From pub.claudio@gmail.com

Hi,

The replacement of the 'whatever' syntax as sugar for consuming Channels
through 'react' locks very often (5-10% of the time). The thread seems to
be stuck on 'shift', probably from the Channel). When rewriting the exact
same functionality with the channel.poll syntax, the problem does not occur.

Attached you'll find 2 scripts​: loop_channelreact.p6 reproduces the
problem, while loop_channelpoll uses the old tried and tested approach.

More info at #perl6 (El_Che and also timotimo).

Regards,

C.

Tested on rakudo 2015.11 on Ubuntu 15.10 and Centos 7.0.
$ perl6 -v
This is rakudo version 2015.11-422-ge19741a built on MoarVM version
2015.11-34-gc3eea17 implementing Perl v6.b.
$ uname -a
Linux ubuntu 4.2.0-18-generic #​22-Ubuntu SMP Fri Nov 6 18​:25​:50 UTC 2015
x86_64 x86_64 x86_64 GNU/Linux

From pub.claudio@gmail.com

loop_channelpoll.p6

From pub.claudio@gmail.com

loop_channelreact.p6

From @lizmat

On 08 Dec 2015, at 14​:40, Claudio Ramirez (via RT) <perl6-bugs-followup@​perl.org> wrote​:

# New Ticket Created by Claudio Ramirez
# Please include the string​: [perl #​126842]
# in the subject line of all future correspondence about this issue.
# <URL​: https://rt-archive.perl.org/perl6/Ticket/Display.html?id=126842 >
Tested on rakudo 2015.11 on Ubuntu 15.10 and Centos 7.0.
$ perl6 -v
This is rakudo version 2015.11-422-ge19741a built on MoarVM version
2015.11-34-gc3eea17 implementing Perl v6.b.
$ uname -a
Linux ubuntu 4.2.0-18-generic #​22-Ubuntu SMP Fri Nov 6 18​:25​:50 UTC 2015
x86_64 x86_64 x86_64 GNU/Linux
<loop_channelpoll.p6><loop_channelreact.p6>

Some extra data-points on OS X​:

- I could not get the channel poll version to break (kept running all night, until at least 300K runs).

- the channel react version consistently breaks on the 77th iteration with a SIGABRT

- running the channel react version with MVM_SPESH_DISABLE=1, makes it consistently break on the 70th iteration with a SIGABRT

- running it with MVM_JIT_DISABLE=1 makes no difference from not specifying it​: it breaks on the 77th iteration with a SIGABRT

Changing the whenever {} to​:

  whenever $channel -> \val {
  print val;

makes it consistently fail on the 75th iteration

All in all, this leads me to suspect some kind of garbage collect issue.

And indeed, if I put an "nqp​::force_gc" just before the “sleep”, it does not seem to fail (still running)

Liz

The RT System itself - Status changed from 'new' to 'open'

From @jnthn

On Tue, 08 Dec 2015 05​:40​:29 -0800, pub.claudio@​gmail.com wrote​:

The replacement of the 'whatever' syntax as sugar for consuming Channels
through 'react' locks very often (5-10% of the time). The thread seems to
be stuck on 'shift', probably from the Channel). When rewriting the exact
same functionality with the channel.poll syntax, the problem does not occur.

Attached you'll find 2 scripts​: loop_channelreact.p6 reproduces the
problem, while loop_channelpoll uses the old tried and tested approach.

All kinds of things have been fixed that could have been to blame since this ticket was field; either way, it works out. I remove the sleep so it'd go faster and ran it for 250,000 iterations without trouble.

Added this example as a test in S17-supply/syntax.t to make sure it doesn't come up again.

@jnthn - Status changed from 'open' to 'resolved'