Comment 3 for bug 303844

$ dmesg|grep -i -e ^mmc -e ^sdhci
sdhci: Secure Digital Host Controller Interface Driver
sdhci: Copyright(c) Pierre Ossman
sdhci-pci 0000:06:04.2: SDHCI controller found [1524:0550] (rev 1)
sdhci-pci 0000:06:04.2: PCI INT B -> GSI 17 (level, low) -> IRQ 17
mmc0: SDHCI controller on PCI [0000:06:04.2] using PIO
sdhci-pci 0000:06:04.4: SDHCI controller found [1524:0551] (rev 1)
sdhci-pci 0000:06:04.4: enabling device (0000 -> 0002)
sdhci-pci 0000:06:04.4: PCI INT B -> GSI 17 (level, low) -> IRQ 17
mmc1: SDHCI controller on PCI [0000:06:04.4] using PIO
$ lspci -ns 0:6:4.2; lspci -ns 0:6:4.4
06:04.2 0805: 1524:0550 (rev 01)
06:04.4 0501: 1524:0551 (rev 01)
$ dmesg|tail -n5
pci 0000:00:02.0: setting latency timer to 64
[drm] Initialized i915 1.6.0 20060119 on minor 0
mmc0: new SDHC card at address 0002
mmcblk0: mmc0:0002 SD4GB 4018176KiB
 mmcblk0:
$

No problems, except for access speed (given that the kernel has decided to use PIO, that's not surprising).

Test kernels are a custom-compiled 2.6.27.7 and a kernel from a Debian live USB image (linux-image-2.6.26-1-686 2.6.26-5, I think), although I've not seen any problems other than with kernels 2.6.18 and 2.6.19 (ref. the lkml link).

The hardware does appear to be the same as in bug 99648. The CardBus controller is listed as CB-712/4 and has PCI ID 1524:1412; the next three IDs match, but are listed as "(rev 01)". This could be important :-)