Hi,
I use, from some year, the msp430fxxx, and now I have a question about
the internal RAM.
Is there a tecnique to test the internal ram?
I must certify my project, and the requirement is to test the ram.
In the specification, there are some tecniques, for example:
- parity-bit for ram
- ram test "galpat" or "transparent galpat"
- ram test "abraham"
- ram test "walk-path"
- crc
but It don't specifies how do it.

Is there an axample of this?

thank you
------------------------------------



(You need to be a member of msp430 -- send a blank email to msp430-subscribe@yahoogroups.com )

The RAM inside MSP430 do not have parity bit. All the other ways you
mentioned can be used on MSP430. All these are used for all kinds of
RAM, not limited to the RAM in MSP430. Thus a little search on the web
will answer your quetions.

--- In m...@yahoogroups.com, "plutoterraneo"
wrote:
>
> Hi,
> I use, from some year, the msp430fxxx, and now I have a question about
> the internal RAM.
> Is there a tecnique to test the internal ram?
> I must certify my project, and the requirement is to test the ram.
> In the specification, there are some tecniques, for example:
> - parity-bit for ram
> - ram test "galpat" or "transparent galpat"
> - ram test "abraham"
> - ram test "walk-path"
> - crc
> but It don't specifies how do it.
>
> Is there an axample of this?
>
> thank you
>

------------------------------------



(You need to be a member of msp430 -- send a blank email to msp430-subscribe@yahoogroups.com )

Hi,
I search on the web, but I don't find an explample of
procedure/function (write in C language) that execute the ram test of
microprocessor.
Can you help me?

--- In m...@yahoogroups.com, "old_cow_yellow"
wrote:
>
> The RAM inside MSP430 do not have parity bit. All the other ways you
> mentioned can be used on MSP430. All these are used for all kinds of
> RAM, not limited to the RAM in MSP430. Thus a little search on the web
> will answer your quetions.
>
> --- In m...@yahoogroups.com, "plutoterraneo"
> wrote:
> >
> > Hi,
> > I use, from some year, the msp430fxxx, and now I have a question about
> > the internal RAM.
> > Is there a tecnique to test the internal ram?
> > I must certify my project, and the requirement is to test the ram.
> > In the specification, there are some tecniques, for example:
> > - parity-bit for ram
> > - ram test "galpat" or "transparent galpat"
> > - ram test "abraham"
> > - ram test "walk-path"
> > - crc
> > but It don't specifies how do it.
> >
> > Is there an axample of this?
> >
> > thank you
>
------------------------------------



(You need to be a member of msp430 -- send a blank email to msp430-subscribe@yahoogroups.com )

Hi,

There's quite a bit of available information on the algorithms you request.
For example, I found the following text using Google in less than a minute.
It should be straightforward converting the description of the algorithm
into C or assembly. Just put some effort into it.

Good luck!

Pablo

Test Pattern Generation Algorithm

The testing algorithm for this project is known as the GALPAT test or
Galloping 1's and 0's. This is a very popular algorithm used by industry to
test RAM's. The GALPAT procedure is as follows:

http://www.eecis.udel.edu/%7Exingle/Projects/GALPAT/mini2.1.gif

Where n is the number of cells in the memory and C k is the k th memory
cell.

This procedure requires a solid background pattern of zeros to first be
written into the memory. The zero value in one bit position is then
verified. A single 1 is then written into that bit position of one word.
Every location in the memory is then read and verified, with the 1 value
being verified between each read. The single 1 is then moved to the next bit
position and the verification procedure is continued. Once this is complete,
the procedure is repeated starting with a solid background pattern of 1's
then writing and verifying a single 0.

Note: In the RAM under test, single bits cannot be directly accessed. You
will read from and write to entire words. For example, to write a one to a
single bit, address one word and write the 1 with 0's in all other positions
of the word. This test verifies that every cell can be set to both 0 and 1
without causing any other cell to change its state; thus it proves that all
the memory cells exist and there is no coupling between any two cells. Also,
all stuck-at faults in the decoder and in the memory array are covered by
the test.

De: m...@yahoogroups.com [mailto:m...@yahoogroups.com] En nombre de
plutoterraneo
Enviado el: Monday, August 04, 2008 10:23 AM
Para: m...@yahoogroups.com
Asunto: [msp430] Re: internal RAM test

Hi,
I search on the web, but I don't find an explample of
procedure/function (write in C language) that execute the ram test of
microprocessor.
Can you help me?

--- In m...@yahoogroups.com ,
"old_cow_yellow"
wrote:
>
> The RAM inside MSP430 do not have parity bit. All the other ways you
> mentioned can be used on MSP430. All these are used for all kinds of
> RAM, not limited to the RAM in MSP430. Thus a little search on the web
> will answer your quetions.
>
> --- In m...@yahoogroups.com ,
"plutoterraneo"
> wrote:
> >
> > Hi,
> > I use, from some year, the msp430fxxx, and now I have a question about
> > the internal RAM.
> > Is there a tecnique to test the internal ram?
> > I must certify my project, and the requirement is to test the ram.
> > In the specification, there are some tecniques, for example:
> > - parity-bit for ram
> > - ram test "galpat" or "transparent galpat"
> > - ram test "abraham"
> > - ram test "walk-path"
> > - crc
> > but It don't specifies how do it.
> >
> > Is there an axample of this?
> >
> > thank you
>

[Non-text portions of this message have been removed]
------------------------------------



(You need to be a member of msp430 -- send a blank email to msp430-subscribe@yahoogroups.com )

The trick to RAM tests on little micros like the MSP430, where there
are no real RAM "chips" or modules to test, is that you have to test
the same RAM that your code is executing out of. The stack usually
causes the biggest problem.

One clean but difficult way to do this is to force your code to
operate in one chunk of RAM and test the rest of the RAM. Then, move
the code RAM into another chunk and test the old chunk out of which
the program executed.

We take an easier route, but it's not quite as clean. During the
test, interrupts are disabled. A few variables are declared as
"registers", just in case the compiler decided to use regular RAM. We
then test every word in RAM (easier and just as fast to do 16-bit
words) by reading it back into the register variable, writing a test
pattern (the one's complement of the address) to the word, reading it
back and comparing it to the 1's complement of the register variable.

We use the 1's complement of the address for a few reasons. We wanted
a changing pattern. We didn't want to write the address itself since
there is the remote possibility of just reading back the address
instead of the data; that's a throwback to the days of external RAMs,
buses, etc.

You MUST spend a lot of time looking at the dissassembly to make sure
the code is doing what you expect. (If in doubt you can write it in
assembly, but that is not my preference.) We spent a good amount of
time verifying this test function, breaking the code and mangling
registers, etc. You should, too. It sounds like yout application is
mission-critical.

Good luck.

Stuart

--- In m...@yahoogroups.com, "plutoterraneo"
wrote:
>
> Hi,
> I use, from some year, the msp430fxxx, and now I have a question about
> the internal RAM.
> Is there a tecnique to test the internal ram?
> I must certify my project, and the requirement is to test the ram.
> In the specification, there are some tecniques, for example:
> - parity-bit for ram
> - ram test "galpat" or "transparent galpat"
> - ram test "abraham"
> - ram test "walk-path"
> - crc
> but It don't specifies how do it.
>
> Is there an axample of this?
>
> thank you
>

------------------------------------



(You need to be a member of msp430 -- send a blank email to msp430-subscribe@yahoogroups.com )

Hi,
I understand the Algorithm, but I have a question:
-If I change the value of ram (1 or 0) for the test, finally my ram
isn't valid, becuase before testing in my ram there were the values of
my ram variables.

--- In m...@yahoogroups.com, "Pablo Villaverde Padilla"
wrote:
>
> Hi,
>
>
>
> There's quite a bit of available information on the algorithms you
request.
> For example, I found the following text using Google in less than a
minute.
> It should be straightforward converting the description of the algorithm
> into C or assembly. Just put some effort into it.
>
>
>
> Good luck!
>
>
>
> Pablo
>
>
>
> Test Pattern Generation Algorithm
>
> The testing algorithm for this project is known as the GALPAT test or
> Galloping 1's and 0's. This is a very popular algorithm used by
industry to
> test RAM's. The GALPAT procedure is as follows:
>
>
>
> http://www.eecis.udel.edu/%7Exingle/Projects/GALPAT/mini2.1.gif
>
> Where n is the number of cells in the memory and C k is the k th memory
> cell.
>
>
>
> This procedure requires a solid background pattern of zeros to first be
> written into the memory. The zero value in one bit position is then
> verified. A single 1 is then written into that bit position of one word.
> Every location in the memory is then read and verified, with the 1 value
> being verified between each read. The single 1 is then moved to the
next bit
> position and the verification procedure is continued. Once this is
complete,
> the procedure is repeated starting with a solid background pattern
of 1's
> then writing and verifying a single 0.
>
>
>
> Note: In the RAM under test, single bits cannot be directly
accessed. You
> will read from and write to entire words. For example, to write a
one to a
> single bit, address one word and write the 1 with 0's in all other
positions
> of the word. This test verifies that every cell can be set to both 0
and 1
> without causing any other cell to change its state; thus it proves
that all
> the memory cells exist and there is no coupling between any two
cells. Also,
> all stuck-at faults in the decoder and in the memory array are
covered by
> the test.
>
>
>
>
>
> De: m...@yahoogroups.com [mailto:m...@yahoogroups.com] En nombre de
> plutoterraneo
> Enviado el: Monday, August 04, 2008 10:23 AM
> Para: m...@yahoogroups.com
> Asunto: [msp430] Re: internal RAM test
>
>
>
> Hi,
> I search on the web, but I don't find an explample of
> procedure/function (write in C language) that execute the ram test of
> microprocessor.
> Can you help me?
>
> --- In m...@yahoogroups.com ,
> "old_cow_yellow"
> wrote:
> >
> > The RAM inside MSP430 do not have parity bit. All the other ways you
> > mentioned can be used on MSP430. All these are used for all kinds of
> > RAM, not limited to the RAM in MSP430. Thus a little search on the web
> > will answer your quetions.
> >
> > --- In m...@yahoogroups.com ,
> "plutoterraneo"
> > wrote:
> > >
> > > Hi,
> > > I use, from some year, the msp430fxxx, and now I have a question
about
> > > the internal RAM.
> > > Is there a tecnique to test the internal ram?
> > > I must certify my project, and the requirement is to test the ram.
> > > In the specification, there are some tecniques, for example:
> > > - parity-bit for ram
> > > - ram test "galpat" or "transparent galpat"
> > > - ram test "abraham"
> > > - ram test "walk-path"
> > > - crc
> > > but It don't specifies how do it.
> > >
> > > Is there an axample of this?
> > >
> > > thank you
> > >
> >
>
> [Non-text portions of this message have been removed]
>

------------------------------------



(You need to be a member of msp430 -- send a blank email to msp430-subscribe@yahoogroups.com )