Previously I had made the struct device point to the input device, but
after talking with Dmitry, he said that the USB device would make more
sense for this driver to point to. So converted it to use that instead.
CC: Alessandro Rubini <rubini@ipvvis.unipv.it>
CC: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Previously I had made the struct device point to the input device, but
after talking with Dmitry, he said that the USB device would make more
sense for this driver to point to. So converted it to use that instead.
CC: Henk Vergonet <Henk.Vergonet@gmail.com>
CC: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Previously I had made the struct device point to the input device, but
after talking with Dmitry, he said that the USB device would make more
sense for this driver to point to. So converted it to use that instead.
CC: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Previously I had made the struct device point to the input device, but
after talking with Dmitry, he said that the USB device would make more
sense for this driver to point to. So converted it to use that instead.
CC: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Previously I had made the struct device point to the input device, but
after talking with Dmitry, he said that the USB device would make more
sense for this driver to point to. So converted it to use that instead.
CC: Dmitry Torokhov <dmitry.torokhov@gmail.com>
CC: Axel Lin <axel.lin@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Previously I had made the struct device point to the input device, but
after talking with Dmitry, he said that the USB device would make more
sense for this driver to point to. So converted it to use that instead.
CC: Dmitry Torokhov <dmitry.torokhov@gmail.com>
CC: "Magnus Hörlin" <magnus@alefors.se>
CC: Rusty Russell <rusty@rustcorp.com.au>
CC: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Previously I had made the struct device point to the input device, but
after talking with Dmitry, he said that the USB device would make more
sense for this driver to point to. So converted it to use that instead.
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The normal read_seqcount_begin() function will wait for any current
writers to exit their critical region by looping until the sequence
count is even.
That "wait for sequence count to stabilize" is the right thing to do if
the read-locker will just retry the whole operation on contention: no
point in doing a potentially expensive reader sequence if we know at the
beginning that we'll just end up re-doing it all.
HOWEVER. Some users don't actually retry the operation, but instead
will abort and do the operation with proper locking. So the sequence
count case may be the optimistic quick case, but in the presense of
writers you may want to do full locking in order to guarantee forward
progress. The prime example of this would be the RCU name lookup.
And in that case, you may well be better off without the "retry early",
and are in a rush to instead get to the failure handling. Thus this
"raw" interface that just returns the sequence number without testing it
- it just forces the low bit to zero so that read_seqcount_retry() will
always fail such a "active concurrent writer" scenario.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We really need to use a ACCESS_ONCE() on the sequence value read in
__read_seqcount_begin(), because otherwise the compiler might end up
reloading the value in between the test and the return of it. As a
result, it might end up returning an odd value (which means that a write
is in progress).
If the reader is then fast enough that that odd value is still the
current one when the read_seqcount_retry() is done, we might end up with
a "successful" read sequence, even despite the concurrent write being
active.
In practice this probably never really happens - there just isn't
anything else going on around the read of the sequence count, and the
common case is that we end up having a read barrier immediately
afterwards.
So the code sequence in which gcc might decide to reaload from memory is
small, and there's no reason to believe it would ever actually do the
reload. But if the compiler ever were to decide to do so, it would be
incredibly annoying to debug. Let's just make sure.
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
So that the power button still wakes up the platform.
Signed-off-by: Pierre Tardy <pierre.tardy@intel.com>
Link: http://lkml.kernel.org/r/20120504210244.F2EA5A018B@akpm.mtv.corp.google.com
Tested-by: Kangkai Yin <kangkai.yin@intel.com>
Tested-by: Yong Wang <yong.y.wang@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Cc: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
It seems that there was an error with the active_low = 1 for the
LED, since it should be set to 0 (meaning that active is high,
since 0 is false, hence the confusion.
The wiki article about it confuses it, since it contradicts itself,
regarding what turns on the LED.
I have tested 3.4-rc2 on my net5501 with this patch, and it makes the LED
behave correctly, where "none" turns it off, and "default-on" turns it on,
when echoed onto the trigger "file" in /sys/class/leds.
Signed-off-by: Bjarke Istrup Pedersen <gurligebis@gentoo.org>
Link: http://lkml.kernel.org/r/20120504210146.62186A018B@akpm.mtv.corp.google.com
Cc: Philip Prindeville <philipp@redfish-solutions.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Fix that when scrub tries to repair an I/O or checksum error and one of
the devices containing the mirror is missing, it crashes in bio_add_page
because the bdev is a NULL pointer for missing devices.
Reported-by: Marco L. Crociani <marco.crociani@gmail.com>
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Fix the size members of btrfs_ioctl_ino_path_args and
btrfs_ioctl_logical_ino_args. The user space btrfs-progs utilities used
__u64 and the kernel headers used __u32 before.
Signed-off-by: Alexander Block <ablock84@googlemail.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
If we happen to alloc a extent buffer and then alloc a page and notice that
page is already attached to an extent buffer, we will only unlock it and
free our existing eb. Any pages currently attached to that eb will be
properly freed, but we don't do the page_cache_release() on the page where
we noticed the other extent buffer which can cause us to leak pages and I
hope cause the weird issues we've been seeing in this area. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
add_root_to_dirty_list happens once at the very beginning of the
transaction, but it is still racey.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Pull drm fixes from Dave Airlie:
"Some minor fixes from Intel and a radeon fix.
I have the nouveau fix for the i2c regression queued for next week,
its mostly a revert and seems to work on the system it was originally
introduced for thanks to some i2c core changes."
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
drm/radeon: clarify and extend wb setup on APUs and NI+ asics
drm/i915: enable dip before writing data on gen4
fixing dmi match for hp t5745 and hp st5747 thin client
drm/i915: Only enable IPS polling for gen5
drm/i915: Do not read non-existent DPLL registers on PCH hardware
This fixes a regression that was introduced in the merge window.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.18 (GNU/Linux)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=8qhg
-----END PGP SIGNATURE-----
Merge tag 'md-3.4-fixes' of git://neil.brown.name/md
Pull one small fix for md/bitmaps from NeilBrown:
"This fixes a regression that was introduced in the merge window."
* tag 'md-3.4-fixes' of git://neil.brown.name/md:
md/bitmap: fix calculation of 'chunks' - missing shift.
Jana Saout confirmed that this fixes the page faults he saw.
His problem was triggered by ocfs2 and autofs symlink lookups, where the
symlink allocation was at the end of a page. But the deeper reason
seems to be the use of Xen-PV, which is what then causes him to have all
these unmapped pages, which is what then makes it a problem when the
unaligned word-at-a-time code fetches data past the end of a page.
* fix-unmapped-word-at-a-time:
vfs: make word-at-a-time accesses handle a non-existing page
Missing handler for freeing requested IRQ added.
Moreover clk_ calls has been reorganized.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
This commit adjust the s3c-hsotg to new clock API.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
This code removes the S3C_ prefix from s3c-hsotg driver. This change
provides more architecture independent code.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Add release function to prevent kernel warning.
Kfree is performed when all references are gone.
Signed-off-by: Sangwook Lee <sangwook.lee@linaro.org>
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
The our_hsotg global pointer to hsotg USB device state is removed.
It has been replaced with to_hsotg(gadget) function.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Refactor all comments to comply with kernel codding style.
Moreover doxygen descriptions have been added for selected functions.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Replace of deprecated start and stop callbacks with a udc_start and
udc_stop ones.
Now the bind from composite driver is NOT called explicitly, so more
work needs to be done at s3c_udc_probe. Especially enabling SoC clocks
and power for runtime determination of EP number.
After probing, those sources are disabled and enabled again at udc_start
and pullup afterwards.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Sangwook Lee <sangwook.lee@linaro.org>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
This commit adds support for determining of EPs number during run time.
Configuration is read from a HW configuration register in a specially
created s3c_hsotg_hw_cfg function.
Moreover it was necessary to defer at probe allocation of the
struct s3c_hsotg_ep instances until number of endpoints is known.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
This code allows Samsung SoC's to recover its state when
device is disconnected and connected during transfer.
It is necessary, in such a scenario, to reinitialize the USB core
to assure correct initial state of the driver.
This operation is needed since the disconnect interrupt is only
available at HOST mode, which is not supported by this driver.
A simple mechanism with jiffies has been used to perform core reset
only once.
Tested with:
- DFU gadget (various size of the sent data - also packet = MPS)
- Ethernet gadget (CDC and RNDIS)
- Multi Function Gadget (g_multi)
HW:
- Samsung's C210 Universal rev.0
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
The s3c_hsotg_disconnect_irq function has been renamed to
reflect, that it can be used not only during the host
disconnect irq.
The s3c_hsotg_disconnect shall be used as a fall back for
scenario when USB cable is unplugged and plugged to the
device.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
The USB Disconnect Interrupt handler, according to specification,
is only working at HOST mode.
Samsung SoCs (e.g. Exynos4) are working at device mode, so
this interrupt is never caught.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
The s3c_hsotg_core_init function has been added to exclude
code responsible for Samsung's SoCs USB core initialization.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
This commit targets following scenarios for IN requests:
1. HOST requests e.g. 256B (which is a multiple of MPS = 64B).
Then NO ZLP shall be sent, since host expects exact number of bytes.
2. HOST requested 4096B, but our data for sending is 256B. In this
situation ZLP shall be send to tell HOST that no more data is available
and it shall not wait for more data. This prevents HOST from hanging.
Tested with:
- DFU gadget (various size of the sent data - also packet = MPS)
- Ethernet gadget (CDC and RNDIS)
- Multi Function Gadget (g_multi)
HW:
- Samsung's C210 Universal rev.0
- Samsung's C110 GONI
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
For SETUP stage of USB control transmission, the NAK shall NOT be
CLEAR.
The SNAK/CNAK control is crucial for this type of driver,
since data arrives to earlier defined requests.
Tested with:
- DFU gadget (various size of the sent data - also packet = MPS)
- Ethernet gadget (CDC and RNDIS)
- Multi Function Gadget (g_multi)
HW:
- Samsung's C210 Universal rev.0
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
This commit adds support for supply voltage management for s3c-hsotg IP
block. For that purpose a convenient regulator_bulk_* functions have been
used.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: Liam Girdwood <lrg@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Wrappers for PHY methods have been added for readability and reduction
of code repetition.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Two functions - namely: s3c_hsotg_gate and s3c_hsotg_otgreset are platform
dependent and therefore removed from Samsung's generic s3c-hsotg code.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
This code removes platform dependency from s3c-hsotg driver.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
functionfs was leaking request objects created by autoconfig.
Signed-off-by: Peter Korsgaard <peter.korsgaard@barco.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Add two isochronous endpoints to the gadget zero source/sink
function. They are enabled by selecting alternate interface 1, so
by default they are not enabled. Module parameters for setting all
the isoc endpoint characteristics are also provided.
Signed-off-by: Pratyush Anand <pratyush.anand@st.com>
Signed-off-by: Paul Zimmerman <paulz@synopsys.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
The original code works fine, but Sparse complains because it isn't
annotated properly.
drivers/usb/gadget/composite.c:793:26: warning: incorrect type in assignment (different base types)
drivers/usb/gadget/composite.c:793:26: expected unsigned short [unsigned] [usertype] language
devel/drivers/usb/gadget/composite.c:793:26: got restricted __le16 [usertype] <noident>
devel/drivers/usb/gadget/composite.c:795:29: warning: restricted __le16 degrades to integer
drivers/usb/gadget/composite.c:798:24: warning: incorrect type in assignment (different base types)
drivers/usb/gadget/composite.c:798:24: expected restricted __le16 [usertype] <noident>
devel/drivers/usb/gadget/composite.c:798:24: got unsigned short [unsigned] [usertype] language
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
The previous code always set to USB_DEVICE_SELF_POWERED in GET_STATUS.
So, this patch adds set_selfpowered().
Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
USB_FSL_USB2 driver can be used on PowerPC and i.MX processors.
Include i.MX processors in the USB_FSL_USB2 help text.
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
dTD's next dtd pointer need to be updated once CPU writes it, or this
request may not be handled by controller, then host will get NAK from
device forever.
This problem occurs when there is a request is handling, we need to add
a new request to dTD list, if this new request is added before the current
one is finished, the new request is intended to added as next dtd pointer
at current dTD, but without wmb(), the dTD's next dtd pointer may not be
updated when the controller reads it. In that case, the controller will
still get Terminate Bit is 1 at dTD's next dtd pointer, that means there is
no next request, then this new request is missed by controller.
CC: <stable@vger.kernel.org>
Signed-off-by: Peter Chen <peter.chen@freescale.com>
Acked-by: Li Yang <leoli@freescale.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
In some USB composite gadget drivers, the configuration's bind function called
by the usb_add_config() calls multiple bind config functions. (for example cdc2
configuration bind function in the cdc_do_config() of the cdc2.c has two
functionality bind config functions.
- the ecm_bind_config() & the acm_bind_config())
In each functionality bind config function, new instance is allocated and
finally added by the usb_add_function().
So if an error occurred during the second functionality bind config (for
example an error occurred at the acm_bind_config() after succeeding of the
ecm_bind_function()), the instance created by the acm_bind_config() cannot be
freed creating a memory leak.
This patch fixes this issue.
Signed-off-by: Yongsul Oh <yongsul96.oh@samsung.com>
Acked-by: Michal Nazarewicz <mina86@mina86.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
There are no in-tree fsg_add() users and it has been deprecated
since 2.6.35 [1dc90985d1: fsg_add() renamed to fsg_bind_config()] so
out-of-tree users had more then enough time to convert. Removing.
Signed-off-by: Michal Nazarewicz <mina86@mina86.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
The previous code always set to USB_DEVICE_SELF_POWERED in GET_STATUS.
So, this patch adds set_selfpowered().
Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
ep->desc is set to NULL on endpoint disable. That means once an endpoint
is disabled it is not possible to free requests. In my target gadget I
first disable endpoints to make sure I have no requests on the fly and
then free frequests. On dummy I am leaking memory here.
Since I can't imagine a reason why it should be a bad thing, lets allow
to free requests on disabled endpoints. On removal of composite the ep0
request is removed so lets allow that here as well.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Felipe Balbi <balbi@ti.com>
This patch converts the g_printer to make use of the compoiste framework
for descriptor parsing instead of its own implementation of it.
This gadget contains now one function which is the printer gadget.
Compile tested only.
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Signed-off-by: Felipe Balbi <balbi@ti.com>
This patch removes the DUALSPEED macro and makes the HS (and FS) case
the default. This is one little step before composite can be used for
descriptor management.
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Signed-off-by: Felipe Balbi <balbi@ti.com>
This patches converts the driver into the new style start/stop interface.
As a result the driver no longer uses the static global controller
variable in start/stop code. I kept the gloval controller variable because
it keeps init simple.
Compile tested only.
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Signed-off-by: Felipe Balbi <balbi@ti.com>
All of this callbacks which I remove here are not implemented and return
an error code. The gadget code returns an error code if a callback is
missing so there is no need to implement this twice.
Cc: Li Yang <leoli@freescale.com>
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Signed-off-by: Felipe Balbi <balbi@ti.com>
