On 05/06/2017 08:43 AM, Aurelien Jarno wrote:
> Export the rotational qdev property in the IDENTIFY request.
>
> Signed-off-by: Aurelien Jarno <aurel...@aurel32.net>
> ---
> hw/ide/core.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/hw/ide/core.c b/hw/ide/core.c
> index 0b48b64d3a..1aa76b0d90 100644
> --- a/hw/ide/core.c
> +++ b/hw/ide/core.c
> @@ -191,6 +191,9 @@ static void ide_identify(IDEState *s)
> if (dev && dev->conf.discard_granularity) {
> put_le16(p + 169, 1); /* TRIM support */
> }
> + if (dev && !dev->conf.rotational) {
> + put_le16(p + 217, 1); /* non-rotating device */
> + }
>
> ide_identify_size(s);
> s->identify_set = 1;
>
7.16.7.80 Word 217: Nominal media rotation rate
Word 217 is a copy of Nominal Media Rotation Rate (see A.11.5.4).
*flips page like it's a choose-your-own-adventure book*
OK, so 0x0001 just means "Non-rotational." I guess it probably does not
matter that our default is "Rate not reported."
Looks like there's no relevant bits for packet identify data either, so:
Acked-by: John Snow <js...@redhat.com>
A question about the series as a whole, though: if by default rotational
is set to false, we will newly be setting this bit in IDENTIFY data with
identical CLI invocations. Why is that OK?
