2011年7月1日 星期五

RU v4.89

This is still a minor bug fix version. The hanging problem is still not solved yet. (Fixed @v4.92)

1. Display PCIe base address in command line while detecting
Please report the address when it hangs.

2. Fixed F6: list PCI will have last page empty sometimes.

7 則留言:

banova 提到...
作者已經移除這則留言。
banova 提到...

Hi James,

I found there might be no rule about the RU489 report with this symptom.

What I got is as below.

Our project:
S30(one Processor) hang at
Scanning PCIe ECAM base = F2700000

D30(Two Processor) hang at
Scanning PCIe ECAM base = D0300000

And checked some function Enable/Disable

Found that
"Enable" the "Above 4G Decoding" in Setup Menu
will solve this symptom.

Thanks !!

James Wang 提到...

Hi Banova,

Yes that's why I can't solve it in a perfect way.

Thanks for the clue. I may need to find another way to get the PCIe base if this can't be resolved. Or ask user to enable 4GB+ decoding.

Thanks!

匿名 提到...

Could you implement the algorithm in the 4G memory search.
It's much faster than the brutal way and simple to implement.

http://en.wikipedia.org/wiki/Boyer%E2%80%93Moore%E2%80%93Horspool_algorithm

James Wang 提到...

Hello Anonymous,

Yes RU search performance is very poor because it doesn't use any algorithm. I will check out the Boyer–Moore–Horspool algorithm and try to implement it.

P.S. I just think that no one use the search function in my RU. XD

匿名 提到...

Indeed, sometimes I want to search 0x55, 0xAA, but it seems that I can't do so. Is that possible to implement it also?

Thanks for your brilliant work.

James Wang 提到...

Yes I think I can add a new command or option to search values instead of always string.

Probably add a dialog also, not only in the command line.