* Re: [Qemu-devel] [PATCH] Fix iSCSI crash on SG_IO with an iovector
2013-06-22 1:37 ` [Qemu-devel] [PATCH] Fix iSCSI crash on SG_IO with an iovector Ronnie Sahlberg
@ 2013-06-23 10:01 ` Laszlo Ersek
2013-06-25 9:21 ` Stefan Hajnoczi
1 sibling, 0 replies; 4+ messages in thread
From: Laszlo Ersek @ 2013-06-23 10:01 UTC (permalink / raw)
To: Ronnie Sahlberg; +Cc: stefanha, 1191606, qemu-devel, pbonzini
On 06/22/13 03:37, Ronnie Sahlberg wrote:
> Don't assume that SG_IO is always invoked with a simple buffer,
> check the iovec_count and if it is > 1 then we need to pass an array
> of iovectors to libiscsi instead of just a plain buffer.
>
> Signed-off-by: Ronnie Sahlberg <ronniesahlberg@gmail.com>
> ---
> block/iscsi.c | 56 +++++++++++++++++++++++++++++++++++++++++++++++++-------
> 1 files changed, 49 insertions(+), 7 deletions(-)
Looks okay to me, but of course I'm not too familiar with this code.
You (or the maintainer with jurisdiction) might want to change the
commit message:
- check the iovec_count and if it is > 1 then we need to pass an array
+ check the iovec_count and if it is >= 1 then we need to pass an array
But I won't insist on a repost naturally!
I assume you tested the code for both definednesses of
LIBISCSI_FEATURE_IOVECTOR.
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Thanks
Laszlo
>
> diff --git a/block/iscsi.c b/block/iscsi.c
> index 0bbf0b1..cbe2e8f 100644
> --- a/block/iscsi.c
> +++ b/block/iscsi.c
> @@ -651,6 +651,9 @@ iscsi_aio_ioctl_cb(struct iscsi_context *iscsi, int status,
> {
> IscsiAIOCB *acb = opaque;
>
> + g_free(acb->buf);
> + acb->buf = NULL;
> +
> if (acb->canceled != 0) {
> return;
> }
> @@ -727,14 +730,36 @@ static BlockDriverAIOCB *iscsi_aio_ioctl(BlockDriverState *bs,
> memcpy(&acb->task->cdb[0], acb->ioh->cmdp, acb->ioh->cmd_len);
> acb->task->expxferlen = acb->ioh->dxfer_len;
>
> + data.size = 0;
> if (acb->task->xfer_dir == SCSI_XFER_WRITE) {
> - data.data = acb->ioh->dxferp;
> - data.size = acb->ioh->dxfer_len;
> + if (acb->ioh->iovec_count == 0) {
> + data.data = acb->ioh->dxferp;
> + data.size = acb->ioh->dxfer_len;
> + } else {
> +#if defined(LIBISCSI_FEATURE_IOVECTOR)
> + scsi_task_set_iov_out(acb->task,
> + (struct scsi_iovec *) acb->ioh->dxferp,
> + acb->ioh->iovec_count);
> + #else
> + int i;
> + char *buf;
> + struct scsi_iovec *iov = (struct scsi_iovec *)acb->ioh->dxferp;
> +
> + acb->buf = g_malloc(acb->ioh->dxfer_len);
> + buf = acb->buf;
> + for (i = 0; i < acb->ioh->iovec_count; i++) {
> + memcpy(buf, iov[i].iov_base, iov[i].iov_len);
> + buf += iov[i].iov_len;
> + }
> + data.data = acb->buf;
> + data.size = acb->ioh->dxfer_len;
> +#endif
> + }
> }
> +
> if (iscsi_scsi_command_async(iscsi, iscsilun->lun, acb->task,
> iscsi_aio_ioctl_cb,
> - (acb->task->xfer_dir == SCSI_XFER_WRITE) ?
> - &data : NULL,
> + (data.size > 0) ? &data : NULL,
> acb) != 0) {
> scsi_free_scsi_task(acb->task);
> qemu_aio_release(acb);
> @@ -743,9 +768,26 @@ static BlockDriverAIOCB *iscsi_aio_ioctl(BlockDriverState *bs,
>
> /* tell libiscsi to read straight into the buffer we got from ioctl */
> if (acb->task->xfer_dir == SCSI_XFER_READ) {
> - scsi_task_add_data_in_buffer(acb->task,
> - acb->ioh->dxfer_len,
> - acb->ioh->dxferp);
> + if (acb->ioh->iovec_count == 0) {
> + scsi_task_add_data_in_buffer(acb->task,
> + acb->ioh->dxfer_len,
> + acb->ioh->dxferp);
> + } else {
> +#if defined(LIBISCSI_FEATURE_IOVECTOR)
> + scsi_task_set_iov_in(acb->task,
> + (struct scsi_iovec *) acb->ioh->dxferp,
> + acb->ioh->iovec_count);
> +#else
> + int i;
> + for (i = 0; i < acb->ioh->iovec_count; i++) {
> + struct scsi_iovec *iov = (struct scsi_iovec *)acb->ioh->dxferp;
> +
> + scsi_task_add_data_in_buffer(acb->task,
> + iov[i].iov_len,
> + iov[i].iov_base);
> + }
> +#endif
> + }
> }
>
> iscsi_set_events(iscsilun);
>
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [Qemu-devel] [PATCH] Fix iSCSI crash on SG_IO with an iovector
2013-06-22 1:37 ` [Qemu-devel] [PATCH] Fix iSCSI crash on SG_IO with an iovector Ronnie Sahlberg
2013-06-23 10:01 ` Laszlo Ersek
@ 2013-06-25 9:21 ` Stefan Hajnoczi
1 sibling, 0 replies; 4+ messages in thread
From: Stefan Hajnoczi @ 2013-06-25 9:21 UTC (permalink / raw)
To: Ronnie Sahlberg; +Cc: pbonzini, 1191606, lersek, qemu-devel
On Fri, Jun 21, 2013 at 06:37:18PM -0700, Ronnie Sahlberg wrote:
Looks fine except these whitespace issues:
> @@ -727,14 +730,36 @@ static BlockDriverAIOCB *iscsi_aio_ioctl(BlockDriverState *bs,
> memcpy(&acb->task->cdb[0], acb->ioh->cmdp, acb->ioh->cmd_len);
> acb->task->expxferlen = acb->ioh->dxfer_len;
>
> + data.size = 0;
> if (acb->task->xfer_dir == SCSI_XFER_WRITE) {
> - data.data = acb->ioh->dxferp;
> - data.size = acb->ioh->dxfer_len;
> + if (acb->ioh->iovec_count == 0) {
> + data.data = acb->ioh->dxferp;
> + data.size = acb->ioh->dxfer_len;
> + } else {
> +#if defined(LIBISCSI_FEATURE_IOVECTOR)
> + scsi_task_set_iov_out(acb->task,
Indentation should be 4 spaces.
> + (struct scsi_iovec *) acb->ioh->dxferp,
> + acb->ioh->iovec_count);
> + #else
Space before #else?
^ permalink raw reply [flat|nested] 4+ messages in thread