Jeff Garzik wrote:
It is unwise to free the struct before the ports are even detached.
Right, theoretically something bad could happen here (though not
likely). Here's a fix. Sorry for attaching with something so trivial,
but Thunderbird isn't very cooperative..
---
The suspend/resume change for sata_nv introduced a potential bug where
the hpriv structure could be used after it was freed in nv_remove_one.
Fix that.
Signed-off-by: Robert Hancock <hancockr@xxxxxxx>
---
Robert Hancock Saskatoon, SK, Canada
To email, remove "nospam" from hancockr@xxxxxxxxxxxxx
Home Page: http://www.roberthancock.com/
--- linux-2.6.19-rc6-mm2/drivers/ata/sata_nv.c 2006-12-11 22:13:26.000000000 -0600
+++ linux-2.6.19-rc6-mm2admafix/drivers/ata/sata_nv.c 2006-12-11 22:15:58.000000000 -0600
@@ -1555,8 +1555,8 @@ static void nv_remove_one (struct pci_de
struct ata_host *host = dev_get_drvdata(&pdev->dev);
struct nv_host_priv *hpriv = host->private_data;
- kfree(hpriv);
ata_pci_remove_one(pdev);
+ kfree(hpriv);
}
static int nv_pci_device_resume(struct pci_dev *pdev)
