From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.remote-ip=100.103.45.18 ARC-Seal: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=none; t=1795085855; b=WSidgroAVaSZJxnv9ifO/9YYspDjaTYTHA0Ejbao++oMSWmlbE7zHHPo8FI67ysclJzB dC+mDQiY/OKWlt6+CiF92odjK/61giODHxV8pBu8r+3GGvz2QudR2t5as9truvUlo7f2z iVHx/mOIIelPD9unlzBuOxF5WULeaseGAvhY1AQubuFzR95y/izxdbFTH5ILd3NXPJxTi 1QFOMg7Y72NBJLpCAd9siLtRCWtpAcbGaTCXx+WZT2d4sijYhnEqA2lD5lUViw+DEZLhU XBCrMjHA6UcXDiiYPrcqn9VPiWLvnHoxnx05Cn/DylmhBqNnhyGDpliBmqDNvXInlTQ== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1795085855; h=Received:DKIM-Signature:Message-ID:From:Subject:Reply-To:To:Cc: In-Reply-To:References:Content-Type:Content-Transfer-Encoding:Date; bh=6lj1ZS90H9VJapZVfQ7UItzB0FRkdpix0cZWX6LCseQ=; b=iThCMfruBa2E6xuE0c+GqATi1TSabaokZpqXxrtotxpyq/hY3O5REcP1aFh4N0w03iNO QXGsOAlmAFD72iRVMOXGtDWqOayuRyvnvR5RyN1aVLz86pZur+qxVdLfu2TKaBCnSKhqz g1kdW8zcF5Q0r9XVCi4lBluIERmocoR6WTN2GNzoJwCh4qZMYh9jCkG9xtHnbcFDC7F85 PntqKH4lZCb0l0TZMCUlX5+Yh8cdSt3MlYLb6fxxsdPkB0igE1ZmtJsS76lm3u1KOGJjV jdjZUYOxx53JWCwf56Jf0QYPDSIzbVzBWYHHHxrd6CQ40LyvuvHXZWXyJxAux2e4vgg== ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=HmqdnD75; arc=none smtp.remote-ip=100.103.45.18 Received: from smtp.kernel.org (quasi.space.kernel.org [100.103.45.18]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (prime256v1)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id BB954606A320 for ; Thu, 19 Nov 2026 10:57:35 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=HmqdnD75 Received: by smtp.kernel.org (Postfix) with ESMTPSA id 3DFCC1F000C4; Thu, 19 Nov 2026 10:57:35 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1795085855; bh=6lj1ZS90H9VJapZVfQ7UItzB0FRkdpix0cZWX6LCseQ=; h=From:Subject:Reply-To:To:Cc:In-Reply-To:References:Date; b=HmqdnD75XY44IPPVgtRkAPOxoU8FnxP2Ay9PdwzpaJQ1pkoxmlu9zqRyokcThm6wN D+hgtxAb2Q0WbCPL2QhgVr71Pw+VANxoWN+gxQDbwj9VwizgaAifLPCid8WxBP4qGE uITGr/HQWS55DKCadkpbPfFyhEb+93+HTW7IopPDEkptgbtrDjemPbqyJWSvOx0riY 8OSK6Fy+RCllaV7J/c7SOSkyzldRWcDart08uYIZXUrkdre3BRloVsxDryhvTPM+nO 6Kio7oqu+ynH4cL8pFagaI1HsZCdTyB1PcYwTqx9zW/1MX8Bfp8UcMKypdb5PT7Wbu YZ+oDY6KolnjQ== Message-ID: From: sashiko-bot@kernel.org Subject: Re: [PATCH net-next V3 1/2] net/mlx5e: Batch NAPI RCU grace periods on channel close Reply-To: sashiko-reviews@lists.linux.dev To: "Tariq Toukan" Cc: linux-rdma@vger.kernel.org In-Reply-To: <20261119103935.2190627-2-tariqt@nvidia.com> References: <20261119103935.2190627-1-tariqt@nvidia.com> <20261119103935.2190627-2-tariqt@nvidia.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Date: Thu, 19 Nov 2026 10:57:34 +0000 X-Mailing-List: linux-rdma@vger.kernel.org List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list > From: Carolina Jubran >=20 > Channel close currently invokes synchronize_net() repeatedly while > tearing down each channel. RQ, ICOSQ, TX SQ and NAPI teardown each > require a grace period, causing the number of waits to scale with the > number of channels. TX SQ teardown adds a wait for each traffic class, > while XDP and QoS can add further calls. >=20 > Batch these waits across all channels. Split channel deactivate and > close into pre_sync and post_sync phases. First clear the ENABLED state > on all queues, then wait for a single grace period before continuing > with the teardown. Once NAPI is disabled on all channels, wait for one > more grace period, as the poll may still arm the CQs after > napi_complete_done(). >=20 > Similarly, close the queues and unlink all NAPI instances with > __netif_napi_del_locked(), issue one synchronize_net() after all of > them have been unlinked, and only then free the channels. >=20 > Keep the existing per object synchronize_net() calls for single queue > paths. Drop the one in mlx5e_close_xdpsq(), as its callers already > wait for a grace period or run before NAPI is enabled. >=20 > 248 channels, rcu_normal=3D1, ip link set eth2 down: > Before: 11.837s > After: 0.817s (~14.5x faster) >=20 > Signed-off-by: Carolina Jubran > Signed-off-by: Tariq Toukan Sashiko has reviewed this patch and found no issues. --=20 Sashiko AI review =C2=B7 https://sashiko.dev/#/patchset/20261119103935.2190= 627-1-tariqt@nvidia.com?part=3D1