From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.5-pre1 (2020-06-20) on
	ip-172-31-74-118.ec2.internal
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=3.0 tests=BAYES_00 autolearn=ham
	autolearn_force=no version=3.4.5-pre1
Path: eternal-september.org!reader02.eternal-september.org!feeder.eternal-september.org!news.swapon.de!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail
From: Niklas Holsti <niklas.holsti@tidorum.invalid>
Newsgroups: comp.lang.ada
Subject: Re: is there a version of unix written in Ada
Date: Thu, 1 Oct 2020 14:38:27 +0300
Organization: Tidorum Ltd
Message-ID: <htltdlFel31U1@mid.individual.net>
References: <e3c51e65-b71e-4623-bd58-c94cdbcc3ba6o@googlegroups.com>
 <00cd3aaa-d518-43a2-b321-58d6fae70aebo@googlegroups.com>
 <db65f550-c1d9-4ddb-b128-7e4e9c1f818bn@googlegroups.com>
 <57eb7a65-51ea-4624-b9dc-9c4dda0fee59n@googlegroups.com>
 <rkq9gr$s0u$1@gioia.aioe.org> <5f70fd3b$0$13541$426a74cc@news.free.fr>
 <rks433$ndr$1@gioia.aioe.org>
 <daa04c57-0dfa-4c95-84ef-0e90f3c1732cn@googlegroups.com>
 <rkst48$171u$1@gioia.aioe.org> <87wo0d3iac.fsf@nightsong.com>
 <rkth03$uk5$1@gioia.aioe.org> <87sgb02l7b.fsf@nightsong.com>
 <rl1f1i$le8$1@gioia.aioe.org> <875z7vyy1u.fsf@nightsong.com>
 <rl2n2q$ddt$1@gioia.aioe.org> <87wo0bkns3.fsf@nightsong.com>
 <rl2rqt$jdq$1@gioia.aioe.org> <rl31jv$cue$1@franka.jacob-sparre.dk>
 <rl47f9$1nn9$2@gioia.aioe.org> <rl49ar$q51$1@dont-email.me>
 <rl4ajp$1d9d$1@gioia.aioe.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
X-Trace: individual.net dkSyTJ4I30UDJl8YduGA2QRd+MzCRbd0egU/0FIe/YaKK9VnPe
Cancel-Lock: sha1:YegihVr2e14o0m3xDMDCIzIRIWg=
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.14; rv:68.0)
 Gecko/20100101 Thunderbird/68.10.0
In-Reply-To: <rl4ajp$1d9d$1@gioia.aioe.org>
Content-Language: en-US
Xref: reader02.eternal-september.org comp.lang.ada:60357
List-Id: <comp.lang.ada>

On 2020-10-01 13:21, Dmitry A. Kazakov wrote:
> On 01/10/2020 11:59, J-P. Rosen wrote:
>> Le 01/10/2020 à 11:28, Dmitry A. Kazakov a écrit :
>>> BTW, I still do not know to design an Ada-conform tracing/logging
>>> facility such that you could trace/log from anywhere, protected action
>>> included, and without knowing statically which protected object is
>>> involved.
>>>
>> Did you have a look at package Debug?
>> (https://www.adalog.fr/en/components#Debug)
> 
> Thanks
> 
>> It features, among others, a trace routine which is guaranteed to not be
>> potentially blocking.
> 
> It calls a protected operation on a different protected object, yes, 
> this is non-blocking, and I considered same, but is this legal?


Yes.

If the program is using ceiling-priority-based protection, the priority 
of the calling object must be less or equal to the priority of the 
called object.


> Or is that limited to protected entries only?


An entry call is potentially blocking and therefore not allowed in a 
protected operation.


> Another issue is having two different calls: Trace and protected Trace. 
> If one used instead of another, you have a ticking bomb in the 
> production code.


I assume that is a "feature" of the referenced Debug package, not of the 
basic method it uses to implement a logging facility.

I haven't looked at the Debug package, but I would have suggested a 
logging facility that consists of:

1. A LIFO queue of log entries implemented in a protected object of 
highest priority. The object has a procedure "Write_Log_Entry".

2. A task that empties the LIFO queue into a log file. The task calls an 
entry of the LIFO protected object to get a log entry from the queue, 
but executes the file-writing operations in task context, not in a 
protected operation.