rotor: report fec masks & sample rtt off shred tile arrival time by emwang-jump · Pull Request #12414 · firedancer-io/firedancer

emwang-jump · 2026-10-19T17:15:35Z

No description provided.

Copilot

🟢 Approval recommended

Static review found no blocking issues in mask propagation, initialization, or preservation; tests were not executed.

No open issues 🎉

What changed in this PR

Adds received-shred masks to Firedancer’s rotor telemetry, carrying them from FEC completion messages into block-received events.

Changes:

Propagates data, parity, and repair-data masks from the shred tile.
Preserves masks from the first FEC completion.
Extends telemetry tests to check mask storage and duplicate-completion handling.

File	Description
src/discof/rotor/fd_rotor_tile.c	Reports masks in block-received events.
src/disco/shred/fd_shred_tile.h	Adds masks to completion messages.
src/disco/shred/fd_shred_tile.c	Populates completion-message masks.
src/choreo/rotor/test_fd_rotor.c	Tests mask storage and preservation.
src/choreo/rotor/fd_rotor.h	Extends FEC state and completion API.
src/choreo/rotor/fd_rotor.c	Initializes masks and stores first-completion values.

_{Review effort: Balanced}

github-actions · 2026-10-19T18:58:10Z

 ┌─ ⚡ PERF · fc21c34 vs main@21c05bf ─────────────────────────────────
 │ SUITE                               BASELINE          NEW         Δ
 │ replay tps, mainnet               30,148 tps   30,030 tps  ·  -0.39%
 │ bench tps, localnet              665,585 tps  670,181 tps  ·  +0.69%
 │ snapshot load, testnet                3.77 s       3.77 s  ·  +0.06%
 │ mem total, mainnet                146.36 GiB   146.36 GiB  ·   0.00%
!│ mem total, testnet                 87.25 GiB    87.25 GiB  ▲   0.00%
!│ mem total, ag mainnet             147.17 GiB   147.17 GiB  ▲   0.00%
 │ clean compile, firedancer             3.44 s       3.44 s  ·  -0.15%
 │ binary size, firedancer             59.89 MB     59.89 MB  ·   0.00%
 ├─────────────────────────────────────────────────────────────────────
@@ 0 REGRESSIONS · 2 WARNINGS · 0 IMPROVED · 6 NOISE @@
 └─────────────────────────────────────────────────────────────────────

history · 10 pushes

 ┌─ HISTORY · Δ vs main, per push, newest first ──────────────────────────────────
 │ HEAD         TPS    BENCH     SNAP    MEM·M    MEM·T     AG·M  COMPILE   BINARY
!│ fc21c34   -0.39%   +0.69%   +0.06%    0.00%    0.00%    0.00%   -0.15%    0.00%
!│ df46180   +0.07%   -0.43%   -0.01%    0.00%    0.00%    0.00%   -0.10%    0.00%
-│ 89a2490        …        …        …        …        …        …        …        …
!│ 194e834   -0.66%   +0.55%   -0.53%    0.00%   +0.05%   +0.03%   -0.74%    0.00%
!│ 7b5541f   +0.47%   +0.18%   +0.22%    0.00%   +0.05%   +0.03%   -0.33%    0.00%
!│ 3064581   -0.08%   +0.47%   +0.01%    0.00%   +0.05%   +0.03%   +0.77%    0.00%
!│ 0f95aa7   +0.72%   +0.43%   -0.23%    0.00%   +0.05%   +0.03%   -0.09%    0.00%
 │ 6f6573d        …        …        …        …        …        …        …        …
!│ eeb7edb   +0.06%   -0.05%        …    0.00%   +0.05%   +0.03%   +0.57%    0.00%
!│ d45f46b   -0.59%   -0.41%   +1.66%    0.00%   +0.04%   +0.03%   +0.14%    0.00%
 └────────────────────────────────────────────────────────────────────────────────

Copilot

🟢 Approval recommended

Static review found no blocking issues in mask propagation or preservation; runtime tests were not run.

No open issues 🎉

_{Review effort: Balanced}

Copilot

🟡 Changes recommended

Partially answered requests inflate the first-response in-flight gauge, and generated metric documentation remains outdated.

2 open issues

_{Review effort: Balanced}

+    <int value="0" name="ForBlockId" label="Response to a pending WindowIndexForBlockId request" />
+    <int value="1" name="SlotIndex" label="Response to a pending WindowIndex or HighestWindowIndex request" />
+    <int value="2" name="Unmatched" label="Matched no pending request: a duplicate, or an answer to an expired or evicted request" />


Copilot

🔵 Needs a closer look

The in-flight request gauge still counts partially answered requests despite being defined as attempts awaiting their first response.

2 open issues

_{Review effort: Balanced}

Copilot

🟡 Changes recommended

The stale generated event protobuf will cause the generated-code CI check to fail.

3 open issues

_{Review effort: Balanced}

Copilot

🟡 Changes recommended

Partially answered by-ID requests incorrectly increment the unanswered-expiration counter.

4 open issues

_{Review effort: Balanced}

Copilot

🔵 Needs a closer look

Partially answered by-ID requests still increment the counter defined for unanswered expirations.

4 open issues

_{Review effort: Balanced}

Copilot

🔵 Needs a closer look

Timing-sensitive request-lifetime and peer-scoring changes warrant runtime validation and human review.

1 open issue

Regenerate metrics documentation for per-shred accounting

3 resolved since last review

_{Review effort: Balanced}

lidatong · 2026-10-21T18:26:50Z

+  uint      data_rcvd;     /* shred tile masks at first completion: bit j set if the data shred with in-type index j was received (vs recovered) */
+  uint      parity_rcvd;   /* ... if the parity shred with in-type index j was received */
+  uint      repair_rcvd;   /* ... if the data shred with in-type index j arrived by repair */
  ulong     next;
  ulong     prev;
 };
 typedef struct fd_rotor_fec fd_rotor_fec_t;
-FD_STATIC_ASSERT( sizeof(fd_rotor_fec_t)==120UL, fd_rotor_fec );


i think if you just jiggle these fields around, you'll find it fits in exactly 128, it probably is just interleaving padding

reordered it so theres no interleaved padding, but its 136 because w/out padding the size of the struct is 132, could cut it down other ways

Copilot

🔵 Needs a closer look

Stateful repair-response lifecycle changes need targeted runtime validation that was unavailable during review.

1 open issue

Regenerate metrics documentation for per-shred accounting

_{Review effort: Balanced}

Copilot

🔵 Needs a closer look

Repair-request lifecycle and latency-driven peer selection changes need targeted runtime validation and human review.

1 open issue

Regenerate metrics documentation for per-shred accounting

_{Review effort: Balanced}

Copilot

🔵 Needs a closer look

Repair-request lifecycle and peer-scoring changes need targeted runtime validation and human review.

1 open issue

Regenerate metrics documentation for per-shred accounting

_{Review effort: Balanced}

Copilot

🔵 Needs a closer look

Cross-tile timing and repair-request lifecycle changes warrant human review and targeted runtime validation despite no confirmed defects.

1 open issue

Regenerate metrics documentation for per-shred accounting

_{Review effort: Balanced}

emwang-jump marked this pull request as ready for review October 19, 2026 17:15

Copilot AI balanced review requested due to automatic review settings October 19, 2026 17:15

Copilot started reviewing on behalf of emwang-jump October 19, 2026 17:16

View session

Copilot AI reviewed Oct 19, 2026

View reviewed changes

emwang-jump force-pushed the emwang/rotor-fec-mas branch from 3dd185c to d45f46b Compare October 19, 2026 18:57

Copilot AI balanced review requested due to automatic review settings October 19, 2026 18:57

Copilot started reviewing on behalf of emwang-jump October 19, 2026 18:58

View session

Copilot AI reviewed Oct 19, 2026

View reviewed changes

Copilot AI balanced review requested due to automatic review settings October 19, 2026 23:48

Copilot started reviewing on behalf of emwang-jump October 19, 2026 23:49

View session

Copilot AI reviewed Oct 19, 2026

View reviewed changes

Copilot AI balanced review requested due to automatic review settings October 20, 2026 17:49

Copilot started reviewing on behalf of emwang-jump October 20, 2026 17:50

View session

Copilot AI reviewed Oct 20, 2026

View reviewed changes

Copilot AI balanced review requested due to automatic review settings October 20, 2026 18:27

Copilot started reviewing on behalf of emwang-jump October 20, 2026 18:28

View session

Copilot AI reviewed Oct 20, 2026

View reviewed changes

Comment thread src/disco/events/schema/metrics_rotor.json

emwang-jump changed the title ~~rotor: report fec masks~~ rotor: report fec masks & sample rtt off shred tile arrival time Oct 20, 2026

Copilot AI balanced review requested due to automatic review settings October 20, 2026 21:07

emwang-jump force-pushed the emwang/rotor-fec-mas branch from 2869902 to eeb7edb Compare October 20, 2026 21:07

Copilot started reviewing on behalf of emwang-jump October 20, 2026 21:07

View session

emwang-jump force-pushed the emwang/rotor-fec-mas branch from eeb7edb to 6f6573d Compare October 20, 2026 21:09

Copilot AI reviewed Oct 20, 2026

View reviewed changes

Comment thread src/discof/rotor/fd_rotor_tile.c Outdated

Copilot AI balanced review requested due to automatic review settings October 20, 2026 21:10

Copilot started reviewing on behalf of emwang-jump October 20, 2026 21:11

View session

Copilot AI reviewed Oct 20, 2026

View reviewed changes

Copilot AI balanced review requested due to automatic review settings October 20, 2026 22:52

Copilot started reviewing on behalf of emwang-jump October 20, 2026 22:52

View session

Copilot AI reviewed Oct 20, 2026

View reviewed changes

emwang-jump force-pushed the emwang/rotor-fec-mas branch from 3064581 to 7b5541f Compare October 20, 2026 23:25

lidatong reviewed Oct 21, 2026

View reviewed changes

Copilot AI balanced review requested due to automatic review settings October 21, 2026 18:39

emwang-jump force-pushed the emwang/rotor-fec-mas branch from 7b5541f to 194e834 Compare October 21, 2026 18:39

Copilot started reviewing on behalf of emwang-jump October 21, 2026 18:40

View session

emwang-jump enabled auto-merge (rebase) October 21, 2026 18:41

Copilot AI reviewed Oct 21, 2026

View reviewed changes

Copilot AI balanced review requested due to automatic review settings October 21, 2026 18:50

emwang-jump force-pushed the emwang/rotor-fec-mas branch from 194e834 to 89a2490 Compare October 21, 2026 18:50

Copilot started reviewing on behalf of emwang-jump October 21, 2026 18:51

View session

Copilot AI reviewed Oct 21, 2026

View reviewed changes

Copilot AI balanced review requested due to automatic review settings October 21, 2026 18:57

Copilot started reviewing on behalf of emwang-jump October 21, 2026 18:58

View session

rotor: report fec masks & sample rtt off shred tile arrival time

fc21c34

emwang-jump force-pushed the emwang/rotor-fec-mas branch from df46180 to fc21c34 Compare October 21, 2026 18:58

Copilot AI reviewed Oct 21, 2026

View reviewed changes

Copilot AI balanced review requested due to automatic review settings October 21, 2026 19:00

Copilot started reviewing on behalf of emwang-jump October 21, 2026 19:01

View session

lidatong approved these changes Oct 21, 2026

View reviewed changes

Copilot AI reviewed Oct 21, 2026

View reviewed changes

emwang-jump merged commit ba0e99f into main Oct 21, 2026
21 checks passed

emwang-jump deleted the emwang/rotor-fec-mas branch October 21, 2026 19:04

Conversation

emwang-jump commented Oct 19, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🟢 Approval recommended

Uh oh!

github-actions Bot commented Oct 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🟢 Approval recommended

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🟡 Changes recommended

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🔵 Needs a closer look

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🟡 Changes recommended

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🟡 Changes recommended

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🔵 Needs a closer look

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🔵 Needs a closer look

Uh oh!

lidatong Oct 21, 2026

Choose a reason for hiding this comment

Uh oh!

emwang-jump Oct 21, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🔵 Needs a closer look

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🔵 Needs a closer look

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🔵 Needs a closer look

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🔵 Needs a closer look

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

github-actions Bot commented Oct 19, 2026 •

edited

Loading