Add rotationrate to geom disk dumpconf

[rpokala]

Summary

This adds struct disk's rotation_rate to the output of geom disk list, which is useful to collect into inventory databases or make decisions on with configuration management software.

Test Plan

[e-mail]:~# geom disk list
Geom name: da0
Providers:

1. Name: da0 Mediasize: 480103981056 (447G) Sectorsize: 512 Stripesize: 4096 Stripeoffset: 0 Mode: r4w4e11 descr: ATA SAMSUNG MZ7WD480 lunid: 5002538500234d15 ident: S1G1NYAF930242 rotationrate: 1 fwsectors: 63 fwheads: 255

[rpokala]

I was actually going to take a look at this one this weekend. My concern is that SSDs report a rotationrate of 1, not 0:

Value	Description
0000h	Rate not reported
0001h	Non-rotating media (e.g., solid state device)
0002h-0400h	Reserved
0401h-FFFEh	Nominal media rotation rate in rotations per minute (rpm) (e.g., 7200 rpm = 1C20h)
FFFFh	Reserved

(That wording is copied from the ATA specs, but SCSI uses the same values for the meaning.)

I was going to validate the d_rotation_rate value before using it. i.e.:

sbuf_printf(sb, "%s<rotationrate>", indent);
if (dp->d_rotatation_rate == 0)
    sbuf_printf(sb, "unknown");
else if (dp->d_rotation_rate == 1)
    sbuf_printf(sb, "0");
else if ((dp->d_rotation_rate >= 0x041) && (dp->d_rotation_rate <= 0xfffe))
    sbuf_printf(sb, "%u", dp->d_rotation_rate);
else
    sbuf_printf(sb, "invalid");
sbuf_printf(sb, "</rotationrate>\n");

[kbowling]

Your logic makes sense to me. Ravi, can you integrate this and MFC to 10-STABLE during slush?

[rpokala]

I'm going to steal this from @kevin.bowling_kev009.com because I have a newer patch, and I can eventually commit it.

[rpokala]

Correctly interpret the d_rotation_rate field, rather than just using the raw value.

[sbruno]

Yup. Go for it.

[jhb]

Just some minor nits.

sys/geom/geom_disk.c

552

style(9) would put a blank line before the comment block.

556

This extra blank line in the comment block can probably be removed.

[rpokala]

Fix some style(9) violations found by jhb.