Skip to content
Snippets Groups Projects

Resolve L2SDP-294

Merged Resolve L2SDP-294
L2SDP-294 into master
3 unresolved threads
Merged Reinier van der Walle requested to merge L2SDP-294 into master
3 unresolved threads

Closes L2SDP-294

Merge request reports

Pipeline #25001 passed

Stage: simulation
Stage: synthesis

Pipeline passed for b2e1332e on L2SDP-294

Merged by Eric KooistraEric Kooistra 3 years ago (Feb 14, 2022 2:42pm UTC)

Loading

Pipeline #25003 passed

Stage: simulation
Stage: synthesis

Pipeline passed for 6e08fa92 on master

Activity

Filter activity
  • Approvals
  • Assignees & reviewers
  • Comments (from bots)
  • Comments (from users)
  • Commits & branches
  • Edits
  • Labels
  • Lock status
  • Mentions
  • Merge request status
  • Tracking
applications/lofar2/libraries/sdp/src/vhdl/sdp_pkg.vhd
109 109 CONSTANT c_sdp_W_bf_weight : NATURAL := 16; -- = w in s(w, p), s = signed
110 110 CONSTANT c_sdp_W_bf_weight_fraction : NATURAL := 14; -- = p in s(w, p)
111 111 CONSTANT c_sdp_W_bf_weight_magnitude : NATURAL := c_sdp_W_bf_weight - c_sdp_W_bf_weight_fraction - 1; -- = 1
112 CONSTANT c_sdp_N_ring_mac : NATURAL := 3; -- for sdp_station_xsub_ring design.
  • Eric Kooistra
    Eric Kooistra @kooistra ·
    Owner

    c_sdp_N_ring_mac --> c_sdp_N_ring_nof_mac10g ? -- Waarom 3, is dat 2 voor ring via PCB + 1 voor ring via QSFP

    -- for sdp_station_xsub_ring design. --> klopt dit comment wel, het is toch 3 MAC10Gs voor 1 QSFP ring

  • Reinier van der Walle
    Reinier van der Walle @walle ·
    Author Contributor

    Waarom 3, is dat 2 voor ring via PCB + 1 voor ring via QSFP. Ja inderdaad. klopt dit comment wel, het is toch 3 MAC10Gs voor 1 QSFP ring. ja lijkt me wel het xsub_ring design gebruikt 1 lane waarvoor er 3 MACs gebruikt zijn.

    Met 3 MACs kun je een lane in positieve richting gebruiken en een lane in negatieve richting. Straks voor het "bf_ring" design waar we 2 lanes in positieve richting gebruiken, heb je dus 6 MACs nodig. Voor de full met XSUB en BF hebben we dan 9 MACs nodig.

  • Reinier van der Walle changed this line in version 2 of the diff

    changed this line in version 2 of the diff

  • Please register or sign in to reply
  • Eric Kooistra
    Eric Kooistra @kooistra started a thread on an old version of the diff
    • libraries/base/dp/src/vhdl/dp_fifo_fill_eop.vhd
    254 254 -- No need to transfer eop counter across clock domains for single clock
    255 255 ELSE
    256 256 wr_eop_busy <= '0'; -- To prevent inferred latch.
  • Reinier van der Walle added 2 commits

    added 2 commits

    Compare with previous version

  • Eric Kooistra
    Eric Kooistra @kooistra started a thread on an old version of the diff
    • applications/lofar2/libraries/sdp/src/vhdl/sdp_pkg.vhd
    109 109 CONSTANT c_sdp_W_bf_weight : NATURAL := 16; -- = w in s(w, p), s = signed
    110 110 CONSTANT c_sdp_W_bf_weight_fraction : NATURAL := 14; -- = p in s(w, p)
    111 111 CONSTANT c_sdp_W_bf_weight_magnitude : NATURAL := c_sdp_W_bf_weight - c_sdp_W_bf_weight_fraction - 1; -- = 1
    112 CONSTANT c_sdp_N_ring_mac10g : NATURAL := 3; -- for sdp_station_xsub_ring design.
  • Reinier van der Walle added 1 commit

    added 1 commit

    Compare with previous version

  • Eric Kooistra mentioned in commit 6e08fa92

    mentioned in commit 6e08fa92

  • Eric Kooistra merged

    merged

  • Daniel van der Schuur unassigned @walle

    unassigned @walle

    • Please register or sign in to reply