[GIT PULL] USB fixes for v4.2-rc4

[Felipe Balbi <balbi@ti.com>]

Hi Greg,

A second fixes pull request, seems like there won't be any
more fixes coming through me; either that or I'll be proven
wrong :-)

cheers

The following changes since commit 52721d9d3334c1cb1f76219a161084094ec634dc:

  Linux 4.2-rc3 (2015-07-19 14:45:02 -0700)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb.git tags/fixes-for-v4.2-rc4

for you to fetch changes up to aebda618718157a69c0dc0adb978d69bc2b8723c:

  usb: dwc3: Reset the transfer resource index on SET_INTERFACE (2015-07-22 08:52:42 -0500)

----------------------------------------------------------------
usb: fixes for v4.2-rc4

Only four fixes this time so I'll describe them all.

We have an iomen resource leak fix in mv_udc_core. This bug
exists since v3.3.

Renesas got a fix for how they use dma_map_single() with IOMMU.

The new ulpi bus got an ordering fix, so drivers don't try to
probe ahead of the bus.

And finally, we have a fix for a really old regression with
dwc3, one which could only be exposed by a recent patch from
Subbaraya. Basically, we were startving the controller of
transfer resources.

Signed-off-by: Felipe Balbi <balbi@ti.com>

----------------------------------------------------------------
Alexey Khoroshilov (1):
      usb: gadget: mv_udc_core: fix phy_regs I/O memory leak

John Youn (1):
      usb: dwc3: Reset the transfer resource index on SET_INTERFACE

Lu Baolu (1):
      usb: ulpi: ulpi_init should be executed in subsys_initcall

Yoshihiro Shimoda (1):
      usb: gadget: udc: core: Fix argument of dma_map_single for IOMMU

 drivers/usb/common/ulpi.c            |  2 +-
 drivers/usb/dwc3/ep0.c               |  4 ++++
 drivers/usb/gadget/udc/mv_udc_core.c |  2 +-
 drivers/usb/gadget/udc/udc-core.c    | 14 ++++++++------
 4 files changed, 14 insertions(+), 8 deletions(-)