From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: T10 PI offloading seems to be broken in iser/isert in 4.5/4.6-rc Date: Fri, 8 Apr 2016 17:14:56 -0700 Message-ID: <20160409001456.GA1752@infradead.org> References: <20160408221553.GA3716@infradead.org> <1460159290.901.5.camel@haakon3.risingtidesystems.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <1460159290.901.5.camel@haakon3.risingtidesystems.com> Sender: target-devel-owner@vger.kernel.org To: "Nicholas A. Bellinger" Cc: Christoph Hellwig , sagig@mellanox.com, linux-rdma@vger.kernel.org, target-devel@vger.kernel.org List-Id: linux-rdma@vger.kernel.org On Fri, Apr 08, 2016 at 04:48:10PM -0700, Nicholas A. Bellinger wrote: > It looks like the isert_set_dif_domain() assignments of > ib_sig_domain->sig.dif.* from dev->dev_attrib.block_size + friends is > being zeroed before making it into mlx5_ib_post_send() -> > IB_WR_REG_SIG_MR -> set_sig_umr_wr() -> set_sig_data_segment(). >>From my limited understanding that's not the case, even if the weird printk about the bad block size might suggest it. I think what we have is the TARGET_PROT_DIN_INSERT / TARGET_PROT_DOUT_STRIP case (probably the first), and thus sig_attrs->mem.sig_type is set to IB_SIG_TYPE_NONE, but for some reason the prot pointer is set, which will lead to this misleading printk. Does anyone know a recent working version of T10 PI in isert? There aren't any recent-ish changes that look related.