From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from out-189.mta0.migadu.com (out-189.mta0.migadu.com [91.218.175.189]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 344883DB313 for ; Tue, 14 Jul 2026 06:41:17 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=91.218.175.189 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1784011279; cv=none; b=Gfq4Wt/I1mmIYtzeDJlYum0RbXARloQBMmbgzeF/0MRi9bKJPsVAsIjzhUdRzb5qQUBoKmnk1wvdJ3s1fQDE7IsfIq3V5nyXxer2FOqaAzIZ2IiOf8a+ZzHF+huX7MqTQ9+jZPdYSNs8edZHPfwz2vDvCaE5q3gXvhwFUwIno0c= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1784011279; c=relaxed/simple; bh=bkwP5nYboJKLZttgwm6FIm5eYW0GGZtVCMUIDOwO6xA=; h=Content-Type:Mime-Version:Subject:From:In-Reply-To:Date:Cc: Message-Id:References:To; b=HuZvuYWfmXkAFVSlFmT2iW4aYZbuRQDywQYYNB9LXKEImqJJJ2TrW66tjSp8Z+AEAh30sXW+U0Fg4uyfZsotrc/jI/DAda/tK1Ik+ylwn7HUCn4eJkIfLj0QLzWmcMpikBXzbxOpygc/GbJRx7Dv0ZIYFGhif/ovTxHgguxArKQ= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev; spf=pass smtp.mailfrom=linux.dev; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b=rBDobzER; arc=none smtp.client-ip=91.218.175.189 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.dev Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b="rBDobzER" Content-Type: text/plain; charset=utf-8 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.dev; s=key1; t=1784011275; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=bkwP5nYboJKLZttgwm6FIm5eYW0GGZtVCMUIDOwO6xA=; b=rBDobzERfqXt13rpR2Ehjlzpy/zktyZ+Mb6QOJQDJBtt6ms9AYnO7aIshqLDqT5J+NwgWC 3qPNcYTTHDQEy0pCcCO5nzsv+JlejATb8F+VXwXE6EsEITMDWbGISMOPDzwbCjAfLYSowR R7wHhb82rl9HP//fF83TguXApJ1SsFQ= Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3864.600.51.1.1\)) Subject: Re: [PATCH 00/18] Another attempt at HVO support on arm64 X-Report-Abuse: Please report any abuse attempt to abuse@migadu.com and include these headers. From: Muchun Song In-Reply-To: Date: Tue, 14 Jul 2026 14:40:24 +0800 Cc: Will Deacon , Catalin Marinas , Oscar Salvador , Nikos Nikoleris , Linu Cherian , Mark Rutland , David Hildenbrand , Andrew Morton , Ryan Roberts , Nanyong Sun , Yu Zhao , Frank van der Linden , David Rientjes , linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-mm@kvack.org Content-Transfer-Encoding: quoted-printable Message-Id: References: <20260708031129.3503195-1-jthoughton@google.com> To: James Houghton X-Migadu-Flow: FLOW_OUT > On Jul 13, 2026, at 12:09, James Houghton = wrote: >=20 > On Thu, Jul 9, 2026 at 8:41=E2=80=AFPM Muchun Song = wrote: >>=20 >>> On Jul 10, 2026, at 03:04, James Houghton = wrote: >>>=20 >>> On Thu, Jul 9, 2026 at 2:55=E2=80=AFAM Muchun Song = wrote: >>>>> On Jul 9, 2026, at 00:49, James Houghton = wrote: >>>>>=20 >>>>> On Wed, Jul 8, 2026 at 1:41=E2=80=AFAM Muchun Song = wrote: >>>>>> Do you mean that the support for AF might vary across different = CPUs? >>>>>> I'm not that familiar with arm64, so it seems a bit strange to me = that >>>>>> such basic hardware features can differ so much from one CPU to = another. >>>>>=20 >>>>> Yes, hardware updates of the Access Flag is a per-CPU feature. It = is >>>>> available for a CPU to use if TCR_EL1.HA is set. TCR_EL1 is a = system >>>>> register; each CPU has its own. (Linux will always enable HW AF = for a >>>>> CPU when it is onlined[1] if support is advertised, so we simply = need >>>>> to check if support is advertised to know that it is in fact = enabled.) >>>>>=20 >>>>> These days it is not uncommon for a system to have two (or more?) >>>>> different core implementations, like with "fast" cores and = "efficient" >>>>> cores. >>>>>=20 >>>>> [1] See the CONFIG_ARM64_HW_AFDBM bits in arch/arm64/mm/proc.S >>>>=20 >>>> Thanks for your detailed explanation. When enabling HVO via the = cmdline, >>>> can we simply prevent CPUs that do not support AF from coming = online? >>>> Would implementing it this way be much simpler? In practice, = developers >>>> definitely know whether the current system is suitable for enabling = HVO. >>>> If some CPUs do not support AF, they would just need to evaluate = the >>>> trade-off between memory savings and having fewer online CPUs than = expected. >>>>=20 >>>> For scenarios where HVO is enabled via sysctl, we simply need to = check >>>> if all CPUs support AF. If any do not, the system should return an = error. >>>>=20 >>>> Then, we can proceed with the Pre-HVO. >>>=20 >>> I don't think it makes sense to try to implement pre-HVO. >>>=20 >>> We cannot do HVO if any boot CPUs do not support HW AF, as HW AF = will >>> be required to free the HugeTLB pages later, which we should = continue >>> to support. Pre-HVO (today anyway) happens before all boot CPUs are >>> onlined. IMO it is not okay to prevent boot CPUs from onlining. >>=20 >> When the boot CPU starts, it first checks if the hardware (the = current >> boot CPU) supports AF (The selection of the Boot CPU is critical here >> If the user really wants to enable HVO). If AF is supported and HVO = is >> enabled via cmdline, we perform a Pre-HVO. Subsequently, any CPUs = that >> do not support AF will be prevented from coming online. >=20 > I understand. I don't think we should do this. I think it's much > better to always be able to start all boot CPUs[1] and simply rely on > "normal" HVO. >=20 > [1] Just to be clear on terminology, I'm using "boot CPUs" to refer to > all CPUs that are onlined at boot-time, not just to *the* boot CPU > that starts the kernel. Thanks for the clarification. We previously had different understandings of the boot CPU, but that should be resolved now. >=20 >>=20 >> If the boot CPU does not support AF, enabling HVO is not = permitted=E2=80=94neither >> via cmdline nor sysctl. All CPUs are allowed to be online. >>=20 >> Once the system has started, there is only one way to enable HVO, and >> that is through sysctl. 1) If the current system includes any CPUs = that do >> not support AF, enabling HVO is not permitted (unless the user = chooses to >> take these CPUs that don't support AF offline). 2) If all online CPUs = in the >> current system support AF, then HVO can be enabled. In this case, the >> system must also block any CPUs that do not support AF from coming = online >> in the future. >=20 > An approach similar to this is implemented in this series: if there > are (or were) any HVOed folios, onlining a CPU without HW AF is not > allowed. If HW AF is not supported by any onlined CPU, HVO cannot > proceed (though it will appear to be enabled). >=20 > I probably should have integrated > `arch_hugetlb_vmemmap_optimization_supported()` and a `hvo_status` > check into the sysctl handler for HVO. I'll do this if it is still > relevant in v2. >=20 >> Of course, if a user specifically wants those CPUs to be >> allowed to come online, they can choose to disable HVO first (At the = same time, >> this means that the HVO-optimized HugeTLB must be freed first as = well). >=20 > I could make it so that, if there are no HVOed pages anymore, onlining > incompatible CPUs becomes permitted again. I don't really see the need > to do this. >=20 >> My proposal assumes that even though the system supports different = types of >> CPUs, their support for AF (I suppose this is a very basic feature = nowadays) >> is generally the same. I believe that for servers (Only servers stand = to >> benefit more from HVO), there shouldn't be a mix of those that = support AF >> and those that don't. At the very least, such a situation would be = extremely >> uncommon. I believe this may be acceptable. That said, if we go this = route, >> is there a way to simplify the code implementation even further?. Of = course, >> this is just my personal speculation. Please let me know if I've = missed >> anything. >=20 > I agree that systems where HVO is relevant are very unlikely to have > mixed support for HW AF. >=20 > I don't think the approach you're suggesting in this email (as I've > understood it) simplifies this series. >=20 > I don't want to mess with the arm64 cpufeature bits more than I've > done here; preventing boot CPUs from onlining will certainly be more > work. :) And then your other suggestions seem to be quite similar to > what is already in these patches. >=20 > I am happy to get rid of the "disable HVO if it is unused and a CPU > without support is onlined" part though and replace it with something > similar (the simplification from the other thread[2]). Thanks for the explanation! My main goal right now is to keep the = initial implementation as simple as possible. If any of my suggestions aren't = quite on point, feel free to adjust them as you see fit. Thanks. >=20 > [2] = https://lore.kernel.org/linux-mm/674EC6D6-B1F7-447D-B462-861ECA5A99B7@linu= x.dev/