Hi all, My reply below. Best, Paul Leopardi On Sat, 7 May 2005 10:13 pm, Muge Unel wrote: > HI gcc-guru's, > > I have a question: > i am having my first compiling experiences on an opteron machine+64bit > arch and testing various compilers and options. > This is on > RH ES x86_64 x86_64 x86_64 GNU/Linux AMD Opteron Processor 248, using > "/usr/bin/gcc -O3" with either an -m32 or -m64 option. > My main concern is how to obtain the maximum precision from the > architecture, expecially 128-bit quad. precision. > The first thing i checked is the sizes of data types. > I think i get the correct behaviour for that: > for example, 16B with 64bit compilation and 12B for 32, for a long > double size. > however, the results of my simple calculations do not seem to make > sense. Can somebody on this list point me to any documentation which says that long double on amd64 is supposed to have 128 bits of arithmetic precision? I've always thought that it was only 80 bits, as per this statement, "The size of long double is 128 bits with only ïrst 80 bits used to match native x87 format. The rest is just padding to keep long double values 128- bit aligned so loads and stores are effective." Ref: Jan Hubicka, "Porting GCC to the AMD64 architecture", http://www.ucw.cz/~hubicka/papers/amd64-2.pdf
