Name: 88SE6111 1-port PATA133(IDE) and 1-port SATA II Controllers
I don't think that this ID actually exists. This seems to me to be the same issue as with 6101/6102.
I own a "P35 Neo2" ("MS-7345") board from MSI which has – according to the manual – an "88SE6111" chip with 1 SATA and 1 PATA port.
Looking at the "dmesg" output, there is also no evidence that there might actually be 2 ports with one of them disabled.
I unfortunately can't have a look at the chip itsself, because it's hidden below interconnected heatsinks which are risky to remove.
So I believe that I actually have an "88SE6111" with the ID "6121" and that this is always the case.
This device works well with the "pata_marvell" Linux kernel driver. I'll add further evidence below.
Unfortunately, the article mentioned by the original author of this ID entry has been deleted just a few days ago.
Fortunately, Wikipedia is free, so I found the seemingly unchanged information here:
… and there:
It's probably only an assumption of its author that the chips' numbers always match their IDs.
"pata_marvell" supports 2 more IDs today, but still not "6102" and "6111":
I think therefore, that the "88SE6111" and "88SE6121" both share the ID "11ab:6111",
while the "88SE6101" and "88SE6102" both have got the same ID "11ab:6101".
Note that SATA devices may be auto-detected by the device class + prog-if if they use the ahci driver and those do not need to have an ID in the kernel. But given this device also has PATA function you are likely right.
More evidence at https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=500982
The Marvell 88SE* are weird. My 88SE9123 PCIe (2 SATA ports, no PATA) (ID: 1b4b:9123:dc93:600e) isn't handeled by the AHCI driver.
I don't know why… Possible reasons without any proof so far:
1) It identifies itsself as a RAID bus controller .
2) Whenever "pata_marvell" is loaded (in my case for a PATA port on the mainboard) this chip can't be handeled by the generic AHCI module. Evidence: some Linux marvell driver source code comments.