linux-kernel - Re: 2.6.25-rc6: CONFIG_USB

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Tue, 10 Jun 2008 08:50:05 -0700 (PDT)
From:	Linus Torvalds <torvalds@...ux-foundation.org>
To:	Oliver Neukum <oliver@...kum.org>
cc:	Pavel Machek <pavel@...e.cz>, linux-usb@...r.kernel.org,
	Greg KH <gregkh@...e.de>,
	Alan Stern <stern@...land.harvard.edu>,
	Linus Torvalds <torvalds@...uxfoundation.org>,
	Andrew Morton <akpm@...uxfoundation.org>,
	kernel list <linux-kernel@...r.kernel.org>,
	"Rafael J. Wysocki" <rjw@...k.pl>
Subject: Re: 2.6.25-rc6: CONFIG_USB_PERSIST forced on



On Tue, 10 Jun 2008, Oliver Neukum wrote:
> 
> If a hotfix it must be, here's my take. It works for me, but it isn't
> tested well.
> Alan, what do you think?

Hmm. I hate making functions more complex. Especially if it's an area 
where we'd expect it to change in the future (ie I think we should aim for 
taking into account whether a device is actually open or not). So here's a 
cleaned-up alternative of your approach with the whole "is it persistent" 
logic separated out into a function of its own.

Then, some day, if we get back-pointers to open devices, or we get drievr 
hooks to say "am I mounted" or something, we have a logical place to do 
those things.

Is there perhaps already a way to know whether a driver is actually 
*active* or not (ie not just registered, but somebody has then opened it)? 
I guess there isn't. Oh, well.

		Linus

---
 drivers/usb/core/hub.c |   44 ++++++++++++++++++++++++++++++++++++++++++--
 1 files changed, 42 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 8eb4da3..3b0b58e 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -644,6 +644,46 @@ static void hub_stop(struct usb_hub *hub)
 
 #ifdef CONFIG_PM
 
+static int persistent_device(struct usb_device *udev)
+{
+	int i, retval;
+	struct usb_host_config *actconfig;
+
+	/* Explicitly not marked persistent? */
+	if (!udev->persist_enabled)
+		return 0;
+
+	/* No active config? */
+	actconfig = udev->actconfig;
+	if (!actconfig)
+		return 0;
+
+	/* FIXME! We should check whether it's open here or not! */
+
+	/*
+	 * Check that all drivers on the active interface have a
+	 * 'reset_resume' entrypoint
+	 */
+	retval = 0;
+	for (i = 0; i < actconfig->desc.bNumInterfaces; i++) {
+		struct usb_interface *intf;
+		struct usb_driver *driver;
+
+		intf = actconfig->interface[i];
+		driver = to_usb_driver(intf->dev.driver);
+		if (!driver)
+			continue;
+		if (!driver->reset_resume)
+			return 0;
+		/*
+		 * We have at least one driver, and that one
+		 * supports 'reset_resume'
+		 */
+		retval = 1;
+	}
+	return retval;
+}
+
 static void hub_restart(struct usb_hub *hub, int type)
 {
 	struct usb_device *hdev = hub->hdev;
@@ -689,8 +729,8 @@ static void hub_restart(struct usb_hub *hub, int type)
 		 * turn off the various status changes to prevent
 		 * khubd from disconnecting it later.
 		 */
-		if (udev->persist_enabled && status == 0 &&
-				!(portstatus & USB_PORT_STAT_ENABLE)) {
+		if (status == 0 && !(portstatus & USB_PORT_STAT_ENABLE) &&
+		    persistent_device(udev)) {
 			if (portchange & USB_PORT_STAT_C_ENABLE)
 				clear_port_feature(hub->hdev, port1,
 						USB_PORT_FEAT_C_ENABLE);
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/