2003-03-11 19:19:11

by Adrian Bunk

Subject: Would the real 82801E_9 please stand up. (fwd)

The issue described in Dave's mail below is still present:

2.4.21-pre5-ac1:
#define PCI_DEVICE_ID_INTEL_82801E_9 0x2459
#define PCI_DEVICE_ID_INTEL_82801E_11 0x245B


2.5.64-ac3:
#define PCI_DEVICE_ID_INTEL_82801E_9 0x245b
#define PCI_DEVICE_ID_INTEL_82801E_11 PCI_DEVICE_ID_INTEL_82801E_9


Jens:
The patch that did the
#define PCI_DEVICE_ID_INTEL_82801E_11 PCI_DEVICE_ID_INTEL_82801E_9
in 2.5 was sent by you, could you comment on this issue?

TIA
Adrian



----- Forwarded message from Dave Jones <[email protected]> -----

Date: Mon, 24 Jun 2002 23:58:52 +0200
From: Dave Jones <[email protected]>
To: Linux Kernel <[email protected]>
Subject: Would the real 82801E_9 please stand up.

Whilst syncing 2.4.19rc1, I spotted this problem
in pci_ids.h

2.4..

#define PCI_DEVICE_ID_INTEL_82801E_9 0x2459

2.5..

#define PCI_DEVICE_ID_INTEL_82801E_9 0x245b

In a word.. eughhh

Dave.

--
| Dave Jones. http://www.codemonkey.org.uk
| SuSE Labs

2003-03-11 19:24:10

by Jens Axboe

Subject: Re: Would the real 82801E_9 please stand up. (fwd)

On Tue, Mar 11 2003, Adrian Bunk wrote:
> The issue described in Dave's mail below is still present:
>
> 2.4.21-pre5-ac1:
> #define PCI_DEVICE_ID_INTEL_82801E_9 0x2459
> #define PCI_DEVICE_ID_INTEL_82801E_11 0x245B
>
>
> 2.5.64-ac3:
> #define PCI_DEVICE_ID_INTEL_82801E_9 0x245b
> #define PCI_DEVICE_ID_INTEL_82801E_11 PCI_DEVICE_ID_INTEL_82801E_9
>
>
> Jens:
> The patch that did the
> #define PCI_DEVICE_ID_INTEL_82801E_11 PCI_DEVICE_ID_INTEL_82801E_9
> in 2.5 was sent by you, could you comment on this issue?

This was during the nasty 2.4-ac IDE merge I did in 2.5, so I wouldn't
trust that particular change completely. The 2.4.21-pre5-ac1 is most
likely the correct one.

--
Jens Axboe

2003-03-11 21:24:29

by Andries Brouwer

Subject: Re: Would the real 82801E_9 please stand up. (fwd)

On Tue, Mar 11, 2003 at 08:29:45PM +0100, Adrian Bunk wrote:
> The issue described in Dave's mail below is still present:
>
> 2.4.21-pre5-ac1:
> #define PCI_DEVICE_ID_INTEL_82801E_9 0x2459
> #define PCI_DEVICE_ID_INTEL_82801E_11 0x245B
>
>
> 2.5.64-ac3:
> #define PCI_DEVICE_ID_INTEL_82801E_9 0x245b
> #define PCI_DEVICE_ID_INTEL_82801E_11 PCI_DEVICE_ID_INTEL_82801E_9

The 82801E has seven device IDs:
B1:D8:F0 2459 Ethernet Controller 0
B1:D9:F0 245d Ethernet Controller 1
D30:F0 245e PCI Hub
D31:F0 2450 LPC I/F
D31:F1 245b IDE
D31:F2 2452 USB
D31:F3 2453 SMBus

Clearly, the 2.5.64 uses are for IDE and must be 0x245b.
PCI_DEVICE_ID_INTEL_82801E_9 does not occur in 2.5.64,
apart from its definition.

Conclusion: 2.4 is right.

--- /linux/2.5/linux-2.5.64/linux/include/linux/pci_ids.h Wed Mar 5 10:47:31 2003
+++ /linux/2.5/linux-2.5.64a/linux/include/linux/pci_ids.h Tue Mar 11 22:31:53 2003
@@ -1831,8 +1831,8 @@
#define PCI_DEVICE_ID_INTEL_82801E_0 0x2450
#define PCI_DEVICE_ID_INTEL_82801E_2 0x2452
#define PCI_DEVICE_ID_INTEL_82801E_3 0x2453
-#define PCI_DEVICE_ID_INTEL_82801E_9 0x245b
-#define PCI_DEVICE_ID_INTEL_82801E_11 PCI_DEVICE_ID_INTEL_82801E_9
+#define PCI_DEVICE_ID_INTEL_82801E_9 0x2459
+#define PCI_DEVICE_ID_INTEL_82801E_11 0x245b
#define PCI_DEVICE_ID_INTEL_82801E_13 0x245d
#define PCI_DEVICE_ID_INTEL_82801E_14 0x245e
#define PCI_DEVICE_ID_INTEL_82801CA_0 0x2480

Andries