Continuous signature monitoring: low-cost concurrent detection of processor control errors

Wilken, Kent; Shen, John Paul

doi:10.1109/43.55193

Cited by 127 publications

(50 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 1 illustrates the CFG concept by a simple example. Thus, in the first step of signature checking, which is based upon the CFG, the compiler pre-computes the signatures associated with each node of the CFG, and then either embeds those signatures into the original codes [12,14,16,17,19,22], or provides that information directly to the watchdog [18,21]. It should be noted that at this step, we …”

Section: Control Flow Errorsmentioning

confidence: 99%

“…From an instruction set architecture standpoint, an abstraction of program execution behavior can be based on various considerations which include control flow, memory and I/O access, and object type and range [11,12]. We first examine how to protect against control flow errors (those which cause a processor to violate the correct sequencing of instructions).…”

Section: Control Flow Errorsmentioning

confidence: 99%

“…Various signature checking techniques have been proposed in the recent past [12,[14][15][16][17][18][19][20][21][22]. Basically, there are two steps in signature checking: compile-time signature generation followed by run-time signature verification.…”

Section: Control Flow Errorsmentioning

confidence: 99%

“…Although the second step of both the assigned-signature checking algorithm and the derived-signature checking algorithm are essentially the same, assigned signature checking techniques have two major drawbacks: the need for registers to hold signatures and the performance overhead due to the need to execute extra instructions related to the assigned-signature checking [12,22]. For example, Oh et al [22] have shown that the overhead in terms of code size ranges from 26.6 to 61.9% while the overhead in terms of execution time ranges from 16.2 to 58.1%.…”

mentioning

confidence: 99%

“…Conversely, the second technique, derived-signature control flow checking [12,14,16,17,21], derives signatures from the nodes themselves. For example, we derive a checksum from the binary code of the instruction inside a node and then use that checksum as the corresponding node signature.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Tolerating Radiation-Induced Transient Faults in Modern Processors

Gaudiot

2009

Int J Parallel Prog

View full text Add to dashboard Cite

As MOS device sizes continue shrinking, lower charges, for example those charges carried by single ionizing particles of naturally occurring radiation, are sufficient to upset the functioning of complex modern microprocessors. In order to handle these inevitable errors, designs should include fault-tolerant features so that the processors can continue to correctly perform despite the occurrence of errors. The main goal of this work is to develop architecture mechanisms to protect processors against the effect of such radiation-induced transient faults. It should first be noted that, from a program execution perspective, many faults manifest themselves as control flow errors that cause processors to violate the correct sequencing of instructions. We present here at first a basic compile-time signature assignment algorithm and describe a novel approach to improve the fault detection coverage of the basic algorithm. Moreover, to allow the processor to efficiently check the run-time sequence and detect control flow errors, we introduce an on-chip assigned-signature checker which is capable of executing three additional instructions (SIC, SIJ, SIJC). Second, since the very concept of simultaneous multi-threading (SMT) provides the necessary redundancy, some proposals have been made to run two copies of the same thread on top of SMT platforms in order to detect and correct soft errors. This allows, upon detection of an error, the rolling back of the processor state to a known safe point, and then a retry of the instructions, thereby effecting a completely error-free execution. This paper has focused on two crucial implementation issues introduced by this

show abstract