enc_table which manages Encoding information. rb_encoding_list
also manages Encoding objects. Both are accessed/modified by ractors
simultaneously so that they should be synchronized.
For enc_table, this patch introduced GLOBAL_ENC_TABLE_ENTER/LEAVE/EVAL
to access this table with VM lock. To make shortcut, three new global
variables global_enc_ascii, global_enc_utf_8, global_enc_us_ascii are
also introduced.
For rb_encoding_list, we split it to rb_default_encoding_list (256 entries)
and rb_additional_encoding_list. rb_default_encoding_list is fixed sized Array
so we don't need to synchronized (and most of apps only needs it). To manage
257 or more encoding objects, they are stored into rb_additional_encoding_list.
To access rb_additional_encoding_list., VM lock is needed.
If a ractor getting a VM lock (monitor) raises an exception,
unlock can be skipped. To release VM lock correctly on exception
(or other jumps with JUMP_TAG), EC_POP_TAG() releases VM lock.
Using Fiber#transfer with Fiber#resume for a same Fiber is
limited (once Fiber#transfer is called for a fiber, the fiber
can not be resumed more). This restriction was introduced to
protect the resume/yield chain, but we realized that it is too much
to protect the chain. Instead of the current restriction, we
introduce some other protections.
(1) can not transfer to the resuming fiber.
(2) can not transfer to the yielding fiber.
(3) can not resume transferred fiber.
(4) can not yield from not-resumed fiber.
[Bug #17221]
Also at the end of a transferred fiber, it had continued on root fiber.
However, if the root fiber resumed a fiber (and that fiber can resumed
another fiber), this behavior also breaks the resume/yield chain.
So at the end of a transferred fiber, switch to the edge of resume
chain from root fiber.
For example, root fiber resumed f1 and f1 resumed f2, transferred to
f3 and f3 terminated, then continue from the fiber f2 (it was continued
from root fiber without this patch).
It uses random to determine if the bignum is sparse or not.
It is arguable if three-digit samples are enough or not to determine it,
but anyway, consuming Random source implicitly is not good.
I introduced the random sampling mechanism, and I don't know any
significant reason to do so. So, let's remove it.
This change makes the sampling points fixed: 40th, 50th, and 60th
percentiles.
When `min: 0` is given to `IO#getch` with `time:` option, it is
expected to return nil if timed out, and needed for source code
the compatibility with unixen platforms.
https://github.com/ruby/io-console/commit/a2afbe72bd
Fixed TypeError when IO#getch timed out
`rb_io_wait` returns a bit-flags Integer representing available
events, or Qfalse if timed out. Also the result of `NUM2INT` is
not a `VALUE`.
```
$ ./bin/ruby -v -rio/console -e "p IO.console.getch(intr: true, time: 0.1)"
ruby 3.0.0dev (2020-10-09T20:27:30Z master 5ea2ea74cc) [x64-mingw32]
-e:1:in `getch': no implicit conversion of false into Integer (TypeError)
from -e:1:in `<main>'
```
https://github.com/ruby/io-console/commit/3bdfaf62df
This fixes the lldb disassembler script so that it doesn't need a live
process when disassembling rb_iseq_t. I also added the PC to the output
so you can tell what the VM is executing when it crashed.
For example:
```
(lldb) rbdisasm ec->cfp->iseq
PC IDX insn_name(operands)
0x56039f0a1720 0000 nop
0x56039f0a1728 0001 getlocal_WC_1( 5 )
0x56039f0a1738 0003 branchunless( 7 )
0x56039f0a1748 0005 getlocal_WC_0( 3 )
0x56039f0a1758 0007 putstring( (VALUE)0x56039f0c7eb8 )
0x56039f0a1768 0009 opt_send_without_block( (struct rb_call_data *)0x56039f09f140 )
0x56039f0a1778 0011 pop
0x56039f0a1780 0012 getglobal( ID: 0x7fd7 )
0x56039f0a1790 0014 branchunless( 7 )
0x56039f0a17a0 0016 getlocal_WC_0( 3 )
0x56039f0a17b0 0018 putstring( (VALUE)0x56039f0c7e90 )
0x56039f0a17c0 0020 opt_send_without_block( (struct rb_call_data *)0x56039f09f150 )
0x56039f0a17d0 0022 pop
0x56039f0a17d8 0023 getlocal_WC_0( 3 )
0x56039f0a17e8 0025 putobject( (VALUE)0x56039f0c7e68 )
0x56039f0a17f8 0027 getlocal_WC_1( 6 )
0x56039f0a1808 0029 dup
0x56039f0a1810 0030 checktype( 5 )
0x56039f0a1820 0032 branchif( 4 )
0x56039f0a1830 0034 dup
0x56039f0a1838 0035 opt_send_without_block( (struct rb_call_data *)0x56039f09f160 )
0x56039f0a1848 0037 tostring
0x56039f0a1850 0038 putobject( (VALUE)0x56039f0c7e40 )
0x56039f0a1860 0040 concatstrings( 3 )
0x56039f0a1870 0042 opt_send_without_block( (struct rb_call_data *)0x56039f09f170 )
0x56039f0a1880 0044 nop
0x56039f0a1888 0045 leave
(lldb) p ec->cfp->pc
(const VALUE *) $146 = 0x000056039f0a1848
```
Here we can see the VM is currently executing `opt_send_without_block`
(because the PC is one ahead of the current instruction)
The write barrier wasn't being called for this object, so add the
missing WB. Automatic compaction moved the reference because it didn't
know about the relationship (that's how I found the missing WB).
This iterator uses an st_table, but if objects move the references in
the st table won't be updated. This patch just changes the st table to
an identity hash.
Hiroshi SHIBATA
Koichi Sasada
git
Nobuyoshi Nakada
Yusuke Endoh
Cristian Greco
tompng
MSP-Greg
Victor Goff
Benoit Daloze
Soutaro Matsumoto
S.H
Ikko Ashimine
Aaron Patterson
Burdette Lamar
Chris Seaton