On Mon, 31 Dec 2007 09:59:12 +0800 "javen.wu" <[EMAIL PROTECTED]> wrote:
> Hi Andrew, > > Thanks for your information. > I think the problem caused by "inquiry page83" response always same > in ips driver. > You know IBM serveRAID controller is not a real SCSI controller and > the logic volumes don't recognize SCSI command actually. > But we put the ips driver under Solaris SCSA storage driver stack and > try to prevent the ips driver as a SCSI HBA driver. > Consequently, I faked some SCSI commands response in the ips driver. > "devid" is generated according the response of inquiry page83(GUID). > I guess the problem you met is caused by generating GUID incorrectly > in the ips driver. I need take a look... I thought this might be it. Is there any potential solution to this? I'm not sure what you can base the GUID on... Right now do you just return some constant GUID, or is it based one something? Could you possibly just take into account the track number in addition to whatever's done currently? > ips driver is not open sourced right now because I ported the ips > driver from FreeBSD and I have no legal agreement. I tried to get > legal agreement more than two years ago, but I didn't get that. :( Agreement from Sun or from FreeBSD? I would assume it's under a BSD-3 license, which I couldn't imagine to be a problem, at least from FreeBSD's end. > I don't remember the exact reason why I didn't get legal agreement. > But I'd like try again if you think it's worth to do.(ips is to > support old IBM serveRAID controllers). Well, it wasn't a big problem for me; I certainly wasn't intending on compiling it myself. It just would have perhaps let me know what was going on more quickly, since I would have liked to see how ips was coming up with the devid or GUID that it returns. And I was just a little confused by there being a bullet point for 'source code' on the page for ips, but no link. So, it'd certainly be nice if ips were open, but it's not a big issue and I don't want to create extra unnecessary work for you. -- Andrew Deason [EMAIL PROTECTED] _______________________________________________ driver-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/driver-discuss
