_t)info->worker_thread[i] == (uint32_t)myThreadId)
+		{
+			dprintf("equal\n");
+			myTID=i;
+			break;
+		}
+	}
+//	eprintf("here,myTID=%d\n",myTID);
+	struct sheepdog_access_info *ai = (struct sheepdog_access_info *)(info + 1);
+//	eprintf("ai_base:%p,TID:%d,ai_addr:%p,fd=%d\n",ai,myTID,ai+myTID,(ai+myTID)->fd);
+
+
+	ret = length = 0;
+	key = asc = 0;
+
+	switch (cmd->scb[0])
+	{
+	case SYNCHRONIZE_CACHE:
+	case SYNCHRONIZE_CACHE_16:
+		break;
+	case WRITE_6:
+	case WRITE_10:
+	case WRITE_12:
+	case WRITE_16:
+		/* eprintf("%s, %s, %s, %x %u %lu\n", cmd->dev->path, ai->s_token, ai->s_url, */
+		/* cmd->scb[0], scsi_get_out_length(cmd), cmd->offset); */
+
+		length = scsi_get_out_length(cmd);
+		ret = sd_io(ai, 1, scsi_get_out_buffer(cmd), length, cmd->offset);
+		if (ret)
+			set_medium_error(&result, &key, &asc);
+		break;
+	case READ_6:
+	case READ_10:
+	case READ_12:
+	case READ_16:
+		/* eprintf("%s, %s, %s, %x %u %lu\n", cmd->dev->path, ai->s_token, ai->s_url, */
+		/* cmd->scb[0], scsi_get_in_length(cmd), cmd->offset); */
+
+		length = scsi_get_in_length(cmd);
+		ret = sd_io(ai, 0, scsi_get_in_buffer(cmd), length, cmd->offset);
+		if (ret)
+			set_medium_error(&result, &key, &asc);
+		break;
+	default:
+		break;
+	}
+
+	dprintf("io done %p %x %d %u\n", cmd, cmd->scb[0], ret, length);
+
+	scsi_set_result(cmd, result);
+
+	if (result != SAM_STAT_GOOD) {
+		eprintf("io error %p %x %d %d %" PRIu64 ", %m\n",
+			cmd, cmd->scb[0], ret, length, cmd->offset);
+		sense_data_build(cmd, key, asc);
+	}
+}
+
+static int bs_sheepdog_open(struct scsi_lu *lu, char *path,
+			    int *fd, uint64_t *size)
+{
+	eprintf("in bs_sheep_open\n");
+	struct bs_thread_info *info = BS_THREAD_I(lu);
+	struct sheepdog_access_info *ai =
+		(struct sheepdog_access_info *)(info + 1);
+	int ret;
+
+	ret = sd_open(ai, path, nr_iothreads);
+	if (ret)
+		return ret;
+
+	*size = ai[0].inode.vdi_size;
+
+	return 0;
+}
+
+static void bs_sheepdog_close(struct scsi_lu *lu)
+{
+	eprintf("in bs_sheep_close");
+}
+
+static int bs_sheepdog_init(struct scsi_lu *lu)
+{
+	eprintf("in bs_sheep_init\n");
+	struct bs_thread_info *info = BS_THREAD_I(lu);
+
+	if(nr_iothreads > WORK_THREAD_MAX_NUM)
+	{
+		eprintf("ERROR:TOO MANY THREADS\n");
+		return -1;
+	}
+	return bs_thread_open(info, bs_sheepdog_request, nr_iothreads);
+}
+
+static void bs_sheepdog_exit(struct scsi_lu *lu)
+{
+	eprintf("in bs_sheep_exit\n");
+	struct bs_thread_info *info = BS_THREAD_I(lu);
+
+	bs_thread_close(info);
+}
+
+static int bs_sheepdog_cmd_done(struct scsi_cmd *cmd)
+{
+	eprintf("in bs_sheep_cmd_done\n");
+	return 0;
+}
+
+static struct backingstore_template sheepdog_bst = {
+	.bs_name		= "sheepdog",
+	.bs_datasize		= sizeof(struct bs_thread_info)	+ sizeof(struct sheepdog_access_info) * WORK_THREAD_MAX_NUM,
+	.bs_open		= bs_sheepdog_open,
+	.bs_close		= bs_sheepdog_close,
+	.bs_init		= bs_sheepdog_init,
+	.bs_exit		= bs_sheepdog_exit,
+	.bs_cmd_submit		= bs_thread_cmd_submit,
+	//.bs_cmd_done		= bs_sheepdog_cmd_done,
+};
+
+__attribute__((constructor)) static void __constructor(void)
+{
+	register_backingstore_template(&sheepdog_bst);
+}
diff --git a/usr/sheepdog.c b/usr/sheepdog.c
new file mode 100644
index 0000000..a7e3531
--- /dev/null
+++ b/usr/sheepdog.c
@@ -0,0 +1,553 @@
+#include <errno.h>
+#include <fcntl.h>
+#include <inttypes.h>
+#include <netdb.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <arpa/inet.h>
+#include <netinet/in.h>
+#include <netinet/tcp.h>
+#include <sys/epoll.h>
+#include <sys/socket.h>
+
+#include "list.h"
+#include "tgtd.h"
+#include "util.h"
+#include "log.h"
+#include "sheepdog.h"
+
+/*
+ * 64 bit FNV-1a non-zero initial basis
+ */
+#define FNV1A_64_INIT ((uint64_t)0xcbf29ce484222325ULL)
+
+/*
+ * 64 bit Fowler/Noll/Vo FNV-1a hash code
+ */
+static inline uint64_t fnv_64a_buf(void *buf, size_t len, uint64_t hval)
+{
+    unsigned char *bp = buf;
+    unsigned char *be = bp + len;
+    while (bp < be) {
+        hval ^= (uint64_t) *bp++;
+        hval += (hval << 1) + (hval << 4) + (hval << 5) +
+            (hval << 7) + (hval << 8) + (hval << 40);
+    }
+    return hval;
+}
+
+static inline int is_data_obj_writeable(SheepdogInode *inode, unsigned int idx)
+{
+    return inode->vdi_id == inode->data_vdi_id[idx];
+}
+
+static inline int is_data_obj(uint64_t oid)
+{
+    return !(VDI_BIT & oid);
+}
+
+static inline uint64_t data_oid_to_idx(uint64_t oid)
+{
+    return oid & (MAX_DATA_OBJS - 1);
+}
+
+static inline uint64_t vid_to_vdi_oid(uint32_t vid)
+{
+    return VDI_BIT | ((uint64_t)vid << VDI_SPACE_SHIFT);
+}
+
+static inline uint64_t vid_to_vmstate_oid(uint32_t vid, uint32_t idx)
+{
+    return VMSTATE_BIT | ((uint64_t)vid << VDI_SPACE_SHIFT) | idx;
+}
+
+static inline uint64_t vid_to_data_oid(uint32_t vid, uint32_t idx)
+{
+    return ((uint64_t)vid << VDI_SPACE_SHIFT) | idx;
+}
+
+static const char * sd_strerror(int err)
+{
+    int i;
+
+    static const struct {
+        int err;
+        const char *desc;
+    } errors[] = {
+        {SD_RES_SUCCESS, "Success"},
+        {SD_RES_UNKNOWN, "Unknown error"},
+        {SD_RES_NO_OBJ, "No object found"},
+        {SD_RES_EIO, "I/O error"},
+        {SD_RES_VDI_EXIST, "VDI exists already"},
+        {SD_RES_INVALID_PARMS, "Invalid parameters"},
+        {SD_RES_SYSTEM_ERROR, "System error"},
+        {SD_RES_VDI_LOCKED, "VDI is already locked"},
+        {SD_RES_NO_VDI, "No vdi found"},
+        {SD_RES_NO_BASE_VDI, "No base VDI found"},
+        {SD_RES_VDI_READ, "Failed read the requested VDI"},
+        {SD_RES_VDI_WRITE, "Failed to write the requested VDI"},
+        {SD_RES_BASE_VDI_READ, "Failed to read the base VDI"},
+        {SD_RES_BASE_VDI_WRITE, "Failed to write the base VDI"},
+        {SD_RES_NO_TAG, "Failed to find the requested tag"},
+        {SD_RES_STARTUP, "The system is still booting"},
+        {SD_RES_VDI_NOT_LOCKED, "VDI isn't locked"},
+        {SD_RES_SHUTDOWN, "The system is shutting down"},
+        {SD_RES_NO_MEM, "Out of memory on the server"},
+        {SD_RES_FULL_VDI, "We already have the maximum vdis"},
+        {SD_RES_VER_MISMATCH, "Protocol version mismatch"},
+        {SD_RES_NO_SPACE, "Server has no space for new objects"},
+        {SD_RES_WAIT_FOR_FORMAT, "Sheepdog is waiting for a format operation"},
+        {SD_RES_WAIT_FOR_JOIN, "Sheepdog is waiting for other nodes joining"},
+        {SD_RES_JOIN_FAILED, "Target node had failed to join sheepdog"},
+    };
+
+    for (i = 0; i < ARRAY_SIZE(errors); ++i) {
+        if (errors[i].err == err) {
+            return errors[i].desc;
+        }
+    }
+
+    return "Invalid error code";
+}
+
+/*
+ * Send/recv data with iovec buffers
+ *
+ * This function send/recv data from/to the iovec buffer directly.
+ * The first `offset' bytes in the iovec buffer are skipped and next
+ * `len' bytes are used.
+ *
+ * For example,
+ *
+ *   do_send_recv(sockfd, iov, len, offset, 1);
+ *
+ * is equals to
+ *
+ *   char *buf = malloc(size);
+ *   iov_to_buf(iov, iovcnt, buf, offset, size);
+ *   send(sockfd, buf, size, 0);
+ *   free(buf);
+ */
+static int do_send_recv(int sockfd, struct iovec *iov, int len, int offset,
+                        int write)
+{
+    struct msghdr msg;
+    int ret, diff;
+
+    memset(&msg, 0, sizeof(msg));
+    msg.msg_iov = iov;
+    msg.msg_iovlen = 1;
+
+    len += offset;
+
+    while (iov->iov_len < len) {
+        len -= iov->iov_len;
+
+        iov++;
+        msg.msg_iovlen++;
+    }
+
+    diff = iov->iov_len - len;
+    iov->iov_len -= diff;
+
+    while (msg.msg_iov->iov_len <= offset) {
+        offset -= msg.msg_iov->iov_len;
+
+        msg.msg_iov++;
+        msg.msg_iovlen--;
+    }
+
+    msg.msg_iov->iov_base = (char *) msg.msg_iov->iov_base + offset;
+    msg.msg_iov->iov_len -= offset;
+
+    if (write) {
+        ret = sendmsg(sockfd, &msg, 0);
+    } else {
+        ret = recvmsg(sockfd, &msg, 0);
+    }
+
+    msg.msg_iov->iov_base = (char *) msg.msg_iov->iov_base - offset;
+    msg.msg_iov->iov_len += offset;
+
+    iov->iov_len += diff;
+    return ret;
+}
+
+static int connect_to_sdog(const char *addr, const char *port)
+{
+    char hbuf[NI_MAXHOST], sbuf[NI_MAXSERV];
+    int fd, ret;
+    struct addrinfo hints, *res, *res0;
+
+    if (!addr) {
+        addr = SD_DEFAULT_ADDR;
+        port = SD_DEFAULT_PORT;
+    }
+
+    memset(&hints, 0, sizeof(hints));
+    hints.ai_socktype = SOCK_STREAM;
+
+    ret = getaddrinfo(addr, port, &hints, &res0);
+    if (ret) {
+        eprintf("unable to get address info %s, %s\n",
+                     addr, strerror(errno));
+        return -1;
+    }
+
+    for (res = res0; res; res = res->ai_next) {
+        ret = getnameinfo(res->ai_addr, res->ai_addrlen, hbuf, sizeof(hbuf),
+                          sbuf, sizeof(sbuf), NI_NUMERICHOST | NI_NUMERICSERV);
+        if (ret) {
+            continue;
+        }
+
+        fd = socket(res->ai_family, res->ai_socktype, res->ai_protocol);
+        if (fd < 0) {
+            continue;
+        }
+
+    reconnect:
+        ret = connect(fd, res->ai_addr, res->ai_addrlen);
+        if (ret < 0) {
+            if (errno == EINTR) {
+                goto reconnect;
+            }
+            break;
+        }
+
+        dprintf("connected to %s:%s\n", addr, port);
+        goto success;
+    }
+    fd = -1;
+    eprintf("failed connect to %s:%s\n", addr, port);
+success:
+    freeaddrinfo(res0);
+    return fd;
+}
+
+static int do_readv_writev(int sockfd, struct iovec *iov, int len,
+                           int iov_offset, int write)
+{
+    int ret;
+again:
+    ret = do_send_recv(sockfd, iov, len, iov_offset, write);
+    if (ret < 0) {
+        if (errno == EINTR || errno == EAGAIN) {
+            goto again;
+        }
+        eprintf("failed to recv a rsp, %s\n", strerror(errno));
+        return 1;
+    }
+
+    iov_offset += ret;
+    len -= ret;
+    if (len) {
+        goto again;
+    }
+
+    return 0;
+}
+
+/* static int do_readv(int sockfd, struct iovec *iov, int len, int iov_offset) */
+/* { */
+/*     return do_readv_writev(sockfd, iov, len, iov_offset, 0); */
+/* } */
+
+static int do_writev(int sockfd, struct iovec *iov, int len, int iov_offset)
+{
+    return do_readv_writev(sockfd, iov, len, iov_offset, 1);
+}
+
+static int do_read_write(int sockfd, void *buf, int len, int write)
+{
+    struct iovec iov;
+
+    iov.iov_base = buf;
+    iov.iov_len = len;
+
+    return do_readv_writev(sockfd, &iov, len, 0, write);
+}
+
+static int do_read(int sockfd, void *buf, int len)
+{
+    return do_read_write(sockfd, buf, len, 0);
+}
+
+/* static int do_write(int sockfd, void *buf, int len) */
+/* { */
+/*     return do_read_write(sockfd, buf, len, 1); */
+/* } */
+
+static int send_req(int sockfd, SheepdogReq *hdr, void *data,
+                    unsigned int *wlen)
+{
+    int ret;
+    struct iovec iov[2];
+
+    iov[0].iov_base = hdr;
+    iov[0].iov_len = sizeof(*hdr);
+
+    if (*wlen) {
+        iov[1].iov_base = data;
+        iov[1].iov_len = *wlen;
+    }
+
+    ret = do_writev(sockfd, iov, sizeof(*hdr) + *wlen, 0);
+    if (ret) {
+        eprintf("failed to send a req, %s\n", strerror(errno));
+        ret = -1;
+    }
+
+    return ret;
+}
+
+static int do_req(int sockfd, SheepdogReq *hdr, void *data,
+                  unsigned int *wlen, unsigned int *rlen)
+{
+    int ret;
+
+    ret = send_req(sockfd, hdr, data, wlen);
+    if (ret) {
+        ret = -1;
+        goto out;
+    }
+
+    ret = do_read(sockfd, hdr, sizeof(*hdr));
+    if (ret) {
+        eprintf("failed to get a rsp, %s\n", strerror(errno));
+        ret = -1;
+        goto out;
+    }
+
+    if (*rlen > hdr->data_length) {
+        *rlen = hdr->data_length;
+    }
+
+    if (*rlen) {
+        ret = do_read(sockfd, data, *rlen);
+        if (ret) {
+            eprintf("failed to get the data, %s\n", strerror(errno));
+            ret = -1;
+            goto out;
+        }
+    }
+    ret = 0;
+out:
+    return ret;
+}
+
+static int read_write_object(int fd, char *buf, uint64_t oid, int copies,
+                             unsigned int datalen, uint64_t offset,
+                             int write, int create)
+{
+    SheepdogObjReq hdr;
+    SheepdogObjRsp *rsp = (SheepdogObjRsp *)&hdr;
+    unsigned int wlen, rlen;
+    int ret;
+
+    memset(&hdr, 0, sizeof(hdr));
+
+    if (write) {
+        wlen = datalen;
+        rlen = 0;
+        hdr.flags = SD_FLAG_CMD_WRITE;
+        if (create) {
+            hdr.opcode = SD_OP_CREATE_AND_WRITE_OBJ;
+        } else {
+            hdr.opcode = SD_OP_WRITE_OBJ;
+        }
+    } else {
+        wlen = 0;
+        rlen = datalen;
+        hdr.opcode = SD_OP_READ_OBJ;
+    }
+    hdr.oid = oid;
+    hdr.data_length = datalen;
+    hdr.offset = offset;
+    hdr.copies = copies;
+
+    ret = do_req(fd, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
+    if (ret) {
+        eprintf("failed to send a request to the sheep\n");
+        return -1;
+    }
+
+    switch (rsp->result) {
+    case SD_RES_SUCCESS:
+        return 0;
+    default:
+        eprintf("%s\n", sd_strerror(rsp->result));
+        return -1;
+    }
+}
+
+static int read_object(int fd, char *buf, uint64_t oid, int copies,
+                       unsigned int datalen, uint64_t offset)
+{
+    return read_write_object(fd, buf, oid, copies, datalen, offset, 0, 0);
+}
+
+static int write_object(int fd, char *buf, uint64_t oid, int copies,
+                        unsigned int datalen, uint64_t offset, int create)
+{
+    return read_write_object(fd, buf, oid, copies, datalen, offset, 1, create);
+}
+
+int sd_io(struct sheepdog_access_info *ai, int write, char *buf,
+	  int len, uint64_t offset)
+{
+	uint32_t vid = ai->inode.vdi_id;
+	unsigned long idx = offset / SD_DATA_OBJ_SIZE;
+	unsigned long max =
+		(offset + len + (SD_DATA_OBJ_SIZE - 1)) / SD_DATA_OBJ_SIZE;
+	unsigned obj_offset = offset % SD_DATA_OBJ_SIZE;
+	size_t size, rest = len;
+	int ret;
+
+	for (; idx < max; idx++) {
+		size = SD_DATA_OBJ_SIZE - obj_offset;
+		size = min_t(size_t, size, rest);
+
+		if (write)
+			ret = write_object(ai->fd, buf + (len - rest),
+					   vid_to_data_oid(vid, idx), 1,
+					   size, obj_offset, 0);
+		else
+			ret = read_object(ai->fd, buf + (len - rest),
+					  vid_to_data_oid(vid, idx), 1,
+					  size, obj_offset);
+		if (ret) {
+			eprintf("%lu %d\n", idx, ret);
+			return -1;
+		}
+
+		rest -= size;
+		obj_offset = 0;
+	}
+
+	return 0;
+}
+
+static int find_vdi_name(char *filename, uint32_t snapid,
+                         char *tag, uint32_t *vid, int for_snapshot)
+{
+    int ret, fd;
+    SheepdogVdiReq hdr;
+    SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr;
+    unsigned int wlen, rlen = 0;
+    char buf[SD_MAX_VDI_LEN + SD_MAX_VDI_TAG_LEN];
+
+    fd = connect_to_sdog(NULL, NULL);
+    if (fd < 0) {
+        return -1;
+    }
+
+    memset(buf, 0, sizeof(buf));
+    strncpy(buf, filename, SD_MAX_VDI_LEN);
+    strncpy(buf + SD_MAX_VDI_LEN, tag, SD_MAX_VDI_TAG_LEN);
+
+    memset(&hdr, 0, sizeof(hdr));
+    if (for_snapshot) {
+        hdr.opcode = SD_OP_GET_VDI_INFO;
+    } else {
+        hdr.opcode = SD_OP_LOCK_VDI;
+    }
+    wlen = SD_MAX_VDI_LEN + SD_MAX_VDI_TAG_LEN;
+    hdr.proto_ver = SD_PROTO_VER;
+    hdr.data_length = wlen;
+    hdr.snapid = snapid;
+    hdr.flags = SD_FLAG_CMD_WRITE;
+
+    ret = do_req(fd, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
+    if (ret) {
+        ret = -1;
+        goto out;
+    }
+
+    if (rsp->result != SD_RES_SUCCESS) {
+        eprintf("cannot get vdi info, %s, %s %d %s\n",
+		sd_strerror(rsp->result), filename, snapid, tag);
+        ret = -1;
+        goto out;
+    }
+    *vid = rsp->vdi_id;
+
+    ret = 0;
+out:
+    close(fd);
+    return ret;
+}
+
+int sd_open(struct sheepdog_access_info *ai, char *filename, int flags)
+{
+    int i;
+    int ret, fd;
+    uint32_t vid = 0;
+    char vdi[SD_MAX_VDI_LEN], tag[SD_MAX_VDI_TAG_LEN];
+
+    struct sheepdog_access_info *ai_list=(struct sheepdog_access_info *)ai;
+
+    memset(vdi, 0, sizeof(vdi));
+    memset(tag, 0, sizeof(tag));
+    
+    /* if (parse_vdiname(s, filename, vdi, &snapid, tag) < 0) { */
+    /*     goto out; */
+    /* } */
+    /* s->fd = get_sheep_fd(s); */
+    /* if (s->fd < 0) { */
+    /*     goto out; */
+    /* } */
+
+    ret = find_vdi_name(filename, CURRENT_VDI_ID, tag, &vid, 0);
+    if (ret) {
+        goto out;
+    }
+
+    /* if (snapid) { */
+    /*     dprintf("%" PRIx32 " snapshot inode was open.\n", vid); */
+    /*     s->is_snapshot = 1; */
+    /* } */
+
+    for(i=0;i<flags;i++)
+    { 
+    	fd = connect_to_sdog(NULL, NULL);
+    	if (fd < 0) {
+        	eprintf("failed to connect\n");
+        	goto out;
+    	}
+	ai_list[i].fd=fd;
+	dprintf("thread %d,fd %d\n",i,fd);
+
+    ret = read_object(ai_list[i].fd, (char *)&(ai_list[i].inode), vid_to_vdi_oid(vid),
+		      0, SD_INODE_SIZE, 0);
+
+    }
+/*
+    for(i=1; i<flags ;i++)
+    {
+	memcpy(&ai_list[0].inode,&ai_list[i].inode,sizeof(struct SheepdogInode));
+    }
+*/  
+  if (ret) {
+        goto out;
+    }
+
+    /* memcpy(&s->inode, buf, sizeof(s->inode)); */
+    /* s->min_dirty_data_idx = UINT32_MAX; */
+    /* s->max_dirty_data_idx = 0; */
+
+    /* bs->total_sectors = s->inode.vdi_size / SECTOR_SIZE; */
+    /* strncpy(s->name, vdi, sizeof(s->name)); */
+    /* qemu_free(buf); */
+    return 0;
+out:
+/*     qemu_aio_set_fd_handler(s->fd, NULL, NULL, NULL, NULL, NULL); */
+/*     if (s->fd >= 0) { */
+/*         closesocket(s->fd); */
+/*     } */
+/*     qemu_free(buf); */
+    return -1;
+}
+
diff --git a/usr/sheepdog.h b/usr/sheepdog.h
new file mode 100644
index 0000000..4251ee4
--- /dev/null
+++ b/usr/sheepdog.h
@@ -0,0 +1,163 @@
+#define SD_PROTO_VER 0x01
+
+//xiaoxi added here
+#define	WORK_THREAD_MAX_NUM 10
+//end of add
+
+#define SD_DEFAULT_ADDR "localhost"
+#define SD_DEFAULT_PORT "7000"
+
+#define SD_OP_CREATE_AND_WRITE_OBJ  0x01
+#define SD_OP_READ_OBJ       0x02
+#define SD_OP_WRITE_OBJ      0x03
+
+#define SD_OP_NEW_VDI        0x11
+#define SD_OP_LOCK_VDI       0x12
+#define SD_OP_RELEASE_VDI    0x13
+#define SD_OP_GET_VDI_INFO   0x14
+#define SD_OP_READ_VDIS      0x15
+
+#define SD_FLAG_CMD_WRITE    0x01
+#define SD_FLAG_CMD_COW      0x02
+
+#define SD_RES_SUCCESS       0x00 /* Success */
+#define SD_RES_UNKNOWN       0x01 /* Unknown error */
+#define SD_RES_NO_OBJ        0x02 /* No object found */
+#define SD_RES_EIO           0x03 /* I/O error */
+#define SD_RES_VDI_EXIST     0x04 /* Vdi exists already */
+#define SD_RES_INVALID_PARMS 0x05 /* Invalid parameters */
+#define SD_RES_SYSTEM_ERROR  0x06 /* System error */
+#define SD_RES_VDI_LOCKED    0x07 /* Vdi is locked */
+#define SD_RES_NO_VDI        0x08 /* No vdi found */
+#define SD_RES_NO_BASE_VDI   0x09 /* No base vdi found */
+#define SD_RES_VDI_READ      0x0A /* Cannot read requested vdi */
+#define SD_RES_VDI_WRITE     0x0B /* Cannot write requested vdi */
+#define SD_RES_BASE_VDI_READ 0x0C /* Cannot read base vdi */
+#define SD_RES_BASE_VDI_WRITE   0x0D /* Cannot write base vdi */
+#define SD_RES_NO_TAG        0x0E /* Requested tag is not found */
+#define SD_RES_STARTUP       0x0F /* Sheepdog is on starting up */
+#define SD_RES_VDI_NOT_LOCKED   0x10 /* Vdi is not locked */
+#define SD_RES_SHUTDOWN      0x11 /* Sheepdog is shutting down */
+#define SD_RES_NO_MEM        0x12 /* Cannot allocate memory */
+#define SD_RES_FULL_VDI      0x13 /* we already have the maximum vdis */
+#define SD_RES_VER_MISMATCH  0x14 /* Protocol version mismatch */
+#define SD_RES_NO_SPACE      0x15 /* Server has no room for new objects */
+#define SD_RES_WAIT_FOR_FORMAT  0x16 /* Waiting for a format operation */
+#define SD_RES_WAIT_FOR_JOIN    0x17 /* Waiting for other nodes joining */
+#define SD_RES_JOIN_FAILED   0x18 /* Target node had failed to join sheepdog */
+
+#define VDI_SPACE_SHIFT   32
+#define VDI_BIT (UINT64_C(1) << 63)
+#define VMSTATE_BIT (UINT64_C(1) << 62)
+#define MAX_DATA_OBJS (UINT64_C(1) << 20)
+#define MAX_CHILDREN 1024
+#define SD_MAX_VDI_LEN 256
+#define SD_MAX_VDI_TAG_LEN 256
+#define SD_NR_VDIS   (1U << 24)
+#define SD_DATA_OBJ_SIZE (UINT64_C(1) << 22)
+#define SD_MAX_VDI_SIZE (SD_DATA_OBJ_SIZE * MAX_DATA_OBJS)
+#define SECTOR_SIZE 512
+
+#define SD_INODE_SIZE (sizeof(SheepdogInode))
+#define CURRENT_VDI_ID 0
+
+typedef struct SheepdogReq {
+    uint8_t proto_ver;
+    uint8_t opcode;
+    uint16_t flags;
+    uint32_t epoch;
+    uint32_t id;
+    uint32_t data_length;
+    uint32_t opcode_specific[8];
+} SheepdogReq;
+
+typedef struct SheepdogRsp {
+    uint8_t proto_ver;
+    uint8_t opcode;
+    uint16_t flags;
+    uint32_t epoch;
+    uint32_t id;
+    uint32_t data_length;
+    uint32_t result;
+    uint32_t opcode_specific[7];
+} SheepdogRsp;
+
+typedef struct SheepdogObjReq {
+    uint8_t proto_ver;
+    uint8_t opcode;
+    uint16_t flags;
+    uint32_t epoch;
+    uint32_t id;
+    uint32_t data_length;
+    uint64_t oid;
+    uint64_t cow_oid;
+    uint32_t copies;
+    uint32_t rsvd;
+    uint64_t offset;
+} SheepdogObjReq;
+
+typedef struct SheepdogObjRsp {
+    uint8_t proto_ver;
+    uint8_t opcode;
+    uint16_t flags;
+    uint32_t epoch;
+    uint32_t id;
+    uint32_t data_length;
+    uint32_t result;
+    uint32_t copies;
+    uint32_t pad[6];
+} SheepdogObjRsp;
+
+typedef struct SheepdogVdiReq {
+    uint8_t proto_ver;
+    uint8_t opcode;
+    uint16_t flags;
+    uint32_t epoch;
+    uint32_t id;
+    uint32_t data_length;
+    uint64_t vdi_size;
+    uint32_t base_vdi_id;
+    uint32_t copies;
+    uint32_t snapid;
+    uint32_t pad[3];
+} SheepdogVdiReq;
+
+typedef struct SheepdogVdiRsp {
+    uint8_t proto_ver;
+    uint8_t opcode;
+    uint16_t flags;
+    uint32_t epoch;
+    uint32_t id;
+    uint32_t data_length;
+    uint32_t result;
+    uint32_t rsvd;
+    uint32_t vdi_id;
+    uint32_t pad[5];
+} SheepdogVdiRsp;
+
+typedef struct SheepdogInode {
+    char name[SD_MAX_VDI_LEN];
+    char tag[SD_MAX_VDI_TAG_LEN];
+    uint64_t ctime;
+    uint64_t snap_ctime;
+    uint64_t vm_clock_nsec;
+    uint64_t vdi_size;
+    uint64_t vm_state_size;
+    uint16_t copy_policy;
+    uint8_t nr_copies;
+    uint8_t block_size_shift;
+    uint32_t snap_id;
+    uint32_t vdi_id;
+    uint32_t parent_vdi_id;
+    uint32_t child_vdi_id[MAX_CHILDREN];
+    uint32_t data_vdi_id[MAX_DATA_OBJS];
+} SheepdogInode;
+
+struct sheepdog_access_info {
+	int fd;
+	struct SheepdogInode inode;
+};
+
+int sd_open(struct sheepdog_access_info *ai, char *filename, int flags);
+int sd_io(struct sheepdog_access_info *ai, int write, char *buf,
+	  int len, uint64_t offset);
-- 
1.7.11.2

--
To unsubscribe from this list: send the line "unsubscribe stgt" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html



More information about the stgt mailing list