From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Justin T. Gibbs" Subject: scsi_malloc fragmentation in 2.4.X Date: Fri, 20 Dec 2002 13:31:58 -0700 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <70300000.1040416318@aslan.btc.adaptec.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-path: Received: from redfish.adaptec.com (redfish.adaptec.com [162.62.50.11]) by magic.adaptec.com (8.11.6+Sun/8.11.6) with ESMTP id gBKKW2j25474 for ; Fri, 20 Dec 2002 12:32:02 -0800 (PST) Received: from btc.btc.adaptec.com (btc.btc.adaptec.com [10.100.0.52]) by redfish.adaptec.com (8.8.8+Sun/8.8.8) with ESMTP id MAA26775 for ; Fri, 20 Dec 2002 12:32:01 -0800 (PST) Received: from [10.100.253.70] (aslan [10.100.253.70]) by btc.btc.adaptec.com (8.8.8+Sun/8.8.8) with ESMTP id NAA29716 for ; Fri, 20 Dec 2002 13:31:57 -0700 (MST) Content-Disposition: inline List-Id: linux-scsi@vger.kernel.org To: linux-scsi@vger.kernel.org Are there any plans to correct the scsi_malloc fragmentation problems in 2.4.X? It looks like this problem has been addressed in 2.5.X by using memory pools, but the current 2.4.X solution of just "backing off a bit" if scsi_malloc returns NULL, is much less than ideal. At least the user isn't assaulted by lots of "Running really short on DMA buffers" messages, but performance does suffer when the only reason you can't queue an I/O is that the pre-allocated S/G list memory is so poorly managed that the available memory can't be used. -- Justin