From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Mokryn Subject: Highmem SCSI driver Date: Thu, 06 Nov 2003 18:05:03 +0200 Sender: linux-kernel-owner@vger.kernel.org Message-ID: <3FAA712F.2030409@il.marvell.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Return-path: To: linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@vger.kernel.org List-Id: linux-scsi@vger.kernel.org We are trying to test 64-bit PCI DMA for a SCSI driver on a Xeon box, RH9 2.4.20-8 bigmem kernel, 6GB RAM. The machine shows 6GB in top, we set highmem_io in the driver, PCI DMA mask covers 64-bit range, etc. Of course we're trying to make sure that the system does not create bounce buffers unnecessarily. On a 64-bit box (AMD64) everything works as expected. On the Xeon, no matter what we try, we never see I/Os mapped above 4GB. Any ideas on how we can drive I/Os mapped above 4GB down to our driver? Thanks, -Mark From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arjan van de Ven Subject: Re: Highmem SCSI driver Date: Thu, 06 Nov 2003 18:01:47 +0100 Sender: linux-kernel-owner@vger.kernel.org Message-ID: <1068138107.5234.6.camel@laptop.fenrus.com> References: <3FAA712F.2030409@il.marvell.com> Reply-To: arjanv@redhat.com Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-CCwyfmWjAPPFahdBEn8q" Return-path: In-Reply-To: <3FAA712F.2030409@il.marvell.com> To: Mark Mokryn Cc: linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@vger.kernel.org List-Id: linux-scsi@vger.kernel.org --=-CCwyfmWjAPPFahdBEn8q Content-Type: text/plain Content-Transfer-Encoding: quoted-printable On Thu, 2003-11-06 at 17:05, Mark Mokryn wrote: > We are trying to test 64-bit PCI DMA for a SCSI driver on a Xeon box,=20 > RH9 2.4.20-8 bigmem kernel, 6GB RAM. > The machine shows 6GB in top, we set highmem_io in the driver, PCI DMA=20 > mask covers 64-bit range, etc. can you give an URL to the driver so that we can see why it breaks ? --=-CCwyfmWjAPPFahdBEn8q Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQA/qn57xULwo51rQBIRAl73AJ4iREs+pKaE4FnosuvXo2lUnpdhCwCdG1Oy j+INZ5WOJB7OZhZqIkzBmGw= =Jh1t -----END PGP SIGNATURE----- --=-CCwyfmWjAPPFahdBEn8q-- From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from il.marvell.com ([10.2.1.184]) by mail.galileo.co.il (8.12.6/8.12.6) with ESMTP id hA6G8iGn018892 for ; Thu, 6 Nov 2003 18:08:45 +0200 (IST) Message-ID: <3FAA7237.1030006@il.marvell.com> Date: Thu, 06 Nov 2003 18:09:27 +0200 From: Mark Mokryn MIME-Version: 1.0 Subject: Highmem SCSI driver Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: linux-mm@kvack.org List-ID: We are trying to test 64-bit PCI DMA for a SCSI driver on a Xeon box, RH9 2.4.20-8 bigmem kernel, 6GB RAM. The machine shows 6GB in top, we set highmem_io in the driver, PCI DMA mask covers 64-bit range, etc. Of course we're trying to make sure that the system does not create bounce buffers unnecessarily. On a 64-bit box (AMD64) everything works as expected. On the Xeon, no matter what we try, we never see I/Os mapped above 4GB. Any ideas on how we can drive I/Os mapped above 4GB down to our driver? Thanks, -Mark -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: aart@kvack.org