linux/drivers/base/firmware_class.c, branch v2.6.33.16

firmware_class: fix memory leak - free allocated pages

2010-08-02T17:26:46Z

commit dd336c554d8926c3348a2d5f2a5ef5597f6d1a06 upstream. fix memory leak introduced by the patch 6e03a201bbe: firmware: speed up request_firmware() 1. vfree won't release pages there were allocated explicitly and mapped using vmap. The memory has to be vunmap-ed and the pages needs to be freed explicitly 2. page array is moved into the 'struct firmware' so that we can free it from release_firmware() and not only in fw_dev_release() The fix doesn't break the firmware load speed. Cc: Johannes Berg Cc: Ming Lei Cc: Catalin Marinas Singed-off-by: Kay Sievers Signed-off-by: David Woodhouse Signed-off-by: Tomas Winkler Signed-off-by: Greg Kroah-Hartman

firmware_class: make request_firmware_nowait more useful

2009-12-11T19:24:52Z

Unfortunately, one cannot hold on to the struct firmware that request_firmware_nowait() hands off, which is needed in some cases. Allow this by requiring the callback to free it (via release_firmware). Additionally, give it a gfp_t parameter -- all the current users call it from a GFP_KERNEL context so the GFP_ATOMIC isn't necessary. This also marks an API break which is useful in a sense, although that is obviously not the primary purpose of this change. Signed-off-by: Johannes Berg Acked-by: Marcel Holtmann Cc: Ming Lei Cc: Catalin Marinas Cc: David Woodhouse Cc: Pavel Roskin Cc: Abhay Salunke Signed-off-by: Greg Kroah-Hartman

driver core: firmware_class:fix memory leak of page pointers array

2009-07-28T20:45:22Z

The page pointers array is allocated in fw_realloc_buffer() called by firmware_data_write(), and should be freed in release function of firmware device. Signed-off-by: Ming Lei Reported-by: Catalin Marinas Acked-by: David Woodhouse Signed-off-by: Greg Kroah-Hartman

Firmware: firmware_class, fix lock imbalance

2009-07-12T20:02:10Z

Add omitted unlock in firmware_data_read. Signed-off-by: Jiri Slaby Signed-off-by: Greg Kroah-Hartman

Free struct device in fw_dev_release()

2009-07-08T16:34:08Z

The f_dev in _request_firmware() is allocated via the fw_setup_device() and fw_register_device() calls and its class set to firmware_class (the class release function is fw_dev_release). Commit 6acf70f078ca replaced the kfree(dev) in fw_dev_release() with a put_device() call but my understanding is that the release function is called via put_device -> kobject_put -> kref_put -> koject_release etc. and it should call kfree since it's the last to see this device structure alive. Because of that, the _request_firmware() function on its -ENOENT error path only calls device_unregister(f_dev) which would eventually call fw_dev_release() but there is no kfree (the subsequent put_device call would just make the kref negative). Signed-off-by: Catalin Marinas Acked-by: Cornelia Huck Acked-by: Ming Lei Signed-off-by: Linus Torvalds

driver core: fix gcc 4.3.3 warnings about string literals

2009-06-16T04:30:25Z

This removes the warning: format not a string literal and no format arguments warnings in the driver core that gcc 4.3.3 complains about. Signed-off-by: Greg Kroah-Hartman

firmware: allocate firmware id dynamically

2009-06-16T04:30:24Z

The firmware loader has a statically allocated 30 bytes long string for the firmware id (a.k.a. the firmware file name). There is no reason why we couldnt allocate it dynamically, and avoid having restrictions on the firmware names lengths. Signed-off-by: Samuel Ortiz Cc: Marcel Holtmann Cc: Zhu Yi , Cc: John Linville Signed-off-by: Greg Kroah-Hartman

driver core: fix documentation of request_firmware_nowait

2009-06-16T04:30:24Z

request_firmware_nowait declares it can be called in non-sleep contexts, but kthead_run called by request_firmware_nowait may sleep. So fix its documentation and comment to make callers clear about it. Signed-off-by: Ming Lei Signed-off-by: Greg Kroah-Hartman

driver core: firmware_class: replace kfree(dev) with put_device(dev)

2009-06-16T04:30:24Z

against v2.6.30-rc3-next tree. Signed-off-by: Ming Lei Signed-off-by: Greg Kroah-Hartman

firmware: speed up request_firmware(), v3

2009-05-15T11:51:29Z

Rather than calling vmalloc() repeatedly to grow the firmware image as we receive data from userspace, just allocate and fill individual pages. Then vmap() the whole lot in one go when we're done. A quick test with a 337KiB iwlagn firmware shows the time taken for request_firmware() going from ~32ms to ~5ms after I apply this patch. [v2: define PAGE_KERNEL_RO as PAGE_KERNEL where necessary, use min_t()] [v3: kunmap() takes the struct page *, not the virtual address] Signed-off-by: David Woodhouse Tested-by: Sachin Sant