This commit makes sure the batch requests are always GCed before sending
back the responses for them. In particular,
- `removeRequest` didn't remove the item from `batchList`, which is now
fixed.
- `Send` didn't remove the request from `requestMap`, which is now fixed
by using `removeRequest`.
This commit creates a `RejectReasonMap` to map the errors returned from
`btcd` to bitcoind's `testmempoolaccept` so the `RejectReason` is
unified at the RPC level. To make sure the map keys are unique, the
error strings are modified in `btcd`.
This commit adds bitcoind version 22.0 and 25.0 to our `BackendVersion`
set to handle the `testmempoolaccept` RPC calls. A unit test is added to
make sure the parser works as expected.
This commit breaks the `maybeAcceptTransaction` into two parts - the
first is reading the mempool to validate the transaction, and the
relevant logic has been moved into the new method
`checkMempoolAcceptance`. The second part is writing to the mempool, and
is kept in the method `maybeAcceptTransaction`.
On startup when the headers-first mode is off, when receiving the first
block, the periodic flush will trigger. The lastflushtime wasn't set
which resulted in the flush being triggered on the first block on
restart.
This commit adds a new NextAvailablePortForProcess function that takes a
process ID and then assures unique (non-occupied) port numbers are
returned per process.
This uses a temporary file that contains the latest used port and a
secondary temporary lock file to assure only a single goroutine can
request a new port at a time.
The GenerateProcessUniqueListenerAddresses is intened to be used as a
package-level override for the ListenAddressGenerator variable. We don't
use it by default to make sure we don't break any existing assumptions.
btcutil.Block caches the serialized raw bytes of the block during ibd.
This serialized block bytes includes the serialized tx. The current tx
hash generation will re-serialized the de-serialized tx to create the
raw bytes and it'll only then hash that.
This commit changes the code so that the re-serialization never happens,
saving tons of cpu and memory overhead.
We used to use a lot of small buffers for serialization, but now we'll
use one buffer large enough, and slice into it when needed.
``
name old time/op new time/op delta
CalcWitnessSigHash-8 31.5µs ± 0% 29.2µs ± 0% -7.05% (p=0.000 n=10+10)
name old alloc/op new alloc/op delta
CalcWitnessSigHash-8 19.9kB ± 0% 18.5kB ± 0% -7.14% (p=0.000 n=10+10)
name old allocs/op new allocs/op delta
CalcWitnessSigHash-8 801 ± 0% 445 ± 0% -44.44% (p=0.000 n=10+10)
```
In this commit, we optimize the sighash calc further by writing directly
into the buffer used for serialization by the sha256.New() instance
rather than to an intermediate buffer, which is then write to the hash
buffer.
In this commit, we update the top-level btcd package to use the latest
version of btcutil and also the chainhash package. With this version
bump, we can now use the new optimized dsha256 routine where applicable.
With this commit, I've covered most of the areas we'll hash an entire
transaction/block/header, but we may want to optimize some other areas
further, in particular, the witness sighash calc.
