Some people think that this line is not compatible with the GPL. The
statement was required due to the Buenos Aires Convention and is now
deprecated. I remove it because it is said that it is pointless nowdays.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Use "ep0in" and "ep0out" instead "ep1in" and "ep0out" which is confusing
and not consistent with the remaining output.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
We already give requests back in dwc3_ep0_stall_and_restart() so
doing it again here will most likely corrupt the list.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
This patch adds the TSC-40 serial touchscreen driver and should be
compatible with TSC-10 and TSC-25.
The driver was written by Linutronix on behalf of Bachmann electronic GmbH.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Christian Gmeiner <[email protected]>
Signed-off-by: Dmitry Torokhov <[email protected]>
This removes the need of ifdefs within the init function and with it the
headache about the correct clean without bus X but with bus/platform Y &
Z.
xhci-pci is only compiled if CONFIG_PCI is selected which can be
de-selected now without trouble. For now the result is kinda useless
because we have no other glue code. However, since nobody is using
USB_ARCH_HAS_XHCI then it should not be an issue :)
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Sarah Sharp <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
xhci_gen_setup() is generic so it can be used to perform the bare xhci
setup even on non-pci based platform. The typedef for the function
pointer is moved into the headerfile
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Sarah Sharp <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
xhci_pci_setup() is split into three pieces:
- xhci_gen_setup()
The major remaining of xhci_pci_setup() is now containing the generic
part of the xhci setup. It allocates the xhci struct, setup
hcs_params? and friends, performs xhci_halt(), xhci_init and so one.
It also obtains the quirks via a callback
- xhci_pci_quirks()
It checks the origin of the xhci core and sets core specific quirks.
- xhci_pci_setup()
PCI specific setup functions. Besides calling xhci_gen_setup() with
xhci_pci_quirks() as an argument it performs PCI specific setup like
obtaining the address of sbrn via a PCI config space.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Sarah Sharp <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
pci_*_consistent() calls dma_*_coherent() with GFP_ATOMIC and requires
pci_dev struct. This is a preparion for later where we no longer have
the pci struct around.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Sarah Sharp <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
The MSI related fuctionality requires a few structs which are not
available if CONFIG_PCI is not enabled. This is a prepartion to allow
xhci be built without CONFIG_PCI set.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Sarah Sharp <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
This patch moves the complete MSI/MSI-X/Legacy dance into its own
function. There is however one difference: If the XHCI_BROKEN_MSI flag
is set then we don't free and register the irq, we simply return.
This is preparation for later PCI decouple.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Sarah Sharp <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
instead of reading the xhci interface version each time _even_ if the
quirk is not required, simply check if the quirk flag is set. This flag
is only set of the module parameter is set and here is where I moved the
version check to.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Sarah Sharp <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
If the gadget drivers sends a ZLP we are trying to map this this request
which does not work on all implementations. So we simply skip mapping
it.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
There are two spots where we wait until the HW finishes processing a
certain command. Initially we had a few problems and we used 500ms as a
limit to be on a the safe side. Paul Zimmerman mentioned this is little too
much. After a debugging session, we noticed that we hardly ever go over 20us
and didn't pass 30usec so far. Using mdelay() seems way overloaded.
Giving the current numbers 500usec as the upper limit is more than enough.
Should it ever timeout then something is definitely wrong.
While here, also replace the type with u32 since long does not really
fit here.
Cc: Paul Zimmerman <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
- since a while we are disabling an endpoint and purging every requests on
RESET and DISCONNECT which leads to a warning since the endpoint was
disabled twice (once by the UDC, and second time by the gadget). I
think UDC should nuke all requests because all those requests
become invalid. It's gadget driver's responsability, though, to disable
its used endpoints. This is done by merging dwc3_stop_active_transfer()
and dwc3_gadget_nuke_reqs() into dwc3_remove_requests().
- dwc3_stop_active_transfer() is now no longer called unconditionaly.
This has the advantage that it is always called to disable an active
transfer which means if res_trans_idx 0 than something went wrong and
it is an error condition because we can't clean up the requests.
- Remove the DWC3_EP_WILL_SHUTDOWN which was introduced while
introducing the command complete part for dequeue. All requests on
req_queued list should be removed during the dwc3_cleanup_done_reqs()
callback so there is no reason to go through the list again.
We consider it an error condition if requests are still on this
list since we never queue TRB without LST=1 (the last requests has
always LST=1, there are no requests with LST=0 behind it).
[ [email protected] : reworked commit log a bit, made patch apply ]
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
We read the DWC3_GSNPSID register to make sure we got the correct
register offset passed. One of the recent commits moved the soft reset
before this so in case of the wrong offset we end up with "reset timed
out". This patch moves the "id" check before the reset again.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
There are some issues around for enabling/disabling this mode and
handling it. It does not work perfectly (yet). However we have a few
gadgets tested successfuly so far. That means we are quite confident
that we won't need this in near future.
So I'm for removing it and bringing a working version back once there is
a need for it.
Thanks to Dan Carpenter who spotted the wrong memory handling here.
[ [email protected] : made it actually apply ]
Cc: Dan Carpenter <[email protected]>
Cc: [email protected]
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Without this patch we won't clear that bit and instead will
clear all other bits on our endpoint flag.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Type 6 should be used for the first transfer during an interval. This is
also what the reference driver is using. Type 7 seems to be for following
or additional transfers within the same interval.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
If we collected two requests together (i.e. only the last of them has
LST=1) then we only have to stop transfer once: The clean-up code will
cleanup everything until first TRB with the LST bit set.
After XferComplete this index should be no longer valid since there is
no transfer pending.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
A TRB which is dequeued seems to have its HWO bits set to 1. Therefore
we ignore it if we dequeue it after the command is completed.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Executing
| testusb -a -c 1 -t 3 -v 421 -s 2048
does not complete on the gadget side.
g_zero enqueues a 4096 bytes long buffer. The host sends 2048bytes which
is a multiple of wMaxPacketSize (either 64 or 512 bytes). The host is
done with sending data but the gadget waits for more.
Since the protocol does not include transfer-length-field sending a
terminating zero packet seems the only way out.
Reviewed-by: Felipe Balbi <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
If id_entry is available then it is used. However if we remove first the
driver followed by the device, then the id_entry is pointing to driver's
memory which is long gone.
Since id->name and plat->name are equal there is no point in
distinguishing them.
Cc: Grant Likely <[email protected]>
Cc: [email protected]
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
not to confuse with Table 9-7 in USB 2.0 spec
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
In case one "forgot" to load the receiver i.e. doing
|modprobe omap2430
|modprobe musb_hdrc
he ends up with:
|musb-hdrc: version 6.0, ?dma?, otg (peripheral+host)
|HS USB OTG: no transceiver configured
|musb-hdrc musb-hdrc: musb_init_controller failed with status -19
|(NULL device *): gadget not registered.
|Unable to handle kernel NULL pointer dereference at virtual address 0000001c
|Internal error: Oops: 17 [#1] SMP
|[<c011383c>] (sysfs_find_dirent+0x4/0x60) from [<c01138c0>] (sysfs_get_dirent+0x28/0x78)
|[<c01138c0>] (sysfs_get_dirent+0x28/0x78) from [<c0115b78>] (sysfs_unmerge_group+0x1c/0x90)
|[<c0115b78>] (sysfs_unmerge_group+0x1c/0x90) from [<c0179ba4>] (dpm_sysfs_remove+0x14/0x3c)
|[<c0179ba4>] (dpm_sysfs_remove+0x14/0x3c) from [<c01742f8>] (device_del+0x40/0x1b4)
|[<c01742f8>] (device_del+0x40/0x1b4) from [<c0174478>] (device_unregister+0xc/0x18)
|[<c0174478>] (device_unregister+0xc/0x18) from [<bf0489b4>] (musb_free+0x24/0x88 [musb_hdrc])
|[<bf0489b4>] (musb_free+0x24/0x88 [musb_hdrc]) from [<bf057d18>] (musb_probe+0xb50/0xe3c [musb_hdrc])
|[<bf057d18>] (musb_probe+0xb50/0xe3c [musb_hdrc]) from [<c01779c4>] (platform_drv_probe+0x1c/0x24)
The problem is that musb_free() tries to figure out what was
initializued and what wasn't and clean up only the initialized part.
This works well for usb_del_gadget_udc() but device_unregister() can't
deal with it. Therefore we rely on the fact the we always have a parent
device and only then remove the device.
I broke this in 0f91349 ("usb: gadget: convert all users to the new udc
infrastructure")
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
For bMaxPacketSize0 we usually take what is specified in ep0->maxpacket.
This is fine in most cases, however on SuperSpeed bMaxPacketSize0
specifies the exponent instead of the actual size in bytes. The only
valid value on SS is 9 which denotes 512 bytes.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
The code in this block is unused and the Author is fine with removing:
| These functions were used to debug unstable hw fifo while developing
| fusb300. It's much more stable now.
| So these functions can be removed.
Cc: "Wendy Yuan-Hsin Chen" <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
For bMaxPacketSize0 we usually take what is specified in ep0->maxpacket.
This is fine in most cases, however on SuperSpeed bMaxPacketSize0
specifies the exponent instead of the actual size in bytes. The only
valid value on SS is 9 which denotes 512 bytes.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
The code in this block is unused and the Author is fine with removing:
| These functions were used to debug unstable hw fifo while developing
| fusb300. It's much more stable now.
| So these functions can be removed.
Cc: "Wendy Yuan-Hsin Chen" <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
wMaxPacketSize is __le16 and should be accessed as such. Also fix the
wBytesPerInterval assignment while here.
v2: also fix the wBytesPerInterval assigment, noticed by Matt Evans
This patch should be backported to the 3.0 kernel.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Sarah Sharp <[email protected]>
Acked-by: Matt Evans <[email protected]>
Cc: [email protected]
If no primary handler is specified then a default one is assigned
which always returns IRQ_WAKE_THREAD. This handler requires the
IRQF_ONESHOT flag on LEVEL / EIO typed irqs because the source of
interrupt is not disabled. Since it is required for those users and
there is no difference for others it makes sense to add this flag
unconditionally.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Thomas Gleixner <[email protected]>
This is not compile tested as I failed at it. I added an #if 0 block
because I did not find the struct device of the device.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Both fusb300 and langwell udcs seem to only
work with 32-bit address space.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
- remove pointer u32 abuse in fusb300_fill_idma_prdtbl().
It is assigned the dma_addr to a pointer and then back.
Poor families may have to recycle variables but we don't
- don't free req.buf in error case. We don't do it in the
ok case so it is probably wrong to do it in error case.
- return in error case. There is no reason to continue
without data and performing ops on an invalid pointer.
- The if (d) statement is bogus since an invalid DMA pointer
is ~0 on some architecutres. And since we return for the
invalid case we don't need it.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Fix the following compile warning:
| usb/gadget/ci13xxx_udc.c: In function 'show_registers':
| usb/gadget/ci13xxx_udc.c:1242:1: warning: the frame size of 2064 bytes is larger than 1024 bytes
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
bind() and pull is moved to udc core, call callbacks are verified by the
upper layer.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
udc-core checks for valid callbacks so there is no need for the driver
to do so. Also "can-be-bound-once" is verified by udc-core. The pull-up
callback is called by udc-core afterwords.
[ [email protected] : keep holding gadget_driver pointer for now
remove the stupid check for gadget_driver otherwise
we don't handle IRQs ]
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
udc_start() should only trigger the internal state machine and make
minimal house keeping. Before that call udc-core calls the bind()
callback and after the callback the pullup().
udc_stop() is simillar, udc-core calls pullup(), unbind() and finally
udc_stop().
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
The configuration is static however we only know the speed after we have
connected with the other side.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
This value is now assigned during bind(). The configuration depends on
static values assigned by dummy driver itself. So there is no need to
defer this assigment until one know the actuall speed since the
configuration is static and known early.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
This is only required to be done once. There is no counter part to this
in ->stop() so there is no need to re-do it next time. While here also
init the max_stream size to 0 on SS speed.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
maxpacket is set by the udc driver for ep0 very early. This value is
copied by the function gadget used later for the USB_DT_DEVICE and
USB_DT_DEVICE_QUALIFIER query. This seems to work fine so far. For USB3
we need set a different value here. In SS speed it is 2^x with x=9 and
in HS we set something <= 64. If the UDC starts in SS and continues in
HS after the cable has been plugged it will report a too small value.
There setting of this value is defered and taken automaticly from the
ep0 pointer where the UDC driver can update it according to the speed it
detected _after_ a cable has been plugged.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
This function has no user in my tree. It looks like it belongs to
net2280 but it somehow morphed into the dummy_hcd. So I remove it
before it spreads into more drivers.
After some digging I figured out that the only user was removed in
|commit 9079e91b5b
|Author: David Brownell <[email protected]>
|Date: Wed May 7 16:00:36 2008 -0700
|
| USB: serial gadget: cleanup/reorg
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Taking the correct struct once avoids doing the speed dance.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
gadget_to_dummy_hcd() already does the speed check, so
it's unnecessary to unroll that all the time.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Before commit 53832daea ("usb: gadget: dummy_hcd: use the shared_hcd
infrastructure") the is_otg field was set in dummy_udc_probe(). It seems
to me that this field is used in gadget's bind function. Therefore I'm
moving it before the bind() callback is called.
Cc: Tatyana Brokhman <[email protected]
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
The removed code does the same thing as dummy_pullup(). The only
difference is that in dummy_udc_stop() the first dummy_pullup()
did not call usb_hcd_poll_rh_status().
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
This also fixes the error path: If the second device fails to register
we never remove the first one.
This is compile-tested only. I don't see any difference between those
two. Maybe we should just use one name instead?
Cc: Ben Dooks <[email protected]>
Cc: Kukjin Kim <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Without it udev won't be able to load the driver once it notices the
device unbound.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
In case pci_register_driver() fails it error will be 0 in case
platform_driver_register() was fine. Also without PCI
pci_register_driver() evaluates to 1 which is well, special. If
platform_driver_register() returns EINVAL or EBUSY we end up with 0.
Acked-by: Mike Frysinger <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Can't find evidence that this is actually done.
Cc: Kuninori Morimoto <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
I doubt the clock is optional. In case it is it should not return with
an error code because we leak everything.
Cc: Jingoo Han <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
this is more backwords than it has to be.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|drivers/usb/gadget/mv_udc_core.c:2108: error: label `error' used but not defined
This seems to be broken since the initial commit. I changed this to a
simple return. The other user is the probe code which lets ->probe()
fail on error here.
|drivers/usb/gadget/mv_udc_core.c:2107: warning: passing argument 1 of `dev_err' from incompatible pointer type
|drivers/usb/gadget/mv_udc_core.c:2118: warning: initialization from incompatible pointer type
|drivers/usb/gadget/mv_udc_core.c:2119: warning: initialization from incompatible pointer type
|drivers/usb/gadget/mv_udc_core.c:2130: error: initializer element is not constant
|drivers/usb/gadget/mv_udc_core.c:2130: error: (near initialization for `udc_driver.driver.pm')
Cc: Chao Xie <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
while going through Tatyana's changes for the gadget framework I noticed
that this type is not defined as __le16.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Sarah Sharp <[email protected]>
the_controller is allocated in dummy_hcd_probe() and is NULL if the
allocation failed. The probe function of the udc driver is dereferencing
this pointer and fault.
Alan Stern suggested to abort the dummy_hcd driver probing so the module
is not loaded. The is abort-on-error has been also added to the udc
driver.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Acked-by: Alan Stern <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
If USB type detections fails, we run into default and return 0.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Acked-by: Alan Stern <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
We use io_apic_setup_irq_pin() in order to configure pin's interrupt
number polarity and type. This is done on every irq_create_of_mapping()
which happens for instance during pci enable calls. Level typed
interrupts are masked by default, edge are unmasked.
On the first ->xlate() call the level interrupt is configured and
masked. The driver calls request_irq() and the line is unmasked. Lets
assume the interrupt line is shared with another device and we call
pci_enable_device() for this device. The ->xlate() configures the pin
again and it is masked. request_irq() does not unmask the line because
it _is_ already unmasked according to its internal state. So the
interrupt will never be unmasked again.
This patch is based on an earlier work by Torben Hohn and solves the
problem by configuring the pin only once. Since all devices must agree
on the same type and polarity there is no point in configuring the pin
more than once.
[ tglx: Split out the ce4100 part into a separate patch ]
Cc: Torben Hohn <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Link: http://lkml.kernel.org/r/%3C20110427143052.GA15211%40linutronix.de%3E
Signed-off-by: Thomas Gleixner <[email protected]>
the of_node will auto-publish devices which are added to the device
tree.
Commit 925bb9c6 aka ("of/i2c: Fix module load order issue caused by
of_i2c.c) moved the of_i2c_register_devices() function from the i2c core
back to the drivers. This patch does the same thing for the pxa driver.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
Signed-off-by: Ben Dooks <[email protected]>
timeout here maybe 0 if the event occured and a task with a higher
priority stole the cpu and we were sleeping longer than the timeout
value we specified.
In case of a real timeout I changed the error code to I2C_RETRY so we
retry the transfer.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
Signed-off-by: Ben Dooks <[email protected]>
Sodaville has three of them on a single IRQ. IRQF_DISABLED is removed
because it is a NOP allready and scheduled for removal.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
Signed-off-by: Ben Dooks <[email protected]>
The Sodaville I2C controller is almost the same as found on PXA2xx. The
difference:
- the register are at a different offset
- no slave support
The PCI probe code adds three platform devices which are probed then by
the platform code.
The X86 part also adds dummy clock defines because we don't have HW
clock support.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
Signed-off-by: Ben Dooks <[email protected]>
This patch moves the platform data definition from
arch/arm/plat-pxa/include/plat/i2c.h to include/linux/i2c/pxa-i2c.h so
it can be accessed from x86 the same way as on ARM.
This change should make no functional change to the PXA code. The move
is verified by building the following defconfigs:
cm_x2xx_defconfig corgi_defconfig em_x270_defconfig ezx_defconfig
imote2_defconfig pxa3xx_defconfig spitz_defconfig zeus_defconfig
raumfeld_defconfig magician_defconfig mmp2_defconfig pxa168_defconfig
pxa910_defconfig
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
Signed-off-by: Ben Dooks <[email protected]>
This will prepare the driver to handle register layouts where certain
registers are not available at all.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Ben Dooks <[email protected]>
Setting the pci ops on subsys initcall unconditionally will break
multi platform kernels on anything except ce4100.
Use x86_init.pci.init ops to call this only on real ce4100 platforms.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Cc: [email protected]
LKML-Reference: <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Until "x86: dt: Cleanup local apic setup" we read the local apic
address from the MSR and ignored the entry in DT. Reflect this change
in the documentation.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Both OLPC and CE4100 activate CONFIG_OF. OLPC uses PROMTREE while CE
uses FLATTREE. Compiling for OLPC only breaks due to missing flat tree
functions and variables.
Use proper wrappers and provide an empty x86_flattree_get_config()
inline so OF=y FLATTREE=n builds and works.
[ tglx: Make it work with HPET_TIMER=n and make a function static ]
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
This allows to load the OF driver based informations from the device
tree. Systems without BIOS may need to perform some initialization.
PowerPC creates a PNP device from the OF information and performs this
kind of initialization in their private PCI quirk. This looks more
generic.
This patch also avoids registering the platform RTC driver on X86 if
we have a device tree blob. Otherwise we would setup the device based
on the hardcoded information in arch/x86 rather than the device tree
based one.
[ tglx: Changed "int of_have_populated_dt()" to bool as recommended by
Grant ]
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
Acked-by: Grant Likely <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: Alessandro Zummo <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
ioapic_xlate provides a translation from the information in device tree
to ioapic related informations. This includes
- obtaining hw irq which is the vector number "=> pin number + gsi"
- obtaining type (level/edge/..)
- programming this information into ioapic
ioapic_add_ofnode adds an irq_domain based on informations from the device
tree. This information (irq_domain) is required in order to map a device to
its proper interrupt controller.
[ tglx: Adapted to the io_apic changes, which let us move that whole code
to devicetree.c ]
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
Acked-by: Grant Likely <[email protected]>
Cc: [email protected]
Cc: [email protected]
LKML-Reference: <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
x86_of_pci_init() does two things:
- it provides a generic irq enable and disable function. enable queries
the device tree for the interrupt information, calls ->xlate on the
irq host and updates the pci->irq information for the device.
- it walks through PCI bus(es) in the device tree and adds its children
(device) nodes to appropriate pci_dev nodes in kernel. So the dtb
node information is available at probe time of the PCI device.
Adding a PCI bus based on the information in the device tree is
currently not supported. Right now direct access via ioports is used.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Tested-by: Dirk Brandewie <[email protected]>
Acked-by: Grant Likely <[email protected]>
Cc: [email protected]
Cc: [email protected]
LKML-Reference: <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
APIC and IO_APIC have to be added to the system early because
native_init_IRQ() requires it.
In order to obtain the address of the ioapic the device tree has to be
unflattened so of_address_to_resource() works.
The device tree is relocated to ensure it is always covered by the
kernel mapping. That way the boot loader does not have to make
any assumptions about kernel's memory layout.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Acked-by: Grant Likely <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: Dirk Brandewie <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
The here introduced irq_domain abstraction represents a generic irq
controller. It is a subset of powerpc's irq_host which is going to be
renamed to irq_domain and then become generic. This implementation will
be removed once it is generic.
The xlate callback is resposible to parse irq informations like irq type
and number and returns the hardware irq number which is reported by the
hardware as active.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Tested-by: Dirk Brandewie <[email protected]>
Acked-by: Grant Likely <[email protected]>
Cc: [email protected]
Cc: [email protected]
LKML-Reference: <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
History:
v1..v2:
- dropped device_type except for cpu & pci. I have the compatible string
for pci so I can drop the device_type once it is possible
- I lowercased all compatible types. I will need to resend some patches
which have upper case intel
- The cpu had the same compatible string as the soc node. So I added to
the soc node -immr for internel memory mapped registers.
- I added generic names for all parts.
- I reworked the i2c bars matching the way you suggested. I added a
compatible node for the PCI device which only the PCI ids in its
compatible string. The bars (each represents a complete i2c
controller) have a "intel,ce4100-i2c-controller" compatible node. It
is not used by the driver.
The driver is probed via PCI ids (by the pci subsystem not OF) and
matches the bar address against the ressource in the child node. Once
there is a hit the node is attached.
- The SPI driver is also probed via pci. However I also attached a
compatible property based on PCI ids
v2..v3:
- intel,ce4100-immr become intel,ce4100-cp. cp stands for core
peripherals. The Atom data sheet talks here about ACPI devices. Since
we don't have ACPI this does not apply here.
- The interrupt map is gone. There are now plenty of device nodes.
- The "unit address string" got fixed, it uses not DD,V format.
v3..v4:
- added descriptions for compatible nodes introduced here:
- intel,ce4100-ioapic
- intel,ce4100-lapic
- intel,ce4100-hpet
- intel,ce4100
- intel,ce4100-cp
- intel,ce4100-pci
- added a description about I2C controller magic.
- Added gpio-controller and gpio-cells property to gpio devices. Those
properties are not (yet) used.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
Acked-by: Grant Likely <[email protected]>
Cc: [email protected]
Cc: [email protected]
LKML-Reference: <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
This patch adds minimal support for device tree on x86. The device
tree blob is passed to the kernel via setup_data which requires at
least boot protocol 2.09.
Memory size, restricted memory regions, boot arguments are gathered
the traditional way so things like cmd_line are just here to let the
code compile.
The current plan is use the device tree as an extension and to gather
information which can not be enumerated and would have to be hardcoded
otherwise. This includes things like
- which devices are on this I2C/SPI bus?
- how are the interrupts wired to IO APIC?
- where could my hpet be?
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
Acked-by: Grant Likely <[email protected]>
Cc: [email protected]
Cc: [email protected]
LKML-Reference: <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
This patch ensures that the memory passed from parse_setup_data() is
large enough to cover the complete data structure. That means that the
conditional mapping in parse_e820_ext() can go.
While here, I also attempt not to map two pages if the address is not
aligned to a page boundary.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
Cc: [email protected]
Cc: [email protected]
LKML-Reference: <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
The document says:
|2.1 Problem description
| When at least two USB devices are simultaneously running, it is observed that
| sometimes the INT corresponding to one of the USB devices stops occurring. This may
| be observed sometimes with USB-to-serial or USB-to-network devices.
| The problem is not noticed when only USB mass storage devices are running.
|2.2 Implication
| This issue is because of the clearing of the respective Done Map bit on reading the ATL
| PTD Done Map register when an INT is generated by another PTD completion, but is not
| found set on that read access. In this situation, the respective Done Map bit will remain
| reset and no further INT will be asserted so the data transfer corresponding to that USB
| device will stop.
|2.3 Workaround
| An SOF INT can be used instead of an ATL INT with polling on Done bits. A time-out can
| be implemented and if a certain Done bit is never set, verification of the PTD completion
| can be done by reading PTD contents (valid bit).
| This is a proven workaround implemented in software.
Russell King run into this with an USB-to-serial converter. This patch
implements his suggestion to enable the high frequent SOF interrupt only
at the time we have ATL packages queued. It goes even one step further
and enables the SOF interrupt only if we have more than one ATL packet
queued at the same time.
Cc: <[email protected]> # [2.6.35.x, 2.6.36.x, 2.6.37.x]
Tested-by: Russell King <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Right now the platform device and its platform data is included in one big
struct which requires its custom ->release function. The problem with the
release function within the driver is that it might be called after the
driver was removed because someone was holding a reference to it and it
was not called right after platform_device_unregister(). So we also free
the platform device memory to which one might hold a reference.
This patch uses the normal pdev functions so this kind of race does not
occur.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Grant Likely <[email protected]>
This patch changes u32 to __be32 for all "ranges", "prop" and "addr" and
such. Those variables are pointing to the device tree which contains
integers in big endian format.
Most functions are doing it right because of_read_number() is doing the
right thing for them. of_bus_isa_get_flags(), of_bus_pci_get_flags() and
of_bus_isa_map() were accessing the data directly and were doing it wrong.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Acked-by: Benjamin Herrenschmidt <[email protected]>
Signed-off-by: Grant Likely <[email protected]>
This was fixed by David Lamparter in v2.6.36-rc5 3486008 ("spi: free
children in spi_unregister_master, not siblings") and broken again in
v2.6.37-rc1~2^2~4 during the merge of 2b9603a0 ("spi: enable
spi_board_info to be registered after spi_master").
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: David Lamparter <[email protected]>
Cc: Grant Likely <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This patch changes u32 to __be32 for all "ranges", "prop" and "addr" and
such. Those variables are pointing to the device tree which containts
intergers in big endian format.
Most functions are doing it right because of_read_number() is doing the
right thing for them. of_bus_isa_get_flags(), of_bus_pci_get_flags() and
of_bus_isa_map() were accessing the data directly and were doing it wrong.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Acked-by: Benjamin Herrenschmidt <[email protected]>
Signed-off-by: Benjamin Herrenschmidt <[email protected]>
Sodaville needs to setup the IO_APIC ids as the boot loader leaves
them uninitialized. Split out the setter function so it can be called
unconditionally from the sodaville board code.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Cc: Yinghai Lu <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
After all TX bytes are sent, the driver spins while the SPI core is busy
and then it spins for a "short" period of time until RX bytes are
available.
On Sodavile the busy flag disappears pretty quick and after that it
takes approx ~130ms (sometimes less but not much) until there are bytes
available in the RX FIFO.
This patch removes the busy loop and modifies the RX threshould so we
get woken up once the remainings bytes arrived.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
The SPI core on Sodaville supports chip selects. Its configuration
moved into the SSSR register at bit 0 and 1. Thus Sodaville can be hooked
up with up to 4 devices.
This patch ensures that the bits which are otherwiese reserved are only
touched on Sodaville and not on any other PXAs. Also it makes sure that
the status register does not lose the CS information while clearing the
ROR bit.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
For PXA the default threshold is FIFO_DEPTH / 2. Adjust this value for
CE4100.
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Signed-off-by: Dirk Brandewie <[email protected]>
