usb: dwc3: gadget: ISOC transfers should be stopped before starting a transfer
For ISOC transfers the requests are not queued until the HOST requests
for data and XferNotReady event is generated .But XferNotReady event
is not getting generated for ISOC transfers for the second time after
Endpoint configuration. Since ISOC packets depend on the XferNotReady
events, they will not be queued to controller. Because of this issue
timeout happens on the application layer.
This patch fixes this issue by issuing END TRANSFER command before
starting any ISOC transfers. Doing so will make the controller clear
the previous allocated endpoint resources and reallocate resources
when the transfer is requested. Because of this change XferNotReady
events will be generated when host requests for the ISOC transfer.
Signed-off-by: Anurag Kumar Vulisha <anurag.kumar.vulisha@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Dhaval Shah [Thu, 26 Apr 2018 07:00:31 +0000 (00:00 -0700)]
soc: xilinx: vcu: Optimize the VCU PLL calculation.
Calculate the maximum and minimum possible FBDIV
values so that number of iteration can be reduce to
calculate FBDIV and divisor for the provided clock
information.
Signed-off-by: Dhaval Shah <dhaval.shah@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
This patch adds interlaced support to the Xilinx V4L dma client. In case
of capture pipeline, the field id is read from the callback.
A check is present to find and correct sequence number in case
a frame is dropped i.e. fid is repeated. For this the prev_fid member is
used to store the previous fid value to be compared with one returned.
In case of output pipeline, the field id is set per dma descriptor.
A dma descriptor pointer is added to xvip_dma_buffer so that it may be
passed as a reference while getting the fid.
The video node gets the field type of the subdev prior and checks if it
is of V4L2_FIELD_ALTERNATE type. If yes then height is halved.
Some other minor fixes for checkpatch are also applied here.
Use the IOCTL Macros provided by Linux Kernel when handling IOCTL requests.
Plus for all parameter passing for IOCTL commands use pointers to be
consistent.
Signed-off-by: Derek Kiernan <derek.kiernan@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
misc: xilinx-sdfec: Use "code_id" to dev_err calls
code_id is a multiplier used when calculating the reg_addr in existing
dev_err calls in the following functions:
- xsdfec_collect_ldpc_reg0
- xsdfec_collect_ldpc_reg1
- xsdfec_collect_ldpc_reg2
- xsdfec_collect_ldpc_reg3
Signed-off-by: Derek Kiernan <derek.kiernan@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Dhaval Shah [Wed, 25 Apr 2018 07:46:07 +0000 (00:46 -0700)]
soc: xilinx: vcu: Divisor calculation for MCU clock is updated.
Divisor of the mcu clock is calculated in such a way that
mcu clock derived from the calculation is greater than
or equal to provided values from the logicoreIP.
Signed-off-by: Dhaval Shah <dhaval.shah@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
dma: xilinx: Get DMA alignment from device tree for Framebuffer
Patch adds xlnx,pixels-per-clock(ppc) and xlnx,dma-align device tree
property decoding to calculate the required DMA alignment.
xlnx,pixels-per-clock is mandatory property now.
Minimum alignment required is 8 * pixels per clock in bytes.
dma-align property is optional.
In case present, dma-align must be a power of 2 and be >= 8 * ppc.
If absent, alignment is set as ppc * 8.
Added a new compatible string and structure which contains dma direction
and flag. The flag would be a bitmask of properties.
When ever a new IP or device tree property is to be supported, a bitmask
for the same is to be created and added to this flag.
For the new v2.1 compatible string, a new XILINX_PPC_PROP bitmask is
set in the flags. If the flag is set then, pixels-per-clock property
would be checked.
Documentation: devicetree: bindings: dma: Add properties for DMA alignment
Adds the pixels per clock and dma-align device tree properties. These
are used to determine the DMA memory alignment.
New compatible strings xlnx,axi-frmbuf-rd-v2.1 and
xlnx,axi-frmbuf-wr-v2.1 have also been added for this.
This patch removes quirk which indicates a broken base clock. This was
making the kernel report wrong base clock of ~187MHz instead of 200MHz
even as the measurement on the hardware was showing 200MHz.
Signed-off-by: Manish Narani <manish.narani@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Rajan Vaja [Thu, 19 Apr 2018 18:13:11 +0000 (11:13 -0700)]
firmware: zynqmp: Add sysfs and IOCTL to set boot health status
Add sysfs interface to set boot health status from userspace.
Add IOCTL ID used by this interface to communicate with firmware.
If PMUFW is compiled with CHECK_HEALTHY_BOOT, it will check the
healthy bit on FPD WDT expiration. If healthy bit is set by a user
application running in Linux, PMUFW will do APU only restart. If
healthy bit is not set during FPD WDT expiration, PMUFW will do
system restart.
Signed-off-by: Rajan Vaja <rajan.vaja@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
By default, the device names derive from the dt description. The DP
pcm devices are represented as a child node of DP node, which results
in both parent and child node names in the device name. The name
created in this way gets too long to fit into the sound component name,
and last characters where IDs are located are not included. This gives
the same component name for two different components, and the debugfs
fails to create entries with same name as warning below:
xilinx-dp-snd-pcm fd4a0000.zynqmp-display:zynqmp_dp_snd_pcm1: ASoC: Failed to create component debugfs directory
This fixes it by setting the child node name as a device name.
drm: xlnx: fb: Prefer the current format depth over depth from fb helper
The drm fb helper has specific preference of bpp and depth. For example,
for 32bit bpp, the depth is hard-coded to be 24. If it's not aligned
with the supported format of a drm device, it fails to initialize fbdev.
So override the depth value from fb helper with the current format
of the drm device. This will allow to initialize the fbdev with
preferred format that matches with actual format.
Currently the driver depends on the bootloader to enable the clocks.
Add support for clocking. The patch enables the clock at probe and
disables them at remove.
Signed-off-by: Shubhrajyoti Datta <shubhrajyoti.datta@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Glenn Langedock [Tue, 27 Mar 2018 14:09:25 +0000 (16:09 +0200)]
gpio: zynq: protect direction in/out with a spinlock
Fix race condition when changing the direction (in/out) of the GPIO pin.
The read-modify-write sequence (as coded in the driver) isn't atomic and
requires synchronization (spinlock).
Signed-off-by: Glenn Langedock <Glenn.Langedock@barco.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Michal Simek [Tue, 27 Mar 2018 12:31:42 +0000 (14:31 +0200)]
arm: zynq: Remove 0x prefixes from cc108
The patch fixing issues reported by DTC:
arch/arm/boot/dts/zynq-cc108.dtb: Warning (unit_address_format): Node
/amba/spi@e000d000/flash@0/partition@0x400000 unit name should not have
leading "0x"
arch/arm/boot/dts/zynq-cc108.dtb: Warning (unit_address_format): Node
/amba/spi@e000d000/flash@0/partition@0x800000 unit name should not have
leading "0x"
arch/arm/boot/dts/zynq-cc108.dtb: Warning (unit_address_format): Node
/amba/spi@e000d000/flash@0/partition@0xc00000 unit name should not have
leading "0x"
arch/arm/boot/dts/zynq-cc108.dtb: Warning (unit_address_format): Node
/amba/spi@e000d000/flash@0/partition@0xd00000 unit name should not have
leading "0x"
arch/arm/boot/dts/zynq-cc108.dtb: Warning (unit_address_format): Node
/amba/spi@e000d000/flash@0/partition@0xf00000 unit name should not have
leading "0x"
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
ARM: dts: zynq: Add generic compatible string for I2C EEPROM
The at24 driver allows to register I2C EEPROM chips using different vendor
and devices, but the I2C subsystem does not take the vendor into account
when matching using the I2C table since it only has device entries.
But when matching using an OF table, both the vendor and device has to be
taken into account so the driver defines only a set of compatible strings
using the "atmel" vendor as a generic fallback for compatible I2C devices.
So add this generic fallback to the device node compatible string to make
the device to match the driver using the OF device ID table.
Signed-off-by: Javier Martinez Canillas <javier@dowhile0.org> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Michal Simek [Wed, 17 Jan 2018 14:21:38 +0000 (15:21 +0100)]
arm: zynq: Add missing address node name in microzed board
This patch is fixing issue reported by dtc:
arch/arm/boot/dts/zynq-microzed.dtb: Warning (unit_address_vs_reg): Node
/memory has a reg or ranges property, but no unit name
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
fpga: region: Add reset support to the fpga region
Many of the fpga regions have a reset that has to be
asserted after the bit file programming. Add support
for the same in case there is no reset phandle passed
no action is taken so it is backward compatible.
Signed-off-by: Shubhrajyoti Datta <shubhrajyoti.datta@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
drm: xlnx: zynqmp: Disable a plane when the fb format changes
The drm core doesn't explicitly disable a plane when format changes.
So add a check in the plane update functions if the new framebuffer
format has changed, and disable the plane for the format change.
Signed-off-by: Hyun Kwon <hyun.kwon@xilinx.com> Tested-by: Kuldeep Dave <kuldeepd@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
The driver's suspend/resume functions were buggy.
If UART node contains any child node in the DT and
the child is established a communication path with
the parent UART. The relevant /dev/ttyPS* node will
be not available for other operations.
If the driver is trying to do any operations like
suspend/resume without checking the tty->dev status
it leads to the kernel crash/hang.
This patch fix this issue by call the device_may_wake()
with the generic parameter of type struct device.
in the uart suspend and resume paths.
It also fixes a race condition in the uart suspend
path(i.e uart_suspend_port() should be called at the
end of cdns_uart_suspend API this path updates the same)
Signed-off-by: Nava kishore Manne <navam@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Hyun Kwon [Thu, 29 Mar 2018 00:49:19 +0000 (17:49 -0700)]
drm: xlnx: drv: Shutdown the drm device properly
If a drm device is released immediately in shutdown callback by calling
drm_put_device(), it gives a warnning as below because there are some
resources, drm gem objects, that are still used by fbdev emulation and
attached to the drm device.
xilinx: v4l2: dma: Bug fix in single plane support
This patch fixes bugs in single plane handling.
a. Passing v4l2 pixel format to framebuffer driver
b. Making dst_icg is zero, though it does not matter
in case of single plane formats
c. Correcting bytes per line for single plane 10 bit
formats
Derek Kiernan [Fri, 23 Mar 2018 19:43:22 +0000 (19:43 +0000)]
misc: xilinx-sdfec: Setting FEC_CODE change
Setting of the FEC_Code (LDPC/Turbo) is decided at using the device tree
only (at design time only), remove the setting of the value in the function
xsdfec_set_turbo and xsdfec_add_ldpc to function xsdfec_parse_of to ensure
that this is OK.
Signed-off-by: Derek Kiernan <Derek.Kiernan@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Derek Kiernan [Fri, 23 Mar 2018 19:43:19 +0000 (19:43 +0000)]
misc: xilinx-sdfec: Memory leak fix
In function xsdfec_add_ldpc and xsdfec_get_ldpc_code_params all failures
after successfully allocating "ldpc/ldpc_params" memory should free the
allocated memory before returning from the function.
Signed-off-by: Derek Kiernan <Derek.Kiernan@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Derek Kiernan [Thu, 15 Mar 2018 17:39:17 +0000 (17:39 +0000)]
misc: xilinx-sdfec: Fix open_count decrement when already open
In function xsdfec_dev_open if the atomic_dec_and_test detects that driver
is already opened a increment is required. This ensures the driver
instance can the reopened after it is released by the user.
Signed-off-by: Derek Kiernan <Derek.Kiernan@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
arm64: zynqmp: Set Dual Role Device Mode by default
This patch enables the dwc3 DRD mode by default. Doing so, will
make the same Image to be used for both Host & Device modes
based on the dr_mode dts parameter.
Signed-off-by: Anurag Kumar Vulisha <anuragku@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Hyun Kwon [Wed, 21 Mar 2018 17:18:09 +0000 (10:18 -0700)]
drm: xlnx: zynqmp_dp: Use the correct number of lanes
The number of lanes is configurable, thus the dynamically configured
value should be used instead of hard-coded max number of lanes. Use
the number of assigned lanes correctly.
While at it, don't print an error for phy probe deferral.
v4l: xilinx: dma: Return from mplane handling for enum formats
In current driver implemenataion, it is always falling back to
single plane handling after mplane handling. This patch ensures
that mplane handling return gracefully.
Michal Simek [Mon, 12 Mar 2018 15:13:57 +0000 (16:13 +0100)]
Bluetooth: hci_ll: Extend timeout between start and communication
Extend timeout between bt enable and communication with chip.
Flow control is automatic and driver is responding that device is ready
to accept messages but it doesn't need to be truth. Wait more time to
ensure that communication cat start.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Series-prefix: LINUX
Cover-letter:
zcu100 BT fixes
Hi,
I have cherry-picked mainline patches to make sure that we are using the
latest hci_ll code with fixes.
This adds an optional nvmem consumer to get a BD address from an external
source. The BD address is then set in the Bluetooth chip after the
firmware has been loaded.
This has been tested working with a TI CC2560A chip (in a LEGO MINDSTORMS
EV3).
Signed-off-by: David Lechner <david@lechnology.com> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
David Lechner [Fri, 8 Dec 2017 02:22:19 +0000 (20:22 -0600)]
Bluetooth: hci_ll: Add endianness conversion when setting baudrate
This adds an endianness conversion when setting the baudrate using a
vendor-specific command. Otherwise, bad things might happen on a big-
endian system.
In preparation to enabling -Wimplicit-fallthrough, mark switch cases
where we are expecting to fall through.
In this particular case, notice that I replaced the
"deliberate fall-through..." comment with a "fall through"
comment, which is what GCC is expecting to find.
Signed-off-by: Gustavo A. R. Silva <garsilva@embeddedor.com> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
David Lechner [Tue, 12 Dec 2017 21:59:16 +0000 (15:59 -0600)]
Bluetooth: hci_ll: add support for setting public address
This adds support for setting the public address on Texas Instruments
Bluetooth chips using a vendor-specific command.
This has been tested on a CC2560A chip. The TI wiki also indicates that
this command should work on TI WL17xx/WL18xx Bluetooth chips.
During review, there was some question as to the correctness of the byte
swapping since TI's documentation is not clear on this matter. This can
be tested with the btmgmt utility from bluez. The adapter must be powered
off to change the address. If the baswap() is omitted, address is reversed.
In case there is a issue in the future, here is the output of btmon during
the command `btmgmt public-addr 00:11:22:33:44:55`:
Bluetooth monitor ver 5.43
= Note: Linux version 4.15.0-rc2-08561-gcb132a1-dirty (armv5tejl) 0.707043
= Note: Bluetooth subsystem version 2.22 0.707091
= New Index: 00:17:E7:BD:1C:8E (Primary,UART,hci0) [hci0] 0.707106
@ MGMT Open: btmgmt (privileged) version 1.14 {0x0002} 0.707124
@ MGMT Open: bluetoothd (privileged) version 1.14 {0x0001} 0.707137
@ MGMT Open: btmon (privileged) version 1.14 {0x0003} 0.707540
@ MGMT Command: Set Public Address (0x0039) plen 6 {0x0002} [hci0] 11.167991
Address: 00:11:22:33:44:55 (CIMSYS Inc)
@ MGMT Event: Command Complete (0x0001) plen 7 {0x0002} [hci0] 11.175681
Set Public Address (0x0039) plen 4
Status: Success (0x00)
Missing options: 0x00000000
@ MGMT Event: Index Removed (0x0005) plen 0 {0x0003} [hci0] 11.175757
@ MGMT Event: Index Removed (0x0005) plen 0 {0x0002} [hci0] 11.175757
@ MGMT Event: Index Removed (0x0005) plen 0 {0x0001} [hci0] 11.175757
= Open Index: 00:17:E7:BD:1C:8E [hci0] 11.176807
< HCI Command: Vendor (0x3f|0x0006) plen 6 [hci0] 11.176975
00 11 22 33 44 55 .."3DU
> HCI Event: Command Complete (0x0e) plen 4 [hci0] 11.188260
Vendor (0x3f|0x0006) ncmd 1
Status: Success (0x00)
...
< HCI Command: Read Local Version Info.. (0x04|0x0001) plen 0 [hci0] 11.189859
> HCI Event: Command Complete (0x0e) plen 12 [hci0] 11.190732
Read Local Version Information (0x04|0x0001) ncmd 1
Status: Success (0x00)
HCI version: Bluetooth 2.1 (0x04) - Revision 0 (0x0000)
LMP version: Bluetooth 2.1 (0x04) - Subversion 6431 (0x191f)
Manufacturer: Texas Instruments Inc. (13)
< HCI Command: Read BD ADDR (0x04|0x0009) plen 0 [hci0] 11.191027
> HCI Event: Command Complete (0x0e) plen 10 [hci0] 11.192101
Read BD ADDR (0x04|0x0009) ncmd 1
Status: Success (0x00)
Address: 00:11:22:33:44:55 (CIMSYS Inc)
...
Signed-off-by: David Lechner <david@lechnology.com> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
projects / zynq / linux.git / log