public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Chris Wright <chrisw@osdl.org>
To: linux-kernel@vger.kernel.org, stable@kernel.org
Cc: Justin Forbes <jmforbes@linuxtx.org>,
	Zwane Mwaikambo <zwane@arm.linux.org.uk>,
	"Theodore Ts'o" <tytso@mit.edu>,
	Randy Dunlap <rdunlap@xenotime.net>,
	Chuck Wolber <chuckw@quantumlinux.com>,
	torvalds@osdl.org, akpm@osdl.org, alan@lxorguk.ukuu.org.uk,
	olof@austin.ibm.com, dsd@gentoo.org
Subject: [patch 05/16] PPC64: Fix LPAR IOMMU setup code for p630
Date: Mon, 23 May 2005 16:21:03 -0700	[thread overview]
Message-ID: <20050523232103.GQ27549@shell0.pdx.osdl.net> (raw)
In-Reply-To: <20050523231529.GL27549@shell0.pdx.osdl.net>

Here's a fix to deal with p630 systems in LPAR mode.  They're to date the
only system that in some cases might lack a dma-window property for the
bus, but contain an overriding property in the device node for the specific
adapter/slot.  This makes the device setup code a bit more complex since it
needs to do some of the things that the bus setup code has already done.

Signed-off-by: Olof Johansson <olof@austin.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 arch/ppc64/kernel/pSeries_iommu.c |   55 +++++++++++++++++++++++++++++++++++++-
 1 files changed, 54 insertions(+), 1 deletion(-)

--- linux-2.6.11.10.orig/arch/ppc64/kernel/pSeries_iommu.c	2005-05-16 10:50:31.000000000 -0700
+++ linux-2.6.11.10/arch/ppc64/kernel/pSeries_iommu.c	2005-05-20 09:36:25.091359360 -0700
@@ -401,6 +401,8 @@
 	struct device_node *dn, *pdn;
 	unsigned int *dma_window = NULL;
 
+	DBG("iommu_bus_setup_pSeriesLP, bus %p, bus->self %p\n", bus, bus->self);
+
 	dn = pci_bus_to_OF_node(bus);
 
 	/* Find nearest ibm,dma-window, walking up the device tree */
@@ -455,6 +457,56 @@
 	}
 }
 
+static void iommu_dev_setup_pSeriesLP(struct pci_dev *dev)
+{
+	struct device_node *pdn, *dn;
+	struct iommu_table *tbl;
+	int *dma_window = NULL;
+
+	DBG("iommu_dev_setup_pSeriesLP, dev %p (%s)\n", dev, dev->pretty_name);
+
+	/* dev setup for LPAR is a little tricky, since the device tree might
+	 * contain the dma-window properties per-device and not neccesarily
+	 * for the bus. So we need to search upwards in the tree until we
+	 * either hit a dma-window property, OR find a parent with a table
+	 * already allocated.
+	 */
+	dn = pci_device_to_OF_node(dev);
+
+	for (pdn = dn; pdn && !pdn->iommu_table; pdn = pdn->parent) {
+		dma_window = (unsigned int *)get_property(pdn, "ibm,dma-window", NULL);
+		if (dma_window)
+			break;
+	}
+
+	/* Check for parent == NULL so we don't try to setup the empty EADS
+	 * slots on POWER4 machines.
+	 */
+	if (dma_window == NULL || pdn->parent == NULL) {
+		/* Fall back to regular (non-LPAR) dev setup */
+		DBG("No dma window for device, falling back to regular setup\n");
+		iommu_dev_setup_pSeries(dev);
+		return;
+	} else {
+		DBG("Found DMA window, allocating table\n");
+	}
+
+	if (!pdn->iommu_table) {
+		/* iommu_table_setparms_lpar needs bussubno. */
+		pdn->bussubno = pdn->phb->bus->number;
+
+		tbl = (struct iommu_table *)kmalloc(sizeof(struct iommu_table),
+						    GFP_KERNEL);
+
+		iommu_table_setparms_lpar(pdn->phb, pdn, tbl, dma_window);
+
+		pdn->iommu_table = iommu_init_table(tbl);
+	}
+
+	if (pdn != dn)
+		dn->iommu_table = pdn->iommu_table;
+}
+
 static void iommu_bus_setup_null(struct pci_bus *b) { }
 static void iommu_dev_setup_null(struct pci_dev *d) { }
 
@@ -479,13 +531,14 @@
 			ppc_md.tce_free	 = tce_free_pSeriesLP;
 		}
 		ppc_md.iommu_bus_setup = iommu_bus_setup_pSeriesLP;
+		ppc_md.iommu_dev_setup = iommu_dev_setup_pSeriesLP;
 	} else {
 		ppc_md.tce_build = tce_build_pSeries;
 		ppc_md.tce_free  = tce_free_pSeries;
 		ppc_md.iommu_bus_setup = iommu_bus_setup_pSeries;
+		ppc_md.iommu_dev_setup = iommu_dev_setup_pSeries;
 	}
 
-	ppc_md.iommu_dev_setup = iommu_dev_setup_pSeries;
 
 	pci_iommu_init();
 }

  parent reply	other threads:[~2005-05-23 23:26 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-05-23 23:15 [00/16] -stable review Chris Wright
2005-05-23 23:17 ` [patch 01/16] Fix get_unmapped_area sanity tests Chris Wright
2005-05-23 23:18 ` [patch 02/16] 3c59x: only put the device into D3 when we're actually using WOL Chris Wright
2005-05-25 17:30   ` Daniel Ritz
2005-05-23 23:19 ` [patch 03/16] [EBTABLES]: Fix smp race Chris Wright
2005-05-23 23:20 ` [patch 04/16] ext3: fix race between ext3 make block reservation and reservation window discard Chris Wright
2005-05-30 13:28   ` Rodrigo Steinmüller Wanderley
2005-05-31  6:23     ` Mingming Cao
2005-05-23 23:21 ` Chris Wright [this message]
2005-05-23 23:22 ` [patch 06/16] Fix matroxfb on big-endian hardware Chris Wright
2005-05-23 23:50   ` Al Viro
2005-05-24  1:17     ` Chris Wright
2005-05-24 10:15       ` Petr Vandrovec
2005-05-23 23:24 ` [patch 07/16] ide-disk: Fix LBA8 DMA Chris Wright
2005-05-23 23:25 ` [patch 08/16] [ROSE]: Fix minor security hole Chris Wright
2005-05-23 23:26 ` [patch 09/16] usbaudio: prevent oops & dead keyboard on usb unplugging Chris Wright
2005-05-23 23:27 ` [patch 10/16] usbusx2y: " Chris Wright
2005-05-23 23:28 ` [patch 11/16] USB: fix bug in visor driver with throttle/unthrottle causing oopses Chris Wright
2005-05-23 23:29 ` [patch 12/16] x86_64: check if ptrace RIP is canonical Chris Wright
2005-05-23 23:30 ` [patch 13/16] x86_64: Fix canonical checking for segment registers in ptrace Chris Wright
2005-05-23 23:30 ` [patch 14/16] x86_64: Add a guard page at the end of the 47bit address space Chris Wright
2005-05-23 23:31 ` [patch 15/16] x86_64: When checking vmalloc mappings don't use pte_page Chris Wright
2005-05-23 23:32 ` [patch 16/16] x86_64: Don't look up struct page pointer of physical address in iounmap Chris Wright

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20050523232103.GQ27549@shell0.pdx.osdl.net \
    --to=chrisw@osdl.org \
    --cc=akpm@osdl.org \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=chuckw@quantumlinux.com \
    --cc=dsd@gentoo.org \
    --cc=jmforbes@linuxtx.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=olof@austin.ibm.com \
    --cc=rdunlap@xenotime.net \
    --cc=stable@kernel.org \
    --cc=torvalds@osdl.org \
    --cc=tytso@mit.edu \
    --cc=zwane@arm.linux.org.uk \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox