[Oberon] Memory

skulski at pas.rochester.edu skulski at pas.rochester.edu
Tue Sep 1 18:32:32 CEST 2015


Peter:

thank you for the web pointers. These memories are very interesting.

We are using off-the-shelf chips in our designs. If the phase change
memory becomes one of these then we will start using it. I can imagine
that the compiler in the embedded system can live in slow memory whose
performance would be similar to an SD card. One can use a regular SPI
flash for doing that, but without the file system but rather
execute-in-place like Chris is doing. It may seem like the SD card is a
more flexible solution, but in a high reliability system a flash chip may
be better because it cannot fall out.

So I think that this discussion is pointing at the concept of memory
hierarchy in a system. The case in point is not so much flash (the SD card
can serve this role) but the BRAM-SRAM-SDRAM hierarchy, transparently
mapped to the Oberon System memory space. Can Oberon System do that? The
companion concept is how to control which variables and which pieces of
code are placed in which memory chunk. Can the loader do that? Does the
Oberon System need to be changed, and to what extent? Is this a small
change, or a big one?

I am truly grateful for such a great discussion on this list. I am
learning a lot from the Oberon experts who are chiming in.

Thank you all!

Wojtek

> Hello Wojtek,
>
> Eventually phase change memory might replace both slow and
> fast memories in your context.
> https://urldefense.proofpoint.com/v2/url?u=https-3A__en.wikipedia.org_wiki_Phase-2Dchange-5Fmemory&d=BQICAg&c=kbmfwr1Yojg42sGEpaQh5ofMHBeTl9EI2eaqQZhHbOU&r=uUiA_zLpwaGJIlq-_BM9w1wVOuyqPwHi3XzJRa-ybV0&m=kAta350wAUPw3XNBa8nVD8FGvDMnq1rERLB2cRfIhnU&s=oAN3WKOn10CYWT4o0XvdUaihaFdkLZf7YH3nQp4cEcA&e=
>
> https://urldefense.proofpoint.com/v2/url?u=http-3A__www.research.ibm.com_labs_zurich_sto_pcm_&d=BQICAg&c=kbmfwr1Yojg42sGEpaQh5ofMHBeTl9EI2eaqQZhHbOU&r=uUiA_zLpwaGJIlq-_BM9w1wVOuyqPwHi3XzJRa-ybV0&m=kAta350wAUPw3XNBa8nVD8FGvDMnq1rERLB2cRfIhnU&s=7n65zLlShgA11gwrs0ODeOV73_nO69v8l3ba_-8De54&e=
> Currently it is too slow or is too expensive for your requirement?
>
> Regards,                         ... Peter E.
>
> --
> 123456789 123456789 123456789 123456789 123456789 123456789 123456789 12
> Tel +13606390202
> https://urldefense.proofpoint.com/v2/url?u=http-3A__easthope.ca_peter.html&d=BQICAg&c=kbmfwr1Yojg42sGEpaQh5ofMHBeTl9EI2eaqQZhHbOU&r=uUiA_zLpwaGJIlq-_BM9w1wVOuyqPwHi3XzJRa-ybV0&m=kAta350wAUPw3XNBa8nVD8FGvDMnq1rERLB2cRfIhnU&s=TDvdABDtK_bqkMX6rW1BxgdmaNitk8Ed5AVtZUuJGyU&e=
>  Bcc: peter at easthope. ca
>
>
> --
> 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=BQICAg&c=kbmfwr1Yojg42sGEpaQh5ofMHBeTl9EI2eaqQZhHbOU&r=uUiA_zLpwaGJIlq-_BM9w1wVOuyqPwHi3XzJRa-ybV0&m=kAta350wAUPw3XNBa8nVD8FGvDMnq1rERLB2cRfIhnU&s=72sC10woSpp-bOYtYPXLwljwT6Xh2JmxL4GHxUbWPZs&e=
>
>





More information about the Oberon mailing list