On 03/13/2012 01:06 PM, Kevin Wolf wrote:
Am 13.03.2012 11:29, schrieb Stefan Hajnoczi:
On Mon, Mar 12, 2012 at 8:05 PM, Floris Bos<b...@je-eigen-domein.nl> wrote:
diff --git a/hw/ide/core.c b/hw/ide/core.c
index 3e50c52..b48e5c2 100644
--- a/hw/ide/core.c
+++ b/hw/ide/core.c
@@ -166,6 +166,13 @@ static void ide_identify(IDEState *s)
if (dev&& dev->conf.discard_granularity) {
put_le16(p + 169, 1); /* TRIM support */
}
+
+ if (s->wwn) {
+ put_le16(p + 108, s->wwn>> 48);
+ put_le16(p + 109, s->wwn>> 32);
+ put_le16(p + 110, s->wwn>> 16);
+ put_le16(p + 111, s->wwn);
+ }
Little-endian 16-bit seems weird at first but the spec requires it, so
it's fine. A comment would be nice.
ATA8-ACS says:
"Bit 8 of word 84 shall be set to one indicating the mandatory World
Wide Name in words 108-111 is supported."
I think we're missing this.
And this:
"Bit 8 of word 87 is a copy of word 84 bit 8."
Otherwise the series looks good to me.
Kevin
Oops, you are right, overlooked those bits.
Linux only seems to care that the WWN starts with a 5, and doesn't look
at the bits, so it didn't turn up in my testing either.
Will send a fixed patch.
Yours sincerely,
Floris Bos
