Hi This is possibly gcc 6.2 related: https://mail.coreboot.org/pipermail/coreboot/2016-December/082739.html
And a possible solution is Build GMP --with-pic, (only) if GCC defaults to -pie (preferred solution from reading the rest of above thread to the end). Regards, Jan On 21 September 2017 at 08:54, Jan Groenewald <j...@aims.ac.za> wrote: > Hi > > > > On 20 September 2017 at 13:32, Dima Pasechnik <dimp...@gmail.com> wrote: > >> >> >> On Wednesday, September 20, 2017 at 8:19:48 AM UTC+1, Jan Groenewald >> wrote: >>> >>> Hi >>> >>> On 20 September 2017 at 08:56, Jan Groenewald <j...@aims.ac.za> wrote: >>> >>>> Hi >>>> >>>> On 19 September 2017 at 19:25, Dima Pasechnik <dim...@gmail.com> wrote: >>>> >>>> Probably your gmp or mpir was not built correctly: >>>>> Indeed: >>>>> >>>>> /usr/bin/ld: /srv/sage-8.0/local/lib/libgmp.a(fat_entry.o): relocation >>>>> R_X86_64_32S against symbol `__gmpn_cpuvec' can not be used when making a >>>>> shared object; recompile with -fPIC >>>>> >>>>> >>>> mpir log attached. There is no gmp log. >>>> >>> /bin/bash ../libtool --mode=compile --tag=CC ../mpn/m4-ccas --m4="m4" >> gcc -c -DHAVE_CONFIG_H -D__GMP_WITHIN_GMP -I.. -DOPERATION_`echo fat_entry >> | sed 's/_$//'` -m64 -O2 -g -I. -I.. `test -f 'fat_entry.asm' || echo >> './'`fat_entry.asm >> >> libtool: compile: ../mpn/m4-ccas --m4=m4 gcc -c -DHAVE_CONFIG_H >> -D__GMP_WITHIN_GMP -I.. -DOPERATION_fat_entry -m64 -O2 -g -I. -I.. >> fat_entry.asm -fPIC -DPIC -o .libs/fat_entry.o >> m4 -DHAVE_CONFIG_H -D__GMP_WITHIN_GMP -DOPERATION_fat_entry -DPIC >> fat_entry.asm >tmp-fat_entry.s >> gcc -c -DHAVE_CONFIG_H -D__GMP_WITHIN_GMP -I.. -DOPERATION_fat_entry >> -m64 -O2 -g -I. -I.. tmp-fat_entry.s -fPIC -DPIC -o .libs/fat_entry.o >> >> libtool: compile: ../mpn/m4-ccas --m4=m4 gcc -c -DHAVE_CONFIG_H >> -D__GMP_WITHIN_GMP -I.. -DOPERATION_fat_entry -m64 -O2 -g -I. -I.. >> fat_entry.asm -o fat_entry.o >/dev/null 2>&1 >> >> these are lines in the log corresponding to compiling fat_entry.o, and it >> looks to me as if the last line does not >> say -fPIC anywhere. No idea whether this is the cause of the problem... >> > > I do see there is a trac issue already: https://trac.sagemath.org/ > ticket/23519 > > > >> >> >>> >>>> This persists with SAGE_FAT_BINARY=no and rerunning make as well. >>>> >>>> This occurs inside a libvirt/qemu virtual machine. >>>> Someone else posted this error related to docker in July: >>>> https://groups.google.com/d/msg/sage-devel/Jl071EqambM/yr4_RvjbBQAJ >>>> >>> >>> Just like that previous poster (in docker), if inside my VM I do: >>> >>> cd upstream; tar xf ecm*; cd ecm*; ./configure; make >>> >> >> but this is against another libgmp, I presume. >> > > Yes, must be the debian libgmp, it was not in a sage shell. > > Regards, > Jan > > > -- > .~. > /V\ Jan Groenewald > /( )\ www.aims.ac.za > ^^-^^ > -- .~. /V\ Jan Groenewald /( )\ www.aims.ac.za ^^-^^ -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
