- 10 Aug, 2010 40 commits
-
-
Andrea Righi authored
If we use the HCD_LOCAL_MEM flag and dma_declare_coherent_memory() to enforce the host controller's local memory utilization we also need to disable native scatter-gather support, otherwise hcd_alloc_coherent() in map_urb_for_dma() is called with urb->transfer_buffer == NULL, that triggers a NULL pointer dereference. We can also consider to add a WARN_ON() and return an error code to better catch this problem in the future. At the moment no driver seems to hit this bug, so I should consider this a low-priority fix. Signed-off-by:
Andrea Righi <arighi@develer.com> Acked-by:
Alan Stern <stern@rowland.harvard.edu> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Kulikov Vasiliy authored
Change sizeof(x) / sizeof(*x) to ARRAY_SIZE(x). Signed-off-by:
Kulikov Vasiliy <segooon@gmail.com> Cc: Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by:
-
Nicolas Kaiser authored
Don't descend to the EARLY_PRINTK_DBGP directory unless it is actually used. Signed-off-by:
Nicolas Kaiser <nikai@nikai.net> Signed-off-by:
-
Nicolas Kaiser authored
Fixed switch, brace, whitespace and spacing coding style issues. Signed-off-by:
Pete Zaitcev <zaitcev@redhat.com> Signed-off-by:
-
Alan Stern authored
This patch (as1386) adds runtime-PM support for PCI-based USB host controllers. By default autosuspend is disallowed; the user must enable it by writing "auto" to the controller's power/control sysfs attribute. Signed-off-by:
-
Alan Stern authored
This patch (as1396) adds code to uhci-hcd to support the vendor-specific wakeup settings found in Intel's ICHx hardware. A couple of unnecessary memory barriers are removed. And the root hub isn't put back into the "suspended" state if power was lost during a system sleep -- there's not much point in doing so because the root hub will be resumed shortly. Signed-off-by:
-
Alan Stern authored
This patch (as1395) adds code to hcd_pci_suspend() for handling wakeup races. This is another general race pattern, similar to the "open vs. unregister" race we're all familiar with. Here, the race is between suspending a device and receiving a wakeup request from one of the device's suspended children. In particular, if a root-hub wakeup is requested at about the same time as the corresponding USB controller is suspended, and if the controller is enabled for wakeup, then the controller should either fail to suspend or else wake right back up again. During system sleep this won't happen very much, especially since host controllers generally aren't enabled for wakeup during sleep. However it is definitely an issue for runtime PM. Something like this will be needed to prevent the controller from autosuspending while waiting for a root-hub resume to take place. (That is, in fact, the common case, for which there is an extra test.) Signed-off-by:
-
Alan Stern authored
This patch (as1394) adds code to ehci-hcd, ohci-hcd, and uhci-hcd for automatically resuming the root hub when the controller is resumed, if the root hub has a wakeup request pending on some port. During resume from system sleep this doesn't matter, because the root hubs will naturally be resumed along with every other device in the system. However it _will_ matter for runtime PM: If the controller is suspended and a remote wakeup request is received then the controller will autoresume, but we need to ensure that the root hub also autoresumes. Otherwise the wakeup request would be ignored, the controller would go back to sleep, and the cycle would repeat a large number of times (I saw this happen before the patch was written). Signed-off-by:
-
Alan Stern authored
This patch (as1385) adds a "do_wakeup" parameter to the pci_suspend method used by PCI-based host controller drivers. ehci-hcd in particular needs to know whether or not to enable wakeup when suspending a controller. Although that information is currently available through device_may_wakeup(), when support is added for runtime suspend this will no longer be true. Signed-off-by:
-
Alan Stern authored
This patch (as1384) moves the resume_common() routine in hcd-pci.c a little higher in the source file to avoid forward references in an upcoming patch. It also replaces the "hibernated" argument with a more general "event" argument, which will be useful when the routine is called during a runtime resume. Signed-off-by:
-
Alan Stern authored
This patch (as1383) takes the powermac-specific code from the PCI HCD glue layer and encapsulates it in its own subroutine. Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Michal Nazarewicz authored
This commit cleans the g_fs gadget hopefully making it more readable. This is achieved by usage of the usb_string_ids_tab() function for batch string IDs registration as well as generalising configuration so that a single routine is used to add each configuration and bind interfaces. As an effect, the code is shorter and has fewer #ifdefs. Moreover, in some circumstances previous code #defined CONFIG_USB_FUNCTIONFS_GENERIC macro to prevent a situation where gadget with no configurations is built. This code removes the #define form source code and achieves the same effect using select in Kconfig. This patch also changes wording and names of the Kconfig options. Signed-off-by:
Michal Nazarewicz <m.nazarewicz@samsung.com> Signed-off-by:
Kyungmin Park <kyungmin.park@samsung.com> Signed-off-by:
-
Alan Stern authored
This patch (as1393) converts several of the single-bit fields in struct usb_hcd to atomic flags. This is for safety's sake; not all CPUs can update bitfield values atomically, and these flags are used in multiple contexts. The flag fields that are set only during registration or removal can remain as they are, since non-atomic accesses at those times will not cause any problems. (Strictly speaking, the authorized_default flag should become atomic as well. I didn't bother with it because it gets changed only via sysfs. It can be done later, if anyone wants.) Signed-off-by:
-
Michal Nazarewicz authored
In may gadgets bind and bind like functions were in a init section as they were only run during initialisation. However, being callback functions they were referenced from structures in “normal” sections. Changing the tag from “__init” to “__ref” fixes the warnings. Signed-off-by:
-
Michal Nazarewicz authored
Added pre_eject() and post_eject() callbacks which are called before and after removable logical unit is ejected. The first can prevent logical unit from being ejected. This commit also changes the way callbacks are passed to the function from gadget. A fsg_operations structure has been created which lists all callbacks -- this is passed to the fsg_config. This is important because it changes the way thread_exits() callback is passed. Signed-off-by:
-
Michal Nazarewicz authored
Added a disconnect() callback to composite devices which is called by composite glue when its disconnect callback is called by gadget. Signed-off-by:
David Brownell <dbrownell@users.sourceforge.net> Signed-off-by:
-
Michal Nazarewicz authored
A short documentation of the g_multi driver along with INF files for Windows XP SP3 are provided. Signed-off-by:
-
Michal Nazarewicz authored
Updated the INF file for g_serial gadget. It should work with most recent Windows systems now. Signed-off-by:
-
Michal Nazarewicz authored
Updated the INF file for the g_ether gadget. It should work with most recent Windows systems now. Signed-off-by:
-
Michal Nazarewicz authored
The Multifunction Composite Gadget have been cleaned up and refactored so hopefully it looks prettier and works at least as good as before changes. A Kconfig has also been fixed to make it impossible to build FunctionFS gadget with no configurations. With this patch, if RNDIS is not chosen by the user CDC is force-selected. Signed-off-by:
-
Michal Nazarewicz authored
The ep0req_name was never used in f_mass_storage hence it may be safely removed from the code. It was a leftover from File Storage Gadget which used it for debug messages. Signed-off-by:
-
Michal Nazarewicz authored
Use usb_string_ids_n() function to simplify string ids registeration. Signed-off-by:
-
Michal Nazarewicz authored
usb_string_ids_tab() and usb_string_ids_n() functions added to the composite framework. The first accepts an array of usb_string object and for each registeres a string id and the second registeres a given number of ids and returns the first. This may simplify string ids registration since gadgets and composite functions won't have to call usb_string_id() several times and each time check for errer status -- all this will be done with a single call. Signed-off-by:
-
Michal Nazarewicz authored
FunctionFS had a bit unique name for function used to add it to USB configuration. Renamed as to match naming convention of other functions. Signed-off-by:
-
Michal Nazarewicz authored
Mass Storage Function had a bit unique name for function used to add it to USB configuration. Renamed as to match naming convention of other functions. Signed-off-by:
-
Michal Nazarewicz authored
This patch changes msg_do_config() function so that it uses a static object for a fsg_common structure instead of dynamically allocated. This is a micro-optimisation. Signed-off-by:
-
Arnd Bergmann authored
compat_ioctl does not use the BKL, so I assume that the native function does not need it either. The open function is already protected by the driver's mutex, the BKL is probably not needed here either. Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Pete Zaitcev <zaitcev@redhat.com> Signed-off-by:
-
Arnd Bergmann authored
There is no gadget driver in the tree that actually implements the ioctl operation, so obviously it is not necessary to hold the BKL around the call. Signed-off-by:
-
Andi Kleen authored
BKL was not really needed, just came from earlier push downs. The only part that's a bit dodgy is the lseek function. Would need another lock or atomic access to fpos on 32bit? Better to have a libfs lseek Signed-off-by:
Andi Kleen <ak@linux.intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Alan Stern <stern@rowland.harvard.edu> Signed-off-by:
-
Andi Kleen authored
The usb serial driver initialization tried to use the BKL to stop driver modules from unloading, but that didn't work anyways. There was already some code to do proper try_module_get, but it was conditional on having a new probe interface. I checked all the low level drivers and they all have proper .owner = THIS_MODULE, so it's ok to always use. The other problem was the usb_serial_driver_list needing protection by a lock. This was broken anyways because unregister did not necessarily have the BKL. I extended the extending table_lock mutex to protect this case too. With these changes the BKL can be removed here. Signed-off-by:
-
Andi Kleen authored
And audit all the users. None needed the BKL. That was easy because there was only very few around. Tested with allmodconfig build on x86-64 Signed-off-by: -
Andi Kleen authored
The code this is attempting to lock against does not use the BKL, so it's not needed. Most likely this code is still broken/racy (Al Viro also thinks so), but removing the BKL should not make it worse than before. Signed-off-by:
-
Nicolas Kaiser authored
Fixed spacing and indentation coding style issues. Signed-off-by:
-
Yann Cantin authored
This patch add a serial number parameter to the g_file_storage module. There's validity checks against the string passed to comply with the specs. Signed-off-by:
Yann Cantin <yann.cantin@laposte.net> Cc: Michał Nazarewicz <m.nazarewicz@samsung.com> Cc: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by:
-
Nicolas Kaiser authored
Fixed spacing and brace coding style issues. Signed-off-by:
-
Joe Perches authored
Signed-off-by:
Joe Perches <joe@perches.com> Signed-off-by:
-
Leann Ogasawara authored
ipaq already had this device id defined: { USB_DEVICE(0x0BB4, 0x00CF) }, /* HTC USB Modem */ Revert the commit which adds the duplicate entry. This reverts commit 04cab132. Originally-by:Ben Collins <ben.collins@canonical.com> Signed-off-by:
Leann Ogasawara <leann.ogasawara@canonical.com> Signed-off-by:
-
Leann Ogasawara authored
AMOI_VENDOR_ID is defined twice. Remove the duplicate entry and move the AMOI_PRODUCT_9508 definition to be grouped with the other AMOI product definitions. Originally-by:
Ben Collins <ben.collins@ubuntu.com> Signed-off-by:
-
Alek Du authored
This patch will enable Per-port event feature defined in EHCI 1.1 addendum. This feature addresses an issue where HCD is currently required to read and parse PORTSC for all enabled root hub ports. With this patch, the overhead will be reduced. Signed-off-by:
Alek Du <alek.du@intel.com> Signed-off-by:
-
Alek Du authored
With this patch, the LPM capable EHCI host controller can put device into L1 sleep state which is a mode that can enter/exit quickly, and reduce power consumption. Signed-off-by:
Jacob Pan <jacob.jun.pan@intel.com> Signed-off-by:
-
