Re: user-defined alignment in gfortran
- From: Timo Schneider <timos@xxxxxxxxxxxxxxx>
- Date: Tue, 29 May 2007 14:03:39 +0000 (UTC)
glen herrmannsfeldt <gah@xxxxxxxxxxxxxxxx> schrieb:
Since a COMPLEX*16 variable is operated on, as far as Fortran is
concerned, as two eight byte real values it would seem in that
sense that eight byte alignment would be good enough. (It would
be interesting to have a processor with complex operations.)
Mm, why? If I have a COMPLEX*16 variable placed somwhere with eight byte
alignment, there can be two cases: It is 16 Byte aligned or not, in
that case only the second part of the variable falls together wih a 16
bit boundary.
Maybe I didn't say it quite right.
No hardware that I know of has instruction that operate on
complex data. There is no complex add or complex multiply
instruction. Operations are done on separate real and imaginary
parts using ordinary floating point instructions.
I think this is rather irrelevant for my problem. On the Cell-BE
architecture DMA transfers have to be 16 byte aligned. So I want my
Array or whatever I allocate in Fortran 16 byte aligned so that I can
DMA-transfer it to the Cell's SPU's later.
Regards,
Timo
.
- References:
- user-defined alignment in gfortran
- From: Timo Schneider
- Re: user-defined alignment in gfortran
- From: Dr Ivan D. Reid
- Re: user-defined alignment in gfortran
- From: glen herrmannsfeldt
- Re: user-defined alignment in gfortran
- From: Timo Schneider
- Re: user-defined alignment in gfortran
- From: glen herrmannsfeldt
- user-defined alignment in gfortran
- Prev by Date: Re: user-defined alignment in gfortran
- Next by Date: Re: user-defined alignment in gfortran
- Previous by thread: Re: user-defined alignment in gfortran
- Next by thread: Re: user-defined alignment in gfortran
- Index(es):
Relevant Pages
|