source: trunk/packages/xen-3.1/xen-3.1/tools/blktap/drivers/block-aio.c @ 34

/* block-aio.c
 *
 * libaio-based raw disk implementation.
 *
 * (c) 2006 Andrew Warfield and Julian Chesterfield
 *
 * NB: This code is not thread-safe.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License version 2
 * as published by the Free Software Foundation; or, when distributed
 * separately from the Linux kernel or incorporated into other
 * software packages, subject to the following license:
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this source file (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.
 */


#include <errno.h>
#include <libaio.h>
#include <fcntl.h>
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <sys/statvfs.h>
#include <sys/stat.h>
#include <sys/ioctl.h>
#include <linux/fs.h>
#include "tapdisk.h"


/**
 * We used a kernel patch to return an fd associated with the AIO context
 * so that we can concurrently poll on synchronous and async descriptors.
 * This is signalled by passing 1 as the io context to io_setup.
 */
#define REQUEST_ASYNC_FD 1

#define MAX_AIO_REQS (MAX_REQUESTS * MAX_SEGMENTS_PER_REQ)

struct pending_aio {
        td_callback_t cb;
        int id;
        void *private;
        uint64_t lsec;
};

struct tdaio_state {
        int fd;
        
        /* libaio state */
        io_context_t       aio_ctx;
        struct iocb        iocb_list  [MAX_AIO_REQS];
        struct iocb       *iocb_free  [MAX_AIO_REQS];
        struct pending_aio pending_aio[MAX_AIO_REQS];
        int                iocb_free_count;
        struct iocb       *iocb_queue[MAX_AIO_REQS];
        int                iocb_queued;
        int                poll_fd; /* NB: we require aio_poll support */
        struct io_event    aio_events[MAX_AIO_REQS];
};

#define IOCB_IDX(_s, _io) ((_io) - (_s)->iocb_list)

/*Get Image size, secsize*/
static int get_image_info(struct td_state *s, int fd)
{
        int ret;
        long size;
        unsigned long total_size;
        struct statvfs statBuf;
        struct stat stat;

        ret = fstat(fd, &stat);
        if (ret != 0) {
                DPRINTF("ERROR: fstat failed, Couldn't stat image");
                return -EINVAL;
        }

        if (S_ISBLK(stat.st_mode)) {
                /*Accessing block device directly*/
                s->size = 0;
                if (ioctl(fd,BLKGETSIZE,&s->size)!=0) {
                        DPRINTF("ERR: BLKGETSIZE failed, couldn't stat image");
                        return -EINVAL;
                }

                DPRINTF("Image size: \n\tpre sector_shift  [%llu]\n\tpost "
                        "sector_shift [%llu]\n",
                        (long long unsigned)(s->size << SECTOR_SHIFT),
                        (long long unsigned)s->size);

                /*Get the sector size*/
#if defined(BLKSSZGET)
                {
                        int arg;
                        s->sector_size = DEFAULT_SECTOR_SIZE;
                        ioctl(fd, BLKSSZGET, &s->sector_size);
                        
                        if (s->sector_size != DEFAULT_SECTOR_SIZE)
                                DPRINTF("Note: sector size is %ld (not %d)\n",
                                        s->sector_size, DEFAULT_SECTOR_SIZE);
                }
#else
                s->sector_size = DEFAULT_SECTOR_SIZE;
#endif

        } else {
                /*Local file? try fstat instead*/
                s->size = (stat.st_size >> SECTOR_SHIFT);
                s->sector_size = DEFAULT_SECTOR_SIZE;
                DPRINTF("Image size: \n\tpre sector_shift  [%llu]\n\tpost "
                        "sector_shift [%llu]\n",
                        (long long unsigned)(s->size << SECTOR_SHIFT),
                        (long long unsigned)s->size);
        }

        if (s->size == 0) {             
                s->size =((uint64_t) 16836057);
                s->sector_size = DEFAULT_SECTOR_SIZE;
        }
        s->info = 0;

        return 0;
}

static inline void init_fds(struct disk_driver *dd)
{
        int i;
        struct tdaio_state *prv = (struct tdaio_state *)dd->private;

        for(i = 0; i < MAX_IOFD; i++) 
                dd->io_fd[i] = 0;

        dd->io_fd[0] = prv->poll_fd;
}

/* Open the disk file and initialize aio state. */
int tdaio_open (struct disk_driver *dd, const char *name, td_flag_t flags)
{
        int i, fd, ret = 0, o_flags;
        struct td_state    *s   = dd->td_state;
        struct tdaio_state *prv = (struct tdaio_state *)dd->private;

        DPRINTF("block-aio open('%s')", name);
        /* Initialize AIO */
        prv->iocb_free_count = MAX_AIO_REQS;
        prv->iocb_queued     = 0;
        
        prv->aio_ctx = (io_context_t) REQUEST_ASYNC_FD;
        prv->poll_fd = io_setup(MAX_AIO_REQS, &prv->aio_ctx);

        if (prv->poll_fd < 0) {
                ret = prv->poll_fd;
                if (ret == -EAGAIN) {
                        DPRINTF("Couldn't setup AIO context.  If you are "
                                "trying to concurrently use a large number "
                                "of blktap-based disks, you may need to "
                                "increase the system-wide aio request limit. "
                                "(e.g. 'echo echo 1048576 > /proc/sys/fs/"
                                "aio-max-nr')\n");
                } else {
                        DPRINTF("Couldn't get fd for AIO poll support.  This "
                                "is probably because your kernel does not "
                                "have the aio-poll patch applied.\n");
                }
                goto done;
        }

        for (i=0;i<MAX_AIO_REQS;i++)
                prv->iocb_free[i] = &prv->iocb_list[i];

        /* Open the file */
        o_flags = O_DIRECT | O_LARGEFILE | 
                ((flags == TD_RDONLY) ? O_RDONLY : O_RDWR);
        fd = open(name, o_flags);

        if ( (fd == -1) && (errno == EINVAL) ) {

                /* Maybe O_DIRECT isn't supported. */
                o_flags &= ~O_DIRECT;
                fd = open(name, o_flags);
                if (fd != -1) DPRINTF("WARNING: Accessing image without"
                                     "O_DIRECT! (%s)\n", name);

        } else if (fd != -1) DPRINTF("open(%s) with O_DIRECT\n", name);
        
        if (fd == -1) {
                DPRINTF("Unable to open [%s] (%d)!\n", name, 0 - errno);
                ret = 0 - errno;
                goto done;
        }

        prv->fd = fd;

        init_fds(dd);
        ret = get_image_info(s, fd);

done:
        return ret;     
}

int tdaio_queue_read(struct disk_driver *dd, uint64_t sector,
                     int nb_sectors, char *buf, td_callback_t cb,
                     int id, void *private)
{
        struct   iocb *io;
        struct   pending_aio *pio;
        struct   td_state    *s   = dd->td_state;
        struct   tdaio_state *prv = (struct tdaio_state *)dd->private;
        int      size    = nb_sectors * s->sector_size;
        uint64_t offset  = sector * (uint64_t)s->sector_size;
        long     ioidx;
        
        if (prv->iocb_free_count == 0)
                return -ENOMEM;
        io = prv->iocb_free[--prv->iocb_free_count];
        
        ioidx = IOCB_IDX(prv, io);
        pio = &prv->pending_aio[ioidx];
        pio->cb = cb;
        pio->id = id;
        pio->private = private;
        pio->lsec = sector;
        
        io_prep_pread(io, prv->fd, buf, size, offset);
        io->data = (void *)ioidx;
        
        prv->iocb_queue[prv->iocb_queued++] = io;

        return 0;
}
                        
int tdaio_queue_write(struct disk_driver *dd, uint64_t sector,
                      int nb_sectors, char *buf, td_callback_t cb,
                      int id, void *private)
{
        struct   iocb *io;
        struct   pending_aio *pio;
        struct   td_state    *s   = dd->td_state;
        struct   tdaio_state *prv = (struct tdaio_state *)dd->private;
        int      size    = nb_sectors * s->sector_size;
        uint64_t offset  = sector * (uint64_t)s->sector_size;
        long     ioidx;
        
        if (prv->iocb_free_count == 0)
                return -ENOMEM;
        io = prv->iocb_free[--prv->iocb_free_count];
        
        ioidx = IOCB_IDX(prv, io);
        pio = &prv->pending_aio[ioidx];
        pio->cb = cb;
        pio->id = id;
        pio->private = private;
        pio->lsec = sector;
        
        io_prep_pwrite(io, prv->fd, buf, size, offset);
        io->data = (void *)ioidx;
        
        prv->iocb_queue[prv->iocb_queued++] = io;

        return 0;
}
                        
int tdaio_submit(struct disk_driver *dd)
{
        int ret;
        struct tdaio_state *prv = (struct tdaio_state *)dd->private;

        if (!prv->iocb_queued)
                return 0;

        ret = io_submit(prv->aio_ctx, prv->iocb_queued, prv->iocb_queue);
        
        /* XXX: TODO: Handle error conditions here. */
        
        /* Success case: */
        prv->iocb_queued = 0;
        
        return 0;
}

int tdaio_close(struct disk_driver *dd)
{
        struct tdaio_state *prv = (struct tdaio_state *)dd->private;
        
        io_destroy(prv->aio_ctx);
        close(prv->fd);

        return 0;
}

int tdaio_do_callbacks(struct disk_driver *dd, int sid)
{
        int ret, i, rsp = 0;
        struct io_event *ep;
        struct tdaio_state *prv = (struct tdaio_state *)dd->private;

        /* Non-blocking test for completed io. */
        ret = io_getevents(prv->aio_ctx, 0, MAX_AIO_REQS, prv->aio_events,
                           NULL);
                        
        for (ep=prv->aio_events,i=ret; i-->0; ep++) {
                struct iocb        *io  = ep->obj;
                struct pending_aio *pio;
                
                pio = &prv->pending_aio[(long)io->data];
                rsp += pio->cb(dd, ep->res == io->u.c.nbytes ? 0 : 1,
                               pio->lsec, io->u.c.nbytes >> 9, 
                               pio->id, pio->private);

                prv->iocb_free[prv->iocb_free_count++] = io;
        }
        return rsp;
}

int tdaio_get_parent_id(struct disk_driver *dd, struct disk_id *id)
{
        return TD_NO_PARENT;
}

int tdaio_validate_parent(struct disk_driver *dd, 
                          struct disk_driver *parent, td_flag_t flags)
{
        return -EINVAL;
}

struct tap_disk tapdisk_aio = {
        .disk_type          = "tapdisk_aio",
        .private_data_size  = sizeof(struct tdaio_state),
        .td_open            = tdaio_open,
        .td_queue_read      = tdaio_queue_read,
        .td_queue_write     = tdaio_queue_write,
        .td_submit          = tdaio_submit,
        .td_close           = tdaio_close,
        .td_do_callbacks    = tdaio_do_callbacks,
        .td_get_parent_id   = tdaio_get_parent_id,
        .td_validate_parent = tdaio_validate_parent
};