For the source controller use the pause signal that has been properly
transferred to the source clock domain rather than the pause signal from
the request clock domain.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
When having multiple DMA cores sharing the same constraint file Vivado
seems to apply the constraints from the first core to all the other cores
when re-running synthesis and implementation from within the Vivado GUI.
This causes wrong timing constraints if the DMA cores have different
configurations. To avoid this issue use a TTCL template that generates a
custom constraint file for each DMA core instance.
This also allows us to drop the asynchronous clock detection hack from the
constraint file and move it to the template and only generate the CDC
constraints if the clock domains are asynchronous.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
The converters on the usdrx1 generate 2.5GB/s. This more than we can
transport over the HP interconnects to the system memory.
Add a dedicated DDR FIFO to design which can be used to buffer the data
before it is transferred to the main memory.
Also increase the interconnect clock rate from 100MHz to 200MHz and the DMA
FIFO size from 4 to 8, so we can transfer the captured data faster to the
main memory.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
The memory mapped AXI interfaces for the AXI-DMAC are uni-directional.
Which means they are either write-only or read-only. Unfortunately the
Altera tools can't handle this, so we had to add dummy signals for the
other direction.
The Xilinx tools on the other hand handle uni-directional AXI interfaces
and in fact IPI can do a better job and use less resources when creating
the AXI interconnects when it knows that the interface is uni-directional.
So always disable the dummy ports for the IPI package.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Supported carrier are ZC706 and RFSOM.
The synchronization pulse is automatically generated by the master terminal, when TDD mode is enabled.
By default a terminal is slave, software must write 0x01 into TDD register 0x50.
Add tdd_gated_[tx/rx]_dmapath control bits to the TDD logic. With these control line, the user can choose between gated and free-running (like in FDD mode) data flow control.
With lower buswidth, if all 4 channels are captured some samples are lost
With fifo size of 64, there are timing violations in the DMAC
With this configuration, 65536 samples could be captured from all 4 channels with no sample lost
Because of the DMAC destination bus is 256, the number of samples to be captured must be a multiple of 16, otherwise the system will freeze. This will be corrected in software
- removed the delay controller from the top file and added it inside the util_gmii_to_rgmii core
- removed delay related xdc constraints as they are not needed
Since we are just doing a loopback all the logic is contained within the
IO bank. By using a BUFIO instead of a BUFG we avoid having to route the
clock signal from the IO bank to the middle of the FPGA and back to the IO
bank. This reduces the skew between clock and the data signals and makes
sure that the we can use the same design over a range of different
resolutions without having to calibrate the delay.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
The ADV7511 samples on the rising edge. Update the data on the falling
edge, this gives us a larger margin and improved signal stability.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Those were added by mistake. It does not seem to be legal Verilog, but for
some reason Vivado accepts it.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
ffff and 0000 are always reserved control characters when using embedded
syncs. So make sure that we never have them in the pixel data, even when
running in full-range mode.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Add a control bit to the register map that allows to bypass the chroma
sub-sampler in the axi_hdmi_tx core. This is primarily interned to be used
to send the test-pattern directly to the HDMI transmitter without modifying
it.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Connect the raw HDMI data as generated by the ADV7604 to the ILA. For
debugging it is quite useful to be able to compare the data before and
after conversion pipeline.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Istvan Csomortani
Lars-Peter Clausen
Adrian Costina
Paul Cercueil