In particular, we track separately whether each AtExitEntry has already
been called, through a separate Bitmap. This has several side-effects:
- We now call malloc() during __cxa_finalize(). I believe this is fine,
and at that point during program execution memory pressure should be
low anyway.
- An attacker could prevent arbitrary entries from executing by writing
to atexit_called_entries. However, this already was possible (by
setting atexit_entry_count to zero), and this path is even more
troublesome (the attacker needs to overwrite atexit_called_entries,
and a region serving as *atexit_called_entries.m_data, and magically
know exactly how many entries already exist.)
- This reduces the size of AtExitEntry from 16 to 12 (on i686). As such,
we can reduce the initial memory allocation from two to one page,
reducing the initial capacity from 512 to 341 entries (or 256 to 170,
on x86_64). It seems that most programs only use 36-47 entries anyway.
For 'true', this shaves off about 69 syscalls, as measured by strace.
Previously, a collection of notes (Vector or Array) would be created and
promptly deleted for every sample (at least 44 thousand times per
second!). This was measured to be one of the most significant
performance drawbacks as well as the most obvious performance
improvement I could currently find here. Although it will not cause
Piano to lag currently (at least on virtualized systems), I see an
incoming issue once we get the capability to use more processors.
Now, we use a HashMap correlating pitches to notes, and Track reuses the
data structure in order to avoid reallocations. That is the reason for
introducing the fast clear_with_capacity to HashMap.
Always throws at the moment, because parse_temporal_month_day_string()
is basically a stub, and parse_iso_date_time() isn't functional either.
The spec issue has been resolved though, so I figured we might as well
get one small step further :^)
Same as Vector, ByteBuffer now also signals allocation failure by
returning an ENOMEM Error instead of a bool, allowing us to use the
TRY() and MUST() patterns.
Instead of signalling allocation failure with a bool return value
(false), we now use ErrorOr<void> and return ENOMEM as appropriate.
This allows us to use TRY() and MUST() with Vector. :^)
First off, this verifies that an initial value is always provided in
Properties.json for each property.
Second, it verifies that parsing that initial value succeeds.
This means that a call to `property_initial_value()` will always return
a valid StyleValue. :^)
It's a little verbose to repeat these in cases like the borders, but if
everything has an initial value, we can guarantee that
`property_initial_value()` will return something! :^)
- `align-items`: `normal` is the initial value in the CSS-ALIGN spec,
but `stretch` is in CSS-FLEXBOX. The FLEXBOX spec is the one we've
actually implemented elsewhere, and ALIGN adds new values with special
syntax, so it's not trivial to add it here.
- `border-spacing`: `0` is equivalent to `0px 0px` and we don't yet
parse the double-value syntax.
- `text-decoration-thickness`: Had the wrong value.
In particular, fstatvfs used to assume that a file that was earlier
opened using some path will forever be at that path. This is wrong, and
in the meantime new mounts and new filesystems could take up the
filename or directories, leading to a completely inaccurate result.
This commit improves the situation:
- All filesystem information is now always accurate.
- The mount flags *might* be erroneously zero, if the custody for the
open file is not available. I don't know when that might happen, but
it is definitely not the typical case.
This patch introduces table joins. It uses a pretty dumb algorithm-
starting with a singleton '__unity__' row consisting of a single boolean
value, a cartesian product of all tables in the 'FROM' clause is built.
This cartesian product is then filtered through the 'WHERE' clause,
again without any smarts just using brute force.
This patch required a bunch of busy work to allow for example the
ColumnNameExpression having to deal with multiple tables potentially
having columns with the same name.
Because SQL is the craptastic language that it is, sometimes expressions
need to know details about the calling statement. For example the tables
in the 'FROM' clause may be needed to determine which columns are
referenced in 'WHERE' expressions. So the current statement is added
to the ExecutionContext and a new 'execute' overload on Statement is
created which takes the Database and the Statement and builds an
ExecutionContaxt from those.
These are needed to distinguish columns from different tables with the
same column name in one and the same (joined) Tuple. Not quite happy
yet with this API; I think some sort of hierarchical structure would be
better but we'll burn that bridge when we get there :^)
These files contain the same SQL statements as the similarly named tests
in Tests/LibSQL/TestSqlStatementExecution.cpp test suite. They can be
fed to the sql utility to assist in troubleshooting failing tests.
