This optimizes unbalanced multiple assignment cases such as:
```ruby
a.b, c.d = e, f, g
a.b, c.d, e.f = g, h
```
Previously, this would use:
```
newarray(3)
expandarray(2, 0)
newarray(2)
expandarray(3, 0)
```
These would both allocate arrays. This switches to opt_reverse
with either pop or putnil:
```
pop
opt_reverse(2)
putnil
opt_reverse(3)
```
This avoids an unnecessary array allocation, and results in a 35-76%
performance increase in these types of unbalanced cases (tested with
benchmark/masgn.yml).
This allows them to show the effect of the previous newarray/expandarray
to swap/opt_reverse optimization. This shows an 35-83% performance
improvement in the four multiple assignment benchmarks that use this
optimization.
This renames the reverse instruction to opt_reverse, since now it
is only added by the optimizer. Then it uses as a more general
form of swap. This optimizes multiple assignment in the popped
case with more than two elements.
An optimization for multiple assignment in the popped case to avoid
array allocation was lost in my fix to make multiple assignment follow
left-to-right evaluation (50c54d40a8).
Before, in the two element case, swap was used. Afterward, newarray(2)
and expandarray(2, 0) were used, which is the same as swap, with the
addition of an unnecessary allocation.
Because this issue is not specific to multiple assignment, and the
multiple assignment code is complex enough as it is, this updates
the peephole optimizer to do the newarray(2)/expandarray(2, 0) -> swap
conversion.
A more general optimization pass for
newarray(X)/expandarray(X, 0) -> reverse(X) will follow, but that
requires readding the reverse instruction.
Poisoned regions cannot be accessed without unpoisoning outside gc.c.
Specifically, debug.gem is terminated by AddressSanitizer.
```
SUMMARY: AddressSanitizer: use-after-poison iseq_collector.c:39 in iseq_i
```
dev is Shopify's internal tool that doesn't work if you use Intel
Homebrew on M1 (or rbenv, btw). Now that we maintain this outside
Shopify's repository, we should stop talking about it here.
- The method was renamed from `get` to `get_value`
- Comparing to `String#unpack` isn't quite equivalent, `unpack1` is closer.
- Use frozen_string_literal to avoid allocating a format string every time.
- Use `N` format which is equivalent to `:U32` (`uint_32_t` big-endian).
- Disable experimental warnings to not mess up the output.
e4e054e3ce used four footnotes
without blank lines. And the ChangeLog generated from that commit
resulted in ``undefined method `parts' for nil`` error.
For now, let a footnote terminated by the next footnote mark.
Also refined the error message when undefined footnote is used.
a7f290130b
Since the change at f310ac1cb2 to show
the backtraces by default, this test started to show the backtraces.
As the backtraces are not the subject of this test, silence them by
using Gem::SilentUI.
The argument of `rb_syswait` is now `rb_pid_t` which may differ from
`int`. Also it is an undefined behavior to take the result of casted
void function (in `rb_protect`).
This is a regression from a change intended to raise errors when user
puts a gem under an incorrect source in the Gemfile by mistake. To fix
the issue, we revert the change that caused it and implement it in a
different way that restores the resolver independency from real
specifications. Now it deals only with names and versions and does not
try to materialize anything into real specifications before resolving.
d2bf1b86eb
GCC warns of empty format strings, perhaps because they have no
effects in printf() and there are better ways than sprintf().
However, ruby_debug_log() adds informations other than the format,
this warning is not the case.
