From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-qk1-f172.google.com (mail-qk1-f172.google.com [209.85.222.172]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 4F669192B75; Wed, 2 Apr 2025 16:29:25 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.222.172 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1743611367; cv=none; b=ilxYQ9x+EIoOII1tpQrGIwHtwc5n4q9TmslWmQnZO2uwN1KzO+7+SzY5XoPhW4Cn6p5Tkn2MdbA8TkWVS20PqyhdPUeGYieHeXOZ83hvsuqiPQX50OrEd+5f3FZ5uGbpLFeknj8Bl/UOZDbf6wz3bs1nzrWkKZZ9CgNBP2DA7LU= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1743611367; c=relaxed/simple; bh=hpNglcGCD/ImR4MbnHVfaGtof3Zs/Gn5mxoDcxHVy5M=; h=Date:From:To:Cc:Subject:Message-ID:References:MIME-Version: Content-Type:Content-Disposition:In-Reply-To; b=DMTLhK2IkGubVE2Cw5iB9YB1a1nZrTfnHXEpIptXds39VsK7YwreRNLqXf7TwzdhfbQBxSHuOPs2c2GIginwaH2SjWK01E1dwgd++Isd5CtWSUhGXdok5iQVucM3kxEfGiAIri1GlftO/pJRZgFNL41u941WBp8OqlmKzvUmD5M= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=EKNP9K5O; arc=none smtp.client-ip=209.85.222.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="EKNP9K5O" Received: by mail-qk1-f172.google.com with SMTP id af79cd13be357-7c546334bdeso513844785a.2; Wed, 02 Apr 2025 09:29:25 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1743611364; x=1744216164; darn=vger.kernel.org; h=in-reply-to:content-disposition:mime-version:references:message-id :subject:cc:to:from:date:feedback-id:from:to:cc:subject:date :message-id:reply-to; bh=FNud7VaCUDeA+FyZPSt2bHJdSWzIBv4EDqJHrTjrcSI=; b=EKNP9K5O2o3ZWV1AtLjwedC37+zomjYHxZqt/6WfqY7fLtfHyZ0RPqMO5zVBUGFOal JZBegA5LDj5SLnzbCa42TU7igPjUbalYoVNQAYQbdrdGouussVRaFBi7TPr9DGWCBayi i9fQoJwKb4cJh++P5lOOdP6p2og1LHs0ro4KMRfKNxtEtiSkGRcpbuw+4xN4geX/0Sza WUO3B95LvvOEK34bhHP1zKJ+WmKSCjkNP8gAUNQ0+hLICTLExIgLRpnCQYj91ooelrpm vBoxg3WEEM/6g/y13IiydpQcyMOPJJyGI/FhpzmiL09FfI6pGD7ylgaqSLFMNz/zZxvb wv1g== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1743611364; x=1744216164; h=in-reply-to:content-disposition:mime-version:references:message-id :subject:cc:to:from:date:feedback-id:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=FNud7VaCUDeA+FyZPSt2bHJdSWzIBv4EDqJHrTjrcSI=; b=kD+VSykTy1piL9jmLa38D2DPe/EAl2lMwPUyO5JhskZ6jXBwmq8GUdHEEO7cbwHs0y yjwbeRzyke31q8XLgyUlQVm/N7btlUGlpIt+0qu0RS0wOnO0NWUq4FzqkVsAk84EpQDY ri/fJqZeqnJ6QYV74bFad3XghQGMtfME+Ewa4Riw2/UgHzx4SUKhszfwuoAAWBrEv5JD tXb+1dkXdocyFrfy1BgGx05YYB5S+zbMkPzN9uMyjZVfDKEIjSlZ34cEyVWr0ipFsy+2 nOFNX75lAtghdZZAyvto6HLDySPHF8TQqWb/xutVburXiydCKwGjasnisLdGCCuVPk3a XSkA== X-Forwarded-Encrypted: i=1; AJvYcCV9aXEUyv7GLYPuoc+Aa4sP9jF0q/OUnf23Knk+o5pmV0ot1PH+paAIgwYlHykHTpDTIWkmH1yvt27EQs4=@vger.kernel.org, AJvYcCVjy6VaHiWWN1pnIGSxY0ECR0JawoBfdP2xcm/ESFw0AvEig+qZErVIvQDvx7NybzqvfHBR/ePOuUhzhAdBuXc=@vger.kernel.org, AJvYcCWMyPc40PGAtT/FmZCkv4RaY4dwNrdi3b0j/Fmr3wpzFbXihI8GYJWuqIk103lzSNW9G/0ViTp3@vger.kernel.org X-Gm-Message-State: AOJu0Yzk+XMBIgeDjEReFYHwjpIXfFnWFm2aVT63SAMikI3EHAPcz6lQ 9XEXyNDt+n2DMAjgConlU3zh+jVWb44E7iJ/ptxZxi2ymW9AM3vk X-Gm-Gg: ASbGnctud2EjVlZ4tN/0IBJrM5Tx8pGgTYYXdCFDGRt5Pp9/0jx+sIDgA9GoNMcVzyB ggTSMKMDcVFwNI9/YxU0Mmrgs2lWbZI8v1C64INECsRyLcbEaboSYGijePdvydgU/8551Ya4dET N0s5T0f/K9mNWzRfVfY3XOJbWewc72N6//RZJgTAfmTjNfJC1onuddvCCJObjRBQUKbgq6LRGdj xyzdZtuULN/idJK9WGaKzzu+3bjS9u3OegLOyt8kzQWT30P0CD/5ZhH3yqj8gZxPyMiX9NE3rE/ NjE2g2R7W3Aiyn88XoyaIdvmrDI+58u/LnYsWx/HFSZSCh54vOrg4bPvXoDor3HxUjlbtqMG7Fo Q+CwEuU1vjNNXbweuzItgmTVKQgy36ZPMB4M= X-Google-Smtp-Source: AGHT+IEp0HORSr1/7mlmiTWemxCFPiCAqmQppHiDsaiqEfJoplVhG4xWytAzoUDxpOw1aGDKv6oBhg== X-Received: by 2002:a05:620a:c51:b0:7c5:b90a:54a6 with SMTP id af79cd13be357-7c6865ecef4mr2469143785a.13.1743611364038; Wed, 02 Apr 2025 09:29:24 -0700 (PDT) Received: from fauth-a2-smtp.messagingengine.com (fauth-a2-smtp.messagingengine.com. [103.168.172.201]) by smtp.gmail.com with ESMTPSA id af79cd13be357-7c5f7682e57sm805481085a.39.2025.04.02.09.29.23 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 02 Apr 2025 09:29:23 -0700 (PDT) Received: from phl-compute-11.internal (phl-compute-11.phl.internal [10.202.2.51]) by mailfauth.phl.internal (Postfix) with ESMTP id AA1F41200043; Wed, 2 Apr 2025 12:29:22 -0400 (EDT) Received: from phl-mailfrontend-01 ([10.202.2.162]) by phl-compute-11.internal (MEProxy); Wed, 02 Apr 2025 12:29:22 -0400 X-ME-Sender: X-ME-Received: X-ME-Proxy-Cause: gggruggvucftvghtrhhoucdtuddrgeefvddrtddtgddukeeiudegucetufdoteggodetrf dotffvucfrrhhofhhilhgvmecuhfgrshhtofgrihhlpdggtfgfnhhsuhgsshgtrhhisggv pdfurfetoffkrfgpnffqhgenuceurghilhhouhhtmecufedttdenucesvcftvggtihhpih gvnhhtshculddquddttddmnecujfgurhepfffhvfevuffkfhggtggujgesthdtredttddt vdenucfhrhhomhepuehoqhhunhcuhfgvnhhguceosghoqhhunhdrfhgvnhhgsehgmhgrih hlrdgtohhmqeenucggtffrrghtthgvrhhnpedvgeeifeehtddtffehffffvdfgjedtieek heefgedtteeluddvheefjefgffejteenucffohhmrghinheprhhushhtqdhfohhrqdhlih hnuhigrdgtohhmpdhgihhthhhusgdrtghomhdpudeirdihohhunecuvehluhhsthgvrhfu ihiivgeptdenucfrrghrrghmpehmrghilhhfrhhomhepsghoqhhunhdomhgvshhmthhprg huthhhphgvrhhsohhnrghlihhthidqieelvdeghedtieegqddujeejkeehheehvddqsgho qhhunhdrfhgvnhhgpeepghhmrghilhdrtghomhesfhhigihmvgdrnhgrmhgvpdhnsggprh gtphhtthhopeefgedpmhhouggvpehsmhhtphhouhhtpdhrtghpthhtohepfhhujhhithgr rdhtohhmohhnohhrihesghhmrghilhdrtghomhdprhgtphhtthhopegrrdhhihhnuggsoh hrgheskhgvrhhnvghlrdhorhhgpdhrtghpthhtohepthhglhigsehlihhnuhhtrhhonhhi gidruggvpdhrtghpthhtoheplhhinhhugidqkhgvrhhnvghlsehvghgvrhdrkhgvrhhnvg hlrdhorhhgpdhrtghpthhtoheprhhushhtqdhfohhrqdhlihhnuhigsehvghgvrhdrkhgv rhhnvghlrdhorhhgpdhrtghpthhtohepnhgvthguvghvsehvghgvrhdrkhgvrhhnvghlrd horhhgpdhrtghpthhtoheprghnughrvgifsehluhhnnhdrtghhpdhrtghpthhtohephhhk rghllhifvghithdusehgmhgrihhlrdgtohhmpdhrtghpthhtohepthhmghhrohhsshesuh hmihgthhdrvgguuh X-ME-Proxy: Feedback-ID: iad51458e:Fastmail Received: by mail.messagingengine.com (Postfix) with ESMTPA; Wed, 2 Apr 2025 12:29:22 -0400 (EDT) Date: Wed, 2 Apr 2025 09:29:18 -0700 From: Boqun Feng To: FUJITA Tomonori Cc: a.hindborg@kernel.org, tglx@linutronix.de, linux-kernel@vger.kernel.org, rust-for-linux@vger.kernel.org, netdev@vger.kernel.org, andrew@lunn.ch, hkallweit1@gmail.com, tmgross@umich.edu, ojeda@kernel.org, alex.gaynor@gmail.com, gary@garyguo.net, bjorn3_gh@protonmail.com, benno.lossin@proton.me, a.hindborg@samsung.com, aliceryhl@google.com, anna-maria@linutronix.de, frederic@kernel.org, arnd@arndb.de, jstultz@google.com, sboyd@kernel.org, mingo@redhat.com, peterz@infradead.org, juri.lelli@redhat.com, vincent.guittot@linaro.org, dietmar.eggemann@arm.com, rostedt@goodmis.org, bsegall@google.com, mgorman@suse.de, vschneid@redhat.com, tgunders@redhat.com, me@kloenk.dev, david.laight.linux@gmail.com Subject: Re: [PATCH v11 6/8] MAINTAINERS: rust: Add new sections for DELAY/SLEEP and TIMEKEEPING API Message-ID: References: <871puoelnj.fsf@kernel.org> <20250402.231627.270393242231849699.fujita.tomonori@gmail.com> Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20250402.231627.270393242231849699.fujita.tomonori@gmail.com> On Wed, Apr 02, 2025 at 11:16:27PM +0900, FUJITA Tomonori wrote: > On Mon, 31 Mar 2025 07:03:15 -0700 > Boqun Feng wrote: > > >> My recommendation would be to take all of `rust/kernel/time` under one > >> entry for now. I suggest the following, folding in the hrtimer entry as > >> well: > >> > >> DELAY, SLEEP, TIMEKEEPING, TIMERS [RUST] > >> M: Andreas Hindborg > > > > Given you're the one who would handle the patches, I think this make > > more sense. > > > >> R: Boqun Feng > >> R: FUJITA Tomonori > > > > Tomo, does this look good to you? > > Fine by me. > > So a single entry for all the Rust time stuff, which isn't aligned > with C's MAINTAINERS entries. It's just for now? > Given Andreas is the one who's going to handle the PRs, and he will put all the things in one branch. I think it's fine even for long term, and we got all relevant reviewers covered. If the Rust timekeeping + hrtimer community expands in the future, we can also add more entries. We don't necessarily need to copy all maintainer structures from C ;-) > > >> R: Lyude Paul > >> R: Frederic Weisbecker > >> R: Thomas Gleixner > >> R: Anna-Maria Behnsen > >> R: John Stultz > > > > We should add: > > > > R: Stephen Boyd > > > > If Stephen is not against it. > > > >> L: rust-for-linux@vger.kernel.org > >> S: Supported > >> W: https://rust-for-linux.com > >> B: https://github.com/Rust-for-Linux/linux/issues > >> T: git https://github.com/Rust-for-Linux/linux.git rust-timekeeping-next > >> F: rust/kernel/time.rs > >> F: rust/kernel/time/ > >> > >> If that is acceptable to everyone, it is very likely that I can pick 2-6 > >> for v6.16. > >> > > > > You will need to fix something because patch 2-6 removes `Ktime` ;-) > > I'll take care of it in the next version. > Thanks! > >> I assume patch 1 will go through the sched/core tree, and then Miguel > >> can pick 7. > >> > > > > Patch 1 & 7 probably should go together, but we can decide it later. > > Since nothing has moved forward for quite a while, maybe it's time to > drop patch 1. No, I think we should keep it. Because otherwise we will use a macro version of read_poll_timeout(), which is strictly worse. I'm happy to collect patch #1 and the cpu_relax() patch of patch #7, and send an PR to tip. Could you split them a bit: * Move the Rust might_sleep() in patch #7 to patch #1 and put it at kernel::task, also if we EXPORT_SYMBOL(__might_sleep_precision), we don't need the rust_helper for it. * Have a separate containing the cpu_relax() bit. * Also you may want to put #[inline] at cpu_relax() and might_resched(). and we can start from there. Sounds good? Regards, Boqun