![]() |
![]() |
#1 |
Oct 2006
vomit_frame_pointer
23·32·5 Posts |
![]()
I'm running 64-bit Linux on AMD Opterons.
Running "make x86_64" produces a slightly slow set of gnfs-lasieve4I1[234]e binaries, and also makes a procrels which consistently crashes. The secret may lie in the target "common" in the top-level makefile. Code:
common : echo "#define GGNFS_VERSION \"$(VERSION)-$(ARCH)\"" > include/version.h @cd src/lasieve4 && rm -f -r asm && ln -s piii asm @HOST=generic ARCH=$(ARCH) $(MAKE) -C src Am I looking in the right direction? Any pointers? Last fiddled with by FactorEyes on 2008-02-19 at 23:18 |
![]() |
![]() |
![]() |
#2 |
Jul 2003
So Cal
1000000101002 Posts |
![]()
You must be using an old source. The common target should use the ppc32 directory:
Code:
common : echo "#define GGNFS_VERSION \"$(VERSION)-$(ARCH)\"" > include/version.h @cd src/lasieve4 && rm -f -r asm && ln -s ppc32 asm @HOST=generic ARCH=$(ARCH) $(MAKE) -C src Greg |
![]() |
![]() |
![]() |
#3 |
Oct 2006
vomit_frame_pointer
23·32·5 Posts |
![]()
My error. I had replaced "ppc32" with "generic", and then switched it back to "piii" for some weird reason.
It doesn't work anyway -- I think the lack of assembler code files in the /generic directory is one among many reasons for this. Your post on another thread just dealt with this anyway: the assembler code for AMD64 hasn't been written yet, at least for public consumption. |
![]() |
![]() |
![]() |
Thread Tools | |
![]() |
||||
Thread | Thread Starter | Forum | Replies | Last Post |
Compiling GNFS sievers on AArch64 platform | wombatman | Programming | 11 | 2017-03-11 03:12 |
compiling GMP-ECM | ATH | GMP-ECM | 69 | 2017-01-04 12:03 |
Need help compiling YAFU from SVN. | Stargate38 | YAFU | 14 | 2016-01-20 21:46 |
Compiling 24.14 | CBoland | Software | 6 | 2007-08-01 00:11 |
Compiling GMP-ECM With GWNUM | tmorrow | GMP-ECM | 5 | 2007-04-04 00:39 |