rsz: Global sizing by erendn · Pull Request #10599 · The-OpenROAD-Project/OpenROAD

erendn · 2026-06-05T20:06:28Z

Summary

This PR introduces a new repair timing policy, Lagrangian relaxation-based global sizing.

Type of Change

New feature

Impact

The new global sizing policy is disabled by default now. Initial QoR experiments show promising results when enabled with repair_timing -phases "GLOBAL_SIZING LEGACY LAST_GASP".

Verification

I have verified that the local build succeeds (./etc/Build.sh).
I have run the relevant tests and they pass.
My code follows the repository's formatting guidelines.
I have included tests to prevent regressions.
I have signed my commits (DCO).

Signed-off-by: Eren Dogan <erendogan@google.com>

gemini-code-assist

Code Review

This pull request introduces a Lagrangian-Relaxation-driven global sizing and Vt assignment optimization phase (GLOBAL_SIZING) to the resizer. It adds the LRSubproblem class to evaluate per-gate Lagrangian subproblems concurrently and the GlobalSizingPolicy class to drive the outer optimization loop. It also introduces worker-safe overloads for gate delay calculations in Resizer to support parallel evaluation. The review comments correctly identify critical null-pointer dereference risks in both LRSubproblem::isDataArc and GlobalSizingPolicy::isDataArc where edge->role() can be null.

erendn · 2026-06-05T20:57:37Z

@QuantamHD @mguthaus FYI

Signed-off-by: Eren Dogan <erendogan@google.com>

maliberty · 2026-06-06T03:08:27Z

I've merged with master and started a secure CI where the repair_timing in cts.tcl uses the above phases.

maliberty · 2026-06-06T03:08:53Z

@codex review

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 4c3e619379

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-06-06T03:15:56Z

+        = r.best_is_downsize ? downsize_accept_tol : upsize_accept_tol;
+    if (r.best_cost < r.baseline_cost * (1.0f - tol)) {
+      sta::LibertyCell* prev = network_->libertyCell(r.inst);
+      if (subproblem_->applyReplacement(r.inst, r.best_cell)) {


Re-check DRC before applying stale Jacobi decisions

When a driver and one of its fanout loads are both resized in the same Jacobi sweep, each GateDecision was DRC-filtered against the pre-sweep snapshot, but this serial apply step installs the cell without revalidating against the live design after earlier replacements. A load upsized earlier in the loop can increase the driver's actual output load after the driver's snapshot passed checkOutputMaxCap/slew, so GLOBAL_SIZING can leave max-cap/max-slew violations that repair_timing does not check at finish.

Useful? React with 👍 / 👎.

chatgpt-codex-connector · 2026-06-06T03:15:56Z

+        = r.best_is_downsize ? downsize_accept_tol : upsize_accept_tol;
+    if (r.best_cost < r.baseline_cost * (1.0f - tol)) {
+      sta::LibertyCell* prev = network_->libertyCell(r.inst);
+      if (subproblem_->applyReplacement(r.inst, r.best_cell)) {


Stop applying upsizes once max utilization is reached

For repair_timing -setup -max_utilization ... -phases GLOBAL_SIZING, this loop can keep accepting area-increasing replacements without any resizer_.overMaxArea() guard or rollback. Other setup policies stop when the area limit is hit, but here the phase may run past the requested utilization and then fail in OptimizationPolicy::finish() with RSZ-0025, even though a partial set of replacements could have respected the user limit.

Useful? React with 👍 / 👎.

maliberty · 2026-06-06T11:27:07Z

The two significant issues in the public CI are cva6/asap7 crashed:

[2026-06-06T04:06:53.732Z] /usr/include/c++/11/bits/stl_vector.h:1045: std::vector<_Tp, _Alloc>::reference std::vector<_Tp, _Alloc>::operator[](std::vector<_Tp, _Alloc>::size_type) [with _Tp = float; _Alloc = std::allocator<float>; std::vector<_Tp, _Alloc>::reference = float&; std::vector<_Tp, _Alloc>::size_type = long unsigned int]: Assertion '__n < this->size()' failed.

[2026-06-06T04:06:53.732Z] Signal 6 received

[2026-06-06T04:06:53.732Z] Stack trace:

[2026-06-06T04:06:53.732Z]  0# 0x0000000000EDBC37 in /OpenROAD-flow-scripts/tools/install/OpenROAD/bin/openroad

[2026-06-06T04:06:53.732Z]  1# 0x0000000000042520 in /lib/x86_64-linux-gnu/libc.so.6

[2026-06-06T04:06:53.732Z]  2# pthread_kill in /lib/x86_64-linux-gnu/libc.so.6

[2026-06-06T04:06:53.732Z]  3# raise in /lib/x86_64-linux-gnu/libc.so.6

[2026-06-06T04:06:53.732Z]  4# abort in /lib/x86_64-linux-gnu/libc.so.6

[2026-06-06T04:06:53.732Z]  5# 0x0000000000ED1418 in /OpenROAD-flow-scripts/tools/install/OpenROAD/bin/openroad

[2026-06-06T04:06:53.732Z]  6# rsz::GlobalSizingPolicy::projectFlowBalance(rsz::LRParams const&) in /OpenROAD-flow-scripts/tools/install/OpenROAD/bin/openroad

[2026-06-06T04:06:53.732Z]  7# rsz::GlobalSizingPolicy::iterate() in /OpenROAD-flow-scripts/tools/install/OpenROAD/bin/openroad

[2026-06-06T04:06:53.732Z]  8# rsz::Optimizer::run() in /OpenROAD-flow-scripts/tools/install/OpenROAD/bin/openroad

[2026-06-06T04:06:53.732Z]  9# rsz::Resizer::repairSetup(double, double, int, int, int, bool, bool, std::vector<rsz::MoveType, std::allocator<rsz::MoveType> > const&, char const*, bool, bool, bool, bool, bool, bool, bool, bool) in /OpenROAD-flow-scripts/tools/install/OpenROAD/bin/openroad

and ibex/ihp fails in legalization

 [ERROR DPL-0036] Detailed placement failed inside DPL.

To reproduce, use your branch with

diff --git a/flow/scripts/cts.tcl b/flow/scripts/cts.tcl
index 1378c9e7b9..e9cdaea7e5 100644
--- a/flow/scripts/cts.tcl
+++ b/flow/scripts/cts.tcl
@@ -66,7 +66,7 @@ if { !$::env(SKIP_CTS_REPAIR_TIMING) } {
     write_lec_verilog 4_before_rsz_lec.v
   }
 
-  repair_timing_helper
+  repair_timing_helper -phases "GLOBAL_SIZING LEGACY LAST_GASP"

maliberty · 2026-06-06T11:44:04Z

Public results are mixed

maliberty · 2026-06-06T11:46:22Z

Even at CTS where the change is applied

Signed-off-by: Eren Dogan <erendogan@google.com>

QuantamHD · 2026-06-07T04:17:10Z

Don't most of the public ci designs meet timing?

maliberty · 2026-06-07T04:33:20Z

Don't most of the public ci designs meet timing?

no

Signed-off-by: Eren Dogan <erendogan@google.com>

erendn · 2026-06-17T21:32:50Z

@maliberty @jhkim-pii I think this is ready now.

maliberty · 2026-06-17T22:45:47Z

@codex review

maliberty · 2026-06-17T22:50:09Z

+  lr_params_.include_clock_network = utl::readEnvarBool(
+      kIncludeClockNetworkEnv, lr_params_.include_clock_network);
+  lr_params_.setup_slack_margin = utl::readEnvarFloat(
+      kLrSetupSlackMarginEnv, lr_params_.setup_slack_margin);
+  lr_params_.max_iterations
+      = utl::readEnvarInt(kLrMaxIterationsEnv, lr_params_.max_iterations);
+  lr_params_.beta = utl::readEnvarFloat(kLrBetaEnv, lr_params_.beta);
+  lr_params_.mu_exponent
+      = utl::readEnvarFloat(kLrMuExponentEnv, lr_params_.mu_exponent);
+  lr_params_.lambda_floor
+      = utl::readEnvarFloat(kLrLambdaFloorEnv, lr_params_.lambda_floor);
+  lr_params_.timing_bias
+      = utl::readEnvarFloat(kLrTimingBiasEnv, lr_params_.timing_bias);
+  lr_params_.budget_safety_factor = utl::readEnvarFloat(
+      kLrBudgetSafetyFactorEnv, lr_params_.budget_safety_factor);
+}


Passing controls through envars is fine for development testing but should be removed for merge. These should be promoted to command options.

Perhaps add these to set_opt_config/reset_opt_config/report_opt_config?

Or we may take this approach. Replacing the envars to constants and merge the PR. For development, revert back to envars locally.

What's wrong with command args?

I think command arguments are good, as long as there aren't too many of them.

So long as they have reasonable defaults they are preferrable to hidden implicit ones. These seem like auto-tunable knobs.

I will remove these and add Tcl command options.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: c39faddb64

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-06-17T23:06:09Z

+        if (static_cast<size_t>(id) >= lambda_.size()) {
+          continue;
+        }


Rebuild LR state after replacement creates new edges

When a sweep replaces cells, this path explicitly skips any timing edge whose id was minted after allocate() sized lambda_; those same new edges are also ignored when snapshots sum output pressure, so subsequent LR iterations can treat already-resized gates as having no timing multiplier even if they are still on the critical path. In multi-iteration GLOBAL_SIZING runs that change Liberty timing arcs, this makes later sizing decisions use an incomplete objective rather than the current timing graph.

Useful? React with 👍 / 👎.

This is a known performance issue. It won't crash but potentially ignore new arcs. I will fix this in the future.

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Add global sizing

9cf00eb

Signed-off-by: Eren Dogan <erendogan@google.com>

github-actions Bot added the size/XL label Jun 5, 2026

gemini-code-assist Bot reviewed Jun 5, 2026

View reviewed changes

Comment thread src/rsz/src/LRSubproblem.cc

Comment thread src/rsz/src/policy/GlobalSizingPolicy.cc

rsz: clang-tidy

4c3e619

Signed-off-by: Eren Dogan <erendogan@google.com>

chatgpt-codex-connector Bot reviewed Jun 6, 2026

View reviewed changes

erendn added 3 commits June 6, 2026 12:29

rsz: Fix edge id overflow

9a56ff0

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Prevent max area overflow

2dd3854

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Early exit when timing is met

75c93ef

Signed-off-by: Eren Dogan <erendogan@google.com>

erendn added 4 commits June 7, 2026 17:38

rsz: Exclude clock drivers

844d2cf

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Add downsizing budget

7cabee9

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Fix vertex id overflow

0b9d52c

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: clang-tidy

7992074

Signed-off-by: Eren Dogan <erendogan@google.com>

jhkim-pii mentioned this pull request Jun 16, 2026

rsz: Add presizing mode for global sizing (Not for merge) #10669

Closed

erendn added 5 commits June 16, 2026 20:01

rsz: Fix global sizing tests

f214a84

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Add pre-sizing option to global sizing

ef8a792

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Add global sizing env vars

bd99741

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Restore best checkpoint with pre-size

0eec239

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Add flag to include clock networks in global sizing

c39fadd

Signed-off-by: Eren Dogan <erendogan@google.com>

erendn marked this pull request as ready for review June 17, 2026 21:32

erendn requested review from a team as code owners June 17, 2026 21:32

erendn requested a review from osamahammad21 June 17, 2026 21:32

erendn requested a review from precisionmoon June 17, 2026 21:32

maliberty reviewed Jun 17, 2026

View reviewed changes

chatgpt-codex-connector Bot reviewed Jun 17, 2026

View reviewed changes

erendn added 5 commits June 18, 2026 13:20

rsz: Honor clock network exclusion during pre-size

088c003

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Delete unused file

d859b4c

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Add Tcl commands for global sizing

f304eb3

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Update BUILD

53e378c

Signed-off-by: Eren Dogan <erendogan@google.com>

rsz: Update README check

6054762

Signed-off-by: Eren Dogan <erendogan@google.com>

jhkim-pii approved these changes Jun 23, 2026

View reviewed changes

maliberty approved these changes Jun 24, 2026

View reviewed changes

maliberty merged commit 9eaf7b1 into The-OpenROAD-Project:master Jun 24, 2026
16 checks passed

Uh oh!

Conversation

erendn commented Jun 5, 2026

Summary

Type of Change

Impact

Verification

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Uh oh!

Uh oh!

erendn commented Jun 5, 2026

Uh oh!

maliberty commented Jun 6, 2026

Uh oh!

maliberty commented Jun 6, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Jun 6, 2026

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector Bot Jun 6, 2026

Choose a reason for hiding this comment

Uh oh!

maliberty commented Jun 6, 2026

Uh oh!

maliberty commented Jun 6, 2026

Uh oh!

maliberty commented Jun 6, 2026

Uh oh!

QuantamHD commented Jun 7, 2026

Uh oh!

maliberty commented Jun 7, 2026

Uh oh!

erendn commented Jun 17, 2026

Uh oh!

maliberty commented Jun 17, 2026

Uh oh!

maliberty Jun 17, 2026

Choose a reason for hiding this comment

Uh oh!

precisionmoon Jun 17, 2026

Choose a reason for hiding this comment

Uh oh!

jhkim-pii Jun 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

maliberty Jun 18, 2026

Choose a reason for hiding this comment

Uh oh!

jhkim-pii Jun 18, 2026

Choose a reason for hiding this comment

Uh oh!

maliberty Jun 18, 2026

Choose a reason for hiding this comment

Uh oh!

erendn Jun 18, 2026

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Jun 17, 2026

Choose a reason for hiding this comment

Uh oh!

erendn Jun 18, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Reviewers

jhkim-pii Jun 18, 2026 •

edited

Loading