ruby.git/missing/dtoa.c, branch v4.0.2 The Ruby Programming Language Fix a possible memory leak in dtoa 2025-12-24T05:05:41+00:00 Nobuyoshi Nakada nobu@ruby-lang.org 2025-12-24T02:30:11+00:00 44693ee32990fdd609e91ebb2970b3110b3f08e5 Fix GH-15061 
Fix GH-15061
dtoa.c: Check integer underflow 2025-09-10T07:45:25+00:00 Nobuyoshi Nakada nobu@ruby-lang.org 2025-09-10T03:57:34+00:00 624538ba927b3d5fa021af0df3c8c29b11ef14f7 Reported at https://hackerone.com/reports/3288162 This underflow does not occur in Ruby because: * This function is `static` and not accessible other than from ruby internal. * Ruby uses mode 0 when calling this function directly. * For `%f` in vsnprintf.c using mode 3, this parameter comes from the precision, but negative precision is meaningless and ignored. 
Reported at https://hackerone.com/reports/3288162

This underflow does not occur in Ruby because:

* This function is `static` and not accessible other than from ruby
  internal.

* Ruby uses mode 0 when calling this function directly.

* For `%f` in vsnprintf.c using mode 3, this parameter comes from the
  precision, but negative precision is meaningless and ignored.
dtoa.c: Check memory allocation failures 2025-09-10T06:34:30+00:00 Nobuyoshi Nakada nobu@ruby-lang.org 2025-09-09T15:52:24+00:00 f047174cf16c42843807c8430d18a83d2aa6175e 






dtoa.c: Extract macro to update 5powers Bigint cache atomically
2025-09-10T06:34:30+00:00

Nobuyoshi Nakada
nobu@ruby-lang.org

2025-09-09T15:26:56+00:00

d0c966adcd8cd04830e75c6b720c0c38427b3419












dtoa.c: Add shortcut if arguments are zero
2025-09-10T06:34:29+00:00

Nobuyoshi Nakada
nobu@ruby-lang.org

2025-09-09T15:23:56+00:00

0887d82406c9a9701acd235e8f510480cea9f218












Remove spinlock in dtoa's Balloc, use xmalloc
2025-04-07T23:56:44+00:00

John Hawthorn
john@hawthorn.email

2025-03-27T01:23:04+00:00

b3b1d7b2952037727abfa66236d43e434003524a

The spinlock here performs poorly when there are multiple Ractors. The
improvement on single threaded performance doesn't seem that
significant, so I think we should just use malloc.





The spinlock here performs poorly when there are multiple Ractors. The
improvement on single threaded performance doesn't seem that
significant, so I think we should just use malloc.







[Bug #20705] Update `strtod` implementation
2024-10-05T14:29:42+00:00

Nobuyoshi Nakada
nobu@ruby-lang.org

2024-09-05T13:08:41+00:00

d17edf3a170b733356836353508319443d12c53c

The absence of either the integer or fractional part should be
allowed.





The absence of either the integer or fractional part should be
allowed.







Make dtoa.c buildable alone
2023-07-14T09:35:23+00:00

Nobuyoshi Nakada
nobu@ruby-lang.org

2023-07-14T05:33:21+00:00

bc8cc68aeff7ae3be1353f4077633cda4d30b54b












Fix dtoa buffer overrun
2022-04-12T12:30:49+00:00

Nobuyoshi Nakada
nobu@ruby-lang.org

2021-06-30T21:39:17+00:00

d0a822eec524522d81ffc7da2bb1baf906b0318a

https://hackerone.com/reports/1248108





https://hackerone.com/reports/1248108







Fixed race in dtoa [Bug #17612]
2021-02-10T10:44:47+00:00

Nobuyoshi Nakada
nobu@ruby-lang.org

2021-02-10T06:24:23+00:00

3acc81d9e41b18380b9e0168fe2b5e5e0c727256

Fixed the race condition when replacing `freelist` entry with its
chained next element.  At acquiring an entry, hold the entry once
with the special value, then release by replacing it with the next
element again after acquired.  If another thread is holding the
same entry at that time, spinning until the entry gets released.

Co-Authored-By: Koichi Sasada <ko1@atdot.net>





Fixed the race condition when replacing `freelist` entry with its
chained next element.  At acquiring an entry, hold the entry once
with the special value, then release by replacing it with the next
element again after acquired.  If another thread is holding the
same entry at that time, spinning until the entry gets released.

Co-Authored-By: Koichi Sasada <ko1@atdot.net>