TA的每日心情 | 擦汗 2020-1-14 15:59 |
---|
签到天数: 1 天 [LV.1]初来乍到
|
2#

楼主 |
发表于 2007-12-18 21:50
|
只看该作者
The biggest problem with asynchronous resets is that they are asynchronous, both at the& _. |' t' V9 g; w) K7 F* y
assertion and at the de-assertion of the reset. The assertion is a non issue, the de-assertion is the0 q( R( N" g3 v
issue. If the asynchronous reset is released at or near the active clock edge of a flip-flop, the
1 Y0 J5 }6 f/ _4 {4 _ Q( q- X; m$ Noutput of the flip-flop could go metastable and thus the reset state of the ASIC could be lost.
* R: j2 s2 ^' W& ^, [, e* n$ }0 FAnother problem that an asynchronous reset can have, depending on its source, is spurious resets
# X6 H$ f6 [: I$ a5 {due to noise or glitches on the board or system reset. See section 8.0 for a possible solution to* S6 [( i' _3 \' c8 x
reset glitches. If this is a real problem in a system, then one might think that using synchronous" T% q1 q9 }. b, Z2 S. D
resets is the solution. A different but similar problem exists for synchronous resets if these2 j& u, M1 I* L) D( }
spurious reset pulses occur near a clock edge, the flip-flops can still go metastable (but this is
8 y0 C8 I) q6 D0 [4 P* qtrue of any data input that violates setup requirements). |
|