From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (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 28BE21C1AAA for ; Mon, 13 Jan 2025 21:44:28 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1736804669; cv=none; b=hElp4NzA7XYI/SCWq2d2XzNIHOr08FLncztbQICXxsvR2y9+xnp7QLpEHPL2LPTgbmiQCdgnbrPaOu+csnwthw5G/AbTf/Urb9OhrmzVn7QihSGBR1yTgQwgfP/SDVB/E75O1uCG6KqVBmu8hWdV/qmCMrEMYu6ST+hmOHRG/TU= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1736804669; c=relaxed/simple; bh=5J4FIZHJNtdwTF6zwzfIgKGrq/YLgfdmKM3PSrrX3Es=; h=Message-ID:Subject:From:To:Cc:Date:In-Reply-To:References: Content-Type:MIME-Version; b=qqxRW37EEXuIjHDXgVuj8v7ZwD1nLh/2zoI0id0crEfWzrLEk9hJNMXsPz/MN5CPNPZ4SXn9r3ObuKnSWVOARFMrAq0YnoDE/gBTGpTzlSIGMOvEFvRWmV3Imx6xmbU08W7BJnANB1aPLY/2ga2Y3xAWDu0yBgY0MkAFys12lYk= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=SjYV0XVJ; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="SjYV0XVJ" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 7834DC4CED6; Mon, 13 Jan 2025 21:44:27 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1736804668; bh=5J4FIZHJNtdwTF6zwzfIgKGrq/YLgfdmKM3PSrrX3Es=; h=Subject:From:To:Cc:Date:In-Reply-To:References:From; b=SjYV0XVJ9ygFqF8YCCw5Rtm96Ciknj6cXG/E9o4mfQ77Qe5sYbSJtcE0Oe5KO8rza 3/+TVVjsWsqhX8DszlFEdn6SQ2GfP3m8U5jN0D5neIEETu+jATTAq+RAUnmM4wqMfc bs6Kpz9E1RSz6Lo0xWjASSuq+PBRlycBfa84JRqCk15GSZELFm5omAgo6y1ITNLPM0 qecyYBZu0u+f5/nktHyBFD/R+3TNUhSpkAvPu8nkEQNna+V4aZ8T6V1t9fD83xwBPA owWqSH2obDzD3quB3x5rMGiUgut8fU3ZOhxzJzWR2QvsDlto7M/3aLuN1c6hRtMWUo oLxhforL+oHRA== Message-ID: Subject: Re: [PATCH v6 4/5] mm/migrate: skip migrating folios under writeback with AS_WRITEBACK_INDETERMINATE mappings From: Jeff Layton To: David Hildenbrand , Shakeel Butt Cc: Miklos Szeredi , Joanne Koong , Bernd Schubert , Zi Yan , linux-fsdevel@vger.kernel.org, jefflexu@linux.alibaba.com, josef@toxicpanda.com, linux-mm@kvack.org, kernel-team@meta.com, Matthew Wilcox , Oscar Salvador , Michal Hocko Date: Mon, 13 Jan 2025 16:44:26 -0500 In-Reply-To: <2848b566-3cae-4e89-916c-241508054402@redhat.com> References: <791d4056-cac1-4477-a8e3-3a2392ed34db@redhat.com> <1fdc9d50-584c-45f4-9acd-3041d0b4b804@redhat.com> <54ebdef4205781d3351e4a38e5551046482dbba0.camel@kernel.org> <2848b566-3cae-4e89-916c-241508054402@redhat.com> Autocrypt: addr=jlayton@kernel.org; prefer-encrypt=mutual; keydata=mQINBE6V0TwBEADXhJg7s8wFDwBMEvn0qyhAnzFLTOCHooMZyx7XO7dAiIhDSi7G1NPxw n8jdFUQMCR/GlpozMFlSFiZXiObE7sef9rTtM68ukUyZM4pJ9l0KjQNgDJ6Fr342Htkjxu/kFV1Wv egyjnSsFt7EGoDjdKqr1TS9syJYFjagYtvWk/UfHlW09X+jOh4vYtfX7iYSx/NfqV3W1D7EDi0PqV T2h6v8i8YqsATFPwO4nuiTmL6I40ZofxVd+9wdRI4Db8yUNA4ZSP2nqLcLtFjClYRBoJvRWvsv4lm 0OX6MYPtv76hka8lW4mnRmZqqx3UtfHX/hF/zH24Gj7A6sYKYLCU3YrI2Ogiu7/ksKcl7goQjpvtV YrOOI5VGLHge0awt7bhMCTM9KAfPc+xL/ZxAMVWd3NCk5SamL2cE99UWgtvNOIYU8m6EjTLhsj8sn VluJH0/RcxEeFbnSaswVChNSGa7mXJrTR22lRL6ZPjdMgS2Km90haWPRc8Wolcz07Y2se0xpGVLEQ cDEsvv5IMmeMe1/qLZ6NaVkNuL3WOXvxaVT9USW1+/SGipO2IpKJjeDZfehlB/kpfF24+RrK+seQf CBYyUE8QJpvTZyfUHNYldXlrjO6n5MdOempLqWpfOmcGkwnyNRBR46g/jf8KnPRwXs509yAqDB6sE LZH+yWr9LQZEwARAQABtCVKZWZmIExheXRvbiA8amxheXRvbkBwb29jaGllcmVkcy5uZXQ+iQI7BB MBAgAlAhsDBgsJCAcDAgYVCAIJCgsEFgIDAQIeAQIXgAUCTpXWPAIZAQAKCRAADmhBGVaCFc65D/4 gBLNMHopQYgG/9RIM3kgFCCQV0pLv0hcg1cjr+bPI5f1PzJoOVi9s0wBDHwp8+vtHgYhM54yt43uI 7Htij0RHFL5eFqoVT4TSfAg2qlvNemJEOY0e4daljjmZM7UtmpGs9NN0r9r50W82eb5Kw5bc/r0km R/arUS2st+ecRsCnwAOj6HiURwIgfDMHGPtSkoPpu3DDp/cjcYUg3HaOJuTjtGHFH963B+f+hyQ2B rQZBBE76ErgTDJ2Db9Ey0kw7VEZ4I2nnVUY9B5dE2pJFVO5HJBMp30fUGKvwaKqYCU2iAKxdmJXRI ONb7dSde8LqZahuunPDMZyMA5+mkQl7kpIpR6kVDIiqmxzRuPeiMP7O2FCUlS2DnJnRVrHmCljLkZ Wf7ZUA22wJpepBligemtSRSbqCyZ3B48zJ8g5B8xLEntPo/NknSJaYRvfEQqGxgk5kkNWMIMDkfQO lDSXZvoxqU9wFH/9jTv1/6p8dHeGM0BsbBLMqQaqnWiVt5mG92E1zkOW69LnoozE6Le+12DsNW7Rj iR5K+27MObjXEYIW7FIvNN/TQ6U1EOsdxwB8o//Yfc3p2QqPr5uS93SDDan5ehH59BnHpguTc27Xi QQZ9EGiieCUx6Zh2ze3X2UW9YNzE15uKwkkuEIj60NvQRmEDfweYfOfPVOueC+iFifbQgSmVmZiBM YXl0b24gPGpsYXl0b25AcmVkaGF0LmNvbT6JAjgEEwECACIFAk6V0q0CGwMGCwkIBwMCBhUIAgkKC wQWAgMBAh4BAheAAAoJEAAOaEEZVoIViKUQALpvsacTMWWOd7SlPFzIYy2/fjvKlfB/Xs4YdNcf9q LqF+lk2RBUHdR/dGwZpvw/OLmnZ8TryDo2zXVJNWEEUFNc7wQpl3i78r6UU/GUY/RQmOgPhs3epQC 3PMJj4xFx+VuVcf/MXgDDdBUHaCTT793hyBeDbQuciARDJAW24Q1RCmjcwWIV/pgrlFa4lAXsmhoa c8UPc82Ijrs6ivlTweFf16VBc4nSLX5FB3ls7S5noRhm5/Zsd4PGPgIHgCZcPgkAnU1S/A/rSqf3F LpU+CbVBDvlVAnOq9gfNF+QiTlOHdZVIe4gEYAU3CUjbleywQqV02BKxPVM0C5/oVjMVx3bri75n1 TkBYGmqAXy9usCkHIsG5CBHmphv9MHmqMZQVsxvCzfnI5IO1+7MoloeeW/lxuyd0pU88dZsV/riHw 87i2GJUJtVlMl5IGBNFpqoNUoqmvRfEMeXhy/kUX4Xc03I1coZIgmwLmCSXwx9MaCPFzV/dOOrju2 xjO+2sYyB5BNtxRqUEyXglpujFZqJxxau7E0eXoYgoY9gtFGsspzFkVNntamVXEWVVgzJJr/EWW0y +jNd54MfPRqH+eCGuqlnNLktSAVz1MvVRY1dxUltSlDZT7P2bUoMorIPu8p7ZCg9dyX1+9T6Muc5d Hxf/BBP/ir+3e8JTFQBFOiLNdFtB9KZWZmIExheXRvbiA8amxheXRvbkBzYW1iYS5vcmc+iQI4BBM BAgAiBQJOldK9AhsDBgsJCAcDAgYVCAIJCgsEFgIDAQIeAQIXgAAKCRAADmhBGVaCFWgWD/0ZRi4h N9FK2BdQs9RwNnFZUr7JidAWfCrs37XrA/56olQl3ojn0fQtrP4DbTmCuh0SfMijB24psy1GnkPep naQ6VRf7Dxg/Y8muZELSOtsv2CKt3/02J1BBitrkkqmHyni5fLLYYg6fub0T/8Kwo1qGPdu1hx2BQ RERYtQ/S5d/T0cACdlzi6w8rs5f09hU9Tu4qV1JLKmBTgUWKN969HPRkxiojLQziHVyM/weR5Reu6 FZVNuVBGqBD+sfk/c98VJHjsQhYJijcsmgMb1NohAzwrBKcSGKOWJToGEO/1RkIN8tqGnYNp2G+aR 685D0chgTl1WzPRM6mFG1+n2b2RR95DxumKVpwBwdLPoCkI24JkeDJ7lXSe3uFWISstFGt0HL8Eew P8RuGC8s5h7Ct91HMNQTbjgA+Vi1foWUVXpEintAKgoywaIDlJfTZIl6Ew8ETN/7DLy8bXYgq0Xzh aKg3CnOUuGQV5/nl4OAX/3jocT5Cz/OtAiNYj5mLPeL5z2ZszjoCAH6caqsF2oLyAnLqRgDgR+wTQ T6gMhr2IRsl+cp8gPHBwQ4uZMb+X00c/Amm9VfviT+BI7B66cnC7Zv6Gvmtu2rEjWDGWPqUgccB7h dMKnKDthkA227/82tYoFiFMb/NwtgGrn5n2vwJyKN6SEoygGrNt0SI84y6hEVbQlSmVmZiBMYXl0b 24gPGpsYXl0b25AcHJpbWFyeWRhdGEuY29tPokCOQQTAQIAIwUCU4xmKQIbAwcLCQgHAwIBBhUIAg kKCwQWAgMBAh4BAheAAAoJEAAOaEEZVoIV1H0P/j4OUTwFd7BBbpoSp695qb6HqCzWMuExsp8nZjr uymMaeZbGr3OWMNEXRI1FWNHMtcMHWLP/RaDqCJil28proO+PQ/yPhsr2QqJcW4nr91tBrv/MqItu AXLYlsgXqp4BxLP67bzRJ1Bd2x0bWXurpEXY//VBOLnODqThGEcL7jouwjmnRh9FTKZfBDpFRaEfD FOXIfAkMKBa/c9TQwRpx2DPsl3eFWVCNuNGKeGsirLqCxUg5kWTxEorROppz9oU4HPicL6rRH22Ce 6nOAON2vHvhkUuO3GbffhrcsPD4DaYup4ic+DxWm+DaSSRJ+e1yJvwi6NmQ9P9UAuLG93S2MdNNbo sZ9P8k2mTOVKMc+GooI9Ve/vH8unwitwo7ORMVXhJeU6Q0X7zf3SjwDq2lBhn1DSuTsn2DbsNTiDv qrAaCvbsTsw+SZRwF85eG67eAwouYk+dnKmp1q57LDKMyzysij2oDKbcBlwB/TeX16p8+LxECv51a sjS9TInnipssssUDrHIvoTTXWcz7Y5wIngxDFwT8rPY3EggzLGfK5Zx2Q5S/N0FfmADmKknG/D8qG IcJE574D956tiUDKN4I+/g125ORR1v7bP+OIaayAvq17RP+qcAqkxc0x8iCYVCYDouDyNvWPGRhbL UO7mlBpjW9jK9e2fvZY9iw3QzIPGKtClKZWZmIExheXRvbiA8amVmZi5sYXl0b25AcHJpbWFyeWRh dGEuY29tPokCOQQTAQIAIwUCU4xmUAIbAwcLCQgHAwIBBhUIAgkKCwQWAgMBAh4BAheAAAoJEAAOa EEZVoIVzJoQALFCS6n/FHQS+hIzHIb56JbokhK0AFqoLVzLKzrnaeXhE5isWcVg0eoV2oTScIwUSU apy94if69tnUo4Q7YNt8/6yFM6hwZAxFjOXR0ciGE3Q+Z1zi49Ox51yjGMQGxlakV9ep4sV/d5a50 M+LFTmYSAFp6HY23JN9PkjVJC4PUv5DYRbOZ6Y1+TfXKBAewMVqtwT1Y+LPlfmI8dbbbuUX/kKZ5d dhV2736fgyfpslvJKYl0YifUOVy4D1G/oSycyHkJG78OvX4JKcf2kKzVvg7/Rnv+AueCfFQ6nGwPn 0P91I7TEOC4XfZ6a1K3uTp4fPPs1Wn75X7K8lzJP/p8lme40uqwAyBjk+IA5VGd+CVRiyJTpGZwA0 jwSYLyXboX+Dqm9pSYzmC9+/AE7lIgpWj+3iNisp1SWtHc4pdtQ5EU2SEz8yKvDbD0lNDbv4ljI7e flPsvN6vOrxz24mCliEco5DwhpaaSnzWnbAPXhQDWb/lUgs/JNk8dtwmvWnqCwRqElMLVisAbJmC0 BhZ/Ab4sph3EaiZfdXKhiQqSGdK4La3OTJOJYZphPdGgnkvDV9Pl1QZ0ijXQrVIy3zd6VCNaKYq7B AKidn5g/2Q8oio9Tf4XfdZ9dtwcB+bwDJFgvvDYaZ5bI3ln4V3EyW5i2NfXazz/GA/I/ZtbsigCFc 8ftCBKZWZmIExheXRvbiA8amxheXRvbkBrZXJuZWwub3JnPokCOAQTAQIAIgUCWe8u6AIbAwYLCQg HAwIGFQgCCQoLBBYCAwECHgECF4AACgkQAA5oQRlWghUuCg/+Lb/xGxZD2Q1oJVAE37uW308UpVSD 2tAMJUvFTdDbfe3zKlPDTuVsyNsALBGclPLagJ5ZTP+Vp2irAN9uwBuacBOTtmOdz4ZN2tdvNgozz uxp4CHBDVzAslUi2idy+xpsp47DWPxYFIRP3M8QG/aNW052LaPc0cedYxp8+9eiVUNpxF4SiU4i9J DfX/sn9XcfoVZIxMpCRE750zvJvcCUz9HojsrMQ1NFc7MFT1z3MOW2/RlzPcog7xvR5ENPH19ojRD CHqumUHRry+RF0lH00clzX/W8OrQJZtoBPXv9ahka/Vp7kEulcBJr1cH5Wz/WprhsIM7U9pse1f1g Yy9YbXtWctUz8uvDR7shsQxAhX3qO7DilMtuGo1v97I/Kx4gXQ52syh/w6EBny71CZrOgD6kJwPVV AaM1LRC28muq91WCFhs/nzHozpbzcheyGtMUI2Ao4K6mnY+3zIuXPygZMFr9KXE6fF7HzKxKuZMJO aEZCiDOq0anx6FmOzs5E6Jqdpo/mtI8beK+BE7Va6ni7YrQlnT0i3vaTVMTiCThbqsB20VrbMjlhp f8lfK1XVNbRq/R7GZ9zHESlsa35ha60yd/j3pu5hT2xyy8krV8vGhHvnJ1XRMJBAB/UYb6FyC7S+m QZIQXVeAA+smfTT0tDrisj1U5x6ZB9b3nBg65kc= Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable User-Agent: Evolution 3.54.2 (3.54.2-1.fc41) Precedence: bulk X-Mailing-List: linux-fsdevel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 On Mon, 2025-01-13 at 16:27 +0100, David Hildenbrand wrote: > On 10.01.25 23:00, Shakeel Butt wrote: > > On Fri, Jan 10, 2025 at 10:13:17PM +0100, David Hildenbrand wrote: > > > On 10.01.25 21:28, Jeff Layton wrote: > > > > On Thu, 2025-01-09 at 12:22 +0100, David Hildenbrand wrote: > > > > > On 07.01.25 19:07, Shakeel Butt wrote: > > > > > > On Tue, Jan 07, 2025 at 09:34:49AM +0100, David Hildenbrand wro= te: > > > > > > > On 06.01.25 19:17, Shakeel Butt wrote: > > > > > > > > On Mon, Jan 06, 2025 at 11:19:42AM +0100, Miklos Szeredi wr= ote: > > > > > > > > > On Fri, 3 Jan 2025 at 21:31, David Hildenbrand wrote: > > > > > > > > > > In any case, having movable pages be turned unmovable d= ue to persistent > > > > > > > > > > writaback is something that must be fixed, not worked a= round. Likely a > > > > > > > > > > good topic for LSF/MM. > > > > > > > > >=20 > > > > > > > > > Yes, this seems a good cross fs-mm topic. > > > > > > > > >=20 > > > > > > > > > So the issue discussed here is that movable pages used fo= r fuse > > > > > > > > > page-cache cause a problems when memory needs to be compa= cted. The > > > > > > > > > problem is either that > > > > > > > > >=20 > > > > > > > > > - the page is skipped, leaving the physical memory b= lock unmovable > > > > > > > > >=20 > > > > > > > > > - the compaction is blocked for an unbounded time > > > > > > > > >=20 > > > > > > > > > While the new AS_WRITEBACK_INDETERMINATE could potentiall= y make things > > > > > > > > > worse, the same thing happens on readahead, since the new= page can be > > > > > > > > > locked for an indeterminate amount of time, which can als= o block > > > > > > > > > compaction, right? > > > > > > >=20 > > > > > > > Yes, as memory hotplug + virtio-mem maintainer my bigger conc= ern is these > > > > > > > pages residing in ZONE_MOVABLE / MIGRATE_CMA areas where ther= e *must not be > > > > > > > unmovable pages ever*. Not triggered by an untrusted source, = not triggered > > > > > > > by an trusted source. > > > > > > >=20 > > > > > > > It's a violation of core-mm principles. > > > > > >=20 > > > > > > The "must not be unmovable pages ever" is a very strong stateme= nt and we > > > > > > are violating it today and will keep violating it in future. An= y > > > > > > page/folio under lock or writeback or have reference taken or h= ave been > > > > > > isolated from their LRU is unmovable (most of the time for smal= l period > > > > > > of time). > > > > >=20 > > > > > ^ this: "small period of time" is what I meant. > > > > >=20 > > > > > Most of these things are known to not be problematic: retrying a = couple > > > > > of times makes it work, that's why migration keeps retrying. > > > > >=20 > > > > > Again, as an example, we allow short-term O_DIRECT but disallow > > > > > long-term page pinning. I think there were concerns at some point= if > > > > > O_DIRECT might also be problematic (I/O might take a while), but = so far > > > > > it was not a problem in practice that would make CMA allocations = easily > > > > > fail. > > > > >=20 > > > > > vmsplice() is a known problem, because it behaves like O_DIRECT b= ut > > > > > actually triggers long-term pinning; IIRC David Howells has this = on his > > > > > todo list to fix. [I recall that seccomp disallows vmsplice by de= fault > > > > > right now] > > > > >=20 > > > > > These operations are being done all over the place in kernel. > > > > > > Miklos gave an example of readahead. > > > > >=20 > > > > > I assume you mean "unmovable for a short time", correct, or can y= ou > > > > > point me at that specific example; I think I missed that. > >=20 > > Please see https://lore.kernel.org/all/CAJfpegthP2enc9o1hV-izyAG9nHcD_t= T8dKFxxzhdQws6pcyhQ@mail.gmail.com/ > >=20 > > > > >=20 > > > > > > The per-CPU LRU caches are another > > > > > > case where folios can get stuck for long period of time. > > > > >=20 > > > > > Which is why memory offlining disables the lru cache. See > > > > > lru_cache_disable(). Other users that care about that drain the L= RU on > > > > > all cpus. > > > > >=20 > > > > > > Reclaim and > > > > > > compaction can isolate a lot of folios that they need to have > > > > > > too_many_isolated() checks. So, "must not be unmovable pages ev= er" is > > > > > > impractical. > > > > >=20 > > > > > "must only be short-term unmovable", better? > >=20 > > Yes and you have clarified further below of the actual amount. > >=20 > > > > >=20 > > > >=20 > > > > Still a little ambiguous. > > > >=20 > > > > How short is "short-term"? Are we talking milliseconds or minutes? > > >=20 > > > Usually a couple of seconds, max. For memory offlining, slightly long= er > > > times are acceptable; other things (in particular compaction or CMA > > > allocations) will give up much faster. > > >=20 > > > >=20 > > > > Imposing a hard timeout on writeback requests to unprivileged FUSE > > > > servers might give us a better guarantee of forward-progress, but i= t > > > > would probably have to be on the order of at least a minute or so t= o be > > > > workable. > > >=20 > > > Yes, and that might already be a bit too much, especially if stuck on > > > waiting for folio writeback ... so ideally we could find a way to mig= rate > > > these folios that are under writeback and it's not your ordinary disk= driver > > > that responds rather quickly. > > >=20 > > > Right now we do it via these temp pages, and I can see how that's > > > undesirable. > > >=20 > > > For NFS etc. we probably never ran into this, because it's all used i= n > > > fairly well managed environments and, well, I assume NFS easily outda= tes CMA > > > and ZONE_MOVABLE :) > > >=20 > > > > > > >=20 > > > > > > The point is that, yes we should aim to improve things but in i= terations > > > > > > and "must not be unmovable pages ever" is not something we can = achieve > > > > > > in one step. > > > > >=20 > > > > > I agree with the "improve things in iterations", but as > > > > > AS_WRITEBACK_INDETERMINATE has the FOLL_LONGTERM smell to it, I t= hink we > > > > > are making things worse. > >=20 > > AS_WRITEBACK_INDETERMINATE is really a bad name we picked as it is stil= l > > causing confusion. It is a simple flag to avoid deadlock in the reclaim > > code path and does not say anything about movability. > >=20 > > > > >=20 > > > > > And as this discussion has been going on for too long, to summari= ze my > > > > > point: there exist conditions where pages are short-term unmovabl= e, and > > > > > possibly some to be fixed that turn pages long-term unmovable (e.= g., > > > > > vmsplice); that does not mean that we can freely add new conditio= ns that > > > > > turn movable pages unmovable long-term or even forever. > > > > >=20 > > > > > Again, this might be a good LSF/MM topic. If I would have the cap= acity I > > > > > would suggest a topic around which things are know to cause pages= to be > > > > > short-term or long-term unmovable/unsplittable, and which can be > > > > > handled, which not. Maybe I'll find the time to propose that as a= topic. > > > > >=20 > > > >=20 > > > >=20 > > > > This does sound like great LSF/MM fodder! I predict that this sessi= on > > > > will run long! ;) > > >=20 > > > Heh, fully agreed! :) > >=20 > > I would like more targeted topic and for that I want us to at least > > agree where we are disagring. Let me write down two statements and > > please tell me where you disagree: >=20 > I think we're mostly in agreement! >=20 > >=20 > > 1. For a normal running FUSE server (without tmp pages), the lifetime o= f > > writeback state of fuse folios falls under "short-term unmovable" bucke= t > > as it does not differ in anyway from anyother filesystems handling > > writeback folios. >=20 > That's the expectation, yes. As long as the FUSE server is able to make= =20 > progress, the expectation is that it's just like NFS etc. If it isn't=20 > able to make progress (i.e., crash), the expectation is that everything= =20 > will get cleaned up either way. >=20 > I wonder if there could be valid scenario where the FUSE server is no=20 > longer able to make progress (ignoring network outages), or the progress= =20 > might start being extremely slow such that it becomes a problem. In=20 > contrast to in-kernel FSs, one can do some fancy stuff with fuse where= =20 > writing a page could possibly consume a lot of memory in user-space.=20 > Likely, in this case we might just blame it on the admin that agreed to= =20 > running this (trusted) fuse server. >=20 > >=20 > > 2. For a buggy or untrusted FUSE server (without tmp pages), the > > lifetime of writeback state of fuse folios can be arbitrarily long and > > we need some mechanism to limit it. >=20 > Yes. >=20 >=20 > Especially in 1), we really want to wait for writeback to finish, just= =20 > like for any other filesystem. For 2), we want a way so writeback will= =20 > not get stuck for a long time, but are able to make progress and migrate= =20 > these pages. >=20 What if we were to allow the kernel to kill off an unprivileged FUSE server that was "misbehaving" [1], clean any dirty pagecache pages that it has, and set writeback errors on the corresponding FUSE inodes [2]? We'd still need a rather long timeout (on the order of at least a minute or so, by default). Would that be enough to assuage concerns about unprivileged servers pinning pages indefinitely? Buggy servers are still a problem, but there's not much we can do about that. There are a lot of details we'd have to sort out, so I'm also interested in whether anyone (Miklos? Bernd?) would find this basic approach objectionable. [1]: for some definition of misbehavior. Probably a writeback timeout=C2=A0of some sort but maybe there would be other criteria too. [2]: or maybe just make them eligible to be cleaned without talking to the server, should the VM wish it. --=20 Jeff Layton