From 2daf7a9bb73c59dfe6bb736a150336000b55301e Mon Sep 17 00:00:00 2001 From: dsengupta0628 Date: Tue, 21 Apr 2026 22:00:36 +0000 Subject: [PATCH] remove design specific info from doc Signed-off-by: dsengupta0628 --- docs/KahnsBFS_Spec.txt | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) diff --git a/docs/KahnsBFS_Spec.txt b/docs/KahnsBFS_Spec.txt index 5e18bef1..ab5b7598 100644 --- a/docs/KahnsBFS_Spec.txt +++ b/docs/KahnsBFS_Spec.txt @@ -237,19 +237,20 @@ Extracts design-size metrics (instance count, net count, IO count, cell area) at Running a full sweep across several designs: 1. Build OpenROAD with Kahn's: the flag sta_use_kahns_bfs defaults to 1. - 2. From flow/, choose the target and the design list. For example: - util/kahns_benchmark.sh -t finish -o kahns_bench_gf12 \ - $(ls -d designs/gf12/*/config.mk) + 2. From flow/, choose the target, the output directory, and the design list. + For example, pick one or more config.mk paths from a platform of interest + (under flow/designs//) and pass them on the command line: + util/kahns_benchmark.sh -t finish -o ... 3. While it runs, tail the most recent per-design stdout log to follow progress and verify the wrapper breadcrumb: - tail -f "$(ls -t kahns_bench_gf12/*.log | head -1)" | grep -i "kahns-wrap\|error" + tail -f "$(ls -t /*.log | head -1)" | grep -i "kahns-wrap\|error" 4. Validate correctness once designs finish: - util/kahns_size.sh --check kahns_bench_gf12 + util/kahns_size.sh --check Address any FAIL before trusting the runtime numbers. 5. Compare per-step runtimes: - util/kahns_compare.sh kahns_bench_gf12 + util/kahns_compare.sh Interpret in the context of design size: - util/kahns_size.sh kahns_bench_gf12 + util/kahns_size.sh 9.6 Additional conventions @@ -274,7 +275,7 @@ OpenSTA standalone: 6109/6109 tests PASS with Kahn's enabled. OpenROAD full regression: All tests PASS, including rmp.gcd_restructure (the test that surfaced both findings in Section 7), rsz (incremental netlist modification), and cts (buffer insertion with re-timing). -ORFS A/B runtime benchmarks (Section 9): in progress. An initial sweep across the gf12 and rapidus2hp design sets is running using util/kahns_benchmark.sh. Completed designs to date show Kahn's at parity or slightly slower on small designs (e.g. gf12/aes: +12s / +3% total over 375s baseline), consistent with the Section 8 prediction that the discovery-pass overhead dominates when the active graph is small. Larger designs (gf12/bp_quad, gf12/ariane133, bp_dual) are still pending; this section will be updated with their numbers and the per-step breakdown as each finishes. Correctness (netlist-size match between OFF and ON) is verified after each design via util/kahns_size.sh --check. +ORFS A/B runtime benchmarks (Section 9): in progress. An initial sweep across several platform/design combinations is running using util/kahns_benchmark.sh. Completed designs to date show Kahn's at parity or slightly slower on small designs (for a representative small design the measured overhead was roughly +3% / +12s over a ~375s baseline), consistent with the Section 8 prediction that the discovery-pass overhead dominates when the active graph is small. Larger designs are still pending; this section will be updated with their numbers and the per-step breakdown as each finishes. Correctness (netlist-size match between OFF and ON) is verified after each design via util/kahns_size.sh --check. 11. LIMITATIONS OF THE CURRENT APPROACH