Home » Mailing list archives » linux-kernel » 2007 » September » 18

[PATCH] pci: Fix e100 interrupt quirk

Previous thread: [GIT PULL] XFS update for 2.6.23 by Tim Shimmin on Tuesday, September 18, 2007 - 7:24 am. (1 message)

Next thread: What can be done to reduce the huge number of build fixes required to release an MM tree? by Miles Lane on Tuesday, September 18, 2007 - 8:23 am. (3 messages)
[view original message]
From: Valentine Barshak <vbarshak@...>
To: <linux-kernel@...>
Subject: [PATCH] pci: Fix e100 interrupt quirk
Date: Tuesday, September 18, 2007 - 7:17 am

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 <vbarshak@ru.mvista.com>

---


--- 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);


-
[view original message]
From: Andrew Morton <akpm@...>
To: Valentine Barshak <vbarshak@...>
Cc: <linux-kernel@...>, Auke Kok <auke-jan.h.kok@...>, <netdev@...>
Subject: Re: [PATCH] pci: Fix e100 interrupt quirk
Date: Thursday, September 20, 2007 - 2:37 am

Really?  So if pci_request_region() failed and !(command & PCI_COMMAND_MEMORY),

-
[view original message]
From: Valentine Barshak <vbarshak@...>
To: Andrew Morton <akpm@...>
Cc: <linux-kernel@...>, Auke Kok <auke-jan.h.kok@...>, <netdev@...>
Subject: Re: [PATCH] pci: Fix e100 interrupt quirk
Date: Thursday, September 20, 2007 - 7:33 am

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,


-
[view original message]
From: Jarek Poplawski <jarkao2@...>
To: Valentine Barshak <vbarshak@...>
Cc: <linux-kernel@...>
Subject: Re: [PATCH] pci: Fix e100 interrupt quirk
Date: Wednesday, September 19, 2007 - 6:43 am

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.

-
[view original message]
From: Valentine Barshak <vbarshak@...>
To: <linux-kernel@...>
Cc: <jarkao2@...>
Subject: [PATCH] pci: Fix e100 interrupt quirk
Date: Wednesday, September 19, 2007 - 8:36 am

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 <vbarshak@ru.mvista.com>

---


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);


-
[view original message]
From: Valentine Barshak <vbarshak@...>
To: Jarek Poplawski <jarkao2@...>
Cc: <linux-kernel@...>
Subject: Re: [PATCH] pci: Fix e100 interrupt quirk
Date: Wednesday, September 19, 2007 - 6:59 am

I'll split these 2 checks and submit new patch in a minute.


-
Previous thread: [GIT PULL] XFS update for 2.6.23 by Tim Shimmin on Tuesday, September 18, 2007 - 7:24 am. (1 message)

Next thread: What can be done to reduce the huge number of build fixes required to release an MM tree? by Miles Lane on Tuesday, September 18, 2007 - 8:23 am. (3 messages)