- 15 Jun, 2015 5 commits
-
-
Lucas Russo authored
-
Lucas Russo authored
This is necessary as hdlmake 2.1 is not working as expect when generating Vivado project files.
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
-
- 13 May, 2015 5 commits
-
-
Lucas Russo authored
Now, the dsp-cores repository changed the name of the position calculation core. So, we update it here.
-
Lucas Russo authored
Now, dsp-cores repository employs a new variable in which we select the machine we want. Currently, we support UVX (130MSPS ADC) and Sirius (130MSPS ADC).
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
These cosntraints were missing the correct wildcard "*", in order to be applied to all input pins. In fact, using wildcard "?" was obviously applying the constraints only to fmc2_adc3_data_i[0] to fmc2_adc3_data_i[9], leaving fmc2_adc3_data_i[10] to fmc2_adc3_data_i[15] out
-
- 22 Apr, 2015 2 commits
-
-
Lucas Russo authored
-
Lucas Russo authored
As we have transfered all of the BPM project to another account, we also update the submodule remotes.
-
- 14 Apr, 2015 3 commits
-
-
Lucas Russo authored
Now, we only have one delta-sigma component instantiation at the FOFB rate, working iterativelly with one clock cycle per input bit. So, we need a total of 32 clock cycles, corresponding to 32 input bits. The FOFB decimation rate is 980, so we could have a CE up to 980/32 and still produce correct results. However, in order to minimize latency we use the minimum CE avaiable to give an extra timing margin and still reduce latency. In our case, this value is 2, the ADC CE.
-
Lucas Russo authored
-
Lucas Russo authored
-
- 07 Apr, 2015 2 commits
-
-
Lucas Russo authored
-
Lucas Russo authored
-
- 27 Mar, 2015 3 commits
-
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
Now, we have the possibility to clock each wb_acq_core module independently and synchronize everyone to an external memory clock (ext_clk_i). This should improve the decoupling of FMCs clock trees.
-
- 25 Mar, 2015 1 commit
-
-
Lucas Russo authored
-
- 24 Mar, 2015 1 commit
-
-
Lucas Russo authored
These setting were causing flashing erros in our current boards (AFCv3)
-
- 17 Mar, 2015 16 commits
-
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
Previously, the RAM current address was read in the same cycle as STB and CYC. As the default Wishbone signals are 0's, the RAM read address was 0, which its contents are 0's for this module. As the RAM takes one clock cycle to return the requested data, the Wishbone data output was reading the previous RAM data, which corresponds to the data in RAM address 0. This was happening regardless of the Wishbone address and the Wishbone data output was always one clock cycle too late, which is not interpreted as valid by the Wishbone Crossbar Switch module.
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
-
Lucas Russo authored
-
Andrzej Wojeński authored
This module recives some information (uTCA board ID, uTCA slot number and some sensor readings) from LPC via SPI.
-
- 10 Mar, 2015 1 commit
-
-
Lucas Russo authored
-
- 09 Mar, 2015 1 commit
-
-
Lucas Russo authored
-