From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (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 977D73D9DD1 for ; Tue, 6 Oct 2026 13:47:47 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1791294468; cv=none; b=tTz5FjTqwrK09u74qrCjUr7HDCG8PsHO2KiSZTmxISE/5O2FttExWb434p/gDN+/55/6OFm/xu5A8DyFBQkY6j9nxIIL8PVyb3f8IXGwvn+kMsaKtXamfa7tya7FVIKmk1Ows06K2JgGm8YfmwREA3Al6oXg/UwB0yB+zwi4KI4= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1791294468; c=relaxed/simple; bh=MxyGa2i0GFPM1T2Ny30asuy+UCUNkZHMxNy6WfbG/GY=; h=MIME-Version:Date:From:To:Cc:Message-Id:In-Reply-To:References: Subject:Content-Type; b=kw0/MWZ5I1xlgcTcEPWaA5IZzJ3kfEcMhw0nhjIakPp+q5vGjg2xdS/PDKUXoF5qD7iWRGepA8gHVmhCuJFHUZWRUuttEi6Ni6eGfLgFLofiFhlJG1dB6Fp5tMhs2Otk5FhbpvtIpPpf01ps0gd2DxTRzLU8ym3/0BejIxiPOD0= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=TlXEotUy; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="TlXEotUy" Received: by smtp.kernel.org (Postfix) with ESMTPSA id F1CF91F00893; Tue, 6 Oct 2026 13:47:46 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1791294467; bh=4Gtx1novXP6NqQnE1CX/522SAS8aNnUdLs4H64hBEnI=; h=Date:From:To:Cc:In-Reply-To:References:Subject; b=TlXEotUyfN4Ui8BtrraBmYMlC9P3jQs2WXBc897AfBo6she2Q0cULcJBjw/A82sgt MZkg6A1XoUu+KyrH/sxKIvJqP8bgkR+5k2gZZ0Cr29TX8ZcPXG02E2ydp5lyM377Li 1d++6uRHKxTZ7iCq75Hed5xo/RMxSaPo2jaTIA2ujAAvZuq5zj0g7ik4ZygO5wVNQ6 xfENvhRbb+oHnzWdXkJYiA7gCp/spmNCESnPrMi2WnHuTA7GL5B00tuOtPiEX/qCRH uIh2dTypCoiy1zYdCWPE3taEILtmhMYMQRTDk5UocsenuY5RjAqenm9ANXzJBYMaKI 1UvmqcRbligyA== Received: from phl-compute-10.internal (phl-compute-10.internal [10.202.2.50]) by mailfauth.phl.internal (Postfix) with ESMTP id 38455F40068; Tue, 6 Oct 2026 09:47:46 -0400 (EDT) Received: from phl-imap-15 ([10.202.2.104]) by phl-compute-10.internal (MEProxy); Tue, 06 Oct 2026 09:47:46 -0400 X-ME-Sender: X-ME-Proxy-Cause: dmFkZTFre9ofM+lzYYLcctWi1Mf9vIDpVpzAQnegcTHCGRTPoVIa7iVaZyBIeYH6XI6anR kT7+/hCBpSOtTlzv1f1kog+0sGLuhQm2MIWuTCmBNwigr/Rdzh4E8OmBsddbWqGvSvRSDj XxPVG2EJv8gtl33qDII7a0rdD0Qs2ISfKfGCBQImC6TabGg+dMmqjS1l2jdVmG5dXkO0Re ZNGJh3kKD5ni3hzsZCEsWWlG9X217CGxOWWulHbNTjsaQ+cm049OJW0XLsgKw+qK4fXFhX 48JB58dMeCT4TtAcmMw/AcBWVuxz+6p9j1ZNnJj+Y+eTTh+GKP199TqALT+swGW+mHyAMk zPdfpzwMNCVsCrxa0ld/tyt6S65DHINoPhzYuVKJMTUPPei3eR3spECwjtxoVVrYJKTGfa w5y2kCQugxk3gGCLp942QYY0dG5zbknpoRCzfNHvgl7qg5pKg08g6dCQjN8aw6sdK6bwzf R47odG02mSQCuQtY3Gy9ZX9KywebqZFyzCm/rSwazNGYqY8LvMuJA0G0kDkEap+vHwsRq3 oln9LeOptCj2tXC/kxcq5xhw3gnSALJeroOzy91hazDTgLleqzC+jc6iCHFTCVncT3xJ7Q 1bzjBQZy7gZwytDsScwVZTTLvkfUW7ztaAl25+BXhpfdT0eCxy2bbozbanUA X-ME-Proxy: Feedback-ID: ifa6e4810:Fastmail Received: by mailuser.phl.internal (Postfix, from userid 501) id 13211780070; Tue, 6 Oct 2026 09:47:46 -0400 (EDT) X-Mailer: MessagingEngine.com Webmail Interface Precedence: bulk X-Mailing-List: linux-rdma@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-ThreadId: AFCZ3x6p1CiR Date: Tue, 06 Oct 2026 09:47:25 -0400 From: "Chuck Lever" To: "J Louis Kaplan" , "Dai Ngo" , "Jeff Layton" , NeilBrown , "Olga Kornievskaia" , "Tom Talpey" Cc: linux-nfs@vger.kernel.org, linux-rdma@vger.kernel.org, "Anna Schumaker" , "Jason Gunthorpe" , "Leon Romanovsky" , "Trond Myklebust" Message-Id: In-Reply-To: <20261006090028.3412544-1-Louis.Kaplan@arm.com> References: <20261006090028.3412544-1-Louis.Kaplan@arm.com> Subject: Re: [RFC PATCH 0/6] Improve NFS server direct throughput with passthrough enabled Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable On Tue, Oct 6, 2026, at 5:00 AM, J Louis Kaplan wrote: > For NFSoRDMA using RoCE, and with IOMMU passthrough enabled, use of a > 64KiB base page size was found to provide higher NFS read throughput > than a 4KiB base page size. This patch series recovers some of that > throughput with a 4k base page size by opportunistic use of large foli= os > for the svc reply buffer. > LLM Usage > =3D=3D=3D=3D=3D=3D=3D=3D=3D > > The idea of folio usage here was human-generated from observations of > performance differences between 4KiB and 64KiB page size kernels. Actually a similar approach has been tried before, but on x86_64: https://lore.kernel.org/linux-nfs/20260319133610.2556826-1-cel@kernel.or= g/ And it had to be reverted: - Mike Snitzer's report (4 Jun=C2=A02026): https://lore.kernel.org/linux-nfs/aiHlPmeZq3WgMwoJ@kernel.org/ - Jonathan Flynn's benchmark and perf data (5 Jun=C2=A02026), showing se= rver CPU going from 8.54% to 76.35% with the commit present: https://lore.kernel.org/linux-nfs/3cb119b4b2a8aada30c0c60286778a54@m= ail.gmail.com/ These two are the Closes: targets in the revert, which is a39f0ce0c9da upstream and 284d5ba931a5 in stable. Lock contention analysis and the decision to revert - "[PATCH] svcrdma: Cap Read sink allocations at PAGE_ALLOC_COSTLY_ORDER" (5 Jun=C2=A02026), whose description carries the zone->lock analysis: https://lore.kernel.org/linux-nfs/20260606035722.83175-1-cel@kernel.= org/ - Flynn's results for that fix (25.4 GiB/s, against 30.3 regressed and 73.9 reverted): https://lore.kernel.org/linux-nfs/65a2cdb132b0c28e69a29955e3bd37e7@m= ail.gmail.com/ - Your reply saying the two failed fixes mean 18755b8c2f24 has to be reverted (6 Jun=C2=A02026): https://lore.kernel.org/linux-nfs/096a2b91-7a19-48da-a06a-dc60e71509= 56@app.fastmail.com/ The earlier failed fix, "[PATCH] svcrdma: Avoid direct reclaim when allocating Read sink buffers", is at https://lore.kernel.org/linux-nfs/20260605223118.75092-1-cel@kernel.= org/. So: Yes, we would like to ensure that svcrdma's RDMA Read buffers (at least, and maybe RDMA Write) are contiguous so that the overhead of a DMA mapping operation will be low. I'm not expert enough with the page allocator's free path to address the lock contention problem, however. --=20 Chuck Lever (Come to NFS bake-a-thon! https://nfsv4bat.org)