linux-kernel - [PATCH v7 11/15] agp/intel: Use CPU physical address, not bus address, for ioremap()

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20140107005630.10786.35580.stgit@bhelgaas-glaptop.roam.corp.google.com>
Date:	Mon, 06 Jan 2014 17:56:31 -0700
From:	Bjorn Helgaas <bhelgaas@...gle.com>
To:	linux-pci@...r.kernel.org, Daniel Vetter <daniel.vetter@...ll.ch>
Cc:	David Airlie <airlied@...ux.ie>, Yinghai Lu <yinghai@...nel.org>,
	Guo Chao <yan@...ux.vnet.ibm.com>, linux-kernel@...r.kernel.org
Subject: [PATCH v7 11/15] agp/intel: Use CPU physical address,
 not bus address, for ioremap()

In i810_setup(), i830_setup(), and i9xx_setup(), we use the result of
pci_bus_address() as an argument to ioremap() and to compute gtt_phys_addr.
These should use pci_resource_start() instead because we want the CPU
physical address, not the bus address.

If there were an AGP device behind a host bridge that translated addresses,
e.g., a PNP0A08 device with _TRA != 0, this would fix a bug.  I'm not aware
of any of those, but they are possible.

Signed-off-by: Bjorn Helgaas <bhelgaas@...gle.com>
---
 drivers/char/agp/intel-gtt.c |   14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c
index dd8b66a617dc..ad5da1ffcbe9 100644
--- a/drivers/char/agp/intel-gtt.c
+++ b/drivers/char/agp/intel-gtt.c
@@ -172,7 +172,7 @@ static void i8xx_destroy_pages(struct page *page)
 #define I810_GTT_ORDER 4
 static int i810_setup(void)
 {
-	u32 reg_addr;
+	phys_addr_t reg_addr;
 	char *gtt_table;
 
 	/* i81x does not preallocate the gtt. It's always 64kb in size. */
@@ -181,7 +181,7 @@ static int i810_setup(void)
 		return -ENOMEM;
 	intel_private.i81x_gtt_table = gtt_table;
 
-	reg_addr = pci_bus_address(intel_private.pcidev, I810_MMADR_BAR);
+	reg_addr = pci_resource_start(intel_private.pcidev, I810_MMADR_BAR);
 
 	intel_private.registers = ioremap(reg_addr, KB(64));
 	if (!intel_private.registers)
@@ -782,9 +782,9 @@ EXPORT_SYMBOL(intel_enable_gtt);
 
 static int i830_setup(void)
 {
-	u32 reg_addr;
+	phys_addr_t reg_addr;
 
-	reg_addr = pci_bus_address(intel_private.pcidev, I810_MMADR_BAR);
+	reg_addr = pci_resource_start(intel_private.pcidev, I810_MMADR_BAR);
 
 	intel_private.registers = ioremap(reg_addr, KB(64));
 	if (!intel_private.registers)
@@ -1102,10 +1102,10 @@ static void i965_write_entry(dma_addr_t addr,
 
 static int i9xx_setup(void)
 {
-	u32 reg_addr;
+	phys_addr_t reg_addr;
 	int size = KB(512);
 
-	reg_addr = pci_bus_address(intel_private.pcidev, I915_MMADR_BAR);
+	reg_addr = pci_resource_start(intel_private.pcidev, I915_MMADR_BAR);
 
 	intel_private.registers = ioremap(reg_addr, size);
 	if (!intel_private.registers)
@@ -1114,7 +1114,7 @@ static int i9xx_setup(void)
 	switch (INTEL_GTT_GEN) {
 	case 3:
 		intel_private.gtt_phys_addr =
-			pci_bus_address(intel_private.pcidev, I915_PTE_BAR);
+			pci_resource_start(intel_private.pcidev, I915_PTE_BAR);
 		break;
 	case 5:
 		intel_private.gtt_phys_addr = reg_addr + MB(2);

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/