[#81492] [Ruby trunk Feature#13618] [PATCH] auto fiber schedule for rb_wait_for_single_fd and rb_waitpid — normalperson@...
Issue #13618 has been reported by normalperson (Eric Wong).
12 messages
2017/06/01
[#88695] Re: [Ruby trunk Feature#13618] [PATCH] auto fiber schedule for rb_wait_for_single_fd and rb_waitpid
— Eric Wong <normalperson@...>
2018/08/27
> https://bugs.ruby-lang.org/issues/13618
[#81569] [Ruby trunk Feature#12589] VM performance improvement proposal — vmakarov@...
Issue #12589 has been updated by vmakarov (Vladimir Makarov).
3 messages
2017/06/04
[#81581] [Ruby trunk Bug#13632] Not processable interrupt queue for a thread after it's notified that FD is closed in some other thread. — sir.nickolas@...
Issue #13632 has been reported by nvashchenko (Nikolay Vashchenko).
4 messages
2017/06/05
[#81590] Re: [ruby-cvs:66197] ko1:r59023 (trunk): revert r59020 because it may fail some tests sometimes on some environment (http://ci.rvm.jp/). This revert is to check the reason of failures. — Eric Wong <normalperson@...>
ko1@ruby-lang.org wrote:
5 messages
2017/06/06
[#81591] Re: [ruby-cvs:66197] ko1:r59023 (trunk): revert r59020 because it may fail some tests sometimes on some environment (http://ci.rvm.jp/). This revert is to check the reason of failures.
— Eric Wong <normalperson@...>
2017/06/06
Eric Wong <normalperson@yhbt.net> wrote:
[#81596] Re: [ruby-cvs:66203] Re: Re: ko1:r59023 (trunk): revert r59020 because it may fail some tests sometimes on some environment (http://ci.rvm.jp/). This revert is to check the reason of failures.
— Eric Wong <normalperson@...>
2017/06/06
Eric Wong <normalperson@yhbt.net> wrote:
[#81825] [Ruby trunk Feature#13697] [PATCH]: futex based thread primitives — normalperson@...
Issue #13697 has been reported by normalperson (Eric Wong).
3 messages
2017/06/29
[ruby-core:81689] [Ruby trunk Bug#13662] prohibit machine stackoverflow while GC
From:
ko1@...
Date:
2017-06-15 07:14:12 UTC
List:
ruby-core #81689
Issue #13662 has been reported by ko1 (Koichi Sasada).
----------------------------------------
Bug #13662: prohibit machine stackoverflow while GC
https://bugs.ruby-lang.org/issues/13662
* Author: ko1 (Koichi Sasada)
* Status: Open
* Priority: Normal
* Assignee:
* Target version:
* ruby -v:
* Backport: 2.2: UNKNOWN, 2.3: UNKNOWN, 2.4: UNKNOWN
----------------------------------------
We have encounter GC marking miss after [Feature #13637] r59047 at test-rubyspec.
> I, [2017-06-15T04:31:33.467582 #2820] INFO -- : ........................................................................................................................................................................................................................................................................................................................................................................................................../home/ko1/ruby/src/trunk/spec/rubyspec/core/encoding/converter/convpath_spec.rb:7: [BUG] rb_gc_mark(): 0x0000000001cbbf78 is T_NONE
http://ci.rvm.jp/results/trunk@P895/1497485282
http://ci.rvm.jp/results/trunk@P895/1497479452
http://ci.rvm.jp/results/trunk@P895/1497468696
... and more (not frequent).
My current assumption is machine stack overflow while GC and finishing marking phase incomplete.
To prevent such situation (incomplete GC), I'll prohibit machine stack overflow while running GC.
```
Index: vm_insnhelper.c
===================================================================
--- vm_insnhelper.c (revision 59088)
+++ vm_insnhelper.c (working copy)
@@ -56,8 +56,11 @@
void
rb_threadptr_stack_overflow(rb_thread_t *th)
{
+ if (rb_during_gc()) {
+ rb_fatal("machine stack overflow while GC is running.");
+ }
#ifdef USE_SIGALTSTACK
- threadptr_stack_overflow(th, !rb_threadptr_during_gc(th));
+ threadptr_stack_overflow(th, TRUE);
#else
threadptr_stack_overflow(th, FALSE);
#endif
```
--
https://bugs.ruby-lang.org/
Unsubscribe: <mailto:ruby-core-request@ruby-lang.org?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>