Commit Graph

201 Commits

Author SHA1 Message Date
gatecat
ddb084e9a8 archapi: Use arbitrary rather than actual placement in predictDelay
This makes predictDelay be based on an arbitrary belpin pair rather
than a arc of a net based on cell placement. This way 'what-if'
decisions can be evaluated without actually changing placement;
potentially useful for parallel placement.

A new helper predictArcDelay behaves like the old predictDelay to
minimise the impact on existing passes; only arches need be updated.

Signed-off-by: gatecat <gatecat@ds0.me>
2021-12-19 17:15:15 +00:00
Maciej Kurc
1db3a87c62 Code formatting
Signed-off-by: Maciej Kurc <mkurc@antmicro.com>
2021-09-29 14:59:09 +02:00
Maciej Kurc
76f5874ffc Brought back printout of critical path source file references, added clk-to-q, source and setup segment types
Signed-off-by: Maciej Kurc <mkurc@antmicro.com>
2021-09-29 10:16:45 +02:00
Maciej Kurc
1ed692aca9 Shifted moving of data containers after printing
Signed-off-by: Maciej Kurc <mkurc@antmicro.com>
2021-09-28 18:01:30 +02:00
Maciej Kurc
9018782eaa Added a commandline option controlled writeout of per-net timing details
Signed-off-by: Maciej Kurc <mkurc@antmicro.com>
2021-09-28 17:42:51 +02:00
Maciej Kurc
6deff56e83 Moved timing result report storage to the context, added its writeout to the current utilization and fmax report
Signed-off-by: Maciej Kurc <mkurc@antmicro.com>
2021-09-28 17:42:51 +02:00
Maciej Kurc
c6dc1f535a Added reporting critical paths in JSON format
Signed-off-by: Maciej Kurc <mkurc@antmicro.com>
2021-09-28 17:42:51 +02:00
Maciej Kurc
d8571b6c00 Decoupled critical path report generation from its printing
Signed-off-by: Maciej Kurc <mkurc@antmicro.com>
2021-09-28 17:42:51 +02:00
Maciej Kurc
12adbb81b1 Switched to JSON format for timing analysis report
Signed-off-by: Maciej Kurc <mkurc@antmicro.com>
2021-09-28 17:42:51 +02:00
Maciej Kurc
99ae5ef38e Added writing a CSV report with timing analysis of each net branch
Signed-off-by: Maciej Kurc <mkurc@antmicro.com>
2021-09-28 17:42:51 +02:00
gatecat
4730a4f339 timing: Always use max delay for required time
Signed-off-by: gatecat <gatecat@ds0.me>
2021-09-18 11:04:39 +01:00
gatecat
287a860283 timing: Fix slack for unconstrained clocks
Signed-off-by: gatecat <gatecat@ds0.me>
2021-09-18 10:43:19 +01:00
gatecat
f207068ee2 router2: Add experimental timing-driven ripup option
Signed-off-by: gatecat <gatecat@ds0.me>
2021-08-15 16:22:03 +01:00
gatecat
4ac00af6fa basectx: Add a field to store timing results
Signed-off-by: gatecat <gatecat@ds0.me>
2021-07-29 12:52:13 +01:00
gatecat
14c676ab49 timing: Allow overriding of route delays
Signed-off-by: gatecat <gatecat@ds0.me>
2021-07-28 11:20:28 +01:00
gatecat
76070a7647 timing: Fix out-of-bounds histogram bins in all cases
Signed-off-by: gatecat <gatecat@ds0.me>
2021-07-10 23:44:21 +01:00
gatecat
889c295baf sta: Fix a memory error introduced by the dict move
Signed-off-by: gatecat <gatecat@ds0.me>
2021-06-17 12:06:26 +01:00
gatecat
2ffb081442 Fixing old emails and names in copyrights
Signed-off-by: gatecat <gatecat@ds0.me>
2021-06-12 13:22:38 +01:00
gatecat
b8a68f5f35 Using hashlib in timing
Signed-off-by: gatecat <gatecat@ds0.me>
2021-06-02 15:05:20 +01:00
gatecat
579b98c596 Use hashlib for core netlist structures
Signed-off-by: gatecat <gatecat@ds0.me>
2021-06-02 14:27:56 +01:00
gatecat
ece10c3e04 timing: Fix domain init when loops are present
Signed-off-by: gatecat <gatecat@ds0.me>
2021-04-13 09:23:08 +01:00
gatecat
8a4bf3a780 timing: Integration tweaks
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-05 10:04:35 +00:00
gatecat
98d1c5a411 timing: Skip route delays for unplaced/nullptr cells
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 11:34:20 +00:00
gatecat
1ff2023f32 timing: Replace all users of criticality with new engine
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 11:29:11 +00:00
gatecat
ebc2527368 timing: Use new engine for HeAP
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 10:29:36 +00:00
gatecat
bbf5a7d461 timing: Add support for critical path printing
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 10:29:36 +00:00
gatecat
e681e0f14c timing: Slack and criticality computation
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 10:29:36 +00:00
gatecat
296e6d10c2 timing: Produce plausible Fmax figure
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 10:29:36 +00:00
gatecat
541376f8cc timing: Add Fmax printing for debugging
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 10:29:36 +00:00
gatecat
16e7bba87b timing: Add backwards path walking
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 10:29:36 +00:00
gatecat
0528ceead1 timing: Add forward path walking
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 10:29:36 +00:00
gatecat
9c8d1bd6e3 timing: Compute domain pairs
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 10:29:36 +00:00
gatecat
534e69fbff timing: Add port-domain tracking
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 10:29:36 +00:00
gatecat
7a546b1554 timing: Add topological sort from Yosys
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 10:29:36 +00:00
gatecat
d0772ce1e3 timing: Import cell delays to our own structures
Signed-off-by: gatecat <gatecat@ds0.me>
2021-03-04 10:29:36 +00:00
gatecat
c0a7cff304 Demote the 'no clocks' warning to info and make clearer
Signed-off-by: gatecat <gatecat@ds0.me>
2021-02-20 20:15:52 +00:00
gatecat
7922b3bfc4 Replace DelayInfo with DelayPair/DelayQuad
This replaces the arch-specific DelayInfo structure with new DelayPair
(min/max only) and DelayQuad (min/max for both rise and fall) structures
that form part of common code.

This further reduces the amount of arch-specific code; and also provides
useful data structures for timing analysis which will need to delay
with pairs/quads of delays as it is improved.

While there may be a small performance cost to arches that didn't
separate the rise/fall cases (arches that aren't currently separating
the min/max cases just need to be fixed...) in DelayInfo, my expectation
is that inlining will mean this doesn't make much difference.

Signed-off-by: gatecat <gatecat@ds0.me>
2021-02-19 11:31:33 +00:00
gatecat
535723f414 Start making use of getBelPinsForCellPin API
This replaces getNetinfoSinkWire with 3 new functions for different use
cases.

At the moment all existing code has been moved to getNetinfoSinkWire
with phys_idx=0 so the build doesn't break; but this won't yet function
properly with more than one sink. But it provides a base on which to
work on refactoring the routers to support this case.

Signed-off-by: gatecat <gatecat@ds0.me>
2021-02-10 14:18:12 +00:00
D. Shah
9388df19d3 refactor: Replace getXName().c_str(ctx) with ctx->nameOfX
This makes the ongoing migration to IdStringList easier.

Signed-off-by: D. Shah <dave@ds0.me>
2021-02-02 17:00:14 +00:00
D. Shah
94e8847d67 cleanup: Spelling fixes
Signed-off-by: D. Shah <dave@ds0.me>
2021-01-28 15:19:06 +00:00
D. Shah
6ecf7f86c8 cleanup: Remove dead/unused code
Note that some '#if 0' code that might still be useful for debugging in
the future has been retained.

Signed-off-by: D. Shah <dave@ds0.me>
2021-01-28 14:59:13 +00:00
Ross Schlaikjer
c30cadd19c
No longer need fstream include 2020-08-30 18:22:05 -04:00
Ross Schlaikjer
cba4753c22
Only print filenames for now, default on 2020-08-30 18:19:41 -04:00
Ross Schlaikjer
a8c110b045
Add option to print critical path source code
In order to make debugging the critical path easier, add an option that
will log the location each net was defined, if known.
If the file that contains the definition is known, and is readable, also
print the part of the source HDL responsible for the signal definition.
2020-08-30 17:43:29 -04:00
David Shah
109aa63474 clangformat
Signed-off-by: David Shah <dave@ds0.me>
2020-08-20 09:46:49 +01:00
David Shah
3611b7120c timing: Fix counting of fanin in out-of-context mode
Signed-off-by: David Shah <dave@ds0.me>
2020-08-12 10:42:26 +01:00
Eddie Hung
e6b85f1bc0 Fix embarassing use of topographical when meaning topological 2020-05-14 08:55:28 -07:00
David Shah
564f40f6db timing: Improve robustness to dangling/undriven logic
Signed-off-by: David Shah <dave@ds0.me>
2020-03-17 10:07:21 +00:00
David Shah
f28186bb1b timing: Fix critical path walking for ECP5
Signed-off-by: David Shah <dave@ds0.me>
2019-11-18 21:34:32 +00:00
David Shah
fba7c2caef timing: Improve support for out-of-context flows
Signed-off-by: David Shah <dave@ds0.me>
2019-08-07 16:51:29 +01:00