>Sorry it took me so long (washing dishes after christmas dinner takes much
>longer thank breaking Linux ramdisk :) - I just reproduced it properly so
>here are the steps:
You are triggering a known bug that is present also in 2.2.x. Basically
the problem is that the protected data is considered as freeable by the
balance_dirty() stuff (or equivalent mark_buffer_dirty stuff in 2.2.x).
That's not a ramdisk problem but more a buffer cache bug.
The hash lookup to take the backend memory in the regular buffer cache is
very fine. I had the same idea in mind (after I rejected the idea of the
global vmalloc) but never had the time to implement it :).
Andrea
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/