PCI memory space may have a 64-bit offset on some architectures
(for example, PowerPC 440) and the actual PCI memory address
has to fixed up (an offset to PCI mem space shuld be added)
before remapping. So, pci_iomap should be used instead of
reading and remapping PCI BAR directly. This has been tested
on Sequoia PowerPC 440EPx board.
Signed-off-by: Valentine Barshak <[email protected]>
---
--- linux-2.6.orig/drivers/pci/quirks.c 2007-09-04 21:15:43.000000000 +0400
+++ linux-2.6.bld/drivers/pci/quirks.c 2007-09-05 20:46:14.000000000 +0400
@@ -1444,9 +1444,9 @@
static void __devinit quirk_e100_interrupt(struct pci_dev *dev)
{
u16 command;
- u32 bar;
u8 __iomem *csr;
u8 cmd_hi;
+ int rc;
switch (dev->device) {
/* PCI IDs taken from drivers/net/e100.c */
@@ -1476,16 +1476,17 @@
* re-enable them when it's ready.
*/
pci_read_config_word(dev, PCI_COMMAND, &command);
- pci_read_config_dword(dev, PCI_BASE_ADDRESS_0, &bar);
- if (!(command & PCI_COMMAND_MEMORY) || !bar)
+ rc = pci_request_region(dev, 0, "e100_quirk");
+
+ if (!(command & PCI_COMMAND_MEMORY) || (rc < 0))
return;
- csr = ioremap(bar, 8);
+ csr = pci_iomap(dev, 0, 8);
if (!csr) {
printk(KERN_WARNING "PCI: Can't map %s e100 registers\n",
pci_name(dev));
- return;
+ goto e100_quirk_exit;
}
cmd_hi = readb(csr + 3);
@@ -1495,7 +1496,9 @@
writeb(1, csr + 3);
}
- iounmap(csr);
+ pci_iounmap(dev, csr);
+e100_quirk_exit:
+ pci_release_region(dev, 0);
}
DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_ANY_ID, quirk_e100_interrupt);
On 18-09-2007 13:17, Valentine Barshak wrote:
> PCI memory space may have a 64-bit offset on some architectures
> (for example, PowerPC 440) and the actual PCI memory address
> has to fixed up (an offset to PCI mem space shuld be added)
> before remapping. So, pci_iomap should be used instead of
> reading and remapping PCI BAR directly. This has been tested
> on Sequoia PowerPC 440EPx board.
>
> Signed-off-by: Valentine Barshak <[email protected]>
> ---
>
> --- linux-2.6.orig/drivers/pci/quirks.c 2007-09-04 21:15:43.000000000 +0400
> +++ linux-2.6.bld/drivers/pci/quirks.c 2007-09-05 20:46:14.000000000 +0400
> @@ -1444,9 +1444,9 @@
> static void __devinit quirk_e100_interrupt(struct pci_dev *dev)
> {
> u16 command;
> - u32 bar;
> u8 __iomem *csr;
> u8 cmd_hi;
> + int rc;
>
> switch (dev->device) {
> /* PCI IDs taken from drivers/net/e100.c */
> @@ -1476,16 +1476,17 @@
> * re-enable them when it's ready.
> */
> pci_read_config_word(dev, PCI_COMMAND, &command);
> - pci_read_config_dword(dev, PCI_BASE_ADDRESS_0, &bar);
>
> - if (!(command & PCI_COMMAND_MEMORY) || !bar)
> + rc = pci_request_region(dev, 0, "e100_quirk");
> +
> + if (!(command & PCI_COMMAND_MEMORY) || (rc < 0))
> return;
I didn't look at this too much, but isn't something like:
if (rc >= 0)
goto e100_quirk_exit;
needed before this return?
Regards,
Jarek P.
Jarek Poplawski wrote:
> On 18-09-2007 13:17, Valentine Barshak wrote:
>> PCI memory space may have a 64-bit offset on some architectures
>> (for example, PowerPC 440) and the actual PCI memory address
>> has to fixed up (an offset to PCI mem space shuld be added)
>> before remapping. So, pci_iomap should be used instead of
>> reading and remapping PCI BAR directly. This has been tested
>> on Sequoia PowerPC 440EPx board.
>>
>> Signed-off-by: Valentine Barshak <[email protected]>
>> ---
>>
>> --- linux-2.6.orig/drivers/pci/quirks.c 2007-09-04 21:15:43.000000000 +0400
>> +++ linux-2.6.bld/drivers/pci/quirks.c 2007-09-05 20:46:14.000000000 +0400
>> @@ -1444,9 +1444,9 @@
>> static void __devinit quirk_e100_interrupt(struct pci_dev *dev)
>> {
>> u16 command;
>> - u32 bar;
>> u8 __iomem *csr;
>> u8 cmd_hi;
>> + int rc;
>>
>> switch (dev->device) {
>> /* PCI IDs taken from drivers/net/e100.c */
>> @@ -1476,16 +1476,17 @@
>> * re-enable them when it's ready.
>> */
>> pci_read_config_word(dev, PCI_COMMAND, &command);
>> - pci_read_config_dword(dev, PCI_BASE_ADDRESS_0, &bar);
>>
>> - if (!(command & PCI_COMMAND_MEMORY) || !bar)
>> + rc = pci_request_region(dev, 0, "e100_quirk");
>> +
>> + if (!(command & PCI_COMMAND_MEMORY) || (rc < 0))
>> return;
>
> I didn't look at this too much, but isn't something like:
> if (rc >= 0)
> goto e100_quirk_exit;
> needed before this return?
I'll split these 2 checks and submit new patch in a minute.
Thanks.
>
> Regards,
> Jarek P.
PCI memory space may have a 64-bit offset on some architectures
(for example, PowerPC 440) and the actual PCI memory address
has to fixed up (an offset to PCI mem space shuld be added)
before remapping. So, pci_iomap should be used instead of
reading and remapping PCI BAR directly. This has been tested
on Sequoia PowerPC 440EPx board.
Signed-off-by: Valentine Barshak <[email protected]>
---
diff -ruN linux-2.6.orig/drivers/pci/quirks.c linux-2.6/drivers/pci/quirks.c
--- linux-2.6.orig/drivers/pci/quirks.c 2007-09-18 15:32:48.000000000 +0400
+++ linux-2.6/drivers/pci/quirks.c 2007-09-19 15:57:26.000000000 +0400
@@ -1444,7 +1444,6 @@
static void __devinit quirk_e100_interrupt(struct pci_dev *dev)
{
u16 command;
- u32 bar;
u8 __iomem *csr;
u8 cmd_hi;
@@ -1476,16 +1475,18 @@
* re-enable them when it's ready.
*/
pci_read_config_word(dev, PCI_COMMAND, &command);
- pci_read_config_dword(dev, PCI_BASE_ADDRESS_0, &bar);
- if (!(command & PCI_COMMAND_MEMORY) || !bar)
+ if (!(command & PCI_COMMAND_MEMORY))
return;
- csr = ioremap(bar, 8);
+ if (pci_request_region(dev, 0, "e100_quirk"))
+ return;
+
+ csr = pci_iomap(dev, 0, 8);
if (!csr) {
printk(KERN_WARNING "PCI: Can't map %s e100 registers\n",
pci_name(dev));
- return;
+ goto e100_quirk_exit;
}
cmd_hi = readb(csr + 3);
@@ -1495,7 +1496,9 @@
writeb(1, csr + 3);
}
- iounmap(csr);
+ pci_iounmap(dev, csr);
+e100_quirk_exit:
+ pci_release_region(dev, 0);
}
DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_ANY_ID, quirk_e100_interrupt);
On Tue, 18 Sep 2007 15:17:37 +0400 Valentine Barshak <[email protected]> wrote:
> PCI memory space may have a 64-bit offset on some architectures
> (for example, PowerPC 440) and the actual PCI memory address
> has to fixed up (an offset to PCI mem space shuld be added)
> before remapping. So, pci_iomap should be used instead of
> reading and remapping PCI BAR directly. This has been tested
> on Sequoia PowerPC 440EPx board.
>
> Signed-off-by: Valentine Barshak <[email protected]>
> ---
>
> --- linux-2.6.orig/drivers/pci/quirks.c 2007-09-04 21:15:43.000000000 +0400
> +++ linux-2.6.bld/drivers/pci/quirks.c 2007-09-05 20:46:14.000000000 +0400
> @@ -1444,9 +1444,9 @@
> static void __devinit quirk_e100_interrupt(struct pci_dev *dev)
> {
> u16 command;
> - u32 bar;
> u8 __iomem *csr;
> u8 cmd_hi;
> + int rc;
>
> switch (dev->device) {
> /* PCI IDs taken from drivers/net/e100.c */
> @@ -1476,16 +1476,17 @@
> * re-enable them when it's ready.
> */
> pci_read_config_word(dev, PCI_COMMAND, &command);
> - pci_read_config_dword(dev, PCI_BASE_ADDRESS_0, &bar);
>
> - if (!(command & PCI_COMMAND_MEMORY) || !bar)
> + rc = pci_request_region(dev, 0, "e100_quirk");
> +
> + if (!(command & PCI_COMMAND_MEMORY) || (rc < 0))
> return;
Really? So if pci_request_region() failed and !(command & PCI_COMMAND_MEMORY),
we leak the region? So the next call to this function will fail?
> - csr = ioremap(bar, 8);
> + csr = pci_iomap(dev, 0, 8);
> if (!csr) {
> printk(KERN_WARNING "PCI: Can't map %s e100 registers\n",
> pci_name(dev));
> - return;
> + goto e100_quirk_exit;
> }
>
> cmd_hi = readb(csr + 3);
> @@ -1495,7 +1496,9 @@
> writeb(1, csr + 3);
> }
>
> - iounmap(csr);
> + pci_iounmap(dev, csr);
> +e100_quirk_exit:
> + pci_release_region(dev, 0);
> }
> DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_ANY_ID, quirk_e100_interrupt);
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
Andrew Morton wrote:
> On Tue, 18 Sep 2007 15:17:37 +0400 Valentine Barshak <[email protected]> wrote:
>
>> PCI memory space may have a 64-bit offset on some architectures
>> (for example, PowerPC 440) and the actual PCI memory address
>> has to fixed up (an offset to PCI mem space shuld be added)
>> before remapping. So, pci_iomap should be used instead of
>> reading and remapping PCI BAR directly. This has been tested
>> on Sequoia PowerPC 440EPx board.
>>
>> Signed-off-by: Valentine Barshak <[email protected]>
>> ---
>>
>> --- linux-2.6.orig/drivers/pci/quirks.c 2007-09-04 21:15:43.000000000 +0400
>> +++ linux-2.6.bld/drivers/pci/quirks.c 2007-09-05 20:46:14.000000000 +0400
>> @@ -1444,9 +1444,9 @@
>> static void __devinit quirk_e100_interrupt(struct pci_dev *dev)
>> {
>> u16 command;
>> - u32 bar;
>> u8 __iomem *csr;
>> u8 cmd_hi;
>> + int rc;
>>
>> switch (dev->device) {
>> /* PCI IDs taken from drivers/net/e100.c */
>> @@ -1476,16 +1476,17 @@
>> * re-enable them when it's ready.
>> */
>> pci_read_config_word(dev, PCI_COMMAND, &command);
>> - pci_read_config_dword(dev, PCI_BASE_ADDRESS_0, &bar);
>>
>> - if (!(command & PCI_COMMAND_MEMORY) || !bar)
>> + rc = pci_request_region(dev, 0, "e100_quirk");
>> +
>> + if (!(command & PCI_COMMAND_MEMORY) || (rc < 0))
>> return;
>
> Really? So if pci_request_region() failed and !(command & PCI_COMMAND_MEMORY),
> we leak the region? So the next call to this function will fail?
>
I've split command and request region checks and submitted new patch:
http://lkml.org/lkml/2007/9/19/106
Please, take a look,
Thanks,
Valentine.
>
>> - csr = ioremap(bar, 8);
>> + csr = pci_iomap(dev, 0, 8);
>> if (!csr) {
>> printk(KERN_WARNING "PCI: Can't map %s e100 registers\n",
>> pci_name(dev));
>> - return;
>> + goto e100_quirk_exit;
>> }
>>
>> cmd_hi = readb(csr + 3);
>> @@ -1495,7 +1496,9 @@
>> writeb(1, csr + 3);
>> }
>>
>> - iounmap(csr);
>> + pci_iounmap(dev, csr);
>> +e100_quirk_exit:
>> + pci_release_region(dev, 0);
>> }
>> DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_ANY_ID, quirk_e100_interrupt);
>>
>> -
>> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>> the body of a message to [email protected]
>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>> Please read the FAQ at http://www.tux.org/lkml/