feat!: new semantic for bit-shifts by guipublic · Pull Request #9373 · noir-lang/noir

guipublic · 2025-08-01T09:22:33Z

Description

Problem*

Resolves #9022

Summary*

Bit-shifts now requires that rhs and lhs have the same type (instead of u8 for rhs). This is a breaking change.
If rhs > bit_size of rhs type (or bit_size -1 for signed types), then bit-shift will overflow.

Additional Context

Documentation*

Check one:

No documentation needed.
Documentation included in this PR.
[For Experimental Features] Documentation to be submitted in a separate PR.

PR Checklist*

I have tested the changes locally.
I have formatted the changes with Prettier and/or cargo fmt on default settings.

…ts bit-size

github-actions · 2025-08-01T09:27:20Z

Changes to Brillig bytecode sizes

Generated at commit: 8172f2128350a775bc9e738b4c8173ceaa6f4cc7, compared to commit: 4f8dbbc20bb40115d87c499154655ebe7e962099

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
bit_shifts_runtime_inliner_max	+59 ❌	+10.99%
bit_shifts_runtime_inliner_min	+59 ❌	+10.99%
bit_shifts_runtime_inliner_zero	+59 ❌	+10.99%
shift_right_overflow_inliner_max	-7 ✅	-24.14%
shift_right_overflow_inliner_min	-7 ✅	-24.14%
shift_right_overflow_inliner_zero	-7 ✅	-24.14%

Full diff report 👇

Program	Brillig opcodes (+/-)	%
bit_shifts_runtime_inliner_max	596 (+59)	+10.99%
bit_shifts_runtime_inliner_min	596 (+59)	+10.99%
bit_shifts_runtime_inliner_zero	596 (+59)	+10.99%
conditional_regression_underflow_inliner_max	68 (+6)	+9.68%
conditional_regression_underflow_inliner_min	68 (+6)	+9.68%
conditional_regression_underflow_inliner_zero	68 (+6)	+9.68%
bit_shifts_comptime_inliner_max	119 (+2)	+1.71%
bit_shifts_comptime_inliner_min	119 (+2)	+1.71%
bit_shifts_comptime_inliner_zero	119 (+2)	+1.71%
array_dynamic_blackbox_input_inliner_min	388 (-1)	-0.26%
array_dynamic_blackbox_input_inliner_max	367 (-1)	-0.27%
array_dynamic_blackbox_input_inliner_zero	367 (-1)	-0.27%
u128_type_inliner_max	138 (-1)	-0.72%
u128_type_inliner_min	138 (-1)	-0.72%
u128_type_inliner_zero	138 (-1)	-0.72%
binary_operator_overloading_inliner_max	380 (-3)	-0.78%
binary_operator_overloading_inliner_min	221 (-3)	-1.34%
binary_operator_overloading_inliner_zero	221 (-3)	-1.34%
u16_support_inliner_min	56 (-3)	-5.08%
u16_support_inliner_zero	56 (-3)	-5.08%
u16_support_inliner_max	48 (-3)	-5.88%
shift_right_overflow_inliner_max	22 (-7)	-24.14%
shift_right_overflow_inliner_min	22 (-7)	-24.14%
shift_right_overflow_inliner_zero	22 (-7)	-24.14%

github-actions · 2025-08-01T09:27:21Z

Changes to number of Brillig opcodes executed

Generated at commit: 8172f2128350a775bc9e738b4c8173ceaa6f4cc7, compared to commit: 4f8dbbc20bb40115d87c499154655ebe7e962099

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
bit_shifts_runtime_inliner_max	+677 ❌	+298.24%
bit_shifts_runtime_inliner_min	+677 ❌	+298.24%
bit_shifts_runtime_inliner_zero	+677 ❌	+298.24%
shift_right_overflow_inliner_max	-5 ✅	-20.00%
shift_right_overflow_inliner_min	-5 ✅	-20.00%
shift_right_overflow_inliner_zero	-5 ✅	-20.00%

Full diff report 👇

Program	Brillig opcodes (+/-)	%
bit_shifts_runtime_inliner_max	904 (+677)	+298.24%
bit_shifts_runtime_inliner_min	904 (+677)	+298.24%
bit_shifts_runtime_inliner_zero	904 (+677)	+298.24%
conditional_regression_underflow_inliner_max	36 (+4)	+12.50%
conditional_regression_underflow_inliner_min	36 (+4)	+12.50%
conditional_regression_underflow_inliner_zero	36 (+4)	+12.50%
array_dynamic_blackbox_input_inliner_min	6,557 (-1)	-0.02%
array_dynamic_blackbox_input_inliner_max	6,528 (-1)	-0.02%
array_dynamic_blackbox_input_inliner_zero	6,528 (-1)	-0.02%
bit_shifts_comptime_inliner_max	462 (-2)	-0.43%
bit_shifts_comptime_inliner_min	462 (-2)	-0.43%
bit_shifts_comptime_inliner_zero	462 (-2)	-0.43%
u128_type_inliner_max	92 (-1)	-1.08%
u128_type_inliner_min	92 (-1)	-1.08%
u128_type_inliner_zero	92 (-1)	-1.08%
binary_operator_overloading_inliner_min	225 (-3)	-1.32%
binary_operator_overloading_inliner_zero	225 (-3)	-1.32%
binary_operator_overloading_inliner_max	209 (-3)	-1.42%
u16_support_inliner_min	48 (-3)	-5.88%
u16_support_inliner_zero	48 (-3)	-5.88%
u16_support_inliner_max	36 (-3)	-7.69%
shift_right_overflow_inliner_max	20 (-5)	-20.00%
shift_right_overflow_inliner_min	20 (-5)	-20.00%
shift_right_overflow_inliner_zero	20 (-5)	-20.00%

github-actions · 2025-08-01T09:27:46Z

FYI @noir-lang/developerrelations on Noir doc changes.

github-actions

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'Test Suite Duration'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.20.

Benchmark suite	Current: `925d517`	Previous: `4f8dbbc`	Ratio
`test_report_zkpassport_noir-ecdsa_`	`2` s	`1` s	`2`

This comment was automatically generated by workflow using github-action-benchmark.

CC: @TomAFrench

github-actions · 2025-08-01T10:00:18Z

🚀 Deployed on https://689e0dac900f62db5aeffada--noir-docs.netlify.app

github-actions

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'Execution Time'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.20.

Benchmark suite	Current: `d2200ea`	Previous: `ee2ac1a`	Ratio
`sha512-100-bytes`	`0.142` s	`0.096` s	`1.48`

This comment was automatically generated by workflow using github-action-benchmark.

CC: @TomAFrench

jfecher · 2025-08-04T18:54:46Z

Looks like this is timing out all the debug tests locally

TomAFrench

~~I think we've got some undesirable behaviour here which means that 4u8>>2u8 is fine but 1u8>>2u8 is disallowed. I would expect the latter to result in zero.~~ I was misreading comments.

compiler/noirc_evaluator/src/ssa/ssa_gen/context.rs

compiler/noirc_evaluator/src/ssa/ir/dfg/simplify/binary.rs

test_programs/execution_success/shift_right_overflow/src/main.nr

TomAFrench · 2025-08-08T09:52:08Z

Program Brillig opcodes (+/-) %

bit_shifts_runtime_inliner_max +677 ❌ +298.24%

bit_shifts_runtime_inliner_min +677 ❌ +298.24%

bit_shifts_runtime_inliner_zero +677 ❌ +298.24%

Strange we're getting such a large regression here as while there's been a few changes to this program it shouldn't result in ~700 new opcodes afaict.

Edit: due to the removals above?

Edit Edit: We're now doing more runtime signed bitshifts so the overflow checks could be the cause of this.

compiler/noirc_evaluator/src/ssa/interpreter/errors.rs

compiler/noirc_evaluator/src/ssa/ir/instruction/binary.rs

compiler/noirc_evaluator/src/ssa/opt/remove_bit_shifts.rs

github-actions

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'Compilation Time'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.20.

Benchmark suite	Current: `d9ba69c`	Previous: `ee2ac1a`	Ratio
`semaphore-depth-10`	`0.958` s	`0.759` s	`1.26`

This comment was automatically generated by workflow using github-action-benchmark.

CC: @TomAFrench

jfecher

LGTM - looks like external repos are failing to build though

…imization (#9495)

Automated pull of nightly from the [noir](https://github.com/noir-lang/noir) programming language, a dependency of Aztec. BEGIN_COMMIT_OVERRIDE chore: restore noir libs in CI (noir-lang/noir#9499) feat!: new semantic for bit-shifts (noir-lang/noir#9373) fix(ssa): Replace side effects with defaults when disabled (noir-lang/noir#9462) fix(ssa): Replace pop from 0-length slice with constraint and defaults (noir-lang/noir#9489) fix: assert types are not mutated in constant folding (noir-lang/noir#9481) fix: remove shadowing in `BoundedVec::any` causing returning false unconditionally (noir-lang/noir#9478) END_COMMIT_OVERRIDE

guipublic and others added 2 commits August 1, 2025 11:10

rhs must have same type as lhs in bit-shifts, and overflows if over i…

6a1e9dc

…ts bit-size

Merge branch 'master' into gd/issue_9022

530947b

github-actions bot reviewed Aug 1, 2025

View reviewed changes

update docs

5c3b5dc

github-actions bot added the documentation Improvements or additions to documentation label Aug 1, 2025

guipublic added 3 commits August 1, 2025 13:18

fix ssa interpreter and a bunch of tests

e7c081a

small fixes for CI

e79aad5

update fuzzer

d6723e8

github-actions bot reviewed Aug 1, 2025

View reviewed changes

guipublic and others added 3 commits August 1, 2025 16:39

clippy

e2057f3

Merge branch 'master' into gd/issue_9022

23d1faf

Merge branch 'master' into gd/issue_9022

a53e7fa

guipublic mentioned this pull request Aug 4, 2025

chore: update ex in docs #9385

Merged

5 tasks

guipublic and others added 4 commits August 7, 2025 17:00

Merge branch 'master' into gd/issue_9022

d98bcb3

snapshots

e6a6f70

chore: update ast_fuzzer to use new semantics

a6393c8

Merge branch 'master' into gd/issue_9022

91f957b

TomAFrench reviewed Aug 8, 2025

View reviewed changes

compiler/noirc_evaluator/src/ssa/ssa_gen/context.rs Show resolved Hide resolved

compiler/noirc_evaluator/src/ssa/ir/dfg/simplify/binary.rs Show resolved Hide resolved

test_programs/execution_success/shift_right_overflow/src/main.nr Show resolved Hide resolved

TomAFrench added 2 commits August 8, 2025 11:18

Merge branch 'master' into gd/issue_9022

99612a8

.

2444c0a

TomAFrench reviewed Aug 8, 2025

View reviewed changes

compiler/noirc_evaluator/src/ssa/interpreter/errors.rs Outdated Show resolved Hide resolved

compiler/noirc_evaluator/src/ssa/ir/instruction/binary.rs Outdated Show resolved Hide resolved

compiler/noirc_evaluator/src/ssa/opt/remove_bit_shifts.rs Outdated Show resolved Hide resolved

chore: sanity check on brillig opcode

bd073ba

This was referenced Aug 8, 2025

fix: switch to new bit-shift semantic noir-lang/sha256#27

Merged

fix!: switch to new bit shifts semantic noir-lang/noir-bignum#196

Merged

TomAFrench added 3 commits August 12, 2025 21:03

.

5db938d

.

5dd5032

.

2cd47fd

TomAFrench changed the title ~~chore!: new semantic for bit-shifts~~ feat!: new semantic for bit-shifts Aug 13, 2025

.

4b41ad4

TomAFrench requested a review from a team August 13, 2025 14:37

TomAFrench added 2 commits August 13, 2025 14:38

chore: rename test

c003350

.

3fa6198

github-actions bot reviewed Aug 13, 2025

View reviewed changes

jfecher reviewed Aug 13, 2025

View reviewed changes

.

d9ba69c

TomAFrench approved these changes Aug 14, 2025

View reviewed changes

.

d2200ea

TomAFrench enabled auto-merge August 14, 2025 10:22

TomAFrench added 3 commits August 14, 2025 12:19

chore: add regression test

fa61e25

.

9ee5662

Merge branch 'master' into gd/issue_9022

0821205

vezenovm mentioned this pull request Aug 14, 2025

fix(ssa): Use reverse post order when traversing blocks in simple_optimization #9495

Merged

5 tasks

fix(ssa): Use reverse post order when traversing blocks in simple_opt…

925d517

…imization (#9495)

TomAFrench added this pull request to the merge queue Aug 14, 2025

Merged via the queue into master with commit f6fed8b Aug 14, 2025
101 checks passed

TomAFrench deleted the gd/issue_9022 branch August 14, 2025 16:52

noirwhal mentioned this pull request Aug 14, 2025

chore: Release Noir(1.0.0-beta.11) #9457

Merged

AztecBot mentioned this pull request Aug 14, 2025

chore: Bump Noir reference AztecProtocol/aztec-packages#16382

Merged

TomAFrench mentioned this pull request Aug 15, 2025

feat: simplify bitshift logic to remove leftover overflow handling #9506

Merged

5 tasks

Conversation

guipublic commented Aug 1, 2025

Description

Problem*

Summary*

Additional Context

Documentation*

PR Checklist*

Uh oh!

github-actions bot commented Aug 1, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changes to Brillig bytecode sizes

🧾 Summary (10% most significant diffs)

Uh oh!

github-actions bot commented Aug 1, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changes to number of Brillig opcodes executed

🧾 Summary (10% most significant diffs)

Uh oh!

github-actions bot commented Aug 1, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

⚠️ Performance Alert ⚠️

Uh oh!

github-actions bot commented Aug 1, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

⚠️ Performance Alert ⚠️

Uh oh!

jfecher commented Aug 4, 2025

Uh oh!

TomAFrench left a comment • 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!

Uh oh!

Uh oh!

Uh oh!

TomAFrench commented Aug 8, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

github-actions bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

⚠️ Performance Alert ⚠️

Uh oh!

jfecher left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

github-actions bot commented Aug 1, 2025 •

edited

Loading

github-actions bot commented Aug 1, 2025 •

edited

Loading

github-actions bot commented Aug 1, 2025 •

edited

Loading

github-actions bot left a comment •

edited

Loading

github-actions bot commented Aug 1, 2025 •

edited

Loading

github-actions bot left a comment •

edited

Loading

TomAFrench left a comment •

edited

Loading

TomAFrench commented Aug 8, 2025 •

edited

Loading

github-actions bot left a comment •

edited

Loading