[Libre-soc-dev] Unexpected clock connexions.

[Staf Verhaegen (FibraServi)]

On 3/06/2021 17:21, Jean-Paul Chaput wrote:
> On Thu, 2021-06-03 at 17:06 +0200, Staf Verhaegen (FibraServi) wrote:
>> 	Error verifying signature: Cannot verify message signature: Incorrect message
>> format
>> On 3/06/2021 16:18, Luke Kenneth Casson Leighton wrote:
>>> Jean-Paul i am having difficulty getting the domains synchronised with
>>> nmigen under time pressure.
>>>
>>> i am recommending we *BYPASS* the PLL_OUT (coresync_clk) and connect
>>> SYS_CLK *DIRECTLY* to TestIssuer and DMI/JTAG.
>> Can't you define the PLL input clock as pll_clk_in (or ref_clk) and then connect
>> pll_clk to sys_clk and let pll_clk_in (or ref_clk) come from IO ?
>> If at all possible I would like to use the PLL for clocking ls180.
>    Given the time constraints, I will directly patch the VHDL generated
>    files after Yosys+blif2vst. As such, isolating the PLL *or* reconnecting
>    everything to the PLL clock output will, I think, give the same amount
>    of work. We just have to decide quickly.
Can you easily connect sys_clk to PLL clock input and then PLL output to 
sys_clk ?
Staf.

-- 
Chips want to be free.