|Cray-2 Fast Memory email@example.com (1993-05-13)|
|Re: Cray-2 Fast Memory firstname.lastname@example.org (1993-05-14)|
|Re: Cray-2 Fast Memory email@example.com (1993-05-26)|
|Re: Cray-2 Fast Memory firstname.lastname@example.org (1993-05-26)|
|Re: Cray-2 Fast Memory email@example.com (1993-05-27)|
|Re: Cray-2 Fast Memory firstname.lastname@example.org (1993-05-31)|
|From:||email@example.com (David desJardins)|
|Keywords:||registers, optimize, Cray|
|Organization:||IDA Center for Communications Research, Princeton|
|Date:||Wed, 26 May 1993 02:58:56 GMT|
Patrick Delano <firstname.lastname@example.org> writes:
> Apparently the Cray-2 had a fast memory that unlike cache memory was
> explicitly managed by the compiler. Can anyone tell me what software
> techniques were used, or point me to some references?
Basically, no software techniques were used. The compiler does very
little to take advantage of the local memory. As far as I am aware, the
only ways in which it is used are the following:
o Temporary storage for register spillage.
o As a means of extracting scalar values from vector registers
(which can be done directly on the X-MP and Y-MP).
o When the programmer, by directive, explicitly indicates that a
variable is to be placed in local rather than common memory.
I believe that the primary reason that more sophisticated techniques
were not used by the compiler is that less than 40 Cray-2 machines were
manufactured and sold, compared to hundreds of X-MP and Y-MP type
> This paper talks a lot about techniques to minimize the use of
> temporary storage to one or several lengths of a vector register (which
> obviously fits well with the availability of local memory).
I don't understand this. Minimizing temporary storage is _less_
important with a fast local memory, because register spillage is less
expensive than if it had to go to common memory.
I agree that minimizing spillage is an important issue, but I don't see
what it has to do with local memory, or the lack of it.
Return to the
Search the comp.compilers archives again.