runtime: implement vat pt1 by ibhatt-jumptrading · Pull Request #8740 · firedancer-io/firedancer

ibhatt-jumptrading · 2026-03-05T14:43:43Z

todo: ledger with vat enabled, pass vat set of validators across replay_epoch,

github-actions · 2026-03-05T14:49:04Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.089377 s`	`0.089175 s`	`-0.226%` ✅
backtest `mainnet-368528500-perf` snapshot load	`5.684 s`	`4.383 s`	`-22.889%` ✅
backtest `mainnet-368528500-perf` total elapsed	`89.376603 s`	`89.174998 s`	`-0.226%` ✅
`firedancer mem` usage with `mainnet.toml`	`963.38 GiB`	`963.38 GiB`	`0.000%` ✅

Copilot

Pull request overview

This PR starts integrating “VAT” support into the runtime by introducing a new CoW-managed fd_top_votes_t state tracked per bank and stored in a shared pool alongside existing CoW fields (e.g. epoch leaders, cost tracker).

Changes:

Add a fixed maximum footprint constant for fd_top_votes_t storage.
Extend fd_bank/fd_banks shared-memory layout with a new top_votes CoW pool, per-bank dirty/index tracking, and a dedicated pool lock.
Implement fd_bank_top_votes_query() / fd_bank_top_votes_modify() and wire pool initialization/join/release paths into banks lifecycle (new/join/advance_root/prune/clear).

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 6 comments.

File	Description
`src/flamenco/stakes/fd_top_votes.h`	Introduces `FD_TOP_VOTES_MAX_FOOTPRINT` used to size bank-owned top-votes storage.
`src/flamenco/runtime/fd_bank.h`	Adds `fd_bank_top_votes_t` pool type, bank/banks offsets + locks + query/modify API declarations.
`src/flamenco/runtime/fd_bank.c`	Allocates/joins top-votes pool in shared memory and implements bank-level query/modify + pool release logic.

Copilot · 2026-03-05T14:53:21Z

+  bank->top_votes_pool_offset = (ulong)top_votes_pool - (ulong)bank;
+}
+
+static inline fd_bank_top_votes_t *
+fd_bank_get_top_votes_pool( fd_bank_data_t * bank ) {
+  return fd_type_pun( (uchar *)bank + bank->top_votes_pool_offset );


fd_bank_set_top_votes_pool()/fd_bank_get_top_votes_pool() store and recover the pool pointer via fd_type_pun rather than using fd_bank_top_votes_pool_leave()/fd_bank_top_votes_pool_join(). This deviates from the existing epoch_leaders/cost_tracker pool pattern and makes the NULL checks in fd_bank_top_votes_{query,modify} ineffective (the getter cannot fail/validate magic). Consider mirroring the established pattern: store the "leave" pointer (shared-memory base) and have the getter return fd_bank_top_votes_pool_join(...), so misconfiguration/corruption is detected early and behavior is consistent across CoW pools.

Suggested change

bank->top_votes_pool_offset = (ulong)top_votes_pool - (ulong)bank;

}

static inline fd_bank_top_votes_t *

fd_bank_get_top_votes_pool( fd_bank_data_t * bank ) {

return fd_type_pun( (uchar *)bank + bank->top_votes_pool_offset );

void * top_votes_pool_mem = fd_bank_top_votes_pool_leave( top_votes_pool );

if( FD_UNLIKELY( !top_votes_pool_mem ) ) {

FD_LOG_CRIT(( "Failed to leave top votes pool" ));

}

bank->top_votes_pool_offset = (ulong)top_votes_pool_mem - (ulong)bank;

}

static inline fd_bank_top_votes_t *

fd_bank_get_top_votes_pool( fd_bank_data_t * bank ) {

return fd_bank_top_votes_pool_join( (uchar *)bank + bank->top_votes_pool_offset );

Copilot · 2026-03-05T14:53:21Z

 #define FD_TOP_VOTES_ALIGN (128UL)

+#define FD_TOP_VOTES_MAX_FOOTPRINT (1000000UL)
+


FD_TOP_VOTES_MAX_FOOTPRINT is set to a hard-coded 1,000,000 bytes and is used to size every fd_bank_top_votes pool element. Given fd_top_votes is always constructed with vote_accounts_max=2000 in this PR, this constant likely over-allocates shared memory per fork by a large margin. Consider defining an explicit FD_TOP_VOTES_VOTE_ACCOUNTS_MAX and sizing the footprint constant to match fd_top_votes_footprint(FD_TOP_VOTES_VOTE_ACCOUNTS_MAX) (and/or adding a runtime assert that the computed footprint fits) to avoid unnecessary memory/IPC pressure as max_fork_width grows.

Suggested change

#define FD_TOP_VOTES_ALIGN (128UL)

#define FD_TOP_VOTES_MAX_FOOTPRINT (1000000UL)

#define FD_TOP_VOTES_ALIGN (128UL)

#define FD_TOP_VOTES_VOTE_ACCOUNTS_MAX (2000UL)

#define FD_TOP_VOTES_MAX_FOOTPRINT \

fd_top_votes_footprint( FD_TOP_VOTES_VOTE_ACCOUNTS_MAX )

Copilot · 2026-03-05T14:53:21Z

+  for( ulong i=0UL; i<max_fork_width; i++ ) {
+    fd_bank_top_votes_t * top_votes_ele = fd_bank_top_votes_pool_ele( top_votes_pool, i );
+    fd_top_votes_t * top_votes = fd_top_votes_join( fd_top_votes_new( top_votes_ele->data, 2000UL, seed ) ); /* TODO:FIXME: Magic number*/
+    if( FD_UNLIKELY( !top_votes ) ) {
+      FD_LOG_WARNING(( "Failed to create top votes" ));
+      return NULL;


In fd_banks_new(), fd_top_votes_new() is called with a hard-coded 2000UL ("TODO:FIXME: Magic number"). This should be a named constant shared with the VAT design (e.g. FD_TOP_VOTES_VOTE_ACCOUNTS_MAX) so the value is not duplicated and future changes don’t risk inconsistent sizing/logic (especially since the backing buffer size is also fixed via FD_TOP_VOTES_MAX_FOOTPRINT).

Copilot · 2026-03-05T14:53:21Z

+  fd_bank_top_votes_t * top_votes_pool = fd_bank_top_votes_pool_join( fd_bank_top_votes_pool_new( top_votes_pool_mem, max_fork_width ) );
+  if( FD_UNLIKELY( !top_votes_pool ) ) {
+    FD_LOG_WARNING(( "Failed to create top votes pool" ));
+    return NULL;
+  }
+  fd_banks_set_top_votes_pool( banks_data, top_votes_pool );
+  for( ulong i=0UL; i<max_fork_width; i++ ) {
+    fd_bank_top_votes_t * top_votes_ele = fd_bank_top_votes_pool_ele( top_votes_pool, i );
+    fd_top_votes_t * top_votes = fd_top_votes_join( fd_top_votes_new( top_votes_ele->data, 2000UL, seed ) ); /* TODO:FIXME: Magic number*/
+    if( FD_UNLIKELY( !top_votes ) ) {
+      FD_LOG_WARNING(( "Failed to create top votes" ));
+      return NULL;
+    }
+    fd_top_votes_init( top_votes );
+  }
+


The new top-votes CoW pool integration adds non-trivial behavior (pool init in fd_banks_new, CoW modify/query, and pool-element release during advance_root/prune/clear) but there are no corresponding bank-level tests. There are existing unit tests in runtime/test_bank.c that exercise epoch_leaders CoW/pool behavior; adding analogous coverage for top_votes would help catch regressions (e.g., cloning vs reset, correct release on root advance).

Copilot · 2026-03-05T14:53:22Z

+static inline fd_bank_top_votes_t *
+fd_banks_get_top_votes_pool( fd_banks_data_t * banks_data ) {
+  return fd_type_pun( (uchar *)banks_data + banks_data->top_votes_pool_offset );
+}
+
+static inline void
+fd_banks_set_top_votes_pool( fd_banks_data_t *     banks_data,
+                             fd_bank_top_votes_t * top_votes_pool ) {
+  banks_data->top_votes_pool_offset = (ulong)top_votes_pool - (ulong)banks_data;
+}


fd_banks_set_top_votes_pool()/fd_banks_get_top_votes_pool() also use raw pointer arithmetic + fd_type_pun instead of the fd_bank_top_votes_pool_{leave,join} APIs. This makes joins less robust (no magic/alignment validation) and is inconsistent with fd_banks_set_epoch_leaders_pool()/get_epoch_leaders_pool. Consider storing the pool "leave" pointer and re-joining via fd_bank_top_votes_pool_join() for consistency and better corruption detection.

github-actions · 2026-03-05T15:19:34Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.07477 s`	`0.074981 s`	`0.282%` ✅
backtest `mainnet-368528500-perf` snapshot load	`4.268 s`	`3.102 s`	`-27.320%` ✅
backtest `mainnet-368528500-perf` total elapsed	`74.77018 s`	`74.980905 s`	`0.282%` ✅
`firedancer mem` usage with `mainnet.toml`	`963.38 GiB`	`963.38 GiB`	`0.000%` ✅

github-actions · 2026-03-05T15:51:50Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.09022 s`	`0.0885 s`	`-1.906%` ✅
backtest `mainnet-368528500-perf` snapshot load	`5.775 s`	`4.359 s`	`-24.519%` ✅
backtest `mainnet-368528500-perf` total elapsed	`90.219846 s`	`88.500164 s`	`-1.906%` ✅
`firedancer mem` usage with `mainnet.toml`	`963.38 GiB`	`963.38 GiB`	`0.000%` ✅

Copilot

Pull request overview

Copilot reviewed 10 out of 10 changed files in this pull request and generated 7 comments.

Copilot · 2026-03-05T15:53:57Z

+  FD_TEST( fd_top_votes_query( top_votes, vote_pubkey, &node_out, &stake_out, &slot_out, &timestamp_out ) );
  FD_TEST( !memcmp( &node_out, expected_node, sizeof(fd_pubkey_t) ) );
  FD_TEST( stake_out==expected_stake );
+  FD_TEST( slot_out==expected_stake );
+  FD_TEST( timestamp_out==(long)expected_stake );


assert_vote_present compares slot_out and timestamp_out against expected_stake, which means the test will still pass even if last_vote_slot/last_vote_timestamp are wired incorrectly. Consider extending assert_vote_present to take expected_slot and expected_timestamp (and use different values in inserts) so the metadata fields are actually validated.

Copilot · 2026-03-05T15:53:58Z

+
+    ulong last_vote_slot;
+    long  last_vote_timestamp;
+    int   found = fd_top_votes_query( top_votes, &pubkey, NULL, NULL, &last_vote_slot, &last_vote_timestamp );


fd_sysvar_clock.c calls fd_top_votes_query(top_votes, ...) unconditionally, but fd_bank_top_votes_query() can return NULL when the bank hasn't allocated a top-votes pool element yet. This will dereference a NULL pointer inside fd_top_votes_query; add a NULL check and fall back to the accdb path when top_votes is NULL.

Suggested change

int found = fd_top_votes_query( top_votes, &pubkey, NULL, NULL, &last_vote_slot, &last_vote_timestamp );

int found = 0;

if( FD_LIKELY( top_votes ) )

found = fd_top_votes_query( top_votes, &pubkey, NULL, NULL, &last_vote_slot, &last_vote_timestamp );

Copilot · 2026-03-05T15:53:59Z

+/* fd_top_votes_update inserts a new vote account into the top votes set
+   given a vote account, node account, last vote slot, last vote
+   timestamp, and a stake.  The node account, last vote slot, and last
+   vote timestamp are just metadata for the structure.  If the vote
+   account isn't in the top max_vote_accounts in terms of stake, it is
+   ignored and is treated as a no-op.  If the vote account ties the
+   minimum stake and the struct is full, all elements with that stake
+   are removed.  */


The doc comment for fd_top_votes_insert still says "fd_top_votes_update inserts ...". This is confusing now that fd_top_votes_update exists again with different semantics; update the comment to refer to fd_top_votes_insert (and consider tightening wording around the parameters order: stake vs last_vote_slot/timestamp).

Copilot · 2026-03-05T15:53:59Z

+void
+fd_top_votes_update( fd_top_votes_t *    top_votes,
+                     fd_pubkey_t const * pubkey,
+                     ulong               last_vote_slot,
+                     long                last_vote_timestamp ) {
  vote_ele_t * pool = get_pool( top_votes );
  map_t *      map  = get_map( top_votes );

+  ushort idx = (ushort)map_idx_query_const( map, pubkey, USHORT_MAX, pool );
+  if( FD_UNLIKELY( idx==USHORT_MAX ) ) return;
+  vote_ele_t * ele = pool_ele( pool, idx );
+  ele->last_vote_slot      = last_vote_slot;
+  ele->last_vote_timestamp = last_vote_timestamp;
+}


fd_top_votes_update is new behavior (updates slot/timestamp for an existing entry) but there are no unit tests exercising it or verifying that fd_top_votes_query returns the updated metadata. Add a test case that inserts an entry, calls fd_top_votes_update with different values, and then queries the metadata to ensure it changes as expected.

…to ibhatt/vat

github-actions · 2026-03-05T19:38:16Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.061158 s`	`0.060466 s`	`-1.131%` ✅
backtest `mainnet-368528500-perf` snapshot load	`3.226 s`	`2.766 s`	`-14.259%` ✅
backtest `mainnet-368528500-perf` total elapsed	`61.158493 s`	`60.466187 s`	`-1.132%` ✅
`firedancer mem` usage with `mainnet.toml`	`963.38 GiB`	`963.38 GiB`	`0.000%` ✅

Copilot

Pull request overview

Copilot reviewed 11 out of 11 changed files in this pull request and generated 6 comments.

github-actions · 2026-03-05T20:02:18Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.072235 s`	`0.071998 s`	`-0.328%` ✅
backtest `mainnet-368528500-perf` snapshot load	`3.204 s`	`2.787 s`	`-13.015%` ✅
backtest `mainnet-368528500-perf` total elapsed	`72.234639 s`	`71.997765 s`	`-0.328%` ✅
`firedancer mem` usage with `mainnet.toml`	`963.38 GiB`	`963.38 GiB`	`0.000%` ✅

github-actions · 2026-03-05T20:27:46Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.083966 s`	`0.082917 s`	`-1.249%` ✅
backtest `mainnet-368528500-perf` snapshot load	`3.855 s`	`3.324 s`	`-13.774%` ✅
backtest `mainnet-368528500-perf` total elapsed	`83.966161 s`	`82.9167 s`	`-1.250%` ✅
`firedancer mem` usage with `mainnet.toml`	`963.38 GiB`	`963.38 GiB`	`0.000%` ✅

Copilot

Pull request overview

Copilot reviewed 14 out of 14 changed files in this pull request and generated 8 comments.

github-actions · 2026-03-05T23:09:24Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.061173 s`	`0.060897 s`	`-0.451%` ✅
backtest `mainnet-368528500-perf` snapshot load	`3.155 s`	`2.39 s`	`-24.247%` ✅
backtest `mainnet-368528500-perf` total elapsed	`61.172539 s`	`60.897139 s`	`-0.450%` ✅
`firedancer mem` usage with `mainnet.toml`	`963.38 GiB`	`963.38 GiB`	`0.000%` ✅

github-actions · 2026-03-09T19:25:42Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.089631 s`	`0.088689 s`	`-1.051%` ✅
backtest `mainnet-368528500-perf` snapshot load	`5.788 s`	`4.42 s`	`-23.635%` ✅
backtest `mainnet-368528500-perf` total elapsed	`89.630558 s`	`88.688961 s`	`-1.051%` ✅
`firedancer mem` usage with `mainnet.toml`	`963.38 GiB`	`963.38 GiB`	`0.000%` ✅

github-actions · 2026-03-10T14:30:35Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.061152 s`	`0.060947 s`	`-0.335%` ✅
backtest `mainnet-368528500-perf` snapshot load	`3.216 s`	`2.422 s`	`-24.689%` ✅
backtest `mainnet-368528500-perf` total elapsed	`61.152432 s`	`60.94654 s`	`-0.337%` ✅
`firedancer mem` usage with `mainnet.toml`	`963.38 GiB`	`963.38 GiB`	`0.000%` ✅

Copilot

Pull request overview

Copilot reviewed 19 out of 19 changed files in this pull request and generated 3 comments.

github-actions · 2026-03-10T15:11:26Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.06107 s`	`0.060889 s`	`-0.296%` ✅
backtest `mainnet-368528500-perf` snapshot load	`3.101 s`	`2.384 s`	`-23.122%` ✅
backtest `mainnet-368528500-perf` total elapsed	`61.069675 s`	`60.889374 s`	`-0.295%` ✅
`firedancer mem` usage with `mainnet.toml`	`963.4 GiB`	`963.4 GiB`	`0.000%` ✅

github-actions · 2026-03-10T15:53:38Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.060897 s`	`0.060569 s`	`-0.539%` ✅
backtest `mainnet-368528500-perf` snapshot load	`3.122 s`	`2.376 s`	`-23.895%` ✅
backtest `mainnet-368528500-perf` total elapsed	`60.897493 s`	`60.569148 s`	`-0.539%` ✅
`firedancer mem` usage with `mainnet.toml`	`963.4 GiB`	`963.4 GiB`	`0.000%` ✅

Copilot

Pull request overview

Copilot reviewed 19 out of 19 changed files in this pull request and generated 5 comments.

Copilot · 2026-03-10T15:54:53Z

  FD_SCRATCH_ALLOC_FINI( l, fd_top_votes_align() );

+  if( FD_UNLIKELY( FD_SCRATCH_ALLOC_FINI( l, fd_top_votes_align() ) != (ulong)top_votes + fd_top_votes_footprint( vote_accounts_max ) ) ) {


fd_top_votes_new calls FD_SCRATCH_ALLOC_FINI( l, fd_top_votes_align() ) twice in a row. The macro mutates l, so calling it twice is redundant/confusing and makes it easier to accidentally introduce layout bugs later. Consider capturing the return value from a single FD_SCRATCH_ALLOC_FINI call and using that for the layout check (or remove the first call).

Suggested change

FD_SCRATCH_ALLOC_FINI( l, fd_top_votes_align() );

if( FD_UNLIKELY( FD_SCRATCH_ALLOC_FINI( l, fd_top_votes_align() ) != (ulong)top_votes + fd_top_votes_footprint( vote_accounts_max ) ) ) {

ulong l_end = FD_SCRATCH_ALLOC_FINI( l, fd_top_votes_align() );

if( FD_UNLIKELY( l_end != (ulong)top_votes + fd_top_votes_footprint( vote_accounts_max ) ) ) {

github-actions · 2026-03-10T16:04:09Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.083843 s`	`0.083092 s`	`-0.896%` ✅
backtest `mainnet-368528500-perf` snapshot load	`3.9 s`	`3.432 s`	`-12.000%` ✅
backtest `mainnet-368528500-perf` total elapsed	`83.843314 s`	`83.091833 s`	`-0.896%` ✅
`firedancer mem` usage with `mainnet.toml`	`963.4 GiB`	`963.4 GiB`	`0.000%` ✅

github-actions · 2026-03-10T19:32:44Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.089903 s`	`0.088946 s`	`-1.064%` ✅
backtest `mainnet-368528500-perf` snapshot load	`5.814 s`	`4.42 s`	`-23.977%` ✅
backtest `mainnet-368528500-perf` total elapsed	`89.903214 s`	`88.945981 s`	`-1.065%` ✅
`firedancer mem` usage with `mainnet.toml`	`994.41 GiB`	`994.41 GiB`	`0.000%` ✅

Copilot

Pull request overview

Copilot reviewed 20 out of 20 changed files in this pull request and generated 4 comments.

github-actions · 2026-03-10T19:52:15Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.061288 s`	`0.060802 s`	`-0.793%` ✅
backtest `mainnet-368528500-perf` snapshot load	`3.186 s`	`2.383 s`	`-25.204%` ✅
backtest `mainnet-368528500-perf` total elapsed	`61.28805 s`	`60.801991 s`	`-0.793%` ✅
`firedancer mem` usage with `mainnet.toml`	`994.41 GiB`	`994.41 GiB`	`0.000%` ✅

github-actions · 2026-03-10T20:16:22Z

Performance Measurements ⏳

Suite	Baseline	New	Change
backtest `mainnet-368528500-perf` per slot	`0.061286 s`	`0.060904 s`	`-0.623%` ✅
backtest `mainnet-368528500-perf` snapshot load	`3.246 s`	`2.479 s`	`-23.629%` ✅
backtest `mainnet-368528500-perf` total elapsed	`61.286082 s`	`60.904384 s`	`-0.623%` ✅
`firedancer mem` usage with `mainnet.toml`	`994.41 GiB`	`994.41 GiB`	`0.000%` ✅

Copilot

Pull request overview

Copilot reviewed 21 out of 21 changed files in this pull request and generated 4 comments.

Copilot · 2026-03-10T20:19:33Z

+    fd_top_votes_t * top_votes = fd_bank_top_votes_modify( bank );
    for( ushort i=0; i<txn_out->accounts.cnt; i++ ) {
      /* We are only interested in saving writable accounts and the fee


fd_runtime_commit_txn unconditionally calls fd_bank_top_votes_modify( bank ) before scanning accounts. This can trigger an unnecessary CoW allocation + ~194KB memcpy for transactions that don't touch any vote accounts. Consider lazily calling fd_bank_top_votes_modify only when a vote-program-owned account is encountered (or when the VAT feature is active and the cache is actually needed).

wip

4efcb7e

Copilot AI review requested due to automatic review settings March 5, 2026 14:43

Copilot started reviewing on behalf of ibhatt-jumptrading March 5, 2026 14:44 View session

Copilot AI reviewed Mar 5, 2026

View reviewed changes

wip

23e2a59

wiip

965676b

Copilot AI review requested due to automatic review settings March 5, 2026 15:46

Copilot started reviewing on behalf of ibhatt-jumptrading March 5, 2026 15:47 View session

Copilot AI reviewed Mar 5, 2026

View reviewed changes

ibhatt-jumptrading added 2 commits March 5, 2026 18:50

Merge branch 'main' of https://github.com/firedancer-io/firedancer in…

596bc59

…to ibhatt/vat

wip

4b97747

removal logic

0ea5f16

Copilot AI review requested due to automatic review settings March 5, 2026 19:55

wip

48b2b95

Copilot AI reviewed Mar 5, 2026

View reviewed changes

wip

f294c38

wip

8ae08f4

Copilot AI review requested due to automatic review settings March 5, 2026 22:57

Copilot AI reviewed Mar 5, 2026

View reviewed changes

Copilot started reviewing on behalf of ibhatt-jumptrading March 5, 2026 23:18 View session

wip

472ec31

wip

2c39257

Copilot started reviewing on behalf of ibhatt-jumptrading March 10, 2026 14:27 View session

Copilot AI reviewed Mar 10, 2026

View reviewed changes

Comment thread src/flamenco/runtime/fd_bank.c

Comment thread src/flamenco/runtime/sysvar/fd_sysvar_clock.c

Comment thread src/flamenco/runtime/fd_runtime.c

cleanup

646da59

wip

43662a5

Copilot AI review requested due to automatic review settings March 10, 2026 15:49

Copilot started reviewing on behalf of ibhatt-jumptrading March 10, 2026 15:50 View session

ibhatt-jumptrading mentioned this pull request Mar 10, 2026

vote-stakes: fix idx assignment #8791

Closed

Copilot AI reviewed Mar 10, 2026

View reviewed changes

wip

a598396

ibhatt-jumptrading changed the title ~~runtime: implement vat [wip]~~ runtime: implement vat Mar 10, 2026

wip

718682a

Copilot AI review requested due to automatic review settings March 10, 2026 19:27

Copilot started reviewing on behalf of ibhatt-jumptrading March 10, 2026 19:28 View session

Copilot AI reviewed Mar 10, 2026

View reviewed changes

Comment thread src/flamenco/stakes/fd_stakes.c

Comment thread src/flamenco/stakes/fd_stakes.c Outdated

Comment thread src/discof/replay/fd_replay_tile.c

Comment thread src/flamenco/runtime/fd_bank.c

wip

02fabe7

wip

786b577

Copilot AI review requested due to automatic review settings March 10, 2026 20:12

Copilot started reviewing on behalf of ibhatt-jumptrading March 10, 2026 20:13 View session

Copilot AI reviewed Mar 10, 2026

View reviewed changes

ibhatt-jumptrading changed the title ~~runtime: implement vat~~ runtime: implement vat pt1 Mar 10, 2026

mjain-jump approved these changes Mar 10, 2026

View reviewed changes

ibhatt-jumptrading merged commit 0d394a6 into main Mar 10, 2026
21 checks passed

ibhatt-jumptrading deleted the ibhatt/vat branch March 10, 2026 21:21

		#define FD_TOP_VOTES_ALIGN (128UL)

		#define FD_TOP_VOTES_MAX_FOOTPRINT (1000000UL)

		FD_SCRATCH_ALLOC_FINI( l, fd_top_votes_align() );

		if( FD_UNLIKELY( FD_SCRATCH_ALLOC_FINI( l, fd_top_votes_align() ) != (ulong)top_votes + fd_top_votes_footprint( vote_accounts_max ) ) ) {

Conversation

ibhatt-jumptrading commented Mar 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented Mar 5, 2026

Performance Measurements ⏳

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

github-actions Bot commented Mar 5, 2026

Performance Measurements ⏳

Uh oh!

github-actions Bot commented Mar 5, 2026

Performance Measurements ⏳

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

github-actions Bot commented Mar 5, 2026

Performance Measurements ⏳

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

github-actions Bot commented Mar 5, 2026

Performance Measurements ⏳

Uh oh!

github-actions Bot commented Mar 5, 2026

Performance Measurements ⏳

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

ibhatt-jumptrading commented Mar 5, 2026 •

edited

Loading