[J-core] We have started!
D. Jeff Dionne
jeff at coresemi.io
Mon Mar 16 02:57:45 UTC 2020
Forgot to attach this inital, quick and dirty synthesis report
2.52. Printing statistics.
=== cpu ===
Number of wires: 10384
Number of wire bits: 16211
Number of public wires: 10384
Number of public wire bits: 16211
Number of memories: 0
Number of memory bits: 0
Number of processes: 0
Number of cells: 11281
CCU2C 199
L6MUX21 1087
LUT4 6745
MULT18X18D 2
PFUMX 2156
TRELLIS_DPR16X4 32
TRELLIS_FF 1060
J.
On 月, 3月 16, 2020 at 11:56 午前, D. Jeff Dionne
<jeff at coresemi.io> wrote:
> I did a quick feasibility check of the CPU with GHDL synth and yosys,
> it looks like the 45k gate device would be better for a full SMP
> system with DDR RAM. The GHDL and Yosys tool flow is getting pretty
> good now, but we've still not managed to get a bitstream out of it
> and nextpnr for somehting meaningful... rapidly advancing though.
>
> The sticking point for this board is the SDRAM... J-Core memory
> controller only supports DDR, DDR2 and LPDDR. There are a few boards
> with this issue that J-Core could support if we were to write a
> simple SDRAM controller, or I suppose we could bolt someone else's
> on. Still a bit of work to teach the SoC generator how to connect
> the bus bridges and cache controllers.
>
> J.
>
> On 日, 3月 15, 2020 at 11:44 午後, Goran Mahovlić
> <goran.mahovlic at gmail.com> wrote:
>> Hopefully someone will now port J-CORE to ULX3S
>>
>> https://www.crowdsupply.com/radiona/ulx3s
>>
>> 12F should be more then enough!
>>
>> We already have Litex linux and SaxonSoc linux samples
>>
>> https://twitter.com/lawriegriffiths/status/1238030666719207425
>>
>> Goran
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.j-core.org/pipermail/j-core/attachments/20200316/c6c2523c/attachment.html>
More information about the J-core
mailing list