How are core defect reports handled?
137 views
Skip to first unread message
Kazutoshi Satoda
Jan 28, 2016, 11:12:38 AM1/28/16
to std-dis...@isocpp.org
I sent a defect report to core chair <wmm at edg.com> with subject
"Defect Report: What is said to be "atomic" or not, in definition of data race?"
at Fri, 22 Jan 2016 01:01:52 +0900 (a week ago), but there is no
response yet.
I followed this guide.
https://isocpp.org/std/submit-issue
I posted to std-discussion (this forum) first, and it seemed to be a
defect.
https://groups.google.com/a/isocpp.org/d/topic/std-discussion/o676rv3iX7Q/discussion
Now I'm afraid that I did something wrong. Please help if you know what
should I do in this situation.
Beside my problem, I wonder how are core defect reports handled now in
general.
In comp.std.c++ era, there was a FAQ describing this.
http://www.comeaucomputing.com/csc/faq.html#B14
It had some negotiation mechanisms including automatic reply from the
dedicated submission address.
In the current procedure, library submission address is dedicated one,
and live status is published.
https://github.com/cplusplus/LWG/
But core submission address seems not dedicated one (looks like a cause
of oversight), and I can't find public live status. (I know
<http://wg21.cmeerw.net/cwg/>, but it seems updated after the list was
published on www.open-std.org.)
Would someone please explain how are core defect reports handled now?
--
k_satoda
"Defect Report: What is said to be "atomic" or not, in definition of data race?"
at Fri, 22 Jan 2016 01:01:52 +0900 (a week ago), but there is no
response yet.
I followed this guide.
https://isocpp.org/std/submit-issue
I posted to std-discussion (this forum) first, and it seemed to be a
defect.
https://groups.google.com/a/isocpp.org/d/topic/std-discussion/o676rv3iX7Q/discussion
Now I'm afraid that I did something wrong. Please help if you know what
should I do in this situation.
Beside my problem, I wonder how are core defect reports handled now in
general.
In comp.std.c++ era, there was a FAQ describing this.
http://www.comeaucomputing.com/csc/faq.html#B14
It had some negotiation mechanisms including automatic reply from the
dedicated submission address.
In the current procedure, library submission address is dedicated one,
and live status is published.
https://github.com/cplusplus/LWG/
But core submission address seems not dedicated one (looks like a cause
of oversight), and I can't find public live status. (I know
<http://wg21.cmeerw.net/cwg/>, but it seems updated after the list was
published on www.open-std.org.)
Would someone please explain how are core defect reports handled now?
--
k_satoda
Daniel Krügler
Jan 28, 2016, 12:58:07 PM1/28/16
to std-dis...@isocpp.org
2016-01-28 17:12 GMT+01:00 Kazutoshi Satoda <k_sa...@f2.dion.ne.jp>:
> I sent a defect report to core chair <wmm at edg.com> with subject
> "Defect Report: What is said to be "atomic" or not, in definition of data race?"
> at Fri, 22 Jan 2016 01:01:52 +0900 (a week ago), but there is no
> response yet.
>
> I followed this guide.
> https://isocpp.org/std/submit-issue
> I posted to std-discussion (this forum) first, and it seemed to be a
> defect.
> https://groups.google.com/a/isocpp.org/d/topic/std-discussion/o676rv3iX7Q/discussion
>
> Now I'm afraid that I did something wrong. Please help if you know what
> should I do in this situation.
>
> Beside my problem, I wonder how are core defect reports handled now in
> general.
Sending an email to the owner of the CWG issue list as you described
> I sent a defect report to core chair <wmm at edg.com> with subject
> "Defect Report: What is said to be "atomic" or not, in definition of data race?"
> at Fri, 22 Jan 2016 01:01:52 +0900 (a week ago), but there is no
> response yet.
>
> I followed this guide.
> https://isocpp.org/std/submit-issue
> I posted to std-discussion (this forum) first, and it seemed to be a
> defect.
> https://groups.google.com/a/isocpp.org/d/topic/std-discussion/o676rv3iX7Q/discussion
>
> Now I'm afraid that I did something wrong. Please help if you know what
> should I do in this situation.
>
> Beside my problem, I wonder how are core defect reports handled now in
> general.
above looks exactly right to me.
> In comp.std.c++ era, there was a FAQ describing this.
> http://www.comeaucomputing.com/csc/faq.html#B14
> It had some negotiation mechanisms including automatic reply from the
> dedicated submission address.
>
> In the current procedure, library submission address is dedicated one,
> and live status is published.
> https://github.com/cplusplus/LWG/
>
> But core submission address seems not dedicated one (looks like a cause
> of oversight), and I can't find public live status. (I know
> <http://wg21.cmeerw.net/cwg/>, but it seems updated after the list was
> published on www.open-std.org.)
>
> Would someone please explain how are core defect reports handled now?
suggest to could send a second one at any time before the next mailing
deadline as specified in the News section of this page:
http://www.open-std.org/jtc1/sc22/wg21/
to double-check that your previous submission attempt had been
received. Mike might be on vacations, so I would recommend to send
your second one not before the begin of the next week, maybe not
before the begin of the week where the next deadline ends.
Thanks,
- Daniel
Robert Haberlach
Jan 28, 2016, 1:10:30 PM1/28/16
to std-dis...@isocpp.org
I filed a core issue on January 14 and didn't get any response yet. Vacation, perhaps?
Kazutoshi Satoda
Jan 30, 2016, 5:38:49 AM1/30/16
to std-dis...@isocpp.org
On 2016/01/29 2:58 +0900, Daniel Krügler wrote:
> I think you have done it correctly. If you didn't got any response, I
> suggest to could send a second one at any time before the next mailing
> deadline as specified in the News section of this page:
>
> http://www.open-std.org/jtc1/sc22/wg21/
>
> to double-check that your previous submission attempt had been
> received. Mike might be on vacations, so I would recommend to send
> your second one not before the begin of the next week, maybe not
> before the begin of the week where the next deadline ends.
Thank you for the advice. I'll resend my issue periodically until I get
> I think you have done it correctly. If you didn't got any response, I
> suggest to could send a second one at any time before the next mailing
> deadline as specified in the News section of this page:
>
> http://www.open-std.org/jtc1/sc22/wg21/
>
> to double-check that your previous submission attempt had been
> received. Mike might be on vacations, so I would recommend to send
> your second one not before the begin of the next week, maybe not
> before the begin of the week where the next deadline ends.
a response or see the issue in published list.
If this is the expected action on this situation, I think it should be
written in <https://isocpp.org/std/submit-issue>. I'll send a comment
from <https://isocpp.org/about/contact> about this.
I also hope that the procedure become more publicly visible, tolerant of
personal condition, and unified with the procedure for library issues.
But I don't know whom to ask for this.
--
k_satoda
Jens Maurer
Feb 1, 2016, 1:24:01 PM2/1/16
to std-dis...@isocpp.org
On 01/28/2016 07:10 PM, Robert Haberlach wrote:
> I filed a core issue on January 14 and didn't get any response yet. Vacation, perhaps?
Mike is usually processing the pending core issues in a batch
> I filed a core issue on January 14 and didn't get any response yet. Vacation, perhaps?
just before the next mailing. He rarely loses an issue,
so please be patient.
While posting to std-discussion won't give you the attention
of the maintainer of the core issues list, it might solicit some
replies; maybe you just misread or there's some other corner
of the standard that sufficiently clarifies your question.
Jens
Kazutoshi Satoda
Feb 19, 2016, 9:42:31 PM2/19/16
to std-dis...@isocpp.org
On 2016/01/30 19:38 +0900, Kazutoshi Satoda wrote:
> On 2016/01/29 2:58 +0900, Daniel Krügler wrote:
>> I think you have done it correctly. If you didn't got any response, I
>> suggest to could send a second one at any time before the next mailing
>> deadline as specified in the News section of this page:
>>
>> http://www.open-std.org/jtc1/sc22/wg21/
...
> On 2016/01/29 2:58 +0900, Daniel Krügler wrote:
>> I think you have done it correctly. If you didn't got any response, I
>> suggest to could send a second one at any time before the next mailing
>> deadline as specified in the News section of this page:
>>
>> http://www.open-std.org/jtc1/sc22/wg21/
> Thank you for the advice. I'll resend my issue periodically until I get
> a response or see the issue in published list.
I resent my report at 2016-02-08.
> a response or see the issue in published list.
Now the issue list Revision 96 (2016-02-15) is published. No issues have
been added since Revision 95. My report (and also Robert's) seems get
lost without any response so far.
I'll resend the report again.
> If this is the expected action on this situation, I think it should be
> written in <https://isocpp.org/std/submit-issue>. I'll send a comment
> from <https://isocpp.org/about/contact> about this.
> I also hope that the procedure become more publicly visible, tolerant of
> personal condition, and unified with the procedure for library issues.
> But I don't know whom to ask for this.
issue at 2016-02-12, but it also doesn't appear at
<https://cplusplus.github.io/LWG/>.
Could someone please explain what could happen (or what is happening),
and/or advise any other ways to proceed the issues?
--
k_satoda
Jens Maurer
Feb 20, 2016, 3:23:12 AM2/20/16
to std-dis...@isocpp.org
On 02/20/2016 03:42 AM, Kazutoshi Satoda wrote:
> On 2016/01/30 19:38 +0900, Kazutoshi Satoda wrote:
>> On 2016/01/29 2:58 +0900, Daniel Krügler wrote:
>>> I think you have done it correctly. If you didn't got any response, I
>>> suggest to could send a second one at any time before the next mailing
>>> deadline as specified in the News section of this page:
>>>
>>> http://www.open-std.org/jtc1/sc22/wg21/
> ...
>> Thank you for the advice. I'll resend my issue periodically until I get
>> a response or see the issue in published list.
>
> I resent my report at 2016-02-08.
>
> Now the issue list Revision 96 (2016-02-15) is published. No issues have
> been added since Revision 95. My report (and also Robert's) seems get
> lost without any response so far.
>
> I'll resend the report again.
That is very unlikely to happen. Mike will add the new issues
> On 2016/01/30 19:38 +0900, Kazutoshi Satoda wrote:
>> On 2016/01/29 2:58 +0900, Daniel Krügler wrote:
>>> I think you have done it correctly. If you didn't got any response, I
>>> suggest to could send a second one at any time before the next mailing
>>> deadline as specified in the News section of this page:
>>>
>>> http://www.open-std.org/jtc1/sc22/wg21/
> ...
>> Thank you for the advice. I'll resend my issue periodically until I get
>> a response or see the issue in published list.
>
> I resent my report at 2016-02-08.
>
> Now the issue list Revision 96 (2016-02-15) is published. No issues have
> been added since Revision 95. My report (and also Robert's) seems get
> lost without any response so far.
>
> I'll resend the report again.
to the issues list just prior to the Jacksonville meeting.
This update was for publishing the Ready and Tentatively Ready
issues as papers in the pre-meeting mailing.
Also, it is always a good idea to post a (potential) issue
in this forum first; maybe there is just some misreading
of the standard's text.
Jens
Kazutoshi Satoda
Feb 20, 2016, 4:44:22 AM2/20/16
to std-dis...@isocpp.org
On 2016/02/20 17:23 +0900, Jens Maurer wrote:
> On 02/20/2016 03:42 AM, Kazutoshi Satoda wrote:
...
> On 02/20/2016 03:42 AM, Kazutoshi Satoda wrote:
>> I resent my report at 2016-02-08.
>>
>> Now the issue list Revision 96 (2016-02-15) is published. No issues have
>> been added since Revision 95. My report (and also Robert's) seems get
>> lost without any response so far.
>>
>> I'll resend the report again.
>
> That is very unlikely to happen. Mike will add the new issues
> to the issues list just prior to the Jacksonville meeting.
> This update was for publishing the Ready and Tentatively Ready
> issues as papers in the pre-meeting mailing.
I hope so. But my doubt is still not cleared, since I have not get an
>>
>> Now the issue list Revision 96 (2016-02-15) is published. No issues have
>> been added since Revision 95. My report (and also Robert's) seems get
>> lost without any response so far.
>>
>> I'll resend the report again.
>
> That is very unlikely to happen. Mike will add the new issues
> to the issues list just prior to the Jacksonville meeting.
> This update was for publishing the Ready and Tentatively Ready
> issues as papers in the pre-meeting mailing.
ACK while I put explicit request in the second try: "I want an ACK when
you accepted this report to know I don't need sending more. An empty
reply is enough."
> Also, it is always a good idea to post a (potential) issue
> in this forum first; maybe there is just some misreading
> of the standard's text.
I sent this issue (in non-question form) to Mike:
Defect Report: What is said to be "atomic" or not, in definition of data race?
https://groups.google.com/a/isocpp.org/d/topic/std-discussion/o676rv3iX7Q/discussion
And this old issue (with updates against latest draft) to lwgchair:
Defect Report: Contradicting definition of empty shared_ptr on shared_ptr(nullptr, d)
http://compgroups.net/comp.std.c++/contradicting-definition-of-empty-shared_ptr-on/2126110
https://groups.google.com/forum/#!topic/comp.std.c++/4Cv99sQPjw4
(The latter URL sometimes work or don't work. Google groups seems
have some problems with forum name with '++'.)
--
k_satoda
Daniel Krügler
Feb 20, 2016, 5:07:42 AM2/20/16
to std-dis...@isocpp.org
2016-02-20 10:44 GMT+01:00 Kazutoshi Satoda <k_sa...@f2.dion.ne.jp>:
> I sent this issue (in non-question form) to Mike:
> Defect Report: What is said to be "atomic" or not, in definition of data race?
> https://groups.google.com/a/isocpp.org/d/topic/std-discussion/o676rv3iX7Q/discussion
This is core language stuff and I cannot tell you more than I already
> I sent this issue (in non-question form) to Mike:
> Defect Report: What is said to be "atomic" or not, in definition of data race?
> https://groups.google.com/a/isocpp.org/d/topic/std-discussion/o676rv3iX7Q/discussion
did before.
> And this old issue (with updates against latest draft) to lwgchair:
> Defect Report: Contradicting definition of empty shared_ptr on shared_ptr(nullptr, d)
> http://compgroups.net/comp.std.c++/contradicting-definition-of-empty-shared_ptr-on/2126110
> https://groups.google.com/forum/#!topic/comp.std.c++/4Cv99sQPjw4
> (The latter URL sometimes work or don't work. Google groups seems
> have some problems with forum name with '++'.)
procedure is explained here:
http://www.open-std.org/jtc1/sc22/wg21/docs/lwg-active.html#submit_issue
I searched for any incoming emails to the lwgchair address containing
the string "Contradicting definition of empty shared_ptr" but I didn't
find any.
Thanks,
- Daniel
Kazutoshi Satoda
Feb 20, 2016, 5:32:14 AM2/20/16
to std-dis...@isocpp.org
On 2016/02/20 19:07 +0900, Daniel Krügler wrote:
> 2016-02-20 10:44 GMT+01:00 Kazutoshi Satoda <k_sa...@f2.dion.ne.jp>:
...
> 2016-02-20 10:44 GMT+01:00 Kazutoshi Satoda <k_sa...@f2.dion.ne.jp>:
>> And this old issue (with updates against latest draft) to lwgchair:
>> Defect Report: Contradicting definition of empty shared_ptr on shared_ptr(nullptr, d)
>> http://compgroups.net/comp.std.c++/contradicting-definition-of-empty-shared_ptr-on/2126110
>> https://groups.google.com/forum/#!topic/comp.std.c++/4Cv99sQPjw4
>> (The latter URL sometimes work or don't work. Google groups seems
>> have some problems with forum name with '++'.)
>
> For library issues you need to contact the lwgchair address. The
> procedure is explained here:
>
> http://www.open-std.org/jtc1/sc22/wg21/docs/lwg-active.html#submit_issue
I want to see there how to know the report was received, and what
>> Defect Report: Contradicting definition of empty shared_ptr on shared_ptr(nullptr, d)
>> http://compgroups.net/comp.std.c++/contradicting-definition-of-empty-shared_ptr-on/2126110
>> https://groups.google.com/forum/#!topic/comp.std.c++/4Cv99sQPjw4
>> (The latter URL sometimes work or don't work. Google groups seems
>> have some problems with forum name with '++'.)
>
> For library issues you need to contact the lwgchair address. The
> procedure is explained here:
>
> http://www.open-std.org/jtc1/sc22/wg21/docs/lwg-active.html#submit_issue
should a reporter do if it seems not received. For example, "Watch
https://cplusplus.github.io/LWG and re-send the report if it doesn't
appear in a week."
> I searched for any incoming emails to the lwgchair address containing
> the string "Contradicting definition of empty shared_ptr" but I didn't
> find any.
I resent the report, with CC to you. Please check it again.
--
k_satoda
T. C.
Feb 21, 2016, 1:21:19 AM2/21/16
to ISO C++ Standard - Discussion
On Saturday, February 20, 2016 at 10:42:31 AM UTC+8, Kazutoshi SATODA wrote:
On 2016/01/30 19:38 +0900, Kazutoshi Satoda wrote:
> On 2016/01/29 2:58 +0900, Daniel Krügler wrote:
>> I think you have done it correctly. If you didn't got any response, I
>> suggest to could send a second one at any time before the next mailing
>> deadline as specified in the News section of this page:
>>
>> http://www.open-std.org/jtc1/sc22/wg21/
...
> Thank you for the advice. I'll resend my issue periodically until I get
> a response or see the issue in published list.
I resent my report at 2016-02-08.
Now the issue list Revision 96 (2016-02-15) is published. No issues have
been added since Revision 95. My report (and also Robert's) seems get
lost without any response so far.
Just speculating, but looking at the history for the last several revisions, it seems that new issues are added to the list in the post-meeting one (R92 - post-Urbana, R94 - post-Lenexa, R95 - post-Kona). Many of them were already in NAD/drafting status when added, suggesting that they were discussed at the meeting.
Kazutoshi Satoda
Feb 28, 2016, 5:15:38 AM2/28/16
to std-dis...@isocpp.org
On 2016/02/20 18:44 +0900, Kazutoshi Satoda wrote:
> On 2016/02/20 17:23 +0900, Jens Maurer wrote:
>> On 02/20/2016 03:42 AM, Kazutoshi Satoda wrote:
> ...
>>> I resent my report at 2016-02-08.
>>>
>>> Now the issue list Revision 96 (2016-02-15) is published. No issues have
>>> been added since Revision 95. My report (and also Robert's) seems get
>>> lost without any response so far.
>>>
>>> I'll resend the report again.
>>
>> That is very unlikely to happen. Mike will add the new issues
>> to the issues list just prior to the Jacksonville meeting.
>> This update was for publishing the Ready and Tentatively Ready
>> issues as papers in the pre-meeting mailing.
>
> I hope so. But my doubt is still not cleared, since I have not get an
> ACK while I put explicit request in the second try: "I want an ACK when
> you accepted this report to know I don't need sending more. An empty
> reply is enough."
I finally got ACK after the third attempt at 2016-02-26.
> On 2016/02/20 17:23 +0900, Jens Maurer wrote:
>> On 02/20/2016 03:42 AM, Kazutoshi Satoda wrote:
> ...
>>> I resent my report at 2016-02-08.
>>>
>>> Now the issue list Revision 96 (2016-02-15) is published. No issues have
>>> been added since Revision 95. My report (and also Robert's) seems get
>>> lost without any response so far.
>>>
>>> I'll resend the report again.
>>
>> That is very unlikely to happen. Mike will add the new issues
>> to the issues list just prior to the Jacksonville meeting.
>> This update was for publishing the Ready and Tentatively Ready
>> issues as papers in the pre-meeting mailing.
>
> I hope so. But my doubt is still not cleared, since I have not get an
> ACK while I put explicit request in the second try: "I want an ACK when
> you accepted this report to know I don't need sending more. An empty
> reply is enough."
I hope the procedure will be improved to avoid a single point of failure
(for example, with a dedicated address and multiple responders behind it
like LWG), and to have a way to quickly recover from such a failure (for
example, by showing roughly expected response time in the procedure).
--
k_satoda
Reply all
Reply to author
Forward
0 new messages