Could you please _try_ and live in the real world? MTBF ain't got a thing to do with it. The article talks about disrupting the memory operation by various real world means such as directing heat or radiation onto the memory module housings. The vitual machine is running on top of an operating system which is running simultaneouly inside the same memory module. I believe that it would be virtually impossible to induce by any external means small numbers of bit errors in the memory without a 99% probablilty of first crashing the operating system. Maybe it would be possible in a lab environment, but not in the typical hacker-cracks-system environment. He pulls out the hair dry, gradually blows in some heat, and presto, the computer crashes.
Doesn't matter, since the attacker has no idea of physically where to apply the error producing agent on the memory module package's outside surface and since the agent most likely cannot be restricted to extremely small localized areas on the silicon ( heat distributes in all directions, shorting gates with radiation disrupts nearby gate operation ), there is a very high probability that the results will be to only crash the operating system and/or the hard drive.
we just ask the little monkeys inside the memory chip who are in charge of steering the data to guide our thermal rays to just the right CMOS gates used by the JVM process so as not to crash the computer...
Real practical exploit NOT.
Slashdot Mirror
Could you please _try_ and live in the real world? MTBF ain't got a thing to do with it. The article talks about disrupting the memory operation by various real world means such as directing heat or radiation onto the memory module housings. The vitual machine is running on top of an operating system which is running simultaneouly inside the same memory module. I believe that it would be virtually impossible to induce by any external means small numbers of bit errors in the memory without a 99% probablilty of first crashing the operating system. Maybe it would be possible in a lab environment, but not in the typical hacker-cracks-system environment. He pulls out the hair dry, gradually blows in some heat, and presto, the computer crashes.
Doesn't matter, since the attacker has no idea of physically where to apply the error producing agent on the memory module package's outside surface and since the agent most likely cannot be restricted to extremely small localized areas on the silicon ( heat distributes in all directions, shorting gates with radiation disrupts nearby gate operation ), there is a very high probability that the results will be to only crash the operating system and/or the hard drive.
we just ask the little monkeys inside the memory chip who are in charge of steering the data to guide our thermal rays to just the right CMOS gates used by the JVM process so as not to crash the computer... Real practical exploit NOT.