Bug 13775
| Summary: | Support for Marvell 88SE6102 IDE controller | ||
|---|---|---|---|
| Product: | IO/Storage | Reporter: | Jean Delvare (jdelvare) |
| Component: | IDE | Assignee: | Alan (alan) |
| Status: | REJECTED INVALID | ||
| Severity: | enhancement | CC: | alan, htejun, tj |
| Priority: | P1 | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Kernel Version: | 2.6.30.1 | Subsystem: | |
| Regression: | No | Bisected commit-id: | |
|
Description
Jean Delvare
2009-07-15 07:39:57 UTC
Should be sufficient to add PCI_DEVICE(0x11AB, 0x6102) to the pata_marvell.c driver. You may also need to change the following just above that list to include both 6101/2
if (pdev->device == 0x6101)
ppi[1] = &ata_dummy_port_info;
just needs someone with hw to test
I managed to boot the board off a USB drive, so I could run lspci. To my great surprise, the PCI device ID of the Marvell controller is 11ab:6101. This means that the device is _already_ supported by the pata_marvell driver, at least in theory. I double-checked the top marking of the chip and it really reads: 88SE6102. Now the question is: is there actually a PCI device with ID 11ab:6101 out there, or is the pci.ids database just wrong? Searching the web for 11ab:6102 doesn't return anything, so I suspect the latter. Hmm, my previous comment was supposed to read: "is there actually a PCI device with ID 11ab:610_2_ out there", of course. The problem, then, is why the hell doesn't the controller work? Does the kernel complain about anything during boot? I booted the board on an old test system hard disk drive. It could as well be a boot loader (old-style IDE vs. PATA maybe?) or kernel configuration problem. I plan to find a CD-ROM drive and try to install a recent distribution on the machine later today and see how it goes. I'll report when I have more information. Unfortunately I have some unrelated hardware problems on the system in question so it might take some time before I can report progress. As it turns out, I may not have access to the system in question for a long time and odds are that the kernel supports the Marvell 88SE6102 IDE controller just fine. So I'm closing this bug as invalid for the time being. I will reopen it later if really needed. I was finally able to boot the board again, and it turns out that the Marvell 88SE6102 IDE controller is supported just fine by the Linux kernel (2.6.34, driver pata_marvell). Sorry for the noise. |