[Unmoderated] ArticiaS: mystery finaly solved ? | ANN.lu |
Posted on 08-Jul-2004 06:47 GMT by brotheris | 140 comments View flat View list |
Here's the summary of the last posts from hot topic. It may finaly put some dots on I's. Up to now we have heared a lot of small bits from variuos parties and finaly we can put the puzzle together. Read more about it.
I'll play Amon_Re of the past:
It all started when Chris Hogdes started explaining few things (in this thread and @226 comment).
During DMA transfers, the ArticiaS does not flag accessed memory as "dirty", therefore the CPU does not automatically know, that it has to update/flush its caches
Later (@ comment 247, 248 and others) Bernie Meyer explained how such a lack of feature (or call it a bug) affects stability, performance and may cause data corruption even in AmigaOS-like enviroment while using CachePreDMA()/CachePostDMA().
And then we discover quotes from ArticiaS documentation:
"The snoop cycle is used to probe the primary and secondary cache for updated data when the PCI
accesses DRAM. This is done to maintain data coherency between the Floating Buffer, DRAM and both
caches. The Articia S performs the Snoop cycle. When there is a snoop hit on a modified cache line in
either level one or two cache, the contents are written back directly to the Floating Buffer. A PCI Bus
master can subsequently later on fetch the data directly from the Floating Buffer. The Floating Buffer is
flushed back to DRAM during a PCI write cycle. The corresponding line in level one or level two cache is
thus invalidated. Snoops are hidden, meaning the CPU can continue its current data access without
being interrupted while the Articia S simultaneously queries both caches."
You can find similar information using google cache. It seems like some people lied. Is lack of Cache Coherency a bug or a feature (it was advertised that there is Cache Coherency, so it had to work) ? We may now put this case to rest.
|
|
List of all comments to this article |
ArticiaS: mystery finaly solved ? : Comment 102 of 140 | ANN.lu |
Posted by Janne on 11-Jul-2004 07:46 GMT | In reply to Comment 97 (Sammy Nordström): >No, you have facts proving certain circumstances, but not the cause for those
>circumstances. You are making use of those facts for specualtions that goes
>way beyond what we know as a matter of fact, that's all. But then, you're
>free to *prove* me wrong anytime.
I guess, if you wish to continue with this court analogy (which I find a bit melodramatic but gladly play along :-), what Johan has is circumstantial evidence.
- To date, and this is more than two year old chip, we have no fully working DMA hard-disk driver in the public for Articia S.
That is certainly circumstantial evidence, not a smoking gun or an eye-witness. I think we both agree on that. Probably not enough to convict a murderer beyond "reasonable doubt", right? How about coupled with other testimony? We have differing eye-witness accounts:
- bPlan engineers saying it is broken and making an expensive switch-over
- AFAIK Tratech/Barbie engineers saying it is broken, canceling product
- Eyetech/MAI saying it is not broken, continuing with their product
I guess, in a court of law, it would come down to a jury deciding who to trust. They have the so far indisputable circumstantial evidence mentioned above, and they have these eye-witnesses saying different things.
And I guess that is the way many things unfold in life as well. The facts that you, Sammy, so eagerly are asking and looking for, very often are not so clear. There are many things you would consider fact but someone else might still dispute them. Both might have evidence. Funny thing is, this is the whole meaning of science - we think we know something, yet science spends a lot of its time trying to prove the current science wrong and redefine our understanding of the universe. And obviously, so it should be. So, really, who to trust?
I don't think Johan's question to you is all that unreasonable. Nor do I think his stance is. Once a working driver is out there and independently verified, we at least know the chip can be made to work (whether there are defects or not). If this, for one reason or another, never happens, or some people still have issues with the driver and many do not, we may never really know for sure.
Personally I think a working Linux driver should be a requirement as well (considering the chip's/board's target market), but that's just my opinion.
I guess Johan's question (or mine at least) is: Will you, if things remain even a tad bit unclear (e.g. as long as one party out there claims otherwise), always consider the chip defect-free, or is there some limit? How much time, how many people, what kind of evidence (or who specifically) would it take for you to believe otherwise or at least believe in the possibility of it?
And yes, Sammy, I must agree with those who think you degenerate threads with meaningless word games. Not so much your message, but the volume of it. |
|
List of all comments to this article (continued) |
|
- User Menu
-
- About ANN archives
- The ANN archives is powered by #AmigaZeux. It was updated daily (news last: 22-Oct-2004; comments last: 18-May-2005).
ANN.lu was created, previously owned and maintained by Christian Kemp, www.ckemp.com.
- Contribute
- Not possible at this time!
- Search ANN archives
- Advanced search
- Hosting
- ANN.lu was hosted by Dreamhost. Sign up through this link, mention "ckemp" as referrer and he will get a 10% commission on any account you purchase.
Please show your appreciation for any past, present and future work on ANN.lu by making a contribution via PayPal.
|