From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from out-182.mta1.migadu.com (out-182.mta1.migadu.com [95.215.58.182]) (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 7696E127B67 for ; Thu, 28 Mar 2024 15:17:07 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=95.215.58.182 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1711639030; cv=none; b=F2l1XRpM69bAFqtE2RIgPXE11TLUQHEMRQsPKtO49q38wdz4ZeCnqpcfBRlipPEIouYy0HMLabztS5VEGXgFPEVrMlhuqjUqlMvbn5Rk/2hw+DPW3cK8c+RMx7Cg5bAt35O53B2aBpXtfWgBVFEwHFZEyps/DBpswUc7tHVn+2U= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1711639030; c=relaxed/simple; bh=QQjLBj2C0qQKA5i/iZiQq6ajP7ELeJUF6DhYJP7iQjY=; h=MIME-Version:Date:Content-Type:From:Message-ID:Subject:To:Cc: In-Reply-To:References; b=FmDZxXx2Tqxrb73tEv3Z7dr3ARck1xDw1Lmg4oyN06JRFOFo9xkqVd+mGf8zoGPBlw0/T2Nhhtw5Cx/5JPgPx6J9FeZNvH6LaawFA/l/9HRjjALgqizkbhjIUOjpsP2IisRnyfE3QTV0g+EszInCIvARqo9+8WHJ00w5MD15gFY= 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=NifsqhM1; arc=none smtp.client-ip=95.215.58.182 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="NifsqhM1" Precedence: bulk X-Mailing-List: v9fs@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.dev; s=key1; t=1711639025; 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=LXggh5/kcSlBOy0okYkks8aUrI4wvQzi55n43pRbQZ8=; b=NifsqhM1EO47z3laL4aEGqv9Fw+ya9JMSFaky+AvWGR46nFaSjmJEpLeO44++lqLLVHHQu PTfcN49fbyxO62MB3pREkLG+4nC65hifJ+ooMceTDKFX8Sr+i2i5zKB528ofJh5ZV3AVyM M8hiXQGSCOG7IPYlL/upeqtJX9PYLvI= Date: Thu, 28 Mar 2024 15:17:03 +0000 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable X-Report-Abuse: Please report any abuse attempt to abuse@migadu.com and include these headers. From: "Eric Van Hensbergen" Message-ID: TLS-Required: No Subject: Re: [PATCH v2 2/4] fs/9p: drop inodes immediately on non-.L too To: "Joakim Sindholt" , v9fs@lists.linux.dev Cc: "Joakim Sindholt" In-Reply-To: <145fc52b2452095c44260825670141d0d3b42c7d@linux.dev> References: <20240318112542.18863-1-opensource@zhasha.com> <20240318112542.18863-3-opensource@zhasha.com> <145fc52b2452095c44260825670141d0d3b42c7d@linux.dev> X-Migadu-Flow: FLOW_OUT Which server are you testing against? I'm wondering if the underlying pr= oblem might actually be i_nlink being maintained incorrectly by the server whic= h would lead to inodes lingering since that is the primary way the generic_inode_= drop differentiates. It is possible that we could could add an v9fs_inode_alw= ays_drop in for legacy if the servers weren't reporting i_nlink a compatible fashi= on although that might result in never caching legacy 9p2000 (which is proba= bly what most legacy folks want anyways). -eric =20 March=2028, 2024 at 10:08 AM, "Eric Van Hensbergen" wrote: >=20 >=20Slowly parsing through these, thanks for the fixes. >=20 >=20This one has a bit of a problem in that we don't have a v9fs specific >=20 >=20drop_inode anymore (either in legacy or .L). The release of the fid s= hould=20 >=20 > be handled by v9fs_dir_release in both versions of the protocol.=20 >=20 > I had convinced myself we could just use generic_drop_inode because we= =20 >=20 > decoupled clunking fids from the dentry/inode structures and just handl= ed >=20 >=20them directly in v9fs_dir_release -- so really by recovering the inod= e=20 >=20 > structure every time we were just forcing churn in the inode alloc/deal= loc >=20 >=20routines that seemed unnecessary (even in the uncached mode). >=20 >=20Was your concern the performance of the client side lookup of the ino= de >=20 >=20based on qid or the server side based on fid and can you give me a bi= t >=20 >=20more information on what you are seeing? Are you not seeing fid clunk= s >=20 >=20for certain conditions (ie. transient fids, etc.)? >=20 >=20 -eric >=20 >=20March 18, 2024 at 6:22 AM, "Joakim Sindholt" = wrote: >=20 >=20>=20 >=20> Signed-off-by: Joakim Sindholt > >=20 >=20>=20=20 >=20>=20 >=20> --- > >=20 >=20>=20=20 >=20>=20 >=20> fs/9p/vfs_super.c | 1 + > >=20 >=20>=20=20 >=20>=20 >=20> 1 file changed, 1 insertion(+) > >=20 >=20>=20=20 >=20>=20 >=20> diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c > >=20 >=20>=20=20 >=20>=20 >=20> index 941f7d0e0bfa..23cc67f29af2 100644 > >=20 >=20>=20=20 >=20>=20 >=20> --- a/fs/9p/vfs_super.c > >=20 >=20>=20=20 >=20>=20 >=20> +++ b/fs/9p/vfs_super.c > >=20 >=20>=20=20 >=20>=20 >=20> @@ -310,6 +310,7 @@ static const struct super_operations v9fs_supe= r_ops =3D { > >=20 >=20>=20=20 >=20>=20 >=20> .alloc_inode =3D v9fs_alloc_inode, > >=20 >=20>=20=20 >=20>=20 >=20> .free_inode =3D v9fs_free_inode, > >=20 >=20>=20=20 >=20>=20 >=20> .statfs =3D simple_statfs, > >=20 >=20>=20=20 >=20>=20 >=20> + .drop_inode =3D v9fs_drop_inode, > >=20 >=20>=20=20 >=20>=20 >=20> .evict_inode =3D v9fs_evict_inode, > >=20 >=20>=20=20 >=20>=20 >=20> .show_options =3D v9fs_show_options, > >=20 >=20>=20=20 >=20>=20 >=20> .umount_begin =3D v9fs_umount_begin, > >=20 >=20>=20=20 >=20>=20 >=20> --=20 >=20>=20 >=20>=20=20 >=20>=20 >=20> 2.43.2 > > >