信息是如何从闪存进入缓存的？

任何计算机体系结构教科书都将阐明处理器缓存如何工作的机制。当缓存丢失发生时，缓存控制器将从内存中填充该缓存行。这是完全透明的。PIC32MZ中的缓存允许锁定单独的缓存行，但是缓存数组不能直接寻址。使用“Cache”汇编指令执行缓存操作，读取MIPS架构手册以获取详细信息。锁定缓存将消除任何闪存等待状态，并减少总线争用，但是您可以通过从RAM执行代码来获得几乎相同的性能改进，而不会增加复杂性。锁定高速缓存行也可以减少其他地方的性能。原则是相同的，但是实现细节不同。例如，一些处理器允许部分或全部缓存用作暂存RAM。



      以下为原文

    Any computer architecture textbook will decribe the mechanics of how processor caches work.
 

When a cache miss occurs, the cache controller will fill that cache line from memory. This happens completely transparently.
 

The cache in the PIC32MZ allows individual cache lines to be locked, but the cache arrays are not directly addressable. Cache manipulation is performed using the "cache" assembly instruction, read the MIPS architecture manuals for details. Locking the cache would remove any flash wait states, and reduce bus contention, but you can get almost the same performance improvement by executing code from RAM, without the added complexity. Locking cache lines can also reduce performance elsewhere.
 

The principles are the same, but the implementation details differ. For example, some processors allow some or all of the cache to be used as scratchpad RAM.

谢谢你的帮助。我发现了一些关于缓存如何工作的好文章，但是他们只是浏览了一下表面，我想知道代码最初是如何进入的。



      以下为原文

    Thanks for the help. I found some good articles about how the cache works, but they only skimmed the surface and I wanted to know how code actually got in there in the first place.
 
Appreciate the explanation,
 
Judd

MX预取Cache M1 L1高速缓存



      以下为原文

    MX Prefetch Cache
 
MZ L1 Cache