Problem: SDHC partition table corruption after suspend/resume
Solution: a Kernel compiled with CONFIG_MMC_UNSAFE_RESUME enabled.
Hi all,
I tried toying around with a quick SDHC card, and the original idea was to mount it as my home directory. The idea seemed a brilliant one, as it practically removed all the waitings involved with heavy firefox-caching or downloading large files, so the basic idea of "halving" the SSD-workload proved to work nicely, the overall responsiveness and speed of the system rose significantly. (if anyone has a good idea of how to benchmark these, let me know and I'm willing to try)
...but it was too good to be true. The problems arose when I first suspend my laptop, the ext2 system on the SD-card is getting corrupted due to unclean unmounts and the card reader device resetting itself (hint: partition table corrupts)
I tried some solutions, that I thought were potential ones. Had the reader been an USB-device, this is exactly what USB_PERSIST would have solved, but that doesn't help with PCIE-devices, obviously.
I tried also adding the sdhci, mmc_block and mmc_core modules to acpi whitelist, no luck -> the home partition cannot be unmounted for reasons unknown to me, and the laptop doesn't suspend at all.
Has anyone else tried to mount vital parts of the system, such as the root partition to an SD card and has managed to get the machine actually working this way? If so, could we try writing a small howto on the subject?
Solution: a Kernel compiled with CONFIG_MMC_UNSAFE_RESUME enabled.
Hi all,
I tried toying around with a quick SDHC card, and the original idea was to mount it as my home directory. The idea seemed a brilliant one, as it practically removed all the waitings involved with heavy firefox-caching or downloading large files, so the basic idea of "halving" the SSD-workload proved to work nicely, the overall responsiveness and speed of the system rose significantly. (if anyone has a good idea of how to benchmark these, let me know and I'm willing to try)
...but it was too good to be true. The problems arose when I first suspend my laptop, the ext2 system on the SD-card is getting corrupted due to unclean unmounts and the card reader device resetting itself (hint: partition table corrupts)
I tried some solutions, that I thought were potential ones. Had the reader been an USB-device, this is exactly what USB_PERSIST would have solved, but that doesn't help with PCIE-devices, obviously.
I tried also adding the sdhci, mmc_block and mmc_core modules to acpi whitelist, no luck -> the home partition cannot be unmounted for reasons unknown to me, and the laptop doesn't suspend at all.
Has anyone else tried to mount vital parts of the system, such as the root partition to an SD card and has managed to get the machine actually working this way? If so, could we try writing a small howto on the subject?