linux-kernel - RE: [patch] PCI: disable MSI on more ATI NorthBridges

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Tue, 23 Oct 2007 18:56:03 +0800
From:	"Shane Huang" <Shane.Huang@....com>
To:	"David Gaarenstroom" <david.gaarenstroom@...il.com>,
	linux-kernel@...r.kernel.org
cc:	"Linas Vepstas" <linas@...tin.ibm.com>, davem@...emloft.net,
	gregkh@...e.de, htejun@...il.com, brice.goglin@...il.com,
	linux-pci@...ey.karlin.mff.cuni.cz, jgarzik@...ox.com,
	"Shane Huang" <Shane.Huang@....com>
Subject: RE: [patch] PCI: disable MSI on more ATI NorthBridges


> If the pci_intx change will be applied to the SATA driver, can it be
> applied for the ATI USB-HCDs too? See
> http://lkml.org/lkml/2006/12/21/47 for more details. That should help
> most of the ATI MSI quirks. It helped me (Acer Aspire 502x laptop with
> ATI RS480/SB400).

I checked the USB MSI problem above(also has reported to our hardware
team),
the cause seems same as our SB700 SATA controller MSI problem.

I also did some small USB MSI debug on our SB700 board with 2.6.23-rc5:
The USB part of the second patch(attached at the end of this mail)
does not work for SB700 USB EHCI/OHCI controllers, no matter INTx is
enabled or disabled before enter MSI. The USB host controllers are
always using IO-APIC, which is different from SB400. I don't know why.

[root@...alhost ~]# cat /proc/interrupts 
           CPU0       CPU1  
17:          0        133   IO-APIC-fasteoi   ohci_hcd:usb1,
ohci_hcd:usb2, ehci_hcd:usb6
18:          1          2   IO-APIC-fasteoi   ohci_hcd:usb3,
ohci_hcd:usb4, ohci_hcd:usb5
19:          0          0   IO-APIC-fasteoi   ehci_hcd:usb7

Also I wonder why the USB MSI patch is not added into kernel at last?
Will it lead to other bugs?


Thanks
Best Regards
Shane


======> USB part of the second patch in lkml.org/lkml/2006/12/21/47
diff -uprdN linux/drivers/usb/core/hcd-pci.c
linux/drivers/usb/core/hcd-pci.c
--- linux/drivers/usb/core/hcd-pci.c	2006-12-16 13:34:57.000000000
-0800
+++ linux/drivers/usb/core/hcd-pci.c	2006-12-16 13:57:09.000000000
-0800
@@ -69,6 +69,7 @@ int usb_hcd_pci_probe (struct pci_dev *d
 
 	if (pci_enable_device (dev) < 0)
 		return -ENODEV;
+	pci_enable_msi(dev);
 	dev->current_state = PCI_D0;
 	dev->dev.power.power_state = PMSG_ON;
 	
@@ -139,6 +140,7 @@ int usb_hcd_pci_probe (struct pci_dev *d
 		release_region (hcd->rsrc_start, hcd->rsrc_len);
  err2:
 	usb_put_hcd (hcd);
+	pci_disable_msi (dev);
  err1:
 	pci_disable_device (dev);
 	dev_err (&dev->dev, "init %s fail, %d\n", pci_name(dev),
retval);
@@ -177,6 +179,7 @@ void usb_hcd_pci_remove (struct pci_dev 
 		release_region (hcd->rsrc_start, hcd->rsrc_len);
 	}
 	usb_put_hcd (hcd);
+	pci_disable_msi(dev);
 	pci_disable_device(dev);
 }
 EXPORT_SYMBOL (usb_hcd_pci_remove);
@@ -391,6 +394,7 @@ int usb_hcd_pci_resume (struct pci_dev *
 			"can't re-enable after resume, %d!\n", retval);
 		return retval;
 	}
+	pci_enable_msi (dev);
 	pci_set_master (dev);
 	pci_restore_state (dev);






-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/