PR #28317 previously attempted to fix a case where blockstore processing
would create children banks for slots past the halt_at_slot.
However, the previous fix didn't handle the case where a slot could be
strictly less than the halt_at_slot, but have children that were greater
than the halt_at_slot. For example, this could happen if a child of slot
S is S+n where n > 1.
Thus, this change covers our processing logic to cover this second case
as well.
* add Eq, PartialEq to EpochInfo
* add Eq, PartialEq related to transactions and messages
* add Eq, PartialEq to Ed25519SignatureOffsets
* add Eq, PartialEq to FeatureSet
* add Eq, PartialEq to fee.rs
* add Eq, PartialEq to GenesisConfig
* don't add Eq to GenesisConfig because floats
* add Eq, PartialEq to PohConfig
* add Eq, PartialEq to SecpSignatureOffsets
* add some missing derives to transaction_context.rs
* add Eq, PartialEq to ProcessedSiblingInstruction
This reverts commit 16853acf35.
We get similar coverage from the slot field of bank-forks_set_root;
additionally, we can see banks with bank-new_from_parent-heights.
load_frozen_forks() finds new slots to process by creating new Banks for
the children of the current slot in process_next_slots(). Prior to this
change, we would then immediately check if we had reached the
halt_at_slot and correctly halt processing when appropriate. As such, it
would be possible for Banks to be created for slots beyond the
halt_at_slot.
While a potential child slot that is past halt_at_slot wouldn't be
replayed, the Bank being created still alters some universal state in
AccountsDb. So, this change moves the halt_at_slot check before we
create children Banks in process_next_slots().
* Moves CachedExecutors, related structs, consts and tests into the program-runtime crate.
* Moves TransactionExecutor, related enum and type defs into executor_cache mod.
#### Summary of Changes
This PR makes two FIFO-related validator arguments public:
--rocksdb_shred_compaction and --rocksdb_fifo_shred_storage_size.
#### Test Plan
* There're already ~26 validators running FIFO in mainnet-beta for more than 30 days,
no issues were reported so far.
* Ran a validator with FIFO for days and observed it's able to consistently catch up
and create new roots.
* Use infallible, unchecked methods to write into the serialization buffer
We serialize in two steps: first we compute the size of the buffer, then
we write into it. Therefore there's no need to check if each individual
write fits the buffer - we know it does we just computed the required
size.
* serialize_parameters: remove extra loop/borrows
Remove one extra loop over accounts to gather account lengths. Also
gather all accounts at once and avoid temporary borrows.
* Move creating MemoryRegions for serialized parameters from create_vm to serialize_parameters
This is in preparation of using multiple MemoryRegions once we land direct account mapping.
* bpf_loader: introduce internal API to build serialization buffer/regions
This is prep work for landing the direct_mapping feature, which maps account
data in their own memory regions.
* serialization: fix after API changes