Am 21.03.10 23:14, schrieb Thomas Keller:
> Am 21.03.10 21:19, schrieb Timothy Brownawell:
>> Thomas Keller wrote:
>>> #
>>> # old_revision [dda54058811fd39356e8755ad0d68998a5159dd3]
>>> #
>>> # patch "network/reactor.cc"
>>> # from [a90cee5442d3c77d8082d7bb91d4d23170bf25fb]
>>> # to [4f424db0f326e61b015ff042880f2d77589c6429]
>>> #
>>> ============================================================
>>> --- network/reactor.cc a90cee5442d3c77d8082d7bb91d4d23170bf25fb
>>> +++ network/reactor.cc 4f424db0f326e61b015ff042880f2d77589c6429
>>> @@ -87,8 +87,9 @@ void reactor::remove(shared_ptr<reactabl
>>> if (i != items.end())
>>> {
>>> items.erase(i);
>>> + bool had_pipe = have_pipe;
>>> have_pipe = false;
>>> - if (readying)
>>> + if (readying && !had_pipe)
>>> item->remove_from_probe(probe);
>>> }
>>> }
>>>
>>> @Timothy: I was a bit careful about setting have_pipe after `if
>>> (readying) ...` here - if you think this would not introduce a problem,
>>> then this patch could be simplyfied of course.
>>
>> The simpler way should be fine, neither version of remove_from_probe
>> cares about the reactor's internal state.
>
> Ok, the simplified version has been committed in dbb536a64.After checking the output with the above patch again more closely I've seen that the error is still present, just that it no longer leads to a fatal error with the patch. I need to examine this further and have disapproved the above revision until then. Thomas. -- GPG-Key 0x160D1092 | tommyd3...@jabber.ccc.de | http://thomaskeller.biz Please note that according to the EU law on data retention, information on every electronic information exchange might be retained for a period of six months or longer: http://www.vorratsdatenspeicherung.de/?lang=en
signature.asc
Description: OpenPGP digital signature
