The following changes since commit 848a6caa88b9f082c89c9b41afa975761262981d:

Merge tag 'migration-20230602-pull-request' of https://gitlab.com/juan.quintela/qemu into staging (2023-06-02 17:33:29 -0700)

https://gitlab.com/hreitz/qemu.git tags/pull-block-2023-06-05

for you to fetch changes up to 42a2890a76f4783cd1c212f27856edcf2b5e8a75:

qcow2: add discard-no-unref option (2023-06-05 13:15:42 +0200)

Block patches

- Fix padding of unaligned vectored requests to match the host alignment

for vectors with 1023 or 1024 buffers

- Refactor and fix bugs in parallels's image check functionality

- Add an option to the qcow2 driver to retain (qcow2-level) allocations

on discard requests from the guest (while still forwarding the discard

to the lower level and marking the range as zero)

Hanna Czenczek (4):

util/iov: Make qiov_slice() public

block: Collapse padded I/O vecs exceeding IOV_MAX

util/iov: Remove qemu_iovec_init_extended()

iotests/iov-padding: New test

Jean-Louis Dupond (1):

qcow2: add discard-no-unref option

qapi/block-core.json | 12 ++

block/qcow2.h | 3 +

include/qemu/iov.h | 8 +-

block/io.c | 166 ++++++++++++++++++--

block/parallels.c | 190 ++++++++++++++++-------

block/qcow2-cluster.c | 32 +++-

block/qcow2.c | 18 +++

util/iov.c | 89 ++---------

qemu-options.hx | 12 ++

tests/qemu-iotests/tests/iov-padding | 85 ++++++++++

tests/qemu-iotests/tests/iov-padding.out | 59 +++++++

11 files changed, 523 insertions(+), 151 deletions(-)

create mode 100755 tests/qemu-iotests/tests/iov-padding

create mode 100644 tests/qemu-iotests/tests/iov-padding.out

2.40.1

We want to inline qemu_iovec_init_extended() in block/io.c for padding

requests, and having access to qiov_slice() is useful for this. As a

public function, it is renamed to qemu_iovec_slice().

(We will need to count the number of I/O vector elements of a slice

there, and then later process this slice. Without qiov_slice(), we

would need to call qemu_iovec_subvec_niov(), and all further

IOV-processing functions may need to skip prefixing elements to

accomodate for a qiov_offset. Because qemu_iovec_subvec_niov()

internally calls qiov_slice(), we can just have the block/io.c code call

qiov_slice() itself, thus get the number of elements, and also create an

iovec array with the superfluous prefixing elements stripped, so the

following processing functions no longer need to skip them.)

Reviewed-by: Eric Blake <eblake@redhat.com>

Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>

Signed-off-by: Hanna Czenczek <hreitz@redhat.com>

Message-Id: <20230411173418.19549-2-hreitz@redhat.com>

---

include/qemu/iov.h | 3 +++

util/iov.c | 14 +++++++-------

2 files changed, 10 insertions(+), 7 deletions(-)

diff --git a/include/qemu/iov.h b/include/qemu/iov.h

--- a/include/qemu/iov.h

+++ b/include/qemu/iov.h

@@ -XXX,XX +XXX,XX @@ int qemu_iovec_init_extended(

void *tail_buf, size_t tail_len);

void qemu_iovec_init_slice(QEMUIOVector *qiov, QEMUIOVector *source,

size_t offset, size_t len);

+struct iovec *qemu_iovec_slice(QEMUIOVector *qiov,

+ size_t offset, size_t len,

+ size_t *head, size_t *tail, int *niov);

int qemu_iovec_subvec_niov(QEMUIOVector *qiov, size_t offset, size_t len);

void qemu_iovec_add(QEMUIOVector *qiov, void *base, size_t len);

void qemu_iovec_concat(QEMUIOVector *dst,

diff --git a/util/iov.c b/util/iov.c

--- a/util/iov.c

+++ b/util/iov.c

@@ -XXX,XX +XXX,XX @@ static struct iovec *iov_skip_offset(struct iovec *iov, size_t offset,

/*

- * qiov_slice

+ * qemu_iovec_slice

*

* Find subarray of iovec's, containing requested range. @head would

* be offset in first iov (returned by the function), @tail would be

* count of extra bytes in last iovec (returned iov + @niov - 1).

*/

-static struct iovec *qiov_slice(QEMUIOVector *qiov,

- size_t offset, size_t len,

- size_t *head, size_t *tail, int *niov)

+struct iovec *qemu_iovec_slice(QEMUIOVector *qiov,

+ size_t offset, size_t len,

+ size_t *head, size_t *tail, int *niov)

struct iovec *iov, *end_iov;

@@ -XXX,XX +XXX,XX @@ int qemu_iovec_subvec_niov(QEMUIOVector *qiov, size_t offset, size_t len)

size_t head, tail;

int niov;

- qiov_slice(qiov, offset, len, &head, &tail, &niov);

+ qemu_iovec_slice(qiov, offset, len, &head, &tail, &niov);

return niov;

}

@@ -XXX,XX +XXX,XX @@ int qemu_iovec_init_extended(

if (mid_len) {

- mid_iov = qiov_slice(mid_qiov, mid_offset, mid_len,

- &mid_head, &mid_tail, &mid_niov);

+ mid_iov = qemu_iovec_slice(mid_qiov, mid_offset, mid_len,

+ &mid_head, &mid_tail, &mid_niov);

total_niov = !!head_len + mid_niov + !!tail_len;

2.40.1

When processing vectored guest requests that are not aligned to the

storage request alignment, we pad them by adding head and/or tail

buffers for a read-modify-write cycle.

The guest can submit I/O vectors up to IOV_MAX (1024) in length, but

with this padding, the vector can exceed that limit. As of

4c002cef0e9abe7135d7916c51abce47f7fc1ee2 ("util/iov: make

qemu_iovec_init_extended() honest"), we refuse to pad vectors beyond the

limit, instead returning an error to the guest.

To the guest, this appears as a random I/O error. We should not return

an I/O error to the guest when it issued a perfectly valid request.

Before 4c002cef0e9abe7135d7916c51abce47f7fc1ee2, we just made the vector

longer than IOV_MAX, which generally seems to work (because the guest

assumes a smaller alignment than we really have, file-posix's

raw_co_prw() will generally see bdrv_qiov_is_aligned() return false, and

so emulate the request, so that the IOV_MAX does not matter). However,

that does not seem exactly great.

I see two ways to fix this problem:

1. We split such long requests into two requests.

2. We join some elements of the vector into new buffers to make it

shorter.

I am wary of (1), because it seems like it may have unintended side

effects.

(2) on the other hand seems relatively simple to implement, with

hopefully few side effects, so this patch does that.

To do this, the use of qemu_iovec_init_extended() in bdrv_pad_request()

is effectively replaced by the new function bdrv_create_padded_qiov(),

which not only wraps the request IOV with padding head/tail, but also

ensures that the resulting vector will not have more than IOV_MAX

elements. Putting that functionality into qemu_iovec_init_extended() is

infeasible because it requires allocating a bounce buffer; doing so

would require many more parameters (buffer alignment, how to initialize

the buffer, and out parameters like the buffer, its length, and the

original elements), which is not reasonable.

Conversely, it is not difficult to move qemu_iovec_init_extended()'s

functionality into bdrv_create_padded_qiov() by using public

qemu_iovec_* functions, so that is what this patch does.

Because bdrv_pad_request() was the only "serious" user of

qemu_iovec_init_extended(), the next patch will remove the latter

function, so the functionality is not implemented twice.

Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2141964

Signed-off-by: Hanna Czenczek <hreitz@redhat.com>

Message-Id: <20230411173418.19549-3-hreitz@redhat.com>

Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>

block/io.c | 166 ++++++++++++++++++++++++++++++++++++++++++++++++-----

1 file changed, 151 insertions(+), 15 deletions(-)

@@ -XXX,XX +XXX,XX @@ out:

* @merge_reads is true for small requests,

* if @buf_len == @head + bytes + @tail. In this case it is possible that both

* head and tail exist but @buf_len == align and @tail_buf == @buf.

+ *

+ * @write is true for write requests, false for read requests.

+ *

+ * If padding makes the vector too long (exceeding IOV_MAX), then we need to

+ * merge existing vector elements into a single one. @collapse_bounce_buf acts

+ * as the bounce buffer in such cases. @pre_collapse_qiov has the pre-collapse

+ * I/O vector elements so for read requests, the data can be copied back after

+ * the read is done.

*/

typedef struct BdrvRequestPadding {

uint8_t *buf;

@@ -XXX,XX +XXX,XX @@ typedef struct BdrvRequestPadding {

size_t head;

size_t tail;

bool merge_reads;

+ bool write;

QEMUIOVector local_qiov;

+

+ uint8_t *collapse_bounce_buf;

+ size_t collapse_len;

+ QEMUIOVector pre_collapse_qiov;

} BdrvRequestPadding;

static bool bdrv_init_padding(BlockDriverState *bs,

int64_t offset, int64_t bytes,

+ bool write,

BdrvRequestPadding *pad)

{

int64_t align = bs->bl.request_alignment;

@@ -XXX,XX +XXX,XX @@ static bool bdrv_init_padding(BlockDriverState *bs,

pad->tail_buf = pad->buf + pad->buf_len - align;

}

+ pad->write = write;

+

return true;

}

@@ -XXX,XX +XXX,XX @@ zero_mem:

return 0;

}

-static void bdrv_padding_destroy(BdrvRequestPadding *pad)

+/**

+ * Free *pad's associated buffers, and perform any necessary finalization steps.

+ */

+static void bdrv_padding_finalize(BdrvRequestPadding *pad)

{

+ if (pad->collapse_bounce_buf) {

+ if (!pad->write) {

+ /*

+ * If padding required elements in the vector to be collapsed into a

+ * bounce buffer, copy the bounce buffer content back

+ */

+ qemu_iovec_from_buf(&pad->pre_collapse_qiov, 0,

+ pad->collapse_bounce_buf, pad->collapse_len);

+ }

+ qemu_vfree(pad->collapse_bounce_buf);

+ qemu_iovec_destroy(&pad->pre_collapse_qiov);

+ }

if (pad->buf) {

qemu_vfree(pad->buf);

qemu_iovec_destroy(&pad->local_qiov);

@@ -XXX,XX +XXX,XX @@ static void bdrv_padding_destroy(BdrvRequestPadding *pad)

memset(pad, 0, sizeof(*pad));

}

+/*

+ * Create pad->local_qiov by wrapping @iov in the padding head and tail, while

+ * ensuring that the resulting vector will not exceed IOV_MAX elements.

+ *

+ * To ensure this, when necessary, the first two or three elements of @iov are

+ * merged into pad->collapse_bounce_buf and replaced by a reference to that

+ * bounce buffer in pad->local_qiov.

+ *

+ * After performing a read request, the data from the bounce buffer must be

+ * copied back into pad->pre_collapse_qiov (e.g. by bdrv_padding_finalize()).

+ */

+static int bdrv_create_padded_qiov(BlockDriverState *bs,

+ BdrvRequestPadding *pad,

+ struct iovec *iov, int niov,

+ size_t iov_offset, size_t bytes)

+ int padded_niov, surplus_count, collapse_count;

+

+ /* Assert this invariant */

+ assert(niov <= IOV_MAX);

+

+ /*

+ * Cannot pad if resulting length would exceed SIZE_MAX. Returning an error

+ * to the guest is not ideal, but there is little else we can do. At least

+ * this will practically never happen on 64-bit systems.

+ */

+ if (SIZE_MAX - pad->head < bytes ||

+ SIZE_MAX - pad->head - bytes < pad->tail)

+ {

+ return -EINVAL;

+ }

+

+ /* Length of the resulting IOV if we just concatenated everything */

+ padded_niov = !!pad->head + niov + !!pad->tail;

+

+ qemu_iovec_init(&pad->local_qiov, MIN(padded_niov, IOV_MAX));

+

+ if (pad->head) {

+ qemu_iovec_add(&pad->local_qiov, pad->buf, pad->head);

+ }

+

+ /*

+ * If padded_niov > IOV_MAX, we cannot just concatenate everything.

+ * Instead, merge the first two or three elements of @iov to reduce the

+ * number of vector elements as necessary.

+ */

+ if (padded_niov > IOV_MAX) {

+ /*

+ * Only head and tail can have lead to the number of entries exceeding

+ * IOV_MAX, so we can exceed it by the head and tail at most. We need

+ * to reduce the number of elements by `surplus_count`, so we merge that

+ * many elements plus one into one element.

+ */

+ surplus_count = padded_niov - IOV_MAX;

+ assert(surplus_count <= !!pad->head + !!pad->tail);

+ collapse_count = surplus_count + 1;

+

+ /*

+ * Move the elements to collapse into `pad->pre_collapse_qiov`, then

+ * advance `iov` (and associated variables) by those elements.

+ */

+ qemu_iovec_init(&pad->pre_collapse_qiov, collapse_count);

+ qemu_iovec_concat_iov(&pad->pre_collapse_qiov, iov,

+ collapse_count, iov_offset, SIZE_MAX);

+ iov += collapse_count;

+ iov_offset = 0;

+ niov -= collapse_count;

+ bytes -= pad->pre_collapse_qiov.size;

+

+ /*

+ * Construct the bounce buffer to match the length of the to-collapse

+ * vector elements, and for write requests, initialize it with the data

+ * from those elements. Then add it to `pad->local_qiov`.

+ */

+ pad->collapse_len = pad->pre_collapse_qiov.size;

+ pad->collapse_bounce_buf = qemu_blockalign(bs, pad->collapse_len);

+ if (pad->write) {

+ qemu_iovec_to_buf(&pad->pre_collapse_qiov, 0,

+ pad->collapse_bounce_buf, pad->collapse_len);

+ }

+ qemu_iovec_add(&pad->local_qiov,

+ pad->collapse_bounce_buf, pad->collapse_len);

+ }

+

+ qemu_iovec_concat_iov(&pad->local_qiov, iov, niov, iov_offset, bytes);

+

+ if (pad->tail) {

+ qemu_iovec_add(&pad->local_qiov,

+ pad->buf + pad->buf_len - pad->tail, pad->tail);

+ }

+

+ assert(pad->local_qiov.niov == MIN(padded_niov, IOV_MAX));

+ return 0;

* bdrv_pad_request

*

@@ -XXX,XX +XXX,XX @@ static void bdrv_padding_destroy(BdrvRequestPadding *pad)

* read of padding, bdrv_padding_rmw_read() should be called separately if

* needed.

*

+ * @write is true for write requests, false for read requests.

+ *

* Request parameters (@qiov, &qiov_offset, &offset, &bytes) are in-out:

* - on function start they represent original request

* - on failure or when padding is not needed they are unchanged

@@ -XXX,XX +XXX,XX @@ static void bdrv_padding_destroy(BdrvRequestPadding *pad)

static int bdrv_pad_request(BlockDriverState *bs,

QEMUIOVector **qiov, size_t *qiov_offset,

int64_t *offset, int64_t *bytes,

+ bool write,

BdrvRequestPadding *pad, bool *padded,

BdrvRequestFlags *flags)

+ struct iovec *sliced_iov;

+ int sliced_niov;

+ size_t sliced_head, sliced_tail;

bdrv_check_qiov_request(*offset, *bytes, *qiov, *qiov_offset, &error_abort);

- if (!bdrv_init_padding(bs, *offset, *bytes, pad)) {

+ if (!bdrv_init_padding(bs, *offset, *bytes, write, pad)) {

if (padded) {

*padded = false;

}

return 0;

}

- ret = qemu_iovec_init_extended(&pad->local_qiov, pad->buf, pad->head,

- *qiov, *qiov_offset, *bytes,

- pad->buf + pad->buf_len - pad->tail,

- pad->tail);

+ sliced_iov = qemu_iovec_slice(*qiov, *qiov_offset, *bytes,

+ &sliced_head, &sliced_tail,

+ &sliced_niov);

+

+ /* Guaranteed by bdrv_check_qiov_request() */

+ assert(*bytes <= SIZE_MAX);

+ ret = bdrv_create_padded_qiov(bs, pad, sliced_iov, sliced_niov,

+ sliced_head, *bytes);

*bytes += pad->head + pad->tail;

@@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_preadv_part(BdrvChild *child,

flags |= BDRV_REQ_COPY_ON_READ;

}

- ret = bdrv_pad_request(bs, &qiov, &qiov_offset, &offset, &bytes, &pad,

- NULL, &flags);

+ ret = bdrv_pad_request(bs, &qiov, &qiov_offset, &offset, &bytes, false,

+ &pad, NULL, &flags);

goto fail;

}

@@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_preadv_part(BdrvChild *child,

bs->bl.request_alignment,

qiov, qiov_offset, flags);

tracked_request_end(&req);

- bdrv_padding_destroy(&pad);

+ bdrv_padding_finalize(&pad);

fail:

bdrv_dec_in_flight(bs);

@@ -XXX,XX +XXX,XX @@ bdrv_co_do_zero_pwritev(BdrvChild *child, int64_t offset, int64_t bytes,

/* This flag doesn't make sense for padding or zero writes */

flags &= ~BDRV_REQ_REGISTERED_BUF;

- padding = bdrv_init_padding(bs, offset, bytes, &pad);

+ padding = bdrv_init_padding(bs, offset, bytes, true, &pad);

if (padding) {

assert(!(flags & BDRV_REQ_NO_WAIT));

bdrv_make_request_serialising(req, align);

@@ -XXX,XX +XXX,XX @@ bdrv_co_do_zero_pwritev(BdrvChild *child, int64_t offset, int64_t bytes,

}

out:

- bdrv_padding_destroy(&pad);

+ bdrv_padding_finalize(&pad);

@@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_pwritev_part(BdrvChild *child,

* bdrv_co_do_zero_pwritev() does aligning by itself, so, we do

* alignment only if there is no ZERO flag.

*/

- ret = bdrv_pad_request(bs, &qiov, &qiov_offset, &offset, &bytes, &pad,

- &padded, &flags);

+ ret = bdrv_pad_request(bs, &qiov, &qiov_offset, &offset, &bytes, true,

+ &pad, &padded, &flags);

if (ret < 0) {

return ret;

}

@@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_pwritev_part(BdrvChild *child,

ret = bdrv_aligned_pwritev(child, &req, offset, bytes, align,

qiov, qiov_offset, flags);

- bdrv_padding_destroy(&pad);

+ bdrv_padding_finalize(&pad);

out:

tracked_request_end(&req);

2.40.1

bdrv_pad_request() was the main user of qemu_iovec_init_extended().

HEAD^ has removed that use, so we can remove qemu_iovec_init_extended()

now.

The only remaining user is qemu_iovec_init_slice(), which can easily

inline the small part it really needs.

Note that qemu_iovec_init_extended() offered a memcpy() optimization to

initialize the new I/O vector. qemu_iovec_concat_iov(), which is used

to replace its functionality, does not, but calls qemu_iovec_add() for

every single element. If we decide this optimization was important, we

will need to re-implement it in qemu_iovec_concat_iov(), which might

also benefit its pre-existing users.

Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>

Signed-off-by: Hanna Czenczek <hreitz@redhat.com>

Message-Id: <20230411173418.19549-4-hreitz@redhat.com>

include/qemu/iov.h | 5 ---

util/iov.c | 79 +++++++---------------------------------------

2 files changed, 11 insertions(+), 73 deletions(-)

diff --git a/include/qemu/iov.h b/include/qemu/iov.h

--- a/include/qemu/iov.h

+++ b/include/qemu/iov.h

@@ -XXX,XX +XXX,XX @@ static inline void *qemu_iovec_buf(QEMUIOVector *qiov)

void qemu_iovec_init(QEMUIOVector *qiov, int alloc_hint);

void qemu_iovec_init_external(QEMUIOVector *qiov, struct iovec *iov, int niov);

-int qemu_iovec_init_extended(

- QEMUIOVector *qiov,

- void *head_buf, size_t head_len,

- QEMUIOVector *mid_qiov, size_t mid_offset, size_t mid_len,

- void *tail_buf, size_t tail_len);

void qemu_iovec_init_slice(QEMUIOVector *qiov, QEMUIOVector *source,

size_t offset, size_t len);

struct iovec *qemu_iovec_slice(QEMUIOVector *qiov,

diff --git a/util/iov.c b/util/iov.c

--- a/util/iov.c

+++ b/util/iov.c

@@ -XXX,XX +XXX,XX @@ int qemu_iovec_subvec_niov(QEMUIOVector *qiov, size_t offset, size_t len)

return niov;

-/*

- * Compile new iovec, combining @head_buf buffer, sub-qiov of @mid_qiov,

- * and @tail_buf buffer into new qiov.

- */

-int qemu_iovec_init_extended(

- QEMUIOVector *qiov,

- void *head_buf, size_t head_len,

- QEMUIOVector *mid_qiov, size_t mid_offset, size_t mid_len,

- void *tail_buf, size_t tail_len)

-{

- size_t mid_head, mid_tail;

- int total_niov, mid_niov = 0;

- struct iovec *p, *mid_iov = NULL;

- assert(mid_qiov->niov <= IOV_MAX);

- if (SIZE_MAX - head_len < mid_len ||

- SIZE_MAX - head_len - mid_len < tail_len)

- {

- return -EINVAL;

- if (mid_len) {

- mid_iov = qemu_iovec_slice(mid_qiov, mid_offset, mid_len,

- &mid_head, &mid_tail, &mid_niov);

- }

-

- total_niov = !!head_len + mid_niov + !!tail_len;

- if (total_niov > IOV_MAX) {

- return -EINVAL;

- }

-

- if (total_niov == 1) {

- qemu_iovec_init_buf(qiov, NULL, 0);

- p = &qiov->local_iov;

- } else {

- qiov->niov = qiov->nalloc = total_niov;

- qiov->size = head_len + mid_len + tail_len;

- p = qiov->iov = g_new(struct iovec, qiov->niov);

- }

-

- if (head_len) {

- p->iov_base = head_buf;

- p->iov_len = head_len;

- p++;

- }

-

- assert(!mid_niov == !mid_len);

- if (mid_niov) {

- memcpy(p, mid_iov, mid_niov * sizeof(*p));

- p[0].iov_base = (uint8_t *)p[0].iov_base + mid_head;

- p[0].iov_len -= mid_head;

- p[mid_niov - 1].iov_len -= mid_tail;

- p += mid_niov;

- }

-

- if (tail_len) {

- p->iov_base = tail_buf;

- p->iov_len = tail_len;

- }

-

- return 0;

* Check if the contents of subrange of qiov data is all zeroes.

*/

@@ -XXX,XX +XXX,XX @@ bool qemu_iovec_is_zero(QEMUIOVector *qiov, size_t offset, size_t bytes)

void qemu_iovec_init_slice(QEMUIOVector *qiov, QEMUIOVector *source,

size_t offset, size_t len)

{

- int ret;

+ struct iovec *slice_iov;

+ int slice_niov;

+ size_t slice_head, slice_tail;

assert(source->size >= len);

assert(source->size - len >= offset);

- /* We shrink the request, so we can't overflow neither size_t nor MAX_IOV */

- ret = qemu_iovec_init_extended(qiov, NULL, 0, source, offset, len, NULL, 0);

- assert(ret == 0);

+ slice_iov = qemu_iovec_slice(source, offset, len,

+ &slice_head, &slice_tail, &slice_niov);

+ if (slice_niov == 1) {

+ qemu_iovec_init_buf(qiov, slice_iov[0].iov_base + slice_head, len);

+ } else {

+ qemu_iovec_init(qiov, slice_niov);

+ qemu_iovec_concat_iov(qiov, slice_iov, slice_niov, slice_head, len);

+ }

void qemu_iovec_destroy(QEMUIOVector *qiov)

2.40.1

Test that even vectored IO requests with 1024 vector elements that are

not aligned to the device's request alignment will succeed.

Reviewed-by: Eric Blake <eblake@redhat.com>

Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>

Signed-off-by: Hanna Czenczek <hreitz@redhat.com>

Message-Id: <20230411173418.19549-5-hreitz@redhat.com>

---

tests/qemu-iotests/tests/iov-padding | 85 ++++++++++++++++++++++++

tests/qemu-iotests/tests/iov-padding.out | 59 ++++++++++++++++

2 files changed, 144 insertions(+)

create mode 100755 tests/qemu-iotests/tests/iov-padding

create mode 100644 tests/qemu-iotests/tests/iov-padding.out

diff --git a/tests/qemu-iotests/tests/iov-padding b/tests/qemu-iotests/tests/iov-padding

new file mode 100755

index XXXXXXX..XXXXXXX

--- /dev/null

+++ b/tests/qemu-iotests/tests/iov-padding

@@ -XXX,XX +XXX,XX @@

+#!/usr/bin/env bash

+# group: rw quick

+#

+# Check the interaction of request padding (to fit alignment restrictions) with

+# vectored I/O from the guest

+#

+# Copyright Red Hat

+#

+# 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; either version 2 of the License, or

+# (at your option) any later version.

+#

+# 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, see <http://www.gnu.org/licenses/>.

+#

+

+seq=$(basename $0)

+echo "QA output created by $seq"

+

+status=1    # failure is the default!

+

+_cleanup()

+{

+ _cleanup_test_img

+}

+trap "_cleanup; exit \$status" 0 1 2 3 15

+

+# get standard environment, filters and checks

+cd ..

+. ./common.rc

+. ./common.filter

+