[ 3.5.y.z extended stable ] Patch "usb-storage: CY7C68300A chips do not support Cypress ATACB" has been added to staging queue

[Luis Henriques]

This is a note to let you know that I have just added a patch titled

    usb-storage: CY7C68300A chips do not support Cypress ATACB

to the linux-3.5.y-queue branch of the 3.5.y.z extended stable tree 
which can be found at:

 http://kernel.ubuntu.com/git?p=ubuntu/linux.git;a=shortlog;h=refs/heads/linux-3.5.y-queue

If you, or anyone else, feels it should not be added to this tree, please 
reply to this email.

For more information about the 3.5.y.z tree, see
https://wiki.ubuntu.com/Kernel/Dev/ExtendedStable

Thanks.
-Luis

------

>From 36c784688602cdc340cfdbb8c424f785b759b16a Mon Sep 17 00:00:00 2001
From: Tormod Volden <debian.tormod at gmail.com>
Date: Sat, 20 Apr 2013 14:24:04 +0200
Subject: [PATCH] usb-storage: CY7C68300A chips do not support Cypress ATACB

commit 671b4b2ba9266cbcfe7210a704e9ea487dcaa988 upstream.

Many cards based on CY7C68300A/B/C use the USB ID 04b4:6830 but only the
B and C variants (EZ-USB AT2LP) support the ATA Command Block
functionality, according to the data sheets. The A variant (EZ-USB AT2)
locks up if ATACB is attempted, until a typical 30 seconds timeout runs
out and a USB reset is performed.

https://bugs.launchpad.net/bugs/428469

It seems that one way to spot a CY7C68300A (at least where the card
manufacturer left Cypress' EEPROM default vaules, against Cypress'
recommendations) is to look at the USB string descriptor indices.

A http://media.digikey.com/pdf/Data%20Sheets/Cypress%20PDFs/CY7C68300A.pdf
B http://www.farnell.com/datasheets/43456.pdf
C http://www.cypress.com/?rID=14189

Note that a CY7C68300B/C chip appears as CY7C68300A if it is running
in Backward Compatibility Mode, and if ATACB would be supported in this
case there is anyway no way to tell which chip it really is.

For 5 years my external USB drive has been locking up for half a minute
when plugged in and ata_id is run by udev, or anytime hdparm or similar
is run on it.

Finally looking at the /correct/ datasheet I think I found the reason. I
am aware the quirk in this patch is a bit hacky, but the hardware
manufacturers haven't made it easy for us.

Signed-off-by: Tormod Volden <debian.tormod at gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh at linuxfoundation.org>
Signed-off-by: Luis Henriques <luis.henriques at canonical.com>
---
 drivers/usb/storage/cypress_atacb.c | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/storage/cypress_atacb.c b/drivers/usb/storage/cypress_atacb.c
index 5fe451d..c1f6751 100644
--- a/drivers/usb/storage/cypress_atacb.c
+++ b/drivers/usb/storage/cypress_atacb.c
@@ -248,14 +248,26 @@ static int cypress_probe(struct usb_interface *intf,
 {
 	struct us_data *us;
 	int result;
+	struct usb_device *device;

 	result = usb_stor_probe1(&us, intf, id,
 			(id - cypress_usb_ids) + cypress_unusual_dev_list);
 	if (result)
 		return result;

-	us->protocol_name = "Transparent SCSI with Cypress ATACB";
-	us->proto_handler = cypress_atacb_passthrough;
+	/* Among CY7C68300 chips, the A revision does not support Cypress ATACB
+	 * Filter out this revision from EEPROM default descriptor values
+	 */
+	device = interface_to_usbdev(intf);
+	if (device->descriptor.iManufacturer != 0x38 ||
+	    device->descriptor.iProduct != 0x4e ||
+	    device->descriptor.iSerialNumber != 0x64) {
+		us->protocol_name = "Transparent SCSI with Cypress ATACB";
+		us->proto_handler = cypress_atacb_passthrough;
+	} else {
+		us->protocol_name = "Transparent SCSI";
+		us->proto_handler = usb_stor_transparent_scsi_command;
+	}

 	result = usb_stor_probe2(us);
 	return result;
--
1.8.1.2