Deutsch English Français Italiano |
<vbb30g$5q1f$1@dont-email.me> View for Bookmarking (what is this?) Look up another Usenet article |
Path: ...!news.mixmin.net!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail From: "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> Newsgroups: comp.arch Subject: Re: arm ldxr/stxr vs cas Date: Wed, 4 Sep 2024 18:54:23 -0700 Organization: A noiseless patient Spider Lines: 29 Message-ID: <vbb30g$5q1f$1@dont-email.me> References: <vb4sit$2u7e2$1@dont-email.me> <07d60bd0a63b903820013ae60792fb7a@www.novabbs.org> <vbarl0$13lk$1@dont-email.me> <b4db68f6d3ab76eb6645905a7108eb75@www.novabbs.org> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit Injection-Date: Thu, 05 Sep 2024 03:54:24 +0200 (CEST) Injection-Info: dont-email.me; posting-host="22bbe074495a5ed8376623fc400f655d"; logging-data="190511"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX1+ffhC1yRwtOCdw/Oxy4nhXF0Gp33BwHu4=" User-Agent: Mozilla Thunderbird Cancel-Lock: sha1:aibdBb1gVoQTwMWqQp64qYy9uKo= In-Reply-To: <b4db68f6d3ab76eb6645905a7108eb75@www.novabbs.org> Content-Language: en-US Bytes: 2454 On 9/4/2024 5:59 PM, MitchAlsup1 wrote: > On Wed, 4 Sep 2024 23:48:48 +0000, Chris M. Thomasson wrote: > >> On 9/4/2024 2:27 PM, MitchAlsup1 wrote: >>> On Mon, 2 Sep 2024 17:27:57 +0000, jseigh wrote: >>> >>>> I read that arm added the cas instruction because they didn't think >>>> ldxr/stxr would scale well. It wasn't clear to me as to why that >>>> would be the case. I would think the memory lock mechanism would >>>> have really low overhead vs cas having to do an interlocked load >>>> and store. Unless maybe the memory lock size might be large >>>> enough to cause false sharing issues. Any ideas? >>> >>> A pipeline lock between the LD part of a CAS and the ST part of a >>> CAS is essentially FREE. But the same is true for LL followed by >>> a later SC. >> >> 100% sure on that? No way to break the reservation from an unrelated >> aspect wrt LL/SC? > > I have been building pipelines like that since 1983. > > Now, it is completely possible to build a pipeline that gives > one grief (lesser or greater) in doing these things--but it is > definitely possible to build a grief free pipeline for LL- SC, > and by extension CAS. Well, the software must play along as well? Padded and aligned on certain boundaries helps things out. Big time.