[Oberon] PO2013 - Real time measurement

[Skulski, Wojciech]

John:

  BF535 was the first Blackfin. I dare not call it junk, but I believe it can be justifiably named a prototype. It was superseded by the newer Blackfins long ago.

I used BF532, BF533, and BF561. Most of my work was with BF561. I really appreciated Analog Devices documentation. Their Technical Reference Manual was fun to read, if this kind of literature can be fun. I designed a board based on this TRM and some reference design examples. It worked. Tuning the memory interface was straightforward. I stumbled across a silicon bug in a Marvell GbE PHY and I was able to fix this bug in the Linux kernel driver in about two weeks, despite the fact that it was my first encounter with the Linux kernel, and one of my first serious encounters with C. 

All in all, Blackfin BF561 was a a very good chip. My only real problem was that they did not integrate the GbE MAC on any Blackfin chip. So I switched to Sitara to gain performance. If I knew then what I know now, I would probably stay with Blackfin and try solving the data readout challenge in some other way. 

Wojtek
________________________________________
From: Oberon [oberon-bounces at lists.inf.ethz.ch] on behalf of John R. Strohm [strohm at airmail.net]
Sent: Tuesday, December 18, 2018 12:08 AM
To: ETH Oberon and related systems
Subject: Re: [Oberon] PO2013 - Real time measurement

My experience was with the BF535, bare metal, some 12-13 years ago.  uCLinux
was not an option for what we were doing.

-----Original Message-----
From: Skulski, Wojciech
Sent: Monday, December 17, 2018 10:57 AM
To: ETH Oberon and related systems
Subject: Re: [Oberon] PO2013 - Real time measurement

John:
>Analog Devices Blackfin SPI devices are a nightmare.
>(Friends don't let friends do Blackfin SPI.  Those words are written in my
>personal blood, thank you for asking.)

It is not that bad in my experience. I used Blackfin a lot with very few
problems. The trick was to use Analog Devices software, in particular their
uCLinux. It used to be very well supported by Robin Getz and his cohort,
until Analog Devices management stepped in and made "improvements" to the
user community structure. The support immediately went down drain. (Sorry,
Robin, it is my sincere experience.) I moved to ARM Sitara because of that.
Well, if you think that Blackfin was tough then try Sitara instead. Sitara
gives a whole new meaning to the word "nightmare".

Back to the topic, we used spidev under Blackfin uCLinux and it solved all
our problems at the Blackfin side. The remaining problems were due to the
ingenious modification of the SPI interface which Analog Devices made in
their ADC chips. Namely, they combined the input and output pins into one IO
pin. It saved one pin and created need to invert the pin direction with a
command. What a genius has invented this idea.

Yes, SPI can be an endless topic over a beer and around a campfire.

Wojtek
--
Oberon at lists.inf.ethz.ch mailing list for ETH Oberon and related systems
https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.inf.ethz.ch_mailman_listinfo_oberon&d=DwICAg&c=kbmfwr1Yojg42sGEpaQh5ofMHBeTl9EI2eaqQZhHbOU&r=uUiA_zLpwaGJIlq-_BM9w1wVOuyqPwHi3XzJRa-ybV0&m=YHxnBuD8Q84reJtzjulF1A5MG02DEE_T9UqFReCw1YQ&s=b-k1P2ZvK2fTUrFU1AOH-sUwPOoqNv4v2uff-NaNU08&e=


---
This email has been checked for viruses by Avast antivirus software.
https://urldefense.proofpoint.com/v2/url?u=https-3A__www.avast.com_antivirus&d=DwICAg&c=kbmfwr1Yojg42sGEpaQh5ofMHBeTl9EI2eaqQZhHbOU&r=uUiA_zLpwaGJIlq-_BM9w1wVOuyqPwHi3XzJRa-ybV0&m=YHxnBuD8Q84reJtzjulF1A5MG02DEE_T9UqFReCw1YQ&s=YnBdMsPdWYaxYwuocr4YCoU43b6gHy55sWLxNwU7gqY&e=

--
Oberon at lists.inf.ethz.ch mailing list for ETH Oberon and related systems
https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.inf.ethz.ch_mailman_listinfo_oberon&d=DwICAg&c=kbmfwr1Yojg42sGEpaQh5ofMHBeTl9EI2eaqQZhHbOU&r=uUiA_zLpwaGJIlq-_BM9w1wVOuyqPwHi3XzJRa-ybV0&m=YHxnBuD8Q84reJtzjulF1A5MG02DEE_T9UqFReCw1YQ&s=b-k1P2ZvK2fTUrFU1AOH-sUwPOoqNv4v2uff-NaNU08&e=