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!aioe.org!.POSTED.2uCIJahv+a4XEBqttj5Vkw.user.gioia.aioe.org!not-for-mail From: "Dmitry A. Kazakov" Newsgroups: comp.lang.ada Subject: Re: is there a version of unix written in Ada Date: Thu, 1 Oct 2020 17:44:52 +0200 Organization: Aioe.org NNTP Server Message-ID: References: <00cd3aaa-d518-43a2-b321-58d6fae70aebo@googlegroups.com> <57eb7a65-51ea-4624-b9dc-9c4dda0fee59n@googlegroups.com> <5f70fd3b$0$13541$426a74cc@news.free.fr> <87wo0d3iac.fsf@nightsong.com> <87sgb02l7b.fsf@nightsong.com> <875z7vyy1u.fsf@nightsong.com> <87wo0bkns3.fsf@nightsong.com> NNTP-Posting-Host: 2uCIJahv+a4XEBqttj5Vkw.user.gioia.aioe.org Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit X-Complaints-To: abuse@aioe.org User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:78.0) Gecko/20100101 Thunderbird/78.3.1 X-Notice: Filtered by postfilter v. 0.9.2 Content-Language: en-US Xref: reader02.eternal-september.org comp.lang.ada:60365 List-Id: On 01/10/2020 16:18, J-P. Rosen wrote: > Le 01/10/2020 à 14:51, Dmitry A. Kazakov a écrit : > >> My mental picture was protected procedure calls executed concurrently on >> different cores of a multi-core processor. Would that sort of >> implementation be legal? > No. Protected objects guarantee that only one task at a time can be > inside (ignoring functions). Multi-cores don't come into play. Inside one protected object, or inside any protected object? Is it effectively one single protected object and all protected object are its facets. >> If so, then let there be protected procedure P1 of the object O1 and P2 >> of O2. If P1 and P2 call to P3 of O3 that would be a problem. Ergo ether >> wandering or concurrent protected protected calls must be illegal. > But it's not the case... The scenario is: Task1 calls P1 on O1. Task2 calls P2 on O2. Both P1 and P2 call Protected_Trace on the protected Tracer object. If tasks occupy two different cores, would/should one suspend (in order not use the reserved word "block") another? >>> 1. A LIFO queue of log entries implemented in a protected object of >>> highest priority. The object has a procedure "Write_Log_Entry". >> >> Yes, that was what I thought and what Debug.adb does. However Debug.adb >> allocates the body of the FIFO element in the pool. I would rather use >> my implementation of indefinite FIFO which does not use pools. I don't >> want allocators/deallocators inside protected stuff. > As surprising as it may seem, allocators/deallocators are NOT > potentially blocking operations. But I understand your concerns... And this raises the same question. Pool must be interlocked, but not block. What is the semantics of this "non-blocking" interlocking on a multi-core machine? -- Regards, Dmitry A. Kazakov http://www.dmitry-kazakov.de