Re: [Qemu-devel] [PATCH 3/4] atapi: implement READ DISC INFORMATION

[Kevin Wolf <kwolf@redhat.com>]

Am 11.06.2012 07:24, schrieb Paolo Bonzini:
> This command is not necessary for CD-ROM and DVD-ROM, but some versions of
> udev trip on its absence.
> 
> Cc: Kevin Wolf <kwolf@redhat.com>
> Cc: Markus Armbruster <armbru@redhat.com>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
> ---
>         Kevin, ok to merge these ATAPI changes via the SCSI tree?
> 
>  hw/ide/atapi.c |   31 +++++++++++++++++++++++++++++++
>  1 file changed, 31 insertions(+)
> 
> diff --git a/hw/ide/atapi.c b/hw/ide/atapi.c
> index 5919cf5..7510886 100644
> --- a/hw/ide/atapi.c
> +++ b/hw/ide/atapi.c
> @@ -956,6 +956,36 @@ static void cmd_read_cdvd_capacity(IDEState *s, uint8_t* buf)
>      ide_atapi_cmd_reply(s, 8, 8);
>  }
>  
> +static void cmd_read_disc_information(IDEState *s, uint8_t* buf)
> +{
> +    uint8_t type = buf[1] & 7;
> +    uint32_t max_len = ube16_to_cpu(buf + 7);
> +
> +    /* Types 1/2 are only defined for Blu-Ray.  */
> +    if (type != 0) {
> +        ide_atapi_cmd_error(s, ILLEGAL_REQUEST,
> +                            ASC_INV_FIELD_IN_CMD_PACKET);
> +        return;
> +    }
> +
> +    memset(buf, 0, 34);
> +    buf[1] = 32;
> +    buf[2] = 0xe; /* last session complete, disc finalized */
> +    buf[3] = 1;   /* first track on disc */
> +    buf[4] = 1;   /* # of sessions */
> +    buf[5] = 1;   /* first track of last session */
> +    buf[6] = 1;   /* last track of last session */
> +    buf[7] = 0x20; /* unrestricted use */
> +    buf[8] = 0x00; /* CD-ROM or DVD-ROM */
> +    /* 9-10-11: most significant byte corresponding bytes 4-5-6 */
> +    /* 12-23: not meaningful for CD-ROM or DVD-ROM */
> +    /* 24-31: disc bar code */
> +    /* 32: disc application code */
> +    /* 33: number of OPC tables */
> +
> +    ide_atapi_cmd_reply(s, 34, max_len);
> +}

Looks correct to me.

> +
>  static void cmd_read_dvd_structure(IDEState *s, uint8_t* buf)
>  {
>      int max_len;

> @@ -1048,6 +1078,7 @@ static const struct {
>      [ 0x5a ] = { cmd_mode_sense, /* (10) */         0 },
>      [ 0xa8 ] = { cmd_read, /* (12) */               CHECK_READY },
>      [ 0xad ] = { cmd_read_dvd_structure,            CHECK_READY },
> +    [ 0x51 ] = { cmd_read_disc_information,         CHECK_READY },
>      [ 0xbb ] = { cmd_set_speed,                     0 },
>      [ 0xbd ] = { cmd_mechanism_status,              0 },
>      [ 0xbe ] = { cmd_read_cd,                       CHECK_READY },

Please keep the table sorted.

Also, if I was the SCSI maintainer, I would require qtest cases for all
four patches. ;-)

Kevin