On 07/15/2015 02:52 PM, Matthew Rosato wrote:
> Currently, scsi_get_configuration always returns a current
> profile (DVD or CD), even when there is actually no media present.
> By comparison, ide/atapi uses a default profile of 0 (MMC_PROFILE_NONE)
> for this case and checks for tray_open, so let's do the same for scsi.
>
> This fixes a problem I'm seeing with Fedora 22 guests where systemd
> cdrom_id fails to unmount after a QEMU-initiated eject against a
> scsi cdrom device because it believes the media is still present
> (but unreadable).
>
> Signed-off-by: Matthew Rosato <mjros...@linux.vnet.ibm.com>
Ping
> ---
> hw/scsi/scsi-disk.c | 16 +++++++++++++++-
> 1 file changed, 15 insertions(+), 1 deletion(-)
>
> diff --git a/hw/scsi/scsi-disk.c b/hw/scsi/scsi-disk.c
> index 54d71f4..64f0694 100644
> --- a/hw/scsi/scsi-disk.c
> +++ b/hw/scsi/scsi-disk.c
> @@ -765,6 +765,9 @@ static inline bool media_is_dvd(SCSIDiskState *s)
> if (!blk_is_inserted(s->qdev.conf.blk)) {
> return false;
> }
> + if (s->tray_open) {
> + return false;
> + }
> blk_get_geometry(s->qdev.conf.blk, &nb_sectors);
> return nb_sectors > CD_MAX_SECTORS;
> }
> @@ -778,6 +781,9 @@ static inline bool media_is_cd(SCSIDiskState *s)
> if (!blk_is_inserted(s->qdev.conf.blk)) {
> return false;
> }
> + if (s->tray_open) {
> + return false;
> + }
> blk_get_geometry(s->qdev.conf.blk, &nb_sectors);
> return nb_sectors <= CD_MAX_SECTORS;
> }
> @@ -975,7 +981,15 @@ static int scsi_get_configuration(SCSIDiskState *s,
> uint8_t *outbuf)
> if (s->qdev.type != TYPE_ROM) {
> return -1;
> }
> - current = media_is_dvd(s) ? MMC_PROFILE_DVD_ROM : MMC_PROFILE_CD_ROM;
> +
> + if (media_is_dvd(s)) {
> + current = MMC_PROFILE_DVD_ROM;
> + } else if (media_is_cd(s)) {
> + current = MMC_PROFILE_CD_ROM;
> + } else {
> + current = MMC_PROFILE_NONE;
> + }
> +
> memset(outbuf, 0, 40);
> stl_be_p(&outbuf[0], 36); /* Bytes after the data length field */
> stw_be_p(&outbuf[6], current);
>
