Received: from mail-qv1-f54.google.com (mail-qv1-f54.google.com [209.85.219.54]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 7906D30FF36 for ; Fri, 7 Aug 2026 17:48:33 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.219.54 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1786124915; cv=none; b=dgTRdKDgDLermAb7/teJCr6xoVYPRsZc9pLI3q63Dg7tVvIhVE+icJxDfr33YLkArG4B0cdcXgBmLVro/+xR/e8Hdo8J2uom0EfPX0iI/FLZXIGNWUZOlPOfw5vxuJIlz+NSZbmiNsRcnnwCIASbUbJTOFQZ7WlZV2NefSDupog= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1786124915; c=relaxed/simple; bh=B2dhcUAy99nJ6Q8/2A/3L1wWVEyE/AbW7wE9iaKuQWk=; h=Date:From:To:Cc:Subject:Message-ID:References:MIME-Version: Content-Type:Content-Disposition:In-Reply-To; b=hsB75Mo/BnUJV6zaNaq8PP8Xj3729Egnz0amtlU186ywjhnDK5Clz3I50KZXvkTRlfbPC1pyAPsdGRkXyGQBZaYTQ0CI99CTdKzU4VtUpEkoTt1mOQnqpLtbEkU66J4K31ZP7zL/r7rHfYs1X/3lrBmNWW82UdfZieuNf8L6XLs= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=ziepe.ca; spf=pass smtp.mailfrom=ziepe.ca; dkim=pass (2048-bit key) header.d=ziepe.ca header.i=@ziepe.ca header.b=RUNoTSIQ; arc=none smtp.client-ip=209.85.219.54 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=ziepe.ca Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=ziepe.ca Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=ziepe.ca header.i=@ziepe.ca header.b="RUNoTSIQ" Received: by mail-qv1-f54.google.com with SMTP id 6a1803df08f44-8eeadbc5e21so24620976d6.3 for ; Fri, 07 Aug 2026 10:48:33 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ziepe.ca; s=google; t=1786124912; x=1786729712; darn=vger.kernel.org; h=in-reply-to:content-transfer-encoding:content-disposition :content-type:mime-version:references:message-id:subject:cc:to:from :date:from:to:cc:subject:date:message-id:reply-to:content-type; bh=a5IgWT1wGp6cqflJGNJAuwwfXTXvrfSYmZSZMvhYv0I=; b=RUNoTSIQOr7hxR2MvzQYz+TKhubjcVPdYCRIhq/p9n+anptqVMW96Z5V/aUxhOqZyn jqlL+lYkbpJRHRD8dReO6q+HVKtGOTbiCAuoygUs0QR7fvOI6XrUHbC2XSHopJ9FbNKK jyL+Vfnt3fc5nAV26iXex3Le4zf/gxiOhtu0dRsiDCuCrd6d6rJrbZDXw/ZJBBKRoHG9 a+yCPFGdl9oyyxudCjgwNOU+ehf2tI+cLSeKOXk3Kt0tvK1PjdfUtSJGFB8WC9ZJFAmg xteMd/qhUx5ovrhmqZ9X3W83nMzKgSRnIkjNwTN3CnwL4mSuAo05MCQugHRCHc36mumz d3lQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1786124912; x=1786729712; h=in-reply-to:content-transfer-encoding:content-disposition :content-type:mime-version:references:message-id:subject:cc:to:from :date:x-gm-gg:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to:content-type; bh=a5IgWT1wGp6cqflJGNJAuwwfXTXvrfSYmZSZMvhYv0I=; b=oSGw2aTOQKDfo+1cpTJopMsDZKunPEyNUlol2F4AiHZR1OSUDxons7v6CvU1LC6Pm/ VkUNvcDM9VVkujLR8RmdGPTwqEoGRkt/auoFMAhXfc/PSFcXe/pveHmpDfYpNK8XVYS9 ddyFmccgqr49EqLPpy/qTV6pgGETjxRnBsm4oZAfDtenLCSRXO1F2L2I+KQ2v70sJ6Ou F1dyxN/gSuKgY9ZKKt7zA75J3+6KzsIUNAZ+IYSKG37PDr1hGjiPzKP3tOmi7dZ0QEkd p0iRL/4VdeFG58374ugnUDfKD6/k9FT5B9RBJdQ8aC6v5k1ayM/XUWpyXo2nVBjpArSN JSgw== X-Forwarded-Encrypted: i=1; AHgh+Rp5Q+x/00Bvcg5meIXxTqoJLBmCv9L2tCXMymD3hzRlIDm0Fga93UjVPAIwdeYF+7EW+KNIaJr+Ex8L@vger.kernel.org X-Gm-Message-State: AOJu0Ywk194TvLw4GVsFj65OI0qgVVi3i2Ae6huJDXBW9MXSQtuUHO6j u4ngA1IyN+D0LshaUlYXK1fOrSpUpS2zbg+hpcrsTaBzSlzP78Y98qYAX7Ufi/+HHiI= X-Gm-Gg: AR+sD11/Kf95rdJLVie3itCAAcAZa4z2BpoaDMVjaM9lgCAJIuaGw98lPbQIfRygvgh 6bx+7xeX9hrVyX9fhT3u9Rmq4/Mv0SuoZRjtNoY+82KwX9tAuOjm08/WGSaHvBLUDHivzHoGNvy /tXE0crS7TVoS3kRLDJ6vUwjTaygSUTwHbtvW0IrGzlQxZwTODl8LDrLmTuqUVKttTpPVkDii8M gqDgOVmKqshaP92DEJFYRPYN881TY9BzalGxhZY4JKV12xy/IZIeMIoqgiICxHyyXMlhQXvelII 2aBvxR+KKtvf2ZbB2ysKzpnXqiykUMGDc9GSZW0gBExUu/5upfWekBnW910elcCzkVXF7M2ebUn 7XmKpQ1CsLCbUkIFxYgMA3M1THAmtVsb8Evaeu5nHmws4VIrYdOubTEIv8+aAguPIJUHBWx6wsU Gzh+9RQOzAxTFX4H5UzhfwDjl5+VUKlFjk7tQVeQ== X-Received: by 2002:a05:622a:c8c:b0:51c:2340:bbdb with SMTP id d75a77b69052e-52d18fd59d2mr61313441cf.26.1786124912297; Fri, 07 Aug 2026 10:48:32 -0700 (PDT) Received: from ziepe.ca ([142.166.156.215]) by smtp.gmail.com with ESMTPSA id d75a77b69052e-52d166297b7sm17136451cf.26.2026.08.07.10.48.31 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 07 Aug 2026 10:48:31 -0700 (PDT) Received: from jgg by wakko with local (Exim 4.97) (envelope-from ) id 1wsOgI-00000001ArF-3vYh; Fri, 07 Aug 2026 14:48:30 -0300 Date: Fri, 7 Aug 2026 14:48:30 -0300 From: Jason Gunthorpe To: Leon Romanovsky Cc: Chenguang Zhao , andrew+netdev@lunn.ch, davem@davemloft.net, edumazet@google.com, kuba@kernel.org, pabeni@redhat.com, linux-rdma@vger.kernel.org, netdev@vger.kernel.org, tariqt@nvidia.com, mbloch@nvidia.com, dtatulea@nvidia.com, shayd@nvidia.com, moshe@nvidia.com, Chenguang Zhao Subject: Re: [PATCH rdma-next v3] RDMA/mlx5: quiesce CQ polling before device shutdown on reboot Message-ID: <20260807174830.GF200537@ziepe.ca> References: <20260715082307.1593915-1-chenguang.zhao@linux.dev> <20260716084211.GC70906@unreal> <20260716105008.GD357857@unreal> Precedence: bulk X-Mailing-List: linux-rdma@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20260716105008.GD357857@unreal> On Thu, Jul 16, 2026 at 01:50:08PM +0300, Leon Romanovsky wrote: > On Thu, Jul 16, 2026 at 05:03:37PM +0800, Chenguang Zhao wrote: > > Hi, Leon > > reboot -f skips orderly shutdown and goes directly to: > > > > kernel_restart_prepare() -> device_shutdown() -> mlx5 shutdown > > Upper layers may still hold live CQs, while ib-comp-wq keeps > > polling — a use-after-free race. > > The point is that this flow is neither RDMA- nor mlx5-specific, and it > works as expected. mlx5 shutdown() stops the FW/HW, while the kernel > stops and tears down the running threads. I'm pretty sure we are missing stuff in the driver to synchronize everything while it is doing a tear down or health recovery. Ripping a driver out from under a still-running subsystem is hard, this does seem like a real bug, but I also don't think the shutdown handler is the right place to fix it.. What was the actual UAF scenario, we should take a look at that more directly? Jason