/* This is where we get a request from a caller to read something */
static BlockDriverAIOCB *tar_aio_readv(BlockDriverState *bs,
int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
BlockDriverCompletionFunc *cb, void *opaque)
{
BDRVTarState *s = bs->opaque;
SparseCache *sparse;
int64_t sec_file = sector_num + s->file_sec;
int64_t start = sector_num * SECTOR_SIZE;
int64_t end = start + (nb_sectors * SECTOR_SIZE);
int i;
TarAIOCB *acb;
for (i = 0; i < s->sparse_num; i++) {
sparse = &s->sparse[i];
if (sparse->start > end) {
/* We expect the cache to be start increasing */
break;
} else if ((sparse->start < start) && (sparse->end <= start)) {
/* sparse before our offset */
sec_file -= (sparse->end - sparse->start) / SECTOR_SIZE;
} else if ((sparse->start <= start) && (sparse->end >= end)) {
/* all our sectors are sparse */
char *buf = g_malloc0(nb_sectors * SECTOR_SIZE);
acb = qemu_aio_get(&tar_aiocb_info, bs, cb, opaque);
qemu_iovec_from_buf(qiov, 0, buf, nb_sectors * SECTOR_SIZE);
g_free(buf);
acb->bh = qemu_bh_new(tar_sparse_cb, acb);
qemu_bh_schedule(acb->bh);
return &acb->common;
} else if (((sparse->start >= start) && (sparse->start < end)) ||
((sparse->end >= start) && (sparse->end < end))) {
/* we're semi-sparse (worst case) */
/* let's go synchronous and read all sectors individually */
char *buf = g_malloc(nb_sectors * SECTOR_SIZE);
uint64_t offs;
for (offs = 0; offs < (nb_sectors * SECTOR_SIZE);
offs += SECTOR_SIZE) {
bdrv_pread(bs, (sector_num * SECTOR_SIZE) + offs,
buf + offs, SECTOR_SIZE);
}
qemu_iovec_from_buf(qiov, 0, buf, nb_sectors * SECTOR_SIZE);
acb = qemu_aio_get(&tar_aiocb_info, bs, cb, opaque);
acb->bh = qemu_bh_new(tar_sparse_cb, acb);
qemu_bh_schedule(acb->bh);
return &acb->common;
}
}
return bdrv_aio_readv(s->hd, sec_file, qiov, nb_sectors,
cb, opaque);
}
static int coroutine_fn
cloop_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
QEMUIOVector *qiov, int flags)
{
BDRVCloopState *s = bs->opaque;
uint64_t sector_num = offset >> BDRV_SECTOR_BITS;
int nb_sectors = bytes >> BDRV_SECTOR_BITS;
int ret, i;
assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
qemu_co_mutex_lock(&s->lock);
for (i = 0; i < nb_sectors; i++) {
void *data;
uint32_t sector_offset_in_block =
((sector_num + i) % s->sectors_per_block),
block_num = (sector_num + i) / s->sectors_per_block;
if (cloop_read_block(bs, block_num) != 0) {
ret = -EIO;
goto fail;
}
data = s->uncompressed_block + sector_offset_in_block * 512;
qemu_iovec_from_buf(qiov, i * 512, data, 512);
}
ret = 0;
fail:
qemu_co_mutex_unlock(&s->lock);
return ret;
}
static size_t curl_read_cb(void *ptr, size_t size, size_t nmemb, void *opaque)
{
CURLState *s = ((CURLState*)opaque);
size_t realsize = size * nmemb;
int i;
DPRINTF("CURL: Just reading %zd bytes\n", realsize);
if (!s || !s->orig_buf)
goto read_end;
memcpy(s->orig_buf + s->buf_off, ptr, realsize);
s->buf_off += realsize;
for(i=0; i<CURL_NUM_ACB; i++) {
CURLAIOCB *acb = s->acb[i];
if (!acb)
continue;
if ((s->buf_off >= acb->end)) {
qemu_iovec_from_buf(acb->qiov, 0, s->orig_buf + acb->start,
acb->end - acb->start);
acb->common.cb(acb->common.opaque, 0);
qemu_aio_release(acb);
s->acb[i] = NULL;
}
}
read_end:
return realsize;
}
static void complete_request(struct iocb *iocb, ssize_t ret, void *opaque)
{
VirtIOBlockDataPlane *s = opaque;
VirtIOBlockRequest *req = container_of(iocb, VirtIOBlockRequest, iocb);
struct virtio_blk_inhdr hdr;
int len;
if (likely(ret >= 0)) {
hdr.status = VIRTIO_BLK_S_OK;
len = ret;
} else {
hdr.status = VIRTIO_BLK_S_IOERR;
len = 0;
}
trace_virtio_blk_data_plane_complete_request(s, req->head, ret);
if (req->read_qiov) {
assert(req->bounce_iov);
qemu_iovec_from_buf(req->read_qiov, 0, req->bounce_iov->iov_base, len);
qemu_iovec_destroy(req->read_qiov);
g_slice_free(QEMUIOVector, req->read_qiov);
}
if (req->bounce_iov) {
qemu_vfree(req->bounce_iov->iov_base);
g_slice_free(struct iovec, req->bounce_iov);
}
static int curl_find_buf(BDRVCURLState *s, size_t start, size_t len,
CURLAIOCB *acb)
{
int i;
size_t end = start + len;
for (i=0; i<CURL_NUM_STATES; i++) {
CURLState *state = &s->states[i];
size_t buf_end = (state->buf_start + state->buf_off);
size_t buf_fend = (state->buf_start + state->buf_len);
if (!state->orig_buf)
continue;
if (!state->buf_off)
continue;
// Does the existing buffer cover our section?
if ((start >= state->buf_start) &&
(start <= buf_end) &&
(end >= state->buf_start) &&
(end <= buf_end))
{
char *buf = state->orig_buf + (start - state->buf_start);
qemu_iovec_from_buf(acb->qiov, 0, buf, len);
acb->common.cb(acb->common.opaque, 0);
return FIND_RET_OK;
}
// Wait for unfinished chunks
if (state->in_use &&
(start >= state->buf_start) &&
(start <= buf_fend) &&
(end >= state->buf_start) &&
(end <= buf_fend))
{
int j;
acb->start = start - state->buf_start;
acb->end = acb->start + len;
for (j=0; j<CURL_NUM_ACB; j++) {
if (!state->acb[j]) {
state->acb[j] = acb;
return FIND_RET_WAIT;
}
}
}
}
return FIND_RET_NONE;
}
static void rbd_aio_bh_cb(void *opaque)
{
RBDAIOCB *acb = opaque;
if (acb->cmd == RBD_AIO_READ) {
qemu_iovec_from_buf(acb->qiov, 0, acb->bounce, acb->qiov->size);
}
qemu_vfree(acb->bounce);
acb->common.cb(acb->common.opaque, (acb->ret > 0 ? 0 : acb->ret));
qemu_bh_delete(acb->bh);
acb->bh = NULL;
qemu_aio_release(acb);
}
/*
* This aio completion is being called from rbd_finish_bh() and runs in qemu
* BH context.
*/
static void qemu_rbd_complete_aio(RADOSCB *rcb)
{
RBDAIOCB *acb = rcb->acb;
int64_t r;
r = rcb->ret;
if (acb->cmd != RBD_AIO_READ) {
if (r < 0) {
acb->ret = r;
acb->error = 1;
} else if (!acb->error) {
acb->ret = rcb->size;
}
} else {
if (r < 0) {
qemu_rbd_memset(rcb, 0);
acb->ret = r;
acb->error = 1;
} else if (r < rcb->size) {
qemu_rbd_memset(rcb, r);
if (!acb->error) {
acb->ret = rcb->size;
}
} else if (!acb->error) {
acb->ret = r;
}
}
g_free(rcb);
if (!LIBRBD_USE_IOVEC) {
if (acb->cmd == RBD_AIO_READ) {
qemu_iovec_from_buf(acb->qiov, 0, acb->bounce, acb->qiov->size);
}
qemu_vfree(acb->bounce);
}
acb->common.cb(acb->common.opaque, (acb->ret > 0 ? 0 : acb->ret));
qemu_aio_unref(acb);
}
/*
* This aio completion is being called from rbd_finish_bh() and runs in qemu
* BH context.
*/
static void qemu_rbd_complete_aio(RADOSCB *rcb)
{
RBDAIOCB *acb = rcb->acb;
int64_t r;
r = rcb->ret;
if (acb->cmd != RBD_AIO_READ) {
if (r < 0) {
acb->ret = r;
acb->error = 1;
} else if (!acb->error) {
acb->ret = rcb->size;
}
} else {
if (r < 0) {
memset(rcb->buf, 0, rcb->size);
acb->ret = r;
acb->error = 1;
} else if (r < rcb->size) {
memset(rcb->buf + r, 0, rcb->size - r);
if (!acb->error) {
acb->ret = rcb->size;
}
} else if (!acb->error) {
acb->ret = r;
}
}
g_free(rcb);
if (acb->cmd == RBD_AIO_READ) {
qemu_iovec_from_buf(acb->qiov, 0, acb->bounce, acb->qiov->size);
}
qemu_vfree(acb->bounce);
acb->common.cb(acb->common.opaque, (acb->ret > 0 ? 0 : acb->ret));
acb->status = 0;
if (!acb->cancelled) {
qemu_aio_release(acb);
}
}
static int coroutine_fn
dmg_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
QEMUIOVector *qiov, int flags)
{
BDRVDMGState *s = bs->opaque;
uint64_t sector_num = offset >> BDRV_SECTOR_BITS;
int nb_sectors = bytes >> BDRV_SECTOR_BITS;
int ret, i;
assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
qemu_co_mutex_lock(&s->lock);
for (i = 0; i < nb_sectors; i++) {
uint32_t sector_offset_in_chunk;
void *data;
if (dmg_read_chunk(bs, sector_num + i) != 0) {
ret = -EIO;
goto fail;
}
/* Special case: current chunk is all zeroes. Do not perform a memcpy as
* s->uncompressed_chunk may be too small to cover the large all-zeroes
* section. dmg_read_chunk is called to find s->current_chunk */
if (s->types[s->current_chunk] == 2) { /* all zeroes block entry */
qemu_iovec_memset(qiov, i * 512, 0, 512);
continue;
}
sector_offset_in_chunk = sector_num + i - s->sectors[s->current_chunk];
data = s->uncompressed_chunk + sector_offset_in_chunk * 512;
qemu_iovec_from_buf(qiov, i * 512, data, 512);
}
ret = 0;
fail:
qemu_co_mutex_unlock(&s->lock);
return ret;
}
static void xseg_request_handler(void *state)
{
BDRVArchipelagoState *s = (BDRVArchipelagoState *) state;
void *psd = xseg_get_signal_desc(s->xseg, s->port);
qemu_mutex_lock(&s->request_mutex);
while (!s->stopping) {
struct xseg_request *req;
void *data;
xseg_prepare_wait(s->xseg, s->srcport);
req = xseg_receive(s->xseg, s->srcport, X_NONBLOCK);
if (req) {
AIORequestData *reqdata;
ArchipelagoSegmentedRequest *segreq;
xseg_get_req_data(s->xseg, req, (void **)&reqdata);
switch (reqdata->op) {
case ARCHIP_OP_READ:
data = xseg_get_data(s->xseg, req);
segreq = reqdata->segreq;
segreq->count += req->serviced;
qemu_iovec_from_buf(reqdata->aio_cb->qiov, reqdata->bufidx,
data,
req->serviced);
xseg_put_request(s->xseg, req, s->srcport);
if ((__sync_add_and_fetch(&segreq->ref, -1)) == 0) {
if (!segreq->failed) {
reqdata->aio_cb->ret = segreq->count;
archipelago_finish_aiocb(reqdata);
g_free(segreq);
} else {
g_free(segreq);
g_free(reqdata);
}
} else {
g_free(reqdata);
}
break;
case ARCHIP_OP_WRITE:
case ARCHIP_OP_FLUSH:
segreq = reqdata->segreq;
segreq->count += req->serviced;
xseg_put_request(s->xseg, req, s->srcport);
if ((__sync_add_and_fetch(&segreq->ref, -1)) == 0) {
if (!segreq->failed) {
reqdata->aio_cb->ret = segreq->count;
archipelago_finish_aiocb(reqdata);
g_free(segreq);
} else {
g_free(segreq);
g_free(reqdata);
}
} else {
g_free(reqdata);
}
break;
case ARCHIP_OP_VOLINFO:
s->is_signaled = true;
qemu_cond_signal(&s->archip_cond);
break;
}
} else {
xseg_wait_signal(s->xseg, psd, 100000UL);
}
xseg_cancel_wait(s->xseg, s->srcport);
}
s->th_is_signaled = true;
qemu_cond_signal(&s->request_cond);
qemu_mutex_unlock(&s->request_mutex);
qemu_thread_exit(NULL);
}
static coroutine_fn int
block_crypto_co_readv(BlockDriverState *bs, int64_t sector_num,
int remaining_sectors, QEMUIOVector *qiov)
{
BlockCrypto *crypto = bs->opaque;
int cur_nr_sectors; /* number of sectors in current iteration */
uint64_t bytes_done = 0;
uint8_t *cipher_data = NULL;
QEMUIOVector hd_qiov;
int ret = 0;
size_t payload_offset =
qcrypto_block_get_payload_offset(crypto->block) / 512;
qemu_iovec_init(&hd_qiov, qiov->niov);
/* Bounce buffer so we have a linear mem region for
* entire sector. XXX optimize so we avoid bounce
* buffer in case that qiov->niov == 1
*/
cipher_data =
qemu_try_blockalign(bs->file->bs, MIN(BLOCK_CRYPTO_MAX_SECTORS * 512,
qiov->size));
if (cipher_data == NULL) {
ret = -ENOMEM;
goto cleanup;
}
while (remaining_sectors) {
cur_nr_sectors = remaining_sectors;
if (cur_nr_sectors > BLOCK_CRYPTO_MAX_SECTORS) {
cur_nr_sectors = BLOCK_CRYPTO_MAX_SECTORS;
}
qemu_iovec_reset(&hd_qiov);
qemu_iovec_add(&hd_qiov, cipher_data, cur_nr_sectors * 512);
ret = bdrv_co_readv(bs->file,
payload_offset + sector_num,
cur_nr_sectors, &hd_qiov);
if (ret < 0) {
goto cleanup;
}
if (qcrypto_block_decrypt(crypto->block,
sector_num,
cipher_data, cur_nr_sectors * 512,
NULL) < 0) {
ret = -EIO;
goto cleanup;
}
qemu_iovec_from_buf(qiov, bytes_done,
cipher_data, cur_nr_sectors * 512);
remaining_sectors -= cur_nr_sectors;
sector_num += cur_nr_sectors;
bytes_done += cur_nr_sectors * 512;
}
cleanup:
qemu_iovec_destroy(&hd_qiov);
qemu_vfree(cipher_data);
return ret;
}
