Releases: slaclab/lcls-timing-core
Patch Release v3.8.4
Pull Requests Since v3.8.3
Unlabeled
- #176 - V3.8.3 trigger pipeline fix
- #173 - Add Ruckus Flag to Load XCI Instead of DCP
- #172 - Updating code comments for Ultrascale/Ultrascale+ GTs wrapper
Pull Request Details
Updating code comments for Ultrascale/Ultrascale+ GTs wrapper
Author: | Larry Ruckman ruckman@slac.stanford.edu |
Date: | Wed Sep 25 11:10:48 2024 -0700 |
Pull: | #172 (4 additions, 3 deletions, 4 files changed) |
Branch: | slaclab/stableClk-gt-comment |
Notes:
Description
- StableClk (which is GT's drpClk) in the IP core configured for 156.25MHz/2 (78.125MHz)
Add Ruckus Flag to Load XCI Instead of DCP
Author: | Larry Ruckman ruckman@slac.stanford.edu |
Date: | Wed Oct 2 09:59:32 2024 -0700 |
Pull: | #173 (29 additions, 19 deletions, 3 files changed) |
Branch: | slaclab/xci-option |
Notes:
The ruckus.tcl files that load IP cores now look for an
LCLS_TIMING_XCI
environment variable.
If it is set to 1, then the XCI file is loaded instead of the DCP.This is useful in simulation. For example, add this line to the target makefile so that the XCI is loaded when building the project for VCS simulation.
vcs: export LCLS_TIMING_XCI=1
V3.8.3 trigger pipeline fix
Author: | Larry Ruckman ruckman@slac.stanford.edu |
Date: | Thu Nov 14 12:48:51 2024 -0800 |
Pull: | #176 (90 additions, 50 deletions, 1 files changed) |
Branch: | slaclab/v3.8.3-trigger-pipeline-fix |
Notes:
I fixed a long standing problem (and a newly discovered one) where changing the trigger configuration parameters, namely delay and width, while the trigger pipeline is enabled causes errors that can overflow the pipeline (negative delay change) or produce the wrong delay (width change).
The one cycle push of delay onto the fifo was replaced with two pushes, each representing a transition of the output trigger state.
Patch Release v3.8.3
Pull Requests Since v3.8.2
Unlabeled
- #170 - Update TimingGtCoreWrapper.vhd
Pull Request Details
Update TimingGtCoreWrapper.vhd
Author: | Larry Ruckman ruckman@slac.stanford.edu |
Date: | Fri Sep 20 09:42:42 2024 -0700 |
Pull: | #170 (1 additions, 1 deletions, 1 files changed) |
Branch: | slaclab/ruck314/RX-RX_EQUALIZER_G |
Notes:
Description
- Updating the RX RX_EQUALIZER_G from
DFE
toLPM
- Related to slaclab/evr-card-g2#11
Patch Release v3.8.2
Pull Requests Since v3.8.1
Unlabeled
Pull Request Details
Update timing_ci.yml
Author: | Larry Ruckman ruckman@slac.stanford.edu |
Date: | Thu Jun 27 15:39:48 2024 -0700 |
Pull: | #166 (15 additions, 80 deletions, 1 files changed) |
Branch: | slaclab/ruck314-patch-1 |
Notes:
Description
- Using the reusable YAML CI scripts
updated clktime_186MHz
Author: | Larry Ruckman ruckman@slac.stanford.edu |
Date: | Wed Aug 14 08:27:37 2024 -0700 |
Pull: | #167 (5 additions, 5 deletions, 2 files changed) |
Branch: | slaclab/tcore_tokenizer |
Notes:
Updated the name of the module ClockTime
added ClockTime ports
Author: | Larry Ruckman ruckman@slac.stanford.edu |
Date: | Fri Aug 16 12:49:51 2024 -0700 |
Pull: | #169 (2 additions, 2 deletions, 1 files changed) |
Branch: | slaclab/tcore_tokenizerxx |
Issues: | #169 |
Notes:
Added ports: "steps, remainder, divisor" to ClockTime.vhd
Patch Release v3.8.1
Pull Requests Since v3.8.0
Unlabeled
- #164 - V3.7.0 streamtx evr
Pull Request Details
V3.7.0 streamtx evr
Author: | Matt Weaver weaver@slac.stanford.edu |
Date: | Mon Mar 11 10:39:57 2024 -0500 |
Pull: | #164 (32 additions, 22 deletions, 2 files changed) |
Branch: | slaclab/v3.7.0-streamtx-evr |
Notes:
Description
Minor Release v3.8.0
Pull Requests Since v3.7.7
Unlabeled
- #162 - Fix Rx issues
Pull Request Details
Fix Rx issues
Author: | Matt Weaver weaver@slac.stanford.edu |
Date: | Tue Dec 5 14:05:38 2023 -0600 |
Pull: | #162 (1876 additions, 2982 deletions, 11 files changed) |
Branch: | slaclab/debuggingTimingRx |
Notes:
Fixes a reset scheme issue on the GTH Rx alignment process
Patch Release v3.7.7
Patch Release v3.7.4
Full Changelog: v3.7.3...v3.7.4
Pull Requests Since v3.7.3
Unlabeled
Pull Request Details
Updating smurf/TimingGty_fixedlat.[xci,dcp]
Author: | Larry Ruckman ruckman@slac.stanford.edu |
Date: | Sat Jul 8 10:03:35 2023 -0700 |
Pull: | #159 (8 additions, 8 deletions, 2 files changed) |
Branch: | slaclab/TimingGty_fixedlat-smurf-update |
Notes:
Description
- adding cpllrefclksel_in & gtgrefclk_in ports to match LCLS-II API
Fixes for ultrascale+gth
Author: | Larry Ruckman ruckman@slac.stanford.edu |
Date: | Mon Jul 10 09:47:46 2023 -0700 |
Pull: | #160 (109 additions, 42 deletions, 5 files changed) |
Branch: | slaclab/fixesForUltrascale+gth |
Notes:
These were the changes I needed to make on the Ultrascale+ GTH for it to work with the detector boards.
Key changes:
1 set the clock to 156.25MHz to match the clock used
2 set SIM_CPLL_CAL_BYPASS to true
3 Timing receiver module now gets a new generic that selects from gty and gth
Patch Release v3.7.3
Pull Requests Since v3.7.2
Unlabeled
- #157 - V3.6.3 trigfifo flush
Pull Request Details
V3.6.3 trigfifo flush
Author: | Larry Ruckman ruckman@slac.stanford.edu |
Date: | Tue Jun 20 13:03:07 2023 -0700 |
Pull: | #157 (19 additions, 9 deletions, 4 files changed) |
Branch: | slaclab/v3.6.3-trigfifo-flush |
Notes:
Reset the serial delay FIFOs when an overflow is detected. Also, allow more status count registers to roll over.
Patch Release v3.7.2
Pull Requests Since v3.7.1
Unlabeled
Pull Request Details
Update timing_ci.yml
Author: | Larry Ruckman ruckman@slac.stanford.edu |
Date: | Sat Jun 3 08:23:10 2023 -0700 |
Pull: | #154 (2 additions, 9 deletions, 1 files changed) |
Branch: | slaclab/ESROGUE-619 |
Jira: | https://jira.slac.stanford.edu/issues/ESROGUE-619 |
Notes:
Description
- No longer generating anaconda for pre-release
Use Gtye4ChannelDummy in gtyUltrascale+ core wrapper
Author: | Larry Ruckman ruckman@slac.stanford.edu |
Date: | Wed Jun 14 14:12:02 2023 -0700 |
Pull: | #155 (1 additions, 1 deletions, 1 files changed) |
Branch: | slaclab/gtye4-fix |
Notes:
The
gtyUltrascle+/TimingGtCoreWrapper
module was usingGthe3ChannelDummy
instead ofGtye4ChannelDummy
.
ButGthe3ChannelDummy
doesn't even get loaded by ruckus for ultrascale+ designs.
This is now fixed.