From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eli Cohen Subject: Re: [Patch 1/2] IB/mlx5: Implementation of PCI error handler Date: Thu, 13 Mar 2014 18:03:59 +0200 Message-ID: <20140313160359.GB28066@mtldesk30> References: <20140312034219.637916521@linux.vnet.ibm.com> <20140312034512.065218504@linux.vnet.ibm.com> <1394649252.23624.36.camel@deadeye.wl.decadent.org.uk> <20140313064521.GH20224@mtldesk30> <5321CAD3.2070301@linux.vnet.ibm.com> <20140313154002.GA28066@mtldesk30> <5321D412.1040501@linux.vnet.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <5321D412.1040501@linux.vnet.ibm.com> Sender: netdev-owner@vger.kernel.org To: Carol Soto Cc: Ben Hutchings , eli@mellanox.com, roland@kernel.org, sean.hefty@intel.com, hal.rosenstock@gmail.com, linux-rdma@vger.kernel.org, netdev@vger.kernel.org, brking@linux.vnet.ibm.com List-Id: linux-rdma@vger.kernel.org On Thu, Mar 13, 2014 at 10:51:46AM -0500, Carol Soto wrote: > > The motivation to reduce that timeout is that if there is a process > in the middle of a HW command > in the middle of the PCI error, I probably did not want to wait 2hrs > since the command will never complete > since the card is dead. Now you are right, I forgot the case of big > memory registration where commands can > take longer than that. Do you have an idea of what is the longest > time that a command can take in mlx5? > There is no guranteed time for command completions. With current driver/firmware, registration of 4.5 GB can take around 2 minutes; for larger regions it can take even more time. As I mentioned earlier, I will soon send a patch that reduces registation times for large regions.