>We broke write-ahead with the BH_Dirty handling - it can mark the buffer
>clean without ever actually writing it out. I bet that's what you
Great!! :)) That almost sure was the bug!
BTW, without the WRITEA the LRU-page patch is running fine on UP for 3
hours and 17 minutes now with the memoryhog+bonnie testcase! :)) I'll
left it running all the night...
Since you given a sense to the corruption, tomorrow I'll try the LRU-page
patch on SMP as well :).
Thanks.
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/