[PATCH] fix error management in 8250_pci.c pciserial_resume_one

[PATCH] fix error management in 8250_pci.c pciserial_resume_one

[Stefano Brivio]

  CC      drivers/serial/8250_pci.o
drivers/serial/8250_pci.c: In function 'pciserial_resume_one':
drivers/serial/8250_pci.c:1830: warning: ignoring return value of 'pci_enable_device', declared with attribute warn_unused_result

This patch fixes it.

Signed-off-by: Stefano Brivio <stefano.brivio@polimi.it>
---

--- drivers/serial/8250_pci.c.orig	2007-02-13 11:04:03.579805618 +0100
+++ drivers/serial/8250_pci.c	2007-02-13 11:12:52.128454010 +0100
@@ -1819,6 +1819,7 @@
 static int pciserial_resume_one(struct pci_dev *dev)
 {
 	struct serial_private *priv = pci_get_drvdata(dev);
+	int rc;
 
 	pci_set_power_state(dev, PCI_D0);
 	pci_restore_state(dev);
@@ -1827,7 +1828,9 @@
 		/*
 		 * The device may have been disabled.  Re-enable it.
 		 */
-		pci_enable_device(dev);
+		rc = pci_enable_device(dev);
+		if (rc)
+			return rc;
 
 		pciserial_resume_ports(priv);
 	}

Re: [PATCH] fix error management in 8250_pci.c pciserial_resume_one

[Russell King]

On Tue, Feb 13, 2007 at 11:30:10AM +0100, Stefano Brivio wrote:
>   CC      drivers/serial/8250_pci.o
> drivers/serial/8250_pci.c: In function 'pciserial_resume_one':
> drivers/serial/8250_pci.c:1830: warning: ignoring return value of 'pci_enable_device', declared with attribute warn_unused_result
> 
> This patch fixes it.

Oh for god sake, this does not fix anything.  It creates more problems
than it solves.

Next time, check the mailing list archives and you'll see this is not
the first time this has come up.

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

Re: [PATCH] fix error management in 8250_pci.c pciserial_resume_one

[Alan]

On Tue, 13 Feb 2007 11:30:10 +0100
Stefano Brivio <stefano.brivio@polimi.it> wrote:

>   CC      drivers/serial/8250_pci.o
> drivers/serial/8250_pci.c: In function 'pciserial_resume_one':
> drivers/serial/8250_pci.c:1830: warning: ignoring return value of 'pci_enable_device', declared with attribute warn_unused_result
> 
> This patch fixes it.

NAK. as ith the past several times people proposed this. The bug is in
the pci_enable_device header being must_check