Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!eternal-september.org!.POSTED!not-for-mail
From: BGB <cr88192@gmail.com>
Newsgroups: comp.arch
Subject: Re: Is Parallel Programming Hard, And, If So, What Can You Do About
 It?
Date: Tue, 20 May 2025 20:08:28 -0500
Organization: A noiseless patient Spider
Lines: 18
Message-ID: <100j9e7$2gdpc$1@dont-email.me>
References: <vvnds6$3gism$1@dont-email.me>
 <edb59b7854474033c748f0fd668badaa@www.novabbs.org>
 <w32UP.481123$C51b.217868@fx17.iad> <vvqdas$g9oh$1@dont-email.me>
 <vvrcs9$msmc$2@dont-email.me>
 <0ec5d195f4732e6c92da77b7e2fa986d@www.novabbs.org>
 <vvribg$npn4$1@dont-email.me> <vvs343$ulkk$1@dont-email.me>
 <vvtt4d$1b8s7$4@dont-email.me> <2025May13.094035@mips.complang.tuwien.ac.at>
 <vvuuua$1mt7m$1@dont-email.me> <vvvons$3uvs3$2@dont-email.me>
 <1000nfp$2440u$1@dont-email.me> <1000pae$3uvs3$3@dont-email.me>
 <100bdhq$lhdb$3@dont-email.me>
 <91c8a31fc5d04a1fadf210b2dd6d4875@www.novabbs.org>
 <100e0it$19264$1@dont-email.me>
 <fa7e33d953bc6f545387d862e19c2bd2@www.novabbs.org>
 <100gipr$1sbnn$10@dont-email.me> <100h3jm$23ehu$1@dont-email.me>
 <jwv7c2bjqcx.fsf-monnier+comp.arch@gnu.org> <100j6pq$2fqhj$9@dont-email.me>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Injection-Date: Wed, 21 May 2025 03:14:47 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="c0e131e10a4d7b51b230e184d42b6367";
	logging-data="2635564"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX185MypiLxTeo1fmonLKNz/3kQH5GU3GLFU="
User-Agent: Mozilla Thunderbird
Cancel-Lock: sha1:otnYLdSE/yt0DgzqtVsxU/2hcXU=
In-Reply-To: <100j6pq$2fqhj$9@dont-email.me>
Content-Language: en-US
Bytes: 2666

On 5/20/2025 7:29 PM, Lawrence D'Oliveiro wrote:
> On Tue, 20 May 2025 10:49:54 -0400, Stefan Monnier wrote:
> 
>> Indeed, AFAIK, what we call "HDD cache" is actually just the RAM used by
>> the embedded CPU inside the drive for its operation.
> 
> If it were just I/O buffers for operations in progress, that would be
> fine. The problem is when it keeps data around instead of immediately
> writing it out, and what’s worse, lies about it, so it tells the OS that
> the write has completed when it hasn’t.

Note that (with SATA and similar) the OS can request that the drive 
flush its caches, and (in theory) drive should not respond to more 
requests until everything has been fully written back to disk.

Though, admittedly, I don't have much first-hand experience with 
interfacing directly with SATA drives.