ata pass-through

• Subject: ata pass-through
• From: Joel Reymont <email@hidden>
• Date: Wed, 16 Nov 2011 16:54:08 +0000
• Delivered-to: email@hidden
• Delivered-to: email@hidden
• Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=from:content-type:content-transfer-encoding:subject:date:message-id :to:mime-version:x-mailer; bh=Y+gt3pPd+j+QxVwhWlWXxS7HW8TQwipDT1MkfWfDzrY=; b=CmIPETK8QdD0MDY/kLKfLookv+Jqv0KtRZbSpH0Owu2KnoxGFDcf21hMB86WT9Ke2D X3u9OdKFW+2GQ2+wne8bbrcOh59AxXoqXeLZnIguUyMKOEHxR6bJgjkQxFgkg2rZl4Rb yz9iZ7b37LP9iJHWYBNaQoeqNq8AcPTib0hFI=

I need to implement a driver to send arbitrary ATA commands [1] to an attached device.

Is there sample code that I can start from?

I would subclass IOSCSIPeripheralDeviceType05 for SCSI pass-through, what about ATA?

Is there a logical unit driver or do I need to deal with the protocol service driver?

Should I subclass ATADeviceNub?

What direction should I generally dig in?

	Thanks, Joel

[1] http://www.opensource.apple.com/source/IOATAFamily/IOATAFamily-151.3.1/IOATACommand.h

--------------------------------------------------------------------------
- for hire: mac osx device driver ninja, kernel extensions and usb drivers
---------------------+------------+---------------------------------------
http://wagerlabs.com | @wagerlabs | http://www.linkedin.com/in/joelreymont
---------------------+------------+---------------------------------------

 _______________________________________________
Do not post admin requests to the list. They will be ignored.
Darwin-drivers mailing list      (email@hidden)
Help/Unsubscribe/Update your Subscription:

This email sent to email@hidden

• Prev by Date: I am finally became Boss
• Next by Date: Problems with shutting down user clientsHi all, I have a driver in development for a multiport PCIe HBA.  I have a custom logging application that communicates with one user-client instance for each port. When I invoke kextunload for my driver, my logging application detects a handshake error with one of its user-client instances and attempts to close all user-client connections (assuming that the driver is probably being unloaded).  The CloseUserClient() call for the last user client to be terminated, consistently returns a result of 0x10000003.  I haven't been able to find any reference that explains what this result code means.  Retrying the failed CloseUserClient() call doesn't seem to help.   CloseUserClient() calls for all user-client instances except the last have no problem.  As an experiment, I reversed the order in which I tried to close the instances, and it was still the last one I tried to close that failed. kextunload fails, saying that there are outstanding instances for the user client and the driver.  When I check, I find one user-client instance and all of the driver instances.  I assume that the failure to unload the last user-client instance prevented unloading of any of the driver instances. I have not completely characterized the problem at this point, but I'm making this early post in the hope the information I have will ring a bell with somebody and that I won't need to analyze further. Thanks for any ideas, David Lee
• Previous by thread: I am finally became Boss
• Next by thread: Problems with shutting down user clientsHi all, I have a driver in development for a multiport PCIe HBA.  I have a custom logging application that communicates with one user-client instance for each port. When I invoke kextunload for my driver, my logging application detects a handshake error with one of its user-client instances and attempts to close all user-client connections (assuming that the driver is probably being unloaded).  The CloseUserClient() call for the last user client to be terminated, consistently returns a result of 0x10000003.  I haven't been able to find any reference that explains what this result code means.  Retrying the failed CloseUserClient() call doesn't seem to help.   CloseUserClient() calls for all user-client instances except the last have no problem.  As an experiment, I reversed the order in which I tried to close the instances, and it was still the last one I tried to close that failed. kextunload fails, saying that there are outstanding instances for the user client and the driver.  When I check, I find one user-client instance and all of the driver instances.  I assume that the failure to unload the last user-client instance prevented unloading of any of the driver instances. I have not completely characterized the problem at this point, but I'm making this early post in the hope the information I have will ring a bell with somebody and that I won't need to analyze further. Thanks for any ideas, David Lee
• Index(es):
  • Date
  • Thread