General

Profile

kivikakk (Asherah Connor)

  • Login: kivikakk
  • Registered on: 01/24/2014
  • Last sign in: 10/27/2024

Issues

open closed Total
Assigned issues 0 0 0
Reported issues 0 2 2

Activity

07/19/2018

06:29 AM Ruby Feature #5446: at_fork callback API
Are we still interested in pursuing something like this? I'd be happy to push it forward. kivikakk (Asherah Connor)

06/25/2018

05:47 AM Ruby Bug #14834: rb_profile_frames SEGV when PC adjusted on IFUNC
Here's an easy way to reproduce the segfault on current trunk:
`Gemfile`
~~~ ruby
source "https://rubygems.org"
gem "stackprof"
~~~
`test.rb`
~~~ ruby
require 'stackprof'
StackProf.run(mode: :object) do
[123].grou...
kivikakk (Asherah Connor)

06/08/2018

05:44 AM Ruby Bug #14834: rb_profile_frames SEGV when PC adjusted on IFUNC
Further:
* is the original change correct? Some instructions have the `handle_frame?` property, which means they *do* increment the PC by the instruction width first. It's those that don't have that property which now increment the PC...
kivikakk (Asherah Connor)
05:13 AM Ruby Bug #14834: rb_profile_frames SEGV when PC adjusted on IFUNC
It's also worth noting:
* is the increment of `pc` with `ec->cfp->pc++` correct? What if the instruction is multiple values wide?
* there are similar `pc` increment/decrement pairs around `EXEC_EVENT_HOOK` calls in `vm_trace`. Do we ...
kivikakk (Asherah Connor)
05:07 AM Ruby Bug #14834 (Closed): rb_profile_frames SEGV when PC adjusted on IFUNC
Since r62052, we increment `ec->cfp->pc` by one pointer width (e.g. 8 bytes) in `gc_event_hook_body` around the `EXEC_EVENT_HOOK` call.
This becomes a problem when the hook is on an IFUNC: in this case, `pc == 0x0`, meaning we increme...
kivikakk (Asherah Connor)

03/08/2018

01:55 AM Ruby Bug #14586 (Closed): URI::RFC2396_Parser#unescape raises an exception if the input is mixed Unicode and percent-escapes
Currently, the following test case passes:
~~~ ruby
def test_unescape
p1 = URI::Parser.new
assert_equal("\xe3\x83\x90", p1.unescape("\xe3\x83\x90"))
assert_equal("\xe3\x83\x90", p1.unescape('%e3%83%90'))
end
~~~
But the...
kivikakk (Asherah Connor)

01/25/2016

01:22 AM Ruby Misc #12004: Code of Conduct
The new guidelines look good. I support Shyouhei's suggestion to leave the final decision up to Matz. kivikakk (Asherah Connor)

01/24/2014

03:25 AM Ruby Bug #9424: ruby 1.9 & 2.x has insecure SSL/TLS client defaults
Shyouhei Urabe wrote:
> I don't wanna be raped like them.
How on earth can you think this is appropriate?
kivikakk (Asherah Connor)

Also available in: Atom