We had a long discussion of this, so trying to document what seems to be
the conclusion... and also clean up the exceptions that I could find.
---------
Co-authored-by: Dana Jansens <danakj@orodu.net>
When using a byte-encoding for matched group metadata we need to mask
down to a single bit in each matching byte to make the iteration of a
range of match indices work. In most cases, this mask can be folded into
the overall match computation, but for Arm Neon, there is avoidable
overhead from this. Instead, we can defer the mask until starting to
iterate. Doing more than one iteration is relative rare so this doesn't
accumulate much waste and makes common paths a bit faster.
For the M1 this makes the SIMD match path about 2-4% faster. This isn't
enough to catch the portable match code path on the M1 though.
For some Neoverse cores the difference here is more significant (>10%
improvement) and it makes the SIMD and scalar code paths have comparable
latency. Still not clear which is better as the latency is comparable
and beyond latency the factors are very hard to analyze -- port pressure
on different parts of the CPU, etc.
Leaving the selected code path as portable since that's so much better
on the M1, and I'm hoping to avoid different code paths for different
Arm CPUs for a while.
---------
Co-authored-by: Danila Kutenin <danilak@google.com>
Disables three new warnings because they lean more towards style
conflicts than fixes. I've brought these up on #style.
Other than that, mostly fixing basic issues, and things that
clang-tidy-20 seems to fire where clang-tiday-16 didn't. One particular
curious case is `llvm::StringLiteral::data()` uses, which are flagged as
not strictly null-terminated; I'm switching to `const char*` in those
spots which matches `llvm::formatv`'s format argument, but feels worse.
I'm removing `run_clang_tidy.py` here because I'm observing it give
fewer warnings than `bazel build --config=clang-tidy -k
//toolchain/...`. The latter matches how we enforce in GitHub actions
(and also caches results, and suppresses output for files that have no
issues), so I'm dropping the bespoke script.
Rely on implicit conversion in call sites and initialization.
Removing the explicit conversions is only code simplication.
Moving from `auto x = Y(z)` to `Y x = z;` helps ensure that only
implicit constructors/conversions are happening (whereas the prior
syntax allows explicit conversions) which can help with readability
since implicit conversions are generally "less
complex"/risky/attention-requiring.
The hash table design is heavily based on Abseil's ["Swiss
Tables"][swiss-tables] design. It uses an array of bytes storing
metadata about each entry and an array of entries where each is a pair
of key and value. The metadata byte consists of 7-bits of hash of the
key (distinct from the bits used to index the table), and one bit
indicating the presence of a special entry -- either empty or deleted.
[swiss-tables]: https://abseil.io/about/design/swisstables
There are a large range of optimizations and other nuanced aspects of
this hash table design and implementation, a good point to understand
that context is `raw_hashtable.h` which has an overview of the design
and references to various other files for relevant details.
---------
Co-authored-by: josh11b <15258583+josh11b@users.noreply.github.com>
Thomas Köppe
Jon Ross-PerkinsandDana Jansens
Chandler CarruthandDanila Kutenin
David Blaikie
