This is a proof-of-concept of switching to a new memory layout for
tokens and AST nodes. The goal is threefold:
* smaller memory footprint
* faster performance for tokenization and parsing
* most importantly, a proof-of-concept that can be also applied to ZIR
and TZIR to improve the entire compiler pipeline in this way.
I had a few key insights here:
* Underlying premise: using less memory will make things faster, because
of fewer allocations and better cache utilization. Also using less
memory is valuable in and of itself.
* Using a Struct-Of-Arrays for tokens and AST nodes, saves the bytes of
padding between the enum tag (which kind of token is it; which kind
of AST node is it) and the next fields in the struct. It also improves
cache coherence, since one can peek ahead in the tokens array without
having to load the source locations of tokens.
* Token memory can be conserved by only having the tag (1 byte) and byte
offset (4 bytes) for a total of 5 bytes per token. It is not necessary
to store the token ending byte offset because one can always re-tokenize
later, but also most tokens the length can be trivially determined from
the tag alone, and for ones where it doesn't, string literals for
example, one must parse the string literal again later anyway in
astgen, making it free to re-tokenize.
* AST nodes do not actually need to store more than 1 token index because
one can poke left and right in the tokens array very cheaply.
So far we are left with one big problem though: how can we put AST nodes
into an array, since different AST nodes are different sizes?
This is where my key observation comes in: one can have a hash table for
the extra data for the less common AST nodes! But it gets even better than
that:
I defined this data that is always present for every AST Node:
* tag (1 byte)
- which AST node is it
* main_token (4 bytes, index into tokens array)
- the tag determines which token this points to
* struct{lhs: u32, rhs: u32}
- enough to store 2 indexes to other AST nodes, the tag determines
how to interpret this data
You can see how a binary operation, such as `a * b` would fit into this
structure perfectly. A unary operation, such as `*a` would also fit,
and leave `rhs` unused. So this is a total of 13 bytes per AST node.
And again, we don't have to pay for the padding to round up to 16 because
we store in struct-of-arrays format.
I made a further observation: the only kind of data AST nodes need to
store other than the main_token is indexes to sub-expressions. That's it.
The only purpose of an AST is to bring a tree structure to a list of tokens.
This observation means all the data that nodes store are only sets of u32
indexes to other nodes. The other tokens can be found later by the compiler,
by poking around in the tokens array, which again is super fast because it
is struct-of-arrays, so you often only need to look at the token tags array,
which is an array of bytes, very cache friendly.
So for nearly every kind of AST node, you can store it in 13 bytes. For the
rarer AST nodes that have 3 or more indexes to other nodes to store, either
the lhs or the rhs will be repurposed to be an index into an extra_data array
which contains the extra AST node indexes. In other words, no hash table needed,
it's just 1 big ArrayList with the extra data for AST Nodes.
Final observation, no need to have a canonical tag for a given AST. For example:
The expression `foo(bar)` is a function call. Function calls can have any
number of parameters. However in this example, we can encode the function
call into the AST with a tag called `FunctionCallOnlyOneParam`, and use lhs
for the function expr and rhs for the only parameter expr. Meanwhile if the
code was `foo(bar, baz)` then the AST node would have to be `FunctionCall`
with lhs still being the function expr, but rhs being the index into
`extra_data`. Then because the tag is `FunctionCall` it means
`extra_data[rhs]` is the "start" and `extra_data[rhs+1]` is the "end".
Now the range `extra_data[start..end]` describes the list of parameters
to the function.
Point being, you only have to pay for the extra bytes if the AST actually
requires it. There's no limit to the number of different AST tag encodings.
Preliminary results:
* 15% improvement on cache-misses
* 28% improvement on total instructions executed
* 26% improvement on total CPU cycles
* 22% improvement on wall clock time
This is 1/4 items on the checklist before this can actually be merged:
* [x] parser
* [ ] render (zig fmt)
* [ ] astgen
* [ ] translate-c
It now uses the log scope "gpa" instead of "std".
Additionally, there is a new config option `verbose_log` which enables
info log messages for every allocation. Can be useful when debugging.
This option is off by default.
Also known as "Struct-Of-Arrays" or "SOA". The purpose of this data
structure is to provide a similar API to ArrayList but instead of
the element type being a struct, the fields of the struct are in N
different arrays, all with the same length and capacity.
Having this abstraction means we can put them in the same allocation,
avoiding overhead with the allocator. It also saves a tiny bit of
overhead from the redundant capacity and length fields, since each
struct element shares the same value.
This is an alternate implementation to #7854.
This temporary patch fixes a segfault caused by miscompilation
by the LLD when generating stubs for initialization of thread local
storage. We effectively bypass TLS in the default panic handler
so that no segfault is generated and the stack trace is correctly
reported back to the user.
Note that, this is linked directly to a bigger issue with LLD
ziglang/zig#7527 and when resolved, we only need to remove the
`comptime` code path introduced with this patch to use the default
panic handler that relies on TLS.
Co-authored-by: Andrew Kelley <andrew@ziglang.org>
ideal capacity is now determined by e.g.
x += x / f
rather than
x = x * b / a
This turns a multiplication into an addition, making it less likely to
overflow the integer. This commit also introduces padToIdeal() which
does saturating arithmetic so that no overflow is possible when
calculating ideal capacity.
closes#7830
* move concurrency primitives that always operate on kernel threads to
the std.Thread namespace
* remove std.SpinLock. Nobody should use this in a non-freestanding
environment; the other primitives are always preferable. In
freestanding, it will be necessary to put custom spin logic in there,
so there are no use cases for a std lib version.
* move some std lib files to the top level fields convention
* add std.Thread.spinLoopHint
* add std.Thread.Condition
* add std.Thread.Semaphore
* new implementation of std.Thread.Mutex for Windows and non-pthreads Linux
* add std.Thread.RwLock
Implementations provided by @kprotty
