This PR revives cockroachdb#26916 (copying over the commit message from there).
It's not strictly necessary, but is a quality of life improvement.
There are also miscellaneous cleanups around TeeEngine included
therein.

Read-only batches, as created by engine.NewReadOnly, are the wrong
abstraction. They conflate two unrelated goals:

  - Ensure read-only commands do not accidentally write data.
  - Improve performance of code paths that can reuse iterators by
  caching the iterator object returned by NewIterator.

The fact that engine.NewReadOnly must return a ReadWriter, instead of a
Reader, to satisfy both of these goals is evidence that these two goals
should be addressed separately.

To better accomplish the first goal, we can leverage the Go type system.
This commit adjusts command registration so that commands declare up
front whether they are read-only or mutating. Read-only commands take an
engine.Reader, while mutating commands take an engine.ReadWriter. This
prevents accidental writes at compile time, whereas the read-only batch
could only check for misuse at runtime.

To accomplish the second goal, this commit replaces read-only batches
with an engine wrapper, rocksDBIterCacher, that does nothing but cache
iterators. This wrapper might also prove useful to cache iterators in
code paths that wish to perform mutations but do not need a full-fledged
batch.

NB: There's still the unaddressed wart that is replica read code paths
operating off an engine.ReadWriter. Changing this out would require a
refactor of replica evaluation code paths, which I'm not doing now.

Release note: None

This PR revives cockroachdb#26916 (copying over the commit message from there).
It's not strictly necessary, but is a quality of life improvement.
There are also miscellaneous cleanups around TeeEngine included
therein.

Read-only batches, as created by engine.NewReadOnly, are the wrong
abstraction. They conflate two unrelated goals:

  - Ensure read-only commands do not accidentally write data.
  - Improve performance of code paths that can reuse iterators by
  caching the iterator object returned by NewIterator.

The fact that engine.NewReadOnly must return a ReadWriter, instead of a
Reader, to satisfy both of these goals is evidence that these two goals
should be addressed separately.

To better accomplish the first goal, we can leverage the Go type system.
This commit adjusts command registration so that commands declare up
front whether they are read-only or mutating. Read-only commands take an
engine.Reader, while mutating commands take an engine.ReadWriter. This
prevents accidental writes at compile time, whereas the read-only batch
could only check for misuse at runtime.

To accomplish the second goal, this commit replaces read-only batches
with an engine wrapper, rocksDBIterCacher, that does nothing but cache
iterators. This wrapper might also prove useful to cache iterators in
code paths that wish to perform mutations but do not need a full-fledged
batch.

NB: There's still the unaddressed wart that is replica read code paths
operating off an engine.ReadWriter. Changing this out would require a
refactor of replica evaluation code paths, which I'm not doing now.

Release note: None

This PR reworks parts of cockroachdb#26916. It's not strictly necessary, but is a
quality of life improvement. There are also miscellaneous cleanups
around TeeEngine, RocksDB benchmarks, etc.

Previous to this we relied on engine.NewReadOnly's runtime panics when
receiving writes to ensure that read-only commands do not accidentally
write data. By changing the command registration process to have
commands declare upfront whether they are read-only or read-write, we
can plumb in engine.Reader or engine.ReadWriter respectively. It's nice
rely on the type system to assert these guarantees.

NB: There's still the unaddressed wart that is replica read code paths
operating off an engine.ReadWriter. Changing this out would require a
refactor of replica evaluation code paths, which I'm not doing now. This
PR however paves the way for that.

Release note: None

This PR reworks parts of cockroachdb#26916. It's not strictly necessary, but is a
quality of life improvement. There are also miscellaneous cleanups
around TeeEngine, RocksDB benchmarks, etc.

Previous to this we relied on engine.NewReadOnly's runtime panics when
receiving writes to ensure that read-only commands do not accidentally
write data. By changing the command registration process to have
commands declare upfront whether they are read-only or read-write, we
can plumb in engine.Reader or engine.ReadWriter respectively. It's nice
rely on the type system to assert these guarantees.

NB: There's still the unaddressed wart that is replica read code paths
operating off an engine.ReadWriter. Changing this out would require a
refactor of replica evaluation code paths, which I'm not doing now. This
PR however paves the way for that.

Release note: None

This PR reworks parts of cockroachdb#26916. It's not strictly necessary, but is a
quality of life improvement. There are also miscellaneous cleanups
around TeeEngine, RocksDB benchmarks, etc.

Previous to this we relied on engine.NewReadOnly's runtime panics when
receiving writes to ensure that read-only commands do not accidentally
write data. By changing the command registration process to have
commands declare upfront whether they are read-only or read-write, we
can plumb in engine.Reader or engine.ReadWriter respectively. It's nice
rely on the type system to assert these guarantees.

NB: There's still the unaddressed wart that is replica read code paths
operating off an engine.ReadWriter. Changing this out would require a
refactor of replica evaluation code paths, which I'm not doing now. This
PR however paves the way for that.

Release note: None

43764: storage: change command registration to declare if ro/rw r=irfansharif a=irfansharif

This PR reworks parts of #26916. It's not strictly necessary, but is a
quality of life improvement. There are also miscellaneous cleanups
around TeeEngine, RocksDB benchmarks, etc.

Previous to this we relied on engine.NewReadOnly's runtime panics when
receiving writes to ensure that read-only commands do not accidentally
write data. By changing the command registration process to have
commands declare upfront whether they are read-only or read-write, we
can plumb in engine.Reader or engine.ReadWriter respectively. It's nice
rely on the type system to assert these guarantees.

NB: There's still the unaddressed wart that is replica read code paths
operating off an engine.ReadWriter. Changing this out would require a
refactor of replica evaluation code paths, which I'm not doing now. This
PR however paves the way for that.

Release note: None


43779: pgwire/hba: fix the HBA parser r=knz a=knz

Previously, the HBA rule parser would not properly support the pg
split format for separate address and netmask.

When the 4th column of a rule in
the setting `server.host_based_authentication.configuration` is
an IP address, the 5th column is expected to be an IP netmask.

Prior to this patch, the HBA parser would not recognize the netmask
syntax and instead expect a simple number of bits set (which is not
valid pg syntax).

This patch fixes this and adds the corresponding tests.

Additionally, this patch replaces the HBA rule parser by one inspired
from pg's own implementation, so as to ease future maintainance.

Release note (bug fix): When the 4th column of a rule in the setting
`server.host_based_authentication.configuration` is an IP address
without a mask length (e.g. `1.2.3.4` instead of `1.2.0.0/16`),
CockroachDB now properly interprets the 5th column as an IP netmask,
as per https://www.postgresql.org/docs/current/auth-pg-hba-conf.html.

Co-authored-by: irfan sharif <irfanmahmoudsharif@gmail.com>
Co-authored-by: Raphael 'kena' Poss <knz@thaumogen.net>