Skip to contentSkip to conversation

Commit 7a192ec is build breaking

PATCH Russell King 4 participants 7 messages Mar 28, 2009 – Apr 4, 2009

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

The commit:

platform driver: fix incorrect use of 'platform_bus_type' with 'struct device_driver'

contains this:

-static int __exit pxa2xx_flash_remove(struct device *dev) +static int __exit pxa2xx_flash_remove(struct platform_device *dev)
...

-       .remove         = __exit_p(pxa2xx_flash_remove),
+       .remove         = __devexit_p(pxa2xx_flash_remove),

which leads to the following build error:

pxa2xx_flash_remove' referenced in section .data' of drivers/built-in.o: defined in discarded section `.exit.text' of drivers/built-in.o

This is not the only instance of it in this patch - all __exit_p's touched by this patch have been converted to __devexit_p's without regard to the original function.

Let's revert this change and, if we are going to convert functions to be __devexit/__devinit, lets have that as a _separate_ patch doing just that change.

---
 arch/mips/basler/excite/excite_iodev.c |    2 +-
 drivers/mtd/maps/pxa2xx-flash.c        |    2 +-
 drivers/mtd/nand/excite_nandflash.c    |    3 +--
 3 files changed, 3 insertions(+), 4 deletions(-)
arch/mips/basler/excite/excite_iodev.c +1 -1
index dfbfd7e..2ab56ea 100644
@@ -71,7 +71,7 @@ static struct platform_driver iodev_driver = {
 		.owner		= THIS_MODULE,
 	},
 	.probe		= iodev_probe,
-	.remove		= __devexit_p(iodev_remove),
+	.remove		= __exit_p(iodev_remove),
 };
 
 
drivers/mtd/maps/pxa2xx-flash.c +1 -1
index e9026cb..bf933a9 100644
@@ -177,7 +177,7 @@ static struct platform_driver pxa2xx_flash_driver = {
 		.owner		= THIS_MODULE,
 	},
 	.probe		= pxa2xx_flash_probe,
-	.remove		= __devexit_p(pxa2xx_flash_remove),
+	.remove		= __exit_p(pxa2xx_flash_remove),
 	.suspend	= pxa2xx_flash_suspend,
 	.resume		= pxa2xx_flash_resume,
 	.shutdown	= pxa2xx_flash_shutdown,
drivers/mtd/nand/excite_nandflash.c +1 -1
index 72446fb..630b4d6 100644
@@ -224,7 +224,7 @@ static struct platform_driver excite_nand_driver = {
 		.owner		= THIS_MODULE,
 	},
 	.probe = excite_nand_probe,
-	.remove = __devexit_p(excite_nand_remove)
+	.remove = __exit_p(excite_nand_remove)
 };
 
 static int __init excite_nand_init(void)
Signature · 1 line
Russell King
#1 lore raw

Ping?

Quoted 74 lines · Russell King

The commit:

platform driver: fix incorrect use of 'platform_bus_type' with 'struct device_driver'

contains this:

(quoted diff) +2 -2
-static int __exit pxa2xx_flash_remove(struct device *dev)
+static int __exit pxa2xx_flash_remove(struct platform_device *dev)
...
-       .remove         = __exit_p(pxa2xx_flash_remove),
+       .remove         = __devexit_p(pxa2xx_flash_remove),

which leads to the following build error:

pxa2xx_flash_remove' referenced in section .data' of drivers/built-in.o: defined in discarded section `.exit.text' of drivers/built-in.o

This is not the only instance of it in this patch - all __exit_p's touched by this patch have been converted to __devexit_p's without regard to the original function.

Let's revert this change and, if we are going to convert functions to be __devexit/__devinit, lets have that as a _separate_ patch doing just that change.

---
 arch/mips/basler/excite/excite_iodev.c |    2 +-
 drivers/mtd/maps/pxa2xx-flash.c        |    2 +-
 drivers/mtd/nand/excite_nandflash.c    |    3 +--
 3 files changed, 3 insertions(+), 4 deletions(-)
arch/mips/basler/excite/excite_iodev.c +1 -1
index dfbfd7e..2ab56ea 100644
@@ -71,7 +71,7 @@ static struct platform_driver iodev_driver = {
 		.owner		= THIS_MODULE,
 	},
 	.probe		= iodev_probe,
-	.remove		= __devexit_p(iodev_remove),
+	.remove		= __exit_p(iodev_remove),
 };
 
 
drivers/mtd/maps/pxa2xx-flash.c +1 -1
index e9026cb..bf933a9 100644
@@ -177,7 +177,7 @@ static struct platform_driver pxa2xx_flash_driver = {
 		.owner		= THIS_MODULE,
 	},
 	.probe		= pxa2xx_flash_probe,
-	.remove		= __devexit_p(pxa2xx_flash_remove),
+	.remove		= __exit_p(pxa2xx_flash_remove),
 	.suspend	= pxa2xx_flash_suspend,
 	.resume		= pxa2xx_flash_resume,
 	.shutdown	= pxa2xx_flash_shutdown,
drivers/mtd/nand/excite_nandflash.c +1 -1
index 72446fb..630b4d6 100644
@@ -224,7 +224,7 @@ static struct platform_driver excite_nand_driver = {
 		.owner		= THIS_MODULE,
 	},
 	.probe = excite_nand_probe,
-	.remove = __devexit_p(excite_nand_remove)
+	.remove = __exit_p(excite_nand_remove)
 };
 
 static int __init excite_nand_init(void)
Signature · 2 lines
-- 
Russell King
Signature · 3 lines
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:
#2 lore raw
In reply to Russell King
Quoted 16 lines · Russell King

The commit:

platform driver: fix incorrect use of 'platform_bus_type' with 'struct device_driver'

contains this:

(quoted diff) +2 -2
-static int __exit pxa2xx_flash_remove(struct device *dev)
+static int __exit pxa2xx_flash_remove(struct platform_device *dev)
...
-       .remove         = __exit_p(pxa2xx_flash_remove),
+       .remove         = __devexit_p(pxa2xx_flash_remove),

which leads to the following build error:

pxa2xx_flash_remove' referenced in section .data' of drivers/built-in.o: defined in discarded section `.exit.text' of drivers/built-in.o

Quoted 7 lines

This is not the only instance of it in this patch - all __exit_p's touched by this patch have been converted to __devexit_p's without regard to the original function.

Let's revert this change and, if we are going to convert functions to be __devexit/__devinit, lets have that as a _separate_ patch doing just that change.

Ick, Ming, any thoughts? I'll revert your patch as it looks like it is incorrect in this way.

thanks,

greg k-h

#3 lore raw

2009/4/4 Greg KH <[email protected]>:

Quoted 21 lines · Russell King
Quoted 20 lines

The commit:

    platform driver: fix incorrect use of 'platform_bus_type' with 'struct device_driver'

contains this:

(quoted diff) +2 -2
-static int __exit pxa2xx_flash_remove(struct device *dev)
+static int __exit pxa2xx_flash_remove(struct platform_device *dev)
...
-       .remove         = __exit_p(pxa2xx_flash_remove),
+       .remove         = __devexit_p(pxa2xx_flash_remove),

which leads to the following build error:

pxa2xx_flash_remove' referenced in section .data' of drivers/built-in.o: defined in discarded section `.exit.text' of drivers/built-in.o

This is not the only instance of it in this patch - all __exit_p's touched by this patch have been converted to __devexit_p's without regard to the original function.

Quoted 6 lines
Quoted 3 lines

Let's revert this change and, if we are going to convert functions to be __devexit/__devinit, lets have that as a _separate_ patch doing just that change.

Ick, Ming, any thoughts?  I'll revert your patch as it looks like it is incorrect in this way.

OK, revert the patch ,please.
I'll resend a patch to fix incorrect use of 'platform_bus_type' , which does not touch
__exit_p.

Thanks!

Quoted 5 lines

thanks,

greg k-h

Signature · 1 line
Lei Ming
#4 lore raw

Note that since I was sending a new set of ARM fixes a couple of days ago and there didn't seem to be anyone listening to my emails, I fixed the PXA flash driver myself. Linus now has that fix.

Quoted 46 lines · Ming Lei

2009/4/4 Greg KH <[email protected]>:

Quoted 21 lines · Russell King
Quoted 20 lines

The commit:

    platform driver: fix incorrect use of 'platform_bus_type' with 'struct device_driver'

contains this:

(quoted diff) +2 -2
-static int __exit pxa2xx_flash_remove(struct device *dev)
+static int __exit pxa2xx_flash_remove(struct platform_device *dev)
...
-       .remove         = __exit_p(pxa2xx_flash_remove),
+       .remove         = __devexit_p(pxa2xx_flash_remove),

which leads to the following build error:

pxa2xx_flash_remove' referenced in section .data' of drivers/built-in.o: defined in discarded section `.exit.text' of drivers/built-in.o

This is not the only instance of it in this patch - all __exit_p's touched by this patch have been converted to __devexit_p's without regard to the original function.

Quoted 6 lines
Quoted 3 lines

Let's revert this change and, if we are going to convert functions to be __devexit/__devinit, lets have that as a _separate_ patch doing just that change.

Ick, Ming, any thoughts?  I'll revert your patch as it looks like it is incorrect in this way.

OK, revert the patch ,please.
I'll resend a patch to fix incorrect use of 'platform_bus_type' , which does not touch
__exit_p.

Thanks!

Quoted 5 lines

thanks,

greg k-h

Signature · 1 line
Lei Ming
Signature · 3 lines
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:
#5 lore raw
Quoted 3 lines · Russell King

Note that since I was sending a new set of ARM fixes a couple of days ago and there didn't seem to be anyone listening to my emails, I fixed the PXA flash driver myself. Linus now has that fix.

Sorry, I was out of the country for a week.

So, we don't need to revert the original change now?

Again, sorry for the delay.

greg k-h

#6 lore raw
Quoted 8 lines · Greg KH
Quoted 3 lines · Russell King

Note that since I was sending a new set of ARM fixes a couple of days ago and there didn't seem to be anyone listening to my emails, I fixed the PXA flash driver myself. Linus now has that fix.

Sorry, I was out of the country for a week.

So, we don't need to revert the original change now?

Well, I didn't apply the non-ARM bits of the patch. So the way forward is to either revert both mine and the original patch and apply a new one (which means a total of three commits) or just apply the remainder of my patch I sent at the top of this thread.

Signature · 3 lines
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of: