qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: qemu-devel@nongnu.org
Cc: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>,
	Richard Henderson <rth@twiddle.net>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	David Gibson <david@gibson.dropbear.id.au>
Subject: [Qemu-devel] [PATCH 11/13] iommu: Allow PCI to use IOMMU infrastructure
Date: Thu, 10 May 2012 14:49:05 +1000	[thread overview]
Message-ID: <1336625347-10169-12-git-send-email-benh@kernel.crashing.org> (raw)
In-Reply-To: <1336625347-10169-1-git-send-email-benh@kernel.crashing.org>

From: David Gibson <david@gibson.dropbear.id.au>

This patch adds some hooks to let PCI devices and busses use the new IOMMU
infrastructure.  When IOMMU support is enabled, each PCI device now
contains a DMAContext * which is used by the pci_dma_*() wrapper functions.

By default, the contexts are initialized to NULL, assuming no IOMMU.
However the platform or host bridge code which sets up the PCI bus can use
pci_setup_iommu() to set a function which will determine the correct
DMAContext for a given PCI device.

Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 hw/pci.c           |    9 +++++++++
 hw/pci.h           |    9 +++++++--
 hw/pci_internals.h |    2 ++
 3 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/hw/pci.c b/hw/pci.c
index b706e69..8901c01 100644
--- a/hw/pci.c
+++ b/hw/pci.c
@@ -761,6 +761,9 @@ static PCIDevice *do_pci_register_device(PCIDevice *pci_dev, PCIBus *bus,
         return NULL;
     }
     pci_dev->bus = bus;
+    if (bus->dma_context_fn) {
+        pci_dev->dma = bus->dma_context_fn(bus, bus->dma_context_opaque, devfn);
+    }
     pci_dev->devfn = devfn;
     pstrcpy(pci_dev->name, sizeof(pci_dev->name), name);
     pci_dev->irq_state = 0;
@@ -2004,6 +2007,12 @@ static void pci_device_class_init(ObjectClass *klass, void *data)
     k->bus_info = &pci_bus_info;
 }
 
+void pci_setup_iommu(PCIBus *bus, PCIDMAContextFunc fn, void *opaque)
+{
+    bus->dma_context_fn = fn;
+    bus->dma_context_opaque = opaque;
+}
+
 static TypeInfo pci_device_type_info = {
     .name = TYPE_PCI_DEVICE,
     .parent = TYPE_DEVICE,
diff --git a/hw/pci.h b/hw/pci.h
index 7e36c53..8c96438 100644
--- a/hw/pci.h
+++ b/hw/pci.h
@@ -175,6 +175,7 @@ typedef struct PCIDeviceClass {
 
 struct PCIDevice {
     DeviceState qdev;
+
     /* PCI config space */
     uint8_t *config;
 
@@ -196,6 +197,7 @@ struct PCIDevice {
     uint32_t devfn;
     char name[64];
     PCIIORegion io_regions[PCI_NUM_REGIONS];
+    DMAContext *dma;
 
     /* do not access the following fields */
     PCIConfigReadFunc *config_read;
@@ -316,6 +318,10 @@ int pci_read_devaddr(Monitor *mon, const char *addr, int *domp, int *busp,
 
 void pci_device_deassert_intx(PCIDevice *dev);
 
+typedef DMAContext *(*PCIDMAContextFunc)(PCIBus *, void *, int);
+
+void pci_setup_iommu(PCIBus *bus, PCIDMAContextFunc fn, void *opaque);
+
 static inline void
 pci_set_byte(uint8_t *config, uint8_t val)
 {
@@ -552,8 +558,7 @@ static inline uint32_t pci_config_size(const PCIDevice *d)
 /* DMA access functions */
 static inline DMAContext *pci_dma_context(PCIDevice *dev)
 {
-    /* Stub for when we have no PCI iommu support */
-    return NULL;
+    return dev->dma;
 }
 
 static inline int pci_dma_rw(PCIDevice *dev, dma_addr_t addr,
diff --git a/hw/pci_internals.h b/hw/pci_internals.h
index 96690b7..983594b 100644
--- a/hw/pci_internals.h
+++ b/hw/pci_internals.h
@@ -16,6 +16,8 @@ extern struct BusInfo pci_bus_info;
 
 struct PCIBus {
     BusState qbus;
+    PCIDMAContextFunc dma_context_fn;
+    void *dma_context_opaque;
     uint8_t devfn_min;
     pci_set_irq_fn set_irq;
     pci_map_irq_fn map_irq;
-- 
1.7.9.5

  parent reply	other threads:[~2012-05-10  4:50 UTC|newest]

Thread overview: 90+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-05-10  4:48 [Qemu-devel] [PATCH 00/13] IOMMU infrastructure Benjamin Herrenschmidt
2012-05-10  4:48 ` [Qemu-devel] [PATCH 01/13] Better support for dma_addr_t variables Benjamin Herrenschmidt
2012-05-10  4:48 ` [Qemu-devel] [PATCH 02/13] Implement cpu_physical_memory_zero() Benjamin Herrenschmidt
2012-05-15  0:42   ` Anthony Liguori
2012-05-15  1:23     ` David Gibson
2012-05-15  2:03       ` Anthony Liguori
2012-05-10  4:48 ` [Qemu-devel] [PATCH 03/13] iommu: Add universal DMA helper functions Benjamin Herrenschmidt
2012-05-10  4:48 ` [Qemu-devel] [PATCH 04/13] usb-ohci: Use " Benjamin Herrenschmidt
2012-05-10  4:48 ` [Qemu-devel] [PATCH 05/13] iommu: Make sglists and dma_bdrv helpers use new universal DMA helpers Benjamin Herrenschmidt
2012-05-10  4:49 ` [Qemu-devel] [PATCH 06/13] ide/ahci: Use universal DMA helper functions Benjamin Herrenschmidt
2012-05-21  1:51   ` [Qemu-devel] [PATCH 06/13 - UPDATED] " Benjamin Herrenschmidt
2012-05-10  4:49 ` [Qemu-devel] [PATCH 07/13] usb: Convert usb_packet_{map, unmap} to universal DMA helpers Benjamin Herrenschmidt
2012-05-10  4:49 ` [Qemu-devel] [PATCH 08/13] iommu: Introduce IOMMU emulation infrastructure Benjamin Herrenschmidt
2012-05-15  0:49   ` Anthony Liguori
2012-05-15  1:42     ` David Gibson
2012-05-15  2:03       ` Anthony Liguori
2012-05-15  2:32         ` Benjamin Herrenschmidt
2012-05-15  2:50           ` Anthony Liguori
2012-05-15  3:02             ` Benjamin Herrenschmidt
2012-05-15 14:02               ` Anthony Liguori
2012-05-15 21:55                 ` Benjamin Herrenschmidt
2012-05-15 22:02                   ` Anthony Liguori
2012-05-15 23:08                     ` Benjamin Herrenschmidt
2012-05-15 23:58                       ` Anthony Liguori
2012-05-16  0:41                         ` Benjamin Herrenschmidt
2012-05-16  0:54                           ` Anthony Liguori
2012-05-16  1:20                             ` Benjamin Herrenschmidt
2012-05-16 19:36                               ` Anthony Liguori
2012-05-10  4:49 ` [Qemu-devel] [PATCH 09/13] iommu: Add facility to cancel in-use dma memory maps Benjamin Herrenschmidt
2012-05-10  4:49 ` [Qemu-devel] [PATCH 10/13] pseries: Convert sPAPR TCEs to use generic IOMMU infrastructure Benjamin Herrenschmidt
2012-05-10  4:49 ` Benjamin Herrenschmidt [this message]
2012-05-10  4:49 ` [Qemu-devel] [PATCH 12/13] pseries: Implement IOMMU and DMA for PAPR PCI devices Benjamin Herrenschmidt
2012-05-10  4:49 ` [Qemu-devel] [PATCH 13/13] iommu: Add a memory barrier to DMA RW function Benjamin Herrenschmidt
2012-05-15  0:52   ` Anthony Liguori
2012-05-15  1:11     ` Benjamin Herrenschmidt
2012-05-15  1:44     ` David Gibson
2012-05-16  4:35       ` Benjamin Herrenschmidt
2012-05-16  5:51         ` David Gibson
2012-05-16 19:39         ` Anthony Liguori
2012-05-16 21:10           ` Benjamin Herrenschmidt
2012-05-16 21:12             ` Benjamin Herrenschmidt
2012-05-17  0:07           ` Benjamin Herrenschmidt
2012-05-17  0:24             ` Benjamin Herrenschmidt
2012-05-17  0:52               ` [Qemu-devel] [RFC/PATCH] Add a memory barrier to guest memory access functions Benjamin Herrenschmidt
2012-05-17  2:28                 ` Anthony Liguori
2012-05-17  2:44                   ` Benjamin Herrenschmidt
2012-05-17 22:09                     ` Anthony Liguori
2012-05-18  1:04                       ` David Gibson
2012-05-18  1:16                       ` Benjamin Herrenschmidt
2012-05-17  3:35                   ` David Gibson
2012-05-18  6:53               ` [Qemu-devel] [PATCH 13/13] iommu: Add a memory barrier to DMA RW function Paolo Bonzini
2012-05-18  8:18                 ` Benjamin Herrenschmidt
2012-05-18  8:57                   ` Paolo Bonzini
2012-05-18 22:26                     ` Benjamin Herrenschmidt
2012-05-19  7:24                       ` Paolo Bonzini
2012-05-20 21:36                         ` Benjamin Herrenschmidt
2012-05-21  1:56                           ` [Qemu-devel] [PATCH] Add a memory barrier to guest memory access functions Benjamin Herrenschmidt
2012-05-21  8:11                             ` Paolo Bonzini
2012-05-21  8:31                               ` Michael S. Tsirkin
2012-05-21  8:58                                 ` Benjamin Herrenschmidt
2012-05-21  9:07                                   ` Benjamin Herrenschmidt
2012-05-21  9:16                                     ` Benjamin Herrenschmidt
2012-05-21  9:34                                       ` Michael S. Tsirkin
2012-05-21  9:53                                         ` Benjamin Herrenschmidt
2012-05-21 10:31                                           ` Michael S. Tsirkin
2012-05-21 11:45                                             ` Benjamin Herrenschmidt
2012-05-21 12:18                                               ` Michael S. Tsirkin
2012-05-21 15:16                                                 ` Paolo Bonzini
2012-05-21 21:58                                                 ` [Qemu-devel] [PATCH] Add a memory barrier to guest memory access function Benjamin Herrenschmidt
2012-05-21 22:22                                                   ` Michael S. Tsirkin
2012-05-21 22:56                                                     ` Benjamin Herrenschmidt
2012-05-22  5:11                                                       ` Michael S. Tsirkin
2012-05-22  0:00                                                     ` Benjamin Herrenschmidt
2012-05-22  4:19                                                   ` Rusty Russell
2012-05-21 22:18                                 ` [Qemu-devel] [PATCH] Add a memory barrier to guest memory access functions Anthony Liguori
2012-05-21 22:26                                   ` Benjamin Herrenschmidt
2012-05-21 22:31                                     ` Anthony Liguori
2012-05-21 22:44                                       ` Michael S. Tsirkin
2012-05-21 23:02                                         ` Benjamin Herrenschmidt
2012-05-22  4:34                                         ` [Qemu-devel] [PATCH] Add a memory barrier to DMA functions Benjamin Herrenschmidt
2012-05-22  4:51                                           ` Benjamin Herrenschmidt
2012-05-22  7:17                                           ` Benjamin Herrenschmidt
2012-05-22 11:14                                             ` Michael S. Tsirkin
2012-05-22 11:41                                               ` Benjamin Herrenschmidt
2012-05-22 12:03                                                 ` Michael S. Tsirkin
2012-05-22 21:24                                                   ` Benjamin Herrenschmidt
2012-05-22 21:40                                                   ` Anthony Liguori
2012-05-21 22:37                                   ` [Qemu-devel] [PATCH] Add a memory barrier to guest memory access functions Michael S. Tsirkin
2012-05-15  0:52 ` [Qemu-devel] [PATCH 00/13] IOMMU infrastructure Anthony Liguori
  -- strict thread matches above, loose matches on Subject: below --
2012-06-19  6:39 [Qemu-devel] [PATCH 00/13] iommu series Benjamin Herrenschmidt
2012-06-19  6:39 ` [Qemu-devel] [PATCH 11/13] iommu: Allow PCI to use IOMMU infrastructure Benjamin Herrenschmidt

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=1336625347-10169-12-git-send-email-benh@kernel.crashing.org \
    --to=benh@kernel.crashing.org \
    --cc=david@gibson.dropbear.id.au \
    --cc=eduard.munteanu@linux360.ro \
    --cc=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=rth@twiddle.net \
    /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;
as well as URLs for NNTP newsgroup(s).